From cf8f3d4deb02a8fdc806c46d4112b69868544697 Mon Sep 17 00:00:00 2001
From: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Date: Wed, 15 Feb 2023 20:31:36 +0200
Subject: wifi: ath11k: Set ext passive scan flag to adjust passive scan start
 time

Set the WMI_SCAN_FLAG_EXT_PASSIVE_SCAN_START_TIME_ENHANCE flag
while sending the scan command.  If this flag is enabled when the
incoming scan request comes with a strict start time and its duration
overlaps with next TBTT, then target adjust the start time accordingly
for passive scan. Target supporting this feature will advertise
WMI_TLV_SERVICE_PASSIVE_SCAN_START_TIME_ENHANCE.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.4.0.1-01467-QCAHKSWPL_SILICONZ-1

Signed-off-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20221222131720.11368-1-quic_tamizhr@quicinc.com
---
 drivers/net/wireless/ath/ath11k/wmi.c | 8 ++++++++
 drivers/net/wireless/ath/ath11k/wmi.h | 3 +++
 2 files changed, 11 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index b3a7d7bfe17c..7057c7916d5d 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -2068,6 +2068,12 @@ void ath11k_wmi_start_scan_init(struct ath11k *ar,
 				  WMI_SCAN_EVENT_FOREIGN_CHAN |
 				  WMI_SCAN_EVENT_DEQUEUED;
 	arg->scan_flags |= WMI_SCAN_CHAN_STAT_EVENT;
+
+	if (test_bit(WMI_TLV_SERVICE_PASSIVE_SCAN_START_TIME_ENHANCE,
+		     ar->ab->wmi_ab.svc_map))
+		arg->scan_ctrl_flags_ext |=
+			WMI_SCAN_FLAG_EXT_PASSIVE_SCAN_START_TIME_ENHANCE;
+
 	arg->num_bssid = 1;
 
 	/* fill bssid_list[0] with 0xff, otherwise bssid and RA will be
@@ -2149,6 +2155,8 @@ ath11k_wmi_copy_scan_event_cntrl_flags(struct wmi_start_scan_cmd *cmd,
 	/* for adaptive scan mode using 3 bits (21 - 23 bits) */
 	WMI_SCAN_SET_DWELL_MODE(cmd->scan_ctrl_flags,
 				param->adaptive_dwell_time_mode);
+
+	cmd->scan_ctrl_flags_ext = param->scan_ctrl_flags_ext;
 }
 
 int ath11k_wmi_send_scan_start_cmd(struct ath11k *ar,
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 0a045af5419b..2273d885f7fa 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2093,6 +2093,7 @@ enum wmi_tlv_service {
 	WMI_TLV_SERVICE_EXT2_MSG = 220,
 	WMI_TLV_SERVICE_PEER_POWER_SAVE_DURATION_SUPPORT = 246,
 	WMI_TLV_SERVICE_SRG_SRP_SPATIAL_REUSE_SUPPORT = 249,
+	WMI_TLV_SERVICE_PASSIVE_SCAN_START_TIME_ENHANCE = 263,
 
 	/* The second 128 bits */
 	WMI_MAX_EXT_SERVICE = 256,
@@ -3223,6 +3224,7 @@ struct  wmi_start_scan_cmd {
 
 #define WMI_SCAN_DWELL_MODE_MASK 0x00E00000
 #define WMI_SCAN_DWELL_MODE_SHIFT        21
+#define WMI_SCAN_FLAG_EXT_PASSIVE_SCAN_START_TIME_ENHANCE   0x00000800
 
 enum {
 	WMI_SCAN_DWELL_MODE_DEFAULT      = 0,
@@ -3270,6 +3272,7 @@ struct scan_req_params {
 		};
 		u32 scan_events;
 	};
+	u32 scan_ctrl_flags_ext;
 	u32 dwell_time_active;
 	u32 dwell_time_active_2g;
 	u32 dwell_time_passive;
-- 
cgit v1.2.3-70-g09d2


From 778f83f889e7fca37780d9640fcbd0229ae38eaa Mon Sep 17 00:00:00 2001
From: "Alexey V. Vissarionov" <gremlin@altlinux.org>
Date: Wed, 15 Feb 2023 20:31:37 +0200
Subject: wifi: ath6kl: minor fix for allocation size

Although the "param" pointer occupies more or equal space compared
to "*param", the allocation size should use the size of variable
itself.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: bdcd81707973cf8a ("Add ath6kl cleaned up driver")
Signed-off-by: Alexey V. Vissarionov <gremlin@altlinux.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230117110414.GC12547@altlinux.org
---
 drivers/net/wireless/ath/ath6kl/bmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath6kl/bmi.c b/drivers/net/wireless/ath/ath6kl/bmi.c
index bde5a10d470c..af98e871199d 100644
--- a/drivers/net/wireless/ath/ath6kl/bmi.c
+++ b/drivers/net/wireless/ath/ath6kl/bmi.c
@@ -246,7 +246,7 @@ int ath6kl_bmi_execute(struct ath6kl *ar, u32 addr, u32 *param)
 		return -EACCES;
 	}
 
-	size = sizeof(cid) + sizeof(addr) + sizeof(param);
+	size = sizeof(cid) + sizeof(addr) + sizeof(*param);
 	if (size > ar->bmi.max_cmd_size) {
 		WARN_ON(1);
 		return -EINVAL;
-- 
cgit v1.2.3-70-g09d2


From 480c9df5778774117546f6389be1a8dc8cc935db Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Wed, 15 Feb 2023 20:31:37 +0200
Subject: wifi: ath12k: Fix spelling mistakes in warning messages and comments

There are quite a few spelling mistakes in warning messages and a lot
of the comments. Fix these.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230214092122.265336-1-colin.i.king@gmail.com
---
 drivers/net/wireless/ath/ath12k/ce.c       |  2 +-
 drivers/net/wireless/ath/ath12k/core.h     |  2 +-
 drivers/net/wireless/ath/ath12k/dp.c       |  2 +-
 drivers/net/wireless/ath/ath12k/dp.h       |  6 +++---
 drivers/net/wireless/ath/ath12k/dp_mon.c   | 10 +++++-----
 drivers/net/wireless/ath/ath12k/dp_rx.c    |  8 ++++----
 drivers/net/wireless/ath/ath12k/hal.c      |  2 +-
 drivers/net/wireless/ath/ath12k/hal.h      | 12 ++++++------
 drivers/net/wireless/ath/ath12k/hal_desc.h | 10 +++++-----
 drivers/net/wireless/ath/ath12k/rx_desc.h  |  2 +-
 drivers/net/wireless/ath/ath12k/wmi.c      |  6 +++---
 drivers/net/wireless/ath/ath12k/wmi.h      |  4 ++--
 12 files changed, 33 insertions(+), 33 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/ce.c b/drivers/net/wireless/ath/ath12k/ce.c
index aed6987804bf..be0d669d31fc 100644
--- a/drivers/net/wireless/ath/ath12k/ce.c
+++ b/drivers/net/wireless/ath/ath12k/ce.c
@@ -946,7 +946,7 @@ int ath12k_ce_alloc_pipes(struct ath12k_base *ab)
 
 		ret = ath12k_ce_alloc_pipe(ab, i);
 		if (ret) {
-			/* Free any parial successful allocation */
+			/* Free any partial successful allocation */
 			ath12k_ce_free_pipes(ab);
 			return ret;
 		}
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index a54ae74543c1..dffa687ee40e 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -691,7 +691,7 @@ struct ath12k_base {
 
 	/* Below regd's are protected by ab->data_lock */
 	/* This is the regd set for every radio
-	 * by the firmware during initializatin
+	 * by the firmware during initialization
 	 */
 	struct ieee80211_regdomain *default_regd[MAX_RADIOS];
 	/* This regd is set during dynamic country setting
diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index eb0261500ab0..9926d81c5fe4 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -1429,7 +1429,7 @@ static int ath12k_dp_cc_init(struct ath12k_base *ab)
 		}
 
 		if (dp->spt_info[i].paddr & ATH12K_SPT_4K_ALIGN_CHECK) {
-			ath12k_warn(ab, "SPT allocated memoty is not 4K aligned");
+			ath12k_warn(ab, "SPT allocated memory is not 4K aligned");
 			ret = -EINVAL;
 			goto free;
 		}
diff --git a/drivers/net/wireless/ath/ath12k/dp.h b/drivers/net/wireless/ath/ath12k/dp.h
index 36a876d7f61d..7c5dafce5a68 100644
--- a/drivers/net/wireless/ath/ath12k/dp.h
+++ b/drivers/net/wireless/ath/ath12k/dp.h
@@ -371,7 +371,7 @@ struct ath12k_dp {
 
 #define HTT_TX_WBM_COMP_STATUS_OFFSET 8
 
-/* HTT tx completion is overlayed in wbm_release_ring */
+/* HTT tx completion is overlaid in wbm_release_ring */
 #define HTT_TX_WBM_COMP_INFO0_STATUS		GENMASK(16, 13)
 #define HTT_TX_WBM_COMP_INFO1_REINJECT_REASON	GENMASK(3, 0)
 #define HTT_TX_WBM_COMP_INFO1_EXCEPTION_FRAME	BIT(4)
@@ -545,7 +545,7 @@ enum htt_srng_ring_id {
  *                     3'b010: 4 usec
  *                     3'b011: 8 usec (default)
  *                     3'b100: 16 usec
- *                     Others: Reserverd
+ *                     Others: Reserved
  *           b'19    - response_required:
  *                     Host needs HTT_T2H_MSG_TYPE_SRING_SETUP_DONE as response
  *           b'20:31 - reserved:  reserved for future use
@@ -1126,7 +1126,7 @@ struct htt_tx_ring_selection_cfg_cmd {
 	__le32 tlv_filter_mask_in1;
 	__le32 tlv_filter_mask_in2;
 	__le32 tlv_filter_mask_in3;
-	__le32 reserverd[3];
+	__le32 reserved[3];
 } __packed;
 
 #define HTT_TX_RING_TLV_FILTER_MGMT_DMA_LEN	GENMASK(3, 0)
diff --git a/drivers/net/wireless/ath/ath12k/dp_mon.c b/drivers/net/wireless/ath/ath12k/dp_mon.c
index a214797c96a2..4f93e4c95fed 100644
--- a/drivers/net/wireless/ath/ath12k/dp_mon.c
+++ b/drivers/net/wireless/ath/ath12k/dp_mon.c
@@ -813,7 +813,7 @@ ath12k_dp_mon_rx_parse_status_tlv(struct ath12k_base *ab,
 		spin_unlock_bh(&buf_ring->idr_lock);
 
 		if (unlikely(!msdu)) {
-			ath12k_warn(ab, "montior destination with invalid buf_id %d\n",
+			ath12k_warn(ab, "monitor destination with invalid buf_id %d\n",
 				    buf_id);
 			return HAL_RX_MON_STATUS_PPDU_NOT_DONE;
 		}
@@ -1124,7 +1124,7 @@ static void ath12k_dp_mon_rx_deliver_msdu(struct ath12k *ar, struct napi_struct
 
 	/* PN for multicast packets are not validate in HW,
 	 * so skip 802.3 rx path
-	 * Also, fast_rx expectes the STA to be authorized, hence
+	 * Also, fast_rx expects the STA to be authorized, hence
 	 * eapol packets are sent in slow path.
 	 */
 	if (decap == DP_RX_DECAP_TYPE_ETHERNET2_DIX && !is_eapol_tkip &&
@@ -1917,7 +1917,7 @@ ath12k_dp_mon_tx_parse_status_tlv(struct ath12k_base *ab,
 		spin_unlock_bh(&buf_ring->idr_lock);
 
 		if (unlikely(!msdu)) {
-			ath12k_warn(ab, "montior destination with invalid buf_id %d\n",
+			ath12k_warn(ab, "monitor destination with invalid buf_id %d\n",
 				    buf_id);
 			return DP_MON_TX_STATUS_PPDU_NOT_DONE;
 		}
@@ -2110,7 +2110,7 @@ int ath12k_dp_mon_srng_process(struct ath12k *ar, int mac_id, int *budget,
 		spin_unlock_bh(&buf_ring->idr_lock);
 
 		if (unlikely(!skb)) {
-			ath12k_warn(ab, "montior destination with invalid buf_id %d\n",
+			ath12k_warn(ab, "monitor destination with invalid buf_id %d\n",
 				    buf_id);
 			goto move_next;
 		}
@@ -2511,7 +2511,7 @@ int ath12k_dp_mon_rx_process_stats(struct ath12k *ar, int mac_id,
 		spin_unlock_bh(&buf_ring->idr_lock);
 
 		if (unlikely(!skb)) {
-			ath12k_warn(ab, "montior destination with invalid buf_id %d\n",
+			ath12k_warn(ab, "monitor destination with invalid buf_id %d\n",
 				    buf_id);
 			goto move_next;
 		}
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 83a43ad48c51..eb67b3409f85 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2443,7 +2443,7 @@ static void ath12k_dp_rx_deliver_msdu(struct ath12k *ar, struct napi_struct *nap
 
 	/* PN for multicast packets are not validate in HW,
 	 * so skip 802.3 rx path
-	 * Also, fast_rx expectes the STA to be authorized, hence
+	 * Also, fast_rx expects the STA to be authorized, hence
 	 * eapol packets are sent in slow path.
 	 */
 	if (decap == DP_RX_DECAP_TYPE_ETHERNET2_DIX && !is_eapol &&
@@ -2611,7 +2611,7 @@ try_again:
 		if (!desc_info) {
 			desc_info = ath12k_dp_get_rx_desc(ab, cookie);
 			if (!desc_info) {
-				ath12k_warn(ab, "Invalid cookie in manual desc retrival");
+				ath12k_warn(ab, "Invalid cookie in manual desc retrieval");
 				continue;
 			}
 		}
@@ -3297,7 +3297,7 @@ ath12k_dp_process_rx_err_buf(struct ath12k *ar, struct hal_reo_dest_ring *desc,
 	if (!desc_info) {
 		desc_info = ath12k_dp_get_rx_desc(ab, cookie);
 		if (!desc_info) {
-			ath12k_warn(ab, "Invalid cookie in manual desc retrival");
+			ath12k_warn(ab, "Invalid cookie in manual desc retrieval");
 			return -EINVAL;
 		}
 	}
@@ -3718,7 +3718,7 @@ int ath12k_dp_rx_process_wbm_err(struct ath12k_base *ab,
 		if (!desc_info) {
 			desc_info = ath12k_dp_get_rx_desc(ab, err_info.cookie);
 			if (!desc_info) {
-				ath12k_warn(ab, "Invalid cookie in manual desc retrival");
+				ath12k_warn(ab, "Invalid cookie in manual desc retrieval");
 				continue;
 			}
 		}
diff --git a/drivers/net/wireless/ath/ath12k/hal.c b/drivers/net/wireless/ath/ath12k/hal.c
index 95d04819083f..0ec53afe9915 100644
--- a/drivers/net/wireless/ath/ath12k/hal.c
+++ b/drivers/net/wireless/ath/ath12k/hal.c
@@ -609,7 +609,7 @@ static int ath12k_hal_srng_create_config_qcn9274(struct ath12k_base *ab)
 		HAL_WBM0_RELEASE_RING_BASE_LSB(ab);
 	s->reg_size[1] = HAL_WBM1_RELEASE_RING_HP - HAL_WBM0_RELEASE_RING_HP;
 
-	/* Some LMAC rings are not accesed from the host:
+	/* Some LMAC rings are not accessed from the host:
 	 * RXDMA_BUG, RXDMA_DST, RXDMA_MONITOR_BUF, RXDMA_MONITOR_STATUS,
 	 * RXDMA_MONITOR_DST, RXDMA_MONITOR_DESC, RXDMA_DIR_BUF_SRC,
 	 * RXDMA_RX_MONITOR_BUF, TX_MONITOR_BUF, TX_MONITOR_DST, SW2RXDMA
diff --git a/drivers/net/wireless/ath/ath12k/hal.h b/drivers/net/wireless/ath/ath12k/hal.h
index dfbd8bce70e5..0d4fa12ea622 100644
--- a/drivers/net/wireless/ath/ath12k/hal.h
+++ b/drivers/net/wireless/ath/ath12k/hal.h
@@ -270,7 +270,7 @@ struct ath12k_base;
 #define HAL_WBM_SW_COOKIE_CONV_CFG_WBM2SW4_EN		BIT(5)
 #define HAL_WBM_SW_COOKIE_CONV_CFG_GLOBAL_EN		BIT(8)
 
-/* TCL ring feild mask and offset */
+/* TCL ring field mask and offset */
 #define HAL_TCL1_RING_BASE_MSB_RING_SIZE		GENMASK(27, 8)
 #define HAL_TCL1_RING_BASE_MSB_RING_BASE_ADDR_MSB	GENMASK(7, 0)
 #define HAL_TCL1_RING_ID_ENTRY_SIZE			GENMASK(7, 0)
@@ -296,7 +296,7 @@ struct ath12k_base;
 #define HAL_TCL1_RING_FIELD_DSCP_TID_MAP6		GENMASK(20, 18)
 #define HAL_TCL1_RING_FIELD_DSCP_TID_MAP7		GENMASK(23, 21)
 
-/* REO ring feild mask and offset */
+/* REO ring field mask and offset */
 #define HAL_REO1_RING_BASE_MSB_RING_SIZE		GENMASK(27, 8)
 #define HAL_REO1_RING_BASE_MSB_RING_BASE_ADDR_MSB	GENMASK(7, 0)
 #define HAL_REO1_RING_ID_RING_ID			GENMASK(15, 8)
@@ -738,7 +738,7 @@ struct hal_srng {
 	} u;
 };
 
-/* Interrupt mitigation - Batch threshold in terms of numer of frames */
+/* Interrupt mitigation - Batch threshold in terms of number of frames */
 #define HAL_SRNG_INT_BATCH_THRESHOLD_TX 256
 #define HAL_SRNG_INT_BATCH_THRESHOLD_RX 128
 #define HAL_SRNG_INT_BATCH_THRESHOLD_OTHER 1
@@ -813,7 +813,7 @@ enum hal_rx_buf_return_buf_manager {
 #define HAL_REO_CMD_FLG_UNBLK_RESOURCE		BIT(7)
 #define HAL_REO_CMD_FLG_UNBLK_CACHE		BIT(8)
 
-/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO0_UPD_* feilds */
+/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO0_UPD_* fields */
 #define HAL_REO_CMD_UPD0_RX_QUEUE_NUM		BIT(8)
 #define HAL_REO_CMD_UPD0_VLD			BIT(9)
 #define HAL_REO_CMD_UPD0_ALDC			BIT(10)
@@ -838,7 +838,7 @@ enum hal_rx_buf_return_buf_manager {
 #define HAL_REO_CMD_UPD0_PN_VALID		BIT(29)
 #define HAL_REO_CMD_UPD0_PN			BIT(30)
 
-/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO1_* feilds */
+/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO1_* fields */
 #define HAL_REO_CMD_UPD1_VLD			BIT(16)
 #define HAL_REO_CMD_UPD1_ALDC			GENMASK(18, 17)
 #define HAL_REO_CMD_UPD1_DIS_DUP_DETECTION	BIT(19)
@@ -854,7 +854,7 @@ enum hal_rx_buf_return_buf_manager {
 #define HAL_REO_CMD_UPD1_PN_HANDLE_ENABLE	BIT(30)
 #define HAL_REO_CMD_UPD1_IGNORE_AMPDU_FLG	BIT(31)
 
-/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO2_* feilds */
+/* Should be matching with HAL_REO_UPD_RX_QUEUE_INFO2_* fields */
 #define HAL_REO_CMD_UPD2_SVLD			BIT(10)
 #define HAL_REO_CMD_UPD2_SSN			GENMASK(22, 11)
 #define HAL_REO_CMD_UPD2_SEQ_2K_ERR		BIT(23)
diff --git a/drivers/net/wireless/ath/ath12k/hal_desc.h b/drivers/net/wireless/ath/ath12k/hal_desc.h
index 2250ca2d19a3..6c17adc6d60b 100644
--- a/drivers/net/wireless/ath/ath12k/hal_desc.h
+++ b/drivers/net/wireless/ath/ath12k/hal_desc.h
@@ -706,7 +706,7 @@ struct rx_msdu_desc {
  *
  * msdu_continuation
  *		When set, this MSDU buffer was not able to hold the entire MSDU.
- *		The next buffer will therefor contain additional information
+ *		The next buffer will therefore contain additional information
  *		related to this MSDU.
  *
  * msdu_length
@@ -1294,7 +1294,7 @@ struct hal_tcl_data_cmd {
  *		link descriptor.
  *
  * tcl_cmd_type
- *		used to select the type of TCL Command decriptor
+ *		used to select the type of TCL Command descriptor
  *
  * desc_type
  *		Indicates the type of address provided in the buf_addr_info.
@@ -1408,7 +1408,7 @@ struct hal_tcl_data_cmd {
  * index_loop_override
  *		When set, address search and packet routing is forced to use
  *		'search_index' instead of following the register configuration
- *		seleced by Bank_id.
+ *		selected by Bank_id.
  *
  * ring_id
  *		The buffer pointer ring ID.
@@ -1990,7 +1990,7 @@ struct hal_wbm_release_ring {
  *	Producer: SW/TQM/RXDMA/REO/SWITCH
  *	Consumer: WBM/SW/FW
  *
- * HTT tx status is overlayed on wbm_release ring on 4-byte words 2, 3, 4 and 5
+ * HTT tx status is overlaid on wbm_release ring on 4-byte words 2, 3, 4 and 5
  * for software based completions.
  *
  * buf_addr_info
@@ -2552,7 +2552,7 @@ struct hal_reo_status_hdr {
  *		commands.
  *
  * execution_time (in us)
- *		The amount of time REO took to excecute the command. Note that
+ *		The amount of time REO took to execute the command. Note that
  *		this time does not include the duration of the command waiting
  *		in the command ring, before the execution started.
  *
diff --git a/drivers/net/wireless/ath/ath12k/rx_desc.h b/drivers/net/wireless/ath/ath12k/rx_desc.h
index 5feaff6450ad..f99556a253e5 100644
--- a/drivers/net/wireless/ath/ath12k/rx_desc.h
+++ b/drivers/net/wireless/ath/ath12k/rx_desc.h
@@ -1072,7 +1072,7 @@ struct rx_msdu_end_qcn9274 {
  *
  * l4_offset
  *		Depending upon mode bit, this field either indicates the
- *		L4 offset nin bytes from the start of RX_HEADER (only valid
+ *		L4 offset in bytes from the start of RX_HEADER (only valid
  *		if either ipv4_proto or ipv6_proto is set to 1) or indicates
  *		the offset in bytes to the start of TCP or UDP header from
  *		the start of the IP header after decapsulation (Only valid if
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index f6df14149531..3e6991120e53 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -494,7 +494,7 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 
 	/* tx/rx chainmask reported from fw depends on the actual hw chains used,
 	 * For example, for 4x4 capable macphys, first 4 chains can be used for first
-	 * mac and the remaing 4 chains can be used for the second mac or vice-versa.
+	 * mac and the remaining 4 chains can be used for the second mac or vice-versa.
 	 * In this case, tx/rx chainmask 0xf will be advertised for first mac and 0xf0
 	 * will be advertised for second mac or vice-versa. Compute the shift value
 	 * for tx/rx chainmask which will be used to advertise supported ht/vht rates to
@@ -1743,7 +1743,7 @@ int ath12k_wmi_vdev_install_key(struct ath12k *ar,
 	int ret, len, key_len_aligned;
 
 	/* WMI_TAG_ARRAY_BYTE needs to be aligned with 4, the actual key
-	 * length is specifed in cmd->key_len.
+	 * length is specified in cmd->key_len.
 	 */
 	key_len_aligned = roundup(arg->key_len, 4);
 
@@ -5995,7 +5995,7 @@ static void ath12k_service_available_event(struct ath12k_base *ab, struct sk_buf
 	}
 
 	/* TODO: Use wmi_service_segment_offset information to get the service
-	 * especially when more services are advertised in multiple sevice
+	 * especially when more services are advertised in multiple service
 	 * available events.
 	 */
 	for (i = 0, j = WMI_MAX_SERVICE;
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 84e3fb918e43..08a8c9e0f59f 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -4002,7 +4002,7 @@ struct ath12k_wmi_pdev_radar_event {
 } __packed;
 
 struct wmi_pdev_temperature_event {
-	/* temperature value in Celcius degree */
+	/* temperature value in Celsius degree */
 	a_sle32 temp;
 	__le32 pdev_id;
 } __packed;
@@ -4192,7 +4192,7 @@ enum wmi_sta_ps_param_tx_wake_threshold {
  */
 enum wmi_sta_ps_param_pspoll_count {
 	WMI_STA_PS_PSPOLL_COUNT_NO_MAX = 0,
-	/* Values greater than 0 indicate the maximum numer of PS-Poll frames
+	/* Values greater than 0 indicate the maximum number of PS-Poll frames
 	 * FW will send before waking up.
 	 */
 };
-- 
cgit v1.2.3-70-g09d2


From 731e1b36656a30ffa9c77de673695cdc733d6c92 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 15 Feb 2023 20:31:37 +0200
Subject: wifi: ath12k: dp_mon: Fix unsigned comparison with less than zero

The return value from the call to idr_alloc() is int.
However, the return value is being assigned to an unsigned
int variable 'buf_id', so making 'buf_id' an int.

Eliminate the following warning:
./drivers/net/wireless/ath/ath12k/dp_mon.c:1300:15-21: WARNING: Unsigned expression compared with zero: buf_id < 0

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4060
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230215011453.73466-1-yang.lee@linux.alibaba.com
---
 drivers/net/wireless/ath/ath12k/dp_mon.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_mon.c b/drivers/net/wireless/ath/ath12k/dp_mon.c
index 4f93e4c95fed..85f8f802340d 100644
--- a/drivers/net/wireless/ath/ath12k/dp_mon.c
+++ b/drivers/net/wireless/ath/ath12k/dp_mon.c
@@ -1268,7 +1268,8 @@ int ath12k_dp_mon_buf_replenish(struct ath12k_base *ab,
 	struct sk_buff *skb;
 	struct hal_srng *srng;
 	dma_addr_t paddr;
-	u32 cookie, buf_id;
+	u32 cookie;
+	int buf_id;
 
 	srng = &ab->hal.srng_list[buf_ring->refill_buf_ring.ring_id];
 	spin_lock_bh(&srng->lock);
-- 
cgit v1.2.3-70-g09d2


From df8e8db22c25f3d0f75ee0c79342d4a0841f4757 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 15 Feb 2023 20:31:38 +0200
Subject: wifi: ath12k: dp_mon: clean up some inconsistent indentings

drivers/net/wireless/ath/ath12k/dp_mon.c:532 ath12k_dp_mon_parse_he_sig_su() warn: inconsistent indenting

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4062
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230215012548.90989-1-yang.lee@linux.alibaba.com
---
 drivers/net/wireless/ath/ath12k/dp_mon.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_mon.c b/drivers/net/wireless/ath/ath12k/dp_mon.c
index 85f8f802340d..f1e57e98bdc6 100644
--- a/drivers/net/wireless/ath/ath12k/dp_mon.c
+++ b/drivers/net/wireless/ath/ath12k/dp_mon.c
@@ -529,12 +529,12 @@ static void ath12k_dp_mon_parse_he_sig_su(u8 *tlv_data,
 	case 3:
 		if (he_dcm && he_stbc) {
 			he_gi = HE_GI_0_8;
-					he_ltf = HE_LTF_4_X;
+			he_ltf = HE_LTF_4_X;
 		} else {
 			he_gi = HE_GI_3_2;
 			he_ltf = HE_LTF_4_X;
-			}
-			break;
+		}
+		break;
 	}
 	ppdu_info->gi = he_gi;
 	value = he_gi << HE_GI_SHIFT;
-- 
cgit v1.2.3-70-g09d2


From 3c3ab8c9a29662ef25f8f62b8f118cd9b0eadd88 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 15 Feb 2023 20:31:38 +0200
Subject: wifi: ath10k: Remove the unused function shadow_dst_wr_ind_addr() and
 ath10k_ce_error_intr_enable()

The function shadow_dst_wr_ind_addr() and ath10k_ce_error_intr_enable()
are defined in the ce.c file, the code calling them has been removed,
so remove these unused functions.

Eliminate the following warnings:
drivers/net/wireless/ath/ath10k/ce.c:80:19: warning: unused function 'shadow_dst_wr_ind_addr'
drivers/net/wireless/ath/ath10k/ce.c:441:20: warning: unused function 'ath10k_ce_error_intr_enable'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4063
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230215014058.116775-1-yang.lee@linux.alibaba.com
---
 drivers/net/wireless/ath/ath10k/ce.c | 52 ------------------------------------
 1 file changed, 52 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index c2f3bd35c392..b656cfc03648 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -77,45 +77,6 @@ static inline u32 shadow_sr_wr_ind_addr(struct ath10k *ar,
 	return addr;
 }
 
-static inline u32 shadow_dst_wr_ind_addr(struct ath10k *ar,
-					 struct ath10k_ce_pipe *ce_state)
-{
-	u32 ce_id = ce_state->id;
-	u32 addr = 0;
-
-	switch (ce_id) {
-	case 1:
-		addr = 0x00032034;
-		break;
-	case 2:
-		addr = 0x00032038;
-		break;
-	case 5:
-		addr = 0x00032044;
-		break;
-	case 7:
-		addr = 0x0003204C;
-		break;
-	case 8:
-		addr = 0x00032050;
-		break;
-	case 9:
-		addr = 0x00032054;
-		break;
-	case 10:
-		addr = 0x00032058;
-		break;
-	case 11:
-		addr = 0x0003205C;
-		break;
-	default:
-		ath10k_warn(ar, "invalid CE id: %d", ce_id);
-		break;
-	}
-
-	return addr;
-}
-
 static inline unsigned int
 ath10k_set_ring_byte(unsigned int offset,
 		     struct ath10k_hw_ce_regs_addr_map *addr_map)
@@ -438,19 +399,6 @@ static inline void ath10k_ce_watermark_intr_disable(struct ath10k *ar,
 			  host_ie_addr & ~(wm_regs->wm_mask));
 }
 
-static inline void ath10k_ce_error_intr_enable(struct ath10k *ar,
-					       u32 ce_ctrl_addr)
-{
-	struct ath10k_hw_ce_misc_regs *misc_regs = ar->hw_ce_regs->misc_regs;
-
-	u32 misc_ie_addr = ath10k_ce_read32(ar, ce_ctrl_addr +
-					    ar->hw_ce_regs->misc_ie_addr);
-
-	ath10k_ce_write32(ar,
-			  ce_ctrl_addr + ar->hw_ce_regs->misc_ie_addr,
-			  misc_ie_addr | misc_regs->err_mask);
-}
-
 static inline void ath10k_ce_error_intr_disable(struct ath10k *ar,
 						u32 ce_ctrl_addr)
 {
-- 
cgit v1.2.3-70-g09d2


From bfcc8ba45eb87bfaaff900bbad2b87b204899d41 Mon Sep 17 00:00:00 2001
From: Kees Cook <keescook@chromium.org>
Date: Wed, 15 Feb 2023 20:31:38 +0200
Subject: wifi: ath: Silence memcpy run-time false positive warning

The memcpy() in ath_key_config() was attempting to write across
neighboring struct members in struct ath_keyval. Introduce a wrapping
struct_group, kv_values, to be the addressable target of the memcpy
without overflowing an individual member. Silences the false positive
run-time warning:

  memcpy: detected field-spanning write (size 32) of single field "hk.kv_val" at drivers/net/wireless/ath/key.c:506 (size 16)

Link: https://bbs.archlinux.org/viewtopic.php?id=282254
Cc: Kalle Valo <kvalo@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: linux-wireless@vger.kernel.org
Cc: netdev@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230210054310.never.554-kees@kernel.org
---
 drivers/net/wireless/ath/ath.h | 12 +++++++-----
 drivers/net/wireless/ath/key.c |  2 +-
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath.h b/drivers/net/wireless/ath/ath.h
index f083fb9038c3..f02a308a9ffc 100644
--- a/drivers/net/wireless/ath/ath.h
+++ b/drivers/net/wireless/ath/ath.h
@@ -96,11 +96,13 @@ struct ath_keyval {
 	u8 kv_type;
 	u8 kv_pad;
 	u16 kv_len;
-	u8 kv_val[16]; /* TK */
-	u8 kv_mic[8]; /* Michael MIC key */
-	u8 kv_txmic[8]; /* Michael MIC TX key (used only if the hardware
-			 * supports both MIC keys in the same key cache entry;
-			 * in that case, kv_mic is the RX key) */
+	struct_group(kv_values,
+		u8 kv_val[16]; /* TK */
+		u8 kv_mic[8]; /* Michael MIC key */
+		u8 kv_txmic[8]; /* Michael MIC TX key (used only if the hardware
+				 * supports both MIC keys in the same key cache entry;
+				 * in that case, kv_mic is the RX key) */
+	);
 };
 
 enum ath_cipher {
diff --git a/drivers/net/wireless/ath/key.c b/drivers/net/wireless/ath/key.c
index 61b59a804e30..b7b61d4f02ba 100644
--- a/drivers/net/wireless/ath/key.c
+++ b/drivers/net/wireless/ath/key.c
@@ -503,7 +503,7 @@ int ath_key_config(struct ath_common *common,
 
 	hk.kv_len = key->keylen;
 	if (key->keylen)
-		memcpy(hk.kv_val, key->key, key->keylen);
+		memcpy(&hk.kv_values, key->key, key->keylen);
 
 	if (!(key->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {
 		switch (vif->type) {
-- 
cgit v1.2.3-70-g09d2


From 8c464d16809fa02982f6341ea598ec5d07457f19 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Thu, 16 Feb 2023 15:16:14 +0300
Subject: wifi: ath12k: use kfree_skb() instead of kfree()

Sk_buffs are supposed to be freed with kfree_skb().

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/Y+4ejiYakhEvEw7c@kili
---
 drivers/net/wireless/ath/ath12k/dp_tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index 95294f35155c..fd8d850f9818 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -270,7 +270,7 @@ tcl_ring_sel:
 					  skb_ext_desc->len, DMA_TO_DEVICE);
 		ret = dma_mapping_error(ab->dev, ti.paddr);
 		if (ret) {
-			kfree(skb_ext_desc);
+			kfree_skb(skb_ext_desc);
 			goto fail_unmap_dma;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 342fcde9d91460f01f65707e16368a1571271a3a Mon Sep 17 00:00:00 2001
From: Yang Yingliang <yangyingliang@huawei.com>
Date: Fri, 17 Feb 2023 11:00:31 +0800
Subject: wifi: ath11k: fix return value check in ath11k_ahb_probe()

ioremap() returns NULL pointer not PTR_ERR() when it fails,
so replace the IS_ERR() check with NULL pointer check.

Fixes: b42b3678c91f ("wifi: ath11k: remap ce register space for IPQ5018")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230217030031.4021289-1-yangyingliang@huawei.com
---
 drivers/net/wireless/ath/ath11k/ahb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
index cd48eca494ed..022caacd2822 100644
--- a/drivers/net/wireless/ath/ath11k/ahb.c
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -1174,7 +1174,7 @@ static int ath11k_ahb_probe(struct platform_device *pdev)
 		 * to a new space for accessing them.
 		 */
 		ab->mem_ce = ioremap(ce_remap->base, ce_remap->size);
-		if (IS_ERR(ab->mem_ce)) {
+		if (!ab->mem_ce) {
 			dev_err(&pdev->dev, "ce ioremap error\n");
 			ret = -ENOMEM;
 			goto err_core_free;
-- 
cgit v1.2.3-70-g09d2


From 7654cc03eb699297130b693ec34e25f77b17c947 Mon Sep 17 00:00:00 2001
From: Fedor Pchelkin <pchelkin@ispras.ru>
Date: Thu, 16 Feb 2023 22:23:01 +0300
Subject: wifi: ath9k: hif_usb: fix memory leak of remain_skbs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

hif_dev->remain_skb is allocated and used exclusively in
ath9k_hif_usb_rx_stream(). It is implied that an allocated remain_skb is
processed and subsequently freed (in error paths) only during the next
call of ath9k_hif_usb_rx_stream().

So, if the urbs are deallocated between those two calls due to the device
deinitialization or suspend, it is possible that ath9k_hif_usb_rx_stream()
is not called next time and the allocated remain_skb is leaked. Our local
Syzkaller instance was able to trigger that.

remain_skb makes sense when receiving two consecutive urbs which are
logically linked together, i.e. a specific data field from the first skb
indicates a cached skb to be allocated, memcpy'd with some data and
subsequently processed in the next call to ath9k_hif_usb_rx_stream(). Urbs
deallocation supposedly makes that link irrelevant so we need to free the
cached skb in those cases.

Fix the leak by introducing a function to explicitly free remain_skb (if
it is not NULL) when the rx urbs have been deallocated. remain_skb is NULL
when it has not been allocated at all (hif_dev struct is kzalloced) or
when it has been processed in next call to ath9k_hif_usb_rx_stream().

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes: fb9987d0f748 ("ath9k_htc: Support for AR9271 chipset.")
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230216192301.171225-1-pchelkin@ispras.ru
---
 drivers/net/wireless/ath/ath9k/hif_usb.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index f521dfa2f194..e0130beb304d 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -534,6 +534,24 @@ static struct ath9k_htc_hif hif_usb = {
 	.send = hif_usb_send,
 };
 
+/* Need to free remain_skb allocated in ath9k_hif_usb_rx_stream
+ * in case ath9k_hif_usb_rx_stream wasn't called next time to
+ * process the buffer and subsequently free it.
+ */
+static void ath9k_hif_usb_free_rx_remain_skb(struct hif_device_usb *hif_dev)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&hif_dev->rx_lock, flags);
+	if (hif_dev->remain_skb) {
+		dev_kfree_skb_any(hif_dev->remain_skb);
+		hif_dev->remain_skb = NULL;
+		hif_dev->rx_remain_len = 0;
+		RX_STAT_INC(hif_dev, skb_dropped);
+	}
+	spin_unlock_irqrestore(&hif_dev->rx_lock, flags);
+}
+
 static void ath9k_hif_usb_rx_stream(struct hif_device_usb *hif_dev,
 				    struct sk_buff *skb)
 {
@@ -868,6 +886,7 @@ err:
 static void ath9k_hif_usb_dealloc_rx_urbs(struct hif_device_usb *hif_dev)
 {
 	usb_kill_anchored_urbs(&hif_dev->rx_submitted);
+	ath9k_hif_usb_free_rx_remain_skb(hif_dev);
 }
 
 static int ath9k_hif_usb_alloc_rx_urbs(struct hif_device_usb *hif_dev)
-- 
cgit v1.2.3-70-g09d2


From 95a389e2ff3212d866cc51c77d682d2934074eb8 Mon Sep 17 00:00:00 2001
From: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Date: Sun, 22 Jan 2023 07:19:36 +0530
Subject: wifi: ath12k: Handle lock during peer_id find

ath12k_peer_find_by_id() requires that the caller hold the
ab->base_lock. Currently the WBM error path does not hold
the lock and calling that function, leads to the
following lockdep_assert()in QCN9274:

[105162.160893] ------------[ cut here ]------------
[105162.160916] WARNING: CPU: 3 PID: 0 at drivers/net/wireless/ath/ath12k/peer.c:71 ath12k_peer_find_by_id+0x52/0x60 [ath12k]
[105162.160933] Modules linked in: ath12k(O) qrtr_mhi qrtr mac80211 cfg80211 mhi qmi_helpers libarc4 nvme nvme_core [last unloaded: ath12k(O)]
[105162.160967] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G        W  O       6.1.0-rc2+ #3
[105162.160972] Hardware name: Intel(R) Client Systems NUC8i7HVK/NUC8i7HVB, BIOS HNKBLi70.86A.0056.2019.0506.1527 05/06/2019
[105162.160977] RIP: 0010:ath12k_peer_find_by_id+0x52/0x60 [ath12k]
[105162.160990] Code: 07 eb 0f 39 68 24 74 0a 48 8b 00 48 39 f8 75 f3 31 c0 5b 5d c3 48 8d bf b0 f2 00 00 be ff ff ff ff e8 22 20 c4 e2 85 c0 75 bf <0f> 0b eb bb 66 2e 0f 1f 84 00 00 00 00 00 41 54 4c 8d a7 98 f2 00
[105162.160996] RSP: 0018:ffffa223001acc60 EFLAGS: 00010246
[105162.161003] RAX: 0000000000000000 RBX: ffff9f0573940000 RCX: 0000000000000000
[105162.161008] RDX: 0000000000000001 RSI: ffffffffa3951c8e RDI: ffffffffa39a96d7
[105162.161013] RBP: 000000000000000a R08: 0000000000000000 R09: 0000000000000000
[105162.161017] R10: ffffa223001acb40 R11: ffffffffa3d57c60 R12: ffff9f057394f2e0
[105162.161022] R13: ffff9f0573940000 R14: ffff9f04ecd659c0 R15: ffff9f04d5a9b040
[105162.161026] FS:  0000000000000000(0000) GS:ffff9f0575600000(0000) knlGS:0000000000000000
[105162.161031] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[105162.161036] CR2: 00001d5c8277a008 CR3: 00000001e6224006 CR4: 00000000003706e0
[105162.161041] Call Trace:
[105162.161046]  <IRQ>
[105162.161051]  ath12k_dp_rx_process_wbm_err+0x6da/0xaf0 [ath12k]
[105162.161072]  ? ath12k_dp_rx_process_err+0x80e/0x15a0 [ath12k]
[105162.161084]  ? __lock_acquire+0x4ca/0x1a60
[105162.161104]  ath12k_dp_service_srng+0x263/0x310 [ath12k]
[105162.161120]  ath12k_pci_ext_grp_napi_poll+0x1c/0x70 [ath12k]
[105162.161133]  __napi_poll+0x22/0x260
[105162.161141]  net_rx_action+0x2f8/0x380
[105162.161153]  __do_softirq+0xd0/0x4c9
[105162.161162]  irq_exit_rcu+0x88/0xe0
[105162.161169]  common_interrupt+0xa5/0xc0
[105162.161174]  </IRQ>
[105162.161179]  <TASK>
[105162.161184]  asm_common_interrupt+0x22/0x40

Handle spin lock/unlock in WBM error path to hold the necessary lock
expected by ath12k_peer_find_by_id().

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0-03171-QCAHKSWPL_SILICONZ-1

Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230122014936.3594-1-quic_rgnanase@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index eb67b3409f85..0adcbcfa0db5 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -3494,11 +3494,14 @@ static int ath12k_dp_rx_h_null_q_desc(struct ath12k *ar, struct sk_buff *msdu,
 	msdu_len = ath12k_dp_rx_h_msdu_len(ab, desc);
 	peer_id = ath12k_dp_rx_h_peer_id(ab, desc);
 
+	spin_lock(&ab->base_lock);
 	if (!ath12k_peer_find_by_id(ab, peer_id)) {
+		spin_unlock(&ab->base_lock);
 		ath12k_dbg(ab, ATH12K_DBG_DATA, "invalid peer id received in wbm err pkt%d\n",
 			   peer_id);
 		return -EINVAL;
 	}
+	spin_unlock(&ab->base_lock);
 
 	if (!rxcb->is_frag && ((msdu_len + hal_rx_desc_sz) > DP_RX_BUFFER_SIZE)) {
 		/* First buffer will be freed by the caller, so deduct it's length */
-- 
cgit v1.2.3-70-g09d2


From 80e396586d0a94c42015dd9472176d89a3b0e4ca Mon Sep 17 00:00:00 2001
From: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Date: Mon, 23 Jan 2023 15:21:41 +0530
Subject: wifi: ath12k: PCI ops for wakeup/release MHI

Wakeup/release MHI is not needed before pci_read/write for QCN9274.
Since wakeup & release MHI is enabled for all QCN9274 and
WCN7850, below MHI assert is seen in QCN9274

[  784.906613] BUG: sleeping function called from invalid context at drivers/bus/mhi/host/pm.c:989
[  784.906633] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 0, name: swapper/3
[  784.906637] preempt_count: 503, expected: 0
[  784.906641] RCU nest depth: 0, expected: 0
[  784.906644] 2 locks held by swapper/3/0:
[  784.906646]  #0: ffff8ed348e429e0 (&ab->ce.ce_lock){+.-.}-{2:2}, at: ath12k_ce_recv_process_cb+0xb3/0x2f0 [ath12k]
[  784.906664]  #1: ffff8ed348e491f0 (&srng->lock_key#3){+.-.}-{2:2}, at: ath12k_ce_recv_process_cb+0xfb/0x2f0 [ath12k]
[  784.906678] Preemption disabled at:
[  784.906680] [<0000000000000000>] 0x0
[  784.906686] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G        W  O       6.1.0-rc2+ #3
[  784.906688] Hardware name: Intel(R) Client Systems NUC8i7HVK/NUC8i7HVB, BIOS HNKBLi70.86A.0056.2019.0506.1527 05/06/2019
[  784.906690] Call Trace:
[  784.906691]  <IRQ>
[  784.906693]  dump_stack_lvl+0x56/0x7b
[  784.906698]  __might_resched+0x21c/0x270
[  784.906704]  __mhi_device_get_sync+0x7d/0x1c0 [mhi]
[  784.906714]  mhi_device_get_sync+0xd/0x20 [mhi]
[  784.906719]  ath12k_pci_write32+0x75/0x170 [ath12k]
[  784.906729]  ath12k_hal_srng_access_end+0x55/0xc0 [ath12k]
[  784.906737]  ath12k_ce_recv_process_cb+0x1f3/0x2f0 [ath12k]
[  784.906776]  ? ath12k_pci_ce_tasklet+0x11/0x30 [ath12k]
[  784.906788]  ath12k_pci_ce_tasklet+0x11/0x30 [ath12k]
[  784.906813]  tasklet_action_common.isra.18+0xb7/0xe0
[  784.906820]  __do_softirq+0xd0/0x4c9
[  784.906826]  irq_exit_rcu+0x88/0xe0
[  784.906828]  common_interrupt+0xa5/0xc0
[  784.906831]  </IRQ>
[  784.906832]  <TASK>

Adding function callbacks for MHI wakeup and release operations.
QCN9274 does not need wakeup/release, function callbacks are initialized
to NULL. In case of WCN7850, shadow registers are used to access rings.
Since, shadow register's offset is less than ACCESS_ALWAYS_OFF,
mhi_device_get_sync() or mhi_device_put() to wakeup
and release mhi will not be called during service ring accesses.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0-03171-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230123095141.5310-1-quic_rgnanase@quicinc.com
---
 drivers/net/wireless/ath/ath12k/pci.c | 47 ++++++++++++++++++++++++++++-------
 drivers/net/wireless/ath/ath12k/pci.h |  6 +++++
 2 files changed, 44 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c
index ae7f6083c9fc..d32637b0113d 100644
--- a/drivers/net/wireless/ath/ath12k/pci.c
+++ b/drivers/net/wireless/ath/ath12k/pci.c
@@ -119,6 +119,30 @@ static const char *irq_name[ATH12K_IRQ_NUM_MAX] = {
 	"tcl2host-status-ring",
 };
 
+static int ath12k_pci_bus_wake_up(struct ath12k_base *ab)
+{
+	struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
+
+	return mhi_device_get_sync(ab_pci->mhi_ctrl->mhi_dev);
+}
+
+static void ath12k_pci_bus_release(struct ath12k_base *ab)
+{
+	struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
+
+	mhi_device_put(ab_pci->mhi_ctrl->mhi_dev);
+}
+
+static const struct ath12k_pci_ops ath12k_pci_ops_qcn9274 = {
+	.wakeup = NULL,
+	.release = NULL,
+};
+
+static const struct ath12k_pci_ops ath12k_pci_ops_wcn7850 = {
+	.wakeup = ath12k_pci_bus_wake_up,
+	.release = ath12k_pci_bus_release,
+};
+
 static void ath12k_pci_select_window(struct ath12k_pci *ab_pci, u32 offset)
 {
 	struct ath12k_base *ab = ab_pci->ab;
@@ -989,13 +1013,14 @@ u32 ath12k_pci_read32(struct ath12k_base *ab, u32 offset)
 {
 	struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
 	u32 val, window_start;
+	int ret = 0;
 
 	/* for offset beyond BAR + 4K - 32, may
 	 * need to wakeup MHI to access.
 	 */
 	if (test_bit(ATH12K_PCI_FLAG_INIT_DONE, &ab_pci->flags) &&
-	    offset >= ACCESS_ALWAYS_OFF)
-		mhi_device_get_sync(ab_pci->mhi_ctrl->mhi_dev);
+	    offset >= ACCESS_ALWAYS_OFF && ab_pci->pci_ops->wakeup)
+		ret = ab_pci->pci_ops->wakeup(ab);
 
 	if (offset < WINDOW_START) {
 		val = ioread32(ab->mem + offset);
@@ -1023,9 +1048,9 @@ u32 ath12k_pci_read32(struct ath12k_base *ab, u32 offset)
 	}
 
 	if (test_bit(ATH12K_PCI_FLAG_INIT_DONE, &ab_pci->flags) &&
-	    offset >= ACCESS_ALWAYS_OFF)
-		mhi_device_put(ab_pci->mhi_ctrl->mhi_dev);
-
+	    offset >= ACCESS_ALWAYS_OFF && ab_pci->pci_ops->release &&
+	    !ret)
+		ab_pci->pci_ops->release(ab);
 	return val;
 }
 
@@ -1033,13 +1058,14 @@ void ath12k_pci_write32(struct ath12k_base *ab, u32 offset, u32 value)
 {
 	struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
 	u32 window_start;
+	int ret = 0;
 
 	/* for offset beyond BAR + 4K - 32, may
 	 * need to wakeup MHI to access.
 	 */
 	if (test_bit(ATH12K_PCI_FLAG_INIT_DONE, &ab_pci->flags) &&
-	    offset >= ACCESS_ALWAYS_OFF)
-		mhi_device_get_sync(ab_pci->mhi_ctrl->mhi_dev);
+	    offset >= ACCESS_ALWAYS_OFF && ab_pci->pci_ops->wakeup)
+		ret = ab_pci->pci_ops->wakeup(ab);
 
 	if (offset < WINDOW_START) {
 		iowrite32(value, ab->mem + offset);
@@ -1067,8 +1093,9 @@ void ath12k_pci_write32(struct ath12k_base *ab, u32 offset, u32 value)
 	}
 
 	if (test_bit(ATH12K_PCI_FLAG_INIT_DONE, &ab_pci->flags) &&
-	    offset >= ACCESS_ALWAYS_OFF)
-		mhi_device_put(ab_pci->mhi_ctrl->mhi_dev);
+	    offset >= ACCESS_ALWAYS_OFF && ab_pci->pci_ops->release &&
+	    !ret)
+		ab_pci->pci_ops->release(ab);
 }
 
 int ath12k_pci_power_up(struct ath12k_base *ab)
@@ -1182,6 +1209,7 @@ static int ath12k_pci_probe(struct pci_dev *pdev,
 	case QCN9274_DEVICE_ID:
 		ab_pci->msi_config = &ath12k_msi_config[0];
 		ab->static_window_map = true;
+		ab_pci->pci_ops = &ath12k_pci_ops_qcn9274;
 		ath12k_pci_read_hw_version(ab, &soc_hw_version_major,
 					   &soc_hw_version_minor);
 		switch (soc_hw_version_major) {
@@ -1202,6 +1230,7 @@ static int ath12k_pci_probe(struct pci_dev *pdev,
 		ab_pci->msi_config = &ath12k_msi_config[0];
 		ab->static_window_map = false;
 		ab->hw_rev = ATH12K_HW_WCN7850_HW20;
+		ab_pci->pci_ops = &ath12k_pci_ops_wcn7850;
 		break;
 
 	default:
diff --git a/drivers/net/wireless/ath/ath12k/pci.h b/drivers/net/wireless/ath/ath12k/pci.h
index 0d9e40ab31f2..0f24fd9395cd 100644
--- a/drivers/net/wireless/ath/ath12k/pci.h
+++ b/drivers/net/wireless/ath/ath12k/pci.h
@@ -86,6 +86,11 @@ enum ath12k_pci_flags {
 	ATH12K_PCI_ASPM_RESTORE,
 };
 
+struct ath12k_pci_ops {
+	int (*wakeup)(struct ath12k_base *ab);
+	void (*release)(struct ath12k_base *ab);
+};
+
 struct ath12k_pci {
 	struct pci_dev *pdev;
 	struct ath12k_base *ab;
@@ -103,6 +108,7 @@ struct ath12k_pci {
 	/* enum ath12k_pci_flags */
 	unsigned long flags;
 	u16 link_ctl;
+	const struct ath12k_pci_ops *pci_ops;
 };
 
 static inline struct ath12k_pci *ath12k_pci_priv(struct ath12k_base *ab)
-- 
cgit v1.2.3-70-g09d2


From f117276638b7600b981b3fe28550823cfbe1ef23 Mon Sep 17 00:00:00 2001
From: Douglas Anderson <dianders@chromium.org>
Date: Wed, 1 Feb 2023 08:54:42 -0800
Subject: wifi: ath11k: Use platform_get_irq() to get the interrupt

As of commit a1a2b7125e10 ("of/platform: Drop static setup of IRQ
resource from DT core"), we need to use platform_get_irq() instead of
platform_get_resource() to get our IRQs because
platform_get_resource() simply won't get them anymore.

This was already fixed in several other Atheros WiFi drivers,
apparently in response to Zeal Robot reports. An example of another
fix is commit 9503a1fc123d ("ath9k: Use platform_get_irq() to get the
interrupt"). ath11k seems to have been missed in this effort, though.

Without this change, WiFi wasn't coming up on my Qualcomm sc7280-based
hardware. Specifically, "platform_get_resource(pdev, IORESOURCE_IRQ,
i)" was failing even for i=0.

Tested-on: WCN6750 hw1.0 AHB WLAN.MSL.1.0.1-00887-QCAMSLSWPLZ-1

Fixes: a1a2b7125e10 ("of/platform: Drop static setup of IRQ resource from DT core")
Fixes: 00402f49d26f ("ath11k: Add support for WCN6750 device")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Jun Yu <junyuu@chromium.org>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230201084131.v2.1.I69cf3d56c97098287fe3a70084ee515098390b70@changeid
---
 drivers/net/wireless/ath/ath11k/ahb.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
index 022caacd2822..6b6059d5dc38 100644
--- a/drivers/net/wireless/ath/ath11k/ahb.c
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -874,11 +874,11 @@ static int ath11k_ahb_setup_msi_resources(struct ath11k_base *ab)
 	ab->pci.msi.ep_base_data = int_prop + 32;
 
 	for (i = 0; i < ab->pci.msi.config->total_vectors; i++) {
-		res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
-		if (!res)
-			return -ENODEV;
+		ret = platform_get_irq(pdev, i);
+		if (ret < 0)
+			return ret;
 
-		ab->pci.msi.irqs[i] = res->start;
+		ab->pci.msi.irqs[i] = ret;
 	}
 
 	set_bit(ATH11K_FLAG_MULTI_MSI_VECTORS, &ab->dev_flags);
-- 
cgit v1.2.3-70-g09d2


From 95c95251d0547b46d6571e4fbd51b42865c15a4a Mon Sep 17 00:00:00 2001
From: Douglas Anderson <dianders@chromium.org>
Date: Wed, 1 Feb 2023 08:54:43 -0800
Subject: wifi: ath5k: Use platform_get_irq() to get the interrupt

As of commit a1a2b7125e10 ("of/platform: Drop static setup of IRQ
resource from DT core"), we need to use platform_get_irq() instead of
platform_get_resource() to get our IRQs because
platform_get_resource() simply won't get them anymore.

This was already fixed in several other Atheros WiFi drivers,
apparently in response to Zeal Robot reports. An example of another
fix is commit 9503a1fc123d ("ath9k: Use platform_get_irq() to get the
interrupt"). ath5k seems to have been missed in this effort, though.

Fixes: a1a2b7125e10 ("of/platform: Drop static setup of IRQ resource from DT core")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230201084131.v2.2.Ic4f8542b0588d7eb4bc6e322d4af3d2064e84ff0@changeid
---
 drivers/net/wireless/ath/ath5k/ahb.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath5k/ahb.c b/drivers/net/wireless/ath/ath5k/ahb.c
index 2c9cec8b53d9..28a1e5eff204 100644
--- a/drivers/net/wireless/ath/ath5k/ahb.c
+++ b/drivers/net/wireless/ath/ath5k/ahb.c
@@ -113,15 +113,13 @@ static int ath_ahb_probe(struct platform_device *pdev)
 		goto err_out;
 	}
 
-	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
-	if (res == NULL) {
-		dev_err(&pdev->dev, "no IRQ resource found\n");
-		ret = -ENXIO;
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(&pdev->dev, "no IRQ resource found: %d\n", irq);
+		ret = irq;
 		goto err_iounmap;
 	}
 
-	irq = res->start;
-
 	hw = ieee80211_alloc_hw(sizeof(struct ath5k_hw), &ath5k_hw_ops);
 	if (hw == NULL) {
 		dev_err(&pdev->dev, "no memory for ieee80211_hw\n");
-- 
cgit v1.2.3-70-g09d2


From 4c856ee12df85aabd437c3836ed9f68d94268358 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Mon, 6 Feb 2023 16:15:48 +0300
Subject: wifi: ath5k: fix an off by one check in ath5k_eeprom_read_freq_list()

This loop checks that i < max at the start of loop but then it does
i++ which could put it past the end of the array.  It's harmless to
check again and prevent a potential out of bounds.

Fixes: 1048643ea94d ("ath5k: Clean up eeprom parsing and add missing calibration data")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/Y+D9hPQrHfWBJhXz@kili
---
 drivers/net/wireless/ath/ath5k/eeprom.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath5k/eeprom.c b/drivers/net/wireless/ath/ath5k/eeprom.c
index d444b3d70ba2..58d3e86f6256 100644
--- a/drivers/net/wireless/ath/ath5k/eeprom.c
+++ b/drivers/net/wireless/ath/ath5k/eeprom.c
@@ -529,7 +529,7 @@ ath5k_eeprom_read_freq_list(struct ath5k_hw *ah, int *offset, int max,
 		ee->ee_n_piers[mode]++;
 
 		freq2 = (val >> 8) & 0xff;
-		if (!freq2)
+		if (!freq2 || i >= max)
 			break;
 
 		pc[i++].freq = ath5k_eeprom_bin2freq(ee,
-- 
cgit v1.2.3-70-g09d2


From 8c68fe00344c01bfba95b9f978af0aa236a821f7 Mon Sep 17 00:00:00 2001
From: Abhishek Kumar <kuabhs@chromium.org>
Date: Fri, 3 Feb 2023 00:01:40 +0000
Subject: wifi: ath10k: snoc: enable threaded napi on WCN3990

NAPI poll can be done in threaded context along with soft irq
context. Threaded context can be scheduled efficiently, thus
creating less of bottleneck during Rx processing. This patch is
to enable threaded NAPI on ath10k driver.

Based on testing, it was observed that on WCN3990, the CPU0 reaches
100% utilization when napi runs in softirq context. At the same
time the other CPUs are at low consumption percentage. This
does not allow device to reach its maximum throughput potential.
After enabling threaded napi, CPU load is balanced across all CPUs
and following improvments were observed:
- UDP_RX increase by ~22-25%
- TCP_RX increase by ~15%

Here are some of the additional raw data with and without threaded napi:
==================================================
udp_rx(Without threaded NAPI)
435.98+-5.16 : Channel 44
439.06+-0.66 : Channel 157

udp_rx(With threaded NAPI)
509.73+-41.03 : Channel 44
549.97+-7.62 : Channel 157
===================================================
udp_tx(Without threaded NAPI)
461.31+-0.69  : Channel 44
461.46+-0.78 : Channel 157

udp_tx(With threaded NAPI)
459.20+-0.77 : Channel 44
459.78+-1.08 : Channel 157
===================================================
tcp_rx(Without threaded NAPI)
472.63+-2.35 : Channel 44
469.29+-6.31 : Channel 157

tcp_rx(With threaded NAPI)
498.49+-2.44 : Channel 44
541.14+-40.65 : Channel 157
===================================================
tcp_tx(Without threaded NAPI)
317.34+-2.37 : Channel 44
317.01+-2.56 : Channel 157

tcp_tx(With threaded NAPI)
371.34+-2.36 : Channel 44
376.95+-9.40 : Channel 157
===================================================

Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.3.2.2-00696-QCAHLSWMTPL-1
Signed-off-by: Abhishek Kumar <kuabhs@chromium.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230203000116.v2.1.I5bb9c164a2d2025655dee810b983e01ecd81c14e@changeid
---
 drivers/net/wireless/ath/ath10k/snoc.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/ath/ath10k/snoc.c b/drivers/net/wireless/ath/ath10k/snoc.c
index cfcb759a87de..0f6d2f67ff6b 100644
--- a/drivers/net/wireless/ath/ath10k/snoc.c
+++ b/drivers/net/wireless/ath/ath10k/snoc.c
@@ -927,6 +927,7 @@ static int ath10k_snoc_hif_start(struct ath10k *ar)
 
 	bitmap_clear(ar_snoc->pending_ce_irqs, 0, CE_COUNT_MAX);
 
+	dev_set_threaded(&ar->napi_dev, true);
 	ath10k_core_napi_enable(ar);
 	ath10k_snoc_irq_enable(ar);
 	ath10k_snoc_rx_post(ar);
-- 
cgit v1.2.3-70-g09d2


From 7c4c511f74ba65a18f0beeb9b2fc0e34f28ee79d Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Tue, 7 Feb 2023 13:24:10 +0800
Subject: wifi: ath10k: Remove redundant assignment to changed_flags

Variable changed_flags is assigned, but is not effectively used, so
delete it.

drivers/net/wireless/ath/ath10k/mac.c:6024:22: warning: parameter 'changed_flags' set but not used.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=3963
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230207052410.26337-1-jiapeng.chong@linux.alibaba.com
---
 drivers/net/wireless/ath/ath10k/mac.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index ec8d5b29bc72..7675858f069b 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -6030,7 +6030,6 @@ static void ath10k_configure_filter(struct ieee80211_hw *hw,
 
 	mutex_lock(&ar->conf_mutex);
 
-	changed_flags &= SUPPORTED_FILTERS;
 	*total_flags &= SUPPORTED_FILTERS;
 	ar->filter_flags = *total_flags;
 
-- 
cgit v1.2.3-70-g09d2


From 60b7d62ba8cdbd073997bff0f1cdae8d844002c0 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Thu, 9 Feb 2023 23:26:22 +0100
Subject: wifi: ath11k: fix SAC bug on peer addition with sta band migration

Fix sleep in atomic context warning detected by Smatch static checker
analyzer.

Following the locking pattern for peer_rhash_add lock tbl_mtx_lock mutex
always even if sta is not transitioning to another band.
This is peer_add function and a more secure locking should not cause
performance regression.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.5.0.1-01208-QCAHKSWPL_SILICONZ-1

Fixes: d673cb6fe6c0 ("wifi: ath11k: fix peer addition/deletion error on sta band migration")
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230209222622.1751-1-ansuelsmth@gmail.com
---
 drivers/net/wireless/ath/ath11k/peer.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/peer.c b/drivers/net/wireless/ath/ath11k/peer.c
index 1ae7af02c364..1380811827a8 100644
--- a/drivers/net/wireless/ath/ath11k/peer.c
+++ b/drivers/net/wireless/ath/ath11k/peer.c
@@ -382,22 +382,23 @@ int ath11k_peer_create(struct ath11k *ar, struct ath11k_vif *arvif,
 		return -ENOBUFS;
 	}
 
+	mutex_lock(&ar->ab->tbl_mtx_lock);
 	spin_lock_bh(&ar->ab->base_lock);
 	peer = ath11k_peer_find_by_addr(ar->ab, param->peer_addr);
 	if (peer) {
 		if (peer->vdev_id == param->vdev_id) {
 			spin_unlock_bh(&ar->ab->base_lock);
+			mutex_unlock(&ar->ab->tbl_mtx_lock);
 			return -EINVAL;
 		}
 
 		/* Assume sta is transitioning to another band.
 		 * Remove here the peer from rhash.
 		 */
-		mutex_lock(&ar->ab->tbl_mtx_lock);
 		ath11k_peer_rhash_delete(ar->ab, peer);
-		mutex_unlock(&ar->ab->tbl_mtx_lock);
 	}
 	spin_unlock_bh(&ar->ab->base_lock);
+	mutex_unlock(&ar->ab->tbl_mtx_lock);
 
 	ret = ath11k_wmi_send_peer_create_cmd(ar, param);
 	if (ret) {
-- 
cgit v1.2.3-70-g09d2


From 59e6ded57cc1b3b5117cef66dabe6d7977208ba0 Mon Sep 17 00:00:00 2001
From: Lu jicong <jiconglu58@gmail.com>
Date: Tue, 14 Feb 2023 06:36:02 +0000
Subject: wifi: rtlwifi: rtl8192ce: fix dealing empty EEPROM values

On OpenWRT platform, RTL8192CE could be soldered on board, but not standard PCI
module. In this case, some EEPROM values aren't programmed and left 0xff.
Load default values when the EEPROM values are empty to avoid problems.

Signed-off-by: Lu jicong <jiconglu58@gmail.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214063602.2257263-1-jiconglu58@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192ce/hw.c    | 25 +++++++++++++++-------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c
index b9c62640d2cb..dc480323c9cb 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c
@@ -1428,7 +1428,9 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 
 	for (rf_path = 0; rf_path < 2; rf_path++) {
 		for (i = 0; i < 3; i++) {
-			if (!autoload_fail) {
+			if (!autoload_fail &&
+			    hwinfo[EEPROM_TXPOWERCCK + rf_path * 3 + i] != 0xff &&
+			    hwinfo[EEPROM_TXPOWERHT40_1S + rf_path * 3 + i] != 0xff) {
 				rtlefuse->
 				    eeprom_chnlarea_txpwr_cck[rf_path][i] =
 				    hwinfo[EEPROM_TXPOWERCCK + rf_path * 3 + i];
@@ -1448,7 +1450,8 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 	}
 
 	for (i = 0; i < 3; i++) {
-		if (!autoload_fail)
+		if (!autoload_fail &&
+		    hwinfo[EEPROM_TXPOWERHT40_2SDIFF + i] != 0xff)
 			tempval = hwinfo[EEPROM_TXPOWERHT40_2SDIFF + i];
 		else
 			tempval = EEPROM_DEFAULT_HT40_2SDIFF;
@@ -1518,7 +1521,9 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 	}
 
 	for (i = 0; i < 3; i++) {
-		if (!autoload_fail) {
+		if (!autoload_fail &&
+		    hwinfo[EEPROM_TXPWR_GROUP + i] != 0xff &&
+		    hwinfo[EEPROM_TXPWR_GROUP + 3 + i] != 0xff) {
 			rtlefuse->eeprom_pwrlimit_ht40[i] =
 			    hwinfo[EEPROM_TXPWR_GROUP + i];
 			rtlefuse->eeprom_pwrlimit_ht20[i] =
@@ -1563,7 +1568,8 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 	for (i = 0; i < 14; i++) {
 		index = rtl92c_get_chnl_group((u8)i);
 
-		if (!autoload_fail)
+		if (!autoload_fail &&
+		    hwinfo[EEPROM_TXPOWERHT20DIFF + index] != 0xff)
 			tempval = hwinfo[EEPROM_TXPOWERHT20DIFF + index];
 		else
 			tempval = EEPROM_DEFAULT_HT20_DIFF;
@@ -1580,7 +1586,8 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 
 		index = rtl92c_get_chnl_group((u8)i);
 
-		if (!autoload_fail)
+		if (!autoload_fail &&
+		    hwinfo[EEPROM_TXPOWER_OFDMDIFF + index] != 0xff)
 			tempval = hwinfo[EEPROM_TXPOWER_OFDMDIFF + index];
 		else
 			tempval = EEPROM_DEFAULT_LEGACYHTTXPOWERDIFF;
@@ -1610,14 +1617,16 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 			"RF-B Legacy to HT40 Diff[%d] = 0x%x\n",
 			i, rtlefuse->txpwr_legacyhtdiff[RF90_PATH_B][i]);
 
-	if (!autoload_fail)
+	if (!autoload_fail && hwinfo[RF_OPTION1] != 0xff)
 		rtlefuse->eeprom_regulatory = (hwinfo[RF_OPTION1] & 0x7);
 	else
 		rtlefuse->eeprom_regulatory = 0;
 	RTPRINT(rtlpriv, FINIT, INIT_TXPOWER,
 		"eeprom_regulatory = 0x%x\n", rtlefuse->eeprom_regulatory);
 
-	if (!autoload_fail) {
+	if (!autoload_fail &&
+	    hwinfo[EEPROM_TSSI_A] != 0xff &&
+	    hwinfo[EEPROM_TSSI_B] != 0xff) {
 		rtlefuse->eeprom_tssi[RF90_PATH_A] = hwinfo[EEPROM_TSSI_A];
 		rtlefuse->eeprom_tssi[RF90_PATH_B] = hwinfo[EEPROM_TSSI_B];
 	} else {
@@ -1628,7 +1637,7 @@ static void _rtl92ce_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 		rtlefuse->eeprom_tssi[RF90_PATH_A],
 		rtlefuse->eeprom_tssi[RF90_PATH_B]);
 
-	if (!autoload_fail)
+	if (!autoload_fail && hwinfo[EEPROM_THERMAL_METER] != 0xff)
 		tempval = hwinfo[EEPROM_THERMAL_METER];
 	else
 		tempval = EEPROM_DEFAULT_THERMALMETER;
-- 
cgit v1.2.3-70-g09d2


From 015bf4df8ea67080668ac8314e692076dea1ce6d Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Wed, 15 Feb 2023 13:34:37 +0100
Subject: wifi: wfx: Remove some dead code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

wait_for_completion_timeout() can not return a <0 value.
So simplify the logic and remove dead code.

-ERESTARTSYS can not be returned by do_wait_for_common() for tasks with
TASK_UNINTERRUPTIBLE, which is the case for wait_for_completion_timeout()

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/809c4a645c8d1306c0d256345515865c40ec731c.1676464422.git.christophe.jaillet@wanadoo.fr
---
 drivers/net/wireless/silabs/wfx/main.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/silabs/wfx/main.c b/drivers/net/wireless/silabs/wfx/main.c
index 6b9864e478ac..0b50f7058bbb 100644
--- a/drivers/net/wireless/silabs/wfx/main.c
+++ b/drivers/net/wireless/silabs/wfx/main.c
@@ -358,13 +358,9 @@ int wfx_probe(struct wfx_dev *wdev)
 
 	wfx_bh_poll_irq(wdev);
 	err = wait_for_completion_timeout(&wdev->firmware_ready, 1 * HZ);
-	if (err <= 0) {
-		if (err == 0) {
-			dev_err(wdev->dev, "timeout while waiting for startup indication\n");
-			err = -ETIMEDOUT;
-		} else if (err == -ERESTARTSYS) {
-			dev_info(wdev->dev, "probe interrupted by user\n");
-		}
+	if (err == 0) {
+		dev_err(wdev->dev, "timeout while waiting for startup indication\n");
+		err = -ETIMEDOUT;
 		goto bh_unregister;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 398eb19415ebb01ccc8e5bb75ff04cb760b01e05 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 16 Feb 2023 08:46:54 +0800
Subject: wifi: rtl8xxxu: 8188e: parse single one element of RA report for
 station mode

Intentionally parsing single one element of RA report by breaking loop
causes a smatch warning:
  drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c:1678 rtl8188e_handle_ra_tx_report2() warn:
  ignoring unreachable code.

With existing comments, it intends to process single one element for
station mode, but it will parse more elements in AP mode if it's
implemented. Implement program logic according to existing comment to avoid
smatch warning, and also be usable for both AP and stations modes.

Compile test only.

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/r/202302142135.LCqUTVGY-lkp@intel.com/
Cc: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Tested-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230216004654.4642-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
index a99ddb41cd24..f15b099899e5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
@@ -1699,6 +1699,12 @@ void rtl8188e_handle_ra_tx_report2(struct rtl8xxxu_priv *priv, struct sk_buff *s
 
 	dev_dbg(dev, "%s: len: %d items: %d\n", __func__, tx_rpt_len, items);
 
+	/* We only use macid 0, so only the first item is relevant.
+	 * AP mode will use more of them if it's ever implemented.
+	 */
+	if (!priv->vif || priv->vif->type == NL80211_IFTYPE_STATION)
+		items = 1;
+
 	for (macid = 0; macid < items; macid++) {
 		valid = false;
 
@@ -1741,12 +1747,6 @@ void rtl8188e_handle_ra_tx_report2(struct rtl8xxxu_priv *priv, struct sk_buff *s
 			min_rpt_time = ra->rpt_time;
 
 		rpt += TX_RPT2_ITEM_SIZE;
-
-		/*
-		 * We only use macid 0, so only the first item is relevant.
-		 * AP mode will use more of them if it's ever implemented.
-		 */
-		break;
 	}
 
 	if (min_rpt_time != ra->pre_min_rpt_time) {
-- 
cgit v1.2.3-70-g09d2


From db5e4b3645534bc8f8d6c24537e9564a62073ac8 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Fri, 17 Feb 2023 17:25:29 +0800
Subject: wifi: rtlwifi: rtl8192de: Remove the unused variable bcnfunc_enable

Variable bcnfunc_enable is not effectively used, so delete it.

drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c:1050:5: warning: variable 'bcnfunc_enable' set but not used.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4110
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230217092529.105899-1-jiapeng.chong@linux.alibaba.com
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c
index 2aecb2583f75..df1e36fbc348 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c
@@ -1047,7 +1047,6 @@ static int _rtl92de_set_media_status(struct ieee80211_hw *hw,
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	u8 bt_msr = rtl_read_byte(rtlpriv, MSR);
 	enum led_ctl_mode ledaction = LED_CTL_NO_LINK;
-	u8 bcnfunc_enable;
 
 	bt_msr &= 0xfc;
 
@@ -1064,31 +1063,26 @@ static int _rtl92de_set_media_status(struct ieee80211_hw *hw,
 			"Set HW_VAR_MEDIA_STATUS: No such media status(%x)\n",
 			type);
 	}
-	bcnfunc_enable = rtl_read_byte(rtlpriv, REG_BCN_CTRL);
 	switch (type) {
 	case NL80211_IFTYPE_UNSPECIFIED:
 		bt_msr |= MSR_NOLINK;
 		ledaction = LED_CTL_LINK;
-		bcnfunc_enable &= 0xF7;
 		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
 			"Set Network type to NO LINK!\n");
 		break;
 	case NL80211_IFTYPE_ADHOC:
 		bt_msr |= MSR_ADHOC;
-		bcnfunc_enable |= 0x08;
 		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
 			"Set Network type to Ad Hoc!\n");
 		break;
 	case NL80211_IFTYPE_STATION:
 		bt_msr |= MSR_INFRA;
 		ledaction = LED_CTL_LINK;
-		bcnfunc_enable &= 0xF7;
 		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
 			"Set Network type to STA!\n");
 		break;
 	case NL80211_IFTYPE_AP:
 		bt_msr |= MSR_AP;
-		bcnfunc_enable |= 0x08;
 		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
 			"Set Network type to AP!\n");
 		break;
-- 
cgit v1.2.3-70-g09d2


From c6aa9a9c47252ac7b07ed6d10459027e2f2a2de0 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 17 Feb 2023 20:00:07 +0800
Subject: wifi: rtw89: add RNR support for 6 GHz scan

Since 6 GHz band has around 60 channels and more strict rules for
active probing. Reduced neighbor report can be used to reduce the
channels we scan and get specific target BSS info to probe for.

Declare flag WIPHY_FLAG_SPLIT_SCAN_6GHZ so the scan request could be
divided into two portions: legacy bands and 6 GHz bands. So RNR
information from legacy bands could later be used when 6 GHz scan.

When the scan flag NL80211_SCAN_FLAG_COLOCATED_6GHZ is set, cfg80211
will pass down a reduced channel set which contains PSCs and non-PSC
with RNR info received in the 2 GHz/5 GHz band. This reduces the
scan duration by allowing us to only scan for channels in which APs
are currently operating.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230217120007.8835-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c |  35 +++++++-
 drivers/net/wireless/realtek/rtw89/fw.c   | 136 ++++++++++++++++++++++++++----
 drivers/net/wireless/realtek/rtw89/fw.h   |   7 ++
 3 files changed, 162 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index f09361bc4a4d..489fa7a86160 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -1400,6 +1400,34 @@ static void rtw89_stats_trigger_frame(struct rtw89_dev *rtwdev,
 	}
 }
 
+static void rtw89_core_cancel_6ghz_probe_tx(struct rtw89_dev *rtwdev,
+					    struct sk_buff *skb)
+{
+	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
+	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)skb->data;
+	struct list_head *pkt_list = rtwdev->scan_info.pkt_list;
+	struct rtw89_pktofld_info *info;
+	const u8 *ies = mgmt->u.beacon.variable, *ssid_ie;
+
+	if (rx_status->band != NL80211_BAND_6GHZ)
+		return;
+
+	ssid_ie = cfg80211_find_ie(WLAN_EID_SSID, ies, skb->len);
+
+	list_for_each_entry(info, &pkt_list[NL80211_BAND_6GHZ], list) {
+		if (ether_addr_equal(info->bssid, mgmt->bssid)) {
+			rtw89_fw_h2c_del_pkt_offload(rtwdev, info->id);
+			continue;
+		}
+
+		if (!ssid_ie || ssid_ie[1] != info->ssid_len || info->ssid_len == 0)
+			continue;
+
+		if (memcmp(&ssid_ie[2], info->ssid, info->ssid_len) == 0)
+			rtw89_fw_h2c_del_pkt_offload(rtwdev, info->id);
+	}
+}
+
 static void rtw89_vif_rx_stats_iter(void *data, u8 *mac,
 				    struct ieee80211_vif *vif)
 {
@@ -1412,6 +1440,11 @@ static void rtw89_vif_rx_stats_iter(void *data, u8 *mac,
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
 	const u8 *bssid = iter_data->bssid;
 
+	if (rtwdev->scanning &&
+	    (ieee80211_is_beacon(hdr->frame_control) ||
+	     ieee80211_is_probe_resp(hdr->frame_control)))
+		rtw89_core_cancel_6ghz_probe_tx(rtwdev, skb);
+
 	if (!vif->bss_conf.bssid)
 		return;
 
@@ -3372,7 +3405,7 @@ static int rtw89_core_register_hw(struct rtw89_dev *rtwdev)
 
 	hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_TDLS |
 			    WIPHY_FLAG_TDLS_EXTERNAL_SETUP |
-			    WIPHY_FLAG_AP_UAPSD;
+			    WIPHY_FLAG_AP_UAPSD | WIPHY_FLAG_SPLIT_SCAN_6GHZ;
 	hw->wiphy->features |= NL80211_FEATURE_SCAN_RANDOM_MAC_ADDR;
 
 	hw->wiphy->max_scan_ssids = RTW89_SCANOFLD_MAX_SSID;
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 0b73dc2e9ad7..a88f7974c1d6 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2702,9 +2702,29 @@ static void rtw89_release_pkt_list(struct rtw89_dev *rtwdev)
 	}
 }
 
+static bool rtw89_is_6ghz_wildcard_probe_req(struct rtw89_dev *rtwdev,
+					     struct rtw89_vif *rtwvif,
+					     struct rtw89_pktofld_info *info,
+					     enum nl80211_band band, u8 ssid_idx)
+{
+	struct cfg80211_scan_request *req = rtwvif->scan_req;
+
+	if (band != NL80211_BAND_6GHZ)
+		return false;
+
+	if (req->ssids[ssid_idx].ssid_len) {
+		memcpy(info->ssid, req->ssids[ssid_idx].ssid,
+		       req->ssids[ssid_idx].ssid_len);
+		info->ssid_len = req->ssids[ssid_idx].ssid_len;
+		return false;
+	} else {
+		return true;
+	}
+}
+
 static int rtw89_append_probe_req_ie(struct rtw89_dev *rtwdev,
 				     struct rtw89_vif *rtwvif,
-				     struct sk_buff *skb)
+				     struct sk_buff *skb, u8 ssid_idx)
 {
 	struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info;
 	struct ieee80211_scan_ies *ies = rtwvif->scan_ies;
@@ -2732,6 +2752,13 @@ static int rtw89_append_probe_req_ie(struct rtw89_dev *rtwdev,
 			goto out;
 		}
 
+		if (rtw89_is_6ghz_wildcard_probe_req(rtwdev, rtwvif, info, band,
+						     ssid_idx)) {
+			kfree_skb(new);
+			kfree(info);
+			goto out;
+		}
+
 		ret = rtw89_fw_h2c_add_pkt_offload(rtwdev, &info->id, new);
 		if (ret) {
 			kfree_skb(new);
@@ -2762,7 +2789,7 @@ static int rtw89_hw_scan_update_probe_req(struct rtw89_dev *rtwdev,
 		if (!skb)
 			return -ENOMEM;
 
-		ret = rtw89_append_probe_req_ie(rtwdev, rtwvif, skb);
+		ret = rtw89_append_probe_req_ie(rtwdev, rtwvif, skb, i);
 		kfree_skb(skb);
 
 		if (ret)
@@ -2772,6 +2799,77 @@ static int rtw89_hw_scan_update_probe_req(struct rtw89_dev *rtwdev,
 	return 0;
 }
 
+static int rtw89_update_6ghz_rnr_chan(struct rtw89_dev *rtwdev,
+				      struct cfg80211_scan_request *req,
+				      struct rtw89_mac_chinfo *ch_info)
+{
+	struct ieee80211_vif *vif = rtwdev->scan_info.scanning_vif;
+	struct list_head *pkt_list = rtwdev->scan_info.pkt_list;
+	struct rtw89_vif *rtwvif = vif_to_rtwvif_safe(vif);
+	struct ieee80211_scan_ies *ies = rtwvif->scan_ies;
+	struct cfg80211_scan_6ghz_params *params;
+	struct rtw89_pktofld_info *info, *tmp;
+	struct ieee80211_hdr *hdr;
+	struct sk_buff *skb;
+	bool found;
+	int ret = 0;
+	u8 i;
+
+	if (!req->n_6ghz_params)
+		return 0;
+
+	for (i = 0; i < req->n_6ghz_params; i++) {
+		params = &req->scan_6ghz_params[i];
+
+		if (req->channels[params->channel_idx]->hw_value !=
+		    ch_info->pri_ch)
+			continue;
+
+		found = false;
+		list_for_each_entry(tmp, &pkt_list[NL80211_BAND_6GHZ], list) {
+			if (ether_addr_equal(tmp->bssid, params->bssid)) {
+				found = true;
+				break;
+			}
+		}
+		if (found)
+			continue;
+
+		skb = ieee80211_probereq_get(rtwdev->hw, rtwvif->mac_addr,
+					     NULL, 0, req->ie_len);
+		skb_put_data(skb, ies->ies[NL80211_BAND_6GHZ], ies->len[NL80211_BAND_6GHZ]);
+		skb_put_data(skb, ies->common_ies, ies->common_ie_len);
+		hdr = (struct ieee80211_hdr *)skb->data;
+		ether_addr_copy(hdr->addr3, params->bssid);
+
+		info = kzalloc(sizeof(*info), GFP_KERNEL);
+		if (!info) {
+			ret = -ENOMEM;
+			kfree_skb(skb);
+			goto out;
+		}
+
+		ret = rtw89_fw_h2c_add_pkt_offload(rtwdev, &info->id, skb);
+		if (ret) {
+			kfree_skb(skb);
+			kfree(info);
+			goto out;
+		}
+
+		ether_addr_copy(info->bssid, params->bssid);
+		info->channel_6ghz = req->channels[params->channel_idx]->hw_value;
+		list_add_tail(&info->list, &rtwdev->scan_info.pkt_list[NL80211_BAND_6GHZ]);
+
+		ch_info->tx_pkt = true;
+		ch_info->period = RTW89_CHANNEL_TIME_6G + RTW89_DWELL_TIME_6G;
+
+		kfree_skb(skb);
+	}
+
+out:
+	return ret;
+}
+
 static void rtw89_hw_scan_add_chan(struct rtw89_dev *rtwdev, int chan_type,
 				   int ssid_num,
 				   struct rtw89_mac_chinfo *ch_info)
@@ -2782,6 +2880,7 @@ static void rtw89_hw_scan_add_chan(struct rtw89_dev *rtwdev, int chan_type,
 	struct cfg80211_scan_request *req = rtwvif->scan_req;
 	struct rtw89_pktofld_info *info;
 	u8 band, probe_count = 0;
+	int ret;
 
 	ch_info->notify_action = RTW89_SCANOFLD_DEBUG_MASK;
 	ch_info->dfs_ch = chan_type == RTW89_CHAN_DFS;
@@ -2793,25 +2892,31 @@ static void rtw89_hw_scan_add_chan(struct rtw89_dev *rtwdev, int chan_type,
 	ch_info->pause_data = false;
 	ch_info->probe_id = RTW89_SCANOFLD_PKT_NONE;
 
+	if (ch_info->ch_band == RTW89_BAND_6G) {
+		if ((ssid_num == 1 && req->ssids[0].ssid_len == 0) ||
+		    !ch_info->is_psc) {
+			ch_info->tx_pkt = false;
+			if (!req->duration_mandatory)
+				ch_info->period -= RTW89_DWELL_TIME_6G;
+		}
+	}
+
+	ret = rtw89_update_6ghz_rnr_chan(rtwdev, req, ch_info);
+	if (ret)
+		rtw89_warn(rtwdev, "RNR fails: %d\n", ret);
+
 	if (ssid_num) {
-		ch_info->num_pkt = ssid_num;
 		band = rtw89_hw_to_nl80211_band(ch_info->ch_band);
 
 		list_for_each_entry(info, &scan_info->pkt_list[band], list) {
-			ch_info->pkt_id[probe_count] = info->id;
-			if (++probe_count >= ssid_num)
+			if (info->channel_6ghz &&
+			    ch_info->pri_ch != info->channel_6ghz)
+				continue;
+			ch_info->pkt_id[probe_count++] = info->id;
+			if (probe_count >= RTW89_SCANOFLD_MAX_SSID)
 				break;
 		}
-		if (probe_count != ssid_num)
-			rtw89_err(rtwdev, "SSID num differs from list len\n");
-	}
-
-	if (ch_info->ch_band == RTW89_BAND_6G) {
-		if (ssid_num == 1 && req->ssids[0].ssid_len == 0) {
-			ch_info->tx_pkt = false;
-			if (!req->duration_mandatory)
-				ch_info->period -= RTW89_DWELL_TIME_6G;
-		}
+		ch_info->num_pkt = probe_count;
 	}
 
 	switch (chan_type) {
@@ -2872,6 +2977,7 @@ static int rtw89_hw_scan_add_chan_list(struct rtw89_dev *rtwdev,
 		ch_info->central_ch = channel->hw_value;
 		ch_info->pri_ch = channel->hw_value;
 		ch_info->rand_seq_num = random_seq;
+		ch_info->is_psc = cfg80211_channel_is_psc(channel);
 
 		if (channel->flags &
 		    (IEEE80211_CHAN_RADAR | IEEE80211_CHAN_NO_IR))
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index cae07e325326..3f6e0871381d 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -237,6 +237,7 @@ struct rtw89_mac_chinfo {
 	u16 tx_pwr_idx;
 	u8 rsvd1;
 	struct list_head list;
+	bool is_psc;
 };
 
 struct rtw89_scan_option {
@@ -247,6 +248,12 @@ struct rtw89_scan_option {
 struct rtw89_pktofld_info {
 	struct list_head list;
 	u8 id;
+
+	/* Below fields are for 6 GHz RNR use only */
+	u8 ssid[IEEE80211_MAX_SSID_LEN];
+	u8 ssid_len;
+	u8 bssid[ETH_ALEN];
+	u16 channel_6ghz;
 };
 
 static inline void RTW89_SET_FWCMD_RA_IS_DIS(void *cmd, u32 val)
-- 
cgit v1.2.3-70-g09d2


From b9b1e4fe2957f361c86e288ecf373dc7895cf7c7 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Fri, 17 Feb 2023 20:49:12 +0200
Subject: wifi: rtl8xxxu: Remove always true condition in
 rtl8xxxu_print_chipinfo

Fix a new smatch warning:
drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c:1580 rtl8xxxu_print_chipinfo() warn: always true condition '(priv->chip_cut <= 15) => (0-15 <= 15)'

Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202302140753.71IgU77A-lkp@intel.com/
Fixes: 7b0ac469e331 ("wifi: rtl8xxxu: Recognise all possible chip cuts")
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/68eff98b-a022-5a00-f330-adf623a35772@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 620a5cc2bfdd..54ca6f2ced3f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -1575,11 +1575,7 @@ rtl8xxxu_set_spec_sifs(struct rtl8xxxu_priv *priv, u16 cck, u16 ofdm)
 static void rtl8xxxu_print_chipinfo(struct rtl8xxxu_priv *priv)
 {
 	struct device *dev = &priv->udev->dev;
-	char cut = '?';
-
-	/* Currently always true: chip_cut is 4 bits. */
-	if (priv->chip_cut <= 15)
-		cut = 'A' + priv->chip_cut;
+	char cut = 'A' + priv->chip_cut;
 
 	dev_info(dev,
 		 "RTL%s rev %c (%s) romver %d, %iT%iR, TX queues %i, WiFi=%i, BT=%i, GPS=%i, HI PA=%i\n",
-- 
cgit v1.2.3-70-g09d2


From 96c79da2e4d1a25e73916c656fe4ccf7fc8176ae Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sat, 18 Feb 2023 16:29:40 +0100
Subject: wifi: rtw88: mac: Add support for the SDIO HCI in
 rtw_pwr_seq_parser()

rtw_pwr_seq_parser() needs to know about the HCI bus interface mask for
the SDIO bus so it can parse the chip state change sequences.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230218152944.48842-2-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index dae64901bac5..c08071a67b41 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -222,6 +222,9 @@ static int rtw_pwr_seq_parser(struct rtw_dev *rtwdev,
 	case RTW_HCI_TYPE_USB:
 		intf_mask = RTW_PWR_INTF_USB_MSK;
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		intf_mask = RTW_PWR_INTF_SDIO_MSK;
+		break;
 	default:
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-70-g09d2


From 8599ea40582d49afe437badde76e31bd7429c607 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sat, 18 Feb 2023 16:29:41 +0100
Subject: wifi: rtw88: mac: Add SDIO HCI support in the TX/page table setup

txdma_queue_mapping() and priority_queue_cfg() can use the first entry
of each chip's rqpn_table and page_table. Add this mapping so data
transmission is possible on SDIO based chipsets.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230218152944.48842-3-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index c08071a67b41..1c9530a0eb69 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -1043,6 +1043,9 @@ static int txdma_queue_mapping(struct rtw_dev *rtwdev)
 		else
 			return -EINVAL;
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		rqpn = &chip->rqpn_table[0];
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -1205,6 +1208,9 @@ static int priority_queue_cfg(struct rtw_dev *rtwdev)
 		else
 			return -EINVAL;
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		pg_tbl = &chip->page_table[0];
+		break;
 	default:
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-70-g09d2


From 64e9d564653566812dfade36d20f1794407e6ca1 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sat, 18 Feb 2023 16:29:42 +0100
Subject: wifi: rtw88: rtw8821c: Implement RTL8821CS (SDIO) efuse parsing

The efuse of the SDIO RTL8821CS chip has only one known member: the mac
address is at offset 0x11a. Add a struct rtw8821cs_efuse describing this
and use it for copying the mac address when the SDIO bus is used.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230218152944.48842-4-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/rtw8821c.c | 9 +++++++++
 drivers/net/wireless/realtek/rtw88/rtw8821c.h | 6 ++++++
 2 files changed, 15 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.c b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
index 17f800f6efbd..7ae0541d7b99 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
@@ -32,6 +32,12 @@ static void rtw8821cu_efuse_parsing(struct rtw_efuse *efuse,
 	ether_addr_copy(efuse->addr, map->u.mac_addr);
 }
 
+static void rtw8821cs_efuse_parsing(struct rtw_efuse *efuse,
+				    struct rtw8821c_efuse *map)
+{
+	ether_addr_copy(efuse->addr, map->s.mac_addr);
+}
+
 enum rtw8821ce_rf_set {
 	SWITCH_TO_BTG,
 	SWITCH_TO_WLG,
@@ -77,6 +83,9 @@ static int rtw8821c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 	case RTW_HCI_TYPE_USB:
 		rtw8821cu_efuse_parsing(efuse, map);
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		rtw8821cs_efuse_parsing(efuse, map);
+		break;
 	default:
 		/* unsupported now */
 		return -ENOTSUPP;
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.h b/drivers/net/wireless/realtek/rtw88/rtw8821c.h
index 1c81260f3a54..fcff31688c45 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.h
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.h
@@ -65,6 +65,11 @@ struct rtw8821ce_efuse {
 	u8 res7;
 };
 
+struct rtw8821cs_efuse {
+	u8 res4[0x4a];			/* 0xd0 */
+	u8 mac_addr[ETH_ALEN];		/* 0x11a */
+} __packed;
+
 struct rtw8821c_efuse {
 	__le16 rtl_id;
 	u8 res0[0x0e];
@@ -94,6 +99,7 @@ struct rtw8821c_efuse {
 	union {
 		struct rtw8821ce_efuse e;
 		struct rtw8821cu_efuse u;
+		struct rtw8821cs_efuse s;
 	};
 };
 
-- 
cgit v1.2.3-70-g09d2


From 9e688784b8a13515ee186d0ee1cfab12b6d05241 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sat, 18 Feb 2023 16:29:43 +0100
Subject: wifi: rtw88: rtw8822b: Implement RTL8822BS (SDIO) efuse parsing

The efuse of the SDIO RTL8822BS chip has only one known member: the mac
address is at offset 0x11a. Add a struct rtw8822bs_efuse describing this
and use it for copying the mac address when the SDIO bus is used.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230218152944.48842-5-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/rtw8822b.c | 9 +++++++++
 drivers/net/wireless/realtek/rtw88/rtw8822b.h | 8 +++++++-
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822b.c b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
index 74dfb89b2c94..531b67787e2e 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822b.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
@@ -32,6 +32,12 @@ static void rtw8822bu_efuse_parsing(struct rtw_efuse *efuse,
 	ether_addr_copy(efuse->addr, map->u.mac_addr);
 }
 
+static void rtw8822bs_efuse_parsing(struct rtw_efuse *efuse,
+				    struct rtw8822b_efuse *map)
+{
+	ether_addr_copy(efuse->addr, map->s.mac_addr);
+}
+
 static int rtw8822b_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 {
 	struct rtw_efuse *efuse = &rtwdev->efuse;
@@ -65,6 +71,9 @@ static int rtw8822b_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 	case RTW_HCI_TYPE_USB:
 		rtw8822bu_efuse_parsing(efuse, map);
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		rtw8822bs_efuse_parsing(efuse, map);
+		break;
 	default:
 		/* unsupported now */
 		return -ENOTSUPP;
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822b.h b/drivers/net/wireless/realtek/rtw88/rtw8822b.h
index 01d3644e0c94..2dc3a6660f06 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822b.h
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822b.h
@@ -65,6 +65,11 @@ struct rtw8822be_efuse {
 	u8 res7;
 };
 
+struct rtw8822bs_efuse {
+	u8 res4[0x4a];			/* 0xd0 */
+	u8 mac_addr[ETH_ALEN];		/* 0x11a */
+} __packed;
+
 struct rtw8822b_efuse {
 	__le16 rtl_id;
 	u8 res0[0x0e];
@@ -92,8 +97,9 @@ struct rtw8822b_efuse {
 	u8 country_code[2];
 	u8 res[3];
 	union {
-		struct rtw8822bu_efuse u;
 		struct rtw8822be_efuse e;
+		struct rtw8822bu_efuse u;
+		struct rtw8822bs_efuse s;
 	};
 };
 
-- 
cgit v1.2.3-70-g09d2


From ad0a677bce20c7fa2e8af3fd7f23c0686018202b Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sat, 18 Feb 2023 16:29:44 +0100
Subject: wifi: rtw88: rtw8822c: Implement RTL8822CS (SDIO) efuse parsing

The efuse of the SDIO RTL8822CS chip has only one known member: the mac
address is at offset 0x16a. Add a struct rtw8822cs_efuse describing this
and use it for copying the mac address when the SDIO bus is used.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230218152944.48842-6-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/rtw8822c.c | 9 +++++++++
 drivers/net/wireless/realtek/rtw88/rtw8822c.h | 8 +++++++-
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 964e27887fe2..5a2c004b12df 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -35,6 +35,12 @@ static void rtw8822cu_efuse_parsing(struct rtw_efuse *efuse,
 	ether_addr_copy(efuse->addr, map->u.mac_addr);
 }
 
+static void rtw8822cs_efuse_parsing(struct rtw_efuse *efuse,
+				    struct rtw8822c_efuse *map)
+{
+	ether_addr_copy(efuse->addr, map->s.mac_addr);
+}
+
 static int rtw8822c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 {
 	struct rtw_efuse *efuse = &rtwdev->efuse;
@@ -67,6 +73,9 @@ static int rtw8822c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 	case RTW_HCI_TYPE_USB:
 		rtw8822cu_efuse_parsing(efuse, map);
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		rtw8822cs_efuse_parsing(efuse, map);
+		break;
 	default:
 		/* unsupported now */
 		return -ENOTSUPP;
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.h b/drivers/net/wireless/realtek/rtw88/rtw8822c.h
index 479d5d769c52..1bc0e7f5d6bb 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.h
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.h
@@ -16,6 +16,11 @@ struct rtw8822cu_efuse {
 	u8 res2[0x3d];
 };
 
+struct rtw8822cs_efuse {
+	u8 res0[0x4a];			/* 0x120 */
+	u8 mac_addr[ETH_ALEN];		/* 0x16a */
+} __packed;
+
 struct rtw8822ce_efuse {
 	u8 mac_addr[ETH_ALEN];		/* 0x120 */
 	u8 vender_id[2];
@@ -91,8 +96,9 @@ struct rtw8822c_efuse {
 	u8 res9;
 	u8 res10[0x42];
 	union {
-		struct rtw8822cu_efuse u;
 		struct rtw8822ce_efuse e;
+		struct rtw8822cu_efuse u;
+		struct rtw8822cs_efuse s;
 	};
 };
 
-- 
cgit v1.2.3-70-g09d2


From aa4e055945462e645224795e174c25c82f6002ac Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Mon, 20 Feb 2023 15:01:57 +0800
Subject: wifi: rtw89: add tx_wake notify for 8852B

8852B has the same issue: management frames get stuck when wifi
chip enters low ps mode, so we alse add notify wake function to
trigger wifi chip wake before forwarding management frames.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230220070202.29868-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index a88f7974c1d6..9dc0abc1fc37 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -258,6 +258,7 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 36, 0, CRASH_TRIGGER),
 	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 38, 0, PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, NO_LPS_PG),
+	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, TX_WAKE),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 20, 0, PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852C, le, 0, 27, 33, 0, NO_DEEP_PS),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 34, 0, TX_WAKE),
-- 
cgit v1.2.3-70-g09d2


From 31c416e69dbf416583d0d8245d91e7e82b7846b1 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Mon, 20 Feb 2023 15:01:58 +0800
Subject: wifi: rtw89: fw: configure CRASH_TRIGGER feature for 8852B

RTL8852B firmware supports CRASH_TRIGGER feature from v0.29.29.0.
After this is configured, debugfs fw_crash can support type 1 on
RTL8852B to trigger firmware crash and verify L2 recovery through
simulation.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230220070202.29868-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 9dc0abc1fc37..d5b1f1b0fd67 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -259,6 +259,7 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 38, 0, PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, NO_LPS_PG),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, TX_WAKE),
+	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 29, 0, CRASH_TRIGGER),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 20, 0, PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852C, le, 0, 27, 33, 0, NO_DEEP_PS),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 34, 0, TX_WAKE),
-- 
cgit v1.2.3-70-g09d2


From bb9040b3ff970529ceaa20b064c113d5ffd5520f Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Mon, 20 Feb 2023 15:01:59 +0800
Subject: wifi: rtw89: adjust channel encoding to common function

Since the range of channel table is identical among ICs. Make channel
encode/decode function common and not IC dependent. So all ICs with
matching firmware that needs this kind of coding can use it directly.
This patch doesn't change logic at all.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230220070202.29868-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/phy.c      | 72 ++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/phy.h      |  3 ++
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 74 +--------------------------
 3 files changed, 77 insertions(+), 72 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index d9f61ba3d176..d8b035972dd4 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -4294,3 +4294,75 @@ void rtw89_phy_tssi_ctrl_set_bandedge_cfg(struct rtw89_dev *rtwdev,
 					      data[RTW89_TSSI_SBW20]);
 }
 EXPORT_SYMBOL(rtw89_phy_tssi_ctrl_set_bandedge_cfg);
+
+static
+const u8 rtw89_ch_base_table[16] = {1, 0xff,
+				    36, 100, 132, 149, 0xff,
+				    1, 33, 65, 97, 129, 161, 193, 225, 0xff};
+#define RTW89_CH_BASE_IDX_2G		0
+#define RTW89_CH_BASE_IDX_5G_FIRST	2
+#define RTW89_CH_BASE_IDX_5G_LAST	5
+#define RTW89_CH_BASE_IDX_6G_FIRST	7
+#define RTW89_CH_BASE_IDX_6G_LAST	14
+
+#define RTW89_CH_BASE_IDX_MASK		GENMASK(7, 4)
+#define RTW89_CH_OFFSET_MASK		GENMASK(3, 0)
+
+u8 rtw89_encode_chan_idx(struct rtw89_dev *rtwdev, u8 central_ch, u8 band)
+{
+	u8 chan_idx;
+	u8 last, first;
+	u8 idx;
+
+	switch (band) {
+	case RTW89_BAND_2G:
+		chan_idx = FIELD_PREP(RTW89_CH_BASE_IDX_MASK, RTW89_CH_BASE_IDX_2G) |
+			   FIELD_PREP(RTW89_CH_OFFSET_MASK, central_ch);
+		return chan_idx;
+	case RTW89_BAND_5G:
+		first = RTW89_CH_BASE_IDX_5G_FIRST;
+		last = RTW89_CH_BASE_IDX_5G_LAST;
+		break;
+	case RTW89_BAND_6G:
+		first = RTW89_CH_BASE_IDX_6G_FIRST;
+		last = RTW89_CH_BASE_IDX_6G_LAST;
+		break;
+	default:
+		rtw89_warn(rtwdev, "Unsupported band %d\n", band);
+		return 0;
+	}
+
+	for (idx = last; idx >= first; idx--)
+		if (central_ch >= rtw89_ch_base_table[idx])
+			break;
+
+	if (idx < first) {
+		rtw89_warn(rtwdev, "Unknown band %d channel %d\n", band, central_ch);
+		return 0;
+	}
+
+	chan_idx = FIELD_PREP(RTW89_CH_BASE_IDX_MASK, idx) |
+		   FIELD_PREP(RTW89_CH_OFFSET_MASK,
+			      (central_ch - rtw89_ch_base_table[idx]) >> 1);
+	return chan_idx;
+}
+EXPORT_SYMBOL(rtw89_encode_chan_idx);
+
+void rtw89_decode_chan_idx(struct rtw89_dev *rtwdev, u8 chan_idx,
+			   u8 *ch, enum nl80211_band *band)
+{
+	u8 idx, offset;
+
+	idx = FIELD_GET(RTW89_CH_BASE_IDX_MASK, chan_idx);
+	offset = FIELD_GET(RTW89_CH_OFFSET_MASK, chan_idx);
+
+	if (idx == RTW89_CH_BASE_IDX_2G) {
+		*band = NL80211_BAND_2GHZ;
+		*ch = offset;
+		return;
+	}
+
+	*band = idx <= RTW89_CH_BASE_IDX_5G_LAST ? NL80211_BAND_5GHZ : NL80211_BAND_6GHZ;
+	*ch = rtw89_ch_base_table[idx] + (offset << 1);
+}
+EXPORT_SYMBOL(rtw89_decode_chan_idx);
diff --git a/drivers/net/wireless/realtek/rtw89/phy.h b/drivers/net/wireless/realtek/rtw89/phy.h
index 21233f094644..de0a9abf646e 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.h
+++ b/drivers/net/wireless/realtek/rtw89/phy.h
@@ -555,5 +555,8 @@ void rtw89_phy_tssi_ctrl_set_bandedge_cfg(struct rtw89_dev *rtwdev,
 					  enum rtw89_tssi_bandedge_cfg bandedge_cfg);
 void rtw89_phy_ul_tb_assoc(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
 void rtw89_phy_ul_tb_ctrl_track(struct rtw89_dev *rtwdev);
+u8 rtw89_encode_chan_idx(struct rtw89_dev *rtwdev, u8 central_ch, u8 band);
+void rtw89_decode_chan_idx(struct rtw89_dev *rtwdev, u8 chan_idx,
+			   u8 *ch, enum nl80211_band *band);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index d2dde21d3daf..8af813132f71 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -852,76 +852,6 @@ static void rtw8852c_set_gain_error(struct rtw89_dev *rtwdev,
 	}
 }
 
-static
-const u8 rtw8852c_ch_base_table[16] = {1, 0xff,
-				       36, 100, 132, 149, 0xff,
-				       1, 33, 65, 97, 129, 161, 193, 225, 0xff};
-#define RTW8852C_CH_BASE_IDX_2G		0
-#define RTW8852C_CH_BASE_IDX_5G_FIRST	2
-#define RTW8852C_CH_BASE_IDX_5G_LAST	5
-#define RTW8852C_CH_BASE_IDX_6G_FIRST	7
-#define RTW8852C_CH_BASE_IDX_6G_LAST	14
-
-#define RTW8852C_CH_BASE_IDX_MASK	GENMASK(7, 4)
-#define RTW8852C_CH_OFFSET_MASK		GENMASK(3, 0)
-
-static u8 rtw8852c_encode_chan_idx(struct rtw89_dev *rtwdev, u8 central_ch, u8 band)
-{
-	u8 chan_idx;
-	u8 last, first;
-	u8 idx;
-
-	switch (band) {
-	case RTW89_BAND_2G:
-		chan_idx = FIELD_PREP(RTW8852C_CH_BASE_IDX_MASK, RTW8852C_CH_BASE_IDX_2G) |
-			   FIELD_PREP(RTW8852C_CH_OFFSET_MASK, central_ch);
-		return chan_idx;
-	case RTW89_BAND_5G:
-		first = RTW8852C_CH_BASE_IDX_5G_FIRST;
-		last = RTW8852C_CH_BASE_IDX_5G_LAST;
-		break;
-	case RTW89_BAND_6G:
-		first = RTW8852C_CH_BASE_IDX_6G_FIRST;
-		last = RTW8852C_CH_BASE_IDX_6G_LAST;
-		break;
-	default:
-		rtw89_warn(rtwdev, "Unsupported band %d\n", band);
-		return 0;
-	}
-
-	for (idx = last; idx >= first; idx--)
-		if (central_ch >= rtw8852c_ch_base_table[idx])
-			break;
-
-	if (idx < first) {
-		rtw89_warn(rtwdev, "Unknown band %d channel %d\n", band, central_ch);
-		return 0;
-	}
-
-	chan_idx = FIELD_PREP(RTW8852C_CH_BASE_IDX_MASK, idx) |
-		   FIELD_PREP(RTW8852C_CH_OFFSET_MASK,
-			      (central_ch - rtw8852c_ch_base_table[idx]) >> 1);
-	return chan_idx;
-}
-
-static void rtw8852c_decode_chan_idx(struct rtw89_dev *rtwdev, u8 chan_idx,
-				     u8 *ch, enum nl80211_band *band)
-{
-	u8 idx, offset;
-
-	idx = FIELD_GET(RTW8852C_CH_BASE_IDX_MASK, chan_idx);
-	offset = FIELD_GET(RTW8852C_CH_OFFSET_MASK, chan_idx);
-
-	if (idx == RTW8852C_CH_BASE_IDX_2G) {
-		*band = NL80211_BAND_2GHZ;
-		*ch = offset;
-		return;
-	}
-
-	*band = idx <= RTW8852C_CH_BASE_IDX_5G_LAST ? NL80211_BAND_5GHZ : NL80211_BAND_6GHZ;
-	*ch = rtw8852c_ch_base_table[idx] + (offset << 1);
-}
-
 static void rtw8852c_set_gain_offset(struct rtw89_dev *rtwdev,
 				     const struct rtw89_chan *chan,
 				     enum rtw89_phy_idx phy_idx,
@@ -1084,7 +1014,7 @@ static void rtw8852c_ctrl_ch(struct rtw89_dev *rtwdev,
 		}
 	}
 
-	chan_idx = rtw8852c_encode_chan_idx(rtwdev, chan->primary_channel, band);
+	chan_idx = rtw89_encode_chan_idx(rtwdev, chan->primary_channel, band);
 	rtw89_phy_write32_idx(rtwdev, R_MAC_PIN_SEL, B_CH_IDX_SEG0, chan_idx, phy_idx);
 }
 
@@ -2730,7 +2660,7 @@ static void rtw8852c_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
 	if (chan_idx == 0)
 		return;
 
-	rtw8852c_decode_chan_idx(rtwdev, chan_idx, &ch, &band);
+	rtw89_decode_chan_idx(rtwdev, chan_idx, &ch, &band);
 	status->freq = ieee80211_channel_to_frequency(ch, band);
 	status->band = band;
 }
-- 
cgit v1.2.3-70-g09d2


From 4f24d7aa575a3c01192faa7f4fb6a91c54f4ef47 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Mon, 20 Feb 2023 15:02:00 +0800
Subject: wifi: rtw89: 8852b: add channel encoding for hw_scan

To obtain correct packet frequency for hw_scan, 52b needs to decode
the channel index obtained from hardware. Change the driver related
set channel part as well to make driver/firmware compatible.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230220070202.29868-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index ee8dba7e0074..499ae0389c71 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -1422,6 +1422,7 @@ static void rtw8852b_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89
 {
 	bool cck_en = chan->channel <= 14;
 	u8 pri_ch_idx = chan->pri_ch_idx;
+	u8 band = chan->band_type, chan_idx;
 
 	if (cck_en)
 		rtw8852b_ctrl_sco_cck(rtwdev,  chan->primary_channel);
@@ -1444,8 +1445,8 @@ static void rtw8852b_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89
 				       B_BT_DYN_DC_EST_EN_MSK, 0x0);
 		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x0);
 	}
-	rtw89_phy_write32_mask(rtwdev, R_MAC_PIN_SEL, B_CH_IDX_SEG0,
-			       chan->primary_channel);
+	chan_idx = rtw89_encode_chan_idx(rtwdev, chan->primary_channel, band);
+	rtw89_phy_write32_mask(rtwdev, R_MAC_PIN_SEL, B_CH_IDX_SEG0, chan_idx);
 	rtw8852b_5m_mask(rtwdev, chan, phy_idx);
 	rtw8852b_bb_set_pop(rtwdev);
 	rtw8852b_bb_reset_all(rtwdev, phy_idx);
@@ -2299,13 +2300,14 @@ static void rtw8852b_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
 					 struct ieee80211_rx_status *status)
 {
 	u16 chan = phy_ppdu->chan_idx;
-	u8 band;
+	enum nl80211_band band;
+	u8 ch;
 
 	if (chan == 0)
 		return;
 
-	band = chan <= 14 ? NL80211_BAND_2GHZ : NL80211_BAND_5GHZ;
-	status->freq = ieee80211_channel_to_frequency(chan, band);
+	rtw89_decode_chan_idx(rtwdev, chan, &ch, &band);
+	status->freq = ieee80211_channel_to_frequency(ch, band);
 	status->band = band;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 357277e1afdae0f56d08846e818579f5a020709d Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Mon, 20 Feb 2023 15:02:01 +0800
Subject: wifi: rtw89: 8852b: enable hw_scan support

This enables hw_scan for 8852b after firmware version 0.29.29.0.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230220070202.29868-6-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index d5b1f1b0fd67..dceb033acbc8 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -260,6 +260,7 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, NO_LPS_PG),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, TX_WAKE),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 29, 0, CRASH_TRIGGER),
+	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 29, 0, SCAN_OFFLOAD),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 20, 0, PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852C, le, 0, 27, 33, 0, NO_DEEP_PS),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 34, 0, TX_WAKE),
-- 
cgit v1.2.3-70-g09d2


From 0d1f7ff19d4f8a6da5a6b60d2afd1d34b5d5ebfc Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Mon, 20 Feb 2023 15:02:02 +0800
Subject: wifi: rtw89: refine FW feature judgement on packet drop

The newer chips use the newer firmware branches, e.g. v027, v029.
And, those firmware branches are supposed to support packet drop
when they are just branched out.

The initial firmware branch used by each chip is as below.
* 8852A: v009
* 8852C: v027
* 8852B: v027

So, only 8852A may use firmware which doesn't support packet drop at
runtime. To save trivial positive listing in firmware feature table,
we change to reverse judgment.

Besides, rtw89_mac_ptk_drop_by_band_and_wait() missed to check firmware
feature before calling rtw89_fw_h2c_pkt_drop(). We also fix it.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230220070202.29868-7-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h     | 2 +-
 drivers/net/wireless/realtek/rtw89/fw.c       | 4 ++--
 drivers/net/wireless/realtek/rtw89/mac.c      | 2 +-
 drivers/net/wireless/realtek/rtw89/mac80211.c | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 41365ffb7e5e..b1a886898c5a 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3023,7 +3023,7 @@ enum rtw89_fw_feature {
 	RTW89_FW_FEATURE_SCAN_OFFLOAD,
 	RTW89_FW_FEATURE_TX_WAKE,
 	RTW89_FW_FEATURE_CRASH_TRIGGER,
-	RTW89_FW_FEATURE_PACKET_DROP,
+	RTW89_FW_FEATURE_NO_PACKET_DROP,
 	RTW89_FW_FEATURE_NO_DEEP_PS,
 	RTW89_FW_FEATURE_NO_LPS_PG,
 };
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index dceb033acbc8..1a4ff24078fb 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -235,6 +235,7 @@ static bool __fw_feat_cond_ ## __cond(u32 suit_ver_code, u32 comp_ver_code) \
 
 __DEF_FW_FEAT_COND(ge, >=); /* greater or equal */
 __DEF_FW_FEAT_COND(le, <=); /* less or equal */
+__DEF_FW_FEAT_COND(lt, <); /* less than */
 
 struct __fw_feat_cfg {
 	enum rtw89_core_chip_id chip_id;
@@ -256,12 +257,11 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 35, 0, SCAN_OFFLOAD),
 	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 35, 0, TX_WAKE),
 	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 36, 0, CRASH_TRIGGER),
-	__CFG_FW_FEAT(RTL8852A, ge, 0, 13, 38, 0, PACKET_DROP),
+	__CFG_FW_FEAT(RTL8852A, lt, 0, 13, 38, 0, NO_PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, NO_LPS_PG),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 26, 0, TX_WAKE),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 29, 0, CRASH_TRIGGER),
 	__CFG_FW_FEAT(RTL8852B, ge, 0, 29, 29, 0, SCAN_OFFLOAD),
-	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 20, 0, PACKET_DROP),
 	__CFG_FW_FEAT(RTL8852C, le, 0, 27, 33, 0, NO_DEEP_PS),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 34, 0, TX_WAKE),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 36, 0, SCAN_OFFLOAD),
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 2e2a2b6eab09..3d1e4ffef1b1 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -5426,7 +5426,7 @@ int rtw89_mac_ptk_drop_by_band_and_wait(struct rtw89_dev *rtwdev,
 	for (i = 0; i < try_cnt; i++) {
 		ret = read_poll_timeout(mac_is_txq_empty, empty, empty, 50,
 					50000, false, rtwdev);
-		if (ret)
+		if (ret && !RTW89_CHK_FW_FEATURE(NO_PACKET_DROP, &rtwdev->fw))
 			rtw89_fw_h2c_pkt_drop(rtwdev, &params);
 		else
 			return 0;
diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index d43281f7335b..367a7bf319da 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -676,7 +676,7 @@ static void rtw89_ops_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	rtw89_leave_lps(rtwdev);
 	rtw89_hci_flush_queues(rtwdev, queues, drop);
 
-	if (drop && RTW89_CHK_FW_FEATURE(PACKET_DROP, &rtwdev->fw))
+	if (drop && !RTW89_CHK_FW_FEATURE(NO_PACKET_DROP, &rtwdev->fw))
 		__rtw89_drop_packets(rtwdev, vif);
 	else
 		rtw89_mac_flush_txq(rtwdev, queues, drop);
-- 
cgit v1.2.3-70-g09d2


From 0a09a2f933c73dc76ab0b72da6855f44342a8903 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Tue, 21 Feb 2023 21:06:42 +0100
Subject: bpf: Annotate data races in bpf_local_storage

There are a few cases where hlist_node is checked to be unhashed without
holding the lock protecting its modification. In this case, one must use
hlist_unhashed_lockless to avoid load tearing and KCSAN reports. Fix
this by using lockless variant in places not protected by the lock.

Since this is not prompted by any actual KCSAN reports but only from
code review, I have not included a fixes tag.

Cc: Martin KaFai Lau <martin.lau@kernel.org>
Cc: KP Singh <kpsingh@kernel.org>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230221200646.2500777-4-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 35f4138a54dc..58da17ae5124 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -51,11 +51,21 @@ owner_storage(struct bpf_local_storage_map *smap, void *owner)
 	return map->ops->map_owner_storage_ptr(owner);
 }
 
+static bool selem_linked_to_storage_lockless(const struct bpf_local_storage_elem *selem)
+{
+	return !hlist_unhashed_lockless(&selem->snode);
+}
+
 static bool selem_linked_to_storage(const struct bpf_local_storage_elem *selem)
 {
 	return !hlist_unhashed(&selem->snode);
 }
 
+static bool selem_linked_to_map_lockless(const struct bpf_local_storage_elem *selem)
+{
+	return !hlist_unhashed_lockless(&selem->map_node);
+}
+
 static bool selem_linked_to_map(const struct bpf_local_storage_elem *selem)
 {
 	return !hlist_unhashed(&selem->map_node);
@@ -174,7 +184,7 @@ static void __bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 	bool free_local_storage = false;
 	unsigned long flags;
 
-	if (unlikely(!selem_linked_to_storage(selem)))
+	if (unlikely(!selem_linked_to_storage_lockless(selem)))
 		/* selem has already been unlinked from sk */
 		return;
 
@@ -208,7 +218,7 @@ void bpf_selem_unlink_map(struct bpf_local_storage_elem *selem)
 	struct bpf_local_storage_map_bucket *b;
 	unsigned long flags;
 
-	if (unlikely(!selem_linked_to_map(selem)))
+	if (unlikely(!selem_linked_to_map_lockless(selem)))
 		/* selem has already be unlinked from smap */
 		return;
 
@@ -420,7 +430,7 @@ bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap,
 		err = check_flags(old_sdata, map_flags);
 		if (err)
 			return ERR_PTR(err);
-		if (old_sdata && selem_linked_to_storage(SELEM(old_sdata))) {
+		if (old_sdata && selem_linked_to_storage_lockless(SELEM(old_sdata))) {
 			copy_map_value_locked(&smap->map, old_sdata->data,
 					      value, false);
 			return old_sdata;
-- 
cgit v1.2.3-70-g09d2


From 521d3c0a1730c29c96870919a7a115577e17f8c7 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Tue, 21 Feb 2023 21:06:43 +0100
Subject: bpf: Remove unused MEM_ALLOC | PTR_TRUSTED checks

The plan is to supposedly tag everything with PTR_TRUSTED eventually,
however those changes should bring in their respective code, instead
of leaving it around right now. It is arguable whether PTR_TRUSTED is
required for all types, when it's only use case is making PTR_TO_BTF_ID
a bit stronger, while all other types are trusted by default.

Hence, just drop the two instances which do not occur in the verifier
for now to avoid reader confusion.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230221200646.2500777-5-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d517d13878cf..477c22c9bbd7 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -6656,7 +6656,6 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env,
 	case PTR_TO_BTF_ID | MEM_ALLOC:
 	case PTR_TO_BTF_ID | PTR_TRUSTED:
 	case PTR_TO_BTF_ID | MEM_RCU:
-	case PTR_TO_BTF_ID | MEM_ALLOC | PTR_TRUSTED:
 	case PTR_TO_BTF_ID | MEM_ALLOC | NON_OWN_REF:
 		/* When referenced PTR_TO_BTF_ID is passed to release function,
 		 * its fixed offset must be 0. In the other cases, fixed offset
@@ -9211,7 +9210,6 @@ static int check_reg_allocation_locked(struct bpf_verifier_env *env, struct bpf_
 		ptr = reg->map_ptr;
 		break;
 	case PTR_TO_BTF_ID | MEM_ALLOC:
-	case PTR_TO_BTF_ID | MEM_ALLOC | PTR_TRUSTED:
 		ptr = reg->btf;
 		break;
 	default:
-- 
cgit v1.2.3-70-g09d2


From da03e43a8c500fcfb11ac5eeb03c1b4a9c1dd958 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Tue, 21 Feb 2023 21:06:44 +0100
Subject: bpf: Fix check_reg_type for PTR_TO_BTF_ID

The current code does type matching for the case where reg->type is
PTR_TO_BTF_ID or has the PTR_TRUSTED flag. However, this only needs to
occur for non-MEM_ALLOC and non-MEM_PERCPU cases, but will include both
as per the current code.

The MEM_ALLOC case with or without PTR_TRUSTED needs to be handled
specially by the code for type_is_alloc case, while MEM_PERCPU case must
be ignored. Hence, to restore correct behavior and for clarity,
explicitly list out the handled PTR_TO_BTF_ID types which should be
handled for each case using a switch statement.

Helpers currently only take:
	PTR_TO_BTF_ID
	PTR_TO_BTF_ID | PTR_TRUSTED
	PTR_TO_BTF_ID | MEM_RCU
	PTR_TO_BTF_ID | MEM_ALLOC
	PTR_TO_BTF_ID | MEM_PERCPU
	PTR_TO_BTF_ID | MEM_PERCPU | PTR_TRUSTED

This fix was also described (for the MEM_ALLOC case) in [0].

  [0]: https://lore.kernel.org/bpf/20221121160657.h6z7xuvedybp5y7s@apollo

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230221200646.2500777-6-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 477c22c9bbd7..062fd4a05234 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -6527,7 +6527,14 @@ static int check_reg_type(struct bpf_verifier_env *env, u32 regno,
 	return -EACCES;
 
 found:
-	if (reg->type == PTR_TO_BTF_ID || reg->type & PTR_TRUSTED) {
+	if (base_type(reg->type) != PTR_TO_BTF_ID)
+		return 0;
+
+	switch ((int)reg->type) {
+	case PTR_TO_BTF_ID:
+	case PTR_TO_BTF_ID | PTR_TRUSTED:
+	case PTR_TO_BTF_ID | MEM_RCU:
+	{
 		/* For bpf_sk_release, it needs to match against first member
 		 * 'struct sock_common', hence make an exception for it. This
 		 * allows bpf_sk_release to work for multiple socket types.
@@ -6563,13 +6570,23 @@ found:
 				return -EACCES;
 			}
 		}
-	} else if (type_is_alloc(reg->type)) {
+		break;
+	}
+	case PTR_TO_BTF_ID | MEM_ALLOC:
 		if (meta->func_id != BPF_FUNC_spin_lock && meta->func_id != BPF_FUNC_spin_unlock) {
 			verbose(env, "verifier internal error: unimplemented handling of MEM_ALLOC\n");
 			return -EFAULT;
 		}
+		/* Handled by helper specific checks */
+		break;
+	case PTR_TO_BTF_ID | MEM_PERCPU:
+	case PTR_TO_BTF_ID | MEM_PERCPU | PTR_TRUSTED:
+		/* Handled by helper specific checks */
+		break;
+	default:
+		verbose(env, "verifier internal error: invalid PTR_TO_BTF_ID register for type match\n");
+		return -EFAULT;
 	}
-
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From dbd8d22863e83ee2834642e4cfd3bdacb8a1c975 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Tue, 21 Feb 2023 21:06:45 +0100
Subject: bpf: Wrap register invalidation with a helper

Typically, verifier should use env->allow_ptr_leaks when invaliding
registers for users that don't have CAP_PERFMON or CAP_SYS_ADMIN to
avoid leaking the pointer value. This is similar in spirit to
c67cae551f0d ("bpf: Tighten ptr_to_btf_id checks."). In a lot of the
existing checks, we know the capabilities are present, hence we don't do
the check.

Instead of being inconsistent in the application of the check, wrap the
action of invalidating a register into a helper named 'mark_invalid_reg'
and use it in a uniform fashion to replace open coded invalidation
operations, so that the check is always made regardless of the call site
and we don't have to remember whether it needs to be done or not for
each case.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230221200646.2500777-7-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 062fd4a05234..741cb5107536 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -895,6 +895,14 @@ static int unmark_stack_slots_dynptr(struct bpf_verifier_env *env, struct bpf_re
 static void __mark_reg_unknown(const struct bpf_verifier_env *env,
 			       struct bpf_reg_state *reg);
 
+static void mark_reg_invalid(const struct bpf_verifier_env *env, struct bpf_reg_state *reg)
+{
+	if (!env->allow_ptr_leaks)
+		__mark_reg_not_init(env, reg);
+	else
+		__mark_reg_unknown(env, reg);
+}
+
 static int destroy_if_dynptr_stack_slot(struct bpf_verifier_env *env,
 				        struct bpf_func_state *state, int spi)
 {
@@ -934,12 +942,8 @@ static int destroy_if_dynptr_stack_slot(struct bpf_verifier_env *env,
 		/* Dynptr slices are only PTR_TO_MEM_OR_NULL and PTR_TO_MEM */
 		if (dreg->type != (PTR_TO_MEM | PTR_MAYBE_NULL) && dreg->type != PTR_TO_MEM)
 			continue;
-		if (dreg->dynptr_id == dynptr_id) {
-			if (!env->allow_ptr_leaks)
-				__mark_reg_not_init(env, dreg);
-			else
-				__mark_reg_unknown(env, dreg);
-		}
+		if (dreg->dynptr_id == dynptr_id)
+			mark_reg_invalid(env, dreg);
 	}));
 
 	/* Do not release reference state, we are destroying dynptr on stack,
@@ -7384,7 +7388,7 @@ static void clear_all_pkt_pointers(struct bpf_verifier_env *env)
 
 	bpf_for_each_reg_in_vstate(env->cur_state, state, reg, ({
 		if (reg_is_pkt_pointer_any(reg))
-			__mark_reg_unknown(env, reg);
+			mark_reg_invalid(env, reg);
 	}));
 }
 
@@ -7429,12 +7433,8 @@ static int release_reference(struct bpf_verifier_env *env,
 		return err;
 
 	bpf_for_each_reg_in_vstate(env->cur_state, state, reg, ({
-		if (reg->ref_obj_id == ref_obj_id) {
-			if (!env->allow_ptr_leaks)
-				__mark_reg_not_init(env, reg);
-			else
-				__mark_reg_unknown(env, reg);
-		}
+		if (reg->ref_obj_id == ref_obj_id)
+			mark_reg_invalid(env, reg);
 	}));
 
 	return 0;
@@ -7447,7 +7447,7 @@ static void invalidate_non_owning_refs(struct bpf_verifier_env *env)
 
 	bpf_for_each_reg_in_vstate(env->cur_state, unused, reg, ({
 		if (type_is_non_owning_ref(reg->type))
-			__mark_reg_unknown(env, reg);
+			mark_reg_invalid(env, reg);
 	}));
 }
 
-- 
cgit v1.2.3-70-g09d2


From 5d5de3a431d87ac51d43da8d796891d014975ab7 Mon Sep 17 00:00:00 2001
From: Hou Tao <houtao1@huawei.com>
Date: Thu, 16 Feb 2023 10:48:21 +0800
Subject: bpf: Only allocate one bpf_mem_cache for bpf_cpumask_ma

The size of bpf_cpumask is fixed, so there is no need to allocate many
bpf_mem_caches for bpf_cpumask_ma, just one bpf_mem_cache is enough.
Also add comments for bpf_mem_alloc_init() in bpf_mem_alloc.h to prevent
future miuse.

Signed-off-by: Hou Tao <houtao1@huawei.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20230216024821.2202916-1-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_mem_alloc.h | 7 +++++++
 kernel/bpf/cpumask.c          | 6 +++---
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/include/linux/bpf_mem_alloc.h b/include/linux/bpf_mem_alloc.h
index 3e164b8efaa9..a7104af61ab4 100644
--- a/include/linux/bpf_mem_alloc.h
+++ b/include/linux/bpf_mem_alloc.h
@@ -14,6 +14,13 @@ struct bpf_mem_alloc {
 	struct work_struct work;
 };
 
+/* 'size != 0' is for bpf_mem_alloc which manages fixed-size objects.
+ * Alloc and free are done with bpf_mem_cache_{alloc,free}().
+ *
+ * 'size = 0' is for bpf_mem_alloc which manages many fixed-size objects.
+ * Alloc and free are done with bpf_mem_{alloc,free}() and the size of
+ * the returned object is given by the size argument of bpf_mem_alloc().
+ */
 int bpf_mem_alloc_init(struct bpf_mem_alloc *ma, int size, bool percpu);
 void bpf_mem_alloc_destroy(struct bpf_mem_alloc *ma);
 
diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c
index 52b981512a35..2b3fbbfebdc5 100644
--- a/kernel/bpf/cpumask.c
+++ b/kernel/bpf/cpumask.c
@@ -55,7 +55,7 @@ __bpf_kfunc struct bpf_cpumask *bpf_cpumask_create(void)
 	/* cpumask must be the first element so struct bpf_cpumask be cast to struct cpumask. */
 	BUILD_BUG_ON(offsetof(struct bpf_cpumask, cpumask) != 0);
 
-	cpumask = bpf_mem_alloc(&bpf_cpumask_ma, sizeof(*cpumask));
+	cpumask = bpf_mem_cache_alloc(&bpf_cpumask_ma);
 	if (!cpumask)
 		return NULL;
 
@@ -123,7 +123,7 @@ __bpf_kfunc void bpf_cpumask_release(struct bpf_cpumask *cpumask)
 
 	if (refcount_dec_and_test(&cpumask->usage)) {
 		migrate_disable();
-		bpf_mem_free(&bpf_cpumask_ma, cpumask);
+		bpf_mem_cache_free(&bpf_cpumask_ma, cpumask);
 		migrate_enable();
 	}
 }
@@ -468,7 +468,7 @@ static int __init cpumask_kfunc_init(void)
 		},
 	};
 
-	ret = bpf_mem_alloc_init(&bpf_cpumask_ma, 0, false);
+	ret = bpf_mem_alloc_init(&bpf_cpumask_ma, sizeof(struct bpf_cpumask), false);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_TRACING, &cpumask_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &cpumask_kfunc_set);
 	return  ret ?: register_btf_id_dtor_kfuncs(cpumask_dtors,
-- 
cgit v1.2.3-70-g09d2


From 1f265d2aea0dff1f2f9ecd41f545d601869462c5 Mon Sep 17 00:00:00 2001
From: Tiezhu Yang <yangtiezhu@loongson.cn>
Date: Thu, 16 Feb 2023 15:51:01 +0800
Subject: selftests/bpf: Remove not used headers

The following three uapi headers:

    tools/arch/arm64/include/uapi/asm/bpf_perf_event.h
    tools/arch/s390/include/uapi/asm/bpf_perf_event.h
    tools/arch/s390/include/uapi/asm/ptrace.h

were introduced in commit 618e165b2a8e ("selftests/bpf: sync kernel headers
and introduce arch support in Makefile"), they are not used any more after
commit 720f228e8d31 ("bpf: fix broken BPF selftest build"), so remove them.

Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Link: https://lore.kernel.org/r/1676533861-27508-1-git-send-email-yangtiezhu@loongson.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/arch/arm64/include/uapi/asm/bpf_perf_event.h |   9 -
 tools/arch/s390/include/uapi/asm/bpf_perf_event.h  |   9 -
 tools/arch/s390/include/uapi/asm/ptrace.h          | 458 ---------------------
 3 files changed, 476 deletions(-)
 delete mode 100644 tools/arch/arm64/include/uapi/asm/bpf_perf_event.h
 delete mode 100644 tools/arch/s390/include/uapi/asm/bpf_perf_event.h
 delete mode 100644 tools/arch/s390/include/uapi/asm/ptrace.h

diff --git a/tools/arch/arm64/include/uapi/asm/bpf_perf_event.h b/tools/arch/arm64/include/uapi/asm/bpf_perf_event.h
deleted file mode 100644
index b551b741653d..000000000000
--- a/tools/arch/arm64/include/uapi/asm/bpf_perf_event.h
+++ /dev/null
@@ -1,9 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _UAPI__ASM_BPF_PERF_EVENT_H__
-#define _UAPI__ASM_BPF_PERF_EVENT_H__
-
-#include <asm/ptrace.h>
-
-typedef struct user_pt_regs bpf_user_pt_regs_t;
-
-#endif /* _UAPI__ASM_BPF_PERF_EVENT_H__ */
diff --git a/tools/arch/s390/include/uapi/asm/bpf_perf_event.h b/tools/arch/s390/include/uapi/asm/bpf_perf_event.h
deleted file mode 100644
index 0a8e37a519f2..000000000000
--- a/tools/arch/s390/include/uapi/asm/bpf_perf_event.h
+++ /dev/null
@@ -1,9 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _UAPI__ASM_BPF_PERF_EVENT_H__
-#define _UAPI__ASM_BPF_PERF_EVENT_H__
-
-#include "ptrace.h"
-
-typedef user_pt_regs bpf_user_pt_regs_t;
-
-#endif /* _UAPI__ASM_BPF_PERF_EVENT_H__ */
diff --git a/tools/arch/s390/include/uapi/asm/ptrace.h b/tools/arch/s390/include/uapi/asm/ptrace.h
deleted file mode 100644
index ad64d673b5e6..000000000000
--- a/tools/arch/s390/include/uapi/asm/ptrace.h
+++ /dev/null
@@ -1,458 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
-/*
- *  S390 version
- *    Copyright IBM Corp. 1999, 2000
- *    Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com)
- */
-
-#ifndef _UAPI_S390_PTRACE_H
-#define _UAPI_S390_PTRACE_H
-
-/*
- * Offsets in the user_regs_struct. They are used for the ptrace
- * system call and in entry.S
- */
-#ifndef __s390x__
-
-#define PT_PSWMASK  0x00
-#define PT_PSWADDR  0x04
-#define PT_GPR0     0x08
-#define PT_GPR1     0x0C
-#define PT_GPR2     0x10
-#define PT_GPR3     0x14
-#define PT_GPR4     0x18
-#define PT_GPR5     0x1C
-#define PT_GPR6     0x20
-#define PT_GPR7     0x24
-#define PT_GPR8     0x28
-#define PT_GPR9     0x2C
-#define PT_GPR10    0x30
-#define PT_GPR11    0x34
-#define PT_GPR12    0x38
-#define PT_GPR13    0x3C
-#define PT_GPR14    0x40
-#define PT_GPR15    0x44
-#define PT_ACR0     0x48
-#define PT_ACR1     0x4C
-#define PT_ACR2     0x50
-#define PT_ACR3     0x54
-#define PT_ACR4	    0x58
-#define PT_ACR5	    0x5C
-#define PT_ACR6	    0x60
-#define PT_ACR7	    0x64
-#define PT_ACR8	    0x68
-#define PT_ACR9	    0x6C
-#define PT_ACR10    0x70
-#define PT_ACR11    0x74
-#define PT_ACR12    0x78
-#define PT_ACR13    0x7C
-#define PT_ACR14    0x80
-#define PT_ACR15    0x84
-#define PT_ORIGGPR2 0x88
-#define PT_FPC	    0x90
-/*
- * A nasty fact of life that the ptrace api
- * only supports passing of longs.
- */
-#define PT_FPR0_HI  0x98
-#define PT_FPR0_LO  0x9C
-#define PT_FPR1_HI  0xA0
-#define PT_FPR1_LO  0xA4
-#define PT_FPR2_HI  0xA8
-#define PT_FPR2_LO  0xAC
-#define PT_FPR3_HI  0xB0
-#define PT_FPR3_LO  0xB4
-#define PT_FPR4_HI  0xB8
-#define PT_FPR4_LO  0xBC
-#define PT_FPR5_HI  0xC0
-#define PT_FPR5_LO  0xC4
-#define PT_FPR6_HI  0xC8
-#define PT_FPR6_LO  0xCC
-#define PT_FPR7_HI  0xD0
-#define PT_FPR7_LO  0xD4
-#define PT_FPR8_HI  0xD8
-#define PT_FPR8_LO  0XDC
-#define PT_FPR9_HI  0xE0
-#define PT_FPR9_LO  0xE4
-#define PT_FPR10_HI 0xE8
-#define PT_FPR10_LO 0xEC
-#define PT_FPR11_HI 0xF0
-#define PT_FPR11_LO 0xF4
-#define PT_FPR12_HI 0xF8
-#define PT_FPR12_LO 0xFC
-#define PT_FPR13_HI 0x100
-#define PT_FPR13_LO 0x104
-#define PT_FPR14_HI 0x108
-#define PT_FPR14_LO 0x10C
-#define PT_FPR15_HI 0x110
-#define PT_FPR15_LO 0x114
-#define PT_CR_9	    0x118
-#define PT_CR_10    0x11C
-#define PT_CR_11    0x120
-#define PT_IEEE_IP  0x13C
-#define PT_LASTOFF  PT_IEEE_IP
-#define PT_ENDREGS  0x140-1
-
-#define GPR_SIZE	4
-#define CR_SIZE		4
-
-#define STACK_FRAME_OVERHEAD	96	/* size of minimum stack frame */
-
-#else /* __s390x__ */
-
-#define PT_PSWMASK  0x00
-#define PT_PSWADDR  0x08
-#define PT_GPR0     0x10
-#define PT_GPR1     0x18
-#define PT_GPR2     0x20
-#define PT_GPR3     0x28
-#define PT_GPR4     0x30
-#define PT_GPR5     0x38
-#define PT_GPR6     0x40
-#define PT_GPR7     0x48
-#define PT_GPR8     0x50
-#define PT_GPR9     0x58
-#define PT_GPR10    0x60
-#define PT_GPR11    0x68
-#define PT_GPR12    0x70
-#define PT_GPR13    0x78
-#define PT_GPR14    0x80
-#define PT_GPR15    0x88
-#define PT_ACR0     0x90
-#define PT_ACR1     0x94
-#define PT_ACR2     0x98
-#define PT_ACR3     0x9C
-#define PT_ACR4	    0xA0
-#define PT_ACR5	    0xA4
-#define PT_ACR6	    0xA8
-#define PT_ACR7	    0xAC
-#define PT_ACR8	    0xB0
-#define PT_ACR9	    0xB4
-#define PT_ACR10    0xB8
-#define PT_ACR11    0xBC
-#define PT_ACR12    0xC0
-#define PT_ACR13    0xC4
-#define PT_ACR14    0xC8
-#define PT_ACR15    0xCC
-#define PT_ORIGGPR2 0xD0
-#define PT_FPC	    0xD8
-#define PT_FPR0     0xE0
-#define PT_FPR1     0xE8
-#define PT_FPR2     0xF0
-#define PT_FPR3     0xF8
-#define PT_FPR4     0x100
-#define PT_FPR5     0x108
-#define PT_FPR6     0x110
-#define PT_FPR7     0x118
-#define PT_FPR8     0x120
-#define PT_FPR9     0x128
-#define PT_FPR10    0x130
-#define PT_FPR11    0x138
-#define PT_FPR12    0x140
-#define PT_FPR13    0x148
-#define PT_FPR14    0x150
-#define PT_FPR15    0x158
-#define PT_CR_9     0x160
-#define PT_CR_10    0x168
-#define PT_CR_11    0x170
-#define PT_IEEE_IP  0x1A8
-#define PT_LASTOFF  PT_IEEE_IP
-#define PT_ENDREGS  0x1B0-1
-
-#define GPR_SIZE	8
-#define CR_SIZE		8
-
-#define STACK_FRAME_OVERHEAD	160	 /* size of minimum stack frame */
-
-#endif /* __s390x__ */
-
-#define NUM_GPRS	16
-#define NUM_FPRS	16
-#define NUM_CRS		16
-#define NUM_ACRS	16
-
-#define NUM_CR_WORDS	3
-
-#define FPR_SIZE	8
-#define FPC_SIZE	4
-#define FPC_PAD_SIZE	4 /* gcc insists on aligning the fpregs */
-#define ACR_SIZE	4
-
-
-#define PTRACE_OLDSETOPTIONS		21
-#define PTRACE_SYSEMU			31
-#define PTRACE_SYSEMU_SINGLESTEP	32
-#ifndef __ASSEMBLY__
-#include <linux/stddef.h>
-#include <linux/types.h>
-
-typedef union {
-	float	f;
-	double	d;
-	__u64	ui;
-	struct
-	{
-		__u32 hi;
-		__u32 lo;
-	} fp;
-} freg_t;
-
-typedef struct {
-	__u32	fpc;
-	__u32	pad;
-	freg_t	fprs[NUM_FPRS];
-} s390_fp_regs;
-
-#define FPC_EXCEPTION_MASK	0xF8000000
-#define FPC_FLAGS_MASK		0x00F80000
-#define FPC_DXC_MASK		0x0000FF00
-#define FPC_RM_MASK		0x00000003
-
-/* this typedef defines how a Program Status Word looks like */
-typedef struct {
-	unsigned long mask;
-	unsigned long addr;
-} __attribute__ ((aligned(8))) psw_t;
-
-#ifndef __s390x__
-
-#define PSW_MASK_PER		0x40000000UL
-#define PSW_MASK_DAT		0x04000000UL
-#define PSW_MASK_IO		0x02000000UL
-#define PSW_MASK_EXT		0x01000000UL
-#define PSW_MASK_KEY		0x00F00000UL
-#define PSW_MASK_BASE		0x00080000UL	/* always one */
-#define PSW_MASK_MCHECK		0x00040000UL
-#define PSW_MASK_WAIT		0x00020000UL
-#define PSW_MASK_PSTATE		0x00010000UL
-#define PSW_MASK_ASC		0x0000C000UL
-#define PSW_MASK_CC		0x00003000UL
-#define PSW_MASK_PM		0x00000F00UL
-#define PSW_MASK_RI		0x00000000UL
-#define PSW_MASK_EA		0x00000000UL
-#define PSW_MASK_BA		0x00000000UL
-
-#define PSW_MASK_USER		0x0000FF00UL
-
-#define PSW_ADDR_AMODE		0x80000000UL
-#define PSW_ADDR_INSN		0x7FFFFFFFUL
-
-#define PSW_DEFAULT_KEY		(((unsigned long) PAGE_DEFAULT_ACC) << 20)
-
-#define PSW_ASC_PRIMARY		0x00000000UL
-#define PSW_ASC_ACCREG		0x00004000UL
-#define PSW_ASC_SECONDARY	0x00008000UL
-#define PSW_ASC_HOME		0x0000C000UL
-
-#else /* __s390x__ */
-
-#define PSW_MASK_PER		0x4000000000000000UL
-#define PSW_MASK_DAT		0x0400000000000000UL
-#define PSW_MASK_IO		0x0200000000000000UL
-#define PSW_MASK_EXT		0x0100000000000000UL
-#define PSW_MASK_BASE		0x0000000000000000UL
-#define PSW_MASK_KEY		0x00F0000000000000UL
-#define PSW_MASK_MCHECK		0x0004000000000000UL
-#define PSW_MASK_WAIT		0x0002000000000000UL
-#define PSW_MASK_PSTATE		0x0001000000000000UL
-#define PSW_MASK_ASC		0x0000C00000000000UL
-#define PSW_MASK_CC		0x0000300000000000UL
-#define PSW_MASK_PM		0x00000F0000000000UL
-#define PSW_MASK_RI		0x0000008000000000UL
-#define PSW_MASK_EA		0x0000000100000000UL
-#define PSW_MASK_BA		0x0000000080000000UL
-
-#define PSW_MASK_USER		0x0000FF0180000000UL
-
-#define PSW_ADDR_AMODE		0x0000000000000000UL
-#define PSW_ADDR_INSN		0xFFFFFFFFFFFFFFFFUL
-
-#define PSW_DEFAULT_KEY		(((unsigned long) PAGE_DEFAULT_ACC) << 52)
-
-#define PSW_ASC_PRIMARY		0x0000000000000000UL
-#define PSW_ASC_ACCREG		0x0000400000000000UL
-#define PSW_ASC_SECONDARY	0x0000800000000000UL
-#define PSW_ASC_HOME		0x0000C00000000000UL
-
-#endif /* __s390x__ */
-
-
-/*
- * The s390_regs structure is used to define the elf_gregset_t.
- */
-typedef struct {
-	psw_t psw;
-	unsigned long gprs[NUM_GPRS];
-	unsigned int  acrs[NUM_ACRS];
-	unsigned long orig_gpr2;
-} s390_regs;
-
-/*
- * The user_pt_regs structure exports the beginning of
- * the in-kernel pt_regs structure to user space.
- */
-typedef struct {
-	unsigned long args[1];
-	psw_t psw;
-	unsigned long gprs[NUM_GPRS];
-} user_pt_regs;
-
-/*
- * Now for the user space program event recording (trace) definitions.
- * The following structures are used only for the ptrace interface, don't
- * touch or even look at it if you don't want to modify the user-space
- * ptrace interface. In particular stay away from it for in-kernel PER.
- */
-typedef struct {
-	unsigned long cr[NUM_CR_WORDS];
-} per_cr_words;
-
-#define PER_EM_MASK 0xE8000000UL
-
-typedef struct {
-#ifdef __s390x__
-	unsigned		       : 32;
-#endif /* __s390x__ */
-	unsigned em_branching	       : 1;
-	unsigned em_instruction_fetch  : 1;
-	/*
-	 * Switching on storage alteration automatically fixes
-	 * the storage alteration event bit in the users std.
-	 */
-	unsigned em_storage_alteration : 1;
-	unsigned em_gpr_alt_unused     : 1;
-	unsigned em_store_real_address : 1;
-	unsigned		       : 3;
-	unsigned branch_addr_ctl       : 1;
-	unsigned		       : 1;
-	unsigned storage_alt_space_ctl : 1;
-	unsigned		       : 21;
-	unsigned long starting_addr;
-	unsigned long ending_addr;
-} per_cr_bits;
-
-typedef struct {
-	unsigned short perc_atmid;
-	unsigned long address;
-	unsigned char access_id;
-} per_lowcore_words;
-
-typedef struct {
-	unsigned perc_branching		 : 1;
-	unsigned perc_instruction_fetch  : 1;
-	unsigned perc_storage_alteration : 1;
-	unsigned perc_gpr_alt_unused	 : 1;
-	unsigned perc_store_real_address : 1;
-	unsigned			 : 3;
-	unsigned atmid_psw_bit_31	 : 1;
-	unsigned atmid_validity_bit	 : 1;
-	unsigned atmid_psw_bit_32	 : 1;
-	unsigned atmid_psw_bit_5	 : 1;
-	unsigned atmid_psw_bit_16	 : 1;
-	unsigned atmid_psw_bit_17	 : 1;
-	unsigned si			 : 2;
-	unsigned long address;
-	unsigned			 : 4;
-	unsigned access_id		 : 4;
-} per_lowcore_bits;
-
-typedef struct {
-	union {
-		per_cr_words   words;
-		per_cr_bits    bits;
-	} control_regs;
-	/*
-	 * The single_step and instruction_fetch bits are obsolete,
-	 * the kernel always sets them to zero. To enable single
-	 * stepping use ptrace(PTRACE_SINGLESTEP) instead.
-	 */
-	unsigned  single_step	    : 1;
-	unsigned  instruction_fetch : 1;
-	unsigned		    : 30;
-	/*
-	 * These addresses are copied into cr10 & cr11 if single
-	 * stepping is switched off
-	 */
-	unsigned long starting_addr;
-	unsigned long ending_addr;
-	union {
-		per_lowcore_words words;
-		per_lowcore_bits  bits;
-	} lowcore;
-} per_struct;
-
-typedef struct {
-	unsigned int  len;
-	unsigned long kernel_addr;
-	unsigned long process_addr;
-} ptrace_area;
-
-/*
- * S/390 specific non posix ptrace requests. I chose unusual values so
- * they are unlikely to clash with future ptrace definitions.
- */
-#define PTRACE_PEEKUSR_AREA	      0x5000
-#define PTRACE_POKEUSR_AREA	      0x5001
-#define PTRACE_PEEKTEXT_AREA	      0x5002
-#define PTRACE_PEEKDATA_AREA	      0x5003
-#define PTRACE_POKETEXT_AREA	      0x5004
-#define PTRACE_POKEDATA_AREA	      0x5005
-#define PTRACE_GET_LAST_BREAK	      0x5006
-#define PTRACE_PEEK_SYSTEM_CALL       0x5007
-#define PTRACE_POKE_SYSTEM_CALL	      0x5008
-#define PTRACE_ENABLE_TE	      0x5009
-#define PTRACE_DISABLE_TE	      0x5010
-#define PTRACE_TE_ABORT_RAND	      0x5011
-
-/*
- * The numbers chosen here are somewhat arbitrary but absolutely MUST
- * not overlap with any of the number assigned in <linux/ptrace.h>.
- */
-#define PTRACE_SINGLEBLOCK	12	/* resume execution until next branch */
-
-/*
- * PT_PROT definition is loosely based on hppa bsd definition in
- * gdb/hppab-nat.c
- */
-#define PTRACE_PROT			  21
-
-typedef enum {
-	ptprot_set_access_watchpoint,
-	ptprot_set_write_watchpoint,
-	ptprot_disable_watchpoint
-} ptprot_flags;
-
-typedef struct {
-	unsigned long lowaddr;
-	unsigned long hiaddr;
-	ptprot_flags prot;
-} ptprot_area;
-
-/* Sequence of bytes for breakpoint illegal instruction.  */
-#define S390_BREAKPOINT     {0x0,0x1}
-#define S390_BREAKPOINT_U16 ((__u16)0x0001)
-#define S390_SYSCALL_OPCODE ((__u16)0x0a00)
-#define S390_SYSCALL_SIZE   2
-
-/*
- * The user_regs_struct defines the way the user registers are
- * store on the stack for signal handling.
- */
-struct user_regs_struct {
-	psw_t psw;
-	unsigned long gprs[NUM_GPRS];
-	unsigned int  acrs[NUM_ACRS];
-	unsigned long orig_gpr2;
-	s390_fp_regs fp_regs;
-	/*
-	 * These per registers are in here so that gdb can modify them
-	 * itself as there is no "official" ptrace interface for hardware
-	 * watchpoints. This is the way intel does it.
-	 */
-	per_struct per_info;
-	unsigned long ieee_instruction_pointer;	/* obsolete, always 0 */
-};
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* _UAPI_S390_PTRACE_H */
-- 
cgit v1.2.3-70-g09d2


From b539a287baaa8501b3af4f7f99aba3c0b1d822f8 Mon Sep 17 00:00:00 2001
From: Florent Revest <revest@chromium.org>
Date: Fri, 17 Feb 2023 16:18:32 +0100
Subject: selftests/bpf: Fix cross compilation with CLANG_CROSS_FLAGS

I cross-compile my BPF selftests with the following command:

CLANG_CROSS_FLAGS="--target=aarch64-linux-gnu --sysroot=/sysroot/" \
  make LLVM=1 CC=clang CROSS_COMPILE=aarch64-linux-gnu- SRCARCH=arm64

(Note the use of CLANG_CROSS_FLAGS to specify a custom sysroot instead
of letting clang use gcc's default sysroot)

However, CLANG_CROSS_FLAGS gets propagated to host tools builds (libbpf
and bpftool) and because they reference it directly in their Makefiles,
they end up cross-compiling host objects which results in linking
errors.

This patch ensures that CLANG_CROSS_FLAGS is reset if CROSS_COMPILE
isn't set (for example when reaching a BPF host tool build).

Signed-off-by: Florent Revest <revest@chromium.org>
Link: https://lore.kernel.org/r/20230217151832.27784-1-revest@chromium.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/scripts/Makefile.include | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/scripts/Makefile.include b/tools/scripts/Makefile.include
index 0efb8f2b33ce..ff527ac065cf 100644
--- a/tools/scripts/Makefile.include
+++ b/tools/scripts/Makefile.include
@@ -108,6 +108,8 @@ endif # GCC_TOOLCHAIN_DIR
 endif # CLANG_CROSS_FLAGS
 CFLAGS += $(CLANG_CROSS_FLAGS)
 AFLAGS += $(CLANG_CROSS_FLAGS)
+else
+CLANG_CROSS_FLAGS :=
 endif # CROSS_COMPILE
 
 # Hack to avoid type-punned warnings on old systems such as RHEL5:
-- 
cgit v1.2.3-70-g09d2


From bb035ef0cc91e115faa80187ac8886a7f1914d06 Mon Sep 17 00:00:00 2001
From: Hengqi Chen <hengqi.chen@gmail.com>
Date: Sat, 18 Feb 2023 10:53:17 +0000
Subject: LoongArch: BPF: Support mixing bpf2bpf and tailcalls

The current implementation already allow such mixing.
Let's enable it in JIT.

Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Link: https://lore.kernel.org/r/20230218105317.4139666-1-hengqi.chen@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 arch/loongarch/net/bpf_jit.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/arch/loongarch/net/bpf_jit.c b/arch/loongarch/net/bpf_jit.c
index 288003a9f0ca..e70c846efaa1 100644
--- a/arch/loongarch/net/bpf_jit.c
+++ b/arch/loongarch/net/bpf_jit.c
@@ -1248,3 +1248,9 @@ out:
 
 	return prog;
 }
+
+/* Indicate the JIT backend supports mixing bpf2bpf and tailcalls. */
+bool bpf_jit_supports_subprog_tailcalls(void)
+{
+	return true;
+}
-- 
cgit v1.2.3-70-g09d2


From df2ccc180a2e6f6e4343ebee99dcfab4f8af2816 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Mon, 20 Feb 2023 17:37:56 +0100
Subject: bpf: Check for helper calls in check_subprogs()

The condition src_reg != BPF_PSEUDO_CALL && imm == BPF_FUNC_tail_call
may be satisfied by a kfunc call. This would lead to unnecessarily
setting has_tail_call. Use src_reg == 0 instead.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20230220163756.753713-1-iii@linux.ibm.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 741cb5107536..5cb8b623f639 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -2479,8 +2479,8 @@ static int check_subprogs(struct bpf_verifier_env *env)
 		u8 code = insn[i].code;
 
 		if (code == (BPF_JMP | BPF_CALL) &&
-		    insn[i].imm == BPF_FUNC_tail_call &&
-		    insn[i].src_reg != BPF_PSEUDO_CALL)
+		    insn[i].src_reg == 0 &&
+		    insn[i].imm == BPF_FUNC_tail_call)
 			subprog[cur_subprog].has_tail_call = true;
 		if (BPF_CLASS(code) == BPF_LD &&
 		    (BPF_MODE(code) == BPF_ABS || BPF_MODE(code) == BPF_IND))
-- 
cgit v1.2.3-70-g09d2


From d40c3847b485acc3522b62b020f77dcd38ca357f Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Tue, 21 Feb 2023 22:06:56 +0800
Subject: riscv, bpf: Add kfunc support for RV64
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds kernel function call support for RV64. Since the offset
from RV64 kernel and module functions to bpf programs is almost within
the range of s32, the current infrastructure of RV64 is already
sufficient for kfunc, so let's turn it on.

Suggested-by: Björn Töpel <bjorn@rivosinc.com>
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Acked-by: Björn Töpel <bjorn@rivosinc.com>
Link: https://lore.kernel.org/r/20230221140656.3480496-1-pulehui@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 arch/riscv/net/bpf_jit_comp64.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index f5a668736c79..a9270366dc57 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -1751,3 +1751,8 @@ void bpf_jit_build_epilogue(struct rv_jit_context *ctx)
 {
 	__build_epilogue(false, ctx);
 }
+
+bool bpf_jit_supports_kfunc_call(void)
+{
+	return true;
+}
-- 
cgit v1.2.3-70-g09d2


From 9fa02892857ae2b3b699630e5ede28f72106e7e7 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Tue, 21 Feb 2023 10:05:18 -0800
Subject: selftests/bpf: Fix BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL for empty
 flow label

Kernel's flow dissector continues to parse the packet when
the (optional) IPv6 flow label is empty even when instructed
to stop (via BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL). Do
the same in our reference BPF reimplementation.

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20230221180518.2139026-1-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/flow_dissector.c      | 24 ++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/bpf_flow.c       |  2 +-
 2 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/flow_dissector.c b/tools/testing/selftests/bpf/prog_tests/flow_dissector.c
index 7acca37a3d2b..c4773173a4e4 100644
--- a/tools/testing/selftests/bpf/prog_tests/flow_dissector.c
+++ b/tools/testing/selftests/bpf/prog_tests/flow_dissector.c
@@ -345,6 +345,30 @@ struct test tests[] = {
 		.flags = BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL,
 		.retval = BPF_OK,
 	},
+	{
+		.name = "ipv6-empty-flow-label",
+		.pkt.ipv6 = {
+			.eth.h_proto = __bpf_constant_htons(ETH_P_IPV6),
+			.iph.nexthdr = IPPROTO_TCP,
+			.iph.payload_len = __bpf_constant_htons(MAGIC_BYTES),
+			.iph.flow_lbl = { 0x00, 0x00, 0x00 },
+			.tcp.doff = 5,
+			.tcp.source = 80,
+			.tcp.dest = 8080,
+		},
+		.keys = {
+			.flags = BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL,
+			.nhoff = ETH_HLEN,
+			.thoff = ETH_HLEN + sizeof(struct ipv6hdr),
+			.addr_proto = ETH_P_IPV6,
+			.ip_proto = IPPROTO_TCP,
+			.n_proto = __bpf_constant_htons(ETH_P_IPV6),
+			.sport = 80,
+			.dport = 8080,
+		},
+		.flags = BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL,
+		.retval = BPF_OK,
+	},
 	{
 		.name = "ipip-encap",
 		.pkt.ipip = {
diff --git a/tools/testing/selftests/bpf/progs/bpf_flow.c b/tools/testing/selftests/bpf/progs/bpf_flow.c
index a20c5ed5e454..b04e092fac94 100644
--- a/tools/testing/selftests/bpf/progs/bpf_flow.c
+++ b/tools/testing/selftests/bpf/progs/bpf_flow.c
@@ -337,7 +337,7 @@ PROG(IPV6)(struct __sk_buff *skb)
 	keys->ip_proto = ip6h->nexthdr;
 	keys->flow_label = ip6_flowlabel(ip6h);
 
-	if (keys->flags & BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL)
+	if (keys->flow_label && keys->flags & BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL)
 		return export_flow_keys(keys, BPF_OK);
 
 	return parse_ipv6_proto(skb, ip6h->nexthdr);
-- 
cgit v1.2.3-70-g09d2


From 746ce767128598711a00d8df5713d4c3b3d9e9a7 Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler@microsoft.com>
Date: Mon, 20 Feb 2023 22:37:42 +0000
Subject: bpf, docs: Add explanation of endianness

Document the discussion from the email thread on the IETF bpf list,
where it was explained that the raw format varies by endianness
of the processor.

Signed-off-by: Dave Thaler <dthaler@microsoft.com>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230220223742.1347-1-dthaler1968@googlemail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/instruction-set.rst | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/Documentation/bpf/instruction-set.rst b/Documentation/bpf/instruction-set.rst
index af515de5fc38..01802ed9b29b 100644
--- a/Documentation/bpf/instruction-set.rst
+++ b/Documentation/bpf/instruction-set.rst
@@ -38,8 +38,9 @@ eBPF has two instruction encodings:
 * the wide instruction encoding, which appends a second 64-bit immediate (i.e.,
   constant) value after the basic instruction for a total of 128 bits.
 
-The basic instruction encoding is as follows, where MSB and LSB mean the most significant
-bits and least significant bits, respectively:
+The basic instruction encoding looks as follows for a little-endian processor,
+where MSB and LSB mean the most significant bits and least significant bits,
+respectively:
 
 =============  =======  =======  =======  ============
 32 bits (MSB)  16 bits  4 bits   4 bits   8 bits (LSB)
@@ -63,6 +64,17 @@ imm            offset   src_reg  dst_reg  opcode
 **opcode**
   operation to perform
 
+and as follows for a big-endian processor:
+
+=============  =======  =======  =======  ============
+32 bits (MSB)  16 bits  4 bits   4 bits   8 bits (LSB)
+=============  =======  =======  =======  ============
+imm            offset   dst_reg  src_reg  opcode
+=============  =======  =======  =======  ============
+
+Multi-byte fields ('imm' and 'offset') are similarly stored in
+the byte order of the processor.
+
 Note that most instructions do not use all of the fields.
 Unused fields shall be cleared to zero.
 
-- 
cgit v1.2.3-70-g09d2


From 332ea1f697be148bd5e66475d82b5ecc5084da65 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 22 Feb 2023 15:29:12 -1000
Subject: bpf: Add bpf_cgroup_from_id() kfunc

cgroup ID is an userspace-visible 64bit value uniquely identifying a given
cgroup. As the IDs are used widely, it's useful to be able to look up the
matching cgroups. Add bpf_cgroup_from_id().

v2: Separate out selftest into its own patch as suggested by Alexei.

Signed-off-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/Y/bBaG96t0/gQl9/@slm.duckdns.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst | 10 +++++++---
 kernel/bpf/helpers.c         | 18 ++++++++++++++++++
 2 files changed, 25 insertions(+), 3 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index ca96ef3f6896..226313747be5 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -583,13 +583,17 @@ Here's an example of how it can be used:
 
 ----
 
-Another kfunc available for interacting with ``struct cgroup *`` objects is
-bpf_cgroup_ancestor(). This allows callers to access the ancestor of a cgroup,
-and return it as a cgroup kptr.
+Other kfuncs available for interacting with ``struct cgroup *`` objects are
+bpf_cgroup_ancestor() and bpf_cgroup_from_id(), allowing callers to access
+the ancestor of a cgroup and find a cgroup by its ID, respectively. Both
+return a cgroup kptr.
 
 .. kernel-doc:: kernel/bpf/helpers.c
    :identifiers: bpf_cgroup_ancestor
 
+.. kernel-doc:: kernel/bpf/helpers.c
+   :identifiers: bpf_cgroup_from_id
+
 Eventually, BPF should be updated to allow this to happen with a normal memory
 load in the program itself. This is currently not possible without more work in
 the verifier. bpf_cgroup_ancestor() can be used as follows:
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 5b278a38ae58..a784be6f8bac 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2101,6 +2101,23 @@ __bpf_kfunc struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level)
 	cgroup_get(ancestor);
 	return ancestor;
 }
+
+/**
+ * bpf_cgroup_from_id - Find a cgroup from its ID. A cgroup returned by this
+ * kfunc which is not subsequently stored in a map, must be released by calling
+ * bpf_cgroup_release().
+ * @cgrp: The cgroup for which we're performing a lookup.
+ * @level: The level of ancestor to look up.
+ */
+__bpf_kfunc struct cgroup *bpf_cgroup_from_id(u64 cgid)
+{
+	struct cgroup *cgrp;
+
+	cgrp = cgroup_get_from_id(cgid);
+	if (IS_ERR(cgrp))
+		return NULL;
+	return cgrp;
+}
 #endif /* CONFIG_CGROUPS */
 
 /**
@@ -2167,6 +2184,7 @@ BTF_ID_FLAGS(func, bpf_cgroup_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_cgroup_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_release, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_cgroup_ancestor, KF_ACQUIRE | KF_TRUSTED_ARGS | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_cgroup_from_id, KF_ACQUIRE | KF_RET_NULL)
 #endif
 BTF_ID_FLAGS(func, bpf_task_from_pid, KF_ACQUIRE | KF_RET_NULL)
 BTF_SET8_END(generic_btf_ids)
-- 
cgit v1.2.3-70-g09d2


From d0093aaefa35b80990c05a424dad2396ba4549d7 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 22 Feb 2023 15:29:58 -1000
Subject: selftests/bpf: Add a test case for bpf_cgroup_from_id()

Add a test case for bpf_cgroup_from_id.

Signed-off-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/Y/bBlt+tPozcQgws@slm.duckdns.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/cgrp_kfunc.c  |  1 +
 .../selftests/bpf/progs/cgrp_kfunc_common.h        |  1 +
 .../selftests/bpf/progs/cgrp_kfunc_success.c       | 42 ++++++++++++++++++++++
 3 files changed, 44 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/cgrp_kfunc.c b/tools/testing/selftests/bpf/prog_tests/cgrp_kfunc.c
index b3f7985c8504..adda85f97058 100644
--- a/tools/testing/selftests/bpf/prog_tests/cgrp_kfunc.c
+++ b/tools/testing/selftests/bpf/prog_tests/cgrp_kfunc.c
@@ -84,6 +84,7 @@ static const char * const success_tests[] = {
 	"test_cgrp_xchg_release",
 	"test_cgrp_get_release",
 	"test_cgrp_get_ancestors",
+	"test_cgrp_from_id",
 };
 
 void test_cgrp_kfunc(void)
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
index 7d30855bfe78..2f8de933b957 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
@@ -24,6 +24,7 @@ struct cgroup *bpf_cgroup_acquire(struct cgroup *p) __ksym;
 struct cgroup *bpf_cgroup_kptr_get(struct cgroup **pp) __ksym;
 void bpf_cgroup_release(struct cgroup *p) __ksym;
 struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level) __ksym;
+struct cgroup *bpf_cgroup_from_id(u64 cgid) __ksym;
 
 static inline struct __cgrps_kfunc_map_value *cgrps_kfunc_map_value_lookup(struct cgroup *cgrp)
 {
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
index 0c23ea32df9f..42e13aebdd62 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
@@ -168,3 +168,45 @@ int BPF_PROG(test_cgrp_get_ancestors, struct cgroup *cgrp, const char *path)
 
 	return 0;
 }
+
+SEC("tp_btf/cgroup_mkdir")
+int BPF_PROG(test_cgrp_from_id, struct cgroup *cgrp, const char *path)
+{
+	struct cgroup *parent, *res;
+	u64 parent_cgid;
+
+	if (!is_test_kfunc_task())
+		return 0;
+
+	/* @cgrp's ID is not visible yet, let's test with the parent */
+	parent = bpf_cgroup_ancestor(cgrp, cgrp->level - 1);
+	if (!parent) {
+		err = 1;
+		return 0;
+	}
+
+	parent_cgid = parent->kn->id;
+	bpf_cgroup_release(parent);
+
+	res = bpf_cgroup_from_id(parent_cgid);
+	if (!res) {
+		err = 2;
+		return 0;
+	}
+
+	bpf_cgroup_release(res);
+
+	if (res != parent) {
+		err = 3;
+		return 0;
+	}
+
+	res = bpf_cgroup_from_id((u64)-1);
+	if (res) {
+		bpf_cgroup_release(res);
+		err = 4;
+		return 0;
+	}
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From 4a51e66fe96dfde76894c4eb445ef1b14d312014 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Thu, 2 Feb 2023 21:58:10 +0100
Subject: wifi: wcn36xx: Slightly optimize PREPARE_HAL_BUF()

In most (likely all) cases, INIT_HAL_MSG() is called before
PREPARE_HAL_BUF().
In such cases calling memset() is useless because:
   msg_body.header.len = sizeof(msg_body)

So, instead of writing twice the memory, we just have a sanity check to
make sure that some potential trailing memory is zeroed.
It even gives the opportunity to see that by itself and optimize it away.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Acked-by: Loic Poulain <loic.poulain@linaro.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/7d8ab7fee45222cdbaf80c507525f2d3941587c1.1675371372.git.christophe.jaillet@wanadoo.fr
---
 drivers/net/wireless/ath/wcn36xx/smd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c
index 566f0b9c1584..17e1919d1cd8 100644
--- a/drivers/net/wireless/ath/wcn36xx/smd.c
+++ b/drivers/net/wireless/ath/wcn36xx/smd.c
@@ -475,8 +475,8 @@ out:
 
 #define PREPARE_HAL_BUF(send_buf, msg_body) \
 	do {							\
-		memset(send_buf, 0, msg_body.header.len);	\
-		memcpy(send_buf, &msg_body, sizeof(msg_body));	\
+		memcpy_and_pad(send_buf, msg_body.header.len,	\
+			       &msg_body, sizeof(msg_body), 0);	\
 	} while (0)						\
 
 #define PREPARE_HAL_PTT_MSG_BUF(send_buf, p_msg_body) \
-- 
cgit v1.2.3-70-g09d2


From 28013c35629647679e10cb1bbf36c3feec2af756 Mon Sep 17 00:00:00 2001
From: Kalle Valo <quic_kvalo@quicinc.com>
Date: Wed, 22 Feb 2023 18:40:14 +0200
Subject: wifi: ath12k: remove memset with byte count of 278528

Sparse warns:

drivers/net/wireless/ath/ath12k/dp.c:1471:15: warning: memset with byte count of 278528

There's no need to use memset() here, instead call dma_alloc_coherent() with __GFP_ZERO.

While at it, remove an extra line before the error handler.

Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230222164014.860-1-kvalo@kernel.org
---
 drivers/net/wireless/ath/ath12k/dp.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index 9926d81c5fe4..ae1645d0f42a 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -1461,15 +1461,12 @@ static int ath12k_dp_reoq_lut_setup(struct ath12k_base *ab)
 	dp->reoq_lut.vaddr = dma_alloc_coherent(ab->dev,
 						DP_REOQ_LUT_SIZE,
 						&dp->reoq_lut.paddr,
-						GFP_KERNEL);
-
+						GFP_KERNEL | __GFP_ZERO);
 	if (!dp->reoq_lut.vaddr) {
 		ath12k_warn(ab, "failed to allocate memory for reoq table");
 		return -ENOMEM;
 	}
 
-	memset(dp->reoq_lut.vaddr, 0, DP_REOQ_LUT_SIZE);
-
 	ath12k_hif_write32(ab, HAL_SEQ_WCSS_UMAC_REO_REG + HAL_REO1_QDESC_LUT_BASE0(ab),
 			   dp->reoq_lut.paddr);
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From b61987d37cbee3c44e80304598c60b163553926b Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Fri, 24 Feb 2023 14:13:42 +0800
Subject: selftests/bpf: move SYS() macro into the test_progs.h

A lot of tests defined SYS() macro to run system calls with goto label.
Let's move this macro to test_progs.h and add configurable
"goto_label" as the first arg.

Suggested-by: Martin KaFai Lau <martin.lau@linux.dev>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20230224061343.506571-2-liuhangbin@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 .../selftests/bpf/prog_tests/decap_sanity.c        |  16 +---
 tools/testing/selftests/bpf/prog_tests/empty_skb.c |  25 +++---
 .../testing/selftests/bpf/prog_tests/fib_lookup.c  |  28 +++---
 .../testing/selftests/bpf/prog_tests/tc_redirect.c | 100 ++++++++++-----------
 .../testing/selftests/bpf/prog_tests/test_tunnel.c |  71 ++++++---------
 .../testing/selftests/bpf/prog_tests/xdp_bonding.c |  40 ++++-----
 .../selftests/bpf/prog_tests/xdp_do_redirect.c     |  30 +++----
 .../selftests/bpf/prog_tests/xdp_metadata.c        |  23 ++---
 .../selftests/bpf/prog_tests/xdp_synproxy.c        |  41 ++++-----
 tools/testing/selftests/bpf/prog_tests/xfrm_info.c |  67 ++++++--------
 tools/testing/selftests/bpf/test_progs.h           |  15 ++++
 11 files changed, 193 insertions(+), 263 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/decap_sanity.c b/tools/testing/selftests/bpf/prog_tests/decap_sanity.c
index 2853883b7cbb..5c0ebe6ba866 100644
--- a/tools/testing/selftests/bpf/prog_tests/decap_sanity.c
+++ b/tools/testing/selftests/bpf/prog_tests/decap_sanity.c
@@ -10,14 +10,6 @@
 #include "network_helpers.h"
 #include "decap_sanity.skel.h"
 
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			goto fail;				\
-	})
-
 #define NS_TEST "decap_sanity_ns"
 #define IPV6_IFACE_ADDR "face::1"
 #define UDP_TEST_PORT 7777
@@ -37,9 +29,9 @@ void test_decap_sanity(void)
 	if (!ASSERT_OK_PTR(skel, "skel open_and_load"))
 		return;
 
-	SYS("ip netns add %s", NS_TEST);
-	SYS("ip -net %s -6 addr add %s/128 dev lo nodad", NS_TEST, IPV6_IFACE_ADDR);
-	SYS("ip -net %s link set dev lo up", NS_TEST);
+	SYS(fail, "ip netns add %s", NS_TEST);
+	SYS(fail, "ip -net %s -6 addr add %s/128 dev lo nodad", NS_TEST, IPV6_IFACE_ADDR);
+	SYS(fail, "ip -net %s link set dev lo up", NS_TEST);
 
 	nstoken = open_netns(NS_TEST);
 	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
@@ -80,6 +72,6 @@ fail:
 		bpf_tc_hook_destroy(&qdisc_hook);
 		close_netns(nstoken);
 	}
-	system("ip netns del " NS_TEST " &> /dev/null");
+	SYS_NOFAIL("ip netns del " NS_TEST " &> /dev/null");
 	decap_sanity__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/empty_skb.c b/tools/testing/selftests/bpf/prog_tests/empty_skb.c
index 32dd731e9070..3b77d8a422db 100644
--- a/tools/testing/selftests/bpf/prog_tests/empty_skb.c
+++ b/tools/testing/selftests/bpf/prog_tests/empty_skb.c
@@ -4,11 +4,6 @@
 #include <net/if.h>
 #include "empty_skb.skel.h"
 
-#define SYS(cmd) ({ \
-	if (!ASSERT_OK(system(cmd), (cmd))) \
-		goto out; \
-})
-
 void test_empty_skb(void)
 {
 	LIBBPF_OPTS(bpf_test_run_opts, tattr);
@@ -93,18 +88,18 @@ void test_empty_skb(void)
 		},
 	};
 
-	SYS("ip netns add empty_skb");
+	SYS(out, "ip netns add empty_skb");
 	tok = open_netns("empty_skb");
-	SYS("ip link add veth0 type veth peer veth1");
-	SYS("ip link set dev veth0 up");
-	SYS("ip link set dev veth1 up");
-	SYS("ip addr add 10.0.0.1/8 dev veth0");
-	SYS("ip addr add 10.0.0.2/8 dev veth1");
+	SYS(out, "ip link add veth0 type veth peer veth1");
+	SYS(out, "ip link set dev veth0 up");
+	SYS(out, "ip link set dev veth1 up");
+	SYS(out, "ip addr add 10.0.0.1/8 dev veth0");
+	SYS(out, "ip addr add 10.0.0.2/8 dev veth1");
 	veth_ifindex = if_nametoindex("veth0");
 
-	SYS("ip link add ipip0 type ipip local 10.0.0.1 remote 10.0.0.2");
-	SYS("ip link set ipip0 up");
-	SYS("ip addr add 192.168.1.1/16 dev ipip0");
+	SYS(out, "ip link add ipip0 type ipip local 10.0.0.1 remote 10.0.0.2");
+	SYS(out, "ip link set ipip0 up");
+	SYS(out, "ip addr add 192.168.1.1/16 dev ipip0");
 	ipip_ifindex = if_nametoindex("ipip0");
 
 	bpf_obj = empty_skb__open_and_load();
@@ -142,5 +137,5 @@ out:
 		empty_skb__destroy(bpf_obj);
 	if (tok)
 		close_netns(tok);
-	system("ip netns del empty_skb");
+	SYS_NOFAIL("ip netns del empty_skb");
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/fib_lookup.c b/tools/testing/selftests/bpf/prog_tests/fib_lookup.c
index 61ccddccf485..429393caf612 100644
--- a/tools/testing/selftests/bpf/prog_tests/fib_lookup.c
+++ b/tools/testing/selftests/bpf/prog_tests/fib_lookup.c
@@ -8,14 +8,6 @@
 #include "network_helpers.h"
 #include "fib_lookup.skel.h"
 
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			goto fail;				\
-	})
-
 #define NS_TEST			"fib_lookup_ns"
 #define IPV6_IFACE_ADDR		"face::face"
 #define IPV6_NUD_FAILED_ADDR	"face::1"
@@ -59,16 +51,16 @@ static int setup_netns(void)
 {
 	int err;
 
-	SYS("ip link add veth1 type veth peer name veth2");
-	SYS("ip link set dev veth1 up");
+	SYS(fail, "ip link add veth1 type veth peer name veth2");
+	SYS(fail, "ip link set dev veth1 up");
 
-	SYS("ip addr add %s/64 dev veth1 nodad", IPV6_IFACE_ADDR);
-	SYS("ip neigh add %s dev veth1 nud failed", IPV6_NUD_FAILED_ADDR);
-	SYS("ip neigh add %s dev veth1 lladdr %s nud stale", IPV6_NUD_STALE_ADDR, DMAC);
+	SYS(fail, "ip addr add %s/64 dev veth1 nodad", IPV6_IFACE_ADDR);
+	SYS(fail, "ip neigh add %s dev veth1 nud failed", IPV6_NUD_FAILED_ADDR);
+	SYS(fail, "ip neigh add %s dev veth1 lladdr %s nud stale", IPV6_NUD_STALE_ADDR, DMAC);
 
-	SYS("ip addr add %s/24 dev veth1 nodad", IPV4_IFACE_ADDR);
-	SYS("ip neigh add %s dev veth1 nud failed", IPV4_NUD_FAILED_ADDR);
-	SYS("ip neigh add %s dev veth1 lladdr %s nud stale", IPV4_NUD_STALE_ADDR, DMAC);
+	SYS(fail, "ip addr add %s/24 dev veth1 nodad", IPV4_IFACE_ADDR);
+	SYS(fail, "ip neigh add %s dev veth1 nud failed", IPV4_NUD_FAILED_ADDR);
+	SYS(fail, "ip neigh add %s dev veth1 lladdr %s nud stale", IPV4_NUD_STALE_ADDR, DMAC);
 
 	err = write_sysctl("/proc/sys/net/ipv4/conf/veth1/forwarding", "1");
 	if (!ASSERT_OK(err, "write_sysctl(net.ipv4.conf.veth1.forwarding)"))
@@ -140,7 +132,7 @@ void test_fib_lookup(void)
 		return;
 	prog_fd = bpf_program__fd(skel->progs.fib_lookup);
 
-	SYS("ip netns add %s", NS_TEST);
+	SYS(fail, "ip netns add %s", NS_TEST);
 
 	nstoken = open_netns(NS_TEST);
 	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
@@ -182,6 +174,6 @@ void test_fib_lookup(void)
 fail:
 	if (nstoken)
 		close_netns(nstoken);
-	system("ip netns del " NS_TEST " &> /dev/null");
+	SYS_NOFAIL("ip netns del " NS_TEST " &> /dev/null");
 	fib_lookup__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/tc_redirect.c b/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
index bca5e6839ac4..6ee22c3b251a 100644
--- a/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
@@ -137,24 +137,16 @@ static int get_ifaddr(const char *name, char *ifaddr)
 	return 0;
 }
 
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			goto fail;				\
-	})
-
 static int netns_setup_links_and_routes(struct netns_setup_result *result)
 {
 	struct nstoken *nstoken = NULL;
 	char veth_src_fwd_addr[IFADDR_STR_LEN+1] = {};
 
-	SYS("ip link add veth_src type veth peer name veth_src_fwd");
-	SYS("ip link add veth_dst type veth peer name veth_dst_fwd");
+	SYS(fail, "ip link add veth_src type veth peer name veth_src_fwd");
+	SYS(fail, "ip link add veth_dst type veth peer name veth_dst_fwd");
 
-	SYS("ip link set veth_dst_fwd address " MAC_DST_FWD);
-	SYS("ip link set veth_dst address " MAC_DST);
+	SYS(fail, "ip link set veth_dst_fwd address " MAC_DST_FWD);
+	SYS(fail, "ip link set veth_dst address " MAC_DST);
 
 	if (get_ifaddr("veth_src_fwd", veth_src_fwd_addr))
 		goto fail;
@@ -175,27 +167,27 @@ static int netns_setup_links_and_routes(struct netns_setup_result *result)
 	if (!ASSERT_GT(result->ifindex_veth_dst_fwd, 0, "ifindex_veth_dst_fwd"))
 		goto fail;
 
-	SYS("ip link set veth_src netns " NS_SRC);
-	SYS("ip link set veth_src_fwd netns " NS_FWD);
-	SYS("ip link set veth_dst_fwd netns " NS_FWD);
-	SYS("ip link set veth_dst netns " NS_DST);
+	SYS(fail, "ip link set veth_src netns " NS_SRC);
+	SYS(fail, "ip link set veth_src_fwd netns " NS_FWD);
+	SYS(fail, "ip link set veth_dst_fwd netns " NS_FWD);
+	SYS(fail, "ip link set veth_dst netns " NS_DST);
 
 	/** setup in 'src' namespace */
 	nstoken = open_netns(NS_SRC);
 	if (!ASSERT_OK_PTR(nstoken, "setns src"))
 		goto fail;
 
-	SYS("ip addr add " IP4_SRC "/32 dev veth_src");
-	SYS("ip addr add " IP6_SRC "/128 dev veth_src nodad");
-	SYS("ip link set dev veth_src up");
+	SYS(fail, "ip addr add " IP4_SRC "/32 dev veth_src");
+	SYS(fail, "ip addr add " IP6_SRC "/128 dev veth_src nodad");
+	SYS(fail, "ip link set dev veth_src up");
 
-	SYS("ip route add " IP4_DST "/32 dev veth_src scope global");
-	SYS("ip route add " IP4_NET "/16 dev veth_src scope global");
-	SYS("ip route add " IP6_DST "/128 dev veth_src scope global");
+	SYS(fail, "ip route add " IP4_DST "/32 dev veth_src scope global");
+	SYS(fail, "ip route add " IP4_NET "/16 dev veth_src scope global");
+	SYS(fail, "ip route add " IP6_DST "/128 dev veth_src scope global");
 
-	SYS("ip neigh add " IP4_DST " dev veth_src lladdr %s",
+	SYS(fail, "ip neigh add " IP4_DST " dev veth_src lladdr %s",
 	    veth_src_fwd_addr);
-	SYS("ip neigh add " IP6_DST " dev veth_src lladdr %s",
+	SYS(fail, "ip neigh add " IP6_DST " dev veth_src lladdr %s",
 	    veth_src_fwd_addr);
 
 	close_netns(nstoken);
@@ -209,15 +201,15 @@ static int netns_setup_links_and_routes(struct netns_setup_result *result)
 	 * needs v4 one in order to start ARP probing. IP4_NET route is added
 	 * to the endpoints so that the ARP processing will reply.
 	 */
-	SYS("ip addr add " IP4_SLL "/32 dev veth_src_fwd");
-	SYS("ip addr add " IP4_DLL "/32 dev veth_dst_fwd");
-	SYS("ip link set dev veth_src_fwd up");
-	SYS("ip link set dev veth_dst_fwd up");
+	SYS(fail, "ip addr add " IP4_SLL "/32 dev veth_src_fwd");
+	SYS(fail, "ip addr add " IP4_DLL "/32 dev veth_dst_fwd");
+	SYS(fail, "ip link set dev veth_src_fwd up");
+	SYS(fail, "ip link set dev veth_dst_fwd up");
 
-	SYS("ip route add " IP4_SRC "/32 dev veth_src_fwd scope global");
-	SYS("ip route add " IP6_SRC "/128 dev veth_src_fwd scope global");
-	SYS("ip route add " IP4_DST "/32 dev veth_dst_fwd scope global");
-	SYS("ip route add " IP6_DST "/128 dev veth_dst_fwd scope global");
+	SYS(fail, "ip route add " IP4_SRC "/32 dev veth_src_fwd scope global");
+	SYS(fail, "ip route add " IP6_SRC "/128 dev veth_src_fwd scope global");
+	SYS(fail, "ip route add " IP4_DST "/32 dev veth_dst_fwd scope global");
+	SYS(fail, "ip route add " IP6_DST "/128 dev veth_dst_fwd scope global");
 
 	close_netns(nstoken);
 
@@ -226,16 +218,16 @@ static int netns_setup_links_and_routes(struct netns_setup_result *result)
 	if (!ASSERT_OK_PTR(nstoken, "setns dst"))
 		goto fail;
 
-	SYS("ip addr add " IP4_DST "/32 dev veth_dst");
-	SYS("ip addr add " IP6_DST "/128 dev veth_dst nodad");
-	SYS("ip link set dev veth_dst up");
+	SYS(fail, "ip addr add " IP4_DST "/32 dev veth_dst");
+	SYS(fail, "ip addr add " IP6_DST "/128 dev veth_dst nodad");
+	SYS(fail, "ip link set dev veth_dst up");
 
-	SYS("ip route add " IP4_SRC "/32 dev veth_dst scope global");
-	SYS("ip route add " IP4_NET "/16 dev veth_dst scope global");
-	SYS("ip route add " IP6_SRC "/128 dev veth_dst scope global");
+	SYS(fail, "ip route add " IP4_SRC "/32 dev veth_dst scope global");
+	SYS(fail, "ip route add " IP4_NET "/16 dev veth_dst scope global");
+	SYS(fail, "ip route add " IP6_SRC "/128 dev veth_dst scope global");
 
-	SYS("ip neigh add " IP4_SRC " dev veth_dst lladdr " MAC_DST_FWD);
-	SYS("ip neigh add " IP6_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+	SYS(fail, "ip neigh add " IP4_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+	SYS(fail, "ip neigh add " IP6_SRC " dev veth_dst lladdr " MAC_DST_FWD);
 
 	close_netns(nstoken);
 
@@ -375,7 +367,7 @@ done:
 
 static int test_ping(int family, const char *addr)
 {
-	SYS("ip netns exec " NS_SRC " %s " PING_ARGS " %s > /dev/null", ping_command(family), addr);
+	SYS(fail, "ip netns exec " NS_SRC " %s " PING_ARGS " %s > /dev/null", ping_command(family), addr);
 	return 0;
 fail:
 	return -1;
@@ -953,7 +945,7 @@ static int tun_open(char *name)
 	if (!ASSERT_OK(err, "ioctl TUNSETIFF"))
 		goto fail;
 
-	SYS("ip link set dev %s up", name);
+	SYS(fail, "ip link set dev %s up", name);
 
 	return fd;
 fail:
@@ -1076,23 +1068,23 @@ static void test_tc_redirect_peer_l3(struct netns_setup_result *setup_result)
 	XGRESS_FILTER_ADD(&qdisc_veth_dst_fwd, BPF_TC_EGRESS, skel->progs.tc_chk, 0);
 
 	/* Setup route and neigh tables */
-	SYS("ip -netns " NS_SRC " addr add dev tun_src " IP4_TUN_SRC "/24");
-	SYS("ip -netns " NS_FWD " addr add dev tun_fwd " IP4_TUN_FWD "/24");
+	SYS(fail, "ip -netns " NS_SRC " addr add dev tun_src " IP4_TUN_SRC "/24");
+	SYS(fail, "ip -netns " NS_FWD " addr add dev tun_fwd " IP4_TUN_FWD "/24");
 
-	SYS("ip -netns " NS_SRC " addr add dev tun_src " IP6_TUN_SRC "/64 nodad");
-	SYS("ip -netns " NS_FWD " addr add dev tun_fwd " IP6_TUN_FWD "/64 nodad");
+	SYS(fail, "ip -netns " NS_SRC " addr add dev tun_src " IP6_TUN_SRC "/64 nodad");
+	SYS(fail, "ip -netns " NS_FWD " addr add dev tun_fwd " IP6_TUN_FWD "/64 nodad");
 
-	SYS("ip -netns " NS_SRC " route del " IP4_DST "/32 dev veth_src scope global");
-	SYS("ip -netns " NS_SRC " route add " IP4_DST "/32 via " IP4_TUN_FWD
+	SYS(fail, "ip -netns " NS_SRC " route del " IP4_DST "/32 dev veth_src scope global");
+	SYS(fail, "ip -netns " NS_SRC " route add " IP4_DST "/32 via " IP4_TUN_FWD
 	    " dev tun_src scope global");
-	SYS("ip -netns " NS_DST " route add " IP4_TUN_SRC "/32 dev veth_dst scope global");
-	SYS("ip -netns " NS_SRC " route del " IP6_DST "/128 dev veth_src scope global");
-	SYS("ip -netns " NS_SRC " route add " IP6_DST "/128 via " IP6_TUN_FWD
+	SYS(fail, "ip -netns " NS_DST " route add " IP4_TUN_SRC "/32 dev veth_dst scope global");
+	SYS(fail, "ip -netns " NS_SRC " route del " IP6_DST "/128 dev veth_src scope global");
+	SYS(fail, "ip -netns " NS_SRC " route add " IP6_DST "/128 via " IP6_TUN_FWD
 	    " dev tun_src scope global");
-	SYS("ip -netns " NS_DST " route add " IP6_TUN_SRC "/128 dev veth_dst scope global");
+	SYS(fail, "ip -netns " NS_DST " route add " IP6_TUN_SRC "/128 dev veth_dst scope global");
 
-	SYS("ip -netns " NS_DST " neigh add " IP4_TUN_SRC " dev veth_dst lladdr " MAC_DST_FWD);
-	SYS("ip -netns " NS_DST " neigh add " IP6_TUN_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+	SYS(fail, "ip -netns " NS_DST " neigh add " IP4_TUN_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+	SYS(fail, "ip -netns " NS_DST " neigh add " IP6_TUN_SRC " dev veth_dst lladdr " MAC_DST_FWD);
 
 	if (!ASSERT_OK(set_forwarding(false), "disable forwarding"))
 		goto fail;
diff --git a/tools/testing/selftests/bpf/prog_tests/test_tunnel.c b/tools/testing/selftests/bpf/prog_tests/test_tunnel.c
index 07ad457f3370..47f1d482fe39 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_tunnel.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_tunnel.c
@@ -91,30 +91,15 @@
 
 #define PING_ARGS "-i 0.01 -c 3 -w 10 -q"
 
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			goto fail;				\
-	})
-
-#define SYS_NOFAIL(fmt, ...)					\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		system(cmd);					\
-	})
-
 static int config_device(void)
 {
-	SYS("ip netns add at_ns0");
-	SYS("ip link add veth0 address " MAC_VETH1 " type veth peer name veth1");
-	SYS("ip link set veth0 netns at_ns0");
-	SYS("ip addr add " IP4_ADDR1_VETH1 "/24 dev veth1");
-	SYS("ip link set dev veth1 up mtu 1500");
-	SYS("ip netns exec at_ns0 ip addr add " IP4_ADDR_VETH0 "/24 dev veth0");
-	SYS("ip netns exec at_ns0 ip link set dev veth0 up mtu 1500");
+	SYS(fail, "ip netns add at_ns0");
+	SYS(fail, "ip link add veth0 address " MAC_VETH1 " type veth peer name veth1");
+	SYS(fail, "ip link set veth0 netns at_ns0");
+	SYS(fail, "ip addr add " IP4_ADDR1_VETH1 "/24 dev veth1");
+	SYS(fail, "ip link set dev veth1 up mtu 1500");
+	SYS(fail, "ip netns exec at_ns0 ip addr add " IP4_ADDR_VETH0 "/24 dev veth0");
+	SYS(fail, "ip netns exec at_ns0 ip link set dev veth0 up mtu 1500");
 
 	return 0;
 fail:
@@ -132,23 +117,23 @@ static void cleanup(void)
 static int add_vxlan_tunnel(void)
 {
 	/* at_ns0 namespace */
-	SYS("ip netns exec at_ns0 ip link add dev %s type vxlan external gbp dstport 4789",
+	SYS(fail, "ip netns exec at_ns0 ip link add dev %s type vxlan external gbp dstport 4789",
 	    VXLAN_TUNL_DEV0);
-	SYS("ip netns exec at_ns0 ip link set dev %s address %s up",
+	SYS(fail, "ip netns exec at_ns0 ip link set dev %s address %s up",
 	    VXLAN_TUNL_DEV0, MAC_TUNL_DEV0);
-	SYS("ip netns exec at_ns0 ip addr add dev %s %s/24",
+	SYS(fail, "ip netns exec at_ns0 ip addr add dev %s %s/24",
 	    VXLAN_TUNL_DEV0, IP4_ADDR_TUNL_DEV0);
-	SYS("ip netns exec at_ns0 ip neigh add %s lladdr %s dev %s",
+	SYS(fail, "ip netns exec at_ns0 ip neigh add %s lladdr %s dev %s",
 	    IP4_ADDR_TUNL_DEV1, MAC_TUNL_DEV1, VXLAN_TUNL_DEV0);
-	SYS("ip netns exec at_ns0 ip neigh add %s lladdr %s dev veth0",
+	SYS(fail, "ip netns exec at_ns0 ip neigh add %s lladdr %s dev veth0",
 	    IP4_ADDR2_VETH1, MAC_VETH1);
 
 	/* root namespace */
-	SYS("ip link add dev %s type vxlan external gbp dstport 4789",
+	SYS(fail, "ip link add dev %s type vxlan external gbp dstport 4789",
 	    VXLAN_TUNL_DEV1);
-	SYS("ip link set dev %s address %s up", VXLAN_TUNL_DEV1, MAC_TUNL_DEV1);
-	SYS("ip addr add dev %s %s/24", VXLAN_TUNL_DEV1, IP4_ADDR_TUNL_DEV1);
-	SYS("ip neigh add %s lladdr %s dev %s",
+	SYS(fail, "ip link set dev %s address %s up", VXLAN_TUNL_DEV1, MAC_TUNL_DEV1);
+	SYS(fail, "ip addr add dev %s %s/24", VXLAN_TUNL_DEV1, IP4_ADDR_TUNL_DEV1);
+	SYS(fail, "ip neigh add %s lladdr %s dev %s",
 	    IP4_ADDR_TUNL_DEV0, MAC_TUNL_DEV0, VXLAN_TUNL_DEV1);
 
 	return 0;
@@ -165,26 +150,26 @@ static void delete_vxlan_tunnel(void)
 
 static int add_ip6vxlan_tunnel(void)
 {
-	SYS("ip netns exec at_ns0 ip -6 addr add %s/96 dev veth0",
+	SYS(fail, "ip netns exec at_ns0 ip -6 addr add %s/96 dev veth0",
 	    IP6_ADDR_VETH0);
-	SYS("ip netns exec at_ns0 ip link set dev veth0 up");
-	SYS("ip -6 addr add %s/96 dev veth1", IP6_ADDR1_VETH1);
-	SYS("ip -6 addr add %s/96 dev veth1", IP6_ADDR2_VETH1);
-	SYS("ip link set dev veth1 up");
+	SYS(fail, "ip netns exec at_ns0 ip link set dev veth0 up");
+	SYS(fail, "ip -6 addr add %s/96 dev veth1", IP6_ADDR1_VETH1);
+	SYS(fail, "ip -6 addr add %s/96 dev veth1", IP6_ADDR2_VETH1);
+	SYS(fail, "ip link set dev veth1 up");
 
 	/* at_ns0 namespace */
-	SYS("ip netns exec at_ns0 ip link add dev %s type vxlan external dstport 4789",
+	SYS(fail, "ip netns exec at_ns0 ip link add dev %s type vxlan external dstport 4789",
 	    IP6VXLAN_TUNL_DEV0);
-	SYS("ip netns exec at_ns0 ip addr add dev %s %s/24",
+	SYS(fail, "ip netns exec at_ns0 ip addr add dev %s %s/24",
 	    IP6VXLAN_TUNL_DEV0, IP4_ADDR_TUNL_DEV0);
-	SYS("ip netns exec at_ns0 ip link set dev %s address %s up",
+	SYS(fail, "ip netns exec at_ns0 ip link set dev %s address %s up",
 	    IP6VXLAN_TUNL_DEV0, MAC_TUNL_DEV0);
 
 	/* root namespace */
-	SYS("ip link add dev %s type vxlan external dstport 4789",
+	SYS(fail, "ip link add dev %s type vxlan external dstport 4789",
 	    IP6VXLAN_TUNL_DEV1);
-	SYS("ip addr add dev %s %s/24", IP6VXLAN_TUNL_DEV1, IP4_ADDR_TUNL_DEV1);
-	SYS("ip link set dev %s address %s up",
+	SYS(fail, "ip addr add dev %s %s/24", IP6VXLAN_TUNL_DEV1, IP4_ADDR_TUNL_DEV1);
+	SYS(fail, "ip link set dev %s address %s up",
 	    IP6VXLAN_TUNL_DEV1, MAC_TUNL_DEV1);
 
 	return 0;
@@ -205,7 +190,7 @@ static void delete_ip6vxlan_tunnel(void)
 
 static int test_ping(int family, const char *addr)
 {
-	SYS("%s %s %s > /dev/null", ping_command(family), PING_ARGS, addr);
+	SYS(fail, "%s %s %s > /dev/null", ping_command(family), PING_ARGS, addr);
 	return 0;
 fail:
 	return -1;
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_bonding.c b/tools/testing/selftests/bpf/prog_tests/xdp_bonding.c
index 5e3a26b15ec6..d19f79048ff6 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_bonding.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_bonding.c
@@ -141,41 +141,33 @@ static const char * const xmit_policy_names[] = {
 static int bonding_setup(struct skeletons *skeletons, int mode, int xmit_policy,
 			 int bond_both_attach)
 {
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			return -1;				\
-	})
-
-	SYS("ip netns add ns_dst");
-	SYS("ip link add veth1_1 type veth peer name veth2_1 netns ns_dst");
-	SYS("ip link add veth1_2 type veth peer name veth2_2 netns ns_dst");
-
-	SYS("ip link add bond1 type bond mode %s xmit_hash_policy %s",
+	SYS(fail, "ip netns add ns_dst");
+	SYS(fail, "ip link add veth1_1 type veth peer name veth2_1 netns ns_dst");
+	SYS(fail, "ip link add veth1_2 type veth peer name veth2_2 netns ns_dst");
+
+	SYS(fail, "ip link add bond1 type bond mode %s xmit_hash_policy %s",
 	    mode_names[mode], xmit_policy_names[xmit_policy]);
-	SYS("ip link set bond1 up address " BOND1_MAC_STR " addrgenmode none");
-	SYS("ip -netns ns_dst link add bond2 type bond mode %s xmit_hash_policy %s",
+	SYS(fail, "ip link set bond1 up address " BOND1_MAC_STR " addrgenmode none");
+	SYS(fail, "ip -netns ns_dst link add bond2 type bond mode %s xmit_hash_policy %s",
 	    mode_names[mode], xmit_policy_names[xmit_policy]);
-	SYS("ip -netns ns_dst link set bond2 up address " BOND2_MAC_STR " addrgenmode none");
+	SYS(fail, "ip -netns ns_dst link set bond2 up address " BOND2_MAC_STR " addrgenmode none");
 
-	SYS("ip link set veth1_1 master bond1");
+	SYS(fail, "ip link set veth1_1 master bond1");
 	if (bond_both_attach == BOND_BOTH_AND_ATTACH) {
-		SYS("ip link set veth1_2 master bond1");
+		SYS(fail, "ip link set veth1_2 master bond1");
 	} else {
-		SYS("ip link set veth1_2 up addrgenmode none");
+		SYS(fail, "ip link set veth1_2 up addrgenmode none");
 
 		if (xdp_attach(skeletons, skeletons->xdp_dummy->progs.xdp_dummy_prog, "veth1_2"))
 			return -1;
 	}
 
-	SYS("ip -netns ns_dst link set veth2_1 master bond2");
+	SYS(fail, "ip -netns ns_dst link set veth2_1 master bond2");
 
 	if (bond_both_attach == BOND_BOTH_AND_ATTACH)
-		SYS("ip -netns ns_dst link set veth2_2 master bond2");
+		SYS(fail, "ip -netns ns_dst link set veth2_2 master bond2");
 	else
-		SYS("ip -netns ns_dst link set veth2_2 up addrgenmode none");
+		SYS(fail, "ip -netns ns_dst link set veth2_2 up addrgenmode none");
 
 	/* Load a dummy program on sending side as with veth peer needs to have a
 	 * XDP program loaded as well.
@@ -194,8 +186,8 @@ static int bonding_setup(struct skeletons *skeletons, int mode, int xmit_policy,
 	}
 
 	return 0;
-
-#undef SYS
+fail:
+	return -1;
 }
 
 static void bonding_cleanup(struct skeletons *skeletons)
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
index 2666c84dbd01..856cbc29e6a1 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
@@ -12,14 +12,6 @@
 #include <uapi/linux/netdev.h>
 #include "test_xdp_do_redirect.skel.h"
 
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			goto out;				\
-	})
-
 struct udp_packet {
 	struct ethhdr eth;
 	struct ipv6hdr iph;
@@ -126,19 +118,19 @@ void test_xdp_do_redirect(void)
 	 * iface and NUM_PKTS-2 in the TC hook. We match the packets on the UDP
 	 * payload.
 	 */
-	SYS("ip netns add testns");
+	SYS(out, "ip netns add testns");
 	nstoken = open_netns("testns");
 	if (!ASSERT_OK_PTR(nstoken, "setns"))
 		goto out;
 
-	SYS("ip link add veth_src type veth peer name veth_dst");
-	SYS("ip link set dev veth_src address 00:11:22:33:44:55");
-	SYS("ip link set dev veth_dst address 66:77:88:99:aa:bb");
-	SYS("ip link set dev veth_src up");
-	SYS("ip link set dev veth_dst up");
-	SYS("ip addr add dev veth_src fc00::1/64");
-	SYS("ip addr add dev veth_dst fc00::2/64");
-	SYS("ip neigh add fc00::2 dev veth_src lladdr 66:77:88:99:aa:bb");
+	SYS(out, "ip link add veth_src type veth peer name veth_dst");
+	SYS(out, "ip link set dev veth_src address 00:11:22:33:44:55");
+	SYS(out, "ip link set dev veth_dst address 66:77:88:99:aa:bb");
+	SYS(out, "ip link set dev veth_src up");
+	SYS(out, "ip link set dev veth_dst up");
+	SYS(out, "ip addr add dev veth_src fc00::1/64");
+	SYS(out, "ip addr add dev veth_dst fc00::2/64");
+	SYS(out, "ip neigh add fc00::2 dev veth_src lladdr 66:77:88:99:aa:bb");
 
 	/* We enable forwarding in the test namespace because that will cause
 	 * the packets that go through the kernel stack (with XDP_PASS) to be
@@ -151,7 +143,7 @@ void test_xdp_do_redirect(void)
 	 * code didn't have this, so we keep the test behaviour to make sure the
 	 * bug doesn't resurface.
 	 */
-	SYS("sysctl -qw net.ipv6.conf.all.forwarding=1");
+	SYS(out, "sysctl -qw net.ipv6.conf.all.forwarding=1");
 
 	ifindex_src = if_nametoindex("veth_src");
 	ifindex_dst = if_nametoindex("veth_dst");
@@ -225,6 +217,6 @@ out_tc:
 out:
 	if (nstoken)
 		close_netns(nstoken);
-	system("ip netns del testns");
+	SYS_NOFAIL("ip netns del testns");
 	test_xdp_do_redirect__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_metadata.c b/tools/testing/selftests/bpf/prog_tests/xdp_metadata.c
index aa4beae99f4f..490e851dc27d 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_metadata.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_metadata.c
@@ -34,11 +34,6 @@
 #define PREFIX_LEN "8"
 #define FAMILY AF_INET
 
-#define SYS(cmd) ({ \
-	if (!ASSERT_OK(system(cmd), (cmd))) \
-		goto out; \
-})
-
 struct xsk {
 	void *umem_area;
 	struct xsk_umem *umem;
@@ -298,16 +293,16 @@ void test_xdp_metadata(void)
 
 	/* Setup new networking namespace, with a veth pair. */
 
-	SYS("ip netns add xdp_metadata");
+	SYS(out, "ip netns add xdp_metadata");
 	tok = open_netns("xdp_metadata");
-	SYS("ip link add numtxqueues 1 numrxqueues 1 " TX_NAME
+	SYS(out, "ip link add numtxqueues 1 numrxqueues 1 " TX_NAME
 	    " type veth peer " RX_NAME " numtxqueues 1 numrxqueues 1");
-	SYS("ip link set dev " TX_NAME " address 00:00:00:00:00:01");
-	SYS("ip link set dev " RX_NAME " address 00:00:00:00:00:02");
-	SYS("ip link set dev " TX_NAME " up");
-	SYS("ip link set dev " RX_NAME " up");
-	SYS("ip addr add " TX_ADDR "/" PREFIX_LEN " dev " TX_NAME);
-	SYS("ip addr add " RX_ADDR "/" PREFIX_LEN " dev " RX_NAME);
+	SYS(out, "ip link set dev " TX_NAME " address 00:00:00:00:00:01");
+	SYS(out, "ip link set dev " RX_NAME " address 00:00:00:00:00:02");
+	SYS(out, "ip link set dev " TX_NAME " up");
+	SYS(out, "ip link set dev " RX_NAME " up");
+	SYS(out, "ip addr add " TX_ADDR "/" PREFIX_LEN " dev " TX_NAME);
+	SYS(out, "ip addr add " RX_ADDR "/" PREFIX_LEN " dev " RX_NAME);
 
 	rx_ifindex = if_nametoindex(RX_NAME);
 	tx_ifindex = if_nametoindex(TX_NAME);
@@ -405,5 +400,5 @@ out:
 	xdp_metadata__destroy(bpf_obj);
 	if (tok)
 		close_netns(tok);
-	system("ip netns del xdp_metadata");
+	SYS_NOFAIL("ip netns del xdp_metadata");
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_synproxy.c b/tools/testing/selftests/bpf/prog_tests/xdp_synproxy.c
index c72083885b6d..8b50a992d233 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_synproxy.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_synproxy.c
@@ -8,11 +8,6 @@
 
 #define CMD_OUT_BUF_SIZE 1023
 
-#define SYS(cmd) ({ \
-	if (!ASSERT_OK(system(cmd), (cmd))) \
-		goto out; \
-})
-
 #define SYS_OUT(cmd, ...) ({ \
 	char buf[1024]; \
 	snprintf(buf, sizeof(buf), (cmd), ##__VA_ARGS__); \
@@ -69,37 +64,37 @@ static void test_synproxy(bool xdp)
 	char buf[CMD_OUT_BUF_SIZE];
 	size_t size;
 
-	SYS("ip netns add synproxy");
+	SYS(out, "ip netns add synproxy");
 
-	SYS("ip link add tmp0 type veth peer name tmp1");
-	SYS("ip link set tmp1 netns synproxy");
-	SYS("ip link set tmp0 up");
-	SYS("ip addr replace 198.18.0.1/24 dev tmp0");
+	SYS(out, "ip link add tmp0 type veth peer name tmp1");
+	SYS(out, "ip link set tmp1 netns synproxy");
+	SYS(out, "ip link set tmp0 up");
+	SYS(out, "ip addr replace 198.18.0.1/24 dev tmp0");
 
 	/* When checksum offload is enabled, the XDP program sees wrong
 	 * checksums and drops packets.
 	 */
-	SYS("ethtool -K tmp0 tx off");
+	SYS(out, "ethtool -K tmp0 tx off");
 	if (xdp)
 		/* Workaround required for veth. */
-		SYS("ip link set tmp0 xdp object xdp_dummy.bpf.o section xdp 2> /dev/null");
+		SYS(out, "ip link set tmp0 xdp object xdp_dummy.bpf.o section xdp 2> /dev/null");
 
 	ns = open_netns("synproxy");
 	if (!ASSERT_OK_PTR(ns, "setns"))
 		goto out;
 
-	SYS("ip link set lo up");
-	SYS("ip link set tmp1 up");
-	SYS("ip addr replace 198.18.0.2/24 dev tmp1");
-	SYS("sysctl -w net.ipv4.tcp_syncookies=2");
-	SYS("sysctl -w net.ipv4.tcp_timestamps=1");
-	SYS("sysctl -w net.netfilter.nf_conntrack_tcp_loose=0");
-	SYS("iptables-legacy -t raw -I PREROUTING \
+	SYS(out, "ip link set lo up");
+	SYS(out, "ip link set tmp1 up");
+	SYS(out, "ip addr replace 198.18.0.2/24 dev tmp1");
+	SYS(out, "sysctl -w net.ipv4.tcp_syncookies=2");
+	SYS(out, "sysctl -w net.ipv4.tcp_timestamps=1");
+	SYS(out, "sysctl -w net.netfilter.nf_conntrack_tcp_loose=0");
+	SYS(out, "iptables-legacy -t raw -I PREROUTING \
 	    -i tmp1 -p tcp -m tcp --syn --dport 8080 -j CT --notrack");
-	SYS("iptables-legacy -t filter -A INPUT \
+	SYS(out, "iptables-legacy -t filter -A INPUT \
 	    -i tmp1 -p tcp -m tcp --dport 8080 -m state --state INVALID,UNTRACKED \
 	    -j SYNPROXY --sack-perm --timestamp --wscale 7 --mss 1460");
-	SYS("iptables-legacy -t filter -A INPUT \
+	SYS(out, "iptables-legacy -t filter -A INPUT \
 	    -i tmp1 -m state --state INVALID -j DROP");
 
 	ctrl_file = SYS_OUT("./xdp_synproxy --iface tmp1 --ports 8080 \
@@ -170,8 +165,8 @@ out:
 	if (ns)
 		close_netns(ns);
 
-	system("ip link del tmp0");
-	system("ip netns del synproxy");
+	SYS_NOFAIL("ip link del tmp0");
+	SYS_NOFAIL("ip netns del synproxy");
 }
 
 void test_xdp_synproxy(void)
diff --git a/tools/testing/selftests/bpf/prog_tests/xfrm_info.c b/tools/testing/selftests/bpf/prog_tests/xfrm_info.c
index 8b03c9bb4862..d37f5394e199 100644
--- a/tools/testing/selftests/bpf/prog_tests/xfrm_info.c
+++ b/tools/testing/selftests/bpf/prog_tests/xfrm_info.c
@@ -69,21 +69,6 @@
     "proto esp aead 'rfc4106(gcm(aes))' " \
     "0xe4d8f4b4da1df18a3510b3781496daa82488b713 128 mode tunnel "
 
-#define SYS(fmt, ...)						\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		if (!ASSERT_OK(system(cmd), cmd))		\
-			goto fail;				\
-	})
-
-#define SYS_NOFAIL(fmt, ...)					\
-	({							\
-		char cmd[1024];					\
-		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
-		system(cmd);					\
-	})
-
 static int attach_tc_prog(struct bpf_tc_hook *hook, int igr_fd, int egr_fd)
 {
 	LIBBPF_OPTS(bpf_tc_opts, opts1, .handle = 1, .priority = 1,
@@ -126,23 +111,23 @@ static void cleanup(void)
 
 static int config_underlay(void)
 {
-	SYS("ip netns add " NS0);
-	SYS("ip netns add " NS1);
-	SYS("ip netns add " NS2);
+	SYS(fail, "ip netns add " NS0);
+	SYS(fail, "ip netns add " NS1);
+	SYS(fail, "ip netns add " NS2);
 
 	/* NS0 <-> NS1 [veth01 <-> veth10] */
-	SYS("ip link add veth01 netns " NS0 " type veth peer name veth10 netns " NS1);
-	SYS("ip -net " NS0 " addr add " IP4_ADDR_VETH01 "/24 dev veth01");
-	SYS("ip -net " NS0 " link set dev veth01 up");
-	SYS("ip -net " NS1 " addr add " IP4_ADDR_VETH10 "/24 dev veth10");
-	SYS("ip -net " NS1 " link set dev veth10 up");
+	SYS(fail, "ip link add veth01 netns " NS0 " type veth peer name veth10 netns " NS1);
+	SYS(fail, "ip -net " NS0 " addr add " IP4_ADDR_VETH01 "/24 dev veth01");
+	SYS(fail, "ip -net " NS0 " link set dev veth01 up");
+	SYS(fail, "ip -net " NS1 " addr add " IP4_ADDR_VETH10 "/24 dev veth10");
+	SYS(fail, "ip -net " NS1 " link set dev veth10 up");
 
 	/* NS0 <-> NS2 [veth02 <-> veth20] */
-	SYS("ip link add veth02 netns " NS0 " type veth peer name veth20 netns " NS2);
-	SYS("ip -net " NS0 " addr add " IP4_ADDR_VETH02 "/24 dev veth02");
-	SYS("ip -net " NS0 " link set dev veth02 up");
-	SYS("ip -net " NS2 " addr add " IP4_ADDR_VETH20 "/24 dev veth20");
-	SYS("ip -net " NS2 " link set dev veth20 up");
+	SYS(fail, "ip link add veth02 netns " NS0 " type veth peer name veth20 netns " NS2);
+	SYS(fail, "ip -net " NS0 " addr add " IP4_ADDR_VETH02 "/24 dev veth02");
+	SYS(fail, "ip -net " NS0 " link set dev veth02 up");
+	SYS(fail, "ip -net " NS2 " addr add " IP4_ADDR_VETH20 "/24 dev veth20");
+	SYS(fail, "ip -net " NS2 " link set dev veth20 up");
 
 	return 0;
 fail:
@@ -153,20 +138,20 @@ static int setup_xfrm_tunnel_ns(const char *ns, const char *ipv4_local,
 				const char *ipv4_remote, int if_id)
 {
 	/* State: local -> remote */
-	SYS("ip -net %s xfrm state add src %s dst %s spi 1 "
+	SYS(fail, "ip -net %s xfrm state add src %s dst %s spi 1 "
 	    ESP_DUMMY_PARAMS "if_id %d", ns, ipv4_local, ipv4_remote, if_id);
 
 	/* State: local <- remote */
-	SYS("ip -net %s xfrm state add src %s dst %s spi 1 "
+	SYS(fail, "ip -net %s xfrm state add src %s dst %s spi 1 "
 	    ESP_DUMMY_PARAMS "if_id %d", ns, ipv4_remote, ipv4_local, if_id);
 
 	/* Policy: local -> remote */
-	SYS("ip -net %s xfrm policy add dir out src 0.0.0.0/0 dst 0.0.0.0/0 "
+	SYS(fail, "ip -net %s xfrm policy add dir out src 0.0.0.0/0 dst 0.0.0.0/0 "
 	    "if_id %d tmpl src %s dst %s proto esp mode tunnel if_id %d", ns,
 	    if_id, ipv4_local, ipv4_remote, if_id);
 
 	/* Policy: local <- remote */
-	SYS("ip -net %s xfrm policy add dir in src 0.0.0.0/0 dst 0.0.0.0/0 "
+	SYS(fail, "ip -net %s xfrm policy add dir in src 0.0.0.0/0 dst 0.0.0.0/0 "
 	    "if_id %d tmpl src %s dst %s proto esp mode tunnel if_id %d", ns,
 	    if_id, ipv4_remote, ipv4_local, if_id);
 
@@ -274,16 +259,16 @@ static int config_overlay(void)
 	if (!ASSERT_OK(setup_xfrmi_external_dev(NS0), "xfrmi"))
 		goto fail;
 
-	SYS("ip -net " NS0 " addr add 192.168.1.100/24 dev ipsec0");
-	SYS("ip -net " NS0 " link set dev ipsec0 up");
+	SYS(fail, "ip -net " NS0 " addr add 192.168.1.100/24 dev ipsec0");
+	SYS(fail, "ip -net " NS0 " link set dev ipsec0 up");
 
-	SYS("ip -net " NS1 " link add ipsec0 type xfrm if_id %d", IF_ID_1);
-	SYS("ip -net " NS1 " addr add 192.168.1.200/24 dev ipsec0");
-	SYS("ip -net " NS1 " link set dev ipsec0 up");
+	SYS(fail, "ip -net " NS1 " link add ipsec0 type xfrm if_id %d", IF_ID_1);
+	SYS(fail, "ip -net " NS1 " addr add 192.168.1.200/24 dev ipsec0");
+	SYS(fail, "ip -net " NS1 " link set dev ipsec0 up");
 
-	SYS("ip -net " NS2 " link add ipsec0 type xfrm if_id %d", IF_ID_2);
-	SYS("ip -net " NS2 " addr add 192.168.1.200/24 dev ipsec0");
-	SYS("ip -net " NS2 " link set dev ipsec0 up");
+	SYS(fail, "ip -net " NS2 " link add ipsec0 type xfrm if_id %d", IF_ID_2);
+	SYS(fail, "ip -net " NS2 " addr add 192.168.1.200/24 dev ipsec0");
+	SYS(fail, "ip -net " NS2 " link set dev ipsec0 up");
 
 	return 0;
 fail:
@@ -294,7 +279,7 @@ static int test_xfrm_ping(struct xfrm_info *skel, u32 if_id)
 {
 	skel->bss->req_if_id = if_id;
 
-	SYS("ping -i 0.01 -c 3 -w 10 -q 192.168.1.200 > /dev/null");
+	SYS(fail, "ping -i 0.01 -c 3 -w 10 -q 192.168.1.200 > /dev/null");
 
 	if (!ASSERT_EQ(skel->bss->resp_if_id, if_id, "if_id"))
 		goto fail;
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index d5d51ec97ec8..9fbdc57c5b57 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -376,6 +376,21 @@ int test__join_cgroup(const char *path);
 	___ok;								\
 })
 
+#define SYS(goto_label, fmt, ...)					\
+	({								\
+		char cmd[1024];						\
+		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);		\
+		if (!ASSERT_OK(system(cmd), cmd))			\
+			goto goto_label;				\
+	})
+
+#define SYS_NOFAIL(fmt, ...)						\
+	({								\
+		char cmd[1024];						\
+		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);		\
+		system(cmd);						\
+	})
+
 static inline __u64 ptr_to_u64(const void *ptr)
 {
 	return (__u64) (unsigned long) ptr;
-- 
cgit v1.2.3-70-g09d2


From 02d6a057c7bee44902c843949de6bbd439e33092 Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Fri, 24 Feb 2023 14:13:43 +0800
Subject: selftests/bpf: run mptcp in a dedicated netns

The current mptcp test is run in init netns. If the user or default
system config disabled mptcp, the test will fail. Let's run the mptcp
test in a dedicated netns to avoid none kernel default mptcp setting.

Suggested-by: Martin KaFai Lau <martin.lau@linux.dev>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Link: https://lore.kernel.org/r/20230224061343.506571-3-liuhangbin@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/mptcp.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c
index 59f08d6d1d53..cd0c42fff7c0 100644
--- a/tools/testing/selftests/bpf/prog_tests/mptcp.c
+++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c
@@ -7,6 +7,8 @@
 #include "network_helpers.h"
 #include "mptcp_sock.skel.h"
 
+#define NS_TEST "mptcp_ns"
+
 #ifndef TCP_CA_NAME_MAX
 #define TCP_CA_NAME_MAX	16
 #endif
@@ -138,12 +140,20 @@ out:
 
 static void test_base(void)
 {
+	struct nstoken *nstoken = NULL;
 	int server_fd, cgroup_fd;
 
 	cgroup_fd = test__join_cgroup("/mptcp");
 	if (!ASSERT_GE(cgroup_fd, 0, "test__join_cgroup"))
 		return;
 
+	SYS(fail, "ip netns add %s", NS_TEST);
+	SYS(fail, "ip -net %s link set dev lo up", NS_TEST);
+
+	nstoken = open_netns(NS_TEST);
+	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
+		goto fail;
+
 	/* without MPTCP */
 	server_fd = start_server(AF_INET, SOCK_STREAM, NULL, 0, 0);
 	if (!ASSERT_GE(server_fd, 0, "start_server"))
@@ -157,13 +167,18 @@ with_mptcp:
 	/* with MPTCP */
 	server_fd = start_mptcp_server(AF_INET, NULL, 0, 0);
 	if (!ASSERT_GE(server_fd, 0, "start_mptcp_server"))
-		goto close_cgroup_fd;
+		goto fail;
 
 	ASSERT_OK(run_test(cgroup_fd, server_fd, true), "run_test mptcp");
 
 	close(server_fd);
 
-close_cgroup_fd:
+fail:
+	if (nstoken)
+		close_netns(nstoken);
+
+	SYS_NOFAIL("ip netns del " NS_TEST " &> /dev/null");
+
 	close(cgroup_fd);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 0f485805d008aa56644f68179a7e6579fc1515e7 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 17:00:33 +0900
Subject: wifi: brcmfmac: acpi: Add support for fetching Apple ACPI properties

On DT platforms, the module-instance and antenna-sku-info properties
are passed in the DT. On ACPI platforms, module-instance is passed via
the analogous Apple device property mechanism, while the antenna SKU
info is instead obtained via an ACPI method that grabs it from
non-volatile storage.

Add support for this, to allow proper firmware selection on Apple
platforms.

Signed-off-by: Hector Martin <marcan@marcan.st>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214080034.3828-2-marcan@marcan.st
---
 .../wireless/broadcom/brcm80211/brcmfmac/Makefile  |  2 +
 .../wireless/broadcom/brcm80211/brcmfmac/acpi.c    | 51 ++++++++++++++++++++++
 .../wireless/broadcom/brcm80211/brcmfmac/common.c  |  1 +
 .../wireless/broadcom/brcm80211/brcmfmac/common.h  |  9 ++++
 4 files changed, 63 insertions(+)
 create mode 100644 drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile
index 0e996cf24f88..dc6d27a36faa 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile
@@ -48,6 +48,8 @@ brcmfmac-$(CONFIG_OF) += \
 		of.o
 brcmfmac-$(CONFIG_DMI) += \
 		dmi.o
+brcmfmac-$(CONFIG_ACPI) += \
+		acpi.o
 
 ifeq ($(CONFIG_BRCMFMAC),m)
 obj-m += wcc/
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
new file mode 100644
index 000000000000..c4a54861bfb4
--- /dev/null
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
@@ -0,0 +1,51 @@
+// SPDX-License-Identifier: ISC
+/*
+ * Copyright The Asahi Linux Contributors
+ */
+
+#include <linux/acpi.h>
+#include "debug.h"
+#include "core.h"
+#include "common.h"
+
+void brcmf_acpi_probe(struct device *dev, enum brcmf_bus_type bus_type,
+		      struct brcmf_mp_device *settings)
+{
+	acpi_status status;
+	const union acpi_object *o;
+	struct acpi_buffer buf = {ACPI_ALLOCATE_BUFFER, NULL};
+	struct acpi_device *adev = ACPI_COMPANION(dev);
+
+	if (!adev)
+		return;
+
+	if (!ACPI_FAILURE(acpi_dev_get_property(adev, "module-instance",
+						ACPI_TYPE_STRING, &o))) {
+		brcmf_dbg(INFO, "ACPI module-instance=%s\n", o->string.pointer);
+		settings->board_type = devm_kasprintf(dev, GFP_KERNEL,
+						      "apple,%s",
+						      o->string.pointer);
+	} else {
+		brcmf_dbg(INFO, "No ACPI module-instance\n");
+		return;
+	}
+
+	status = acpi_evaluate_object(adev->handle, "RWCV", NULL, &buf);
+	o = buf.pointer;
+	if (!ACPI_FAILURE(status) && o && o->type == ACPI_TYPE_BUFFER &&
+	    o->buffer.length >= 2) {
+		char *antenna_sku = devm_kzalloc(dev, 3, GFP_KERNEL);
+
+		if (antenna_sku) {
+			memcpy(antenna_sku, o->buffer.pointer, 2);
+			brcmf_dbg(INFO, "ACPI RWCV data=%*phN antenna-sku=%s\n",
+				  (int)o->buffer.length, o->buffer.pointer,
+				  antenna_sku);
+			settings->antenna_sku = antenna_sku;
+		}
+
+		kfree(buf.pointer);
+	} else {
+		brcmf_dbg(INFO, "No ACPI antenna-sku\n");
+	}
+}
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index f235beaddddb..4fea6f0ea8c8 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -487,6 +487,7 @@ struct brcmf_mp_device *brcmf_get_module_param(struct device *dev,
 		/* No platform data for this device, try OF and DMI data */
 		brcmf_dmi_probe(settings, chip, chiprev);
 		brcmf_of_probe(dev, bus_type, settings);
+		brcmf_acpi_probe(dev, bus_type, settings);
 	}
 	return settings;
 }
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index aa25abffcc7d..7167fd4f8c63 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -77,6 +77,15 @@ static inline void
 brcmf_dmi_probe(struct brcmf_mp_device *settings, u32 chip, u32 chiprev) {}
 #endif
 
+#ifdef CONFIG_ACPI
+void brcmf_acpi_probe(struct device *dev, enum brcmf_bus_type bus_type,
+		      struct brcmf_mp_device *settings);
+#else
+static inline void brcmf_acpi_probe(struct device *dev,
+				    enum brcmf_bus_type bus_type,
+				    struct brcmf_mp_device *settings) {}
+#endif
+
 u8 brcmf_map_prio_to_prec(void *cfg, u8 prio);
 
 u8 brcmf_map_prio_to_aci(void *cfg, u8 prio);
-- 
cgit v1.2.3-70-g09d2


From 91918ce88d9fef408bb12c46a27c73d79b604c20 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 17:00:34 +0900
Subject: wifi: brcmfmac: pcie: Provide a buffer of random bytes to the device

Newer Apple firmwares on chipsets without a hardware RNG require the
host to provide a buffer of 256 random bytes to the device on
initialization. This buffer is present immediately before NVRAM,
suffixed by a footer containing a magic number and the buffer length.

This won't affect chips/firmwares that do not use this feature, so do it
unconditionally for all Apple platforms (those with an Apple OTP).

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Hector Martin <marcan@marcan.st>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214080034.3828-3-marcan@marcan.st
---
 .../wireless/broadcom/brcm80211/brcmfmac/pcie.c    | 32 ++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index a9b9b2dc62d4..2835ef4edb18 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -15,6 +15,7 @@
 #include <linux/sched/signal.h>
 #include <linux/kthread.h>
 #include <linux/io.h>
+#include <linux/random.h>
 #include <asm/unaligned.h>
 
 #include <soc.h>
@@ -1653,6 +1654,13 @@ brcmf_pcie_init_share_ram_info(struct brcmf_pciedev_info *devinfo,
 	return 0;
 }
 
+struct brcmf_random_seed_footer {
+	__le32 length;
+	__le32 magic;
+};
+
+#define BRCMF_RANDOM_SEED_MAGIC		0xfeedc0de
+#define BRCMF_RANDOM_SEED_LENGTH	0x100
 
 static int brcmf_pcie_download_fw_nvram(struct brcmf_pciedev_info *devinfo,
 					const struct firmware *fw, void *nvram,
@@ -1689,6 +1697,30 @@ static int brcmf_pcie_download_fw_nvram(struct brcmf_pciedev_info *devinfo,
 			  nvram_len;
 		memcpy_toio(devinfo->tcm + address, nvram, nvram_len);
 		brcmf_fw_nvram_free(nvram);
+
+		if (devinfo->otp.valid) {
+			size_t rand_len = BRCMF_RANDOM_SEED_LENGTH;
+			struct brcmf_random_seed_footer footer = {
+				.length = cpu_to_le32(rand_len),
+				.magic = cpu_to_le32(BRCMF_RANDOM_SEED_MAGIC),
+			};
+			void *randbuf;
+
+			/* Some Apple chips/firmwares expect a buffer of random
+			 * data to be present before NVRAM
+			 */
+			brcmf_dbg(PCIE, "Download random seed\n");
+
+			address -= sizeof(footer);
+			memcpy_toio(devinfo->tcm + address, &footer,
+				    sizeof(footer));
+
+			address -= rand_len;
+			randbuf = kzalloc(rand_len, GFP_KERNEL);
+			get_random_bytes(randbuf, rand_len);
+			memcpy_toio(devinfo->tcm + address, randbuf, rand_len);
+			kfree(randbuf);
+		}
 	} else {
 		brcmf_dbg(PCIE, "No matching NVRAM file found %s\n",
 			  devinfo->nvram_name);
-- 
cgit v1.2.3-70-g09d2


From ec52d77d077529f198fd874c550a26b9cc86a331 Mon Sep 17 00:00:00 2001
From: John Keeping <john@metanate.com>
Date: Tue, 24 Jan 2023 10:42:48 +0000
Subject: wifi: brcmfmac: support CQM RSSI notification with older firmware

Using the BCM4339 firmware from linux-firmware (version "BCM4339/2 wl0:
Sep  5 2019 11:05:52 version 6.37.39.113 (r722271 CY)" from
cypress/cyfmac4339-sdio.bin) the RSSI respose is only 4 bytes, which
results in an error being logged.

It seems that older devices send only the RSSI field and neither SNR nor
noise is included.  Handle this by accepting a 4 byte message and
reading only the RSSI from it.

Fixes: 7dd56ea45a66 ("brcmfmac: add support for CQM RSSI notifications")
Signed-off-by: John Keeping <john@metanate.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230124104248.2917465-1-john@metanate.com
---
 .../net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c    | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index a9690ec4c850..d80d2b9b15c7 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6489,18 +6489,20 @@ static s32 brcmf_notify_rssi(struct brcmf_if *ifp,
 {
 	struct brcmf_cfg80211_vif *vif = ifp->vif;
 	struct brcmf_rssi_be *info = data;
-	s32 rssi, snr, noise;
+	s32 rssi, snr = 0, noise = 0;
 	s32 low, high, last;
 
-	if (e->datalen < sizeof(*info)) {
+	if (e->datalen >= sizeof(*info)) {
+		rssi = be32_to_cpu(info->rssi);
+		snr = be32_to_cpu(info->snr);
+		noise = be32_to_cpu(info->noise);
+	} else if (e->datalen >= sizeof(rssi)) {
+		rssi = be32_to_cpu(*(__be32 *)data);
+	} else {
 		brcmf_err("insufficient RSSI event data\n");
 		return 0;
 	}
 
-	rssi = be32_to_cpu(info->rssi);
-	snr = be32_to_cpu(info->snr);
-	noise = be32_to_cpu(info->noise);
-
 	low = vif->cqm_rssi_low;
 	high = vif->cqm_rssi_high;
 	last = vif->cqm_rssi_last;
-- 
cgit v1.2.3-70-g09d2


From 75c4a8154cb6c7239fb55d5550f481f6765fb83c Mon Sep 17 00:00:00 2001
From: Fedor Pchelkin <pchelkin@ispras.ru>
Date: Fri, 24 Feb 2023 12:28:05 +0200
Subject: wifi: ath6kl: reduce WARN to dev_dbg() in callback

The warn is triggered on a known race condition, documented in the code above
the test, that is correctly handled.  Using WARN() hinders automated testing.
Reducing severity.

Fixes: de2070fc4aa7 ("ath6kl: Fix kernel panic on continuous driver load/unload")
Reported-and-tested-by: syzbot+555908813b2ea35dae9a@syzkaller.appspotmail.com
Signed-off-by: Oliver Neukum <oneukum@suse.com>
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230126182431.867984-1-pchelkin@ispras.ru
---
 drivers/net/wireless/ath/ath6kl/htc_pipe.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath6kl/htc_pipe.c b/drivers/net/wireless/ath/ath6kl/htc_pipe.c
index c68848819a52..9b88d96bfe96 100644
--- a/drivers/net/wireless/ath/ath6kl/htc_pipe.c
+++ b/drivers/net/wireless/ath/ath6kl/htc_pipe.c
@@ -960,8 +960,8 @@ static int ath6kl_htc_pipe_rx_complete(struct ath6kl *ar, struct sk_buff *skb,
 	 * Thus the possibility of ar->htc_target being NULL
 	 * via ath6kl_recv_complete -> ath6kl_usb_io_comp_work.
 	 */
-	if (WARN_ON_ONCE(!target)) {
-		ath6kl_err("Target not yet initialized\n");
+	if (!target) {
+		ath6kl_dbg(ATH6KL_DBG_HTC, "Target not yet initialized\n");
 		status = -EINVAL;
 		goto free_skb;
 	}
-- 
cgit v1.2.3-70-g09d2


From a96f10422e74cde27c100b321b127ec32ae75747 Mon Sep 17 00:00:00 2001
From: Muna Sinada <quic_msinada@quicinc.com>
Date: Fri, 24 Feb 2023 12:28:03 +0200
Subject: wifi: ath11k: modify accessor macros to match index size

HE PHY is only 11 bytes, therefore it should be using byte indexes
instead of dword. Change corresponding macros to reflect this.

Signed-off-by: Muna Sinada <quic_msinada@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/1666128501-12364-2-git-send-email-quic_msinada@quicinc.com
---
 drivers/net/wireless/ath/ath11k/wmi.h | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 2273d885f7fa..9f50be411a8c 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2859,30 +2859,32 @@ struct rx_reorder_queue_remove_params {
 #define WMI_VDEV_PARAM_TXBF_SU_TX_BFER BIT(2)
 #define WMI_VDEV_PARAM_TXBF_MU_TX_BFER BIT(3)
 
-#define HECAP_PHYDWORD_0	0
-#define HECAP_PHYDWORD_1	1
-#define HECAP_PHYDWORD_2	2
+#define HE_PHYCAP_BYTE_0	0
+#define HE_PHYCAP_BYTE_1	1
+#define HE_PHYCAP_BYTE_2	2
+#define HE_PHYCAP_BYTE_3	3
+#define HE_PHYCAP_BYTE_4	4
 
-#define HECAP_PHY_SU_BFER		BIT(31)
+#define HECAP_PHY_SU_BFER		BIT(7)
 #define HECAP_PHY_SU_BFEE		BIT(0)
 #define HECAP_PHY_MU_BFER		BIT(1)
-#define HECAP_PHY_UL_MUMIMO		BIT(22)
-#define HECAP_PHY_UL_MUOFDMA		BIT(23)
+#define HECAP_PHY_UL_MUMIMO		BIT(6)
+#define HECAP_PHY_UL_MUOFDMA		BIT(7)
 
 #define HECAP_PHY_SUBFMR_GET(hecap_phy) \
-	FIELD_GET(HECAP_PHY_SU_BFER, hecap_phy[HECAP_PHYDWORD_0])
+	FIELD_GET(HECAP_PHY_SU_BFER, hecap_phy[HE_PHYCAP_BYTE_3])
 
 #define HECAP_PHY_SUBFME_GET(hecap_phy) \
-	FIELD_GET(HECAP_PHY_SU_BFEE, hecap_phy[HECAP_PHYDWORD_1])
+	FIELD_GET(HECAP_PHY_SU_BFEE, hecap_phy[HE_PHYCAP_BYTE_4])
 
 #define HECAP_PHY_MUBFMR_GET(hecap_phy) \
-	FIELD_GET(HECAP_PHY_MU_BFER, hecap_phy[HECAP_PHYDWORD_1])
+	FIELD_GET(HECAP_PHY_MU_BFER, hecap_phy[HE_PHYCAP_BYTE_4])
 
 #define HECAP_PHY_ULMUMIMO_GET(hecap_phy) \
-	FIELD_GET(HECAP_PHY_UL_MUMIMO, hecap_phy[HECAP_PHYDWORD_0])
+	FIELD_GET(HECAP_PHY_UL_MUMIMO, hecap_phy[HE_PHYCAP_BYTE_2])
 
 #define HECAP_PHY_ULOFDMA_GET(hecap_phy) \
-	FIELD_GET(HECAP_PHY_UL_MUOFDMA, hecap_phy[HECAP_PHYDWORD_0])
+	FIELD_GET(HECAP_PHY_UL_MUOFDMA, hecap_phy[HE_PHYCAP_BYTE_2])
 
 #define HE_MODE_SU_TX_BFEE	BIT(0)
 #define HE_MODE_SU_TX_BFER	BIT(1)
-- 
cgit v1.2.3-70-g09d2


From 38dfe775d0abf511341f37c1cb77b919a3ad410b Mon Sep 17 00:00:00 2001
From: Muna Sinada <quic_msinada@quicinc.com>
Date: Fri, 24 Feb 2023 12:28:04 +0200
Subject: wifi: ath11k: push MU-MIMO params from hostapd to hardware

In the previous behaviour only HE IE in management frames are changed
regarding MU-MIMO configurations and not in hardware. Adding push of
MU-MIMO configurations to the hardware as well.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.4.0.1-00356-QCAHKSWPL_SILICONZ-1

Co-developed-by: Anilkumar Kolli <quic_akolli@quicinc.com>
Signed-off-by: Anilkumar Kolli <quic_akolli@quicinc.com>
Signed-off-by: Muna Sinada <quic_msinada@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/1666128501-12364-3-git-send-email-quic_msinada@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 200 +++++++++++++++++++++-------------
 drivers/net/wireless/ath/ath11k/wmi.h |   3 +
 2 files changed, 130 insertions(+), 73 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 110a38cce0a7..7c4f424180dc 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -2699,6 +2699,117 @@ static int ath11k_setup_peer_smps(struct ath11k *ar, struct ath11k_vif *arvif,
 					 ath11k_smps_map[smps]);
 }
 
+static bool ath11k_mac_set_he_txbf_conf(struct ath11k_vif *arvif)
+{
+	struct ath11k *ar = arvif->ar;
+	u32 param, value;
+	int ret;
+
+	if (!arvif->vif->bss_conf.he_support)
+		return true;
+
+	param = WMI_VDEV_PARAM_SET_HEMU_MODE;
+	value = 0;
+	if (arvif->vif->bss_conf.he_su_beamformer) {
+		value |= FIELD_PREP(HE_MODE_SU_TX_BFER, HE_SU_BFER_ENABLE);
+		if (arvif->vif->bss_conf.he_mu_beamformer &&
+		    arvif->vdev_type == WMI_VDEV_TYPE_AP)
+			value |= FIELD_PREP(HE_MODE_MU_TX_BFER, HE_MU_BFER_ENABLE);
+	}
+
+	if (arvif->vif->type != NL80211_IFTYPE_MESH_POINT) {
+		value |= FIELD_PREP(HE_MODE_DL_OFDMA, HE_DL_MUOFDMA_ENABLE) |
+			 FIELD_PREP(HE_MODE_UL_OFDMA, HE_UL_MUOFDMA_ENABLE);
+
+		if (arvif->vif->bss_conf.he_full_ul_mumimo)
+			value |= FIELD_PREP(HE_MODE_UL_MUMIMO, HE_UL_MUMIMO_ENABLE);
+
+		if (arvif->vif->bss_conf.he_su_beamformee)
+			value |= FIELD_PREP(HE_MODE_SU_TX_BFEE, HE_SU_BFEE_ENABLE);
+	}
+
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, param, value);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set vdev %d HE MU mode: %d\n",
+			    arvif->vdev_id, ret);
+		return false;
+	}
+
+	param = WMI_VDEV_PARAM_SET_HE_SOUNDING_MODE;
+	value =	FIELD_PREP(HE_VHT_SOUNDING_MODE, HE_VHT_SOUNDING_MODE_ENABLE) |
+		FIELD_PREP(HE_TRIG_NONTRIG_SOUNDING_MODE,
+			   HE_TRIG_NONTRIG_SOUNDING_MODE_ENABLE);
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
+					    param, value);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to set vdev %d sounding mode: %d\n",
+			    arvif->vdev_id, ret);
+		return false;
+	}
+	return true;
+}
+
+static bool ath11k_mac_vif_recalc_sta_he_txbf(struct ath11k *ar,
+					      struct ieee80211_vif *vif,
+					      struct ieee80211_sta_he_cap *he_cap)
+{
+	struct ath11k_vif *arvif = (void *)vif->drv_priv;
+	struct ieee80211_he_cap_elem he_cap_elem = {0};
+	struct ieee80211_sta_he_cap *cap_band = NULL;
+	struct cfg80211_chan_def def;
+	u32 param = WMI_VDEV_PARAM_SET_HEMU_MODE;
+	u32 hemode = 0;
+	int ret;
+
+	if (!vif->bss_conf.he_support)
+		return true;
+
+	if (vif->type != NL80211_IFTYPE_STATION)
+		return false;
+
+	if (WARN_ON(ath11k_mac_vif_chan(vif, &def)))
+		return false;
+
+	if (def.chan->band == NL80211_BAND_2GHZ)
+		cap_band = &ar->mac.iftype[NL80211_BAND_2GHZ][vif->type].he_cap;
+	else
+		cap_band = &ar->mac.iftype[NL80211_BAND_5GHZ][vif->type].he_cap;
+
+	memcpy(&he_cap_elem, &cap_band->he_cap_elem, sizeof(he_cap_elem));
+
+	if (HECAP_PHY_SUBFME_GET(he_cap_elem.phy_cap_info)) {
+		if (HECAP_PHY_SUBFMR_GET(he_cap->he_cap_elem.phy_cap_info))
+			hemode |= FIELD_PREP(HE_MODE_SU_TX_BFEE, HE_SU_BFEE_ENABLE);
+		if (HECAP_PHY_MUBFMR_GET(he_cap->he_cap_elem.phy_cap_info))
+			hemode |= FIELD_PREP(HE_MODE_MU_TX_BFEE, HE_MU_BFEE_ENABLE);
+	}
+
+	if (vif->type != NL80211_IFTYPE_MESH_POINT) {
+		hemode |= FIELD_PREP(HE_MODE_DL_OFDMA, HE_DL_MUOFDMA_ENABLE) |
+			  FIELD_PREP(HE_MODE_UL_OFDMA, HE_UL_MUOFDMA_ENABLE);
+
+		if (HECAP_PHY_ULMUMIMO_GET(he_cap_elem.phy_cap_info))
+			if (HECAP_PHY_ULMUMIMO_GET(he_cap->he_cap_elem.phy_cap_info))
+				hemode |= FIELD_PREP(HE_MODE_UL_MUMIMO,
+						     HE_UL_MUMIMO_ENABLE);
+
+		if (FIELD_GET(HE_MODE_MU_TX_BFEE, hemode))
+			hemode |= FIELD_PREP(HE_MODE_SU_TX_BFEE, HE_SU_BFEE_ENABLE);
+
+		if (FIELD_GET(HE_MODE_MU_TX_BFER, hemode))
+			hemode |= FIELD_PREP(HE_MODE_SU_TX_BFER, HE_SU_BFER_ENABLE);
+	}
+
+	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, param, hemode);
+	if (ret) {
+		ath11k_warn(ar->ab, "failed to submit vdev param txbf 0x%x: %d\n",
+			    hemode, ret);
+		return false;
+	}
+
+	return true;
+}
+
 static void ath11k_bss_assoc(struct ieee80211_hw *hw,
 			     struct ieee80211_vif *vif,
 			     struct ieee80211_bss_conf *bss_conf)
@@ -2709,6 +2820,7 @@ static void ath11k_bss_assoc(struct ieee80211_hw *hw,
 	struct ieee80211_sta *ap_sta;
 	struct ath11k_peer *peer;
 	bool is_auth = false;
+	struct ieee80211_sta_he_cap  he_cap;
 	int ret;
 
 	lockdep_assert_held(&ar->conf_mutex);
@@ -2726,6 +2838,9 @@ static void ath11k_bss_assoc(struct ieee80211_hw *hw,
 		return;
 	}
 
+	/* he_cap here is updated at assoc success for sta mode only */
+	he_cap  = ap_sta->deflink.he_cap;
+
 	ath11k_peer_assoc_prepare(ar, vif, ap_sta, &peer_arg, false);
 
 	rcu_read_unlock();
@@ -2753,6 +2868,12 @@ static void ath11k_bss_assoc(struct ieee80211_hw *hw,
 		return;
 	}
 
+	if (!ath11k_mac_vif_recalc_sta_he_txbf(ar, vif, &he_cap)) {
+		ath11k_warn(ar->ab, "failed to recalc he txbf for vdev %i on bss %pM\n",
+			    arvif->vdev_id, bss_conf->bssid);
+		return;
+	}
+
 	WARN_ON(arvif->is_up);
 
 	arvif->aid = vif->cfg.aid;
@@ -3202,6 +3323,8 @@ static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 		ether_addr_copy(arvif->bssid, info->bssid);
 
 	if (changed & BSS_CHANGED_BEACON_ENABLED) {
+		if (info->enable_beacon)
+			ath11k_mac_set_he_txbf_conf(arvif);
 		ath11k_control_beaconing(arvif, info);
 
 		if (arvif->is_up && vif->bss_conf.he_support &&
@@ -5392,6 +5515,10 @@ static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 
 		he_cap_elem->mac_cap_info[1] &=
 			IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_MASK;
+		he_cap_elem->phy_cap_info[0] &=
+			~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G;
+		he_cap_elem->phy_cap_info[0] &=
+			~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G;
 
 		he_cap_elem->phy_cap_info[5] &=
 			~IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_MASK;
@@ -6026,69 +6153,6 @@ ath11k_mac_setup_vdev_create_params(struct ath11k_vif *arvif,
 	}
 }
 
-static u32
-ath11k_mac_prepare_he_mode(struct ath11k_pdev *pdev, u32 viftype)
-{
-	struct ath11k_pdev_cap *pdev_cap = &pdev->cap;
-	struct ath11k_band_cap *cap_band = NULL;
-	u32 *hecap_phy_ptr = NULL;
-	u32 hemode = 0;
-
-	if (pdev->cap.supported_bands & WMI_HOST_WLAN_2G_CAP)
-		cap_band = &pdev_cap->band[NL80211_BAND_2GHZ];
-	else
-		cap_band = &pdev_cap->band[NL80211_BAND_5GHZ];
-
-	hecap_phy_ptr = &cap_band->he_cap_phy_info[0];
-
-	hemode = FIELD_PREP(HE_MODE_SU_TX_BFEE, HE_SU_BFEE_ENABLE) |
-		 FIELD_PREP(HE_MODE_SU_TX_BFER, HECAP_PHY_SUBFMR_GET(hecap_phy_ptr)) |
-		 FIELD_PREP(HE_MODE_UL_MUMIMO, HECAP_PHY_ULMUMIMO_GET(hecap_phy_ptr));
-
-	/* TODO WDS and other modes */
-	if (viftype == NL80211_IFTYPE_AP) {
-		hemode |= FIELD_PREP(HE_MODE_MU_TX_BFER,
-			  HECAP_PHY_MUBFMR_GET(hecap_phy_ptr)) |
-			  FIELD_PREP(HE_MODE_DL_OFDMA, HE_DL_MUOFDMA_ENABLE) |
-			  FIELD_PREP(HE_MODE_UL_OFDMA, HE_UL_MUOFDMA_ENABLE);
-	} else {
-		hemode |= FIELD_PREP(HE_MODE_MU_TX_BFEE, HE_MU_BFEE_ENABLE);
-	}
-
-	return hemode;
-}
-
-static int ath11k_set_he_mu_sounding_mode(struct ath11k *ar,
-					  struct ath11k_vif *arvif)
-{
-	u32 param_id, param_value;
-	struct ath11k_base *ab = ar->ab;
-	int ret = 0;
-
-	param_id = WMI_VDEV_PARAM_SET_HEMU_MODE;
-	param_value = ath11k_mac_prepare_he_mode(ar->pdev, arvif->vif->type);
-	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
-					    param_id, param_value);
-	if (ret) {
-		ath11k_warn(ab, "failed to set vdev %d HE MU mode: %d param_value %x\n",
-			    arvif->vdev_id, ret, param_value);
-		return ret;
-	}
-	param_id = WMI_VDEV_PARAM_SET_HE_SOUNDING_MODE;
-	param_value =
-		FIELD_PREP(HE_VHT_SOUNDING_MODE, HE_VHT_SOUNDING_MODE_ENABLE) |
-		FIELD_PREP(HE_TRIG_NONTRIG_SOUNDING_MODE,
-			   HE_TRIG_NONTRIG_SOUNDING_MODE_ENABLE);
-	ret = ath11k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id,
-					    param_id, param_value);
-	if (ret) {
-		ath11k_warn(ab, "failed to set vdev %d HE MU mode: %d\n",
-			    arvif->vdev_id, ret);
-		return ret;
-	}
-	return ret;
-}
-
 static void ath11k_mac_op_update_vif_offload(struct ieee80211_hw *hw,
 					     struct ieee80211_vif *vif)
 {
@@ -6757,7 +6821,6 @@ ath11k_mac_vdev_start_restart(struct ath11k_vif *arvif,
 	struct ath11k_base *ab = ar->ab;
 	struct wmi_vdev_start_req_arg arg = {};
 	const struct cfg80211_chan_def *chandef = &ctx->def;
-	int he_support = arvif->vif->bss_conf.he_support;
 	int ret = 0;
 
 	lockdep_assert_held(&ar->conf_mutex);
@@ -6798,15 +6861,6 @@ ath11k_mac_vdev_start_restart(struct ath11k_vif *arvif,
 		spin_lock_bh(&ab->base_lock);
 		arg.regdomain = ar->ab->dfs_region;
 		spin_unlock_bh(&ab->base_lock);
-
-		if (he_support) {
-			ret = ath11k_set_he_mu_sounding_mode(ar, arvif);
-			if (ret) {
-				ath11k_warn(ar->ab, "failed to set he mode vdev %i\n",
-					    arg.vdev_id);
-				return ret;
-			}
-		}
 	}
 
 	arg.channel.passive |= !!(chandef->chan->flags & IEEE80211_CHAN_NO_IR);
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 9f50be411a8c..d3b74f1346f5 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2897,8 +2897,11 @@ struct rx_reorder_queue_remove_params {
 #define HE_DL_MUOFDMA_ENABLE	1
 #define HE_UL_MUOFDMA_ENABLE	1
 #define HE_DL_MUMIMO_ENABLE	1
+#define HE_UL_MUMIMO_ENABLE	1
 #define HE_MU_BFEE_ENABLE	1
 #define HE_SU_BFEE_ENABLE	1
+#define HE_MU_BFER_ENABLE	1
+#define HE_SU_BFER_ENABLE	1
 
 #define HE_VHT_SOUNDING_MODE_ENABLE		1
 #define HE_SU_MU_SOUNDING_MODE_ENABLE		1
-- 
cgit v1.2.3-70-g09d2


From 8077c1bbbc28e527fb29143c46f32c6a9d6cadf0 Mon Sep 17 00:00:00 2001
From: Muna Sinada <quic_msinada@quicinc.com>
Date: Fri, 24 Feb 2023 12:28:04 +0200
Subject: wifi: ath11k: move HE MCS mapper to a separate function

Move HE MCS mapper to a separate function and call new function
in ath11k_mac_copy_he_cap().

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.4.0.1-00356-QCAHKSWPL_SILICONZ-1

Signed-off-by: Muna Sinada <quic_msinada@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/1666128501-12364-4-git-send-email-quic_msinada@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 34 ++++++++++++++++++++++------------
 1 file changed, 22 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 7c4f424180dc..0835ab8b389a 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5483,6 +5483,27 @@ static __le16 ath11k_mac_setup_he_6ghz_cap(struct ath11k_pdev_cap *pcap,
 	return cpu_to_le16(bcap->he_6ghz_capa);
 }
 
+static void ath11k_mac_set_hemcsmap(struct ath11k *ar,
+				    struct ath11k_pdev_cap *cap,
+				    struct ieee80211_sta_he_cap *he_cap,
+				    int band)
+{
+	struct ath11k_band_cap *band_cap = &cap->band[band];
+
+	he_cap->he_mcs_nss_supp.rx_mcs_80 =
+		cpu_to_le16(band_cap->he_mcs & 0xffff);
+	he_cap->he_mcs_nss_supp.tx_mcs_80 =
+		cpu_to_le16(band_cap->he_mcs & 0xffff);
+	he_cap->he_mcs_nss_supp.rx_mcs_160 =
+		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+	he_cap->he_mcs_nss_supp.tx_mcs_160 =
+		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+	he_cap->he_mcs_nss_supp.rx_mcs_80p80 =
+		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+	he_cap->he_mcs_nss_supp.tx_mcs_80p80 =
+		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+}
+
 static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 				  struct ath11k_pdev_cap *cap,
 				  struct ieee80211_sband_iftype_data *data,
@@ -5544,18 +5565,7 @@ static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 			break;
 		}
 
-		he_cap->he_mcs_nss_supp.rx_mcs_80 =
-			cpu_to_le16(band_cap->he_mcs & 0xffff);
-		he_cap->he_mcs_nss_supp.tx_mcs_80 =
-			cpu_to_le16(band_cap->he_mcs & 0xffff);
-		he_cap->he_mcs_nss_supp.rx_mcs_160 =
-			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
-		he_cap->he_mcs_nss_supp.tx_mcs_160 =
-			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
-		he_cap->he_mcs_nss_supp.rx_mcs_80p80 =
-			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
-		he_cap->he_mcs_nss_supp.tx_mcs_80p80 =
-			cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		ath11k_mac_set_hemcsmap(ar, cap, he_cap, band);
 
 		memset(he_cap->ppe_thres, 0, sizeof(he_cap->ppe_thres));
 		if (he_cap_elem->phy_cap_info[6] &
-- 
cgit v1.2.3-70-g09d2


From ebf82988f844dd98e6b007cffcc5e95986056995 Mon Sep 17 00:00:00 2001
From: Muna Sinada <quic_msinada@quicinc.com>
Date: Fri, 24 Feb 2023 12:28:04 +0200
Subject: wifi: ath11k: generate rx and tx mcs maps for supported HE mcs

Generate rx and tx mcs maps in ath11k_mac_set_hemcsmap() and set them
in supported mcs/nss for HE capabilities.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.4.0.1-00356-QCAHKSWPL_SILICONZ-1

Signed-off-by: Muna Sinada <quic_msinada@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/1666128501-12364-5-git-send-email-quic_msinada@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 30 +++++++++++++++++++++++-------
 1 file changed, 23 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 0835ab8b389a..fd1a23ead5e2 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5488,20 +5488,36 @@ static void ath11k_mac_set_hemcsmap(struct ath11k *ar,
 				    struct ieee80211_sta_he_cap *he_cap,
 				    int band)
 {
-	struct ath11k_band_cap *band_cap = &cap->band[band];
+	u16 txmcs_map, rxmcs_map;
+	u32 i;
 
+	rxmcs_map = 0;
+	txmcs_map = 0;
+	for (i = 0; i < 8; i++) {
+		if (i < ar->num_tx_chains &&
+		    (ar->cfg_tx_chainmask >> cap->tx_chain_mask_shift) & BIT(i))
+			txmcs_map |= IEEE80211_HE_MCS_SUPPORT_0_11 << (i * 2);
+		else
+			txmcs_map |= IEEE80211_HE_MCS_NOT_SUPPORTED << (i * 2);
+
+		if (i < ar->num_rx_chains &&
+		    (ar->cfg_rx_chainmask >> cap->tx_chain_mask_shift) & BIT(i))
+			rxmcs_map |= IEEE80211_HE_MCS_SUPPORT_0_11 << (i * 2);
+		else
+			rxmcs_map |= IEEE80211_HE_MCS_NOT_SUPPORTED << (i * 2);
+	}
 	he_cap->he_mcs_nss_supp.rx_mcs_80 =
-		cpu_to_le16(band_cap->he_mcs & 0xffff);
+		cpu_to_le16(rxmcs_map & 0xffff);
 	he_cap->he_mcs_nss_supp.tx_mcs_80 =
-		cpu_to_le16(band_cap->he_mcs & 0xffff);
+		cpu_to_le16(txmcs_map & 0xffff);
 	he_cap->he_mcs_nss_supp.rx_mcs_160 =
-		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		cpu_to_le16(rxmcs_map & 0xffff);
 	he_cap->he_mcs_nss_supp.tx_mcs_160 =
-		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		cpu_to_le16(txmcs_map & 0xffff);
 	he_cap->he_mcs_nss_supp.rx_mcs_80p80 =
-		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		cpu_to_le16(rxmcs_map & 0xffff);
 	he_cap->he_mcs_nss_supp.tx_mcs_80p80 =
-		cpu_to_le16((band_cap->he_mcs >> 16) & 0xffff);
+		cpu_to_le16(txmcs_map & 0xffff);
 }
 
 static int ath11k_mac_copy_he_cap(struct ath11k *ar,
-- 
cgit v1.2.3-70-g09d2


From 3c7c07ca7ab144ef25402858078b27806322b752 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:14 +0900
Subject: wifi: brcmfmac: chip: Only disable D11 cores; handle an arbitrary
 number

At least on BCM4387, the D11 cores are held in reset on cold startup and
firmware expects to release reset itself. Just assert reset here and let
firmware deassert it. Premature deassertion results in the firmware
failing to initialize properly some of the time, with strange AXI bus
errors.

Also, BCM4387 has 3 cores, up from 2. The logic for handling that is in
brcmf_chip_ai_resetcore(), but since we aren't using that any more, just
handle it here.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-1-marcan@marcan.st
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index 8073f31be27d..a6239051404b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -1292,15 +1292,18 @@ static bool brcmf_chip_cm3_set_active(struct brcmf_chip_priv *chip)
 static inline void
 brcmf_chip_cr4_set_passive(struct brcmf_chip_priv *chip)
 {
+	int i;
 	struct brcmf_core *core;
 
 	brcmf_chip_disable_arm(chip, BCMA_CORE_ARM_CR4);
 
-	core = brcmf_chip_get_core(&chip->pub, BCMA_CORE_80211);
-	brcmf_chip_resetcore(core, D11_BCMA_IOCTL_PHYRESET |
-				   D11_BCMA_IOCTL_PHYCLOCKEN,
-			     D11_BCMA_IOCTL_PHYCLOCKEN,
-			     D11_BCMA_IOCTL_PHYCLOCKEN);
+	/* Disable the cores only and let the firmware enable them.
+	 * Releasing reset ourselves breaks BCM4387 in weird ways.
+	 */
+	for (i = 0; (core = brcmf_chip_get_d11core(&chip->pub, i)); i++)
+		brcmf_chip_coredisable(core, D11_BCMA_IOCTL_PHYRESET |
+				       D11_BCMA_IOCTL_PHYCLOCKEN,
+				       D11_BCMA_IOCTL_PHYCLOCKEN);
 }
 
 static bool brcmf_chip_cr4_set_active(struct brcmf_chip_priv *chip, u32 rstvec)
-- 
cgit v1.2.3-70-g09d2


From 098e0b105ce1047ad9984dc79287573e313b1232 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:15 +0900
Subject: wifi: brcmfmac: chip: Handle 1024-unit sizes for TCM blocks

BCM4387 has trailing odd-sized blocks as part of TCM which have
their size described as a multiple of 1024 instead of 8192. Handle this
so we can compute the TCM size properly.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-2-marcan@marcan.st
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index a6239051404b..50e0c015cf43 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -212,8 +212,9 @@ struct sbsocramregs {
 #define	ARMCR4_TCBANB_MASK	0xf
 #define	ARMCR4_TCBANB_SHIFT	0
 
-#define	ARMCR4_BSZ_MASK		0x3f
+#define	ARMCR4_BSZ_MASK		0x7f
 #define	ARMCR4_BSZ_MULT		8192
+#define	ARMCR4_BLK_1K_MASK	0x200
 
 struct brcmf_core_priv {
 	struct brcmf_core pub;
@@ -684,6 +685,7 @@ static u32 brcmf_chip_tcm_ramsize(struct brcmf_core_priv *cr4)
 	u32 nbb;
 	u32 totb;
 	u32 bxinfo;
+	u32 blksize;
 	u32 idx;
 
 	corecap = brcmf_chip_core_read32(cr4, ARMCR4_CAP);
@@ -695,7 +697,11 @@ static u32 brcmf_chip_tcm_ramsize(struct brcmf_core_priv *cr4)
 	for (idx = 0; idx < totb; idx++) {
 		brcmf_chip_core_write32(cr4, ARMCR4_BANKIDX, idx);
 		bxinfo = brcmf_chip_core_read32(cr4, ARMCR4_BANKINFO);
-		memsize += ((bxinfo & ARMCR4_BSZ_MASK) + 1) * ARMCR4_BSZ_MULT;
+		blksize = ARMCR4_BSZ_MULT;
+		if (bxinfo & ARMCR4_BLK_1K_MASK)
+			blksize >>= 3;
+
+		memsize += ((bxinfo & ARMCR4_BSZ_MASK) + 1) * blksize;
 	}
 
 	return memsize;
-- 
cgit v1.2.3-70-g09d2


From 398ce273d6b16a57dee99e4054a2be37f0a958ed Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:16 +0900
Subject: wifi: brcmfmac: cfg80211: Add support for scan params v2

This new API version is required for at least the BCM4387 firmware. Add
support for it, with a fallback to the v1 API.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-3-marcan@marcan.st
---
 .../broadcom/brcm80211/brcmfmac/cfg80211.c         | 245 +++++++++++++--------
 .../wireless/broadcom/brcm80211/brcmfmac/feature.c |   1 +
 .../wireless/broadcom/brcm80211/brcmfmac/feature.h |   4 +-
 .../broadcom/brcm80211/brcmfmac/fwil_types.h       |  49 ++++-
 4 files changed, 209 insertions(+), 90 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index d80d2b9b15c7..77706e204a8f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -1039,12 +1039,134 @@ void brcmf_set_mpc(struct brcmf_if *ifp, int mpc)
 	}
 }
 
+static void brcmf_scan_params_v2_to_v1(struct brcmf_scan_params_v2_le *params_v2_le,
+				       struct brcmf_scan_params_le *params_le)
+{
+	size_t params_size;
+	u32 ch;
+	int n_channels, n_ssids;
+
+	memcpy(&params_le->ssid_le, &params_v2_le->ssid_le,
+	       sizeof(params_le->ssid_le));
+	memcpy(&params_le->bssid, &params_v2_le->bssid,
+	       sizeof(params_le->bssid));
+
+	params_le->bss_type = params_v2_le->bss_type;
+	params_le->scan_type = le32_to_cpu(params_v2_le->scan_type);
+	params_le->nprobes = params_v2_le->nprobes;
+	params_le->active_time = params_v2_le->active_time;
+	params_le->passive_time = params_v2_le->passive_time;
+	params_le->home_time = params_v2_le->home_time;
+	params_le->channel_num = params_v2_le->channel_num;
+
+	ch = le32_to_cpu(params_v2_le->channel_num);
+	n_channels = ch & BRCMF_SCAN_PARAMS_COUNT_MASK;
+	n_ssids = ch >> BRCMF_SCAN_PARAMS_NSSID_SHIFT;
+
+	params_size = sizeof(u16) * n_channels;
+	if (n_ssids > 0) {
+		params_size = roundup(params_size, sizeof(u32));
+		params_size += sizeof(struct brcmf_ssid_le) * n_ssids;
+	}
+
+	memcpy(&params_le->channel_list[0],
+	       &params_v2_le->channel_list[0], params_size);
+}
+
+static void brcmf_escan_prep(struct brcmf_cfg80211_info *cfg,
+			     struct brcmf_scan_params_v2_le *params_le,
+			     struct cfg80211_scan_request *request)
+{
+	u32 n_ssids;
+	u32 n_channels;
+	s32 i;
+	s32 offset;
+	u16 chanspec;
+	char *ptr;
+	int length;
+	struct brcmf_ssid_le ssid_le;
+
+	eth_broadcast_addr(params_le->bssid);
+
+	length = BRCMF_SCAN_PARAMS_V2_FIXED_SIZE;
+
+	params_le->version = cpu_to_le16(BRCMF_SCAN_PARAMS_VERSION_V2);
+	params_le->bss_type = DOT11_BSSTYPE_ANY;
+	params_le->scan_type = cpu_to_le32(BRCMF_SCANTYPE_ACTIVE);
+	params_le->channel_num = 0;
+	params_le->nprobes = cpu_to_le32(-1);
+	params_le->active_time = cpu_to_le32(-1);
+	params_le->passive_time = cpu_to_le32(-1);
+	params_le->home_time = cpu_to_le32(-1);
+	memset(&params_le->ssid_le, 0, sizeof(params_le->ssid_le));
+
+	/* Scan abort */
+	if (!request) {
+		length += sizeof(u16);
+		params_le->channel_num = cpu_to_le32(1);
+		params_le->channel_list[0] = cpu_to_le16(-1);
+		params_le->length = cpu_to_le16(length);
+		return;
+	}
+
+	n_ssids = request->n_ssids;
+	n_channels = request->n_channels;
+
+	/* Copy channel array if applicable */
+	brcmf_dbg(SCAN, "### List of channelspecs to scan ### %d\n",
+		  n_channels);
+	if (n_channels > 0) {
+		length += roundup(sizeof(u16) * n_channels, sizeof(u32));
+		for (i = 0; i < n_channels; i++) {
+			chanspec = channel_to_chanspec(&cfg->d11inf,
+						       request->channels[i]);
+			brcmf_dbg(SCAN, "Chan : %d, Channel spec: %x\n",
+				  request->channels[i]->hw_value, chanspec);
+			params_le->channel_list[i] = cpu_to_le16(chanspec);
+		}
+	} else {
+		brcmf_dbg(SCAN, "Scanning all channels\n");
+	}
+
+	/* Copy ssid array if applicable */
+	brcmf_dbg(SCAN, "### List of SSIDs to scan ### %d\n", n_ssids);
+	if (n_ssids > 0) {
+		offset = offsetof(struct brcmf_scan_params_v2_le, channel_list) +
+				n_channels * sizeof(u16);
+		offset = roundup(offset, sizeof(u32));
+		length += sizeof(ssid_le) * n_ssids,
+		ptr = (char *)params_le + offset;
+		for (i = 0; i < n_ssids; i++) {
+			memset(&ssid_le, 0, sizeof(ssid_le));
+			ssid_le.SSID_len =
+					cpu_to_le32(request->ssids[i].ssid_len);
+			memcpy(ssid_le.SSID, request->ssids[i].ssid,
+			       request->ssids[i].ssid_len);
+			if (!ssid_le.SSID_len)
+				brcmf_dbg(SCAN, "%d: Broadcast scan\n", i);
+			else
+				brcmf_dbg(SCAN, "%d: scan for  %.32s size=%d\n",
+					  i, ssid_le.SSID, ssid_le.SSID_len);
+			memcpy(ptr, &ssid_le, sizeof(ssid_le));
+			ptr += sizeof(ssid_le);
+		}
+	} else {
+		brcmf_dbg(SCAN, "Performing passive scan\n");
+		params_le->scan_type = cpu_to_le32(BRCMF_SCANTYPE_PASSIVE);
+	}
+	params_le->length = cpu_to_le16(length);
+	/* Adding mask to channel numbers */
+	params_le->channel_num =
+		cpu_to_le32((n_ssids << BRCMF_SCAN_PARAMS_NSSID_SHIFT) |
+			(n_channels & BRCMF_SCAN_PARAMS_COUNT_MASK));
+}
+
 s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg,
 				struct brcmf_if *ifp, bool aborted,
 				bool fw_abort)
 {
 	struct brcmf_pub *drvr = cfg->pub;
-	struct brcmf_scan_params_le params_le;
+	struct brcmf_scan_params_v2_le params_v2_le;
 	struct cfg80211_scan_request *scan_request;
 	u64 reqid;
 	u32 bucket;
@@ -1063,20 +1185,23 @@ s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg,
 	if (fw_abort) {
 		/* Do a scan abort to stop the driver's scan engine */
 		brcmf_dbg(SCAN, "ABORT scan in firmware\n");
-		memset(&params_le, 0, sizeof(params_le));
-		eth_broadcast_addr(params_le.bssid);
-		params_le.bss_type = DOT11_BSSTYPE_ANY;
-		params_le.scan_type = 0;
-		params_le.channel_num = cpu_to_le32(1);
-		params_le.nprobes = cpu_to_le32(1);
-		params_le.active_time = cpu_to_le32(-1);
-		params_le.passive_time = cpu_to_le32(-1);
-		params_le.home_time = cpu_to_le32(-1);
-		/* Scan is aborted by setting channel_list[0] to -1 */
-		params_le.channel_list[0] = cpu_to_le16(-1);
+
+		brcmf_escan_prep(cfg, &params_v2_le, NULL);
+
 		/* E-Scan (or anyother type) can be aborted by SCAN */
-		err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SCAN,
-					     &params_le, sizeof(params_le));
+		if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_SCAN_V2)) {
+			err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SCAN,
+						     &params_v2_le,
+						     sizeof(params_v2_le));
+		} else {
+			struct brcmf_scan_params_le params_le;
+
+			brcmf_scan_params_v2_to_v1(&params_v2_le, &params_le);
+			err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SCAN,
+						     &params_le,
+						     sizeof(params_le));
+		}
+
 		if (err)
 			bphy_err(drvr, "Scan abort failed\n");
 	}
@@ -1295,83 +1420,13 @@ done:
 	return err;
 }
 
-static void brcmf_escan_prep(struct brcmf_cfg80211_info *cfg,
-			     struct brcmf_scan_params_le *params_le,
-			     struct cfg80211_scan_request *request)
-{
-	u32 n_ssids;
-	u32 n_channels;
-	s32 i;
-	s32 offset;
-	u16 chanspec;
-	char *ptr;
-	struct brcmf_ssid_le ssid_le;
-
-	eth_broadcast_addr(params_le->bssid);
-	params_le->bss_type = DOT11_BSSTYPE_ANY;
-	params_le->scan_type = BRCMF_SCANTYPE_ACTIVE;
-	params_le->channel_num = 0;
-	params_le->nprobes = cpu_to_le32(-1);
-	params_le->active_time = cpu_to_le32(-1);
-	params_le->passive_time = cpu_to_le32(-1);
-	params_le->home_time = cpu_to_le32(-1);
-	memset(&params_le->ssid_le, 0, sizeof(params_le->ssid_le));
-
-	n_ssids = request->n_ssids;
-	n_channels = request->n_channels;
-
-	/* Copy channel array if applicable */
-	brcmf_dbg(SCAN, "### List of channelspecs to scan ### %d\n",
-		  n_channels);
-	if (n_channels > 0) {
-		for (i = 0; i < n_channels; i++) {
-			chanspec = channel_to_chanspec(&cfg->d11inf,
-						       request->channels[i]);
-			brcmf_dbg(SCAN, "Chan : %d, Channel spec: %x\n",
-				  request->channels[i]->hw_value, chanspec);
-			params_le->channel_list[i] = cpu_to_le16(chanspec);
-		}
-	} else {
-		brcmf_dbg(SCAN, "Scanning all channels\n");
-	}
-	/* Copy ssid array if applicable */
-	brcmf_dbg(SCAN, "### List of SSIDs to scan ### %d\n", n_ssids);
-	if (n_ssids > 0) {
-		offset = offsetof(struct brcmf_scan_params_le, channel_list) +
-				n_channels * sizeof(u16);
-		offset = roundup(offset, sizeof(u32));
-		ptr = (char *)params_le + offset;
-		for (i = 0; i < n_ssids; i++) {
-			memset(&ssid_le, 0, sizeof(ssid_le));
-			ssid_le.SSID_len =
-					cpu_to_le32(request->ssids[i].ssid_len);
-			memcpy(ssid_le.SSID, request->ssids[i].ssid,
-			       request->ssids[i].ssid_len);
-			if (!ssid_le.SSID_len)
-				brcmf_dbg(SCAN, "%d: Broadcast scan\n", i);
-			else
-				brcmf_dbg(SCAN, "%d: scan for  %.32s size=%d\n",
-					  i, ssid_le.SSID, ssid_le.SSID_len);
-			memcpy(ptr, &ssid_le, sizeof(ssid_le));
-			ptr += sizeof(ssid_le);
-		}
-	} else {
-		brcmf_dbg(SCAN, "Performing passive scan\n");
-		params_le->scan_type = BRCMF_SCANTYPE_PASSIVE;
-	}
-	/* Adding mask to channel numbers */
-	params_le->channel_num =
-		cpu_to_le32((n_ssids << BRCMF_SCAN_PARAMS_NSSID_SHIFT) |
-			(n_channels & BRCMF_SCAN_PARAMS_COUNT_MASK));
-}
-
 static s32
 brcmf_run_escan(struct brcmf_cfg80211_info *cfg, struct brcmf_if *ifp,
 		struct cfg80211_scan_request *request)
 {
 	struct brcmf_pub *drvr = cfg->pub;
-	s32 params_size = BRCMF_SCAN_PARAMS_FIXED_SIZE +
-			  offsetof(struct brcmf_escan_params_le, params_le);
+	s32 params_size = BRCMF_SCAN_PARAMS_V2_FIXED_SIZE +
+			  offsetof(struct brcmf_escan_params_le, params_v2_le);
 	struct brcmf_escan_params_le *params;
 	s32 err = 0;
 
@@ -1391,8 +1446,22 @@ brcmf_run_escan(struct brcmf_cfg80211_info *cfg, struct brcmf_if *ifp,
 		goto exit;
 	}
 	BUG_ON(params_size + sizeof("escan") >= BRCMF_DCMD_MEDLEN);
-	brcmf_escan_prep(cfg, &params->params_le, request);
-	params->version = cpu_to_le32(BRCMF_ESCAN_REQ_VERSION);
+	brcmf_escan_prep(cfg, &params->params_v2_le, request);
+
+	params->version = cpu_to_le32(BRCMF_ESCAN_REQ_VERSION_V2);
+
+	if (!brcmf_feat_is_enabled(ifp, BRCMF_FEAT_SCAN_V2)) {
+		struct brcmf_escan_params_le *params_v1;
+
+		params_size -= BRCMF_SCAN_PARAMS_V2_FIXED_SIZE;
+		params_size += BRCMF_SCAN_PARAMS_FIXED_SIZE;
+		params_v1 = kzalloc(params_size, GFP_KERNEL);
+		params_v1->version = cpu_to_le32(BRCMF_ESCAN_REQ_VERSION);
+		brcmf_scan_params_v2_to_v1(&params->params_v2_le, &params_v1->params_le);
+		kfree(params);
+		params = params_v1;
+	}
+
 	params->action = cpu_to_le16(WL_ESCAN_ACTION_START);
 	params->sync_id = cpu_to_le16(0x1234);
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
index 10bac865d724..b6797f800e55 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
@@ -290,6 +290,7 @@ void brcmf_feat_attach(struct brcmf_pub *drvr)
 		ifp->drvr->feat_flags |= BIT(BRCMF_FEAT_SCAN_RANDOM_MAC);
 
 	brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_FWSUP, "sup_wpa");
+	brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_SCAN_V2, "scan_ver");
 
 	if (drvr->settings->feature_disable) {
 		brcmf_dbg(INFO, "Features: 0x%02x, disable: 0x%02x\n",
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
index f1b086a69d73..549298c55b55 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
@@ -30,6 +30,7 @@
  * SAE: simultaneous authentication of equals
  * FWAUTH: Firmware authenticator
  * DUMP_OBSS: Firmware has capable to dump obss info to support ACS
+ * SCAN_V2: Version 2 scan params
  */
 #define BRCMF_FEAT_LIST \
 	BRCMF_FEAT_DEF(MBSS) \
@@ -53,7 +54,8 @@
 	BRCMF_FEAT_DEF(DOT11H) \
 	BRCMF_FEAT_DEF(SAE) \
 	BRCMF_FEAT_DEF(FWAUTH) \
-	BRCMF_FEAT_DEF(DUMP_OBSS)
+	BRCMF_FEAT_DEF(DUMP_OBSS) \
+	BRCMF_FEAT_DEF(SCAN_V2)
 
 /*
  * Quirks:
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
index 04e1beedfd81..b3844d0d1adb 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
@@ -48,6 +48,10 @@
 
 /* size of brcmf_scan_params not including variable length array */
 #define BRCMF_SCAN_PARAMS_FIXED_SIZE	64
+#define BRCMF_SCAN_PARAMS_V2_FIXED_SIZE	72
+
+/* version of brcmf_scan_params structure */
+#define BRCMF_SCAN_PARAMS_VERSION_V2	2
 
 /* masks for channel and ssid count */
 #define BRCMF_SCAN_PARAMS_COUNT_MASK	0x0000ffff
@@ -67,6 +71,7 @@
 #define BRCMF_PRIMARY_KEY		(1 << 1)
 #define DOT11_BSSTYPE_ANY		2
 #define BRCMF_ESCAN_REQ_VERSION		1
+#define BRCMF_ESCAN_REQ_VERSION_V2	2
 
 #define BRCMF_MAXRATES_IN_SET		16	/* max # of rates in rateset */
 
@@ -386,6 +391,45 @@ struct brcmf_scan_params_le {
 	__le16 channel_list[1];	/* list of chanspecs */
 };
 
+struct brcmf_scan_params_v2_le {
+	__le16 version;		/* structure version */
+	__le16 length;		/* structure length */
+	struct brcmf_ssid_le ssid_le;	/* default: {0, ""} */
+	u8 bssid[ETH_ALEN];	/* default: bcast */
+	s8 bss_type;		/* default: any,
+				 * DOT11_BSSTYPE_ANY/INFRASTRUCTURE/INDEPENDENT
+				 */
+	u8 pad;
+	__le32 scan_type;	/* flags, 0 use default */
+	__le32 nprobes;		/* -1 use default, number of probes per channel */
+	__le32 active_time;	/* -1 use default, dwell time per channel for
+				 * active scanning
+				 */
+	__le32 passive_time;	/* -1 use default, dwell time per channel
+				 * for passive scanning
+				 */
+	__le32 home_time;	/* -1 use default, dwell time for the
+				 * home channel between channel scans
+				 */
+	__le32 channel_num;	/* count of channels and ssids that follow
+				 *
+				 * low half is count of channels in
+				 * channel_list, 0 means default (use all
+				 * available channels)
+				 *
+				 * high half is entries in struct brcmf_ssid
+				 * array that follows channel_list, aligned for
+				 * s32 (4 bytes) meaning an odd channel count
+				 * implies a 2-byte pad between end of
+				 * channel_list and first ssid
+				 *
+				 * if ssid count is zero, single ssid in the
+				 * fixed parameter portion is assumed, otherwise
+				 * ssid in the fixed portion is ignored
+				 */
+	__le16 channel_list[1];	/* list of chanspecs */
+};
+
 struct brcmf_scan_results {
 	u32 buflen;
 	u32 version;
@@ -397,7 +441,10 @@ struct brcmf_escan_params_le {
 	__le32 version;
 	__le16 action;
 	__le16 sync_id;
-	struct brcmf_scan_params_le params_le;
+	union {
+		struct brcmf_scan_params_le params_le;
+		struct brcmf_scan_params_v2_le params_v2_le;
+	};
 };
 
 struct brcmf_escan_result_le {
-- 
cgit v1.2.3-70-g09d2


From d75ef1f81e42dfccfeb97952e2ab70376832cf7a Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:17 +0900
Subject: wifi: brcmfmac: feature: Add support for setting feats based on WLC
 version

The "wlc_ver" iovar returns information on the WLC and EPI versions.
This can be used to determine whether the PMKID_V2 and _V3 features are
supported.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-4-marcan@marcan.st
---
 .../wireless/broadcom/brcm80211/brcmfmac/feature.c | 48 ++++++++++++++++++++++
 .../wireless/broadcom/brcm80211/brcmfmac/feature.h |  4 +-
 .../broadcom/brcm80211/brcmfmac/fwil_types.h       | 25 +++++++++++
 3 files changed, 76 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
index b6797f800e55..6d10c9efbe93 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
@@ -126,6 +126,53 @@ static void brcmf_feat_firmware_overrides(struct brcmf_pub *drv)
 	drv->feat_flags |= feat_flags;
 }
 
+struct brcmf_feat_wlcfeat {
+	u16 min_ver_major;
+	u16 min_ver_minor;
+	u32 feat_flags;
+};
+
+static const struct brcmf_feat_wlcfeat brcmf_feat_wlcfeat_map[] = {
+	{ 12, 0, BIT(BRCMF_FEAT_PMKID_V2) },
+	{ 13, 0, BIT(BRCMF_FEAT_PMKID_V3) },
+};
+
+static void brcmf_feat_wlc_version_overrides(struct brcmf_pub *drv)
+{
+	struct brcmf_if *ifp = brcmf_get_ifp(drv, 0);
+	const struct brcmf_feat_wlcfeat *e;
+	struct brcmf_wlc_version_le ver;
+	u32 feat_flags = 0;
+	int i, err, major, minor;
+
+	err = brcmf_fil_iovar_data_get(ifp, "wlc_ver", &ver, sizeof(ver));
+	if (err)
+		return;
+
+	major = le16_to_cpu(ver.wlc_ver_major);
+	minor = le16_to_cpu(ver.wlc_ver_minor);
+
+	brcmf_dbg(INFO, "WLC version: %d.%d\n", major, minor);
+
+	for (i = 0; i < ARRAY_SIZE(brcmf_feat_wlcfeat_map); i++) {
+		e = &brcmf_feat_wlcfeat_map[i];
+		if (major > e->min_ver_major ||
+		    (major == e->min_ver_major &&
+		     minor >= e->min_ver_minor)) {
+			feat_flags |= e->feat_flags;
+		}
+	}
+
+	if (!feat_flags)
+		return;
+
+	for (i = 0; i < BRCMF_FEAT_LAST; i++)
+		if (feat_flags & BIT(i))
+			brcmf_dbg(INFO, "enabling firmware feature: %s\n",
+				  brcmf_feat_names[i]);
+	drv->feat_flags |= feat_flags;
+}
+
 /**
  * brcmf_feat_iovar_int_get() - determine feature through iovar query.
  *
@@ -299,6 +346,7 @@ void brcmf_feat_attach(struct brcmf_pub *drvr)
 		ifp->drvr->feat_flags &= ~drvr->settings->feature_disable;
 	}
 
+	brcmf_feat_wlc_version_overrides(drvr);
 	brcmf_feat_firmware_overrides(drvr);
 
 	/* set chip related quirks */
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
index 549298c55b55..7f4f0b3e4a7b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
@@ -55,7 +55,9 @@
 	BRCMF_FEAT_DEF(SAE) \
 	BRCMF_FEAT_DEF(FWAUTH) \
 	BRCMF_FEAT_DEF(DUMP_OBSS) \
-	BRCMF_FEAT_DEF(SCAN_V2)
+	BRCMF_FEAT_DEF(SCAN_V2) \
+	BRCMF_FEAT_DEF(PMKID_V2) \
+	BRCMF_FEAT_DEF(PMKID_V3)
 
 /*
  * Quirks:
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
index b3844d0d1adb..801709c26b7b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
@@ -788,6 +788,31 @@ struct brcmf_rev_info_le {
 	__le32 nvramrev;
 };
 
+/**
+ * struct brcmf_wlc_version_le - firmware revision info.
+ *
+ * @version: structure version.
+ * @length: structure length.
+ * @epi_ver_major: EPI major version
+ * @epi_ver_minor: EPI minor version
+ * @epi_ver_rc: EPI rc version
+ * @epi_ver_incr: EPI increment version
+ * @wlc_ver_major: WLC major version
+ * @wlc_ver_minor: WLC minor version
+ */
+struct brcmf_wlc_version_le {
+	__le16 version;
+	__le16 length;
+
+	__le16 epi_ver_major;
+	__le16 epi_ver_minor;
+	__le16 epi_ver_rc;
+	__le16 epi_ver_incr;
+
+	__le16 wlc_ver_major;
+	__le16 wlc_ver_minor;
+};
+
 /**
  * struct brcmf_assoclist_le - request assoc list.
  *
-- 
cgit v1.2.3-70-g09d2


From a96202acaea47fa8377088e0952bb63bd02a3bab Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:18 +0900
Subject: wifi: brcmfmac: cfg80211: Add support for PMKID_V3 operations

Add support for the new PMKID_V3 API, which allows performing PMKID
mutations individually, instead of requiring the driver to keep track of
the full list. This new API is required by at least BCM4387.

Note that PMKID_V2 is not implemented yet.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-5-marcan@marcan.st
---
 .../broadcom/brcm80211/brcmfmac/cfg80211.c         | 52 +++++++++++++-
 .../broadcom/brcm80211/brcmfmac/fwil_types.h       | 83 ++++++++++++++++++++++
 2 files changed, 132 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 77706e204a8f..b561ca6b467e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -4306,6 +4306,37 @@ exit:
 	return 0;
 }
 
+static s32
+brcmf_pmksa_v3_op(struct brcmf_if *ifp, struct cfg80211_pmksa *pmksa,
+		  bool alive)
+{
+	struct brcmf_pmk_op_v3_le *pmk_op;
+	int length = offsetof(struct brcmf_pmk_op_v3_le, pmk);
+	int ret;
+
+	pmk_op = kzalloc(sizeof(*pmk_op), GFP_KERNEL);
+	pmk_op->version = cpu_to_le16(BRCMF_PMKSA_VER_3);
+
+	if (!pmksa) {
+		/* Flush operation, operate on entire list */
+		pmk_op->count = cpu_to_le16(0);
+	} else {
+		/* Single PMK operation */
+		pmk_op->count = cpu_to_le16(1);
+		length += sizeof(struct brcmf_pmksa_v3);
+		memcpy(pmk_op->pmk[0].bssid, pmksa->bssid, ETH_ALEN);
+		memcpy(pmk_op->pmk[0].pmkid, pmksa->pmkid, WLAN_PMKID_LEN);
+		pmk_op->pmk[0].pmkid_len = WLAN_PMKID_LEN;
+		pmk_op->pmk[0].time_left = cpu_to_le32(alive ? BRCMF_PMKSA_NO_EXPIRY : 0);
+	}
+
+	pmk_op->length = cpu_to_le16(length);
+
+	ret = brcmf_fil_iovar_data_set(ifp, "pmkid_info", pmk_op, sizeof(*pmk_op));
+	kfree(pmk_op);
+	return ret;
+}
+
 static __used s32
 brcmf_update_pmklist(struct brcmf_cfg80211_info *cfg, struct brcmf_if *ifp)
 {
@@ -4339,6 +4370,14 @@ brcmf_cfg80211_set_pmksa(struct wiphy *wiphy, struct net_device *ndev,
 	if (!check_vif_up(ifp->vif))
 		return -EIO;
 
+	brcmf_dbg(CONN, "set_pmksa - PMK bssid: %pM =\n", pmksa->bssid);
+	brcmf_dbg(CONN, "%*ph\n", WLAN_PMKID_LEN, pmksa->pmkid);
+
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_PMKID_V3))
+		return brcmf_pmksa_v3_op(ifp, pmksa, true);
+
+	/* TODO: implement PMKID_V2 */
+
 	npmk = le32_to_cpu(cfg->pmk_list.npmk);
 	for (i = 0; i < npmk; i++)
 		if (!memcmp(pmksa->bssid, pmk[i].bssid, ETH_ALEN))
@@ -4355,9 +4394,6 @@ brcmf_cfg80211_set_pmksa(struct wiphy *wiphy, struct net_device *ndev,
 		return -EINVAL;
 	}
 
-	brcmf_dbg(CONN, "set_pmksa - PMK bssid: %pM =\n", pmk[npmk].bssid);
-	brcmf_dbg(CONN, "%*ph\n", WLAN_PMKID_LEN, pmk[npmk].pmkid);
-
 	err = brcmf_update_pmklist(cfg, ifp);
 
 	brcmf_dbg(TRACE, "Exit\n");
@@ -4381,6 +4417,11 @@ brcmf_cfg80211_del_pmksa(struct wiphy *wiphy, struct net_device *ndev,
 
 	brcmf_dbg(CONN, "del_pmksa - PMK bssid = %pM\n", pmksa->bssid);
 
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_PMKID_V3))
+		return brcmf_pmksa_v3_op(ifp, pmksa, false);
+
+	/* TODO: implement PMKID_V2 */
+
 	npmk = le32_to_cpu(cfg->pmk_list.npmk);
 	for (i = 0; i < npmk; i++)
 		if (!memcmp(pmksa->bssid, pmk[i].bssid, ETH_ALEN))
@@ -4417,6 +4458,11 @@ brcmf_cfg80211_flush_pmksa(struct wiphy *wiphy, struct net_device *ndev)
 	if (!check_vif_up(ifp->vif))
 		return -EIO;
 
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_PMKID_V3))
+		return brcmf_pmksa_v3_op(ifp, NULL, false);
+
+	/* TODO: implement PMKID_V2 */
+
 	memset(&cfg->pmk_list, 0, sizeof(cfg->pmk_list));
 	err = brcmf_update_pmklist(cfg, ifp);
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
index 801709c26b7b..792adaf880b4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
@@ -174,6 +174,10 @@
 
 #define BRCMF_HE_CAP_MCS_MAP_NSS_MAX	8
 
+#define BRCMF_PMKSA_VER_2		2
+#define BRCMF_PMKSA_VER_3		3
+#define BRCMF_PMKSA_NO_EXPIRY		0xffffffff
+
 /* MAX_CHUNK_LEN is the maximum length for data passing to firmware in each
  * ioctl. It is relatively small because firmware has small maximum size input
  * playload restriction for ioctls.
@@ -355,6 +359,12 @@ struct brcmf_ssid_le {
 	unsigned char SSID[IEEE80211_MAX_SSID_LEN];
 };
 
+/* Alternate SSID structure used in some places... */
+struct brcmf_ssid8_le {
+	u8 SSID_len;
+	unsigned char SSID[IEEE80211_MAX_SSID_LEN];
+};
+
 struct brcmf_scan_params_le {
 	struct brcmf_ssid_le ssid_le;	/* default: {0, ""} */
 	u8 bssid[ETH_ALEN];	/* default: bcast */
@@ -875,6 +885,51 @@ struct brcmf_pmksa {
 	u8 pmkid[WLAN_PMKID_LEN];
 };
 
+/**
+ * struct brcmf_pmksa_v2 - PMK Security Association
+ *
+ * @length: Length of the structure.
+ * @bssid: The AP's BSSID.
+ * @pmkid: The PMK ID.
+ * @pmk: PMK material for FILS key derivation.
+ * @pmk_len: Length of PMK data.
+ * @ssid: The AP's SSID.
+ * @fils_cache_id: FILS cache identifier
+ */
+struct brcmf_pmksa_v2 {
+	__le16 length;
+	u8 bssid[ETH_ALEN];
+	u8 pmkid[WLAN_PMKID_LEN];
+	u8 pmk[WLAN_PMK_LEN_SUITE_B_192];
+	__le16 pmk_len;
+	struct brcmf_ssid8_le ssid;
+	u16 fils_cache_id;
+};
+
+/**
+ * struct brcmf_pmksa_v3 - PMK Security Association
+ *
+ * @bssid: The AP's BSSID.
+ * @pmkid: The PMK ID.
+ * @pmkid_len: The length of the PMK ID.
+ * @pmk: PMK material for FILS key derivation.
+ * @pmk_len: Length of PMK data.
+ * @fils_cache_id: FILS cache identifier
+ * @ssid: The AP's SSID.
+ * @time_left: Remaining time until expiry. 0 = expired, ~0 = no expiry.
+ */
+struct brcmf_pmksa_v3 {
+	u8 bssid[ETH_ALEN];
+	u8 pmkid[WLAN_PMKID_LEN];
+	u8 pmkid_len;
+	u8 pmk[WLAN_PMK_LEN_SUITE_B_192];
+	u8 pmk_len;
+	__le16 fils_cache_id;
+	u8 pad;
+	struct brcmf_ssid8_le ssid;
+	__le32 time_left;
+};
+
 /**
  * struct brcmf_pmk_list_le - List of pmksa's.
  *
@@ -886,6 +941,34 @@ struct brcmf_pmk_list_le {
 	struct brcmf_pmksa pmk[BRCMF_MAXPMKID];
 };
 
+/**
+ * struct brcmf_pmk_list_v2_le - List of pmksa's.
+ *
+ * @version: Request version.
+ * @length: Length of this structure.
+ * @pmk: PMK SA information.
+ */
+struct brcmf_pmk_list_v2_le {
+	__le16 version;
+	__le16 length;
+	struct brcmf_pmksa_v2 pmk[BRCMF_MAXPMKID];
+};
+
+/**
+ * struct brcmf_pmk_op_v3_le - Operation on PMKSA list.
+ *
+ * @version: Request version.
+ * @length: Length of this structure.
+ * @pmk: PMK SA information.
+ */
+struct brcmf_pmk_op_v3_le {
+	__le16 version;
+	__le16 length;
+	__le16 count;
+	__le16 pad;
+	struct brcmf_pmksa_v3 pmk[BRCMF_MAXPMKID];
+};
+
 /**
  * struct brcmf_pno_param_le - PNO scan configuration parameters
  *
-- 
cgit v1.2.3-70-g09d2


From 89b89e52153fda2733562776c7c9d9d3ebf8dd6d Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:19 +0900
Subject: wifi: brcmfmac: cfg80211: Pass the PMK in binary instead of hex

Apparently the hex passphrase mechanism does not work on newer
chips/firmware (e.g. BCM4387). It seems there was a simple way of
passing it in binary all along, so use that and avoid the hexification.

OpenBSD has been doing it like this from the beginning, so this should
work on all chips.

Also clear the structure before setting the PMK. This was leaking
uninitialized stack contents to the device.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-6-marcan@marcan.st
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index b561ca6b467e..e0a70a671550 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -1686,13 +1686,14 @@ static int brcmf_set_pmk(struct brcmf_if *ifp, const u8 *pmk_data, u16 pmk_len)
 {
 	struct brcmf_pub *drvr = ifp->drvr;
 	struct brcmf_wsec_pmk_le pmk;
-	int i, err;
+	int err;
+
+	memset(&pmk, 0, sizeof(pmk));
 
-	/* convert to firmware key format */
-	pmk.key_len = cpu_to_le16(pmk_len << 1);
-	pmk.flags = cpu_to_le16(BRCMF_WSEC_PASSPHRASE);
-	for (i = 0; i < pmk_len; i++)
-		snprintf(&pmk.key[2 * i], 3, "%02x", pmk_data[i]);
+	/* pass pmk directly */
+	pmk.key_len = cpu_to_le16(pmk_len);
+	pmk.flags = cpu_to_le16(0);
+	memcpy(pmk.key, pmk_data, pmk_len);
 
 	/* store psk in firmware */
 	err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SET_WSEC_PMK,
-- 
cgit v1.2.3-70-g09d2


From 117ace4014cce3fb78b40eb8028bb0f4fc37dd6f Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:20 +0900
Subject: wifi: brcmfmac: pcie: Add IDs/properties for BCM4387

This chip is present on Apple M1 Pro/Max (t600x) platforms:

* maldives   (apple,j314s): MacBook Pro (14-inch, M1 Pro, 2021)
* maldives   (apple,j314c): MacBook Pro (14-inch, M1 Max, 2021)
* madagascar (apple,j316s): MacBook Pro (16-inch, M1 Pro, 2021)
* madagascar (apple,j316c): MacBook Pro (16-inch, M1 Max, 2021)

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-7-marcan@marcan.st
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c       | 2 ++
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c       | 8 ++++++++
 drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h | 2 ++
 3 files changed, 12 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index 50e0c015cf43..9f9bf08a70bb 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -743,6 +743,8 @@ static u32 brcmf_chip_tcm_rambase(struct brcmf_chip_priv *ci)
 		return 0x170000;
 	case BRCM_CC_4378_CHIP_ID:
 		return 0x352000;
+	case BRCM_CC_4387_CHIP_ID:
+		return 0x740000;
 	default:
 		brcmf_err("unknown chip: %s\n", ci->pub.name);
 		break;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 2835ef4edb18..d2dad5414f39 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -67,6 +67,7 @@ BRCMF_FW_DEF(4366C, "brcmfmac4366c-pcie");
 BRCMF_FW_DEF(4371, "brcmfmac4371-pcie");
 BRCMF_FW_CLM_DEF(4377B3, "brcmfmac4377b3-pcie");
 BRCMF_FW_CLM_DEF(4378B1, "brcmfmac4378b1-pcie");
+BRCMF_FW_CLM_DEF(4387C2, "brcmfmac4387c2-pcie");
 
 /* firmware config files */
 MODULE_FIRMWARE(BRCMF_FW_DEFAULT_PATH "brcmfmac*-pcie.txt");
@@ -101,6 +102,7 @@ static const struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 	BRCMF_FW_ENTRY(BRCM_CC_4371_CHIP_ID, 0xFFFFFFFF, 4371),
 	BRCMF_FW_ENTRY(BRCM_CC_4377_CHIP_ID, 0xFFFFFFFF, 4377B3), /* revision ID 4 */
 	BRCMF_FW_ENTRY(BRCM_CC_4378_CHIP_ID, 0xFFFFFFFF, 4378B1), /* revision ID 3 */
+	BRCMF_FW_ENTRY(BRCM_CC_4387_CHIP_ID, 0xFFFFFFFF, 4387C2), /* revision ID 7 */
 };
 
 #define BRCMF_PCIE_FW_UP_TIMEOUT		5000 /* msec */
@@ -2048,6 +2050,11 @@ static int brcmf_pcie_read_otp(struct brcmf_pciedev_info *devinfo)
 		base = 0x1120;
 		words = 0x170;
 		break;
+	case BRCM_CC_4387_CHIP_ID:
+		coreid = BCMA_CORE_GCI;
+		base = 0x113c;
+		words = 0x170;
+		break;
 	default:
 		/* OTP not supported on this chip */
 		return 0;
@@ -2662,6 +2669,7 @@ static const struct pci_device_id brcmf_pcie_devid_table[] = {
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_43596_DEVICE_ID, CYW),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4377_DEVICE_ID, WCC),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4378_DEVICE_ID, WCC),
+	BRCMF_PCIE_DEVICE(BRCM_PCIE_4387_DEVICE_ID, WCC),
 
 	{ /* end: all zeroes */ }
 };
diff --git a/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h b/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h
index 896615f57952..44684bf1b9ac 100644
--- a/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h
+++ b/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h
@@ -54,6 +54,7 @@
 #define BRCM_CC_4371_CHIP_ID		0x4371
 #define BRCM_CC_4377_CHIP_ID		0x4377
 #define BRCM_CC_4378_CHIP_ID		0x4378
+#define BRCM_CC_4387_CHIP_ID		0x4387
 #define CY_CC_4373_CHIP_ID		0x4373
 #define CY_CC_43012_CHIP_ID		43012
 #define CY_CC_43439_CHIP_ID		43439
@@ -95,6 +96,7 @@
 #define BRCM_PCIE_43596_DEVICE_ID	0x4415
 #define BRCM_PCIE_4377_DEVICE_ID	0x4488
 #define BRCM_PCIE_4378_DEVICE_ID	0x4425
+#define BRCM_PCIE_4387_DEVICE_ID	0x4433
 
 /* brcmsmac IDs */
 #define BCM4313_D11N2G_ID	0x4727	/* 4313 802.11n 2.4G device */
-- 
cgit v1.2.3-70-g09d2


From dd7e55401fec58a2d03e5fdcb0c6d20e8fbe450a Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:21 +0900
Subject: wifi: brcmfmac: common: Add support for downloading TxCap blobs

The TxCap blobs are additional data blobs used on Apple devices, and
are uploaded analogously to CLM blobs. Add core support for doing this.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-8-marcan@marcan.st
---
 .../net/wireless/broadcom/brcm80211/brcmfmac/bus.h |  1 +
 .../wireless/broadcom/brcm80211/brcmfmac/common.c  | 93 ++++++++++++++++------
 2 files changed, 70 insertions(+), 24 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h
index 501136e011b5..fe31051a9e11 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h
@@ -55,6 +55,7 @@ enum brcmf_bus_protocol_type {
 /* Firmware blobs that may be available */
 enum brcmf_blob_type {
 	BRCMF_BLOB_CLM,
+	BRCMF_BLOB_TXCAP,
 };
 
 struct brcmf_mp_device;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index 4fea6f0ea8c8..88e81b83af9d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -101,7 +101,7 @@ void brcmf_c_set_joinpref_default(struct brcmf_if *ifp)
 
 static int brcmf_c_download(struct brcmf_if *ifp, u16 flag,
 			    struct brcmf_dload_data_le *dload_buf,
-			    u32 len)
+			    u32 len, const char *var)
 {
 	s32 err;
 
@@ -111,18 +111,18 @@ static int brcmf_c_download(struct brcmf_if *ifp, u16 flag,
 	dload_buf->len = cpu_to_le32(len);
 	dload_buf->crc = cpu_to_le32(0);
 
-	err = brcmf_fil_iovar_data_set(ifp, "clmload", dload_buf,
+	err = brcmf_fil_iovar_data_set(ifp, var, dload_buf,
 				       struct_size(dload_buf, data, len));
 
 	return err;
 }
 
-static int brcmf_c_process_clm_blob(struct brcmf_if *ifp)
+static int brcmf_c_download_blob(struct brcmf_if *ifp,
+				 const void *data, size_t size,
+				 const char *loadvar, const char *statvar)
 {
 	struct brcmf_pub *drvr = ifp->drvr;
-	struct brcmf_bus *bus = drvr->bus_if;
 	struct brcmf_dload_data_le *chunk_buf;
-	const struct firmware *clm = NULL;
 	u32 chunk_len;
 	u32 datalen;
 	u32 cumulative_len;
@@ -132,21 +132,14 @@ static int brcmf_c_process_clm_blob(struct brcmf_if *ifp)
 
 	brcmf_dbg(TRACE, "Enter\n");
 
-	err = brcmf_bus_get_blob(bus, &clm, BRCMF_BLOB_CLM);
-	if (err || !clm) {
-		brcmf_info("no clm_blob available (err=%d), device may have limited channels available\n",
-			   err);
-		return 0;
-	}
-
 	chunk_buf = kzalloc(struct_size(chunk_buf, data, MAX_CHUNK_LEN),
 			    GFP_KERNEL);
 	if (!chunk_buf) {
 		err = -ENOMEM;
-		goto done;
+		return -ENOMEM;
 	}
 
-	datalen = clm->size;
+	datalen = size;
 	cumulative_len = 0;
 	do {
 		if (datalen > MAX_CHUNK_LEN) {
@@ -155,9 +148,10 @@ static int brcmf_c_process_clm_blob(struct brcmf_if *ifp)
 			chunk_len = datalen;
 			dl_flag |= DL_END;
 		}
-		memcpy(chunk_buf->data, clm->data + cumulative_len, chunk_len);
+		memcpy(chunk_buf->data, data + cumulative_len, chunk_len);
 
-		err = brcmf_c_download(ifp, dl_flag, chunk_buf, chunk_len);
+		err = brcmf_c_download(ifp, dl_flag, chunk_buf, chunk_len,
+				       loadvar);
 
 		dl_flag &= ~DL_BEGIN;
 
@@ -166,20 +160,64 @@ static int brcmf_c_process_clm_blob(struct brcmf_if *ifp)
 	} while ((datalen > 0) && (err == 0));
 
 	if (err) {
-		bphy_err(drvr, "clmload (%zu byte file) failed (%d)\n",
-			 clm->size, err);
-		/* Retrieve clmload_status and print */
-		err = brcmf_fil_iovar_int_get(ifp, "clmload_status", &status);
+		bphy_err(drvr, "%s (%zu byte file) failed (%d)\n",
+			 loadvar, size, err);
+		/* Retrieve status and print */
+		err = brcmf_fil_iovar_int_get(ifp, statvar, &status);
 		if (err)
-			bphy_err(drvr, "get clmload_status failed (%d)\n", err);
+			bphy_err(drvr, "get %s failed (%d)\n", statvar, err);
 		else
-			brcmf_dbg(INFO, "clmload_status=%d\n", status);
+			brcmf_dbg(INFO, "%s=%d\n", statvar, status);
 		err = -EIO;
 	}
 
 	kfree(chunk_buf);
-done:
-	release_firmware(clm);
+	return err;
+}
+
+static int brcmf_c_process_clm_blob(struct brcmf_if *ifp)
+{
+	struct brcmf_pub *drvr = ifp->drvr;
+	struct brcmf_bus *bus = drvr->bus_if;
+	const struct firmware *fw = NULL;
+	s32 err;
+
+	brcmf_dbg(TRACE, "Enter\n");
+
+	err = brcmf_bus_get_blob(bus, &fw, BRCMF_BLOB_CLM);
+	if (err || !fw) {
+		brcmf_info("no clm_blob available (err=%d), device may have limited channels available\n",
+			   err);
+		return 0;
+	}
+
+	err = brcmf_c_download_blob(ifp, fw->data, fw->size,
+				    "clmload", "clmload_status");
+
+	release_firmware(fw);
+	return err;
+}
+
+static int brcmf_c_process_txcap_blob(struct brcmf_if *ifp)
+{
+	struct brcmf_pub *drvr = ifp->drvr;
+	struct brcmf_bus *bus = drvr->bus_if;
+	const struct firmware *fw = NULL;
+	s32 err;
+
+	brcmf_dbg(TRACE, "Enter\n");
+
+	err = brcmf_bus_get_blob(bus, &fw, BRCMF_BLOB_TXCAP);
+	if (err || !fw) {
+		brcmf_info("no txcap_blob available (err=%d)\n", err);
+		return 0;
+	}
+
+	brcmf_info("TxCap blob found, loading\n");
+	err = brcmf_c_download_blob(ifp, fw->data, fw->size,
+				    "txcapload", "txcapload_status");
+
+	release_firmware(fw);
 	return err;
 }
 
@@ -291,6 +329,13 @@ int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
 		goto done;
 	}
 
+	/* Do TxCap downloading, if needed */
+	err = brcmf_c_process_txcap_blob(ifp);
+	if (err < 0) {
+		bphy_err(drvr, "download TxCap blob file failed, %d\n", err);
+		goto done;
+	}
+
 	/* query for 'ver' to get version info from firmware */
 	memset(buf, 0, sizeof(buf));
 	err = brcmf_fil_iovar_data_get(ifp, "ver", buf, sizeof(buf));
-- 
cgit v1.2.3-70-g09d2


From 75102b7543ed87cf7d599a382e7340e572da6987 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:22 +0900
Subject: wifi: brcmfmac: pcie: Load and provide TxCap blobs

These blobs are named .txcap_blob, and exist alongside the existing
.clm_blob files. Use the existing firmware machinery to provide them to
the core.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-9-marcan@marcan.st
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index d2dad5414f39..0fcdd84bbed0 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -76,6 +76,7 @@ MODULE_FIRMWARE(BRCMF_FW_DEFAULT_PATH "brcmfmac*-pcie.*.txt");
 /* per-board firmware binaries */
 MODULE_FIRMWARE(BRCMF_FW_DEFAULT_PATH "brcmfmac*-pcie.*.bin");
 MODULE_FIRMWARE(BRCMF_FW_DEFAULT_PATH "brcmfmac*-pcie.*.clm_blob");
+MODULE_FIRMWARE(BRCMF_FW_DEFAULT_PATH "brcmfmac*-pcie.*.txcap_blob");
 
 static const struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 	BRCMF_FW_ENTRY(BRCM_CC_43602_CHIP_ID, 0xFFFFFFFF, 43602),
@@ -329,7 +330,9 @@ struct brcmf_pciedev_info {
 	char fw_name[BRCMF_FW_NAME_LEN];
 	char nvram_name[BRCMF_FW_NAME_LEN];
 	char clm_name[BRCMF_FW_NAME_LEN];
+	char txcap_name[BRCMF_FW_NAME_LEN];
 	const struct firmware *clm_fw;
+	const struct firmware *txcap_fw;
 	const struct brcmf_pcie_reginfo *reginfo;
 	void __iomem *regs;
 	void __iomem *tcm;
@@ -1520,6 +1523,10 @@ static int brcmf_pcie_get_blob(struct device *dev, const struct firmware **fw,
 		*fw = devinfo->clm_fw;
 		devinfo->clm_fw = NULL;
 		break;
+	case BRCMF_BLOB_TXCAP:
+		*fw = devinfo->txcap_fw;
+		devinfo->txcap_fw = NULL;
+		break;
 	default:
 		return -ENOENT;
 	}
@@ -2112,6 +2119,7 @@ static int brcmf_pcie_read_otp(struct brcmf_pciedev_info *devinfo)
 #define BRCMF_PCIE_FW_CODE	0
 #define BRCMF_PCIE_FW_NVRAM	1
 #define BRCMF_PCIE_FW_CLM	2
+#define BRCMF_PCIE_FW_TXCAP	3
 
 static void brcmf_pcie_setup(struct device *dev, int ret,
 			     struct brcmf_fw_request *fwreq)
@@ -2138,6 +2146,7 @@ static void brcmf_pcie_setup(struct device *dev, int ret,
 	nvram = fwreq->items[BRCMF_PCIE_FW_NVRAM].nv_data.data;
 	nvram_len = fwreq->items[BRCMF_PCIE_FW_NVRAM].nv_data.len;
 	devinfo->clm_fw = fwreq->items[BRCMF_PCIE_FW_CLM].binary;
+	devinfo->txcap_fw = fwreq->items[BRCMF_PCIE_FW_TXCAP].binary;
 	kfree(fwreq);
 
 	ret = brcmf_chip_get_raminfo(devinfo->ci);
@@ -2219,6 +2228,7 @@ brcmf_pcie_prepare_fw_request(struct brcmf_pciedev_info *devinfo)
 		{ ".bin", devinfo->fw_name },
 		{ ".txt", devinfo->nvram_name },
 		{ ".clm_blob", devinfo->clm_name },
+		{ ".txcap_blob", devinfo->txcap_name },
 	};
 
 	fwreq = brcmf_fw_alloc_request(devinfo->ci->chip, devinfo->ci->chiprev,
@@ -2233,6 +2243,8 @@ brcmf_pcie_prepare_fw_request(struct brcmf_pciedev_info *devinfo)
 	fwreq->items[BRCMF_PCIE_FW_NVRAM].flags = BRCMF_FW_REQF_OPTIONAL;
 	fwreq->items[BRCMF_PCIE_FW_CLM].type = BRCMF_FW_TYPE_BINARY;
 	fwreq->items[BRCMF_PCIE_FW_CLM].flags = BRCMF_FW_REQF_OPTIONAL;
+	fwreq->items[BRCMF_PCIE_FW_TXCAP].type = BRCMF_FW_TYPE_BINARY;
+	fwreq->items[BRCMF_PCIE_FW_TXCAP].flags = BRCMF_FW_REQF_OPTIONAL;
 	/* NVRAM reserves PCI domain 0 for Broadcom's SDK faked bus */
 	fwreq->domain_nr = pci_domain_nr(devinfo->pdev->bus) + 1;
 	fwreq->bus_nr = devinfo->pdev->bus->number;
@@ -2530,6 +2542,7 @@ brcmf_pcie_remove(struct pci_dev *pdev)
 	brcmf_pcie_reset_device(devinfo);
 	brcmf_pcie_release_resource(devinfo);
 	release_firmware(devinfo->clm_fw);
+	release_firmware(devinfo->txcap_fw);
 
 	if (devinfo->ci)
 		brcmf_chip_detach(devinfo->ci);
-- 
cgit v1.2.3-70-g09d2


From 5b3ee9987f5856080509e62968f812961173d336 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:24:23 +0900
Subject: wifi: brcmfmac: common: Add support for external calibration blobs

The calibration blob for a chip is normally stored in SROM and loaded
internally by the firmware. However, Apple ARM64 platforms instead store
it as part of platform configuration data, and provide it via the Apple
Device Tree. We forward this into the Linux DT in the bootloader.

Add support for taking this blob from the DT and loading it into the
dongle. The loading mechanism is the same as used for the CLM and TxCap
blobs.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Hector Martin <marcan@marcan.st>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092423.15175-10-marcan@marcan.st
---
 .../wireless/broadcom/brcm80211/brcmfmac/common.c  | 24 ++++++++++++++++++++++
 .../wireless/broadcom/brcm80211/brcmfmac/common.h  |  2 ++
 .../net/wireless/broadcom/brcm80211/brcmfmac/of.c  |  7 +++++++
 3 files changed, 33 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index 88e81b83af9d..a194b0e68eb5 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -246,6 +246,23 @@ static const u8 brcmf_default_mac_address[ETH_ALEN] = {
 	0x00, 0x90, 0x4c, 0xc5, 0x12, 0x38
 };
 
+static int brcmf_c_process_cal_blob(struct brcmf_if *ifp)
+{
+	struct brcmf_pub *drvr = ifp->drvr;
+	struct brcmf_mp_device *settings = drvr->settings;
+	s32 err;
+
+	brcmf_dbg(TRACE, "Enter\n");
+
+	if (!settings->cal_blob || !settings->cal_size)
+		return 0;
+
+	brcmf_info("Calibration blob provided by platform, loading\n");
+	err = brcmf_c_download_blob(ifp, settings->cal_blob, settings->cal_size,
+				    "calload", "calload_status");
+	return err;
+}
+
 int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
 {
 	struct brcmf_pub *drvr = ifp->drvr;
@@ -336,6 +353,13 @@ int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
 		goto done;
 	}
 
+	/* Download external calibration blob, if available */
+	err = brcmf_c_process_cal_blob(ifp);
+	if (err < 0) {
+		bphy_err(drvr, "download calibration blob file failed, %d\n", err);
+		goto done;
+	}
+
 	/* query for 'ver' to get version info from firmware */
 	memset(buf, 0, sizeof(buf));
 	err = brcmf_fil_iovar_data_get(ifp, "ver", buf, sizeof(buf));
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index 7167fd4f8c63..2be2986d2110 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -54,6 +54,8 @@ struct brcmf_mp_device {
 	const char	*board_type;
 	unsigned char	mac[ETH_ALEN];
 	const char	*antenna_sku;
+	const void	*cal_blob;
+	int		cal_size;
 	union {
 		struct brcmfmac_sdio_pd sdio;
 	} bus;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index fdd0c9abc1a1..52527b61341e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -86,6 +86,13 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
 	if (!of_property_read_string(np, "apple,antenna-sku", &prop))
 		settings->antenna_sku = prop;
 
+	/* The WLAN calibration blob is normally stored in SROM, but Apple
+	 * ARM64 platforms pass it via the DT instead.
+	 */
+	prop = of_get_property(np, "brcm,cal-blob", &settings->cal_size);
+	if (prop && settings->cal_size)
+		settings->cal_blob = prop;
+
 	/* Set board-type to the first string of the machine compatible prop */
 	root = of_find_node_by_path("/");
 	if (root && err) {
-- 
cgit v1.2.3-70-g09d2


From 1d5003d05f983eee28756896328e4949d9a97b7f Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Tue, 14 Feb 2023 18:28:38 +0900
Subject: wifi: brcmfmac: pcie: Add BCM4378B3 support

BCM4378B3 is a new silicon revision of BCM4378 present on the Apple M2
13" MacBook Pro "kyushu". Its PCI revision number is 5.

Signed-off-by: Hector Martin <marcan@marcan.st>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230214092838.17869-1-marcan@marcan.st
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 0fcdd84bbed0..848f05ec1268 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -67,6 +67,7 @@ BRCMF_FW_DEF(4366C, "brcmfmac4366c-pcie");
 BRCMF_FW_DEF(4371, "brcmfmac4371-pcie");
 BRCMF_FW_CLM_DEF(4377B3, "brcmfmac4377b3-pcie");
 BRCMF_FW_CLM_DEF(4378B1, "brcmfmac4378b1-pcie");
+BRCMF_FW_CLM_DEF(4378B3, "brcmfmac4378b3-pcie");
 BRCMF_FW_CLM_DEF(4387C2, "brcmfmac4387c2-pcie");
 
 /* firmware config files */
@@ -102,7 +103,8 @@ static const struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 	BRCMF_FW_ENTRY(BRCM_CC_43666_CHIP_ID, 0xFFFFFFF0, 4366C),
 	BRCMF_FW_ENTRY(BRCM_CC_4371_CHIP_ID, 0xFFFFFFFF, 4371),
 	BRCMF_FW_ENTRY(BRCM_CC_4377_CHIP_ID, 0xFFFFFFFF, 4377B3), /* revision ID 4 */
-	BRCMF_FW_ENTRY(BRCM_CC_4378_CHIP_ID, 0xFFFFFFFF, 4378B1), /* revision ID 3 */
+	BRCMF_FW_ENTRY(BRCM_CC_4378_CHIP_ID, 0x0000000F, 4378B1), /* revision ID 3 */
+	BRCMF_FW_ENTRY(BRCM_CC_4378_CHIP_ID, 0xFFFFFFE0, 4378B3), /* revision ID 5 */
 	BRCMF_FW_ENTRY(BRCM_CC_4387_CHIP_ID, 0xFFFFFFFF, 4387C2), /* revision ID 7 */
 };
 
-- 
cgit v1.2.3-70-g09d2


From 29c66ad1c3ad1698becf81182f73fb4bd50bceb2 Mon Sep 17 00:00:00 2001
From: Tiezhu Yang <yangtiezhu@loongson.cn>
Date: Fri, 24 Feb 2023 18:36:53 +0800
Subject: libbpf: Use struct user_pt_regs to define __PT_REGS_CAST() for
 LoongArch

LoongArch provides struct user_pt_regs instead of struct pt_regs
to userspace, use struct user_pt_regs to define __PT_REGS_CAST()
to fix the following build error:

     CLNG-BPF [test_maps] loop1.bpf.o
  progs/loop1.c:22:9: error: incomplete definition of type 'struct pt_regs'
                                  m = PT_REGS_RC(ctx);
                                      ^~~~~~~~~~~~~~~
  tools/testing/selftests/bpf/tools/include/bpf/bpf_tracing.h:493:41: note: expanded from macro 'PT_REGS_RC'
  #define PT_REGS_RC(x) (__PT_REGS_CAST(x)->__PT_RC_REG)
                         ~~~~~~~~~~~~~~~~~^
  tools/testing/selftests/bpf/tools/include/bpf/bpf_helper_defs.h:20:8: note: forward declaration of 'struct pt_regs'
  struct pt_regs;
         ^
  1 error generated.
  make: *** [Makefile:572: tools/testing/selftests/bpf/loop1.bpf.o] Error 1
  make: Leaving directory 'tools/testing/selftests/bpf'

Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/1677235015-21717-2-git-send-email-yangtiezhu@loongson.cn
---
 tools/lib/bpf/bpf_tracing.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h
index 6db88f41fa0d..137b13de0807 100644
--- a/tools/lib/bpf/bpf_tracing.h
+++ b/tools/lib/bpf/bpf_tracing.h
@@ -415,6 +415,8 @@ struct pt_regs___arm64 {
  * https://loongson.github.io/LoongArch-Documentation/LoongArch-ELF-ABI-EN.html
  */
 
+/* loongarch provides struct user_pt_regs instead of struct pt_regs to userspace */
+#define __PT_REGS_CAST(x) ((const struct user_pt_regs *)(x))
 #define __PT_PARM1_REG regs[4]
 #define __PT_PARM2_REG regs[5]
 #define __PT_PARM3_REG regs[6]
-- 
cgit v1.2.3-70-g09d2


From 84c22fa83f9cb97061359f97d85f53a9ccde02c4 Mon Sep 17 00:00:00 2001
From: Tiezhu Yang <yangtiezhu@loongson.cn>
Date: Fri, 24 Feb 2023 18:36:55 +0800
Subject: selftests/bpf: Use __NR_prlimit64 instead of __NR_getrlimit in
 user_ringbuf test

After commit 80d7da1cac62 ("asm-generic: Drop getrlimit and setrlimit
syscalls from default list"), new architectures won't need to include
getrlimit and setrlimit, they are superseded with prlimit64.

In order to maintain compatibility for the new architectures, such as
LoongArch which does not define __NR_getrlimit, it is better to use
__NR_prlimit64 instead of __NR_getrlimit in user_ringbuf test to fix
the following build error:

    TEST-OBJ [test_progs] user_ringbuf.test.o
  tools/testing/selftests/bpf/prog_tests/user_ringbuf.c: In function 'kick_kernel_cb':
  tools/testing/selftests/bpf/prog_tests/user_ringbuf.c:593:17: error: '__NR_getrlimit' undeclared (first use in this function)
    593 |         syscall(__NR_getrlimit);
        |                 ^~~~~~~~~~~~~~
  tools/testing/selftests/bpf/prog_tests/user_ringbuf.c:593:17: note: each undeclared identifier is reported only once for each function it appears in
  make: *** [Makefile:573: tools/testing/selftests/bpf/user_ringbuf.test.o] Error 1
  make: Leaving directory 'tools/testing/selftests/bpf'

Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/1677235015-21717-4-git-send-email-yangtiezhu@loongson.cn
---
 tools/testing/selftests/bpf/prog_tests/user_ringbuf.c    | 2 +-
 tools/testing/selftests/bpf/progs/user_ringbuf_success.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/user_ringbuf.c b/tools/testing/selftests/bpf/prog_tests/user_ringbuf.c
index 3a13e102c149..e51721df14fc 100644
--- a/tools/testing/selftests/bpf/prog_tests/user_ringbuf.c
+++ b/tools/testing/selftests/bpf/prog_tests/user_ringbuf.c
@@ -590,7 +590,7 @@ static void *kick_kernel_cb(void *arg)
 	/* Kick the kernel, causing it to drain the ring buffer and then wake
 	 * up the test thread waiting on epoll.
 	 */
-	syscall(__NR_getrlimit);
+	syscall(__NR_prlimit64);
 
 	return NULL;
 }
diff --git a/tools/testing/selftests/bpf/progs/user_ringbuf_success.c b/tools/testing/selftests/bpf/progs/user_ringbuf_success.c
index b39093dd5715..0ade1110613b 100644
--- a/tools/testing/selftests/bpf/progs/user_ringbuf_success.c
+++ b/tools/testing/selftests/bpf/progs/user_ringbuf_success.c
@@ -202,7 +202,7 @@ do_nothing_cb(struct bpf_dynptr *dynptr, void *context)
 	return 0;
 }
 
-SEC("fentry/" SYS_PREFIX "sys_getrlimit")
+SEC("fentry/" SYS_PREFIX "sys_prlimit64")
 int test_user_ringbuf_epoll(void *ctx)
 {
 	long num_samples;
-- 
cgit v1.2.3-70-g09d2


From c679bbd611c08b0559ffae079330bc4e5574696a Mon Sep 17 00:00:00 2001
From: Luis Gerhorst <gerhorst@cs.fau.de>
Date: Mon, 27 Feb 2023 16:08:54 +0100
Subject: tools: bpftool: Remove invalid \' json escape

RFC8259 ("The JavaScript Object Notation (JSON) Data Interchange
Format") only specifies \", \\, \/, \b, \f, \n, \r, and \r as valid
two-character escape sequences. This does not include \', which is not
required in JSON because it exclusively uses double quotes as string
separators.

Solidus (/) may be escaped, but does not have to. Only reverse
solidus (\), double quotes ("), and the control characters have to be
escaped. Therefore, with this fix, bpftool correctly supports all valid
two-character escape sequences (but still does not support characters
that require multi-character escape sequences).

Witout this fix, attempting to load a JSON file generated by bpftool
using Python 3.10.6's default json.load() may fail with the error
"Invalid \escape" if the file contains the invalid escaped single
quote (\').

Fixes: b66e907cfee2 ("tools: bpftool: copy JSON writer from iproute2 repository")
Signed-off-by: Luis Gerhorst <gerhorst@cs.fau.de>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/bpf/20230227150853.16863-1-gerhorst@cs.fau.de
---
 tools/bpf/bpftool/json_writer.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/tools/bpf/bpftool/json_writer.c b/tools/bpf/bpftool/json_writer.c
index 7fea83bedf48..bca5dd0a59e3 100644
--- a/tools/bpf/bpftool/json_writer.c
+++ b/tools/bpf/bpftool/json_writer.c
@@ -80,9 +80,6 @@ static void jsonw_puts(json_writer_t *self, const char *str)
 		case '"':
 			fputs("\\\"", self->out);
 			break;
-		case '\'':
-			fputs("\\\'", self->out);
-			break;
 		default:
 			putc(*str, self->out);
 		}
-- 
cgit v1.2.3-70-g09d2


From 11e456cae91e9044cb12c2b037b52c9b268925f7 Mon Sep 17 00:00:00 2001
From: Rong Tao <rongtao@cestc.cn>
Date: Fri, 24 Feb 2023 23:10:02 +0800
Subject: selftests/bpf: Fix compilation errors: Assign a value to a constant

Commit bc292ab00f6c("mm: introduce vma->vm_flags wrapper functions")
turns the vm_flags into a const variable.

Added bpf_find_vma test in commit f108662b27c9("selftests/bpf: Add tests
for bpf_find_vma") to assign values to variables that declare const in
find_vma_fail1.c programs, which is an error to the compiler and does not
test BPF verifiers. It is better to replace 'const vm_flags_t vm_flags'
with 'unsigned long vm_start' for testing.

    $ make -C tools/testing/selftests/bpf/ -j8
    ...
    progs/find_vma_fail1.c:16:16: error: cannot assign to non-static data
    member 'vm_flags' with const-qualified type 'const vm_flags_t' (aka
    'const unsigned long')
            vma->vm_flags |= 0x55;
            ~~~~~~~~~~~~~ ^
    ../tools/testing/selftests/bpf/tools/include/vmlinux.h:1898:20:
    note: non-static data member 'vm_flags' declared const here
                    const vm_flags_t vm_flags;
                    ~~~~~~~~~~~`~~~~~~^~~~~~~~

Signed-off-by: Rong Tao <rongtao@cestc.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/tencent_CB281722B3C1BD504C16CDE586CACC2BE706@qq.com
---
 tools/testing/selftests/bpf/progs/find_vma_fail1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/find_vma_fail1.c b/tools/testing/selftests/bpf/progs/find_vma_fail1.c
index b3b326b8e2d1..47d5dedff554 100644
--- a/tools/testing/selftests/bpf/progs/find_vma_fail1.c
+++ b/tools/testing/selftests/bpf/progs/find_vma_fail1.c
@@ -13,7 +13,7 @@ static long write_vma(struct task_struct *task, struct vm_area_struct *vma,
 		      struct callback_ctx *data)
 {
 	/* writing to vma, which is illegal */
-	vma->vm_flags |= 0x55;
+	vma->vm_start = 0xffffffffff600000;
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 06943ae675945c762bb8d5edc93d203f2b041d8d Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay12@gmail.com>
Date: Thu, 23 Feb 2023 09:53:46 +0000
Subject: libbpf: Fix arm syscall regs spec in bpf_tracing.h

The syscall register definitions for ARM in bpf_tracing.h doesn't define
the fifth parameter for the syscalls. Because of this some KPROBES based
selftests fail to compile for ARM architecture.

Define the fifth parameter that is passed in the R5 register (uregs[4]).

Fixes: 3a95c42d65d5 ("libbpf: Define arm syscall regs spec in bpf_tracing.h")
Signed-off-by: Puranjay Mohan <puranjay12@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230223095346.10129-1-puranjay12@gmail.com
---
 tools/lib/bpf/bpf_tracing.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h
index 137b13de0807..6fb3d0f9af17 100644
--- a/tools/lib/bpf/bpf_tracing.h
+++ b/tools/lib/bpf/bpf_tracing.h
@@ -204,6 +204,7 @@ struct pt_regs___s390 {
 #define __PT_PARM2_SYSCALL_REG __PT_PARM2_REG
 #define __PT_PARM3_SYSCALL_REG __PT_PARM3_REG
 #define __PT_PARM4_SYSCALL_REG __PT_PARM4_REG
+#define __PT_PARM5_SYSCALL_REG uregs[4]
 #define __PT_PARM6_SYSCALL_REG uregs[5]
 #define __PT_PARM7_SYSCALL_REG uregs[6]
 
-- 
cgit v1.2.3-70-g09d2


From 0a504fa1a780332760f2a4369a1adc3ee9f620d0 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 21 Feb 2023 00:49:58 +0100
Subject: libbpf: Document bpf_{btf,link,map,prog}_get_info_by_fd()

Replace the short informal description with the proper doc comments.

Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230220234958.764997-1-iii@linux.ibm.com
---
 tools/lib/bpf/bpf.h | 69 +++++++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 64 insertions(+), 5 deletions(-)

diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 9ed9bceb4111..f0f786373238 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
 
 /*
- * common eBPF ELF operations.
+ * Common BPF ELF operations.
  *
  * Copyright (C) 2013-2015 Alexei Starovoitov <ast@kernel.org>
  * Copyright (C) 2015 Wang Nan <wangnan0@huawei.com>
@@ -386,14 +386,73 @@ LIBBPF_API int bpf_link_get_fd_by_id(__u32 id);
 LIBBPF_API int bpf_link_get_fd_by_id_opts(__u32 id,
 				const struct bpf_get_fd_by_id_opts *opts);
 LIBBPF_API int bpf_obj_get_info_by_fd(int bpf_fd, void *info, __u32 *info_len);
-/* Type-safe variants of bpf_obj_get_info_by_fd(). The callers still needs to
- * pass info_len, which should normally be
- * sizeof(struct bpf_{prog,map,btf,link}_info), in order to be compatible with
- * different libbpf and kernel versions.
+
+/**
+ * @brief **bpf_prog_get_info_by_fd()** obtains information about the BPF
+ * program corresponding to *prog_fd*.
+ *
+ * Populates up to *info_len* bytes of *info* and updates *info_len* with the
+ * actual number of bytes written to *info*.
+ *
+ * @param prog_fd BPF program file descriptor
+ * @param info pointer to **struct bpf_prog_info** that will be populated with
+ * BPF program information
+ * @param info_len pointer to the size of *info*; on success updated with the
+ * number of bytes written to *info*
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
  */
 LIBBPF_API int bpf_prog_get_info_by_fd(int prog_fd, struct bpf_prog_info *info, __u32 *info_len);
+
+/**
+ * @brief **bpf_map_get_info_by_fd()** obtains information about the BPF
+ * map corresponding to *map_fd*.
+ *
+ * Populates up to *info_len* bytes of *info* and updates *info_len* with the
+ * actual number of bytes written to *info*.
+ *
+ * @param map_fd BPF map file descriptor
+ * @param info pointer to **struct bpf_map_info** that will be populated with
+ * BPF map information
+ * @param info_len pointer to the size of *info*; on success updated with the
+ * number of bytes written to *info*
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
+ */
 LIBBPF_API int bpf_map_get_info_by_fd(int map_fd, struct bpf_map_info *info, __u32 *info_len);
+
+/**
+ * @brief **bpf_btf_get_info_by_fd()** obtains information about the 
+ * BTF object corresponding to *btf_fd*.
+ *
+ * Populates up to *info_len* bytes of *info* and updates *info_len* with the
+ * actual number of bytes written to *info*.
+ *
+ * @param btf_fd BTF object file descriptor
+ * @param info pointer to **struct bpf_btf_info** that will be populated with
+ * BTF object information
+ * @param info_len pointer to the size of *info*; on success updated with the
+ * number of bytes written to *info*
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
+ */
 LIBBPF_API int bpf_btf_get_info_by_fd(int btf_fd, struct bpf_btf_info *info, __u32 *info_len);
+
+/**
+ * @brief **bpf_btf_get_info_by_fd()** obtains information about the BPF
+ * link corresponding to *link_fd*.
+ *
+ * Populates up to *info_len* bytes of *info* and updates *info_len* with the
+ * actual number of bytes written to *info*.
+ *
+ * @param link_fd BPF link file descriptor
+ * @param info pointer to **struct bpf_link_info** that will be populated with
+ * BPF link information
+ * @param info_len pointer to the size of *info*; on success updated with the
+ * number of bytes written to *info*
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
+ */
 LIBBPF_API int bpf_link_get_info_by_fd(int link_fd, struct bpf_link_info *info, __u32 *info_len);
 
 struct bpf_prog_query_opts {
-- 
cgit v1.2.3-70-g09d2


From c8ee37bde4021a275d2e4f33bd48d54912bb00c4 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Mon, 27 Feb 2023 14:49:43 -0800
Subject: libbpf: Fix bpf_xdp_query() in old kernels

Commit 04d58f1b26a4("libbpf: add API to get XDP/XSK supported features")
added feature_flags to struct bpf_xdp_query_opts. If a user uses
bpf_xdp_query_opts with feature_flags member, the bpf_xdp_query()
will check whether 'netdev' family exists or not in the kernel.
If it does not exist, the bpf_xdp_query() will return -ENOENT.

But 'netdev' family does not exist in old kernels as it is
introduced in the same patch set as Commit 04d58f1b26a4.
So old kernel with newer libbpf won't work properly with
bpf_xdp_query() api call.

To fix this issue, if the return value of
libbpf_netlink_resolve_genl_family_id() is -ENOENT, bpf_xdp_query()
will just return 0, skipping the rest of xdp feature query.
This preserves backward compatibility.

Fixes: 04d58f1b26a4 ("libbpf: add API to get XDP/XSK supported features")
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230227224943.1153459-1-yhs@fb.com
---
 tools/lib/bpf/netlink.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c
index 1653e7a8b0a1..84dd5fa14905 100644
--- a/tools/lib/bpf/netlink.c
+++ b/tools/lib/bpf/netlink.c
@@ -468,8 +468,13 @@ int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
 		return 0;
 
 	err = libbpf_netlink_resolve_genl_family_id("netdev", sizeof("netdev"), &id);
-	if (err < 0)
+	if (err < 0) {
+		if (err == -ENOENT) {
+			opts->feature_flags = 0;
+			goto skip_feature_flags;
+		}
 		return libbpf_err(err);
+	}
 
 	memset(&req, 0, sizeof(req));
 	req.nh.nlmsg_len = NLMSG_LENGTH(GENL_HDRLEN);
@@ -489,6 +494,7 @@ int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
 
 	opts->feature_flags = md.flags;
 
+skip_feature_flags:
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From bbefef2f07080cd502a93cb1c529e1c8a6c4ac8e Mon Sep 17 00:00:00 2001
From: Jiaxun Yang <jiaxun.yang@flygoat.com>
Date: Tue, 28 Feb 2023 11:33:04 +0000
Subject: bpf, mips: Implement DADDI workarounds for JIT
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For DADDI errata we just workaround by disable immediate operation
for BPF_ADD / BPF_SUB to avoid generation of DADDIU.

All other use cases in JIT won't cause overflow thus they are all safe.

Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Johan Almbladh <johan.almbladh@anyfinetworks.com>
Link: https://lore.kernel.org/bpf/20230228113305.83751-2-jiaxun.yang@flygoat.com
---
 arch/mips/Kconfig            | 1 -
 arch/mips/net/bpf_jit_comp.c | 4 ++++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 15cb692b0a09..b89c4bf2fa0a 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -64,7 +64,6 @@ config MIPS
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_EBPF_JIT if !CPU_MICROMIPS && \
-				!CPU_DADDI_WORKAROUNDS && \
 				!CPU_R4000_WORKAROUNDS && \
 				!CPU_R4400_WORKAROUNDS
 	select HAVE_EXIT_THREAD
diff --git a/arch/mips/net/bpf_jit_comp.c b/arch/mips/net/bpf_jit_comp.c
index b17130d510d4..a40d926b6513 100644
--- a/arch/mips/net/bpf_jit_comp.c
+++ b/arch/mips/net/bpf_jit_comp.c
@@ -218,9 +218,13 @@ bool valid_alu_i(u8 op, s32 imm)
 		/* All legal eBPF values are valid */
 		return true;
 	case BPF_ADD:
+		if (IS_ENABLED(CONFIG_CPU_DADDI_WORKAROUNDS))
+			return false;
 		/* imm must be 16 bits */
 		return imm >= -0x8000 && imm <= 0x7fff;
 	case BPF_SUB:
+		if (IS_ENABLED(CONFIG_CPU_DADDI_WORKAROUNDS))
+			return false;
 		/* -imm must be 16 bits */
 		return imm >= -0x7fff && imm <= 0x8000;
 	case BPF_AND:
-- 
cgit v1.2.3-70-g09d2


From 7364d60c26618d7465602ac86717a5a325b342f3 Mon Sep 17 00:00:00 2001
From: Jiaxun Yang <jiaxun.yang@flygoat.com>
Date: Tue, 28 Feb 2023 11:33:05 +0000
Subject: bpf, mips: Implement R4000 workarounds for JIT
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For R4000 erratas around multiplication and division instructions,
as our use of those instructions are always followed by mflo/mfhi
instructions, the only issue we need care is

"MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0" Errata 28:
"A double-word or a variable shift may give an incorrect result if
executed while an integer multiplication is in progress."

We just emit a mfhi $0 to ensure the operation is completed after
every multiplication instruction according to workaround suggestion
in the document.

Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Johan Almbladh <johan.almbladh@anyfinetworks.com>
Link: https://lore.kernel.org/bpf/20230228113305.83751-3-jiaxun.yang@flygoat.com
---
 arch/mips/Kconfig              | 4 +---
 arch/mips/net/bpf_jit_comp64.c | 3 +++
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index b89c4bf2fa0a..363cd59a3bc0 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -63,9 +63,7 @@ config MIPS
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DYNAMIC_FTRACE
-	select HAVE_EBPF_JIT if !CPU_MICROMIPS && \
-				!CPU_R4000_WORKAROUNDS && \
-				!CPU_R4400_WORKAROUNDS
+	select HAVE_EBPF_JIT if !CPU_MICROMIPS
 	select HAVE_EXIT_THREAD
 	select HAVE_FAST_GUP
 	select HAVE_FTRACE_MCOUNT_RECORD
diff --git a/arch/mips/net/bpf_jit_comp64.c b/arch/mips/net/bpf_jit_comp64.c
index 0e7c1bdcf914..fa7e9aa37f49 100644
--- a/arch/mips/net/bpf_jit_comp64.c
+++ b/arch/mips/net/bpf_jit_comp64.c
@@ -228,6 +228,9 @@ static void emit_alu_r64(struct jit_context *ctx, u8 dst, u8 src, u8 op)
 		} else {
 			emit(ctx, dmultu, dst, src);
 			emit(ctx, mflo, dst);
+			/* Ensure multiplication is completed */
+			if (IS_ENABLED(CONFIG_CPU_R4000_WORKAROUNDS))
+				emit(ctx, mfhi, MIPS_R_ZERO);
 		}
 		break;
 	/* dst = dst / src */
-- 
cgit v1.2.3-70-g09d2


From 30a2d8328d8ac1bb0a6bf73f4f4cf03f4f5977cc Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Tue, 28 Feb 2023 09:28:45 -0600
Subject: bpf: Fix bpf_cgroup_from_id() doxygen header

In commit 332ea1f697be ("bpf: Add bpf_cgroup_from_id() kfunc"), a new
bpf_cgroup_from_id() kfunc was added which allows a BPF program to
lookup and acquire a reference to a cgroup from a cgroup id. The
commit's doxygen comment seems to have copy-pasted fields, which causes
BPF kfunc helper documentation to fail to render:

<snip>/helpers.c:2114: warning: Excess function parameter 'cgrp'...
<snip>/helpers.c:2114: warning: Excess function parameter 'level'...

<snip>

<snip>/helpers.c:2114: warning: Excess function parameter 'level'...

This patch fixes the doxygen header.

Fixes: 332ea1f697be ("bpf: Add bpf_cgroup_from_id() kfunc")
Signed-off-by: David Vernet <void@manifault.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20230228152845.294695-1-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index a784be6f8bac..abdcc52f90a6 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2106,8 +2106,7 @@ __bpf_kfunc struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level)
  * bpf_cgroup_from_id - Find a cgroup from its ID. A cgroup returned by this
  * kfunc which is not subsequently stored in a map, must be released by calling
  * bpf_cgroup_release().
- * @cgrp: The cgroup for which we're performing a lookup.
- * @level: The level of ancestor to look up.
+ * @cgid: cgroup id.
  */
 __bpf_kfunc struct cgroup *bpf_cgroup_from_id(u64 cgid)
 {
-- 
cgit v1.2.3-70-g09d2


From ae256f95478e07d49dae5036bb83c09dfbd686d4 Mon Sep 17 00:00:00 2001
From: "Jose E. Marchesi" <jose.marchesi@oracle.com>
Date: Tue, 28 Feb 2023 10:51:29 +0100
Subject: bpf, docs: Document BPF insn encoding in term of stored bytes

[Changes from V4:
- s/regs:16/regs:8 in figure.]

[Changes from V3:
- Back to src_reg and dst_reg, since they denote register numbers
  as opposed to the values stored in these registers.]

[Changes from V2:
- Use src and dst consistently in the document.
- Use a more graphical depiction of the 128-bit instruction.
- Remove `Where:' fragment.
- Clarify that unused bits are reserved and shall be zeroed.]

[Changes from V1:
- Use rst literal blocks for figures.
- Avoid using | in the basic instruction/pseudo instruction figure.
- Rebased to today's bpf-next master branch.]

This patch modifies instruction-set.rst so it documents the encoding
of BPF instructions in terms of how the bytes are stored (be it in an
ELF file or as bytes in a memory buffer to be loaded into the kernel
or some other BPF consumer) as opposed to how the instruction looks
like once loaded.

This is hopefully easier to understand by implementors looking to
generate and/or consume bytes conforming BPF instructions.

The patch also clarifies that the unused bytes in a pseudo-instruction
shall be cleared with zeros.

Signed-off-by: Jose E. Marchesi <jose.marchesi@oracle.com>
Acked-by: Yonghong Song <yhs@fb.com>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/87h6v6i0da.fsf_-_@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/instruction-set.rst | 46 ++++++++++++++++++-----------------
 1 file changed, 24 insertions(+), 22 deletions(-)

diff --git a/Documentation/bpf/instruction-set.rst b/Documentation/bpf/instruction-set.rst
index 01802ed9b29b..db8789e6969e 100644
--- a/Documentation/bpf/instruction-set.rst
+++ b/Documentation/bpf/instruction-set.rst
@@ -38,15 +38,11 @@ eBPF has two instruction encodings:
 * the wide instruction encoding, which appends a second 64-bit immediate (i.e.,
   constant) value after the basic instruction for a total of 128 bits.
 
-The basic instruction encoding looks as follows for a little-endian processor,
-where MSB and LSB mean the most significant bits and least significant bits,
-respectively:
+The fields conforming an encoded basic instruction are stored in the
+following order::
 
-=============  =======  =======  =======  ============
-32 bits (MSB)  16 bits  4 bits   4 bits   8 bits (LSB)
-=============  =======  =======  =======  ============
-imm            offset   src_reg  dst_reg  opcode
-=============  =======  =======  =======  ============
+  opcode:8 src_reg:4 dst_reg:4 offset:16 imm:32 // In little-endian BPF.
+  opcode:8 dst_reg:4 src_reg:4 offset:16 imm:32 // In big-endian BPF.
 
 **imm**
   signed integer immediate value
@@ -64,16 +60,17 @@ imm            offset   src_reg  dst_reg  opcode
 **opcode**
   operation to perform
 
-and as follows for a big-endian processor:
+Note that the contents of multi-byte fields ('imm' and 'offset') are
+stored using big-endian byte ordering in big-endian BPF and
+little-endian byte ordering in little-endian BPF.
 
-=============  =======  =======  =======  ============
-32 bits (MSB)  16 bits  4 bits   4 bits   8 bits (LSB)
-=============  =======  =======  =======  ============
-imm            offset   dst_reg  src_reg  opcode
-=============  =======  =======  =======  ============
+For example::
 
-Multi-byte fields ('imm' and 'offset') are similarly stored in
-the byte order of the processor.
+  opcode                  offset imm          assembly
+         src_reg dst_reg
+  07     0       1        00 00  44 33 22 11  r1 += 0x11223344 // little
+         dst_reg src_reg
+  07     1       0        00 00  11 22 33 44  r1 += 0x11223344 // big
 
 Note that most instructions do not use all of the fields.
 Unused fields shall be cleared to zero.
@@ -84,18 +81,23 @@ The 64 bits following the basic instruction contain a pseudo instruction
 using the same format but with opcode, dst_reg, src_reg, and offset all set to zero,
 and imm containing the high 32 bits of the immediate value.
 
-=================  ==================
-64 bits (MSB)      64 bits (LSB)
-=================  ==================
-basic instruction  pseudo instruction
-=================  ==================
+This is depicted in the following figure::
+
+        basic_instruction
+  .-----------------------------.
+  |                             |
+  code:8 regs:8 offset:16 imm:32 unused:32 imm:32
+                                 |              |
+                                 '--------------'
+                                pseudo instruction
 
 Thus the 64-bit immediate value is constructed as follows:
 
   imm64 = (next_imm << 32) | imm
 
 where 'next_imm' refers to the imm value of the pseudo instruction
-following the basic instruction.
+following the basic instruction.  The unused bytes in the pseudo
+instruction are reserved and shall be cleared to zero.
 
 Instruction classes
 -------------------
-- 
cgit v1.2.3-70-g09d2


From 01c6c9fccbd51c1d9eab0f5794b0271b026178df Mon Sep 17 00:00:00 2001
From: Abinaya Kalaiselvan <quic_akalaise@quicinc.com>
Date: Mon, 19 Dec 2022 11:08:44 +0530
Subject: wifi: ath11k: Add tx ack signal support for management packets

Add support to notify tx ack signal values for management
packets to userspace through nl80211 interface.

Advertise NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT flag
to enable this feature and it will be used for data
packets as well.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Abinaya Kalaiselvan <quic_akalaise@quicinc.com>
Signed-off-by: Maharaja Kennadyrajan <quic_mkenna@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20221219053844.4084486-1-quic_mkenna@quicinc.com
---
 drivers/net/wireless/ath/ath11k/hw.c  |  1 +
 drivers/net/wireless/ath/ath11k/mac.c |  5 +++++
 drivers/net/wireless/ath/ath11k/wmi.c | 27 ++++++++++++++++-----------
 drivers/net/wireless/ath/ath11k/wmi.h |  3 +++
 4 files changed, 25 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/hw.c b/drivers/net/wireless/ath/ath11k/hw.c
index ab8f0ccacc6b..60ac215e0678 100644
--- a/drivers/net/wireless/ath/ath11k/hw.c
+++ b/drivers/net/wireless/ath/ath11k/hw.c
@@ -201,6 +201,7 @@ static void ath11k_init_wmi_config_ipq8074(struct ath11k_base *ab,
 	config->twt_ap_pdev_count = ab->num_radios;
 	config->twt_ap_sta_count = 1000;
 	config->flag1 |= WMI_RSRC_CFG_FLAG1_BSS_CHANNEL_INFO_64;
+	config->flag1 |= WMI_RSRC_CFG_FLAG1_ACK_RSSI;
 }
 
 static int ath11k_hw_mac_id_to_pdev_id_ipq8074(struct ath11k_hw_params *hw,
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index fd1a23ead5e2..cad832e0e6b8 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -9174,6 +9174,11 @@ static int __ath11k_mac_register(struct ath11k *ar)
 		goto err_free_if_combs;
 	}
 
+	if (test_bit(WMI_TLV_SERVICE_TX_DATA_MGMT_ACK_RSSI,
+		     ar->ab->wmi_ab.svc_map))
+		wiphy_ext_feature_set(ar->hw->wiphy,
+				      NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT);
+
 	ar->hw->queues = ATH11K_HW_MAX_QUEUES;
 	ar->hw->wiphy->tx_queue_len = ATH11K_QUEUE_LEN;
 	ar->hw->offchannel_tx_hw_queue = ATH11K_HW_MAX_QUEUES - 1;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 7057c7916d5d..815f49c3b3ad 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -5229,8 +5229,8 @@ static int ath11k_pull_mgmt_rx_params_tlv(struct ath11k_base *ab,
 	return 0;
 }
 
-static int wmi_process_mgmt_tx_comp(struct ath11k *ar, u32 desc_id,
-				    u32 status)
+static int wmi_process_mgmt_tx_comp(struct ath11k *ar,
+				    struct wmi_mgmt_tx_compl_event *tx_compl_param)
 {
 	struct sk_buff *msdu;
 	struct ieee80211_tx_info *info;
@@ -5238,24 +5238,29 @@ static int wmi_process_mgmt_tx_comp(struct ath11k *ar, u32 desc_id,
 	int num_mgmt;
 
 	spin_lock_bh(&ar->txmgmt_idr_lock);
-	msdu = idr_find(&ar->txmgmt_idr, desc_id);
+	msdu = idr_find(&ar->txmgmt_idr, tx_compl_param->desc_id);
 
 	if (!msdu) {
 		ath11k_warn(ar->ab, "received mgmt tx compl for invalid msdu_id: %d\n",
-			    desc_id);
+			    tx_compl_param->desc_id);
 		spin_unlock_bh(&ar->txmgmt_idr_lock);
 		return -ENOENT;
 	}
 
-	idr_remove(&ar->txmgmt_idr, desc_id);
+	idr_remove(&ar->txmgmt_idr, tx_compl_param->desc_id);
 	spin_unlock_bh(&ar->txmgmt_idr_lock);
 
 	skb_cb = ATH11K_SKB_CB(msdu);
 	dma_unmap_single(ar->ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
 
 	info = IEEE80211_SKB_CB(msdu);
-	if ((!(info->flags & IEEE80211_TX_CTL_NO_ACK)) && !status)
+	if ((!(info->flags & IEEE80211_TX_CTL_NO_ACK)) &&
+	    !tx_compl_param->status) {
 		info->flags |= IEEE80211_TX_STAT_ACK;
+		if (test_bit(WMI_TLV_SERVICE_TX_DATA_MGMT_ACK_RSSI,
+			     ar->ab->wmi_ab.svc_map))
+			info->status.ack_signal = tx_compl_param->ack_rssi;
+	}
 
 	ieee80211_tx_status_irqsafe(ar->hw, msdu);
 
@@ -5267,7 +5272,7 @@ static int wmi_process_mgmt_tx_comp(struct ath11k *ar, u32 desc_id,
 
 	ath11k_dbg(ar->ab, ATH11K_DBG_WMI,
 		   "wmi mgmt tx comp pending %d desc id %d\n",
-		   num_mgmt, desc_id);
+		   num_mgmt, tx_compl_param->desc_id);
 
 	if (!num_mgmt)
 		wake_up(&ar->txmgmt_empty_waitq);
@@ -5300,6 +5305,7 @@ static int ath11k_pull_mgmt_tx_compl_param_tlv(struct ath11k_base *ab,
 	param->pdev_id = ev->pdev_id;
 	param->desc_id = ev->desc_id;
 	param->status = ev->status;
+	param->ack_rssi = ev->ack_rssi;
 
 	kfree(tb);
 	return 0;
@@ -7070,13 +7076,12 @@ static void ath11k_mgmt_tx_compl_event(struct ath11k_base *ab, struct sk_buff *s
 		goto exit;
 	}
 
-	wmi_process_mgmt_tx_comp(ar, tx_compl_param.desc_id,
-				 tx_compl_param.status);
+	wmi_process_mgmt_tx_comp(ar, &tx_compl_param);
 
 	ath11k_dbg(ab, ATH11K_DBG_MGMT,
-		   "mgmt tx compl ev pdev_id %d, desc_id %d, status %d",
+		   "mgmt tx compl ev pdev_id %d, desc_id %d, status %d ack_rssi %d",
 		   tx_compl_param.pdev_id, tx_compl_param.desc_id,
-		   tx_compl_param.status);
+		   tx_compl_param.status, tx_compl_param.ack_rssi);
 
 exit:
 	rcu_read_unlock();
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index d3b74f1346f5..63dbb085949f 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2311,6 +2311,7 @@ struct wmi_init_cmd {
 } __packed;
 
 #define WMI_RSRC_CFG_FLAG1_BSS_CHANNEL_INFO_64 BIT(5)
+#define WMI_RSRC_CFG_FLAG1_ACK_RSSI BIT(18)
 
 struct wmi_resource_config {
 	u32 tlv_header;
@@ -4550,6 +4551,8 @@ struct wmi_mgmt_tx_compl_event {
 	u32 desc_id;
 	u32 status;
 	u32 pdev_id;
+	u32 ppdu_id;
+	u32 ack_rssi;
 } __packed;
 
 struct wmi_scan_event {
-- 
cgit v1.2.3-70-g09d2


From 2f46439346700a2b41cf0fa9432f110f42fd8821 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:44 -0800
Subject: bpf: Support "sk_buff" and "xdp_buff" as valid kfunc arg types

The bpf mirror of the in-kernel sk_buff and xdp_buff data structures are
__sk_buff and xdp_md. Currently, when we pass in the program ctx to a
kfunc where the program ctx is a skb or xdp buffer, we reject the
program if the in-kernel definition is sk_buff/xdp_buff instead of
__sk_buff/xdp_md.

This change allows "sk_buff <--> __sk_buff" and "xdp_buff <--> xdp_md"
to be recognized as valid matches. The user program may pass in their
program ctx as a __sk_buff or xdp_md, and the in-kernel definition
of the kfunc may define this arg as a sk_buff or xdp_buff.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-2-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index fa22ec79ac0e..84cca8473873 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5683,6 +5683,10 @@ again:
 	 * int socket_filter_bpf_prog(struct __sk_buff *skb)
 	 * { // no fields of skb are ever used }
 	 */
+	if (strcmp(ctx_tname, "__sk_buff") == 0 && strcmp(tname, "sk_buff") == 0)
+		return ctx_type;
+	if (strcmp(ctx_tname, "xdp_md") == 0 && strcmp(tname, "xdp_buff") == 0)
+		return ctx_type;
 	if (strcmp(ctx_tname, tname)) {
 		/* bpf_user_pt_regs_t is a typedef, so resolve it to
 		 * underlying struct and check name again
-- 
cgit v1.2.3-70-g09d2


From 7e0dac2807e6c4ae8c56941d74971fdb0763b4f9 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:45 -0800
Subject: bpf: Refactor process_dynptr_func

This change cleans up process_dynptr_func's flow to be more intuitive
and updates some comments with more context.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-3-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_verifier.h |  3 ---
 kernel/bpf/verifier.c        | 62 ++++++++++++++++++++++----------------------
 2 files changed, 31 insertions(+), 34 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index cf1bb1cf4a7b..b26ff2a8f63b 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -616,9 +616,6 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env,
 			   enum bpf_arg_type arg_type);
 int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
 		   u32 regno, u32 mem_size);
-struct bpf_call_arg_meta;
-int process_dynptr_func(struct bpf_verifier_env *env, int regno,
-			enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta);
 
 /* this lives here instead of in bpf.h because it needs to dereference tgt_prog */
 static inline u64 bpf_trampoline_compute_key(const struct bpf_prog *tgt_prog,
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5cb8b623f639..e0e00509846b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -959,39 +959,49 @@ static int destroy_if_dynptr_stack_slot(struct bpf_verifier_env *env,
 	return 0;
 }
 
-static bool is_dynptr_reg_valid_uninit(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
-				       int spi)
+static bool is_dynptr_reg_valid_uninit(struct bpf_verifier_env *env, struct bpf_reg_state *reg)
 {
+	int spi;
+
 	if (reg->type == CONST_PTR_TO_DYNPTR)
 		return false;
 
-	/* For -ERANGE (i.e. spi not falling into allocated stack slots), we
-	 * will do check_mem_access to check and update stack bounds later, so
-	 * return true for that case.
+	spi = dynptr_get_spi(env, reg);
+
+	/* -ERANGE (i.e. spi not falling into allocated stack slots) isn't an
+	 * error because this just means the stack state hasn't been updated yet.
+	 * We will do check_mem_access to check and update stack bounds later.
 	 */
-	if (spi < 0)
-		return spi == -ERANGE;
-	/* We allow overwriting existing unreferenced STACK_DYNPTR slots, see
-	 * mark_stack_slots_dynptr which calls destroy_if_dynptr_stack_slot to
-	 * ensure dynptr objects at the slots we are touching are completely
-	 * destructed before we reinitialize them for a new one. For referenced
-	 * ones, destroy_if_dynptr_stack_slot returns an error early instead of
-	 * delaying it until the end where the user will get "Unreleased
+	if (spi < 0 && spi != -ERANGE)
+		return false;
+
+	/* We don't need to check if the stack slots are marked by previous
+	 * dynptr initializations because we allow overwriting existing unreferenced
+	 * STACK_DYNPTR slots, see mark_stack_slots_dynptr which calls
+	 * destroy_if_dynptr_stack_slot to ensure dynptr objects at the slots we are
+	 * touching are completely destructed before we reinitialize them for a new
+	 * one. For referenced ones, destroy_if_dynptr_stack_slot returns an error early
+	 * instead of delaying it until the end where the user will get "Unreleased
 	 * reference" error.
 	 */
 	return true;
 }
 
-static bool is_dynptr_reg_valid_init(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
-				     int spi)
+static bool is_dynptr_reg_valid_init(struct bpf_verifier_env *env, struct bpf_reg_state *reg)
 {
 	struct bpf_func_state *state = func(env, reg);
-	int i;
+	int i, spi;
 
-	/* This already represents first slot of initialized bpf_dynptr */
+	/* This already represents first slot of initialized bpf_dynptr.
+	 *
+	 * CONST_PTR_TO_DYNPTR already has fixed and var_off as 0 due to
+	 * check_func_arg_reg_off's logic, so we don't need to check its
+	 * offset and alignment.
+	 */
 	if (reg->type == CONST_PTR_TO_DYNPTR)
 		return true;
 
+	spi = dynptr_get_spi(env, reg);
 	if (spi < 0)
 		return false;
 	if (!state->stack[spi].spilled_ptr.dynptr.first_slot)
@@ -6215,11 +6225,10 @@ static int process_kptr_func(struct bpf_verifier_env *env, int regno,
  * Helpers which do not mutate the bpf_dynptr set MEM_RDONLY in their argument
  * type, and declare it as 'const struct bpf_dynptr *' in their prototype.
  */
-int process_dynptr_func(struct bpf_verifier_env *env, int regno,
-			enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta)
+static int process_dynptr_func(struct bpf_verifier_env *env, int regno,
+			       enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta)
 {
 	struct bpf_reg_state *regs = cur_regs(env), *reg = &regs[regno];
-	int spi = 0;
 
 	/* MEM_UNINIT and MEM_RDONLY are exclusive, when applied to an
 	 * ARG_PTR_TO_DYNPTR (or ARG_PTR_TO_DYNPTR | DYNPTR_TYPE_*):
@@ -6228,15 +6237,6 @@ int process_dynptr_func(struct bpf_verifier_env *env, int regno,
 		verbose(env, "verifier internal error: misconfigured dynptr helper type flags\n");
 		return -EFAULT;
 	}
-	/* CONST_PTR_TO_DYNPTR already has fixed and var_off as 0 due to
-	 * check_func_arg_reg_off's logic. We only need to check offset
-	 * and its alignment for PTR_TO_STACK.
-	 */
-	if (reg->type == PTR_TO_STACK) {
-		spi = dynptr_get_spi(env, reg);
-		if (spi < 0 && spi != -ERANGE)
-			return spi;
-	}
 
 	/*  MEM_UNINIT - Points to memory that is an appropriate candidate for
 	 *		 constructing a mutable bpf_dynptr object.
@@ -6254,7 +6254,7 @@ int process_dynptr_func(struct bpf_verifier_env *env, int regno,
 	 *		 to.
 	 */
 	if (arg_type & MEM_UNINIT) {
-		if (!is_dynptr_reg_valid_uninit(env, reg, spi)) {
+		if (!is_dynptr_reg_valid_uninit(env, reg)) {
 			verbose(env, "Dynptr has to be an uninitialized dynptr\n");
 			return -EINVAL;
 		}
@@ -6277,7 +6277,7 @@ int process_dynptr_func(struct bpf_verifier_env *env, int regno,
 			return -EINVAL;
 		}
 
-		if (!is_dynptr_reg_valid_init(env, reg, spi)) {
+		if (!is_dynptr_reg_valid_init(env, reg)) {
 			verbose(env,
 				"Expected an initialized dynptr as arg #%d\n",
 				regno);
-- 
cgit v1.2.3-70-g09d2


From 1d18feb2c915c5ad0a9a61d04b8560e8efb78ce8 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:46 -0800
Subject: bpf: Allow initializing dynptrs in kfuncs

This change allows kfuncs to take in an uninitialized dynptr as a
parameter. Before this change, only helper functions could successfully
use uninitialized dynptrs. This change moves the memory access check
(including stack state growing and slot marking) into
process_dynptr_func(), which both helpers and kfuncs call into.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-4-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 67 +++++++++++++++++----------------------------------
 1 file changed, 22 insertions(+), 45 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index e0e00509846b..82e39fc5ed05 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -268,7 +268,6 @@ struct bpf_call_arg_meta {
 	u32 ret_btf_id;
 	u32 subprogno;
 	struct btf_field *kptr_field;
-	u8 uninit_dynptr_regno;
 };
 
 struct btf *btf_vmlinux;
@@ -6225,10 +6224,11 @@ static int process_kptr_func(struct bpf_verifier_env *env, int regno,
  * Helpers which do not mutate the bpf_dynptr set MEM_RDONLY in their argument
  * type, and declare it as 'const struct bpf_dynptr *' in their prototype.
  */
-static int process_dynptr_func(struct bpf_verifier_env *env, int regno,
-			       enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta)
+static int process_dynptr_func(struct bpf_verifier_env *env, int regno, int insn_idx,
+			       enum bpf_arg_type arg_type)
 {
 	struct bpf_reg_state *regs = cur_regs(env), *reg = &regs[regno];
+	int err;
 
 	/* MEM_UNINIT and MEM_RDONLY are exclusive, when applied to an
 	 * ARG_PTR_TO_DYNPTR (or ARG_PTR_TO_DYNPTR | DYNPTR_TYPE_*):
@@ -6254,23 +6254,23 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno,
 	 *		 to.
 	 */
 	if (arg_type & MEM_UNINIT) {
+		int i;
+
 		if (!is_dynptr_reg_valid_uninit(env, reg)) {
 			verbose(env, "Dynptr has to be an uninitialized dynptr\n");
 			return -EINVAL;
 		}
 
-		/* We only support one dynptr being uninitialized at the moment,
-		 * which is sufficient for the helper functions we have right now.
-		 */
-		if (meta->uninit_dynptr_regno) {
-			verbose(env, "verifier internal error: multiple uninitialized dynptr args\n");
-			return -EFAULT;
+		/* we write BPF_DW bits (8 bytes) at a time */
+		for (i = 0; i < BPF_DYNPTR_SIZE; i += 8) {
+			err = check_mem_access(env, insn_idx, regno,
+					       i, BPF_DW, BPF_WRITE, -1, false);
+			if (err)
+				return err;
 		}
 
-		meta->uninit_dynptr_regno = regno;
+		err = mark_stack_slots_dynptr(env, reg, arg_type, insn_idx);
 	} else /* MEM_RDONLY and None case from above */ {
-		int err;
-
 		/* For the reg->type == PTR_TO_STACK case, bpf_dynptr is never const */
 		if (reg->type == CONST_PTR_TO_DYNPTR && !(arg_type & MEM_RDONLY)) {
 			verbose(env, "cannot pass pointer to const bpf_dynptr, the helper mutates it\n");
@@ -6306,10 +6306,8 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno,
 		}
 
 		err = mark_dynptr_read(env, reg);
-		if (err)
-			return err;
 	}
-	return 0;
+	return err;
 }
 
 static bool arg_type_is_mem_size(enum bpf_arg_type type)
@@ -6719,7 +6717,8 @@ static int dynptr_ref_obj_id(struct bpf_verifier_env *env, struct bpf_reg_state
 
 static int check_func_arg(struct bpf_verifier_env *env, u32 arg,
 			  struct bpf_call_arg_meta *meta,
-			  const struct bpf_func_proto *fn)
+			  const struct bpf_func_proto *fn,
+			  int insn_idx)
 {
 	u32 regno = BPF_REG_1 + arg;
 	struct bpf_reg_state *regs = cur_regs(env), *reg = &regs[regno];
@@ -6932,7 +6931,7 @@ skip_type_check:
 		err = check_mem_size_reg(env, reg, regno, true, meta);
 		break;
 	case ARG_PTR_TO_DYNPTR:
-		err = process_dynptr_func(env, regno, arg_type, meta);
+		err = process_dynptr_func(env, regno, insn_idx, arg_type);
 		if (err)
 			return err;
 		break;
@@ -8218,7 +8217,7 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn
 	meta.func_id = func_id;
 	/* check args */
 	for (i = 0; i < MAX_BPF_FUNC_REG_ARGS; i++) {
-		err = check_func_arg(env, i, &meta, fn);
+		err = check_func_arg(env, i, &meta, fn, insn_idx);
 		if (err)
 			return err;
 	}
@@ -8243,30 +8242,6 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn
 
 	regs = cur_regs(env);
 
-	/* This can only be set for PTR_TO_STACK, as CONST_PTR_TO_DYNPTR cannot
-	 * be reinitialized by any dynptr helper. Hence, mark_stack_slots_dynptr
-	 * is safe to do directly.
-	 */
-	if (meta.uninit_dynptr_regno) {
-		if (regs[meta.uninit_dynptr_regno].type == CONST_PTR_TO_DYNPTR) {
-			verbose(env, "verifier internal error: CONST_PTR_TO_DYNPTR cannot be initialized\n");
-			return -EFAULT;
-		}
-		/* we write BPF_DW bits (8 bytes) at a time */
-		for (i = 0; i < BPF_DYNPTR_SIZE; i += 8) {
-			err = check_mem_access(env, insn_idx, meta.uninit_dynptr_regno,
-					       i, BPF_DW, BPF_WRITE, -1, false);
-			if (err)
-				return err;
-		}
-
-		err = mark_stack_slots_dynptr(env, &regs[meta.uninit_dynptr_regno],
-					      fn->arg_type[meta.uninit_dynptr_regno - BPF_REG_1],
-					      insn_idx);
-		if (err)
-			return err;
-	}
-
 	if (meta.release_regno) {
 		err = -EINVAL;
 		/* This can only be set for PTR_TO_STACK, as CONST_PTR_TO_DYNPTR cannot
@@ -9475,7 +9450,8 @@ static int process_kf_arg_ptr_to_rbtree_node(struct bpf_verifier_env *env,
 						  &meta->arg_rbtree_root.field);
 }
 
-static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_arg_meta *meta)
+static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_arg_meta *meta,
+			    int insn_idx)
 {
 	const char *func_name = meta->func_name, *ref_tname;
 	const struct btf *btf = meta->btf;
@@ -9672,7 +9648,8 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 				return -EINVAL;
 			}
 
-			ret = process_dynptr_func(env, regno, ARG_PTR_TO_DYNPTR | MEM_RDONLY, NULL);
+			ret = process_dynptr_func(env, regno, insn_idx,
+						  ARG_PTR_TO_DYNPTR | MEM_RDONLY);
 			if (ret < 0)
 				return ret;
 			break;
@@ -9880,7 +9857,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 	}
 
 	/* Check the arguments */
-	err = check_kfunc_args(env, &meta);
+	err = check_kfunc_args(env, &meta, insn_idx);
 	if (err < 0)
 		return err;
 	/* In case of release function, we get register number of refcounted
-- 
cgit v1.2.3-70-g09d2


From 8357b366cbb09b17c90e2cd758360a6bd2ea7507 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:47 -0800
Subject: bpf: Define no-ops for externally called bpf dynptr functions

Some bpf dynptr functions will be called from places where
if CONFIG_BPF_SYSCALL is not set, then the dynptr function is
undefined. For example, when skb type dynptrs are added in the
next commit, dynptr functions are called from net/core/filter.c

This patch defines no-op implementations of these dynptr functions
so that they do not break compilation by being an undefined reference.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-5-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h | 75 ++++++++++++++++++++++++++++++++---------------------
 1 file changed, 45 insertions(+), 30 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 520b238abd5a..296841a31749 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1124,6 +1124,33 @@ static __always_inline __nocfi unsigned int bpf_dispatcher_nop_func(
 	return bpf_func(ctx, insnsi);
 }
 
+/* the implementation of the opaque uapi struct bpf_dynptr */
+struct bpf_dynptr_kern {
+	void *data;
+	/* Size represents the number of usable bytes of dynptr data.
+	 * If for example the offset is at 4 for a local dynptr whose data is
+	 * of type u64, the number of usable bytes is 4.
+	 *
+	 * The upper 8 bits are reserved. It is as follows:
+	 * Bits 0 - 23 = size
+	 * Bits 24 - 30 = dynptr type
+	 * Bit 31 = whether dynptr is read-only
+	 */
+	u32 size;
+	u32 offset;
+} __aligned(8);
+
+enum bpf_dynptr_type {
+	BPF_DYNPTR_TYPE_INVALID,
+	/* Points to memory that is local to the bpf program */
+	BPF_DYNPTR_TYPE_LOCAL,
+	/* Underlying data is a ringbuf record */
+	BPF_DYNPTR_TYPE_RINGBUF,
+};
+
+int bpf_dynptr_check_size(u32 size);
+u32 bpf_dynptr_get_size(const struct bpf_dynptr_kern *ptr);
+
 #ifdef CONFIG_BPF_JIT
 int bpf_trampoline_link_prog(struct bpf_tramp_link *link, struct bpf_trampoline *tr);
 int bpf_trampoline_unlink_prog(struct bpf_tramp_link *link, struct bpf_trampoline *tr);
@@ -2266,6 +2293,11 @@ static inline bool has_current_bpf_ctx(void)
 }
 
 void notrace bpf_prog_inc_misses_counter(struct bpf_prog *prog);
+
+void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
+		     enum bpf_dynptr_type type, u32 offset, u32 size);
+void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr);
+void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr);
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
 {
@@ -2495,6 +2527,19 @@ static inline void bpf_prog_inc_misses_counter(struct bpf_prog *prog)
 static inline void bpf_cgrp_storage_free(struct cgroup *cgroup)
 {
 }
+
+static inline void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
+				   enum bpf_dynptr_type type, u32 offset, u32 size)
+{
+}
+
+static inline void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr)
+{
+}
+
+static inline void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr)
+{
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 void __bpf_free_used_btfs(struct bpf_prog_aux *aux,
@@ -2913,36 +2958,6 @@ int bpf_bprintf_prepare(char *fmt, u32 fmt_size, const u64 *raw_args,
 			u32 num_args, struct bpf_bprintf_data *data);
 void bpf_bprintf_cleanup(struct bpf_bprintf_data *data);
 
-/* the implementation of the opaque uapi struct bpf_dynptr */
-struct bpf_dynptr_kern {
-	void *data;
-	/* Size represents the number of usable bytes of dynptr data.
-	 * If for example the offset is at 4 for a local dynptr whose data is
-	 * of type u64, the number of usable bytes is 4.
-	 *
-	 * The upper 8 bits are reserved. It is as follows:
-	 * Bits 0 - 23 = size
-	 * Bits 24 - 30 = dynptr type
-	 * Bit 31 = whether dynptr is read-only
-	 */
-	u32 size;
-	u32 offset;
-} __aligned(8);
-
-enum bpf_dynptr_type {
-	BPF_DYNPTR_TYPE_INVALID,
-	/* Points to memory that is local to the bpf program */
-	BPF_DYNPTR_TYPE_LOCAL,
-	/* Underlying data is a kernel-produced ringbuf record */
-	BPF_DYNPTR_TYPE_RINGBUF,
-};
-
-void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
-		     enum bpf_dynptr_type type, u32 offset, u32 size);
-void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr);
-int bpf_dynptr_check_size(u32 size);
-u32 bpf_dynptr_get_size(const struct bpf_dynptr_kern *ptr);
-
 #ifdef CONFIG_BPF_LSM
 void bpf_cgroup_atype_get(u32 attach_btf_id, int cgroup_atype);
 void bpf_cgroup_atype_put(int cgroup_atype);
-- 
cgit v1.2.3-70-g09d2


From 485ec51ef9764c0f67d35cabba0a963936b9126e Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:48 -0800
Subject: bpf: Refactor verifier dynptr into get_dynptr_arg_reg

This commit refactors the logic for determining which register in a
function is the dynptr into "get_dynptr_arg_reg". This will be used
in the future when the dynptr reg for BPF_FUNC_dynptr_write will need
to be obtained in order to support writes for skb dynptrs.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-6-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 80 ++++++++++++++++++++++++++++++++-------------------
 1 file changed, 50 insertions(+), 30 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 82e39fc5ed05..8fd2f26a8977 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -6689,6 +6689,28 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env,
 	}
 }
 
+static struct bpf_reg_state *get_dynptr_arg_reg(struct bpf_verifier_env *env,
+						const struct bpf_func_proto *fn,
+						struct bpf_reg_state *regs)
+{
+	struct bpf_reg_state *state = NULL;
+	int i;
+
+	for (i = 0; i < MAX_BPF_FUNC_REG_ARGS; i++)
+		if (arg_type_is_dynptr(fn->arg_type[i])) {
+			if (state) {
+				verbose(env, "verifier internal error: multiple dynptr args\n");
+				return NULL;
+			}
+			state = &regs[BPF_REG_1 + i];
+		}
+
+	if (!state)
+		verbose(env, "verifier internal error: no dynptr arg found\n");
+
+	return state;
+}
+
 static int dynptr_id(struct bpf_verifier_env *env, struct bpf_reg_state *reg)
 {
 	struct bpf_func_state *state = func(env, reg);
@@ -8326,43 +8348,41 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn
 		}
 		break;
 	case BPF_FUNC_dynptr_data:
-		for (i = 0; i < MAX_BPF_FUNC_REG_ARGS; i++) {
-			if (arg_type_is_dynptr(fn->arg_type[i])) {
-				struct bpf_reg_state *reg = &regs[BPF_REG_1 + i];
-				int id, ref_obj_id;
-
-				if (meta.dynptr_id) {
-					verbose(env, "verifier internal error: meta.dynptr_id already set\n");
-					return -EFAULT;
-				}
-
-				if (meta.ref_obj_id) {
-					verbose(env, "verifier internal error: meta.ref_obj_id already set\n");
-					return -EFAULT;
-				}
+	{
+		struct bpf_reg_state *reg;
+		int id, ref_obj_id;
 
-				id = dynptr_id(env, reg);
-				if (id < 0) {
-					verbose(env, "verifier internal error: failed to obtain dynptr id\n");
-					return id;
-				}
+		reg = get_dynptr_arg_reg(env, fn, regs);
+		if (!reg)
+			return -EFAULT;
 
-				ref_obj_id = dynptr_ref_obj_id(env, reg);
-				if (ref_obj_id < 0) {
-					verbose(env, "verifier internal error: failed to obtain dynptr ref_obj_id\n");
-					return ref_obj_id;
-				}
 
-				meta.dynptr_id = id;
-				meta.ref_obj_id = ref_obj_id;
-				break;
-			}
+		if (meta.dynptr_id) {
+			verbose(env, "verifier internal error: meta.dynptr_id already set\n");
+			return -EFAULT;
 		}
-		if (i == MAX_BPF_FUNC_REG_ARGS) {
-			verbose(env, "verifier internal error: no dynptr in bpf_dynptr_data()\n");
+		if (meta.ref_obj_id) {
+			verbose(env, "verifier internal error: meta.ref_obj_id already set\n");
 			return -EFAULT;
 		}
+
+		id = dynptr_id(env, reg);
+		if (id < 0) {
+			verbose(env, "verifier internal error: failed to obtain dynptr id\n");
+			return id;
+		}
+
+		ref_obj_id = dynptr_ref_obj_id(env, reg);
+		if (ref_obj_id < 0) {
+			verbose(env, "verifier internal error: failed to obtain dynptr ref_obj_id\n");
+			return ref_obj_id;
+		}
+
+		meta.dynptr_id = id;
+		meta.ref_obj_id = ref_obj_id;
+
 		break;
+	}
 	case BPF_FUNC_user_ringbuf_drain:
 		err = __check_func_call(env, insn, insn_idx_p, meta.subprogno,
 					set_user_ringbuf_callback_state);
-- 
cgit v1.2.3-70-g09d2


From d96d937d7c5c12237dce1f14bf0fc9900cabba09 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:49 -0800
Subject: bpf: Add __uninit kfunc annotation

This patch adds __uninit as a kfunc annotation.

This will be useful for scenarios such as for example in dynptrs,
indicating whether the dynptr should be checked by the verifier as an
initialized or an uninitialized dynptr.

Without this annotation, the alternative would be needing to hard-code
in the verifier the specific kfunc to indicate that arg should be
treated as an uninitialized arg.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-7-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst | 17 +++++++++++++++++
 kernel/bpf/verifier.c        | 18 ++++++++++++++++--
 2 files changed, 33 insertions(+), 2 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index 226313747be5..9a78533d25ac 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -100,6 +100,23 @@ Hence, whenever a constant scalar argument is accepted by a kfunc which is not a
 size parameter, and the value of the constant matters for program safety, __k
 suffix should be used.
 
+2.2.2 __uninit Annotation
+--------------------
+
+This annotation is used to indicate that the argument will be treated as
+uninitialized.
+
+An example is given below::
+
+        __bpf_kfunc int bpf_dynptr_from_skb(..., struct bpf_dynptr_kern *ptr__uninit)
+        {
+        ...
+        }
+
+Here, the dynptr will be treated as an uninitialized dynptr. Without this
+annotation, the verifier will reject the program if the dynptr passed in is
+not initialized.
+
 .. _BPF_kfunc_nodef:
 
 2.3 Using an existing kernel function
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 8fd2f26a8977..d052aa5800de 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -8727,6 +8727,11 @@ static bool is_kfunc_arg_alloc_obj(const struct btf *btf, const struct btf_param
 	return __kfunc_param_match_suffix(btf, arg, "__alloc");
 }
 
+static bool is_kfunc_arg_uninit(const struct btf *btf, const struct btf_param *arg)
+{
+	return __kfunc_param_match_suffix(btf, arg, "__uninit");
+}
+
 static bool is_kfunc_arg_scalar_with_name(const struct btf *btf,
 					  const struct btf_param *arg,
 					  const char *name)
@@ -9662,17 +9667,26 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 				return ret;
 			break;
 		case KF_ARG_PTR_TO_DYNPTR:
+		{
+			enum bpf_arg_type dynptr_arg_type = ARG_PTR_TO_DYNPTR;
+
 			if (reg->type != PTR_TO_STACK &&
 			    reg->type != CONST_PTR_TO_DYNPTR) {
 				verbose(env, "arg#%d expected pointer to stack or dynptr_ptr\n", i);
 				return -EINVAL;
 			}
 
-			ret = process_dynptr_func(env, regno, insn_idx,
-						  ARG_PTR_TO_DYNPTR | MEM_RDONLY);
+			if (reg->type == CONST_PTR_TO_DYNPTR)
+				dynptr_arg_type |= MEM_RDONLY;
+
+			if (is_kfunc_arg_uninit(btf, &args[i]))
+				dynptr_arg_type |= MEM_UNINIT;
+
+			ret = process_dynptr_func(env, regno, insn_idx, dynptr_arg_type);
 			if (ret < 0)
 				return ret;
 			break;
+		}
 		case KF_ARG_PTR_TO_LIST_HEAD:
 			if (reg->type != PTR_TO_MAP_VALUE &&
 			    reg->type != (PTR_TO_BTF_ID | MEM_ALLOC)) {
-- 
cgit v1.2.3-70-g09d2


From b5964b968ac64c2ec2debee7518499113b27c34e Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:50 -0800
Subject: bpf: Add skb dynptrs

Add skb dynptrs, which are dynptrs whose underlying pointer points
to a skb. The dynptr acts on skb data. skb dynptrs have two main
benefits. One is that they allow operations on sizes that are not
statically known at compile-time (eg variable-sized accesses).
Another is that parsing the packet data through dynptrs (instead of
through direct access of skb->data and skb->data_end) can be more
ergonomic and less brittle (eg does not need manual if checking for
being within bounds of data_end).

For bpf prog types that don't support writes on skb data, the dynptr is
read-only (bpf_dynptr_write() will return an error)

For reads and writes through the bpf_dynptr_read() and bpf_dynptr_write()
interfaces, reading and writing from/to data in the head as well as from/to
non-linear paged buffers is supported. Data slices through the
bpf_dynptr_data API are not supported; instead bpf_dynptr_slice() and
bpf_dynptr_slice_rdwr() (added in subsequent commit) should be used.

For examples of how skb dynptrs can be used, please see the attached
selftests.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-8-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h            | 14 +++++++-
 include/linux/filter.h         | 18 ++++++++++
 include/uapi/linux/bpf.h       | 13 ++++++--
 kernel/bpf/btf.c               | 18 ++++++++++
 kernel/bpf/helpers.c           | 76 ++++++++++++++++++++++++++++++++++--------
 kernel/bpf/verifier.c          | 61 +++++++++++++++++++++++++++++++++
 net/core/filter.c              | 67 +++++++++++++++++++++++++++++++++++++
 tools/include/uapi/linux/bpf.h | 13 ++++++--
 8 files changed, 261 insertions(+), 19 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 296841a31749..e7436d7615b0 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -607,11 +607,14 @@ enum bpf_type_flag {
 	 */
 	NON_OWN_REF		= BIT(14 + BPF_BASE_TYPE_BITS),
 
+	/* DYNPTR points to sk_buff */
+	DYNPTR_TYPE_SKB		= BIT(15 + BPF_BASE_TYPE_BITS),
+
 	__BPF_TYPE_FLAG_MAX,
 	__BPF_TYPE_LAST_FLAG	= __BPF_TYPE_FLAG_MAX - 1,
 };
 
-#define DYNPTR_TYPE_FLAG_MASK	(DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF)
+#define DYNPTR_TYPE_FLAG_MASK	(DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF | DYNPTR_TYPE_SKB)
 
 /* Max number of base types. */
 #define BPF_BASE_TYPE_LIMIT	(1UL << BPF_BASE_TYPE_BITS)
@@ -1146,6 +1149,8 @@ enum bpf_dynptr_type {
 	BPF_DYNPTR_TYPE_LOCAL,
 	/* Underlying data is a ringbuf record */
 	BPF_DYNPTR_TYPE_RINGBUF,
+	/* Underlying data is a sk_buff */
+	BPF_DYNPTR_TYPE_SKB,
 };
 
 int bpf_dynptr_check_size(u32 size);
@@ -2846,6 +2851,8 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 				struct bpf_insn *insn_buf,
 				struct bpf_prog *prog,
 				u32 *target_size);
+int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
+			       struct bpf_dynptr_kern *ptr);
 #else
 static inline bool bpf_sock_common_is_valid_access(int off, int size,
 						   enum bpf_access_type type,
@@ -2867,6 +2874,11 @@ static inline u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 {
 	return 0;
 }
+static inline int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
+					     struct bpf_dynptr_kern *ptr)
+{
+	return -EOPNOTSUPP;
+}
 #endif
 
 #ifdef CONFIG_INET
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 1727898f1641..de18e844d15e 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1542,4 +1542,22 @@ static __always_inline int __bpf_xdp_redirect_map(struct bpf_map *map, u64 index
 	return XDP_REDIRECT;
 }
 
+#ifdef CONFIG_NET
+int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len);
+int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from,
+			  u32 len, u64 flags);
+#else /* CONFIG_NET */
+static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset,
+				       void *to, u32 len)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset,
+					const void *from, u32 len, u64 flags)
+{
+	return -EOPNOTSUPP;
+}
+#endif /* CONFIG_NET */
+
 #endif /* __LINUX_FILTER_H__ */
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 62ce1f5d1b1d..d0351d30e551 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -5325,11 +5325,17 @@ union bpf_attr {
  *	Description
  *		Write *len* bytes from *src* into *dst*, starting from *offset*
  *		into *dst*.
- *		*flags* is currently unused.
+ *
+ *		*flags* must be 0 except for skb-type dynptrs.
+ *
+ *		For skb-type dynptrs:
+ *		    *  For *flags*, please see the flags accepted by
+ *		       **bpf_skb_store_bytes**\ ().
  *	Return
  *		0 on success, -E2BIG if *offset* + *len* exceeds the length
  *		of *dst*'s data, -EINVAL if *dst* is an invalid dynptr or if *dst*
- *		is a read-only dynptr or if *flags* is not 0.
+ *		is a read-only dynptr or if *flags* is not correct. For skb-type dynptrs,
+ *		other errors correspond to errors returned by **bpf_skb_store_bytes**\ ().
  *
  * void *bpf_dynptr_data(const struct bpf_dynptr *ptr, u32 offset, u32 len)
  *	Description
@@ -5337,6 +5343,9 @@ union bpf_attr {
  *
  *		*len* must be a statically known value. The returned data slice
  *		is invalidated whenever the dynptr is invalidated.
+ *
+ *		skb type dynptrs may not use bpf_dynptr_data. They should
+ *		instead use bpf_dynptr_slice and bpf_dynptr_slice_rdwr.
  *	Return
  *		Pointer to the underlying dynptr data, NULL if the dynptr is
  *		read-only, if the dynptr is invalid, or if the offset and length
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 84cca8473873..ef2d8969ed1f 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -207,6 +207,11 @@ enum btf_kfunc_hook {
 	BTF_KFUNC_HOOK_TRACING,
 	BTF_KFUNC_HOOK_SYSCALL,
 	BTF_KFUNC_HOOK_FMODRET,
+	BTF_KFUNC_HOOK_CGROUP_SKB,
+	BTF_KFUNC_HOOK_SCHED_ACT,
+	BTF_KFUNC_HOOK_SK_SKB,
+	BTF_KFUNC_HOOK_SOCKET_FILTER,
+	BTF_KFUNC_HOOK_LWT,
 	BTF_KFUNC_HOOK_MAX,
 };
 
@@ -7708,6 +7713,19 @@ static int bpf_prog_type_to_kfunc_hook(enum bpf_prog_type prog_type)
 		return BTF_KFUNC_HOOK_TRACING;
 	case BPF_PROG_TYPE_SYSCALL:
 		return BTF_KFUNC_HOOK_SYSCALL;
+	case BPF_PROG_TYPE_CGROUP_SKB:
+		return BTF_KFUNC_HOOK_CGROUP_SKB;
+	case BPF_PROG_TYPE_SCHED_ACT:
+		return BTF_KFUNC_HOOK_SCHED_ACT;
+	case BPF_PROG_TYPE_SK_SKB:
+		return BTF_KFUNC_HOOK_SK_SKB;
+	case BPF_PROG_TYPE_SOCKET_FILTER:
+		return BTF_KFUNC_HOOK_SOCKET_FILTER;
+	case BPF_PROG_TYPE_LWT_OUT:
+	case BPF_PROG_TYPE_LWT_IN:
+	case BPF_PROG_TYPE_LWT_XMIT:
+	case BPF_PROG_TYPE_LWT_SEG6LOCAL:
+		return BTF_KFUNC_HOOK_LWT;
 	default:
 		return BTF_KFUNC_HOOK_MAX;
 	}
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index abdcc52f90a6..e8e2414d1587 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1420,11 +1420,21 @@ static bool bpf_dynptr_is_rdonly(const struct bpf_dynptr_kern *ptr)
 	return ptr->size & DYNPTR_RDONLY_BIT;
 }
 
+void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr)
+{
+	ptr->size |= DYNPTR_RDONLY_BIT;
+}
+
 static void bpf_dynptr_set_type(struct bpf_dynptr_kern *ptr, enum bpf_dynptr_type type)
 {
 	ptr->size |= type << DYNPTR_TYPE_SHIFT;
 }
 
+static enum bpf_dynptr_type bpf_dynptr_get_type(const struct bpf_dynptr_kern *ptr)
+{
+	return (ptr->size & ~(DYNPTR_RDONLY_BIT)) >> DYNPTR_TYPE_SHIFT;
+}
+
 u32 bpf_dynptr_get_size(const struct bpf_dynptr_kern *ptr)
 {
 	return ptr->size & DYNPTR_SIZE_MASK;
@@ -1497,6 +1507,7 @@ static const struct bpf_func_proto bpf_dynptr_from_mem_proto = {
 BPF_CALL_5(bpf_dynptr_read, void *, dst, u32, len, const struct bpf_dynptr_kern *, src,
 	   u32, offset, u64, flags)
 {
+	enum bpf_dynptr_type type;
 	int err;
 
 	if (!src->data || flags)
@@ -1506,13 +1517,23 @@ BPF_CALL_5(bpf_dynptr_read, void *, dst, u32, len, const struct bpf_dynptr_kern
 	if (err)
 		return err;
 
-	/* Source and destination may possibly overlap, hence use memmove to
-	 * copy the data. E.g. bpf_dynptr_from_mem may create two dynptr
-	 * pointing to overlapping PTR_TO_MAP_VALUE regions.
-	 */
-	memmove(dst, src->data + src->offset + offset, len);
+	type = bpf_dynptr_get_type(src);
 
-	return 0;
+	switch (type) {
+	case BPF_DYNPTR_TYPE_LOCAL:
+	case BPF_DYNPTR_TYPE_RINGBUF:
+		/* Source and destination may possibly overlap, hence use memmove to
+		 * copy the data. E.g. bpf_dynptr_from_mem may create two dynptr
+		 * pointing to overlapping PTR_TO_MAP_VALUE regions.
+		 */
+		memmove(dst, src->data + src->offset + offset, len);
+		return 0;
+	case BPF_DYNPTR_TYPE_SKB:
+		return __bpf_skb_load_bytes(src->data, src->offset + offset, dst, len);
+	default:
+		WARN_ONCE(true, "bpf_dynptr_read: unknown dynptr type %d\n", type);
+		return -EFAULT;
+	}
 }
 
 static const struct bpf_func_proto bpf_dynptr_read_proto = {
@@ -1529,22 +1550,36 @@ static const struct bpf_func_proto bpf_dynptr_read_proto = {
 BPF_CALL_5(bpf_dynptr_write, const struct bpf_dynptr_kern *, dst, u32, offset, void *, src,
 	   u32, len, u64, flags)
 {
+	enum bpf_dynptr_type type;
 	int err;
 
-	if (!dst->data || flags || bpf_dynptr_is_rdonly(dst))
+	if (!dst->data || bpf_dynptr_is_rdonly(dst))
 		return -EINVAL;
 
 	err = bpf_dynptr_check_off_len(dst, offset, len);
 	if (err)
 		return err;
 
-	/* Source and destination may possibly overlap, hence use memmove to
-	 * copy the data. E.g. bpf_dynptr_from_mem may create two dynptr
-	 * pointing to overlapping PTR_TO_MAP_VALUE regions.
-	 */
-	memmove(dst->data + dst->offset + offset, src, len);
+	type = bpf_dynptr_get_type(dst);
 
-	return 0;
+	switch (type) {
+	case BPF_DYNPTR_TYPE_LOCAL:
+	case BPF_DYNPTR_TYPE_RINGBUF:
+		if (flags)
+			return -EINVAL;
+		/* Source and destination may possibly overlap, hence use memmove to
+		 * copy the data. E.g. bpf_dynptr_from_mem may create two dynptr
+		 * pointing to overlapping PTR_TO_MAP_VALUE regions.
+		 */
+		memmove(dst->data + dst->offset + offset, src, len);
+		return 0;
+	case BPF_DYNPTR_TYPE_SKB:
+		return __bpf_skb_store_bytes(dst->data, dst->offset + offset, src, len,
+					     flags);
+	default:
+		WARN_ONCE(true, "bpf_dynptr_write: unknown dynptr type %d\n", type);
+		return -EFAULT;
+	}
 }
 
 static const struct bpf_func_proto bpf_dynptr_write_proto = {
@@ -1560,6 +1595,7 @@ static const struct bpf_func_proto bpf_dynptr_write_proto = {
 
 BPF_CALL_3(bpf_dynptr_data, const struct bpf_dynptr_kern *, ptr, u32, offset, u32, len)
 {
+	enum bpf_dynptr_type type;
 	int err;
 
 	if (!ptr->data)
@@ -1572,7 +1608,19 @@ BPF_CALL_3(bpf_dynptr_data, const struct bpf_dynptr_kern *, ptr, u32, offset, u3
 	if (bpf_dynptr_is_rdonly(ptr))
 		return 0;
 
-	return (unsigned long)(ptr->data + ptr->offset + offset);
+	type = bpf_dynptr_get_type(ptr);
+
+	switch (type) {
+	case BPF_DYNPTR_TYPE_LOCAL:
+	case BPF_DYNPTR_TYPE_RINGBUF:
+		return (unsigned long)(ptr->data + ptr->offset + offset);
+	case BPF_DYNPTR_TYPE_SKB:
+		/* skb dynptrs should use bpf_dynptr_slice / bpf_dynptr_slice_rdwr */
+		return 0;
+	default:
+		WARN_ONCE(true, "bpf_dynptr_data: unknown dynptr type %d\n", type);
+		return 0;
+	}
 }
 
 static const struct bpf_func_proto bpf_dynptr_data_proto = {
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d052aa5800de..4f5fce16543b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -750,6 +750,8 @@ static enum bpf_dynptr_type arg_to_dynptr_type(enum bpf_arg_type arg_type)
 		return BPF_DYNPTR_TYPE_LOCAL;
 	case DYNPTR_TYPE_RINGBUF:
 		return BPF_DYNPTR_TYPE_RINGBUF;
+	case DYNPTR_TYPE_SKB:
+		return BPF_DYNPTR_TYPE_SKB;
 	default:
 		return BPF_DYNPTR_TYPE_INVALID;
 	}
@@ -6295,6 +6297,9 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno, int insn
 			case DYNPTR_TYPE_RINGBUF:
 				err_extra = "ringbuf";
 				break;
+			case DYNPTR_TYPE_SKB:
+				err_extra = "skb ";
+				break;
 			default:
 				err_extra = "<unknown>";
 				break;
@@ -6737,6 +6742,24 @@ static int dynptr_ref_obj_id(struct bpf_verifier_env *env, struct bpf_reg_state
 	return state->stack[spi].spilled_ptr.ref_obj_id;
 }
 
+static enum bpf_dynptr_type dynptr_get_type(struct bpf_verifier_env *env,
+					    struct bpf_reg_state *reg)
+{
+	struct bpf_func_state *state = func(env, reg);
+	int spi;
+
+	if (reg->type == CONST_PTR_TO_DYNPTR)
+		return reg->dynptr.type;
+
+	spi = __get_spi(reg->off);
+	if (spi < 0) {
+		verbose(env, "verifier internal error: invalid spi when querying dynptr type\n");
+		return BPF_DYNPTR_TYPE_INVALID;
+	}
+
+	return state->stack[spi].spilled_ptr.dynptr.type;
+}
+
 static int check_func_arg(struct bpf_verifier_env *env, u32 arg,
 			  struct bpf_call_arg_meta *meta,
 			  const struct bpf_func_proto *fn,
@@ -8383,6 +8406,27 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn
 
 		break;
 	}
+	case BPF_FUNC_dynptr_write:
+	{
+		enum bpf_dynptr_type dynptr_type;
+		struct bpf_reg_state *reg;
+
+		reg = get_dynptr_arg_reg(env, fn, regs);
+		if (!reg)
+			return -EFAULT;
+
+		dynptr_type = dynptr_get_type(env, reg);
+		if (dynptr_type == BPF_DYNPTR_TYPE_INVALID)
+			return -EFAULT;
+
+		if (dynptr_type == BPF_DYNPTR_TYPE_SKB)
+			/* this will trigger clear_all_pkt_pointers(), which will
+			 * invalidate all dynptr slices associated with the skb
+			 */
+			changes_data = true;
+
+		break;
+	}
 	case BPF_FUNC_user_ringbuf_drain:
 		err = __check_func_call(env, insn, insn_idx_p, meta.subprogno,
 					set_user_ringbuf_callback_state);
@@ -8898,6 +8942,7 @@ enum special_kfunc_type {
 	KF_bpf_rbtree_remove,
 	KF_bpf_rbtree_add,
 	KF_bpf_rbtree_first,
+	KF_bpf_dynptr_from_skb,
 };
 
 BTF_SET_START(special_kfunc_set)
@@ -8912,6 +8957,7 @@ BTF_ID(func, bpf_rdonly_cast)
 BTF_ID(func, bpf_rbtree_remove)
 BTF_ID(func, bpf_rbtree_add)
 BTF_ID(func, bpf_rbtree_first)
+BTF_ID(func, bpf_dynptr_from_skb)
 BTF_SET_END(special_kfunc_set)
 
 BTF_ID_LIST(special_kfunc_list)
@@ -8928,6 +8974,7 @@ BTF_ID(func, bpf_rcu_read_unlock)
 BTF_ID(func, bpf_rbtree_remove)
 BTF_ID(func, bpf_rbtree_add)
 BTF_ID(func, bpf_rbtree_first)
+BTF_ID(func, bpf_dynptr_from_skb)
 
 static bool is_kfunc_bpf_rcu_read_lock(struct bpf_kfunc_call_arg_meta *meta)
 {
@@ -9682,6 +9729,9 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 			if (is_kfunc_arg_uninit(btf, &args[i]))
 				dynptr_arg_type |= MEM_UNINIT;
 
+			if (meta->func_id == special_kfunc_list[KF_bpf_dynptr_from_skb])
+				dynptr_arg_type |= DYNPTR_TYPE_SKB;
+
 			ret = process_dynptr_func(env, regno, insn_idx, dynptr_arg_type);
 			if (ret < 0)
 				return ret;
@@ -16356,6 +16406,17 @@ static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		   desc->func_id == special_kfunc_list[KF_bpf_rdonly_cast]) {
 		insn_buf[0] = BPF_MOV64_REG(BPF_REG_0, BPF_REG_1);
 		*cnt = 1;
+	} else if (desc->func_id == special_kfunc_list[KF_bpf_dynptr_from_skb]) {
+		bool seen_direct_write = env->seen_direct_write;
+		bool is_rdonly = !may_access_direct_pkt_data(env, NULL, BPF_WRITE);
+
+		if (is_rdonly)
+			insn->imm = BPF_CALL_IMM(bpf_dynptr_from_skb_rdonly);
+
+		/* restore env->seen_direct_write to its original value, since
+		 * may_access_direct_pkt_data mutates it
+		 */
+		env->seen_direct_write = seen_direct_write;
 	}
 	return 0;
 }
diff --git a/net/core/filter.c b/net/core/filter.c
index 1d6f165923bf..f3afa31a9b10 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1721,6 +1721,12 @@ static const struct bpf_func_proto bpf_skb_store_bytes_proto = {
 	.arg5_type	= ARG_ANYTHING,
 };
 
+int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from,
+			  u32 len, u64 flags)
+{
+	return ____bpf_skb_store_bytes(skb, offset, from, len, flags);
+}
+
 BPF_CALL_4(bpf_skb_load_bytes, const struct sk_buff *, skb, u32, offset,
 	   void *, to, u32, len)
 {
@@ -1751,6 +1757,11 @@ static const struct bpf_func_proto bpf_skb_load_bytes_proto = {
 	.arg4_type	= ARG_CONST_SIZE,
 };
 
+int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len)
+{
+	return ____bpf_skb_load_bytes(skb, offset, to, len);
+}
+
 BPF_CALL_4(bpf_flow_dissector_load_bytes,
 	   const struct bpf_flow_dissector *, ctx, u32, offset,
 	   void *, to, u32, len)
@@ -11621,3 +11632,59 @@ bpf_sk_base_func_proto(enum bpf_func_id func_id)
 
 	return func;
 }
+
+__diag_push();
+__diag_ignore_all("-Wmissing-prototypes",
+		  "Global functions as their definitions will be in vmlinux BTF");
+__bpf_kfunc int bpf_dynptr_from_skb(struct sk_buff *skb, u64 flags,
+				    struct bpf_dynptr_kern *ptr__uninit)
+{
+	if (flags) {
+		bpf_dynptr_set_null(ptr__uninit);
+		return -EINVAL;
+	}
+
+	bpf_dynptr_init(ptr__uninit, skb, BPF_DYNPTR_TYPE_SKB, 0, skb->len);
+
+	return 0;
+}
+__diag_pop();
+
+int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
+			       struct bpf_dynptr_kern *ptr__uninit)
+{
+	int err;
+
+	err = bpf_dynptr_from_skb(skb, flags, ptr__uninit);
+	if (err)
+		return err;
+
+	bpf_dynptr_set_rdonly(ptr__uninit);
+
+	return 0;
+}
+
+BTF_SET8_START(bpf_kfunc_check_set_skb)
+BTF_ID_FLAGS(func, bpf_dynptr_from_skb)
+BTF_SET8_END(bpf_kfunc_check_set_skb)
+
+static const struct btf_kfunc_id_set bpf_kfunc_set_skb = {
+	.owner = THIS_MODULE,
+	.set = &bpf_kfunc_check_set_skb,
+};
+
+static int __init bpf_kfunc_init(void)
+{
+	int ret;
+
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_ACT, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SK_SKB, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SOCKET_FILTER, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SKB, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_OUT, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_IN, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_XMIT, &bpf_kfunc_set_skb);
+	return ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_SEG6LOCAL, &bpf_kfunc_set_skb);
+}
+late_initcall(bpf_kfunc_init);
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 62ce1f5d1b1d..d0351d30e551 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -5325,11 +5325,17 @@ union bpf_attr {
  *	Description
  *		Write *len* bytes from *src* into *dst*, starting from *offset*
  *		into *dst*.
- *		*flags* is currently unused.
+ *
+ *		*flags* must be 0 except for skb-type dynptrs.
+ *
+ *		For skb-type dynptrs:
+ *		    *  For *flags*, please see the flags accepted by
+ *		       **bpf_skb_store_bytes**\ ().
  *	Return
  *		0 on success, -E2BIG if *offset* + *len* exceeds the length
  *		of *dst*'s data, -EINVAL if *dst* is an invalid dynptr or if *dst*
- *		is a read-only dynptr or if *flags* is not 0.
+ *		is a read-only dynptr or if *flags* is not correct. For skb-type dynptrs,
+ *		other errors correspond to errors returned by **bpf_skb_store_bytes**\ ().
  *
  * void *bpf_dynptr_data(const struct bpf_dynptr *ptr, u32 offset, u32 len)
  *	Description
@@ -5337,6 +5343,9 @@ union bpf_attr {
  *
  *		*len* must be a statically known value. The returned data slice
  *		is invalidated whenever the dynptr is invalidated.
+ *
+ *		skb type dynptrs may not use bpf_dynptr_data. They should
+ *		instead use bpf_dynptr_slice and bpf_dynptr_slice_rdwr.
  *	Return
  *		Pointer to the underlying dynptr data, NULL if the dynptr is
  *		read-only, if the dynptr is invalid, or if the offset and length
-- 
cgit v1.2.3-70-g09d2


From 05421aecd4ed65da0dc17b0c3c13779ef334e9e5 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:51 -0800
Subject: bpf: Add xdp dynptrs

Add xdp dynptrs, which are dynptrs whose underlying pointer points
to a xdp_buff. The dynptr acts on xdp data. xdp dynptrs have two main
benefits. One is that they allow operations on sizes that are not
statically known at compile-time (eg variable-sized accesses).
Another is that parsing the packet data through dynptrs (instead of
through direct access of xdp->data and xdp->data_end) can be more
ergonomic and less brittle (eg does not need manual if checking for
being within bounds of data_end).

For reads and writes on the dynptr, this includes reading/writing
from/to and across fragments. Data slices through the bpf_dynptr_data
API are not supported; instead bpf_dynptr_slice() and
bpf_dynptr_slice_rdwr() should be used.

For examples of how xdp dynptrs can be used, please see the attached
selftests.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-9-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h            |  8 +++++++-
 include/linux/filter.h         | 14 ++++++++++++++
 include/uapi/linux/bpf.h       |  2 +-
 kernel/bpf/helpers.c           |  9 ++++++++-
 kernel/bpf/verifier.c          | 10 ++++++++++
 net/core/filter.c              | 37 +++++++++++++++++++++++++++++++++++--
 tools/include/uapi/linux/bpf.h |  2 +-
 7 files changed, 76 insertions(+), 6 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index e7436d7615b0..23ec684e660d 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -610,11 +610,15 @@ enum bpf_type_flag {
 	/* DYNPTR points to sk_buff */
 	DYNPTR_TYPE_SKB		= BIT(15 + BPF_BASE_TYPE_BITS),
 
+	/* DYNPTR points to xdp_buff */
+	DYNPTR_TYPE_XDP		= BIT(16 + BPF_BASE_TYPE_BITS),
+
 	__BPF_TYPE_FLAG_MAX,
 	__BPF_TYPE_LAST_FLAG	= __BPF_TYPE_FLAG_MAX - 1,
 };
 
-#define DYNPTR_TYPE_FLAG_MASK	(DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF | DYNPTR_TYPE_SKB)
+#define DYNPTR_TYPE_FLAG_MASK	(DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF | DYNPTR_TYPE_SKB \
+				 | DYNPTR_TYPE_XDP)
 
 /* Max number of base types. */
 #define BPF_BASE_TYPE_LIMIT	(1UL << BPF_BASE_TYPE_BITS)
@@ -1151,6 +1155,8 @@ enum bpf_dynptr_type {
 	BPF_DYNPTR_TYPE_RINGBUF,
 	/* Underlying data is a sk_buff */
 	BPF_DYNPTR_TYPE_SKB,
+	/* Underlying data is a xdp_buff */
+	BPF_DYNPTR_TYPE_XDP,
 };
 
 int bpf_dynptr_check_size(u32 size);
diff --git a/include/linux/filter.h b/include/linux/filter.h
index de18e844d15e..3f6992261ec5 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1546,6 +1546,8 @@ static __always_inline int __bpf_xdp_redirect_map(struct bpf_map *map, u64 index
 int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len);
 int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from,
 			  u32 len, u64 flags);
+int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len);
+int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len);
 #else /* CONFIG_NET */
 static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset,
 				       void *to, u32 len)
@@ -1558,6 +1560,18 @@ static inline int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset,
 {
 	return -EOPNOTSUPP;
 }
+
+static inline int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset,
+				       void *buf, u32 len)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset,
+					void *buf, u32 len)
+{
+	return -EOPNOTSUPP;
+}
 #endif /* CONFIG_NET */
 
 #endif /* __LINUX_FILTER_H__ */
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index d0351d30e551..faa304c926cf 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -5344,7 +5344,7 @@ union bpf_attr {
  *		*len* must be a statically known value. The returned data slice
  *		is invalidated whenever the dynptr is invalidated.
  *
- *		skb type dynptrs may not use bpf_dynptr_data. They should
+ *		skb and xdp type dynptrs may not use bpf_dynptr_data. They should
  *		instead use bpf_dynptr_slice and bpf_dynptr_slice_rdwr.
  *	Return
  *		Pointer to the underlying dynptr data, NULL if the dynptr is
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index e8e2414d1587..114a875a05b1 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1530,6 +1530,8 @@ BPF_CALL_5(bpf_dynptr_read, void *, dst, u32, len, const struct bpf_dynptr_kern
 		return 0;
 	case BPF_DYNPTR_TYPE_SKB:
 		return __bpf_skb_load_bytes(src->data, src->offset + offset, dst, len);
+	case BPF_DYNPTR_TYPE_XDP:
+		return __bpf_xdp_load_bytes(src->data, src->offset + offset, dst, len);
 	default:
 		WARN_ONCE(true, "bpf_dynptr_read: unknown dynptr type %d\n", type);
 		return -EFAULT;
@@ -1576,6 +1578,10 @@ BPF_CALL_5(bpf_dynptr_write, const struct bpf_dynptr_kern *, dst, u32, offset, v
 	case BPF_DYNPTR_TYPE_SKB:
 		return __bpf_skb_store_bytes(dst->data, dst->offset + offset, src, len,
 					     flags);
+	case BPF_DYNPTR_TYPE_XDP:
+		if (flags)
+			return -EINVAL;
+		return __bpf_xdp_store_bytes(dst->data, dst->offset + offset, src, len);
 	default:
 		WARN_ONCE(true, "bpf_dynptr_write: unknown dynptr type %d\n", type);
 		return -EFAULT;
@@ -1615,7 +1621,8 @@ BPF_CALL_3(bpf_dynptr_data, const struct bpf_dynptr_kern *, ptr, u32, offset, u3
 	case BPF_DYNPTR_TYPE_RINGBUF:
 		return (unsigned long)(ptr->data + ptr->offset + offset);
 	case BPF_DYNPTR_TYPE_SKB:
-		/* skb dynptrs should use bpf_dynptr_slice / bpf_dynptr_slice_rdwr */
+	case BPF_DYNPTR_TYPE_XDP:
+		/* skb and xdp dynptrs should use bpf_dynptr_slice / bpf_dynptr_slice_rdwr */
 		return 0;
 	default:
 		WARN_ONCE(true, "bpf_dynptr_data: unknown dynptr type %d\n", type);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 4f5fce16543b..5e42946e53ab 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -752,6 +752,8 @@ static enum bpf_dynptr_type arg_to_dynptr_type(enum bpf_arg_type arg_type)
 		return BPF_DYNPTR_TYPE_RINGBUF;
 	case DYNPTR_TYPE_SKB:
 		return BPF_DYNPTR_TYPE_SKB;
+	case DYNPTR_TYPE_XDP:
+		return BPF_DYNPTR_TYPE_XDP;
 	default:
 		return BPF_DYNPTR_TYPE_INVALID;
 	}
@@ -6300,6 +6302,9 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno, int insn
 			case DYNPTR_TYPE_SKB:
 				err_extra = "skb ";
 				break;
+			case DYNPTR_TYPE_XDP:
+				err_extra = "xdp ";
+				break;
 			default:
 				err_extra = "<unknown>";
 				break;
@@ -8943,6 +8948,7 @@ enum special_kfunc_type {
 	KF_bpf_rbtree_add,
 	KF_bpf_rbtree_first,
 	KF_bpf_dynptr_from_skb,
+	KF_bpf_dynptr_from_xdp,
 };
 
 BTF_SET_START(special_kfunc_set)
@@ -8958,6 +8964,7 @@ BTF_ID(func, bpf_rbtree_remove)
 BTF_ID(func, bpf_rbtree_add)
 BTF_ID(func, bpf_rbtree_first)
 BTF_ID(func, bpf_dynptr_from_skb)
+BTF_ID(func, bpf_dynptr_from_xdp)
 BTF_SET_END(special_kfunc_set)
 
 BTF_ID_LIST(special_kfunc_list)
@@ -8975,6 +8982,7 @@ BTF_ID(func, bpf_rbtree_remove)
 BTF_ID(func, bpf_rbtree_add)
 BTF_ID(func, bpf_rbtree_first)
 BTF_ID(func, bpf_dynptr_from_skb)
+BTF_ID(func, bpf_dynptr_from_xdp)
 
 static bool is_kfunc_bpf_rcu_read_lock(struct bpf_kfunc_call_arg_meta *meta)
 {
@@ -9731,6 +9739,8 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 
 			if (meta->func_id == special_kfunc_list[KF_bpf_dynptr_from_skb])
 				dynptr_arg_type |= DYNPTR_TYPE_SKB;
+			else if (meta->func_id == special_kfunc_list[KF_bpf_dynptr_from_xdp])
+				dynptr_arg_type |= DYNPTR_TYPE_XDP;
 
 			ret = process_dynptr_func(env, regno, insn_idx, dynptr_arg_type);
 			if (ret < 0)
diff --git a/net/core/filter.c b/net/core/filter.c
index f3afa31a9b10..c692046fa7f6 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3839,7 +3839,7 @@ static const struct bpf_func_proto sk_skb_change_head_proto = {
 	.arg3_type	= ARG_ANYTHING,
 };
 
-BPF_CALL_1(bpf_xdp_get_buff_len, struct  xdp_buff*, xdp)
+BPF_CALL_1(bpf_xdp_get_buff_len, struct xdp_buff*, xdp)
 {
 	return xdp_get_buff_len(xdp);
 }
@@ -3999,6 +3999,11 @@ static const struct bpf_func_proto bpf_xdp_load_bytes_proto = {
 	.arg4_type	= ARG_CONST_SIZE,
 };
 
+int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len)
+{
+	return ____bpf_xdp_load_bytes(xdp, offset, buf, len);
+}
+
 BPF_CALL_4(bpf_xdp_store_bytes, struct xdp_buff *, xdp, u32, offset,
 	   void *, buf, u32, len)
 {
@@ -4026,6 +4031,11 @@ static const struct bpf_func_proto bpf_xdp_store_bytes_proto = {
 	.arg4_type	= ARG_CONST_SIZE,
 };
 
+int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len)
+{
+	return ____bpf_xdp_store_bytes(xdp, offset, buf, len);
+}
+
 static int bpf_xdp_frags_increase_tail(struct xdp_buff *xdp, int offset)
 {
 	struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
@@ -11648,6 +11658,19 @@ __bpf_kfunc int bpf_dynptr_from_skb(struct sk_buff *skb, u64 flags,
 
 	return 0;
 }
+
+__bpf_kfunc int bpf_dynptr_from_xdp(struct xdp_buff *xdp, u64 flags,
+				    struct bpf_dynptr_kern *ptr__uninit)
+{
+	if (flags) {
+		bpf_dynptr_set_null(ptr__uninit);
+		return -EINVAL;
+	}
+
+	bpf_dynptr_init(ptr__uninit, xdp, BPF_DYNPTR_TYPE_XDP, 0, xdp_get_buff_len(xdp));
+
+	return 0;
+}
 __diag_pop();
 
 int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
@@ -11668,11 +11691,20 @@ BTF_SET8_START(bpf_kfunc_check_set_skb)
 BTF_ID_FLAGS(func, bpf_dynptr_from_skb)
 BTF_SET8_END(bpf_kfunc_check_set_skb)
 
+BTF_SET8_START(bpf_kfunc_check_set_xdp)
+BTF_ID_FLAGS(func, bpf_dynptr_from_xdp)
+BTF_SET8_END(bpf_kfunc_check_set_xdp)
+
 static const struct btf_kfunc_id_set bpf_kfunc_set_skb = {
 	.owner = THIS_MODULE,
 	.set = &bpf_kfunc_check_set_skb,
 };
 
+static const struct btf_kfunc_id_set bpf_kfunc_set_xdp = {
+	.owner = THIS_MODULE,
+	.set = &bpf_kfunc_check_set_xdp,
+};
+
 static int __init bpf_kfunc_init(void)
 {
 	int ret;
@@ -11685,6 +11717,7 @@ static int __init bpf_kfunc_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_OUT, &bpf_kfunc_set_skb);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_IN, &bpf_kfunc_set_skb);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_XMIT, &bpf_kfunc_set_skb);
-	return ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_SEG6LOCAL, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_SEG6LOCAL, &bpf_kfunc_set_skb);
+	return ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_XDP, &bpf_kfunc_set_xdp);
 }
 late_initcall(bpf_kfunc_init);
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index d0351d30e551..faa304c926cf 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -5344,7 +5344,7 @@ union bpf_attr {
  *		*len* must be a statically known value. The returned data slice
  *		is invalidated whenever the dynptr is invalidated.
  *
- *		skb type dynptrs may not use bpf_dynptr_data. They should
+ *		skb and xdp type dynptrs may not use bpf_dynptr_data. They should
  *		instead use bpf_dynptr_slice and bpf_dynptr_slice_rdwr.
  *	Return
  *		Pointer to the underlying dynptr data, NULL if the dynptr is
-- 
cgit v1.2.3-70-g09d2


From 66e3a13e7c2c44d0c9dd6bb244680ca7529a8845 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:52 -0800
Subject: bpf: Add bpf_dynptr_slice and bpf_dynptr_slice_rdwr

Two new kfuncs are added, bpf_dynptr_slice and bpf_dynptr_slice_rdwr.
The user must pass in a buffer to store the contents of the data slice
if a direct pointer to the data cannot be obtained.

For skb and xdp type dynptrs, these two APIs are the only way to obtain
a data slice. However, for other types of dynptrs, there is no
difference between bpf_dynptr_slice(_rdwr) and bpf_dynptr_data.

For skb type dynptrs, the data is copied into the user provided buffer
if any of the data is not in the linear portion of the skb. For xdp type
dynptrs, the data is copied into the user provided buffer if the data is
between xdp frags.

If the skb is cloned and a call to bpf_dynptr_data_rdwr is made, then
the skb will be uncloned (see bpf_unclone_prologue()).

Please note that any bpf_dynptr_write() automatically invalidates any prior
data slices of the skb dynptr. This is because the skb may be cloned or
may need to pull its paged buffer into the head. As such, any
bpf_dynptr_write() will automatically have its prior data slices
invalidated, even if the write is to data in the skb head of an uncloned
skb. Please note as well that any other helper calls that change the
underlying packet buffer (eg bpf_skb_pull_data()) invalidates any data
slices of the skb dynptr as well, for the same reasons.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Link: https://lore.kernel.org/r/20230301154953.641654-10-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/filter.h         |  14 +++++
 include/uapi/linux/bpf.h       |   5 ++
 kernel/bpf/helpers.c           | 138 +++++++++++++++++++++++++++++++++++++++++
 kernel/bpf/verifier.c          | 127 +++++++++++++++++++++++++++++++++++--
 net/core/filter.c              |   6 +-
 tools/include/uapi/linux/bpf.h |   5 ++
 6 files changed, 288 insertions(+), 7 deletions(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 3f6992261ec5..efa5d4a1677e 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1548,6 +1548,9 @@ int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from,
 			  u32 len, u64 flags);
 int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len);
 int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len);
+void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len);
+void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off,
+		      void *buf, unsigned long len, bool flush);
 #else /* CONFIG_NET */
 static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset,
 				       void *to, u32 len)
@@ -1572,6 +1575,17 @@ static inline int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset,
 {
 	return -EOPNOTSUPP;
 }
+
+static inline void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len)
+{
+	return NULL;
+}
+
+static inline void *bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off, void *buf,
+				     unsigned long len, bool flush)
+{
+	return NULL;
+}
 #endif /* CONFIG_NET */
 
 #endif /* __LINUX_FILTER_H__ */
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index faa304c926cf..c9699304aed2 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -5329,6 +5329,11 @@ union bpf_attr {
  *		*flags* must be 0 except for skb-type dynptrs.
  *
  *		For skb-type dynptrs:
+ *		    *  All data slices of the dynptr are automatically
+ *		       invalidated after **bpf_dynptr_write**\ (). This is
+ *		       because writing may pull the skb and change the
+ *		       underlying packet buffer.
+ *
  *		    *  For *flags*, please see the flags accepted by
  *		       **bpf_skb_store_bytes**\ ().
  *	Return
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 114a875a05b1..648b29e78b84 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2193,6 +2193,142 @@ __bpf_kfunc struct task_struct *bpf_task_from_pid(s32 pid)
 	return p;
 }
 
+/**
+ * bpf_dynptr_slice - Obtain a read-only pointer to the dynptr data.
+ *
+ * For non-skb and non-xdp type dynptrs, there is no difference between
+ * bpf_dynptr_slice and bpf_dynptr_data.
+ *
+ * If the intention is to write to the data slice, please use
+ * bpf_dynptr_slice_rdwr.
+ *
+ * The user must check that the returned pointer is not null before using it.
+ *
+ * Please note that in the case of skb and xdp dynptrs, bpf_dynptr_slice
+ * does not change the underlying packet data pointers, so a call to
+ * bpf_dynptr_slice will not invalidate any ctx->data/data_end pointers in
+ * the bpf program.
+ *
+ * @ptr: The dynptr whose data slice to retrieve
+ * @offset: Offset into the dynptr
+ * @buffer: User-provided buffer to copy contents into
+ * @buffer__szk: Size (in bytes) of the buffer. This is the length of the
+ * requested slice. This must be a constant.
+ *
+ * @returns: NULL if the call failed (eg invalid dynptr), pointer to a read-only
+ * data slice (can be either direct pointer to the data or a pointer to the user
+ * provided buffer, with its contents containing the data, if unable to obtain
+ * direct pointer)
+ */
+__bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset,
+				   void *buffer, u32 buffer__szk)
+{
+	enum bpf_dynptr_type type;
+	u32 len = buffer__szk;
+	int err;
+
+	if (!ptr->data)
+		return 0;
+
+	err = bpf_dynptr_check_off_len(ptr, offset, len);
+	if (err)
+		return 0;
+
+	type = bpf_dynptr_get_type(ptr);
+
+	switch (type) {
+	case BPF_DYNPTR_TYPE_LOCAL:
+	case BPF_DYNPTR_TYPE_RINGBUF:
+		return ptr->data + ptr->offset + offset;
+	case BPF_DYNPTR_TYPE_SKB:
+		return skb_header_pointer(ptr->data, ptr->offset + offset, len, buffer);
+	case BPF_DYNPTR_TYPE_XDP:
+	{
+		void *xdp_ptr = bpf_xdp_pointer(ptr->data, ptr->offset + offset, len);
+		if (xdp_ptr)
+			return xdp_ptr;
+
+		bpf_xdp_copy_buf(ptr->data, ptr->offset + offset, buffer, len, false);
+		return buffer;
+	}
+	default:
+		WARN_ONCE(true, "unknown dynptr type %d\n", type);
+		return 0;
+	}
+}
+
+/**
+ * bpf_dynptr_slice_rdwr - Obtain a writable pointer to the dynptr data.
+ *
+ * For non-skb and non-xdp type dynptrs, there is no difference between
+ * bpf_dynptr_slice and bpf_dynptr_data.
+ *
+ * The returned pointer is writable and may point to either directly the dynptr
+ * data at the requested offset or to the buffer if unable to obtain a direct
+ * data pointer to (example: the requested slice is to the paged area of an skb
+ * packet). In the case where the returned pointer is to the buffer, the user
+ * is responsible for persisting writes through calling bpf_dynptr_write(). This
+ * usually looks something like this pattern:
+ *
+ * struct eth_hdr *eth = bpf_dynptr_slice_rdwr(&dynptr, 0, buffer, sizeof(buffer));
+ * if (!eth)
+ *	return TC_ACT_SHOT;
+ *
+ * // mutate eth header //
+ *
+ * if (eth == buffer)
+ *	bpf_dynptr_write(&ptr, 0, buffer, sizeof(buffer), 0);
+ *
+ * Please note that, as in the example above, the user must check that the
+ * returned pointer is not null before using it.
+ *
+ * Please also note that in the case of skb and xdp dynptrs, bpf_dynptr_slice_rdwr
+ * does not change the underlying packet data pointers, so a call to
+ * bpf_dynptr_slice_rdwr will not invalidate any ctx->data/data_end pointers in
+ * the bpf program.
+ *
+ * @ptr: The dynptr whose data slice to retrieve
+ * @offset: Offset into the dynptr
+ * @buffer: User-provided buffer to copy contents into
+ * @buffer__szk: Size (in bytes) of the buffer. This is the length of the
+ * requested slice. This must be a constant.
+ *
+ * @returns: NULL if the call failed (eg invalid dynptr), pointer to a
+ * data slice (can be either direct pointer to the data or a pointer to the user
+ * provided buffer, with its contents containing the data, if unable to obtain
+ * direct pointer)
+ */
+__bpf_kfunc void *bpf_dynptr_slice_rdwr(const struct bpf_dynptr_kern *ptr, u32 offset,
+					void *buffer, u32 buffer__szk)
+{
+	if (!ptr->data || bpf_dynptr_is_rdonly(ptr))
+		return 0;
+
+	/* bpf_dynptr_slice_rdwr is the same logic as bpf_dynptr_slice.
+	 *
+	 * For skb-type dynptrs, it is safe to write into the returned pointer
+	 * if the bpf program allows skb data writes. There are two possiblities
+	 * that may occur when calling bpf_dynptr_slice_rdwr:
+	 *
+	 * 1) The requested slice is in the head of the skb. In this case, the
+	 * returned pointer is directly to skb data, and if the skb is cloned, the
+	 * verifier will have uncloned it (see bpf_unclone_prologue()) already.
+	 * The pointer can be directly written into.
+	 *
+	 * 2) Some portion of the requested slice is in the paged buffer area.
+	 * In this case, the requested data will be copied out into the buffer
+	 * and the returned pointer will be a pointer to the buffer. The skb
+	 * will not be pulled. To persist the write, the user will need to call
+	 * bpf_dynptr_write(), which will pull the skb and commit the write.
+	 *
+	 * Similarly for xdp programs, if the requested slice is not across xdp
+	 * fragments, then a direct pointer will be returned, otherwise the data
+	 * will be copied out into the buffer and the user will need to call
+	 * bpf_dynptr_write() to commit changes.
+	 */
+	return bpf_dynptr_slice(ptr, offset, buffer, buffer__szk);
+}
+
 __bpf_kfunc void *bpf_cast_to_kern_ctx(void *obj)
 {
 	return obj;
@@ -2262,6 +2398,8 @@ BTF_ID_FLAGS(func, bpf_cast_to_kern_ctx)
 BTF_ID_FLAGS(func, bpf_rdonly_cast)
 BTF_ID_FLAGS(func, bpf_rcu_read_lock)
 BTF_ID_FLAGS(func, bpf_rcu_read_unlock)
+BTF_ID_FLAGS(func, bpf_dynptr_slice, KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_dynptr_slice_rdwr, KF_RET_NULL)
 BTF_SET8_END(common_btf_ids)
 
 static const struct btf_kfunc_id_set common_kfunc_set = {
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5e42946e53ab..a856896e835a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -759,6 +759,22 @@ static enum bpf_dynptr_type arg_to_dynptr_type(enum bpf_arg_type arg_type)
 	}
 }
 
+static enum bpf_type_flag get_dynptr_type_flag(enum bpf_dynptr_type type)
+{
+	switch (type) {
+	case BPF_DYNPTR_TYPE_LOCAL:
+		return DYNPTR_TYPE_LOCAL;
+	case BPF_DYNPTR_TYPE_RINGBUF:
+		return DYNPTR_TYPE_RINGBUF;
+	case BPF_DYNPTR_TYPE_SKB:
+		return DYNPTR_TYPE_SKB;
+	case BPF_DYNPTR_TYPE_XDP:
+		return DYNPTR_TYPE_XDP;
+	default:
+		return 0;
+	}
+}
+
 static bool dynptr_type_refcounted(enum bpf_dynptr_type type)
 {
 	return type == BPF_DYNPTR_TYPE_RINGBUF;
@@ -1681,6 +1697,12 @@ static bool reg_is_pkt_pointer_any(const struct bpf_reg_state *reg)
 	       reg->type == PTR_TO_PACKET_END;
 }
 
+static bool reg_is_dynptr_slice_pkt(const struct bpf_reg_state *reg)
+{
+	return base_type(reg->type) == PTR_TO_MEM &&
+		(reg->type & DYNPTR_TYPE_SKB || reg->type & DYNPTR_TYPE_XDP);
+}
+
 /* Unmodified PTR_TO_PACKET[_META,_END] register from ctx access. */
 static bool reg_is_init_pkt_pointer(const struct bpf_reg_state *reg,
 				    enum bpf_reg_type which)
@@ -7429,6 +7451,9 @@ static int check_func_proto(const struct bpf_func_proto *fn, int func_id)
 
 /* Packet data might have moved, any old PTR_TO_PACKET[_META,_END]
  * are now invalid, so turn them into unknown SCALAR_VALUE.
+ *
+ * This also applies to dynptr slices belonging to skb and xdp dynptrs,
+ * since these slices point to packet data.
  */
 static void clear_all_pkt_pointers(struct bpf_verifier_env *env)
 {
@@ -7436,7 +7461,7 @@ static void clear_all_pkt_pointers(struct bpf_verifier_env *env)
 	struct bpf_reg_state *reg;
 
 	bpf_for_each_reg_in_vstate(env->cur_state, state, reg, ({
-		if (reg_is_pkt_pointer_any(reg))
+		if (reg_is_pkt_pointer_any(reg) || reg_is_dynptr_slice_pkt(reg))
 			mark_reg_invalid(env, reg);
 	}));
 }
@@ -8688,6 +8713,11 @@ struct bpf_kfunc_call_arg_meta {
 	struct {
 		struct btf_field *field;
 	} arg_rbtree_root;
+	struct {
+		enum bpf_dynptr_type type;
+		u32 id;
+	} initialized_dynptr;
+	u64 mem_size;
 };
 
 static bool is_kfunc_acquire(struct bpf_kfunc_call_arg_meta *meta)
@@ -8761,6 +8791,19 @@ static bool is_kfunc_arg_mem_size(const struct btf *btf,
 	return __kfunc_param_match_suffix(btf, arg, "__sz");
 }
 
+static bool is_kfunc_arg_const_mem_size(const struct btf *btf,
+					const struct btf_param *arg,
+					const struct bpf_reg_state *reg)
+{
+	const struct btf_type *t;
+
+	t = btf_type_skip_modifiers(btf, arg->type, NULL);
+	if (!btf_type_is_scalar(t) || reg->type != SCALAR_VALUE)
+		return false;
+
+	return __kfunc_param_match_suffix(btf, arg, "__szk");
+}
+
 static bool is_kfunc_arg_constant(const struct btf *btf, const struct btf_param *arg)
 {
 	return __kfunc_param_match_suffix(btf, arg, "__k");
@@ -8949,6 +8992,8 @@ enum special_kfunc_type {
 	KF_bpf_rbtree_first,
 	KF_bpf_dynptr_from_skb,
 	KF_bpf_dynptr_from_xdp,
+	KF_bpf_dynptr_slice,
+	KF_bpf_dynptr_slice_rdwr,
 };
 
 BTF_SET_START(special_kfunc_set)
@@ -8965,6 +9010,8 @@ BTF_ID(func, bpf_rbtree_add)
 BTF_ID(func, bpf_rbtree_first)
 BTF_ID(func, bpf_dynptr_from_skb)
 BTF_ID(func, bpf_dynptr_from_xdp)
+BTF_ID(func, bpf_dynptr_slice)
+BTF_ID(func, bpf_dynptr_slice_rdwr)
 BTF_SET_END(special_kfunc_set)
 
 BTF_ID_LIST(special_kfunc_list)
@@ -8983,6 +9030,8 @@ BTF_ID(func, bpf_rbtree_add)
 BTF_ID(func, bpf_rbtree_first)
 BTF_ID(func, bpf_dynptr_from_skb)
 BTF_ID(func, bpf_dynptr_from_xdp)
+BTF_ID(func, bpf_dynptr_slice)
+BTF_ID(func, bpf_dynptr_slice_rdwr)
 
 static bool is_kfunc_bpf_rcu_read_lock(struct bpf_kfunc_call_arg_meta *meta)
 {
@@ -9062,7 +9111,10 @@ get_kfunc_ptr_arg_type(struct bpf_verifier_env *env,
 	if (is_kfunc_arg_callback(env, meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_CALLBACK;
 
-	if (argno + 1 < nargs && is_kfunc_arg_mem_size(meta->btf, &args[argno + 1], &regs[regno + 1]))
+
+	if (argno + 1 < nargs &&
+	    (is_kfunc_arg_mem_size(meta->btf, &args[argno + 1], &regs[regno + 1]) ||
+	     is_kfunc_arg_const_mem_size(meta->btf, &args[argno + 1], &regs[regno + 1])))
 		arg_mem_size = true;
 
 	/* This is the catch all argument type of register types supported by
@@ -9745,6 +9797,18 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 			ret = process_dynptr_func(env, regno, insn_idx, dynptr_arg_type);
 			if (ret < 0)
 				return ret;
+
+			if (!(dynptr_arg_type & MEM_UNINIT)) {
+				int id = dynptr_id(env, reg);
+
+				if (id < 0) {
+					verbose(env, "verifier internal error: failed to obtain dynptr id\n");
+					return id;
+				}
+				meta->initialized_dynptr.id = id;
+				meta->initialized_dynptr.type = dynptr_get_type(env, reg);
+			}
+
 			break;
 		}
 		case KF_ARG_PTR_TO_LIST_HEAD:
@@ -9840,14 +9904,33 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 				return ret;
 			break;
 		case KF_ARG_PTR_TO_MEM_SIZE:
-			ret = check_kfunc_mem_size_reg(env, &regs[regno + 1], regno + 1);
+		{
+			struct bpf_reg_state *size_reg = &regs[regno + 1];
+			const struct btf_param *size_arg = &args[i + 1];
+
+			ret = check_kfunc_mem_size_reg(env, size_reg, regno + 1);
 			if (ret < 0) {
 				verbose(env, "arg#%d arg#%d memory, len pair leads to invalid memory access\n", i, i + 1);
 				return ret;
 			}
-			/* Skip next '__sz' argument */
+
+			if (is_kfunc_arg_const_mem_size(meta->btf, size_arg, size_reg)) {
+				if (meta->arg_constant.found) {
+					verbose(env, "verifier internal error: only one constant argument permitted\n");
+					return -EFAULT;
+				}
+				if (!tnum_is_const(size_reg->var_off)) {
+					verbose(env, "R%d must be a known constant\n", regno + 1);
+					return -EINVAL;
+				}
+				meta->arg_constant.found = true;
+				meta->arg_constant.value = size_reg->var_off.value;
+			}
+
+			/* Skip next '__sz' or '__szk' argument */
 			i++;
 			break;
+		}
 		case KF_ARG_PTR_TO_CALLBACK:
 			meta->subprogno = reg->subprogno;
 			break;
@@ -10082,6 +10165,42 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 				regs[BPF_REG_0].type = PTR_TO_BTF_ID | PTR_UNTRUSTED;
 				regs[BPF_REG_0].btf = desc_btf;
 				regs[BPF_REG_0].btf_id = meta.arg_constant.value;
+			} else if (meta.func_id == special_kfunc_list[KF_bpf_dynptr_slice] ||
+				   meta.func_id == special_kfunc_list[KF_bpf_dynptr_slice_rdwr]) {
+				enum bpf_type_flag type_flag = get_dynptr_type_flag(meta.initialized_dynptr.type);
+
+				mark_reg_known_zero(env, regs, BPF_REG_0);
+
+				if (!meta.arg_constant.found) {
+					verbose(env, "verifier internal error: bpf_dynptr_slice(_rdwr) no constant size\n");
+					return -EFAULT;
+				}
+
+				regs[BPF_REG_0].mem_size = meta.arg_constant.value;
+
+				/* PTR_MAYBE_NULL will be added when is_kfunc_ret_null is checked */
+				regs[BPF_REG_0].type = PTR_TO_MEM | type_flag;
+
+				if (meta.func_id == special_kfunc_list[KF_bpf_dynptr_slice]) {
+					regs[BPF_REG_0].type |= MEM_RDONLY;
+				} else {
+					/* this will set env->seen_direct_write to true */
+					if (!may_access_direct_pkt_data(env, NULL, BPF_WRITE)) {
+						verbose(env, "the prog does not allow writes to packet data\n");
+						return -EINVAL;
+					}
+				}
+
+				if (!meta.initialized_dynptr.id) {
+					verbose(env, "verifier internal error: no dynptr id\n");
+					return -EFAULT;
+				}
+				regs[BPF_REG_0].dynptr_id = meta.initialized_dynptr.id;
+
+				/* we don't need to set BPF_REG_0's ref obj id
+				 * because packet slices are not refcounted (see
+				 * dynptr_type_refcounted)
+				 */
 			} else {
 				verbose(env, "kernel function %s unhandled dynamic return type\n",
 					meta.func_name);
diff --git a/net/core/filter.c b/net/core/filter.c
index c692046fa7f6..8f3124e06133 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3894,8 +3894,8 @@ static const struct bpf_func_proto bpf_xdp_adjust_head_proto = {
 	.arg2_type	= ARG_ANYTHING,
 };
 
-static void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off,
-			     void *buf, unsigned long len, bool flush)
+void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off,
+		      void *buf, unsigned long len, bool flush)
 {
 	unsigned long ptr_len, ptr_off = 0;
 	skb_frag_t *next_frag, *end_frag;
@@ -3941,7 +3941,7 @@ static void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off,
 	}
 }
 
-static void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len)
+void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len)
 {
 	struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
 	u32 size = xdp->data_end - xdp->data;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index faa304c926cf..c9699304aed2 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -5329,6 +5329,11 @@ union bpf_attr {
  *		*flags* must be 0 except for skb-type dynptrs.
  *
  *		For skb-type dynptrs:
+ *		    *  All data slices of the dynptr are automatically
+ *		       invalidated after **bpf_dynptr_write**\ (). This is
+ *		       because writing may pull the skb and change the
+ *		       underlying packet buffer.
+ *
  *		    *  For *flags*, please see the flags accepted by
  *		       **bpf_skb_store_bytes**\ ().
  *	Return
-- 
cgit v1.2.3-70-g09d2


From cfa7b011894d689cccfa88a25da324fa5c34e4ed Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 07:49:53 -0800
Subject: selftests/bpf: tests for using dynptrs to parse skb and xdp buffers

Test skb and xdp dynptr functionality in the following ways:

1) progs/test_cls_redirect_dynptr.c
   * Rewrite "progs/test_cls_redirect.c" test to use dynptrs to parse
     skb data

   * This is a great example of how dynptrs can be used to simplify a
     lot of the parsing logic for non-statically known values.

     When measuring the user + system time between the original version
     vs. using dynptrs, and averaging the time for 10 runs (using
     "time ./test_progs -t cls_redirect"):
         original version: 0.092 sec
         with dynptrs: 0.078 sec

2) progs/test_xdp_dynptr.c
   * Rewrite "progs/test_xdp.c" test to use dynptrs to parse xdp data

     When measuring the user + system time between the original version
     vs. using dynptrs, and averaging the time for 10 runs (using
     "time ./test_progs -t xdp_attach"):
         original version: 0.118 sec
         with dynptrs: 0.094 sec

3) progs/test_l4lb_noinline_dynptr.c
   * Rewrite "progs/test_l4lb_noinline.c" test to use dynptrs to parse
     skb data

     When measuring the user + system time between the original version
     vs. using dynptrs, and averaging the time for 10 runs (using
     "time ./test_progs -t l4lb_all"):
         original version: 0.062 sec
         with dynptrs: 0.081 sec

     For number of processed verifier instructions:
         original version: 6268 insns
         with dynptrs: 2588 insns

4) progs/test_parse_tcp_hdr_opt_dynptr.c
   * Add sample code for parsing tcp hdr opt lookup using dynptrs.
     This logic is lifted from a real-world use case of packet parsing
     in katran [0], a layer 4 load balancer. The original version
     "progs/test_parse_tcp_hdr_opt.c" (not using dynptrs) is included
     here as well, for comparison.

     When measuring the user + system time between the original version
     vs. using dynptrs, and averaging the time for 10 runs (using
     "time ./test_progs -t parse_tcp_hdr_opt"):
         original version: 0.031 sec
         with dynptrs: 0.045 sec

5) progs/dynptr_success.c
   * Add test case "test_skb_readonly" for testing attempts at writes
     on a prog type with read-only skb ctx.
   * Add "test_dynptr_skb_data" for testing that bpf_dynptr_data isn't
     supported for skb progs.

6) progs/dynptr_fail.c
   * Add test cases "skb_invalid_data_slice{1,2,3,4}" and
     "xdp_invalid_data_slice{1,2}" for testing that helpers that modify the
     underlying packet buffer automatically invalidate the associated
     data slice.
   * Add test cases "skb_invalid_ctx" and "xdp_invalid_ctx" for testing
     that prog types that do not support bpf_dynptr_from_skb/xdp don't
     have access to the API.
   * Add test case "dynptr_slice_var_len{1,2}" for testing that
     variable-sized len can't be passed in to bpf_dynptr_slice
   * Add test case "skb_invalid_slice_write" for testing that writes to a
     read-only data slice are rejected by the verifier.
   * Add test case "data_slice_out_of_bounds_skb" for testing that
     writes to an area outside the slice are rejected.
   * Add test case "invalid_slice_rdwr_rdonly" for testing that prog
     types that don't allow writes to packet data don't accept any calls
     to bpf_dynptr_slice_rdwr.

[0] https://github.com/facebookincubator/katran/blob/main/katran/lib/bpf/pckt_parsing.h

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230301154953.641654-11-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/DENYLIST.s390x         |   2 +
 tools/testing/selftests/bpf/bpf_kfuncs.h           |  38 +
 .../selftests/bpf/prog_tests/cls_redirect.c        |  25 +
 tools/testing/selftests/bpf/prog_tests/dynptr.c    |  74 +-
 tools/testing/selftests/bpf/prog_tests/l4lb_all.c  |   2 +
 .../selftests/bpf/prog_tests/parse_tcp_hdr_opt.c   |  93 ++
 .../testing/selftests/bpf/prog_tests/xdp_attach.c  |  11 +-
 tools/testing/selftests/bpf/progs/dynptr_fail.c    | 287 +++++-
 tools/testing/selftests/bpf/progs/dynptr_success.c |  55 +-
 .../selftests/bpf/progs/test_cls_redirect_dynptr.c | 980 +++++++++++++++++++++
 .../bpf/progs/test_l4lb_noinline_dynptr.c          | 487 ++++++++++
 .../selftests/bpf/progs/test_parse_tcp_hdr_opt.c   | 119 +++
 .../bpf/progs/test_parse_tcp_hdr_opt_dynptr.c      | 114 +++
 .../testing/selftests/bpf/progs/test_xdp_dynptr.c  | 257 ++++++
 tools/testing/selftests/bpf/test_tcp_hdr_options.h |   1 +
 15 files changed, 2522 insertions(+), 23 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/bpf_kfuncs.h
 create mode 100644 tools/testing/selftests/bpf/prog_tests/parse_tcp_hdr_opt.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_l4lb_noinline_dynptr.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_dynptr.c

diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
index b89eb87034e4..a02a085e7f32 100644
--- a/tools/testing/selftests/bpf/DENYLIST.s390x
+++ b/tools/testing/selftests/bpf/DENYLIST.s390x
@@ -4,6 +4,8 @@ bloom_filter_map                         # failed to find kernel BTF type ID of
 bpf_cookie                               # failed to open_and_load program: -524 (trampoline)
 bpf_loop                                 # attaches to __x64_sys_nanosleep
 cgrp_local_storage                       # prog_attach unexpected error: -524                                          (trampoline)
+dynptr/test_dynptr_skb_data
+dynptr/test_skb_readonly
 fexit_sleep                              # fexit_skel_load fexit skeleton failed                                       (trampoline)
 get_stack_raw_tp                         # user_stack corrupted user stack                                             (no backchain userspace)
 kprobe_multi_bench_attach                # bpf_program__attach_kprobe_multi_opts unexpected error: -95
diff --git a/tools/testing/selftests/bpf/bpf_kfuncs.h b/tools/testing/selftests/bpf/bpf_kfuncs.h
new file mode 100644
index 000000000000..8c993ec8ceea
--- /dev/null
+++ b/tools/testing/selftests/bpf/bpf_kfuncs.h
@@ -0,0 +1,38 @@
+#ifndef __BPF_KFUNCS__
+#define __BPF_KFUNCS__
+
+/* Description
+ *  Initializes an skb-type dynptr
+ * Returns
+ *  Error code
+ */
+extern int bpf_dynptr_from_skb(struct __sk_buff *skb, __u64 flags,
+    struct bpf_dynptr *ptr__uninit) __ksym;
+
+/* Description
+ *  Initializes an xdp-type dynptr
+ * Returns
+ *  Error code
+ */
+extern int bpf_dynptr_from_xdp(struct xdp_md *xdp, __u64 flags,
+			       struct bpf_dynptr *ptr__uninit) __ksym;
+
+/* Description
+ *  Obtain a read-only pointer to the dynptr's data
+ * Returns
+ *  Either a direct pointer to the dynptr data or a pointer to the user-provided
+ *  buffer if unable to obtain a direct pointer
+ */
+extern void *bpf_dynptr_slice(const struct bpf_dynptr *ptr, __u32 offset,
+			      void *buffer, __u32 buffer__szk) __ksym;
+
+/* Description
+ *  Obtain a read-write pointer to the dynptr's data
+ * Returns
+ *  Either a direct pointer to the dynptr data or a pointer to the user-provided
+ *  buffer if unable to obtain a direct pointer
+ */
+extern void *bpf_dynptr_slice_rdwr(const struct bpf_dynptr *ptr, __u32 offset,
+			      void *buffer, __u32 buffer__szk) __ksym;
+
+#endif
diff --git a/tools/testing/selftests/bpf/prog_tests/cls_redirect.c b/tools/testing/selftests/bpf/prog_tests/cls_redirect.c
index 224f016b0a53..2a55f717fc07 100644
--- a/tools/testing/selftests/bpf/prog_tests/cls_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/cls_redirect.c
@@ -13,6 +13,7 @@
 
 #include "progs/test_cls_redirect.h"
 #include "test_cls_redirect.skel.h"
+#include "test_cls_redirect_dynptr.skel.h"
 #include "test_cls_redirect_subprogs.skel.h"
 
 #define ENCAP_IP INADDR_LOOPBACK
@@ -446,6 +447,28 @@ cleanup:
 	close_fds((int *)conns, sizeof(conns) / sizeof(conns[0][0]));
 }
 
+static void test_cls_redirect_dynptr(void)
+{
+	struct test_cls_redirect_dynptr *skel;
+	int err;
+
+	skel = test_cls_redirect_dynptr__open();
+	if (!ASSERT_OK_PTR(skel, "skel_open"))
+		return;
+
+	skel->rodata->ENCAPSULATION_IP = htonl(ENCAP_IP);
+	skel->rodata->ENCAPSULATION_PORT = htons(ENCAP_PORT);
+
+	err = test_cls_redirect_dynptr__load(skel);
+	if (!ASSERT_OK(err, "skel_load"))
+		goto cleanup;
+
+	test_cls_redirect_common(skel->progs.cls_redirect);
+
+cleanup:
+	test_cls_redirect_dynptr__destroy(skel);
+}
+
 static void test_cls_redirect_inlined(void)
 {
 	struct test_cls_redirect *skel;
@@ -496,4 +519,6 @@ void test_cls_redirect(void)
 		test_cls_redirect_inlined();
 	if (test__start_subtest("cls_redirect_subprogs"))
 		test_cls_redirect_subprogs();
+	if (test__start_subtest("cls_redirect_dynptr"))
+		test_cls_redirect_dynptr();
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/dynptr.c b/tools/testing/selftests/bpf/prog_tests/dynptr.c
index b99264ec0d9c..d176c34a7d2e 100644
--- a/tools/testing/selftests/bpf/prog_tests/dynptr.c
+++ b/tools/testing/selftests/bpf/prog_tests/dynptr.c
@@ -2,20 +2,32 @@
 /* Copyright (c) 2022 Facebook */
 
 #include <test_progs.h>
+#include <network_helpers.h>
 #include "dynptr_fail.skel.h"
 #include "dynptr_success.skel.h"
 
-static const char * const success_tests[] = {
-	"test_read_write",
-	"test_data_slice",
-	"test_ringbuf",
+enum test_setup_type {
+	SETUP_SYSCALL_SLEEP,
+	SETUP_SKB_PROG,
 };
 
-static void verify_success(const char *prog_name)
+static struct {
+	const char *prog_name;
+	enum test_setup_type type;
+} success_tests[] = {
+	{"test_read_write", SETUP_SYSCALL_SLEEP},
+	{"test_dynptr_data", SETUP_SYSCALL_SLEEP},
+	{"test_ringbuf", SETUP_SYSCALL_SLEEP},
+	{"test_skb_readonly", SETUP_SKB_PROG},
+	{"test_dynptr_skb_data", SETUP_SKB_PROG},
+};
+
+static void verify_success(const char *prog_name, enum test_setup_type setup_type)
 {
 	struct dynptr_success *skel;
 	struct bpf_program *prog;
 	struct bpf_link *link;
+       int err;
 
 	skel = dynptr_success__open();
 	if (!ASSERT_OK_PTR(skel, "dynptr_success__open"))
@@ -23,23 +35,53 @@ static void verify_success(const char *prog_name)
 
 	skel->bss->pid = getpid();
 
-	dynptr_success__load(skel);
-	if (!ASSERT_OK_PTR(skel, "dynptr_success__load"))
-		goto cleanup;
-
 	prog = bpf_object__find_program_by_name(skel->obj, prog_name);
 	if (!ASSERT_OK_PTR(prog, "bpf_object__find_program_by_name"))
 		goto cleanup;
 
-	link = bpf_program__attach(prog);
-	if (!ASSERT_OK_PTR(link, "bpf_program__attach"))
+       bpf_program__set_autoload(prog, true);
+
+	err = dynptr_success__load(skel);
+	if (!ASSERT_OK(err, "dynptr_success__load"))
 		goto cleanup;
 
-	usleep(1);
+	switch (setup_type) {
+	case SETUP_SYSCALL_SLEEP:
+		link = bpf_program__attach(prog);
+		if (!ASSERT_OK_PTR(link, "bpf_program__attach"))
+			goto cleanup;
 
-	ASSERT_EQ(skel->bss->err, 0, "err");
+		usleep(1);
+
+		bpf_link__destroy(link);
+		break;
+	case SETUP_SKB_PROG:
+	{
+		int prog_fd;
+		char buf[64];
+
+		LIBBPF_OPTS(bpf_test_run_opts, topts,
+			    .data_in = &pkt_v4,
+			    .data_size_in = sizeof(pkt_v4),
+			    .data_out = buf,
+			    .data_size_out = sizeof(buf),
+			    .repeat = 1,
+		);
 
-	bpf_link__destroy(link);
+		prog_fd = bpf_program__fd(prog);
+		if (!ASSERT_GE(prog_fd, 0, "prog_fd"))
+			goto cleanup;
+
+		err = bpf_prog_test_run_opts(prog_fd, &topts);
+
+		if (!ASSERT_OK(err, "test_run"))
+			goto cleanup;
+
+		break;
+	}
+	}
+
+	ASSERT_EQ(skel->bss->err, 0, "err");
 
 cleanup:
 	dynptr_success__destroy(skel);
@@ -50,10 +92,10 @@ void test_dynptr(void)
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(success_tests); i++) {
-		if (!test__start_subtest(success_tests[i]))
+		if (!test__start_subtest(success_tests[i].prog_name))
 			continue;
 
-		verify_success(success_tests[i]);
+		verify_success(success_tests[i].prog_name, success_tests[i].type);
 	}
 
 	RUN_TESTS(dynptr_fail);
diff --git a/tools/testing/selftests/bpf/prog_tests/l4lb_all.c b/tools/testing/selftests/bpf/prog_tests/l4lb_all.c
index 9c1a18573ffd..1eab286b14fe 100644
--- a/tools/testing/selftests/bpf/prog_tests/l4lb_all.c
+++ b/tools/testing/selftests/bpf/prog_tests/l4lb_all.c
@@ -93,4 +93,6 @@ void test_l4lb_all(void)
 		test_l4lb("test_l4lb.bpf.o");
 	if (test__start_subtest("l4lb_noinline"))
 		test_l4lb("test_l4lb_noinline.bpf.o");
+	if (test__start_subtest("l4lb_noinline_dynptr"))
+		test_l4lb("test_l4lb_noinline_dynptr.bpf.o");
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/parse_tcp_hdr_opt.c b/tools/testing/selftests/bpf/prog_tests/parse_tcp_hdr_opt.c
new file mode 100644
index 000000000000..daa952711d8f
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/parse_tcp_hdr_opt.c
@@ -0,0 +1,93 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+#include <network_helpers.h>
+#include "test_parse_tcp_hdr_opt.skel.h"
+#include "test_parse_tcp_hdr_opt_dynptr.skel.h"
+#include "test_tcp_hdr_options.h"
+
+struct test_pkt {
+	struct ipv6_packet pk6_v6;
+	u8 options[16];
+} __packed;
+
+struct test_pkt pkt = {
+	.pk6_v6.eth.h_proto = __bpf_constant_htons(ETH_P_IPV6),
+	.pk6_v6.iph.nexthdr = IPPROTO_TCP,
+	.pk6_v6.iph.payload_len = __bpf_constant_htons(MAGIC_BYTES),
+	.pk6_v6.tcp.urg_ptr = 123,
+	.pk6_v6.tcp.doff = 9, /* 16 bytes of options */
+
+	.options = {
+		TCPOPT_MSS, 4, 0x05, 0xB4, TCPOPT_NOP, TCPOPT_NOP,
+		0, 6, 0xBB, 0xBB, 0xBB, 0xBB, TCPOPT_EOL
+	},
+};
+
+static void test_parse_opt(void)
+{
+	struct test_parse_tcp_hdr_opt *skel;
+	struct bpf_program *prog;
+	char buf[128];
+	int err;
+
+	LIBBPF_OPTS(bpf_test_run_opts, topts,
+		    .data_in = &pkt,
+		    .data_size_in = sizeof(pkt),
+		    .data_out = buf,
+		    .data_size_out = sizeof(buf),
+		    .repeat = 3,
+	);
+
+	skel = test_parse_tcp_hdr_opt__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel_open_and_load"))
+		return;
+
+	pkt.options[6] = skel->rodata->tcp_hdr_opt_kind_tpr;
+	prog = skel->progs.xdp_ingress_v6;
+
+	err = bpf_prog_test_run_opts(bpf_program__fd(prog), &topts);
+	ASSERT_OK(err, "ipv6 test_run");
+	ASSERT_EQ(topts.retval, XDP_PASS, "ipv6 test_run retval");
+	ASSERT_EQ(skel->bss->server_id, 0xBBBBBBBB, "server id");
+
+	test_parse_tcp_hdr_opt__destroy(skel);
+}
+
+static void test_parse_opt_dynptr(void)
+{
+	struct test_parse_tcp_hdr_opt_dynptr *skel;
+	struct bpf_program *prog;
+	char buf[128];
+	int err;
+
+	LIBBPF_OPTS(bpf_test_run_opts, topts,
+		    .data_in = &pkt,
+		    .data_size_in = sizeof(pkt),
+		    .data_out = buf,
+		    .data_size_out = sizeof(buf),
+		    .repeat = 3,
+	);
+
+	skel = test_parse_tcp_hdr_opt_dynptr__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel_open_and_load"))
+		return;
+
+	pkt.options[6] = skel->rodata->tcp_hdr_opt_kind_tpr;
+	prog = skel->progs.xdp_ingress_v6;
+
+	err = bpf_prog_test_run_opts(bpf_program__fd(prog), &topts);
+	ASSERT_OK(err, "ipv6 test_run");
+	ASSERT_EQ(topts.retval, XDP_PASS, "ipv6 test_run retval");
+	ASSERT_EQ(skel->bss->server_id, 0xBBBBBBBB, "server id");
+
+	test_parse_tcp_hdr_opt_dynptr__destroy(skel);
+}
+
+void test_parse_tcp_hdr_opt(void)
+{
+	if (test__start_subtest("parse_tcp_hdr_opt"))
+		test_parse_opt();
+	if (test__start_subtest("parse_tcp_hdr_opt_dynptr"))
+		test_parse_opt_dynptr();
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_attach.c
index d4cd9f873c14..fa3cac5488f5 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_attach.c
@@ -4,11 +4,10 @@
 #define IFINDEX_LO 1
 #define XDP_FLAGS_REPLACE		(1U << 4)
 
-void serial_test_xdp_attach(void)
+static void test_xdp_attach(const char *file)
 {
 	__u32 duration = 0, id1, id2, id0 = 0, len;
 	struct bpf_object *obj1, *obj2, *obj3;
-	const char *file = "./test_xdp.bpf.o";
 	struct bpf_prog_info info = {};
 	int err, fd1, fd2, fd3;
 	LIBBPF_OPTS(bpf_xdp_attach_opts, opts);
@@ -85,3 +84,11 @@ out_2:
 out_1:
 	bpf_object__close(obj1);
 }
+
+void serial_test_xdp_attach(void)
+{
+	if (test__start_subtest("xdp_attach"))
+		test_xdp_attach("./test_xdp.bpf.o");
+	if (test__start_subtest("xdp_attach_dynptr"))
+		test_xdp_attach("./test_xdp_dynptr.bpf.o");
+}
diff --git a/tools/testing/selftests/bpf/progs/dynptr_fail.c b/tools/testing/selftests/bpf/progs/dynptr_fail.c
index aa5b69354b91..20ce920d891d 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_fail.c
@@ -5,7 +5,9 @@
 #include <string.h>
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
+#include <linux/if_ether.h>
 #include "bpf_misc.h"
+#include "bpf_kfuncs.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -244,6 +246,27 @@ done:
 	return 0;
 }
 
+/* A data slice can't be accessed out of bounds */
+SEC("?tc")
+__failure __msg("value is outside of the allowed memory range")
+int data_slice_out_of_bounds_skb(struct __sk_buff *skb)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	/* this should fail */
+	*(__u8*)(hdr + 1) = 1;
+
+	return SK_PASS;
+}
+
 SEC("?raw_tp")
 __failure __msg("value is outside of the allowed memory range")
 int data_slice_out_of_bounds_map_value(void *ctx)
@@ -399,7 +422,6 @@ int invalid_helper2(void *ctx)
 
 	/* this should fail */
 	bpf_dynptr_read(read_data, sizeof(read_data), (void *)&ptr + 8, 0, 0);
-
 	return 0;
 }
 
@@ -1044,6 +1066,193 @@ int dynptr_read_into_slot(void *ctx)
 	return 0;
 }
 
+/* bpf_dynptr_slice()s are read-only and cannot be written to */
+SEC("?tc")
+__failure __msg("R0 cannot write into rdonly_mem")
+int skb_invalid_slice_write(struct __sk_buff *skb)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	hdr = bpf_dynptr_slice(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	/* this should fail */
+	hdr->h_proto = 1;
+
+	return SK_PASS;
+}
+
+/* The read-only data slice is invalidated whenever a helper changes packet data */
+SEC("?tc")
+__failure __msg("invalid mem access 'scalar'")
+int skb_invalid_data_slice1(struct __sk_buff *skb)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	hdr = bpf_dynptr_slice(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	val = hdr->h_proto;
+
+	if (bpf_skb_pull_data(skb, skb->len))
+		return SK_DROP;
+
+	/* this should fail */
+	val = hdr->h_proto;
+
+	return SK_PASS;
+}
+
+/* The read-write data slice is invalidated whenever a helper changes packet data */
+SEC("?tc")
+__failure __msg("invalid mem access 'scalar'")
+int skb_invalid_data_slice2(struct __sk_buff *skb)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	hdr->h_proto = 123;
+
+	if (bpf_skb_pull_data(skb, skb->len))
+		return SK_DROP;
+
+	/* this should fail */
+	hdr->h_proto = 1;
+
+	return SK_PASS;
+}
+
+/* The read-only data slice is invalidated whenever bpf_dynptr_write() is called */
+SEC("?tc")
+__failure __msg("invalid mem access 'scalar'")
+int skb_invalid_data_slice3(struct __sk_buff *skb)
+{
+	char write_data[64] = "hello there, world!!";
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	hdr = bpf_dynptr_slice(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	val = hdr->h_proto;
+
+	bpf_dynptr_write(&ptr, 0, write_data, sizeof(write_data), 0);
+
+	/* this should fail */
+	val = hdr->h_proto;
+
+	return SK_PASS;
+}
+
+/* The read-write data slice is invalidated whenever bpf_dynptr_write() is called */
+SEC("?tc")
+__failure __msg("invalid mem access 'scalar'")
+int skb_invalid_data_slice4(struct __sk_buff *skb)
+{
+	char write_data[64] = "hello there, world!!";
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+	hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	hdr->h_proto = 123;
+
+	bpf_dynptr_write(&ptr, 0, write_data, sizeof(write_data), 0);
+
+	/* this should fail */
+	hdr->h_proto = 1;
+
+	return SK_PASS;
+}
+
+/* The read-only data slice is invalidated whenever a helper changes packet data */
+SEC("?xdp")
+__failure __msg("invalid mem access 'scalar'")
+int xdp_invalid_data_slice1(struct xdp_md *xdp)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_xdp(xdp, 0, &ptr);
+	hdr = bpf_dynptr_slice(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	val = hdr->h_proto;
+
+	if (bpf_xdp_adjust_head(xdp, 0 - (int)sizeof(*hdr)))
+		return XDP_DROP;
+
+	/* this should fail */
+	val = hdr->h_proto;
+
+	return XDP_PASS;
+}
+
+/* The read-write data slice is invalidated whenever a helper changes packet data */
+SEC("?xdp")
+__failure __msg("invalid mem access 'scalar'")
+int xdp_invalid_data_slice2(struct xdp_md *xdp)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_xdp(xdp, 0, &ptr);
+	hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+	if (!hdr)
+		return SK_DROP;
+
+	hdr->h_proto = 9;
+
+	if (bpf_xdp_adjust_head(xdp, 0 - (int)sizeof(*hdr)))
+		return XDP_DROP;
+
+	/* this should fail */
+	hdr->h_proto = 1;
+
+	return XDP_PASS;
+}
+
+/* Only supported prog type can create skb-type dynptrs */
+SEC("?raw_tp")
+__failure __msg("calling kernel function bpf_dynptr_from_skb is not allowed")
+int skb_invalid_ctx(void *ctx)
+{
+	struct bpf_dynptr ptr;
+
+	/* this should fail */
+	bpf_dynptr_from_skb(ctx, 0, &ptr);
+
+	return 0;
+}
+
 /* Reject writes to dynptr slot for uninit arg */
 SEC("?raw_tp")
 __failure __msg("potential write to dynptr at off=-16")
@@ -1061,6 +1270,61 @@ int uninit_write_into_slot(void *ctx)
 	return 0;
 }
 
+/* Only supported prog type can create xdp-type dynptrs */
+SEC("?raw_tp")
+__failure __msg("calling kernel function bpf_dynptr_from_xdp is not allowed")
+int xdp_invalid_ctx(void *ctx)
+{
+	struct bpf_dynptr ptr;
+
+	/* this should fail */
+	bpf_dynptr_from_xdp(ctx, 0, &ptr);
+
+	return 0;
+}
+
+__u32 hdr_size = sizeof(struct ethhdr);
+/* Can't pass in variable-sized len to bpf_dynptr_slice */
+SEC("?tc")
+__failure __msg("unbounded memory access")
+int dynptr_slice_var_len1(struct __sk_buff *skb)
+{
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+	char buffer[sizeof(*hdr)] = {};
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	/* this should fail */
+	hdr = bpf_dynptr_slice(&ptr, 0, buffer, hdr_size);
+	if (!hdr)
+		return SK_DROP;
+
+	return SK_PASS;
+}
+
+/* Can't pass in variable-sized len to bpf_dynptr_slice */
+SEC("?tc")
+__failure __msg("must be a known constant")
+int dynptr_slice_var_len2(struct __sk_buff *skb)
+{
+	char buffer[sizeof(struct ethhdr)] = {};
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	if (hdr_size <= sizeof(buffer)) {
+		/* this should fail */
+		hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, hdr_size);
+		if (!hdr)
+			return SK_DROP;
+		hdr->h_proto = 12;
+	}
+
+	return SK_PASS;
+}
+
 static int callback(__u32 index, void *data)
 {
         *(__u32 *)data = 123;
@@ -1092,3 +1356,24 @@ int invalid_data_slices(void *ctx)
 
 	return 0;
 }
+
+/* Program types that don't allow writes to packet data should fail if
+ * bpf_dynptr_slice_rdwr is called
+ */
+SEC("cgroup_skb/ingress")
+__failure __msg("the prog does not allow writes to packet data")
+int invalid_slice_rdwr_rdonly(struct __sk_buff *skb)
+{
+	char buffer[sizeof(struct ethhdr)] = {};
+	struct bpf_dynptr ptr;
+	struct ethhdr *hdr;
+
+	bpf_dynptr_from_skb(skb, 0, &ptr);
+
+	/* this should fail since cgroup_skb doesn't allow
+	 * changing packet data
+	 */
+	hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/dynptr_success.c b/tools/testing/selftests/bpf/progs/dynptr_success.c
index 35db7c6c1fc7..c8358a7c7924 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_success.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_success.c
@@ -5,6 +5,7 @@
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
 #include "bpf_misc.h"
+#include "bpf_kfuncs.h"
 #include "errno.h"
 
 char _license[] SEC("license") = "GPL";
@@ -30,7 +31,7 @@ struct {
 	__type(value, __u32);
 } array_map SEC(".maps");
 
-SEC("tp/syscalls/sys_enter_nanosleep")
+SEC("?tp/syscalls/sys_enter_nanosleep")
 int test_read_write(void *ctx)
 {
 	char write_data[64] = "hello there, world!!";
@@ -61,8 +62,8 @@ int test_read_write(void *ctx)
 	return 0;
 }
 
-SEC("tp/syscalls/sys_enter_nanosleep")
-int test_data_slice(void *ctx)
+SEC("?tp/syscalls/sys_enter_nanosleep")
+int test_dynptr_data(void *ctx)
 {
 	__u32 key = 0, val = 235, *map_val;
 	struct bpf_dynptr ptr;
@@ -131,7 +132,7 @@ static int ringbuf_callback(__u32 index, void *data)
 	return 0;
 }
 
-SEC("tp/syscalls/sys_enter_nanosleep")
+SEC("?tp/syscalls/sys_enter_nanosleep")
 int test_ringbuf(void *ctx)
 {
 	struct bpf_dynptr ptr;
@@ -163,3 +164,49 @@ done:
 	bpf_ringbuf_discard_dynptr(&ptr, 0);
 	return 0;
 }
+
+SEC("?cgroup_skb/egress")
+int test_skb_readonly(struct __sk_buff *skb)
+{
+	__u8 write_data[2] = {1, 2};
+	struct bpf_dynptr ptr;
+	__u64 *data;
+	int ret;
+
+	if (bpf_dynptr_from_skb(skb, 0, &ptr)) {
+		err = 1;
+		return 1;
+	}
+
+	/* since cgroup skbs are read only, writes should fail */
+	ret = bpf_dynptr_write(&ptr, 0, write_data, sizeof(write_data), 0);
+	if (ret != -EINVAL) {
+		err = 2;
+		return 1;
+	}
+
+	return 1;
+}
+
+SEC("?cgroup_skb/egress")
+int test_dynptr_skb_data(struct __sk_buff *skb)
+{
+	__u8 write_data[2] = {1, 2};
+	struct bpf_dynptr ptr;
+	__u64 *data;
+	int ret;
+
+	if (bpf_dynptr_from_skb(skb, 0, &ptr)) {
+		err = 1;
+		return 1;
+	}
+
+	/* This should return NULL. Must use bpf_dynptr_slice API */
+	data = bpf_dynptr_data(&ptr, 0, 1);
+	if (data) {
+		err = 2;
+		return 1;
+	}
+
+	return 1;
+}
diff --git a/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c b/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c
new file mode 100644
index 000000000000..f45a7095de7a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c
@@ -0,0 +1,980 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+// Copyright (c) 2019, 2020 Cloudflare
+
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <string.h>
+
+#include <linux/bpf.h>
+#include <linux/icmp.h>
+#include <linux/icmpv6.h>
+#include <linux/if_ether.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/pkt_cls.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
+
+#include "test_cls_redirect.h"
+#include "bpf_kfuncs.h"
+
+#define offsetofend(TYPE, MEMBER) \
+	(offsetof(TYPE, MEMBER) + sizeof((((TYPE *)0)->MEMBER)))
+
+#define IP_OFFSET_MASK (0x1FFF)
+#define IP_MF (0x2000)
+
+char _license[] SEC("license") = "Dual BSD/GPL";
+
+/**
+ * Destination port and IP used for UDP encapsulation.
+ */
+volatile const __be16 ENCAPSULATION_PORT;
+volatile const __be32 ENCAPSULATION_IP;
+
+typedef struct {
+	uint64_t processed_packets_total;
+	uint64_t l3_protocol_packets_total_ipv4;
+	uint64_t l3_protocol_packets_total_ipv6;
+	uint64_t l4_protocol_packets_total_tcp;
+	uint64_t l4_protocol_packets_total_udp;
+	uint64_t accepted_packets_total_syn;
+	uint64_t accepted_packets_total_syn_cookies;
+	uint64_t accepted_packets_total_last_hop;
+	uint64_t accepted_packets_total_icmp_echo_request;
+	uint64_t accepted_packets_total_established;
+	uint64_t forwarded_packets_total_gue;
+	uint64_t forwarded_packets_total_gre;
+
+	uint64_t errors_total_unknown_l3_proto;
+	uint64_t errors_total_unknown_l4_proto;
+	uint64_t errors_total_malformed_ip;
+	uint64_t errors_total_fragmented_ip;
+	uint64_t errors_total_malformed_icmp;
+	uint64_t errors_total_unwanted_icmp;
+	uint64_t errors_total_malformed_icmp_pkt_too_big;
+	uint64_t errors_total_malformed_tcp;
+	uint64_t errors_total_malformed_udp;
+	uint64_t errors_total_icmp_echo_replies;
+	uint64_t errors_total_malformed_encapsulation;
+	uint64_t errors_total_encap_adjust_failed;
+	uint64_t errors_total_encap_buffer_too_small;
+	uint64_t errors_total_redirect_loop;
+	uint64_t errors_total_encap_mtu_violate;
+} metrics_t;
+
+typedef enum {
+	INVALID = 0,
+	UNKNOWN,
+	ECHO_REQUEST,
+	SYN,
+	SYN_COOKIE,
+	ESTABLISHED,
+} verdict_t;
+
+typedef struct {
+	uint16_t src, dst;
+} flow_ports_t;
+
+_Static_assert(
+	sizeof(flow_ports_t) !=
+		offsetofend(struct bpf_sock_tuple, ipv4.dport) -
+			offsetof(struct bpf_sock_tuple, ipv4.sport) - 1,
+	"flow_ports_t must match sport and dport in struct bpf_sock_tuple");
+_Static_assert(
+	sizeof(flow_ports_t) !=
+		offsetofend(struct bpf_sock_tuple, ipv6.dport) -
+			offsetof(struct bpf_sock_tuple, ipv6.sport) - 1,
+	"flow_ports_t must match sport and dport in struct bpf_sock_tuple");
+
+struct iphdr_info {
+	void *hdr;
+	__u64 len;
+};
+
+typedef int ret_t;
+
+/* This is a bit of a hack. We need a return value which allows us to
+ * indicate that the regular flow of the program should continue,
+ * while allowing functions to use XDP_PASS and XDP_DROP, etc.
+ */
+static const ret_t CONTINUE_PROCESSING = -1;
+
+/* Convenience macro to call functions which return ret_t.
+ */
+#define MAYBE_RETURN(x)                           \
+	do {                                      \
+		ret_t __ret = x;                  \
+		if (__ret != CONTINUE_PROCESSING) \
+			return __ret;             \
+	} while (0)
+
+static bool ipv4_is_fragment(const struct iphdr *ip)
+{
+	uint16_t frag_off = ip->frag_off & bpf_htons(IP_OFFSET_MASK);
+	return (ip->frag_off & bpf_htons(IP_MF)) != 0 || frag_off > 0;
+}
+
+static int pkt_parse_ipv4(struct bpf_dynptr *dynptr, __u64 *offset, struct iphdr *iphdr)
+{
+	if (bpf_dynptr_read(iphdr, sizeof(*iphdr), dynptr, *offset, 0))
+		return -1;
+
+	*offset += sizeof(*iphdr);
+
+	if (iphdr->ihl < 5)
+		return -1;
+
+	/* skip ipv4 options */
+	*offset += (iphdr->ihl - 5) * 4;
+
+	return 0;
+}
+
+/* Parse the L4 ports from a packet, assuming a layout like TCP or UDP. */
+static bool pkt_parse_icmp_l4_ports(struct bpf_dynptr *dynptr, __u64 *offset, flow_ports_t *ports)
+{
+	if (bpf_dynptr_read(ports, sizeof(*ports), dynptr, *offset, 0))
+		return false;
+
+	*offset += sizeof(*ports);
+
+	/* Ports in the L4 headers are reversed, since we are parsing an ICMP
+	 * payload which is going towards the eyeball.
+	 */
+	uint16_t dst = ports->src;
+	ports->src = ports->dst;
+	ports->dst = dst;
+	return true;
+}
+
+static uint16_t pkt_checksum_fold(uint32_t csum)
+{
+	/* The highest reasonable value for an IPv4 header
+	 * checksum requires two folds, so we just do that always.
+	 */
+	csum = (csum & 0xffff) + (csum >> 16);
+	csum = (csum & 0xffff) + (csum >> 16);
+	return (uint16_t)~csum;
+}
+
+static void pkt_ipv4_checksum(struct iphdr *iph)
+{
+	iph->check = 0;
+
+	/* An IP header without options is 20 bytes. Two of those
+	 * are the checksum, which we always set to zero. Hence,
+	 * the maximum accumulated value is 18 / 2 * 0xffff = 0x8fff7,
+	 * which fits in 32 bit.
+	 */
+	_Static_assert(sizeof(struct iphdr) == 20, "iphdr must be 20 bytes");
+	uint32_t acc = 0;
+	uint16_t *ipw = (uint16_t *)iph;
+
+	for (size_t i = 0; i < sizeof(struct iphdr) / 2; i++)
+		acc += ipw[i];
+
+	iph->check = pkt_checksum_fold(acc);
+}
+
+static bool pkt_skip_ipv6_extension_headers(struct bpf_dynptr *dynptr, __u64 *offset,
+					    const struct ipv6hdr *ipv6, uint8_t *upper_proto,
+					    bool *is_fragment)
+{
+	/* We understand five extension headers.
+	 * https://tools.ietf.org/html/rfc8200#section-4.1 states that all
+	 * headers should occur once, except Destination Options, which may
+	 * occur twice. Hence we give up after 6 headers.
+	 */
+	struct {
+		uint8_t next;
+		uint8_t len;
+	} exthdr = {
+		.next = ipv6->nexthdr,
+	};
+	*is_fragment = false;
+
+	for (int i = 0; i < 6; i++) {
+		switch (exthdr.next) {
+		case IPPROTO_FRAGMENT:
+			*is_fragment = true;
+			/* NB: We don't check that hdrlen == 0 as per spec. */
+			/* fallthrough; */
+
+		case IPPROTO_HOPOPTS:
+		case IPPROTO_ROUTING:
+		case IPPROTO_DSTOPTS:
+		case IPPROTO_MH:
+			if (bpf_dynptr_read(&exthdr, sizeof(exthdr), dynptr, *offset, 0))
+				return false;
+
+			/* hdrlen is in 8-octet units, and excludes the first 8 octets. */
+			*offset += (exthdr.len + 1) * 8;
+
+			/* Decode next header */
+			break;
+
+		default:
+			/* The next header is not one of the known extension
+			 * headers, treat it as the upper layer header.
+			 *
+			 * This handles IPPROTO_NONE.
+			 *
+			 * Encapsulating Security Payload (50) and Authentication
+			 * Header (51) also end up here (and will trigger an
+			 * unknown proto error later). They have a custom header
+			 * format and seem too esoteric to care about.
+			 */
+			*upper_proto = exthdr.next;
+			return true;
+		}
+	}
+
+	/* We never found an upper layer header. */
+	return false;
+}
+
+static int pkt_parse_ipv6(struct bpf_dynptr *dynptr, __u64 *offset, struct ipv6hdr *ipv6,
+			  uint8_t *proto, bool *is_fragment)
+{
+	if (bpf_dynptr_read(ipv6, sizeof(*ipv6), dynptr, *offset, 0))
+		return -1;
+
+	*offset += sizeof(*ipv6);
+
+	if (!pkt_skip_ipv6_extension_headers(dynptr, offset, ipv6, proto, is_fragment))
+		return -1;
+
+	return 0;
+}
+
+/* Global metrics, per CPU
+ */
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, unsigned int);
+	__type(value, metrics_t);
+} metrics_map SEC(".maps");
+
+static metrics_t *get_global_metrics(void)
+{
+	uint64_t key = 0;
+	return bpf_map_lookup_elem(&metrics_map, &key);
+}
+
+static ret_t accept_locally(struct __sk_buff *skb, encap_headers_t *encap)
+{
+	const int payload_off =
+		sizeof(*encap) +
+		sizeof(struct in_addr) * encap->unigue.hop_count;
+	int32_t encap_overhead = payload_off - sizeof(struct ethhdr);
+
+	/* Changing the ethertype if the encapsulated packet is ipv6 */
+	if (encap->gue.proto_ctype == IPPROTO_IPV6)
+		encap->eth.h_proto = bpf_htons(ETH_P_IPV6);
+
+	if (bpf_skb_adjust_room(skb, -encap_overhead, BPF_ADJ_ROOM_MAC,
+				BPF_F_ADJ_ROOM_FIXED_GSO |
+				BPF_F_ADJ_ROOM_NO_CSUM_RESET) ||
+	    bpf_csum_level(skb, BPF_CSUM_LEVEL_DEC))
+		return TC_ACT_SHOT;
+
+	return bpf_redirect(skb->ifindex, BPF_F_INGRESS);
+}
+
+static ret_t forward_with_gre(struct __sk_buff *skb, struct bpf_dynptr *dynptr,
+			      encap_headers_t *encap, struct in_addr *next_hop,
+			      metrics_t *metrics)
+{
+	const int payload_off =
+		sizeof(*encap) +
+		sizeof(struct in_addr) * encap->unigue.hop_count;
+	int32_t encap_overhead =
+		payload_off - sizeof(struct ethhdr) - sizeof(struct iphdr);
+	int32_t delta = sizeof(struct gre_base_hdr) - encap_overhead;
+	__u8 encap_buffer[sizeof(encap_gre_t)] = {};
+	uint16_t proto = ETH_P_IP;
+	uint32_t mtu_len = 0;
+	encap_gre_t *encap_gre;
+
+	metrics->forwarded_packets_total_gre++;
+
+	/* Loop protection: the inner packet's TTL is decremented as a safeguard
+	 * against any forwarding loop. As the only interesting field is the TTL
+	 * hop limit for IPv6, it is easier to use bpf_skb_load_bytes/bpf_skb_store_bytes
+	 * as they handle the split packets if needed (no need for the data to be
+	 * in the linear section).
+	 */
+	if (encap->gue.proto_ctype == IPPROTO_IPV6) {
+		proto = ETH_P_IPV6;
+		uint8_t ttl;
+		int rc;
+
+		rc = bpf_skb_load_bytes(
+			skb, payload_off + offsetof(struct ipv6hdr, hop_limit),
+			&ttl, 1);
+		if (rc != 0) {
+			metrics->errors_total_malformed_encapsulation++;
+			return TC_ACT_SHOT;
+		}
+
+		if (ttl == 0) {
+			metrics->errors_total_redirect_loop++;
+			return TC_ACT_SHOT;
+		}
+
+		ttl--;
+		rc = bpf_skb_store_bytes(
+			skb, payload_off + offsetof(struct ipv6hdr, hop_limit),
+			&ttl, 1, 0);
+		if (rc != 0) {
+			metrics->errors_total_malformed_encapsulation++;
+			return TC_ACT_SHOT;
+		}
+	} else {
+		uint8_t ttl;
+		int rc;
+
+		rc = bpf_skb_load_bytes(
+			skb, payload_off + offsetof(struct iphdr, ttl), &ttl,
+			1);
+		if (rc != 0) {
+			metrics->errors_total_malformed_encapsulation++;
+			return TC_ACT_SHOT;
+		}
+
+		if (ttl == 0) {
+			metrics->errors_total_redirect_loop++;
+			return TC_ACT_SHOT;
+		}
+
+		/* IPv4 also has a checksum to patch. While the TTL is only one byte,
+		 * this function only works for 2 and 4 bytes arguments (the result is
+		 * the same).
+		 */
+		rc = bpf_l3_csum_replace(
+			skb, payload_off + offsetof(struct iphdr, check), ttl,
+			ttl - 1, 2);
+		if (rc != 0) {
+			metrics->errors_total_malformed_encapsulation++;
+			return TC_ACT_SHOT;
+		}
+
+		ttl--;
+		rc = bpf_skb_store_bytes(
+			skb, payload_off + offsetof(struct iphdr, ttl), &ttl, 1,
+			0);
+		if (rc != 0) {
+			metrics->errors_total_malformed_encapsulation++;
+			return TC_ACT_SHOT;
+		}
+	}
+
+	if (bpf_check_mtu(skb, skb->ifindex, &mtu_len, delta, 0)) {
+		metrics->errors_total_encap_mtu_violate++;
+		return TC_ACT_SHOT;
+	}
+
+	if (bpf_skb_adjust_room(skb, delta, BPF_ADJ_ROOM_NET,
+				BPF_F_ADJ_ROOM_FIXED_GSO |
+				BPF_F_ADJ_ROOM_NO_CSUM_RESET) ||
+	    bpf_csum_level(skb, BPF_CSUM_LEVEL_INC)) {
+		metrics->errors_total_encap_adjust_failed++;
+		return TC_ACT_SHOT;
+	}
+
+	if (bpf_skb_pull_data(skb, sizeof(encap_gre_t))) {
+		metrics->errors_total_encap_buffer_too_small++;
+		return TC_ACT_SHOT;
+	}
+
+	encap_gre = bpf_dynptr_slice_rdwr(dynptr, 0, encap_buffer, sizeof(encap_buffer));
+	if (!encap_gre) {
+		metrics->errors_total_encap_buffer_too_small++;
+		return TC_ACT_SHOT;
+	}
+
+	encap_gre->ip.protocol = IPPROTO_GRE;
+	encap_gre->ip.daddr = next_hop->s_addr;
+	encap_gre->ip.saddr = ENCAPSULATION_IP;
+	encap_gre->ip.tot_len =
+		bpf_htons(bpf_ntohs(encap_gre->ip.tot_len) + delta);
+	encap_gre->gre.flags = 0;
+	encap_gre->gre.protocol = bpf_htons(proto);
+	pkt_ipv4_checksum((void *)&encap_gre->ip);
+
+	if (encap_gre == encap_buffer)
+		bpf_dynptr_write(dynptr, 0, encap_buffer, sizeof(encap_buffer), 0);
+
+	return bpf_redirect(skb->ifindex, 0);
+}
+
+static ret_t forward_to_next_hop(struct __sk_buff *skb, struct bpf_dynptr *dynptr,
+				 encap_headers_t *encap, struct in_addr *next_hop,
+				 metrics_t *metrics)
+{
+	/* swap L2 addresses */
+	/* This assumes that packets are received from a router.
+	 * So just swapping the MAC addresses here will make the packet go back to
+	 * the router, which will send it to the appropriate machine.
+	 */
+	unsigned char temp[ETH_ALEN];
+	memcpy(temp, encap->eth.h_dest, sizeof(temp));
+	memcpy(encap->eth.h_dest, encap->eth.h_source,
+	       sizeof(encap->eth.h_dest));
+	memcpy(encap->eth.h_source, temp, sizeof(encap->eth.h_source));
+
+	if (encap->unigue.next_hop == encap->unigue.hop_count - 1 &&
+	    encap->unigue.last_hop_gre) {
+		return forward_with_gre(skb, dynptr, encap, next_hop, metrics);
+	}
+
+	metrics->forwarded_packets_total_gue++;
+	uint32_t old_saddr = encap->ip.saddr;
+	encap->ip.saddr = encap->ip.daddr;
+	encap->ip.daddr = next_hop->s_addr;
+	if (encap->unigue.next_hop < encap->unigue.hop_count) {
+		encap->unigue.next_hop++;
+	}
+
+	/* Remove ip->saddr, add next_hop->s_addr */
+	const uint64_t off = offsetof(typeof(*encap), ip.check);
+	int ret = bpf_l3_csum_replace(skb, off, old_saddr, next_hop->s_addr, 4);
+	if (ret < 0) {
+		return TC_ACT_SHOT;
+	}
+
+	return bpf_redirect(skb->ifindex, 0);
+}
+
+static ret_t skip_next_hops(__u64 *offset, int n)
+{
+	__u32 res;
+	switch (n) {
+	case 1:
+		*offset += sizeof(struct in_addr);
+	case 0:
+		return CONTINUE_PROCESSING;
+
+	default:
+		return TC_ACT_SHOT;
+	}
+}
+
+/* Get the next hop from the GLB header.
+ *
+ * Sets next_hop->s_addr to 0 if there are no more hops left.
+ * pkt is positioned just after the variable length GLB header
+ * iff the call is successful.
+ */
+static ret_t get_next_hop(struct bpf_dynptr *dynptr, __u64 *offset, encap_headers_t *encap,
+			  struct in_addr *next_hop)
+{
+	if (encap->unigue.next_hop > encap->unigue.hop_count)
+		return TC_ACT_SHOT;
+
+	/* Skip "used" next hops. */
+	MAYBE_RETURN(skip_next_hops(offset, encap->unigue.next_hop));
+
+	if (encap->unigue.next_hop == encap->unigue.hop_count) {
+		/* No more next hops, we are at the end of the GLB header. */
+		next_hop->s_addr = 0;
+		return CONTINUE_PROCESSING;
+	}
+
+	if (bpf_dynptr_read(next_hop, sizeof(*next_hop), dynptr, *offset, 0))
+		return TC_ACT_SHOT;
+
+	*offset += sizeof(*next_hop);
+
+	/* Skip the remainig next hops (may be zero). */
+	return skip_next_hops(offset, encap->unigue.hop_count - encap->unigue.next_hop - 1);
+}
+
+/* Fill a bpf_sock_tuple to be used with the socket lookup functions.
+ * This is a kludge that let's us work around verifier limitations:
+ *
+ *    fill_tuple(&t, foo, sizeof(struct iphdr), 123, 321)
+ *
+ * clang will substitue a costant for sizeof, which allows the verifier
+ * to track it's value. Based on this, it can figure out the constant
+ * return value, and calling code works while still being "generic" to
+ * IPv4 and IPv6.
+ */
+static uint64_t fill_tuple(struct bpf_sock_tuple *tuple, void *iph,
+				    uint64_t iphlen, uint16_t sport, uint16_t dport)
+{
+	switch (iphlen) {
+	case sizeof(struct iphdr): {
+		struct iphdr *ipv4 = (struct iphdr *)iph;
+		tuple->ipv4.daddr = ipv4->daddr;
+		tuple->ipv4.saddr = ipv4->saddr;
+		tuple->ipv4.sport = sport;
+		tuple->ipv4.dport = dport;
+		return sizeof(tuple->ipv4);
+	}
+
+	case sizeof(struct ipv6hdr): {
+		struct ipv6hdr *ipv6 = (struct ipv6hdr *)iph;
+		memcpy(&tuple->ipv6.daddr, &ipv6->daddr,
+		       sizeof(tuple->ipv6.daddr));
+		memcpy(&tuple->ipv6.saddr, &ipv6->saddr,
+		       sizeof(tuple->ipv6.saddr));
+		tuple->ipv6.sport = sport;
+		tuple->ipv6.dport = dport;
+		return sizeof(tuple->ipv6);
+	}
+
+	default:
+		return 0;
+	}
+}
+
+static verdict_t classify_tcp(struct __sk_buff *skb, struct bpf_sock_tuple *tuple,
+			      uint64_t tuplen, void *iph, struct tcphdr *tcp)
+{
+	struct bpf_sock *sk =
+		bpf_skc_lookup_tcp(skb, tuple, tuplen, BPF_F_CURRENT_NETNS, 0);
+
+	if (sk == NULL)
+		return UNKNOWN;
+
+	if (sk->state != BPF_TCP_LISTEN) {
+		bpf_sk_release(sk);
+		return ESTABLISHED;
+	}
+
+	if (iph != NULL && tcp != NULL) {
+		/* Kludge: we've run out of arguments, but need the length of the ip header. */
+		uint64_t iphlen = sizeof(struct iphdr);
+
+		if (tuplen == sizeof(tuple->ipv6))
+			iphlen = sizeof(struct ipv6hdr);
+
+		if (bpf_tcp_check_syncookie(sk, iph, iphlen, tcp,
+					    sizeof(*tcp)) == 0) {
+			bpf_sk_release(sk);
+			return SYN_COOKIE;
+		}
+	}
+
+	bpf_sk_release(sk);
+	return UNKNOWN;
+}
+
+static verdict_t classify_udp(struct __sk_buff *skb, struct bpf_sock_tuple *tuple, uint64_t tuplen)
+{
+	struct bpf_sock *sk =
+		bpf_sk_lookup_udp(skb, tuple, tuplen, BPF_F_CURRENT_NETNS, 0);
+
+	if (sk == NULL)
+		return UNKNOWN;
+
+	if (sk->state == BPF_TCP_ESTABLISHED) {
+		bpf_sk_release(sk);
+		return ESTABLISHED;
+	}
+
+	bpf_sk_release(sk);
+	return UNKNOWN;
+}
+
+static verdict_t classify_icmp(struct __sk_buff *skb, uint8_t proto, struct bpf_sock_tuple *tuple,
+			       uint64_t tuplen, metrics_t *metrics)
+{
+	switch (proto) {
+	case IPPROTO_TCP:
+		return classify_tcp(skb, tuple, tuplen, NULL, NULL);
+
+	case IPPROTO_UDP:
+		return classify_udp(skb, tuple, tuplen);
+
+	default:
+		metrics->errors_total_malformed_icmp++;
+		return INVALID;
+	}
+}
+
+static verdict_t process_icmpv4(struct __sk_buff *skb, struct bpf_dynptr *dynptr, __u64 *offset,
+				metrics_t *metrics)
+{
+	struct icmphdr icmp;
+	struct iphdr ipv4;
+
+	if (bpf_dynptr_read(&icmp, sizeof(icmp), dynptr, *offset, 0)) {
+		metrics->errors_total_malformed_icmp++;
+		return INVALID;
+	}
+
+	*offset += sizeof(icmp);
+
+	/* We should never receive encapsulated echo replies. */
+	if (icmp.type == ICMP_ECHOREPLY) {
+		metrics->errors_total_icmp_echo_replies++;
+		return INVALID;
+	}
+
+	if (icmp.type == ICMP_ECHO)
+		return ECHO_REQUEST;
+
+	if (icmp.type != ICMP_DEST_UNREACH || icmp.code != ICMP_FRAG_NEEDED) {
+		metrics->errors_total_unwanted_icmp++;
+		return INVALID;
+	}
+
+	if (pkt_parse_ipv4(dynptr, offset, &ipv4)) {
+		metrics->errors_total_malformed_icmp_pkt_too_big++;
+		return INVALID;
+	}
+
+	/* The source address in the outer IP header is from the entity that
+	 * originated the ICMP message. Use the original IP header to restore
+	 * the correct flow tuple.
+	 */
+	struct bpf_sock_tuple tuple;
+	tuple.ipv4.saddr = ipv4.daddr;
+	tuple.ipv4.daddr = ipv4.saddr;
+
+	if (!pkt_parse_icmp_l4_ports(dynptr, offset, (flow_ports_t *)&tuple.ipv4.sport)) {
+		metrics->errors_total_malformed_icmp_pkt_too_big++;
+		return INVALID;
+	}
+
+	return classify_icmp(skb, ipv4.protocol, &tuple,
+			     sizeof(tuple.ipv4), metrics);
+}
+
+static verdict_t process_icmpv6(struct bpf_dynptr *dynptr, __u64 *offset, struct __sk_buff *skb,
+				metrics_t *metrics)
+{
+	struct bpf_sock_tuple tuple;
+	struct ipv6hdr ipv6;
+	struct icmp6hdr icmp6;
+	bool is_fragment;
+	uint8_t l4_proto;
+
+	if (bpf_dynptr_read(&icmp6, sizeof(icmp6), dynptr, *offset, 0)) {
+		metrics->errors_total_malformed_icmp++;
+		return INVALID;
+	}
+
+	/* We should never receive encapsulated echo replies. */
+	if (icmp6.icmp6_type == ICMPV6_ECHO_REPLY) {
+		metrics->errors_total_icmp_echo_replies++;
+		return INVALID;
+	}
+
+	if (icmp6.icmp6_type == ICMPV6_ECHO_REQUEST) {
+		return ECHO_REQUEST;
+	}
+
+	if (icmp6.icmp6_type != ICMPV6_PKT_TOOBIG) {
+		metrics->errors_total_unwanted_icmp++;
+		return INVALID;
+	}
+
+	if (pkt_parse_ipv6(dynptr, offset, &ipv6, &l4_proto, &is_fragment)) {
+		metrics->errors_total_malformed_icmp_pkt_too_big++;
+		return INVALID;
+	}
+
+	if (is_fragment) {
+		metrics->errors_total_fragmented_ip++;
+		return INVALID;
+	}
+
+	/* Swap source and dest addresses. */
+	memcpy(&tuple.ipv6.saddr, &ipv6.daddr, sizeof(tuple.ipv6.saddr));
+	memcpy(&tuple.ipv6.daddr, &ipv6.saddr, sizeof(tuple.ipv6.daddr));
+
+	if (!pkt_parse_icmp_l4_ports(dynptr, offset, (flow_ports_t *)&tuple.ipv6.sport)) {
+		metrics->errors_total_malformed_icmp_pkt_too_big++;
+		return INVALID;
+	}
+
+	return classify_icmp(skb, l4_proto, &tuple, sizeof(tuple.ipv6),
+			     metrics);
+}
+
+static verdict_t process_tcp(struct bpf_dynptr *dynptr, __u64 *offset, struct __sk_buff *skb,
+			     struct iphdr_info *info, metrics_t *metrics)
+{
+	struct bpf_sock_tuple tuple;
+	struct tcphdr tcp;
+	uint64_t tuplen;
+
+	metrics->l4_protocol_packets_total_tcp++;
+
+	if (bpf_dynptr_read(&tcp, sizeof(tcp), dynptr, *offset, 0)) {
+		metrics->errors_total_malformed_tcp++;
+		return INVALID;
+	}
+
+	*offset += sizeof(tcp);
+
+	if (tcp.syn)
+		return SYN;
+
+	tuplen = fill_tuple(&tuple, info->hdr, info->len, tcp.source, tcp.dest);
+	return classify_tcp(skb, &tuple, tuplen, info->hdr, &tcp);
+}
+
+static verdict_t process_udp(struct bpf_dynptr *dynptr, __u64 *offset, struct __sk_buff *skb,
+			     struct iphdr_info *info, metrics_t *metrics)
+{
+	struct bpf_sock_tuple tuple;
+	struct udphdr udph;
+	uint64_t tuplen;
+
+	metrics->l4_protocol_packets_total_udp++;
+
+	if (bpf_dynptr_read(&udph, sizeof(udph), dynptr, *offset, 0)) {
+		metrics->errors_total_malformed_udp++;
+		return INVALID;
+	}
+	*offset += sizeof(udph);
+
+	tuplen = fill_tuple(&tuple, info->hdr, info->len, udph.source, udph.dest);
+	return classify_udp(skb, &tuple, tuplen);
+}
+
+static verdict_t process_ipv4(struct __sk_buff *skb, struct bpf_dynptr *dynptr,
+			      __u64 *offset, metrics_t *metrics)
+{
+	struct iphdr ipv4;
+	struct iphdr_info info = {
+		.hdr = &ipv4,
+		.len = sizeof(ipv4),
+	};
+
+	metrics->l3_protocol_packets_total_ipv4++;
+
+	if (pkt_parse_ipv4(dynptr, offset, &ipv4)) {
+		metrics->errors_total_malformed_ip++;
+		return INVALID;
+	}
+
+	if (ipv4.version != 4) {
+		metrics->errors_total_malformed_ip++;
+		return INVALID;
+	}
+
+	if (ipv4_is_fragment(&ipv4)) {
+		metrics->errors_total_fragmented_ip++;
+		return INVALID;
+	}
+
+	switch (ipv4.protocol) {
+	case IPPROTO_ICMP:
+		return process_icmpv4(skb, dynptr, offset, metrics);
+
+	case IPPROTO_TCP:
+		return process_tcp(dynptr, offset, skb, &info, metrics);
+
+	case IPPROTO_UDP:
+		return process_udp(dynptr, offset, skb, &info, metrics);
+
+	default:
+		metrics->errors_total_unknown_l4_proto++;
+		return INVALID;
+	}
+}
+
+static verdict_t process_ipv6(struct __sk_buff *skb, struct bpf_dynptr *dynptr,
+			      __u64 *offset, metrics_t *metrics)
+{
+	struct ipv6hdr ipv6;
+	struct iphdr_info info = {
+		.hdr = &ipv6,
+		.len = sizeof(ipv6),
+	};
+	uint8_t l4_proto;
+	bool is_fragment;
+
+	metrics->l3_protocol_packets_total_ipv6++;
+
+	if (pkt_parse_ipv6(dynptr, offset, &ipv6, &l4_proto, &is_fragment)) {
+		metrics->errors_total_malformed_ip++;
+		return INVALID;
+	}
+
+	if (ipv6.version != 6) {
+		metrics->errors_total_malformed_ip++;
+		return INVALID;
+	}
+
+	if (is_fragment) {
+		metrics->errors_total_fragmented_ip++;
+		return INVALID;
+	}
+
+	switch (l4_proto) {
+	case IPPROTO_ICMPV6:
+		return process_icmpv6(dynptr, offset, skb, metrics);
+
+	case IPPROTO_TCP:
+		return process_tcp(dynptr, offset, skb, &info, metrics);
+
+	case IPPROTO_UDP:
+		return process_udp(dynptr, offset, skb, &info, metrics);
+
+	default:
+		metrics->errors_total_unknown_l4_proto++;
+		return INVALID;
+	}
+}
+
+SEC("tc")
+int cls_redirect(struct __sk_buff *skb)
+{
+	__u8 encap_buffer[sizeof(encap_headers_t)] = {};
+	struct bpf_dynptr dynptr;
+	struct in_addr next_hop;
+	/* Tracks offset of the dynptr. This will be unnecessary once
+	 * bpf_dynptr_advance() is available.
+	 */
+	__u64 off = 0;
+	ret_t ret;
+
+	bpf_dynptr_from_skb(skb, 0, &dynptr);
+
+	metrics_t *metrics = get_global_metrics();
+	if (metrics == NULL)
+		return TC_ACT_SHOT;
+
+	metrics->processed_packets_total++;
+
+	/* Pass bogus packets as long as we're not sure they're
+	 * destined for us.
+	 */
+	if (skb->protocol != bpf_htons(ETH_P_IP))
+		return TC_ACT_OK;
+
+	encap_headers_t *encap;
+
+	/* Make sure that all encapsulation headers are available in
+	 * the linear portion of the skb. This makes it easy to manipulate them.
+	 */
+	if (bpf_skb_pull_data(skb, sizeof(*encap)))
+		return TC_ACT_OK;
+
+	encap = bpf_dynptr_slice_rdwr(&dynptr, 0, encap_buffer, sizeof(encap_buffer));
+	if (!encap)
+		return TC_ACT_OK;
+
+	off += sizeof(*encap);
+
+	if (encap->ip.ihl != 5)
+		/* We never have any options. */
+		return TC_ACT_OK;
+
+	if (encap->ip.daddr != ENCAPSULATION_IP ||
+	    encap->ip.protocol != IPPROTO_UDP)
+		return TC_ACT_OK;
+
+	/* TODO Check UDP length? */
+	if (encap->udp.dest != ENCAPSULATION_PORT)
+		return TC_ACT_OK;
+
+	/* We now know that the packet is destined to us, we can
+	 * drop bogus ones.
+	 */
+	if (ipv4_is_fragment((void *)&encap->ip)) {
+		metrics->errors_total_fragmented_ip++;
+		return TC_ACT_SHOT;
+	}
+
+	if (encap->gue.variant != 0) {
+		metrics->errors_total_malformed_encapsulation++;
+		return TC_ACT_SHOT;
+	}
+
+	if (encap->gue.control != 0) {
+		metrics->errors_total_malformed_encapsulation++;
+		return TC_ACT_SHOT;
+	}
+
+	if (encap->gue.flags != 0) {
+		metrics->errors_total_malformed_encapsulation++;
+		return TC_ACT_SHOT;
+	}
+
+	if (encap->gue.hlen !=
+	    sizeof(encap->unigue) / 4 + encap->unigue.hop_count) {
+		metrics->errors_total_malformed_encapsulation++;
+		return TC_ACT_SHOT;
+	}
+
+	if (encap->unigue.version != 0) {
+		metrics->errors_total_malformed_encapsulation++;
+		return TC_ACT_SHOT;
+	}
+
+	if (encap->unigue.reserved != 0)
+		return TC_ACT_SHOT;
+
+	MAYBE_RETURN(get_next_hop(&dynptr, &off, encap, &next_hop));
+
+	if (next_hop.s_addr == 0) {
+		metrics->accepted_packets_total_last_hop++;
+		return accept_locally(skb, encap);
+	}
+
+	verdict_t verdict;
+	switch (encap->gue.proto_ctype) {
+	case IPPROTO_IPIP:
+		verdict = process_ipv4(skb, &dynptr, &off, metrics);
+		break;
+
+	case IPPROTO_IPV6:
+		verdict = process_ipv6(skb, &dynptr, &off, metrics);
+		break;
+
+	default:
+		metrics->errors_total_unknown_l3_proto++;
+		return TC_ACT_SHOT;
+	}
+
+	switch (verdict) {
+	case INVALID:
+		/* metrics have already been bumped */
+		return TC_ACT_SHOT;
+
+	case UNKNOWN:
+		return forward_to_next_hop(skb, &dynptr, encap, &next_hop, metrics);
+
+	case ECHO_REQUEST:
+		metrics->accepted_packets_total_icmp_echo_request++;
+		break;
+
+	case SYN:
+		if (encap->unigue.forward_syn) {
+			return forward_to_next_hop(skb, &dynptr, encap, &next_hop,
+						   metrics);
+		}
+
+		metrics->accepted_packets_total_syn++;
+		break;
+
+	case SYN_COOKIE:
+		metrics->accepted_packets_total_syn_cookies++;
+		break;
+
+	case ESTABLISHED:
+		metrics->accepted_packets_total_established++;
+		break;
+	}
+
+	ret = accept_locally(skb, encap);
+
+	if (encap == encap_buffer)
+		bpf_dynptr_write(&dynptr, 0, encap_buffer, sizeof(encap_buffer), 0);
+
+	return ret;
+}
diff --git a/tools/testing/selftests/bpf/progs/test_l4lb_noinline_dynptr.c b/tools/testing/selftests/bpf/progs/test_l4lb_noinline_dynptr.c
new file mode 100644
index 000000000000..f997f5080748
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_l4lb_noinline_dynptr.c
@@ -0,0 +1,487 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2017 Facebook
+#include <stddef.h>
+#include <stdbool.h>
+#include <string.h>
+#include <linux/pkt_cls.h>
+#include <linux/bpf.h>
+#include <linux/in.h>
+#include <linux/if_ether.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/icmp.h>
+#include <linux/icmpv6.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+#include <bpf/bpf_helpers.h>
+#include "test_iptunnel_common.h"
+#include <bpf/bpf_endian.h>
+
+#include "bpf_kfuncs.h"
+
+static __always_inline __u32 rol32(__u32 word, unsigned int shift)
+{
+	return (word << shift) | (word >> ((-shift) & 31));
+}
+
+/* copy paste of jhash from kernel sources to make sure llvm
+ * can compile it into valid sequence of bpf instructions
+ */
+#define __jhash_mix(a, b, c)			\
+{						\
+	a -= c;  a ^= rol32(c, 4);  c += b;	\
+	b -= a;  b ^= rol32(a, 6);  a += c;	\
+	c -= b;  c ^= rol32(b, 8);  b += a;	\
+	a -= c;  a ^= rol32(c, 16); c += b;	\
+	b -= a;  b ^= rol32(a, 19); a += c;	\
+	c -= b;  c ^= rol32(b, 4);  b += a;	\
+}
+
+#define __jhash_final(a, b, c)			\
+{						\
+	c ^= b; c -= rol32(b, 14);		\
+	a ^= c; a -= rol32(c, 11);		\
+	b ^= a; b -= rol32(a, 25);		\
+	c ^= b; c -= rol32(b, 16);		\
+	a ^= c; a -= rol32(c, 4);		\
+	b ^= a; b -= rol32(a, 14);		\
+	c ^= b; c -= rol32(b, 24);		\
+}
+
+#define JHASH_INITVAL		0xdeadbeef
+
+typedef unsigned int u32;
+
+static __noinline u32 jhash(const void *key, u32 length, u32 initval)
+{
+	u32 a, b, c;
+	const unsigned char *k = key;
+
+	a = b = c = JHASH_INITVAL + length + initval;
+
+	while (length > 12) {
+		a += *(u32 *)(k);
+		b += *(u32 *)(k + 4);
+		c += *(u32 *)(k + 8);
+		__jhash_mix(a, b, c);
+		length -= 12;
+		k += 12;
+	}
+	switch (length) {
+	case 12: c += (u32)k[11]<<24;
+	case 11: c += (u32)k[10]<<16;
+	case 10: c += (u32)k[9]<<8;
+	case 9:  c += k[8];
+	case 8:  b += (u32)k[7]<<24;
+	case 7:  b += (u32)k[6]<<16;
+	case 6:  b += (u32)k[5]<<8;
+	case 5:  b += k[4];
+	case 4:  a += (u32)k[3]<<24;
+	case 3:  a += (u32)k[2]<<16;
+	case 2:  a += (u32)k[1]<<8;
+	case 1:  a += k[0];
+		 __jhash_final(a, b, c);
+	case 0: /* Nothing left to add */
+		break;
+	}
+
+	return c;
+}
+
+static __noinline u32 __jhash_nwords(u32 a, u32 b, u32 c, u32 initval)
+{
+	a += initval;
+	b += initval;
+	c += initval;
+	__jhash_final(a, b, c);
+	return c;
+}
+
+static __noinline u32 jhash_2words(u32 a, u32 b, u32 initval)
+{
+	return __jhash_nwords(a, b, 0, initval + JHASH_INITVAL + (2 << 2));
+}
+
+#define PCKT_FRAGMENTED 65343
+#define IPV4_HDR_LEN_NO_OPT 20
+#define IPV4_PLUS_ICMP_HDR 28
+#define IPV6_PLUS_ICMP_HDR 48
+#define RING_SIZE 2
+#define MAX_VIPS 12
+#define MAX_REALS 5
+#define CTL_MAP_SIZE 16
+#define CH_RINGS_SIZE (MAX_VIPS * RING_SIZE)
+#define F_IPV6 (1 << 0)
+#define F_HASH_NO_SRC_PORT (1 << 0)
+#define F_ICMP (1 << 0)
+#define F_SYN_SET (1 << 1)
+
+struct packet_description {
+	union {
+		__be32 src;
+		__be32 srcv6[4];
+	};
+	union {
+		__be32 dst;
+		__be32 dstv6[4];
+	};
+	union {
+		__u32 ports;
+		__u16 port16[2];
+	};
+	__u8 proto;
+	__u8 flags;
+};
+
+struct ctl_value {
+	union {
+		__u64 value;
+		__u32 ifindex;
+		__u8 mac[6];
+	};
+};
+
+struct vip_meta {
+	__u32 flags;
+	__u32 vip_num;
+};
+
+struct real_definition {
+	union {
+		__be32 dst;
+		__be32 dstv6[4];
+	};
+	__u8 flags;
+};
+
+struct vip_stats {
+	__u64 bytes;
+	__u64 pkts;
+};
+
+struct eth_hdr {
+	unsigned char eth_dest[ETH_ALEN];
+	unsigned char eth_source[ETH_ALEN];
+	unsigned short eth_proto;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, MAX_VIPS);
+	__type(key, struct vip);
+	__type(value, struct vip_meta);
+} vip_map SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, CH_RINGS_SIZE);
+	__type(key, __u32);
+	__type(value, __u32);
+} ch_rings SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, MAX_REALS);
+	__type(key, __u32);
+	__type(value, struct real_definition);
+} reals SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__uint(max_entries, MAX_VIPS);
+	__type(key, __u32);
+	__type(value, struct vip_stats);
+} stats SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, CTL_MAP_SIZE);
+	__type(key, __u32);
+	__type(value, struct ctl_value);
+} ctl_array SEC(".maps");
+
+static __noinline __u32 get_packet_hash(struct packet_description *pckt, bool ipv6)
+{
+	if (ipv6)
+		return jhash_2words(jhash(pckt->srcv6, 16, MAX_VIPS),
+				    pckt->ports, CH_RINGS_SIZE);
+	else
+		return jhash_2words(pckt->src, pckt->ports, CH_RINGS_SIZE);
+}
+
+static __noinline bool get_packet_dst(struct real_definition **real,
+				      struct packet_description *pckt,
+				      struct vip_meta *vip_info,
+				      bool is_ipv6)
+{
+	__u32 hash = get_packet_hash(pckt, is_ipv6);
+	__u32 key = RING_SIZE * vip_info->vip_num + hash % RING_SIZE;
+	__u32 *real_pos;
+
+	if (hash != 0x358459b7 /* jhash of ipv4 packet */  &&
+	    hash != 0x2f4bc6bb /* jhash of ipv6 packet */)
+		return false;
+
+	real_pos = bpf_map_lookup_elem(&ch_rings, &key);
+	if (!real_pos)
+		return false;
+	key = *real_pos;
+	*real = bpf_map_lookup_elem(&reals, &key);
+	if (!(*real))
+		return false;
+	return true;
+}
+
+static __noinline int parse_icmpv6(struct bpf_dynptr *skb_ptr, __u64 off,
+				   struct packet_description *pckt)
+{
+	__u8 buffer[sizeof(struct ipv6hdr)] = {};
+	struct icmp6hdr *icmp_hdr;
+	struct ipv6hdr *ip6h;
+
+	icmp_hdr = bpf_dynptr_slice(skb_ptr, off, buffer, sizeof(buffer));
+	if (!icmp_hdr)
+		return TC_ACT_SHOT;
+
+	if (icmp_hdr->icmp6_type != ICMPV6_PKT_TOOBIG)
+		return TC_ACT_OK;
+	off += sizeof(struct icmp6hdr);
+	ip6h = bpf_dynptr_slice(skb_ptr, off, buffer, sizeof(buffer));
+	if (!ip6h)
+		return TC_ACT_SHOT;
+	pckt->proto = ip6h->nexthdr;
+	pckt->flags |= F_ICMP;
+	memcpy(pckt->srcv6, ip6h->daddr.s6_addr32, 16);
+	memcpy(pckt->dstv6, ip6h->saddr.s6_addr32, 16);
+	return TC_ACT_UNSPEC;
+}
+
+static __noinline int parse_icmp(struct bpf_dynptr *skb_ptr, __u64 off,
+				 struct packet_description *pckt)
+{
+	__u8 buffer_icmp[sizeof(struct iphdr)] = {};
+	__u8 buffer_ip[sizeof(struct iphdr)] = {};
+	struct icmphdr *icmp_hdr;
+	struct iphdr *iph;
+
+	icmp_hdr = bpf_dynptr_slice(skb_ptr, off, buffer_icmp, sizeof(buffer_icmp));
+	if (!icmp_hdr)
+		return TC_ACT_SHOT;
+	if (icmp_hdr->type != ICMP_DEST_UNREACH ||
+	    icmp_hdr->code != ICMP_FRAG_NEEDED)
+		return TC_ACT_OK;
+	off += sizeof(struct icmphdr);
+	iph = bpf_dynptr_slice(skb_ptr, off, buffer_ip, sizeof(buffer_ip));
+	if (!iph || iph->ihl != 5)
+		return TC_ACT_SHOT;
+	pckt->proto = iph->protocol;
+	pckt->flags |= F_ICMP;
+	pckt->src = iph->daddr;
+	pckt->dst = iph->saddr;
+	return TC_ACT_UNSPEC;
+}
+
+static __noinline bool parse_udp(struct bpf_dynptr *skb_ptr, __u64 off,
+				 struct packet_description *pckt)
+{
+	__u8 buffer[sizeof(struct udphdr)] = {};
+	struct udphdr *udp;
+
+	udp = bpf_dynptr_slice(skb_ptr, off, buffer, sizeof(buffer));
+	if (!udp)
+		return false;
+
+	if (!(pckt->flags & F_ICMP)) {
+		pckt->port16[0] = udp->source;
+		pckt->port16[1] = udp->dest;
+	} else {
+		pckt->port16[0] = udp->dest;
+		pckt->port16[1] = udp->source;
+	}
+	return true;
+}
+
+static __noinline bool parse_tcp(struct bpf_dynptr *skb_ptr, __u64 off,
+				 struct packet_description *pckt)
+{
+	__u8 buffer[sizeof(struct tcphdr)] = {};
+	struct tcphdr *tcp;
+
+	tcp = bpf_dynptr_slice(skb_ptr, off, buffer, sizeof(buffer));
+	if (!tcp)
+		return false;
+
+	if (tcp->syn)
+		pckt->flags |= F_SYN_SET;
+
+	if (!(pckt->flags & F_ICMP)) {
+		pckt->port16[0] = tcp->source;
+		pckt->port16[1] = tcp->dest;
+	} else {
+		pckt->port16[0] = tcp->dest;
+		pckt->port16[1] = tcp->source;
+	}
+	return true;
+}
+
+static __noinline int process_packet(struct bpf_dynptr *skb_ptr,
+				     struct eth_hdr *eth, __u64 off,
+				     bool is_ipv6, struct __sk_buff *skb)
+{
+	struct packet_description pckt = {};
+	struct bpf_tunnel_key tkey = {};
+	struct vip_stats *data_stats;
+	struct real_definition *dst;
+	struct vip_meta *vip_info;
+	struct ctl_value *cval;
+	__u32 v4_intf_pos = 1;
+	__u32 v6_intf_pos = 2;
+	struct ipv6hdr *ip6h;
+	struct vip vip = {};
+	struct iphdr *iph;
+	int tun_flag = 0;
+	__u16 pkt_bytes;
+	__u64 iph_len;
+	__u32 ifindex;
+	__u8 protocol;
+	__u32 vip_num;
+	int action;
+
+	tkey.tunnel_ttl = 64;
+	if (is_ipv6) {
+		__u8 buffer[sizeof(struct ipv6hdr)] = {};
+
+		ip6h = bpf_dynptr_slice(skb_ptr, off, buffer, sizeof(buffer));
+		if (!ip6h)
+			return TC_ACT_SHOT;
+
+		iph_len = sizeof(struct ipv6hdr);
+		protocol = ip6h->nexthdr;
+		pckt.proto = protocol;
+		pkt_bytes = bpf_ntohs(ip6h->payload_len);
+		off += iph_len;
+		if (protocol == IPPROTO_FRAGMENT) {
+			return TC_ACT_SHOT;
+		} else if (protocol == IPPROTO_ICMPV6) {
+			action = parse_icmpv6(skb_ptr, off, &pckt);
+			if (action >= 0)
+				return action;
+			off += IPV6_PLUS_ICMP_HDR;
+		} else {
+			memcpy(pckt.srcv6, ip6h->saddr.s6_addr32, 16);
+			memcpy(pckt.dstv6, ip6h->daddr.s6_addr32, 16);
+		}
+	} else {
+		__u8 buffer[sizeof(struct iphdr)] = {};
+
+		iph = bpf_dynptr_slice(skb_ptr, off, buffer, sizeof(buffer));
+		if (!iph || iph->ihl != 5)
+			return TC_ACT_SHOT;
+
+		protocol = iph->protocol;
+		pckt.proto = protocol;
+		pkt_bytes = bpf_ntohs(iph->tot_len);
+		off += IPV4_HDR_LEN_NO_OPT;
+
+		if (iph->frag_off & PCKT_FRAGMENTED)
+			return TC_ACT_SHOT;
+		if (protocol == IPPROTO_ICMP) {
+			action = parse_icmp(skb_ptr, off, &pckt);
+			if (action >= 0)
+				return action;
+			off += IPV4_PLUS_ICMP_HDR;
+		} else {
+			pckt.src = iph->saddr;
+			pckt.dst = iph->daddr;
+		}
+	}
+	protocol = pckt.proto;
+
+	if (protocol == IPPROTO_TCP) {
+		if (!parse_tcp(skb_ptr, off, &pckt))
+			return TC_ACT_SHOT;
+	} else if (protocol == IPPROTO_UDP) {
+		if (!parse_udp(skb_ptr, off, &pckt))
+			return TC_ACT_SHOT;
+	} else {
+		return TC_ACT_SHOT;
+	}
+
+	if (is_ipv6)
+		memcpy(vip.daddr.v6, pckt.dstv6, 16);
+	else
+		vip.daddr.v4 = pckt.dst;
+
+	vip.dport = pckt.port16[1];
+	vip.protocol = pckt.proto;
+	vip_info = bpf_map_lookup_elem(&vip_map, &vip);
+	if (!vip_info) {
+		vip.dport = 0;
+		vip_info = bpf_map_lookup_elem(&vip_map, &vip);
+		if (!vip_info)
+			return TC_ACT_SHOT;
+		pckt.port16[1] = 0;
+	}
+
+	if (vip_info->flags & F_HASH_NO_SRC_PORT)
+		pckt.port16[0] = 0;
+
+	if (!get_packet_dst(&dst, &pckt, vip_info, is_ipv6))
+		return TC_ACT_SHOT;
+
+	if (dst->flags & F_IPV6) {
+		cval = bpf_map_lookup_elem(&ctl_array, &v6_intf_pos);
+		if (!cval)
+			return TC_ACT_SHOT;
+		ifindex = cval->ifindex;
+		memcpy(tkey.remote_ipv6, dst->dstv6, 16);
+		tun_flag = BPF_F_TUNINFO_IPV6;
+	} else {
+		cval = bpf_map_lookup_elem(&ctl_array, &v4_intf_pos);
+		if (!cval)
+			return TC_ACT_SHOT;
+		ifindex = cval->ifindex;
+		tkey.remote_ipv4 = dst->dst;
+	}
+	vip_num = vip_info->vip_num;
+	data_stats = bpf_map_lookup_elem(&stats, &vip_num);
+	if (!data_stats)
+		return TC_ACT_SHOT;
+	data_stats->pkts++;
+	data_stats->bytes += pkt_bytes;
+	bpf_skb_set_tunnel_key(skb, &tkey, sizeof(tkey), tun_flag);
+	*(u32 *)eth->eth_dest = tkey.remote_ipv4;
+	return bpf_redirect(ifindex, 0);
+}
+
+SEC("tc")
+int balancer_ingress(struct __sk_buff *ctx)
+{
+	__u8 buffer[sizeof(struct eth_hdr)] = {};
+	struct bpf_dynptr ptr;
+	struct eth_hdr *eth;
+	__u32 eth_proto;
+	__u32 nh_off;
+	int err;
+
+	nh_off = sizeof(struct eth_hdr);
+
+	bpf_dynptr_from_skb(ctx, 0, &ptr);
+	eth = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+	if (!eth)
+		return TC_ACT_SHOT;
+	eth_proto = eth->eth_proto;
+	if (eth_proto == bpf_htons(ETH_P_IP))
+		err = process_packet(&ptr, eth, nh_off, false, ctx);
+	else if (eth_proto == bpf_htons(ETH_P_IPV6))
+		err = process_packet(&ptr, eth, nh_off, true, ctx);
+	else
+		return TC_ACT_SHOT;
+
+	if (eth == buffer)
+		bpf_dynptr_write(&ptr, 0, buffer, sizeof(buffer), 0);
+
+	return err;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c
new file mode 100644
index 000000000000..79bab9b50e9e
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c
@@ -0,0 +1,119 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* This parsing logic is taken from the open source library katran, a layer 4
+ * load balancer.
+ *
+ * This code logic using dynptrs can be found in test_parse_tcp_hdr_opt_dynptr.c
+ *
+ * https://github.com/facebookincubator/katran/blob/main/katran/lib/bpf/pckt_parsing.h
+ */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <linux/tcp.h>
+#include <stdbool.h>
+#include <linux/ipv6.h>
+#include <linux/if_ether.h>
+#include "test_tcp_hdr_options.h"
+
+char _license[] SEC("license") = "GPL";
+
+/* Kind number used for experiments */
+const __u32 tcp_hdr_opt_kind_tpr = 0xFD;
+/* Length of the tcp header option */
+const __u32 tcp_hdr_opt_len_tpr = 6;
+/* maximum number of header options to check to lookup server_id */
+const __u32 tcp_hdr_opt_max_opt_checks = 15;
+
+__u32 server_id;
+
+struct hdr_opt_state {
+	__u32 server_id;
+	__u8 byte_offset;
+	__u8 hdr_bytes_remaining;
+};
+
+static int parse_hdr_opt(const struct xdp_md *xdp, struct hdr_opt_state *state)
+{
+	const void *data = (void *)(long)xdp->data;
+	const void *data_end = (void *)(long)xdp->data_end;
+	__u8 *tcp_opt, kind, hdr_len;
+
+	tcp_opt = (__u8 *)(data + state->byte_offset);
+	if (tcp_opt + 1 > data_end)
+		return -1;
+
+	kind = tcp_opt[0];
+
+	if (kind == TCPOPT_EOL)
+		return -1;
+
+	if (kind == TCPOPT_NOP) {
+		state->hdr_bytes_remaining--;
+		state->byte_offset++;
+		return 0;
+	}
+
+	if (state->hdr_bytes_remaining < 2 ||
+	    tcp_opt + sizeof(__u8) + sizeof(__u8) > data_end)
+		return -1;
+
+	hdr_len = tcp_opt[1];
+	if (hdr_len > state->hdr_bytes_remaining)
+		return -1;
+
+	if (kind == tcp_hdr_opt_kind_tpr) {
+		if (hdr_len != tcp_hdr_opt_len_tpr)
+			return -1;
+
+		if (tcp_opt + tcp_hdr_opt_len_tpr > data_end)
+			return -1;
+
+		state->server_id = *(__u32 *)&tcp_opt[2];
+		return 1;
+	}
+
+	state->hdr_bytes_remaining -= hdr_len;
+	state->byte_offset += hdr_len;
+	return 0;
+}
+
+SEC("xdp")
+int xdp_ingress_v6(struct xdp_md *xdp)
+{
+	const void *data = (void *)(long)xdp->data;
+	const void *data_end = (void *)(long)xdp->data_end;
+	struct hdr_opt_state opt_state = {};
+	__u8 tcp_hdr_opt_len = 0;
+	struct tcphdr *tcp_hdr;
+	__u64 tcp_offset = 0;
+	__u32 off;
+	int err;
+
+	tcp_offset = sizeof(struct ethhdr) + sizeof(struct ipv6hdr);
+	tcp_hdr = (struct tcphdr *)(data + tcp_offset);
+	if (tcp_hdr + 1 > data_end)
+		return XDP_DROP;
+
+	tcp_hdr_opt_len = (tcp_hdr->doff * 4) - sizeof(struct tcphdr);
+	if (tcp_hdr_opt_len < tcp_hdr_opt_len_tpr)
+		return XDP_DROP;
+
+	opt_state.hdr_bytes_remaining = tcp_hdr_opt_len;
+	opt_state.byte_offset = sizeof(struct tcphdr) + tcp_offset;
+
+	/* max number of bytes of options in tcp header is 40 bytes */
+	for (int i = 0; i < tcp_hdr_opt_max_opt_checks; i++) {
+		err = parse_hdr_opt(xdp, &opt_state);
+
+		if (err || !opt_state.hdr_bytes_remaining)
+			break;
+	}
+
+	if (!opt_state.server_id)
+		return XDP_DROP;
+
+	server_id = opt_state.server_id;
+
+	return XDP_PASS;
+}
diff --git a/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c
new file mode 100644
index 000000000000..d3b319722e30
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c
@@ -0,0 +1,114 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* This logic is lifted from a real-world use case of packet parsing, used in
+ * the open source library katran, a layer 4 load balancer.
+ *
+ * This test demonstrates how to parse packet contents using dynptrs. The
+ * original code (parsing without dynptrs) can be found in test_parse_tcp_hdr_opt.c
+ */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <linux/tcp.h>
+#include <stdbool.h>
+#include <linux/ipv6.h>
+#include <linux/if_ether.h>
+#include "test_tcp_hdr_options.h"
+#include "bpf_kfuncs.h"
+
+char _license[] SEC("license") = "GPL";
+
+/* Kind number used for experiments */
+const __u32 tcp_hdr_opt_kind_tpr = 0xFD;
+/* Length of the tcp header option */
+const __u32 tcp_hdr_opt_len_tpr = 6;
+/* maximum number of header options to check to lookup server_id */
+const __u32 tcp_hdr_opt_max_opt_checks = 15;
+
+__u32 server_id;
+
+static int parse_hdr_opt(struct bpf_dynptr *ptr, __u32 *off, __u8 *hdr_bytes_remaining,
+			 __u32 *server_id)
+{
+	__u8 *tcp_opt, kind, hdr_len;
+	__u8 buffer[sizeof(kind) + sizeof(hdr_len) + sizeof(*server_id)];
+	__u8 *data;
+
+	__builtin_memset(buffer, 0, sizeof(buffer));
+
+	data = bpf_dynptr_slice(ptr, *off, buffer, sizeof(buffer));
+	if (!data)
+		return -1;
+
+	kind = data[0];
+
+	if (kind == TCPOPT_EOL)
+		return -1;
+
+	if (kind == TCPOPT_NOP) {
+		*off += 1;
+		*hdr_bytes_remaining -= 1;
+		return 0;
+	}
+
+	if (*hdr_bytes_remaining < 2)
+		return -1;
+
+	hdr_len = data[1];
+	if (hdr_len > *hdr_bytes_remaining)
+		return -1;
+
+	if (kind == tcp_hdr_opt_kind_tpr) {
+		if (hdr_len != tcp_hdr_opt_len_tpr)
+			return -1;
+
+		__builtin_memcpy(server_id, (__u32 *)(data + 2), sizeof(*server_id));
+		return 1;
+	}
+
+	*off += hdr_len;
+	*hdr_bytes_remaining -= hdr_len;
+	return 0;
+}
+
+SEC("xdp")
+int xdp_ingress_v6(struct xdp_md *xdp)
+{
+	__u8 buffer[sizeof(struct tcphdr)] = {};
+	__u8 hdr_bytes_remaining;
+	struct tcphdr *tcp_hdr;
+	__u8 tcp_hdr_opt_len;
+	int err = 0;
+	__u32 off;
+
+	struct bpf_dynptr ptr;
+
+	bpf_dynptr_from_xdp(xdp, 0, &ptr);
+
+	off = sizeof(struct ethhdr) + sizeof(struct ipv6hdr);
+
+	tcp_hdr = bpf_dynptr_slice(&ptr, off, buffer, sizeof(buffer));
+	if (!tcp_hdr)
+		return XDP_DROP;
+
+	tcp_hdr_opt_len = (tcp_hdr->doff * 4) - sizeof(struct tcphdr);
+	if (tcp_hdr_opt_len < tcp_hdr_opt_len_tpr)
+		return XDP_DROP;
+
+	hdr_bytes_remaining = tcp_hdr_opt_len;
+
+	off += sizeof(struct tcphdr);
+
+	/* max number of bytes of options in tcp header is 40 bytes */
+	for (int i = 0; i < tcp_hdr_opt_max_opt_checks; i++) {
+		err = parse_hdr_opt(&ptr, &off, &hdr_bytes_remaining, &server_id);
+
+		if (err || !hdr_bytes_remaining)
+			break;
+	}
+
+	if (!server_id)
+		return XDP_DROP;
+
+	return XDP_PASS;
+}
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c b/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c
new file mode 100644
index 000000000000..7521a805b506
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c
@@ -0,0 +1,257 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2022 Meta */
+#include <stddef.h>
+#include <string.h>
+#include <linux/bpf.h>
+#include <linux/if_ether.h>
+#include <linux/if_packet.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/in.h>
+#include <linux/udp.h>
+#include <linux/tcp.h>
+#include <linux/pkt_cls.h>
+#include <sys/socket.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
+#include "test_iptunnel_common.h"
+#include "bpf_kfuncs.h"
+
+const size_t tcphdr_sz = sizeof(struct tcphdr);
+const size_t udphdr_sz = sizeof(struct udphdr);
+const size_t ethhdr_sz = sizeof(struct ethhdr);
+const size_t iphdr_sz = sizeof(struct iphdr);
+const size_t ipv6hdr_sz = sizeof(struct ipv6hdr);
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__uint(max_entries, 256);
+	__type(key, __u32);
+	__type(value, __u64);
+} rxcnt SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, MAX_IPTNL_ENTRIES);
+	__type(key, struct vip);
+	__type(value, struct iptnl_info);
+} vip2tnl SEC(".maps");
+
+static __always_inline void count_tx(__u32 protocol)
+{
+	__u64 *rxcnt_count;
+
+	rxcnt_count = bpf_map_lookup_elem(&rxcnt, &protocol);
+	if (rxcnt_count)
+		*rxcnt_count += 1;
+}
+
+static __always_inline int get_dport(void *trans_data, __u8 protocol)
+{
+	struct tcphdr *th;
+	struct udphdr *uh;
+
+	switch (protocol) {
+	case IPPROTO_TCP:
+		th = (struct tcphdr *)trans_data;
+		return th->dest;
+	case IPPROTO_UDP:
+		uh = (struct udphdr *)trans_data;
+		return uh->dest;
+	default:
+		return 0;
+	}
+}
+
+static __always_inline void set_ethhdr(struct ethhdr *new_eth,
+				       const struct ethhdr *old_eth,
+				       const struct iptnl_info *tnl,
+				       __be16 h_proto)
+{
+	memcpy(new_eth->h_source, old_eth->h_dest, sizeof(new_eth->h_source));
+	memcpy(new_eth->h_dest, tnl->dmac, sizeof(new_eth->h_dest));
+	new_eth->h_proto = h_proto;
+}
+
+static __always_inline int handle_ipv4(struct xdp_md *xdp, struct bpf_dynptr *xdp_ptr)
+{
+	__u8 eth_buffer[ethhdr_sz + iphdr_sz + ethhdr_sz];
+	__u8 iph_buffer_tcp[iphdr_sz + tcphdr_sz];
+	__u8 iph_buffer_udp[iphdr_sz + udphdr_sz];
+	struct bpf_dynptr new_xdp_ptr;
+	struct iptnl_info *tnl;
+	struct ethhdr *new_eth;
+	struct ethhdr *old_eth;
+	__u32 transport_hdr_sz;
+	struct iphdr *iph;
+	__u16 *next_iph;
+	__u16 payload_len;
+	struct vip vip = {};
+	int dport;
+	__u32 csum = 0;
+	int i;
+
+	__builtin_memset(eth_buffer, 0, sizeof(eth_buffer));
+	__builtin_memset(iph_buffer_tcp, 0, sizeof(iph_buffer_tcp));
+	__builtin_memset(iph_buffer_udp, 0, sizeof(iph_buffer_udp));
+
+	if (ethhdr_sz + iphdr_sz + tcphdr_sz > xdp->data_end - xdp->data)
+		iph = bpf_dynptr_slice(xdp_ptr, ethhdr_sz, iph_buffer_udp, sizeof(iph_buffer_udp));
+	else
+		iph = bpf_dynptr_slice(xdp_ptr, ethhdr_sz, iph_buffer_tcp, sizeof(iph_buffer_tcp));
+
+	if (!iph)
+		return XDP_DROP;
+
+	dport = get_dport(iph + 1, iph->protocol);
+	if (dport == -1)
+		return XDP_DROP;
+
+	vip.protocol = iph->protocol;
+	vip.family = AF_INET;
+	vip.daddr.v4 = iph->daddr;
+	vip.dport = dport;
+	payload_len = bpf_ntohs(iph->tot_len);
+
+	tnl = bpf_map_lookup_elem(&vip2tnl, &vip);
+	/* It only does v4-in-v4 */
+	if (!tnl || tnl->family != AF_INET)
+		return XDP_PASS;
+
+	if (bpf_xdp_adjust_head(xdp, 0 - (int)iphdr_sz))
+		return XDP_DROP;
+
+	bpf_dynptr_from_xdp(xdp, 0, &new_xdp_ptr);
+	new_eth = bpf_dynptr_slice_rdwr(&new_xdp_ptr, 0, eth_buffer, sizeof(eth_buffer));
+	if (!new_eth)
+		return XDP_DROP;
+
+	iph = (struct iphdr *)(new_eth + 1);
+	old_eth = (struct ethhdr *)(iph + 1);
+
+	set_ethhdr(new_eth, old_eth, tnl, bpf_htons(ETH_P_IP));
+
+	if (new_eth == eth_buffer)
+		bpf_dynptr_write(&new_xdp_ptr, 0, eth_buffer, sizeof(eth_buffer), 0);
+
+	iph->version = 4;
+	iph->ihl = iphdr_sz >> 2;
+	iph->frag_off =	0;
+	iph->protocol = IPPROTO_IPIP;
+	iph->check = 0;
+	iph->tos = 0;
+	iph->tot_len = bpf_htons(payload_len + iphdr_sz);
+	iph->daddr = tnl->daddr.v4;
+	iph->saddr = tnl->saddr.v4;
+	iph->ttl = 8;
+
+	next_iph = (__u16 *)iph;
+	for (i = 0; i < iphdr_sz >> 1; i++)
+		csum += *next_iph++;
+
+	iph->check = ~((csum & 0xffff) + (csum >> 16));
+
+	count_tx(vip.protocol);
+
+	return XDP_TX;
+}
+
+static __always_inline int handle_ipv6(struct xdp_md *xdp, struct bpf_dynptr *xdp_ptr)
+{
+	__u8 eth_buffer[ethhdr_sz + ipv6hdr_sz + ethhdr_sz];
+	__u8 ip6h_buffer_tcp[ipv6hdr_sz + tcphdr_sz];
+	__u8 ip6h_buffer_udp[ipv6hdr_sz + udphdr_sz];
+	struct bpf_dynptr new_xdp_ptr;
+	struct iptnl_info *tnl;
+	struct ethhdr *new_eth;
+	struct ethhdr *old_eth;
+	__u32 transport_hdr_sz;
+	struct ipv6hdr *ip6h;
+	__u16 payload_len;
+	struct vip vip = {};
+	int dport;
+
+	__builtin_memset(eth_buffer, 0, sizeof(eth_buffer));
+	__builtin_memset(ip6h_buffer_tcp, 0, sizeof(ip6h_buffer_tcp));
+	__builtin_memset(ip6h_buffer_udp, 0, sizeof(ip6h_buffer_udp));
+
+	if (ethhdr_sz + iphdr_sz + tcphdr_sz > xdp->data_end - xdp->data)
+		ip6h = bpf_dynptr_slice(xdp_ptr, ethhdr_sz, ip6h_buffer_udp, sizeof(ip6h_buffer_udp));
+	else
+		ip6h = bpf_dynptr_slice(xdp_ptr, ethhdr_sz, ip6h_buffer_tcp, sizeof(ip6h_buffer_tcp));
+
+	if (!ip6h)
+		return XDP_DROP;
+
+	dport = get_dport(ip6h + 1, ip6h->nexthdr);
+	if (dport == -1)
+		return XDP_DROP;
+
+	vip.protocol = ip6h->nexthdr;
+	vip.family = AF_INET6;
+	memcpy(vip.daddr.v6, ip6h->daddr.s6_addr32, sizeof(vip.daddr));
+	vip.dport = dport;
+	payload_len = ip6h->payload_len;
+
+	tnl = bpf_map_lookup_elem(&vip2tnl, &vip);
+	/* It only does v6-in-v6 */
+	if (!tnl || tnl->family != AF_INET6)
+		return XDP_PASS;
+
+	if (bpf_xdp_adjust_head(xdp, 0 - (int)ipv6hdr_sz))
+		return XDP_DROP;
+
+	bpf_dynptr_from_xdp(xdp, 0, &new_xdp_ptr);
+	new_eth = bpf_dynptr_slice_rdwr(&new_xdp_ptr, 0, eth_buffer, sizeof(eth_buffer));
+	if (!new_eth)
+		return XDP_DROP;
+
+	ip6h = (struct ipv6hdr *)(new_eth + 1);
+	old_eth = (struct ethhdr *)(ip6h + 1);
+
+	set_ethhdr(new_eth, old_eth, tnl, bpf_htons(ETH_P_IPV6));
+
+	if (new_eth == eth_buffer)
+		bpf_dynptr_write(&new_xdp_ptr, 0, eth_buffer, sizeof(eth_buffer), 0);
+
+	ip6h->version = 6;
+	ip6h->priority = 0;
+	memset(ip6h->flow_lbl, 0, sizeof(ip6h->flow_lbl));
+	ip6h->payload_len = bpf_htons(bpf_ntohs(payload_len) + ipv6hdr_sz);
+	ip6h->nexthdr = IPPROTO_IPV6;
+	ip6h->hop_limit = 8;
+	memcpy(ip6h->saddr.s6_addr32, tnl->saddr.v6, sizeof(tnl->saddr.v6));
+	memcpy(ip6h->daddr.s6_addr32, tnl->daddr.v6, sizeof(tnl->daddr.v6));
+
+	count_tx(vip.protocol);
+
+	return XDP_TX;
+}
+
+SEC("xdp")
+int _xdp_tx_iptunnel(struct xdp_md *xdp)
+{
+	__u8 buffer[ethhdr_sz];
+	struct bpf_dynptr ptr;
+	struct ethhdr *eth;
+	__u16 h_proto;
+
+	__builtin_memset(buffer, 0, sizeof(buffer));
+
+	bpf_dynptr_from_xdp(xdp, 0, &ptr);
+	eth = bpf_dynptr_slice(&ptr, 0, buffer, sizeof(buffer));
+	if (!eth)
+		return XDP_DROP;
+
+	h_proto = eth->h_proto;
+
+	if (h_proto == bpf_htons(ETH_P_IP))
+		return handle_ipv4(xdp, &ptr);
+	else if (h_proto == bpf_htons(ETH_P_IPV6))
+
+		return handle_ipv6(xdp, &ptr);
+	else
+		return XDP_DROP;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/test_tcp_hdr_options.h b/tools/testing/selftests/bpf/test_tcp_hdr_options.h
index 6118e3ab61fc..56c9f8a3ad3d 100644
--- a/tools/testing/selftests/bpf/test_tcp_hdr_options.h
+++ b/tools/testing/selftests/bpf/test_tcp_hdr_options.h
@@ -50,6 +50,7 @@ struct linum_err {
 
 #define TCPOPT_EOL		0
 #define TCPOPT_NOP		1
+#define TCPOPT_MSS		2
 #define TCPOPT_WINDOW		3
 #define TCPOPT_EXP		254
 
-- 
cgit v1.2.3-70-g09d2


From 65334e64a493c6a0976de7ad56bf8b7a9ff04b4a Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Sat, 25 Feb 2023 16:40:08 +0100
Subject: bpf: Support kptrs in percpu hashmap and percpu LRU hashmap

Enable support for kptrs in percpu BPF hashmap and percpu BPF LRU
hashmap by wiring up the freeing of these kptrs from percpu map
elements.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230225154010.391965-2-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/hashtab.c | 59 ++++++++++++++++++++++++++++++++--------------------
 kernel/bpf/syscall.c |  2 ++
 2 files changed, 39 insertions(+), 22 deletions(-)

diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 5dfcb5ad0d06..653aeb481c79 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -249,7 +249,18 @@ static void htab_free_prealloced_fields(struct bpf_htab *htab)
 		struct htab_elem *elem;
 
 		elem = get_htab_elem(htab, i);
-		bpf_obj_free_fields(htab->map.record, elem->key + round_up(htab->map.key_size, 8));
+		if (htab_is_percpu(htab)) {
+			void __percpu *pptr = htab_elem_get_ptr(elem, htab->map.key_size);
+			int cpu;
+
+			for_each_possible_cpu(cpu) {
+				bpf_obj_free_fields(htab->map.record, per_cpu_ptr(pptr, cpu));
+				cond_resched();
+			}
+		} else {
+			bpf_obj_free_fields(htab->map.record, elem->key + round_up(htab->map.key_size, 8));
+			cond_resched();
+		}
 		cond_resched();
 	}
 }
@@ -759,9 +770,17 @@ static int htab_lru_map_gen_lookup(struct bpf_map *map,
 static void check_and_free_fields(struct bpf_htab *htab,
 				  struct htab_elem *elem)
 {
-	void *map_value = elem->key + round_up(htab->map.key_size, 8);
+	if (htab_is_percpu(htab)) {
+		void __percpu *pptr = htab_elem_get_ptr(elem, htab->map.key_size);
+		int cpu;
 
-	bpf_obj_free_fields(htab->map.record, map_value);
+		for_each_possible_cpu(cpu)
+			bpf_obj_free_fields(htab->map.record, per_cpu_ptr(pptr, cpu));
+	} else {
+		void *map_value = elem->key + round_up(htab->map.key_size, 8);
+
+		bpf_obj_free_fields(htab->map.record, map_value);
+	}
 }
 
 /* It is called from the bpf_lru_list when the LRU needs to delete
@@ -858,9 +877,9 @@ find_first_elem:
 
 static void htab_elem_free(struct bpf_htab *htab, struct htab_elem *l)
 {
+	check_and_free_fields(htab, l);
 	if (htab->map.map_type == BPF_MAP_TYPE_PERCPU_HASH)
 		bpf_mem_cache_free(&htab->pcpu_ma, l->ptr_to_pptr);
-	check_and_free_fields(htab, l);
 	bpf_mem_cache_free(&htab->ma, l);
 }
 
@@ -918,14 +937,13 @@ static void pcpu_copy_value(struct bpf_htab *htab, void __percpu *pptr,
 {
 	if (!onallcpus) {
 		/* copy true value_size bytes */
-		memcpy(this_cpu_ptr(pptr), value, htab->map.value_size);
+		copy_map_value(&htab->map, this_cpu_ptr(pptr), value);
 	} else {
 		u32 size = round_up(htab->map.value_size, 8);
 		int off = 0, cpu;
 
 		for_each_possible_cpu(cpu) {
-			bpf_long_memcpy(per_cpu_ptr(pptr, cpu),
-					value + off, size);
+			copy_map_value_long(&htab->map, per_cpu_ptr(pptr, cpu), value + off);
 			off += size;
 		}
 	}
@@ -940,16 +958,14 @@ static void pcpu_init_value(struct bpf_htab *htab, void __percpu *pptr,
 	 * (onallcpus=false always when coming from bpf prog).
 	 */
 	if (!onallcpus) {
-		u32 size = round_up(htab->map.value_size, 8);
 		int current_cpu = raw_smp_processor_id();
 		int cpu;
 
 		for_each_possible_cpu(cpu) {
 			if (cpu == current_cpu)
-				bpf_long_memcpy(per_cpu_ptr(pptr, cpu), value,
-						size);
-			else
-				memset(per_cpu_ptr(pptr, cpu), 0, size);
+				copy_map_value_long(&htab->map, per_cpu_ptr(pptr, cpu), value);
+			else /* Since elem is preallocated, we cannot touch special fields */
+				zero_map_value(&htab->map, per_cpu_ptr(pptr, cpu));
 		}
 	} else {
 		pcpu_copy_value(htab, pptr, value, onallcpus);
@@ -1575,9 +1591,8 @@ static int __htab_map_lookup_and_delete_elem(struct bpf_map *map, void *key,
 
 			pptr = htab_elem_get_ptr(l, key_size);
 			for_each_possible_cpu(cpu) {
-				bpf_long_memcpy(value + off,
-						per_cpu_ptr(pptr, cpu),
-						roundup_value_size);
+				copy_map_value_long(&htab->map, value + off, per_cpu_ptr(pptr, cpu));
+				check_and_init_map_value(&htab->map, value + off);
 				off += roundup_value_size;
 			}
 		} else {
@@ -1772,8 +1787,8 @@ again_nocopy:
 
 			pptr = htab_elem_get_ptr(l, map->key_size);
 			for_each_possible_cpu(cpu) {
-				bpf_long_memcpy(dst_val + off,
-						per_cpu_ptr(pptr, cpu), size);
+				copy_map_value_long(&htab->map, dst_val + off, per_cpu_ptr(pptr, cpu));
+				check_and_init_map_value(&htab->map, dst_val + off);
 				off += size;
 			}
 		} else {
@@ -2046,9 +2061,9 @@ static int __bpf_hash_map_seq_show(struct seq_file *seq, struct htab_elem *elem)
 				roundup_value_size = round_up(map->value_size, 8);
 				pptr = htab_elem_get_ptr(elem, map->key_size);
 				for_each_possible_cpu(cpu) {
-					bpf_long_memcpy(info->percpu_value_buf + off,
-							per_cpu_ptr(pptr, cpu),
-							roundup_value_size);
+					copy_map_value_long(map, info->percpu_value_buf + off,
+							    per_cpu_ptr(pptr, cpu));
+					check_and_init_map_value(map, info->percpu_value_buf + off);
 					off += roundup_value_size;
 				}
 				ctx.value = info->percpu_value_buf;
@@ -2292,8 +2307,8 @@ int bpf_percpu_hash_copy(struct bpf_map *map, void *key, void *value)
 	 */
 	pptr = htab_elem_get_ptr(l, map->key_size);
 	for_each_possible_cpu(cpu) {
-		bpf_long_memcpy(value + off,
-				per_cpu_ptr(pptr, cpu), size);
+		copy_map_value_long(map, value + off, per_cpu_ptr(pptr, cpu));
+		check_and_init_map_value(map, value + off);
 		off += size;
 	}
 	ret = 0;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index e3fcdc9836a6..da117a2a83b2 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1059,7 +1059,9 @@ static int map_check_btf(struct bpf_map *map, const struct btf *btf,
 			case BPF_KPTR_UNREF:
 			case BPF_KPTR_REF:
 				if (map->map_type != BPF_MAP_TYPE_HASH &&
+				    map->map_type != BPF_MAP_TYPE_PERCPU_HASH &&
 				    map->map_type != BPF_MAP_TYPE_LRU_HASH &&
+				    map->map_type != BPF_MAP_TYPE_LRU_PERCPU_HASH &&
 				    map->map_type != BPF_MAP_TYPE_ARRAY &&
 				    map->map_type != BPF_MAP_TYPE_PERCPU_ARRAY) {
 					ret = -EOPNOTSUPP;
-- 
cgit v1.2.3-70-g09d2


From 9db44fdd8105da00669d425acab887c668df75f6 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Sat, 25 Feb 2023 16:40:09 +0100
Subject: bpf: Support kptrs in local storage maps

Enable support for kptrs in local storage maps by wiring up the freeing
of these kptrs from map value. Freeing of bpf_local_storage_map is only
delayed in case there are special fields, therefore bpf_selem_free_*
path can also only dereference smap safely in that case. This is
recorded using a bool utilizing a hole in bpF_local_storage_elem. It
could have been tagged in the pointer value smap using the lowest bit
(since alignment > 1), but since there was already a hole I went with
the simpler option. Only the map structure freeing is delayed using RCU
barriers, as the buckets aren't used when selem is being freed, so they
can be freed once all readers of the bucket lists can no longer access
it.

Cc: Martin KaFai Lau <martin.lau@kernel.org>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230225154010.391965-3-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h |  6 +++++
 kernel/bpf/bpf_local_storage.c    | 48 +++++++++++++++++++++++++++++++++++----
 kernel/bpf/syscall.c              |  6 ++++-
 kernel/bpf/verifier.c             | 12 ++++++----
 4 files changed, 63 insertions(+), 9 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 6d37a40cd90e..0fe92986412b 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -74,6 +74,12 @@ struct bpf_local_storage_elem {
 	struct hlist_node snode;	/* Linked to bpf_local_storage */
 	struct bpf_local_storage __rcu *local_storage;
 	struct rcu_head rcu;
+	bool can_use_smap; /* Is it safe to access smap in bpf_selem_free_* RCU
+			    * callbacks? bpf_local_storage_map_free only
+			    * executes rcu_barrier when there are special
+			    * fields, this field remembers that to ensure we
+			    * don't access already freed smap in sdata.
+			    */
 	/* 8 bytes hole */
 	/* The data is stored in another cacheline to minimize
 	 * the number of cachelines access during a cache hit.
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 58da17ae5124..2bdd722fe293 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -85,6 +85,7 @@ bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner,
 	if (selem) {
 		if (value)
 			copy_map_value(&smap->map, SDATA(selem)->data, value);
+		/* No need to call check_and_init_map_value as memory is zero init */
 		return selem;
 	}
 
@@ -113,10 +114,25 @@ static void bpf_selem_free_rcu(struct rcu_head *rcu)
 	struct bpf_local_storage_elem *selem;
 
 	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
+	/* The can_use_smap bool is set whenever we need to free additional
+	 * fields in selem data before freeing selem. bpf_local_storage_map_free
+	 * only executes rcu_barrier to wait for RCU callbacks when it has
+	 * special fields, hence we can only conditionally dereference smap, as
+	 * by this time the map might have already been freed without waiting
+	 * for our call_rcu callback if it did not have any special fields.
+	 */
+	if (selem->can_use_smap)
+		bpf_obj_free_fields(SDATA(selem)->smap->map.record, SDATA(selem)->data);
+	kfree(selem);
+}
+
+static void bpf_selem_free_tasks_trace_rcu(struct rcu_head *rcu)
+{
+	/* Free directly if Tasks Trace RCU GP also implies RCU GP */
 	if (rcu_trace_implies_rcu_gp())
-		kfree(selem);
+		bpf_selem_free_rcu(rcu);
 	else
-		kfree_rcu(selem, rcu);
+		call_rcu(rcu, bpf_selem_free_rcu);
 }
 
 /* local_storage->lock must be held and selem->local_storage == local_storage.
@@ -170,9 +186,9 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 		RCU_INIT_POINTER(local_storage->cache[smap->cache_idx], NULL);
 
 	if (use_trace_rcu)
-		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_rcu);
+		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_tasks_trace_rcu);
 	else
-		kfree_rcu(selem, rcu);
+		call_rcu(&selem->rcu, bpf_selem_free_rcu);
 
 	return free_local_storage;
 }
@@ -240,6 +256,11 @@ void bpf_selem_link_map(struct bpf_local_storage_map *smap,
 	RCU_INIT_POINTER(SDATA(selem)->smap, smap);
 	hlist_add_head_rcu(&selem->map_node, &b->list);
 	raw_spin_unlock_irqrestore(&b->lock, flags);
+
+	/* If our data will have special fields, smap will wait for us to use
+	 * its record in bpf_selem_free_* RCU callbacks before freeing itself.
+	 */
+	selem->can_use_smap = !IS_ERR_OR_NULL(smap->map.record);
 }
 
 void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu)
@@ -723,6 +744,25 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 	 */
 	synchronize_rcu();
 
+	/* Only delay freeing of smap, buckets are not needed anymore */
 	kvfree(smap->buckets);
+
+	/* When local storage has special fields, callbacks for
+	 * bpf_selem_free_rcu and bpf_selem_free_tasks_trace_rcu will keep using
+	 * the map BTF record, we need to execute an RCU barrier to wait for
+	 * them as the record will be freed right after our map_free callback.
+	 */
+	if (!IS_ERR_OR_NULL(smap->map.record)) {
+		rcu_barrier_tasks_trace();
+		/* We cannot skip rcu_barrier() when rcu_trace_implies_rcu_gp()
+		 * is true, because while call_rcu invocation is skipped in that
+		 * case in bpf_selem_free_tasks_trace_rcu (and all local storage
+		 * maps pass use_trace_rcu = true), there can be call_rcu
+		 * callbacks based on use_trace_rcu = false in the earlier while
+		 * ((selem = ...)) loop or from bpf_local_storage_unlink_nolock
+		 * called from owner's free path.
+		 */
+		rcu_barrier();
+	}
 	bpf_map_area_free(smap);
 }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index da117a2a83b2..eb50025b03c1 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1063,7 +1063,11 @@ static int map_check_btf(struct bpf_map *map, const struct btf *btf,
 				    map->map_type != BPF_MAP_TYPE_LRU_HASH &&
 				    map->map_type != BPF_MAP_TYPE_LRU_PERCPU_HASH &&
 				    map->map_type != BPF_MAP_TYPE_ARRAY &&
-				    map->map_type != BPF_MAP_TYPE_PERCPU_ARRAY) {
+				    map->map_type != BPF_MAP_TYPE_PERCPU_ARRAY &&
+				    map->map_type != BPF_MAP_TYPE_SK_STORAGE &&
+				    map->map_type != BPF_MAP_TYPE_INODE_STORAGE &&
+				    map->map_type != BPF_MAP_TYPE_TASK_STORAGE &&
+				    map->map_type != BPF_MAP_TYPE_CGRP_STORAGE) {
 					ret = -EOPNOTSUPP;
 					goto free_map_tab;
 				}
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index a856896e835a..bf580f246a01 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -7222,22 +7222,26 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 		break;
 	case BPF_MAP_TYPE_SK_STORAGE:
 		if (func_id != BPF_FUNC_sk_storage_get &&
-		    func_id != BPF_FUNC_sk_storage_delete)
+		    func_id != BPF_FUNC_sk_storage_delete &&
+		    func_id != BPF_FUNC_kptr_xchg)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_INODE_STORAGE:
 		if (func_id != BPF_FUNC_inode_storage_get &&
-		    func_id != BPF_FUNC_inode_storage_delete)
+		    func_id != BPF_FUNC_inode_storage_delete &&
+		    func_id != BPF_FUNC_kptr_xchg)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_TASK_STORAGE:
 		if (func_id != BPF_FUNC_task_storage_get &&
-		    func_id != BPF_FUNC_task_storage_delete)
+		    func_id != BPF_FUNC_task_storage_delete &&
+		    func_id != BPF_FUNC_kptr_xchg)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_CGRP_STORAGE:
 		if (func_id != BPF_FUNC_cgrp_storage_get &&
-		    func_id != BPF_FUNC_cgrp_storage_delete)
+		    func_id != BPF_FUNC_cgrp_storage_delete &&
+		    func_id != BPF_FUNC_kptr_xchg)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_BLOOM_FILTER:
-- 
cgit v1.2.3-70-g09d2


From 85521e1ea4d0d7d8e62bbb0999f91e31ae421d76 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Sat, 25 Feb 2023 16:40:10 +0100
Subject: selftests/bpf: Add more tests for kptrs in maps

Firstly, ensure programs successfully load when using all of the
supported maps. Then, extend existing tests to test more cases at
runtime. We are currently testing both the synchronous freeing of items
and asynchronous destruction when map is freed, but the code needs to be
adjusted a bit to be able to also accomodate support for percpu maps.

We now do a delete on the item (and update for array maps which has a
similar effect for kptrs) to perform a synchronous free of the kptr, and
test destruction both for the synchronous and asynchronous deletion.
Next time the program runs, it should observe the refcount as 1 since
all existing references should have been released by then. By running
the program after both possible paths freeing kptrs, we establish that
they correctly release resources. Next, we augment the existing test to
also test the same code path shared by all local storage maps using a
task local storage map.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230225154010.391965-4-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/map_kptr.c  | 136 ++++++--
 tools/testing/selftests/bpf/progs/map_kptr.c       | 344 ++++++++++++++++++---
 .../selftests/bpf/progs/rcu_tasks_trace_gp.c       |  36 +++
 3 files changed, 451 insertions(+), 65 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/rcu_tasks_trace_gp.c

diff --git a/tools/testing/selftests/bpf/prog_tests/map_kptr.c b/tools/testing/selftests/bpf/prog_tests/map_kptr.c
index 3533a4ecad01..8743df599567 100644
--- a/tools/testing/selftests/bpf/prog_tests/map_kptr.c
+++ b/tools/testing/selftests/bpf/prog_tests/map_kptr.c
@@ -4,70 +4,160 @@
 
 #include "map_kptr.skel.h"
 #include "map_kptr_fail.skel.h"
+#include "rcu_tasks_trace_gp.skel.h"
 
 static void test_map_kptr_success(bool test_run)
 {
+	LIBBPF_OPTS(bpf_test_run_opts, lopts);
 	LIBBPF_OPTS(bpf_test_run_opts, opts,
 		.data_in = &pkt_v4,
 		.data_size_in = sizeof(pkt_v4),
 		.repeat = 1,
 	);
+	int key = 0, ret, cpu;
 	struct map_kptr *skel;
-	int key = 0, ret;
-	char buf[16];
+	char buf[16], *pbuf;
 
 	skel = map_kptr__open_and_load();
 	if (!ASSERT_OK_PTR(skel, "map_kptr__open_and_load"))
 		return;
 
-	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref), &opts);
-	ASSERT_OK(ret, "test_map_kptr_ref refcount");
-	ASSERT_OK(opts.retval, "test_map_kptr_ref retval");
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref1), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref1 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref1 retval");
 	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref2), &opts);
 	ASSERT_OK(ret, "test_map_kptr_ref2 refcount");
 	ASSERT_OK(opts.retval, "test_map_kptr_ref2 retval");
 
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_ls_map_kptr_ref1), &lopts);
+	ASSERT_OK(ret, "test_ls_map_kptr_ref1 refcount");
+	ASSERT_OK(lopts.retval, "test_ls_map_kptr_ref1 retval");
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_ls_map_kptr_ref2), &lopts);
+	ASSERT_OK(ret, "test_ls_map_kptr_ref2 refcount");
+	ASSERT_OK(lopts.retval, "test_ls_map_kptr_ref2 retval");
+
 	if (test_run)
 		goto exit;
 
+	cpu = libbpf_num_possible_cpus();
+	if (!ASSERT_GT(cpu, 0, "libbpf_num_possible_cpus"))
+		goto exit;
+
+	pbuf = calloc(cpu, sizeof(buf));
+	if (!ASSERT_OK_PTR(pbuf, "calloc(pbuf)"))
+		goto exit;
+
 	ret = bpf_map__update_elem(skel->maps.array_map,
 				   &key, sizeof(key), buf, sizeof(buf), 0);
 	ASSERT_OK(ret, "array_map update");
-	ret = bpf_map__update_elem(skel->maps.array_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "array_map update2");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__update_elem(skel->maps.pcpu_array_map,
+				   &key, sizeof(key), pbuf, cpu * sizeof(buf), 0);
+	ASSERT_OK(ret, "pcpu_array_map update");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
-	ret = bpf_map__update_elem(skel->maps.hash_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "hash_map update");
 	ret = bpf_map__delete_elem(skel->maps.hash_map, &key, sizeof(key), 0);
 	ASSERT_OK(ret, "hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__delete_elem(skel->maps.pcpu_hash_map, &key, sizeof(key), 0);
+	ASSERT_OK(ret, "pcpu_hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
-	ret = bpf_map__update_elem(skel->maps.hash_malloc_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "hash_malloc_map update");
 	ret = bpf_map__delete_elem(skel->maps.hash_malloc_map, &key, sizeof(key), 0);
 	ASSERT_OK(ret, "hash_malloc_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__delete_elem(skel->maps.pcpu_hash_malloc_map, &key, sizeof(key), 0);
+	ASSERT_OK(ret, "pcpu_hash_malloc_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
-	ret = bpf_map__update_elem(skel->maps.lru_hash_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "lru_hash_map update");
 	ret = bpf_map__delete_elem(skel->maps.lru_hash_map, &key, sizeof(key), 0);
 	ASSERT_OK(ret, "lru_hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__delete_elem(skel->maps.lru_pcpu_hash_map, &key, sizeof(key), 0);
+	ASSERT_OK(ret, "lru_pcpu_hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_ls_map_kptr_ref_del), &lopts);
+	ASSERT_OK(ret, "test_ls_map_kptr_ref_del delete");
+	skel->data->ref--;
+	ASSERT_OK(lopts.retval, "test_ls_map_kptr_ref_del retval");
+
+	free(pbuf);
 exit:
 	map_kptr__destroy(skel);
 }
 
-void test_map_kptr(void)
+static int kern_sync_rcu_tasks_trace(struct rcu_tasks_trace_gp *rcu)
 {
-	if (test__start_subtest("success")) {
+	long gp_seq = READ_ONCE(rcu->bss->gp_seq);
+	LIBBPF_OPTS(bpf_test_run_opts, opts);
+
+	if (!ASSERT_OK(bpf_prog_test_run_opts(bpf_program__fd(rcu->progs.do_call_rcu_tasks_trace),
+					      &opts), "do_call_rcu_tasks_trace"))
+		return -EFAULT;
+	if (!ASSERT_OK(opts.retval, "opts.retval == 0"))
+		return -EFAULT;
+	while (gp_seq == READ_ONCE(rcu->bss->gp_seq))
+		sched_yield();
+	return 0;
+}
+
+void serial_test_map_kptr(void)
+{
+	struct rcu_tasks_trace_gp *skel;
+
+	RUN_TESTS(map_kptr_fail);
+
+	skel = rcu_tasks_trace_gp__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "rcu_tasks_trace_gp__open_and_load"))
+		return;
+	if (!ASSERT_OK(rcu_tasks_trace_gp__attach(skel), "rcu_tasks_trace_gp__attach"))
+		goto end;
+
+	if (test__start_subtest("success-map")) {
+		test_map_kptr_success(true);
+
+		ASSERT_OK(kern_sync_rcu_tasks_trace(skel), "sync rcu_tasks_trace");
+		ASSERT_OK(kern_sync_rcu(), "sync rcu");
+		/* Observe refcount dropping to 1 on bpf_map_free_deferred */
 		test_map_kptr_success(false);
-		/* Do test_run twice, so that we see refcount going back to 1
-		 * after we leave it in map from first iteration.
-		 */
+
+		ASSERT_OK(kern_sync_rcu_tasks_trace(skel), "sync rcu_tasks_trace");
+		ASSERT_OK(kern_sync_rcu(), "sync rcu");
+		/* Observe refcount dropping to 1 on synchronous delete elem */
 		test_map_kptr_success(true);
 	}
 
-	RUN_TESTS(map_kptr_fail);
+end:
+	rcu_tasks_trace_gp__destroy(skel);
+	return;
 }
diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c
index 228ec45365a8..a24d17bc17eb 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr.c
@@ -15,6 +15,13 @@ struct array_map {
 	__uint(max_entries, 1);
 } array_map SEC(".maps");
 
+struct pcpu_array_map {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} pcpu_array_map SEC(".maps");
+
 struct hash_map {
 	__uint(type, BPF_MAP_TYPE_HASH);
 	__type(key, int);
@@ -22,6 +29,13 @@ struct hash_map {
 	__uint(max_entries, 1);
 } hash_map SEC(".maps");
 
+struct pcpu_hash_map {
+	__uint(type, BPF_MAP_TYPE_PERCPU_HASH);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} pcpu_hash_map SEC(".maps");
+
 struct hash_malloc_map {
 	__uint(type, BPF_MAP_TYPE_HASH);
 	__type(key, int);
@@ -30,6 +44,14 @@ struct hash_malloc_map {
 	__uint(map_flags, BPF_F_NO_PREALLOC);
 } hash_malloc_map SEC(".maps");
 
+struct pcpu_hash_malloc_map {
+	__uint(type, BPF_MAP_TYPE_PERCPU_HASH);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+} pcpu_hash_malloc_map SEC(".maps");
+
 struct lru_hash_map {
 	__uint(type, BPF_MAP_TYPE_LRU_HASH);
 	__type(key, int);
@@ -37,6 +59,41 @@ struct lru_hash_map {
 	__uint(max_entries, 1);
 } lru_hash_map SEC(".maps");
 
+struct lru_pcpu_hash_map {
+	__uint(type, BPF_MAP_TYPE_LRU_PERCPU_HASH);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} lru_pcpu_hash_map SEC(".maps");
+
+struct cgrp_ls_map {
+	__uint(type, BPF_MAP_TYPE_CGRP_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} cgrp_ls_map SEC(".maps");
+
+struct task_ls_map {
+	__uint(type, BPF_MAP_TYPE_TASK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} task_ls_map SEC(".maps");
+
+struct inode_ls_map {
+	__uint(type, BPF_MAP_TYPE_INODE_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} inode_ls_map SEC(".maps");
+
+struct sk_ls_map {
+	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} sk_ls_map SEC(".maps");
+
 #define DEFINE_MAP_OF_MAP(map_type, inner_map_type, name)       \
 	struct {                                                \
 		__uint(type, map_type);                         \
@@ -160,6 +217,58 @@ int test_map_kptr(struct __sk_buff *ctx)
 	return 0;
 }
 
+SEC("tp_btf/cgroup_mkdir")
+int BPF_PROG(test_cgrp_map_kptr, struct cgroup *cgrp, const char *path)
+{
+	struct map_value *v;
+
+	v = bpf_cgrp_storage_get(&cgrp_ls_map, cgrp, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
+SEC("lsm/inode_unlink")
+int BPF_PROG(test_task_map_kptr, struct inode *inode, struct dentry *victim)
+{
+	struct task_struct *task;
+	struct map_value *v;
+
+	task = bpf_get_current_task_btf();
+	if (!task)
+		return 0;
+	v = bpf_task_storage_get(&task_ls_map, task, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
+SEC("lsm/inode_unlink")
+int BPF_PROG(test_inode_map_kptr, struct inode *inode, struct dentry *victim)
+{
+	struct map_value *v;
+
+	v = bpf_inode_storage_get(&inode_ls_map, inode, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
+SEC("tc")
+int test_sk_map_kptr(struct __sk_buff *ctx)
+{
+	struct map_value *v;
+	struct bpf_sock *sk;
+
+	sk = ctx->sk;
+	if (!sk)
+		return 0;
+	v = bpf_sk_storage_get(&sk_ls_map, sk, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
 SEC("tc")
 int test_map_in_map_kptr(struct __sk_buff *ctx)
 {
@@ -189,106 +298,257 @@ int test_map_in_map_kptr(struct __sk_buff *ctx)
 	return 0;
 }
 
-SEC("tc")
-int test_map_kptr_ref(struct __sk_buff *ctx)
+int ref = 1;
+
+static __always_inline
+int test_map_kptr_ref_pre(struct map_value *v)
 {
 	struct prog_test_ref_kfunc *p, *p_st;
 	unsigned long arg = 0;
-	struct map_value *v;
-	int key = 0, ret;
+	int ret;
 
 	p = bpf_kfunc_call_test_acquire(&arg);
 	if (!p)
 		return 1;
+	ref++;
 
 	p_st = p->next;
-	if (p_st->cnt.refs.counter != 2) {
+	if (p_st->cnt.refs.counter != ref) {
 		ret = 2;
 		goto end;
 	}
 
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v) {
-		ret = 3;
-		goto end;
-	}
-
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
-		ret = 4;
+		ret = 3;
 		goto end;
 	}
-	if (p_st->cnt.refs.counter != 2)
-		return 5;
+	if (p_st->cnt.refs.counter != ref)
+		return 4;
 
 	p = bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0);
 	if (!p)
-		return 6;
-	if (p_st->cnt.refs.counter != 3) {
-		ret = 7;
+		return 5;
+	ref++;
+	if (p_st->cnt.refs.counter != ref) {
+		ret = 6;
 		goto end;
 	}
 	bpf_kfunc_call_test_release(p);
-	if (p_st->cnt.refs.counter != 2)
-		return 8;
+	ref--;
+	if (p_st->cnt.refs.counter != ref)
+		return 7;
 
 	p = bpf_kptr_xchg(&v->ref_ptr, NULL);
 	if (!p)
-		return 9;
+		return 8;
 	bpf_kfunc_call_test_release(p);
-	if (p_st->cnt.refs.counter != 1)
-		return 10;
+	ref--;
+	if (p_st->cnt.refs.counter != ref)
+		return 9;
 
 	p = bpf_kfunc_call_test_acquire(&arg);
 	if (!p)
-		return 11;
+		return 10;
+	ref++;
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
-		ret = 12;
+		ret = 11;
 		goto end;
 	}
-	if (p_st->cnt.refs.counter != 2)
-		return 13;
+	if (p_st->cnt.refs.counter != ref)
+		return 12;
 	/* Leave in map */
 
 	return 0;
 end:
+	ref--;
 	bpf_kfunc_call_test_release(p);
 	return ret;
 }
 
-SEC("tc")
-int test_map_kptr_ref2(struct __sk_buff *ctx)
+static __always_inline
+int test_map_kptr_ref_post(struct map_value *v)
 {
 	struct prog_test_ref_kfunc *p, *p_st;
-	struct map_value *v;
-	int key = 0;
-
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v)
-		return 1;
 
 	p_st = v->ref_ptr;
-	if (!p_st || p_st->cnt.refs.counter != 2)
-		return 2;
+	if (!p_st || p_st->cnt.refs.counter != ref)
+		return 1;
 
 	p = bpf_kptr_xchg(&v->ref_ptr, NULL);
 	if (!p)
-		return 3;
-	if (p_st->cnt.refs.counter != 2) {
+		return 2;
+	if (p_st->cnt.refs.counter != ref) {
 		bpf_kfunc_call_test_release(p);
-		return 4;
+		return 3;
 	}
 
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
 		bpf_kfunc_call_test_release(p);
-		return 5;
+		return 4;
 	}
-	if (p_st->cnt.refs.counter != 2)
-		return 6;
+	if (p_st->cnt.refs.counter != ref)
+		return 5;
+
+	return 0;
+}
+
+#define TEST(map)                            \
+	v = bpf_map_lookup_elem(&map, &key); \
+	if (!v)                              \
+		return -1;                   \
+	ret = test_map_kptr_ref_pre(v);      \
+	if (ret)                             \
+		return ret;
+
+#define TEST_PCPU(map)                                 \
+	v = bpf_map_lookup_percpu_elem(&map, &key, 0); \
+	if (!v)                                        \
+		return -1;                             \
+	ret = test_map_kptr_ref_pre(v);                \
+	if (ret)                                       \
+		return ret;
+
+SEC("tc")
+int test_map_kptr_ref1(struct __sk_buff *ctx)
+{
+	struct map_value *v, val = {};
+	int key = 0, ret;
+
+	bpf_map_update_elem(&hash_map, &key, &val, 0);
+	bpf_map_update_elem(&hash_malloc_map, &key, &val, 0);
+	bpf_map_update_elem(&lru_hash_map, &key, &val, 0);
+
+	bpf_map_update_elem(&pcpu_hash_map, &key, &val, 0);
+	bpf_map_update_elem(&pcpu_hash_malloc_map, &key, &val, 0);
+	bpf_map_update_elem(&lru_pcpu_hash_map, &key, &val, 0);
+
+	TEST(array_map);
+	TEST(hash_map);
+	TEST(hash_malloc_map);
+	TEST(lru_hash_map);
+
+	TEST_PCPU(pcpu_array_map);
+	TEST_PCPU(pcpu_hash_map);
+	TEST_PCPU(pcpu_hash_malloc_map);
+	TEST_PCPU(lru_pcpu_hash_map);
+
+	return 0;
+}
+
+#undef TEST
+#undef TEST_PCPU
+
+#define TEST(map)                            \
+	v = bpf_map_lookup_elem(&map, &key); \
+	if (!v)                              \
+		return -1;                   \
+	ret = test_map_kptr_ref_post(v);     \
+	if (ret)                             \
+		return ret;
+
+#define TEST_PCPU(map)                                 \
+	v = bpf_map_lookup_percpu_elem(&map, &key, 0); \
+	if (!v)                                        \
+		return -1;                             \
+	ret = test_map_kptr_ref_post(v);               \
+	if (ret)                                       \
+		return ret;
+
+SEC("tc")
+int test_map_kptr_ref2(struct __sk_buff *ctx)
+{
+	struct map_value *v;
+	int key = 0, ret;
+
+	TEST(array_map);
+	TEST(hash_map);
+	TEST(hash_malloc_map);
+	TEST(lru_hash_map);
+
+	TEST_PCPU(pcpu_array_map);
+	TEST_PCPU(pcpu_hash_map);
+	TEST_PCPU(pcpu_hash_malloc_map);
+	TEST_PCPU(lru_pcpu_hash_map);
 
 	return 0;
 }
 
+#undef TEST
+#undef TEST_PCPU
+
+SEC("tc")
+int test_map_kptr_ref3(struct __sk_buff *ctx)
+{
+	struct prog_test_ref_kfunc *p;
+	unsigned long sp = 0;
+
+	p = bpf_kfunc_call_test_acquire(&sp);
+	if (!p)
+		return 1;
+	ref++;
+	if (p->cnt.refs.counter != ref) {
+		bpf_kfunc_call_test_release(p);
+		return 2;
+	}
+	bpf_kfunc_call_test_release(p);
+	ref--;
+	return 0;
+}
+
+SEC("syscall")
+int test_ls_map_kptr_ref1(void *ctx)
+{
+	struct task_struct *current;
+	struct map_value *v;
+	int ret;
+
+	current = bpf_get_current_task_btf();
+	if (!current)
+		return 100;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, 0);
+	if (v)
+		return 150;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (!v)
+		return 200;
+	return test_map_kptr_ref_pre(v);
+}
+
+SEC("syscall")
+int test_ls_map_kptr_ref2(void *ctx)
+{
+	struct task_struct *current;
+	struct map_value *v;
+	int ret;
+
+	current = bpf_get_current_task_btf();
+	if (!current)
+		return 100;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, 0);
+	if (!v)
+		return 200;
+	return test_map_kptr_ref_post(v);
+}
+
+SEC("syscall")
+int test_ls_map_kptr_ref_del(void *ctx)
+{
+	struct task_struct *current;
+	struct map_value *v;
+	int ret;
+
+	current = bpf_get_current_task_btf();
+	if (!current)
+		return 100;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, 0);
+	if (!v)
+		return 200;
+	if (!v->ref_ptr)
+		return 300;
+	return bpf_task_storage_delete(&task_ls_map, current);
+}
+
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/rcu_tasks_trace_gp.c b/tools/testing/selftests/bpf/progs/rcu_tasks_trace_gp.c
new file mode 100644
index 000000000000..df4873558634
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/rcu_tasks_trace_gp.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+
+struct task_ls_map {
+	__uint(type, BPF_MAP_TYPE_TASK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, int);
+} task_ls_map SEC(".maps");
+
+long gp_seq;
+
+SEC("syscall")
+int do_call_rcu_tasks_trace(void *ctx)
+{
+    struct task_struct *current;
+    int *v;
+
+    current = bpf_get_current_task_btf();
+    v = bpf_task_storage_get(&task_ls_map, current, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+    if (!v)
+        return 1;
+    /* Invoke call_rcu_tasks_trace */
+    return bpf_task_storage_delete(&task_ls_map, current);
+}
+
+SEC("kprobe/rcu_tasks_trace_postgp")
+int rcu_tasks_trace_postgp(void *ctx)
+{
+    __sync_add_and_fetch(&gp_seq, 1);
+    return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From be35f4af719c94df137cd611bf497d658eb3adc2 Mon Sep 17 00:00:00 2001
From: Tiezhu Yang <yangtiezhu@loongson.cn>
Date: Tue, 28 Feb 2023 20:03:01 +0800
Subject: selftests/bpf: Set __BITS_PER_LONG if target is bpf for LoongArch

If target is bpf, there is no __loongarch__ definition, __BITS_PER_LONG
defaults to 32, __NR_nanosleep is not defined:

  #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32
  #define __NR_nanosleep 101
  __SC_3264(__NR_nanosleep, sys_nanosleep_time32, sys_nanosleep)
  #endif

Work around this problem, by explicitly setting __BITS_PER_LONG to
__loongarch_grlen which is defined by compiler as 64 for LA64.

This is similar with commit 36e70b9b06bf ("selftests, bpf: Fix broken
riscv build").

Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/1677585781-21628-1-git-send-email-yangtiezhu@loongson.cn
---
 tools/testing/selftests/bpf/Makefile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index b677dcd0b77a..f40606a85a0f 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -338,7 +338,8 @@ $(RESOLVE_BTFIDS): $(HOST_BPFOBJ) | $(HOST_BUILD_DIR)/resolve_btfids	\
 define get_sys_includes
 $(shell $(1) $(2) -v -E - </dev/null 2>&1 \
 	| sed -n '/<...> search starts here:/,/End of search list./{ s| \(/.*\)|-idirafter \1|p }') \
-$(shell $(1) $(2) -dM -E - </dev/null | grep '__riscv_xlen ' | awk '{printf("-D__riscv_xlen=%d -D__BITS_PER_LONG=%d", $$3, $$3)}')
+$(shell $(1) $(2) -dM -E - </dev/null | grep '__riscv_xlen ' | awk '{printf("-D__riscv_xlen=%d -D__BITS_PER_LONG=%d", $$3, $$3)}') \
+$(shell $(1) $(2) -dM -E - </dev/null | grep '__loongarch_grlen ' | awk '{printf("-D__BITS_PER_LONG=%d", $$3)}')
 endef
 
 # Determine target endianness.
-- 
cgit v1.2.3-70-g09d2


From 40e1bcab1e4ce7d7e4d601782a963e8d59d09695 Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Wed, 1 Mar 2023 09:53:53 +0100
Subject: libbpf: Remove unnecessary ternary operator

Coverity reports that the first check of 'err' in bpf_object__init_maps
is always false as 'err' is initialized to 0 at that point. Remove the
unnecessary ternary operator.

Signed-off-by: Viktor Malik <vmalik@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/78a3702f2ea9f32a84faaae9b674c56269d330a7.1677658777.git.vmalik@redhat.com
---
 tools/lib/bpf/libbpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 05c4db355f28..905193d98885 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -2615,7 +2615,7 @@ static int bpf_object__init_maps(struct bpf_object *obj,
 	strict = !OPTS_GET(opts, relaxed_maps, false);
 	pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
 
-	err = err ?: bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
+	err = bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
 	err = err ?: bpf_object__init_global_data_maps(obj);
 	err = err ?: bpf_object__init_kconfig_map(obj);
 	err = err ?: bpf_object__init_struct_ops_maps(obj);
-- 
cgit v1.2.3-70-g09d2


From 7832d06bd9f9080d9ceae388bb81bf33adce3850 Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Wed, 1 Mar 2023 09:53:54 +0100
Subject: libbpf: Remove several dead assignments

Clang Static Analyzer (scan-build) reports several dead assignments in
libbpf where the assigned value is unconditionally overridden by another
value before it is read. Remove these assignments.

Signed-off-by: Viktor Malik <vmalik@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/5503d18966583e55158471ebbb2f67374b11bf5e.1677658777.git.vmalik@redhat.com
---
 tools/lib/bpf/btf.c       | 2 --
 tools/lib/bpf/libbpf.c    | 1 -
 tools/lib/bpf/relo_core.c | 3 ---
 3 files changed, 6 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 9181d36118d2..0a2c079244b6 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -1000,8 +1000,6 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 		}
 	}
 
-	err = 0;
-
 	if (!btf_data) {
 		pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
 		err = -ENODATA;
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 905193d98885..ba9e7e2b7951 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -798,7 +798,6 @@ bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data,
 	progs = obj->programs;
 	nr_progs = obj->nr_programs;
 	nr_syms = symbols->d_size / sizeof(Elf64_Sym);
-	sec_off = 0;
 
 	for (i = 0; i < nr_syms; i++) {
 		sym = elf_sym_by_idx(obj, i);
diff --git a/tools/lib/bpf/relo_core.c b/tools/lib/bpf/relo_core.c
index c4b0e81ae293..a26b2f5fa0fc 100644
--- a/tools/lib/bpf/relo_core.c
+++ b/tools/lib/bpf/relo_core.c
@@ -1551,9 +1551,6 @@ int __bpf_core_types_match(const struct btf *local_btf, __u32 local_id, const st
 	if (level <= 0)
 		return -EINVAL;
 
-	local_t = btf_type_by_id(local_btf, local_id);
-	targ_t = btf_type_by_id(targ_btf, targ_id);
-
 recur:
 	depth--;
 	if (depth < 0)
-- 
cgit v1.2.3-70-g09d2


From 4672129127eed0d16ab1b4b70b4e49014e49e8bb Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Wed, 1 Mar 2023 09:53:55 +0100
Subject: libbpf: Cleanup linker_append_elf_relos

Clang Static Analyser (scan-build) reports some unused symbols and dead
assignments in the linker_append_elf_relos function. Clean these up.

Signed-off-by: Viktor Malik <vmalik@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/c5c8fe9f411b69afada8399d23bb048ef2a70535.1677658777.git.vmalik@redhat.com
---
 tools/lib/bpf/linker.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/tools/lib/bpf/linker.c b/tools/lib/bpf/linker.c
index 4ac02c28e152..d7069780984a 100644
--- a/tools/lib/bpf/linker.c
+++ b/tools/lib/bpf/linker.c
@@ -1997,7 +1997,6 @@ add_sym:
 static int linker_append_elf_relos(struct bpf_linker *linker, struct src_obj *obj)
 {
 	struct src_sec *src_symtab = &obj->secs[obj->symtab_sec_idx];
-	struct dst_sec *dst_symtab;
 	int i, err;
 
 	for (i = 1; i < obj->sec_cnt; i++) {
@@ -2030,9 +2029,6 @@ static int linker_append_elf_relos(struct bpf_linker *linker, struct src_obj *ob
 			return -1;
 		}
 
-		/* add_dst_sec() above could have invalidated linker->secs */
-		dst_symtab = &linker->secs[linker->symtab_sec_idx];
-
 		/* shdr->sh_link points to SYMTAB */
 		dst_sec->shdr->sh_link = linker->symtab_sec_idx;
 
@@ -2049,16 +2045,13 @@ static int linker_append_elf_relos(struct bpf_linker *linker, struct src_obj *ob
 		dst_rel = dst_sec->raw_data + src_sec->dst_off;
 		n = src_sec->shdr->sh_size / src_sec->shdr->sh_entsize;
 		for (j = 0; j < n; j++, src_rel++, dst_rel++) {
-			size_t src_sym_idx = ELF64_R_SYM(src_rel->r_info);
-			size_t sym_type = ELF64_R_TYPE(src_rel->r_info);
-			Elf64_Sym *src_sym, *dst_sym;
-			size_t dst_sym_idx;
+			size_t src_sym_idx, dst_sym_idx, sym_type;
+			Elf64_Sym *src_sym;
 
 			src_sym_idx = ELF64_R_SYM(src_rel->r_info);
 			src_sym = src_symtab->data->d_buf + sizeof(*src_sym) * src_sym_idx;
 
 			dst_sym_idx = obj->sym_map[src_sym_idx];
-			dst_sym = dst_symtab->raw_data + sizeof(*dst_sym) * dst_sym_idx;
 			dst_rel->r_offset += src_linked_sec->dst_off;
 			sym_type = ELF64_R_TYPE(src_rel->r_info);
 			dst_rel->r_info = ELF64_R_INFO(dst_sym_idx, sym_type);
-- 
cgit v1.2.3-70-g09d2


From 35cbf7f9156893f55b885f6fc678b2ab8b0d8918 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 1 Mar 2023 19:54:17 +0200
Subject: selftests/bpf: Support custom per-test flags and multiple expected
 messages

Extend __flag attribute by allowing to specify one of the following:
 * BPF_F_STRICT_ALIGNMENT
 * BPF_F_ANY_ALIGNMENT
 * BPF_F_TEST_RND_HI32
 * BPF_F_TEST_STATE_FREQ
 * BPF_F_SLEEPABLE
 * BPF_F_XDP_HAS_FRAGS
 * Some numeric value

Extend __msg attribute by allowing to specify multiple exepcted messages.
All messages are expected to be present in the verifier log in the
order of application.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230301175417.3146070-2-eddyz87@gmail.com

[ Eduard: added commit message, formatting, comments ]
---
 tools/testing/selftests/bpf/progs/bpf_misc.h | 23 ++++++++++
 tools/testing/selftests/bpf/test_loader.c    | 69 ++++++++++++++++++++++++----
 tools/testing/selftests/bpf/test_progs.h     |  1 +
 3 files changed, 84 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 14e28f991451..f704885aa534 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -2,10 +2,33 @@
 #ifndef __BPF_MISC_H__
 #define __BPF_MISC_H__
 
+/* This set of attributes controls behavior of the
+ * test_loader.c:test_loader__run_subtests().
+ *
+ * __msg             Message expected to be found in the verifier log.
+ *                   Multiple __msg attributes could be specified.
+ *
+ * __success         Expect program load success in privileged mode.
+ *
+ * __failure         Expect program load failure in privileged mode.
+ *
+ * __log_level       Log level to use for the program, numeric value expected.
+ *
+ * __flag            Adds one flag use for the program, the following values are valid:
+ *                   - BPF_F_STRICT_ALIGNMENT;
+ *                   - BPF_F_TEST_RND_HI32;
+ *                   - BPF_F_TEST_STATE_FREQ;
+ *                   - BPF_F_SLEEPABLE;
+ *                   - BPF_F_XDP_HAS_FRAGS;
+ *                   - A numeric value.
+ *                   Multiple __flag attributes could be specified, the final flags
+ *                   value is derived by applying binary "or" to all specified values.
+ */
 #define __msg(msg)		__attribute__((btf_decl_tag("comment:test_expect_msg=" msg)))
 #define __failure		__attribute__((btf_decl_tag("comment:test_expect_failure")))
 #define __success		__attribute__((btf_decl_tag("comment:test_expect_success")))
 #define __log_level(lvl)	__attribute__((btf_decl_tag("comment:test_log_level="#lvl)))
+#define __flag(flag)		__attribute__((btf_decl_tag("comment:test_prog_flags="#flag)))
 
 /* Convenience macro for use with 'asm volatile' blocks */
 #define __naked __attribute__((naked))
diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index 679efb3aa785..bf41390157bf 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -13,12 +13,15 @@
 #define TEST_TAG_EXPECT_SUCCESS "comment:test_expect_success"
 #define TEST_TAG_EXPECT_MSG_PFX "comment:test_expect_msg="
 #define TEST_TAG_LOG_LEVEL_PFX "comment:test_log_level="
+#define TEST_TAG_PROG_FLAGS_PFX "comment:test_prog_flags="
 
 struct test_spec {
 	const char *name;
 	bool expect_failure;
-	const char *expect_msg;
+	const char **expect_msgs;
+	size_t expect_msg_cnt;
 	int log_level;
+	int prog_flags;
 };
 
 static int tester_init(struct test_loader *tester)
@@ -67,7 +70,8 @@ static int parse_test_spec(struct test_loader *tester,
 
 	for (i = 1; i < btf__type_cnt(btf); i++) {
 		const struct btf_type *t;
-		const char *s;
+		const char *s, *val;
+		char *e;
 
 		t = btf__type_by_id(btf, i);
 		if (!btf_is_decl_tag(t))
@@ -82,14 +86,48 @@ static int parse_test_spec(struct test_loader *tester,
 		} else if (strcmp(s, TEST_TAG_EXPECT_SUCCESS) == 0) {
 			spec->expect_failure = false;
 		} else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX)) {
-			spec->expect_msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
+			void *tmp;
+			const char **msg;
+
+			tmp = realloc(spec->expect_msgs,
+				      (1 + spec->expect_msg_cnt) * sizeof(void *));
+			if (!tmp) {
+				ASSERT_FAIL("failed to realloc memory for messages\n");
+				return -ENOMEM;
+			}
+			spec->expect_msgs = tmp;
+			msg = &spec->expect_msgs[spec->expect_msg_cnt++];
+			*msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
 		} else if (str_has_pfx(s, TEST_TAG_LOG_LEVEL_PFX)) {
+			val = s + sizeof(TEST_TAG_LOG_LEVEL_PFX) - 1;
 			errno = 0;
-			spec->log_level = strtol(s + sizeof(TEST_TAG_LOG_LEVEL_PFX) - 1, NULL, 0);
-			if (errno) {
+			spec->log_level = strtol(val, &e, 0);
+			if (errno || e[0] != '\0') {
 				ASSERT_FAIL("failed to parse test log level from '%s'", s);
 				return -EINVAL;
 			}
+		} else if (str_has_pfx(s, TEST_TAG_PROG_FLAGS_PFX)) {
+			val = s + sizeof(TEST_TAG_PROG_FLAGS_PFX) - 1;
+			if (strcmp(val, "BPF_F_STRICT_ALIGNMENT") == 0) {
+				spec->prog_flags |= BPF_F_STRICT_ALIGNMENT;
+			} else if (strcmp(val, "BPF_F_ANY_ALIGNMENT") == 0) {
+				spec->prog_flags |= BPF_F_ANY_ALIGNMENT;
+			} else if (strcmp(val, "BPF_F_TEST_RND_HI32") == 0) {
+				spec->prog_flags |= BPF_F_TEST_RND_HI32;
+			} else if (strcmp(val, "BPF_F_TEST_STATE_FREQ") == 0) {
+				spec->prog_flags |= BPF_F_TEST_STATE_FREQ;
+			} else if (strcmp(val, "BPF_F_SLEEPABLE") == 0) {
+				spec->prog_flags |= BPF_F_SLEEPABLE;
+			} else if (strcmp(val, "BPF_F_XDP_HAS_FRAGS") == 0) {
+				spec->prog_flags |= BPF_F_XDP_HAS_FRAGS;
+			} else /* assume numeric value */ {
+				errno = 0;
+				spec->prog_flags |= strtol(val, &e, 0);
+				if (errno || e[0] != '\0') {
+					ASSERT_FAIL("failed to parse test prog flags from '%s'", s);
+					return -EINVAL;
+				}
+			}
 		}
 	}
 
@@ -101,7 +139,7 @@ static void prepare_case(struct test_loader *tester,
 			 struct bpf_object *obj,
 			 struct bpf_program *prog)
 {
-	int min_log_level = 0;
+	int min_log_level = 0, prog_flags;
 
 	if (env.verbosity > VERBOSE_NONE)
 		min_log_level = 1;
@@ -119,7 +157,11 @@ static void prepare_case(struct test_loader *tester,
 	else
 		bpf_program__set_log_level(prog, spec->log_level);
 
+	prog_flags = bpf_program__flags(prog);
+	bpf_program__set_flags(prog, prog_flags | spec->prog_flags);
+
 	tester->log_buf[0] = '\0';
+	tester->next_match_pos = 0;
 }
 
 static void emit_verifier_log(const char *log_buf, bool force)
@@ -135,17 +177,26 @@ static void validate_case(struct test_loader *tester,
 			  struct bpf_program *prog,
 			  int load_err)
 {
-	if (spec->expect_msg) {
+	int i, j;
+
+	for (i = 0; i < spec->expect_msg_cnt; i++) {
 		char *match;
+		const char *expect_msg;
+
+		expect_msg = spec->expect_msgs[i];
 
-		match = strstr(tester->log_buf, spec->expect_msg);
+		match = strstr(tester->log_buf + tester->next_match_pos, expect_msg);
 		if (!ASSERT_OK_PTR(match, "expect_msg")) {
 			/* if we are in verbose mode, we've already emitted log */
 			if (env.verbosity == VERBOSE_NONE)
 				emit_verifier_log(tester->log_buf, true /*force*/);
-			fprintf(stderr, "EXPECTED MSG: '%s'\n", spec->expect_msg);
+			for (j = 0; j < i; j++)
+				fprintf(stderr, "MATCHED  MSG: '%s'\n", spec->expect_msgs[j]);
+			fprintf(stderr, "EXPECTED MSG: '%s'\n", expect_msg);
 			return;
 		}
+
+		tester->next_match_pos = match - tester->log_buf + strlen(expect_msg);
 	}
 }
 
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 9fbdc57c5b57..3cbf005747ed 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -427,6 +427,7 @@ int get_bpf_max_tramp_links(void);
 struct test_loader {
 	char *log_buf;
 	size_t log_buf_sz;
+	size_t next_match_pos;
 
 	struct bpf_object *obj;
 };
-- 
cgit v1.2.3-70-g09d2


From 7ce60b110eece1d7b3d5c322fd11f6d41a29d17b Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Wed, 1 Mar 2023 13:49:09 -0600
Subject: bpf: Fix doxygen comments for dynptr slice kfuncs

In commit 66e3a13e7c2c ("bpf: Add bpf_dynptr_slice and
bpf_dynptr_slice_rdwr"), the bpf_dynptr_slice() and
bpf_dynptr_slice_rdwr() kfuncs were added to BPF. These kfuncs included
doxygen headers, but unfortunately those headers are not properly
formatted according to [0], and causes the following warnings during the
docs build:

./kernel/bpf/helpers.c:2225: warning: \
    Excess function parameter 'returns' description in 'bpf_dynptr_slice'
./kernel/bpf/helpers.c:2303: warning: \
    Excess function parameter 'returns' description in 'bpf_dynptr_slice_rdwr'
...

This patch fixes those doxygen comments.

[0]: https://docs.kernel.org/doc-guide/kernel-doc.html#function-documentation

Fixes: 66e3a13e7c2c ("bpf: Add bpf_dynptr_slice and bpf_dynptr_slice_rdwr")
Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230301194910.602738-1-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 648b29e78b84..58431a92bb65 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2194,7 +2194,12 @@ __bpf_kfunc struct task_struct *bpf_task_from_pid(s32 pid)
 }
 
 /**
- * bpf_dynptr_slice - Obtain a read-only pointer to the dynptr data.
+ * bpf_dynptr_slice() - Obtain a read-only pointer to the dynptr data.
+ * @ptr: The dynptr whose data slice to retrieve
+ * @offset: Offset into the dynptr
+ * @buffer: User-provided buffer to copy contents into
+ * @buffer__szk: Size (in bytes) of the buffer. This is the length of the
+ *		 requested slice. This must be a constant.
  *
  * For non-skb and non-xdp type dynptrs, there is no difference between
  * bpf_dynptr_slice and bpf_dynptr_data.
@@ -2209,13 +2214,7 @@ __bpf_kfunc struct task_struct *bpf_task_from_pid(s32 pid)
  * bpf_dynptr_slice will not invalidate any ctx->data/data_end pointers in
  * the bpf program.
  *
- * @ptr: The dynptr whose data slice to retrieve
- * @offset: Offset into the dynptr
- * @buffer: User-provided buffer to copy contents into
- * @buffer__szk: Size (in bytes) of the buffer. This is the length of the
- * requested slice. This must be a constant.
- *
- * @returns: NULL if the call failed (eg invalid dynptr), pointer to a read-only
+ * Return: NULL if the call failed (eg invalid dynptr), pointer to a read-only
  * data slice (can be either direct pointer to the data or a pointer to the user
  * provided buffer, with its contents containing the data, if unable to obtain
  * direct pointer)
@@ -2258,7 +2257,12 @@ __bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset
 }
 
 /**
- * bpf_dynptr_slice_rdwr - Obtain a writable pointer to the dynptr data.
+ * bpf_dynptr_slice_rdwr() - Obtain a writable pointer to the dynptr data.
+ * @ptr: The dynptr whose data slice to retrieve
+ * @offset: Offset into the dynptr
+ * @buffer: User-provided buffer to copy contents into
+ * @buffer__szk: Size (in bytes) of the buffer. This is the length of the
+ *		 requested slice. This must be a constant.
  *
  * For non-skb and non-xdp type dynptrs, there is no difference between
  * bpf_dynptr_slice and bpf_dynptr_data.
@@ -2287,13 +2291,7 @@ __bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset
  * bpf_dynptr_slice_rdwr will not invalidate any ctx->data/data_end pointers in
  * the bpf program.
  *
- * @ptr: The dynptr whose data slice to retrieve
- * @offset: Offset into the dynptr
- * @buffer: User-provided buffer to copy contents into
- * @buffer__szk: Size (in bytes) of the buffer. This is the length of the
- * requested slice. This must be a constant.
- *
- * @returns: NULL if the call failed (eg invalid dynptr), pointer to a
+ * Return: NULL if the call failed (eg invalid dynptr), pointer to a
  * data slice (can be either direct pointer to the data or a pointer to the user
  * provided buffer, with its contents containing the data, if unable to obtain
  * direct pointer)
-- 
cgit v1.2.3-70-g09d2


From db52b587c67f40e4bd6e8167f2334d4500617bdc Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Wed, 1 Mar 2023 13:49:10 -0600
Subject: bpf, docs: Fix __uninit kfunc doc section

In commit d96d937d7c5c ("bpf: Add __uninit kfunc annotation"), the
__uninit kfunc annotation was documented in kfuncs.rst. You have to
fully underline a section in rst, or the build will issue a warning that
the title underline is too short:

./Documentation/bpf/kfuncs.rst:104: WARNING: Title underline too short.

2.2.2 __uninit Annotation
--------------------

This patch fixes that title underline.

Fixes: d96d937d7c5c ("bpf: Add __uninit kfunc annotation")
Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230301194910.602738-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index 9a78533d25ac..9d85bbc3b771 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -101,7 +101,7 @@ size parameter, and the value of the constant matters for program safety, __k
 suffix should be used.
 
 2.2.2 __uninit Annotation
---------------------
+-------------------------
 
 This annotation is used to indicate that the argument will be treated as
 uninitialized.
-- 
cgit v1.2.3-70-g09d2


From 1eebcb60633fd469ee27b0fbd7ee4f271feedeca Mon Sep 17 00:00:00 2001
From: Daniel Müller <deso@posteo.net>
Date: Wed, 1 Mar 2023 21:23:06 +0000
Subject: libbpf: Implement basic zip archive parsing support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This change implements support for reading zip archives, including
opening an archive, finding an entry based on its path and name in it,
and closing it.
The code was copied from https://github.com/iovisor/bcc/pull/4440, which
implements similar functionality for bcc. The author confirmed that he
is fine with this usage and the corresponding relicensing. I adjusted it
to adhere to libbpf coding standards.

Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Michał Gregorczyk <michalgr@meta.com>
Link: https://lore.kernel.org/bpf/20230301212308.1839139-2-deso@posteo.net
---
 tools/lib/bpf/Build |   2 +-
 tools/lib/bpf/zip.c | 328 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 tools/lib/bpf/zip.h |  47 ++++++++
 3 files changed, 376 insertions(+), 1 deletion(-)
 create mode 100644 tools/lib/bpf/zip.c
 create mode 100644 tools/lib/bpf/zip.h

diff --git a/tools/lib/bpf/Build b/tools/lib/bpf/Build
index 5a3dfb56d78f..b8b0a6369363 100644
--- a/tools/lib/bpf/Build
+++ b/tools/lib/bpf/Build
@@ -1,4 +1,4 @@
 libbpf-y := libbpf.o bpf.o nlattr.o btf.o libbpf_errno.o str_error.o \
 	    netlink.o bpf_prog_linfo.o libbpf_probes.o hashmap.o \
 	    btf_dump.o ringbuf.o strset.o linker.o gen_loader.o relo_core.o \
-	    usdt.o
+	    usdt.o zip.o
diff --git a/tools/lib/bpf/zip.c b/tools/lib/bpf/zip.c
new file mode 100644
index 000000000000..8458c2dd0e3b
--- /dev/null
+++ b/tools/lib/bpf/zip.c
@@ -0,0 +1,328 @@
+// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
+/*
+ * Routines for dealing with .zip archives.
+ *
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <unistd.h>
+
+#include "libbpf_internal.h"
+#include "zip.h"
+
+/* Specification of ZIP file format can be found here:
+ * https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT
+ * For a high level overview of the structure of a ZIP file see
+ * sections 4.3.1 - 4.3.6.
+ *
+ * Data structures appearing in ZIP files do not contain any
+ * padding and they might be misaligned. To allow us to safely
+ * operate on pointers to such structures and their members, we
+ * declare the types as packed.
+ */
+
+#define END_OF_CD_RECORD_MAGIC 0x06054b50
+
+/* See section 4.3.16 of the spec. */
+struct end_of_cd_record {
+	/* Magic value equal to END_OF_CD_RECORD_MAGIC */
+	__u32 magic;
+
+	/* Number of the file containing this structure or 0xFFFF if ZIP64 archive.
+	 * Zip archive might span multiple files (disks).
+	 */
+	__u16 this_disk;
+
+	/* Number of the file containing the beginning of the central directory or
+	 * 0xFFFF if ZIP64 archive.
+	 */
+	__u16 cd_disk;
+
+	/* Number of central directory records on this disk or 0xFFFF if ZIP64
+	 * archive.
+	 */
+	__u16 cd_records;
+
+	/* Number of central directory records on all disks or 0xFFFF if ZIP64
+	 * archive.
+	 */
+	__u16 cd_records_total;
+
+	/* Size of the central directory record or 0xFFFFFFFF if ZIP64 archive. */
+	__u32 cd_size;
+
+	/* Offset of the central directory from the beginning of the archive or
+	 * 0xFFFFFFFF if ZIP64 archive.
+	 */
+	__u32 cd_offset;
+
+	/* Length of comment data following end of central directory record. */
+	__u16 comment_length;
+
+	/* Up to 64k of arbitrary bytes. */
+	/* uint8_t comment[comment_length] */
+} __attribute__((packed));
+
+#define CD_FILE_HEADER_MAGIC 0x02014b50
+#define FLAG_ENCRYPTED (1 << 0)
+#define FLAG_HAS_DATA_DESCRIPTOR (1 << 3)
+
+/* See section 4.3.12 of the spec. */
+struct cd_file_header {
+	/* Magic value equal to CD_FILE_HEADER_MAGIC. */
+	__u32 magic;
+	__u16 version;
+	/* Minimum zip version needed to extract the file. */
+	__u16 min_version;
+	__u16 flags;
+	__u16 compression;
+	__u16 last_modified_time;
+	__u16 last_modified_date;
+	__u32 crc;
+	__u32 compressed_size;
+	__u32 uncompressed_size;
+	__u16 file_name_length;
+	__u16 extra_field_length;
+	__u16 file_comment_length;
+	/* Number of the disk where the file starts or 0xFFFF if ZIP64 archive. */
+	__u16 disk;
+	__u16 internal_attributes;
+	__u32 external_attributes;
+	/* Offset from the start of the disk containing the local file header to the
+	 * start of the local file header.
+	 */
+	__u32 offset;
+} __attribute__((packed));
+
+#define LOCAL_FILE_HEADER_MAGIC 0x04034b50
+
+/* See section 4.3.7 of the spec. */
+struct local_file_header {
+	/* Magic value equal to LOCAL_FILE_HEADER_MAGIC. */
+	__u32 magic;
+	/* Minimum zip version needed to extract the file. */
+	__u16 min_version;
+	__u16 flags;
+	__u16 compression;
+	__u16 last_modified_time;
+	__u16 last_modified_date;
+	__u32 crc;
+	__u32 compressed_size;
+	__u32 uncompressed_size;
+	__u16 file_name_length;
+	__u16 extra_field_length;
+} __attribute__((packed));
+
+struct zip_archive {
+	void *data;
+	__u32 size;
+	__u32 cd_offset;
+	__u32 cd_records;
+};
+
+static void *check_access(struct zip_archive *archive, __u32 offset, __u32 size)
+{
+	if (offset + size > archive->size || offset > offset + size)
+		return NULL;
+
+	return archive->data + offset;
+}
+
+/* Returns 0 on success, -EINVAL on error and -ENOTSUP if the eocd indicates the
+ * archive uses features which are not supported.
+ */
+static int try_parse_end_of_cd(struct zip_archive *archive, __u32 offset)
+{
+	__u16 comment_length, cd_records;
+	struct end_of_cd_record *eocd;
+	__u32 cd_offset, cd_size;
+
+	eocd = check_access(archive, offset, sizeof(*eocd));
+	if (!eocd || eocd->magic != END_OF_CD_RECORD_MAGIC)
+		return -EINVAL;
+
+	comment_length = eocd->comment_length;
+	if (offset + sizeof(*eocd) + comment_length != archive->size)
+		return -EINVAL;
+
+	cd_records = eocd->cd_records;
+	if (eocd->this_disk != 0 || eocd->cd_disk != 0 || eocd->cd_records_total != cd_records)
+		/* This is a valid eocd, but we only support single-file non-ZIP64 archives. */
+		return -ENOTSUP;
+
+	cd_offset = eocd->cd_offset;
+	cd_size = eocd->cd_size;
+	if (!check_access(archive, cd_offset, cd_size))
+		return -EINVAL;
+
+	archive->cd_offset = cd_offset;
+	archive->cd_records = cd_records;
+	return 0;
+}
+
+static int find_cd(struct zip_archive *archive)
+{
+	int rc = -EINVAL;
+	int64_t limit;
+	__u32 offset;
+
+	if (archive->size <= sizeof(struct end_of_cd_record))
+		return -EINVAL;
+
+	/* Because the end of central directory ends with a variable length array of
+	 * up to 0xFFFF bytes we can't know exactly where it starts and need to
+	 * search for it at the end of the file, scanning the (limit, offset] range.
+	 */
+	offset = archive->size - sizeof(struct end_of_cd_record);
+	limit = (int64_t)offset - (1 << 16);
+
+	for (; offset >= 0 && offset > limit && rc != 0; offset--) {
+		rc = try_parse_end_of_cd(archive, offset);
+		if (rc == -ENOTSUP)
+			break;
+	}
+	return rc;
+}
+
+struct zip_archive *zip_archive_open(const char *path)
+{
+	struct zip_archive *archive;
+	int err, fd;
+	off_t size;
+	void *data;
+
+	fd = open(path, O_RDONLY | O_CLOEXEC);
+	if (fd < 0)
+		return ERR_PTR(-errno);
+
+	size = lseek(fd, 0, SEEK_END);
+	if (size == (off_t)-1 || size > UINT32_MAX) {
+		close(fd);
+		return ERR_PTR(-EINVAL);
+	}
+
+	data = mmap(NULL, size, PROT_READ, MAP_PRIVATE, fd, 0);
+	err = -errno;
+	close(fd);
+
+	if (data == MAP_FAILED)
+		return ERR_PTR(err);
+
+	archive = malloc(sizeof(*archive));
+	if (!archive) {
+		munmap(data, size);
+		return ERR_PTR(-ENOMEM);
+	};
+
+	archive->data = data;
+	archive->size = size;
+
+	err = find_cd(archive);
+	if (err) {
+		munmap(data, size);
+		free(archive);
+		return ERR_PTR(err);
+	}
+
+	return archive;
+}
+
+void zip_archive_close(struct zip_archive *archive)
+{
+	munmap(archive->data, archive->size);
+	free(archive);
+}
+
+static struct local_file_header *local_file_header_at_offset(struct zip_archive *archive,
+							     __u32 offset)
+{
+	struct local_file_header *lfh;
+
+	lfh = check_access(archive, offset, sizeof(*lfh));
+	if (!lfh || lfh->magic != LOCAL_FILE_HEADER_MAGIC)
+		return NULL;
+
+	return lfh;
+}
+
+static int get_entry_at_offset(struct zip_archive *archive, __u32 offset, struct zip_entry *out)
+{
+	struct local_file_header *lfh;
+	__u32 compressed_size;
+	const char *name;
+	void *data;
+
+	lfh = local_file_header_at_offset(archive, offset);
+	if (!lfh)
+		return -EINVAL;
+
+	offset += sizeof(*lfh);
+	if ((lfh->flags & FLAG_ENCRYPTED) || (lfh->flags & FLAG_HAS_DATA_DESCRIPTOR))
+		return -EINVAL;
+
+	name = check_access(archive, offset, lfh->file_name_length);
+	if (!name)
+		return -EINVAL;
+
+	offset += lfh->file_name_length;
+	if (!check_access(archive, offset, lfh->extra_field_length))
+		return -EINVAL;
+
+	offset += lfh->extra_field_length;
+	compressed_size = lfh->compressed_size;
+	data = check_access(archive, offset, compressed_size);
+	if (!data)
+		return -EINVAL;
+
+	out->compression = lfh->compression;
+	out->name_length = lfh->file_name_length;
+	out->name = name;
+	out->data = data;
+	out->data_length = compressed_size;
+	out->data_offset = offset;
+
+	return 0;
+}
+
+int zip_archive_find_entry(struct zip_archive *archive, const char *file_name,
+			   struct zip_entry *out)
+{
+	size_t file_name_length = strlen(file_name);
+	__u32 i, offset = archive->cd_offset;
+
+	for (i = 0; i < archive->cd_records; ++i) {
+		__u16 cdfh_name_length, cdfh_flags;
+		struct cd_file_header *cdfh;
+		const char *cdfh_name;
+
+		cdfh = check_access(archive, offset, sizeof(*cdfh));
+		if (!cdfh || cdfh->magic != CD_FILE_HEADER_MAGIC)
+			return -EINVAL;
+
+		offset += sizeof(*cdfh);
+		cdfh_name_length = cdfh->file_name_length;
+		cdfh_name = check_access(archive, offset, cdfh_name_length);
+		if (!cdfh_name)
+			return -EINVAL;
+
+		cdfh_flags = cdfh->flags;
+		if ((cdfh_flags & FLAG_ENCRYPTED) == 0 &&
+		    (cdfh_flags & FLAG_HAS_DATA_DESCRIPTOR) == 0 &&
+		    file_name_length == cdfh_name_length &&
+		    memcmp(file_name, archive->data + offset, file_name_length) == 0) {
+			return get_entry_at_offset(archive, cdfh->offset, out);
+		}
+
+		offset += cdfh_name_length;
+		offset += cdfh->extra_field_length;
+		offset += cdfh->file_comment_length;
+	}
+
+	return -ENOENT;
+}
diff --git a/tools/lib/bpf/zip.h b/tools/lib/bpf/zip.h
new file mode 100644
index 000000000000..1c1bb21fba76
--- /dev/null
+++ b/tools/lib/bpf/zip.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+#ifndef __LIBBPF_ZIP_H
+#define __LIBBPF_ZIP_H
+
+#include <linux/types.h>
+
+/* Represents an open zip archive.
+ * Only basic ZIP files are supported, in particular the following are not
+ * supported:
+ * - encryption
+ * - streaming
+ * - multi-part ZIP files
+ * - ZIP64
+ */
+struct zip_archive;
+
+/* Carries information on name, compression method, and data corresponding to a
+ * file in a zip archive.
+ */
+struct zip_entry {
+	/* Compression method as defined in pkzip spec. 0 means data is uncompressed. */
+	__u16 compression;
+
+	/* Non-null terminated name of the file. */
+	const char *name;
+	/* Length of the file name. */
+	__u16 name_length;
+
+	/* Pointer to the file data. */
+	const void *data;
+	/* Length of the file data. */
+	__u32 data_length;
+	/* Offset of the file data within the archive. */
+	__u32 data_offset;
+};
+
+/* Open a zip archive. Returns NULL in case of an error. */
+struct zip_archive *zip_archive_open(const char *path);
+
+/* Close a zip archive and release resources. */
+void zip_archive_close(struct zip_archive *archive);
+
+/* Look up an entry corresponding to a file in given zip archive. */
+int zip_archive_find_entry(struct zip_archive *archive, const char *name, struct zip_entry *out);
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From 434fdcead73515a76885418ffe2d96b4f3ed0f49 Mon Sep 17 00:00:00 2001
From: Daniel Müller <deso@posteo.net>
Date: Wed, 1 Mar 2023 21:23:07 +0000
Subject: libbpf: Introduce elf_find_func_offset_from_file() function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This change splits the elf_find_func_offset() function in two:
elf_find_func_offset(), which now accepts an already opened Elf object
instead of a path to a file that is to be opened, as well as
elf_find_func_offset_from_file(), which opens a binary based on a
path and then invokes elf_find_func_offset() on the Elf object. Having
this split in responsibilities will allow us to call
elf_find_func_offset() from other code paths on Elf objects that did not
necessarily come from a file on disk.

Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230301212308.1839139-3-deso@posteo.net
---
 tools/lib/bpf/libbpf.c | 57 +++++++++++++++++++++++++++++++-------------------
 1 file changed, 36 insertions(+), 21 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index ba9e7e2b7951..1abef3dc9a6e 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -10530,32 +10530,19 @@ static Elf_Scn *elf_find_next_scn_by_type(Elf *elf, int sh_type, Elf_Scn *scn)
 	return NULL;
 }
 
-/* Find offset of function name in object specified by path.  "name" matches
- * symbol name or name@@LIB for library functions.
+/* Find offset of function name in the provided ELF object. "binary_path" is
+ * the path to the ELF binary represented by "elf", and only used for error
+ * reporting matters. "name" matches symbol name or name@@LIB for library
+ * functions.
  */
-static long elf_find_func_offset(const char *binary_path, const char *name)
+static long elf_find_func_offset(Elf *elf, const char *binary_path, const char *name)
 {
-	int fd, i, sh_types[2] = { SHT_DYNSYM, SHT_SYMTAB };
+	int i, sh_types[2] = { SHT_DYNSYM, SHT_SYMTAB };
 	bool is_shared_lib, is_name_qualified;
-	char errmsg[STRERR_BUFSIZE];
 	long ret = -ENOENT;
 	size_t name_len;
 	GElf_Ehdr ehdr;
-	Elf *elf;
 
-	fd = open(binary_path, O_RDONLY | O_CLOEXEC);
-	if (fd < 0) {
-		ret = -errno;
-		pr_warn("failed to open %s: %s\n", binary_path,
-			libbpf_strerror_r(ret, errmsg, sizeof(errmsg)));
-		return ret;
-	}
-	elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
-	if (!elf) {
-		pr_warn("elf: could not read elf from %s: %s\n", binary_path, elf_errmsg(-1));
-		close(fd);
-		return -LIBBPF_ERRNO__FORMAT;
-	}
 	if (!gelf_getehdr(elf, &ehdr)) {
 		pr_warn("elf: failed to get ehdr from %s: %s\n", binary_path, elf_errmsg(-1));
 		ret = -LIBBPF_ERRNO__FORMAT;
@@ -10568,7 +10555,7 @@ static long elf_find_func_offset(const char *binary_path, const char *name)
 	/* Does name specify "@@LIB"? */
 	is_name_qualified = strstr(name, "@@") != NULL;
 
-	/* Search SHT_DYNSYM, SHT_SYMTAB for symbol.  This search order is used because if
+	/* Search SHT_DYNSYM, SHT_SYMTAB for symbol. This search order is used because if
 	 * a binary is stripped, it may only have SHT_DYNSYM, and a fully-statically
 	 * linked binary may not have SHT_DYMSYM, so absence of a section should not be
 	 * reported as a warning/error.
@@ -10681,6 +10668,34 @@ static long elf_find_func_offset(const char *binary_path, const char *name)
 		}
 	}
 out:
+	return ret;
+}
+
+/* Find offset of function name in ELF object specified by path. "name" matches
+ * symbol name or name@@LIB for library functions.
+ */
+static long elf_find_func_offset_from_file(const char *binary_path, const char *name)
+{
+	char errmsg[STRERR_BUFSIZE];
+	long ret = -ENOENT;
+	Elf *elf;
+	int fd;
+
+	fd = open(binary_path, O_RDONLY | O_CLOEXEC);
+	if (fd < 0) {
+		ret = -errno;
+		pr_warn("failed to open %s: %s\n", binary_path,
+			libbpf_strerror_r(ret, errmsg, sizeof(errmsg)));
+		return ret;
+	}
+	elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
+	if (!elf) {
+		pr_warn("elf: could not read elf from %s: %s\n", binary_path, elf_errmsg(-1));
+		close(fd);
+		return -LIBBPF_ERRNO__FORMAT;
+	}
+
+	ret = elf_find_func_offset(elf, binary_path, name);
 	elf_end(elf);
 	close(fd);
 	return ret;
@@ -10804,7 +10819,7 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 	if (func_name) {
 		long sym_off;
 
-		sym_off = elf_find_func_offset(binary_path, func_name);
+		sym_off = elf_find_func_offset_from_file(binary_path, func_name);
 		if (sym_off < 0)
 			return libbpf_err_ptr(sym_off);
 		func_offset += sym_off;
-- 
cgit v1.2.3-70-g09d2


From c44fd84507637f07ca9a7e16f83706e109f6b8ff Mon Sep 17 00:00:00 2001
From: Daniel Müller <deso@posteo.net>
Date: Wed, 1 Mar 2023 21:23:08 +0000
Subject: libbpf: Add support for attaching uprobes to shared objects in APKs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This change adds support for attaching uprobes to shared objects located
in APKs, which is relevant for Android systems where various libraries
may reside in APKs. To make that happen, we extend the syntax for the
"binary path" argument to attach to with that supported by various
Android tools:
  <archive>!/<binary-in-archive>

For example:
  /system/app/test-app/test-app.apk!/lib/arm64-v8a/libc++_shared.so

APKs need to be specified via full path, i.e., we do not attempt to
resolve mere file names by searching system directories.

We cannot currently test this functionality end-to-end in an automated
fashion, because it relies on an Android system being present, but there
is no support for that in CI. I have tested the functionality manually,
by creating a libbpf program containing a uretprobe, attaching it to a
function inside a shared object inside an APK, and verifying the sanity
of the returned values.

Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230301212308.1839139-4-deso@posteo.net
---
 tools/lib/bpf/libbpf.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 84 insertions(+), 7 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 1abef3dc9a6e..dacaae31b76a 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -53,6 +53,7 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 #include "bpf_gen_internal.h"
+#include "zip.h"
 
 #ifndef BPF_FS_MAGIC
 #define BPF_FS_MAGIC		0xcafe4a11
@@ -10701,6 +10702,68 @@ static long elf_find_func_offset_from_file(const char *binary_path, const char *
 	return ret;
 }
 
+/* Find offset of function name in archive specified by path. Currently
+ * supported are .zip files that do not compress their contents, as used on
+ * Android in the form of APKs, for example. "file_name" is the name of the ELF
+ * file inside the archive. "func_name" matches symbol name or name@@LIB for
+ * library functions.
+ *
+ * An overview of the APK format specifically provided here:
+ * https://en.wikipedia.org/w/index.php?title=Apk_(file_format)&oldid=1139099120#Package_contents
+ */
+static long elf_find_func_offset_from_archive(const char *archive_path, const char *file_name,
+					      const char *func_name)
+{
+	struct zip_archive *archive;
+	struct zip_entry entry;
+	long ret;
+	Elf *elf;
+
+	archive = zip_archive_open(archive_path);
+	if (IS_ERR(archive)) {
+		ret = PTR_ERR(archive);
+		pr_warn("zip: failed to open %s: %ld\n", archive_path, ret);
+		return ret;
+	}
+
+	ret = zip_archive_find_entry(archive, file_name, &entry);
+	if (ret) {
+		pr_warn("zip: could not find archive member %s in %s: %ld\n", file_name,
+			archive_path, ret);
+		goto out;
+	}
+	pr_debug("zip: found entry for %s in %s at 0x%lx\n", file_name, archive_path,
+		 (unsigned long)entry.data_offset);
+
+	if (entry.compression) {
+		pr_warn("zip: entry %s of %s is compressed and cannot be handled\n", file_name,
+			archive_path);
+		ret = -LIBBPF_ERRNO__FORMAT;
+		goto out;
+	}
+
+	elf = elf_memory((void *)entry.data, entry.data_length);
+	if (!elf) {
+		pr_warn("elf: could not read elf file %s from %s: %s\n", file_name, archive_path,
+			elf_errmsg(-1));
+		ret = -LIBBPF_ERRNO__LIBELF;
+		goto out;
+	}
+
+	ret = elf_find_func_offset(elf, file_name, func_name);
+	if (ret > 0) {
+		pr_debug("elf: symbol address match for %s of %s in %s: 0x%x + 0x%lx = 0x%lx\n",
+			 func_name, file_name, archive_path, entry.data_offset, ret,
+			 ret + entry.data_offset);
+		ret += entry.data_offset;
+	}
+	elf_end(elf);
+
+out:
+	zip_archive_close(archive);
+	return ret;
+}
+
 static const char *arch_specific_lib_paths(void)
 {
 	/*
@@ -10786,9 +10849,10 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 				const char *binary_path, size_t func_offset,
 				const struct bpf_uprobe_opts *opts)
 {
-	DECLARE_LIBBPF_OPTS(bpf_perf_event_opts, pe_opts);
+	const char *archive_path = NULL, *archive_sep = NULL;
 	char errmsg[STRERR_BUFSIZE], *legacy_probe = NULL;
-	char full_binary_path[PATH_MAX];
+	DECLARE_LIBBPF_OPTS(bpf_perf_event_opts, pe_opts);
+	char full_path[PATH_MAX];
 	struct bpf_link *link;
 	size_t ref_ctr_off;
 	int pfd, err;
@@ -10805,21 +10869,34 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 	if (!binary_path)
 		return libbpf_err_ptr(-EINVAL);
 
-	if (!strchr(binary_path, '/')) {
-		err = resolve_full_path(binary_path, full_binary_path,
-					sizeof(full_binary_path));
+	/* Check if "binary_path" refers to an archive. */
+	archive_sep = strstr(binary_path, "!/");
+	if (archive_sep) {
+		full_path[0] = '\0';
+		libbpf_strlcpy(full_path, binary_path,
+			       min(sizeof(full_path), (size_t)(archive_sep - binary_path + 1)));
+		archive_path = full_path;
+		binary_path = archive_sep + 2;
+	} else if (!strchr(binary_path, '/')) {
+		err = resolve_full_path(binary_path, full_path, sizeof(full_path));
 		if (err) {
 			pr_warn("prog '%s': failed to resolve full path for '%s': %d\n",
 				prog->name, binary_path, err);
 			return libbpf_err_ptr(err);
 		}
-		binary_path = full_binary_path;
+		binary_path = full_path;
 	}
 	func_name = OPTS_GET(opts, func_name, NULL);
 	if (func_name) {
 		long sym_off;
 
-		sym_off = elf_find_func_offset_from_file(binary_path, func_name);
+		if (archive_path) {
+			sym_off = elf_find_func_offset_from_archive(archive_path, binary_path,
+								    func_name);
+			binary_path = archive_path;
+		} else {
+			sym_off = elf_find_func_offset_from_file(binary_path, func_name);
+		}
 		if (sym_off < 0)
 			return libbpf_err_ptr(sym_off);
 		func_offset += sym_off;
-- 
cgit v1.2.3-70-g09d2


From c45eac537bd8b4977d335c123212140bc5257670 Mon Sep 17 00:00:00 2001
From: Joanne Koong <joannelkoong@gmail.com>
Date: Wed, 1 Mar 2023 21:30:14 -0800
Subject: bpf: Fix bpf_dynptr_slice{_rdwr} to return NULL instead of 0

Change bpf_dynptr_slice and bpf_dynptr_slice_rdwr to return NULL instead
of 0, in accordance with the codebase guidelines.

Fixes: 66e3a13e7c2c ("bpf: Add bpf_dynptr_slice and bpf_dynptr_slice_rdwr")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230302053014.1726219-1-joannelkoong@gmail.com
---
 kernel/bpf/helpers.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 58431a92bb65..de9ef8476e29 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2227,11 +2227,11 @@ __bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset
 	int err;
 
 	if (!ptr->data)
-		return 0;
+		return NULL;
 
 	err = bpf_dynptr_check_off_len(ptr, offset, len);
 	if (err)
-		return 0;
+		return NULL;
 
 	type = bpf_dynptr_get_type(ptr);
 
@@ -2252,7 +2252,7 @@ __bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset
 	}
 	default:
 		WARN_ONCE(true, "unknown dynptr type %d\n", type);
-		return 0;
+		return NULL;
 	}
 }
 
@@ -2300,7 +2300,7 @@ __bpf_kfunc void *bpf_dynptr_slice_rdwr(const struct bpf_dynptr_kern *ptr, u32 o
 					void *buffer, u32 buffer__szk)
 {
 	if (!ptr->data || bpf_dynptr_is_rdonly(ptr))
-		return 0;
+		return NULL;
 
 	/* bpf_dynptr_slice_rdwr is the same logic as bpf_dynptr_slice.
 	 *
-- 
cgit v1.2.3-70-g09d2


From d56b0c461d19dae917fa0bba76cbe8ad7a44712e Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 2 Mar 2023 12:39:17 -0600
Subject: bpf, docs: Fix link to netdev-FAQ target

The BPF devel Q&A documentation page makes frequent reference to the
netdev-QA page via the netdev-FAQ rst link. This link is currently
broken, as is evidenced by the build output when making BPF docs:

./Documentation/bpf/bpf_devel_QA.rst:150: WARNING: undefined label: 'netdev-faq'
./Documentation/bpf/bpf_devel_QA.rst:206: WARNING: undefined label: 'netdev-faq'
./Documentation/bpf/bpf_devel_QA.rst:231: WARNING: undefined label: 'netdev-faq'
./Documentation/bpf/bpf_devel_QA.rst:396: WARNING: undefined label: 'netdev-faq'
./Documentation/bpf/bpf_devel_QA.rst:412: WARNING: undefined label: 'netdev-faq'

Fix the links to point to the actual netdev-faq page.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230302183918.54190-1-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/bpf_devel_QA.rst | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/Documentation/bpf/bpf_devel_QA.rst b/Documentation/bpf/bpf_devel_QA.rst
index 03d4993eda6f..5f5f9ccc3862 100644
--- a/Documentation/bpf/bpf_devel_QA.rst
+++ b/Documentation/bpf/bpf_devel_QA.rst
@@ -128,7 +128,7 @@ into the bpf-next tree will make their way into net-next tree. net and
 net-next are both run by David S. Miller. From there, they will go
 into the kernel mainline tree run by Linus Torvalds. To read up on the
 process of net and net-next being merged into the mainline tree, see
-the :ref:`netdev-FAQ`
+the `netdev-FAQ`_.
 
 
 
@@ -147,7 +147,7 @@ request)::
 Q: How do I indicate which tree (bpf vs. bpf-next) my patch should be applied to?
 ---------------------------------------------------------------------------------
 
-A: The process is the very same as described in the :ref:`netdev-FAQ`,
+A: The process is the very same as described in the `netdev-FAQ`_,
 so please read up on it. The subject line must indicate whether the
 patch is a fix or rather "next-like" content in order to let the
 maintainers know whether it is targeted at bpf or bpf-next.
@@ -206,7 +206,7 @@ ii) run extensive BPF test suite and
 Once the BPF pull request was accepted by David S. Miller, then
 the patches end up in net or net-next tree, respectively, and
 make their way from there further into mainline. Again, see the
-:ref:`netdev-FAQ` for additional information e.g. on how often they are
+`netdev-FAQ`_ for additional information e.g. on how often they are
 merged to mainline.
 
 Q: How long do I need to wait for feedback on my BPF patches?
@@ -230,7 +230,7 @@ Q: Are patches applied to bpf-next when the merge window is open?
 -----------------------------------------------------------------
 A: For the time when the merge window is open, bpf-next will not be
 processed. This is roughly analogous to net-next patch processing,
-so feel free to read up on the :ref:`netdev-FAQ` about further details.
+so feel free to read up on the `netdev-FAQ`_ about further details.
 
 During those two weeks of merge window, we might ask you to resend
 your patch series once bpf-next is open again. Once Linus released
@@ -394,7 +394,7 @@ netdev kernel mailing list in Cc and ask for the fix to be queued up:
   netdev@vger.kernel.org
 
 The process in general is the same as on netdev itself, see also the
-:ref:`netdev-FAQ`.
+`netdev-FAQ`_.
 
 Q: Do you also backport to kernels not currently maintained as stable?
 ----------------------------------------------------------------------
@@ -410,7 +410,7 @@ Q: The BPF patch I am about to submit needs to go to stable as well
 What should I do?
 
 A: The same rules apply as with netdev patch submissions in general, see
-the :ref:`netdev-FAQ`.
+the `netdev-FAQ`_.
 
 Never add "``Cc: stable@vger.kernel.org``" to the patch description, but
 ask the BPF maintainers to queue the patches instead. This can be done
@@ -685,7 +685,7 @@ when:
 
 .. Links
 .. _Documentation/process/: https://www.kernel.org/doc/html/latest/process/
-.. _netdev-FAQ: Documentation/process/maintainer-netdev.rst
+.. _netdev-FAQ: https://www.kernel.org/doc/html/latest/process/maintainer-netdev.html
 .. _selftests:
    https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/testing/selftests/bpf/
 .. _Documentation/dev-tools/kselftest.rst:
-- 
cgit v1.2.3-70-g09d2


From cacad346f67ce9604dcc9db10f1f1769dabb3891 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 2 Mar 2023 12:39:18 -0600
Subject: bpf, docs: Fix final bpf docs build failure

maps.rst in the BPF documentation links to the
/userspace-api/ebpf/syscall document
(Documentation/userspace-api/ebpf/syscall.rst). For some reason, if you
try to reference the document with :doc:, the docs build emits the
following warning:

./Documentation/bpf/maps.rst:13: WARNING: \
    unknown document: '/userspace-api/ebpf/syscall'

It appears that other places in the docs tree also don't support using
:doc:. Elsewhere in the BPF documentation, we just reference the kernel
docs page directly. Let's do that here to clean up the last remaining
noise in the docs build.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230302183918.54190-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/maps.rst | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/Documentation/bpf/maps.rst b/Documentation/bpf/maps.rst
index 4906ff0f8382..6f069f3d6f4b 100644
--- a/Documentation/bpf/maps.rst
+++ b/Documentation/bpf/maps.rst
@@ -11,9 +11,9 @@ maps are accessed from BPF programs via BPF helpers which are documented in the
 `man-pages`_ for `bpf-helpers(7)`_.
 
 BPF maps are accessed from user space via the ``bpf`` syscall, which provides
-commands to create maps, lookup elements, update elements and delete
-elements. More details of the BPF syscall are available in
-:doc:`/userspace-api/ebpf/syscall` and in the `man-pages`_ for `bpf(2)`_.
+commands to create maps, lookup elements, update elements and delete elements.
+More details of the BPF syscall are available in `ebpf-syscall`_ and in the
+`man-pages`_ for `bpf(2)`_.
 
 Map Types
 =========
@@ -79,3 +79,4 @@ Find and delete element by key in a given map using ``attr->map_fd``,
 .. _man-pages: https://www.kernel.org/doc/man-pages/
 .. _bpf(2): https://man7.org/linux/man-pages/man2/bpf.2.html
 .. _bpf-helpers(7): https://man7.org/linux/man-pages/man7/bpf-helpers.7.html
+.. _ebpf-syscall: https://docs.kernel.org/userspace-api/ebpf/syscall.html
-- 
cgit v1.2.3-70-g09d2


From c501bf55c88b834adefda870c7c092ec9052a437 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 2 Mar 2023 09:42:59 -1000
Subject: bpf: Make bpf_get_current_[ancestor_]cgroup_id() available for all
 program types

These helpers are safe to call from any context and there's no reason to
restrict access to them. Remove them from bpf_trace and filter lists and add
to bpf_base_func_proto() under perfmon_capable().

v2: After consulting with Andrii, relocated in bpf_base_func_proto() so that
    they require bpf_capable() but not perfomon_capable() as it doesn't read
    from or affect others on the system.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/ZAD8QyoszMZiTzBY@slm.duckdns.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/cgroup.c      | 4 ----
 kernel/bpf/helpers.c     | 4 ++++
 kernel/trace/bpf_trace.c | 4 ----
 net/core/filter.c        | 6 ------
 4 files changed, 4 insertions(+), 14 deletions(-)

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index bf2fdb33fb31..a4ae422b8f12 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -2529,10 +2529,6 @@ cgroup_current_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_get_current_pid_tgid_proto;
 	case BPF_FUNC_get_current_comm:
 		return &bpf_get_current_comm_proto;
-	case BPF_FUNC_get_current_cgroup_id:
-		return &bpf_get_current_cgroup_id_proto;
-	case BPF_FUNC_get_current_ancestor_cgroup_id:
-		return &bpf_get_current_ancestor_cgroup_id_proto;
 #ifdef CONFIG_CGROUP_NET_CLASSID
 	case BPF_FUNC_get_cgroup_classid:
 		return &bpf_get_cgroup_classid_curr_proto;
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index de9ef8476e29..6fc0d6c44e4c 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1748,6 +1748,10 @@ bpf_base_func_proto(enum bpf_func_id func_id)
 		return &bpf_cgrp_storage_get_proto;
 	case BPF_FUNC_cgrp_storage_delete:
 		return &bpf_cgrp_storage_delete_proto;
+	case BPF_FUNC_get_current_cgroup_id:
+		return &bpf_get_current_cgroup_id_proto;
+	case BPF_FUNC_get_current_ancestor_cgroup_id:
+		return &bpf_get_current_ancestor_cgroup_id_proto;
 #endif
 	default:
 		break;
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index e8da032bb6fc..bcf91bc7bf71 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -1453,10 +1453,6 @@ bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		       NULL : &bpf_probe_read_compat_str_proto;
 #endif
 #ifdef CONFIG_CGROUPS
-	case BPF_FUNC_get_current_cgroup_id:
-		return &bpf_get_current_cgroup_id_proto;
-	case BPF_FUNC_get_current_ancestor_cgroup_id:
-		return &bpf_get_current_ancestor_cgroup_id_proto;
 	case BPF_FUNC_cgrp_storage_get:
 		return &bpf_cgrp_storage_get_proto;
 	case BPF_FUNC_cgrp_storage_delete:
diff --git a/net/core/filter.c b/net/core/filter.c
index 8f3124e06133..a2dc44e70ea0 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -8165,12 +8165,6 @@ sk_msg_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_sk_storage_delete_proto;
 	case BPF_FUNC_get_netns_cookie:
 		return &bpf_get_netns_cookie_sk_msg_proto;
-#ifdef CONFIG_CGROUPS
-	case BPF_FUNC_get_current_cgroup_id:
-		return &bpf_get_current_cgroup_id_proto;
-	case BPF_FUNC_get_current_ancestor_cgroup_id:
-		return &bpf_get_current_ancestor_cgroup_id_proto;
-#endif
 #ifdef CONFIG_CGROUP_NET_CLASSID
 	case BPF_FUNC_get_cgroup_classid:
 		return &bpf_get_cgroup_classid_curr_proto;
-- 
cgit v1.2.3-70-g09d2


From ec97a76f113ee0d8fd17178b4e4ffbf0ab9e5452 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Thu, 2 Mar 2023 16:55:00 -0800
Subject: selftests/bpf: Add -Wuninitialized flag to bpf prog flags

Per C99 standard [0], Section 6.7.8, Paragraph 10:

  If an object that has automatic storage duration is not initialized
  explicitly, its value is indeterminate.

And in the same document, in appendix "J.2 Undefined behavior":

  The behavior is undefined in the following circumstances:
  [...]
  The value of an object with automatic storage duration is used while
  it is indeterminate (6.2.4, 6.7.8, 6.8).

This means that use of an uninitialized stack variable is undefined
behavior, and therefore that clang can choose to do a variety of scary
things, such as not generating bytecode for "bunch of useful code" in
the below example:

  void some_func()
  {
    int i;
    if (!i)
      return;
    // bunch of useful code
  }

To add insult to injury, if some_func above is a helper function for
some BPF program, clang can choose to not generate an "exit" insn,
causing verifier to fail with "last insn is not an exit or jmp". Going
from that verification failure to the root cause of uninitialized use
is certain to be frustrating.

This patch adds -Wuninitialized to the cflags for selftest BPF progs and
fixes up existing instances of uninitialized use.

  [0]: https://www.open-std.org/jtc1/sc22/WG14/www/docs/n1256.pdf

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Cc: David Vernet <void@manifault.com>
Cc: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230303005500.1614874-1-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/Makefile                        |  2 +-
 tools/testing/selftests/bpf/progs/rbtree.c                  |  2 +-
 tools/testing/selftests/bpf/progs/rbtree_fail.c             |  7 +++++--
 tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c |  2 +-
 tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c     |  2 +-
 tools/testing/selftests/bpf/progs/test_tunnel_kern.c        | 10 +++++-----
 6 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index f40606a85a0f..eab3cf5399f5 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -357,7 +357,7 @@ BPF_CFLAGS = -g -Werror -D__TARGET_ARCH_$(SRCARCH) $(MENDIAN) 		\
 	     -I$(abspath $(OUTPUT)/../usr/include)
 
 CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \
-	       -Wno-compare-distinct-pointer-types
+	       -Wno-compare-distinct-pointer-types -Wuninitialized
 
 $(OUTPUT)/test_l4lb_noinline.o: BPF_CFLAGS += -fno-inline
 $(OUTPUT)/test_xdp_noinline.o: BPF_CFLAGS += -fno-inline
diff --git a/tools/testing/selftests/bpf/progs/rbtree.c b/tools/testing/selftests/bpf/progs/rbtree.c
index e5db1a4287e5..4c90aa6abddd 100644
--- a/tools/testing/selftests/bpf/progs/rbtree.c
+++ b/tools/testing/selftests/bpf/progs/rbtree.c
@@ -75,7 +75,7 @@ SEC("tc")
 long rbtree_add_and_remove(void *ctx)
 {
 	struct bpf_rb_node *res = NULL;
-	struct node_data *n, *m;
+	struct node_data *n, *m = NULL;
 
 	n = bpf_obj_new(typeof(*n));
 	if (!n)
diff --git a/tools/testing/selftests/bpf/progs/rbtree_fail.c b/tools/testing/selftests/bpf/progs/rbtree_fail.c
index bf3cba115897..1ced900f3fce 100644
--- a/tools/testing/selftests/bpf/progs/rbtree_fail.c
+++ b/tools/testing/selftests/bpf/progs/rbtree_fail.c
@@ -232,8 +232,11 @@ long rbtree_api_first_release_unlock_escape(void *ctx)
 
 	bpf_spin_lock(&glock);
 	res = bpf_rbtree_first(&groot);
-	if (res)
-		n = container_of(res, struct node_data, node);
+	if (!res) {
+		bpf_spin_unlock(&glock);
+		return 1;
+	}
+	n = container_of(res, struct node_data, node);
 	bpf_spin_unlock(&glock);
 
 	bpf_spin_lock(&glock);
diff --git a/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c b/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c
index 2fbef3cc7ad8..2dde8e3fe4c9 100644
--- a/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c
+++ b/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c
@@ -48,7 +48,7 @@ SEC("?lsm.s/bpf")
 __failure __msg("arg#0 expected pointer to stack or dynptr_ptr")
 int BPF_PROG(not_ptr_to_stack, int cmd, union bpf_attr *attr, unsigned int size)
 {
-	unsigned long val;
+	unsigned long val = 0;
 
 	return bpf_verify_pkcs7_signature((struct bpf_dynptr *)val,
 					  (struct bpf_dynptr *)val, NULL);
diff --git a/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c b/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
index b502e5c92e33..6ccf6d546074 100644
--- a/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
@@ -23,8 +23,8 @@ static struct bpf_sock_tuple *get_tuple(void *data, __u64 nh_off,
 					bool *ipv4)
 {
 	struct bpf_sock_tuple *result;
+	__u64 ihl_len = 0;
 	__u8 proto = 0;
-	__u64 ihl_len;
 
 	if (eth_proto == bpf_htons(ETH_P_IP)) {
 		struct iphdr *iph = (struct iphdr *)(data + nh_off);
diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
index 508da4a23c4f..95b4aa0928ba 100644
--- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
@@ -324,11 +324,11 @@ int ip4ip6erspan_get_tunnel(struct __sk_buff *skb)
 SEC("tc")
 int vxlan_set_tunnel_dst(struct __sk_buff *skb)
 {
-	int ret;
 	struct bpf_tunnel_key key;
 	struct vxlan_metadata md;
 	__u32 index = 0;
 	__u32 *local_ip = NULL;
+	int ret = 0;
 
 	local_ip = bpf_map_lookup_elem(&local_ip_map, &index);
 	if (!local_ip) {
@@ -363,11 +363,11 @@ int vxlan_set_tunnel_dst(struct __sk_buff *skb)
 SEC("tc")
 int vxlan_set_tunnel_src(struct __sk_buff *skb)
 {
-	int ret;
 	struct bpf_tunnel_key key;
 	struct vxlan_metadata md;
 	__u32 index = 0;
 	__u32 *local_ip = NULL;
+	int ret = 0;
 
 	local_ip = bpf_map_lookup_elem(&local_ip_map, &index);
 	if (!local_ip) {
@@ -494,9 +494,9 @@ SEC("tc")
 int ip6vxlan_set_tunnel_dst(struct __sk_buff *skb)
 {
 	struct bpf_tunnel_key key;
-	int ret;
 	__u32 index = 0;
 	__u32 *local_ip;
+	int ret = 0;
 
 	local_ip = bpf_map_lookup_elem(&local_ip_map, &index);
 	if (!local_ip) {
@@ -525,9 +525,9 @@ SEC("tc")
 int ip6vxlan_set_tunnel_src(struct __sk_buff *skb)
 {
 	struct bpf_tunnel_key key;
-	int ret;
 	__u32 index = 0;
 	__u32 *local_ip;
+	int ret = 0;
 
 	local_ip = bpf_map_lookup_elem(&local_ip_map, &index);
 	if (!local_ip) {
@@ -556,9 +556,9 @@ SEC("tc")
 int ip6vxlan_get_tunnel_src(struct __sk_buff *skb)
 {
 	struct bpf_tunnel_key key;
-	int ret;
 	__u32 index = 0;
 	__u32 *local_ip;
+	int ret = 0;
 
 	local_ip = bpf_map_lookup_elem(&local_ip_map, &index);
 	if (!local_ip) {
-- 
cgit v1.2.3-70-g09d2


From f71f8530494bb5ab43d3369ef0ce8373eb1ee077 Mon Sep 17 00:00:00 2001
From: Tero Kristo <tero.kristo@linux.intel.com>
Date: Thu, 2 Mar 2023 13:46:13 +0200
Subject: bpf: Add support for absolute value BPF timers

Add a new flag BPF_F_TIMER_ABS that can be passed to bpf_timer_start()
to start an absolute value timer instead of the default relative value.
This makes the timer expire at an exact point in time, instead of a time
with latencies induced by both the BPF and timer subsystems.

Suggested-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Tero Kristo <tero.kristo@linux.intel.com>
Link: https://lore.kernel.org/r/20230302114614.2985072-2-tero.kristo@linux.intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/uapi/linux/bpf.h       | 15 +++++++++++++++
 kernel/bpf/helpers.c           | 11 +++++++++--
 tools/include/uapi/linux/bpf.h | 15 +++++++++++++++
 3 files changed, 39 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index c9699304aed2..976b194eb775 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -4969,6 +4969,12 @@ union bpf_attr {
  *		different maps if key/value layout matches across maps.
  *		Every bpf_timer_set_callback() can have different callback_fn.
  *
+ *		*flags* can be one of:
+ *
+ *		**BPF_F_TIMER_ABS**
+ *			Start the timer in absolute expire value instead of the
+ *			default relative one.
+ *
  *	Return
  *		0 on success.
  *		**-EINVAL** if *timer* was not initialized with bpf_timer_init() earlier
@@ -7097,4 +7103,13 @@ struct bpf_core_relo {
 	enum bpf_core_relo_kind kind;
 };
 
+/*
+ * Flags to control bpf_timer_start() behaviour.
+ *     - BPF_F_TIMER_ABS: Timeout passed is absolute time, by default it is
+ *       relative to current time.
+ */
+enum {
+	BPF_F_TIMER_ABS = (1ULL << 0),
+};
+
 #endif /* _UAPI__LINUX_BPF_H__ */
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 6fc0d6c44e4c..12f12e879bcf 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1264,10 +1264,11 @@ BPF_CALL_3(bpf_timer_start, struct bpf_timer_kern *, timer, u64, nsecs, u64, fla
 {
 	struct bpf_hrtimer *t;
 	int ret = 0;
+	enum hrtimer_mode mode;
 
 	if (in_nmi())
 		return -EOPNOTSUPP;
-	if (flags)
+	if (flags > BPF_F_TIMER_ABS)
 		return -EINVAL;
 	__bpf_spin_lock_irqsave(&timer->lock);
 	t = timer->timer;
@@ -1275,7 +1276,13 @@ BPF_CALL_3(bpf_timer_start, struct bpf_timer_kern *, timer, u64, nsecs, u64, fla
 		ret = -EINVAL;
 		goto out;
 	}
-	hrtimer_start(&t->timer, ns_to_ktime(nsecs), HRTIMER_MODE_REL_SOFT);
+
+	if (flags & BPF_F_TIMER_ABS)
+		mode = HRTIMER_MODE_ABS_SOFT;
+	else
+		mode = HRTIMER_MODE_REL_SOFT;
+
+	hrtimer_start(&t->timer, ns_to_ktime(nsecs), mode);
 out:
 	__bpf_spin_unlock_irqrestore(&timer->lock);
 	return ret;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index c9699304aed2..976b194eb775 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -4969,6 +4969,12 @@ union bpf_attr {
  *		different maps if key/value layout matches across maps.
  *		Every bpf_timer_set_callback() can have different callback_fn.
  *
+ *		*flags* can be one of:
+ *
+ *		**BPF_F_TIMER_ABS**
+ *			Start the timer in absolute expire value instead of the
+ *			default relative one.
+ *
  *	Return
  *		0 on success.
  *		**-EINVAL** if *timer* was not initialized with bpf_timer_init() earlier
@@ -7097,4 +7103,13 @@ struct bpf_core_relo {
 	enum bpf_core_relo_kind kind;
 };
 
+/*
+ * Flags to control bpf_timer_start() behaviour.
+ *     - BPF_F_TIMER_ABS: Timeout passed is absolute time, by default it is
+ *       relative to current time.
+ */
+enum {
+	BPF_F_TIMER_ABS = (1ULL << 0),
+};
+
 #endif /* _UAPI__LINUX_BPF_H__ */
-- 
cgit v1.2.3-70-g09d2


From 944459e88b4f5c71683b56710f96e39756afae31 Mon Sep 17 00:00:00 2001
From: Tero Kristo <tero.kristo@linux.intel.com>
Date: Thu, 2 Mar 2023 13:46:14 +0200
Subject: selftests/bpf: Add absolute timer test

Add test for the absolute BPF timer under the existing timer tests. This
will run the timer two times with 1us expiration time, and then re-arm
the timer at ~35s in the future. At the end, it is verified that the
absolute timer expired exactly two times.

Signed-off-by: Tero Kristo <tero.kristo@linux.intel.com>
Link: https://lore.kernel.org/r/20230302114614.2985072-3-tero.kristo@linux.intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/timer.c |  3 ++
 tools/testing/selftests/bpf/progs/timer.c      | 45 ++++++++++++++++++++++++++
 2 files changed, 48 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/timer.c b/tools/testing/selftests/bpf/prog_tests/timer.c
index 7eb049214859..290c21dbe65a 100644
--- a/tools/testing/selftests/bpf/prog_tests/timer.c
+++ b/tools/testing/selftests/bpf/prog_tests/timer.c
@@ -29,6 +29,9 @@ static int timer(struct timer *timer_skel)
 	/* check that timer_cb2() was executed twice */
 	ASSERT_EQ(timer_skel->bss->bss_data, 10, "bss_data");
 
+	/* check that timer_cb3() was executed twice */
+	ASSERT_EQ(timer_skel->bss->abs_data, 12, "abs_data");
+
 	/* check that there were no errors in timer execution */
 	ASSERT_EQ(timer_skel->bss->err, 0, "err");
 
diff --git a/tools/testing/selftests/bpf/progs/timer.c b/tools/testing/selftests/bpf/progs/timer.c
index acda5c9cea93..9a16d95213e1 100644
--- a/tools/testing/selftests/bpf/progs/timer.c
+++ b/tools/testing/selftests/bpf/progs/timer.c
@@ -46,7 +46,15 @@ struct {
 	__type(value, struct elem);
 } lru SEC(".maps");
 
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct elem);
+} abs_timer SEC(".maps");
+
 __u64 bss_data;
+__u64 abs_data;
 __u64 err;
 __u64 ok;
 __u64 callback_check = 52;
@@ -284,3 +292,40 @@ int BPF_PROG2(test2, int, a, int, b)
 
 	return bpf_timer_test();
 }
+
+/* callback for absolute timer */
+static int timer_cb3(void *map, int *key, struct bpf_timer *timer)
+{
+	abs_data += 6;
+
+	if (abs_data < 12) {
+		bpf_timer_start(timer, bpf_ktime_get_boot_ns() + 1000,
+				BPF_F_TIMER_ABS);
+	} else {
+		/* Re-arm timer ~35 seconds in future */
+		bpf_timer_start(timer, bpf_ktime_get_boot_ns() + (1ull << 35),
+				BPF_F_TIMER_ABS);
+	}
+
+	return 0;
+}
+
+SEC("fentry/bpf_fentry_test3")
+int BPF_PROG2(test3, int, a)
+{
+	int key = 0;
+	struct bpf_timer *timer;
+
+	bpf_printk("test3");
+
+	timer = bpf_map_lookup_elem(&abs_timer, &key);
+	if (timer) {
+		if (bpf_timer_init(timer, &abs_timer, CLOCK_BOOTTIME) != 0)
+			err |= 2048;
+		bpf_timer_set_callback(timer, timer_cb3);
+		bpf_timer_start(timer, bpf_ktime_get_boot_ns() + 1000,
+				BPF_F_TIMER_ABS);
+	}
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From 03b77e17aeb22a5935ea20d585ca6a1f2947e62b Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 2 Mar 2023 20:14:41 -0800
Subject: bpf: Rename __kptr_ref -> __kptr and __kptr -> __kptr_untrusted.

__kptr meant to store PTR_UNTRUSTED kernel pointers inside bpf maps.
The concept felt useful, but didn't get much traction,
since bpf_rdonly_cast() was added soon after and bpf programs received
a simpler way to access PTR_UNTRUSTED kernel pointers
without going through restrictive __kptr usage.

Rename __kptr_ref -> __kptr and __kptr -> __kptr_untrusted to indicate
its intended usage.
The main goal of __kptr_untrusted was to read/write such pointers
directly while bpf_kptr_xchg was a mechanism to access refcnted
kernel pointers. The next patch will allow RCU protected __kptr access
with direct read. At that point __kptr_untrusted will be deprecated.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230303041446.3630-2-alexei.starovoitov@gmail.com
---
 Documentation/bpf/bpf_design_QA.rst                |  4 ++--
 Documentation/bpf/cpumasks.rst                     |  4 ++--
 Documentation/bpf/kfuncs.rst                       |  2 +-
 kernel/bpf/btf.c                                   |  4 ++--
 tools/lib/bpf/bpf_helpers.h                        |  2 +-
 tools/testing/selftests/bpf/progs/cb_refs.c        |  2 +-
 .../selftests/bpf/progs/cgrp_kfunc_common.h        |  2 +-
 tools/testing/selftests/bpf/progs/cpumask_common.h |  2 +-
 tools/testing/selftests/bpf/progs/jit_probe_mem.c  |  2 +-
 tools/testing/selftests/bpf/progs/lru_bug.c        |  2 +-
 tools/testing/selftests/bpf/progs/map_kptr.c       |  4 ++--
 tools/testing/selftests/bpf/progs/map_kptr_fail.c  |  6 +++---
 .../selftests/bpf/progs/task_kfunc_common.h        |  2 +-
 tools/testing/selftests/bpf/test_verifier.c        | 22 +++++++++++-----------
 14 files changed, 30 insertions(+), 30 deletions(-)

diff --git a/Documentation/bpf/bpf_design_QA.rst b/Documentation/bpf/bpf_design_QA.rst
index bfff0e7e37c2..38372a956d65 100644
--- a/Documentation/bpf/bpf_design_QA.rst
+++ b/Documentation/bpf/bpf_design_QA.rst
@@ -314,7 +314,7 @@ Q: What is the compatibility story for special BPF types in map values?
 Q: Users are allowed to embed bpf_spin_lock, bpf_timer fields in their BPF map
 values (when using BTF support for BPF maps). This allows to use helpers for
 such objects on these fields inside map values. Users are also allowed to embed
-pointers to some kernel types (with __kptr and __kptr_ref BTF tags). Will the
+pointers to some kernel types (with __kptr_untrusted and __kptr BTF tags). Will the
 kernel preserve backwards compatibility for these features?
 
 A: It depends. For bpf_spin_lock, bpf_timer: YES, for kptr and everything else:
@@ -324,7 +324,7 @@ For struct types that have been added already, like bpf_spin_lock and bpf_timer,
 the kernel will preserve backwards compatibility, as they are part of UAPI.
 
 For kptrs, they are also part of UAPI, but only with respect to the kptr
-mechanism. The types that you can use with a __kptr and __kptr_ref tagged
+mechanism. The types that you can use with a __kptr_untrusted and __kptr tagged
 pointer in your struct are NOT part of the UAPI contract. The supported types can
 and will change across kernel releases. However, operations like accessing kptr
 fields and bpf_kptr_xchg() helper will continue to be supported across kernel
diff --git a/Documentation/bpf/cpumasks.rst b/Documentation/bpf/cpumasks.rst
index 24bef9cbbeee..75344cd230e5 100644
--- a/Documentation/bpf/cpumasks.rst
+++ b/Documentation/bpf/cpumasks.rst
@@ -51,7 +51,7 @@ For example:
 .. code-block:: c
 
         struct cpumask_map_value {
-                struct bpf_cpumask __kptr_ref * cpumask;
+                struct bpf_cpumask __kptr * cpumask;
         };
 
         struct array_map {
@@ -128,7 +128,7 @@ Here is an example of a ``struct bpf_cpumask *`` being retrieved from a map:
 
 	/* struct containing the struct bpf_cpumask kptr which is stored in the map. */
 	struct cpumasks_kfunc_map_value {
-		struct bpf_cpumask __kptr_ref * bpf_cpumask;
+		struct bpf_cpumask __kptr * bpf_cpumask;
 	};
 
 	/* The map containing struct cpumasks_kfunc_map_value entries. */
diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index 9d85bbc3b771..b5d9b0d446bc 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -544,7 +544,7 @@ Here's an example of how it can be used:
 
 	/* struct containing the struct task_struct kptr which is actually stored in the map. */
 	struct __cgroups_kfunc_map_value {
-		struct cgroup __kptr_ref * cgroup;
+		struct cgroup __kptr * cgroup;
 	};
 
 	/* The map containing struct __cgroups_kfunc_map_value entries. */
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index ef2d8969ed1f..c5e1d6955491 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3288,9 +3288,9 @@ static int btf_find_kptr(const struct btf *btf, const struct btf_type *t,
 	/* Reject extra tags */
 	if (btf_type_is_type_tag(btf_type_by_id(btf, t->type)))
 		return -EINVAL;
-	if (!strcmp("kptr", __btf_name_by_offset(btf, t->name_off)))
+	if (!strcmp("kptr_untrusted", __btf_name_by_offset(btf, t->name_off)))
 		type = BPF_KPTR_UNREF;
-	else if (!strcmp("kptr_ref", __btf_name_by_offset(btf, t->name_off)))
+	else if (!strcmp("kptr", __btf_name_by_offset(btf, t->name_off)))
 		type = BPF_KPTR_REF;
 	else
 		return -EINVAL;
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index 5ec1871acb2f..7d12d3e620cc 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -174,8 +174,8 @@ enum libbpf_tristate {
 
 #define __kconfig __attribute__((section(".kconfig")))
 #define __ksym __attribute__((section(".ksyms")))
+#define __kptr_untrusted __attribute__((btf_type_tag("kptr_untrusted")))
 #define __kptr __attribute__((btf_type_tag("kptr")))
-#define __kptr_ref __attribute__((btf_type_tag("kptr_ref")))
 
 #ifndef ___bpf_concat
 #define ___bpf_concat(a, b) a ## b
diff --git a/tools/testing/selftests/bpf/progs/cb_refs.c b/tools/testing/selftests/bpf/progs/cb_refs.c
index 7653df1bc787..ce96b33e38d6 100644
--- a/tools/testing/selftests/bpf/progs/cb_refs.c
+++ b/tools/testing/selftests/bpf/progs/cb_refs.c
@@ -4,7 +4,7 @@
 #include <bpf/bpf_helpers.h>
 
 struct map_value {
-	struct prog_test_ref_kfunc __kptr_ref *ptr;
+	struct prog_test_ref_kfunc __kptr *ptr;
 };
 
 struct {
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
index 2f8de933b957..d0b7cd0d09d7 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
@@ -10,7 +10,7 @@
 #include <bpf/bpf_tracing.h>
 
 struct __cgrps_kfunc_map_value {
-	struct cgroup __kptr_ref * cgrp;
+	struct cgroup __kptr * cgrp;
 };
 
 struct hash_map {
diff --git a/tools/testing/selftests/bpf/progs/cpumask_common.h b/tools/testing/selftests/bpf/progs/cpumask_common.h
index ad34f3b602be..65e5496ca1b2 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_common.h
+++ b/tools/testing/selftests/bpf/progs/cpumask_common.h
@@ -10,7 +10,7 @@
 int err;
 
 struct __cpumask_map_value {
-	struct bpf_cpumask __kptr_ref * cpumask;
+	struct bpf_cpumask __kptr * cpumask;
 };
 
 struct array_map {
diff --git a/tools/testing/selftests/bpf/progs/jit_probe_mem.c b/tools/testing/selftests/bpf/progs/jit_probe_mem.c
index 2d2e61470794..13f00ca2ed0a 100644
--- a/tools/testing/selftests/bpf/progs/jit_probe_mem.c
+++ b/tools/testing/selftests/bpf/progs/jit_probe_mem.c
@@ -4,7 +4,7 @@
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_helpers.h>
 
-static struct prog_test_ref_kfunc __kptr_ref *v;
+static struct prog_test_ref_kfunc __kptr *v;
 long total_sum = -1;
 
 extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
diff --git a/tools/testing/selftests/bpf/progs/lru_bug.c b/tools/testing/selftests/bpf/progs/lru_bug.c
index 687081a724b3..ad73029cb1e3 100644
--- a/tools/testing/selftests/bpf/progs/lru_bug.c
+++ b/tools/testing/selftests/bpf/progs/lru_bug.c
@@ -4,7 +4,7 @@
 #include <bpf/bpf_helpers.h>
 
 struct map_value {
-	struct task_struct __kptr *ptr;
+	struct task_struct __kptr_untrusted *ptr;
 };
 
 struct {
diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c
index a24d17bc17eb..3fe7cde4cbfd 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr.c
@@ -4,8 +4,8 @@
 #include <bpf/bpf_helpers.h>
 
 struct map_value {
-	struct prog_test_ref_kfunc __kptr *unref_ptr;
-	struct prog_test_ref_kfunc __kptr_ref *ref_ptr;
+	struct prog_test_ref_kfunc __kptr_untrusted *unref_ptr;
+	struct prog_test_ref_kfunc __kptr *ref_ptr;
 };
 
 struct array_map {
diff --git a/tools/testing/selftests/bpf/progs/map_kptr_fail.c b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
index 760e41e1a632..e19e2a5f38cf 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
@@ -7,9 +7,9 @@
 
 struct map_value {
 	char buf[8];
-	struct prog_test_ref_kfunc __kptr *unref_ptr;
-	struct prog_test_ref_kfunc __kptr_ref *ref_ptr;
-	struct prog_test_member __kptr_ref *ref_memb_ptr;
+	struct prog_test_ref_kfunc __kptr_untrusted *unref_ptr;
+	struct prog_test_ref_kfunc __kptr *ref_ptr;
+	struct prog_test_member __kptr *ref_memb_ptr;
 };
 
 struct array_map {
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_common.h b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
index c0ffd171743e..4c2a4b0e3a25 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
@@ -10,7 +10,7 @@
 #include <bpf/bpf_tracing.h>
 
 struct __tasks_kfunc_map_value {
-	struct task_struct __kptr_ref * task;
+	struct task_struct __kptr * task;
 };
 
 struct hash_map {
diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c
index 8b9949bb833d..49a70d9beb0b 100644
--- a/tools/testing/selftests/bpf/test_verifier.c
+++ b/tools/testing/selftests/bpf/test_verifier.c
@@ -699,13 +699,13 @@ static int create_cgroup_storage(bool percpu)
  *   struct bpf_timer t;
  * };
  * struct btf_ptr {
+ *   struct prog_test_ref_kfunc __kptr_untrusted *ptr;
  *   struct prog_test_ref_kfunc __kptr *ptr;
- *   struct prog_test_ref_kfunc __kptr_ref *ptr;
- *   struct prog_test_member __kptr_ref *ptr;
+ *   struct prog_test_member __kptr *ptr;
  * }
  */
 static const char btf_str_sec[] = "\0bpf_spin_lock\0val\0cnt\0l\0bpf_timer\0timer\0t"
-				  "\0btf_ptr\0prog_test_ref_kfunc\0ptr\0kptr\0kptr_ref"
+				  "\0btf_ptr\0prog_test_ref_kfunc\0ptr\0kptr\0kptr_untrusted"
 				  "\0prog_test_member";
 static __u32 btf_raw_types[] = {
 	/* int */
@@ -724,20 +724,20 @@ static __u32 btf_raw_types[] = {
 	BTF_MEMBER_ENC(41, 4, 0), /* struct bpf_timer t; */
 	/* struct prog_test_ref_kfunc */		/* [6] */
 	BTF_STRUCT_ENC(51, 0, 0),
-	BTF_STRUCT_ENC(89, 0, 0),			/* [7] */
+	BTF_STRUCT_ENC(95, 0, 0),			/* [7] */
+	/* type tag "kptr_untrusted" */
+	BTF_TYPE_TAG_ENC(80, 6),			/* [8] */
 	/* type tag "kptr" */
-	BTF_TYPE_TAG_ENC(75, 6),			/* [8] */
-	/* type tag "kptr_ref" */
-	BTF_TYPE_TAG_ENC(80, 6),			/* [9] */
-	BTF_TYPE_TAG_ENC(80, 7),			/* [10] */
+	BTF_TYPE_TAG_ENC(75, 6),			/* [9] */
+	BTF_TYPE_TAG_ENC(75, 7),			/* [10] */
 	BTF_PTR_ENC(8),					/* [11] */
 	BTF_PTR_ENC(9),					/* [12] */
 	BTF_PTR_ENC(10),				/* [13] */
 	/* struct btf_ptr */				/* [14] */
 	BTF_STRUCT_ENC(43, 3, 24),
-	BTF_MEMBER_ENC(71, 11, 0), /* struct prog_test_ref_kfunc __kptr *ptr; */
-	BTF_MEMBER_ENC(71, 12, 64), /* struct prog_test_ref_kfunc __kptr_ref *ptr; */
-	BTF_MEMBER_ENC(71, 13, 128), /* struct prog_test_member __kptr_ref *ptr; */
+	BTF_MEMBER_ENC(71, 11, 0), /* struct prog_test_ref_kfunc __kptr_untrusted *ptr; */
+	BTF_MEMBER_ENC(71, 12, 64), /* struct prog_test_ref_kfunc __kptr *ptr; */
+	BTF_MEMBER_ENC(71, 13, 128), /* struct prog_test_member __kptr *ptr; */
 };
 
 static char bpf_vlog[UINT_MAX >> 8];
-- 
cgit v1.2.3-70-g09d2


From 8d093b4e95a2a16a2cfcd36869b348a17112fabe Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 2 Mar 2023 20:14:42 -0800
Subject: bpf: Mark cgroups and dfl_cgrp fields as trusted.

bpf programs sometimes do:
bpf_cgrp_storage_get(&map, task->cgroups->dfl_cgrp, ...);
It is safe to do, because cgroups->dfl_cgrp pointer is set diring init and
never changes. The task->cgroups is also never NULL. It is also set during init
and will change when task switches cgroups. For any trusted task pointer
dereference of cgroups and dfl_cgrp should yield trusted pointers. The verifier
wasn't aware of this. Hence in gcc compiled kernels task->cgroups dereference
was producing PTR_TO_BTF_ID without modifiers while in clang compiled kernels
the verifier recognizes __rcu tag in cgroups field and produces
PTR_TO_BTF_ID | MEM_RCU | MAYBE_NULL.
Tag cgroups and dfl_cgrp as trusted to equalize clang and gcc behavior.
When GCC supports btf_type_tag such tagging will done directly in the type.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/bpf/20230303041446.3630-3-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index bf580f246a01..b834f3d2d81a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5046,6 +5046,11 @@ static int bpf_map_direct_read(struct bpf_map *map, int off, int size, u64 *val)
 
 BTF_TYPE_SAFE_NESTED(struct task_struct) {
 	const cpumask_t *cpus_ptr;
+	struct css_set __rcu *cgroups;
+};
+
+BTF_TYPE_SAFE_NESTED(struct css_set) {
+	struct cgroup *dfl_cgrp;
 };
 
 static bool nested_ptr_is_trusted(struct bpf_verifier_env *env,
@@ -5057,6 +5062,7 @@ static bool nested_ptr_is_trusted(struct bpf_verifier_env *env,
 		return false;
 
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_NESTED(struct task_struct));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_NESTED(struct css_set));
 
 	return btf_nested_type_is_trusted(&env->log, reg, off);
 }
-- 
cgit v1.2.3-70-g09d2


From 20c09d92faeefb8536f705d3a4629e0dc314c8a1 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 2 Mar 2023 20:14:43 -0800
Subject: bpf: Introduce kptr_rcu.

The life time of certain kernel structures like 'struct cgroup' is protected by RCU.
Hence it's safe to dereference them directly from __kptr tagged pointers in bpf maps.
The resulting pointer is MEM_RCU and can be passed to kfuncs that expect KF_RCU.
Derefrence of other kptr-s returns PTR_UNTRUSTED.

For example:
struct map_value {
   struct cgroup __kptr *cgrp;
};

SEC("tp_btf/cgroup_mkdir")
int BPF_PROG(test_cgrp_get_ancestors, struct cgroup *cgrp_arg, const char *path)
{
  struct cgroup *cg, *cg2;

  cg = bpf_cgroup_acquire(cgrp_arg); // cg is PTR_TRUSTED and ref_obj_id > 0
  bpf_kptr_xchg(&v->cgrp, cg);

  cg2 = v->cgrp; // This is new feature introduced by this patch.
  // cg2 is PTR_MAYBE_NULL | MEM_RCU.
  // When cg2 != NULL, it's a valid cgroup, but its percpu_ref could be zero

  if (cg2)
    bpf_cgroup_ancestor(cg2, level); // safe to do.
}

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230303041446.3630-4-alexei.starovoitov@gmail.com
---
 Documentation/bpf/kfuncs.rst                       | 12 +++--
 include/linux/btf.h                                |  2 +-
 kernel/bpf/helpers.c                               |  6 ++-
 kernel/bpf/verifier.c                              | 55 ++++++++++++++++++----
 net/bpf/test_run.c                                 |  3 +-
 .../selftests/bpf/progs/cgrp_kfunc_failure.c       |  2 +-
 tools/testing/selftests/bpf/progs/map_kptr_fail.c  |  4 +-
 tools/testing/selftests/bpf/verifier/calls.c       |  2 +-
 tools/testing/selftests/bpf/verifier/map_kptr.c    |  2 +-
 9 files changed, 65 insertions(+), 23 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index b5d9b0d446bc..69eccf6f98ef 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -249,11 +249,13 @@ added later.
 2.4.8 KF_RCU flag
 -----------------
 
-The KF_RCU flag is used for kfuncs which have a rcu ptr as its argument.
-When used together with KF_ACQUIRE, it indicates the kfunc should have a
-single argument which must be a trusted argument or a MEM_RCU pointer.
-The argument may have reference count of 0 and the kfunc must take this
-into consideration.
+The KF_RCU flag is a weaker version of KF_TRUSTED_ARGS. The kfuncs marked with
+KF_RCU expect either PTR_TRUSTED or MEM_RCU arguments. The verifier guarantees
+that the objects are valid and there is no use-after-free. The pointers are not
+NULL, but the object's refcount could have reached zero. The kfuncs need to
+consider doing refcnt != 0 check, especially when returning a KF_ACQUIRE
+pointer. Note as well that a KF_ACQUIRE kfunc that is KF_RCU should very likely
+also be KF_RET_NULL.
 
 .. _KF_deprecated_flag:
 
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 49e0fe6d8274..556b3e2e7471 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -70,7 +70,7 @@
 #define KF_TRUSTED_ARGS (1 << 4) /* kfunc only takes trusted pointer arguments */
 #define KF_SLEEPABLE    (1 << 5) /* kfunc may sleep */
 #define KF_DESTRUCTIVE  (1 << 6) /* kfunc performs destructive actions */
-#define KF_RCU          (1 << 7) /* kfunc only takes rcu pointer arguments */
+#define KF_RCU          (1 << 7) /* kfunc takes either rcu or trusted pointer arguments */
 
 /*
  * Tag marking a kernel function as a kfunc. This is meant to minimize the
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 12f12e879bcf..637ac4e92e75 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2163,8 +2163,10 @@ __bpf_kfunc struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level)
 	if (level > cgrp->level || level < 0)
 		return NULL;
 
+	/* cgrp's refcnt could be 0 here, but ancestors can still be accessed */
 	ancestor = cgrp->ancestors[level];
-	cgroup_get(ancestor);
+	if (!cgroup_tryget(ancestor))
+		return NULL;
 	return ancestor;
 }
 
@@ -2382,7 +2384,7 @@ BTF_ID_FLAGS(func, bpf_rbtree_first, KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_cgroup_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_release, KF_RELEASE)
-BTF_ID_FLAGS(func, bpf_cgroup_ancestor, KF_ACQUIRE | KF_TRUSTED_ARGS | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_cgroup_ancestor, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_from_id, KF_ACQUIRE | KF_RET_NULL)
 #endif
 BTF_ID_FLAGS(func, bpf_task_from_pid, KF_ACQUIRE | KF_RET_NULL)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index b834f3d2d81a..a095055d7ef4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4218,7 +4218,7 @@ static int map_kptr_match_type(struct bpf_verifier_env *env,
 			       struct bpf_reg_state *reg, u32 regno)
 {
 	const char *targ_name = kernel_type_name(kptr_field->kptr.btf, kptr_field->kptr.btf_id);
-	int perm_flags = PTR_MAYBE_NULL | PTR_TRUSTED;
+	int perm_flags = PTR_MAYBE_NULL | PTR_TRUSTED | MEM_RCU;
 	const char *reg_name = "";
 
 	/* Only unreferenced case accepts untrusted pointers */
@@ -4285,6 +4285,34 @@ bad_type:
 	return -EINVAL;
 }
 
+/* The non-sleepable programs and sleepable programs with explicit bpf_rcu_read_lock()
+ * can dereference RCU protected pointers and result is PTR_TRUSTED.
+ */
+static bool in_rcu_cs(struct bpf_verifier_env *env)
+{
+	return env->cur_state->active_rcu_lock || !env->prog->aux->sleepable;
+}
+
+/* Once GCC supports btf_type_tag the following mechanism will be replaced with tag check */
+BTF_SET_START(rcu_protected_types)
+BTF_ID(struct, prog_test_ref_kfunc)
+BTF_ID(struct, cgroup)
+BTF_SET_END(rcu_protected_types)
+
+static bool rcu_protected_object(const struct btf *btf, u32 btf_id)
+{
+	if (!btf_is_kernel(btf))
+		return false;
+	return btf_id_set_contains(&rcu_protected_types, btf_id);
+}
+
+static bool rcu_safe_kptr(const struct btf_field *field)
+{
+	const struct btf_field_kptr *kptr = &field->kptr;
+
+	return field->type == BPF_KPTR_REF && rcu_protected_object(kptr->btf, kptr->btf_id);
+}
+
 static int check_map_kptr_access(struct bpf_verifier_env *env, u32 regno,
 				 int value_regno, int insn_idx,
 				 struct btf_field *kptr_field)
@@ -4319,7 +4347,10 @@ static int check_map_kptr_access(struct bpf_verifier_env *env, u32 regno,
 		 * value from map as PTR_TO_BTF_ID, with the correct type.
 		 */
 		mark_btf_ld_reg(env, cur_regs(env), value_regno, PTR_TO_BTF_ID, kptr_field->kptr.btf,
-				kptr_field->kptr.btf_id, PTR_MAYBE_NULL | PTR_UNTRUSTED);
+				kptr_field->kptr.btf_id,
+				rcu_safe_kptr(kptr_field) && in_rcu_cs(env) ?
+				PTR_MAYBE_NULL | MEM_RCU :
+				PTR_MAYBE_NULL | PTR_UNTRUSTED);
 		/* For mark_ptr_or_null_reg */
 		val_reg->id = ++env->id_gen;
 	} else if (class == BPF_STX) {
@@ -5163,10 +5194,17 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 	 * An RCU-protected pointer can also be deemed trusted if we are in an
 	 * RCU read region. This case is handled below.
 	 */
-	if (nested_ptr_is_trusted(env, reg, off))
+	if (nested_ptr_is_trusted(env, reg, off)) {
 		flag |= PTR_TRUSTED;
-	else
+		/*
+		 * task->cgroups is trusted. It provides a stronger guarantee
+		 * than __rcu tag on 'cgroups' field in 'struct task_struct'.
+		 * Clear MEM_RCU in such case.
+		 */
+		flag &= ~MEM_RCU;
+	} else {
 		flag &= ~PTR_TRUSTED;
+	}
 
 	if (flag & MEM_RCU) {
 		/* Mark value register as MEM_RCU only if it is protected by
@@ -5175,11 +5213,10 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 		 * read lock region. Also mark rcu pointer as PTR_MAYBE_NULL since
 		 * it could be null in some cases.
 		 */
-		if (!env->cur_state->active_rcu_lock ||
-		    !(is_trusted_reg(reg) || is_rcu_reg(reg)))
-			flag &= ~MEM_RCU;
-		else
+		if (in_rcu_cs(env) && (is_trusted_reg(reg) || is_rcu_reg(reg)))
 			flag |= PTR_MAYBE_NULL;
+		else
+			flag &= ~MEM_RCU;
 	} else if (reg->type & MEM_RCU) {
 		/* ptr (reg) is marked as MEM_RCU, but the struct field is not tagged
 		 * with __rcu. Mark the flag as PTR_UNTRUSTED conservatively.
@@ -9676,7 +9713,7 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 			return -EINVAL;
 		}
 
-		if (is_kfunc_trusted_args(meta) &&
+		if ((is_kfunc_trusted_args(meta) || is_kfunc_rcu(meta)) &&
 		    (register_is_null(reg) || type_may_be_null(reg->type))) {
 			verbose(env, "Possibly NULL pointer passed to trusted arg%d\n", i);
 			return -EACCES;
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 6f3d654b3339..6a8b33a103a4 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -737,6 +737,7 @@ __bpf_kfunc void bpf_kfunc_call_test_mem_len_fail2(u64 *mem, int len)
 
 __bpf_kfunc void bpf_kfunc_call_test_ref(struct prog_test_ref_kfunc *p)
 {
+	/* p != NULL, but p->cnt could be 0 */
 }
 
 __bpf_kfunc void bpf_kfunc_call_test_destructive(void)
@@ -784,7 +785,7 @@ BTF_ID_FLAGS(func, bpf_kfunc_call_test_fail3)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_mem_len_pass1)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_mem_len_fail1)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_mem_len_fail2)
-BTF_ID_FLAGS(func, bpf_kfunc_call_test_ref, KF_TRUSTED_ARGS)
+BTF_ID_FLAGS(func, bpf_kfunc_call_test_ref, KF_TRUSTED_ARGS | KF_RCU)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_destructive, KF_DESTRUCTIVE)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_static_unused_arg)
 BTF_SET8_END(test_sk_check_kfunc_ids)
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
index 4ad7fe24966d..b42291ed9586 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
@@ -205,7 +205,7 @@ int BPF_PROG(cgrp_kfunc_get_unreleased, struct cgroup *cgrp, const char *path)
 }
 
 SEC("tp_btf/cgroup_mkdir")
-__failure __msg("arg#0 is untrusted_ptr_or_null_ expected ptr_ or socket")
+__failure __msg("expects refcounted")
 int BPF_PROG(cgrp_kfunc_release_untrusted, struct cgroup *cgrp, const char *path)
 {
 	struct __cgrps_kfunc_map_value *v;
diff --git a/tools/testing/selftests/bpf/progs/map_kptr_fail.c b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
index e19e2a5f38cf..08f9ec18c345 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
@@ -281,7 +281,7 @@ int reject_kptr_get_bad_type_match(struct __sk_buff *ctx)
 }
 
 SEC("?tc")
-__failure __msg("R1 type=untrusted_ptr_or_null_ expected=percpu_ptr_")
+__failure __msg("R1 type=rcu_ptr_or_null_ expected=percpu_ptr_")
 int mark_ref_as_untrusted_or_null(struct __sk_buff *ctx)
 {
 	struct map_value *v;
@@ -316,7 +316,7 @@ int reject_untrusted_store_to_ref(struct __sk_buff *ctx)
 }
 
 SEC("?tc")
-__failure __msg("R2 type=untrusted_ptr_ expected=ptr_")
+__failure __msg("R2 must be referenced")
 int reject_untrusted_xchg(struct __sk_buff *ctx)
 {
 	struct prog_test_ref_kfunc *p;
diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c
index 289ed202ec66..9a326a800e5c 100644
--- a/tools/testing/selftests/bpf/verifier/calls.c
+++ b/tools/testing/selftests/bpf/verifier/calls.c
@@ -243,7 +243,7 @@
 	},
 	.result_unpriv = REJECT,
 	.result = REJECT,
-	.errstr = "R1 must be referenced",
+	.errstr = "R1 must be",
 },
 {
 	"calls: valid kfunc call: referenced arg needs refcounted PTR_TO_BTF_ID",
diff --git a/tools/testing/selftests/bpf/verifier/map_kptr.c b/tools/testing/selftests/bpf/verifier/map_kptr.c
index 6914904344c0..d775ccb01989 100644
--- a/tools/testing/selftests/bpf/verifier/map_kptr.c
+++ b/tools/testing/selftests/bpf/verifier/map_kptr.c
@@ -336,7 +336,7 @@
 	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
 	.fixup_map_kptr = { 1 },
 	.result = REJECT,
-	.errstr = "R1 type=untrusted_ptr_or_null_ expected=percpu_ptr_",
+	.errstr = "R1 type=rcu_ptr_or_null_ expected=percpu_ptr_",
 },
 {
 	"map_kptr: ref: reject off != 0",
-- 
cgit v1.2.3-70-g09d2


From 838bd4ac9aa35bdf43bf0199fa8eef9d3a004611 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 2 Mar 2023 20:14:44 -0800
Subject: selftests/bpf: Add a test case for kptr_rcu.

Tweak existing map_kptr test to check kptr_rcu.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230303041446.3630-5-alexei.starovoitov@gmail.com
---
 tools/testing/selftests/bpf/progs/map_kptr.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c
index 3fe7cde4cbfd..3903d30217b8 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr.c
@@ -118,6 +118,7 @@ extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp
 extern struct prog_test_ref_kfunc *
 bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **p, int a, int b) __ksym;
 extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
+void bpf_kfunc_call_test_ref(struct prog_test_ref_kfunc *p) __ksym;
 
 #define WRITE_ONCE(x, val) ((*(volatile typeof(x) *) &(x)) = (val))
 
@@ -147,12 +148,23 @@ static void test_kptr_ref(struct map_value *v)
 	WRITE_ONCE(v->unref_ptr, p);
 	if (!p)
 		return;
+	/*
+	 * p is rcu_ptr_prog_test_ref_kfunc,
+	 * because bpf prog is non-sleepable and runs in RCU CS.
+	 * p can be passed to kfunc that requires KF_RCU.
+	 */
+	bpf_kfunc_call_test_ref(p);
 	if (p->a + p->b > 100)
 		return;
 	/* store NULL */
 	p = bpf_kptr_xchg(&v->ref_ptr, NULL);
 	if (!p)
 		return;
+	/*
+	 * p is trusted_ptr_prog_test_ref_kfunc.
+	 * p can be passed to kfunc that requires KF_RCU.
+	 */
+	bpf_kfunc_call_test_ref(p);
 	if (p->a + p->b > 100) {
 		bpf_kfunc_call_test_release(p);
 		return;
-- 
cgit v1.2.3-70-g09d2


From 0047d8343f6042c4feea24072ef254d47b8a33b3 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 2 Mar 2023 20:14:45 -0800
Subject: selftests/bpf: Tweak cgroup kfunc test.

Adjust cgroup kfunc test to dereference RCU protected cgroup pointer
as PTR_TRUSTED and pass into KF_TRUSTED_ARGS kfunc.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230303041446.3630-6-alexei.starovoitov@gmail.com
---
 tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
index 42e13aebdd62..030aff700084 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
@@ -61,7 +61,7 @@ int BPF_PROG(test_cgrp_acquire_leave_in_map, struct cgroup *cgrp, const char *pa
 SEC("tp_btf/cgroup_mkdir")
 int BPF_PROG(test_cgrp_xchg_release, struct cgroup *cgrp, const char *path)
 {
-	struct cgroup *kptr;
+	struct cgroup *kptr, *cg;
 	struct __cgrps_kfunc_map_value *v;
 	long status;
 
@@ -80,6 +80,16 @@ int BPF_PROG(test_cgrp_xchg_release, struct cgroup *cgrp, const char *path)
 		return 0;
 	}
 
+	kptr = v->cgrp;
+	if (!kptr) {
+		err = 4;
+		return 0;
+	}
+
+	cg = bpf_cgroup_ancestor(kptr, 1);
+	if (cg)	/* verifier only check */
+		bpf_cgroup_release(cg);
+
 	kptr = bpf_kptr_xchg(&v->cgrp, NULL);
 	if (!kptr) {
 		err = 3;
-- 
cgit v1.2.3-70-g09d2


From 6fcd486b3a0a628c41f12b3a7329a18a2c74b351 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 2 Mar 2023 20:14:46 -0800
Subject: bpf: Refactor RCU enforcement in the verifier.

bpf_rcu_read_lock/unlock() are only available in clang compiled kernels. Lack
of such key mechanism makes it impossible for sleepable bpf programs to use RCU
pointers.

Allow bpf_rcu_read_lock/unlock() in GCC compiled kernels (though GCC doesn't
support btf_type_tag yet) and allowlist certain field dereferences in important
data structures like tast_struct, cgroup, socket that are used by sleepable
programs either as RCU pointer or full trusted pointer (which is valid outside
of RCU CS). Use BTF_TYPE_SAFE_RCU and BTF_TYPE_SAFE_TRUSTED macros for such
tagging. They will be removed once GCC supports btf_type_tag.

With that refactor check_ptr_to_btf_access(). Make it strict in enforcing
PTR_TRUSTED and PTR_UNTRUSTED while deprecating old PTR_TO_BTF_ID without
modifier flags. There is a chance that this strict enforcement might break
existing programs (especially on GCC compiled kernels), but this cleanup has to
start sooner than later. Note PTR_TO_CTX access still yields old deprecated
PTR_TO_BTF_ID. Once it's converted to strict PTR_TRUSTED or PTR_UNTRUSTED the
kfuncs and helpers will be able to default to KF_TRUSTED_ARGS. KF_RCU will
remain as a weaker version of KF_TRUSTED_ARGS where obj refcnt could be 0.

Adjust rcu_read_lock selftest to run on gcc and clang compiled kernels.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230303041446.3630-7-alexei.starovoitov@gmail.com
---
 include/linux/bpf.h                                |   2 +-
 include/linux/bpf_verifier.h                       |   1 -
 kernel/bpf/btf.c                                   |  16 +-
 kernel/bpf/cpumask.c                               |  40 ++---
 kernel/bpf/verifier.c                              | 178 ++++++++++++++-------
 .../selftests/bpf/prog_tests/cgrp_local_storage.c  |  14 +-
 .../selftests/bpf/prog_tests/rcu_read_lock.c       |  16 +-
 .../selftests/bpf/progs/cgrp_ls_sleepable.c        |   4 +-
 .../testing/selftests/bpf/progs/cpumask_failure.c  |   2 +-
 .../selftests/bpf/progs/nested_trust_failure.c     |   2 +-
 tools/testing/selftests/bpf/progs/rcu_read_lock.c  |   6 +-
 tools/testing/selftests/bpf/verifier/calls.c       |   2 +-
 12 files changed, 173 insertions(+), 110 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 23ec684e660d..d3456804f7aa 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2279,7 +2279,7 @@ struct bpf_core_ctx {
 
 bool btf_nested_type_is_trusted(struct bpf_verifier_log *log,
 				const struct bpf_reg_state *reg,
-				int off);
+				int off, const char *suffix);
 
 bool btf_type_ids_nocast_alias(struct bpf_verifier_log *log,
 			       const struct btf *reg_btf, u32 reg_id,
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index b26ff2a8f63b..18538bad2b8c 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -537,7 +537,6 @@ struct bpf_verifier_env {
 	bool bypass_spec_v1;
 	bool bypass_spec_v4;
 	bool seen_direct_write;
-	bool rcu_tag_supported;
 	struct bpf_insn_aux_data *insn_aux_data; /* array of per-insn state */
 	const struct bpf_line_info *prev_linfo;
 	struct bpf_verifier_log log;
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index c5e1d6955491..a8cb09e5973b 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -6163,6 +6163,7 @@ static int btf_struct_walk(struct bpf_verifier_log *log, const struct btf *btf,
 	const char *tname, *mname, *tag_value;
 	u32 vlen, elem_id, mid;
 
+	*flag = 0;
 again:
 	tname = __btf_name_by_offset(btf, t->name_off);
 	if (!btf_type_is_struct(t)) {
@@ -6329,6 +6330,15 @@ error:
 		 * of this field or inside of this struct
 		 */
 		if (btf_type_is_struct(mtype)) {
+			if (BTF_INFO_KIND(mtype->info) == BTF_KIND_UNION &&
+			    btf_type_vlen(mtype) != 1)
+				/*
+				 * walking unions yields untrusted pointers
+				 * with exception of __bpf_md_ptr and other
+				 * unions with a single member
+				 */
+				*flag |= PTR_UNTRUSTED;
+
 			/* our field must be inside that union or struct */
 			t = mtype;
 
@@ -6373,7 +6383,7 @@ error:
 			stype = btf_type_skip_modifiers(btf, mtype->type, &id);
 			if (btf_type_is_struct(stype)) {
 				*next_btf_id = id;
-				*flag = tmp_flag;
+				*flag |= tmp_flag;
 				return WALK_PTR;
 			}
 		}
@@ -8357,7 +8367,7 @@ out:
 
 bool btf_nested_type_is_trusted(struct bpf_verifier_log *log,
 				const struct bpf_reg_state *reg,
-				int off)
+				int off, const char *suffix)
 {
 	struct btf *btf = reg->btf;
 	const struct btf_type *walk_type, *safe_type;
@@ -8374,7 +8384,7 @@ bool btf_nested_type_is_trusted(struct bpf_verifier_log *log,
 
 	tname = btf_name_by_offset(btf, walk_type->name_off);
 
-	ret = snprintf(safe_tname, sizeof(safe_tname), "%s__safe_fields", tname);
+	ret = snprintf(safe_tname, sizeof(safe_tname), "%s%s", tname, suffix);
 	if (ret < 0)
 		return false;
 
diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c
index 2b3fbbfebdc5..b6587ec40f1b 100644
--- a/kernel/bpf/cpumask.c
+++ b/kernel/bpf/cpumask.c
@@ -427,26 +427,26 @@ BTF_ID_FLAGS(func, bpf_cpumask_create, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cpumask_release, KF_RELEASE | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_cpumask_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_cpumask_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
-BTF_ID_FLAGS(func, bpf_cpumask_first, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_first_zero, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_set_cpu, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_clear_cpu, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_test_cpu, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_test_and_set_cpu, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_test_and_clear_cpu, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_setall, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_clear, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_and, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_or, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_xor, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_equal, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_intersects, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_subset, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_empty, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_full, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_copy, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_any, KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_any_and, KF_TRUSTED_ARGS)
+BTF_ID_FLAGS(func, bpf_cpumask_first, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_first_zero, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_set_cpu, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_clear_cpu, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_test_cpu, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_test_and_set_cpu, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_test_and_clear_cpu, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_setall, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_clear, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_and, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_or, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_xor, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_equal, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_intersects, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_subset, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_empty, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_full, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_copy, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_any, KF_RCU)
+BTF_ID_FLAGS(func, bpf_cpumask_any_and, KF_RCU)
 BTF_SET8_END(cpumask_kfunc_btf_ids)
 
 static const struct btf_kfunc_id_set cpumask_kfunc_set = {
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index a095055d7ef4..c2adf3c24c64 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5073,29 +5073,76 @@ static int bpf_map_direct_read(struct bpf_map *map, int off, int size, u64 *val)
 	return 0;
 }
 
-#define BTF_TYPE_SAFE_NESTED(__type)  __PASTE(__type, __safe_fields)
+#define BTF_TYPE_SAFE_RCU(__type)  __PASTE(__type, __safe_rcu)
+#define BTF_TYPE_SAFE_TRUSTED(__type)  __PASTE(__type, __safe_trusted)
 
-BTF_TYPE_SAFE_NESTED(struct task_struct) {
+/*
+ * Allow list few fields as RCU trusted or full trusted.
+ * This logic doesn't allow mix tagging and will be removed once GCC supports
+ * btf_type_tag.
+ */
+
+/* RCU trusted: these fields are trusted in RCU CS and never NULL */
+BTF_TYPE_SAFE_RCU(struct task_struct) {
 	const cpumask_t *cpus_ptr;
 	struct css_set __rcu *cgroups;
+	struct task_struct __rcu *real_parent;
+	struct task_struct *group_leader;
 };
 
-BTF_TYPE_SAFE_NESTED(struct css_set) {
+BTF_TYPE_SAFE_RCU(struct css_set) {
 	struct cgroup *dfl_cgrp;
 };
 
-static bool nested_ptr_is_trusted(struct bpf_verifier_env *env,
-				  struct bpf_reg_state *reg,
-				  int off)
+/* full trusted: these fields are trusted even outside of RCU CS and never NULL */
+BTF_TYPE_SAFE_TRUSTED(struct bpf_iter_meta) {
+	__bpf_md_ptr(struct seq_file *, seq);
+};
+
+BTF_TYPE_SAFE_TRUSTED(struct bpf_iter__task) {
+	__bpf_md_ptr(struct bpf_iter_meta *, meta);
+	__bpf_md_ptr(struct task_struct *, task);
+};
+
+BTF_TYPE_SAFE_TRUSTED(struct linux_binprm) {
+	struct file *file;
+};
+
+BTF_TYPE_SAFE_TRUSTED(struct file) {
+	struct inode *f_inode;
+};
+
+BTF_TYPE_SAFE_TRUSTED(struct dentry) {
+	/* no negative dentry-s in places where bpf can see it */
+	struct inode *d_inode;
+};
+
+BTF_TYPE_SAFE_TRUSTED(struct socket) {
+	struct sock *sk;
+};
+
+static bool type_is_rcu(struct bpf_verifier_env *env,
+			struct bpf_reg_state *reg,
+			int off)
 {
-	/* If its parent is not trusted, it can't regain its trusted status. */
-	if (!is_trusted_reg(reg))
-		return false;
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct task_struct));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct css_set));
 
-	BTF_TYPE_EMIT(BTF_TYPE_SAFE_NESTED(struct task_struct));
-	BTF_TYPE_EMIT(BTF_TYPE_SAFE_NESTED(struct css_set));
+	return btf_nested_type_is_trusted(&env->log, reg, off, "__safe_rcu");
+}
 
-	return btf_nested_type_is_trusted(&env->log, reg, off);
+static bool type_is_trusted(struct bpf_verifier_env *env,
+			    struct bpf_reg_state *reg,
+			    int off)
+{
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct bpf_iter_meta));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct bpf_iter__task));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct linux_binprm));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct file));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct dentry));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct socket));
+
+	return btf_nested_type_is_trusted(&env->log, reg, off, "__safe_trusted");
 }
 
 static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
@@ -5181,49 +5228,58 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 	if (ret < 0)
 		return ret;
 
-	/* If this is an untrusted pointer, all pointers formed by walking it
-	 * also inherit the untrusted flag.
-	 */
-	if (type_flag(reg->type) & PTR_UNTRUSTED)
-		flag |= PTR_UNTRUSTED;
+	if (ret != PTR_TO_BTF_ID) {
+		/* just mark; */
 
-	/* By default any pointer obtained from walking a trusted pointer is no
-	 * longer trusted, unless the field being accessed has explicitly been
-	 * marked as inheriting its parent's state of trust.
-	 *
-	 * An RCU-protected pointer can also be deemed trusted if we are in an
-	 * RCU read region. This case is handled below.
-	 */
-	if (nested_ptr_is_trusted(env, reg, off)) {
-		flag |= PTR_TRUSTED;
-		/*
-		 * task->cgroups is trusted. It provides a stronger guarantee
-		 * than __rcu tag on 'cgroups' field in 'struct task_struct'.
-		 * Clear MEM_RCU in such case.
+	} else if (type_flag(reg->type) & PTR_UNTRUSTED) {
+		/* If this is an untrusted pointer, all pointers formed by walking it
+		 * also inherit the untrusted flag.
+		 */
+		flag = PTR_UNTRUSTED;
+
+	} else if (is_trusted_reg(reg) || is_rcu_reg(reg)) {
+		/* By default any pointer obtained from walking a trusted pointer is no
+		 * longer trusted, unless the field being accessed has explicitly been
+		 * marked as inheriting its parent's state of trust (either full or RCU).
+		 * For example:
+		 * 'cgroups' pointer is untrusted if task->cgroups dereference
+		 * happened in a sleepable program outside of bpf_rcu_read_lock()
+		 * section. In a non-sleepable program it's trusted while in RCU CS (aka MEM_RCU).
+		 * Note bpf_rcu_read_unlock() converts MEM_RCU pointers to PTR_UNTRUSTED.
+		 *
+		 * A regular RCU-protected pointer with __rcu tag can also be deemed
+		 * trusted if we are in an RCU CS. Such pointer can be NULL.
 		 */
-		flag &= ~MEM_RCU;
+		if (type_is_trusted(env, reg, off)) {
+			flag |= PTR_TRUSTED;
+		} else if (in_rcu_cs(env) && !type_may_be_null(reg->type)) {
+			if (type_is_rcu(env, reg, off)) {
+				/* ignore __rcu tag and mark it MEM_RCU */
+				flag |= MEM_RCU;
+			} else if (flag & MEM_RCU) {
+				/* __rcu tagged pointers can be NULL */
+				flag |= PTR_MAYBE_NULL;
+			} else if (flag & (MEM_PERCPU | MEM_USER)) {
+				/* keep as-is */
+			} else {
+				/* walking unknown pointers yields untrusted pointer */
+				flag = PTR_UNTRUSTED;
+			}
+		} else {
+			/*
+			 * If not in RCU CS or MEM_RCU pointer can be NULL then
+			 * aggressively mark as untrusted otherwise such
+			 * pointers will be plain PTR_TO_BTF_ID without flags
+			 * and will be allowed to be passed into helpers for
+			 * compat reasons.
+			 */
+			flag = PTR_UNTRUSTED;
+		}
 	} else {
+		/* Old compat. Deprecated */
 		flag &= ~PTR_TRUSTED;
 	}
 
-	if (flag & MEM_RCU) {
-		/* Mark value register as MEM_RCU only if it is protected by
-		 * bpf_rcu_read_lock() and the ptr reg is rcu or trusted. MEM_RCU
-		 * itself can already indicate trustedness inside the rcu
-		 * read lock region. Also mark rcu pointer as PTR_MAYBE_NULL since
-		 * it could be null in some cases.
-		 */
-		if (in_rcu_cs(env) && (is_trusted_reg(reg) || is_rcu_reg(reg)))
-			flag |= PTR_MAYBE_NULL;
-		else
-			flag &= ~MEM_RCU;
-	} else if (reg->type & MEM_RCU) {
-		/* ptr (reg) is marked as MEM_RCU, but the struct field is not tagged
-		 * with __rcu. Mark the flag as PTR_UNTRUSTED conservatively.
-		 */
-		flag |= PTR_UNTRUSTED;
-	}
-
 	if (atype == BPF_READ && value_regno >= 0)
 		mark_btf_ld_reg(env, regs, value_regno, ret, reg->btf, btf_id, flag);
 
@@ -10049,10 +10105,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 
 	rcu_lock = is_kfunc_bpf_rcu_read_lock(&meta);
 	rcu_unlock = is_kfunc_bpf_rcu_read_unlock(&meta);
-	if ((rcu_lock || rcu_unlock) && !env->rcu_tag_supported) {
-		verbose(env, "no vmlinux btf rcu tag support for kfunc %s\n", func_name);
-		return -EACCES;
-	}
 
 	if (env->cur_state->active_rcu_lock) {
 		struct bpf_func_state *state;
@@ -14911,8 +14963,22 @@ static int do_check(struct bpf_verifier_env *env)
 				 * src_reg == stack|map in some other branch.
 				 * Reject it.
 				 */
-				verbose(env, "same insn cannot be used with different pointers\n");
-				return -EINVAL;
+				if (base_type(src_reg_type) == PTR_TO_BTF_ID &&
+				    base_type(*prev_src_type) == PTR_TO_BTF_ID) {
+					/*
+					 * Have to support a use case when one path through
+					 * the program yields TRUSTED pointer while another
+					 * is UNTRUSTED. Fallback to UNTRUSTED to generate
+					 * BPF_PROBE_MEM.
+					 */
+					*prev_src_type = PTR_TO_BTF_ID | PTR_UNTRUSTED;
+				} else {
+					verbose(env,
+						"The same insn cannot be used with different pointers: %s",
+						reg_type_str(env, src_reg_type));
+					verbose(env, " != %s\n", reg_type_str(env, *prev_src_type));
+					return -EINVAL;
+				}
 			}
 
 		} else if (class == BPF_STX) {
@@ -17984,8 +18050,6 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, bpfptr_t uattr)
 	env->bypass_spec_v1 = bpf_bypass_spec_v1();
 	env->bypass_spec_v4 = bpf_bypass_spec_v4();
 	env->bpf_capable = bpf_capable();
-	env->rcu_tag_supported = btf_vmlinux &&
-		btf_find_by_name_kind(btf_vmlinux, "rcu", BTF_KIND_TYPE_TAG) > 0;
 
 	if (is_priv)
 		env->test_state_freq = attr->prog_flags & BPF_F_TEST_STATE_FREQ;
diff --git a/tools/testing/selftests/bpf/prog_tests/cgrp_local_storage.c b/tools/testing/selftests/bpf/prog_tests/cgrp_local_storage.c
index 2cc759956e3b..63e776f4176e 100644
--- a/tools/testing/selftests/bpf/prog_tests/cgrp_local_storage.c
+++ b/tools/testing/selftests/bpf/prog_tests/cgrp_local_storage.c
@@ -193,7 +193,7 @@ out:
 	cgrp_ls_sleepable__destroy(skel);
 }
 
-static void test_no_rcu_lock(__u64 cgroup_id)
+static void test_yes_rcu_lock(__u64 cgroup_id)
 {
 	struct cgrp_ls_sleepable *skel;
 	int err;
@@ -204,7 +204,7 @@ static void test_no_rcu_lock(__u64 cgroup_id)
 
 	skel->bss->target_pid = syscall(SYS_gettid);
 
-	bpf_program__set_autoload(skel->progs.no_rcu_lock, true);
+	bpf_program__set_autoload(skel->progs.yes_rcu_lock, true);
 	err = cgrp_ls_sleepable__load(skel);
 	if (!ASSERT_OK(err, "skel_load"))
 		goto out;
@@ -220,7 +220,7 @@ out:
 	cgrp_ls_sleepable__destroy(skel);
 }
 
-static void test_rcu_lock(void)
+static void test_no_rcu_lock(void)
 {
 	struct cgrp_ls_sleepable *skel;
 	int err;
@@ -229,7 +229,7 @@ static void test_rcu_lock(void)
 	if (!ASSERT_OK_PTR(skel, "skel_open"))
 		return;
 
-	bpf_program__set_autoload(skel->progs.yes_rcu_lock, true);
+	bpf_program__set_autoload(skel->progs.no_rcu_lock, true);
 	err = cgrp_ls_sleepable__load(skel);
 	ASSERT_ERR(err, "skel_load");
 
@@ -256,10 +256,10 @@ void test_cgrp_local_storage(void)
 		test_negative();
 	if (test__start_subtest("cgroup_iter_sleepable"))
 		test_cgroup_iter_sleepable(cgroup_fd, cgroup_id);
+	if (test__start_subtest("yes_rcu_lock"))
+		test_yes_rcu_lock(cgroup_id);
 	if (test__start_subtest("no_rcu_lock"))
-		test_no_rcu_lock(cgroup_id);
-	if (test__start_subtest("rcu_lock"))
-		test_rcu_lock();
+		test_no_rcu_lock();
 
 	close(cgroup_fd);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/rcu_read_lock.c b/tools/testing/selftests/bpf/prog_tests/rcu_read_lock.c
index 447d8560ecb6..3f1f58d3a729 100644
--- a/tools/testing/selftests/bpf/prog_tests/rcu_read_lock.c
+++ b/tools/testing/selftests/bpf/prog_tests/rcu_read_lock.c
@@ -25,10 +25,10 @@ static void test_success(void)
 
 	bpf_program__set_autoload(skel->progs.get_cgroup_id, true);
 	bpf_program__set_autoload(skel->progs.task_succ, true);
-	bpf_program__set_autoload(skel->progs.no_lock, true);
 	bpf_program__set_autoload(skel->progs.two_regions, true);
 	bpf_program__set_autoload(skel->progs.non_sleepable_1, true);
 	bpf_program__set_autoload(skel->progs.non_sleepable_2, true);
+	bpf_program__set_autoload(skel->progs.task_trusted_non_rcuptr, true);
 	err = rcu_read_lock__load(skel);
 	if (!ASSERT_OK(err, "skel_load"))
 		goto out;
@@ -69,6 +69,7 @@ out:
 
 static const char * const inproper_region_tests[] = {
 	"miss_lock",
+	"no_lock",
 	"miss_unlock",
 	"non_sleepable_rcu_mismatch",
 	"inproper_sleepable_helper",
@@ -99,7 +100,6 @@ out:
 }
 
 static const char * const rcuptr_misuse_tests[] = {
-	"task_untrusted_non_rcuptr",
 	"task_untrusted_rcuptr",
 	"cross_rcu_region",
 };
@@ -128,17 +128,8 @@ out:
 
 void test_rcu_read_lock(void)
 {
-	struct btf *vmlinux_btf;
 	int cgroup_fd;
 
-	vmlinux_btf = btf__load_vmlinux_btf();
-	if (!ASSERT_OK_PTR(vmlinux_btf, "could not load vmlinux BTF"))
-		return;
-	if (btf__find_by_name_kind(vmlinux_btf, "rcu", BTF_KIND_TYPE_TAG) < 0) {
-		test__skip();
-		goto out;
-	}
-
 	cgroup_fd = test__join_cgroup("/rcu_read_lock");
 	if (!ASSERT_GE(cgroup_fd, 0, "join_cgroup /rcu_read_lock"))
 		goto out;
@@ -153,6 +144,5 @@ void test_rcu_read_lock(void)
 	if (test__start_subtest("negative_tests_rcuptr_misuse"))
 		test_rcuptr_misuse();
 	close(cgroup_fd);
-out:
-	btf__free(vmlinux_btf);
+out:;
 }
diff --git a/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c b/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c
index 2d11ed528b6f..7615dc23d301 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c
@@ -49,7 +49,7 @@ int no_rcu_lock(void *ctx)
 	if (task->pid != target_pid)
 		return 0;
 
-	/* ptr_to_btf_id semantics. should work. */
+	/* task->cgroups is untrusted in sleepable prog outside of RCU CS */
 	cgrp = task->cgroups->dfl_cgrp;
 	ptr = bpf_cgrp_storage_get(&map_a, cgrp, 0,
 				   BPF_LOCAL_STORAGE_GET_F_CREATE);
@@ -71,7 +71,7 @@ int yes_rcu_lock(void *ctx)
 
 	bpf_rcu_read_lock();
 	cgrp = task->cgroups->dfl_cgrp;
-	/* cgrp is untrusted and cannot pass to bpf_cgrp_storage_get() helper. */
+	/* cgrp is trusted under RCU CS */
 	ptr = bpf_cgrp_storage_get(&map_a, cgrp, 0, BPF_LOCAL_STORAGE_GET_F_CREATE);
 	if (ptr)
 		cgroup_id = cgrp->kn->id;
diff --git a/tools/testing/selftests/bpf/progs/cpumask_failure.c b/tools/testing/selftests/bpf/progs/cpumask_failure.c
index 33e8e86dd090..c16f7563b84e 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_failure.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_failure.c
@@ -44,7 +44,7 @@ int BPF_PROG(test_alloc_double_release, struct task_struct *task, u64 clone_flag
 }
 
 SEC("tp_btf/task_newtask")
-__failure __msg("bpf_cpumask_acquire args#0 expected pointer to STRUCT bpf_cpumask")
+__failure __msg("must be referenced")
 int BPF_PROG(test_acquire_wrong_cpumask, struct task_struct *task, u64 clone_flags)
 {
 	struct bpf_cpumask *cpumask;
diff --git a/tools/testing/selftests/bpf/progs/nested_trust_failure.c b/tools/testing/selftests/bpf/progs/nested_trust_failure.c
index 14aff7676436..0d1aa6bbace4 100644
--- a/tools/testing/selftests/bpf/progs/nested_trust_failure.c
+++ b/tools/testing/selftests/bpf/progs/nested_trust_failure.c
@@ -17,7 +17,7 @@ char _license[] SEC("license") = "GPL";
  */
 
 SEC("tp_btf/task_newtask")
-__failure __msg("R2 must be referenced or trusted")
+__failure __msg("R2 must be")
 int BPF_PROG(test_invalid_nested_user_cpus, struct task_struct *task, u64 clone_flags)
 {
 	bpf_cpumask_test_cpu(0, task->user_cpus_ptr);
diff --git a/tools/testing/selftests/bpf/progs/rcu_read_lock.c b/tools/testing/selftests/bpf/progs/rcu_read_lock.c
index 5cecbdbbb16e..7250bb76d18a 100644
--- a/tools/testing/selftests/bpf/progs/rcu_read_lock.c
+++ b/tools/testing/selftests/bpf/progs/rcu_read_lock.c
@@ -81,7 +81,7 @@ int no_lock(void *ctx)
 {
 	struct task_struct *task, *real_parent;
 
-	/* no bpf_rcu_read_lock(), old code still works */
+	/* old style ptr_to_btf_id is not allowed in sleepable */
 	task = bpf_get_current_task_btf();
 	real_parent = task->real_parent;
 	(void)bpf_task_storage_get(&map_a, real_parent, 0, 0);
@@ -286,13 +286,13 @@ out:
 }
 
 SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
-int task_untrusted_non_rcuptr(void *ctx)
+int task_trusted_non_rcuptr(void *ctx)
 {
 	struct task_struct *task, *group_leader;
 
 	task = bpf_get_current_task_btf();
 	bpf_rcu_read_lock();
-	/* the pointer group_leader marked as untrusted */
+	/* the pointer group_leader is explicitly marked as trusted */
 	group_leader = task->real_parent->group_leader;
 	(void)bpf_task_storage_get(&map_a, group_leader, 0, 0);
 	bpf_rcu_read_unlock();
diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c
index 9a326a800e5c..5702fc9761ef 100644
--- a/tools/testing/selftests/bpf/verifier/calls.c
+++ b/tools/testing/selftests/bpf/verifier/calls.c
@@ -181,7 +181,7 @@
 	},
 	.result_unpriv = REJECT,
 	.result = REJECT,
-	.errstr = "negative offset ptr_ ptr R1 off=-4 disallowed",
+	.errstr = "ptr R1 off=-4 disallowed",
 },
 {
 	"calls: invalid kfunc call: PTR_TO_BTF_ID with variable offset",
-- 
cgit v1.2.3-70-g09d2


From e768e3c5aab44ee63f58649d4c8cbbb3270e5c06 Mon Sep 17 00:00:00 2001
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Date: Fri, 3 Mar 2023 15:15:42 +0100
Subject: bpf: Use separate RCU callbacks for freeing selem

Martin suggested that instead of using a byte in the hole (which he has
a use for in his future patch) in bpf_local_storage_elem, we can
dispatch a different call_rcu callback based on whether we need to free
special fields in bpf_local_storage_elem data. The free path, described
in commit 9db44fdd8105 ("bpf: Support kptrs in local storage maps"),
only waits for call_rcu callbacks when there are special (kptrs, etc.)
fields in the map value, hence it is necessary that we only access
smap in this case.

Therefore, dispatch different RCU callbacks based on the BPF map has a
valid btf_record, which dereference and use smap's btf_record only when
it is valid.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20230303141542.300068-1-memxor@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf_local_storage.h |  6 ---
 kernel/bpf/bpf_local_storage.c    | 79 ++++++++++++++++++++++++---------------
 2 files changed, 49 insertions(+), 36 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 0fe92986412b..6d37a40cd90e 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -74,12 +74,6 @@ struct bpf_local_storage_elem {
 	struct hlist_node snode;	/* Linked to bpf_local_storage */
 	struct bpf_local_storage __rcu *local_storage;
 	struct rcu_head rcu;
-	bool can_use_smap; /* Is it safe to access smap in bpf_selem_free_* RCU
-			    * callbacks? bpf_local_storage_map_free only
-			    * executes rcu_barrier when there are special
-			    * fields, this field remembers that to ensure we
-			    * don't access already freed smap in sdata.
-			    */
 	/* 8 bytes hole */
 	/* The data is stored in another cacheline to minimize
 	 * the number of cachelines access during a cache hit.
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 2bdd722fe293..3d320393a12c 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -109,30 +109,36 @@ void bpf_local_storage_free_rcu(struct rcu_head *rcu)
 		kfree_rcu(local_storage, rcu);
 }
 
-static void bpf_selem_free_rcu(struct rcu_head *rcu)
+static void bpf_selem_free_fields_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage_elem *selem;
+	struct bpf_local_storage_map *smap;
 
 	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
-	/* The can_use_smap bool is set whenever we need to free additional
-	 * fields in selem data before freeing selem. bpf_local_storage_map_free
-	 * only executes rcu_barrier to wait for RCU callbacks when it has
-	 * special fields, hence we can only conditionally dereference smap, as
-	 * by this time the map might have already been freed without waiting
-	 * for our call_rcu callback if it did not have any special fields.
-	 */
-	if (selem->can_use_smap)
-		bpf_obj_free_fields(SDATA(selem)->smap->map.record, SDATA(selem)->data);
+	/* protected by the rcu_barrier*() */
+	smap = rcu_dereference_protected(SDATA(selem)->smap, true);
+	bpf_obj_free_fields(smap->map.record, SDATA(selem)->data);
 	kfree(selem);
 }
 
-static void bpf_selem_free_tasks_trace_rcu(struct rcu_head *rcu)
+static void bpf_selem_free_fields_trace_rcu(struct rcu_head *rcu)
 {
 	/* Free directly if Tasks Trace RCU GP also implies RCU GP */
 	if (rcu_trace_implies_rcu_gp())
-		bpf_selem_free_rcu(rcu);
+		bpf_selem_free_fields_rcu(rcu);
+	else
+		call_rcu(rcu, bpf_selem_free_fields_rcu);
+}
+
+static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
+{
+	struct bpf_local_storage_elem *selem;
+
+	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
+	if (rcu_trace_implies_rcu_gp())
+		kfree(selem);
 	else
-		call_rcu(rcu, bpf_selem_free_rcu);
+		kfree_rcu(selem, rcu);
 }
 
 /* local_storage->lock must be held and selem->local_storage == local_storage.
@@ -145,6 +151,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 {
 	struct bpf_local_storage_map *smap;
 	bool free_local_storage;
+	struct btf_record *rec;
 	void *owner;
 
 	smap = rcu_dereference_check(SDATA(selem)->smap, bpf_rcu_lock_held());
@@ -185,10 +192,26 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	    SDATA(selem))
 		RCU_INIT_POINTER(local_storage->cache[smap->cache_idx], NULL);
 
-	if (use_trace_rcu)
-		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_tasks_trace_rcu);
-	else
-		call_rcu(&selem->rcu, bpf_selem_free_rcu);
+	/* A different RCU callback is chosen whenever we need to free
+	 * additional fields in selem data before freeing selem.
+	 * bpf_local_storage_map_free only executes rcu_barrier to wait for RCU
+	 * callbacks when it has special fields, hence we can only conditionally
+	 * dereference smap, as by this time the map might have already been
+	 * freed without waiting for our call_rcu callback if it did not have
+	 * any special fields.
+	 */
+	rec = smap->map.record;
+	if (use_trace_rcu) {
+		if (!IS_ERR_OR_NULL(rec))
+			call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_fields_trace_rcu);
+		else
+			call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
+	} else {
+		if (!IS_ERR_OR_NULL(rec))
+			call_rcu(&selem->rcu, bpf_selem_free_fields_rcu);
+		else
+			kfree_rcu(selem, rcu);
+	}
 
 	return free_local_storage;
 }
@@ -256,11 +279,6 @@ void bpf_selem_link_map(struct bpf_local_storage_map *smap,
 	RCU_INIT_POINTER(SDATA(selem)->smap, smap);
 	hlist_add_head_rcu(&selem->map_node, &b->list);
 	raw_spin_unlock_irqrestore(&b->lock, flags);
-
-	/* If our data will have special fields, smap will wait for us to use
-	 * its record in bpf_selem_free_* RCU callbacks before freeing itself.
-	 */
-	selem->can_use_smap = !IS_ERR_OR_NULL(smap->map.record);
 }
 
 void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu)
@@ -748,19 +766,20 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 	kvfree(smap->buckets);
 
 	/* When local storage has special fields, callbacks for
-	 * bpf_selem_free_rcu and bpf_selem_free_tasks_trace_rcu will keep using
-	 * the map BTF record, we need to execute an RCU barrier to wait for
-	 * them as the record will be freed right after our map_free callback.
+	 * bpf_selem_free_fields_rcu and bpf_selem_free_fields_trace_rcu will
+	 * keep using the map BTF record, we need to execute an RCU barrier to
+	 * wait for them as the record will be freed right after our map_free
+	 * callback.
 	 */
 	if (!IS_ERR_OR_NULL(smap->map.record)) {
 		rcu_barrier_tasks_trace();
 		/* We cannot skip rcu_barrier() when rcu_trace_implies_rcu_gp()
 		 * is true, because while call_rcu invocation is skipped in that
-		 * case in bpf_selem_free_tasks_trace_rcu (and all local storage
-		 * maps pass use_trace_rcu = true), there can be call_rcu
-		 * callbacks based on use_trace_rcu = false in the earlier while
-		 * ((selem = ...)) loop or from bpf_local_storage_unlink_nolock
-		 * called from owner's free path.
+		 * case in bpf_selem_free_fields_trace_rcu (and all local
+		 * storage maps pass use_trace_rcu = true), there can be
+		 * call_rcu callbacks based on use_trace_rcu = false in the
+		 * while ((selem = ...)) loop above or when owner's free path
+		 * calls bpf_local_storage_unlink_nolock.
 		 */
 		rcu_barrier();
 	}
-- 
cgit v1.2.3-70-g09d2


From 0d80a619c113d0e216dbffa56b2d5ccc079ee520 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 4 Mar 2023 03:12:45 +0200
Subject: bpf: allow ctx writes using BPF_ST_MEM instruction

Lift verifier restriction to use BPF_ST_MEM instructions to write to
context data structures. This requires the following changes:
 - verifier.c:do_check() for BPF_ST updated to:
   - no longer forbid writes to registers of type PTR_TO_CTX;
   - track dst_reg type in the env->insn_aux_data[...].ptr_type field
     (same way it is done for BPF_STX and BPF_LDX instructions).
 - verifier.c:convert_ctx_access() and various callbacks invoked by
   it are updated to handled BPF_ST instruction alongside BPF_STX.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230304011247.566040-2-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/cgroup.c                        |  49 ++++++++-----
 kernel/bpf/verifier.c                      | 110 ++++++++++++++---------------
 net/core/filter.c                          |  79 ++++++++++++---------
 tools/testing/selftests/bpf/verifier/ctx.c |  11 ---
 4 files changed, 126 insertions(+), 123 deletions(-)

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index a4ae422b8f12..53edb8ad2471 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -2223,10 +2223,12 @@ static u32 sysctl_convert_ctx_access(enum bpf_access_type type,
 				BPF_FIELD_SIZEOF(struct bpf_sysctl_kern, ppos),
 				treg, si->dst_reg,
 				offsetof(struct bpf_sysctl_kern, ppos));
-			*insn++ = BPF_STX_MEM(
-				BPF_SIZEOF(u32), treg, si->src_reg,
+			*insn++ = BPF_RAW_INSN(
+				BPF_CLASS(si->code) | BPF_MEM | BPF_SIZEOF(u32),
+				treg, si->src_reg,
 				bpf_ctx_narrow_access_offset(
-					0, sizeof(u32), sizeof(loff_t)));
+					0, sizeof(u32), sizeof(loff_t)),
+				si->imm);
 			*insn++ = BPF_LDX_MEM(
 				BPF_DW, treg, si->dst_reg,
 				offsetof(struct bpf_sysctl_kern, tmp_reg));
@@ -2376,10 +2378,17 @@ static bool cg_sockopt_is_valid_access(int off, int size,
 	return true;
 }
 
-#define CG_SOCKOPT_ACCESS_FIELD(T, F)					\
-	T(BPF_FIELD_SIZEOF(struct bpf_sockopt_kern, F),			\
-	  si->dst_reg, si->src_reg,					\
-	  offsetof(struct bpf_sockopt_kern, F))
+#define CG_SOCKOPT_READ_FIELD(F)					\
+	BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_sockopt_kern, F),	\
+		    si->dst_reg, si->src_reg,				\
+		    offsetof(struct bpf_sockopt_kern, F))
+
+#define CG_SOCKOPT_WRITE_FIELD(F)					\
+	BPF_RAW_INSN((BPF_FIELD_SIZEOF(struct bpf_sockopt_kern, F) |	\
+		      BPF_MEM | BPF_CLASS(si->code)),			\
+		     si->dst_reg, si->src_reg,				\
+		     offsetof(struct bpf_sockopt_kern, F),		\
+		     si->imm)
 
 static u32 cg_sockopt_convert_ctx_access(enum bpf_access_type type,
 					 const struct bpf_insn *si,
@@ -2391,25 +2400,25 @@ static u32 cg_sockopt_convert_ctx_access(enum bpf_access_type type,
 
 	switch (si->off) {
 	case offsetof(struct bpf_sockopt, sk):
-		*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, sk);
+		*insn++ = CG_SOCKOPT_READ_FIELD(sk);
 		break;
 	case offsetof(struct bpf_sockopt, level):
 		if (type == BPF_WRITE)
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_STX_MEM, level);
+			*insn++ = CG_SOCKOPT_WRITE_FIELD(level);
 		else
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, level);
+			*insn++ = CG_SOCKOPT_READ_FIELD(level);
 		break;
 	case offsetof(struct bpf_sockopt, optname):
 		if (type == BPF_WRITE)
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_STX_MEM, optname);
+			*insn++ = CG_SOCKOPT_WRITE_FIELD(optname);
 		else
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, optname);
+			*insn++ = CG_SOCKOPT_READ_FIELD(optname);
 		break;
 	case offsetof(struct bpf_sockopt, optlen):
 		if (type == BPF_WRITE)
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_STX_MEM, optlen);
+			*insn++ = CG_SOCKOPT_WRITE_FIELD(optlen);
 		else
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, optlen);
+			*insn++ = CG_SOCKOPT_READ_FIELD(optlen);
 		break;
 	case offsetof(struct bpf_sockopt, retval):
 		BUILD_BUG_ON(offsetof(struct bpf_cg_run_ctx, run_ctx) != 0);
@@ -2429,9 +2438,11 @@ static u32 cg_sockopt_convert_ctx_access(enum bpf_access_type type,
 			*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct task_struct, bpf_ctx),
 					      treg, treg,
 					      offsetof(struct task_struct, bpf_ctx));
-			*insn++ = BPF_STX_MEM(BPF_FIELD_SIZEOF(struct bpf_cg_run_ctx, retval),
-					      treg, si->src_reg,
-					      offsetof(struct bpf_cg_run_ctx, retval));
+			*insn++ = BPF_RAW_INSN(BPF_CLASS(si->code) | BPF_MEM |
+					       BPF_FIELD_SIZEOF(struct bpf_cg_run_ctx, retval),
+					       treg, si->src_reg,
+					       offsetof(struct bpf_cg_run_ctx, retval),
+					       si->imm);
 			*insn++ = BPF_LDX_MEM(BPF_DW, treg, si->dst_reg,
 					      offsetof(struct bpf_sockopt_kern, tmp_reg));
 		} else {
@@ -2447,10 +2458,10 @@ static u32 cg_sockopt_convert_ctx_access(enum bpf_access_type type,
 		}
 		break;
 	case offsetof(struct bpf_sockopt, optval):
-		*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, optval);
+		*insn++ = CG_SOCKOPT_READ_FIELD(optval);
 		break;
 	case offsetof(struct bpf_sockopt, optval_end):
-		*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, optval_end);
+		*insn++ = CG_SOCKOPT_READ_FIELD(optval_end);
 		break;
 	}
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index c2adf3c24c64..4c5d2b5e25c8 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -14813,6 +14813,44 @@ static bool reg_type_mismatch(enum bpf_reg_type src, enum bpf_reg_type prev)
 			       !reg_type_mismatch_ok(prev));
 }
 
+static int save_aux_ptr_type(struct bpf_verifier_env *env, enum bpf_reg_type type,
+			     bool allow_trust_missmatch)
+{
+	enum bpf_reg_type *prev_type = &env->insn_aux_data[env->insn_idx].ptr_type;
+
+	if (*prev_type == NOT_INIT) {
+		/* Saw a valid insn
+		 * dst_reg = *(u32 *)(src_reg + off)
+		 * save type to validate intersecting paths
+		 */
+		*prev_type = type;
+	} else if (reg_type_mismatch(type, *prev_type)) {
+		/* Abuser program is trying to use the same insn
+		 * dst_reg = *(u32*) (src_reg + off)
+		 * with different pointer types:
+		 * src_reg == ctx in one branch and
+		 * src_reg == stack|map in some other branch.
+		 * Reject it.
+		 */
+		if (allow_trust_missmatch &&
+		    base_type(type) == PTR_TO_BTF_ID &&
+		    base_type(*prev_type) == PTR_TO_BTF_ID) {
+			/*
+			 * Have to support a use case when one path through
+			 * the program yields TRUSTED pointer while another
+			 * is UNTRUSTED. Fallback to UNTRUSTED to generate
+			 * BPF_PROBE_MEM.
+			 */
+			*prev_type = PTR_TO_BTF_ID | PTR_UNTRUSTED;
+		} else {
+			verbose(env, "same insn cannot be used with different pointers\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
 static int do_check(struct bpf_verifier_env *env)
 {
 	bool pop_log = !(env->log.level & BPF_LOG_LEVEL2);
@@ -14922,7 +14960,7 @@ static int do_check(struct bpf_verifier_env *env)
 				return err;
 
 		} else if (class == BPF_LDX) {
-			enum bpf_reg_type *prev_src_type, src_reg_type;
+			enum bpf_reg_type src_reg_type;
 
 			/* check for reserved fields is already done */
 
@@ -14946,43 +14984,11 @@ static int do_check(struct bpf_verifier_env *env)
 			if (err)
 				return err;
 
-			prev_src_type = &env->insn_aux_data[env->insn_idx].ptr_type;
-
-			if (*prev_src_type == NOT_INIT) {
-				/* saw a valid insn
-				 * dst_reg = *(u32 *)(src_reg + off)
-				 * save type to validate intersecting paths
-				 */
-				*prev_src_type = src_reg_type;
-
-			} else if (reg_type_mismatch(src_reg_type, *prev_src_type)) {
-				/* ABuser program is trying to use the same insn
-				 * dst_reg = *(u32*) (src_reg + off)
-				 * with different pointer types:
-				 * src_reg == ctx in one branch and
-				 * src_reg == stack|map in some other branch.
-				 * Reject it.
-				 */
-				if (base_type(src_reg_type) == PTR_TO_BTF_ID &&
-				    base_type(*prev_src_type) == PTR_TO_BTF_ID) {
-					/*
-					 * Have to support a use case when one path through
-					 * the program yields TRUSTED pointer while another
-					 * is UNTRUSTED. Fallback to UNTRUSTED to generate
-					 * BPF_PROBE_MEM.
-					 */
-					*prev_src_type = PTR_TO_BTF_ID | PTR_UNTRUSTED;
-				} else {
-					verbose(env,
-						"The same insn cannot be used with different pointers: %s",
-						reg_type_str(env, src_reg_type));
-					verbose(env, " != %s\n", reg_type_str(env, *prev_src_type));
-					return -EINVAL;
-				}
-			}
-
+			err = save_aux_ptr_type(env, src_reg_type, true);
+			if (err)
+				return err;
 		} else if (class == BPF_STX) {
-			enum bpf_reg_type *prev_dst_type, dst_reg_type;
+			enum bpf_reg_type dst_reg_type;
 
 			if (BPF_MODE(insn->code) == BPF_ATOMIC) {
 				err = check_atomic(env, env->insn_idx, insn);
@@ -15015,16 +15021,12 @@ static int do_check(struct bpf_verifier_env *env)
 			if (err)
 				return err;
 
-			prev_dst_type = &env->insn_aux_data[env->insn_idx].ptr_type;
-
-			if (*prev_dst_type == NOT_INIT) {
-				*prev_dst_type = dst_reg_type;
-			} else if (reg_type_mismatch(dst_reg_type, *prev_dst_type)) {
-				verbose(env, "same insn cannot be used with different pointers\n");
-				return -EINVAL;
-			}
-
+			err = save_aux_ptr_type(env, dst_reg_type, false);
+			if (err)
+				return err;
 		} else if (class == BPF_ST) {
+			enum bpf_reg_type dst_reg_type;
+
 			if (BPF_MODE(insn->code) != BPF_MEM ||
 			    insn->src_reg != BPF_REG_0) {
 				verbose(env, "BPF_ST uses reserved fields\n");
@@ -15035,12 +15037,7 @@ static int do_check(struct bpf_verifier_env *env)
 			if (err)
 				return err;
 
-			if (is_ctx_reg(env, insn->dst_reg)) {
-				verbose(env, "BPF_ST stores into R%d %s is not allowed\n",
-					insn->dst_reg,
-					reg_type_str(env, reg_state(env, insn->dst_reg)->type));
-				return -EACCES;
-			}
+			dst_reg_type = regs[insn->dst_reg].type;
 
 			/* check that memory (dst_reg + off) is writeable */
 			err = check_mem_access(env, env->insn_idx, insn->dst_reg,
@@ -15049,6 +15046,9 @@ static int do_check(struct bpf_verifier_env *env)
 			if (err)
 				return err;
 
+			err = save_aux_ptr_type(env, dst_reg_type, false);
+			if (err)
+				return err;
 		} else if (class == BPF_JMP || class == BPF_JMP32) {
 			u8 opcode = BPF_OP(insn->code);
 
@@ -16157,14 +16157,12 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env)
 
 	for (i = 0; i < insn_cnt; i++, insn++) {
 		bpf_convert_ctx_access_t convert_ctx_access;
-		bool ctx_access;
 
 		if (insn->code == (BPF_LDX | BPF_MEM | BPF_B) ||
 		    insn->code == (BPF_LDX | BPF_MEM | BPF_H) ||
 		    insn->code == (BPF_LDX | BPF_MEM | BPF_W) ||
 		    insn->code == (BPF_LDX | BPF_MEM | BPF_DW)) {
 			type = BPF_READ;
-			ctx_access = true;
 		} else if (insn->code == (BPF_STX | BPF_MEM | BPF_B) ||
 			   insn->code == (BPF_STX | BPF_MEM | BPF_H) ||
 			   insn->code == (BPF_STX | BPF_MEM | BPF_W) ||
@@ -16174,7 +16172,6 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env)
 			   insn->code == (BPF_ST | BPF_MEM | BPF_W) ||
 			   insn->code == (BPF_ST | BPF_MEM | BPF_DW)) {
 			type = BPF_WRITE;
-			ctx_access = BPF_CLASS(insn->code) == BPF_STX;
 		} else {
 			continue;
 		}
@@ -16197,9 +16194,6 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env)
 			continue;
 		}
 
-		if (!ctx_access)
-			continue;
-
 		switch ((int)env->insn_aux_data[i + delta].ptr_type) {
 		case PTR_TO_CTX:
 			if (!ops->convert_ctx_access)
diff --git a/net/core/filter.c b/net/core/filter.c
index a2dc44e70ea0..50f649f1b4a9 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -9279,11 +9279,15 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog,
 #endif
 
 	/* <store>: skb->tstamp = tstamp */
-	*insn++ = BPF_STX_MEM(BPF_DW, skb_reg, value_reg,
-			      offsetof(struct sk_buff, tstamp));
+	*insn++ = BPF_RAW_INSN(BPF_CLASS(si->code) | BPF_DW | BPF_MEM,
+			       skb_reg, value_reg, offsetof(struct sk_buff, tstamp), si->imm);
 	return insn;
 }
 
+#define BPF_EMIT_STORE(size, si, off)					\
+	BPF_RAW_INSN(BPF_CLASS((si)->code) | (size) | BPF_MEM,		\
+		     (si)->dst_reg, (si)->src_reg, (off), (si)->imm)
+
 static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 				  const struct bpf_insn *si,
 				  struct bpf_insn *insn_buf,
@@ -9313,9 +9317,9 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 
 	case offsetof(struct __sk_buff, priority):
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg,
-					      bpf_target_off(struct sk_buff, priority, 4,
-							     target_size));
+			*insn++ = BPF_EMIT_STORE(BPF_W, si,
+						 bpf_target_off(struct sk_buff, priority, 4,
+								target_size));
 		else
 			*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
 					      bpf_target_off(struct sk_buff, priority, 4,
@@ -9346,9 +9350,9 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 
 	case offsetof(struct __sk_buff, mark):
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg,
-					      bpf_target_off(struct sk_buff, mark, 4,
-							     target_size));
+			*insn++ = BPF_EMIT_STORE(BPF_W, si,
+						 bpf_target_off(struct sk_buff, mark, 4,
+								target_size));
 		else
 			*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
 					      bpf_target_off(struct sk_buff, mark, 4,
@@ -9367,11 +9371,16 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 
 	case offsetof(struct __sk_buff, queue_mapping):
 		if (type == BPF_WRITE) {
-			*insn++ = BPF_JMP_IMM(BPF_JGE, si->src_reg, NO_QUEUE_MAPPING, 1);
-			*insn++ = BPF_STX_MEM(BPF_H, si->dst_reg, si->src_reg,
-					      bpf_target_off(struct sk_buff,
-							     queue_mapping,
-							     2, target_size));
+			u32 off = bpf_target_off(struct sk_buff, queue_mapping, 2, target_size);
+
+			if (BPF_CLASS(si->code) == BPF_ST && si->imm >= NO_QUEUE_MAPPING) {
+				*insn++ = BPF_JMP_A(0); /* noop */
+				break;
+			}
+
+			if (BPF_CLASS(si->code) == BPF_STX)
+				*insn++ = BPF_JMP_IMM(BPF_JGE, si->src_reg, NO_QUEUE_MAPPING, 1);
+			*insn++ = BPF_EMIT_STORE(BPF_H, si, off);
 		} else {
 			*insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->src_reg,
 					      bpf_target_off(struct sk_buff,
@@ -9407,8 +9416,7 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 		off += offsetof(struct sk_buff, cb);
 		off += offsetof(struct qdisc_skb_cb, data);
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_SIZE(si->code), si->dst_reg,
-					      si->src_reg, off);
+			*insn++ = BPF_EMIT_STORE(BPF_SIZE(si->code), si, off);
 		else
 			*insn++ = BPF_LDX_MEM(BPF_SIZE(si->code), si->dst_reg,
 					      si->src_reg, off);
@@ -9423,8 +9431,7 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 		off += offsetof(struct qdisc_skb_cb, tc_classid);
 		*target_size = 2;
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_H, si->dst_reg,
-					      si->src_reg, off);
+			*insn++ = BPF_EMIT_STORE(BPF_H, si, off);
 		else
 			*insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg,
 					      si->src_reg, off);
@@ -9457,9 +9464,9 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 	case offsetof(struct __sk_buff, tc_index):
 #ifdef CONFIG_NET_SCHED
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_H, si->dst_reg, si->src_reg,
-					      bpf_target_off(struct sk_buff, tc_index, 2,
-							     target_size));
+			*insn++ = BPF_EMIT_STORE(BPF_H, si,
+						 bpf_target_off(struct sk_buff, tc_index, 2,
+								target_size));
 		else
 			*insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->src_reg,
 					      bpf_target_off(struct sk_buff, tc_index, 2,
@@ -9660,8 +9667,8 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 		BUILD_BUG_ON(sizeof_field(struct sock, sk_bound_dev_if) != 4);
 
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg,
-					offsetof(struct sock, sk_bound_dev_if));
+			*insn++ = BPF_EMIT_STORE(BPF_W, si,
+						 offsetof(struct sock, sk_bound_dev_if));
 		else
 			*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
 				      offsetof(struct sock, sk_bound_dev_if));
@@ -9671,8 +9678,8 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 		BUILD_BUG_ON(sizeof_field(struct sock, sk_mark) != 4);
 
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg,
-					offsetof(struct sock, sk_mark));
+			*insn++ = BPF_EMIT_STORE(BPF_W, si,
+						 offsetof(struct sock, sk_mark));
 		else
 			*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
 				      offsetof(struct sock, sk_mark));
@@ -9682,8 +9689,8 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 		BUILD_BUG_ON(sizeof_field(struct sock, sk_priority) != 4);
 
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg,
-					offsetof(struct sock, sk_priority));
+			*insn++ = BPF_EMIT_STORE(BPF_W, si,
+						 offsetof(struct sock, sk_priority));
 		else
 			*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
 				      offsetof(struct sock, sk_priority));
@@ -9948,10 +9955,12 @@ static u32 xdp_convert_ctx_access(enum bpf_access_type type,
 				      offsetof(S, TF));			       \
 		*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(S, F), tmp_reg,	       \
 				      si->dst_reg, offsetof(S, F));	       \
-		*insn++ = BPF_STX_MEM(SIZE, tmp_reg, si->src_reg,	       \
+		*insn++ = BPF_RAW_INSN(SIZE | BPF_MEM | BPF_CLASS(si->code),   \
+				       tmp_reg, si->src_reg,		       \
 			bpf_target_off(NS, NF, sizeof_field(NS, NF),	       \
 				       target_size)			       \
-				+ OFF);					       \
+				       + OFF,				       \
+				       si->imm);			       \
 		*insn++ = BPF_LDX_MEM(BPF_DW, tmp_reg, si->dst_reg,	       \
 				      offsetof(S, TF));			       \
 	} while (0)
@@ -10186,9 +10195,11 @@ static u32 sock_ops_convert_ctx_access(enum bpf_access_type type,
 						struct bpf_sock_ops_kern, sk),\
 				      reg, si->dst_reg,			      \
 				      offsetof(struct bpf_sock_ops_kern, sk));\
-		*insn++ = BPF_STX_MEM(BPF_FIELD_SIZEOF(OBJ, OBJ_FIELD),	      \
-				      reg, si->src_reg,			      \
-				      offsetof(OBJ, OBJ_FIELD));	      \
+		*insn++ = BPF_RAW_INSN(BPF_FIELD_SIZEOF(OBJ, OBJ_FIELD) |     \
+				       BPF_MEM | BPF_CLASS(si->code),	      \
+				       reg, si->src_reg,		      \
+				       offsetof(OBJ, OBJ_FIELD),	      \
+				       si->imm);			      \
 		*insn++ = BPF_LDX_MEM(BPF_DW, reg, si->dst_reg,		      \
 				      offsetof(struct bpf_sock_ops_kern,      \
 					       temp));			      \
@@ -10220,8 +10231,7 @@ static u32 sock_ops_convert_ctx_access(enum bpf_access_type type,
 		off -= offsetof(struct bpf_sock_ops, replylong[0]);
 		off += offsetof(struct bpf_sock_ops_kern, replylong[0]);
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg,
-					      off);
+			*insn++ = BPF_EMIT_STORE(BPF_W, si, off);
 		else
 			*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg,
 					      off);
@@ -10578,8 +10588,7 @@ static u32 sk_skb_convert_ctx_access(enum bpf_access_type type,
 		off += offsetof(struct sk_buff, cb);
 		off += offsetof(struct sk_skb_cb, data);
 		if (type == BPF_WRITE)
-			*insn++ = BPF_STX_MEM(BPF_SIZE(si->code), si->dst_reg,
-					      si->src_reg, off);
+			*insn++ = BPF_EMIT_STORE(BPF_SIZE(si->code), si, off);
 		else
 			*insn++ = BPF_LDX_MEM(BPF_SIZE(si->code), si->dst_reg,
 					      si->src_reg, off);
diff --git a/tools/testing/selftests/bpf/verifier/ctx.c b/tools/testing/selftests/bpf/verifier/ctx.c
index c8eaf0536c24..2fd31612c0b8 100644
--- a/tools/testing/selftests/bpf/verifier/ctx.c
+++ b/tools/testing/selftests/bpf/verifier/ctx.c
@@ -1,14 +1,3 @@
-{
-	"context stores via ST",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_1, offsetof(struct __sk_buff, mark), 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "BPF_ST stores into R1 ctx is not allowed",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
 {
 	"context stores via BPF_ATOMIC",
 	.insns = {
-- 
cgit v1.2.3-70-g09d2


From 806f81cd1ee30c66a3d2a4cd18b13c97429397a0 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 4 Mar 2023 03:12:46 +0200
Subject: selftests/bpf: test if pointer type is tracked for BPF_ST_MEM

Check that verifier tracks pointer types for BPF_ST_MEM instructions
and reports error if pointer types do not match for different
execution branches.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230304011247.566040-3-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/verifier/unpriv.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/tools/testing/selftests/bpf/verifier/unpriv.c b/tools/testing/selftests/bpf/verifier/unpriv.c
index 878ca26c3f0a..af0c0f336625 100644
--- a/tools/testing/selftests/bpf/verifier/unpriv.c
+++ b/tools/testing/selftests/bpf/verifier/unpriv.c
@@ -239,6 +239,29 @@
 	.errstr = "same insn cannot be used with different pointers",
 	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
 },
+{
+	/* Same as above, but use BPF_ST_MEM to save 42
+	 * instead of BPF_STX_MEM.
+	 */
+	"unpriv: spill/fill of different pointers st",
+	.insns = {
+	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
+	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
+	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -16),
+	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
+	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
+	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
+	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
+	BPF_ST_MEM(BPF_W, BPF_REG_1, offsetof(struct __sk_buff, mark), 42),
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = REJECT,
+	.errstr = "same insn cannot be used with different pointers",
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+},
 {
 	"unpriv: spill/fill of different pointers stx - ctx and sock",
 	.insns = {
-- 
cgit v1.2.3-70-g09d2


From 71cf4d027ad53a1e2847191ac14e50132d35a6a7 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 4 Mar 2023 03:12:47 +0200
Subject: selftests/bpf: Disassembler tests for verifier.c:convert_ctx_access()

Function verifier.c:convert_ctx_access() applies some rewrites to BPF
instructions that read or write BPF program context. This commit adds
machinery to allow test cases that inspect BPF program after these
rewrites are applied.

An example of a test case:

  {
        // Shorthand for field offset and size specification
	N(CGROUP_SOCKOPT, struct bpf_sockopt, retval),

        // Pattern generated for field read
	.read  = "$dst = *(u64 *)($ctx + bpf_sockopt_kern::current_task);"
		 "$dst = *(u64 *)($dst + task_struct::bpf_ctx);"
		 "$dst = *(u32 *)($dst + bpf_cg_run_ctx::retval);",

        // Pattern generated for field write
	.write = "*(u64 *)($ctx + bpf_sockopt_kern::tmp_reg) = r9;"
		 "r9 = *(u64 *)($ctx + bpf_sockopt_kern::current_task);"
		 "r9 = *(u64 *)(r9 + task_struct::bpf_ctx);"
		 "*(u32 *)(r9 + bpf_cg_run_ctx::retval) = $src;"
		 "r9 = *(u64 *)($ctx + bpf_sockopt_kern::tmp_reg);" ,
  },

For each test case, up to three programs are created:
- One that uses BPF_LDX_MEM to read the context field.
- One that uses BPF_STX_MEM to write to the context field.
- One that uses BPF_ST_MEM to write to the context field.

The disassembly of each program is compared with the pattern specified
in the test case.

Kernel code for disassembly is reused (as is in the bpftool).
To keep Makefile changes to the minimum, symbolic links to
`kernel/bpf/disasm.c` and `kernel/bpf/disasm.h ` are added.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230304011247.566040-4-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/Makefile               |   2 +-
 tools/testing/selftests/bpf/disasm.c               |   1 +
 tools/testing/selftests/bpf/disasm.h               |   1 +
 .../testing/selftests/bpf/prog_tests/ctx_rewrite.c | 917 +++++++++++++++++++++
 4 files changed, 920 insertions(+), 1 deletion(-)
 create mode 120000 tools/testing/selftests/bpf/disasm.c
 create mode 120000 tools/testing/selftests/bpf/disasm.h
 create mode 100644 tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index eab3cf5399f5..16f404aa1b23 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -559,7 +559,7 @@ TRUNNER_BPF_PROGS_DIR := progs
 TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c	\
 			 network_helpers.c testing_helpers.c		\
 			 btf_helpers.c flow_dissector_load.h		\
-			 cap_helpers.c test_loader.c xsk.c
+			 cap_helpers.c test_loader.c xsk.c disasm.c
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read $(OUTPUT)/bpf_testmod.ko	\
 		       $(OUTPUT)/liburandom_read.so			\
 		       $(OUTPUT)/xdp_synproxy				\
diff --git a/tools/testing/selftests/bpf/disasm.c b/tools/testing/selftests/bpf/disasm.c
new file mode 120000
index 000000000000..b1571927bd54
--- /dev/null
+++ b/tools/testing/selftests/bpf/disasm.c
@@ -0,0 +1 @@
+../../../../kernel/bpf/disasm.c
\ No newline at end of file
diff --git a/tools/testing/selftests/bpf/disasm.h b/tools/testing/selftests/bpf/disasm.h
new file mode 120000
index 000000000000..8054fd497340
--- /dev/null
+++ b/tools/testing/selftests/bpf/disasm.h
@@ -0,0 +1 @@
+../../../../kernel/bpf/disasm.h
\ No newline at end of file
diff --git a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
new file mode 100644
index 000000000000..d5fe3d4b936c
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
@@ -0,0 +1,917 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <limits.h>
+#include <stdio.h>
+#include <string.h>
+#include <ctype.h>
+#include <regex.h>
+#include <test_progs.h>
+
+#include "bpf/btf.h"
+#include "bpf_util.h"
+#include "linux/filter.h"
+#include "disasm.h"
+
+#define MAX_PROG_TEXT_SZ (32 * 1024)
+
+/* The code in this file serves the sole purpose of executing test cases
+ * specified in the test_cases array. Each test case specifies a program
+ * type, context field offset, and disassembly patterns that correspond
+ * to read and write instructions generated by
+ * verifier.c:convert_ctx_access() for accessing that field.
+ *
+ * For each test case, up to three programs are created:
+ * - One that uses BPF_LDX_MEM to read the context field.
+ * - One that uses BPF_STX_MEM to write to the context field.
+ * - One that uses BPF_ST_MEM to write to the context field.
+ *
+ * The disassembly of each program is then compared with the pattern
+ * specified in the test case.
+ */
+struct test_case {
+	char *name;
+	enum bpf_prog_type prog_type;
+	enum bpf_attach_type expected_attach_type;
+	int field_offset;
+	int field_sz;
+	/* Program generated for BPF_ST_MEM uses value 42 by default,
+	 * this field allows to specify custom value.
+	 */
+	struct {
+		bool use;
+		int value;
+	} st_value;
+	/* Pattern for BPF_LDX_MEM(field_sz, dst, ctx, field_offset) */
+	char *read;
+	/* Pattern for BPF_STX_MEM(field_sz, ctx, src, field_offset) and
+	 *             BPF_ST_MEM (field_sz, ctx, src, field_offset)
+	 */
+	char *write;
+	/* Pattern for BPF_ST_MEM(field_sz, ctx, src, field_offset),
+	 * takes priority over `write`.
+	 */
+	char *write_st;
+	/* Pattern for BPF_STX_MEM (field_sz, ctx, src, field_offset),
+	 * takes priority over `write`.
+	 */
+	char *write_stx;
+};
+
+#define N(_prog_type, type, field, name_extra...)	\
+	.name = #_prog_type "." #field name_extra,	\
+	.prog_type = BPF_PROG_TYPE_##_prog_type,	\
+	.field_offset = offsetof(type, field),		\
+	.field_sz = sizeof(typeof(((type *)NULL)->field))
+
+static struct test_case test_cases[] = {
+/* Sign extension on s390 changes the pattern */
+#if defined(__x86_64__) || defined(__aarch64__)
+	{
+		N(SCHED_CLS, struct __sk_buff, tstamp),
+		.read  = "r11 = *(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset);"
+			 "w11 &= 160;"
+			 "if w11 != 0xa0 goto pc+2;"
+			 "$dst = 0;"
+			 "goto pc+1;"
+			 "$dst = *(u64 *)($ctx + sk_buff::tstamp);",
+		.write = "r11 = *(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset);"
+			 "if w11 & 0x80 goto pc+1;"
+			 "goto pc+2;"
+			 "w11 &= -33;"
+			 "*(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset) = r11;"
+			 "*(u64 *)($ctx + sk_buff::tstamp) = $src;",
+	},
+#endif
+	{
+		N(SCHED_CLS, struct __sk_buff, priority),
+		.read  = "$dst = *(u32 *)($ctx + sk_buff::priority);",
+		.write = "*(u32 *)($ctx + sk_buff::priority) = $src;",
+	},
+	{
+		N(SCHED_CLS, struct __sk_buff, mark),
+		.read  = "$dst = *(u32 *)($ctx + sk_buff::mark);",
+		.write = "*(u32 *)($ctx + sk_buff::mark) = $src;",
+	},
+	{
+		N(SCHED_CLS, struct __sk_buff, cb[0]),
+		.read  = "$dst = *(u32 *)($ctx + $(sk_buff::cb + qdisc_skb_cb::data));",
+		.write = "*(u32 *)($ctx + $(sk_buff::cb + qdisc_skb_cb::data)) = $src;",
+	},
+	{
+		N(SCHED_CLS, struct __sk_buff, tc_classid),
+		.read  = "$dst = *(u16 *)($ctx + $(sk_buff::cb + qdisc_skb_cb::tc_classid));",
+		.write = "*(u16 *)($ctx + $(sk_buff::cb + qdisc_skb_cb::tc_classid)) = $src;",
+	},
+	{
+		N(SCHED_CLS, struct __sk_buff, tc_index),
+		.read  = "$dst = *(u16 *)($ctx + sk_buff::tc_index);",
+		.write = "*(u16 *)($ctx + sk_buff::tc_index) = $src;",
+	},
+	{
+		N(SCHED_CLS, struct __sk_buff, queue_mapping),
+		.read      = "$dst = *(u16 *)($ctx + sk_buff::queue_mapping);",
+		.write_stx = "if $src >= 0xffff goto pc+1;"
+			     "*(u16 *)($ctx + sk_buff::queue_mapping) = $src;",
+		.write_st  = "*(u16 *)($ctx + sk_buff::queue_mapping) = $src;",
+	},
+	{
+		/* This is a corner case in filter.c:bpf_convert_ctx_access() */
+		N(SCHED_CLS, struct __sk_buff, queue_mapping, ".ushrt_max"),
+		.st_value = { true, USHRT_MAX },
+		.write_st = "goto pc+0;",
+	},
+	{
+		N(CGROUP_SOCK, struct bpf_sock, bound_dev_if),
+		.read  = "$dst = *(u32 *)($ctx + sock_common::skc_bound_dev_if);",
+		.write = "*(u32 *)($ctx + sock_common::skc_bound_dev_if) = $src;",
+	},
+	{
+		N(CGROUP_SOCK, struct bpf_sock, mark),
+		.read  = "$dst = *(u32 *)($ctx + sock::sk_mark);",
+		.write = "*(u32 *)($ctx + sock::sk_mark) = $src;",
+	},
+	{
+		N(CGROUP_SOCK, struct bpf_sock, priority),
+		.read  = "$dst = *(u32 *)($ctx + sock::sk_priority);",
+		.write = "*(u32 *)($ctx + sock::sk_priority) = $src;",
+	},
+	{
+		N(SOCK_OPS, struct bpf_sock_ops, replylong[0]),
+		.read  = "$dst = *(u32 *)($ctx + bpf_sock_ops_kern::replylong);",
+		.write = "*(u32 *)($ctx + bpf_sock_ops_kern::replylong) = $src;",
+	},
+	{
+		N(CGROUP_SYSCTL, struct bpf_sysctl, file_pos),
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+		.read  = "$dst = *(u64 *)($ctx + bpf_sysctl_kern::ppos);"
+			 "$dst = *(u32 *)($dst +0);",
+		.write = "*(u64 *)($ctx + bpf_sysctl_kern::tmp_reg) = r9;"
+			 "r9 = *(u64 *)($ctx + bpf_sysctl_kern::ppos);"
+			 "*(u32 *)(r9 +0) = $src;"
+			 "r9 = *(u64 *)($ctx + bpf_sysctl_kern::tmp_reg);",
+#else
+		.read  = "$dst = *(u64 *)($ctx + bpf_sysctl_kern::ppos);"
+			 "$dst = *(u32 *)($dst +4);",
+		.write = "*(u64 *)($ctx + bpf_sysctl_kern::tmp_reg) = r9;"
+			 "r9 = *(u64 *)($ctx + bpf_sysctl_kern::ppos);"
+			 "*(u32 *)(r9 +4) = $src;"
+			 "r9 = *(u64 *)($ctx + bpf_sysctl_kern::tmp_reg);",
+#endif
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, sk),
+		.read  = "$dst = *(u64 *)($ctx + bpf_sockopt_kern::sk);",
+		.expected_attach_type = BPF_CGROUP_GETSOCKOPT,
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, level),
+		.read  = "$dst = *(u32 *)($ctx + bpf_sockopt_kern::level);",
+		.write = "*(u32 *)($ctx + bpf_sockopt_kern::level) = $src;",
+		.expected_attach_type = BPF_CGROUP_SETSOCKOPT,
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, optname),
+		.read  = "$dst = *(u32 *)($ctx + bpf_sockopt_kern::optname);",
+		.write = "*(u32 *)($ctx + bpf_sockopt_kern::optname) = $src;",
+		.expected_attach_type = BPF_CGROUP_SETSOCKOPT,
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, optlen),
+		.read  = "$dst = *(u32 *)($ctx + bpf_sockopt_kern::optlen);",
+		.write = "*(u32 *)($ctx + bpf_sockopt_kern::optlen) = $src;",
+		.expected_attach_type = BPF_CGROUP_SETSOCKOPT,
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, retval),
+		.read  = "$dst = *(u64 *)($ctx + bpf_sockopt_kern::current_task);"
+			 "$dst = *(u64 *)($dst + task_struct::bpf_ctx);"
+			 "$dst = *(u32 *)($dst + bpf_cg_run_ctx::retval);",
+		.write = "*(u64 *)($ctx + bpf_sockopt_kern::tmp_reg) = r9;"
+			 "r9 = *(u64 *)($ctx + bpf_sockopt_kern::current_task);"
+			 "r9 = *(u64 *)(r9 + task_struct::bpf_ctx);"
+			 "*(u32 *)(r9 + bpf_cg_run_ctx::retval) = $src;"
+			 "r9 = *(u64 *)($ctx + bpf_sockopt_kern::tmp_reg);",
+		.expected_attach_type = BPF_CGROUP_GETSOCKOPT,
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, optval),
+		.read  = "$dst = *(u64 *)($ctx + bpf_sockopt_kern::optval);",
+		.expected_attach_type = BPF_CGROUP_GETSOCKOPT,
+	},
+	{
+		N(CGROUP_SOCKOPT, struct bpf_sockopt, optval_end),
+		.read  = "$dst = *(u64 *)($ctx + bpf_sockopt_kern::optval_end);",
+		.expected_attach_type = BPF_CGROUP_GETSOCKOPT,
+	},
+};
+
+#undef N
+
+static regex_t *ident_regex;
+static regex_t *field_regex;
+
+static char *skip_space(char *str)
+{
+	while (*str && isspace(*str))
+		++str;
+	return str;
+}
+
+static char *skip_space_and_semi(char *str)
+{
+	while (*str && (isspace(*str) || *str == ';'))
+		++str;
+	return str;
+}
+
+static char *match_str(char *str, char *prefix)
+{
+	while (*str && *prefix && *str == *prefix) {
+		++str;
+		++prefix;
+	}
+	if (*prefix)
+		return NULL;
+	return str;
+}
+
+static char *match_number(char *str, int num)
+{
+	char *next;
+	int snum = strtol(str, &next, 10);
+
+	if (next - str == 0 || num != snum)
+		return NULL;
+
+	return next;
+}
+
+static int find_field_offset_aux(struct btf *btf, int btf_id, char *field_name, int off)
+{
+	const struct btf_type *type = btf__type_by_id(btf, btf_id);
+	const struct btf_member *m;
+	__u16 mnum;
+	int i;
+
+	if (!type) {
+		PRINT_FAIL("Can't find btf_type for id %d\n", btf_id);
+		return -1;
+	}
+
+	if (!btf_is_struct(type) && !btf_is_union(type)) {
+		PRINT_FAIL("BTF id %d is not struct or union\n", btf_id);
+		return -1;
+	}
+
+	m = btf_members(type);
+	mnum = btf_vlen(type);
+
+	for (i = 0; i < mnum; ++i, ++m) {
+		const char *mname = btf__name_by_offset(btf, m->name_off);
+
+		if (strcmp(mname, "") == 0) {
+			int msize = find_field_offset_aux(btf, m->type, field_name,
+							  off + m->offset);
+			if (msize >= 0)
+				return msize;
+		}
+
+		if (strcmp(mname, field_name))
+			continue;
+
+		return (off + m->offset) / 8;
+	}
+
+	return -1;
+}
+
+static int find_field_offset(struct btf *btf, char *pattern, regmatch_t *matches)
+{
+	int type_sz  = matches[1].rm_eo - matches[1].rm_so;
+	int field_sz = matches[2].rm_eo - matches[2].rm_so;
+	char *type   = pattern + matches[1].rm_so;
+	char *field  = pattern + matches[2].rm_so;
+	char field_str[128] = {};
+	char type_str[128] = {};
+	int btf_id, field_offset;
+
+	if (type_sz >= sizeof(type_str)) {
+		PRINT_FAIL("Malformed pattern: type ident is too long: %d\n", type_sz);
+		return -1;
+	}
+
+	if (field_sz >= sizeof(field_str)) {
+		PRINT_FAIL("Malformed pattern: field ident is too long: %d\n", field_sz);
+		return -1;
+	}
+
+	strncpy(type_str, type, type_sz);
+	strncpy(field_str, field, field_sz);
+	btf_id = btf__find_by_name(btf, type_str);
+	if (btf_id < 0) {
+		PRINT_FAIL("No BTF info for type %s\n", type_str);
+		return -1;
+	}
+
+	field_offset = find_field_offset_aux(btf, btf_id, field_str, 0);
+	if (field_offset < 0) {
+		PRINT_FAIL("No BTF info for field %s::%s\n", type_str, field_str);
+		return -1;
+	}
+
+	return field_offset;
+}
+
+static regex_t *compile_regex(char *pat)
+{
+	regex_t *re;
+	int err;
+
+	re = malloc(sizeof(regex_t));
+	if (!re) {
+		PRINT_FAIL("Can't alloc regex\n");
+		return NULL;
+	}
+
+	err = regcomp(re, pat, REG_EXTENDED);
+	if (err) {
+		char errbuf[512];
+
+		regerror(err, re, errbuf, sizeof(errbuf));
+		PRINT_FAIL("Can't compile regex: %s\n", errbuf);
+		free(re);
+		return NULL;
+	}
+
+	return re;
+}
+
+static void free_regex(regex_t *re)
+{
+	if (!re)
+		return;
+
+	regfree(re);
+	free(re);
+}
+
+static u32 max_line_len(char *str)
+{
+	u32 max_line = 0;
+	char *next = str;
+
+	while (next) {
+		next = strchr(str, '\n');
+		if (next) {
+			max_line = max_t(u32, max_line, (next - str));
+			str = next + 1;
+		} else {
+			max_line = max_t(u32, max_line, strlen(str));
+		}
+	}
+
+	return min(max_line, 60u);
+}
+
+/* Print strings `pattern_origin` and `text_origin` side by side,
+ * assume `pattern_pos` and `text_pos` designate location within
+ * corresponding origin string where match diverges.
+ * The output should look like:
+ *
+ *   Can't match disassembly(left) with pattern(right):
+ *   r2 = *(u64 *)(r1 +0)  ;  $dst = *(u64 *)($ctx + bpf_sockopt_kern::sk1)
+ *                     ^                             ^
+ *   r0 = 0                ;
+ *   exit                  ;
+ */
+static void print_match_error(FILE *out,
+			      char *pattern_origin, char *text_origin,
+			      char *pattern_pos, char *text_pos)
+{
+	char *pattern = pattern_origin;
+	char *text = text_origin;
+	int middle = max_line_len(text) + 2;
+
+	fprintf(out, "Can't match disassembly(left) with pattern(right):\n");
+	while (*pattern || *text) {
+		int column = 0;
+		int mark1 = -1;
+		int mark2 = -1;
+
+		/* Print one line from text */
+		while (*text && *text != '\n') {
+			if (text == text_pos)
+				mark1 = column;
+			fputc(*text, out);
+			++text;
+			++column;
+		}
+		if (text == text_pos)
+			mark1 = column;
+
+		/* Pad to the middle */
+		while (column < middle) {
+			fputc(' ', out);
+			++column;
+		}
+		fputs(";  ", out);
+		column += 3;
+
+		/* Print one line from pattern, pattern lines are terminated by ';' */
+		while (*pattern && *pattern != ';') {
+			if (pattern == pattern_pos)
+				mark2 = column;
+			fputc(*pattern, out);
+			++pattern;
+			++column;
+		}
+		if (pattern == pattern_pos)
+			mark2 = column;
+
+		fputc('\n', out);
+		if (*pattern)
+			++pattern;
+		if (*text)
+			++text;
+
+		/* If pattern and text diverge at this line, print an
+		 * additional line with '^' marks, highlighting
+		 * positions where match fails.
+		 */
+		if (mark1 > 0 || mark2 > 0) {
+			for (column = 0; column <= max(mark1, mark2); ++column) {
+				if (column == mark1 || column == mark2)
+					fputc('^', out);
+				else
+					fputc(' ', out);
+			}
+			fputc('\n', out);
+		}
+	}
+}
+
+/* Test if `text` matches `pattern`. Pattern consists of the following elements:
+ *
+ * - Field offset references:
+ *
+ *     <type>::<field>
+ *
+ *   When such reference is encountered BTF is used to compute numerical
+ *   value for the offset of <field> in <type>. The `text` is expected to
+ *   contain matching numerical value.
+ *
+ * - Field groups:
+ *
+ *     $(<type>::<field> [+ <type>::<field>]*)
+ *
+ *   Allows to specify an offset that is a sum of multiple field offsets.
+ *   The `text` is expected to contain matching numerical value.
+ *
+ * - Variable references, e.g. `$src`, `$dst`, `$ctx`.
+ *   These are substitutions specified in `reg_map` array.
+ *   If a substring of pattern is equal to `reg_map[i][0]` the `text` is
+ *   expected to contain `reg_map[i][1]` in the matching position.
+ *
+ * - Whitespace is ignored, ';' counts as whitespace for `pattern`.
+ *
+ * - Any other characters, `pattern` and `text` should match one-to-one.
+ *
+ * Example of a pattern:
+ *
+ *                    __________ fields group ________________
+ *                   '                                        '
+ *   *(u16 *)($ctx + $(sk_buff::cb + qdisc_skb_cb::tc_classid)) = $src;
+ *            ^^^^                   '______________________'
+ *     variable reference             field offset reference
+ */
+static bool match_pattern(struct btf *btf, char *pattern, char *text, char *reg_map[][2])
+{
+	char *pattern_origin = pattern;
+	char *text_origin = text;
+	regmatch_t matches[3];
+
+_continue:
+	while (*pattern) {
+		if (!*text)
+			goto err;
+
+		/* Skip whitespace */
+		if (isspace(*pattern) || *pattern == ';') {
+			if (!isspace(*text) && text != text_origin && isalnum(text[-1]))
+				goto err;
+			pattern = skip_space_and_semi(pattern);
+			text = skip_space(text);
+			continue;
+		}
+
+		/* Check for variable references */
+		for (int i = 0; reg_map[i][0]; ++i) {
+			char *pattern_next, *text_next;
+
+			pattern_next = match_str(pattern, reg_map[i][0]);
+			if (!pattern_next)
+				continue;
+
+			text_next = match_str(text, reg_map[i][1]);
+			if (!text_next)
+				goto err;
+
+			pattern = pattern_next;
+			text = text_next;
+			goto _continue;
+		}
+
+		/* Match field group:
+		 *   $(sk_buff::cb + qdisc_skb_cb::tc_classid)
+		 */
+		if (strncmp(pattern, "$(", 2) == 0) {
+			char *group_start = pattern, *text_next;
+			int acc_offset = 0;
+
+			pattern += 2;
+
+			for (;;) {
+				int field_offset;
+
+				pattern = skip_space(pattern);
+				if (!*pattern) {
+					PRINT_FAIL("Unexpected end of pattern\n");
+					goto err;
+				}
+
+				if (*pattern == ')') {
+					++pattern;
+					break;
+				}
+
+				if (*pattern == '+') {
+					++pattern;
+					continue;
+				}
+
+				printf("pattern: %s\n", pattern);
+				if (regexec(field_regex, pattern, 3, matches, 0) != 0) {
+					PRINT_FAIL("Field reference expected\n");
+					goto err;
+				}
+
+				field_offset = find_field_offset(btf, pattern, matches);
+				if (field_offset < 0)
+					goto err;
+
+				pattern += matches[0].rm_eo;
+				acc_offset += field_offset;
+			}
+
+			text_next = match_number(text, acc_offset);
+			if (!text_next) {
+				PRINT_FAIL("No match for group offset %.*s (%d)\n",
+					   (int)(pattern - group_start),
+					   group_start,
+					   acc_offset);
+				goto err;
+			}
+			text = text_next;
+		}
+
+		/* Match field reference:
+		 *   sk_buff::cb
+		 */
+		if (regexec(field_regex, pattern, 3, matches, 0) == 0) {
+			int field_offset;
+			char *text_next;
+
+			field_offset = find_field_offset(btf, pattern, matches);
+			if (field_offset < 0)
+				goto err;
+
+			text_next = match_number(text, field_offset);
+			if (!text_next) {
+				PRINT_FAIL("No match for field offset %.*s (%d)\n",
+					   (int)matches[0].rm_eo, pattern, field_offset);
+				goto err;
+			}
+
+			pattern += matches[0].rm_eo;
+			text = text_next;
+			continue;
+		}
+
+		/* If pattern points to identifier not followed by '::'
+		 * skip the identifier to avoid n^2 application of the
+		 * field reference rule.
+		 */
+		if (regexec(ident_regex, pattern, 1, matches, 0) == 0) {
+			if (strncmp(pattern, text, matches[0].rm_eo) != 0)
+				goto err;
+
+			pattern += matches[0].rm_eo;
+			text += matches[0].rm_eo;
+			continue;
+		}
+
+		/* Match literally */
+		if (*pattern != *text)
+			goto err;
+
+		++pattern;
+		++text;
+	}
+
+	return true;
+
+err:
+	test__fail();
+	print_match_error(stdout, pattern_origin, text_origin, pattern, text);
+	return false;
+}
+
+/* Request BPF program instructions after all rewrites are applied,
+ * e.g. verifier.c:convert_ctx_access() is done.
+ */
+static int get_xlated_program(int fd_prog, struct bpf_insn **buf, __u32 *cnt)
+{
+	struct bpf_prog_info info = {};
+	__u32 info_len = sizeof(info);
+	__u32 xlated_prog_len;
+	__u32 buf_element_size = sizeof(struct bpf_insn);
+
+	if (bpf_prog_get_info_by_fd(fd_prog, &info, &info_len)) {
+		perror("bpf_prog_get_info_by_fd failed");
+		return -1;
+	}
+
+	xlated_prog_len = info.xlated_prog_len;
+	if (xlated_prog_len % buf_element_size) {
+		printf("Program length %d is not multiple of %d\n",
+		       xlated_prog_len, buf_element_size);
+		return -1;
+	}
+
+	*cnt = xlated_prog_len / buf_element_size;
+	*buf = calloc(*cnt, buf_element_size);
+	if (!buf) {
+		perror("can't allocate xlated program buffer");
+		return -ENOMEM;
+	}
+
+	bzero(&info, sizeof(info));
+	info.xlated_prog_len = xlated_prog_len;
+	info.xlated_prog_insns = (__u64)(unsigned long)*buf;
+	if (bpf_prog_get_info_by_fd(fd_prog, &info, &info_len)) {
+		perror("second bpf_prog_get_info_by_fd failed");
+		goto out_free_buf;
+	}
+
+	return 0;
+
+out_free_buf:
+	free(*buf);
+	return -1;
+}
+
+static void print_insn(void *private_data, const char *fmt, ...)
+{
+	va_list args;
+
+	va_start(args, fmt);
+	vfprintf((FILE *)private_data, fmt, args);
+	va_end(args);
+}
+
+/* Disassemble instructions to a stream */
+static void print_xlated(FILE *out, struct bpf_insn *insn, __u32 len)
+{
+	const struct bpf_insn_cbs cbs = {
+		.cb_print	= print_insn,
+		.cb_call	= NULL,
+		.cb_imm		= NULL,
+		.private_data	= out,
+	};
+	bool double_insn = false;
+	int i;
+
+	for (i = 0; i < len; i++) {
+		if (double_insn) {
+			double_insn = false;
+			continue;
+		}
+
+		double_insn = insn[i].code == (BPF_LD | BPF_IMM | BPF_DW);
+		print_bpf_insn(&cbs, insn + i, true);
+	}
+}
+
+/* We share code with kernel BPF disassembler, it adds '(FF) ' prefix
+ * for each instruction (FF stands for instruction `code` byte).
+ * This function removes the prefix inplace for each line in `str`.
+ */
+static void remove_insn_prefix(char *str, int size)
+{
+	const int prefix_size = 5;
+
+	int write_pos = 0, read_pos = prefix_size;
+	int len = strlen(str);
+	char c;
+
+	size = min(size, len);
+
+	while (read_pos < size) {
+		c = str[read_pos++];
+		if (c == 0)
+			break;
+		str[write_pos++] = c;
+		if (c == '\n')
+			read_pos += prefix_size;
+	}
+	str[write_pos] = 0;
+}
+
+struct prog_info {
+	char *prog_kind;
+	enum bpf_prog_type prog_type;
+	enum bpf_attach_type expected_attach_type;
+	struct bpf_insn *prog;
+	u32 prog_len;
+};
+
+static void match_program(struct btf *btf,
+			  struct prog_info *pinfo,
+			  char *pattern,
+			  char *reg_map[][2],
+			  bool skip_first_insn)
+{
+	struct bpf_insn *buf = NULL;
+	int err = 0, prog_fd = 0;
+	FILE *prog_out = NULL;
+	char *text = NULL;
+	__u32 cnt = 0;
+
+	text = calloc(MAX_PROG_TEXT_SZ, 1);
+	if (!text) {
+		PRINT_FAIL("Can't allocate %d bytes\n", MAX_PROG_TEXT_SZ);
+		goto out;
+	}
+
+	// TODO: log level
+	LIBBPF_OPTS(bpf_prog_load_opts, opts);
+	opts.log_buf = text;
+	opts.log_size = MAX_PROG_TEXT_SZ;
+	opts.log_level = 1 | 2 | 4;
+	opts.expected_attach_type = pinfo->expected_attach_type;
+
+	prog_fd = bpf_prog_load(pinfo->prog_type, NULL, "GPL",
+				pinfo->prog, pinfo->prog_len, &opts);
+	if (prog_fd < 0) {
+		PRINT_FAIL("Can't load program, errno %d (%s), verifier log:\n%s\n",
+			   errno, strerror(errno), text);
+		goto out;
+	}
+
+	memset(text, 0, MAX_PROG_TEXT_SZ);
+
+	err = get_xlated_program(prog_fd, &buf, &cnt);
+	if (err) {
+		PRINT_FAIL("Can't load back BPF program\n");
+		goto out;
+	}
+
+	prog_out = fmemopen(text, MAX_PROG_TEXT_SZ - 1, "w");
+	if (!prog_out) {
+		PRINT_FAIL("Can't open memory stream\n");
+		goto out;
+	}
+	if (skip_first_insn)
+		print_xlated(prog_out, buf + 1, cnt - 1);
+	else
+		print_xlated(prog_out, buf, cnt);
+	fclose(prog_out);
+	remove_insn_prefix(text, MAX_PROG_TEXT_SZ);
+
+	ASSERT_TRUE(match_pattern(btf, pattern, text, reg_map),
+		    pinfo->prog_kind);
+
+out:
+	if (prog_fd)
+		close(prog_fd);
+	free(buf);
+	free(text);
+}
+
+static void run_one_testcase(struct btf *btf, struct test_case *test)
+{
+	struct prog_info pinfo = {};
+	int bpf_sz;
+
+	if (!test__start_subtest(test->name))
+		return;
+
+	switch (test->field_sz) {
+	case 8:
+		bpf_sz = BPF_DW;
+		break;
+	case 4:
+		bpf_sz = BPF_W;
+		break;
+	case 2:
+		bpf_sz = BPF_H;
+		break;
+	case 1:
+		bpf_sz = BPF_B;
+		break;
+	default:
+		PRINT_FAIL("Unexpected field size: %d, want 8,4,2 or 1\n", test->field_sz);
+		return;
+	}
+
+	pinfo.prog_type = test->prog_type;
+	pinfo.expected_attach_type = test->expected_attach_type;
+
+	if (test->read) {
+		struct bpf_insn ldx_prog[] = {
+			BPF_LDX_MEM(bpf_sz, BPF_REG_2, BPF_REG_1, test->field_offset),
+			BPF_MOV64_IMM(BPF_REG_0, 0),
+			BPF_EXIT_INSN(),
+		};
+		char *reg_map[][2] = {
+			{ "$ctx", "r1" },
+			{ "$dst", "r2" },
+			{}
+		};
+
+		pinfo.prog_kind = "LDX";
+		pinfo.prog = ldx_prog;
+		pinfo.prog_len = ARRAY_SIZE(ldx_prog);
+		match_program(btf, &pinfo, test->read, reg_map, false);
+	}
+
+	if (test->write || test->write_st || test->write_stx) {
+		struct bpf_insn stx_prog[] = {
+			BPF_MOV64_IMM(BPF_REG_2, 0),
+			BPF_STX_MEM(bpf_sz, BPF_REG_1, BPF_REG_2, test->field_offset),
+			BPF_MOV64_IMM(BPF_REG_0, 0),
+			BPF_EXIT_INSN(),
+		};
+		char *stx_reg_map[][2] = {
+			{ "$ctx", "r1" },
+			{ "$src", "r2" },
+			{}
+		};
+		struct bpf_insn st_prog[] = {
+			BPF_ST_MEM(bpf_sz, BPF_REG_1, test->field_offset,
+				   test->st_value.use ? test->st_value.value : 42),
+			BPF_MOV64_IMM(BPF_REG_0, 0),
+			BPF_EXIT_INSN(),
+		};
+		char *st_reg_map[][2] = {
+			{ "$ctx", "r1" },
+			{ "$src", "42" },
+			{}
+		};
+
+		if (test->write || test->write_stx) {
+			char *pattern = test->write_stx ? test->write_stx : test->write;
+
+			pinfo.prog_kind = "STX";
+			pinfo.prog = stx_prog;
+			pinfo.prog_len = ARRAY_SIZE(stx_prog);
+			match_program(btf, &pinfo, pattern, stx_reg_map, true);
+		}
+
+		if (test->write || test->write_st) {
+			char *pattern = test->write_st ? test->write_st : test->write;
+
+			pinfo.prog_kind = "ST";
+			pinfo.prog = st_prog;
+			pinfo.prog_len = ARRAY_SIZE(st_prog);
+			match_program(btf, &pinfo, pattern, st_reg_map, false);
+		}
+	}
+
+	test__end_subtest();
+}
+
+void test_ctx_rewrite(void)
+{
+	struct btf *btf;
+	int i;
+
+	field_regex = compile_regex("^([[:alpha:]_][[:alnum:]_]+)::([[:alpha:]_][[:alnum:]_]+)");
+	ident_regex = compile_regex("^[[:alpha:]_][[:alnum:]_]+");
+	if (!field_regex || !ident_regex)
+		return;
+
+	btf = btf__load_vmlinux_btf();
+	if (!btf) {
+		PRINT_FAIL("Can't load vmlinux BTF, errno %d (%s)\n", errno, strerror(errno));
+		goto out;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); ++i)
+		run_one_testcase(btf, &test_cases[i]);
+
+out:
+	btf__free(btf);
+	free_regex(field_regex);
+	free_regex(ident_regex);
+}
-- 
cgit v1.2.3-70-g09d2


From d54e0f6c1adffbf72f2cf4aebe6122899c3b851c Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:49:59 -0800
Subject: bpf: improve stack slot state printing

Improve stack slot state printing to provide more useful and relevant
information, especially for dynptrs. While previously we'd see something
like:

  8: (85) call bpf_ringbuf_reserve_dynptr#198   ; R0_w=scalar() fp-8_w=dddddddd fp-16_w=dddddddd refs=2

Now we'll see way more useful:

  8: (85) call bpf_ringbuf_reserve_dynptr#198   ; R0_w=scalar() fp-16_w=dynptr_ringbuf(ref_id=2) refs=2

I experimented with printing the range of slots taken by dynptr,
something like:

  fp-16..8_w=dynptr_ringbuf(ref_id=2)

But it felt very awkward and pretty useless. So we print the lowest
address (most negative offset) only.

The general structure of this code is now also set up for easier
extension and will accommodate ITER slots naturally.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 75 +++++++++++++++++++++++++++++++++------------------
 1 file changed, 49 insertions(+), 26 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 4c5d2b5e25c8..4f71b6b61ef4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -705,6 +705,25 @@ static const char *kernel_type_name(const struct btf* btf, u32 id)
 	return btf_name_by_offset(btf, btf_type_by_id(btf, id)->name_off);
 }
 
+static const char *dynptr_type_str(enum bpf_dynptr_type type)
+{
+	switch (type) {
+	case BPF_DYNPTR_TYPE_LOCAL:
+		return "local";
+	case BPF_DYNPTR_TYPE_RINGBUF:
+		return "ringbuf";
+	case BPF_DYNPTR_TYPE_SKB:
+		return "skb";
+	case BPF_DYNPTR_TYPE_XDP:
+		return "xdp";
+	case BPF_DYNPTR_TYPE_INVALID:
+		return "<invalid>";
+	default:
+		WARN_ONCE(1, "unknown dynptr type %d\n", type);
+		return "<unknown>";
+	}
+}
+
 static void mark_reg_scratched(struct bpf_verifier_env *env, u32 regno)
 {
 	env->scratched_regs |= 1U << regno;
@@ -1176,26 +1195,49 @@ static void print_verifier_state(struct bpf_verifier_env *env,
 		for (j = 0; j < BPF_REG_SIZE; j++) {
 			if (state->stack[i].slot_type[j] != STACK_INVALID)
 				valid = true;
-			types_buf[j] = slot_type_char[
-					state->stack[i].slot_type[j]];
+			types_buf[j] = slot_type_char[state->stack[i].slot_type[j]];
 		}
 		types_buf[BPF_REG_SIZE] = 0;
 		if (!valid)
 			continue;
 		if (!print_all && !stack_slot_scratched(env, i))
 			continue;
-		verbose(env, " fp%d", (-i - 1) * BPF_REG_SIZE);
-		print_liveness(env, state->stack[i].spilled_ptr.live);
-		if (is_spilled_reg(&state->stack[i])) {
+		switch (state->stack[i].slot_type[BPF_REG_SIZE - 1]) {
+		case STACK_SPILL:
 			reg = &state->stack[i].spilled_ptr;
 			t = reg->type;
+
+			verbose(env, " fp%d", (-i - 1) * BPF_REG_SIZE);
+			print_liveness(env, reg->live);
 			verbose(env, "=%s", t == SCALAR_VALUE ? "" : reg_type_str(env, t));
 			if (t == SCALAR_VALUE && reg->precise)
 				verbose(env, "P");
 			if (t == SCALAR_VALUE && tnum_is_const(reg->var_off))
 				verbose(env, "%lld", reg->var_off.value + reg->off);
-		} else {
+			break;
+		case STACK_DYNPTR:
+			i += BPF_DYNPTR_NR_SLOTS - 1;
+			reg = &state->stack[i].spilled_ptr;
+
+			verbose(env, " fp%d", (-i - 1) * BPF_REG_SIZE);
+			print_liveness(env, reg->live);
+			verbose(env, "=dynptr_%s", dynptr_type_str(reg->dynptr.type));
+			if (reg->ref_obj_id)
+				verbose(env, "(ref_id=%d)", reg->ref_obj_id);
+			break;
+		case STACK_MISC:
+		case STACK_ZERO:
+		default:
+			reg = &state->stack[i].spilled_ptr;
+
+			for (j = 0; j < BPF_REG_SIZE; j++)
+				types_buf[j] = slot_type_char[state->stack[i].slot_type[j]];
+			types_buf[BPF_REG_SIZE] = 0;
+
+			verbose(env, " fp%d", (-i - 1) * BPF_REG_SIZE);
+			print_liveness(env, reg->live);
 			verbose(env, "=%s", types_buf);
+			break;
 		}
 	}
 	if (state->acquired_refs && state->refs[0].id) {
@@ -6411,28 +6453,9 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno, int insn
 
 		/* Fold modifiers (in this case, MEM_RDONLY) when checking expected type */
 		if (!is_dynptr_type_expected(env, reg, arg_type & ~MEM_RDONLY)) {
-			const char *err_extra = "";
-
-			switch (arg_type & DYNPTR_TYPE_FLAG_MASK) {
-			case DYNPTR_TYPE_LOCAL:
-				err_extra = "local";
-				break;
-			case DYNPTR_TYPE_RINGBUF:
-				err_extra = "ringbuf";
-				break;
-			case DYNPTR_TYPE_SKB:
-				err_extra = "skb ";
-				break;
-			case DYNPTR_TYPE_XDP:
-				err_extra = "xdp ";
-				break;
-			default:
-				err_extra = "<unknown>";
-				break;
-			}
 			verbose(env,
 				"Expected a dynptr of type %s as arg #%d\n",
-				err_extra, regno);
+				dynptr_type_str(arg_to_dynptr_type(arg_type)), regno);
 			return -EINVAL;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 567da5d253cd6b41c6d015adac1af653725bef9d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:00 -0800
Subject: bpf: improve regsafe() checks for PTR_TO_{MEM,BUF,TP_BUFFER}

Teach regsafe() logic to handle PTR_TO_MEM, PTR_TO_BUF, and
PTR_TO_TP_BUFFER similarly to PTR_TO_MAP_{KEY,VALUE}. That is, instead of
exact match for var_off and range, use tnum_in() and range_within()
checks, allowing more general verified state to subsume more specific
current state. This allows to match wider range of valid and safe
states, speeding up verification and detecting wider range of equivalent
states for upcoming open-coded iteration looping logic.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 4f71b6b61ef4..b071b922848b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -14209,13 +14209,17 @@ static bool regsafe(struct bpf_verifier_env *env, struct bpf_reg_state *rold,
 		       tnum_in(rold->var_off, rcur->var_off);
 	case PTR_TO_MAP_KEY:
 	case PTR_TO_MAP_VALUE:
+	case PTR_TO_MEM:
+	case PTR_TO_BUF:
+	case PTR_TO_TP_BUFFER:
 		/* If the new min/max/var_off satisfy the old ones and
 		 * everything else matches, we are OK.
 		 */
 		return memcmp(rold, rcur, offsetof(struct bpf_reg_state, var_off)) == 0 &&
 		       range_within(rold, rcur) &&
 		       tnum_in(rold->var_off, rcur->var_off) &&
-		       check_ids(rold->id, rcur->id, idmap);
+		       check_ids(rold->id, rcur->id, idmap) &&
+		       check_ids(rold->ref_obj_id, rcur->ref_obj_id, idmap);
 	case PTR_TO_PACKET_META:
 	case PTR_TO_PACKET:
 		/* We must have at least as much range as the old ptr
-- 
cgit v1.2.3-70-g09d2


From 6f876e75d316a75957f3d43c3a8c2a6fe9bc18b2 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:01 -0800
Subject: selftests/bpf: enhance align selftest's expected log matching

Allow to search for expected register state in all the verifier log
output that's related to specified instruction number.

See added comment for an example of possible situation that is happening
due to a simple enhancement done in the next patch, which fixes handling
of env->test_state_freq flag in state checkpointing logic.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/align.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/align.c b/tools/testing/selftests/bpf/prog_tests/align.c
index 4666f88f2bb4..c94fa8d6c4f6 100644
--- a/tools/testing/selftests/bpf/prog_tests/align.c
+++ b/tools/testing/selftests/bpf/prog_tests/align.c
@@ -660,16 +660,22 @@ static int do_test_single(struct bpf_align_test *test)
 			 * func#0 @0
 			 * 0: R1=ctx(off=0,imm=0) R10=fp0
 			 * 0: (b7) r3 = 2                 ; R3_w=2
+			 *
+			 * Sometimes it's actually two lines below, e.g. when
+			 * searching for "6: R3_w=scalar(umax=255,var_off=(0x0; 0xff))":
+			 *   from 4 to 6: R0_w=pkt(off=8,r=8,imm=0) R1=ctx(off=0,imm=0) R2_w=pkt(off=0,r=8,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
+			 *   6: R0_w=pkt(off=8,r=8,imm=0) R1=ctx(off=0,imm=0) R2_w=pkt(off=0,r=8,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
+			 *   6: (71) r3 = *(u8 *)(r2 +0)           ; R2_w=pkt(off=0,r=8,imm=0) R3_w=scalar(umax=255,var_off=(0x0; 0xff))
 			 */
-			if (!strstr(line_ptr, m.match)) {
+			while (!strstr(line_ptr, m.match)) {
 				cur_line = -1;
 				line_ptr = strtok(NULL, "\n");
-				sscanf(line_ptr, "%u: ", &cur_line);
+				sscanf(line_ptr ?: "", "%u: ", &cur_line);
+				if (!line_ptr || cur_line != m.line)
+					break;
 			}
-			if (cur_line != m.line || !line_ptr ||
-			    !strstr(line_ptr, m.match)) {
-				printf("Failed to find match %u: %s\n",
-				       m.line, m.match);
+			if (cur_line != m.line || !line_ptr || !strstr(line_ptr, m.match)) {
+				printf("Failed to find match %u: %s\n", m.line, m.match);
 				ret = 1;
 				printf("%s", bpf_vlog);
 				break;
-- 
cgit v1.2.3-70-g09d2


From 98ddcf389d1bb7a407d49c23dfe6443680812f24 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:02 -0800
Subject: bpf: honor env->test_state_freq flag in is_state_visited()

env->test_state_freq flag can be set by user by passing
BPF_F_TEST_STATE_FREQ program flag. This is used in a bunch of selftests
to have predictable state checkpoints at every jump and so on.

Currently, bounded loop handling heuristic ignores this flag if number
of processed jumps and/or number of processed instructions is below some
thresholds, which throws off that reliable state checkpointing.

Honor this flag in all circumstances by disabling heuristic if
env->test_state_freq is set.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index b071b922848b..fa93ba10762d 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -14651,7 +14651,8 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 			 * This threshold shouldn't be too high either, since states
 			 * at the end of the loop are likely to be useful in pruning.
 			 */
-			if (env->jmps_processed - env->prev_jmps_processed < 20 &&
+			if (!env->test_state_freq &&
+			    env->jmps_processed - env->prev_jmps_processed < 20 &&
 			    env->insn_processed - env->prev_insn_processed < 100)
 				add_new_state = false;
 			goto miss;
-- 
cgit v1.2.3-70-g09d2


From fffc893b6bf29162aca76842238868b131fcb477 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:03 -0800
Subject: selftests/bpf: adjust log_fixup's buffer size for proper truncation

Adjust log_fixup's expected buffer length to fix the test. It's pretty
finicky in its length expectation, but it doesn't break often. So just
adjust the length to work on current kernel and with follow up iterator
changes as well.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-6-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/log_fixup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/log_fixup.c b/tools/testing/selftests/bpf/prog_tests/log_fixup.c
index f4ffdcabf4e4..239e1c5753b0 100644
--- a/tools/testing/selftests/bpf/prog_tests/log_fixup.c
+++ b/tools/testing/selftests/bpf/prog_tests/log_fixup.c
@@ -141,7 +141,7 @@ void test_log_fixup(void)
 	if (test__start_subtest("bad_core_relo_trunc_partial"))
 		bad_core_relo(300, TRUNC_PARTIAL /* truncate original log a bit */);
 	if (test__start_subtest("bad_core_relo_trunc_full"))
-		bad_core_relo(250, TRUNC_FULL  /* truncate also libbpf's message patch */);
+		bad_core_relo(210, TRUNC_FULL  /* truncate also libbpf's message patch */);
 	if (test__start_subtest("bad_core_relo_subprog"))
 		bad_core_relo_subprog();
 	if (test__start_subtest("missing_map"))
-- 
cgit v1.2.3-70-g09d2


From 653ae3a874aca6764a4c1f5a8bf1b072ade0d6f4 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:04 -0800
Subject: bpf: clean up visit_insn()'s instruction processing

Instead of referencing processed instruction repeatedly as insns[t]
throughout entire visit_insn() function, take a local insn pointer and
work with it in a cleaner way.

It makes enhancing this function further a bit easier as well.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-7-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 25 ++++++++++++-------------
 1 file changed, 12 insertions(+), 13 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index fa93ba10762d..6188d5604ed4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -13484,44 +13484,43 @@ static int visit_func_call_insn(int t, struct bpf_insn *insns,
  */
 static int visit_insn(int t, struct bpf_verifier_env *env)
 {
-	struct bpf_insn *insns = env->prog->insnsi;
+	struct bpf_insn *insns = env->prog->insnsi, *insn = &insns[t];
 	int ret;
 
-	if (bpf_pseudo_func(insns + t))
+	if (bpf_pseudo_func(insn))
 		return visit_func_call_insn(t, insns, env, true);
 
 	/* All non-branch instructions have a single fall-through edge. */
-	if (BPF_CLASS(insns[t].code) != BPF_JMP &&
-	    BPF_CLASS(insns[t].code) != BPF_JMP32)
+	if (BPF_CLASS(insn->code) != BPF_JMP &&
+	    BPF_CLASS(insn->code) != BPF_JMP32)
 		return push_insn(t, t + 1, FALLTHROUGH, env, false);
 
-	switch (BPF_OP(insns[t].code)) {
+	switch (BPF_OP(insn->code)) {
 	case BPF_EXIT:
 		return DONE_EXPLORING;
 
 	case BPF_CALL:
-		if (insns[t].imm == BPF_FUNC_timer_set_callback)
+		if (insn->imm == BPF_FUNC_timer_set_callback)
 			/* Mark this call insn as a prune point to trigger
 			 * is_state_visited() check before call itself is
 			 * processed by __check_func_call(). Otherwise new
 			 * async state will be pushed for further exploration.
 			 */
 			mark_prune_point(env, t);
-		return visit_func_call_insn(t, insns, env,
-					    insns[t].src_reg == BPF_PSEUDO_CALL);
+		return visit_func_call_insn(t, insns, env, insn->src_reg == BPF_PSEUDO_CALL);
 
 	case BPF_JA:
-		if (BPF_SRC(insns[t].code) != BPF_K)
+		if (BPF_SRC(insn->code) != BPF_K)
 			return -EINVAL;
 
 		/* unconditional jump with single edge */
-		ret = push_insn(t, t + insns[t].off + 1, FALLTHROUGH, env,
+		ret = push_insn(t, t + insn->off + 1, FALLTHROUGH, env,
 				true);
 		if (ret)
 			return ret;
 
-		mark_prune_point(env, t + insns[t].off + 1);
-		mark_jmp_point(env, t + insns[t].off + 1);
+		mark_prune_point(env, t + insn->off + 1);
+		mark_jmp_point(env, t + insn->off + 1);
 
 		return ret;
 
@@ -13533,7 +13532,7 @@ static int visit_insn(int t, struct bpf_verifier_env *env)
 		if (ret)
 			return ret;
 
-		return push_insn(t, t + insns[t].off + 1, BRANCH, env, true);
+		return push_insn(t, t + insn->off + 1, BRANCH, env, true);
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From c1ee85a9806a720aa054f68fe7f9c79418f36c2b Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:05 -0800
Subject: bpf: fix visit_insn()'s detection of BPF_FUNC_timer_set_callback
 helper

It's not correct to assume that any BPF_CALL instruction is a helper
call. Fix visit_insn()'s detection of bpf_timer_set_callback() helper by
also checking insn->code == 0. For kfuncs insn->code would be set to
BPF_PSEUDO_KFUNC_CALL, and for subprog calls it will be BPF_PSEUDO_CALL.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-8-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 6188d5604ed4..34fd808e9692 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -13500,7 +13500,7 @@ static int visit_insn(int t, struct bpf_verifier_env *env)
 		return DONE_EXPLORING;
 
 	case BPF_CALL:
-		if (insn->imm == BPF_FUNC_timer_set_callback)
+		if (insn->src_reg == 0 && insn->imm == BPF_FUNC_timer_set_callback)
 			/* Mark this call insn as a prune point to trigger
 			 * is_state_visited() check before call itself is
 			 * processed by __check_func_call(). Otherwise new
-- 
cgit v1.2.3-70-g09d2


From 553a64a85c5d1dac277325a0f51a31c056593048 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:06 -0800
Subject: bpf: ensure that r0 is marked scratched after any function call

r0 is important (unless called function is void-returning, but that's
taken care of by print_verifier_state() anyways) in verifier logs.
Currently for helpers we seem to print it in verifier log, but for
kfuncs we don't.

Instead of figuring out where in the maze of code we accidentally set r0
as scratched for helpers and why we don't do that for kfuncs, just
enforce that after any function call r0 is marked as scratched.

Also, perhaps, we should reconsider "scratched" terminology, as it's
mightily confusing. "Touched" would seem more appropriate. But I left
that for follow ups for now.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-9-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 34fd808e9692..db0c37e9bc3a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -15110,6 +15110,8 @@ static int do_check(struct bpf_verifier_env *env)
 					err = check_helper_call(env, insn, &env->insn_idx);
 				if (err)
 					return err;
+
+				mark_reg_scratched(env, BPF_REG_0);
 			} else if (opcode == BPF_JA) {
 				if (BPF_SRC(insn->code) != BPF_K ||
 				    insn->imm != 0 ||
-- 
cgit v1.2.3-70-g09d2


From d0e1ac227945c6af616c003365c6feb986dc0839 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:07 -0800
Subject: bpf: move kfunc_call_arg_meta higher in the file

Move struct bpf_kfunc_call_arg_meta higher in the file and put it next
to struct bpf_call_arg_meta, so it can be used from more functions.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-10-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 70 +++++++++++++++++++++++++--------------------------
 1 file changed, 35 insertions(+), 35 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index db0c37e9bc3a..ed9a96ad7b5b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -270,6 +270,41 @@ struct bpf_call_arg_meta {
 	struct btf_field *kptr_field;
 };
 
+struct bpf_kfunc_call_arg_meta {
+	/* In parameters */
+	struct btf *btf;
+	u32 func_id;
+	u32 kfunc_flags;
+	const struct btf_type *func_proto;
+	const char *func_name;
+	/* Out parameters */
+	u32 ref_obj_id;
+	u8 release_regno;
+	bool r0_rdonly;
+	u32 ret_btf_id;
+	u64 r0_size;
+	u32 subprogno;
+	struct {
+		u64 value;
+		bool found;
+	} arg_constant;
+	struct {
+		struct btf *btf;
+		u32 btf_id;
+	} arg_obj_drop;
+	struct {
+		struct btf_field *field;
+	} arg_list_head;
+	struct {
+		struct btf_field *field;
+	} arg_rbtree_root;
+	struct {
+		enum bpf_dynptr_type type;
+		u32 id;
+	} initialized_dynptr;
+	u64 mem_size;
+};
+
 struct btf *btf_vmlinux;
 
 static DEFINE_MUTEX(bpf_verifier_lock);
@@ -8811,41 +8846,6 @@ static void mark_btf_func_reg_size(struct bpf_verifier_env *env, u32 regno,
 	}
 }
 
-struct bpf_kfunc_call_arg_meta {
-	/* In parameters */
-	struct btf *btf;
-	u32 func_id;
-	u32 kfunc_flags;
-	const struct btf_type *func_proto;
-	const char *func_name;
-	/* Out parameters */
-	u32 ref_obj_id;
-	u8 release_regno;
-	bool r0_rdonly;
-	u32 ret_btf_id;
-	u64 r0_size;
-	u32 subprogno;
-	struct {
-		u64 value;
-		bool found;
-	} arg_constant;
-	struct {
-		struct btf *btf;
-		u32 btf_id;
-	} arg_obj_drop;
-	struct {
-		struct btf_field *field;
-	} arg_list_head;
-	struct {
-		struct btf_field *field;
-	} arg_rbtree_root;
-	struct {
-		enum bpf_dynptr_type type;
-		u32 id;
-	} initialized_dynptr;
-	u64 mem_size;
-};
-
 static bool is_kfunc_acquire(struct bpf_kfunc_call_arg_meta *meta)
 {
 	return meta->kfunc_flags & KF_ACQUIRE;
-- 
cgit v1.2.3-70-g09d2


From d5271c5b1950b887def1663b75e2d710cc16535f Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:08 -0800
Subject: bpf: mark PTR_TO_MEM as non-null register type

PTR_TO_MEM register without PTR_MAYBE_NULL is indeed non-null. This is
important for BPF verifier to be able to prune guaranteed not to be
taken branches. This is always the case with open-coded iterators.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-11-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index ed9a96ad7b5b..d95975cbcc19 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -487,7 +487,8 @@ static bool reg_type_not_null(enum bpf_reg_type type)
 		type == PTR_TO_TCP_SOCK ||
 		type == PTR_TO_MAP_VALUE ||
 		type == PTR_TO_MAP_KEY ||
-		type == PTR_TO_SOCK_COMMON;
+		type == PTR_TO_SOCK_COMMON ||
+		type == PTR_TO_MEM;
 }
 
 static bool type_is_ptr_alloc_obj(u32 type)
-- 
cgit v1.2.3-70-g09d2


From a461f5adf17756e99ee0903d1a40961b0342ebb3 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:09 -0800
Subject: bpf: generalize dynptr_get_spi to be usable for iters

Generalize the logic of fetching special stack slot object state using
spi (stack slot index). This will be used by STACK_ITER logic next.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-12-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d95975cbcc19..c4151c9efe24 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -710,32 +710,38 @@ static bool is_spi_bounds_valid(struct bpf_func_state *state, int spi, int nr_sl
        return spi - nr_slots + 1 >= 0 && spi < allocated_slots;
 }
 
-static int dynptr_get_spi(struct bpf_verifier_env *env, struct bpf_reg_state *reg)
+static int stack_slot_obj_get_spi(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+			          const char *obj_kind, int nr_slots)
 {
 	int off, spi;
 
 	if (!tnum_is_const(reg->var_off)) {
-		verbose(env, "dynptr has to be at a constant offset\n");
+		verbose(env, "%s has to be at a constant offset\n", obj_kind);
 		return -EINVAL;
 	}
 
 	off = reg->off + reg->var_off.value;
 	if (off % BPF_REG_SIZE) {
-		verbose(env, "cannot pass in dynptr at an offset=%d\n", off);
+		verbose(env, "cannot pass in %s at an offset=%d\n", obj_kind, off);
 		return -EINVAL;
 	}
 
 	spi = __get_spi(off);
-	if (spi < 1) {
-		verbose(env, "cannot pass in dynptr at an offset=%d\n", off);
+	if (spi + 1 < nr_slots) {
+		verbose(env, "cannot pass in %s at an offset=%d\n", obj_kind, off);
 		return -EINVAL;
 	}
 
-	if (!is_spi_bounds_valid(func(env, reg), spi, BPF_DYNPTR_NR_SLOTS))
+	if (!is_spi_bounds_valid(func(env, reg), spi, nr_slots))
 		return -ERANGE;
 	return spi;
 }
 
+static int dynptr_get_spi(struct bpf_verifier_env *env, struct bpf_reg_state *reg)
+{
+	return stack_slot_obj_get_spi(env, reg, "dynptr", BPF_DYNPTR_NR_SLOTS);
+}
+
 static const char *kernel_type_name(const struct btf* btf, u32 id)
 {
 	return btf_name_by_offset(btf, btf_type_by_id(btf, id)->name_off);
-- 
cgit v1.2.3-70-g09d2


From f4b4eee6169bb33c5157ebe07e53d7e4be7631c0 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 2 Mar 2023 15:50:10 -0800
Subject: bpf: add support for fixed-size memory pointer returns for kfuncs

Support direct fixed-size (and for now, read-only) memory access when
kfunc's return type is a pointer to non-struct type. Calculate type size
and let BPF program access that many bytes directly. This is crucial for
numbers iterator.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230302235015.2044271-13-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index c4151c9efe24..b2116ca78d9a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -10336,6 +10336,14 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 				return -EFAULT;
 			}
 		} else if (!__btf_type_is_struct(ptr_type)) {
+			if (!meta.r0_size) {
+				__u32 sz;
+
+				if (!IS_ERR(btf_resolve_size(desc_btf, ptr_type, &sz))) {
+					meta.r0_size = sz;
+					meta.r0_rdonly = true;
+				}
+			}
 			if (!meta.r0_size) {
 				ptr_type_name = btf_name_by_offset(desc_btf,
 								   ptr_type->name_off);
-- 
cgit v1.2.3-70-g09d2


From 5c48f9432d06bf85ca934d7c4ecb14a7ec0c7f5d Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Fri, 24 Feb 2023 16:21:17 +0800
Subject: wifi: rtw89: fix SER L1 might stop entering LPS issue

When SER L1 triggered, driver need to stop Rx and clear RTW89_FLAG_RUNNING
flag. If track_work check RTW89_FLAG_RUNNING simultaneously, it will check
failed and never queue track_work again, and LPS won't enter either.
Therefore, we cancel delayed work when enter SER L1, and queue delayed work
when leave SER L1.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230224082117.21241-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/ser.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/ser.c b/drivers/net/wireless/realtek/rtw89/ser.c
index 61db7189fdab..9e9f6947e7f1 100644
--- a/drivers/net/wireless/realtek/rtw89/ser.c
+++ b/drivers/net/wireless/realtek/rtw89/ser.c
@@ -414,8 +414,11 @@ static void ser_idle_st_hdl(struct rtw89_ser *ser, u8 evt)
 
 static void ser_reset_trx_st_hdl(struct rtw89_ser *ser, u8 evt)
 {
+	struct rtw89_dev *rtwdev = container_of(ser, struct rtw89_dev, ser);
+
 	switch (evt) {
 	case SER_EV_STATE_IN:
+		cancel_delayed_work_sync(&rtwdev->track_work);
 		drv_stop_tx(ser);
 
 		if (hal_stop_dma(ser)) {
@@ -446,6 +449,8 @@ static void ser_reset_trx_st_hdl(struct rtw89_ser *ser, u8 evt)
 		hal_enable_dma(ser);
 		drv_resume_rx(ser);
 		drv_resume_tx(ser);
+		ieee80211_queue_delayed_work(rtwdev->hw, &rtwdev->track_work,
+					     RTW89_TRACK_WORK_PERIOD);
 		break;
 
 	default:
-- 
cgit v1.2.3-70-g09d2


From e5c3da9abd44bee298070d4e66b502067a8f1715 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 24 Feb 2023 13:59:17 +0100
Subject: wifi: brcmfmac: pcie: Add 4359C0 firmware definition

Some phones from around 2016, as well as other random devices have
this chip called 43956 or 4359C0 or 43596A0, which is more or less
just a rev bump (v9) of the already-supported 4359. Add a corresponding
firmware definition to allow for choosing the correct blob.

Suggested-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230224-topic-brcm_tone-v1-1-333b0ac67934@linaro.org
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 848f05ec1268..59f3e9c5e139 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -58,6 +58,7 @@ BRCMF_FW_CLM_DEF(4356, "brcmfmac4356-pcie");
 BRCMF_FW_CLM_DEF(43570, "brcmfmac43570-pcie");
 BRCMF_FW_DEF(4358, "brcmfmac4358-pcie");
 BRCMF_FW_DEF(4359, "brcmfmac4359-pcie");
+BRCMF_FW_DEF(4359C, "brcmfmac4359c-pcie");
 BRCMF_FW_CLM_DEF(4364B2, "brcmfmac4364b2-pcie");
 BRCMF_FW_CLM_DEF(4364B3, "brcmfmac4364b3-pcie");
 BRCMF_FW_DEF(4365B, "brcmfmac4365b-pcie");
@@ -92,7 +93,8 @@ static const struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 	BRCMF_FW_ENTRY(BRCM_CC_43569_CHIP_ID, 0xFFFFFFFF, 43570),
 	BRCMF_FW_ENTRY(BRCM_CC_43570_CHIP_ID, 0xFFFFFFFF, 43570),
 	BRCMF_FW_ENTRY(BRCM_CC_4358_CHIP_ID, 0xFFFFFFFF, 4358),
-	BRCMF_FW_ENTRY(BRCM_CC_4359_CHIP_ID, 0xFFFFFFFF, 4359),
+	BRCMF_FW_ENTRY(BRCM_CC_4359_CHIP_ID, 0x000001FF, 4359),
+	BRCMF_FW_ENTRY(BRCM_CC_4359_CHIP_ID, 0xFFFFFE00, 4359C),
 	BRCMF_FW_ENTRY(BRCM_CC_4364_CHIP_ID, 0x0000000F, 4364B2), /* 3 */
 	BRCMF_FW_ENTRY(BRCM_CC_4364_CHIP_ID, 0xFFFFFFF0, 4364B3), /* 4 */
 	BRCMF_FW_ENTRY(BRCM_CC_4365_CHIP_ID, 0x0000000F, 4365B),
-- 
cgit v1.2.3-70-g09d2


From b7ed9fa2cb76ca7a3c3cd4a6d35748fe1fbda9f6 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sun, 26 Feb 2023 23:10:03 +0100
Subject: wifi: rtw88: mac: Return the original error from rtw_pwr_seq_parser()

rtw_pwr_seq_parser() calls rtw_sub_pwr_seq_parser() which can either
return -EBUSY, -EINVAL or 0. Propagate the original error code instead
of unconditionally returning -EBUSY in case of an error.

Fixes: e3037485c68e ("rtw88: new Realtek 802.11ac driver")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230226221004.138331-2-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 1c9530a0eb69..4749d75fefee 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -236,7 +236,7 @@ static int rtw_pwr_seq_parser(struct rtw_dev *rtwdev,
 
 		ret = rtw_sub_pwr_seq_parser(rtwdev, intf_mask, cut_mask, cmd);
 		if (ret)
-			return -EBUSY;
+			return ret;
 
 		idx++;
 	} while (1);
-- 
cgit v1.2.3-70-g09d2


From 15c8e267dfa62f207ee1db666c822324e3362b84 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sun, 26 Feb 2023 23:10:04 +0100
Subject: wifi: rtw88: mac: Return the original error from
 rtw_mac_power_switch()

rtw_mac_power_switch() calls rtw_pwr_seq_parser() which can return
-EINVAL, -EBUSY or 0. Propagate the original error code instead of
unconditionally returning -EINVAL in case of an error.

Fixes: e3037485c68e ("rtw88: new Realtek 802.11ac driver")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230226221004.138331-3-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 4749d75fefee..f3a566cf979b 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -250,6 +250,7 @@ static int rtw_mac_power_switch(struct rtw_dev *rtwdev, bool pwr_on)
 	const struct rtw_pwr_seq_cmd **pwr_seq;
 	u8 rpwm;
 	bool cur_pwr;
+	int ret;
 
 	if (rtw_chip_wcpu_11ac(rtwdev)) {
 		rpwm = rtw_read8(rtwdev, rtwdev->hci.rpwm_addr);
@@ -273,8 +274,9 @@ static int rtw_mac_power_switch(struct rtw_dev *rtwdev, bool pwr_on)
 		return -EALREADY;
 
 	pwr_seq = pwr_on ? chip->pwr_on_seq : chip->pwr_off_seq;
-	if (rtw_pwr_seq_parser(rtwdev, pwr_seq))
-		return -EINVAL;
+	ret = rtw_pwr_seq_parser(rtwdev, pwr_seq);
+	if (ret)
+		return ret;
 
 	if (pwr_on)
 		set_bit(RTW_FLAG_POWERON, rtwdev->flags);
-- 
cgit v1.2.3-70-g09d2


From ff6f38eb920bd2b86e1d3157a01f409e2dce1320 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Thu, 2 Mar 2023 10:39:11 +0800
Subject: wifi: rtlwifi: rtl8192se: Remove some unused variables

Variables bcntime_cfg, bcn_cw and bcn_ifs are not effectively used, so
delete them.

drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c:1555:6: warning: variable 'bcntime_cfg' set but not used.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4240
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230302023911.59278-1-jiapeng.chong@linux.alibaba.com
---
 drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c
index bd0b7e365edb..a8b5bf45b1bb 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c
@@ -1552,8 +1552,6 @@ void rtl92se_set_beacon_related_registers(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
-	u16 bcntime_cfg = 0;
-	u16 bcn_cw = 6, bcn_ifs = 0xf;
 	u16 atim_window = 2;
 
 	/* ATIM Window (in unit of TU). */
@@ -1576,13 +1574,6 @@ void rtl92se_set_beacon_related_registers(struct ieee80211_hw *hw)
 	 * other ad hoc STA */
 	rtl_write_byte(rtlpriv, BCN_ERR_THRESH, 100);
 
-	/* Beacon Time Configuration */
-	if (mac->opmode == NL80211_IFTYPE_ADHOC)
-		bcntime_cfg |= (bcn_cw << BCN_TCFG_CW_SHIFT);
-
-	/* TODO: bcn_ifs may required to be changed on ASIC */
-	bcntime_cfg |= bcn_ifs << BCN_TCFG_IFS;
-
 	/*for beacon changed */
 	rtl92s_phy_set_beacon_hwreg(hw, mac->beacon_interval);
 }
-- 
cgit v1.2.3-70-g09d2


From fd4cb29f2a3d54ec7b4e012300321601af10bd68 Mon Sep 17 00:00:00 2001
From: Rong Tao <rongtao@cestc.cn>
Date: Sat, 4 Mar 2023 23:17:04 +0800
Subject: tools/resolve_btfids: Add /libsubcmd to .gitignore

Add libsubcmd to .gitignore, otherwise after compiling the kernel it
would result in the following:

    # bpf-next...bpf-next/master
    ?? tools/bpf/resolve_btfids/libsubcmd/

Signed-off-by: Rong Tao <rongtao@cestc.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/tencent_F13D670D5D7AA9C4BD868D3220921AAC090A@qq.com
---
 tools/bpf/resolve_btfids/.gitignore | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/bpf/resolve_btfids/.gitignore b/tools/bpf/resolve_btfids/.gitignore
index 16913fffc985..52d5e9721d92 100644
--- a/tools/bpf/resolve_btfids/.gitignore
+++ b/tools/bpf/resolve_btfids/.gitignore
@@ -1,3 +1,4 @@
 /fixdep
 /resolve_btfids
 /libbpf/
+/libsubcmd/
-- 
cgit v1.2.3-70-g09d2


From f8b299bc6a0f8695d4005ecff29773c79a1d29af Mon Sep 17 00:00:00 2001
From: Menglong Dong <imagedong@tencent.com>
Date: Mon, 6 Mar 2023 14:48:31 +0800
Subject: libbpf: Add support to set kprobe/uprobe attach mode

By default, libbpf will attach the kprobe/uprobe BPF program in the
latest mode that supported by kernel. In this patch, we add the support
to let users manually attach kprobe/uprobe in legacy or perf mode.

There are 3 mode that supported by the kernel to attach kprobe/uprobe:

  LEGACY: create perf event in legacy way and don't use bpf_link
  PERF: create perf event with perf_event_open() and don't use bpf_link

Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Biao Jiang <benbjiang@tencent.com>
Link: create perf event with perf_event_open() and use bpf_link
Link: https://lore.kernel.org/bpf/20230113093427.1666466-1-imagedong@tencent.com/
Link: https://lore.kernel.org/bpf/20230306064833.7932-2-imagedong@tencent.com

Users now can manually choose the mode with
bpf_program__attach_uprobe_opts()/bpf_program__attach_kprobe_opts().
---
 tools/lib/bpf/libbpf.c | 48 +++++++++++++++++++++++++++++++++++++++++++++++-
 tools/lib/bpf/libbpf.h | 50 +++++++++++++++++++++++++++++++++++++-------------
 2 files changed, 84 insertions(+), 14 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index dacaae31b76a..a557718401e4 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -9724,6 +9724,7 @@ struct bpf_link *bpf_program__attach_perf_event_opts(const struct bpf_program *p
 	char errmsg[STRERR_BUFSIZE];
 	struct bpf_link_perf *link;
 	int prog_fd, link_fd = -1, err;
+	bool force_ioctl_attach;
 
 	if (!OPTS_VALID(opts, bpf_perf_event_opts))
 		return libbpf_err_ptr(-EINVAL);
@@ -9747,7 +9748,8 @@ struct bpf_link *bpf_program__attach_perf_event_opts(const struct bpf_program *p
 	link->link.dealloc = &bpf_link_perf_dealloc;
 	link->perf_event_fd = pfd;
 
-	if (kernel_supports(prog->obj, FEAT_PERF_LINK)) {
+	force_ioctl_attach = OPTS_GET(opts, force_ioctl_attach, false);
+	if (kernel_supports(prog->obj, FEAT_PERF_LINK) && !force_ioctl_attach) {
 		DECLARE_LIBBPF_OPTS(bpf_link_create_opts, link_opts,
 			.perf_event.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0));
 
@@ -10106,6 +10108,7 @@ bpf_program__attach_kprobe_opts(const struct bpf_program *prog,
 				const struct bpf_kprobe_opts *opts)
 {
 	DECLARE_LIBBPF_OPTS(bpf_perf_event_opts, pe_opts);
+	enum probe_attach_mode attach_mode;
 	char errmsg[STRERR_BUFSIZE];
 	char *legacy_probe = NULL;
 	struct bpf_link *link;
@@ -10116,11 +10119,32 @@ bpf_program__attach_kprobe_opts(const struct bpf_program *prog,
 	if (!OPTS_VALID(opts, bpf_kprobe_opts))
 		return libbpf_err_ptr(-EINVAL);
 
+	attach_mode = OPTS_GET(opts, attach_mode, PROBE_ATTACH_MODE_DEFAULT);
 	retprobe = OPTS_GET(opts, retprobe, false);
 	offset = OPTS_GET(opts, offset, 0);
 	pe_opts.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0);
 
 	legacy = determine_kprobe_perf_type() < 0;
+	switch (attach_mode) {
+	case PROBE_ATTACH_MODE_LEGACY:
+		legacy = true;
+		pe_opts.force_ioctl_attach = true;
+		break;
+	case PROBE_ATTACH_MODE_PERF:
+		if (legacy)
+			return libbpf_err_ptr(-ENOTSUP);
+		pe_opts.force_ioctl_attach = true;
+		break;
+	case PROBE_ATTACH_MODE_LINK:
+		if (legacy || !kernel_supports(prog->obj, FEAT_PERF_LINK))
+			return libbpf_err_ptr(-ENOTSUP);
+		break;
+	case PROBE_ATTACH_MODE_DEFAULT:
+		break;
+	default:
+		return libbpf_err_ptr(-EINVAL);
+	}
+
 	if (!legacy) {
 		pfd = perf_event_open_probe(false /* uprobe */, retprobe,
 					    func_name, offset,
@@ -10852,6 +10876,7 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 	const char *archive_path = NULL, *archive_sep = NULL;
 	char errmsg[STRERR_BUFSIZE], *legacy_probe = NULL;
 	DECLARE_LIBBPF_OPTS(bpf_perf_event_opts, pe_opts);
+	enum probe_attach_mode attach_mode;
 	char full_path[PATH_MAX];
 	struct bpf_link *link;
 	size_t ref_ctr_off;
@@ -10862,6 +10887,7 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 	if (!OPTS_VALID(opts, bpf_uprobe_opts))
 		return libbpf_err_ptr(-EINVAL);
 
+	attach_mode = OPTS_GET(opts, attach_mode, PROBE_ATTACH_MODE_DEFAULT);
 	retprobe = OPTS_GET(opts, retprobe, false);
 	ref_ctr_off = OPTS_GET(opts, ref_ctr_offset, 0);
 	pe_opts.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0);
@@ -10903,6 +10929,26 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 	}
 
 	legacy = determine_uprobe_perf_type() < 0;
+	switch (attach_mode) {
+	case PROBE_ATTACH_MODE_LEGACY:
+		legacy = true;
+		pe_opts.force_ioctl_attach = true;
+		break;
+	case PROBE_ATTACH_MODE_PERF:
+		if (legacy)
+			return libbpf_err_ptr(-ENOTSUP);
+		pe_opts.force_ioctl_attach = true;
+		break;
+	case PROBE_ATTACH_MODE_LINK:
+		if (legacy || !kernel_supports(prog->obj, FEAT_PERF_LINK))
+			return libbpf_err_ptr(-ENOTSUP);
+		break;
+	case PROBE_ATTACH_MODE_DEFAULT:
+		break;
+	default:
+		return libbpf_err_ptr(-EINVAL);
+	}
+
 	if (!legacy) {
 		pfd = perf_event_open_probe(true /* uprobe */, retprobe, binary_path,
 					    func_offset, pid, ref_ctr_off);
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 2efd80f6f7b9..db4992a036f8 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -447,12 +447,15 @@ LIBBPF_API struct bpf_link *
 bpf_program__attach(const struct bpf_program *prog);
 
 struct bpf_perf_event_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 	/* custom user-provided value fetchable through bpf_get_attach_cookie() */
 	__u64 bpf_cookie;
+	/* don't use BPF link when attach BPF program */
+	bool force_ioctl_attach;
+	size_t :0;
 };
-#define bpf_perf_event_opts__last_field bpf_cookie
+#define bpf_perf_event_opts__last_field force_ioctl_attach
 
 LIBBPF_API struct bpf_link *
 bpf_program__attach_perf_event(const struct bpf_program *prog, int pfd);
@@ -461,8 +464,25 @@ LIBBPF_API struct bpf_link *
 bpf_program__attach_perf_event_opts(const struct bpf_program *prog, int pfd,
 				    const struct bpf_perf_event_opts *opts);
 
+/**
+ * enum probe_attach_mode - the mode to attach kprobe/uprobe
+ *
+ * force libbpf to attach kprobe/uprobe in specific mode, -ENOTSUP will
+ * be returned if it is not supported by the kernel.
+ */
+enum probe_attach_mode {
+	/* attach probe in latest supported mode by kernel */
+	PROBE_ATTACH_MODE_DEFAULT = 0,
+	/* attach probe in legacy mode, using debugfs/tracefs */
+	PROBE_ATTACH_MODE_LEGACY,
+	/* create perf event with perf_event_open() syscall */
+	PROBE_ATTACH_MODE_PERF,
+	/* attach probe with BPF link */
+	PROBE_ATTACH_MODE_LINK,
+};
+
 struct bpf_kprobe_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 	/* custom user-provided value fetchable through bpf_get_attach_cookie() */
 	__u64 bpf_cookie;
@@ -470,9 +490,11 @@ struct bpf_kprobe_opts {
 	size_t offset;
 	/* kprobe is return probe */
 	bool retprobe;
+	/* kprobe attach mode */
+	enum probe_attach_mode attach_mode;
 	size_t :0;
 };
-#define bpf_kprobe_opts__last_field retprobe
+#define bpf_kprobe_opts__last_field attach_mode
 
 LIBBPF_API struct bpf_link *
 bpf_program__attach_kprobe(const struct bpf_program *prog, bool retprobe,
@@ -506,7 +528,7 @@ bpf_program__attach_kprobe_multi_opts(const struct bpf_program *prog,
 				      const struct bpf_kprobe_multi_opts *opts);
 
 struct bpf_ksyscall_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 	/* custom user-provided value fetchable through bpf_get_attach_cookie() */
 	__u64 bpf_cookie;
@@ -552,7 +574,7 @@ bpf_program__attach_ksyscall(const struct bpf_program *prog,
 			     const struct bpf_ksyscall_opts *opts);
 
 struct bpf_uprobe_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 	/* offset of kernel reference counted USDT semaphore, added in
 	 * a6ca88b241d5 ("trace_uprobe: support reference counter in fd-based uprobe")
@@ -570,9 +592,11 @@ struct bpf_uprobe_opts {
 	 * binary_path.
 	 */
 	const char *func_name;
+	/* uprobe attach mode */
+	enum probe_attach_mode attach_mode;
 	size_t :0;
 };
-#define bpf_uprobe_opts__last_field func_name
+#define bpf_uprobe_opts__last_field attach_mode
 
 /**
  * @brief **bpf_program__attach_uprobe()** attaches a BPF program
@@ -646,7 +670,7 @@ bpf_program__attach_usdt(const struct bpf_program *prog,
 			 const struct bpf_usdt_opts *opts);
 
 struct bpf_tracepoint_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 	/* custom user-provided value fetchable through bpf_get_attach_cookie() */
 	__u64 bpf_cookie;
@@ -1110,7 +1134,7 @@ struct user_ring_buffer;
 typedef int (*ring_buffer_sample_fn)(void *ctx, void *data, size_t size);
 
 struct ring_buffer_opts {
-	size_t sz; /* size of this struct, for forward/backward compatiblity */
+	size_t sz; /* size of this struct, for forward/backward compatibility */
 };
 
 #define ring_buffer_opts__last_field sz
@@ -1475,7 +1499,7 @@ LIBBPF_API void
 bpf_object__destroy_subskeleton(struct bpf_object_subskeleton *s);
 
 struct gen_loader_opts {
-	size_t sz; /* size of this struct, for forward/backward compatiblity */
+	size_t sz; /* size of this struct, for forward/backward compatibility */
 	const char *data;
 	const char *insns;
 	__u32 data_sz;
@@ -1493,13 +1517,13 @@ enum libbpf_tristate {
 };
 
 struct bpf_linker_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 };
 #define bpf_linker_opts__last_field sz
 
 struct bpf_linker_file_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 };
 #define bpf_linker_file_opts__last_field sz
@@ -1542,7 +1566,7 @@ typedef int (*libbpf_prog_attach_fn_t)(const struct bpf_program *prog, long cook
 				       struct bpf_link **link);
 
 struct libbpf_prog_handler_opts {
-	/* size of this struct, for forward/backward compatiblity */
+	/* size of this struct, for forward/backward compatibility */
 	size_t sz;
 	/* User-provided value that is passed to prog_setup_fn,
 	 * prog_prepare_load_fn, and prog_attach_fn callbacks. Allows user to
-- 
cgit v1.2.3-70-g09d2


From 7391ec6391e2b129aeaee5462487c404f61157aa Mon Sep 17 00:00:00 2001
From: Menglong Dong <imagedong@tencent.com>
Date: Mon, 6 Mar 2023 14:48:32 +0800
Subject: selftests/bpf: Split test_attach_probe into multi subtests

In order to adapt to the older kernel, now we split the "attach_probe"
testing into multi subtests:

  manual // manual attach tests for kprobe/uprobe
  auto // auto-attach tests for kprobe and uprobe
  kprobe-sleepable // kprobe sleepable test
  uprobe-lib // uprobe tests for library function by name
  uprobe-sleepable // uprobe sleepable test
  uprobe-ref_ctr // uprobe ref_ctr test

As sleepable kprobe needs to set BPF_F_SLEEPABLE flag before loading,
we need to move it to a stand alone skel file, in case of it is not
supported by kernel and make the whole loading fail.

Therefore, we can only enable part of the subtests for older kernel.

Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Biao Jiang <benbjiang@tencent.com>
Link: https://lore.kernel.org/bpf/20230306064833.7932-3-imagedong@tencent.com
---
 .../selftests/bpf/prog_tests/attach_probe.c        | 260 ++++++++++++++-------
 .../bpf/progs/test_attach_kprobe_sleepable.c       |  23 ++
 .../selftests/bpf/progs/test_attach_probe.c        |  23 +-
 3 files changed, 205 insertions(+), 101 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/test_attach_kprobe_sleepable.c

diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index 56374c8b5436..c374759f39ce 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_attach_kprobe_sleepable.skel.h"
 #include "test_attach_probe.skel.h"
 
 /* this is how USDT semaphore is actually defined, except volatile modifier */
@@ -23,52 +24,24 @@ static noinline void trigger_func3(void)
 	asm volatile ("");
 }
 
+/* attach point for ref_ctr */
+static noinline void trigger_func4(void)
+{
+	asm volatile ("");
+}
+
 static char test_data[] = "test_data";
 
-void test_attach_probe(void)
+/* manual attach kprobe/kretprobe/uprobe/uretprobe testings */
+static void test_attach_probe_manual(struct test_attach_probe *skel)
 {
 	DECLARE_LIBBPF_OPTS(bpf_uprobe_opts, uprobe_opts);
 	struct bpf_link *kprobe_link, *kretprobe_link;
 	struct bpf_link *uprobe_link, *uretprobe_link;
-	struct test_attach_probe* skel;
-	ssize_t uprobe_offset, ref_ctr_offset;
-	struct bpf_link *uprobe_err_link;
-	FILE *devnull;
-	bool legacy;
-
-	/* Check if new-style kprobe/uprobe API is supported.
-	 * Kernels that support new FD-based kprobe and uprobe BPF attachment
-	 * through perf_event_open() syscall expose
-	 * /sys/bus/event_source/devices/kprobe/type and
-	 * /sys/bus/event_source/devices/uprobe/type files, respectively. They
-	 * contain magic numbers that are passed as "type" field of
-	 * perf_event_attr. Lack of such file in the system indicates legacy
-	 * kernel with old-style kprobe/uprobe attach interface through
-	 * creating per-probe event through tracefs. For such cases
-	 * ref_ctr_offset feature is not supported, so we don't test it.
-	 */
-	legacy = access("/sys/bus/event_source/devices/kprobe/type", F_OK) != 0;
+	ssize_t uprobe_offset;
 
 	uprobe_offset = get_uprobe_offset(&trigger_func);
 	if (!ASSERT_GE(uprobe_offset, 0, "uprobe_offset"))
-		return;
-
-	ref_ctr_offset = get_rel_offset((uintptr_t)&uprobe_ref_ctr);
-	if (!ASSERT_GE(ref_ctr_offset, 0, "ref_ctr_offset"))
-		return;
-
-	skel = test_attach_probe__open();
-	if (!ASSERT_OK_PTR(skel, "skel_open"))
-		return;
-
-	/* sleepable kprobe test case needs flags set before loading */
-	if (!ASSERT_OK(bpf_program__set_flags(skel->progs.handle_kprobe_sleepable,
-		BPF_F_SLEEPABLE), "kprobe_sleepable_flags"))
-		goto cleanup;
-
-	if (!ASSERT_OK(test_attach_probe__load(skel), "skel_load"))
-		goto cleanup;
-	if (!ASSERT_OK_PTR(skel->bss, "check_bss"))
 		goto cleanup;
 
 	/* manual-attach kprobe/kretprobe */
@@ -86,18 +59,9 @@ void test_attach_probe(void)
 		goto cleanup;
 	skel->links.handle_kretprobe = kretprobe_link;
 
-	/* auto-attachable kprobe and kretprobe */
-	skel->links.handle_kprobe_auto = bpf_program__attach(skel->progs.handle_kprobe_auto);
-	ASSERT_OK_PTR(skel->links.handle_kprobe_auto, "attach_kprobe_auto");
-
-	skel->links.handle_kretprobe_auto = bpf_program__attach(skel->progs.handle_kretprobe_auto);
-	ASSERT_OK_PTR(skel->links.handle_kretprobe_auto, "attach_kretprobe_auto");
-
-	if (!legacy)
-		ASSERT_EQ(uprobe_ref_ctr, 0, "uprobe_ref_ctr_before");
-
+	/* manual-attach uprobe/uretprobe */
+	uprobe_opts.ref_ctr_offset = 0;
 	uprobe_opts.retprobe = false;
-	uprobe_opts.ref_ctr_offset = legacy ? 0 : ref_ctr_offset;
 	uprobe_link = bpf_program__attach_uprobe_opts(skel->progs.handle_uprobe,
 						      0 /* self pid */,
 						      "/proc/self/exe",
@@ -107,12 +71,7 @@ void test_attach_probe(void)
 		goto cleanup;
 	skel->links.handle_uprobe = uprobe_link;
 
-	if (!legacy)
-		ASSERT_GT(uprobe_ref_ctr, 0, "uprobe_ref_ctr_after");
-
-	/* if uprobe uses ref_ctr, uretprobe has to use ref_ctr as well */
 	uprobe_opts.retprobe = true;
-	uprobe_opts.ref_ctr_offset = legacy ? 0 : ref_ctr_offset;
 	uretprobe_link = bpf_program__attach_uprobe_opts(skel->progs.handle_uretprobe,
 							 -1 /* any pid */,
 							 "/proc/self/exe",
@@ -121,12 +80,7 @@ void test_attach_probe(void)
 		goto cleanup;
 	skel->links.handle_uretprobe = uretprobe_link;
 
-	/* verify auto-attach fails for old-style uprobe definition */
-	uprobe_err_link = bpf_program__attach(skel->progs.handle_uprobe_byname);
-	if (!ASSERT_EQ(libbpf_get_error(uprobe_err_link), -EOPNOTSUPP,
-		       "auto-attach should fail for old-style name"))
-		goto cleanup;
-
+	/* attach uprobe by function name manually */
 	uprobe_opts.func_name = "trigger_func2";
 	uprobe_opts.retprobe = false;
 	uprobe_opts.ref_ctr_offset = 0;
@@ -138,11 +92,62 @@ void test_attach_probe(void)
 	if (!ASSERT_OK_PTR(skel->links.handle_uprobe_byname, "attach_uprobe_byname"))
 		goto cleanup;
 
+	/* trigger & validate kprobe && kretprobe */
+	usleep(1);
+
+	/* trigger & validate uprobe & uretprobe */
+	trigger_func();
+
+	/* trigger & validate uprobe attached by name */
+	trigger_func2();
+
+	ASSERT_EQ(skel->bss->kprobe_res, 1, "check_kprobe_res");
+	ASSERT_EQ(skel->bss->kretprobe_res, 2, "check_kretprobe_res");
+	ASSERT_EQ(skel->bss->uprobe_res, 3, "check_uprobe_res");
+	ASSERT_EQ(skel->bss->uretprobe_res, 4, "check_uretprobe_res");
+	ASSERT_EQ(skel->bss->uprobe_byname_res, 5, "check_uprobe_byname_res");
+
+cleanup:
+}
+
+static void test_attach_probe_auto(struct test_attach_probe *skel)
+{
+	struct bpf_link *uprobe_err_link;
+
+	/* auto-attachable kprobe and kretprobe */
+	skel->links.handle_kprobe_auto = bpf_program__attach(skel->progs.handle_kprobe_auto);
+	ASSERT_OK_PTR(skel->links.handle_kprobe_auto, "attach_kprobe_auto");
+
+	skel->links.handle_kretprobe_auto = bpf_program__attach(skel->progs.handle_kretprobe_auto);
+	ASSERT_OK_PTR(skel->links.handle_kretprobe_auto, "attach_kretprobe_auto");
+
+	/* verify auto-attach fails for old-style uprobe definition */
+	uprobe_err_link = bpf_program__attach(skel->progs.handle_uprobe_byname);
+	if (!ASSERT_EQ(libbpf_get_error(uprobe_err_link), -EOPNOTSUPP,
+		       "auto-attach should fail for old-style name"))
+		return;
+
 	/* verify auto-attach works */
 	skel->links.handle_uretprobe_byname =
 			bpf_program__attach(skel->progs.handle_uretprobe_byname);
 	if (!ASSERT_OK_PTR(skel->links.handle_uretprobe_byname, "attach_uretprobe_byname"))
-		goto cleanup;
+		return;
+
+	/* trigger & validate kprobe && kretprobe */
+	usleep(1);
+
+	/* trigger & validate uprobe attached by name */
+	trigger_func2();
+
+	ASSERT_EQ(skel->bss->kprobe2_res, 11, "check_kprobe_auto_res");
+	ASSERT_EQ(skel->bss->kretprobe2_res, 22, "check_kretprobe_auto_res");
+	ASSERT_EQ(skel->bss->uretprobe_byname_res, 6, "check_uretprobe_byname_res");
+}
+
+static void test_uprobe_lib(struct test_attach_probe *skel)
+{
+	DECLARE_LIBBPF_OPTS(bpf_uprobe_opts, uprobe_opts);
+	FILE *devnull;
 
 	/* test attach by name for a library function, using the library
 	 * as the binary argument. libc.so.6 will be resolved via dlopen()/dlinfo().
@@ -155,7 +160,7 @@ void test_attach_probe(void)
 							"libc.so.6",
 							0, &uprobe_opts);
 	if (!ASSERT_OK_PTR(skel->links.handle_uprobe_byname2, "attach_uprobe_byname2"))
-		goto cleanup;
+		return;
 
 	uprobe_opts.func_name = "fclose";
 	uprobe_opts.retprobe = true;
@@ -165,62 +170,137 @@ void test_attach_probe(void)
 							"libc.so.6",
 							0, &uprobe_opts);
 	if (!ASSERT_OK_PTR(skel->links.handle_uretprobe_byname2, "attach_uretprobe_byname2"))
+		return;
+
+	/* trigger & validate shared library u[ret]probes attached by name */
+	devnull = fopen("/dev/null", "r");
+	fclose(devnull);
+
+	ASSERT_EQ(skel->bss->uprobe_byname2_res, 7, "check_uprobe_byname2_res");
+	ASSERT_EQ(skel->bss->uretprobe_byname2_res, 8, "check_uretprobe_byname2_res");
+}
+
+static void test_uprobe_ref_ctr(struct test_attach_probe *skel)
+{
+	DECLARE_LIBBPF_OPTS(bpf_uprobe_opts, uprobe_opts);
+	struct bpf_link *uprobe_link, *uretprobe_link;
+	ssize_t uprobe_offset, ref_ctr_offset;
+
+	uprobe_offset = get_uprobe_offset(&trigger_func4);
+	if (!ASSERT_GE(uprobe_offset, 0, "uprobe_offset_ref_ctr"))
+		return;
+
+	ref_ctr_offset = get_rel_offset((uintptr_t)&uprobe_ref_ctr);
+	if (!ASSERT_GE(ref_ctr_offset, 0, "ref_ctr_offset"))
+		return;
+
+	ASSERT_EQ(uprobe_ref_ctr, 0, "uprobe_ref_ctr_before");
+
+	uprobe_opts.retprobe = false;
+	uprobe_opts.ref_ctr_offset = ref_ctr_offset;
+	uprobe_link = bpf_program__attach_uprobe_opts(skel->progs.handle_uprobe_ref_ctr,
+						      0 /* self pid */,
+						      "/proc/self/exe",
+						      uprobe_offset,
+						      &uprobe_opts);
+	if (!ASSERT_OK_PTR(uprobe_link, "attach_uprobe_ref_ctr"))
+		return;
+	skel->links.handle_uprobe_ref_ctr = uprobe_link;
+
+	ASSERT_GT(uprobe_ref_ctr, 0, "uprobe_ref_ctr_after");
+
+	/* if uprobe uses ref_ctr, uretprobe has to use ref_ctr as well */
+	uprobe_opts.retprobe = true;
+	uprobe_opts.ref_ctr_offset = ref_ctr_offset;
+	uretprobe_link = bpf_program__attach_uprobe_opts(skel->progs.handle_uretprobe_ref_ctr,
+							 -1 /* any pid */,
+							 "/proc/self/exe",
+							 uprobe_offset, &uprobe_opts);
+	if (!ASSERT_OK_PTR(uretprobe_link, "attach_uretprobe_ref_ctr"))
+		return;
+	skel->links.handle_uretprobe_ref_ctr = uretprobe_link;
+}
+
+static void test_kprobe_sleepable(void)
+{
+	struct test_attach_kprobe_sleepable *skel;
+
+	skel = test_attach_kprobe_sleepable__open();
+	if (!ASSERT_OK_PTR(skel, "skel_kprobe_sleepable_open"))
+		return;
+
+	/* sleepable kprobe test case needs flags set before loading */
+	if (!ASSERT_OK(bpf_program__set_flags(skel->progs.handle_kprobe_sleepable,
+		BPF_F_SLEEPABLE), "kprobe_sleepable_flags"))
+		goto cleanup;
+
+	if (!ASSERT_OK(test_attach_kprobe_sleepable__load(skel),
+		       "skel_kprobe_sleepable_load"))
 		goto cleanup;
 
 	/* sleepable kprobes should not attach successfully */
 	skel->links.handle_kprobe_sleepable = bpf_program__attach(skel->progs.handle_kprobe_sleepable);
-	if (!ASSERT_ERR_PTR(skel->links.handle_kprobe_sleepable, "attach_kprobe_sleepable"))
-		goto cleanup;
+	ASSERT_ERR_PTR(skel->links.handle_kprobe_sleepable, "attach_kprobe_sleepable");
 
+cleanup:
+	test_attach_kprobe_sleepable__destroy(skel);
+}
+
+static void test_uprobe_sleepable(struct test_attach_probe *skel)
+{
 	/* test sleepable uprobe and uretprobe variants */
 	skel->links.handle_uprobe_byname3_sleepable = bpf_program__attach(skel->progs.handle_uprobe_byname3_sleepable);
 	if (!ASSERT_OK_PTR(skel->links.handle_uprobe_byname3_sleepable, "attach_uprobe_byname3_sleepable"))
-		goto cleanup;
+		return;
 
 	skel->links.handle_uprobe_byname3 = bpf_program__attach(skel->progs.handle_uprobe_byname3);
 	if (!ASSERT_OK_PTR(skel->links.handle_uprobe_byname3, "attach_uprobe_byname3"))
-		goto cleanup;
+		return;
 
 	skel->links.handle_uretprobe_byname3_sleepable = bpf_program__attach(skel->progs.handle_uretprobe_byname3_sleepable);
 	if (!ASSERT_OK_PTR(skel->links.handle_uretprobe_byname3_sleepable, "attach_uretprobe_byname3_sleepable"))
-		goto cleanup;
+		return;
 
 	skel->links.handle_uretprobe_byname3 = bpf_program__attach(skel->progs.handle_uretprobe_byname3);
 	if (!ASSERT_OK_PTR(skel->links.handle_uretprobe_byname3, "attach_uretprobe_byname3"))
-		goto cleanup;
+		return;
 
 	skel->bss->user_ptr = test_data;
 
-	/* trigger & validate kprobe && kretprobe */
-	usleep(1);
-
-	/* trigger & validate shared library u[ret]probes attached by name */
-	devnull = fopen("/dev/null", "r");
-	fclose(devnull);
-
-	/* trigger & validate uprobe & uretprobe */
-	trigger_func();
-
-	/* trigger & validate uprobe attached by name */
-	trigger_func2();
-
 	/* trigger & validate sleepable uprobe attached by name */
 	trigger_func3();
 
-	ASSERT_EQ(skel->bss->kprobe_res, 1, "check_kprobe_res");
-	ASSERT_EQ(skel->bss->kprobe2_res, 11, "check_kprobe_auto_res");
-	ASSERT_EQ(skel->bss->kretprobe_res, 2, "check_kretprobe_res");
-	ASSERT_EQ(skel->bss->kretprobe2_res, 22, "check_kretprobe_auto_res");
-	ASSERT_EQ(skel->bss->uprobe_res, 3, "check_uprobe_res");
-	ASSERT_EQ(skel->bss->uretprobe_res, 4, "check_uretprobe_res");
-	ASSERT_EQ(skel->bss->uprobe_byname_res, 5, "check_uprobe_byname_res");
-	ASSERT_EQ(skel->bss->uretprobe_byname_res, 6, "check_uretprobe_byname_res");
-	ASSERT_EQ(skel->bss->uprobe_byname2_res, 7, "check_uprobe_byname2_res");
-	ASSERT_EQ(skel->bss->uretprobe_byname2_res, 8, "check_uretprobe_byname2_res");
 	ASSERT_EQ(skel->bss->uprobe_byname3_sleepable_res, 9, "check_uprobe_byname3_sleepable_res");
 	ASSERT_EQ(skel->bss->uprobe_byname3_res, 10, "check_uprobe_byname3_res");
 	ASSERT_EQ(skel->bss->uretprobe_byname3_sleepable_res, 11, "check_uretprobe_byname3_sleepable_res");
 	ASSERT_EQ(skel->bss->uretprobe_byname3_res, 12, "check_uretprobe_byname3_res");
+}
+
+void test_attach_probe(void)
+{
+	struct test_attach_probe *skel;
+
+	skel = test_attach_probe__open();
+	if (!ASSERT_OK_PTR(skel, "skel_open"))
+		return;
+
+	if (!ASSERT_OK(test_attach_probe__load(skel), "skel_load"))
+		goto cleanup;
+	if (!ASSERT_OK_PTR(skel->bss, "check_bss"))
+		goto cleanup;
+
+	if (test__start_subtest("manual"))
+		test_attach_probe_manual(skel);
+	if (test__start_subtest("auto"))
+		test_attach_probe_auto(skel);
+	if (test__start_subtest("kprobe-sleepable"))
+		test_kprobe_sleepable();
+	if (test__start_subtest("uprobe-lib"))
+		test_uprobe_lib(skel);
+	if (test__start_subtest("uprobe-sleepable"))
+		test_uprobe_sleepable(skel);
+	if (test__start_subtest("uprobe-ref_ctr"))
+		test_uprobe_ref_ctr(skel);
 
 cleanup:
 	test_attach_probe__destroy(skel);
diff --git a/tools/testing/selftests/bpf/progs/test_attach_kprobe_sleepable.c b/tools/testing/selftests/bpf/progs/test_attach_kprobe_sleepable.c
new file mode 100644
index 000000000000..f548b7446218
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_attach_kprobe_sleepable.c
@@ -0,0 +1,23 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2017 Facebook
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_core_read.h>
+#include "bpf_misc.h"
+
+int kprobe_res = 0;
+
+/**
+ * This program will be manually made sleepable on the userspace side
+ * and should thus be unattachable.
+ */
+SEC("kprobe/" SYS_PREFIX "sys_nanosleep")
+int handle_kprobe_sleepable(struct pt_regs *ctx)
+{
+	kprobe_res = 1;
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index 3b5dc34d23e9..9e1e7163bb67 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -37,17 +37,6 @@ int BPF_KSYSCALL(handle_kprobe_auto, struct __kernel_timespec *req, struct __ker
 	return 0;
 }
 
-/**
- * This program will be manually made sleepable on the userspace side
- * and should thus be unattachable.
- */
-SEC("kprobe/" SYS_PREFIX "sys_nanosleep")
-int handle_kprobe_sleepable(struct pt_regs *ctx)
-{
-	kprobe_res = 2;
-	return 0;
-}
-
 SEC("kretprobe")
 int handle_kretprobe(struct pt_regs *ctx)
 {
@@ -76,6 +65,18 @@ int handle_uretprobe(struct pt_regs *ctx)
 	return 0;
 }
 
+SEC("uprobe")
+int handle_uprobe_ref_ctr(struct pt_regs *ctx)
+{
+	return 0;
+}
+
+SEC("uretprobe")
+int handle_uretprobe_ref_ctr(struct pt_regs *ctx)
+{
+	return 0;
+}
+
 SEC("uprobe")
 int handle_uprobe_byname(struct pt_regs *ctx)
 {
-- 
cgit v1.2.3-70-g09d2


From c7aec81b31e43a0aa94ee55d9bb33d70b1046f76 Mon Sep 17 00:00:00 2001
From: Menglong Dong <imagedong@tencent.com>
Date: Mon, 6 Mar 2023 14:48:33 +0800
Subject: selftests/bpf: Add test for legacy/perf kprobe/uprobe attach mode

Add the testing for kprobe/uprobe attaching in default, legacy, perf and
link mode. And the testing passed:

./test_progs -t attach_probe
$5/1     attach_probe/manual-default:OK
$5/2     attach_probe/manual-legacy:OK
$5/3     attach_probe/manual-perf:OK
$5/4     attach_probe/manual-link:OK
$5/5     attach_probe/auto:OK
$5/6     attach_probe/kprobe-sleepable:OK
$5/7     attach_probe/uprobe-lib:OK
$5/8     attach_probe/uprobe-sleepable:OK
$5/9     attach_probe/uprobe-ref_ctr:OK
$5       attach_probe:OK
Summary: 1/9 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Biao Jiang <benbjiang@tencent.com>
Link: https://lore.kernel.org/bpf/20230306064833.7932-4-imagedong@tencent.com
---
 .../selftests/bpf/prog_tests/attach_probe.c        | 37 +++++++++++----
 .../selftests/bpf/progs/test_attach_probe.c        | 32 -------------
 .../selftests/bpf/progs/test_attach_probe_manual.c | 53 ++++++++++++++++++++++
 3 files changed, 81 insertions(+), 41 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/test_attach_probe_manual.c

diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index c374759f39ce..7175af39134f 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
 #include "test_attach_kprobe_sleepable.skel.h"
+#include "test_attach_probe_manual.skel.h"
 #include "test_attach_probe.skel.h"
 
 /* this is how USDT semaphore is actually defined, except volatile modifier */
@@ -33,33 +34,43 @@ static noinline void trigger_func4(void)
 static char test_data[] = "test_data";
 
 /* manual attach kprobe/kretprobe/uprobe/uretprobe testings */
-static void test_attach_probe_manual(struct test_attach_probe *skel)
+static void test_attach_probe_manual(enum probe_attach_mode attach_mode)
 {
 	DECLARE_LIBBPF_OPTS(bpf_uprobe_opts, uprobe_opts);
+	DECLARE_LIBBPF_OPTS(bpf_kprobe_opts, kprobe_opts);
 	struct bpf_link *kprobe_link, *kretprobe_link;
 	struct bpf_link *uprobe_link, *uretprobe_link;
+	struct test_attach_probe_manual *skel;
 	ssize_t uprobe_offset;
 
+	skel = test_attach_probe_manual__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel_kprobe_manual_open_and_load"))
+		return;
+
 	uprobe_offset = get_uprobe_offset(&trigger_func);
 	if (!ASSERT_GE(uprobe_offset, 0, "uprobe_offset"))
 		goto cleanup;
 
 	/* manual-attach kprobe/kretprobe */
-	kprobe_link = bpf_program__attach_kprobe(skel->progs.handle_kprobe,
-						 false /* retprobe */,
-						 SYS_NANOSLEEP_KPROBE_NAME);
+	kprobe_opts.attach_mode = attach_mode;
+	kprobe_opts.retprobe = false;
+	kprobe_link = bpf_program__attach_kprobe_opts(skel->progs.handle_kprobe,
+						      SYS_NANOSLEEP_KPROBE_NAME,
+						      &kprobe_opts);
 	if (!ASSERT_OK_PTR(kprobe_link, "attach_kprobe"))
 		goto cleanup;
 	skel->links.handle_kprobe = kprobe_link;
 
-	kretprobe_link = bpf_program__attach_kprobe(skel->progs.handle_kretprobe,
-						    true /* retprobe */,
-						    SYS_NANOSLEEP_KPROBE_NAME);
+	kprobe_opts.retprobe = true;
+	kretprobe_link = bpf_program__attach_kprobe_opts(skel->progs.handle_kretprobe,
+							 SYS_NANOSLEEP_KPROBE_NAME,
+							 &kprobe_opts);
 	if (!ASSERT_OK_PTR(kretprobe_link, "attach_kretprobe"))
 		goto cleanup;
 	skel->links.handle_kretprobe = kretprobe_link;
 
 	/* manual-attach uprobe/uretprobe */
+	uprobe_opts.attach_mode = attach_mode;
 	uprobe_opts.ref_ctr_offset = 0;
 	uprobe_opts.retprobe = false;
 	uprobe_link = bpf_program__attach_uprobe_opts(skel->progs.handle_uprobe,
@@ -108,6 +119,7 @@ static void test_attach_probe_manual(struct test_attach_probe *skel)
 	ASSERT_EQ(skel->bss->uprobe_byname_res, 5, "check_uprobe_byname_res");
 
 cleanup:
+	test_attach_probe_manual__destroy(skel);
 }
 
 static void test_attach_probe_auto(struct test_attach_probe *skel)
@@ -289,8 +301,15 @@ void test_attach_probe(void)
 	if (!ASSERT_OK_PTR(skel->bss, "check_bss"))
 		goto cleanup;
 
-	if (test__start_subtest("manual"))
-		test_attach_probe_manual(skel);
+	if (test__start_subtest("manual-default"))
+		test_attach_probe_manual(PROBE_ATTACH_MODE_DEFAULT);
+	if (test__start_subtest("manual-legacy"))
+		test_attach_probe_manual(PROBE_ATTACH_MODE_LEGACY);
+	if (test__start_subtest("manual-perf"))
+		test_attach_probe_manual(PROBE_ATTACH_MODE_PERF);
+	if (test__start_subtest("manual-link"))
+		test_attach_probe_manual(PROBE_ATTACH_MODE_LINK);
+
 	if (test__start_subtest("auto"))
 		test_attach_probe_auto(skel);
 	if (test__start_subtest("kprobe-sleepable"))
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index 9e1e7163bb67..68466a6ad18c 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -7,12 +7,8 @@
 #include <bpf/bpf_core_read.h>
 #include "bpf_misc.h"
 
-int kprobe_res = 0;
 int kprobe2_res = 0;
-int kretprobe_res = 0;
 int kretprobe2_res = 0;
-int uprobe_res = 0;
-int uretprobe_res = 0;
 int uprobe_byname_res = 0;
 int uretprobe_byname_res = 0;
 int uprobe_byname2_res = 0;
@@ -23,13 +19,6 @@ int uretprobe_byname3_sleepable_res = 0;
 int uretprobe_byname3_res = 0;
 void *user_ptr = 0;
 
-SEC("kprobe")
-int handle_kprobe(struct pt_regs *ctx)
-{
-	kprobe_res = 1;
-	return 0;
-}
-
 SEC("ksyscall/nanosleep")
 int BPF_KSYSCALL(handle_kprobe_auto, struct __kernel_timespec *req, struct __kernel_timespec *rem)
 {
@@ -37,13 +26,6 @@ int BPF_KSYSCALL(handle_kprobe_auto, struct __kernel_timespec *req, struct __ker
 	return 0;
 }
 
-SEC("kretprobe")
-int handle_kretprobe(struct pt_regs *ctx)
-{
-	kretprobe_res = 2;
-	return 0;
-}
-
 SEC("kretsyscall/nanosleep")
 int BPF_KRETPROBE(handle_kretprobe_auto, int ret)
 {
@@ -51,20 +33,6 @@ int BPF_KRETPROBE(handle_kretprobe_auto, int ret)
 	return ret;
 }
 
-SEC("uprobe")
-int handle_uprobe(struct pt_regs *ctx)
-{
-	uprobe_res = 3;
-	return 0;
-}
-
-SEC("uretprobe")
-int handle_uretprobe(struct pt_regs *ctx)
-{
-	uretprobe_res = 4;
-	return 0;
-}
-
 SEC("uprobe")
 int handle_uprobe_ref_ctr(struct pt_regs *ctx)
 {
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe_manual.c b/tools/testing/selftests/bpf/progs/test_attach_probe_manual.c
new file mode 100644
index 000000000000..7f08bce94596
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe_manual.c
@@ -0,0 +1,53 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2017 Facebook
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_core_read.h>
+#include "bpf_misc.h"
+
+int kprobe_res = 0;
+int kretprobe_res = 0;
+int uprobe_res = 0;
+int uretprobe_res = 0;
+int uprobe_byname_res = 0;
+void *user_ptr = 0;
+
+SEC("kprobe")
+int handle_kprobe(struct pt_regs *ctx)
+{
+	kprobe_res = 1;
+	return 0;
+}
+
+SEC("kretprobe")
+int handle_kretprobe(struct pt_regs *ctx)
+{
+	kretprobe_res = 2;
+	return 0;
+}
+
+SEC("uprobe")
+int handle_uprobe(struct pt_regs *ctx)
+{
+	uprobe_res = 3;
+	return 0;
+}
+
+SEC("uretprobe")
+int handle_uretprobe(struct pt_regs *ctx)
+{
+	uretprobe_res = 4;
+	return 0;
+}
+
+SEC("uprobe")
+int handle_uprobe_byname(struct pt_regs *ctx)
+{
+	uprobe_byname_res = 5;
+	return 0;
+}
+
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From d509c55cda22096a1836e35b03f66e1ef411c0c2 Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Wed, 1 Mar 2023 12:09:13 +0200
Subject: wifi: nl80211: Update the documentation of
 NL80211_SCAN_FLAG_COLOCATED_6GHZ

Add a detailed description of NL80211_SCAN_FLAG_COLOCATED_6GHZ
flag.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.487ab04feb39.I5129fd61841332474693046241586f057b134c3c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/uapi/linux/nl80211.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index f14621a954e1..c22eeb18b996 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -6510,8 +6510,14 @@ enum nl80211_timeout_reason {
  * @NL80211_SCAN_FLAG_FREQ_KHZ: report scan results with
  *	%NL80211_ATTR_SCAN_FREQ_KHZ. This also means
  *	%NL80211_ATTR_SCAN_FREQUENCIES will not be included.
- * @NL80211_SCAN_FLAG_COLOCATED_6GHZ: scan for colocated APs reported by
- *	2.4/5 GHz APs
+ * @NL80211_SCAN_FLAG_COLOCATED_6GHZ: scan for collocated APs reported by
+ *	2.4/5 GHz APs. When the flag is set, the scan logic will use the
+ *	information from the RNR element found in beacons/probe responses
+ *	received on the 2.4/5 GHz channels to actively scan only the 6GHz
+ *	channels on which APs are expected to be found. Note that when not set,
+ *	the scan logic would scan all 6GHz channels, but since transmission of
+ *	probe requests on non PSC channels is limited, it is highly likely that
+ *	these channels would passively be scanned.
  */
 enum nl80211_scan_flags {
 	NL80211_SCAN_FLAG_LOW_PRIORITY				= 1<<0,
-- 
cgit v1.2.3-70-g09d2


From 0fd3af61731567dc0ad798c88c20f6d63fe331ea Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:14 +0200
Subject: wifi: mac80211: adjust scan cancel comment/check

Instead of the comment about holding RTNL, which is now wrong,
add a proper lockdep assertion for the wiphy mutex.

Fixes: a05829a7222e ("cfg80211: avoid holding the RTNL when calling the driver")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.84352e46f342.Id90fef8c581cebe19cb30274340cf43885d55c74@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/scan.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c
index dc3cdee51e66..32fa8aca7005 100644
--- a/net/mac80211/scan.c
+++ b/net/mac80211/scan.c
@@ -9,7 +9,7 @@
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2013-2015  Intel Mobile Communications GmbH
  * Copyright 2016-2017  Intel Deutschland GmbH
- * Copyright (C) 2018-2021 Intel Corporation
+ * Copyright (C) 2018-2022 Intel Corporation
  */
 
 #include <linux/if_arp.h>
@@ -1246,11 +1246,11 @@ int ieee80211_request_ibss_scan(struct ieee80211_sub_if_data *sdata,
 	return ret;
 }
 
-/*
- * Only call this function when a scan can't be queued -- under RTNL.
- */
 void ieee80211_scan_cancel(struct ieee80211_local *local)
 {
+	/* ensure a new scan cannot be queued */
+	lockdep_assert_wiphy(local->hw.wiphy);
+
 	/*
 	 * We are canceling software scan, or deferred scan that was not
 	 * yet really started (see __ieee80211_start_scan ).
-- 
cgit v1.2.3-70-g09d2


From 3a867c7eef812ad4684dccd8825f1099a60b8c98 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Wed, 1 Mar 2023 12:09:15 +0200
Subject: wifi: mac80211: clear all bits that relate rtap fields on skb

Since we remove radiotap from skb data, clear all RX_FLAG_X related
info that indicate info on the skb data.

Also we need to do it only once so remove the clear from cooked_monitor.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.74d3efe19eae.Ie17a35864d2e120f9858516a2e3d3047d83cf805@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/rx.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index f7fdfe710951..095bcd2552bb 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -43,6 +43,7 @@ static struct sk_buff *ieee80211_clean_skb(struct sk_buff *skb,
 					   unsigned int present_fcs_len,
 					   unsigned int rtap_space)
 {
+	struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb);
 	struct ieee80211_hdr *hdr;
 	unsigned int hdrlen;
 	__le16 fc;
@@ -51,6 +52,14 @@ static struct sk_buff *ieee80211_clean_skb(struct sk_buff *skb,
 		__pskb_trim(skb, skb->len - present_fcs_len);
 	pskb_pull(skb, rtap_space);
 
+	/* After pulling radiotap header, clear all flags that indicate
+	 * info in skb->data.
+	 */
+	status->flag &= ~(RX_FLAG_RADIOTAP_VENDOR_DATA |
+			  RX_FLAG_RADIOTAP_LSIG |
+			  RX_FLAG_RADIOTAP_HE_MU |
+			  RX_FLAG_RADIOTAP_HE);
+
 	hdr = (void *)skb->data;
 	fc = hdr->frame_control;
 
@@ -3916,8 +3925,6 @@ static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx,
 	if (!local->cooked_mntrs)
 		goto out_free_skb;
 
-	/* vendor data is long removed here */
-	status->flag &= ~RX_FLAG_RADIOTAP_VENDOR_DATA;
 	/* room for the radiotap header based on driver features */
 	needed_headroom = ieee80211_rx_radiotap_hdrlen(local, status, skb);
 
-- 
cgit v1.2.3-70-g09d2


From 3ffcc659dc6fa84da7db57f56fd48756c5a60fba Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:16 +0200
Subject: wifi: mac80211: check key taint for beacon protection

This will likely never happen, but for completeness check
the key taint flag before using a key for beacon protection.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.cf2c3fee6f1f.I2f19b3e04e31c99bed3c9dc71935bf513b2cd177@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/tx.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 7699fb410670..1f9294f5baa4 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -5115,6 +5115,12 @@ static int ieee80211_beacon_protect(struct sk_buff *skb,
 	tx.key = rcu_dereference(link->default_beacon_key);
 	if (!tx.key)
 		return 0;
+
+	if (unlikely(tx.key->flags & KEY_FLAG_TAINTED)) {
+		tx.key = NULL;
+		return -EINVAL;
+	}
+
 	tx.local = local;
 	tx.sdata = sdata;
 	__skb_queue_head_init(&tx.skbs);
-- 
cgit v1.2.3-70-g09d2


From d1b9bb6520fe0c295128229f5d3a9298127d5c8a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:17 +0200
Subject: wifi: mac80211: allow beacon protection HW offload

In case of beacon protection, check if the key was offloaded
to the hardware and in that case set control.hw_key so that
the encryption function will see it and only do the needed
steps that aren't done in hardware.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.b2becd9a22fb.I6c0b9c50c6a481128ba912a11cb7afc92c4b6da7@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/tx.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 1f9294f5baa4..628d60f3db2a 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -5121,6 +5121,10 @@ static int ieee80211_beacon_protect(struct sk_buff *skb,
 		return -EINVAL;
 	}
 
+	if (!(tx.key->conf.flags & IEEE80211_KEY_FLAG_SW_MGMT_TX) &&
+	    tx.key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE)
+		IEEE80211_SKB_CB(skb)->control.hw_key = &tx.key->conf;
+
 	tx.local = local;
 	tx.sdata = sdata;
 	__skb_queue_head_init(&tx.skbs);
-- 
cgit v1.2.3-70-g09d2


From 724a486cd24581928f4a82619806939119304645 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Wed, 1 Mar 2023 12:09:18 +0200
Subject: wifi: wireless: return primary channel regardless of DUP

Currently in case DUP bit is not set we don't return the primary
channel for 6 GHz Band, but the spec says that the DUP bit has
no effect on this field:

9.4.2.249 HE Operation element:
The Duplicate Beacon subfield is set to 1 if the AP transmits
Beacon frames in non-HT duplicate PPDU with a TXVECTOR parameter
CH_BANDWIDTH value that is up to the BSS bandwidth and is set to
0 otherwise.

So remove the condition for returning primary channel based on DUP.

Since the caller code already marks the signal as invalid in case
the indicated frequency is not the tuned frequency, there's no need
to additionally handle this case here since that's already true for
duplicated beacons on the non-primary channel(s).

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.66d7f05f7d11.I5e0add054f72ede95611391b99804c61c40cc959@changeid
[clarify commit message]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/scan.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 790bc31cf82e..72e9af6158ef 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -1830,9 +1830,7 @@ int cfg80211_get_ies_channel_number(const u8 *ie, size_t ielen,
 			if (!he_6ghz_oper)
 				return -1;
 
-			if (ftype != CFG80211_BSS_FTYPE_BEACON ||
-			    he_6ghz_oper->control & IEEE80211_HE_6GHZ_OPER_CTRL_DUP_BEACON)
-				return he_6ghz_oper->primary;
+			return he_6ghz_oper->primary;
 		}
 	} else if (band == NL80211_BAND_S1GHZ) {
 		tmp = cfg80211_find_elem(WLAN_EID_S1G_OPERATION, ie, ielen);
-- 
cgit v1.2.3-70-g09d2


From fb4b441c5d764813370e381310d9f82b38e7cd3f Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Wed, 1 Mar 2023 12:09:19 +0200
Subject: wifi: wireless: correct primary channel validation on 6 GHz

The check that beacon primary channel is in the range of 80 MHz
(abs < 80) is invalid for 320 MHz since duplicate beacon transmit
means that the AP transmits it on all the 20 MHz sub-channels:

  9.4.2.249 HE Operation element - ... AP transmits Beacon frames in
  non-HT duplicate PPDU with a TXVECTOR parameter CH_BANDWIDTH value
  that is up to the BSS bandwidth.

So in case of 320 MHz the DUP beacon can be in upper 160 for primary
channel in the lower 160 giving possibly an absolute range of over
80 MHz.

Also this check is redundant alltogether, if AP has a wrong primary
channel in the beacon it's a faulty AP, and we would fail in next
steps to connect.

While at it, fix the frequency comparison to no longer compare
between KHz and MHz, which was introduced by commit 7f599aeccbd2
("cfg80211: Use the HE operation IE to determine a 6GHz BSS channel").

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.314faf725255.I5e27251ac558297553b590d3917a7b6d1aae0e74@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/scan.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 72e9af6158ef..c23709d852bc 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -1886,22 +1886,21 @@ cfg80211_get_bss_channel(struct wiphy *wiphy, const u8 *ie, size_t ielen,
 	freq = ieee80211_channel_to_freq_khz(channel_number, channel->band);
 
 	/*
-	 * In 6GHz, duplicated beacon indication is relevant for
-	 * beacons only.
+	 * Frame info (beacon/prob res) is the same as received channel,
+	 * no need for further processing.
 	 */
-	if (channel->band == NL80211_BAND_6GHZ &&
-	    (freq == channel->center_freq ||
-	     abs(freq - channel->center_freq) > 80))
+	if (freq == ieee80211_channel_to_khz(channel))
 		return channel;
 
 	alt_channel = ieee80211_get_channel_khz(wiphy, freq);
 	if (!alt_channel) {
-		if (channel->band == NL80211_BAND_2GHZ) {
+		if (channel->band == NL80211_BAND_2GHZ ||
+		    channel->band == NL80211_BAND_6GHZ) {
 			/*
 			 * Better not allow unexpected channels when that could
 			 * be going beyond the 1-11 range (e.g., discovering
 			 * BSS on channel 12 when radio is configured for
-			 * channel 11.
+			 * channel 11) or beyond the 6 GHz channel range.
 			 */
 			return NULL;
 		}
-- 
cgit v1.2.3-70-g09d2


From 6ff9efcfc2dc256480b252321818e0111b9399a2 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Wed, 1 Mar 2023 12:09:20 +0200
Subject: wifi: wireless: cleanup unused function parameters

In the past ftype was used for deciding about 6G DUP beacon, but the
logic was removed and ftype is not needed anymore.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.98d4761b809b.I255f5ecd77cb24fcf2f1641bb5833ea2d121296e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h |  4 +---
 net/wireless/scan.c    | 21 +++++----------------
 2 files changed, 6 insertions(+), 19 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f115b2550309..03b911abd772 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -6814,13 +6814,11 @@ enum cfg80211_bss_frame_type {
  * @ie: IEs
  * @ielen: length of IEs
  * @band: enum nl80211_band of the channel
- * @ftype: frame type
  *
  * Returns the channel number, or -1 if none could be determined.
  */
 int cfg80211_get_ies_channel_number(const u8 *ie, size_t ielen,
-				    enum nl80211_band band,
-				    enum cfg80211_bss_frame_type ftype);
+				    enum nl80211_band band);
 
 /**
  * cfg80211_inform_bss_data - inform cfg80211 of a new BSS
diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index c23709d852bc..a1382255fab3 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -1810,8 +1810,7 @@ cfg80211_bss_update(struct cfg80211_registered_device *rdev,
 }
 
 int cfg80211_get_ies_channel_number(const u8 *ie, size_t ielen,
-				    enum nl80211_band band,
-				    enum cfg80211_bss_frame_type ftype)
+				    enum nl80211_band band)
 {
 	const struct element *tmp;
 
@@ -1868,15 +1867,14 @@ EXPORT_SYMBOL(cfg80211_get_ies_channel_number);
 static struct ieee80211_channel *
 cfg80211_get_bss_channel(struct wiphy *wiphy, const u8 *ie, size_t ielen,
 			 struct ieee80211_channel *channel,
-			 enum nl80211_bss_scan_width scan_width,
-			 enum cfg80211_bss_frame_type ftype)
+			 enum nl80211_bss_scan_width scan_width)
 {
 	u32 freq;
 	int channel_number;
 	struct ieee80211_channel *alt_channel;
 
 	channel_number = cfg80211_get_ies_channel_number(ie, ielen,
-							 channel->band, ftype);
+							 channel->band);
 
 	if (channel_number < 0) {
 		/* No channel information in frame payload */
@@ -1954,7 +1952,7 @@ cfg80211_inform_single_bss_data(struct wiphy *wiphy,
 		return NULL;
 
 	channel = cfg80211_get_bss_channel(wiphy, ie, ielen, data->chan,
-					   data->scan_width, ftype);
+					   data->scan_width);
 	if (!channel)
 		return NULL;
 
@@ -2388,7 +2386,6 @@ cfg80211_inform_single_bss_frame_data(struct wiphy *wiphy,
 	size_t ielen, min_hdr_len = offsetof(struct ieee80211_mgmt,
 					     u.probe_resp.variable);
 	int bss_type;
-	enum cfg80211_bss_frame_type ftype;
 
 	BUILD_BUG_ON(offsetof(struct ieee80211_mgmt, u.probe_resp.variable) !=
 			offsetof(struct ieee80211_mgmt, u.beacon.variable));
@@ -2425,16 +2422,8 @@ cfg80211_inform_single_bss_frame_data(struct wiphy *wiphy,
 			variable = ext->u.s1g_beacon.variable;
 	}
 
-	if (ieee80211_is_beacon(mgmt->frame_control))
-		ftype = CFG80211_BSS_FTYPE_BEACON;
-	else if (ieee80211_is_probe_resp(mgmt->frame_control))
-		ftype = CFG80211_BSS_FTYPE_PRESP;
-	else
-		ftype = CFG80211_BSS_FTYPE_UNKNOWN;
-
 	channel = cfg80211_get_bss_channel(wiphy, variable,
-					   ielen, data->chan, data->scan_width,
-					   ftype);
+					   ielen, data->chan, data->scan_width);
 	if (!channel)
 		return NULL;
 
-- 
cgit v1.2.3-70-g09d2


From cbbaf2bb829b6c4ef911d4a725fc9b1fadc1e43f Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 1 Mar 2023 12:09:21 +0200
Subject: wifi: nl80211: add a command to enable/disable HW timestamping

Add a command to enable and disable HW timestamping of TM and FTM
frames. HW timestamping can be enabled for a specific mac address
or for all addresses.

The low level driver will indicate how many peers HW timestamping
can be enabled concurrently, and this information will be passed
to userspace.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.05678d7b1c17.Iccc08869ea8156f1c71a3111a47f86dd56234bd0@changeid
[switch to needing netdev UP, minor edits]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h       | 27 +++++++++++++++++++++++++++
 include/uapi/linux/nl80211.h | 22 ++++++++++++++++++++++
 net/wireless/nl80211.c       | 37 +++++++++++++++++++++++++++++++++++++
 net/wireless/rdev-ops.h      | 17 +++++++++++++++++
 net/wireless/trace.h         | 25 +++++++++++++++++++++++++
 5 files changed, 128 insertions(+)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 03b911abd772..f0da61c6ec4b 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -827,6 +827,18 @@ struct cfg80211_fils_aad {
 	const u8 *anonce;
 };
 
+/**
+ * struct cfg80211_set_hw_timestamp - enable/disable HW timestamping
+ * @macaddr: peer MAC address. NULL to enable/disable HW timestamping for all
+ *	addresses.
+ * @enable: if set, enable HW timestamping for the specified MAC address.
+ *	Otherwise disable HW timestamping for the specified MAC address.
+ */
+struct cfg80211_set_hw_timestamp {
+	const u8 *macaddr;
+	bool enable;
+};
+
 /**
  * cfg80211_get_chandef_type - return old channel type from chandef
  * @chandef: the channel definition
@@ -4330,6 +4342,8 @@ struct mgmt_frame_regs {
  * @add_link_station: Add a link to a station.
  * @mod_link_station: Modify a link of a station.
  * @del_link_station: Remove a link of a station.
+ *
+ * @set_hw_timestamp: Enable/disable HW timestamping of TM/FTM frames.
  */
 struct cfg80211_ops {
 	int	(*suspend)(struct wiphy *wiphy, struct cfg80211_wowlan *wow);
@@ -4683,6 +4697,8 @@ struct cfg80211_ops {
 				    struct link_station_parameters *params);
 	int	(*del_link_station)(struct wiphy *wiphy, struct net_device *dev,
 				    struct link_station_del_parameters *params);
+	int	(*set_hw_timestamp)(struct wiphy *wiphy, struct net_device *dev,
+				    struct cfg80211_set_hw_timestamp *hwts);
 };
 
 /*
@@ -5139,6 +5155,8 @@ struct wiphy_iftype_akm_suites {
 	int n_akm_suites;
 };
 
+#define CFG80211_HW_TIMESTAMP_ALL_PEERS	0xffff
+
 /**
  * struct wiphy - wireless hardware description
  * @mtx: mutex for the data (structures) of this device
@@ -5348,6 +5366,13 @@ struct wiphy_iftype_akm_suites {
  *	NL80211_MAX_NR_AKM_SUITES in order to avoid compatibility issues with
  *	legacy userspace and maximum allowed value is
  *	CFG80211_MAX_NUM_AKM_SUITES.
+ *
+ * @hw_timestamp_max_peers: maximum number of peers that the driver supports
+ *	enabling HW timestamping for concurrently. Setting this field to a
+ *	non-zero value indicates that the driver supports HW timestamping.
+ *	A value of %CFG80211_HW_TIMESTAMP_ALL_PEERS indicates the driver
+ *	supports enabling HW timestamping for all peers (i.e. no need to
+ *	specify a mac address).
  */
 struct wiphy {
 	struct mutex mtx;
@@ -5496,6 +5521,8 @@ struct wiphy {
 	u8 ema_max_profile_periodicity;
 	u16 max_num_akm_suites;
 
+	u16 hw_timestamp_max_peers;
+
 	char priv[] __aligned(NETDEV_ALIGN);
 };
 
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index c22eeb18b996..c8520c150f9c 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -1299,6 +1299,16 @@
  * @NL80211_CMD_MODIFY_LINK_STA: Modify a link of an MLD station
  * @NL80211_CMD_REMOVE_LINK_STA: Remove a link of an MLD station
  *
+ * @NL80211_CMD_SET_HW_TIMESTAMP: Enable/disable HW timestamping of Timing
+ *	measurement and Fine timing measurement frames. If %NL80211_ATTR_MAC
+ *	is included, enable/disable HW timestamping only for frames to/from the
+ *	specified MAC address. Otherwise enable/disable HW timestamping for
+ *	all TM/FTM frames (including ones that were enabled with specific MAC
+ *	address). If %NL80211_ATTR_HW_TIMESTAMP_ENABLED is not included, disable
+ *	HW timestamping.
+ *	The number of peers that HW timestamping can be enabled for concurrently
+ *	is indicated by %NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS.
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -1550,6 +1560,8 @@ enum nl80211_commands {
 	NL80211_CMD_MODIFY_LINK_STA,
 	NL80211_CMD_REMOVE_LINK_STA,
 
+	NL80211_CMD_SET_HW_TIMESTAMP,
+
 	/* add new commands above here */
 
 	/* used to define NL80211_CMD_MAX below */
@@ -2775,6 +2787,13 @@ enum nl80211_commands {
  *	indicates that the sub-channel is punctured. Higher 16 bits are
  *	reserved.
  *
+ * @NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS: Maximum number of peers that HW
+ *	timestamping can be enabled for concurrently (u16), a wiphy attribute.
+ *	A value of 0xffff indicates setting for all peers (i.e. not specifying
+ *	an address with %NL80211_CMD_SET_HW_TIMESTAMP) is supported.
+ * @NL80211_ATTR_HW_TIMESTAMP_ENABLED: Indicates whether HW timestamping should
+ *	be enabled or not (flag attribute).
+ *
  * @NUM_NL80211_ATTR: total number of nl80211_attrs available
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
@@ -3306,6 +3325,9 @@ enum nl80211_attrs {
 
 	NL80211_ATTR_PUNCT_BITMAP,
 
+	NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS,
+	NL80211_ATTR_HW_TIMESTAMP_ENABLED,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 112b4bb009c8..ab0497efdd37 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -806,6 +806,9 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = {
 	[NL80211_ATTR_MLO_SUPPORT] = { .type = NLA_FLAG },
 	[NL80211_ATTR_MAX_NUM_AKM_SUITES] = { .type = NLA_REJECT },
 	[NL80211_ATTR_PUNCT_BITMAP] = NLA_POLICY_RANGE(NLA_U8, 0, 0xffff),
+
+	[NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS] = { .type = NLA_U16 },
+	[NL80211_ATTR_HW_TIMESTAMP_ENABLED] = { .type = NLA_FLAG },
 };
 
 /* policy for the key attributes */
@@ -2964,6 +2967,11 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev,
 		if (rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_MLO)
 			nla_put_flag(msg, NL80211_ATTR_MLO_SUPPORT);
 
+		if (rdev->wiphy.hw_timestamp_max_peers &&
+		    nla_put_u16(msg, NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS,
+				rdev->wiphy.hw_timestamp_max_peers))
+			goto nla_put_failure;
+
 		/* done */
 		state->split_start = 0;
 		break;
@@ -16162,6 +16170,29 @@ nl80211_remove_link_station(struct sk_buff *skb, struct genl_info *info)
 	return ret;
 }
 
+static int nl80211_set_hw_timestamp(struct sk_buff *skb,
+				    struct genl_info *info)
+{
+	struct cfg80211_registered_device *rdev = info->user_ptr[0];
+	struct net_device *dev = info->user_ptr[1];
+	struct cfg80211_set_hw_timestamp hwts = {};
+
+	if (!rdev->wiphy.hw_timestamp_max_peers)
+		return -EOPNOTSUPP;
+
+	if (!info->attrs[NL80211_ATTR_MAC] &&
+	    rdev->wiphy.hw_timestamp_max_peers != CFG80211_HW_TIMESTAMP_ALL_PEERS)
+		return -EOPNOTSUPP;
+
+	if (info->attrs[NL80211_ATTR_MAC])
+		hwts.macaddr = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	hwts.enable =
+		nla_get_flag(info->attrs[NL80211_ATTR_HW_TIMESTAMP_ENABLED]);
+
+	return rdev_set_hw_timestamp(rdev, dev, &hwts);
+}
+
 #define NL80211_FLAG_NEED_WIPHY		0x01
 #define NL80211_FLAG_NEED_NETDEV	0x02
 #define NL80211_FLAG_NEED_RTNL		0x04
@@ -17336,6 +17367,12 @@ static const struct genl_small_ops nl80211_small_ops[] = {
 		.internal_flags = IFLAGS(NL80211_FLAG_NEED_NETDEV_UP |
 					 NL80211_FLAG_MLO_VALID_LINK_ID),
 	},
+	{
+		.cmd = NL80211_CMD_SET_HW_TIMESTAMP,
+		.doit = nl80211_set_hw_timestamp,
+		.flags = GENL_UNS_ADMIN_PERM,
+		.internal_flags = IFLAGS(NL80211_FLAG_NEED_NETDEV_UP),
+	},
 };
 
 static struct genl_family nl80211_fam __ro_after_init = {
diff --git a/net/wireless/rdev-ops.h b/net/wireless/rdev-ops.h
index 13b209a8db28..2e497cf26ef2 100644
--- a/net/wireless/rdev-ops.h
+++ b/net/wireless/rdev-ops.h
@@ -1494,4 +1494,21 @@ rdev_del_link_station(struct cfg80211_registered_device *rdev,
 	return ret;
 }
 
+static inline int
+rdev_set_hw_timestamp(struct cfg80211_registered_device *rdev,
+		      struct net_device *dev,
+		      struct cfg80211_set_hw_timestamp *hwts)
+{
+	struct wiphy *wiphy = &rdev->wiphy;
+	int ret;
+
+	if (!rdev->ops->set_hw_timestamp)
+		return -EOPNOTSUPP;
+
+	trace_rdev_set_hw_timestamp(wiphy, dev, hwts);
+	ret = rdev->ops->set_hw_timestamp(wiphy, dev, hwts);
+	trace_rdev_return_int(wiphy, ret);
+
+	return ret;
+}
 #endif /* __CFG80211_RDEV_OPS */
diff --git a/net/wireless/trace.h b/net/wireless/trace.h
index ca7474eec723..f3fcfc4fcce5 100644
--- a/net/wireless/trace.h
+++ b/net/wireless/trace.h
@@ -3918,6 +3918,31 @@ TRACE_EVENT(rdev_del_link_station,
 		  __entry->link_id)
 );
 
+TRACE_EVENT(rdev_set_hw_timestamp,
+	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
+		 struct cfg80211_set_hw_timestamp *hwts),
+
+	TP_ARGS(wiphy, netdev, hwts),
+
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		NETDEV_ENTRY
+		MAC_ENTRY(macaddr)
+		__field(bool, enable)
+	),
+
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		NETDEV_ASSIGN;
+		MAC_ASSIGN(macaddr, hwts->macaddr);
+		__entry->enable = hwts->enable;
+	),
+
+	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", mac %pM, enable: %u",
+		  WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->macaddr,
+		  __entry->enable)
+);
+
 #endif /* !__RDEV_OPS_TRACE || TRACE_HEADER_MULTI_READ */
 
 #undef TRACE_INCLUDE_PATH
-- 
cgit v1.2.3-70-g09d2


From 81202305f7c282c356c337dded8472d884acd94b Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 1 Mar 2023 12:09:22 +0200
Subject: wifi: mac80211: add support for set_hw_timestamp command

Support the set_hw_timestamp callback for enabling and disabling HW
timestamping if the low level driver supports it.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.700ded7badde.Ib2f7c228256ce313a04d3d9f9ecc6c7b9aa602bb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h |  6 ++++++
 net/mac80211/cfg.c     | 17 +++++++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 219fd15893b0..6946c9d95aec 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -4230,6 +4230,9 @@ struct ieee80211_prep_tx_info {
  *	Note that a sta can also be inserted or removed with valid links,
  *	i.e. passed to @sta_add/@sta_state with sta->valid_links not zero.
  *	In fact, cannot change from having valid_links and not having them.
+ * @set_hw_timestamp: Enable/disable HW timestamping of TM/FTM frames. This is
+ *	not restored at HW reset by mac80211 so drivers need to take care of
+ *	that.
  */
 struct ieee80211_ops {
 	void (*tx)(struct ieee80211_hw *hw,
@@ -4589,6 +4592,9 @@ struct ieee80211_ops {
 				struct ieee80211_vif *vif,
 				struct ieee80211_sta *sta,
 				u16 old_links, u16 new_links);
+	int (*set_hw_timestamp)(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct cfg80211_set_hw_timestamp *hwts);
 };
 
 /**
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 8eb342300868..7e90f4a81962 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -4904,6 +4904,22 @@ ieee80211_del_link_station(struct wiphy *wiphy, struct net_device *dev,
 	return ret;
 }
 
+static int ieee80211_set_hw_timestamp(struct wiphy *wiphy,
+				      struct net_device *dev,
+				      struct cfg80211_set_hw_timestamp *hwts)
+{
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_local *local = sdata->local;
+
+	if (!local->ops->set_hw_timestamp)
+		return -EOPNOTSUPP;
+
+	if (!check_sdata_in_driver(sdata))
+		return -EIO;
+
+	return local->ops->set_hw_timestamp(&local->hw, &sdata->vif, hwts);
+}
+
 const struct cfg80211_ops mac80211_config_ops = {
 	.add_virtual_intf = ieee80211_add_iface,
 	.del_virtual_intf = ieee80211_del_iface,
@@ -5014,4 +5030,5 @@ const struct cfg80211_ops mac80211_config_ops = {
 	.add_link_station = ieee80211_add_link_station,
 	.mod_link_station = ieee80211_mod_link_station,
 	.del_link_station = ieee80211_del_link_station,
+	.set_hw_timestamp = ieee80211_set_hw_timestamp,
 };
-- 
cgit v1.2.3-70-g09d2


From 4c532321bf90288dae6b07a3f52279bfde842a80 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:23 +0200
Subject: wifi: cfg80211/mac80211: report link ID on control port RX

For control port RX, report the link ID for MLO.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.fe06dfc3791b.Iddcab94789cafe336417be406072ce8a6312fc2d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h |  5 +++--
 net/mac80211/rx.c      |  2 +-
 net/wireless/nl80211.c | 15 ++++++++++-----
 net/wireless/trace.h   | 11 +++++++----
 4 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f0da61c6ec4b..7cebba1c4135 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -8126,6 +8126,7 @@ void cfg80211_control_port_tx_status(struct wireless_dev *wdev, u64 cookie,
  *	responsible for any cleanup.  The caller must also ensure that
  *	skb->protocol is set appropriately.
  * @unencrypted: Whether the frame was received unencrypted
+ * @link_id: the link the frame was received on, -1 if not applicable or unknown
  *
  * This function is used to inform userspace about a received control port
  * frame.  It should only be used if userspace indicated it wants to receive
@@ -8136,8 +8137,8 @@ void cfg80211_control_port_tx_status(struct wireless_dev *wdev, u64 cookie,
  *
  * Return: %true if the frame was passed to userspace
  */
-bool cfg80211_rx_control_port(struct net_device *dev,
-			      struct sk_buff *skb, bool unencrypted);
+bool cfg80211_rx_control_port(struct net_device *dev, struct sk_buff *skb,
+			      bool unencrypted, int link_id);
 
 /**
  * cfg80211_cqm_rssi_notify - connection quality monitoring rssi event
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 095bcd2552bb..f63ed6b91d9b 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2591,7 +2591,7 @@ static void ieee80211_deliver_skb_to_local_stack(struct sk_buff *skb,
 		struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb);
 		bool noencrypt = !(status->flag & RX_FLAG_DECRYPTED);
 
-		cfg80211_rx_control_port(dev, skb, noencrypt);
+		cfg80211_rx_control_port(dev, skb, noencrypt, rx->link_id);
 		dev_kfree_skb(skb);
 	} else {
 		struct ethhdr *ehdr = (void *)skb_mac_header(skb);
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index ab0497efdd37..85f714e1af87 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -18754,7 +18754,9 @@ EXPORT_SYMBOL(cfg80211_mgmt_tx_status_ext);
 
 static int __nl80211_rx_control_port(struct net_device *dev,
 				     struct sk_buff *skb,
-				     bool unencrypted, gfp_t gfp)
+				     bool unencrypted,
+				     int link_id,
+				     gfp_t gfp)
 {
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
 	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
@@ -18786,6 +18788,8 @@ static int __nl80211_rx_control_port(struct net_device *dev,
 			      NL80211_ATTR_PAD) ||
 	    nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, addr) ||
 	    nla_put_u16(msg, NL80211_ATTR_CONTROL_PORT_ETHERTYPE, proto) ||
+	    (link_id >= 0 &&
+	     nla_put_u8(msg, NL80211_ATTR_MLO_LINK_ID, link_id)) ||
 	    (unencrypted && nla_put_flag(msg,
 					 NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT)))
 		goto nla_put_failure;
@@ -18804,13 +18808,14 @@ static int __nl80211_rx_control_port(struct net_device *dev,
 	return -ENOBUFS;
 }
 
-bool cfg80211_rx_control_port(struct net_device *dev,
-			      struct sk_buff *skb, bool unencrypted)
+bool cfg80211_rx_control_port(struct net_device *dev, struct sk_buff *skb,
+			      bool unencrypted, int link_id)
 {
 	int ret;
 
-	trace_cfg80211_rx_control_port(dev, skb, unencrypted);
-	ret = __nl80211_rx_control_port(dev, skb, unencrypted, GFP_ATOMIC);
+	trace_cfg80211_rx_control_port(dev, skb, unencrypted, link_id);
+	ret = __nl80211_rx_control_port(dev, skb, unencrypted, link_id,
+					GFP_ATOMIC);
 	trace_cfg80211_return_bool(ret == 0);
 	return ret == 0;
 }
diff --git a/net/wireless/trace.h b/net/wireless/trace.h
index f3fcfc4fcce5..716a1fa70069 100644
--- a/net/wireless/trace.h
+++ b/net/wireless/trace.h
@@ -3165,14 +3165,15 @@ TRACE_EVENT(cfg80211_control_port_tx_status,
 
 TRACE_EVENT(cfg80211_rx_control_port,
 	TP_PROTO(struct net_device *netdev, struct sk_buff *skb,
-		 bool unencrypted),
-	TP_ARGS(netdev, skb, unencrypted),
+		 bool unencrypted, int link_id),
+	TP_ARGS(netdev, skb, unencrypted, link_id),
 	TP_STRUCT__entry(
 		NETDEV_ENTRY
 		__field(int, len)
 		MAC_ENTRY(from)
 		__field(u16, proto)
 		__field(bool, unencrypted)
+		__field(int, link_id)
 	),
 	TP_fast_assign(
 		NETDEV_ASSIGN;
@@ -3180,10 +3181,12 @@ TRACE_EVENT(cfg80211_rx_control_port,
 		MAC_ASSIGN(from, eth_hdr(skb)->h_source);
 		__entry->proto = be16_to_cpu(skb->protocol);
 		__entry->unencrypted = unencrypted;
+		__entry->link_id = link_id;
 	),
-	TP_printk(NETDEV_PR_FMT ", len=%d, %pM, proto: 0x%x, unencrypted: %s",
+	TP_printk(NETDEV_PR_FMT ", len=%d, %pM, proto: 0x%x, unencrypted: %s, link: %d",
 		  NETDEV_PR_ARG, __entry->len, __entry->from,
-		  __entry->proto, BOOL_TO_STR(__entry->unencrypted))
+		  __entry->proto, BOOL_TO_STR(__entry->unencrypted),
+		  __entry->link_id)
 );
 
 TRACE_EVENT(cfg80211_cqm_rssi_notify,
-- 
cgit v1.2.3-70-g09d2


From a1e91ef92392e5da15a1a16f8545ede2c02f7049 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:24 +0200
Subject: wifi: mac80211: warn only once on AP probe

We should perhaps support this API for MLO, but it's not
clear that it makes sense, in any case then we'd have to
update it to probe the correct BSS.

For now, if it happens, warn only once so that we don't
get flooded with messages if the driver misbehaves and
calls this.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.1c8499b6fbe6.I1a76a2be3b42ff93904870ac069f0319507adc23@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 60792dfabc9d..cdf842c0849b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3227,7 +3227,7 @@ static void ieee80211_mgd_probe_ap(struct ieee80211_sub_if_data *sdata,
 	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
 	bool already = false;
 
-	if (WARN_ON(sdata->vif.valid_links))
+	if (WARN_ON_ONCE(sdata->vif.valid_links))
 		return;
 
 	if (!ieee80211_sdata_running(sdata))
-- 
cgit v1.2.3-70-g09d2


From e1f113cc67870375eae0c7b84c2a40cc6388d903 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:25 +0200
Subject: wifi: mac80211: add pointer from bss_conf to vif

While often not needed, this considerably simplifies going from a link
specific bss_config to the vif. This helps with e.g. creating link
specific debugfs entries inside drivers.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.46f701a10ed5.I20390b2a8165ff222d66585915689206ea93222b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h | 3 +++
 net/mac80211/link.c    | 1 +
 2 files changed, 4 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 6946c9d95aec..2a1874d994e0 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -534,6 +534,7 @@ struct ieee80211_fils_discovery {
  * This structure keeps information about a BSS (and an association
  * to that BSS) that can change during the lifetime of the BSS.
  *
+ * @vif: reference to owning VIF
  * @addr: (link) address used locally
  * @link_id: link ID, or 0 for non-MLO
  * @htc_trig_based_pkt_ext: default PE in 4us units, if BSS supports HE
@@ -675,6 +676,8 @@ struct ieee80211_fils_discovery {
  *	bandwidth
  */
 struct ieee80211_bss_conf {
+	struct ieee80211_vif *vif;
+
 	const u8 *bssid;
 	unsigned int link_id;
 	u8 addr[ETH_ALEN] __aligned(2);
diff --git a/net/mac80211/link.c b/net/mac80211/link.c
index 8c8869cc1fb4..e6f9fce1dadb 100644
--- a/net/mac80211/link.c
+++ b/net/mac80211/link.c
@@ -34,6 +34,7 @@ void ieee80211_link_init(struct ieee80211_sub_if_data *sdata,
 	link->link_id = link_id;
 	link->conf = link_conf;
 	link_conf->link_id = link_id;
+	link_conf->vif = &sdata->vif;
 
 	INIT_WORK(&link->csa_finalize_work,
 		  ieee80211_csa_finalize_work);
-- 
cgit v1.2.3-70-g09d2


From 586100ad85fed0041e203d6006d45e0ec4c5a0b4 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:26 +0200
Subject: wifi: mac80211: remove SMPS from AP debugfs

The spatial multiplexing power save feature does not apply to
AP mode. Remove it from debugfs in this case.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.01b167027dd5.Iee69f2e4df98581f259ef2c76309b940b20174be@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/debugfs_netdev.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c
index 0bac9af3ca96..187bb22c0bbd 100644
--- a/net/mac80211/debugfs_netdev.c
+++ b/net/mac80211/debugfs_netdev.c
@@ -698,7 +698,6 @@ static void add_sta_files(struct ieee80211_sub_if_data *sdata)
 static void add_ap_files(struct ieee80211_sub_if_data *sdata)
 {
 	DEBUGFS_ADD(num_mcast_sta);
-	DEBUGFS_ADD_MODE(smps, 0600);
 	DEBUGFS_ADD(num_sta_ps);
 	DEBUGFS_ADD(dtim_count);
 	DEBUGFS_ADD(num_buffered_multicast);
-- 
cgit v1.2.3-70-g09d2


From 170cd6a66d9a164180eb4dc72d50afa6ce1ce566 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:27 +0200
Subject: wifi: mac80211: add netdev per-link debugfs data and driver hook

This adds the infrastructure to have netdev specific per-link data both
for mac80211 and the driver in debugfs. For the driver, a new callback
is added which is only used if MLO is supported.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.fb4c947e4df8.I69b3516ddf4c8a7501b395f652d6063444ecad63@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h        |  10 ++
 net/mac80211/debugfs_netdev.c | 222 ++++++++++++++++++++++++++++++++----------
 net/mac80211/debugfs_netdev.h |  16 +++
 net/mac80211/driver-ops.c     |  25 ++++-
 net/mac80211/driver-ops.h     |  16 +++
 net/mac80211/ieee80211_i.h    |   4 +
 net/mac80211/link.c           |   4 +
 7 files changed, 242 insertions(+), 55 deletions(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 2a1874d994e0..5df9eb828a58 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -3844,6 +3844,12 @@ struct ieee80211_prep_tx_info {
  *	the station. See @sta_pre_rcu_remove if needed.
  *	This callback can sleep.
  *
+ * @link_add_debugfs: Drivers can use this callback to add debugfs files
+ *	when a link is added to a mac80211 vif. This callback should be within
+ *	a CONFIG_MAC80211_DEBUGFS conditional. This callback can sleep.
+ *	For non-MLO the callback will be called once for the default bss_conf
+ *	with the vif's directory rather than a separate subdirectory.
+ *
  * @sta_add_debugfs: Drivers can use this callback to add debugfs files
  *	when a station is added to mac80211's station list. This callback
  *	should be within a CONFIG_MAC80211_DEBUGFS conditional. This
@@ -4325,6 +4331,10 @@ struct ieee80211_ops {
 	int (*sta_remove)(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			  struct ieee80211_sta *sta);
 #ifdef CONFIG_MAC80211_DEBUGFS
+	void (*link_add_debugfs)(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_bss_conf *link_conf,
+				 struct dentry *dir);
 	void (*sta_add_debugfs)(struct ieee80211_hw *hw,
 				struct ieee80211_vif *vif,
 				struct ieee80211_sta *sta,
diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c
index 187bb22c0bbd..b0cef37eb394 100644
--- a/net/mac80211/debugfs_netdev.c
+++ b/net/mac80211/debugfs_netdev.c
@@ -23,16 +23,16 @@
 #include "driver-ops.h"
 
 static ssize_t ieee80211_if_read(
-	struct ieee80211_sub_if_data *sdata,
+	void *data,
 	char __user *userbuf,
 	size_t count, loff_t *ppos,
-	ssize_t (*format)(const struct ieee80211_sub_if_data *, char *, int))
+	ssize_t (*format)(const void *, char *, int))
 {
 	char buf[200];
 	ssize_t ret = -EINVAL;
 
 	read_lock(&dev_base_lock);
-	ret = (*format)(sdata, buf, sizeof(buf));
+	ret = (*format)(data, buf, sizeof(buf));
 	read_unlock(&dev_base_lock);
 
 	if (ret >= 0)
@@ -42,10 +42,10 @@ static ssize_t ieee80211_if_read(
 }
 
 static ssize_t ieee80211_if_write(
-	struct ieee80211_sub_if_data *sdata,
+	void *data,
 	const char __user *userbuf,
 	size_t count, loff_t *ppos,
-	ssize_t (*write)(struct ieee80211_sub_if_data *, const char *, int))
+	ssize_t (*write)(void *, const char *, int))
 {
 	char buf[64];
 	ssize_t ret;
@@ -58,64 +58,64 @@ static ssize_t ieee80211_if_write(
 	buf[count] = '\0';
 
 	rtnl_lock();
-	ret = (*write)(sdata, buf, count);
+	ret = (*write)(data, buf, count);
 	rtnl_unlock();
 
 	return ret;
 }
 
-#define IEEE80211_IF_FMT(name, field, format_string)			\
+#define IEEE80211_IF_FMT(name, type, field, format_string)		\
 static ssize_t ieee80211_if_fmt_##name(					\
-	const struct ieee80211_sub_if_data *sdata, char *buf,		\
+	const type *data, char *buf,					\
 	int buflen)							\
 {									\
-	return scnprintf(buf, buflen, format_string, sdata->field);	\
+	return scnprintf(buf, buflen, format_string, data->field);	\
 }
-#define IEEE80211_IF_FMT_DEC(name, field)				\
-		IEEE80211_IF_FMT(name, field, "%d\n")
-#define IEEE80211_IF_FMT_HEX(name, field)				\
-		IEEE80211_IF_FMT(name, field, "%#x\n")
-#define IEEE80211_IF_FMT_LHEX(name, field)				\
-		IEEE80211_IF_FMT(name, field, "%#lx\n")
+#define IEEE80211_IF_FMT_DEC(name, type, field)				\
+		IEEE80211_IF_FMT(name, type, field, "%d\n")
+#define IEEE80211_IF_FMT_HEX(name, type, field)				\
+		IEEE80211_IF_FMT(name, type, field, "%#x\n")
+#define IEEE80211_IF_FMT_LHEX(name, type, field)			\
+		IEEE80211_IF_FMT(name, type, field, "%#lx\n")
 
-#define IEEE80211_IF_FMT_HEXARRAY(name, field)				\
+#define IEEE80211_IF_FMT_HEXARRAY(name, type, field)			\
 static ssize_t ieee80211_if_fmt_##name(					\
-	const struct ieee80211_sub_if_data *sdata,			\
+	const type *data,						\
 	char *buf, int buflen)						\
 {									\
 	char *p = buf;							\
 	int i;								\
-	for (i = 0; i < sizeof(sdata->field); i++) {			\
+	for (i = 0; i < sizeof(data->field); i++) {			\
 		p += scnprintf(p, buflen + buf - p, "%.2x ",		\
-				 sdata->field[i]);			\
+				 data->field[i]);			\
 	}								\
 	p += scnprintf(p, buflen + buf - p, "\n");			\
 	return p - buf;							\
 }
 
-#define IEEE80211_IF_FMT_ATOMIC(name, field)				\
+#define IEEE80211_IF_FMT_ATOMIC(name, type, field)			\
 static ssize_t ieee80211_if_fmt_##name(					\
-	const struct ieee80211_sub_if_data *sdata,			\
+	const type *data,						\
 	char *buf, int buflen)						\
 {									\
-	return scnprintf(buf, buflen, "%d\n", atomic_read(&sdata->field));\
+	return scnprintf(buf, buflen, "%d\n", atomic_read(&data->field));\
 }
 
-#define IEEE80211_IF_FMT_MAC(name, field)				\
+#define IEEE80211_IF_FMT_MAC(name, type, field)				\
 static ssize_t ieee80211_if_fmt_##name(					\
-	const struct ieee80211_sub_if_data *sdata, char *buf,		\
+	const type *data, char *buf,					\
 	int buflen)							\
 {									\
-	return scnprintf(buf, buflen, "%pM\n", sdata->field);		\
+	return scnprintf(buf, buflen, "%pM\n", data->field);		\
 }
 
-#define IEEE80211_IF_FMT_JIFFIES_TO_MS(name, field)			\
+#define IEEE80211_IF_FMT_JIFFIES_TO_MS(name, type, field)		\
 static ssize_t ieee80211_if_fmt_##name(					\
-	const struct ieee80211_sub_if_data *sdata,			\
+	const type *data,						\
 	char *buf, int buflen)						\
 {									\
 	return scnprintf(buf, buflen, "%d\n",				\
-			 jiffies_to_msecs(sdata->field));		\
+			 jiffies_to_msecs(data->field));		\
 }
 
 #define _IEEE80211_IF_FILE_OPS(name, _read, _write)			\
@@ -126,43 +126,67 @@ static const struct file_operations name##_ops = {			\
 	.llseek = generic_file_llseek,					\
 }
 
-#define _IEEE80211_IF_FILE_R_FN(name)					\
+#define _IEEE80211_IF_FILE_R_FN(name, type)				\
 static ssize_t ieee80211_if_read_##name(struct file *file,		\
 					char __user *userbuf,		\
 					size_t count, loff_t *ppos)	\
 {									\
+	ssize_t (*fn)(const void *, char *, int) = (void *)		\
+		((ssize_t (*)(const type, char *, int))			\
+		 ieee80211_if_fmt_##name);				\
 	return ieee80211_if_read(file->private_data,			\
-				 userbuf, count, ppos,			\
-				 ieee80211_if_fmt_##name);		\
+				 userbuf, count, ppos, fn);		\
 }
 
-#define _IEEE80211_IF_FILE_W_FN(name)					\
+#define _IEEE80211_IF_FILE_W_FN(name, type)				\
 static ssize_t ieee80211_if_write_##name(struct file *file,		\
 					 const char __user *userbuf,	\
 					 size_t count, loff_t *ppos)	\
 {									\
+	ssize_t (*fn)(void *, const char *, int) = (void *)		\
+		((ssize_t (*)(type, const char *, int))			\
+		 ieee80211_if_parse_##name);				\
 	return ieee80211_if_write(file->private_data, userbuf, count,	\
-				  ppos, ieee80211_if_parse_##name);	\
+				  ppos, fn);				\
 }
 
 #define IEEE80211_IF_FILE_R(name)					\
-	_IEEE80211_IF_FILE_R_FN(name)					\
+	_IEEE80211_IF_FILE_R_FN(name, struct ieee80211_sub_if_data *)	\
 	_IEEE80211_IF_FILE_OPS(name, ieee80211_if_read_##name, NULL)
 
 #define IEEE80211_IF_FILE_W(name)					\
-	_IEEE80211_IF_FILE_W_FN(name)					\
+	_IEEE80211_IF_FILE_W_FN(name, struct ieee80211_sub_if_data *)	\
 	_IEEE80211_IF_FILE_OPS(name, NULL, ieee80211_if_write_##name)
 
 #define IEEE80211_IF_FILE_RW(name)					\
-	_IEEE80211_IF_FILE_R_FN(name)					\
-	_IEEE80211_IF_FILE_W_FN(name)					\
+	_IEEE80211_IF_FILE_R_FN(name, struct ieee80211_sub_if_data *)	\
+	_IEEE80211_IF_FILE_W_FN(name, struct ieee80211_sub_if_data *)	\
 	_IEEE80211_IF_FILE_OPS(name, ieee80211_if_read_##name,		\
 			       ieee80211_if_write_##name)
 
 #define IEEE80211_IF_FILE(name, field, format)				\
-	IEEE80211_IF_FMT_##format(name, field)				\
+	IEEE80211_IF_FMT_##format(name, struct ieee80211_sub_if_data, field) \
 	IEEE80211_IF_FILE_R(name)
 
+/* Same but with a link_ prefix in the ops variable name and different type */
+#define IEEE80211_IF_LINK_FILE_R(name)					\
+	_IEEE80211_IF_FILE_R_FN(name, struct ieee80211_link_data *)	\
+	_IEEE80211_IF_FILE_OPS(link_##name, ieee80211_if_read_##name, NULL)
+
+#define IEEE80211_IF_LINK_FILE_W(name)					\
+	_IEEE80211_IF_FILE_W_FN(name)					\
+	_IEEE80211_IF_FILE_OPS(link_##name, NULL, ieee80211_if_write_##name)
+
+#define IEEE80211_IF_LINK_FILE_RW(name)					\
+	_IEEE80211_IF_FILE_R_FN(name, struct ieee80211_link_data *)	\
+	_IEEE80211_IF_FILE_W_FN(name, struct ieee80211_link_data *)	\
+	_IEEE80211_IF_FILE_OPS(link_##name, ieee80211_if_read_##name,	\
+			       ieee80211_if_write_##name)
+
+#define IEEE80211_IF_LINK_FILE(name, field, format)				\
+	IEEE80211_IF_FMT_##format(name, struct ieee80211_link_data, field) \
+	IEEE80211_IF_LINK_FILE_R(name)
+
 /* common attributes */
 IEEE80211_IF_FILE(rc_rateidx_mask_2ghz, rc_rateidx_mask[NL80211_BAND_2GHZ],
 		  HEX);
@@ -207,9 +231,9 @@ IEEE80211_IF_FILE_R(rc_rateidx_vht_mcs_mask_5ghz);
 
 IEEE80211_IF_FILE(flags, flags, HEX);
 IEEE80211_IF_FILE(state, state, LHEX);
-IEEE80211_IF_FILE(txpower, vif.bss_conf.txpower, DEC);
-IEEE80211_IF_FILE(ap_power_level, deflink.ap_power_level, DEC);
-IEEE80211_IF_FILE(user_power_level, deflink.user_power_level, DEC);
+IEEE80211_IF_LINK_FILE(txpower, conf->txpower, DEC);
+IEEE80211_IF_LINK_FILE(ap_power_level, ap_power_level, DEC);
+IEEE80211_IF_LINK_FILE(user_power_level, user_power_level, DEC);
 
 static ssize_t
 ieee80211_if_fmt_hw_queues(const struct ieee80211_sub_if_data *sdata,
@@ -236,9 +260,10 @@ IEEE80211_IF_FILE(bssid, deflink.u.mgd.bssid, MAC);
 IEEE80211_IF_FILE(aid, vif.cfg.aid, DEC);
 IEEE80211_IF_FILE(beacon_timeout, u.mgd.beacon_timeout, JIFFIES_TO_MS);
 
-static int ieee80211_set_smps(struct ieee80211_sub_if_data *sdata,
+static int ieee80211_set_smps(struct ieee80211_link_data *link,
 			      enum ieee80211_smps_mode smps_mode)
 {
+	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_local *local = sdata->local;
 	int err;
 
@@ -256,7 +281,7 @@ static int ieee80211_set_smps(struct ieee80211_sub_if_data *sdata,
 		return -EOPNOTSUPP;
 
 	sdata_lock(sdata);
-	err = __ieee80211_request_smps_mgd(sdata, &sdata->deflink, smps_mode);
+	err = __ieee80211_request_smps_mgd(link->sdata, link, smps_mode);
 	sdata_unlock(sdata);
 
 	return err;
@@ -269,24 +294,24 @@ static const char *smps_modes[IEEE80211_SMPS_NUM_MODES] = {
 	[IEEE80211_SMPS_DYNAMIC] = "dynamic",
 };
 
-static ssize_t ieee80211_if_fmt_smps(const struct ieee80211_sub_if_data *sdata,
+static ssize_t ieee80211_if_fmt_smps(const struct ieee80211_link_data *link,
 				     char *buf, int buflen)
 {
-	if (sdata->vif.type == NL80211_IFTYPE_STATION)
+	if (link->sdata->vif.type == NL80211_IFTYPE_STATION)
 		return snprintf(buf, buflen, "request: %s\nused: %s\n",
-				smps_modes[sdata->deflink.u.mgd.req_smps],
-				smps_modes[sdata->deflink.smps_mode]);
+				smps_modes[link->u.mgd.req_smps],
+				smps_modes[link->smps_mode]);
 	return -EINVAL;
 }
 
-static ssize_t ieee80211_if_parse_smps(struct ieee80211_sub_if_data *sdata,
+static ssize_t ieee80211_if_parse_smps(struct ieee80211_link_data *link,
 				       const char *buf, int buflen)
 {
 	enum ieee80211_smps_mode mode;
 
 	for (mode = 0; mode < IEEE80211_SMPS_NUM_MODES; mode++) {
 		if (strncmp(buf, smps_modes[mode], buflen) == 0) {
-			int err = ieee80211_set_smps(sdata, mode);
+			int err = ieee80211_set_smps(link, mode);
 			if (!err)
 				return buflen;
 			return err;
@@ -295,7 +320,7 @@ static ssize_t ieee80211_if_parse_smps(struct ieee80211_sub_if_data *sdata,
 
 	return -EINVAL;
 }
-IEEE80211_IF_FILE_RW(smps);
+IEEE80211_IF_LINK_FILE_RW(smps);
 
 static ssize_t ieee80211_if_parse_tkip_mic_test(
 	struct ieee80211_sub_if_data *sdata, const char *buf, int buflen)
@@ -595,6 +620,8 @@ static ssize_t ieee80211_if_parse_active_links(struct ieee80211_sub_if_data *sda
 }
 IEEE80211_IF_FILE_RW(active_links);
 
+IEEE80211_IF_LINK_FILE(addr, conf->addr, MAC);
+
 #ifdef CONFIG_MAC80211_MESH
 IEEE80211_IF_FILE(estab_plinks, u.mesh.estab_plinks, ATOMIC);
 
@@ -685,7 +712,6 @@ static void add_sta_files(struct ieee80211_sub_if_data *sdata)
 	DEBUGFS_ADD(bssid);
 	DEBUGFS_ADD(aid);
 	DEBUGFS_ADD(beacon_timeout);
-	DEBUGFS_ADD_MODE(smps, 0600);
 	DEBUGFS_ADD_MODE(tkip_mic_test, 0200);
 	DEBUGFS_ADD_MODE(beacon_loss, 0200);
 	DEBUGFS_ADD_MODE(uapsd_queues, 0600);
@@ -788,9 +814,6 @@ static void add_files(struct ieee80211_sub_if_data *sdata)
 
 	DEBUGFS_ADD(flags);
 	DEBUGFS_ADD(state);
-	DEBUGFS_ADD(txpower);
-	DEBUGFS_ADD(user_power_level);
-	DEBUGFS_ADD(ap_power_level);
 
 	if (sdata->vif.type != NL80211_IFTYPE_MONITOR)
 		add_common_files(sdata);
@@ -820,6 +843,31 @@ static void add_files(struct ieee80211_sub_if_data *sdata)
 	}
 }
 
+#undef DEBUGFS_ADD_MODE
+#undef DEBUGFS_ADD
+
+#define DEBUGFS_ADD_MODE(dentry, name, mode) \
+	debugfs_create_file(#name, mode, dentry, \
+			    link, &link_##name##_ops)
+
+#define DEBUGFS_ADD(dentry, name) DEBUGFS_ADD_MODE(dentry, name, 0400)
+
+static void add_link_files(struct ieee80211_link_data *link,
+			   struct dentry *dentry)
+{
+	DEBUGFS_ADD(dentry, txpower);
+	DEBUGFS_ADD(dentry, user_power_level);
+	DEBUGFS_ADD(dentry, ap_power_level);
+
+	switch (link->sdata->vif.type) {
+	case NL80211_IFTYPE_STATION:
+		DEBUGFS_ADD_MODE(dentry, smps, 0600);
+		break;
+	default:
+		break;
+	}
+}
+
 void ieee80211_debugfs_add_netdev(struct ieee80211_sub_if_data *sdata)
 {
 	char buf[10+IFNAMSIZ];
@@ -830,6 +878,9 @@ void ieee80211_debugfs_add_netdev(struct ieee80211_sub_if_data *sdata)
 	sdata->debugfs.subdir_stations = debugfs_create_dir("stations",
 							sdata->vif.debugfs_dir);
 	add_files(sdata);
+
+	if (!(sdata->local->hw.wiphy->flags & WIPHY_FLAG_SUPPORTS_MLO))
+		add_link_files(&sdata->deflink, sdata->vif.debugfs_dir);
 }
 
 void ieee80211_debugfs_remove_netdev(struct ieee80211_sub_if_data *sdata)
@@ -855,3 +906,66 @@ void ieee80211_debugfs_rename_netdev(struct ieee80211_sub_if_data *sdata)
 	sprintf(buf, "netdev:%s", sdata->name);
 	debugfs_rename(dir->d_parent, dir, dir->d_parent, buf);
 }
+
+void ieee80211_link_debugfs_add(struct ieee80211_link_data *link)
+{
+	char link_dir_name[10];
+
+	if (WARN_ON(!link->sdata->vif.debugfs_dir))
+		return;
+
+	/* For now, this should not be called for non-MLO capable drivers */
+	if (WARN_ON(!(link->sdata->local->hw.wiphy->flags & WIPHY_FLAG_SUPPORTS_MLO)))
+		return;
+
+	snprintf(link_dir_name, sizeof(link_dir_name),
+		 "link-%d", link->link_id);
+
+	link->debugfs_dir =
+		debugfs_create_dir(link_dir_name,
+				   link->sdata->vif.debugfs_dir);
+
+	DEBUGFS_ADD(link->debugfs_dir, addr);
+	add_link_files(link, link->debugfs_dir);
+}
+
+void ieee80211_link_debugfs_remove(struct ieee80211_link_data *link)
+{
+	if (!link->sdata->vif.debugfs_dir || !link->debugfs_dir) {
+		link->debugfs_dir = NULL;
+		return;
+	}
+
+	if (link->debugfs_dir == link->sdata->vif.debugfs_dir) {
+		WARN_ON(link != &link->sdata->deflink);
+		link->debugfs_dir = NULL;
+		return;
+	}
+
+	debugfs_remove_recursive(link->debugfs_dir);
+	link->debugfs_dir = NULL;
+}
+
+void ieee80211_link_debugfs_drv_add(struct ieee80211_link_data *link)
+{
+	if (WARN_ON(!link->debugfs_dir))
+		return;
+
+	drv_link_add_debugfs(link->sdata->local, link->sdata,
+			     link->conf, link->debugfs_dir);
+}
+
+void ieee80211_link_debugfs_drv_remove(struct ieee80211_link_data *link)
+{
+	if (!link || !link->debugfs_dir)
+		return;
+
+	if (WARN_ON(link->debugfs_dir == link->sdata->vif.debugfs_dir))
+		return;
+
+	/* Recreate the directory excluding the driver data */
+	debugfs_remove_recursive(link->debugfs_dir);
+	link->debugfs_dir = NULL;
+
+	ieee80211_link_debugfs_add(link);
+}
diff --git a/net/mac80211/debugfs_netdev.h b/net/mac80211/debugfs_netdev.h
index a7e9d8d518f9..99e688dcabd6 100644
--- a/net/mac80211/debugfs_netdev.h
+++ b/net/mac80211/debugfs_netdev.h
@@ -10,6 +10,12 @@
 void ieee80211_debugfs_add_netdev(struct ieee80211_sub_if_data *sdata);
 void ieee80211_debugfs_remove_netdev(struct ieee80211_sub_if_data *sdata);
 void ieee80211_debugfs_rename_netdev(struct ieee80211_sub_if_data *sdata);
+
+void ieee80211_link_debugfs_add(struct ieee80211_link_data *link);
+void ieee80211_link_debugfs_remove(struct ieee80211_link_data *link);
+
+void ieee80211_link_debugfs_drv_add(struct ieee80211_link_data *link);
+void ieee80211_link_debugfs_drv_remove(struct ieee80211_link_data *link);
 #else
 static inline void ieee80211_debugfs_add_netdev(
 	struct ieee80211_sub_if_data *sdata)
@@ -20,6 +26,16 @@ static inline void ieee80211_debugfs_remove_netdev(
 static inline void ieee80211_debugfs_rename_netdev(
 	struct ieee80211_sub_if_data *sdata)
 {}
+
+static inline void ieee80211_link_debugfs_add(struct ieee80211_link_data *link)
+{}
+static inline void ieee80211_link_debugfs_remove(struct ieee80211_link_data *link)
+{}
+
+static inline void ieee80211_link_debugfs_drv_add(struct ieee80211_link_data *link)
+{}
+static inline void ieee80211_link_debugfs_drv_remove(struct ieee80211_link_data *link)
+{}
 #endif
 
 #endif /* __IEEE80211_DEBUGFS_NETDEV_H */
diff --git a/net/mac80211/driver-ops.c b/net/mac80211/driver-ops.c
index cfb09e4aed4d..30cd0c905a24 100644
--- a/net/mac80211/driver-ops.c
+++ b/net/mac80211/driver-ops.c
@@ -8,6 +8,7 @@
 #include "trace.h"
 #include "driver-ops.h"
 #include "debugfs_sta.h"
+#include "debugfs_netdev.h"
 
 int drv_start(struct ieee80211_local *local)
 {
@@ -477,6 +478,10 @@ int drv_change_vif_links(struct ieee80211_local *local,
 			 u16 old_links, u16 new_links,
 			 struct ieee80211_bss_conf *old[IEEE80211_MLD_MAX_NUM_LINKS])
 {
+	struct ieee80211_link_data *link;
+	unsigned long links_to_add;
+	unsigned long links_to_rem;
+	unsigned int link_id;
 	int ret = -EOPNOTSUPP;
 
 	might_sleep();
@@ -487,13 +492,31 @@ int drv_change_vif_links(struct ieee80211_local *local,
 	if (old_links == new_links)
 		return 0;
 
+	links_to_add = ~old_links & new_links;
+	links_to_rem = old_links & ~new_links;
+
+	for_each_set_bit(link_id, &links_to_rem, IEEE80211_MLD_MAX_NUM_LINKS) {
+		link = rcu_access_pointer(sdata->link[link_id]);
+
+		ieee80211_link_debugfs_drv_remove(link);
+	}
+
 	trace_drv_change_vif_links(local, sdata, old_links, new_links);
 	if (local->ops->change_vif_links)
 		ret = local->ops->change_vif_links(&local->hw, &sdata->vif,
 						   old_links, new_links, old);
 	trace_drv_return_int(local, ret);
 
-	return ret;
+	if (ret)
+		return ret;
+
+	for_each_set_bit(link_id, &links_to_add, IEEE80211_MLD_MAX_NUM_LINKS) {
+		link = rcu_access_pointer(sdata->link[link_id]);
+
+		ieee80211_link_debugfs_drv_add(link);
+	}
+
+	return 0;
 }
 
 int drv_change_sta_links(struct ieee80211_local *local,
diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h
index 5d13a3dfd366..a68d606e6987 100644
--- a/net/mac80211/driver-ops.h
+++ b/net/mac80211/driver-ops.h
@@ -465,6 +465,22 @@ static inline void drv_sta_remove(struct ieee80211_local *local,
 }
 
 #ifdef CONFIG_MAC80211_DEBUGFS
+static inline void drv_link_add_debugfs(struct ieee80211_local *local,
+					struct ieee80211_sub_if_data *sdata,
+					struct ieee80211_bss_conf *link_conf,
+					struct dentry *dir)
+{
+	might_sleep();
+
+	sdata = get_bss_sdata(sdata);
+	if (!check_sdata_in_driver(sdata))
+		return;
+
+	if (local->ops->link_add_debugfs)
+		local->ops->link_add_debugfs(&local->hw, &sdata->vif,
+					     link_conf, dir);
+}
+
 static inline void drv_sta_add_debugfs(struct ieee80211_local *local,
 				       struct ieee80211_sub_if_data *sdata,
 				       struct ieee80211_sta *sta,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index ecc232eb1ee8..3d4edc25a69e 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -999,6 +999,10 @@ struct ieee80211_link_data {
 	struct ieee80211_tx_queue_params tx_conf[IEEE80211_NUM_ACS];
 
 	struct ieee80211_bss_conf *conf;
+
+#ifdef CONFIG_MAC80211_DEBUGFS
+	struct dentry *debugfs_dir;
+#endif
 };
 
 struct ieee80211_sub_if_data {
diff --git a/net/mac80211/link.c b/net/mac80211/link.c
index e6f9fce1dadb..e82db88a47f8 100644
--- a/net/mac80211/link.c
+++ b/net/mac80211/link.c
@@ -10,6 +10,7 @@
 #include "ieee80211_i.h"
 #include "driver-ops.h"
 #include "key.h"
+#include "debugfs_netdev.h"
 
 void ieee80211_link_setup(struct ieee80211_link_data *link)
 {
@@ -61,6 +62,8 @@ void ieee80211_link_init(struct ieee80211_sub_if_data *sdata,
 		default:
 			WARN_ON(1);
 		}
+
+		ieee80211_link_debugfs_add(link);
 	}
 }
 
@@ -94,6 +97,7 @@ static void ieee80211_tear_down_links(struct ieee80211_sub_if_data *sdata,
 		if (WARN_ON(!link))
 			continue;
 		ieee80211_remove_link_keys(link, &keys);
+		ieee80211_link_debugfs_remove(link);
 		ieee80211_link_stop(link);
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 5cf10940a47985260ebbdc4665a8081c6b575e17 Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Wed, 1 Mar 2023 12:09:28 +0200
Subject: wifi: mac80211_hwsim: Indicate support for
 NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT

mac80211_hwsim always uses the probe request coming from mac80211,
which has support for minimal probe request content.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.67d31166ffd1.I908b68c5127525da3388c0d61a3d5d221ba451f6@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/mac80211_hwsim.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index 4cc4eaf80b14..e9b9340a97dc 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -4446,6 +4446,9 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 	wiphy_ext_feature_set(hw->wiphy,
 			      NL80211_EXT_FEATURE_BEACON_RATE_LEGACY);
 
+	wiphy_ext_feature_set(hw->wiphy,
+			      NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT);
+
 	hw->wiphy->interface_modes = param->iftypes;
 
 	/* ask mac80211 to reserve space for magic */
-- 
cgit v1.2.3-70-g09d2


From 15f9b3ef5190bacc8a39628bc57c6da085877ffe Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:30 +0200
Subject: wifi: mac80211: mlme: remove pointless sta check

We already exited the function if sta ended up NULL,
so just remove the extra check.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.4cbac9cfd03a.I21ec81c96d246afdabc2b0807d3856e6b1182cb7@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index cdf842c0849b..78adbacf8538 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -5893,7 +5893,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 		goto free;
 	}
 
-	if (sta && elems->opmode_notif)
+	if (elems->opmode_notif)
 		ieee80211_vht_handle_opmode(sdata, link_sta,
 					    *elems->opmode_notif,
 					    rx_status->band);
-- 
cgit v1.2.3-70-g09d2


From e8edb34640eeeefc74a3d767b61b4d4ac3b94eea Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:31 +0200
Subject: wifi: mac80211: simplify reasoning about EHT capa handling

Given the code in cfg80211, EHT capa cannot be non-NULL when
HE capa is NULL, but it's easier to reason about it if both
are checked and the compiler will likely integrate the check
with the previous one for HE capa anyway.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.7413d50d23bc.I6fef7484721be9bd5364f64921fc5e9168495f62@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 7e90f4a81962..2c8fa19c2f24 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1788,7 +1788,7 @@ static int sta_link_apply_parameters(struct ieee80211_local *local,
 						  (void *)params->he_6ghz_capa,
 						  link_sta);
 
-	if (params->eht_capa)
+	if (params->he_capa && params->eht_capa)
 		ieee80211_eht_cap_ie_to_sta_eht_cap(sdata, sband,
 						    (u8 *)params->he_capa,
 						    params->he_capa_len,
-- 
cgit v1.2.3-70-g09d2


From e820373a4fd06ebb6ed0331e97693048cfd16652 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 1 Mar 2023 12:09:32 +0200
Subject: wifi: mac80211: fix ieee80211_link_set_associated() type

The return type here should be u64 for the flags, even
if it doesn't matter right now because it doesn't return
any flags that don't fit into u32.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.d67ccae57d60.Ia4768e547ba8b1deb2b84ce3bbfbe216d5bfff6a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 78adbacf8538..e13a0354c397 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2744,7 +2744,7 @@ static u32 ieee80211_handle_bss_capability(struct ieee80211_link_data *link,
 	return changed;
 }
 
-static u32 ieee80211_link_set_associated(struct ieee80211_link_data *link,
+static u64 ieee80211_link_set_associated(struct ieee80211_link_data *link,
 					 struct cfg80211_bss *cbss)
 {
 	struct ieee80211_sub_if_data *sdata = link->sdata;
-- 
cgit v1.2.3-70-g09d2


From 18cbf7c089ba70fefe1b4c01af28753cabfbf38f Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Wed, 1 Mar 2023 12:09:34 +0200
Subject: wifi: radiotap: Add EHT radiotap definitions

This is based on https://www.radiotap.org/fields/EHT.html and
https://www.radiotap.org/fields/U-SIG.html new EHT TLV
definition for 11be standard.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.254b19fffe41.I4ce78e2c558da6e5a708a8d68d61b5d7b3eb3746@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/ieee80211_radiotap.h | 187 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 185 insertions(+), 2 deletions(-)

diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h
index 598f53d2a3a0..0fc2667a9a5d 100644
--- a/include/net/ieee80211_radiotap.h
+++ b/include/net/ieee80211_radiotap.h
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2017		Intel Deutschland GmbH
- * Copyright (c) 2018-2019, 2021 Intel Corporation
+ * Copyright (c) 2018-2019, 2021-2022 Intel Corporation
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -82,11 +82,14 @@ enum ieee80211_radiotap_presence {
 	IEEE80211_RADIOTAP_HE_MU = 24,
 	IEEE80211_RADIOTAP_ZERO_LEN_PSDU = 26,
 	IEEE80211_RADIOTAP_LSIG = 27,
+	IEEE80211_RADIOTAP_TLV = 28,
 
 	/* valid in every it_present bitmap, even vendor namespaces */
 	IEEE80211_RADIOTAP_RADIOTAP_NAMESPACE = 29,
 	IEEE80211_RADIOTAP_VENDOR_NAMESPACE = 30,
-	IEEE80211_RADIOTAP_EXT = 31
+	IEEE80211_RADIOTAP_EXT = 31,
+	IEEE80211_RADIOTAP_EHT_USIG = 33,
+	IEEE80211_RADIOTAP_EHT = 34,
 };
 
 /* for IEEE80211_RADIOTAP_FLAGS */
@@ -360,6 +363,186 @@ enum ieee80211_radiotap_zero_len_psdu_type {
 	IEEE80211_RADIOTAP_ZERO_LEN_PSDU_VENDOR			= 0xff,
 };
 
+struct ieee80211_radiotap_tlv {
+	__le16 type;
+	__le16 len;
+	u8 data[];
+} __packed;
+
+/* ieee80211_radiotap_eht_usig - content of U-SIG tlv (type 33)
+ * see www.radiotap.org/fields/U-SIG.html for details
+ */
+struct ieee80211_radiotap_eht_usig {
+	__le32 common;
+	__le32 value;
+	__le32 mask;
+} __packed;
+
+/* ieee80211_radiotap_eht - content of EHT tlv (type 34)
+ * see www.radiotap.org/fields/EHT.html for details
+ */
+struct ieee80211_radiotap_eht {
+	__le32 known;
+	__le32 data[9];
+	__le32 user_info[];
+} __packed;
+
+/* Known field for EHT TLV
+ * The ending defines for what the field applies as following
+ * O - OFDMA (including TB), M - MU-MIMO, S - EHT sounding.
+ */
+enum ieee80211_radiotap_eht_known {
+	IEEE80211_RADIOTAP_EHT_KNOWN_SPATIAL_REUSE		= 0x00000002,
+	IEEE80211_RADIOTAP_EHT_KNOWN_GI				= 0x00000004,
+	IEEE80211_RADIOTAP_EHT_KNOWN_EHT_LTF			= 0x00000010,
+	IEEE80211_RADIOTAP_EHT_KNOWN_LDPC_EXTRA_SYM_OM		= 0x00000020,
+	IEEE80211_RADIOTAP_EHT_KNOWN_PRE_PADD_FACOR_OM		= 0x00000040,
+	IEEE80211_RADIOTAP_EHT_KNOWN_PE_DISAMBIGUITY_OM		= 0x00000080,
+	IEEE80211_RADIOTAP_EHT_KNOWN_DISREGARD_O		= 0x00000100,
+	IEEE80211_RADIOTAP_EHT_KNOWN_DISREGARD_S		= 0x00000200,
+	IEEE80211_RADIOTAP_EHT_KNOWN_CRC1			= 0x00002000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_TAIL1			= 0x00004000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_CRC2_O			= 0x00008000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_TAIL2_O			= 0x00010000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_NSS_S			= 0x00020000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_BEAMFORMED_S		= 0x00040000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_NR_NON_OFDMA_USERS_M	= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_ENCODING_BLOCK_CRC_M	= 0x00100000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_ENCODING_BLOCK_TAIL_M	= 0x00200000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_RU_MRU_SIZE_OM		= 0x00400000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_RU_MRU_INDEX_OM		= 0x00800000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_RU_ALLOC_TB_FMT		= 0x01000000,
+	IEEE80211_RADIOTAP_EHT_KNOWN_PRIMARY_80			= 0x02000000,
+};
+
+enum ieee80211_radiotap_eht_data {
+	/* Data 0 */
+	IEEE80211_RADIOTAP_EHT_DATA0_SPATIAL_REUSE		= 0x00000078,
+	IEEE80211_RADIOTAP_EHT_DATA0_GI				= 0x00000180,
+	IEEE80211_RADIOTAP_EHT_DATA0_LTF			= 0x00000600,
+	IEEE80211_RADIOTAP_EHT_DATA0_EHT_LTF			= 0x00003800,
+	IEEE80211_RADIOTAP_EHT_DATA0_LDPC_EXTRA_SYM_OM		= 0x00004000,
+	IEEE80211_RADIOTAP_EHT_DATA0_PRE_PADD_FACOR_OM		= 0x00018000,
+	IEEE80211_RADIOTAP_EHT_DATA0_PE_DISAMBIGUITY_OM		= 0x00020000,
+	IEEE80211_RADIOTAP_EHT_DATA0_DISREGARD_S		= 0x000c0000,
+	IEEE80211_RADIOTAP_EHT_DATA0_DISREGARD_O		= 0x003c0000,
+	IEEE80211_RADIOTAP_EHT_DATA0_CRC1_O			= 0x03c00000,
+	IEEE80211_RADIOTAP_EHT_DATA0_TAIL1_O			= 0xfc000000,
+	/* Data 1 */
+	IEEE80211_RADIOTAP_EHT_DATA1_RU_SIZE			= 0x0000001f,
+	IEEE80211_RADIOTAP_EHT_DATA1_RU_INDEX			= 0x00001fe0,
+	IEEE80211_RADIOTAP_EHT_DATA1_RU_ALLOC_CC_1_1_1		= 0x003fe000,
+	IEEE80211_RADIOTAP_EHT_DATA1_RU_ALLOC_CC_1_1_1_KNOWN	= 0x00400000,
+	IEEE80211_RADIOTAP_EHT_DATA1_PRIMARY_80			= 0xc0000000,
+	/* Data 2 */
+	IEEE80211_RADIOTAP_EHT_DATA2_RU_ALLOC_CC_2_1_1		= 0x000001ff,
+	IEEE80211_RADIOTAP_EHT_DATA2_RU_ALLOC_CC_2_1_1_KNOWN	= 0x00000200,
+	IEEE80211_RADIOTAP_EHT_DATA2_RU_ALLOC_CC_1_1_2		= 0x0007fc00,
+	IEEE80211_RADIOTAP_EHT_DATA2_RU_ALLOC_CC_1_1_2_KNOWN	= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_DATA2_RU_ALLOC_CC_2_1_2		= 0x1ff00000,
+	IEEE80211_RADIOTAP_EHT_DATA2_RU_ALLOC_CC_2_1_2_KNOWN	= 0x20000000,
+	/* Data 3 */
+	IEEE80211_RADIOTAP_EHT_DATA3_RU_ALLOC_CC_1_2_1		= 0x000001ff,
+	IEEE80211_RADIOTAP_EHT_DATA3_RU_ALLOC_CC_1_2_1_KNOWN	= 0x00000200,
+	IEEE80211_RADIOTAP_EHT_DATA3_RU_ALLOC_CC_2_2_1		= 0x0007fc00,
+	IEEE80211_RADIOTAP_EHT_DATA3_RU_ALLOC_CC_2_2_1_KNOWN	= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_DATA3_RU_ALLOC_CC_1_2_2		= 0x1ff00000,
+	IEEE80211_RADIOTAP_EHT_DATA3_RU_ALLOC_CC_1_2_2_KNOWN	= 0x20000000,
+	/* Data 4 */
+	IEEE80211_RADIOTAP_EHT_DATA4_RU_ALLOC_CC_2_2_2		= 0x000001ff,
+	IEEE80211_RADIOTAP_EHT_DATA4_RU_ALLOC_CC_2_2_2_KNOWN	= 0x00000200,
+	IEEE80211_RADIOTAP_EHT_DATA4_RU_ALLOC_CC_1_2_3		= 0x0007fc00,
+	IEEE80211_RADIOTAP_EHT_DATA4_RU_ALLOC_CC_1_2_3_KNOWN	= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_DATA4_RU_ALLOC_CC_2_2_3		= 0x1ff00000,
+	IEEE80211_RADIOTAP_EHT_DATA4_RU_ALLOC_CC_2_2_3_KNOWN	= 0x20000000,
+	/* Data 5 */
+	IEEE80211_RADIOTAP_EHT_DATA5_RU_ALLOC_CC_1_2_4		= 0x000001ff,
+	IEEE80211_RADIOTAP_EHT_DATA5_RU_ALLOC_CC_1_2_4_KNOWN	= 0x00000200,
+	IEEE80211_RADIOTAP_EHT_DATA5_RU_ALLOC_CC_2_2_4		= 0x0007fc00,
+	IEEE80211_RADIOTAP_EHT_DATA5_RU_ALLOC_CC_2_2_4_KNOWN	= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_DATA5_RU_ALLOC_CC_1_2_5		= 0x1ff00000,
+	IEEE80211_RADIOTAP_EHT_DATA5_RU_ALLOC_CC_1_2_5_KNOWN	= 0x20000000,
+	/* Data 6 */
+	IEEE80211_RADIOTAP_EHT_DATA6_RU_ALLOC_CC_2_2_5		= 0x000001ff,
+	IEEE80211_RADIOTAP_EHT_DATA6_RU_ALLOC_CC_2_2_5_KNOWN	= 0x00000200,
+	IEEE80211_RADIOTAP_EHT_DATA6_RU_ALLOC_CC_1_2_6		= 0x0007fc00,
+	IEEE80211_RADIOTAP_EHT_DATA6_RU_ALLOC_CC_1_2_6_KNOWN	= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_DATA6_RU_ALLOC_CC_2_2_6		= 0x1ff00000,
+	IEEE80211_RADIOTAP_EHT_DATA6_RU_ALLOC_CC_2_2_6_KNOWN	= 0x20000000,
+	/* Data 7 */
+	IEEE80211_RADIOTAP_EHT_DATA7_CRC2_O			= 0x0000000f,
+	IEEE80211_RADIOTAP_EHT_DATA7_TAIL_2_O			= 0x000003f0,
+	IEEE80211_RADIOTAP_EHT_DATA7_NSS_S			= 0x0000f000,
+	IEEE80211_RADIOTAP_EHT_DATA7_BEAMFORMED_S		= 0x00010000,
+	IEEE80211_RADIOTAP_EHT_DATA7_NUM_OF_NON_OFDMA_USERS	= 0x000e0000,
+	IEEE80211_RADIOTAP_EHT_DATA7_USER_ENCODING_BLOCK_CRC	= 0x00f00000,
+	IEEE80211_RADIOTAP_EHT_DATA7_USER_ENCODING_BLOCK_TAIL	= 0x3f000000,
+	/* Data 8 */
+	IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_PS_160	= 0x00000001,
+	IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B0		= 0x00000002,
+	IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1	= 0x000001fc,
+};
+
+enum ieee80211_radiotap_eht_user_info {
+	IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID_KNOWN		= 0x00000001,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_MCS_KNOWN		= 0x00000002,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_CODING_KNOWN		= 0x00000004,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_KNOWN_O		= 0x00000010,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_KNOWN_O	= 0x00000020,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_SPATIAL_CONFIG_KNOWN_M	= 0x00000040,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_DATA_FOR_USER		= 0x00000080,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID			= 0x0007ff00,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_CODING			= 0x00080000,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_MCS			= 0x00f00000,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O			= 0x0f000000,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_O		= 0x20000000,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_SPATIAL_CONFIG_M	= 0x3f000000,
+	IEEE80211_RADIOTAP_EHT_USER_INFO_RESEVED_c0000000	= 0xc0000000,
+};
+
+enum ieee80211_radiotap_eht_usig_common {
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER_KNOWN	= 0x00000001,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_KNOWN		= 0x00000002,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL_KNOWN		= 0x00000004,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN	= 0x00000008,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN		= 0x00000010,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_BAD_USIG_CRC		= 0x00000020,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER		= 0x00007000,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW			= 0x00038000,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL		= 0x00040000,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR		= 0x01f80000,
+	IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP			= 0xfe000000,
+};
+
+enum ieee80211_radiotap_eht_usig_mu {
+	/* MU-USIG-1 */
+	IEEE80211_RADIOTAP_EHT_USIG1_MU_B20_B24_DISREGARD	= 0x0000001f,
+	IEEE80211_RADIOTAP_EHT_USIG1_MU_B25_VALIDATE		= 0x00000020,
+	/* MU-USIG-2 */
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B0_B1_PPDU_TYPE		= 0x000000c0,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B2_VALIDATE		= 0x00000100,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B3_B7_PUNCTURED_INFO	= 0x00003e00,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B8_VALIDATE		= 0x00004000,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B9_B10_SIG_MCS		= 0x00018000,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS = 0x003e0000,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B16_B19_CRC		= 0x03c00000,
+	IEEE80211_RADIOTAP_EHT_USIG2_MU_B20_B25_TAIL		= 0xfc000000,
+};
+
+enum ieee80211_radiotap_eht_usig_tb {
+	/* TB-USIG-1 */
+	IEEE80211_RADIOTAP_EHT_USIG1_TB_B20_B25_DISREGARD	= 0x0000001f,
+
+	/* TB-USIG-2 */
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B0_B1_PPDU_TYPE		= 0x000000c0,
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B2_VALIDATE		= 0x00000100,
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B3_B6_SPATIAL_REUSE_1	= 0x00001e00,
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B7_B10_SPATIAL_REUSE_2	= 0x0001e000,
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B11_B15_DISREGARD	= 0x003e0000,
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B16_B19_CRC		= 0x03c00000,
+	IEEE80211_RADIOTAP_EHT_USIG2_TB_B20_B25_TAIL		= 0xfc000000,
+};
+
 /**
  * ieee80211_get_radiotap_len - get radiotap header length
  */
-- 
cgit v1.2.3-70-g09d2


From 9179dff82598ab8b4e88dcc93c9e26a2594efd1a Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Wed, 1 Mar 2023 12:09:35 +0200
Subject: wifi: mac80211: add support for driver adding radiotap TLVs

The new TLV format enables adding TLVs after the fixed
fields in radiotap, as part of the radiotap header.
Support this and move vendor data to the TLV format,
allowing a reuse of the RX_FLAG_RADIOTAP_VENDOR_DATA as
the new RX_FLAG_RADIOTAP_TLV_AT_END flag.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230301115906.b18fd5da8477.I576400ec40a7b35ef97a3b09a99b3a49e9174786@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 27 ++++-----
 drivers/net/wireless/mac80211_hwsim.c         | 45 ++++++++-------
 include/net/ieee80211_radiotap.h              | 20 +++++++
 include/net/mac80211.h                        | 44 +++-----------
 net/mac80211/rx.c                             | 82 +++++++++------------------
 5 files changed, 91 insertions(+), 127 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 549dbe0be223..d1769464d75b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -209,33 +209,34 @@ static void iwl_mvm_add_rtap_sniffer_config(struct iwl_mvm *mvm,
 					    struct sk_buff *skb)
 {
 	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
-	struct ieee80211_vendor_radiotap *radiotap;
-	const int size = sizeof(*radiotap) + sizeof(__le16);
+	struct ieee80211_radiotap_vendor_tlv *radiotap;
+	const u16 vendor_data_len = sizeof(mvm->cur_aid);
+	const u16 padding = ALIGN(vendor_data_len, 4) - vendor_data_len;
 
 	if (!mvm->cur_aid)
 		return;
 
-	/* ensure alignment */
-	BUILD_BUG_ON((size + 2) % 4);
+	radiotap = skb_put(skb, sizeof(*radiotap) + vendor_data_len + padding);
+	radiotap->type = cpu_to_le16(IEEE80211_RADIOTAP_VENDOR_NAMESPACE);
+	radiotap->len = cpu_to_le16(sizeof(*radiotap) -
+				    sizeof(struct ieee80211_radiotap_tlv) +
+				    vendor_data_len);
 
-	radiotap = skb_put(skb, size + 2);
-	radiotap->align = 1;
 	/* Intel OUI */
 	radiotap->oui[0] = 0xf6;
 	radiotap->oui[1] = 0x54;
 	radiotap->oui[2] = 0x25;
 	/* radiotap sniffer config sub-namespace */
-	radiotap->subns = 1;
-	radiotap->present = 0x1;
-	radiotap->len = size - sizeof(*radiotap);
-	radiotap->pad = 2;
-
+	radiotap->oui_subtype = 1;
+	radiotap->vendor_type = 0;
+	/* clear reserved field */
+	radiotap->reserved = 0;
 	/* fill the data now */
 	memcpy(radiotap->data, &mvm->cur_aid, sizeof(mvm->cur_aid));
 	/* and clear the padding */
-	memset(radiotap->data + sizeof(__le16), 0, radiotap->pad);
+	memset(radiotap->data + vendor_data_len, 0, padding);
 
-	rx_status->flag |= RX_FLAG_RADIOTAP_VENDOR_DATA;
+	rx_status->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
 }
 
 /* iwl_mvm_pass_packet_to_mac80211 - passes the packet for mac80211 */
diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index e9b9340a97dc..152617034d19 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -1534,37 +1534,38 @@ static void mac80211_hwsim_add_vendor_rtap(struct sk_buff *skb)
 	 * the values accordingly.
 	 */
 #ifdef HWSIM_RADIOTAP_OUI
-	struct ieee80211_vendor_radiotap *rtap;
+	struct ieee80211_radiotap_vendor_tlv *rtap;
+	static const char vendor_data[8] = "ABCDEFGH";
+
+	// Make sure no padding is needed
+	BUILD_BUG_ON(sizeof(vendor_data) % 4);
+	/* this is last radiotap info before the mac header, so
+	 * skb_reset_mac_header for mac8022 to know the end of
+	 * the radiotap TLV/beginning of the 802.11 header
+	 */
+	skb_reset_mac_header(skb);
 
 	/*
 	 * Note that this code requires the headroom in the SKB
 	 * that was allocated earlier.
 	 */
-	rtap = skb_push(skb, sizeof(*rtap) + 8 + 4);
+	rtap = skb_push(skb, sizeof(*rtap) + sizeof(vendor_data));
+
+	rtap->len = cpu_to_le16(sizeof(*rtap) -
+				sizeof(struct ieee80211_radiotap_tlv) +
+				sizeof(vendor_data));
+	rtap->type = cpu_to_le16(IEEE80211_RADIOTAP_VENDOR_NAMESPACE);
+
 	rtap->oui[0] = HWSIM_RADIOTAP_OUI[0];
 	rtap->oui[1] = HWSIM_RADIOTAP_OUI[1];
 	rtap->oui[2] = HWSIM_RADIOTAP_OUI[2];
-	rtap->subns = 127;
+	rtap->oui_subtype = 127;
+	/* clear reserved field */
+	rtap->reserved = 0;
+	rtap->vendor_type = 0;
+	memcpy(rtap->data, vendor_data, sizeof(vendor_data));
 
-	/*
-	 * Radiotap vendor namespaces can (and should) also be
-	 * split into fields by using the standard radiotap
-	 * presence bitmap mechanism. Use just BIT(0) here for
-	 * the presence bitmap.
-	 */
-	rtap->present = BIT(0);
-	/* We have 8 bytes of (dummy) data */
-	rtap->len = 8;
-	/* For testing, also require it to be aligned */
-	rtap->align = 8;
-	/* And also test that padding works, 4 bytes */
-	rtap->pad = 4;
-	/* push the data */
-	memcpy(rtap->data, "ABCDEFGH", 8);
-	/* make sure to clear padding, mac80211 doesn't */
-	memset(rtap->data + 8, 0, 4);
-
-	IEEE80211_SKB_RXCB(skb)->flag |= RX_FLAG_RADIOTAP_VENDOR_DATA;
+	IEEE80211_SKB_RXCB(skb)->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
 #endif
 }
 
diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h
index 0fc2667a9a5d..95436686d3fe 100644
--- a/include/net/ieee80211_radiotap.h
+++ b/include/net/ieee80211_radiotap.h
@@ -369,6 +369,26 @@ struct ieee80211_radiotap_tlv {
 	u8 data[];
 } __packed;
 
+/**
+ * struct ieee80211_radiotap_vendor_tlv - vendor radiotap data information
+ * @type: should always be set to IEEE80211_RADIOTAP_VENDOR_NAMESPACE
+ * @len: length of data
+ * @oui: radiotap vendor namespace OUI
+ * @oui_subtype: radiotap vendor sub namespace
+ * @vendor_type: radiotap vendor type
+ * @reserved: should always be set to zero (to avoid leaking memory)
+ * @data: the actual vendor namespace data
+ */
+struct ieee80211_radiotap_vendor_tlv {
+	__le16 type; /* IEEE80211_RADIOTAP_VENDOR_NAMESPACE */
+	__le16 len;
+	u8 oui[3];
+	u8 oui_subtype;
+	__le16 vendor_type;
+	__le16 reserved;
+	u8 data[];
+} __packed;
+
 /* ieee80211_radiotap_eht_usig - content of U-SIG tlv (type 33)
  * see www.radiotap.org/fields/U-SIG.html for details
  */
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5df9eb828a58..3a43ce5fd4ec 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1375,9 +1375,12 @@ ieee80211_tx_info_clear_status(struct ieee80211_tx_info *info)
  *	subframes share the same sequence number. Reported subframes can be
  *	either regular MSDU or singly A-MSDUs. Subframes must not be
  *	interleaved with other frames.
- * @RX_FLAG_RADIOTAP_VENDOR_DATA: This frame contains vendor-specific
- *	radiotap data in the skb->data (before the frame) as described by
- *	the &struct ieee80211_vendor_radiotap.
+ * @RX_FLAG_RADIOTAP_TLV_AT_END: This frame contains radiotap TLVs in the
+ *	skb->data (before the 802.11 header).
+ *	If used, the SKB's mac_header pointer must be set to point
+ *	to the 802.11 header after the TLVs, and any padding added after TLV
+ *	data to align to 4 must be cleared by the driver putting the TLVs
+ *	in the skb.
  * @RX_FLAG_ALLOW_SAME_PN: Allow the same PN as same packet before.
  *	This is used for AMSDU subframes which can have the same PN as
  *	the first subframe.
@@ -1429,7 +1432,7 @@ enum mac80211_rx_flags {
 	RX_FLAG_ONLY_MONITOR		= BIT(17),
 	RX_FLAG_SKIP_MONITOR		= BIT(18),
 	RX_FLAG_AMSDU_MORE		= BIT(19),
-	RX_FLAG_RADIOTAP_VENDOR_DATA	= BIT(20),
+	RX_FLAG_RADIOTAP_TLV_AT_END	= BIT(20),
 	RX_FLAG_MIC_STRIPPED		= BIT(21),
 	RX_FLAG_ALLOW_SAME_PN		= BIT(22),
 	RX_FLAG_ICV_STRIPPED		= BIT(23),
@@ -1569,39 +1572,6 @@ ieee80211_rx_status_to_khz(struct ieee80211_rx_status *rx_status)
 	       (rx_status->freq_offset ? 500 : 0);
 }
 
-/**
- * struct ieee80211_vendor_radiotap - vendor radiotap data information
- * @present: presence bitmap for this vendor namespace
- *	(this could be extended in the future if any vendor needs more
- *	 bits, the radiotap spec does allow for that)
- * @align: radiotap vendor namespace alignment. This defines the needed
- *	alignment for the @data field below, not for the vendor namespace
- *	description itself (which has a fixed 2-byte alignment)
- *	Must be a power of two, and be set to at least 1!
- * @oui: radiotap vendor namespace OUI
- * @subns: radiotap vendor sub namespace
- * @len: radiotap vendor sub namespace skip length, if alignment is done
- *	then that's added to this, i.e. this is only the length of the
- *	@data field.
- * @pad: number of bytes of padding after the @data, this exists so that
- *	the skb data alignment can be preserved even if the data has odd
- *	length
- * @data: the actual vendor namespace data
- *
- * This struct, including the vendor data, goes into the skb->data before
- * the 802.11 header. It's split up in mac80211 using the align/oui/subns
- * data.
- */
-struct ieee80211_vendor_radiotap {
-	u32 present;
-	u8 align;
-	u8 oui[3];
-	u8 subns;
-	u8 pad;
-	u16 len;
-	u8 data[];
-} __packed;
-
 /**
  * enum ieee80211_conf_flags - configuration flags
  *
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index f63ed6b91d9b..0255c5745e1c 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -55,7 +55,7 @@ static struct sk_buff *ieee80211_clean_skb(struct sk_buff *skb,
 	/* After pulling radiotap header, clear all flags that indicate
 	 * info in skb->data.
 	 */
-	status->flag &= ~(RX_FLAG_RADIOTAP_VENDOR_DATA |
+	status->flag &= ~(RX_FLAG_RADIOTAP_TLV_AT_END |
 			  RX_FLAG_RADIOTAP_LSIG |
 			  RX_FLAG_RADIOTAP_HE_MU |
 			  RX_FLAG_RADIOTAP_HE);
@@ -126,9 +126,6 @@ ieee80211_rx_radiotap_hdrlen(struct ieee80211_local *local,
 	/* allocate extra bitmaps */
 	if (status->chains)
 		len += 4 * hweight8(status->chains);
-	/* vendor presence bitmap */
-	if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA)
-		len += 4;
 
 	if (ieee80211_have_rx_timestamp(status)) {
 		len = ALIGN(len, 8);
@@ -190,34 +187,28 @@ ieee80211_rx_radiotap_hdrlen(struct ieee80211_local *local,
 		len += 2 * hweight8(status->chains);
 	}
 
-	if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA) {
-		struct ieee80211_vendor_radiotap *rtap;
-		int vendor_data_offset = 0;
+	if (status->flag & RX_FLAG_RADIOTAP_TLV_AT_END) {
+		int tlv_offset = 0;
 
 		/*
 		 * The position to look at depends on the existence (or non-
 		 * existence) of other elements, so take that into account...
 		 */
 		if (status->flag & RX_FLAG_RADIOTAP_HE)
-			vendor_data_offset +=
+			tlv_offset +=
 				sizeof(struct ieee80211_radiotap_he);
 		if (status->flag & RX_FLAG_RADIOTAP_HE_MU)
-			vendor_data_offset +=
+			tlv_offset +=
 				sizeof(struct ieee80211_radiotap_he_mu);
 		if (status->flag & RX_FLAG_RADIOTAP_LSIG)
-			vendor_data_offset +=
+			tlv_offset +=
 				sizeof(struct ieee80211_radiotap_lsig);
 
-		rtap = (void *)&skb->data[vendor_data_offset];
+		/* ensure 4 byte alignment for TLV */
+		len = ALIGN(len, 4);
 
-		/* alignment for fixed 6-byte vendor data header */
-		len = ALIGN(len, 2);
-		/* vendor data header */
-		len += 6;
-		if (WARN_ON(rtap->align == 0))
-			rtap->align = 1;
-		len = ALIGN(len, rtap->align);
-		len += rtap->len + rtap->pad;
+		/* TLVs until the mac header */
+		len += skb_mac_header(skb) - &skb->data[tlv_offset];
 	}
 
 	return len;
@@ -313,9 +304,9 @@ ieee80211_add_rx_radiotap_header(struct ieee80211_local *local,
 	u32 it_present_val;
 	u16 rx_flags = 0;
 	u16 channel_flags = 0;
+	u32 tlvs_len = 0;
 	int mpdulen, chain;
 	unsigned long chains = status->chains;
-	struct ieee80211_vendor_radiotap rtap = {};
 	struct ieee80211_radiotap_he he = {};
 	struct ieee80211_radiotap_he_mu he_mu = {};
 	struct ieee80211_radiotap_lsig lsig = {};
@@ -336,18 +327,17 @@ ieee80211_add_rx_radiotap_header(struct ieee80211_local *local,
 		skb_pull(skb, sizeof(lsig));
 	}
 
-	if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA) {
-		rtap = *(struct ieee80211_vendor_radiotap *)skb->data;
-		/* rtap.len and rtap.pad are undone immediately */
-		skb_pull(skb, sizeof(rtap) + rtap.len + rtap.pad);
+	if (status->flag & RX_FLAG_RADIOTAP_TLV_AT_END) {
+		/* data is pointer at tlv all other info was pulled off */
+		tlvs_len = skb_mac_header(skb) - skb->data;
 	}
 
 	mpdulen = skb->len;
 	if (!(has_fcs && ieee80211_hw_check(&local->hw, RX_INCLUDES_FCS)))
 		mpdulen += FCS_LEN;
 
-	rthdr = skb_push(skb, rtap_len);
-	memset(rthdr, 0, rtap_len - rtap.len - rtap.pad);
+	rthdr = skb_push(skb, rtap_len - tlvs_len);
+	memset(rthdr, 0, rtap_len - tlvs_len);
 	it_present = &rthdr->it_present;
 
 	/* radiotap header, set always present flags */
@@ -369,13 +359,8 @@ ieee80211_add_rx_radiotap_header(struct ieee80211_local *local,
 				 BIT(IEEE80211_RADIOTAP_DBM_ANTSIGNAL);
 	}
 
-	if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA) {
-		it_present_val |= BIT(IEEE80211_RADIOTAP_VENDOR_NAMESPACE) |
-				  BIT(IEEE80211_RADIOTAP_EXT);
-		put_unaligned_le32(it_present_val, it_present);
-		it_present++;
-		it_present_val = rtap.present;
-	}
+	if (status->flag & RX_FLAG_RADIOTAP_TLV_AT_END)
+		it_present_val |= BIT(IEEE80211_RADIOTAP_TLV);
 
 	put_unaligned_le32(it_present_val, it_present);
 
@@ -706,22 +691,6 @@ ieee80211_add_rx_radiotap_header(struct ieee80211_local *local,
 		*pos++ = status->chain_signal[chain];
 		*pos++ = chain;
 	}
-
-	if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA) {
-		/* ensure 2 byte alignment for the vendor field as required */
-		if ((pos - (u8 *)rthdr) & 1)
-			*pos++ = 0;
-		*pos++ = rtap.oui[0];
-		*pos++ = rtap.oui[1];
-		*pos++ = rtap.oui[2];
-		*pos++ = rtap.subns;
-		put_unaligned_le16(rtap.len, pos);
-		pos += 2;
-		/* align the actual payload as requested */
-		while ((pos - (u8 *)rthdr) & (rtap.align - 1))
-			*pos++ = 0;
-		/* data (and possible padding) already follows */
-	}
 }
 
 static struct sk_buff *
@@ -797,6 +766,13 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 	bool only_monitor = false;
 	unsigned int min_head_len;
 
+	if (WARN_ON_ONCE(status->flag & RX_FLAG_RADIOTAP_TLV_AT_END &&
+			 !skb_mac_header_was_set(origskb))) {
+		/* with this skb no way to know where frame payload starts */
+		dev_kfree_skb(origskb);
+		return NULL;
+	}
+
 	if (status->flag & RX_FLAG_RADIOTAP_HE)
 		rtap_space += sizeof(struct ieee80211_radiotap_he);
 
@@ -806,12 +782,8 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 	if (status->flag & RX_FLAG_RADIOTAP_LSIG)
 		rtap_space += sizeof(struct ieee80211_radiotap_lsig);
 
-	if (unlikely(status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA)) {
-		struct ieee80211_vendor_radiotap *rtap =
-			(void *)(origskb->data + rtap_space);
-
-		rtap_space += sizeof(*rtap) + rtap->len + rtap->pad;
-	}
+	if (status->flag & RX_FLAG_RADIOTAP_TLV_AT_END)
+		rtap_space += skb_mac_header(origskb) - &origskb->data[rtap_space];
 
 	min_head_len = rtap_space;
 
-- 
cgit v1.2.3-70-g09d2


From 5383bfff5261422f843de72a4089da9b152291b0 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sat, 18 Feb 2023 01:50:05 +0800
Subject: wifi: mac80211: introduce ieee80211_refresh_tx_agg_session_timer()

This allows low level drivers to refresh the tx agg session timer, based on
querying stats from the firmware usually. Especially for some mt76 devices
support .net_fill_forward_path would bypass mac80211, which leads to tx BA
session timeout clients that set a timeout in their AddBA response to our
request, even if our request is without a timeout.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/7c3f72eac1c34921cd84a462e60d71e125862152.1676616450.git.ryder.lee@mediatek.com
[slightly clarify commit message, add note about RCU]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h | 14 ++++++++++++++
 net/mac80211/agg-tx.c  | 17 +++++++++++++++++
 2 files changed, 31 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 3a43ce5fd4ec..722b99e54ad2 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -5973,6 +5973,20 @@ void ieee80211_queue_delayed_work(struct ieee80211_hw *hw,
 				  struct delayed_work *dwork,
 				  unsigned long delay);
 
+/**
+ * ieee80211_refresh_tx_agg_session_timer - Refresh a tx agg session timer.
+ * @sta: the station for which to start a BA session
+ * @tid: the TID to BA on.
+ *
+ * This function allows low level driver to refresh tx agg session timer
+ * to maintain BA session, the session level will still be managed by the
+ * mac80211.
+ *
+ * Note: must be called in an RCU critical section.
+ */
+void ieee80211_refresh_tx_agg_session_timer(struct ieee80211_sta *sta,
+					    u16 tid);
+
 /**
  * ieee80211_start_tx_ba_session - Start a tx Block Ack session.
  * @sta: the station for which to start a BA session
diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index f9514bacbd4a..3b651e7f5a73 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -554,6 +554,23 @@ void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 	ieee80211_send_addba_with_timeout(sta, tid_tx);
 }
 
+void ieee80211_refresh_tx_agg_session_timer(struct ieee80211_sta *pubsta,
+					    u16 tid)
+{
+	struct sta_info *sta = container_of(pubsta, struct sta_info, sta);
+	struct tid_ampdu_tx *tid_tx;
+
+	if (WARN_ON_ONCE(tid >= IEEE80211_NUM_TIDS))
+		return;
+
+	tid_tx = rcu_dereference(sta->ampdu_mlme.tid_tx[tid]);
+	if (!tid_tx)
+		return;
+
+	tid_tx->last_tx = jiffies;
+}
+EXPORT_SYMBOL(ieee80211_refresh_tx_agg_session_timer);
+
 /*
  * After accepting the AddBA Response we activated a timer,
  * resetting it after each frame that we send.
-- 
cgit v1.2.3-70-g09d2


From f4d1181e4759c9c6c97c86cda2cf2d1ddb6a74d2 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sat, 18 Feb 2023 01:48:59 +0800
Subject: wifi: mac80211: add EHT MU-MIMO related flags in ieee80211_bss_conf

Similar to VHT/HE. This is utilized to pass MU-MIMO configurations
from user space (i.e. hostapd) to driver.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/8d9966c4c1e77cb1ade77d42bdc49905609192e9.1676628065.git.ryder.lee@mediatek.com
[move into combined if statement, reset on !eht]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h |  9 +++++++++
 net/mac80211/cfg.c     | 16 ++++++++++++++++
 2 files changed, 25 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 722b99e54ad2..410cd3daaa59 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -674,6 +674,12 @@ struct ieee80211_fils_discovery {
  * @he_full_ul_mumimo: does this BSS support the reception (AP) or transmission
  *	(non-AP STA) of an HE TB PPDU on an RU that spans the entire PPDU
  *	bandwidth
+ * @eht_su_beamformer: in AP-mode, does this BSS enable operation as an EHT SU
+ *	beamformer
+ * @eht_su_beamformee: in AP-mode, does this BSS enable operation as an EHT SU
+ *	beamformee
+ * @eht_mu_beamformer: in AP-mode, does this BSS enable operation as an EHT MU
+ *	beamformer
  */
 struct ieee80211_bss_conf {
 	struct ieee80211_vif *vif;
@@ -761,6 +767,9 @@ struct ieee80211_bss_conf {
 	bool he_su_beamformee;
 	bool he_mu_beamformer;
 	bool he_full_ul_mumimo;
+	bool eht_su_beamformer;
+	bool eht_su_beamformee;
+	bool eht_mu_beamformer;
 };
 
 /**
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 2c8fa19c2f24..9789008626a5 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1299,6 +1299,22 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev,
 	if (params->eht_cap) {
 		link_conf->eht_puncturing = params->punct_bitmap;
 		changed |= BSS_CHANGED_EHT_PUNCTURING;
+
+		link_conf->eht_su_beamformer =
+			params->eht_cap->fixed.phy_cap_info[0] &
+				IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER;
+		link_conf->eht_su_beamformee =
+			params->eht_cap->fixed.phy_cap_info[0] &
+				IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE;
+		link_conf->eht_mu_beamformer =
+			params->eht_cap->fixed.phy_cap_info[7] &
+				(IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
+				 IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ |
+				 IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ);
+	} else {
+		link_conf->eht_su_beamformer = false;
+		link_conf->eht_su_beamformee = false;
+		link_conf->eht_mu_beamformer = false;
 	}
 
 	if (sdata->vif.type == NL80211_IFTYPE_AP &&
-- 
cgit v1.2.3-70-g09d2


From 2ad7dd9425408bf0ca524102808059c05bad169e Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sat, 18 Feb 2023 01:49:25 +0800
Subject: wifi: mac80211: add LDPC related flags in ieee80211_bss_conf

This is utilized to pass LDPC configurations from user space
(i.e. hostapd) to driver.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/1de696aaa34efd77a926eb657b8c0fda05aaa177.1676628065.git.ryder.lee@mediatek.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h |  6 ++++++
 net/mac80211/cfg.c     | 11 +++++++++++
 2 files changed, 17 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 410cd3daaa59..f12edca660ba 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -657,6 +657,9 @@ struct ieee80211_fils_discovery {
  *	write-protected by sdata_lock and local->mtx so holding either is fine
  *	for read access.
  * @color_change_color: the bss color that will be used after the change.
+ * @ht_ldpc: in AP mode, indicates interface has HT LDPC capability.
+ * @vht_ldpc: in AP mode, indicates interface has VHT LDPC capability.
+ * @he_ldpc: in AP mode, indicates interface has HE LDPC capability.
  * @vht_su_beamformer: in AP mode, does this BSS support operation as an VHT SU
  *	beamformer
  * @vht_su_beamformee: in AP mode, does this BSS support operation as an VHT SU
@@ -759,6 +762,9 @@ struct ieee80211_bss_conf {
 	bool color_change_active;
 	u8 color_change_color;
 
+	bool ht_ldpc;
+	bool vht_ldpc;
+	bool he_ldpc;
 	bool vht_su_beamformer;
 	bool vht_su_beamformee;
 	bool vht_mu_beamformer;
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 9789008626a5..760ad934f9e1 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1252,7 +1252,15 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev,
 	prev_beacon_int = link_conf->beacon_int;
 	link_conf->beacon_int = params->beacon_interval;
 
+	if (params->ht_cap)
+		link_conf->ht_ldpc =
+			params->ht_cap->cap_info &
+				cpu_to_le16(IEEE80211_HT_CAP_LDPC_CODING);
+
 	if (params->vht_cap) {
+		link_conf->vht_ldpc =
+			params->vht_cap->vht_cap_info &
+				cpu_to_le32(IEEE80211_VHT_CAP_RXLDPC);
 		link_conf->vht_su_beamformer =
 			params->vht_cap->vht_cap_info &
 				cpu_to_le32(IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE);
@@ -1282,6 +1290,9 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev,
 	}
 
 	if (params->he_cap) {
+		link_conf->he_ldpc =
+			params->he_cap->phy_cap_info[1] &
+				IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD;
 		link_conf->he_su_beamformer =
 			params->he_cap->phy_cap_info[3] &
 				IEEE80211_HE_PHY_CAP3_SU_BEAMFORMER;
-- 
cgit v1.2.3-70-g09d2


From 6933486133ecf71bbe273d7ac72cfc4a51286af3 Mon Sep 17 00:00:00 2001
From: Veerendranath Jakkam <quic_vjakkam@quicinc.com>
Date: Thu, 12 Jan 2023 06:54:13 +0530
Subject: wifi: nl80211: Add support for randomizing TA of auth and deauth
 frames

Add support to use a random local address in authentication and
deauthentication frames sent to unassociated peer when the driver
supports.

The driver needs to configure receive behavior to accept frames with
random transmit address specified in TX path authentication frames
during the time of the frame exchange is pending and such frames need to
be acknowledged similarly to frames sent to the local permanent address
when this random address functionality is used.

This capability allows use of randomized transmit address for PASN
authentication frames to improve privacy of WLAN clients.

Signed-off-by: Veerendranath Jakkam <quic_vjakkam@quicinc.com>
Link: https://lore.kernel.org/r/20230112012415.167556-2-quic_vjakkam@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/uapi/linux/nl80211.h |  5 ++++
 net/wireless/mlme.c          | 55 +++++++++++++++++++++++++++++---------------
 2 files changed, 41 insertions(+), 19 deletions(-)

diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index c8520c150f9c..9a0ac0363f1f 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -6348,6 +6348,10 @@ enum nl80211_feature_flags {
  * @NL80211_EXT_FEATURE_SECURE_NAN: Device supports NAN Pairing which enables
  *	authentication, data encryption and message integrity.
  *
+ * @NL80211_EXT_FEATURE_AUTH_AND_DEAUTH_RANDOM_TA: Device supports randomized TA
+ *	in authentication and deauthentication frames sent to unassociated peer
+ *	using @NL80211_CMD_FRAME.
+ *
  * @NUM_NL80211_EXT_FEATURES: number of extended features.
  * @MAX_NL80211_EXT_FEATURES: highest extended feature index.
  */
@@ -6418,6 +6422,7 @@ enum nl80211_ext_feature_index {
 	NL80211_EXT_FEATURE_POWERED_ADDR_CHANGE,
 	NL80211_EXT_FEATURE_PUNCT,
 	NL80211_EXT_FEATURE_SECURE_NAN,
+	NL80211_EXT_FEATURE_AUTH_AND_DEAUTH_RANDOM_TA,
 
 	/* add new features before the definition below */
 	NUM_NL80211_EXT_FEATURES,
diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c
index 81d3f40d6235..ac059cefbeb3 100644
--- a/net/wireless/mlme.c
+++ b/net/wireless/mlme.c
@@ -673,6 +673,39 @@ static bool cfg80211_allowed_address(struct wireless_dev *wdev, const u8 *addr)
 	return ether_addr_equal(addr, wdev_address(wdev));
 }
 
+static bool cfg80211_allowed_random_address(struct wireless_dev *wdev,
+					    const struct ieee80211_mgmt *mgmt)
+{
+	if (ieee80211_is_auth(mgmt->frame_control) ||
+	    ieee80211_is_deauth(mgmt->frame_control)) {
+		/* Allow random TA to be used with authentication and
+		 * deauthentication frames if the driver has indicated support.
+		 */
+		if (wiphy_ext_feature_isset(
+			    wdev->wiphy,
+			    NL80211_EXT_FEATURE_AUTH_AND_DEAUTH_RANDOM_TA))
+			return true;
+	} else if (ieee80211_is_action(mgmt->frame_control) &&
+		   mgmt->u.action.category == WLAN_CATEGORY_PUBLIC) {
+		/* Allow random TA to be used with Public Action frames if the
+		 * driver has indicated support.
+		 */
+		if (!wdev->connected &&
+		    wiphy_ext_feature_isset(
+			    wdev->wiphy,
+			    NL80211_EXT_FEATURE_MGMT_TX_RANDOM_TA))
+			return true;
+
+		if (wdev->connected &&
+		    wiphy_ext_feature_isset(
+			    wdev->wiphy,
+			    NL80211_EXT_FEATURE_MGMT_TX_RANDOM_TA_CONNECTED))
+			return true;
+	}
+
+	return false;
+}
+
 int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev,
 			  struct wireless_dev *wdev,
 			  struct cfg80211_mgmt_tx_params *params, u64 *cookie)
@@ -774,25 +807,9 @@ int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev,
 			return err;
 	}
 
-	if (!cfg80211_allowed_address(wdev, mgmt->sa)) {
-		/* Allow random TA to be used with Public Action frames if the
-		 * driver has indicated support for this. Otherwise, only allow
-		 * the local address to be used.
-		 */
-		if (!ieee80211_is_action(mgmt->frame_control) ||
-		    mgmt->u.action.category != WLAN_CATEGORY_PUBLIC)
-			return -EINVAL;
-		if (!wdev->connected &&
-		    !wiphy_ext_feature_isset(
-			    &rdev->wiphy,
-			    NL80211_EXT_FEATURE_MGMT_TX_RANDOM_TA))
-			return -EINVAL;
-		if (wdev->connected &&
-		    !wiphy_ext_feature_isset(
-			    &rdev->wiphy,
-			    NL80211_EXT_FEATURE_MGMT_TX_RANDOM_TA_CONNECTED))
-			return -EINVAL;
-	}
+	if (!cfg80211_allowed_address(wdev, mgmt->sa) &&
+	    !cfg80211_allowed_random_address(wdev, mgmt))
+		return -EINVAL;
 
 	/* Transmit the management frame as requested by user space */
 	return rdev_mgmt_tx(rdev, wdev, params, cookie);
-- 
cgit v1.2.3-70-g09d2


From 2d5bcdcda8799cf21f9ab84598c946dd320207a2 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Tue, 7 Mar 2023 08:14:06 -0700
Subject: bpf: Increase size of BTF_ID_LIST without CONFIG_DEBUG_INFO_BTF again

After commit 66e3a13e7c2c ("bpf: Add bpf_dynptr_slice and bpf_dynptr_slice_rdwr"),
clang builds without CONFIG_DEBUG_INFO_BTF warn:

  kernel/bpf/verifier.c:10298:24: warning: array index 16 is past the end of the array (that has type 'u32[16]' (aka 'unsigned int[16]')) [-Warray-bounds]
                                     meta.func_id == special_kfunc_list[KF_bpf_dynptr_slice_rdwr]) {
                                                     ^                  ~~~~~~~~~~~~~~~~~~~~~~~~
  kernel/bpf/verifier.c:9150:1: note: array 'special_kfunc_list' declared here
  BTF_ID_LIST(special_kfunc_list)
  ^
  include/linux/btf_ids.h:207:27: note: expanded from macro 'BTF_ID_LIST'
  #define BTF_ID_LIST(name) static u32 __maybe_unused name[16];
                            ^
  1 warning generated.

A warning of this nature was previously addressed by
commit beb3d47d1d3d ("bpf: Fix a BTF_ID_LIST bug with CONFIG_DEBUG_INFO_BTF not set")
but there have been new kfuncs added since then.

Quadruple the size of the CONFIG_DEBUG_INFO_BTF=n definition so that
this problem is unlikely to show up for some time.

Link: https://github.com/ClangBuiltLinux/linux/issues/1810
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Tom Rix <trix@redhat.com>
Link: https://lore.kernel.org/r/20230307-bpf-kfuncs-warray-bounds-v1-1-00ad3191f3a6@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/btf_ids.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/btf_ids.h b/include/linux/btf_ids.h
index 3a4f7cd882ca..00950cc03bff 100644
--- a/include/linux/btf_ids.h
+++ b/include/linux/btf_ids.h
@@ -204,7 +204,7 @@ extern struct btf_id_set8 name;
 
 #else
 
-#define BTF_ID_LIST(name) static u32 __maybe_unused name[16];
+#define BTF_ID_LIST(name) static u32 __maybe_unused name[64];
 #define BTF_ID(prefix, name)
 #define BTF_ID_FLAGS(prefix, name, ...)
 #define BTF_ID_UNUSED
-- 
cgit v1.2.3-70-g09d2


From 14e05beb10326eeb997ec5ebbf10edf08e85c3f8 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 28 Feb 2023 08:28:27 -0800
Subject: wifi: nl80211: convert cfg80211_scan_request allocation to *_size
 macros

The cfg80211_scan_request structure is followed by a flexible array member
as well as several other arrays that are then stored into pointers in the
structure. These are allocated currently using a simple sequence of
multiplications.

Replace the calculations with struct_size(), size_add(), and array_size()
helper macros. These macros saturate the calculation at SIZE_MAX rather
than overflowing.

Note that we can't use flex_array_size() instead of array_size() because
the fields are not arrays, but simple pointers.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20230228162827.3876606-3-jacob.e.keller@intel.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/nl80211.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 85f714e1af87..0a31b1d2845d 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -9027,7 +9027,7 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info)
 	struct nlattr *attr;
 	struct wiphy *wiphy;
 	int err, tmp, n_ssids = 0, n_channels, i;
-	size_t ie_len;
+	size_t ie_len, size;
 
 	wiphy = &rdev->wiphy;
 
@@ -9072,10 +9072,10 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info)
 	if (ie_len > wiphy->max_scan_ie_len)
 		return -EINVAL;
 
-	request = kzalloc(sizeof(*request)
-			+ sizeof(*request->ssids) * n_ssids
-			+ sizeof(*request->channels) * n_channels
-			+ ie_len, GFP_KERNEL);
+	size = struct_size(request, channels, n_channels);
+	size = size_add(size, array_size(sizeof(*request->ssids), n_ssids));
+	size = size_add(size, ie_len);
+	request = kzalloc(size, GFP_KERNEL);
 	if (!request)
 		return -ENOMEM;
 
@@ -9408,7 +9408,7 @@ nl80211_parse_sched_scan(struct wiphy *wiphy, struct wireless_dev *wdev,
 	struct nlattr *attr;
 	int err, tmp, n_ssids = 0, n_match_sets = 0, n_channels, i, n_plans = 0;
 	enum nl80211_band band;
-	size_t ie_len;
+	size_t ie_len, size;
 	struct nlattr *tb[NL80211_SCHED_SCAN_MATCH_ATTR_MAX + 1];
 	s32 default_match_rssi = NL80211_SCAN_RSSI_THOLD_OFF;
 
@@ -9517,12 +9517,14 @@ nl80211_parse_sched_scan(struct wiphy *wiphy, struct wireless_dev *wdev,
 	     attrs[NL80211_ATTR_SCHED_SCAN_RSSI_ADJUST]))
 		return ERR_PTR(-EINVAL);
 
-	request = kzalloc(sizeof(*request)
-			+ sizeof(*request->ssids) * n_ssids
-			+ sizeof(*request->match_sets) * n_match_sets
-			+ sizeof(*request->scan_plans) * n_plans
-			+ sizeof(*request->channels) * n_channels
-			+ ie_len, GFP_KERNEL);
+	size = struct_size(request, channels, n_channels);
+	size = size_add(size, array_size(sizeof(*request->ssids), n_ssids));
+	size = size_add(size, array_size(sizeof(*request->match_sets),
+					 n_match_sets));
+	size = size_add(size, array_size(sizeof(*request->scan_plans),
+					 n_plans));
+	size = size_add(size, ie_len);
+	request = kzalloc(size, GFP_KERNEL);
 	if (!request)
 		return ERR_PTR(-ENOMEM);
 
-- 
cgit v1.2.3-70-g09d2


From eccfe0176cc11432f1bc7e8b035164d6263df715 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:15 +0200
Subject: wifi: iwlwifi: mvm: add LSIG info to radio tap info in EHT

Nothing change with LSIG in EHT so just extend the switch case to
include EHT standard.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124406.53edcfc049fb.Iec0401faead108aa5130c23c604cfb88ef2943de@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index d1769464d75b..3d7717fdb4d5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1594,6 +1594,10 @@ static void iwl_mvm_decode_lsig(struct sk_buff *skb,
 	case IWL_RX_PHY_INFO_TYPE_HE_MU:
 	case IWL_RX_PHY_INFO_TYPE_HE_MU_EXT:
 	case IWL_RX_PHY_INFO_TYPE_HE_TB:
+	case IWL_RX_PHY_INFO_TYPE_EHT_MU:
+	case IWL_RX_PHY_INFO_TYPE_EHT_TB:
+	case IWL_RX_PHY_INFO_TYPE_EHT_MU_EXT:
+	case IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT:
 		lsig = skb_put(skb, sizeof(*lsig));
 		lsig->data1 = cpu_to_le16(IEEE80211_RADIOTAP_LSIG_DATA1_LENGTH_KNOWN);
 		lsig->data2 = le16_encode_bits(le32_get_bits(phy_data->d1,
-- 
cgit v1.2.3-70-g09d2


From f567b9b46c9e8ecc12190a109ed16006c9f441fa Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:16 +0200
Subject: wifi: iwlwifi: mvm: mark mac header with no data frames

Although no data is presented in the skb, but upper layers need it for
calculating where radio tap header are done, so we mark it.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124406.8f53cc8579bf.I8819c005b4953cea03346d0aff82b31d94f79fe5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 3d7717fdb4d5..ba9f79ada82f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -2147,11 +2147,8 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 	 *
 	 * We mark it as mac header, for upper layers to know where
 	 * all radio tap header ends.
-	 *
-	 * Since data doesn't move data while putting data on skb and that is
-	 * the only way we use, data + len is the next place that hdr would be put
 	 */
-	skb_set_mac_header(skb, skb->len);
+	skb_reset_mac_header(skb);
 
 	/*
 	 * Override the nss from the rx_vec since the rate_n_flags has
-- 
cgit v1.2.3-70-g09d2


From 7f165fdf2967a5b7aa4b123cc9586e9f5068325b Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Sun, 5 Mar 2023 14:16:17 +0200
Subject: wifi: iwlwifi: Adding the code to get RF name for MsP device

Add missing RF name extraction for MsP device from hardware rf-id.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124406.ae831f21ec55.Iad71cc30f9930f7ebe079d210ebe6c968a159273@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-csr.h         | 1 +
 drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c | 5 ++++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index 3e1f011e93aa..bece76b1a514 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -348,6 +348,7 @@ enum {
 #define CSR_HW_RF_ID_TYPE_HRCDB		(0x00109F00)
 #define CSR_HW_RF_ID_TYPE_GF		(0x0010D000)
 #define CSR_HW_RF_ID_TYPE_GF4		(0x0010E000)
+#define CSR_HW_RF_ID_TYPE_MS		(0x00111000)
 
 /* HW_RF CHIP STEP  */
 #define CSR_HW_RF_STEP(_val) (((_val) >> 8) & 0xF)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index 94f40c4d2421..1e263154e9eb 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2021 Intel Corporation
+ * Copyright (C) 2018-2022 Intel Corporation
  */
 #include "iwl-trans.h"
 #include "iwl-prph.h"
@@ -277,6 +277,9 @@ static void iwl_pcie_get_rf_name(struct iwl_trans *trans)
 	case CSR_HW_RFID_TYPE(CSR_HW_RF_ID_TYPE_HRCDB):
 		pos = scnprintf(buf, buflen, "HRCDB");
 		break;
+	case CSR_HW_RFID_TYPE(CSR_HW_RF_ID_TYPE_MS):
+		pos = scnprintf(buf, buflen, "MS");
+		break;
 	default:
 		return;
 	}
-- 
cgit v1.2.3-70-g09d2


From 876882b51569bf686ae4c421b0a738b78a536aa3 Mon Sep 17 00:00:00 2001
From: Golan Ben Ami <golan.ben.ami@intel.com>
Date: Sun, 5 Mar 2023 14:16:18 +0200
Subject: wifi: iwlwifi: reduce verbosity of some logging events

These are cases in which we'd like to warn that something
unexpected happened but they may not be errors.

Reduce verbosity.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.5eea0f58a74f.Ifb6b35903a5a452a757bfe50b6a7e58b3fd2ef23@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 6 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c      | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c    | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 48e7376a5fea..8e0bc1f5f6c6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -350,9 +350,9 @@ void iwl_dbg_tlv_alloc(struct iwl_trans *trans, const struct iwl_ucode_tlv *tlv,
 
 	ret = dbg_tlv_alloc[tlv_idx](trans, tlv);
 	if (ret) {
-		IWL_ERR(trans,
-			"WRT: Failed to allocate TLV 0x%x, ret %d, (ext=%d)\n",
-			type, ret, ext);
+		IWL_WARN(trans,
+			 "WRT: Failed to allocate TLV 0x%x, ret %d, (ext=%d)\n",
+			 type, ret, ext);
 		goto out_err;
 	}
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index 49ca1e168fc5..ab9a51375c8a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -190,7 +190,7 @@ static u32 iwl_mvm_set_mac80211_rx_flag(struct iwl_mvm *mvm,
 	default:
 		/* Expected in monitor (not having the keys) */
 		if (!mvm->monitor_on)
-			IWL_ERR(mvm, "Unhandled alg: 0x%x\n", rx_pkt_status);
+			IWL_WARN(mvm, "Unhandled alg: 0x%x\n", rx_pkt_status);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index ba9f79ada82f..2db4f68becff 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -444,7 +444,7 @@ static int iwl_mvm_rx_crypto(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		 */
 		if (!is_multicast_ether_addr(hdr->addr1) &&
 		    !mvm->monitor_on && net_ratelimit())
-			IWL_ERR(mvm, "Unhandled alg: 0x%x\n", status);
+			IWL_WARN(mvm, "Unhandled alg: 0x%x\n", status);
 	}
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 90a5527d7686d3ebe0dd2a831356a6c7d7dc31bc Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:45:58 +0000
Subject: bpf: add new map ops ->map_mem_usage

Add a new map ops ->map_mem_usage to print the memory usage of a
bpf map.

This is a preparation for the followup change.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-2-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  |  2 ++
 kernel/bpf/syscall.c | 15 +++++++--------
 2 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index d3456804f7aa..9059520bbb5e 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -161,6 +161,8 @@ struct bpf_map_ops {
 				     bpf_callback_t callback_fn,
 				     void *callback_ctx, u64 flags);
 
+	u64 (*map_mem_usage)(const struct bpf_map *map);
+
 	/* BTF id of struct allocated by map_alloc */
 	int *map_btf_id;
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index ede5f987484f..7c96e68859ec 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -771,16 +771,15 @@ static fmode_t map_get_sys_perms(struct bpf_map *map, struct fd f)
 }
 
 #ifdef CONFIG_PROC_FS
-/* Provides an approximation of the map's memory footprint.
- * Used only to provide a backward compatibility and display
- * a reasonable "memlock" info.
- */
-static unsigned long bpf_map_memory_footprint(const struct bpf_map *map)
+/* Show the memory usage of a bpf map */
+static u64 bpf_map_memory_usage(const struct bpf_map *map)
 {
 	unsigned long size;
 
-	size = round_up(map->key_size + bpf_map_value_size(map), 8);
+	if (map->ops->map_mem_usage)
+		return map->ops->map_mem_usage(map);
 
+	size = round_up(map->key_size + bpf_map_value_size(map), 8);
 	return round_up(map->max_entries * size, PAGE_SIZE);
 }
 
@@ -803,7 +802,7 @@ static void bpf_map_show_fdinfo(struct seq_file *m, struct file *filp)
 		   "max_entries:\t%u\n"
 		   "map_flags:\t%#x\n"
 		   "map_extra:\t%#llx\n"
-		   "memlock:\t%lu\n"
+		   "memlock:\t%llu\n"
 		   "map_id:\t%u\n"
 		   "frozen:\t%u\n",
 		   map->map_type,
@@ -812,7 +811,7 @@ static void bpf_map_show_fdinfo(struct seq_file *m, struct file *filp)
 		   map->max_entries,
 		   map->map_flags,
 		   (unsigned long long)map->map_extra,
-		   bpf_map_memory_footprint(map),
+		   bpf_map_memory_usage(map),
 		   map->id,
 		   READ_ONCE(map->frozen));
 	if (type) {
-- 
cgit v1.2.3-70-g09d2


From 41d5941e7f9a27f3d28220f08b843e8d87df1be4 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:45:59 +0000
Subject: bpf: lpm_trie memory usage

trie_mem_usage() is introduced to calculate the lpm_trie memory usage.
Some small memory allocations are ignored. The inner node is also
ignored.

The result as follows,

- before
10: lpm_trie  flags 0x1
        key 8B  value 8B  max_entries 65536  memlock 1048576B

- after
10: lpm_trie  flags 0x1
        key 8B  value 8B  max_entries 65536  memlock 2291536B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-3-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/lpm_trie.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c
index d833496e9e42..dc23f2ac9cde 100644
--- a/kernel/bpf/lpm_trie.c
+++ b/kernel/bpf/lpm_trie.c
@@ -720,6 +720,16 @@ static int trie_check_btf(const struct bpf_map *map,
 	       -EINVAL : 0;
 }
 
+static u64 trie_mem_usage(const struct bpf_map *map)
+{
+	struct lpm_trie *trie = container_of(map, struct lpm_trie, map);
+	u64 elem_size;
+
+	elem_size = sizeof(struct lpm_trie_node) + trie->data_size +
+			    trie->map.value_size;
+	return elem_size * READ_ONCE(trie->n_entries);
+}
+
 BTF_ID_LIST_SINGLE(trie_map_btf_ids, struct, lpm_trie)
 const struct bpf_map_ops trie_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -733,5 +743,6 @@ const struct bpf_map_ops trie_map_ops = {
 	.map_update_batch = generic_map_update_batch,
 	.map_delete_batch = generic_map_delete_batch,
 	.map_check_btf = trie_check_btf,
+	.map_mem_usage = trie_mem_usage,
 	.map_btf_id = &trie_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From 304849a27b341cf22d5c15096144cc8c1b69e0c0 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:00 +0000
Subject: bpf: hashtab memory usage

htab_map_mem_usage() is introduced to calculate hashmap memory usage. In
this helper, some small memory allocations are ignore, as their size is
quite small compared with the total size. The inner_map_meta in
hash_of_map is also ignored.

The result for hashtab as follows,

- before this change
1: hash  name count_map  flags 0x1  <<<< no prealloc, fully set
        key 16B  value 24B  max_entries 1048576  memlock 41943040B
2: hash  name count_map  flags 0x1  <<<< no prealloc, none set
        key 16B  value 24B  max_entries 1048576  memlock 41943040B
3: hash  name count_map  flags 0x0  <<<< prealloc
        key 16B  value 24B  max_entries 1048576  memlock 41943040B

The memlock is always a fixed size whatever it is preallocated or
not, and whatever the count of allocated elements is.

- after this change
1: hash  name count_map  flags 0x1    <<<< non prealloc, fully set
        key 16B  value 24B  max_entries 1048576  memlock 117441536B
2: hash  name count_map  flags 0x1    <<<< non prealloc, non set
        key 16B  value 24B  max_entries 1048576  memlock 16778240B
3: hash  name count_map  flags 0x0    <<<< prealloc
        key 16B  value 24B  max_entries 1048576  memlock 109056000B

The memlock now is hashtab actually allocated.

The result for percpu hash map as follows,
- before this change
4: percpu_hash  name count_map  flags 0x0       <<<< prealloc
        key 16B  value 24B  max_entries 1048576  memlock 822083584B
5: percpu_hash  name count_map  flags 0x1       <<<< no prealloc
        key 16B  value 24B  max_entries 1048576  memlock 822083584B

- after this change
4: percpu_hash  name count_map  flags 0x0
        key 16B  value 24B  max_entries 1048576  memlock 897582080B
5: percpu_hash  name count_map  flags 0x1
        key 16B  value 24B  max_entries 1048576  memlock 922748736B

At worst, the difference can be 10x, for example,
- before this change
6: hash  name count_map  flags 0x0
        key 4B  value 4B  max_entries 1048576  memlock 8388608B

- after this change
6: hash  name count_map  flags 0x0
        key 4B  value 4B  max_entries 1048576  memlock 83889408B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Hou Tao <houtao1@huawei.com>
Link: https://lore.kernel.org/r/20230305124615.12358-4-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/hashtab.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 653aeb481c79..0df4b0c10f59 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -2190,6 +2190,44 @@ out:
 	return num_elems;
 }
 
+static u64 htab_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
+	u32 value_size = round_up(htab->map.value_size, 8);
+	bool prealloc = htab_is_prealloc(htab);
+	bool percpu = htab_is_percpu(htab);
+	bool lru = htab_is_lru(htab);
+	u64 num_entries;
+	u64 usage = sizeof(struct bpf_htab);
+
+	usage += sizeof(struct bucket) * htab->n_buckets;
+	usage += sizeof(int) * num_possible_cpus() * HASHTAB_MAP_LOCK_COUNT;
+	if (prealloc) {
+		num_entries = map->max_entries;
+		if (htab_has_extra_elems(htab))
+			num_entries += num_possible_cpus();
+
+		usage += htab->elem_size * num_entries;
+
+		if (percpu)
+			usage += value_size * num_possible_cpus() * num_entries;
+		else if (!lru)
+			usage += sizeof(struct htab_elem *) * num_possible_cpus();
+	} else {
+#define LLIST_NODE_SZ sizeof(struct llist_node)
+
+		num_entries = htab->use_percpu_counter ?
+					  percpu_counter_sum(&htab->pcount) :
+					  atomic_read(&htab->count);
+		usage += (htab->elem_size + LLIST_NODE_SZ) * num_entries;
+		if (percpu) {
+			usage += (LLIST_NODE_SZ + sizeof(void *)) * num_entries;
+			usage += value_size * num_possible_cpus() * num_entries;
+		}
+	}
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(htab_map_btf_ids, struct, bpf_htab)
 const struct bpf_map_ops htab_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -2206,6 +2244,7 @@ const struct bpf_map_ops htab_map_ops = {
 	.map_seq_show_elem = htab_map_seq_show_elem,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_hash_elem,
+	.map_mem_usage = htab_map_mem_usage,
 	BATCH_OPS(htab),
 	.map_btf_id = &htab_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
@@ -2227,6 +2266,7 @@ const struct bpf_map_ops htab_lru_map_ops = {
 	.map_seq_show_elem = htab_map_seq_show_elem,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_hash_elem,
+	.map_mem_usage = htab_map_mem_usage,
 	BATCH_OPS(htab_lru),
 	.map_btf_id = &htab_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
@@ -2378,6 +2418,7 @@ const struct bpf_map_ops htab_percpu_map_ops = {
 	.map_seq_show_elem = htab_percpu_map_seq_show_elem,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_hash_elem,
+	.map_mem_usage = htab_map_mem_usage,
 	BATCH_OPS(htab_percpu),
 	.map_btf_id = &htab_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
@@ -2397,6 +2438,7 @@ const struct bpf_map_ops htab_lru_percpu_map_ops = {
 	.map_seq_show_elem = htab_percpu_map_seq_show_elem,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_hash_elem,
+	.map_mem_usage = htab_map_mem_usage,
 	BATCH_OPS(htab_lru_percpu),
 	.map_btf_id = &htab_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
@@ -2534,6 +2576,7 @@ const struct bpf_map_ops htab_of_maps_map_ops = {
 	.map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem,
 	.map_gen_lookup = htab_of_map_gen_lookup,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = htab_map_mem_usage,
 	BATCH_OPS(htab),
 	.map_btf_id = &htab_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From 1746d0555a8795c7ecfeab6a2c3e3c824cf57535 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:01 +0000
Subject: bpf: arraymap memory usage

Introduce array_map_mem_usage() to calculate arraymap memory usage. In
this helper, some small memory allocations are ignored, like the
allocation of struct bpf_array_aux in prog_array. The inner_map_meta in
array_of_map is also ignored.

The result as follows,

- before
11: array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B
12: percpu_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 8912896B
13: perf_event_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B
14: prog_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B
15: cgroup_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B

- after
11: array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524608B
12: percpu_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 17301824B
13: perf_event_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524608B
14: prog_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524608B
15: cgroup_array  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524608B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-5-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/arraymap.c | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 484706959556..1588c793a715 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -721,6 +721,28 @@ static int bpf_for_each_array_elem(struct bpf_map *map, bpf_callback_t callback_
 	return num_elems;
 }
 
+static u64 array_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_array *array = container_of(map, struct bpf_array, map);
+	bool percpu = map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY;
+	u32 elem_size = array->elem_size;
+	u64 entries = map->max_entries;
+	u64 usage = sizeof(*array);
+
+	if (percpu) {
+		usage += entries * sizeof(void *);
+		usage += entries * elem_size * num_possible_cpus();
+	} else {
+		if (map->map_flags & BPF_F_MMAPABLE) {
+			usage = PAGE_ALIGN(usage);
+			usage += PAGE_ALIGN(entries * elem_size);
+		} else {
+			usage += entries * elem_size;
+		}
+	}
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(array_map_btf_ids, struct, bpf_array)
 const struct bpf_map_ops array_map_ops = {
 	.map_meta_equal = array_map_meta_equal,
@@ -742,6 +764,7 @@ const struct bpf_map_ops array_map_ops = {
 	.map_update_batch = generic_map_update_batch,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_array_elem,
+	.map_mem_usage = array_map_mem_usage,
 	.map_btf_id = &array_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
 };
@@ -762,6 +785,7 @@ const struct bpf_map_ops percpu_array_map_ops = {
 	.map_update_batch = generic_map_update_batch,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_array_elem,
+	.map_mem_usage = array_map_mem_usage,
 	.map_btf_id = &array_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
 };
@@ -1156,6 +1180,7 @@ const struct bpf_map_ops prog_array_map_ops = {
 	.map_fd_sys_lookup_elem = prog_fd_array_sys_lookup_elem,
 	.map_release_uref = prog_array_map_clear,
 	.map_seq_show_elem = prog_array_map_seq_show_elem,
+	.map_mem_usage = array_map_mem_usage,
 	.map_btf_id = &array_map_btf_ids[0],
 };
 
@@ -1257,6 +1282,7 @@ const struct bpf_map_ops perf_event_array_map_ops = {
 	.map_fd_put_ptr = perf_event_fd_array_put_ptr,
 	.map_release = perf_event_fd_array_release,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = array_map_mem_usage,
 	.map_btf_id = &array_map_btf_ids[0],
 };
 
@@ -1291,6 +1317,7 @@ const struct bpf_map_ops cgroup_array_map_ops = {
 	.map_fd_get_ptr = cgroup_fd_array_get_ptr,
 	.map_fd_put_ptr = cgroup_fd_array_put_ptr,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = array_map_mem_usage,
 	.map_btf_id = &array_map_btf_ids[0],
 };
 #endif
@@ -1379,5 +1406,6 @@ const struct bpf_map_ops array_of_maps_map_ops = {
 	.map_lookup_batch = generic_map_lookup_batch,
 	.map_update_batch = generic_map_update_batch,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = array_map_mem_usage,
 	.map_btf_id = &array_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From cbb9b6068c68332eefb0ef4c0ebf6f96c35d9bde Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:02 +0000
Subject: bpf: stackmap memory usage

A new helper is introduced to get stackmap memory usage. Some small
memory allocations are ignored as their memory size is quite small
compared to the totol usage.

The result as follows,
- before
16: stack_trace  name count_map  flags 0x0
        key 4B  value 8B  max_entries 65536  memlock 1048576B

- after
16: stack_trace  name count_map  flags 0x0
        key 4B  value 8B  max_entries 65536  memlock 2097472B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-6-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/stackmap.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index aecea7451b61..0f1d8dced933 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -654,6 +654,19 @@ static void stack_map_free(struct bpf_map *map)
 	put_callchain_buffers();
 }
 
+static u64 stack_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
+	u64 value_size = map->value_size;
+	u64 n_buckets = smap->n_buckets;
+	u64 enties = map->max_entries;
+	u64 usage = sizeof(*smap);
+
+	usage += n_buckets * sizeof(struct stack_map_bucket *);
+	usage += enties * (sizeof(struct stack_map_bucket) + value_size);
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(stack_trace_map_btf_ids, struct, bpf_stack_map)
 const struct bpf_map_ops stack_trace_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -664,5 +677,6 @@ const struct bpf_map_ops stack_trace_map_ops = {
 	.map_update_elem = stack_map_update_elem,
 	.map_delete_elem = stack_map_delete_elem,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = stack_map_mem_usage,
 	.map_btf_id = &stack_trace_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From 2e89caf055a64cb531cbd8ab7a01c3e9b0b3133d Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:03 +0000
Subject: bpf: reuseport_array memory usage

A new helper is introduced to calculate reuseport_array memory usage.

The result as follows,
- before
14: reuseport_sockarray  name count_map  flags 0x0
        key 4B  value 8B  max_entries 65536  memlock 1048576B

- after
14: reuseport_sockarray  name count_map  flags 0x0
        key 4B  value 8B  max_entries 65536  memlock 524544B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-7-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/reuseport_array.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/kernel/bpf/reuseport_array.c b/kernel/bpf/reuseport_array.c
index 82c61612f382..71cb72f5b733 100644
--- a/kernel/bpf/reuseport_array.c
+++ b/kernel/bpf/reuseport_array.c
@@ -335,6 +335,13 @@ static int reuseport_array_get_next_key(struct bpf_map *map, void *key,
 	return 0;
 }
 
+static u64 reuseport_array_mem_usage(const struct bpf_map *map)
+{
+	struct reuseport_array *array;
+
+	return struct_size(array, ptrs, map->max_entries);
+}
+
 BTF_ID_LIST_SINGLE(reuseport_array_map_btf_ids, struct, reuseport_array)
 const struct bpf_map_ops reuseport_array_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -344,5 +351,6 @@ const struct bpf_map_ops reuseport_array_ops = {
 	.map_lookup_elem = reuseport_array_lookup_elem,
 	.map_get_next_key = reuseport_array_get_next_key,
 	.map_delete_elem = reuseport_array_delete_elem,
+	.map_mem_usage = reuseport_array_mem_usage,
 	.map_btf_id = &reuseport_array_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From 2f7e4ab2caa9a692ffc843f801cea90632cdc782 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:04 +0000
Subject: bpf: ringbuf memory usage

A new helper ringbuf_map_mem_usage() is introduced to calculate ringbuf
memory usage.

The result as follows,
- before
15: ringbuf  name count_map  flags 0x0
        key 0B  value 0B  max_entries 65536  memlock 0B

- after
15: ringbuf  name count_map  flags 0x0
        key 0B  value 0B  max_entries 65536  memlock 78424B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230305124615.12358-8-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/ringbuf.c | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c
index 8732e0aadf36..0d2a45ff83f1 100644
--- a/kernel/bpf/ringbuf.c
+++ b/kernel/bpf/ringbuf.c
@@ -19,6 +19,7 @@
 	(offsetof(struct bpf_ringbuf, consumer_pos) >> PAGE_SHIFT)
 /* consumer page and producer page */
 #define RINGBUF_POS_PAGES 2
+#define RINGBUF_NR_META_PAGES (RINGBUF_PGOFF + RINGBUF_POS_PAGES)
 
 #define RINGBUF_MAX_RECORD_SZ (UINT_MAX/4)
 
@@ -96,7 +97,7 @@ static struct bpf_ringbuf *bpf_ringbuf_area_alloc(size_t data_sz, int numa_node)
 {
 	const gfp_t flags = GFP_KERNEL_ACCOUNT | __GFP_RETRY_MAYFAIL |
 			    __GFP_NOWARN | __GFP_ZERO;
-	int nr_meta_pages = RINGBUF_PGOFF + RINGBUF_POS_PAGES;
+	int nr_meta_pages = RINGBUF_NR_META_PAGES;
 	int nr_data_pages = data_sz >> PAGE_SHIFT;
 	int nr_pages = nr_meta_pages + nr_data_pages;
 	struct page **pages, *page;
@@ -336,6 +337,21 @@ static __poll_t ringbuf_map_poll_user(struct bpf_map *map, struct file *filp,
 	return 0;
 }
 
+static u64 ringbuf_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_ringbuf *rb;
+	int nr_data_pages;
+	int nr_meta_pages;
+	u64 usage = sizeof(struct bpf_ringbuf_map);
+
+	rb = container_of(map, struct bpf_ringbuf_map, map)->rb;
+	usage += (u64)rb->nr_pages << PAGE_SHIFT;
+	nr_meta_pages = RINGBUF_NR_META_PAGES;
+	nr_data_pages = map->max_entries >> PAGE_SHIFT;
+	usage += (nr_meta_pages + 2 * nr_data_pages) * sizeof(struct page *);
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(ringbuf_map_btf_ids, struct, bpf_ringbuf_map)
 const struct bpf_map_ops ringbuf_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -347,6 +363,7 @@ const struct bpf_map_ops ringbuf_map_ops = {
 	.map_update_elem = ringbuf_map_update_elem,
 	.map_delete_elem = ringbuf_map_delete_elem,
 	.map_get_next_key = ringbuf_map_get_next_key,
+	.map_mem_usage = ringbuf_map_mem_usage,
 	.map_btf_id = &ringbuf_map_btf_ids[0],
 };
 
@@ -361,6 +378,7 @@ const struct bpf_map_ops user_ringbuf_map_ops = {
 	.map_update_elem = ringbuf_map_update_elem,
 	.map_delete_elem = ringbuf_map_delete_elem,
 	.map_get_next_key = ringbuf_map_get_next_key,
+	.map_mem_usage = ringbuf_map_mem_usage,
 	.map_btf_id = &user_ringbuf_map_btf_ids[0],
 };
 
-- 
cgit v1.2.3-70-g09d2


From 71a49abe73cb56b15bf0701b5e98b6103480f762 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:05 +0000
Subject: bpf: bloom_filter memory usage

Introduce a new helper to calculate the bloom_filter memory usage.

The result as follows,
- before
16: bloom_filter  flags 0x0
        key 0B  value 8B  max_entries 65536  memlock 524288B

- after
16: bloom_filter  flags 0x0
        key 0B  value 8B  max_entries 65536  memlock 65856B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-9-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bloom_filter.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/kernel/bpf/bloom_filter.c b/kernel/bpf/bloom_filter.c
index 48ee750849f2..6350c5d35a9b 100644
--- a/kernel/bpf/bloom_filter.c
+++ b/kernel/bpf/bloom_filter.c
@@ -193,6 +193,17 @@ static int bloom_map_check_btf(const struct bpf_map *map,
 	return btf_type_is_void(key_type) ? 0 : -EINVAL;
 }
 
+static u64 bloom_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_bloom_filter *bloom;
+	u64 bitset_bytes;
+
+	bloom = container_of(map, struct bpf_bloom_filter, map);
+	bitset_bytes = BITS_TO_BYTES((u64)bloom->bitset_mask + 1);
+	bitset_bytes = roundup(bitset_bytes, sizeof(unsigned long));
+	return sizeof(*bloom) + bitset_bytes;
+}
+
 BTF_ID_LIST_SINGLE(bpf_bloom_map_btf_ids, struct, bpf_bloom_filter)
 const struct bpf_map_ops bloom_filter_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -206,5 +217,6 @@ const struct bpf_map_ops bloom_filter_map_ops = {
 	.map_update_elem = bloom_map_update_elem,
 	.map_delete_elem = bloom_map_delete_elem,
 	.map_check_btf = bloom_map_check_btf,
+	.map_mem_usage = bloom_map_mem_usage,
 	.map_btf_id = &bpf_bloom_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From 835f1fca951303930b2c74c5b0abe3c03a3aeadf Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:06 +0000
Subject: bpf: cpumap memory usage

A new helper is introduced to calculate cpumap memory usage. The size of
cpu_entries can be dynamically changed when we update or delete a cpumap
element, but this patch doesn't include the memory size of cpu_entry
yet. We can dynamically calculate the memory usage when we alloc or free
a cpu_entry, but it will take extra runtime overhead, so let just put it
aside currently. Note that the size of different cpu_entry may be
different as well.

The result as follows,
- before
48: cpumap  name count_map  flags 0x4
        key 4B  value 4B  max_entries 64  memlock 4096B

- after
48: cpumap  name count_map  flags 0x4
        key 4B  value 4B  max_entries 64  memlock 832B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-10-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/cpumap.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index d2110c1f6fa6..871809e71b4e 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -673,6 +673,15 @@ static int cpu_map_redirect(struct bpf_map *map, u64 index, u64 flags)
 				      __cpu_map_lookup_elem);
 }
 
+static u64 cpu_map_mem_usage(const struct bpf_map *map)
+{
+	u64 usage = sizeof(struct bpf_cpu_map);
+
+	/* Currently the dynamically allocated elements are not counted */
+	usage += (u64)map->max_entries * sizeof(struct bpf_cpu_map_entry *);
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(cpu_map_btf_ids, struct, bpf_cpu_map)
 const struct bpf_map_ops cpu_map_ops = {
 	.map_meta_equal		= bpf_map_meta_equal,
@@ -683,6 +692,7 @@ const struct bpf_map_ops cpu_map_ops = {
 	.map_lookup_elem	= cpu_map_lookup_elem,
 	.map_get_next_key	= cpu_map_get_next_key,
 	.map_check_btf		= map_check_no_btf,
+	.map_mem_usage		= cpu_map_mem_usage,
 	.map_btf_id		= &cpu_map_btf_ids[0],
 	.map_redirect		= cpu_map_redirect,
 };
-- 
cgit v1.2.3-70-g09d2


From fa5e83df173beb6c1334737a463fc2b4ef4efa8b Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:07 +0000
Subject: bpf: devmap memory usage

A new helper is introduced to calculate the memory usage of devmap and
devmap_hash. The number of dynamically allocated elements are recored
for devmap_hash already, but not for devmap. To track the memory size of
dynamically allocated elements, this patch also count the numbers for
devmap.

The result as follows,
- before
40: devmap  name count_map  flags 0x80
        key 4B  value 4B  max_entries 65536  memlock 524288B
41: devmap_hash  name count_map  flags 0x80
        key 4B  value 4B  max_entries 65536  memlock 524288B

- after
40: devmap  name count_map  flags 0x80  <<<< no elements
        key 4B  value 4B  max_entries 65536  memlock 524608B
41: devmap_hash  name count_map  flags 0x80 <<<< no elements
        key 4B  value 4B  max_entries 65536  memlock 524608B

Note that the number of buckets is same with max_entries for devmap_hash
in this case.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-11-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/devmap.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 2675fefc6cb6..19b036a228f7 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -819,8 +819,10 @@ static int dev_map_delete_elem(struct bpf_map *map, void *key)
 		return -EINVAL;
 
 	old_dev = unrcu_pointer(xchg(&dtab->netdev_map[k], NULL));
-	if (old_dev)
+	if (old_dev) {
 		call_rcu(&old_dev->rcu, __dev_map_entry_free);
+		atomic_dec((atomic_t *)&dtab->items);
+	}
 	return 0;
 }
 
@@ -931,6 +933,8 @@ static int __dev_map_update_elem(struct net *net, struct bpf_map *map,
 	old_dev = unrcu_pointer(xchg(&dtab->netdev_map[i], RCU_INITIALIZER(dev)));
 	if (old_dev)
 		call_rcu(&old_dev->rcu, __dev_map_entry_free);
+	else
+		atomic_inc((atomic_t *)&dtab->items);
 
 	return 0;
 }
@@ -1016,6 +1020,20 @@ static int dev_hash_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags)
 				      __dev_map_hash_lookup_elem);
 }
 
+static u64 dev_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
+	u64 usage = sizeof(struct bpf_dtab);
+
+	if (map->map_type == BPF_MAP_TYPE_DEVMAP_HASH)
+		usage += (u64)dtab->n_buckets * sizeof(struct hlist_head);
+	else
+		usage += (u64)map->max_entries * sizeof(struct bpf_dtab_netdev *);
+	usage += atomic_read((atomic_t *)&dtab->items) *
+			 (u64)sizeof(struct bpf_dtab_netdev);
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(dev_map_btf_ids, struct, bpf_dtab)
 const struct bpf_map_ops dev_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -1026,6 +1044,7 @@ const struct bpf_map_ops dev_map_ops = {
 	.map_update_elem = dev_map_update_elem,
 	.map_delete_elem = dev_map_delete_elem,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = dev_map_mem_usage,
 	.map_btf_id = &dev_map_btf_ids[0],
 	.map_redirect = dev_map_redirect,
 };
@@ -1039,6 +1058,7 @@ const struct bpf_map_ops dev_map_hash_ops = {
 	.map_update_elem = dev_map_hash_update_elem,
 	.map_delete_elem = dev_map_hash_delete_elem,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = dev_map_mem_usage,
 	.map_btf_id = &dev_map_btf_ids[0],
 	.map_redirect = dev_hash_map_redirect,
 };
@@ -1109,9 +1129,11 @@ static int dev_map_notification(struct notifier_block *notifier,
 				if (!dev || netdev != dev->dev)
 					continue;
 				odev = unrcu_pointer(cmpxchg(&dtab->netdev_map[i], RCU_INITIALIZER(dev), NULL));
-				if (dev == odev)
+				if (dev == odev) {
 					call_rcu(&dev->rcu,
 						 __dev_map_entry_free);
+					atomic_dec((atomic_t *)&dtab->items);
+				}
 			}
 		}
 		rcu_read_unlock();
-- 
cgit v1.2.3-70-g09d2


From c6e66b42a348125fd41131de75f84fa67b6579ce Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:08 +0000
Subject: bpf: queue_stack_maps memory usage

A new helper is introduced to calculate queue_stack_maps memory usage.

The result as follows,

- before
20: queue  name count_map  flags 0x0
        key 0B  value 4B  max_entries 65536  memlock 266240B
21: stack  name count_map  flags 0x0
        key 0B  value 4B  max_entries 65536  memlock 266240B

- after
20: queue  name count_map  flags 0x0
        key 0B  value 4B  max_entries 65536  memlock 524288B
21: stack  name count_map  flags 0x0
        key 0B  value 4B  max_entries 65536  memlock 524288B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-12-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/queue_stack_maps.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/kernel/bpf/queue_stack_maps.c b/kernel/bpf/queue_stack_maps.c
index 8a5e060de63b..63ecbbcb349d 100644
--- a/kernel/bpf/queue_stack_maps.c
+++ b/kernel/bpf/queue_stack_maps.c
@@ -246,6 +246,14 @@ static int queue_stack_map_get_next_key(struct bpf_map *map, void *key,
 	return -EINVAL;
 }
 
+static u64 queue_stack_map_mem_usage(const struct bpf_map *map)
+{
+	u64 usage = sizeof(struct bpf_queue_stack);
+
+	usage += ((u64)map->max_entries + 1) * map->value_size;
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(queue_map_btf_ids, struct, bpf_queue_stack)
 const struct bpf_map_ops queue_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -259,6 +267,7 @@ const struct bpf_map_ops queue_map_ops = {
 	.map_pop_elem = queue_map_pop_elem,
 	.map_peek_elem = queue_map_peek_elem,
 	.map_get_next_key = queue_stack_map_get_next_key,
+	.map_mem_usage = queue_stack_map_mem_usage,
 	.map_btf_id = &queue_map_btf_ids[0],
 };
 
@@ -274,5 +283,6 @@ const struct bpf_map_ops stack_map_ops = {
 	.map_pop_elem = stack_map_pop_elem,
 	.map_peek_elem = stack_map_peek_elem,
 	.map_get_next_key = queue_stack_map_get_next_key,
+	.map_mem_usage = queue_stack_map_mem_usage,
 	.map_btf_id = &queue_map_btf_ids[0],
 };
-- 
cgit v1.2.3-70-g09d2


From f062226d8d59b521ddc946ad791048188a16722a Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:09 +0000
Subject: bpf: bpf_struct_ops memory usage

A new helper is introduced to calculate bpf_struct_ops memory usage.

The result as follows,

- before
1: struct_ops  name count_map  flags 0x0
        key 4B  value 256B  max_entries 1  memlock 4096B
        btf_id 73

- after
1: struct_ops  name count_map  flags 0x0
        key 4B  value 256B  max_entries 1  memlock 5016B
        btf_id 73

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-13-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_struct_ops.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index ece9870cab68..38903fb52f98 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -641,6 +641,21 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 	return map;
 }
 
+static u64 bpf_struct_ops_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
+	const struct bpf_struct_ops *st_ops = st_map->st_ops;
+	const struct btf_type *vt = st_ops->value_type;
+	u64 usage;
+
+	usage = sizeof(*st_map) +
+			vt->size - sizeof(struct bpf_struct_ops_value);
+	usage += vt->size;
+	usage += btf_type_vlen(vt) * sizeof(struct bpf_links *);
+	usage += PAGE_SIZE;
+	return usage;
+}
+
 BTF_ID_LIST_SINGLE(bpf_struct_ops_map_btf_ids, struct, bpf_struct_ops_map)
 const struct bpf_map_ops bpf_struct_ops_map_ops = {
 	.map_alloc_check = bpf_struct_ops_map_alloc_check,
@@ -651,6 +666,7 @@ const struct bpf_map_ops bpf_struct_ops_map_ops = {
 	.map_delete_elem = bpf_struct_ops_map_delete_elem,
 	.map_update_elem = bpf_struct_ops_map_update_elem,
 	.map_seq_show_elem = bpf_struct_ops_map_seq_show_elem,
+	.map_mem_usage = bpf_struct_ops_map_mem_usage,
 	.map_btf_id = &bpf_struct_ops_map_btf_ids[0],
 };
 
-- 
cgit v1.2.3-70-g09d2


From 2f536977d6f1481f0a149140c8311103ddebe36a Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:10 +0000
Subject: bpf: local_storage memory usage

A new helper is introduced to calculate local_storage map memory usage.
Currently the dynamically allocated elements are not counted, since it
will take runtime overhead in the element update or delete path. So
let's put it aside currently, and implement it in the future if the user
really needs it.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-14-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/local_storage.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/kernel/bpf/local_storage.c b/kernel/bpf/local_storage.c
index e90d9f63edc5..a993560f200a 100644
--- a/kernel/bpf/local_storage.c
+++ b/kernel/bpf/local_storage.c
@@ -446,6 +446,12 @@ static void cgroup_storage_seq_show_elem(struct bpf_map *map, void *key,
 	rcu_read_unlock();
 }
 
+static u64 cgroup_storage_map_usage(const struct bpf_map *map)
+{
+	/* Currently the dynamically allocated elements are not counted. */
+	return sizeof(struct bpf_cgroup_storage_map);
+}
+
 BTF_ID_LIST_SINGLE(cgroup_storage_map_btf_ids, struct,
 		   bpf_cgroup_storage_map)
 const struct bpf_map_ops cgroup_storage_map_ops = {
@@ -457,6 +463,7 @@ const struct bpf_map_ops cgroup_storage_map_ops = {
 	.map_delete_elem = cgroup_storage_delete_elem,
 	.map_check_btf = cgroup_storage_check_btf,
 	.map_seq_show_elem = cgroup_storage_seq_show_elem,
+	.map_mem_usage = cgroup_storage_map_usage,
 	.map_btf_id = &cgroup_storage_map_btf_ids[0],
 };
 
-- 
cgit v1.2.3-70-g09d2


From 7490b7f1c02ef825ef98f7230662049d4a464a21 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:11 +0000
Subject: bpf, net: bpf_local_storage memory usage

A new helper is introduced into bpf_local_storage map to calculate the
memory usage. This helper is also used by other maps like
bpf_cgrp_storage, bpf_inode_storage, bpf_task_storage and etc.

Note that currently the dynamically allocated storage elements are not
counted in the usage, since it will take extra runtime overhead in the
elements update or delete path. So let's put it aside now, and implement
it in the future when someone really need it.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-15-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h |  1 +
 kernel/bpf/bpf_cgrp_storage.c     |  1 +
 kernel/bpf/bpf_inode_storage.c    |  1 +
 kernel/bpf/bpf_local_storage.c    | 10 ++++++++++
 kernel/bpf/bpf_task_storage.c     |  1 +
 net/core/bpf_sk_storage.c         |  1 +
 6 files changed, 15 insertions(+)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 6d37a40cd90e..d934248b8e81 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -164,5 +164,6 @@ bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap,
 			 void *value, u64 map_flags, gfp_t gfp_flags);
 
 void bpf_local_storage_free_rcu(struct rcu_head *rcu);
+u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map);
 
 #endif /* _BPF_LOCAL_STORAGE_H */
diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c
index 6cdf6d9ed91d..9ae07aedaf23 100644
--- a/kernel/bpf/bpf_cgrp_storage.c
+++ b/kernel/bpf/bpf_cgrp_storage.c
@@ -221,6 +221,7 @@ const struct bpf_map_ops cgrp_storage_map_ops = {
 	.map_update_elem = bpf_cgrp_storage_update_elem,
 	.map_delete_elem = bpf_cgrp_storage_delete_elem,
 	.map_check_btf = bpf_local_storage_map_check_btf,
+	.map_mem_usage = bpf_local_storage_map_mem_usage,
 	.map_btf_id = &bpf_local_storage_map_btf_id[0],
 	.map_owner_storage_ptr = cgroup_storage_ptr,
 };
diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c
index 05f4c66c9089..43e2619c8167 100644
--- a/kernel/bpf/bpf_inode_storage.c
+++ b/kernel/bpf/bpf_inode_storage.c
@@ -223,6 +223,7 @@ const struct bpf_map_ops inode_storage_map_ops = {
 	.map_update_elem = bpf_fd_inode_storage_update_elem,
 	.map_delete_elem = bpf_fd_inode_storage_delete_elem,
 	.map_check_btf = bpf_local_storage_map_check_btf,
+	.map_mem_usage = bpf_local_storage_map_mem_usage,
 	.map_btf_id = &bpf_local_storage_map_btf_id[0],
 	.map_owner_storage_ptr = inode_storage_ptr,
 };
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 3d320393a12c..d3ba3f2db640 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -685,6 +685,16 @@ bool bpf_local_storage_unlink_nolock(struct bpf_local_storage *local_storage)
 	return free_storage;
 }
 
+u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_local_storage_map *smap = (struct bpf_local_storage_map *)map;
+	u64 usage = sizeof(*smap);
+
+	/* The dynamically callocated selems are not counted currently. */
+	usage += sizeof(*smap->buckets) * (1ULL << smap->bucket_log);
+	return usage;
+}
+
 struct bpf_map *
 bpf_local_storage_map_alloc(union bpf_attr *attr,
 			    struct bpf_local_storage_cache *cache)
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index 1e486055a523..20f942229f3c 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -335,6 +335,7 @@ const struct bpf_map_ops task_storage_map_ops = {
 	.map_update_elem = bpf_pid_task_storage_update_elem,
 	.map_delete_elem = bpf_pid_task_storage_delete_elem,
 	.map_check_btf = bpf_local_storage_map_check_btf,
+	.map_mem_usage = bpf_local_storage_map_mem_usage,
 	.map_btf_id = &bpf_local_storage_map_btf_id[0],
 	.map_owner_storage_ptr = task_storage_ptr,
 };
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index bb378c33f542..7a36353dbc22 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -324,6 +324,7 @@ const struct bpf_map_ops sk_storage_map_ops = {
 	.map_local_storage_charge = bpf_sk_storage_charge,
 	.map_local_storage_uncharge = bpf_sk_storage_uncharge,
 	.map_owner_storage_ptr = bpf_sk_storage_ptr,
+	.map_mem_usage = bpf_local_storage_map_mem_usage,
 };
 
 const struct bpf_func_proto bpf_sk_storage_get_proto = {
-- 
cgit v1.2.3-70-g09d2


From 73d2c61919e9aeffad1b826ec23b1a4a07c1e0dd Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:12 +0000
Subject: bpf, net: sock_map memory usage

sockmap and sockhash don't have something in common in allocation, so let's
introduce different helpers to calculate their memory usage.

The reuslt as follows,

- before
28: sockmap  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B
29: sockhash  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B

- after
28: sockmap  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524608B
29: sockhash  name count_map  flags 0x0  <<<< no updated elements
        key 4B  value 4B  max_entries 65536  memlock 1048896B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-16-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/core/sock_map.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index a68a7290a3b2..9b854e236d23 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -797,6 +797,14 @@ static void sock_map_fini_seq_private(void *priv_data)
 	bpf_map_put_with_uref(info->map);
 }
 
+static u64 sock_map_mem_usage(const struct bpf_map *map)
+{
+	u64 usage = sizeof(struct bpf_stab);
+
+	usage += (u64)map->max_entries * sizeof(struct sock *);
+	return usage;
+}
+
 static const struct bpf_iter_seq_info sock_map_iter_seq_info = {
 	.seq_ops		= &sock_map_seq_ops,
 	.init_seq_private	= sock_map_init_seq_private,
@@ -816,6 +824,7 @@ const struct bpf_map_ops sock_map_ops = {
 	.map_lookup_elem	= sock_map_lookup,
 	.map_release_uref	= sock_map_release_progs,
 	.map_check_btf		= map_check_no_btf,
+	.map_mem_usage		= sock_map_mem_usage,
 	.map_btf_id		= &sock_map_btf_ids[0],
 	.iter_seq_info		= &sock_map_iter_seq_info,
 };
@@ -1397,6 +1406,16 @@ static void sock_hash_fini_seq_private(void *priv_data)
 	bpf_map_put_with_uref(info->map);
 }
 
+static u64 sock_hash_mem_usage(const struct bpf_map *map)
+{
+	struct bpf_shtab *htab = container_of(map, struct bpf_shtab, map);
+	u64 usage = sizeof(*htab);
+
+	usage += htab->buckets_num * sizeof(struct bpf_shtab_bucket);
+	usage += atomic_read(&htab->count) * (u64)htab->elem_size;
+	return usage;
+}
+
 static const struct bpf_iter_seq_info sock_hash_iter_seq_info = {
 	.seq_ops		= &sock_hash_seq_ops,
 	.init_seq_private	= sock_hash_init_seq_private,
@@ -1416,6 +1435,7 @@ const struct bpf_map_ops sock_hash_ops = {
 	.map_lookup_elem_sys_only = sock_hash_lookup_sys,
 	.map_release_uref	= sock_hash_release_progs,
 	.map_check_btf		= map_check_no_btf,
+	.map_mem_usage		= sock_hash_mem_usage,
 	.map_btf_id		= &sock_hash_map_btf_ids[0],
 	.iter_seq_info		= &sock_hash_iter_seq_info,
 };
-- 
cgit v1.2.3-70-g09d2


From b4fd0d672bca001632d7291b5b162b08e065b815 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:13 +0000
Subject: bpf, net: xskmap memory usage

A new helper is introduced to calculate xskmap memory usage.

The xfsmap memory usage can be dynamically changed when we add or remove
a xsk_map_node. Hence we need to track the count of xsk_map_node to get
its memory usage.

The result as follows,
- before
10: xskmap  name count_map  flags 0x0
        key 4B  value 4B  max_entries 65536  memlock 524288B

- after
10: xskmap  name count_map  flags 0x0 <<< no elements case
        key 4B  value 4B  max_entries 65536  memlock 524608B

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-17-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/net/xdp_sock.h |  1 +
 net/xdp/xskmap.c       | 13 +++++++++++++
 2 files changed, 14 insertions(+)

diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index 3057e1a4a11c..e96a1151ec75 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -38,6 +38,7 @@ struct xdp_umem {
 struct xsk_map {
 	struct bpf_map map;
 	spinlock_t lock; /* Synchronize map updates */
+	atomic_t count;
 	struct xdp_sock __rcu *xsk_map[];
 };
 
diff --git a/net/xdp/xskmap.c b/net/xdp/xskmap.c
index 771d0fa90ef5..0c38d7175922 100644
--- a/net/xdp/xskmap.c
+++ b/net/xdp/xskmap.c
@@ -24,6 +24,7 @@ static struct xsk_map_node *xsk_map_node_alloc(struct xsk_map *map,
 		return ERR_PTR(-ENOMEM);
 
 	bpf_map_inc(&map->map);
+	atomic_inc(&map->count);
 
 	node->map = map;
 	node->map_entry = map_entry;
@@ -32,8 +33,11 @@ static struct xsk_map_node *xsk_map_node_alloc(struct xsk_map *map,
 
 static void xsk_map_node_free(struct xsk_map_node *node)
 {
+	struct xsk_map *map = node->map;
+
 	bpf_map_put(&node->map->map);
 	kfree(node);
+	atomic_dec(&map->count);
 }
 
 static void xsk_map_sock_add(struct xdp_sock *xs, struct xsk_map_node *node)
@@ -85,6 +89,14 @@ static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 	return &m->map;
 }
 
+static u64 xsk_map_mem_usage(const struct bpf_map *map)
+{
+	struct xsk_map *m = container_of(map, struct xsk_map, map);
+
+	return struct_size(m, xsk_map, map->max_entries) +
+		   (u64)atomic_read(&m->count) * sizeof(struct xsk_map_node);
+}
+
 static void xsk_map_free(struct bpf_map *map)
 {
 	struct xsk_map *m = container_of(map, struct xsk_map, map);
@@ -267,6 +279,7 @@ const struct bpf_map_ops xsk_map_ops = {
 	.map_update_elem = xsk_map_update_elem,
 	.map_delete_elem = xsk_map_delete_elem,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = xsk_map_mem_usage,
 	.map_btf_id = &xsk_map_btf_ids[0],
 	.map_redirect = xsk_map_redirect,
 };
-- 
cgit v1.2.3-70-g09d2


From 9629363cd05642fe43aded44938adec067ad1da3 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:14 +0000
Subject: bpf: offload map memory usage

A new helper is introduced to calculate offload map memory usage. But
currently the memory dynamically allocated in netdev dev_ops, like
nsim_map_update_elem, is not counted. Let's just put it aside now.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-18-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  | 6 ++++++
 kernel/bpf/offload.c | 6 ++++++
 kernel/bpf/syscall.c | 1 +
 3 files changed, 13 insertions(+)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 9059520bbb5e..6792a7940e1e 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2624,6 +2624,7 @@ static inline bool bpf_map_is_offloaded(struct bpf_map *map)
 
 struct bpf_map *bpf_map_offload_map_alloc(union bpf_attr *attr);
 void bpf_map_offload_map_free(struct bpf_map *map);
+u64 bpf_map_offload_map_mem_usage(const struct bpf_map *map);
 int bpf_prog_test_run_syscall(struct bpf_prog *prog,
 			      const union bpf_attr *kattr,
 			      union bpf_attr __user *uattr);
@@ -2695,6 +2696,11 @@ static inline void bpf_map_offload_map_free(struct bpf_map *map)
 {
 }
 
+static inline u64 bpf_map_offload_map_mem_usage(const struct bpf_map *map)
+{
+	return 0;
+}
+
 static inline int bpf_prog_test_run_syscall(struct bpf_prog *prog,
 					    const union bpf_attr *kattr,
 					    union bpf_attr __user *uattr)
diff --git a/kernel/bpf/offload.c b/kernel/bpf/offload.c
index 0c85e06f7ea7..d9c9f45e3529 100644
--- a/kernel/bpf/offload.c
+++ b/kernel/bpf/offload.c
@@ -563,6 +563,12 @@ void bpf_map_offload_map_free(struct bpf_map *map)
 	bpf_map_area_free(offmap);
 }
 
+u64 bpf_map_offload_map_mem_usage(const struct bpf_map *map)
+{
+	/* The memory dynamically allocated in netdev dev_ops is not counted */
+	return sizeof(struct bpf_offloaded_map);
+}
+
 int bpf_map_offload_lookup_elem(struct bpf_map *map, void *key, void *value)
 {
 	struct bpf_offloaded_map *offmap = map_to_offmap(map);
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 7c96e68859ec..053409d951d2 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -105,6 +105,7 @@ const struct bpf_map_ops bpf_map_offload_ops = {
 	.map_alloc = bpf_map_offload_map_alloc,
 	.map_free = bpf_map_offload_map_free,
 	.map_check_btf = map_check_no_btf,
+	.map_mem_usage = bpf_map_offload_map_mem_usage,
 };
 
 static struct bpf_map *find_and_alloc_map(union bpf_attr *attr)
-- 
cgit v1.2.3-70-g09d2


From 6b4a6ea2c62d34272d64161d43a19c02355576e2 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Sun, 5 Mar 2023 12:46:15 +0000
Subject: bpf: enforce all maps having memory usage callback

We have implemented memory usage callback for all maps, and we enforce
any newly added map having a callback as well. We check this callback at
map creation time. If it doesn't have the callback, we will return
EINVAL.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230305124615.12358-19-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/syscall.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 053409d951d2..f406dfa13792 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -129,6 +129,8 @@ static struct bpf_map *find_and_alloc_map(union bpf_attr *attr)
 	}
 	if (attr->map_ifindex)
 		ops = &bpf_map_offload_ops;
+	if (!ops->map_mem_usage)
+		return ERR_PTR(-EINVAL);
 	map = ops->map_alloc(attr);
 	if (IS_ERR(map))
 		return map;
@@ -775,13 +777,7 @@ static fmode_t map_get_sys_perms(struct bpf_map *map, struct fd f)
 /* Show the memory usage of a bpf map */
 static u64 bpf_map_memory_usage(const struct bpf_map *map)
 {
-	unsigned long size;
-
-	if (map->ops->map_mem_usage)
-		return map->ops->map_mem_usage(map);
-
-	size = round_up(map->key_size + bpf_map_value_size(map), 8);
-	return round_up(map->max_entries * size, PAGE_SIZE);
+	return map->ops->map_mem_usage(map);
 }
 
 static void bpf_map_show_fdinfo(struct seq_file *m, struct file *filp)
-- 
cgit v1.2.3-70-g09d2


From 11a2638d120b9d998916efb6fc55c6422e469ffa Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:19 +0200
Subject: wifi: radiotap: separate vendor TLV into header/content

To be able to use a general function later for any kind of
TLV, separate the vendor TLV header/content in the structs.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.8ac5195bb3e6.I19ad99c1ad3108453aede64bddf6ef1a7c4a0b74@changeid
[separate from the original combined patch]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 16 ++++++++--------
 drivers/net/wireless/mac80211_hwsim.c         | 14 +++++++-------
 include/net/ieee80211_radiotap.h              | 20 ++++++++++++++------
 3 files changed, 29 insertions(+), 21 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 2db4f68becff..71a6555f90d9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -223,18 +223,18 @@ static void iwl_mvm_add_rtap_sniffer_config(struct iwl_mvm *mvm,
 				    vendor_data_len);
 
 	/* Intel OUI */
-	radiotap->oui[0] = 0xf6;
-	radiotap->oui[1] = 0x54;
-	radiotap->oui[2] = 0x25;
+	radiotap->content.oui[0] = 0xf6;
+	radiotap->content.oui[1] = 0x54;
+	radiotap->content.oui[2] = 0x25;
 	/* radiotap sniffer config sub-namespace */
-	radiotap->oui_subtype = 1;
-	radiotap->vendor_type = 0;
+	radiotap->content.oui_subtype = 1;
+	radiotap->content.vendor_type = 0;
 	/* clear reserved field */
-	radiotap->reserved = 0;
+	radiotap->content.reserved = 0;
 	/* fill the data now */
-	memcpy(radiotap->data, &mvm->cur_aid, sizeof(mvm->cur_aid));
+	memcpy(radiotap->content.data, &mvm->cur_aid, sizeof(mvm->cur_aid));
 	/* and clear the padding */
-	memset(radiotap->data + vendor_data_len, 0, padding);
+	memset(radiotap->content.data + vendor_data_len, 0, padding);
 
 	rx_status->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
 }
diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index 152617034d19..f4bdc243ea0d 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -1556,14 +1556,14 @@ static void mac80211_hwsim_add_vendor_rtap(struct sk_buff *skb)
 				sizeof(vendor_data));
 	rtap->type = cpu_to_le16(IEEE80211_RADIOTAP_VENDOR_NAMESPACE);
 
-	rtap->oui[0] = HWSIM_RADIOTAP_OUI[0];
-	rtap->oui[1] = HWSIM_RADIOTAP_OUI[1];
-	rtap->oui[2] = HWSIM_RADIOTAP_OUI[2];
-	rtap->oui_subtype = 127;
+	rtap->content.oui[0] = HWSIM_RADIOTAP_OUI[0];
+	rtap->content.oui[1] = HWSIM_RADIOTAP_OUI[1];
+	rtap->content.oui[2] = HWSIM_RADIOTAP_OUI[2];
+	rtap->content.oui_subtype = 127;
 	/* clear reserved field */
-	rtap->reserved = 0;
-	rtap->vendor_type = 0;
-	memcpy(rtap->data, vendor_data, sizeof(vendor_data));
+	rtap->content.reserved = 0;
+	rtap->content.vendor_type = 0;
+	memcpy(rtap->content.data, vendor_data, sizeof(vendor_data));
 
 	IEEE80211_SKB_RXCB(skb)->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
 #endif
diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h
index 95436686d3fe..f980a72f2ce6 100644
--- a/include/net/ieee80211_radiotap.h
+++ b/include/net/ieee80211_radiotap.h
@@ -370,18 +370,14 @@ struct ieee80211_radiotap_tlv {
 } __packed;
 
 /**
- * struct ieee80211_radiotap_vendor_tlv - vendor radiotap data information
- * @type: should always be set to IEEE80211_RADIOTAP_VENDOR_NAMESPACE
- * @len: length of data
+ * struct ieee80211_radiotap_vendor_content - radiotap vendor data content
  * @oui: radiotap vendor namespace OUI
  * @oui_subtype: radiotap vendor sub namespace
  * @vendor_type: radiotap vendor type
  * @reserved: should always be set to zero (to avoid leaking memory)
  * @data: the actual vendor namespace data
  */
-struct ieee80211_radiotap_vendor_tlv {
-	__le16 type; /* IEEE80211_RADIOTAP_VENDOR_NAMESPACE */
-	__le16 len;
+struct ieee80211_radiotap_vendor_content {
 	u8 oui[3];
 	u8 oui_subtype;
 	__le16 vendor_type;
@@ -389,6 +385,18 @@ struct ieee80211_radiotap_vendor_tlv {
 	u8 data[];
 } __packed;
 
+/**
+ * struct ieee80211_radiotap_vendor_tlv - vendor radiotap data information
+ * @type: should always be set to IEEE80211_RADIOTAP_VENDOR_NAMESPACE
+ * @len: length of data
+ * @content: vendor content see @ieee80211_radiotap_vendor_content
+ */
+struct ieee80211_radiotap_vendor_tlv {
+	__le16 type; /* IEEE80211_RADIOTAP_VENDOR_NAMESPACE */
+	__le16 len;
+	struct ieee80211_radiotap_vendor_content content;
+};
+
 /* ieee80211_radiotap_eht_usig - content of U-SIG tlv (type 33)
  * see www.radiotap.org/fields/U-SIG.html for details
  */
-- 
cgit v1.2.3-70-g09d2


From 056805bcc6bc439de1c9fbb1854042fdf40c8638 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:19 +0200
Subject: wifi: iwlwifi: mvm: add an helper function radiotap TLVs

Add a helper function setting type, length, zeroing out
TLV data and including adding padding if necessary.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.8ac5195bb3e6.I19ad99c1ad3108453aede64bddf6ef1a7c4a0b74@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 43 ++++++++++++++++-----------
 1 file changed, 26 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 71a6555f90d9..206b56ca8db8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -205,36 +205,45 @@ static int iwl_mvm_create_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 	return 0;
 }
 
+/* put a TLV on the skb and return data pointer
+ *
+ * Also pad to 4 the len and zero out all data part
+ */
+static void *
+iwl_mvm_radiotap_put_tlv(struct sk_buff *skb, u16 type, u16 len)
+{
+	struct ieee80211_radiotap_tlv *tlv;
+
+	tlv = skb_put(skb, sizeof(*tlv));
+	tlv->type = cpu_to_le16(type);
+	tlv->len = cpu_to_le16(len);
+	return skb_put_zero(skb, ALIGN(len, 4));
+}
+
 static void iwl_mvm_add_rtap_sniffer_config(struct iwl_mvm *mvm,
 					    struct sk_buff *skb)
 {
 	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
-	struct ieee80211_radiotap_vendor_tlv *radiotap;
+	struct ieee80211_radiotap_vendor_content *radiotap;
 	const u16 vendor_data_len = sizeof(mvm->cur_aid);
-	const u16 padding = ALIGN(vendor_data_len, 4) - vendor_data_len;
 
 	if (!mvm->cur_aid)
 		return;
 
-	radiotap = skb_put(skb, sizeof(*radiotap) + vendor_data_len + padding);
-	radiotap->type = cpu_to_le16(IEEE80211_RADIOTAP_VENDOR_NAMESPACE);
-	radiotap->len = cpu_to_le16(sizeof(*radiotap) -
-				    sizeof(struct ieee80211_radiotap_tlv) +
-				    vendor_data_len);
+	radiotap = iwl_mvm_radiotap_put_tlv(skb,
+					    IEEE80211_RADIOTAP_VENDOR_NAMESPACE,
+					    sizeof(*radiotap) + vendor_data_len);
 
 	/* Intel OUI */
-	radiotap->content.oui[0] = 0xf6;
-	radiotap->content.oui[1] = 0x54;
-	radiotap->content.oui[2] = 0x25;
+	radiotap->oui[0] = 0xf6;
+	radiotap->oui[1] = 0x54;
+	radiotap->oui[2] = 0x25;
 	/* radiotap sniffer config sub-namespace */
-	radiotap->content.oui_subtype = 1;
-	radiotap->content.vendor_type = 0;
-	/* clear reserved field */
-	radiotap->content.reserved = 0;
+	radiotap->oui_subtype = 1;
+	radiotap->vendor_type = 0;
+
 	/* fill the data now */
-	memcpy(radiotap->content.data, &mvm->cur_aid, sizeof(mvm->cur_aid));
-	/* and clear the padding */
-	memset(radiotap->content.data + vendor_data_len, 0, padding);
+	memcpy(radiotap->data, &mvm->cur_aid, sizeof(mvm->cur_aid));
 
 	rx_status->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
 }
-- 
cgit v1.2.3-70-g09d2


From 24f7f6e3ed3b91381307c649a32aff9a69b7cae6 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:20 +0200
Subject: wifi: iwlwifi: mvm: add EHT radiotap info based on rate_n_flags

rate_n_flags is always present in the data so at least give all of
the information we can extract from rate_n_flags

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.b1c7d49ad35e.Ie2412ac6f88700aa3767ff95ffb52a806b13b7ce@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rs.h |  24 +++--
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c  | 135 +++++++++++++++++++++++++
 2 files changed, 152 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
index ddacd5b45aea..1d372a09ebb4 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
@@ -449,11 +449,16 @@ enum {
  *	1			2xLTF+0.8us
  *	2			2xLTF+1.6us
  *	3			4xLTF+3.2us
- * HE TRIG:
+ * HE-EHT TRIG:
  *	0			1xLTF+1.6us
  *	1			2xLTF+1.6us
  *	2			4xLTF+3.2us
  *	3			(does not occur)
+ * EHT MU:
+ *	0			2xLTF+0.8us
+ *	1			2xLTF+1.6us
+ *	2			4xLTF+0.8us
+ *	3			4xLTF+3.2us
  */
 #define RATE_MCS_HE_GI_LTF_POS		20
 #define RATE_MCS_HE_GI_LTF_MSK_V1		(3 << RATE_MCS_HE_GI_LTF_POS)
@@ -546,12 +551,17 @@ enum {
 /*
  * Bits 13-11: (0) 20MHz, (1) 40MHz, (2) 80MHz, (3) 160MHz, (4) 320MHz
  */
-#define RATE_MCS_CHAN_WIDTH_MSK			(0x7 << RATE_MCS_CHAN_WIDTH_POS)
-#define RATE_MCS_CHAN_WIDTH_20			(0 << RATE_MCS_CHAN_WIDTH_POS)
-#define RATE_MCS_CHAN_WIDTH_40			(1 << RATE_MCS_CHAN_WIDTH_POS)
-#define RATE_MCS_CHAN_WIDTH_80			(2 << RATE_MCS_CHAN_WIDTH_POS)
-#define RATE_MCS_CHAN_WIDTH_160			(3 << RATE_MCS_CHAN_WIDTH_POS)
-#define RATE_MCS_CHAN_WIDTH_320			(4 << RATE_MCS_CHAN_WIDTH_POS)
+#define RATE_MCS_CHAN_WIDTH_MSK		(0x7 << RATE_MCS_CHAN_WIDTH_POS)
+#define RATE_MCS_CHAN_WIDTH_20_VAL	0
+#define RATE_MCS_CHAN_WIDTH_20		(RATE_MCS_CHAN_WIDTH_20_VAL << RATE_MCS_CHAN_WIDTH_POS)
+#define RATE_MCS_CHAN_WIDTH_40_VAL	1
+#define RATE_MCS_CHAN_WIDTH_40		(RATE_MCS_CHAN_WIDTH_40_VAL << RATE_MCS_CHAN_WIDTH_POS)
+#define RATE_MCS_CHAN_WIDTH_80_VAL	2
+#define RATE_MCS_CHAN_WIDTH_80		(RATE_MCS_CHAN_WIDTH_80_VAL << RATE_MCS_CHAN_WIDTH_POS)
+#define RATE_MCS_CHAN_WIDTH_160_VAL	3
+#define RATE_MCS_CHAN_WIDTH_160		(RATE_MCS_CHAN_WIDTH_160_VAL << RATE_MCS_CHAN_WIDTH_POS)
+#define RATE_MCS_CHAN_WIDTH_320_VAL	4
+#define RATE_MCS_CHAN_WIDTH_320		(RATE_MCS_CHAN_WIDTH_320_VAL << RATE_MCS_CHAN_WIDTH_POS)
 
 /* Bit 15-14: Antenna selection:
  * Bit 14: Ant A active
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 206b56ca8db8..aa72910f8ff2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1456,6 +1456,137 @@ static void iwl_mvm_decode_he_phy_data(struct iwl_mvm *mvm,
 	}
 }
 
+static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
+			   struct iwl_mvm_rx_phy_data *phy_data,
+			   int queue)
+{
+	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
+
+	struct ieee80211_radiotap_eht *eht;
+	struct ieee80211_radiotap_eht_usig *usig;
+
+	u32 rate_n_flags = phy_data->rate_n_flags;
+	u32 he_type = rate_n_flags & RATE_MCS_HE_TYPE_MSK;
+	/* EHT and HE have the same valus for LTF */
+	u8 ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_UNKNOWN;
+	u16 phy_info = phy_data->phy_info;
+	u32 bw;
+
+	/* u32 for 1 user_info */
+	eht = iwl_mvm_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT,
+				       sizeof(*eht) + sizeof(u32));
+
+	usig = iwl_mvm_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT_USIG,
+					sizeof(*usig));
+	rx_status->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
+	usig->common |=
+		cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_KNOWN);
+
+	/* specific handling for 320MHz */
+	bw = FIELD_GET(RATE_MCS_CHAN_WIDTH_MSK, rate_n_flags);
+	if (bw == RATE_MCS_CHAN_WIDTH_320_VAL)
+		bw += FIELD_GET(IWL_RX_PHY_DATA0_EHT_BW320_SLOT,
+				le32_to_cpu(phy_data->d0));
+
+	usig->common |= cpu_to_le32
+		(FIELD_PREP(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW, bw));
+
+	/* report the AMPDU-EOF bit on single frames */
+	if (!queue && !(phy_info & IWL_RX_MPDU_PHY_AMPDU)) {
+		rx_status->flag |= RX_FLAG_AMPDU_DETAILS;
+		rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN;
+		if (phy_data->d0 & cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF))
+			rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT;
+	}
+
+	/* update aggregation data for monitor sake on default queue */
+	if (!queue && (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) &&
+	    (phy_info & IWL_RX_MPDU_PHY_AMPDU)) {
+		bool toggle_bit = phy_info & IWL_RX_MPDU_PHY_AMPDU_TOGGLE;
+
+		/* toggle is switched whenever new aggregation starts */
+		if (toggle_bit != mvm->ampdu_toggle) {
+			rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN;
+			if (phy_data->d0 & cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF))
+				rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT;
+		}
+	}
+
+	/* TODO: fill usig info (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) */
+
+#define CHECK_TYPE(F)							\
+	BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_DATA1_FORMAT_ ## F !=	\
+		     (RATE_MCS_HE_TYPE_ ## F >> RATE_MCS_HE_TYPE_POS))
+
+	CHECK_TYPE(SU);
+	CHECK_TYPE(EXT_SU);
+	CHECK_TYPE(MU);
+	CHECK_TYPE(TRIG);
+
+	switch (FIELD_GET(RATE_MCS_HE_GI_LTF_MSK, rate_n_flags)) {
+	case 0:
+		if (he_type == RATE_MCS_HE_TYPE_TRIG) {
+			rx_status->eht.gi = NL80211_RATE_INFO_EHT_GI_1_6;
+			ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_1X;
+		} else {
+			rx_status->eht.gi = NL80211_RATE_INFO_EHT_GI_0_8;
+			ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_2X;
+		}
+		break;
+	case 1:
+		rx_status->eht.gi = NL80211_RATE_INFO_EHT_GI_1_6;
+		ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_2X;
+		break;
+	case 2:
+		ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_4X;
+		if (he_type == RATE_MCS_HE_TYPE_TRIG)
+			rx_status->eht.gi = NL80211_RATE_INFO_EHT_GI_3_2;
+		else
+			rx_status->eht.gi = NL80211_RATE_INFO_EHT_GI_0_8;
+		break;
+	case 3:
+		if (he_type != RATE_MCS_HE_TYPE_TRIG) {
+			ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_4X;
+			rx_status->eht.gi = NL80211_RATE_INFO_EHT_GI_3_2;
+		}
+		break;
+	default:
+		/* nothing here */
+		break;
+	}
+
+	if (ltf != IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_UNKNOWN) {
+		eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_GI);
+		eht->data[0] |= cpu_to_le32
+			(FIELD_PREP(IEEE80211_RADIOTAP_EHT_DATA0_LTF,
+				    ltf) |
+			 FIELD_PREP(IEEE80211_RADIOTAP_EHT_DATA0_GI,
+				    rx_status->eht.gi));
+	}
+
+	eht->user_info[0] |= cpu_to_le32
+		(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS_KNOWN |
+		 IEEE80211_RADIOTAP_EHT_USER_INFO_CODING_KNOWN |
+		 IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_KNOWN_O |
+		 IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_KNOWN_O |
+		 IEEE80211_RADIOTAP_EHT_USER_INFO_DATA_FOR_USER);
+
+	if (rate_n_flags & RATE_MCS_BF_MSK)
+		eht->user_info[0] |=
+			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_O);
+
+	if (rate_n_flags & RATE_MCS_LDPC_MSK)
+		eht->user_info[0] |=
+			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_CODING);
+
+	eht->user_info[0] |= cpu_to_le32
+		(FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS,
+			    FIELD_GET(RATE_VHT_MCS_RATE_CODE_MSK,
+				      rate_n_flags)) |
+		 FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O,
+			    FIELD_GET(RATE_MCS_NSS_MSK, rate_n_flags)));
+}
+
 static void iwl_mvm_rx_he(struct iwl_mvm *mvm, struct sk_buff *skb,
 			  struct iwl_mvm_rx_phy_data *phy_data,
 			  int queue)
@@ -1703,6 +1834,10 @@ static void iwl_mvm_rx_fill_status(struct iwl_mvm *mvm,
 	iwl_mvm_get_signal_strength(mvm, rx_status, rate_n_flags,
 				    phy_data->energy_a, phy_data->energy_b);
 
+	/* using TLV format and must be after all fixed len fields */
+	if (format == RATE_MCS_EHT_MSK)
+		iwl_mvm_rx_eht(mvm, skb, phy_data, queue);
+
 	if (unlikely(mvm->monitor_on))
 		iwl_mvm_add_rtap_sniffer_config(mvm, skb);
 
-- 
cgit v1.2.3-70-g09d2


From 4ec825854c3c1e7a2cb3fedd6d1acbba0a3e2d74 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:21 +0200
Subject: wifi: iwlwifi: mvm: add all EHT based on data0 info from HW

Update all radiotap EHT TLVs that we can extract from data0 in HW.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.730f219e02ee.Ife3dd85c65758694d7602e8bc8660887d77faacf@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rx.h |  3 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c  | 73 +++++++++++++++++++++++++-
 2 files changed, 73 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
index 1c4e84932058..544d22472a6f 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
@@ -367,7 +367,8 @@ enum iwl_rx_phy_eht_data1 {
 	/* number of EHT-LTF symbols 0 - 1 EHT-LTF, 1 - 2 EHT-LTFs, 2 - 4 EHT-LTFs,
 	 * 3 - 6 EHT-LTFs, 4 - 8 EHT-LTFs */
 	IWL_RX_PHY_DATA1_EHT_SIG_LTF_NUM		= 0x000000e0,
-	IWL_RX_PHY_DATA1_EHT_RU_ALLOC			= 0x0000ff00,
+	IWL_RX_PHY_DATA1_EHT_B0				= 0x00000100,
+	IWL_RX_PHY_DATA1_EHT_RU_B1_B7_ALLOC		= 0x0000fe00,
 };
 
 /* goes into Metadata DW 7 */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index aa72910f8ff2..174bed9b5074 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2012-2014, 2018-2022 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2023 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2015-2017 Intel Deutschland GmbH
  */
@@ -1456,6 +1456,74 @@ static void iwl_mvm_decode_he_phy_data(struct iwl_mvm *mvm,
 	}
 }
 
+#define LE32_DEC_ENC(value, dec_bits, enc_bits) \
+	le32_encode_bits(le32_get_bits(value, dec_bits), enc_bits)
+
+static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
+					struct iwl_mvm_rx_phy_data *phy_data,
+					struct ieee80211_rx_status *rx_status,
+					struct ieee80211_radiotap_eht *eht,
+					struct ieee80211_radiotap_eht_usig *usig)
+
+{
+	__le32 data0 = phy_data->d0;
+	__le32 data1 = phy_data->d1;
+	u8 info_type = phy_data->info_type;
+
+	/* Not in EHT range */
+	if (info_type < IWL_RX_PHY_INFO_TYPE_EHT_MU ||
+	    info_type > IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT)
+		return;
+
+	usig->common |= cpu_to_le32
+		(IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL_KNOWN |
+		 IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN);
+	usig->common |= LE32_DEC_ENC(data0,
+				     IWL_RX_PHY_DATA0_EHT_UPLINK,
+				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL);
+	usig->common |= LE32_DEC_ENC(data0,
+				     IWL_RX_PHY_DATA0_EHT_BSS_COLOR_MASK,
+				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN);
+
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_SPATIAL_REUSE);
+	eht->data[0] |= LE32_DEC_ENC(data0,
+				     IWL_RX_PHY_DATA0_ETH_SPATIAL_REUSE_MASK,
+				     IEEE80211_RADIOTAP_EHT_DATA0_SPATIAL_REUSE);
+
+	/* All RU allocating size/index is in TB format */
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_RU_ALLOC_TB_FMT);
+	eht->data[8] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PS160,
+				     IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_PS_160);
+	eht->data[8] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_B0,
+				     IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B0);
+	eht->data[8] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_RU_B1_B7_ALLOC,
+				     IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1);
+
+	usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN);
+	usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_TXOP_DUR_MASK,
+				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP);
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_LDPC_EXTRA_SYM_OM);
+	eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_LDPC_EXT_SYM,
+				     IEEE80211_RADIOTAP_EHT_DATA0_LDPC_EXTRA_SYM_OM);
+
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_PRE_PADD_FACOR_OM);
+	eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PRE_FEC_PAD_MASK,
+				    IEEE80211_RADIOTAP_EHT_DATA0_PRE_PADD_FACOR_OM);
+
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_PE_DISAMBIGUITY_OM);
+	eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PE_DISAMBIG,
+				     IEEE80211_RADIOTAP_EHT_DATA0_PE_DISAMBIGUITY_OM);
+
+	/* TODO: what about IWL_RX_PHY_DATA0_EHT_BW320_SLOT */
+
+	if (!le32_get_bits(data0, IWL_RX_PHY_DATA0_EHT_SIGA_CRC_OK))
+		usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BAD_USIG_CRC);
+
+	usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER_KNOWN);
+	usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PHY_VER,
+				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER);
+}
+
 static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
 			   struct iwl_mvm_rx_phy_data *phy_data,
 			   int queue)
@@ -1512,7 +1580,8 @@ static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
 		}
 	}
 
-	/* TODO: fill usig info (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) */
+	if (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD)
+		iwl_mvm_decode_eht_phy_data(mvm, phy_data, rx_status, eht, usig);
 
 #define CHECK_TYPE(F)							\
 	BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_DATA1_FORMAT_ ## F !=	\
-- 
cgit v1.2.3-70-g09d2


From 3ecf34118dc80a0508dc82e284617513cbb270e8 Mon Sep 17 00:00:00 2001
From: Alon Giladi <alon.giladi@intel.com>
Date: Sun, 5 Mar 2023 14:16:22 +0200
Subject: wifi: iwlwifi: mvm: allow Microsoft to use TAS

Add Microsoft to the list of OEMs which allowed to use TAS.

Signed-off-by: Alon Giladi <alon.giladi@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.662967fec1cc.Icb30cddc049cb5402fd5ab2ce7f95033e478b1b9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 0c6b49fcb00d..45981e22b2db 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1084,6 +1084,11 @@ static const struct dmi_system_id dmi_tas_approved_list[] = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
 		},
 	},
+	{ .ident = "MSFT",
+	  .matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Microsoft Corporation"),
+		},
+	},
 
 	/* keep last */
 	{}
-- 
cgit v1.2.3-70-g09d2


From c0da321b601a92453f928c819dbd65c5712480d3 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:23 +0200
Subject: wifi: iwlwifi: mvm: rename define to generic name

The type RX_NO_DATA_INFO_TYPE_HE_TB_UNMATCHED is applied to all TB
frames including EHT mode, so rename accordingly.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.e4f51f347e48.I2d6ecb6eadc95666d2ef9794662ee779488ceac1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rx.h | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
index 544d22472a6f..0f190266fffd 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
@@ -726,7 +726,7 @@ struct iwl_rx_mpdu_desc {
 #define RX_NO_DATA_INFO_TYPE_RX_ERR	1
 #define RX_NO_DATA_INFO_TYPE_NDP	2
 #define RX_NO_DATA_INFO_TYPE_MU_UNMATCHED	3
-#define RX_NO_DATA_INFO_TYPE_HE_TB_UNMATCHED	4
+#define RX_NO_DATA_INFO_TYPE_TB_UNMATCHED	4
 
 #define RX_NO_DATA_INFO_ERR_POS		8
 #define RX_NO_DATA_INFO_ERR_MSK		(0xff << RX_NO_DATA_INFO_ERR_POS)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 174bed9b5074..1fb77cda568a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -2341,7 +2341,7 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 			IEEE80211_RADIOTAP_ZERO_LEN_PSDU_SOUNDING;
 		break;
 	case RX_NO_DATA_INFO_TYPE_MU_UNMATCHED:
-	case RX_NO_DATA_INFO_TYPE_HE_TB_UNMATCHED:
+	case RX_NO_DATA_INFO_TYPE_TB_UNMATCHED:
 		rx_status->zero_length_psdu_type =
 			IEEE80211_RADIOTAP_ZERO_LEN_PSDU_NOT_CAPTURED;
 		break;
-- 
cgit v1.2.3-70-g09d2


From e8c0a6fd08d7dc6c178e77f154794cdd55575780 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:24 +0200
Subject: wifi: iwlwifi: mvm: decode USIG_B1_B7 RU to nl80211 RU width

This is based on 802.11be D1.5 table 9-53a

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.0b720d6d6a48.I0034dd108696223494799d3ffe4f09685800b831@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 71 +++++++++++++++++++++++++++
 1 file changed, 71 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 1fb77cda568a..c069e81bf3f1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1459,6 +1459,75 @@ static void iwl_mvm_decode_he_phy_data(struct iwl_mvm *mvm,
 #define LE32_DEC_ENC(value, dec_bits, enc_bits) \
 	le32_encode_bits(le32_get_bits(value, dec_bits), enc_bits)
 
+static void iwl_mvm_decode_eht_ru(struct iwl_mvm *mvm,
+				  struct ieee80211_rx_status *rx_status,
+				  struct ieee80211_radiotap_eht *eht)
+{
+	u32 ru = le32_get_bits(eht->data[8],
+			       IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1);
+	enum nl80211_eht_ru_alloc nl_ru;
+
+	/* Using D1.5 Table 9-53a - Encoding of PS160 and RU Allocation subfields
+	 * in an EHT variant User Info field
+	 */
+
+	switch (ru) {
+	case 0 ... 36:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_26;
+		break;
+	case 37 ... 52:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_52;
+		break;
+	case 53 ... 60:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_106;
+		break;
+	case 61 ... 64:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_242;
+		break;
+	case 65 ... 66:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_484;
+		break;
+	case 67:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_996;
+		break;
+	case 68:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_2x996;
+		break;
+	case 69:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_4x996;
+		break;
+	case 70 ... 81:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_52P26;
+		break;
+	case 82 ... 89:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_106P26;
+		break;
+	case 90 ... 93:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_484P242;
+		break;
+	case 94 ... 95:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_996P484;
+		break;
+	case 96 ... 99:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_996P484P242;
+		break;
+	case 100 ... 103:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_2x996P484;
+		break;
+	case 104:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_3x996;
+		break;
+	case 105 ... 106:
+		nl_ru = NL80211_RATE_INFO_EHT_RU_ALLOC_3x996P484;
+		break;
+	default:
+		return;
+	}
+
+	rx_status->bw = RATE_INFO_BW_EHT_RU;
+	rx_status->eht.ru = nl_ru;
+}
+
 static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 					struct iwl_mvm_rx_phy_data *phy_data,
 					struct ieee80211_rx_status *rx_status,
@@ -1499,6 +1568,8 @@ static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 	eht->data[8] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_RU_B1_B7_ALLOC,
 				     IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1);
 
+	iwl_mvm_decode_eht_ru(mvm, rx_status, eht);
+
 	usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN);
 	usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_TXOP_DUR_MASK,
 				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP);
-- 
cgit v1.2.3-70-g09d2


From feb4a0e215fa98f72c37b51344d3c013bb43e013 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:25 +0200
Subject: wifi: iwlwifi: mvm: parse FW frame metadata for EHT sniffer mode

In EHT sniffer mode DW4 is all used for sniffer data (unlike we have in
HE mode), so move the full DW4 into a union, and we extract the new data5
used for parsing USIG info and set all to radiotap TLVs with the
extracted data.

Also parse OFDM_RX_VECTOR_USIG_A1_OUT and OFDM_RX_VECTOR_USIG_A2_OUT for
rx_no_data notification.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.557d3870753b.I4e9fa4d21900a187753529d46956ba2a7ee75fda@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rx.h |  66 ++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c  | 170 +++++++++++++++++++++++--
 2 files changed, 213 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
index 0f190266fffd..97e946e70279 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
@@ -674,22 +674,31 @@ struct iwl_rx_mpdu_desc {
 	 * @mac_phy_idx: MAC/PHY index
 	 */
 	u8 mac_phy_idx;
-	/* DW4 - carries csum data only when rpa_en == 1 */
-	/**
-	 * @raw_csum: raw checksum (alledgedly unreliable)
-	 */
-	__le16 raw_csum;
-
+	/* DW4 */
 	union {
+		struct {
+			/* carries csum data only when rpa_en == 1 */
+			/**
+			 * @raw_csum: raw checksum (alledgedly unreliable)
+			 */
+			__le16 raw_csum;
+
+			union {
+				/**
+				 * @l3l4_flags: &enum iwl_rx_l3l4_flags
+				 */
+				__le16 l3l4_flags;
+
+				/**
+				 * @phy_data4: depends on info type, see phy_data1
+				 */
+				__le16 phy_data4;
+			};
+		};
 		/**
-		 * @l3l4_flags: &enum iwl_rx_l3l4_flags
-		 */
-		__le16 l3l4_flags;
-
-		/**
-		 * @phy_data4: depends on info type, see phy_data1
+		 * @phy_eht_data4: depends on info type, see phy_data1
 		 */
-		__le16 phy_data4;
+		__le32 phy_eht_data4;
 	};
 	/* DW5 */
 	/**
@@ -744,6 +753,35 @@ struct iwl_rx_mpdu_desc {
 #define RX_NO_DATA_RX_VEC0_VHT_NSTS_MSK	0x38000000
 #define RX_NO_DATA_RX_VEC2_EHT_NSTS_MSK	0x00f00000
 
+/* content of OFDM_RX_VECTOR_USIG_A1_OUT */
+enum iwl_rx_usig_a1 {
+	IWL_RX_USIG_A1_ENHANCED_WIFI_VER_ID	= 0x00000007,
+	IWL_RX_USIG_A1_BANDWIDTH		= 0x00000038,
+	IWL_RX_USIG_A1_UL_FLAG			= 0x00000040,
+	IWL_RX_USIG_A1_BSS_COLOR		= 0x00001f80,
+	IWL_RX_USIG_A1_TXOP_DURATION		= 0x000fe000,
+	IWL_RX_USIG_A1_DISREGARD		= 0x01f00000,
+	IWL_RX_USIG_A1_VALIDATE			= 0x02000000,
+	IWL_RX_USIG_A1_EHT_BW320_SLOT		= 0x04000000,
+	IWL_RX_USIG_A1_EHT_TYPE			= 0x18000000,
+	IWL_RX_USIG_A1_RDY			= 0x80000000,
+};
+
+/* content of OFDM_RX_VECTOR_USIG_A2_EHT_OUT */
+enum iwl_rx_usig_a2_eht {
+	IWL_RX_USIG_A2_EHT_PPDU_TYPE		= 0x00000003,
+	IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B2	= 0x00000004,
+	IWL_RX_USIG_A2_EHT_PUNC_CHANNEL		= 0x000000f8,
+	IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B8	= 0x00000100,
+	IWL_RX_USIG_A2_EHT_SIG_MCS		= 0x00000600,
+	IWL_RX_USIG_A2_EHT_SIG_SYM_NUM		= 0x0000f800,
+	IWL_RX_USIG_A2_EHT_TRIG_SPATIAL_REUSE_1 = 0x000f0000,
+	IWL_RX_USIG_A2_EHT_TRIG_SPATIAL_REUSE_2 = 0x00f00000,
+	IWL_RX_USIG_A2_EHT_TRIG_USIG2_DISREGARD	= 0x1f000000,
+	IWL_RX_USIG_A2_EHT_CRC_OK		= 0x40000000,
+	IWL_RX_USIG_A2_EHT_RDY			= 0x80000000,
+};
+
 /**
  * struct iwl_rx_no_data - RX no data descriptor
  * @info: 7:0 frame type, 15:8 RX error type
@@ -781,7 +819,7 @@ struct iwl_rx_no_data {
  * @rx_vec: DW-12:9 raw RX vectors from DSP according to modulation type.
  *	for VHT: OFDM_RX_VECTOR_SIGA1_OUT, OFDM_RX_VECTOR_SIGA2_OUT
  *	for HE: OFDM_RX_VECTOR_HE_SIGA1_OUT, OFDM_RX_VECTOR_HE_SIGA2_OUT
- *	for EHT: OFDM_RX_VECTOR_USIG_A1_OUT, OFDM_RX_VECTOR_USIG_A2_OUT,
+ *	for EHT: OFDM_RX_VECTOR_USIG_A1_OUT, OFDM_RX_VECTOR_USIG_A2_EHT_OUT,
  *	OFDM_RX_VECTOR_EHT_OUT, OFDM_RX_VECTOR_EHT_USER_FIELD_OUT
  */
 struct iwl_rx_no_data_ver_3 {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index c069e81bf3f1..1334c96098a5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1177,8 +1177,10 @@ static void iwl_mvm_flip_address(u8 *addr)
 
 struct iwl_mvm_rx_phy_data {
 	enum iwl_rx_phy_info_type info_type;
-	__le32 d0, d1, d2, d3;
+	__le32 d0, d1, d2, d3, eht_d4, d5;
 	__le16 d4;
+	bool with_data;
+	__le32 rx_vec[4];
 
 	u32 rate_n_flags;
 	u32 gp2_on_air_rise;
@@ -1459,6 +1461,119 @@ static void iwl_mvm_decode_he_phy_data(struct iwl_mvm *mvm,
 #define LE32_DEC_ENC(value, dec_bits, enc_bits) \
 	le32_encode_bits(le32_get_bits(value, dec_bits), enc_bits)
 
+#define IWL_MVM_ENC_USIG_VALUE_MASK(usig, in_value, dec_bits, enc_bits) do { \
+	typeof(enc_bits) _enc_bits = enc_bits; \
+	typeof(usig) _usig = usig; \
+	(_usig)->mask |= cpu_to_le32(_enc_bits); \
+	(_usig)->value |= LE32_DEC_ENC(in_value, dec_bits, _enc_bits); \
+} while (0)
+
+static void iwl_mvm_decode_eht_ext_mu(struct iwl_mvm *mvm,
+				      struct iwl_mvm_rx_phy_data *phy_data,
+				      struct ieee80211_rx_status *rx_status,
+				      struct ieee80211_radiotap_eht *eht,
+				      struct ieee80211_radiotap_eht_usig *usig)
+{
+	__le32 data1 = phy_data->d1;
+
+	if (phy_data->with_data) {
+		__le32 data4 = phy_data->eht_d4;
+		__le32 data5 = phy_data->d5;
+
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data5,
+					    IWL_RX_PHY_DATA5_EHT_TYPE_AND_COMP,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B0_B1_PPDU_TYPE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data5,
+					    IWL_RX_PHY_DATA5_EHT_MU_PUNC_CH_CODE,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B3_B7_PUNCTURED_INFO);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data4,
+					    IWL_RX_PHY_DATA4_EHT_MU_EXT_SIGB_MCS,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B9_B10_SIG_MCS);
+		IWL_MVM_ENC_USIG_VALUE_MASK
+			(usig, data1, IWL_RX_PHY_DATA1_EHT_MU_NUM_SIG_SYM_USIGA2,
+			 IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS);
+
+	} else {
+		__le32 usig_a1 = phy_data->rx_vec[0];
+		__le32 usig_a2 = phy_data->rx_vec[1];
+
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a1,
+					    IWL_RX_USIG_A1_DISREGARD,
+					    IEEE80211_RADIOTAP_EHT_USIG1_MU_B20_B24_DISREGARD);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a1,
+					    IWL_RX_USIG_A1_VALIDATE,
+					    IEEE80211_RADIOTAP_EHT_USIG1_MU_B25_VALIDATE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_PPDU_TYPE,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B0_B1_PPDU_TYPE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B2,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B2_VALIDATE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_PUNC_CHANNEL,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B3_B7_PUNCTURED_INFO);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B8,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B8_VALIDATE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_SIG_MCS,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B9_B10_SIG_MCS);
+		IWL_MVM_ENC_USIG_VALUE_MASK
+			(usig, usig_a2, IWL_RX_USIG_A2_EHT_SIG_SYM_NUM,
+			 IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_CRC_OK,
+					    IEEE80211_RADIOTAP_EHT_USIG2_MU_B16_B19_CRC);
+	}
+}
+
+static void iwl_mvm_decode_eht_ext_tb(struct iwl_mvm *mvm,
+				      struct iwl_mvm_rx_phy_data *phy_data,
+				      struct ieee80211_rx_status *rx_status,
+				      struct ieee80211_radiotap_eht *eht,
+				      struct ieee80211_radiotap_eht_usig *usig)
+{
+	if (phy_data->with_data) {
+		__le32 data5 = phy_data->d5;
+
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data5,
+					    IWL_RX_PHY_DATA5_EHT_TYPE_AND_COMP,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B0_B1_PPDU_TYPE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data5,
+					    IWL_RX_PHY_DATA5_EHT_TB_SPATIAL_REUSE1,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B3_B6_SPATIAL_REUSE_1);
+
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data5,
+					    IWL_RX_PHY_DATA5_EHT_TB_SPATIAL_REUSE2,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B7_B10_SPATIAL_REUSE_2);
+	} else {
+		__le32 usig_a1 = phy_data->rx_vec[0];
+		__le32 usig_a2 = phy_data->rx_vec[1];
+
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a1,
+					    IWL_RX_USIG_A1_DISREGARD,
+					    IEEE80211_RADIOTAP_EHT_USIG1_TB_B20_B25_DISREGARD);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_PPDU_TYPE,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B0_B1_PPDU_TYPE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B2,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B2_VALIDATE);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_TRIG_SPATIAL_REUSE_1,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B3_B6_SPATIAL_REUSE_1);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_TRIG_SPATIAL_REUSE_2,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B7_B10_SPATIAL_REUSE_2);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_TRIG_USIG2_DISREGARD,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B11_B15_DISREGARD);
+		IWL_MVM_ENC_USIG_VALUE_MASK(usig, usig_a2,
+					    IWL_RX_USIG_A2_EHT_CRC_OK,
+					    IEEE80211_RADIOTAP_EHT_USIG2_TB_B16_B19_CRC);
+	}
+}
+
 static void iwl_mvm_decode_eht_ru(struct iwl_mvm *mvm,
 				  struct ieee80211_rx_status *rx_status,
 				  struct ieee80211_radiotap_eht *eht)
@@ -1537,6 +1652,7 @@ static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 {
 	__le32 data0 = phy_data->d0;
 	__le32 data1 = phy_data->d1;
+	__le32 usig_a1 = phy_data->rx_vec[0];
 	u8 info_type = phy_data->info_type;
 
 	/* Not in EHT range */
@@ -1547,12 +1663,21 @@ static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 	usig->common |= cpu_to_le32
 		(IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL_KNOWN |
 		 IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN);
-	usig->common |= LE32_DEC_ENC(data0,
-				     IWL_RX_PHY_DATA0_EHT_UPLINK,
-				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL);
-	usig->common |= LE32_DEC_ENC(data0,
-				     IWL_RX_PHY_DATA0_EHT_BSS_COLOR_MASK,
-				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN);
+	if (phy_data->with_data) {
+		usig->common |= LE32_DEC_ENC(data0,
+					     IWL_RX_PHY_DATA0_EHT_UPLINK,
+					     IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL);
+		usig->common |= LE32_DEC_ENC(data0,
+					     IWL_RX_PHY_DATA0_EHT_BSS_COLOR_MASK,
+					     IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR);
+	} else {
+		usig->common |= LE32_DEC_ENC(usig_a1,
+					     IWL_RX_USIG_A1_UL_FLAG,
+					     IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL);
+		usig->common |= LE32_DEC_ENC(usig_a1,
+					     IWL_RX_USIG_A1_BSS_COLOR,
+					     IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR);
+	}
 
 	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_SPATIAL_REUSE);
 	eht->data[0] |= LE32_DEC_ENC(data0,
@@ -1571,8 +1696,13 @@ static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 	iwl_mvm_decode_eht_ru(mvm, rx_status, eht);
 
 	usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN);
-	usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_TXOP_DUR_MASK,
-				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP);
+	if (phy_data->with_data)
+		usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_TXOP_DUR_MASK,
+					     IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP);
+	else
+		usig->common |= LE32_DEC_ENC(usig_a1, IWL_RX_USIG_A1_TXOP_DURATION,
+					     IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP);
+
 	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_LDPC_EXTRA_SYM_OM);
 	eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_LDPC_EXT_SYM,
 				     IEEE80211_RADIOTAP_EHT_DATA0_LDPC_EXTRA_SYM_OM);
@@ -1593,6 +1723,23 @@ static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 	usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER_KNOWN);
 	usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PHY_VER,
 				     IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER);
+
+	/*
+	 * TODO: what about TB - IWL_RX_PHY_DATA1_EHT_TB_PILOT_TYPE,
+	 *			 IWL_RX_PHY_DATA1_EHT_TB_LOW_SS
+	 */
+
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_EHT_LTF);
+	eht->data[0] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_SIG_LTF_NUM,
+				     IEEE80211_RADIOTAP_EHT_DATA0_EHT_LTF);
+
+	if (info_type == IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT ||
+	    info_type == IWL_RX_PHY_INFO_TYPE_EHT_TB)
+		iwl_mvm_decode_eht_ext_tb(mvm, phy_data, rx_status, eht, usig);
+
+	if (info_type == IWL_RX_PHY_INFO_TYPE_EHT_MU_EXT ||
+	    info_type == IWL_RX_PHY_INFO_TYPE_EHT_MU)
+		iwl_mvm_decode_eht_ext_mu(mvm, phy_data, rx_status, eht, usig);
 }
 
 static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
@@ -2077,6 +2224,8 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 		phy_data.d1 = desc->v3.phy_data1;
 		phy_data.d2 = desc->v3.phy_data2;
 		phy_data.d3 = desc->v3.phy_data3;
+		phy_data.eht_d4 = desc->phy_eht_data4;
+		phy_data.d5 = desc->v3.phy_data5;
 	} else {
 		phy_data.rate_n_flags = le32_to_cpu(desc->v1.rate_n_flags);
 		phy_data.channel = desc->v1.channel;
@@ -2106,6 +2255,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 		return;
 	}
 
+	phy_data.with_data = true;
 	phy_data.phy_info = le16_to_cpu(desc->phy_info);
 	phy_data.d4 = desc->phy_data4;
 
@@ -2368,6 +2518,7 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 	phy_data.energy_a = u32_get_bits(rssi, RX_NO_DATA_CHAIN_A_MSK);
 	phy_data.energy_b = u32_get_bits(rssi, RX_NO_DATA_CHAIN_B_MSK);
 	phy_data.channel = u32_get_bits(rssi, RX_NO_DATA_CHANNEL_MSK);
+	phy_data.with_data = false;
 
 	if (iwl_fw_lookup_notif_ver(mvm->fw, DATA_PATH_GROUP,
 				    RX_NO_DATA_NOTIF, 0) < 2) {
@@ -2386,6 +2537,7 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 		    sizeof(struct iwl_rx_no_data_ver_3)))
 		/* invalid len for ver 3 */
 			return;
+		memcpy(phy_data.rx_vec, desc->rx_vec, sizeof(phy_data.rx_vec));
 	} else {
 		if (format == RATE_MCS_EHT_MSK)
 			/* no support for EHT before version 3 API */
-- 
cgit v1.2.3-70-g09d2


From 5abf31544a4d2bcb3a0dd2fd080f9448d4026975 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:26 +0200
Subject: wifi: iwlwifi: mvm: add primary 80 known for EHT radiotap

Calculate the position of the control channel in the wide channel
based on the chandef, this is used to obtain the value of N in
802.11be D1.5 Table 9-53a in the column PHY MU/MRU index.

To avoid the need to calculate every frame the value, do it once
monitor vif is added.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.fe9a5b58e241.I291ee480252d098f62d9ec39040284d3e521d88e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 27 ++++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  5 +++++
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c     |  9 ++++++++
 3 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 565522466eba..51c7652df0ef 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1362,6 +1362,28 @@ static void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk)
 	ieee80211_chswitch_done(vif, false);
 }
 
+static u8
+iwl_mvm_chandef_get_primary_80(struct cfg80211_chan_def *chandef)
+{
+	int data_start;
+	int control_start;
+	int bw;
+
+	if (chandef->width == NL80211_CHAN_WIDTH_320)
+		bw = 320;
+	else if (chandef->width == NL80211_CHAN_WIDTH_160)
+		bw = 160;
+	else
+		return 0;
+
+	/* data is bw wide so the start is half the width */
+	data_start = chandef->center_freq1 - bw / 2;
+	/* control is 20Mhz width */
+	control_start = chandef->chan->center_freq - 10;
+
+	return (control_start - data_start) / 80;
+}
+
 static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 				     struct ieee80211_vif *vif)
 {
@@ -1478,8 +1500,11 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 	INIT_DELAYED_WORK(&mvmvif->csa_work,
 			  iwl_mvm_channel_switch_disconnect_wk);
 
-	if (vif->type == NL80211_IFTYPE_MONITOR)
+	if (vif->type == NL80211_IFTYPE_MONITOR) {
 		mvm->monitor_on = true;
+		mvm->monitor_p80 =
+			iwl_mvm_chandef_get_primary_80(&vif->bss_conf.chandef);
+	}
 
 	iwl_mvm_vif_dbgfs_register(mvm, vif);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ce6b701f3f4c..301f39d9efe7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1096,6 +1096,11 @@ struct iwl_mvm {
 
 	/* does a monitor vif exist (only one can exist hence bool) */
 	bool monitor_on;
+	/*
+	 * primary channel position relative to he whole bandwidth,
+	 * in steps of 80 MHz
+	 */
+	u8 monitor_p80;
 
 	/* sniffer data to include in radiotap */
 	__le16 cur_aid;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 1334c96098a5..b545673fb150 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1695,6 +1695,15 @@ static void iwl_mvm_decode_eht_phy_data(struct iwl_mvm *mvm,
 
 	iwl_mvm_decode_eht_ru(mvm, rx_status, eht);
 
+	/* We only get here in case of IWL_RX_MPDU_PHY_TSF_OVERLOAD is set
+	 * which is on only in case of monitor mode so no need to check monitor
+	 * mode
+	 */
+	eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_PRIMARY_80);
+	eht->data[1] |=
+		le32_encode_bits(mvm->monitor_p80,
+				 IEEE80211_RADIOTAP_EHT_DATA1_PRIMARY_80);
+
 	usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN);
 	if (phy_data->with_data)
 		usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_TXOP_DUR_MASK,
-- 
cgit v1.2.3-70-g09d2


From b85f7ebb24974e016520b5ea39394ab4cb08af0c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 5 Mar 2023 14:16:27 +0200
Subject: wifi: iwlwifi: mvm: avoid UB shift of snif_queue

For the old TX API we need the tfd_queue_msk, but for the
new TX API we don't need it here because we add it to the
station later. However, for the new API mvm->snif_queue is
set to IWL_MVM_INVALID_QUEUE == 0xffff, so the BIT() here
is undefined behaviour.

Since we don't need the tfd_queue_msk value for the new TX
API at all, simply fill it in only for the old API.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.b8da0b7eb194.I53744fd7cfb6e146a9393272a2a61852841238d9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index aa791dbc3066..114c96ba39ee 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -654,7 +654,7 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct iwl_mvm *mvm,
 					 u32 action)
 {
 	struct iwl_mac_ctx_cmd cmd = {};
-	u32 tfd_queue_msk = BIT(mvm->snif_queue);
+	u32 tfd_queue_msk = 0;
 	int ret;
 
 	WARN_ON(vif->type != NL80211_IFTYPE_MONITOR);
@@ -669,6 +669,14 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct iwl_mvm *mvm,
 				       MAC_FILTER_ACCEPT_GRP);
 	ieee80211_hw_set(mvm->hw, RX_INCLUDES_FCS);
 
+	/*
+	 * the queue mask is only relevant for old TX API, and
+	 * mvm->snif_queue isn't set here (it's still set to
+	 * IWL_MVM_INVALID_QUEUE so the BIT() of it is UB)
+	 */
+	if (!iwl_mvm_has_new_tx_api(mvm))
+		tfd_queue_msk = BIT(mvm->snif_queue);
+
 	/* Allocate sniffer station */
 	ret = iwl_mvm_allocate_int_sta(mvm, &mvm->snif_sta, tfd_queue_msk,
 				       vif->type, IWL_STA_GENERAL_PURPOSE);
-- 
cgit v1.2.3-70-g09d2


From f7bd883b3fca50f8429a67b4bfdf82a22e8dac53 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 5 Mar 2023 14:16:28 +0200
Subject: wifi: iwlwifi: mvm: make flush code a bit clearer

The mask building here is only relevant for the old TX API,
so move it into the else branch.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.c0795543f254.I302124a8584dd049577b0c2c74ecd7c48ddf4f3e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 51c7652df0ef..ab02c6076276 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -5058,9 +5058,10 @@ static void iwl_mvm_mac_flush(struct ieee80211_hw *hw,
 			if (iwl_mvm_flush_sta(mvm, mvmsta, false))
 				IWL_ERR(mvm, "flush request fail\n");
 		} else {
-			msk |= mvmsta->tfd_queue_msk;
 			if (iwl_mvm_has_new_tx_api(mvm))
 				iwl_mvm_wait_sta_queues_empty(mvm, mvmsta);
+			else /* only used for !iwl_mvm_has_new_tx_api() below */
+				msk |= mvmsta->tfd_queue_msk;
 		}
 	}
 
-- 
cgit v1.2.3-70-g09d2


From b96e516ccf9f5a227756989d4e47297f65e12e60 Mon Sep 17 00:00:00 2001
From: Golan Ben Ami <golan.ben.ami@intel.com>
Date: Sun, 5 Mar 2023 14:16:29 +0200
Subject: wifi: iwlwifi: Add support for B step of BnJ-Fm4

Support new HW step of BnJ-Fm4 device

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.bb0591c59898.If04d7a45707ba008981f8c8ea7f7f107880f146c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  | 12 ++++++++++++
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  1 +
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   |  5 +++++
 3 files changed, 18 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 3bdd6774716d..05720352e49f 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -62,6 +62,7 @@
 #define IWL_BZ_Z_GF_A_FW_PRE		"iwlwifi-bz-z0-gf-a0-"
 #define IWL_BNJ_A_FM_A_FW_PRE		"iwlwifi-BzBnj-a0-fm-a0-"
 #define IWL_BNJ_A_FM4_A_FW_PRE		"iwlwifi-BzBnj-a0-fm4-a0-"
+#define IWL_BNJ_B_FM4_B_FW_PRE		"iwlwifi-BzBnj-b0-fm4-b0-"
 #define IWL_BNJ_A_GF_A_FW_PRE		"iwlwifi-BzBnj-a0-gf-a0-"
 #define IWL_BNJ_A_GF4_A_FW_PRE		"iwlwifi-BzBnj-a0-gf4-a0-"
 #define IWL_BNJ_A_HR_B_FW_PRE		"iwlwifi-BzBnj-a0-hr-b0-"
@@ -132,6 +133,8 @@
 	IWL_BNJ_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_FM4_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_FM4_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_BNJ_B_FM4_B_MODULE_FIRMWARE(api) \
+	IWL_BNJ_B_FM4_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_GF_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_GF_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(api) \
@@ -998,6 +1001,14 @@ const struct iwl_cfg iwl_cfg_bnj_a0_fm4_a0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bnj_b0_fm4_b0 = {
+	.fw_name_pre = IWL_BNJ_B_FM4_B_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bnj_a0_gf_a0 = {
 	.fw_name_pre = IWL_BNJ_A_GF_A_FW_PRE,
 	.uhb_supported = true,
@@ -1059,6 +1070,7 @@ MODULE_FIRMWARE(IWL_BZ_A_FM_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_GL_A_FM_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_FM_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_FM4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BNJ_B_FM4_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index cfa5e1b3c3f6..eaa0ff2736c5 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -659,6 +659,7 @@ extern const struct iwl_cfg iwl_cfg_bnj_a0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_gf4_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_fm_b0;
+extern const struct iwl_cfg iwl_cfg_bnj_b0_fm4_b0;
 #endif /* CONFIG_IWLMVM */
 
 #endif /* __IWL_CONFIG_H__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 99768d6a6032..8aa8a678475c 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1193,6 +1193,11 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_fm4_a0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
+		      iwl_cfg_bnj_b0_fm4_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
-- 
cgit v1.2.3-70-g09d2


From beddcdc489864e234af844d906547b1cb70e9036 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:30 +0200
Subject: wifi: iwlwifi: rs-fw: break out for unsupported bandwidth

Currently the for loop runs also over unsupported bandwidth in the
command, shorten the path in case we don't support it.

Also use the right macro for setting BW20.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.0264ba9df63b.I6c7c9efc806e0ffb7cb3b6051b2d109646e8708c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index f30eeab5505b..e3fb1b2cea6d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -337,10 +337,14 @@ static void rs_fw_eht_set_enabled_rates(const struct ieee80211_sta *sta,
 		const struct ieee80211_eht_mcs_nss_supp_bw *mcs_tx =
 			rs_fw_rs_mcs2eht_mcs(bw, eht_tx_mcs);
 
-		/* got unsuppored index for bw */
+		/* got unsupported index for bw */
 		if (!mcs_rx || !mcs_tx)
 			continue;
 
+		/* break out if we don't support the bandwidth */
+		if (cmd->max_ch_width < (bw + IWL_TLC_MNG_CH_WIDTH_80MHZ))
+			break;
+
 		rs_fw_set_eht_mcs_nss(cmd->ht_rates, bw,
 				      MAX_NSS_MCS(9, mcs_rx, mcs_tx), GENMASK(9, 0));
 		rs_fw_set_eht_mcs_nss(cmd->ht_rates, bw,
@@ -550,7 +554,7 @@ void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	struct iwl_tlc_config_cmd_v4 cfg_cmd = {
 		.sta_id = mvmsta->sta_id,
 		.max_ch_width = update ?
-			rs_fw_bw_from_sta_bw(sta) : RATE_MCS_CHAN_WIDTH_20,
+			rs_fw_bw_from_sta_bw(sta) : IWL_TLC_MNG_CH_WIDTH_20MHZ,
 		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, sta, sband)),
 		.chains = rs_fw_set_active_chains(iwl_mvm_get_valid_tx_ant(mvm)),
 		.sgi_ch_width_supp = rs_fw_sgi_cw_support(sta),
-- 
cgit v1.2.3-70-g09d2


From 774302d2d3ad916f415d651362550daf22f630b7 Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:31 +0200
Subject: wifi: iwlwifi: mvm: clean up duplicated defines

VHT, HE and EHT rates use the same bits for NSS, so no need for
defines per PHY version.

Also use spatch to replace bit manipulation with FIELD_GET:

  @@
  identifier rate;
  @@
  -((rate & RATE_MCS_NSS_MSK) >> RATE_MCS_NSS_POS)
  +FIELD_GET(RATE_MCS_NSS_MSK, rate)

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.167ed9477aa8.Ibd8e71d31896e8d8f067ce4e3a6e9a0e86c78f3f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rs.h |  3 ---
 drivers/net/wireless/intel/iwlwifi/fw/rs.c     |  4 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c    | 17 ++++++-----------
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c    |  6 ++----
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c    |  7 +++----
 5 files changed, 13 insertions(+), 24 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
index 1d372a09ebb4..c9a48fc5fac8 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
@@ -373,9 +373,6 @@ enum {
 
 /* Bit 4-5: (0) SISO, (1) MIMO2 (2) MIMO3 */
 #define RATE_VHT_MCS_RATE_CODE_MSK	0xf
-#define RATE_VHT_MCS_NSS_POS		4
-#define RATE_VHT_MCS_NSS_MSK		(3 << RATE_VHT_MCS_NSS_POS)
-#define RATE_VHT_MCS_MIMO2_MSK		BIT(RATE_VHT_MCS_NSS_POS)
 
 /*
  * Legacy OFDM rate format for bits 7:0
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/rs.c b/drivers/net/wireless/intel/iwlwifi/fw/rs.c
index e128d2e07f38..b09e68dbf5a9 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/rs.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2021 Intel Corporation
+ * Copyright (C) 2021-2022 Intel Corporation
  */
 
 #include <net/mac80211.h>
@@ -126,7 +126,7 @@ u32 iwl_new_rate_from_v1(u32 rate_v1)
 		   rate_v1 & RATE_MCS_HE_MSK_V1) {
 		rate_v2 |= rate_v1 & RATE_VHT_MCS_RATE_CODE_MSK;
 
-		rate_v2 |= rate_v1 & RATE_VHT_MCS_MIMO2_MSK;
+		rate_v2 |= rate_v1 & RATE_MCS_NSS_MSK;
 
 		if (rate_v1 & RATE_MCS_HE_MSK_V1) {
 			u32 he_type_bits = rate_v1 & RATE_MCS_HE_TYPE_MSK_V1;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 0b50b816684a..1f81dff71bc4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /******************************************************************************
  *
- * Copyright(c) 2005 - 2014, 2018 - 2021 Intel Corporation. All rights reserved.
+ * Copyright(c) 2005 - 2014, 2018 - 2022 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
  * Copyright(c) 2016 - 2017 Intel Deutschland GmbH
  *****************************************************************************/
@@ -895,8 +895,7 @@ static int rs_rate_from_ucode_rate(const u32 ucode_rate,
 			WARN_ON_ONCE(1);
 		}
 	} else if (ucode_rate & RATE_MCS_VHT_MSK_V1) {
-		nss = ((ucode_rate & RATE_VHT_MCS_NSS_MSK) >>
-		       RATE_VHT_MCS_NSS_POS) + 1;
+		nss = FIELD_GET(RATE_MCS_NSS_MSK, ucode_rate) + 1;
 
 		if (nss == 1) {
 			rate->type = LQ_VHT_SISO;
@@ -910,8 +909,7 @@ static int rs_rate_from_ucode_rate(const u32 ucode_rate,
 			WARN_ON_ONCE(1);
 		}
 	} else if (ucode_rate & RATE_MCS_HE_MSK_V1) {
-		nss = ((ucode_rate & RATE_VHT_MCS_NSS_MSK) >>
-		      RATE_VHT_MCS_NSS_POS) + 1;
+		nss = FIELD_GET(RATE_MCS_NSS_MSK, ucode_rate) + 1;
 
 		if (nss == 1) {
 			rate->type = LQ_HE_SISO;
@@ -2885,8 +2883,7 @@ void iwl_mvm_update_frame_stats(struct iwl_mvm *mvm, u32 rate, bool agg)
 		nss = ((rate & RATE_HT_MCS_NSS_MSK_V1) >> RATE_HT_MCS_NSS_POS_V1) + 1;
 	} else if (rate & RATE_MCS_VHT_MSK_V1) {
 		mvm->drv_rx_stats.vht_frames++;
-		nss = ((rate & RATE_VHT_MCS_NSS_MSK) >>
-		       RATE_VHT_MCS_NSS_POS) + 1;
+		nss = FIELD_GET(RATE_MCS_NSS_MSK, rate) + 1;
 	} else {
 		mvm->drv_rx_stats.legacy_frames++;
 	}
@@ -3665,8 +3662,7 @@ int rs_pretty_print_rate_v1(char *buf, int bufsz, const u32 rate)
 	if (rate & RATE_MCS_VHT_MSK_V1) {
 		type = "VHT";
 		mcs = rate & RATE_VHT_MCS_RATE_CODE_MSK;
-		nss = ((rate & RATE_VHT_MCS_NSS_MSK)
-		       >> RATE_VHT_MCS_NSS_POS) + 1;
+		nss = FIELD_GET(RATE_MCS_NSS_MSK, rate) + 1;
 	} else if (rate & RATE_MCS_HT_MSK_V1) {
 		type = "HT";
 		mcs = rate & RATE_HT_MCS_INDEX_MSK_V1;
@@ -3675,8 +3671,7 @@ int rs_pretty_print_rate_v1(char *buf, int bufsz, const u32 rate)
 	} else if (rate & RATE_MCS_HE_MSK_V1) {
 		type = "HE";
 		mcs = rate & RATE_VHT_MCS_RATE_CODE_MSK;
-		nss = ((rate & RATE_VHT_MCS_NSS_MSK)
-		       >> RATE_VHT_MCS_NSS_POS) + 1;
+		nss = FIELD_GET(RATE_MCS_NSS_MSK, rate) + 1;
 	} else {
 		type = "Unknown"; /* shouldn't happen */
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index ab9a51375c8a..d2ce414879aa 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -253,8 +253,7 @@ static void iwl_mvm_rx_handle_tcm(struct iwl_mvm *mvm,
 				ARRAY_SIZE(thresh_tpt)))
 			return;
 		thr = thresh_tpt[rate_n_flags & RATE_VHT_MCS_RATE_CODE_MSK];
-		thr *= 1 + ((rate_n_flags & RATE_VHT_MCS_NSS_MSK) >>
-					RATE_VHT_MCS_NSS_POS);
+		thr *= 1 + FIELD_GET(RATE_MCS_NSS_MSK, rate_n_flags);
 	}
 
 	thr <<= ((rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK_V1) >>
@@ -500,8 +499,7 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi,
 		u8 stbc = (rate_n_flags & RATE_MCS_STBC_MSK) >>
 				RATE_MCS_STBC_POS;
 		rx_status->nss =
-			((rate_n_flags & RATE_VHT_MCS_NSS_MSK) >>
-						RATE_VHT_MCS_NSS_POS) + 1;
+			FIELD_GET(RATE_MCS_NSS_MSK, rate_n_flags) + 1;
 		rx_status->rate_idx = rate_n_flags & RATE_VHT_MCS_RATE_CODE_MSK;
 		rx_status->encoding = RX_ENC_VHT;
 		rx_status->enc_flags |= stbc << RX_ENC_FLAG_STBC_SHIFT;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 9813d7fa1800..a6d69885cd3f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -1396,8 +1396,8 @@ void iwl_mvm_hwrate_to_tx_rate(u32 rate_n_flags,
 		r->idx = rate;
 	} else if (format ==  RATE_MCS_VHT_MSK) {
 		ieee80211_rate_set_vht(r, rate,
-				       ((rate_n_flags & RATE_MCS_NSS_MSK) >>
-					RATE_MCS_NSS_POS) + 1);
+				       FIELD_GET(RATE_MCS_NSS_MSK,
+						 rate_n_flags) + 1);
 		r->flags |= IEEE80211_TX_RC_VHT_MCS;
 	} else if (format == RATE_MCS_HE_MSK) {
 		/* mac80211 cannot do this without ieee80211_tx_status_ext()
@@ -1428,8 +1428,7 @@ void iwl_mvm_hwrate_to_tx_rate_v1(u32 rate_n_flags,
 	} else if (rate_n_flags & RATE_MCS_VHT_MSK_V1) {
 		ieee80211_rate_set_vht(
 			r, rate_n_flags & RATE_VHT_MCS_RATE_CODE_MSK,
-			((rate_n_flags & RATE_VHT_MCS_NSS_MSK) >>
-						RATE_VHT_MCS_NSS_POS) + 1);
+			FIELD_GET(RATE_MCS_NSS_MSK, rate_n_flags) + 1);
 		r->flags |= IEEE80211_TX_RC_VHT_MCS;
 	} else {
 		r->idx = iwl_mvm_legacy_rate_to_mac80211_idx(rate_n_flags,
-- 
cgit v1.2.3-70-g09d2


From 558f874ea049af3c0e2c25b4de2b7fcf3f40e43b Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Sun, 5 Mar 2023 14:16:32 +0200
Subject: wifi: iwlwifi: Update logs for yoyo reset sw changes

Update the log category for the reset-fw changes.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.852a6b5f95fa.Ie67bd28da65c7e42424cacb37495930475de2dad@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 8e0bc1f5f6c6..87366b70b17f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -1218,11 +1218,12 @@ iwl_dbg_tlv_tp_trigger(struct iwl_fw_runtime *fwrt, bool sync,
 		}
 
 		fwrt->trans->dbg.restart_required = FALSE;
-		IWL_DEBUG_INFO(fwrt, "WRT: tp %d, reset_fw %d\n",
-			       tp, dump_data.trig->reset_fw);
-		IWL_DEBUG_INFO(fwrt, "WRT: restart_required %d, last_tp_resetfw %d\n",
-			       fwrt->trans->dbg.restart_required,
-			       fwrt->trans->dbg.last_tp_resetfw);
+		IWL_DEBUG_FW(fwrt, "WRT: tp %d, reset_fw %d\n",
+			     tp, dump_data.trig->reset_fw);
+		IWL_DEBUG_FW(fwrt,
+			     "WRT: restart_required %d, last_tp_resetfw %d\n",
+			     fwrt->trans->dbg.restart_required,
+			     fwrt->trans->dbg.last_tp_resetfw);
 
 		if (fwrt->trans->trans_cfg->device_family ==
 		    IWL_DEVICE_FAMILY_9000) {
@@ -1235,18 +1236,19 @@ iwl_dbg_tlv_tp_trigger(struct iwl_fw_runtime *fwrt, bool sync,
 			IWL_DEBUG_FW(fwrt, "WRT: FW_ASSERT due to reset_fw_mode-no restart\n");
 		} else if (le32_to_cpu(dump_data.trig->reset_fw) ==
 			   IWL_FW_INI_RESET_FW_MODE_STOP_AND_RELOAD_FW) {
-			IWL_DEBUG_INFO(fwrt, "WRT: stop and reload firmware\n");
+			IWL_DEBUG_FW(fwrt, "WRT: stop and reload firmware\n");
 			fwrt->trans->dbg.restart_required = TRUE;
 		} else if (le32_to_cpu(dump_data.trig->reset_fw) ==
 			   IWL_FW_INI_RESET_FW_MODE_STOP_FW_ONLY) {
-			IWL_DEBUG_INFO(fwrt, "WRT: stop only and no reload firmware\n");
+			IWL_DEBUG_FW(fwrt,
+				     "WRT: stop only and no reload firmware\n");
 			fwrt->trans->dbg.restart_required = FALSE;
 			fwrt->trans->dbg.last_tp_resetfw =
 				le32_to_cpu(dump_data.trig->reset_fw);
 		} else if (le32_to_cpu(dump_data.trig->reset_fw) ==
 			   IWL_FW_INI_RESET_FW_MODE_NOTHING) {
-			IWL_DEBUG_INFO(fwrt,
-				       "WRT: nothing need to be done after debug collection\n");
+			IWL_DEBUG_FW(fwrt,
+				     "WRT: nothing need to be done after debug collection\n");
 		} else {
 			IWL_ERR(fwrt, "WRT: wrong resetfw %d\n",
 				le32_to_cpu(dump_data.trig->reset_fw));
-- 
cgit v1.2.3-70-g09d2


From 7696c07bfbded95ce22aa41b6411534d0d06990d Mon Sep 17 00:00:00 2001
From: Mordechay Goodstein <mordechay.goodstein@intel.com>
Date: Sun, 5 Mar 2023 14:16:33 +0200
Subject: wifi: iwlwifi: mvm: add EHT RU allocation to radiotap

FW new API added the info missing for update RU allocation,
so use the new API to update radiotap information.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.b16acaa4bad1.I53afa03058dbd2cd8afbaf5e82596c8ed501a476@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rx.h | 15 ++---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c  | 90 +++++++++++++++++++++++++-
 2 files changed, 95 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
index 97e946e70279..fdd8b01f09e4 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
@@ -414,7 +414,7 @@ enum iwl_rx_phy_eht_data2 {
 	/* OFDM_RX_VECTOR_COMMON_RU_ALLOC_0_OUT */
 	IWL_RX_PHY_DATA2_EHT_MU_EXT_RU_ALLOC_A1	= 0x000001ff,
 	IWL_RX_PHY_DATA2_EHT_MU_EXT_RU_ALLOC_A2	= 0x0003fe00,
-	IWL_RX_PHY_DATA2_EHT_MU_EXT_RU_ALLOC_A3	= 0x01fc0000,
+	IWL_RX_PHY_DATA2_EHT_MU_EXT_RU_ALLOC_B1	= 0x07fc0000,
 
 	/* info type: EHT-TB-EXT */
 	IWL_RX_PHY_DATA2_EHT_TB_EXT_TRIG_SIGA1	= 0xffffffff,
@@ -424,19 +424,18 @@ enum iwl_rx_phy_eht_data2 {
 enum iwl_rx_phy_eht_data3 {
 	/* info type: EHT-MU-EXT */
 	/* OFDM_RX_VECTOR_COMMON_RU_ALLOC_1_OUT */
-	IWL_RX_PHY_DATA3_EHT_MU_EXT_RU_ALLOC_B1	= 0x000001ff,
-	IWL_RX_PHY_DATA3_EHT_MU_EXT_RU_ALLOC_B2	= 0x0003fe00,
-	IWL_RX_PHY_DATA3_EHT_MU_EXT_RU_ALLOC_B3	= 0x01fc0000,
+	IWL_RX_PHY_DATA3_EHT_MU_EXT_RU_ALLOC_B2	= 0x000001ff,
+	IWL_RX_PHY_DATA3_EHT_MU_EXT_RU_ALLOC_C1	= 0x0003fe00,
+	IWL_RX_PHY_DATA3_EHT_MU_EXT_RU_ALLOC_C2	= 0x07fc0000,
 };
 
 /* goes into Metadata DW 4 */
 enum iwl_rx_phy_eht_data4 {
 	/* info type: EHT-MU-EXT */
 	/* OFDM_RX_VECTOR_COMMON_RU_ALLOC_2_OUT */
-	IWL_RX_PHY_DATA4_EHT_MU_EXT_RU_ALLOC_C1	= 0x000001ff,
-	IWL_RX_PHY_DATA4_EHT_MU_EXT_RU_ALLOC_C2	= 0x0003fe00,
-	IWL_RX_PHY_DATA4_EHT_MU_EXT_RU_ALLOC_C3	= 0x01fc0000,
-	IWL_RX_PHY_DATA4_EHT_MU_EXT_SIGB_MCS	= 0x18000000,
+	IWL_RX_PHY_DATA4_EHT_MU_EXT_RU_ALLOC_D1	= 0x000001ff,
+	IWL_RX_PHY_DATA4_EHT_MU_EXT_RU_ALLOC_D2	= 0x0003fe00,
+	IWL_RX_PHY_DATA4_EHT_MU_EXT_SIGB_MCS	= 0x000c0000,
 };
 
 /* goes into Metadata DW 16 */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index b545673fb150..c07ed33395b6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1468,17 +1468,54 @@ static void iwl_mvm_decode_he_phy_data(struct iwl_mvm *mvm,
 	(_usig)->value |= LE32_DEC_ENC(in_value, dec_bits, _enc_bits); \
 } while (0)
 
+#define __IWL_MVM_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) \
+	eht->data[(rt_data)] |= \
+		(cpu_to_le32 \
+		 (IEEE80211_RADIOTAP_EHT_DATA ## rt_data ## _RU_ALLOC_CC_ ## rt_ru ## _KNOWN) | \
+		 LE32_DEC_ENC(data ## fw_data, \
+			      IWL_RX_PHY_DATA ## fw_data ## _EHT_MU_EXT_RU_ALLOC_ ## fw_ru, \
+			      IEEE80211_RADIOTAP_EHT_DATA ## rt_data ## _RU_ALLOC_CC_ ## rt_ru))
+
+#define _IWL_MVM_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru)	\
+	__IWL_MVM_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru)
+
+#define IEEE80211_RADIOTAP_RU_DATA_1_1_1	1
+#define IEEE80211_RADIOTAP_RU_DATA_2_1_1	2
+#define IEEE80211_RADIOTAP_RU_DATA_1_1_2	2
+#define IEEE80211_RADIOTAP_RU_DATA_2_1_2	2
+#define IEEE80211_RADIOTAP_RU_DATA_1_2_1	3
+#define IEEE80211_RADIOTAP_RU_DATA_2_2_1	3
+#define IEEE80211_RADIOTAP_RU_DATA_1_2_2	3
+#define IEEE80211_RADIOTAP_RU_DATA_2_2_2	4
+
+#define IWL_RX_RU_DATA_A1			2
+#define IWL_RX_RU_DATA_A2			2
+#define IWL_RX_RU_DATA_B1			2
+#define IWL_RX_RU_DATA_B2			3
+#define IWL_RX_RU_DATA_C1			3
+#define IWL_RX_RU_DATA_C2			3
+#define IWL_RX_RU_DATA_D1			4
+#define IWL_RX_RU_DATA_D2			4
+
+#define IWL_MVM_ENC_EHT_RU(rt_ru, fw_ru)				\
+	_IWL_MVM_ENC_EHT_RU(IEEE80211_RADIOTAP_RU_DATA_ ## rt_ru,	\
+			    rt_ru,					\
+			    IWL_RX_RU_DATA_ ## fw_ru,			\
+			    fw_ru)
+
 static void iwl_mvm_decode_eht_ext_mu(struct iwl_mvm *mvm,
 				      struct iwl_mvm_rx_phy_data *phy_data,
 				      struct ieee80211_rx_status *rx_status,
 				      struct ieee80211_radiotap_eht *eht,
 				      struct ieee80211_radiotap_eht_usig *usig)
 {
-	__le32 data1 = phy_data->d1;
-
 	if (phy_data->with_data) {
+		__le32 data1 = phy_data->d1;
+		__le32 data2 = phy_data->d2;
+		__le32 data3 = phy_data->d3;
 		__le32 data4 = phy_data->eht_d4;
 		__le32 data5 = phy_data->d5;
+		u32 phy_bw = phy_data->rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK;
 
 		IWL_MVM_ENC_USIG_VALUE_MASK(usig, data5,
 					    IWL_RX_PHY_DATA5_EHT_TYPE_AND_COMP,
@@ -1493,6 +1530,55 @@ static void iwl_mvm_decode_eht_ext_mu(struct iwl_mvm *mvm,
 			(usig, data1, IWL_RX_PHY_DATA1_EHT_MU_NUM_SIG_SYM_USIGA2,
 			 IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS);
 
+		eht->user_info[0] |=
+			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID_KNOWN) |
+			LE32_DEC_ENC(data5, IWL_RX_PHY_DATA5_EHT_MU_STA_ID_USR,
+				     IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID);
+
+		eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_NR_NON_OFDMA_USERS_M);
+		eht->data[7] |= LE32_DEC_ENC
+			(data5, IWL_RX_PHY_DATA5_EHT_MU_NUM_USR_NON_OFDMA,
+			 IEEE80211_RADIOTAP_EHT_DATA7_NUM_OF_NON_OFDMA_USERS);
+
+		/*
+		 * Hardware labels the content channels/RU allocation values
+		 * as follows:
+		 *           Content Channel 1		Content Channel 2
+		 *   20 MHz: A1
+		 *   40 MHz: A1				B1
+		 *   80 MHz: A1 C1			B1 D1
+		 *  160 MHz: A1 C1 A2 C2		B1 D1 B2 D2
+		 *  320 MHz: A1 C1 A2 C2 A3 C3 A4 C4	B1 D1 B2 D2 B3 D3 B4 D4
+		 *
+		 * However firmware can only give us A1-D2, so the higher
+		 * frequencies are missing.
+		 */
+
+		switch (phy_bw) {
+		case RATE_MCS_CHAN_WIDTH_320:
+			/* additional values are missing in RX metadata */
+		case RATE_MCS_CHAN_WIDTH_160:
+			/* content channel 1 */
+			IWL_MVM_ENC_EHT_RU(1_2_1, A2);
+			IWL_MVM_ENC_EHT_RU(1_2_2, C2);
+			/* content channel 2 */
+			IWL_MVM_ENC_EHT_RU(2_2_1, B2);
+			IWL_MVM_ENC_EHT_RU(2_2_2, D2);
+			fallthrough;
+		case RATE_MCS_CHAN_WIDTH_80:
+			/* content channel 1 */
+			IWL_MVM_ENC_EHT_RU(1_1_2, C1);
+			/* content channel 2 */
+			IWL_MVM_ENC_EHT_RU(2_1_2, D1);
+			fallthrough;
+		case RATE_MCS_CHAN_WIDTH_40:
+			/* content channel 2 */
+			IWL_MVM_ENC_EHT_RU(2_1_1, B1);
+			fallthrough;
+		case RATE_MCS_CHAN_WIDTH_20:
+			IWL_MVM_ENC_EHT_RU(1_1_1, A1);
+			break;
+		}
 	} else {
 		__le32 usig_a1 = phy_data->rx_vec[0];
 		__le32 usig_a2 = phy_data->rx_vec[1];
-- 
cgit v1.2.3-70-g09d2


From b55c1f4ec535f87a4063a8cbb75b014cdfc41bcb Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Sun, 5 Mar 2023 14:16:34 +0200
Subject: wifi: iwlwifi: Do not include radiotap EHT user info if not needed

Do not include user information in radtiotap EHT data for EHT sounding
NDP as the frame doesn't include the user specific field. Instead,
encode the NSS and the beamforming information in the EHT data.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.ac6474ded9bd.I9655589e9afbacc16820f35f6f5d90c6a91b8b05@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 58 +++++++++++++++++----------
 1 file changed, 37 insertions(+), 21 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index c07ed33395b6..2ea1fbf81d3b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1845,6 +1845,7 @@ static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	struct ieee80211_radiotap_eht *eht;
 	struct ieee80211_radiotap_eht_usig *usig;
+	size_t eht_len = sizeof(*eht);
 
 	u32 rate_n_flags = phy_data->rate_n_flags;
 	u32 he_type = rate_n_flags & RATE_MCS_HE_TYPE_MSK;
@@ -1854,8 +1855,10 @@ static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
 	u32 bw;
 
 	/* u32 for 1 user_info */
-	eht = iwl_mvm_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT,
-				       sizeof(*eht) + sizeof(u32));
+	if (phy_data->with_data)
+		eht_len += sizeof(u32);
+
+	eht = iwl_mvm_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT, eht_len);
 
 	usig = iwl_mvm_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT_USIG,
 					sizeof(*usig));
@@ -1946,27 +1949,40 @@ static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
 				    rx_status->eht.gi));
 	}
 
-	eht->user_info[0] |= cpu_to_le32
-		(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS_KNOWN |
-		 IEEE80211_RADIOTAP_EHT_USER_INFO_CODING_KNOWN |
-		 IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_KNOWN_O |
-		 IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_KNOWN_O |
-		 IEEE80211_RADIOTAP_EHT_USER_INFO_DATA_FOR_USER);
-
-	if (rate_n_flags & RATE_MCS_BF_MSK)
-		eht->user_info[0] |=
-			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_O);
 
-	if (rate_n_flags & RATE_MCS_LDPC_MSK)
+	if (!phy_data->with_data) {
+		eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_NSS_S |
+					  IEEE80211_RADIOTAP_EHT_KNOWN_BEAMFORMED_S);
+		eht->data[7] |=
+			le32_encode_bits(le32_get_bits(phy_data->rx_vec[2],
+						       RX_NO_DATA_RX_VEC2_EHT_NSTS_MSK),
+					 IEEE80211_RADIOTAP_EHT_DATA7_NSS_S);
+		if (rate_n_flags & RATE_MCS_BF_MSK)
+			eht->data[7] |=
+				cpu_to_le32(IEEE80211_RADIOTAP_EHT_DATA7_BEAMFORMED_S);
+	} else {
 		eht->user_info[0] |=
-			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_CODING);
-
-	eht->user_info[0] |= cpu_to_le32
-		(FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS,
-			    FIELD_GET(RATE_VHT_MCS_RATE_CODE_MSK,
-				      rate_n_flags)) |
-		 FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O,
-			    FIELD_GET(RATE_MCS_NSS_MSK, rate_n_flags)));
+			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS_KNOWN |
+				    IEEE80211_RADIOTAP_EHT_USER_INFO_CODING_KNOWN |
+				    IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_KNOWN_O |
+				    IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_KNOWN_O |
+				    IEEE80211_RADIOTAP_EHT_USER_INFO_DATA_FOR_USER);
+
+		if (rate_n_flags & RATE_MCS_BF_MSK)
+			eht->user_info[0] |=
+				cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_BEAMFORMING_O);
+
+		if (rate_n_flags & RATE_MCS_LDPC_MSK)
+			eht->user_info[0] |=
+				cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_CODING);
+
+		eht->user_info[0] |= cpu_to_le32
+			(FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS,
+				    FIELD_GET(RATE_VHT_MCS_RATE_CODE_MSK,
+					      rate_n_flags)) |
+			 FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O,
+				    FIELD_GET(RATE_MCS_NSS_MSK, rate_n_flags)));
+	}
 }
 
 static void iwl_mvm_rx_he(struct iwl_mvm *mvm, struct sk_buff *skb,
-- 
cgit v1.2.3-70-g09d2


From da1185449c669076276027c600666286124eef9f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 5 Mar 2023 14:16:35 +0200
Subject: wifi: iwlwifi: mvm: fix EOF bit reporting

In monitor mode, we try to report the EOF bit on the
first MPDU of an A-MPDU (hardware duplicates this bit
over all MPDUs, so it's only trustable on the first).

However, due to reshuffling in an ealier commit, the
toggle_bit != mvm->ampdu_toggle logic can no longer
work since mvm->ampdu_toggle is now set before this
code runs.

Fix this by tracking the first_subframe status in the
phy data struct and using that instead of checking.

Fixes: f1490546bec9 ("wifi: iwlwifi: mvm: rxmq: refactor mac80211 rx_status setting")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230305124407.e273aa0d3fdc.I77db4cc247898eae8a98b80659386d6737052b95@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 28 ++++++++++-----------------
 1 file changed, 10 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 2ea1fbf81d3b..91556d43735a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1180,6 +1180,7 @@ struct iwl_mvm_rx_phy_data {
 	__le32 d0, d1, d2, d3, eht_d4, d5;
 	__le16 d4;
 	bool with_data;
+	bool first_subframe;
 	__le32 rx_vec[4];
 
 	u32 rate_n_flags;
@@ -1885,15 +1886,10 @@ static void iwl_mvm_rx_eht(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	/* update aggregation data for monitor sake on default queue */
 	if (!queue && (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) &&
-	    (phy_info & IWL_RX_MPDU_PHY_AMPDU)) {
-		bool toggle_bit = phy_info & IWL_RX_MPDU_PHY_AMPDU_TOGGLE;
-
-		/* toggle is switched whenever new aggregation starts */
-		if (toggle_bit != mvm->ampdu_toggle) {
-			rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN;
-			if (phy_data->d0 & cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF))
-				rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT;
-		}
+	    (phy_info & IWL_RX_MPDU_PHY_AMPDU) && phy_data->first_subframe) {
+		rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN;
+		if (phy_data->d0 & cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF))
+			rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT;
 	}
 
 	if (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD)
@@ -2036,15 +2032,10 @@ static void iwl_mvm_rx_he(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	/* update aggregation data for monitor sake on default queue */
 	if (!queue && (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) &&
-	    (phy_info & IWL_RX_MPDU_PHY_AMPDU)) {
-		bool toggle_bit = phy_info & IWL_RX_MPDU_PHY_AMPDU_TOGGLE;
-
-		/* toggle is switched whenever new aggregation starts */
-		if (toggle_bit != mvm->ampdu_toggle) {
-			rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN;
-			if (phy_data->d0 & cpu_to_le32(IWL_RX_PHY_DATA0_HE_DELIM_EOF))
-				rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT;
-		}
+	    (phy_info & IWL_RX_MPDU_PHY_AMPDU) && phy_data->first_subframe) {
+		rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN;
+		if (phy_data->d0 & cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF))
+			rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT;
 	}
 
 	if (he_type == RATE_MCS_HE_TYPE_EXT_SU &&
@@ -2447,6 +2438,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			if (mvm->ampdu_ref == 0)
 				mvm->ampdu_ref++;
 			mvm->ampdu_toggle = toggle_bit;
+			phy_data.first_subframe = true;
 		}
 		rx_status->ampdu_reference = mvm->ampdu_ref;
 	}
-- 
cgit v1.2.3-70-g09d2


From ae4fe46983007bc46d87dcb284a5e5851c3e1c84 Mon Sep 17 00:00:00 2001
From: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com>
Date: Thu, 16 Feb 2023 10:07:31 +0800
Subject: igc: Add qbv_config_change_errors counter

Add ConfigChangeError(qbv_config_change_errors) when user try to set the
AdminBaseTime to past value while the current GCL is still running.

The ConfigChangeError counter should not be increased when a gate control
list is scheduled into the future.

User can use "ethtool -S <interface> | grep qbv_config_change_errors"
command to check the counter values.

Signed-off-by: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h         |  1 +
 drivers/net/ethernet/intel/igc/igc_ethtool.c |  1 +
 drivers/net/ethernet/intel/igc/igc_main.c    |  1 +
 drivers/net/ethernet/intel/igc/igc_tsn.c     | 12 ++++++++++++
 4 files changed, 15 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index df3e26c0cf01..79cc99af4317 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -185,6 +185,7 @@ struct igc_adapter {
 	ktime_t base_time;
 	ktime_t cycle_time;
 	bool qbv_enable;
+	u32 qbv_config_change_errors;
 
 	/* OS defined structs */
 	struct pci_dev *pdev;
diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index 5a26a7805ef8..0e2cb00622d1 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -67,6 +67,7 @@ static const struct igc_stats igc_gstrings_stats[] = {
 	IGC_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared),
 	IGC_STAT("tx_lpi_counter", stats.tlpic),
 	IGC_STAT("rx_lpi_counter", stats.rlpic),
+	IGC_STAT("qbv_config_change_errors", qbv_config_change_errors),
 };
 
 #define IGC_NETDEV_STAT(_net_stat) { \
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 2928a6c73692..4992cca4029d 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -6049,6 +6049,7 @@ static int igc_tsn_clear_schedule(struct igc_adapter *adapter)
 
 	adapter->base_time = 0;
 	adapter->cycle_time = NSEC_PER_SEC;
+	adapter->qbv_config_change_errors = 0;
 
 	for (i = 0; i < adapter->num_tx_queues; i++) {
 		struct igc_ring *ring = adapter->tx_ring[i];
diff --git a/drivers/net/ethernet/intel/igc/igc_tsn.c b/drivers/net/ethernet/intel/igc/igc_tsn.c
index a386c8d61dbf..94a2b0dfb54d 100644
--- a/drivers/net/ethernet/intel/igc/igc_tsn.c
+++ b/drivers/net/ethernet/intel/igc/igc_tsn.c
@@ -114,6 +114,7 @@ static int igc_tsn_disable_offload(struct igc_adapter *adapter)
 static int igc_tsn_enable_offload(struct igc_adapter *adapter)
 {
 	struct igc_hw *hw = &adapter->hw;
+	bool tsn_mode_reconfig = false;
 	u32 tqavctrl, baset_l, baset_h;
 	u32 sec, nsec, cycle;
 	ktime_t base_time, systim;
@@ -226,6 +227,10 @@ skip_cbs:
 	}
 
 	tqavctrl = rd32(IGC_TQAVCTRL) & ~IGC_TQAVCTRL_FUTSCDDIS;
+
+	if (tqavctrl & IGC_TQAVCTRL_TRANSMIT_MODE_TSN)
+		tsn_mode_reconfig = true;
+
 	tqavctrl |= IGC_TQAVCTRL_TRANSMIT_MODE_TSN | IGC_TQAVCTRL_ENHANCED_QAV;
 
 	cycle = adapter->cycle_time;
@@ -239,6 +244,13 @@ skip_cbs:
 		s64 n = div64_s64(ktime_sub_ns(systim, base_time), cycle);
 
 		base_time = ktime_add_ns(base_time, (n + 1) * cycle);
+
+		/* Increase the counter if scheduling into the past while
+		 * Gate Control List (GCL) is running.
+		 */
+		if ((rd32(IGC_BASET_H) || rd32(IGC_BASET_L)) &&
+		    tsn_mode_reconfig)
+			adapter->qbv_config_change_errors++;
 	} else {
 		/* According to datasheet section 7.5.2.9.3.3, FutScdDis bit
 		 * has to be configured before the cycle time and base time.
-- 
cgit v1.2.3-70-g09d2


From 92a0dcb8427dc52eb0ec806206e2bdb23cc13840 Mon Sep 17 00:00:00 2001
From: Tan Tee Min <tee.min.tan@linux.intel.com>
Date: Thu, 16 Feb 2023 09:16:24 +0800
Subject: igc: offload queue max SDU from tc-taprio

Add support for configuring the max SDU for each Tx queue.
If not specified, keep the default.

Signed-off-by: Tan Tee Min <tee.min.tan@linux.intel.com>
Signed-off-by: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h      |  1 +
 drivers/net/ethernet/intel/igc/igc_hw.h   |  1 +
 drivers/net/ethernet/intel/igc/igc_main.c | 33 +++++++++++++++++++++++++++----
 3 files changed, 31 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 79cc99af4317..c0c00b8bd8d8 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -99,6 +99,7 @@ struct igc_ring {
 
 	u32 start_time;
 	u32 end_time;
+	u32 max_sdu;
 
 	/* CBS parameters */
 	bool cbs_enable;                /* indicates if CBS is enabled */
diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index 88680e3d613d..e1c572e0d4ef 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -273,6 +273,7 @@ struct igc_hw_stats {
 	u64 o2bspc;
 	u64 b2ospc;
 	u64 b2ogprc;
+	u64 txdrop;
 };
 
 struct net_device *igc_get_hw_dev(struct igc_hw *hw);
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 4992cca4029d..1e1245085a36 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1501,6 +1501,7 @@ static int igc_tso(struct igc_ring *tx_ring,
 static netdev_tx_t igc_xmit_frame_ring(struct sk_buff *skb,
 				       struct igc_ring *tx_ring)
 {
+	struct igc_adapter *adapter = netdev_priv(tx_ring->netdev);
 	bool first_flag = false, insert_empty = false;
 	u16 count = TXD_USE_COUNT(skb_headlen(skb));
 	__be16 protocol = vlan_get_protocol(skb);
@@ -1563,9 +1564,19 @@ done:
 	first->bytecount = skb->len;
 	first->gso_segs = 1;
 
-	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
-		struct igc_adapter *adapter = netdev_priv(tx_ring->netdev);
+	if (tx_ring->max_sdu > 0) {
+		u32 max_sdu = 0;
+
+		max_sdu = tx_ring->max_sdu +
+			  (skb_vlan_tagged(first->skb) ? VLAN_HLEN : 0);
 
+		if (first->bytecount > max_sdu) {
+			adapter->stats.txdrop++;
+			goto out_drop;
+		}
+	}
+
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
 		/* FIXME: add support for retrieving timestamps from
 		 * the other timer registers before skipping the
 		 * timestamping request.
@@ -4920,7 +4931,8 @@ void igc_update_stats(struct igc_adapter *adapter)
 	net_stats->tx_window_errors = adapter->stats.latecol;
 	net_stats->tx_carrier_errors = adapter->stats.tncrs;
 
-	/* Tx Dropped needs to be maintained elsewhere */
+	/* Tx Dropped */
+	net_stats->tx_dropped = adapter->stats.txdrop;
 
 	/* Management Stats */
 	adapter->stats.mgptc += rd32(IGC_MGTPTC);
@@ -6056,6 +6068,7 @@ static int igc_tsn_clear_schedule(struct igc_adapter *adapter)
 
 		ring->start_time = 0;
 		ring->end_time = NSEC_PER_SEC;
+		ring->max_sdu = 0;
 	}
 
 	return 0;
@@ -6139,6 +6152,16 @@ static int igc_save_qbv_schedule(struct igc_adapter *adapter,
 		}
 	}
 
+	for (i = 0; i < adapter->num_tx_queues; i++) {
+		struct igc_ring *ring = adapter->tx_ring[i];
+		struct net_device *dev = adapter->netdev;
+
+		if (qopt->max_sdu[i])
+			ring->max_sdu = qopt->max_sdu[i] + dev->hard_header_len;
+		else
+			ring->max_sdu = 0;
+	}
+
 	return 0;
 }
 
@@ -6237,8 +6260,10 @@ static int igc_tc_query_caps(struct igc_adapter *adapter,
 
 		caps->broken_mqprio = true;
 
-		if (hw->mac.type == igc_i225)
+		if (hw->mac.type == igc_i225) {
+			caps->supports_queue_max_sdu = true;
 			caps->gate_mask_per_txq = true;
+		}
 
 		return 0;
 	}
-- 
cgit v1.2.3-70-g09d2


From 6cc1b2fd736dbc4c846078459c3c00a0372247da Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Sun, 20 Nov 2022 10:15:11 +0200
Subject: igc: Clean up and optimize watchdog task

i225/i226 parts used only one media type copper. The copper media type is
not replaceable. Clean up the code accordingly, and remove the obsolete
media replacement and reset options.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h      |  2 --
 drivers/net/ethernet/intel/igc/igc_main.c | 17 -----------------
 2 files changed, 19 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index c0c00b8bd8d8..34aebf00a512 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -294,8 +294,6 @@ extern char igc_driver_name[];
 #define IGC_FLAG_PTP			BIT(8)
 #define IGC_FLAG_WOL_SUPPORTED		BIT(8)
 #define IGC_FLAG_NEED_LINK_UPDATE	BIT(9)
-#define IGC_FLAG_MEDIA_RESET		BIT(10)
-#define IGC_FLAG_MAS_ENABLE		BIT(12)
 #define IGC_FLAG_HAS_MSIX		BIT(13)
 #define IGC_FLAG_EEE			BIT(14)
 #define IGC_FLAG_VLAN_PROMISC		BIT(15)
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 1e1245085a36..2c40796c151f 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -5578,25 +5578,8 @@ no_wait:
 				mod_timer(&adapter->phy_info_timer,
 					  round_jiffies(jiffies + 2 * HZ));
 
-			/* link is down, time to check for alternate media */
-			if (adapter->flags & IGC_FLAG_MAS_ENABLE) {
-				if (adapter->flags & IGC_FLAG_MEDIA_RESET) {
-					schedule_work(&adapter->reset_task);
-					/* return immediately */
-					return;
-				}
-			}
 			pm_schedule_suspend(netdev->dev.parent,
 					    MSEC_PER_SEC * 5);
-
-		/* also check for alternate media here */
-		} else if (!netif_carrier_ok(netdev) &&
-			   (adapter->flags & IGC_FLAG_MAS_ENABLE)) {
-			if (adapter->flags & IGC_FLAG_MEDIA_RESET) {
-				schedule_work(&adapter->reset_task);
-				/* return immediately */
-				return;
-			}
 		}
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 3ecde2182adbb12b52b777d8fb4a599b43faf4f1 Mon Sep 17 00:00:00 2001
From: Daniel Müller <deso@posteo.net>
Date: Tue, 7 Mar 2023 21:55:04 +0000
Subject: libbpf: Fix theoretical u32 underflow in find_cd() function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Coverity reported a potential underflow of the offset variable used in
the find_cd() function. Switch to using a signed 64 bit integer for the
representation of offset to make sure we can never underflow.

Fixes: 1eebcb60633f ("libbpf: Implement basic zip archive parsing support")
Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230307215504.837321-1-deso@posteo.net
---
 tools/lib/bpf/zip.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/tools/lib/bpf/zip.c b/tools/lib/bpf/zip.c
index 8458c2dd0e3b..f561aa07438f 100644
--- a/tools/lib/bpf/zip.c
+++ b/tools/lib/bpf/zip.c
@@ -168,9 +168,8 @@ static int try_parse_end_of_cd(struct zip_archive *archive, __u32 offset)
 
 static int find_cd(struct zip_archive *archive)
 {
+	int64_t limit, offset;
 	int rc = -EINVAL;
-	int64_t limit;
-	__u32 offset;
 
 	if (archive->size <= sizeof(struct end_of_cd_record))
 		return -EINVAL;
-- 
cgit v1.2.3-70-g09d2


From 98e678e9bc5859dec605794730ad37d26789bd30 Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay12@gmail.com>
Date: Tue, 7 Mar 2023 12:04:39 +0000
Subject: libbpf: Refactor parse_usdt_arg() to re-use code

The parse_usdt_arg() function is defined differently for each
architecture but the last part of the function is repeated
verbatim for each architecture.

Refactor parse_usdt_arg() to fill the arg_sz and then do the repeated
post-processing in parse_usdt_spec().

Signed-off-by: Puranjay Mohan <puranjay12@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230307120440.25941-2-puranjay12@gmail.com
---
 tools/lib/bpf/usdt.c | 124 +++++++++++++++++----------------------------------
 1 file changed, 42 insertions(+), 82 deletions(-)

diff --git a/tools/lib/bpf/usdt.c b/tools/lib/bpf/usdt.c
index 75b411fc2c77..72faa8162e34 100644
--- a/tools/lib/bpf/usdt.c
+++ b/tools/lib/bpf/usdt.c
@@ -1141,12 +1141,13 @@ static int parse_usdt_note(Elf *elf, const char *path, GElf_Nhdr *nhdr,
 	return 0;
 }
 
-static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg);
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz);
 
 static int parse_usdt_spec(struct usdt_spec *spec, const struct usdt_note *note, __u64 usdt_cookie)
 {
+	struct usdt_arg_spec *arg;
 	const char *s;
-	int len;
+	int arg_sz, len;
 
 	spec->usdt_cookie = usdt_cookie;
 	spec->arg_cnt = 0;
@@ -1159,10 +1160,25 @@ static int parse_usdt_spec(struct usdt_spec *spec, const struct usdt_note *note,
 			return -E2BIG;
 		}
 
-		len = parse_usdt_arg(s, spec->arg_cnt, &spec->args[spec->arg_cnt]);
+		arg = &spec->args[spec->arg_cnt];
+		len = parse_usdt_arg(s, spec->arg_cnt, arg, &arg_sz);
 		if (len < 0)
 			return len;
 
+		arg->arg_signed = arg_sz < 0;
+		if (arg_sz < 0)
+			arg_sz = -arg_sz;
+
+		switch (arg_sz) {
+		case 1: case 2: case 4: case 8:
+			arg->arg_bitshift = 64 - arg_sz * 8;
+			break;
+		default:
+			pr_warn("usdt: unsupported arg #%d (spec '%s') size: %d\n",
+				spec->arg_cnt, s, arg_sz);
+			return -EINVAL;
+		}
+
 		s += len;
 		spec->arg_cnt++;
 	}
@@ -1219,13 +1235,13 @@ static int calc_pt_regs_off(const char *reg_name)
 	return -ENOENT;
 }
 
-static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg)
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
 {
 	char reg_name[16];
-	int arg_sz, len, reg_off;
+	int len, reg_off;
 	long off;
 
-	if (sscanf(arg_str, " %d @ %ld ( %%%15[^)] ) %n", &arg_sz, &off, reg_name, &len) == 3) {
+	if (sscanf(arg_str, " %d @ %ld ( %%%15[^)] ) %n", arg_sz, &off, reg_name, &len) == 3) {
 		/* Memory dereference case, e.g., -4@-20(%rbp) */
 		arg->arg_type = USDT_ARG_REG_DEREF;
 		arg->val_off = off;
@@ -1233,7 +1249,7 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		if (reg_off < 0)
 			return reg_off;
 		arg->reg_off = reg_off;
-	} else if (sscanf(arg_str, " %d @ ( %%%15[^)] ) %n", &arg_sz, reg_name, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ ( %%%15[^)] ) %n", arg_sz, reg_name, &len) == 2) {
 		/* Memory dereference case without offset, e.g., 8@(%rsp) */
 		arg->arg_type = USDT_ARG_REG_DEREF;
 		arg->val_off = 0;
@@ -1241,7 +1257,7 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		if (reg_off < 0)
 			return reg_off;
 		arg->reg_off = reg_off;
-	} else if (sscanf(arg_str, " %d @ %%%15s %n", &arg_sz, reg_name, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %%%15s %n", arg_sz, reg_name, &len) == 2) {
 		/* Register read case, e.g., -4@%eax */
 		arg->arg_type = USDT_ARG_REG;
 		arg->val_off = 0;
@@ -1250,7 +1266,7 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		if (reg_off < 0)
 			return reg_off;
 		arg->reg_off = reg_off;
-	} else if (sscanf(arg_str, " %d @ $%ld %n", &arg_sz, &off, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ $%ld %n", arg_sz, &off, &len) == 2) {
 		/* Constant value case, e.g., 4@$71 */
 		arg->arg_type = USDT_ARG_CONST;
 		arg->val_off = off;
@@ -1260,20 +1276,6 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		return -EINVAL;
 	}
 
-	arg->arg_signed = arg_sz < 0;
-	if (arg_sz < 0)
-		arg_sz = -arg_sz;
-
-	switch (arg_sz) {
-	case 1: case 2: case 4: case 8:
-		arg->arg_bitshift = 64 - arg_sz * 8;
-		break;
-	default:
-		pr_warn("usdt: unsupported arg #%d (spec '%s') size: %d\n",
-			arg_num, arg_str, arg_sz);
-		return -EINVAL;
-	}
-
 	return len;
 }
 
@@ -1281,13 +1283,13 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 
 /* Do not support __s390__ for now, since user_pt_regs is broken with -m31. */
 
-static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg)
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
 {
 	unsigned int reg;
-	int arg_sz, len;
+	int len;
 	long off;
 
-	if (sscanf(arg_str, " %d @ %ld ( %%r%u ) %n", &arg_sz, &off, &reg, &len) == 3) {
+	if (sscanf(arg_str, " %d @ %ld ( %%r%u ) %n", arg_sz, &off, &reg, &len) == 3) {
 		/* Memory dereference case, e.g., -2@-28(%r15) */
 		arg->arg_type = USDT_ARG_REG_DEREF;
 		arg->val_off = off;
@@ -1296,7 +1298,7 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 			return -EINVAL;
 		}
 		arg->reg_off = offsetof(user_pt_regs, gprs[reg]);
-	} else if (sscanf(arg_str, " %d @ %%r%u %n", &arg_sz, &reg, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %%r%u %n", arg_sz, &reg, &len) == 2) {
 		/* Register read case, e.g., -8@%r0 */
 		arg->arg_type = USDT_ARG_REG;
 		arg->val_off = 0;
@@ -1305,7 +1307,7 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 			return -EINVAL;
 		}
 		arg->reg_off = offsetof(user_pt_regs, gprs[reg]);
-	} else if (sscanf(arg_str, " %d @ %ld %n", &arg_sz, &off, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %ld %n", arg_sz, &off, &len) == 2) {
 		/* Constant value case, e.g., 4@71 */
 		arg->arg_type = USDT_ARG_CONST;
 		arg->val_off = off;
@@ -1315,20 +1317,6 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		return -EINVAL;
 	}
 
-	arg->arg_signed = arg_sz < 0;
-	if (arg_sz < 0)
-		arg_sz = -arg_sz;
-
-	switch (arg_sz) {
-	case 1: case 2: case 4: case 8:
-		arg->arg_bitshift = 64 - arg_sz * 8;
-		break;
-	default:
-		pr_warn("usdt: unsupported arg #%d (spec '%s') size: %d\n",
-			arg_num, arg_str, arg_sz);
-		return -EINVAL;
-	}
-
 	return len;
 }
 
@@ -1348,13 +1336,13 @@ static int calc_pt_regs_off(const char *reg_name)
 	return -ENOENT;
 }
 
-static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg)
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
 {
 	char reg_name[16];
-	int arg_sz, len, reg_off;
+	int len, reg_off;
 	long off;
 
-	if (sscanf(arg_str, " %d @ \[ %15[a-z0-9], %ld ] %n", &arg_sz, reg_name, &off, &len) == 3) {
+	if (sscanf(arg_str, " %d @ \[ %15[a-z0-9] , %ld ] %n", arg_sz, reg_name, &off, &len) == 3) {
 		/* Memory dereference case, e.g., -4@[sp, 96] */
 		arg->arg_type = USDT_ARG_REG_DEREF;
 		arg->val_off = off;
@@ -1362,7 +1350,7 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		if (reg_off < 0)
 			return reg_off;
 		arg->reg_off = reg_off;
-	} else if (sscanf(arg_str, " %d @ \[ %15[a-z0-9] ] %n", &arg_sz, reg_name, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ \[ %15[a-z0-9] ] %n", arg_sz, reg_name, &len) == 2) {
 		/* Memory dereference case, e.g., -4@[sp] */
 		arg->arg_type = USDT_ARG_REG_DEREF;
 		arg->val_off = 0;
@@ -1370,12 +1358,12 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		if (reg_off < 0)
 			return reg_off;
 		arg->reg_off = reg_off;
-	} else if (sscanf(arg_str, " %d @ %ld %n", &arg_sz, &off, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %ld %n", arg_sz, &off, &len) == 2) {
 		/* Constant value case, e.g., 4@5 */
 		arg->arg_type = USDT_ARG_CONST;
 		arg->val_off = off;
 		arg->reg_off = 0;
-	} else if (sscanf(arg_str, " %d @ %15[a-z0-9] %n", &arg_sz, reg_name, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %15[a-z0-9] %n", arg_sz, reg_name, &len) == 2) {
 		/* Register read case, e.g., -8@x4 */
 		arg->arg_type = USDT_ARG_REG;
 		arg->val_off = 0;
@@ -1388,20 +1376,6 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		return -EINVAL;
 	}
 
-	arg->arg_signed = arg_sz < 0;
-	if (arg_sz < 0)
-		arg_sz = -arg_sz;
-
-	switch (arg_sz) {
-	case 1: case 2: case 4: case 8:
-		arg->arg_bitshift = 64 - arg_sz * 8;
-		break;
-	default:
-		pr_warn("usdt: unsupported arg #%d (spec '%s') size: %d\n",
-			arg_num, arg_str, arg_sz);
-		return -EINVAL;
-	}
-
 	return len;
 }
 
@@ -1456,13 +1430,13 @@ static int calc_pt_regs_off(const char *reg_name)
 	return -ENOENT;
 }
 
-static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg)
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
 {
 	char reg_name[16];
-	int arg_sz, len, reg_off;
+	int len, reg_off;
 	long off;
 
-	if (sscanf(arg_str, " %d @ %ld ( %15[a-z0-9] ) %n", &arg_sz, &off, reg_name, &len) == 3) {
+	if (sscanf(arg_str, " %d @ %ld ( %15[a-z0-9] ) %n", arg_sz, &off, reg_name, &len) == 3) {
 		/* Memory dereference case, e.g., -8@-88(s0) */
 		arg->arg_type = USDT_ARG_REG_DEREF;
 		arg->val_off = off;
@@ -1470,12 +1444,12 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		if (reg_off < 0)
 			return reg_off;
 		arg->reg_off = reg_off;
-	} else if (sscanf(arg_str, " %d @ %ld %n", &arg_sz, &off, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %ld %n", arg_sz, &off, &len) == 2) {
 		/* Constant value case, e.g., 4@5 */
 		arg->arg_type = USDT_ARG_CONST;
 		arg->val_off = off;
 		arg->reg_off = 0;
-	} else if (sscanf(arg_str, " %d @ %15[a-z0-9] %n", &arg_sz, reg_name, &len) == 2) {
+	} else if (sscanf(arg_str, " %d @ %15[a-z0-9] %n", arg_sz, reg_name, &len) == 2) {
 		/* Register read case, e.g., -8@a1 */
 		arg->arg_type = USDT_ARG_REG;
 		arg->val_off = 0;
@@ -1488,26 +1462,12 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 		return -EINVAL;
 	}
 
-	arg->arg_signed = arg_sz < 0;
-	if (arg_sz < 0)
-		arg_sz = -arg_sz;
-
-	switch (arg_sz) {
-	case 1: case 2: case 4: case 8:
-		arg->arg_bitshift = 64 - arg_sz * 8;
-		break;
-	default:
-		pr_warn("usdt: unsupported arg #%d (spec '%s') size: %d\n",
-			arg_num, arg_str, arg_sz);
-		return -EINVAL;
-	}
-
 	return len;
 }
 
 #else
 
-static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg)
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
 {
 	pr_warn("usdt: libbpf doesn't support USDTs on current architecture\n");
 	return -ENOTSUP;
-- 
cgit v1.2.3-70-g09d2


From 720d93b60aec22abcb2a5b6d8de472f3a50694b1 Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay12@gmail.com>
Date: Tue, 7 Mar 2023 12:04:40 +0000
Subject: libbpf: USDT arm arg parsing support

Parsing of USDT arguments is architecture-specific; on arm it is
relatively easy since registers used are r[0-10], fp, ip, sp, lr,
pc. Format is slightly different compared to aarch64; forms are

- "size @ [ reg, #offset ]" for dereferences, for example
  "-8 @ [ sp, #76 ]" ; " -4 @ [ sp ]"
- "size @ reg" for register values; for example
  "-4@r0"
- "size @ #value" for raw values; for example
  "-8@#1"

Add support for parsing USDT arguments for ARM architecture.

To test the above changes QEMU's virt[1] board with cortex-a15
CPU was used. libbpf-bootstrap's usdt example[2] was modified to attach
to a test program with DTRACE_PROBE1/2/3/4... probes to test different
combinations.

[1] https://www.qemu.org/docs/master/system/arm/virt.html
[2] https://github.com/libbpf/libbpf-bootstrap/blob/master/examples/c/usdt.bpf.c

Signed-off-by: Puranjay Mohan <puranjay12@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230307120440.25941-3-puranjay12@gmail.com
---
 tools/lib/bpf/usdt.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 80 insertions(+)

diff --git a/tools/lib/bpf/usdt.c b/tools/lib/bpf/usdt.c
index 72faa8162e34..b8402e3f9eb2 100644
--- a/tools/lib/bpf/usdt.c
+++ b/tools/lib/bpf/usdt.c
@@ -1465,6 +1465,86 @@ static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec
 	return len;
 }
 
+#elif defined(__arm__)
+
+static int calc_pt_regs_off(const char *reg_name)
+{
+	static struct {
+		const char *name;
+		size_t pt_regs_off;
+	} reg_map[] = {
+		{ "r0", offsetof(struct pt_regs, uregs[0]) },
+		{ "r1", offsetof(struct pt_regs, uregs[1]) },
+		{ "r2", offsetof(struct pt_regs, uregs[2]) },
+		{ "r3", offsetof(struct pt_regs, uregs[3]) },
+		{ "r4", offsetof(struct pt_regs, uregs[4]) },
+		{ "r5", offsetof(struct pt_regs, uregs[5]) },
+		{ "r6", offsetof(struct pt_regs, uregs[6]) },
+		{ "r7", offsetof(struct pt_regs, uregs[7]) },
+		{ "r8", offsetof(struct pt_regs, uregs[8]) },
+		{ "r9", offsetof(struct pt_regs, uregs[9]) },
+		{ "r10", offsetof(struct pt_regs, uregs[10]) },
+		{ "fp", offsetof(struct pt_regs, uregs[11]) },
+		{ "ip", offsetof(struct pt_regs, uregs[12]) },
+		{ "sp", offsetof(struct pt_regs, uregs[13]) },
+		{ "lr", offsetof(struct pt_regs, uregs[14]) },
+		{ "pc", offsetof(struct pt_regs, uregs[15]) },
+	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(reg_map); i++) {
+		if (strcmp(reg_name, reg_map[i].name) == 0)
+			return reg_map[i].pt_regs_off;
+	}
+
+	pr_warn("usdt: unrecognized register '%s'\n", reg_name);
+	return -ENOENT;
+}
+
+static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
+{
+	char reg_name[16];
+	int len, reg_off;
+	long off;
+
+	if (sscanf(arg_str, " %d @ \[ %15[a-z0-9] , #%ld ] %n",
+		   arg_sz, reg_name, &off, &len) == 3) {
+		/* Memory dereference case, e.g., -4@[fp, #96] */
+		arg->arg_type = USDT_ARG_REG_DEREF;
+		arg->val_off = off;
+		reg_off = calc_pt_regs_off(reg_name);
+		if (reg_off < 0)
+			return reg_off;
+		arg->reg_off = reg_off;
+	} else if (sscanf(arg_str, " %d @ \[ %15[a-z0-9] ] %n", arg_sz, reg_name, &len) == 2) {
+		/* Memory dereference case, e.g., -4@[sp] */
+		arg->arg_type = USDT_ARG_REG_DEREF;
+		arg->val_off = 0;
+		reg_off = calc_pt_regs_off(reg_name);
+		if (reg_off < 0)
+			return reg_off;
+		arg->reg_off = reg_off;
+	} else if (sscanf(arg_str, " %d @ #%ld %n", arg_sz, &off, &len) == 2) {
+		/* Constant value case, e.g., 4@#5 */
+		arg->arg_type = USDT_ARG_CONST;
+		arg->val_off = off;
+		arg->reg_off = 0;
+	} else if (sscanf(arg_str, " %d @ %15[a-z0-9] %n", arg_sz, reg_name, &len) == 2) {
+		/* Register read case, e.g., -8@r4 */
+		arg->arg_type = USDT_ARG_REG;
+		arg->val_off = 0;
+		reg_off = calc_pt_regs_off(reg_name);
+		if (reg_off < 0)
+			return reg_off;
+		arg->reg_off = reg_off;
+	} else {
+		pr_warn("usdt: unrecognized arg #%d spec '%s'\n", arg_num, arg_str);
+		return -EINVAL;
+	}
+
+	return len;
+}
+
 #else
 
 static int parse_usdt_arg(const char *arg_str, int arg_num, struct usdt_arg_spec *arg, int *arg_sz)
-- 
cgit v1.2.3-70-g09d2


From 0194b64578e905dc8f112e641a71c306bd58ddde Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:51:35 +0100
Subject: net: phy: improve phy_read_poll_timeout

cond sometimes is (val & MASK) what may result in a false positive
if val is a negative errno. We shouldn't evaluate cond if val < 0.
This has no functional impact here, but it's not nice.
Therefore switch order of the checks.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/6d8274ac-4344-23b4-d9a3-cad4c39517d4@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/phy.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/include/linux/phy.h b/include/linux/phy.h
index 36bf0bbc8efa..fefd5091bc24 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1130,16 +1130,15 @@ static inline int phy_read(struct phy_device *phydev, u32 regnum)
 #define phy_read_poll_timeout(phydev, regnum, val, cond, sleep_us, \
 				timeout_us, sleep_before_read) \
 ({ \
-	int __ret = read_poll_timeout(phy_read, val, (cond) || val < 0, \
+	int __ret = read_poll_timeout(phy_read, val, val < 0 || (cond), \
 		sleep_us, timeout_us, sleep_before_read, phydev, regnum); \
-	if (val <  0) \
+	if (val < 0) \
 		__ret = val; \
 	if (__ret) \
 		phydev_err(phydev, "%s failed: %d\n", __func__, __ret); \
 	__ret; \
 })
 
-
 /**
  * __phy_read - convenience function for reading a given PHY register
  * @phydev: the phy_device struct
-- 
cgit v1.2.3-70-g09d2


From 40bbae583ec38ea31e728bf42a4ea72bded22ab6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 6 Mar 2023 20:43:13 +0000
Subject: net: remove enum skb_free_reason

enum skb_drop_reason is more generic, we can adopt it instead.

Provide dev_kfree_skb_irq_reason() and dev_kfree_skb_any_reason().

This means drivers can use more precise drop reasons if they want to.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Yunsheng Lin <linyunsheng@huawei.com>
Link: https://lore.kernel.org/r/20230306204313.10492-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h | 18 +++++++-----------
 net/core/dev.c            | 20 +++++++++-----------
 2 files changed, 16 insertions(+), 22 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 6a14b7b11766..ee483071cf59 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -52,6 +52,7 @@
 #include <linux/rbtree.h>
 #include <net/net_trackers.h>
 #include <net/net_debug.h>
+#include <net/dropreason.h>
 
 struct netpoll_info;
 struct device;
@@ -3804,13 +3805,8 @@ static inline unsigned int get_netdev_rx_queue_index(
 
 int netif_get_num_default_rss_queues(void);
 
-enum skb_free_reason {
-	SKB_REASON_CONSUMED,
-	SKB_REASON_DROPPED,
-};
-
-void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason);
-void __dev_kfree_skb_any(struct sk_buff *skb, enum skb_free_reason reason);
+void dev_kfree_skb_irq_reason(struct sk_buff *skb, enum skb_drop_reason reason);
+void dev_kfree_skb_any_reason(struct sk_buff *skb, enum skb_drop_reason reason);
 
 /*
  * It is not allowed to call kfree_skb() or consume_skb() from hardware
@@ -3833,22 +3829,22 @@ void __dev_kfree_skb_any(struct sk_buff *skb, enum skb_free_reason reason);
  */
 static inline void dev_kfree_skb_irq(struct sk_buff *skb)
 {
-	__dev_kfree_skb_irq(skb, SKB_REASON_DROPPED);
+	dev_kfree_skb_irq_reason(skb, SKB_DROP_REASON_NOT_SPECIFIED);
 }
 
 static inline void dev_consume_skb_irq(struct sk_buff *skb)
 {
-	__dev_kfree_skb_irq(skb, SKB_REASON_CONSUMED);
+	dev_kfree_skb_irq_reason(skb, SKB_CONSUMED);
 }
 
 static inline void dev_kfree_skb_any(struct sk_buff *skb)
 {
-	__dev_kfree_skb_any(skb, SKB_REASON_DROPPED);
+	dev_kfree_skb_any_reason(skb, SKB_DROP_REASON_NOT_SPECIFIED);
 }
 
 static inline void dev_consume_skb_any(struct sk_buff *skb)
 {
-	__dev_kfree_skb_any(skb, SKB_REASON_CONSUMED);
+	dev_kfree_skb_any_reason(skb, SKB_CONSUMED);
 }
 
 u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp,
diff --git a/net/core/dev.c b/net/core/dev.c
index 253584777101..c7853192563d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3075,7 +3075,7 @@ void __netif_schedule(struct Qdisc *q)
 EXPORT_SYMBOL(__netif_schedule);
 
 struct dev_kfree_skb_cb {
-	enum skb_free_reason reason;
+	enum skb_drop_reason reason;
 };
 
 static struct dev_kfree_skb_cb *get_kfree_skb_cb(const struct sk_buff *skb)
@@ -3108,7 +3108,7 @@ void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 }
 EXPORT_SYMBOL(netif_tx_wake_queue);
 
-void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason)
+void dev_kfree_skb_irq_reason(struct sk_buff *skb, enum skb_drop_reason reason)
 {
 	unsigned long flags;
 
@@ -3128,18 +3128,16 @@ void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason)
 	raise_softirq_irqoff(NET_TX_SOFTIRQ);
 	local_irq_restore(flags);
 }
-EXPORT_SYMBOL(__dev_kfree_skb_irq);
+EXPORT_SYMBOL(dev_kfree_skb_irq_reason);
 
-void __dev_kfree_skb_any(struct sk_buff *skb, enum skb_free_reason reason)
+void dev_kfree_skb_any_reason(struct sk_buff *skb, enum skb_drop_reason reason)
 {
 	if (in_hardirq() || irqs_disabled())
-		__dev_kfree_skb_irq(skb, reason);
-	else if (unlikely(reason == SKB_REASON_DROPPED))
-		kfree_skb(skb);
+		dev_kfree_skb_irq_reason(skb, reason);
 	else
-		consume_skb(skb);
+		kfree_skb_reason(skb, reason);
 }
-EXPORT_SYMBOL(__dev_kfree_skb_any);
+EXPORT_SYMBOL(dev_kfree_skb_any_reason);
 
 
 /**
@@ -5020,11 +5018,11 @@ static __latent_entropy void net_tx_action(struct softirq_action *h)
 			clist = clist->next;
 
 			WARN_ON(refcount_read(&skb->users));
-			if (likely(get_kfree_skb_cb(skb)->reason == SKB_REASON_CONSUMED))
+			if (likely(get_kfree_skb_cb(skb)->reason == SKB_CONSUMED))
 				trace_consume_skb(skb, net_tx_action);
 			else
 				trace_kfree_skb(skb, net_tx_action,
-						SKB_DROP_REASON_NOT_SPECIFIED);
+						get_kfree_skb_cb(skb)->reason);
 
 			if (skb->fclone != SKB_FCLONE_UNAVAILABLE)
 				__kfree_skb(skb);
-- 
cgit v1.2.3-70-g09d2


From 4310e2f42030ffda0997fb87e2ebdd9711030218 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 23:10:57 +0100
Subject: net: phy: smsc: simplify lan95xx_config_aneg_ext

lan95xx_config_aneg_ext() can be simplified by using phy_set_bits().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/3da785c7-3ef8-b5d3-89a0-340f550be3c2@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/smsc.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index ac7481ce2fc1..af89f3ef1c4f 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -178,18 +178,15 @@ static int lan87xx_config_aneg(struct phy_device *phydev)
 
 static int lan95xx_config_aneg_ext(struct phy_device *phydev)
 {
-	int rc;
-
-	if (phydev->phy_id != 0x0007c0f0) /* not (LAN9500A or LAN9505A) */
-		return lan87xx_config_aneg(phydev);
+	if (phydev->phy_id == 0x0007c0f0) { /* LAN9500A or LAN9505A */
+		/* Extend Manual AutoMDIX timer */
+		int rc = phy_set_bits(phydev, PHY_EDPD_CONFIG,
+				      PHY_EDPD_CONFIG_EXT_CROSSOVER_);
 
-	/* Extend Manual AutoMDIX timer */
-	rc = phy_read(phydev, PHY_EDPD_CONFIG);
-	if (rc < 0)
-		return rc;
+		if (rc < 0)
+			return rc;
+	}
 
-	rc |= PHY_EDPD_CONFIG_EXT_CROSSOVER_;
-	phy_write(phydev, PHY_EDPD_CONFIG, rc);
 	return lan87xx_config_aneg(phydev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 2549347972a8e94a9261c125e494955d873b1a9f Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Mon, 6 Mar 2023 17:40:28 -0600
Subject: netxen_nic: Replace fake flex-array with flexible-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Transform zero-length array into flexible-array member in struct
nx_cardrsp_rx_ctx_t.

Address the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c:361:26: warning: array subscript <unknown> is outside array bounds of ‘char[0]’ [-Warray-bounds=]
drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c:372:25: warning: array subscript <unknown> is outside array bounds of ‘char[0]’ [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/265
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/ZAZ57I6WdQEwWh7v@work
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/netxen/netxen_nic.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h
index f13fa7396aef..3d36d23df0c6 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h
@@ -854,7 +854,7 @@ typedef struct {
 	   The following is packed:
 	   - N cardrsp_rds_rings
 	   - N cardrs_sds_rings */
-	char data[0];
+	char data[];
 } nx_cardrsp_rx_ctx_t;
 
 #define SIZEOF_HOSTRQ_RX(HOSTRQ_RX, rds_rings, sds_rings)	\
-- 
cgit v1.2.3-70-g09d2


From 8ca5a5790b9a1ce147484d2a2c4e66d2553f3d6c Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Mon, 6 Mar 2023 08:07:38 -0800
Subject: net-timestamp: extend SOF_TIMESTAMPING_OPT_ID to HW timestamps

When the feature was added it was enabled for SW timestamps only but
with current hardware the same out-of-order timestamps can be seen.
Let's expand the area for the feature to all types of timestamps.

Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/ip_output.c  | 2 +-
 net/ipv6/ip6_output.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 4e4e308c3230..e7bef36ce26f 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -990,7 +990,7 @@ static int __ip_append_data(struct sock *sk,
 	mtu = cork->gso_size ? IP_MAX_MTU : cork->fragsize;
 	paged = !!cork->gso_size;
 
-	if (cork->tx_flags & SKBTX_ANY_SW_TSTAMP &&
+	if (cork->tx_flags & SKBTX_ANY_TSTAMP &&
 	    sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID)
 		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index c314fdde0097..4ce3f9d3bc8a 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1500,7 +1500,7 @@ static int __ip6_append_data(struct sock *sk,
 	mtu = cork->gso_size ? IP6_MAX_MTU : cork->fragsize;
 	orig_mtu = mtu;
 
-	if (cork->tx_flags & SKBTX_ANY_SW_TSTAMP &&
+	if (cork->tx_flags & SKBTX_ANY_TSTAMP &&
 	    sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID)
 		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
 
-- 
cgit v1.2.3-70-g09d2


From 91c8643578a21e435c412ffbe902bb4b4773e262 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:23:15 +0100
Subject: r8169: use spinlock to protect mac ocp register access
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For disabling ASPM during NAPI poll we'll have to access mac ocp
registers in atomic context. This could result in races because
a mac ocp read consists of a write to register OCPDR, followed
by a read from the same register. Therefore add a spinlock to
protect access to mac ocp registers.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 37 +++++++++++++++++++++++++++----
 1 file changed, 33 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 45147a1016be..259eac5b0616 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -613,6 +613,8 @@ struct rtl8169_private {
 		struct work_struct work;
 	} wk;
 
+	spinlock_t mac_ocp_lock;
+
 	unsigned supports_gmii:1;
 	unsigned aspm_manageable:1;
 	dma_addr_t counters_phys_addr;
@@ -847,7 +849,7 @@ static int r8168_phy_ocp_read(struct rtl8169_private *tp, u32 reg)
 		(RTL_R32(tp, GPHY_OCP) & 0xffff) : -ETIMEDOUT;
 }
 
-static void r8168_mac_ocp_write(struct rtl8169_private *tp, u32 reg, u32 data)
+static void __r8168_mac_ocp_write(struct rtl8169_private *tp, u32 reg, u32 data)
 {
 	if (rtl_ocp_reg_failure(reg))
 		return;
@@ -855,7 +857,16 @@ static void r8168_mac_ocp_write(struct rtl8169_private *tp, u32 reg, u32 data)
 	RTL_W32(tp, OCPDR, OCPAR_FLAG | (reg << 15) | data);
 }
 
-static u16 r8168_mac_ocp_read(struct rtl8169_private *tp, u32 reg)
+static void r8168_mac_ocp_write(struct rtl8169_private *tp, u32 reg, u32 data)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&tp->mac_ocp_lock, flags);
+	__r8168_mac_ocp_write(tp, reg, data);
+	spin_unlock_irqrestore(&tp->mac_ocp_lock, flags);
+}
+
+static u16 __r8168_mac_ocp_read(struct rtl8169_private *tp, u32 reg)
 {
 	if (rtl_ocp_reg_failure(reg))
 		return 0;
@@ -865,12 +876,28 @@ static u16 r8168_mac_ocp_read(struct rtl8169_private *tp, u32 reg)
 	return RTL_R32(tp, OCPDR);
 }
 
+static u16 r8168_mac_ocp_read(struct rtl8169_private *tp, u32 reg)
+{
+	unsigned long flags;
+	u16 val;
+
+	spin_lock_irqsave(&tp->mac_ocp_lock, flags);
+	val = __r8168_mac_ocp_read(tp, reg);
+	spin_unlock_irqrestore(&tp->mac_ocp_lock, flags);
+
+	return val;
+}
+
 static void r8168_mac_ocp_modify(struct rtl8169_private *tp, u32 reg, u16 mask,
 				 u16 set)
 {
-	u16 data = r8168_mac_ocp_read(tp, reg);
+	unsigned long flags;
+	u16 data;
 
-	r8168_mac_ocp_write(tp, reg, (data & ~mask) | set);
+	spin_lock_irqsave(&tp->mac_ocp_lock, flags);
+	data = __r8168_mac_ocp_read(tp, reg);
+	__r8168_mac_ocp_write(tp, reg, (data & ~mask) | set);
+	spin_unlock_irqrestore(&tp->mac_ocp_lock, flags);
 }
 
 /* Work around a hw issue with RTL8168g PHY, the quirk disables
@@ -5176,6 +5203,8 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tp->eee_adv = -1;
 	tp->ocp_base = OCP_STD_PHY_BASE;
 
+	spin_lock_init(&tp->mac_ocp_lock);
+
 	dev->tstats = devm_netdev_alloc_pcpu_stats(&pdev->dev,
 						   struct pcpu_sw_netstats);
 	if (!dev->tstats)
-- 
cgit v1.2.3-70-g09d2


From 6bc6c4e6893ee79a9862c61d1635e7da6d5a3333 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:24:00 +0100
Subject: r8169: use spinlock to protect access to registers Config2 and
 Config5
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For disabling ASPM during NAPI poll we'll have to access both registers
in atomic context. Use a spinlock to protect access.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 47 ++++++++++++++++++++++++-------
 1 file changed, 37 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 259eac5b0616..e6f3f1947488 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -613,6 +613,7 @@ struct rtl8169_private {
 		struct work_struct work;
 	} wk;
 
+	spinlock_t config25_lock;
 	spinlock_t mac_ocp_lock;
 
 	unsigned supports_gmii:1;
@@ -677,6 +678,28 @@ static void rtl_pci_commit(struct rtl8169_private *tp)
 	RTL_R8(tp, ChipCmd);
 }
 
+static void rtl_mod_config2(struct rtl8169_private *tp, u8 clear, u8 set)
+{
+	unsigned long flags;
+	u8 val;
+
+	spin_lock_irqsave(&tp->config25_lock, flags);
+	val = RTL_R8(tp, Config2);
+	RTL_W8(tp, Config2, (val & ~clear) | set);
+	spin_unlock_irqrestore(&tp->config25_lock, flags);
+}
+
+static void rtl_mod_config5(struct rtl8169_private *tp, u8 clear, u8 set)
+{
+	unsigned long flags;
+	u8 val;
+
+	spin_lock_irqsave(&tp->config25_lock, flags);
+	val = RTL_R8(tp, Config5);
+	RTL_W8(tp, Config5, (val & ~clear) | set);
+	spin_unlock_irqrestore(&tp->config25_lock, flags);
+}
+
 static bool rtl_is_8125(struct rtl8169_private *tp)
 {
 	return tp->mac_version >= RTL_GIGA_MAC_VER_61;
@@ -1363,6 +1386,7 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
 		{ WAKE_MAGIC, Config3, MagicPacket }
 	};
 	unsigned int i, tmp = ARRAY_SIZE(cfg);
+	unsigned long flags;
 	u8 options;
 
 	rtl_unlock_config_regs(tp);
@@ -1381,12 +1405,14 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
 			r8168_mac_ocp_modify(tp, 0xc0b6, BIT(0), 0);
 	}
 
+	spin_lock_irqsave(&tp->config25_lock, flags);
 	for (i = 0; i < tmp; i++) {
 		options = RTL_R8(tp, cfg[i].reg) & ~cfg[i].mask;
 		if (wolopts & cfg[i].opt)
 			options |= cfg[i].mask;
 		RTL_W8(tp, cfg[i].reg, options);
 	}
+	spin_unlock_irqrestore(&tp->config25_lock, flags);
 
 	switch (tp->mac_version) {
 	case RTL_GIGA_MAC_VER_02 ... RTL_GIGA_MAC_VER_06:
@@ -1398,10 +1424,10 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
 	case RTL_GIGA_MAC_VER_34:
 	case RTL_GIGA_MAC_VER_37:
 	case RTL_GIGA_MAC_VER_39 ... RTL_GIGA_MAC_VER_63:
-		options = RTL_R8(tp, Config2) & ~PME_SIGNAL;
 		if (wolopts)
-			options |= PME_SIGNAL;
-		RTL_W8(tp, Config2, options);
+			rtl_mod_config2(tp, 0, PME_SIGNAL);
+		else
+			rtl_mod_config2(tp, PME_SIGNAL, 0);
 		break;
 	default:
 		break;
@@ -2704,8 +2730,8 @@ static void rtl_hw_aspm_clkreq_enable(struct rtl8169_private *tp, bool enable)
 {
 	/* Don't enable ASPM in the chip if OS can't control ASPM */
 	if (enable && tp->aspm_manageable) {
-		RTL_W8(tp, Config5, RTL_R8(tp, Config5) | ASPM_en);
-		RTL_W8(tp, Config2, RTL_R8(tp, Config2) | ClkReqEn);
+		rtl_mod_config5(tp, 0, ASPM_en);
+		rtl_mod_config2(tp, 0, ClkReqEn);
 
 		switch (tp->mac_version) {
 		case RTL_GIGA_MAC_VER_46 ... RTL_GIGA_MAC_VER_48:
@@ -2728,8 +2754,8 @@ static void rtl_hw_aspm_clkreq_enable(struct rtl8169_private *tp, bool enable)
 			break;
 		}
 
-		RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
-		RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+		rtl_mod_config2(tp, ClkReqEn, 0);
+		rtl_mod_config5(tp, ASPM_en, 0);
 	}
 
 	udelay(10);
@@ -2890,7 +2916,7 @@ static void rtl_hw_start_8168e_1(struct rtl8169_private *tp)
 	RTL_W32(tp, MISC, RTL_R32(tp, MISC) | TXPLA_RST);
 	RTL_W32(tp, MISC, RTL_R32(tp, MISC) & ~TXPLA_RST);
 
-	RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~Spi_en);
+	rtl_mod_config5(tp, Spi_en, 0);
 }
 
 static void rtl_hw_start_8168e_2(struct rtl8169_private *tp)
@@ -2923,7 +2949,7 @@ static void rtl_hw_start_8168e_2(struct rtl8169_private *tp)
 
 	RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) | PFM_EN);
 	RTL_W32(tp, MISC, RTL_R32(tp, MISC) | PWM_EN);
-	RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~Spi_en);
+	rtl_mod_config5(tp, Spi_en, 0);
 
 	rtl_hw_aspm_clkreq_enable(tp, true);
 }
@@ -2946,7 +2972,7 @@ static void rtl_hw_start_8168f(struct rtl8169_private *tp)
 	RTL_W8(tp, MCU, RTL_R8(tp, MCU) & ~NOW_IS_OOB);
 	RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) | PFM_EN);
 	RTL_W32(tp, MISC, RTL_R32(tp, MISC) | PWM_EN);
-	RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~Spi_en);
+	rtl_mod_config5(tp, Spi_en, 0);
 
 	rtl8168_config_eee_mac(tp);
 }
@@ -5203,6 +5229,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tp->eee_adv = -1;
 	tp->ocp_base = OCP_STD_PHY_BASE;
 
+	spin_lock_init(&tp->config25_lock);
 	spin_lock_init(&tp->mac_ocp_lock);
 
 	dev->tstats = devm_netdev_alloc_pcpu_stats(&pdev->dev,
-- 
cgit v1.2.3-70-g09d2


From 59ee97c0c1a8d0dadc092897ca7d5fe3a80e1bc3 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:24:49 +0100
Subject: r8169: enable cfg9346 config register access in atomic context
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For disabling ASPM during NAPI poll we'll have to unlock access
to the config registers in atomic context. Other code parts
running with config register access unlocked are partially
longer and can sleep. Add a usage counter to enable parallel
execution of code parts requiring unlocked config registers.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index e6f3f1947488..61cbf498fd07 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -616,6 +616,9 @@ struct rtl8169_private {
 	spinlock_t config25_lock;
 	spinlock_t mac_ocp_lock;
 
+	spinlock_t cfg9346_usage_lock;
+	int cfg9346_usage_count;
+
 	unsigned supports_gmii:1;
 	unsigned aspm_manageable:1;
 	dma_addr_t counters_phys_addr;
@@ -664,12 +667,22 @@ static inline struct device *tp_to_dev(struct rtl8169_private *tp)
 
 static void rtl_lock_config_regs(struct rtl8169_private *tp)
 {
-	RTL_W8(tp, Cfg9346, Cfg9346_Lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&tp->cfg9346_usage_lock, flags);
+	if (!--tp->cfg9346_usage_count)
+		RTL_W8(tp, Cfg9346, Cfg9346_Lock);
+	spin_unlock_irqrestore(&tp->cfg9346_usage_lock, flags);
 }
 
 static void rtl_unlock_config_regs(struct rtl8169_private *tp)
 {
-	RTL_W8(tp, Cfg9346, Cfg9346_Unlock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&tp->cfg9346_usage_lock, flags);
+	if (!tp->cfg9346_usage_count++)
+		RTL_W8(tp, Cfg9346, Cfg9346_Unlock);
+	spin_unlock_irqrestore(&tp->cfg9346_usage_lock, flags);
 }
 
 static void rtl_pci_commit(struct rtl8169_private *tp)
@@ -5229,6 +5242,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tp->eee_adv = -1;
 	tp->ocp_base = OCP_STD_PHY_BASE;
 
+	spin_lock_init(&tp->cfg9346_usage_lock);
 	spin_lock_init(&tp->config25_lock);
 	spin_lock_init(&tp->mac_ocp_lock);
 
-- 
cgit v1.2.3-70-g09d2


From 49ef7d846d4bd77b0b9f1f801fc765b004690a07 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:25:49 +0100
Subject: r8169: prepare rtl_hw_aspm_clkreq_enable for usage in atomic context
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Bail out if the function is used with chip versions that don't support
ASPM configuration. In addition remove the delay, it tuned out that
it's not needed, also vendor driver r8125 doesn't have it.

Suggested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 61cbf498fd07..96af31aeab0a 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2741,6 +2741,9 @@ static void rtl_disable_exit_l1(struct rtl8169_private *tp)
 
 static void rtl_hw_aspm_clkreq_enable(struct rtl8169_private *tp, bool enable)
 {
+	if (tp->mac_version < RTL_GIGA_MAC_VER_32)
+		return;
+
 	/* Don't enable ASPM in the chip if OS can't control ASPM */
 	if (enable && tp->aspm_manageable) {
 		rtl_mod_config5(tp, 0, ASPM_en);
@@ -2770,8 +2773,6 @@ static void rtl_hw_aspm_clkreq_enable(struct rtl8169_private *tp, bool enable)
 		rtl_mod_config2(tp, ClkReqEn, 0);
 		rtl_mod_config5(tp, ASPM_en, 0);
 	}
-
-	udelay(10);
 }
 
 static void rtl_set_fifo_size(struct rtl8169_private *tp, u16 rx_stat,
-- 
cgit v1.2.3-70-g09d2


From e1ed3e4d91112027b90c7ee61479141b3f948e6a Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:26:47 +0100
Subject: r8169: disable ASPM during NAPI poll
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Several chip versions have problems with ASPM, what may result in
rx_missed errors or tx timeouts. The root cause isn't known but
experience shows that disabling ASPM during NAPI poll can avoid
these problems.

Suggested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 96af31aeab0a..2897b9bf29af 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -4577,6 +4577,10 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
 	}
 
 	if (napi_schedule_prep(&tp->napi)) {
+		rtl_unlock_config_regs(tp);
+		rtl_hw_aspm_clkreq_enable(tp, false);
+		rtl_lock_config_regs(tp);
+
 		rtl_irq_disable(tp);
 		__napi_schedule(&tp->napi);
 	}
@@ -4636,9 +4640,14 @@ static int rtl8169_poll(struct napi_struct *napi, int budget)
 
 	work_done = rtl_rx(dev, tp, budget);
 
-	if (work_done < budget && napi_complete_done(napi, work_done))
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
 		rtl_irq_enable(tp);
 
+		rtl_unlock_config_regs(tp);
+		rtl_hw_aspm_clkreq_enable(tp, true);
+		rtl_lock_config_regs(tp);
+	}
+
 	return work_done;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 2ab19de62d67e403105ba860971e5ff0d511ad15 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 6 Mar 2023 22:28:06 +0100
Subject: r8169: remove ASPM restrictions now that ASPM is disabled during NAPI
 poll
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that  ASPM is disabled during NAPI poll, we can remove all ASPM
restrictions. This allows for higher power savings if the network
isn't fully loaded.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 27 +--------------------------
 1 file changed, 1 insertion(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 2897b9bf29af..6563e4c6a136 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -620,7 +620,6 @@ struct rtl8169_private {
 	int cfg9346_usage_count;
 
 	unsigned supports_gmii:1;
-	unsigned aspm_manageable:1;
 	dma_addr_t counters_phys_addr;
 	struct rtl8169_counters *counters;
 	struct rtl8169_tc_offsets tc_offset;
@@ -2744,8 +2743,7 @@ static void rtl_hw_aspm_clkreq_enable(struct rtl8169_private *tp, bool enable)
 	if (tp->mac_version < RTL_GIGA_MAC_VER_32)
 		return;
 
-	/* Don't enable ASPM in the chip if OS can't control ASPM */
-	if (enable && tp->aspm_manageable) {
+	if (enable) {
 		rtl_mod_config5(tp, 0, ASPM_en);
 		rtl_mod_config2(tp, 0, ClkReqEn);
 
@@ -5221,16 +5219,6 @@ done:
 	rtl_rar_set(tp, mac_addr);
 }
 
-/* register is set if system vendor successfully tested ASPM 1.2 */
-static bool rtl_aspm_is_safe(struct rtl8169_private *tp)
-{
-	if (tp->mac_version >= RTL_GIGA_MAC_VER_61 &&
-	    r8168_mac_ocp_read(tp, 0xc0b2) & 0xf)
-		return true;
-
-	return false;
-}
-
 static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct rtl8169_private *tp;
@@ -5302,19 +5290,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	tp->mac_version = chipset;
 
-	/* Disable ASPM L1 as that cause random device stop working
-	 * problems as well as full system hangs for some PCIe devices users.
-	 * Chips from RTL8168h partially have issues with L1.2, but seem
-	 * to work fine with L1 and L1.1.
-	 */
-	if (rtl_aspm_is_safe(tp))
-		rc = 0;
-	else if (tp->mac_version >= RTL_GIGA_MAC_VER_46)
-		rc = pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2);
-	else
-		rc = pci_disable_link_state(pdev, PCIE_LINK_STATE_L1);
-	tp->aspm_manageable = !rc;
-
 	tp->dash_type = rtl_check_dash(tp);
 
 	tp->cp_cmd = RTL_R16(tp, CPlusCmd) & CPCMD_MASK;
-- 
cgit v1.2.3-70-g09d2


From 966b6b809f900b6c41f3b559fd57ec567d735dc9 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Mon, 6 Mar 2023 17:51:52 -0600
Subject: net/mlx4_en: Replace fake flex-array with flexible-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Transform zero-length array into flexible-array member in struct
mlx4_en_rx_desc.

Address the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
drivers/net/ethernet/mellanox/mlx4/en_rx.c:88:30: warning: array subscript i is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]
drivers/net/ethernet/mellanox/mlx4/en_rx.c:149:30: warning: array subscript 0 is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]
drivers/net/ethernet/mellanox/mlx4/en_rx.c:127:30: warning: array subscript i is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]
drivers/net/ethernet/mellanox/mlx4/en_rx.c:128:30: warning: array subscript i is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]
drivers/net/ethernet/mellanox/mlx4/en_rx.c:129:30: warning: array subscript i is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]
drivers/net/ethernet/mellanox/mlx4/en_rx.c:117:30: warning: array subscript i is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]
drivers/net/ethernet/mellanox/mlx4/en_rx.c:119:30: warning: array subscript i is outside array bounds of ‘struct mlx4_wqe_data_seg[0]’ [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/264
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 544e09b97483..034733b13b1a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -323,7 +323,7 @@ struct mlx4_en_tx_ring {
 
 struct mlx4_en_rx_desc {
 	/* actual number of entries depends on rx ring stride */
-	struct mlx4_wqe_data_seg data[0];
+	DECLARE_FLEX_ARRAY(struct mlx4_wqe_data_seg, data);
 };
 
 struct mlx4_en_rx_ring {
-- 
cgit v1.2.3-70-g09d2


From ecf729f93bd5a2117ef82d1e281bfc780c643e36 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Tue, 7 Mar 2023 13:41:38 +0800
Subject: emulex/benet: clean up some inconsistent indenting

No functional modification involved.

drivers/net/ethernet/emulex/benet/be_cmds.c:1120 be_cmd_pmac_add() warn: inconsistent indenting.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4396
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/emulex/benet/be_cmds.c | 27 ++++++++++++++-------------
 1 file changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 08ec84cd21c0..61adcebeef01 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -135,7 +135,8 @@ static int be_mcc_notify(struct be_adapter *adapter)
 
 /* To check if valid bit is set, check the entire word as we don't know
  * the endianness of the data (old entry is host endian while a new entry is
- * little endian) */
+ * little endian)
+ */
 static inline bool be_mcc_compl_is_new(struct be_mcc_compl *compl)
 {
 	u32 flags;
@@ -248,7 +249,8 @@ static int be_mcc_compl_process(struct be_adapter *adapter,
 	u8 opcode = 0, subsystem = 0;
 
 	/* Just swap the status to host endian; mcc tag is opaquely copied
-	 * from mcc_wrb */
+	 * from mcc_wrb
+	 */
 	be_dws_le_to_cpu(compl, 4);
 
 	base_status = base_status(compl->status);
@@ -657,8 +659,7 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db)
 	return 0;
 }
 
-/*
- * Insert the mailbox address into the doorbell in two steps
+/* Insert the mailbox address into the doorbell in two steps
  * Polls on the mbox doorbell till a command completion (or a timeout) occurs
  */
 static int be_mbox_notify_wait(struct be_adapter *adapter)
@@ -802,7 +803,7 @@ static void be_wrb_cmd_hdr_prepare(struct be_cmd_req_hdr *req_hdr,
 	req_hdr->subsystem = subsystem;
 	req_hdr->request_length = cpu_to_le32(cmd_len - sizeof(*req_hdr));
 	req_hdr->version = 0;
-	fill_wrb_tags(wrb, (ulong) req_hdr);
+	fill_wrb_tags(wrb, (ulong)req_hdr);
 	wrb->payload_length = cmd_len;
 	if (mem) {
 		wrb->embedded |= (1 & MCC_WRB_SGE_CNT_MASK) <<
@@ -832,8 +833,8 @@ static void be_cmd_page_addrs_prepare(struct phys_addr *pages, u32 max_pages,
 static inline struct be_mcc_wrb *wrb_from_mbox(struct be_adapter *adapter)
 {
 	struct be_dma_mem *mbox_mem = &adapter->mbox_mem;
-	struct be_mcc_wrb *wrb
-		= &((struct be_mcc_mailbox *)(mbox_mem->va))->wrb;
+	struct be_mcc_wrb *wrb = &((struct be_mcc_mailbox *)(mbox_mem->va))->wrb;
+
 	memset(wrb, 0, sizeof(*wrb));
 	return wrb;
 }
@@ -896,7 +897,7 @@ static struct be_mcc_wrb *be_cmd_copy(struct be_adapter *adapter,
 
 	memcpy(dest_wrb, wrb, sizeof(*wrb));
 	if (wrb->embedded & cpu_to_le32(MCC_WRB_EMBEDDED_MASK))
-		fill_wrb_tags(dest_wrb, (ulong) embedded_payload(wrb));
+		fill_wrb_tags(dest_wrb, (ulong)embedded_payload(wrb));
 
 	return dest_wrb;
 }
@@ -1114,7 +1115,7 @@ int be_cmd_pmac_add(struct be_adapter *adapter, const u8 *mac_addr,
 err:
 	mutex_unlock(&adapter->mcc_lock);
 
-	 if (base_status(status) == MCC_STATUS_UNAUTHORIZED_REQUEST)
+	if (base_status(status) == MCC_STATUS_UNAUTHORIZED_REQUEST)
 		status = -EPERM;
 
 	return status;
@@ -1803,7 +1804,7 @@ int be_cmd_get_fat_dump(struct be_adapter *adapter, u32 buf_len, void *buf)
 
 	total_size = buf_len;
 
-	get_fat_cmd.size = sizeof(struct be_cmd_req_get_fat) + 60*1024;
+	get_fat_cmd.size = sizeof(struct be_cmd_req_get_fat) + 60 * 1024;
 	get_fat_cmd.va = dma_alloc_coherent(&adapter->pdev->dev,
 					    get_fat_cmd.size,
 					    &get_fat_cmd.dma, GFP_ATOMIC);
@@ -1813,7 +1814,7 @@ int be_cmd_get_fat_dump(struct be_adapter *adapter, u32 buf_len, void *buf)
 	mutex_lock(&adapter->mcc_lock);
 
 	while (total_size) {
-		buf_size = min(total_size, (u32)60*1024);
+		buf_size = min(total_size, (u32)60 * 1024);
 		total_size -= buf_size;
 
 		wrb = wrb_from_mccq(adapter);
@@ -3362,7 +3363,7 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
 	req->pattern = cpu_to_le64(pattern);
 	req->byte_count = cpu_to_le32(byte_cnt);
 	for (i = 0; i < byte_cnt; i++) {
-		req->snd_buff[i] = (u8)(pattern >> (j*8));
+		req->snd_buff[i] = (u8)(pattern >> (j * 8));
 		j++;
 		if (j > 7)
 			j = 0;
@@ -3846,7 +3847,7 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array,
 	req->hdr.domain = domain;
 	req->mac_count = mac_count;
 	if (mac_count)
-		memcpy(req->mac, mac_array, ETH_ALEN*mac_count);
+		memcpy(req->mac, mac_array, ETH_ALEN * mac_count);
 
 	status = be_mcc_notify_wait(adapter);
 
-- 
cgit v1.2.3-70-g09d2


From 7d8c48917a9576b5fc8871aa4946149b0e4a4927 Mon Sep 17 00:00:00 2001
From: Arınç ÜNAL <arinc.unal@arinc9.com>
Date: Tue, 7 Mar 2023 12:56:19 +0300
Subject: dt-bindings: net: dsa: mediatek,mt7530: change some descriptions to
 literal
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The line endings must be preserved on gpio-controller, io-supply, and
reset-gpios properties to look proper when the YAML file is parsed.

Currently it's interpreted as a single line when parsed. Change the style
of the description of these properties to literal style to preserve the
line endings.

Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml b/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
index 449ee0735012..5ae9cd8f99a2 100644
--- a/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
@@ -93,7 +93,7 @@ properties:
 
   gpio-controller:
     type: boolean
-    description:
+    description: |
       If defined, LED controller of the MT7530 switch will run on GPIO mode.
 
       There are 15 controllable pins.
@@ -112,7 +112,7 @@ properties:
     maxItems: 1
 
   io-supply:
-    description:
+    description: |
       Phandle to the regulator node necessary for the I/O power.
       See Documentation/devicetree/bindings/regulator/mt6323-regulator.txt for
       details for the regulator setup on these boards.
@@ -124,7 +124,7 @@ properties:
       switch is a part of the multi-chip module.
 
   reset-gpios:
-    description:
+    description: |
       GPIO to reset the switch. Use this if mediatek,mcm is not used.
       This property is optional because some boards share the reset line with
       other components which makes it impossible to probe the switch if the
-- 
cgit v1.2.3-70-g09d2


From fbd3dce9580882580b7d4779569062e371e8be48 Mon Sep 17 00:00:00 2001
From: Steen Hegelund <steen.hegelund@microchip.com>
Date: Tue, 7 Mar 2023 14:40:59 +0100
Subject: net: microchip: sparx5: Correct the spelling of the keysets in
 debugfs

Correct the name used in the debugfs output.

Signed-off-by: Steen Hegelund <steen.hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microchip/sparx5/sparx5_vcap_debugfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_debugfs.c b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_debugfs.c
index 07b472c84a47..12722f728ef7 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_debugfs.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_debugfs.c
@@ -198,7 +198,7 @@ static void sparx5_vcap_is2_port_keys(struct sparx5 *sparx5,
 			out->prf(out->dst, "ip6_std");
 			break;
 		case VCAP_IS2_PS_IPV6_MC_IP4_TCP_UDP_OTHER:
-			out->prf(out->dst, "ip4_tcp_udp ipv4_other");
+			out->prf(out->dst, "ip4_tcp_udp ip4_other");
 			break;
 		}
 		out->prf(out->dst, "\n      ipv6_uc: ");
-- 
cgit v1.2.3-70-g09d2


From bfcb94aacc92f7025bd86a545037e5eb8db00e32 Mon Sep 17 00:00:00 2001
From: Steen Hegelund <steen.hegelund@microchip.com>
Date: Tue, 7 Mar 2023 14:41:00 +0100
Subject: net: microchip: sparx5: Provide rule count, key removal and keyset
 select

This provides these 3 functions in the VCAP API:

- Count the number of rules in a VCAP lookup (chain)
- Remove a key from a VCAP rule
- Find the keyset that gives the smallest rule list from a list of keysets

Signed-off-by: Steen Hegelund <steen.hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microchip/vcap/vcap_api.c     | 61 ++++++++++++++++++++++
 .../net/ethernet/microchip/vcap/vcap_api_client.h  | 11 ++++
 2 files changed, 72 insertions(+)

diff --git a/drivers/net/ethernet/microchip/vcap/vcap_api.c b/drivers/net/ethernet/microchip/vcap/vcap_api.c
index 4847d0d99ec9..5675b0962bc3 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_api.c
+++ b/drivers/net/ethernet/microchip/vcap/vcap_api.c
@@ -976,6 +976,25 @@ int vcap_lookup_rule_by_cookie(struct vcap_control *vctrl, u64 cookie)
 }
 EXPORT_SYMBOL_GPL(vcap_lookup_rule_by_cookie);
 
+/* Get number of rules in a vcap instance lookup chain id range */
+int vcap_admin_rule_count(struct vcap_admin *admin, int cid)
+{
+	int max_cid = roundup(cid + 1, VCAP_CID_LOOKUP_SIZE);
+	int min_cid = rounddown(cid, VCAP_CID_LOOKUP_SIZE);
+	struct vcap_rule_internal *elem;
+	int count = 0;
+
+	list_for_each_entry(elem, &admin->rules, list) {
+		mutex_lock(&admin->lock);
+		if (elem->data.vcap_chain_id >= min_cid &&
+		    elem->data.vcap_chain_id < max_cid)
+			++count;
+		mutex_unlock(&admin->lock);
+	}
+	return count;
+}
+EXPORT_SYMBOL_GPL(vcap_admin_rule_count);
+
 /* Make a copy of the rule, shallow or full */
 static struct vcap_rule_internal *vcap_dup_rule(struct vcap_rule_internal *ri,
 						bool full)
@@ -3403,6 +3422,25 @@ int vcap_rule_mod_key_u32(struct vcap_rule *rule, enum vcap_key_field key,
 }
 EXPORT_SYMBOL_GPL(vcap_rule_mod_key_u32);
 
+/* Remove a key field with value and mask in the rule */
+int vcap_rule_rem_key(struct vcap_rule *rule, enum vcap_key_field key)
+{
+	struct vcap_rule_internal *ri = to_intrule(rule);
+	struct vcap_client_keyfield *field;
+
+	field = vcap_find_keyfield(rule, key);
+	if (!field) {
+		pr_err("%s:%d: key %s is not in the rule\n",
+		       __func__, __LINE__, vcap_keyfield_name(ri->vctrl, key));
+		return -EINVAL;
+	}
+	/* Deallocate the key field */
+	list_del(&field->ctrl.list);
+	kfree(field);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(vcap_rule_rem_key);
+
 static int vcap_rule_mod_action(struct vcap_rule *rule,
 				enum vcap_action_field action,
 				enum vcap_field_type ftype,
@@ -3475,6 +3513,29 @@ int vcap_filter_rule_keys(struct vcap_rule *rule,
 }
 EXPORT_SYMBOL_GPL(vcap_filter_rule_keys);
 
+/* Select the keyset from the list that results in the smallest rule size */
+enum vcap_keyfield_set
+vcap_select_min_rule_keyset(struct vcap_control *vctrl,
+			    enum vcap_type vtype,
+			    struct vcap_keyset_list *kslist)
+{
+	enum vcap_keyfield_set ret = VCAP_KFS_NO_VALUE;
+	const struct vcap_set *kset;
+	int max = 100, idx;
+
+	for (idx = 0; idx < kslist->cnt; ++idx) {
+		kset = vcap_keyfieldset(vctrl, vtype, kslist->keysets[idx]);
+		if (!kset)
+			continue;
+		if (kset->sw_per_item >= max)
+			continue;
+		max = kset->sw_per_item;
+		ret = kslist->keysets[idx];
+	}
+	return ret;
+}
+EXPORT_SYMBOL_GPL(vcap_select_min_rule_keyset);
+
 /* Make a full copy of an existing rule with a new rule id */
 struct vcap_rule *vcap_copy_rule(struct vcap_rule *erule)
 {
diff --git a/drivers/net/ethernet/microchip/vcap/vcap_api_client.h b/drivers/net/ethernet/microchip/vcap/vcap_api_client.h
index 417af9754bcc..d9d1f7c9d762 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_api_client.h
+++ b/drivers/net/ethernet/microchip/vcap/vcap_api_client.h
@@ -201,6 +201,9 @@ int vcap_rule_add_action_bit(struct vcap_rule *rule,
 int vcap_rule_add_action_u32(struct vcap_rule *rule,
 			     enum vcap_action_field action, u32 value);
 
+/* Get number of rules in a vcap instance lookup chain id range */
+int vcap_admin_rule_count(struct vcap_admin *admin, int cid);
+
 /* VCAP rule counter operations */
 int vcap_get_rule_count_by_cookie(struct vcap_control *vctrl,
 				  struct vcap_counter *ctr, u64 cookie);
@@ -269,6 +272,14 @@ int vcap_rule_mod_action_u32(struct vcap_rule *rule,
 int vcap_rule_get_key_u32(struct vcap_rule *rule, enum vcap_key_field key,
 			  u32 *value, u32 *mask);
 
+/* Remove a key field with value and mask in the rule */
+int vcap_rule_rem_key(struct vcap_rule *rule, enum vcap_key_field key);
+
+/* Select the keyset from the list that results in the smallest rule size */
+enum vcap_keyfield_set
+vcap_select_min_rule_keyset(struct vcap_control *vctrl, enum vcap_type vtype,
+			    struct vcap_keyset_list *kslist);
+
 struct vcap_client_actionfield *
 vcap_find_actionfield(struct vcap_rule *rule, enum vcap_action_field act);
 #endif /* __VCAP_API_CLIENT__ */
-- 
cgit v1.2.3-70-g09d2


From 1c14432dcefaf44c2e3fa11ac61097ee342eecfe Mon Sep 17 00:00:00 2001
From: Steen Hegelund <steen.hegelund@microchip.com>
Date: Tue, 7 Mar 2023 14:41:01 +0100
Subject: net: microchip: sparx5: Add TC template list to a port

This adds a list that is used to collect the templates that are active on a
port.

This allows the template creation to change the port configuration
and the template destruction to change it back.

Signed-off-by: Steen Hegelund <steen.hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microchip/sparx5/sparx5_main.c | 1 +
 drivers/net/ethernet/microchip/sparx5/sparx5_main.h | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_main.c b/drivers/net/ethernet/microchip/sparx5/sparx5_main.c
index 42b77ba9b572..a7edf524eedb 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_main.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_main.c
@@ -282,6 +282,7 @@ static int sparx5_create_port(struct sparx5 *sparx5,
 	spx5_port->phylink_pcs.poll = true;
 	spx5_port->phylink_pcs.ops = &sparx5_phylink_pcs_ops;
 	spx5_port->is_mrouter = false;
+	INIT_LIST_HEAD(&spx5_port->tc_templates);
 	sparx5->ports[config->portno] = spx5_port;
 
 	err = sparx5_port_init(sparx5, spx5_port, &config->conf);
diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_main.h b/drivers/net/ethernet/microchip/sparx5/sparx5_main.h
index 72e7928912eb..62c85463b634 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_main.h
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_main.h
@@ -192,6 +192,7 @@ struct sparx5_port {
 	u16 ts_id;
 	struct sk_buff_head tx_skbs;
 	bool is_mrouter;
+	struct list_head tc_templates; /* list of TC templates on this port */
 };
 
 enum sparx5_core_clockfreq {
-- 
cgit v1.2.3-70-g09d2


From d9f175b0dffbf70add634319a81d683be782c2e1 Mon Sep 17 00:00:00 2001
From: Steen Hegelund <steen.hegelund@microchip.com>
Date: Tue, 7 Mar 2023 14:41:02 +0100
Subject: net: microchip: sparx5: Add port keyset changing functionality

With this its is now possible for clients (like TC) to change the port
keyset configuration in the Sparx5 VCAPs.

This is typically done per traffic class which is guided with the L3
protocol information.
Before the change the current keyset configuration is collected in a list
that is handed back to the client.

Signed-off-by: Steen Hegelund <steen.hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/microchip/sparx5/sparx5_vcap_impl.c   | 270 +++++++++++++++++++++
 .../ethernet/microchip/sparx5/sparx5_vcap_impl.h   |   6 +
 2 files changed, 276 insertions(+)

diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c
index d0d4e0385ac7..187efa1fc904 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c
@@ -1519,6 +1519,276 @@ static struct vcap_operations sparx5_vcap_ops = {
 	.port_info = sparx5_port_info,
 };
 
+static u32 sparx5_vcap_is0_keyset_to_etype_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_NORMAL_7TUPLE:
+		return VCAP_IS0_PS_ETYPE_NORMAL_7TUPLE;
+	case VCAP_KFS_NORMAL_5TUPLE_IP4:
+		return VCAP_IS0_PS_ETYPE_NORMAL_5TUPLE_IP4;
+	default:
+		return VCAP_IS0_PS_ETYPE_NORMAL_7TUPLE;
+	}
+}
+
+static void sparx5_vcap_is0_set_port_keyset(struct net_device *ndev, int lookup,
+					    enum vcap_keyfield_set keyset,
+					    int l3_proto)
+{
+	struct sparx5_port *port = netdev_priv(ndev);
+	struct sparx5 *sparx5 = port->sparx5;
+	int portno = port->portno;
+	u32 value;
+
+	switch (l3_proto) {
+	case ETH_P_IP:
+		value = sparx5_vcap_is0_keyset_to_etype_ps(keyset);
+		spx5_rmw(ANA_CL_ADV_CL_CFG_IP4_CLM_KEY_SEL_SET(value),
+			 ANA_CL_ADV_CL_CFG_IP4_CLM_KEY_SEL,
+			 sparx5,
+			 ANA_CL_ADV_CL_CFG(portno, lookup));
+		break;
+	case ETH_P_IPV6:
+		value = sparx5_vcap_is0_keyset_to_etype_ps(keyset);
+		spx5_rmw(ANA_CL_ADV_CL_CFG_IP6_CLM_KEY_SEL_SET(value),
+			 ANA_CL_ADV_CL_CFG_IP6_CLM_KEY_SEL,
+			 sparx5,
+			 ANA_CL_ADV_CL_CFG(portno, lookup));
+		break;
+	default:
+		value = sparx5_vcap_is0_keyset_to_etype_ps(keyset);
+		spx5_rmw(ANA_CL_ADV_CL_CFG_ETYPE_CLM_KEY_SEL_SET(value),
+			 ANA_CL_ADV_CL_CFG_ETYPE_CLM_KEY_SEL,
+			 sparx5,
+			 ANA_CL_ADV_CL_CFG(portno, lookup));
+		break;
+	}
+}
+
+static u32 sparx5_vcap_is2_keyset_to_arp_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_ARP:
+		return VCAP_IS2_PS_ARP_ARP;
+	default:
+		return VCAP_IS2_PS_ARP_MAC_ETYPE;
+	}
+}
+
+static u32 sparx5_vcap_is2_keyset_to_ipv4_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_MAC_ETYPE:
+		return VCAP_IS2_PS_IPV4_UC_MAC_ETYPE;
+	case VCAP_KFS_IP4_OTHER:
+	case VCAP_KFS_IP4_TCP_UDP:
+		return VCAP_IS2_PS_IPV4_UC_IP4_TCP_UDP_OTHER;
+	case VCAP_KFS_IP_7TUPLE:
+		return VCAP_IS2_PS_IPV4_UC_IP_7TUPLE;
+	default:
+		return VCAP_KFS_NO_VALUE;
+	}
+}
+
+static u32 sparx5_vcap_is2_keyset_to_ipv6_uc_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_MAC_ETYPE:
+		return VCAP_IS2_PS_IPV6_UC_MAC_ETYPE;
+	case VCAP_KFS_IP4_OTHER:
+	case VCAP_KFS_IP4_TCP_UDP:
+		return VCAP_IS2_PS_IPV6_UC_IP4_TCP_UDP_OTHER;
+	case VCAP_KFS_IP_7TUPLE:
+		return VCAP_IS2_PS_IPV6_UC_IP_7TUPLE;
+	default:
+		return VCAP_KFS_NO_VALUE;
+	}
+}
+
+static u32 sparx5_vcap_is2_keyset_to_ipv6_mc_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_MAC_ETYPE:
+		return VCAP_IS2_PS_IPV6_MC_MAC_ETYPE;
+	case VCAP_KFS_IP4_OTHER:
+	case VCAP_KFS_IP4_TCP_UDP:
+		return VCAP_IS2_PS_IPV6_MC_IP4_TCP_UDP_OTHER;
+	case VCAP_KFS_IP_7TUPLE:
+		return VCAP_IS2_PS_IPV6_MC_IP_7TUPLE;
+	default:
+		return VCAP_KFS_NO_VALUE;
+	}
+}
+
+static void sparx5_vcap_is2_set_port_keyset(struct net_device *ndev, int lookup,
+					    enum vcap_keyfield_set keyset,
+					    int l3_proto)
+{
+	struct sparx5_port *port = netdev_priv(ndev);
+	struct sparx5 *sparx5 = port->sparx5;
+	int portno = port->portno;
+	u32 value;
+
+	switch (l3_proto) {
+	case ETH_P_ARP:
+		value = sparx5_vcap_is2_keyset_to_arp_ps(keyset);
+		spx5_rmw(ANA_ACL_VCAP_S2_KEY_SEL_ARP_KEY_SEL_SET(value),
+			 ANA_ACL_VCAP_S2_KEY_SEL_ARP_KEY_SEL,
+			 sparx5,
+			 ANA_ACL_VCAP_S2_KEY_SEL(portno, lookup));
+		break;
+	case ETH_P_IP:
+		value = sparx5_vcap_is2_keyset_to_ipv4_ps(keyset);
+		spx5_rmw(ANA_ACL_VCAP_S2_KEY_SEL_IP4_UC_KEY_SEL_SET(value),
+			 ANA_ACL_VCAP_S2_KEY_SEL_IP4_UC_KEY_SEL,
+			 sparx5,
+			 ANA_ACL_VCAP_S2_KEY_SEL(portno, lookup));
+		spx5_rmw(ANA_ACL_VCAP_S2_KEY_SEL_IP4_MC_KEY_SEL_SET(value),
+			 ANA_ACL_VCAP_S2_KEY_SEL_IP4_MC_KEY_SEL,
+			 sparx5,
+			 ANA_ACL_VCAP_S2_KEY_SEL(portno, lookup));
+		break;
+	case ETH_P_IPV6:
+		value = sparx5_vcap_is2_keyset_to_ipv6_uc_ps(keyset);
+		spx5_rmw(ANA_ACL_VCAP_S2_KEY_SEL_IP6_UC_KEY_SEL_SET(value),
+			 ANA_ACL_VCAP_S2_KEY_SEL_IP6_UC_KEY_SEL,
+			 sparx5,
+			 ANA_ACL_VCAP_S2_KEY_SEL(portno, lookup));
+		value = sparx5_vcap_is2_keyset_to_ipv6_mc_ps(keyset);
+		spx5_rmw(ANA_ACL_VCAP_S2_KEY_SEL_IP6_MC_KEY_SEL_SET(value),
+			 ANA_ACL_VCAP_S2_KEY_SEL_IP6_MC_KEY_SEL,
+			 sparx5,
+			 ANA_ACL_VCAP_S2_KEY_SEL(portno, lookup));
+		break;
+	default:
+		value = VCAP_IS2_PS_NONETH_MAC_ETYPE;
+		spx5_rmw(ANA_ACL_VCAP_S2_KEY_SEL_NON_ETH_KEY_SEL_SET(value),
+			 ANA_ACL_VCAP_S2_KEY_SEL_NON_ETH_KEY_SEL,
+			 sparx5,
+			 ANA_ACL_VCAP_S2_KEY_SEL(portno, lookup));
+		break;
+	}
+}
+
+static u32 sparx5_vcap_es2_keyset_to_arp_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_ARP:
+		return VCAP_ES2_PS_ARP_ARP;
+	default:
+		return VCAP_ES2_PS_ARP_MAC_ETYPE;
+	}
+}
+
+static u32 sparx5_vcap_es2_keyset_to_ipv4_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_MAC_ETYPE:
+		return VCAP_ES2_PS_IPV4_MAC_ETYPE;
+	case VCAP_KFS_IP_7TUPLE:
+		return VCAP_ES2_PS_IPV4_IP_7TUPLE;
+	case VCAP_KFS_IP4_TCP_UDP:
+		return VCAP_ES2_PS_IPV4_IP4_TCP_UDP_OTHER;
+	case VCAP_KFS_IP4_OTHER:
+		return VCAP_ES2_PS_IPV4_IP4_OTHER;
+	default:
+		return VCAP_ES2_PS_IPV4_MAC_ETYPE;
+	}
+}
+
+static u32 sparx5_vcap_es2_keyset_to_ipv6_ps(enum vcap_keyfield_set keyset)
+{
+	switch (keyset) {
+	case VCAP_KFS_MAC_ETYPE:
+		return VCAP_ES2_PS_IPV6_MAC_ETYPE;
+	case VCAP_KFS_IP4_TCP_UDP:
+	case VCAP_KFS_IP4_OTHER:
+		return VCAP_ES2_PS_IPV6_IP4_DOWNGRADE;
+	case VCAP_KFS_IP_7TUPLE:
+		return VCAP_ES2_PS_IPV6_IP_7TUPLE;
+	case VCAP_KFS_IP6_STD:
+		return VCAP_ES2_PS_IPV6_IP6_STD;
+	default:
+		return VCAP_ES2_PS_IPV6_MAC_ETYPE;
+	}
+}
+
+static void sparx5_vcap_es2_set_port_keyset(struct net_device *ndev, int lookup,
+					    enum vcap_keyfield_set keyset,
+					    int l3_proto)
+{
+	struct sparx5_port *port = netdev_priv(ndev);
+	struct sparx5 *sparx5 = port->sparx5;
+	int portno = port->portno;
+	u32 value;
+
+	switch (l3_proto) {
+	case ETH_P_IP:
+		value = sparx5_vcap_es2_keyset_to_ipv4_ps(keyset);
+		spx5_rmw(EACL_VCAP_ES2_KEY_SEL_IP4_KEY_SEL_SET(value),
+			 EACL_VCAP_ES2_KEY_SEL_IP4_KEY_SEL,
+			 sparx5,
+			 EACL_VCAP_ES2_KEY_SEL(portno, lookup));
+		break;
+	case ETH_P_IPV6:
+		value = sparx5_vcap_es2_keyset_to_ipv6_ps(keyset);
+		spx5_rmw(EACL_VCAP_ES2_KEY_SEL_IP6_KEY_SEL_SET(value),
+			 EACL_VCAP_ES2_KEY_SEL_IP6_KEY_SEL,
+			 sparx5,
+			 EACL_VCAP_ES2_KEY_SEL(portno, lookup));
+		break;
+	case ETH_P_ARP:
+		value = sparx5_vcap_es2_keyset_to_arp_ps(keyset);
+		spx5_rmw(EACL_VCAP_ES2_KEY_SEL_ARP_KEY_SEL_SET(value),
+			 EACL_VCAP_ES2_KEY_SEL_ARP_KEY_SEL,
+			 sparx5,
+			 EACL_VCAP_ES2_KEY_SEL(portno, lookup));
+		break;
+	}
+}
+
+/* Change the port keyset for the lookup and protocol */
+void sparx5_vcap_set_port_keyset(struct net_device *ndev,
+				 struct vcap_admin *admin,
+				 int cid,
+				 u16 l3_proto,
+				 enum vcap_keyfield_set keyset,
+				 struct vcap_keyset_list *orig)
+{
+	struct sparx5_port *port;
+	int lookup;
+
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS0:
+		lookup = sparx5_vcap_is0_cid_to_lookup(cid);
+		if (orig)
+			sparx5_vcap_is0_get_port_keysets(ndev, lookup, orig,
+							 l3_proto);
+		sparx5_vcap_is0_set_port_keyset(ndev, lookup, keyset, l3_proto);
+		break;
+	case VCAP_TYPE_IS2:
+		lookup = sparx5_vcap_is2_cid_to_lookup(cid);
+		if (orig)
+			sparx5_vcap_is2_get_port_keysets(ndev, lookup, orig,
+							 l3_proto);
+		sparx5_vcap_is2_set_port_keyset(ndev, lookup, keyset, l3_proto);
+		break;
+	case VCAP_TYPE_ES0:
+		break;
+	case VCAP_TYPE_ES2:
+		lookup = sparx5_vcap_es2_cid_to_lookup(cid);
+		if (orig)
+			sparx5_vcap_es2_get_port_keysets(ndev, lookup, orig,
+							 l3_proto);
+		sparx5_vcap_es2_set_port_keyset(ndev, lookup, keyset, l3_proto);
+		break;
+	default:
+		port = netdev_priv(ndev);
+		sparx5_vcap_type_err(port->sparx5, admin, __func__);
+		break;
+	}
+}
+
 /* Enable IS0 lookups per port and set the keyset generation */
 static void sparx5_vcap_is0_port_key_selection(struct sparx5 *sparx5,
 					       struct vcap_admin *admin)
diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.h b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.h
index 3260ab5e3a82..2684d9199b05 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.h
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.h
@@ -195,6 +195,12 @@ int sparx5_vcap_get_port_keyset(struct net_device *ndev,
 				u16 l3_proto,
 				struct vcap_keyset_list *kslist);
 
+/* Change the port keyset for the lookup and protocol */
+void sparx5_vcap_set_port_keyset(struct net_device *ndev,
+				 struct vcap_admin *admin, int cid,
+				 u16 l3_proto, enum vcap_keyfield_set keyset,
+				 struct vcap_keyset_list *orig);
+
 /* Check if the ethertype is supported by the vcap port classification */
 bool sparx5_vcap_is_known_etype(struct vcap_admin *admin, u16 etype);
 
-- 
cgit v1.2.3-70-g09d2


From e1d597ecbe5c5d7128630d4c9ec27c42018613ed Mon Sep 17 00:00:00 2001
From: Steen Hegelund <steen.hegelund@microchip.com>
Date: Tue, 7 Mar 2023 14:41:03 +0100
Subject: net: microchip: sparx5: Add TC template support

This adds support for using the "template add" and "template destroy"
functionality to change the port keyset configuration.

If the VCAP lookup already contains rules, the port keyset is left
unchanged, as a change would make these rules unusable.

When the template is destroyed the port keyset configuration is restored.
The filters using the template chain will automatically be deleted by the
TC framework.

Signed-off-by: Steen Hegelund <steen.hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/microchip/sparx5/sparx5_tc_flower.c   | 209 ++++++++++++++++++++-
 1 file changed, 202 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_tc_flower.c b/drivers/net/ethernet/microchip/sparx5/sparx5_tc_flower.c
index b36819aafaca..3f87a5285a6d 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_tc_flower.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_tc_flower.c
@@ -28,6 +28,14 @@ struct sparx5_multiple_rules {
 	struct sparx5_wildcard_rule rule[SPX5_MAX_RULE_SIZE];
 };
 
+struct sparx5_tc_flower_template {
+	struct list_head list; /* for insertion in the list of templates */
+	int cid; /* chain id */
+	enum vcap_keyfield_set orig; /* keyset used before the template */
+	enum vcap_keyfield_set keyset; /* new keyset used by template */
+	u16 l3_proto; /* protocol specified in the template */
+};
+
 static int
 sparx5_tc_flower_es0_tpid(struct vcap_tc_flower_parse_usage *st)
 {
@@ -382,7 +390,7 @@ static int sparx5_tc_select_protocol_keyset(struct net_device *ndev,
 	/* Find the keysets that the rule can use */
 	matches.keysets = keysets;
 	matches.max = ARRAY_SIZE(keysets);
-	if (vcap_rule_find_keysets(vrule, &matches) == 0)
+	if (!vcap_rule_find_keysets(vrule, &matches))
 		return -EINVAL;
 
 	/* Find the keysets that the port configuration supports */
@@ -996,6 +1004,73 @@ static int sparx5_tc_action_vlan_push(struct vcap_admin *admin,
 	return err;
 }
 
+/* Remove rule keys that may prevent templates from matching a keyset */
+static void sparx5_tc_flower_simplify_rule(struct vcap_admin *admin,
+					   struct vcap_rule *vrule,
+					   u16 l3_proto)
+{
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS0:
+		vcap_rule_rem_key(vrule, VCAP_KF_ETYPE);
+		switch (l3_proto) {
+		case ETH_P_IP:
+			break;
+		case ETH_P_IPV6:
+			vcap_rule_rem_key(vrule, VCAP_KF_IP_SNAP_IS);
+			break;
+		default:
+			break;
+		}
+		break;
+	case VCAP_TYPE_ES2:
+		switch (l3_proto) {
+		case ETH_P_IP:
+			if (vrule->keyset == VCAP_KFS_IP4_OTHER)
+				vcap_rule_rem_key(vrule, VCAP_KF_TCP_IS);
+			break;
+		case ETH_P_IPV6:
+			if (vrule->keyset == VCAP_KFS_IP6_STD)
+				vcap_rule_rem_key(vrule, VCAP_KF_TCP_IS);
+			vcap_rule_rem_key(vrule, VCAP_KF_IP4_IS);
+			break;
+		default:
+			break;
+		}
+		break;
+	case VCAP_TYPE_IS2:
+		switch (l3_proto) {
+		case ETH_P_IP:
+		case ETH_P_IPV6:
+			vcap_rule_rem_key(vrule, VCAP_KF_IP4_IS);
+			break;
+		default:
+			break;
+		}
+		break;
+	default:
+		break;
+	}
+}
+
+static bool sparx5_tc_flower_use_template(struct net_device *ndev,
+					  struct flow_cls_offload *fco,
+					  struct vcap_admin *admin,
+					  struct vcap_rule *vrule)
+{
+	struct sparx5_port *port = netdev_priv(ndev);
+	struct sparx5_tc_flower_template *ftp;
+
+	list_for_each_entry(ftp, &port->tc_templates, list) {
+		if (ftp->cid != fco->common.chain_index)
+			continue;
+
+		vcap_set_rule_set_keyset(vrule, ftp->keyset);
+		sparx5_tc_flower_simplify_rule(admin, vrule, ftp->l3_proto);
+		return true;
+	}
+	return false;
+}
+
 static int sparx5_tc_flower_replace(struct net_device *ndev,
 				    struct flow_cls_offload *fco,
 				    struct vcap_admin *admin,
@@ -1122,12 +1197,14 @@ static int sparx5_tc_flower_replace(struct net_device *ndev,
 			goto out;
 	}
 
-	err = sparx5_tc_select_protocol_keyset(ndev, vrule, admin,
-					       state.l3_proto, &multi);
-	if (err) {
-		NL_SET_ERR_MSG_MOD(fco->common.extack,
-				   "No matching port keyset for filter protocol and keys");
-		goto out;
+	if (!sparx5_tc_flower_use_template(ndev, fco, admin, vrule)) {
+		err = sparx5_tc_select_protocol_keyset(ndev, vrule, admin,
+						       state.l3_proto, &multi);
+		if (err) {
+			NL_SET_ERR_MSG_MOD(fco->common.extack,
+					   "No matching port keyset for filter protocol and keys");
+			goto out;
+		}
 	}
 
 	/* provide the l3 protocol to guide the keyset selection */
@@ -1259,6 +1336,120 @@ static int sparx5_tc_flower_stats(struct net_device *ndev,
 	return err;
 }
 
+static int sparx5_tc_flower_template_create(struct net_device *ndev,
+					    struct flow_cls_offload *fco,
+					    struct vcap_admin *admin)
+{
+	struct sparx5_port *port = netdev_priv(ndev);
+	struct vcap_tc_flower_parse_usage state = {
+		.fco = fco,
+		.l3_proto = ETH_P_ALL,
+		.admin = admin,
+	};
+	struct sparx5_tc_flower_template *ftp;
+	struct vcap_keyset_list kslist = {};
+	enum vcap_keyfield_set keysets[10];
+	struct vcap_control *vctrl;
+	struct vcap_rule *vrule;
+	int count, err;
+
+	if (admin->vtype == VCAP_TYPE_ES0) {
+		pr_err("%s:%d: %s\n", __func__, __LINE__,
+		       "VCAP does not support templates");
+		return -EINVAL;
+	}
+
+	count = vcap_admin_rule_count(admin, fco->common.chain_index);
+	if (count > 0) {
+		pr_err("%s:%d: %s\n", __func__, __LINE__,
+		       "Filters are already present");
+		return -EBUSY;
+	}
+
+	ftp = kzalloc(sizeof(*ftp), GFP_KERNEL);
+	if (!ftp)
+		return -ENOMEM;
+
+	ftp->cid = fco->common.chain_index;
+	ftp->orig = VCAP_KFS_NO_VALUE;
+	ftp->keyset = VCAP_KFS_NO_VALUE;
+
+	vctrl = port->sparx5->vcap_ctrl;
+	vrule = vcap_alloc_rule(vctrl, ndev, fco->common.chain_index,
+				VCAP_USER_TC, fco->common.prio, 0);
+	if (IS_ERR(vrule)) {
+		err = PTR_ERR(vrule);
+		goto err_rule;
+	}
+
+	state.vrule = vrule;
+	state.frule = flow_cls_offload_flow_rule(fco);
+	err = sparx5_tc_use_dissectors(&state, admin, vrule);
+	if (err) {
+		pr_err("%s:%d: key error: %d\n", __func__, __LINE__, err);
+		goto out;
+	}
+
+	ftp->l3_proto = state.l3_proto;
+
+	sparx5_tc_flower_simplify_rule(admin, vrule, state.l3_proto);
+
+	/* Find the keysets that the rule can use */
+	kslist.keysets = keysets;
+	kslist.max = ARRAY_SIZE(keysets);
+	if (!vcap_rule_find_keysets(vrule, &kslist)) {
+		pr_err("%s:%d: %s\n", __func__, __LINE__,
+		       "Could not find a suitable keyset");
+		err = -ENOENT;
+		goto out;
+	}
+
+	ftp->keyset = vcap_select_min_rule_keyset(vctrl, admin->vtype, &kslist);
+	kslist.cnt = 0;
+	sparx5_vcap_set_port_keyset(ndev, admin, fco->common.chain_index,
+				    state.l3_proto,
+				    ftp->keyset,
+				    &kslist);
+
+	if (kslist.cnt > 0)
+		ftp->orig = kslist.keysets[0];
+
+	/* Store new template */
+	list_add_tail(&ftp->list, &port->tc_templates);
+	vcap_free_rule(vrule);
+	return 0;
+
+out:
+	vcap_free_rule(vrule);
+err_rule:
+	kfree(ftp);
+	return err;
+}
+
+static int sparx5_tc_flower_template_destroy(struct net_device *ndev,
+					     struct flow_cls_offload *fco,
+					     struct vcap_admin *admin)
+{
+	struct sparx5_port *port = netdev_priv(ndev);
+	struct sparx5_tc_flower_template *ftp, *tmp;
+	int err = -ENOENT;
+
+	/* Rules using the template are removed by the tc framework */
+	list_for_each_entry_safe(ftp, tmp, &port->tc_templates, list) {
+		if (ftp->cid != fco->common.chain_index)
+			continue;
+
+		sparx5_vcap_set_port_keyset(ndev, admin,
+					    fco->common.chain_index,
+					    ftp->l3_proto, ftp->orig,
+					    NULL);
+		list_del(&ftp->list);
+		kfree(ftp);
+		break;
+	}
+	return err;
+}
+
 int sparx5_tc_flower(struct net_device *ndev, struct flow_cls_offload *fco,
 		     bool ingress)
 {
@@ -1282,6 +1473,10 @@ int sparx5_tc_flower(struct net_device *ndev, struct flow_cls_offload *fco,
 		return sparx5_tc_flower_destroy(ndev, fco, admin);
 	case FLOW_CLS_STATS:
 		return sparx5_tc_flower_stats(ndev, fco, admin);
+	case FLOW_CLS_TMPLT_CREATE:
+		return sparx5_tc_flower_template_create(ndev, fco, admin);
+	case FLOW_CLS_TMPLT_DESTROY:
+		return sparx5_tc_flower_template_destroy(ndev, fco, admin);
 	default:
 		return -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3-70-g09d2


From 4386b921857793440ebd4db3d6b70639149c7074 Mon Sep 17 00:00:00 2001
From: Sriram Yagnaraman <sriram.yagnaraman@est.tech>
Date: Fri, 24 Feb 2023 10:52:51 +0100
Subject: netfilter: bridge: introduce broute meta statement

nftables equivalent for ebtables -t broute.

Implement broute meta statement to set br_netfilter_broute flag
in skb to force a packet to be routed instead of being bridged.

Signed-off-by: Sriram Yagnaraman <sriram.yagnaraman@est.tech>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/uapi/linux/netfilter/nf_tables.h |  2 +
 net/bridge/netfilter/nft_meta_bridge.c   | 71 ++++++++++++++++++++++++++++++--
 2 files changed, 70 insertions(+), 3 deletions(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index ff677f3a6cad..9c6f02c26054 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -931,6 +931,7 @@ enum nft_exthdr_attributes {
  * @NFT_META_TIME_HOUR: hour of day (in seconds)
  * @NFT_META_SDIF: slave device interface index
  * @NFT_META_SDIFNAME: slave device interface name
+ * @NFT_META_BRI_BROUTE: packet br_netfilter_broute bit
  */
 enum nft_meta_keys {
 	NFT_META_LEN,
@@ -969,6 +970,7 @@ enum nft_meta_keys {
 	NFT_META_TIME_HOUR,
 	NFT_META_SDIF,
 	NFT_META_SDIFNAME,
+	NFT_META_BRI_BROUTE,
 	__NFT_META_IIFTYPE,
 };
 
diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c
index c3ecd77e25cb..bd4d1b4d745f 100644
--- a/net/bridge/netfilter/nft_meta_bridge.c
+++ b/net/bridge/netfilter/nft_meta_bridge.c
@@ -8,6 +8,9 @@
 #include <net/netfilter/nf_tables.h>
 #include <net/netfilter/nft_meta.h>
 #include <linux/if_bridge.h>
+#include <uapi/linux/netfilter_bridge.h> /* NF_BR_PRE_ROUTING */
+
+#include "../br_private.h"
 
 static const struct net_device *
 nft_meta_get_bridge(const struct net_device *dev)
@@ -102,6 +105,50 @@ static const struct nft_expr_ops nft_meta_bridge_get_ops = {
 	.reduce		= nft_meta_get_reduce,
 };
 
+static void nft_meta_bridge_set_eval(const struct nft_expr *expr,
+				     struct nft_regs *regs,
+				     const struct nft_pktinfo *pkt)
+{
+	const struct nft_meta *meta = nft_expr_priv(expr);
+	u32 *sreg = &regs->data[meta->sreg];
+	struct sk_buff *skb = pkt->skb;
+	u8 value8;
+
+	switch (meta->key) {
+	case NFT_META_BRI_BROUTE:
+		value8 = nft_reg_load8(sreg);
+		BR_INPUT_SKB_CB(skb)->br_netfilter_broute = !!value8;
+		break;
+	default:
+		nft_meta_set_eval(expr, regs, pkt);
+	}
+}
+
+static int nft_meta_bridge_set_init(const struct nft_ctx *ctx,
+				    const struct nft_expr *expr,
+				    const struct nlattr * const tb[])
+{
+	struct nft_meta *priv = nft_expr_priv(expr);
+	unsigned int len;
+	int err;
+
+	priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY]));
+	switch (priv->key) {
+	case NFT_META_BRI_BROUTE:
+		len = sizeof(u8);
+		break;
+	default:
+		return nft_meta_set_init(ctx, expr, tb);
+	}
+
+	priv->len = len;
+	err = nft_parse_register_load(tb[NFTA_META_SREG], &priv->sreg, len);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
 static bool nft_meta_bridge_set_reduce(struct nft_regs_track *track,
 				       const struct nft_expr *expr)
 {
@@ -120,15 +167,33 @@ static bool nft_meta_bridge_set_reduce(struct nft_regs_track *track,
 	return false;
 }
 
+static int nft_meta_bridge_set_validate(const struct nft_ctx *ctx,
+					const struct nft_expr *expr,
+					const struct nft_data **data)
+{
+	struct nft_meta *priv = nft_expr_priv(expr);
+	unsigned int hooks;
+
+	switch (priv->key) {
+	case NFT_META_BRI_BROUTE:
+		hooks = 1 << NF_BR_PRE_ROUTING;
+		break;
+	default:
+		return nft_meta_set_validate(ctx, expr, data);
+	}
+
+	return nft_chain_validate_hooks(ctx->chain, hooks);
+}
+
 static const struct nft_expr_ops nft_meta_bridge_set_ops = {
 	.type		= &nft_meta_bridge_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_meta)),
-	.eval		= nft_meta_set_eval,
-	.init		= nft_meta_set_init,
+	.eval		= nft_meta_bridge_set_eval,
+	.init		= nft_meta_bridge_set_init,
 	.destroy	= nft_meta_set_destroy,
 	.dump		= nft_meta_set_dump,
 	.reduce		= nft_meta_bridge_set_reduce,
-	.validate	= nft_meta_set_validate,
+	.validate	= nft_meta_bridge_set_validate,
 };
 
 static const struct nft_expr_ops *
-- 
cgit v1.2.3-70-g09d2


From 10369080454d87ee5b2db211ce947cb3118f0e13 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 7 Mar 2023 14:59:59 +0000
Subject: net: reclaim skb->scm_io_uring bit

Commit 0091bfc81741 ("io_uring/af_unix: defer registered
files gc to io_uring release") added one bit to struct sk_buff.

This structure is critical for networking, and we try very hard
to not add bloat on it, unless absolutely required.

For instance, we can use a specific destructor as a wrapper
around unix_destruct_scm(), to identify skbs that unix_gc()
has to special case.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Pavel Begunkov <asml.silence@gmail.com>
Cc: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Cc: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 2 --
 include/net/af_unix.h  | 1 +
 io_uring/rsrc.c        | 3 +--
 net/unix/garbage.c     | 2 +-
 net/unix/scm.c         | 6 ++++++
 5 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index ff7ad331fb82..fe661011644b 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -810,7 +810,6 @@ typedef unsigned char *sk_buff_data_t;
  *	@csum_level: indicates the number of consecutive checksums found in
  *		the packet minus one that have been verified as
  *		CHECKSUM_UNNECESSARY (max 3)
- *	@scm_io_uring: SKB holds io_uring registered files
  *	@dst_pending_confirm: need to confirm neighbour
  *	@decrypted: Decrypted SKB
  *	@slow_gro: state present at GRO time, slower prepare step required
@@ -989,7 +988,6 @@ struct sk_buff {
 #endif
 	__u8			slow_gro:1;
 	__u8			csum_not_inet:1;
-	__u8			scm_io_uring:1;
 
 #ifdef CONFIG_NET_SCHED
 	__u16			tc_index;	/* traffic control index */
diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index 480fa579787e..45ebde587138 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -11,6 +11,7 @@
 void unix_inflight(struct user_struct *user, struct file *fp);
 void unix_notinflight(struct user_struct *user, struct file *fp);
 void unix_destruct_scm(struct sk_buff *skb);
+void io_uring_destruct_scm(struct sk_buff *skb);
 void unix_gc(void);
 void wait_for_unix_gc(void);
 struct sock *unix_get_socket(struct file *filp);
diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c
index a59fc02de598..27ceda3b50cf 100644
--- a/io_uring/rsrc.c
+++ b/io_uring/rsrc.c
@@ -867,8 +867,7 @@ int __io_scm_file_account(struct io_ring_ctx *ctx, struct file *file)
 
 		UNIXCB(skb).fp = fpl;
 		skb->sk = sk;
-		skb->scm_io_uring = 1;
-		skb->destructor = unix_destruct_scm;
+		skb->destructor = io_uring_destruct_scm;
 		refcount_add(skb->truesize, &sk->sk_wmem_alloc);
 	}
 
diff --git a/net/unix/garbage.c b/net/unix/garbage.c
index dc2763540393..2405f0f9af31 100644
--- a/net/unix/garbage.c
+++ b/net/unix/garbage.c
@@ -305,7 +305,7 @@ void unix_gc(void)
 	 * release.path eventually putting registered files.
 	 */
 	skb_queue_walk_safe(&hitlist, skb, next_skb) {
-		if (skb->scm_io_uring) {
+		if (skb->destructor == io_uring_destruct_scm) {
 			__skb_unlink(skb, &hitlist);
 			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
 		}
diff --git a/net/unix/scm.c b/net/unix/scm.c
index aa27a02478dc..f9152881d77f 100644
--- a/net/unix/scm.c
+++ b/net/unix/scm.c
@@ -152,3 +152,9 @@ void unix_destruct_scm(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 EXPORT_SYMBOL(unix_destruct_scm);
+
+void io_uring_destruct_scm(struct sk_buff *skb)
+{
+	unix_destruct_scm(skb);
+}
+EXPORT_SYMBOL(io_uring_destruct_scm);
-- 
cgit v1.2.3-70-g09d2


From 9ccff83b1322f95da7a74784cf6f47a481e03dc5 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:31:27 -0500
Subject: netfilter: bridge: call pskb_may_pull in br_nf_check_hbh_len

When checking Hop-by-hop option header, if the option data is in
nonlinear area, it should do pskb_may_pull instead of discarding
the skb as a bad IPv6 packet.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/bridge/br_netfilter_ipv6.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/net/bridge/br_netfilter_ipv6.c b/net/bridge/br_netfilter_ipv6.c
index 6b07f30675bb..afd1c718b683 100644
--- a/net/bridge/br_netfilter_ipv6.c
+++ b/net/bridge/br_netfilter_ipv6.c
@@ -45,14 +45,18 @@
  */
 static int br_nf_check_hbh_len(struct sk_buff *skb)
 {
-	unsigned char *raw = (u8 *)(ipv6_hdr(skb) + 1);
+	int len, off = sizeof(struct ipv6hdr);
+	unsigned char *nh;
 	u32 pkt_len;
-	const unsigned char *nh = skb_network_header(skb);
-	int off = raw - nh;
-	int len = (raw[1] + 1) << 3;
 
-	if ((raw + len) - skb->data > skb_headlen(skb))
+	if (!pskb_may_pull(skb, off + 8))
 		goto bad;
+	nh = (unsigned char *)(ipv6_hdr(skb) + 1);
+	len = (nh[1] + 1) << 3;
+
+	if (!pskb_may_pull(skb, off + len))
+		goto bad;
+	nh = skb_network_header(skb);
 
 	off += 2;
 	len -= 2;
-- 
cgit v1.2.3-70-g09d2


From a7f1a2f43e683c8ffca691d45f2cb32c052158fa Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:31:28 -0500
Subject: netfilter: bridge: check len before accessing more nh data

In the while loop of br_nf_check_hbh_len(), similar to ip6_parse_tlv(),
before accessing 'nh[off + 1]', it should add a check 'len < 2'; and
before parsing IPV6_TLV_JUMBO, it should add a check 'optlen > len',
in case of overflows.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/bridge/br_netfilter_ipv6.c | 45 +++++++++++++++++++-----------------------
 1 file changed, 20 insertions(+), 25 deletions(-)

diff --git a/net/bridge/br_netfilter_ipv6.c b/net/bridge/br_netfilter_ipv6.c
index afd1c718b683..8be3c5c8b925 100644
--- a/net/bridge/br_netfilter_ipv6.c
+++ b/net/bridge/br_netfilter_ipv6.c
@@ -50,54 +50,49 @@ static int br_nf_check_hbh_len(struct sk_buff *skb)
 	u32 pkt_len;
 
 	if (!pskb_may_pull(skb, off + 8))
-		goto bad;
+		return -1;
 	nh = (unsigned char *)(ipv6_hdr(skb) + 1);
 	len = (nh[1] + 1) << 3;
 
 	if (!pskb_may_pull(skb, off + len))
-		goto bad;
+		return -1;
 	nh = skb_network_header(skb);
 
 	off += 2;
 	len -= 2;
-
 	while (len > 0) {
-		int optlen = nh[off + 1] + 2;
-
-		switch (nh[off]) {
-		case IPV6_TLV_PAD1:
-			optlen = 1;
-			break;
+		int optlen;
 
-		case IPV6_TLV_PADN:
-			break;
+		if (nh[off] == IPV6_TLV_PAD1) {
+			off++;
+			len--;
+			continue;
+		}
+		if (len < 2)
+			return -1;
+		optlen = nh[off + 1] + 2;
+		if (optlen > len)
+			return -1;
 
-		case IPV6_TLV_JUMBO:
+		if (nh[off] == IPV6_TLV_JUMBO) {
 			if (nh[off + 1] != 4 || (off & 3) != 2)
-				goto bad;
+				return -1;
 			pkt_len = ntohl(*(__be32 *)(nh + off + 2));
 			if (pkt_len <= IPV6_MAXPLEN ||
 			    ipv6_hdr(skb)->payload_len)
-				goto bad;
+				return -1;
 			if (pkt_len > skb->len - sizeof(struct ipv6hdr))
-				goto bad;
+				return -1;
 			if (pskb_trim_rcsum(skb,
 					    pkt_len + sizeof(struct ipv6hdr)))
-				goto bad;
+				return -1;
 			nh = skb_network_header(skb);
-			break;
-		default:
-			if (optlen > len)
-				goto bad;
-			break;
 		}
 		off += optlen;
 		len -= optlen;
 	}
-	if (len == 0)
-		return 0;
-bad:
-	return -1;
+
+	return len ? -1 : 0;
 }
 
 int br_validate_ipv6(struct net *net, struct sk_buff *skb)
-- 
cgit v1.2.3-70-g09d2


From 0b24bd71a6c0214aaa2115302dd6598b89d2fa8a Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:31:29 -0500
Subject: netfilter: bridge: move pskb_trim_rcsum out of br_nf_check_hbh_len

br_nf_check_hbh_len() is a function to check the Hop-by-hop option
header, and shouldn't do pskb_trim_rcsum() there. This patch is to
pass pkt_len out to br_validate_ipv6() and do pskb_trim_rcsum()
after calling br_validate_ipv6() instead.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/bridge/br_netfilter_ipv6.c | 33 ++++++++++++++-------------------
 1 file changed, 14 insertions(+), 19 deletions(-)

diff --git a/net/bridge/br_netfilter_ipv6.c b/net/bridge/br_netfilter_ipv6.c
index 8be3c5c8b925..a0d6dfb3e255 100644
--- a/net/bridge/br_netfilter_ipv6.c
+++ b/net/bridge/br_netfilter_ipv6.c
@@ -43,11 +43,10 @@
 /* We only check the length. A bridge shouldn't do any hop-by-hop stuff
  * anyway
  */
-static int br_nf_check_hbh_len(struct sk_buff *skb)
+static int br_nf_check_hbh_len(struct sk_buff *skb, u32 *plen)
 {
 	int len, off = sizeof(struct ipv6hdr);
 	unsigned char *nh;
-	u32 pkt_len;
 
 	if (!pskb_may_pull(skb, off + 8))
 		return -1;
@@ -75,6 +74,8 @@ static int br_nf_check_hbh_len(struct sk_buff *skb)
 			return -1;
 
 		if (nh[off] == IPV6_TLV_JUMBO) {
+			u32 pkt_len;
+
 			if (nh[off + 1] != 4 || (off & 3) != 2)
 				return -1;
 			pkt_len = ntohl(*(__be32 *)(nh + off + 2));
@@ -83,10 +84,7 @@ static int br_nf_check_hbh_len(struct sk_buff *skb)
 				return -1;
 			if (pkt_len > skb->len - sizeof(struct ipv6hdr))
 				return -1;
-			if (pskb_trim_rcsum(skb,
-					    pkt_len + sizeof(struct ipv6hdr)))
-				return -1;
-			nh = skb_network_header(skb);
+			*plen = pkt_len;
 		}
 		off += optlen;
 		len -= optlen;
@@ -114,22 +112,19 @@ int br_validate_ipv6(struct net *net, struct sk_buff *skb)
 		goto inhdr_error;
 
 	pkt_len = ntohs(hdr->payload_len);
+	if (hdr->nexthdr == NEXTHDR_HOP && br_nf_check_hbh_len(skb, &pkt_len))
+		goto drop;
 
-	if (pkt_len || hdr->nexthdr != NEXTHDR_HOP) {
-		if (pkt_len + ip6h_len > skb->len) {
-			__IP6_INC_STATS(net, idev,
-					IPSTATS_MIB_INTRUNCATEDPKTS);
-			goto drop;
-		}
-		if (pskb_trim_rcsum(skb, pkt_len + ip6h_len)) {
-			__IP6_INC_STATS(net, idev,
-					IPSTATS_MIB_INDISCARDS);
-			goto drop;
-		}
-		hdr = ipv6_hdr(skb);
+	if (pkt_len + ip6h_len > skb->len) {
+		__IP6_INC_STATS(net, idev,
+				IPSTATS_MIB_INTRUNCATEDPKTS);
+		goto drop;
 	}
-	if (hdr->nexthdr == NEXTHDR_HOP && br_nf_check_hbh_len(skb))
+	if (pskb_trim_rcsum(skb, pkt_len + ip6h_len)) {
+		__IP6_INC_STATS(net, idev,
+				IPSTATS_MIB_INDISCARDS);
 		goto drop;
+	}
 
 	memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm));
 	/* No IP options in IPv6 header; however it should be
-- 
cgit v1.2.3-70-g09d2


From 28e144cf5f72ce1c304571bc448e37c27495903a Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:31:30 -0500
Subject: netfilter: move br_nf_check_hbh_len to utils

Rename br_nf_check_hbh_len() to nf_ip6_check_hbh_len() and move it
to netfilter utils, so that it can be used by other modules, like
ovs and tc.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/linux/netfilter_ipv6.h |  2 ++
 net/bridge/br_netfilter_ipv6.c | 55 +-----------------------------------------
 net/netfilter/utils.c          | 52 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 55 insertions(+), 54 deletions(-)

diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index 48314ade1506..7834c0be2831 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -197,6 +197,8 @@ static inline int nf_cookie_v6_check(const struct ipv6hdr *iph,
 __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 			unsigned int dataoff, u_int8_t protocol);
 
+int nf_ip6_check_hbh_len(struct sk_buff *skb, u32 *plen);
+
 int ipv6_netfilter_init(void);
 void ipv6_netfilter_fini(void);
 
diff --git a/net/bridge/br_netfilter_ipv6.c b/net/bridge/br_netfilter_ipv6.c
index a0d6dfb3e255..550039dfc31a 100644
--- a/net/bridge/br_netfilter_ipv6.c
+++ b/net/bridge/br_netfilter_ipv6.c
@@ -40,59 +40,6 @@
 #include <linux/sysctl.h>
 #endif
 
-/* We only check the length. A bridge shouldn't do any hop-by-hop stuff
- * anyway
- */
-static int br_nf_check_hbh_len(struct sk_buff *skb, u32 *plen)
-{
-	int len, off = sizeof(struct ipv6hdr);
-	unsigned char *nh;
-
-	if (!pskb_may_pull(skb, off + 8))
-		return -1;
-	nh = (unsigned char *)(ipv6_hdr(skb) + 1);
-	len = (nh[1] + 1) << 3;
-
-	if (!pskb_may_pull(skb, off + len))
-		return -1;
-	nh = skb_network_header(skb);
-
-	off += 2;
-	len -= 2;
-	while (len > 0) {
-		int optlen;
-
-		if (nh[off] == IPV6_TLV_PAD1) {
-			off++;
-			len--;
-			continue;
-		}
-		if (len < 2)
-			return -1;
-		optlen = nh[off + 1] + 2;
-		if (optlen > len)
-			return -1;
-
-		if (nh[off] == IPV6_TLV_JUMBO) {
-			u32 pkt_len;
-
-			if (nh[off + 1] != 4 || (off & 3) != 2)
-				return -1;
-			pkt_len = ntohl(*(__be32 *)(nh + off + 2));
-			if (pkt_len <= IPV6_MAXPLEN ||
-			    ipv6_hdr(skb)->payload_len)
-				return -1;
-			if (pkt_len > skb->len - sizeof(struct ipv6hdr))
-				return -1;
-			*plen = pkt_len;
-		}
-		off += optlen;
-		len -= optlen;
-	}
-
-	return len ? -1 : 0;
-}
-
 int br_validate_ipv6(struct net *net, struct sk_buff *skb)
 {
 	const struct ipv6hdr *hdr;
@@ -112,7 +59,7 @@ int br_validate_ipv6(struct net *net, struct sk_buff *skb)
 		goto inhdr_error;
 
 	pkt_len = ntohs(hdr->payload_len);
-	if (hdr->nexthdr == NEXTHDR_HOP && br_nf_check_hbh_len(skb, &pkt_len))
+	if (hdr->nexthdr == NEXTHDR_HOP && nf_ip6_check_hbh_len(skb, &pkt_len))
 		goto drop;
 
 	if (pkt_len + ip6h_len > skb->len) {
diff --git a/net/netfilter/utils.c b/net/netfilter/utils.c
index 2182d361e273..acef4155f0da 100644
--- a/net/netfilter/utils.c
+++ b/net/netfilter/utils.c
@@ -215,3 +215,55 @@ int nf_reroute(struct sk_buff *skb, struct nf_queue_entry *entry)
 	}
 	return ret;
 }
+
+/* Only get and check the lengths, not do any hop-by-hop stuff. */
+int nf_ip6_check_hbh_len(struct sk_buff *skb, u32 *plen)
+{
+	int len, off = sizeof(struct ipv6hdr);
+	unsigned char *nh;
+
+	if (!pskb_may_pull(skb, off + 8))
+		return -ENOMEM;
+	nh = (unsigned char *)(ipv6_hdr(skb) + 1);
+	len = (nh[1] + 1) << 3;
+
+	if (!pskb_may_pull(skb, off + len))
+		return -ENOMEM;
+	nh = skb_network_header(skb);
+
+	off += 2;
+	len -= 2;
+	while (len > 0) {
+		int optlen;
+
+		if (nh[off] == IPV6_TLV_PAD1) {
+			off++;
+			len--;
+			continue;
+		}
+		if (len < 2)
+			return -EBADMSG;
+		optlen = nh[off + 1] + 2;
+		if (optlen > len)
+			return -EBADMSG;
+
+		if (nh[off] == IPV6_TLV_JUMBO) {
+			u32 pkt_len;
+
+			if (nh[off + 1] != 4 || (off & 3) != 2)
+				return -EBADMSG;
+			pkt_len = ntohl(*(__be32 *)(nh + off + 2));
+			if (pkt_len <= IPV6_MAXPLEN ||
+			    ipv6_hdr(skb)->payload_len)
+				return -EBADMSG;
+			if (pkt_len > skb->len - sizeof(struct ipv6hdr))
+				return -EBADMSG;
+			*plen = pkt_len;
+		}
+		off += optlen;
+		len -= optlen;
+	}
+
+	return len ? -EBADMSG : 0;
+}
+EXPORT_SYMBOL_GPL(nf_ip6_check_hbh_len);
-- 
cgit v1.2.3-70-g09d2


From eaafdaa3e92234b877b645431957549a1f87e2bf Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:31:31 -0500
Subject: netfilter: use nf_ip6_check_hbh_len in nf_ct_skb_network_trim

For IPv6 Jumbo packets, the ipv6_hdr(skb)->payload_len is always 0,
and its real payload_len ( > 65535) is saved in hbh exthdr. With 0
length for the jumbo packets, all data and exthdr will be trimmed
in nf_ct_skb_network_trim().

This patch is to call nf_ip6_check_hbh_len() to get real pkt_len
of the IPv6 packet, similar to br_validate_ipv6().

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nf_conntrack_ovs.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_conntrack_ovs.c b/net/netfilter/nf_conntrack_ovs.c
index 52b776bdf526..068e9489e1c2 100644
--- a/net/netfilter/nf_conntrack_ovs.c
+++ b/net/netfilter/nf_conntrack_ovs.c
@@ -6,6 +6,7 @@
 #include <net/netfilter/ipv6/nf_defrag_ipv6.h>
 #include <net/ipv6_frag.h>
 #include <net/ip.h>
+#include <linux/netfilter_ipv6.h>
 
 /* 'skb' should already be pulled to nh_ofs. */
 int nf_ct_helper(struct sk_buff *skb, struct nf_conn *ct,
@@ -120,8 +121,14 @@ int nf_ct_skb_network_trim(struct sk_buff *skb, int family)
 		len = skb_ip_totlen(skb);
 		break;
 	case NFPROTO_IPV6:
-		len = sizeof(struct ipv6hdr)
-			+ ntohs(ipv6_hdr(skb)->payload_len);
+		len = ntohs(ipv6_hdr(skb)->payload_len);
+		if (ipv6_hdr(skb)->nexthdr == NEXTHDR_HOP) {
+			int err = nf_ip6_check_hbh_len(skb, &len);
+
+			if (err)
+				return err;
+		}
+		len += sizeof(struct ipv6hdr);
 		break;
 	default:
 		len = skb->len;
-- 
cgit v1.2.3-70-g09d2


From 6bb382bcf742de5c17209848325653059c995a04 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:31:32 -0500
Subject: selftests: add a selftest for big tcp

This test runs on the client-router-server topo, and monitors the traffic
on the RX devices of router and server while sending BIG TCP packets with
netperf from client to server. Meanwhile, it changes 'tso' on the TX devs
and 'gro' on the RX devs. Then it checks if any BIG TCP packets appears
on the RX devs with 'ip/ip6tables -m length ! --length 0:65535' for each
case.

Note that we also add tc action ct in link1 ingress to cover the ipv6
jumbo packets process in nf_ct_skb_network_trim() of nf_conntrack_ovs.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 tools/testing/selftests/net/Makefile   |   1 +
 tools/testing/selftests/net/big_tcp.sh | 180 +++++++++++++++++++++++++++++++++
 2 files changed, 181 insertions(+)
 create mode 100755 tools/testing/selftests/net/big_tcp.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 6cd8993454d7..099741290184 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -48,6 +48,7 @@ TEST_PROGS += l2_tos_ttl_inherit.sh
 TEST_PROGS += bind_bhash.sh
 TEST_PROGS += ip_local_port_range.sh
 TEST_PROGS += rps_default_mask.sh
+TEST_PROGS += big_tcp.sh
 TEST_PROGS_EXTENDED := in_netns.sh setup_loopback.sh setup_veth.sh
 TEST_PROGS_EXTENDED += toeplitz_client.sh toeplitz.sh
 TEST_GEN_FILES =  socket nettest
diff --git a/tools/testing/selftests/net/big_tcp.sh b/tools/testing/selftests/net/big_tcp.sh
new file mode 100755
index 000000000000..cde9a91c4797
--- /dev/null
+++ b/tools/testing/selftests/net/big_tcp.sh
@@ -0,0 +1,180 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Testing For IPv4 and IPv6 BIG TCP.
+# TOPO: CLIENT_NS (link0)<--->(link1) ROUTER_NS (link2)<--->(link3) SERVER_NS
+
+CLIENT_NS=$(mktemp -u client-XXXXXXXX)
+CLIENT_IP4="198.51.100.1"
+CLIENT_IP6="2001:db8:1::1"
+
+SERVER_NS=$(mktemp -u server-XXXXXXXX)
+SERVER_IP4="203.0.113.1"
+SERVER_IP6="2001:db8:2::1"
+
+ROUTER_NS=$(mktemp -u router-XXXXXXXX)
+SERVER_GW4="203.0.113.2"
+CLIENT_GW4="198.51.100.2"
+SERVER_GW6="2001:db8:2::2"
+CLIENT_GW6="2001:db8:1::2"
+
+MAX_SIZE=128000
+CHK_SIZE=65535
+
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+setup() {
+	ip netns add $CLIENT_NS
+	ip netns add $SERVER_NS
+	ip netns add $ROUTER_NS
+	ip -net $ROUTER_NS link add link1 type veth peer name link0 netns $CLIENT_NS
+	ip -net $ROUTER_NS link add link2 type veth peer name link3 netns $SERVER_NS
+
+	ip -net $CLIENT_NS link set link0 up
+	ip -net $CLIENT_NS link set link0 mtu 1442
+	ip -net $CLIENT_NS addr add $CLIENT_IP4/24 dev link0
+	ip -net $CLIENT_NS addr add $CLIENT_IP6/64 dev link0 nodad
+	ip -net $CLIENT_NS route add $SERVER_IP4 dev link0 via $CLIENT_GW4
+	ip -net $CLIENT_NS route add $SERVER_IP6 dev link0 via $CLIENT_GW6
+	ip -net $CLIENT_NS link set dev link0 \
+		gro_ipv4_max_size $MAX_SIZE gso_ipv4_max_size $MAX_SIZE
+	ip -net $CLIENT_NS link set dev link0 \
+		gro_max_size $MAX_SIZE gso_max_size $MAX_SIZE
+	ip net exec $CLIENT_NS sysctl -wq net.ipv4.tcp_window_scaling=10
+
+	ip -net $ROUTER_NS link set link1 up
+	ip -net $ROUTER_NS link set link2 up
+	ip -net $ROUTER_NS addr add $CLIENT_GW4/24 dev link1
+	ip -net $ROUTER_NS addr add $CLIENT_GW6/64 dev link1 nodad
+	ip -net $ROUTER_NS addr add $SERVER_GW4/24 dev link2
+	ip -net $ROUTER_NS addr add $SERVER_GW6/64 dev link2 nodad
+	ip -net $ROUTER_NS link set dev link1 \
+		gro_ipv4_max_size $MAX_SIZE gso_ipv4_max_size $MAX_SIZE
+	ip -net $ROUTER_NS link set dev link2 \
+		gro_ipv4_max_size $MAX_SIZE gso_ipv4_max_size $MAX_SIZE
+	ip -net $ROUTER_NS link set dev link1 \
+		gro_max_size $MAX_SIZE gso_max_size $MAX_SIZE
+	ip -net $ROUTER_NS link set dev link2 \
+		gro_max_size $MAX_SIZE gso_max_size $MAX_SIZE
+	# test for nf_ct_skb_network_trim in nf_conntrack_ovs used by TC ct action.
+	ip net exec $ROUTER_NS tc qdisc add dev link1 ingress
+	ip net exec $ROUTER_NS tc filter add dev link1 ingress \
+		proto ip flower ip_proto tcp action ct
+	ip net exec $ROUTER_NS tc filter add dev link1 ingress \
+		proto ipv6 flower ip_proto tcp action ct
+	ip net exec $ROUTER_NS sysctl -wq net.ipv4.ip_forward=1
+	ip net exec $ROUTER_NS sysctl -wq net.ipv6.conf.all.forwarding=1
+
+	ip -net $SERVER_NS link set link3 up
+	ip -net $SERVER_NS addr add $SERVER_IP4/24 dev link3
+	ip -net $SERVER_NS addr add $SERVER_IP6/64 dev link3 nodad
+	ip -net $SERVER_NS route add $CLIENT_IP4 dev link3 via $SERVER_GW4
+	ip -net $SERVER_NS route add $CLIENT_IP6 dev link3 via $SERVER_GW6
+	ip -net $SERVER_NS link set dev link3 \
+		gro_ipv4_max_size $MAX_SIZE gso_ipv4_max_size $MAX_SIZE
+	ip -net $SERVER_NS link set dev link3 \
+		gro_max_size $MAX_SIZE gso_max_size $MAX_SIZE
+	ip net exec $SERVER_NS sysctl -wq net.ipv4.tcp_window_scaling=10
+	ip net exec $SERVER_NS netserver 2>&1 >/dev/null
+}
+
+cleanup() {
+	ip net exec $SERVER_NS pkill netserver
+	ip -net $ROUTER_NS link del link1
+	ip -net $ROUTER_NS link del link2
+	ip netns del "$CLIENT_NS"
+	ip netns del "$SERVER_NS"
+	ip netns del "$ROUTER_NS"
+}
+
+start_counter() {
+	local ipt="iptables"
+	local iface=$1
+	local netns=$2
+
+	[ "$NF" = "6" ] && ipt="ip6tables"
+	ip net exec $netns $ipt -t raw -A PREROUTING -i $iface \
+		-m length ! --length 0:$CHK_SIZE -j ACCEPT
+}
+
+check_counter() {
+	local ipt="iptables"
+	local iface=$1
+	local netns=$2
+
+	[ "$NF" = "6" ] && ipt="ip6tables"
+	test `ip net exec $netns $ipt -t raw -L -v |grep $iface | awk '{print $1}'` != "0"
+}
+
+stop_counter() {
+	local ipt="iptables"
+	local iface=$1
+	local netns=$2
+
+	[ "$NF" = "6" ] && ipt="ip6tables"
+	ip net exec $netns $ipt -t raw -D PREROUTING -i $iface \
+		-m length ! --length 0:$CHK_SIZE -j ACCEPT
+}
+
+do_netperf() {
+	local serip=$SERVER_IP4
+	local netns=$1
+
+	[ "$NF" = "6" ] && serip=$SERVER_IP6
+	ip net exec $netns netperf -$NF -t TCP_STREAM -H $serip 2>&1 >/dev/null
+}
+
+do_test() {
+	local cli_tso=$1
+	local gw_gro=$2
+	local gw_tso=$3
+	local ser_gro=$4
+	local ret="PASS"
+
+	ip net exec $CLIENT_NS ethtool -K link0 tso $cli_tso
+	ip net exec $ROUTER_NS ethtool -K link1 gro $gw_gro
+	ip net exec $ROUTER_NS ethtool -K link2 tso $gw_tso
+	ip net exec $SERVER_NS ethtool -K link3 gro $ser_gro
+
+	start_counter link1 $ROUTER_NS
+	start_counter link3 $SERVER_NS
+	do_netperf $CLIENT_NS
+
+	if check_counter link1 $ROUTER_NS; then
+		check_counter link3 $SERVER_NS || ret="FAIL_on_link3"
+	else
+		ret="FAIL_on_link1"
+	fi
+
+	stop_counter link1 $ROUTER_NS
+	stop_counter link3 $SERVER_NS
+	printf "%-9s %-8s %-8s %-8s: [%s]\n" \
+		$cli_tso $gw_gro $gw_tso $ser_gro $ret
+	test $ret = "PASS"
+}
+
+testup() {
+	echo "CLI GSO | GW GRO | GW GSO | SER GRO" && \
+	do_test "on"  "on"  "on"  "on"  && \
+	do_test "on"  "off" "on"  "off" && \
+	do_test "off" "on"  "on"  "on"  && \
+	do_test "on"  "on"  "off" "on"  && \
+	do_test "off" "on"  "off" "on"
+}
+
+if ! netperf -V &> /dev/null; then
+	echo "SKIP: Could not run test without netperf tool"
+	exit $ksft_skip
+fi
+
+if ! ip link help 2>&1 | grep gso_ipv4_max_size &> /dev/null; then
+	echo "SKIP: Could not run test without gso/gro_ipv4_max_size supported in ip-link"
+	exit $ksft_skip
+fi
+
+trap cleanup EXIT
+setup && echo "Testing for BIG TCP:" && \
+NF=4 testup && echo "***v4 Tests Done***" && \
+NF=6 testup && echo "***v6 Tests Done***"
+exit $?
-- 
cgit v1.2.3-70-g09d2


From e5d015a114dad6a50c22cf282007f7d92086df46 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Tue, 7 Mar 2023 23:30:48 +0000
Subject: netfilter: conntrack: fix typo

There's a spelling mistake in a comment.  Fix it.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nf_conntrack_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 7250082e7de5..004c54132a3b 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1294,7 +1294,7 @@ dying:
 }
 EXPORT_SYMBOL_GPL(__nf_conntrack_confirm);
 
-/* Returns true if a connection correspondings to the tuple (required
+/* Returns true if a connection corresponds to the tuple (required
    for NAT). */
 int
 nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple,
-- 
cgit v1.2.3-70-g09d2


From b0ca200077b3872056e6a8291c9a50f803658c2a Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Tue, 7 Mar 2023 23:30:49 +0000
Subject: netfilter: nat: fix indentation of function arguments

A couple of arguments to a function call are incorrectly indented.
Fix them.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nf_nat_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c
index e29e4ccb5c5a..ce829d434f13 100644
--- a/net/netfilter/nf_nat_core.c
+++ b/net/netfilter/nf_nat_core.c
@@ -549,8 +549,8 @@ get_unique_tuple(struct nf_conntrack_tuple *tuple,
 		if (range->flags & NF_NAT_RANGE_PROTO_SPECIFIED) {
 			if (!(range->flags & NF_NAT_RANGE_PROTO_OFFSET) &&
 			    l4proto_in_range(tuple, maniptype,
-			          &range->min_proto,
-			          &range->max_proto) &&
+					     &range->min_proto,
+					     &range->max_proto) &&
 			    (range->min_proto.all == range->max_proto.all ||
 			     !nf_nat_used_tuple(tuple, ct)))
 				return;
-- 
cgit v1.2.3-70-g09d2


From 12fabae03ca6474fd571bf6ddb37d009533305d6 Mon Sep 17 00:00:00 2001
From: Roberto Sassu <roberto.sassu@huawei.com>
Date: Wed, 8 Mar 2023 11:37:13 +0100
Subject: selftests/bpf: Fix IMA test

Commit 62622dab0a28 ("ima: return IMA digest value only when IMA_COLLECTED
flag is set") caused bpf_ima_inode_hash() to refuse to give non-fresh
digests. IMA test #3 assumed the old behavior, that bpf_ima_inode_hash()
still returned also non-fresh digests.

Correct the test by accepting both cases. If the samples returned are 1,
assume that the commit above is applied and that the returned digest is
fresh. If the samples returned are 2, assume that the commit above is not
applied, and check both the non-fresh and fresh digest.

Fixes: 62622dab0a28 ("ima: return IMA digest value only when IMA_COLLECTED flag is set")
Reported-by: David Vernet <void@manifault.com>
Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Matt Bobrowski <mattbobrowski@google.com>
Link: https://lore.kernel.org/bpf/20230308103713.1681200-1-roberto.sassu@huaweicloud.com
---
 tools/testing/selftests/bpf/prog_tests/test_ima.c | 29 ++++++++++++++++-------
 1 file changed, 21 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_ima.c b/tools/testing/selftests/bpf/prog_tests/test_ima.c
index b13feceb38f1..810b14981c2e 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_ima.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_ima.c
@@ -70,7 +70,7 @@ void test_test_ima(void)
 	u64 bin_true_sample;
 	char cmd[256];
 
-	int err, duration = 0;
+	int err, duration = 0, fresh_digest_idx = 0;
 	struct ima *skel = NULL;
 
 	skel = ima__open_and_load();
@@ -129,7 +129,15 @@ void test_test_ima(void)
 	/*
 	 * Test #3
 	 * - Goal: confirm that bpf_ima_inode_hash() returns a non-fresh digest
-	 * - Expected result: 2 samples (/bin/true: non-fresh, fresh)
+	 * - Expected result:
+	 *   1 sample (/bin/true: fresh) if commit 62622dab0a28 applied
+	 *   2 samples (/bin/true: non-fresh, fresh) if commit 62622dab0a28 is
+	 *     not applied
+	 *
+	 * If commit 62622dab0a28 ("ima: return IMA digest value only when
+	 * IMA_COLLECTED flag is set") is applied, bpf_ima_inode_hash() refuses
+	 * to give a non-fresh digest, hence the correct result is 1 instead of
+	 * 2.
 	 */
 	test_init(skel->bss);
 
@@ -144,13 +152,18 @@ void test_test_ima(void)
 		goto close_clean;
 
 	err = ring_buffer__consume(ringbuf);
-	ASSERT_EQ(err, 2, "num_samples_or_err");
-	ASSERT_NEQ(ima_hash_from_bpf[0], 0, "ima_hash");
-	ASSERT_NEQ(ima_hash_from_bpf[1], 0, "ima_hash");
-	ASSERT_EQ(ima_hash_from_bpf[0], bin_true_sample, "sample_equal_or_err");
+	ASSERT_GE(err, 1, "num_samples_or_err");
+	if (err == 2) {
+		ASSERT_NEQ(ima_hash_from_bpf[0], 0, "ima_hash");
+		ASSERT_EQ(ima_hash_from_bpf[0], bin_true_sample,
+			  "sample_equal_or_err");
+		fresh_digest_idx = 1;
+	}
+
+	ASSERT_NEQ(ima_hash_from_bpf[fresh_digest_idx], 0, "ima_hash");
 	/* IMA refreshed the digest. */
-	ASSERT_NEQ(ima_hash_from_bpf[1], bin_true_sample,
-		   "sample_different_or_err");
+	ASSERT_NEQ(ima_hash_from_bpf[fresh_digest_idx], bin_true_sample,
+		   "sample_equal_or_err");
 
 	/*
 	 * Test #4
-- 
cgit v1.2.3-70-g09d2


From 07236eab7a3139da97aef9f5f21f403be82a82ea Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:14 -0800
Subject: bpf: factor out fetching basic kfunc metadata

Factor out logic to fetch basic kfunc metadata based on struct bpf_insn.
This is not exactly short or trivial code to just copy/paste and this
information is sometimes necessary in other parts of the verifier logic.
Subsequent patches will rely on this to determine if an instruction is
a kfunc call to iterator next method.

No functional changes intended, including that verbose() warning
behavior when kfunc is not allowed for a particular program type.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 92 +++++++++++++++++++++++++++++++++------------------
 1 file changed, 59 insertions(+), 33 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index b2116ca78d9a..8d40fba6a1c0 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -10079,24 +10079,21 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 	return 0;
 }
 
-static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
-			    int *insn_idx_p)
+static int fetch_kfunc_meta(struct bpf_verifier_env *env,
+			    struct bpf_insn *insn,
+			    struct bpf_kfunc_call_arg_meta *meta,
+			    const char **kfunc_name)
 {
-	const struct btf_type *t, *func, *func_proto, *ptr_type;
-	u32 i, nargs, func_id, ptr_type_id, release_ref_obj_id;
-	struct bpf_reg_state *regs = cur_regs(env);
-	const char *func_name, *ptr_type_name;
-	bool sleepable, rcu_lock, rcu_unlock;
-	struct bpf_kfunc_call_arg_meta meta;
-	int err, insn_idx = *insn_idx_p;
-	const struct btf_param *args;
-	const struct btf_type *ret_t;
+	const struct btf_type *func, *func_proto;
+	u32 func_id, *kfunc_flags;
+	const char *func_name;
 	struct btf *desc_btf;
-	u32 *kfunc_flags;
 
-	/* skip for now, but return error when we find this in fixup_kfunc_call */
+	if (kfunc_name)
+		*kfunc_name = NULL;
+
 	if (!insn->imm)
-		return 0;
+		return -EINVAL;
 
 	desc_btf = find_kfunc_desc_btf(env, insn->off);
 	if (IS_ERR(desc_btf))
@@ -10105,22 +10102,51 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 	func_id = insn->imm;
 	func = btf_type_by_id(desc_btf, func_id);
 	func_name = btf_name_by_offset(desc_btf, func->name_off);
+	if (kfunc_name)
+		*kfunc_name = func_name;
 	func_proto = btf_type_by_id(desc_btf, func->type);
 
 	kfunc_flags = btf_kfunc_id_set_contains(desc_btf, resolve_prog_type(env->prog), func_id);
 	if (!kfunc_flags) {
-		verbose(env, "calling kernel function %s is not allowed\n",
-			func_name);
 		return -EACCES;
 	}
 
-	/* Prepare kfunc call metadata */
-	memset(&meta, 0, sizeof(meta));
-	meta.btf = desc_btf;
-	meta.func_id = func_id;
-	meta.kfunc_flags = *kfunc_flags;
-	meta.func_proto = func_proto;
-	meta.func_name = func_name;
+	memset(meta, 0, sizeof(*meta));
+	meta->btf = desc_btf;
+	meta->func_id = func_id;
+	meta->kfunc_flags = *kfunc_flags;
+	meta->func_proto = func_proto;
+	meta->func_name = func_name;
+
+	return 0;
+}
+
+static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
+			    int *insn_idx_p)
+{
+	const struct btf_type *t, *ptr_type;
+	u32 i, nargs, ptr_type_id, release_ref_obj_id;
+	struct bpf_reg_state *regs = cur_regs(env);
+	const char *func_name, *ptr_type_name;
+	bool sleepable, rcu_lock, rcu_unlock;
+	struct bpf_kfunc_call_arg_meta meta;
+	struct bpf_insn_aux_data *insn_aux;
+	int err, insn_idx = *insn_idx_p;
+	const struct btf_param *args;
+	const struct btf_type *ret_t;
+	struct btf *desc_btf;
+
+	/* skip for now, but return error when we find this in fixup_kfunc_call */
+	if (!insn->imm)
+		return 0;
+
+	err = fetch_kfunc_meta(env, insn, &meta, &func_name);
+	if (err == -EACCES && func_name)
+		verbose(env, "calling kernel function %s is not allowed\n", func_name);
+	if (err)
+		return err;
+	desc_btf = meta.btf;
+	insn_aux = &env->insn_aux_data[insn_idx];
 
 	if (is_kfunc_destructive(&meta) && !capable(CAP_SYS_BOOT)) {
 		verbose(env, "destructive kfunc calls require CAP_SYS_BOOT capability\n");
@@ -10173,7 +10199,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		err = release_reference(env, regs[meta.release_regno].ref_obj_id);
 		if (err) {
 			verbose(env, "kfunc %s#%d reference has not been acquired before\n",
-				func_name, func_id);
+				func_name, meta.func_id);
 			return err;
 		}
 	}
@@ -10185,14 +10211,14 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		err = ref_convert_owning_non_owning(env, release_ref_obj_id);
 		if (err) {
 			verbose(env, "kfunc %s#%d conversion of owning ref to non-owning failed\n",
-				func_name, func_id);
+				func_name, meta.func_id);
 			return err;
 		}
 
 		err = release_reference(env, release_ref_obj_id);
 		if (err) {
 			verbose(env, "kfunc %s#%d reference has not been acquired before\n",
-				func_name, func_id);
+				func_name, meta.func_id);
 			return err;
 		}
 	}
@@ -10202,7 +10228,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 					set_rbtree_add_callback_state);
 		if (err) {
 			verbose(env, "kfunc %s#%d failed callback verification\n",
-				func_name, func_id);
+				func_name, meta.func_id);
 			return err;
 		}
 	}
@@ -10211,7 +10237,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		mark_reg_not_init(env, regs, caller_saved[i]);
 
 	/* Check return type */
-	t = btf_type_skip_modifiers(desc_btf, func_proto->type, NULL);
+	t = btf_type_skip_modifiers(desc_btf, meta.func_proto->type, NULL);
 
 	if (is_kfunc_acquire(&meta) && !btf_type_is_struct_ptr(meta.btf, t)) {
 		/* Only exception is bpf_obj_new_impl */
@@ -10260,11 +10286,11 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 				regs[BPF_REG_0].btf = ret_btf;
 				regs[BPF_REG_0].btf_id = ret_btf_id;
 
-				env->insn_aux_data[insn_idx].obj_new_size = ret_t->size;
-				env->insn_aux_data[insn_idx].kptr_struct_meta =
+				insn_aux->obj_new_size = ret_t->size;
+				insn_aux->kptr_struct_meta =
 					btf_find_struct_meta(ret_btf, ret_btf_id);
 			} else if (meta.func_id == special_kfunc_list[KF_bpf_obj_drop_impl]) {
-				env->insn_aux_data[insn_idx].kptr_struct_meta =
+				insn_aux->kptr_struct_meta =
 					btf_find_struct_meta(meta.arg_obj_drop.btf,
 							     meta.arg_obj_drop.btf_id);
 			} else if (meta.func_id == special_kfunc_list[KF_bpf_list_pop_front] ||
@@ -10397,8 +10423,8 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			regs[BPF_REG_0].id = ++env->id_gen;
 	} /* else { add_kfunc_call() ensures it is btf_type_is_void(t) } */
 
-	nargs = btf_type_vlen(func_proto);
-	args = (const struct btf_param *)(func_proto + 1);
+	nargs = btf_type_vlen(meta.func_proto);
+	args = (const struct btf_param *)(meta.func_proto + 1);
 	for (i = 0; i < nargs; i++) {
 		u32 regno = i + 1;
 
-- 
cgit v1.2.3-70-g09d2


From 215bf4962f6c9605710012fad222a5fec001b3ad Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:15 -0800
Subject: bpf: add iterator kfuncs registration and validation logic

Add ability to register kfuncs that implement BPF open-coded iterator
contract and enforce naming and function proto convention. Enforcement
happens at the time of kfunc registration and significantly simplifies
the rest of iterators logic in the verifier.

More details follow in subsequent patches, but we enforce the following
conditions.

All kfuncs (constructor, next, destructor) have to be named consistenly
as bpf_iter_<type>_{new,next,destroy}(), respectively. <type> represents
iterator type, and iterator state should be represented as a matching
`struct bpf_iter_<type>` state type. Also, all iter kfuncs should have
a pointer to this `struct bpf_iter_<type>` as the very first argument.

Additionally:
  - Constructor, i.e., bpf_iter_<type>_new(), can have arbitrary extra
  number of arguments. Return type is not enforced either.
  - Next method, i.e., bpf_iter_<type>_next(), has to return a pointer
  type and should have exactly one argument: `struct bpf_iter_<type> *`
  (const/volatile/restrict and typedefs are ignored).
  - Destructor, i.e., bpf_iter_<type>_destroy(), should return void and
  should have exactly one argument, similar to the next method.
  - struct bpf_iter_<type> size is enforced to be positive and
  a multiple of 8 bytes (to fit stack slots correctly).

Such strictness and consistency allows to build generic helpers
abstracting important, but boilerplate, details to be able to use
open-coded iterators effectively and ergonomically (see bpf_for_each()
in subsequent patches). It also simplifies the verifier logic in some
places. At the same time, this doesn't hurt generality of possible
iterator implementations. Win-win.

Constructor kfunc is marked with a new KF_ITER_NEW flags, next method is
marked with KF_ITER_NEXT (and should also have KF_RET_NULL, of course),
while destructor kfunc is marked as KF_ITER_DESTROY.

Additionally, we add a trivial kfunc name validation: it should be
a valid non-NULL and non-empty string.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_verifier.h |   2 +
 include/linux/btf.h          |   4 ++
 kernel/bpf/btf.c             | 112 ++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 117 insertions(+), 1 deletion(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 18538bad2b8c..e2dc7f064449 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -59,6 +59,8 @@ struct bpf_active_lock {
 	u32 id;
 };
 
+#define ITER_PREFIX "bpf_iter_"
+
 struct bpf_reg_state {
 	/* Ordering of fields matters.  See states_equal() */
 	enum bpf_reg_type type;
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 556b3e2e7471..1bba0827e8c4 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -71,6 +71,10 @@
 #define KF_SLEEPABLE    (1 << 5) /* kfunc may sleep */
 #define KF_DESTRUCTIVE  (1 << 6) /* kfunc performs destructive actions */
 #define KF_RCU          (1 << 7) /* kfunc takes either rcu or trusted pointer arguments */
+/* only one of KF_ITER_{NEW,NEXT,DESTROY} could be specified per kfunc */
+#define KF_ITER_NEW     (1 << 8) /* kfunc implements BPF iter constructor */
+#define KF_ITER_NEXT    (1 << 9) /* kfunc implements BPF iter next method */
+#define KF_ITER_DESTROY (1 << 10) /* kfunc implements BPF iter destructor */
 
 /*
  * Tag marking a kernel function as a kfunc. This is meant to minimize the
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index a8cb09e5973b..71758cd15b07 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -7596,6 +7596,108 @@ BTF_ID_LIST_GLOBAL(btf_tracing_ids, MAX_BTF_TRACING_TYPE)
 BTF_TRACING_TYPE_xxx
 #undef BTF_TRACING_TYPE
 
+static int btf_check_iter_kfuncs(struct btf *btf, const char *func_name,
+				 const struct btf_type *func, u32 func_flags)
+{
+	u32 flags = func_flags & (KF_ITER_NEW | KF_ITER_NEXT | KF_ITER_DESTROY);
+	const char *name, *sfx, *iter_name;
+	const struct btf_param *arg;
+	const struct btf_type *t;
+	char exp_name[128];
+	u32 nr_args;
+
+	/* exactly one of KF_ITER_{NEW,NEXT,DESTROY} can be set */
+	if (!flags || (flags & (flags - 1)))
+		return -EINVAL;
+
+	/* any BPF iter kfunc should have `struct bpf_iter_<type> *` first arg */
+	nr_args = btf_type_vlen(func);
+	if (nr_args < 1)
+		return -EINVAL;
+
+	arg = &btf_params(func)[0];
+	t = btf_type_skip_modifiers(btf, arg->type, NULL);
+	if (!t || !btf_type_is_ptr(t))
+		return -EINVAL;
+	t = btf_type_skip_modifiers(btf, t->type, NULL);
+	if (!t || !__btf_type_is_struct(t))
+		return -EINVAL;
+
+	name = btf_name_by_offset(btf, t->name_off);
+	if (!name || strncmp(name, ITER_PREFIX, sizeof(ITER_PREFIX) - 1))
+		return -EINVAL;
+
+	/* sizeof(struct bpf_iter_<type>) should be a multiple of 8 to
+	 * fit nicely in stack slots
+	 */
+	if (t->size == 0 || (t->size % 8))
+		return -EINVAL;
+
+	/* validate bpf_iter_<type>_{new,next,destroy}(struct bpf_iter_<type> *)
+	 * naming pattern
+	 */
+	iter_name = name + sizeof(ITER_PREFIX) - 1;
+	if (flags & KF_ITER_NEW)
+		sfx = "new";
+	else if (flags & KF_ITER_NEXT)
+		sfx = "next";
+	else /* (flags & KF_ITER_DESTROY) */
+		sfx = "destroy";
+
+	snprintf(exp_name, sizeof(exp_name), "bpf_iter_%s_%s", iter_name, sfx);
+	if (strcmp(func_name, exp_name))
+		return -EINVAL;
+
+	/* only iter constructor should have extra arguments */
+	if (!(flags & KF_ITER_NEW) && nr_args != 1)
+		return -EINVAL;
+
+	if (flags & KF_ITER_NEXT) {
+		/* bpf_iter_<type>_next() should return pointer */
+		t = btf_type_skip_modifiers(btf, func->type, NULL);
+		if (!t || !btf_type_is_ptr(t))
+			return -EINVAL;
+	}
+
+	if (flags & KF_ITER_DESTROY) {
+		/* bpf_iter_<type>_destroy() should return void */
+		t = btf_type_by_id(btf, func->type);
+		if (!t || !btf_type_is_void(t))
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int btf_check_kfunc_protos(struct btf *btf, u32 func_id, u32 func_flags)
+{
+	const struct btf_type *func;
+	const char *func_name;
+	int err;
+
+	/* any kfunc should be FUNC -> FUNC_PROTO */
+	func = btf_type_by_id(btf, func_id);
+	if (!func || !btf_type_is_func(func))
+		return -EINVAL;
+
+	/* sanity check kfunc name */
+	func_name = btf_name_by_offset(btf, func->name_off);
+	if (!func_name || !func_name[0])
+		return -EINVAL;
+
+	func = btf_type_by_id(btf, func->type);
+	if (!func || !btf_type_is_func_proto(func))
+		return -EINVAL;
+
+	if (func_flags & (KF_ITER_NEW | KF_ITER_NEXT | KF_ITER_DESTROY)) {
+		err = btf_check_iter_kfuncs(btf, func_name, func, func_flags);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 /* Kernel Function (kfunc) BTF ID set registration API */
 
 static int btf_populate_kfunc_set(struct btf *btf, enum btf_kfunc_hook hook,
@@ -7772,7 +7874,7 @@ static int __register_btf_kfunc_id_set(enum btf_kfunc_hook hook,
 				       const struct btf_kfunc_id_set *kset)
 {
 	struct btf *btf;
-	int ret;
+	int ret, i;
 
 	btf = btf_get_module_btf(kset->owner);
 	if (!btf) {
@@ -7789,7 +7891,15 @@ static int __register_btf_kfunc_id_set(enum btf_kfunc_hook hook,
 	if (IS_ERR(btf))
 		return PTR_ERR(btf);
 
+	for (i = 0; i < kset->set->cnt; i++) {
+		ret = btf_check_kfunc_protos(btf, kset->set->pairs[i].id,
+					     kset->set->pairs[i].flags);
+		if (ret)
+			goto err_out;
+	}
+
 	ret = btf_populate_kfunc_set(btf, hook, kset->set);
+err_out:
 	btf_put(btf);
 	return ret;
 }
-- 
cgit v1.2.3-70-g09d2


From 06accc8779c1d558a5b5a21f2ac82b0c95827ddd Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:16 -0800
Subject: bpf: add support for open-coded iterator loops

Teach verifier about the concept of the open-coded (or inline) iterators.

This patch adds generic iterator loop verification logic, new STACK_ITER
stack slot type to contain iterator state, and necessary kfunc plumbing
for iterator's constructor, destructor and next methods. Next patch
implements first specific iterator (numbers iterator for implementing
for() loop logic). Such split allows to have more focused commits for
verifier logic and separate commit that we could point later to
demonstrating  what does it take to add a new kind of iterator.

Each kind of iterator has its own associated struct bpf_iter_<type>,
where <type> denotes a specific type of iterator. struct bpf_iter_<type>
state is supposed to live on BPF program stack, so there will be no way
to change its size later on without breaking backwards compatibility, so
choose wisely! But given this struct is specific to a given <type> of
iterator, this allows a lot of flexibility: simple iterators could be
fine with just one stack slot (8 bytes), like numbers iterator in the
next patch, while some other more complicated iterators might need way
more to keep their iterator state. Either way, such design allows to
avoid runtime memory allocations, which otherwise would be necessary if
we fixed on-the-stack size and it turned out to be too small for a given
iterator implementation.

The way BPF verifier logic is implemented, there are no artificial
restrictions on a number of active iterators, it should work correctly
using multiple active iterators at the same time. This also means you
can have multiple nested iteration loops. struct bpf_iter_<type>
reference can be safely passed to subprograms as well.

General flow is easiest to demonstrate with a simple example using
number iterator implemented in next patch. Here's the simplest possible
loop:

  struct bpf_iter_num it;
  int *v;

  bpf_iter_num_new(&it, 2, 5);
  while ((v = bpf_iter_num_next(&it))) {
      bpf_printk("X = %d", *v);
  }
  bpf_iter_num_destroy(&it);

Above snippet should output "X = 2", "X = 3", "X = 4". Note that 5 is
exclusive and is not returned. This matches similar APIs (e.g., slices
in Go or Rust) that implement a range of elements, where end index is
non-inclusive.

In the above example, we see a trio of function:
  - constructor, bpf_iter_num_new(), which initializes iterator state
  (struct bpf_iter_num it) on the stack. If any of the input arguments
  are invalid, constructor should make sure to still initialize it such
  that subsequent bpf_iter_num_next() calls will return NULL. I.e., on
  error, return error and construct empty iterator.
  - next method, bpf_iter_num_next(), which accepts pointer to iterator
  state and produces an element. Next method should always return
  a pointer. The contract between BPF verifier is that next method will
  always eventually return NULL when elements are exhausted. Once NULL is
  returned, subsequent next calls should keep returning NULL. In the
  case of numbers iterator, bpf_iter_num_next() returns a pointer to an int
  (storage for this integer is inside the iterator state itself),
  which can be dereferenced after corresponding NULL check.
  - once done with the iterator, it's mandated that user cleans up its
  state with the call to destructor, bpf_iter_num_destroy() in this
  case. Destructor frees up any resources and marks stack space used by
  struct bpf_iter_num as usable for something else.

Any other iterator implementation will have to implement at least these
three methods. It is enforced that for any given type of iterator only
applicable constructor/destructor/next are callable. I.e., verifier
ensures you can't pass number iterator state into, say, cgroup
iterator's next method.

It is important to keep the naming pattern consistent to be able to
create generic macros to help with BPF iter usability. E.g., one
of the follow up patches adds generic bpf_for_each() macro to bpf_misc.h
in selftests, which allows to utilize iterator "trio" nicely without
having to code the above somewhat tedious loop explicitly every time.
This is enforced at kfunc registration point by one of the previous
patches in this series.

At the implementation level, iterator state tracking for verification
purposes is very similar to dynptr. We add STACK_ITER stack slot type,
reserve necessary number of slots, depending on
sizeof(struct bpf_iter_<type>), and keep track of necessary extra state
in the "main" slot, which is marked with non-zero ref_obj_id. Other
slots are also marked as STACK_ITER, but have zero ref_obj_id. This is
simpler than having a separate "is_first_slot" flag.

Another big distinction is that STACK_ITER is *always refcounted*, which
simplifies implementation without sacrificing usability. So no need for
extra "iter_id", no need to anticipate reuse of STACK_ITER slots for new
constructors, etc. Keeping it simple here.

As far as the verification logic goes, there are two extensive comments:
in process_iter_next_call() and iter_active_depths_differ() explaining
some important and sometimes subtle aspects. Please refer to them for
details.

But from 10,000-foot point of view, next methods are the points of
forking a verification state, which are conceptually similar to what
verifier is doing when validating conditional jump. We branch out at
a `call bpf_iter_<type>_next` instruction and simulate two outcomes:
NULL (iteration is done) and non-NULL (new element is returned). NULL is
simulated first and is supposed to reach exit without looping. After
that non-NULL case is validated and it either reaches exit (for trivial
examples with no real loop), or reaches another `call bpf_iter_<type>_next`
instruction with the state equivalent to already (partially) validated
one. State equivalency at that point means we technically are going to
be looping forever without "breaking out" out of established "state
envelope" (i.e., subsequent iterations don't add any new knowledge or
constraints to the verifier state, so running 1, 2, 10, or a million of
them doesn't matter). But taking into account the contract stating that
iterator next method *has to* return NULL eventually, we can conclude
that loop body is safe and will eventually terminate. Given we validated
logic outside of the loop (NULL case), and concluded that loop body is
safe (though potentially looping many times), verifier can claim safety
of the overall program logic.

The rest of the patch is necessary plumbing for state tracking, marking,
validation, and necessary further kfunc plumbing to allow implementing
iterator constructor, destructor, and next methods.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_verifier.h |  23 ++
 kernel/bpf/verifier.c        | 595 ++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 610 insertions(+), 8 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index e2dc7f064449..0c052bc79940 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -61,6 +61,12 @@ struct bpf_active_lock {
 
 #define ITER_PREFIX "bpf_iter_"
 
+enum bpf_iter_state {
+	BPF_ITER_STATE_INVALID, /* for non-first slot */
+	BPF_ITER_STATE_ACTIVE,
+	BPF_ITER_STATE_DRAINED,
+};
+
 struct bpf_reg_state {
 	/* Ordering of fields matters.  See states_equal() */
 	enum bpf_reg_type type;
@@ -105,6 +111,18 @@ struct bpf_reg_state {
 			bool first_slot;
 		} dynptr;
 
+		/* For bpf_iter stack slots */
+		struct {
+			/* BTF container and BTF type ID describing
+			 * struct bpf_iter_<type> of an iterator state
+			 */
+			struct btf *btf;
+			u32 btf_id;
+			/* packing following two fields to fit iter state into 16 bytes */
+			enum bpf_iter_state state:2;
+			int depth:30;
+		} iter;
+
 		/* Max size from any of the above. */
 		struct {
 			unsigned long raw1;
@@ -143,6 +161,8 @@ struct bpf_reg_state {
 	 * same reference to the socket, to determine proper reference freeing.
 	 * For stack slots that are dynptrs, this is used to track references to
 	 * the dynptr to determine proper reference freeing.
+	 * Similarly to dynptrs, we use ID to track "belonging" of a reference
+	 * to a specific instance of bpf_iter.
 	 */
 	u32 id;
 	/* PTR_TO_SOCKET and PTR_TO_TCP_SOCK could be a ptr returned
@@ -213,9 +233,11 @@ enum bpf_stack_slot_type {
 	 * is stored in bpf_stack_state->spilled_ptr.dynptr.type
 	 */
 	STACK_DYNPTR,
+	STACK_ITER,
 };
 
 #define BPF_REG_SIZE 8	/* size of eBPF register in bytes */
+
 #define BPF_DYNPTR_SIZE		sizeof(struct bpf_dynptr_kern)
 #define BPF_DYNPTR_NR_SLOTS		(BPF_DYNPTR_SIZE / BPF_REG_SIZE)
 
@@ -450,6 +472,7 @@ struct bpf_insn_aux_data {
 	bool sanitize_stack_spill; /* subject to Spectre v4 sanitation */
 	bool zext_dst; /* this insn zero extends dst reg */
 	bool storage_get_func_atomic; /* bpf_*_storage_get() with atomic memory alloc */
+	bool is_iter_next; /* bpf_iter_<type>_next() kfunc call */
 	u8 alu_state; /* used in combination with alu_limit */
 
 	/* below fields are initialized once */
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 8d40fba6a1c0..45a082284464 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -302,6 +302,10 @@ struct bpf_kfunc_call_arg_meta {
 		enum bpf_dynptr_type type;
 		u32 id;
 	} initialized_dynptr;
+	struct {
+		u8 spi;
+		u8 frameno;
+	} iter;
 	u64 mem_size;
 };
 
@@ -668,6 +672,7 @@ static char slot_type_char[] = {
 	[STACK_MISC]	= 'm',
 	[STACK_ZERO]	= '0',
 	[STACK_DYNPTR]	= 'd',
+	[STACK_ITER]	= 'i',
 };
 
 static void print_liveness(struct bpf_verifier_env *env,
@@ -742,6 +747,11 @@ static int dynptr_get_spi(struct bpf_verifier_env *env, struct bpf_reg_state *re
 	return stack_slot_obj_get_spi(env, reg, "dynptr", BPF_DYNPTR_NR_SLOTS);
 }
 
+static int iter_get_spi(struct bpf_verifier_env *env, struct bpf_reg_state *reg, int nr_slots)
+{
+	return stack_slot_obj_get_spi(env, reg, "iter", nr_slots);
+}
+
 static const char *kernel_type_name(const struct btf* btf, u32 id)
 {
 	return btf_name_by_offset(btf, btf_type_by_id(btf, id)->name_off);
@@ -766,6 +776,30 @@ static const char *dynptr_type_str(enum bpf_dynptr_type type)
 	}
 }
 
+static const char *iter_type_str(const struct btf *btf, u32 btf_id)
+{
+	if (!btf || btf_id == 0)
+		return "<invalid>";
+
+	/* we already validated that type is valid and has conforming name */
+	return kernel_type_name(btf, btf_id) + sizeof(ITER_PREFIX) - 1;
+}
+
+static const char *iter_state_str(enum bpf_iter_state state)
+{
+	switch (state) {
+	case BPF_ITER_STATE_ACTIVE:
+		return "active";
+	case BPF_ITER_STATE_DRAINED:
+		return "drained";
+	case BPF_ITER_STATE_INVALID:
+		return "<invalid>";
+	default:
+		WARN_ONCE(1, "unknown iter state %d\n", state);
+		return "<unknown>";
+	}
+}
+
 static void mark_reg_scratched(struct bpf_verifier_env *env, u32 regno)
 {
 	env->scratched_regs |= 1U << regno;
@@ -1118,6 +1152,157 @@ static bool is_dynptr_type_expected(struct bpf_verifier_env *env, struct bpf_reg
 	}
 }
 
+static void __mark_reg_known_zero(struct bpf_reg_state *reg);
+
+static int mark_stack_slots_iter(struct bpf_verifier_env *env,
+				 struct bpf_reg_state *reg, int insn_idx,
+				 struct btf *btf, u32 btf_id, int nr_slots)
+{
+	struct bpf_func_state *state = func(env, reg);
+	int spi, i, j, id;
+
+	spi = iter_get_spi(env, reg, nr_slots);
+	if (spi < 0)
+		return spi;
+
+	id = acquire_reference_state(env, insn_idx);
+	if (id < 0)
+		return id;
+
+	for (i = 0; i < nr_slots; i++) {
+		struct bpf_stack_state *slot = &state->stack[spi - i];
+		struct bpf_reg_state *st = &slot->spilled_ptr;
+
+		__mark_reg_known_zero(st);
+		st->type = PTR_TO_STACK; /* we don't have dedicated reg type */
+		st->live |= REG_LIVE_WRITTEN;
+		st->ref_obj_id = i == 0 ? id : 0;
+		st->iter.btf = btf;
+		st->iter.btf_id = btf_id;
+		st->iter.state = BPF_ITER_STATE_ACTIVE;
+		st->iter.depth = 0;
+
+		for (j = 0; j < BPF_REG_SIZE; j++)
+			slot->slot_type[j] = STACK_ITER;
+
+		mark_stack_slot_scratched(env, spi - i);
+	}
+
+	return 0;
+}
+
+static int unmark_stack_slots_iter(struct bpf_verifier_env *env,
+				   struct bpf_reg_state *reg, int nr_slots)
+{
+	struct bpf_func_state *state = func(env, reg);
+	int spi, i, j;
+
+	spi = iter_get_spi(env, reg, nr_slots);
+	if (spi < 0)
+		return spi;
+
+	for (i = 0; i < nr_slots; i++) {
+		struct bpf_stack_state *slot = &state->stack[spi - i];
+		struct bpf_reg_state *st = &slot->spilled_ptr;
+
+		if (i == 0)
+			WARN_ON_ONCE(release_reference(env, st->ref_obj_id));
+
+		__mark_reg_not_init(env, st);
+
+		/* see unmark_stack_slots_dynptr() for why we need to set REG_LIVE_WRITTEN */
+		st->live |= REG_LIVE_WRITTEN;
+
+		for (j = 0; j < BPF_REG_SIZE; j++)
+			slot->slot_type[j] = STACK_INVALID;
+
+		mark_stack_slot_scratched(env, spi - i);
+	}
+
+	return 0;
+}
+
+static bool is_iter_reg_valid_uninit(struct bpf_verifier_env *env,
+				     struct bpf_reg_state *reg, int nr_slots)
+{
+	struct bpf_func_state *state = func(env, reg);
+	int spi, i, j;
+
+	/* For -ERANGE (i.e. spi not falling into allocated stack slots), we
+	 * will do check_mem_access to check and update stack bounds later, so
+	 * return true for that case.
+	 */
+	spi = iter_get_spi(env, reg, nr_slots);
+	if (spi == -ERANGE)
+		return true;
+	if (spi < 0)
+		return false;
+
+	for (i = 0; i < nr_slots; i++) {
+		struct bpf_stack_state *slot = &state->stack[spi - i];
+
+		for (j = 0; j < BPF_REG_SIZE; j++)
+			if (slot->slot_type[j] == STACK_ITER)
+				return false;
+	}
+
+	return true;
+}
+
+static bool is_iter_reg_valid_init(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+				   struct btf *btf, u32 btf_id, int nr_slots)
+{
+	struct bpf_func_state *state = func(env, reg);
+	int spi, i, j;
+
+	spi = iter_get_spi(env, reg, nr_slots);
+	if (spi < 0)
+		return false;
+
+	for (i = 0; i < nr_slots; i++) {
+		struct bpf_stack_state *slot = &state->stack[spi - i];
+		struct bpf_reg_state *st = &slot->spilled_ptr;
+
+		/* only main (first) slot has ref_obj_id set */
+		if (i == 0 && !st->ref_obj_id)
+			return false;
+		if (i != 0 && st->ref_obj_id)
+			return false;
+		if (st->iter.btf != btf || st->iter.btf_id != btf_id)
+			return false;
+
+		for (j = 0; j < BPF_REG_SIZE; j++)
+			if (slot->slot_type[j] != STACK_ITER)
+				return false;
+	}
+
+	return true;
+}
+
+/* Check if given stack slot is "special":
+ *   - spilled register state (STACK_SPILL);
+ *   - dynptr state (STACK_DYNPTR);
+ *   - iter state (STACK_ITER).
+ */
+static bool is_stack_slot_special(const struct bpf_stack_state *stack)
+{
+	enum bpf_stack_slot_type type = stack->slot_type[BPF_REG_SIZE - 1];
+
+	switch (type) {
+	case STACK_SPILL:
+	case STACK_DYNPTR:
+	case STACK_ITER:
+		return true;
+	case STACK_INVALID:
+	case STACK_MISC:
+	case STACK_ZERO:
+		return false;
+	default:
+		WARN_ONCE(1, "unknown stack slot type %d\n", type);
+		return true;
+	}
+}
+
 /* The reg state of a pointer or a bounded scalar was saved when
  * it was spilled to the stack.
  */
@@ -1267,6 +1452,19 @@ static void print_verifier_state(struct bpf_verifier_env *env,
 			if (reg->ref_obj_id)
 				verbose(env, "(ref_id=%d)", reg->ref_obj_id);
 			break;
+		case STACK_ITER:
+			/* only main slot has ref_obj_id set; skip others */
+			reg = &state->stack[i].spilled_ptr;
+			if (!reg->ref_obj_id)
+				continue;
+
+			verbose(env, " fp%d", (-i - 1) * BPF_REG_SIZE);
+			print_liveness(env, reg->live);
+			verbose(env, "=iter_%s(ref_id=%d,state=%s,depth=%u)",
+				iter_type_str(reg->iter.btf, reg->iter.btf_id),
+				reg->ref_obj_id, iter_state_str(reg->iter.state),
+				reg->iter.depth);
+			break;
 		case STACK_MISC:
 		case STACK_ZERO:
 		default:
@@ -2710,6 +2908,25 @@ static int mark_dynptr_read(struct bpf_verifier_env *env, struct bpf_reg_state *
 			     state->stack[spi - 1].spilled_ptr.parent, REG_LIVE_READ64);
 }
 
+static int mark_iter_read(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+			  int spi, int nr_slots)
+{
+	struct bpf_func_state *state = func(env, reg);
+	int err, i;
+
+	for (i = 0; i < nr_slots; i++) {
+		struct bpf_reg_state *st = &state->stack[spi - i].spilled_ptr;
+
+		err = mark_reg_read(env, st, st->parent, REG_LIVE_READ64);
+		if (err)
+			return err;
+
+		mark_stack_slot_scratched(env, spi - i);
+	}
+
+	return 0;
+}
+
 /* This function is supposed to be used by the following 32-bit optimization
  * code only. It returns TRUE if the source or destination register operates
  * on 64-bit, otherwise return FALSE.
@@ -3691,8 +3908,8 @@ static int check_stack_write_fixed_off(struct bpf_verifier_env *env,
 
 		/* regular write of data into stack destroys any spilled ptr */
 		state->stack[spi].spilled_ptr.type = NOT_INIT;
-		/* Mark slots as STACK_MISC if they belonged to spilled ptr. */
-		if (is_spilled_reg(&state->stack[spi]))
+		/* Mark slots as STACK_MISC if they belonged to spilled ptr/dynptr/iter. */
+		if (is_stack_slot_special(&state->stack[spi]))
 			for (i = 0; i < BPF_REG_SIZE; i++)
 				scrub_spilled_slot(&state->stack[spi].slot_type[i]);
 
@@ -6506,6 +6723,203 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno, int insn
 	return err;
 }
 
+static u32 iter_ref_obj_id(struct bpf_verifier_env *env, struct bpf_reg_state *reg, int spi)
+{
+	struct bpf_func_state *state = func(env, reg);
+
+	return state->stack[spi].spilled_ptr.ref_obj_id;
+}
+
+static bool is_iter_kfunc(struct bpf_kfunc_call_arg_meta *meta)
+{
+	return meta->kfunc_flags & (KF_ITER_NEW | KF_ITER_NEXT | KF_ITER_DESTROY);
+}
+
+static bool is_iter_new_kfunc(struct bpf_kfunc_call_arg_meta *meta)
+{
+	return meta->kfunc_flags & KF_ITER_NEW;
+}
+
+static bool is_iter_next_kfunc(struct bpf_kfunc_call_arg_meta *meta)
+{
+	return meta->kfunc_flags & KF_ITER_NEXT;
+}
+
+static bool is_iter_destroy_kfunc(struct bpf_kfunc_call_arg_meta *meta)
+{
+	return meta->kfunc_flags & KF_ITER_DESTROY;
+}
+
+static bool is_kfunc_arg_iter(struct bpf_kfunc_call_arg_meta *meta, int arg)
+{
+	/* btf_check_iter_kfuncs() guarantees that first argument of any iter
+	 * kfunc is iter state pointer
+	 */
+	return arg == 0 && is_iter_kfunc(meta);
+}
+
+static int process_iter_arg(struct bpf_verifier_env *env, int regno, int insn_idx,
+			    struct bpf_kfunc_call_arg_meta *meta)
+{
+	struct bpf_reg_state *regs = cur_regs(env), *reg = &regs[regno];
+	const struct btf_type *t;
+	const struct btf_param *arg;
+	int spi, err, i, nr_slots;
+	u32 btf_id;
+
+	/* btf_check_iter_kfuncs() ensures we don't need to validate anything here */
+	arg = &btf_params(meta->func_proto)[0];
+	t = btf_type_skip_modifiers(meta->btf, arg->type, NULL);	/* PTR */
+	t = btf_type_skip_modifiers(meta->btf, t->type, &btf_id);	/* STRUCT */
+	nr_slots = t->size / BPF_REG_SIZE;
+
+	spi = iter_get_spi(env, reg, nr_slots);
+	if (spi < 0 && spi != -ERANGE)
+		return spi;
+
+	meta->iter.spi = spi;
+	meta->iter.frameno = reg->frameno;
+
+	if (is_iter_new_kfunc(meta)) {
+		/* bpf_iter_<type>_new() expects pointer to uninit iter state */
+		if (!is_iter_reg_valid_uninit(env, reg, nr_slots)) {
+			verbose(env, "expected uninitialized iter_%s as arg #%d\n",
+				iter_type_str(meta->btf, btf_id), regno);
+			return -EINVAL;
+		}
+
+		for (i = 0; i < nr_slots * 8; i += BPF_REG_SIZE) {
+			err = check_mem_access(env, insn_idx, regno,
+					       i, BPF_DW, BPF_WRITE, -1, false);
+			if (err)
+				return err;
+		}
+
+		err = mark_stack_slots_iter(env, reg, insn_idx, meta->btf, btf_id, nr_slots);
+		if (err)
+			return err;
+	} else {
+		/* iter_next() or iter_destroy() expect initialized iter state*/
+		if (!is_iter_reg_valid_init(env, reg, meta->btf, btf_id, nr_slots)) {
+			verbose(env, "expected an initialized iter_%s as arg #%d\n",
+				iter_type_str(meta->btf, btf_id), regno);
+			return -EINVAL;
+		}
+
+		err = mark_iter_read(env, reg, spi, nr_slots);
+		if (err)
+			return err;
+
+		meta->ref_obj_id = iter_ref_obj_id(env, reg, spi);
+
+		if (is_iter_destroy_kfunc(meta)) {
+			err = unmark_stack_slots_iter(env, reg, nr_slots);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+/* process_iter_next_call() is called when verifier gets to iterator's next
+ * "method" (e.g., bpf_iter_num_next() for numbers iterator) call. We'll refer
+ * to it as just "iter_next()" in comments below.
+ *
+ * BPF verifier relies on a crucial contract for any iter_next()
+ * implementation: it should *eventually* return NULL, and once that happens
+ * it should keep returning NULL. That is, once iterator exhausts elements to
+ * iterate, it should never reset or spuriously return new elements.
+ *
+ * With the assumption of such contract, process_iter_next_call() simulates
+ * a fork in the verifier state to validate loop logic correctness and safety
+ * without having to simulate infinite amount of iterations.
+ *
+ * In current state, we first assume that iter_next() returned NULL and
+ * iterator state is set to DRAINED (BPF_ITER_STATE_DRAINED). In such
+ * conditions we should not form an infinite loop and should eventually reach
+ * exit.
+ *
+ * Besides that, we also fork current state and enqueue it for later
+ * verification. In a forked state we keep iterator state as ACTIVE
+ * (BPF_ITER_STATE_ACTIVE) and assume non-NULL return from iter_next(). We
+ * also bump iteration depth to prevent erroneous infinite loop detection
+ * later on (see iter_active_depths_differ() comment for details). In this
+ * state we assume that we'll eventually loop back to another iter_next()
+ * calls (it could be in exactly same location or in some other instruction,
+ * it doesn't matter, we don't make any unnecessary assumptions about this,
+ * everything revolves around iterator state in a stack slot, not which
+ * instruction is calling iter_next()). When that happens, we either will come
+ * to iter_next() with equivalent state and can conclude that next iteration
+ * will proceed in exactly the same way as we just verified, so it's safe to
+ * assume that loop converges. If not, we'll go on another iteration
+ * simulation with a different input state, until all possible starting states
+ * are validated or we reach maximum number of instructions limit.
+ *
+ * This way, we will either exhaustively discover all possible input states
+ * that iterator loop can start with and eventually will converge, or we'll
+ * effectively regress into bounded loop simulation logic and either reach
+ * maximum number of instructions if loop is not provably convergent, or there
+ * is some statically known limit on number of iterations (e.g., if there is
+ * an explicit `if n > 100 then break;` statement somewhere in the loop).
+ *
+ * One very subtle but very important aspect is that we *always* simulate NULL
+ * condition first (as the current state) before we simulate non-NULL case.
+ * This has to do with intricacies of scalar precision tracking. By simulating
+ * "exit condition" of iter_next() returning NULL first, we make sure all the
+ * relevant precision marks *that will be set **after** we exit iterator loop*
+ * are propagated backwards to common parent state of NULL and non-NULL
+ * branches. Thanks to that, state equivalence checks done later in forked
+ * state, when reaching iter_next() for ACTIVE iterator, can assume that
+ * precision marks are finalized and won't change. Because simulating another
+ * ACTIVE iterator iteration won't change them (because given same input
+ * states we'll end up with exactly same output states which we are currently
+ * comparing; and verification after the loop already propagated back what
+ * needs to be **additionally** tracked as precise). It's subtle, grok
+ * precision tracking for more intuitive understanding.
+ */
+static int process_iter_next_call(struct bpf_verifier_env *env, int insn_idx,
+				  struct bpf_kfunc_call_arg_meta *meta)
+{
+	struct bpf_verifier_state *cur_st = env->cur_state, *queued_st;
+	struct bpf_func_state *cur_fr = cur_st->frame[cur_st->curframe], *queued_fr;
+	struct bpf_reg_state *cur_iter, *queued_iter;
+	int iter_frameno = meta->iter.frameno;
+	int iter_spi = meta->iter.spi;
+
+	BTF_TYPE_EMIT(struct bpf_iter);
+
+	cur_iter = &env->cur_state->frame[iter_frameno]->stack[iter_spi].spilled_ptr;
+
+	if (cur_iter->iter.state != BPF_ITER_STATE_ACTIVE &&
+	    cur_iter->iter.state != BPF_ITER_STATE_DRAINED) {
+		verbose(env, "verifier internal error: unexpected iterator state %d (%s)\n",
+			cur_iter->iter.state, iter_state_str(cur_iter->iter.state));
+		return -EFAULT;
+	}
+
+	if (cur_iter->iter.state == BPF_ITER_STATE_ACTIVE) {
+		/* branch out active iter state */
+		queued_st = push_stack(env, insn_idx + 1, insn_idx, false);
+		if (!queued_st)
+			return -ENOMEM;
+
+		queued_iter = &queued_st->frame[iter_frameno]->stack[iter_spi].spilled_ptr;
+		queued_iter->iter.state = BPF_ITER_STATE_ACTIVE;
+		queued_iter->iter.depth++;
+
+		queued_fr = queued_st->frame[queued_st->curframe];
+		mark_ptr_not_null_reg(&queued_fr->regs[BPF_REG_0]);
+	}
+
+	/* switch to DRAINED state, but keep the depth unchanged */
+	/* mark current iter state as drained and assume returned NULL */
+	cur_iter->iter.state = BPF_ITER_STATE_DRAINED;
+	__mark_reg_const_zero(&cur_fr->regs[BPF_REG_0]);
+
+	return 0;
+}
+
 static bool arg_type_is_mem_size(enum bpf_arg_type type)
 {
 	return type == ARG_CONST_SIZE ||
@@ -9099,6 +9513,7 @@ enum kfunc_ptr_arg_type {
 	KF_ARG_PTR_TO_ALLOC_BTF_ID,  /* Allocated object */
 	KF_ARG_PTR_TO_KPTR,	     /* PTR_TO_KPTR but type specific */
 	KF_ARG_PTR_TO_DYNPTR,
+	KF_ARG_PTR_TO_ITER,
 	KF_ARG_PTR_TO_LIST_HEAD,
 	KF_ARG_PTR_TO_LIST_NODE,
 	KF_ARG_PTR_TO_BTF_ID,	     /* Also covers reg2btf_ids conversions */
@@ -9220,6 +9635,9 @@ get_kfunc_ptr_arg_type(struct bpf_verifier_env *env,
 	if (is_kfunc_arg_dynptr(meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_DYNPTR;
 
+	if (is_kfunc_arg_iter(meta, argno))
+		return KF_ARG_PTR_TO_ITER;
+
 	if (is_kfunc_arg_list_head(meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_LIST_HEAD;
 
@@ -9848,6 +10266,7 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 			break;
 		case KF_ARG_PTR_TO_KPTR:
 		case KF_ARG_PTR_TO_DYNPTR:
+		case KF_ARG_PTR_TO_ITER:
 		case KF_ARG_PTR_TO_LIST_HEAD:
 		case KF_ARG_PTR_TO_LIST_NODE:
 		case KF_ARG_PTR_TO_RB_ROOT:
@@ -9944,6 +10363,11 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 
 			break;
 		}
+		case KF_ARG_PTR_TO_ITER:
+			ret = process_iter_arg(env, regno, insn_idx, meta);
+			if (ret < 0)
+				return ret;
+			break;
 		case KF_ARG_PTR_TO_LIST_HEAD:
 			if (reg->type != PTR_TO_MAP_VALUE &&
 			    reg->type != (PTR_TO_BTF_ID | MEM_ALLOC)) {
@@ -10148,6 +10572,8 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 	desc_btf = meta.btf;
 	insn_aux = &env->insn_aux_data[insn_idx];
 
+	insn_aux->is_iter_next = is_iter_next_kfunc(&meta);
+
 	if (is_kfunc_destructive(&meta) && !capable(CAP_SYS_BOOT)) {
 		verbose(env, "destructive kfunc calls require CAP_SYS_BOOT capability\n");
 		return -EACCES;
@@ -10436,6 +10862,12 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			mark_btf_func_reg_size(env, regno, t->size);
 	}
 
+	if (is_iter_next_kfunc(&meta)) {
+		err = process_iter_next_call(env, insn_idx, &meta);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
@@ -13548,6 +13980,13 @@ static int visit_insn(int t, struct bpf_verifier_env *env)
 			 * async state will be pushed for further exploration.
 			 */
 			mark_prune_point(env, t);
+		if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL) {
+			struct bpf_kfunc_call_arg_meta meta;
+
+			ret = fetch_kfunc_meta(env, insn, &meta, NULL);
+			if (ret == 0 && is_iter_next_kfunc(&meta))
+				mark_prune_point(env, t);
+		}
 		return visit_func_call_insn(t, insns, env, insn->src_reg == BPF_PSEUDO_CALL);
 
 	case BPF_JA:
@@ -14301,6 +14740,8 @@ static bool stacksafe(struct bpf_verifier_env *env, struct bpf_func_state *old,
 	 * didn't use them
 	 */
 	for (i = 0; i < old->allocated_stack; i++) {
+		struct bpf_reg_state *old_reg, *cur_reg;
+
 		spi = i / BPF_REG_SIZE;
 
 		if (!(old->stack[spi].spilled_ptr.live & REG_LIVE_READ)) {
@@ -14357,9 +14798,6 @@ static bool stacksafe(struct bpf_verifier_env *env, struct bpf_func_state *old,
 				return false;
 			break;
 		case STACK_DYNPTR:
-		{
-			const struct bpf_reg_state *old_reg, *cur_reg;
-
 			old_reg = &old->stack[spi].spilled_ptr;
 			cur_reg = &cur->stack[spi].spilled_ptr;
 			if (old_reg->dynptr.type != cur_reg->dynptr.type ||
@@ -14367,7 +14805,22 @@ static bool stacksafe(struct bpf_verifier_env *env, struct bpf_func_state *old,
 			    !check_ids(old_reg->ref_obj_id, cur_reg->ref_obj_id, idmap))
 				return false;
 			break;
-		}
+		case STACK_ITER:
+			old_reg = &old->stack[spi].spilled_ptr;
+			cur_reg = &cur->stack[spi].spilled_ptr;
+			/* iter.depth is not compared between states as it
+			 * doesn't matter for correctness and would otherwise
+			 * prevent convergence; we maintain it only to prevent
+			 * infinite loop check triggering, see
+			 * iter_active_depths_differ()
+			 */
+			if (old_reg->iter.btf != cur_reg->iter.btf ||
+			    old_reg->iter.btf_id != cur_reg->iter.btf_id ||
+			    old_reg->iter.state != cur_reg->iter.state ||
+			    /* ignore {old_reg,cur_reg}->iter.depth, see above */
+			    !check_ids(old_reg->ref_obj_id, cur_reg->ref_obj_id, idmap))
+				return false;
+			break;
 		case STACK_MISC:
 		case STACK_ZERO:
 		case STACK_INVALID:
@@ -14626,6 +15079,92 @@ static bool states_maybe_looping(struct bpf_verifier_state *old,
 	return true;
 }
 
+static bool is_iter_next_insn(struct bpf_verifier_env *env, int insn_idx)
+{
+	return env->insn_aux_data[insn_idx].is_iter_next;
+}
+
+/* is_state_visited() handles iter_next() (see process_iter_next_call() for
+ * terminology) calls specially: as opposed to bounded BPF loops, it *expects*
+ * states to match, which otherwise would look like an infinite loop. So while
+ * iter_next() calls are taken care of, we still need to be careful and
+ * prevent erroneous and too eager declaration of "ininite loop", when
+ * iterators are involved.
+ *
+ * Here's a situation in pseudo-BPF assembly form:
+ *
+ *   0: again:                          ; set up iter_next() call args
+ *   1:   r1 = &it                      ; <CHECKPOINT HERE>
+ *   2:   call bpf_iter_num_next        ; this is iter_next() call
+ *   3:   if r0 == 0 goto done
+ *   4:   ... something useful here ...
+ *   5:   goto again                    ; another iteration
+ *   6: done:
+ *   7:   r1 = &it
+ *   8:   call bpf_iter_num_destroy     ; clean up iter state
+ *   9:   exit
+ *
+ * This is a typical loop. Let's assume that we have a prune point at 1:,
+ * before we get to `call bpf_iter_num_next` (e.g., because of that `goto
+ * again`, assuming other heuristics don't get in a way).
+ *
+ * When we first time come to 1:, let's say we have some state X. We proceed
+ * to 2:, fork states, enqueue ACTIVE, validate NULL case successfully, exit.
+ * Now we come back to validate that forked ACTIVE state. We proceed through
+ * 3-5, come to goto, jump to 1:. Let's assume our state didn't change, so we
+ * are converging. But the problem is that we don't know that yet, as this
+ * convergence has to happen at iter_next() call site only. So if nothing is
+ * done, at 1: verifier will use bounded loop logic and declare infinite
+ * looping (and would be *technically* correct, if not for iterator's
+ * "eventual sticky NULL" contract, see process_iter_next_call()). But we
+ * don't want that. So what we do in process_iter_next_call() when we go on
+ * another ACTIVE iteration, we bump slot->iter.depth, to mark that it's
+ * a different iteration. So when we suspect an infinite loop, we additionally
+ * check if any of the *ACTIVE* iterator states depths differ. If yes, we
+ * pretend we are not looping and wait for next iter_next() call.
+ *
+ * This only applies to ACTIVE state. In DRAINED state we don't expect to
+ * loop, because that would actually mean infinite loop, as DRAINED state is
+ * "sticky", and so we'll keep returning into the same instruction with the
+ * same state (at least in one of possible code paths).
+ *
+ * This approach allows to keep infinite loop heuristic even in the face of
+ * active iterator. E.g., C snippet below is and will be detected as
+ * inifintely looping:
+ *
+ *   struct bpf_iter_num it;
+ *   int *p, x;
+ *
+ *   bpf_iter_num_new(&it, 0, 10);
+ *   while ((p = bpf_iter_num_next(&t))) {
+ *       x = p;
+ *       while (x--) {} // <<-- infinite loop here
+ *   }
+ *
+ */
+static bool iter_active_depths_differ(struct bpf_verifier_state *old, struct bpf_verifier_state *cur)
+{
+	struct bpf_reg_state *slot, *cur_slot;
+	struct bpf_func_state *state;
+	int i, fr;
+
+	for (fr = old->curframe; fr >= 0; fr--) {
+		state = old->frame[fr];
+		for (i = 0; i < state->allocated_stack / BPF_REG_SIZE; i++) {
+			if (state->stack[i].slot_type[0] != STACK_ITER)
+				continue;
+
+			slot = &state->stack[i].spilled_ptr;
+			if (slot->iter.state != BPF_ITER_STATE_ACTIVE)
+				continue;
+
+			cur_slot = &cur->frame[fr]->stack[i].spilled_ptr;
+			if (cur_slot->iter.depth != slot->iter.depth)
+				return true;
+		}
+	}
+	return false;
+}
 
 static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 {
@@ -14673,8 +15212,46 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 				 * Since the verifier still needs to catch infinite loops
 				 * inside async callbacks.
 				 */
-			} else if (states_maybe_looping(&sl->state, cur) &&
-				   states_equal(env, &sl->state, cur)) {
+				goto skip_inf_loop_check;
+			}
+			/* BPF open-coded iterators loop detection is special.
+			 * states_maybe_looping() logic is too simplistic in detecting
+			 * states that *might* be equivalent, because it doesn't know
+			 * about ID remapping, so don't even perform it.
+			 * See process_iter_next_call() and iter_active_depths_differ()
+			 * for overview of the logic. When current and one of parent
+			 * states are detected as equivalent, it's a good thing: we prove
+			 * convergence and can stop simulating further iterations.
+			 * It's safe to assume that iterator loop will finish, taking into
+			 * account iter_next() contract of eventually returning
+			 * sticky NULL result.
+			 */
+			if (is_iter_next_insn(env, insn_idx)) {
+				if (states_equal(env, &sl->state, cur)) {
+					struct bpf_func_state *cur_frame;
+					struct bpf_reg_state *iter_state, *iter_reg;
+					int spi;
+
+					cur_frame = cur->frame[cur->curframe];
+					/* btf_check_iter_kfuncs() enforces that
+					 * iter state pointer is always the first arg
+					 */
+					iter_reg = &cur_frame->regs[BPF_REG_1];
+					/* current state is valid due to states_equal(),
+					 * so we can assume valid iter and reg state,
+					 * no need for extra (re-)validations
+					 */
+					spi = __get_spi(iter_reg->off + iter_reg->var_off.value);
+					iter_state = &func(env, iter_reg)->stack[spi].spilled_ptr;
+					if (iter_state->iter.state == BPF_ITER_STATE_ACTIVE)
+						goto hit;
+				}
+				goto skip_inf_loop_check;
+			}
+			/* attempt to detect infinite loop to avoid unnecessary doomed work */
+			if (states_maybe_looping(&sl->state, cur) &&
+			    states_equal(env, &sl->state, cur) &&
+			    !iter_active_depths_differ(&sl->state, cur)) {
 				verbose_linfo(env, insn_idx, "; ");
 				verbose(env, "infinite loop detected at insn %d\n", insn_idx);
 				return -EINVAL;
@@ -14691,6 +15268,7 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 			 * This threshold shouldn't be too high either, since states
 			 * at the end of the loop are likely to be useful in pruning.
 			 */
+skip_inf_loop_check:
 			if (!env->test_state_freq &&
 			    env->jmps_processed - env->prev_jmps_processed < 20 &&
 			    env->insn_processed - env->prev_insn_processed < 100)
@@ -14698,6 +15276,7 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 			goto miss;
 		}
 		if (states_equal(env, &sl->state, cur)) {
+hit:
 			sl->hit_cnt++;
 			/* reached equivalent register/stack state,
 			 * prune the search.
-- 
cgit v1.2.3-70-g09d2


From 6018e1f407cccf39b804d1f75ad4de7be4e6cc45 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:17 -0800
Subject: bpf: implement numbers iterator

Implement the first open-coded iterator type over a range of integers.

It's public API consists of:
  - bpf_iter_num_new() constructor, which accepts [start, end) range
    (that is, start is inclusive, end is exclusive).
  - bpf_iter_num_next() which will keep returning read-only pointer to int
    until the range is exhausted, at which point NULL will be returned.
    If bpf_iter_num_next() is kept calling after this, NULL will be
    persistently returned.
  - bpf_iter_num_destroy() destructor, which needs to be called at some
    point to clean up iterator state. BPF verifier enforces that iterator
    destructor is called at some point before BPF program exits.

Note that `start = end = X` is a valid combination to setup an empty
iterator. bpf_iter_num_new() will return 0 (success) for any such
combination.

If bpf_iter_num_new() detects invalid combination of input arguments, it
returns error, resets iterator state to, effectively, empty iterator, so
any subsequent call to bpf_iter_num_next() will keep returning NULL.

BPF verifier has no knowledge that returned integers are in the
[start, end) value range, as both `start` and `end` are not statically
known and enforced: they are runtime values.

While the implementation is pretty trivial, some care needs to be taken
to avoid overflows and underflows. Subsequent selftests will validate
correctness of [start, end) semantics, especially around extremes
(INT_MIN and INT_MAX).

Similarly to bpf_loop(), we enforce that no more than BPF_MAX_LOOPS can
be specified.

bpf_iter_num_{new,next,destroy}() is a logical evolution from bounded
BPF loops and bpf_loop() helper and is the basis for implementing
ergonomic BPF loops with no statically known or verified bounds.
Subsequent patches implement bpf_for() macro, demonstrating how this can
be wrapped into something that works and feels like a normal for() loop
in C language.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h            |  8 +++--
 include/uapi/linux/bpf.h       |  8 +++++
 kernel/bpf/bpf_iter.c          | 70 ++++++++++++++++++++++++++++++++++++++++++
 kernel/bpf/helpers.c           |  3 ++
 tools/include/uapi/linux/bpf.h |  8 +++++
 5 files changed, 95 insertions(+), 2 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 6792a7940e1e..e64ff1e89fb2 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1617,8 +1617,12 @@ struct bpf_array {
 #define BPF_COMPLEXITY_LIMIT_INSNS      1000000 /* yes. 1M insns */
 #define MAX_TAIL_CALL_CNT 33
 
-/* Maximum number of loops for bpf_loop */
-#define BPF_MAX_LOOPS	BIT(23)
+/* Maximum number of loops for bpf_loop and bpf_iter_num.
+ * It's enum to expose it (and thus make it discoverable) through BTF.
+ */
+enum {
+	BPF_MAX_LOOPS = 8 * 1024 * 1024,
+};
 
 #define BPF_F_ACCESS_MASK	(BPF_F_RDONLY |		\
 				 BPF_F_RDONLY_PROG |	\
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 976b194eb775..4abddb668a10 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -7112,4 +7112,12 @@ enum {
 	BPF_F_TIMER_ABS = (1ULL << 0),
 };
 
+/* BPF numbers iterator state */
+struct bpf_iter_num {
+	/* opaque iterator state; having __u64 here allows to preserve correct
+	 * alignment requirements in vmlinux.h, generated from BTF
+	 */
+	__u64 __opaque[1];
+} __attribute__((aligned(8)));
+
 #endif /* _UAPI__LINUX_BPF_H__ */
diff --git a/kernel/bpf/bpf_iter.c b/kernel/bpf/bpf_iter.c
index 5dc307bdeaeb..96856f130cbf 100644
--- a/kernel/bpf/bpf_iter.c
+++ b/kernel/bpf/bpf_iter.c
@@ -776,3 +776,73 @@ const struct bpf_func_proto bpf_loop_proto = {
 	.arg3_type	= ARG_PTR_TO_STACK_OR_NULL,
 	.arg4_type	= ARG_ANYTHING,
 };
+
+struct bpf_iter_num_kern {
+	int cur; /* current value, inclusive */
+	int end; /* final value, exclusive */
+} __aligned(8);
+
+__diag_push();
+__diag_ignore_all("-Wmissing-prototypes",
+		  "Global functions as their definitions will be in vmlinux BTF");
+
+__bpf_kfunc int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end)
+{
+	struct bpf_iter_num_kern *s = (void *)it;
+
+	BUILD_BUG_ON(sizeof(struct bpf_iter_num_kern) != sizeof(struct bpf_iter_num));
+	BUILD_BUG_ON(__alignof__(struct bpf_iter_num_kern) != __alignof__(struct bpf_iter_num));
+
+	BTF_TYPE_EMIT(struct btf_iter_num);
+
+	/* start == end is legit, it's an empty range and we'll just get NULL
+	 * on first (and any subsequent) bpf_iter_num_next() call
+	 */
+	if (start > end) {
+		s->cur = s->end = 0;
+		return -EINVAL;
+	}
+
+	/* avoid overflows, e.g., if start == INT_MIN and end == INT_MAX */
+	if ((s64)end - (s64)start > BPF_MAX_LOOPS) {
+		s->cur = s->end = 0;
+		return -E2BIG;
+	}
+
+	/* user will call bpf_iter_num_next() first,
+	 * which will set s->cur to exactly start value;
+	 * underflow shouldn't matter
+	 */
+	s->cur = start - 1;
+	s->end = end;
+
+	return 0;
+}
+
+__bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it)
+{
+	struct bpf_iter_num_kern *s = (void *)it;
+
+	/* check failed initialization or if we are done (same behavior);
+	 * need to be careful about overflow, so convert to s64 for checks,
+	 * e.g., if s->cur == s->end == INT_MAX, we can't just do
+	 * s->cur + 1 >= s->end
+	 */
+	if ((s64)(s->cur + 1) >= s->end) {
+		s->cur = s->end = 0;
+		return NULL;
+	}
+
+	s->cur++;
+
+	return &s->cur;
+}
+
+__bpf_kfunc void bpf_iter_num_destroy(struct bpf_iter_num *it)
+{
+	struct bpf_iter_num_kern *s = (void *)it;
+
+	s->cur = s->end = 0;
+}
+
+__diag_pop();
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 637ac4e92e75..f9b7eeedce08 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2411,6 +2411,9 @@ BTF_ID_FLAGS(func, bpf_rcu_read_lock)
 BTF_ID_FLAGS(func, bpf_rcu_read_unlock)
 BTF_ID_FLAGS(func, bpf_dynptr_slice, KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_dynptr_slice_rdwr, KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_iter_num_new, KF_ITER_NEW)
+BTF_ID_FLAGS(func, bpf_iter_num_next, KF_ITER_NEXT | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_iter_num_destroy, KF_ITER_DESTROY)
 BTF_SET8_END(common_btf_ids)
 
 static const struct btf_kfunc_id_set common_kfunc_set = {
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 976b194eb775..4abddb668a10 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -7112,4 +7112,12 @@ enum {
 	BPF_F_TIMER_ABS = (1ULL << 0),
 };
 
+/* BPF numbers iterator state */
+struct bpf_iter_num {
+	/* opaque iterator state; having __u64 here allows to preserve correct
+	 * alignment requirements in vmlinux.h, generated from BTF
+	 */
+	__u64 __opaque[1];
+} __attribute__((aligned(8)));
+
 #endif /* _UAPI__LINUX_BPF_H__ */
-- 
cgit v1.2.3-70-g09d2


From 8c2b5e90505e474f36ecc3b7f3f8298b59d72e91 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:18 -0800
Subject: selftests/bpf: add bpf_for_each(), bpf_for(), and bpf_repeat() macros

Add bpf_for_each(), bpf_for(), and bpf_repeat() macros that make writing
open-coded iterator-based loops much more convenient and natural. These
macros utilize cleanup attribute to ensure proper destruction of the
iterator and thanks to that manage to provide the ergonomics that is
very close to C language's for() construct. Typical loop would look like:

  int i;
  int arr[N];

  bpf_for(i, 0, N) {
      /* verifier will know that i >= 0 && i < N, so could be used to
       * directly access array elements with no extra checks
       */
       arr[i] = i;
  }

bpf_repeat() is very similar, but it doesn't expose iteration number and
is meant as a simple "repeat action N times" loop:

  bpf_repeat(N) { /* whatever, N times */ }

Note that `break` and `continue` statements inside the {} block work as
expected.

bpf_for_each() is a generalization over any kind of BPF open-coded
iterator allowing to use for-each-like approach instead of calling
low-level bpf_iter_<type>_{new,next,destroy}() APIs explicitly. E.g.:

  struct cgroup *cg;

  bpf_for_each(cgroup, cg, some, input, args) {
      /* do something with each cg */
  }

would call (not-yet-implemented) bpf_iter_cgroup_{new,next,destroy}()
functions to form a loop over cgroups, where `some, input, args` are
passed verbatim into constructor as

  bpf_iter_cgroup_new(&it, some, input, args).

As a first demonstration, add pyperf variant based on the bpf_for() loop.

Also clean up a few tests that either included bpf_misc.h header
unnecessarily from the user-space, which is unsupported, or included it
before any common types are defined (and thus leading to unnecessary
compilation warnings, potentially).

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-6-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/bpf_verif_scale.c     |  6 ++
 .../selftests/bpf/prog_tests/uprobe_autoattach.c   |  1 -
 tools/testing/selftests/bpf/progs/bpf_misc.h       | 99 ++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/lsm.c            |  4 +-
 tools/testing/selftests/bpf/progs/pyperf.h         | 14 ++-
 tools/testing/selftests/bpf/progs/pyperf600_iter.c |  7 ++
 .../selftests/bpf/progs/pyperf600_nounroll.c       |  3 -
 7 files changed, 124 insertions(+), 10 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/pyperf600_iter.c

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c b/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
index 5ca252823294..731c343897d8 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
@@ -144,6 +144,12 @@ void test_verif_scale_pyperf600_nounroll()
 	scale_test("pyperf600_nounroll.bpf.o", BPF_PROG_TYPE_RAW_TRACEPOINT, false);
 }
 
+void test_verif_scale_pyperf600_iter()
+{
+	/* open-coded BPF iterator version */
+	scale_test("pyperf600_iter.bpf.o", BPF_PROG_TYPE_RAW_TRACEPOINT, false);
+}
+
 void test_verif_scale_loop1()
 {
 	scale_test("loop1.bpf.o", BPF_PROG_TYPE_RAW_TRACEPOINT, false);
diff --git a/tools/testing/selftests/bpf/prog_tests/uprobe_autoattach.c b/tools/testing/selftests/bpf/prog_tests/uprobe_autoattach.c
index 6558c857e620..d5b3377aa33c 100644
--- a/tools/testing/selftests/bpf/prog_tests/uprobe_autoattach.c
+++ b/tools/testing/selftests/bpf/prog_tests/uprobe_autoattach.c
@@ -3,7 +3,6 @@
 
 #include <test_progs.h>
 #include "test_uprobe_autoattach.skel.h"
-#include "progs/bpf_misc.h"
 
 /* uprobe attach point */
 static noinline int autoattach_trigger_func(int arg1, int arg2, int arg3,
diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index f704885aa534..597688a188ae 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -75,5 +75,104 @@
 #define FUNC_REG_ARG_CNT 5
 #endif
 
+struct bpf_iter_num;
+
+extern int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end) __ksym;
+extern int *bpf_iter_num_next(struct bpf_iter_num *it) __ksym;
+extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
+
+#ifndef bpf_for_each
+/* bpf_for_each(iter_type, cur_elem, args...) provides generic construct for
+ * using BPF open-coded iterators without having to write mundane explicit
+ * low-level loop logic. Instead, it provides for()-like generic construct
+ * that can be used pretty naturally. E.g., for some hypothetical cgroup
+ * iterator, you'd write:
+ *
+ * struct cgroup *cg, *parent_cg = <...>;
+ *
+ * bpf_for_each(cgroup, cg, parent_cg, CG_ITER_CHILDREN) {
+ *     bpf_printk("Child cgroup id = %d", cg->cgroup_id);
+ *     if (cg->cgroup_id == 123)
+ *         break;
+ * }
+ *
+ * I.e., it looks almost like high-level for each loop in other languages,
+ * supports continue/break, and is verifiable by BPF verifier.
+ *
+ * For iterating integers, the difference betwen bpf_for_each(num, i, N, M)
+ * and bpf_for(i, N, M) is in that bpf_for() provides additional proof to
+ * verifier that i is in [N, M) range, and in bpf_for_each() case i is `int
+ * *`, not just `int`. So for integers bpf_for() is more convenient.
+ *
+ * Note: this macro relies on C99 feature of allowing to declare variables
+ * inside for() loop, bound to for() loop lifetime. It also utilizes GCC
+ * extension: __attribute__((cleanup(<func>))), supported by both GCC and
+ * Clang.
+ */
+#define bpf_for_each(type, cur, args...) for (							\
+	/* initialize and define destructor */							\
+	struct bpf_iter_##type ___it __attribute__((aligned(8), /* enforce, just in case */,	\
+						    cleanup(bpf_iter_##type##_destroy))),	\
+	/* ___p pointer is just to call bpf_iter_##type##_new() *once* to init ___it */		\
+			       *___p = (bpf_iter_##type##_new(&___it, ##args),			\
+	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
+	/* for bpf_iter_##type##_destroy() when used from cleanup() attribute */		\
+					(void)bpf_iter_##type##_destroy, (void *)0);		\
+	/* iteration and termination check */							\
+	(((cur) = bpf_iter_##type##_next(&___it)));						\
+)
+#endif /* bpf_for_each */
+
+#ifndef bpf_for
+/* bpf_for(i, start, end) implements a for()-like looping construct that sets
+ * provided integer variable *i* to values starting from *start* through,
+ * but not including, *end*. It also proves to BPF verifier that *i* belongs
+ * to range [start, end), so this can be used for accessing arrays without
+ * extra checks.
+ *
+ * Note: *start* and *end* are assumed to be expressions with no side effects
+ * and whose values do not change throughout bpf_for() loop execution. They do
+ * not have to be statically known or constant, though.
+ *
+ * Note: similarly to bpf_for_each(), it relies on C99 feature of declaring for()
+ * loop bound variables and cleanup attribute, supported by GCC and Clang.
+ */
+#define bpf_for(i, start, end) for (								\
+	/* initialize and define destructor */							\
+	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
+						 cleanup(bpf_iter_num_destroy))),		\
+	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
+			    *___p = (bpf_iter_num_new(&___it, (start), (end)),			\
+	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
+	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
+				(void)bpf_iter_num_destroy, (void *)0);				\
+	({											\
+		/* iteration step */								\
+		int *___t = bpf_iter_num_next(&___it);						\
+		/* termination and bounds check */						\
+		(___t && ((i) = *___t, (i) >= (start) && (i) < (end)));				\
+	});											\
+)
+#endif /* bpf_for */
+
+#ifndef bpf_repeat
+/* bpf_repeat(N) performs N iterations without exposing iteration number
+ *
+ * Note: similarly to bpf_for_each(), it relies on C99 feature of declaring for()
+ * loop bound variables and cleanup attribute, supported by GCC and Clang.
+ */
+#define bpf_repeat(N) for (									\
+	/* initialize and define destructor */							\
+	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
+						 cleanup(bpf_iter_num_destroy))),		\
+	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
+			    *___p = (bpf_iter_num_new(&___it, 0, (N)),				\
+	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
+	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
+				(void)bpf_iter_num_destroy, (void *)0);				\
+	bpf_iter_num_next(&___it);								\
+	/* nothing here  */									\
+)
+#endif /* bpf_repeat */
 
 #endif
diff --git a/tools/testing/selftests/bpf/progs/lsm.c b/tools/testing/selftests/bpf/progs/lsm.c
index dc93887ed34c..fadfdd98707c 100644
--- a/tools/testing/selftests/bpf/progs/lsm.c
+++ b/tools/testing/selftests/bpf/progs/lsm.c
@@ -4,12 +4,12 @@
  * Copyright 2020 Google LLC.
  */
 
-#include "bpf_misc.h"
 #include "vmlinux.h"
+#include <errno.h>
 #include <bpf/bpf_core_read.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
-#include <errno.h>
+#include "bpf_misc.h"
 
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/pyperf.h b/tools/testing/selftests/bpf/progs/pyperf.h
index 6c7b1fb268d6..f2e7a31c8d75 100644
--- a/tools/testing/selftests/bpf/progs/pyperf.h
+++ b/tools/testing/selftests/bpf/progs/pyperf.h
@@ -7,6 +7,7 @@
 #include <stdbool.h>
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
 
 #define FUNCTION_NAME_LEN 64
 #define FILE_NAME_LEN 128
@@ -294,17 +295,22 @@ int __on_event(struct bpf_raw_tracepoint_args *ctx)
 	if (ctx.done)
 		return 0;
 #else
-#ifdef NO_UNROLL
+#if defined(USE_ITER)
+/* no for loop, no unrolling */
+#elif defined(NO_UNROLL)
 #pragma clang loop unroll(disable)
-#else
-#ifdef UNROLL_COUNT
+#elif defined(UNROLL_COUNT)
 #pragma clang loop unroll_count(UNROLL_COUNT)
 #else
 #pragma clang loop unroll(full)
-#endif
 #endif /* NO_UNROLL */
 		/* Unwind python stack */
+#ifdef USE_ITER
+		int i;
+		bpf_for(i, 0, STACK_MAX_LEN) {
+#else /* !USE_ITER */
 		for (int i = 0; i < STACK_MAX_LEN; ++i) {
+#endif
 			if (frame_ptr && get_frame_data(frame_ptr, pidData, &frame, &sym)) {
 				int32_t new_symbol_id = *symbol_counter * 64 + cur_cpu;
 				int32_t *symbol_id = bpf_map_lookup_elem(&symbolmap, &sym);
diff --git a/tools/testing/selftests/bpf/progs/pyperf600_iter.c b/tools/testing/selftests/bpf/progs/pyperf600_iter.c
new file mode 100644
index 000000000000..d62e1b200c30
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/pyperf600_iter.c
@@ -0,0 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2023 Meta Platforms, Inc. and affiliates.
+#define STACK_MAX_LEN 600
+#define SUBPROGS
+#define NO_UNROLL
+#define USE_ITER
+#include "pyperf.h"
diff --git a/tools/testing/selftests/bpf/progs/pyperf600_nounroll.c b/tools/testing/selftests/bpf/progs/pyperf600_nounroll.c
index 6beff7502f4d..520b58c4f8db 100644
--- a/tools/testing/selftests/bpf/progs/pyperf600_nounroll.c
+++ b/tools/testing/selftests/bpf/progs/pyperf600_nounroll.c
@@ -2,7 +2,4 @@
 // Copyright (c) 2019 Facebook
 #define STACK_MAX_LEN 600
 #define NO_UNROLL
-/* clang will not unroll at all.
- * Total program size is around 2k insns
- */
 #include "pyperf.h"
-- 
cgit v1.2.3-70-g09d2


From 57400dcce6c2cf3985120c4ee28b37a1f4238dbb Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:19 -0800
Subject: selftests/bpf: add iterators tests

Add various tests for open-coded iterators. Some of them excercise
various possible coding patterns in C, some go down to low-level
assembly for more control over various conditions, especially invalid
ones.

We also make use of bpf_for(), bpf_for_each(), bpf_repeat() macros in
some of these tests.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-7-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/iters.c     |  15 +
 tools/testing/selftests/bpf/progs/bpf_misc.h       |   1 +
 tools/testing/selftests/bpf/progs/iters.c          | 720 +++++++++++++++++++++
 tools/testing/selftests/bpf/progs/iters_looping.c  | 163 +++++
 .../selftests/bpf/progs/iters_state_safety.c       | 426 ++++++++++++
 5 files changed, 1325 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/iters.c
 create mode 100644 tools/testing/selftests/bpf/progs/iters.c
 create mode 100644 tools/testing/selftests/bpf/progs/iters_looping.c
 create mode 100644 tools/testing/selftests/bpf/progs/iters_state_safety.c

diff --git a/tools/testing/selftests/bpf/prog_tests/iters.c b/tools/testing/selftests/bpf/prog_tests/iters.c
new file mode 100644
index 000000000000..414fb8d82145
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/iters.c
@@ -0,0 +1,15 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <test_progs.h>
+
+#include "iters.skel.h"
+#include "iters_state_safety.skel.h"
+#include "iters_looping.skel.h"
+
+void test_iters(void)
+{
+	RUN_TESTS(iters_state_safety);
+	RUN_TESTS(iters_looping);
+	RUN_TESTS(iters);
+}
diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 597688a188ae..43b154a639e7 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -36,6 +36,7 @@
 #define __clobber_common "r0", "r1", "r2", "r3", "r4", "r5", "memory"
 #define __imm(name) [name]"i"(name)
 #define __imm_addr(name) [name]"i"(&name)
+#define __imm_ptr(name) [name]"p"(&name)
 
 #if defined(__TARGET_ARCH_x86)
 #define SYSCALL_WRAPPER 1
diff --git a/tools/testing/selftests/bpf/progs/iters.c b/tools/testing/selftests/bpf/progs/iters.c
new file mode 100644
index 000000000000..84e5dc10243c
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/iters.c
@@ -0,0 +1,720 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <stdbool.h>
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+
+static volatile int zero = 0;
+
+int my_pid;
+int arr[256];
+int small_arr[16] SEC(".data.small_arr");
+
+#ifdef REAL_TEST
+#define MY_PID_GUARD() if (my_pid != (bpf_get_current_pid_tgid() >> 32)) return 0
+#else
+#define MY_PID_GUARD() ({ })
+#endif
+
+SEC("?raw_tp")
+__failure __msg("math between map_value pointer and register with unbounded min value is not allowed")
+int iter_err_unsafe_c_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i = zero; /* obscure initial value of i */
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 1000);
+	while ((v = bpf_iter_num_next(&it))) {
+		i++;
+	}
+	bpf_iter_num_destroy(&it);
+
+	small_arr[i] = 123; /* invalid */
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("unbounded memory access")
+int iter_err_unsafe_asm_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i = 0;
+
+	MY_PID_GUARD();
+
+	asm volatile (
+		"r6 = %[zero];" /* iteration counter */
+		"r1 = %[it];" /* iterator state */
+		"r2 = 0;"
+		"r3 = 1000;"
+		"r4 = 1;"
+		"call %[bpf_iter_num_new];"
+	"loop:"
+		"r1 = %[it];"
+		"call %[bpf_iter_num_next];"
+		"if r0 == 0 goto out;"
+		"r6 += 1;"
+		"goto loop;"
+	"out:"
+		"r1 = %[it];"
+		"call %[bpf_iter_num_destroy];"
+		"r1 = %[small_arr];"
+		"r2 = r6;"
+		"r2 <<= 2;"
+		"r1 += r2;"
+		"*(u32 *)(r1 + 0) = r6;" /* invalid */
+		:
+		: [it]"r"(&it),
+		  [small_arr]"p"(small_arr),
+		  [zero]"p"(zero),
+		  __imm(bpf_iter_num_new),
+		  __imm(bpf_iter_num_next),
+		  __imm(bpf_iter_num_destroy)
+		: __clobber_common, "r6"
+	);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_while_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 3);
+	while ((v = bpf_iter_num_next(&it))) {
+		bpf_printk("ITER_BASIC: E1 VAL: v=%d", *v);
+	}
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_while_loop_auto_cleanup(const void *ctx)
+{
+	__attribute__((cleanup(bpf_iter_num_destroy))) struct bpf_iter_num it;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 3);
+	while ((v = bpf_iter_num_next(&it))) {
+		bpf_printk("ITER_BASIC: E1 VAL: v=%d", *v);
+	}
+	/* (!) no explicit bpf_iter_num_destroy() */
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_for_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 5, 10);
+	for (v = bpf_iter_num_next(&it); v; v = bpf_iter_num_next(&it)) {
+		bpf_printk("ITER_BASIC: E2 VAL: v=%d", *v);
+	}
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_bpf_for_each_macro(const void *ctx)
+{
+	int *v;
+
+	MY_PID_GUARD();
+
+	bpf_for_each(num, v, 5, 10) {
+		bpf_printk("ITER_BASIC: E2 VAL: v=%d", *v);
+	}
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_bpf_for_macro(const void *ctx)
+{
+	int i;
+
+	MY_PID_GUARD();
+
+	bpf_for(i, 5, 10) {
+		bpf_printk("ITER_BASIC: E2 VAL: v=%d", i);
+	}
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_pragma_unroll_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 2);
+#pragma nounroll
+	for (i = 0; i < 3; i++) {
+		v = bpf_iter_num_next(&it);
+		bpf_printk("ITER_BASIC: E3 VAL: i=%d v=%d", i, v ? *v : -1);
+	}
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_manual_unroll_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 100, 200);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d", v ? *v : -1);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d", v ? *v : -1);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d", v ? *v : -1);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d\n", v ? *v : -1);
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_multiple_sequential_loops(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 3);
+	while ((v = bpf_iter_num_next(&it))) {
+		bpf_printk("ITER_BASIC: E1 VAL: v=%d", *v);
+	}
+	bpf_iter_num_destroy(&it);
+
+	bpf_iter_num_new(&it, 5, 10);
+	for (v = bpf_iter_num_next(&it); v; v = bpf_iter_num_next(&it)) {
+		bpf_printk("ITER_BASIC: E2 VAL: v=%d", *v);
+	}
+	bpf_iter_num_destroy(&it);
+
+	bpf_iter_num_new(&it, 0, 2);
+#pragma nounroll
+	for (i = 0; i < 3; i++) {
+		v = bpf_iter_num_next(&it);
+		bpf_printk("ITER_BASIC: E3 VAL: i=%d v=%d", i, v ? *v : -1);
+	}
+	bpf_iter_num_destroy(&it);
+
+	bpf_iter_num_new(&it, 100, 200);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d", v ? *v : -1);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d", v ? *v : -1);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d", v ? *v : -1);
+	v = bpf_iter_num_next(&it);
+	bpf_printk("ITER_BASIC: E4 VAL: v=%d\n", v ? *v : -1);
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_limit_cond_break_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, i = 0, sum = 0;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 10);
+	while ((v = bpf_iter_num_next(&it))) {
+		bpf_printk("ITER_SIMPLE: i=%d v=%d", i, *v);
+		sum += *v;
+
+		i++;
+		if (i > 3)
+			break;
+	}
+	bpf_iter_num_destroy(&it);
+
+	bpf_printk("ITER_SIMPLE: sum=%d\n", sum);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_obfuscate_counter(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, sum = 0;
+	/* Make i's initial value unknowable for verifier to prevent it from
+	 * pruning if/else branch inside the loop body and marking i as precise.
+	 */
+	int i = zero;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 10);
+	while ((v = bpf_iter_num_next(&it))) {
+		int x;
+
+		i += 1;
+
+		/* If we initialized i as `int i = 0;` above, verifier would
+		 * track that i becomes 1 on first iteration after increment
+		 * above, and here verifier would eagerly prune else branch
+		 * and mark i as precise, ruining open-coded iterator logic
+		 * completely, as each next iteration would have a different
+		 * *precise* value of i, and thus there would be no
+		 * convergence of state. This would result in reaching maximum
+		 * instruction limit, no matter what the limit is.
+		 */
+		if (i == 1)
+			x = 123;
+		else
+			x = i * 3 + 1;
+
+		bpf_printk("ITER_OBFUSCATE_COUNTER: i=%d v=%d x=%d", i, *v, x);
+
+		sum += x;
+	}
+	bpf_iter_num_destroy(&it);
+
+	bpf_printk("ITER_OBFUSCATE_COUNTER: sum=%d\n", sum);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_search_loop(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, *elem = NULL;
+	bool found = false;
+
+	MY_PID_GUARD();
+
+	bpf_iter_num_new(&it, 0, 10);
+
+	while ((v = bpf_iter_num_next(&it))) {
+		bpf_printk("ITER_SEARCH_LOOP: v=%d", *v);
+
+		if (*v == 2) {
+			found = true;
+			elem = v;
+			barrier_var(elem);
+		}
+	}
+
+	/* should fail to verify if bpf_iter_num_destroy() is here */
+
+	if (found)
+		/* here found element will be wrong, we should have copied
+		 * value to a variable, but here we want to make sure we can
+		 * access memory after the loop anyways
+		 */
+		bpf_printk("ITER_SEARCH_LOOP: FOUND IT = %d!\n", *elem);
+	else
+		bpf_printk("ITER_SEARCH_LOOP: NOT FOUND IT!\n");
+
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_array_fill(const void *ctx)
+{
+	int sum, i;
+
+	MY_PID_GUARD();
+
+	bpf_for(i, 0, ARRAY_SIZE(arr)) {
+		arr[i] = i * 2;
+	}
+
+	sum = 0;
+	bpf_for(i, 0, ARRAY_SIZE(arr)) {
+		sum += arr[i];
+	}
+
+	bpf_printk("ITER_ARRAY_FILL: sum=%d (should be %d)\n", sum, 255 * 256);
+
+	return 0;
+}
+
+static int arr2d[4][5];
+static int arr2d_row_sums[4];
+static int arr2d_col_sums[5];
+
+SEC("raw_tp")
+__success
+int iter_nested_iters(const void *ctx)
+{
+	int sum, row, col;
+
+	MY_PID_GUARD();
+
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		bpf_for( col, 0, ARRAY_SIZE(arr2d[0])) {
+			arr2d[row][col] = row * col;
+		}
+	}
+
+	/* zero-initialize sums */
+	sum = 0;
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		arr2d_row_sums[row] = 0;
+	}
+	bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+		arr2d_col_sums[col] = 0;
+	}
+
+	/* calculate sums */
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+			sum += arr2d[row][col];
+			arr2d_row_sums[row] += arr2d[row][col];
+			arr2d_col_sums[col] += arr2d[row][col];
+		}
+	}
+
+	bpf_printk("ITER_NESTED_ITERS: total sum=%d", sum);
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		bpf_printk("ITER_NESTED_ITERS: row #%d sum=%d", row, arr2d_row_sums[row]);
+	}
+	bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+		bpf_printk("ITER_NESTED_ITERS: col #%d sum=%d%s",
+			   col, arr2d_col_sums[col],
+			   col == ARRAY_SIZE(arr2d[0]) - 1 ? "\n" : "");
+	}
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_nested_deeply_iters(const void *ctx)
+{
+	int sum = 0;
+
+	MY_PID_GUARD();
+
+	bpf_repeat(10) {
+		bpf_repeat(10) {
+			bpf_repeat(10) {
+				bpf_repeat(10) {
+					bpf_repeat(10) {
+						sum += 1;
+					}
+				}
+			}
+		}
+		/* validate that we can break from inside bpf_repeat() */
+		break;
+	}
+
+	return sum;
+}
+
+static __noinline void fill_inner_dimension(int row)
+{
+	int col;
+
+	bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+		arr2d[row][col] = row * col;
+	}
+}
+
+static __noinline int sum_inner_dimension(int row)
+{
+	int sum = 0, col;
+
+	bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+		sum += arr2d[row][col];
+		arr2d_row_sums[row] += arr2d[row][col];
+		arr2d_col_sums[col] += arr2d[row][col];
+	}
+
+	return sum;
+}
+
+SEC("raw_tp")
+__success
+int iter_subprog_iters(const void *ctx)
+{
+	int sum, row, col;
+
+	MY_PID_GUARD();
+
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		fill_inner_dimension(row);
+	}
+
+	/* zero-initialize sums */
+	sum = 0;
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		arr2d_row_sums[row] = 0;
+	}
+	bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+		arr2d_col_sums[col] = 0;
+	}
+
+	/* calculate sums */
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		sum += sum_inner_dimension(row);
+	}
+
+	bpf_printk("ITER_SUBPROG_ITERS: total sum=%d", sum);
+	bpf_for(row, 0, ARRAY_SIZE(arr2d)) {
+		bpf_printk("ITER_SUBPROG_ITERS: row #%d sum=%d",
+			   row, arr2d_row_sums[row]);
+	}
+	bpf_for(col, 0, ARRAY_SIZE(arr2d[0])) {
+		bpf_printk("ITER_SUBPROG_ITERS: col #%d sum=%d%s",
+			   col, arr2d_col_sums[col],
+			   col == ARRAY_SIZE(arr2d[0]) - 1 ? "\n" : "");
+	}
+
+	return 0;
+}
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, int);
+	__type(value, int);
+	__uint(max_entries, 1000);
+} arr_map SEC(".maps");
+
+SEC("?raw_tp")
+__failure __msg("invalid mem access 'scalar'")
+int iter_err_too_permissive1(const void *ctx)
+{
+	int *map_val = NULL;
+	int key = 0;
+
+	MY_PID_GUARD();
+
+	map_val = bpf_map_lookup_elem(&arr_map, &key);
+	if (!map_val)
+		return 0;
+
+	bpf_repeat(1000000) {
+		map_val = NULL;
+	}
+
+	*map_val = 123;
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("invalid mem access 'map_value_or_null'")
+int iter_err_too_permissive2(const void *ctx)
+{
+	int *map_val = NULL;
+	int key = 0;
+
+	MY_PID_GUARD();
+
+	map_val = bpf_map_lookup_elem(&arr_map, &key);
+	if (!map_val)
+		return 0;
+
+	bpf_repeat(1000000) {
+		map_val = bpf_map_lookup_elem(&arr_map, &key);
+	}
+
+	*map_val = 123;
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("invalid mem access 'map_value_or_null'")
+int iter_err_too_permissive3(const void *ctx)
+{
+	int *map_val = NULL;
+	int key = 0;
+	bool found = false;
+
+	MY_PID_GUARD();
+
+	bpf_repeat(1000000) {
+		map_val = bpf_map_lookup_elem(&arr_map, &key);
+		found = true;
+	}
+
+	if (found)
+		*map_val = 123;
+
+	return 0;
+}
+
+SEC("raw_tp")
+__success
+int iter_tricky_but_fine(const void *ctx)
+{
+	int *map_val = NULL;
+	int key = 0;
+	bool found = false;
+
+	MY_PID_GUARD();
+
+	bpf_repeat(1000000) {
+		map_val = bpf_map_lookup_elem(&arr_map, &key);
+		if (map_val) {
+			found = true;
+			break;
+		}
+	}
+
+	if (found)
+		*map_val = 123;
+
+	return 0;
+}
+
+#define __bpf_memzero(p, sz) bpf_probe_read_kernel((p), (sz), 0)
+
+SEC("raw_tp")
+__success
+int iter_stack_array_loop(const void *ctx)
+{
+	long arr1[16], arr2[16], sum = 0;
+	int *v, i;
+
+	MY_PID_GUARD();
+
+	/* zero-init arr1 and arr2 in such a way that verifier doesn't know
+	 * it's all zeros; if we don't do that, we'll make BPF verifier track
+	 * all combination of zero/non-zero stack slots for arr1/arr2, which
+	 * will lead to O(2^(ARRAY_SIZE(arr1)+ARRAY_SIZE(arr2))) different
+	 * states
+	 */
+	__bpf_memzero(arr1, sizeof(arr1));
+	__bpf_memzero(arr2, sizeof(arr1));
+
+	/* validate that we can break and continue when using bpf_for() */
+	bpf_for(i, 0, ARRAY_SIZE(arr1)) {
+		if (i & 1) {
+			arr1[i] = i;
+			continue;
+		} else {
+			arr2[i] = i;
+			break;
+		}
+	}
+
+	bpf_for(i, 0, ARRAY_SIZE(arr1)) {
+		sum += arr1[i] + arr2[i];
+	}
+
+	return sum;
+}
+
+static __noinline void fill(struct bpf_iter_num *it, int *arr, __u32 n, int mul)
+{
+	int *t, i;
+
+	while ((t = bpf_iter_num_next(it))) {
+		i = *t;
+		if (i >= n)
+			break;
+		arr[i] =  i * mul;
+	}
+}
+
+static __noinline int sum(struct bpf_iter_num *it, int *arr, __u32 n)
+{
+	int *t, i, sum = 0;;
+
+	while ((t = bpf_iter_num_next(it))) {
+		i = *t;
+		if (i >= n)
+			break;
+		sum += arr[i];
+	}
+
+	return sum;
+}
+
+SEC("raw_tp")
+__success
+int iter_pass_iter_ptr_to_subprog(const void *ctx)
+{
+	int arr1[16], arr2[32];
+	struct bpf_iter_num it;
+	int n, sum1, sum2;
+
+	MY_PID_GUARD();
+
+	/* fill arr1 */
+	n = ARRAY_SIZE(arr1);
+	bpf_iter_num_new(&it, 0, n);
+	fill(&it, arr1, n, 2);
+	bpf_iter_num_destroy(&it);
+
+	/* fill arr2 */
+	n = ARRAY_SIZE(arr2);
+	bpf_iter_num_new(&it, 0, n);
+	fill(&it, arr2, n, 10);
+	bpf_iter_num_destroy(&it);
+
+	/* sum arr1 */
+	n = ARRAY_SIZE(arr1);
+	bpf_iter_num_new(&it, 0, n);
+	sum1 = sum(&it, arr1, n);
+	bpf_iter_num_destroy(&it);
+
+	/* sum arr2 */
+	n = ARRAY_SIZE(arr2);
+	bpf_iter_num_new(&it, 0, n);
+	sum2 = sum(&it, arr2, n);
+	bpf_iter_num_destroy(&it);
+
+	bpf_printk("sum1=%d, sum2=%d", sum1, sum2);
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/iters_looping.c b/tools/testing/selftests/bpf/progs/iters_looping.c
new file mode 100644
index 000000000000..05fa5ce7fc59
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/iters_looping.c
@@ -0,0 +1,163 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <errno.h>
+#include <string.h>
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+char _license[] SEC("license") = "GPL";
+
+#define ITER_HELPERS						\
+	  __imm(bpf_iter_num_new),				\
+	  __imm(bpf_iter_num_next),				\
+	  __imm(bpf_iter_num_destroy)
+
+SEC("?raw_tp")
+__success
+int force_clang_to_emit_btf_for_externs(void *ctx)
+{
+	/* we need this as a workaround to enforce compiler emitting BTF
+	 * information for bpf_iter_num_{new,next,destroy}() kfuncs,
+	 * as, apparently, it doesn't emit it for symbols only referenced from
+	 * assembly (or cleanup attribute, for that matter, as well)
+	 */
+	bpf_repeat(0);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__success
+int consume_first_item_only(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* consume first item */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_next];"
+
+		"if r0 == 0 goto +1;"
+		"r0 = *(u32 *)(r0 + 0);"
+
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("R0 invalid mem access 'scalar'")
+int missing_null_check_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* consume first element */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_next];"
+
+		/* FAIL: deref with no NULL check */
+		"r1 = *(u32 *)(r0 + 0);"
+
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure
+__msg("invalid access to memory, mem_size=4 off=0 size=8")
+__msg("R0 min value is outside of the allowed memory range")
+int wrong_sized_read_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* consume first element */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_next];"
+
+		"if r0 == 0 goto +1;"
+		/* FAIL: deref more than available 4 bytes */
+		"r0 = *(u64 *)(r0 + 0);"
+
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__success __log_level(2)
+__flag(BPF_F_TEST_STATE_FREQ)
+int simplest_loop(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		"r6 = 0;" /* init sum */
+
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 10;"
+		"call %[bpf_iter_num_new];"
+
+	"1:"
+		/* consume next item */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_next];"
+
+		"if r0 == 0 goto 2f;"
+		"r0 = *(u32 *)(r0 + 0);"
+		"r6 += r0;" /* accumulate sum */
+		"goto 1b;"
+
+	"2:"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common, "r6"
+	);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/iters_state_safety.c b/tools/testing/selftests/bpf/progs/iters_state_safety.c
new file mode 100644
index 000000000000..d47e59aba6de
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/iters_state_safety.c
@@ -0,0 +1,426 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2022 Facebook */
+
+#include <errno.h>
+#include <string.h>
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+char _license[] SEC("license") = "GPL";
+
+#define ITER_HELPERS						\
+	  __imm(bpf_iter_num_new),				\
+	  __imm(bpf_iter_num_next),				\
+	  __imm(bpf_iter_num_destroy)
+
+SEC("?raw_tp")
+__success
+int force_clang_to_emit_btf_for_externs(void *ctx)
+{
+	/* we need this as a workaround to enforce compiler emitting BTF
+	 * information for bpf_iter_num_{new,next,destroy}() kfuncs,
+	 * as, apparently, it doesn't emit it for symbols only referenced from
+	 * assembly (or cleanup attribute, for that matter, as well)
+	 */
+	bpf_repeat(0);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__success __log_level(2)
+__msg("fp-8_w=iter_num(ref_id=1,state=active,depth=0)")
+int create_and_destroy(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("Unreleased reference id=1")
+int create_and_forget_to_destroy_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected an initialized iter_num as arg #1")
+int destroy_without_creating_fail(void *ctx)
+{
+	/* init with zeros to stop verifier complaining about uninit stack */
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected an initialized iter_num as arg #1")
+int compromise_iter_w_direct_write_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* directly write over first half of iter state */
+		"*(u64 *)(%[iter] + 0) = r0;"
+
+		/* (attempt to) destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("Unreleased reference id=1")
+int compromise_iter_w_direct_write_and_skip_destroy_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* directly write over first half of iter state */
+		"*(u64 *)(%[iter] + 0) = r0;"
+
+		/* don't destroy iter, leaking ref, which should fail */
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected an initialized iter_num as arg #1")
+int compromise_iter_w_helper_write_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* overwrite 8th byte with bpf_probe_read_kernel() */
+		"r1 = %[iter];"
+		"r1 += 7;"
+		"r2 = 1;"
+		"r3 = 0;" /* NULL */
+		"call %[bpf_probe_read_kernel];"
+
+		/* (attempt to) destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS, __imm(bpf_probe_read_kernel)
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+static __noinline void subprog_with_iter(void)
+{
+	struct bpf_iter_num iter;
+
+	bpf_iter_num_new(&iter, 0, 1);
+
+	return;
+}
+
+SEC("?raw_tp")
+__failure
+/* ensure there was a call to subprog, which might happen without __noinline */
+__msg("returning from callee:")
+__msg("Unreleased reference id=1")
+int leak_iter_from_subprog_fail(void *ctx)
+{
+	subprog_with_iter();
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__success __log_level(2)
+__msg("fp-8_w=iter_num(ref_id=1,state=active,depth=0)")
+int valid_stack_reuse(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+
+		/* now reuse same stack slots */
+
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected uninitialized iter_num as arg #1")
+int double_create_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* (attempt to) create iterator again */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected an initialized iter_num as arg #1")
+int double_destroy_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		/* (attempt to) destroy iterator again */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected an initialized iter_num as arg #1")
+int next_without_new_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* don't create iterator and try to iterate*/
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_next];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("expected an initialized iter_num as arg #1")
+int next_after_destroy_fail(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* create iterator */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+		/* destroy iterator */
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_destroy];"
+		/* don't create iterator and try to iterate*/
+		"r1 = %[iter];"
+		"call %[bpf_iter_num_next];"
+		:
+		: __imm_ptr(iter), ITER_HELPERS
+		: __clobber_common
+	);
+
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("invalid read from stack")
+int __naked read_from_iter_slot_fail(void)
+{
+	asm volatile (
+		/* r6 points to struct bpf_iter_num on the stack */
+		"r6 = r10;"
+		"r6 += -24;"
+
+		/* create iterator */
+		"r1 = r6;"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* attemp to leak bpf_iter_num state */
+		"r7 = *(u64 *)(r6 + 0);"
+		"r8 = *(u64 *)(r6 + 8);"
+
+		/* destroy iterator */
+		"r1 = r6;"
+		"call %[bpf_iter_num_destroy];"
+
+		/* leak bpf_iter_num state */
+		"r0 = r7;"
+		"if r7 > r8 goto +1;"
+		"r0 = r8;"
+		"exit;"
+		:
+		: ITER_HELPERS
+		: __clobber_common, "r6", "r7", "r8"
+	);
+}
+
+int zero;
+
+SEC("?raw_tp")
+__failure
+__flag(BPF_F_TEST_STATE_FREQ)
+__msg("Unreleased reference")
+int stacksafe_should_not_conflate_stack_spill_and_iter(void *ctx)
+{
+	struct bpf_iter_num iter;
+
+	asm volatile (
+		/* Create a fork in logic, with general setup as follows:
+		 *   - fallthrough (first) path is valid;
+		 *   - branch (second) path is invalid.
+		 * Then depending on what we do in fallthrough vs branch path,
+		 * we try to detect bugs in func_states_equal(), regsafe(),
+		 * refsafe(), stack_safe(), and similar by tricking verifier
+		 * into believing that branch state is a valid subset of
+		 * a fallthrough state. Verifier should reject overall
+		 * validation, unless there is a bug somewhere in verifier
+		 * logic.
+		 */
+		"call %[bpf_get_prandom_u32];"
+		"r6 = r0;"
+		"call %[bpf_get_prandom_u32];"
+		"r7 = r0;"
+
+		"if r6 > r7 goto bad;" /* fork */
+
+		/* spill r6 into stack slot of bpf_iter_num var */
+		"*(u64 *)(%[iter] + 0) = r6;"
+
+		"goto skip_bad;"
+
+	"bad:"
+		/* create iterator in the same stack slot */
+		"r1 = %[iter];"
+		"r2 = 0;"
+		"r3 = 1000;"
+		"call %[bpf_iter_num_new];"
+
+		/* but then forget about it and overwrite it back to r6 spill */
+		"*(u64 *)(%[iter] + 0) = r6;"
+
+	"skip_bad:"
+		"goto +0;" /* force checkpoint */
+
+		/* corrupt stack slots, if they are really dynptr */
+		"*(u64 *)(%[iter] + 0) = r6;"
+		:
+		: __imm_ptr(iter),
+		  __imm_addr(zero),
+		  __imm(bpf_get_prandom_u32),
+		  __imm(bpf_dynptr_from_mem),
+		  ITER_HELPERS
+		: __clobber_common, "r6", "r7"
+	);
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From f59b146092653bcf014ccdc9bd8bc94e79065ce3 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:20 -0800
Subject: selftests/bpf: add number iterator tests

Add number iterator (bpf_iter_num_{new,next,destroy}()) tests,
validating the correct handling of various corner and common cases
*at runtime*.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-8-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/iters.c |  49 +++++
 tools/testing/selftests/bpf/progs/iters_num.c  | 242 +++++++++++++++++++++++++
 2 files changed, 291 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/iters_num.c

diff --git a/tools/testing/selftests/bpf/prog_tests/iters.c b/tools/testing/selftests/bpf/prog_tests/iters.c
index 414fb8d82145..2e7caff9523e 100644
--- a/tools/testing/selftests/bpf/prog_tests/iters.c
+++ b/tools/testing/selftests/bpf/prog_tests/iters.c
@@ -6,10 +6,59 @@
 #include "iters.skel.h"
 #include "iters_state_safety.skel.h"
 #include "iters_looping.skel.h"
+#include "iters_num.skel.h"
+
+static void subtest_num_iters(void)
+{
+	struct iters_num *skel;
+	int err;
+
+	skel = iters_num__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel_open_and_load"))
+		return;
+
+	err = iters_num__attach(skel);
+	if (!ASSERT_OK(err, "skel_attach"))
+		goto cleanup;
+
+	usleep(1);
+	iters_num__detach(skel);
+
+#define VALIDATE_CASE(case_name)					\
+	ASSERT_EQ(skel->bss->res_##case_name,				\
+		  skel->rodata->exp_##case_name,			\
+		  #case_name)
+
+	VALIDATE_CASE(empty_zero);
+	VALIDATE_CASE(empty_int_min);
+	VALIDATE_CASE(empty_int_max);
+	VALIDATE_CASE(empty_minus_one);
+
+	VALIDATE_CASE(simple_sum);
+	VALIDATE_CASE(neg_sum);
+	VALIDATE_CASE(very_neg_sum);
+	VALIDATE_CASE(neg_pos_sum);
+
+	VALIDATE_CASE(invalid_range);
+	VALIDATE_CASE(max_range);
+	VALIDATE_CASE(e2big_range);
+
+	VALIDATE_CASE(succ_elem_cnt);
+	VALIDATE_CASE(overfetched_elem_cnt);
+	VALIDATE_CASE(fail_elem_cnt);
+
+#undef VALIDATE_CASE
+
+cleanup:
+	iters_num__destroy(skel);
+}
 
 void test_iters(void)
 {
 	RUN_TESTS(iters_state_safety);
 	RUN_TESTS(iters_looping);
 	RUN_TESTS(iters);
+
+	if (test__start_subtest("num"))
+		subtest_num_iters();
 }
diff --git a/tools/testing/selftests/bpf/progs/iters_num.c b/tools/testing/selftests/bpf/progs/iters_num.c
new file mode 100644
index 000000000000..7a77a8daee0d
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/iters_num.c
@@ -0,0 +1,242 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <limits.h>
+#include <linux/errno.h>
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+const volatile __s64 exp_empty_zero = 0 + 1;
+__s64 res_empty_zero;
+
+SEC("raw_tp/sys_enter")
+int num_empty_zero(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, 0, 0) sum += i;
+	res_empty_zero = 1 + sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_empty_int_min = 0 + 2;
+__s64 res_empty_int_min;
+
+SEC("raw_tp/sys_enter")
+int num_empty_int_min(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, INT_MIN, INT_MIN) sum += i;
+	res_empty_int_min = 2 + sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_empty_int_max = 0 + 3;
+__s64 res_empty_int_max;
+
+SEC("raw_tp/sys_enter")
+int num_empty_int_max(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, INT_MAX, INT_MAX) sum += i;
+	res_empty_int_max = 3 + sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_empty_minus_one = 0 + 4;
+__s64 res_empty_minus_one;
+
+SEC("raw_tp/sys_enter")
+int num_empty_minus_one(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, -1, -1) sum += i;
+	res_empty_minus_one = 4 + sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_simple_sum = 9 * 10 / 2;
+__s64 res_simple_sum;
+
+SEC("raw_tp/sys_enter")
+int num_simple_sum(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, 0, 10) sum += i;
+	res_simple_sum = sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_neg_sum = -11 * 10 / 2;
+__s64 res_neg_sum;
+
+SEC("raw_tp/sys_enter")
+int num_neg_sum(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, -10, 0) sum += i;
+	res_neg_sum = sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_very_neg_sum = INT_MIN + (__s64)(INT_MIN + 1);
+__s64 res_very_neg_sum;
+
+SEC("raw_tp/sys_enter")
+int num_very_neg_sum(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, INT_MIN, INT_MIN + 2) sum += i;
+	res_very_neg_sum = sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_very_big_sum = (__s64)(INT_MAX - 1) + (__s64)(INT_MAX - 2);
+__s64 res_very_big_sum;
+
+SEC("raw_tp/sys_enter")
+int num_very_big_sum(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, INT_MAX - 2, INT_MAX) sum += i;
+	res_very_big_sum = sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_neg_pos_sum = -3;
+__s64 res_neg_pos_sum;
+
+SEC("raw_tp/sys_enter")
+int num_neg_pos_sum(const void *ctx)
+{
+	__s64 sum = 0, i;
+
+	bpf_for(i, -3, 3) sum += i;
+	res_neg_pos_sum = sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_invalid_range = -EINVAL;
+__s64 res_invalid_range;
+
+SEC("raw_tp/sys_enter")
+int num_invalid_range(const void *ctx)
+{
+	struct bpf_iter_num it;
+
+	res_invalid_range = bpf_iter_num_new(&it, 1, 0);
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+const volatile __s64 exp_max_range = 0 + 10;
+__s64 res_max_range;
+
+SEC("raw_tp/sys_enter")
+int num_max_range(const void *ctx)
+{
+	struct bpf_iter_num it;
+
+	res_max_range = 10 + bpf_iter_num_new(&it, 0, BPF_MAX_LOOPS);
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+const volatile __s64 exp_e2big_range = -E2BIG;
+__s64 res_e2big_range;
+
+SEC("raw_tp/sys_enter")
+int num_e2big_range(const void *ctx)
+{
+	struct bpf_iter_num it;
+
+	res_e2big_range = bpf_iter_num_new(&it, -1, BPF_MAX_LOOPS);
+	bpf_iter_num_destroy(&it);
+
+	return 0;
+}
+
+const volatile __s64 exp_succ_elem_cnt = 10;
+__s64 res_succ_elem_cnt;
+
+SEC("raw_tp/sys_enter")
+int num_succ_elem_cnt(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int cnt = 0, *v;
+
+	bpf_iter_num_new(&it, 0, 10);
+	while ((v = bpf_iter_num_next(&it))) {
+		cnt++;
+	}
+	bpf_iter_num_destroy(&it);
+
+	res_succ_elem_cnt = cnt;
+
+	return 0;
+}
+
+const volatile __s64 exp_overfetched_elem_cnt = 5;
+__s64 res_overfetched_elem_cnt;
+
+SEC("raw_tp/sys_enter")
+int num_overfetched_elem_cnt(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int cnt = 0, *v, i;
+
+	bpf_iter_num_new(&it, 0, 5);
+	for (i = 0; i < 10; i++) {
+		v = bpf_iter_num_next(&it);
+		if (v)
+			cnt++;
+	}
+	bpf_iter_num_destroy(&it);
+
+	res_overfetched_elem_cnt = cnt;
+
+	return 0;
+}
+
+const volatile __s64 exp_fail_elem_cnt = 20 + 0;
+__s64 res_fail_elem_cnt;
+
+SEC("raw_tp/sys_enter")
+int num_fail_elem_cnt(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int cnt = 0, *v, i;
+
+	bpf_iter_num_new(&it, 100, 10);
+	for (i = 0; i < 10; i++) {
+		v = bpf_iter_num_next(&it);
+		if (v)
+			cnt++;
+	}
+	bpf_iter_num_destroy(&it);
+
+	res_fail_elem_cnt = 20 + cnt;
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 7e86a8c4ac8d5dcf7dd58f5a4779d1a6ff0a827d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 10:41:21 -0800
Subject: selftests/bpf: implement and test custom testmod_seq iterator

Implement a trivial iterator returning same specified integer value
N times as part of bpf_testmod kernel module. Add selftests to validate
everything works end to end.

We also reuse these tests as "verification-only" tests to validate that
kernel prints the state of custom kernel module-defined iterator correctly:

  fp-16=iter_testmod_seq(ref_id=1,state=drained,depth=0)

"testmod_seq" part is an iterator type, and is coming from module's BTF
data dynamically at runtime.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230308184121.1165081-9-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/DENYLIST.s390x         |  1 +
 .../selftests/bpf/bpf_testmod/bpf_testmod.c        | 42 +++++++++++-
 .../selftests/bpf/bpf_testmod/bpf_testmod.h        |  6 ++
 tools/testing/selftests/bpf/prog_tests/iters.c     | 42 ++++++++++++
 .../selftests/bpf/progs/iters_testmod_seq.c        | 79 ++++++++++++++++++++++
 5 files changed, 169 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/bpf/progs/iters_testmod_seq.c

diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
index a02a085e7f32..34cb8b2de8ca 100644
--- a/tools/testing/selftests/bpf/DENYLIST.s390x
+++ b/tools/testing/selftests/bpf/DENYLIST.s390x
@@ -8,6 +8,7 @@ dynptr/test_dynptr_skb_data
 dynptr/test_skb_readonly
 fexit_sleep                              # fexit_skel_load fexit skeleton failed                                       (trampoline)
 get_stack_raw_tp                         # user_stack corrupted user stack                                             (no backchain userspace)
+iters/testmod_seq*                       # s390x doesn't support kfuncs in modules yet
 kprobe_multi_bench_attach                # bpf_program__attach_kprobe_multi_opts unexpected error: -95
 kprobe_multi_test                        # relies on fentry
 ksyms_module                             # test_ksyms_module__open_and_load unexpected error: -9                       (?)
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 46500636d8cd..5e6e85c8d77d 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -65,6 +65,34 @@ bpf_testmod_test_mod_kfunc(int i)
 	*(int *)this_cpu_ptr(&bpf_testmod_ksym_percpu) = i;
 }
 
+__bpf_kfunc int bpf_iter_testmod_seq_new(struct bpf_iter_testmod_seq *it, s64 value, int cnt)
+{
+	if (cnt < 0) {
+		it->cnt = 0;
+		return -EINVAL;
+	}
+
+	it->value = value;
+	it->cnt = cnt;
+
+	return 0;
+}
+
+__bpf_kfunc s64 *bpf_iter_testmod_seq_next(struct bpf_iter_testmod_seq* it)
+{
+	if (it->cnt <= 0)
+		return NULL;
+
+	it->cnt--;
+
+	return &it->value;
+}
+
+__bpf_kfunc void bpf_iter_testmod_seq_destroy(struct bpf_iter_testmod_seq *it)
+{
+	it->cnt = 0;
+}
+
 struct bpf_testmod_btf_type_tag_1 {
 	int a;
 };
@@ -220,6 +248,17 @@ static struct bin_attribute bin_attr_bpf_testmod_file __ro_after_init = {
 	.write = bpf_testmod_test_write,
 };
 
+BTF_SET8_START(bpf_testmod_common_kfunc_ids)
+BTF_ID_FLAGS(func, bpf_iter_testmod_seq_new, KF_ITER_NEW)
+BTF_ID_FLAGS(func, bpf_iter_testmod_seq_next, KF_ITER_NEXT | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_iter_testmod_seq_destroy, KF_ITER_DESTROY)
+BTF_SET8_END(bpf_testmod_common_kfunc_ids)
+
+static const struct btf_kfunc_id_set bpf_testmod_common_kfunc_set = {
+	.owner = THIS_MODULE,
+	.set   = &bpf_testmod_common_kfunc_ids,
+};
+
 BTF_SET8_START(bpf_testmod_check_kfunc_ids)
 BTF_ID_FLAGS(func, bpf_testmod_test_mod_kfunc)
 BTF_SET8_END(bpf_testmod_check_kfunc_ids)
@@ -235,7 +274,8 @@ static int bpf_testmod_init(void)
 {
 	int ret;
 
-	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, &bpf_testmod_kfunc_set);
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_UNSPEC, &bpf_testmod_common_kfunc_set);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, &bpf_testmod_kfunc_set);
 	if (ret < 0)
 		return ret;
 	if (bpf_fentry_test1(0) < 0)
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.h b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.h
index 0d71e2607832..f32793efe095 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.h
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.h
@@ -22,4 +22,10 @@ struct bpf_testmod_test_writable_ctx {
 	int val;
 };
 
+/* BPF iter that returns *value* *n* times in a row */
+struct bpf_iter_testmod_seq {
+	s64 value;
+	int cnt;
+};
+
 #endif /* _BPF_TESTMOD_H */
diff --git a/tools/testing/selftests/bpf/prog_tests/iters.c b/tools/testing/selftests/bpf/prog_tests/iters.c
index 2e7caff9523e..10804ae5ae97 100644
--- a/tools/testing/selftests/bpf/prog_tests/iters.c
+++ b/tools/testing/selftests/bpf/prog_tests/iters.c
@@ -7,6 +7,7 @@
 #include "iters_state_safety.skel.h"
 #include "iters_looping.skel.h"
 #include "iters_num.skel.h"
+#include "iters_testmod_seq.skel.h"
 
 static void subtest_num_iters(void)
 {
@@ -53,12 +54,53 @@ cleanup:
 	iters_num__destroy(skel);
 }
 
+static void subtest_testmod_seq_iters(void)
+{
+	struct iters_testmod_seq *skel;
+	int err;
+
+	if (!env.has_testmod) {
+		test__skip();
+		return;
+	}
+
+	skel = iters_testmod_seq__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel_open_and_load"))
+		return;
+
+	err = iters_testmod_seq__attach(skel);
+	if (!ASSERT_OK(err, "skel_attach"))
+		goto cleanup;
+
+	usleep(1);
+	iters_testmod_seq__detach(skel);
+
+#define VALIDATE_CASE(case_name)					\
+	ASSERT_EQ(skel->bss->res_##case_name,				\
+		  skel->rodata->exp_##case_name,			\
+		  #case_name)
+
+	VALIDATE_CASE(empty);
+	VALIDATE_CASE(full);
+	VALIDATE_CASE(truncated);
+
+#undef VALIDATE_CASE
+
+cleanup:
+	iters_testmod_seq__destroy(skel);
+}
+
 void test_iters(void)
 {
 	RUN_TESTS(iters_state_safety);
 	RUN_TESTS(iters_looping);
 	RUN_TESTS(iters);
 
+	if (env.has_testmod)
+		RUN_TESTS(iters_testmod_seq);
+
 	if (test__start_subtest("num"))
 		subtest_num_iters();
+	if (test__start_subtest("testmod_seq"))
+		subtest_testmod_seq_iters();
 }
diff --git a/tools/testing/selftests/bpf/progs/iters_testmod_seq.c b/tools/testing/selftests/bpf/progs/iters_testmod_seq.c
new file mode 100644
index 000000000000..3873fb6c292a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/iters_testmod_seq.c
@@ -0,0 +1,79 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct bpf_iter_testmod_seq {
+	u64 :64;
+	u64 :64;
+};
+
+extern int bpf_iter_testmod_seq_new(struct bpf_iter_testmod_seq *it, s64 value, int cnt) __ksym;
+extern s64 *bpf_iter_testmod_seq_next(struct bpf_iter_testmod_seq *it) __ksym;
+extern void bpf_iter_testmod_seq_destroy(struct bpf_iter_testmod_seq *it) __ksym;
+
+const volatile __s64 exp_empty = 0 + 1;
+__s64 res_empty;
+
+SEC("raw_tp/sys_enter")
+__success __log_level(2)
+__msg("fp-16_w=iter_testmod_seq(ref_id=1,state=active,depth=0)")
+__msg("fp-16=iter_testmod_seq(ref_id=1,state=drained,depth=0)")
+__msg("call bpf_iter_testmod_seq_destroy")
+int testmod_seq_empty(const void *ctx)
+{
+	__s64 sum = 0, *i;
+
+	bpf_for_each(testmod_seq, i, 1000, 0) sum += *i;
+	res_empty = 1 + sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_full = 1000000;
+__s64 res_full;
+
+SEC("raw_tp/sys_enter")
+__success __log_level(2)
+__msg("fp-16_w=iter_testmod_seq(ref_id=1,state=active,depth=0)")
+__msg("fp-16=iter_testmod_seq(ref_id=1,state=drained,depth=0)")
+__msg("call bpf_iter_testmod_seq_destroy")
+int testmod_seq_full(const void *ctx)
+{
+	__s64 sum = 0, *i;
+
+	bpf_for_each(testmod_seq, i, 1000, 1000) sum += *i;
+	res_full = sum;
+
+	return 0;
+}
+
+const volatile __s64 exp_truncated = 10 * 1000000;
+__s64 res_truncated;
+
+static volatile int zero = 0;
+
+SEC("raw_tp/sys_enter")
+__success __log_level(2)
+__msg("fp-16_w=iter_testmod_seq(ref_id=1,state=active,depth=0)")
+__msg("fp-16=iter_testmod_seq(ref_id=1,state=drained,depth=0)")
+__msg("call bpf_iter_testmod_seq_destroy")
+int testmod_seq_truncated(const void *ctx)
+{
+	__s64 sum = 0, *i;
+	int cnt = zero;
+
+	bpf_for_each(testmod_seq, i, 10, 2000000) {
+		sum += *i;
+		cnt++;
+		if (cnt >= 1000000)
+			break;
+	}
+	res_truncated = sum;
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 6bf0ad7f29173869de6a5a359554426a7127c247 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 7 Mar 2023 17:30:35 +0100
Subject: ravb: remove R-Car H3 ES1.* handling

R-Car H3 ES1.* was only available to an internal development group and
needed a lot of quirks and workarounds. These become a maintenance
burden now, so our development group decided to remove upstream support
and disable booting for this SoC. Public users only have ES2 onwards.

Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://lore.kernel.org/all/20230307163041.3815-8-wsa+renesas@sang-engineering.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/renesas/ravb_main.c | 15 ---------------
 1 file changed, 15 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 0f54849a3823..b81f0d8dfda8 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -28,7 +28,6 @@
 #include <linux/pm_runtime.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
-#include <linux/sys_soc.h>
 #include <linux/reset.h>
 #include <linux/math64.h>
 
@@ -1390,11 +1389,6 @@ static void ravb_adjust_link(struct net_device *ndev)
 		phy_print_status(phydev);
 }
 
-static const struct soc_device_attribute r8a7795es10[] = {
-	{ .soc_id = "r8a7795", .revision = "ES1.0", },
-	{ /* sentinel */ }
-};
-
 /* PHY init function */
 static int ravb_phy_init(struct net_device *ndev)
 {
@@ -1434,15 +1428,6 @@ static int ravb_phy_init(struct net_device *ndev)
 		goto err_deregister_fixed_link;
 	}
 
-	/* This driver only support 10/100Mbit speeds on R-Car H3 ES1.0
-	 * at this time.
-	 */
-	if (soc_device_match(r8a7795es10)) {
-		phy_set_max_speed(phydev, SPEED_100);
-
-		netdev_info(ndev, "limited PHY to 100Mbit/s\n");
-	}
-
 	if (!info->half_duplex) {
 		/* 10BASE, Pause and Asym Pause is not supported */
 		phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_10baseT_Half_BIT);
-- 
cgit v1.2.3-70-g09d2


From 1de2a84dd06091cee943ab796660ba64c8479d33 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:12 -0600
Subject: alx: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Chris Snook <chris.snook@gmail.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/atheros/alx/main.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index 306393f8eeca..49bb9a8f00e6 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -39,7 +39,6 @@
 #include <linux/ipv6.h>
 #include <linux/if_vlan.h>
 #include <linux/mdio.h>
-#include <linux/aer.h>
 #include <linux/bitops.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
@@ -1745,7 +1744,6 @@ static int alx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_pci_disable;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
 	pci_set_master(pdev);
 
 	if (!pdev->pm_cap) {
@@ -1879,7 +1877,6 @@ out_free_netdev:
 	free_netdev(netdev);
 out_pci_release:
 	pci_release_mem_regions(pdev);
-	pci_disable_pcie_error_reporting(pdev);
 out_pci_disable:
 	pci_disable_device(pdev);
 	return err;
@@ -1897,7 +1894,6 @@ static void alx_remove(struct pci_dev *pdev)
 	iounmap(hw->hw_addr);
 	pci_release_mem_regions(pdev);
 
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
 
 	mutex_destroy(&alx->mtx);
-- 
cgit v1.2.3-70-g09d2


From b4e24578b4841b6abac38fd40fddba6caa9a656a Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:13 -0600
Subject: be2net: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration,  so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Ajit Khaparde <ajit.khaparde@broadcom.com>
Cc: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Cc: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/emulex/benet/be_main.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 46fe3d74e2e9..aed1b622f51f 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -16,7 +16,6 @@
 #include "be.h"
 #include "be_cmds.h"
 #include <asm/div64.h>
-#include <linux/aer.h>
 #include <linux/if_bridge.h>
 #include <net/busy_poll.h>
 #include <net/vxlan.h>
@@ -5726,8 +5725,6 @@ static void be_remove(struct pci_dev *pdev)
 	be_unmap_pci_bars(adapter);
 	be_drv_cleanup(adapter);
 
-	pci_disable_pcie_error_reporting(pdev);
-
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 
@@ -5845,10 +5842,6 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id)
 		goto free_netdev;
 	}
 
-	status = pci_enable_pcie_error_reporting(pdev);
-	if (!status)
-		dev_info(&pdev->dev, "PCIe error reporting enabled\n");
-
 	status = be_map_pci_bars(adapter);
 	if (status)
 		goto free_netdev;
@@ -5893,7 +5886,6 @@ drv_cleanup:
 unmap_bars:
 	be_unmap_pci_bars(adapter);
 free_netdev:
-	pci_disable_pcie_error_reporting(pdev);
 	free_netdev(netdev);
 rel_reg:
 	pci_release_regions(pdev);
-- 
cgit v1.2.3-70-g09d2


From 5f00358b5e905a1db1077006d01cd5d207f91fb0 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:14 -0600
Subject: bnx2: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

cd709aa90648 ("bnx2: Add PCI Advanced Error Reporting support.") added
pci_enable_pcie_error_reporting() for all devices, and c239f279e571 ("bnx2:
Enable AER on PCIE devices only") restricted it to BNX2_CHIP_5709 devices
to avoid an error message when it failed on non-PCIe devices.  The PCI core
only enables PCIe error reporting on PCIe devices, which I assume means
BNX2_CHIP_5709.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Rasesh Mody <rmody@marvell.com>
Cc: GR-Linux-NIC-Dev@marvell.com
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnx2.c | 21 ---------------------
 drivers/net/ethernet/broadcom/bnx2.h |  1 -
 2 files changed, 22 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index 9f473854b0f4..a66137b8d1a6 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -48,7 +48,6 @@
 #include <linux/cache.h>
 #include <linux/firmware.h>
 #include <linux/log2.h>
-#include <linux/aer.h>
 #include <linux/crash_dump.h>
 
 #if IS_ENABLED(CONFIG_CNIC)
@@ -8093,7 +8092,6 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 	int rc, i, j;
 	u32 reg;
 	u64 dma_mask, persist_dma_mask;
-	int err;
 
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	bp = netdev_priv(dev);
@@ -8176,12 +8174,6 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		bp->flags |= BNX2_FLAG_PCIE;
 		if (BNX2_CHIP_REV(bp) == BNX2_CHIP_REV_Ax)
 			bp->flags |= BNX2_FLAG_JUMBO_BROKEN;
-
-		/* AER (Advanced Error Reporting) hooks */
-		err = pci_enable_pcie_error_reporting(pdev);
-		if (!err)
-			bp->flags |= BNX2_FLAG_AER_ENABLED;
-
 	} else {
 		bp->pcix_cap = pci_find_capability(pdev, PCI_CAP_ID_PCIX);
 		if (bp->pcix_cap == 0) {
@@ -8460,11 +8452,6 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 	return 0;
 
 err_out_unmap:
-	if (bp->flags & BNX2_FLAG_AER_ENABLED) {
-		pci_disable_pcie_error_reporting(pdev);
-		bp->flags &= ~BNX2_FLAG_AER_ENABLED;
-	}
-
 	pci_iounmap(pdev, bp->regview);
 	bp->regview = NULL;
 
@@ -8638,11 +8625,6 @@ bnx2_remove_one(struct pci_dev *pdev)
 	bnx2_free_stats_blk(dev);
 	kfree(bp->temp_stats_blk);
 
-	if (bp->flags & BNX2_FLAG_AER_ENABLED) {
-		pci_disable_pcie_error_reporting(pdev);
-		bp->flags &= ~BNX2_FLAG_AER_ENABLED;
-	}
-
 	bnx2_release_firmware(bp);
 
 	free_netdev(dev);
@@ -8766,9 +8748,6 @@ static pci_ers_result_t bnx2_io_slot_reset(struct pci_dev *pdev)
 	}
 	rtnl_unlock();
 
-	if (!(bp->flags & BNX2_FLAG_AER_ENABLED))
-		return result;
-
 	return result;
 }
 
diff --git a/drivers/net/ethernet/broadcom/bnx2.h b/drivers/net/ethernet/broadcom/bnx2.h
index a09ec47461c9..315b08c64edd 100644
--- a/drivers/net/ethernet/broadcom/bnx2.h
+++ b/drivers/net/ethernet/broadcom/bnx2.h
@@ -6808,7 +6808,6 @@ struct bnx2 {
 #define BNX2_FLAG_JUMBO_BROKEN		0x00000800
 #define BNX2_FLAG_CAN_KEEP_VLAN		0x00001000
 #define BNX2_FLAG_BROKEN_STATS		0x00002000
-#define BNX2_FLAG_AER_ENABLED		0x00004000
 
 	struct bnx2_napi	bnx2_napi[BNX2_MAX_MSIX_VEC];
 
-- 
cgit v1.2.3-70-g09d2


From 2fba753cc9b57d82800db0997b9271898323c57e Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:15 -0600
Subject: bnx2x: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Ariel Elior <aelior@marvell.com>
Cc: Sudarsana Kalluru <skalluru@marvell.com>
Cc: Manish Chopra <manishc@marvell.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x.h      |  1 -
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 19 -------------------
 2 files changed, 20 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index dd5945c4bfec..8bcde0a6e011 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -1486,7 +1486,6 @@ struct bnx2x {
 #define IS_VF_FLAG			(1 << 22)
 #define BC_SUPPORTS_RMMOD_CMD		(1 << 23)
 #define HAS_PHYS_PORT_ID		(1 << 24)
-#define AER_ENABLED			(1 << 25)
 #define PTP_SUPPORTED			(1 << 26)
 #define TX_TIMESTAMPING_EN		(1 << 27)
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 5d1e4fe335aa..3bb5ea570c87 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -29,7 +29,6 @@
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/init.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
@@ -13037,14 +13036,6 @@ static const struct net_device_ops bnx2x_netdev_ops = {
 	.ndo_features_check	= bnx2x_features_check,
 };
 
-static void bnx2x_disable_pcie_error_reporting(struct bnx2x *bp)
-{
-	if (bp->flags & AER_ENABLED) {
-		pci_disable_pcie_error_reporting(bp->pdev);
-		bp->flags &= ~AER_ENABLED;
-	}
-}
-
 static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 			  struct net_device *dev, unsigned long board_type)
 {
@@ -13157,13 +13148,6 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 	/* Set PCIe reset type to fundamental for EEH recovery */
 	pdev->needs_freset = 1;
 
-	/* AER (Advanced Error reporting) configuration */
-	rc = pci_enable_pcie_error_reporting(pdev);
-	if (!rc)
-		bp->flags |= AER_ENABLED;
-	else
-		BNX2X_DEV_INFO("Failed To configure PCIe AER [%d]\n", rc);
-
 	/*
 	 * Clean the following indirect addresses for all functions since it
 	 * is not used by the driver.
@@ -14020,8 +14004,6 @@ init_one_freemem:
 	bnx2x_free_mem_bp(bp);
 
 init_one_exit:
-	bnx2x_disable_pcie_error_reporting(bp);
-
 	if (bp->regview)
 		iounmap(bp->regview);
 
@@ -14102,7 +14084,6 @@ static void __bnx2x_remove(struct pci_dev *pdev,
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
 
-	bnx2x_disable_pcie_error_reporting(bp);
 	if (remove_netdev) {
 		if (bp->regview)
 			iounmap(bp->regview);
-- 
cgit v1.2.3-70-g09d2


From 5f29b73d4eba2926ab99d7bf5f2028810af3c66c Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:16 -0600
Subject: bnxt: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 5d4b1f2ebeac..7245fee13ad0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -48,7 +48,6 @@
 #include <linux/prefetch.h>
 #include <linux/cache.h>
 #include <linux/log2.h>
-#include <linux/aer.h>
 #include <linux/bitmap.h>
 #include <linux/cpu_rmap.h>
 #include <linux/cpumask.h>
@@ -12706,8 +12705,6 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 		goto init_err_release;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
-
 	INIT_WORK(&bp->sp_task, bnxt_sp_task);
 	INIT_DELAYED_WORK(&bp->fw_reset_task, bnxt_fw_reset_task);
 
@@ -13187,7 +13184,6 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 	bnxt_rdma_aux_device_uninit(bp);
 
 	bnxt_ptp_clear(bp);
-	pci_disable_pcie_error_reporting(pdev);
 	unregister_netdev(dev);
 	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 	/* Flush any pending tasks */
-- 
cgit v1.2.3-70-g09d2


From ca7f175fc24eea4638f93a4a49d229c34ae0c770 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:17 -0600
Subject: cxgb4: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Raju Rangoju <rajur@chelsio.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 7db2403c4c9c..f0bc7396ce2b 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -51,7 +51,6 @@
 #include <linux/mutex.h>
 #include <linux/netdevice.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/rtnetlink.h>
 #include <linux/sched.h>
 #include <linux/seq_file.h>
@@ -6687,7 +6686,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_free_adapter;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
 	pci_set_master(pdev);
 	pci_save_state(pdev);
 	adap_idx++;
@@ -7092,7 +7090,6 @@ fw_attach_fail:
  out_unmap_bar0:
 	iounmap(regs);
  out_disable_device:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
  out_release_regions:
 	pci_release_regions(pdev);
@@ -7171,7 +7168,6 @@ static void remove_one(struct pci_dev *pdev)
 	}
 #endif
 	iounmap(adapter->regs);
-	pci_disable_pcie_error_reporting(pdev);
 	if ((adapter->flags & CXGB4_DEV_ENABLED)) {
 		pci_disable_device(pdev);
 		adapter->flags &= ~CXGB4_DEV_ENABLED;
-- 
cgit v1.2.3-70-g09d2


From 49f79ac22f89476cb2eac018676490e33a7c4a10 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:18 -0600
Subject: net/fungible: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Dimitris Michailidis <dmichail@fungible.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/fungible/funcore/fun_dev.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/fungible/funcore/fun_dev.c b/drivers/net/ethernet/fungible/funcore/fun_dev.c
index fb5120d90f26..3680f83feba2 100644
--- a/drivers/net/ethernet/fungible/funcore/fun_dev.c
+++ b/drivers/net/ethernet/fungible/funcore/fun_dev.c
@@ -1,6 +1,5 @@
 // SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
 
-#include <linux/aer.h>
 #include <linux/bitmap.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
@@ -748,7 +747,6 @@ void fun_dev_disable(struct fun_dev *fdev)
 	pci_free_irq_vectors(pdev);
 
 	pci_clear_master(pdev);
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
 
 	fun_unmap_bars(fdev);
@@ -781,8 +779,6 @@ int fun_dev_enable(struct fun_dev *fdev, struct pci_dev *pdev,
 		goto unmap;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
-
 	rc = sanitize_dev(fdev);
 	if (rc)
 		goto disable_dev;
@@ -830,7 +826,6 @@ free_irq_mgr:
 free_irqs:
 	pci_free_irq_vectors(pdev);
 disable_dev:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
 unmap:
 	fun_unmap_bars(fdev);
-- 
cgit v1.2.3-70-g09d2


From c183033f631a9265e5f85827b1cfcdb8d5a7ee35 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:19 -0600
Subject: net: hns3: remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Yisen Zhuang <yisen.zhuang@huawei.com>
Cc: Salil Mehta <salil.mehta@huawei.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 25be7f8ac7cd..5caea154362f 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -13,7 +13,6 @@
 #include <linux/ipv6.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/skbuff.h>
 #include <linux/sctp.h>
 #include <net/gre.h>
-- 
cgit v1.2.3-70-g09d2


From 2d0e0372069d3050479d62d48f3b04445f60083a Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:20 -0600
Subject: netxen_nic: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Also note that the driver only called these for NX_IS_REVISION_P3 devices,
so since f26e58bf6f54, error reporting has been enabled for devices other
than NX_IS_REVISION_P3.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Manish Chopra <manishc@marvell.com>
Cc: Rahul Verma <rahulv@marvell.com>
Cc: GR-Linux-NIC-Dev@marvell.com
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index de8d54b23f73..59d0dd862fd1 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -18,7 +18,6 @@
 #include <linux/ipv6.h>
 #include <linux/inetdevice.h>
 #include <linux/sysfs.h>
-#include <linux/aer.h>
 
 MODULE_DESCRIPTION("QLogic/NetXen (1/10) GbE Intelligent Ethernet Driver");
 MODULE_LICENSE("GPL");
@@ -1464,9 +1463,6 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if ((err = pci_request_regions(pdev, netxen_nic_driver_name)))
 		goto err_out_disable_pdev;
 
-	if (NX_IS_REVISION_P3(pdev->revision))
-		pci_enable_pcie_error_reporting(pdev);
-
 	pci_set_master(pdev);
 
 	netdev = alloc_etherdev(sizeof(struct netxen_adapter));
@@ -1603,8 +1599,6 @@ err_out_free_netdev:
 	free_netdev(netdev);
 
 err_out_free_res:
-	if (NX_IS_REVISION_P3(pdev->revision))
-		pci_disable_pcie_error_reporting(pdev);
 	pci_release_regions(pdev);
 
 err_out_disable_pdev:
@@ -1659,10 +1653,8 @@ static void netxen_nic_remove(struct pci_dev *pdev)
 
 	netxen_release_firmware(adapter);
 
-	if (NX_IS_REVISION_P3(pdev->revision)) {
+	if (NX_IS_REVISION_P3(pdev->revision))
 		netxen_cleanup_minidump(adapter);
-		pci_disable_pcie_error_reporting(pdev);
-	}
 
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-70-g09d2


From fe3f4c292da1908eda5d322e60b3c09ed9508288 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:21 -0600
Subject: octeon_ep: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Veerasenareddy Burru <vburru@marvell.com>
Cc: Abhijit Ayarekar <aayarekar@marvell.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/octeon_ep/octep_main.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index 5a898fb88e37..fdce78ceea87 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -8,7 +8,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/rtnetlink.h>
@@ -1050,7 +1049,6 @@ static int octep_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_pci_regions;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
 	pci_set_master(pdev);
 
 	netdev = alloc_etherdev_mq(sizeof(struct octep_device),
@@ -1106,7 +1104,6 @@ register_dev_err:
 err_octep_config:
 	free_netdev(netdev);
 err_alloc_netdev:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_release_mem_regions(pdev);
 err_pci_regions:
 err_dma_mask:
@@ -1139,7 +1136,6 @@ static void octep_remove(struct pci_dev *pdev)
 	octep_device_cleanup(oct);
 	pci_release_mem_regions(pdev);
 	free_netdev(netdev);
-	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 1263c7b78315f6682a106d56e9878b3a7b176660 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:22 -0600
Subject: qed: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Ariel Elior <aelior@marvell.com>
Cc: Manish Chopra <manishc@marvell.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/qed/qed_main.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index c91898be7c03..f5af83342856 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -23,7 +23,6 @@
 #include <linux/qed/qed_if.h>
 #include <linux/qed/qed_ll2_if.h>
 #include <net/devlink.h>
-#include <linux/aer.h>
 #include <linux/phylink.h>
 
 #include "qed.h"
@@ -259,8 +258,6 @@ static void qed_free_pci(struct qed_dev *cdev)
 {
 	struct pci_dev *pdev = cdev->pdev;
 
-	pci_disable_pcie_error_reporting(pdev);
-
 	if (cdev->doorbells && cdev->db_size)
 		iounmap(cdev->doorbells);
 	if (cdev->regview)
@@ -366,12 +363,6 @@ static int qed_init_pci(struct qed_dev *cdev, struct pci_dev *pdev)
 		return -ENOMEM;
 	}
 
-	/* AER (Advanced Error reporting) configuration */
-	rc = pci_enable_pcie_error_reporting(pdev);
-	if (rc)
-		DP_VERBOSE(cdev, NETIF_MSG_DRV,
-			   "Failed to configure PCIe AER [%d]\n", rc);
-
 	return 0;
 
 err2:
-- 
cgit v1.2.3-70-g09d2


From 5f1fbdc168f4bf9705b81a8844d0f8ca201a7363 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:23 -0600
Subject: net: qede: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Ariel Elior <aelior@marvell.com>
Cc: Manish Chopra <manishc@marvell.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/qede/qede_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 261f982ca40d..4c6c685820e3 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -35,7 +35,6 @@
 #include <net/ip6_checksum.h>
 #include <linux/bitops.h>
 #include <linux/vmalloc.h>
-#include <linux/aer.h>
 #include "qede.h"
 #include "qede_ptp.h"
 
-- 
cgit v1.2.3-70-g09d2


From 95e35f5994075090ed086ad3bcef33878218784e Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:24 -0600
Subject: qlcnic: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Shahed Shaikh <shshaikh@marvell.com>
Cc: Manish Chopra <manishc@marvell.com>
Cc: GR-Linux-NIC-Dev@marvell.com
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c  | 4 ----
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c | 1 -
 2 files changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 44dac3c0908e..90df4a0909fa 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -12,7 +12,6 @@
 #include <net/ip.h>
 #include <linux/ipv6.h>
 #include <linux/inetdevice.h>
-#include <linux/aer.h>
 #include <linux/log2.h>
 #include <linux/pci.h>
 #include <net/vxlan.h>
@@ -2445,7 +2444,6 @@ qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_out_disable_pdev;
 
 	pci_set_master(pdev);
-	pci_enable_pcie_error_reporting(pdev);
 
 	ahw = kzalloc(sizeof(struct qlcnic_hardware_context), GFP_KERNEL);
 	if (!ahw) {
@@ -2675,7 +2673,6 @@ err_out_free_hw_res:
 	kfree(ahw);
 
 err_out_free_res:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_release_regions(pdev);
 
 err_out_disable_pdev:
@@ -2757,7 +2754,6 @@ static void qlcnic_remove(struct pci_dev *pdev)
 
 	qlcnic_release_firmware(adapter);
 
-	pci_disable_pcie_error_reporting(pdev);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
index 5c2edb715d3e..74125188beb8 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
@@ -12,7 +12,6 @@
 #include <linux/ipv6.h>
 #include <linux/inetdevice.h>
 #include <linux/sysfs.h>
-#include <linux/aer.h>
 #include <linux/log2.h>
 #ifdef CONFIG_QLCNIC_HWMON
 #include <linux/hwmon.h>
-- 
cgit v1.2.3-70-g09d2


From e07ce5567194dd14f4fcbbaac4a1c48bb97da8e3 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:25 -0600
Subject: qlcnic: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Shahed Shaikh <shshaikh@marvell.com>
Cc: Manish Chopra <manishc@marvell.com>
Cc: GR-Linux-NIC-Dev@marvell.com
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
index 2fd5c6fdb500..bcef8ab715bf 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
@@ -8,7 +8,6 @@
 #include <linux/ipv6.h>
 #include <linux/ethtool.h>
 #include <linux/interrupt.h>
-#include <linux/aer.h>
 
 #include "qlcnic.h"
 #include "qlcnic_sriov.h"
-- 
cgit v1.2.3-70-g09d2


From bdedf705688c7704f2efc3ce4930a32612e4efdc Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:26 -0600
Subject: sfc: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Martin Habets <habetsm.xilinx@gmail.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/efx.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 02c2adeb0a12..92c390ec4735 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -18,7 +18,6 @@
 #include <linux/ethtool.h>
 #include <linux/topology.h>
 #include <linux/gfp.h>
-#include <linux/aer.h>
 #include <linux/interrupt.h>
 #include "net_driver.h"
 #include <net/gre.h>
@@ -892,8 +891,6 @@ static void efx_pci_remove(struct pci_dev *pci_dev)
 	free_netdev(efx->net_dev);
 	probe_data = container_of(efx, struct efx_probe_data, efx);
 	kfree(probe_data);
-
-	pci_disable_pcie_error_reporting(pci_dev);
 };
 
 /* NIC VPD information
@@ -1126,8 +1123,6 @@ static int efx_pci_probe(struct pci_dev *pci_dev,
 		netif_warn(efx, probe, efx->net_dev,
 			   "failed to create MTDs (%d)\n", rc);
 
-	(void)pci_enable_pcie_error_reporting(pci_dev);
-
 	if (efx->type->udp_tnl_push_ports)
 		efx->type->udp_tnl_push_ports(efx);
 
-- 
cgit v1.2.3-70-g09d2


From 4ac9272691a402d5e862be627bac54d2cd6dacbc Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:27 -0600
Subject: sfc: falcon: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Martin Habets <habetsm.xilinx@gmail.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/falcon/efx.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/net/ethernet/sfc/falcon/efx.c b/drivers/net/ethernet/sfc/falcon/efx.c
index e151b0957751..e001f27085c6 100644
--- a/drivers/net/ethernet/sfc/falcon/efx.c
+++ b/drivers/net/ethernet/sfc/falcon/efx.c
@@ -17,7 +17,6 @@
 #include <linux/ethtool.h>
 #include <linux/topology.h>
 #include <linux/gfp.h>
-#include <linux/aer.h>
 #include <linux/interrupt.h>
 #include "net_driver.h"
 #include "efx.h"
@@ -2765,8 +2764,6 @@ static void ef4_pci_remove(struct pci_dev *pci_dev)
 
 	ef4_fini_struct(efx);
 	free_netdev(efx->net_dev);
-
-	pci_disable_pcie_error_reporting(pci_dev);
 };
 
 /* NIC VPD information
@@ -2927,12 +2924,6 @@ static int ef4_pci_probe(struct pci_dev *pci_dev,
 		netif_warn(efx, probe, efx->net_dev,
 			   "failed to create MTDs (%d)\n", rc);
 
-	rc = pci_enable_pcie_error_reporting(pci_dev);
-	if (rc && rc != -EINVAL)
-		netif_notice(efx, probe, efx->net_dev,
-			     "PCIE error reporting unavailable (%d).\n",
-			     rc);
-
 	return 0;
 
  fail4:
-- 
cgit v1.2.3-70-g09d2


From ecded61ceb89880f0b0490a168d18c9845b60eb3 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:28 -0600
Subject: sfc/siena: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Martin Habets <habetsm.xilinx@gmail.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/siena/efx.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/sfc/siena/efx.c b/drivers/net/ethernet/sfc/siena/efx.c
index ef52ec71d197..8c557f6a183c 100644
--- a/drivers/net/ethernet/sfc/siena/efx.c
+++ b/drivers/net/ethernet/sfc/siena/efx.c
@@ -18,7 +18,6 @@
 #include <linux/ethtool.h>
 #include <linux/topology.h>
 #include <linux/gfp.h>
-#include <linux/aer.h>
 #include <linux/interrupt.h>
 #include "net_driver.h"
 #include <net/gre.h>
@@ -874,8 +873,6 @@ static void efx_pci_remove(struct pci_dev *pci_dev)
 
 	efx_siena_fini_struct(efx);
 	free_netdev(efx->net_dev);
-
-	pci_disable_pcie_error_reporting(pci_dev);
 };
 
 /* NIC VPD information
@@ -1094,8 +1091,6 @@ static int efx_pci_probe(struct pci_dev *pci_dev,
 		netif_warn(efx, probe, efx->net_dev,
 			   "failed to create MTDs (%d)\n", rc);
 
-	(void)pci_enable_pcie_error_reporting(pci_dev);
-
 	if (efx->type->udp_tnl_push_ports)
 		efx->type->udp_tnl_push_ports(efx);
 
-- 
cgit v1.2.3-70-g09d2


From c39abdd396bc2b299798e8422538806a8dc29d20 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:29 -0600
Subject: sfc_ef100: Drop redundant pci_disable_pcie_error_reporting()

51b35a454efd ("sfc: skeleton EF100 PF driver") added a call to
pci_disable_pcie_error_reporting() in ef100_pci_remove().

Remove this call since there's no apparent reason to disable error
reporting when it was not previously enabled.

Note that since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core enables PCIe error reporting for all devices during
enumeration, so the driver doesn't need to do it itself.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Martin Habets <habetsm.xilinx@gmail.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ef100.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef100.c b/drivers/net/ethernet/sfc/ef100.c
index 71aab3d0480f..6334992b0af4 100644
--- a/drivers/net/ethernet/sfc/ef100.c
+++ b/drivers/net/ethernet/sfc/ef100.c
@@ -11,7 +11,6 @@
 
 #include "net_driver.h"
 #include <linux/module.h>
-#include <linux/aer.h>
 #include "efx_common.h"
 #include "efx_channels.h"
 #include "io.h"
@@ -440,8 +439,6 @@ static void ef100_pci_remove(struct pci_dev *pci_dev)
 
 	pci_dbg(pci_dev, "shutdown successful\n");
 
-	pci_disable_pcie_error_reporting(pci_dev);
-
 	pci_set_drvdata(pci_dev, NULL);
 	efx_fini_struct(efx);
 	kfree(probe_data);
-- 
cgit v1.2.3-70-g09d2


From a7edf8e5142f2037db944ec87c0f2491dadfe2a6 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:30 -0600
Subject: net: ngbe: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jiawen Wu <jiawenwu@trustnetic.com>
Cc: Mengyuan Lou <mengyuanlou@net-swift.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/wangxun/ngbe/ngbe_main.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
index 5b564d348c09..0e4163e1106f 100644
--- a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
+++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
@@ -6,7 +6,6 @@
 #include <linux/pci.h>
 #include <linux/netdevice.h>
 #include <linux/string.h>
-#include <linux/aer.h>
 #include <linux/etherdevice.h>
 #include <net/ip.h>
 #include <linux/phy.h>
@@ -520,7 +519,6 @@ static int ngbe_probe(struct pci_dev *pdev,
 		goto err_pci_disable_dev;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
 	pci_set_master(pdev);
 
 	netdev = devm_alloc_etherdev_mqs(&pdev->dev,
@@ -669,7 +667,6 @@ err_clear_interrupt_scheme:
 err_free_mac_table:
 	kfree(wx->mac_table);
 err_pci_release_regions:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_release_selected_regions(pdev,
 				     pci_select_bars(pdev, IORESOURCE_MEM));
 err_pci_disable_dev:
@@ -698,7 +695,6 @@ static void ngbe_remove(struct pci_dev *pdev)
 
 	kfree(wx->mac_table);
 	wx_clear_interrupt_scheme(wx);
-	pci_disable_pcie_error_reporting(pdev);
 
 	pci_disable_device(pdev);
 }
-- 
cgit v1.2.3-70-g09d2


From 1fccc781bf7e32b89ecb289987670b040f97d7c0 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:31 -0600
Subject: net: txgbe: Drop redundant pci_enable_pcie_error_reporting()

pci_enable_pcie_error_reporting() enables the device to send ERR_*
Messages.  Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is
native"), the PCI core does this for all devices during enumeration, so the
driver doesn't need to do it itself.

Remove the redundant pci_enable_pcie_error_reporting() call from the
driver.  Also remove the corresponding pci_disable_pcie_error_reporting()
from the driver .remove() path.

Note that this only controls ERR_* Messages from the device.  An ERR_*
Message may cause the Root Port to generate an interrupt, depending on the
AER Root Error Command register managed by the AER service driver.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jiawen Wu <jiawenwu@trustnetic.com>
Cc: Mengyuan Lou <mengyuanlou@net-swift.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/wangxun/txgbe/txgbe_main.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
index 6c0a98230557..859feaafd350 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
@@ -6,7 +6,6 @@
 #include <linux/pci.h>
 #include <linux/netdevice.h>
 #include <linux/string.h>
-#include <linux/aer.h>
 #include <linux/etherdevice.h>
 #include <net/ip.h>
 
@@ -538,7 +537,6 @@ static int txgbe_probe(struct pci_dev *pdev,
 		goto err_pci_disable_dev;
 	}
 
-	pci_enable_pcie_error_reporting(pdev);
 	pci_set_master(pdev);
 
 	netdev = devm_alloc_etherdev_mqs(&pdev->dev,
@@ -698,7 +696,6 @@ err_release_hw:
 err_free_mac_table:
 	kfree(wx->mac_table);
 err_pci_release_regions:
-	pci_disable_pcie_error_reporting(pdev);
 	pci_release_selected_regions(pdev,
 				     pci_select_bars(pdev, IORESOURCE_MEM));
 err_pci_disable_dev:
@@ -729,8 +726,6 @@ static void txgbe_remove(struct pci_dev *pdev)
 	kfree(wx->mac_table);
 	wx_clear_interrupt_scheme(wx);
 
-	pci_disable_pcie_error_reporting(pdev);
-
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From ab76f2bff0f3d165428fb71532c7e6ba81df8317 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:32 -0600
Subject: e1000e: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Cc: intel-wired-lan@lists.osuosl.org
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/e1000e/netdev.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index e1eb1de88bf9..6f5c16aebcbf 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -23,7 +23,6 @@
 #include <linux/smp.h>
 #include <linux/pm_qos.h>
 #include <linux/pm_runtime.h>
-#include <linux/aer.h>
 #include <linux/prefetch.h>
 #include <linux/suspend.h>
 
-- 
cgit v1.2.3-70-g09d2


From 8be901a6715f290131ca919bef425717538382b5 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:33 -0600
Subject: fm10k: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/fm10k/fm10k_pci.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
index 027d721feb18..d748b98274e7 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
@@ -3,7 +3,6 @@
 
 #include <linux/module.h>
 #include <linux/interrupt.h>
-#include <linux/aer.h>
 
 #include "fm10k.h"
 
-- 
cgit v1.2.3-70-g09d2


From acd2bb015fae8d67ff11d11dfe250b0fb6c6081c Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:34 -0600
Subject: i40e: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/i40e/i40e.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 60ce4d15d82a..6e310a539467 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -10,7 +10,6 @@
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/netdevice.h>
 #include <linux/ioport.h>
 #include <linux/iommu.h>
-- 
cgit v1.2.3-70-g09d2


From 495b72c79302bb827ff14686000edbba1dd9e372 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:35 -0600
Subject: iavf: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/iavf/iavf.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf.h b/drivers/net/ethernet/intel/iavf/iavf.h
index 232bc61d9eee..2cdce251472c 100644
--- a/drivers/net/ethernet/intel/iavf/iavf.h
+++ b/drivers/net/ethernet/intel/iavf/iavf.h
@@ -6,7 +6,6 @@
 
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/aer.h>
 #include <linux/netdevice.h>
 #include <linux/vmalloc.h>
 #include <linux/interrupt.h>
-- 
cgit v1.2.3-70-g09d2


From ddd652ef30e396efe006494a67e43e174099b419 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:36 -0600
Subject: ice: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index b0e29e342401..d79a48d27f1a 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -20,7 +20,6 @@
 #include <linux/pci.h>
 #include <linux/workqueue.h>
 #include <linux/wait.h>
-#include <linux/aer.h>
 #include <linux/interrupt.h>
 #include <linux/ethtool.h>
 #include <linux/timer.h>
-- 
cgit v1.2.3-70-g09d2


From 648a2020fdac17591b909849513a10a65daa8b27 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:37 -0600
Subject: igb: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/igb/igb_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 03bc1e8af575..a2914298dd69 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -28,7 +28,6 @@
 #include <linux/tcp.h>
 #include <linux/sctp.h>
 #include <linux/if_ether.h>
-#include <linux/aer.h>
 #include <linux/prefetch.h>
 #include <linux/bpf.h>
 #include <linux/bpf_trace.h>
-- 
cgit v1.2.3-70-g09d2


From 1530522f101f5f58f73bb3c9f5b1be39cb1c2a62 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:38 -0600
Subject: igc: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 2928a6c73692..d7ee06d28b50 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4,7 +4,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/if_vlan.h>
-#include <linux/aer.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
 #include <linux/ip.h>
-- 
cgit v1.2.3-70-g09d2


From f3468e3944398c4de9b2deff73041394fba632e7 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 12:19:39 -0600
Subject: ixgbe: Remove unnecessary aer.h include

<linux/aer.h> is unused, so remove it.

Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 8736ca4b2628..63d4e32df029 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -9,7 +9,6 @@
 #include <linux/pci.h>
 #include <linux/netdevice.h>
 #include <linux/cpumask.h>
-#include <linux/aer.h>
 #include <linux/if_vlan.h>
 #include <linux/jiffies.h>
 #include <linux/phy.h>
-- 
cgit v1.2.3-70-g09d2


From 04eb3d1cede014defaa8c259263439cc7e9ceda6 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 7 Mar 2023 16:19:26 +0000
Subject: net: mtk_eth_soc: tidy mtk_gmac0_rgmii_adjust()

Get rid of the multiple tenary operators in mtk_gmac0_rgmii_adjust()
replacing them with a single if(), thus making the code easier to read.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 34 ++++++++++++++++-------------
 1 file changed, 19 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 14be6ea51b88..c63f17929ccf 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -397,38 +397,42 @@ static int mt7621_gmac0_rgmii_adjust(struct mtk_eth *eth,
 static void mtk_gmac0_rgmii_adjust(struct mtk_eth *eth,
 				   phy_interface_t interface, int speed)
 {
-	u32 val;
+	unsigned long rate;
+	u32 tck, rck, intf;
 	int ret;
 
 	if (interface == PHY_INTERFACE_MODE_TRGMII) {
 		mtk_w32(eth, TRGMII_MODE, INTF_MODE);
-		val = 500000000;
-		ret = clk_set_rate(eth->clks[MTK_CLK_TRGPLL], val);
+		ret = clk_set_rate(eth->clks[MTK_CLK_TRGPLL], 500000000);
 		if (ret)
 			dev_err(eth->dev, "Failed to set trgmii pll: %d\n", ret);
 		return;
 	}
 
-	val = (speed == SPEED_1000) ?
-		INTF_MODE_RGMII_1000 : INTF_MODE_RGMII_10_100;
-	mtk_w32(eth, val, INTF_MODE);
+	if (speed == SPEED_1000) {
+		intf = INTF_MODE_RGMII_1000;
+		rate = 250000000;
+		rck = RCK_CTRL_RGMII_1000;
+		tck = TCK_CTRL_RGMII_1000;
+	} else {
+		intf = INTF_MODE_RGMII_10_100;
+		rate = 500000000;
+		rck = RCK_CTRL_RGMII_10_100;
+		tck = TCK_CTRL_RGMII_10_100;
+	}
+
+	mtk_w32(eth, intf, INTF_MODE);
 
 	regmap_update_bits(eth->ethsys, ETHSYS_CLKCFG0,
 			   ETHSYS_TRGMII_CLK_SEL362_5,
 			   ETHSYS_TRGMII_CLK_SEL362_5);
 
-	val = (speed == SPEED_1000) ? 250000000 : 500000000;
-	ret = clk_set_rate(eth->clks[MTK_CLK_TRGPLL], val);
+	ret = clk_set_rate(eth->clks[MTK_CLK_TRGPLL], rate);
 	if (ret)
 		dev_err(eth->dev, "Failed to set trgmii pll: %d\n", ret);
 
-	val = (speed == SPEED_1000) ?
-		RCK_CTRL_RGMII_1000 : RCK_CTRL_RGMII_10_100;
-	mtk_w32(eth, val, TRGMII_RCK_CTRL);
-
-	val = (speed == SPEED_1000) ?
-		TCK_CTRL_RGMII_1000 : TCK_CTRL_RGMII_10_100;
-	mtk_w32(eth, val, TRGMII_TCK_CTRL);
+	mtk_w32(eth, rck, TRGMII_RCK_CTRL);
+	mtk_w32(eth, tck, TRGMII_TCK_CTRL);
 }
 
 static struct phylink_pcs *mtk_mac_select_pcs(struct phylink_config *config,
-- 
cgit v1.2.3-70-g09d2


From 7910898e1b2ae8ba0646434294931bf69cc0fdb5 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 7 Mar 2023 16:19:31 +0000
Subject: net: mtk_eth_soc: move trgmii ddr2 check to probe function

If TRGMII mode is not permitted when using DDR2 mode, we should handle
that when setting up phylink's ->supported_interfaces so phylink knows
that this is not supported by the hardware. Move this check to
mtk_add_mac().

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index c63f17929ccf..1b385dfe620f 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -374,17 +374,6 @@ static int mt7621_gmac0_rgmii_adjust(struct mtk_eth *eth,
 {
 	u32 val;
 
-	/* Check DDR memory type.
-	 * Currently TRGMII mode with DDR2 memory is not supported.
-	 */
-	regmap_read(eth->ethsys, ETHSYS_SYSCFG, &val);
-	if (interface == PHY_INTERFACE_MODE_TRGMII &&
-	    val & SYSCFG_DRAM_TYPE_DDR2) {
-		dev_err(eth->dev,
-			"TRGMII mode with DDR2 memory is not supported!\n");
-		return -EOPNOTSUPP;
-	}
-
 	val = (interface == PHY_INTERFACE_MODE_TRGMII) ?
 		ETHSYS_TRGMII_MT7621_DDR_PLL : 0;
 
@@ -4333,6 +4322,7 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np)
 	struct mtk_mac *mac;
 	int id, err;
 	int txqs = 1;
+	u32 val;
 
 	if (!_id) {
 		dev_err(eth->dev, "missing mac id\n");
@@ -4409,6 +4399,15 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np)
 		__set_bit(PHY_INTERFACE_MODE_TRGMII,
 			  mac->phylink_config.supported_interfaces);
 
+	/* TRGMII is not permitted on MT7621 if using DDR2 */
+	if (MTK_HAS_CAPS(mac->hw->soc->caps, MTK_GMAC1_TRGMII) &&
+	    MTK_HAS_CAPS(mac->hw->soc->caps, MTK_TRGMII_MT7621_CLK)) {
+		regmap_read(eth->ethsys, ETHSYS_SYSCFG, &val);
+		if (val & SYSCFG_DRAM_TYPE_DDR2)
+			__clear_bit(PHY_INTERFACE_MODE_TRGMII,
+				    mac->phylink_config.supported_interfaces);
+	}
+
 	if (MTK_HAS_CAPS(mac->hw->soc->caps, MTK_SGMII)) {
 		__set_bit(PHY_INTERFACE_MODE_SGMII,
 			  mac->phylink_config.supported_interfaces);
-- 
cgit v1.2.3-70-g09d2


From c9f9e3a3289f2f068db45ea1219aa5b295762f44 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 7 Mar 2023 16:19:36 +0000
Subject: net: mtk_eth_soc: remove unnecessary checks in mtk_mac_config()

mtk_mac_config() checks that the interface mode is permitted for the
capabilities, but we already do these checks in mtk_add_mac() when
initialising phylink's supported_interfaces bitmap. Remove the
unnecessary tests.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 1b385dfe620f..f78810717f66 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -458,12 +458,6 @@ static void mtk_mac_config(struct phylink_config *config, unsigned int mode,
 		/* Setup soc pin functions */
 		switch (state->interface) {
 		case PHY_INTERFACE_MODE_TRGMII:
-			if (mac->id)
-				goto err_phy;
-			if (!MTK_HAS_CAPS(mac->hw->soc->caps,
-					  MTK_GMAC1_TRGMII))
-				goto err_phy;
-			fallthrough;
 		case PHY_INTERFACE_MODE_RGMII_TXID:
 		case PHY_INTERFACE_MODE_RGMII_RXID:
 		case PHY_INTERFACE_MODE_RGMII_ID:
@@ -480,11 +474,9 @@ static void mtk_mac_config(struct phylink_config *config, unsigned int mode,
 		case PHY_INTERFACE_MODE_1000BASEX:
 		case PHY_INTERFACE_MODE_2500BASEX:
 		case PHY_INTERFACE_MODE_SGMII:
-			if (MTK_HAS_CAPS(eth->soc->caps, MTK_SGMII)) {
-				err = mtk_gmac_sgmii_path_setup(eth, mac->id);
-				if (err)
-					goto init_err;
-			}
+			err = mtk_gmac_sgmii_path_setup(eth, mac->id);
+			if (err)
+				goto init_err;
 			break;
 		case PHY_INTERFACE_MODE_GMII:
 			if (MTK_HAS_CAPS(eth->soc->caps, MTK_GEPHY)) {
-- 
cgit v1.2.3-70-g09d2


From 8cd9de08ccf6e1575d4994ba8347fa9ac083f900 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 7 Mar 2023 16:19:41 +0000
Subject: net: mtk_eth_soc: remove support for RMII and REVMII modes

Since the conversion of mtk_eth_soc to phylink's supported_interfaces
bitmap, these two modes have not been selectable. No one has raised
this as an issue. Checking the in-kernel DT files, none of them use
either of these modes with this hardware.

Daniel Golle concurs:

 A quick grep through the device trees of the more than 650 ramips and
 mediatek boards we support in OpenWrt has revealed that *none* of them
 uses either reduced-MII or reverse-MII PHY modes. I could imaging that
 some more specialized ramips boards may use the RMII 100M PHY mode to
 connect with exotic PHYs for industrial or automotive applications
 (think: for 100BASE-T1 PHY connected via RMII). I have never seen or
 touched such boards, but there are hints that they do exist.

 For reverse-MII there are cases in which the Ralink SoC (Rt305x, for
 example) is used in iNIC mode, ie. connected as a PHY to another SoC,
 and running only a minimal firmware rather than running Linux. Due to
 the lack of external DRAM for the Ralink SoC on this kind of boards,
 the Ralink SoC there will anyway never be able to boot Linux.
 I've seen this e.g. in multimedia devices like early WiFi-connected
 not-yet-so-smart TVs.

Consequently, the conclusion is that no one uses these modes with this
hardware, so we might as well drop support for them.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index f78810717f66..b22cd160554e 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -463,8 +463,6 @@ static void mtk_mac_config(struct phylink_config *config, unsigned int mode,
 		case PHY_INTERFACE_MODE_RGMII_ID:
 		case PHY_INTERFACE_MODE_RGMII:
 		case PHY_INTERFACE_MODE_MII:
-		case PHY_INTERFACE_MODE_REVMII:
-		case PHY_INTERFACE_MODE_RMII:
 			if (MTK_HAS_CAPS(eth->soc->caps, MTK_RGMII)) {
 				err = mtk_gmac_rgmii_path_setup(eth, mac->id);
 				if (err)
@@ -524,21 +522,13 @@ static void mtk_mac_config(struct phylink_config *config, unsigned int mode,
 			}
 		}
 
-		ge_mode = 0;
 		switch (state->interface) {
 		case PHY_INTERFACE_MODE_MII:
 		case PHY_INTERFACE_MODE_GMII:
 			ge_mode = 1;
 			break;
-		case PHY_INTERFACE_MODE_REVMII:
-			ge_mode = 2;
-			break;
-		case PHY_INTERFACE_MODE_RMII:
-			if (mac->id)
-				goto err_phy;
-			ge_mode = 3;
-			break;
 		default:
+			ge_mode = 0;
 			break;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 4821a076eb602a6238528e9ebafeac853c833415 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:23:26 -0500
Subject: sctp: add fair capacity stream scheduler

As it says in rfc8260#section-3.5 about the fair capacity scheduler:

   A fair capacity distribution between the streams is used.  This
   scheduler considers the lengths of the messages of each stream and
   schedules them in a specific way to maintain an equal capacity for
   all streams.  The details are implementation dependent.  interleaving
   user messages allows for a better realization of the fair capacity
   usage.

This patch adds Fair Capacity Scheduler based on the foundations added
by commit 5bbbbe32a431 ("sctp: introduce stream scheduler foundations"):

A fc_list and a fc_length are added into struct sctp_stream_out_ext and
a fc_list is added into struct sctp_stream. In .enqueue, when there are
chunks enqueued into a stream, this stream will be linked into stream->
fc_list by its fc_list ordered by its fc_length. In .dequeue, it always
picks up the 1st skb from stream->fc_list. In .dequeue_done, fc_length
is increased by chunk's len and update its location in stream->fc_list
according to the its new fc_length.

Note that when the new fc_length overflows in .dequeue_done, instead of
resetting all fc_lengths to 0, we only reduced them by U32_MAX / 4 to
avoid a moment of imbalance in the scheduling, as Marcelo suggested.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/sctp/stream_sched.h |   1 +
 include/net/sctp/structs.h      |   7 ++
 include/uapi/linux/sctp.h       |   3 +-
 net/sctp/Makefile               |   3 +-
 net/sctp/stream_sched.c         |   1 +
 net/sctp/stream_sched_fc.c      | 183 ++++++++++++++++++++++++++++++++++++++++
 6 files changed, 196 insertions(+), 2 deletions(-)
 create mode 100644 net/sctp/stream_sched_fc.c

diff --git a/include/net/sctp/stream_sched.h b/include/net/sctp/stream_sched.h
index fa00dc20a0d7..913170710adb 100644
--- a/include/net/sctp/stream_sched.h
+++ b/include/net/sctp/stream_sched.h
@@ -58,5 +58,6 @@ void sctp_sched_ops_register(enum sctp_sched_type sched,
 			     struct sctp_sched_ops *sched_ops);
 void sctp_sched_ops_prio_init(void);
 void sctp_sched_ops_rr_init(void);
+void sctp_sched_ops_fc_init(void);
 
 #endif /* __sctp_stream_sched_h__ */
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index e1f6e7fc2b11..2f1c9f50b352 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1429,6 +1429,10 @@ struct sctp_stream_out_ext {
 		struct {
 			struct list_head rr_list;
 		};
+		struct {
+			struct list_head fc_list;
+			__u32 fc_length;
+		};
 	};
 };
 
@@ -1475,6 +1479,9 @@ struct sctp_stream {
 			/* The next stream in line */
 			struct sctp_stream_out_ext *rr_next;
 		};
+		struct {
+			struct list_head fc_list;
+		};
 	};
 	struct sctp_stream_interleave *si;
 };
diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h
index ed7d4ecbf53d..6814c5a1c4bc 100644
--- a/include/uapi/linux/sctp.h
+++ b/include/uapi/linux/sctp.h
@@ -1211,7 +1211,8 @@ enum sctp_sched_type {
 	SCTP_SS_DEFAULT = SCTP_SS_FCFS,
 	SCTP_SS_PRIO,
 	SCTP_SS_RR,
-	SCTP_SS_MAX = SCTP_SS_RR
+	SCTP_SS_FC,
+	SCTP_SS_MAX = SCTP_SS_FC
 };
 
 /* Probe Interval socket option */
diff --git a/net/sctp/Makefile b/net/sctp/Makefile
index e845e4588535..0448398408d8 100644
--- a/net/sctp/Makefile
+++ b/net/sctp/Makefile
@@ -13,7 +13,8 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \
 	  tsnmap.o bind_addr.o socket.o primitive.o \
 	  output.o input.o debug.o stream.o auth.o \
 	  offload.o stream_sched.o stream_sched_prio.o \
-	  stream_sched_rr.o stream_interleave.o
+	  stream_sched_rr.o stream_sched_fc.o \
+	  stream_interleave.o
 
 sctp_diag-y := diag.o
 
diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c
index 330067002deb..1ebd14ef8daa 100644
--- a/net/sctp/stream_sched.c
+++ b/net/sctp/stream_sched.c
@@ -124,6 +124,7 @@ void sctp_sched_ops_init(void)
 	sctp_sched_ops_fcfs_init();
 	sctp_sched_ops_prio_init();
 	sctp_sched_ops_rr_init();
+	sctp_sched_ops_fc_init();
 }
 
 static void sctp_sched_free_sched(struct sctp_stream *stream)
diff --git a/net/sctp/stream_sched_fc.c b/net/sctp/stream_sched_fc.c
new file mode 100644
index 000000000000..b336c2f5486b
--- /dev/null
+++ b/net/sctp/stream_sched_fc.c
@@ -0,0 +1,183 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* SCTP kernel implementation
+ * (C) Copyright Red Hat Inc. 2022
+ *
+ * This file is part of the SCTP kernel implementation
+ *
+ * These functions manipulate sctp stream queue/scheduling.
+ *
+ * Please send any bug reports or fixes you make to the
+ * email addresched(es):
+ *    lksctp developers <linux-sctp@vger.kernel.org>
+ *
+ * Written or modified by:
+ *    Xin Long <lucien.xin@gmail.com>
+ */
+
+#include <linux/list.h>
+#include <net/sctp/sctp.h>
+#include <net/sctp/sm.h>
+#include <net/sctp/stream_sched.h>
+
+/* Fair Capacity handling
+ * RFC 8260 section 3.5
+ */
+static void sctp_sched_fc_unsched_all(struct sctp_stream *stream);
+
+static int sctp_sched_fc_set(struct sctp_stream *stream, __u16 sid,
+			     __u16 weight, gfp_t gfp)
+{
+	return 0;
+}
+
+static int sctp_sched_fc_get(struct sctp_stream *stream, __u16 sid,
+			     __u16 *value)
+{
+	return 0;
+}
+
+static int sctp_sched_fc_init(struct sctp_stream *stream)
+{
+	INIT_LIST_HEAD(&stream->fc_list);
+
+	return 0;
+}
+
+static int sctp_sched_fc_init_sid(struct sctp_stream *stream, __u16 sid,
+				  gfp_t gfp)
+{
+	struct sctp_stream_out_ext *soute = SCTP_SO(stream, sid)->ext;
+
+	INIT_LIST_HEAD(&soute->fc_list);
+	soute->fc_length = 0;
+
+	return 0;
+}
+
+static void sctp_sched_fc_free_sid(struct sctp_stream *stream, __u16 sid)
+{
+}
+
+static void sctp_sched_fc_sched(struct sctp_stream *stream,
+				struct sctp_stream_out_ext *soute)
+{
+	struct sctp_stream_out_ext *pos;
+
+	if (!list_empty(&soute->fc_list))
+		return;
+
+	list_for_each_entry(pos, &stream->fc_list, fc_list)
+		if (pos->fc_length >= soute->fc_length)
+			break;
+	list_add_tail(&soute->fc_list, &pos->fc_list);
+}
+
+static void sctp_sched_fc_enqueue(struct sctp_outq *q,
+				  struct sctp_datamsg *msg)
+{
+	struct sctp_stream *stream;
+	struct sctp_chunk *ch;
+	__u16 sid;
+
+	ch = list_first_entry(&msg->chunks, struct sctp_chunk, frag_list);
+	sid = sctp_chunk_stream_no(ch);
+	stream = &q->asoc->stream;
+	sctp_sched_fc_sched(stream, SCTP_SO(stream, sid)->ext);
+}
+
+static struct sctp_chunk *sctp_sched_fc_dequeue(struct sctp_outq *q)
+{
+	struct sctp_stream *stream = &q->asoc->stream;
+	struct sctp_stream_out_ext *soute;
+	struct sctp_chunk *ch;
+
+	/* Bail out quickly if queue is empty */
+	if (list_empty(&q->out_chunk_list))
+		return NULL;
+
+	/* Find which chunk is next */
+	if (stream->out_curr)
+		soute = stream->out_curr->ext;
+	else
+		soute = list_entry(stream->fc_list.next, struct sctp_stream_out_ext, fc_list);
+	ch = list_entry(soute->outq.next, struct sctp_chunk, stream_list);
+
+	sctp_sched_dequeue_common(q, ch);
+	return ch;
+}
+
+static void sctp_sched_fc_dequeue_done(struct sctp_outq *q,
+				       struct sctp_chunk *ch)
+{
+	struct sctp_stream *stream = &q->asoc->stream;
+	struct sctp_stream_out_ext *soute, *pos;
+	__u16 sid, i;
+
+	sid = sctp_chunk_stream_no(ch);
+	soute = SCTP_SO(stream, sid)->ext;
+	/* reduce all fc_lengths by U32_MAX / 4 if the current fc_length overflows. */
+	if (soute->fc_length > U32_MAX - ch->skb->len) {
+		for (i = 0; i < stream->outcnt; i++) {
+			pos = SCTP_SO(stream, i)->ext;
+			if (!pos)
+				continue;
+			if (pos->fc_length <= (U32_MAX >> 2)) {
+				pos->fc_length = 0;
+				continue;
+			}
+			pos->fc_length -= (U32_MAX >> 2);
+		}
+	}
+	soute->fc_length += ch->skb->len;
+
+	if (list_empty(&soute->outq)) {
+		list_del_init(&soute->fc_list);
+		return;
+	}
+
+	pos = soute;
+	list_for_each_entry_continue(pos, &stream->fc_list, fc_list)
+		if (pos->fc_length >= soute->fc_length)
+			break;
+	list_move_tail(&soute->fc_list, &pos->fc_list);
+}
+
+static void sctp_sched_fc_sched_all(struct sctp_stream *stream)
+{
+	struct sctp_association *asoc;
+	struct sctp_chunk *ch;
+
+	asoc = container_of(stream, struct sctp_association, stream);
+	list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) {
+		__u16 sid = sctp_chunk_stream_no(ch);
+
+		if (SCTP_SO(stream, sid)->ext)
+			sctp_sched_fc_sched(stream, SCTP_SO(stream, sid)->ext);
+	}
+}
+
+static void sctp_sched_fc_unsched_all(struct sctp_stream *stream)
+{
+	struct sctp_stream_out_ext *soute, *tmp;
+
+	list_for_each_entry_safe(soute, tmp, &stream->fc_list, fc_list)
+		list_del_init(&soute->fc_list);
+}
+
+static struct sctp_sched_ops sctp_sched_fc = {
+	.set = sctp_sched_fc_set,
+	.get = sctp_sched_fc_get,
+	.init = sctp_sched_fc_init,
+	.init_sid = sctp_sched_fc_init_sid,
+	.free_sid = sctp_sched_fc_free_sid,
+	.enqueue = sctp_sched_fc_enqueue,
+	.dequeue = sctp_sched_fc_dequeue,
+	.dequeue_done = sctp_sched_fc_dequeue_done,
+	.sched_all = sctp_sched_fc_sched_all,
+	.unsched_all = sctp_sched_fc_unsched_all,
+};
+
+void sctp_sched_ops_fc_init(void)
+{
+	sctp_sched_ops_register(SCTP_SS_FC, &sctp_sched_fc);
+}
-- 
cgit v1.2.3-70-g09d2


From 42d452e7709fdb4d42376d2a97369e22cc80a5d2 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Tue, 7 Mar 2023 16:23:27 -0500
Subject: sctp: add weighted fair queueing stream scheduler

As it says in rfc8260#section-3.6 about the weighted fair queueing
scheduler:

   A Weighted Fair Queueing scheduler between the streams is used.  The
   weight is configurable per outgoing SCTP stream.  This scheduler
   considers the lengths of the messages of each stream and schedules
   them in a specific way to use the capacity according to the given
   weights.  If the weight of stream S1 is n times the weight of stream
   S2, the scheduler should assign to stream S1 n times the capacity it
   assigns to stream S2.  The details are implementation dependent.
   Interleaving user messages allows for a better realization of the
   capacity usage according to the given weights.

This patch adds Weighted Fair Queueing Scheduler actually based on
the code of Fair Capacity Scheduler by adding fc_weight into struct
sctp_stream_out_ext and taking it into account when sorting stream->
fc_list in sctp_sched_fc_sched() and sctp_sched_fc_dequeue_done().

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/sctp/stream_sched.h |  1 +
 include/net/sctp/structs.h      |  1 +
 include/uapi/linux/sctp.h       |  3 ++-
 net/sctp/stream_sched.c         |  1 +
 net/sctp/stream_sched_fc.c      | 50 +++++++++++++++++++++++++++++++++++++----
 5 files changed, 51 insertions(+), 5 deletions(-)

diff --git a/include/net/sctp/stream_sched.h b/include/net/sctp/stream_sched.h
index 913170710adb..572d73fdcd5e 100644
--- a/include/net/sctp/stream_sched.h
+++ b/include/net/sctp/stream_sched.h
@@ -59,5 +59,6 @@ void sctp_sched_ops_register(enum sctp_sched_type sched,
 void sctp_sched_ops_prio_init(void);
 void sctp_sched_ops_rr_init(void);
 void sctp_sched_ops_fc_init(void);
+void sctp_sched_ops_wfq_init(void);
 
 #endif /* __sctp_stream_sched_h__ */
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 2f1c9f50b352..a0933efd93c3 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1432,6 +1432,7 @@ struct sctp_stream_out_ext {
 		struct {
 			struct list_head fc_list;
 			__u32 fc_length;
+			__u16 fc_weight;
 		};
 	};
 };
diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h
index 6814c5a1c4bc..b7d91d4cf0db 100644
--- a/include/uapi/linux/sctp.h
+++ b/include/uapi/linux/sctp.h
@@ -1212,7 +1212,8 @@ enum sctp_sched_type {
 	SCTP_SS_PRIO,
 	SCTP_SS_RR,
 	SCTP_SS_FC,
-	SCTP_SS_MAX = SCTP_SS_FC
+	SCTP_SS_WFQ,
+	SCTP_SS_MAX = SCTP_SS_WFQ
 };
 
 /* Probe Interval socket option */
diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c
index 1ebd14ef8daa..e843760e9aaa 100644
--- a/net/sctp/stream_sched.c
+++ b/net/sctp/stream_sched.c
@@ -125,6 +125,7 @@ void sctp_sched_ops_init(void)
 	sctp_sched_ops_prio_init();
 	sctp_sched_ops_rr_init();
 	sctp_sched_ops_fc_init();
+	sctp_sched_ops_wfq_init();
 }
 
 static void sctp_sched_free_sched(struct sctp_stream *stream)
diff --git a/net/sctp/stream_sched_fc.c b/net/sctp/stream_sched_fc.c
index b336c2f5486b..4bd18a497a6d 100644
--- a/net/sctp/stream_sched_fc.c
+++ b/net/sctp/stream_sched_fc.c
@@ -19,11 +19,32 @@
 #include <net/sctp/sm.h>
 #include <net/sctp/stream_sched.h>
 
-/* Fair Capacity handling
- * RFC 8260 section 3.5
+/* Fair Capacity and Weighted Fair Queueing handling
+ * RFC 8260 section 3.5 and 3.6
  */
 static void sctp_sched_fc_unsched_all(struct sctp_stream *stream);
 
+static int sctp_sched_wfq_set(struct sctp_stream *stream, __u16 sid,
+			      __u16 weight, gfp_t gfp)
+{
+	struct sctp_stream_out_ext *soute = SCTP_SO(stream, sid)->ext;
+
+	if (!weight)
+		return -EINVAL;
+
+	soute->fc_weight = weight;
+	return 0;
+}
+
+static int sctp_sched_wfq_get(struct sctp_stream *stream, __u16 sid,
+			      __u16 *value)
+{
+	struct sctp_stream_out_ext *soute = SCTP_SO(stream, sid)->ext;
+
+	*value = soute->fc_weight;
+	return 0;
+}
+
 static int sctp_sched_fc_set(struct sctp_stream *stream, __u16 sid,
 			     __u16 weight, gfp_t gfp)
 {
@@ -50,6 +71,7 @@ static int sctp_sched_fc_init_sid(struct sctp_stream *stream, __u16 sid,
 
 	INIT_LIST_HEAD(&soute->fc_list);
 	soute->fc_length = 0;
+	soute->fc_weight = 1;
 
 	return 0;
 }
@@ -67,7 +89,8 @@ static void sctp_sched_fc_sched(struct sctp_stream *stream,
 		return;
 
 	list_for_each_entry(pos, &stream->fc_list, fc_list)
-		if (pos->fc_length >= soute->fc_length)
+		if ((__u64)pos->fc_length * soute->fc_weight >=
+		    (__u64)soute->fc_length * pos->fc_weight)
 			break;
 	list_add_tail(&soute->fc_list, &pos->fc_list);
 }
@@ -137,7 +160,8 @@ static void sctp_sched_fc_dequeue_done(struct sctp_outq *q,
 
 	pos = soute;
 	list_for_each_entry_continue(pos, &stream->fc_list, fc_list)
-		if (pos->fc_length >= soute->fc_length)
+		if ((__u64)pos->fc_length * soute->fc_weight >=
+		    (__u64)soute->fc_length * pos->fc_weight)
 			break;
 	list_move_tail(&soute->fc_list, &pos->fc_list);
 }
@@ -181,3 +205,21 @@ void sctp_sched_ops_fc_init(void)
 {
 	sctp_sched_ops_register(SCTP_SS_FC, &sctp_sched_fc);
 }
+
+static struct sctp_sched_ops sctp_sched_wfq = {
+	.set = sctp_sched_wfq_set,
+	.get = sctp_sched_wfq_get,
+	.init = sctp_sched_fc_init,
+	.init_sid = sctp_sched_fc_init_sid,
+	.free_sid = sctp_sched_fc_free_sid,
+	.enqueue = sctp_sched_fc_enqueue,
+	.dequeue = sctp_sched_fc_dequeue,
+	.dequeue_done = sctp_sched_fc_dequeue_done,
+	.sched_all = sctp_sched_fc_sched_all,
+	.unsched_all = sctp_sched_fc_unsched_all,
+};
+
+void sctp_sched_ops_wfq_init(void)
+{
+	sctp_sched_ops_register(SCTP_SS_WFQ, &sctp_sched_wfq);
+}
-- 
cgit v1.2.3-70-g09d2


From a6865576317f6249f3f83cf4c10ab56e627ee153 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 8 Mar 2023 22:02:44 -0800
Subject: selftests/bpf: Fix flaky fib_lookup test

There is a report that fib_lookup test is flaky when running in parallel.
A symptom of slowness or delay. An example:

Testing IPv6 stale neigh
set_lookup_params:PASS:inet_pton(IPV6_IFACE_ADDR) 0 nsec
test_fib_lookup:PASS:bpf_prog_test_run_opts 0 nsec
test_fib_lookup:FAIL:fib_lookup_ret unexpected fib_lookup_ret: actual 0 != expected 7
test_fib_lookup:FAIL:dmac not match unexpected dmac not match: actual 1 != expected 0
dmac expected 11:11:11:11:11:11 actual 00:00:00:00:00:00

[ Note that the "fib_lookup_ret unexpected fib_lookup_ret actual 0 ..."
  is reversed in terms of expected and actual value. Fixing in this
  patch also. ]

One possibility is the testing stale neigh entry was marked dead by the
gc (in neigh_periodic_work). The default gc_stale_time sysctl is 60s.
This patch increases it to 15 mins.

It also:

- fixes the reversed arg (actual vs expected) in one of the
  ASSERT_EQ test
- removes the nodad command arg when adding v4 neigh entry which
  currently has a warning.

Fixes: 168de0233586 ("selftests/bpf: Add bpf_fib_lookup test")
Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230309060244.3242491-1-martin.lau@linux.dev
---
 tools/testing/selftests/bpf/prog_tests/fib_lookup.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/fib_lookup.c b/tools/testing/selftests/bpf/prog_tests/fib_lookup.c
index 429393caf612..a1e712105811 100644
--- a/tools/testing/selftests/bpf/prog_tests/fib_lookup.c
+++ b/tools/testing/selftests/bpf/prog_tests/fib_lookup.c
@@ -54,11 +54,19 @@ static int setup_netns(void)
 	SYS(fail, "ip link add veth1 type veth peer name veth2");
 	SYS(fail, "ip link set dev veth1 up");
 
+	err = write_sysctl("/proc/sys/net/ipv4/neigh/veth1/gc_stale_time", "900");
+	if (!ASSERT_OK(err, "write_sysctl(net.ipv4.neigh.veth1.gc_stale_time)"))
+		goto fail;
+
+	err = write_sysctl("/proc/sys/net/ipv6/neigh/veth1/gc_stale_time", "900");
+	if (!ASSERT_OK(err, "write_sysctl(net.ipv6.neigh.veth1.gc_stale_time)"))
+		goto fail;
+
 	SYS(fail, "ip addr add %s/64 dev veth1 nodad", IPV6_IFACE_ADDR);
 	SYS(fail, "ip neigh add %s dev veth1 nud failed", IPV6_NUD_FAILED_ADDR);
 	SYS(fail, "ip neigh add %s dev veth1 lladdr %s nud stale", IPV6_NUD_STALE_ADDR, DMAC);
 
-	SYS(fail, "ip addr add %s/24 dev veth1 nodad", IPV4_IFACE_ADDR);
+	SYS(fail, "ip addr add %s/24 dev veth1", IPV4_IFACE_ADDR);
 	SYS(fail, "ip neigh add %s dev veth1 nud failed", IPV4_NUD_FAILED_ADDR);
 	SYS(fail, "ip neigh add %s dev veth1 lladdr %s nud stale", IPV4_NUD_STALE_ADDR, DMAC);
 
@@ -158,7 +166,7 @@ void test_fib_lookup(void)
 		if (!ASSERT_OK(err, "bpf_prog_test_run_opts"))
 			continue;
 
-		ASSERT_EQ(tests[i].expected_ret, skel->bss->fib_lookup_ret,
+		ASSERT_EQ(skel->bss->fib_lookup_ret, tests[i].expected_ret,
 			  "fib_lookup_ret");
 
 		ret = memcmp(tests[i].dmac, fib_params->dmac, sizeof(tests[i].dmac));
-- 
cgit v1.2.3-70-g09d2


From 5a70f4a63000ba68004fb3c1aaf2f90303dd228f Mon Sep 17 00:00:00 2001
From: Michael Weiß <michael.weiss@aisec.fraunhofer.de>
Date: Thu, 9 Mar 2023 14:38:23 +0100
Subject: bpf: Fix a typo for BPF_F_ANY_ALIGNMENT in bpf.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix s/BPF_PROF_LOAD/BPF_PROG_LOAD/ typo in the documentation comment
for BPF_F_ANY_ALIGNMENT in bpf.h.

Signed-off-by: Michael Weiß <michael.weiss@aisec.fraunhofer.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230309133823.944097-1-michael.weiss@aisec.fraunhofer.de
---
 include/uapi/linux/bpf.h       | 2 +-
 tools/include/uapi/linux/bpf.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 4abddb668a10..d8c534e05b0a 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1108,7 +1108,7 @@ enum bpf_link_type {
  */
 #define BPF_F_STRICT_ALIGNMENT	(1U << 0)
 
-/* If BPF_F_ANY_ALIGNMENT is used in BPF_PROF_LOAD command, the
+/* If BPF_F_ANY_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will allow any alignment whatsoever.  On platforms
  * with strict alignment requirements for loads ands stores (such
  * as sparc and mips) the verifier validates that all loads and
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 4abddb668a10..d8c534e05b0a 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1108,7 +1108,7 @@ enum bpf_link_type {
  */
 #define BPF_F_STRICT_ALIGNMENT	(1U << 0)
 
-/* If BPF_F_ANY_ALIGNMENT is used in BPF_PROF_LOAD command, the
+/* If BPF_F_ANY_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will allow any alignment whatsoever.  On platforms
  * with strict alignment requirements for loads ands stores (such
  * as sparc and mips) the verifier validates that all loads and
-- 
cgit v1.2.3-70-g09d2


From 27a36bc3cdd5e0420eea90762d69bea34daf97e1 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Thu, 9 Mar 2023 18:32:40 +0100
Subject: selftests/bpf: Use ifname instead of ifindex in XDP compliance test
 tool

Rely on interface name instead of interface index in error messages or
logs from XDP compliance test tool.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/7dc5a8ff56c252b1a7ae29b059d0b2b1543c8b5d.1678382940.git.lorenzo@kernel.org
---
 tools/testing/selftests/bpf/xdp_features.c | 44 ++++++++++++++++++------------
 1 file changed, 27 insertions(+), 17 deletions(-)

diff --git a/tools/testing/selftests/bpf/xdp_features.c b/tools/testing/selftests/bpf/xdp_features.c
index fce12165213b..b060a0d24e44 100644
--- a/tools/testing/selftests/bpf/xdp_features.c
+++ b/tools/testing/selftests/bpf/xdp_features.c
@@ -25,6 +25,7 @@
 
 static struct env {
 	bool verbosity;
+	char ifname[IF_NAMESIZE];
 	int ifindex;
 	bool is_tester;
 	struct {
@@ -179,7 +180,7 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 		env.ifindex = if_nametoindex(arg);
 		if (!env.ifindex)
 			env.ifindex = strtoul(arg, NULL, 0);
-		if (!env.ifindex) {
+		if (!env.ifindex || !if_indextoname(env.ifindex, env.ifname)) {
 			fprintf(stderr,
 				"Bad interface index or name (%d): %s\n",
 				errno, strerror(errno));
@@ -205,6 +206,7 @@ static void set_env_default(void)
 	env.feature.drv_feature = NETDEV_XDP_ACT_NDO_XMIT;
 	env.feature.action = -EINVAL;
 	env.ifindex = -ENODEV;
+	strcpy(env.ifname, "unknown");
 	make_sockaddr(AF_INET6, "::ffff:127.0.0.1", DUT_CTRL_PORT,
 		      &env.dut_ctrl_addr, NULL);
 	make_sockaddr(AF_INET6, "::ffff:127.0.0.1", DUT_ECHO_PORT,
@@ -248,15 +250,18 @@ static int dut_run_echo_thread(pthread_t *t, int *sockfd)
 	sockfd = start_reuseport_server(AF_INET6, SOCK_DGRAM, NULL,
 					DUT_ECHO_PORT, 0, 1);
 	if (!sockfd) {
-		fprintf(stderr, "Failed to create echo socket\n");
+		fprintf(stderr,
+			"Failed creating data UDP socket on device %s\n",
+			env.ifname);
 		return -errno;
 	}
 
 	/* start echo channel */
 	err = pthread_create(t, NULL, dut_echo_thread, sockfd);
 	if (err) {
-		fprintf(stderr, "Failed creating dut_echo thread: %s\n",
-			strerror(-err));
+		fprintf(stderr,
+			"Failed creating data UDP thread on device %s: %s\n",
+			env.ifname, strerror(-err));
 		free_fds(sockfd, 1);
 		return -EINVAL;
 	}
@@ -320,9 +325,8 @@ static int dut_attach_xdp_prog(struct xdp_features *skel, int flags)
 
 	err = bpf_xdp_attach(env.ifindex, bpf_program__fd(prog), flags, NULL);
 	if (err)
-		fprintf(stderr,
-			"Failed to attach XDP program to ifindex %d\n",
-			env.ifindex);
+		fprintf(stderr, "Failed attaching XDP program to device %s\n",
+			env.ifname);
 	return err;
 }
 
@@ -358,13 +362,16 @@ static int dut_run(struct xdp_features *skel)
 	sockfd = start_reuseport_server(AF_INET6, SOCK_STREAM, NULL,
 					DUT_CTRL_PORT, 0, 1);
 	if (!sockfd) {
-		fprintf(stderr, "Failed to create DUT socket\n");
+		fprintf(stderr,
+			"Failed creating control socket on device %s\n", env.ifname);
 		return -errno;
 	}
 
 	ctrl_sockfd = accept(*sockfd, (struct sockaddr *)&ctrl_addr, &addrlen);
 	if (ctrl_sockfd < 0) {
-		fprintf(stderr, "Failed to accept connection on DUT socket\n");
+		fprintf(stderr,
+			"Failed accepting connections on device %s control socket\n",
+			env.ifname);
 		free_fds(sockfd, 1);
 		return -errno;
 	}
@@ -422,8 +429,8 @@ static int dut_run(struct xdp_features *skel)
 					    &opts);
 			if (err) {
 				fprintf(stderr,
-					"Failed to query XDP cap for ifindex %d\n",
-					env.ifindex);
+					"Failed querying XDP cap for device %s\n",
+					env.ifname);
 				goto end_thread;
 			}
 
@@ -540,7 +547,9 @@ static int send_echo_msg(void)
 
 	sockfd = socket(AF_INET6, SOCK_DGRAM, 0);
 	if (sockfd < 0) {
-		fprintf(stderr, "Failed to create echo socket\n");
+		fprintf(stderr,
+			"Failed creating data UDP socket on device %s\n",
+			env.ifname);
 		return -errno;
 	}
 
@@ -596,8 +605,8 @@ static int tester_run(struct xdp_features *skel)
 
 	err = bpf_xdp_attach(env.ifindex, bpf_program__fd(prog), flags, NULL);
 	if (err) {
-		fprintf(stderr, "Failed to attach XDP program to ifindex %d\n",
-			env.ifindex);
+		fprintf(stderr, "Failed attaching XDP program to device %s\n",
+			env.ifname);
 		goto out;
 	}
 
@@ -653,7 +662,7 @@ int main(int argc, char **argv)
 		return err;
 
 	if (env.ifindex < 0) {
-		fprintf(stderr, "Invalid ifindex\n");
+		fprintf(stderr, "Invalid device name %s\n", env.ifname);
 		return -ENODEV;
 	}
 
@@ -684,11 +693,12 @@ int main(int argc, char **argv)
 
 	if (env.is_tester) {
 		/* Tester */
-		fprintf(stdout, "Starting tester on device %d\n", env.ifindex);
+		fprintf(stdout, "Starting tester service on device %s\n",
+			env.ifname);
 		err = tester_run(skel);
 	} else {
 		/* DUT */
-		fprintf(stdout, "Starting DUT on device %d\n", env.ifindex);
+		fprintf(stdout, "Starting test on device %s\n", env.ifname);
 		err = dut_run(skel);
 	}
 
-- 
cgit v1.2.3-70-g09d2


From c1cd734c1bb3f4d9db75c51c23306e29d8749783 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Thu, 9 Mar 2023 18:32:41 +0100
Subject: selftests/bpf: Improve error logs in XDP compliance test tool

Improve some error logs reported in the XDP compliance test tool.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/212fc5bd214ff706f6ef1acbe7272cf4d803ca9c.1678382940.git.lorenzo@kernel.org
---
 tools/testing/selftests/bpf/xdp_features.c | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/bpf/xdp_features.c b/tools/testing/selftests/bpf/xdp_features.c
index b060a0d24e44..b449788fbd39 100644
--- a/tools/testing/selftests/bpf/xdp_features.c
+++ b/tools/testing/selftests/bpf/xdp_features.c
@@ -152,20 +152,26 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 	case 'D':
 		if (make_sockaddr(AF_INET6, arg, DUT_ECHO_PORT,
 				  &env.dut_addr, NULL)) {
-			fprintf(stderr, "Invalid DUT address: %s\n", arg);
+			fprintf(stderr,
+				"Invalid address assigned to the Device Under Test: %s\n",
+				arg);
 			return ARGP_ERR_UNKNOWN;
 		}
 		break;
 	case 'C':
 		if (make_sockaddr(AF_INET6, arg, DUT_CTRL_PORT,
 				  &env.dut_ctrl_addr, NULL)) {
-			fprintf(stderr, "Invalid DUT CTRL address: %s\n", arg);
+			fprintf(stderr,
+				"Invalid address assigned to the Device Under Test: %s\n",
+				arg);
 			return ARGP_ERR_UNKNOWN;
 		}
 		break;
 	case 'T':
 		if (make_sockaddr(AF_INET6, arg, 0, &env.tester_addr, NULL)) {
-			fprintf(stderr, "Invalid Tester address: %s\n", arg);
+			fprintf(stderr,
+				"Invalid address assigned to the Tester device: %s\n",
+				arg);
 			return ARGP_ERR_UNKNOWN;
 		}
 		break;
@@ -454,7 +460,8 @@ static int dut_run(struct xdp_features *skel)
 						   &key, sizeof(key),
 						   &val, sizeof(val), 0);
 			if (err) {
-				fprintf(stderr, "bpf_map_lookup_elem failed\n");
+				fprintf(stderr,
+					"bpf_map_lookup_elem failed (%d)\n", err);
 				goto end_thread;
 			}
 
@@ -496,7 +503,7 @@ static bool tester_collect_detected_cap(struct xdp_features *skel,
 	err = bpf_map__lookup_elem(skel->maps.stats, &key, sizeof(key),
 				   &val, sizeof(val), 0);
 	if (err) {
-		fprintf(stderr, "bpf_map_lookup_elem failed\n");
+		fprintf(stderr, "bpf_map_lookup_elem failed (%d)\n", err);
 		return false;
 	}
 
@@ -574,7 +581,8 @@ static int tester_run(struct xdp_features *skel)
 
 	sockfd = socket(AF_INET6, SOCK_STREAM, 0);
 	if (sockfd < 0) {
-		fprintf(stderr, "Failed to create tester socket\n");
+		fprintf(stderr,
+			"Failed creating tester service control socket\n");
 		return -errno;
 	}
 
@@ -584,7 +592,8 @@ static int tester_run(struct xdp_features *skel)
 	err = connect(sockfd, (struct sockaddr *)&env.dut_ctrl_addr,
 		      sizeof(env.dut_ctrl_addr));
 	if (err) {
-		fprintf(stderr, "Failed to connect to the DUT\n");
+		fprintf(stderr,
+			"Failed connecting to the Device Under Test control socket\n");
 		return -errno;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From a4ba62906db80246b24408e5c8a9e9d78b647b00 Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:08 +0530
Subject: i40e: consolidate maximum frame size calculation for vsi

Introduce new helper function to calculate max frame size for validating
and setting of vsi frame size. This is used while configuring vsi,
changing the MTU and attaching an XDP program to the vsi.

This is in preparation of the legacy rx and multi-buffer changes to be
introduced in later patches.

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 71 +++++++++++++++--------------
 1 file changed, 38 insertions(+), 33 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 467001db5070..d375d7940308 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2896,15 +2896,35 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf)
 }
 
 /**
- * i40e_max_xdp_frame_size - returns the maximum allowed frame size for XDP
+ * i40e_calculate_vsi_rx_buf_len - Calculates buffer length
+ *
+ * @vsi: VSI to calculate rx_buf_len from
+ */
+static u16 i40e_calculate_vsi_rx_buf_len(struct i40e_vsi *vsi)
+{
+	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
+		return I40E_RXBUFFER_2048;
+
+	return PAGE_SIZE < 8192 ? I40E_RXBUFFER_3072 : I40E_RXBUFFER_2048;
+}
+
+/**
+ * i40e_max_vsi_frame_size - returns the maximum allowed frame size for VSI
  * @vsi: the vsi
+ * @xdp_prog: XDP program
  **/
-static int i40e_max_xdp_frame_size(struct i40e_vsi *vsi)
+static int i40e_max_vsi_frame_size(struct i40e_vsi *vsi,
+				   struct bpf_prog *xdp_prog)
 {
-	if (PAGE_SIZE >= 8192 || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
-		return I40E_RXBUFFER_2048;
+	u16 rx_buf_len = i40e_calculate_vsi_rx_buf_len(vsi);
+	u16 chain_len;
+
+	if (xdp_prog)
+		chain_len = 1;
 	else
-		return I40E_RXBUFFER_3072;
+		chain_len = I40E_MAX_CHAINED_RX_BUFFERS;
+
+	return min_t(u16, rx_buf_len * chain_len, I40E_MAX_RXBUFFER);
 }
 
 /**
@@ -2919,12 +2939,13 @@ static int i40e_change_mtu(struct net_device *netdev, int new_mtu)
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
+	int frame_size;
 
-	if (i40e_enabled_xdp_vsi(vsi)) {
-		int frame_size = new_mtu + I40E_PACKET_HDR_PAD;
-
-		if (frame_size > i40e_max_xdp_frame_size(vsi))
-			return -EINVAL;
+	frame_size = i40e_max_vsi_frame_size(vsi, vsi->xdp_prog);
+	if (new_mtu > frame_size - I40E_PACKET_HDR_PAD) {
+		netdev_err(netdev, "Error changing mtu to %d, Max is %d\n",
+			   new_mtu, frame_size - I40E_PACKET_HDR_PAD);
+		return -EINVAL;
 	}
 
 	netdev_dbg(netdev, "changing MTU from %d to %d\n",
@@ -3693,24 +3714,6 @@ static int i40e_vsi_configure_tx(struct i40e_vsi *vsi)
 	return err;
 }
 
-/**
- * i40e_calculate_vsi_rx_buf_len - Calculates buffer length
- *
- * @vsi: VSI to calculate rx_buf_len from
- */
-static u16 i40e_calculate_vsi_rx_buf_len(struct i40e_vsi *vsi)
-{
-	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
-		return I40E_RXBUFFER_2048;
-
-#if (PAGE_SIZE < 8192)
-	if (!I40E_2K_TOO_SMALL_WITH_PADDING && vsi->netdev->mtu <= ETH_DATA_LEN)
-		return I40E_RXBUFFER_1536 - NET_IP_ALIGN;
-#endif
-
-	return PAGE_SIZE < 8192 ? I40E_RXBUFFER_3072 : I40E_RXBUFFER_2048;
-}
-
 /**
  * i40e_vsi_configure_rx - Configure the VSI for Rx
  * @vsi: the VSI being configured
@@ -3722,13 +3725,15 @@ static int i40e_vsi_configure_rx(struct i40e_vsi *vsi)
 	int err = 0;
 	u16 i;
 
-	vsi->max_frame = I40E_MAX_RXBUFFER;
+	vsi->max_frame = i40e_max_vsi_frame_size(vsi, vsi->xdp_prog);
 	vsi->rx_buf_len = i40e_calculate_vsi_rx_buf_len(vsi);
 
 #if (PAGE_SIZE < 8192)
 	if (vsi->netdev && !I40E_2K_TOO_SMALL_WITH_PADDING &&
-	    vsi->netdev->mtu <= ETH_DATA_LEN)
-		vsi->max_frame = I40E_RXBUFFER_1536 - NET_IP_ALIGN;
+	    vsi->netdev->mtu <= ETH_DATA_LEN) {
+		vsi->rx_buf_len = I40E_RXBUFFER_1536 - NET_IP_ALIGN;
+		vsi->max_frame = vsi->rx_buf_len;
+	}
 #endif
 
 	/* set up individual rings */
@@ -13316,14 +13321,14 @@ out_err:
 static int i40e_xdp_setup(struct i40e_vsi *vsi, struct bpf_prog *prog,
 			  struct netlink_ext_ack *extack)
 {
-	int frame_size = vsi->netdev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
+	int frame_size = i40e_max_vsi_frame_size(vsi, prog);
 	struct i40e_pf *pf = vsi->back;
 	struct bpf_prog *old_prog;
 	bool need_reset;
 	int i;
 
 	/* Don't allow frames that span over multiple buffers */
-	if (frame_size > i40e_calculate_vsi_rx_buf_len(vsi)) {
+	if (vsi->netdev->mtu > frame_size - I40E_PACKET_HDR_PAD) {
 		NL_SET_ERR_MSG_MOD(extack, "MTU too large to enable XDP");
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-70-g09d2


From f7f732a7196d2e876639b1af453b30a6a5fe4a90 Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:09 +0530
Subject: i40e: change Rx buffer size for legacy-rx to support XDP multi-buffer

Adding support for XDP multi-buffer entails adding information of all
the fragments of the packet in the xdp_buff. This approach implies that
underlying buffer has to provide tailroom for skb_shared_info.

In the legacy-rx mode, driver can only configure up to 2k sized Rx buffers
and with the current configuration of 2k sized Rx buffers there is no way
to do tailroom reservation for skb_shared_info. Hence size of Rx buffers
is now lowered to 2048 - sizeof(skb_shared_info). Also, driver can only
chain up to 5 Rx buffers and this means max MTU supported for legacy-rx
is now 8614 (5 * rx_buffer_len  - ETH header with VLAN).

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c |  7 +++++++
 drivers/net/ethernet/intel/i40e/i40e_main.c    | 12 +++++++++---
 2 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 4934ff58332c..afc4fa8c66af 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -5402,6 +5402,13 @@ flags_complete:
 		return -EOPNOTSUPP;
 	}
 
+	if ((changed_flags & I40E_FLAG_LEGACY_RX) &&
+	    I40E_2K_TOO_SMALL_WITH_PADDING) {
+		dev_warn(&pf->pdev->dev,
+			 "2k Rx buffer is too small to fit standard MTU and skb_shared_info\n");
+		return -EOPNOTSUPP;
+	}
+
 	if ((changed_flags & new_flags &
 	     I40E_FLAG_LINK_DOWN_ON_CLOSE_ENABLED) &&
 	    (new_flags & I40E_FLAG_MFP_ENABLED))
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index d375d7940308..e8cf5644bf10 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2903,7 +2903,7 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf)
 static u16 i40e_calculate_vsi_rx_buf_len(struct i40e_vsi *vsi)
 {
 	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
-		return I40E_RXBUFFER_2048;
+		return SKB_WITH_OVERHEAD(I40E_RXBUFFER_2048);
 
 	return PAGE_SIZE < 8192 ? I40E_RXBUFFER_3072 : I40E_RXBUFFER_2048;
 }
@@ -3661,10 +3661,16 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring)
 	}
 
 	/* configure Rx buffer alignment */
-	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
+	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX)) {
+		if (I40E_2K_TOO_SMALL_WITH_PADDING) {
+			dev_info(&vsi->back->pdev->dev,
+				 "2k Rx buffer is too small to fit standard MTU and skb_shared_info\n");
+			return -EOPNOTSUPP;
+		}
 		clear_ring_build_skb_enabled(ring);
-	else
+	} else {
 		set_ring_build_skb_enabled(ring);
+	}
 
 	ring->rx_offset = i40e_rx_offset(ring);
 
-- 
cgit v1.2.3-70-g09d2


From e2843f037127c3a9db26718aaa293d2a8e5881e4 Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:10 +0530
Subject: i40e: add pre-xdp page_count in rx_buffer

Page count of rx_buffer needs to be stored prior to XDP call to prevent
page recycling in case that buffer would be freed within xdp redirect
path. Instead of storing it on the stack, now it is stored in the
rx_buffer struct. This will help in processing multi-buffers as the page
counts of all rx_buffers (of the same packet) don't need to be stored on
stack.

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c | 23 ++++++++---------------
 drivers/net/ethernet/intel/i40e/i40e_txrx.h |  1 +
 2 files changed, 9 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 924f972b91fa..a7fba294a8f4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1970,7 +1970,6 @@ static bool i40e_cleanup_headers(struct i40e_ring *rx_ring, struct sk_buff *skb,
  * i40e_can_reuse_rx_page - Determine if page can be reused for another Rx
  * @rx_buffer: buffer containing the page
  * @rx_stats: rx stats structure for the rx ring
- * @rx_buffer_pgcnt: buffer page refcount pre xdp_do_redirect() call
  *
  * If page is reusable, we have a green light for calling i40e_reuse_rx_page,
  * which will assign the current buffer to the buffer that next_to_alloc is
@@ -1981,8 +1980,7 @@ static bool i40e_cleanup_headers(struct i40e_ring *rx_ring, struct sk_buff *skb,
  * or busy if it could not be reused.
  */
 static bool i40e_can_reuse_rx_page(struct i40e_rx_buffer *rx_buffer,
-				   struct i40e_rx_queue_stats *rx_stats,
-				   int rx_buffer_pgcnt)
+				   struct i40e_rx_queue_stats *rx_stats)
 {
 	unsigned int pagecnt_bias = rx_buffer->pagecnt_bias;
 	struct page *page = rx_buffer->page;
@@ -1995,7 +1993,7 @@ static bool i40e_can_reuse_rx_page(struct i40e_rx_buffer *rx_buffer,
 
 #if (PAGE_SIZE < 8192)
 	/* if we are only owner of page we can reuse it */
-	if (unlikely((rx_buffer_pgcnt - pagecnt_bias) > 1)) {
+	if (unlikely((rx_buffer->page_count - pagecnt_bias) > 1)) {
 		rx_stats->page_busy_count++;
 		return false;
 	}
@@ -2058,19 +2056,17 @@ static void i40e_add_rx_frag(struct i40e_ring *rx_ring,
  * i40e_get_rx_buffer - Fetch Rx buffer and synchronize data for use
  * @rx_ring: rx descriptor ring to transact packets on
  * @size: size of buffer to add to skb
- * @rx_buffer_pgcnt: buffer page refcount
  *
  * This function will pull an Rx buffer from the ring and synchronize it
  * for use by the CPU.
  */
 static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring *rx_ring,
-						 const unsigned int size,
-						 int *rx_buffer_pgcnt)
+						 const unsigned int size)
 {
 	struct i40e_rx_buffer *rx_buffer;
 
 	rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean);
-	*rx_buffer_pgcnt =
+	rx_buffer->page_count =
 #if (PAGE_SIZE < 8192)
 		page_count(rx_buffer->page);
 #else
@@ -2226,16 +2222,14 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
  * i40e_put_rx_buffer - Clean up used buffer and either recycle or free
  * @rx_ring: rx descriptor ring to transact packets on
  * @rx_buffer: rx buffer to pull data from
- * @rx_buffer_pgcnt: rx buffer page refcount pre xdp_do_redirect() call
  *
  * This function will clean up the contents of the rx_buffer.  It will
  * either recycle the buffer or unmap it and free the associated resources.
  */
 static void i40e_put_rx_buffer(struct i40e_ring *rx_ring,
-			       struct i40e_rx_buffer *rx_buffer,
-			       int rx_buffer_pgcnt)
+			       struct i40e_rx_buffer *rx_buffer)
 {
-	if (i40e_can_reuse_rx_page(rx_buffer, &rx_ring->rx_stats, rx_buffer_pgcnt)) {
+	if (i40e_can_reuse_rx_page(rx_buffer, &rx_ring->rx_stats)) {
 		/* hand second half of page back to the ring */
 		i40e_reuse_rx_page(rx_ring, rx_buffer);
 	} else {
@@ -2457,7 +2451,6 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 	while (likely(total_rx_packets < (unsigned int)budget)) {
 		struct i40e_rx_buffer *rx_buffer;
 		union i40e_rx_desc *rx_desc;
-		int rx_buffer_pgcnt;
 		unsigned int size;
 		u64 qword;
 
@@ -2500,7 +2493,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 			break;
 
 		i40e_trace(clean_rx_irq, rx_ring, rx_desc, skb);
-		rx_buffer = i40e_get_rx_buffer(rx_ring, size, &rx_buffer_pgcnt);
+		rx_buffer = i40e_get_rx_buffer(rx_ring, size);
 
 		/* retrieve a buffer from the ring */
 		if (!skb) {
@@ -2541,7 +2534,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 			break;
 		}
 
-		i40e_put_rx_buffer(rx_ring, rx_buffer, rx_buffer_pgcnt);
+		i40e_put_rx_buffer(rx_ring, rx_buffer);
 		cleaned_count++;
 
 		i40e_inc_ntc(rx_ring);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 768290dc6f48..eec4a4a99b9c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -277,6 +277,7 @@ struct i40e_rx_buffer {
 	struct page *page;
 	__u32 page_offset;
 	__u16 pagecnt_bias;
+	__u32 page_count;
 };
 
 struct i40e_queue_stats {
-- 
cgit v1.2.3-70-g09d2


From 03e88c8a791c0655ddea07fd4c7cd6cb16087c8f Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:11 +0530
Subject: i40e: Change size to truesize when using i40e_rx_buffer_flip()

Truesize is now passed directly to i40e_rx_buffer_flip() instead of size
so that it does not need to recalculate truesize from size using
i40e_rx_frame_truesize() before adjusting page offset.

With these change the function can now be used during skb building and
adding frags. In later patches it will also be easier for adjusting
page offsets for multi-buffers.

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c | 54 ++++++++++-------------------
 1 file changed, 19 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index a7fba294a8f4..fa44a2c353b2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2018,6 +2018,21 @@ static bool i40e_can_reuse_rx_page(struct i40e_rx_buffer *rx_buffer,
 	return true;
 }
 
+/**
+ * i40e_rx_buffer_flip - adjusted rx_buffer to point to an unused region
+ * @rx_buffer: Rx buffer to adjust
+ * @truesize: Size of adjustment
+ **/
+static void i40e_rx_buffer_flip(struct i40e_rx_buffer *rx_buffer,
+				unsigned int truesize)
+{
+#if (PAGE_SIZE < 8192)
+	rx_buffer->page_offset ^= truesize;
+#else
+	rx_buffer->page_offset += truesize;
+#endif
+}
+
 /**
  * i40e_add_rx_frag - Add contents of Rx buffer to sk_buff
  * @rx_ring: rx descriptor ring to transact packets on
@@ -2045,11 +2060,7 @@ static void i40e_add_rx_frag(struct i40e_ring *rx_ring,
 			rx_buffer->page_offset, size, truesize);
 
 	/* page is being used so we must update the page offset */
-#if (PAGE_SIZE < 8192)
-	rx_buffer->page_offset ^= truesize;
-#else
-	rx_buffer->page_offset += truesize;
-#endif
+	i40e_rx_buffer_flip(rx_buffer, truesize);
 }
 
 /**
@@ -2154,11 +2165,7 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring,
 				size, truesize);
 
 		/* buffer is used by skb, update page_offset */
-#if (PAGE_SIZE < 8192)
-		rx_buffer->page_offset ^= truesize;
-#else
-		rx_buffer->page_offset += truesize;
-#endif
+		i40e_rx_buffer_flip(rx_buffer, truesize);
 	} else {
 		/* buffer is unused, reset bias back to rx_buffer */
 		rx_buffer->pagecnt_bias++;
@@ -2209,11 +2216,7 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
 		skb_metadata_set(skb, metasize);
 
 	/* buffer is used by skb, update page_offset */
-#if (PAGE_SIZE < 8192)
-	rx_buffer->page_offset ^= truesize;
-#else
-	rx_buffer->page_offset += truesize;
-#endif
+	i40e_rx_buffer_flip(rx_buffer, truesize);
 
 	return skb;
 }
@@ -2326,25 +2329,6 @@ xdp_out:
 	return result;
 }
 
-/**
- * i40e_rx_buffer_flip - adjusted rx_buffer to point to an unused region
- * @rx_ring: Rx ring
- * @rx_buffer: Rx buffer to adjust
- * @size: Size of adjustment
- **/
-static void i40e_rx_buffer_flip(struct i40e_ring *rx_ring,
-				struct i40e_rx_buffer *rx_buffer,
-				unsigned int size)
-{
-	unsigned int truesize = i40e_rx_frame_truesize(rx_ring, size);
-
-#if (PAGE_SIZE < 8192)
-	rx_buffer->page_offset ^= truesize;
-#else
-	rx_buffer->page_offset += truesize;
-#endif
-}
-
 /**
  * i40e_xdp_ring_update_tail - Updates the XDP Tx ring tail register
  * @xdp_ring: XDP Tx ring
@@ -2513,7 +2497,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 		if (xdp_res) {
 			if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) {
 				xdp_xmit |= xdp_res;
-				i40e_rx_buffer_flip(rx_ring, rx_buffer, size);
+				i40e_rx_buffer_flip(rx_buffer, xdp.frame_sz);
 			} else {
 				rx_buffer->pagecnt_bias++;
 			}
-- 
cgit v1.2.3-70-g09d2


From 2bc0de9aca3ebdf24674f5a2a7890fde6304f5ca Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:12 +0530
Subject: i40e: use frame_sz instead of recalculating truesize for building skb

In skb path truesize is calculated while building skb. This is now
avoided and xdp->frame_is used instead for both i40e_build_skb() and
i40e_construct_skb().

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c | 20 ++++----------------
 1 file changed, 4 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index fa44a2c353b2..e34595ca4fbe 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2113,11 +2113,6 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring,
 					  struct xdp_buff *xdp)
 {
 	unsigned int size = xdp->data_end - xdp->data;
-#if (PAGE_SIZE < 8192)
-	unsigned int truesize = i40e_rx_pg_size(rx_ring) / 2;
-#else
-	unsigned int truesize = SKB_DATA_ALIGN(size);
-#endif
 	unsigned int headlen;
 	struct sk_buff *skb;
 
@@ -2162,10 +2157,10 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring,
 	if (size) {
 		skb_add_rx_frag(skb, 0, rx_buffer->page,
 				rx_buffer->page_offset + headlen,
-				size, truesize);
+				size, xdp->frame_sz);
 
 		/* buffer is used by skb, update page_offset */
-		i40e_rx_buffer_flip(rx_buffer, truesize);
+		i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
 	} else {
 		/* buffer is unused, reset bias back to rx_buffer */
 		rx_buffer->pagecnt_bias++;
@@ -2188,13 +2183,6 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
 				      struct xdp_buff *xdp)
 {
 	unsigned int metasize = xdp->data - xdp->data_meta;
-#if (PAGE_SIZE < 8192)
-	unsigned int truesize = i40e_rx_pg_size(rx_ring) / 2;
-#else
-	unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
-				SKB_DATA_ALIGN(xdp->data_end -
-					       xdp->data_hard_start);
-#endif
 	struct sk_buff *skb;
 
 	/* Prefetch first cache line of first page. If xdp->data_meta
@@ -2205,7 +2193,7 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
 	net_prefetch(xdp->data_meta);
 
 	/* build an skb around the page buffer */
-	skb = napi_build_skb(xdp->data_hard_start, truesize);
+	skb = napi_build_skb(xdp->data_hard_start, xdp->frame_sz);
 	if (unlikely(!skb))
 		return NULL;
 
@@ -2216,7 +2204,7 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
 		skb_metadata_set(skb, metasize);
 
 	/* buffer is used by skb, update page_offset */
-	i40e_rx_buffer_flip(rx_buffer, truesize);
+	i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
 
 	return skb;
 }
-- 
cgit v1.2.3-70-g09d2


From e9031f2da1aef34b0b4c659ead613c335b46ae92 Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:13 +0530
Subject: i40e: introduce next_to_process to i40e_ring

Add a new field called next_to_process in the i40e_ring that is
advanced for every buffer and change the semantics of next_to_clean to
point to the first buffer of a packet. Driver will use next_to_process
in the same way next_to_clean was used previously.

For the non multi-buffer case, next_to_process and next_to_clean will
always be the same since each packet consists of a single buffer.

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c | 26 +++++++++++++++-----------
 drivers/net/ethernet/intel/i40e/i40e_txrx.h |  4 ++++
 2 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index e34595ca4fbe..7fa35ff52689 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1524,6 +1524,7 @@ skip_free:
 
 	rx_ring->next_to_alloc = 0;
 	rx_ring->next_to_clean = 0;
+	rx_ring->next_to_process = 0;
 	rx_ring->next_to_use = 0;
 }
 
@@ -1576,6 +1577,7 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring)
 
 	rx_ring->next_to_alloc = 0;
 	rx_ring->next_to_clean = 0;
+	rx_ring->next_to_process = 0;
 	rx_ring->next_to_use = 0;
 
 	/* XDP RX-queue info only needed for RX rings exposed to XDP */
@@ -2076,7 +2078,7 @@ static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring *rx_ring,
 {
 	struct i40e_rx_buffer *rx_buffer;
 
-	rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean);
+	rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_process);
 	rx_buffer->page_count =
 #if (PAGE_SIZE < 8192)
 		page_count(rx_buffer->page);
@@ -2375,16 +2377,16 @@ void i40e_finalize_xdp_rx(struct i40e_ring *rx_ring, unsigned int xdp_res)
 }
 
 /**
- * i40e_inc_ntc: Advance the next_to_clean index
+ * i40e_inc_ntp: Advance the next_to_process index
  * @rx_ring: Rx ring
  **/
-static void i40e_inc_ntc(struct i40e_ring *rx_ring)
+static void i40e_inc_ntp(struct i40e_ring *rx_ring)
 {
-	u32 ntc = rx_ring->next_to_clean + 1;
+	u32 ntp = rx_ring->next_to_process + 1;
 
-	ntc = (ntc < rx_ring->count) ? ntc : 0;
-	rx_ring->next_to_clean = ntc;
-	prefetch(I40E_RX_DESC(rx_ring, ntc));
+	ntp = (ntp < rx_ring->count) ? ntp : 0;
+	rx_ring->next_to_process = ntp;
+	prefetch(I40E_RX_DESC(rx_ring, ntp));
 }
 
 /**
@@ -2421,6 +2423,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 	xdp_prog = READ_ONCE(rx_ring->xdp_prog);
 
 	while (likely(total_rx_packets < (unsigned int)budget)) {
+		u16 ntp = rx_ring->next_to_process;
 		struct i40e_rx_buffer *rx_buffer;
 		union i40e_rx_desc *rx_desc;
 		unsigned int size;
@@ -2433,7 +2436,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 			cleaned_count = 0;
 		}
 
-		rx_desc = I40E_RX_DESC(rx_ring, rx_ring->next_to_clean);
+		rx_desc = I40E_RX_DESC(rx_ring, ntp);
 
 		/* status_error_len will always be zero for unused descriptors
 		 * because it's cleared in cleanup, and overlaps with hdr_addr
@@ -2452,8 +2455,8 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 			i40e_clean_programming_status(rx_ring,
 						      rx_desc->raw.qword[0],
 						      qword);
-			rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean);
-			i40e_inc_ntc(rx_ring);
+			rx_buffer = i40e_rx_bi(rx_ring, ntp);
+			i40e_inc_ntp(rx_ring);
 			i40e_reuse_rx_page(rx_ring, rx_buffer);
 			cleaned_count++;
 			continue;
@@ -2509,7 +2512,8 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 		i40e_put_rx_buffer(rx_ring, rx_buffer);
 		cleaned_count++;
 
-		i40e_inc_ntc(rx_ring);
+		i40e_inc_ntp(rx_ring);
+		rx_ring->next_to_clean = rx_ring->next_to_process;
 		if (i40e_is_non_eop(rx_ring, rx_desc))
 			continue;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index eec4a4a99b9c..75888a9ab06d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -337,6 +337,10 @@ struct i40e_ring {
 	u8 dcb_tc;			/* Traffic class of ring */
 	u8 __iomem *tail;
 
+	/* Next descriptor to be processed; next_to_clean is updated only on
+	 * processing EOP descriptor
+	 */
+	u16 next_to_process;
 	/* high bit set means dynamic, use accessor routines to read/write.
 	 * hardware only supports 2us resolution for the ITR registers.
 	 * these values always store the USER setting, and must be converted
-- 
cgit v1.2.3-70-g09d2


From 01aa49e31e1674e22dd9c868ca6b4b945acd621e Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:14 +0530
Subject: i40e: add xdp_buff to i40e_ring struct

Store xdp_buff on Rx ring struct in preparation for XDP multi-buffer
support. This will allow us to combine fragmented frames across
separate NAPI cycles in the same way as currently skb fragments are
handled. This means that skb pointer on Rx ring will become redundant
and will be removed in a later patch. As a consequence i40e_trace() now
uses xdp instead of skb pointer.

Truesize only needs to be calculated for page sizes bigger than 4k as it
is always half-page for 4k pages. With xdp_buff on ring, frame size can
now be set during xdp_init_buff() and need not be repopulated in each
NAPI call for 4k pages. As a consequence i40e_rx_frame_truesize() is now
used only for bigger pages.

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c  |  2 ++
 drivers/net/ethernet/intel/i40e/i40e_trace.h | 20 ++++++++---------
 drivers/net/ethernet/intel/i40e/i40e_txrx.c  | 33 +++++++++++-----------------
 drivers/net/ethernet/intel/i40e/i40e_txrx.h  |  7 ++++++
 4 files changed, 32 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index e8cf5644bf10..f2c151a01e77 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3616,6 +3616,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring)
 		}
 	}
 
+	xdp_init_buff(&ring->xdp, i40e_rx_pg_size(ring) / 2, &ring->xdp_rxq);
+
 	rx_ctx.dbuff = DIV_ROUND_UP(ring->rx_buf_len,
 				    BIT_ULL(I40E_RXQ_CTX_DBUFF_SHIFT));
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_trace.h b/drivers/net/ethernet/intel/i40e/i40e_trace.h
index 79d587ad5409..33b4e30f5e00 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_trace.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_trace.h
@@ -162,45 +162,45 @@ DECLARE_EVENT_CLASS(
 
 	TP_PROTO(struct i40e_ring *ring,
 		 union i40e_16byte_rx_desc *desc,
-		 struct sk_buff *skb),
+		 struct xdp_buff *xdp),
 
-	TP_ARGS(ring, desc, skb),
+	TP_ARGS(ring, desc, xdp),
 
 	TP_STRUCT__entry(
 		__field(void*, ring)
 		__field(void*, desc)
-		__field(void*, skb)
+		__field(void*, xdp)
 		__string(devname, ring->netdev->name)
 	),
 
 	TP_fast_assign(
 		__entry->ring = ring;
 		__entry->desc = desc;
-		__entry->skb = skb;
+		__entry->xdp = xdp;
 		__assign_str(devname, ring->netdev->name);
 	),
 
 	TP_printk(
-		"netdev: %s ring: %p desc: %p skb %p",
+		"netdev: %s ring: %p desc: %p xdp %p",
 		__get_str(devname), __entry->ring,
-		__entry->desc, __entry->skb)
+		__entry->desc, __entry->xdp)
 );
 
 DEFINE_EVENT(
 	i40e_rx_template, i40e_clean_rx_irq,
 	TP_PROTO(struct i40e_ring *ring,
 		 union i40e_16byte_rx_desc *desc,
-		 struct sk_buff *skb),
+		 struct xdp_buff *xdp),
 
-	TP_ARGS(ring, desc, skb));
+	TP_ARGS(ring, desc, xdp));
 
 DEFINE_EVENT(
 	i40e_rx_template, i40e_clean_rx_irq_rx,
 	TP_PROTO(struct i40e_ring *ring,
 		 union i40e_16byte_rx_desc *desc,
-		 struct sk_buff *skb),
+		 struct xdp_buff *xdp),
 
-	TP_ARGS(ring, desc, skb));
+	TP_ARGS(ring, desc, xdp));
 
 DECLARE_EVENT_CLASS(
 	i40e_xmit_template,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 7fa35ff52689..5544c2d43a92 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1619,21 +1619,19 @@ void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val)
 	writel(val, rx_ring->tail);
 }
 
+#if (PAGE_SIZE >= 8192)
 static unsigned int i40e_rx_frame_truesize(struct i40e_ring *rx_ring,
 					   unsigned int size)
 {
 	unsigned int truesize;
 
-#if (PAGE_SIZE < 8192)
-	truesize = i40e_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */
-#else
 	truesize = rx_ring->rx_offset ?
 		SKB_DATA_ALIGN(size + rx_ring->rx_offset) +
 		SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) :
 		SKB_DATA_ALIGN(size);
-#endif
 	return truesize;
 }
+#endif
 
 /**
  * i40e_alloc_mapped_page - recycle or make a new page
@@ -2405,21 +2403,16 @@ static void i40e_inc_ntp(struct i40e_ring *rx_ring)
 static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 			     unsigned int *rx_cleaned)
 {
-	unsigned int total_rx_bytes = 0, total_rx_packets = 0, frame_sz = 0;
+	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 	u16 cleaned_count = I40E_DESC_UNUSED(rx_ring);
 	unsigned int offset = rx_ring->rx_offset;
+	struct xdp_buff *xdp = &rx_ring->xdp;
 	struct sk_buff *skb = rx_ring->skb;
 	unsigned int xdp_xmit = 0;
 	struct bpf_prog *xdp_prog;
 	bool failure = false;
-	struct xdp_buff xdp;
 	int xdp_res = 0;
 
-#if (PAGE_SIZE < 8192)
-	frame_sz = i40e_rx_frame_truesize(rx_ring, 0);
-#endif
-	xdp_init_buff(&xdp, frame_sz, &rx_ring->xdp_rxq);
-
 	xdp_prog = READ_ONCE(rx_ring->xdp_prog);
 
 	while (likely(total_rx_packets < (unsigned int)budget)) {
@@ -2467,7 +2460,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 		if (!size)
 			break;
 
-		i40e_trace(clean_rx_irq, rx_ring, rx_desc, skb);
+		i40e_trace(clean_rx_irq, rx_ring, rx_desc, xdp);
 		rx_buffer = i40e_get_rx_buffer(rx_ring, size);
 
 		/* retrieve a buffer from the ring */
@@ -2476,19 +2469,19 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 
 			hard_start = page_address(rx_buffer->page) +
 				     rx_buffer->page_offset - offset;
-			xdp_prepare_buff(&xdp, hard_start, offset, size, true);
-			xdp_buff_clear_frags_flag(&xdp);
+			xdp_prepare_buff(xdp, hard_start, offset, size, true);
+			xdp_buff_clear_frags_flag(xdp);
 #if (PAGE_SIZE > 4096)
 			/* At larger PAGE_SIZE, frame_sz depend on len size */
-			xdp.frame_sz = i40e_rx_frame_truesize(rx_ring, size);
+			xdp->frame_sz = i40e_rx_frame_truesize(rx_ring, size);
 #endif
-			xdp_res = i40e_run_xdp(rx_ring, &xdp, xdp_prog);
+			xdp_res = i40e_run_xdp(rx_ring, xdp, xdp_prog);
 		}
 
 		if (xdp_res) {
 			if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) {
 				xdp_xmit |= xdp_res;
-				i40e_rx_buffer_flip(rx_buffer, xdp.frame_sz);
+				i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
 			} else {
 				rx_buffer->pagecnt_bias++;
 			}
@@ -2497,9 +2490,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 		} else if (skb) {
 			i40e_add_rx_frag(rx_ring, rx_buffer, skb, size);
 		} else if (ring_uses_build_skb(rx_ring)) {
-			skb = i40e_build_skb(rx_ring, rx_buffer, &xdp);
+			skb = i40e_build_skb(rx_ring, rx_buffer, xdp);
 		} else {
-			skb = i40e_construct_skb(rx_ring, rx_buffer, &xdp);
+			skb = i40e_construct_skb(rx_ring, rx_buffer, xdp);
 		}
 
 		/* exit if we failed to retrieve a buffer */
@@ -2528,7 +2521,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 		/* populate checksum, VLAN, and protocol */
 		i40e_process_skb_fields(rx_ring, rx_desc, skb);
 
-		i40e_trace(clean_rx_irq_rx, rx_ring, rx_desc, skb);
+		i40e_trace(clean_rx_irq_rx, rx_ring, rx_desc, xdp);
 		napi_gro_receive(&rx_ring->q_vector->napi, skb);
 		skb = NULL;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 75888a9ab06d..55da9fde9d17 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -337,6 +337,13 @@ struct i40e_ring {
 	u8 dcb_tc;			/* Traffic class of ring */
 	u8 __iomem *tail;
 
+	/* Storing xdp_buff on ring helps in saving the state of partially built
+	 * packet when i40e_clean_rx_ring_irq() must return before it sees EOP
+	 * and to resume packet building for this ring in the next call to
+	 * i40e_clean_rx_ring_irq().
+	 */
+	struct xdp_buff xdp;
+
 	/* Next descriptor to be processed; next_to_clean is updated only on
 	 * processing EOP descriptor
 	 */
-- 
cgit v1.2.3-70-g09d2


From e213ced19befc09d6d6913799053b67896596cd1 Mon Sep 17 00:00:00 2001
From: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Date: Fri, 10 Mar 2023 00:26:15 +0530
Subject: i40e: add support for XDP multi-buffer Rx

This patch adds multi-buffer support for the i40e_driver.

i40e_clean_rx_irq() is modified to collate all the buffers of a packet
before calling the XDP program. xdp_buff is built for the first frag of
the packet and subsequent frags are added to it. 'next_to_process' is
incremented for all non-EOP frags while 'next_to_clean' stays at the
first descriptor of the packet. XDP program is called only on receiving
EOP frag.

New functions are added for adding frags to xdp_buff and for post
processing of the buffers once the xdp prog has run. For XDP_PASS this
results in a skb with multiple fragments.

i40e_build_skb() builds the skb around xdp buffer that already contains
frags data. So i40e_add_rx_frag() helper function is now removed. Since
fields before 'dataref' in skb_shared_info are cleared during
napi_skb_build(), xdp_update_skb_shared_info() is called to set those.

For i40e_construct_skb(), all the frags data needs to be copied from
xdp_buffer's shared_skb_info to newly constructed skb's shared_skb_info.

This also means 'skb' does not need to be preserved across i40e_napi_poll()
calls and hence is removed from i40e_ring structure.

Previously i40e_alloc_rx_buffers() was called for every 32 cleaned
buffers. For multi-buffers this may not be optimal as there may be more
cleaned buffers in each i40e_clean_rx_irq() call. So this is now called
when at least half of the ring size has been cleaned.

Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c |   7 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c | 314 ++++++++++++++++++----------
 drivers/net/ethernet/intel/i40e/i40e_txrx.h |   8 -
 3 files changed, 211 insertions(+), 118 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index f2c151a01e77..053be338e4ba 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2919,7 +2919,7 @@ static int i40e_max_vsi_frame_size(struct i40e_vsi *vsi,
 	u16 rx_buf_len = i40e_calculate_vsi_rx_buf_len(vsi);
 	u16 chain_len;
 
-	if (xdp_prog)
+	if (xdp_prog && !xdp_prog->aux->xdp_has_frags)
 		chain_len = 1;
 	else
 		chain_len = I40E_MAX_CHAINED_RX_BUFFERS;
@@ -13337,7 +13337,7 @@ static int i40e_xdp_setup(struct i40e_vsi *vsi, struct bpf_prog *prog,
 
 	/* Don't allow frames that span over multiple buffers */
 	if (vsi->netdev->mtu > frame_size - I40E_PACKET_HDR_PAD) {
-		NL_SET_ERR_MSG_MOD(extack, "MTU too large to enable XDP");
+		NL_SET_ERR_MSG_MOD(extack, "MTU too large for linear frames and XDP prog does not support frags");
 		return -EINVAL;
 	}
 
@@ -13823,7 +13823,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 
 		netdev->xdp_features = NETDEV_XDP_ACT_BASIC |
 				       NETDEV_XDP_ACT_REDIRECT |
-				       NETDEV_XDP_ACT_XSK_ZEROCOPY;
+				       NETDEV_XDP_ACT_XSK_ZEROCOPY |
+				       NETDEV_XDP_ACT_RX_SG;
 	} else {
 		/* Relate the VSI_VMDQ name to the VSI_MAIN name. Note that we
 		 * are still limited by IFNAMSIZ, but we're adding 'v%d\0' to
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 5544c2d43a92..32cce90abbb4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1477,9 +1477,6 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring)
 	if (!rx_ring->rx_bi)
 		return;
 
-	dev_kfree_skb(rx_ring->skb);
-	rx_ring->skb = NULL;
-
 	if (rx_ring->xsk_pool) {
 		i40e_xsk_clean_rx_ring(rx_ring);
 		goto skip_free;
@@ -2033,36 +2030,6 @@ static void i40e_rx_buffer_flip(struct i40e_rx_buffer *rx_buffer,
 #endif
 }
 
-/**
- * i40e_add_rx_frag - Add contents of Rx buffer to sk_buff
- * @rx_ring: rx descriptor ring to transact packets on
- * @rx_buffer: buffer containing page to add
- * @skb: sk_buff to place the data into
- * @size: packet length from rx_desc
- *
- * This function will add the data contained in rx_buffer->page to the skb.
- * It will just attach the page as a frag to the skb.
- *
- * The function will then update the page offset.
- **/
-static void i40e_add_rx_frag(struct i40e_ring *rx_ring,
-			     struct i40e_rx_buffer *rx_buffer,
-			     struct sk_buff *skb,
-			     unsigned int size)
-{
-#if (PAGE_SIZE < 8192)
-	unsigned int truesize = i40e_rx_pg_size(rx_ring) / 2;
-#else
-	unsigned int truesize = SKB_DATA_ALIGN(size + rx_ring->rx_offset);
-#endif
-
-	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
-			rx_buffer->page_offset, size, truesize);
-
-	/* page is being used so we must update the page offset */
-	i40e_rx_buffer_flip(rx_buffer, truesize);
-}
-
 /**
  * i40e_get_rx_buffer - Fetch Rx buffer and synchronize data for use
  * @rx_ring: rx descriptor ring to transact packets on
@@ -2099,20 +2066,82 @@ static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring *rx_ring,
 }
 
 /**
- * i40e_construct_skb - Allocate skb and populate it
+ * i40e_put_rx_buffer - Clean up used buffer and either recycle or free
  * @rx_ring: rx descriptor ring to transact packets on
  * @rx_buffer: rx buffer to pull data from
+ *
+ * This function will clean up the contents of the rx_buffer.  It will
+ * either recycle the buffer or unmap it and free the associated resources.
+ */
+static void i40e_put_rx_buffer(struct i40e_ring *rx_ring,
+			       struct i40e_rx_buffer *rx_buffer)
+{
+	if (i40e_can_reuse_rx_page(rx_buffer, &rx_ring->rx_stats)) {
+		/* hand second half of page back to the ring */
+		i40e_reuse_rx_page(rx_ring, rx_buffer);
+	} else {
+		/* we are not reusing the buffer so unmap it */
+		dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
+				     i40e_rx_pg_size(rx_ring),
+				     DMA_FROM_DEVICE, I40E_RX_DMA_ATTR);
+		__page_frag_cache_drain(rx_buffer->page,
+					rx_buffer->pagecnt_bias);
+		/* clear contents of buffer_info */
+		rx_buffer->page = NULL;
+	}
+}
+
+/**
+ * i40e_process_rx_buffs- Processing of buffers post XDP prog or on error
+ * @rx_ring: Rx descriptor ring to transact packets on
+ * @xdp_res: Result of the XDP program
+ * @xdp: xdp_buff pointing to the data
+ **/
+static void i40e_process_rx_buffs(struct i40e_ring *rx_ring, int xdp_res,
+				  struct xdp_buff *xdp)
+{
+	u32 next = rx_ring->next_to_clean;
+	struct i40e_rx_buffer *rx_buffer;
+
+	xdp->flags = 0;
+
+	while (1) {
+		rx_buffer = i40e_rx_bi(rx_ring, next);
+		if (++next == rx_ring->count)
+			next = 0;
+
+		if (!rx_buffer->page)
+			continue;
+
+		if (xdp_res == I40E_XDP_CONSUMED)
+			rx_buffer->pagecnt_bias++;
+		else
+			i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
+
+		/* EOP buffer will be put in i40e_clean_rx_irq() */
+		if (next == rx_ring->next_to_process)
+			return;
+
+		i40e_put_rx_buffer(rx_ring, rx_buffer);
+	}
+}
+
+/**
+ * i40e_construct_skb - Allocate skb and populate it
+ * @rx_ring: rx descriptor ring to transact packets on
  * @xdp: xdp_buff pointing to the data
+ * @nr_frags: number of buffers for the packet
  *
  * This function allocates an skb.  It then populates it with the page
  * data from the current receive descriptor, taking care to set up the
  * skb correctly.
  */
 static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring,
-					  struct i40e_rx_buffer *rx_buffer,
-					  struct xdp_buff *xdp)
+					  struct xdp_buff *xdp,
+					  u32 nr_frags)
 {
 	unsigned int size = xdp->data_end - xdp->data;
+	struct i40e_rx_buffer *rx_buffer;
 	unsigned int headlen;
 	struct sk_buff *skb;
 
@@ -2152,13 +2181,17 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring,
 	memcpy(__skb_put(skb, headlen), xdp->data,
 	       ALIGN(headlen, sizeof(long)));
 
+	rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean);
 	/* update all of the pointers */
 	size -= headlen;
 	if (size) {
+		if (unlikely(nr_frags >= MAX_SKB_FRAGS)) {
+			dev_kfree_skb(skb);
+			return NULL;
+		}
 		skb_add_rx_frag(skb, 0, rx_buffer->page,
 				rx_buffer->page_offset + headlen,
 				size, xdp->frame_sz);
-
 		/* buffer is used by skb, update page_offset */
 		i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
 	} else {
@@ -2166,21 +2199,40 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring,
 		rx_buffer->pagecnt_bias++;
 	}
 
+	if (unlikely(xdp_buff_has_frags(xdp))) {
+		struct skb_shared_info *sinfo, *skinfo = skb_shinfo(skb);
+
+		sinfo = xdp_get_shared_info_from_buff(xdp);
+		memcpy(&skinfo->frags[skinfo->nr_frags], &sinfo->frags[0],
+		       sizeof(skb_frag_t) * nr_frags);
+
+		xdp_update_skb_shared_info(skb, skinfo->nr_frags + nr_frags,
+					   sinfo->xdp_frags_size,
+					   nr_frags * xdp->frame_sz,
+					   xdp_buff_is_frag_pfmemalloc(xdp));
+
+		/* First buffer has already been processed, so bump ntc */
+		if (++rx_ring->next_to_clean == rx_ring->count)
+			rx_ring->next_to_clean = 0;
+
+		i40e_process_rx_buffs(rx_ring, I40E_XDP_PASS, xdp);
+	}
+
 	return skb;
 }
 
 /**
  * i40e_build_skb - Build skb around an existing buffer
  * @rx_ring: Rx descriptor ring to transact packets on
- * @rx_buffer: Rx buffer to pull data from
  * @xdp: xdp_buff pointing to the data
+ * @nr_frags: number of buffers for the packet
  *
  * This function builds an skb around an existing Rx buffer, taking care
  * to set up the skb correctly and avoid any memcpy overhead.
  */
 static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
-				      struct i40e_rx_buffer *rx_buffer,
-				      struct xdp_buff *xdp)
+				      struct xdp_buff *xdp,
+				      u32 nr_frags)
 {
 	unsigned int metasize = xdp->data - xdp->data_meta;
 	struct sk_buff *skb;
@@ -2203,36 +2255,25 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring,
 	if (metasize)
 		skb_metadata_set(skb, metasize);
 
-	/* buffer is used by skb, update page_offset */
-	i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
+	if (unlikely(xdp_buff_has_frags(xdp))) {
+		struct skb_shared_info *sinfo;
 
-	return skb;
-}
+		sinfo = xdp_get_shared_info_from_buff(xdp);
+		xdp_update_skb_shared_info(skb, nr_frags,
+					   sinfo->xdp_frags_size,
+					   nr_frags * xdp->frame_sz,
+					   xdp_buff_is_frag_pfmemalloc(xdp));
 
-/**
- * i40e_put_rx_buffer - Clean up used buffer and either recycle or free
- * @rx_ring: rx descriptor ring to transact packets on
- * @rx_buffer: rx buffer to pull data from
- *
- * This function will clean up the contents of the rx_buffer.  It will
- * either recycle the buffer or unmap it and free the associated resources.
- */
-static void i40e_put_rx_buffer(struct i40e_ring *rx_ring,
-			       struct i40e_rx_buffer *rx_buffer)
-{
-	if (i40e_can_reuse_rx_page(rx_buffer, &rx_ring->rx_stats)) {
-		/* hand second half of page back to the ring */
-		i40e_reuse_rx_page(rx_ring, rx_buffer);
+		i40e_process_rx_buffs(rx_ring, I40E_XDP_PASS, xdp);
 	} else {
-		/* we are not reusing the buffer so unmap it */
-		dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
-				     i40e_rx_pg_size(rx_ring),
-				     DMA_FROM_DEVICE, I40E_RX_DMA_ATTR);
-		__page_frag_cache_drain(rx_buffer->page,
-					rx_buffer->pagecnt_bias);
-		/* clear contents of buffer_info */
-		rx_buffer->page = NULL;
+		struct i40e_rx_buffer *rx_buffer;
+
+		rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean);
+		/* buffer is used by skb, update page_offset */
+		i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
 	}
+
+	return skb;
 }
 
 /**
@@ -2387,6 +2428,55 @@ static void i40e_inc_ntp(struct i40e_ring *rx_ring)
 	prefetch(I40E_RX_DESC(rx_ring, ntp));
 }
 
+/**
+ * i40e_add_xdp_frag: Add a frag to xdp_buff
+ * @xdp: xdp_buff pointing to the data
+ * @nr_frags: return number of buffers for the packet
+ * @rx_buffer: rx_buffer holding data of the current frag
+ * @size: size of data of current frag
+ */
+static int i40e_add_xdp_frag(struct xdp_buff *xdp, u32 *nr_frags,
+			     struct i40e_rx_buffer *rx_buffer, u32 size)
+{
+	struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
+
+	if (!xdp_buff_has_frags(xdp)) {
+		sinfo->nr_frags = 0;
+		sinfo->xdp_frags_size = 0;
+		xdp_buff_set_frags_flag(xdp);
+	} else if (unlikely(sinfo->nr_frags >= MAX_SKB_FRAGS)) {
+		/* Overflowing packet: All frags need to be dropped */
+		return -ENOMEM;
+	}
+
+	__skb_fill_page_desc_noacc(sinfo, sinfo->nr_frags++, rx_buffer->page,
+				   rx_buffer->page_offset, size);
+
+	sinfo->xdp_frags_size += size;
+
+	if (page_is_pfmemalloc(rx_buffer->page))
+		xdp_buff_set_frag_pfmemalloc(xdp);
+	*nr_frags = sinfo->nr_frags;
+
+	return 0;
+}
+
+/**
+ * i40e_consume_xdp_buff - Consume all the buffers of the packet and update ntc
+ * @rx_ring: rx descriptor ring to transact packets on
+ * @xdp: xdp_buff pointing to the data
+ * @rx_buffer: rx_buffer of eop desc
+ */
+static void i40e_consume_xdp_buff(struct i40e_ring *rx_ring,
+				  struct xdp_buff *xdp,
+				  struct i40e_rx_buffer *rx_buffer)
+{
+	i40e_process_rx_buffs(rx_ring, I40E_XDP_CONSUMED, xdp);
+	i40e_put_rx_buffer(rx_ring, rx_buffer);
+	rx_ring->next_to_clean = rx_ring->next_to_process;
+	xdp->data = NULL;
+}
+
 /**
  * i40e_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf
  * @rx_ring: rx descriptor ring to transact packets on
@@ -2405,9 +2495,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 {
 	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 	u16 cleaned_count = I40E_DESC_UNUSED(rx_ring);
+	u16 clean_threshold = rx_ring->count / 2;
 	unsigned int offset = rx_ring->rx_offset;
 	struct xdp_buff *xdp = &rx_ring->xdp;
-	struct sk_buff *skb = rx_ring->skb;
 	unsigned int xdp_xmit = 0;
 	struct bpf_prog *xdp_prog;
 	bool failure = false;
@@ -2419,11 +2509,14 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 		u16 ntp = rx_ring->next_to_process;
 		struct i40e_rx_buffer *rx_buffer;
 		union i40e_rx_desc *rx_desc;
+		struct sk_buff *skb;
 		unsigned int size;
+		u32 nfrags = 0;
+		bool neop;
 		u64 qword;
 
 		/* return some buffers to hardware, one at a time is too slow */
-		if (cleaned_count >= I40E_RX_BUFFER_WRITE) {
+		if (cleaned_count >= clean_threshold) {
 			failure = failure ||
 				  i40e_alloc_rx_buffers(rx_ring, cleaned_count);
 			cleaned_count = 0;
@@ -2461,76 +2554,83 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
 			break;
 
 		i40e_trace(clean_rx_irq, rx_ring, rx_desc, xdp);
+		/* retrieve a buffer from the ring */
 		rx_buffer = i40e_get_rx_buffer(rx_ring, size);
 
-		/* retrieve a buffer from the ring */
-		if (!skb) {
+		neop = i40e_is_non_eop(rx_ring, rx_desc);
+		i40e_inc_ntp(rx_ring);
+
+		if (!xdp->data) {
 			unsigned char *hard_start;
 
 			hard_start = page_address(rx_buffer->page) +
 				     rx_buffer->page_offset - offset;
 			xdp_prepare_buff(xdp, hard_start, offset, size, true);
-			xdp_buff_clear_frags_flag(xdp);
 #if (PAGE_SIZE > 4096)
 			/* At larger PAGE_SIZE, frame_sz depend on len size */
 			xdp->frame_sz = i40e_rx_frame_truesize(rx_ring, size);
 #endif
-			xdp_res = i40e_run_xdp(rx_ring, xdp, xdp_prog);
+		} else if (i40e_add_xdp_frag(xdp, &nfrags, rx_buffer, size) &&
+			   !neop) {
+			/* Overflowing packet: Drop all frags on EOP */
+			i40e_consume_xdp_buff(rx_ring, xdp, rx_buffer);
+			break;
 		}
 
+		if (neop)
+			continue;
+
+		xdp_res = i40e_run_xdp(rx_ring, xdp, xdp_prog);
+
 		if (xdp_res) {
-			if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) {
-				xdp_xmit |= xdp_res;
+			xdp_xmit |= xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR);
+
+			if (unlikely(xdp_buff_has_frags(xdp))) {
+				i40e_process_rx_buffs(rx_ring, xdp_res, xdp);
+				size = xdp_get_buff_len(xdp);
+			} else if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) {
 				i40e_rx_buffer_flip(rx_buffer, xdp->frame_sz);
 			} else {
 				rx_buffer->pagecnt_bias++;
 			}
 			total_rx_bytes += size;
-			total_rx_packets++;
-		} else if (skb) {
-			i40e_add_rx_frag(rx_ring, rx_buffer, skb, size);
-		} else if (ring_uses_build_skb(rx_ring)) {
-			skb = i40e_build_skb(rx_ring, rx_buffer, xdp);
 		} else {
-			skb = i40e_construct_skb(rx_ring, rx_buffer, xdp);
-		}
+			if (ring_uses_build_skb(rx_ring))
+				skb = i40e_build_skb(rx_ring, xdp, nfrags);
+			else
+				skb = i40e_construct_skb(rx_ring, xdp, nfrags);
+
+			/* drop if we failed to retrieve a buffer */
+			if (!skb) {
+				rx_ring->rx_stats.alloc_buff_failed++;
+				i40e_consume_xdp_buff(rx_ring, xdp, rx_buffer);
+				break;
+			}
 
-		/* exit if we failed to retrieve a buffer */
-		if (!xdp_res && !skb) {
-			rx_ring->rx_stats.alloc_buff_failed++;
-			rx_buffer->pagecnt_bias++;
-			break;
-		}
+			if (i40e_cleanup_headers(rx_ring, skb, rx_desc))
+				goto process_next;
 
-		i40e_put_rx_buffer(rx_ring, rx_buffer);
-		cleaned_count++;
+			/* probably a little skewed due to removing CRC */
+			total_rx_bytes += skb->len;
 
-		i40e_inc_ntp(rx_ring);
-		rx_ring->next_to_clean = rx_ring->next_to_process;
-		if (i40e_is_non_eop(rx_ring, rx_desc))
-			continue;
+			/* populate checksum, VLAN, and protocol */
+			i40e_process_skb_fields(rx_ring, rx_desc, skb);
 
-		if (xdp_res || i40e_cleanup_headers(rx_ring, skb, rx_desc)) {
-			skb = NULL;
-			continue;
+			i40e_trace(clean_rx_irq_rx, rx_ring, rx_desc, xdp);
+			napi_gro_receive(&rx_ring->q_vector->napi, skb);
 		}
 
-		/* probably a little skewed due to removing CRC */
-		total_rx_bytes += skb->len;
-
-		/* populate checksum, VLAN, and protocol */
-		i40e_process_skb_fields(rx_ring, rx_desc, skb);
-
-		i40e_trace(clean_rx_irq_rx, rx_ring, rx_desc, xdp);
-		napi_gro_receive(&rx_ring->q_vector->napi, skb);
-		skb = NULL;
-
 		/* update budget accounting */
 		total_rx_packets++;
+process_next:
+		cleaned_count += nfrags + 1;
+		i40e_put_rx_buffer(rx_ring, rx_buffer);
+		rx_ring->next_to_clean = rx_ring->next_to_process;
+
+		xdp->data = NULL;
 	}
 
 	i40e_finalize_xdp_rx(rx_ring, xdp_xmit);
-	rx_ring->skb = skb;
 
 	i40e_update_rx_stats(rx_ring, total_rx_bytes, total_rx_packets);
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 55da9fde9d17..8c3d24012c54 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -392,14 +392,6 @@ struct i40e_ring {
 
 	struct rcu_head rcu;		/* to avoid race on free */
 	u16 next_to_alloc;
-	struct sk_buff *skb;		/* When i40e_clean_rx_ring_irq() must
-					 * return before it sees the EOP for
-					 * the current packet, we save that skb
-					 * here and resume receiving this
-					 * packet the next time
-					 * i40e_clean_rx_ring_irq() is called
-					 * for this ring.
-					 */
 
 	struct i40e_channel *ch;
 	u16 rx_offset;
-- 
cgit v1.2.3-70-g09d2


From 63d78b7e8ca2d0eb8c687a355fa19d01b6fcc723 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Thu, 9 Mar 2023 17:24:10 -0800
Subject: selftests/bpf: Workaround verification failure for
 fexit_bpf2bpf/func_replace_return_code

With latest llvm17, selftest fexit_bpf2bpf/func_replace_return_code
has the following verification failure:

  0: R1=ctx(off=0,imm=0) R10=fp0
  ; int connect_v4_prog(struct bpf_sock_addr *ctx)
  0: (bf) r7 = r1                       ; R1=ctx(off=0,imm=0) R7_w=ctx(off=0,imm=0)
  1: (b4) w6 = 0                        ; R6_w=0
  ; memset(&tuple.ipv4.saddr, 0, sizeof(tuple.ipv4.saddr));
  ...
  ; return do_bind(ctx) ? 1 : 0;
  179: (bf) r1 = r7                     ; R1=ctx(off=0,imm=0) R7=ctx(off=0,imm=0)
  180: (85) call pc+147
  Func#3 is global and valid. Skipping.
  181: R0_w=scalar()
  181: (bc) w6 = w0                     ; R0_w=scalar() R6_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
  182: (05) goto pc-129
  ; }
  54: (bc) w0 = w6                      ; R0_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff)) R6_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
  55: (95) exit
  At program exit the register R0 has value (0x0; 0xffffffff) should have been in (0x0; 0x1)
  processed 281 insns (limit 1000000) max_states_per_insn 1 total_states 26 peak_states 26 mark_read 13
  -- END PROG LOAD LOG --
  libbpf: prog 'connect_v4_prog': failed to load: -22

The corresponding source code:

  __attribute__ ((noinline))
  int do_bind(struct bpf_sock_addr *ctx)
  {
        struct sockaddr_in sa = {};

        sa.sin_family = AF_INET;
        sa.sin_port = bpf_htons(0);
        sa.sin_addr.s_addr = bpf_htonl(SRC_REWRITE_IP4);

        if (bpf_bind(ctx, (struct sockaddr *)&sa, sizeof(sa)) != 0)
                return 0;

        return 1;
  }
  ...
  SEC("cgroup/connect4")
  int connect_v4_prog(struct bpf_sock_addr *ctx)
  {
  ...
        return do_bind(ctx) ? 1 : 0;
  }

Insn 180 is a call to 'do_bind'. The call's return value is also the return value
for the program. Since do_bind() returns 0/1, so it is legitimate for compiler to
optimize 'return do_bind(ctx) ? 1 : 0' to 'return do_bind(ctx)'. However, such
optimization breaks verifier as the return value of 'do_bind()' is marked as any
scalar which violates the requirement of prog return value 0/1.

There are two ways to fix this problem, (1) changing 'return 1' in do_bind() to
e.g. 'return 10' so the compiler has to do 'do_bind(ctx) ? 1 :0', or (2)
suggested by Andrii, marking do_bind() with __weak attribute so the compiler
cannot make any assumption on do_bind() return value.

This patch adopted adding __weak approach which is simpler and more resistant
to potential compiler optimizations.

Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230310012410.2920570-1-yhs@fb.com
---
 tools/testing/selftests/bpf/progs/connect4_prog.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/connect4_prog.c b/tools/testing/selftests/bpf/progs/connect4_prog.c
index ec25371de789..7ef49ec04838 100644
--- a/tools/testing/selftests/bpf/progs/connect4_prog.c
+++ b/tools/testing/selftests/bpf/progs/connect4_prog.c
@@ -32,7 +32,7 @@
 #define IFNAMSIZ 16
 #endif
 
-__attribute__ ((noinline))
+__attribute__ ((noinline)) __weak
 int do_bind(struct bpf_sock_addr *ctx)
 {
 	struct sockaddr_in sa = {};
-- 
cgit v1.2.3-70-g09d2


From 14296c7d72ec26d18cbe200a43c6d5b7d967ab89 Mon Sep 17 00:00:00 2001
From: Nick Alcock <nick.alcock@oracle.com>
Date: Wed, 8 Mar 2023 12:12:30 +0000
Subject: mctp: remove MODULE_LICENSE in non-modules

Since commit 8b41fc4454e ("kbuild: create modules.builtin without
Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
are used to identify modules. As a consequence, uses of the macro
in non-modules will cause modprobe to misidentify their containing
object file as a module when it is not (false positives), and modprobe
might succeed rather than failing with a suitable error message.

So remove it in the files in this commit, none of which can be built as
modules.

Signed-off-by: Nick Alcock <nick.alcock@oracle.com>
Suggested-by: Luis Chamberlain <mcgrof@kernel.org>
Cc: Hitomi Hasegawa <hasegawa-hitomi@fujitsu.com>
Cc: Jeremy Kerr <jk@codeconstruct.com.au>
Cc: Matt Johnston <matt@codeconstruct.com.au>
Link: https://lore.kernel.org/r/20230308121230.5354-2-nick.alcock@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/mctp/af_mctp.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c
index 3150f3f0c872..bb4bd0b6a4f7 100644
--- a/net/mctp/af_mctp.c
+++ b/net/mctp/af_mctp.c
@@ -704,7 +704,6 @@ subsys_initcall(mctp_init);
 module_exit(mctp_exit);
 
 MODULE_DESCRIPTION("MCTP core");
-MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Jeremy Kerr <jk@codeconstruct.com.au>");
 
 MODULE_ALIAS_NETPROTO(PF_MCTP);
-- 
cgit v1.2.3-70-g09d2


From efb5b62d72719fd1df1f927542c58d1e21d69e19 Mon Sep 17 00:00:00 2001
From: Nick Alcock <nick.alcock@oracle.com>
Date: Wed, 8 Mar 2023 12:12:29 +0000
Subject: lib: packing: remove MODULE_LICENSE in non-modules

Since commit 8b41fc4454e ("kbuild: create modules.builtin without
Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
are used to identify modules. As a consequence, uses of the macro
in non-modules will cause modprobe to misidentify their containing
object file as a module when it is not (false positives), and modprobe
might succeed rather than failing with a suitable error message.

So remove it in the files in this commit, none of which can be built as
modules.

Signed-off-by: Nick Alcock <nick.alcock@oracle.com>
Suggested-by: Luis Chamberlain <mcgrof@kernel.org>
Cc: Hitomi Hasegawa <hasegawa-hitomi@fujitsu.com>
Cc: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20230308121230.5354-1-nick.alcock@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 lib/packing.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/lib/packing.c b/lib/packing.c
index a96169237ae6..3f656167c17e 100644
--- a/lib/packing.c
+++ b/lib/packing.c
@@ -198,5 +198,4 @@ int packing(void *pbuf, u64 *uval, int startbit, int endbit, size_t pbuflen,
 }
 EXPORT_SYMBOL(packing);
 
-MODULE_LICENSE("GPL v2");
 MODULE_DESCRIPTION("Generic bitfield packing and unpacking");
-- 
cgit v1.2.3-70-g09d2


From 6978052448f9eb19f7b03243ac0416104e5ee50d Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 8 Mar 2023 15:20:06 +0100
Subject: netlink: remove unused 'compare' function

No users in the tree.  Tested with allmodconfig build.

Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230308142006.20879-1-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netlink.h  | 1 -
 net/netlink/af_netlink.c | 2 --
 net/netlink/af_netlink.h | 1 -
 3 files changed, 4 deletions(-)

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index c43ac7690eca..3e8743252167 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -50,7 +50,6 @@ struct netlink_kernel_cfg {
 	struct mutex	*cb_mutex;
 	int		(*bind)(struct net *net, int group);
 	void		(*unbind)(struct net *net, int group);
-	bool		(*compare)(struct net *net, struct sock *sk);
 };
 
 struct sock *__netlink_kernel_create(struct net *net, int unit,
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index c64277659753..877f1da1a8ac 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -2097,8 +2097,6 @@ __netlink_kernel_create(struct net *net, int unit, struct module *module,
 			nl_table[unit].bind = cfg->bind;
 			nl_table[unit].unbind = cfg->unbind;
 			nl_table[unit].flags = cfg->flags;
-			if (cfg->compare)
-				nl_table[unit].compare = cfg->compare;
 		}
 		nl_table[unit].registered = 1;
 	} else {
diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h
index 5f454c8de6a4..90a3198a9b7f 100644
--- a/net/netlink/af_netlink.h
+++ b/net/netlink/af_netlink.h
@@ -64,7 +64,6 @@ struct netlink_table {
 	struct module		*module;
 	int			(*bind)(struct net *net, int group);
 	void			(*unbind)(struct net *net, int group);
-	bool			(*compare)(struct net *net, struct sock *sock);
 	int			registered;
 };
 
-- 
cgit v1.2.3-70-g09d2


From 513bdd9473888ba4b1441923c2711cb5ecd06cfd Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Wed, 8 Mar 2023 21:19:55 +0100
Subject: net: phy: smsc: use phy_set_bits in smsc_phy_config_init

Simplify the code by using phy_set_bits().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/b64d9f86-d029-b911-bbe9-6ca6889399d7@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/smsc.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index feb6ac3939ac..ac951d67edcc 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -99,20 +99,13 @@ static irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev)
 static int smsc_phy_config_init(struct phy_device *phydev)
 {
 	struct smsc_phy_priv *priv = phydev->priv;
-	int rc;
 
 	if (!priv->energy_enable || phydev->irq != PHY_POLL)
 		return 0;
 
-	rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
-
-	if (rc < 0)
-		return rc;
-
-	/* Enable energy detect mode for this SMSC Transceivers */
-	rc = phy_write(phydev, MII_LAN83C185_CTRL_STATUS,
-		       rc | MII_LAN83C185_EDPWRDOWN);
-	return rc;
+	/* Enable energy detect power down mode */
+	return phy_set_bits(phydev, MII_LAN83C185_CTRL_STATUS,
+			    MII_LAN83C185_EDPWRDOWN);
 }
 
 static int smsc_phy_reset(struct phy_device *phydev)
-- 
cgit v1.2.3-70-g09d2


From 62423bd2d2e231951245d77740a58027a2d81ef9 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 8 Mar 2023 18:26:48 +0000
Subject: net: sched: remove qdisc_watchdog->last_expires

This field mirrors hrtimer softexpires, we can instead
use the existing helpers.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230308182648.1150762-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/pkt_sched.h | 1 -
 net/sched/sch_api.c     | 6 ++++--
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 2016839991a4..bb0bd69fb655 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -64,7 +64,6 @@ static inline psched_time_t psched_get_time(void)
 }
 
 struct qdisc_watchdog {
-	u64		last_expires;
 	struct hrtimer	timer;
 	struct Qdisc	*qdisc;
 };
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index aba789c30a2e..fdb8f429333d 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -639,14 +639,16 @@ void qdisc_watchdog_schedule_range_ns(struct qdisc_watchdog *wd, u64 expires,
 		return;
 
 	if (hrtimer_is_queued(&wd->timer)) {
+		u64 softexpires;
+
+		softexpires = ktime_to_ns(hrtimer_get_softexpires(&wd->timer));
 		/* If timer is already set in [expires, expires + delta_ns],
 		 * do not reprogram it.
 		 */
-		if (wd->last_expires - expires <= delta_ns)
+		if (softexpires - expires <= delta_ns)
 			return;
 	}
 
-	wd->last_expires = expires;
 	hrtimer_start_range_ns(&wd->timer,
 			       ns_to_ktime(expires),
 			       delta_ns,
-- 
cgit v1.2.3-70-g09d2


From 76b9bf965c98c9b53ef7420b3b11438dbd764f92 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Wed, 8 Mar 2023 11:23:13 +0200
Subject: neighbour: delete neigh_lookup_nodev as not used

neigh_lookup_nodev isn't used in the kernel after removal
of DECnet. So let's remove it.

Fixes: 1202cdd66531 ("Remove DECnet support from kernel")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://lore.kernel.org/r/eb5656200d7964b2d177a36b77efa3c597d6d72d.1678267343.git.leonro@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/neighbour.h |  2 --
 net/core/neighbour.c    | 31 -------------------------------
 2 files changed, 33 deletions(-)

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 2f2a6023fb0e..234799ca527e 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -336,8 +336,6 @@ void neigh_table_init(int index, struct neigh_table *tbl);
 int neigh_table_clear(int index, struct neigh_table *tbl);
 struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
 			       struct net_device *dev);
-struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
-				     const void *pkey);
 struct neighbour *__neigh_create(struct neigh_table *tbl, const void *pkey,
 				 struct net_device *dev, bool want_ref);
 static inline struct neighbour *neigh_create(struct neigh_table *tbl,
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 6798f6d2423b..0116b0ff91a7 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -627,37 +627,6 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
 }
 EXPORT_SYMBOL(neigh_lookup);
 
-struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
-				     const void *pkey)
-{
-	struct neighbour *n;
-	unsigned int key_len = tbl->key_len;
-	u32 hash_val;
-	struct neigh_hash_table *nht;
-
-	NEIGH_CACHE_STAT_INC(tbl, lookups);
-
-	rcu_read_lock_bh();
-	nht = rcu_dereference_bh(tbl->nht);
-	hash_val = tbl->hash(pkey, NULL, nht->hash_rnd) >> (32 - nht->hash_shift);
-
-	for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]);
-	     n != NULL;
-	     n = rcu_dereference_bh(n->next)) {
-		if (!memcmp(n->primary_key, pkey, key_len) &&
-		    net_eq(dev_net(n->dev), net)) {
-			if (!refcount_inc_not_zero(&n->refcnt))
-				n = NULL;
-			NEIGH_CACHE_STAT_INC(tbl, hits);
-			break;
-		}
-	}
-
-	rcu_read_unlock_bh();
-	return n;
-}
-EXPORT_SYMBOL(neigh_lookup_nodev);
-
 static struct neighbour *
 ___neigh_create(struct neigh_table *tbl, const void *pkey,
 		struct net_device *dev, u32 flags,
-- 
cgit v1.2.3-70-g09d2


From b3a8df9f27c094dd41fe88b7b97f64d8dcb7d696 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 8 Mar 2023 14:04:52 +0100
Subject: net: ethernet: ti: am65-cpsw: Convert to devm_of_phy_optional_get()

Use the new devm_of_phy_optional_get() helper instead of open-coding the
same operation.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/01605ea233ff7fc09bb0ea34fc8126af73db83f9.1678280599.git.geert+renesas@glider.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 4e3861c47708..25996826edab 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -1470,11 +1470,9 @@ static int am65_cpsw_init_serdes_phy(struct device *dev, struct device_node *por
 	struct phy *phy;
 	int ret;
 
-	phy = devm_of_phy_get(dev, port_np, name);
-	if (PTR_ERR(phy) == -ENODEV)
-		return 0;
-	if (IS_ERR(phy))
-		return PTR_ERR(phy);
+	phy = devm_of_phy_optional_get(dev, port_np, name);
+	if (IS_ERR_OR_NULL(phy))
+		return PTR_ERR_OR_ZERO(phy);
 
 	/* Serdes PHY exists. Store it. */
 	port->slave.serdes_phy = phy;
-- 
cgit v1.2.3-70-g09d2


From fd9c31f834416e3060061dbcb45ababaa25cdfe2 Mon Sep 17 00:00:00 2001
From: Jason Xing <kernelxing@tencent.com>
Date: Wed, 8 Mar 2023 10:11:53 +0800
Subject: udp: introduce __sk_mem_schedule() usage

Keep the accounting schema consistent across different protocols
with __sk_mem_schedule(). Besides, it adjusts a little bit on how
to calculate forward allocated memory compared to before. After
applied this patch, we could avoid receive path scheduling extra
amount of memory.

Link: https://lore.kernel.org/lkml/20230221110344.82818-1-kerneljasonxing@gmail.com/
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230308021153.99777-1-kerneljasonxing@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/udp.c | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index c605d171eb2d..dc8feb54d835 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1531,10 +1531,21 @@ static void busylock_release(spinlock_t *busy)
 		spin_unlock(busy);
 }
 
+static int udp_rmem_schedule(struct sock *sk, int size)
+{
+	int delta;
+
+	delta = size - sk->sk_forward_alloc;
+	if (delta > 0 && !__sk_mem_schedule(sk, delta, SK_MEM_RECV))
+		return -ENOBUFS;
+
+	return 0;
+}
+
 int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 {
 	struct sk_buff_head *list = &sk->sk_receive_queue;
-	int rmem, delta, amt, err = -ENOMEM;
+	int rmem, err = -ENOMEM;
 	spinlock_t *busy = NULL;
 	int size;
 
@@ -1567,16 +1578,10 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 		goto uncharge_drop;
 
 	spin_lock(&list->lock);
-	if (size >= sk->sk_forward_alloc) {
-		amt = sk_mem_pages(size);
-		delta = amt << PAGE_SHIFT;
-		if (!__sk_mem_raise_allocated(sk, delta, amt, SK_MEM_RECV)) {
-			err = -ENOBUFS;
-			spin_unlock(&list->lock);
-			goto uncharge_drop;
-		}
-
-		sk->sk_forward_alloc += delta;
+	err = udp_rmem_schedule(sk, size);
+	if (err) {
+		spin_unlock(&list->lock);
+		goto uncharge_drop;
 	}
 
 	sk->sk_forward_alloc -= size;
-- 
cgit v1.2.3-70-g09d2


From a1331535aeb41b08fe0c2c78af51885edc93615b Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bhelgaas@google.com>
Date: Tue, 7 Mar 2023 16:10:51 -0600
Subject: net: restore alpha order to Ethernet devices in config

The filename "wangxun" sorts between "intel" and "xscale", but
xscale/Kconfig contains "Intel XScale" prompts, so Wangxun ends up in the
wrong place in the config front-ends.

Move wangxun/Kconfig so the Wangxun devices appear in order in the user
interface.

Fixes: 3ce7547e5b71 ("net: txgbe: Add build support for txgbe")
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20230307221051.890135-1-helgaas@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 1917da784191..5a274b99f299 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -84,7 +84,6 @@ source "drivers/net/ethernet/huawei/Kconfig"
 source "drivers/net/ethernet/i825xx/Kconfig"
 source "drivers/net/ethernet/ibm/Kconfig"
 source "drivers/net/ethernet/intel/Kconfig"
-source "drivers/net/ethernet/wangxun/Kconfig"
 source "drivers/net/ethernet/xscale/Kconfig"
 
 config JME
@@ -189,6 +188,7 @@ source "drivers/net/ethernet/toshiba/Kconfig"
 source "drivers/net/ethernet/tundra/Kconfig"
 source "drivers/net/ethernet/vertexcom/Kconfig"
 source "drivers/net/ethernet/via/Kconfig"
+source "drivers/net/ethernet/wangxun/Kconfig"
 source "drivers/net/ethernet/wiznet/Kconfig"
 source "drivers/net/ethernet/xilinx/Kconfig"
 source "drivers/net/ethernet/xircom/Kconfig"
-- 
cgit v1.2.3-70-g09d2


From 2498e6231bfd44f8f85afbc838b37441551a4028 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 21:40:12 -0800
Subject: selftests/bpf: prevent unused variable warning in bpf_for()

Add __attribute__((unused)) to inner __p variable inside bpf_for(),
bpf_for_each(), and bpf_repeat() macros to avoid compiler warnings about
unused variable.

Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230309054015.4068562-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_misc.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 43b154a639e7..c95eb603403c 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -115,7 +115,8 @@ extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
 	struct bpf_iter_##type ___it __attribute__((aligned(8), /* enforce, just in case */,	\
 						    cleanup(bpf_iter_##type##_destroy))),	\
 	/* ___p pointer is just to call bpf_iter_##type##_new() *once* to init ___it */		\
-			       *___p = (bpf_iter_##type##_new(&___it, ##args),			\
+			       *___p __attribute__((unused)) = (				\
+					bpf_iter_##type##_new(&___it, ##args),			\
 	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
 	/* for bpf_iter_##type##_destroy() when used from cleanup() attribute */		\
 					(void)bpf_iter_##type##_destroy, (void *)0);		\
@@ -143,7 +144,8 @@ extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
 	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
 						 cleanup(bpf_iter_num_destroy))),		\
 	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
-			    *___p = (bpf_iter_num_new(&___it, (start), (end)),			\
+			    *___p __attribute__((unused)) = (					\
+				bpf_iter_num_new(&___it, (start), (end)),			\
 	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
 	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
 				(void)bpf_iter_num_destroy, (void *)0);				\
@@ -167,7 +169,8 @@ extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
 	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
 						 cleanup(bpf_iter_num_destroy))),		\
 	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
-			    *___p = (bpf_iter_num_new(&___it, 0, (N)),				\
+			    *___p __attribute__((unused)) = (					\
+				bpf_iter_num_new(&___it, 0, (N)),				\
 	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
 	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
 				(void)bpf_iter_num_destroy, (void *)0);				\
-- 
cgit v1.2.3-70-g09d2


From 713461b895ef958ef444b00cc2d979f3ca3a82e2 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 21:40:13 -0800
Subject: selftests/bpf: add __sink() macro to fake variable consumption

Add __sink(expr) macro that forces compiler to believe that passed in
expression is both read and written. It used a simple embedded asm for
this. This is useful in a lot of tests where we assign value to some variable
to trigger some action, but later don't read variable, causing compiler
to complain (if corresponding compiler warnings are turned on, which
we'll do in the next patch).

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230309054015.4068562-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_misc.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index c95eb603403c..3c03ec8056ce 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -76,6 +76,9 @@
 #define FUNC_REG_ARG_CNT 5
 #endif
 
+/* make it look to compiler like value is read and written */
+#define __sink(expr) asm volatile("" : "+g"(expr))
+
 struct bpf_iter_num;
 
 extern int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end) __ksym;
-- 
cgit v1.2.3-70-g09d2


From c8ed66859397237c649998c58a68a86b8ea5f417 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 21:40:14 -0800
Subject: selftests/bpf: fix lots of silly mistakes pointed out by compiler

Once we enable -Wall for BPF sources, compiler will complain about lots
of unused variables, variables that are set but never read, etc.

Fix all these issues first before enabling -Wall in Makefile.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230309054015.4068562-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_iter_ksym.c  |  1 -
 .../selftests/bpf/progs/bpf_iter_setsockopt.c      |  1 -
 tools/testing/selftests/bpf/progs/bpf_loop.c       |  2 -
 tools/testing/selftests/bpf/progs/cb_refs.c        |  1 -
 .../bpf/progs/cgroup_skb_sk_lookup_kern.c          |  1 -
 .../selftests/bpf/progs/cgrp_kfunc_failure.c       |  1 +
 .../selftests/bpf/progs/cgrp_ls_attach_cgroup.c    |  1 -
 .../selftests/bpf/progs/cgrp_ls_sleepable.c        |  1 -
 tools/testing/selftests/bpf/progs/core_kern.c      |  2 +-
 .../testing/selftests/bpf/progs/cpumask_failure.c  |  3 ++
 .../testing/selftests/bpf/progs/cpumask_success.c  |  1 -
 tools/testing/selftests/bpf/progs/dynptr_fail.c    |  5 ++-
 tools/testing/selftests/bpf/progs/dynptr_success.c |  5 +--
 tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c  |  2 -
 .../selftests/bpf/progs/freplace_attach_probe.c    |  2 +-
 tools/testing/selftests/bpf/progs/iters.c          | 11 +++---
 tools/testing/selftests/bpf/progs/linked_funcs1.c  |  3 ++
 tools/testing/selftests/bpf/progs/linked_funcs2.c  |  3 ++
 tools/testing/selftests/bpf/progs/linked_list.c    |  4 --
 .../testing/selftests/bpf/progs/linked_list_fail.c |  1 -
 tools/testing/selftests/bpf/progs/local_storage.c  |  1 -
 tools/testing/selftests/bpf/progs/map_kptr.c       |  3 --
 tools/testing/selftests/bpf/progs/netcnt_prog.c    |  1 -
 .../selftests/bpf/progs/netif_receive_skb.c        |  1 -
 tools/testing/selftests/bpf/progs/perfbuf_bench.c  |  1 -
 tools/testing/selftests/bpf/progs/pyperf.h         |  2 +-
 .../bpf/progs/rbtree_btf_fail__wrong_node_type.c   | 11 ------
 tools/testing/selftests/bpf/progs/rbtree_fail.c    |  3 +-
 tools/testing/selftests/bpf/progs/rcu_read_lock.c  |  4 --
 .../bpf/progs/read_bpf_task_storage_busy.c         |  1 -
 tools/testing/selftests/bpf/progs/recvmsg4_prog.c  |  2 -
 tools/testing/selftests/bpf/progs/recvmsg6_prog.c  |  2 -
 tools/testing/selftests/bpf/progs/sendmsg4_prog.c  |  2 -
 .../selftests/bpf/progs/sockmap_verdict_prog.c     |  4 ++
 tools/testing/selftests/bpf/progs/strobemeta.h     |  1 -
 .../selftests/bpf/progs/tailcall_bpf2bpf3.c        | 11 ++++++
 .../selftests/bpf/progs/tailcall_bpf2bpf6.c        |  3 ++
 .../selftests/bpf/progs/task_kfunc_failure.c       |  1 +
 .../selftests/bpf/progs/task_kfunc_success.c       |  6 ---
 tools/testing/selftests/bpf/progs/test_bpf_nf.c    |  1 -
 .../selftests/bpf/progs/test_cls_redirect_dynptr.c |  1 -
 .../bpf/progs/test_core_reloc_bitfields_probed.c   |  1 -
 .../selftests/bpf/progs/test_global_func1.c        |  4 ++
 .../selftests/bpf/progs/test_global_func2.c        |  4 ++
 .../selftests/bpf/progs/test_hash_large_key.c      |  2 +-
 .../bpf/progs/test_ksyms_btf_write_check.c         |  1 -
 .../selftests/bpf/progs/test_legacy_printk.c       |  2 +-
 tools/testing/selftests/bpf/progs/test_map_lock.c  |  2 +-
 tools/testing/selftests/bpf/progs/test_obj_id.c    |  2 +
 .../selftests/bpf/progs/test_parse_tcp_hdr_opt.c   |  1 -
 .../bpf/progs/test_parse_tcp_hdr_opt_dynptr.c      |  2 +-
 .../testing/selftests/bpf/progs/test_pkt_access.c  |  5 +++
 tools/testing/selftests/bpf/progs/test_ringbuf.c   |  1 -
 .../selftests/bpf/progs/test_ringbuf_map_key.c     |  1 +
 .../selftests/bpf/progs/test_ringbuf_multi.c       |  1 -
 .../bpf/progs/test_select_reuseport_kern.c         |  2 +-
 tools/testing/selftests/bpf/progs/test_sk_assign.c |  4 +-
 tools/testing/selftests/bpf/progs/test_sk_lookup.c |  9 +----
 .../selftests/bpf/progs/test_sk_lookup_kern.c      |  2 -
 .../testing/selftests/bpf/progs/test_sock_fields.c |  2 +-
 .../selftests/bpf/progs/test_sockmap_kern.h        | 14 +++++--
 tools/testing/selftests/bpf/progs/test_spin_lock.c |  3 ++
 tools/testing/selftests/bpf/progs/test_tc_dtime.c  |  4 +-
 tools/testing/selftests/bpf/progs/test_tc_neigh.c  |  4 +-
 .../testing/selftests/bpf/progs/test_tcpbpf_kern.c |  2 -
 .../testing/selftests/bpf/progs/test_tunnel_kern.c |  6 ---
 .../selftests/bpf/progs/test_usdt_multispec.c      |  2 -
 .../selftests/bpf/progs/test_verif_scale1.c        |  2 +-
 .../selftests/bpf/progs/test_verif_scale2.c        |  2 +-
 .../selftests/bpf/progs/test_verif_scale3.c        |  2 +-
 .../bpf/progs/test_xdp_adjust_tail_grow.c          |  2 -
 .../testing/selftests/bpf/progs/test_xdp_bpf2bpf.c |  2 -
 .../testing/selftests/bpf/progs/test_xdp_dynptr.c  |  2 -
 .../selftests/bpf/progs/test_xdp_noinline.c        | 43 ----------------------
 tools/testing/selftests/bpf/progs/test_xdp_vlan.c  | 13 -------
 tools/testing/selftests/bpf/progs/type_cast.c      |  1 -
 tools/testing/selftests/bpf/progs/udp_limit.c      |  2 -
 .../selftests/bpf/progs/user_ringbuf_success.c     |  6 ---
 tools/testing/selftests/bpf/progs/xdp_features.c   |  1 -
 tools/testing/selftests/bpf/progs/xdping_kern.c    |  2 -
 tools/testing/selftests/bpf/progs/xdpwall.c        |  1 -
 81 files changed, 90 insertions(+), 187 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_ksym.c b/tools/testing/selftests/bpf/progs/bpf_iter_ksym.c
index 9ba14c37bbcc..5ddcc46fd886 100644
--- a/tools/testing/selftests/bpf/progs/bpf_iter_ksym.c
+++ b/tools/testing/selftests/bpf/progs/bpf_iter_ksym.c
@@ -33,7 +33,6 @@ int dump_ksym(struct bpf_iter__ksym *ctx)
 	__u32 seq_num = ctx->meta->seq_num;
 	unsigned long value;
 	char type;
-	int ret;
 
 	if (!iter)
 		return 0;
diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_setsockopt.c b/tools/testing/selftests/bpf/progs/bpf_iter_setsockopt.c
index b77adfd55d73..ec7f91850dec 100644
--- a/tools/testing/selftests/bpf/progs/bpf_iter_setsockopt.c
+++ b/tools/testing/selftests/bpf/progs/bpf_iter_setsockopt.c
@@ -42,7 +42,6 @@ int change_tcp_cc(struct bpf_iter__tcp *ctx)
 	char cur_cc[TCP_CA_NAME_MAX];
 	struct tcp_sock *tp;
 	struct sock *sk;
-	int ret;
 
 	if (!bpf_tcp_sk(ctx->sk_common))
 		return 0;
diff --git a/tools/testing/selftests/bpf/progs/bpf_loop.c b/tools/testing/selftests/bpf/progs/bpf_loop.c
index de1fc82d2710..1d194455b109 100644
--- a/tools/testing/selftests/bpf/progs/bpf_loop.c
+++ b/tools/testing/selftests/bpf/progs/bpf_loop.c
@@ -138,8 +138,6 @@ static int callback_set_0f(int i, void *ctx)
 SEC("fentry/" SYS_PREFIX "sys_nanosleep")
 int prog_non_constant_callback(void *ctx)
 {
-	struct callback_ctx data = {};
-
 	if (bpf_get_current_pid_tgid() >> 32 != pid)
 		return 0;
 
diff --git a/tools/testing/selftests/bpf/progs/cb_refs.c b/tools/testing/selftests/bpf/progs/cb_refs.c
index ce96b33e38d6..50f95ec61165 100644
--- a/tools/testing/selftests/bpf/progs/cb_refs.c
+++ b/tools/testing/selftests/bpf/progs/cb_refs.c
@@ -52,7 +52,6 @@ int leak_prog(void *ctx)
 {
 	struct prog_test_ref_kfunc *p;
 	struct map_value *v;
-	unsigned long sl;
 
 	v = bpf_map_lookup_elem(&array_map, &(int){0});
 	if (!v)
diff --git a/tools/testing/selftests/bpf/progs/cgroup_skb_sk_lookup_kern.c b/tools/testing/selftests/bpf/progs/cgroup_skb_sk_lookup_kern.c
index 88638315c582..ac86a8a61605 100644
--- a/tools/testing/selftests/bpf/progs/cgroup_skb_sk_lookup_kern.c
+++ b/tools/testing/selftests/bpf/progs/cgroup_skb_sk_lookup_kern.c
@@ -66,7 +66,6 @@ static inline int is_allowed_peer_cg(struct __sk_buff *skb,
 SEC("cgroup_skb/ingress")
 int ingress_lookup(struct __sk_buff *skb)
 {
-	__u32 serv_port_key = 0;
 	struct ipv6hdr ip6h;
 	struct tcphdr tcph;
 
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
index b42291ed9586..807fb0ac41e9 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
@@ -109,6 +109,7 @@ int BPF_PROG(cgrp_kfunc_acquire_unreleased, struct cgroup *cgrp, const char *pat
 	acquired = bpf_cgroup_acquire(cgrp);
 
 	/* Acquired cgroup is never released. */
+	__sink(acquired);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/cgrp_ls_attach_cgroup.c b/tools/testing/selftests/bpf/progs/cgrp_ls_attach_cgroup.c
index 6652d18465b2..8aeba1b75c83 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_ls_attach_cgroup.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_ls_attach_cgroup.c
@@ -84,7 +84,6 @@ int BPF_PROG(update_cookie_tracing, struct socket *sock,
 	     struct sockaddr *uaddr, int addr_len, int flags)
 {
 	struct socket_cookie *p;
-	struct tcp_sock *tcp_sk;
 
 	if (uaddr->sa_family != AF_INET6)
 		return 0;
diff --git a/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c b/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c
index 7615dc23d301..4c7844e1dbfa 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_ls_sleepable.c
@@ -24,7 +24,6 @@ void bpf_rcu_read_unlock(void) __ksym;
 SEC("?iter.s/cgroup")
 int cgroup_iter(struct bpf_iter__cgroup *ctx)
 {
-	struct seq_file *seq = ctx->meta->seq;
 	struct cgroup *cgrp = ctx->cgroup;
 	long *ptr;
 
diff --git a/tools/testing/selftests/bpf/progs/core_kern.c b/tools/testing/selftests/bpf/progs/core_kern.c
index 2715fe27d4cf..004f2acef2eb 100644
--- a/tools/testing/selftests/bpf/progs/core_kern.c
+++ b/tools/testing/selftests/bpf/progs/core_kern.c
@@ -77,7 +77,7 @@ int balancer_ingress(struct __sk_buff *ctx)
 	void *data_end = (void *)(long)ctx->data_end;
 	void *data = (void *)(long)ctx->data;
 	void *ptr;
-	int ret = 0, nh_off, i = 0;
+	int nh_off, i = 0;
 
 	nh_off = 14;
 
diff --git a/tools/testing/selftests/bpf/progs/cpumask_failure.c b/tools/testing/selftests/bpf/progs/cpumask_failure.c
index c16f7563b84e..cfe83f0ef9e2 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_failure.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_failure.c
@@ -23,6 +23,7 @@ int BPF_PROG(test_alloc_no_release, struct task_struct *task, u64 clone_flags)
 	struct bpf_cpumask *cpumask;
 
 	cpumask = create_cpumask();
+	__sink(cpumask);
 
 	/* cpumask is never released. */
 	return 0;
@@ -51,6 +52,7 @@ int BPF_PROG(test_acquire_wrong_cpumask, struct task_struct *task, u64 clone_fla
 
 	/* Can't acquire a non-struct bpf_cpumask. */
 	cpumask = bpf_cpumask_acquire((struct bpf_cpumask *)task->cpus_ptr);
+	__sink(cpumask);
 
 	return 0;
 }
@@ -63,6 +65,7 @@ int BPF_PROG(test_mutate_cpumask, struct task_struct *task, u64 clone_flags)
 
 	/* Can't set the CPU of a non-struct bpf_cpumask. */
 	bpf_cpumask_set_cpu(0, (struct bpf_cpumask *)task->cpus_ptr);
+	__sink(cpumask);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/cpumask_success.c b/tools/testing/selftests/bpf/progs/cpumask_success.c
index 1d38bc65d4b0..97ed08c4ff03 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_success.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_success.c
@@ -353,7 +353,6 @@ SEC("tp_btf/task_newtask")
 int BPF_PROG(test_insert_leave, struct task_struct *task, u64 clone_flags)
 {
 	struct bpf_cpumask *cpumask;
-	struct __cpumask_map_value *v;
 
 	cpumask = create_cpumask();
 	if (!cpumask)
diff --git a/tools/testing/selftests/bpf/progs/dynptr_fail.c b/tools/testing/selftests/bpf/progs/dynptr_fail.c
index 20ce920d891d..759eb5c245cd 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_fail.c
@@ -271,7 +271,7 @@ SEC("?raw_tp")
 __failure __msg("value is outside of the allowed memory range")
 int data_slice_out_of_bounds_map_value(void *ctx)
 {
-	__u32 key = 0, map_val;
+	__u32 map_val;
 	struct bpf_dynptr ptr;
 	void *data;
 
@@ -388,7 +388,6 @@ int data_slice_missing_null_check2(void *ctx)
 		/* this should fail */
 		*data2 = 3;
 
-done:
 	bpf_ringbuf_discard_dynptr(&ptr, 0);
 	return 0;
 }
@@ -440,6 +439,7 @@ int invalid_write1(void *ctx)
 
 	/* this should fail */
 	data = bpf_dynptr_data(&ptr, 0, 1);
+	__sink(data);
 
 	return 0;
 }
@@ -1374,6 +1374,7 @@ int invalid_slice_rdwr_rdonly(struct __sk_buff *skb)
 	 * changing packet data
 	 */
 	hdr = bpf_dynptr_slice_rdwr(&ptr, 0, buffer, sizeof(buffer));
+	__sink(hdr);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/dynptr_success.c b/tools/testing/selftests/bpf/progs/dynptr_success.c
index c8358a7c7924..b2fa6c47ecc0 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_success.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_success.c
@@ -35,7 +35,7 @@ SEC("?tp/syscalls/sys_enter_nanosleep")
 int test_read_write(void *ctx)
 {
 	char write_data[64] = "hello there, world!!";
-	char read_data[64] = {}, buf[64] = {};
+	char read_data[64] = {};
 	struct bpf_dynptr ptr;
 	int i;
 
@@ -170,7 +170,6 @@ int test_skb_readonly(struct __sk_buff *skb)
 {
 	__u8 write_data[2] = {1, 2};
 	struct bpf_dynptr ptr;
-	__u64 *data;
 	int ret;
 
 	if (bpf_dynptr_from_skb(skb, 0, &ptr)) {
@@ -191,10 +190,8 @@ int test_skb_readonly(struct __sk_buff *skb)
 SEC("?cgroup_skb/egress")
 int test_dynptr_skb_data(struct __sk_buff *skb)
 {
-	__u8 write_data[2] = {1, 2};
 	struct bpf_dynptr ptr;
 	__u64 *data;
-	int ret;
 
 	if (bpf_dynptr_from_skb(skb, 0, &ptr)) {
 		err = 1;
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
index 4547b059d487..983b7c233382 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
@@ -120,8 +120,6 @@ int new_get_skb_ifindex(int val, struct __sk_buff *skb, int var)
 	void *data = (void *)(long)skb->data;
 	struct ipv6hdr ip6, *ip6p;
 	int ifindex = skb->ifindex;
-	__u32 eth_proto;
-	__u32 nh_off;
 
 	/* check that BPF extension can read packet via direct packet access */
 	if (data + 14 + sizeof(ip6) > data_end)
diff --git a/tools/testing/selftests/bpf/progs/freplace_attach_probe.c b/tools/testing/selftests/bpf/progs/freplace_attach_probe.c
index bb2a77c5b62b..370a0e1922e0 100644
--- a/tools/testing/selftests/bpf/progs/freplace_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/freplace_attach_probe.c
@@ -23,7 +23,7 @@ struct {
 SEC("freplace/handle_kprobe")
 int new_handle_kprobe(struct pt_regs *ctx)
 {
-	struct hmap_elem zero = {}, *val;
+	struct hmap_elem *val;
 	int key = 0;
 
 	val = bpf_map_lookup_elem(&hash_map, &key);
diff --git a/tools/testing/selftests/bpf/progs/iters.c b/tools/testing/selftests/bpf/progs/iters.c
index 84e5dc10243c..6b9b3c56f009 100644
--- a/tools/testing/selftests/bpf/progs/iters.c
+++ b/tools/testing/selftests/bpf/progs/iters.c
@@ -45,7 +45,6 @@ __failure __msg("unbounded memory access")
 int iter_err_unsafe_asm_loop(const void *ctx)
 {
 	struct bpf_iter_num it;
-	int *v, i = 0;
 
 	MY_PID_GUARD();
 
@@ -88,7 +87,7 @@ __success
 int iter_while_loop(const void *ctx)
 {
 	struct bpf_iter_num it;
-	int *v, i;
+	int *v;
 
 	MY_PID_GUARD();
 
@@ -106,7 +105,7 @@ __success
 int iter_while_loop_auto_cleanup(const void *ctx)
 {
 	__attribute__((cleanup(bpf_iter_num_destroy))) struct bpf_iter_num it;
-	int *v, i;
+	int *v;
 
 	MY_PID_GUARD();
 
@@ -124,7 +123,7 @@ __success
 int iter_for_loop(const void *ctx)
 {
 	struct bpf_iter_num it;
-	int *v, i;
+	int *v;
 
 	MY_PID_GUARD();
 
@@ -192,7 +191,7 @@ __success
 int iter_manual_unroll_loop(const void *ctx)
 {
 	struct bpf_iter_num it;
-	int *v, i;
+	int *v;
 
 	MY_PID_GUARD();
 
@@ -621,7 +620,7 @@ __success
 int iter_stack_array_loop(const void *ctx)
 {
 	long arr1[16], arr2[16], sum = 0;
-	int *v, i;
+	int i;
 
 	MY_PID_GUARD();
 
diff --git a/tools/testing/selftests/bpf/progs/linked_funcs1.c b/tools/testing/selftests/bpf/progs/linked_funcs1.c
index b05571bc67d5..c4b49ceea967 100644
--- a/tools/testing/selftests/bpf/progs/linked_funcs1.c
+++ b/tools/testing/selftests/bpf/progs/linked_funcs1.c
@@ -5,6 +5,7 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_core_read.h>
+#include "bpf_misc.h"
 
 /* weak and shared between two files */
 const volatile int my_tid __weak;
@@ -51,6 +52,7 @@ __weak int set_output_weak(int x)
 	 * cause problems for BPF static linker
 	 */
 	whatever = bpf_core_type_size(struct task_struct);
+	__sink(whatever);
 
 	output_weak1 = x;
 	return x;
@@ -71,6 +73,7 @@ int BPF_PROG(handler1, struct pt_regs *regs, long id)
 
 	/* make sure we have CO-RE relocations in main program */
 	whatever = bpf_core_type_size(struct task_struct);
+	__sink(whatever);
 
 	set_output_val2(1000);
 	set_output_ctx2(ctx); /* ctx definition is hidden in BPF_PROG macro */
diff --git a/tools/testing/selftests/bpf/progs/linked_funcs2.c b/tools/testing/selftests/bpf/progs/linked_funcs2.c
index ee7e3848ee4f..013ff0645f0c 100644
--- a/tools/testing/selftests/bpf/progs/linked_funcs2.c
+++ b/tools/testing/selftests/bpf/progs/linked_funcs2.c
@@ -5,6 +5,7 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_core_read.h>
+#include "bpf_misc.h"
 
 /* weak and shared between both files */
 const volatile int my_tid __weak;
@@ -51,6 +52,7 @@ __weak int set_output_weak(int x)
 	 * cause problems for BPF static linker
 	 */
 	whatever = 2 * bpf_core_type_size(struct task_struct);
+	__sink(whatever);
 
 	output_weak2 = x;
 	return 2 * x;
@@ -71,6 +73,7 @@ int BPF_PROG(handler2, struct pt_regs *regs, long id)
 
 	/* make sure we have CO-RE relocations in main program */
 	whatever = bpf_core_type_size(struct task_struct);
+	__sink(whatever);
 
 	set_output_val1(2000);
 	set_output_ctx1(ctx); /* ctx definition is hidden in BPF_PROG macro */
diff --git a/tools/testing/selftests/bpf/progs/linked_list.c b/tools/testing/selftests/bpf/progs/linked_list.c
index 4fa4a9b01bde..53ded51a3abb 100644
--- a/tools/testing/selftests/bpf/progs/linked_list.c
+++ b/tools/testing/selftests/bpf/progs/linked_list.c
@@ -313,7 +313,6 @@ SEC("tc")
 int map_list_push_pop_multiple(void *ctx)
 {
 	struct map_value *v;
-	int ret;
 
 	v = bpf_map_lookup_elem(&array_map, &(int){0});
 	if (!v)
@@ -326,7 +325,6 @@ int inner_map_list_push_pop_multiple(void *ctx)
 {
 	struct map_value *v;
 	void *map;
-	int ret;
 
 	map = bpf_map_lookup_elem(&map_of_maps, &(int){0});
 	if (!map)
@@ -352,7 +350,6 @@ SEC("tc")
 int map_list_in_list(void *ctx)
 {
 	struct map_value *v;
-	int ret;
 
 	v = bpf_map_lookup_elem(&array_map, &(int){0});
 	if (!v)
@@ -365,7 +362,6 @@ int inner_map_list_in_list(void *ctx)
 {
 	struct map_value *v;
 	void *map;
-	int ret;
 
 	map = bpf_map_lookup_elem(&map_of_maps, &(int){0});
 	if (!map)
diff --git a/tools/testing/selftests/bpf/progs/linked_list_fail.c b/tools/testing/selftests/bpf/progs/linked_list_fail.c
index 69cdc07cba13..41978b46f58e 100644
--- a/tools/testing/selftests/bpf/progs/linked_list_fail.c
+++ b/tools/testing/selftests/bpf/progs/linked_list_fail.c
@@ -557,7 +557,6 @@ SEC("?tc")
 int incorrect_head_off2(void *ctx)
 {
 	struct foo *f;
-	struct bar *b;
 
 	f = bpf_obj_new(typeof(*f));
 	if (!f)
diff --git a/tools/testing/selftests/bpf/progs/local_storage.c b/tools/testing/selftests/bpf/progs/local_storage.c
index 19423ed862e3..01c74bc870ae 100644
--- a/tools/testing/selftests/bpf/progs/local_storage.c
+++ b/tools/testing/selftests/bpf/progs/local_storage.c
@@ -77,7 +77,6 @@ int BPF_PROG(inode_rename, struct inode *old_dir, struct dentry *old_dentry,
 	     struct inode *new_dir, struct dentry *new_dentry,
 	     unsigned int flags)
 {
-	__u32 pid = bpf_get_current_pid_tgid() >> 32;
 	struct local_storage *storage;
 	int err;
 
diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c
index 3903d30217b8..dae5dab1bbf7 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr.c
@@ -515,7 +515,6 @@ int test_ls_map_kptr_ref1(void *ctx)
 {
 	struct task_struct *current;
 	struct map_value *v;
-	int ret;
 
 	current = bpf_get_current_task_btf();
 	if (!current)
@@ -534,7 +533,6 @@ int test_ls_map_kptr_ref2(void *ctx)
 {
 	struct task_struct *current;
 	struct map_value *v;
-	int ret;
 
 	current = bpf_get_current_task_btf();
 	if (!current)
@@ -550,7 +548,6 @@ int test_ls_map_kptr_ref_del(void *ctx)
 {
 	struct task_struct *current;
 	struct map_value *v;
-	int ret;
 
 	current = bpf_get_current_task_btf();
 	if (!current)
diff --git a/tools/testing/selftests/bpf/progs/netcnt_prog.c b/tools/testing/selftests/bpf/progs/netcnt_prog.c
index f718b2c212dc..f9ef8aee56f1 100644
--- a/tools/testing/selftests/bpf/progs/netcnt_prog.c
+++ b/tools/testing/selftests/bpf/progs/netcnt_prog.c
@@ -26,7 +26,6 @@ SEC("cgroup/skb")
 int bpf_nextcnt(struct __sk_buff *skb)
 {
 	union percpu_net_cnt *percpu_cnt;
-	char fmt[] = "%d %llu %llu\n";
 	union net_cnt *cnt;
 	__u64 ts, dt;
 	int ret;
diff --git a/tools/testing/selftests/bpf/progs/netif_receive_skb.c b/tools/testing/selftests/bpf/progs/netif_receive_skb.c
index 1d8918dfbd3f..c0062645fc68 100644
--- a/tools/testing/selftests/bpf/progs/netif_receive_skb.c
+++ b/tools/testing/selftests/bpf/progs/netif_receive_skb.c
@@ -53,7 +53,6 @@ static int __strncmp(const void *m1, const void *m2, size_t len)
 	do {								\
 		static const char _expectedval[EXPECTED_STRSIZE] =	\
 							_expected;	\
-		static const char _ptrtype[64] = #_type;		\
 		__u64 _hflags = _flags | BTF_F_COMPACT;			\
 		static _type _ptrdata = __VA_ARGS__;			\
 		static struct btf_ptr _ptr = { };			\
diff --git a/tools/testing/selftests/bpf/progs/perfbuf_bench.c b/tools/testing/selftests/bpf/progs/perfbuf_bench.c
index 45204fe0c570..29c1639fc78a 100644
--- a/tools/testing/selftests/bpf/progs/perfbuf_bench.c
+++ b/tools/testing/selftests/bpf/progs/perfbuf_bench.c
@@ -22,7 +22,6 @@ long dropped __attribute__((aligned(128))) = 0;
 SEC("fentry/" SYS_PREFIX "sys_getpgid")
 int bench_perfbuf(void *ctx)
 {
-	__u64 *sample;
 	int i;
 
 	for (i = 0; i < batch_cnt; i++) {
diff --git a/tools/testing/selftests/bpf/progs/pyperf.h b/tools/testing/selftests/bpf/progs/pyperf.h
index f2e7a31c8d75..026d573ce179 100644
--- a/tools/testing/selftests/bpf/progs/pyperf.h
+++ b/tools/testing/selftests/bpf/progs/pyperf.h
@@ -345,7 +345,7 @@ int __on_event(struct bpf_raw_tracepoint_args *ctx)
 SEC("raw_tracepoint/kfree_skb")
 int on_event(struct bpf_raw_tracepoint_args* ctx)
 {
-	int i, ret = 0;
+	int ret = 0;
 	ret |= __on_event(ctx);
 	ret |= __on_event(ctx);
 	ret |= __on_event(ctx);
diff --git a/tools/testing/selftests/bpf/progs/rbtree_btf_fail__wrong_node_type.c b/tools/testing/selftests/bpf/progs/rbtree_btf_fail__wrong_node_type.c
index 340f97da1084..7651843f5a80 100644
--- a/tools/testing/selftests/bpf/progs/rbtree_btf_fail__wrong_node_type.c
+++ b/tools/testing/selftests/bpf/progs/rbtree_btf_fail__wrong_node_type.c
@@ -16,17 +16,6 @@ struct node_data {
 	struct bpf_list_node node;
 };
 
-static bool less(struct bpf_rb_node *a, const struct bpf_rb_node *b)
-{
-	struct node_data *node_a;
-	struct node_data *node_b;
-
-	node_a = container_of(a, struct node_data, node);
-	node_b = container_of(b, struct node_data, node);
-
-	return node_a->key < node_b->key;
-}
-
 #define private(name) SEC(".data." #name) __hidden __attribute__((aligned(8)))
 private(A) struct bpf_spin_lock glock;
 private(A) struct bpf_rb_root groot __contains(node_data, node);
diff --git a/tools/testing/selftests/bpf/progs/rbtree_fail.c b/tools/testing/selftests/bpf/progs/rbtree_fail.c
index 1ced900f3fce..46d7d18a218f 100644
--- a/tools/testing/selftests/bpf/progs/rbtree_fail.c
+++ b/tools/testing/selftests/bpf/progs/rbtree_fail.c
@@ -105,7 +105,7 @@ long rbtree_api_remove_unadded_node(void *ctx)
 }
 
 SEC("?tc")
-__failure __msg("Unreleased reference id=2 alloc_insn=11")
+__failure __msg("Unreleased reference id=2 alloc_insn=10")
 long rbtree_api_remove_no_drop(void *ctx)
 {
 	struct bpf_rb_node *res;
@@ -119,6 +119,7 @@ long rbtree_api_remove_no_drop(void *ctx)
 	res = bpf_rbtree_remove(&groot, res);
 
 	n = container_of(res, struct node_data, node);
+	__sink(n);
 	bpf_spin_unlock(&glock);
 
 	/* bpf_obj_drop(n) is missing here */
diff --git a/tools/testing/selftests/bpf/progs/rcu_read_lock.c b/tools/testing/selftests/bpf/progs/rcu_read_lock.c
index 7250bb76d18a..6a8c88e58df2 100644
--- a/tools/testing/selftests/bpf/progs/rcu_read_lock.c
+++ b/tools/testing/selftests/bpf/progs/rcu_read_lock.c
@@ -179,8 +179,6 @@ SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
 int miss_lock(void *ctx)
 {
 	struct task_struct *task;
-	struct css_set *cgroups;
-	struct cgroup *dfl_cgrp;
 
 	/* missing bpf_rcu_read_lock() */
 	task = bpf_get_current_task_btf();
@@ -195,8 +193,6 @@ SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
 int miss_unlock(void *ctx)
 {
 	struct task_struct *task;
-	struct css_set *cgroups;
-	struct cgroup *dfl_cgrp;
 
 	/* missing bpf_rcu_read_unlock() */
 	task = bpf_get_current_task_btf();
diff --git a/tools/testing/selftests/bpf/progs/read_bpf_task_storage_busy.c b/tools/testing/selftests/bpf/progs/read_bpf_task_storage_busy.c
index a47bb0120719..76556e0b42b2 100644
--- a/tools/testing/selftests/bpf/progs/read_bpf_task_storage_busy.c
+++ b/tools/testing/selftests/bpf/progs/read_bpf_task_storage_busy.c
@@ -23,7 +23,6 @@ SEC("raw_tp/sys_enter")
 int BPF_PROG(read_bpf_task_storage_busy)
 {
 	int *value;
-	int key;
 
 	if (!CONFIG_PREEMPT)
 		return 0;
diff --git a/tools/testing/selftests/bpf/progs/recvmsg4_prog.c b/tools/testing/selftests/bpf/progs/recvmsg4_prog.c
index 3d1ae8b3402f..59748c95471a 100644
--- a/tools/testing/selftests/bpf/progs/recvmsg4_prog.c
+++ b/tools/testing/selftests/bpf/progs/recvmsg4_prog.c
@@ -17,8 +17,6 @@ SEC("cgroup/recvmsg4")
 int recvmsg4_prog(struct bpf_sock_addr *ctx)
 {
 	struct bpf_sock *sk;
-	__u32 user_ip4;
-	__u16 user_port;
 
 	sk = ctx->sk;
 	if (!sk)
diff --git a/tools/testing/selftests/bpf/progs/recvmsg6_prog.c b/tools/testing/selftests/bpf/progs/recvmsg6_prog.c
index 27dfb21b21b4..d9a4016596d5 100644
--- a/tools/testing/selftests/bpf/progs/recvmsg6_prog.c
+++ b/tools/testing/selftests/bpf/progs/recvmsg6_prog.c
@@ -20,8 +20,6 @@ SEC("cgroup/recvmsg6")
 int recvmsg6_prog(struct bpf_sock_addr *ctx)
 {
 	struct bpf_sock *sk;
-	__u32 user_ip4;
-	__u16 user_port;
 
 	sk = ctx->sk;
 	if (!sk)
diff --git a/tools/testing/selftests/bpf/progs/sendmsg4_prog.c b/tools/testing/selftests/bpf/progs/sendmsg4_prog.c
index ea75a44cb7fc..351e79aef2fa 100644
--- a/tools/testing/selftests/bpf/progs/sendmsg4_prog.c
+++ b/tools/testing/selftests/bpf/progs/sendmsg4_prog.c
@@ -21,8 +21,6 @@
 SEC("cgroup/sendmsg4")
 int sendmsg_v4_prog(struct bpf_sock_addr *ctx)
 {
-	int prio;
-
 	if (ctx->type != SOCK_DGRAM)
 		return 0;
 
diff --git a/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c b/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c
index e2468a6d01a5..0660f29dca95 100644
--- a/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c
+++ b/tools/testing/selftests/bpf/progs/sockmap_verdict_prog.c
@@ -1,6 +1,7 @@
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_endian.h>
+#include "bpf_misc.h"
 
 struct {
 	__uint(type, BPF_MAP_TYPE_SOCKMAP);
@@ -40,6 +41,9 @@ int bpf_prog2(struct __sk_buff *skb)
 	__u8 *d = data;
 	__u8 sk, map;
 
+	__sink(lport);
+	__sink(rport);
+
 	if (data + 8 > data_end)
 		return SK_DROP;
 
diff --git a/tools/testing/selftests/bpf/progs/strobemeta.h b/tools/testing/selftests/bpf/progs/strobemeta.h
index e562be6356f3..e02cfd380746 100644
--- a/tools/testing/selftests/bpf/progs/strobemeta.h
+++ b/tools/testing/selftests/bpf/progs/strobemeta.h
@@ -391,7 +391,6 @@ static __always_inline void *read_map_var(struct strobemeta_cfg *cfg,
 	struct strobe_map_raw map;
 	void *location;
 	uint64_t len;
-	int i;
 
 	descr->tag_len = 0; /* presume no tag is set */
 	descr->cnt = -1; /* presume no value is set */
diff --git a/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf3.c b/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf3.c
index 7fab39a3bb12..99c8d1d8a187 100644
--- a/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf3.c
+++ b/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf3.c
@@ -2,6 +2,7 @@
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
 #include "bpf_legacy.h"
+#include "bpf_misc.h"
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
@@ -20,6 +21,8 @@ int subprog_tail2(struct __sk_buff *skb)
 	else
 		bpf_tail_call_static(skb, &jmp_table, 1);
 
+	__sink(arr[sizeof(arr) - 1]);
+
 	return skb->len;
 }
 
@@ -30,6 +33,8 @@ int subprog_tail(struct __sk_buff *skb)
 
 	bpf_tail_call_static(skb, &jmp_table, 0);
 
+	__sink(arr[sizeof(arr) - 1]);
+
 	return skb->len * 2;
 }
 
@@ -38,6 +43,8 @@ int classifier_0(struct __sk_buff *skb)
 {
 	volatile char arr[128] = {};
 
+	__sink(arr[sizeof(arr) - 1]);
+
 	return subprog_tail2(skb);
 }
 
@@ -46,6 +53,8 @@ int classifier_1(struct __sk_buff *skb)
 {
 	volatile char arr[128] = {};
 
+	__sink(arr[sizeof(arr) - 1]);
+
 	return skb->len * 3;
 }
 
@@ -54,6 +63,8 @@ int entry(struct __sk_buff *skb)
 {
 	volatile char arr[128] = {};
 
+	__sink(arr[sizeof(arr) - 1]);
+
 	return subprog_tail(skb);
 }
 
diff --git a/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c b/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
index 41ce83da78e8..4a9f63bea66c 100644
--- a/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
+++ b/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
 
 #define __unused __attribute__((unused))
 
@@ -36,6 +37,8 @@ int entry(struct __sk_buff *skb)
 	/* Have data on stack which size is not a multiple of 8 */
 	volatile char arr[1] = {};
 
+	__sink(arr[0]);
+
 	return subprog_tail(skb);
 }
 
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
index f19d54eda4f1..002c7f69e47f 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
@@ -109,6 +109,7 @@ int BPF_PROG(task_kfunc_acquire_unreleased, struct task_struct *task, u64 clone_
 	acquired = bpf_task_acquire(task);
 
 	/* Acquired task is never released. */
+	__sink(acquired);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index 9f359cfd29e7..aebc4bb14e7d 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -171,8 +171,6 @@ static void lookup_compare_pid(const struct task_struct *p)
 SEC("tp_btf/task_newtask")
 int BPF_PROG(test_task_from_pid_arg, struct task_struct *task, u64 clone_flags)
 {
-	struct task_struct *acquired;
-
 	if (!is_test_kfunc_task())
 		return 0;
 
@@ -183,8 +181,6 @@ int BPF_PROG(test_task_from_pid_arg, struct task_struct *task, u64 clone_flags)
 SEC("tp_btf/task_newtask")
 int BPF_PROG(test_task_from_pid_current, struct task_struct *task, u64 clone_flags)
 {
-	struct task_struct *current, *acquired;
-
 	if (!is_test_kfunc_task())
 		return 0;
 
@@ -208,8 +204,6 @@ static int is_pid_lookup_valid(s32 pid)
 SEC("tp_btf/task_newtask")
 int BPF_PROG(test_task_from_pid_invalid, struct task_struct *task, u64 clone_flags)
 {
-	struct task_struct *acquired;
-
 	if (!is_test_kfunc_task())
 		return 0;
 
diff --git a/tools/testing/selftests/bpf/progs/test_bpf_nf.c b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
index 9fc603c9d673..77ad8adf68da 100644
--- a/tools/testing/selftests/bpf/progs/test_bpf_nf.c
+++ b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
@@ -75,7 +75,6 @@ nf_ct_test(struct nf_conn *(*lookup_fn)(void *, struct bpf_sock_tuple *, u32,
 	struct bpf_ct_opts___local opts_def = { .l4proto = IPPROTO_TCP, .netns_id = -1 };
 	struct bpf_sock_tuple bpf_tuple;
 	struct nf_conn *ct;
-	int err;
 
 	__builtin_memset(&bpf_tuple, 0, sizeof(bpf_tuple.ipv4));
 
diff --git a/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c b/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c
index f45a7095de7a..f41c81212ee9 100644
--- a/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c
+++ b/tools/testing/selftests/bpf/progs/test_cls_redirect_dynptr.c
@@ -455,7 +455,6 @@ static ret_t forward_to_next_hop(struct __sk_buff *skb, struct bpf_dynptr *dynpt
 
 static ret_t skip_next_hops(__u64 *offset, int n)
 {
-	__u32 res;
 	switch (n) {
 	case 1:
 		*offset += sizeof(struct in_addr);
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c b/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c
index ab1e647aeb31..b86fdda2a6ea 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_bitfields_probed.c
@@ -42,7 +42,6 @@ int test_core_bitfields(void *ctx)
 {
 	struct core_reloc_bitfields *in = (void *)&data.in;
 	struct core_reloc_bitfields_output *out = (void *)&data.out;
-	uint64_t res;
 
 	out->ub1 = BPF_CORE_READ_BITFIELD_PROBED(in, ub1);
 	out->ub2 = BPF_CORE_READ_BITFIELD_PROBED(in, ub2);
diff --git a/tools/testing/selftests/bpf/progs/test_global_func1.c b/tools/testing/selftests/bpf/progs/test_global_func1.c
index 23970a20b324..b85fc8c423ba 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func1.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func1.c
@@ -18,6 +18,8 @@ int f1(struct __sk_buff *skb)
 {
 	volatile char buf[MAX_STACK] = {};
 
+	__sink(buf[MAX_STACK - 1]);
+
 	return f0(0, skb) + skb->len;
 }
 
@@ -34,6 +36,8 @@ int f3(int val, struct __sk_buff *skb, int var)
 {
 	volatile char buf[MAX_STACK] = {};
 
+	__sink(buf[MAX_STACK - 1]);
+
 	return skb->ifindex * val * var;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/test_global_func2.c b/tools/testing/selftests/bpf/progs/test_global_func2.c
index 3dce97fb52a4..2beab9c3b68a 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func2.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func2.c
@@ -18,6 +18,8 @@ int f1(struct __sk_buff *skb)
 {
 	volatile char buf[MAX_STACK] = {};
 
+	__sink(buf[MAX_STACK - 1]);
+
 	return f0(0, skb) + skb->len;
 }
 
@@ -34,6 +36,8 @@ int f3(int val, struct __sk_buff *skb, int var)
 {
 	volatile char buf[MAX_STACK] = {};
 
+	__sink(buf[MAX_STACK - 1]);
+
 	return skb->ifindex * val * var;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/test_hash_large_key.c b/tools/testing/selftests/bpf/progs/test_hash_large_key.c
index 473a22794a62..8b438128f46b 100644
--- a/tools/testing/selftests/bpf/progs/test_hash_large_key.c
+++ b/tools/testing/selftests/bpf/progs/test_hash_large_key.c
@@ -28,7 +28,7 @@ struct bigelement {
 SEC("raw_tracepoint/sys_enter")
 int bpf_hash_large_key_test(void *ctx)
 {
-	int zero = 0, err = 1, value = 42;
+	int zero = 0, value = 42;
 	struct bigelement *key;
 
 	key = bpf_map_lookup_elem(&key_map, &zero);
diff --git a/tools/testing/selftests/bpf/progs/test_ksyms_btf_write_check.c b/tools/testing/selftests/bpf/progs/test_ksyms_btf_write_check.c
index a72a5bf3812a..27109b877714 100644
--- a/tools/testing/selftests/bpf/progs/test_ksyms_btf_write_check.c
+++ b/tools/testing/selftests/bpf/progs/test_ksyms_btf_write_check.c
@@ -35,7 +35,6 @@ SEC("raw_tp/sys_enter")
 int handler2(const void *ctx)
 {
 	int *active;
-	__u32 cpu;
 
 	active = bpf_this_cpu_ptr(&bpf_prog_active);
 	write_active(active);
diff --git a/tools/testing/selftests/bpf/progs/test_legacy_printk.c b/tools/testing/selftests/bpf/progs/test_legacy_printk.c
index 64c2d9ced529..42718cd8e6a4 100644
--- a/tools/testing/selftests/bpf/progs/test_legacy_printk.c
+++ b/tools/testing/selftests/bpf/progs/test_legacy_printk.c
@@ -56,7 +56,7 @@ int handle_legacy(void *ctx)
 SEC("tp/raw_syscalls/sys_enter")
 int handle_modern(void *ctx)
 {
-	int zero = 0, cur_pid;
+	int cur_pid;
 
 	cur_pid = bpf_get_current_pid_tgid() >> 32;
 	if (cur_pid != my_pid_var)
diff --git a/tools/testing/selftests/bpf/progs/test_map_lock.c b/tools/testing/selftests/bpf/progs/test_map_lock.c
index acf073db9e8b..1c02511b73cd 100644
--- a/tools/testing/selftests/bpf/progs/test_map_lock.c
+++ b/tools/testing/selftests/bpf/progs/test_map_lock.c
@@ -33,7 +33,7 @@ struct {
 SEC("cgroup/skb")
 int bpf_map_lock_test(struct __sk_buff *skb)
 {
-	struct hmap_elem zero = {}, *val;
+	struct hmap_elem *val;
 	int rnd = bpf_get_prandom_u32();
 	int key = 0, err = 1, i;
 	struct array_elem *q;
diff --git a/tools/testing/selftests/bpf/progs/test_obj_id.c b/tools/testing/selftests/bpf/progs/test_obj_id.c
index ded71b3ff6b4..2850ae788a91 100644
--- a/tools/testing/selftests/bpf/progs/test_obj_id.c
+++ b/tools/testing/selftests/bpf/progs/test_obj_id.c
@@ -4,6 +4,7 @@
 #include <stddef.h>
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
 
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
@@ -19,6 +20,7 @@ int test_obj_id(void *ctx)
 	__u64 *value;
 
 	value = bpf_map_lookup_elem(&test_map_id, &key);
+	__sink(value);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c
index 79bab9b50e9e..d9b2ba7ac340 100644
--- a/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c
+++ b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt.c
@@ -87,7 +87,6 @@ int xdp_ingress_v6(struct xdp_md *xdp)
 	__u8 tcp_hdr_opt_len = 0;
 	struct tcphdr *tcp_hdr;
 	__u64 tcp_offset = 0;
-	__u32 off;
 	int err;
 
 	tcp_offset = sizeof(struct ethhdr) + sizeof(struct ipv6hdr);
diff --git a/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c
index d3b319722e30..dc6e43bc6a62 100644
--- a/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c
+++ b/tools/testing/selftests/bpf/progs/test_parse_tcp_hdr_opt_dynptr.c
@@ -30,7 +30,7 @@ __u32 server_id;
 static int parse_hdr_opt(struct bpf_dynptr *ptr, __u32 *off, __u8 *hdr_bytes_remaining,
 			 __u32 *server_id)
 {
-	__u8 *tcp_opt, kind, hdr_len;
+	__u8 kind, hdr_len;
 	__u8 buffer[sizeof(kind) + sizeof(hdr_len) + sizeof(*server_id)];
 	__u8 *data;
 
diff --git a/tools/testing/selftests/bpf/progs/test_pkt_access.c b/tools/testing/selftests/bpf/progs/test_pkt_access.c
index 5cd7c096f62d..bce7173152c6 100644
--- a/tools/testing/selftests/bpf/progs/test_pkt_access.c
+++ b/tools/testing/selftests/bpf/progs/test_pkt_access.c
@@ -13,6 +13,7 @@
 #include <linux/pkt_cls.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_endian.h>
+#include "bpf_misc.h"
 
 /* llvm will optimize both subprograms into exactly the same BPF assembly
  *
@@ -51,6 +52,8 @@ int get_skb_len(struct __sk_buff *skb)
 {
 	volatile char buf[MAX_STACK] = {};
 
+	__sink(buf[MAX_STACK - 1]);
+
 	return skb->len;
 }
 
@@ -73,6 +76,8 @@ int get_skb_ifindex(int val, struct __sk_buff *skb, int var)
 {
 	volatile char buf[MAX_STACK] = {};
 
+	__sink(buf[MAX_STACK - 1]);
+
 	return skb->ifindex * val * var;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/test_ringbuf.c b/tools/testing/selftests/bpf/progs/test_ringbuf.c
index 5bdc0d38efc0..501cefa97633 100644
--- a/tools/testing/selftests/bpf/progs/test_ringbuf.c
+++ b/tools/testing/selftests/bpf/progs/test_ringbuf.c
@@ -41,7 +41,6 @@ int test_ringbuf(void *ctx)
 {
 	int cur_pid = bpf_get_current_pid_tgid() >> 32;
 	struct sample *sample;
-	int zero = 0;
 
 	if (cur_pid != pid)
 		return 0;
diff --git a/tools/testing/selftests/bpf/progs/test_ringbuf_map_key.c b/tools/testing/selftests/bpf/progs/test_ringbuf_map_key.c
index 2760bf60d05a..21bb7da90ea5 100644
--- a/tools/testing/selftests/bpf/progs/test_ringbuf_map_key.c
+++ b/tools/testing/selftests/bpf/progs/test_ringbuf_map_key.c
@@ -53,6 +53,7 @@ int test_ringbuf_mem_map_key(void *ctx)
 	/* test using 'sample' (PTR_TO_MEM | MEM_ALLOC) as map key arg
 	 */
 	lookup_val = (int *)bpf_map_lookup_elem(&hash_map, sample);
+	__sink(lookup_val);
 
 	/* workaround - memcpy is necessary so that verifier doesn't
 	 * complain with:
diff --git a/tools/testing/selftests/bpf/progs/test_ringbuf_multi.c b/tools/testing/selftests/bpf/progs/test_ringbuf_multi.c
index e416e0ce12b7..9626baa6779c 100644
--- a/tools/testing/selftests/bpf/progs/test_ringbuf_multi.c
+++ b/tools/testing/selftests/bpf/progs/test_ringbuf_multi.c
@@ -59,7 +59,6 @@ int test_ringbuf(void *ctx)
 	int cur_pid = bpf_get_current_pid_tgid() >> 32;
 	struct sample *sample;
 	void *rb;
-	int zero = 0;
 
 	if (cur_pid != pid)
 		return 0;
diff --git a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
index 7d56ed47cd4d..5eb25c6ad75b 100644
--- a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
@@ -64,7 +64,7 @@ SEC("sk_reuseport")
 int _select_by_skb_data(struct sk_reuseport_md *reuse_md)
 {
 	__u32 linum, index = 0, flags = 0, index_zero = 0;
-	__u32 *result_cnt, *linum_value;
+	__u32 *result_cnt;
 	struct data_check data_check = {};
 	struct cmd *cmd, cmd_copy;
 	void *data, *data_end;
diff --git a/tools/testing/selftests/bpf/progs/test_sk_assign.c b/tools/testing/selftests/bpf/progs/test_sk_assign.c
index 21b19b758c4e..3079244c7f96 100644
--- a/tools/testing/selftests/bpf/progs/test_sk_assign.c
+++ b/tools/testing/selftests/bpf/progs/test_sk_assign.c
@@ -15,6 +15,7 @@
 #include <sys/socket.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_endian.h>
+#include "bpf_misc.h"
 
 #if defined(IPROUTE2_HAVE_LIBBPF)
 /* Use a new-style map definition. */
@@ -57,7 +58,6 @@ get_tuple(struct __sk_buff *skb, bool *ipv4, bool *tcp)
 	void *data = (void *)(long)skb->data;
 	struct bpf_sock_tuple *result;
 	struct ethhdr *eth;
-	__u64 tuple_len;
 	__u8 proto = 0;
 	__u64 ihl_len;
 
@@ -94,6 +94,7 @@ get_tuple(struct __sk_buff *skb, bool *ipv4, bool *tcp)
 		return NULL;
 
 	*tcp = (proto == IPPROTO_TCP);
+	__sink(ihl_len);
 	return result;
 }
 
@@ -173,7 +174,6 @@ int bpf_sk_assign_test(struct __sk_buff *skb)
 	struct bpf_sock_tuple *tuple;
 	bool ipv4 = false;
 	bool tcp = false;
-	int tuple_len;
 	int ret = 0;
 
 	tuple = get_tuple(skb, &ipv4, &tcp);
diff --git a/tools/testing/selftests/bpf/progs/test_sk_lookup.c b/tools/testing/selftests/bpf/progs/test_sk_lookup.c
index 6058dcb11b36..71f844b9b902 100644
--- a/tools/testing/selftests/bpf/progs/test_sk_lookup.c
+++ b/tools/testing/selftests/bpf/progs/test_sk_lookup.c
@@ -391,7 +391,6 @@ SEC("sk_lookup")
 int ctx_narrow_access(struct bpf_sk_lookup *ctx)
 {
 	struct bpf_sock *sk;
-	int err, family;
 	__u32 val_u32;
 	bool v4;
 
@@ -645,9 +644,7 @@ static __always_inline int select_server_a(struct bpf_sk_lookup *ctx)
 SEC("sk_lookup")
 int multi_prog_redir1(struct bpf_sk_lookup *ctx)
 {
-	int ret;
-
-	ret = select_server_a(ctx);
+	(void)select_server_a(ctx);
 	bpf_map_update_elem(&run_map, &KEY_PROG1, &PROG_DONE, BPF_ANY);
 	return SK_PASS;
 }
@@ -655,9 +652,7 @@ int multi_prog_redir1(struct bpf_sk_lookup *ctx)
 SEC("sk_lookup")
 int multi_prog_redir2(struct bpf_sk_lookup *ctx)
 {
-	int ret;
-
-	ret = select_server_a(ctx);
+	(void)select_server_a(ctx);
 	bpf_map_update_elem(&run_map, &KEY_PROG2, &PROG_DONE, BPF_ANY);
 	return SK_PASS;
 }
diff --git a/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c b/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
index 6ccf6d546074..e9efc3263022 100644
--- a/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_sk_lookup_kern.c
@@ -110,7 +110,6 @@ int err_modify_sk_pointer(struct __sk_buff *skb)
 {
 	struct bpf_sock_tuple tuple = {};
 	struct bpf_sock *sk;
-	__u32 family;
 
 	sk = bpf_sk_lookup_tcp(skb, &tuple, sizeof(tuple), BPF_F_CURRENT_NETNS, 0);
 	if (sk) {
@@ -125,7 +124,6 @@ int err_modify_sk_or_null_pointer(struct __sk_buff *skb)
 {
 	struct bpf_sock_tuple tuple = {};
 	struct bpf_sock *sk;
-	__u32 family;
 
 	sk = bpf_sk_lookup_tcp(skb, &tuple, sizeof(tuple), BPF_F_CURRENT_NETNS, 0);
 	sk += 1;
diff --git a/tools/testing/selftests/bpf/progs/test_sock_fields.c b/tools/testing/selftests/bpf/progs/test_sock_fields.c
index 9f4b8f9f1181..bbad3c2d9aa5 100644
--- a/tools/testing/selftests/bpf/progs/test_sock_fields.c
+++ b/tools/testing/selftests/bpf/progs/test_sock_fields.c
@@ -121,7 +121,7 @@ static void tpcpy(struct bpf_tcp_sock *dst,
 SEC("cgroup_skb/egress")
 int egress_read_sock_fields(struct __sk_buff *skb)
 {
-	struct bpf_spinlock_cnt cli_cnt_init = { .lock = 0, .cnt = 0xeB9F };
+	struct bpf_spinlock_cnt cli_cnt_init = { .lock = {}, .cnt = 0xeB9F };
 	struct bpf_spinlock_cnt *pkt_out_cnt, *pkt_out_cnt10;
 	struct bpf_tcp_sock *tp, *tp_ret;
 	struct bpf_sock *sk, *sk_ret;
diff --git a/tools/testing/selftests/bpf/progs/test_sockmap_kern.h b/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
index 6c85b00f27b2..baf9ebc6d903 100644
--- a/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
+++ b/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
@@ -14,6 +14,7 @@
 #include <sys/socket.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_endian.h>
+#include "bpf_misc.h"
 
 /* Sockmap sample program connects a client and a backend together
  * using cgroups.
@@ -111,12 +112,15 @@ int bpf_prog2(struct __sk_buff *skb)
 	int len, *f, ret, zero = 0;
 	__u64 flags = 0;
 
+	__sink(rport);
 	if (lport == 10000)
 		ret = 10;
 	else
 		ret = 1;
 
 	len = (__u32)skb->data_end - (__u32)skb->data;
+	__sink(len);
+
 	f = bpf_map_lookup_elem(&sock_skb_opts, &zero);
 	if (f && *f) {
 		ret = 3;
@@ -180,7 +184,6 @@ int bpf_prog3(struct __sk_buff *skb)
 	if (err)
 		return SK_DROP;
 	bpf_write_pass(skb, 13);
-tls_out:
 	return ret;
 }
 
@@ -188,8 +191,7 @@ SEC("sockops")
 int bpf_sockmap(struct bpf_sock_ops *skops)
 {
 	__u32 lport, rport;
-	int op, err = 0, index, key, ret;
-
+	int op, err, ret;
 
 	op = (int) skops->op;
 
@@ -228,6 +230,8 @@ int bpf_sockmap(struct bpf_sock_ops *skops)
 		break;
 	}
 
+	__sink(err);
+
 	return 0;
 }
 
@@ -321,6 +325,10 @@ int bpf_prog8(struct sk_msg_md *msg)
 	} else {
 		return SK_DROP;
 	}
+
+	__sink(data_end);
+	__sink(data);
+
 	return SK_PASS;
 }
 SEC("sk_msg4")
diff --git a/tools/testing/selftests/bpf/progs/test_spin_lock.c b/tools/testing/selftests/bpf/progs/test_spin_lock.c
index 5bd10409285b..b2440a0ff422 100644
--- a/tools/testing/selftests/bpf/progs/test_spin_lock.c
+++ b/tools/testing/selftests/bpf/progs/test_spin_lock.c
@@ -3,6 +3,7 @@
 #include <linux/bpf.h>
 #include <linux/version.h>
 #include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
 
 struct hmap_elem {
 	volatile int cnt;
@@ -89,6 +90,8 @@ int bpf_spin_lock_test(struct __sk_buff *skb)
 	credit = q->credit;
 	bpf_spin_unlock(&q->lock);
 
+	__sink(credit);
+
 	/* spin_lock in cgroup local storage */
 	cls = bpf_get_local_storage(&cls_map, 0);
 	bpf_spin_lock(&cls->lock);
diff --git a/tools/testing/selftests/bpf/progs/test_tc_dtime.c b/tools/testing/selftests/bpf/progs/test_tc_dtime.c
index 125beec31834..74ec09f040b7 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_dtime.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_dtime.c
@@ -163,9 +163,9 @@ static int skb_get_type(struct __sk_buff *skb)
 		ip6h = data + sizeof(struct ethhdr);
 		if (ip6h + 1 > data_end)
 			return -1;
-		if (v6_equal(ip6h->saddr, (struct in6_addr)ip6_src))
+		if (v6_equal(ip6h->saddr, (struct in6_addr){{ip6_src}}))
 			ns = SRC_NS;
-		else if (v6_equal(ip6h->saddr, (struct in6_addr)ip6_dst))
+		else if (v6_equal(ip6h->saddr, (struct in6_addr){{ip6_dst}}))
 			ns = DST_NS;
 		inet_proto = ip6h->nexthdr;
 		trans = ip6h + 1;
diff --git a/tools/testing/selftests/bpf/progs/test_tc_neigh.c b/tools/testing/selftests/bpf/progs/test_tc_neigh.c
index 3e32ea375ab4..de15155f2609 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_neigh.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_neigh.c
@@ -94,7 +94,7 @@ int tc_dst(struct __sk_buff *skb)
 		redirect = is_remote_ep_v4(skb, __bpf_constant_htonl(ip4_src));
 		break;
 	case __bpf_constant_htons(ETH_P_IPV6):
-		redirect = is_remote_ep_v6(skb, (struct in6_addr)ip6_src);
+		redirect = is_remote_ep_v6(skb, (struct in6_addr){{ip6_src}});
 		break;
 	}
 
@@ -119,7 +119,7 @@ int tc_src(struct __sk_buff *skb)
 		redirect = is_remote_ep_v4(skb, __bpf_constant_htonl(ip4_dst));
 		break;
 	case __bpf_constant_htons(ETH_P_IPV6):
-		redirect = is_remote_ep_v6(skb, (struct in6_addr)ip6_dst);
+		redirect = is_remote_ep_v6(skb, (struct in6_addr){{ip6_dst}});
 		break;
 	}
 
diff --git a/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c b/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c
index 3ded05280757..cf7ed8cbb1fe 100644
--- a/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c
@@ -46,8 +46,6 @@ int bpf_testcb(struct bpf_sock_ops *skops)
 	struct bpf_sock_ops *reuse = skops;
 	struct tcphdr *thdr;
 	int window_clamp = 9216;
-	int good_call_rv = 0;
-	int bad_call_rv = 0;
 	int save_syn = 1;
 	int rv = -1;
 	int v = 0;
diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
index 95b4aa0928ba..9ab2d55ab7c0 100644
--- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
@@ -209,7 +209,6 @@ int erspan_get_tunnel(struct __sk_buff *skb)
 {
 	struct bpf_tunnel_key key;
 	struct erspan_metadata md;
-	__u32 index;
 	int ret;
 
 	ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key), 0);
@@ -289,7 +288,6 @@ int ip4ip6erspan_get_tunnel(struct __sk_buff *skb)
 {
 	struct bpf_tunnel_key key;
 	struct erspan_metadata md;
-	__u32 index;
 	int ret;
 
 	ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key),
@@ -405,8 +403,6 @@ int vxlan_get_tunnel_src(struct __sk_buff *skb)
 	int ret;
 	struct bpf_tunnel_key key;
 	struct vxlan_metadata md;
-	__u32 orig_daddr;
-	__u32 index = 0;
 
 	ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key),
 				     BPF_F_TUNINFO_FLAGS);
@@ -443,9 +439,7 @@ int veth_set_outer_dst(struct __sk_buff *skb)
 	void *data_end = (void *)(long)skb->data_end;
 	struct udphdr *udph;
 	struct iphdr *iph;
-	__u32 index = 0;
 	int ret = 0;
-	int shrink;
 	__s64 csum;
 
 	if ((void *)eth + sizeof(*eth) > data_end) {
diff --git a/tools/testing/selftests/bpf/progs/test_usdt_multispec.c b/tools/testing/selftests/bpf/progs/test_usdt_multispec.c
index aa6de32b50d1..962f3462066a 100644
--- a/tools/testing/selftests/bpf/progs/test_usdt_multispec.c
+++ b/tools/testing/selftests/bpf/progs/test_usdt_multispec.c
@@ -18,8 +18,6 @@ int usdt_100_sum;
 SEC("usdt//proc/self/exe:test:usdt_100")
 int BPF_USDT(usdt_100, int x)
 {
-	long tmp;
-
 	if (my_pid != (bpf_get_current_pid_tgid() >> 32))
 		return 0;
 
diff --git a/tools/testing/selftests/bpf/progs/test_verif_scale1.c b/tools/testing/selftests/bpf/progs/test_verif_scale1.c
index ac6135d9374c..323a73fb2e8c 100644
--- a/tools/testing/selftests/bpf/progs/test_verif_scale1.c
+++ b/tools/testing/selftests/bpf/progs/test_verif_scale1.c
@@ -11,7 +11,7 @@ int balancer_ingress(struct __sk_buff *ctx)
 	void *data_end = (void *)(long)ctx->data_end;
 	void *data = (void *)(long)ctx->data;
 	void *ptr;
-	int ret = 0, nh_off, i = 0;
+	int nh_off, i = 0;
 
 	nh_off = 14;
 
diff --git a/tools/testing/selftests/bpf/progs/test_verif_scale2.c b/tools/testing/selftests/bpf/progs/test_verif_scale2.c
index f90ffcafd1e8..f5318f757084 100644
--- a/tools/testing/selftests/bpf/progs/test_verif_scale2.c
+++ b/tools/testing/selftests/bpf/progs/test_verif_scale2.c
@@ -11,7 +11,7 @@ int balancer_ingress(struct __sk_buff *ctx)
 	void *data_end = (void *)(long)ctx->data_end;
 	void *data = (void *)(long)ctx->data;
 	void *ptr;
-	int ret = 0, nh_off, i = 0;
+	int nh_off, i = 0;
 
 	nh_off = 14;
 
diff --git a/tools/testing/selftests/bpf/progs/test_verif_scale3.c b/tools/testing/selftests/bpf/progs/test_verif_scale3.c
index ca33a9b711c4..2e06dbb1ad5c 100644
--- a/tools/testing/selftests/bpf/progs/test_verif_scale3.c
+++ b/tools/testing/selftests/bpf/progs/test_verif_scale3.c
@@ -11,7 +11,7 @@ int balancer_ingress(struct __sk_buff *ctx)
 	void *data_end = (void *)(long)ctx->data_end;
 	void *data = (void *)(long)ctx->data;
 	void *ptr;
-	int ret = 0, nh_off, i = 0;
+	int nh_off, i = 0;
 
 	nh_off = 32;
 
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_adjust_tail_grow.c b/tools/testing/selftests/bpf/progs/test_xdp_adjust_tail_grow.c
index 297c260fc364..81bb38d72ced 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_adjust_tail_grow.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_adjust_tail_grow.c
@@ -5,8 +5,6 @@
 SEC("xdp")
 int _xdp_adjust_tail_grow(struct xdp_md *xdp)
 {
-	void *data_end = (void *)(long)xdp->data_end;
-	void *data = (void *)(long)xdp->data;
 	int data_len = bpf_xdp_get_buff_len(xdp);
 	int offset = 0;
 	/* SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) */
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
index 3379d303f41a..ee48c4963971 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
@@ -45,8 +45,6 @@ SEC("fentry/FUNC")
 int BPF_PROG(trace_on_entry, struct xdp_buff *xdp)
 {
 	struct meta meta;
-	void *data_end = (void *)(long)xdp->data_end;
-	void *data = (void *)(long)xdp->data;
 
 	meta.ifindex = xdp->rxq->dev->ifindex;
 	meta.pkt_len = bpf_xdp_get_buff_len((struct xdp_md *)xdp);
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c b/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c
index 7521a805b506..25ee4a22e48d 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_dynptr.c
@@ -82,7 +82,6 @@ static __always_inline int handle_ipv4(struct xdp_md *xdp, struct bpf_dynptr *xd
 	struct iptnl_info *tnl;
 	struct ethhdr *new_eth;
 	struct ethhdr *old_eth;
-	__u32 transport_hdr_sz;
 	struct iphdr *iph;
 	__u16 *next_iph;
 	__u16 payload_len;
@@ -165,7 +164,6 @@ static __always_inline int handle_ipv6(struct xdp_md *xdp, struct bpf_dynptr *xd
 	struct iptnl_info *tnl;
 	struct ethhdr *new_eth;
 	struct ethhdr *old_eth;
-	__u32 transport_hdr_sz;
 	struct ipv6hdr *ip6h;
 	__u16 payload_len;
 	struct vip vip = {};
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_noinline.c b/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
index ba48fcb98ab2..42c8f6ded0e4 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_noinline.c
@@ -371,45 +371,6 @@ bool encap_v4(struct xdp_md *xdp, struct ctl_value *cval,
 	return true;
 }
 
-static __attribute__ ((noinline))
-bool decap_v6(struct xdp_md *xdp, void **data, void **data_end, bool inner_v4)
-{
-	struct eth_hdr *new_eth;
-	struct eth_hdr *old_eth;
-
-	old_eth = *data;
-	new_eth = *data + sizeof(struct ipv6hdr);
-	memcpy(new_eth->eth_source, old_eth->eth_source, 6);
-	memcpy(new_eth->eth_dest, old_eth->eth_dest, 6);
-	if (inner_v4)
-		new_eth->eth_proto = 8;
-	else
-		new_eth->eth_proto = 56710;
-	if (bpf_xdp_adjust_head(xdp, (int)sizeof(struct ipv6hdr)))
-		return false;
-	*data = (void *)(long)xdp->data;
-	*data_end = (void *)(long)xdp->data_end;
-	return true;
-}
-
-static __attribute__ ((noinline))
-bool decap_v4(struct xdp_md *xdp, void **data, void **data_end)
-{
-	struct eth_hdr *new_eth;
-	struct eth_hdr *old_eth;
-
-	old_eth = *data;
-	new_eth = *data + sizeof(struct iphdr);
-	memcpy(new_eth->eth_source, old_eth->eth_source, 6);
-	memcpy(new_eth->eth_dest, old_eth->eth_dest, 6);
-	new_eth->eth_proto = 8;
-	if (bpf_xdp_adjust_head(xdp, (int)sizeof(struct iphdr)))
-		return false;
-	*data = (void *)(long)xdp->data;
-	*data_end = (void *)(long)xdp->data_end;
-	return true;
-}
-
 static __attribute__ ((noinline))
 int swap_mac_and_send(void *data, void *data_end)
 {
@@ -430,7 +391,6 @@ int send_icmp_reply(void *data, void *data_end)
 	__u16 *next_iph_u16;
 	__u32 tmp_addr = 0;
 	struct iphdr *iph;
-	__u32 csum1 = 0;
 	__u32 csum = 0;
 	__u64 off = 0;
 
@@ -662,7 +622,6 @@ static int process_l3_headers_v4(struct packet_description *pckt,
 				 void *data_end)
 {
 	struct iphdr *iph;
-	__u64 iph_len;
 	int action;
 
 	iph = data + off;
@@ -696,7 +655,6 @@ static int process_packet(void *data, __u64 off, void *data_end,
 	struct packet_description pckt = { };
 	struct vip_definition vip = { };
 	struct lb_stats *data_stats;
-	struct eth_hdr *eth = data;
 	void *lru_map = &lru_cache;
 	struct vip_meta *vip_info;
 	__u32 lru_stats_key = 513;
@@ -704,7 +662,6 @@ static int process_packet(void *data, __u64 off, void *data_end,
 	__u32 stats_key = 512;
 	struct ctl_value *cval;
 	__u16 pkt_bytes;
-	__u64 iph_len;
 	__u8 protocol;
 	__u32 vip_num;
 	int action;
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_vlan.c b/tools/testing/selftests/bpf/progs/test_xdp_vlan.c
index 4ddcb6dfe500..f3ec8086482d 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_vlan.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_vlan.c
@@ -209,19 +209,6 @@ int  xdp_prognum2(struct xdp_md *ctx)
 	return XDP_PASS;
 }
 
-static __always_inline
-void shift_mac_4bytes_16bit(void *data)
-{
-	__u16 *p = data;
-
-	p[7] = p[5]; /* delete p[7] was vlan_hdr->h_vlan_TCI */
-	p[6] = p[4]; /* delete p[6] was ethhdr->h_proto */
-	p[5] = p[3];
-	p[4] = p[2];
-	p[3] = p[1];
-	p[2] = p[0];
-}
-
 static __always_inline
 void shift_mac_4bytes_32bit(void *data)
 {
diff --git a/tools/testing/selftests/bpf/progs/type_cast.c b/tools/testing/selftests/bpf/progs/type_cast.c
index eb78e6f03129..a9629ac230fd 100644
--- a/tools/testing/selftests/bpf/progs/type_cast.c
+++ b/tools/testing/selftests/bpf/progs/type_cast.c
@@ -63,7 +63,6 @@ SEC("?tp_btf/sys_enter")
 int BPF_PROG(untrusted_ptr, struct pt_regs *regs, long id)
 {
 	struct task_struct *task, *task_dup;
-	long *ptr;
 
 	task = bpf_get_current_task_btf();
 	task_dup = bpf_rdonly_cast(task, bpf_core_type_id_kernel(struct task_struct));
diff --git a/tools/testing/selftests/bpf/progs/udp_limit.c b/tools/testing/selftests/bpf/progs/udp_limit.c
index 165e3c2dd9a3..4767451b59ac 100644
--- a/tools/testing/selftests/bpf/progs/udp_limit.c
+++ b/tools/testing/selftests/bpf/progs/udp_limit.c
@@ -17,7 +17,6 @@ SEC("cgroup/sock_create")
 int sock(struct bpf_sock *ctx)
 {
 	int *sk_storage;
-	__u32 key;
 
 	if (ctx->type != SOCK_DGRAM)
 		return 1;
@@ -46,7 +45,6 @@ SEC("cgroup/sock_release")
 int sock_release(struct bpf_sock *ctx)
 {
 	int *sk_storage;
-	__u32 key;
 
 	if (ctx->type != SOCK_DGRAM)
 		return 1;
diff --git a/tools/testing/selftests/bpf/progs/user_ringbuf_success.c b/tools/testing/selftests/bpf/progs/user_ringbuf_success.c
index 0ade1110613b..dd3bdf672633 100644
--- a/tools/testing/selftests/bpf/progs/user_ringbuf_success.c
+++ b/tools/testing/selftests/bpf/progs/user_ringbuf_success.c
@@ -162,8 +162,6 @@ SEC("fentry/" SYS_PREFIX "sys_prctl")
 int test_user_ringbuf_protocol(void *ctx)
 {
 	long status = 0;
-	struct sample *sample = NULL;
-	struct bpf_dynptr ptr;
 
 	if (!is_test_process())
 		return 0;
@@ -183,10 +181,6 @@ int test_user_ringbuf_protocol(void *ctx)
 SEC("fentry/" SYS_PREFIX "sys_getpgid")
 int test_user_ringbuf(void *ctx)
 {
-	int status = 0;
-	struct sample *sample = NULL;
-	struct bpf_dynptr ptr;
-
 	if (!is_test_process())
 		return 0;
 
diff --git a/tools/testing/selftests/bpf/progs/xdp_features.c b/tools/testing/selftests/bpf/progs/xdp_features.c
index 87c247d56f72..67424084a38a 100644
--- a/tools/testing/selftests/bpf/progs/xdp_features.c
+++ b/tools/testing/selftests/bpf/progs/xdp_features.c
@@ -70,7 +70,6 @@ xdp_process_echo_packet(struct xdp_md *xdp, bool dut)
 	struct tlv_hdr *tlv;
 	struct udphdr *uh;
 	__be16 port;
-	__u8 *cmd;
 
 	if (eh + 1 > (struct ethhdr *)data_end)
 		return -EINVAL;
diff --git a/tools/testing/selftests/bpf/progs/xdping_kern.c b/tools/testing/selftests/bpf/progs/xdping_kern.c
index 4ad73847b8a5..54cf1765118b 100644
--- a/tools/testing/selftests/bpf/progs/xdping_kern.c
+++ b/tools/testing/selftests/bpf/progs/xdping_kern.c
@@ -89,7 +89,6 @@ static __always_inline int icmp_check(struct xdp_md *ctx, int type)
 SEC("xdp")
 int xdping_client(struct xdp_md *ctx)
 {
-	void *data_end = (void *)(long)ctx->data_end;
 	void *data = (void *)(long)ctx->data;
 	struct pinginfo *pinginfo = NULL;
 	struct ethhdr *eth = data;
@@ -153,7 +152,6 @@ int xdping_client(struct xdp_md *ctx)
 SEC("xdp")
 int xdping_server(struct xdp_md *ctx)
 {
-	void *data_end = (void *)(long)ctx->data_end;
 	void *data = (void *)(long)ctx->data;
 	struct ethhdr *eth = data;
 	struct icmphdr *icmph;
diff --git a/tools/testing/selftests/bpf/progs/xdpwall.c b/tools/testing/selftests/bpf/progs/xdpwall.c
index 7a891a0c3a39..c2dd0c28237a 100644
--- a/tools/testing/selftests/bpf/progs/xdpwall.c
+++ b/tools/testing/selftests/bpf/progs/xdpwall.c
@@ -321,7 +321,6 @@ int edgewall(struct xdp_md *ctx)
 	void *data = (void *)(long)(ctx->data);
 	struct fw_match_info match_info = {};
 	struct pkt_info info = {};
-	__u8 parse_err = NO_ERR;
 	void *transport_hdr;
 	struct ethhdr *eth;
 	bool filter_res;
-- 
cgit v1.2.3-70-g09d2


From 3d5a55ddc25508fe950991603d0224c0bba60558 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 8 Mar 2023 21:40:15 -0800
Subject: selftests/bpf: make BPF compiler flags stricter

We recently added -Wuninitialized, but it's not enough to catch various
silly mistakes or omissions. Let's go all the way to -Wall, just like we
do for user-space code.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230309054015.4068562-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 16f404aa1b23..606e2d738dd8 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -352,12 +352,12 @@ CLANG_TARGET_ARCH = --target=$(notdir $(CROSS_COMPILE:%-=%))
 endif
 
 CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG),$(CLANG_TARGET_ARCH))
-BPF_CFLAGS = -g -Werror -D__TARGET_ARCH_$(SRCARCH) $(MENDIAN) 		\
+BPF_CFLAGS = -g -Wall -Werror -D__TARGET_ARCH_$(SRCARCH) $(MENDIAN)	\
 	     -I$(INCLUDE_DIR) -I$(CURDIR) -I$(APIDIR)			\
 	     -I$(abspath $(OUTPUT)/../usr/include)
 
 CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \
-	       -Wno-compare-distinct-pointer-types -Wuninitialized
+	       -Wno-compare-distinct-pointer-types
 
 $(OUTPUT)/test_l4lb_noinline.o: BPF_CFLAGS += -fno-inline
 $(OUTPUT)/test_xdp_noinline.o: BPF_CFLAGS += -fno-inline
-- 
cgit v1.2.3-70-g09d2


From 4b5ce570dbef57a20acdd71b0c65376009012354 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 9 Mar 2023 22:01:49 -0800
Subject: bpf: ensure state checkpointing at iter_next() call sites

State equivalence check and checkpointing performed in is_state_visited()
employs certain heuristics to try to save memory by avoiding state checkpoints
if not enough jumps and instructions happened since last checkpoint. This leads
to unpredictability of whether a particular instruction will be checkpointed
and how regularly. While normally this is not causing much problems (except
inconveniences for predictable verifier tests, which we overcome with
BPF_F_TEST_STATE_FREQ flag), turns out it's not the case for open-coded
iterators.

Checking and saving state checkpoints at iter_next() call is crucial for fast
convergence of open-coded iterator loop logic, so we need to force it. If we
don't do that, is_state_visited() might skip saving a checkpoint, causing
unnecessarily long sequence of not checkpointed instructions and jumps, leading
to exhaustion of jump history buffer, and potentially other undesired outcomes.
It is expected that with correct open-coded iterators convergence will happen
quickly, so we don't run a risk of exhausting memory.

This patch adds, in addition to prune and jump instruction marks, also a
"forced checkpoint" mark, and makes sure that any iter_next() call instruction
is marked as such.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230310060149.625887-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_verifier.h |  6 +++++-
 kernel/bpf/verifier.c        | 31 ++++++++++++++++++++++++++++---
 2 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 0c052bc79940..81d525d057c7 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -477,8 +477,12 @@ struct bpf_insn_aux_data {
 
 	/* below fields are initialized once */
 	unsigned int orig_idx; /* original instruction index */
-	bool prune_point;
 	bool jmp_point;
+	bool prune_point;
+	/* ensure we check state equivalence and save state checkpoint and
+	 * this instruction, regardless of any heuristics
+	 */
+	bool force_checkpoint;
 };
 
 #define MAX_USED_MAPS 64 /* max number of maps accessed by one eBPF program */
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 45a082284464..13fd4c893f3b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -13865,6 +13865,17 @@ static bool is_prune_point(struct bpf_verifier_env *env, int insn_idx)
 	return env->insn_aux_data[insn_idx].prune_point;
 }
 
+static void mark_force_checkpoint(struct bpf_verifier_env *env, int idx)
+{
+	env->insn_aux_data[idx].force_checkpoint = true;
+}
+
+static bool is_force_checkpoint(struct bpf_verifier_env *env, int insn_idx)
+{
+	return env->insn_aux_data[insn_idx].force_checkpoint;
+}
+
+
 enum {
 	DONE_EXPLORING = 0,
 	KEEP_EXPLORING = 1,
@@ -13984,8 +13995,21 @@ static int visit_insn(int t, struct bpf_verifier_env *env)
 			struct bpf_kfunc_call_arg_meta meta;
 
 			ret = fetch_kfunc_meta(env, insn, &meta, NULL);
-			if (ret == 0 && is_iter_next_kfunc(&meta))
+			if (ret == 0 && is_iter_next_kfunc(&meta)) {
 				mark_prune_point(env, t);
+				/* Checking and saving state checkpoints at iter_next() call
+				 * is crucial for fast convergence of open-coded iterator loop
+				 * logic, so we need to force it. If we don't do that,
+				 * is_state_visited() might skip saving a checkpoint, causing
+				 * unnecessarily long sequence of not checkpointed
+				 * instructions and jumps, leading to exhaustion of jump
+				 * history buffer, and potentially other undesired outcomes.
+				 * It is expected that with correct open-coded iterators
+				 * convergence will happen quickly, so we don't run a risk of
+				 * exhausting memory.
+				 */
+				mark_force_checkpoint(env, t);
+			}
 		}
 		return visit_func_call_insn(t, insns, env, insn->src_reg == BPF_PSEUDO_CALL);
 
@@ -15172,7 +15196,8 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 	struct bpf_verifier_state_list *sl, **pprev;
 	struct bpf_verifier_state *cur = env->cur_state, *new;
 	int i, j, err, states_cnt = 0;
-	bool add_new_state = env->test_state_freq ? true : false;
+	bool force_new_state = env->test_state_freq || is_force_checkpoint(env, insn_idx);
+	bool add_new_state = force_new_state;
 
 	/* bpf progs typically have pruning point every 4 instructions
 	 * http://vger.kernel.org/bpfconf2019.html#session-1
@@ -15269,7 +15294,7 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx)
 			 * at the end of the loop are likely to be useful in pruning.
 			 */
 skip_inf_loop_check:
-			if (!env->test_state_freq &&
+			if (!force_new_state &&
 			    env->jmps_processed - env->prev_jmps_processed < 20 &&
 			    env->insn_processed - env->prev_insn_processed < 100)
 				add_new_state = false;
-- 
cgit v1.2.3-70-g09d2


From 52c2b005a3c18c565fc70cfd0ca49375f301e952 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 9 Mar 2023 14:41:31 -0800
Subject: bpf: take into account liveness when propagating precision

When doing state comparison, if old state has register that is not
marked as REG_LIVE_READ, then we just skip comparison, regardless what's
the state of corresponing register in current state. This is because not
REG_LIVE_READ register is irrelevant for further program execution and
correctness. All good here.

But when we get to precision propagation, after two states were declared
equivalent, we don't take into account old register's liveness, and thus
attempt to propagate precision for register in current state even if
that register in old state was not REG_LIVE_READ anymore. This is bad,
because register in current state could be anything at all and this
could cause -EFAULT due to internal logic bugs.

Fix by taking into account REG_LIVE_READ liveness mark to keep the logic
in state comparison in sync with precision propagation.

Fixes: a3ce685dd01a ("bpf: fix precision tracking")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230309224131.57449-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 13fd4c893f3b..0aaf7703326b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -15058,7 +15058,8 @@ static int propagate_precision(struct bpf_verifier_env *env,
 		state_reg = state->regs;
 		for (i = 0; i < BPF_REG_FP; i++, state_reg++) {
 			if (state_reg->type != SCALAR_VALUE ||
-			    !state_reg->precise)
+			    !state_reg->precise ||
+			    !(state_reg->live & REG_LIVE_READ))
 				continue;
 			if (env->log.level & BPF_LOG_LEVEL2)
 				verbose(env, "frame %d: propagating r%d\n", i, fr);
@@ -15072,7 +15073,8 @@ static int propagate_precision(struct bpf_verifier_env *env,
 				continue;
 			state_reg = &state->stack[i].spilled_ptr;
 			if (state_reg->type != SCALAR_VALUE ||
-			    !state_reg->precise)
+			    !state_reg->precise ||
+			    !(state_reg->live & REG_LIVE_READ))
 				continue;
 			if (env->log.level & BPF_LOG_LEVEL2)
 				verbose(env, "frame %d: propagating fp%d\n",
-- 
cgit v1.2.3-70-g09d2


From 4a54de65964d37c3929379271ab31355e93ccddf Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Fri, 10 Mar 2023 00:19:09 -0600
Subject: bpf/selftests: Fix send_signal tracepoint tests

The send_signal tracepoint tests are non-deterministically failing in
CI. The test works as follows:

1. Two pairs of file descriptors are created using the pipe() function.
   One pair is used to communicate between a parent process -> child
   process, and the other for the reverse direction.

2. A child is fork()'ed. The child process registers a signal handler,
   notifies its parent that the signal handler is registered, and then
   and waits for its parent to have enabled a BPF program that sends a
   signal.

3. The parent opens and loads a BPF skeleton with programs that send
   signals to the child process. The different programs are triggered by
   different perf events (either NMI or normal perf), or by regular
   tracepoints. The signal is delivered to the child whenever the child
   triggers the program.

4. The child's signal handler is invoked, which sets a flag saying that
   the signal handler was reached. The child then signals to the parent
   that it received the signal, and the test ends.

The perf testcases (send_signal_perf{_thread} and
send_signal_nmi{_thread}) work 100% of the time, but the tracepoint
testcases fail non-deterministically because the tracepoint is not
always being fired for the child.

There are two tracepoint programs registered in the test:
'tracepoint/sched/sched_switch', and
'tracepoint/syscalls/sys_enter_nanosleep'. The child never intentionally
blocks, nor sleeps, so neither tracepoint is guaranteed to be triggered.
To fix this, we can have the child trigger the nanosleep program with a
usleep().

Before this patch, the test would fail locally every 2-3 runs. Now, it
doesn't fail after more than 1000 runs.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230310061909.1420887-1-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/send_signal.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/send_signal.c b/tools/testing/selftests/bpf/prog_tests/send_signal.c
index d63a20fbed33..b15b343ebb6b 100644
--- a/tools/testing/selftests/bpf/prog_tests/send_signal.c
+++ b/tools/testing/selftests/bpf/prog_tests/send_signal.c
@@ -64,8 +64,12 @@ static void test_send_signal_common(struct perf_event_attr *attr,
 		ASSERT_EQ(read(pipe_p2c[0], buf, 1), 1, "pipe_read");
 
 		/* wait a little for signal handler */
-		for (int i = 0; i < 1000000000 && !sigusr1_received; i++)
+		for (int i = 0; i < 1000000000 && !sigusr1_received; i++) {
 			j /= i + j + 1;
+			if (!attr)
+				/* trigger the nanosleep tracepoint program. */
+				usleep(1);
+		}
 
 		buf[0] = sigusr1_received ? '2' : '0';
 		ASSERT_EQ(sigusr1_received, 1, "sigusr1_received");
-- 
cgit v1.2.3-70-g09d2


From 4cbd23cc92c49173e402753cab62b8a7754ed18f Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:20 -0800
Subject: bpf: Move a few bpf_local_storage functions to static scope

This patch moves the bpf_local_storage_free_rcu() and
bpf_selem_unlink_map() to static because they are
not used outside of bpf_local_storage.c.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-2-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h | 3 ---
 kernel/bpf/bpf_local_storage.c    | 4 ++--
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index d934248b8e81..502ad7093f13 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -147,8 +147,6 @@ void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu);
 void bpf_selem_link_map(struct bpf_local_storage_map *smap,
 			struct bpf_local_storage_elem *selem);
 
-void bpf_selem_unlink_map(struct bpf_local_storage_elem *selem);
-
 struct bpf_local_storage_elem *
 bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner, void *value,
 		bool charge_mem, gfp_t gfp_flags);
@@ -163,7 +161,6 @@ struct bpf_local_storage_data *
 bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap,
 			 void *value, u64 map_flags, gfp_t gfp_flags);
 
-void bpf_local_storage_free_rcu(struct rcu_head *rcu);
 u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map);
 
 #endif /* _BPF_LOCAL_STORAGE_H */
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index d3ba3f2db640..1904a4245ebe 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -95,7 +95,7 @@ bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner,
 	return NULL;
 }
 
-void bpf_local_storage_free_rcu(struct rcu_head *rcu)
+static void bpf_local_storage_free_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage *local_storage;
 
@@ -251,7 +251,7 @@ void bpf_selem_link_storage_nolock(struct bpf_local_storage *local_storage,
 	hlist_add_head_rcu(&selem->snode, &local_storage->list);
 }
 
-void bpf_selem_unlink_map(struct bpf_local_storage_elem *selem)
+static void bpf_selem_unlink_map(struct bpf_local_storage_elem *selem)
 {
 	struct bpf_local_storage_map *smap;
 	struct bpf_local_storage_map_bucket *b;
-- 
cgit v1.2.3-70-g09d2


From 2ffcb6fc50174d1efc8f98633eb2647d84483c68 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:21 -0800
Subject: bpf: Refactor codes into bpf_local_storage_destroy

This patch first renames bpf_local_storage_unlink_nolock to
bpf_local_storage_destroy(). It better reflects that it is only
used when the storage's owner (sk/task/cgrp/inode) is being kfree().

All bpf_local_storage_destroy's caller is taking the spin lock and
then free the storage. This patch also moves these two steps into
the bpf_local_storage_destroy.

This is a preparation work for a later patch that uses
bpf_mem_cache_alloc/free in the bpf_local_storage.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-3-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h | 2 +-
 kernel/bpf/bpf_cgrp_storage.c     | 9 +--------
 kernel/bpf/bpf_inode_storage.c    | 8 +-------
 kernel/bpf/bpf_local_storage.c    | 8 ++++++--
 kernel/bpf/bpf_task_storage.c     | 9 +--------
 net/core/bpf_sk_storage.c         | 8 +-------
 6 files changed, 11 insertions(+), 33 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 502ad7093f13..5908a954ddc2 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -128,7 +128,7 @@ bpf_local_storage_lookup(struct bpf_local_storage *local_storage,
 			 struct bpf_local_storage_map *smap,
 			 bool cacheit_lockit);
 
-bool bpf_local_storage_unlink_nolock(struct bpf_local_storage *local_storage);
+void bpf_local_storage_destroy(struct bpf_local_storage *local_storage);
 
 void bpf_local_storage_map_free(struct bpf_map *map,
 				struct bpf_local_storage_cache *cache,
diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c
index 9ae07aedaf23..492594d69a86 100644
--- a/kernel/bpf/bpf_cgrp_storage.c
+++ b/kernel/bpf/bpf_cgrp_storage.c
@@ -46,8 +46,6 @@ static struct bpf_local_storage __rcu **cgroup_storage_ptr(void *owner)
 void bpf_cgrp_storage_free(struct cgroup *cgroup)
 {
 	struct bpf_local_storage *local_storage;
-	bool free_cgroup_storage = false;
-	unsigned long flags;
 
 	rcu_read_lock();
 	local_storage = rcu_dereference(cgroup->bpf_cgrp_storage);
@@ -57,14 +55,9 @@ void bpf_cgrp_storage_free(struct cgroup *cgroup)
 	}
 
 	bpf_cgrp_storage_lock();
-	raw_spin_lock_irqsave(&local_storage->lock, flags);
-	free_cgroup_storage = bpf_local_storage_unlink_nolock(local_storage);
-	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
+	bpf_local_storage_destroy(local_storage);
 	bpf_cgrp_storage_unlock();
 	rcu_read_unlock();
-
-	if (free_cgroup_storage)
-		kfree_rcu(local_storage, rcu);
 }
 
 static struct bpf_local_storage_data *
diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c
index 43e2619c8167..2d25bcfa371b 100644
--- a/kernel/bpf/bpf_inode_storage.c
+++ b/kernel/bpf/bpf_inode_storage.c
@@ -57,7 +57,6 @@ static struct bpf_local_storage_data *inode_storage_lookup(struct inode *inode,
 void bpf_inode_storage_free(struct inode *inode)
 {
 	struct bpf_local_storage *local_storage;
-	bool free_inode_storage = false;
 	struct bpf_storage_blob *bsb;
 
 	bsb = bpf_inode(inode);
@@ -72,13 +71,8 @@ void bpf_inode_storage_free(struct inode *inode)
 		return;
 	}
 
-	raw_spin_lock_bh(&local_storage->lock);
-	free_inode_storage = bpf_local_storage_unlink_nolock(local_storage);
-	raw_spin_unlock_bh(&local_storage->lock);
+	bpf_local_storage_destroy(local_storage);
 	rcu_read_unlock();
-
-	if (free_inode_storage)
-		kfree_rcu(local_storage, rcu);
 }
 
 static void *bpf_fd_inode_storage_lookup_elem(struct bpf_map *map, void *key)
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 1904a4245ebe..e19f9f50a60d 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -652,11 +652,12 @@ int bpf_local_storage_map_check_btf(const struct bpf_map *map,
 	return 0;
 }
 
-bool bpf_local_storage_unlink_nolock(struct bpf_local_storage *local_storage)
+void bpf_local_storage_destroy(struct bpf_local_storage *local_storage)
 {
 	struct bpf_local_storage_elem *selem;
 	bool free_storage = false;
 	struct hlist_node *n;
+	unsigned long flags;
 
 	/* Neither the bpf_prog nor the bpf_map's syscall
 	 * could be modifying the local_storage->list now.
@@ -667,6 +668,7 @@ bool bpf_local_storage_unlink_nolock(struct bpf_local_storage *local_storage)
 	 * when unlinking elem from the local_storage->list and
 	 * the map's bucket->list.
 	 */
+	raw_spin_lock_irqsave(&local_storage->lock, flags);
 	hlist_for_each_entry_safe(selem, n, &local_storage->list, snode) {
 		/* Always unlink from map before unlinking from
 		 * local_storage.
@@ -681,8 +683,10 @@ bool bpf_local_storage_unlink_nolock(struct bpf_local_storage *local_storage)
 		free_storage = bpf_selem_unlink_storage_nolock(
 			local_storage, selem, false, false);
 	}
+	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
-	return free_storage;
+	if (free_storage)
+		kfree_rcu(local_storage, rcu);
 }
 
 u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map)
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index 20f942229f3c..4dcef28744d1 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -72,8 +72,6 @@ task_storage_lookup(struct task_struct *task, struct bpf_map *map,
 void bpf_task_storage_free(struct task_struct *task)
 {
 	struct bpf_local_storage *local_storage;
-	bool free_task_storage = false;
-	unsigned long flags;
 
 	rcu_read_lock();
 
@@ -84,14 +82,9 @@ void bpf_task_storage_free(struct task_struct *task)
 	}
 
 	bpf_task_storage_lock();
-	raw_spin_lock_irqsave(&local_storage->lock, flags);
-	free_task_storage = bpf_local_storage_unlink_nolock(local_storage);
-	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
+	bpf_local_storage_destroy(local_storage);
 	bpf_task_storage_unlock();
 	rcu_read_unlock();
-
-	if (free_task_storage)
-		kfree_rcu(local_storage, rcu);
 }
 
 static void *bpf_pid_task_storage_lookup_elem(struct bpf_map *map, void *key)
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index 7a36353dbc22..8f56438c104b 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -49,7 +49,6 @@ static int bpf_sk_storage_del(struct sock *sk, struct bpf_map *map)
 void bpf_sk_storage_free(struct sock *sk)
 {
 	struct bpf_local_storage *sk_storage;
-	bool free_sk_storage = false;
 
 	rcu_read_lock();
 	sk_storage = rcu_dereference(sk->sk_bpf_storage);
@@ -58,13 +57,8 @@ void bpf_sk_storage_free(struct sock *sk)
 		return;
 	}
 
-	raw_spin_lock_bh(&sk_storage->lock);
-	free_sk_storage = bpf_local_storage_unlink_nolock(sk_storage);
-	raw_spin_unlock_bh(&sk_storage->lock);
+	bpf_local_storage_destroy(sk_storage);
 	rcu_read_unlock();
-
-	if (free_sk_storage)
-		kfree_rcu(sk_storage, rcu);
 }
 
 static void bpf_sk_storage_map_free(struct bpf_map *map)
-- 
cgit v1.2.3-70-g09d2


From 62827d612ae525695799b3635a087cb49c55e977 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:22 -0800
Subject: bpf: Remove __bpf_local_storage_map_alloc

bpf_local_storage_map_alloc() is the only caller of
__bpf_local_storage_map_alloc().  The remaining logic in
bpf_local_storage_map_alloc() is only a one liner setting
the smap->cache_idx.

Remove __bpf_local_storage_map_alloc() to simplify code.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-4-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 63 +++++++++++++++++-------------------------
 1 file changed, 26 insertions(+), 37 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index e19f9f50a60d..f7234a8d4959 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -601,40 +601,6 @@ int bpf_local_storage_map_alloc_check(union bpf_attr *attr)
 	return 0;
 }
 
-static struct bpf_local_storage_map *__bpf_local_storage_map_alloc(union bpf_attr *attr)
-{
-	struct bpf_local_storage_map *smap;
-	unsigned int i;
-	u32 nbuckets;
-
-	smap = bpf_map_area_alloc(sizeof(*smap), NUMA_NO_NODE);
-	if (!smap)
-		return ERR_PTR(-ENOMEM);
-	bpf_map_init_from_attr(&smap->map, attr);
-
-	nbuckets = roundup_pow_of_two(num_possible_cpus());
-	/* Use at least 2 buckets, select_bucket() is undefined behavior with 1 bucket */
-	nbuckets = max_t(u32, 2, nbuckets);
-	smap->bucket_log = ilog2(nbuckets);
-
-	smap->buckets = bpf_map_kvcalloc(&smap->map, sizeof(*smap->buckets),
-					 nbuckets, GFP_USER | __GFP_NOWARN);
-	if (!smap->buckets) {
-		bpf_map_area_free(smap);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	for (i = 0; i < nbuckets; i++) {
-		INIT_HLIST_HEAD(&smap->buckets[i].list);
-		raw_spin_lock_init(&smap->buckets[i].lock);
-	}
-
-	smap->elem_size = offsetof(struct bpf_local_storage_elem,
-				   sdata.data[attr->value_size]);
-
-	return smap;
-}
-
 int bpf_local_storage_map_check_btf(const struct bpf_map *map,
 				    const struct btf *btf,
 				    const struct btf_type *key_type,
@@ -704,10 +670,33 @@ bpf_local_storage_map_alloc(union bpf_attr *attr,
 			    struct bpf_local_storage_cache *cache)
 {
 	struct bpf_local_storage_map *smap;
+	unsigned int i;
+	u32 nbuckets;
+
+	smap = bpf_map_area_alloc(sizeof(*smap), NUMA_NO_NODE);
+	if (!smap)
+		return ERR_PTR(-ENOMEM);
+	bpf_map_init_from_attr(&smap->map, attr);
+
+	nbuckets = roundup_pow_of_two(num_possible_cpus());
+	/* Use at least 2 buckets, select_bucket() is undefined behavior with 1 bucket */
+	nbuckets = max_t(u32, 2, nbuckets);
+	smap->bucket_log = ilog2(nbuckets);
 
-	smap = __bpf_local_storage_map_alloc(attr);
-	if (IS_ERR(smap))
-		return ERR_CAST(smap);
+	smap->buckets = bpf_map_kvcalloc(&smap->map, sizeof(*smap->buckets),
+					 nbuckets, GFP_USER | __GFP_NOWARN);
+	if (!smap->buckets) {
+		bpf_map_area_free(smap);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	for (i = 0; i < nbuckets; i++) {
+		INIT_HLIST_HEAD(&smap->buckets[i].list);
+		raw_spin_lock_init(&smap->buckets[i].lock);
+	}
+
+	smap->elem_size = offsetof(struct bpf_local_storage_elem,
+				   sdata.data[attr->value_size]);
 
 	smap->cache_idx = bpf_local_storage_cache_idx_get(cache);
 	return &smap->map;
-- 
cgit v1.2.3-70-g09d2


From 121f31f3e00dfc1acbca43f6f35779e050b56cfc Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:23 -0800
Subject: bpf: Remove the preceding __ from __bpf_selem_unlink_storage

__bpf_selem_unlink_storage is taking the spin lock and there is
no name collision also. Having the preceding '__' is confusing
when reviewing the later patch.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-5-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index f7234a8d4959..70df8dcb2066 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -216,8 +216,8 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	return free_local_storage;
 }
 
-static void __bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
-				       bool use_trace_rcu)
+static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
+				     bool use_trace_rcu)
 {
 	struct bpf_local_storage *local_storage;
 	bool free_local_storage = false;
@@ -288,7 +288,7 @@ void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu)
 	 * the local_storage.
 	 */
 	bpf_selem_unlink_map(selem);
-	__bpf_selem_unlink_storage(selem, use_trace_rcu);
+	bpf_selem_unlink_storage(selem, use_trace_rcu);
 }
 
 /* If cacheit_lockit is false, this lookup function is lockless */
-- 
cgit v1.2.3-70-g09d2


From fc6652aab6ad545de70b772550da9043d0b47f1c Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:24 -0800
Subject: bpf: Remember smap in bpf_local_storage

This patch remembers which smap triggers the allocation
of a 'struct bpf_local_storage' object. The local_storage is
allocated during the very first selem added to the owner.
The smap pointer is needed when using the bpf_mem_cache_free
in a later patch because it needs to free to the correct
smap's bpf_mem_alloc object.

When a selem is being removed, it needs to check if it is
the selem that triggers the creation of the local_storage.
If it is, the local_storage->smap pointer will be reset to NULL.
This NULL reset is done under the local_storage->lock in
bpf_selem_unlink_storage_nolock() when a selem is being removed.
Also note that the local_storage may not go away even
local_storage->smap is NULL because there may be other
selem still stored in the local_storage.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-6-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h | 1 +
 kernel/bpf/bpf_local_storage.c    | 4 ++++
 2 files changed, 5 insertions(+)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 5908a954ddc2..613b1805ed9f 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -83,6 +83,7 @@ struct bpf_local_storage_elem {
 
 struct bpf_local_storage {
 	struct bpf_local_storage_data __rcu *cache[BPF_LOCAL_STORAGE_CACHE_SIZE];
+	struct bpf_local_storage_map __rcu *smap;
 	struct hlist_head list; /* List of bpf_local_storage_elem */
 	void *owner;		/* The object that owns the above "list" of
 				 * bpf_local_storage_elem.
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 70df8dcb2066..5585dbfd9c66 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -213,6 +213,9 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 			kfree_rcu(selem, rcu);
 	}
 
+	if (rcu_access_pointer(local_storage->smap) == smap)
+		RCU_INIT_POINTER(local_storage->smap, NULL);
+
 	return free_local_storage;
 }
 
@@ -368,6 +371,7 @@ int bpf_local_storage_alloc(void *owner,
 		goto uncharge;
 	}
 
+	RCU_INIT_POINTER(storage->smap, smap);
 	INIT_HLIST_HEAD(&storage->list);
 	raw_spin_lock_init(&storage->lock);
 	storage->owner = owner;
-- 
cgit v1.2.3-70-g09d2


From a47eabf216f77cb6f22ceb38d46f1bb95968579c Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:25 -0800
Subject: bpf: Repurpose use_trace_rcu to reuse_now in bpf_local_storage

This patch re-purpose the use_trace_rcu to mean
if the freed memory can be reused immediately or not.
The use_trace_rcu is renamed to reuse_now. Other than
the boolean test is reversed, it should be a no-op.

The following explains the reason for the rename and how it will
be used in a later patch.

In a later patch, bpf_mem_cache_alloc/free will be used
in the bpf_local_storage. The bpf mem allocator will reuse
the freed memory immediately. Some of the free paths in
bpf_local_storage does not support memory to be reused immediately.
These paths are the "delete" elem cases from the bpf_*_storage_delete()
helper and the map_delete_elem() syscall. Note that "delete" elem
before the owner's (sk/task/cgrp/inode) lifetime ended is not
the common usage for the local storage.

The common free path, bpf_local_storage_destroy(), can reuse the
memory immediately. This common path means the storage stays with
its owner until the owner is destroyed.

The above mentioned "delete" elem paths that cannot
reuse immediately always has the 'use_trace_rcu ==  true'.
The cases that is safe for immediate reuse always have
'use_trace_rcu == false'. Instead of adding another arg
in a later patch, this patch re-purpose this arg
to reuse_now and have the test logic reversed.

In a later patch, 'reuse_now == true' will free to the
bpf_mem_cache_free() where the memory can be reused
immediately. 'reuse_now == false' will go through the
call_rcu_tasks_trace().

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-7-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h |  2 +-
 kernel/bpf/bpf_cgrp_storage.c     |  2 +-
 kernel/bpf/bpf_inode_storage.c    |  2 +-
 kernel/bpf/bpf_local_storage.c    | 24 ++++++++++++------------
 kernel/bpf/bpf_task_storage.c     |  2 +-
 net/core/bpf_sk_storage.c         |  2 +-
 6 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 613b1805ed9f..18a31add2255 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -143,7 +143,7 @@ int bpf_local_storage_map_check_btf(const struct bpf_map *map,
 void bpf_selem_link_storage_nolock(struct bpf_local_storage *local_storage,
 				   struct bpf_local_storage_elem *selem);
 
-void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu);
+void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now);
 
 void bpf_selem_link_map(struct bpf_local_storage_map *smap,
 			struct bpf_local_storage_elem *selem);
diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c
index 492594d69a86..c975cacdd16b 100644
--- a/kernel/bpf/bpf_cgrp_storage.c
+++ b/kernel/bpf/bpf_cgrp_storage.c
@@ -121,7 +121,7 @@ static int cgroup_storage_delete(struct cgroup *cgroup, struct bpf_map *map)
 	if (!sdata)
 		return -ENOENT;
 
-	bpf_selem_unlink(SELEM(sdata), true);
+	bpf_selem_unlink(SELEM(sdata), false);
 	return 0;
 }
 
diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c
index 2d25bcfa371b..ad2ab0187e45 100644
--- a/kernel/bpf/bpf_inode_storage.c
+++ b/kernel/bpf/bpf_inode_storage.c
@@ -122,7 +122,7 @@ static int inode_storage_delete(struct inode *inode, struct bpf_map *map)
 	if (!sdata)
 		return -ENOENT;
 
-	bpf_selem_unlink(SELEM(sdata), true);
+	bpf_selem_unlink(SELEM(sdata), false);
 
 	return 0;
 }
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 5585dbfd9c66..70c34a948c3c 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -147,7 +147,7 @@ static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
  */
 static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_storage,
 					    struct bpf_local_storage_elem *selem,
-					    bool uncharge_mem, bool use_trace_rcu)
+					    bool uncharge_mem, bool reuse_now)
 {
 	struct bpf_local_storage_map *smap;
 	bool free_local_storage;
@@ -201,7 +201,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	 * any special fields.
 	 */
 	rec = smap->map.record;
-	if (use_trace_rcu) {
+	if (!reuse_now) {
 		if (!IS_ERR_OR_NULL(rec))
 			call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_fields_trace_rcu);
 		else
@@ -220,7 +220,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 }
 
 static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
-				     bool use_trace_rcu)
+				     bool reuse_now)
 {
 	struct bpf_local_storage *local_storage;
 	bool free_local_storage = false;
@@ -235,11 +235,11 @@ static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 	raw_spin_lock_irqsave(&local_storage->lock, flags);
 	if (likely(selem_linked_to_storage(selem)))
 		free_local_storage = bpf_selem_unlink_storage_nolock(
-			local_storage, selem, true, use_trace_rcu);
+			local_storage, selem, true, reuse_now);
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
 	if (free_local_storage) {
-		if (use_trace_rcu)
+		if (!reuse_now)
 			call_rcu_tasks_trace(&local_storage->rcu,
 				     bpf_local_storage_free_rcu);
 		else
@@ -284,14 +284,14 @@ void bpf_selem_link_map(struct bpf_local_storage_map *smap,
 	raw_spin_unlock_irqrestore(&b->lock, flags);
 }
 
-void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu)
+void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now)
 {
 	/* Always unlink from map before unlinking from local_storage
 	 * because selem will be freed after successfully unlinked from
 	 * the local_storage.
 	 */
 	bpf_selem_unlink_map(selem);
-	bpf_selem_unlink_storage(selem, use_trace_rcu);
+	bpf_selem_unlink_storage(selem, reuse_now);
 }
 
 /* If cacheit_lockit is false, this lookup function is lockless */
@@ -538,7 +538,7 @@ bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap,
 	if (old_sdata) {
 		bpf_selem_unlink_map(SELEM(old_sdata));
 		bpf_selem_unlink_storage_nolock(local_storage, SELEM(old_sdata),
-						false, true);
+						false, false);
 	}
 
 unlock:
@@ -651,7 +651,7 @@ void bpf_local_storage_destroy(struct bpf_local_storage *local_storage)
 		 * of the loop will set the free_cgroup_storage to true.
 		 */
 		free_storage = bpf_selem_unlink_storage_nolock(
-			local_storage, selem, false, false);
+			local_storage, selem, false, true);
 	}
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
@@ -745,7 +745,7 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 				migrate_disable();
 				this_cpu_inc(*busy_counter);
 			}
-			bpf_selem_unlink(selem, false);
+			bpf_selem_unlink(selem, true);
 			if (busy_counter) {
 				this_cpu_dec(*busy_counter);
 				migrate_enable();
@@ -783,8 +783,8 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 		/* We cannot skip rcu_barrier() when rcu_trace_implies_rcu_gp()
 		 * is true, because while call_rcu invocation is skipped in that
 		 * case in bpf_selem_free_fields_trace_rcu (and all local
-		 * storage maps pass use_trace_rcu = true), there can be
-		 * call_rcu callbacks based on use_trace_rcu = false in the
+		 * storage maps pass reuse_now = false), there can be
+		 * call_rcu callbacks based on reuse_now = true in the
 		 * while ((selem = ...)) loop above or when owner's free path
 		 * calls bpf_local_storage_unlink_nolock.
 		 */
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index 4dcef28744d1..c88cc04c17c1 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -168,7 +168,7 @@ static int task_storage_delete(struct task_struct *task, struct bpf_map *map,
 	if (!nobusy)
 		return -EBUSY;
 
-	bpf_selem_unlink(SELEM(sdata), true);
+	bpf_selem_unlink(SELEM(sdata), false);
 
 	return 0;
 }
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index 8f56438c104b..a5f185b8e50a 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -40,7 +40,7 @@ static int bpf_sk_storage_del(struct sock *sk, struct bpf_map *map)
 	if (!sdata)
 		return -ENOENT;
 
-	bpf_selem_unlink(SELEM(sdata), true);
+	bpf_selem_unlink(SELEM(sdata), false);
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From c609981342dca634e5dea8c6ca175b6533581261 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:26 -0800
Subject: bpf: Remove bpf_selem_free_fields*_rcu

This patch removes the bpf_selem_free_fields*_rcu. The
bpf_obj_free_fields() can be done before the call_rcu_trasks_trace()
and kfree_rcu(). It is needed when a later patch uses
bpf_mem_cache_alloc/free. In bpf hashtab, bpf_obj_free_fields()
is also called before calling bpf_mem_cache_free. The discussion
can be found in
https://lore.kernel.org/bpf/f67021ee-21d9-bfae-6134-4ca542fab843@linux.dev/

Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-8-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 67 ++++--------------------------------------
 1 file changed, 5 insertions(+), 62 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 70c34a948c3c..715deaaefe13 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -109,27 +109,6 @@ static void bpf_local_storage_free_rcu(struct rcu_head *rcu)
 		kfree_rcu(local_storage, rcu);
 }
 
-static void bpf_selem_free_fields_rcu(struct rcu_head *rcu)
-{
-	struct bpf_local_storage_elem *selem;
-	struct bpf_local_storage_map *smap;
-
-	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
-	/* protected by the rcu_barrier*() */
-	smap = rcu_dereference_protected(SDATA(selem)->smap, true);
-	bpf_obj_free_fields(smap->map.record, SDATA(selem)->data);
-	kfree(selem);
-}
-
-static void bpf_selem_free_fields_trace_rcu(struct rcu_head *rcu)
-{
-	/* Free directly if Tasks Trace RCU GP also implies RCU GP */
-	if (rcu_trace_implies_rcu_gp())
-		bpf_selem_free_fields_rcu(rcu);
-	else
-		call_rcu(rcu, bpf_selem_free_fields_rcu);
-}
-
 static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage_elem *selem;
@@ -151,7 +130,6 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 {
 	struct bpf_local_storage_map *smap;
 	bool free_local_storage;
-	struct btf_record *rec;
 	void *owner;
 
 	smap = rcu_dereference_check(SDATA(selem)->smap, bpf_rcu_lock_held());
@@ -192,26 +170,11 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	    SDATA(selem))
 		RCU_INIT_POINTER(local_storage->cache[smap->cache_idx], NULL);
 
-	/* A different RCU callback is chosen whenever we need to free
-	 * additional fields in selem data before freeing selem.
-	 * bpf_local_storage_map_free only executes rcu_barrier to wait for RCU
-	 * callbacks when it has special fields, hence we can only conditionally
-	 * dereference smap, as by this time the map might have already been
-	 * freed without waiting for our call_rcu callback if it did not have
-	 * any special fields.
-	 */
-	rec = smap->map.record;
-	if (!reuse_now) {
-		if (!IS_ERR_OR_NULL(rec))
-			call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_fields_trace_rcu);
-		else
-			call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
-	} else {
-		if (!IS_ERR_OR_NULL(rec))
-			call_rcu(&selem->rcu, bpf_selem_free_fields_rcu);
-		else
-			kfree_rcu(selem, rcu);
-	}
+	bpf_obj_free_fields(smap->map.record, SDATA(selem)->data);
+	if (!reuse_now)
+		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
+	else
+		kfree_rcu(selem, rcu);
 
 	if (rcu_access_pointer(local_storage->smap) == smap)
 		RCU_INIT_POINTER(local_storage->smap, NULL);
@@ -769,26 +732,6 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 	 */
 	synchronize_rcu();
 
-	/* Only delay freeing of smap, buckets are not needed anymore */
 	kvfree(smap->buckets);
-
-	/* When local storage has special fields, callbacks for
-	 * bpf_selem_free_fields_rcu and bpf_selem_free_fields_trace_rcu will
-	 * keep using the map BTF record, we need to execute an RCU barrier to
-	 * wait for them as the record will be freed right after our map_free
-	 * callback.
-	 */
-	if (!IS_ERR_OR_NULL(smap->map.record)) {
-		rcu_barrier_tasks_trace();
-		/* We cannot skip rcu_barrier() when rcu_trace_implies_rcu_gp()
-		 * is true, because while call_rcu invocation is skipped in that
-		 * case in bpf_selem_free_fields_trace_rcu (and all local
-		 * storage maps pass reuse_now = false), there can be
-		 * call_rcu callbacks based on reuse_now = true in the
-		 * while ((selem = ...)) loop above or when owner's free path
-		 * calls bpf_local_storage_unlink_nolock.
-		 */
-		rcu_barrier();
-	}
 	bpf_map_area_free(smap);
 }
-- 
cgit v1.2.3-70-g09d2


From f8ccf30c179ec1ac16654f6e6ceb40cce1530b91 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:27 -0800
Subject: bpf: Add bpf_selem_free_rcu callback

Add bpf_selem_free_rcu() callback to do the kfree() instead
of using kfree_rcu. It is a preparation work for using
bpf_mem_cache_alloc/free in a later patch.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-9-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 715deaaefe13..146e9caeda96 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -109,15 +109,20 @@ static void bpf_local_storage_free_rcu(struct rcu_head *rcu)
 		kfree_rcu(local_storage, rcu);
 }
 
-static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
+static void bpf_selem_free_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage_elem *selem;
 
 	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
+	kfree(selem);
+}
+
+static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
+{
 	if (rcu_trace_implies_rcu_gp())
-		kfree(selem);
+		bpf_selem_free_rcu(rcu);
 	else
-		kfree_rcu(selem, rcu);
+		call_rcu(rcu, bpf_selem_free_rcu);
 }
 
 /* local_storage->lock must be held and selem->local_storage == local_storage.
@@ -174,7 +179,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	if (!reuse_now)
 		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
 	else
-		kfree_rcu(selem, rcu);
+		call_rcu(&selem->rcu, bpf_selem_free_rcu);
 
 	if (rcu_access_pointer(local_storage->smap) == smap)
 		RCU_INIT_POINTER(local_storage->smap, NULL);
-- 
cgit v1.2.3-70-g09d2


From c0d63f309186d8492577c67c67984c714b6b72bc Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:28 -0800
Subject: bpf: Add bpf_selem_free()

This patch refactors the selem freeing logic into bpf_selem_free().
It is a preparation work for a later patch using
bpf_mem_cache_alloc/free. The other kfree(selem) cases
are also changed to bpf_selem_free(..., reuse_now = true).

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-10-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h |  4 ++++
 kernel/bpf/bpf_local_storage.c    | 21 ++++++++++++++-------
 net/core/bpf_sk_storage.c         |  2 +-
 3 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 18a31add2255..a34f61467a2f 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -152,6 +152,10 @@ struct bpf_local_storage_elem *
 bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner, void *value,
 		bool charge_mem, gfp_t gfp_flags);
 
+void bpf_selem_free(struct bpf_local_storage_elem *selem,
+		    struct bpf_local_storage_map *smap,
+		    bool reuse_now);
+
 int
 bpf_local_storage_alloc(void *owner,
 			struct bpf_local_storage_map *smap,
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 146e9caeda96..512943aac435 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -125,6 +125,17 @@ static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
 		call_rcu(rcu, bpf_selem_free_rcu);
 }
 
+void bpf_selem_free(struct bpf_local_storage_elem *selem,
+		    struct bpf_local_storage_map *smap,
+		    bool reuse_now)
+{
+	bpf_obj_free_fields(smap->map.record, SDATA(selem)->data);
+	if (!reuse_now)
+		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
+	else
+		call_rcu(&selem->rcu, bpf_selem_free_rcu);
+}
+
 /* local_storage->lock must be held and selem->local_storage == local_storage.
  * The caller must ensure selem->smap is still valid to be
  * dereferenced for its smap->elem_size and smap->cache_idx.
@@ -175,11 +186,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	    SDATA(selem))
 		RCU_INIT_POINTER(local_storage->cache[smap->cache_idx], NULL);
 
-	bpf_obj_free_fields(smap->map.record, SDATA(selem)->data);
-	if (!reuse_now)
-		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
-	else
-		call_rcu(&selem->rcu, bpf_selem_free_rcu);
+	bpf_selem_free(selem, smap, reuse_now);
 
 	if (rcu_access_pointer(local_storage->smap) == smap)
 		RCU_INIT_POINTER(local_storage->smap, NULL);
@@ -423,7 +430,7 @@ bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap,
 
 		err = bpf_local_storage_alloc(owner, smap, selem, gfp_flags);
 		if (err) {
-			kfree(selem);
+			bpf_selem_free(selem, smap, true);
 			mem_uncharge(smap, owner, smap->elem_size);
 			return ERR_PTR(err);
 		}
@@ -517,7 +524,7 @@ unlock_err:
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 	if (selem) {
 		mem_uncharge(smap, owner, smap->elem_size);
-		kfree(selem);
+		bpf_selem_free(selem, smap, true);
 	}
 	return ERR_PTR(err);
 }
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index a5f185b8e50a..24c3dc0d62e5 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -197,7 +197,7 @@ int bpf_sk_storage_clone(const struct sock *sk, struct sock *newsk)
 		} else {
 			ret = bpf_local_storage_alloc(newsk, smap, copy_selem, GFP_ATOMIC);
 			if (ret) {
-				kfree(copy_selem);
+				bpf_selem_free(copy_selem, smap, true);
 				atomic_sub(smap->elem_size,
 					   &newsk->sk_omem_alloc);
 				bpf_map_put(map);
-- 
cgit v1.2.3-70-g09d2


From 1288aaa2786b1e58c9e88e53f7654d520ebe0f3b Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:29 -0800
Subject: bpf: Add bpf_local_storage_rcu callback

The existing bpf_local_storage_free_rcu is renamed to
bpf_local_storage_free_trace_rcu. A new bpf_local_storage_rcu
callback is added to do the kfree instead of using kfree_rcu.
It is a preparation work for a later patch using
bpf_mem_cache_alloc/free.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-11-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 512943aac435..0fbc477c8b27 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -99,14 +99,19 @@ static void bpf_local_storage_free_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage *local_storage;
 
+	local_storage = container_of(rcu, struct bpf_local_storage, rcu);
+	kfree(local_storage);
+}
+
+static void bpf_local_storage_free_trace_rcu(struct rcu_head *rcu)
+{
 	/* If RCU Tasks Trace grace period implies RCU grace period, do
 	 * kfree(), else do kfree_rcu().
 	 */
-	local_storage = container_of(rcu, struct bpf_local_storage, rcu);
 	if (rcu_trace_implies_rcu_gp())
-		kfree(local_storage);
+		bpf_local_storage_free_rcu(rcu);
 	else
-		kfree_rcu(local_storage, rcu);
+		call_rcu(rcu, bpf_local_storage_free_rcu);
 }
 
 static void bpf_selem_free_rcu(struct rcu_head *rcu)
@@ -216,9 +221,9 @@ static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 	if (free_local_storage) {
 		if (!reuse_now)
 			call_rcu_tasks_trace(&local_storage->rcu,
-				     bpf_local_storage_free_rcu);
+				     bpf_local_storage_free_trace_rcu);
 		else
-			kfree_rcu(local_storage, rcu);
+			call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
 	}
 }
 
@@ -631,7 +636,7 @@ void bpf_local_storage_destroy(struct bpf_local_storage *local_storage)
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
 	if (free_storage)
-		kfree_rcu(local_storage, rcu);
+		call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
 }
 
 u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map)
-- 
cgit v1.2.3-70-g09d2


From 7e30a8477b0bdd13dfd0b24e4f32b26d22b96e6c Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:30 -0800
Subject: bpf: Add bpf_local_storage_free()

This patch refactors local_storage freeing logic into
bpf_local_storage_free(). It is a preparation work for a later
patch that uses bpf_mem_cache_alloc/free. The other kfree(local_storage)
cases are also changed to bpf_local_storage_free(..., reuse_now = true).

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-12-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 0fbc477c8b27..351d991694cb 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -114,6 +114,16 @@ static void bpf_local_storage_free_trace_rcu(struct rcu_head *rcu)
 		call_rcu(rcu, bpf_local_storage_free_rcu);
 }
 
+static void bpf_local_storage_free(struct bpf_local_storage *local_storage,
+				   bool reuse_now)
+{
+	if (!reuse_now)
+		call_rcu_tasks_trace(&local_storage->rcu,
+				     bpf_local_storage_free_trace_rcu);
+	else
+		call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
+}
+
 static void bpf_selem_free_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage_elem *selem;
@@ -218,13 +228,8 @@ static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 			local_storage, selem, true, reuse_now);
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
-	if (free_local_storage) {
-		if (!reuse_now)
-			call_rcu_tasks_trace(&local_storage->rcu,
-				     bpf_local_storage_free_trace_rcu);
-		else
-			call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
-	}
+	if (free_local_storage)
+		bpf_local_storage_free(local_storage, reuse_now);
 }
 
 void bpf_selem_link_storage_nolock(struct bpf_local_storage *local_storage,
@@ -391,7 +396,7 @@ int bpf_local_storage_alloc(void *owner,
 	return 0;
 
 uncharge:
-	kfree(storage);
+	bpf_local_storage_free(storage, true);
 	mem_uncharge(smap, owner, sizeof(*storage));
 	return err;
 }
@@ -636,7 +641,7 @@ void bpf_local_storage_destroy(struct bpf_local_storage *local_storage)
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
 	if (free_storage)
-		call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
+		bpf_local_storage_free(local_storage, true);
 }
 
 u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map)
-- 
cgit v1.2.3-70-g09d2


From 57ef77152b58770cbd54d624babd8f5d90805ea7 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:34 -0800
Subject: selftests/bpf: Replace CHECK with ASSERT in test_local_storage

This patch migrates the CHECK macro to ASSERT macro.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-16-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/test_local_storage.c  | 47 ++++++++--------------
 1 file changed, 17 insertions(+), 30 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_local_storage.c b/tools/testing/selftests/bpf/prog_tests/test_local_storage.c
index 9c77cd6b1eaf..563a9c746b7b 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_local_storage.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_local_storage.c
@@ -13,8 +13,6 @@
 #include "network_helpers.h"
 #include "task_local_storage_helpers.h"
 
-static unsigned int duration;
-
 #define TEST_STORAGE_VALUE 0xbeefdead
 
 struct storage {
@@ -60,36 +58,30 @@ static bool check_syscall_operations(int map_fd, int obj_fd)
 
 	/* Looking up an existing element should fail initially */
 	err = bpf_map_lookup_elem_flags(map_fd, &obj_fd, &lookup_val, 0);
-	if (CHECK(!err || errno != ENOENT, "bpf_map_lookup_elem",
-		  "err:%d errno:%d\n", err, errno))
+	if (!ASSERT_EQ(err, -ENOENT, "bpf_map_lookup_elem"))
 		return false;
 
 	/* Create a new element */
 	err = bpf_map_update_elem(map_fd, &obj_fd, &val, BPF_NOEXIST);
-	if (CHECK(err < 0, "bpf_map_update_elem", "err:%d errno:%d\n", err,
-		  errno))
+	if (!ASSERT_OK(err, "bpf_map_update_elem"))
 		return false;
 
 	/* Lookup the newly created element */
 	err = bpf_map_lookup_elem_flags(map_fd, &obj_fd, &lookup_val, 0);
-	if (CHECK(err < 0, "bpf_map_lookup_elem", "err:%d errno:%d", err,
-		  errno))
+	if (!ASSERT_OK(err, "bpf_map_lookup_elem"))
 		return false;
 
 	/* Check the value of the newly created element */
-	if (CHECK(lookup_val.value != val.value, "bpf_map_lookup_elem",
-		  "value got = %x errno:%d", lookup_val.value, val.value))
+	if (!ASSERT_EQ(lookup_val.value, val.value, "bpf_map_lookup_elem"))
 		return false;
 
 	err = bpf_map_delete_elem(map_fd, &obj_fd);
-	if (CHECK(err, "bpf_map_delete_elem()", "err:%d errno:%d\n", err,
-		  errno))
+	if (!ASSERT_OK(err, "bpf_map_delete_elem()"))
 		return false;
 
 	/* The lookup should fail, now that the element has been deleted */
 	err = bpf_map_lookup_elem_flags(map_fd, &obj_fd, &lookup_val, 0);
-	if (CHECK(!err || errno != ENOENT, "bpf_map_lookup_elem",
-		  "err:%d errno:%d\n", err, errno))
+	if (!ASSERT_EQ(err, -ENOENT, "bpf_map_lookup_elem"))
 		return false;
 
 	return true;
@@ -104,35 +96,32 @@ void test_test_local_storage(void)
 	char cmd[256];
 
 	skel = local_storage__open_and_load();
-	if (CHECK(!skel, "skel_load", "lsm skeleton failed\n"))
+	if (!ASSERT_OK_PTR(skel, "skel_load"))
 		goto close_prog;
 
 	err = local_storage__attach(skel);
-	if (CHECK(err, "attach", "lsm attach failed: %d\n", err))
+	if (!ASSERT_OK(err, "attach"))
 		goto close_prog;
 
 	task_fd = sys_pidfd_open(getpid(), 0);
-	if (CHECK(task_fd < 0, "pidfd_open",
-		  "failed to get pidfd err:%d, errno:%d", task_fd, errno))
+	if (!ASSERT_GE(task_fd, 0, "pidfd_open"))
 		goto close_prog;
 
 	if (!check_syscall_operations(bpf_map__fd(skel->maps.task_storage_map),
 				      task_fd))
 		goto close_prog;
 
-	if (CHECK(!mkdtemp(tmp_dir_path), "mkdtemp",
-		  "unable to create tmpdir: %d\n", errno))
+	if (!ASSERT_OK_PTR(mkdtemp(tmp_dir_path), "mkdtemp"))
 		goto close_prog;
 
 	snprintf(tmp_exec_path, sizeof(tmp_exec_path), "%s/copy_of_rm",
 		 tmp_dir_path);
 	snprintf(cmd, sizeof(cmd), "cp /bin/rm %s", tmp_exec_path);
-	if (CHECK_FAIL(system(cmd)))
+	if (!ASSERT_OK(system(cmd), "system(cp)"))
 		goto close_prog_rmdir;
 
 	rm_fd = open(tmp_exec_path, O_RDONLY);
-	if (CHECK(rm_fd < 0, "open", "failed to open %s err:%d, errno:%d",
-		  tmp_exec_path, rm_fd, errno))
+	if (!ASSERT_GE(rm_fd, 0, "open(tmp_exec_path)"))
 		goto close_prog_rmdir;
 
 	if (!check_syscall_operations(bpf_map__fd(skel->maps.inode_storage_map),
@@ -145,7 +134,7 @@ void test_test_local_storage(void)
 	 * LSM program.
 	 */
 	err = run_self_unlink(&skel->bss->monitored_pid, tmp_exec_path);
-	if (CHECK(err != EPERM, "run_self_unlink", "err %d want EPERM\n", err))
+	if (!ASSERT_EQ(err, EPERM, "run_self_unlink"))
 		goto close_prog_rmdir;
 
 	/* Set the process being monitored to be the current process */
@@ -156,18 +145,16 @@ void test_test_local_storage(void)
 	 */
 	snprintf(cmd, sizeof(cmd), "mv %s/copy_of_rm %s/check_null_ptr",
 		 tmp_dir_path, tmp_dir_path);
-	if (CHECK_FAIL(system(cmd)))
+	if (!ASSERT_OK(system(cmd), "system(mv)"))
 		goto close_prog_rmdir;
 
-	CHECK(skel->data->inode_storage_result != 0, "inode_storage_result",
-	      "inode_local_storage not set\n");
+	ASSERT_EQ(skel->data->inode_storage_result, 0, "inode_storage_result");
 
 	serv_sk = start_server(AF_INET6, SOCK_STREAM, NULL, 0, 0);
-	if (CHECK(serv_sk < 0, "start_server", "failed to start server\n"))
+	if (!ASSERT_GE(serv_sk, 0, "start_server"))
 		goto close_prog_rmdir;
 
-	CHECK(skel->data->sk_storage_result != 0, "sk_storage_result",
-	      "sk_local_storage not set\n");
+	ASSERT_EQ(skel->data->sk_storage_result, 0, "sk_storage_result");
 
 	if (!check_syscall_operations(bpf_map__fd(skel->maps.sk_storage_map),
 				      serv_sk))
-- 
cgit v1.2.3-70-g09d2


From 1f443d0f2b5702bad5f03aab544858ddd33999b7 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:35 -0800
Subject: selftests/bpf: Check freeing sk->sk_local_storage with
 sk_local_storage->smap is NULL

This patch tweats the socket_bind bpf prog to test the
local_storage->smap == NULL case in the bpf_local_storage_free()
code path. The idea is to create the local_storage with
the sk_storage_map's selem first. Then add the sk_storage_map2's selem
and then delete the earlier sk_storeage_map's selem.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-17-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/local_storage.c | 29 ++++++++++++++++-------
 1 file changed, 20 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/local_storage.c b/tools/testing/selftests/bpf/progs/local_storage.c
index 01c74bc870ae..c8ba7207f5a5 100644
--- a/tools/testing/selftests/bpf/progs/local_storage.c
+++ b/tools/testing/selftests/bpf/progs/local_storage.c
@@ -108,18 +108,17 @@ int BPF_PROG(socket_bind, struct socket *sock, struct sockaddr *address,
 {
 	__u32 pid = bpf_get_current_pid_tgid() >> 32;
 	struct local_storage *storage;
-	int err;
 
 	if (pid != monitored_pid)
 		return 0;
 
-	storage = bpf_sk_storage_get(&sk_storage_map, sock->sk, 0,
-				     BPF_LOCAL_STORAGE_GET_F_CREATE);
+	storage = bpf_sk_storage_get(&sk_storage_map, sock->sk, 0, 0);
 	if (!storage)
 		return 0;
 
+	sk_storage_result = -1;
 	if (storage->value != DUMMY_STORAGE_VALUE)
-		sk_storage_result = -1;
+		return 0;
 
 	/* This tests that we can associate multiple elements
 	 * with the local storage.
@@ -129,14 +128,26 @@ int BPF_PROG(socket_bind, struct socket *sock, struct sockaddr *address,
 	if (!storage)
 		return 0;
 
-	err = bpf_sk_storage_delete(&sk_storage_map, sock->sk);
-	if (err)
+	if (bpf_sk_storage_delete(&sk_storage_map2, sock->sk))
 		return 0;
 
-	err = bpf_sk_storage_delete(&sk_storage_map2, sock->sk);
-	if (!err)
-		sk_storage_result = err;
+	storage = bpf_sk_storage_get(&sk_storage_map2, sock->sk, 0,
+				     BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (!storage)
+		return 0;
+
+	if (bpf_sk_storage_delete(&sk_storage_map, sock->sk))
+		return 0;
+
+	/* Ensure that the sk_storage_map is disconnected from the storage.
+	 * The storage memory should not be freed back to the
+	 * bpf_mem_alloc of the sk_bpf_storage_map because
+	 * sk_bpf_storage_map may have been gone.
+	 */
+	if (!sock->sk->sk_bpf_storage || sock->sk->sk_bpf_storage->smap)
+		return 0;
 
+	sk_storage_result = 0;
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 4659fba121dac21a3516986a3c2cf8459c7ac3bc Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Tue, 7 Mar 2023 22:59:36 -0800
Subject: selftests/bpf: Add local-storage-create benchmark
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch tests how many kmallocs is needed to create and free
a batch of UDP sockets and each socket has a 64bytes bpf storage.
It also measures how fast the UDP sockets can be created.

The result is from my qemu setup.

Before bpf_mem_cache_alloc/free:
./bench -p 1 local-storage-create
Setting up benchmark 'local-storage-create'...
Benchmark 'local-storage-create' started.
Iter   0 ( 73.193us): creates  213.552k/s (213.552k/prod), 3.09 kmallocs/create
Iter   1 (-20.724us): creates  211.908k/s (211.908k/prod), 3.09 kmallocs/create
Iter   2 (  9.280us): creates  212.574k/s (212.574k/prod), 3.12 kmallocs/create
Iter   3 ( 11.039us): creates  213.209k/s (213.209k/prod), 3.12 kmallocs/create
Iter   4 (-11.411us): creates  213.351k/s (213.351k/prod), 3.12 kmallocs/create
Iter   5 ( -7.915us): creates  214.754k/s (214.754k/prod), 3.12 kmallocs/create
Iter   6 ( 11.317us): creates  210.942k/s (210.942k/prod), 3.12 kmallocs/create
Summary: creates  212.789 ± 1.310k/s (212.789k/prod), 3.12 kmallocs/create

After bpf_mem_cache_alloc/free:
./bench -p 1 local-storage-create
Setting up benchmark 'local-storage-create'...
Benchmark 'local-storage-create' started.
Iter   0 ( 68.265us): creates  243.984k/s (243.984k/prod), 1.04 kmallocs/create
Iter   1 ( 30.357us): creates  238.424k/s (238.424k/prod), 1.04 kmallocs/create
Iter   2 (-18.712us): creates  232.963k/s (232.963k/prod), 1.04 kmallocs/create
Iter   3 (-15.885us): creates  238.879k/s (238.879k/prod), 1.04 kmallocs/create
Iter   4 (  5.590us): creates  237.490k/s (237.490k/prod), 1.04 kmallocs/create
Iter   5 (  8.577us): creates  237.521k/s (237.521k/prod), 1.04 kmallocs/create
Iter   6 ( -6.263us): creates  238.508k/s (238.508k/prod), 1.04 kmallocs/create
Summary: creates  237.298 ± 2.198k/s (237.298k/prod), 1.04 kmallocs/create

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230308065936.1550103-18-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/Makefile               |   2 +
 tools/testing/selftests/bpf/bench.c                |   2 +
 .../bpf/benchs/bench_local_storage_create.c        | 141 +++++++++++++++++++++
 .../bpf/progs/bench_local_storage_create.c         |  57 +++++++++
 4 files changed, 202 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
 create mode 100644 tools/testing/selftests/bpf/progs/bench_local_storage_create.c

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 606e2d738dd8..55811c448eb7 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -639,6 +639,7 @@ $(OUTPUT)/bench_strncmp.o: $(OUTPUT)/strncmp_bench.skel.h
 $(OUTPUT)/bench_bpf_hashmap_full_update.o: $(OUTPUT)/bpf_hashmap_full_update_bench.skel.h
 $(OUTPUT)/bench_local_storage.o: $(OUTPUT)/local_storage_bench.skel.h
 $(OUTPUT)/bench_local_storage_rcu_tasks_trace.o: $(OUTPUT)/local_storage_rcu_tasks_trace_bench.skel.h
+$(OUTPUT)/bench_local_storage_create.o: $(OUTPUT)/bench_local_storage_create.skel.h
 $(OUTPUT)/bench_bpf_hashmap_lookup.o: $(OUTPUT)/bpf_hashmap_lookup.skel.h
 $(OUTPUT)/bench.o: bench.h testing_helpers.h $(BPFOBJ)
 $(OUTPUT)/bench: LDLIBS += -lm
@@ -656,6 +657,7 @@ $(OUTPUT)/bench: $(OUTPUT)/bench.o \
 		 $(OUTPUT)/bench_local_storage.o \
 		 $(OUTPUT)/bench_local_storage_rcu_tasks_trace.o \
 		 $(OUTPUT)/bench_bpf_hashmap_lookup.o \
+		 $(OUTPUT)/bench_local_storage_create.o \
 		 #
 	$(call msg,BINARY,,$@)
 	$(Q)$(CC) $(CFLAGS) $(LDFLAGS) $(filter %.a %.o,$^) $(LDLIBS) -o $@
diff --git a/tools/testing/selftests/bpf/bench.c b/tools/testing/selftests/bpf/bench.c
index 0b2a53bb8460..dc3827c1f139 100644
--- a/tools/testing/selftests/bpf/bench.c
+++ b/tools/testing/selftests/bpf/bench.c
@@ -515,6 +515,7 @@ extern const struct bench bench_local_storage_cache_interleaved_get;
 extern const struct bench bench_local_storage_cache_hashmap_control;
 extern const struct bench bench_local_storage_tasks_trace;
 extern const struct bench bench_bpf_hashmap_lookup;
+extern const struct bench bench_local_storage_create;
 
 static const struct bench *benchs[] = {
 	&bench_count_global,
@@ -555,6 +556,7 @@ static const struct bench *benchs[] = {
 	&bench_local_storage_cache_hashmap_control,
 	&bench_local_storage_tasks_trace,
 	&bench_bpf_hashmap_lookup,
+	&bench_local_storage_create,
 };
 
 static void find_benchmark(void)
diff --git a/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c b/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
new file mode 100644
index 000000000000..f8b2a640ccbe
--- /dev/null
+++ b/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
@@ -0,0 +1,141 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <sys/types.h>
+#include <sys/socket.h>
+
+#include "bench.h"
+#include "bench_local_storage_create.skel.h"
+
+#define BATCH_SZ 32
+
+struct thread {
+	int fds[BATCH_SZ];
+};
+
+static struct bench_local_storage_create *skel;
+static struct thread *threads;
+static long socket_errs;
+
+static void validate(void)
+{
+	if (env.consumer_cnt > 1) {
+		fprintf(stderr,
+			"local-storage-create benchmark does not need consumer\n");
+		exit(1);
+	}
+}
+
+static void setup(void)
+{
+	skel = bench_local_storage_create__open_and_load();
+	if (!skel) {
+		fprintf(stderr, "error loading skel\n");
+		exit(1);
+	}
+
+	skel->bss->bench_pid = getpid();
+
+	if (!bpf_program__attach(skel->progs.socket_post_create)) {
+		fprintf(stderr, "Error attaching bpf program\n");
+		exit(1);
+	}
+
+	if (!bpf_program__attach(skel->progs.kmalloc)) {
+		fprintf(stderr, "Error attaching bpf program\n");
+		exit(1);
+	}
+
+	threads = calloc(env.producer_cnt, sizeof(*threads));
+
+	if (!threads) {
+		fprintf(stderr, "cannot alloc thread_res\n");
+		exit(1);
+	}
+}
+
+static void measure(struct bench_res *res)
+{
+	res->hits = atomic_swap(&skel->bss->create_cnts, 0);
+	res->drops = atomic_swap(&skel->bss->kmalloc_cnts, 0);
+}
+
+static void *consumer(void *input)
+{
+	return NULL;
+}
+
+static void *producer(void *input)
+{
+	struct thread *t = &threads[(long)(input)];
+	int *fds = t->fds;
+	int i;
+
+	while (true) {
+		for (i = 0; i < BATCH_SZ; i++) {
+			fds[i] = socket(AF_INET6, SOCK_DGRAM, 0);
+			if (fds[i] == -1)
+				atomic_inc(&socket_errs);
+		}
+
+		for (i = 0; i < BATCH_SZ; i++) {
+			if (fds[i] != -1)
+				close(fds[i]);
+		}
+	}
+
+	return NULL;
+}
+
+static void report_progress(int iter, struct bench_res *res, long delta_ns)
+{
+	double creates_per_sec, kmallocs_per_create;
+
+	creates_per_sec = res->hits / 1000.0 / (delta_ns / 1000000000.0);
+	kmallocs_per_create = (double)res->drops / res->hits;
+
+	printf("Iter %3d (%7.3lfus): ",
+	       iter, (delta_ns - 1000000000) / 1000.0);
+	printf("creates %8.3lfk/s (%7.3lfk/prod), ",
+	       creates_per_sec, creates_per_sec / env.producer_cnt);
+	printf("%3.2lf kmallocs/create\n", kmallocs_per_create);
+}
+
+static void report_final(struct bench_res res[], int res_cnt)
+{
+	double creates_mean = 0.0, creates_stddev = 0.0;
+	long total_creates = 0, total_kmallocs = 0;
+	int i;
+
+	for (i = 0; i < res_cnt; i++) {
+		creates_mean += res[i].hits / 1000.0 / (0.0 + res_cnt);
+		total_creates += res[i].hits;
+		total_kmallocs += res[i].drops;
+	}
+
+	if (res_cnt > 1)  {
+		for (i = 0; i < res_cnt; i++)
+			creates_stddev += (creates_mean - res[i].hits / 1000.0) *
+				       (creates_mean - res[i].hits / 1000.0) /
+				       (res_cnt - 1.0);
+		creates_stddev = sqrt(creates_stddev);
+	}
+	printf("Summary: creates %8.3lf \u00B1 %5.3lfk/s (%7.3lfk/prod), ",
+	       creates_mean, creates_stddev, creates_mean / env.producer_cnt);
+	printf("%4.2lf kmallocs/create\n", (double)total_kmallocs / total_creates);
+	if (socket_errs || skel->bss->create_errs)
+		printf("socket() errors %ld create_errs %ld\n", socket_errs,
+		       skel->bss->create_errs);
+}
+
+/* Benchmark performance of creating bpf local storage  */
+const struct bench bench_local_storage_create = {
+	.name = "local-storage-create",
+	.validate = validate,
+	.setup = setup,
+	.producer_thread = producer,
+	.consumer_thread = consumer,
+	.measure = measure,
+	.report_progress = report_progress,
+	.report_final = report_final,
+};
diff --git a/tools/testing/selftests/bpf/progs/bench_local_storage_create.c b/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
new file mode 100644
index 000000000000..2814bab54d28
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include "vmlinux.h"
+#include "bpf_tracing_net.h"
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+
+long create_errs = 0;
+long create_cnts = 0;
+long kmalloc_cnts = 0;
+__u32 bench_pid = 0;
+
+struct storage {
+	__u8 data[64];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct storage);
+} sk_storage_map SEC(".maps");
+
+SEC("raw_tp/kmalloc")
+int BPF_PROG(kmalloc, unsigned long call_site, const void *ptr,
+	     size_t bytes_req, size_t bytes_alloc, gfp_t gfp_flags,
+	     int node)
+{
+	__sync_fetch_and_add(&kmalloc_cnts, 1);
+
+	return 0;
+}
+
+SEC("lsm.s/socket_post_create")
+int BPF_PROG(socket_post_create, struct socket *sock, int family, int type,
+	     int protocol, int kern)
+{
+	struct storage *stg;
+	__u32 pid;
+
+	pid = bpf_get_current_pid_tgid() >> 32;
+	if (pid != bench_pid)
+		return 0;
+
+	stg = bpf_sk_storage_get(&sk_storage_map, sock->sk, NULL,
+				 BPF_LOCAL_STORAGE_GET_F_CREATE);
+
+	if (stg)
+		__sync_fetch_and_add(&create_cnts, 1);
+	else
+		__sync_fetch_and_add(&create_errs, 1);
+
+	return 0;
+}
+
+char __license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From b32a5dae44cc7346835839c2e92356ff4609c823 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Thu, 9 Mar 2023 10:01:06 -0800
Subject: bpf: verifier: Rename kernel_type_name helper to btf_type_name

kernel_type_name was introduced in commit 9e15db66136a ("bpf: Implement accurate raw_tp context access via BTF")
with type signature:

  const char *kernel_type_name(u32 id)

At that time the function used global btf_vmlinux BTF for all id lookups. Later,
in commit 22dc4a0f5ed1 ("bpf: Remove hard-coded btf_vmlinux assumption from BPF verifier"),
the type signature was changed to:

  static const char *kernel_type_name(const struct btf* btf, u32 id)

With the btf parameter used for lookups instead of global btf_vmlinux.

The helper will function as expected for type name lookup using non-kernel BTFs,
and will be used for such in further patches in the series. Let's rename it to
avoid incorrect assumptions that might arise when seeing the current name.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230309180111.1618459-2-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 0aaf7703326b..073c56dbec63 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -752,7 +752,7 @@ static int iter_get_spi(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
 	return stack_slot_obj_get_spi(env, reg, "iter", nr_slots);
 }
 
-static const char *kernel_type_name(const struct btf* btf, u32 id)
+static const char *btf_type_name(const struct btf *btf, u32 id)
 {
 	return btf_name_by_offset(btf, btf_type_by_id(btf, id)->name_off);
 }
@@ -782,7 +782,7 @@ static const char *iter_type_str(const struct btf *btf, u32 btf_id)
 		return "<invalid>";
 
 	/* we already validated that type is valid and has conforming name */
-	return kernel_type_name(btf, btf_id) + sizeof(ITER_PREFIX) - 1;
+	return btf_type_name(btf, btf_id) + sizeof(ITER_PREFIX) - 1;
 }
 
 static const char *iter_state_str(enum bpf_iter_state state)
@@ -1349,7 +1349,7 @@ static void print_verifier_state(struct bpf_verifier_env *env,
 
 			verbose(env, "%s", reg_type_str(env, t));
 			if (base_type(t) == PTR_TO_BTF_ID)
-				verbose(env, "%s", kernel_type_name(reg->btf, reg->btf_id));
+				verbose(env, "%s", btf_type_name(reg->btf, reg->btf_id));
 			verbose(env, "(");
 /*
  * _a stands for append, was shortened to avoid multiline statements below.
@@ -4518,7 +4518,7 @@ static int map_kptr_match_type(struct bpf_verifier_env *env,
 			       struct btf_field *kptr_field,
 			       struct bpf_reg_state *reg, u32 regno)
 {
-	const char *targ_name = kernel_type_name(kptr_field->kptr.btf, kptr_field->kptr.btf_id);
+	const char *targ_name = btf_type_name(kptr_field->kptr.btf, kptr_field->kptr.btf_id);
 	int perm_flags = PTR_MAYBE_NULL | PTR_TRUSTED | MEM_RCU;
 	const char *reg_name = "";
 
@@ -4534,7 +4534,7 @@ static int map_kptr_match_type(struct bpf_verifier_env *env,
 		return -EINVAL;
 	}
 	/* We need to verify reg->type and reg->btf, before accessing reg->btf */
-	reg_name = kernel_type_name(reg->btf, reg->btf_id);
+	reg_name = btf_type_name(reg->btf, reg->btf_id);
 
 	/* For ref_ptr case, release function check should ensure we get one
 	 * referenced PTR_TO_BTF_ID, and that its fixed offset is 0. For the
@@ -7177,8 +7177,8 @@ found:
 						  btf_vmlinux, *arg_btf_id,
 						  strict_type_match)) {
 				verbose(env, "R%d is of type %s but %s is expected\n",
-					regno, kernel_type_name(reg->btf, reg->btf_id),
-					kernel_type_name(btf_vmlinux, *arg_btf_id));
+					regno, btf_type_name(reg->btf, reg->btf_id),
+					btf_type_name(btf_vmlinux, *arg_btf_id));
 				return -EACCES;
 			}
 		}
@@ -7248,7 +7248,7 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env,
 			verbose(env, "R%d must have zero offset when passed to release func\n",
 				regno);
 			verbose(env, "No graph node or root found at R%d type:%s off:%d\n", regno,
-				kernel_type_name(reg->btf, reg->btf_id), reg->off);
+				btf_type_name(reg->btf, reg->btf_id), reg->off);
 			return -EINVAL;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From a4aa38897b6a6dad4318bed036edc7ed0c8a4578 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Thu, 9 Mar 2023 10:01:07 -0800
Subject: bpf: btf: Remove unused btf_field_info_type enum

This enum was added and used in commit aa3496accc41 ("bpf: Refactor kptr_off_tab
into btf_record"). Later refactoring in commit db559117828d ("bpf: Consolidate
spin_lock, timer management into btf_record") resulted in the enum
values no longer being used anywhere.

Let's remove them.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230309180111.1618459-3-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 71758cd15b07..37779ceefd09 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3231,12 +3231,6 @@ static void btf_struct_log(struct btf_verifier_env *env,
 	btf_verifier_log(env, "size=%u vlen=%u", t->size, btf_type_vlen(t));
 }
 
-enum btf_field_info_type {
-	BTF_FIELD_SPIN_LOCK,
-	BTF_FIELD_TIMER,
-	BTF_FIELD_KPTR,
-};
-
 enum {
 	BTF_FIELD_IGNORE = 0,
 	BTF_FIELD_FOUND  = 1,
-- 
cgit v1.2.3-70-g09d2


From 74843b57ec70af7b67b7e6153374834ee18d139f Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Thu, 9 Mar 2023 10:01:08 -0800
Subject: bpf: Change btf_record_find enum parameter to field_mask

btf_record_find's 3rd parameter can be multiple enum btf_field_type's
masked together. The function is called with BPF_KPTR in two places in
verifier.c, so it works with masked values already.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230309180111.1618459-4-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  | 2 +-
 kernel/bpf/syscall.c | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index e64ff1e89fb2..3a38db315f7f 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1925,7 +1925,7 @@ void bpf_prog_free_id(struct bpf_prog *prog);
 void bpf_map_free_id(struct bpf_map *map);
 
 struct btf_field *btf_record_find(const struct btf_record *rec,
-				  u32 offset, enum btf_field_type type);
+				  u32 offset, u32 field_mask);
 void btf_record_free(struct btf_record *rec);
 void bpf_map_free_record(struct bpf_map *map);
 struct btf_record *btf_record_dup(const struct btf_record *rec);
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index f406dfa13792..cc4b7684910c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -520,14 +520,14 @@ static int btf_field_cmp(const void *a, const void *b)
 }
 
 struct btf_field *btf_record_find(const struct btf_record *rec, u32 offset,
-				  enum btf_field_type type)
+				  u32 field_mask)
 {
 	struct btf_field *field;
 
-	if (IS_ERR_OR_NULL(rec) || !(rec->field_mask & type))
+	if (IS_ERR_OR_NULL(rec) || !(rec->field_mask & field_mask))
 		return NULL;
 	field = bsearch(&offset, rec->fields, rec->cnt, sizeof(rec->fields[0]), btf_field_cmp);
-	if (!field || !(field->type & type))
+	if (!field || !(field->type & field_mask))
 		return NULL;
 	return field;
 }
-- 
cgit v1.2.3-70-g09d2


From c1f9e14e3b676eb88fe1c9488c0b5f4fc9108a1c Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler@microsoft.com>
Date: Wed, 8 Mar 2023 20:53:03 +0000
Subject: bpf, docs: Explain helper functions

Add brief text about existence of helper functions, with details to go in
separate psABI text.

Note that text about runtime functions (kfuncs) is part of a separate patch,
not this one.

Signed-off-by: Dave Thaler <dthaler@microsoft.com>
Link: https://lore.kernel.org/r/20230308205303.1308-1-dthaler1968@googlemail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/clang-notes.rst     | 6 ++++++
 Documentation/bpf/instruction-set.rst | 9 ++++++++-
 Documentation/bpf/linux-notes.rst     | 8 ++++++++
 3 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/Documentation/bpf/clang-notes.rst b/Documentation/bpf/clang-notes.rst
index 528feddf2db9..2c872a1ee08e 100644
--- a/Documentation/bpf/clang-notes.rst
+++ b/Documentation/bpf/clang-notes.rst
@@ -20,6 +20,12 @@ Arithmetic instructions
 For CPU versions prior to 3, Clang v7.0 and later can enable ``BPF_ALU`` support with
 ``-Xclang -target-feature -Xclang +alu32``.  In CPU version 3, support is automatically included.
 
+Jump instructions
+=================
+
+If ``-O0`` is used, Clang will generate the ``BPF_CALL | BPF_X | BPF_JMP`` (0x8d)
+instruction, which is not supported by the Linux kernel verifier.
+
 Atomic operations
 =================
 
diff --git a/Documentation/bpf/instruction-set.rst b/Documentation/bpf/instruction-set.rst
index db8789e6969e..5e43e14abe80 100644
--- a/Documentation/bpf/instruction-set.rst
+++ b/Documentation/bpf/instruction-set.rst
@@ -253,7 +253,7 @@ BPF_JSET  0x40   PC += off if dst & src
 BPF_JNE   0x50   PC += off if dst != src
 BPF_JSGT  0x60   PC += off if dst > src     signed
 BPF_JSGE  0x70   PC += off if dst >= src    signed
-BPF_CALL  0x80   function call
+BPF_CALL  0x80   function call              see `Helper functions`_
 BPF_EXIT  0x90   function / program return  BPF_JMP only
 BPF_JLT   0xa0   PC += off if dst < src     unsigned
 BPF_JLE   0xb0   PC += off if dst <= src    unsigned
@@ -264,6 +264,13 @@ BPF_JSLE  0xd0   PC += off if dst <= src    signed
 The eBPF program needs to store the return value into register R0 before doing a
 BPF_EXIT.
 
+Helper functions
+~~~~~~~~~~~~~~~~
+
+Helper functions are a concept whereby BPF programs can call into a
+set of function calls exposed by the runtime.  Each helper
+function is identified by an integer used in a ``BPF_CALL`` instruction.
+The available helper functions may differ for each program type.
 
 Load and store instructions
 ===========================
diff --git a/Documentation/bpf/linux-notes.rst b/Documentation/bpf/linux-notes.rst
index 956b0c86699d..f43b9c797bcb 100644
--- a/Documentation/bpf/linux-notes.rst
+++ b/Documentation/bpf/linux-notes.rst
@@ -12,6 +12,14 @@ Byte swap instructions
 
 ``BPF_FROM_LE`` and ``BPF_FROM_BE`` exist as aliases for ``BPF_TO_LE`` and ``BPF_TO_BE`` respectively.
 
+Jump instructions
+=================
+
+``BPF_CALL | BPF_X | BPF_JMP`` (0x8d), where the helper function
+integer would be read from a specified register, is not currently supported
+by the verifier.  Any programs with this instruction will fail to load
+until such support is added.
+
 Legacy BPF Packet access instructions
 =====================================
 
-- 
cgit v1.2.3-70-g09d2


From cc4342f60f1a6d0f4a30ae1887a75834d0109444 Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Tue, 7 Mar 2023 20:29:27 +0100
Subject: net: mvpp2: Defer probe if MAC address source is not yet ready

NVMEM layouts are no longer registered early, and thus may not yet be
available when Ethernet drivers (or any other consumer) probe, leading
to possible probe deferrals errors. Forward the error code if this
happens. All other errors being discarded, the driver will eventually
use a random MAC address if no other source was considered valid (no
functional change on this regard).

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Marcin Wojtas <mw@semihalf.com>
Link: https://lore.kernel.org/r/20230307192927.512757-1-miquel.raynal@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 9b4ecbe4f36d..e7c7652ffac5 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -6081,18 +6081,19 @@ static bool mvpp2_port_has_irqs(struct mvpp2 *priv,
 	return true;
 }
 
-static void mvpp2_port_copy_mac_addr(struct net_device *dev, struct mvpp2 *priv,
-				     struct fwnode_handle *fwnode,
-				     char **mac_from)
+static int mvpp2_port_copy_mac_addr(struct net_device *dev, struct mvpp2 *priv,
+				    struct fwnode_handle *fwnode,
+				    char **mac_from)
 {
 	struct mvpp2_port *port = netdev_priv(dev);
 	char hw_mac_addr[ETH_ALEN] = {0};
 	char fw_mac_addr[ETH_ALEN];
+	int ret;
 
 	if (!fwnode_get_mac_address(fwnode, fw_mac_addr)) {
 		*mac_from = "firmware node";
 		eth_hw_addr_set(dev, fw_mac_addr);
-		return;
+		return 0;
 	}
 
 	if (priv->hw_version == MVPP21) {
@@ -6100,19 +6101,24 @@ static void mvpp2_port_copy_mac_addr(struct net_device *dev, struct mvpp2 *priv,
 		if (is_valid_ether_addr(hw_mac_addr)) {
 			*mac_from = "hardware";
 			eth_hw_addr_set(dev, hw_mac_addr);
-			return;
+			return 0;
 		}
 	}
 
 	/* Only valid on OF enabled platforms */
-	if (!of_get_mac_address_nvmem(to_of_node(fwnode), fw_mac_addr)) {
+	ret = of_get_mac_address_nvmem(to_of_node(fwnode), fw_mac_addr);
+	if (ret == -EPROBE_DEFER)
+		return ret;
+	if (!ret) {
 		*mac_from = "nvmem cell";
 		eth_hw_addr_set(dev, fw_mac_addr);
-		return;
+		return 0;
 	}
 
 	*mac_from = "random";
 	eth_hw_addr_random(dev);
+
+	return 0;
 }
 
 static struct mvpp2_port *mvpp2_phylink_to_port(struct phylink_config *config)
@@ -6815,7 +6821,9 @@ static int mvpp2_port_probe(struct platform_device *pdev,
 	mutex_init(&port->gather_stats_lock);
 	INIT_DELAYED_WORK(&port->stats_work, mvpp2_gather_hw_statistics);
 
-	mvpp2_port_copy_mac_addr(dev, priv, port_fwnode, &mac_from);
+	err = mvpp2_port_copy_mac_addr(dev, priv, port_fwnode, &mac_from);
+	if (err < 0)
+		goto err_free_stats;
 
 	port->tx_ring_size = MVPP2_MAX_TXD_DFLT;
 	port->rx_ring_size = MVPP2_MAX_RXD_DFLT;
-- 
cgit v1.2.3-70-g09d2


From c8e18754091479fac3f5b6c053c6bc4be0b7fb11 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Fri, 10 Mar 2023 15:07:41 -0800
Subject: bpf: Support __kptr to local kptrs

If a PTR_TO_BTF_ID type comes from program BTF - not vmlinux or module
BTF - it must have been allocated by bpf_obj_new and therefore must be
free'd with bpf_obj_drop. Such a PTR_TO_BTF_ID is considered a "local
kptr" and is tagged with MEM_ALLOC type tag by bpf_obj_new.

This patch adds support for treating __kptr-tagged pointers to "local
kptrs" as having an implicit bpf_obj_drop destructor for referenced kptr
acquire / release semantics. Consider the following example:

  struct node_data {
          long key;
          long data;
          struct bpf_rb_node node;
  };

  struct map_value {
          struct node_data __kptr *node;
  };

  struct {
          __uint(type, BPF_MAP_TYPE_ARRAY);
          __type(key, int);
          __type(value, struct map_value);
          __uint(max_entries, 1);
  } some_nodes SEC(".maps");

If struct node_data had a matching definition in kernel BTF, the verifier would
expect a destructor for the type to be registered. Since struct node_data does
not match any type in kernel BTF, the verifier knows that there is no kfunc
that provides a PTR_TO_BTF_ID to this type, and that such a PTR_TO_BTF_ID can
only come from bpf_obj_new. So instead of searching for a registered dtor,
a bpf_obj_drop dtor can be assumed.

This allows the runtime to properly destruct such kptrs in
bpf_obj_free_fields, which enables maps to clean up map_vals w/ such
kptrs when going away.

Implementation notes:
  * "kernel_btf" variable is renamed to "kptr_btf" in btf_parse_kptr.
    Before this patch, the variable would only ever point to vmlinux or
    module BTFs, but now it can point to some program BTF for local kptr
    type. It's later used to populate the (btf, btf_id) pair in kptr btf
    field.
  * It's necessary to btf_get the program BTF when populating btf_field
    for local kptr. btf_record_free later does a btf_put.
  * Behavior for non-local referenced kptrs is not modified, as
    bpf_find_btf_id helper only searches vmlinux and module BTFs for
    matching BTF type. If such a type is found, btf_field_kptr's btf will
    pass btf_is_kernel check, and the associated release function is
    some one-argument dtor. If btf_is_kernel check fails, associated
    release function is two-arg bpf_obj_drop_impl. Before this patch
    only btf_field_kptr's w/ kernel or module BTFs were created.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230310230743.2320707-2-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  | 11 ++++++++++-
 include/linux/btf.h  |  2 --
 kernel/bpf/btf.c     | 37 ++++++++++++++++++++++++++++---------
 kernel/bpf/helpers.c | 11 ++++++++---
 kernel/bpf/syscall.c | 14 +++++++++++++-
 5 files changed, 59 insertions(+), 16 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 3a38db315f7f..756b85f0d0d3 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -189,10 +189,19 @@ enum btf_field_type {
 				 BPF_RB_NODE | BPF_RB_ROOT,
 };
 
+typedef void (*btf_dtor_kfunc_t)(void *);
+typedef void (*btf_dtor_obj_drop)(void *, const struct btf_record *);
+
 struct btf_field_kptr {
 	struct btf *btf;
 	struct module *module;
-	btf_dtor_kfunc_t dtor;
+	union {
+		/* dtor used if btf_is_kernel(btf), otherwise the type
+		 * is program-allocated and obj_drop is used
+		 */
+		btf_dtor_kfunc_t dtor;
+		btf_dtor_obj_drop obj_drop;
+	};
 	u32 btf_id;
 };
 
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 1bba0827e8c4..d53b10cc55f2 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -121,8 +121,6 @@ struct btf_struct_metas {
 	struct btf_struct_meta types[];
 };
 
-typedef void (*btf_dtor_kfunc_t)(void *);
-
 extern const struct file_operations btf_fops;
 
 void btf_get(struct btf *btf);
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 37779ceefd09..66fad7a16b6c 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3551,12 +3551,17 @@ static int btf_find_field(const struct btf *btf, const struct btf_type *t,
 	return -EINVAL;
 }
 
+extern void __bpf_obj_drop_impl(void *p, const struct btf_record *rec);
+
 static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 			  struct btf_field_info *info)
 {
 	struct module *mod = NULL;
 	const struct btf_type *t;
-	struct btf *kernel_btf;
+	/* If a matching btf type is found in kernel or module BTFs, kptr_ref
+	 * is that BTF, otherwise it's program BTF
+	 */
+	struct btf *kptr_btf;
 	int ret;
 	s32 id;
 
@@ -3565,7 +3570,20 @@ static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 	 */
 	t = btf_type_by_id(btf, info->kptr.type_id);
 	id = bpf_find_btf_id(__btf_name_by_offset(btf, t->name_off), BTF_INFO_KIND(t->info),
-			     &kernel_btf);
+			     &kptr_btf);
+	if (id == -ENOENT) {
+		/* btf_parse_kptr should only be called w/ btf = program BTF */
+		WARN_ON_ONCE(btf_is_kernel(btf));
+
+		/* Type exists only in program BTF. Assume that it's a MEM_ALLOC
+		 * kptr allocated via bpf_obj_new
+		 */
+		field->kptr.dtor = (void *)&__bpf_obj_drop_impl;
+		id = info->kptr.type_id;
+		kptr_btf = (struct btf *)btf;
+		btf_get(kptr_btf);
+		goto found_dtor;
+	}
 	if (id < 0)
 		return id;
 
@@ -3582,20 +3600,20 @@ static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 		 * can be used as a referenced pointer and be stored in a map at
 		 * the same time.
 		 */
-		dtor_btf_id = btf_find_dtor_kfunc(kernel_btf, id);
+		dtor_btf_id = btf_find_dtor_kfunc(kptr_btf, id);
 		if (dtor_btf_id < 0) {
 			ret = dtor_btf_id;
 			goto end_btf;
 		}
 
-		dtor_func = btf_type_by_id(kernel_btf, dtor_btf_id);
+		dtor_func = btf_type_by_id(kptr_btf, dtor_btf_id);
 		if (!dtor_func) {
 			ret = -ENOENT;
 			goto end_btf;
 		}
 
-		if (btf_is_module(kernel_btf)) {
-			mod = btf_try_get_module(kernel_btf);
+		if (btf_is_module(kptr_btf)) {
+			mod = btf_try_get_module(kptr_btf);
 			if (!mod) {
 				ret = -ENXIO;
 				goto end_btf;
@@ -3605,7 +3623,7 @@ static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 		/* We already verified dtor_func to be btf_type_is_func
 		 * in register_btf_id_dtor_kfuncs.
 		 */
-		dtor_func_name = __btf_name_by_offset(kernel_btf, dtor_func->name_off);
+		dtor_func_name = __btf_name_by_offset(kptr_btf, dtor_func->name_off);
 		addr = kallsyms_lookup_name(dtor_func_name);
 		if (!addr) {
 			ret = -EINVAL;
@@ -3614,14 +3632,15 @@ static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 		field->kptr.dtor = (void *)addr;
 	}
 
+found_dtor:
 	field->kptr.btf_id = id;
-	field->kptr.btf = kernel_btf;
+	field->kptr.btf = kptr_btf;
 	field->kptr.module = mod;
 	return 0;
 end_mod:
 	module_put(mod);
 end_btf:
-	btf_put(kernel_btf);
+	btf_put(kptr_btf);
 	return ret;
 }
 
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index f9b7eeedce08..77d64b6951b9 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1896,14 +1896,19 @@ __bpf_kfunc void *bpf_obj_new_impl(u64 local_type_id__k, void *meta__ign)
 	return p;
 }
 
+void __bpf_obj_drop_impl(void *p, const struct btf_record *rec)
+{
+	if (rec)
+		bpf_obj_free_fields(rec, p);
+	bpf_mem_free(&bpf_global_ma, p);
+}
+
 __bpf_kfunc void bpf_obj_drop_impl(void *p__alloc, void *meta__ign)
 {
 	struct btf_struct_meta *meta = meta__ign;
 	void *p = p__alloc;
 
-	if (meta)
-		bpf_obj_free_fields(meta->record, p);
-	bpf_mem_free(&bpf_global_ma, p);
+	__bpf_obj_drop_impl(p, meta ? meta->record : NULL);
 }
 
 static void __bpf_list_add(struct bpf_list_node *node, struct bpf_list_head *head, bool tail)
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index cc4b7684910c..0684febc447a 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -659,8 +659,10 @@ void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 		return;
 	fields = rec->fields;
 	for (i = 0; i < rec->cnt; i++) {
+		struct btf_struct_meta *pointee_struct_meta;
 		const struct btf_field *field = &fields[i];
 		void *field_ptr = obj + field->offset;
+		void *xchgd_field;
 
 		switch (fields[i].type) {
 		case BPF_SPIN_LOCK:
@@ -672,7 +674,17 @@ void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 			WRITE_ONCE(*(u64 *)field_ptr, 0);
 			break;
 		case BPF_KPTR_REF:
-			field->kptr.dtor((void *)xchg((unsigned long *)field_ptr, 0));
+			xchgd_field = (void *)xchg((unsigned long *)field_ptr, 0);
+			if (!btf_is_kernel(field->kptr.btf)) {
+				pointee_struct_meta = btf_find_struct_meta(field->kptr.btf,
+									   field->kptr.btf_id);
+				WARN_ON_ONCE(!pointee_struct_meta);
+				field->kptr.obj_drop(xchgd_field, pointee_struct_meta ?
+								  pointee_struct_meta->record :
+								  NULL);
+			} else {
+				field->kptr.dtor(xchgd_field);
+			}
 			break;
 		case BPF_LIST_HEAD:
 			if (WARN_ON_ONCE(rec->spin_lock_off < 0))
-- 
cgit v1.2.3-70-g09d2


From 738c96d5e2e3700b370f02ac84a9dc159ca81f25 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Fri, 10 Mar 2023 15:07:42 -0800
Subject: bpf: Allow local kptrs to be exchanged via bpf_kptr_xchg

The previous patch added necessary plumbing for verifier and runtime to
know what to do with non-kernel PTR_TO_BTF_IDs in map values, but didn't
provide any way to get such local kptrs into a map value. This patch
modifies verifier handling of bpf_kptr_xchg to allow MEM_ALLOC kptr
types.

check_reg_type is modified accept MEM_ALLOC-flagged input to
bpf_kptr_xchg despite such types not being in btf_ptr_types. This could
have been done with a MAYBE_MEM_ALLOC equivalent to MAYBE_NULL, but
bpf_kptr_xchg is the only helper that I can forsee using
MAYBE_MEM_ALLOC, so keep it special-cased for now.

The verifier tags bpf_kptr_xchg retval MEM_ALLOC if and only if the BTF
associated with the retval is not kernel BTF.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230310230743.2320707-3-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 073c56dbec63..519d465407ef 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -7123,6 +7123,9 @@ static int check_reg_type(struct bpf_verifier_env *env, u32 regno,
 	if (arg_type & PTR_MAYBE_NULL)
 		type &= ~PTR_MAYBE_NULL;
 
+	if (meta->func_id == BPF_FUNC_kptr_xchg && type & MEM_ALLOC)
+		type &= ~MEM_ALLOC;
+
 	for (i = 0; i < ARRAY_SIZE(compatible->types); i++) {
 		expected = compatible->types[i];
 		if (expected == NOT_INIT)
@@ -7185,7 +7188,8 @@ found:
 		break;
 	}
 	case PTR_TO_BTF_ID | MEM_ALLOC:
-		if (meta->func_id != BPF_FUNC_spin_lock && meta->func_id != BPF_FUNC_spin_unlock) {
+		if (meta->func_id != BPF_FUNC_spin_lock && meta->func_id != BPF_FUNC_spin_unlock &&
+		    meta->func_id != BPF_FUNC_kptr_xchg) {
 			verbose(env, "verifier internal error: unimplemented handling of MEM_ALLOC\n");
 			return -EFAULT;
 		}
@@ -9151,6 +9155,8 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn
 		if (func_id == BPF_FUNC_kptr_xchg) {
 			ret_btf = meta.kptr_field->kptr.btf;
 			ret_btf_id = meta.kptr_field->kptr.btf_id;
+			if (!btf_is_kernel(ret_btf))
+				regs[BPF_REG_0].type |= MEM_ALLOC;
 		} else {
 			if (fn->ret_btf_id == BPF_PTR_POISON) {
 				verbose(env, "verifier internal error:");
-- 
cgit v1.2.3-70-g09d2


From 5d8d6634cccf1ebd0db4e220e52e7128b030c7b4 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Fri, 10 Mar 2023 15:07:43 -0800
Subject: selftests/bpf: Add local kptr stashing test

Add a new selftest, local_kptr_stash, which uses bpf_kptr_xchg to stash
a bpf_obj_new-allocated object in a map. Test the following scenarios:

  * Stash two rb_nodes in an arraymap, don't unstash them, rely on map
    free to destruct them
  * Stash two rb_nodes in an arraymap, unstash the second one in a
    separate program, rely on map free to destruct first

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230310230743.2320707-4-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/local_kptr_stash.c    |  60 ++++++++++++
 .../testing/selftests/bpf/progs/local_kptr_stash.c | 108 +++++++++++++++++++++
 2 files changed, 168 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/local_kptr_stash.c
 create mode 100644 tools/testing/selftests/bpf/progs/local_kptr_stash.c

diff --git a/tools/testing/selftests/bpf/prog_tests/local_kptr_stash.c b/tools/testing/selftests/bpf/prog_tests/local_kptr_stash.c
new file mode 100644
index 000000000000..76f1da877f81
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/local_kptr_stash.c
@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <test_progs.h>
+#include <network_helpers.h>
+
+#include "local_kptr_stash.skel.h"
+static void test_local_kptr_stash_simple(void)
+{
+	LIBBPF_OPTS(bpf_test_run_opts, opts,
+		    .data_in = &pkt_v4,
+		    .data_size_in = sizeof(pkt_v4),
+		    .repeat = 1,
+	);
+	struct local_kptr_stash *skel;
+	int ret;
+
+	skel = local_kptr_stash__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "local_kptr_stash__open_and_load"))
+		return;
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.stash_rb_nodes), &opts);
+	ASSERT_OK(ret, "local_kptr_stash_add_nodes run");
+	ASSERT_OK(opts.retval, "local_kptr_stash_add_nodes retval");
+
+	local_kptr_stash__destroy(skel);
+}
+
+static void test_local_kptr_stash_unstash(void)
+{
+	LIBBPF_OPTS(bpf_test_run_opts, opts,
+		    .data_in = &pkt_v4,
+		    .data_size_in = sizeof(pkt_v4),
+		    .repeat = 1,
+	);
+	struct local_kptr_stash *skel;
+	int ret;
+
+	skel = local_kptr_stash__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "local_kptr_stash__open_and_load"))
+		return;
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.stash_rb_nodes), &opts);
+	ASSERT_OK(ret, "local_kptr_stash_add_nodes run");
+	ASSERT_OK(opts.retval, "local_kptr_stash_add_nodes retval");
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.unstash_rb_node), &opts);
+	ASSERT_OK(ret, "local_kptr_stash_add_nodes run");
+	ASSERT_EQ(opts.retval, 42, "local_kptr_stash_add_nodes retval");
+
+	local_kptr_stash__destroy(skel);
+}
+
+void test_local_kptr_stash_success(void)
+{
+	if (test__start_subtest("local_kptr_stash_simple"))
+		test_local_kptr_stash_simple();
+	if (test__start_subtest("local_kptr_stash_unstash"))
+		test_local_kptr_stash_unstash();
+}
diff --git a/tools/testing/selftests/bpf/progs/local_kptr_stash.c b/tools/testing/selftests/bpf/progs/local_kptr_stash.c
new file mode 100644
index 000000000000..0ef286da092b
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/local_kptr_stash.c
@@ -0,0 +1,108 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <vmlinux.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
+#include "bpf_experimental.h"
+
+struct node_data {
+	long key;
+	long data;
+	struct bpf_rb_node node;
+};
+
+struct map_value {
+	struct prog_test_ref_kfunc *not_kptr;
+	struct prog_test_ref_kfunc __kptr *val;
+	struct node_data __kptr *node;
+};
+
+/* This is necessary so that LLVM generates BTF for node_data struct
+ * If it's not included, a fwd reference for node_data will be generated but
+ * no struct. Example BTF of "node" field in map_value when not included:
+ *
+ * [10] PTR '(anon)' type_id=35
+ * [34] FWD 'node_data' fwd_kind=struct
+ * [35] TYPE_TAG 'kptr_ref' type_id=34
+ *
+ * (with no node_data struct defined)
+ * Had to do the same w/ bpf_kfunc_call_test_release below
+ */
+struct node_data *just_here_because_btf_bug;
+
+extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 2);
+} some_nodes SEC(".maps");
+
+static int create_and_stash(int idx, int val)
+{
+	struct map_value *mapval;
+	struct node_data *res;
+
+	mapval = bpf_map_lookup_elem(&some_nodes, &idx);
+	if (!mapval)
+		return 1;
+
+	res = bpf_obj_new(typeof(*res));
+	if (!res)
+		return 1;
+	res->key = val;
+
+	res = bpf_kptr_xchg(&mapval->node, res);
+	if (res)
+		bpf_obj_drop(res);
+	return 0;
+}
+
+SEC("tc")
+long stash_rb_nodes(void *ctx)
+{
+	return create_and_stash(0, 41) ?: create_and_stash(1, 42);
+}
+
+SEC("tc")
+long unstash_rb_node(void *ctx)
+{
+	struct map_value *mapval;
+	struct node_data *res;
+	long retval;
+	int key = 1;
+
+	mapval = bpf_map_lookup_elem(&some_nodes, &key);
+	if (!mapval)
+		return 1;
+
+	res = bpf_kptr_xchg(&mapval->node, NULL);
+	if (res) {
+		retval = res->key;
+		bpf_obj_drop(res);
+		return retval;
+	}
+	return 1;
+}
+
+SEC("tc")
+long stash_test_ref_kfunc(void *ctx)
+{
+	struct prog_test_ref_kfunc *res;
+	struct map_value *mapval;
+	int key = 0;
+
+	mapval = bpf_map_lookup_elem(&some_nodes, &key);
+	if (!mapval)
+		return 1;
+
+	res = bpf_kptr_xchg(&mapval->val, NULL);
+	if (res)
+		bpf_kfunc_call_test_release(res);
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 99ce286d2d30a31eba4171036bc3f32eeb59e5f3 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 7 Mar 2023 23:09:25 +0100
Subject: net: lan966x: Add IS1 VCAP model

Provide IS1 (ingress stage 1) VCAP model for lan966x.
This provides classification actions for lan966x.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../microchip/lan966x/lan966x_vcap_ag_api.c        | 1402 +++++++++++++++++++-
 drivers/net/ethernet/microchip/vcap/vcap_ag_api.h  |  217 ++-
 .../microchip/vcap/vcap_api_debugfs_kunit.c        |    4 +-
 3 files changed, 1563 insertions(+), 60 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_ag_api.c b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_ag_api.c
index 928e711960e6..66400a082d02 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_ag_api.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_ag_api.c
@@ -6,6 +6,965 @@
 #include "lan966x_vcap_ag_api.h"
 
 /* keyfields */
+static const struct vcap_field is1_normal_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 0,
+		.width = 1,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 1,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 3,
+		.width = 9,
+	},
+	[VCAP_KF_L2_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 12,
+		.width = 1,
+	},
+	[VCAP_KF_L2_BC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_IP_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 17,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 18,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 19,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 31,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 32,
+		.width = 3,
+	},
+	[VCAP_KF_L2_SMAC] = {
+		.type = VCAP_FIELD_U48,
+		.offset = 35,
+		.width = 48,
+	},
+	[VCAP_KF_ETYPE_LEN_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 83,
+		.width = 1,
+	},
+	[VCAP_KF_ETYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 84,
+		.width = 16,
+	},
+	[VCAP_KF_IP_SNAP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 100,
+		.width = 1,
+	},
+	[VCAP_KF_IP4_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 101,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAGMENT] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 102,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAG_OFS_GT0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 103,
+		.width = 1,
+	},
+	[VCAP_KF_L3_OPTIONS_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 104,
+		.width = 1,
+	},
+	[VCAP_KF_L3_DSCP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 105,
+		.width = 6,
+	},
+	[VCAP_KF_L3_IP4_SIP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 111,
+		.width = 32,
+	},
+	[VCAP_KF_TCP_UDP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 143,
+		.width = 1,
+	},
+	[VCAP_KF_TCP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 144,
+		.width = 1,
+	},
+	[VCAP_KF_L4_SPORT] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 145,
+		.width = 16,
+	},
+	[VCAP_KF_L4_RNG] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 161,
+		.width = 8,
+	},
+};
+
+static const struct vcap_field is1_5tuple_ip4_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 0,
+		.width = 1,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 1,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 3,
+		.width = 9,
+	},
+	[VCAP_KF_L2_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 12,
+		.width = 1,
+	},
+	[VCAP_KF_L2_BC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_IP_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 17,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 18,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 19,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 31,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 32,
+		.width = 3,
+	},
+	[VCAP_KF_8021Q_TPID1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 35,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 36,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 48,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 49,
+		.width = 3,
+	},
+	[VCAP_KF_IP4_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 52,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAGMENT] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 53,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAG_OFS_GT0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 54,
+		.width = 1,
+	},
+	[VCAP_KF_L3_OPTIONS_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 55,
+		.width = 1,
+	},
+	[VCAP_KF_L3_DSCP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 56,
+		.width = 6,
+	},
+	[VCAP_KF_L3_IP4_DIP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 62,
+		.width = 32,
+	},
+	[VCAP_KF_L3_IP4_SIP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 94,
+		.width = 32,
+	},
+	[VCAP_KF_L3_IP_PROTO] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 126,
+		.width = 8,
+	},
+	[VCAP_KF_TCP_UDP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 134,
+		.width = 1,
+	},
+	[VCAP_KF_TCP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 135,
+		.width = 1,
+	},
+	[VCAP_KF_L4_RNG] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 136,
+		.width = 8,
+	},
+	[VCAP_KF_IP_PAYLOAD_5TUPLE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 144,
+		.width = 32,
+	},
+};
+
+static const struct vcap_field is1_normal_ip6_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 0,
+		.width = 2,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 2,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 4,
+		.width = 9,
+	},
+	[VCAP_KF_L2_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_L2_BC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_IP_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 17,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 18,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 19,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 20,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 32,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 33,
+		.width = 3,
+	},
+	[VCAP_KF_8021Q_TPID1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 36,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 37,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 49,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 50,
+		.width = 3,
+	},
+	[VCAP_KF_L2_SMAC] = {
+		.type = VCAP_FIELD_U48,
+		.offset = 53,
+		.width = 48,
+	},
+	[VCAP_KF_L3_DSCP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 101,
+		.width = 6,
+	},
+	[VCAP_KF_L3_IP6_SIP] = {
+		.type = VCAP_FIELD_U128,
+		.offset = 107,
+		.width = 128,
+	},
+	[VCAP_KF_L3_IP_PROTO] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 235,
+		.width = 8,
+	},
+	[VCAP_KF_TCP_UDP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 243,
+		.width = 1,
+	},
+	[VCAP_KF_L4_RNG] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 244,
+		.width = 8,
+	},
+	[VCAP_KF_IP_PAYLOAD_S1_IP6] = {
+		.type = VCAP_FIELD_U112,
+		.offset = 252,
+		.width = 112,
+	},
+};
+
+static const struct vcap_field is1_7tuple_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 0,
+		.width = 2,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 2,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 4,
+		.width = 9,
+	},
+	[VCAP_KF_L2_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_L2_BC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_IP_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 17,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 18,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 19,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 20,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 32,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 33,
+		.width = 3,
+	},
+	[VCAP_KF_8021Q_TPID1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 36,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 37,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 49,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 50,
+		.width = 3,
+	},
+	[VCAP_KF_L2_DMAC] = {
+		.type = VCAP_FIELD_U48,
+		.offset = 53,
+		.width = 48,
+	},
+	[VCAP_KF_L2_SMAC] = {
+		.type = VCAP_FIELD_U48,
+		.offset = 101,
+		.width = 48,
+	},
+	[VCAP_KF_ETYPE_LEN_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 149,
+		.width = 1,
+	},
+	[VCAP_KF_ETYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 150,
+		.width = 16,
+	},
+	[VCAP_KF_IP_SNAP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 166,
+		.width = 1,
+	},
+	[VCAP_KF_IP4_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 167,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAGMENT] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 168,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAG_OFS_GT0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 169,
+		.width = 1,
+	},
+	[VCAP_KF_L3_OPTIONS_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 170,
+		.width = 1,
+	},
+	[VCAP_KF_L3_DSCP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 171,
+		.width = 6,
+	},
+	[VCAP_KF_L3_IP6_DIP_MSB] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 177,
+		.width = 16,
+	},
+	[VCAP_KF_L3_IP6_DIP] = {
+		.type = VCAP_FIELD_U64,
+		.offset = 193,
+		.width = 64,
+	},
+	[VCAP_KF_L3_IP6_SIP_MSB] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 257,
+		.width = 16,
+	},
+	[VCAP_KF_L3_IP6_SIP] = {
+		.type = VCAP_FIELD_U64,
+		.offset = 273,
+		.width = 64,
+	},
+	[VCAP_KF_TCP_UDP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 337,
+		.width = 1,
+	},
+	[VCAP_KF_TCP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 338,
+		.width = 1,
+	},
+	[VCAP_KF_L4_SPORT] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 339,
+		.width = 16,
+	},
+	[VCAP_KF_L4_RNG] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 355,
+		.width = 8,
+	},
+};
+
+static const struct vcap_field is1_5tuple_ip6_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 0,
+		.width = 2,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 2,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 4,
+		.width = 9,
+	},
+	[VCAP_KF_L2_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_L2_BC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_IP_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 17,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 18,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 19,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 20,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 32,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 33,
+		.width = 3,
+	},
+	[VCAP_KF_8021Q_TPID1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 36,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 37,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 49,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 50,
+		.width = 3,
+	},
+	[VCAP_KF_L3_DSCP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 53,
+		.width = 6,
+	},
+	[VCAP_KF_L3_IP6_DIP] = {
+		.type = VCAP_FIELD_U128,
+		.offset = 59,
+		.width = 128,
+	},
+	[VCAP_KF_L3_IP6_SIP] = {
+		.type = VCAP_FIELD_U128,
+		.offset = 187,
+		.width = 128,
+	},
+	[VCAP_KF_L3_IP_PROTO] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 315,
+		.width = 8,
+	},
+	[VCAP_KF_TCP_UDP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 323,
+		.width = 1,
+	},
+	[VCAP_KF_L4_RNG] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 324,
+		.width = 8,
+	},
+	[VCAP_KF_IP_PAYLOAD_5TUPLE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 332,
+		.width = 32,
+	},
+};
+
+static const struct vcap_field is1_dbl_vid_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 0,
+		.width = 2,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 2,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 4,
+		.width = 9,
+	},
+	[VCAP_KF_L2_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_L2_BC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_IP_MC_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 17,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 18,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 19,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 20,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 32,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 33,
+		.width = 3,
+	},
+	[VCAP_KF_8021Q_TPID1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 36,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 37,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI1] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 49,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP1] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 50,
+		.width = 3,
+	},
+	[VCAP_KF_ETYPE_LEN_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 53,
+		.width = 1,
+	},
+	[VCAP_KF_ETYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 54,
+		.width = 16,
+	},
+	[VCAP_KF_IP_SNAP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 70,
+		.width = 1,
+	},
+	[VCAP_KF_IP4_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 71,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAGMENT] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 72,
+		.width = 1,
+	},
+	[VCAP_KF_L3_FRAG_OFS_GT0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 73,
+		.width = 1,
+	},
+	[VCAP_KF_L3_OPTIONS_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 74,
+		.width = 1,
+	},
+	[VCAP_KF_L3_DSCP] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 75,
+		.width = 6,
+	},
+	[VCAP_KF_TCP_UDP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 81,
+		.width = 1,
+	},
+	[VCAP_KF_TCP_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 82,
+		.width = 1,
+	},
+};
+
+static const struct vcap_field is1_rt_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 0,
+		.width = 2,
+	},
+	[VCAP_KF_LOOKUP_FIRST_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 2,
+		.width = 1,
+	},
+	[VCAP_KF_IF_IGR_PORT] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 3,
+		.width = 3,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 6,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 7,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 8,
+		.width = 1,
+	},
+	[VCAP_KF_L2_MAC] = {
+		.type = VCAP_FIELD_U48,
+		.offset = 9,
+		.width = 48,
+	},
+	[VCAP_KF_RT_VLAN_IDX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 57,
+		.width = 3,
+	},
+	[VCAP_KF_RT_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 60,
+		.width = 2,
+	},
+	[VCAP_KF_RT_FRMID] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 62,
+		.width = 32,
+	},
+};
+
+static const struct vcap_field is1_dmac_vid_keyfield[] = {
+	[VCAP_KF_TYPE] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 0,
+		.width = 2,
+	},
+	[VCAP_KF_LOOKUP_INDEX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 2,
+		.width = 2,
+	},
+	[VCAP_KF_IF_IGR_PORT_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 4,
+		.width = 9,
+	},
+	[VCAP_KF_8021CB_R_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 14,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 15,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_TPID0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 16,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_VID0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 17,
+		.width = 12,
+	},
+	[VCAP_KF_8021Q_DEI0] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 29,
+		.width = 1,
+	},
+	[VCAP_KF_8021Q_PCP0] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 30,
+		.width = 3,
+	},
+	[VCAP_KF_L2_DMAC] = {
+		.type = VCAP_FIELD_U48,
+		.offset = 33,
+		.width = 48,
+	},
+};
+
 static const struct vcap_field is2_mac_etype_keyfield[] = {
 	[VCAP_KF_TYPE] = {
 		.type = VCAP_FIELD_U32,
@@ -1163,6 +2122,49 @@ static const struct vcap_field is2_smac_sip6_keyfield[] = {
 };
 
 /* keyfield_set */
+static const struct vcap_set is1_keyfield_set[] = {
+	[VCAP_KFS_NORMAL] = {
+		.type_id = 0,
+		.sw_per_item = 2,
+		.sw_cnt = 2,
+	},
+	[VCAP_KFS_5TUPLE_IP4] = {
+		.type_id = 1,
+		.sw_per_item = 2,
+		.sw_cnt = 2,
+	},
+	[VCAP_KFS_NORMAL_IP6] = {
+		.type_id = 0,
+		.sw_per_item = 4,
+		.sw_cnt = 1,
+	},
+	[VCAP_KFS_7TUPLE] = {
+		.type_id = 1,
+		.sw_per_item = 4,
+		.sw_cnt = 1,
+	},
+	[VCAP_KFS_5TUPLE_IP6] = {
+		.type_id = 2,
+		.sw_per_item = 4,
+		.sw_cnt = 1,
+	},
+	[VCAP_KFS_DBL_VID] = {
+		.type_id = 0,
+		.sw_per_item = 1,
+		.sw_cnt = 4,
+	},
+	[VCAP_KFS_RT] = {
+		.type_id = 1,
+		.sw_per_item = 1,
+		.sw_cnt = 4,
+	},
+	[VCAP_KFS_DMAC_VID] = {
+		.type_id = 2,
+		.sw_per_item = 1,
+		.sw_cnt = 4,
+	},
+};
+
 static const struct vcap_set is2_keyfield_set[] = {
 	[VCAP_KFS_MAC_ETYPE] = {
 		.type_id = 0,
@@ -1227,6 +2229,17 @@ static const struct vcap_set is2_keyfield_set[] = {
 };
 
 /* keyfield_set map */
+static const struct vcap_field *is1_keyfield_set_map[] = {
+	[VCAP_KFS_NORMAL] = is1_normal_keyfield,
+	[VCAP_KFS_5TUPLE_IP4] = is1_5tuple_ip4_keyfield,
+	[VCAP_KFS_NORMAL_IP6] = is1_normal_ip6_keyfield,
+	[VCAP_KFS_7TUPLE] = is1_7tuple_keyfield,
+	[VCAP_KFS_5TUPLE_IP6] = is1_5tuple_ip6_keyfield,
+	[VCAP_KFS_DBL_VID] = is1_dbl_vid_keyfield,
+	[VCAP_KFS_RT] = is1_rt_keyfield,
+	[VCAP_KFS_DMAC_VID] = is1_dmac_vid_keyfield,
+};
+
 static const struct vcap_field *is2_keyfield_set_map[] = {
 	[VCAP_KFS_MAC_ETYPE] = is2_mac_etype_keyfield,
 	[VCAP_KFS_MAC_LLC] = is2_mac_llc_keyfield,
@@ -1243,6 +2256,17 @@ static const struct vcap_field *is2_keyfield_set_map[] = {
 };
 
 /* keyfield_set map sizes */
+static int is1_keyfield_set_map_size[] = {
+	[VCAP_KFS_NORMAL] = ARRAY_SIZE(is1_normal_keyfield),
+	[VCAP_KFS_5TUPLE_IP4] = ARRAY_SIZE(is1_5tuple_ip4_keyfield),
+	[VCAP_KFS_NORMAL_IP6] = ARRAY_SIZE(is1_normal_ip6_keyfield),
+	[VCAP_KFS_7TUPLE] = ARRAY_SIZE(is1_7tuple_keyfield),
+	[VCAP_KFS_5TUPLE_IP6] = ARRAY_SIZE(is1_5tuple_ip6_keyfield),
+	[VCAP_KFS_DBL_VID] = ARRAY_SIZE(is1_dbl_vid_keyfield),
+	[VCAP_KFS_RT] = ARRAY_SIZE(is1_rt_keyfield),
+	[VCAP_KFS_DMAC_VID] = ARRAY_SIZE(is1_dmac_vid_keyfield),
+};
+
 static int is2_keyfield_set_map_size[] = {
 	[VCAP_KFS_MAC_ETYPE] = ARRAY_SIZE(is2_mac_etype_keyfield),
 	[VCAP_KFS_MAC_LLC] = ARRAY_SIZE(is2_mac_llc_keyfield),
@@ -1259,6 +2283,154 @@ static int is2_keyfield_set_map_size[] = {
 };
 
 /* actionfields */
+static const struct vcap_field is1_s1_actionfield[] = {
+	[VCAP_AF_TYPE] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 0,
+		.width = 1,
+	},
+	[VCAP_AF_DSCP_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 1,
+		.width = 1,
+	},
+	[VCAP_AF_DSCP_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 2,
+		.width = 6,
+	},
+	[VCAP_AF_QOS_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 8,
+		.width = 1,
+	},
+	[VCAP_AF_QOS_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 9,
+		.width = 3,
+	},
+	[VCAP_AF_DP_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 12,
+		.width = 1,
+	},
+	[VCAP_AF_DP_VAL] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 13,
+		.width = 1,
+	},
+	[VCAP_AF_PAG_OVERRIDE_MASK] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 14,
+		.width = 8,
+	},
+	[VCAP_AF_PAG_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 22,
+		.width = 8,
+	},
+	[VCAP_AF_ISDX_REPLACE_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 30,
+		.width = 1,
+	},
+	[VCAP_AF_ISDX_ADD_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 31,
+		.width = 8,
+	},
+	[VCAP_AF_VID_REPLACE_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 39,
+		.width = 1,
+	},
+	[VCAP_AF_VID_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 40,
+		.width = 12,
+	},
+	[VCAP_AF_PCP_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 67,
+		.width = 1,
+	},
+	[VCAP_AF_PCP_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 68,
+		.width = 3,
+	},
+	[VCAP_AF_DEI_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 71,
+		.width = 1,
+	},
+	[VCAP_AF_DEI_VAL] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 72,
+		.width = 1,
+	},
+	[VCAP_AF_VLAN_POP_CNT_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 73,
+		.width = 1,
+	},
+	[VCAP_AF_VLAN_POP_CNT] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 74,
+		.width = 2,
+	},
+	[VCAP_AF_CUSTOM_ACE_TYPE_ENA] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 76,
+		.width = 4,
+	},
+	[VCAP_AF_SFID_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 80,
+		.width = 1,
+	},
+	[VCAP_AF_SFID_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 81,
+		.width = 8,
+	},
+	[VCAP_AF_SGID_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 89,
+		.width = 1,
+	},
+	[VCAP_AF_SGID_VAL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 90,
+		.width = 8,
+	},
+	[VCAP_AF_POLICE_ENA] = {
+		.type = VCAP_FIELD_BIT,
+		.offset = 98,
+		.width = 1,
+	},
+	[VCAP_AF_POLICE_IDX] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 99,
+		.width = 9,
+	},
+	[VCAP_AF_OAM_SEL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 108,
+		.width = 3,
+	},
+	[VCAP_AF_MRP_SEL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 111,
+		.width = 2,
+	},
+	[VCAP_AF_DLR_SEL] = {
+		.type = VCAP_FIELD_U32,
+		.offset = 113,
+		.width = 2,
+	},
+};
+
 static const struct vcap_field is2_base_type_actionfield[] = {
 	[VCAP_AF_HIT_ME_ONCE] = {
 		.type = VCAP_FIELD_BIT,
@@ -1351,6 +2523,14 @@ static const struct vcap_field is2_smac_sip_actionfield[] = {
 };
 
 /* actionfield_set */
+static const struct vcap_set is1_actionfield_set[] = {
+	[VCAP_AFS_S1] = {
+		.type_id = 0,
+		.sw_per_item = 1,
+		.sw_cnt = 4,
+	},
+};
+
 static const struct vcap_set is2_actionfield_set[] = {
 	[VCAP_AFS_BASE_TYPE] = {
 		.type_id = -1,
@@ -1365,18 +2545,73 @@ static const struct vcap_set is2_actionfield_set[] = {
 };
 
 /* actionfield_set map */
+static const struct vcap_field *is1_actionfield_set_map[] = {
+	[VCAP_AFS_S1] = is1_s1_actionfield,
+};
+
 static const struct vcap_field *is2_actionfield_set_map[] = {
 	[VCAP_AFS_BASE_TYPE] = is2_base_type_actionfield,
 	[VCAP_AFS_SMAC_SIP] = is2_smac_sip_actionfield,
 };
 
 /* actionfield_set map size */
+static int is1_actionfield_set_map_size[] = {
+	[VCAP_AFS_S1] = ARRAY_SIZE(is1_s1_actionfield),
+};
+
 static int is2_actionfield_set_map_size[] = {
 	[VCAP_AFS_BASE_TYPE] = ARRAY_SIZE(is2_base_type_actionfield),
 	[VCAP_AFS_SMAC_SIP] = ARRAY_SIZE(is2_smac_sip_actionfield),
 };
 
 /* Type Groups */
+static const struct vcap_typegroup is1_x4_keyfield_set_typegroups[] = {
+	{
+		.offset = 0,
+		.width = 3,
+		.value = 4,
+	},
+	{
+		.offset = 96,
+		.width = 1,
+		.value = 0,
+	},
+	{
+		.offset = 192,
+		.width = 2,
+		.value = 0,
+	},
+	{
+		.offset = 288,
+		.width = 1,
+		.value = 0,
+	},
+	{}
+};
+
+static const struct vcap_typegroup is1_x2_keyfield_set_typegroups[] = {
+	{
+		.offset = 0,
+		.width = 2,
+		.value = 2,
+	},
+	{
+		.offset = 96,
+		.width = 1,
+		.value = 0,
+	},
+	{}
+};
+
+static const struct vcap_typegroup is1_x1_keyfield_set_typegroups[] = {
+	{
+		.offset = 0,
+		.width = 1,
+		.value = 1,
+	},
+	{}
+};
+
 static const struct vcap_typegroup is2_x4_keyfield_set_typegroups[] = {
 	{
 		.offset = 0,
@@ -1424,6 +2659,13 @@ static const struct vcap_typegroup is2_x1_keyfield_set_typegroups[] = {
 	{}
 };
 
+static const struct vcap_typegroup *is1_keyfield_set_typegroups[] = {
+	[4] = is1_x4_keyfield_set_typegroups,
+	[2] = is1_x2_keyfield_set_typegroups,
+	[1] = is1_x1_keyfield_set_typegroups,
+	[5] = NULL,
+};
+
 static const struct vcap_typegroup *is2_keyfield_set_typegroups[] = {
 	[4] = is2_x4_keyfield_set_typegroups,
 	[2] = is2_x2_keyfield_set_typegroups,
@@ -1431,6 +2673,10 @@ static const struct vcap_typegroup *is2_keyfield_set_typegroups[] = {
 	[5] = NULL,
 };
 
+static const struct vcap_typegroup is1_x1_actionfield_set_typegroups[] = {
+	{}
+};
+
 static const struct vcap_typegroup is2_x2_actionfield_set_typegroups[] = {
 	{
 		.offset = 0,
@@ -1454,6 +2700,11 @@ static const struct vcap_typegroup is2_x1_actionfield_set_typegroups[] = {
 	{}
 };
 
+static const struct vcap_typegroup *is1_actionfield_set_typegroups[] = {
+	[1] = is1_x1_actionfield_set_typegroups,
+	[5] = NULL,
+};
+
 static const struct vcap_typegroup *is2_actionfield_set_typegroups[] = {
 	[2] = is2_x2_actionfield_set_typegroups,
 	[1] = is2_x1_actionfield_set_typegroups,
@@ -1463,16 +2714,33 @@ static const struct vcap_typegroup *is2_actionfield_set_typegroups[] = {
 /* Keyfieldset names */
 static const char * const vcap_keyfield_set_names[] = {
 	[VCAP_KFS_NO_VALUE]                      =  "(None)",
+	[VCAP_KFS_5TUPLE_IP4]                    =  "VCAP_KFS_5TUPLE_IP4",
+	[VCAP_KFS_5TUPLE_IP6]                    =  "VCAP_KFS_5TUPLE_IP6",
+	[VCAP_KFS_7TUPLE]                        =  "VCAP_KFS_7TUPLE",
 	[VCAP_KFS_ARP]                           =  "VCAP_KFS_ARP",
+	[VCAP_KFS_DBL_VID]                       =  "VCAP_KFS_DBL_VID",
+	[VCAP_KFS_DMAC_VID]                      =  "VCAP_KFS_DMAC_VID",
+	[VCAP_KFS_ETAG]                          =  "VCAP_KFS_ETAG",
 	[VCAP_KFS_IP4_OTHER]                     =  "VCAP_KFS_IP4_OTHER",
 	[VCAP_KFS_IP4_TCP_UDP]                   =  "VCAP_KFS_IP4_TCP_UDP",
+	[VCAP_KFS_IP4_VID]                       =  "VCAP_KFS_IP4_VID",
 	[VCAP_KFS_IP6_OTHER]                     =  "VCAP_KFS_IP6_OTHER",
 	[VCAP_KFS_IP6_STD]                       =  "VCAP_KFS_IP6_STD",
 	[VCAP_KFS_IP6_TCP_UDP]                   =  "VCAP_KFS_IP6_TCP_UDP",
+	[VCAP_KFS_IP6_VID]                       =  "VCAP_KFS_IP6_VID",
+	[VCAP_KFS_IP_7TUPLE]                     =  "VCAP_KFS_IP_7TUPLE",
+	[VCAP_KFS_ISDX]                          =  "VCAP_KFS_ISDX",
+	[VCAP_KFS_LL_FULL]                       =  "VCAP_KFS_LL_FULL",
 	[VCAP_KFS_MAC_ETYPE]                     =  "VCAP_KFS_MAC_ETYPE",
 	[VCAP_KFS_MAC_LLC]                       =  "VCAP_KFS_MAC_LLC",
 	[VCAP_KFS_MAC_SNAP]                      =  "VCAP_KFS_MAC_SNAP",
+	[VCAP_KFS_NORMAL]                        =  "VCAP_KFS_NORMAL",
+	[VCAP_KFS_NORMAL_5TUPLE_IP4]             =  "VCAP_KFS_NORMAL_5TUPLE_IP4",
+	[VCAP_KFS_NORMAL_7TUPLE]                 =  "VCAP_KFS_NORMAL_7TUPLE",
+	[VCAP_KFS_NORMAL_IP6]                    =  "VCAP_KFS_NORMAL_IP6",
 	[VCAP_KFS_OAM]                           =  "VCAP_KFS_OAM",
+	[VCAP_KFS_PURE_5TUPLE_IP4]               =  "VCAP_KFS_PURE_5TUPLE_IP4",
+	[VCAP_KFS_RT]                            =  "VCAP_KFS_RT",
 	[VCAP_KFS_SMAC_SIP4]                     =  "VCAP_KFS_SMAC_SIP4",
 	[VCAP_KFS_SMAC_SIP6]                     =  "VCAP_KFS_SMAC_SIP6",
 };
@@ -1481,16 +2749,42 @@ static const char * const vcap_keyfield_set_names[] = {
 static const char * const vcap_actionfield_set_names[] = {
 	[VCAP_AFS_NO_VALUE]                      =  "(None)",
 	[VCAP_AFS_BASE_TYPE]                     =  "VCAP_AFS_BASE_TYPE",
+	[VCAP_AFS_CLASSIFICATION]                =  "VCAP_AFS_CLASSIFICATION",
+	[VCAP_AFS_CLASS_REDUCED]                 =  "VCAP_AFS_CLASS_REDUCED",
+	[VCAP_AFS_FULL]                          =  "VCAP_AFS_FULL",
+	[VCAP_AFS_S1]                            =  "VCAP_AFS_S1",
 	[VCAP_AFS_SMAC_SIP]                      =  "VCAP_AFS_SMAC_SIP",
 };
 
 /* Keyfield names */
 static const char * const vcap_keyfield_names[] = {
 	[VCAP_KF_NO_VALUE]                       =  "(None)",
+	[VCAP_KF_8021BR_ECID_BASE]               =  "8021BR_ECID_BASE",
+	[VCAP_KF_8021BR_ECID_EXT]                =  "8021BR_ECID_EXT",
+	[VCAP_KF_8021BR_E_TAGGED]                =  "8021BR_E_TAGGED",
+	[VCAP_KF_8021BR_GRP]                     =  "8021BR_GRP",
+	[VCAP_KF_8021BR_IGR_ECID_BASE]           =  "8021BR_IGR_ECID_BASE",
+	[VCAP_KF_8021BR_IGR_ECID_EXT]            =  "8021BR_IGR_ECID_EXT",
+	[VCAP_KF_8021CB_R_TAGGED_IS]             =  "8021CB_R_TAGGED_IS",
+	[VCAP_KF_8021Q_DEI0]                     =  "8021Q_DEI0",
+	[VCAP_KF_8021Q_DEI1]                     =  "8021Q_DEI1",
+	[VCAP_KF_8021Q_DEI2]                     =  "8021Q_DEI2",
 	[VCAP_KF_8021Q_DEI_CLS]                  =  "8021Q_DEI_CLS",
+	[VCAP_KF_8021Q_PCP0]                     =  "8021Q_PCP0",
+	[VCAP_KF_8021Q_PCP1]                     =  "8021Q_PCP1",
+	[VCAP_KF_8021Q_PCP2]                     =  "8021Q_PCP2",
 	[VCAP_KF_8021Q_PCP_CLS]                  =  "8021Q_PCP_CLS",
+	[VCAP_KF_8021Q_TPID0]                    =  "8021Q_TPID0",
+	[VCAP_KF_8021Q_TPID1]                    =  "8021Q_TPID1",
+	[VCAP_KF_8021Q_TPID2]                    =  "8021Q_TPID2",
+	[VCAP_KF_8021Q_VID0]                     =  "8021Q_VID0",
+	[VCAP_KF_8021Q_VID1]                     =  "8021Q_VID1",
+	[VCAP_KF_8021Q_VID2]                     =  "8021Q_VID2",
 	[VCAP_KF_8021Q_VID_CLS]                  =  "8021Q_VID_CLS",
+	[VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS]       =  "8021Q_VLAN_DBL_TAGGED_IS",
 	[VCAP_KF_8021Q_VLAN_TAGGED_IS]           =  "8021Q_VLAN_TAGGED_IS",
+	[VCAP_KF_8021Q_VLAN_TAGS]                =  "8021Q_VLAN_TAGS",
+	[VCAP_KF_ACL_GRP_ID]                     =  "ACL_GRP_ID",
 	[VCAP_KF_ARP_ADDR_SPACE_OK_IS]           =  "ARP_ADDR_SPACE_OK_IS",
 	[VCAP_KF_ARP_LEN_OK_IS]                  =  "ARP_LEN_OK_IS",
 	[VCAP_KF_ARP_OPCODE]                     =  "ARP_OPCODE",
@@ -1498,32 +2792,57 @@ static const char * const vcap_keyfield_names[] = {
 	[VCAP_KF_ARP_PROTO_SPACE_OK_IS]          =  "ARP_PROTO_SPACE_OK_IS",
 	[VCAP_KF_ARP_SENDER_MATCH_IS]            =  "ARP_SENDER_MATCH_IS",
 	[VCAP_KF_ARP_TGT_MATCH_IS]               =  "ARP_TGT_MATCH_IS",
+	[VCAP_KF_COSID_CLS]                      =  "COSID_CLS",
+	[VCAP_KF_ES0_ISDX_KEY_ENA]               =  "ES0_ISDX_KEY_ENA",
 	[VCAP_KF_ETYPE]                          =  "ETYPE",
+	[VCAP_KF_ETYPE_LEN_IS]                   =  "ETYPE_LEN_IS",
 	[VCAP_KF_HOST_MATCH]                     =  "HOST_MATCH",
+	[VCAP_KF_IF_EGR_PORT_MASK]               =  "IF_EGR_PORT_MASK",
+	[VCAP_KF_IF_EGR_PORT_MASK_RNG]           =  "IF_EGR_PORT_MASK_RNG",
 	[VCAP_KF_IF_IGR_PORT]                    =  "IF_IGR_PORT",
 	[VCAP_KF_IF_IGR_PORT_MASK]               =  "IF_IGR_PORT_MASK",
+	[VCAP_KF_IF_IGR_PORT_MASK_L3]            =  "IF_IGR_PORT_MASK_L3",
+	[VCAP_KF_IF_IGR_PORT_MASK_RNG]           =  "IF_IGR_PORT_MASK_RNG",
+	[VCAP_KF_IF_IGR_PORT_MASK_SEL]           =  "IF_IGR_PORT_MASK_SEL",
+	[VCAP_KF_IF_IGR_PORT_SEL]                =  "IF_IGR_PORT_SEL",
 	[VCAP_KF_IP4_IS]                         =  "IP4_IS",
+	[VCAP_KF_IP_MC_IS]                       =  "IP_MC_IS",
+	[VCAP_KF_IP_PAYLOAD_5TUPLE]              =  "IP_PAYLOAD_5TUPLE",
+	[VCAP_KF_IP_PAYLOAD_S1_IP6]              =  "IP_PAYLOAD_S1_IP6",
+	[VCAP_KF_IP_SNAP_IS]                     =  "IP_SNAP_IS",
+	[VCAP_KF_ISDX_CLS]                       =  "ISDX_CLS",
 	[VCAP_KF_ISDX_GT0_IS]                    =  "ISDX_GT0_IS",
 	[VCAP_KF_L2_BC_IS]                       =  "L2_BC_IS",
 	[VCAP_KF_L2_DMAC]                        =  "L2_DMAC",
 	[VCAP_KF_L2_FRM_TYPE]                    =  "L2_FRM_TYPE",
+	[VCAP_KF_L2_FWD_IS]                      =  "L2_FWD_IS",
 	[VCAP_KF_L2_LLC]                         =  "L2_LLC",
+	[VCAP_KF_L2_MAC]                         =  "L2_MAC",
 	[VCAP_KF_L2_MC_IS]                       =  "L2_MC_IS",
 	[VCAP_KF_L2_PAYLOAD0]                    =  "L2_PAYLOAD0",
 	[VCAP_KF_L2_PAYLOAD1]                    =  "L2_PAYLOAD1",
 	[VCAP_KF_L2_PAYLOAD2]                    =  "L2_PAYLOAD2",
+	[VCAP_KF_L2_PAYLOAD_ETYPE]               =  "L2_PAYLOAD_ETYPE",
 	[VCAP_KF_L2_SMAC]                        =  "L2_SMAC",
 	[VCAP_KF_L2_SNAP]                        =  "L2_SNAP",
 	[VCAP_KF_L3_DIP_EQ_SIP_IS]               =  "L3_DIP_EQ_SIP_IS",
+	[VCAP_KF_L3_DPL_CLS]                     =  "L3_DPL_CLS",
+	[VCAP_KF_L3_DSCP]                        =  "L3_DSCP",
+	[VCAP_KF_L3_DST_IS]                      =  "L3_DST_IS",
 	[VCAP_KF_L3_FRAGMENT]                    =  "L3_FRAGMENT",
+	[VCAP_KF_L3_FRAGMENT_TYPE]               =  "L3_FRAGMENT_TYPE",
+	[VCAP_KF_L3_FRAG_INVLD_L4_LEN]           =  "L3_FRAG_INVLD_L4_LEN",
 	[VCAP_KF_L3_FRAG_OFS_GT0]                =  "L3_FRAG_OFS_GT0",
 	[VCAP_KF_L3_IP4_DIP]                     =  "L3_IP4_DIP",
 	[VCAP_KF_L3_IP4_SIP]                     =  "L3_IP4_SIP",
 	[VCAP_KF_L3_IP6_DIP]                     =  "L3_IP6_DIP",
+	[VCAP_KF_L3_IP6_DIP_MSB]                 =  "L3_IP6_DIP_MSB",
 	[VCAP_KF_L3_IP6_SIP]                     =  "L3_IP6_SIP",
+	[VCAP_KF_L3_IP6_SIP_MSB]                 =  "L3_IP6_SIP_MSB",
 	[VCAP_KF_L3_IP_PROTO]                    =  "L3_IP_PROTO",
 	[VCAP_KF_L3_OPTIONS_IS]                  =  "L3_OPTIONS_IS",
 	[VCAP_KF_L3_PAYLOAD]                     =  "L3_PAYLOAD",
+	[VCAP_KF_L3_RT_IS]                       =  "L3_RT_IS",
 	[VCAP_KF_L3_TOS]                         =  "L3_TOS",
 	[VCAP_KF_L3_TTL_GT0]                     =  "L3_TTL_GT0",
 	[VCAP_KF_L4_1588_DOM]                    =  "L4_1588_DOM",
@@ -1531,6 +2850,7 @@ static const char * const vcap_keyfield_names[] = {
 	[VCAP_KF_L4_ACK]                         =  "L4_ACK",
 	[VCAP_KF_L4_DPORT]                       =  "L4_DPORT",
 	[VCAP_KF_L4_FIN]                         =  "L4_FIN",
+	[VCAP_KF_L4_PAYLOAD]                     =  "L4_PAYLOAD",
 	[VCAP_KF_L4_PSH]                         =  "L4_PSH",
 	[VCAP_KF_L4_RNG]                         =  "L4_RNG",
 	[VCAP_KF_L4_RST]                         =  "L4_RST",
@@ -1540,7 +2860,11 @@ static const char * const vcap_keyfield_names[] = {
 	[VCAP_KF_L4_SYN]                         =  "L4_SYN",
 	[VCAP_KF_L4_URG]                         =  "L4_URG",
 	[VCAP_KF_LOOKUP_FIRST_IS]                =  "LOOKUP_FIRST_IS",
+	[VCAP_KF_LOOKUP_GEN_IDX]                 =  "LOOKUP_GEN_IDX",
+	[VCAP_KF_LOOKUP_GEN_IDX_SEL]             =  "LOOKUP_GEN_IDX_SEL",
+	[VCAP_KF_LOOKUP_INDEX]                   =  "LOOKUP_INDEX",
 	[VCAP_KF_LOOKUP_PAG]                     =  "LOOKUP_PAG",
+	[VCAP_KF_MIRROR_PROBE]                   =  "MIRROR_PROBE",
 	[VCAP_KF_OAM_CCM_CNTS_EQ0]               =  "OAM_CCM_CNTS_EQ0",
 	[VCAP_KF_OAM_DETECTED]                   =  "OAM_DETECTED",
 	[VCAP_KF_OAM_FLAGS]                      =  "OAM_FLAGS",
@@ -1549,7 +2873,12 @@ static const char * const vcap_keyfield_names[] = {
 	[VCAP_KF_OAM_OPCODE]                     =  "OAM_OPCODE",
 	[VCAP_KF_OAM_VER]                        =  "OAM_VER",
 	[VCAP_KF_OAM_Y1731_IS]                   =  "OAM_Y1731_IS",
+	[VCAP_KF_PROT_ACTIVE]                    =  "PROT_ACTIVE",
+	[VCAP_KF_RT_FRMID]                       =  "RT_FRMID",
+	[VCAP_KF_RT_TYPE]                        =  "RT_TYPE",
+	[VCAP_KF_RT_VLAN_IDX]                    =  "RT_VLAN_IDX",
 	[VCAP_KF_TCP_IS]                         =  "TCP_IS",
+	[VCAP_KF_TCP_UDP_IS]                     =  "TCP_UDP_IS",
 	[VCAP_KF_TYPE]                           =  "TYPE",
 };
 
@@ -1557,24 +2886,95 @@ static const char * const vcap_keyfield_names[] = {
 static const char * const vcap_actionfield_names[] = {
 	[VCAP_AF_NO_VALUE]                       =  "(None)",
 	[VCAP_AF_ACL_ID]                         =  "ACL_ID",
+	[VCAP_AF_CLS_VID_SEL]                    =  "CLS_VID_SEL",
+	[VCAP_AF_CNT_ID]                         =  "CNT_ID",
+	[VCAP_AF_COPY_PORT_NUM]                  =  "COPY_PORT_NUM",
+	[VCAP_AF_COPY_QUEUE_NUM]                 =  "COPY_QUEUE_NUM",
 	[VCAP_AF_CPU_COPY_ENA]                   =  "CPU_COPY_ENA",
 	[VCAP_AF_CPU_QUEUE_NUM]                  =  "CPU_QUEUE_NUM",
+	[VCAP_AF_CUSTOM_ACE_TYPE_ENA]            =  "CUSTOM_ACE_TYPE_ENA",
+	[VCAP_AF_DEI_ENA]                        =  "DEI_ENA",
+	[VCAP_AF_DEI_VAL]                        =  "DEI_VAL",
+	[VCAP_AF_DLR_SEL]                        =  "DLR_SEL",
+	[VCAP_AF_DP_ENA]                         =  "DP_ENA",
+	[VCAP_AF_DP_VAL]                         =  "DP_VAL",
+	[VCAP_AF_DSCP_ENA]                       =  "DSCP_ENA",
+	[VCAP_AF_DSCP_VAL]                       =  "DSCP_VAL",
+	[VCAP_AF_ES2_REW_CMD]                    =  "ES2_REW_CMD",
 	[VCAP_AF_FWD_KILL_ENA]                   =  "FWD_KILL_ENA",
+	[VCAP_AF_FWD_MODE]                       =  "FWD_MODE",
 	[VCAP_AF_HIT_ME_ONCE]                    =  "HIT_ME_ONCE",
 	[VCAP_AF_HOST_MATCH]                     =  "HOST_MATCH",
+	[VCAP_AF_IGNORE_PIPELINE_CTRL]           =  "IGNORE_PIPELINE_CTRL",
+	[VCAP_AF_INTR_ENA]                       =  "INTR_ENA",
+	[VCAP_AF_ISDX_ADD_REPLACE_SEL]           =  "ISDX_ADD_REPLACE_SEL",
+	[VCAP_AF_ISDX_ADD_VAL]                   =  "ISDX_ADD_VAL",
 	[VCAP_AF_ISDX_ENA]                       =  "ISDX_ENA",
+	[VCAP_AF_ISDX_REPLACE_ENA]               =  "ISDX_REPLACE_ENA",
+	[VCAP_AF_ISDX_VAL]                       =  "ISDX_VAL",
 	[VCAP_AF_LRN_DIS]                        =  "LRN_DIS",
+	[VCAP_AF_MAP_IDX]                        =  "MAP_IDX",
+	[VCAP_AF_MAP_KEY]                        =  "MAP_KEY",
+	[VCAP_AF_MAP_LOOKUP_SEL]                 =  "MAP_LOOKUP_SEL",
 	[VCAP_AF_MASK_MODE]                      =  "MASK_MODE",
+	[VCAP_AF_MATCH_ID]                       =  "MATCH_ID",
+	[VCAP_AF_MATCH_ID_MASK]                  =  "MATCH_ID_MASK",
 	[VCAP_AF_MIRROR_ENA]                     =  "MIRROR_ENA",
+	[VCAP_AF_MIRROR_PROBE]                   =  "MIRROR_PROBE",
+	[VCAP_AF_MIRROR_PROBE_ID]                =  "MIRROR_PROBE_ID",
+	[VCAP_AF_MRP_SEL]                        =  "MRP_SEL",
+	[VCAP_AF_NXT_IDX]                        =  "NXT_IDX",
+	[VCAP_AF_NXT_IDX_CTRL]                   =  "NXT_IDX_CTRL",
+	[VCAP_AF_OAM_SEL]                        =  "OAM_SEL",
+	[VCAP_AF_PAG_OVERRIDE_MASK]              =  "PAG_OVERRIDE_MASK",
+	[VCAP_AF_PAG_VAL]                        =  "PAG_VAL",
+	[VCAP_AF_PCP_ENA]                        =  "PCP_ENA",
+	[VCAP_AF_PCP_VAL]                        =  "PCP_VAL",
+	[VCAP_AF_PIPELINE_FORCE_ENA]             =  "PIPELINE_FORCE_ENA",
+	[VCAP_AF_PIPELINE_PT]                    =  "PIPELINE_PT",
 	[VCAP_AF_POLICE_ENA]                     =  "POLICE_ENA",
 	[VCAP_AF_POLICE_IDX]                     =  "POLICE_IDX",
+	[VCAP_AF_POLICE_REMARK]                  =  "POLICE_REMARK",
 	[VCAP_AF_POLICE_VCAP_ONLY]               =  "POLICE_VCAP_ONLY",
 	[VCAP_AF_PORT_MASK]                      =  "PORT_MASK",
+	[VCAP_AF_QOS_ENA]                        =  "QOS_ENA",
+	[VCAP_AF_QOS_VAL]                        =  "QOS_VAL",
 	[VCAP_AF_REW_OP]                         =  "REW_OP",
+	[VCAP_AF_RT_DIS]                         =  "RT_DIS",
+	[VCAP_AF_SFID_ENA]                       =  "SFID_ENA",
+	[VCAP_AF_SFID_VAL]                       =  "SFID_VAL",
+	[VCAP_AF_SGID_ENA]                       =  "SGID_ENA",
+	[VCAP_AF_SGID_VAL]                       =  "SGID_VAL",
+	[VCAP_AF_TYPE]                           =  "TYPE",
+	[VCAP_AF_VID_REPLACE_ENA]                =  "VID_REPLACE_ENA",
+	[VCAP_AF_VID_VAL]                        =  "VID_VAL",
+	[VCAP_AF_VLAN_POP_CNT]                   =  "VLAN_POP_CNT",
+	[VCAP_AF_VLAN_POP_CNT_ENA]               =  "VLAN_POP_CNT_ENA",
 };
 
 /* VCAPs */
 const struct vcap_info lan966x_vcaps[] = {
+	[VCAP_TYPE_IS1] = {
+		.name = "is1",
+		.rows = 192,
+		.sw_count = 4,
+		.sw_width = 96,
+		.sticky_width = 32,
+		.act_width = 123,
+		.default_cnt = 0,
+		.require_cnt_dis = 1,
+		.version = 1,
+		.keyfield_set = is1_keyfield_set,
+		.keyfield_set_size = ARRAY_SIZE(is1_keyfield_set),
+		.actionfield_set = is1_actionfield_set,
+		.actionfield_set_size = ARRAY_SIZE(is1_actionfield_set),
+		.keyfield_set_map = is1_keyfield_set_map,
+		.keyfield_set_map_size = is1_keyfield_set_map_size,
+		.actionfield_set_map = is1_actionfield_set_map,
+		.actionfield_set_map_size = is1_actionfield_set_map_size,
+		.keyfield_set_typegroups = is1_keyfield_set_typegroups,
+		.actionfield_set_typegroups = is1_actionfield_set_typegroups,
+	},
 	[VCAP_TYPE_IS2] = {
 		.name = "is2",
 		.rows = 64,
@@ -1600,7 +3000,7 @@ const struct vcap_info lan966x_vcaps[] = {
 
 const struct vcap_statistics lan966x_vcap_stats = {
 	.name = "lan966x",
-	.count = 1,
+	.count = 2,
 	.keyfield_set_names = vcap_keyfield_set_names,
 	.actionfield_set_names = vcap_actionfield_set_names,
 	.keyfield_names = vcap_keyfield_names,
diff --git a/drivers/net/ethernet/microchip/vcap/vcap_ag_api.h b/drivers/net/ethernet/microchip/vcap/vcap_ag_api.h
index 0844fcaeee68..a556c4419986 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_ag_api.h
+++ b/drivers/net/ethernet/microchip/vcap/vcap_ag_api.h
@@ -3,8 +3,8 @@
  * Microchip VCAP API
  */
 
-/* This file is autogenerated by cml-utils 2023-02-10 11:15:56 +0100.
- * Commit ID: c30fb4bf0281cd4a7133bdab6682f9e43c872ada
+/* This file is autogenerated by cml-utils 2023-02-16 11:41:14 +0100.
+ * Commit ID: be85f176b3a151fa748dcaf97c8824a5c2e065f3
  */
 
 #ifndef __VCAP_AG_API__
@@ -14,6 +14,7 @@ enum vcap_type {
 	VCAP_TYPE_ES0,
 	VCAP_TYPE_ES2,
 	VCAP_TYPE_IS0,
+	VCAP_TYPE_IS1,
 	VCAP_TYPE_IS2,
 	VCAP_TYPE_MAX
 };
@@ -21,7 +22,12 @@ enum vcap_type {
 /* Keyfieldset names with origin information */
 enum vcap_keyfield_set {
 	VCAP_KFS_NO_VALUE,          /* initial value */
+	VCAP_KFS_5TUPLE_IP4,        /* lan966x is1 X2 */
+	VCAP_KFS_5TUPLE_IP6,        /* lan966x is1 X4 */
+	VCAP_KFS_7TUPLE,            /* lan966x is1 X4 */
 	VCAP_KFS_ARP,               /* sparx5 is2 X6, sparx5 es2 X6, lan966x is2 X2 */
+	VCAP_KFS_DBL_VID,           /* lan966x is1 X1 */
+	VCAP_KFS_DMAC_VID,          /* lan966x is1 X1 */
 	VCAP_KFS_ETAG,              /* sparx5 is0 X2 */
 	VCAP_KFS_IP4_OTHER,         /* sparx5 is2 X6, sparx5 es2 X6, lan966x is2 X2 */
 	VCAP_KFS_IP4_TCP_UDP,       /* sparx5 is2 X6, sparx5 es2 X6, lan966x is2 X2 */
@@ -36,10 +42,13 @@ enum vcap_keyfield_set {
 	VCAP_KFS_MAC_ETYPE,         /* sparx5 is2 X6, sparx5 es2 X6, lan966x is2 X2 */
 	VCAP_KFS_MAC_LLC,           /* lan966x is2 X2 */
 	VCAP_KFS_MAC_SNAP,          /* lan966x is2 X2 */
+	VCAP_KFS_NORMAL,            /* lan966x is1 X2 */
 	VCAP_KFS_NORMAL_5TUPLE_IP4,  /* sparx5 is0 X6 */
 	VCAP_KFS_NORMAL_7TUPLE,     /* sparx5 is0 X12 */
+	VCAP_KFS_NORMAL_IP6,        /* lan966x is1 X4 */
 	VCAP_KFS_OAM,               /* lan966x is2 X2 */
 	VCAP_KFS_PURE_5TUPLE_IP4,   /* sparx5 is0 X3 */
+	VCAP_KFS_RT,                /* lan966x is1 X1 */
 	VCAP_KFS_SMAC_SIP4,         /* lan966x is2 X1 */
 	VCAP_KFS_SMAC_SIP6,         /* lan966x is2 X2 */
 };
@@ -61,17 +70,20 @@ enum vcap_keyfield_set {
  *   Used by 802.1BR Bridge Port Extension in an E-Tag
  * VCAP_KF_8021BR_IGR_ECID_EXT: W8, sparx5: is0
  *   Used by 802.1BR Bridge Port Extension in an E-Tag
- * VCAP_KF_8021Q_DEI0: W1, sparx5: is0
+ * VCAP_KF_8021CB_R_TAGGED_IS: W1, lan966x: is1
+ *   Set if frame contains an RTAG: IEEE 802.1CB (FRER Redundancy tag, Ethertype
+ *   0xf1c1)
+ * VCAP_KF_8021Q_DEI0: W1, sparx5: is0, lan966x: is1
  *   First DEI in multiple vlan tags (outer tag or default port tag)
- * VCAP_KF_8021Q_DEI1: W1, sparx5: is0
+ * VCAP_KF_8021Q_DEI1: W1, sparx5: is0, lan966x: is1
  *   Second DEI in multiple vlan tags (inner tag)
  * VCAP_KF_8021Q_DEI2: W1, sparx5: is0
  *   Third DEI in multiple vlan tags (not always available)
  * VCAP_KF_8021Q_DEI_CLS: W1, sparx5: is2/es2, lan966x: is2
  *   Classified DEI
- * VCAP_KF_8021Q_PCP0: W3, sparx5: is0
+ * VCAP_KF_8021Q_PCP0: W3, sparx5: is0, lan966x: is1
  *   First PCP in multiple vlan tags (outer tag or default port tag)
- * VCAP_KF_8021Q_PCP1: W3, sparx5: is0
+ * VCAP_KF_8021Q_PCP1: W3, sparx5: is0, lan966x: is1
  *   Second PCP in multiple vlan tags (inner tag)
  * VCAP_KF_8021Q_PCP2: W3, sparx5: is0
  *   Third PCP in multiple vlan tags (not always available)
@@ -79,22 +91,24 @@ enum vcap_keyfield_set {
  *   Classified PCP
  * VCAP_KF_8021Q_TPID: W3, sparx5: es0
  *   TPID for outer tag: 0: Customer TPID 1: Service TPID (88A8 or programmable)
- * VCAP_KF_8021Q_TPID0: W3, sparx5: is0
+ * VCAP_KF_8021Q_TPID0: sparx5 is0 W3, lan966x is1 W1
  *   First TPIC in multiple vlan tags (outer tag or default port tag)
- * VCAP_KF_8021Q_TPID1: W3, sparx5: is0
+ * VCAP_KF_8021Q_TPID1: sparx5 is0 W3, lan966x is1 W1
  *   Second TPID in multiple vlan tags (inner tag)
  * VCAP_KF_8021Q_TPID2: W3, sparx5: is0
  *   Third TPID in multiple vlan tags (not always available)
- * VCAP_KF_8021Q_VID0: W12, sparx5: is0
+ * VCAP_KF_8021Q_VID0: W12, sparx5: is0, lan966x: is1
  *   First VID in multiple vlan tags (outer tag or default port tag)
- * VCAP_KF_8021Q_VID1: W12, sparx5: is0
+ * VCAP_KF_8021Q_VID1: W12, sparx5: is0, lan966x: is1
  *   Second VID in multiple vlan tags (inner tag)
  * VCAP_KF_8021Q_VID2: W12, sparx5: is0
  *   Third VID in multiple vlan tags (not always available)
  * VCAP_KF_8021Q_VID_CLS: sparx5 is2 W13, sparx5 es0 W13, sparx5 es2 W13,
  *   lan966x is2 W12
  *   Classified VID
- * VCAP_KF_8021Q_VLAN_TAGGED_IS: W1, sparx5: is2/es2, lan966x: is2
+ * VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS: W1, lan966x: is1
+ *   Set if frame has two or more Q-tags. Independent of port VLAN awareness
+ * VCAP_KF_8021Q_VLAN_TAGGED_IS: W1, sparx5: is2/es2, lan966x: is1/is2
  *   Sparx5: Set if frame was received with a VLAN tag, LAN966x: Set if frame has
  *   one or more Q-tags. Independent of port VLAN awareness
  * VCAP_KF_8021Q_VLAN_TAGS: W3, sparx5: is0
@@ -120,9 +134,9 @@ enum vcap_keyfield_set {
  *   Class of service
  * VCAP_KF_ES0_ISDX_KEY_ENA: W1, sparx5: es2
  *   The value taken from the IFH .FWD.ES0_ISDX_KEY_ENA
- * VCAP_KF_ETYPE: W16, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_ETYPE: W16, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Ethernet type
- * VCAP_KF_ETYPE_LEN_IS: W1, sparx5: is0/is2/es2
+ * VCAP_KF_ETYPE_LEN_IS: W1, sparx5: is0/is2/es2, lan966x: is1
  *   Set if frame has EtherType >= 0x600
  * VCAP_KF_HOST_MATCH: W1, lan966x: is2
  *   The action from the SMAC_SIP4 or SMAC_SIP6 lookups. Used for IP source
@@ -134,11 +148,12 @@ enum vcap_keyfield_set {
  *   CPU queue)
  * VCAP_KF_IF_EGR_PORT_NO: W7, sparx5: es0
  *   Egress port number
- * VCAP_KF_IF_IGR_PORT: sparx5 is0 W7, sparx5 es2 W9, lan966x is2 W4
+ * VCAP_KF_IF_IGR_PORT: sparx5 is0 W7, sparx5 es2 W9, lan966x is1 W3, lan966x
+ *   is2 W4
  *   Sparx5: Logical ingress port number retrieved from
  *   ANA_CL::PORT_ID_CFG.LPORT_NUM or ERLEG, LAN966x: ingress port nunmber
  * VCAP_KF_IF_IGR_PORT_MASK: sparx5 is0 W65, sparx5 is2 W32, sparx5 is2 W65,
- *   lan966x is2 W9
+ *   lan966x is1 W9, lan966x is2 W9
  *   Ingress port mask, one bit per port/erleg
  * VCAP_KF_IF_IGR_PORT_MASK_L3: W1, sparx5: is2
  *   If set, IF_IGR_PORT_MASK, IF_IGR_PORT_MASK_RNG, and IF_IGR_PORT_MASK_SEL are
@@ -151,24 +166,26 @@ enum vcap_keyfield_set {
  *   Mapping: 0: DEFAULT 1: LOOPBACK 2: MASQUERADE 3: CPU_VD
  * VCAP_KF_IF_IGR_PORT_SEL: W1, sparx5: es2
  *   Selector for IF_IGR_PORT: physical port number or ERLEG
- * VCAP_KF_IP4_IS: W1, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_IP4_IS: W1, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Set if frame has EtherType = 0x800 and IP version = 4
- * VCAP_KF_IP_MC_IS: W1, sparx5: is0
+ * VCAP_KF_IP_MC_IS: W1, sparx5: is0, lan966x: is1
  *   Set if frame is IPv4 frame and frame's destination MAC address is an IPv4
  *   multicast address (0x01005E0 /25). Set if frame is IPv6 frame and frame's
  *   destination MAC address is an IPv6 multicast address (0x3333/16).
- * VCAP_KF_IP_PAYLOAD_5TUPLE: W32, sparx5: is0
+ * VCAP_KF_IP_PAYLOAD_5TUPLE: W32, sparx5: is0, lan966x: is1
  *   Payload bytes after IP header
- * VCAP_KF_IP_SNAP_IS: W1, sparx5: is0
+ * VCAP_KF_IP_PAYLOAD_S1_IP6: W112, lan966x: is1
+ *   Payload after IPv6 header
+ * VCAP_KF_IP_SNAP_IS: W1, sparx5: is0, lan966x: is1
  *   Set if frame is IPv4, IPv6, or SNAP frame
  * VCAP_KF_ISDX_CLS: W12, sparx5: is2/es0/es2
  *   Classified ISDX
  * VCAP_KF_ISDX_GT0_IS: W1, sparx5: is2/es0/es2, lan966x: is2
  *   Set if classified ISDX > 0
- * VCAP_KF_L2_BC_IS: W1, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L2_BC_IS: W1, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Set if frame's destination MAC address is the broadcast address
  *   (FF-FF-FF-FF-FF-FF).
- * VCAP_KF_L2_DMAC: W48, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L2_DMAC: W48, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Destination MAC address
  * VCAP_KF_L2_FRM_TYPE: W4, lan966x: is2
  *   Frame subtype for specific EtherTypes (MRP, DLR)
@@ -176,7 +193,9 @@ enum vcap_keyfield_set {
  *   Set if the frame is allowed to be forwarded to front ports
  * VCAP_KF_L2_LLC: W40, lan966x: is2
  *   LLC header and data after up to two VLAN tags and the type/length field
- * VCAP_KF_L2_MC_IS: W1, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L2_MAC: W48, lan966x: is1
+ *   MAC address (FIRST=1: SMAC, FIRST=0: DMAC)
+ * VCAP_KF_L2_MC_IS: W1, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Set if frame's destination MAC address is a multicast address (bit 40 = 1).
  * VCAP_KF_L2_PAYLOAD0: W16, lan966x: is2
  *   Payload bytes 0-1 after the frame's EtherType
@@ -188,7 +207,7 @@ enum vcap_keyfield_set {
  *   specifically for PTP frames.
  * VCAP_KF_L2_PAYLOAD_ETYPE: W64, sparx5: is2/es2
  *   Byte 0-7 of L2 payload after Type/Len field and overloading for OAM
- * VCAP_KF_L2_SMAC: W48, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L2_SMAC: W48, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Source MAC address
  * VCAP_KF_L2_SNAP: W40, lan966x: is2
  *   SNAP header after LLC header (AA-AA-03)
@@ -196,32 +215,38 @@ enum vcap_keyfield_set {
  *   Set if Src IP matches Dst IP address
  * VCAP_KF_L3_DPL_CLS: W1, sparx5: es0/es2
  *   The frames drop precedence level
- * VCAP_KF_L3_DSCP: W6, sparx5: is0
+ * VCAP_KF_L3_DSCP: W6, sparx5: is0, lan966x: is1
  *   Frame's DSCP value
  * VCAP_KF_L3_DST_IS: W1, sparx5: is2
  *   Set if lookup is done for egress router leg
- * VCAP_KF_L3_FRAGMENT: W1, lan966x: is2
+ * VCAP_KF_L3_FRAGMENT: W1, lan966x: is1/is2
  *   Set if IPv4 frame is fragmented
  * VCAP_KF_L3_FRAGMENT_TYPE: W2, sparx5: is0/is2/es2
  *   L3 Fragmentation type (none, initial, suspicious, valid follow up)
  * VCAP_KF_L3_FRAG_INVLD_L4_LEN: W1, sparx5: is0/is2
  *   Set if frame's L4 length is less than ANA_CL:COMMON:CLM_FRAGMENT_CFG.L4_MIN_L
  *   EN
- * VCAP_KF_L3_FRAG_OFS_GT0: W1, lan966x: is2
+ * VCAP_KF_L3_FRAG_OFS_GT0: W1, lan966x: is1/is2
  *   Set if IPv4 frame is fragmented and it is not the first fragment
- * VCAP_KF_L3_IP4_DIP: W32, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L3_IP4_DIP: W32, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Destination IPv4 Address
- * VCAP_KF_L3_IP4_SIP: W32, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L3_IP4_SIP: W32, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Source IPv4 Address
- * VCAP_KF_L3_IP6_DIP: W128, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L3_IP6_DIP: sparx5 is0 W128, sparx5 is2 W128, sparx5 es2 W128,
+ *   lan966x is1 W64, lan966x is1 W128, lan966x is2 W128
  *   Sparx5: Full IPv6 DIP, LAN966x: Either Full IPv6 DIP or a subset depending on
  *   frame type
- * VCAP_KF_L3_IP6_SIP: W128, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L3_IP6_DIP_MSB: W16, lan966x: is1
+ *   MS 16bits of IPv6 DIP
+ * VCAP_KF_L3_IP6_SIP: sparx5 is0 W128, sparx5 is2 W128, sparx5 es2 W128,
+ *   lan966x is1 W128, lan966x is1 W64, lan966x is2 W128
  *   Sparx5: Full IPv6 SIP, LAN966x: Either Full IPv6 SIP or a subset depending on
  *   frame type
- * VCAP_KF_L3_IP_PROTO: W8, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L3_IP6_SIP_MSB: W16, lan966x: is1
+ *   MS 16bits of IPv6 DIP
+ * VCAP_KF_L3_IP_PROTO: W8, sparx5: is0/is2/es2, lan966x: is1/is2
  *   IPv4 frames: IP protocol. IPv6 frames: Next header, same as for IPV4
- * VCAP_KF_L3_OPTIONS_IS: W1, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L3_OPTIONS_IS: W1, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Set if IPv4 frame contains options (IP len > 5)
  * VCAP_KF_L3_PAYLOAD: sparx5 is2 W96, sparx5 is2 W40, sparx5 es2 W96, sparx5
  *   es2 W40, lan966x is2 W56
@@ -254,7 +279,8 @@ enum vcap_keyfield_set {
  * VCAP_KF_L4_PSH: W1, sparx5: is2/es2, lan966x: is2
  *   Sparx5: TCP flag PSH, LAN966x: TCP: TCP flag PSH. PTP over UDP: flagField bit
  *   1 (twoStepFlag)
- * VCAP_KF_L4_RNG: sparx5 is0 W8, sparx5 is2 W16, sparx5 es2 W16, lan966x is2 W8
+ * VCAP_KF_L4_RNG: sparx5 is0 W8, sparx5 is2 W16, sparx5 es2 W16, lan966x is1
+ *   W8, lan966x is2 W8
  *   Range checker bitmask (one for each range checker). Input into range checkers
  *   is taken from classified results (VID, DSCP) and frame (SPORT, DPORT, ETYPE,
  *   outer VID, inner VID)
@@ -264,7 +290,7 @@ enum vcap_keyfield_set {
  * VCAP_KF_L4_SEQUENCE_EQ0_IS: W1, sparx5: is2/es2, lan966x: is2
  *   Set if TCP sequence number is 0, LAN966x: Overlayed with PTP over UDP:
  *   messageType bit 0
- * VCAP_KF_L4_SPORT: W16, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_L4_SPORT: W16, sparx5: is0/is2/es2, lan966x: is1/is2
  *   TCP/UDP source port
  * VCAP_KF_L4_SPORT_EQ_DPORT_IS: W1, sparx5: is2/es2, lan966x: is2
  *   Set if UDP or TCP source port equals UDP or TCP destination port
@@ -274,13 +300,16 @@ enum vcap_keyfield_set {
  * VCAP_KF_L4_URG: W1, sparx5: is2/es2, lan966x: is2
  *   Sparx5: TCP flag URG, LAN966x: TCP: TCP flag URG. PTP over UDP: flagField bit
  *   7 (reserved)
- * VCAP_KF_LOOKUP_FIRST_IS: W1, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_LOOKUP_FIRST_IS: W1, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Selects between entries relevant for first and second lookup. Set for first
  *   lookup, cleared for second lookup.
  * VCAP_KF_LOOKUP_GEN_IDX: W12, sparx5: is0
  *   Generic index - for chaining CLM instances
  * VCAP_KF_LOOKUP_GEN_IDX_SEL: W2, sparx5: is0
  *   Select the mode of the Generic Index
+ * VCAP_KF_LOOKUP_INDEX: W2, lan966x: is1
+ *   0: First lookup, 1: Second lookup, 2: Third lookup, Similar to VCAP_KF_FIRST
+ *   but with extra info
  * VCAP_KF_LOOKUP_PAG: W8, sparx5: is2, lan966x: is2
  *   Classified Policy Association Group: chains rules from IS1/CLM to IS2
  * VCAP_KF_MIRROR_PROBE: W2, sparx5: es2
@@ -303,14 +332,22 @@ enum vcap_keyfield_set {
  *   Set if frame's EtherType = 0x8902
  * VCAP_KF_PROT_ACTIVE: W1, sparx5: es0/es2
  *   Protection is active
- * VCAP_KF_TCP_IS: W1, sparx5: is0/is2/es2, lan966x: is2
+ * VCAP_KF_RT_FRMID: W32, lan966x: is1
+ *   Profinet or OPC-UA FrameId
+ * VCAP_KF_RT_TYPE: W2, lan966x: is1
+ *   Encoding of frame's EtherType: 0: Other, 1: Profinet, 2: OPC-UA, 3: Custom
+ *   (ANA::RT_CUSTOM)
+ * VCAP_KF_RT_VLAN_IDX: W3, lan966x: is1
+ *   Real-time VLAN index from ANA::RT_VLAN_PCP
+ * VCAP_KF_TCP_IS: W1, sparx5: is0/is2/es2, lan966x: is1/is2
  *   Set if frame is IPv4 TCP frame (IP protocol = 6) or IPv6 TCP frames (Next
  *   header = 6)
- * VCAP_KF_TCP_UDP_IS: W1, sparx5: is0/is2/es2
+ * VCAP_KF_TCP_UDP_IS: W1, sparx5: is0/is2/es2, lan966x: is1
  *   Set if frame is IPv4/IPv6 TCP or UDP frame (IP protocol/next header equals 6
  *   or 17)
  * VCAP_KF_TYPE: sparx5 is0 W2, sparx5 is0 W1, sparx5 is2 W4, sparx5 is2 W2,
- *   sparx5 es0 W1, sparx5 es2 W3, lan966x is2 W4, lan966x is2 W2
+ *   sparx5 es0 W1, sparx5 es2 W3, lan966x is1 W1, lan966x is1 W2, lan966x is2 W4,
+ *   lan966x is2 W2
  *   Keyset type id - set by the API
  */
 
@@ -323,6 +360,7 @@ enum vcap_key_field {
 	VCAP_KF_8021BR_GRP,
 	VCAP_KF_8021BR_IGR_ECID_BASE,
 	VCAP_KF_8021BR_IGR_ECID_EXT,
+	VCAP_KF_8021CB_R_TAGGED_IS,
 	VCAP_KF_8021Q_DEI0,
 	VCAP_KF_8021Q_DEI1,
 	VCAP_KF_8021Q_DEI2,
@@ -339,6 +377,7 @@ enum vcap_key_field {
 	VCAP_KF_8021Q_VID1,
 	VCAP_KF_8021Q_VID2,
 	VCAP_KF_8021Q_VID_CLS,
+	VCAP_KF_8021Q_VLAN_DBL_TAGGED_IS,
 	VCAP_KF_8021Q_VLAN_TAGGED_IS,
 	VCAP_KF_8021Q_VLAN_TAGS,
 	VCAP_KF_ACL_GRP_ID,
@@ -366,6 +405,7 @@ enum vcap_key_field {
 	VCAP_KF_IP4_IS,
 	VCAP_KF_IP_MC_IS,
 	VCAP_KF_IP_PAYLOAD_5TUPLE,
+	VCAP_KF_IP_PAYLOAD_S1_IP6,
 	VCAP_KF_IP_SNAP_IS,
 	VCAP_KF_ISDX_CLS,
 	VCAP_KF_ISDX_GT0_IS,
@@ -374,6 +414,7 @@ enum vcap_key_field {
 	VCAP_KF_L2_FRM_TYPE,
 	VCAP_KF_L2_FWD_IS,
 	VCAP_KF_L2_LLC,
+	VCAP_KF_L2_MAC,
 	VCAP_KF_L2_MC_IS,
 	VCAP_KF_L2_PAYLOAD0,
 	VCAP_KF_L2_PAYLOAD1,
@@ -392,7 +433,9 @@ enum vcap_key_field {
 	VCAP_KF_L3_IP4_DIP,
 	VCAP_KF_L3_IP4_SIP,
 	VCAP_KF_L3_IP6_DIP,
+	VCAP_KF_L3_IP6_DIP_MSB,
 	VCAP_KF_L3_IP6_SIP,
+	VCAP_KF_L3_IP6_SIP_MSB,
 	VCAP_KF_L3_IP_PROTO,
 	VCAP_KF_L3_OPTIONS_IS,
 	VCAP_KF_L3_PAYLOAD,
@@ -416,6 +459,7 @@ enum vcap_key_field {
 	VCAP_KF_LOOKUP_FIRST_IS,
 	VCAP_KF_LOOKUP_GEN_IDX,
 	VCAP_KF_LOOKUP_GEN_IDX_SEL,
+	VCAP_KF_LOOKUP_INDEX,
 	VCAP_KF_LOOKUP_PAG,
 	VCAP_KF_MIRROR_PROBE,
 	VCAP_KF_OAM_CCM_CNTS_EQ0,
@@ -427,6 +471,9 @@ enum vcap_key_field {
 	VCAP_KF_OAM_VER,
 	VCAP_KF_OAM_Y1731_IS,
 	VCAP_KF_PROT_ACTIVE,
+	VCAP_KF_RT_FRMID,
+	VCAP_KF_RT_TYPE,
+	VCAP_KF_RT_VLAN_IDX,
 	VCAP_KF_TCP_IS,
 	VCAP_KF_TCP_UDP_IS,
 	VCAP_KF_TYPE,
@@ -440,6 +487,7 @@ enum vcap_actionfield_set {
 	VCAP_AFS_CLASS_REDUCED,     /* sparx5 is0 X1 */
 	VCAP_AFS_ES0,               /* sparx5 es0 X1 */
 	VCAP_AFS_FULL,              /* sparx5 is0 X3 */
+	VCAP_AFS_S1,                /* lan966x is1 X1 */
 	VCAP_AFS_SMAC_SIP,          /* lan966x is2 X1 */
 };
 
@@ -470,23 +518,31 @@ enum vcap_actionfield_set {
  *   CPU extraction queue. Used when FWD_SEL >0 and PIPELINE_ACT = XTR.
  * VCAP_AF_CPU_QUEUE_NUM: W3, sparx5: is2/es2, lan966x: is2
  *   CPU queue number. Used when CPU_COPY_ENA is set.
+ * VCAP_AF_CUSTOM_ACE_TYPE_ENA: W4, lan966x: is1
+ *   Enables use of custom keys in IS2. Bits 3:2 control second lookup in IS2
+ *   while bits 1:0 control first lookup. Encoding per lookup: 0: Disabled.  1:
+ *   Extract 40 bytes after position corresponding to the location of the IPv4
+ *   header and use as key.  2: Extract 40 bytes after SMAC and use as key
  * VCAP_AF_DEI_A_VAL: W1, sparx5: es0
  *   DEI used in ES0 tag A. See TAG_A_DEI_SEL.
  * VCAP_AF_DEI_B_VAL: W1, sparx5: es0
  *   DEI used in ES0 tag B. See TAG_B_DEI_SEL.
  * VCAP_AF_DEI_C_VAL: W1, sparx5: es0
  *   DEI used in ES0 tag C. See TAG_C_DEI_SEL.
- * VCAP_AF_DEI_ENA: W1, sparx5: is0
+ * VCAP_AF_DEI_ENA: W1, sparx5: is0, lan966x: is1
  *   If set, use DEI_VAL as classified DEI value. Otherwise, DEI from basic
  *   classification is used
- * VCAP_AF_DEI_VAL: W1, sparx5: is0
+ * VCAP_AF_DEI_VAL: W1, sparx5: is0, lan966x: is1
  *   See DEI_ENA
- * VCAP_AF_DP_ENA: W1, sparx5: is0
+ * VCAP_AF_DLR_SEL: W2, lan966x: is1
+ *   0: No changes to port-based selection in ANA:PORT:OAM_CFG.DLR_ENA.  1: Enable
+ *   DLR frame processing 2: Disable DLR processing
+ * VCAP_AF_DP_ENA: W1, sparx5: is0, lan966x: is1
  *   If set, use DP_VAL as classified drop precedence level. Otherwise, drop
  *   precedence level from basic classification is used.
- * VCAP_AF_DP_VAL: W2, sparx5: is0
+ * VCAP_AF_DP_VAL: sparx5 is0 W2, lan966x is1 W1
  *   See DP_ENA.
- * VCAP_AF_DSCP_ENA: W1, sparx5: is0
+ * VCAP_AF_DSCP_ENA: W1, sparx5: is0, lan966x: is1
  *   If set, use DSCP_VAL as classified DSCP value. Otherwise, DSCP value from
  *   basic classification is used.
  * VCAP_AF_DSCP_SEL: W3, sparx5: es0
@@ -495,7 +551,7 @@ enum vcap_actionfield_set {
  *   table 0, otherwise use DSCP_VAL. 5: Mapped using mapping table 1, otherwise
  *   use mapping table 0. 6: Mapped using mapping table 2, otherwise use DSCP_VAL.
  *   7: Mapped using mapping table 3, otherwise use mapping table 2
- * VCAP_AF_DSCP_VAL: W6, sparx5: is0/es0
+ * VCAP_AF_DSCP_VAL: W6, sparx5: is0/es0, lan966x: is1
  *   See DSCP_ENA.
  * VCAP_AF_ES2_REW_CMD: W3, sparx5: es2
  *   Command forwarded to REW: 0: No action. 1: SWAP MAC addresses. 2: Do L2CP
@@ -529,9 +585,16 @@ enum vcap_actionfield_set {
  * VCAP_AF_ISDX_ADD_REPLACE_SEL: W1, sparx5: is0
  *   Controls the classified ISDX. 0: New ISDX = old ISDX + ISDX_VAL. 1: New ISDX
  *   = ISDX_VAL.
+ * VCAP_AF_ISDX_ADD_VAL: W8, lan966x: is1
+ *   If ISDX_REPLACE_ENA is set, ISDX_ADD_VAL is used directly as the new ISDX.
+ *   Encoding: ISDX_REPLACE_ENA=0, ISDX_ADD_VAL=0: Disabled ISDX_EPLACE_ENA=0,
+ *   ISDX_ADD_VAL>0: Add value to classified ISDX. ISDX_REPLACE_ENA=1: Replace
+ *   with ISDX_ADD_VAL value.
  * VCAP_AF_ISDX_ENA: W1, lan966x: is2
  *   Setting this bit to 1 causes the classified ISDX to be set to the value of
  *   POLICE_IDX[8:0].
+ * VCAP_AF_ISDX_REPLACE_ENA: W1, lan966x: is1
+ *   If set, classified ISDX is set to ISDX_ADD_VAL.
  * VCAP_AF_ISDX_VAL: W12, sparx5: is0
  *   See isdx_add_replace_sel
  * VCAP_AF_LOOP_ENA: W1, sparx5: es0
@@ -572,14 +635,22 @@ enum vcap_actionfield_set {
  * VCAP_AF_MIRROR_PROBE_ID: W2, sparx5: es2
  *   Signals a mirror probe to be placed in the IFH. Only possible when FWD_MODE
  *   is copy. 0: No mirroring. 1-3: Use mirror probe 0-2.
+ * VCAP_AF_MRP_SEL: W2, lan966x: is1
+ *   0: No changes to port-based selection in ANA:PORT:OAM_CFG.MRP_ENA.  1: Enable
+ *   MRP frame processing 2: Disable MRP processing
  * VCAP_AF_NXT_IDX: W12, sparx5: is0
  *   Index used as part of key (field G_IDX) in the next lookup.
  * VCAP_AF_NXT_IDX_CTRL: W3, sparx5: is0
  *   Controls the generation of the G_IDX used in the VCAP CLM next lookup
- * VCAP_AF_PAG_OVERRIDE_MASK: W8, sparx5: is0
+ * VCAP_AF_OAM_SEL: W3, lan966x: is1
+ *   0: No changes to port-based selection in ANA:PORT:OAM_CFG.OAM_CFG 1: Enable
+ *   OAM frame processing for untagged frames 2: Enable OAM frame processing for
+ *   single frames 3: Enable OAM frame processing for double frames 4: Disable OAM
+ *   frame processing
+ * VCAP_AF_PAG_OVERRIDE_MASK: W8, sparx5: is0, lan966x: is1
  *   Bits set in this mask will override PAG_VAL from port profile. New PAG = (PAG
  *   (input) AND ~PAG_OVERRIDE_MASK) OR (PAG_VAL AND PAG_OVERRIDE_MASK)
- * VCAP_AF_PAG_VAL: W8, sparx5: is0
+ * VCAP_AF_PAG_VAL: W8, sparx5: is0, lan966x: is1
  *   See PAG_OVERRIDE_MASK.
  * VCAP_AF_PCP_A_VAL: W3, sparx5: es0
  *   PCP used in ES0 tag A. See TAG_A_PCP_SEL.
@@ -587,10 +658,10 @@ enum vcap_actionfield_set {
  *   PCP used in ES0 tag B. See TAG_B_PCP_SEL.
  * VCAP_AF_PCP_C_VAL: W3, sparx5: es0
  *   PCP used in ES0 tag C. See TAG_C_PCP_SEL.
- * VCAP_AF_PCP_ENA: W1, sparx5: is0
+ * VCAP_AF_PCP_ENA: W1, sparx5: is0, lan966x: is1
  *   If set, use PCP_VAL as classified PCP value. Otherwise, PCP from basic
  *   classification is used.
- * VCAP_AF_PCP_VAL: W3, sparx5: is0
+ * VCAP_AF_PCP_VAL: W3, sparx5: is0, lan966x: is1
  *   See PCP_ENA.
  * VCAP_AF_PIPELINE_ACT: W1, sparx5: es0
  *   Pipeline action when FWD_SEL > 0. 0: XTR. CPU_QU selects CPU extraction queue
@@ -600,11 +671,11 @@ enum vcap_actionfield_set {
  *   PIPELINE_PT == NONE. Overrules previous settings of pipeline point.
  * VCAP_AF_PIPELINE_PT: sparx5 is2 W5, sparx5 es0 W2
  *   Pipeline point used if PIPELINE_FORCE_ENA is set
- * VCAP_AF_POLICE_ENA: W1, sparx5: is2/es2, lan966x: is2
- *   Setting this bit to 1 causes frames that hit this action to be policed by the
- *   ACL policer specified in POLICE_IDX. Only applies to the first lookup.
- * VCAP_AF_POLICE_IDX: sparx5 is2 W6, sparx5 es2 W6, lan966x is2 W9
- *   Selects VCAP policer used when policing frames (POLICE_ENA)
+ * VCAP_AF_POLICE_ENA: W1, sparx5: is2/es2, lan966x: is1/is2
+ *   If set, POLICE_IDX is used to lookup ANA::POL.
+ * VCAP_AF_POLICE_IDX: sparx5 is2 W6, sparx5 es2 W6, lan966x is1 W9, lan966x is2
+ *   W9
+ *   Policer index.
  * VCAP_AF_POLICE_REMARK: W1, sparx5: es2
  *   If set, frames exceeding policer rates are marked as yellow but not
  *   discarded.
@@ -628,16 +699,24 @@ enum vcap_actionfield_set {
  *   port. 1: ES0 tag A: Push ES0 tag A. No port tag. 2: Force port tag: Always
  *   push port tag. No ES0 tag A. 3: Force untag: Never push port tag or ES0 tag
  *   A.
- * VCAP_AF_QOS_ENA: W1, sparx5: is0
+ * VCAP_AF_QOS_ENA: W1, sparx5: is0, lan966x: is1
  *   If set, use QOS_VAL as classified QoS class. Otherwise, QoS class from basic
  *   classification is used.
- * VCAP_AF_QOS_VAL: W3, sparx5: is0
+ * VCAP_AF_QOS_VAL: W3, sparx5: is0, lan966x: is1
  *   See QOS_ENA.
  * VCAP_AF_REW_OP: W16, lan966x: is2
  *   Rewriter operation command.
  * VCAP_AF_RT_DIS: W1, sparx5: is2
  *   If set, routing is disallowed. Only applies when IS_INNER_ACL is 0. See also
  *   IGR_ACL_ENA, EGR_ACL_ENA, and RLEG_STAT_IDX.
+ * VCAP_AF_SFID_ENA: W1, lan966x: is1
+ *   If set, SFID_VAL is used to lookup ANA::SFID.
+ * VCAP_AF_SFID_VAL: W8, lan966x: is1
+ *   Stream filter identifier.
+ * VCAP_AF_SGID_ENA: W1, lan966x: is1
+ *   If set, SGID_VAL is used to lookup ANA::SGID.
+ * VCAP_AF_SGID_VAL: W8, lan966x: is1
+ *   Stream gate identifier.
  * VCAP_AF_SWAP_MACS_ENA: W1, sparx5: es0
  *   This setting is only active when FWD_SEL = 1 or FWD_SEL = 2 and PIPELINE_ACT
  *   = LBK_ASM. 0: No action. 1: Swap MACs and clear bit 40 in new SMAC.
@@ -686,7 +765,7 @@ enum vcap_actionfield_set {
  * VCAP_AF_TAG_C_VID_SEL: W2, sparx5: es0
  *   Selects VID for ES0 tag C. The resulting VID is termed C-TAG.VID. 0:
  *   Classified VID. 1: VID_C_VAL. 2: IFH.ENCAP.GVID. 3: Reserved.
- * VCAP_AF_TYPE: W1, sparx5: is0
+ * VCAP_AF_TYPE: W1, sparx5: is0, lan966x: is1
  *   Actionset type id - Set by the API
  * VCAP_AF_UNTAG_VID_ENA: W1, sparx5: es0
  *   Controls insertion of tag C. Untag or insert mode can be selected. See
@@ -697,8 +776,19 @@ enum vcap_actionfield_set {
  *   VID used in ES0 tag B. See TAG_B_VID_SEL.
  * VCAP_AF_VID_C_VAL: W12, sparx5: es0
  *   VID used in ES0 tag C. See TAG_C_VID_SEL.
- * VCAP_AF_VID_VAL: W13, sparx5: is0
+ * VCAP_AF_VID_REPLACE_ENA: W1, lan966x: is1
+ *   Controls the classified VID: VID_REPLACE_ENA=0: Add VID_ADD_VAL to basic
+ *   classified VID and use result as new classified VID. VID_REPLACE_ENA = 1:
+ *   Replace basic classified VID with VID_VAL value and use as new classified
+ *   VID.
+ * VCAP_AF_VID_VAL: sparx5 is0 W13, lan966x is1 W12
  *   New VID Value
+ * VCAP_AF_VLAN_POP_CNT: W2, lan966x: is1
+ *   See VLAN_POP_CNT_ENA
+ * VCAP_AF_VLAN_POP_CNT_ENA: W1, lan966x: is1
+ *   If set, use VLAN_POP_CNT as the number of VLAN tags to pop from the incoming
+ *   frame. This number is used by the Rewriter. Otherwise, VLAN_POP_CNT from
+ *   ANA:PORT:VLAN_CFG.VLAN_POP_CNT is used
  */
 
 /* Actionfield names */
@@ -712,11 +802,13 @@ enum vcap_action_field {
 	VCAP_AF_CPU_COPY_ENA,
 	VCAP_AF_CPU_QU,
 	VCAP_AF_CPU_QUEUE_NUM,
+	VCAP_AF_CUSTOM_ACE_TYPE_ENA,
 	VCAP_AF_DEI_A_VAL,
 	VCAP_AF_DEI_B_VAL,
 	VCAP_AF_DEI_C_VAL,
 	VCAP_AF_DEI_ENA,
 	VCAP_AF_DEI_VAL,
+	VCAP_AF_DLR_SEL,
 	VCAP_AF_DP_ENA,
 	VCAP_AF_DP_VAL,
 	VCAP_AF_DSCP_ENA,
@@ -732,7 +824,9 @@ enum vcap_action_field {
 	VCAP_AF_IGNORE_PIPELINE_CTRL,
 	VCAP_AF_INTR_ENA,
 	VCAP_AF_ISDX_ADD_REPLACE_SEL,
+	VCAP_AF_ISDX_ADD_VAL,
 	VCAP_AF_ISDX_ENA,
+	VCAP_AF_ISDX_REPLACE_ENA,
 	VCAP_AF_ISDX_VAL,
 	VCAP_AF_LOOP_ENA,
 	VCAP_AF_LRN_DIS,
@@ -745,8 +839,10 @@ enum vcap_action_field {
 	VCAP_AF_MIRROR_ENA,
 	VCAP_AF_MIRROR_PROBE,
 	VCAP_AF_MIRROR_PROBE_ID,
+	VCAP_AF_MRP_SEL,
 	VCAP_AF_NXT_IDX,
 	VCAP_AF_NXT_IDX_CTRL,
+	VCAP_AF_OAM_SEL,
 	VCAP_AF_PAG_OVERRIDE_MASK,
 	VCAP_AF_PAG_VAL,
 	VCAP_AF_PCP_A_VAL,
@@ -770,6 +866,10 @@ enum vcap_action_field {
 	VCAP_AF_QOS_VAL,
 	VCAP_AF_REW_OP,
 	VCAP_AF_RT_DIS,
+	VCAP_AF_SFID_ENA,
+	VCAP_AF_SFID_VAL,
+	VCAP_AF_SGID_ENA,
+	VCAP_AF_SGID_VAL,
 	VCAP_AF_SWAP_MACS_ENA,
 	VCAP_AF_TAG_A_DEI_SEL,
 	VCAP_AF_TAG_A_PCP_SEL,
@@ -788,7 +888,10 @@ enum vcap_action_field {
 	VCAP_AF_VID_A_VAL,
 	VCAP_AF_VID_B_VAL,
 	VCAP_AF_VID_C_VAL,
+	VCAP_AF_VID_REPLACE_ENA,
 	VCAP_AF_VID_VAL,
+	VCAP_AF_VLAN_POP_CNT,
+	VCAP_AF_VLAN_POP_CNT_ENA,
 };
 
 #endif /* __VCAP_AG_API__ */
diff --git a/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c b/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c
index 0de3f677135a..b23c11b0647c 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c
+++ b/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c
@@ -387,7 +387,7 @@ static const char * const test_admin_info_expect[] = {
 	"default_cnt: 73\n",
 	"require_cnt_dis: 0\n",
 	"version: 1\n",
-	"vtype: 3\n",
+	"vtype: 4\n",
 	"vinst: 0\n",
 	"ingress: 1\n",
 	"first_cid: 10000\n",
@@ -435,7 +435,7 @@ static const char * const test_admin_expect[] = {
 	"default_cnt: 73\n",
 	"require_cnt_dis: 0\n",
 	"version: 1\n",
-	"vtype: 3\n",
+	"vtype: 4\n",
 	"vinst: 0\n",
 	"ingress: 1\n",
 	"first_cid: 8000000\n",
-- 
cgit v1.2.3-70-g09d2


From a4d9b3ec63de3c36890fb440980ebf8f642bab88 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 7 Mar 2023 23:09:26 +0100
Subject: net: lan966x: Add IS1 VCAP keyset configuration for lan966x

Add IS1 VCAP port keyset configuration for lan966x and also update debug
fs support to show the keyset configuration.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/microchip/lan966x/lan966x_main.h  |  38 ++++
 .../net/ethernet/microchip/lan966x/lan966x_regs.h  |  36 ++++
 .../microchip/lan966x/lan966x_vcap_debugfs.c       | 133 +++++++++++++-
 .../ethernet/microchip/lan966x/lan966x_vcap_impl.c | 192 +++++++++++++++++++--
 4 files changed, 383 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
index 49f5159afbf3..cbdae0ab8bb6 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
@@ -92,6 +92,11 @@
 #define SE_IDX_QUEUE			0  /* 0-79 : Queue scheduler elements */
 #define SE_IDX_PORT			80 /* 80-89 : Port schedular elements */
 
+#define LAN966X_VCAP_CID_IS1_L0 VCAP_CID_INGRESS_L0 /* IS1 lookup 0 */
+#define LAN966X_VCAP_CID_IS1_L1 VCAP_CID_INGRESS_L1 /* IS1 lookup 1 */
+#define LAN966X_VCAP_CID_IS1_L2 VCAP_CID_INGRESS_L2 /* IS1 lookup 2 */
+#define LAN966X_VCAP_CID_IS1_MAX (VCAP_CID_INGRESS_L3 - 1) /* IS1 Max */
+
 #define LAN966X_VCAP_CID_IS2_L0 VCAP_CID_INGRESS_STAGE2_L0 /* IS2 lookup 0 */
 #define LAN966X_VCAP_CID_IS2_L1 VCAP_CID_INGRESS_STAGE2_L1 /* IS2 lookup 1 */
 #define LAN966X_VCAP_CID_IS2_MAX (VCAP_CID_INGRESS_STAGE2_L2 - 1) /* IS2 Max */
@@ -139,6 +144,39 @@ enum vcap_is2_port_sel_ipv6 {
 	VCAP_IS2_PS_IPV6_MAC_ETYPE,
 };
 
+enum vcap_is1_port_sel_other {
+	VCAP_IS1_PS_OTHER_NORMAL,
+	VCAP_IS1_PS_OTHER_7TUPLE,
+	VCAP_IS1_PS_OTHER_DBL_VID,
+	VCAP_IS1_PS_OTHER_DMAC_VID,
+};
+
+enum vcap_is1_port_sel_ipv4 {
+	VCAP_IS1_PS_IPV4_NORMAL,
+	VCAP_IS1_PS_IPV4_7TUPLE,
+	VCAP_IS1_PS_IPV4_5TUPLE_IP4,
+	VCAP_IS1_PS_IPV4_DBL_VID,
+	VCAP_IS1_PS_IPV4_DMAC_VID,
+};
+
+enum vcap_is1_port_sel_ipv6 {
+	VCAP_IS1_PS_IPV6_NORMAL,
+	VCAP_IS1_PS_IPV6_7TUPLE,
+	VCAP_IS1_PS_IPV6_5TUPLE_IP4,
+	VCAP_IS1_PS_IPV6_NORMAL_IP6,
+	VCAP_IS1_PS_IPV6_5TUPLE_IP6,
+	VCAP_IS1_PS_IPV6_DBL_VID,
+	VCAP_IS1_PS_IPV6_DMAC_VID,
+};
+
+enum vcap_is1_port_sel_rt {
+	VCAP_IS1_PS_RT_NORMAL,
+	VCAP_IS1_PS_RT_7TUPLE,
+	VCAP_IS1_PS_RT_DBL_VID,
+	VCAP_IS1_PS_RT_DMAC_VID,
+	VCAP_IS1_PS_RT_FOLLOW_OTHER = 7,
+};
+
 struct lan966x_port;
 
 struct lan966x_db {
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h b/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h
index 9767b5a1c958..f99f88b5caa8 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h
@@ -316,6 +316,42 @@ enum lan966x_target {
 #define ANA_DROP_CFG_DROP_MC_SMAC_ENA_GET(x)\
 	FIELD_GET(ANA_DROP_CFG_DROP_MC_SMAC_ENA, x)
 
+/*      ANA:PORT:VCAP_CFG */
+#define ANA_VCAP_CFG(g)           __REG(TARGET_ANA, 0, 1, 28672, g, 9, 128, 12, 0, 1, 4)
+
+#define ANA_VCAP_CFG_S1_ENA                      BIT(14)
+#define ANA_VCAP_CFG_S1_ENA_SET(x)\
+	FIELD_PREP(ANA_VCAP_CFG_S1_ENA, x)
+#define ANA_VCAP_CFG_S1_ENA_GET(x)\
+	FIELD_GET(ANA_VCAP_CFG_S1_ENA, x)
+
+/*      ANA:PORT:VCAP_S1_KEY_CFG */
+#define ANA_VCAP_S1_CFG(g, r)     __REG(TARGET_ANA, 0, 1, 28672, g, 9, 128, 16, r, 3, 4)
+
+#define ANA_VCAP_S1_CFG_KEY_RT_CFG               GENMASK(11, 9)
+#define ANA_VCAP_S1_CFG_KEY_RT_CFG_SET(x)\
+	FIELD_PREP(ANA_VCAP_S1_CFG_KEY_RT_CFG, x)
+#define ANA_VCAP_S1_CFG_KEY_RT_CFG_GET(x)\
+	FIELD_GET(ANA_VCAP_S1_CFG_KEY_RT_CFG, x)
+
+#define ANA_VCAP_S1_CFG_KEY_IP6_CFG              GENMASK(8, 6)
+#define ANA_VCAP_S1_CFG_KEY_IP6_CFG_SET(x)\
+	FIELD_PREP(ANA_VCAP_S1_CFG_KEY_IP6_CFG, x)
+#define ANA_VCAP_S1_CFG_KEY_IP6_CFG_GET(x)\
+	FIELD_GET(ANA_VCAP_S1_CFG_KEY_IP6_CFG, x)
+
+#define ANA_VCAP_S1_CFG_KEY_IP4_CFG              GENMASK(5, 3)
+#define ANA_VCAP_S1_CFG_KEY_IP4_CFG_SET(x)\
+	FIELD_PREP(ANA_VCAP_S1_CFG_KEY_IP4_CFG, x)
+#define ANA_VCAP_S1_CFG_KEY_IP4_CFG_GET(x)\
+	FIELD_GET(ANA_VCAP_S1_CFG_KEY_IP4_CFG, x)
+
+#define ANA_VCAP_S1_CFG_KEY_OTHER_CFG            GENMASK(2, 0)
+#define ANA_VCAP_S1_CFG_KEY_OTHER_CFG_SET(x)\
+	FIELD_PREP(ANA_VCAP_S1_CFG_KEY_OTHER_CFG, x)
+#define ANA_VCAP_S1_CFG_KEY_OTHER_CFG_GET(x)\
+	FIELD_GET(ANA_VCAP_S1_CFG_KEY_OTHER_CFG, x)
+
 /*      ANA:PORT:VCAP_S2_CFG */
 #define ANA_VCAP_S2_CFG(g)        __REG(TARGET_ANA, 0, 1, 28672, g, 9, 128, 28, 0, 1, 4)
 
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_debugfs.c b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_debugfs.c
index 7a0db58f5513..d90c08cfcf14 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_debugfs.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_debugfs.c
@@ -5,9 +5,124 @@
 #include "vcap_api.h"
 #include "vcap_api_client.h"
 
-static void lan966x_vcap_port_keys(struct lan966x_port *port,
-				   struct vcap_admin *admin,
-				   struct vcap_output_print *out)
+static void lan966x_vcap_is1_port_keys(struct lan966x_port *port,
+				       struct vcap_admin *admin,
+				       struct vcap_output_print *out)
+{
+	struct lan966x *lan966x = port->lan966x;
+	u32 val;
+
+	out->prf(out->dst, "  port[%d] (%s): ", port->chip_port,
+		 netdev_name(port->dev));
+
+	val = lan_rd(lan966x, ANA_VCAP_CFG(port->chip_port));
+	out->prf(out->dst, "\n    state: ");
+	if (ANA_VCAP_CFG_S1_ENA_GET(val))
+		out->prf(out->dst, "on");
+	else
+		out->prf(out->dst, "off");
+
+	for (int l = 0; l < admin->lookups; ++l) {
+		out->prf(out->dst, "\n    Lookup %d: ", l);
+
+		out->prf(out->dst, "\n      other: ");
+		switch (ANA_VCAP_S1_CFG_KEY_OTHER_CFG_GET(val)) {
+		case VCAP_IS1_PS_OTHER_NORMAL:
+			out->prf(out->dst, "normal");
+			break;
+		case VCAP_IS1_PS_OTHER_7TUPLE:
+			out->prf(out->dst, "7tuple");
+			break;
+		case VCAP_IS1_PS_OTHER_DBL_VID:
+			out->prf(out->dst, "dbl_vid");
+			break;
+		case VCAP_IS1_PS_OTHER_DMAC_VID:
+			out->prf(out->dst, "dmac_vid");
+			break;
+		default:
+			out->prf(out->dst, "-");
+			break;
+		}
+
+		out->prf(out->dst, "\n      ipv4: ");
+		switch (ANA_VCAP_S1_CFG_KEY_IP4_CFG_GET(val)) {
+		case VCAP_IS1_PS_IPV4_NORMAL:
+			out->prf(out->dst, "normal");
+			break;
+		case VCAP_IS1_PS_IPV4_7TUPLE:
+			out->prf(out->dst, "7tuple");
+			break;
+		case VCAP_IS1_PS_IPV4_5TUPLE_IP4:
+			out->prf(out->dst, "5tuple_ipv4");
+			break;
+		case VCAP_IS1_PS_IPV4_DBL_VID:
+			out->prf(out->dst, "dbl_vid");
+			break;
+		case VCAP_IS1_PS_IPV4_DMAC_VID:
+			out->prf(out->dst, "dmac_vid");
+			break;
+		default:
+			out->prf(out->dst, "-");
+			break;
+		}
+
+		out->prf(out->dst, "\n      ipv6: ");
+		switch (ANA_VCAP_S1_CFG_KEY_IP6_CFG_GET(val)) {
+		case VCAP_IS1_PS_IPV6_NORMAL:
+			out->prf(out->dst, "normal");
+			break;
+		case VCAP_IS1_PS_IPV6_7TUPLE:
+			out->prf(out->dst, "7tuple");
+			break;
+		case VCAP_IS1_PS_IPV6_5TUPLE_IP4:
+			out->prf(out->dst, "5tuple_ip4");
+			break;
+		case VCAP_IS1_PS_IPV6_NORMAL_IP6:
+			out->prf(out->dst, "normal_ip6");
+			break;
+		case VCAP_IS1_PS_IPV6_5TUPLE_IP6:
+			out->prf(out->dst, "5tuple_ip6");
+			break;
+		case VCAP_IS1_PS_IPV6_DBL_VID:
+			out->prf(out->dst, "dbl_vid");
+			break;
+		case VCAP_IS1_PS_IPV6_DMAC_VID:
+			out->prf(out->dst, "dmac_vid");
+			break;
+		default:
+			out->prf(out->dst, "-");
+			break;
+		}
+
+		out->prf(out->dst, "\n      rt: ");
+		switch (ANA_VCAP_S1_CFG_KEY_RT_CFG_GET(val)) {
+		case VCAP_IS1_PS_RT_NORMAL:
+			out->prf(out->dst, "normal");
+			break;
+		case VCAP_IS1_PS_RT_7TUPLE:
+			out->prf(out->dst, "7tuple");
+			break;
+		case VCAP_IS1_PS_RT_DBL_VID:
+			out->prf(out->dst, "dbl_vid");
+			break;
+		case VCAP_IS1_PS_RT_DMAC_VID:
+			out->prf(out->dst, "dmac_vid");
+			break;
+		case VCAP_IS1_PS_RT_FOLLOW_OTHER:
+			out->prf(out->dst, "follow_other");
+			break;
+		default:
+			out->prf(out->dst, "-");
+			break;
+		}
+	}
+
+	out->prf(out->dst, "\n");
+}
+
+static void lan966x_vcap_is2_port_keys(struct lan966x_port *port,
+				       struct vcap_admin *admin,
+				       struct vcap_output_print *out)
 {
 	struct lan966x *lan966x = port->lan966x;
 	u32 val;
@@ -88,7 +203,17 @@ int lan966x_vcap_port_info(struct net_device *dev,
 	vcap = &vctrl->vcaps[admin->vtype];
 
 	out->prf(out->dst, "%s:\n", vcap->name);
-	lan966x_vcap_port_keys(port, admin, out);
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS2:
+		lan966x_vcap_is2_port_keys(port, admin, out);
+		break;
+	case VCAP_TYPE_IS1:
+		lan966x_vcap_is1_port_keys(port, admin, out);
+		break;
+	default:
+		out->prf(out->dst, "  no info\n");
+		break;
+	}
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c
index 68f9d69fd37b..7ea8e8633609 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c
@@ -8,6 +8,7 @@
 
 #define STREAMSIZE (64 * 4)
 
+#define LAN966X_IS1_LOOKUPS 3
 #define LAN966X_IS2_LOOKUPS 2
 
 static struct lan966x_vcap_inst {
@@ -19,6 +20,15 @@ static struct lan966x_vcap_inst {
 	int count; /* number of available addresses */
 	bool ingress; /* is vcap in the ingress path */
 } lan966x_vcap_inst_cfg[] = {
+	{
+		.vtype = VCAP_TYPE_IS1, /* IS1-0 */
+		.tgt_inst = 1,
+		.lookups = LAN966X_IS1_LOOKUPS,
+		.first_cid = LAN966X_VCAP_CID_IS1_L0,
+		.last_cid = LAN966X_VCAP_CID_IS1_MAX,
+		.count = 768,
+		.ingress = true,
+	},
 	{
 		.vtype = VCAP_TYPE_IS2, /* IS2-0 */
 		.tgt_inst = 2,
@@ -72,7 +82,21 @@ static void __lan966x_vcap_range_init(struct lan966x *lan966x,
 	lan966x_vcap_wait_update(lan966x, admin->tgt_inst);
 }
 
-static int lan966x_vcap_cid_to_lookup(int cid)
+static int lan966x_vcap_is1_cid_to_lookup(int cid)
+{
+	int lookup = 0;
+
+	if (cid >= LAN966X_VCAP_CID_IS1_L1 &&
+	    cid < LAN966X_VCAP_CID_IS1_L2)
+		lookup = 1;
+	else if (cid >= LAN966X_VCAP_CID_IS1_L2 &&
+		 cid < LAN966X_VCAP_CID_IS1_MAX)
+		lookup = 2;
+
+	return lookup;
+}
+
+static int lan966x_vcap_is2_cid_to_lookup(int cid)
 {
 	if (cid >= LAN966X_VCAP_CID_IS2_L1 &&
 	    cid < LAN966X_VCAP_CID_IS2_MAX)
@@ -81,6 +105,67 @@ static int lan966x_vcap_cid_to_lookup(int cid)
 	return 0;
 }
 
+/* Return the list of keysets for the vcap port configuration */
+static int
+lan966x_vcap_is1_get_port_keysets(struct net_device *ndev, int lookup,
+				  struct vcap_keyset_list *keysetlist,
+				  u16 l3_proto)
+{
+	struct lan966x_port *port = netdev_priv(ndev);
+	struct lan966x *lan966x = port->lan966x;
+	u32 val;
+
+	val = lan_rd(lan966x, ANA_VCAP_S1_CFG(port->chip_port, lookup));
+
+	/* Collect all keysets for the port in a list */
+	if (l3_proto == ETH_P_ALL || l3_proto == ETH_P_IP) {
+		switch (ANA_VCAP_S1_CFG_KEY_IP4_CFG_GET(val)) {
+		case VCAP_IS1_PS_IPV4_7TUPLE:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_7TUPLE);
+			break;
+		case VCAP_IS1_PS_IPV4_5TUPLE_IP4:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_5TUPLE_IP4);
+			break;
+		case VCAP_IS1_PS_IPV4_NORMAL:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_NORMAL);
+			break;
+		}
+	}
+
+	if (l3_proto == ETH_P_ALL || l3_proto == ETH_P_IPV6) {
+		switch (ANA_VCAP_S1_CFG_KEY_IP6_CFG_GET(val)) {
+		case VCAP_IS1_PS_IPV6_NORMAL:
+		case VCAP_IS1_PS_IPV6_NORMAL_IP6:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_NORMAL);
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_NORMAL_IP6);
+			break;
+		case VCAP_IS1_PS_IPV6_5TUPLE_IP6:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_5TUPLE_IP6);
+			break;
+		case VCAP_IS1_PS_IPV6_7TUPLE:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_7TUPLE);
+			break;
+		case VCAP_IS1_PS_IPV6_5TUPLE_IP4:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_5TUPLE_IP4);
+			break;
+		case VCAP_IS1_PS_IPV6_DMAC_VID:
+			vcap_keyset_list_add(keysetlist, VCAP_KFS_DMAC_VID);
+			break;
+		}
+	}
+
+	switch (ANA_VCAP_S1_CFG_KEY_OTHER_CFG_GET(val)) {
+	case VCAP_IS1_PS_OTHER_7TUPLE:
+		vcap_keyset_list_add(keysetlist, VCAP_KFS_7TUPLE);
+		break;
+	case VCAP_IS1_PS_OTHER_NORMAL:
+		vcap_keyset_list_add(keysetlist, VCAP_KFS_NORMAL);
+		break;
+	}
+
+	return 0;
+}
+
 static int
 lan966x_vcap_is2_get_port_keysets(struct net_device *dev, int lookup,
 				  struct vcap_keyset_list *keysetlist,
@@ -180,11 +265,26 @@ lan966x_vcap_validate_keyset(struct net_device *dev,
 	if (!kslist || kslist->cnt == 0)
 		return VCAP_KFS_NO_VALUE;
 
-	lookup = lan966x_vcap_cid_to_lookup(rule->vcap_chain_id);
 	keysetlist.max = ARRAY_SIZE(keysets);
 	keysetlist.keysets = keysets;
-	err = lan966x_vcap_is2_get_port_keysets(dev, lookup, &keysetlist,
-						l3_proto);
+
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS1:
+		lookup = lan966x_vcap_is1_cid_to_lookup(rule->vcap_chain_id);
+		err = lan966x_vcap_is1_get_port_keysets(dev, lookup, &keysetlist,
+							l3_proto);
+		break;
+	case VCAP_TYPE_IS2:
+		lookup = lan966x_vcap_is2_cid_to_lookup(rule->vcap_chain_id);
+		err = lan966x_vcap_is2_get_port_keysets(dev, lookup, &keysetlist,
+							l3_proto);
+		break;
+	default:
+		pr_err("vcap type: %s not supported\n",
+		       lan966x_vcaps[admin->vtype].name);
+		return VCAP_KFS_NO_VALUE;
+	}
+
 	if (err)
 		return VCAP_KFS_NO_VALUE;
 
@@ -197,17 +297,32 @@ lan966x_vcap_validate_keyset(struct net_device *dev,
 	return VCAP_KFS_NO_VALUE;
 }
 
-static bool lan966x_vcap_is_first_chain(struct vcap_rule *rule)
+static bool lan966x_vcap_is2_is_first_chain(struct vcap_rule *rule)
 {
 	return (rule->vcap_chain_id >= LAN966X_VCAP_CID_IS2_L0 &&
 		rule->vcap_chain_id < LAN966X_VCAP_CID_IS2_L1);
 }
 
-static void lan966x_vcap_add_default_fields(struct net_device *dev,
-					    struct vcap_admin *admin,
-					    struct vcap_rule *rule)
+static void lan966x_vcap_is1_add_default_fields(struct lan966x_port *port,
+						struct vcap_admin *admin,
+						struct vcap_rule *rule)
+{
+	u32 value, mask;
+	u32 lookup;
+
+	if (vcap_rule_get_key_u32(rule, VCAP_KF_IF_IGR_PORT_MASK,
+				  &value, &mask))
+		vcap_rule_add_key_u32(rule, VCAP_KF_IF_IGR_PORT_MASK, 0,
+				      ~BIT(port->chip_port));
+
+	lookup = lan966x_vcap_is1_cid_to_lookup(rule->vcap_chain_id);
+	vcap_rule_add_key_u32(rule, VCAP_KF_LOOKUP_INDEX, lookup, 0x3);
+}
+
+static void lan966x_vcap_is2_add_default_fields(struct lan966x_port *port,
+						struct vcap_admin *admin,
+						struct vcap_rule *rule)
 {
-	struct lan966x_port *port = netdev_priv(dev);
 	u32 value, mask;
 
 	if (vcap_rule_get_key_u32(rule, VCAP_KF_IF_IGR_PORT_MASK,
@@ -215,7 +330,7 @@ static void lan966x_vcap_add_default_fields(struct net_device *dev,
 		vcap_rule_add_key_u32(rule, VCAP_KF_IF_IGR_PORT_MASK, 0,
 				      ~BIT(port->chip_port));
 
-	if (lan966x_vcap_is_first_chain(rule))
+	if (lan966x_vcap_is2_is_first_chain(rule))
 		vcap_rule_add_key_bit(rule, VCAP_KF_LOOKUP_FIRST_IS,
 				      VCAP_BIT_1);
 	else
@@ -223,6 +338,26 @@ static void lan966x_vcap_add_default_fields(struct net_device *dev,
 				      VCAP_BIT_0);
 }
 
+static void lan966x_vcap_add_default_fields(struct net_device *dev,
+					    struct vcap_admin *admin,
+					    struct vcap_rule *rule)
+{
+	struct lan966x_port *port = netdev_priv(dev);
+
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS1:
+		lan966x_vcap_is1_add_default_fields(port, admin, rule);
+		break;
+	case VCAP_TYPE_IS2:
+		lan966x_vcap_is2_add_default_fields(port, admin, rule);
+		break;
+	default:
+		pr_err("vcap type: %s not supported\n",
+		       lan966x_vcaps[admin->vtype].name);
+		break;
+	}
+}
+
 static void lan966x_vcap_cache_erase(struct vcap_admin *admin)
 {
 	memset(admin->cache.keystream, 0, STREAMSIZE);
@@ -464,8 +599,37 @@ static void lan966x_vcap_block_init(struct lan966x *lan966x,
 static void lan966x_vcap_port_key_deselection(struct lan966x *lan966x,
 					      struct vcap_admin *admin)
 {
-	for (int p = 0; p < lan966x->num_phys_ports; ++p)
-		lan_wr(0, lan966x, ANA_VCAP_S2_CFG(p));
+	u32 val;
+
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS1:
+		val = ANA_VCAP_S1_CFG_KEY_IP6_CFG_SET(VCAP_IS1_PS_IPV6_5TUPLE_IP6) |
+		      ANA_VCAP_S1_CFG_KEY_IP4_CFG_SET(VCAP_IS1_PS_IPV4_5TUPLE_IP4) |
+		      ANA_VCAP_S1_CFG_KEY_OTHER_CFG_SET(VCAP_IS1_PS_OTHER_NORMAL);
+
+		for (int p = 0; p < lan966x->num_phys_ports; ++p) {
+			if (!lan966x->ports[p])
+				continue;
+
+			for (int l = 0; l < LAN966X_IS1_LOOKUPS; ++l)
+				lan_wr(val, lan966x, ANA_VCAP_S1_CFG(p, l));
+
+			lan_rmw(ANA_VCAP_CFG_S1_ENA_SET(true),
+				ANA_VCAP_CFG_S1_ENA, lan966x,
+				ANA_VCAP_CFG(p));
+		}
+
+		break;
+	case VCAP_TYPE_IS2:
+		for (int p = 0; p < lan966x->num_phys_ports; ++p)
+			lan_wr(0, lan966x, ANA_VCAP_S2_CFG(p));
+
+		break;
+	default:
+		pr_err("vcap type: %s not supported\n",
+		       lan966x_vcaps[admin->vtype].name);
+		break;
+	}
 }
 
 int lan966x_vcap_init(struct lan966x *lan966x)
@@ -506,6 +670,10 @@ int lan966x_vcap_init(struct lan966x *lan966x)
 			lan_rmw(ANA_VCAP_S2_CFG_ENA_SET(true),
 				ANA_VCAP_S2_CFG_ENA, lan966x,
 				ANA_VCAP_S2_CFG(lan966x->ports[p]->chip_port));
+
+			lan_rmw(ANA_VCAP_CFG_S1_ENA_SET(true),
+				ANA_VCAP_CFG_S1_ENA, lan966x,
+				ANA_VCAP_CFG(lan966x->ports[p]->chip_port));
 		}
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 135c2116fd03642b74fa2c0cd1cbd2614ca3d80c Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 7 Mar 2023 23:09:27 +0100
Subject: net: lan966x: Add TC support for IS1 VCAP

Enable TC command to use IS1 VCAP

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/microchip/lan966x/lan966x_tc_flower.c | 105 ++++++++++++++++++++-
 1 file changed, 102 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c b/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
index f960727ecaee..8391652c1c45 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
@@ -79,18 +79,61 @@ lan966x_tc_flower_handler_basic_usage(struct vcap_tc_flower_parse_usage *st)
 						    VCAP_BIT_1);
 			if (err)
 				goto out;
+		} else if (st->l3_proto == ETH_P_IPV6 &&
+			   st->admin->vtype == VCAP_TYPE_IS1) {
+			/* Don't set any keys in this case */
+		} else if (st->l3_proto == ETH_P_SNAP &&
+			   st->admin->vtype == VCAP_TYPE_IS1) {
+			err = vcap_rule_add_key_bit(st->vrule,
+						    VCAP_KF_ETYPE_LEN_IS,
+						    VCAP_BIT_0);
+			if (err)
+				goto out;
+
+			err = vcap_rule_add_key_bit(st->vrule,
+						    VCAP_KF_IP_SNAP_IS,
+						    VCAP_BIT_1);
+			if (err)
+				goto out;
+		} else if (st->admin->vtype == VCAP_TYPE_IS1) {
+			err = vcap_rule_add_key_bit(st->vrule,
+						    VCAP_KF_ETYPE_LEN_IS,
+						    VCAP_BIT_1);
+			if (err)
+				goto out;
+
+			err = vcap_rule_add_key_u32(st->vrule, VCAP_KF_ETYPE,
+						    st->l3_proto, ~0);
+			if (err)
+				goto out;
 		}
 	}
 	if (match.mask->ip_proto) {
 		st->l4_proto = match.key->ip_proto;
 
 		if (st->l4_proto == IPPROTO_TCP) {
+			if (st->admin->vtype == VCAP_TYPE_IS1) {
+				err = vcap_rule_add_key_bit(st->vrule,
+							    VCAP_KF_TCP_UDP_IS,
+							    VCAP_BIT_1);
+				if (err)
+					goto out;
+			}
+
 			err = vcap_rule_add_key_bit(st->vrule,
 						    VCAP_KF_TCP_IS,
 						    VCAP_BIT_1);
 			if (err)
 				goto out;
 		} else if (st->l4_proto == IPPROTO_UDP) {
+			if (st->admin->vtype == VCAP_TYPE_IS1) {
+				err = vcap_rule_add_key_bit(st->vrule,
+							    VCAP_KF_TCP_UDP_IS,
+							    VCAP_BIT_1);
+				if (err)
+					goto out;
+			}
+
 			err = vcap_rule_add_key_bit(st->vrule,
 						    VCAP_KF_TCP_IS,
 						    VCAP_BIT_0);
@@ -112,12 +155,30 @@ out:
 	return err;
 }
 
+static int
+lan966x_tc_flower_handler_cvlan_usage(struct vcap_tc_flower_parse_usage *st)
+{
+	if (st->admin->vtype != VCAP_TYPE_IS1) {
+		NL_SET_ERR_MSG_MOD(st->fco->common.extack,
+				   "cvlan not supported in this VCAP");
+		return -EINVAL;
+	}
+
+	return vcap_tc_flower_handler_cvlan_usage(st);
+}
+
 static int
 lan966x_tc_flower_handler_vlan_usage(struct vcap_tc_flower_parse_usage *st)
 {
-	return vcap_tc_flower_handler_vlan_usage(st,
-						 VCAP_KF_8021Q_VID_CLS,
-						 VCAP_KF_8021Q_PCP_CLS);
+	enum vcap_key_field vid_key = VCAP_KF_8021Q_VID_CLS;
+	enum vcap_key_field pcp_key = VCAP_KF_8021Q_PCP_CLS;
+
+	if (st->admin->vtype == VCAP_TYPE_IS1) {
+		vid_key = VCAP_KF_8021Q_VID0;
+		pcp_key = VCAP_KF_8021Q_PCP0;
+	}
+
+	return vcap_tc_flower_handler_vlan_usage(st, vid_key, pcp_key);
 }
 
 static int
@@ -128,6 +189,7 @@ static int
 	[FLOW_DISSECTOR_KEY_CONTROL] = lan966x_tc_flower_handler_control_usage,
 	[FLOW_DISSECTOR_KEY_PORTS] = vcap_tc_flower_handler_portnum_usage,
 	[FLOW_DISSECTOR_KEY_BASIC] = lan966x_tc_flower_handler_basic_usage,
+	[FLOW_DISSECTOR_KEY_CVLAN] = lan966x_tc_flower_handler_cvlan_usage,
 	[FLOW_DISSECTOR_KEY_VLAN] = lan966x_tc_flower_handler_vlan_usage,
 	[FLOW_DISSECTOR_KEY_TCP] = vcap_tc_flower_handler_tcp_usage,
 	[FLOW_DISSECTOR_KEY_ARP] = vcap_tc_flower_handler_arp_usage,
@@ -143,6 +205,7 @@ static int lan966x_tc_flower_use_dissectors(struct flow_cls_offload *f,
 		.fco = f,
 		.vrule = vrule,
 		.l3_proto = ETH_P_ALL,
+		.admin = admin,
 	};
 	int err = 0;
 
@@ -221,6 +284,31 @@ static int lan966x_tc_flower_action_check(struct vcap_control *vctrl,
 	return 0;
 }
 
+/* Add the actionset that is the default for the VCAP type */
+static int lan966x_tc_set_actionset(struct vcap_admin *admin,
+				    struct vcap_rule *vrule)
+{
+	enum vcap_actionfield_set aset;
+	int err = 0;
+
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS1:
+		aset = VCAP_AFS_S1;
+		break;
+	case VCAP_TYPE_IS2:
+		aset = VCAP_AFS_BASE_TYPE;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Do not overwrite any current actionset */
+	if (vrule->actionset == VCAP_AFS_NO_VALUE)
+		err = vcap_set_rule_set_actionset(vrule, aset);
+
+	return err;
+}
+
 static int lan966x_tc_flower_add(struct lan966x_port *port,
 				 struct flow_cls_offload *f,
 				 struct vcap_admin *admin,
@@ -253,6 +341,13 @@ static int lan966x_tc_flower_add(struct lan966x_port *port,
 	flow_action_for_each(idx, act, &frule->action) {
 		switch (act->id) {
 		case FLOW_ACTION_TRAP:
+			if (admin->vtype != VCAP_TYPE_IS2) {
+				NL_SET_ERR_MSG_MOD(f->common.extack,
+						   "Trap action not supported in this VCAP");
+				err = -EOPNOTSUPP;
+				goto out;
+			}
+
 			err = vcap_rule_add_action_bit(vrule,
 						       VCAP_AF_CPU_COPY_ENA,
 						       VCAP_BIT_1);
@@ -266,6 +361,10 @@ static int lan966x_tc_flower_add(struct lan966x_port *port,
 
 			break;
 		case FLOW_ACTION_GOTO:
+			err = lan966x_tc_set_actionset(admin, vrule);
+			if (err)
+				goto out;
+
 			break;
 		default:
 			NL_SET_ERR_MSG_MOD(f->common.extack,
-- 
cgit v1.2.3-70-g09d2


From b3762a9db39c40a2e508a8edb91217676a74f299 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 7 Mar 2023 23:09:28 +0100
Subject: net: lan966x: Add TC filter chaining support for IS1 and IS2 VCAPs

Allow rules to be chained between IS1 VCAP and IS2 VCAP. Chaining
between IS1 lookups or between IS2 lookups are not supported by the
hardware.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/microchip/lan966x/lan966x_tc_flower.c | 80 ++++++++++++++++++++++
 1 file changed, 80 insertions(+)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c b/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
index 8391652c1c45..570ac28736e0 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
@@ -309,6 +309,75 @@ static int lan966x_tc_set_actionset(struct vcap_admin *admin,
 	return err;
 }
 
+static int lan966x_tc_add_rule_link_target(struct vcap_admin *admin,
+					   struct vcap_rule *vrule,
+					   int target_cid)
+{
+	int link_val = target_cid % VCAP_CID_LOOKUP_SIZE;
+	int err;
+
+	if (!link_val)
+		return 0;
+
+	switch (admin->vtype) {
+	case VCAP_TYPE_IS1:
+		/* Choose IS1 specific NXT_IDX key (for chaining rules from IS1) */
+		err = vcap_rule_add_key_u32(vrule, VCAP_KF_LOOKUP_GEN_IDX_SEL,
+					    1, ~0);
+		if (err)
+			return err;
+
+		return vcap_rule_add_key_u32(vrule, VCAP_KF_LOOKUP_GEN_IDX,
+					     link_val, ~0);
+	case VCAP_TYPE_IS2:
+		/* Add IS2 specific PAG key (for chaining rules from IS1) */
+		return vcap_rule_add_key_u32(vrule, VCAP_KF_LOOKUP_PAG,
+					     link_val, ~0);
+	default:
+		break;
+	}
+	return 0;
+}
+
+static int lan966x_tc_add_rule_link(struct vcap_control *vctrl,
+				    struct vcap_admin *admin,
+				    struct vcap_rule *vrule,
+				    struct flow_cls_offload *f,
+				    int to_cid)
+{
+	struct vcap_admin *to_admin = vcap_find_admin(vctrl, to_cid);
+	int diff, err = 0;
+
+	if (!to_admin) {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "Unknown destination chain");
+		return -EINVAL;
+	}
+
+	diff = vcap_chain_offset(vctrl, f->common.chain_index, to_cid);
+	if (!diff)
+		return 0;
+
+	/* Between IS1 and IS2 the PAG value is used */
+	if (admin->vtype == VCAP_TYPE_IS1 && to_admin->vtype == VCAP_TYPE_IS2) {
+		/* This works for IS1->IS2 */
+		err = vcap_rule_add_action_u32(vrule, VCAP_AF_PAG_VAL, diff);
+		if (err)
+			return err;
+
+		err = vcap_rule_add_action_u32(vrule, VCAP_AF_PAG_OVERRIDE_MASK,
+					       0xff);
+		if (err)
+			return err;
+	} else {
+		NL_SET_ERR_MSG_MOD(f->common.extack,
+				   "Unsupported chain destination");
+		return -EOPNOTSUPP;
+	}
+
+	return err;
+}
+
 static int lan966x_tc_flower_add(struct lan966x_port *port,
 				 struct flow_cls_offload *f,
 				 struct vcap_admin *admin,
@@ -336,6 +405,11 @@ static int lan966x_tc_flower_add(struct lan966x_port *port,
 	if (err)
 		goto out;
 
+	err = lan966x_tc_add_rule_link_target(admin, vrule,
+					      f->common.chain_index);
+	if (err)
+		goto out;
+
 	frule = flow_cls_offload_flow_rule(f);
 
 	flow_action_for_each(idx, act, &frule->action) {
@@ -365,6 +439,12 @@ static int lan966x_tc_flower_add(struct lan966x_port *port,
 			if (err)
 				goto out;
 
+			err = lan966x_tc_add_rule_link(port->lan966x->vcap_ctrl,
+						       admin, vrule,
+						       f, act->chain_index);
+			if (err)
+				goto out;
+
 			break;
 		default:
 			NL_SET_ERR_MSG_MOD(f->common.extack,
-- 
cgit v1.2.3-70-g09d2


From 44d706fde75518781f6859febb0979cf5e7327d4 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 7 Mar 2023 23:09:29 +0100
Subject: net: lan966x: Add support for IS1 VCAP ethernet protocol types

IS1 VCAP has it's own list of supported ethernet protocol types which is
different than the IS2 VCAP. Therefore separate the list of known
protocol types based on the VCAP type.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/microchip/lan966x/lan966x_tc_flower.c | 36 +++++++++++++++++-----
 1 file changed, 28 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c b/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
index 570ac28736e0..47b2f7579dd2 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_tc_flower.c
@@ -5,14 +5,34 @@
 #include "vcap_api_client.h"
 #include "vcap_tc.h"
 
-static bool lan966x_tc_is_known_etype(u16 etype)
+static bool lan966x_tc_is_known_etype(struct vcap_tc_flower_parse_usage *st,
+				      u16 etype)
 {
-	switch (etype) {
-	case ETH_P_ALL:
-	case ETH_P_ARP:
-	case ETH_P_IP:
-	case ETH_P_IPV6:
-		return true;
+	switch (st->admin->vtype) {
+	case VCAP_TYPE_IS1:
+		switch (etype) {
+		case ETH_P_ALL:
+		case ETH_P_ARP:
+		case ETH_P_IP:
+		case ETH_P_IPV6:
+			return true;
+		}
+		break;
+	case VCAP_TYPE_IS2:
+		switch (etype) {
+		case ETH_P_ALL:
+		case ETH_P_ARP:
+		case ETH_P_IP:
+		case ETH_P_IPV6:
+		case ETH_P_SNAP:
+		case ETH_P_802_2:
+			return true;
+		}
+		break;
+	default:
+		NL_SET_ERR_MSG_MOD(st->fco->common.extack,
+				   "VCAP type not supported");
+		return false;
 	}
 
 	return false;
@@ -69,7 +89,7 @@ lan966x_tc_flower_handler_basic_usage(struct vcap_tc_flower_parse_usage *st)
 	flow_rule_match_basic(st->frule, &match);
 	if (match.mask->n_proto) {
 		st->l3_proto = be16_to_cpu(match.key->n_proto);
-		if (!lan966x_tc_is_known_etype(st->l3_proto)) {
+		if (!lan966x_tc_is_known_etype(st, st->l3_proto)) {
 			err = vcap_rule_add_key_u32(st->vrule, VCAP_KF_ETYPE,
 						    st->l3_proto, ~0);
 			if (err)
-- 
cgit v1.2.3-70-g09d2


From 438b406055cd21105aad77db7938ee4720b09bee Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Tue, 7 Mar 2023 20:45:56 -0700
Subject: tun: flag the device as supporting FMODE_NOWAIT

tun already checks for both O_NONBLOCK and IOCB_NOWAIT in its read
and write iter handlers, so it's fully ready for FMODE_NOWAIT. But
for some reason it doesn't set it. Rectify that oversight.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/3f7dc1f0-79ca-d85c-4d16-8c12c5bd492d@kernel.dk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/tun.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index ad653b32b2f0..4c7f74904c25 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -3463,6 +3463,8 @@ static int tun_chr_open(struct inode *inode, struct file * file)
 
 	sock_set_flag(&tfile->sk, SOCK_ZEROCOPY);
 
+	/* tun groks IOCB_NOWAIT just fine, mark it as such */
+	file->f_mode |= FMODE_NOWAIT;
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From f758bfec377ad2f15d7683473b1db1cfbf8e1bb0 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Tue, 7 Mar 2023 21:18:21 -0700
Subject: tap: add support for IOCB_NOWAIT

The tap driver already supports passing in nonblocking state based
on O_NONBLOCK, add support for checking IOCB_NOWAIT as well. With that
done, we can flag it with FMODE_NOWAIT as well.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/8f859870-e6e2-09ca-9c0f-a2aa7c984fb2@kernel.dk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/tap.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/net/tap.c b/drivers/net/tap.c
index 8941aa199ea3..ce993cc75bf3 100644
--- a/drivers/net/tap.c
+++ b/drivers/net/tap.c
@@ -555,6 +555,9 @@ static int tap_open(struct inode *inode, struct file *file)
 		goto err_put;
 	}
 
+	/* tap groks IOCB_NOWAIT just fine, mark it as such */
+	file->f_mode |= FMODE_NOWAIT;
+
 	dev_put(tap->dev);
 
 	rtnl_unlock();
@@ -771,8 +774,12 @@ static ssize_t tap_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct tap_queue *q = file->private_data;
+	int noblock = 0;
+
+	if ((file->f_flags & O_NONBLOCK) || (iocb->ki_flags & IOCB_NOWAIT))
+		noblock = 1;
 
-	return tap_get_user(q, NULL, from, file->f_flags & O_NONBLOCK);
+	return tap_get_user(q, NULL, from, noblock);
 }
 
 /* Put packet to the user space buffer */
@@ -888,8 +895,12 @@ static ssize_t tap_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	struct file *file = iocb->ki_filp;
 	struct tap_queue *q = file->private_data;
 	ssize_t len = iov_iter_count(to), ret;
+	int noblock = 0;
+
+	if ((file->f_flags & O_NONBLOCK) || (iocb->ki_flags & IOCB_NOWAIT))
+		noblock = 1;
 
-	ret = tap_do_read(q, to, file->f_flags & O_NONBLOCK, NULL);
+	ret = tap_do_read(q, to, noblock, NULL);
 	ret = min_t(ssize_t, ret, len);
 	if (ret > 0)
 		iocb->ki_pos = ret;
-- 
cgit v1.2.3-70-g09d2


From 566b6701d5dfc823d1e7ee27a7f0c5719f4b93dd Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Wed, 8 Mar 2023 15:17:19 +0200
Subject: skbuff: Replace open-coded skb_propagate_pfmemalloc()s

Use skb_propagate_pfmemalloc() in build_skb()/build_skb_around() instead
of open-coding it.

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/skbuff.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1a31815104d6..de465368fc2c 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -422,8 +422,7 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size)
 
 	if (skb && frag_size) {
 		skb->head_frag = 1;
-		if (page_is_pfmemalloc(virt_to_head_page(data)))
-			skb->pfmemalloc = 1;
+		skb_propagate_pfmemalloc(virt_to_head_page(data), skb);
 	}
 	return skb;
 }
@@ -445,8 +444,7 @@ struct sk_buff *build_skb_around(struct sk_buff *skb,
 
 	if (frag_size) {
 		skb->head_frag = 1;
-		if (page_is_pfmemalloc(virt_to_head_page(data)))
-			skb->pfmemalloc = 1;
+		skb_propagate_pfmemalloc(virt_to_head_page(data), skb);
 	}
 	return skb;
 }
-- 
cgit v1.2.3-70-g09d2


From 3c6401266f91c69771176d3b289abfeaec731611 Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Wed, 8 Mar 2023 15:17:20 +0200
Subject: skbuff: Add likely to skb pointer in build_skb()

Similarly to napi_build_skb(), it is likely the skb allocation in
build_skb() succeeded. frag_size != 0 is also likely, as stated in
__build_skb_around().

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/skbuff.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index de465368fc2c..050a875d09c5 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -420,7 +420,7 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size)
 {
 	struct sk_buff *skb = __build_skb(data, frag_size);
 
-	if (skb && frag_size) {
+	if (likely(skb && frag_size)) {
 		skb->head_frag = 1;
 		skb_propagate_pfmemalloc(virt_to_head_page(data), skb);
 	}
-- 
cgit v1.2.3-70-g09d2


From f94b9bed12e8244717512941494fe83c94773a58 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 9 Mar 2023 15:57:11 +0000
Subject: net: sfp: add A2h presence flag

The hwmon code wants to know when it is safe to access the A2h data
stored in a separate address. We indicate that this is present when
we have SFF-8472 compliance and the lack of an address-change
sequence.,

The same conditions are also true if we want to access other controls
and status in the A2h address. So let's make a flag to indicate whether
we can access it, instead of repeating the conditions throughout the
code.

For now, only convert the hwmon code.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/sfp.c | 42 ++++++++++++++++++++++++------------------
 1 file changed, 24 insertions(+), 18 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index c02cad6478a8..4ff07b5a5590 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -255,6 +255,8 @@ struct sfp {
 	unsigned int module_power_mW;
 	unsigned int module_t_start_up;
 	unsigned int module_t_wait;
+
+	bool have_a2;
 	bool tx_fault_ignore;
 
 	const struct sfp_quirk *quirk;
@@ -1453,20 +1455,10 @@ static void sfp_hwmon_probe(struct work_struct *work)
 
 static int sfp_hwmon_insert(struct sfp *sfp)
 {
-	if (sfp->id.ext.sff8472_compliance == SFP_SFF8472_COMPLIANCE_NONE)
-		return 0;
-
-	if (!(sfp->id.ext.diagmon & SFP_DIAGMON_DDM))
-		return 0;
-
-	if (sfp->id.ext.diagmon & SFP_DIAGMON_ADDRMODE)
-		/* This driver in general does not support address
-		 * change.
-		 */
-		return 0;
-
-	mod_delayed_work(system_wq, &sfp->hwmon_probe, 1);
-	sfp->hwmon_tries = R_PROBE_RETRY_SLOW;
+	if (sfp->have_a2 && sfp->id.ext.diagmon & SFP_DIAGMON_DDM) {
+		mod_delayed_work(system_wq, &sfp->hwmon_probe, 1);
+		sfp->hwmon_tries = R_PROBE_RETRY_SLOW;
+	}
 
 	return 0;
 }
@@ -1916,6 +1908,18 @@ static int sfp_cotsworks_fixup_check(struct sfp *sfp, struct sfp_eeprom_id *id)
 	return 0;
 }
 
+static int sfp_module_parse_sff8472(struct sfp *sfp)
+{
+	/* If the module requires address swap mode, warn about it */
+	if (sfp->id.ext.diagmon & SFP_DIAGMON_ADDRMODE)
+		dev_warn(sfp->dev,
+			 "module address swap to access page 0xA2 is not supported.\n");
+	else
+		sfp->have_a2 = true;
+
+	return 0;
+}
+
 static int sfp_sm_mod_probe(struct sfp *sfp, bool report)
 {
 	/* SFP module inserted - read I2C data */
@@ -2053,10 +2057,11 @@ static int sfp_sm_mod_probe(struct sfp *sfp, bool report)
 		return -EINVAL;
 	}
 
-	/* If the module requires address swap mode, warn about it */
-	if (sfp->id.ext.diagmon & SFP_DIAGMON_ADDRMODE)
-		dev_warn(sfp->dev,
-			 "module address swap to access page 0xA2 is not supported.\n");
+	if (sfp->id.ext.sff8472_compliance != SFP_SFF8472_COMPLIANCE_NONE) {
+		ret = sfp_module_parse_sff8472(sfp);
+		if (ret < 0)
+			return ret;
+	}
 
 	/* Parse the module power requirement */
 	ret = sfp_module_parse_power(sfp);
@@ -2103,6 +2108,7 @@ static void sfp_sm_mod_remove(struct sfp *sfp)
 
 	memset(&sfp->id, 0, sizeof(sfp->id));
 	sfp->module_power_mW = 0;
+	sfp->have_a2 = false;
 
 	dev_info(sfp->dev, "module removed\n");
 }
-- 
cgit v1.2.3-70-g09d2


From 5daed426f012a1c0db0048339e359ee98a2c8752 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 9 Mar 2023 15:57:16 +0000
Subject: net: sfp: only use soft polling if we have A2h access

The soft state bits are stored in the A2h memory space, and require
SFF-8472 compliance. This is what our have_a2 flag tells us, so use
this to indicate whether we should attempt to use the soft signals.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/sfp.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 4ff07b5a5590..39e3095796d0 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -2284,7 +2284,11 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 		    sfp->sm_dev_state != SFP_DEV_UP)
 			break;
 
-		if (!(sfp->id.ext.diagmon & SFP_DIAGMON_ADDRMODE))
+		/* Only use the soft state bits if we have access to the A2h
+		 * memory, which implies that we have some level of SFF-8472
+		 * compliance.
+		 */
+		if (sfp->have_a2)
 			sfp_soft_start_poll(sfp);
 
 		sfp_module_tx_enable(sfp);
-- 
cgit v1.2.3-70-g09d2


From 939a3f2a76e3216253b0a596b67ce9bd9ff6af51 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 9 Mar 2023 02:54:21 -0800
Subject: ptp_ocp: add force_irq to xilinx_spi configuration

Flashing firmware via devlink flash was failing on PTP OCP devices
because it is using Quad SPI mode, but the driver was not properly
behaving. With force_irq flag landed it now can be fixed.

Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Link: https://lore.kernel.org/r/20230309105421.2953451-1-vadfed@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/ptp/ptp_ocp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c
index 4bbaccd543ad..2b63f3487645 100644
--- a/drivers/ptp/ptp_ocp.c
+++ b/drivers/ptp/ptp_ocp.c
@@ -662,6 +662,7 @@ static struct ocp_resource ocp_fb_resource[] = {
 				.num_chipselect = 1,
 				.bits_per_word = 8,
 				.num_devices = 1,
+				.force_irq = true,
 				.devices = &(struct spi_board_info) {
 					.modalias = "spi-nor",
 				},
-- 
cgit v1.2.3-70-g09d2


From aacaf7b3d19daaa91528ab0c598b89a7f82aa47d Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Thu, 9 Mar 2023 13:06:11 +0530
Subject: dt-bindings: net: ti: k3-am654-cpsw-nuss: Document Serdes PHY

Update bindings to include Serdes PHY as an optional PHY, in addition to
the existing CPSW MAC's PHY. The CPSW MAC's PHY is required while the
Serdes PHY is optional. The Serdes PHY handle has to be provided only
when the Serdes is being configured in a Single-Link protocol. Using the
name "serdes-phy" to represent the Serdes PHY handle, the am65-cpsw-nuss
driver can obtain the Serdes PHY and request the Serdes to be
configured.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml     | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml b/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
index 900063411a20..628d63e1eb1f 100644
--- a/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
+++ b/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
@@ -126,8 +126,18 @@ properties:
             description: CPSW port number
 
           phys:
-            maxItems: 1
-            description: phandle on phy-gmii-sel PHY
+            minItems: 1
+            items:
+              - description: CPSW MAC's PHY.
+              - description: Serdes PHY. Serdes PHY is required only if
+                             the Serdes has to be configured in the
+                             Single-Link configuration.
+
+          phy-names:
+            minItems: 1
+            items:
+              - const: mac
+              - const: serdes
 
           label:
             description: label associated with this port
-- 
cgit v1.2.3-70-g09d2


From bca93b20c3976ff5ca9fe373cf7acd0277492ae8 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Thu, 9 Mar 2023 13:06:12 +0530
Subject: net: ethernet: ti: am65-cpsw: Update name of Serdes PHY

The bindings for the am65-cpsw driver use the name "serdes" to refer to
the Serdes PHY. Thus, update the name used for the Serdes PHY within the
am65_cpsw_init_serdes_phy() function from "serdes-phy" to "serdes".

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 25996826edab..4cfbc1c2b1c4 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -1466,7 +1466,7 @@ static void am65_cpsw_disable_serdes_phy(struct am65_cpsw_common *common)
 static int am65_cpsw_init_serdes_phy(struct device *dev, struct device_node *port_np,
 				     struct am65_cpsw_port *port)
 {
-	const char *name = "serdes-phy";
+	const char *name = "serdes";
 	struct phy *phy;
 	int ret;
 
-- 
cgit v1.2.3-70-g09d2


From 05ccd8d8a15e6b9c99b86cf8a2fd78e3f0c60a84 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 9 Mar 2023 11:59:04 +0000
Subject: sfc: support offloading TC VLAN push/pop actions to the MAE

EF100 can pop and/or push up to two VLAN tags.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230309115904.56442-1-edward.cree@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/mae.c  | 16 ++++++++++++++++
 drivers/net/ethernet/sfc/mcdi.h |  5 +++++
 drivers/net/ethernet/sfc/tc.c   | 42 +++++++++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/tc.h   |  4 ++++
 4 files changed, 67 insertions(+)

diff --git a/drivers/net/ethernet/sfc/mae.c b/drivers/net/ethernet/sfc/mae.c
index 2d32abe5f478..c53d354c1fb2 100644
--- a/drivers/net/ethernet/sfc/mae.c
+++ b/drivers/net/ethernet/sfc/mae.c
@@ -682,6 +682,10 @@ int efx_mae_alloc_action_set(struct efx_nic *efx, struct efx_tc_action_set *act)
 	size_t outlen;
 	int rc;
 
+	MCDI_POPULATE_DWORD_2(inbuf, MAE_ACTION_SET_ALLOC_IN_FLAGS,
+			      MAE_ACTION_SET_ALLOC_IN_VLAN_PUSH, act->vlan_push,
+			      MAE_ACTION_SET_ALLOC_IN_VLAN_POP, act->vlan_pop);
+
 	MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_SRC_MAC_ID,
 		       MC_CMD_MAE_MAC_ADDR_ALLOC_OUT_MAC_ID_NULL);
 	MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_DST_MAC_ID,
@@ -694,6 +698,18 @@ int efx_mae_alloc_action_set(struct efx_nic *efx, struct efx_tc_action_set *act)
 			       MC_CMD_MAE_COUNTER_ALLOC_OUT_COUNTER_ID_NULL);
 	MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_COUNTER_LIST_ID,
 		       MC_CMD_MAE_COUNTER_LIST_ALLOC_OUT_COUNTER_LIST_ID_NULL);
+	if (act->vlan_push) {
+		MCDI_SET_WORD_BE(inbuf, MAE_ACTION_SET_ALLOC_IN_VLAN0_TCI_BE,
+				 act->vlan_tci[0]);
+		MCDI_SET_WORD_BE(inbuf, MAE_ACTION_SET_ALLOC_IN_VLAN0_PROTO_BE,
+				 act->vlan_proto[0]);
+	}
+	if (act->vlan_push >= 2) {
+		MCDI_SET_WORD_BE(inbuf, MAE_ACTION_SET_ALLOC_IN_VLAN1_TCI_BE,
+				 act->vlan_tci[1]);
+		MCDI_SET_WORD_BE(inbuf, MAE_ACTION_SET_ALLOC_IN_VLAN1_PROTO_BE,
+				 act->vlan_proto[1]);
+	}
 	MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_ENCAP_HEADER_ID,
 		       MC_CMD_MAE_ENCAP_HEADER_ALLOC_OUT_ENCAP_HEADER_ID_NULL);
 	if (act->deliver)
diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h
index b139b76febff..454e9d51a4c2 100644
--- a/drivers/net/ethernet/sfc/mcdi.h
+++ b/drivers/net/ethernet/sfc/mcdi.h
@@ -233,6 +233,11 @@ void efx_mcdi_sensor_event(struct efx_nic *efx, efx_qword_t *ev);
 	((void)BUILD_BUG_ON_ZERO(_field ## _LEN != 2),  \
 	le16_to_cpu(*(__force const __le16 *)MCDI_STRUCT_PTR(_buf, _field)))
 /* Write a 16-bit field defined in the protocol as being big-endian. */
+#define MCDI_SET_WORD_BE(_buf, _field, _value) do {			\
+	BUILD_BUG_ON(MC_CMD_ ## _field ## _LEN != 2);			\
+	BUILD_BUG_ON(MC_CMD_ ## _field ## _OFST & 1);			\
+	*(__force __be16 *)MCDI_PTR(_buf, _field) = (_value);		\
+	} while (0)
 #define MCDI_STRUCT_SET_WORD_BE(_buf, _field, _value) do {		\
 	BUILD_BUG_ON(_field ## _LEN != 2);				\
 	BUILD_BUG_ON(_field ## _OFST & 1);				\
diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c
index deeaab9ee761..2b07bb2fd735 100644
--- a/drivers/net/ethernet/sfc/tc.c
+++ b/drivers/net/ethernet/sfc/tc.c
@@ -286,6 +286,8 @@ static int efx_tc_flower_parse_match(struct efx_nic *efx,
 
 /* For details of action order constraints refer to SF-123102-TC-1§12.6.1 */
 enum efx_tc_action_order {
+	EFX_TC_AO_VLAN_POP,
+	EFX_TC_AO_VLAN_PUSH,
 	EFX_TC_AO_COUNT,
 	EFX_TC_AO_DELIVER
 };
@@ -294,6 +296,20 @@ static bool efx_tc_flower_action_order_ok(const struct efx_tc_action_set *act,
 					  enum efx_tc_action_order new)
 {
 	switch (new) {
+	case EFX_TC_AO_VLAN_POP:
+		if (act->vlan_pop >= 2)
+			return false;
+		/* If we've already pushed a VLAN, we can't then pop it;
+		 * the hardware would instead try to pop an existing VLAN
+		 * before pushing the new one.
+		 */
+		if (act->vlan_push)
+			return false;
+		fallthrough;
+	case EFX_TC_AO_VLAN_PUSH:
+		if (act->vlan_push >= 2)
+			return false;
+		fallthrough;
 	case EFX_TC_AO_COUNT:
 		if (act->count)
 			return false;
@@ -393,6 +409,7 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 
 	flow_action_for_each(i, fa, &fr->action) {
 		struct efx_tc_action_set save;
+		u16 tci;
 
 		if (!act) {
 			/* more actions after a non-pipe action */
@@ -494,6 +511,31 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 			}
 			*act = save;
 			break;
+		case FLOW_ACTION_VLAN_POP:
+			if (act->vlan_push) {
+				act->vlan_push--;
+			} else if (efx_tc_flower_action_order_ok(act, EFX_TC_AO_VLAN_POP)) {
+				act->vlan_pop++;
+			} else {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "More than two VLAN pops, or action order violated");
+				rc = -EINVAL;
+				goto release;
+			}
+			break;
+		case FLOW_ACTION_VLAN_PUSH:
+			if (!efx_tc_flower_action_order_ok(act, EFX_TC_AO_VLAN_PUSH)) {
+				rc = -EINVAL;
+				NL_SET_ERR_MSG_MOD(extack,
+						   "More than two VLAN pushes, or action order violated");
+				goto release;
+			}
+			tci = fa->vlan.vid & VLAN_VID_MASK;
+			tci |= fa->vlan.prio << VLAN_PRIO_SHIFT;
+			act->vlan_tci[act->vlan_push] = cpu_to_be16(tci);
+			act->vlan_proto[act->vlan_push] = fa->vlan.proto;
+			act->vlan_push++;
+			break;
 		default:
 			NL_SET_ERR_MSG_FMT_MOD(extack, "Unhandled action %u",
 					       fa->id);
diff --git a/drivers/net/ethernet/sfc/tc.h b/drivers/net/ethernet/sfc/tc.h
index 418ce8c13a06..542853f60c2a 100644
--- a/drivers/net/ethernet/sfc/tc.h
+++ b/drivers/net/ethernet/sfc/tc.h
@@ -19,7 +19,11 @@
 #define IS_ALL_ONES(v)	(!(typeof (v))~(v))
 
 struct efx_tc_action_set {
+	u16 vlan_push:2;
+	u16 vlan_pop:2;
 	u16 deliver:1;
+	__be16 vlan_tci[2]; /* TCIs for vlan_push */
+	__be16 vlan_proto[2]; /* Ethertypes for vlan_push */
 	struct efx_tc_counter_index *count;
 	u32 dest_mport;
 	u32 fw_id; /* index of this entry in firmware actions table */
-- 
cgit v1.2.3-70-g09d2


From 95b744508d4d5135ae2a096ff3f0ee882bcc52b3 Mon Sep 17 00:00:00 2001
From: Muhammad Usama Anjum <usama.anjum@collabora.com>
Date: Fri, 10 Mar 2023 03:52:05 +0500
Subject: qede: remove linux/version.h and linux/compiler.h

make versioncheck reports the following:
./drivers/net/ethernet/qlogic/qede/qede.h: 10 linux/version.h not needed.
./drivers/net/ethernet/qlogic/qede/qede_ethtool.c: 7 linux/version.h not needed.

So remove linux/version.h from both of these files. Also remove
linux/compiler.h while at it as it is also not being used.

Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230309225206.2473644-1-usama.anjum@collabora.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/qede/qede.h         | 2 --
 drivers/net/ethernet/qlogic/qede/qede_ethtool.c | 1 -
 2 files changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index f90dcfe9ee68..f9931ecb7baa 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -6,8 +6,6 @@
 
 #ifndef _QEDE_H_
 #define _QEDE_H_
-#include <linux/compiler.h>
-#include <linux/version.h>
 #include <linux/workqueue.h>
 #include <linux/netdevice.h>
 #include <linux/interrupt.h>
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
index 8034d812d5a0..374a86b875a3 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
@@ -4,7 +4,6 @@
  * Copyright (c) 2019-2020 Marvell International Ltd.
  */
 
-#include <linux/version.h>
 #include <linux/types.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
-- 
cgit v1.2.3-70-g09d2


From 5f24f41e8ea62a6a9095f9bbafb8b3aebe265c68 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 10 Mar 2023 17:26:05 +0800
Subject: xfrm: Remove inner/outer modes from input path

The inner/outer modes were added to abstract out common code that
were once duplicated between IPv4 and IPv6.  As time went on the
abstractions have been removed and we are now left with empty
shells that only contain duplicate information.  These can be
removed one-by-one as the same information is already present
elsewhere in the xfrm_state object.

Removing them from the input path actually allows certain valid
combinations that are currently disallowed.  In particular, when
a transport mode SA sits beneath a tunnel mode SA that changes
address families, at present the transport mode SA cannot have
AF_UNSPEC as its selector because it will be erroneously be treated
as inter-family itself even though it simply sits beneath one.

This is a serious problem because you can't set the selector to
non-AF_UNSPEC either as that will cause the selector match to
fail as we always match selectors to the inner-most traffic.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_input.c | 66 ++++++++++++++++++---------------------------------
 1 file changed, 23 insertions(+), 43 deletions(-)

diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 436d29640ac2..39fb91ff23d9 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -231,9 +231,6 @@ static int xfrm4_remove_tunnel_encap(struct xfrm_state *x, struct sk_buff *skb)
 {
 	int err = -EINVAL;
 
-	if (XFRM_MODE_SKB_CB(skb)->protocol != IPPROTO_IPIP)
-		goto out;
-
 	if (!pskb_may_pull(skb, sizeof(struct iphdr)))
 		goto out;
 
@@ -269,8 +266,6 @@ static int xfrm6_remove_tunnel_encap(struct xfrm_state *x, struct sk_buff *skb)
 {
 	int err = -EINVAL;
 
-	if (XFRM_MODE_SKB_CB(skb)->protocol != IPPROTO_IPV6)
-		goto out;
 	if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
 		goto out;
 
@@ -331,22 +326,26 @@ out:
  */
 static int
 xfrm_inner_mode_encap_remove(struct xfrm_state *x,
-			     const struct xfrm_mode *inner_mode,
 			     struct sk_buff *skb)
 {
-	switch (inner_mode->encap) {
+	switch (x->props.mode) {
 	case XFRM_MODE_BEET:
-		if (inner_mode->family == AF_INET)
+		switch (XFRM_MODE_SKB_CB(skb)->protocol) {
+		case IPPROTO_IPIP:
+		case IPPROTO_BEETPH:
 			return xfrm4_remove_beet_encap(x, skb);
-		if (inner_mode->family == AF_INET6)
+		case IPPROTO_IPV6:
 			return xfrm6_remove_beet_encap(x, skb);
+		}
 		break;
 	case XFRM_MODE_TUNNEL:
-		if (inner_mode->family == AF_INET)
+		switch (XFRM_MODE_SKB_CB(skb)->protocol) {
+		case IPPROTO_IPIP:
 			return xfrm4_remove_tunnel_encap(x, skb);
-		if (inner_mode->family == AF_INET6)
+		case IPPROTO_IPV6:
 			return xfrm6_remove_tunnel_encap(x, skb);
 		break;
+		}
 	}
 
 	WARN_ON_ONCE(1);
@@ -355,9 +354,7 @@ xfrm_inner_mode_encap_remove(struct xfrm_state *x,
 
 static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb)
 {
-	const struct xfrm_mode *inner_mode = &x->inner_mode;
-
-	switch (x->outer_mode.family) {
+	switch (x->props.family) {
 	case AF_INET:
 		xfrm4_extract_header(skb);
 		break;
@@ -369,17 +366,12 @@ static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb)
 		return -EAFNOSUPPORT;
 	}
 
-	if (x->sel.family == AF_UNSPEC) {
-		inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol);
-		if (!inner_mode)
-			return -EAFNOSUPPORT;
-	}
-
-	switch (inner_mode->family) {
-	case AF_INET:
+	switch (XFRM_MODE_SKB_CB(skb)->protocol) {
+	case IPPROTO_IPIP:
+	case IPPROTO_BEETPH:
 		skb->protocol = htons(ETH_P_IP);
 		break;
-	case AF_INET6:
+	case IPPROTO_IPV6:
 		skb->protocol = htons(ETH_P_IPV6);
 		break;
 	default:
@@ -387,7 +379,7 @@ static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb)
 		break;
 	}
 
-	return xfrm_inner_mode_encap_remove(x, inner_mode, skb);
+	return xfrm_inner_mode_encap_remove(x, skb);
 }
 
 /* Remove encapsulation header.
@@ -433,17 +425,16 @@ static int xfrm6_transport_input(struct xfrm_state *x, struct sk_buff *skb)
 }
 
 static int xfrm_inner_mode_input(struct xfrm_state *x,
-				 const struct xfrm_mode *inner_mode,
 				 struct sk_buff *skb)
 {
-	switch (inner_mode->encap) {
+	switch (x->props.mode) {
 	case XFRM_MODE_BEET:
 	case XFRM_MODE_TUNNEL:
 		return xfrm_prepare_input(x, skb);
 	case XFRM_MODE_TRANSPORT:
-		if (inner_mode->family == AF_INET)
+		if (x->props.family == AF_INET)
 			return xfrm4_transport_input(x, skb);
-		if (inner_mode->family == AF_INET6)
+		if (x->props.family == AF_INET6)
 			return xfrm6_transport_input(x, skb);
 		break;
 	case XFRM_MODE_ROUTEOPTIMIZATION:
@@ -461,7 +452,6 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 {
 	const struct xfrm_state_afinfo *afinfo;
 	struct net *net = dev_net(skb->dev);
-	const struct xfrm_mode *inner_mode;
 	int err;
 	__be32 seq;
 	__be32 seq_hi;
@@ -491,7 +481,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			goto drop;
 		}
 
-		family = x->outer_mode.family;
+		family = x->props.family;
 
 		/* An encap_type of -1 indicates async resumption. */
 		if (encap_type == -1) {
@@ -676,17 +666,7 @@ resume:
 
 		XFRM_MODE_SKB_CB(skb)->protocol = nexthdr;
 
-		inner_mode = &x->inner_mode;
-
-		if (x->sel.family == AF_UNSPEC) {
-			inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol);
-			if (inner_mode == NULL) {
-				XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMODEERROR);
-				goto drop;
-			}
-		}
-
-		if (xfrm_inner_mode_input(x, inner_mode, skb)) {
+		if (xfrm_inner_mode_input(x, skb)) {
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMODEERROR);
 			goto drop;
 		}
@@ -701,7 +681,7 @@ resume:
 		 * transport mode so the outer address is identical.
 		 */
 		daddr = &x->id.daddr;
-		family = x->outer_mode.family;
+		family = x->props.family;
 
 		err = xfrm_parse_spi(skb, nexthdr, &spi, &seq);
 		if (err < 0) {
@@ -732,7 +712,7 @@ resume:
 
 		err = -EAFNOSUPPORT;
 		rcu_read_lock();
-		afinfo = xfrm_state_afinfo_get_rcu(x->inner_mode.family);
+		afinfo = xfrm_state_afinfo_get_rcu(x->props.family);
 		if (likely(afinfo))
 			err = afinfo->transport_finish(skb, xfrm_gro || async);
 		rcu_read_unlock();
-- 
cgit v1.2.3-70-g09d2


From f4796398f21b9844017a2dac883b1dd6ad6edd60 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 10 Mar 2023 17:40:32 +0800
Subject: xfrm: Remove inner/outer modes from output path

The inner/outer modes were added to abstract out common code that
were once duplicated between IPv4 and IPv6.  As time went on the
abstractions have been removed and we are now left with empty
shells that only contain duplicate information.  These can be
removed one-by-one as the same information is already present
elsewhere in the xfrm_state object.

Just like the input-side, removing this from the output code
makes it possible to use transport-mode SAs underneath an
inter-family tunnel mode SA.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_output.c | 33 +++++++++++----------------------
 1 file changed, 11 insertions(+), 22 deletions(-)

diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index ff114d68cc43..369e5de8558f 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -412,7 +412,7 @@ static int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE;
 	skb->protocol = htons(ETH_P_IP);
 
-	switch (x->outer_mode.encap) {
+	switch (x->props.mode) {
 	case XFRM_MODE_BEET:
 		return xfrm4_beet_encap_add(x, skb);
 	case XFRM_MODE_TUNNEL:
@@ -435,7 +435,7 @@ static int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 	skb->ignore_df = 1;
 	skb->protocol = htons(ETH_P_IPV6);
 
-	switch (x->outer_mode.encap) {
+	switch (x->props.mode) {
 	case XFRM_MODE_BEET:
 		return xfrm6_beet_encap_add(x, skb);
 	case XFRM_MODE_TUNNEL:
@@ -451,22 +451,22 @@ static int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb)
 
 static int xfrm_outer_mode_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	switch (x->outer_mode.encap) {
+	switch (x->props.mode) {
 	case XFRM_MODE_BEET:
 	case XFRM_MODE_TUNNEL:
-		if (x->outer_mode.family == AF_INET)
+		if (x->props.family == AF_INET)
 			return xfrm4_prepare_output(x, skb);
-		if (x->outer_mode.family == AF_INET6)
+		if (x->props.family == AF_INET6)
 			return xfrm6_prepare_output(x, skb);
 		break;
 	case XFRM_MODE_TRANSPORT:
-		if (x->outer_mode.family == AF_INET)
+		if (x->props.family == AF_INET)
 			return xfrm4_transport_output(x, skb);
-		if (x->outer_mode.family == AF_INET6)
+		if (x->props.family == AF_INET6)
 			return xfrm6_transport_output(x, skb);
 		break;
 	case XFRM_MODE_ROUTEOPTIMIZATION:
-		if (x->outer_mode.family == AF_INET6)
+		if (x->props.family == AF_INET6)
 			return xfrm6_ro_output(x, skb);
 		WARN_ON_ONCE(1);
 		break;
@@ -875,21 +875,10 @@ static int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb)
 
 static int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	const struct xfrm_mode *inner_mode;
-
-	if (x->sel.family == AF_UNSPEC)
-		inner_mode = xfrm_ip2inner_mode(x,
-				xfrm_af2proto(skb_dst(skb)->ops->family));
-	else
-		inner_mode = &x->inner_mode;
-
-	if (inner_mode == NULL)
-		return -EAFNOSUPPORT;
-
-	switch (inner_mode->family) {
-	case AF_INET:
+	switch (skb->protocol) {
+	case htons(ETH_P_IP):
 		return xfrm4_extract_output(x, skb);
-	case AF_INET6:
+	case htons(ETH_P_IPV6):
 		return xfrm6_extract_output(x, skb);
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 25e289e1f52e1f4fb1d07622c6a24f8d8a8e420d Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Wed, 1 Mar 2023 16:20:58 +0200
Subject: wifi: ath11k: use proper regulatory reference for bands

Currently, during regulatory event, 2 GHz/5 GHz is referred
to as 2G/5G including variable names. However, there is no
such entity as 2G or 5G.

Re-name such occurences to its proper name. No functional changes.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.5.0.1-01100-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230110121024.14051-2-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath11k/reg.c | 20 ++++++------
 drivers/net/wireless/ath/ath11k/wmi.c | 58 ++++++++++++++++++-----------------
 drivers/net/wireless/ath/ath11k/wmi.h | 28 ++++++++---------
 3 files changed, 54 insertions(+), 52 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
index 6fae4e61ede7..c231abdf49cb 100644
--- a/drivers/net/wireless/ath/ath11k/reg.c
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -619,7 +619,7 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 	u32 flags;
 	char alpha2[3];
 
-	num_rules = reg_info->num_5g_reg_rules + reg_info->num_2g_reg_rules;
+	num_rules = reg_info->num_5ghz_reg_rules + reg_info->num_2ghz_reg_rules;
 
 	if (!num_rules)
 		goto ret;
@@ -644,20 +644,20 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 		   alpha2, ath11k_reg_get_regdom_str(tmp_regd->dfs_region),
 		   reg_info->dfs_region, num_rules);
 	/* Update reg_rules[] below. Firmware is expected to
-	 * send these rules in order(2G rules first and then 5G)
+	 * send these rules in order(2 GHz rules first and then 5 GHz)
 	 */
 	for (; i < num_rules; i++) {
-		if (reg_info->num_2g_reg_rules &&
-		    (i < reg_info->num_2g_reg_rules)) {
-			reg_rule = reg_info->reg_rules_2g_ptr + i;
+		if (reg_info->num_2ghz_reg_rules &&
+		    (i < reg_info->num_2ghz_reg_rules)) {
+			reg_rule = reg_info->reg_rules_2ghz_ptr + i;
 			max_bw = min_t(u16, reg_rule->max_bw,
-				       reg_info->max_bw_2g);
+				       reg_info->max_bw_2ghz);
 			flags = 0;
-		} else if (reg_info->num_5g_reg_rules &&
-			   (j < reg_info->num_5g_reg_rules)) {
-			reg_rule = reg_info->reg_rules_5g_ptr + j++;
+		} else if (reg_info->num_5ghz_reg_rules &&
+			   (j < reg_info->num_5ghz_reg_rules)) {
+			reg_rule = reg_info->reg_rules_5ghz_ptr + j++;
 			max_bw = min_t(u16, reg_rule->max_bw,
-				       reg_info->max_bw_5g);
+				       reg_info->max_bw_5ghz);
 
 			/* FW doesn't pass NL80211_RRF_AUTO_BW flag for
 			 * BW Auto correction, we can enable this by default
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 815f49c3b3ad..ecd51d285103 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -4959,7 +4959,7 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 	const void **tb;
 	const struct wmi_reg_chan_list_cc_event *chan_list_event_hdr;
 	struct wmi_regulatory_rule_struct *wmi_reg_rule;
-	u32 num_2g_reg_rules, num_5g_reg_rules;
+	u32 num_2ghz_reg_rules, num_5ghz_reg_rules;
 	int ret;
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI, "processing regulatory channel list\n");
@@ -4978,10 +4978,10 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 		return -EPROTO;
 	}
 
-	reg_info->num_2g_reg_rules = chan_list_event_hdr->num_2g_reg_rules;
-	reg_info->num_5g_reg_rules = chan_list_event_hdr->num_5g_reg_rules;
+	reg_info->num_2ghz_reg_rules = chan_list_event_hdr->num_2ghz_reg_rules;
+	reg_info->num_5ghz_reg_rules = chan_list_event_hdr->num_5ghz_reg_rules;
 
-	if (!(reg_info->num_2g_reg_rules + reg_info->num_5g_reg_rules)) {
+	if (!(reg_info->num_2ghz_reg_rules + reg_info->num_5ghz_reg_rules)) {
 		ath11k_warn(ab, "No regulatory rules available in the event info\n");
 		kfree(tb);
 		return -EINVAL;
@@ -5008,46 +5008,48 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_FAIL)
 		reg_info->status_code = REG_SET_CC_STATUS_FAIL;
 
-	reg_info->min_bw_2g = chan_list_event_hdr->min_bw_2g;
-	reg_info->max_bw_2g = chan_list_event_hdr->max_bw_2g;
-	reg_info->min_bw_5g = chan_list_event_hdr->min_bw_5g;
-	reg_info->max_bw_5g = chan_list_event_hdr->max_bw_5g;
+	reg_info->min_bw_2ghz = chan_list_event_hdr->min_bw_2ghz;
+	reg_info->max_bw_2ghz = chan_list_event_hdr->max_bw_2ghz;
+	reg_info->min_bw_5ghz = chan_list_event_hdr->min_bw_5ghz;
+	reg_info->max_bw_5ghz = chan_list_event_hdr->max_bw_5ghz;
 
-	num_2g_reg_rules = reg_info->num_2g_reg_rules;
-	num_5g_reg_rules = reg_info->num_5g_reg_rules;
+	num_2ghz_reg_rules = reg_info->num_2ghz_reg_rules;
+	num_5ghz_reg_rules = reg_info->num_5ghz_reg_rules;
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI,
-		   "%s:cc %s dsf %d BW: min_2g %d max_2g %d min_5g %d max_5g %d",
+		   "%s:cc %s dsf %d BW: min_2ghz %d max_2ghz %d min_5ghz %d max_5ghz %d",
 		   __func__, reg_info->alpha2, reg_info->dfs_region,
-		   reg_info->min_bw_2g, reg_info->max_bw_2g,
-		   reg_info->min_bw_5g, reg_info->max_bw_5g);
+		   reg_info->min_bw_2ghz, reg_info->max_bw_2ghz,
+		   reg_info->min_bw_5ghz, reg_info->max_bw_5ghz);
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI,
-		   "%s: num_2g_reg_rules %d num_5g_reg_rules %d", __func__,
-		   num_2g_reg_rules, num_5g_reg_rules);
+		   "%s: num_2ghz_reg_rules %d num_5ghz_reg_rules %d", __func__,
+		   num_2ghz_reg_rules, num_5ghz_reg_rules);
 
 	wmi_reg_rule =
 		(struct wmi_regulatory_rule_struct *)((u8 *)chan_list_event_hdr
 						+ sizeof(*chan_list_event_hdr)
 						+ sizeof(struct wmi_tlv));
 
-	if (num_2g_reg_rules) {
-		reg_info->reg_rules_2g_ptr = create_reg_rules_from_wmi(num_2g_reg_rules,
-								       wmi_reg_rule);
-		if (!reg_info->reg_rules_2g_ptr) {
+	if (num_2ghz_reg_rules) {
+		reg_info->reg_rules_2ghz_ptr =
+				create_reg_rules_from_wmi(num_2ghz_reg_rules,
+							  wmi_reg_rule);
+		if (!reg_info->reg_rules_2ghz_ptr) {
 			kfree(tb);
-			ath11k_warn(ab, "Unable to Allocate memory for 2g rules\n");
+			ath11k_warn(ab, "Unable to Allocate memory for 2 GHz rules\n");
 			return -ENOMEM;
 		}
 	}
 
-	if (num_5g_reg_rules) {
-		wmi_reg_rule += num_2g_reg_rules;
-		reg_info->reg_rules_5g_ptr = create_reg_rules_from_wmi(num_5g_reg_rules,
-								       wmi_reg_rule);
-		if (!reg_info->reg_rules_5g_ptr) {
+	if (num_5ghz_reg_rules) {
+		wmi_reg_rule += num_2ghz_reg_rules;
+		reg_info->reg_rules_5ghz_ptr =
+				create_reg_rules_from_wmi(num_5ghz_reg_rules,
+							  wmi_reg_rule);
+		if (!reg_info->reg_rules_5ghz_ptr) {
 			kfree(tb);
-			ath11k_warn(ab, "Unable to Allocate memory for 5g rules\n");
+			ath11k_warn(ab, "Unable to Allocate memory for 5 GHz rules\n");
 			return -ENOMEM;
 		}
 	}
@@ -6619,8 +6621,8 @@ fallback:
 	WARN_ON(1);
 mem_free:
 	if (reg_info) {
-		kfree(reg_info->reg_rules_2g_ptr);
-		kfree(reg_info->reg_rules_5g_ptr);
+		kfree(reg_info->reg_rules_2ghz_ptr);
+		kfree(reg_info->reg_rules_5ghz_ptr);
 		kfree(reg_info);
 	}
 	return ret;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 63dbb085949f..858403f190c2 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -4129,14 +4129,14 @@ struct cur_regulatory_info {
 	u8 alpha2[REG_ALPHA2_LEN + 1];
 	u32 dfs_region;
 	u32 phybitmap;
-	u32 min_bw_2g;
-	u32 max_bw_2g;
-	u32 min_bw_5g;
-	u32 max_bw_5g;
-	u32 num_2g_reg_rules;
-	u32 num_5g_reg_rules;
-	struct cur_reg_rule *reg_rules_2g_ptr;
-	struct cur_reg_rule *reg_rules_5g_ptr;
+	u32 min_bw_2ghz;
+	u32 max_bw_2ghz;
+	u32 min_bw_5ghz;
+	u32 max_bw_5ghz;
+	u32 num_2ghz_reg_rules;
+	u32 num_5ghz_reg_rules;
+	struct cur_reg_rule *reg_rules_2ghz_ptr;
+	struct cur_reg_rule *reg_rules_5ghz_ptr;
 };
 
 struct wmi_reg_chan_list_cc_event {
@@ -4148,12 +4148,12 @@ struct wmi_reg_chan_list_cc_event {
 	u32 domain_code;
 	u32 dfs_region;
 	u32 phybitmap;
-	u32 min_bw_2g;
-	u32 max_bw_2g;
-	u32 min_bw_5g;
-	u32 max_bw_5g;
-	u32 num_2g_reg_rules;
-	u32 num_5g_reg_rules;
+	u32 min_bw_2ghz;
+	u32 max_bw_2ghz;
+	u32 min_bw_5ghz;
+	u32 max_bw_5ghz;
+	u32 num_2ghz_reg_rules;
+	u32 num_5ghz_reg_rules;
 } __packed;
 
 struct wmi_regulatory_rule_struct {
-- 
cgit v1.2.3-70-g09d2


From 91fa00fa69224aae5afb720c5e68b22e4c4f7333 Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Wed, 1 Mar 2023 16:20:59 +0200
Subject: wifi: ath11k: add support to parse new WMI event for 6 GHz

In order to support different power levels of 6 GHz AP and client,
new WMI event for regulatory - WMI_REG_CHAN_LIST_CC_EXT_EVENTID is
added in firmware. This event provides new parameters required for
6 GHz regulatory rules.

Add support for parsing 2.4 GHz, 5 GHz and 6 GHz reg rules and other
parameters from WMI_REG_CHAN_LIST_CC_EXT_EVENTID.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.5.0.1-01100-QCAHKSWPL_SILICONZ-1

Signed-off-by: Lavanya Suresh <quic_lavaks@quicinc.com>
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230110121024.14051-3-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath11k/reg.c |  37 ++-
 drivers/net/wireless/ath/ath11k/wmi.c | 418 ++++++++++++++++++++++++++++++++--
 drivers/net/wireless/ath/ath11k/wmi.h | 163 ++++++++++++-
 3 files changed, 584 insertions(+), 34 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
index c231abdf49cb..08621bd13a12 100644
--- a/drivers/net/wireless/ath/ath11k/reg.c
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -613,7 +613,7 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 {
 	struct ieee80211_regdomain *tmp_regd, *default_regd, *new_regd = NULL;
 	struct cur_reg_rule *reg_rule;
-	u8 i = 0, j = 0;
+	u8 i = 0, j = 0, k = 0;
 	u8 num_rules;
 	u16 max_bw;
 	u32 flags;
@@ -621,6 +621,12 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 
 	num_rules = reg_info->num_5ghz_reg_rules + reg_info->num_2ghz_reg_rules;
 
+	/* FIXME: Currently taking reg rules for 6 GHz only from Indoor AP mode list.
+	 * This can be updated after complete 6 GHz regulatory support is added.
+	 */
+	if (reg_info->is_ext_reg_event)
+		num_rules += reg_info->num_6ghz_rules_ap[WMI_REG_INDOOR_AP];
+
 	if (!num_rules)
 		goto ret;
 
@@ -666,6 +672,14 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 			 * per other BW rule flags we pass from here
 			 */
 			flags = NL80211_RRF_AUTO_BW;
+		} else if (reg_info->is_ext_reg_event &&
+			   reg_info->num_6ghz_rules_ap[WMI_REG_INDOOR_AP] &&
+			   (k < reg_info->num_6ghz_rules_ap[WMI_REG_INDOOR_AP])) {
+			reg_rule = reg_info->reg_rules_6ghz_ap_ptr[WMI_REG_INDOOR_AP] +
+				   k++;
+			max_bw = min_t(u16, reg_rule->max_bw,
+				       reg_info->max_bw_6ghz_ap[WMI_REG_INDOOR_AP]);
+			flags = NL80211_RRF_AUTO_BW;
 		} else {
 			break;
 		}
@@ -693,12 +707,21 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 			continue;
 		}
 
-		ath11k_dbg(ab, ATH11K_DBG_REG,
-			   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d)\n",
-			   i + 1, reg_rule->start_freq, reg_rule->end_freq,
-			   max_bw, reg_rule->ant_gain, reg_rule->reg_power,
-			   tmp_regd->reg_rules[i].dfs_cac_ms,
-			   flags);
+		if (reg_info->is_ext_reg_event) {
+			ath11k_dbg(ab, ATH11K_DBG_REG,
+				   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d) (%d, %d)\n",
+				   i + 1, reg_rule->start_freq, reg_rule->end_freq,
+				   max_bw, reg_rule->ant_gain, reg_rule->reg_power,
+				   tmp_regd->reg_rules[i].dfs_cac_ms, flags,
+				   reg_rule->psd_flag, reg_rule->psd_eirp);
+		} else {
+			ath11k_dbg(ab, ATH11K_DBG_REG,
+				   "\t%d. (%d - %d @ %d) (%d, %d) (%d ms) (FLAGS %d)\n",
+				   i + 1, reg_rule->start_freq, reg_rule->end_freq,
+				   max_bw, reg_rule->ant_gain, reg_rule->reg_power,
+				   tmp_regd->reg_rules[i].dfs_cac_ms,
+				   flags);
+		}
 	}
 
 	tmp_regd->n_reg_rules = i;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index ecd51d285103..34731d0d1efc 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -105,6 +105,8 @@ static const struct wmi_tlv_policy wmi_tlv_policies[] = {
 		= { .min_len = sizeof(struct wmi_vdev_stopped_event) },
 	[WMI_TAG_REG_CHAN_LIST_CC_EVENT]
 		= { .min_len = sizeof(struct wmi_reg_chan_list_cc_event) },
+	[WMI_TAG_REG_CHAN_LIST_CC_EXT_EVENT]
+		= { .min_len = sizeof(struct wmi_reg_chan_list_cc_ext_event) },
 	[WMI_TAG_MGMT_RX_HDR]
 		= { .min_len = sizeof(struct wmi_mgmt_rx_hdr) },
 	[WMI_TAG_MGMT_TX_COMPL_EVENT]
@@ -3974,6 +3976,10 @@ ath11k_wmi_copy_resource_config(struct wmi_resource_config *wmi_cfg,
 	wmi_cfg->sched_params = tg_cfg->sched_params;
 	wmi_cfg->twt_ap_pdev_count = tg_cfg->twt_ap_pdev_count;
 	wmi_cfg->twt_ap_sta_count = tg_cfg->twt_ap_sta_count;
+	wmi_cfg->host_service_flags &=
+		~(1 << WMI_CFG_HOST_SERVICE_FLAG_REG_CC_EXT);
+	wmi_cfg->host_service_flags |= (tg_cfg->is_reg_cc_ext_event_supported <<
+					WMI_CFG_HOST_SERVICE_FLAG_REG_CC_EXT);
 }
 
 static int ath11k_init_cmd_send(struct ath11k_pdev_wmi *wmi,
@@ -4192,6 +4198,10 @@ int ath11k_wmi_cmd_init(struct ath11k_base *ab)
 
 	ab->hw_params.hw_ops->wmi_init_config(ab, &config);
 
+	if (test_bit(WMI_TLV_SERVICE_REG_CC_EXT_EVENT_SUPPORT,
+		     ab->wmi_ab.svc_map))
+		config.is_reg_cc_ext_event_supported = 1;
+
 	memcpy(&wmi_sc->wlan_resource_config, &config, sizeof(config));
 
 	init_param.res_cfg = &wmi_sc->wlan_resource_config;
@@ -4995,18 +5005,11 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 	reg_info->phy_id = chan_list_event_hdr->phy_id;
 	reg_info->ctry_code = chan_list_event_hdr->country_id;
 	reg_info->reg_dmn_pair = chan_list_event_hdr->domain_code;
-	if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_PASS)
-		reg_info->status_code = REG_SET_CC_STATUS_PASS;
-	else if (chan_list_event_hdr->status_code == WMI_REG_CURRENT_ALPHA2_NOT_FOUND)
-		reg_info->status_code = REG_CURRENT_ALPHA2_NOT_FOUND;
-	else if (chan_list_event_hdr->status_code == WMI_REG_INIT_ALPHA2_NOT_FOUND)
-		reg_info->status_code = REG_INIT_ALPHA2_NOT_FOUND;
-	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_CHANGE_NOT_ALLOWED)
-		reg_info->status_code = REG_SET_CC_CHANGE_NOT_ALLOWED;
-	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_NO_MEMORY)
-		reg_info->status_code = REG_SET_CC_STATUS_NO_MEMORY;
-	else if (chan_list_event_hdr->status_code == WMI_REG_SET_CC_STATUS_FAIL)
-		reg_info->status_code = REG_SET_CC_STATUS_FAIL;
+
+	reg_info->status_code =
+		ath11k_wmi_cc_setting_code_to_reg(chan_list_event_hdr->status_code);
+
+	reg_info->is_ext_reg_event = false;
 
 	reg_info->min_bw_2ghz = chan_list_event_hdr->min_bw_2ghz;
 	reg_info->max_bw_2ghz = chan_list_event_hdr->max_bw_2ghz;
@@ -5060,6 +5063,372 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 	return 0;
 }
 
+static struct cur_reg_rule
+*create_ext_reg_rules_from_wmi(u32 num_reg_rules,
+			       struct wmi_regulatory_ext_rule *wmi_reg_rule)
+{
+	struct cur_reg_rule *reg_rule_ptr;
+	u32 count;
+
+	reg_rule_ptr =  kcalloc(num_reg_rules, sizeof(*reg_rule_ptr), GFP_ATOMIC);
+
+	if (!reg_rule_ptr)
+		return NULL;
+
+	for (count = 0; count < num_reg_rules; count++) {
+		reg_rule_ptr[count].start_freq =
+			u32_get_bits(wmi_reg_rule[count].freq_info,
+				     REG_RULE_START_FREQ);
+		reg_rule_ptr[count].end_freq =
+			u32_get_bits(wmi_reg_rule[count].freq_info,
+				     REG_RULE_END_FREQ);
+		reg_rule_ptr[count].max_bw =
+			u32_get_bits(wmi_reg_rule[count].bw_pwr_info,
+				     REG_RULE_MAX_BW);
+		reg_rule_ptr[count].reg_power =
+			u32_get_bits(wmi_reg_rule[count].bw_pwr_info,
+				     REG_RULE_REG_PWR);
+		reg_rule_ptr[count].ant_gain =
+			u32_get_bits(wmi_reg_rule[count].bw_pwr_info,
+				     REG_RULE_ANT_GAIN);
+		reg_rule_ptr[count].flags =
+			u32_get_bits(wmi_reg_rule[count].flag_info,
+				     REG_RULE_FLAGS);
+		reg_rule_ptr[count].psd_flag =
+			u32_get_bits(wmi_reg_rule[count].psd_power_info,
+				     REG_RULE_PSD_INFO);
+		reg_rule_ptr[count].psd_eirp =
+			u32_get_bits(wmi_reg_rule[count].psd_power_info,
+				     REG_RULE_PSD_EIRP);
+	}
+
+	return reg_rule_ptr;
+}
+
+static u8
+ath11k_invalid_5ghz_reg_ext_rules_from_wmi(u32 num_reg_rules,
+					   const struct wmi_regulatory_ext_rule *rule)
+{
+	u8 num_invalid_5ghz_rules = 0;
+	u32 count, start_freq;
+
+	for (count = 0; count < num_reg_rules; count++) {
+		start_freq = u32_get_bits(rule[count].freq_info,
+					  REG_RULE_START_FREQ);
+
+		if (start_freq >= ATH11K_MIN_6G_FREQ)
+			num_invalid_5ghz_rules++;
+	}
+
+	return num_invalid_5ghz_rules;
+}
+
+static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
+						   struct sk_buff *skb,
+						   struct cur_regulatory_info *reg_info)
+{
+	const void **tb;
+	const struct wmi_reg_chan_list_cc_ext_event *ext_chan_list_event_hdr;
+	struct wmi_regulatory_ext_rule *ext_wmi_reg_rule;
+	u32 num_2ghz_reg_rules, num_5ghz_reg_rules;
+	u32 num_6ghz_reg_rules_ap[WMI_REG_CURRENT_MAX_AP_TYPE];
+	u32 num_6ghz_client[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE];
+	u32 total_reg_rules = 0;
+	int ret, i, j, num_invalid_5ghz_ext_rules = 0;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "processing regulatory ext channel list\n");
+
+	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return ret;
+	}
+
+	ext_chan_list_event_hdr = tb[WMI_TAG_REG_CHAN_LIST_CC_EXT_EVENT];
+	if (!ext_chan_list_event_hdr) {
+		ath11k_warn(ab, "failed to fetch reg chan list ext update ev\n");
+		kfree(tb);
+		return -EPROTO;
+	}
+
+	reg_info->num_2ghz_reg_rules =
+			ext_chan_list_event_hdr->num_2ghz_reg_rules;
+	reg_info->num_5ghz_reg_rules =
+			ext_chan_list_event_hdr->num_5ghz_reg_rules;
+	reg_info->num_6ghz_rules_ap[WMI_REG_INDOOR_AP] =
+			ext_chan_list_event_hdr->num_6ghz_reg_rules_ap_lpi;
+	reg_info->num_6ghz_rules_ap[WMI_REG_STANDARD_POWER_AP] =
+			ext_chan_list_event_hdr->num_6ghz_reg_rules_ap_sp;
+	reg_info->num_6ghz_rules_ap[WMI_REG_VERY_LOW_POWER_AP] =
+			ext_chan_list_event_hdr->num_6ghz_reg_rules_ap_vlp;
+
+	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
+		reg_info->num_6ghz_rules_client[WMI_REG_INDOOR_AP][i] =
+			ext_chan_list_event_hdr->num_6ghz_reg_rules_client_lpi[i];
+		reg_info->num_6ghz_rules_client[WMI_REG_STANDARD_POWER_AP][i] =
+			ext_chan_list_event_hdr->num_6ghz_reg_rules_client_sp[i];
+		reg_info->num_6ghz_rules_client[WMI_REG_VERY_LOW_POWER_AP][i] =
+			ext_chan_list_event_hdr->num_6ghz_reg_rules_client_vlp[i];
+	}
+
+	num_2ghz_reg_rules = reg_info->num_2ghz_reg_rules;
+	num_5ghz_reg_rules = reg_info->num_5ghz_reg_rules;
+
+	total_reg_rules += num_2ghz_reg_rules;
+	total_reg_rules += num_5ghz_reg_rules;
+
+	if ((num_2ghz_reg_rules > MAX_REG_RULES) ||
+	    (num_5ghz_reg_rules > MAX_REG_RULES)) {
+		ath11k_warn(ab, "Num reg rules for 2.4 GHz/5 GHz exceeds max limit (num_2ghz_reg_rules: %d num_5ghz_reg_rules: %d max_rules: %d)\n",
+			    num_2ghz_reg_rules, num_5ghz_reg_rules, MAX_REG_RULES);
+		kfree(tb);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < WMI_REG_CURRENT_MAX_AP_TYPE; i++) {
+		num_6ghz_reg_rules_ap[i] = reg_info->num_6ghz_rules_ap[i];
+
+		if (num_6ghz_reg_rules_ap[i] > MAX_6GHZ_REG_RULES) {
+			ath11k_warn(ab, "Num 6 GHz reg rules for AP mode(%d) exceeds max limit (num_6ghz_reg_rules_ap: %d, max_rules: %d)\n",
+				    i, num_6ghz_reg_rules_ap[i], MAX_6GHZ_REG_RULES);
+			kfree(tb);
+			return -EINVAL;
+		}
+
+		total_reg_rules += num_6ghz_reg_rules_ap[i];
+	}
+
+	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
+		num_6ghz_client[WMI_REG_INDOOR_AP][i] =
+			reg_info->num_6ghz_rules_client[WMI_REG_INDOOR_AP][i];
+		total_reg_rules += num_6ghz_client[WMI_REG_INDOOR_AP][i];
+
+		num_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
+			reg_info->num_6ghz_rules_client[WMI_REG_STANDARD_POWER_AP][i];
+		total_reg_rules += num_6ghz_client[WMI_REG_STANDARD_POWER_AP][i];
+
+		num_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
+			reg_info->num_6ghz_rules_client[WMI_REG_VERY_LOW_POWER_AP][i];
+		total_reg_rules += num_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i];
+
+		if ((num_6ghz_client[WMI_REG_INDOOR_AP][i] > MAX_6GHZ_REG_RULES) ||
+		    (num_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] >
+							     MAX_6GHZ_REG_RULES) ||
+		    (num_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] >
+							     MAX_6GHZ_REG_RULES)) {
+			ath11k_warn(ab,
+				    "Num 6 GHz client reg rules exceeds max limit, for client(type: %d)\n",
+				    i);
+			kfree(tb);
+			return -EINVAL;
+		}
+	}
+
+	if (!total_reg_rules) {
+		ath11k_warn(ab, "No reg rules available\n");
+		kfree(tb);
+		return -EINVAL;
+	}
+
+	memcpy(reg_info->alpha2, &ext_chan_list_event_hdr->alpha2,
+	       REG_ALPHA2_LEN);
+
+	reg_info->dfs_region = ext_chan_list_event_hdr->dfs_region;
+	reg_info->phybitmap = ext_chan_list_event_hdr->phybitmap;
+	reg_info->num_phy = ext_chan_list_event_hdr->num_phy;
+	reg_info->phy_id = ext_chan_list_event_hdr->phy_id;
+	reg_info->ctry_code = ext_chan_list_event_hdr->country_id;
+	reg_info->reg_dmn_pair = ext_chan_list_event_hdr->domain_code;
+
+	reg_info->status_code =
+		ath11k_wmi_cc_setting_code_to_reg(ext_chan_list_event_hdr->status_code);
+
+	reg_info->is_ext_reg_event = true;
+
+	reg_info->min_bw_2ghz = ext_chan_list_event_hdr->min_bw_2ghz;
+	reg_info->max_bw_2ghz = ext_chan_list_event_hdr->max_bw_2ghz;
+	reg_info->min_bw_5ghz = ext_chan_list_event_hdr->min_bw_5ghz;
+	reg_info->max_bw_5ghz = ext_chan_list_event_hdr->max_bw_5ghz;
+
+	reg_info->min_bw_6ghz_ap[WMI_REG_INDOOR_AP] =
+			ext_chan_list_event_hdr->min_bw_6ghz_ap_lpi;
+	reg_info->max_bw_6ghz_ap[WMI_REG_INDOOR_AP] =
+			 ext_chan_list_event_hdr->max_bw_6ghz_ap_lpi;
+	reg_info->min_bw_6ghz_ap[WMI_REG_STANDARD_POWER_AP] =
+			ext_chan_list_event_hdr->min_bw_6ghz_ap_sp;
+	reg_info->max_bw_6ghz_ap[WMI_REG_STANDARD_POWER_AP] =
+			ext_chan_list_event_hdr->max_bw_6ghz_ap_sp;
+	reg_info->min_bw_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP] =
+			ext_chan_list_event_hdr->min_bw_6ghz_ap_vlp;
+	reg_info->max_bw_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP] =
+			ext_chan_list_event_hdr->max_bw_6ghz_ap_vlp;
+
+	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
+		reg_info->min_bw_6ghz_client[WMI_REG_INDOOR_AP][i] =
+				ext_chan_list_event_hdr->min_bw_6ghz_client_lpi[i];
+		reg_info->max_bw_6ghz_client[WMI_REG_INDOOR_AP][i] =
+				ext_chan_list_event_hdr->max_bw_6ghz_client_lpi[i];
+		reg_info->min_bw_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
+				ext_chan_list_event_hdr->min_bw_6ghz_client_sp[i];
+		reg_info->max_bw_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
+				ext_chan_list_event_hdr->max_bw_6ghz_client_sp[i];
+		reg_info->min_bw_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
+				ext_chan_list_event_hdr->min_bw_6ghz_client_vlp[i];
+		reg_info->max_bw_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
+				ext_chan_list_event_hdr->max_bw_6ghz_client_vlp[i];
+	}
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "%s:cc_ext %s dsf %d BW: min_2ghz %d max_2ghz %d min_5ghz %d max_5ghz %d",
+		   __func__, reg_info->alpha2, reg_info->dfs_region,
+		   reg_info->min_bw_2ghz, reg_info->max_bw_2ghz,
+		   reg_info->min_bw_5ghz, reg_info->max_bw_5ghz);
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "num_2ghz_reg_rules %d num_5ghz_reg_rules %d",
+		   num_2ghz_reg_rules, num_5ghz_reg_rules);
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "num_6ghz_reg_rules_ap_lpi: %d num_6ghz_reg_rules_ap_sp: %d num_6ghz_reg_rules_ap_vlp: %d",
+		   num_6ghz_reg_rules_ap[WMI_REG_INDOOR_AP],
+		   num_6ghz_reg_rules_ap[WMI_REG_STANDARD_POWER_AP],
+		   num_6ghz_reg_rules_ap[WMI_REG_VERY_LOW_POWER_AP]);
+
+	j = WMI_REG_DEFAULT_CLIENT;
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "6 GHz Regular client: num_6ghz_reg_rules_lpi: %d num_6ghz_reg_rules_sp: %d num_6ghz_reg_rules_vlp: %d",
+		   num_6ghz_client[WMI_REG_INDOOR_AP][j],
+		   num_6ghz_client[WMI_REG_STANDARD_POWER_AP][j],
+		   num_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][j]);
+
+	j = WMI_REG_SUBORDINATE_CLIENT;
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "6 GHz Subordinate client: num_6ghz_reg_rules_lpi: %d num_6ghz_reg_rules_sp: %d num_6ghz_reg_rules_vlp: %d",
+		   num_6ghz_client[WMI_REG_INDOOR_AP][j],
+		   num_6ghz_client[WMI_REG_STANDARD_POWER_AP][j],
+		   num_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][j]);
+
+	ext_wmi_reg_rule =
+		(struct wmi_regulatory_ext_rule *)((u8 *)ext_chan_list_event_hdr
+						   + sizeof(*ext_chan_list_event_hdr)
+						   + sizeof(struct wmi_tlv));
+	if (num_2ghz_reg_rules) {
+		reg_info->reg_rules_2ghz_ptr =
+			create_ext_reg_rules_from_wmi(num_2ghz_reg_rules,
+						      ext_wmi_reg_rule);
+
+		if (!reg_info->reg_rules_2ghz_ptr) {
+			kfree(tb);
+			ath11k_warn(ab, "Unable to Allocate memory for 2 GHz rules\n");
+			return -ENOMEM;
+		}
+	}
+
+	ext_wmi_reg_rule += num_2ghz_reg_rules;
+
+	/* Firmware might include 6 GHz reg rule in 5 GHz rule list
+	 * for few countries along with separate 6 GHz rule.
+	 * Having same 6 GHz reg rule in 5 GHz and 6 GHz rules list
+	 * causes intersect check to be true, and same rules will be
+	 * shown multiple times in iw cmd.
+	 * Hence, avoid parsing 6 GHz rule from 5 GHz reg rule list
+	 */
+	num_invalid_5ghz_ext_rules =
+		ath11k_invalid_5ghz_reg_ext_rules_from_wmi(num_5ghz_reg_rules,
+							   ext_wmi_reg_rule);
+
+	if (num_invalid_5ghz_ext_rules) {
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "CC: %s 5 GHz reg rules number %d from fw, %d number of invalid 5 GHz rules",
+			   reg_info->alpha2, reg_info->num_5ghz_reg_rules,
+			   num_invalid_5ghz_ext_rules);
+
+		num_5ghz_reg_rules = num_5ghz_reg_rules - num_invalid_5ghz_ext_rules;
+		reg_info->num_5ghz_reg_rules = num_5ghz_reg_rules;
+	}
+
+	if (num_5ghz_reg_rules) {
+		reg_info->reg_rules_5ghz_ptr =
+			create_ext_reg_rules_from_wmi(num_5ghz_reg_rules,
+						      ext_wmi_reg_rule);
+
+		if (!reg_info->reg_rules_5ghz_ptr) {
+			kfree(tb);
+			ath11k_warn(ab, "Unable to Allocate memory for 5 GHz rules\n");
+			return -ENOMEM;
+		}
+	}
+
+	/* We have adjusted the number of 5 GHz reg rules above. But still those
+	 * many rules needs to be adjusted in ext_wmi_reg_rule.
+	 *
+	 * NOTE: num_invalid_5ghz_ext_rules will be 0 for rest other cases.
+	 */
+	ext_wmi_reg_rule += (num_5ghz_reg_rules + num_invalid_5ghz_ext_rules);
+
+	for (i = 0; i < WMI_REG_CURRENT_MAX_AP_TYPE; i++) {
+		reg_info->reg_rules_6ghz_ap_ptr[i] =
+			create_ext_reg_rules_from_wmi(num_6ghz_reg_rules_ap[i],
+						      ext_wmi_reg_rule);
+
+		if (!reg_info->reg_rules_6ghz_ap_ptr[i]) {
+			kfree(tb);
+			ath11k_warn(ab, "Unable to Allocate memory for 6 GHz AP rules\n");
+			return -ENOMEM;
+		}
+
+		ext_wmi_reg_rule += num_6ghz_reg_rules_ap[i];
+	}
+
+	for (j = 0; j < WMI_REG_CURRENT_MAX_AP_TYPE; j++) {
+		for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
+			reg_info->reg_rules_6ghz_client_ptr[j][i] =
+				create_ext_reg_rules_from_wmi(num_6ghz_client[j][i],
+							      ext_wmi_reg_rule);
+
+			if (!reg_info->reg_rules_6ghz_client_ptr[j][i]) {
+				kfree(tb);
+				ath11k_warn(ab, "Unable to Allocate memory for 6 GHz client rules\n");
+				return -ENOMEM;
+			}
+
+			ext_wmi_reg_rule += num_6ghz_client[j][i];
+		}
+	}
+
+	reg_info->client_type = ext_chan_list_event_hdr->client_type;
+	reg_info->rnr_tpe_usable = ext_chan_list_event_hdr->rnr_tpe_usable;
+	reg_info->unspecified_ap_usable =
+			ext_chan_list_event_hdr->unspecified_ap_usable;
+	reg_info->domain_code_6ghz_ap[WMI_REG_INDOOR_AP] =
+			ext_chan_list_event_hdr->domain_code_6ghz_ap_lpi;
+	reg_info->domain_code_6ghz_ap[WMI_REG_STANDARD_POWER_AP] =
+			ext_chan_list_event_hdr->domain_code_6ghz_ap_sp;
+	reg_info->domain_code_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP] =
+			ext_chan_list_event_hdr->domain_code_6ghz_ap_vlp;
+
+	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
+		reg_info->domain_code_6ghz_client[WMI_REG_INDOOR_AP][i] =
+				ext_chan_list_event_hdr->domain_code_6ghz_client_lpi[i];
+		reg_info->domain_code_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
+				ext_chan_list_event_hdr->domain_code_6ghz_client_sp[i];
+		reg_info->domain_code_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
+				ext_chan_list_event_hdr->domain_code_6ghz_client_vlp[i];
+	}
+
+	reg_info->domain_code_6ghz_super_id =
+			ext_chan_list_event_hdr->domain_code_6ghz_super_id;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "6 GHz client_type: %d domain_code_6ghz_super_id: %d",
+		   reg_info->client_type, reg_info->domain_code_6ghz_super_id);
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "processed regulatory ext channel list\n");
+
+	kfree(tb);
+	return 0;
+}
+
 static int ath11k_pull_peer_del_resp_ev(struct ath11k_base *ab, struct sk_buff *skb,
 					struct wmi_peer_delete_resp_event *peer_del_resp)
 {
@@ -6507,12 +6876,14 @@ static bool ath11k_reg_is_world_alpha(char *alpha)
 	return false;
 }
 
-static int ath11k_reg_chan_list_event(struct ath11k_base *ab, struct sk_buff *skb)
+static int ath11k_reg_chan_list_event(struct ath11k_base *ab,
+				      struct sk_buff *skb,
+				      enum wmi_reg_chan_list_cmd_type id)
 {
 	struct cur_regulatory_info *reg_info = NULL;
 	struct ieee80211_regdomain *regd = NULL;
 	bool intersect = false;
-	int ret = 0, pdev_idx;
+	int ret = 0, pdev_idx, i, j;
 	struct ath11k *ar;
 
 	reg_info = kzalloc(sizeof(*reg_info), GFP_ATOMIC);
@@ -6521,7 +6892,11 @@ static int ath11k_reg_chan_list_event(struct ath11k_base *ab, struct sk_buff *sk
 		goto fallback;
 	}
 
-	ret = ath11k_pull_reg_chan_list_update_ev(ab, skb, reg_info);
+	if (id == WMI_REG_CHAN_LIST_CC_ID)
+		ret = ath11k_pull_reg_chan_list_update_ev(ab, skb, reg_info);
+	else
+		ret = ath11k_pull_reg_chan_list_ext_update_ev(ab, skb, reg_info);
+
 	if (ret) {
 		ath11k_warn(ab, "failed to extract regulatory info from received event\n");
 		goto fallback;
@@ -6623,6 +6998,14 @@ mem_free:
 	if (reg_info) {
 		kfree(reg_info->reg_rules_2ghz_ptr);
 		kfree(reg_info->reg_rules_5ghz_ptr);
+		if (reg_info->is_ext_reg_event) {
+			for (i = 0; i < WMI_REG_CURRENT_MAX_AP_TYPE; i++)
+				kfree(reg_info->reg_rules_6ghz_ap_ptr[i]);
+
+			for (j = 0; j < WMI_REG_CURRENT_MAX_AP_TYPE; j++)
+				for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++)
+					kfree(reg_info->reg_rules_6ghz_client_ptr[j][i]);
+		}
 		kfree(reg_info);
 	}
 	return ret;
@@ -8054,7 +8437,10 @@ static void ath11k_wmi_tlv_op_rx(struct ath11k_base *ab, struct sk_buff *skb)
 		ath11k_service_ready_ext2_event(ab, skb);
 		break;
 	case WMI_REG_CHAN_LIST_CC_EVENTID:
-		ath11k_reg_chan_list_event(ab, skb);
+		ath11k_reg_chan_list_event(ab, skb, WMI_REG_CHAN_LIST_CC_ID);
+		break;
+	case WMI_REG_CHAN_LIST_CC_EXT_EVENTID:
+		ath11k_reg_chan_list_event(ab, skb, WMI_REG_CHAN_LIST_CC_EXT_ID);
 		break;
 	case WMI_READY_EVENTID:
 		ath11k_ready_event(ab, skb);
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 858403f190c2..3ec3fa0aa5c8 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -797,6 +797,7 @@ enum wmi_tlv_event_id {
 	WMI_RMC_NEW_LEADER_EVENTID = WMI_TLV_CMD(WMI_GRP_RMC),
 	WMI_REG_CHAN_LIST_CC_EVENTID = WMI_TLV_CMD(WMI_GRP_REGULATORY),
 	WMI_11D_NEW_COUNTRY_EVENTID,
+	WMI_REG_CHAN_LIST_CC_EXT_EVENTID,
 	WMI_NDI_CAP_RSP_EVENTID = WMI_TLV_CMD(WMI_GRP_PROTOTYPE),
 	WMI_NDP_INITIATOR_RSP_EVENTID,
 	WMI_NDP_RESPONDER_RSP_EVENTID,
@@ -1865,6 +1866,8 @@ enum wmi_tlv_tag {
 	WMI_TAG_PDEV_SRG_OBSS_BSSID_ENABLE_BITMAP_CMD,
 	WMI_TAG_PDEV_NON_SRG_OBSS_COLOR_ENABLE_BITMAP_CMD,
 	WMI_TAG_PDEV_NON_SRG_OBSS_BSSID_ENABLE_BITMAP_CMD,
+	WMI_TAG_REGULATORY_RULE_EXT_STRUCT = 0x3A9,
+	WMI_TAG_REG_CHAN_LIST_CC_EXT_EVENT,
 	WMI_TAG_PDEV_SET_BIOS_SAR_TABLE_CMD = 0x3D8,
 	WMI_TAG_PDEV_SET_BIOS_GEO_TABLE_CMD,
 	WMI_TAG_MAX
@@ -2097,6 +2100,7 @@ enum wmi_tlv_service {
 
 	/* The second 128 bits */
 	WMI_MAX_EXT_SERVICE = 256,
+	WMI_TLV_SERVICE_REG_CC_EXT_EVENT_SUPPORT = 281,
 	WMI_TLV_SERVICE_BIOS_SAR_SUPPORT = 326,
 
 	/* The third 128 bits */
@@ -2313,6 +2317,8 @@ struct wmi_init_cmd {
 #define WMI_RSRC_CFG_FLAG1_BSS_CHANNEL_INFO_64 BIT(5)
 #define WMI_RSRC_CFG_FLAG1_ACK_RSSI BIT(18)
 
+#define WMI_CFG_HOST_SERVICE_FLAG_REG_CC_EXT 4
+
 struct wmi_resource_config {
 	u32 tlv_header;
 	u32 num_vdevs;
@@ -2372,6 +2378,15 @@ struct wmi_resource_config {
 	u32 sched_params;
 	u32 twt_ap_pdev_count;
 	u32 twt_ap_sta_count;
+	u32 max_nlo_ssids;
+	u32 num_pkt_filters;
+	u32 num_max_sta_vdevs;
+	u32 max_bssid_indicator;
+	u32 ul_resp_config;
+	u32 msdu_flow_override_config0;
+	u32 msdu_flow_override_config1;
+	u32 flags2;
+	u32 host_service_flags;
 } __packed;
 
 struct wmi_service_ready_event {
@@ -2854,6 +2869,8 @@ struct rx_reorder_queue_remove_params {
 #define REG_RULE_MAX_BW				0x0000ffff
 #define REG_RULE_REG_PWR			0x00ff0000
 #define REG_RULE_ANT_GAIN			0xff000000
+#define REG_RULE_PSD_INFO			BIT(0)
+#define REG_RULE_PSD_EIRP			0xff0000
 
 #define WMI_VDEV_PARAM_TXBF_SU_TX_BFEE BIT(0)
 #define WMI_VDEV_PARAM_TXBF_MU_TX_BFEE BIT(1)
@@ -4049,6 +4066,7 @@ struct wmi_he_rate_set {
 
 #define MAX_REG_RULES 10
 #define REG_ALPHA2_LEN 2
+#define MAX_6GHZ_REG_RULES 5
 
 enum wmi_start_event_param {
 	WMI_VDEV_START_RESP_EVENT = 0,
@@ -4079,16 +4097,6 @@ enum wmi_vdev_start_resp_status_code {
 	WMI_VDEV_START_RESPONSE_INVALID_REGDOMAIN = 4,
 };
 
-;
-enum cc_setting_code {
-	REG_SET_CC_STATUS_PASS = 0,
-	REG_CURRENT_ALPHA2_NOT_FOUND = 1,
-	REG_INIT_ALPHA2_NOT_FOUND = 2,
-	REG_SET_CC_CHANGE_NOT_ALLOWED = 3,
-	REG_SET_CC_STATUS_NO_MEMORY = 4,
-	REG_SET_CC_STATUS_FAIL = 5,
-};
-
 /* Regaulatory Rule Flags Passed by FW */
 #define REGULATORY_CHAN_DISABLED     BIT(0)
 #define REGULATORY_CHAN_NO_IR        BIT(1)
@@ -4102,13 +4110,72 @@ enum cc_setting_code {
 #define REGULATORY_CHAN_NO_20MHZ     BIT(11)
 #define REGULATORY_CHAN_NO_10MHZ     BIT(12)
 
-enum {
+enum wmi_reg_chan_list_cmd_type {
+	WMI_REG_CHAN_LIST_CC_ID = 0,
+	WMI_REG_CHAN_LIST_CC_EXT_ID = 1,
+};
+
+enum wmi_reg_cc_setting_code {
 	WMI_REG_SET_CC_STATUS_PASS = 0,
 	WMI_REG_CURRENT_ALPHA2_NOT_FOUND = 1,
 	WMI_REG_INIT_ALPHA2_NOT_FOUND = 2,
 	WMI_REG_SET_CC_CHANGE_NOT_ALLOWED = 3,
 	WMI_REG_SET_CC_STATUS_NO_MEMORY = 4,
 	WMI_REG_SET_CC_STATUS_FAIL = 5,
+
+	/* add new setting code above, update in
+	 * @enum cc_setting_code as well.
+	 * Also handle it in ath11k_wmi_cc_setting_code_to_reg()
+	 */
+};
+
+enum cc_setting_code {
+	REG_SET_CC_STATUS_PASS = 0,
+	REG_CURRENT_ALPHA2_NOT_FOUND = 1,
+	REG_INIT_ALPHA2_NOT_FOUND = 2,
+	REG_SET_CC_CHANGE_NOT_ALLOWED = 3,
+	REG_SET_CC_STATUS_NO_MEMORY = 4,
+	REG_SET_CC_STATUS_FAIL = 5,
+
+	/* add new setting code above, update in
+	 * @enum wmi_reg_cc_setting_code as well.
+	 */
+};
+
+static inline enum cc_setting_code
+ath11k_wmi_cc_setting_code_to_reg(enum wmi_reg_cc_setting_code status_code)
+{
+	switch (status_code) {
+	case WMI_REG_SET_CC_STATUS_PASS:
+		return REG_SET_CC_STATUS_PASS;
+	case WMI_REG_CURRENT_ALPHA2_NOT_FOUND:
+		return REG_CURRENT_ALPHA2_NOT_FOUND;
+	case WMI_REG_INIT_ALPHA2_NOT_FOUND:
+		return REG_INIT_ALPHA2_NOT_FOUND;
+	case WMI_REG_SET_CC_CHANGE_NOT_ALLOWED:
+		return REG_SET_CC_CHANGE_NOT_ALLOWED;
+	case WMI_REG_SET_CC_STATUS_NO_MEMORY:
+		return REG_SET_CC_STATUS_NO_MEMORY;
+	case WMI_REG_SET_CC_STATUS_FAIL:
+		return REG_SET_CC_STATUS_FAIL;
+	}
+
+	return REG_SET_CC_STATUS_FAIL;
+}
+
+enum wmi_reg_6ghz_ap_type {
+	WMI_REG_INDOOR_AP = 0,
+	WMI_REG_STANDARD_POWER_AP = 1,
+	WMI_REG_VERY_LOW_POWER_AP = 2,
+
+	WMI_REG_CURRENT_MAX_AP_TYPE,
+	WMI_REG_MAX_AP_TYPE = 7,
+};
+
+enum wmi_reg_6ghz_client_type {
+	WMI_REG_DEFAULT_CLIENT = 0,
+	WMI_REG_SUBORDINATE_CLIENT = 1,
+	WMI_REG_MAX_CLIENT_TYPE = 2,
 };
 
 struct cur_reg_rule {
@@ -4118,6 +4185,8 @@ struct cur_reg_rule {
 	u8 reg_power;
 	u8 ant_gain;
 	u16 flags;
+	bool psd_flag;
+	s8 psd_eirp;
 };
 
 struct cur_regulatory_info {
@@ -4137,6 +4206,22 @@ struct cur_regulatory_info {
 	u32 num_5ghz_reg_rules;
 	struct cur_reg_rule *reg_rules_2ghz_ptr;
 	struct cur_reg_rule *reg_rules_5ghz_ptr;
+	bool is_ext_reg_event;
+	enum wmi_reg_6ghz_client_type client_type;
+	bool rnr_tpe_usable;
+	bool unspecified_ap_usable;
+	u8 domain_code_6ghz_ap[WMI_REG_CURRENT_MAX_AP_TYPE];
+	u8 domain_code_6ghz_client[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE];
+	u32 domain_code_6ghz_super_id;
+	u32 min_bw_6ghz_ap[WMI_REG_CURRENT_MAX_AP_TYPE];
+	u32 max_bw_6ghz_ap[WMI_REG_CURRENT_MAX_AP_TYPE];
+	u32 min_bw_6ghz_client[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE];
+	u32 max_bw_6ghz_client[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE];
+	u32 num_6ghz_rules_ap[WMI_REG_CURRENT_MAX_AP_TYPE];
+	u32 num_6ghz_rules_client[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE];
+	struct cur_reg_rule *reg_rules_6ghz_ap_ptr[WMI_REG_CURRENT_MAX_AP_TYPE];
+	struct cur_reg_rule *reg_rules_6ghz_client_ptr
+		[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE];
 };
 
 struct wmi_reg_chan_list_cc_event {
@@ -4163,6 +4248,61 @@ struct wmi_regulatory_rule_struct {
 	u32  flag_info;
 };
 
+#define WMI_REG_CLIENT_MAX 4
+
+struct wmi_reg_chan_list_cc_ext_event {
+	u32 status_code;
+	u32 phy_id;
+	u32 alpha2;
+	u32 num_phy;
+	u32 country_id;
+	u32 domain_code;
+	u32 dfs_region;
+	u32 phybitmap;
+	u32 min_bw_2ghz;
+	u32 max_bw_2ghz;
+	u32 min_bw_5ghz;
+	u32 max_bw_5ghz;
+	u32 num_2ghz_reg_rules;
+	u32 num_5ghz_reg_rules;
+	u32 client_type;
+	u32 rnr_tpe_usable;
+	u32 unspecified_ap_usable;
+	u32 domain_code_6ghz_ap_lpi;
+	u32 domain_code_6ghz_ap_sp;
+	u32 domain_code_6ghz_ap_vlp;
+	u32 domain_code_6ghz_client_lpi[WMI_REG_CLIENT_MAX];
+	u32 domain_code_6ghz_client_sp[WMI_REG_CLIENT_MAX];
+	u32 domain_code_6ghz_client_vlp[WMI_REG_CLIENT_MAX];
+	u32 domain_code_6ghz_super_id;
+	u32 min_bw_6ghz_ap_sp;
+	u32 max_bw_6ghz_ap_sp;
+	u32 min_bw_6ghz_ap_lpi;
+	u32 max_bw_6ghz_ap_lpi;
+	u32 min_bw_6ghz_ap_vlp;
+	u32 max_bw_6ghz_ap_vlp;
+	u32 min_bw_6ghz_client_sp[WMI_REG_CLIENT_MAX];
+	u32 max_bw_6ghz_client_sp[WMI_REG_CLIENT_MAX];
+	u32 min_bw_6ghz_client_lpi[WMI_REG_CLIENT_MAX];
+	u32 max_bw_6ghz_client_lpi[WMI_REG_CLIENT_MAX];
+	u32 min_bw_6ghz_client_vlp[WMI_REG_CLIENT_MAX];
+	u32 max_bw_6ghz_client_vlp[WMI_REG_CLIENT_MAX];
+	u32 num_6ghz_reg_rules_ap_sp;
+	u32 num_6ghz_reg_rules_ap_lpi;
+	u32 num_6ghz_reg_rules_ap_vlp;
+	u32 num_6ghz_reg_rules_client_sp[WMI_REG_CLIENT_MAX];
+	u32 num_6ghz_reg_rules_client_lpi[WMI_REG_CLIENT_MAX];
+	u32 num_6ghz_reg_rules_client_vlp[WMI_REG_CLIENT_MAX];
+} __packed;
+
+struct wmi_regulatory_ext_rule {
+	u32 tlv_header;
+	u32 freq_info;
+	u32 bw_pwr_info;
+	u32 flag_info;
+	u32 psd_power_info;
+} __packed;
+
 struct wmi_vdev_delete_resp_event {
 	u32 vdev_id;
 } __packed;
@@ -5358,6 +5498,7 @@ struct target_resource_config {
 	u32 sched_params;
 	u32 twt_ap_pdev_count;
 	u32 twt_ap_sta_count;
+	u8 is_reg_cc_ext_event_supported;
 };
 
 enum wmi_debug_log_param {
-- 
cgit v1.2.3-70-g09d2


From e238e62ba8868a784e485eb94451c87cd1b85cee Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Wed, 1 Mar 2023 16:20:59 +0200
Subject: wifi: ath11k: add debug prints in regulatory WMI event processing

Add some more debug prints in processing regulatory WMI event in order to
increase more debuggability.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.5.0.1-01100-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230110121024.14051-4-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath11k/reg.c |   2 +-
 drivers/net/wireless/ath/ath11k/wmi.c | 207 ++++++++++++++++++++++++----------
 drivers/net/wireless/ath/ath11k/wmi.h | 142 +++++++++++++++++++++++
 3 files changed, 291 insertions(+), 60 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
index 08621bd13a12..67443457f4da 100644
--- a/drivers/net/wireless/ath/ath11k/reg.c
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -646,7 +646,7 @@ ath11k_reg_build_regd(struct ath11k_base *ab,
 	tmp_regd->dfs_region = ath11k_map_fw_dfs_region(reg_info->dfs_region);
 
 	ath11k_dbg(ab, ATH11K_DBG_REG,
-		   "\r\nCountry %s, CFG Regdomain %s FW Regdomain %d, num_reg_rules %d\n",
+		   "Country %s, CFG Regdomain %s FW Regdomain %d, num_reg_rules %d\n",
 		   alpha2, ath11k_reg_get_regdom_str(tmp_regd->dfs_region),
 		   reg_info->dfs_region, num_rules);
 	/* Update reg_rules[] below. Firmware is expected to
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 34731d0d1efc..27f3fceb33c5 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -4925,6 +4925,26 @@ static int ath11k_pull_vdev_start_resp_tlv(struct ath11k_base *ab, struct sk_buf
 	return 0;
 }
 
+static void ath11k_print_reg_rule(struct ath11k_base *ab, const char *band,
+				  u32 num_reg_rules,
+				  struct cur_reg_rule *reg_rule_ptr)
+{
+	struct cur_reg_rule *reg_rule = reg_rule_ptr;
+	u32 count;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI, "number of reg rules in %s band: %d\n",
+		   band, num_reg_rules);
+
+	for (count = 0; count < num_reg_rules; count++) {
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "reg rule %d: (%d - %d @ %d) (%d, %d) (FLAGS %d)\n",
+			   count + 1, reg_rule->start_freq, reg_rule->end_freq,
+			   reg_rule->max_bw, reg_rule->ant_gain,
+			   reg_rule->reg_power, reg_rule->flags);
+		reg_rule++;
+	}
+}
+
 static struct cur_reg_rule
 *create_reg_rules_from_wmi(u32 num_reg_rules,
 			   struct wmi_regulatory_rule_struct *wmi_reg_rule)
@@ -5006,6 +5026,10 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 	reg_info->ctry_code = chan_list_event_hdr->country_id;
 	reg_info->reg_dmn_pair = chan_list_event_hdr->domain_code;
 
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "status_code %s",
+		   ath11k_cc_status_to_str(reg_info->status_code));
+
 	reg_info->status_code =
 		ath11k_wmi_cc_setting_code_to_reg(chan_list_event_hdr->status_code);
 
@@ -5020,13 +5044,13 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 	num_5ghz_reg_rules = reg_info->num_5ghz_reg_rules;
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI,
-		   "%s:cc %s dsf %d BW: min_2ghz %d max_2ghz %d min_5ghz %d max_5ghz %d",
-		   __func__, reg_info->alpha2, reg_info->dfs_region,
+		   "cc %s dsf %d BW: min_2ghz %d max_2ghz %d min_5ghz %d max_5ghz %d",
+		   reg_info->alpha2, reg_info->dfs_region,
 		   reg_info->min_bw_2ghz, reg_info->max_bw_2ghz,
 		   reg_info->min_bw_5ghz, reg_info->max_bw_5ghz);
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI,
-		   "%s: num_2ghz_reg_rules %d num_5ghz_reg_rules %d", __func__,
+		   "num_2ghz_reg_rules %d num_5ghz_reg_rules %d",
 		   num_2ghz_reg_rules, num_5ghz_reg_rules);
 
 	wmi_reg_rule =
@@ -5043,6 +5067,10 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 			ath11k_warn(ab, "Unable to Allocate memory for 2 GHz rules\n");
 			return -ENOMEM;
 		}
+
+		ath11k_print_reg_rule(ab, "2 GHz",
+				      num_2ghz_reg_rules,
+				      reg_info->reg_rules_2ghz_ptr);
 	}
 
 	if (num_5ghz_reg_rules) {
@@ -5055,6 +5083,10 @@ static int ath11k_pull_reg_chan_list_update_ev(struct ath11k_base *ab,
 			ath11k_warn(ab, "Unable to Allocate memory for 5 GHz rules\n");
 			return -ENOMEM;
 		}
+
+		ath11k_print_reg_rule(ab, "5 GHz",
+				      num_5ghz_reg_rules,
+				      reg_info->reg_rules_5ghz_ptr);
 	}
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI, "processed regulatory channel list\n");
@@ -5128,7 +5160,7 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 						   struct cur_regulatory_info *reg_info)
 {
 	const void **tb;
-	const struct wmi_reg_chan_list_cc_ext_event *ext_chan_list_event_hdr;
+	const struct wmi_reg_chan_list_cc_ext_event *ev;
 	struct wmi_regulatory_ext_rule *ext_wmi_reg_rule;
 	u32 num_2ghz_reg_rules, num_5ghz_reg_rules;
 	u32 num_6ghz_reg_rules_ap[WMI_REG_CURRENT_MAX_AP_TYPE];
@@ -5145,31 +5177,29 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 		return ret;
 	}
 
-	ext_chan_list_event_hdr = tb[WMI_TAG_REG_CHAN_LIST_CC_EXT_EVENT];
-	if (!ext_chan_list_event_hdr) {
+	ev = tb[WMI_TAG_REG_CHAN_LIST_CC_EXT_EVENT];
+	if (!ev) {
 		ath11k_warn(ab, "failed to fetch reg chan list ext update ev\n");
 		kfree(tb);
 		return -EPROTO;
 	}
 
-	reg_info->num_2ghz_reg_rules =
-			ext_chan_list_event_hdr->num_2ghz_reg_rules;
-	reg_info->num_5ghz_reg_rules =
-			ext_chan_list_event_hdr->num_5ghz_reg_rules;
+	reg_info->num_2ghz_reg_rules = ev->num_2ghz_reg_rules;
+	reg_info->num_5ghz_reg_rules = ev->num_5ghz_reg_rules;
 	reg_info->num_6ghz_rules_ap[WMI_REG_INDOOR_AP] =
-			ext_chan_list_event_hdr->num_6ghz_reg_rules_ap_lpi;
+			ev->num_6ghz_reg_rules_ap_lpi;
 	reg_info->num_6ghz_rules_ap[WMI_REG_STANDARD_POWER_AP] =
-			ext_chan_list_event_hdr->num_6ghz_reg_rules_ap_sp;
+			ev->num_6ghz_reg_rules_ap_sp;
 	reg_info->num_6ghz_rules_ap[WMI_REG_VERY_LOW_POWER_AP] =
-			ext_chan_list_event_hdr->num_6ghz_reg_rules_ap_vlp;
+			ev->num_6ghz_reg_rules_ap_vlp;
 
 	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
 		reg_info->num_6ghz_rules_client[WMI_REG_INDOOR_AP][i] =
-			ext_chan_list_event_hdr->num_6ghz_reg_rules_client_lpi[i];
+			ev->num_6ghz_reg_rules_client_lpi[i];
 		reg_info->num_6ghz_rules_client[WMI_REG_STANDARD_POWER_AP][i] =
-			ext_chan_list_event_hdr->num_6ghz_reg_rules_client_sp[i];
+			ev->num_6ghz_reg_rules_client_sp[i];
 		reg_info->num_6ghz_rules_client[WMI_REG_VERY_LOW_POWER_AP][i] =
-			ext_chan_list_event_hdr->num_6ghz_reg_rules_client_vlp[i];
+			ev->num_6ghz_reg_rules_client_vlp[i];
 	}
 
 	num_2ghz_reg_rules = reg_info->num_2ghz_reg_rules;
@@ -5231,57 +5261,79 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 		return -EINVAL;
 	}
 
-	memcpy(reg_info->alpha2, &ext_chan_list_event_hdr->alpha2,
-	       REG_ALPHA2_LEN);
+	memcpy(reg_info->alpha2, &ev->alpha2, REG_ALPHA2_LEN);
 
-	reg_info->dfs_region = ext_chan_list_event_hdr->dfs_region;
-	reg_info->phybitmap = ext_chan_list_event_hdr->phybitmap;
-	reg_info->num_phy = ext_chan_list_event_hdr->num_phy;
-	reg_info->phy_id = ext_chan_list_event_hdr->phy_id;
-	reg_info->ctry_code = ext_chan_list_event_hdr->country_id;
-	reg_info->reg_dmn_pair = ext_chan_list_event_hdr->domain_code;
+	reg_info->dfs_region = ev->dfs_region;
+	reg_info->phybitmap = ev->phybitmap;
+	reg_info->num_phy = ev->num_phy;
+	reg_info->phy_id = ev->phy_id;
+	reg_info->ctry_code = ev->country_id;
+	reg_info->reg_dmn_pair = ev->domain_code;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "status_code %s",
+		   ath11k_cc_status_to_str(reg_info->status_code));
 
 	reg_info->status_code =
-		ath11k_wmi_cc_setting_code_to_reg(ext_chan_list_event_hdr->status_code);
+		ath11k_wmi_cc_setting_code_to_reg(ev->status_code);
 
 	reg_info->is_ext_reg_event = true;
 
-	reg_info->min_bw_2ghz = ext_chan_list_event_hdr->min_bw_2ghz;
-	reg_info->max_bw_2ghz = ext_chan_list_event_hdr->max_bw_2ghz;
-	reg_info->min_bw_5ghz = ext_chan_list_event_hdr->min_bw_5ghz;
-	reg_info->max_bw_5ghz = ext_chan_list_event_hdr->max_bw_5ghz;
+	reg_info->min_bw_2ghz = ev->min_bw_2ghz;
+	reg_info->max_bw_2ghz = ev->max_bw_2ghz;
+	reg_info->min_bw_5ghz = ev->min_bw_5ghz;
+	reg_info->max_bw_5ghz = ev->max_bw_5ghz;
 
 	reg_info->min_bw_6ghz_ap[WMI_REG_INDOOR_AP] =
-			ext_chan_list_event_hdr->min_bw_6ghz_ap_lpi;
+			ev->min_bw_6ghz_ap_lpi;
 	reg_info->max_bw_6ghz_ap[WMI_REG_INDOOR_AP] =
-			 ext_chan_list_event_hdr->max_bw_6ghz_ap_lpi;
+			ev->max_bw_6ghz_ap_lpi;
 	reg_info->min_bw_6ghz_ap[WMI_REG_STANDARD_POWER_AP] =
-			ext_chan_list_event_hdr->min_bw_6ghz_ap_sp;
+			ev->min_bw_6ghz_ap_sp;
 	reg_info->max_bw_6ghz_ap[WMI_REG_STANDARD_POWER_AP] =
-			ext_chan_list_event_hdr->max_bw_6ghz_ap_sp;
+			ev->max_bw_6ghz_ap_sp;
 	reg_info->min_bw_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP] =
-			ext_chan_list_event_hdr->min_bw_6ghz_ap_vlp;
+			ev->min_bw_6ghz_ap_vlp;
 	reg_info->max_bw_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP] =
-			ext_chan_list_event_hdr->max_bw_6ghz_ap_vlp;
+			ev->max_bw_6ghz_ap_vlp;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "6 GHz AP BW: LPI (%d - %d), SP (%d - %d), VLP (%d - %d)\n",
+		   reg_info->min_bw_6ghz_ap[WMI_REG_INDOOR_AP],
+		   reg_info->max_bw_6ghz_ap[WMI_REG_INDOOR_AP],
+		   reg_info->min_bw_6ghz_ap[WMI_REG_STANDARD_POWER_AP],
+		   reg_info->max_bw_6ghz_ap[WMI_REG_STANDARD_POWER_AP],
+		   reg_info->min_bw_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP],
+		   reg_info->max_bw_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP]);
 
 	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
 		reg_info->min_bw_6ghz_client[WMI_REG_INDOOR_AP][i] =
-				ext_chan_list_event_hdr->min_bw_6ghz_client_lpi[i];
+				ev->min_bw_6ghz_client_lpi[i];
 		reg_info->max_bw_6ghz_client[WMI_REG_INDOOR_AP][i] =
-				ext_chan_list_event_hdr->max_bw_6ghz_client_lpi[i];
+				ev->max_bw_6ghz_client_lpi[i];
 		reg_info->min_bw_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
-				ext_chan_list_event_hdr->min_bw_6ghz_client_sp[i];
+				ev->min_bw_6ghz_client_sp[i];
 		reg_info->max_bw_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
-				ext_chan_list_event_hdr->max_bw_6ghz_client_sp[i];
+				ev->max_bw_6ghz_client_sp[i];
 		reg_info->min_bw_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
-				ext_chan_list_event_hdr->min_bw_6ghz_client_vlp[i];
+				ev->min_bw_6ghz_client_vlp[i];
 		reg_info->max_bw_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
-				ext_chan_list_event_hdr->max_bw_6ghz_client_vlp[i];
+				ev->max_bw_6ghz_client_vlp[i];
+
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "6 GHz %s BW: LPI (%d - %d), SP (%d - %d), VLP (%d - %d)\n",
+			   ath11k_6ghz_client_type_to_str(i),
+			   reg_info->min_bw_6ghz_client[WMI_REG_INDOOR_AP][i],
+			   reg_info->max_bw_6ghz_client[WMI_REG_INDOOR_AP][i],
+			   reg_info->min_bw_6ghz_client[WMI_REG_STANDARD_POWER_AP][i],
+			   reg_info->max_bw_6ghz_client[WMI_REG_STANDARD_POWER_AP][i],
+			   reg_info->min_bw_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i],
+			   reg_info->max_bw_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i]);
 	}
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI,
-		   "%s:cc_ext %s dsf %d BW: min_2ghz %d max_2ghz %d min_5ghz %d max_5ghz %d",
-		   __func__, reg_info->alpha2, reg_info->dfs_region,
+		   "cc_ext %s dsf %d BW: min_2ghz %d max_2ghz %d min_5ghz %d max_5ghz %d",
+		   reg_info->alpha2, reg_info->dfs_region,
 		   reg_info->min_bw_2ghz, reg_info->max_bw_2ghz,
 		   reg_info->min_bw_5ghz, reg_info->max_bw_5ghz);
 
@@ -5310,9 +5362,8 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 		   num_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][j]);
 
 	ext_wmi_reg_rule =
-		(struct wmi_regulatory_ext_rule *)((u8 *)ext_chan_list_event_hdr
-						   + sizeof(*ext_chan_list_event_hdr)
-						   + sizeof(struct wmi_tlv));
+		(struct wmi_regulatory_ext_rule *)((u8 *)ev + sizeof(*ev) +
+						   sizeof(struct wmi_tlv));
 	if (num_2ghz_reg_rules) {
 		reg_info->reg_rules_2ghz_ptr =
 			create_ext_reg_rules_from_wmi(num_2ghz_reg_rules,
@@ -5323,6 +5374,10 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 			ath11k_warn(ab, "Unable to Allocate memory for 2 GHz rules\n");
 			return -ENOMEM;
 		}
+
+		ath11k_print_reg_rule(ab, "2 GHz",
+				      num_2ghz_reg_rules,
+				      reg_info->reg_rules_2ghz_ptr);
 	}
 
 	ext_wmi_reg_rule += num_2ghz_reg_rules;
@@ -5358,6 +5413,10 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 			ath11k_warn(ab, "Unable to Allocate memory for 5 GHz rules\n");
 			return -ENOMEM;
 		}
+
+		ath11k_print_reg_rule(ab, "5 GHz",
+				      num_5ghz_reg_rules,
+				      reg_info->reg_rules_5ghz_ptr);
 	}
 
 	/* We have adjusted the number of 5 GHz reg rules above. But still those
@@ -5378,10 +5437,17 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 			return -ENOMEM;
 		}
 
+		ath11k_print_reg_rule(ab, ath11k_6ghz_ap_type_to_str(i),
+				      num_6ghz_reg_rules_ap[i],
+				      reg_info->reg_rules_6ghz_ap_ptr[i]);
+
 		ext_wmi_reg_rule += num_6ghz_reg_rules_ap[i];
 	}
 
 	for (j = 0; j < WMI_REG_CURRENT_MAX_AP_TYPE; j++) {
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "6 GHz AP type %s", ath11k_6ghz_ap_type_to_str(j));
+
 		for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
 			reg_info->reg_rules_6ghz_client_ptr[j][i] =
 				create_ext_reg_rules_from_wmi(num_6ghz_client[j][i],
@@ -5393,35 +5459,58 @@ static int ath11k_pull_reg_chan_list_ext_update_ev(struct ath11k_base *ab,
 				return -ENOMEM;
 			}
 
+			ath11k_print_reg_rule(ab,
+					      ath11k_6ghz_client_type_to_str(i),
+					      num_6ghz_client[j][i],
+					      reg_info->reg_rules_6ghz_client_ptr[j][i]);
+
 			ext_wmi_reg_rule += num_6ghz_client[j][i];
 		}
 	}
 
-	reg_info->client_type = ext_chan_list_event_hdr->client_type;
-	reg_info->rnr_tpe_usable = ext_chan_list_event_hdr->rnr_tpe_usable;
+	reg_info->client_type = ev->client_type;
+	reg_info->rnr_tpe_usable = ev->rnr_tpe_usable;
 	reg_info->unspecified_ap_usable =
-			ext_chan_list_event_hdr->unspecified_ap_usable;
+			ev->unspecified_ap_usable;
 	reg_info->domain_code_6ghz_ap[WMI_REG_INDOOR_AP] =
-			ext_chan_list_event_hdr->domain_code_6ghz_ap_lpi;
+			ev->domain_code_6ghz_ap_lpi;
 	reg_info->domain_code_6ghz_ap[WMI_REG_STANDARD_POWER_AP] =
-			ext_chan_list_event_hdr->domain_code_6ghz_ap_sp;
+			ev->domain_code_6ghz_ap_sp;
 	reg_info->domain_code_6ghz_ap[WMI_REG_VERY_LOW_POWER_AP] =
-			ext_chan_list_event_hdr->domain_code_6ghz_ap_vlp;
+			ev->domain_code_6ghz_ap_vlp;
+
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "6 GHz reg info client type %s rnr_tpe_usable %d unspecified_ap_usable %d AP sub domain: lpi %s, sp %s, vlp %s\n",
+		   ath11k_6ghz_client_type_to_str(reg_info->client_type),
+		   reg_info->rnr_tpe_usable,
+		   reg_info->unspecified_ap_usable,
+		   ath11k_sub_reg_6ghz_to_str(ev->domain_code_6ghz_ap_lpi),
+		   ath11k_sub_reg_6ghz_to_str(ev->domain_code_6ghz_ap_sp),
+		   ath11k_sub_reg_6ghz_to_str(ev->domain_code_6ghz_ap_vlp));
 
 	for (i = 0; i < WMI_REG_MAX_CLIENT_TYPE; i++) {
 		reg_info->domain_code_6ghz_client[WMI_REG_INDOOR_AP][i] =
-				ext_chan_list_event_hdr->domain_code_6ghz_client_lpi[i];
+				ev->domain_code_6ghz_client_lpi[i];
 		reg_info->domain_code_6ghz_client[WMI_REG_STANDARD_POWER_AP][i] =
-				ext_chan_list_event_hdr->domain_code_6ghz_client_sp[i];
+				ev->domain_code_6ghz_client_sp[i];
 		reg_info->domain_code_6ghz_client[WMI_REG_VERY_LOW_POWER_AP][i] =
-				ext_chan_list_event_hdr->domain_code_6ghz_client_vlp[i];
+				ev->domain_code_6ghz_client_vlp[i];
+
+		ath11k_dbg(ab, ATH11K_DBG_WMI,
+			   "6 GHz client type %s client sub domain: lpi %s, sp %s, vlp %s\n",
+			   ath11k_6ghz_client_type_to_str(i),
+			   ath11k_sub_reg_6ghz_to_str(ev->domain_code_6ghz_client_lpi[i]),
+			   ath11k_sub_reg_6ghz_to_str(ev->domain_code_6ghz_client_sp[i]),
+			   ath11k_sub_reg_6ghz_to_str(ev->domain_code_6ghz_client_vlp[i])
+			  );
 	}
 
-	reg_info->domain_code_6ghz_super_id =
-			ext_chan_list_event_hdr->domain_code_6ghz_super_id;
+	reg_info->domain_code_6ghz_super_id = ev->domain_code_6ghz_super_id;
 
-	ath11k_dbg(ab, ATH11K_DBG_WMI, "6 GHz client_type: %d domain_code_6ghz_super_id: %d",
-		   reg_info->client_type, reg_info->domain_code_6ghz_super_id);
+	ath11k_dbg(ab, ATH11K_DBG_WMI,
+		   "6 GHz client_type %s 6 GHz super domain %s",
+		   ath11k_6ghz_client_type_to_str(reg_info->client_type),
+		   ath11k_super_reg_6ghz_to_str(reg_info->domain_code_6ghz_super_id));
 
 	ath11k_dbg(ab, ATH11K_DBG_WMI, "processed regulatory ext channel list\n");
 
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 3ec3fa0aa5c8..b23b7a22bc9a 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -4139,6 +4139,7 @@ enum cc_setting_code {
 
 	/* add new setting code above, update in
 	 * @enum wmi_reg_cc_setting_code as well.
+	 * Also handle it in ath11k_cc_status_to_str()
 	 */
 };
 
@@ -4163,21 +4164,162 @@ ath11k_wmi_cc_setting_code_to_reg(enum wmi_reg_cc_setting_code status_code)
 	return REG_SET_CC_STATUS_FAIL;
 }
 
+static inline const char *ath11k_cc_status_to_str(enum cc_setting_code code)
+{
+	switch (code) {
+	case REG_SET_CC_STATUS_PASS:
+		return "REG_SET_CC_STATUS_PASS";
+	case REG_CURRENT_ALPHA2_NOT_FOUND:
+		return "REG_CURRENT_ALPHA2_NOT_FOUND";
+	case REG_INIT_ALPHA2_NOT_FOUND:
+		return "REG_INIT_ALPHA2_NOT_FOUND";
+	case REG_SET_CC_CHANGE_NOT_ALLOWED:
+		return "REG_SET_CC_CHANGE_NOT_ALLOWED";
+	case REG_SET_CC_STATUS_NO_MEMORY:
+		return "REG_SET_CC_STATUS_NO_MEMORY";
+	case REG_SET_CC_STATUS_FAIL:
+		return "REG_SET_CC_STATUS_FAIL";
+	}
+
+	return "Unknown CC status";
+}
+
 enum wmi_reg_6ghz_ap_type {
 	WMI_REG_INDOOR_AP = 0,
 	WMI_REG_STANDARD_POWER_AP = 1,
 	WMI_REG_VERY_LOW_POWER_AP = 2,
 
+	/* add AP type above, handle in ath11k_6ghz_ap_type_to_str()
+	 */
 	WMI_REG_CURRENT_MAX_AP_TYPE,
 	WMI_REG_MAX_AP_TYPE = 7,
 };
 
+static inline const char *
+ath11k_6ghz_ap_type_to_str(enum wmi_reg_6ghz_ap_type type)
+{
+	switch (type) {
+	case WMI_REG_INDOOR_AP:
+		return "INDOOR AP";
+	case WMI_REG_STANDARD_POWER_AP:
+		return "STANDARD POWER AP";
+	case WMI_REG_VERY_LOW_POWER_AP:
+		return "VERY LOW POWER AP";
+	case WMI_REG_CURRENT_MAX_AP_TYPE:
+		return "CURRENT_MAX_AP_TYPE";
+	case WMI_REG_MAX_AP_TYPE:
+		return "MAX_AP_TYPE";
+	}
+
+	return "unknown 6 GHz AP type";
+}
+
 enum wmi_reg_6ghz_client_type {
 	WMI_REG_DEFAULT_CLIENT = 0,
 	WMI_REG_SUBORDINATE_CLIENT = 1,
 	WMI_REG_MAX_CLIENT_TYPE = 2,
+
+	/* add client type above, handle it in
+	 * ath11k_6ghz_client_type_to_str()
+	 */
 };
 
+static inline const char *
+ath11k_6ghz_client_type_to_str(enum wmi_reg_6ghz_client_type type)
+{
+	switch (type) {
+	case WMI_REG_DEFAULT_CLIENT:
+		return "DEFAULT CLIENT";
+	case WMI_REG_SUBORDINATE_CLIENT:
+		return "SUBORDINATE CLIENT";
+	case WMI_REG_MAX_CLIENT_TYPE:
+		return "MAX_CLIENT_TYPE";
+	}
+
+	return "unknown 6 GHz client type";
+}
+
+enum reg_subdomains_6ghz {
+	EMPTY_6GHZ = 0x0,
+	FCC1_CLIENT_LPI_REGULAR_6GHZ = 0x01,
+	FCC1_CLIENT_SP_6GHZ = 0x02,
+	FCC1_AP_LPI_6GHZ = 0x03,
+	FCC1_CLIENT_LPI_SUBORDINATE = FCC1_AP_LPI_6GHZ,
+	FCC1_AP_SP_6GHZ = 0x04,
+	ETSI1_LPI_6GHZ = 0x10,
+	ETSI1_VLP_6GHZ = 0x11,
+	ETSI2_LPI_6GHZ = 0x12,
+	ETSI2_VLP_6GHZ = 0x13,
+	APL1_LPI_6GHZ = 0x20,
+	APL1_VLP_6GHZ = 0x21,
+
+	/* add sub-domain above, handle it in
+	 * ath11k_sub_reg_6ghz_to_str()
+	 */
+};
+
+static inline const char *
+ath11k_sub_reg_6ghz_to_str(enum reg_subdomains_6ghz sub_id)
+{
+	switch (sub_id) {
+	case EMPTY_6GHZ:
+		return "N/A";
+	case FCC1_CLIENT_LPI_REGULAR_6GHZ:
+		return "FCC1_CLIENT_LPI_REGULAR_6GHZ";
+	case FCC1_CLIENT_SP_6GHZ:
+		return "FCC1_CLIENT_SP_6GHZ";
+	case FCC1_AP_LPI_6GHZ:
+		return "FCC1_AP_LPI_6GHZ/FCC1_CLIENT_LPI_SUBORDINATE";
+	case FCC1_AP_SP_6GHZ:
+		return "FCC1_AP_SP_6GHZ";
+	case ETSI1_LPI_6GHZ:
+		return "ETSI1_LPI_6GHZ";
+	case ETSI1_VLP_6GHZ:
+		return "ETSI1_VLP_6GHZ";
+	case ETSI2_LPI_6GHZ:
+		return "ETSI2_LPI_6GHZ";
+	case ETSI2_VLP_6GHZ:
+		return "ETSI2_VLP_6GHZ";
+	case APL1_LPI_6GHZ:
+		return "APL1_LPI_6GHZ";
+	case APL1_VLP_6GHZ:
+		return "APL1_VLP_6GHZ";
+	}
+
+	return "unknown sub reg id";
+}
+
+enum reg_super_domain_6ghz {
+	FCC1_6GHZ = 0x01,
+	ETSI1_6GHZ = 0x02,
+	ETSI2_6GHZ = 0x03,
+	APL1_6GHZ = 0x04,
+	FCC1_6GHZ_CL = 0x05,
+
+	/* add super domain above, handle it in
+	 * ath11k_super_reg_6ghz_to_str()
+	 */
+};
+
+static inline const char *
+ath11k_super_reg_6ghz_to_str(enum reg_super_domain_6ghz domain_id)
+{
+	switch (domain_id) {
+	case FCC1_6GHZ:
+		return "FCC1_6GHZ";
+	case ETSI1_6GHZ:
+		return "ETSI1_6GHZ";
+	case ETSI2_6GHZ:
+		return "ETSI2_6GHZ";
+	case APL1_6GHZ:
+		return "APL1_6GHZ";
+	case FCC1_6GHZ_CL:
+		return "FCC1_6GHZ_CL";
+	}
+
+	return "unknown domain id";
+}
+
 struct cur_reg_rule {
 	u16 start_freq;
 	u16 end_freq;
-- 
cgit v1.2.3-70-g09d2


From f79cbc77abde22dd01ac90393a7bfb283d9d50a3 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@kernel.org>
Date: Mon, 27 Feb 2023 14:17:31 +0200
Subject: wifi: move mac80211_hwsim and virt_wifi to virtual directory

To clean up drivers/net/wireless move the virtual drivers to a new virtual
directory. I did consider adding CONFIG_WLAN_VENDOR_VIRTUAL like other vendors
have but then dropped the idea as we are not real drivers.

There should be no changes in compilation or in Kconfig options, merely moving
files. The order in menuconfig is slightly changed, the virtual drivers are now
last in the list.

Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230227121732.8967-2-kvalo@kernel.org
---
 drivers/net/wireless/Kconfig                  |   20 +-
 drivers/net/wireless/Makefile                 |    4 +-
 drivers/net/wireless/mac80211_hwsim.c         | 5901 -------------------------
 drivers/net/wireless/mac80211_hwsim.h         |  280 --
 drivers/net/wireless/virt_wifi.c              |  697 ---
 drivers/net/wireless/virtual/Kconfig          |   20 +
 drivers/net/wireless/virtual/Makefile         |    3 +
 drivers/net/wireless/virtual/mac80211_hwsim.c | 5901 +++++++++++++++++++++++++
 drivers/net/wireless/virtual/mac80211_hwsim.h |  280 ++
 drivers/net/wireless/virtual/virt_wifi.c      |  697 +++
 10 files changed, 6903 insertions(+), 6900 deletions(-)
 delete mode 100644 drivers/net/wireless/mac80211_hwsim.c
 delete mode 100644 drivers/net/wireless/mac80211_hwsim.h
 delete mode 100644 drivers/net/wireless/virt_wifi.c
 create mode 100644 drivers/net/wireless/virtual/Kconfig
 create mode 100644 drivers/net/wireless/virtual/Makefile
 create mode 100644 drivers/net/wireless/virtual/mac80211_hwsim.c
 create mode 100644 drivers/net/wireless/virtual/mac80211_hwsim.h
 create mode 100644 drivers/net/wireless/virtual/virt_wifi.c

diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig
index cb1c15012dd0..42b40cc96b21 100644
--- a/drivers/net/wireless/Kconfig
+++ b/drivers/net/wireless/Kconfig
@@ -64,19 +64,6 @@ config PCMCIA_WL3501
 	  It has basic support for Linux wireless extensions and initial
 	  micro support for ethtool.
 
-config MAC80211_HWSIM
-	tristate "Simulated radio testing tool for mac80211"
-	depends on MAC80211
-	help
-	  This driver is a developer testing tool that can be used to test
-	  IEEE 802.11 networking stack (mac80211) functionality. This is not
-	  needed for normal wireless LAN usage and is only for testing. See
-	  Documentation/networking/mac80211_hwsim for more information on how
-	  to use this tool.
-
-	  To compile this driver as a module, choose M here: the module will be
-	  called mac80211_hwsim.  If unsure, say N.
-
 config USB_NET_RNDIS_WLAN
 	tristate "Wireless RNDIS USB support"
 	depends on USB
@@ -106,11 +93,6 @@ config USB_NET_RNDIS_WLAN
 
 	  If you choose to build a module, it'll be called rndis_wlan.
 
-config VIRT_WIFI
-	tristate "Wifi wrapper for ethernet drivers"
-	depends on CFG80211
-	help
-	  This option adds support for ethernet connections to appear as if they
-	  are wifi connections through a special rtnetlink device.
+source "drivers/net/wireless/virtual/Kconfig"
 
 endif # WLAN
diff --git a/drivers/net/wireless/Makefile b/drivers/net/wireless/Makefile
index a61cf6c90343..1b697cfe0a13 100644
--- a/drivers/net/wireless/Makefile
+++ b/drivers/net/wireless/Makefile
@@ -29,6 +29,4 @@ obj-$(CONFIG_PCMCIA_WL3501)	+= wl3501_cs.o
 
 obj-$(CONFIG_USB_NET_RNDIS_WLAN)	+= rndis_wlan.o
 
-obj-$(CONFIG_MAC80211_HWSIM)	+= mac80211_hwsim.o
-
-obj-$(CONFIG_VIRT_WIFI)	+= virt_wifi.o
+obj-$(CONFIG_WLAN) += virtual/
diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
deleted file mode 100644
index f4bdc243ea0d..000000000000
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ /dev/null
@@ -1,5901 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * mac80211_hwsim - software simulator of 802.11 radio(s) for mac80211
- * Copyright (c) 2008, Jouni Malinen <j@w1.fi>
- * Copyright (c) 2011, Javier Lopez <jlopex@gmail.com>
- * Copyright (c) 2016 - 2017 Intel Deutschland GmbH
- * Copyright (C) 2018 - 2022 Intel Corporation
- */
-
-/*
- * TODO:
- * - Add TSF sync and fix IBSS beacon transmission by adding
- *   competition for "air time" at TBTT
- * - RX filtering based on filter configuration (data->rx_filter)
- */
-
-#include <linux/list.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <net/dst.h>
-#include <net/xfrm.h>
-#include <net/mac80211.h>
-#include <net/ieee80211_radiotap.h>
-#include <linux/if_arp.h>
-#include <linux/rtnetlink.h>
-#include <linux/etherdevice.h>
-#include <linux/platform_device.h>
-#include <linux/debugfs.h>
-#include <linux/module.h>
-#include <linux/ktime.h>
-#include <net/genetlink.h>
-#include <net/net_namespace.h>
-#include <net/netns/generic.h>
-#include <linux/rhashtable.h>
-#include <linux/nospec.h>
-#include <linux/virtio.h>
-#include <linux/virtio_ids.h>
-#include <linux/virtio_config.h>
-#include "mac80211_hwsim.h"
-
-#define WARN_QUEUE 100
-#define MAX_QUEUE 200
-
-MODULE_AUTHOR("Jouni Malinen");
-MODULE_DESCRIPTION("Software simulator of 802.11 radio(s) for mac80211");
-MODULE_LICENSE("GPL");
-
-static int radios = 2;
-module_param(radios, int, 0444);
-MODULE_PARM_DESC(radios, "Number of simulated radios");
-
-static int channels = 1;
-module_param(channels, int, 0444);
-MODULE_PARM_DESC(channels, "Number of concurrent channels");
-
-static bool paged_rx = false;
-module_param(paged_rx, bool, 0644);
-MODULE_PARM_DESC(paged_rx, "Use paged SKBs for RX instead of linear ones");
-
-static bool rctbl = false;
-module_param(rctbl, bool, 0444);
-MODULE_PARM_DESC(rctbl, "Handle rate control table");
-
-static bool support_p2p_device = true;
-module_param(support_p2p_device, bool, 0444);
-MODULE_PARM_DESC(support_p2p_device, "Support P2P-Device interface type");
-
-static bool mlo;
-module_param(mlo, bool, 0444);
-MODULE_PARM_DESC(mlo, "Support MLO");
-
-/**
- * enum hwsim_regtest - the type of regulatory tests we offer
- *
- * These are the different values you can use for the regtest
- * module parameter. This is useful to help test world roaming
- * and the driver regulatory_hint() call and combinations of these.
- * If you want to do specific alpha2 regulatory domain tests simply
- * use the userspace regulatory request as that will be respected as
- * well without the need of this module parameter. This is designed
- * only for testing the driver regulatory request, world roaming
- * and all possible combinations.
- *
- * @HWSIM_REGTEST_DISABLED: No regulatory tests are performed,
- * 	this is the default value.
- * @HWSIM_REGTEST_DRIVER_REG_FOLLOW: Used for testing the driver regulatory
- *	hint, only one driver regulatory hint will be sent as such the
- * 	secondary radios are expected to follow.
- * @HWSIM_REGTEST_DRIVER_REG_ALL: Used for testing the driver regulatory
- * 	request with all radios reporting the same regulatory domain.
- * @HWSIM_REGTEST_DIFF_COUNTRY: Used for testing the drivers calling
- * 	different regulatory domains requests. Expected behaviour is for
- * 	an intersection to occur but each device will still use their
- * 	respective regulatory requested domains. Subsequent radios will
- * 	use the resulting intersection.
- * @HWSIM_REGTEST_WORLD_ROAM: Used for testing the world roaming. We accomplish
- *	this by using a custom beacon-capable regulatory domain for the first
- *	radio. All other device world roam.
- * @HWSIM_REGTEST_CUSTOM_WORLD: Used for testing the custom world regulatory
- * 	domain requests. All radios will adhere to this custom world regulatory
- * 	domain.
- * @HWSIM_REGTEST_CUSTOM_WORLD_2: Used for testing 2 custom world regulatory
- * 	domain requests. The first radio will adhere to the first custom world
- * 	regulatory domain, the second one to the second custom world regulatory
- * 	domain. All other devices will world roam.
- * @HWSIM_REGTEST_STRICT_FOLLOW: Used for testing strict regulatory domain
- *	settings, only the first radio will send a regulatory domain request
- *	and use strict settings. The rest of the radios are expected to follow.
- * @HWSIM_REGTEST_STRICT_ALL: Used for testing strict regulatory domain
- *	settings. All radios will adhere to this.
- * @HWSIM_REGTEST_STRICT_AND_DRIVER_REG: Used for testing strict regulatory
- *	domain settings, combined with secondary driver regulatory domain
- *	settings. The first radio will get a strict regulatory domain setting
- *	using the first driver regulatory request and the second radio will use
- *	non-strict settings using the second driver regulatory request. All
- *	other devices should follow the intersection created between the
- *	first two.
- * @HWSIM_REGTEST_ALL: Used for testing every possible mix. You will need
- * 	at least 6 radios for a complete test. We will test in this order:
- * 	1 - driver custom world regulatory domain
- * 	2 - second custom world regulatory domain
- * 	3 - first driver regulatory domain request
- * 	4 - second driver regulatory domain request
- * 	5 - strict regulatory domain settings using the third driver regulatory
- * 	    domain request
- * 	6 and on - should follow the intersection of the 3rd, 4rth and 5th radio
- * 	           regulatory requests.
- */
-enum hwsim_regtest {
-	HWSIM_REGTEST_DISABLED = 0,
-	HWSIM_REGTEST_DRIVER_REG_FOLLOW = 1,
-	HWSIM_REGTEST_DRIVER_REG_ALL = 2,
-	HWSIM_REGTEST_DIFF_COUNTRY = 3,
-	HWSIM_REGTEST_WORLD_ROAM = 4,
-	HWSIM_REGTEST_CUSTOM_WORLD = 5,
-	HWSIM_REGTEST_CUSTOM_WORLD_2 = 6,
-	HWSIM_REGTEST_STRICT_FOLLOW = 7,
-	HWSIM_REGTEST_STRICT_ALL = 8,
-	HWSIM_REGTEST_STRICT_AND_DRIVER_REG = 9,
-	HWSIM_REGTEST_ALL = 10,
-};
-
-/* Set to one of the HWSIM_REGTEST_* values above */
-static int regtest = HWSIM_REGTEST_DISABLED;
-module_param(regtest, int, 0444);
-MODULE_PARM_DESC(regtest, "The type of regulatory test we want to run");
-
-static const char *hwsim_alpha2s[] = {
-	"FI",
-	"AL",
-	"US",
-	"DE",
-	"JP",
-	"AL",
-};
-
-static const struct ieee80211_regdomain hwsim_world_regdom_custom_01 = {
-	.n_reg_rules = 5,
-	.alpha2 =  "99",
-	.reg_rules = {
-		REG_RULE(2412-10, 2462+10, 40, 0, 20, 0),
-		REG_RULE(2484-10, 2484+10, 40, 0, 20, 0),
-		REG_RULE(5150-10, 5240+10, 40, 0, 30, 0),
-		REG_RULE(5745-10, 5825+10, 40, 0, 30, 0),
-		REG_RULE(5855-10, 5925+10, 40, 0, 33, 0),
-	}
-};
-
-static const struct ieee80211_regdomain hwsim_world_regdom_custom_02 = {
-	.n_reg_rules = 3,
-	.alpha2 =  "99",
-	.reg_rules = {
-		REG_RULE(2412-10, 2462+10, 40, 0, 20, 0),
-		REG_RULE(5725-10, 5850+10, 40, 0, 30,
-			 NL80211_RRF_NO_IR),
-		REG_RULE(5855-10, 5925+10, 40, 0, 33, 0),
-	}
-};
-
-static const struct ieee80211_regdomain hwsim_world_regdom_custom_03 = {
-	.n_reg_rules = 6,
-	.alpha2 =  "99",
-	.reg_rules = {
-		REG_RULE(2412 - 10, 2462 + 10, 40, 0, 20, 0),
-		REG_RULE(2484 - 10, 2484 + 10, 40, 0, 20, 0),
-		REG_RULE(5150 - 10, 5240 + 10, 40, 0, 30, 0),
-		REG_RULE(5745 - 10, 5825 + 10, 40, 0, 30, 0),
-		REG_RULE(5855 - 10, 5925 + 10, 40, 0, 33, 0),
-		REG_RULE(5955 - 10, 7125 + 10, 320, 0, 33, 0),
-	}
-};
-
-static const struct ieee80211_regdomain *hwsim_world_regdom_custom[] = {
-	&hwsim_world_regdom_custom_01,
-	&hwsim_world_regdom_custom_02,
-	&hwsim_world_regdom_custom_03,
-};
-
-struct hwsim_vif_priv {
-	u32 magic;
-	u8 bssid[ETH_ALEN];
-	bool assoc;
-	bool bcn_en;
-	u16 aid;
-};
-
-#define HWSIM_VIF_MAGIC	0x69537748
-
-static inline void hwsim_check_magic(struct ieee80211_vif *vif)
-{
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-	WARN(vp->magic != HWSIM_VIF_MAGIC,
-	     "Invalid VIF (%p) magic %#x, %pM, %d/%d\n",
-	     vif, vp->magic, vif->addr, vif->type, vif->p2p);
-}
-
-static inline void hwsim_set_magic(struct ieee80211_vif *vif)
-{
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-	vp->magic = HWSIM_VIF_MAGIC;
-}
-
-static inline void hwsim_clear_magic(struct ieee80211_vif *vif)
-{
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-	vp->magic = 0;
-}
-
-struct hwsim_sta_priv {
-	u32 magic;
-	unsigned int last_link;
-	u16 active_links_rx;
-};
-
-#define HWSIM_STA_MAGIC	0x6d537749
-
-static inline void hwsim_check_sta_magic(struct ieee80211_sta *sta)
-{
-	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-	WARN_ON(sp->magic != HWSIM_STA_MAGIC);
-}
-
-static inline void hwsim_set_sta_magic(struct ieee80211_sta *sta)
-{
-	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-	sp->magic = HWSIM_STA_MAGIC;
-}
-
-static inline void hwsim_clear_sta_magic(struct ieee80211_sta *sta)
-{
-	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-	sp->magic = 0;
-}
-
-struct hwsim_chanctx_priv {
-	u32 magic;
-};
-
-#define HWSIM_CHANCTX_MAGIC 0x6d53774a
-
-static inline void hwsim_check_chanctx_magic(struct ieee80211_chanctx_conf *c)
-{
-	struct hwsim_chanctx_priv *cp = (void *)c->drv_priv;
-	WARN_ON(cp->magic != HWSIM_CHANCTX_MAGIC);
-}
-
-static inline void hwsim_set_chanctx_magic(struct ieee80211_chanctx_conf *c)
-{
-	struct hwsim_chanctx_priv *cp = (void *)c->drv_priv;
-	cp->magic = HWSIM_CHANCTX_MAGIC;
-}
-
-static inline void hwsim_clear_chanctx_magic(struct ieee80211_chanctx_conf *c)
-{
-	struct hwsim_chanctx_priv *cp = (void *)c->drv_priv;
-	cp->magic = 0;
-}
-
-static unsigned int hwsim_net_id;
-
-static DEFINE_IDA(hwsim_netgroup_ida);
-
-struct hwsim_net {
-	int netgroup;
-	u32 wmediumd;
-};
-
-static inline int hwsim_net_get_netgroup(struct net *net)
-{
-	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
-
-	return hwsim_net->netgroup;
-}
-
-static inline int hwsim_net_set_netgroup(struct net *net)
-{
-	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
-
-	hwsim_net->netgroup = ida_alloc(&hwsim_netgroup_ida, GFP_KERNEL);
-	return hwsim_net->netgroup >= 0 ? 0 : -ENOMEM;
-}
-
-static inline u32 hwsim_net_get_wmediumd(struct net *net)
-{
-	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
-
-	return hwsim_net->wmediumd;
-}
-
-static inline void hwsim_net_set_wmediumd(struct net *net, u32 portid)
-{
-	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
-
-	hwsim_net->wmediumd = portid;
-}
-
-static struct class *hwsim_class;
-
-static struct net_device *hwsim_mon; /* global monitor netdev */
-
-#define CHAN2G(_freq)  { \
-	.band = NL80211_BAND_2GHZ, \
-	.center_freq = (_freq), \
-	.hw_value = (_freq), \
-}
-
-#define CHAN5G(_freq) { \
-	.band = NL80211_BAND_5GHZ, \
-	.center_freq = (_freq), \
-	.hw_value = (_freq), \
-}
-
-#define CHAN6G(_freq) { \
-	.band = NL80211_BAND_6GHZ, \
-	.center_freq = (_freq), \
-	.hw_value = (_freq), \
-}
-
-static const struct ieee80211_channel hwsim_channels_2ghz[] = {
-	CHAN2G(2412), /* Channel 1 */
-	CHAN2G(2417), /* Channel 2 */
-	CHAN2G(2422), /* Channel 3 */
-	CHAN2G(2427), /* Channel 4 */
-	CHAN2G(2432), /* Channel 5 */
-	CHAN2G(2437), /* Channel 6 */
-	CHAN2G(2442), /* Channel 7 */
-	CHAN2G(2447), /* Channel 8 */
-	CHAN2G(2452), /* Channel 9 */
-	CHAN2G(2457), /* Channel 10 */
-	CHAN2G(2462), /* Channel 11 */
-	CHAN2G(2467), /* Channel 12 */
-	CHAN2G(2472), /* Channel 13 */
-	CHAN2G(2484), /* Channel 14 */
-};
-
-static const struct ieee80211_channel hwsim_channels_5ghz[] = {
-	CHAN5G(5180), /* Channel 36 */
-	CHAN5G(5200), /* Channel 40 */
-	CHAN5G(5220), /* Channel 44 */
-	CHAN5G(5240), /* Channel 48 */
-
-	CHAN5G(5260), /* Channel 52 */
-	CHAN5G(5280), /* Channel 56 */
-	CHAN5G(5300), /* Channel 60 */
-	CHAN5G(5320), /* Channel 64 */
-
-	CHAN5G(5500), /* Channel 100 */
-	CHAN5G(5520), /* Channel 104 */
-	CHAN5G(5540), /* Channel 108 */
-	CHAN5G(5560), /* Channel 112 */
-	CHAN5G(5580), /* Channel 116 */
-	CHAN5G(5600), /* Channel 120 */
-	CHAN5G(5620), /* Channel 124 */
-	CHAN5G(5640), /* Channel 128 */
-	CHAN5G(5660), /* Channel 132 */
-	CHAN5G(5680), /* Channel 136 */
-	CHAN5G(5700), /* Channel 140 */
-
-	CHAN5G(5745), /* Channel 149 */
-	CHAN5G(5765), /* Channel 153 */
-	CHAN5G(5785), /* Channel 157 */
-	CHAN5G(5805), /* Channel 161 */
-	CHAN5G(5825), /* Channel 165 */
-	CHAN5G(5845), /* Channel 169 */
-
-	CHAN5G(5855), /* Channel 171 */
-	CHAN5G(5860), /* Channel 172 */
-	CHAN5G(5865), /* Channel 173 */
-	CHAN5G(5870), /* Channel 174 */
-
-	CHAN5G(5875), /* Channel 175 */
-	CHAN5G(5880), /* Channel 176 */
-	CHAN5G(5885), /* Channel 177 */
-	CHAN5G(5890), /* Channel 178 */
-	CHAN5G(5895), /* Channel 179 */
-	CHAN5G(5900), /* Channel 180 */
-	CHAN5G(5905), /* Channel 181 */
-
-	CHAN5G(5910), /* Channel 182 */
-	CHAN5G(5915), /* Channel 183 */
-	CHAN5G(5920), /* Channel 184 */
-	CHAN5G(5925), /* Channel 185 */
-};
-
-static const struct ieee80211_channel hwsim_channels_6ghz[] = {
-	CHAN6G(5955), /* Channel 1 */
-	CHAN6G(5975), /* Channel 5 */
-	CHAN6G(5995), /* Channel 9 */
-	CHAN6G(6015), /* Channel 13 */
-	CHAN6G(6035), /* Channel 17 */
-	CHAN6G(6055), /* Channel 21 */
-	CHAN6G(6075), /* Channel 25 */
-	CHAN6G(6095), /* Channel 29 */
-	CHAN6G(6115), /* Channel 33 */
-	CHAN6G(6135), /* Channel 37 */
-	CHAN6G(6155), /* Channel 41 */
-	CHAN6G(6175), /* Channel 45 */
-	CHAN6G(6195), /* Channel 49 */
-	CHAN6G(6215), /* Channel 53 */
-	CHAN6G(6235), /* Channel 57 */
-	CHAN6G(6255), /* Channel 61 */
-	CHAN6G(6275), /* Channel 65 */
-	CHAN6G(6295), /* Channel 69 */
-	CHAN6G(6315), /* Channel 73 */
-	CHAN6G(6335), /* Channel 77 */
-	CHAN6G(6355), /* Channel 81 */
-	CHAN6G(6375), /* Channel 85 */
-	CHAN6G(6395), /* Channel 89 */
-	CHAN6G(6415), /* Channel 93 */
-	CHAN6G(6435), /* Channel 97 */
-	CHAN6G(6455), /* Channel 181 */
-	CHAN6G(6475), /* Channel 105 */
-	CHAN6G(6495), /* Channel 109 */
-	CHAN6G(6515), /* Channel 113 */
-	CHAN6G(6535), /* Channel 117 */
-	CHAN6G(6555), /* Channel 121 */
-	CHAN6G(6575), /* Channel 125 */
-	CHAN6G(6595), /* Channel 129 */
-	CHAN6G(6615), /* Channel 133 */
-	CHAN6G(6635), /* Channel 137 */
-	CHAN6G(6655), /* Channel 141 */
-	CHAN6G(6675), /* Channel 145 */
-	CHAN6G(6695), /* Channel 149 */
-	CHAN6G(6715), /* Channel 153 */
-	CHAN6G(6735), /* Channel 157 */
-	CHAN6G(6755), /* Channel 161 */
-	CHAN6G(6775), /* Channel 165 */
-	CHAN6G(6795), /* Channel 169 */
-	CHAN6G(6815), /* Channel 173 */
-	CHAN6G(6835), /* Channel 177 */
-	CHAN6G(6855), /* Channel 181 */
-	CHAN6G(6875), /* Channel 185 */
-	CHAN6G(6895), /* Channel 189 */
-	CHAN6G(6915), /* Channel 193 */
-	CHAN6G(6935), /* Channel 197 */
-	CHAN6G(6955), /* Channel 201 */
-	CHAN6G(6975), /* Channel 205 */
-	CHAN6G(6995), /* Channel 209 */
-	CHAN6G(7015), /* Channel 213 */
-	CHAN6G(7035), /* Channel 217 */
-	CHAN6G(7055), /* Channel 221 */
-	CHAN6G(7075), /* Channel 225 */
-	CHAN6G(7095), /* Channel 229 */
-	CHAN6G(7115), /* Channel 233 */
-};
-
-#define NUM_S1G_CHANS_US 51
-static struct ieee80211_channel hwsim_channels_s1g[NUM_S1G_CHANS_US];
-
-static const struct ieee80211_sta_s1g_cap hwsim_s1g_cap = {
-	.s1g = true,
-	.cap = { S1G_CAP0_SGI_1MHZ | S1G_CAP0_SGI_2MHZ,
-		 0,
-		 0,
-		 S1G_CAP3_MAX_MPDU_LEN,
-		 0,
-		 S1G_CAP5_AMPDU,
-		 0,
-		 S1G_CAP7_DUP_1MHZ,
-		 S1G_CAP8_TWT_RESPOND | S1G_CAP8_TWT_REQUEST,
-		 0},
-	.nss_mcs = { 0xfc | 1, /* MCS 7 for 1 SS */
-	/* RX Highest Supported Long GI Data Rate 0:7 */
-		     0,
-	/* RX Highest Supported Long GI Data Rate 0:7 */
-	/* TX S1G MCS Map 0:6 */
-		     0xfa,
-	/* TX S1G MCS Map :7 */
-	/* TX Highest Supported Long GI Data Rate 0:6 */
-		     0x80,
-	/* TX Highest Supported Long GI Data Rate 7:8 */
-	/* Rx Single spatial stream and S1G-MCS Map for 1MHz */
-	/* Tx Single spatial stream and S1G-MCS Map for 1MHz */
-		     0 },
-};
-
-static void hwsim_init_s1g_channels(struct ieee80211_channel *chans)
-{
-	int ch, freq;
-
-	for (ch = 0; ch < NUM_S1G_CHANS_US; ch++) {
-		freq = 902000 + (ch + 1) * 500;
-		chans[ch].band = NL80211_BAND_S1GHZ;
-		chans[ch].center_freq = KHZ_TO_MHZ(freq);
-		chans[ch].freq_offset = freq % 1000;
-		chans[ch].hw_value = ch + 1;
-	}
-}
-
-static const struct ieee80211_rate hwsim_rates[] = {
-	{ .bitrate = 10 },
-	{ .bitrate = 20, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
-	{ .bitrate = 55, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
-	{ .bitrate = 110, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
-	{ .bitrate = 60 },
-	{ .bitrate = 90 },
-	{ .bitrate = 120 },
-	{ .bitrate = 180 },
-	{ .bitrate = 240 },
-	{ .bitrate = 360 },
-	{ .bitrate = 480 },
-	{ .bitrate = 540 }
-};
-
-#define DEFAULT_RX_RSSI -50
-
-static const u32 hwsim_ciphers[] = {
-	WLAN_CIPHER_SUITE_WEP40,
-	WLAN_CIPHER_SUITE_WEP104,
-	WLAN_CIPHER_SUITE_TKIP,
-	WLAN_CIPHER_SUITE_CCMP,
-	WLAN_CIPHER_SUITE_CCMP_256,
-	WLAN_CIPHER_SUITE_GCMP,
-	WLAN_CIPHER_SUITE_GCMP_256,
-	WLAN_CIPHER_SUITE_AES_CMAC,
-	WLAN_CIPHER_SUITE_BIP_CMAC_256,
-	WLAN_CIPHER_SUITE_BIP_GMAC_128,
-	WLAN_CIPHER_SUITE_BIP_GMAC_256,
-};
-
-#define OUI_QCA 0x001374
-#define QCA_NL80211_SUBCMD_TEST 1
-enum qca_nl80211_vendor_subcmds {
-	QCA_WLAN_VENDOR_ATTR_TEST = 8,
-	QCA_WLAN_VENDOR_ATTR_MAX = QCA_WLAN_VENDOR_ATTR_TEST
-};
-
-static const struct nla_policy
-hwsim_vendor_test_policy[QCA_WLAN_VENDOR_ATTR_MAX + 1] = {
-	[QCA_WLAN_VENDOR_ATTR_MAX] = { .type = NLA_U32 },
-};
-
-static int mac80211_hwsim_vendor_cmd_test(struct wiphy *wiphy,
-					  struct wireless_dev *wdev,
-					  const void *data, int data_len)
-{
-	struct sk_buff *skb;
-	struct nlattr *tb[QCA_WLAN_VENDOR_ATTR_MAX + 1];
-	int err;
-	u32 val;
-
-	err = nla_parse_deprecated(tb, QCA_WLAN_VENDOR_ATTR_MAX, data,
-				   data_len, hwsim_vendor_test_policy, NULL);
-	if (err)
-		return err;
-	if (!tb[QCA_WLAN_VENDOR_ATTR_TEST])
-		return -EINVAL;
-	val = nla_get_u32(tb[QCA_WLAN_VENDOR_ATTR_TEST]);
-	wiphy_dbg(wiphy, "%s: test=%u\n", __func__, val);
-
-	/* Send a vendor event as a test. Note that this would not normally be
-	 * done within a command handler, but rather, based on some other
-	 * trigger. For simplicity, this command is used to trigger the event
-	 * here.
-	 *
-	 * event_idx = 0 (index in mac80211_hwsim_vendor_commands)
-	 */
-	skb = cfg80211_vendor_event_alloc(wiphy, wdev, 100, 0, GFP_KERNEL);
-	if (skb) {
-		/* skb_put() or nla_put() will fill up data within
-		 * NL80211_ATTR_VENDOR_DATA.
-		 */
-
-		/* Add vendor data */
-		nla_put_u32(skb, QCA_WLAN_VENDOR_ATTR_TEST, val + 1);
-
-		/* Send the event - this will call nla_nest_end() */
-		cfg80211_vendor_event(skb, GFP_KERNEL);
-	}
-
-	/* Send a response to the command */
-	skb = cfg80211_vendor_cmd_alloc_reply_skb(wiphy, 10);
-	if (!skb)
-		return -ENOMEM;
-
-	/* skb_put() or nla_put() will fill up data within
-	 * NL80211_ATTR_VENDOR_DATA
-	 */
-	nla_put_u32(skb, QCA_WLAN_VENDOR_ATTR_TEST, val + 2);
-
-	return cfg80211_vendor_cmd_reply(skb);
-}
-
-static struct wiphy_vendor_command mac80211_hwsim_vendor_commands[] = {
-	{
-		.info = { .vendor_id = OUI_QCA,
-			  .subcmd = QCA_NL80211_SUBCMD_TEST },
-		.flags = WIPHY_VENDOR_CMD_NEED_NETDEV,
-		.doit = mac80211_hwsim_vendor_cmd_test,
-		.policy = hwsim_vendor_test_policy,
-		.maxattr = QCA_WLAN_VENDOR_ATTR_MAX,
-	}
-};
-
-/* Advertise support vendor specific events */
-static const struct nl80211_vendor_cmd_info mac80211_hwsim_vendor_events[] = {
-	{ .vendor_id = OUI_QCA, .subcmd = 1 },
-};
-
-static DEFINE_SPINLOCK(hwsim_radio_lock);
-static LIST_HEAD(hwsim_radios);
-static struct rhashtable hwsim_radios_rht;
-static int hwsim_radio_idx;
-static int hwsim_radios_generation = 1;
-
-static struct platform_driver mac80211_hwsim_driver = {
-	.driver = {
-		.name = "mac80211_hwsim",
-	},
-};
-
-struct mac80211_hwsim_link_data {
-	u32 link_id;
-	u64 beacon_int	/* beacon interval in us */;
-	struct hrtimer beacon_timer;
-};
-
-struct mac80211_hwsim_data {
-	struct list_head list;
-	struct rhash_head rht;
-	struct ieee80211_hw *hw;
-	struct device *dev;
-	struct ieee80211_supported_band bands[NUM_NL80211_BANDS];
-	struct ieee80211_channel channels_2ghz[ARRAY_SIZE(hwsim_channels_2ghz)];
-	struct ieee80211_channel channels_5ghz[ARRAY_SIZE(hwsim_channels_5ghz)];
-	struct ieee80211_channel channels_6ghz[ARRAY_SIZE(hwsim_channels_6ghz)];
-	struct ieee80211_channel channels_s1g[ARRAY_SIZE(hwsim_channels_s1g)];
-	struct ieee80211_rate rates[ARRAY_SIZE(hwsim_rates)];
-	struct ieee80211_iface_combination if_combination;
-	struct ieee80211_iface_limit if_limits[3];
-	int n_if_limits;
-
-	u32 ciphers[ARRAY_SIZE(hwsim_ciphers)];
-
-	struct mac_address addresses[2];
-	int channels, idx;
-	bool use_chanctx;
-	bool destroy_on_close;
-	u32 portid;
-	char alpha2[2];
-	const struct ieee80211_regdomain *regd;
-
-	struct ieee80211_channel *tmp_chan;
-	struct ieee80211_channel *roc_chan;
-	u32 roc_duration;
-	struct delayed_work roc_start;
-	struct delayed_work roc_done;
-	struct delayed_work hw_scan;
-	struct cfg80211_scan_request *hw_scan_request;
-	struct ieee80211_vif *hw_scan_vif;
-	int scan_chan_idx;
-	u8 scan_addr[ETH_ALEN];
-	struct {
-		struct ieee80211_channel *channel;
-		unsigned long next_start, start, end;
-	} survey_data[ARRAY_SIZE(hwsim_channels_2ghz) +
-		      ARRAY_SIZE(hwsim_channels_5ghz) +
-		      ARRAY_SIZE(hwsim_channels_6ghz)];
-
-	struct ieee80211_channel *channel;
-	enum nl80211_chan_width bw;
-	unsigned int rx_filter;
-	bool started, idle, scanning;
-	struct mutex mutex;
-	enum ps_mode {
-		PS_DISABLED, PS_ENABLED, PS_AUTO_POLL, PS_MANUAL_POLL
-	} ps;
-	bool ps_poll_pending;
-	struct dentry *debugfs;
-
-	atomic_t pending_cookie;
-	struct sk_buff_head pending;	/* packets pending */
-	/*
-	 * Only radios in the same group can communicate together (the
-	 * channel has to match too). Each bit represents a group. A
-	 * radio can be in more than one group.
-	 */
-	u64 group;
-
-	/* group shared by radios created in the same netns */
-	int netgroup;
-	/* wmediumd portid responsible for netgroup of this radio */
-	u32 wmediumd;
-
-	/* difference between this hw's clock and the real clock, in usecs */
-	s64 tsf_offset;
-	s64 bcn_delta;
-	/* absolute beacon transmission time. Used to cover up "tx" delay. */
-	u64 abs_bcn_ts;
-
-	/* Stats */
-	u64 tx_pkts;
-	u64 rx_pkts;
-	u64 tx_bytes;
-	u64 rx_bytes;
-	u64 tx_dropped;
-	u64 tx_failed;
-
-	/* RSSI in rx status of the receiver */
-	int rx_rssi;
-
-	struct mac80211_hwsim_link_data link_data[IEEE80211_MLD_MAX_NUM_LINKS];
-};
-
-static const struct rhashtable_params hwsim_rht_params = {
-	.nelem_hint = 2,
-	.automatic_shrinking = true,
-	.key_len = ETH_ALEN,
-	.key_offset = offsetof(struct mac80211_hwsim_data, addresses[1]),
-	.head_offset = offsetof(struct mac80211_hwsim_data, rht),
-};
-
-struct hwsim_radiotap_hdr {
-	struct ieee80211_radiotap_header hdr;
-	__le64 rt_tsft;
-	u8 rt_flags;
-	u8 rt_rate;
-	__le16 rt_channel;
-	__le16 rt_chbitmask;
-} __packed;
-
-struct hwsim_radiotap_ack_hdr {
-	struct ieee80211_radiotap_header hdr;
-	u8 rt_flags;
-	u8 pad;
-	__le16 rt_channel;
-	__le16 rt_chbitmask;
-} __packed;
-
-/* MAC80211_HWSIM netlink family */
-static struct genl_family hwsim_genl_family;
-
-enum hwsim_multicast_groups {
-	HWSIM_MCGRP_CONFIG,
-};
-
-static const struct genl_multicast_group hwsim_mcgrps[] = {
-	[HWSIM_MCGRP_CONFIG] = { .name = "config", },
-};
-
-/* MAC80211_HWSIM netlink policy */
-
-static const struct nla_policy hwsim_genl_policy[HWSIM_ATTR_MAX + 1] = {
-	[HWSIM_ATTR_ADDR_RECEIVER] = NLA_POLICY_ETH_ADDR_COMPAT,
-	[HWSIM_ATTR_ADDR_TRANSMITTER] = NLA_POLICY_ETH_ADDR_COMPAT,
-	[HWSIM_ATTR_FRAME] = { .type = NLA_BINARY,
-			       .len = IEEE80211_MAX_DATA_LEN },
-	[HWSIM_ATTR_FLAGS] = { .type = NLA_U32 },
-	[HWSIM_ATTR_RX_RATE] = { .type = NLA_U32 },
-	[HWSIM_ATTR_SIGNAL] = { .type = NLA_U32 },
-	[HWSIM_ATTR_TX_INFO] = { .type = NLA_BINARY,
-				 .len = IEEE80211_TX_MAX_RATES *
-					sizeof(struct hwsim_tx_rate)},
-	[HWSIM_ATTR_COOKIE] = { .type = NLA_U64 },
-	[HWSIM_ATTR_CHANNELS] = { .type = NLA_U32 },
-	[HWSIM_ATTR_RADIO_ID] = { .type = NLA_U32 },
-	[HWSIM_ATTR_REG_HINT_ALPHA2] = { .type = NLA_STRING, .len = 2 },
-	[HWSIM_ATTR_REG_CUSTOM_REG] = { .type = NLA_U32 },
-	[HWSIM_ATTR_REG_STRICT_REG] = { .type = NLA_FLAG },
-	[HWSIM_ATTR_SUPPORT_P2P_DEVICE] = { .type = NLA_FLAG },
-	[HWSIM_ATTR_USE_CHANCTX] = { .type = NLA_FLAG },
-	[HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE] = { .type = NLA_FLAG },
-	[HWSIM_ATTR_RADIO_NAME] = { .type = NLA_STRING },
-	[HWSIM_ATTR_NO_VIF] = { .type = NLA_FLAG },
-	[HWSIM_ATTR_FREQ] = { .type = NLA_U32 },
-	[HWSIM_ATTR_TX_INFO_FLAGS] = { .type = NLA_BINARY },
-	[HWSIM_ATTR_PERM_ADDR] = NLA_POLICY_ETH_ADDR_COMPAT,
-	[HWSIM_ATTR_IFTYPE_SUPPORT] = { .type = NLA_U32 },
-	[HWSIM_ATTR_CIPHER_SUPPORT] = { .type = NLA_BINARY },
-	[HWSIM_ATTR_MLO_SUPPORT] = { .type = NLA_FLAG },
-};
-
-#if IS_REACHABLE(CONFIG_VIRTIO)
-
-/* MAC80211_HWSIM virtio queues */
-static struct virtqueue *hwsim_vqs[HWSIM_NUM_VQS];
-static bool hwsim_virtio_enabled;
-static DEFINE_SPINLOCK(hwsim_virtio_lock);
-
-static void hwsim_virtio_rx_work(struct work_struct *work);
-static DECLARE_WORK(hwsim_virtio_rx, hwsim_virtio_rx_work);
-
-static int hwsim_tx_virtio(struct mac80211_hwsim_data *data,
-			   struct sk_buff *skb)
-{
-	struct scatterlist sg[1];
-	unsigned long flags;
-	int err;
-
-	spin_lock_irqsave(&hwsim_virtio_lock, flags);
-	if (!hwsim_virtio_enabled) {
-		err = -ENODEV;
-		goto out_free;
-	}
-
-	sg_init_one(sg, skb->head, skb_end_offset(skb));
-	err = virtqueue_add_outbuf(hwsim_vqs[HWSIM_VQ_TX], sg, 1, skb,
-				   GFP_ATOMIC);
-	if (err)
-		goto out_free;
-	virtqueue_kick(hwsim_vqs[HWSIM_VQ_TX]);
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-	return 0;
-
-out_free:
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-	nlmsg_free(skb);
-	return err;
-}
-#else
-/* cause a linker error if this ends up being needed */
-extern int hwsim_tx_virtio(struct mac80211_hwsim_data *data,
-			   struct sk_buff *skb);
-#define hwsim_virtio_enabled false
-#endif
-
-static int hwsim_get_chanwidth(enum nl80211_chan_width bw)
-{
-	switch (bw) {
-	case NL80211_CHAN_WIDTH_20_NOHT:
-	case NL80211_CHAN_WIDTH_20:
-		return 20;
-	case NL80211_CHAN_WIDTH_40:
-		return 40;
-	case NL80211_CHAN_WIDTH_80:
-		return 80;
-	case NL80211_CHAN_WIDTH_80P80:
-	case NL80211_CHAN_WIDTH_160:
-		return 160;
-	case NL80211_CHAN_WIDTH_320:
-		return 320;
-	case NL80211_CHAN_WIDTH_5:
-		return 5;
-	case NL80211_CHAN_WIDTH_10:
-		return 10;
-	case NL80211_CHAN_WIDTH_1:
-		return 1;
-	case NL80211_CHAN_WIDTH_2:
-		return 2;
-	case NL80211_CHAN_WIDTH_4:
-		return 4;
-	case NL80211_CHAN_WIDTH_8:
-		return 8;
-	case NL80211_CHAN_WIDTH_16:
-		return 16;
-	}
-
-	return INT_MAX;
-}
-
-static void mac80211_hwsim_tx_frame(struct ieee80211_hw *hw,
-				    struct sk_buff *skb,
-				    struct ieee80211_channel *chan);
-
-/* sysfs attributes */
-static void hwsim_send_ps_poll(void *dat, u8 *mac, struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *data = dat;
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-	struct sk_buff *skb;
-	struct ieee80211_pspoll *pspoll;
-
-	if (!vp->assoc)
-		return;
-
-	wiphy_dbg(data->hw->wiphy,
-		  "%s: send PS-Poll to %pM for aid %d\n",
-		  __func__, vp->bssid, vp->aid);
-
-	skb = dev_alloc_skb(sizeof(*pspoll));
-	if (!skb)
-		return;
-	pspoll = skb_put(skb, sizeof(*pspoll));
-	pspoll->frame_control = cpu_to_le16(IEEE80211_FTYPE_CTL |
-					    IEEE80211_STYPE_PSPOLL |
-					    IEEE80211_FCTL_PM);
-	pspoll->aid = cpu_to_le16(0xc000 | vp->aid);
-	memcpy(pspoll->bssid, vp->bssid, ETH_ALEN);
-	memcpy(pspoll->ta, mac, ETH_ALEN);
-
-	rcu_read_lock();
-	mac80211_hwsim_tx_frame(data->hw, skb,
-				rcu_dereference(vif->bss_conf.chanctx_conf)->def.chan);
-	rcu_read_unlock();
-}
-
-static void hwsim_send_nullfunc(struct mac80211_hwsim_data *data, u8 *mac,
-				struct ieee80211_vif *vif, int ps)
-{
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-	struct sk_buff *skb;
-	struct ieee80211_hdr *hdr;
-	struct ieee80211_tx_info *cb;
-
-	if (!vp->assoc)
-		return;
-
-	wiphy_dbg(data->hw->wiphy,
-		  "%s: send data::nullfunc to %pM ps=%d\n",
-		  __func__, vp->bssid, ps);
-
-	skb = dev_alloc_skb(sizeof(*hdr));
-	if (!skb)
-		return;
-	hdr = skb_put(skb, sizeof(*hdr) - ETH_ALEN);
-	hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_DATA |
-					 IEEE80211_STYPE_NULLFUNC |
-					 IEEE80211_FCTL_TODS |
-					 (ps ? IEEE80211_FCTL_PM : 0));
-	hdr->duration_id = cpu_to_le16(0);
-	memcpy(hdr->addr1, vp->bssid, ETH_ALEN);
-	memcpy(hdr->addr2, mac, ETH_ALEN);
-	memcpy(hdr->addr3, vp->bssid, ETH_ALEN);
-
-	cb = IEEE80211_SKB_CB(skb);
-	cb->control.rates[0].count = 1;
-	cb->control.rates[1].idx = -1;
-
-	rcu_read_lock();
-	mac80211_hwsim_tx_frame(data->hw, skb,
-				rcu_dereference(vif->bss_conf.chanctx_conf)->def.chan);
-	rcu_read_unlock();
-}
-
-
-static void hwsim_send_nullfunc_ps(void *dat, u8 *mac,
-				   struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *data = dat;
-	hwsim_send_nullfunc(data, mac, vif, 1);
-}
-
-static void hwsim_send_nullfunc_no_ps(void *dat, u8 *mac,
-				      struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *data = dat;
-	hwsim_send_nullfunc(data, mac, vif, 0);
-}
-
-static int hwsim_fops_ps_read(void *dat, u64 *val)
-{
-	struct mac80211_hwsim_data *data = dat;
-	*val = data->ps;
-	return 0;
-}
-
-static int hwsim_fops_ps_write(void *dat, u64 val)
-{
-	struct mac80211_hwsim_data *data = dat;
-	enum ps_mode old_ps;
-
-	if (val != PS_DISABLED && val != PS_ENABLED && val != PS_AUTO_POLL &&
-	    val != PS_MANUAL_POLL)
-		return -EINVAL;
-
-	if (val == PS_MANUAL_POLL) {
-		if (data->ps != PS_ENABLED)
-			return -EINVAL;
-		local_bh_disable();
-		ieee80211_iterate_active_interfaces_atomic(
-			data->hw, IEEE80211_IFACE_ITER_NORMAL,
-			hwsim_send_ps_poll, data);
-		local_bh_enable();
-		return 0;
-	}
-	old_ps = data->ps;
-	data->ps = val;
-
-	local_bh_disable();
-	if (old_ps == PS_DISABLED && val != PS_DISABLED) {
-		ieee80211_iterate_active_interfaces_atomic(
-			data->hw, IEEE80211_IFACE_ITER_NORMAL,
-			hwsim_send_nullfunc_ps, data);
-	} else if (old_ps != PS_DISABLED && val == PS_DISABLED) {
-		ieee80211_iterate_active_interfaces_atomic(
-			data->hw, IEEE80211_IFACE_ITER_NORMAL,
-			hwsim_send_nullfunc_no_ps, data);
-	}
-	local_bh_enable();
-
-	return 0;
-}
-
-DEFINE_DEBUGFS_ATTRIBUTE(hwsim_fops_ps, hwsim_fops_ps_read, hwsim_fops_ps_write,
-			 "%llu\n");
-
-static int hwsim_write_simulate_radar(void *dat, u64 val)
-{
-	struct mac80211_hwsim_data *data = dat;
-
-	ieee80211_radar_detected(data->hw);
-
-	return 0;
-}
-
-DEFINE_DEBUGFS_ATTRIBUTE(hwsim_simulate_radar, NULL,
-			 hwsim_write_simulate_radar, "%llu\n");
-
-static int hwsim_fops_group_read(void *dat, u64 *val)
-{
-	struct mac80211_hwsim_data *data = dat;
-	*val = data->group;
-	return 0;
-}
-
-static int hwsim_fops_group_write(void *dat, u64 val)
-{
-	struct mac80211_hwsim_data *data = dat;
-	data->group = val;
-	return 0;
-}
-
-DEFINE_DEBUGFS_ATTRIBUTE(hwsim_fops_group,
-			 hwsim_fops_group_read, hwsim_fops_group_write,
-			 "%llx\n");
-
-static int hwsim_fops_rx_rssi_read(void *dat, u64 *val)
-{
-	struct mac80211_hwsim_data *data = dat;
-	*val = data->rx_rssi;
-	return 0;
-}
-
-static int hwsim_fops_rx_rssi_write(void *dat, u64 val)
-{
-	struct mac80211_hwsim_data *data = dat;
-	int rssi = (int)val;
-
-	if (rssi >= 0 || rssi < -100)
-		return -EINVAL;
-
-	data->rx_rssi = rssi;
-	return 0;
-}
-
-DEFINE_DEBUGFS_ATTRIBUTE(hwsim_fops_rx_rssi,
-			 hwsim_fops_rx_rssi_read, hwsim_fops_rx_rssi_write,
-			 "%lld\n");
-
-static netdev_tx_t hwsim_mon_xmit(struct sk_buff *skb,
-					struct net_device *dev)
-{
-	/* TODO: allow packet injection */
-	dev_kfree_skb(skb);
-	return NETDEV_TX_OK;
-}
-
-static inline u64 mac80211_hwsim_get_tsf_raw(void)
-{
-	return ktime_to_us(ktime_get_real());
-}
-
-static __le64 __mac80211_hwsim_get_tsf(struct mac80211_hwsim_data *data)
-{
-	u64 now = mac80211_hwsim_get_tsf_raw();
-	return cpu_to_le64(now + data->tsf_offset);
-}
-
-static u64 mac80211_hwsim_get_tsf(struct ieee80211_hw *hw,
-				  struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	return le64_to_cpu(__mac80211_hwsim_get_tsf(data));
-}
-
-static void mac80211_hwsim_set_tsf(struct ieee80211_hw *hw,
-		struct ieee80211_vif *vif, u64 tsf)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	u64 now = mac80211_hwsim_get_tsf(hw, vif);
-	/* MLD not supported here */
-	u32 bcn_int = data->link_data[0].beacon_int;
-	u64 delta = abs(tsf - now);
-
-	/* adjust after beaconing with new timestamp at old TBTT */
-	if (tsf > now) {
-		data->tsf_offset += delta;
-		data->bcn_delta = do_div(delta, bcn_int);
-	} else {
-		data->tsf_offset -= delta;
-		data->bcn_delta = -(s64)do_div(delta, bcn_int);
-	}
-}
-
-static void mac80211_hwsim_monitor_rx(struct ieee80211_hw *hw,
-				      struct sk_buff *tx_skb,
-				      struct ieee80211_channel *chan)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	struct sk_buff *skb;
-	struct hwsim_radiotap_hdr *hdr;
-	u16 flags, bitrate;
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx_skb);
-	struct ieee80211_rate *txrate = ieee80211_get_tx_rate(hw, info);
-
-	if (!txrate)
-		bitrate = 0;
-	else
-		bitrate = txrate->bitrate;
-
-	if (!netif_running(hwsim_mon))
-		return;
-
-	skb = skb_copy_expand(tx_skb, sizeof(*hdr), 0, GFP_ATOMIC);
-	if (skb == NULL)
-		return;
-
-	hdr = skb_push(skb, sizeof(*hdr));
-	hdr->hdr.it_version = PKTHDR_RADIOTAP_VERSION;
-	hdr->hdr.it_pad = 0;
-	hdr->hdr.it_len = cpu_to_le16(sizeof(*hdr));
-	hdr->hdr.it_present = cpu_to_le32((1 << IEEE80211_RADIOTAP_FLAGS) |
-					  (1 << IEEE80211_RADIOTAP_RATE) |
-					  (1 << IEEE80211_RADIOTAP_TSFT) |
-					  (1 << IEEE80211_RADIOTAP_CHANNEL));
-	hdr->rt_tsft = __mac80211_hwsim_get_tsf(data);
-	hdr->rt_flags = 0;
-	hdr->rt_rate = bitrate / 5;
-	hdr->rt_channel = cpu_to_le16(chan->center_freq);
-	flags = IEEE80211_CHAN_2GHZ;
-	if (txrate && txrate->flags & IEEE80211_RATE_ERP_G)
-		flags |= IEEE80211_CHAN_OFDM;
-	else
-		flags |= IEEE80211_CHAN_CCK;
-	hdr->rt_chbitmask = cpu_to_le16(flags);
-
-	skb->dev = hwsim_mon;
-	skb_reset_mac_header(skb);
-	skb->ip_summed = CHECKSUM_UNNECESSARY;
-	skb->pkt_type = PACKET_OTHERHOST;
-	skb->protocol = htons(ETH_P_802_2);
-	memset(skb->cb, 0, sizeof(skb->cb));
-	netif_rx(skb);
-}
-
-
-static void mac80211_hwsim_monitor_ack(struct ieee80211_channel *chan,
-				       const u8 *addr)
-{
-	struct sk_buff *skb;
-	struct hwsim_radiotap_ack_hdr *hdr;
-	u16 flags;
-	struct ieee80211_hdr *hdr11;
-
-	if (!netif_running(hwsim_mon))
-		return;
-
-	skb = dev_alloc_skb(100);
-	if (skb == NULL)
-		return;
-
-	hdr = skb_put(skb, sizeof(*hdr));
-	hdr->hdr.it_version = PKTHDR_RADIOTAP_VERSION;
-	hdr->hdr.it_pad = 0;
-	hdr->hdr.it_len = cpu_to_le16(sizeof(*hdr));
-	hdr->hdr.it_present = cpu_to_le32((1 << IEEE80211_RADIOTAP_FLAGS) |
-					  (1 << IEEE80211_RADIOTAP_CHANNEL));
-	hdr->rt_flags = 0;
-	hdr->pad = 0;
-	hdr->rt_channel = cpu_to_le16(chan->center_freq);
-	flags = IEEE80211_CHAN_2GHZ;
-	hdr->rt_chbitmask = cpu_to_le16(flags);
-
-	hdr11 = skb_put(skb, 10);
-	hdr11->frame_control = cpu_to_le16(IEEE80211_FTYPE_CTL |
-					   IEEE80211_STYPE_ACK);
-	hdr11->duration_id = cpu_to_le16(0);
-	memcpy(hdr11->addr1, addr, ETH_ALEN);
-
-	skb->dev = hwsim_mon;
-	skb_reset_mac_header(skb);
-	skb->ip_summed = CHECKSUM_UNNECESSARY;
-	skb->pkt_type = PACKET_OTHERHOST;
-	skb->protocol = htons(ETH_P_802_2);
-	memset(skb->cb, 0, sizeof(skb->cb));
-	netif_rx(skb);
-}
-
-struct mac80211_hwsim_addr_match_data {
-	u8 addr[ETH_ALEN];
-	bool ret;
-};
-
-static void mac80211_hwsim_addr_iter(void *data, u8 *mac,
-				     struct ieee80211_vif *vif)
-{
-	int i;
-	struct mac80211_hwsim_addr_match_data *md = data;
-
-	if (memcmp(mac, md->addr, ETH_ALEN) == 0) {
-		md->ret = true;
-		return;
-	}
-
-	/* Match the link address */
-	for (i = 0; i < ARRAY_SIZE(vif->link_conf); i++) {
-		struct ieee80211_bss_conf *conf;
-
-		conf = rcu_dereference(vif->link_conf[i]);
-		if (!conf)
-			continue;
-
-		if (memcmp(conf->addr, md->addr, ETH_ALEN) == 0) {
-			md->ret = true;
-			return;
-		}
-	}
-}
-
-static bool mac80211_hwsim_addr_match(struct mac80211_hwsim_data *data,
-				      const u8 *addr)
-{
-	struct mac80211_hwsim_addr_match_data md = {
-		.ret = false,
-	};
-
-	if (data->scanning && memcmp(addr, data->scan_addr, ETH_ALEN) == 0)
-		return true;
-
-	memcpy(md.addr, addr, ETH_ALEN);
-
-	ieee80211_iterate_active_interfaces_atomic(data->hw,
-						   IEEE80211_IFACE_ITER_NORMAL,
-						   mac80211_hwsim_addr_iter,
-						   &md);
-
-	return md.ret;
-}
-
-static bool hwsim_ps_rx_ok(struct mac80211_hwsim_data *data,
-			   struct sk_buff *skb)
-{
-	switch (data->ps) {
-	case PS_DISABLED:
-		return true;
-	case PS_ENABLED:
-		return false;
-	case PS_AUTO_POLL:
-		/* TODO: accept (some) Beacons by default and other frames only
-		 * if pending PS-Poll has been sent */
-		return true;
-	case PS_MANUAL_POLL:
-		/* Allow unicast frames to own address if there is a pending
-		 * PS-Poll */
-		if (data->ps_poll_pending &&
-		    mac80211_hwsim_addr_match(data, skb->data + 4)) {
-			data->ps_poll_pending = false;
-			return true;
-		}
-		return false;
-	}
-
-	return true;
-}
-
-static int hwsim_unicast_netgroup(struct mac80211_hwsim_data *data,
-				  struct sk_buff *skb, int portid)
-{
-	struct net *net;
-	bool found = false;
-	int res = -ENOENT;
-
-	rcu_read_lock();
-	for_each_net_rcu(net) {
-		if (data->netgroup == hwsim_net_get_netgroup(net)) {
-			res = genlmsg_unicast(net, skb, portid);
-			found = true;
-			break;
-		}
-	}
-	rcu_read_unlock();
-
-	if (!found)
-		nlmsg_free(skb);
-
-	return res;
-}
-
-static void mac80211_hwsim_config_mac_nl(struct ieee80211_hw *hw,
-					 const u8 *addr, bool add)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	u32 _portid = READ_ONCE(data->wmediumd);
-	struct sk_buff *skb;
-	void *msg_head;
-
-	WARN_ON(!is_valid_ether_addr(addr));
-
-	if (!_portid && !hwsim_virtio_enabled)
-		return;
-
-	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_ATOMIC);
-	if (!skb)
-		return;
-
-	msg_head = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0,
-			       add ? HWSIM_CMD_ADD_MAC_ADDR :
-				     HWSIM_CMD_DEL_MAC_ADDR);
-	if (!msg_head) {
-		pr_debug("mac80211_hwsim: problem with msg_head\n");
-		goto nla_put_failure;
-	}
-
-	if (nla_put(skb, HWSIM_ATTR_ADDR_TRANSMITTER,
-		    ETH_ALEN, data->addresses[1].addr))
-		goto nla_put_failure;
-
-	if (nla_put(skb, HWSIM_ATTR_ADDR_RECEIVER, ETH_ALEN, addr))
-		goto nla_put_failure;
-
-	genlmsg_end(skb, msg_head);
-
-	if (hwsim_virtio_enabled)
-		hwsim_tx_virtio(data, skb);
-	else
-		hwsim_unicast_netgroup(data, skb, _portid);
-	return;
-nla_put_failure:
-	nlmsg_free(skb);
-}
-
-static inline u16 trans_tx_rate_flags_ieee2hwsim(struct ieee80211_tx_rate *rate)
-{
-	u16 result = 0;
-
-	if (rate->flags & IEEE80211_TX_RC_USE_RTS_CTS)
-		result |= MAC80211_HWSIM_TX_RC_USE_RTS_CTS;
-	if (rate->flags & IEEE80211_TX_RC_USE_CTS_PROTECT)
-		result |= MAC80211_HWSIM_TX_RC_USE_CTS_PROTECT;
-	if (rate->flags & IEEE80211_TX_RC_USE_SHORT_PREAMBLE)
-		result |= MAC80211_HWSIM_TX_RC_USE_SHORT_PREAMBLE;
-	if (rate->flags & IEEE80211_TX_RC_MCS)
-		result |= MAC80211_HWSIM_TX_RC_MCS;
-	if (rate->flags & IEEE80211_TX_RC_GREEN_FIELD)
-		result |= MAC80211_HWSIM_TX_RC_GREEN_FIELD;
-	if (rate->flags & IEEE80211_TX_RC_40_MHZ_WIDTH)
-		result |= MAC80211_HWSIM_TX_RC_40_MHZ_WIDTH;
-	if (rate->flags & IEEE80211_TX_RC_DUP_DATA)
-		result |= MAC80211_HWSIM_TX_RC_DUP_DATA;
-	if (rate->flags & IEEE80211_TX_RC_SHORT_GI)
-		result |= MAC80211_HWSIM_TX_RC_SHORT_GI;
-	if (rate->flags & IEEE80211_TX_RC_VHT_MCS)
-		result |= MAC80211_HWSIM_TX_RC_VHT_MCS;
-	if (rate->flags & IEEE80211_TX_RC_80_MHZ_WIDTH)
-		result |= MAC80211_HWSIM_TX_RC_80_MHZ_WIDTH;
-	if (rate->flags & IEEE80211_TX_RC_160_MHZ_WIDTH)
-		result |= MAC80211_HWSIM_TX_RC_160_MHZ_WIDTH;
-
-	return result;
-}
-
-static void mac80211_hwsim_tx_frame_nl(struct ieee80211_hw *hw,
-				       struct sk_buff *my_skb,
-				       int dst_portid,
-				       struct ieee80211_channel *channel)
-{
-	struct sk_buff *skb;
-	struct mac80211_hwsim_data *data = hw->priv;
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) my_skb->data;
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(my_skb);
-	void *msg_head;
-	unsigned int hwsim_flags = 0;
-	int i;
-	struct hwsim_tx_rate tx_attempts[IEEE80211_TX_MAX_RATES];
-	struct hwsim_tx_rate_flag tx_attempts_flags[IEEE80211_TX_MAX_RATES];
-	uintptr_t cookie;
-
-	if (data->ps != PS_DISABLED)
-		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
-	/* If the queue contains MAX_QUEUE skb's drop some */
-	if (skb_queue_len(&data->pending) >= MAX_QUEUE) {
-		/* Dropping until WARN_QUEUE level */
-		while (skb_queue_len(&data->pending) >= WARN_QUEUE) {
-			ieee80211_free_txskb(hw, skb_dequeue(&data->pending));
-			data->tx_dropped++;
-		}
-	}
-
-	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_ATOMIC);
-	if (skb == NULL)
-		goto nla_put_failure;
-
-	msg_head = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0,
-			       HWSIM_CMD_FRAME);
-	if (msg_head == NULL) {
-		pr_debug("mac80211_hwsim: problem with msg_head\n");
-		goto nla_put_failure;
-	}
-
-	if (nla_put(skb, HWSIM_ATTR_ADDR_TRANSMITTER,
-		    ETH_ALEN, data->addresses[1].addr))
-		goto nla_put_failure;
-
-	/* We get the skb->data */
-	if (nla_put(skb, HWSIM_ATTR_FRAME, my_skb->len, my_skb->data))
-		goto nla_put_failure;
-
-	/* We get the flags for this transmission, and we translate them to
-	   wmediumd flags  */
-
-	if (info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS)
-		hwsim_flags |= HWSIM_TX_CTL_REQ_TX_STATUS;
-
-	if (info->flags & IEEE80211_TX_CTL_NO_ACK)
-		hwsim_flags |= HWSIM_TX_CTL_NO_ACK;
-
-	if (nla_put_u32(skb, HWSIM_ATTR_FLAGS, hwsim_flags))
-		goto nla_put_failure;
-
-	if (nla_put_u32(skb, HWSIM_ATTR_FREQ, channel->center_freq))
-		goto nla_put_failure;
-
-	/* We get the tx control (rate and retries) info*/
-
-	for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) {
-		tx_attempts[i].idx = info->status.rates[i].idx;
-		tx_attempts_flags[i].idx = info->status.rates[i].idx;
-		tx_attempts[i].count = info->status.rates[i].count;
-		tx_attempts_flags[i].flags =
-				trans_tx_rate_flags_ieee2hwsim(
-						&info->status.rates[i]);
-	}
-
-	if (nla_put(skb, HWSIM_ATTR_TX_INFO,
-		    sizeof(struct hwsim_tx_rate)*IEEE80211_TX_MAX_RATES,
-		    tx_attempts))
-		goto nla_put_failure;
-
-	if (nla_put(skb, HWSIM_ATTR_TX_INFO_FLAGS,
-		    sizeof(struct hwsim_tx_rate_flag) * IEEE80211_TX_MAX_RATES,
-		    tx_attempts_flags))
-		goto nla_put_failure;
-
-	/* We create a cookie to identify this skb */
-	cookie = atomic_inc_return(&data->pending_cookie);
-	info->rate_driver_data[0] = (void *)cookie;
-	if (nla_put_u64_64bit(skb, HWSIM_ATTR_COOKIE, cookie, HWSIM_ATTR_PAD))
-		goto nla_put_failure;
-
-	genlmsg_end(skb, msg_head);
-
-	if (hwsim_virtio_enabled) {
-		if (hwsim_tx_virtio(data, skb))
-			goto err_free_txskb;
-	} else {
-		if (hwsim_unicast_netgroup(data, skb, dst_portid))
-			goto err_free_txskb;
-	}
-
-	/* Enqueue the packet */
-	skb_queue_tail(&data->pending, my_skb);
-	data->tx_pkts++;
-	data->tx_bytes += my_skb->len;
-	return;
-
-nla_put_failure:
-	nlmsg_free(skb);
-err_free_txskb:
-	pr_debug("mac80211_hwsim: error occurred in %s\n", __func__);
-	ieee80211_free_txskb(hw, my_skb);
-	data->tx_failed++;
-}
-
-static bool hwsim_chans_compat(struct ieee80211_channel *c1,
-			       struct ieee80211_channel *c2)
-{
-	if (!c1 || !c2)
-		return false;
-
-	return c1->center_freq == c2->center_freq;
-}
-
-struct tx_iter_data {
-	struct ieee80211_channel *channel;
-	bool receive;
-};
-
-static void mac80211_hwsim_tx_iter(void *_data, u8 *addr,
-				   struct ieee80211_vif *vif)
-{
-	struct tx_iter_data *data = _data;
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(vif->link_conf); i++) {
-		struct ieee80211_bss_conf *conf;
-		struct ieee80211_chanctx_conf *chanctx;
-
-		conf = rcu_dereference(vif->link_conf[i]);
-		if (!conf)
-			continue;
-
-		chanctx = rcu_dereference(conf->chanctx_conf);
-		if (!chanctx)
-			continue;
-
-		if (!hwsim_chans_compat(data->channel, chanctx->def.chan))
-			continue;
-
-		data->receive = true;
-		return;
-	}
-}
-
-static void mac80211_hwsim_add_vendor_rtap(struct sk_buff *skb)
-{
-	/*
-	 * To enable this code, #define the HWSIM_RADIOTAP_OUI,
-	 * e.g. like this:
-	 * #define HWSIM_RADIOTAP_OUI "\x02\x00\x00"
-	 * (but you should use a valid OUI, not that)
-	 *
-	 * If anyone wants to 'donate' a radiotap OUI/subns code
-	 * please send a patch removing this #ifdef and changing
-	 * the values accordingly.
-	 */
-#ifdef HWSIM_RADIOTAP_OUI
-	struct ieee80211_radiotap_vendor_tlv *rtap;
-	static const char vendor_data[8] = "ABCDEFGH";
-
-	// Make sure no padding is needed
-	BUILD_BUG_ON(sizeof(vendor_data) % 4);
-	/* this is last radiotap info before the mac header, so
-	 * skb_reset_mac_header for mac8022 to know the end of
-	 * the radiotap TLV/beginning of the 802.11 header
-	 */
-	skb_reset_mac_header(skb);
-
-	/*
-	 * Note that this code requires the headroom in the SKB
-	 * that was allocated earlier.
-	 */
-	rtap = skb_push(skb, sizeof(*rtap) + sizeof(vendor_data));
-
-	rtap->len = cpu_to_le16(sizeof(*rtap) -
-				sizeof(struct ieee80211_radiotap_tlv) +
-				sizeof(vendor_data));
-	rtap->type = cpu_to_le16(IEEE80211_RADIOTAP_VENDOR_NAMESPACE);
-
-	rtap->content.oui[0] = HWSIM_RADIOTAP_OUI[0];
-	rtap->content.oui[1] = HWSIM_RADIOTAP_OUI[1];
-	rtap->content.oui[2] = HWSIM_RADIOTAP_OUI[2];
-	rtap->content.oui_subtype = 127;
-	/* clear reserved field */
-	rtap->content.reserved = 0;
-	rtap->content.vendor_type = 0;
-	memcpy(rtap->content.data, vendor_data, sizeof(vendor_data));
-
-	IEEE80211_SKB_RXCB(skb)->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
-#endif
-}
-
-static void mac80211_hwsim_rx(struct mac80211_hwsim_data *data,
-			      struct ieee80211_rx_status *rx_status,
-			      struct sk_buff *skb)
-{
-	struct ieee80211_hdr *hdr = (void *)skb->data;
-
-	if (!ieee80211_has_morefrags(hdr->frame_control) &&
-	    !is_multicast_ether_addr(hdr->addr1) &&
-	    (ieee80211_is_mgmt(hdr->frame_control) ||
-	     ieee80211_is_data(hdr->frame_control))) {
-		struct ieee80211_sta *sta;
-		unsigned int link_id;
-
-		rcu_read_lock();
-		sta = ieee80211_find_sta_by_link_addrs(data->hw, hdr->addr2,
-						       hdr->addr1, &link_id);
-		if (sta) {
-			struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-
-			if (ieee80211_has_pm(hdr->frame_control))
-				sp->active_links_rx &= ~BIT(link_id);
-			else
-				sp->active_links_rx |= BIT(link_id);
-		}
-		rcu_read_unlock();
-	}
-
-	memcpy(IEEE80211_SKB_RXCB(skb), rx_status, sizeof(*rx_status));
-
-	mac80211_hwsim_add_vendor_rtap(skb);
-
-	data->rx_pkts++;
-	data->rx_bytes += skb->len;
-	ieee80211_rx_irqsafe(data->hw, skb);
-}
-
-static bool mac80211_hwsim_tx_frame_no_nl(struct ieee80211_hw *hw,
-					  struct sk_buff *skb,
-					  struct ieee80211_channel *chan)
-{
-	struct mac80211_hwsim_data *data = hw->priv, *data2;
-	bool ack = false;
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
-	struct ieee80211_rx_status rx_status;
-	u64 now;
-
-	memset(&rx_status, 0, sizeof(rx_status));
-	rx_status.flag |= RX_FLAG_MACTIME_START;
-	rx_status.freq = chan->center_freq;
-	rx_status.freq_offset = chan->freq_offset ? 1 : 0;
-	rx_status.band = chan->band;
-	if (info->control.rates[0].flags & IEEE80211_TX_RC_VHT_MCS) {
-		rx_status.rate_idx =
-			ieee80211_rate_get_vht_mcs(&info->control.rates[0]);
-		rx_status.nss =
-			ieee80211_rate_get_vht_nss(&info->control.rates[0]);
-		rx_status.encoding = RX_ENC_VHT;
-	} else {
-		rx_status.rate_idx = info->control.rates[0].idx;
-		if (info->control.rates[0].flags & IEEE80211_TX_RC_MCS)
-			rx_status.encoding = RX_ENC_HT;
-	}
-	if (info->control.rates[0].flags & IEEE80211_TX_RC_40_MHZ_WIDTH)
-		rx_status.bw = RATE_INFO_BW_40;
-	else if (info->control.rates[0].flags & IEEE80211_TX_RC_80_MHZ_WIDTH)
-		rx_status.bw = RATE_INFO_BW_80;
-	else if (info->control.rates[0].flags & IEEE80211_TX_RC_160_MHZ_WIDTH)
-		rx_status.bw = RATE_INFO_BW_160;
-	else
-		rx_status.bw = RATE_INFO_BW_20;
-	if (info->control.rates[0].flags & IEEE80211_TX_RC_SHORT_GI)
-		rx_status.enc_flags |= RX_ENC_FLAG_SHORT_GI;
-	/* TODO: simulate optional packet loss */
-	rx_status.signal = data->rx_rssi;
-	if (info->control.vif)
-		rx_status.signal += info->control.vif->bss_conf.txpower;
-
-	if (data->ps != PS_DISABLED)
-		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
-
-	/* release the skb's source info */
-	skb_orphan(skb);
-	skb_dst_drop(skb);
-	skb->mark = 0;
-	skb_ext_reset(skb);
-	nf_reset_ct(skb);
-
-	/*
-	 * Get absolute mactime here so all HWs RX at the "same time", and
-	 * absolute TX time for beacon mactime so the timestamp matches.
-	 * Giving beacons a different mactime than non-beacons looks messy, but
-	 * it helps the Toffset be exact and a ~10us mactime discrepancy
-	 * probably doesn't really matter.
-	 */
-	if (ieee80211_is_beacon(hdr->frame_control) ||
-	    ieee80211_is_probe_resp(hdr->frame_control)) {
-		rx_status.boottime_ns = ktime_get_boottime_ns();
-		now = data->abs_bcn_ts;
-	} else {
-		now = mac80211_hwsim_get_tsf_raw();
-	}
-
-	/* Copy skb to all enabled radios that are on the current frequency */
-	spin_lock(&hwsim_radio_lock);
-	list_for_each_entry(data2, &hwsim_radios, list) {
-		struct sk_buff *nskb;
-		struct tx_iter_data tx_iter_data = {
-			.receive = false,
-			.channel = chan,
-		};
-
-		if (data == data2)
-			continue;
-
-		if (!data2->started || (data2->idle && !data2->tmp_chan) ||
-		    !hwsim_ps_rx_ok(data2, skb))
-			continue;
-
-		if (!(data->group & data2->group))
-			continue;
-
-		if (data->netgroup != data2->netgroup)
-			continue;
-
-		if (!hwsim_chans_compat(chan, data2->tmp_chan) &&
-		    !hwsim_chans_compat(chan, data2->channel)) {
-			ieee80211_iterate_active_interfaces_atomic(
-				data2->hw, IEEE80211_IFACE_ITER_NORMAL,
-				mac80211_hwsim_tx_iter, &tx_iter_data);
-			if (!tx_iter_data.receive)
-				continue;
-		}
-
-		/*
-		 * reserve some space for our vendor and the normal
-		 * radiotap header, since we're copying anyway
-		 */
-		if (skb->len < PAGE_SIZE && paged_rx) {
-			struct page *page = alloc_page(GFP_ATOMIC);
-
-			if (!page)
-				continue;
-
-			nskb = dev_alloc_skb(128);
-			if (!nskb) {
-				__free_page(page);
-				continue;
-			}
-
-			memcpy(page_address(page), skb->data, skb->len);
-			skb_add_rx_frag(nskb, 0, page, 0, skb->len, skb->len);
-		} else {
-			nskb = skb_copy(skb, GFP_ATOMIC);
-			if (!nskb)
-				continue;
-		}
-
-		if (mac80211_hwsim_addr_match(data2, hdr->addr1))
-			ack = true;
-
-		rx_status.mactime = now + data2->tsf_offset;
-
-		mac80211_hwsim_rx(data2, &rx_status, nskb);
-	}
-	spin_unlock(&hwsim_radio_lock);
-
-	return ack;
-}
-
-static struct ieee80211_bss_conf *
-mac80211_hwsim_select_tx_link(struct mac80211_hwsim_data *data,
-			      struct ieee80211_vif *vif,
-			      struct ieee80211_sta *sta,
-			      struct ieee80211_hdr *hdr,
-			      struct ieee80211_link_sta **link_sta)
-{
-	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-	int i;
-
-	if (!vif->valid_links)
-		return &vif->bss_conf;
-
-	WARN_ON(is_multicast_ether_addr(hdr->addr1));
-
-	if (WARN_ON_ONCE(!sta->valid_links))
-		return &vif->bss_conf;
-
-	for (i = 0; i < ARRAY_SIZE(vif->link_conf); i++) {
-		struct ieee80211_bss_conf *bss_conf;
-		unsigned int link_id;
-
-		/* round-robin the available link IDs */
-		link_id = (sp->last_link + i + 1) % ARRAY_SIZE(vif->link_conf);
-
-		if (!(vif->active_links & BIT(link_id)))
-			continue;
-
-		if (!(sp->active_links_rx & BIT(link_id)))
-			continue;
-
-		*link_sta = rcu_dereference(sta->link[link_id]);
-		if (!*link_sta)
-			continue;
-
-		bss_conf = rcu_dereference(vif->link_conf[link_id]);
-		if (WARN_ON_ONCE(!bss_conf))
-			continue;
-
-		/* can happen while switching links */
-		if (!rcu_access_pointer(bss_conf->chanctx_conf))
-			continue;
-
-		sp->last_link = link_id;
-		return bss_conf;
-	}
-
-	return NULL;
-}
-
-static void mac80211_hwsim_tx(struct ieee80211_hw *hw,
-			      struct ieee80211_tx_control *control,
-			      struct sk_buff *skb)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	struct ieee80211_tx_info *txi = IEEE80211_SKB_CB(skb);
-	struct ieee80211_hdr *hdr = (void *)skb->data;
-	struct ieee80211_chanctx_conf *chanctx_conf;
-	struct ieee80211_channel *channel;
-	bool ack;
-	enum nl80211_chan_width confbw = NL80211_CHAN_WIDTH_20_NOHT;
-	u32 _portid, i;
-
-	if (WARN_ON(skb->len < 10)) {
-		/* Should not happen; just a sanity check for addr1 use */
-		ieee80211_free_txskb(hw, skb);
-		return;
-	}
-
-	if (!data->use_chanctx) {
-		channel = data->channel;
-		confbw = data->bw;
-	} else if (txi->hw_queue == 4) {
-		channel = data->tmp_chan;
-	} else {
-		u8 link = u32_get_bits(IEEE80211_SKB_CB(skb)->control.flags,
-				       IEEE80211_TX_CTRL_MLO_LINK);
-		struct ieee80211_vif *vif = txi->control.vif;
-		struct ieee80211_link_sta *link_sta = NULL;
-		struct ieee80211_sta *sta = control->sta;
-		struct ieee80211_bss_conf *bss_conf;
-
-		if (link != IEEE80211_LINK_UNSPECIFIED) {
-			bss_conf = rcu_dereference(txi->control.vif->link_conf[link]);
-			if (sta)
-				link_sta = rcu_dereference(sta->link[link]);
-		} else {
-			bss_conf = mac80211_hwsim_select_tx_link(data, vif, sta,
-								 hdr, &link_sta);
-		}
-
-		if (WARN_ON(!bss_conf)) {
-			ieee80211_free_txskb(hw, skb);
-			return;
-		}
-
-		if (sta && sta->mlo) {
-			if (WARN_ON(!link_sta)) {
-				ieee80211_free_txskb(hw, skb);
-				return;
-			}
-			/* address translation to link addresses on TX */
-			ether_addr_copy(hdr->addr1, link_sta->addr);
-			ether_addr_copy(hdr->addr2, bss_conf->addr);
-			/* translate A3 only if it's the BSSID */
-			if (!ieee80211_has_tods(hdr->frame_control) &&
-			    !ieee80211_has_fromds(hdr->frame_control)) {
-				if (ether_addr_equal(hdr->addr3, sta->addr))
-					ether_addr_copy(hdr->addr3, link_sta->addr);
-				else if (ether_addr_equal(hdr->addr3, vif->addr))
-					ether_addr_copy(hdr->addr3, bss_conf->addr);
-			}
-			/* no need to look at A4, if present it's SA */
-		}
-
-		chanctx_conf = rcu_dereference(bss_conf->chanctx_conf);
-		if (chanctx_conf) {
-			channel = chanctx_conf->def.chan;
-			confbw = chanctx_conf->def.width;
-		} else {
-			channel = NULL;
-		}
-	}
-
-	if (WARN(!channel, "TX w/o channel - queue = %d\n", txi->hw_queue)) {
-		ieee80211_free_txskb(hw, skb);
-		return;
-	}
-
-	if (data->idle && !data->tmp_chan) {
-		wiphy_dbg(hw->wiphy, "Trying to TX when idle - reject\n");
-		ieee80211_free_txskb(hw, skb);
-		return;
-	}
-
-	if (txi->control.vif)
-		hwsim_check_magic(txi->control.vif);
-	if (control->sta)
-		hwsim_check_sta_magic(control->sta);
-
-	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE))
-		ieee80211_get_tx_rates(txi->control.vif, control->sta, skb,
-				       txi->control.rates,
-				       ARRAY_SIZE(txi->control.rates));
-
-	for (i = 0; i < ARRAY_SIZE(txi->control.rates); i++) {
-		u16 rflags = txi->control.rates[i].flags;
-		/* initialize to data->bw for 5/10 MHz handling */
-		enum nl80211_chan_width bw = data->bw;
-
-		if (txi->control.rates[i].idx == -1)
-			break;
-
-		if (rflags & IEEE80211_TX_RC_40_MHZ_WIDTH)
-			bw = NL80211_CHAN_WIDTH_40;
-		else if (rflags & IEEE80211_TX_RC_80_MHZ_WIDTH)
-			bw = NL80211_CHAN_WIDTH_80;
-		else if (rflags & IEEE80211_TX_RC_160_MHZ_WIDTH)
-			bw = NL80211_CHAN_WIDTH_160;
-
-		if (WARN_ON(hwsim_get_chanwidth(bw) > hwsim_get_chanwidth(confbw)))
-			return;
-	}
-
-	if (skb->len >= 24 + 8 &&
-	    ieee80211_is_probe_resp(hdr->frame_control)) {
-		/* fake header transmission time */
-		struct ieee80211_mgmt *mgmt;
-		struct ieee80211_rate *txrate;
-		/* TODO: get MCS */
-		int bitrate = 100;
-		u64 ts;
-
-		mgmt = (struct ieee80211_mgmt *)skb->data;
-		txrate = ieee80211_get_tx_rate(hw, txi);
-		if (txrate)
-			bitrate = txrate->bitrate;
-		ts = mac80211_hwsim_get_tsf_raw();
-		mgmt->u.probe_resp.timestamp =
-			cpu_to_le64(ts + data->tsf_offset +
-				    24 * 8 * 10 / bitrate);
-	}
-
-	mac80211_hwsim_monitor_rx(hw, skb, channel);
-
-	/* wmediumd mode check */
-	_portid = READ_ONCE(data->wmediumd);
-
-	if (_portid || hwsim_virtio_enabled)
-		return mac80211_hwsim_tx_frame_nl(hw, skb, _portid, channel);
-
-	/* NO wmediumd detected, perfect medium simulation */
-	data->tx_pkts++;
-	data->tx_bytes += skb->len;
-	ack = mac80211_hwsim_tx_frame_no_nl(hw, skb, channel);
-
-	if (ack && skb->len >= 16)
-		mac80211_hwsim_monitor_ack(channel, hdr->addr2);
-
-	ieee80211_tx_info_clear_status(txi);
-
-	/* frame was transmitted at most favorable rate at first attempt */
-	txi->control.rates[0].count = 1;
-	txi->control.rates[1].idx = -1;
-
-	if (!(txi->flags & IEEE80211_TX_CTL_NO_ACK) && ack)
-		txi->flags |= IEEE80211_TX_STAT_ACK;
-	ieee80211_tx_status_irqsafe(hw, skb);
-}
-
-
-static int mac80211_hwsim_start(struct ieee80211_hw *hw)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	wiphy_dbg(hw->wiphy, "%s\n", __func__);
-	data->started = true;
-	return 0;
-}
-
-
-static void mac80211_hwsim_stop(struct ieee80211_hw *hw)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	int i;
-
-	data->started = false;
-
-	for (i = 0; i < ARRAY_SIZE(data->link_data); i++)
-		hrtimer_cancel(&data->link_data[i].beacon_timer);
-
-	while (!skb_queue_empty(&data->pending))
-		ieee80211_free_txskb(hw, skb_dequeue(&data->pending));
-
-	wiphy_dbg(hw->wiphy, "%s\n", __func__);
-}
-
-
-static int mac80211_hwsim_add_interface(struct ieee80211_hw *hw,
-					struct ieee80211_vif *vif)
-{
-	wiphy_dbg(hw->wiphy, "%s (type=%d mac_addr=%pM)\n",
-		  __func__, ieee80211_vif_type_p2p(vif),
-		  vif->addr);
-	hwsim_set_magic(vif);
-
-	if (vif->type != NL80211_IFTYPE_MONITOR)
-		mac80211_hwsim_config_mac_nl(hw, vif->addr, true);
-
-	vif->cab_queue = 0;
-	vif->hw_queue[IEEE80211_AC_VO] = 0;
-	vif->hw_queue[IEEE80211_AC_VI] = 1;
-	vif->hw_queue[IEEE80211_AC_BE] = 2;
-	vif->hw_queue[IEEE80211_AC_BK] = 3;
-
-	return 0;
-}
-
-
-static int mac80211_hwsim_change_interface(struct ieee80211_hw *hw,
-					   struct ieee80211_vif *vif,
-					   enum nl80211_iftype newtype,
-					   bool newp2p)
-{
-	newtype = ieee80211_iftype_p2p(newtype, newp2p);
-	wiphy_dbg(hw->wiphy,
-		  "%s (old type=%d, new type=%d, mac_addr=%pM)\n",
-		  __func__, ieee80211_vif_type_p2p(vif),
-		    newtype, vif->addr);
-	hwsim_check_magic(vif);
-
-	/*
-	 * interface may change from non-AP to AP in
-	 * which case this needs to be set up again
-	 */
-	vif->cab_queue = 0;
-
-	return 0;
-}
-
-static void mac80211_hwsim_remove_interface(
-	struct ieee80211_hw *hw, struct ieee80211_vif *vif)
-{
-	wiphy_dbg(hw->wiphy, "%s (type=%d mac_addr=%pM)\n",
-		  __func__, ieee80211_vif_type_p2p(vif),
-		  vif->addr);
-	hwsim_check_magic(vif);
-	hwsim_clear_magic(vif);
-	if (vif->type != NL80211_IFTYPE_MONITOR)
-		mac80211_hwsim_config_mac_nl(hw, vif->addr, false);
-}
-
-static void mac80211_hwsim_tx_frame(struct ieee80211_hw *hw,
-				    struct sk_buff *skb,
-				    struct ieee80211_channel *chan)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	u32 _portid = READ_ONCE(data->wmediumd);
-
-	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE)) {
-		struct ieee80211_tx_info *txi = IEEE80211_SKB_CB(skb);
-		ieee80211_get_tx_rates(txi->control.vif, NULL, skb,
-				       txi->control.rates,
-				       ARRAY_SIZE(txi->control.rates));
-	}
-
-	mac80211_hwsim_monitor_rx(hw, skb, chan);
-
-	if (_portid || hwsim_virtio_enabled)
-		return mac80211_hwsim_tx_frame_nl(hw, skb, _portid, chan);
-
-	data->tx_pkts++;
-	data->tx_bytes += skb->len;
-	mac80211_hwsim_tx_frame_no_nl(hw, skb, chan);
-	dev_kfree_skb(skb);
-}
-
-static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
-				     struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_link_data *link_data = arg;
-	u32 link_id = link_data->link_id;
-	struct ieee80211_bss_conf *link_conf;
-	struct mac80211_hwsim_data *data =
-		container_of(link_data, struct mac80211_hwsim_data,
-			     link_data[link_id]);
-	struct ieee80211_hw *hw = data->hw;
-	struct ieee80211_tx_info *info;
-	struct ieee80211_rate *txrate;
-	struct ieee80211_mgmt *mgmt;
-	struct sk_buff *skb;
-	/* TODO: get MCS */
-	int bitrate = 100;
-
-	hwsim_check_magic(vif);
-
-	link_conf = rcu_dereference(vif->link_conf[link_id]);
-	if (!link_conf)
-		return;
-
-	if (vif->type != NL80211_IFTYPE_AP &&
-	    vif->type != NL80211_IFTYPE_MESH_POINT &&
-	    vif->type != NL80211_IFTYPE_ADHOC &&
-	    vif->type != NL80211_IFTYPE_OCB)
-		return;
-
-	skb = ieee80211_beacon_get(hw, vif, link_data->link_id);
-	if (skb == NULL)
-		return;
-	info = IEEE80211_SKB_CB(skb);
-	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE))
-		ieee80211_get_tx_rates(vif, NULL, skb,
-				       info->control.rates,
-				       ARRAY_SIZE(info->control.rates));
-
-	txrate = ieee80211_get_tx_rate(hw, info);
-	if (txrate)
-		bitrate = txrate->bitrate;
-
-	mgmt = (struct ieee80211_mgmt *) skb->data;
-	/* fake header transmission time */
-	data->abs_bcn_ts = mac80211_hwsim_get_tsf_raw();
-	if (ieee80211_is_s1g_beacon(mgmt->frame_control)) {
-		struct ieee80211_ext *ext = (void *) mgmt;
-
-		ext->u.s1g_beacon.timestamp = cpu_to_le32(data->abs_bcn_ts +
-							  data->tsf_offset +
-							  10 * 8 * 10 /
-							  bitrate);
-	} else {
-		mgmt->u.beacon.timestamp = cpu_to_le64(data->abs_bcn_ts +
-						       data->tsf_offset +
-						       24 * 8 * 10 /
-						       bitrate);
-	}
-
-	mac80211_hwsim_tx_frame(hw, skb,
-			rcu_dereference(link_conf->chanctx_conf)->def.chan);
-
-	while ((skb = ieee80211_get_buffered_bc(hw, vif)) != NULL) {
-		mac80211_hwsim_tx_frame(hw, skb,
-			rcu_dereference(link_conf->chanctx_conf)->def.chan);
-	}
-
-	if (link_conf->csa_active && ieee80211_beacon_cntdwn_is_complete(vif))
-		ieee80211_csa_finish(vif);
-}
-
-static enum hrtimer_restart
-mac80211_hwsim_beacon(struct hrtimer *timer)
-{
-	struct mac80211_hwsim_link_data *link_data =
-		container_of(timer, struct mac80211_hwsim_link_data, beacon_timer);
-	struct mac80211_hwsim_data *data =
-		container_of(link_data, struct mac80211_hwsim_data,
-			     link_data[link_data->link_id]);
-	struct ieee80211_hw *hw = data->hw;
-	u64 bcn_int = link_data->beacon_int;
-
-	if (!data->started)
-		return HRTIMER_NORESTART;
-
-	ieee80211_iterate_active_interfaces_atomic(
-		hw, IEEE80211_IFACE_ITER_NORMAL,
-		mac80211_hwsim_beacon_tx, link_data);
-
-	/* beacon at new TBTT + beacon interval */
-	if (data->bcn_delta) {
-		bcn_int -= data->bcn_delta;
-		data->bcn_delta = 0;
-	}
-	hrtimer_forward_now(&link_data->beacon_timer,
-			    ns_to_ktime(bcn_int * NSEC_PER_USEC));
-	return HRTIMER_RESTART;
-}
-
-static const char * const hwsim_chanwidths[] = {
-	[NL80211_CHAN_WIDTH_5] = "ht5",
-	[NL80211_CHAN_WIDTH_10] = "ht10",
-	[NL80211_CHAN_WIDTH_20_NOHT] = "noht",
-	[NL80211_CHAN_WIDTH_20] = "ht20",
-	[NL80211_CHAN_WIDTH_40] = "ht40",
-	[NL80211_CHAN_WIDTH_80] = "vht80",
-	[NL80211_CHAN_WIDTH_80P80] = "vht80p80",
-	[NL80211_CHAN_WIDTH_160] = "vht160",
-	[NL80211_CHAN_WIDTH_1] = "1MHz",
-	[NL80211_CHAN_WIDTH_2] = "2MHz",
-	[NL80211_CHAN_WIDTH_4] = "4MHz",
-	[NL80211_CHAN_WIDTH_8] = "8MHz",
-	[NL80211_CHAN_WIDTH_16] = "16MHz",
-};
-
-static int mac80211_hwsim_config(struct ieee80211_hw *hw, u32 changed)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	struct ieee80211_conf *conf = &hw->conf;
-	static const char *smps_modes[IEEE80211_SMPS_NUM_MODES] = {
-		[IEEE80211_SMPS_AUTOMATIC] = "auto",
-		[IEEE80211_SMPS_OFF] = "off",
-		[IEEE80211_SMPS_STATIC] = "static",
-		[IEEE80211_SMPS_DYNAMIC] = "dynamic",
-	};
-	int idx;
-
-	if (conf->chandef.chan)
-		wiphy_dbg(hw->wiphy,
-			  "%s (freq=%d(%d - %d)/%s idle=%d ps=%d smps=%s)\n",
-			  __func__,
-			  conf->chandef.chan->center_freq,
-			  conf->chandef.center_freq1,
-			  conf->chandef.center_freq2,
-			  hwsim_chanwidths[conf->chandef.width],
-			  !!(conf->flags & IEEE80211_CONF_IDLE),
-			  !!(conf->flags & IEEE80211_CONF_PS),
-			  smps_modes[conf->smps_mode]);
-	else
-		wiphy_dbg(hw->wiphy,
-			  "%s (freq=0 idle=%d ps=%d smps=%s)\n",
-			  __func__,
-			  !!(conf->flags & IEEE80211_CONF_IDLE),
-			  !!(conf->flags & IEEE80211_CONF_PS),
-			  smps_modes[conf->smps_mode]);
-
-	data->idle = !!(conf->flags & IEEE80211_CONF_IDLE);
-
-	WARN_ON(conf->chandef.chan && data->use_chanctx);
-
-	mutex_lock(&data->mutex);
-	if (data->scanning && conf->chandef.chan) {
-		for (idx = 0; idx < ARRAY_SIZE(data->survey_data); idx++) {
-			if (data->survey_data[idx].channel == data->channel) {
-				data->survey_data[idx].start =
-					data->survey_data[idx].next_start;
-				data->survey_data[idx].end = jiffies;
-				break;
-			}
-		}
-
-		data->channel = conf->chandef.chan;
-		data->bw = conf->chandef.width;
-
-		for (idx = 0; idx < ARRAY_SIZE(data->survey_data); idx++) {
-			if (data->survey_data[idx].channel &&
-			    data->survey_data[idx].channel != data->channel)
-				continue;
-			data->survey_data[idx].channel = data->channel;
-			data->survey_data[idx].next_start = jiffies;
-			break;
-		}
-	} else {
-		data->channel = conf->chandef.chan;
-		data->bw = conf->chandef.width;
-	}
-	mutex_unlock(&data->mutex);
-
-	for (idx = 0; idx < ARRAY_SIZE(data->link_data); idx++) {
-		struct mac80211_hwsim_link_data *link_data =
-			&data->link_data[idx];
-
-		if (!data->started || !link_data->beacon_int) {
-			hrtimer_cancel(&link_data->beacon_timer);
-		} else if (!hrtimer_is_queued(&link_data->beacon_timer)) {
-			u64 tsf = mac80211_hwsim_get_tsf(hw, NULL);
-			u32 bcn_int = link_data->beacon_int;
-			u64 until_tbtt = bcn_int - do_div(tsf, bcn_int);
-
-			hrtimer_start(&link_data->beacon_timer,
-				      ns_to_ktime(until_tbtt * NSEC_PER_USEC),
-				      HRTIMER_MODE_REL_SOFT);
-		}
-	}
-
-	return 0;
-}
-
-
-static void mac80211_hwsim_configure_filter(struct ieee80211_hw *hw,
-					    unsigned int changed_flags,
-					    unsigned int *total_flags,u64 multicast)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-
-	wiphy_dbg(hw->wiphy, "%s\n", __func__);
-
-	data->rx_filter = 0;
-	if (*total_flags & FIF_ALLMULTI)
-		data->rx_filter |= FIF_ALLMULTI;
-	if (*total_flags & FIF_MCAST_ACTION)
-		data->rx_filter |= FIF_MCAST_ACTION;
-
-	*total_flags = data->rx_filter;
-}
-
-static void mac80211_hwsim_bcn_en_iter(void *data, u8 *mac,
-				       struct ieee80211_vif *vif)
-{
-	unsigned int *count = data;
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-
-	if (vp->bcn_en)
-		(*count)++;
-}
-
-static void mac80211_hwsim_vif_info_changed(struct ieee80211_hw *hw,
-					    struct ieee80211_vif *vif,
-					    u64 changed)
-{
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-
-	hwsim_check_magic(vif);
-
-	wiphy_dbg(hw->wiphy, "%s(changed=0x%llx vif->addr=%pM)\n",
-		  __func__, changed, vif->addr);
-
-	if (changed & BSS_CHANGED_ASSOC) {
-		wiphy_dbg(hw->wiphy, "  ASSOC: assoc=%d aid=%d\n",
-			  vif->cfg.assoc, vif->cfg.aid);
-		vp->assoc = vif->cfg.assoc;
-		vp->aid = vif->cfg.aid;
-	}
-}
-
-static void mac80211_hwsim_link_info_changed(struct ieee80211_hw *hw,
-					     struct ieee80211_vif *vif,
-					     struct ieee80211_bss_conf *info,
-					     u64 changed)
-{
-	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
-	struct mac80211_hwsim_data *data = hw->priv;
-	unsigned int link_id = info->link_id;
-	struct mac80211_hwsim_link_data *link_data = &data->link_data[link_id];
-
-	hwsim_check_magic(vif);
-
-	wiphy_dbg(hw->wiphy, "%s(changed=0x%llx vif->addr=%pM, link id %u)\n",
-		  __func__, (unsigned long long)changed, vif->addr, link_id);
-
-	if (changed & BSS_CHANGED_BSSID) {
-		wiphy_dbg(hw->wiphy, "%s: BSSID changed: %pM\n",
-			  __func__, info->bssid);
-		memcpy(vp->bssid, info->bssid, ETH_ALEN);
-	}
-
-	if (changed & BSS_CHANGED_BEACON_ENABLED) {
-		wiphy_dbg(hw->wiphy, "  BCN EN: %d (BI=%u)\n",
-			  info->enable_beacon, info->beacon_int);
-		vp->bcn_en = info->enable_beacon;
-		if (data->started &&
-		    !hrtimer_is_queued(&link_data->beacon_timer) &&
-		    info->enable_beacon) {
-			u64 tsf, until_tbtt;
-			u32 bcn_int;
-			link_data->beacon_int = info->beacon_int * 1024;
-			tsf = mac80211_hwsim_get_tsf(hw, vif);
-			bcn_int = link_data->beacon_int;
-			until_tbtt = bcn_int - do_div(tsf, bcn_int);
-
-			hrtimer_start(&link_data->beacon_timer,
-				      ns_to_ktime(until_tbtt * NSEC_PER_USEC),
-				      HRTIMER_MODE_REL_SOFT);
-		} else if (!info->enable_beacon) {
-			unsigned int count = 0;
-			ieee80211_iterate_active_interfaces_atomic(
-				data->hw, IEEE80211_IFACE_ITER_NORMAL,
-				mac80211_hwsim_bcn_en_iter, &count);
-			wiphy_dbg(hw->wiphy, "  beaconing vifs remaining: %u",
-				  count);
-			if (count == 0) {
-				hrtimer_cancel(&link_data->beacon_timer);
-				link_data->beacon_int = 0;
-			}
-		}
-	}
-
-	if (changed & BSS_CHANGED_ERP_CTS_PROT) {
-		wiphy_dbg(hw->wiphy, "  ERP_CTS_PROT: %d\n",
-			  info->use_cts_prot);
-	}
-
-	if (changed & BSS_CHANGED_ERP_PREAMBLE) {
-		wiphy_dbg(hw->wiphy, "  ERP_PREAMBLE: %d\n",
-			  info->use_short_preamble);
-	}
-
-	if (changed & BSS_CHANGED_ERP_SLOT) {
-		wiphy_dbg(hw->wiphy, "  ERP_SLOT: %d\n", info->use_short_slot);
-	}
-
-	if (changed & BSS_CHANGED_HT) {
-		wiphy_dbg(hw->wiphy, "  HT: op_mode=0x%x\n",
-			  info->ht_operation_mode);
-	}
-
-	if (changed & BSS_CHANGED_BASIC_RATES) {
-		wiphy_dbg(hw->wiphy, "  BASIC_RATES: 0x%llx\n",
-			  (unsigned long long) info->basic_rates);
-	}
-
-	if (changed & BSS_CHANGED_TXPOWER)
-		wiphy_dbg(hw->wiphy, "  TX Power: %d dBm\n", info->txpower);
-}
-
-static void
-mac80211_hwsim_sta_rc_update(struct ieee80211_hw *hw,
-			     struct ieee80211_vif *vif,
-			     struct ieee80211_sta *sta,
-			     u32 changed)
-{
-	struct mac80211_hwsim_data *data = hw->priv;
-	u32 bw = U32_MAX;
-	int link_id;
-
-	rcu_read_lock();
-	for (link_id = 0;
-	     link_id < ARRAY_SIZE(vif->link_conf);
-	     link_id++) {
-		enum nl80211_chan_width confbw = NL80211_CHAN_WIDTH_20_NOHT;
-		struct ieee80211_bss_conf *vif_conf;
-		struct ieee80211_link_sta *link_sta;
-
-		link_sta = rcu_dereference(sta->link[link_id]);
-
-		if (!link_sta)
-			continue;
-
-		switch (link_sta->bandwidth) {
-#define C(_bw) case IEEE80211_STA_RX_BW_##_bw: bw = _bw; break
-		C(20);
-		C(40);
-		C(80);
-		C(160);
-		C(320);
-#undef C
-		}
-
-		if (!data->use_chanctx) {
-			confbw = data->bw;
-		} else {
-			struct ieee80211_chanctx_conf *chanctx_conf;
-
-			vif_conf = rcu_dereference(vif->link_conf[link_id]);
-			if (WARN_ON(!vif_conf))
-				continue;
-
-			chanctx_conf = rcu_dereference(vif_conf->chanctx_conf);
-
-			if (!WARN_ON(!chanctx_conf))
-				confbw = chanctx_conf->def.width;
-		}
-
-		WARN(bw > hwsim_get_chanwidth(confbw),
-		     "intf %pM [link=%d]: bad STA %pM bandwidth %d MHz (%d) > channel config %d MHz (%d)\n",
-		     vif->addr, link_id, sta->addr, bw, sta->deflink.bandwidth,
-		     hwsim_get_chanwidth(data->bw), data->bw);
-
-
-	}
-	rcu_read_unlock();
-
-
-}
-
-static int mac80211_hwsim_sta_add(struct ieee80211_hw *hw,
-				  struct ieee80211_vif *vif,
-				  struct ieee80211_sta *sta)
-{
-	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-
-	hwsim_check_magic(vif);
-	hwsim_set_sta_magic(sta);
-	mac80211_hwsim_sta_rc_update(hw, vif, sta, 0);
-
-	if (sta->valid_links) {
-		WARN(hweight16(sta->valid_links) > 1,
-		     "expect to add STA with single link, have 0x%x\n",
-		     sta->valid_links);
-		sp->active_links_rx = sta->valid_links;
-	}
-
-	return 0;
-}
-
-static int mac80211_hwsim_sta_remove(struct ieee80211_hw *hw,
-				     struct ieee80211_vif *vif,
-				     struct ieee80211_sta *sta)
-{
-	hwsim_check_magic(vif);
-	hwsim_clear_sta_magic(sta);
-
-	return 0;
-}
-
-static int mac80211_hwsim_sta_state(struct ieee80211_hw *hw,
-				    struct ieee80211_vif *vif,
-				    struct ieee80211_sta *sta,
-				    enum ieee80211_sta_state old_state,
-				    enum ieee80211_sta_state new_state)
-{
-	if (new_state == IEEE80211_STA_NOTEXIST)
-		return mac80211_hwsim_sta_remove(hw, vif, sta);
-
-	if (old_state == IEEE80211_STA_NOTEXIST)
-		return mac80211_hwsim_sta_add(hw, vif, sta);
-
-	/*
-	 * when client is authorized (AP station marked as such),
-	 * enable all links
-	 */
-	if (vif->type == NL80211_IFTYPE_STATION &&
-	    new_state == IEEE80211_STA_AUTHORIZED && !sta->tdls)
-		ieee80211_set_active_links_async(vif, vif->valid_links);
-
-	return 0;
-}
-
-static void mac80211_hwsim_sta_notify(struct ieee80211_hw *hw,
-				      struct ieee80211_vif *vif,
-				      enum sta_notify_cmd cmd,
-				      struct ieee80211_sta *sta)
-{
-	hwsim_check_magic(vif);
-
-	switch (cmd) {
-	case STA_NOTIFY_SLEEP:
-	case STA_NOTIFY_AWAKE:
-		/* TODO: make good use of these flags */
-		break;
-	default:
-		WARN(1, "Invalid sta notify: %d\n", cmd);
-		break;
-	}
-}
-
-static int mac80211_hwsim_set_tim(struct ieee80211_hw *hw,
-				  struct ieee80211_sta *sta,
-				  bool set)
-{
-	hwsim_check_sta_magic(sta);
-	return 0;
-}
-
-static int mac80211_hwsim_conf_tx(struct ieee80211_hw *hw,
-				  struct ieee80211_vif *vif,
-				  unsigned int link_id, u16 queue,
-				  const struct ieee80211_tx_queue_params *params)
-{
-	wiphy_dbg(hw->wiphy,
-		  "%s (queue=%d txop=%d cw_min=%d cw_max=%d aifs=%d)\n",
-		  __func__, queue,
-		  params->txop, params->cw_min,
-		  params->cw_max, params->aifs);
-	return 0;
-}
-
-static int mac80211_hwsim_get_survey(struct ieee80211_hw *hw, int idx,
-				     struct survey_info *survey)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-
-	if (idx < 0 || idx >= ARRAY_SIZE(hwsim->survey_data))
-		return -ENOENT;
-
-	mutex_lock(&hwsim->mutex);
-	survey->channel = hwsim->survey_data[idx].channel;
-	if (!survey->channel) {
-		mutex_unlock(&hwsim->mutex);
-		return -ENOENT;
-	}
-
-	/*
-	 * Magically conjured dummy values --- this is only ok for simulated hardware.
-	 *
-	 * A real driver which cannot determine real values noise MUST NOT
-	 * report any, especially not a magically conjured ones :-)
-	 */
-	survey->filled = SURVEY_INFO_NOISE_DBM |
-			 SURVEY_INFO_TIME |
-			 SURVEY_INFO_TIME_BUSY;
-	survey->noise = -92;
-	survey->time =
-		jiffies_to_msecs(hwsim->survey_data[idx].end -
-				 hwsim->survey_data[idx].start);
-	/* report 12.5% of channel time is used */
-	survey->time_busy = survey->time/8;
-	mutex_unlock(&hwsim->mutex);
-
-	return 0;
-}
-
-#ifdef CONFIG_NL80211_TESTMODE
-/*
- * This section contains example code for using netlink
- * attributes with the testmode command in nl80211.
- */
-
-/* These enums need to be kept in sync with userspace */
-enum hwsim_testmode_attr {
-	__HWSIM_TM_ATTR_INVALID	= 0,
-	HWSIM_TM_ATTR_CMD	= 1,
-	HWSIM_TM_ATTR_PS	= 2,
-
-	/* keep last */
-	__HWSIM_TM_ATTR_AFTER_LAST,
-	HWSIM_TM_ATTR_MAX	= __HWSIM_TM_ATTR_AFTER_LAST - 1
-};
-
-enum hwsim_testmode_cmd {
-	HWSIM_TM_CMD_SET_PS		= 0,
-	HWSIM_TM_CMD_GET_PS		= 1,
-	HWSIM_TM_CMD_STOP_QUEUES	= 2,
-	HWSIM_TM_CMD_WAKE_QUEUES	= 3,
-};
-
-static const struct nla_policy hwsim_testmode_policy[HWSIM_TM_ATTR_MAX + 1] = {
-	[HWSIM_TM_ATTR_CMD] = { .type = NLA_U32 },
-	[HWSIM_TM_ATTR_PS] = { .type = NLA_U32 },
-};
-
-static int mac80211_hwsim_testmode_cmd(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       void *data, int len)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-	struct nlattr *tb[HWSIM_TM_ATTR_MAX + 1];
-	struct sk_buff *skb;
-	int err, ps;
-
-	err = nla_parse_deprecated(tb, HWSIM_TM_ATTR_MAX, data, len,
-				   hwsim_testmode_policy, NULL);
-	if (err)
-		return err;
-
-	if (!tb[HWSIM_TM_ATTR_CMD])
-		return -EINVAL;
-
-	switch (nla_get_u32(tb[HWSIM_TM_ATTR_CMD])) {
-	case HWSIM_TM_CMD_SET_PS:
-		if (!tb[HWSIM_TM_ATTR_PS])
-			return -EINVAL;
-		ps = nla_get_u32(tb[HWSIM_TM_ATTR_PS]);
-		return hwsim_fops_ps_write(hwsim, ps);
-	case HWSIM_TM_CMD_GET_PS:
-		skb = cfg80211_testmode_alloc_reply_skb(hw->wiphy,
-						nla_total_size(sizeof(u32)));
-		if (!skb)
-			return -ENOMEM;
-		if (nla_put_u32(skb, HWSIM_TM_ATTR_PS, hwsim->ps))
-			goto nla_put_failure;
-		return cfg80211_testmode_reply(skb);
-	case HWSIM_TM_CMD_STOP_QUEUES:
-		ieee80211_stop_queues(hw);
-		return 0;
-	case HWSIM_TM_CMD_WAKE_QUEUES:
-		ieee80211_wake_queues(hw);
-		return 0;
-	default:
-		return -EOPNOTSUPP;
-	}
-
- nla_put_failure:
-	kfree_skb(skb);
-	return -ENOBUFS;
-}
-#endif
-
-static int mac80211_hwsim_ampdu_action(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       struct ieee80211_ampdu_params *params)
-{
-	struct ieee80211_sta *sta = params->sta;
-	enum ieee80211_ampdu_mlme_action action = params->action;
-	u16 tid = params->tid;
-
-	switch (action) {
-	case IEEE80211_AMPDU_TX_START:
-		return IEEE80211_AMPDU_TX_START_IMMEDIATE;
-	case IEEE80211_AMPDU_TX_STOP_CONT:
-	case IEEE80211_AMPDU_TX_STOP_FLUSH:
-	case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
-		ieee80211_stop_tx_ba_cb_irqsafe(vif, sta->addr, tid);
-		break;
-	case IEEE80211_AMPDU_TX_OPERATIONAL:
-		break;
-	case IEEE80211_AMPDU_RX_START:
-	case IEEE80211_AMPDU_RX_STOP:
-		break;
-	default:
-		return -EOPNOTSUPP;
-	}
-
-	return 0;
-}
-
-static void mac80211_hwsim_flush(struct ieee80211_hw *hw,
-				 struct ieee80211_vif *vif,
-				 u32 queues, bool drop)
-{
-	/* Not implemented, queues only on kernel side */
-}
-
-static void hw_scan_work(struct work_struct *work)
-{
-	struct mac80211_hwsim_data *hwsim =
-		container_of(work, struct mac80211_hwsim_data, hw_scan.work);
-	struct cfg80211_scan_request *req = hwsim->hw_scan_request;
-	int dwell, i;
-
-	mutex_lock(&hwsim->mutex);
-	if (hwsim->scan_chan_idx >= req->n_channels) {
-		struct cfg80211_scan_info info = {
-			.aborted = false,
-		};
-
-		wiphy_dbg(hwsim->hw->wiphy, "hw scan complete\n");
-		ieee80211_scan_completed(hwsim->hw, &info);
-		hwsim->hw_scan_request = NULL;
-		hwsim->hw_scan_vif = NULL;
-		hwsim->tmp_chan = NULL;
-		mutex_unlock(&hwsim->mutex);
-		mac80211_hwsim_config_mac_nl(hwsim->hw, hwsim->scan_addr,
-					     false);
-		return;
-	}
-
-	wiphy_dbg(hwsim->hw->wiphy, "hw scan %d MHz\n",
-		  req->channels[hwsim->scan_chan_idx]->center_freq);
-
-	hwsim->tmp_chan = req->channels[hwsim->scan_chan_idx];
-	if (hwsim->tmp_chan->flags & (IEEE80211_CHAN_NO_IR |
-				      IEEE80211_CHAN_RADAR) ||
-	    !req->n_ssids) {
-		dwell = 120;
-	} else {
-		dwell = 30;
-		/* send probes */
-		for (i = 0; i < req->n_ssids; i++) {
-			struct sk_buff *probe;
-			struct ieee80211_mgmt *mgmt;
-
-			probe = ieee80211_probereq_get(hwsim->hw,
-						       hwsim->scan_addr,
-						       req->ssids[i].ssid,
-						       req->ssids[i].ssid_len,
-						       req->ie_len);
-			if (!probe)
-				continue;
-
-			mgmt = (struct ieee80211_mgmt *) probe->data;
-			memcpy(mgmt->da, req->bssid, ETH_ALEN);
-			memcpy(mgmt->bssid, req->bssid, ETH_ALEN);
-
-			if (req->ie_len)
-				skb_put_data(probe, req->ie, req->ie_len);
-
-			rcu_read_lock();
-			if (!ieee80211_tx_prepare_skb(hwsim->hw,
-						      hwsim->hw_scan_vif,
-						      probe,
-						      hwsim->tmp_chan->band,
-						      NULL)) {
-				rcu_read_unlock();
-				kfree_skb(probe);
-				continue;
-			}
-
-			local_bh_disable();
-			mac80211_hwsim_tx_frame(hwsim->hw, probe,
-						hwsim->tmp_chan);
-			rcu_read_unlock();
-			local_bh_enable();
-		}
-	}
-	ieee80211_queue_delayed_work(hwsim->hw, &hwsim->hw_scan,
-				     msecs_to_jiffies(dwell));
-	hwsim->survey_data[hwsim->scan_chan_idx].channel = hwsim->tmp_chan;
-	hwsim->survey_data[hwsim->scan_chan_idx].start = jiffies;
-	hwsim->survey_data[hwsim->scan_chan_idx].end =
-		jiffies + msecs_to_jiffies(dwell);
-	hwsim->scan_chan_idx++;
-	mutex_unlock(&hwsim->mutex);
-}
-
-static int mac80211_hwsim_hw_scan(struct ieee80211_hw *hw,
-				  struct ieee80211_vif *vif,
-				  struct ieee80211_scan_request *hw_req)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-	struct cfg80211_scan_request *req = &hw_req->req;
-
-	mutex_lock(&hwsim->mutex);
-	if (WARN_ON(hwsim->tmp_chan || hwsim->hw_scan_request)) {
-		mutex_unlock(&hwsim->mutex);
-		return -EBUSY;
-	}
-	hwsim->hw_scan_request = req;
-	hwsim->hw_scan_vif = vif;
-	hwsim->scan_chan_idx = 0;
-	if (req->flags & NL80211_SCAN_FLAG_RANDOM_ADDR)
-		get_random_mask_addr(hwsim->scan_addr,
-				     hw_req->req.mac_addr,
-				     hw_req->req.mac_addr_mask);
-	else
-		memcpy(hwsim->scan_addr, vif->addr, ETH_ALEN);
-	memset(hwsim->survey_data, 0, sizeof(hwsim->survey_data));
-	mutex_unlock(&hwsim->mutex);
-
-	mac80211_hwsim_config_mac_nl(hw, hwsim->scan_addr, true);
-	wiphy_dbg(hw->wiphy, "hwsim hw_scan request\n");
-
-	ieee80211_queue_delayed_work(hwsim->hw, &hwsim->hw_scan, 0);
-
-	return 0;
-}
-
-static void mac80211_hwsim_cancel_hw_scan(struct ieee80211_hw *hw,
-					  struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-	struct cfg80211_scan_info info = {
-		.aborted = true,
-	};
-
-	wiphy_dbg(hw->wiphy, "hwsim cancel_hw_scan\n");
-
-	cancel_delayed_work_sync(&hwsim->hw_scan);
-
-	mutex_lock(&hwsim->mutex);
-	ieee80211_scan_completed(hwsim->hw, &info);
-	hwsim->tmp_chan = NULL;
-	hwsim->hw_scan_request = NULL;
-	hwsim->hw_scan_vif = NULL;
-	mutex_unlock(&hwsim->mutex);
-}
-
-static void mac80211_hwsim_sw_scan(struct ieee80211_hw *hw,
-				   struct ieee80211_vif *vif,
-				   const u8 *mac_addr)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-
-	mutex_lock(&hwsim->mutex);
-
-	if (hwsim->scanning) {
-		pr_debug("two hwsim sw_scans detected!\n");
-		goto out;
-	}
-
-	pr_debug("hwsim sw_scan request, prepping stuff\n");
-
-	memcpy(hwsim->scan_addr, mac_addr, ETH_ALEN);
-	mac80211_hwsim_config_mac_nl(hw, hwsim->scan_addr, true);
-	hwsim->scanning = true;
-	memset(hwsim->survey_data, 0, sizeof(hwsim->survey_data));
-
-out:
-	mutex_unlock(&hwsim->mutex);
-}
-
-static void mac80211_hwsim_sw_scan_complete(struct ieee80211_hw *hw,
-					    struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-
-	mutex_lock(&hwsim->mutex);
-
-	pr_debug("hwsim sw_scan_complete\n");
-	hwsim->scanning = false;
-	mac80211_hwsim_config_mac_nl(hw, hwsim->scan_addr, false);
-	eth_zero_addr(hwsim->scan_addr);
-
-	mutex_unlock(&hwsim->mutex);
-}
-
-static void hw_roc_start(struct work_struct *work)
-{
-	struct mac80211_hwsim_data *hwsim =
-		container_of(work, struct mac80211_hwsim_data, roc_start.work);
-
-	mutex_lock(&hwsim->mutex);
-
-	wiphy_dbg(hwsim->hw->wiphy, "hwsim ROC begins\n");
-	hwsim->tmp_chan = hwsim->roc_chan;
-	ieee80211_ready_on_channel(hwsim->hw);
-
-	ieee80211_queue_delayed_work(hwsim->hw, &hwsim->roc_done,
-				     msecs_to_jiffies(hwsim->roc_duration));
-
-	mutex_unlock(&hwsim->mutex);
-}
-
-static void hw_roc_done(struct work_struct *work)
-{
-	struct mac80211_hwsim_data *hwsim =
-		container_of(work, struct mac80211_hwsim_data, roc_done.work);
-
-	mutex_lock(&hwsim->mutex);
-	ieee80211_remain_on_channel_expired(hwsim->hw);
-	hwsim->tmp_chan = NULL;
-	mutex_unlock(&hwsim->mutex);
-
-	wiphy_dbg(hwsim->hw->wiphy, "hwsim ROC expired\n");
-}
-
-static int mac80211_hwsim_roc(struct ieee80211_hw *hw,
-			      struct ieee80211_vif *vif,
-			      struct ieee80211_channel *chan,
-			      int duration,
-			      enum ieee80211_roc_type type)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-
-	mutex_lock(&hwsim->mutex);
-	if (WARN_ON(hwsim->tmp_chan || hwsim->hw_scan_request)) {
-		mutex_unlock(&hwsim->mutex);
-		return -EBUSY;
-	}
-
-	hwsim->roc_chan = chan;
-	hwsim->roc_duration = duration;
-	mutex_unlock(&hwsim->mutex);
-
-	wiphy_dbg(hw->wiphy, "hwsim ROC (%d MHz, %d ms)\n",
-		  chan->center_freq, duration);
-	ieee80211_queue_delayed_work(hw, &hwsim->roc_start, HZ/50);
-
-	return 0;
-}
-
-static int mac80211_hwsim_croc(struct ieee80211_hw *hw,
-			       struct ieee80211_vif *vif)
-{
-	struct mac80211_hwsim_data *hwsim = hw->priv;
-
-	cancel_delayed_work_sync(&hwsim->roc_start);
-	cancel_delayed_work_sync(&hwsim->roc_done);
-
-	mutex_lock(&hwsim->mutex);
-	hwsim->tmp_chan = NULL;
-	mutex_unlock(&hwsim->mutex);
-
-	wiphy_dbg(hw->wiphy, "hwsim ROC canceled\n");
-
-	return 0;
-}
-
-static int mac80211_hwsim_add_chanctx(struct ieee80211_hw *hw,
-				      struct ieee80211_chanctx_conf *ctx)
-{
-	hwsim_set_chanctx_magic(ctx);
-	wiphy_dbg(hw->wiphy,
-		  "add channel context control: %d MHz/width: %d/cfreqs:%d/%d MHz\n",
-		  ctx->def.chan->center_freq, ctx->def.width,
-		  ctx->def.center_freq1, ctx->def.center_freq2);
-	return 0;
-}
-
-static void mac80211_hwsim_remove_chanctx(struct ieee80211_hw *hw,
-					  struct ieee80211_chanctx_conf *ctx)
-{
-	wiphy_dbg(hw->wiphy,
-		  "remove channel context control: %d MHz/width: %d/cfreqs:%d/%d MHz\n",
-		  ctx->def.chan->center_freq, ctx->def.width,
-		  ctx->def.center_freq1, ctx->def.center_freq2);
-	hwsim_check_chanctx_magic(ctx);
-	hwsim_clear_chanctx_magic(ctx);
-}
-
-static void mac80211_hwsim_change_chanctx(struct ieee80211_hw *hw,
-					  struct ieee80211_chanctx_conf *ctx,
-					  u32 changed)
-{
-	hwsim_check_chanctx_magic(ctx);
-	wiphy_dbg(hw->wiphy,
-		  "change channel context control: %d MHz/width: %d/cfreqs:%d/%d MHz\n",
-		  ctx->def.chan->center_freq, ctx->def.width,
-		  ctx->def.center_freq1, ctx->def.center_freq2);
-}
-
-static int mac80211_hwsim_assign_vif_chanctx(struct ieee80211_hw *hw,
-					     struct ieee80211_vif *vif,
-					     struct ieee80211_bss_conf *link_conf,
-					     struct ieee80211_chanctx_conf *ctx)
-{
-	hwsim_check_magic(vif);
-	hwsim_check_chanctx_magic(ctx);
-
-	/* if we activate a link while already associated wake it up */
-	if (vif->type == NL80211_IFTYPE_STATION && vif->cfg.assoc) {
-		struct sk_buff *skb;
-
-		skb = ieee80211_nullfunc_get(hw, vif, link_conf->link_id, true);
-		if (skb) {
-			local_bh_disable();
-			mac80211_hwsim_tx_frame(hw, skb, ctx->def.chan);
-			local_bh_enable();
-		}
-	}
-
-	return 0;
-}
-
-static void mac80211_hwsim_unassign_vif_chanctx(struct ieee80211_hw *hw,
-						struct ieee80211_vif *vif,
-						struct ieee80211_bss_conf *link_conf,
-						struct ieee80211_chanctx_conf *ctx)
-{
-	hwsim_check_magic(vif);
-	hwsim_check_chanctx_magic(ctx);
-
-	/* if we deactivate a link while associated suspend it first */
-	if (vif->type == NL80211_IFTYPE_STATION && vif->cfg.assoc) {
-		struct sk_buff *skb;
-
-		skb = ieee80211_nullfunc_get(hw, vif, link_conf->link_id, true);
-		if (skb) {
-			struct ieee80211_hdr *hdr = (void *)skb->data;
-
-			hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
-
-			local_bh_disable();
-			mac80211_hwsim_tx_frame(hw, skb, ctx->def.chan);
-			local_bh_enable();
-		}
-	}
-}
-
-static const char mac80211_hwsim_gstrings_stats[][ETH_GSTRING_LEN] = {
-	"tx_pkts_nic",
-	"tx_bytes_nic",
-	"rx_pkts_nic",
-	"rx_bytes_nic",
-	"d_tx_dropped",
-	"d_tx_failed",
-	"d_ps_mode",
-	"d_group",
-};
-
-#define MAC80211_HWSIM_SSTATS_LEN ARRAY_SIZE(mac80211_hwsim_gstrings_stats)
-
-static void mac80211_hwsim_get_et_strings(struct ieee80211_hw *hw,
-					  struct ieee80211_vif *vif,
-					  u32 sset, u8 *data)
-{
-	if (sset == ETH_SS_STATS)
-		memcpy(data, *mac80211_hwsim_gstrings_stats,
-		       sizeof(mac80211_hwsim_gstrings_stats));
-}
-
-static int mac80211_hwsim_get_et_sset_count(struct ieee80211_hw *hw,
-					    struct ieee80211_vif *vif, int sset)
-{
-	if (sset == ETH_SS_STATS)
-		return MAC80211_HWSIM_SSTATS_LEN;
-	return 0;
-}
-
-static void mac80211_hwsim_get_et_stats(struct ieee80211_hw *hw,
-					struct ieee80211_vif *vif,
-					struct ethtool_stats *stats, u64 *data)
-{
-	struct mac80211_hwsim_data *ar = hw->priv;
-	int i = 0;
-
-	data[i++] = ar->tx_pkts;
-	data[i++] = ar->tx_bytes;
-	data[i++] = ar->rx_pkts;
-	data[i++] = ar->rx_bytes;
-	data[i++] = ar->tx_dropped;
-	data[i++] = ar->tx_failed;
-	data[i++] = ar->ps;
-	data[i++] = ar->group;
-
-	WARN_ON(i != MAC80211_HWSIM_SSTATS_LEN);
-}
-
-static int mac80211_hwsim_tx_last_beacon(struct ieee80211_hw *hw)
-{
-	return 1;
-}
-
-static int mac80211_hwsim_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
-{
-	return -EOPNOTSUPP;
-}
-
-static int mac80211_hwsim_change_vif_links(struct ieee80211_hw *hw,
-					   struct ieee80211_vif *vif,
-					   u16 old_links, u16 new_links,
-					   struct ieee80211_bss_conf *old[IEEE80211_MLD_MAX_NUM_LINKS])
-{
-	unsigned long rem = old_links & ~new_links;
-	unsigned long add = new_links & ~old_links;
-	int i;
-
-	if (!old_links)
-		rem |= BIT(0);
-	if (!new_links)
-		add |= BIT(0);
-
-	for_each_set_bit(i, &rem, IEEE80211_MLD_MAX_NUM_LINKS)
-		mac80211_hwsim_config_mac_nl(hw, old[i]->addr, false);
-
-	for_each_set_bit(i, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
-		struct ieee80211_bss_conf *link_conf;
-
-		link_conf = link_conf_dereference_protected(vif, i);
-		if (WARN_ON(!link_conf))
-			continue;
-
-		mac80211_hwsim_config_mac_nl(hw, link_conf->addr, true);
-	}
-
-	return 0;
-}
-
-static int mac80211_hwsim_change_sta_links(struct ieee80211_hw *hw,
-					   struct ieee80211_vif *vif,
-					   struct ieee80211_sta *sta,
-					   u16 old_links, u16 new_links)
-{
-	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
-
-	hwsim_check_sta_magic(sta);
-
-	if (vif->type == NL80211_IFTYPE_STATION)
-		sp->active_links_rx = new_links;
-
-	return 0;
-}
-
-#define HWSIM_COMMON_OPS					\
-	.tx = mac80211_hwsim_tx,				\
-	.wake_tx_queue = ieee80211_handle_wake_tx_queue,	\
-	.start = mac80211_hwsim_start,				\
-	.stop = mac80211_hwsim_stop,				\
-	.add_interface = mac80211_hwsim_add_interface,		\
-	.change_interface = mac80211_hwsim_change_interface,	\
-	.remove_interface = mac80211_hwsim_remove_interface,	\
-	.config = mac80211_hwsim_config,			\
-	.configure_filter = mac80211_hwsim_configure_filter,	\
-	.vif_cfg_changed = mac80211_hwsim_vif_info_changed,	\
-	.link_info_changed = mac80211_hwsim_link_info_changed,  \
-	.tx_last_beacon = mac80211_hwsim_tx_last_beacon,	\
-	.sta_notify = mac80211_hwsim_sta_notify,		\
-	.sta_rc_update = mac80211_hwsim_sta_rc_update,		\
-	.conf_tx = mac80211_hwsim_conf_tx,			\
-	.get_survey = mac80211_hwsim_get_survey,		\
-	CFG80211_TESTMODE_CMD(mac80211_hwsim_testmode_cmd)	\
-	.ampdu_action = mac80211_hwsim_ampdu_action,		\
-	.flush = mac80211_hwsim_flush,				\
-	.get_et_sset_count = mac80211_hwsim_get_et_sset_count,	\
-	.get_et_stats = mac80211_hwsim_get_et_stats,		\
-	.get_et_strings = mac80211_hwsim_get_et_strings,
-
-#define HWSIM_NON_MLO_OPS					\
-	.sta_add = mac80211_hwsim_sta_add,			\
-	.sta_remove = mac80211_hwsim_sta_remove,		\
-	.set_tim = mac80211_hwsim_set_tim,			\
-	.get_tsf = mac80211_hwsim_get_tsf,			\
-	.set_tsf = mac80211_hwsim_set_tsf,
-
-static const struct ieee80211_ops mac80211_hwsim_ops = {
-	HWSIM_COMMON_OPS
-	HWSIM_NON_MLO_OPS
-	.sw_scan_start = mac80211_hwsim_sw_scan,
-	.sw_scan_complete = mac80211_hwsim_sw_scan_complete,
-};
-
-#define HWSIM_CHANCTX_OPS					\
-	.hw_scan = mac80211_hwsim_hw_scan,			\
-	.cancel_hw_scan = mac80211_hwsim_cancel_hw_scan,	\
-	.remain_on_channel = mac80211_hwsim_roc,		\
-	.cancel_remain_on_channel = mac80211_hwsim_croc,	\
-	.add_chanctx = mac80211_hwsim_add_chanctx,		\
-	.remove_chanctx = mac80211_hwsim_remove_chanctx,	\
-	.change_chanctx = mac80211_hwsim_change_chanctx,	\
-	.assign_vif_chanctx = mac80211_hwsim_assign_vif_chanctx,\
-	.unassign_vif_chanctx = mac80211_hwsim_unassign_vif_chanctx,
-
-static const struct ieee80211_ops mac80211_hwsim_mchan_ops = {
-	HWSIM_COMMON_OPS
-	HWSIM_NON_MLO_OPS
-	HWSIM_CHANCTX_OPS
-};
-
-static const struct ieee80211_ops mac80211_hwsim_mlo_ops = {
-	HWSIM_COMMON_OPS
-	HWSIM_CHANCTX_OPS
-	.set_rts_threshold = mac80211_hwsim_set_rts_threshold,
-	.change_vif_links = mac80211_hwsim_change_vif_links,
-	.change_sta_links = mac80211_hwsim_change_sta_links,
-	.sta_state = mac80211_hwsim_sta_state,
-};
-
-struct hwsim_new_radio_params {
-	unsigned int channels;
-	const char *reg_alpha2;
-	const struct ieee80211_regdomain *regd;
-	bool reg_strict;
-	bool p2p_device;
-	bool use_chanctx;
-	bool destroy_on_close;
-	const char *hwname;
-	bool no_vif;
-	const u8 *perm_addr;
-	u32 iftypes;
-	u32 *ciphers;
-	u8 n_ciphers;
-	bool mlo;
-};
-
-static void hwsim_mcast_config_msg(struct sk_buff *mcast_skb,
-				   struct genl_info *info)
-{
-	if (info)
-		genl_notify(&hwsim_genl_family, mcast_skb, info,
-			    HWSIM_MCGRP_CONFIG, GFP_KERNEL);
-	else
-		genlmsg_multicast(&hwsim_genl_family, mcast_skb, 0,
-				  HWSIM_MCGRP_CONFIG, GFP_KERNEL);
-}
-
-static int append_radio_msg(struct sk_buff *skb, int id,
-			    struct hwsim_new_radio_params *param)
-{
-	int ret;
-
-	ret = nla_put_u32(skb, HWSIM_ATTR_RADIO_ID, id);
-	if (ret < 0)
-		return ret;
-
-	if (param->channels) {
-		ret = nla_put_u32(skb, HWSIM_ATTR_CHANNELS, param->channels);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (param->reg_alpha2) {
-		ret = nla_put(skb, HWSIM_ATTR_REG_HINT_ALPHA2, 2,
-			      param->reg_alpha2);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (param->regd) {
-		int i;
-
-		for (i = 0; i < ARRAY_SIZE(hwsim_world_regdom_custom); i++) {
-			if (hwsim_world_regdom_custom[i] != param->regd)
-				continue;
-
-			ret = nla_put_u32(skb, HWSIM_ATTR_REG_CUSTOM_REG, i);
-			if (ret < 0)
-				return ret;
-			break;
-		}
-	}
-
-	if (param->reg_strict) {
-		ret = nla_put_flag(skb, HWSIM_ATTR_REG_STRICT_REG);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (param->p2p_device) {
-		ret = nla_put_flag(skb, HWSIM_ATTR_SUPPORT_P2P_DEVICE);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (param->use_chanctx) {
-		ret = nla_put_flag(skb, HWSIM_ATTR_USE_CHANCTX);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (param->hwname) {
-		ret = nla_put(skb, HWSIM_ATTR_RADIO_NAME,
-			      strlen(param->hwname), param->hwname);
-		if (ret < 0)
-			return ret;
-	}
-
-	return 0;
-}
-
-static void hwsim_mcast_new_radio(int id, struct genl_info *info,
-				  struct hwsim_new_radio_params *param)
-{
-	struct sk_buff *mcast_skb;
-	void *data;
-
-	mcast_skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
-	if (!mcast_skb)
-		return;
-
-	data = genlmsg_put(mcast_skb, 0, 0, &hwsim_genl_family, 0,
-			   HWSIM_CMD_NEW_RADIO);
-	if (!data)
-		goto out_err;
-
-	if (append_radio_msg(mcast_skb, id, param) < 0)
-		goto out_err;
-
-	genlmsg_end(mcast_skb, data);
-
-	hwsim_mcast_config_msg(mcast_skb, info);
-	return;
-
-out_err:
-	nlmsg_free(mcast_skb);
-}
-
-static const struct ieee80211_sband_iftype_data sband_capa_2ghz[] = {
-	{
-		.types_mask = BIT(NL80211_IFTYPE_STATION),
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_BSR |
-					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] =
-					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
-					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
-					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xffff),
-				.tx_mcs_160 = cpu_to_le16(0xffff),
-				.rx_mcs_80p80 = cpu_to_le16(0xffff),
-				.tx_mcs_80p80 = cpu_to_le16(0xffff),
-			},
-		},
-		.eht_cap = {
-			.has_eht = true,
-			.eht_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
-					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
-					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
-				.phy_cap_info[0] =
-					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
-					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
-					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE,
-				.phy_cap_info[3] =
-					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
-				.phy_cap_info[4] =
-					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
-					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
-					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
-					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
-				.phy_cap_info[5] =
-					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
-					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
-					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
-				.phy_cap_info[6] =
-					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
-					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
-				.phy_cap_info[7] =
-					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW,
-			},
-
-			/* For all MCS and bandwidth, set 8 NSS for both Tx and
-			 * Rx
-			 */
-			.eht_mcs_nss_supp = {
-				/*
-				 * Since B0, B1, B2 and B3 are not set in
-				 * the supported channel width set field in the
-				 * HE PHY capabilities information field the
-				 * device is a 20MHz only device on 2.4GHz band.
-				 */
-				.only_20mhz = {
-					.rx_tx_mcs7_max_nss = 0x88,
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-			},
-			/* PPE threshold information is not supported */
-		},
-	},
-	{
-		.types_mask = BIT(NL80211_IFTYPE_AP),
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_BSR |
-					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] =
-					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
-					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
-					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xffff),
-				.tx_mcs_160 = cpu_to_le16(0xffff),
-				.rx_mcs_80p80 = cpu_to_le16(0xffff),
-				.tx_mcs_80p80 = cpu_to_le16(0xffff),
-			},
-		},
-		.eht_cap = {
-			.has_eht = true,
-			.eht_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
-					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
-					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
-				.phy_cap_info[0] =
-					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
-					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
-					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE,
-				.phy_cap_info[3] =
-					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
-				.phy_cap_info[4] =
-					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
-					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
-					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
-					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
-				.phy_cap_info[5] =
-					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
-					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
-					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
-				.phy_cap_info[6] =
-					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
-					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
-				.phy_cap_info[7] =
-					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW,
-			},
-
-			/* For all MCS and bandwidth, set 8 NSS for both Tx and
-			 * Rx
-			 */
-			.eht_mcs_nss_supp = {
-				/*
-				 * Since B0, B1, B2 and B3 are not set in
-				 * the supported channel width set field in the
-				 * HE PHY capabilities information field the
-				 * device is a 20MHz only device on 2.4GHz band.
-				 */
-				.only_20mhz = {
-					.rx_tx_mcs7_max_nss = 0x88,
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-			},
-			/* PPE threshold information is not supported */
-		},
-	},
-#ifdef CONFIG_MAC80211_MESH
-	{
-		.types_mask = BIT(NL80211_IFTYPE_MESH_POINT),
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] = 0,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xffff),
-				.tx_mcs_160 = cpu_to_le16(0xffff),
-				.rx_mcs_80p80 = cpu_to_le16(0xffff),
-				.tx_mcs_80p80 = cpu_to_le16(0xffff),
-			},
-		},
-	},
-#endif
-};
-
-static const struct ieee80211_sband_iftype_data sband_capa_5ghz[] = {
-	{
-		/* TODO: should we support other types, e.g., P2P? */
-		.types_mask = BIT(NL80211_IFTYPE_STATION),
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_BSR |
-					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[0] =
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] =
-					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
-					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
-					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xfffa),
-				.tx_mcs_160 = cpu_to_le16(0xfffa),
-				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
-			},
-		},
-		.eht_cap = {
-			.has_eht = true,
-			.eht_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
-					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
-					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
-				.phy_cap_info[0] =
-					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
-					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
-					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
-					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
-				.phy_cap_info[1] =
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK,
-				.phy_cap_info[2] =
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK,
-				.phy_cap_info[3] =
-					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
-				.phy_cap_info[4] =
-					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
-					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
-					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
-					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
-				.phy_cap_info[5] =
-					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
-					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
-					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
-				.phy_cap_info[6] =
-					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
-					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
-				.phy_cap_info[7] =
-					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ,
-			},
-
-			/* For all MCS and bandwidth, set 8 NSS for both Tx and
-			 * Rx
-			 */
-			.eht_mcs_nss_supp = {
-				/*
-				 * As B1 and B2 are set in the supported
-				 * channel width set field in the HE PHY
-				 * capabilities information field include all
-				 * the following MCS/NSS.
-				 */
-				.bw._80 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-				.bw._160 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-			},
-			/* PPE threshold information is not supported */
-		},
-	},
-	{
-		.types_mask = BIT(NL80211_IFTYPE_AP),
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_BSR |
-					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[0] =
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] =
-					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
-					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
-					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xfffa),
-				.tx_mcs_160 = cpu_to_le16(0xfffa),
-				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
-			},
-		},
-		.eht_cap = {
-			.has_eht = true,
-			.eht_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
-					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
-					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
-				.phy_cap_info[0] =
-					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
-					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
-					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
-					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
-				.phy_cap_info[1] =
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK,
-				.phy_cap_info[2] =
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK,
-				.phy_cap_info[3] =
-					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
-				.phy_cap_info[4] =
-					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
-					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
-					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
-					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
-				.phy_cap_info[5] =
-					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
-					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
-					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
-				.phy_cap_info[6] =
-					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
-					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
-				.phy_cap_info[7] =
-					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ,
-			},
-
-			/* For all MCS and bandwidth, set 8 NSS for both Tx and
-			 * Rx
-			 */
-			.eht_mcs_nss_supp = {
-				/*
-				 * As B1 and B2 are set in the supported
-				 * channel width set field in the HE PHY
-				 * capabilities information field include all
-				 * the following MCS/NSS.
-				 */
-				.bw._80 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-				.bw._160 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-			},
-			/* PPE threshold information is not supported */
-		},
-	},
-#ifdef CONFIG_MAC80211_MESH
-	{
-		/* TODO: should we support other types, e.g., IBSS?*/
-		.types_mask = BIT(NL80211_IFTYPE_MESH_POINT),
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[0] =
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] = 0,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xfffa),
-				.tx_mcs_160 = cpu_to_le16(0xfffa),
-				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
-			},
-		},
-	},
-#endif
-};
-
-static const struct ieee80211_sband_iftype_data sband_capa_6ghz[] = {
-	{
-		/* TODO: should we support other types, e.g., P2P? */
-		.types_mask = BIT(NL80211_IFTYPE_STATION),
-		.he_6ghz_capa = {
-			.capa = cpu_to_le16(IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START |
-					    IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP |
-					    IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN |
-					    IEEE80211_HE_6GHZ_CAP_SM_PS |
-					    IEEE80211_HE_6GHZ_CAP_RD_RESPONDER |
-					    IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
-					    IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS),
-		},
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_BSR |
-					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[0] =
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] =
-					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
-					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
-					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xfffa),
-				.tx_mcs_160 = cpu_to_le16(0xfffa),
-				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
-			},
-		},
-		.eht_cap = {
-			.has_eht = true,
-			.eht_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
-					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
-					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
-				.phy_cap_info[0] =
-					IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ |
-					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
-					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
-					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
-					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
-				.phy_cap_info[1] =
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK,
-				.phy_cap_info[2] =
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK,
-				.phy_cap_info[3] =
-					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
-				.phy_cap_info[4] =
-					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
-					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
-					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
-					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
-				.phy_cap_info[5] =
-					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
-					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
-					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
-				.phy_cap_info[6] =
-					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
-					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK |
-					IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP,
-				.phy_cap_info[7] =
-					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ,
-			},
-
-			/* For all MCS and bandwidth, set 8 NSS for both Tx and
-			 * Rx
-			 */
-			.eht_mcs_nss_supp = {
-				/*
-				 * As B1 and B2 are set in the supported
-				 * channel width set field in the HE PHY
-				 * capabilities information field and 320MHz in
-				 * 6GHz is supported include all the following
-				 * MCS/NSS.
-				 */
-				.bw._80 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-				.bw._160 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-				.bw._320 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-			},
-			/* PPE threshold information is not supported */
-		},
-	},
-	{
-		.types_mask = BIT(NL80211_IFTYPE_AP),
-		.he_6ghz_capa = {
-			.capa = cpu_to_le16(IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START |
-					    IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP |
-					    IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN |
-					    IEEE80211_HE_6GHZ_CAP_SM_PS |
-					    IEEE80211_HE_6GHZ_CAP_RD_RESPONDER |
-					    IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
-					    IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS),
-		},
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_BSR |
-					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[0] =
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] =
-					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
-					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
-					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
-					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xfffa),
-				.tx_mcs_160 = cpu_to_le16(0xfffa),
-				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
-			},
-		},
-		.eht_cap = {
-			.has_eht = true,
-			.eht_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
-					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
-					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
-				.phy_cap_info[0] =
-					IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ |
-					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
-					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
-					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
-					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
-					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
-				.phy_cap_info[1] =
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK,
-				.phy_cap_info[2] =
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK |
-					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK,
-				.phy_cap_info[3] =
-					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
-					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
-				.phy_cap_info[4] =
-					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
-					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
-					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
-					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
-					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
-				.phy_cap_info[5] =
-					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
-					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
-					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
-					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
-					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
-				.phy_cap_info[6] =
-					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
-					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK |
-					IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP,
-				.phy_cap_info[7] =
-					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
-					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ |
-					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ,
-			},
-
-			/* For all MCS and bandwidth, set 8 NSS for both Tx and
-			 * Rx
-			 */
-			.eht_mcs_nss_supp = {
-				/*
-				 * As B1 and B2 are set in the supported
-				 * channel width set field in the HE PHY
-				 * capabilities information field and 320MHz in
-				 * 6GHz is supported include all the following
-				 * MCS/NSS.
-				 */
-				.bw._80 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-				.bw._160 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-				.bw._320 = {
-					.rx_tx_mcs9_max_nss = 0x88,
-					.rx_tx_mcs11_max_nss = 0x88,
-					.rx_tx_mcs13_max_nss = 0x88,
-				},
-			},
-			/* PPE threshold information is not supported */
-		},
-	},
-#ifdef CONFIG_MAC80211_MESH
-	{
-		/* TODO: should we support other types, e.g., IBSS?*/
-		.types_mask = BIT(NL80211_IFTYPE_MESH_POINT),
-		.he_6ghz_capa = {
-			.capa = cpu_to_le16(IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START |
-					    IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP |
-					    IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN |
-					    IEEE80211_HE_6GHZ_CAP_SM_PS |
-					    IEEE80211_HE_6GHZ_CAP_RD_RESPONDER |
-					    IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
-					    IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS),
-		},
-		.he_cap = {
-			.has_he = true,
-			.he_cap_elem = {
-				.mac_cap_info[0] =
-					IEEE80211_HE_MAC_CAP0_HTC_HE,
-				.mac_cap_info[1] =
-					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
-				.mac_cap_info[2] =
-					IEEE80211_HE_MAC_CAP2_ACK_EN,
-				.mac_cap_info[3] =
-					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
-				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
-				.phy_cap_info[0] =
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
-					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
-				.phy_cap_info[1] =
-					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
-					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
-					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
-					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
-				.phy_cap_info[2] = 0,
-
-				/* Leave all the other PHY capability bytes
-				 * unset, as DCM, beam forming, RU and PPE
-				 * threshold information are not supported
-				 */
-			},
-			.he_mcs_nss_supp = {
-				.rx_mcs_80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80 = cpu_to_le16(0xfffa),
-				.rx_mcs_160 = cpu_to_le16(0xfffa),
-				.tx_mcs_160 = cpu_to_le16(0xfffa),
-				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
-				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
-			},
-		},
-	},
-#endif
-};
-
-static void mac80211_hwsim_sband_capab(struct ieee80211_supported_band *sband)
-{
-	u16 n_iftype_data;
-
-	if (sband->band == NL80211_BAND_2GHZ) {
-		n_iftype_data = ARRAY_SIZE(sband_capa_2ghz);
-		sband->iftype_data =
-			(struct ieee80211_sband_iftype_data *)sband_capa_2ghz;
-	} else if (sband->band == NL80211_BAND_5GHZ) {
-		n_iftype_data = ARRAY_SIZE(sband_capa_5ghz);
-		sband->iftype_data =
-			(struct ieee80211_sband_iftype_data *)sband_capa_5ghz;
-	} else if (sband->band == NL80211_BAND_6GHZ) {
-		n_iftype_data = ARRAY_SIZE(sband_capa_6ghz);
-		sband->iftype_data =
-			(struct ieee80211_sband_iftype_data *)sband_capa_6ghz;
-	} else {
-		return;
-	}
-
-	sband->n_iftype_data = n_iftype_data;
-}
-
-#ifdef CONFIG_MAC80211_MESH
-#define HWSIM_MESH_BIT BIT(NL80211_IFTYPE_MESH_POINT)
-#else
-#define HWSIM_MESH_BIT 0
-#endif
-
-#define HWSIM_DEFAULT_IF_LIMIT \
-	(BIT(NL80211_IFTYPE_STATION) | \
-	 BIT(NL80211_IFTYPE_P2P_CLIENT) | \
-	 BIT(NL80211_IFTYPE_AP) | \
-	 BIT(NL80211_IFTYPE_P2P_GO) | \
-	 HWSIM_MESH_BIT)
-
-#define HWSIM_IFTYPE_SUPPORT_MASK \
-	(BIT(NL80211_IFTYPE_STATION) | \
-	 BIT(NL80211_IFTYPE_AP) | \
-	 BIT(NL80211_IFTYPE_P2P_CLIENT) | \
-	 BIT(NL80211_IFTYPE_P2P_GO) | \
-	 BIT(NL80211_IFTYPE_ADHOC) | \
-	 BIT(NL80211_IFTYPE_MESH_POINT) | \
-	 BIT(NL80211_IFTYPE_OCB))
-
-static int mac80211_hwsim_new_radio(struct genl_info *info,
-				    struct hwsim_new_radio_params *param)
-{
-	int err;
-	u8 addr[ETH_ALEN];
-	struct mac80211_hwsim_data *data;
-	struct ieee80211_hw *hw;
-	enum nl80211_band band;
-	const struct ieee80211_ops *ops = &mac80211_hwsim_ops;
-	struct net *net;
-	int idx, i;
-	int n_limits = 0;
-
-	if (WARN_ON(param->channels > 1 && !param->use_chanctx))
-		return -EINVAL;
-
-	spin_lock_bh(&hwsim_radio_lock);
-	idx = hwsim_radio_idx++;
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	if (param->mlo)
-		ops = &mac80211_hwsim_mlo_ops;
-	else if (param->use_chanctx)
-		ops = &mac80211_hwsim_mchan_ops;
-	hw = ieee80211_alloc_hw_nm(sizeof(*data), ops, param->hwname);
-	if (!hw) {
-		pr_debug("mac80211_hwsim: ieee80211_alloc_hw failed\n");
-		err = -ENOMEM;
-		goto failed;
-	}
-
-	/* ieee80211_alloc_hw_nm may have used a default name */
-	param->hwname = wiphy_name(hw->wiphy);
-
-	if (info)
-		net = genl_info_net(info);
-	else
-		net = &init_net;
-	wiphy_net_set(hw->wiphy, net);
-
-	data = hw->priv;
-	data->hw = hw;
-
-	data->dev = device_create(hwsim_class, NULL, 0, hw, "hwsim%d", idx);
-	if (IS_ERR(data->dev)) {
-		printk(KERN_DEBUG
-		       "mac80211_hwsim: device_create failed (%ld)\n",
-		       PTR_ERR(data->dev));
-		err = -ENOMEM;
-		goto failed_drvdata;
-	}
-	data->dev->driver = &mac80211_hwsim_driver.driver;
-	err = device_bind_driver(data->dev);
-	if (err != 0) {
-		pr_debug("mac80211_hwsim: device_bind_driver failed (%d)\n",
-		       err);
-		goto failed_bind;
-	}
-
-	skb_queue_head_init(&data->pending);
-
-	SET_IEEE80211_DEV(hw, data->dev);
-	if (!param->perm_addr) {
-		eth_zero_addr(addr);
-		addr[0] = 0x02;
-		addr[3] = idx >> 8;
-		addr[4] = idx;
-		memcpy(data->addresses[0].addr, addr, ETH_ALEN);
-		/* Why need here second address ? */
-		memcpy(data->addresses[1].addr, addr, ETH_ALEN);
-		data->addresses[1].addr[0] |= 0x40;
-		hw->wiphy->n_addresses = 2;
-		hw->wiphy->addresses = data->addresses;
-		/* possible address clash is checked at hash table insertion */
-	} else {
-		memcpy(data->addresses[0].addr, param->perm_addr, ETH_ALEN);
-		/* compatibility with automatically generated mac addr */
-		memcpy(data->addresses[1].addr, param->perm_addr, ETH_ALEN);
-		hw->wiphy->n_addresses = 2;
-		hw->wiphy->addresses = data->addresses;
-	}
-
-	data->channels = param->channels;
-	data->use_chanctx = param->use_chanctx;
-	data->idx = idx;
-	data->destroy_on_close = param->destroy_on_close;
-	if (info)
-		data->portid = info->snd_portid;
-
-	/* setup interface limits, only on interface types we support */
-	if (param->iftypes & BIT(NL80211_IFTYPE_ADHOC)) {
-		data->if_limits[n_limits].max = 1;
-		data->if_limits[n_limits].types = BIT(NL80211_IFTYPE_ADHOC);
-		n_limits++;
-	}
-
-	if (param->iftypes & HWSIM_DEFAULT_IF_LIMIT) {
-		data->if_limits[n_limits].max = 2048;
-		/*
-		 * For this case, we may only support a subset of
-		 * HWSIM_DEFAULT_IF_LIMIT, therefore we only want to add the
-		 * bits that both param->iftype & HWSIM_DEFAULT_IF_LIMIT have.
-		 */
-		data->if_limits[n_limits].types =
-					HWSIM_DEFAULT_IF_LIMIT & param->iftypes;
-		n_limits++;
-	}
-
-	if (param->iftypes & BIT(NL80211_IFTYPE_P2P_DEVICE)) {
-		data->if_limits[n_limits].max = 1;
-		data->if_limits[n_limits].types =
-						BIT(NL80211_IFTYPE_P2P_DEVICE);
-		n_limits++;
-	}
-
-	if (data->use_chanctx) {
-		hw->wiphy->max_scan_ssids = 255;
-		hw->wiphy->max_scan_ie_len = IEEE80211_MAX_DATA_LEN;
-		hw->wiphy->max_remain_on_channel_duration = 1000;
-		data->if_combination.radar_detect_widths = 0;
-		data->if_combination.num_different_channels = data->channels;
-	} else {
-		data->if_combination.num_different_channels = 1;
-		data->if_combination.radar_detect_widths =
-					BIT(NL80211_CHAN_WIDTH_5) |
-					BIT(NL80211_CHAN_WIDTH_10) |
-					BIT(NL80211_CHAN_WIDTH_20_NOHT) |
-					BIT(NL80211_CHAN_WIDTH_20) |
-					BIT(NL80211_CHAN_WIDTH_40) |
-					BIT(NL80211_CHAN_WIDTH_80) |
-					BIT(NL80211_CHAN_WIDTH_160);
-	}
-
-	if (!n_limits) {
-		err = -EINVAL;
-		goto failed_hw;
-	}
-
-	data->if_combination.max_interfaces = 0;
-	for (i = 0; i < n_limits; i++)
-		data->if_combination.max_interfaces +=
-			data->if_limits[i].max;
-
-	data->if_combination.n_limits = n_limits;
-	data->if_combination.limits = data->if_limits;
-
-	/*
-	 * If we actually were asked to support combinations,
-	 * advertise them - if there's only a single thing like
-	 * only IBSS then don't advertise it as combinations.
-	 */
-	if (data->if_combination.max_interfaces > 1) {
-		hw->wiphy->iface_combinations = &data->if_combination;
-		hw->wiphy->n_iface_combinations = 1;
-	}
-
-	if (param->ciphers) {
-		memcpy(data->ciphers, param->ciphers,
-		       param->n_ciphers * sizeof(u32));
-		hw->wiphy->cipher_suites = data->ciphers;
-		hw->wiphy->n_cipher_suites = param->n_ciphers;
-	}
-
-	data->rx_rssi = DEFAULT_RX_RSSI;
-
-	INIT_DELAYED_WORK(&data->roc_start, hw_roc_start);
-	INIT_DELAYED_WORK(&data->roc_done, hw_roc_done);
-	INIT_DELAYED_WORK(&data->hw_scan, hw_scan_work);
-
-	hw->queues = 5;
-	hw->offchannel_tx_hw_queue = 4;
-
-	ieee80211_hw_set(hw, SUPPORT_FAST_XMIT);
-	ieee80211_hw_set(hw, CHANCTX_STA_CSA);
-	ieee80211_hw_set(hw, SUPPORTS_HT_CCK_RATES);
-	ieee80211_hw_set(hw, QUEUE_CONTROL);
-	ieee80211_hw_set(hw, WANT_MONITOR_VIF);
-	ieee80211_hw_set(hw, AMPDU_AGGREGATION);
-	ieee80211_hw_set(hw, MFP_CAPABLE);
-	ieee80211_hw_set(hw, SIGNAL_DBM);
-	ieee80211_hw_set(hw, SUPPORTS_PS);
-	ieee80211_hw_set(hw, REPORTS_TX_ACK_STATUS);
-	ieee80211_hw_set(hw, TDLS_WIDER_BW);
-	ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID);
-
-	if (param->mlo) {
-		hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_MLO;
-		ieee80211_hw_set(hw, HAS_RATE_CONTROL);
-		ieee80211_hw_set(hw, SUPPORTS_DYNAMIC_PS);
-		ieee80211_hw_set(hw, CONNECTION_MONITOR);
-		ieee80211_hw_set(hw, AP_LINK_PS);
-	} else {
-		ieee80211_hw_set(hw, HOST_BROADCAST_PS_BUFFERING);
-		ieee80211_hw_set(hw, PS_NULLFUNC_STACK);
-		if (rctbl)
-			ieee80211_hw_set(hw, SUPPORTS_RC_TABLE);
-	}
-
-	hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
-	hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_TDLS |
-			    WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL |
-			    WIPHY_FLAG_AP_UAPSD |
-			    WIPHY_FLAG_SUPPORTS_5_10_MHZ |
-			    WIPHY_FLAG_HAS_CHANNEL_SWITCH;
-	hw->wiphy->features |= NL80211_FEATURE_ACTIVE_MONITOR |
-			       NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE |
-			       NL80211_FEATURE_STATIC_SMPS |
-			       NL80211_FEATURE_DYNAMIC_SMPS |
-			       NL80211_FEATURE_SCAN_RANDOM_MAC_ADDR;
-	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_VHT_IBSS);
-	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_BEACON_PROTECTION);
-	wiphy_ext_feature_set(hw->wiphy,
-			      NL80211_EXT_FEATURE_MULTICAST_REGISTRATIONS);
-	wiphy_ext_feature_set(hw->wiphy,
-			      NL80211_EXT_FEATURE_BEACON_RATE_LEGACY);
-
-	wiphy_ext_feature_set(hw->wiphy,
-			      NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT);
-
-	hw->wiphy->interface_modes = param->iftypes;
-
-	/* ask mac80211 to reserve space for magic */
-	hw->vif_data_size = sizeof(struct hwsim_vif_priv);
-	hw->sta_data_size = sizeof(struct hwsim_sta_priv);
-	hw->chanctx_data_size = sizeof(struct hwsim_chanctx_priv);
-
-	memcpy(data->channels_2ghz, hwsim_channels_2ghz,
-		sizeof(hwsim_channels_2ghz));
-	memcpy(data->channels_5ghz, hwsim_channels_5ghz,
-		sizeof(hwsim_channels_5ghz));
-	memcpy(data->channels_6ghz, hwsim_channels_6ghz,
-		sizeof(hwsim_channels_6ghz));
-	memcpy(data->channels_s1g, hwsim_channels_s1g,
-	       sizeof(hwsim_channels_s1g));
-	memcpy(data->rates, hwsim_rates, sizeof(hwsim_rates));
-
-	for (band = NL80211_BAND_2GHZ; band < NUM_NL80211_BANDS; band++) {
-		struct ieee80211_supported_band *sband = &data->bands[band];
-
-		sband->band = band;
-
-		switch (band) {
-		case NL80211_BAND_2GHZ:
-			sband->channels = data->channels_2ghz;
-			sband->n_channels = ARRAY_SIZE(hwsim_channels_2ghz);
-			sband->bitrates = data->rates;
-			sband->n_bitrates = ARRAY_SIZE(hwsim_rates);
-			break;
-		case NL80211_BAND_5GHZ:
-			sband->channels = data->channels_5ghz;
-			sband->n_channels = ARRAY_SIZE(hwsim_channels_5ghz);
-			sband->bitrates = data->rates + 4;
-			sband->n_bitrates = ARRAY_SIZE(hwsim_rates) - 4;
-
-			sband->vht_cap.vht_supported = true;
-			sband->vht_cap.cap =
-				IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454 |
-				IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ |
-				IEEE80211_VHT_CAP_RXLDPC |
-				IEEE80211_VHT_CAP_SHORT_GI_80 |
-				IEEE80211_VHT_CAP_SHORT_GI_160 |
-				IEEE80211_VHT_CAP_TXSTBC |
-				IEEE80211_VHT_CAP_RXSTBC_4 |
-				IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK;
-			sband->vht_cap.vht_mcs.rx_mcs_map =
-				cpu_to_le16(IEEE80211_VHT_MCS_SUPPORT_0_9 << 0 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 2 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 4 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 6 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 8 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 10 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 12 |
-					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 14);
-			sband->vht_cap.vht_mcs.tx_mcs_map =
-				sband->vht_cap.vht_mcs.rx_mcs_map;
-			break;
-		case NL80211_BAND_6GHZ:
-			sband->channels = data->channels_6ghz;
-			sband->n_channels = ARRAY_SIZE(hwsim_channels_6ghz);
-			sband->bitrates = data->rates + 4;
-			sband->n_bitrates = ARRAY_SIZE(hwsim_rates) - 4;
-			break;
-		case NL80211_BAND_S1GHZ:
-			memcpy(&sband->s1g_cap, &hwsim_s1g_cap,
-			       sizeof(sband->s1g_cap));
-			sband->channels = data->channels_s1g;
-			sband->n_channels = ARRAY_SIZE(hwsim_channels_s1g);
-			break;
-		default:
-			continue;
-		}
-
-		if (band != NL80211_BAND_6GHZ){
-			sband->ht_cap.ht_supported = true;
-			sband->ht_cap.cap = IEEE80211_HT_CAP_SUP_WIDTH_20_40 |
-					    IEEE80211_HT_CAP_GRN_FLD |
-					    IEEE80211_HT_CAP_SGI_20 |
-					    IEEE80211_HT_CAP_SGI_40 |
-					    IEEE80211_HT_CAP_DSSSCCK40;
-			sband->ht_cap.ampdu_factor = 0x3;
-			sband->ht_cap.ampdu_density = 0x6;
-			memset(&sband->ht_cap.mcs, 0,
-			       sizeof(sband->ht_cap.mcs));
-			sband->ht_cap.mcs.rx_mask[0] = 0xff;
-			sband->ht_cap.mcs.rx_mask[1] = 0xff;
-			sband->ht_cap.mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED;
-		}
-
-		mac80211_hwsim_sband_capab(sband);
-
-		hw->wiphy->bands[band] = sband;
-	}
-
-	/* By default all radios belong to the first group */
-	data->group = 1;
-	mutex_init(&data->mutex);
-
-	data->netgroup = hwsim_net_get_netgroup(net);
-	data->wmediumd = hwsim_net_get_wmediumd(net);
-
-	/* Enable frame retransmissions for lossy channels */
-	hw->max_rates = 4;
-	hw->max_rate_tries = 11;
-
-	hw->wiphy->vendor_commands = mac80211_hwsim_vendor_commands;
-	hw->wiphy->n_vendor_commands =
-		ARRAY_SIZE(mac80211_hwsim_vendor_commands);
-	hw->wiphy->vendor_events = mac80211_hwsim_vendor_events;
-	hw->wiphy->n_vendor_events = ARRAY_SIZE(mac80211_hwsim_vendor_events);
-
-	if (param->reg_strict)
-		hw->wiphy->regulatory_flags |= REGULATORY_STRICT_REG;
-	if (param->regd) {
-		data->regd = param->regd;
-		hw->wiphy->regulatory_flags |= REGULATORY_CUSTOM_REG;
-		wiphy_apply_custom_regulatory(hw->wiphy, param->regd);
-		/* give the regulatory workqueue a chance to run */
-		schedule_timeout_interruptible(1);
-	}
-
-	if (param->no_vif)
-		ieee80211_hw_set(hw, NO_AUTO_VIF);
-
-	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST);
-
-	for (i = 0; i < ARRAY_SIZE(data->link_data); i++) {
-		hrtimer_init(&data->link_data[i].beacon_timer, CLOCK_MONOTONIC,
-			     HRTIMER_MODE_ABS_SOFT);
-		data->link_data[i].beacon_timer.function =
-			mac80211_hwsim_beacon;
-		data->link_data[i].link_id = i;
-	}
-
-	err = ieee80211_register_hw(hw);
-	if (err < 0) {
-		pr_debug("mac80211_hwsim: ieee80211_register_hw failed (%d)\n",
-		       err);
-		goto failed_hw;
-	}
-
-	wiphy_dbg(hw->wiphy, "hwaddr %pM registered\n", hw->wiphy->perm_addr);
-
-	if (param->reg_alpha2) {
-		data->alpha2[0] = param->reg_alpha2[0];
-		data->alpha2[1] = param->reg_alpha2[1];
-		regulatory_hint(hw->wiphy, param->reg_alpha2);
-	}
-
-	data->debugfs = debugfs_create_dir("hwsim", hw->wiphy->debugfsdir);
-	debugfs_create_file("ps", 0666, data->debugfs, data, &hwsim_fops_ps);
-	debugfs_create_file("group", 0666, data->debugfs, data,
-			    &hwsim_fops_group);
-	debugfs_create_file("rx_rssi", 0666, data->debugfs, data,
-			    &hwsim_fops_rx_rssi);
-	if (!data->use_chanctx)
-		debugfs_create_file("dfs_simulate_radar", 0222,
-				    data->debugfs,
-				    data, &hwsim_simulate_radar);
-
-	spin_lock_bh(&hwsim_radio_lock);
-	err = rhashtable_insert_fast(&hwsim_radios_rht, &data->rht,
-				     hwsim_rht_params);
-	if (err < 0) {
-		if (info) {
-			GENL_SET_ERR_MSG(info, "perm addr already present");
-			NL_SET_BAD_ATTR(info->extack,
-					info->attrs[HWSIM_ATTR_PERM_ADDR]);
-		}
-		spin_unlock_bh(&hwsim_radio_lock);
-		goto failed_final_insert;
-	}
-
-	list_add_tail(&data->list, &hwsim_radios);
-	hwsim_radios_generation++;
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	hwsim_mcast_new_radio(idx, info, param);
-
-	return idx;
-
-failed_final_insert:
-	debugfs_remove_recursive(data->debugfs);
-	ieee80211_unregister_hw(data->hw);
-failed_hw:
-	device_release_driver(data->dev);
-failed_bind:
-	device_unregister(data->dev);
-failed_drvdata:
-	ieee80211_free_hw(hw);
-failed:
-	return err;
-}
-
-static void hwsim_mcast_del_radio(int id, const char *hwname,
-				  struct genl_info *info)
-{
-	struct sk_buff *skb;
-	void *data;
-	int ret;
-
-	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
-	if (!skb)
-		return;
-
-	data = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0,
-			   HWSIM_CMD_DEL_RADIO);
-	if (!data)
-		goto error;
-
-	ret = nla_put_u32(skb, HWSIM_ATTR_RADIO_ID, id);
-	if (ret < 0)
-		goto error;
-
-	ret = nla_put(skb, HWSIM_ATTR_RADIO_NAME, strlen(hwname),
-		      hwname);
-	if (ret < 0)
-		goto error;
-
-	genlmsg_end(skb, data);
-
-	hwsim_mcast_config_msg(skb, info);
-
-	return;
-
-error:
-	nlmsg_free(skb);
-}
-
-static void mac80211_hwsim_del_radio(struct mac80211_hwsim_data *data,
-				     const char *hwname,
-				     struct genl_info *info)
-{
-	hwsim_mcast_del_radio(data->idx, hwname, info);
-	debugfs_remove_recursive(data->debugfs);
-	ieee80211_unregister_hw(data->hw);
-	device_release_driver(data->dev);
-	device_unregister(data->dev);
-	ieee80211_free_hw(data->hw);
-}
-
-static int mac80211_hwsim_get_radio(struct sk_buff *skb,
-				    struct mac80211_hwsim_data *data,
-				    u32 portid, u32 seq,
-				    struct netlink_callback *cb, int flags)
-{
-	void *hdr;
-	struct hwsim_new_radio_params param = { };
-	int res = -EMSGSIZE;
-
-	hdr = genlmsg_put(skb, portid, seq, &hwsim_genl_family, flags,
-			  HWSIM_CMD_GET_RADIO);
-	if (!hdr)
-		return -EMSGSIZE;
-
-	if (cb)
-		genl_dump_check_consistent(cb, hdr);
-
-	if (data->alpha2[0] && data->alpha2[1])
-		param.reg_alpha2 = data->alpha2;
-
-	param.reg_strict = !!(data->hw->wiphy->regulatory_flags &
-					REGULATORY_STRICT_REG);
-	param.p2p_device = !!(data->hw->wiphy->interface_modes &
-					BIT(NL80211_IFTYPE_P2P_DEVICE));
-	param.use_chanctx = data->use_chanctx;
-	param.regd = data->regd;
-	param.channels = data->channels;
-	param.hwname = wiphy_name(data->hw->wiphy);
-
-	res = append_radio_msg(skb, data->idx, &param);
-	if (res < 0)
-		goto out_err;
-
-	genlmsg_end(skb, hdr);
-	return 0;
-
-out_err:
-	genlmsg_cancel(skb, hdr);
-	return res;
-}
-
-static void mac80211_hwsim_free(void)
-{
-	struct mac80211_hwsim_data *data;
-
-	spin_lock_bh(&hwsim_radio_lock);
-	while ((data = list_first_entry_or_null(&hwsim_radios,
-						struct mac80211_hwsim_data,
-						list))) {
-		list_del(&data->list);
-		spin_unlock_bh(&hwsim_radio_lock);
-		mac80211_hwsim_del_radio(data, wiphy_name(data->hw->wiphy),
-					 NULL);
-		spin_lock_bh(&hwsim_radio_lock);
-	}
-	spin_unlock_bh(&hwsim_radio_lock);
-	class_destroy(hwsim_class);
-}
-
-static const struct net_device_ops hwsim_netdev_ops = {
-	.ndo_start_xmit 	= hwsim_mon_xmit,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-static void hwsim_mon_setup(struct net_device *dev)
-{
-	u8 addr[ETH_ALEN];
-
-	dev->netdev_ops = &hwsim_netdev_ops;
-	dev->needs_free_netdev = true;
-	ether_setup(dev);
-	dev->priv_flags |= IFF_NO_QUEUE;
-	dev->type = ARPHRD_IEEE80211_RADIOTAP;
-	eth_zero_addr(addr);
-	addr[0] = 0x12;
-	eth_hw_addr_set(dev, addr);
-}
-
-static struct mac80211_hwsim_data *get_hwsim_data_ref_from_addr(const u8 *addr)
-{
-	return rhashtable_lookup_fast(&hwsim_radios_rht,
-				      addr,
-				      hwsim_rht_params);
-}
-
-static void hwsim_register_wmediumd(struct net *net, u32 portid)
-{
-	struct mac80211_hwsim_data *data;
-
-	hwsim_net_set_wmediumd(net, portid);
-
-	spin_lock_bh(&hwsim_radio_lock);
-	list_for_each_entry(data, &hwsim_radios, list) {
-		if (data->netgroup == hwsim_net_get_netgroup(net))
-			data->wmediumd = portid;
-	}
-	spin_unlock_bh(&hwsim_radio_lock);
-}
-
-static int hwsim_tx_info_frame_received_nl(struct sk_buff *skb_2,
-					   struct genl_info *info)
-{
-
-	struct ieee80211_hdr *hdr;
-	struct mac80211_hwsim_data *data2;
-	struct ieee80211_tx_info *txi;
-	struct hwsim_tx_rate *tx_attempts;
-	u64 ret_skb_cookie;
-	struct sk_buff *skb, *tmp;
-	const u8 *src;
-	unsigned int hwsim_flags;
-	int i;
-	unsigned long flags;
-	bool found = false;
-
-	if (!info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER] ||
-	    !info->attrs[HWSIM_ATTR_FLAGS] ||
-	    !info->attrs[HWSIM_ATTR_COOKIE] ||
-	    !info->attrs[HWSIM_ATTR_SIGNAL] ||
-	    !info->attrs[HWSIM_ATTR_TX_INFO])
-		goto out;
-
-	src = (void *)nla_data(info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER]);
-	hwsim_flags = nla_get_u32(info->attrs[HWSIM_ATTR_FLAGS]);
-	ret_skb_cookie = nla_get_u64(info->attrs[HWSIM_ATTR_COOKIE]);
-
-	data2 = get_hwsim_data_ref_from_addr(src);
-	if (!data2)
-		goto out;
-
-	if (!hwsim_virtio_enabled) {
-		if (hwsim_net_get_netgroup(genl_info_net(info)) !=
-		    data2->netgroup)
-			goto out;
-
-		if (info->snd_portid != data2->wmediumd)
-			goto out;
-	}
-
-	/* look for the skb matching the cookie passed back from user */
-	spin_lock_irqsave(&data2->pending.lock, flags);
-	skb_queue_walk_safe(&data2->pending, skb, tmp) {
-		uintptr_t skb_cookie;
-
-		txi = IEEE80211_SKB_CB(skb);
-		skb_cookie = (uintptr_t)txi->rate_driver_data[0];
-
-		if (skb_cookie == ret_skb_cookie) {
-			__skb_unlink(skb, &data2->pending);
-			found = true;
-			break;
-		}
-	}
-	spin_unlock_irqrestore(&data2->pending.lock, flags);
-
-	/* not found */
-	if (!found)
-		goto out;
-
-	/* Tx info received because the frame was broadcasted on user space,
-	 so we get all the necessary info: tx attempts and skb control buff */
-
-	tx_attempts = (struct hwsim_tx_rate *)nla_data(
-		       info->attrs[HWSIM_ATTR_TX_INFO]);
-
-	/* now send back TX status */
-	txi = IEEE80211_SKB_CB(skb);
-
-	ieee80211_tx_info_clear_status(txi);
-
-	for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) {
-		txi->status.rates[i].idx = tx_attempts[i].idx;
-		txi->status.rates[i].count = tx_attempts[i].count;
-	}
-
-	txi->status.ack_signal = nla_get_u32(info->attrs[HWSIM_ATTR_SIGNAL]);
-
-	if (!(hwsim_flags & HWSIM_TX_CTL_NO_ACK) &&
-	   (hwsim_flags & HWSIM_TX_STAT_ACK)) {
-		if (skb->len >= 16) {
-			hdr = (struct ieee80211_hdr *) skb->data;
-			mac80211_hwsim_monitor_ack(data2->channel,
-						   hdr->addr2);
-		}
-		txi->flags |= IEEE80211_TX_STAT_ACK;
-	}
-
-	if (hwsim_flags & HWSIM_TX_CTL_NO_ACK)
-		txi->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
-
-	ieee80211_tx_status_irqsafe(data2->hw, skb);
-	return 0;
-out:
-	return -EINVAL;
-
-}
-
-static int hwsim_cloned_frame_received_nl(struct sk_buff *skb_2,
-					  struct genl_info *info)
-{
-	struct mac80211_hwsim_data *data2;
-	struct ieee80211_rx_status rx_status;
-	struct ieee80211_hdr *hdr;
-	const u8 *dst;
-	int frame_data_len;
-	void *frame_data;
-	struct sk_buff *skb = NULL;
-	struct ieee80211_channel *channel = NULL;
-
-	if (!info->attrs[HWSIM_ATTR_ADDR_RECEIVER] ||
-	    !info->attrs[HWSIM_ATTR_FRAME] ||
-	    !info->attrs[HWSIM_ATTR_RX_RATE] ||
-	    !info->attrs[HWSIM_ATTR_SIGNAL])
-		goto out;
-
-	dst = (void *)nla_data(info->attrs[HWSIM_ATTR_ADDR_RECEIVER]);
-	frame_data_len = nla_len(info->attrs[HWSIM_ATTR_FRAME]);
-	frame_data = (void *)nla_data(info->attrs[HWSIM_ATTR_FRAME]);
-
-	/* Allocate new skb here */
-	skb = alloc_skb(frame_data_len, GFP_KERNEL);
-	if (skb == NULL)
-		goto err;
-
-	if (frame_data_len > IEEE80211_MAX_DATA_LEN)
-		goto err;
-
-	/* Copy the data */
-	skb_put_data(skb, frame_data, frame_data_len);
-
-	data2 = get_hwsim_data_ref_from_addr(dst);
-	if (!data2)
-		goto out;
-
-	if (data2->use_chanctx) {
-		if (data2->tmp_chan)
-			channel = data2->tmp_chan;
-	} else {
-		channel = data2->channel;
-	}
-
-	if (!hwsim_virtio_enabled) {
-		if (hwsim_net_get_netgroup(genl_info_net(info)) !=
-		    data2->netgroup)
-			goto out;
-
-		if (info->snd_portid != data2->wmediumd)
-			goto out;
-	}
-
-	/* check if radio is configured properly */
-
-	if ((data2->idle && !data2->tmp_chan) || !data2->started)
-		goto out;
-
-	/* A frame is received from user space */
-	memset(&rx_status, 0, sizeof(rx_status));
-	if (info->attrs[HWSIM_ATTR_FREQ]) {
-		struct tx_iter_data iter_data = {};
-
-		/* throw away off-channel packets, but allow both the temporary
-		 * ("hw" scan/remain-on-channel), regular channels and links,
-		 * since the internal datapath also allows this
-		 */
-		rx_status.freq = nla_get_u32(info->attrs[HWSIM_ATTR_FREQ]);
-
-		iter_data.channel = ieee80211_get_channel(data2->hw->wiphy,
-							  rx_status.freq);
-		if (!iter_data.channel)
-			goto out;
-		rx_status.band = iter_data.channel->band;
-
-		mutex_lock(&data2->mutex);
-		if (!hwsim_chans_compat(iter_data.channel, channel)) {
-			ieee80211_iterate_active_interfaces_atomic(
-				data2->hw, IEEE80211_IFACE_ITER_NORMAL,
-				mac80211_hwsim_tx_iter, &iter_data);
-			if (!iter_data.receive) {
-				mutex_unlock(&data2->mutex);
-				goto out;
-			}
-		}
-		mutex_unlock(&data2->mutex);
-	} else if (!channel) {
-		goto out;
-	} else {
-		rx_status.freq = channel->center_freq;
-		rx_status.band = channel->band;
-	}
-
-	rx_status.rate_idx = nla_get_u32(info->attrs[HWSIM_ATTR_RX_RATE]);
-	if (rx_status.rate_idx >= data2->hw->wiphy->bands[rx_status.band]->n_bitrates)
-		goto out;
-	rx_status.signal = nla_get_u32(info->attrs[HWSIM_ATTR_SIGNAL]);
-
-	hdr = (void *)skb->data;
-
-	if (ieee80211_is_beacon(hdr->frame_control) ||
-	    ieee80211_is_probe_resp(hdr->frame_control))
-		rx_status.boottime_ns = ktime_get_boottime_ns();
-
-	mac80211_hwsim_rx(data2, &rx_status, skb);
-
-	return 0;
-err:
-	pr_debug("mac80211_hwsim: error occurred in %s\n", __func__);
-out:
-	dev_kfree_skb(skb);
-	return -EINVAL;
-}
-
-static int hwsim_register_received_nl(struct sk_buff *skb_2,
-				      struct genl_info *info)
-{
-	struct net *net = genl_info_net(info);
-	struct mac80211_hwsim_data *data;
-	int chans = 1;
-
-	spin_lock_bh(&hwsim_radio_lock);
-	list_for_each_entry(data, &hwsim_radios, list)
-		chans = max(chans, data->channels);
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	/* In the future we should revise the userspace API and allow it
-	 * to set a flag that it does support multi-channel, then we can
-	 * let this pass conditionally on the flag.
-	 * For current userspace, prohibit it since it won't work right.
-	 */
-	if (chans > 1)
-		return -EOPNOTSUPP;
-
-	if (hwsim_net_get_wmediumd(net))
-		return -EBUSY;
-
-	hwsim_register_wmediumd(net, info->snd_portid);
-
-	pr_debug("mac80211_hwsim: received a REGISTER, "
-	       "switching to wmediumd mode with pid %d\n", info->snd_portid);
-
-	return 0;
-}
-
-/* ensures ciphers only include ciphers listed in 'hwsim_ciphers' array */
-static bool hwsim_known_ciphers(const u32 *ciphers, int n_ciphers)
-{
-	int i;
-
-	for (i = 0; i < n_ciphers; i++) {
-		int j;
-		int found = 0;
-
-		for (j = 0; j < ARRAY_SIZE(hwsim_ciphers); j++) {
-			if (ciphers[i] == hwsim_ciphers[j]) {
-				found = 1;
-				break;
-			}
-		}
-
-		if (!found)
-			return false;
-	}
-
-	return true;
-}
-
-static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info)
-{
-	struct hwsim_new_radio_params param = { 0 };
-	const char *hwname = NULL;
-	int ret;
-
-	param.reg_strict = info->attrs[HWSIM_ATTR_REG_STRICT_REG];
-	param.p2p_device = info->attrs[HWSIM_ATTR_SUPPORT_P2P_DEVICE];
-	param.channels = channels;
-	param.destroy_on_close =
-		info->attrs[HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE];
-
-	if (info->attrs[HWSIM_ATTR_CHANNELS])
-		param.channels = nla_get_u32(info->attrs[HWSIM_ATTR_CHANNELS]);
-
-	if (param.channels < 1) {
-		GENL_SET_ERR_MSG(info, "must have at least one channel");
-		return -EINVAL;
-	}
-
-	if (info->attrs[HWSIM_ATTR_NO_VIF])
-		param.no_vif = true;
-
-	if (info->attrs[HWSIM_ATTR_USE_CHANCTX])
-		param.use_chanctx = true;
-	else
-		param.use_chanctx = (param.channels > 1);
-
-	if (info->attrs[HWSIM_ATTR_REG_HINT_ALPHA2])
-		param.reg_alpha2 =
-			nla_data(info->attrs[HWSIM_ATTR_REG_HINT_ALPHA2]);
-
-	if (info->attrs[HWSIM_ATTR_REG_CUSTOM_REG]) {
-		u32 idx = nla_get_u32(info->attrs[HWSIM_ATTR_REG_CUSTOM_REG]);
-
-		if (idx >= ARRAY_SIZE(hwsim_world_regdom_custom))
-			return -EINVAL;
-
-		idx = array_index_nospec(idx,
-					 ARRAY_SIZE(hwsim_world_regdom_custom));
-		param.regd = hwsim_world_regdom_custom[idx];
-	}
-
-	if (info->attrs[HWSIM_ATTR_PERM_ADDR]) {
-		if (!is_valid_ether_addr(
-				nla_data(info->attrs[HWSIM_ATTR_PERM_ADDR]))) {
-			GENL_SET_ERR_MSG(info,"MAC is no valid source addr");
-			NL_SET_BAD_ATTR(info->extack,
-					info->attrs[HWSIM_ATTR_PERM_ADDR]);
-			return -EINVAL;
-		}
-
-		param.perm_addr = nla_data(info->attrs[HWSIM_ATTR_PERM_ADDR]);
-	}
-
-	if (info->attrs[HWSIM_ATTR_IFTYPE_SUPPORT]) {
-		param.iftypes =
-			nla_get_u32(info->attrs[HWSIM_ATTR_IFTYPE_SUPPORT]);
-
-		if (param.iftypes & ~HWSIM_IFTYPE_SUPPORT_MASK) {
-			NL_SET_ERR_MSG_ATTR(info->extack,
-					    info->attrs[HWSIM_ATTR_IFTYPE_SUPPORT],
-					    "cannot support more iftypes than kernel");
-			return -EINVAL;
-		}
-	} else {
-		param.iftypes = HWSIM_IFTYPE_SUPPORT_MASK;
-	}
-
-	/* ensure both flag and iftype support is honored */
-	if (param.p2p_device ||
-	    param.iftypes & BIT(NL80211_IFTYPE_P2P_DEVICE)) {
-		param.iftypes |= BIT(NL80211_IFTYPE_P2P_DEVICE);
-		param.p2p_device = true;
-	}
-
-	if (info->attrs[HWSIM_ATTR_CIPHER_SUPPORT]) {
-		u32 len = nla_len(info->attrs[HWSIM_ATTR_CIPHER_SUPPORT]);
-
-		param.ciphers =
-			nla_data(info->attrs[HWSIM_ATTR_CIPHER_SUPPORT]);
-
-		if (len % sizeof(u32)) {
-			NL_SET_ERR_MSG_ATTR(info->extack,
-					    info->attrs[HWSIM_ATTR_CIPHER_SUPPORT],
-					    "bad cipher list length");
-			return -EINVAL;
-		}
-
-		param.n_ciphers = len / sizeof(u32);
-
-		if (param.n_ciphers > ARRAY_SIZE(hwsim_ciphers)) {
-			NL_SET_ERR_MSG_ATTR(info->extack,
-					    info->attrs[HWSIM_ATTR_CIPHER_SUPPORT],
-					    "too many ciphers specified");
-			return -EINVAL;
-		}
-
-		if (!hwsim_known_ciphers(param.ciphers, param.n_ciphers)) {
-			NL_SET_ERR_MSG_ATTR(info->extack,
-					    info->attrs[HWSIM_ATTR_CIPHER_SUPPORT],
-					    "unsupported ciphers specified");
-			return -EINVAL;
-		}
-	}
-
-	param.mlo = info->attrs[HWSIM_ATTR_MLO_SUPPORT];
-
-	if (param.mlo)
-		param.use_chanctx = true;
-
-	if (info->attrs[HWSIM_ATTR_RADIO_NAME]) {
-		hwname = kstrndup((char *)nla_data(info->attrs[HWSIM_ATTR_RADIO_NAME]),
-				  nla_len(info->attrs[HWSIM_ATTR_RADIO_NAME]),
-				  GFP_KERNEL);
-		if (!hwname)
-			return -ENOMEM;
-		param.hwname = hwname;
-	}
-
-	ret = mac80211_hwsim_new_radio(info, &param);
-	kfree(hwname);
-	return ret;
-}
-
-static int hwsim_del_radio_nl(struct sk_buff *msg, struct genl_info *info)
-{
-	struct mac80211_hwsim_data *data;
-	s64 idx = -1;
-	const char *hwname = NULL;
-
-	if (info->attrs[HWSIM_ATTR_RADIO_ID]) {
-		idx = nla_get_u32(info->attrs[HWSIM_ATTR_RADIO_ID]);
-	} else if (info->attrs[HWSIM_ATTR_RADIO_NAME]) {
-		hwname = kstrndup((char *)nla_data(info->attrs[HWSIM_ATTR_RADIO_NAME]),
-				  nla_len(info->attrs[HWSIM_ATTR_RADIO_NAME]),
-				  GFP_KERNEL);
-		if (!hwname)
-			return -ENOMEM;
-	} else
-		return -EINVAL;
-
-	spin_lock_bh(&hwsim_radio_lock);
-	list_for_each_entry(data, &hwsim_radios, list) {
-		if (idx >= 0) {
-			if (data->idx != idx)
-				continue;
-		} else {
-			if (!hwname ||
-			    strcmp(hwname, wiphy_name(data->hw->wiphy)))
-				continue;
-		}
-
-		if (!net_eq(wiphy_net(data->hw->wiphy), genl_info_net(info)))
-			continue;
-
-		list_del(&data->list);
-		rhashtable_remove_fast(&hwsim_radios_rht, &data->rht,
-				       hwsim_rht_params);
-		hwsim_radios_generation++;
-		spin_unlock_bh(&hwsim_radio_lock);
-		mac80211_hwsim_del_radio(data, wiphy_name(data->hw->wiphy),
-					 info);
-		kfree(hwname);
-		return 0;
-	}
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	kfree(hwname);
-	return -ENODEV;
-}
-
-static int hwsim_get_radio_nl(struct sk_buff *msg, struct genl_info *info)
-{
-	struct mac80211_hwsim_data *data;
-	struct sk_buff *skb;
-	int idx, res = -ENODEV;
-
-	if (!info->attrs[HWSIM_ATTR_RADIO_ID])
-		return -EINVAL;
-	idx = nla_get_u32(info->attrs[HWSIM_ATTR_RADIO_ID]);
-
-	spin_lock_bh(&hwsim_radio_lock);
-	list_for_each_entry(data, &hwsim_radios, list) {
-		if (data->idx != idx)
-			continue;
-
-		if (!net_eq(wiphy_net(data->hw->wiphy), genl_info_net(info)))
-			continue;
-
-		skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
-		if (!skb) {
-			res = -ENOMEM;
-			goto out_err;
-		}
-
-		res = mac80211_hwsim_get_radio(skb, data, info->snd_portid,
-					       info->snd_seq, NULL, 0);
-		if (res < 0) {
-			nlmsg_free(skb);
-			goto out_err;
-		}
-
-		res = genlmsg_reply(skb, info);
-		break;
-	}
-
-out_err:
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	return res;
-}
-
-static int hwsim_dump_radio_nl(struct sk_buff *skb,
-			       struct netlink_callback *cb)
-{
-	int last_idx = cb->args[0] - 1;
-	struct mac80211_hwsim_data *data = NULL;
-	int res = 0;
-	void *hdr;
-
-	spin_lock_bh(&hwsim_radio_lock);
-	cb->seq = hwsim_radios_generation;
-
-	if (last_idx >= hwsim_radio_idx-1)
-		goto done;
-
-	list_for_each_entry(data, &hwsim_radios, list) {
-		if (data->idx <= last_idx)
-			continue;
-
-		if (!net_eq(wiphy_net(data->hw->wiphy), sock_net(skb->sk)))
-			continue;
-
-		res = mac80211_hwsim_get_radio(skb, data,
-					       NETLINK_CB(cb->skb).portid,
-					       cb->nlh->nlmsg_seq, cb,
-					       NLM_F_MULTI);
-		if (res < 0)
-			break;
-
-		last_idx = data->idx;
-	}
-
-	cb->args[0] = last_idx + 1;
-
-	/* list changed, but no new element sent, set interrupted flag */
-	if (skb->len == 0 && cb->prev_seq && cb->seq != cb->prev_seq) {
-		hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid,
-				  cb->nlh->nlmsg_seq, &hwsim_genl_family,
-				  NLM_F_MULTI, HWSIM_CMD_GET_RADIO);
-		if (hdr) {
-			genl_dump_check_consistent(cb, hdr);
-			genlmsg_end(skb, hdr);
-		} else {
-			res = -EMSGSIZE;
-		}
-	}
-
-done:
-	spin_unlock_bh(&hwsim_radio_lock);
-	return res ?: skb->len;
-}
-
-/* Generic Netlink operations array */
-static const struct genl_small_ops hwsim_ops[] = {
-	{
-		.cmd = HWSIM_CMD_REGISTER,
-		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.doit = hwsim_register_received_nl,
-		.flags = GENL_UNS_ADMIN_PERM,
-	},
-	{
-		.cmd = HWSIM_CMD_FRAME,
-		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.doit = hwsim_cloned_frame_received_nl,
-	},
-	{
-		.cmd = HWSIM_CMD_TX_INFO_FRAME,
-		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.doit = hwsim_tx_info_frame_received_nl,
-	},
-	{
-		.cmd = HWSIM_CMD_NEW_RADIO,
-		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.doit = hwsim_new_radio_nl,
-		.flags = GENL_UNS_ADMIN_PERM,
-	},
-	{
-		.cmd = HWSIM_CMD_DEL_RADIO,
-		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.doit = hwsim_del_radio_nl,
-		.flags = GENL_UNS_ADMIN_PERM,
-	},
-	{
-		.cmd = HWSIM_CMD_GET_RADIO,
-		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
-		.doit = hwsim_get_radio_nl,
-		.dumpit = hwsim_dump_radio_nl,
-	},
-};
-
-static struct genl_family hwsim_genl_family __ro_after_init = {
-	.name = "MAC80211_HWSIM",
-	.version = 1,
-	.maxattr = HWSIM_ATTR_MAX,
-	.policy = hwsim_genl_policy,
-	.netnsok = true,
-	.module = THIS_MODULE,
-	.small_ops = hwsim_ops,
-	.n_small_ops = ARRAY_SIZE(hwsim_ops),
-	.resv_start_op = HWSIM_CMD_DEL_MAC_ADDR + 1,
-	.mcgrps = hwsim_mcgrps,
-	.n_mcgrps = ARRAY_SIZE(hwsim_mcgrps),
-};
-
-static void remove_user_radios(u32 portid)
-{
-	struct mac80211_hwsim_data *entry, *tmp;
-	LIST_HEAD(list);
-
-	spin_lock_bh(&hwsim_radio_lock);
-	list_for_each_entry_safe(entry, tmp, &hwsim_radios, list) {
-		if (entry->destroy_on_close && entry->portid == portid) {
-			list_move(&entry->list, &list);
-			rhashtable_remove_fast(&hwsim_radios_rht, &entry->rht,
-					       hwsim_rht_params);
-			hwsim_radios_generation++;
-		}
-	}
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	list_for_each_entry_safe(entry, tmp, &list, list) {
-		list_del(&entry->list);
-		mac80211_hwsim_del_radio(entry, wiphy_name(entry->hw->wiphy),
-					 NULL);
-	}
-}
-
-static int mac80211_hwsim_netlink_notify(struct notifier_block *nb,
-					 unsigned long state,
-					 void *_notify)
-{
-	struct netlink_notify *notify = _notify;
-
-	if (state != NETLINK_URELEASE)
-		return NOTIFY_DONE;
-
-	remove_user_radios(notify->portid);
-
-	if (notify->portid == hwsim_net_get_wmediumd(notify->net)) {
-		printk(KERN_INFO "mac80211_hwsim: wmediumd released netlink"
-		       " socket, switching to perfect channel medium\n");
-		hwsim_register_wmediumd(notify->net, 0);
-	}
-	return NOTIFY_DONE;
-
-}
-
-static struct notifier_block hwsim_netlink_notifier = {
-	.notifier_call = mac80211_hwsim_netlink_notify,
-};
-
-static int __init hwsim_init_netlink(void)
-{
-	int rc;
-
-	printk(KERN_INFO "mac80211_hwsim: initializing netlink\n");
-
-	rc = genl_register_family(&hwsim_genl_family);
-	if (rc)
-		goto failure;
-
-	rc = netlink_register_notifier(&hwsim_netlink_notifier);
-	if (rc) {
-		genl_unregister_family(&hwsim_genl_family);
-		goto failure;
-	}
-
-	return 0;
-
-failure:
-	pr_debug("mac80211_hwsim: error occurred in %s\n", __func__);
-	return -EINVAL;
-}
-
-static __net_init int hwsim_init_net(struct net *net)
-{
-	return hwsim_net_set_netgroup(net);
-}
-
-static void __net_exit hwsim_exit_net(struct net *net)
-{
-	struct mac80211_hwsim_data *data, *tmp;
-	LIST_HEAD(list);
-
-	spin_lock_bh(&hwsim_radio_lock);
-	list_for_each_entry_safe(data, tmp, &hwsim_radios, list) {
-		if (!net_eq(wiphy_net(data->hw->wiphy), net))
-			continue;
-
-		/* Radios created in init_net are returned to init_net. */
-		if (data->netgroup == hwsim_net_get_netgroup(&init_net))
-			continue;
-
-		list_move(&data->list, &list);
-		rhashtable_remove_fast(&hwsim_radios_rht, &data->rht,
-				       hwsim_rht_params);
-		hwsim_radios_generation++;
-	}
-	spin_unlock_bh(&hwsim_radio_lock);
-
-	list_for_each_entry_safe(data, tmp, &list, list) {
-		list_del(&data->list);
-		mac80211_hwsim_del_radio(data,
-					 wiphy_name(data->hw->wiphy),
-					 NULL);
-	}
-
-	ida_free(&hwsim_netgroup_ida, hwsim_net_get_netgroup(net));
-}
-
-static struct pernet_operations hwsim_net_ops = {
-	.init = hwsim_init_net,
-	.exit = hwsim_exit_net,
-	.id   = &hwsim_net_id,
-	.size = sizeof(struct hwsim_net),
-};
-
-static void hwsim_exit_netlink(void)
-{
-	/* unregister the notifier */
-	netlink_unregister_notifier(&hwsim_netlink_notifier);
-	/* unregister the family */
-	genl_unregister_family(&hwsim_genl_family);
-}
-
-#if IS_REACHABLE(CONFIG_VIRTIO)
-static void hwsim_virtio_tx_done(struct virtqueue *vq)
-{
-	unsigned int len;
-	struct sk_buff *skb;
-	unsigned long flags;
-
-	spin_lock_irqsave(&hwsim_virtio_lock, flags);
-	while ((skb = virtqueue_get_buf(vq, &len)))
-		nlmsg_free(skb);
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-}
-
-static int hwsim_virtio_handle_cmd(struct sk_buff *skb)
-{
-	struct nlmsghdr *nlh;
-	struct genlmsghdr *gnlh;
-	struct nlattr *tb[HWSIM_ATTR_MAX + 1];
-	struct genl_info info = {};
-	int err;
-
-	nlh = nlmsg_hdr(skb);
-	gnlh = nlmsg_data(nlh);
-
-	if (skb->len < nlh->nlmsg_len)
-		return -EINVAL;
-
-	err = genlmsg_parse(nlh, &hwsim_genl_family, tb, HWSIM_ATTR_MAX,
-			    hwsim_genl_policy, NULL);
-	if (err) {
-		pr_err_ratelimited("hwsim: genlmsg_parse returned %d\n", err);
-		return err;
-	}
-
-	info.attrs = tb;
-
-	switch (gnlh->cmd) {
-	case HWSIM_CMD_FRAME:
-		hwsim_cloned_frame_received_nl(skb, &info);
-		break;
-	case HWSIM_CMD_TX_INFO_FRAME:
-		hwsim_tx_info_frame_received_nl(skb, &info);
-		break;
-	default:
-		pr_err_ratelimited("hwsim: invalid cmd: %d\n", gnlh->cmd);
-		return -EPROTO;
-	}
-	return 0;
-}
-
-static void hwsim_virtio_rx_work(struct work_struct *work)
-{
-	struct virtqueue *vq;
-	unsigned int len;
-	struct sk_buff *skb;
-	struct scatterlist sg[1];
-	int err;
-	unsigned long flags;
-
-	spin_lock_irqsave(&hwsim_virtio_lock, flags);
-	if (!hwsim_virtio_enabled)
-		goto out_unlock;
-
-	skb = virtqueue_get_buf(hwsim_vqs[HWSIM_VQ_RX], &len);
-	if (!skb)
-		goto out_unlock;
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-
-	skb->data = skb->head;
-	skb_reset_tail_pointer(skb);
-	skb_put(skb, len);
-	hwsim_virtio_handle_cmd(skb);
-
-	spin_lock_irqsave(&hwsim_virtio_lock, flags);
-	if (!hwsim_virtio_enabled) {
-		nlmsg_free(skb);
-		goto out_unlock;
-	}
-	vq = hwsim_vqs[HWSIM_VQ_RX];
-	sg_init_one(sg, skb->head, skb_end_offset(skb));
-	err = virtqueue_add_inbuf(vq, sg, 1, skb, GFP_ATOMIC);
-	if (WARN(err, "virtqueue_add_inbuf returned %d\n", err))
-		nlmsg_free(skb);
-	else
-		virtqueue_kick(vq);
-	schedule_work(&hwsim_virtio_rx);
-
-out_unlock:
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-}
-
-static void hwsim_virtio_rx_done(struct virtqueue *vq)
-{
-	schedule_work(&hwsim_virtio_rx);
-}
-
-static int init_vqs(struct virtio_device *vdev)
-{
-	vq_callback_t *callbacks[HWSIM_NUM_VQS] = {
-		[HWSIM_VQ_TX] = hwsim_virtio_tx_done,
-		[HWSIM_VQ_RX] = hwsim_virtio_rx_done,
-	};
-	const char *names[HWSIM_NUM_VQS] = {
-		[HWSIM_VQ_TX] = "tx",
-		[HWSIM_VQ_RX] = "rx",
-	};
-
-	return virtio_find_vqs(vdev, HWSIM_NUM_VQS,
-			       hwsim_vqs, callbacks, names, NULL);
-}
-
-static int fill_vq(struct virtqueue *vq)
-{
-	int i, err;
-	struct sk_buff *skb;
-	struct scatterlist sg[1];
-
-	for (i = 0; i < virtqueue_get_vring_size(vq); i++) {
-		skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
-		if (!skb)
-			return -ENOMEM;
-
-		sg_init_one(sg, skb->head, skb_end_offset(skb));
-		err = virtqueue_add_inbuf(vq, sg, 1, skb, GFP_KERNEL);
-		if (err) {
-			nlmsg_free(skb);
-			return err;
-		}
-	}
-	virtqueue_kick(vq);
-	return 0;
-}
-
-static void remove_vqs(struct virtio_device *vdev)
-{
-	int i;
-
-	virtio_reset_device(vdev);
-
-	for (i = 0; i < ARRAY_SIZE(hwsim_vqs); i++) {
-		struct virtqueue *vq = hwsim_vqs[i];
-		struct sk_buff *skb;
-
-		while ((skb = virtqueue_detach_unused_buf(vq)))
-			nlmsg_free(skb);
-	}
-
-	vdev->config->del_vqs(vdev);
-}
-
-static int hwsim_virtio_probe(struct virtio_device *vdev)
-{
-	int err;
-	unsigned long flags;
-
-	spin_lock_irqsave(&hwsim_virtio_lock, flags);
-	if (hwsim_virtio_enabled) {
-		spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-		return -EEXIST;
-	}
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-
-	err = init_vqs(vdev);
-	if (err)
-		return err;
-
-	virtio_device_ready(vdev);
-
-	err = fill_vq(hwsim_vqs[HWSIM_VQ_RX]);
-	if (err)
-		goto out_remove;
-
-	spin_lock_irqsave(&hwsim_virtio_lock, flags);
-	hwsim_virtio_enabled = true;
-	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
-
-	schedule_work(&hwsim_virtio_rx);
-	return 0;
-
-out_remove:
-	remove_vqs(vdev);
-	return err;
-}
-
-static void hwsim_virtio_remove(struct virtio_device *vdev)
-{
-	hwsim_virtio_enabled = false;
-
-	cancel_work_sync(&hwsim_virtio_rx);
-
-	remove_vqs(vdev);
-}
-
-/* MAC80211_HWSIM virtio device id table */
-static const struct virtio_device_id id_table[] = {
-	{ VIRTIO_ID_MAC80211_HWSIM, VIRTIO_DEV_ANY_ID },
-	{ 0 }
-};
-MODULE_DEVICE_TABLE(virtio, id_table);
-
-static struct virtio_driver virtio_hwsim = {
-	.driver.name = KBUILD_MODNAME,
-	.driver.owner = THIS_MODULE,
-	.id_table = id_table,
-	.probe = hwsim_virtio_probe,
-	.remove = hwsim_virtio_remove,
-};
-
-static int hwsim_register_virtio_driver(void)
-{
-	return register_virtio_driver(&virtio_hwsim);
-}
-
-static void hwsim_unregister_virtio_driver(void)
-{
-	unregister_virtio_driver(&virtio_hwsim);
-}
-#else
-static inline int hwsim_register_virtio_driver(void)
-{
-	return 0;
-}
-
-static inline void hwsim_unregister_virtio_driver(void)
-{
-}
-#endif
-
-static int __init init_mac80211_hwsim(void)
-{
-	int i, err;
-
-	if (radios < 0 || radios > 100)
-		return -EINVAL;
-
-	if (channels < 1)
-		return -EINVAL;
-
-	err = rhashtable_init(&hwsim_radios_rht, &hwsim_rht_params);
-	if (err)
-		return err;
-
-	err = register_pernet_device(&hwsim_net_ops);
-	if (err)
-		goto out_free_rht;
-
-	err = platform_driver_register(&mac80211_hwsim_driver);
-	if (err)
-		goto out_unregister_pernet;
-
-	err = hwsim_init_netlink();
-	if (err)
-		goto out_unregister_driver;
-
-	err = hwsim_register_virtio_driver();
-	if (err)
-		goto out_exit_netlink;
-
-	hwsim_class = class_create(THIS_MODULE, "mac80211_hwsim");
-	if (IS_ERR(hwsim_class)) {
-		err = PTR_ERR(hwsim_class);
-		goto out_exit_virtio;
-	}
-
-	hwsim_init_s1g_channels(hwsim_channels_s1g);
-
-	for (i = 0; i < radios; i++) {
-		struct hwsim_new_radio_params param = { 0 };
-
-		param.channels = channels;
-
-		switch (regtest) {
-		case HWSIM_REGTEST_DIFF_COUNTRY:
-			if (i < ARRAY_SIZE(hwsim_alpha2s))
-				param.reg_alpha2 = hwsim_alpha2s[i];
-			break;
-		case HWSIM_REGTEST_DRIVER_REG_FOLLOW:
-			if (!i)
-				param.reg_alpha2 = hwsim_alpha2s[0];
-			break;
-		case HWSIM_REGTEST_STRICT_ALL:
-			param.reg_strict = true;
-			fallthrough;
-		case HWSIM_REGTEST_DRIVER_REG_ALL:
-			param.reg_alpha2 = hwsim_alpha2s[0];
-			break;
-		case HWSIM_REGTEST_WORLD_ROAM:
-			if (i == 0)
-				param.regd = &hwsim_world_regdom_custom_01;
-			break;
-		case HWSIM_REGTEST_CUSTOM_WORLD:
-			param.regd = &hwsim_world_regdom_custom_01;
-			break;
-		case HWSIM_REGTEST_CUSTOM_WORLD_2:
-			if (i == 0)
-				param.regd = &hwsim_world_regdom_custom_01;
-			else if (i == 1)
-				param.regd = &hwsim_world_regdom_custom_02;
-			break;
-		case HWSIM_REGTEST_STRICT_FOLLOW:
-			if (i == 0) {
-				param.reg_strict = true;
-				param.reg_alpha2 = hwsim_alpha2s[0];
-			}
-			break;
-		case HWSIM_REGTEST_STRICT_AND_DRIVER_REG:
-			if (i == 0) {
-				param.reg_strict = true;
-				param.reg_alpha2 = hwsim_alpha2s[0];
-			} else if (i == 1) {
-				param.reg_alpha2 = hwsim_alpha2s[1];
-			}
-			break;
-		case HWSIM_REGTEST_ALL:
-			switch (i) {
-			case 0:
-				param.regd = &hwsim_world_regdom_custom_01;
-				break;
-			case 1:
-				param.regd = &hwsim_world_regdom_custom_02;
-				break;
-			case 2:
-				param.reg_alpha2 = hwsim_alpha2s[0];
-				break;
-			case 3:
-				param.reg_alpha2 = hwsim_alpha2s[1];
-				break;
-			case 4:
-				param.reg_strict = true;
-				param.reg_alpha2 = hwsim_alpha2s[2];
-				break;
-			}
-			break;
-		default:
-			break;
-		}
-
-		param.p2p_device = support_p2p_device;
-		param.mlo = mlo;
-		param.use_chanctx = channels > 1 || mlo;
-		param.iftypes = HWSIM_IFTYPE_SUPPORT_MASK;
-		if (param.p2p_device)
-			param.iftypes |= BIT(NL80211_IFTYPE_P2P_DEVICE);
-
-		err = mac80211_hwsim_new_radio(NULL, &param);
-		if (err < 0)
-			goto out_free_radios;
-	}
-
-	hwsim_mon = alloc_netdev(0, "hwsim%d", NET_NAME_UNKNOWN,
-				 hwsim_mon_setup);
-	if (hwsim_mon == NULL) {
-		err = -ENOMEM;
-		goto out_free_radios;
-	}
-
-	rtnl_lock();
-	err = dev_alloc_name(hwsim_mon, hwsim_mon->name);
-	if (err < 0) {
-		rtnl_unlock();
-		goto out_free_mon;
-	}
-
-	err = register_netdevice(hwsim_mon);
-	if (err < 0) {
-		rtnl_unlock();
-		goto out_free_mon;
-	}
-	rtnl_unlock();
-
-	return 0;
-
-out_free_mon:
-	free_netdev(hwsim_mon);
-out_free_radios:
-	mac80211_hwsim_free();
-out_exit_virtio:
-	hwsim_unregister_virtio_driver();
-out_exit_netlink:
-	hwsim_exit_netlink();
-out_unregister_driver:
-	platform_driver_unregister(&mac80211_hwsim_driver);
-out_unregister_pernet:
-	unregister_pernet_device(&hwsim_net_ops);
-out_free_rht:
-	rhashtable_destroy(&hwsim_radios_rht);
-	return err;
-}
-module_init(init_mac80211_hwsim);
-
-static void __exit exit_mac80211_hwsim(void)
-{
-	pr_debug("mac80211_hwsim: unregister radios\n");
-
-	hwsim_unregister_virtio_driver();
-	hwsim_exit_netlink();
-
-	mac80211_hwsim_free();
-
-	rhashtable_destroy(&hwsim_radios_rht);
-	unregister_netdev(hwsim_mon);
-	platform_driver_unregister(&mac80211_hwsim_driver);
-	unregister_pernet_device(&hwsim_net_ops);
-}
-module_exit(exit_mac80211_hwsim);
diff --git a/drivers/net/wireless/mac80211_hwsim.h b/drivers/net/wireless/mac80211_hwsim.h
deleted file mode 100644
index 527799b2de0f..000000000000
--- a/drivers/net/wireless/mac80211_hwsim.h
+++ /dev/null
@@ -1,280 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * mac80211_hwsim - software simulator of 802.11 radio(s) for mac80211
- * Copyright (c) 2008, Jouni Malinen <j@w1.fi>
- * Copyright (c) 2011, Javier Lopez <jlopex@gmail.com>
- * Copyright (C) 2020, 2022 Intel Corporation
- */
-
-#ifndef __MAC80211_HWSIM_H
-#define __MAC80211_HWSIM_H
-
-/**
- * enum hwsim_tx_control_flags - flags to describe transmission info/status
- *
- * These flags are used to give the wmediumd extra information in order to
- * modify its behavior for each frame
- *
- * @HWSIM_TX_CTL_REQ_TX_STATUS: require TX status callback for this frame.
- * @HWSIM_TX_CTL_NO_ACK: tell the wmediumd not to wait for an ack
- * @HWSIM_TX_STAT_ACK: Frame was acknowledged
- *
- */
-enum hwsim_tx_control_flags {
-	HWSIM_TX_CTL_REQ_TX_STATUS		= BIT(0),
-	HWSIM_TX_CTL_NO_ACK			= BIT(1),
-	HWSIM_TX_STAT_ACK			= BIT(2),
-};
-
-/**
- * DOC: Frame transmission/registration support
- *
- * Frame transmission and registration support exists to allow userspace
- * entities such as wmediumd to receive and process all broadcasted
- * frames from a mac80211_hwsim radio device.
- *
- * This allow user space applications to decide if the frame should be
- * dropped or not and implement a wireless medium simulator at user space.
- *
- * Registration is done by sending a register message to the driver and
- * will be automatically unregistered if the user application doesn't
- * responds to sent frames.
- * Once registered the user application has to take responsibility of
- * broadcasting the frames to all listening mac80211_hwsim radio
- * interfaces.
- *
- * For more technical details, see the corresponding command descriptions
- * below.
- */
-
-/**
- * enum hwsim_commands - supported hwsim commands
- *
- * @HWSIM_CMD_UNSPEC: unspecified command to catch errors
- *
- * @HWSIM_CMD_REGISTER: request to register and received all broadcasted
- *	frames by any mac80211_hwsim radio device.
- * @HWSIM_CMD_FRAME: send/receive a broadcasted frame from/to kernel/user
- *	space, uses:
- *	%HWSIM_ATTR_ADDR_TRANSMITTER, %HWSIM_ATTR_ADDR_RECEIVER,
- *	%HWSIM_ATTR_FRAME, %HWSIM_ATTR_FLAGS, %HWSIM_ATTR_RX_RATE,
- *	%HWSIM_ATTR_SIGNAL, %HWSIM_ATTR_COOKIE, %HWSIM_ATTR_FREQ (optional)
- * @HWSIM_CMD_TX_INFO_FRAME: Transmission info report from user space to
- *	kernel, uses:
- *	%HWSIM_ATTR_ADDR_TRANSMITTER, %HWSIM_ATTR_FLAGS,
- *	%HWSIM_ATTR_TX_INFO, %WSIM_ATTR_TX_INFO_FLAGS,
- *	%HWSIM_ATTR_SIGNAL, %HWSIM_ATTR_COOKIE
- * @HWSIM_CMD_NEW_RADIO: create a new radio with the given parameters,
- *	returns the radio ID (>= 0) or negative on errors, if successful
- *	then multicast the result, uses optional parameter:
- *	%HWSIM_ATTR_REG_STRICT_REG, %HWSIM_ATTR_SUPPORT_P2P_DEVICE,
- *	%HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE, %HWSIM_ATTR_CHANNELS,
- *	%HWSIM_ATTR_NO_VIF, %HWSIM_ATTR_RADIO_NAME, %HWSIM_ATTR_USE_CHANCTX,
- *	%HWSIM_ATTR_REG_HINT_ALPHA2, %HWSIM_ATTR_REG_CUSTOM_REG,
- *	%HWSIM_ATTR_PERM_ADDR
- * @HWSIM_CMD_DEL_RADIO: destroy a radio, reply is multicasted
- * @HWSIM_CMD_GET_RADIO: fetch information about existing radios, uses:
- *	%HWSIM_ATTR_RADIO_ID
- * @HWSIM_CMD_ADD_MAC_ADDR: add a receive MAC address (given in the
- *	%HWSIM_ATTR_ADDR_RECEIVER attribute) to a device identified by
- *	%HWSIM_ATTR_ADDR_TRANSMITTER. This lets wmediumd forward frames
- *	to this receiver address for a given station.
- * @HWSIM_CMD_DEL_MAC_ADDR: remove the MAC address again, the attributes
- *	are the same as to @HWSIM_CMD_ADD_MAC_ADDR.
- * @__HWSIM_CMD_MAX: enum limit
- */
-enum {
-	HWSIM_CMD_UNSPEC,
-	HWSIM_CMD_REGISTER,
-	HWSIM_CMD_FRAME,
-	HWSIM_CMD_TX_INFO_FRAME,
-	HWSIM_CMD_NEW_RADIO,
-	HWSIM_CMD_DEL_RADIO,
-	HWSIM_CMD_GET_RADIO,
-	HWSIM_CMD_ADD_MAC_ADDR,
-	HWSIM_CMD_DEL_MAC_ADDR,
-	__HWSIM_CMD_MAX,
-};
-#define HWSIM_CMD_MAX (_HWSIM_CMD_MAX - 1)
-
-#define HWSIM_CMD_CREATE_RADIO   HWSIM_CMD_NEW_RADIO
-#define HWSIM_CMD_DESTROY_RADIO  HWSIM_CMD_DEL_RADIO
-
-/**
- * enum hwsim_attrs - hwsim netlink attributes
- *
- * @HWSIM_ATTR_UNSPEC: unspecified attribute to catch errors
- *
- * @HWSIM_ATTR_ADDR_RECEIVER: MAC address of the radio device that
- *	the frame is broadcasted to
- * @HWSIM_ATTR_ADDR_TRANSMITTER: MAC address of the radio device that
- *	the frame was broadcasted from
- * @HWSIM_ATTR_FRAME: Data array
- * @HWSIM_ATTR_FLAGS: mac80211 transmission flags, used to process
-	properly the frame at user space
- * @HWSIM_ATTR_RX_RATE: estimated rx rate index for this frame at user
-	space
- * @HWSIM_ATTR_SIGNAL: estimated RX signal for this frame at user
-	space
- * @HWSIM_ATTR_TX_INFO: ieee80211_tx_rate array
- * @HWSIM_ATTR_COOKIE: sk_buff cookie to identify the frame
- * @HWSIM_ATTR_CHANNELS: u32 attribute used with the %HWSIM_CMD_CREATE_RADIO
- *	command giving the number of channels supported by the new radio
- * @HWSIM_ATTR_RADIO_ID: u32 attribute used with %HWSIM_CMD_DESTROY_RADIO
- *	only to destroy a radio
- * @HWSIM_ATTR_REG_HINT_ALPHA2: alpha2 for regulatoro driver hint
- *	(nla string, length 2)
- * @HWSIM_ATTR_REG_CUSTOM_REG: custom regulatory domain index (u32 attribute)
- * @HWSIM_ATTR_REG_STRICT_REG: request REGULATORY_STRICT_REG (flag attribute)
- * @HWSIM_ATTR_SUPPORT_P2P_DEVICE: support P2P Device virtual interface (flag)
- * @HWSIM_ATTR_USE_CHANCTX: used with the %HWSIM_CMD_CREATE_RADIO
- *	command to force use of channel contexts even when only a
- *	single channel is supported
- * @HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE: used with the %HWSIM_CMD_CREATE_RADIO
- *	command to force radio removal when process that created the radio dies
- * @HWSIM_ATTR_RADIO_NAME: Name of radio, e.g. phy666
- * @HWSIM_ATTR_NO_VIF:  Do not create vif (wlanX) when creating radio.
- * @HWSIM_ATTR_FREQ: Frequency at which packet is transmitted or received.
- * @HWSIM_ATTR_TX_INFO_FLAGS: additional flags for corresponding
- *	rates of %HWSIM_ATTR_TX_INFO
- * @HWSIM_ATTR_PERM_ADDR: permanent mac address of new radio
- * @HWSIM_ATTR_IFTYPE_SUPPORT: u32 attribute of supported interface types bits
- * @HWSIM_ATTR_CIPHER_SUPPORT: u32 array of supported cipher types
- * @HWSIM_ATTR_MLO_SUPPORT: claim MLO support (exact parameters TBD) for
- *	the new radio
- * @__HWSIM_ATTR_MAX: enum limit
- */
-
-
-enum {
-	HWSIM_ATTR_UNSPEC,
-	HWSIM_ATTR_ADDR_RECEIVER,
-	HWSIM_ATTR_ADDR_TRANSMITTER,
-	HWSIM_ATTR_FRAME,
-	HWSIM_ATTR_FLAGS,
-	HWSIM_ATTR_RX_RATE,
-	HWSIM_ATTR_SIGNAL,
-	HWSIM_ATTR_TX_INFO,
-	HWSIM_ATTR_COOKIE,
-	HWSIM_ATTR_CHANNELS,
-	HWSIM_ATTR_RADIO_ID,
-	HWSIM_ATTR_REG_HINT_ALPHA2,
-	HWSIM_ATTR_REG_CUSTOM_REG,
-	HWSIM_ATTR_REG_STRICT_REG,
-	HWSIM_ATTR_SUPPORT_P2P_DEVICE,
-	HWSIM_ATTR_USE_CHANCTX,
-	HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE,
-	HWSIM_ATTR_RADIO_NAME,
-	HWSIM_ATTR_NO_VIF,
-	HWSIM_ATTR_FREQ,
-	HWSIM_ATTR_PAD,
-	HWSIM_ATTR_TX_INFO_FLAGS,
-	HWSIM_ATTR_PERM_ADDR,
-	HWSIM_ATTR_IFTYPE_SUPPORT,
-	HWSIM_ATTR_CIPHER_SUPPORT,
-	HWSIM_ATTR_MLO_SUPPORT,
-	__HWSIM_ATTR_MAX,
-};
-#define HWSIM_ATTR_MAX (__HWSIM_ATTR_MAX - 1)
-
-/**
- * struct hwsim_tx_rate - rate selection/status
- *
- * @idx: rate index to attempt to send with
- * @count: number of tries in this rate before going to the next rate
- *
- * A value of -1 for @idx indicates an invalid rate and, if used
- * in an array of retry rates, that no more rates should be tried.
- *
- * When used for transmit status reporting, the driver should
- * always report the rate and number of retries used.
- *
- */
-struct hwsim_tx_rate {
-	s8 idx;
-	u8 count;
-} __packed;
-
-/**
- * enum hwsim_tx_rate_flags - per-rate flags set by the rate control algorithm.
- *	Inspired by structure mac80211_rate_control_flags. New flags may be
- *	appended, but old flags not deleted, to keep compatibility for
- *	userspace.
- *
- * These flags are set by the Rate control algorithm for each rate during tx,
- * in the @flags member of struct ieee80211_tx_rate.
- *
- * @MAC80211_HWSIM_TX_RC_USE_RTS_CTS: Use RTS/CTS exchange for this rate.
- * @MAC80211_HWSIM_TX_RC_USE_CTS_PROTECT: CTS-to-self protection is required.
- *	This is set if the current BSS requires ERP protection.
- * @MAC80211_HWSIM_TX_RC_USE_SHORT_PREAMBLE: Use short preamble.
- * @MAC80211_HWSIM_TX_RC_MCS: HT rate.
- * @MAC80211_HWSIM_TX_RC_VHT_MCS: VHT MCS rate, in this case the idx field is
- *	split into a higher 4 bits (Nss) and lower 4 bits (MCS number)
- * @MAC80211_HWSIM_TX_RC_GREEN_FIELD: Indicates whether this rate should be used
- *	in Greenfield mode.
- * @MAC80211_HWSIM_TX_RC_40_MHZ_WIDTH: Indicates if the Channel Width should be
- *	40 MHz.
- * @MAC80211_HWSIM_TX_RC_80_MHZ_WIDTH: Indicates 80 MHz transmission
- * @MAC80211_HWSIM_TX_RC_160_MHZ_WIDTH: Indicates 160 MHz transmission
- *	(80+80 isn't supported yet)
- * @MAC80211_HWSIM_TX_RC_DUP_DATA: The frame should be transmitted on both of
- *	the adjacent 20 MHz channels, if the current channel type is
- *	NL80211_CHAN_HT40MINUS or NL80211_CHAN_HT40PLUS.
- * @MAC80211_HWSIM_TX_RC_SHORT_GI: Short Guard interval should be used for this
- *	rate.
- */
-enum hwsim_tx_rate_flags {
-	MAC80211_HWSIM_TX_RC_USE_RTS_CTS		= BIT(0),
-	MAC80211_HWSIM_TX_RC_USE_CTS_PROTECT		= BIT(1),
-	MAC80211_HWSIM_TX_RC_USE_SHORT_PREAMBLE	= BIT(2),
-
-	/* rate index is an HT/VHT MCS instead of an index */
-	MAC80211_HWSIM_TX_RC_MCS			= BIT(3),
-	MAC80211_HWSIM_TX_RC_GREEN_FIELD		= BIT(4),
-	MAC80211_HWSIM_TX_RC_40_MHZ_WIDTH		= BIT(5),
-	MAC80211_HWSIM_TX_RC_DUP_DATA		= BIT(6),
-	MAC80211_HWSIM_TX_RC_SHORT_GI		= BIT(7),
-	MAC80211_HWSIM_TX_RC_VHT_MCS			= BIT(8),
-	MAC80211_HWSIM_TX_RC_80_MHZ_WIDTH		= BIT(9),
-	MAC80211_HWSIM_TX_RC_160_MHZ_WIDTH		= BIT(10),
-};
-
-/**
- * struct hwsim_tx_rate - rate selection/status
- *
- * @idx: rate index to attempt to send with
- * @count: number of tries in this rate before going to the next rate
- *
- * A value of -1 for @idx indicates an invalid rate and, if used
- * in an array of retry rates, that no more rates should be tried.
- *
- * When used for transmit status reporting, the driver should
- * always report the rate and number of retries used.
- *
- */
-struct hwsim_tx_rate_flag {
-	s8 idx;
-	u16 flags;
-} __packed;
-
-/**
- * DOC: Frame transmission support over virtio
- *
- * Frame transmission is also supported over virtio to allow communication
- * with external entities.
- */
-
-/**
- * enum hwsim_vqs - queues for virtio frame transmission
- *
- * @HWSIM_VQ_TX: send frames to external entity
- * @HWSIM_VQ_RX: receive frames and transmission info reports
- * @HWSIM_NUM_VQS: enum limit
- */
-enum {
-	HWSIM_VQ_TX,
-	HWSIM_VQ_RX,
-	HWSIM_NUM_VQS,
-};
-#endif /* __MAC80211_HWSIM_H */
diff --git a/drivers/net/wireless/virt_wifi.c b/drivers/net/wireless/virt_wifi.c
deleted file mode 100644
index ba14d83353a4..000000000000
--- a/drivers/net/wireless/virt_wifi.c
+++ /dev/null
@@ -1,697 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* drivers/net/wireless/virt_wifi.c
- *
- * A fake implementation of cfg80211_ops that can be tacked on to an ethernet
- * net_device to make it appear as a wireless connection.
- *
- * Copyright (C) 2018 Google, Inc.
- *
- * Author: schuffelen@google.com
- */
-
-#include <net/cfg80211.h>
-#include <net/rtnetlink.h>
-#include <linux/etherdevice.h>
-#include <linux/math64.h>
-#include <linux/module.h>
-
-static struct wiphy *common_wiphy;
-
-struct virt_wifi_wiphy_priv {
-	struct delayed_work scan_result;
-	struct cfg80211_scan_request *scan_request;
-	bool being_deleted;
-};
-
-static struct ieee80211_channel channel_2ghz = {
-	.band = NL80211_BAND_2GHZ,
-	.center_freq = 2432,
-	.hw_value = 2432,
-	.max_power = 20,
-};
-
-static struct ieee80211_rate bitrates_2ghz[] = {
-	{ .bitrate = 10 },
-	{ .bitrate = 20 },
-	{ .bitrate = 55 },
-	{ .bitrate = 110 },
-	{ .bitrate = 60 },
-	{ .bitrate = 120 },
-	{ .bitrate = 240 },
-};
-
-static struct ieee80211_supported_band band_2ghz = {
-	.channels = &channel_2ghz,
-	.bitrates = bitrates_2ghz,
-	.band = NL80211_BAND_2GHZ,
-	.n_channels = 1,
-	.n_bitrates = ARRAY_SIZE(bitrates_2ghz),
-	.ht_cap = {
-		.ht_supported = true,
-		.cap = IEEE80211_HT_CAP_SUP_WIDTH_20_40 |
-		       IEEE80211_HT_CAP_GRN_FLD |
-		       IEEE80211_HT_CAP_SGI_20 |
-		       IEEE80211_HT_CAP_SGI_40 |
-		       IEEE80211_HT_CAP_DSSSCCK40,
-		.ampdu_factor = 0x3,
-		.ampdu_density = 0x6,
-		.mcs = {
-			.rx_mask = {0xff, 0xff},
-			.tx_params = IEEE80211_HT_MCS_TX_DEFINED,
-		},
-	},
-};
-
-static struct ieee80211_channel channel_5ghz = {
-	.band = NL80211_BAND_5GHZ,
-	.center_freq = 5240,
-	.hw_value = 5240,
-	.max_power = 20,
-};
-
-static struct ieee80211_rate bitrates_5ghz[] = {
-	{ .bitrate = 60 },
-	{ .bitrate = 120 },
-	{ .bitrate = 240 },
-};
-
-#define RX_MCS_MAP (IEEE80211_VHT_MCS_SUPPORT_0_9 << 0 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 2 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 4 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 6 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 8 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 10 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 12 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 14)
-
-#define TX_MCS_MAP (IEEE80211_VHT_MCS_SUPPORT_0_9 << 0 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 2 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 4 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 6 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 8 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 10 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 12 | \
-		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 14)
-
-static struct ieee80211_supported_band band_5ghz = {
-	.channels = &channel_5ghz,
-	.bitrates = bitrates_5ghz,
-	.band = NL80211_BAND_5GHZ,
-	.n_channels = 1,
-	.n_bitrates = ARRAY_SIZE(bitrates_5ghz),
-	.ht_cap = {
-		.ht_supported = true,
-		.cap = IEEE80211_HT_CAP_SUP_WIDTH_20_40 |
-		       IEEE80211_HT_CAP_GRN_FLD |
-		       IEEE80211_HT_CAP_SGI_20 |
-		       IEEE80211_HT_CAP_SGI_40 |
-		       IEEE80211_HT_CAP_DSSSCCK40,
-		.ampdu_factor = 0x3,
-		.ampdu_density = 0x6,
-		.mcs = {
-			.rx_mask = {0xff, 0xff},
-			.tx_params = IEEE80211_HT_MCS_TX_DEFINED,
-		},
-	},
-	.vht_cap = {
-		.vht_supported = true,
-		.cap = IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454 |
-		       IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ |
-		       IEEE80211_VHT_CAP_RXLDPC |
-		       IEEE80211_VHT_CAP_SHORT_GI_80 |
-		       IEEE80211_VHT_CAP_SHORT_GI_160 |
-		       IEEE80211_VHT_CAP_TXSTBC |
-		       IEEE80211_VHT_CAP_RXSTBC_1 |
-		       IEEE80211_VHT_CAP_RXSTBC_2 |
-		       IEEE80211_VHT_CAP_RXSTBC_3 |
-		       IEEE80211_VHT_CAP_RXSTBC_4 |
-		       IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK,
-		.vht_mcs = {
-			.rx_mcs_map = cpu_to_le16(RX_MCS_MAP),
-			.tx_mcs_map = cpu_to_le16(TX_MCS_MAP),
-		}
-	},
-};
-
-/* Assigned at module init. Guaranteed locally-administered and unicast. */
-static u8 fake_router_bssid[ETH_ALEN] __ro_after_init = {};
-
-static void virt_wifi_inform_bss(struct wiphy *wiphy)
-{
-	u64 tsf = div_u64(ktime_get_boottime_ns(), 1000);
-	struct cfg80211_bss *informed_bss;
-	static const struct {
-		u8 tag;
-		u8 len;
-		u8 ssid[8];
-	} __packed ssid = {
-		.tag = WLAN_EID_SSID,
-		.len = 8,
-		.ssid = "VirtWifi",
-	};
-
-	informed_bss = cfg80211_inform_bss(wiphy, &channel_5ghz,
-					   CFG80211_BSS_FTYPE_PRESP,
-					   fake_router_bssid, tsf,
-					   WLAN_CAPABILITY_ESS, 0,
-					   (void *)&ssid, sizeof(ssid),
-					   DBM_TO_MBM(-50), GFP_KERNEL);
-	cfg80211_put_bss(wiphy, informed_bss);
-}
-
-/* Called with the rtnl lock held. */
-static int virt_wifi_scan(struct wiphy *wiphy,
-			  struct cfg80211_scan_request *request)
-{
-	struct virt_wifi_wiphy_priv *priv = wiphy_priv(wiphy);
-
-	wiphy_debug(wiphy, "scan\n");
-
-	if (priv->scan_request || priv->being_deleted)
-		return -EBUSY;
-
-	priv->scan_request = request;
-	schedule_delayed_work(&priv->scan_result, HZ * 2);
-
-	return 0;
-}
-
-/* Acquires and releases the rdev BSS lock. */
-static void virt_wifi_scan_result(struct work_struct *work)
-{
-	struct virt_wifi_wiphy_priv *priv =
-		container_of(work, struct virt_wifi_wiphy_priv,
-			     scan_result.work);
-	struct wiphy *wiphy = priv_to_wiphy(priv);
-	struct cfg80211_scan_info scan_info = { .aborted = false };
-
-	virt_wifi_inform_bss(wiphy);
-
-	/* Schedules work which acquires and releases the rtnl lock. */
-	cfg80211_scan_done(priv->scan_request, &scan_info);
-	priv->scan_request = NULL;
-}
-
-/* May acquire and release the rdev BSS lock. */
-static void virt_wifi_cancel_scan(struct wiphy *wiphy)
-{
-	struct virt_wifi_wiphy_priv *priv = wiphy_priv(wiphy);
-
-	cancel_delayed_work_sync(&priv->scan_result);
-	/* Clean up dangling callbacks if necessary. */
-	if (priv->scan_request) {
-		struct cfg80211_scan_info scan_info = { .aborted = true };
-		/* Schedules work which acquires and releases the rtnl lock. */
-		cfg80211_scan_done(priv->scan_request, &scan_info);
-		priv->scan_request = NULL;
-	}
-}
-
-struct virt_wifi_netdev_priv {
-	struct delayed_work connect;
-	struct net_device *lowerdev;
-	struct net_device *upperdev;
-	u32 tx_packets;
-	u32 tx_failed;
-	u8 connect_requested_bss[ETH_ALEN];
-	bool is_up;
-	bool is_connected;
-	bool being_deleted;
-};
-
-/* Called with the rtnl lock held. */
-static int virt_wifi_connect(struct wiphy *wiphy, struct net_device *netdev,
-			     struct cfg80211_connect_params *sme)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(netdev);
-	bool could_schedule;
-
-	if (priv->being_deleted || !priv->is_up)
-		return -EBUSY;
-
-	could_schedule = schedule_delayed_work(&priv->connect, HZ * 2);
-	if (!could_schedule)
-		return -EBUSY;
-
-	if (sme->bssid) {
-		ether_addr_copy(priv->connect_requested_bss, sme->bssid);
-	} else {
-		virt_wifi_inform_bss(wiphy);
-		eth_zero_addr(priv->connect_requested_bss);
-	}
-
-	wiphy_debug(wiphy, "connect\n");
-
-	return 0;
-}
-
-/* Acquires and releases the rdev event lock. */
-static void virt_wifi_connect_complete(struct work_struct *work)
-{
-	struct virt_wifi_netdev_priv *priv =
-		container_of(work, struct virt_wifi_netdev_priv, connect.work);
-	u8 *requested_bss = priv->connect_requested_bss;
-	bool right_addr = ether_addr_equal(requested_bss, fake_router_bssid);
-	u16 status = WLAN_STATUS_SUCCESS;
-
-	if (is_zero_ether_addr(requested_bss))
-		requested_bss = NULL;
-
-	if (!priv->is_up || (requested_bss && !right_addr))
-		status = WLAN_STATUS_UNSPECIFIED_FAILURE;
-	else
-		priv->is_connected = true;
-
-	/* Schedules an event that acquires the rtnl lock. */
-	cfg80211_connect_result(priv->upperdev, requested_bss, NULL, 0, NULL, 0,
-				status, GFP_KERNEL);
-	netif_carrier_on(priv->upperdev);
-}
-
-/* May acquire and release the rdev event lock. */
-static void virt_wifi_cancel_connect(struct net_device *netdev)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(netdev);
-
-	/* If there is work pending, clean up dangling callbacks. */
-	if (cancel_delayed_work_sync(&priv->connect)) {
-		/* Schedules an event that acquires the rtnl lock. */
-		cfg80211_connect_result(priv->upperdev,
-					priv->connect_requested_bss, NULL, 0,
-					NULL, 0,
-					WLAN_STATUS_UNSPECIFIED_FAILURE,
-					GFP_KERNEL);
-	}
-}
-
-/* Called with the rtnl lock held. Acquires the rdev event lock. */
-static int virt_wifi_disconnect(struct wiphy *wiphy, struct net_device *netdev,
-				u16 reason_code)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(netdev);
-
-	if (priv->being_deleted)
-		return -EBUSY;
-
-	wiphy_debug(wiphy, "disconnect\n");
-	virt_wifi_cancel_connect(netdev);
-
-	cfg80211_disconnected(netdev, reason_code, NULL, 0, true, GFP_KERNEL);
-	priv->is_connected = false;
-	netif_carrier_off(netdev);
-
-	return 0;
-}
-
-/* Called with the rtnl lock held. */
-static int virt_wifi_get_station(struct wiphy *wiphy, struct net_device *dev,
-				 const u8 *mac, struct station_info *sinfo)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-
-	wiphy_debug(wiphy, "get_station\n");
-
-	if (!priv->is_connected || !ether_addr_equal(mac, fake_router_bssid))
-		return -ENOENT;
-
-	sinfo->filled = BIT_ULL(NL80211_STA_INFO_TX_PACKETS) |
-		BIT_ULL(NL80211_STA_INFO_TX_FAILED) |
-		BIT_ULL(NL80211_STA_INFO_SIGNAL) |
-		BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
-	sinfo->tx_packets = priv->tx_packets;
-	sinfo->tx_failed = priv->tx_failed;
-	/* For CFG80211_SIGNAL_TYPE_MBM, value is expressed in _dBm_ */
-	sinfo->signal = -50;
-	sinfo->txrate = (struct rate_info) {
-		.legacy = 10, /* units are 100kbit/s */
-	};
-	return 0;
-}
-
-/* Called with the rtnl lock held. */
-static int virt_wifi_dump_station(struct wiphy *wiphy, struct net_device *dev,
-				  int idx, u8 *mac, struct station_info *sinfo)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-
-	wiphy_debug(wiphy, "dump_station\n");
-
-	if (idx != 0 || !priv->is_connected)
-		return -ENOENT;
-
-	ether_addr_copy(mac, fake_router_bssid);
-	return virt_wifi_get_station(wiphy, dev, fake_router_bssid, sinfo);
-}
-
-static const struct cfg80211_ops virt_wifi_cfg80211_ops = {
-	.scan = virt_wifi_scan,
-
-	.connect = virt_wifi_connect,
-	.disconnect = virt_wifi_disconnect,
-
-	.get_station = virt_wifi_get_station,
-	.dump_station = virt_wifi_dump_station,
-};
-
-/* Acquires and releases the rtnl lock. */
-static struct wiphy *virt_wifi_make_wiphy(void)
-{
-	struct wiphy *wiphy;
-	struct virt_wifi_wiphy_priv *priv;
-	int err;
-
-	wiphy = wiphy_new(&virt_wifi_cfg80211_ops, sizeof(*priv));
-
-	if (!wiphy)
-		return NULL;
-
-	wiphy->max_scan_ssids = 4;
-	wiphy->max_scan_ie_len = 1000;
-	wiphy->signal_type = CFG80211_SIGNAL_TYPE_MBM;
-
-	wiphy->bands[NL80211_BAND_2GHZ] = &band_2ghz;
-	wiphy->bands[NL80211_BAND_5GHZ] = &band_5ghz;
-	wiphy->bands[NL80211_BAND_60GHZ] = NULL;
-
-	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION);
-
-	priv = wiphy_priv(wiphy);
-	priv->being_deleted = false;
-	priv->scan_request = NULL;
-	INIT_DELAYED_WORK(&priv->scan_result, virt_wifi_scan_result);
-
-	err = wiphy_register(wiphy);
-	if (err < 0) {
-		wiphy_free(wiphy);
-		return NULL;
-	}
-
-	return wiphy;
-}
-
-/* Acquires and releases the rtnl lock. */
-static void virt_wifi_destroy_wiphy(struct wiphy *wiphy)
-{
-	struct virt_wifi_wiphy_priv *priv;
-
-	WARN(!wiphy, "%s called with null wiphy", __func__);
-	if (!wiphy)
-		return;
-
-	priv = wiphy_priv(wiphy);
-	priv->being_deleted = true;
-	virt_wifi_cancel_scan(wiphy);
-
-	if (wiphy->registered)
-		wiphy_unregister(wiphy);
-	wiphy_free(wiphy);
-}
-
-/* Enters and exits a RCU-bh critical section. */
-static netdev_tx_t virt_wifi_start_xmit(struct sk_buff *skb,
-					struct net_device *dev)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-
-	priv->tx_packets++;
-	if (!priv->is_connected) {
-		priv->tx_failed++;
-		return NET_XMIT_DROP;
-	}
-
-	skb->dev = priv->lowerdev;
-	return dev_queue_xmit(skb);
-}
-
-/* Called with rtnl lock held. */
-static int virt_wifi_net_device_open(struct net_device *dev)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-
-	priv->is_up = true;
-	return 0;
-}
-
-/* Called with rtnl lock held. */
-static int virt_wifi_net_device_stop(struct net_device *dev)
-{
-	struct virt_wifi_netdev_priv *n_priv = netdev_priv(dev);
-
-	n_priv->is_up = false;
-
-	if (!dev->ieee80211_ptr)
-		return 0;
-
-	virt_wifi_cancel_scan(dev->ieee80211_ptr->wiphy);
-	virt_wifi_cancel_connect(dev);
-	netif_carrier_off(dev);
-
-	return 0;
-}
-
-static int virt_wifi_net_device_get_iflink(const struct net_device *dev)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-
-	return priv->lowerdev->ifindex;
-}
-
-static const struct net_device_ops virt_wifi_ops = {
-	.ndo_start_xmit = virt_wifi_start_xmit,
-	.ndo_open	= virt_wifi_net_device_open,
-	.ndo_stop	= virt_wifi_net_device_stop,
-	.ndo_get_iflink = virt_wifi_net_device_get_iflink,
-};
-
-/* Invoked as part of rtnl lock release. */
-static void virt_wifi_net_device_destructor(struct net_device *dev)
-{
-	/* Delayed past dellink to allow nl80211 to react to the device being
-	 * deleted.
-	 */
-	kfree(dev->ieee80211_ptr);
-	dev->ieee80211_ptr = NULL;
-}
-
-/* No lock interaction. */
-static void virt_wifi_setup(struct net_device *dev)
-{
-	ether_setup(dev);
-	dev->netdev_ops = &virt_wifi_ops;
-	dev->needs_free_netdev  = true;
-}
-
-/* Called in a RCU read critical section from netif_receive_skb */
-static rx_handler_result_t virt_wifi_rx_handler(struct sk_buff **pskb)
-{
-	struct sk_buff *skb = *pskb;
-	struct virt_wifi_netdev_priv *priv =
-		rcu_dereference(skb->dev->rx_handler_data);
-
-	if (!priv->is_connected)
-		return RX_HANDLER_PASS;
-
-	/* GFP_ATOMIC because this is a packet interrupt handler. */
-	skb = skb_share_check(skb, GFP_ATOMIC);
-	if (!skb) {
-		dev_err(&priv->upperdev->dev, "can't skb_share_check\n");
-		return RX_HANDLER_CONSUMED;
-	}
-
-	*pskb = skb;
-	skb->dev = priv->upperdev;
-	skb->pkt_type = PACKET_HOST;
-	return RX_HANDLER_ANOTHER;
-}
-
-/* Called with rtnl lock held. */
-static int virt_wifi_newlink(struct net *src_net, struct net_device *dev,
-			     struct nlattr *tb[], struct nlattr *data[],
-			     struct netlink_ext_ack *extack)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-	int err;
-
-	if (!tb[IFLA_LINK])
-		return -EINVAL;
-
-	netif_carrier_off(dev);
-
-	priv->upperdev = dev;
-	priv->lowerdev = __dev_get_by_index(src_net,
-					    nla_get_u32(tb[IFLA_LINK]));
-
-	if (!priv->lowerdev)
-		return -ENODEV;
-	if (!tb[IFLA_MTU])
-		dev->mtu = priv->lowerdev->mtu;
-	else if (dev->mtu > priv->lowerdev->mtu)
-		return -EINVAL;
-
-	err = netdev_rx_handler_register(priv->lowerdev, virt_wifi_rx_handler,
-					 priv);
-	if (err) {
-		dev_err(&priv->lowerdev->dev,
-			"can't netdev_rx_handler_register: %d\n", err);
-		return err;
-	}
-
-	eth_hw_addr_inherit(dev, priv->lowerdev);
-	netif_stacked_transfer_operstate(priv->lowerdev, dev);
-
-	SET_NETDEV_DEV(dev, &priv->lowerdev->dev);
-	dev->ieee80211_ptr = kzalloc(sizeof(*dev->ieee80211_ptr), GFP_KERNEL);
-
-	if (!dev->ieee80211_ptr) {
-		err = -ENOMEM;
-		goto remove_handler;
-	}
-
-	dev->ieee80211_ptr->iftype = NL80211_IFTYPE_STATION;
-	dev->ieee80211_ptr->wiphy = common_wiphy;
-
-	err = register_netdevice(dev);
-	if (err) {
-		dev_err(&priv->lowerdev->dev, "can't register_netdevice: %d\n",
-			err);
-		goto free_wireless_dev;
-	}
-
-	err = netdev_upper_dev_link(priv->lowerdev, dev, extack);
-	if (err) {
-		dev_err(&priv->lowerdev->dev, "can't netdev_upper_dev_link: %d\n",
-			err);
-		goto unregister_netdev;
-	}
-
-	dev->priv_destructor = virt_wifi_net_device_destructor;
-	priv->being_deleted = false;
-	priv->is_connected = false;
-	priv->is_up = false;
-	INIT_DELAYED_WORK(&priv->connect, virt_wifi_connect_complete);
-	__module_get(THIS_MODULE);
-
-	return 0;
-unregister_netdev:
-	unregister_netdevice(dev);
-free_wireless_dev:
-	kfree(dev->ieee80211_ptr);
-	dev->ieee80211_ptr = NULL;
-remove_handler:
-	netdev_rx_handler_unregister(priv->lowerdev);
-
-	return err;
-}
-
-/* Called with rtnl lock held. */
-static void virt_wifi_dellink(struct net_device *dev,
-			      struct list_head *head)
-{
-	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
-
-	if (dev->ieee80211_ptr)
-		virt_wifi_cancel_scan(dev->ieee80211_ptr->wiphy);
-
-	priv->being_deleted = true;
-	virt_wifi_cancel_connect(dev);
-	netif_carrier_off(dev);
-
-	netdev_rx_handler_unregister(priv->lowerdev);
-	netdev_upper_dev_unlink(priv->lowerdev, dev);
-
-	unregister_netdevice_queue(dev, head);
-	module_put(THIS_MODULE);
-
-	/* Deleting the wiphy is handled in the module destructor. */
-}
-
-static struct rtnl_link_ops virt_wifi_link_ops = {
-	.kind		= "virt_wifi",
-	.setup		= virt_wifi_setup,
-	.newlink	= virt_wifi_newlink,
-	.dellink	= virt_wifi_dellink,
-	.priv_size	= sizeof(struct virt_wifi_netdev_priv),
-};
-
-static bool netif_is_virt_wifi_dev(const struct net_device *dev)
-{
-	return rcu_access_pointer(dev->rx_handler) == virt_wifi_rx_handler;
-}
-
-static int virt_wifi_event(struct notifier_block *this, unsigned long event,
-			   void *ptr)
-{
-	struct net_device *lower_dev = netdev_notifier_info_to_dev(ptr);
-	struct virt_wifi_netdev_priv *priv;
-	struct net_device *upper_dev;
-	LIST_HEAD(list_kill);
-
-	if (!netif_is_virt_wifi_dev(lower_dev))
-		return NOTIFY_DONE;
-
-	switch (event) {
-	case NETDEV_UNREGISTER:
-		priv = rtnl_dereference(lower_dev->rx_handler_data);
-		if (!priv)
-			return NOTIFY_DONE;
-
-		upper_dev = priv->upperdev;
-
-		upper_dev->rtnl_link_ops->dellink(upper_dev, &list_kill);
-		unregister_netdevice_many(&list_kill);
-		break;
-	}
-
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block virt_wifi_notifier = {
-	.notifier_call = virt_wifi_event,
-};
-
-/* Acquires and releases the rtnl lock. */
-static int __init virt_wifi_init_module(void)
-{
-	int err;
-
-	/* Guaranteed to be locally-administered and not multicast. */
-	eth_random_addr(fake_router_bssid);
-
-	err = register_netdevice_notifier(&virt_wifi_notifier);
-	if (err)
-		return err;
-
-	err = -ENOMEM;
-	common_wiphy = virt_wifi_make_wiphy();
-	if (!common_wiphy)
-		goto notifier;
-
-	err = rtnl_link_register(&virt_wifi_link_ops);
-	if (err)
-		goto destroy_wiphy;
-
-	return 0;
-
-destroy_wiphy:
-	virt_wifi_destroy_wiphy(common_wiphy);
-notifier:
-	unregister_netdevice_notifier(&virt_wifi_notifier);
-	return err;
-}
-
-/* Acquires and releases the rtnl lock. */
-static void __exit virt_wifi_cleanup_module(void)
-{
-	/* Will delete any devices that depend on the wiphy. */
-	rtnl_link_unregister(&virt_wifi_link_ops);
-	virt_wifi_destroy_wiphy(common_wiphy);
-	unregister_netdevice_notifier(&virt_wifi_notifier);
-}
-
-module_init(virt_wifi_init_module);
-module_exit(virt_wifi_cleanup_module);
-
-MODULE_LICENSE("GPL v2");
-MODULE_AUTHOR("Cody Schuffelen <schuffelen@google.com>");
-MODULE_DESCRIPTION("Driver for a wireless wrapper of ethernet devices");
-MODULE_ALIAS_RTNL_LINK("virt_wifi");
diff --git a/drivers/net/wireless/virtual/Kconfig b/drivers/net/wireless/virtual/Kconfig
new file mode 100644
index 000000000000..fb3b4b69f26b
--- /dev/null
+++ b/drivers/net/wireless/virtual/Kconfig
@@ -0,0 +1,20 @@
+config MAC80211_HWSIM
+	tristate "Simulated radio testing tool for mac80211"
+	depends on MAC80211
+	help
+	  This driver is a developer testing tool that can be used to test
+	  IEEE 802.11 networking stack (mac80211) functionality. This is not
+	  needed for normal wireless LAN usage and is only for testing. See
+	  Documentation/networking/mac80211_hwsim for more information on how
+	  to use this tool.
+
+	  To compile this driver as a module, choose M here: the module will be
+	  called mac80211_hwsim.  If unsure, say N.
+
+config VIRT_WIFI
+	tristate "Wifi wrapper for ethernet drivers"
+	depends on CFG80211
+	help
+	  This option adds support for ethernet connections to appear as if they
+	  are wifi connections through a special rtnetlink device.
+
diff --git a/drivers/net/wireless/virtual/Makefile b/drivers/net/wireless/virtual/Makefile
new file mode 100644
index 000000000000..5773cc6d643e
--- /dev/null
+++ b/drivers/net/wireless/virtual/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_MAC80211_HWSIM)	+= mac80211_hwsim.o
+
+obj-$(CONFIG_VIRT_WIFI)	+= virt_wifi.o
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
new file mode 100644
index 000000000000..f4bdc243ea0d
--- /dev/null
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -0,0 +1,5901 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * mac80211_hwsim - software simulator of 802.11 radio(s) for mac80211
+ * Copyright (c) 2008, Jouni Malinen <j@w1.fi>
+ * Copyright (c) 2011, Javier Lopez <jlopex@gmail.com>
+ * Copyright (c) 2016 - 2017 Intel Deutschland GmbH
+ * Copyright (C) 2018 - 2022 Intel Corporation
+ */
+
+/*
+ * TODO:
+ * - Add TSF sync and fix IBSS beacon transmission by adding
+ *   competition for "air time" at TBTT
+ * - RX filtering based on filter configuration (data->rx_filter)
+ */
+
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <net/dst.h>
+#include <net/xfrm.h>
+#include <net/mac80211.h>
+#include <net/ieee80211_radiotap.h>
+#include <linux/if_arp.h>
+#include <linux/rtnetlink.h>
+#include <linux/etherdevice.h>
+#include <linux/platform_device.h>
+#include <linux/debugfs.h>
+#include <linux/module.h>
+#include <linux/ktime.h>
+#include <net/genetlink.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <linux/rhashtable.h>
+#include <linux/nospec.h>
+#include <linux/virtio.h>
+#include <linux/virtio_ids.h>
+#include <linux/virtio_config.h>
+#include "mac80211_hwsim.h"
+
+#define WARN_QUEUE 100
+#define MAX_QUEUE 200
+
+MODULE_AUTHOR("Jouni Malinen");
+MODULE_DESCRIPTION("Software simulator of 802.11 radio(s) for mac80211");
+MODULE_LICENSE("GPL");
+
+static int radios = 2;
+module_param(radios, int, 0444);
+MODULE_PARM_DESC(radios, "Number of simulated radios");
+
+static int channels = 1;
+module_param(channels, int, 0444);
+MODULE_PARM_DESC(channels, "Number of concurrent channels");
+
+static bool paged_rx = false;
+module_param(paged_rx, bool, 0644);
+MODULE_PARM_DESC(paged_rx, "Use paged SKBs for RX instead of linear ones");
+
+static bool rctbl = false;
+module_param(rctbl, bool, 0444);
+MODULE_PARM_DESC(rctbl, "Handle rate control table");
+
+static bool support_p2p_device = true;
+module_param(support_p2p_device, bool, 0444);
+MODULE_PARM_DESC(support_p2p_device, "Support P2P-Device interface type");
+
+static bool mlo;
+module_param(mlo, bool, 0444);
+MODULE_PARM_DESC(mlo, "Support MLO");
+
+/**
+ * enum hwsim_regtest - the type of regulatory tests we offer
+ *
+ * These are the different values you can use for the regtest
+ * module parameter. This is useful to help test world roaming
+ * and the driver regulatory_hint() call and combinations of these.
+ * If you want to do specific alpha2 regulatory domain tests simply
+ * use the userspace regulatory request as that will be respected as
+ * well without the need of this module parameter. This is designed
+ * only for testing the driver regulatory request, world roaming
+ * and all possible combinations.
+ *
+ * @HWSIM_REGTEST_DISABLED: No regulatory tests are performed,
+ * 	this is the default value.
+ * @HWSIM_REGTEST_DRIVER_REG_FOLLOW: Used for testing the driver regulatory
+ *	hint, only one driver regulatory hint will be sent as such the
+ * 	secondary radios are expected to follow.
+ * @HWSIM_REGTEST_DRIVER_REG_ALL: Used for testing the driver regulatory
+ * 	request with all radios reporting the same regulatory domain.
+ * @HWSIM_REGTEST_DIFF_COUNTRY: Used for testing the drivers calling
+ * 	different regulatory domains requests. Expected behaviour is for
+ * 	an intersection to occur but each device will still use their
+ * 	respective regulatory requested domains. Subsequent radios will
+ * 	use the resulting intersection.
+ * @HWSIM_REGTEST_WORLD_ROAM: Used for testing the world roaming. We accomplish
+ *	this by using a custom beacon-capable regulatory domain for the first
+ *	radio. All other device world roam.
+ * @HWSIM_REGTEST_CUSTOM_WORLD: Used for testing the custom world regulatory
+ * 	domain requests. All radios will adhere to this custom world regulatory
+ * 	domain.
+ * @HWSIM_REGTEST_CUSTOM_WORLD_2: Used for testing 2 custom world regulatory
+ * 	domain requests. The first radio will adhere to the first custom world
+ * 	regulatory domain, the second one to the second custom world regulatory
+ * 	domain. All other devices will world roam.
+ * @HWSIM_REGTEST_STRICT_FOLLOW: Used for testing strict regulatory domain
+ *	settings, only the first radio will send a regulatory domain request
+ *	and use strict settings. The rest of the radios are expected to follow.
+ * @HWSIM_REGTEST_STRICT_ALL: Used for testing strict regulatory domain
+ *	settings. All radios will adhere to this.
+ * @HWSIM_REGTEST_STRICT_AND_DRIVER_REG: Used for testing strict regulatory
+ *	domain settings, combined with secondary driver regulatory domain
+ *	settings. The first radio will get a strict regulatory domain setting
+ *	using the first driver regulatory request and the second radio will use
+ *	non-strict settings using the second driver regulatory request. All
+ *	other devices should follow the intersection created between the
+ *	first two.
+ * @HWSIM_REGTEST_ALL: Used for testing every possible mix. You will need
+ * 	at least 6 radios for a complete test. We will test in this order:
+ * 	1 - driver custom world regulatory domain
+ * 	2 - second custom world regulatory domain
+ * 	3 - first driver regulatory domain request
+ * 	4 - second driver regulatory domain request
+ * 	5 - strict regulatory domain settings using the third driver regulatory
+ * 	    domain request
+ * 	6 and on - should follow the intersection of the 3rd, 4rth and 5th radio
+ * 	           regulatory requests.
+ */
+enum hwsim_regtest {
+	HWSIM_REGTEST_DISABLED = 0,
+	HWSIM_REGTEST_DRIVER_REG_FOLLOW = 1,
+	HWSIM_REGTEST_DRIVER_REG_ALL = 2,
+	HWSIM_REGTEST_DIFF_COUNTRY = 3,
+	HWSIM_REGTEST_WORLD_ROAM = 4,
+	HWSIM_REGTEST_CUSTOM_WORLD = 5,
+	HWSIM_REGTEST_CUSTOM_WORLD_2 = 6,
+	HWSIM_REGTEST_STRICT_FOLLOW = 7,
+	HWSIM_REGTEST_STRICT_ALL = 8,
+	HWSIM_REGTEST_STRICT_AND_DRIVER_REG = 9,
+	HWSIM_REGTEST_ALL = 10,
+};
+
+/* Set to one of the HWSIM_REGTEST_* values above */
+static int regtest = HWSIM_REGTEST_DISABLED;
+module_param(regtest, int, 0444);
+MODULE_PARM_DESC(regtest, "The type of regulatory test we want to run");
+
+static const char *hwsim_alpha2s[] = {
+	"FI",
+	"AL",
+	"US",
+	"DE",
+	"JP",
+	"AL",
+};
+
+static const struct ieee80211_regdomain hwsim_world_regdom_custom_01 = {
+	.n_reg_rules = 5,
+	.alpha2 =  "99",
+	.reg_rules = {
+		REG_RULE(2412-10, 2462+10, 40, 0, 20, 0),
+		REG_RULE(2484-10, 2484+10, 40, 0, 20, 0),
+		REG_RULE(5150-10, 5240+10, 40, 0, 30, 0),
+		REG_RULE(5745-10, 5825+10, 40, 0, 30, 0),
+		REG_RULE(5855-10, 5925+10, 40, 0, 33, 0),
+	}
+};
+
+static const struct ieee80211_regdomain hwsim_world_regdom_custom_02 = {
+	.n_reg_rules = 3,
+	.alpha2 =  "99",
+	.reg_rules = {
+		REG_RULE(2412-10, 2462+10, 40, 0, 20, 0),
+		REG_RULE(5725-10, 5850+10, 40, 0, 30,
+			 NL80211_RRF_NO_IR),
+		REG_RULE(5855-10, 5925+10, 40, 0, 33, 0),
+	}
+};
+
+static const struct ieee80211_regdomain hwsim_world_regdom_custom_03 = {
+	.n_reg_rules = 6,
+	.alpha2 =  "99",
+	.reg_rules = {
+		REG_RULE(2412 - 10, 2462 + 10, 40, 0, 20, 0),
+		REG_RULE(2484 - 10, 2484 + 10, 40, 0, 20, 0),
+		REG_RULE(5150 - 10, 5240 + 10, 40, 0, 30, 0),
+		REG_RULE(5745 - 10, 5825 + 10, 40, 0, 30, 0),
+		REG_RULE(5855 - 10, 5925 + 10, 40, 0, 33, 0),
+		REG_RULE(5955 - 10, 7125 + 10, 320, 0, 33, 0),
+	}
+};
+
+static const struct ieee80211_regdomain *hwsim_world_regdom_custom[] = {
+	&hwsim_world_regdom_custom_01,
+	&hwsim_world_regdom_custom_02,
+	&hwsim_world_regdom_custom_03,
+};
+
+struct hwsim_vif_priv {
+	u32 magic;
+	u8 bssid[ETH_ALEN];
+	bool assoc;
+	bool bcn_en;
+	u16 aid;
+};
+
+#define HWSIM_VIF_MAGIC	0x69537748
+
+static inline void hwsim_check_magic(struct ieee80211_vif *vif)
+{
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+	WARN(vp->magic != HWSIM_VIF_MAGIC,
+	     "Invalid VIF (%p) magic %#x, %pM, %d/%d\n",
+	     vif, vp->magic, vif->addr, vif->type, vif->p2p);
+}
+
+static inline void hwsim_set_magic(struct ieee80211_vif *vif)
+{
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+	vp->magic = HWSIM_VIF_MAGIC;
+}
+
+static inline void hwsim_clear_magic(struct ieee80211_vif *vif)
+{
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+	vp->magic = 0;
+}
+
+struct hwsim_sta_priv {
+	u32 magic;
+	unsigned int last_link;
+	u16 active_links_rx;
+};
+
+#define HWSIM_STA_MAGIC	0x6d537749
+
+static inline void hwsim_check_sta_magic(struct ieee80211_sta *sta)
+{
+	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+	WARN_ON(sp->magic != HWSIM_STA_MAGIC);
+}
+
+static inline void hwsim_set_sta_magic(struct ieee80211_sta *sta)
+{
+	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+	sp->magic = HWSIM_STA_MAGIC;
+}
+
+static inline void hwsim_clear_sta_magic(struct ieee80211_sta *sta)
+{
+	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+	sp->magic = 0;
+}
+
+struct hwsim_chanctx_priv {
+	u32 magic;
+};
+
+#define HWSIM_CHANCTX_MAGIC 0x6d53774a
+
+static inline void hwsim_check_chanctx_magic(struct ieee80211_chanctx_conf *c)
+{
+	struct hwsim_chanctx_priv *cp = (void *)c->drv_priv;
+	WARN_ON(cp->magic != HWSIM_CHANCTX_MAGIC);
+}
+
+static inline void hwsim_set_chanctx_magic(struct ieee80211_chanctx_conf *c)
+{
+	struct hwsim_chanctx_priv *cp = (void *)c->drv_priv;
+	cp->magic = HWSIM_CHANCTX_MAGIC;
+}
+
+static inline void hwsim_clear_chanctx_magic(struct ieee80211_chanctx_conf *c)
+{
+	struct hwsim_chanctx_priv *cp = (void *)c->drv_priv;
+	cp->magic = 0;
+}
+
+static unsigned int hwsim_net_id;
+
+static DEFINE_IDA(hwsim_netgroup_ida);
+
+struct hwsim_net {
+	int netgroup;
+	u32 wmediumd;
+};
+
+static inline int hwsim_net_get_netgroup(struct net *net)
+{
+	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
+
+	return hwsim_net->netgroup;
+}
+
+static inline int hwsim_net_set_netgroup(struct net *net)
+{
+	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
+
+	hwsim_net->netgroup = ida_alloc(&hwsim_netgroup_ida, GFP_KERNEL);
+	return hwsim_net->netgroup >= 0 ? 0 : -ENOMEM;
+}
+
+static inline u32 hwsim_net_get_wmediumd(struct net *net)
+{
+	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
+
+	return hwsim_net->wmediumd;
+}
+
+static inline void hwsim_net_set_wmediumd(struct net *net, u32 portid)
+{
+	struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id);
+
+	hwsim_net->wmediumd = portid;
+}
+
+static struct class *hwsim_class;
+
+static struct net_device *hwsim_mon; /* global monitor netdev */
+
+#define CHAN2G(_freq)  { \
+	.band = NL80211_BAND_2GHZ, \
+	.center_freq = (_freq), \
+	.hw_value = (_freq), \
+}
+
+#define CHAN5G(_freq) { \
+	.band = NL80211_BAND_5GHZ, \
+	.center_freq = (_freq), \
+	.hw_value = (_freq), \
+}
+
+#define CHAN6G(_freq) { \
+	.band = NL80211_BAND_6GHZ, \
+	.center_freq = (_freq), \
+	.hw_value = (_freq), \
+}
+
+static const struct ieee80211_channel hwsim_channels_2ghz[] = {
+	CHAN2G(2412), /* Channel 1 */
+	CHAN2G(2417), /* Channel 2 */
+	CHAN2G(2422), /* Channel 3 */
+	CHAN2G(2427), /* Channel 4 */
+	CHAN2G(2432), /* Channel 5 */
+	CHAN2G(2437), /* Channel 6 */
+	CHAN2G(2442), /* Channel 7 */
+	CHAN2G(2447), /* Channel 8 */
+	CHAN2G(2452), /* Channel 9 */
+	CHAN2G(2457), /* Channel 10 */
+	CHAN2G(2462), /* Channel 11 */
+	CHAN2G(2467), /* Channel 12 */
+	CHAN2G(2472), /* Channel 13 */
+	CHAN2G(2484), /* Channel 14 */
+};
+
+static const struct ieee80211_channel hwsim_channels_5ghz[] = {
+	CHAN5G(5180), /* Channel 36 */
+	CHAN5G(5200), /* Channel 40 */
+	CHAN5G(5220), /* Channel 44 */
+	CHAN5G(5240), /* Channel 48 */
+
+	CHAN5G(5260), /* Channel 52 */
+	CHAN5G(5280), /* Channel 56 */
+	CHAN5G(5300), /* Channel 60 */
+	CHAN5G(5320), /* Channel 64 */
+
+	CHAN5G(5500), /* Channel 100 */
+	CHAN5G(5520), /* Channel 104 */
+	CHAN5G(5540), /* Channel 108 */
+	CHAN5G(5560), /* Channel 112 */
+	CHAN5G(5580), /* Channel 116 */
+	CHAN5G(5600), /* Channel 120 */
+	CHAN5G(5620), /* Channel 124 */
+	CHAN5G(5640), /* Channel 128 */
+	CHAN5G(5660), /* Channel 132 */
+	CHAN5G(5680), /* Channel 136 */
+	CHAN5G(5700), /* Channel 140 */
+
+	CHAN5G(5745), /* Channel 149 */
+	CHAN5G(5765), /* Channel 153 */
+	CHAN5G(5785), /* Channel 157 */
+	CHAN5G(5805), /* Channel 161 */
+	CHAN5G(5825), /* Channel 165 */
+	CHAN5G(5845), /* Channel 169 */
+
+	CHAN5G(5855), /* Channel 171 */
+	CHAN5G(5860), /* Channel 172 */
+	CHAN5G(5865), /* Channel 173 */
+	CHAN5G(5870), /* Channel 174 */
+
+	CHAN5G(5875), /* Channel 175 */
+	CHAN5G(5880), /* Channel 176 */
+	CHAN5G(5885), /* Channel 177 */
+	CHAN5G(5890), /* Channel 178 */
+	CHAN5G(5895), /* Channel 179 */
+	CHAN5G(5900), /* Channel 180 */
+	CHAN5G(5905), /* Channel 181 */
+
+	CHAN5G(5910), /* Channel 182 */
+	CHAN5G(5915), /* Channel 183 */
+	CHAN5G(5920), /* Channel 184 */
+	CHAN5G(5925), /* Channel 185 */
+};
+
+static const struct ieee80211_channel hwsim_channels_6ghz[] = {
+	CHAN6G(5955), /* Channel 1 */
+	CHAN6G(5975), /* Channel 5 */
+	CHAN6G(5995), /* Channel 9 */
+	CHAN6G(6015), /* Channel 13 */
+	CHAN6G(6035), /* Channel 17 */
+	CHAN6G(6055), /* Channel 21 */
+	CHAN6G(6075), /* Channel 25 */
+	CHAN6G(6095), /* Channel 29 */
+	CHAN6G(6115), /* Channel 33 */
+	CHAN6G(6135), /* Channel 37 */
+	CHAN6G(6155), /* Channel 41 */
+	CHAN6G(6175), /* Channel 45 */
+	CHAN6G(6195), /* Channel 49 */
+	CHAN6G(6215), /* Channel 53 */
+	CHAN6G(6235), /* Channel 57 */
+	CHAN6G(6255), /* Channel 61 */
+	CHAN6G(6275), /* Channel 65 */
+	CHAN6G(6295), /* Channel 69 */
+	CHAN6G(6315), /* Channel 73 */
+	CHAN6G(6335), /* Channel 77 */
+	CHAN6G(6355), /* Channel 81 */
+	CHAN6G(6375), /* Channel 85 */
+	CHAN6G(6395), /* Channel 89 */
+	CHAN6G(6415), /* Channel 93 */
+	CHAN6G(6435), /* Channel 97 */
+	CHAN6G(6455), /* Channel 181 */
+	CHAN6G(6475), /* Channel 105 */
+	CHAN6G(6495), /* Channel 109 */
+	CHAN6G(6515), /* Channel 113 */
+	CHAN6G(6535), /* Channel 117 */
+	CHAN6G(6555), /* Channel 121 */
+	CHAN6G(6575), /* Channel 125 */
+	CHAN6G(6595), /* Channel 129 */
+	CHAN6G(6615), /* Channel 133 */
+	CHAN6G(6635), /* Channel 137 */
+	CHAN6G(6655), /* Channel 141 */
+	CHAN6G(6675), /* Channel 145 */
+	CHAN6G(6695), /* Channel 149 */
+	CHAN6G(6715), /* Channel 153 */
+	CHAN6G(6735), /* Channel 157 */
+	CHAN6G(6755), /* Channel 161 */
+	CHAN6G(6775), /* Channel 165 */
+	CHAN6G(6795), /* Channel 169 */
+	CHAN6G(6815), /* Channel 173 */
+	CHAN6G(6835), /* Channel 177 */
+	CHAN6G(6855), /* Channel 181 */
+	CHAN6G(6875), /* Channel 185 */
+	CHAN6G(6895), /* Channel 189 */
+	CHAN6G(6915), /* Channel 193 */
+	CHAN6G(6935), /* Channel 197 */
+	CHAN6G(6955), /* Channel 201 */
+	CHAN6G(6975), /* Channel 205 */
+	CHAN6G(6995), /* Channel 209 */
+	CHAN6G(7015), /* Channel 213 */
+	CHAN6G(7035), /* Channel 217 */
+	CHAN6G(7055), /* Channel 221 */
+	CHAN6G(7075), /* Channel 225 */
+	CHAN6G(7095), /* Channel 229 */
+	CHAN6G(7115), /* Channel 233 */
+};
+
+#define NUM_S1G_CHANS_US 51
+static struct ieee80211_channel hwsim_channels_s1g[NUM_S1G_CHANS_US];
+
+static const struct ieee80211_sta_s1g_cap hwsim_s1g_cap = {
+	.s1g = true,
+	.cap = { S1G_CAP0_SGI_1MHZ | S1G_CAP0_SGI_2MHZ,
+		 0,
+		 0,
+		 S1G_CAP3_MAX_MPDU_LEN,
+		 0,
+		 S1G_CAP5_AMPDU,
+		 0,
+		 S1G_CAP7_DUP_1MHZ,
+		 S1G_CAP8_TWT_RESPOND | S1G_CAP8_TWT_REQUEST,
+		 0},
+	.nss_mcs = { 0xfc | 1, /* MCS 7 for 1 SS */
+	/* RX Highest Supported Long GI Data Rate 0:7 */
+		     0,
+	/* RX Highest Supported Long GI Data Rate 0:7 */
+	/* TX S1G MCS Map 0:6 */
+		     0xfa,
+	/* TX S1G MCS Map :7 */
+	/* TX Highest Supported Long GI Data Rate 0:6 */
+		     0x80,
+	/* TX Highest Supported Long GI Data Rate 7:8 */
+	/* Rx Single spatial stream and S1G-MCS Map for 1MHz */
+	/* Tx Single spatial stream and S1G-MCS Map for 1MHz */
+		     0 },
+};
+
+static void hwsim_init_s1g_channels(struct ieee80211_channel *chans)
+{
+	int ch, freq;
+
+	for (ch = 0; ch < NUM_S1G_CHANS_US; ch++) {
+		freq = 902000 + (ch + 1) * 500;
+		chans[ch].band = NL80211_BAND_S1GHZ;
+		chans[ch].center_freq = KHZ_TO_MHZ(freq);
+		chans[ch].freq_offset = freq % 1000;
+		chans[ch].hw_value = ch + 1;
+	}
+}
+
+static const struct ieee80211_rate hwsim_rates[] = {
+	{ .bitrate = 10 },
+	{ .bitrate = 20, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 55, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 110, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 60 },
+	{ .bitrate = 90 },
+	{ .bitrate = 120 },
+	{ .bitrate = 180 },
+	{ .bitrate = 240 },
+	{ .bitrate = 360 },
+	{ .bitrate = 480 },
+	{ .bitrate = 540 }
+};
+
+#define DEFAULT_RX_RSSI -50
+
+static const u32 hwsim_ciphers[] = {
+	WLAN_CIPHER_SUITE_WEP40,
+	WLAN_CIPHER_SUITE_WEP104,
+	WLAN_CIPHER_SUITE_TKIP,
+	WLAN_CIPHER_SUITE_CCMP,
+	WLAN_CIPHER_SUITE_CCMP_256,
+	WLAN_CIPHER_SUITE_GCMP,
+	WLAN_CIPHER_SUITE_GCMP_256,
+	WLAN_CIPHER_SUITE_AES_CMAC,
+	WLAN_CIPHER_SUITE_BIP_CMAC_256,
+	WLAN_CIPHER_SUITE_BIP_GMAC_128,
+	WLAN_CIPHER_SUITE_BIP_GMAC_256,
+};
+
+#define OUI_QCA 0x001374
+#define QCA_NL80211_SUBCMD_TEST 1
+enum qca_nl80211_vendor_subcmds {
+	QCA_WLAN_VENDOR_ATTR_TEST = 8,
+	QCA_WLAN_VENDOR_ATTR_MAX = QCA_WLAN_VENDOR_ATTR_TEST
+};
+
+static const struct nla_policy
+hwsim_vendor_test_policy[QCA_WLAN_VENDOR_ATTR_MAX + 1] = {
+	[QCA_WLAN_VENDOR_ATTR_MAX] = { .type = NLA_U32 },
+};
+
+static int mac80211_hwsim_vendor_cmd_test(struct wiphy *wiphy,
+					  struct wireless_dev *wdev,
+					  const void *data, int data_len)
+{
+	struct sk_buff *skb;
+	struct nlattr *tb[QCA_WLAN_VENDOR_ATTR_MAX + 1];
+	int err;
+	u32 val;
+
+	err = nla_parse_deprecated(tb, QCA_WLAN_VENDOR_ATTR_MAX, data,
+				   data_len, hwsim_vendor_test_policy, NULL);
+	if (err)
+		return err;
+	if (!tb[QCA_WLAN_VENDOR_ATTR_TEST])
+		return -EINVAL;
+	val = nla_get_u32(tb[QCA_WLAN_VENDOR_ATTR_TEST]);
+	wiphy_dbg(wiphy, "%s: test=%u\n", __func__, val);
+
+	/* Send a vendor event as a test. Note that this would not normally be
+	 * done within a command handler, but rather, based on some other
+	 * trigger. For simplicity, this command is used to trigger the event
+	 * here.
+	 *
+	 * event_idx = 0 (index in mac80211_hwsim_vendor_commands)
+	 */
+	skb = cfg80211_vendor_event_alloc(wiphy, wdev, 100, 0, GFP_KERNEL);
+	if (skb) {
+		/* skb_put() or nla_put() will fill up data within
+		 * NL80211_ATTR_VENDOR_DATA.
+		 */
+
+		/* Add vendor data */
+		nla_put_u32(skb, QCA_WLAN_VENDOR_ATTR_TEST, val + 1);
+
+		/* Send the event - this will call nla_nest_end() */
+		cfg80211_vendor_event(skb, GFP_KERNEL);
+	}
+
+	/* Send a response to the command */
+	skb = cfg80211_vendor_cmd_alloc_reply_skb(wiphy, 10);
+	if (!skb)
+		return -ENOMEM;
+
+	/* skb_put() or nla_put() will fill up data within
+	 * NL80211_ATTR_VENDOR_DATA
+	 */
+	nla_put_u32(skb, QCA_WLAN_VENDOR_ATTR_TEST, val + 2);
+
+	return cfg80211_vendor_cmd_reply(skb);
+}
+
+static struct wiphy_vendor_command mac80211_hwsim_vendor_commands[] = {
+	{
+		.info = { .vendor_id = OUI_QCA,
+			  .subcmd = QCA_NL80211_SUBCMD_TEST },
+		.flags = WIPHY_VENDOR_CMD_NEED_NETDEV,
+		.doit = mac80211_hwsim_vendor_cmd_test,
+		.policy = hwsim_vendor_test_policy,
+		.maxattr = QCA_WLAN_VENDOR_ATTR_MAX,
+	}
+};
+
+/* Advertise support vendor specific events */
+static const struct nl80211_vendor_cmd_info mac80211_hwsim_vendor_events[] = {
+	{ .vendor_id = OUI_QCA, .subcmd = 1 },
+};
+
+static DEFINE_SPINLOCK(hwsim_radio_lock);
+static LIST_HEAD(hwsim_radios);
+static struct rhashtable hwsim_radios_rht;
+static int hwsim_radio_idx;
+static int hwsim_radios_generation = 1;
+
+static struct platform_driver mac80211_hwsim_driver = {
+	.driver = {
+		.name = "mac80211_hwsim",
+	},
+};
+
+struct mac80211_hwsim_link_data {
+	u32 link_id;
+	u64 beacon_int	/* beacon interval in us */;
+	struct hrtimer beacon_timer;
+};
+
+struct mac80211_hwsim_data {
+	struct list_head list;
+	struct rhash_head rht;
+	struct ieee80211_hw *hw;
+	struct device *dev;
+	struct ieee80211_supported_band bands[NUM_NL80211_BANDS];
+	struct ieee80211_channel channels_2ghz[ARRAY_SIZE(hwsim_channels_2ghz)];
+	struct ieee80211_channel channels_5ghz[ARRAY_SIZE(hwsim_channels_5ghz)];
+	struct ieee80211_channel channels_6ghz[ARRAY_SIZE(hwsim_channels_6ghz)];
+	struct ieee80211_channel channels_s1g[ARRAY_SIZE(hwsim_channels_s1g)];
+	struct ieee80211_rate rates[ARRAY_SIZE(hwsim_rates)];
+	struct ieee80211_iface_combination if_combination;
+	struct ieee80211_iface_limit if_limits[3];
+	int n_if_limits;
+
+	u32 ciphers[ARRAY_SIZE(hwsim_ciphers)];
+
+	struct mac_address addresses[2];
+	int channels, idx;
+	bool use_chanctx;
+	bool destroy_on_close;
+	u32 portid;
+	char alpha2[2];
+	const struct ieee80211_regdomain *regd;
+
+	struct ieee80211_channel *tmp_chan;
+	struct ieee80211_channel *roc_chan;
+	u32 roc_duration;
+	struct delayed_work roc_start;
+	struct delayed_work roc_done;
+	struct delayed_work hw_scan;
+	struct cfg80211_scan_request *hw_scan_request;
+	struct ieee80211_vif *hw_scan_vif;
+	int scan_chan_idx;
+	u8 scan_addr[ETH_ALEN];
+	struct {
+		struct ieee80211_channel *channel;
+		unsigned long next_start, start, end;
+	} survey_data[ARRAY_SIZE(hwsim_channels_2ghz) +
+		      ARRAY_SIZE(hwsim_channels_5ghz) +
+		      ARRAY_SIZE(hwsim_channels_6ghz)];
+
+	struct ieee80211_channel *channel;
+	enum nl80211_chan_width bw;
+	unsigned int rx_filter;
+	bool started, idle, scanning;
+	struct mutex mutex;
+	enum ps_mode {
+		PS_DISABLED, PS_ENABLED, PS_AUTO_POLL, PS_MANUAL_POLL
+	} ps;
+	bool ps_poll_pending;
+	struct dentry *debugfs;
+
+	atomic_t pending_cookie;
+	struct sk_buff_head pending;	/* packets pending */
+	/*
+	 * Only radios in the same group can communicate together (the
+	 * channel has to match too). Each bit represents a group. A
+	 * radio can be in more than one group.
+	 */
+	u64 group;
+
+	/* group shared by radios created in the same netns */
+	int netgroup;
+	/* wmediumd portid responsible for netgroup of this radio */
+	u32 wmediumd;
+
+	/* difference between this hw's clock and the real clock, in usecs */
+	s64 tsf_offset;
+	s64 bcn_delta;
+	/* absolute beacon transmission time. Used to cover up "tx" delay. */
+	u64 abs_bcn_ts;
+
+	/* Stats */
+	u64 tx_pkts;
+	u64 rx_pkts;
+	u64 tx_bytes;
+	u64 rx_bytes;
+	u64 tx_dropped;
+	u64 tx_failed;
+
+	/* RSSI in rx status of the receiver */
+	int rx_rssi;
+
+	struct mac80211_hwsim_link_data link_data[IEEE80211_MLD_MAX_NUM_LINKS];
+};
+
+static const struct rhashtable_params hwsim_rht_params = {
+	.nelem_hint = 2,
+	.automatic_shrinking = true,
+	.key_len = ETH_ALEN,
+	.key_offset = offsetof(struct mac80211_hwsim_data, addresses[1]),
+	.head_offset = offsetof(struct mac80211_hwsim_data, rht),
+};
+
+struct hwsim_radiotap_hdr {
+	struct ieee80211_radiotap_header hdr;
+	__le64 rt_tsft;
+	u8 rt_flags;
+	u8 rt_rate;
+	__le16 rt_channel;
+	__le16 rt_chbitmask;
+} __packed;
+
+struct hwsim_radiotap_ack_hdr {
+	struct ieee80211_radiotap_header hdr;
+	u8 rt_flags;
+	u8 pad;
+	__le16 rt_channel;
+	__le16 rt_chbitmask;
+} __packed;
+
+/* MAC80211_HWSIM netlink family */
+static struct genl_family hwsim_genl_family;
+
+enum hwsim_multicast_groups {
+	HWSIM_MCGRP_CONFIG,
+};
+
+static const struct genl_multicast_group hwsim_mcgrps[] = {
+	[HWSIM_MCGRP_CONFIG] = { .name = "config", },
+};
+
+/* MAC80211_HWSIM netlink policy */
+
+static const struct nla_policy hwsim_genl_policy[HWSIM_ATTR_MAX + 1] = {
+	[HWSIM_ATTR_ADDR_RECEIVER] = NLA_POLICY_ETH_ADDR_COMPAT,
+	[HWSIM_ATTR_ADDR_TRANSMITTER] = NLA_POLICY_ETH_ADDR_COMPAT,
+	[HWSIM_ATTR_FRAME] = { .type = NLA_BINARY,
+			       .len = IEEE80211_MAX_DATA_LEN },
+	[HWSIM_ATTR_FLAGS] = { .type = NLA_U32 },
+	[HWSIM_ATTR_RX_RATE] = { .type = NLA_U32 },
+	[HWSIM_ATTR_SIGNAL] = { .type = NLA_U32 },
+	[HWSIM_ATTR_TX_INFO] = { .type = NLA_BINARY,
+				 .len = IEEE80211_TX_MAX_RATES *
+					sizeof(struct hwsim_tx_rate)},
+	[HWSIM_ATTR_COOKIE] = { .type = NLA_U64 },
+	[HWSIM_ATTR_CHANNELS] = { .type = NLA_U32 },
+	[HWSIM_ATTR_RADIO_ID] = { .type = NLA_U32 },
+	[HWSIM_ATTR_REG_HINT_ALPHA2] = { .type = NLA_STRING, .len = 2 },
+	[HWSIM_ATTR_REG_CUSTOM_REG] = { .type = NLA_U32 },
+	[HWSIM_ATTR_REG_STRICT_REG] = { .type = NLA_FLAG },
+	[HWSIM_ATTR_SUPPORT_P2P_DEVICE] = { .type = NLA_FLAG },
+	[HWSIM_ATTR_USE_CHANCTX] = { .type = NLA_FLAG },
+	[HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE] = { .type = NLA_FLAG },
+	[HWSIM_ATTR_RADIO_NAME] = { .type = NLA_STRING },
+	[HWSIM_ATTR_NO_VIF] = { .type = NLA_FLAG },
+	[HWSIM_ATTR_FREQ] = { .type = NLA_U32 },
+	[HWSIM_ATTR_TX_INFO_FLAGS] = { .type = NLA_BINARY },
+	[HWSIM_ATTR_PERM_ADDR] = NLA_POLICY_ETH_ADDR_COMPAT,
+	[HWSIM_ATTR_IFTYPE_SUPPORT] = { .type = NLA_U32 },
+	[HWSIM_ATTR_CIPHER_SUPPORT] = { .type = NLA_BINARY },
+	[HWSIM_ATTR_MLO_SUPPORT] = { .type = NLA_FLAG },
+};
+
+#if IS_REACHABLE(CONFIG_VIRTIO)
+
+/* MAC80211_HWSIM virtio queues */
+static struct virtqueue *hwsim_vqs[HWSIM_NUM_VQS];
+static bool hwsim_virtio_enabled;
+static DEFINE_SPINLOCK(hwsim_virtio_lock);
+
+static void hwsim_virtio_rx_work(struct work_struct *work);
+static DECLARE_WORK(hwsim_virtio_rx, hwsim_virtio_rx_work);
+
+static int hwsim_tx_virtio(struct mac80211_hwsim_data *data,
+			   struct sk_buff *skb)
+{
+	struct scatterlist sg[1];
+	unsigned long flags;
+	int err;
+
+	spin_lock_irqsave(&hwsim_virtio_lock, flags);
+	if (!hwsim_virtio_enabled) {
+		err = -ENODEV;
+		goto out_free;
+	}
+
+	sg_init_one(sg, skb->head, skb_end_offset(skb));
+	err = virtqueue_add_outbuf(hwsim_vqs[HWSIM_VQ_TX], sg, 1, skb,
+				   GFP_ATOMIC);
+	if (err)
+		goto out_free;
+	virtqueue_kick(hwsim_vqs[HWSIM_VQ_TX]);
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+	return 0;
+
+out_free:
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+	nlmsg_free(skb);
+	return err;
+}
+#else
+/* cause a linker error if this ends up being needed */
+extern int hwsim_tx_virtio(struct mac80211_hwsim_data *data,
+			   struct sk_buff *skb);
+#define hwsim_virtio_enabled false
+#endif
+
+static int hwsim_get_chanwidth(enum nl80211_chan_width bw)
+{
+	switch (bw) {
+	case NL80211_CHAN_WIDTH_20_NOHT:
+	case NL80211_CHAN_WIDTH_20:
+		return 20;
+	case NL80211_CHAN_WIDTH_40:
+		return 40;
+	case NL80211_CHAN_WIDTH_80:
+		return 80;
+	case NL80211_CHAN_WIDTH_80P80:
+	case NL80211_CHAN_WIDTH_160:
+		return 160;
+	case NL80211_CHAN_WIDTH_320:
+		return 320;
+	case NL80211_CHAN_WIDTH_5:
+		return 5;
+	case NL80211_CHAN_WIDTH_10:
+		return 10;
+	case NL80211_CHAN_WIDTH_1:
+		return 1;
+	case NL80211_CHAN_WIDTH_2:
+		return 2;
+	case NL80211_CHAN_WIDTH_4:
+		return 4;
+	case NL80211_CHAN_WIDTH_8:
+		return 8;
+	case NL80211_CHAN_WIDTH_16:
+		return 16;
+	}
+
+	return INT_MAX;
+}
+
+static void mac80211_hwsim_tx_frame(struct ieee80211_hw *hw,
+				    struct sk_buff *skb,
+				    struct ieee80211_channel *chan);
+
+/* sysfs attributes */
+static void hwsim_send_ps_poll(void *dat, u8 *mac, struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *data = dat;
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+	struct sk_buff *skb;
+	struct ieee80211_pspoll *pspoll;
+
+	if (!vp->assoc)
+		return;
+
+	wiphy_dbg(data->hw->wiphy,
+		  "%s: send PS-Poll to %pM for aid %d\n",
+		  __func__, vp->bssid, vp->aid);
+
+	skb = dev_alloc_skb(sizeof(*pspoll));
+	if (!skb)
+		return;
+	pspoll = skb_put(skb, sizeof(*pspoll));
+	pspoll->frame_control = cpu_to_le16(IEEE80211_FTYPE_CTL |
+					    IEEE80211_STYPE_PSPOLL |
+					    IEEE80211_FCTL_PM);
+	pspoll->aid = cpu_to_le16(0xc000 | vp->aid);
+	memcpy(pspoll->bssid, vp->bssid, ETH_ALEN);
+	memcpy(pspoll->ta, mac, ETH_ALEN);
+
+	rcu_read_lock();
+	mac80211_hwsim_tx_frame(data->hw, skb,
+				rcu_dereference(vif->bss_conf.chanctx_conf)->def.chan);
+	rcu_read_unlock();
+}
+
+static void hwsim_send_nullfunc(struct mac80211_hwsim_data *data, u8 *mac,
+				struct ieee80211_vif *vif, int ps)
+{
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+	struct sk_buff *skb;
+	struct ieee80211_hdr *hdr;
+	struct ieee80211_tx_info *cb;
+
+	if (!vp->assoc)
+		return;
+
+	wiphy_dbg(data->hw->wiphy,
+		  "%s: send data::nullfunc to %pM ps=%d\n",
+		  __func__, vp->bssid, ps);
+
+	skb = dev_alloc_skb(sizeof(*hdr));
+	if (!skb)
+		return;
+	hdr = skb_put(skb, sizeof(*hdr) - ETH_ALEN);
+	hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_DATA |
+					 IEEE80211_STYPE_NULLFUNC |
+					 IEEE80211_FCTL_TODS |
+					 (ps ? IEEE80211_FCTL_PM : 0));
+	hdr->duration_id = cpu_to_le16(0);
+	memcpy(hdr->addr1, vp->bssid, ETH_ALEN);
+	memcpy(hdr->addr2, mac, ETH_ALEN);
+	memcpy(hdr->addr3, vp->bssid, ETH_ALEN);
+
+	cb = IEEE80211_SKB_CB(skb);
+	cb->control.rates[0].count = 1;
+	cb->control.rates[1].idx = -1;
+
+	rcu_read_lock();
+	mac80211_hwsim_tx_frame(data->hw, skb,
+				rcu_dereference(vif->bss_conf.chanctx_conf)->def.chan);
+	rcu_read_unlock();
+}
+
+
+static void hwsim_send_nullfunc_ps(void *dat, u8 *mac,
+				   struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *data = dat;
+	hwsim_send_nullfunc(data, mac, vif, 1);
+}
+
+static void hwsim_send_nullfunc_no_ps(void *dat, u8 *mac,
+				      struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *data = dat;
+	hwsim_send_nullfunc(data, mac, vif, 0);
+}
+
+static int hwsim_fops_ps_read(void *dat, u64 *val)
+{
+	struct mac80211_hwsim_data *data = dat;
+	*val = data->ps;
+	return 0;
+}
+
+static int hwsim_fops_ps_write(void *dat, u64 val)
+{
+	struct mac80211_hwsim_data *data = dat;
+	enum ps_mode old_ps;
+
+	if (val != PS_DISABLED && val != PS_ENABLED && val != PS_AUTO_POLL &&
+	    val != PS_MANUAL_POLL)
+		return -EINVAL;
+
+	if (val == PS_MANUAL_POLL) {
+		if (data->ps != PS_ENABLED)
+			return -EINVAL;
+		local_bh_disable();
+		ieee80211_iterate_active_interfaces_atomic(
+			data->hw, IEEE80211_IFACE_ITER_NORMAL,
+			hwsim_send_ps_poll, data);
+		local_bh_enable();
+		return 0;
+	}
+	old_ps = data->ps;
+	data->ps = val;
+
+	local_bh_disable();
+	if (old_ps == PS_DISABLED && val != PS_DISABLED) {
+		ieee80211_iterate_active_interfaces_atomic(
+			data->hw, IEEE80211_IFACE_ITER_NORMAL,
+			hwsim_send_nullfunc_ps, data);
+	} else if (old_ps != PS_DISABLED && val == PS_DISABLED) {
+		ieee80211_iterate_active_interfaces_atomic(
+			data->hw, IEEE80211_IFACE_ITER_NORMAL,
+			hwsim_send_nullfunc_no_ps, data);
+	}
+	local_bh_enable();
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(hwsim_fops_ps, hwsim_fops_ps_read, hwsim_fops_ps_write,
+			 "%llu\n");
+
+static int hwsim_write_simulate_radar(void *dat, u64 val)
+{
+	struct mac80211_hwsim_data *data = dat;
+
+	ieee80211_radar_detected(data->hw);
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(hwsim_simulate_radar, NULL,
+			 hwsim_write_simulate_radar, "%llu\n");
+
+static int hwsim_fops_group_read(void *dat, u64 *val)
+{
+	struct mac80211_hwsim_data *data = dat;
+	*val = data->group;
+	return 0;
+}
+
+static int hwsim_fops_group_write(void *dat, u64 val)
+{
+	struct mac80211_hwsim_data *data = dat;
+	data->group = val;
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(hwsim_fops_group,
+			 hwsim_fops_group_read, hwsim_fops_group_write,
+			 "%llx\n");
+
+static int hwsim_fops_rx_rssi_read(void *dat, u64 *val)
+{
+	struct mac80211_hwsim_data *data = dat;
+	*val = data->rx_rssi;
+	return 0;
+}
+
+static int hwsim_fops_rx_rssi_write(void *dat, u64 val)
+{
+	struct mac80211_hwsim_data *data = dat;
+	int rssi = (int)val;
+
+	if (rssi >= 0 || rssi < -100)
+		return -EINVAL;
+
+	data->rx_rssi = rssi;
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(hwsim_fops_rx_rssi,
+			 hwsim_fops_rx_rssi_read, hwsim_fops_rx_rssi_write,
+			 "%lld\n");
+
+static netdev_tx_t hwsim_mon_xmit(struct sk_buff *skb,
+					struct net_device *dev)
+{
+	/* TODO: allow packet injection */
+	dev_kfree_skb(skb);
+	return NETDEV_TX_OK;
+}
+
+static inline u64 mac80211_hwsim_get_tsf_raw(void)
+{
+	return ktime_to_us(ktime_get_real());
+}
+
+static __le64 __mac80211_hwsim_get_tsf(struct mac80211_hwsim_data *data)
+{
+	u64 now = mac80211_hwsim_get_tsf_raw();
+	return cpu_to_le64(now + data->tsf_offset);
+}
+
+static u64 mac80211_hwsim_get_tsf(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	return le64_to_cpu(__mac80211_hwsim_get_tsf(data));
+}
+
+static void mac80211_hwsim_set_tsf(struct ieee80211_hw *hw,
+		struct ieee80211_vif *vif, u64 tsf)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	u64 now = mac80211_hwsim_get_tsf(hw, vif);
+	/* MLD not supported here */
+	u32 bcn_int = data->link_data[0].beacon_int;
+	u64 delta = abs(tsf - now);
+
+	/* adjust after beaconing with new timestamp at old TBTT */
+	if (tsf > now) {
+		data->tsf_offset += delta;
+		data->bcn_delta = do_div(delta, bcn_int);
+	} else {
+		data->tsf_offset -= delta;
+		data->bcn_delta = -(s64)do_div(delta, bcn_int);
+	}
+}
+
+static void mac80211_hwsim_monitor_rx(struct ieee80211_hw *hw,
+				      struct sk_buff *tx_skb,
+				      struct ieee80211_channel *chan)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	struct sk_buff *skb;
+	struct hwsim_radiotap_hdr *hdr;
+	u16 flags, bitrate;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx_skb);
+	struct ieee80211_rate *txrate = ieee80211_get_tx_rate(hw, info);
+
+	if (!txrate)
+		bitrate = 0;
+	else
+		bitrate = txrate->bitrate;
+
+	if (!netif_running(hwsim_mon))
+		return;
+
+	skb = skb_copy_expand(tx_skb, sizeof(*hdr), 0, GFP_ATOMIC);
+	if (skb == NULL)
+		return;
+
+	hdr = skb_push(skb, sizeof(*hdr));
+	hdr->hdr.it_version = PKTHDR_RADIOTAP_VERSION;
+	hdr->hdr.it_pad = 0;
+	hdr->hdr.it_len = cpu_to_le16(sizeof(*hdr));
+	hdr->hdr.it_present = cpu_to_le32((1 << IEEE80211_RADIOTAP_FLAGS) |
+					  (1 << IEEE80211_RADIOTAP_RATE) |
+					  (1 << IEEE80211_RADIOTAP_TSFT) |
+					  (1 << IEEE80211_RADIOTAP_CHANNEL));
+	hdr->rt_tsft = __mac80211_hwsim_get_tsf(data);
+	hdr->rt_flags = 0;
+	hdr->rt_rate = bitrate / 5;
+	hdr->rt_channel = cpu_to_le16(chan->center_freq);
+	flags = IEEE80211_CHAN_2GHZ;
+	if (txrate && txrate->flags & IEEE80211_RATE_ERP_G)
+		flags |= IEEE80211_CHAN_OFDM;
+	else
+		flags |= IEEE80211_CHAN_CCK;
+	hdr->rt_chbitmask = cpu_to_le16(flags);
+
+	skb->dev = hwsim_mon;
+	skb_reset_mac_header(skb);
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+	skb->pkt_type = PACKET_OTHERHOST;
+	skb->protocol = htons(ETH_P_802_2);
+	memset(skb->cb, 0, sizeof(skb->cb));
+	netif_rx(skb);
+}
+
+
+static void mac80211_hwsim_monitor_ack(struct ieee80211_channel *chan,
+				       const u8 *addr)
+{
+	struct sk_buff *skb;
+	struct hwsim_radiotap_ack_hdr *hdr;
+	u16 flags;
+	struct ieee80211_hdr *hdr11;
+
+	if (!netif_running(hwsim_mon))
+		return;
+
+	skb = dev_alloc_skb(100);
+	if (skb == NULL)
+		return;
+
+	hdr = skb_put(skb, sizeof(*hdr));
+	hdr->hdr.it_version = PKTHDR_RADIOTAP_VERSION;
+	hdr->hdr.it_pad = 0;
+	hdr->hdr.it_len = cpu_to_le16(sizeof(*hdr));
+	hdr->hdr.it_present = cpu_to_le32((1 << IEEE80211_RADIOTAP_FLAGS) |
+					  (1 << IEEE80211_RADIOTAP_CHANNEL));
+	hdr->rt_flags = 0;
+	hdr->pad = 0;
+	hdr->rt_channel = cpu_to_le16(chan->center_freq);
+	flags = IEEE80211_CHAN_2GHZ;
+	hdr->rt_chbitmask = cpu_to_le16(flags);
+
+	hdr11 = skb_put(skb, 10);
+	hdr11->frame_control = cpu_to_le16(IEEE80211_FTYPE_CTL |
+					   IEEE80211_STYPE_ACK);
+	hdr11->duration_id = cpu_to_le16(0);
+	memcpy(hdr11->addr1, addr, ETH_ALEN);
+
+	skb->dev = hwsim_mon;
+	skb_reset_mac_header(skb);
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+	skb->pkt_type = PACKET_OTHERHOST;
+	skb->protocol = htons(ETH_P_802_2);
+	memset(skb->cb, 0, sizeof(skb->cb));
+	netif_rx(skb);
+}
+
+struct mac80211_hwsim_addr_match_data {
+	u8 addr[ETH_ALEN];
+	bool ret;
+};
+
+static void mac80211_hwsim_addr_iter(void *data, u8 *mac,
+				     struct ieee80211_vif *vif)
+{
+	int i;
+	struct mac80211_hwsim_addr_match_data *md = data;
+
+	if (memcmp(mac, md->addr, ETH_ALEN) == 0) {
+		md->ret = true;
+		return;
+	}
+
+	/* Match the link address */
+	for (i = 0; i < ARRAY_SIZE(vif->link_conf); i++) {
+		struct ieee80211_bss_conf *conf;
+
+		conf = rcu_dereference(vif->link_conf[i]);
+		if (!conf)
+			continue;
+
+		if (memcmp(conf->addr, md->addr, ETH_ALEN) == 0) {
+			md->ret = true;
+			return;
+		}
+	}
+}
+
+static bool mac80211_hwsim_addr_match(struct mac80211_hwsim_data *data,
+				      const u8 *addr)
+{
+	struct mac80211_hwsim_addr_match_data md = {
+		.ret = false,
+	};
+
+	if (data->scanning && memcmp(addr, data->scan_addr, ETH_ALEN) == 0)
+		return true;
+
+	memcpy(md.addr, addr, ETH_ALEN);
+
+	ieee80211_iterate_active_interfaces_atomic(data->hw,
+						   IEEE80211_IFACE_ITER_NORMAL,
+						   mac80211_hwsim_addr_iter,
+						   &md);
+
+	return md.ret;
+}
+
+static bool hwsim_ps_rx_ok(struct mac80211_hwsim_data *data,
+			   struct sk_buff *skb)
+{
+	switch (data->ps) {
+	case PS_DISABLED:
+		return true;
+	case PS_ENABLED:
+		return false;
+	case PS_AUTO_POLL:
+		/* TODO: accept (some) Beacons by default and other frames only
+		 * if pending PS-Poll has been sent */
+		return true;
+	case PS_MANUAL_POLL:
+		/* Allow unicast frames to own address if there is a pending
+		 * PS-Poll */
+		if (data->ps_poll_pending &&
+		    mac80211_hwsim_addr_match(data, skb->data + 4)) {
+			data->ps_poll_pending = false;
+			return true;
+		}
+		return false;
+	}
+
+	return true;
+}
+
+static int hwsim_unicast_netgroup(struct mac80211_hwsim_data *data,
+				  struct sk_buff *skb, int portid)
+{
+	struct net *net;
+	bool found = false;
+	int res = -ENOENT;
+
+	rcu_read_lock();
+	for_each_net_rcu(net) {
+		if (data->netgroup == hwsim_net_get_netgroup(net)) {
+			res = genlmsg_unicast(net, skb, portid);
+			found = true;
+			break;
+		}
+	}
+	rcu_read_unlock();
+
+	if (!found)
+		nlmsg_free(skb);
+
+	return res;
+}
+
+static void mac80211_hwsim_config_mac_nl(struct ieee80211_hw *hw,
+					 const u8 *addr, bool add)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	u32 _portid = READ_ONCE(data->wmediumd);
+	struct sk_buff *skb;
+	void *msg_head;
+
+	WARN_ON(!is_valid_ether_addr(addr));
+
+	if (!_portid && !hwsim_virtio_enabled)
+		return;
+
+	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_ATOMIC);
+	if (!skb)
+		return;
+
+	msg_head = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0,
+			       add ? HWSIM_CMD_ADD_MAC_ADDR :
+				     HWSIM_CMD_DEL_MAC_ADDR);
+	if (!msg_head) {
+		pr_debug("mac80211_hwsim: problem with msg_head\n");
+		goto nla_put_failure;
+	}
+
+	if (nla_put(skb, HWSIM_ATTR_ADDR_TRANSMITTER,
+		    ETH_ALEN, data->addresses[1].addr))
+		goto nla_put_failure;
+
+	if (nla_put(skb, HWSIM_ATTR_ADDR_RECEIVER, ETH_ALEN, addr))
+		goto nla_put_failure;
+
+	genlmsg_end(skb, msg_head);
+
+	if (hwsim_virtio_enabled)
+		hwsim_tx_virtio(data, skb);
+	else
+		hwsim_unicast_netgroup(data, skb, _portid);
+	return;
+nla_put_failure:
+	nlmsg_free(skb);
+}
+
+static inline u16 trans_tx_rate_flags_ieee2hwsim(struct ieee80211_tx_rate *rate)
+{
+	u16 result = 0;
+
+	if (rate->flags & IEEE80211_TX_RC_USE_RTS_CTS)
+		result |= MAC80211_HWSIM_TX_RC_USE_RTS_CTS;
+	if (rate->flags & IEEE80211_TX_RC_USE_CTS_PROTECT)
+		result |= MAC80211_HWSIM_TX_RC_USE_CTS_PROTECT;
+	if (rate->flags & IEEE80211_TX_RC_USE_SHORT_PREAMBLE)
+		result |= MAC80211_HWSIM_TX_RC_USE_SHORT_PREAMBLE;
+	if (rate->flags & IEEE80211_TX_RC_MCS)
+		result |= MAC80211_HWSIM_TX_RC_MCS;
+	if (rate->flags & IEEE80211_TX_RC_GREEN_FIELD)
+		result |= MAC80211_HWSIM_TX_RC_GREEN_FIELD;
+	if (rate->flags & IEEE80211_TX_RC_40_MHZ_WIDTH)
+		result |= MAC80211_HWSIM_TX_RC_40_MHZ_WIDTH;
+	if (rate->flags & IEEE80211_TX_RC_DUP_DATA)
+		result |= MAC80211_HWSIM_TX_RC_DUP_DATA;
+	if (rate->flags & IEEE80211_TX_RC_SHORT_GI)
+		result |= MAC80211_HWSIM_TX_RC_SHORT_GI;
+	if (rate->flags & IEEE80211_TX_RC_VHT_MCS)
+		result |= MAC80211_HWSIM_TX_RC_VHT_MCS;
+	if (rate->flags & IEEE80211_TX_RC_80_MHZ_WIDTH)
+		result |= MAC80211_HWSIM_TX_RC_80_MHZ_WIDTH;
+	if (rate->flags & IEEE80211_TX_RC_160_MHZ_WIDTH)
+		result |= MAC80211_HWSIM_TX_RC_160_MHZ_WIDTH;
+
+	return result;
+}
+
+static void mac80211_hwsim_tx_frame_nl(struct ieee80211_hw *hw,
+				       struct sk_buff *my_skb,
+				       int dst_portid,
+				       struct ieee80211_channel *channel)
+{
+	struct sk_buff *skb;
+	struct mac80211_hwsim_data *data = hw->priv;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) my_skb->data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(my_skb);
+	void *msg_head;
+	unsigned int hwsim_flags = 0;
+	int i;
+	struct hwsim_tx_rate tx_attempts[IEEE80211_TX_MAX_RATES];
+	struct hwsim_tx_rate_flag tx_attempts_flags[IEEE80211_TX_MAX_RATES];
+	uintptr_t cookie;
+
+	if (data->ps != PS_DISABLED)
+		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
+	/* If the queue contains MAX_QUEUE skb's drop some */
+	if (skb_queue_len(&data->pending) >= MAX_QUEUE) {
+		/* Dropping until WARN_QUEUE level */
+		while (skb_queue_len(&data->pending) >= WARN_QUEUE) {
+			ieee80211_free_txskb(hw, skb_dequeue(&data->pending));
+			data->tx_dropped++;
+		}
+	}
+
+	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_ATOMIC);
+	if (skb == NULL)
+		goto nla_put_failure;
+
+	msg_head = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0,
+			       HWSIM_CMD_FRAME);
+	if (msg_head == NULL) {
+		pr_debug("mac80211_hwsim: problem with msg_head\n");
+		goto nla_put_failure;
+	}
+
+	if (nla_put(skb, HWSIM_ATTR_ADDR_TRANSMITTER,
+		    ETH_ALEN, data->addresses[1].addr))
+		goto nla_put_failure;
+
+	/* We get the skb->data */
+	if (nla_put(skb, HWSIM_ATTR_FRAME, my_skb->len, my_skb->data))
+		goto nla_put_failure;
+
+	/* We get the flags for this transmission, and we translate them to
+	   wmediumd flags  */
+
+	if (info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS)
+		hwsim_flags |= HWSIM_TX_CTL_REQ_TX_STATUS;
+
+	if (info->flags & IEEE80211_TX_CTL_NO_ACK)
+		hwsim_flags |= HWSIM_TX_CTL_NO_ACK;
+
+	if (nla_put_u32(skb, HWSIM_ATTR_FLAGS, hwsim_flags))
+		goto nla_put_failure;
+
+	if (nla_put_u32(skb, HWSIM_ATTR_FREQ, channel->center_freq))
+		goto nla_put_failure;
+
+	/* We get the tx control (rate and retries) info*/
+
+	for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) {
+		tx_attempts[i].idx = info->status.rates[i].idx;
+		tx_attempts_flags[i].idx = info->status.rates[i].idx;
+		tx_attempts[i].count = info->status.rates[i].count;
+		tx_attempts_flags[i].flags =
+				trans_tx_rate_flags_ieee2hwsim(
+						&info->status.rates[i]);
+	}
+
+	if (nla_put(skb, HWSIM_ATTR_TX_INFO,
+		    sizeof(struct hwsim_tx_rate)*IEEE80211_TX_MAX_RATES,
+		    tx_attempts))
+		goto nla_put_failure;
+
+	if (nla_put(skb, HWSIM_ATTR_TX_INFO_FLAGS,
+		    sizeof(struct hwsim_tx_rate_flag) * IEEE80211_TX_MAX_RATES,
+		    tx_attempts_flags))
+		goto nla_put_failure;
+
+	/* We create a cookie to identify this skb */
+	cookie = atomic_inc_return(&data->pending_cookie);
+	info->rate_driver_data[0] = (void *)cookie;
+	if (nla_put_u64_64bit(skb, HWSIM_ATTR_COOKIE, cookie, HWSIM_ATTR_PAD))
+		goto nla_put_failure;
+
+	genlmsg_end(skb, msg_head);
+
+	if (hwsim_virtio_enabled) {
+		if (hwsim_tx_virtio(data, skb))
+			goto err_free_txskb;
+	} else {
+		if (hwsim_unicast_netgroup(data, skb, dst_portid))
+			goto err_free_txskb;
+	}
+
+	/* Enqueue the packet */
+	skb_queue_tail(&data->pending, my_skb);
+	data->tx_pkts++;
+	data->tx_bytes += my_skb->len;
+	return;
+
+nla_put_failure:
+	nlmsg_free(skb);
+err_free_txskb:
+	pr_debug("mac80211_hwsim: error occurred in %s\n", __func__);
+	ieee80211_free_txskb(hw, my_skb);
+	data->tx_failed++;
+}
+
+static bool hwsim_chans_compat(struct ieee80211_channel *c1,
+			       struct ieee80211_channel *c2)
+{
+	if (!c1 || !c2)
+		return false;
+
+	return c1->center_freq == c2->center_freq;
+}
+
+struct tx_iter_data {
+	struct ieee80211_channel *channel;
+	bool receive;
+};
+
+static void mac80211_hwsim_tx_iter(void *_data, u8 *addr,
+				   struct ieee80211_vif *vif)
+{
+	struct tx_iter_data *data = _data;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(vif->link_conf); i++) {
+		struct ieee80211_bss_conf *conf;
+		struct ieee80211_chanctx_conf *chanctx;
+
+		conf = rcu_dereference(vif->link_conf[i]);
+		if (!conf)
+			continue;
+
+		chanctx = rcu_dereference(conf->chanctx_conf);
+		if (!chanctx)
+			continue;
+
+		if (!hwsim_chans_compat(data->channel, chanctx->def.chan))
+			continue;
+
+		data->receive = true;
+		return;
+	}
+}
+
+static void mac80211_hwsim_add_vendor_rtap(struct sk_buff *skb)
+{
+	/*
+	 * To enable this code, #define the HWSIM_RADIOTAP_OUI,
+	 * e.g. like this:
+	 * #define HWSIM_RADIOTAP_OUI "\x02\x00\x00"
+	 * (but you should use a valid OUI, not that)
+	 *
+	 * If anyone wants to 'donate' a radiotap OUI/subns code
+	 * please send a patch removing this #ifdef and changing
+	 * the values accordingly.
+	 */
+#ifdef HWSIM_RADIOTAP_OUI
+	struct ieee80211_radiotap_vendor_tlv *rtap;
+	static const char vendor_data[8] = "ABCDEFGH";
+
+	// Make sure no padding is needed
+	BUILD_BUG_ON(sizeof(vendor_data) % 4);
+	/* this is last radiotap info before the mac header, so
+	 * skb_reset_mac_header for mac8022 to know the end of
+	 * the radiotap TLV/beginning of the 802.11 header
+	 */
+	skb_reset_mac_header(skb);
+
+	/*
+	 * Note that this code requires the headroom in the SKB
+	 * that was allocated earlier.
+	 */
+	rtap = skb_push(skb, sizeof(*rtap) + sizeof(vendor_data));
+
+	rtap->len = cpu_to_le16(sizeof(*rtap) -
+				sizeof(struct ieee80211_radiotap_tlv) +
+				sizeof(vendor_data));
+	rtap->type = cpu_to_le16(IEEE80211_RADIOTAP_VENDOR_NAMESPACE);
+
+	rtap->content.oui[0] = HWSIM_RADIOTAP_OUI[0];
+	rtap->content.oui[1] = HWSIM_RADIOTAP_OUI[1];
+	rtap->content.oui[2] = HWSIM_RADIOTAP_OUI[2];
+	rtap->content.oui_subtype = 127;
+	/* clear reserved field */
+	rtap->content.reserved = 0;
+	rtap->content.vendor_type = 0;
+	memcpy(rtap->content.data, vendor_data, sizeof(vendor_data));
+
+	IEEE80211_SKB_RXCB(skb)->flag |= RX_FLAG_RADIOTAP_TLV_AT_END;
+#endif
+}
+
+static void mac80211_hwsim_rx(struct mac80211_hwsim_data *data,
+			      struct ieee80211_rx_status *rx_status,
+			      struct sk_buff *skb)
+{
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+
+	if (!ieee80211_has_morefrags(hdr->frame_control) &&
+	    !is_multicast_ether_addr(hdr->addr1) &&
+	    (ieee80211_is_mgmt(hdr->frame_control) ||
+	     ieee80211_is_data(hdr->frame_control))) {
+		struct ieee80211_sta *sta;
+		unsigned int link_id;
+
+		rcu_read_lock();
+		sta = ieee80211_find_sta_by_link_addrs(data->hw, hdr->addr2,
+						       hdr->addr1, &link_id);
+		if (sta) {
+			struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+
+			if (ieee80211_has_pm(hdr->frame_control))
+				sp->active_links_rx &= ~BIT(link_id);
+			else
+				sp->active_links_rx |= BIT(link_id);
+		}
+		rcu_read_unlock();
+	}
+
+	memcpy(IEEE80211_SKB_RXCB(skb), rx_status, sizeof(*rx_status));
+
+	mac80211_hwsim_add_vendor_rtap(skb);
+
+	data->rx_pkts++;
+	data->rx_bytes += skb->len;
+	ieee80211_rx_irqsafe(data->hw, skb);
+}
+
+static bool mac80211_hwsim_tx_frame_no_nl(struct ieee80211_hw *hw,
+					  struct sk_buff *skb,
+					  struct ieee80211_channel *chan)
+{
+	struct mac80211_hwsim_data *data = hw->priv, *data2;
+	bool ack = false;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_rx_status rx_status;
+	u64 now;
+
+	memset(&rx_status, 0, sizeof(rx_status));
+	rx_status.flag |= RX_FLAG_MACTIME_START;
+	rx_status.freq = chan->center_freq;
+	rx_status.freq_offset = chan->freq_offset ? 1 : 0;
+	rx_status.band = chan->band;
+	if (info->control.rates[0].flags & IEEE80211_TX_RC_VHT_MCS) {
+		rx_status.rate_idx =
+			ieee80211_rate_get_vht_mcs(&info->control.rates[0]);
+		rx_status.nss =
+			ieee80211_rate_get_vht_nss(&info->control.rates[0]);
+		rx_status.encoding = RX_ENC_VHT;
+	} else {
+		rx_status.rate_idx = info->control.rates[0].idx;
+		if (info->control.rates[0].flags & IEEE80211_TX_RC_MCS)
+			rx_status.encoding = RX_ENC_HT;
+	}
+	if (info->control.rates[0].flags & IEEE80211_TX_RC_40_MHZ_WIDTH)
+		rx_status.bw = RATE_INFO_BW_40;
+	else if (info->control.rates[0].flags & IEEE80211_TX_RC_80_MHZ_WIDTH)
+		rx_status.bw = RATE_INFO_BW_80;
+	else if (info->control.rates[0].flags & IEEE80211_TX_RC_160_MHZ_WIDTH)
+		rx_status.bw = RATE_INFO_BW_160;
+	else
+		rx_status.bw = RATE_INFO_BW_20;
+	if (info->control.rates[0].flags & IEEE80211_TX_RC_SHORT_GI)
+		rx_status.enc_flags |= RX_ENC_FLAG_SHORT_GI;
+	/* TODO: simulate optional packet loss */
+	rx_status.signal = data->rx_rssi;
+	if (info->control.vif)
+		rx_status.signal += info->control.vif->bss_conf.txpower;
+
+	if (data->ps != PS_DISABLED)
+		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
+
+	/* release the skb's source info */
+	skb_orphan(skb);
+	skb_dst_drop(skb);
+	skb->mark = 0;
+	skb_ext_reset(skb);
+	nf_reset_ct(skb);
+
+	/*
+	 * Get absolute mactime here so all HWs RX at the "same time", and
+	 * absolute TX time for beacon mactime so the timestamp matches.
+	 * Giving beacons a different mactime than non-beacons looks messy, but
+	 * it helps the Toffset be exact and a ~10us mactime discrepancy
+	 * probably doesn't really matter.
+	 */
+	if (ieee80211_is_beacon(hdr->frame_control) ||
+	    ieee80211_is_probe_resp(hdr->frame_control)) {
+		rx_status.boottime_ns = ktime_get_boottime_ns();
+		now = data->abs_bcn_ts;
+	} else {
+		now = mac80211_hwsim_get_tsf_raw();
+	}
+
+	/* Copy skb to all enabled radios that are on the current frequency */
+	spin_lock(&hwsim_radio_lock);
+	list_for_each_entry(data2, &hwsim_radios, list) {
+		struct sk_buff *nskb;
+		struct tx_iter_data tx_iter_data = {
+			.receive = false,
+			.channel = chan,
+		};
+
+		if (data == data2)
+			continue;
+
+		if (!data2->started || (data2->idle && !data2->tmp_chan) ||
+		    !hwsim_ps_rx_ok(data2, skb))
+			continue;
+
+		if (!(data->group & data2->group))
+			continue;
+
+		if (data->netgroup != data2->netgroup)
+			continue;
+
+		if (!hwsim_chans_compat(chan, data2->tmp_chan) &&
+		    !hwsim_chans_compat(chan, data2->channel)) {
+			ieee80211_iterate_active_interfaces_atomic(
+				data2->hw, IEEE80211_IFACE_ITER_NORMAL,
+				mac80211_hwsim_tx_iter, &tx_iter_data);
+			if (!tx_iter_data.receive)
+				continue;
+		}
+
+		/*
+		 * reserve some space for our vendor and the normal
+		 * radiotap header, since we're copying anyway
+		 */
+		if (skb->len < PAGE_SIZE && paged_rx) {
+			struct page *page = alloc_page(GFP_ATOMIC);
+
+			if (!page)
+				continue;
+
+			nskb = dev_alloc_skb(128);
+			if (!nskb) {
+				__free_page(page);
+				continue;
+			}
+
+			memcpy(page_address(page), skb->data, skb->len);
+			skb_add_rx_frag(nskb, 0, page, 0, skb->len, skb->len);
+		} else {
+			nskb = skb_copy(skb, GFP_ATOMIC);
+			if (!nskb)
+				continue;
+		}
+
+		if (mac80211_hwsim_addr_match(data2, hdr->addr1))
+			ack = true;
+
+		rx_status.mactime = now + data2->tsf_offset;
+
+		mac80211_hwsim_rx(data2, &rx_status, nskb);
+	}
+	spin_unlock(&hwsim_radio_lock);
+
+	return ack;
+}
+
+static struct ieee80211_bss_conf *
+mac80211_hwsim_select_tx_link(struct mac80211_hwsim_data *data,
+			      struct ieee80211_vif *vif,
+			      struct ieee80211_sta *sta,
+			      struct ieee80211_hdr *hdr,
+			      struct ieee80211_link_sta **link_sta)
+{
+	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+	int i;
+
+	if (!vif->valid_links)
+		return &vif->bss_conf;
+
+	WARN_ON(is_multicast_ether_addr(hdr->addr1));
+
+	if (WARN_ON_ONCE(!sta->valid_links))
+		return &vif->bss_conf;
+
+	for (i = 0; i < ARRAY_SIZE(vif->link_conf); i++) {
+		struct ieee80211_bss_conf *bss_conf;
+		unsigned int link_id;
+
+		/* round-robin the available link IDs */
+		link_id = (sp->last_link + i + 1) % ARRAY_SIZE(vif->link_conf);
+
+		if (!(vif->active_links & BIT(link_id)))
+			continue;
+
+		if (!(sp->active_links_rx & BIT(link_id)))
+			continue;
+
+		*link_sta = rcu_dereference(sta->link[link_id]);
+		if (!*link_sta)
+			continue;
+
+		bss_conf = rcu_dereference(vif->link_conf[link_id]);
+		if (WARN_ON_ONCE(!bss_conf))
+			continue;
+
+		/* can happen while switching links */
+		if (!rcu_access_pointer(bss_conf->chanctx_conf))
+			continue;
+
+		sp->last_link = link_id;
+		return bss_conf;
+	}
+
+	return NULL;
+}
+
+static void mac80211_hwsim_tx(struct ieee80211_hw *hw,
+			      struct ieee80211_tx_control *control,
+			      struct sk_buff *skb)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	struct ieee80211_tx_info *txi = IEEE80211_SKB_CB(skb);
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+	struct ieee80211_chanctx_conf *chanctx_conf;
+	struct ieee80211_channel *channel;
+	bool ack;
+	enum nl80211_chan_width confbw = NL80211_CHAN_WIDTH_20_NOHT;
+	u32 _portid, i;
+
+	if (WARN_ON(skb->len < 10)) {
+		/* Should not happen; just a sanity check for addr1 use */
+		ieee80211_free_txskb(hw, skb);
+		return;
+	}
+
+	if (!data->use_chanctx) {
+		channel = data->channel;
+		confbw = data->bw;
+	} else if (txi->hw_queue == 4) {
+		channel = data->tmp_chan;
+	} else {
+		u8 link = u32_get_bits(IEEE80211_SKB_CB(skb)->control.flags,
+				       IEEE80211_TX_CTRL_MLO_LINK);
+		struct ieee80211_vif *vif = txi->control.vif;
+		struct ieee80211_link_sta *link_sta = NULL;
+		struct ieee80211_sta *sta = control->sta;
+		struct ieee80211_bss_conf *bss_conf;
+
+		if (link != IEEE80211_LINK_UNSPECIFIED) {
+			bss_conf = rcu_dereference(txi->control.vif->link_conf[link]);
+			if (sta)
+				link_sta = rcu_dereference(sta->link[link]);
+		} else {
+			bss_conf = mac80211_hwsim_select_tx_link(data, vif, sta,
+								 hdr, &link_sta);
+		}
+
+		if (WARN_ON(!bss_conf)) {
+			ieee80211_free_txskb(hw, skb);
+			return;
+		}
+
+		if (sta && sta->mlo) {
+			if (WARN_ON(!link_sta)) {
+				ieee80211_free_txskb(hw, skb);
+				return;
+			}
+			/* address translation to link addresses on TX */
+			ether_addr_copy(hdr->addr1, link_sta->addr);
+			ether_addr_copy(hdr->addr2, bss_conf->addr);
+			/* translate A3 only if it's the BSSID */
+			if (!ieee80211_has_tods(hdr->frame_control) &&
+			    !ieee80211_has_fromds(hdr->frame_control)) {
+				if (ether_addr_equal(hdr->addr3, sta->addr))
+					ether_addr_copy(hdr->addr3, link_sta->addr);
+				else if (ether_addr_equal(hdr->addr3, vif->addr))
+					ether_addr_copy(hdr->addr3, bss_conf->addr);
+			}
+			/* no need to look at A4, if present it's SA */
+		}
+
+		chanctx_conf = rcu_dereference(bss_conf->chanctx_conf);
+		if (chanctx_conf) {
+			channel = chanctx_conf->def.chan;
+			confbw = chanctx_conf->def.width;
+		} else {
+			channel = NULL;
+		}
+	}
+
+	if (WARN(!channel, "TX w/o channel - queue = %d\n", txi->hw_queue)) {
+		ieee80211_free_txskb(hw, skb);
+		return;
+	}
+
+	if (data->idle && !data->tmp_chan) {
+		wiphy_dbg(hw->wiphy, "Trying to TX when idle - reject\n");
+		ieee80211_free_txskb(hw, skb);
+		return;
+	}
+
+	if (txi->control.vif)
+		hwsim_check_magic(txi->control.vif);
+	if (control->sta)
+		hwsim_check_sta_magic(control->sta);
+
+	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE))
+		ieee80211_get_tx_rates(txi->control.vif, control->sta, skb,
+				       txi->control.rates,
+				       ARRAY_SIZE(txi->control.rates));
+
+	for (i = 0; i < ARRAY_SIZE(txi->control.rates); i++) {
+		u16 rflags = txi->control.rates[i].flags;
+		/* initialize to data->bw for 5/10 MHz handling */
+		enum nl80211_chan_width bw = data->bw;
+
+		if (txi->control.rates[i].idx == -1)
+			break;
+
+		if (rflags & IEEE80211_TX_RC_40_MHZ_WIDTH)
+			bw = NL80211_CHAN_WIDTH_40;
+		else if (rflags & IEEE80211_TX_RC_80_MHZ_WIDTH)
+			bw = NL80211_CHAN_WIDTH_80;
+		else if (rflags & IEEE80211_TX_RC_160_MHZ_WIDTH)
+			bw = NL80211_CHAN_WIDTH_160;
+
+		if (WARN_ON(hwsim_get_chanwidth(bw) > hwsim_get_chanwidth(confbw)))
+			return;
+	}
+
+	if (skb->len >= 24 + 8 &&
+	    ieee80211_is_probe_resp(hdr->frame_control)) {
+		/* fake header transmission time */
+		struct ieee80211_mgmt *mgmt;
+		struct ieee80211_rate *txrate;
+		/* TODO: get MCS */
+		int bitrate = 100;
+		u64 ts;
+
+		mgmt = (struct ieee80211_mgmt *)skb->data;
+		txrate = ieee80211_get_tx_rate(hw, txi);
+		if (txrate)
+			bitrate = txrate->bitrate;
+		ts = mac80211_hwsim_get_tsf_raw();
+		mgmt->u.probe_resp.timestamp =
+			cpu_to_le64(ts + data->tsf_offset +
+				    24 * 8 * 10 / bitrate);
+	}
+
+	mac80211_hwsim_monitor_rx(hw, skb, channel);
+
+	/* wmediumd mode check */
+	_portid = READ_ONCE(data->wmediumd);
+
+	if (_portid || hwsim_virtio_enabled)
+		return mac80211_hwsim_tx_frame_nl(hw, skb, _portid, channel);
+
+	/* NO wmediumd detected, perfect medium simulation */
+	data->tx_pkts++;
+	data->tx_bytes += skb->len;
+	ack = mac80211_hwsim_tx_frame_no_nl(hw, skb, channel);
+
+	if (ack && skb->len >= 16)
+		mac80211_hwsim_monitor_ack(channel, hdr->addr2);
+
+	ieee80211_tx_info_clear_status(txi);
+
+	/* frame was transmitted at most favorable rate at first attempt */
+	txi->control.rates[0].count = 1;
+	txi->control.rates[1].idx = -1;
+
+	if (!(txi->flags & IEEE80211_TX_CTL_NO_ACK) && ack)
+		txi->flags |= IEEE80211_TX_STAT_ACK;
+	ieee80211_tx_status_irqsafe(hw, skb);
+}
+
+
+static int mac80211_hwsim_start(struct ieee80211_hw *hw)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	wiphy_dbg(hw->wiphy, "%s\n", __func__);
+	data->started = true;
+	return 0;
+}
+
+
+static void mac80211_hwsim_stop(struct ieee80211_hw *hw)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	int i;
+
+	data->started = false;
+
+	for (i = 0; i < ARRAY_SIZE(data->link_data); i++)
+		hrtimer_cancel(&data->link_data[i].beacon_timer);
+
+	while (!skb_queue_empty(&data->pending))
+		ieee80211_free_txskb(hw, skb_dequeue(&data->pending));
+
+	wiphy_dbg(hw->wiphy, "%s\n", __func__);
+}
+
+
+static int mac80211_hwsim_add_interface(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif)
+{
+	wiphy_dbg(hw->wiphy, "%s (type=%d mac_addr=%pM)\n",
+		  __func__, ieee80211_vif_type_p2p(vif),
+		  vif->addr);
+	hwsim_set_magic(vif);
+
+	if (vif->type != NL80211_IFTYPE_MONITOR)
+		mac80211_hwsim_config_mac_nl(hw, vif->addr, true);
+
+	vif->cab_queue = 0;
+	vif->hw_queue[IEEE80211_AC_VO] = 0;
+	vif->hw_queue[IEEE80211_AC_VI] = 1;
+	vif->hw_queue[IEEE80211_AC_BE] = 2;
+	vif->hw_queue[IEEE80211_AC_BK] = 3;
+
+	return 0;
+}
+
+
+static int mac80211_hwsim_change_interface(struct ieee80211_hw *hw,
+					   struct ieee80211_vif *vif,
+					   enum nl80211_iftype newtype,
+					   bool newp2p)
+{
+	newtype = ieee80211_iftype_p2p(newtype, newp2p);
+	wiphy_dbg(hw->wiphy,
+		  "%s (old type=%d, new type=%d, mac_addr=%pM)\n",
+		  __func__, ieee80211_vif_type_p2p(vif),
+		    newtype, vif->addr);
+	hwsim_check_magic(vif);
+
+	/*
+	 * interface may change from non-AP to AP in
+	 * which case this needs to be set up again
+	 */
+	vif->cab_queue = 0;
+
+	return 0;
+}
+
+static void mac80211_hwsim_remove_interface(
+	struct ieee80211_hw *hw, struct ieee80211_vif *vif)
+{
+	wiphy_dbg(hw->wiphy, "%s (type=%d mac_addr=%pM)\n",
+		  __func__, ieee80211_vif_type_p2p(vif),
+		  vif->addr);
+	hwsim_check_magic(vif);
+	hwsim_clear_magic(vif);
+	if (vif->type != NL80211_IFTYPE_MONITOR)
+		mac80211_hwsim_config_mac_nl(hw, vif->addr, false);
+}
+
+static void mac80211_hwsim_tx_frame(struct ieee80211_hw *hw,
+				    struct sk_buff *skb,
+				    struct ieee80211_channel *chan)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	u32 _portid = READ_ONCE(data->wmediumd);
+
+	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE)) {
+		struct ieee80211_tx_info *txi = IEEE80211_SKB_CB(skb);
+		ieee80211_get_tx_rates(txi->control.vif, NULL, skb,
+				       txi->control.rates,
+				       ARRAY_SIZE(txi->control.rates));
+	}
+
+	mac80211_hwsim_monitor_rx(hw, skb, chan);
+
+	if (_portid || hwsim_virtio_enabled)
+		return mac80211_hwsim_tx_frame_nl(hw, skb, _portid, chan);
+
+	data->tx_pkts++;
+	data->tx_bytes += skb->len;
+	mac80211_hwsim_tx_frame_no_nl(hw, skb, chan);
+	dev_kfree_skb(skb);
+}
+
+static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
+				     struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_link_data *link_data = arg;
+	u32 link_id = link_data->link_id;
+	struct ieee80211_bss_conf *link_conf;
+	struct mac80211_hwsim_data *data =
+		container_of(link_data, struct mac80211_hwsim_data,
+			     link_data[link_id]);
+	struct ieee80211_hw *hw = data->hw;
+	struct ieee80211_tx_info *info;
+	struct ieee80211_rate *txrate;
+	struct ieee80211_mgmt *mgmt;
+	struct sk_buff *skb;
+	/* TODO: get MCS */
+	int bitrate = 100;
+
+	hwsim_check_magic(vif);
+
+	link_conf = rcu_dereference(vif->link_conf[link_id]);
+	if (!link_conf)
+		return;
+
+	if (vif->type != NL80211_IFTYPE_AP &&
+	    vif->type != NL80211_IFTYPE_MESH_POINT &&
+	    vif->type != NL80211_IFTYPE_ADHOC &&
+	    vif->type != NL80211_IFTYPE_OCB)
+		return;
+
+	skb = ieee80211_beacon_get(hw, vif, link_data->link_id);
+	if (skb == NULL)
+		return;
+	info = IEEE80211_SKB_CB(skb);
+	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE))
+		ieee80211_get_tx_rates(vif, NULL, skb,
+				       info->control.rates,
+				       ARRAY_SIZE(info->control.rates));
+
+	txrate = ieee80211_get_tx_rate(hw, info);
+	if (txrate)
+		bitrate = txrate->bitrate;
+
+	mgmt = (struct ieee80211_mgmt *) skb->data;
+	/* fake header transmission time */
+	data->abs_bcn_ts = mac80211_hwsim_get_tsf_raw();
+	if (ieee80211_is_s1g_beacon(mgmt->frame_control)) {
+		struct ieee80211_ext *ext = (void *) mgmt;
+
+		ext->u.s1g_beacon.timestamp = cpu_to_le32(data->abs_bcn_ts +
+							  data->tsf_offset +
+							  10 * 8 * 10 /
+							  bitrate);
+	} else {
+		mgmt->u.beacon.timestamp = cpu_to_le64(data->abs_bcn_ts +
+						       data->tsf_offset +
+						       24 * 8 * 10 /
+						       bitrate);
+	}
+
+	mac80211_hwsim_tx_frame(hw, skb,
+			rcu_dereference(link_conf->chanctx_conf)->def.chan);
+
+	while ((skb = ieee80211_get_buffered_bc(hw, vif)) != NULL) {
+		mac80211_hwsim_tx_frame(hw, skb,
+			rcu_dereference(link_conf->chanctx_conf)->def.chan);
+	}
+
+	if (link_conf->csa_active && ieee80211_beacon_cntdwn_is_complete(vif))
+		ieee80211_csa_finish(vif);
+}
+
+static enum hrtimer_restart
+mac80211_hwsim_beacon(struct hrtimer *timer)
+{
+	struct mac80211_hwsim_link_data *link_data =
+		container_of(timer, struct mac80211_hwsim_link_data, beacon_timer);
+	struct mac80211_hwsim_data *data =
+		container_of(link_data, struct mac80211_hwsim_data,
+			     link_data[link_data->link_id]);
+	struct ieee80211_hw *hw = data->hw;
+	u64 bcn_int = link_data->beacon_int;
+
+	if (!data->started)
+		return HRTIMER_NORESTART;
+
+	ieee80211_iterate_active_interfaces_atomic(
+		hw, IEEE80211_IFACE_ITER_NORMAL,
+		mac80211_hwsim_beacon_tx, link_data);
+
+	/* beacon at new TBTT + beacon interval */
+	if (data->bcn_delta) {
+		bcn_int -= data->bcn_delta;
+		data->bcn_delta = 0;
+	}
+	hrtimer_forward_now(&link_data->beacon_timer,
+			    ns_to_ktime(bcn_int * NSEC_PER_USEC));
+	return HRTIMER_RESTART;
+}
+
+static const char * const hwsim_chanwidths[] = {
+	[NL80211_CHAN_WIDTH_5] = "ht5",
+	[NL80211_CHAN_WIDTH_10] = "ht10",
+	[NL80211_CHAN_WIDTH_20_NOHT] = "noht",
+	[NL80211_CHAN_WIDTH_20] = "ht20",
+	[NL80211_CHAN_WIDTH_40] = "ht40",
+	[NL80211_CHAN_WIDTH_80] = "vht80",
+	[NL80211_CHAN_WIDTH_80P80] = "vht80p80",
+	[NL80211_CHAN_WIDTH_160] = "vht160",
+	[NL80211_CHAN_WIDTH_1] = "1MHz",
+	[NL80211_CHAN_WIDTH_2] = "2MHz",
+	[NL80211_CHAN_WIDTH_4] = "4MHz",
+	[NL80211_CHAN_WIDTH_8] = "8MHz",
+	[NL80211_CHAN_WIDTH_16] = "16MHz",
+};
+
+static int mac80211_hwsim_config(struct ieee80211_hw *hw, u32 changed)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	struct ieee80211_conf *conf = &hw->conf;
+	static const char *smps_modes[IEEE80211_SMPS_NUM_MODES] = {
+		[IEEE80211_SMPS_AUTOMATIC] = "auto",
+		[IEEE80211_SMPS_OFF] = "off",
+		[IEEE80211_SMPS_STATIC] = "static",
+		[IEEE80211_SMPS_DYNAMIC] = "dynamic",
+	};
+	int idx;
+
+	if (conf->chandef.chan)
+		wiphy_dbg(hw->wiphy,
+			  "%s (freq=%d(%d - %d)/%s idle=%d ps=%d smps=%s)\n",
+			  __func__,
+			  conf->chandef.chan->center_freq,
+			  conf->chandef.center_freq1,
+			  conf->chandef.center_freq2,
+			  hwsim_chanwidths[conf->chandef.width],
+			  !!(conf->flags & IEEE80211_CONF_IDLE),
+			  !!(conf->flags & IEEE80211_CONF_PS),
+			  smps_modes[conf->smps_mode]);
+	else
+		wiphy_dbg(hw->wiphy,
+			  "%s (freq=0 idle=%d ps=%d smps=%s)\n",
+			  __func__,
+			  !!(conf->flags & IEEE80211_CONF_IDLE),
+			  !!(conf->flags & IEEE80211_CONF_PS),
+			  smps_modes[conf->smps_mode]);
+
+	data->idle = !!(conf->flags & IEEE80211_CONF_IDLE);
+
+	WARN_ON(conf->chandef.chan && data->use_chanctx);
+
+	mutex_lock(&data->mutex);
+	if (data->scanning && conf->chandef.chan) {
+		for (idx = 0; idx < ARRAY_SIZE(data->survey_data); idx++) {
+			if (data->survey_data[idx].channel == data->channel) {
+				data->survey_data[idx].start =
+					data->survey_data[idx].next_start;
+				data->survey_data[idx].end = jiffies;
+				break;
+			}
+		}
+
+		data->channel = conf->chandef.chan;
+		data->bw = conf->chandef.width;
+
+		for (idx = 0; idx < ARRAY_SIZE(data->survey_data); idx++) {
+			if (data->survey_data[idx].channel &&
+			    data->survey_data[idx].channel != data->channel)
+				continue;
+			data->survey_data[idx].channel = data->channel;
+			data->survey_data[idx].next_start = jiffies;
+			break;
+		}
+	} else {
+		data->channel = conf->chandef.chan;
+		data->bw = conf->chandef.width;
+	}
+	mutex_unlock(&data->mutex);
+
+	for (idx = 0; idx < ARRAY_SIZE(data->link_data); idx++) {
+		struct mac80211_hwsim_link_data *link_data =
+			&data->link_data[idx];
+
+		if (!data->started || !link_data->beacon_int) {
+			hrtimer_cancel(&link_data->beacon_timer);
+		} else if (!hrtimer_is_queued(&link_data->beacon_timer)) {
+			u64 tsf = mac80211_hwsim_get_tsf(hw, NULL);
+			u32 bcn_int = link_data->beacon_int;
+			u64 until_tbtt = bcn_int - do_div(tsf, bcn_int);
+
+			hrtimer_start(&link_data->beacon_timer,
+				      ns_to_ktime(until_tbtt * NSEC_PER_USEC),
+				      HRTIMER_MODE_REL_SOFT);
+		}
+	}
+
+	return 0;
+}
+
+
+static void mac80211_hwsim_configure_filter(struct ieee80211_hw *hw,
+					    unsigned int changed_flags,
+					    unsigned int *total_flags,u64 multicast)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+
+	wiphy_dbg(hw->wiphy, "%s\n", __func__);
+
+	data->rx_filter = 0;
+	if (*total_flags & FIF_ALLMULTI)
+		data->rx_filter |= FIF_ALLMULTI;
+	if (*total_flags & FIF_MCAST_ACTION)
+		data->rx_filter |= FIF_MCAST_ACTION;
+
+	*total_flags = data->rx_filter;
+}
+
+static void mac80211_hwsim_bcn_en_iter(void *data, u8 *mac,
+				       struct ieee80211_vif *vif)
+{
+	unsigned int *count = data;
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+
+	if (vp->bcn_en)
+		(*count)++;
+}
+
+static void mac80211_hwsim_vif_info_changed(struct ieee80211_hw *hw,
+					    struct ieee80211_vif *vif,
+					    u64 changed)
+{
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+
+	hwsim_check_magic(vif);
+
+	wiphy_dbg(hw->wiphy, "%s(changed=0x%llx vif->addr=%pM)\n",
+		  __func__, changed, vif->addr);
+
+	if (changed & BSS_CHANGED_ASSOC) {
+		wiphy_dbg(hw->wiphy, "  ASSOC: assoc=%d aid=%d\n",
+			  vif->cfg.assoc, vif->cfg.aid);
+		vp->assoc = vif->cfg.assoc;
+		vp->aid = vif->cfg.aid;
+	}
+}
+
+static void mac80211_hwsim_link_info_changed(struct ieee80211_hw *hw,
+					     struct ieee80211_vif *vif,
+					     struct ieee80211_bss_conf *info,
+					     u64 changed)
+{
+	struct hwsim_vif_priv *vp = (void *)vif->drv_priv;
+	struct mac80211_hwsim_data *data = hw->priv;
+	unsigned int link_id = info->link_id;
+	struct mac80211_hwsim_link_data *link_data = &data->link_data[link_id];
+
+	hwsim_check_magic(vif);
+
+	wiphy_dbg(hw->wiphy, "%s(changed=0x%llx vif->addr=%pM, link id %u)\n",
+		  __func__, (unsigned long long)changed, vif->addr, link_id);
+
+	if (changed & BSS_CHANGED_BSSID) {
+		wiphy_dbg(hw->wiphy, "%s: BSSID changed: %pM\n",
+			  __func__, info->bssid);
+		memcpy(vp->bssid, info->bssid, ETH_ALEN);
+	}
+
+	if (changed & BSS_CHANGED_BEACON_ENABLED) {
+		wiphy_dbg(hw->wiphy, "  BCN EN: %d (BI=%u)\n",
+			  info->enable_beacon, info->beacon_int);
+		vp->bcn_en = info->enable_beacon;
+		if (data->started &&
+		    !hrtimer_is_queued(&link_data->beacon_timer) &&
+		    info->enable_beacon) {
+			u64 tsf, until_tbtt;
+			u32 bcn_int;
+			link_data->beacon_int = info->beacon_int * 1024;
+			tsf = mac80211_hwsim_get_tsf(hw, vif);
+			bcn_int = link_data->beacon_int;
+			until_tbtt = bcn_int - do_div(tsf, bcn_int);
+
+			hrtimer_start(&link_data->beacon_timer,
+				      ns_to_ktime(until_tbtt * NSEC_PER_USEC),
+				      HRTIMER_MODE_REL_SOFT);
+		} else if (!info->enable_beacon) {
+			unsigned int count = 0;
+			ieee80211_iterate_active_interfaces_atomic(
+				data->hw, IEEE80211_IFACE_ITER_NORMAL,
+				mac80211_hwsim_bcn_en_iter, &count);
+			wiphy_dbg(hw->wiphy, "  beaconing vifs remaining: %u",
+				  count);
+			if (count == 0) {
+				hrtimer_cancel(&link_data->beacon_timer);
+				link_data->beacon_int = 0;
+			}
+		}
+	}
+
+	if (changed & BSS_CHANGED_ERP_CTS_PROT) {
+		wiphy_dbg(hw->wiphy, "  ERP_CTS_PROT: %d\n",
+			  info->use_cts_prot);
+	}
+
+	if (changed & BSS_CHANGED_ERP_PREAMBLE) {
+		wiphy_dbg(hw->wiphy, "  ERP_PREAMBLE: %d\n",
+			  info->use_short_preamble);
+	}
+
+	if (changed & BSS_CHANGED_ERP_SLOT) {
+		wiphy_dbg(hw->wiphy, "  ERP_SLOT: %d\n", info->use_short_slot);
+	}
+
+	if (changed & BSS_CHANGED_HT) {
+		wiphy_dbg(hw->wiphy, "  HT: op_mode=0x%x\n",
+			  info->ht_operation_mode);
+	}
+
+	if (changed & BSS_CHANGED_BASIC_RATES) {
+		wiphy_dbg(hw->wiphy, "  BASIC_RATES: 0x%llx\n",
+			  (unsigned long long) info->basic_rates);
+	}
+
+	if (changed & BSS_CHANGED_TXPOWER)
+		wiphy_dbg(hw->wiphy, "  TX Power: %d dBm\n", info->txpower);
+}
+
+static void
+mac80211_hwsim_sta_rc_update(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_sta *sta,
+			     u32 changed)
+{
+	struct mac80211_hwsim_data *data = hw->priv;
+	u32 bw = U32_MAX;
+	int link_id;
+
+	rcu_read_lock();
+	for (link_id = 0;
+	     link_id < ARRAY_SIZE(vif->link_conf);
+	     link_id++) {
+		enum nl80211_chan_width confbw = NL80211_CHAN_WIDTH_20_NOHT;
+		struct ieee80211_bss_conf *vif_conf;
+		struct ieee80211_link_sta *link_sta;
+
+		link_sta = rcu_dereference(sta->link[link_id]);
+
+		if (!link_sta)
+			continue;
+
+		switch (link_sta->bandwidth) {
+#define C(_bw) case IEEE80211_STA_RX_BW_##_bw: bw = _bw; break
+		C(20);
+		C(40);
+		C(80);
+		C(160);
+		C(320);
+#undef C
+		}
+
+		if (!data->use_chanctx) {
+			confbw = data->bw;
+		} else {
+			struct ieee80211_chanctx_conf *chanctx_conf;
+
+			vif_conf = rcu_dereference(vif->link_conf[link_id]);
+			if (WARN_ON(!vif_conf))
+				continue;
+
+			chanctx_conf = rcu_dereference(vif_conf->chanctx_conf);
+
+			if (!WARN_ON(!chanctx_conf))
+				confbw = chanctx_conf->def.width;
+		}
+
+		WARN(bw > hwsim_get_chanwidth(confbw),
+		     "intf %pM [link=%d]: bad STA %pM bandwidth %d MHz (%d) > channel config %d MHz (%d)\n",
+		     vif->addr, link_id, sta->addr, bw, sta->deflink.bandwidth,
+		     hwsim_get_chanwidth(data->bw), data->bw);
+
+
+	}
+	rcu_read_unlock();
+
+
+}
+
+static int mac80211_hwsim_sta_add(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif,
+				  struct ieee80211_sta *sta)
+{
+	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+
+	hwsim_check_magic(vif);
+	hwsim_set_sta_magic(sta);
+	mac80211_hwsim_sta_rc_update(hw, vif, sta, 0);
+
+	if (sta->valid_links) {
+		WARN(hweight16(sta->valid_links) > 1,
+		     "expect to add STA with single link, have 0x%x\n",
+		     sta->valid_links);
+		sp->active_links_rx = sta->valid_links;
+	}
+
+	return 0;
+}
+
+static int mac80211_hwsim_sta_remove(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_sta *sta)
+{
+	hwsim_check_magic(vif);
+	hwsim_clear_sta_magic(sta);
+
+	return 0;
+}
+
+static int mac80211_hwsim_sta_state(struct ieee80211_hw *hw,
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_sta *sta,
+				    enum ieee80211_sta_state old_state,
+				    enum ieee80211_sta_state new_state)
+{
+	if (new_state == IEEE80211_STA_NOTEXIST)
+		return mac80211_hwsim_sta_remove(hw, vif, sta);
+
+	if (old_state == IEEE80211_STA_NOTEXIST)
+		return mac80211_hwsim_sta_add(hw, vif, sta);
+
+	/*
+	 * when client is authorized (AP station marked as such),
+	 * enable all links
+	 */
+	if (vif->type == NL80211_IFTYPE_STATION &&
+	    new_state == IEEE80211_STA_AUTHORIZED && !sta->tdls)
+		ieee80211_set_active_links_async(vif, vif->valid_links);
+
+	return 0;
+}
+
+static void mac80211_hwsim_sta_notify(struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif,
+				      enum sta_notify_cmd cmd,
+				      struct ieee80211_sta *sta)
+{
+	hwsim_check_magic(vif);
+
+	switch (cmd) {
+	case STA_NOTIFY_SLEEP:
+	case STA_NOTIFY_AWAKE:
+		/* TODO: make good use of these flags */
+		break;
+	default:
+		WARN(1, "Invalid sta notify: %d\n", cmd);
+		break;
+	}
+}
+
+static int mac80211_hwsim_set_tim(struct ieee80211_hw *hw,
+				  struct ieee80211_sta *sta,
+				  bool set)
+{
+	hwsim_check_sta_magic(sta);
+	return 0;
+}
+
+static int mac80211_hwsim_conf_tx(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif,
+				  unsigned int link_id, u16 queue,
+				  const struct ieee80211_tx_queue_params *params)
+{
+	wiphy_dbg(hw->wiphy,
+		  "%s (queue=%d txop=%d cw_min=%d cw_max=%d aifs=%d)\n",
+		  __func__, queue,
+		  params->txop, params->cw_min,
+		  params->cw_max, params->aifs);
+	return 0;
+}
+
+static int mac80211_hwsim_get_survey(struct ieee80211_hw *hw, int idx,
+				     struct survey_info *survey)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+
+	if (idx < 0 || idx >= ARRAY_SIZE(hwsim->survey_data))
+		return -ENOENT;
+
+	mutex_lock(&hwsim->mutex);
+	survey->channel = hwsim->survey_data[idx].channel;
+	if (!survey->channel) {
+		mutex_unlock(&hwsim->mutex);
+		return -ENOENT;
+	}
+
+	/*
+	 * Magically conjured dummy values --- this is only ok for simulated hardware.
+	 *
+	 * A real driver which cannot determine real values noise MUST NOT
+	 * report any, especially not a magically conjured ones :-)
+	 */
+	survey->filled = SURVEY_INFO_NOISE_DBM |
+			 SURVEY_INFO_TIME |
+			 SURVEY_INFO_TIME_BUSY;
+	survey->noise = -92;
+	survey->time =
+		jiffies_to_msecs(hwsim->survey_data[idx].end -
+				 hwsim->survey_data[idx].start);
+	/* report 12.5% of channel time is used */
+	survey->time_busy = survey->time/8;
+	mutex_unlock(&hwsim->mutex);
+
+	return 0;
+}
+
+#ifdef CONFIG_NL80211_TESTMODE
+/*
+ * This section contains example code for using netlink
+ * attributes with the testmode command in nl80211.
+ */
+
+/* These enums need to be kept in sync with userspace */
+enum hwsim_testmode_attr {
+	__HWSIM_TM_ATTR_INVALID	= 0,
+	HWSIM_TM_ATTR_CMD	= 1,
+	HWSIM_TM_ATTR_PS	= 2,
+
+	/* keep last */
+	__HWSIM_TM_ATTR_AFTER_LAST,
+	HWSIM_TM_ATTR_MAX	= __HWSIM_TM_ATTR_AFTER_LAST - 1
+};
+
+enum hwsim_testmode_cmd {
+	HWSIM_TM_CMD_SET_PS		= 0,
+	HWSIM_TM_CMD_GET_PS		= 1,
+	HWSIM_TM_CMD_STOP_QUEUES	= 2,
+	HWSIM_TM_CMD_WAKE_QUEUES	= 3,
+};
+
+static const struct nla_policy hwsim_testmode_policy[HWSIM_TM_ATTR_MAX + 1] = {
+	[HWSIM_TM_ATTR_CMD] = { .type = NLA_U32 },
+	[HWSIM_TM_ATTR_PS] = { .type = NLA_U32 },
+};
+
+static int mac80211_hwsim_testmode_cmd(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       void *data, int len)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+	struct nlattr *tb[HWSIM_TM_ATTR_MAX + 1];
+	struct sk_buff *skb;
+	int err, ps;
+
+	err = nla_parse_deprecated(tb, HWSIM_TM_ATTR_MAX, data, len,
+				   hwsim_testmode_policy, NULL);
+	if (err)
+		return err;
+
+	if (!tb[HWSIM_TM_ATTR_CMD])
+		return -EINVAL;
+
+	switch (nla_get_u32(tb[HWSIM_TM_ATTR_CMD])) {
+	case HWSIM_TM_CMD_SET_PS:
+		if (!tb[HWSIM_TM_ATTR_PS])
+			return -EINVAL;
+		ps = nla_get_u32(tb[HWSIM_TM_ATTR_PS]);
+		return hwsim_fops_ps_write(hwsim, ps);
+	case HWSIM_TM_CMD_GET_PS:
+		skb = cfg80211_testmode_alloc_reply_skb(hw->wiphy,
+						nla_total_size(sizeof(u32)));
+		if (!skb)
+			return -ENOMEM;
+		if (nla_put_u32(skb, HWSIM_TM_ATTR_PS, hwsim->ps))
+			goto nla_put_failure;
+		return cfg80211_testmode_reply(skb);
+	case HWSIM_TM_CMD_STOP_QUEUES:
+		ieee80211_stop_queues(hw);
+		return 0;
+	case HWSIM_TM_CMD_WAKE_QUEUES:
+		ieee80211_wake_queues(hw);
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+ nla_put_failure:
+	kfree_skb(skb);
+	return -ENOBUFS;
+}
+#endif
+
+static int mac80211_hwsim_ampdu_action(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_ampdu_params *params)
+{
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+
+	switch (action) {
+	case IEEE80211_AMPDU_TX_START:
+		return IEEE80211_AMPDU_TX_START_IMMEDIATE;
+	case IEEE80211_AMPDU_TX_STOP_CONT:
+	case IEEE80211_AMPDU_TX_STOP_FLUSH:
+	case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
+		ieee80211_stop_tx_ba_cb_irqsafe(vif, sta->addr, tid);
+		break;
+	case IEEE80211_AMPDU_TX_OPERATIONAL:
+		break;
+	case IEEE80211_AMPDU_RX_START:
+	case IEEE80211_AMPDU_RX_STOP:
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static void mac80211_hwsim_flush(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 u32 queues, bool drop)
+{
+	/* Not implemented, queues only on kernel side */
+}
+
+static void hw_scan_work(struct work_struct *work)
+{
+	struct mac80211_hwsim_data *hwsim =
+		container_of(work, struct mac80211_hwsim_data, hw_scan.work);
+	struct cfg80211_scan_request *req = hwsim->hw_scan_request;
+	int dwell, i;
+
+	mutex_lock(&hwsim->mutex);
+	if (hwsim->scan_chan_idx >= req->n_channels) {
+		struct cfg80211_scan_info info = {
+			.aborted = false,
+		};
+
+		wiphy_dbg(hwsim->hw->wiphy, "hw scan complete\n");
+		ieee80211_scan_completed(hwsim->hw, &info);
+		hwsim->hw_scan_request = NULL;
+		hwsim->hw_scan_vif = NULL;
+		hwsim->tmp_chan = NULL;
+		mutex_unlock(&hwsim->mutex);
+		mac80211_hwsim_config_mac_nl(hwsim->hw, hwsim->scan_addr,
+					     false);
+		return;
+	}
+
+	wiphy_dbg(hwsim->hw->wiphy, "hw scan %d MHz\n",
+		  req->channels[hwsim->scan_chan_idx]->center_freq);
+
+	hwsim->tmp_chan = req->channels[hwsim->scan_chan_idx];
+	if (hwsim->tmp_chan->flags & (IEEE80211_CHAN_NO_IR |
+				      IEEE80211_CHAN_RADAR) ||
+	    !req->n_ssids) {
+		dwell = 120;
+	} else {
+		dwell = 30;
+		/* send probes */
+		for (i = 0; i < req->n_ssids; i++) {
+			struct sk_buff *probe;
+			struct ieee80211_mgmt *mgmt;
+
+			probe = ieee80211_probereq_get(hwsim->hw,
+						       hwsim->scan_addr,
+						       req->ssids[i].ssid,
+						       req->ssids[i].ssid_len,
+						       req->ie_len);
+			if (!probe)
+				continue;
+
+			mgmt = (struct ieee80211_mgmt *) probe->data;
+			memcpy(mgmt->da, req->bssid, ETH_ALEN);
+			memcpy(mgmt->bssid, req->bssid, ETH_ALEN);
+
+			if (req->ie_len)
+				skb_put_data(probe, req->ie, req->ie_len);
+
+			rcu_read_lock();
+			if (!ieee80211_tx_prepare_skb(hwsim->hw,
+						      hwsim->hw_scan_vif,
+						      probe,
+						      hwsim->tmp_chan->band,
+						      NULL)) {
+				rcu_read_unlock();
+				kfree_skb(probe);
+				continue;
+			}
+
+			local_bh_disable();
+			mac80211_hwsim_tx_frame(hwsim->hw, probe,
+						hwsim->tmp_chan);
+			rcu_read_unlock();
+			local_bh_enable();
+		}
+	}
+	ieee80211_queue_delayed_work(hwsim->hw, &hwsim->hw_scan,
+				     msecs_to_jiffies(dwell));
+	hwsim->survey_data[hwsim->scan_chan_idx].channel = hwsim->tmp_chan;
+	hwsim->survey_data[hwsim->scan_chan_idx].start = jiffies;
+	hwsim->survey_data[hwsim->scan_chan_idx].end =
+		jiffies + msecs_to_jiffies(dwell);
+	hwsim->scan_chan_idx++;
+	mutex_unlock(&hwsim->mutex);
+}
+
+static int mac80211_hwsim_hw_scan(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif,
+				  struct ieee80211_scan_request *hw_req)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+	struct cfg80211_scan_request *req = &hw_req->req;
+
+	mutex_lock(&hwsim->mutex);
+	if (WARN_ON(hwsim->tmp_chan || hwsim->hw_scan_request)) {
+		mutex_unlock(&hwsim->mutex);
+		return -EBUSY;
+	}
+	hwsim->hw_scan_request = req;
+	hwsim->hw_scan_vif = vif;
+	hwsim->scan_chan_idx = 0;
+	if (req->flags & NL80211_SCAN_FLAG_RANDOM_ADDR)
+		get_random_mask_addr(hwsim->scan_addr,
+				     hw_req->req.mac_addr,
+				     hw_req->req.mac_addr_mask);
+	else
+		memcpy(hwsim->scan_addr, vif->addr, ETH_ALEN);
+	memset(hwsim->survey_data, 0, sizeof(hwsim->survey_data));
+	mutex_unlock(&hwsim->mutex);
+
+	mac80211_hwsim_config_mac_nl(hw, hwsim->scan_addr, true);
+	wiphy_dbg(hw->wiphy, "hwsim hw_scan request\n");
+
+	ieee80211_queue_delayed_work(hwsim->hw, &hwsim->hw_scan, 0);
+
+	return 0;
+}
+
+static void mac80211_hwsim_cancel_hw_scan(struct ieee80211_hw *hw,
+					  struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+	struct cfg80211_scan_info info = {
+		.aborted = true,
+	};
+
+	wiphy_dbg(hw->wiphy, "hwsim cancel_hw_scan\n");
+
+	cancel_delayed_work_sync(&hwsim->hw_scan);
+
+	mutex_lock(&hwsim->mutex);
+	ieee80211_scan_completed(hwsim->hw, &info);
+	hwsim->tmp_chan = NULL;
+	hwsim->hw_scan_request = NULL;
+	hwsim->hw_scan_vif = NULL;
+	mutex_unlock(&hwsim->mutex);
+}
+
+static void mac80211_hwsim_sw_scan(struct ieee80211_hw *hw,
+				   struct ieee80211_vif *vif,
+				   const u8 *mac_addr)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+
+	mutex_lock(&hwsim->mutex);
+
+	if (hwsim->scanning) {
+		pr_debug("two hwsim sw_scans detected!\n");
+		goto out;
+	}
+
+	pr_debug("hwsim sw_scan request, prepping stuff\n");
+
+	memcpy(hwsim->scan_addr, mac_addr, ETH_ALEN);
+	mac80211_hwsim_config_mac_nl(hw, hwsim->scan_addr, true);
+	hwsim->scanning = true;
+	memset(hwsim->survey_data, 0, sizeof(hwsim->survey_data));
+
+out:
+	mutex_unlock(&hwsim->mutex);
+}
+
+static void mac80211_hwsim_sw_scan_complete(struct ieee80211_hw *hw,
+					    struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+
+	mutex_lock(&hwsim->mutex);
+
+	pr_debug("hwsim sw_scan_complete\n");
+	hwsim->scanning = false;
+	mac80211_hwsim_config_mac_nl(hw, hwsim->scan_addr, false);
+	eth_zero_addr(hwsim->scan_addr);
+
+	mutex_unlock(&hwsim->mutex);
+}
+
+static void hw_roc_start(struct work_struct *work)
+{
+	struct mac80211_hwsim_data *hwsim =
+		container_of(work, struct mac80211_hwsim_data, roc_start.work);
+
+	mutex_lock(&hwsim->mutex);
+
+	wiphy_dbg(hwsim->hw->wiphy, "hwsim ROC begins\n");
+	hwsim->tmp_chan = hwsim->roc_chan;
+	ieee80211_ready_on_channel(hwsim->hw);
+
+	ieee80211_queue_delayed_work(hwsim->hw, &hwsim->roc_done,
+				     msecs_to_jiffies(hwsim->roc_duration));
+
+	mutex_unlock(&hwsim->mutex);
+}
+
+static void hw_roc_done(struct work_struct *work)
+{
+	struct mac80211_hwsim_data *hwsim =
+		container_of(work, struct mac80211_hwsim_data, roc_done.work);
+
+	mutex_lock(&hwsim->mutex);
+	ieee80211_remain_on_channel_expired(hwsim->hw);
+	hwsim->tmp_chan = NULL;
+	mutex_unlock(&hwsim->mutex);
+
+	wiphy_dbg(hwsim->hw->wiphy, "hwsim ROC expired\n");
+}
+
+static int mac80211_hwsim_roc(struct ieee80211_hw *hw,
+			      struct ieee80211_vif *vif,
+			      struct ieee80211_channel *chan,
+			      int duration,
+			      enum ieee80211_roc_type type)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+
+	mutex_lock(&hwsim->mutex);
+	if (WARN_ON(hwsim->tmp_chan || hwsim->hw_scan_request)) {
+		mutex_unlock(&hwsim->mutex);
+		return -EBUSY;
+	}
+
+	hwsim->roc_chan = chan;
+	hwsim->roc_duration = duration;
+	mutex_unlock(&hwsim->mutex);
+
+	wiphy_dbg(hw->wiphy, "hwsim ROC (%d MHz, %d ms)\n",
+		  chan->center_freq, duration);
+	ieee80211_queue_delayed_work(hw, &hwsim->roc_start, HZ/50);
+
+	return 0;
+}
+
+static int mac80211_hwsim_croc(struct ieee80211_hw *hw,
+			       struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_data *hwsim = hw->priv;
+
+	cancel_delayed_work_sync(&hwsim->roc_start);
+	cancel_delayed_work_sync(&hwsim->roc_done);
+
+	mutex_lock(&hwsim->mutex);
+	hwsim->tmp_chan = NULL;
+	mutex_unlock(&hwsim->mutex);
+
+	wiphy_dbg(hw->wiphy, "hwsim ROC canceled\n");
+
+	return 0;
+}
+
+static int mac80211_hwsim_add_chanctx(struct ieee80211_hw *hw,
+				      struct ieee80211_chanctx_conf *ctx)
+{
+	hwsim_set_chanctx_magic(ctx);
+	wiphy_dbg(hw->wiphy,
+		  "add channel context control: %d MHz/width: %d/cfreqs:%d/%d MHz\n",
+		  ctx->def.chan->center_freq, ctx->def.width,
+		  ctx->def.center_freq1, ctx->def.center_freq2);
+	return 0;
+}
+
+static void mac80211_hwsim_remove_chanctx(struct ieee80211_hw *hw,
+					  struct ieee80211_chanctx_conf *ctx)
+{
+	wiphy_dbg(hw->wiphy,
+		  "remove channel context control: %d MHz/width: %d/cfreqs:%d/%d MHz\n",
+		  ctx->def.chan->center_freq, ctx->def.width,
+		  ctx->def.center_freq1, ctx->def.center_freq2);
+	hwsim_check_chanctx_magic(ctx);
+	hwsim_clear_chanctx_magic(ctx);
+}
+
+static void mac80211_hwsim_change_chanctx(struct ieee80211_hw *hw,
+					  struct ieee80211_chanctx_conf *ctx,
+					  u32 changed)
+{
+	hwsim_check_chanctx_magic(ctx);
+	wiphy_dbg(hw->wiphy,
+		  "change channel context control: %d MHz/width: %d/cfreqs:%d/%d MHz\n",
+		  ctx->def.chan->center_freq, ctx->def.width,
+		  ctx->def.center_freq1, ctx->def.center_freq2);
+}
+
+static int mac80211_hwsim_assign_vif_chanctx(struct ieee80211_hw *hw,
+					     struct ieee80211_vif *vif,
+					     struct ieee80211_bss_conf *link_conf,
+					     struct ieee80211_chanctx_conf *ctx)
+{
+	hwsim_check_magic(vif);
+	hwsim_check_chanctx_magic(ctx);
+
+	/* if we activate a link while already associated wake it up */
+	if (vif->type == NL80211_IFTYPE_STATION && vif->cfg.assoc) {
+		struct sk_buff *skb;
+
+		skb = ieee80211_nullfunc_get(hw, vif, link_conf->link_id, true);
+		if (skb) {
+			local_bh_disable();
+			mac80211_hwsim_tx_frame(hw, skb, ctx->def.chan);
+			local_bh_enable();
+		}
+	}
+
+	return 0;
+}
+
+static void mac80211_hwsim_unassign_vif_chanctx(struct ieee80211_hw *hw,
+						struct ieee80211_vif *vif,
+						struct ieee80211_bss_conf *link_conf,
+						struct ieee80211_chanctx_conf *ctx)
+{
+	hwsim_check_magic(vif);
+	hwsim_check_chanctx_magic(ctx);
+
+	/* if we deactivate a link while associated suspend it first */
+	if (vif->type == NL80211_IFTYPE_STATION && vif->cfg.assoc) {
+		struct sk_buff *skb;
+
+		skb = ieee80211_nullfunc_get(hw, vif, link_conf->link_id, true);
+		if (skb) {
+			struct ieee80211_hdr *hdr = (void *)skb->data;
+
+			hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
+
+			local_bh_disable();
+			mac80211_hwsim_tx_frame(hw, skb, ctx->def.chan);
+			local_bh_enable();
+		}
+	}
+}
+
+static const char mac80211_hwsim_gstrings_stats[][ETH_GSTRING_LEN] = {
+	"tx_pkts_nic",
+	"tx_bytes_nic",
+	"rx_pkts_nic",
+	"rx_bytes_nic",
+	"d_tx_dropped",
+	"d_tx_failed",
+	"d_ps_mode",
+	"d_group",
+};
+
+#define MAC80211_HWSIM_SSTATS_LEN ARRAY_SIZE(mac80211_hwsim_gstrings_stats)
+
+static void mac80211_hwsim_get_et_strings(struct ieee80211_hw *hw,
+					  struct ieee80211_vif *vif,
+					  u32 sset, u8 *data)
+{
+	if (sset == ETH_SS_STATS)
+		memcpy(data, *mac80211_hwsim_gstrings_stats,
+		       sizeof(mac80211_hwsim_gstrings_stats));
+}
+
+static int mac80211_hwsim_get_et_sset_count(struct ieee80211_hw *hw,
+					    struct ieee80211_vif *vif, int sset)
+{
+	if (sset == ETH_SS_STATS)
+		return MAC80211_HWSIM_SSTATS_LEN;
+	return 0;
+}
+
+static void mac80211_hwsim_get_et_stats(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					struct ethtool_stats *stats, u64 *data)
+{
+	struct mac80211_hwsim_data *ar = hw->priv;
+	int i = 0;
+
+	data[i++] = ar->tx_pkts;
+	data[i++] = ar->tx_bytes;
+	data[i++] = ar->rx_pkts;
+	data[i++] = ar->rx_bytes;
+	data[i++] = ar->tx_dropped;
+	data[i++] = ar->tx_failed;
+	data[i++] = ar->ps;
+	data[i++] = ar->group;
+
+	WARN_ON(i != MAC80211_HWSIM_SSTATS_LEN);
+}
+
+static int mac80211_hwsim_tx_last_beacon(struct ieee80211_hw *hw)
+{
+	return 1;
+}
+
+static int mac80211_hwsim_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
+{
+	return -EOPNOTSUPP;
+}
+
+static int mac80211_hwsim_change_vif_links(struct ieee80211_hw *hw,
+					   struct ieee80211_vif *vif,
+					   u16 old_links, u16 new_links,
+					   struct ieee80211_bss_conf *old[IEEE80211_MLD_MAX_NUM_LINKS])
+{
+	unsigned long rem = old_links & ~new_links;
+	unsigned long add = new_links & ~old_links;
+	int i;
+
+	if (!old_links)
+		rem |= BIT(0);
+	if (!new_links)
+		add |= BIT(0);
+
+	for_each_set_bit(i, &rem, IEEE80211_MLD_MAX_NUM_LINKS)
+		mac80211_hwsim_config_mac_nl(hw, old[i]->addr, false);
+
+	for_each_set_bit(i, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_bss_conf *link_conf;
+
+		link_conf = link_conf_dereference_protected(vif, i);
+		if (WARN_ON(!link_conf))
+			continue;
+
+		mac80211_hwsim_config_mac_nl(hw, link_conf->addr, true);
+	}
+
+	return 0;
+}
+
+static int mac80211_hwsim_change_sta_links(struct ieee80211_hw *hw,
+					   struct ieee80211_vif *vif,
+					   struct ieee80211_sta *sta,
+					   u16 old_links, u16 new_links)
+{
+	struct hwsim_sta_priv *sp = (void *)sta->drv_priv;
+
+	hwsim_check_sta_magic(sta);
+
+	if (vif->type == NL80211_IFTYPE_STATION)
+		sp->active_links_rx = new_links;
+
+	return 0;
+}
+
+#define HWSIM_COMMON_OPS					\
+	.tx = mac80211_hwsim_tx,				\
+	.wake_tx_queue = ieee80211_handle_wake_tx_queue,	\
+	.start = mac80211_hwsim_start,				\
+	.stop = mac80211_hwsim_stop,				\
+	.add_interface = mac80211_hwsim_add_interface,		\
+	.change_interface = mac80211_hwsim_change_interface,	\
+	.remove_interface = mac80211_hwsim_remove_interface,	\
+	.config = mac80211_hwsim_config,			\
+	.configure_filter = mac80211_hwsim_configure_filter,	\
+	.vif_cfg_changed = mac80211_hwsim_vif_info_changed,	\
+	.link_info_changed = mac80211_hwsim_link_info_changed,  \
+	.tx_last_beacon = mac80211_hwsim_tx_last_beacon,	\
+	.sta_notify = mac80211_hwsim_sta_notify,		\
+	.sta_rc_update = mac80211_hwsim_sta_rc_update,		\
+	.conf_tx = mac80211_hwsim_conf_tx,			\
+	.get_survey = mac80211_hwsim_get_survey,		\
+	CFG80211_TESTMODE_CMD(mac80211_hwsim_testmode_cmd)	\
+	.ampdu_action = mac80211_hwsim_ampdu_action,		\
+	.flush = mac80211_hwsim_flush,				\
+	.get_et_sset_count = mac80211_hwsim_get_et_sset_count,	\
+	.get_et_stats = mac80211_hwsim_get_et_stats,		\
+	.get_et_strings = mac80211_hwsim_get_et_strings,
+
+#define HWSIM_NON_MLO_OPS					\
+	.sta_add = mac80211_hwsim_sta_add,			\
+	.sta_remove = mac80211_hwsim_sta_remove,		\
+	.set_tim = mac80211_hwsim_set_tim,			\
+	.get_tsf = mac80211_hwsim_get_tsf,			\
+	.set_tsf = mac80211_hwsim_set_tsf,
+
+static const struct ieee80211_ops mac80211_hwsim_ops = {
+	HWSIM_COMMON_OPS
+	HWSIM_NON_MLO_OPS
+	.sw_scan_start = mac80211_hwsim_sw_scan,
+	.sw_scan_complete = mac80211_hwsim_sw_scan_complete,
+};
+
+#define HWSIM_CHANCTX_OPS					\
+	.hw_scan = mac80211_hwsim_hw_scan,			\
+	.cancel_hw_scan = mac80211_hwsim_cancel_hw_scan,	\
+	.remain_on_channel = mac80211_hwsim_roc,		\
+	.cancel_remain_on_channel = mac80211_hwsim_croc,	\
+	.add_chanctx = mac80211_hwsim_add_chanctx,		\
+	.remove_chanctx = mac80211_hwsim_remove_chanctx,	\
+	.change_chanctx = mac80211_hwsim_change_chanctx,	\
+	.assign_vif_chanctx = mac80211_hwsim_assign_vif_chanctx,\
+	.unassign_vif_chanctx = mac80211_hwsim_unassign_vif_chanctx,
+
+static const struct ieee80211_ops mac80211_hwsim_mchan_ops = {
+	HWSIM_COMMON_OPS
+	HWSIM_NON_MLO_OPS
+	HWSIM_CHANCTX_OPS
+};
+
+static const struct ieee80211_ops mac80211_hwsim_mlo_ops = {
+	HWSIM_COMMON_OPS
+	HWSIM_CHANCTX_OPS
+	.set_rts_threshold = mac80211_hwsim_set_rts_threshold,
+	.change_vif_links = mac80211_hwsim_change_vif_links,
+	.change_sta_links = mac80211_hwsim_change_sta_links,
+	.sta_state = mac80211_hwsim_sta_state,
+};
+
+struct hwsim_new_radio_params {
+	unsigned int channels;
+	const char *reg_alpha2;
+	const struct ieee80211_regdomain *regd;
+	bool reg_strict;
+	bool p2p_device;
+	bool use_chanctx;
+	bool destroy_on_close;
+	const char *hwname;
+	bool no_vif;
+	const u8 *perm_addr;
+	u32 iftypes;
+	u32 *ciphers;
+	u8 n_ciphers;
+	bool mlo;
+};
+
+static void hwsim_mcast_config_msg(struct sk_buff *mcast_skb,
+				   struct genl_info *info)
+{
+	if (info)
+		genl_notify(&hwsim_genl_family, mcast_skb, info,
+			    HWSIM_MCGRP_CONFIG, GFP_KERNEL);
+	else
+		genlmsg_multicast(&hwsim_genl_family, mcast_skb, 0,
+				  HWSIM_MCGRP_CONFIG, GFP_KERNEL);
+}
+
+static int append_radio_msg(struct sk_buff *skb, int id,
+			    struct hwsim_new_radio_params *param)
+{
+	int ret;
+
+	ret = nla_put_u32(skb, HWSIM_ATTR_RADIO_ID, id);
+	if (ret < 0)
+		return ret;
+
+	if (param->channels) {
+		ret = nla_put_u32(skb, HWSIM_ATTR_CHANNELS, param->channels);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (param->reg_alpha2) {
+		ret = nla_put(skb, HWSIM_ATTR_REG_HINT_ALPHA2, 2,
+			      param->reg_alpha2);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (param->regd) {
+		int i;
+
+		for (i = 0; i < ARRAY_SIZE(hwsim_world_regdom_custom); i++) {
+			if (hwsim_world_regdom_custom[i] != param->regd)
+				continue;
+
+			ret = nla_put_u32(skb, HWSIM_ATTR_REG_CUSTOM_REG, i);
+			if (ret < 0)
+				return ret;
+			break;
+		}
+	}
+
+	if (param->reg_strict) {
+		ret = nla_put_flag(skb, HWSIM_ATTR_REG_STRICT_REG);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (param->p2p_device) {
+		ret = nla_put_flag(skb, HWSIM_ATTR_SUPPORT_P2P_DEVICE);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (param->use_chanctx) {
+		ret = nla_put_flag(skb, HWSIM_ATTR_USE_CHANCTX);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (param->hwname) {
+		ret = nla_put(skb, HWSIM_ATTR_RADIO_NAME,
+			      strlen(param->hwname), param->hwname);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+static void hwsim_mcast_new_radio(int id, struct genl_info *info,
+				  struct hwsim_new_radio_params *param)
+{
+	struct sk_buff *mcast_skb;
+	void *data;
+
+	mcast_skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!mcast_skb)
+		return;
+
+	data = genlmsg_put(mcast_skb, 0, 0, &hwsim_genl_family, 0,
+			   HWSIM_CMD_NEW_RADIO);
+	if (!data)
+		goto out_err;
+
+	if (append_radio_msg(mcast_skb, id, param) < 0)
+		goto out_err;
+
+	genlmsg_end(mcast_skb, data);
+
+	hwsim_mcast_config_msg(mcast_skb, info);
+	return;
+
+out_err:
+	nlmsg_free(mcast_skb);
+}
+
+static const struct ieee80211_sband_iftype_data sband_capa_2ghz[] = {
+	{
+		.types_mask = BIT(NL80211_IFTYPE_STATION),
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_BSR |
+					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] =
+					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
+					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xffff),
+				.tx_mcs_160 = cpu_to_le16(0xffff),
+				.rx_mcs_80p80 = cpu_to_le16(0xffff),
+				.tx_mcs_80p80 = cpu_to_le16(0xffff),
+			},
+		},
+		.eht_cap = {
+			.has_eht = true,
+			.eht_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
+					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
+					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
+				.phy_cap_info[0] =
+					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
+					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
+					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE,
+				.phy_cap_info[3] =
+					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
+				.phy_cap_info[4] =
+					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
+					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
+					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
+					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
+				.phy_cap_info[5] =
+					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
+					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
+					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
+				.phy_cap_info[6] =
+					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
+					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
+				.phy_cap_info[7] =
+					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW,
+			},
+
+			/* For all MCS and bandwidth, set 8 NSS for both Tx and
+			 * Rx
+			 */
+			.eht_mcs_nss_supp = {
+				/*
+				 * Since B0, B1, B2 and B3 are not set in
+				 * the supported channel width set field in the
+				 * HE PHY capabilities information field the
+				 * device is a 20MHz only device on 2.4GHz band.
+				 */
+				.only_20mhz = {
+					.rx_tx_mcs7_max_nss = 0x88,
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+			},
+			/* PPE threshold information is not supported */
+		},
+	},
+	{
+		.types_mask = BIT(NL80211_IFTYPE_AP),
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_BSR |
+					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] =
+					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
+					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xffff),
+				.tx_mcs_160 = cpu_to_le16(0xffff),
+				.rx_mcs_80p80 = cpu_to_le16(0xffff),
+				.tx_mcs_80p80 = cpu_to_le16(0xffff),
+			},
+		},
+		.eht_cap = {
+			.has_eht = true,
+			.eht_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
+					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
+					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
+				.phy_cap_info[0] =
+					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
+					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
+					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE,
+				.phy_cap_info[3] =
+					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
+				.phy_cap_info[4] =
+					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
+					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
+					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
+					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
+				.phy_cap_info[5] =
+					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
+					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
+					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
+				.phy_cap_info[6] =
+					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
+					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
+				.phy_cap_info[7] =
+					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW,
+			},
+
+			/* For all MCS and bandwidth, set 8 NSS for both Tx and
+			 * Rx
+			 */
+			.eht_mcs_nss_supp = {
+				/*
+				 * Since B0, B1, B2 and B3 are not set in
+				 * the supported channel width set field in the
+				 * HE PHY capabilities information field the
+				 * device is a 20MHz only device on 2.4GHz band.
+				 */
+				.only_20mhz = {
+					.rx_tx_mcs7_max_nss = 0x88,
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+			},
+			/* PPE threshold information is not supported */
+		},
+	},
+#ifdef CONFIG_MAC80211_MESH
+	{
+		.types_mask = BIT(NL80211_IFTYPE_MESH_POINT),
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] = 0,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xffff),
+				.tx_mcs_160 = cpu_to_le16(0xffff),
+				.rx_mcs_80p80 = cpu_to_le16(0xffff),
+				.tx_mcs_80p80 = cpu_to_le16(0xffff),
+			},
+		},
+	},
+#endif
+};
+
+static const struct ieee80211_sband_iftype_data sband_capa_5ghz[] = {
+	{
+		/* TODO: should we support other types, e.g., P2P? */
+		.types_mask = BIT(NL80211_IFTYPE_STATION),
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_BSR |
+					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[0] =
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] =
+					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
+					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xfffa),
+				.tx_mcs_160 = cpu_to_le16(0xfffa),
+				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
+			},
+		},
+		.eht_cap = {
+			.has_eht = true,
+			.eht_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
+					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
+					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
+				.phy_cap_info[0] =
+					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
+					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
+					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
+					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
+				.phy_cap_info[1] =
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK,
+				.phy_cap_info[2] =
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK,
+				.phy_cap_info[3] =
+					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
+				.phy_cap_info[4] =
+					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
+					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
+					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
+					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
+				.phy_cap_info[5] =
+					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
+					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
+					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
+				.phy_cap_info[6] =
+					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
+					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
+				.phy_cap_info[7] =
+					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ,
+			},
+
+			/* For all MCS and bandwidth, set 8 NSS for both Tx and
+			 * Rx
+			 */
+			.eht_mcs_nss_supp = {
+				/*
+				 * As B1 and B2 are set in the supported
+				 * channel width set field in the HE PHY
+				 * capabilities information field include all
+				 * the following MCS/NSS.
+				 */
+				.bw._80 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+				.bw._160 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+			},
+			/* PPE threshold information is not supported */
+		},
+	},
+	{
+		.types_mask = BIT(NL80211_IFTYPE_AP),
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_BSR |
+					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[0] =
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] =
+					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
+					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xfffa),
+				.tx_mcs_160 = cpu_to_le16(0xfffa),
+				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
+			},
+		},
+		.eht_cap = {
+			.has_eht = true,
+			.eht_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
+					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
+					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
+				.phy_cap_info[0] =
+					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
+					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
+					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
+					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
+				.phy_cap_info[1] =
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK,
+				.phy_cap_info[2] =
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK,
+				.phy_cap_info[3] =
+					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
+				.phy_cap_info[4] =
+					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
+					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
+					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
+					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
+				.phy_cap_info[5] =
+					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
+					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
+					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
+				.phy_cap_info[6] =
+					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
+					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK,
+				.phy_cap_info[7] =
+					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ,
+			},
+
+			/* For all MCS and bandwidth, set 8 NSS for both Tx and
+			 * Rx
+			 */
+			.eht_mcs_nss_supp = {
+				/*
+				 * As B1 and B2 are set in the supported
+				 * channel width set field in the HE PHY
+				 * capabilities information field include all
+				 * the following MCS/NSS.
+				 */
+				.bw._80 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+				.bw._160 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+			},
+			/* PPE threshold information is not supported */
+		},
+	},
+#ifdef CONFIG_MAC80211_MESH
+	{
+		/* TODO: should we support other types, e.g., IBSS?*/
+		.types_mask = BIT(NL80211_IFTYPE_MESH_POINT),
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[0] =
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] = 0,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xfffa),
+				.tx_mcs_160 = cpu_to_le16(0xfffa),
+				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
+			},
+		},
+	},
+#endif
+};
+
+static const struct ieee80211_sband_iftype_data sband_capa_6ghz[] = {
+	{
+		/* TODO: should we support other types, e.g., P2P? */
+		.types_mask = BIT(NL80211_IFTYPE_STATION),
+		.he_6ghz_capa = {
+			.capa = cpu_to_le16(IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START |
+					    IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP |
+					    IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN |
+					    IEEE80211_HE_6GHZ_CAP_SM_PS |
+					    IEEE80211_HE_6GHZ_CAP_RD_RESPONDER |
+					    IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
+					    IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS),
+		},
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_BSR |
+					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[0] =
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] =
+					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
+					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xfffa),
+				.tx_mcs_160 = cpu_to_le16(0xfffa),
+				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
+			},
+		},
+		.eht_cap = {
+			.has_eht = true,
+			.eht_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
+					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
+					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
+				.phy_cap_info[0] =
+					IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ |
+					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
+					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
+					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
+					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
+				.phy_cap_info[1] =
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK,
+				.phy_cap_info[2] =
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK,
+				.phy_cap_info[3] =
+					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
+				.phy_cap_info[4] =
+					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
+					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
+					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
+					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
+				.phy_cap_info[5] =
+					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
+					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
+					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
+				.phy_cap_info[6] =
+					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
+					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK |
+					IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP,
+				.phy_cap_info[7] =
+					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ,
+			},
+
+			/* For all MCS and bandwidth, set 8 NSS for both Tx and
+			 * Rx
+			 */
+			.eht_mcs_nss_supp = {
+				/*
+				 * As B1 and B2 are set in the supported
+				 * channel width set field in the HE PHY
+				 * capabilities information field and 320MHz in
+				 * 6GHz is supported include all the following
+				 * MCS/NSS.
+				 */
+				.bw._80 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+				.bw._160 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+				.bw._320 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+			},
+			/* PPE threshold information is not supported */
+		},
+	},
+	{
+		.types_mask = BIT(NL80211_IFTYPE_AP),
+		.he_6ghz_capa = {
+			.capa = cpu_to_le16(IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START |
+					    IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP |
+					    IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN |
+					    IEEE80211_HE_6GHZ_CAP_SM_PS |
+					    IEEE80211_HE_6GHZ_CAP_RD_RESPONDER |
+					    IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
+					    IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS),
+		},
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US |
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_BSR |
+					IEEE80211_HE_MAC_CAP2_MU_CASCADING |
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[0] =
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] =
+					IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US |
+					IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ |
+					IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO |
+					IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xfffa),
+				.tx_mcs_160 = cpu_to_le16(0xfffa),
+				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
+			},
+		},
+		.eht_cap = {
+			.has_eht = true,
+			.eht_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS |
+					IEEE80211_EHT_MAC_CAP0_OM_CONTROL |
+					IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1,
+				.phy_cap_info[0] =
+					IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ |
+					IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ |
+					IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI |
+					IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER |
+					IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE |
+					IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK,
+				.phy_cap_info[1] =
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK,
+				.phy_cap_info[2] =
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK |
+					IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK,
+				.phy_cap_info[3] =
+					IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK |
+					IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK,
+				.phy_cap_info[4] =
+					IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO |
+					IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP |
+					IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP |
+					IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI |
+					IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK,
+				.phy_cap_info[5] =
+					IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK |
+					IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP |
+					IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT |
+					IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK |
+					IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK,
+				.phy_cap_info[6] =
+					IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK |
+					IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK |
+					IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP,
+				.phy_cap_info[7] =
+					IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
+					IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ |
+					IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ,
+			},
+
+			/* For all MCS and bandwidth, set 8 NSS for both Tx and
+			 * Rx
+			 */
+			.eht_mcs_nss_supp = {
+				/*
+				 * As B1 and B2 are set in the supported
+				 * channel width set field in the HE PHY
+				 * capabilities information field and 320MHz in
+				 * 6GHz is supported include all the following
+				 * MCS/NSS.
+				 */
+				.bw._80 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+				.bw._160 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+				.bw._320 = {
+					.rx_tx_mcs9_max_nss = 0x88,
+					.rx_tx_mcs11_max_nss = 0x88,
+					.rx_tx_mcs13_max_nss = 0x88,
+				},
+			},
+			/* PPE threshold information is not supported */
+		},
+	},
+#ifdef CONFIG_MAC80211_MESH
+	{
+		/* TODO: should we support other types, e.g., IBSS?*/
+		.types_mask = BIT(NL80211_IFTYPE_MESH_POINT),
+		.he_6ghz_capa = {
+			.capa = cpu_to_le16(IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START |
+					    IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP |
+					    IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN |
+					    IEEE80211_HE_6GHZ_CAP_SM_PS |
+					    IEEE80211_HE_6GHZ_CAP_RD_RESPONDER |
+					    IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
+					    IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS),
+		},
+		.he_cap = {
+			.has_he = true,
+			.he_cap_elem = {
+				.mac_cap_info[0] =
+					IEEE80211_HE_MAC_CAP0_HTC_HE,
+				.mac_cap_info[1] =
+					IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8,
+				.mac_cap_info[2] =
+					IEEE80211_HE_MAC_CAP2_ACK_EN,
+				.mac_cap_info[3] =
+					IEEE80211_HE_MAC_CAP3_OMI_CONTROL |
+					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3,
+				.mac_cap_info[4] = IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU,
+				.phy_cap_info[0] =
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G |
+					IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G,
+				.phy_cap_info[1] =
+					IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK |
+					IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A |
+					IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD |
+					IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS,
+				.phy_cap_info[2] = 0,
+
+				/* Leave all the other PHY capability bytes
+				 * unset, as DCM, beam forming, RU and PPE
+				 * threshold information are not supported
+				 */
+			},
+			.he_mcs_nss_supp = {
+				.rx_mcs_80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80 = cpu_to_le16(0xfffa),
+				.rx_mcs_160 = cpu_to_le16(0xfffa),
+				.tx_mcs_160 = cpu_to_le16(0xfffa),
+				.rx_mcs_80p80 = cpu_to_le16(0xfffa),
+				.tx_mcs_80p80 = cpu_to_le16(0xfffa),
+			},
+		},
+	},
+#endif
+};
+
+static void mac80211_hwsim_sband_capab(struct ieee80211_supported_band *sband)
+{
+	u16 n_iftype_data;
+
+	if (sband->band == NL80211_BAND_2GHZ) {
+		n_iftype_data = ARRAY_SIZE(sband_capa_2ghz);
+		sband->iftype_data =
+			(struct ieee80211_sband_iftype_data *)sband_capa_2ghz;
+	} else if (sband->band == NL80211_BAND_5GHZ) {
+		n_iftype_data = ARRAY_SIZE(sband_capa_5ghz);
+		sband->iftype_data =
+			(struct ieee80211_sband_iftype_data *)sband_capa_5ghz;
+	} else if (sband->band == NL80211_BAND_6GHZ) {
+		n_iftype_data = ARRAY_SIZE(sband_capa_6ghz);
+		sband->iftype_data =
+			(struct ieee80211_sband_iftype_data *)sband_capa_6ghz;
+	} else {
+		return;
+	}
+
+	sband->n_iftype_data = n_iftype_data;
+}
+
+#ifdef CONFIG_MAC80211_MESH
+#define HWSIM_MESH_BIT BIT(NL80211_IFTYPE_MESH_POINT)
+#else
+#define HWSIM_MESH_BIT 0
+#endif
+
+#define HWSIM_DEFAULT_IF_LIMIT \
+	(BIT(NL80211_IFTYPE_STATION) | \
+	 BIT(NL80211_IFTYPE_P2P_CLIENT) | \
+	 BIT(NL80211_IFTYPE_AP) | \
+	 BIT(NL80211_IFTYPE_P2P_GO) | \
+	 HWSIM_MESH_BIT)
+
+#define HWSIM_IFTYPE_SUPPORT_MASK \
+	(BIT(NL80211_IFTYPE_STATION) | \
+	 BIT(NL80211_IFTYPE_AP) | \
+	 BIT(NL80211_IFTYPE_P2P_CLIENT) | \
+	 BIT(NL80211_IFTYPE_P2P_GO) | \
+	 BIT(NL80211_IFTYPE_ADHOC) | \
+	 BIT(NL80211_IFTYPE_MESH_POINT) | \
+	 BIT(NL80211_IFTYPE_OCB))
+
+static int mac80211_hwsim_new_radio(struct genl_info *info,
+				    struct hwsim_new_radio_params *param)
+{
+	int err;
+	u8 addr[ETH_ALEN];
+	struct mac80211_hwsim_data *data;
+	struct ieee80211_hw *hw;
+	enum nl80211_band band;
+	const struct ieee80211_ops *ops = &mac80211_hwsim_ops;
+	struct net *net;
+	int idx, i;
+	int n_limits = 0;
+
+	if (WARN_ON(param->channels > 1 && !param->use_chanctx))
+		return -EINVAL;
+
+	spin_lock_bh(&hwsim_radio_lock);
+	idx = hwsim_radio_idx++;
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	if (param->mlo)
+		ops = &mac80211_hwsim_mlo_ops;
+	else if (param->use_chanctx)
+		ops = &mac80211_hwsim_mchan_ops;
+	hw = ieee80211_alloc_hw_nm(sizeof(*data), ops, param->hwname);
+	if (!hw) {
+		pr_debug("mac80211_hwsim: ieee80211_alloc_hw failed\n");
+		err = -ENOMEM;
+		goto failed;
+	}
+
+	/* ieee80211_alloc_hw_nm may have used a default name */
+	param->hwname = wiphy_name(hw->wiphy);
+
+	if (info)
+		net = genl_info_net(info);
+	else
+		net = &init_net;
+	wiphy_net_set(hw->wiphy, net);
+
+	data = hw->priv;
+	data->hw = hw;
+
+	data->dev = device_create(hwsim_class, NULL, 0, hw, "hwsim%d", idx);
+	if (IS_ERR(data->dev)) {
+		printk(KERN_DEBUG
+		       "mac80211_hwsim: device_create failed (%ld)\n",
+		       PTR_ERR(data->dev));
+		err = -ENOMEM;
+		goto failed_drvdata;
+	}
+	data->dev->driver = &mac80211_hwsim_driver.driver;
+	err = device_bind_driver(data->dev);
+	if (err != 0) {
+		pr_debug("mac80211_hwsim: device_bind_driver failed (%d)\n",
+		       err);
+		goto failed_bind;
+	}
+
+	skb_queue_head_init(&data->pending);
+
+	SET_IEEE80211_DEV(hw, data->dev);
+	if (!param->perm_addr) {
+		eth_zero_addr(addr);
+		addr[0] = 0x02;
+		addr[3] = idx >> 8;
+		addr[4] = idx;
+		memcpy(data->addresses[0].addr, addr, ETH_ALEN);
+		/* Why need here second address ? */
+		memcpy(data->addresses[1].addr, addr, ETH_ALEN);
+		data->addresses[1].addr[0] |= 0x40;
+		hw->wiphy->n_addresses = 2;
+		hw->wiphy->addresses = data->addresses;
+		/* possible address clash is checked at hash table insertion */
+	} else {
+		memcpy(data->addresses[0].addr, param->perm_addr, ETH_ALEN);
+		/* compatibility with automatically generated mac addr */
+		memcpy(data->addresses[1].addr, param->perm_addr, ETH_ALEN);
+		hw->wiphy->n_addresses = 2;
+		hw->wiphy->addresses = data->addresses;
+	}
+
+	data->channels = param->channels;
+	data->use_chanctx = param->use_chanctx;
+	data->idx = idx;
+	data->destroy_on_close = param->destroy_on_close;
+	if (info)
+		data->portid = info->snd_portid;
+
+	/* setup interface limits, only on interface types we support */
+	if (param->iftypes & BIT(NL80211_IFTYPE_ADHOC)) {
+		data->if_limits[n_limits].max = 1;
+		data->if_limits[n_limits].types = BIT(NL80211_IFTYPE_ADHOC);
+		n_limits++;
+	}
+
+	if (param->iftypes & HWSIM_DEFAULT_IF_LIMIT) {
+		data->if_limits[n_limits].max = 2048;
+		/*
+		 * For this case, we may only support a subset of
+		 * HWSIM_DEFAULT_IF_LIMIT, therefore we only want to add the
+		 * bits that both param->iftype & HWSIM_DEFAULT_IF_LIMIT have.
+		 */
+		data->if_limits[n_limits].types =
+					HWSIM_DEFAULT_IF_LIMIT & param->iftypes;
+		n_limits++;
+	}
+
+	if (param->iftypes & BIT(NL80211_IFTYPE_P2P_DEVICE)) {
+		data->if_limits[n_limits].max = 1;
+		data->if_limits[n_limits].types =
+						BIT(NL80211_IFTYPE_P2P_DEVICE);
+		n_limits++;
+	}
+
+	if (data->use_chanctx) {
+		hw->wiphy->max_scan_ssids = 255;
+		hw->wiphy->max_scan_ie_len = IEEE80211_MAX_DATA_LEN;
+		hw->wiphy->max_remain_on_channel_duration = 1000;
+		data->if_combination.radar_detect_widths = 0;
+		data->if_combination.num_different_channels = data->channels;
+	} else {
+		data->if_combination.num_different_channels = 1;
+		data->if_combination.radar_detect_widths =
+					BIT(NL80211_CHAN_WIDTH_5) |
+					BIT(NL80211_CHAN_WIDTH_10) |
+					BIT(NL80211_CHAN_WIDTH_20_NOHT) |
+					BIT(NL80211_CHAN_WIDTH_20) |
+					BIT(NL80211_CHAN_WIDTH_40) |
+					BIT(NL80211_CHAN_WIDTH_80) |
+					BIT(NL80211_CHAN_WIDTH_160);
+	}
+
+	if (!n_limits) {
+		err = -EINVAL;
+		goto failed_hw;
+	}
+
+	data->if_combination.max_interfaces = 0;
+	for (i = 0; i < n_limits; i++)
+		data->if_combination.max_interfaces +=
+			data->if_limits[i].max;
+
+	data->if_combination.n_limits = n_limits;
+	data->if_combination.limits = data->if_limits;
+
+	/*
+	 * If we actually were asked to support combinations,
+	 * advertise them - if there's only a single thing like
+	 * only IBSS then don't advertise it as combinations.
+	 */
+	if (data->if_combination.max_interfaces > 1) {
+		hw->wiphy->iface_combinations = &data->if_combination;
+		hw->wiphy->n_iface_combinations = 1;
+	}
+
+	if (param->ciphers) {
+		memcpy(data->ciphers, param->ciphers,
+		       param->n_ciphers * sizeof(u32));
+		hw->wiphy->cipher_suites = data->ciphers;
+		hw->wiphy->n_cipher_suites = param->n_ciphers;
+	}
+
+	data->rx_rssi = DEFAULT_RX_RSSI;
+
+	INIT_DELAYED_WORK(&data->roc_start, hw_roc_start);
+	INIT_DELAYED_WORK(&data->roc_done, hw_roc_done);
+	INIT_DELAYED_WORK(&data->hw_scan, hw_scan_work);
+
+	hw->queues = 5;
+	hw->offchannel_tx_hw_queue = 4;
+
+	ieee80211_hw_set(hw, SUPPORT_FAST_XMIT);
+	ieee80211_hw_set(hw, CHANCTX_STA_CSA);
+	ieee80211_hw_set(hw, SUPPORTS_HT_CCK_RATES);
+	ieee80211_hw_set(hw, QUEUE_CONTROL);
+	ieee80211_hw_set(hw, WANT_MONITOR_VIF);
+	ieee80211_hw_set(hw, AMPDU_AGGREGATION);
+	ieee80211_hw_set(hw, MFP_CAPABLE);
+	ieee80211_hw_set(hw, SIGNAL_DBM);
+	ieee80211_hw_set(hw, SUPPORTS_PS);
+	ieee80211_hw_set(hw, REPORTS_TX_ACK_STATUS);
+	ieee80211_hw_set(hw, TDLS_WIDER_BW);
+	ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID);
+
+	if (param->mlo) {
+		hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_MLO;
+		ieee80211_hw_set(hw, HAS_RATE_CONTROL);
+		ieee80211_hw_set(hw, SUPPORTS_DYNAMIC_PS);
+		ieee80211_hw_set(hw, CONNECTION_MONITOR);
+		ieee80211_hw_set(hw, AP_LINK_PS);
+	} else {
+		ieee80211_hw_set(hw, HOST_BROADCAST_PS_BUFFERING);
+		ieee80211_hw_set(hw, PS_NULLFUNC_STACK);
+		if (rctbl)
+			ieee80211_hw_set(hw, SUPPORTS_RC_TABLE);
+	}
+
+	hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
+	hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_TDLS |
+			    WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL |
+			    WIPHY_FLAG_AP_UAPSD |
+			    WIPHY_FLAG_SUPPORTS_5_10_MHZ |
+			    WIPHY_FLAG_HAS_CHANNEL_SWITCH;
+	hw->wiphy->features |= NL80211_FEATURE_ACTIVE_MONITOR |
+			       NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE |
+			       NL80211_FEATURE_STATIC_SMPS |
+			       NL80211_FEATURE_DYNAMIC_SMPS |
+			       NL80211_FEATURE_SCAN_RANDOM_MAC_ADDR;
+	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_VHT_IBSS);
+	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_BEACON_PROTECTION);
+	wiphy_ext_feature_set(hw->wiphy,
+			      NL80211_EXT_FEATURE_MULTICAST_REGISTRATIONS);
+	wiphy_ext_feature_set(hw->wiphy,
+			      NL80211_EXT_FEATURE_BEACON_RATE_LEGACY);
+
+	wiphy_ext_feature_set(hw->wiphy,
+			      NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT);
+
+	hw->wiphy->interface_modes = param->iftypes;
+
+	/* ask mac80211 to reserve space for magic */
+	hw->vif_data_size = sizeof(struct hwsim_vif_priv);
+	hw->sta_data_size = sizeof(struct hwsim_sta_priv);
+	hw->chanctx_data_size = sizeof(struct hwsim_chanctx_priv);
+
+	memcpy(data->channels_2ghz, hwsim_channels_2ghz,
+		sizeof(hwsim_channels_2ghz));
+	memcpy(data->channels_5ghz, hwsim_channels_5ghz,
+		sizeof(hwsim_channels_5ghz));
+	memcpy(data->channels_6ghz, hwsim_channels_6ghz,
+		sizeof(hwsim_channels_6ghz));
+	memcpy(data->channels_s1g, hwsim_channels_s1g,
+	       sizeof(hwsim_channels_s1g));
+	memcpy(data->rates, hwsim_rates, sizeof(hwsim_rates));
+
+	for (band = NL80211_BAND_2GHZ; band < NUM_NL80211_BANDS; band++) {
+		struct ieee80211_supported_band *sband = &data->bands[band];
+
+		sband->band = band;
+
+		switch (band) {
+		case NL80211_BAND_2GHZ:
+			sband->channels = data->channels_2ghz;
+			sband->n_channels = ARRAY_SIZE(hwsim_channels_2ghz);
+			sband->bitrates = data->rates;
+			sband->n_bitrates = ARRAY_SIZE(hwsim_rates);
+			break;
+		case NL80211_BAND_5GHZ:
+			sband->channels = data->channels_5ghz;
+			sband->n_channels = ARRAY_SIZE(hwsim_channels_5ghz);
+			sband->bitrates = data->rates + 4;
+			sband->n_bitrates = ARRAY_SIZE(hwsim_rates) - 4;
+
+			sband->vht_cap.vht_supported = true;
+			sband->vht_cap.cap =
+				IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454 |
+				IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ |
+				IEEE80211_VHT_CAP_RXLDPC |
+				IEEE80211_VHT_CAP_SHORT_GI_80 |
+				IEEE80211_VHT_CAP_SHORT_GI_160 |
+				IEEE80211_VHT_CAP_TXSTBC |
+				IEEE80211_VHT_CAP_RXSTBC_4 |
+				IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK;
+			sband->vht_cap.vht_mcs.rx_mcs_map =
+				cpu_to_le16(IEEE80211_VHT_MCS_SUPPORT_0_9 << 0 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 2 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 4 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 6 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 8 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 10 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 12 |
+					    IEEE80211_VHT_MCS_SUPPORT_0_9 << 14);
+			sband->vht_cap.vht_mcs.tx_mcs_map =
+				sband->vht_cap.vht_mcs.rx_mcs_map;
+			break;
+		case NL80211_BAND_6GHZ:
+			sband->channels = data->channels_6ghz;
+			sband->n_channels = ARRAY_SIZE(hwsim_channels_6ghz);
+			sband->bitrates = data->rates + 4;
+			sband->n_bitrates = ARRAY_SIZE(hwsim_rates) - 4;
+			break;
+		case NL80211_BAND_S1GHZ:
+			memcpy(&sband->s1g_cap, &hwsim_s1g_cap,
+			       sizeof(sband->s1g_cap));
+			sband->channels = data->channels_s1g;
+			sband->n_channels = ARRAY_SIZE(hwsim_channels_s1g);
+			break;
+		default:
+			continue;
+		}
+
+		if (band != NL80211_BAND_6GHZ){
+			sband->ht_cap.ht_supported = true;
+			sband->ht_cap.cap = IEEE80211_HT_CAP_SUP_WIDTH_20_40 |
+					    IEEE80211_HT_CAP_GRN_FLD |
+					    IEEE80211_HT_CAP_SGI_20 |
+					    IEEE80211_HT_CAP_SGI_40 |
+					    IEEE80211_HT_CAP_DSSSCCK40;
+			sband->ht_cap.ampdu_factor = 0x3;
+			sband->ht_cap.ampdu_density = 0x6;
+			memset(&sband->ht_cap.mcs, 0,
+			       sizeof(sband->ht_cap.mcs));
+			sband->ht_cap.mcs.rx_mask[0] = 0xff;
+			sband->ht_cap.mcs.rx_mask[1] = 0xff;
+			sband->ht_cap.mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED;
+		}
+
+		mac80211_hwsim_sband_capab(sband);
+
+		hw->wiphy->bands[band] = sband;
+	}
+
+	/* By default all radios belong to the first group */
+	data->group = 1;
+	mutex_init(&data->mutex);
+
+	data->netgroup = hwsim_net_get_netgroup(net);
+	data->wmediumd = hwsim_net_get_wmediumd(net);
+
+	/* Enable frame retransmissions for lossy channels */
+	hw->max_rates = 4;
+	hw->max_rate_tries = 11;
+
+	hw->wiphy->vendor_commands = mac80211_hwsim_vendor_commands;
+	hw->wiphy->n_vendor_commands =
+		ARRAY_SIZE(mac80211_hwsim_vendor_commands);
+	hw->wiphy->vendor_events = mac80211_hwsim_vendor_events;
+	hw->wiphy->n_vendor_events = ARRAY_SIZE(mac80211_hwsim_vendor_events);
+
+	if (param->reg_strict)
+		hw->wiphy->regulatory_flags |= REGULATORY_STRICT_REG;
+	if (param->regd) {
+		data->regd = param->regd;
+		hw->wiphy->regulatory_flags |= REGULATORY_CUSTOM_REG;
+		wiphy_apply_custom_regulatory(hw->wiphy, param->regd);
+		/* give the regulatory workqueue a chance to run */
+		schedule_timeout_interruptible(1);
+	}
+
+	if (param->no_vif)
+		ieee80211_hw_set(hw, NO_AUTO_VIF);
+
+	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST);
+
+	for (i = 0; i < ARRAY_SIZE(data->link_data); i++) {
+		hrtimer_init(&data->link_data[i].beacon_timer, CLOCK_MONOTONIC,
+			     HRTIMER_MODE_ABS_SOFT);
+		data->link_data[i].beacon_timer.function =
+			mac80211_hwsim_beacon;
+		data->link_data[i].link_id = i;
+	}
+
+	err = ieee80211_register_hw(hw);
+	if (err < 0) {
+		pr_debug("mac80211_hwsim: ieee80211_register_hw failed (%d)\n",
+		       err);
+		goto failed_hw;
+	}
+
+	wiphy_dbg(hw->wiphy, "hwaddr %pM registered\n", hw->wiphy->perm_addr);
+
+	if (param->reg_alpha2) {
+		data->alpha2[0] = param->reg_alpha2[0];
+		data->alpha2[1] = param->reg_alpha2[1];
+		regulatory_hint(hw->wiphy, param->reg_alpha2);
+	}
+
+	data->debugfs = debugfs_create_dir("hwsim", hw->wiphy->debugfsdir);
+	debugfs_create_file("ps", 0666, data->debugfs, data, &hwsim_fops_ps);
+	debugfs_create_file("group", 0666, data->debugfs, data,
+			    &hwsim_fops_group);
+	debugfs_create_file("rx_rssi", 0666, data->debugfs, data,
+			    &hwsim_fops_rx_rssi);
+	if (!data->use_chanctx)
+		debugfs_create_file("dfs_simulate_radar", 0222,
+				    data->debugfs,
+				    data, &hwsim_simulate_radar);
+
+	spin_lock_bh(&hwsim_radio_lock);
+	err = rhashtable_insert_fast(&hwsim_radios_rht, &data->rht,
+				     hwsim_rht_params);
+	if (err < 0) {
+		if (info) {
+			GENL_SET_ERR_MSG(info, "perm addr already present");
+			NL_SET_BAD_ATTR(info->extack,
+					info->attrs[HWSIM_ATTR_PERM_ADDR]);
+		}
+		spin_unlock_bh(&hwsim_radio_lock);
+		goto failed_final_insert;
+	}
+
+	list_add_tail(&data->list, &hwsim_radios);
+	hwsim_radios_generation++;
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	hwsim_mcast_new_radio(idx, info, param);
+
+	return idx;
+
+failed_final_insert:
+	debugfs_remove_recursive(data->debugfs);
+	ieee80211_unregister_hw(data->hw);
+failed_hw:
+	device_release_driver(data->dev);
+failed_bind:
+	device_unregister(data->dev);
+failed_drvdata:
+	ieee80211_free_hw(hw);
+failed:
+	return err;
+}
+
+static void hwsim_mcast_del_radio(int id, const char *hwname,
+				  struct genl_info *info)
+{
+	struct sk_buff *skb;
+	void *data;
+	int ret;
+
+	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!skb)
+		return;
+
+	data = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0,
+			   HWSIM_CMD_DEL_RADIO);
+	if (!data)
+		goto error;
+
+	ret = nla_put_u32(skb, HWSIM_ATTR_RADIO_ID, id);
+	if (ret < 0)
+		goto error;
+
+	ret = nla_put(skb, HWSIM_ATTR_RADIO_NAME, strlen(hwname),
+		      hwname);
+	if (ret < 0)
+		goto error;
+
+	genlmsg_end(skb, data);
+
+	hwsim_mcast_config_msg(skb, info);
+
+	return;
+
+error:
+	nlmsg_free(skb);
+}
+
+static void mac80211_hwsim_del_radio(struct mac80211_hwsim_data *data,
+				     const char *hwname,
+				     struct genl_info *info)
+{
+	hwsim_mcast_del_radio(data->idx, hwname, info);
+	debugfs_remove_recursive(data->debugfs);
+	ieee80211_unregister_hw(data->hw);
+	device_release_driver(data->dev);
+	device_unregister(data->dev);
+	ieee80211_free_hw(data->hw);
+}
+
+static int mac80211_hwsim_get_radio(struct sk_buff *skb,
+				    struct mac80211_hwsim_data *data,
+				    u32 portid, u32 seq,
+				    struct netlink_callback *cb, int flags)
+{
+	void *hdr;
+	struct hwsim_new_radio_params param = { };
+	int res = -EMSGSIZE;
+
+	hdr = genlmsg_put(skb, portid, seq, &hwsim_genl_family, flags,
+			  HWSIM_CMD_GET_RADIO);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	if (cb)
+		genl_dump_check_consistent(cb, hdr);
+
+	if (data->alpha2[0] && data->alpha2[1])
+		param.reg_alpha2 = data->alpha2;
+
+	param.reg_strict = !!(data->hw->wiphy->regulatory_flags &
+					REGULATORY_STRICT_REG);
+	param.p2p_device = !!(data->hw->wiphy->interface_modes &
+					BIT(NL80211_IFTYPE_P2P_DEVICE));
+	param.use_chanctx = data->use_chanctx;
+	param.regd = data->regd;
+	param.channels = data->channels;
+	param.hwname = wiphy_name(data->hw->wiphy);
+
+	res = append_radio_msg(skb, data->idx, &param);
+	if (res < 0)
+		goto out_err;
+
+	genlmsg_end(skb, hdr);
+	return 0;
+
+out_err:
+	genlmsg_cancel(skb, hdr);
+	return res;
+}
+
+static void mac80211_hwsim_free(void)
+{
+	struct mac80211_hwsim_data *data;
+
+	spin_lock_bh(&hwsim_radio_lock);
+	while ((data = list_first_entry_or_null(&hwsim_radios,
+						struct mac80211_hwsim_data,
+						list))) {
+		list_del(&data->list);
+		spin_unlock_bh(&hwsim_radio_lock);
+		mac80211_hwsim_del_radio(data, wiphy_name(data->hw->wiphy),
+					 NULL);
+		spin_lock_bh(&hwsim_radio_lock);
+	}
+	spin_unlock_bh(&hwsim_radio_lock);
+	class_destroy(hwsim_class);
+}
+
+static const struct net_device_ops hwsim_netdev_ops = {
+	.ndo_start_xmit 	= hwsim_mon_xmit,
+	.ndo_set_mac_address 	= eth_mac_addr,
+	.ndo_validate_addr	= eth_validate_addr,
+};
+
+static void hwsim_mon_setup(struct net_device *dev)
+{
+	u8 addr[ETH_ALEN];
+
+	dev->netdev_ops = &hwsim_netdev_ops;
+	dev->needs_free_netdev = true;
+	ether_setup(dev);
+	dev->priv_flags |= IFF_NO_QUEUE;
+	dev->type = ARPHRD_IEEE80211_RADIOTAP;
+	eth_zero_addr(addr);
+	addr[0] = 0x12;
+	eth_hw_addr_set(dev, addr);
+}
+
+static struct mac80211_hwsim_data *get_hwsim_data_ref_from_addr(const u8 *addr)
+{
+	return rhashtable_lookup_fast(&hwsim_radios_rht,
+				      addr,
+				      hwsim_rht_params);
+}
+
+static void hwsim_register_wmediumd(struct net *net, u32 portid)
+{
+	struct mac80211_hwsim_data *data;
+
+	hwsim_net_set_wmediumd(net, portid);
+
+	spin_lock_bh(&hwsim_radio_lock);
+	list_for_each_entry(data, &hwsim_radios, list) {
+		if (data->netgroup == hwsim_net_get_netgroup(net))
+			data->wmediumd = portid;
+	}
+	spin_unlock_bh(&hwsim_radio_lock);
+}
+
+static int hwsim_tx_info_frame_received_nl(struct sk_buff *skb_2,
+					   struct genl_info *info)
+{
+
+	struct ieee80211_hdr *hdr;
+	struct mac80211_hwsim_data *data2;
+	struct ieee80211_tx_info *txi;
+	struct hwsim_tx_rate *tx_attempts;
+	u64 ret_skb_cookie;
+	struct sk_buff *skb, *tmp;
+	const u8 *src;
+	unsigned int hwsim_flags;
+	int i;
+	unsigned long flags;
+	bool found = false;
+
+	if (!info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER] ||
+	    !info->attrs[HWSIM_ATTR_FLAGS] ||
+	    !info->attrs[HWSIM_ATTR_COOKIE] ||
+	    !info->attrs[HWSIM_ATTR_SIGNAL] ||
+	    !info->attrs[HWSIM_ATTR_TX_INFO])
+		goto out;
+
+	src = (void *)nla_data(info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER]);
+	hwsim_flags = nla_get_u32(info->attrs[HWSIM_ATTR_FLAGS]);
+	ret_skb_cookie = nla_get_u64(info->attrs[HWSIM_ATTR_COOKIE]);
+
+	data2 = get_hwsim_data_ref_from_addr(src);
+	if (!data2)
+		goto out;
+
+	if (!hwsim_virtio_enabled) {
+		if (hwsim_net_get_netgroup(genl_info_net(info)) !=
+		    data2->netgroup)
+			goto out;
+
+		if (info->snd_portid != data2->wmediumd)
+			goto out;
+	}
+
+	/* look for the skb matching the cookie passed back from user */
+	spin_lock_irqsave(&data2->pending.lock, flags);
+	skb_queue_walk_safe(&data2->pending, skb, tmp) {
+		uintptr_t skb_cookie;
+
+		txi = IEEE80211_SKB_CB(skb);
+		skb_cookie = (uintptr_t)txi->rate_driver_data[0];
+
+		if (skb_cookie == ret_skb_cookie) {
+			__skb_unlink(skb, &data2->pending);
+			found = true;
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&data2->pending.lock, flags);
+
+	/* not found */
+	if (!found)
+		goto out;
+
+	/* Tx info received because the frame was broadcasted on user space,
+	 so we get all the necessary info: tx attempts and skb control buff */
+
+	tx_attempts = (struct hwsim_tx_rate *)nla_data(
+		       info->attrs[HWSIM_ATTR_TX_INFO]);
+
+	/* now send back TX status */
+	txi = IEEE80211_SKB_CB(skb);
+
+	ieee80211_tx_info_clear_status(txi);
+
+	for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) {
+		txi->status.rates[i].idx = tx_attempts[i].idx;
+		txi->status.rates[i].count = tx_attempts[i].count;
+	}
+
+	txi->status.ack_signal = nla_get_u32(info->attrs[HWSIM_ATTR_SIGNAL]);
+
+	if (!(hwsim_flags & HWSIM_TX_CTL_NO_ACK) &&
+	   (hwsim_flags & HWSIM_TX_STAT_ACK)) {
+		if (skb->len >= 16) {
+			hdr = (struct ieee80211_hdr *) skb->data;
+			mac80211_hwsim_monitor_ack(data2->channel,
+						   hdr->addr2);
+		}
+		txi->flags |= IEEE80211_TX_STAT_ACK;
+	}
+
+	if (hwsim_flags & HWSIM_TX_CTL_NO_ACK)
+		txi->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
+
+	ieee80211_tx_status_irqsafe(data2->hw, skb);
+	return 0;
+out:
+	return -EINVAL;
+
+}
+
+static int hwsim_cloned_frame_received_nl(struct sk_buff *skb_2,
+					  struct genl_info *info)
+{
+	struct mac80211_hwsim_data *data2;
+	struct ieee80211_rx_status rx_status;
+	struct ieee80211_hdr *hdr;
+	const u8 *dst;
+	int frame_data_len;
+	void *frame_data;
+	struct sk_buff *skb = NULL;
+	struct ieee80211_channel *channel = NULL;
+
+	if (!info->attrs[HWSIM_ATTR_ADDR_RECEIVER] ||
+	    !info->attrs[HWSIM_ATTR_FRAME] ||
+	    !info->attrs[HWSIM_ATTR_RX_RATE] ||
+	    !info->attrs[HWSIM_ATTR_SIGNAL])
+		goto out;
+
+	dst = (void *)nla_data(info->attrs[HWSIM_ATTR_ADDR_RECEIVER]);
+	frame_data_len = nla_len(info->attrs[HWSIM_ATTR_FRAME]);
+	frame_data = (void *)nla_data(info->attrs[HWSIM_ATTR_FRAME]);
+
+	/* Allocate new skb here */
+	skb = alloc_skb(frame_data_len, GFP_KERNEL);
+	if (skb == NULL)
+		goto err;
+
+	if (frame_data_len > IEEE80211_MAX_DATA_LEN)
+		goto err;
+
+	/* Copy the data */
+	skb_put_data(skb, frame_data, frame_data_len);
+
+	data2 = get_hwsim_data_ref_from_addr(dst);
+	if (!data2)
+		goto out;
+
+	if (data2->use_chanctx) {
+		if (data2->tmp_chan)
+			channel = data2->tmp_chan;
+	} else {
+		channel = data2->channel;
+	}
+
+	if (!hwsim_virtio_enabled) {
+		if (hwsim_net_get_netgroup(genl_info_net(info)) !=
+		    data2->netgroup)
+			goto out;
+
+		if (info->snd_portid != data2->wmediumd)
+			goto out;
+	}
+
+	/* check if radio is configured properly */
+
+	if ((data2->idle && !data2->tmp_chan) || !data2->started)
+		goto out;
+
+	/* A frame is received from user space */
+	memset(&rx_status, 0, sizeof(rx_status));
+	if (info->attrs[HWSIM_ATTR_FREQ]) {
+		struct tx_iter_data iter_data = {};
+
+		/* throw away off-channel packets, but allow both the temporary
+		 * ("hw" scan/remain-on-channel), regular channels and links,
+		 * since the internal datapath also allows this
+		 */
+		rx_status.freq = nla_get_u32(info->attrs[HWSIM_ATTR_FREQ]);
+
+		iter_data.channel = ieee80211_get_channel(data2->hw->wiphy,
+							  rx_status.freq);
+		if (!iter_data.channel)
+			goto out;
+		rx_status.band = iter_data.channel->band;
+
+		mutex_lock(&data2->mutex);
+		if (!hwsim_chans_compat(iter_data.channel, channel)) {
+			ieee80211_iterate_active_interfaces_atomic(
+				data2->hw, IEEE80211_IFACE_ITER_NORMAL,
+				mac80211_hwsim_tx_iter, &iter_data);
+			if (!iter_data.receive) {
+				mutex_unlock(&data2->mutex);
+				goto out;
+			}
+		}
+		mutex_unlock(&data2->mutex);
+	} else if (!channel) {
+		goto out;
+	} else {
+		rx_status.freq = channel->center_freq;
+		rx_status.band = channel->band;
+	}
+
+	rx_status.rate_idx = nla_get_u32(info->attrs[HWSIM_ATTR_RX_RATE]);
+	if (rx_status.rate_idx >= data2->hw->wiphy->bands[rx_status.band]->n_bitrates)
+		goto out;
+	rx_status.signal = nla_get_u32(info->attrs[HWSIM_ATTR_SIGNAL]);
+
+	hdr = (void *)skb->data;
+
+	if (ieee80211_is_beacon(hdr->frame_control) ||
+	    ieee80211_is_probe_resp(hdr->frame_control))
+		rx_status.boottime_ns = ktime_get_boottime_ns();
+
+	mac80211_hwsim_rx(data2, &rx_status, skb);
+
+	return 0;
+err:
+	pr_debug("mac80211_hwsim: error occurred in %s\n", __func__);
+out:
+	dev_kfree_skb(skb);
+	return -EINVAL;
+}
+
+static int hwsim_register_received_nl(struct sk_buff *skb_2,
+				      struct genl_info *info)
+{
+	struct net *net = genl_info_net(info);
+	struct mac80211_hwsim_data *data;
+	int chans = 1;
+
+	spin_lock_bh(&hwsim_radio_lock);
+	list_for_each_entry(data, &hwsim_radios, list)
+		chans = max(chans, data->channels);
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	/* In the future we should revise the userspace API and allow it
+	 * to set a flag that it does support multi-channel, then we can
+	 * let this pass conditionally on the flag.
+	 * For current userspace, prohibit it since it won't work right.
+	 */
+	if (chans > 1)
+		return -EOPNOTSUPP;
+
+	if (hwsim_net_get_wmediumd(net))
+		return -EBUSY;
+
+	hwsim_register_wmediumd(net, info->snd_portid);
+
+	pr_debug("mac80211_hwsim: received a REGISTER, "
+	       "switching to wmediumd mode with pid %d\n", info->snd_portid);
+
+	return 0;
+}
+
+/* ensures ciphers only include ciphers listed in 'hwsim_ciphers' array */
+static bool hwsim_known_ciphers(const u32 *ciphers, int n_ciphers)
+{
+	int i;
+
+	for (i = 0; i < n_ciphers; i++) {
+		int j;
+		int found = 0;
+
+		for (j = 0; j < ARRAY_SIZE(hwsim_ciphers); j++) {
+			if (ciphers[i] == hwsim_ciphers[j]) {
+				found = 1;
+				break;
+			}
+		}
+
+		if (!found)
+			return false;
+	}
+
+	return true;
+}
+
+static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info)
+{
+	struct hwsim_new_radio_params param = { 0 };
+	const char *hwname = NULL;
+	int ret;
+
+	param.reg_strict = info->attrs[HWSIM_ATTR_REG_STRICT_REG];
+	param.p2p_device = info->attrs[HWSIM_ATTR_SUPPORT_P2P_DEVICE];
+	param.channels = channels;
+	param.destroy_on_close =
+		info->attrs[HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE];
+
+	if (info->attrs[HWSIM_ATTR_CHANNELS])
+		param.channels = nla_get_u32(info->attrs[HWSIM_ATTR_CHANNELS]);
+
+	if (param.channels < 1) {
+		GENL_SET_ERR_MSG(info, "must have at least one channel");
+		return -EINVAL;
+	}
+
+	if (info->attrs[HWSIM_ATTR_NO_VIF])
+		param.no_vif = true;
+
+	if (info->attrs[HWSIM_ATTR_USE_CHANCTX])
+		param.use_chanctx = true;
+	else
+		param.use_chanctx = (param.channels > 1);
+
+	if (info->attrs[HWSIM_ATTR_REG_HINT_ALPHA2])
+		param.reg_alpha2 =
+			nla_data(info->attrs[HWSIM_ATTR_REG_HINT_ALPHA2]);
+
+	if (info->attrs[HWSIM_ATTR_REG_CUSTOM_REG]) {
+		u32 idx = nla_get_u32(info->attrs[HWSIM_ATTR_REG_CUSTOM_REG]);
+
+		if (idx >= ARRAY_SIZE(hwsim_world_regdom_custom))
+			return -EINVAL;
+
+		idx = array_index_nospec(idx,
+					 ARRAY_SIZE(hwsim_world_regdom_custom));
+		param.regd = hwsim_world_regdom_custom[idx];
+	}
+
+	if (info->attrs[HWSIM_ATTR_PERM_ADDR]) {
+		if (!is_valid_ether_addr(
+				nla_data(info->attrs[HWSIM_ATTR_PERM_ADDR]))) {
+			GENL_SET_ERR_MSG(info,"MAC is no valid source addr");
+			NL_SET_BAD_ATTR(info->extack,
+					info->attrs[HWSIM_ATTR_PERM_ADDR]);
+			return -EINVAL;
+		}
+
+		param.perm_addr = nla_data(info->attrs[HWSIM_ATTR_PERM_ADDR]);
+	}
+
+	if (info->attrs[HWSIM_ATTR_IFTYPE_SUPPORT]) {
+		param.iftypes =
+			nla_get_u32(info->attrs[HWSIM_ATTR_IFTYPE_SUPPORT]);
+
+		if (param.iftypes & ~HWSIM_IFTYPE_SUPPORT_MASK) {
+			NL_SET_ERR_MSG_ATTR(info->extack,
+					    info->attrs[HWSIM_ATTR_IFTYPE_SUPPORT],
+					    "cannot support more iftypes than kernel");
+			return -EINVAL;
+		}
+	} else {
+		param.iftypes = HWSIM_IFTYPE_SUPPORT_MASK;
+	}
+
+	/* ensure both flag and iftype support is honored */
+	if (param.p2p_device ||
+	    param.iftypes & BIT(NL80211_IFTYPE_P2P_DEVICE)) {
+		param.iftypes |= BIT(NL80211_IFTYPE_P2P_DEVICE);
+		param.p2p_device = true;
+	}
+
+	if (info->attrs[HWSIM_ATTR_CIPHER_SUPPORT]) {
+		u32 len = nla_len(info->attrs[HWSIM_ATTR_CIPHER_SUPPORT]);
+
+		param.ciphers =
+			nla_data(info->attrs[HWSIM_ATTR_CIPHER_SUPPORT]);
+
+		if (len % sizeof(u32)) {
+			NL_SET_ERR_MSG_ATTR(info->extack,
+					    info->attrs[HWSIM_ATTR_CIPHER_SUPPORT],
+					    "bad cipher list length");
+			return -EINVAL;
+		}
+
+		param.n_ciphers = len / sizeof(u32);
+
+		if (param.n_ciphers > ARRAY_SIZE(hwsim_ciphers)) {
+			NL_SET_ERR_MSG_ATTR(info->extack,
+					    info->attrs[HWSIM_ATTR_CIPHER_SUPPORT],
+					    "too many ciphers specified");
+			return -EINVAL;
+		}
+
+		if (!hwsim_known_ciphers(param.ciphers, param.n_ciphers)) {
+			NL_SET_ERR_MSG_ATTR(info->extack,
+					    info->attrs[HWSIM_ATTR_CIPHER_SUPPORT],
+					    "unsupported ciphers specified");
+			return -EINVAL;
+		}
+	}
+
+	param.mlo = info->attrs[HWSIM_ATTR_MLO_SUPPORT];
+
+	if (param.mlo)
+		param.use_chanctx = true;
+
+	if (info->attrs[HWSIM_ATTR_RADIO_NAME]) {
+		hwname = kstrndup((char *)nla_data(info->attrs[HWSIM_ATTR_RADIO_NAME]),
+				  nla_len(info->attrs[HWSIM_ATTR_RADIO_NAME]),
+				  GFP_KERNEL);
+		if (!hwname)
+			return -ENOMEM;
+		param.hwname = hwname;
+	}
+
+	ret = mac80211_hwsim_new_radio(info, &param);
+	kfree(hwname);
+	return ret;
+}
+
+static int hwsim_del_radio_nl(struct sk_buff *msg, struct genl_info *info)
+{
+	struct mac80211_hwsim_data *data;
+	s64 idx = -1;
+	const char *hwname = NULL;
+
+	if (info->attrs[HWSIM_ATTR_RADIO_ID]) {
+		idx = nla_get_u32(info->attrs[HWSIM_ATTR_RADIO_ID]);
+	} else if (info->attrs[HWSIM_ATTR_RADIO_NAME]) {
+		hwname = kstrndup((char *)nla_data(info->attrs[HWSIM_ATTR_RADIO_NAME]),
+				  nla_len(info->attrs[HWSIM_ATTR_RADIO_NAME]),
+				  GFP_KERNEL);
+		if (!hwname)
+			return -ENOMEM;
+	} else
+		return -EINVAL;
+
+	spin_lock_bh(&hwsim_radio_lock);
+	list_for_each_entry(data, &hwsim_radios, list) {
+		if (idx >= 0) {
+			if (data->idx != idx)
+				continue;
+		} else {
+			if (!hwname ||
+			    strcmp(hwname, wiphy_name(data->hw->wiphy)))
+				continue;
+		}
+
+		if (!net_eq(wiphy_net(data->hw->wiphy), genl_info_net(info)))
+			continue;
+
+		list_del(&data->list);
+		rhashtable_remove_fast(&hwsim_radios_rht, &data->rht,
+				       hwsim_rht_params);
+		hwsim_radios_generation++;
+		spin_unlock_bh(&hwsim_radio_lock);
+		mac80211_hwsim_del_radio(data, wiphy_name(data->hw->wiphy),
+					 info);
+		kfree(hwname);
+		return 0;
+	}
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	kfree(hwname);
+	return -ENODEV;
+}
+
+static int hwsim_get_radio_nl(struct sk_buff *msg, struct genl_info *info)
+{
+	struct mac80211_hwsim_data *data;
+	struct sk_buff *skb;
+	int idx, res = -ENODEV;
+
+	if (!info->attrs[HWSIM_ATTR_RADIO_ID])
+		return -EINVAL;
+	idx = nla_get_u32(info->attrs[HWSIM_ATTR_RADIO_ID]);
+
+	spin_lock_bh(&hwsim_radio_lock);
+	list_for_each_entry(data, &hwsim_radios, list) {
+		if (data->idx != idx)
+			continue;
+
+		if (!net_eq(wiphy_net(data->hw->wiphy), genl_info_net(info)))
+			continue;
+
+		skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
+		if (!skb) {
+			res = -ENOMEM;
+			goto out_err;
+		}
+
+		res = mac80211_hwsim_get_radio(skb, data, info->snd_portid,
+					       info->snd_seq, NULL, 0);
+		if (res < 0) {
+			nlmsg_free(skb);
+			goto out_err;
+		}
+
+		res = genlmsg_reply(skb, info);
+		break;
+	}
+
+out_err:
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	return res;
+}
+
+static int hwsim_dump_radio_nl(struct sk_buff *skb,
+			       struct netlink_callback *cb)
+{
+	int last_idx = cb->args[0] - 1;
+	struct mac80211_hwsim_data *data = NULL;
+	int res = 0;
+	void *hdr;
+
+	spin_lock_bh(&hwsim_radio_lock);
+	cb->seq = hwsim_radios_generation;
+
+	if (last_idx >= hwsim_radio_idx-1)
+		goto done;
+
+	list_for_each_entry(data, &hwsim_radios, list) {
+		if (data->idx <= last_idx)
+			continue;
+
+		if (!net_eq(wiphy_net(data->hw->wiphy), sock_net(skb->sk)))
+			continue;
+
+		res = mac80211_hwsim_get_radio(skb, data,
+					       NETLINK_CB(cb->skb).portid,
+					       cb->nlh->nlmsg_seq, cb,
+					       NLM_F_MULTI);
+		if (res < 0)
+			break;
+
+		last_idx = data->idx;
+	}
+
+	cb->args[0] = last_idx + 1;
+
+	/* list changed, but no new element sent, set interrupted flag */
+	if (skb->len == 0 && cb->prev_seq && cb->seq != cb->prev_seq) {
+		hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid,
+				  cb->nlh->nlmsg_seq, &hwsim_genl_family,
+				  NLM_F_MULTI, HWSIM_CMD_GET_RADIO);
+		if (hdr) {
+			genl_dump_check_consistent(cb, hdr);
+			genlmsg_end(skb, hdr);
+		} else {
+			res = -EMSGSIZE;
+		}
+	}
+
+done:
+	spin_unlock_bh(&hwsim_radio_lock);
+	return res ?: skb->len;
+}
+
+/* Generic Netlink operations array */
+static const struct genl_small_ops hwsim_ops[] = {
+	{
+		.cmd = HWSIM_CMD_REGISTER,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_register_received_nl,
+		.flags = GENL_UNS_ADMIN_PERM,
+	},
+	{
+		.cmd = HWSIM_CMD_FRAME,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_cloned_frame_received_nl,
+	},
+	{
+		.cmd = HWSIM_CMD_TX_INFO_FRAME,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_tx_info_frame_received_nl,
+	},
+	{
+		.cmd = HWSIM_CMD_NEW_RADIO,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_new_radio_nl,
+		.flags = GENL_UNS_ADMIN_PERM,
+	},
+	{
+		.cmd = HWSIM_CMD_DEL_RADIO,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_del_radio_nl,
+		.flags = GENL_UNS_ADMIN_PERM,
+	},
+	{
+		.cmd = HWSIM_CMD_GET_RADIO,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_get_radio_nl,
+		.dumpit = hwsim_dump_radio_nl,
+	},
+};
+
+static struct genl_family hwsim_genl_family __ro_after_init = {
+	.name = "MAC80211_HWSIM",
+	.version = 1,
+	.maxattr = HWSIM_ATTR_MAX,
+	.policy = hwsim_genl_policy,
+	.netnsok = true,
+	.module = THIS_MODULE,
+	.small_ops = hwsim_ops,
+	.n_small_ops = ARRAY_SIZE(hwsim_ops),
+	.resv_start_op = HWSIM_CMD_DEL_MAC_ADDR + 1,
+	.mcgrps = hwsim_mcgrps,
+	.n_mcgrps = ARRAY_SIZE(hwsim_mcgrps),
+};
+
+static void remove_user_radios(u32 portid)
+{
+	struct mac80211_hwsim_data *entry, *tmp;
+	LIST_HEAD(list);
+
+	spin_lock_bh(&hwsim_radio_lock);
+	list_for_each_entry_safe(entry, tmp, &hwsim_radios, list) {
+		if (entry->destroy_on_close && entry->portid == portid) {
+			list_move(&entry->list, &list);
+			rhashtable_remove_fast(&hwsim_radios_rht, &entry->rht,
+					       hwsim_rht_params);
+			hwsim_radios_generation++;
+		}
+	}
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	list_for_each_entry_safe(entry, tmp, &list, list) {
+		list_del(&entry->list);
+		mac80211_hwsim_del_radio(entry, wiphy_name(entry->hw->wiphy),
+					 NULL);
+	}
+}
+
+static int mac80211_hwsim_netlink_notify(struct notifier_block *nb,
+					 unsigned long state,
+					 void *_notify)
+{
+	struct netlink_notify *notify = _notify;
+
+	if (state != NETLINK_URELEASE)
+		return NOTIFY_DONE;
+
+	remove_user_radios(notify->portid);
+
+	if (notify->portid == hwsim_net_get_wmediumd(notify->net)) {
+		printk(KERN_INFO "mac80211_hwsim: wmediumd released netlink"
+		       " socket, switching to perfect channel medium\n");
+		hwsim_register_wmediumd(notify->net, 0);
+	}
+	return NOTIFY_DONE;
+
+}
+
+static struct notifier_block hwsim_netlink_notifier = {
+	.notifier_call = mac80211_hwsim_netlink_notify,
+};
+
+static int __init hwsim_init_netlink(void)
+{
+	int rc;
+
+	printk(KERN_INFO "mac80211_hwsim: initializing netlink\n");
+
+	rc = genl_register_family(&hwsim_genl_family);
+	if (rc)
+		goto failure;
+
+	rc = netlink_register_notifier(&hwsim_netlink_notifier);
+	if (rc) {
+		genl_unregister_family(&hwsim_genl_family);
+		goto failure;
+	}
+
+	return 0;
+
+failure:
+	pr_debug("mac80211_hwsim: error occurred in %s\n", __func__);
+	return -EINVAL;
+}
+
+static __net_init int hwsim_init_net(struct net *net)
+{
+	return hwsim_net_set_netgroup(net);
+}
+
+static void __net_exit hwsim_exit_net(struct net *net)
+{
+	struct mac80211_hwsim_data *data, *tmp;
+	LIST_HEAD(list);
+
+	spin_lock_bh(&hwsim_radio_lock);
+	list_for_each_entry_safe(data, tmp, &hwsim_radios, list) {
+		if (!net_eq(wiphy_net(data->hw->wiphy), net))
+			continue;
+
+		/* Radios created in init_net are returned to init_net. */
+		if (data->netgroup == hwsim_net_get_netgroup(&init_net))
+			continue;
+
+		list_move(&data->list, &list);
+		rhashtable_remove_fast(&hwsim_radios_rht, &data->rht,
+				       hwsim_rht_params);
+		hwsim_radios_generation++;
+	}
+	spin_unlock_bh(&hwsim_radio_lock);
+
+	list_for_each_entry_safe(data, tmp, &list, list) {
+		list_del(&data->list);
+		mac80211_hwsim_del_radio(data,
+					 wiphy_name(data->hw->wiphy),
+					 NULL);
+	}
+
+	ida_free(&hwsim_netgroup_ida, hwsim_net_get_netgroup(net));
+}
+
+static struct pernet_operations hwsim_net_ops = {
+	.init = hwsim_init_net,
+	.exit = hwsim_exit_net,
+	.id   = &hwsim_net_id,
+	.size = sizeof(struct hwsim_net),
+};
+
+static void hwsim_exit_netlink(void)
+{
+	/* unregister the notifier */
+	netlink_unregister_notifier(&hwsim_netlink_notifier);
+	/* unregister the family */
+	genl_unregister_family(&hwsim_genl_family);
+}
+
+#if IS_REACHABLE(CONFIG_VIRTIO)
+static void hwsim_virtio_tx_done(struct virtqueue *vq)
+{
+	unsigned int len;
+	struct sk_buff *skb;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwsim_virtio_lock, flags);
+	while ((skb = virtqueue_get_buf(vq, &len)))
+		nlmsg_free(skb);
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+}
+
+static int hwsim_virtio_handle_cmd(struct sk_buff *skb)
+{
+	struct nlmsghdr *nlh;
+	struct genlmsghdr *gnlh;
+	struct nlattr *tb[HWSIM_ATTR_MAX + 1];
+	struct genl_info info = {};
+	int err;
+
+	nlh = nlmsg_hdr(skb);
+	gnlh = nlmsg_data(nlh);
+
+	if (skb->len < nlh->nlmsg_len)
+		return -EINVAL;
+
+	err = genlmsg_parse(nlh, &hwsim_genl_family, tb, HWSIM_ATTR_MAX,
+			    hwsim_genl_policy, NULL);
+	if (err) {
+		pr_err_ratelimited("hwsim: genlmsg_parse returned %d\n", err);
+		return err;
+	}
+
+	info.attrs = tb;
+
+	switch (gnlh->cmd) {
+	case HWSIM_CMD_FRAME:
+		hwsim_cloned_frame_received_nl(skb, &info);
+		break;
+	case HWSIM_CMD_TX_INFO_FRAME:
+		hwsim_tx_info_frame_received_nl(skb, &info);
+		break;
+	default:
+		pr_err_ratelimited("hwsim: invalid cmd: %d\n", gnlh->cmd);
+		return -EPROTO;
+	}
+	return 0;
+}
+
+static void hwsim_virtio_rx_work(struct work_struct *work)
+{
+	struct virtqueue *vq;
+	unsigned int len;
+	struct sk_buff *skb;
+	struct scatterlist sg[1];
+	int err;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwsim_virtio_lock, flags);
+	if (!hwsim_virtio_enabled)
+		goto out_unlock;
+
+	skb = virtqueue_get_buf(hwsim_vqs[HWSIM_VQ_RX], &len);
+	if (!skb)
+		goto out_unlock;
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+
+	skb->data = skb->head;
+	skb_reset_tail_pointer(skb);
+	skb_put(skb, len);
+	hwsim_virtio_handle_cmd(skb);
+
+	spin_lock_irqsave(&hwsim_virtio_lock, flags);
+	if (!hwsim_virtio_enabled) {
+		nlmsg_free(skb);
+		goto out_unlock;
+	}
+	vq = hwsim_vqs[HWSIM_VQ_RX];
+	sg_init_one(sg, skb->head, skb_end_offset(skb));
+	err = virtqueue_add_inbuf(vq, sg, 1, skb, GFP_ATOMIC);
+	if (WARN(err, "virtqueue_add_inbuf returned %d\n", err))
+		nlmsg_free(skb);
+	else
+		virtqueue_kick(vq);
+	schedule_work(&hwsim_virtio_rx);
+
+out_unlock:
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+}
+
+static void hwsim_virtio_rx_done(struct virtqueue *vq)
+{
+	schedule_work(&hwsim_virtio_rx);
+}
+
+static int init_vqs(struct virtio_device *vdev)
+{
+	vq_callback_t *callbacks[HWSIM_NUM_VQS] = {
+		[HWSIM_VQ_TX] = hwsim_virtio_tx_done,
+		[HWSIM_VQ_RX] = hwsim_virtio_rx_done,
+	};
+	const char *names[HWSIM_NUM_VQS] = {
+		[HWSIM_VQ_TX] = "tx",
+		[HWSIM_VQ_RX] = "rx",
+	};
+
+	return virtio_find_vqs(vdev, HWSIM_NUM_VQS,
+			       hwsim_vqs, callbacks, names, NULL);
+}
+
+static int fill_vq(struct virtqueue *vq)
+{
+	int i, err;
+	struct sk_buff *skb;
+	struct scatterlist sg[1];
+
+	for (i = 0; i < virtqueue_get_vring_size(vq); i++) {
+		skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+		if (!skb)
+			return -ENOMEM;
+
+		sg_init_one(sg, skb->head, skb_end_offset(skb));
+		err = virtqueue_add_inbuf(vq, sg, 1, skb, GFP_KERNEL);
+		if (err) {
+			nlmsg_free(skb);
+			return err;
+		}
+	}
+	virtqueue_kick(vq);
+	return 0;
+}
+
+static void remove_vqs(struct virtio_device *vdev)
+{
+	int i;
+
+	virtio_reset_device(vdev);
+
+	for (i = 0; i < ARRAY_SIZE(hwsim_vqs); i++) {
+		struct virtqueue *vq = hwsim_vqs[i];
+		struct sk_buff *skb;
+
+		while ((skb = virtqueue_detach_unused_buf(vq)))
+			nlmsg_free(skb);
+	}
+
+	vdev->config->del_vqs(vdev);
+}
+
+static int hwsim_virtio_probe(struct virtio_device *vdev)
+{
+	int err;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwsim_virtio_lock, flags);
+	if (hwsim_virtio_enabled) {
+		spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+		return -EEXIST;
+	}
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+
+	err = init_vqs(vdev);
+	if (err)
+		return err;
+
+	virtio_device_ready(vdev);
+
+	err = fill_vq(hwsim_vqs[HWSIM_VQ_RX]);
+	if (err)
+		goto out_remove;
+
+	spin_lock_irqsave(&hwsim_virtio_lock, flags);
+	hwsim_virtio_enabled = true;
+	spin_unlock_irqrestore(&hwsim_virtio_lock, flags);
+
+	schedule_work(&hwsim_virtio_rx);
+	return 0;
+
+out_remove:
+	remove_vqs(vdev);
+	return err;
+}
+
+static void hwsim_virtio_remove(struct virtio_device *vdev)
+{
+	hwsim_virtio_enabled = false;
+
+	cancel_work_sync(&hwsim_virtio_rx);
+
+	remove_vqs(vdev);
+}
+
+/* MAC80211_HWSIM virtio device id table */
+static const struct virtio_device_id id_table[] = {
+	{ VIRTIO_ID_MAC80211_HWSIM, VIRTIO_DEV_ANY_ID },
+	{ 0 }
+};
+MODULE_DEVICE_TABLE(virtio, id_table);
+
+static struct virtio_driver virtio_hwsim = {
+	.driver.name = KBUILD_MODNAME,
+	.driver.owner = THIS_MODULE,
+	.id_table = id_table,
+	.probe = hwsim_virtio_probe,
+	.remove = hwsim_virtio_remove,
+};
+
+static int hwsim_register_virtio_driver(void)
+{
+	return register_virtio_driver(&virtio_hwsim);
+}
+
+static void hwsim_unregister_virtio_driver(void)
+{
+	unregister_virtio_driver(&virtio_hwsim);
+}
+#else
+static inline int hwsim_register_virtio_driver(void)
+{
+	return 0;
+}
+
+static inline void hwsim_unregister_virtio_driver(void)
+{
+}
+#endif
+
+static int __init init_mac80211_hwsim(void)
+{
+	int i, err;
+
+	if (radios < 0 || radios > 100)
+		return -EINVAL;
+
+	if (channels < 1)
+		return -EINVAL;
+
+	err = rhashtable_init(&hwsim_radios_rht, &hwsim_rht_params);
+	if (err)
+		return err;
+
+	err = register_pernet_device(&hwsim_net_ops);
+	if (err)
+		goto out_free_rht;
+
+	err = platform_driver_register(&mac80211_hwsim_driver);
+	if (err)
+		goto out_unregister_pernet;
+
+	err = hwsim_init_netlink();
+	if (err)
+		goto out_unregister_driver;
+
+	err = hwsim_register_virtio_driver();
+	if (err)
+		goto out_exit_netlink;
+
+	hwsim_class = class_create(THIS_MODULE, "mac80211_hwsim");
+	if (IS_ERR(hwsim_class)) {
+		err = PTR_ERR(hwsim_class);
+		goto out_exit_virtio;
+	}
+
+	hwsim_init_s1g_channels(hwsim_channels_s1g);
+
+	for (i = 0; i < radios; i++) {
+		struct hwsim_new_radio_params param = { 0 };
+
+		param.channels = channels;
+
+		switch (regtest) {
+		case HWSIM_REGTEST_DIFF_COUNTRY:
+			if (i < ARRAY_SIZE(hwsim_alpha2s))
+				param.reg_alpha2 = hwsim_alpha2s[i];
+			break;
+		case HWSIM_REGTEST_DRIVER_REG_FOLLOW:
+			if (!i)
+				param.reg_alpha2 = hwsim_alpha2s[0];
+			break;
+		case HWSIM_REGTEST_STRICT_ALL:
+			param.reg_strict = true;
+			fallthrough;
+		case HWSIM_REGTEST_DRIVER_REG_ALL:
+			param.reg_alpha2 = hwsim_alpha2s[0];
+			break;
+		case HWSIM_REGTEST_WORLD_ROAM:
+			if (i == 0)
+				param.regd = &hwsim_world_regdom_custom_01;
+			break;
+		case HWSIM_REGTEST_CUSTOM_WORLD:
+			param.regd = &hwsim_world_regdom_custom_01;
+			break;
+		case HWSIM_REGTEST_CUSTOM_WORLD_2:
+			if (i == 0)
+				param.regd = &hwsim_world_regdom_custom_01;
+			else if (i == 1)
+				param.regd = &hwsim_world_regdom_custom_02;
+			break;
+		case HWSIM_REGTEST_STRICT_FOLLOW:
+			if (i == 0) {
+				param.reg_strict = true;
+				param.reg_alpha2 = hwsim_alpha2s[0];
+			}
+			break;
+		case HWSIM_REGTEST_STRICT_AND_DRIVER_REG:
+			if (i == 0) {
+				param.reg_strict = true;
+				param.reg_alpha2 = hwsim_alpha2s[0];
+			} else if (i == 1) {
+				param.reg_alpha2 = hwsim_alpha2s[1];
+			}
+			break;
+		case HWSIM_REGTEST_ALL:
+			switch (i) {
+			case 0:
+				param.regd = &hwsim_world_regdom_custom_01;
+				break;
+			case 1:
+				param.regd = &hwsim_world_regdom_custom_02;
+				break;
+			case 2:
+				param.reg_alpha2 = hwsim_alpha2s[0];
+				break;
+			case 3:
+				param.reg_alpha2 = hwsim_alpha2s[1];
+				break;
+			case 4:
+				param.reg_strict = true;
+				param.reg_alpha2 = hwsim_alpha2s[2];
+				break;
+			}
+			break;
+		default:
+			break;
+		}
+
+		param.p2p_device = support_p2p_device;
+		param.mlo = mlo;
+		param.use_chanctx = channels > 1 || mlo;
+		param.iftypes = HWSIM_IFTYPE_SUPPORT_MASK;
+		if (param.p2p_device)
+			param.iftypes |= BIT(NL80211_IFTYPE_P2P_DEVICE);
+
+		err = mac80211_hwsim_new_radio(NULL, &param);
+		if (err < 0)
+			goto out_free_radios;
+	}
+
+	hwsim_mon = alloc_netdev(0, "hwsim%d", NET_NAME_UNKNOWN,
+				 hwsim_mon_setup);
+	if (hwsim_mon == NULL) {
+		err = -ENOMEM;
+		goto out_free_radios;
+	}
+
+	rtnl_lock();
+	err = dev_alloc_name(hwsim_mon, hwsim_mon->name);
+	if (err < 0) {
+		rtnl_unlock();
+		goto out_free_mon;
+	}
+
+	err = register_netdevice(hwsim_mon);
+	if (err < 0) {
+		rtnl_unlock();
+		goto out_free_mon;
+	}
+	rtnl_unlock();
+
+	return 0;
+
+out_free_mon:
+	free_netdev(hwsim_mon);
+out_free_radios:
+	mac80211_hwsim_free();
+out_exit_virtio:
+	hwsim_unregister_virtio_driver();
+out_exit_netlink:
+	hwsim_exit_netlink();
+out_unregister_driver:
+	platform_driver_unregister(&mac80211_hwsim_driver);
+out_unregister_pernet:
+	unregister_pernet_device(&hwsim_net_ops);
+out_free_rht:
+	rhashtable_destroy(&hwsim_radios_rht);
+	return err;
+}
+module_init(init_mac80211_hwsim);
+
+static void __exit exit_mac80211_hwsim(void)
+{
+	pr_debug("mac80211_hwsim: unregister radios\n");
+
+	hwsim_unregister_virtio_driver();
+	hwsim_exit_netlink();
+
+	mac80211_hwsim_free();
+
+	rhashtable_destroy(&hwsim_radios_rht);
+	unregister_netdev(hwsim_mon);
+	platform_driver_unregister(&mac80211_hwsim_driver);
+	unregister_pernet_device(&hwsim_net_ops);
+}
+module_exit(exit_mac80211_hwsim);
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
new file mode 100644
index 000000000000..527799b2de0f
--- /dev/null
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -0,0 +1,280 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * mac80211_hwsim - software simulator of 802.11 radio(s) for mac80211
+ * Copyright (c) 2008, Jouni Malinen <j@w1.fi>
+ * Copyright (c) 2011, Javier Lopez <jlopex@gmail.com>
+ * Copyright (C) 2020, 2022 Intel Corporation
+ */
+
+#ifndef __MAC80211_HWSIM_H
+#define __MAC80211_HWSIM_H
+
+/**
+ * enum hwsim_tx_control_flags - flags to describe transmission info/status
+ *
+ * These flags are used to give the wmediumd extra information in order to
+ * modify its behavior for each frame
+ *
+ * @HWSIM_TX_CTL_REQ_TX_STATUS: require TX status callback for this frame.
+ * @HWSIM_TX_CTL_NO_ACK: tell the wmediumd not to wait for an ack
+ * @HWSIM_TX_STAT_ACK: Frame was acknowledged
+ *
+ */
+enum hwsim_tx_control_flags {
+	HWSIM_TX_CTL_REQ_TX_STATUS		= BIT(0),
+	HWSIM_TX_CTL_NO_ACK			= BIT(1),
+	HWSIM_TX_STAT_ACK			= BIT(2),
+};
+
+/**
+ * DOC: Frame transmission/registration support
+ *
+ * Frame transmission and registration support exists to allow userspace
+ * entities such as wmediumd to receive and process all broadcasted
+ * frames from a mac80211_hwsim radio device.
+ *
+ * This allow user space applications to decide if the frame should be
+ * dropped or not and implement a wireless medium simulator at user space.
+ *
+ * Registration is done by sending a register message to the driver and
+ * will be automatically unregistered if the user application doesn't
+ * responds to sent frames.
+ * Once registered the user application has to take responsibility of
+ * broadcasting the frames to all listening mac80211_hwsim radio
+ * interfaces.
+ *
+ * For more technical details, see the corresponding command descriptions
+ * below.
+ */
+
+/**
+ * enum hwsim_commands - supported hwsim commands
+ *
+ * @HWSIM_CMD_UNSPEC: unspecified command to catch errors
+ *
+ * @HWSIM_CMD_REGISTER: request to register and received all broadcasted
+ *	frames by any mac80211_hwsim radio device.
+ * @HWSIM_CMD_FRAME: send/receive a broadcasted frame from/to kernel/user
+ *	space, uses:
+ *	%HWSIM_ATTR_ADDR_TRANSMITTER, %HWSIM_ATTR_ADDR_RECEIVER,
+ *	%HWSIM_ATTR_FRAME, %HWSIM_ATTR_FLAGS, %HWSIM_ATTR_RX_RATE,
+ *	%HWSIM_ATTR_SIGNAL, %HWSIM_ATTR_COOKIE, %HWSIM_ATTR_FREQ (optional)
+ * @HWSIM_CMD_TX_INFO_FRAME: Transmission info report from user space to
+ *	kernel, uses:
+ *	%HWSIM_ATTR_ADDR_TRANSMITTER, %HWSIM_ATTR_FLAGS,
+ *	%HWSIM_ATTR_TX_INFO, %WSIM_ATTR_TX_INFO_FLAGS,
+ *	%HWSIM_ATTR_SIGNAL, %HWSIM_ATTR_COOKIE
+ * @HWSIM_CMD_NEW_RADIO: create a new radio with the given parameters,
+ *	returns the radio ID (>= 0) or negative on errors, if successful
+ *	then multicast the result, uses optional parameter:
+ *	%HWSIM_ATTR_REG_STRICT_REG, %HWSIM_ATTR_SUPPORT_P2P_DEVICE,
+ *	%HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE, %HWSIM_ATTR_CHANNELS,
+ *	%HWSIM_ATTR_NO_VIF, %HWSIM_ATTR_RADIO_NAME, %HWSIM_ATTR_USE_CHANCTX,
+ *	%HWSIM_ATTR_REG_HINT_ALPHA2, %HWSIM_ATTR_REG_CUSTOM_REG,
+ *	%HWSIM_ATTR_PERM_ADDR
+ * @HWSIM_CMD_DEL_RADIO: destroy a radio, reply is multicasted
+ * @HWSIM_CMD_GET_RADIO: fetch information about existing radios, uses:
+ *	%HWSIM_ATTR_RADIO_ID
+ * @HWSIM_CMD_ADD_MAC_ADDR: add a receive MAC address (given in the
+ *	%HWSIM_ATTR_ADDR_RECEIVER attribute) to a device identified by
+ *	%HWSIM_ATTR_ADDR_TRANSMITTER. This lets wmediumd forward frames
+ *	to this receiver address for a given station.
+ * @HWSIM_CMD_DEL_MAC_ADDR: remove the MAC address again, the attributes
+ *	are the same as to @HWSIM_CMD_ADD_MAC_ADDR.
+ * @__HWSIM_CMD_MAX: enum limit
+ */
+enum {
+	HWSIM_CMD_UNSPEC,
+	HWSIM_CMD_REGISTER,
+	HWSIM_CMD_FRAME,
+	HWSIM_CMD_TX_INFO_FRAME,
+	HWSIM_CMD_NEW_RADIO,
+	HWSIM_CMD_DEL_RADIO,
+	HWSIM_CMD_GET_RADIO,
+	HWSIM_CMD_ADD_MAC_ADDR,
+	HWSIM_CMD_DEL_MAC_ADDR,
+	__HWSIM_CMD_MAX,
+};
+#define HWSIM_CMD_MAX (_HWSIM_CMD_MAX - 1)
+
+#define HWSIM_CMD_CREATE_RADIO   HWSIM_CMD_NEW_RADIO
+#define HWSIM_CMD_DESTROY_RADIO  HWSIM_CMD_DEL_RADIO
+
+/**
+ * enum hwsim_attrs - hwsim netlink attributes
+ *
+ * @HWSIM_ATTR_UNSPEC: unspecified attribute to catch errors
+ *
+ * @HWSIM_ATTR_ADDR_RECEIVER: MAC address of the radio device that
+ *	the frame is broadcasted to
+ * @HWSIM_ATTR_ADDR_TRANSMITTER: MAC address of the radio device that
+ *	the frame was broadcasted from
+ * @HWSIM_ATTR_FRAME: Data array
+ * @HWSIM_ATTR_FLAGS: mac80211 transmission flags, used to process
+	properly the frame at user space
+ * @HWSIM_ATTR_RX_RATE: estimated rx rate index for this frame at user
+	space
+ * @HWSIM_ATTR_SIGNAL: estimated RX signal for this frame at user
+	space
+ * @HWSIM_ATTR_TX_INFO: ieee80211_tx_rate array
+ * @HWSIM_ATTR_COOKIE: sk_buff cookie to identify the frame
+ * @HWSIM_ATTR_CHANNELS: u32 attribute used with the %HWSIM_CMD_CREATE_RADIO
+ *	command giving the number of channels supported by the new radio
+ * @HWSIM_ATTR_RADIO_ID: u32 attribute used with %HWSIM_CMD_DESTROY_RADIO
+ *	only to destroy a radio
+ * @HWSIM_ATTR_REG_HINT_ALPHA2: alpha2 for regulatoro driver hint
+ *	(nla string, length 2)
+ * @HWSIM_ATTR_REG_CUSTOM_REG: custom regulatory domain index (u32 attribute)
+ * @HWSIM_ATTR_REG_STRICT_REG: request REGULATORY_STRICT_REG (flag attribute)
+ * @HWSIM_ATTR_SUPPORT_P2P_DEVICE: support P2P Device virtual interface (flag)
+ * @HWSIM_ATTR_USE_CHANCTX: used with the %HWSIM_CMD_CREATE_RADIO
+ *	command to force use of channel contexts even when only a
+ *	single channel is supported
+ * @HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE: used with the %HWSIM_CMD_CREATE_RADIO
+ *	command to force radio removal when process that created the radio dies
+ * @HWSIM_ATTR_RADIO_NAME: Name of radio, e.g. phy666
+ * @HWSIM_ATTR_NO_VIF:  Do not create vif (wlanX) when creating radio.
+ * @HWSIM_ATTR_FREQ: Frequency at which packet is transmitted or received.
+ * @HWSIM_ATTR_TX_INFO_FLAGS: additional flags for corresponding
+ *	rates of %HWSIM_ATTR_TX_INFO
+ * @HWSIM_ATTR_PERM_ADDR: permanent mac address of new radio
+ * @HWSIM_ATTR_IFTYPE_SUPPORT: u32 attribute of supported interface types bits
+ * @HWSIM_ATTR_CIPHER_SUPPORT: u32 array of supported cipher types
+ * @HWSIM_ATTR_MLO_SUPPORT: claim MLO support (exact parameters TBD) for
+ *	the new radio
+ * @__HWSIM_ATTR_MAX: enum limit
+ */
+
+
+enum {
+	HWSIM_ATTR_UNSPEC,
+	HWSIM_ATTR_ADDR_RECEIVER,
+	HWSIM_ATTR_ADDR_TRANSMITTER,
+	HWSIM_ATTR_FRAME,
+	HWSIM_ATTR_FLAGS,
+	HWSIM_ATTR_RX_RATE,
+	HWSIM_ATTR_SIGNAL,
+	HWSIM_ATTR_TX_INFO,
+	HWSIM_ATTR_COOKIE,
+	HWSIM_ATTR_CHANNELS,
+	HWSIM_ATTR_RADIO_ID,
+	HWSIM_ATTR_REG_HINT_ALPHA2,
+	HWSIM_ATTR_REG_CUSTOM_REG,
+	HWSIM_ATTR_REG_STRICT_REG,
+	HWSIM_ATTR_SUPPORT_P2P_DEVICE,
+	HWSIM_ATTR_USE_CHANCTX,
+	HWSIM_ATTR_DESTROY_RADIO_ON_CLOSE,
+	HWSIM_ATTR_RADIO_NAME,
+	HWSIM_ATTR_NO_VIF,
+	HWSIM_ATTR_FREQ,
+	HWSIM_ATTR_PAD,
+	HWSIM_ATTR_TX_INFO_FLAGS,
+	HWSIM_ATTR_PERM_ADDR,
+	HWSIM_ATTR_IFTYPE_SUPPORT,
+	HWSIM_ATTR_CIPHER_SUPPORT,
+	HWSIM_ATTR_MLO_SUPPORT,
+	__HWSIM_ATTR_MAX,
+};
+#define HWSIM_ATTR_MAX (__HWSIM_ATTR_MAX - 1)
+
+/**
+ * struct hwsim_tx_rate - rate selection/status
+ *
+ * @idx: rate index to attempt to send with
+ * @count: number of tries in this rate before going to the next rate
+ *
+ * A value of -1 for @idx indicates an invalid rate and, if used
+ * in an array of retry rates, that no more rates should be tried.
+ *
+ * When used for transmit status reporting, the driver should
+ * always report the rate and number of retries used.
+ *
+ */
+struct hwsim_tx_rate {
+	s8 idx;
+	u8 count;
+} __packed;
+
+/**
+ * enum hwsim_tx_rate_flags - per-rate flags set by the rate control algorithm.
+ *	Inspired by structure mac80211_rate_control_flags. New flags may be
+ *	appended, but old flags not deleted, to keep compatibility for
+ *	userspace.
+ *
+ * These flags are set by the Rate control algorithm for each rate during tx,
+ * in the @flags member of struct ieee80211_tx_rate.
+ *
+ * @MAC80211_HWSIM_TX_RC_USE_RTS_CTS: Use RTS/CTS exchange for this rate.
+ * @MAC80211_HWSIM_TX_RC_USE_CTS_PROTECT: CTS-to-self protection is required.
+ *	This is set if the current BSS requires ERP protection.
+ * @MAC80211_HWSIM_TX_RC_USE_SHORT_PREAMBLE: Use short preamble.
+ * @MAC80211_HWSIM_TX_RC_MCS: HT rate.
+ * @MAC80211_HWSIM_TX_RC_VHT_MCS: VHT MCS rate, in this case the idx field is
+ *	split into a higher 4 bits (Nss) and lower 4 bits (MCS number)
+ * @MAC80211_HWSIM_TX_RC_GREEN_FIELD: Indicates whether this rate should be used
+ *	in Greenfield mode.
+ * @MAC80211_HWSIM_TX_RC_40_MHZ_WIDTH: Indicates if the Channel Width should be
+ *	40 MHz.
+ * @MAC80211_HWSIM_TX_RC_80_MHZ_WIDTH: Indicates 80 MHz transmission
+ * @MAC80211_HWSIM_TX_RC_160_MHZ_WIDTH: Indicates 160 MHz transmission
+ *	(80+80 isn't supported yet)
+ * @MAC80211_HWSIM_TX_RC_DUP_DATA: The frame should be transmitted on both of
+ *	the adjacent 20 MHz channels, if the current channel type is
+ *	NL80211_CHAN_HT40MINUS or NL80211_CHAN_HT40PLUS.
+ * @MAC80211_HWSIM_TX_RC_SHORT_GI: Short Guard interval should be used for this
+ *	rate.
+ */
+enum hwsim_tx_rate_flags {
+	MAC80211_HWSIM_TX_RC_USE_RTS_CTS		= BIT(0),
+	MAC80211_HWSIM_TX_RC_USE_CTS_PROTECT		= BIT(1),
+	MAC80211_HWSIM_TX_RC_USE_SHORT_PREAMBLE	= BIT(2),
+
+	/* rate index is an HT/VHT MCS instead of an index */
+	MAC80211_HWSIM_TX_RC_MCS			= BIT(3),
+	MAC80211_HWSIM_TX_RC_GREEN_FIELD		= BIT(4),
+	MAC80211_HWSIM_TX_RC_40_MHZ_WIDTH		= BIT(5),
+	MAC80211_HWSIM_TX_RC_DUP_DATA		= BIT(6),
+	MAC80211_HWSIM_TX_RC_SHORT_GI		= BIT(7),
+	MAC80211_HWSIM_TX_RC_VHT_MCS			= BIT(8),
+	MAC80211_HWSIM_TX_RC_80_MHZ_WIDTH		= BIT(9),
+	MAC80211_HWSIM_TX_RC_160_MHZ_WIDTH		= BIT(10),
+};
+
+/**
+ * struct hwsim_tx_rate - rate selection/status
+ *
+ * @idx: rate index to attempt to send with
+ * @count: number of tries in this rate before going to the next rate
+ *
+ * A value of -1 for @idx indicates an invalid rate and, if used
+ * in an array of retry rates, that no more rates should be tried.
+ *
+ * When used for transmit status reporting, the driver should
+ * always report the rate and number of retries used.
+ *
+ */
+struct hwsim_tx_rate_flag {
+	s8 idx;
+	u16 flags;
+} __packed;
+
+/**
+ * DOC: Frame transmission support over virtio
+ *
+ * Frame transmission is also supported over virtio to allow communication
+ * with external entities.
+ */
+
+/**
+ * enum hwsim_vqs - queues for virtio frame transmission
+ *
+ * @HWSIM_VQ_TX: send frames to external entity
+ * @HWSIM_VQ_RX: receive frames and transmission info reports
+ * @HWSIM_NUM_VQS: enum limit
+ */
+enum {
+	HWSIM_VQ_TX,
+	HWSIM_VQ_RX,
+	HWSIM_NUM_VQS,
+};
+#endif /* __MAC80211_HWSIM_H */
diff --git a/drivers/net/wireless/virtual/virt_wifi.c b/drivers/net/wireless/virtual/virt_wifi.c
new file mode 100644
index 000000000000..ba14d83353a4
--- /dev/null
+++ b/drivers/net/wireless/virtual/virt_wifi.c
@@ -0,0 +1,697 @@
+// SPDX-License-Identifier: GPL-2.0
+/* drivers/net/wireless/virt_wifi.c
+ *
+ * A fake implementation of cfg80211_ops that can be tacked on to an ethernet
+ * net_device to make it appear as a wireless connection.
+ *
+ * Copyright (C) 2018 Google, Inc.
+ *
+ * Author: schuffelen@google.com
+ */
+
+#include <net/cfg80211.h>
+#include <net/rtnetlink.h>
+#include <linux/etherdevice.h>
+#include <linux/math64.h>
+#include <linux/module.h>
+
+static struct wiphy *common_wiphy;
+
+struct virt_wifi_wiphy_priv {
+	struct delayed_work scan_result;
+	struct cfg80211_scan_request *scan_request;
+	bool being_deleted;
+};
+
+static struct ieee80211_channel channel_2ghz = {
+	.band = NL80211_BAND_2GHZ,
+	.center_freq = 2432,
+	.hw_value = 2432,
+	.max_power = 20,
+};
+
+static struct ieee80211_rate bitrates_2ghz[] = {
+	{ .bitrate = 10 },
+	{ .bitrate = 20 },
+	{ .bitrate = 55 },
+	{ .bitrate = 110 },
+	{ .bitrate = 60 },
+	{ .bitrate = 120 },
+	{ .bitrate = 240 },
+};
+
+static struct ieee80211_supported_band band_2ghz = {
+	.channels = &channel_2ghz,
+	.bitrates = bitrates_2ghz,
+	.band = NL80211_BAND_2GHZ,
+	.n_channels = 1,
+	.n_bitrates = ARRAY_SIZE(bitrates_2ghz),
+	.ht_cap = {
+		.ht_supported = true,
+		.cap = IEEE80211_HT_CAP_SUP_WIDTH_20_40 |
+		       IEEE80211_HT_CAP_GRN_FLD |
+		       IEEE80211_HT_CAP_SGI_20 |
+		       IEEE80211_HT_CAP_SGI_40 |
+		       IEEE80211_HT_CAP_DSSSCCK40,
+		.ampdu_factor = 0x3,
+		.ampdu_density = 0x6,
+		.mcs = {
+			.rx_mask = {0xff, 0xff},
+			.tx_params = IEEE80211_HT_MCS_TX_DEFINED,
+		},
+	},
+};
+
+static struct ieee80211_channel channel_5ghz = {
+	.band = NL80211_BAND_5GHZ,
+	.center_freq = 5240,
+	.hw_value = 5240,
+	.max_power = 20,
+};
+
+static struct ieee80211_rate bitrates_5ghz[] = {
+	{ .bitrate = 60 },
+	{ .bitrate = 120 },
+	{ .bitrate = 240 },
+};
+
+#define RX_MCS_MAP (IEEE80211_VHT_MCS_SUPPORT_0_9 << 0 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 2 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 4 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 6 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 8 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 10 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 12 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 14)
+
+#define TX_MCS_MAP (IEEE80211_VHT_MCS_SUPPORT_0_9 << 0 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 2 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 4 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 6 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 8 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 10 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 12 | \
+		    IEEE80211_VHT_MCS_SUPPORT_0_9 << 14)
+
+static struct ieee80211_supported_band band_5ghz = {
+	.channels = &channel_5ghz,
+	.bitrates = bitrates_5ghz,
+	.band = NL80211_BAND_5GHZ,
+	.n_channels = 1,
+	.n_bitrates = ARRAY_SIZE(bitrates_5ghz),
+	.ht_cap = {
+		.ht_supported = true,
+		.cap = IEEE80211_HT_CAP_SUP_WIDTH_20_40 |
+		       IEEE80211_HT_CAP_GRN_FLD |
+		       IEEE80211_HT_CAP_SGI_20 |
+		       IEEE80211_HT_CAP_SGI_40 |
+		       IEEE80211_HT_CAP_DSSSCCK40,
+		.ampdu_factor = 0x3,
+		.ampdu_density = 0x6,
+		.mcs = {
+			.rx_mask = {0xff, 0xff},
+			.tx_params = IEEE80211_HT_MCS_TX_DEFINED,
+		},
+	},
+	.vht_cap = {
+		.vht_supported = true,
+		.cap = IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454 |
+		       IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ |
+		       IEEE80211_VHT_CAP_RXLDPC |
+		       IEEE80211_VHT_CAP_SHORT_GI_80 |
+		       IEEE80211_VHT_CAP_SHORT_GI_160 |
+		       IEEE80211_VHT_CAP_TXSTBC |
+		       IEEE80211_VHT_CAP_RXSTBC_1 |
+		       IEEE80211_VHT_CAP_RXSTBC_2 |
+		       IEEE80211_VHT_CAP_RXSTBC_3 |
+		       IEEE80211_VHT_CAP_RXSTBC_4 |
+		       IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK,
+		.vht_mcs = {
+			.rx_mcs_map = cpu_to_le16(RX_MCS_MAP),
+			.tx_mcs_map = cpu_to_le16(TX_MCS_MAP),
+		}
+	},
+};
+
+/* Assigned at module init. Guaranteed locally-administered and unicast. */
+static u8 fake_router_bssid[ETH_ALEN] __ro_after_init = {};
+
+static void virt_wifi_inform_bss(struct wiphy *wiphy)
+{
+	u64 tsf = div_u64(ktime_get_boottime_ns(), 1000);
+	struct cfg80211_bss *informed_bss;
+	static const struct {
+		u8 tag;
+		u8 len;
+		u8 ssid[8];
+	} __packed ssid = {
+		.tag = WLAN_EID_SSID,
+		.len = 8,
+		.ssid = "VirtWifi",
+	};
+
+	informed_bss = cfg80211_inform_bss(wiphy, &channel_5ghz,
+					   CFG80211_BSS_FTYPE_PRESP,
+					   fake_router_bssid, tsf,
+					   WLAN_CAPABILITY_ESS, 0,
+					   (void *)&ssid, sizeof(ssid),
+					   DBM_TO_MBM(-50), GFP_KERNEL);
+	cfg80211_put_bss(wiphy, informed_bss);
+}
+
+/* Called with the rtnl lock held. */
+static int virt_wifi_scan(struct wiphy *wiphy,
+			  struct cfg80211_scan_request *request)
+{
+	struct virt_wifi_wiphy_priv *priv = wiphy_priv(wiphy);
+
+	wiphy_debug(wiphy, "scan\n");
+
+	if (priv->scan_request || priv->being_deleted)
+		return -EBUSY;
+
+	priv->scan_request = request;
+	schedule_delayed_work(&priv->scan_result, HZ * 2);
+
+	return 0;
+}
+
+/* Acquires and releases the rdev BSS lock. */
+static void virt_wifi_scan_result(struct work_struct *work)
+{
+	struct virt_wifi_wiphy_priv *priv =
+		container_of(work, struct virt_wifi_wiphy_priv,
+			     scan_result.work);
+	struct wiphy *wiphy = priv_to_wiphy(priv);
+	struct cfg80211_scan_info scan_info = { .aborted = false };
+
+	virt_wifi_inform_bss(wiphy);
+
+	/* Schedules work which acquires and releases the rtnl lock. */
+	cfg80211_scan_done(priv->scan_request, &scan_info);
+	priv->scan_request = NULL;
+}
+
+/* May acquire and release the rdev BSS lock. */
+static void virt_wifi_cancel_scan(struct wiphy *wiphy)
+{
+	struct virt_wifi_wiphy_priv *priv = wiphy_priv(wiphy);
+
+	cancel_delayed_work_sync(&priv->scan_result);
+	/* Clean up dangling callbacks if necessary. */
+	if (priv->scan_request) {
+		struct cfg80211_scan_info scan_info = { .aborted = true };
+		/* Schedules work which acquires and releases the rtnl lock. */
+		cfg80211_scan_done(priv->scan_request, &scan_info);
+		priv->scan_request = NULL;
+	}
+}
+
+struct virt_wifi_netdev_priv {
+	struct delayed_work connect;
+	struct net_device *lowerdev;
+	struct net_device *upperdev;
+	u32 tx_packets;
+	u32 tx_failed;
+	u8 connect_requested_bss[ETH_ALEN];
+	bool is_up;
+	bool is_connected;
+	bool being_deleted;
+};
+
+/* Called with the rtnl lock held. */
+static int virt_wifi_connect(struct wiphy *wiphy, struct net_device *netdev,
+			     struct cfg80211_connect_params *sme)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(netdev);
+	bool could_schedule;
+
+	if (priv->being_deleted || !priv->is_up)
+		return -EBUSY;
+
+	could_schedule = schedule_delayed_work(&priv->connect, HZ * 2);
+	if (!could_schedule)
+		return -EBUSY;
+
+	if (sme->bssid) {
+		ether_addr_copy(priv->connect_requested_bss, sme->bssid);
+	} else {
+		virt_wifi_inform_bss(wiphy);
+		eth_zero_addr(priv->connect_requested_bss);
+	}
+
+	wiphy_debug(wiphy, "connect\n");
+
+	return 0;
+}
+
+/* Acquires and releases the rdev event lock. */
+static void virt_wifi_connect_complete(struct work_struct *work)
+{
+	struct virt_wifi_netdev_priv *priv =
+		container_of(work, struct virt_wifi_netdev_priv, connect.work);
+	u8 *requested_bss = priv->connect_requested_bss;
+	bool right_addr = ether_addr_equal(requested_bss, fake_router_bssid);
+	u16 status = WLAN_STATUS_SUCCESS;
+
+	if (is_zero_ether_addr(requested_bss))
+		requested_bss = NULL;
+
+	if (!priv->is_up || (requested_bss && !right_addr))
+		status = WLAN_STATUS_UNSPECIFIED_FAILURE;
+	else
+		priv->is_connected = true;
+
+	/* Schedules an event that acquires the rtnl lock. */
+	cfg80211_connect_result(priv->upperdev, requested_bss, NULL, 0, NULL, 0,
+				status, GFP_KERNEL);
+	netif_carrier_on(priv->upperdev);
+}
+
+/* May acquire and release the rdev event lock. */
+static void virt_wifi_cancel_connect(struct net_device *netdev)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(netdev);
+
+	/* If there is work pending, clean up dangling callbacks. */
+	if (cancel_delayed_work_sync(&priv->connect)) {
+		/* Schedules an event that acquires the rtnl lock. */
+		cfg80211_connect_result(priv->upperdev,
+					priv->connect_requested_bss, NULL, 0,
+					NULL, 0,
+					WLAN_STATUS_UNSPECIFIED_FAILURE,
+					GFP_KERNEL);
+	}
+}
+
+/* Called with the rtnl lock held. Acquires the rdev event lock. */
+static int virt_wifi_disconnect(struct wiphy *wiphy, struct net_device *netdev,
+				u16 reason_code)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(netdev);
+
+	if (priv->being_deleted)
+		return -EBUSY;
+
+	wiphy_debug(wiphy, "disconnect\n");
+	virt_wifi_cancel_connect(netdev);
+
+	cfg80211_disconnected(netdev, reason_code, NULL, 0, true, GFP_KERNEL);
+	priv->is_connected = false;
+	netif_carrier_off(netdev);
+
+	return 0;
+}
+
+/* Called with the rtnl lock held. */
+static int virt_wifi_get_station(struct wiphy *wiphy, struct net_device *dev,
+				 const u8 *mac, struct station_info *sinfo)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+
+	wiphy_debug(wiphy, "get_station\n");
+
+	if (!priv->is_connected || !ether_addr_equal(mac, fake_router_bssid))
+		return -ENOENT;
+
+	sinfo->filled = BIT_ULL(NL80211_STA_INFO_TX_PACKETS) |
+		BIT_ULL(NL80211_STA_INFO_TX_FAILED) |
+		BIT_ULL(NL80211_STA_INFO_SIGNAL) |
+		BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
+	sinfo->tx_packets = priv->tx_packets;
+	sinfo->tx_failed = priv->tx_failed;
+	/* For CFG80211_SIGNAL_TYPE_MBM, value is expressed in _dBm_ */
+	sinfo->signal = -50;
+	sinfo->txrate = (struct rate_info) {
+		.legacy = 10, /* units are 100kbit/s */
+	};
+	return 0;
+}
+
+/* Called with the rtnl lock held. */
+static int virt_wifi_dump_station(struct wiphy *wiphy, struct net_device *dev,
+				  int idx, u8 *mac, struct station_info *sinfo)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+
+	wiphy_debug(wiphy, "dump_station\n");
+
+	if (idx != 0 || !priv->is_connected)
+		return -ENOENT;
+
+	ether_addr_copy(mac, fake_router_bssid);
+	return virt_wifi_get_station(wiphy, dev, fake_router_bssid, sinfo);
+}
+
+static const struct cfg80211_ops virt_wifi_cfg80211_ops = {
+	.scan = virt_wifi_scan,
+
+	.connect = virt_wifi_connect,
+	.disconnect = virt_wifi_disconnect,
+
+	.get_station = virt_wifi_get_station,
+	.dump_station = virt_wifi_dump_station,
+};
+
+/* Acquires and releases the rtnl lock. */
+static struct wiphy *virt_wifi_make_wiphy(void)
+{
+	struct wiphy *wiphy;
+	struct virt_wifi_wiphy_priv *priv;
+	int err;
+
+	wiphy = wiphy_new(&virt_wifi_cfg80211_ops, sizeof(*priv));
+
+	if (!wiphy)
+		return NULL;
+
+	wiphy->max_scan_ssids = 4;
+	wiphy->max_scan_ie_len = 1000;
+	wiphy->signal_type = CFG80211_SIGNAL_TYPE_MBM;
+
+	wiphy->bands[NL80211_BAND_2GHZ] = &band_2ghz;
+	wiphy->bands[NL80211_BAND_5GHZ] = &band_5ghz;
+	wiphy->bands[NL80211_BAND_60GHZ] = NULL;
+
+	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION);
+
+	priv = wiphy_priv(wiphy);
+	priv->being_deleted = false;
+	priv->scan_request = NULL;
+	INIT_DELAYED_WORK(&priv->scan_result, virt_wifi_scan_result);
+
+	err = wiphy_register(wiphy);
+	if (err < 0) {
+		wiphy_free(wiphy);
+		return NULL;
+	}
+
+	return wiphy;
+}
+
+/* Acquires and releases the rtnl lock. */
+static void virt_wifi_destroy_wiphy(struct wiphy *wiphy)
+{
+	struct virt_wifi_wiphy_priv *priv;
+
+	WARN(!wiphy, "%s called with null wiphy", __func__);
+	if (!wiphy)
+		return;
+
+	priv = wiphy_priv(wiphy);
+	priv->being_deleted = true;
+	virt_wifi_cancel_scan(wiphy);
+
+	if (wiphy->registered)
+		wiphy_unregister(wiphy);
+	wiphy_free(wiphy);
+}
+
+/* Enters and exits a RCU-bh critical section. */
+static netdev_tx_t virt_wifi_start_xmit(struct sk_buff *skb,
+					struct net_device *dev)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+
+	priv->tx_packets++;
+	if (!priv->is_connected) {
+		priv->tx_failed++;
+		return NET_XMIT_DROP;
+	}
+
+	skb->dev = priv->lowerdev;
+	return dev_queue_xmit(skb);
+}
+
+/* Called with rtnl lock held. */
+static int virt_wifi_net_device_open(struct net_device *dev)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+
+	priv->is_up = true;
+	return 0;
+}
+
+/* Called with rtnl lock held. */
+static int virt_wifi_net_device_stop(struct net_device *dev)
+{
+	struct virt_wifi_netdev_priv *n_priv = netdev_priv(dev);
+
+	n_priv->is_up = false;
+
+	if (!dev->ieee80211_ptr)
+		return 0;
+
+	virt_wifi_cancel_scan(dev->ieee80211_ptr->wiphy);
+	virt_wifi_cancel_connect(dev);
+	netif_carrier_off(dev);
+
+	return 0;
+}
+
+static int virt_wifi_net_device_get_iflink(const struct net_device *dev)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+
+	return priv->lowerdev->ifindex;
+}
+
+static const struct net_device_ops virt_wifi_ops = {
+	.ndo_start_xmit = virt_wifi_start_xmit,
+	.ndo_open	= virt_wifi_net_device_open,
+	.ndo_stop	= virt_wifi_net_device_stop,
+	.ndo_get_iflink = virt_wifi_net_device_get_iflink,
+};
+
+/* Invoked as part of rtnl lock release. */
+static void virt_wifi_net_device_destructor(struct net_device *dev)
+{
+	/* Delayed past dellink to allow nl80211 to react to the device being
+	 * deleted.
+	 */
+	kfree(dev->ieee80211_ptr);
+	dev->ieee80211_ptr = NULL;
+}
+
+/* No lock interaction. */
+static void virt_wifi_setup(struct net_device *dev)
+{
+	ether_setup(dev);
+	dev->netdev_ops = &virt_wifi_ops;
+	dev->needs_free_netdev  = true;
+}
+
+/* Called in a RCU read critical section from netif_receive_skb */
+static rx_handler_result_t virt_wifi_rx_handler(struct sk_buff **pskb)
+{
+	struct sk_buff *skb = *pskb;
+	struct virt_wifi_netdev_priv *priv =
+		rcu_dereference(skb->dev->rx_handler_data);
+
+	if (!priv->is_connected)
+		return RX_HANDLER_PASS;
+
+	/* GFP_ATOMIC because this is a packet interrupt handler. */
+	skb = skb_share_check(skb, GFP_ATOMIC);
+	if (!skb) {
+		dev_err(&priv->upperdev->dev, "can't skb_share_check\n");
+		return RX_HANDLER_CONSUMED;
+	}
+
+	*pskb = skb;
+	skb->dev = priv->upperdev;
+	skb->pkt_type = PACKET_HOST;
+	return RX_HANDLER_ANOTHER;
+}
+
+/* Called with rtnl lock held. */
+static int virt_wifi_newlink(struct net *src_net, struct net_device *dev,
+			     struct nlattr *tb[], struct nlattr *data[],
+			     struct netlink_ext_ack *extack)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+	int err;
+
+	if (!tb[IFLA_LINK])
+		return -EINVAL;
+
+	netif_carrier_off(dev);
+
+	priv->upperdev = dev;
+	priv->lowerdev = __dev_get_by_index(src_net,
+					    nla_get_u32(tb[IFLA_LINK]));
+
+	if (!priv->lowerdev)
+		return -ENODEV;
+	if (!tb[IFLA_MTU])
+		dev->mtu = priv->lowerdev->mtu;
+	else if (dev->mtu > priv->lowerdev->mtu)
+		return -EINVAL;
+
+	err = netdev_rx_handler_register(priv->lowerdev, virt_wifi_rx_handler,
+					 priv);
+	if (err) {
+		dev_err(&priv->lowerdev->dev,
+			"can't netdev_rx_handler_register: %d\n", err);
+		return err;
+	}
+
+	eth_hw_addr_inherit(dev, priv->lowerdev);
+	netif_stacked_transfer_operstate(priv->lowerdev, dev);
+
+	SET_NETDEV_DEV(dev, &priv->lowerdev->dev);
+	dev->ieee80211_ptr = kzalloc(sizeof(*dev->ieee80211_ptr), GFP_KERNEL);
+
+	if (!dev->ieee80211_ptr) {
+		err = -ENOMEM;
+		goto remove_handler;
+	}
+
+	dev->ieee80211_ptr->iftype = NL80211_IFTYPE_STATION;
+	dev->ieee80211_ptr->wiphy = common_wiphy;
+
+	err = register_netdevice(dev);
+	if (err) {
+		dev_err(&priv->lowerdev->dev, "can't register_netdevice: %d\n",
+			err);
+		goto free_wireless_dev;
+	}
+
+	err = netdev_upper_dev_link(priv->lowerdev, dev, extack);
+	if (err) {
+		dev_err(&priv->lowerdev->dev, "can't netdev_upper_dev_link: %d\n",
+			err);
+		goto unregister_netdev;
+	}
+
+	dev->priv_destructor = virt_wifi_net_device_destructor;
+	priv->being_deleted = false;
+	priv->is_connected = false;
+	priv->is_up = false;
+	INIT_DELAYED_WORK(&priv->connect, virt_wifi_connect_complete);
+	__module_get(THIS_MODULE);
+
+	return 0;
+unregister_netdev:
+	unregister_netdevice(dev);
+free_wireless_dev:
+	kfree(dev->ieee80211_ptr);
+	dev->ieee80211_ptr = NULL;
+remove_handler:
+	netdev_rx_handler_unregister(priv->lowerdev);
+
+	return err;
+}
+
+/* Called with rtnl lock held. */
+static void virt_wifi_dellink(struct net_device *dev,
+			      struct list_head *head)
+{
+	struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
+
+	if (dev->ieee80211_ptr)
+		virt_wifi_cancel_scan(dev->ieee80211_ptr->wiphy);
+
+	priv->being_deleted = true;
+	virt_wifi_cancel_connect(dev);
+	netif_carrier_off(dev);
+
+	netdev_rx_handler_unregister(priv->lowerdev);
+	netdev_upper_dev_unlink(priv->lowerdev, dev);
+
+	unregister_netdevice_queue(dev, head);
+	module_put(THIS_MODULE);
+
+	/* Deleting the wiphy is handled in the module destructor. */
+}
+
+static struct rtnl_link_ops virt_wifi_link_ops = {
+	.kind		= "virt_wifi",
+	.setup		= virt_wifi_setup,
+	.newlink	= virt_wifi_newlink,
+	.dellink	= virt_wifi_dellink,
+	.priv_size	= sizeof(struct virt_wifi_netdev_priv),
+};
+
+static bool netif_is_virt_wifi_dev(const struct net_device *dev)
+{
+	return rcu_access_pointer(dev->rx_handler) == virt_wifi_rx_handler;
+}
+
+static int virt_wifi_event(struct notifier_block *this, unsigned long event,
+			   void *ptr)
+{
+	struct net_device *lower_dev = netdev_notifier_info_to_dev(ptr);
+	struct virt_wifi_netdev_priv *priv;
+	struct net_device *upper_dev;
+	LIST_HEAD(list_kill);
+
+	if (!netif_is_virt_wifi_dev(lower_dev))
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_UNREGISTER:
+		priv = rtnl_dereference(lower_dev->rx_handler_data);
+		if (!priv)
+			return NOTIFY_DONE;
+
+		upper_dev = priv->upperdev;
+
+		upper_dev->rtnl_link_ops->dellink(upper_dev, &list_kill);
+		unregister_netdevice_many(&list_kill);
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block virt_wifi_notifier = {
+	.notifier_call = virt_wifi_event,
+};
+
+/* Acquires and releases the rtnl lock. */
+static int __init virt_wifi_init_module(void)
+{
+	int err;
+
+	/* Guaranteed to be locally-administered and not multicast. */
+	eth_random_addr(fake_router_bssid);
+
+	err = register_netdevice_notifier(&virt_wifi_notifier);
+	if (err)
+		return err;
+
+	err = -ENOMEM;
+	common_wiphy = virt_wifi_make_wiphy();
+	if (!common_wiphy)
+		goto notifier;
+
+	err = rtnl_link_register(&virt_wifi_link_ops);
+	if (err)
+		goto destroy_wiphy;
+
+	return 0;
+
+destroy_wiphy:
+	virt_wifi_destroy_wiphy(common_wiphy);
+notifier:
+	unregister_netdevice_notifier(&virt_wifi_notifier);
+	return err;
+}
+
+/* Acquires and releases the rtnl lock. */
+static void __exit virt_wifi_cleanup_module(void)
+{
+	/* Will delete any devices that depend on the wiphy. */
+	rtnl_link_unregister(&virt_wifi_link_ops);
+	virt_wifi_destroy_wiphy(common_wiphy);
+	unregister_netdevice_notifier(&virt_wifi_notifier);
+}
+
+module_init(virt_wifi_init_module);
+module_exit(virt_wifi_cleanup_module);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Cody Schuffelen <schuffelen@google.com>");
+MODULE_DESCRIPTION("Driver for a wireless wrapper of ethernet devices");
+MODULE_ALIAS_RTNL_LINK("virt_wifi");
-- 
cgit v1.2.3-70-g09d2


From 298e50ad8eb8fa12ea68bb2da45bb8ef4edcd0ec Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@kernel.org>
Date: Mon, 27 Feb 2023 14:17:32 +0200
Subject: wifi: move raycs, wl3501 and rndis_wlan to legacy directory

To clean up drivers/net/wireless move the old drivers drivers left in the
directory to a new "legacy" directory. I did consider adding
CONFIG_WLAN_VENDOR_LEGACY like other vendors have but then dropped the idea as
these are really old drivers and hopefully we get to remove them soon.

There should be no changes in compilation or in Kconfig options, merely moving files.

Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230227121732.8967-3-kvalo@kernel.org
---
 drivers/net/wireless/Kconfig             |   55 +-
 drivers/net/wireless/Makefile            |    7 +-
 drivers/net/wireless/legacy/Kconfig      |   55 +
 drivers/net/wireless/legacy/Makefile     |    6 +
 drivers/net/wireless/legacy/ray_cs.c     | 2819 ++++++++++++++++++++++
 drivers/net/wireless/legacy/ray_cs.h     |   74 +
 drivers/net/wireless/legacy/rayctl.h     |  734 ++++++
 drivers/net/wireless/legacy/rndis_wlan.c | 3760 ++++++++++++++++++++++++++++++
 drivers/net/wireless/legacy/wl3501.h     |  615 +++++
 drivers/net/wireless/legacy/wl3501_cs.c  | 2030 ++++++++++++++++
 drivers/net/wireless/ray_cs.c            | 2819 ----------------------
 drivers/net/wireless/ray_cs.h            |   74 -
 drivers/net/wireless/rayctl.h            |  734 ------
 drivers/net/wireless/rndis_wlan.c        | 3760 ------------------------------
 drivers/net/wireless/wl3501.h            |  615 -----
 drivers/net/wireless/wl3501_cs.c         | 2030 ----------------
 16 files changed, 10095 insertions(+), 10092 deletions(-)
 create mode 100644 drivers/net/wireless/legacy/Kconfig
 create mode 100644 drivers/net/wireless/legacy/Makefile
 create mode 100644 drivers/net/wireless/legacy/ray_cs.c
 create mode 100644 drivers/net/wireless/legacy/ray_cs.h
 create mode 100644 drivers/net/wireless/legacy/rayctl.h
 create mode 100644 drivers/net/wireless/legacy/rndis_wlan.c
 create mode 100644 drivers/net/wireless/legacy/wl3501.h
 create mode 100644 drivers/net/wireless/legacy/wl3501_cs.c
 delete mode 100644 drivers/net/wireless/ray_cs.c
 delete mode 100644 drivers/net/wireless/ray_cs.h
 delete mode 100644 drivers/net/wireless/rayctl.h
 delete mode 100644 drivers/net/wireless/rndis_wlan.c
 delete mode 100644 drivers/net/wireless/wl3501.h
 delete mode 100644 drivers/net/wireless/wl3501_cs.c

diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig
index 42b40cc96b21..7555af5195ec 100644
--- a/drivers/net/wireless/Kconfig
+++ b/drivers/net/wireless/Kconfig
@@ -38,60 +38,7 @@ source "drivers/net/wireless/ti/Kconfig"
 source "drivers/net/wireless/zydas/Kconfig"
 source "drivers/net/wireless/quantenna/Kconfig"
 
-config PCMCIA_RAYCS
-	tristate "Aviator/Raytheon 2.4GHz wireless support"
-	depends on PCMCIA
-	select WIRELESS_EXT
-	select WEXT_SPY
-	select WEXT_PRIV
-	help
-	  Say Y here if you intend to attach an Aviator/Raytheon PCMCIA
-	  (PC-card) wireless Ethernet networking card to your computer.
-	  Please read the file
-	  <file:Documentation/networking/device_drivers/wifi/ray_cs.rst> for
-	  details.
-
-	  To compile this driver as a module, choose M here: the module will be
-	  called ray_cs.  If unsure, say N.
-
-config PCMCIA_WL3501
-	tristate "Planet WL3501 PCMCIA cards"
-	depends on CFG80211 && PCMCIA
-	select WIRELESS_EXT
-	select WEXT_SPY
-	help
-	  A driver for WL3501 PCMCIA 802.11 wireless cards made by Planet.
-	  It has basic support for Linux wireless extensions and initial
-	  micro support for ethtool.
-
-config USB_NET_RNDIS_WLAN
-	tristate "Wireless RNDIS USB support"
-	depends on USB
-	depends on CFG80211
-	select USB_NET_DRIVERS
-	select USB_USBNET
-	select USB_NET_CDCETHER
-	select USB_NET_RNDIS_HOST
-	help
-	  This is a driver for wireless RNDIS devices.
-	  These are USB based adapters found in devices such as:
-
-	  Buffalo WLI-U2-KG125S
-	  U.S. Robotics USR5421
-	  Belkin F5D7051
-	  Linksys WUSB54GSv2
-	  Linksys WUSB54GSC
-	  Asus WL169gE
-	  Eminent EM4045
-	  BT Voyager 1055
-	  Linksys WUSB54GSv1
-	  U.S. Robotics USR5420
-	  BUFFALO WLI-USB-G54
-
-	  All of these devices are based on Broadcom 4320 chip which is the
-	  only wireless RNDIS chip known to date.
-
-	  If you choose to build a module, it'll be called rndis_wlan.
+source "drivers/net/wireless/legacy/Kconfig"
 
 source "drivers/net/wireless/virtual/Kconfig"
 
diff --git a/drivers/net/wireless/Makefile b/drivers/net/wireless/Makefile
index 1b697cfe0a13..4d7374d567d1 100644
--- a/drivers/net/wireless/Makefile
+++ b/drivers/net/wireless/Makefile
@@ -23,10 +23,5 @@ obj-$(CONFIG_WLAN_VENDOR_ST) += st/
 obj-$(CONFIG_WLAN_VENDOR_TI) += ti/
 obj-$(CONFIG_WLAN_VENDOR_ZYDAS) += zydas/
 
-# 16-bit wireless PCMCIA client drivers
-obj-$(CONFIG_PCMCIA_RAYCS)	+= ray_cs.o
-obj-$(CONFIG_PCMCIA_WL3501)	+= wl3501_cs.o
-
-obj-$(CONFIG_USB_NET_RNDIS_WLAN)	+= rndis_wlan.o
-
+obj-$(CONFIG_WLAN) += legacy/
 obj-$(CONFIG_WLAN) += virtual/
diff --git a/drivers/net/wireless/legacy/Kconfig b/drivers/net/wireless/legacy/Kconfig
new file mode 100644
index 000000000000..3a5275941212
--- /dev/null
+++ b/drivers/net/wireless/legacy/Kconfig
@@ -0,0 +1,55 @@
+config PCMCIA_RAYCS
+	tristate "Aviator/Raytheon 2.4GHz wireless support"
+	depends on PCMCIA
+	select WIRELESS_EXT
+	select WEXT_SPY
+	select WEXT_PRIV
+	help
+	  Say Y here if you intend to attach an Aviator/Raytheon PCMCIA
+	  (PC-card) wireless Ethernet networking card to your computer.
+	  Please read the file
+	  <file:Documentation/networking/device_drivers/wifi/ray_cs.rst> for
+	  details.
+
+	  To compile this driver as a module, choose M here: the module will be
+	  called ray_cs.  If unsure, say N.
+
+config PCMCIA_WL3501
+	tristate "Planet WL3501 PCMCIA cards"
+	depends on CFG80211 && PCMCIA
+	select WIRELESS_EXT
+	select WEXT_SPY
+	help
+	  A driver for WL3501 PCMCIA 802.11 wireless cards made by Planet.
+	  It has basic support for Linux wireless extensions and initial
+	  micro support for ethtool.
+
+config USB_NET_RNDIS_WLAN
+	tristate "Wireless RNDIS USB support"
+	depends on USB
+	depends on CFG80211
+	select USB_NET_DRIVERS
+	select USB_USBNET
+	select USB_NET_CDCETHER
+	select USB_NET_RNDIS_HOST
+	help
+	  This is a driver for wireless RNDIS devices.
+	  These are USB based adapters found in devices such as:
+
+	  Buffalo WLI-U2-KG125S
+	  U.S. Robotics USR5421
+	  Belkin F5D7051
+	  Linksys WUSB54GSv2
+	  Linksys WUSB54GSC
+	  Asus WL169gE
+	  Eminent EM4045
+	  BT Voyager 1055
+	  Linksys WUSB54GSv1
+	  U.S. Robotics USR5420
+	  BUFFALO WLI-USB-G54
+
+	  All of these devices are based on Broadcom 4320 chip which is the
+	  only wireless RNDIS chip known to date.
+
+	  If you choose to build a module, it'll be called rndis_wlan.
+
diff --git a/drivers/net/wireless/legacy/Makefile b/drivers/net/wireless/legacy/Makefile
new file mode 100644
index 000000000000..36878f080bfc
--- /dev/null
+++ b/drivers/net/wireless/legacy/Makefile
@@ -0,0 +1,6 @@
+# 16-bit wireless PCMCIA client drivers
+obj-$(CONFIG_PCMCIA_RAYCS)	+= ray_cs.o
+obj-$(CONFIG_PCMCIA_WL3501)	+= wl3501_cs.o
+
+obj-$(CONFIG_USB_NET_RNDIS_WLAN)	+= rndis_wlan.o
+
diff --git a/drivers/net/wireless/legacy/ray_cs.c b/drivers/net/wireless/legacy/ray_cs.c
new file mode 100644
index 000000000000..1f57a0055bbd
--- /dev/null
+++ b/drivers/net/wireless/legacy/ray_cs.c
@@ -0,0 +1,2819 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*=============================================================================
+ *
+ * A  PCMCIA client driver for the Raylink wireless LAN card.
+ * The starting point for this module was the skeleton.c in the
+ * PCMCIA 2.9.12 package written by David Hinds, dahinds@users.sourceforge.net
+ *
+ * Copyright (c) 1998  Corey Thomas (corey@world.std.com)
+ *
+ * Changes:
+ * Arnaldo Carvalho de Melo <acme@conectiva.com.br> - 08/08/2000
+ * - reorganize kmallocs in ray_attach, checking all for failure
+ *   and releasing the previous allocations if one fails
+ *
+ * Daniele Bellucci <bellucda@tiscali.it> - 07/10/2003
+ * - Audit copy_to_user in ioctl(SIOCGIWESSID)
+ *
+=============================================================================*/
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/proc_fs.h>
+#include <linux/ptrace.h>
+#include <linux/seq_file.h>
+#include <linux/string.h>
+#include <linux/timer.h>
+#include <linux/init.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/if_arp.h>
+#include <linux/ioport.h>
+#include <linux/skbuff.h>
+#include <linux/ieee80211.h>
+
+#include <pcmcia/cistpl.h>
+#include <pcmcia/cisreg.h>
+#include <pcmcia/ds.h>
+
+#include <linux/wireless.h>
+#include <net/iw_handler.h>
+
+#include <asm/io.h>
+#include <asm/byteorder.h>
+#include <linux/uaccess.h>
+
+/* Warning : these stuff will slow down the driver... */
+#define WIRELESS_SPY		/* Enable spying addresses */
+/* Definitions we need for spy */
+typedef struct iw_statistics iw_stats;
+typedef u_char mac_addr[ETH_ALEN];	/* Hardware address */
+
+#include "rayctl.h"
+#include "ray_cs.h"
+
+
+/** Prototypes based on PCMCIA skeleton driver *******************************/
+static int ray_config(struct pcmcia_device *link);
+static void ray_release(struct pcmcia_device *link);
+static void ray_detach(struct pcmcia_device *p_dev);
+
+/***** Prototypes indicated by device structure ******************************/
+static int ray_dev_close(struct net_device *dev);
+static int ray_dev_config(struct net_device *dev, struct ifmap *map);
+static struct net_device_stats *ray_get_stats(struct net_device *dev);
+static int ray_dev_init(struct net_device *dev);
+
+static int ray_open(struct net_device *dev);
+static netdev_tx_t ray_dev_start_xmit(struct sk_buff *skb,
+					    struct net_device *dev);
+static void set_multicast_list(struct net_device *dev);
+static void ray_update_multi_list(struct net_device *dev, int all);
+static int translate_frame(ray_dev_t *local, struct tx_msg __iomem *ptx,
+			   unsigned char *data, int len);
+static void ray_build_header(ray_dev_t *local, struct tx_msg __iomem *ptx,
+			     UCHAR msg_type, unsigned char *data);
+static void untranslate(ray_dev_t *local, struct sk_buff *skb, int len);
+static iw_stats *ray_get_wireless_stats(struct net_device *dev);
+static const struct iw_handler_def ray_handler_def;
+
+/***** Prototypes for raylink functions **************************************/
+static void authenticate(ray_dev_t *local);
+static int build_auth_frame(ray_dev_t *local, UCHAR *dest, int auth_type);
+static void authenticate_timeout(struct timer_list *t);
+static int get_free_ccs(ray_dev_t *local);
+static int get_free_tx_ccs(ray_dev_t *local);
+static void init_startup_params(ray_dev_t *local);
+static int parse_addr(char *in_str, UCHAR *out);
+static int ray_hw_xmit(unsigned char *data, int len, struct net_device *dev, UCHAR type);
+static int ray_init(struct net_device *dev);
+static int interrupt_ecf(ray_dev_t *local, int ccs);
+static void ray_reset(struct net_device *dev);
+static void ray_update_parm(struct net_device *dev, UCHAR objid, UCHAR *value, int len);
+static void verify_dl_startup(struct timer_list *t);
+
+/* Prototypes for interrpt time functions **********************************/
+static irqreturn_t ray_interrupt(int reg, void *dev_id);
+static void clear_interrupt(ray_dev_t *local);
+static void rx_deauthenticate(ray_dev_t *local, struct rcs __iomem *prcs,
+			      unsigned int pkt_addr, int rx_len);
+static int copy_from_rx_buff(ray_dev_t *local, UCHAR *dest, int pkt_addr, int len);
+static void ray_rx(struct net_device *dev, ray_dev_t *local, struct rcs __iomem *prcs);
+static void release_frag_chain(ray_dev_t *local, struct rcs __iomem *prcs);
+static void rx_authenticate(ray_dev_t *local, struct rcs __iomem *prcs,
+			    unsigned int pkt_addr, int rx_len);
+static void rx_data(struct net_device *dev, struct rcs __iomem *prcs,
+		    unsigned int pkt_addr, int rx_len);
+static void associate(ray_dev_t *local);
+
+/* Card command functions */
+static int dl_startup_params(struct net_device *dev);
+static void join_net(struct timer_list *t);
+static void start_net(struct timer_list *t);
+
+/*===========================================================================*/
+/* Parameters that can be set with 'insmod' */
+
+/* ADHOC=0, Infrastructure=1 */
+static int net_type = ADHOC;
+
+/* Hop dwell time in Kus (1024 us units defined by 802.11) */
+static int hop_dwell = 128;
+
+/* Beacon period in Kus */
+static int beacon_period = 256;
+
+/* power save mode (0 = off, 1 = save power) */
+static int psm;
+
+/* String for network's Extended Service Set ID. 32 Characters max */
+static char *essid;
+
+/* Default to encapsulation unless translation requested */
+static bool translate = true;
+
+static int country = USA;
+
+static int sniffer;
+
+static int bc;
+
+/* 48 bit physical card address if overriding card's real physical
+ * address is required.  Since IEEE 802.11 addresses are 48 bits
+ * like ethernet, an int can't be used, so a string is used. To
+ * allow use of addresses starting with a decimal digit, the first
+ * character must be a letter and will be ignored. This letter is
+ * followed by up to 12 hex digits which are the address.  If less
+ * than 12 digits are used, the address will be left filled with 0's.
+ * Note that bit 0 of the first byte is the broadcast bit, and evil
+ * things will happen if it is not 0 in a card address.
+ */
+static char *phy_addr = NULL;
+
+static unsigned int ray_mem_speed = 500;
+
+/* WARNING: THIS DRIVER IS NOT CAPABLE OF HANDLING MULTIPLE DEVICES! */
+static struct pcmcia_device *this_device = NULL;
+
+MODULE_AUTHOR("Corey Thomas <corey@world.std.com>");
+MODULE_DESCRIPTION("Raylink/WebGear wireless LAN driver");
+MODULE_LICENSE("GPL");
+
+module_param(net_type, int, 0);
+module_param(hop_dwell, int, 0);
+module_param(beacon_period, int, 0);
+module_param(psm, int, 0);
+module_param(essid, charp, 0);
+module_param(translate, bool, 0);
+module_param(country, int, 0);
+module_param(sniffer, int, 0);
+module_param(bc, int, 0);
+module_param(phy_addr, charp, 0);
+module_param(ray_mem_speed, int, 0);
+
+static const UCHAR b5_default_startup_parms[] = {
+	0, 0,			/* Adhoc station */
+	'L', 'I', 'N', 'U', 'X', 0, 0, 0,	/* 32 char ESSID */
+	0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0,
+	1, 0,			/* Active scan, CA Mode */
+	0, 0, 0, 0, 0, 0,	/* No default MAC addr  */
+	0x7f, 0xff,		/* Frag threshold */
+	0x00, 0x80,		/* Hop time 128 Kus */
+	0x01, 0x00,		/* Beacon period 256 Kus */
+	0x01, 0x07, 0xa3,	/* DTIM, retries, ack timeout */
+	0x1d, 0x82, 0x4e,	/* SIFS, DIFS, PIFS */
+	0x7f, 0xff,		/* RTS threshold */
+	0x04, 0xe2, 0x38, 0xA4,	/* scan_dwell, max_scan_dwell */
+	0x05,			/* assoc resp timeout thresh */
+	0x08, 0x02, 0x08,	/* adhoc, infra, super cycle max */
+	0,			/* Promiscuous mode */
+	0x0c, 0x0bd,		/* Unique word */
+	0x32,			/* Slot time */
+	0xff, 0xff,		/* roam-low snr, low snr count */
+	0x05, 0xff,		/* Infra, adhoc missed bcn thresh */
+	0x01, 0x0b, 0x4f,	/* USA, hop pattern, hop pat length */
+/* b4 - b5 differences start here */
+	0x00, 0x3f,		/* CW max */
+	0x00, 0x0f,		/* CW min */
+	0x04, 0x08,		/* Noise gain, limit offset */
+	0x28, 0x28,		/* det rssi, med busy offsets */
+	7,			/* det sync thresh */
+	0, 2, 2,		/* test mode, min, max */
+	0,			/* allow broadcast SSID probe resp */
+	0, 0,			/* privacy must start, can join */
+	2, 0, 0, 0, 0, 0, 0, 0	/* basic rate set */
+};
+
+static const UCHAR b4_default_startup_parms[] = {
+	0, 0,			/* Adhoc station */
+	'L', 'I', 'N', 'U', 'X', 0, 0, 0,	/* 32 char ESSID */
+	0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0,
+	1, 0,			/* Active scan, CA Mode */
+	0, 0, 0, 0, 0, 0,	/* No default MAC addr  */
+	0x7f, 0xff,		/* Frag threshold */
+	0x02, 0x00,		/* Hop time */
+	0x00, 0x01,		/* Beacon period */
+	0x01, 0x07, 0xa3,	/* DTIM, retries, ack timeout */
+	0x1d, 0x82, 0xce,	/* SIFS, DIFS, PIFS */
+	0x7f, 0xff,		/* RTS threshold */
+	0xfb, 0x1e, 0xc7, 0x5c,	/* scan_dwell, max_scan_dwell */
+	0x05,			/* assoc resp timeout thresh */
+	0x04, 0x02, 0x4,	/* adhoc, infra, super cycle max */
+	0,			/* Promiscuous mode */
+	0x0c, 0x0bd,		/* Unique word */
+	0x4e,			/* Slot time (TBD seems wrong) */
+	0xff, 0xff,		/* roam-low snr, low snr count */
+	0x05, 0xff,		/* Infra, adhoc missed bcn thresh */
+	0x01, 0x0b, 0x4e,	/* USA, hop pattern, hop pat length */
+/* b4 - b5 differences start here */
+	0x3f, 0x0f,		/* CW max, min */
+	0x04, 0x08,		/* Noise gain, limit offset */
+	0x28, 0x28,		/* det rssi, med busy offsets */
+	7,			/* det sync thresh */
+	0, 2, 2,		/* test mode, min, max */
+	0,			/* rx/tx delay */
+	0, 0, 0, 0, 0, 0,	/* current BSS id */
+	0			/* hop set */
+};
+
+/*===========================================================================*/
+static const u8 eth2_llc[] = { 0xaa, 0xaa, 3, 0, 0, 0 };
+
+static const char hop_pattern_length[] = { 1,
+	USA_HOP_MOD, EUROPE_HOP_MOD,
+	JAPAN_HOP_MOD, KOREA_HOP_MOD,
+	SPAIN_HOP_MOD, FRANCE_HOP_MOD,
+	ISRAEL_HOP_MOD, AUSTRALIA_HOP_MOD,
+	JAPAN_TEST_HOP_MOD
+};
+
+static const char rcsid[] =
+    "Raylink/WebGear wireless LAN - Corey <Thomas corey@world.std.com>";
+
+static const struct net_device_ops ray_netdev_ops = {
+	.ndo_init 		= ray_dev_init,
+	.ndo_open 		= ray_open,
+	.ndo_stop 		= ray_dev_close,
+	.ndo_start_xmit		= ray_dev_start_xmit,
+	.ndo_set_config		= ray_dev_config,
+	.ndo_get_stats		= ray_get_stats,
+	.ndo_set_rx_mode	= set_multicast_list,
+	.ndo_set_mac_address 	= eth_mac_addr,
+	.ndo_validate_addr	= eth_validate_addr,
+};
+
+static int ray_probe(struct pcmcia_device *p_dev)
+{
+	ray_dev_t *local;
+	struct net_device *dev;
+
+	dev_dbg(&p_dev->dev, "ray_attach()\n");
+
+	/* Allocate space for private device-specific data */
+	dev = alloc_etherdev(sizeof(ray_dev_t));
+	if (!dev)
+		goto fail_alloc_dev;
+
+	local = netdev_priv(dev);
+	local->finder = p_dev;
+
+	/* The io structure describes IO port mapping. None used here */
+	p_dev->resource[0]->end = 0;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+
+	/* General socket configuration */
+	p_dev->config_flags |= CONF_ENABLE_IRQ;
+	p_dev->config_index = 1;
+
+	p_dev->priv = dev;
+
+	local->finder = p_dev;
+	local->card_status = CARD_INSERTED;
+	local->authentication_state = UNAUTHENTICATED;
+	local->num_multi = 0;
+	dev_dbg(&p_dev->dev, "ray_attach p_dev = %p,  dev = %p,  local = %p, intr = %p\n",
+	      p_dev, dev, local, &ray_interrupt);
+
+	/* Raylink entries in the device structure */
+	dev->netdev_ops = &ray_netdev_ops;
+	dev->wireless_handlers = &ray_handler_def;
+#ifdef WIRELESS_SPY
+	local->wireless_data.spy_data = &local->spy_data;
+	dev->wireless_data = &local->wireless_data;
+#endif /* WIRELESS_SPY */
+
+
+	dev_dbg(&p_dev->dev, "ray_cs ray_attach calling ether_setup.)\n");
+	netif_stop_queue(dev);
+
+	timer_setup(&local->timer, NULL, 0);
+
+	this_device = p_dev;
+	return ray_config(p_dev);
+
+fail_alloc_dev:
+	return -ENOMEM;
+} /* ray_attach */
+
+static void ray_detach(struct pcmcia_device *link)
+{
+	struct net_device *dev;
+	ray_dev_t *local;
+
+	dev_dbg(&link->dev, "ray_detach\n");
+
+	this_device = NULL;
+	dev = link->priv;
+
+	ray_release(link);
+
+	local = netdev_priv(dev);
+	del_timer_sync(&local->timer);
+
+	if (link->priv) {
+		unregister_netdev(dev);
+		free_netdev(dev);
+	}
+	dev_dbg(&link->dev, "ray_cs ray_detach ending\n");
+} /* ray_detach */
+
+#define MAX_TUPLE_SIZE 128
+static int ray_config(struct pcmcia_device *link)
+{
+	int ret = 0;
+	int i;
+	struct net_device *dev = (struct net_device *)link->priv;
+	ray_dev_t *local = netdev_priv(dev);
+
+	dev_dbg(&link->dev, "ray_config\n");
+
+	/* Determine card type and firmware version */
+	printk(KERN_INFO "ray_cs Detected: %s%s%s%s\n",
+	       link->prod_id[0] ? link->prod_id[0] : " ",
+	       link->prod_id[1] ? link->prod_id[1] : " ",
+	       link->prod_id[2] ? link->prod_id[2] : " ",
+	       link->prod_id[3] ? link->prod_id[3] : " ");
+
+	/* Now allocate an interrupt line.  Note that this does not
+	   actually assign a handler to the interrupt.
+	 */
+	ret = pcmcia_request_irq(link, ray_interrupt);
+	if (ret)
+		goto failed;
+	dev->irq = link->irq;
+
+	ret = pcmcia_enable_device(link);
+	if (ret)
+		goto failed;
+
+/*** Set up 32k window for shared memory (transmit and control) ************/
+	link->resource[2]->flags |= WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
+	link->resource[2]->start = 0;
+	link->resource[2]->end = 0x8000;
+	ret = pcmcia_request_window(link, link->resource[2], ray_mem_speed);
+	if (ret)
+		goto failed;
+	ret = pcmcia_map_mem_page(link, link->resource[2], 0);
+	if (ret)
+		goto failed;
+	local->sram = ioremap(link->resource[2]->start,
+			resource_size(link->resource[2]));
+	if (!local->sram)
+		goto failed;
+
+/*** Set up 16k window for shared memory (receive buffer) ***************/
+	link->resource[3]->flags |=
+	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
+	link->resource[3]->start = 0;
+	link->resource[3]->end = 0x4000;
+	ret = pcmcia_request_window(link, link->resource[3], ray_mem_speed);
+	if (ret)
+		goto failed;
+	ret = pcmcia_map_mem_page(link, link->resource[3], 0x8000);
+	if (ret)
+		goto failed;
+	local->rmem = ioremap(link->resource[3]->start,
+			resource_size(link->resource[3]));
+	if (!local->rmem)
+		goto failed;
+
+/*** Set up window for attribute memory ***********************************/
+	link->resource[4]->flags |=
+	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_AM | WIN_ENABLE | WIN_USE_WAIT;
+	link->resource[4]->start = 0;
+	link->resource[4]->end = 0x1000;
+	ret = pcmcia_request_window(link, link->resource[4], ray_mem_speed);
+	if (ret)
+		goto failed;
+	ret = pcmcia_map_mem_page(link, link->resource[4], 0);
+	if (ret)
+		goto failed;
+	local->amem = ioremap(link->resource[4]->start,
+			resource_size(link->resource[4]));
+	if (!local->amem)
+		goto failed;
+
+	dev_dbg(&link->dev, "ray_config sram=%p\n", local->sram);
+	dev_dbg(&link->dev, "ray_config rmem=%p\n", local->rmem);
+	dev_dbg(&link->dev, "ray_config amem=%p\n", local->amem);
+	if (ray_init(dev) < 0) {
+		ray_release(link);
+		return -ENODEV;
+	}
+
+	SET_NETDEV_DEV(dev, &link->dev);
+	i = register_netdev(dev);
+	if (i != 0) {
+		printk("ray_config register_netdev() failed\n");
+		ray_release(link);
+		return i;
+	}
+
+	printk(KERN_INFO "%s: RayLink, irq %d, hw_addr %pM\n",
+	       dev->name, dev->irq, dev->dev_addr);
+
+	return 0;
+
+failed:
+	ray_release(link);
+	return -ENODEV;
+} /* ray_config */
+
+static inline struct ccs __iomem *ccs_base(ray_dev_t *dev)
+{
+	return dev->sram + CCS_BASE;
+}
+
+static inline struct rcs __iomem *rcs_base(ray_dev_t *dev)
+{
+	/*
+	 * This looks nonsensical, since there is a separate
+	 * RCS_BASE. But the difference between a "struct rcs"
+	 * and a "struct ccs" ends up being in the _index_ off
+	 * the base, so the base pointer is the same for both
+	 * ccs/rcs.
+	 */
+	return dev->sram + CCS_BASE;
+}
+
+/*===========================================================================*/
+static int ray_init(struct net_device *dev)
+{
+	int i;
+	struct ccs __iomem *pccs;
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	dev_dbg(&link->dev, "ray_init(0x%p)\n", dev);
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_init - device not present\n");
+		return -1;
+	}
+
+	local->net_type = net_type;
+	local->sta_type = TYPE_STA;
+
+	/* Copy the startup results to local memory */
+	memcpy_fromio(&local->startup_res, local->sram + ECF_TO_HOST_BASE,
+		      sizeof(struct startup_res_6));
+
+	/* Check Power up test status and get mac address from card */
+	if (local->startup_res.startup_word != 0x80) {
+		printk(KERN_INFO "ray_init ERROR card status = %2x\n",
+		       local->startup_res.startup_word);
+		local->card_status = CARD_INIT_ERROR;
+		return -1;
+	}
+
+	local->fw_ver = local->startup_res.firmware_version[0];
+	local->fw_bld = local->startup_res.firmware_version[1];
+	local->fw_var = local->startup_res.firmware_version[2];
+	dev_dbg(&link->dev, "ray_init firmware version %d.%d\n", local->fw_ver,
+	      local->fw_bld);
+
+	local->tib_length = 0x20;
+	if ((local->fw_ver == 5) && (local->fw_bld >= 30))
+		local->tib_length = local->startup_res.tib_length;
+	dev_dbg(&link->dev, "ray_init tib_length = 0x%02x\n", local->tib_length);
+	/* Initialize CCS's to buffer free state */
+	pccs = ccs_base(local);
+	for (i = 0; i < NUMBER_OF_CCS; i++) {
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+	}
+	init_startup_params(local);
+
+	/* copy mac address to startup parameters */
+	if (!parse_addr(phy_addr, local->sparm.b4.a_mac_addr)) {
+		memcpy(&local->sparm.b4.a_mac_addr,
+		       &local->startup_res.station_addr, ADDRLEN);
+	}
+
+	clear_interrupt(local);	/* Clear any interrupt from the card */
+	local->card_status = CARD_AWAITING_PARAM;
+	dev_dbg(&link->dev, "ray_init ending\n");
+	return 0;
+} /* ray_init */
+
+/*===========================================================================*/
+/* Download startup parameters to the card and command it to read them       */
+static int dl_startup_params(struct net_device *dev)
+{
+	int ccsindex;
+	ray_dev_t *local = netdev_priv(dev);
+	struct ccs __iomem *pccs;
+	struct pcmcia_device *link = local->finder;
+
+	dev_dbg(&link->dev, "dl_startup_params entered\n");
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs dl_startup_params - device not present\n");
+		return -1;
+	}
+
+	/* Copy parameters to host to ECF area */
+	if (local->fw_ver == 0x55)
+		memcpy_toio(local->sram + HOST_TO_ECF_BASE, &local->sparm.b4,
+			    sizeof(struct b4_startup_params));
+	else
+		memcpy_toio(local->sram + HOST_TO_ECF_BASE, &local->sparm.b5,
+			    sizeof(struct b5_startup_params));
+
+	/* Fill in the CCS fields for the ECF */
+	if ((ccsindex = get_free_ccs(local)) < 0)
+		return -1;
+	local->dl_param_ccs = ccsindex;
+	pccs = ccs_base(local) + ccsindex;
+	writeb(CCS_DOWNLOAD_STARTUP_PARAMS, &pccs->cmd);
+	dev_dbg(&link->dev, "dl_startup_params start ccsindex = %d\n",
+	      local->dl_param_ccs);
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		printk(KERN_INFO "ray dl_startup_params failed - "
+		       "ECF not ready for intr\n");
+		local->card_status = CARD_DL_PARAM_ERROR;
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+		return -2;
+	}
+	local->card_status = CARD_DL_PARAM;
+	/* Start kernel timer to wait for dl startup to complete. */
+	local->timer.expires = jiffies + HZ / 2;
+	local->timer.function = verify_dl_startup;
+	add_timer(&local->timer);
+	dev_dbg(&link->dev,
+	      "ray_cs dl_startup_params started timer for verify_dl_startup\n");
+	return 0;
+} /* dl_startup_params */
+
+/*===========================================================================*/
+static void init_startup_params(ray_dev_t *local)
+{
+	int i;
+
+	if (country > JAPAN_TEST)
+		country = USA;
+	else if (country < USA)
+		country = USA;
+	/* structure for hop time and beacon period is defined here using
+	 * New 802.11D6.1 format.  Card firmware is still using old format
+	 * until version 6.
+	 *    Before                    After
+	 *    a_hop_time ms byte        a_hop_time ms byte
+	 *    a_hop_time 2s byte        a_hop_time ls byte
+	 *    a_hop_time ls byte        a_beacon_period ms byte
+	 *    a_beacon_period           a_beacon_period ls byte
+	 *
+	 *    a_hop_time = uS           a_hop_time = KuS
+	 *    a_beacon_period = hops    a_beacon_period = KuS
+	 *//* 64ms = 010000 */
+	if (local->fw_ver == 0x55) {
+		memcpy(&local->sparm.b4, b4_default_startup_parms,
+		       sizeof(struct b4_startup_params));
+		/* Translate sane kus input values to old build 4/5 format */
+		/* i = hop time in uS truncated to 3 bytes */
+		i = (hop_dwell * 1024) & 0xffffff;
+		local->sparm.b4.a_hop_time[0] = (i >> 16) & 0xff;
+		local->sparm.b4.a_hop_time[1] = (i >> 8) & 0xff;
+		local->sparm.b4.a_beacon_period[0] = 0;
+		local->sparm.b4.a_beacon_period[1] =
+		    ((beacon_period / hop_dwell) - 1) & 0xff;
+		local->sparm.b4.a_curr_country_code = country;
+		local->sparm.b4.a_hop_pattern_length =
+		    hop_pattern_length[(int)country] - 1;
+		if (bc) {
+			local->sparm.b4.a_ack_timeout = 0x50;
+			local->sparm.b4.a_sifs = 0x3f;
+		}
+	} else { /* Version 5 uses real kus values */
+		memcpy((UCHAR *) &local->sparm.b5, b5_default_startup_parms,
+		       sizeof(struct b5_startup_params));
+
+		local->sparm.b5.a_hop_time[0] = (hop_dwell >> 8) & 0xff;
+		local->sparm.b5.a_hop_time[1] = hop_dwell & 0xff;
+		local->sparm.b5.a_beacon_period[0] =
+		    (beacon_period >> 8) & 0xff;
+		local->sparm.b5.a_beacon_period[1] = beacon_period & 0xff;
+		if (psm)
+			local->sparm.b5.a_power_mgt_state = 1;
+		local->sparm.b5.a_curr_country_code = country;
+		local->sparm.b5.a_hop_pattern_length =
+		    hop_pattern_length[(int)country];
+	}
+
+	local->sparm.b4.a_network_type = net_type & 0x01;
+	local->sparm.b4.a_acting_as_ap_status = TYPE_STA;
+
+	if (essid != NULL)
+		strncpy(local->sparm.b4.a_current_ess_id, essid, ESSID_SIZE);
+} /* init_startup_params */
+
+/*===========================================================================*/
+static void verify_dl_startup(struct timer_list *t)
+{
+	ray_dev_t *local = from_timer(local, t, timer);
+	struct ccs __iomem *pccs = ccs_base(local) + local->dl_param_ccs;
+	UCHAR status;
+	struct pcmcia_device *link = local->finder;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs verify_dl_startup - device not present\n");
+		return;
+	}
+#if 0
+	{
+		int i;
+		printk(KERN_DEBUG
+		       "verify_dl_startup parameters sent via ccs %d:\n",
+		       local->dl_param_ccs);
+		for (i = 0; i < sizeof(struct b5_startup_params); i++) {
+			printk(" %2x",
+			       (unsigned int)readb(local->sram +
+						   HOST_TO_ECF_BASE + i));
+		}
+		printk("\n");
+	}
+#endif
+
+	status = readb(&pccs->buffer_status);
+	if (status != CCS_BUFFER_FREE) {
+		printk(KERN_INFO
+		       "Download startup params failed.  Status = %d\n",
+		       status);
+		local->card_status = CARD_DL_PARAM_ERROR;
+		return;
+	}
+	if (local->sparm.b4.a_network_type == ADHOC)
+		start_net(&local->timer);
+	else
+		join_net(&local->timer);
+} /* end verify_dl_startup */
+
+/*===========================================================================*/
+/* Command card to start a network */
+static void start_net(struct timer_list *t)
+{
+	ray_dev_t *local = from_timer(local, t, timer);
+	struct ccs __iomem *pccs;
+	int ccsindex;
+	struct pcmcia_device *link = local->finder;
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs start_net - device not present\n");
+		return;
+	}
+	/* Fill in the CCS fields for the ECF */
+	if ((ccsindex = get_free_ccs(local)) < 0)
+		return;
+	pccs = ccs_base(local) + ccsindex;
+	writeb(CCS_START_NETWORK, &pccs->cmd);
+	writeb(0, &pccs->var.start_network.update_param);
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		dev_dbg(&link->dev, "ray start net failed - card not ready for intr\n");
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+		return;
+	}
+	local->card_status = CARD_DOING_ACQ;
+} /* end start_net */
+
+/*===========================================================================*/
+/* Command card to join a network */
+static void join_net(struct timer_list *t)
+{
+	ray_dev_t *local = from_timer(local, t, timer);
+
+	struct ccs __iomem *pccs;
+	int ccsindex;
+	struct pcmcia_device *link = local->finder;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs join_net - device not present\n");
+		return;
+	}
+	/* Fill in the CCS fields for the ECF */
+	if ((ccsindex = get_free_ccs(local)) < 0)
+		return;
+	pccs = ccs_base(local) + ccsindex;
+	writeb(CCS_JOIN_NETWORK, &pccs->cmd);
+	writeb(0, &pccs->var.join_network.update_param);
+	writeb(0, &pccs->var.join_network.net_initiated);
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		dev_dbg(&link->dev, "ray join net failed - card not ready for intr\n");
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+		return;
+	}
+	local->card_status = CARD_DOING_ACQ;
+}
+
+
+static void ray_release(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+	ray_dev_t *local = netdev_priv(dev);
+
+	dev_dbg(&link->dev, "ray_release\n");
+
+	del_timer(&local->timer);
+
+	iounmap(local->sram);
+	iounmap(local->rmem);
+	iounmap(local->amem);
+	pcmcia_disable_device(link);
+
+	dev_dbg(&link->dev, "ray_release ending\n");
+}
+
+static int ray_suspend(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+
+	if (link->open)
+		netif_device_detach(dev);
+
+	return 0;
+}
+
+static int ray_resume(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+
+	if (link->open) {
+		ray_reset(dev);
+		netif_device_attach(dev);
+	}
+
+	return 0;
+}
+
+/*===========================================================================*/
+static int ray_dev_init(struct net_device *dev)
+{
+#ifdef RAY_IMMEDIATE_INIT
+	int i;
+#endif /* RAY_IMMEDIATE_INIT */
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+
+	dev_dbg(&link->dev, "ray_dev_init(dev=%p)\n", dev);
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_dev_init - device not present\n");
+		return -1;
+	}
+#ifdef RAY_IMMEDIATE_INIT
+	/* Download startup parameters */
+	if ((i = dl_startup_params(dev)) < 0) {
+		printk(KERN_INFO "ray_dev_init dl_startup_params failed - "
+		       "returns 0x%x\n", i);
+		return -1;
+	}
+#else /* RAY_IMMEDIATE_INIT */
+	/* Postpone the card init so that we can still configure the card,
+	 * for example using the Wireless Extensions. The init will happen
+	 * in ray_open() - Jean II */
+	dev_dbg(&link->dev,
+	      "ray_dev_init: postponing card init to ray_open() ; Status = %d\n",
+	      local->card_status);
+#endif /* RAY_IMMEDIATE_INIT */
+
+	/* copy mac and broadcast addresses to linux device */
+	eth_hw_addr_set(dev, local->sparm.b4.a_mac_addr);
+	eth_broadcast_addr(dev->broadcast);
+
+	dev_dbg(&link->dev, "ray_dev_init ending\n");
+	return 0;
+}
+
+/*===========================================================================*/
+static int ray_dev_config(struct net_device *dev, struct ifmap *map)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	/* Dummy routine to satisfy device structure */
+	dev_dbg(&link->dev, "ray_dev_config(dev=%p,ifmap=%p)\n", dev, map);
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_dev_config - device not present\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+/*===========================================================================*/
+static netdev_tx_t ray_dev_start_xmit(struct sk_buff *skb,
+					    struct net_device *dev)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	short length = skb->len;
+
+	if (!pcmcia_dev_present(link)) {
+		dev_dbg(&link->dev, "ray_dev_start_xmit - device not present\n");
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+
+	dev_dbg(&link->dev, "ray_dev_start_xmit(skb=%p, dev=%p)\n", skb, dev);
+	if (local->authentication_state == NEED_TO_AUTH) {
+		dev_dbg(&link->dev, "ray_cs Sending authentication request.\n");
+		if (!build_auth_frame(local, local->auth_id, OPEN_AUTH_REQUEST)) {
+			local->authentication_state = AUTHENTICATED;
+			netif_stop_queue(dev);
+			return NETDEV_TX_BUSY;
+		}
+	}
+
+	if (length < ETH_ZLEN) {
+		if (skb_padto(skb, ETH_ZLEN))
+			return NETDEV_TX_OK;
+		length = ETH_ZLEN;
+	}
+	switch (ray_hw_xmit(skb->data, length, dev, DATA_TYPE)) {
+	case XMIT_NO_CCS:
+	case XMIT_NEED_AUTH:
+		netif_stop_queue(dev);
+		return NETDEV_TX_BUSY;
+	case XMIT_NO_INTR:
+	case XMIT_MSG_BAD:
+	case XMIT_OK:
+	default:
+		dev_kfree_skb(skb);
+	}
+
+	return NETDEV_TX_OK;
+} /* ray_dev_start_xmit */
+
+/*===========================================================================*/
+static int ray_hw_xmit(unsigned char *data, int len, struct net_device *dev,
+		       UCHAR msg_type)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct ccs __iomem *pccs;
+	int ccsindex;
+	int offset;
+	struct tx_msg __iomem *ptx;	/* Address of xmit buffer in PC space */
+	short int addr;		/* Address of xmit buffer in card space */
+
+	pr_debug("ray_hw_xmit(data=%p, len=%d, dev=%p)\n", data, len, dev);
+	if (len + TX_HEADER_LENGTH > TX_BUF_SIZE) {
+		printk(KERN_INFO "ray_hw_xmit packet too large: %d bytes\n",
+		       len);
+		return XMIT_MSG_BAD;
+	}
+	switch (ccsindex = get_free_tx_ccs(local)) {
+	case ECCSBUSY:
+		pr_debug("ray_hw_xmit tx_ccs table busy\n");
+		fallthrough;
+	case ECCSFULL:
+		pr_debug("ray_hw_xmit No free tx ccs\n");
+		fallthrough;
+	case ECARDGONE:
+		netif_stop_queue(dev);
+		return XMIT_NO_CCS;
+	default:
+		break;
+	}
+	addr = TX_BUF_BASE + (ccsindex << 11);
+
+	if (msg_type == DATA_TYPE) {
+		local->stats.tx_bytes += len;
+		local->stats.tx_packets++;
+	}
+
+	ptx = local->sram + addr;
+
+	ray_build_header(local, ptx, msg_type, data);
+	if (translate) {
+		offset = translate_frame(local, ptx, data, len);
+	} else { /* Encapsulate frame */
+		/* TBD TIB length will move address of ptx->var */
+		memcpy_toio(&ptx->var, data, len);
+		offset = 0;
+	}
+
+	/* fill in the CCS */
+	pccs = ccs_base(local) + ccsindex;
+	len += TX_HEADER_LENGTH + offset;
+	writeb(CCS_TX_REQUEST, &pccs->cmd);
+	writeb(addr >> 8, &pccs->var.tx_request.tx_data_ptr[0]);
+	writeb(local->tib_length, &pccs->var.tx_request.tx_data_ptr[1]);
+	writeb(len >> 8, &pccs->var.tx_request.tx_data_length[0]);
+	writeb(len & 0xff, &pccs->var.tx_request.tx_data_length[1]);
+/* TBD still need psm_cam? */
+	writeb(PSM_CAM, &pccs->var.tx_request.pow_sav_mode);
+	writeb(local->net_default_tx_rate, &pccs->var.tx_request.tx_rate);
+	writeb(0, &pccs->var.tx_request.antenna);
+	pr_debug("ray_hw_xmit default_tx_rate = 0x%x\n",
+	      local->net_default_tx_rate);
+
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		pr_debug("ray_hw_xmit failed - ECF not ready for intr\n");
+/* TBD very inefficient to copy packet to buffer, and then not
+   send it, but the alternative is to queue the messages and that
+   won't be done for a while.  Maybe set tbusy until a CCS is free?
+*/
+		writeb(CCS_BUFFER_FREE, &pccs->buffer_status);
+		return XMIT_NO_INTR;
+	}
+	return XMIT_OK;
+} /* end ray_hw_xmit */
+
+/*===========================================================================*/
+static int translate_frame(ray_dev_t *local, struct tx_msg __iomem *ptx,
+			   unsigned char *data, int len)
+{
+	__be16 proto = ((struct ethhdr *)data)->h_proto;
+	if (ntohs(proto) >= ETH_P_802_3_MIN) { /* DIX II ethernet frame */
+		pr_debug("ray_cs translate_frame DIX II\n");
+		/* Copy LLC header to card buffer */
+		memcpy_toio(&ptx->var, eth2_llc, sizeof(eth2_llc));
+		memcpy_toio(((void __iomem *)&ptx->var) + sizeof(eth2_llc),
+			    (UCHAR *) &proto, 2);
+		if (proto == htons(ETH_P_AARP) || proto == htons(ETH_P_IPX)) {
+			/* This is the selective translation table, only 2 entries */
+			writeb(0xf8,
+			       &((struct snaphdr_t __iomem *)ptx->var)->org[2]);
+		}
+		/* Copy body of ethernet packet without ethernet header */
+		memcpy_toio((void __iomem *)&ptx->var +
+			    sizeof(struct snaphdr_t), data + ETH_HLEN,
+			    len - ETH_HLEN);
+		return (int)sizeof(struct snaphdr_t) - ETH_HLEN;
+	} else { /* already  802 type, and proto is length */
+		pr_debug("ray_cs translate_frame 802\n");
+		if (proto == htons(0xffff)) { /* evil netware IPX 802.3 without LLC */
+			pr_debug("ray_cs translate_frame evil IPX\n");
+			memcpy_toio(&ptx->var, data + ETH_HLEN, len - ETH_HLEN);
+			return 0 - ETH_HLEN;
+		}
+		memcpy_toio(&ptx->var, data + ETH_HLEN, len - ETH_HLEN);
+		return 0 - ETH_HLEN;
+	}
+	/* TBD do other frame types */
+} /* end translate_frame */
+
+/*===========================================================================*/
+static void ray_build_header(ray_dev_t *local, struct tx_msg __iomem *ptx,
+			     UCHAR msg_type, unsigned char *data)
+{
+	writeb(PROTOCOL_VER | msg_type, &ptx->mac.frame_ctl_1);
+/*** IEEE 802.11 Address field assignments *************
+		TODS	FROMDS	addr_1		addr_2		addr_3	addr_4
+Adhoc		0	0	dest		src (terminal)	BSSID	N/A
+AP to Terminal	0	1	dest		AP(BSSID)	source	N/A
+Terminal to AP	1	0	AP(BSSID)	src (terminal)	dest	N/A
+AP to AP	1	1	dest AP		src AP		dest	source
+*******************************************************/
+	if (local->net_type == ADHOC) {
+		writeb(0, &ptx->mac.frame_ctl_2);
+		memcpy_toio(ptx->mac.addr_1, ((struct ethhdr *)data)->h_dest,
+			    ADDRLEN);
+		memcpy_toio(ptx->mac.addr_2, ((struct ethhdr *)data)->h_source,
+			    ADDRLEN);
+		memcpy_toio(ptx->mac.addr_3, local->bss_id, ADDRLEN);
+	} else { /* infrastructure */
+
+		if (local->sparm.b4.a_acting_as_ap_status) {
+			writeb(FC2_FROM_DS, &ptx->mac.frame_ctl_2);
+			memcpy_toio(ptx->mac.addr_1,
+				    ((struct ethhdr *)data)->h_dest, ADDRLEN);
+			memcpy_toio(ptx->mac.addr_2, local->bss_id, 6);
+			memcpy_toio(ptx->mac.addr_3,
+				    ((struct ethhdr *)data)->h_source, ADDRLEN);
+		} else { /* Terminal */
+
+			writeb(FC2_TO_DS, &ptx->mac.frame_ctl_2);
+			memcpy_toio(ptx->mac.addr_1, local->bss_id, ADDRLEN);
+			memcpy_toio(ptx->mac.addr_2,
+				    ((struct ethhdr *)data)->h_source, ADDRLEN);
+			memcpy_toio(ptx->mac.addr_3,
+				    ((struct ethhdr *)data)->h_dest, ADDRLEN);
+		}
+	}
+} /* end encapsulate_frame */
+
+/*====================================================================*/
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get protocol name
+ */
+static int ray_get_name(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	strcpy(wrqu->name, "IEEE 802.11-FH");
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : set frequency
+ */
+static int ray_set_freq(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	int err = -EINPROGRESS;	/* Call commit handler */
+
+	/* Reject if card is already initialised */
+	if (local->card_status != CARD_AWAITING_PARAM)
+		return -EBUSY;
+
+	/* Setting by channel number */
+	if ((wrqu->freq.m > USA_HOP_MOD) || (wrqu->freq.e > 0))
+		err = -EOPNOTSUPP;
+	else
+		local->sparm.b5.a_hop_pattern = wrqu->freq.m;
+
+	return err;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get frequency
+ */
+static int ray_get_freq(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	wrqu->freq.m = local->sparm.b5.a_hop_pattern;
+	wrqu->freq.e = 0;
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : set ESSID
+ */
+static int ray_set_essid(struct net_device *dev, struct iw_request_info *info,
+			 union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	/* Reject if card is already initialised */
+	if (local->card_status != CARD_AWAITING_PARAM)
+		return -EBUSY;
+
+	/* Check if we asked for `any' */
+	if (wrqu->essid.flags == 0)
+		/* Corey : can you do that ? */
+		return -EOPNOTSUPP;
+
+	/* Check the size of the string */
+	if (wrqu->essid.length > IW_ESSID_MAX_SIZE)
+		return -E2BIG;
+
+	/* Set the ESSID in the card */
+	memset(local->sparm.b5.a_current_ess_id, 0, IW_ESSID_MAX_SIZE);
+	memcpy(local->sparm.b5.a_current_ess_id, extra, wrqu->essid.length);
+
+	return -EINPROGRESS;	/* Call commit handler */
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get ESSID
+ */
+static int ray_get_essid(struct net_device *dev, struct iw_request_info *info,
+			 union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	UCHAR tmp[IW_ESSID_MAX_SIZE + 1];
+
+	/* Get the essid that was set */
+	memcpy(extra, local->sparm.b5.a_current_ess_id, IW_ESSID_MAX_SIZE);
+	memcpy(tmp, local->sparm.b5.a_current_ess_id, IW_ESSID_MAX_SIZE);
+	tmp[IW_ESSID_MAX_SIZE] = '\0';
+
+	/* Push it out ! */
+	wrqu->essid.length = strlen(tmp);
+	wrqu->essid.flags = 1;	/* active */
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get AP address
+ */
+static int ray_get_wap(struct net_device *dev, struct iw_request_info *info,
+		       union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	memcpy(wrqu->ap_addr.sa_data, local->bss_id, ETH_ALEN);
+	wrqu->ap_addr.sa_family = ARPHRD_ETHER;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : set Bit-Rate
+ */
+static int ray_set_rate(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	/* Reject if card is already initialised */
+	if (local->card_status != CARD_AWAITING_PARAM)
+		return -EBUSY;
+
+	/* Check if rate is in range */
+	if ((wrqu->bitrate.value != 1000000) && (wrqu->bitrate.value != 2000000))
+		return -EINVAL;
+
+	/* Hack for 1.5 Mb/s instead of 2 Mb/s */
+	if ((local->fw_ver == 0x55) &&	/* Please check */
+	    (wrqu->bitrate.value == 2000000))
+		local->net_default_tx_rate = 3;
+	else
+		local->net_default_tx_rate = wrqu->bitrate.value / 500000;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get Bit-Rate
+ */
+static int ray_get_rate(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	if (local->net_default_tx_rate == 3)
+		wrqu->bitrate.value = 2000000;	/* Hum... */
+	else
+		wrqu->bitrate.value = local->net_default_tx_rate * 500000;
+	wrqu->bitrate.fixed = 0;	/* We are in auto mode */
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : set RTS threshold
+ */
+static int ray_set_rts(struct net_device *dev, struct iw_request_info *info,
+		       union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	int rthr = wrqu->rts.value;
+
+	/* Reject if card is already initialised */
+	if (local->card_status != CARD_AWAITING_PARAM)
+		return -EBUSY;
+
+	/* if(wrq->u.rts.fixed == 0) we should complain */
+	if (wrqu->rts.disabled)
+		rthr = 32767;
+	else {
+		if ((rthr < 0) || (rthr > 2347))   /* What's the max packet size ??? */
+			return -EINVAL;
+	}
+	local->sparm.b5.a_rts_threshold[0] = (rthr >> 8) & 0xFF;
+	local->sparm.b5.a_rts_threshold[1] = rthr & 0xFF;
+
+	return -EINPROGRESS;	/* Call commit handler */
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get RTS threshold
+ */
+static int ray_get_rts(struct net_device *dev, struct iw_request_info *info,
+		       union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	wrqu->rts.value = (local->sparm.b5.a_rts_threshold[0] << 8)
+	    + local->sparm.b5.a_rts_threshold[1];
+	wrqu->rts.disabled = (wrqu->rts.value == 32767);
+	wrqu->rts.fixed = 1;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : set Fragmentation threshold
+ */
+static int ray_set_frag(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	int fthr = wrqu->frag.value;
+
+	/* Reject if card is already initialised */
+	if (local->card_status != CARD_AWAITING_PARAM)
+		return -EBUSY;
+
+	/* if(wrq->u.frag.fixed == 0) should complain */
+	if (wrqu->frag.disabled)
+		fthr = 32767;
+	else {
+		if ((fthr < 256) || (fthr > 2347))	/* To check out ! */
+			return -EINVAL;
+	}
+	local->sparm.b5.a_frag_threshold[0] = (fthr >> 8) & 0xFF;
+	local->sparm.b5.a_frag_threshold[1] = fthr & 0xFF;
+
+	return -EINPROGRESS;	/* Call commit handler */
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get Fragmentation threshold
+ */
+static int ray_get_frag(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	wrqu->frag.value = (local->sparm.b5.a_frag_threshold[0] << 8)
+	    + local->sparm.b5.a_frag_threshold[1];
+	wrqu->frag.disabled = (wrqu->frag.value == 32767);
+	wrqu->frag.fixed = 1;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : set Mode of Operation
+ */
+static int ray_set_mode(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	int err = -EINPROGRESS;	/* Call commit handler */
+	char card_mode = 1;
+
+	/* Reject if card is already initialised */
+	if (local->card_status != CARD_AWAITING_PARAM)
+		return -EBUSY;
+
+	switch (wrqu->mode) {
+	case IW_MODE_ADHOC:
+		card_mode = 0;
+		fallthrough;
+	case IW_MODE_INFRA:
+		local->sparm.b5.a_network_type = card_mode;
+		break;
+	default:
+		err = -EINVAL;
+	}
+
+	return err;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get Mode of Operation
+ */
+static int ray_get_mode(struct net_device *dev, struct iw_request_info *info,
+			union iwreq_data *wrqu, char *extra)
+{
+	ray_dev_t *local = netdev_priv(dev);
+
+	if (local->sparm.b5.a_network_type)
+		wrqu->mode = IW_MODE_INFRA;
+	else
+		wrqu->mode = IW_MODE_ADHOC;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Handler : get range info
+ */
+static int ray_get_range(struct net_device *dev, struct iw_request_info *info,
+			 union iwreq_data *wrqu, char *extra)
+{
+	struct iw_range *range = (struct iw_range *)extra;
+
+	memset(range, 0, sizeof(struct iw_range));
+
+	/* Set the length (very important for backward compatibility) */
+	wrqu->data.length = sizeof(struct iw_range);
+
+	/* Set the Wireless Extension versions */
+	range->we_version_compiled = WIRELESS_EXT;
+	range->we_version_source = 9;
+
+	/* Set information in the range struct */
+	range->throughput = 1.1 * 1000 * 1000;	/* Put the right number here */
+	range->num_channels = hop_pattern_length[(int)country];
+	range->num_frequency = 0;
+	range->max_qual.qual = 0;
+	range->max_qual.level = 255;	/* What's the correct value ? */
+	range->max_qual.noise = 255;	/* Idem */
+	range->num_bitrates = 2;
+	range->bitrate[0] = 1000000;	/* 1 Mb/s */
+	range->bitrate[1] = 2000000;	/* 2 Mb/s */
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Private Handler : set framing mode
+ */
+static int ray_set_framing(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	translate = !!*(extra);	/* Set framing mode */
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Private Handler : get framing mode
+ */
+static int ray_get_framing(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	*(extra) = translate;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Wireless Private Handler : get country
+ */
+static int ray_get_country(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	*(extra) = country;
+
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Commit handler : called after a bunch of SET operations
+ */
+static int ray_commit(struct net_device *dev, struct iw_request_info *info,
+		      union iwreq_data *wrqu, char *extra)
+{
+	return 0;
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * Stats handler : return Wireless Stats
+ */
+static iw_stats *ray_get_wireless_stats(struct net_device *dev)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	struct status __iomem *p = local->sram + STATUS_BASE;
+
+	local->wstats.status = local->card_status;
+#ifdef WIRELESS_SPY
+	if ((local->spy_data.spy_number > 0)
+	    && (local->sparm.b5.a_network_type == 0)) {
+		/* Get it from the first node in spy list */
+		local->wstats.qual.qual = local->spy_data.spy_stat[0].qual;
+		local->wstats.qual.level = local->spy_data.spy_stat[0].level;
+		local->wstats.qual.noise = local->spy_data.spy_stat[0].noise;
+		local->wstats.qual.updated =
+		    local->spy_data.spy_stat[0].updated;
+	}
+#endif /* WIRELESS_SPY */
+
+	if (pcmcia_dev_present(link)) {
+		local->wstats.qual.noise = readb(&p->rxnoise);
+		local->wstats.qual.updated |= 4;
+	}
+
+	return &local->wstats;
+} /* end ray_get_wireless_stats */
+
+/*------------------------------------------------------------------*/
+/*
+ * Structures to export the Wireless Handlers
+ */
+
+static const iw_handler ray_handler[] = {
+	IW_HANDLER(SIOCSIWCOMMIT, ray_commit),
+	IW_HANDLER(SIOCGIWNAME, ray_get_name),
+	IW_HANDLER(SIOCSIWFREQ, ray_set_freq),
+	IW_HANDLER(SIOCGIWFREQ, ray_get_freq),
+	IW_HANDLER(SIOCSIWMODE, ray_set_mode),
+	IW_HANDLER(SIOCGIWMODE, ray_get_mode),
+	IW_HANDLER(SIOCGIWRANGE, ray_get_range),
+#ifdef WIRELESS_SPY
+	IW_HANDLER(SIOCSIWSPY, iw_handler_set_spy),
+	IW_HANDLER(SIOCGIWSPY, iw_handler_get_spy),
+	IW_HANDLER(SIOCSIWTHRSPY, iw_handler_set_thrspy),
+	IW_HANDLER(SIOCGIWTHRSPY, iw_handler_get_thrspy),
+#endif /* WIRELESS_SPY */
+	IW_HANDLER(SIOCGIWAP, ray_get_wap),
+	IW_HANDLER(SIOCSIWESSID, ray_set_essid),
+	IW_HANDLER(SIOCGIWESSID, ray_get_essid),
+	IW_HANDLER(SIOCSIWRATE, ray_set_rate),
+	IW_HANDLER(SIOCGIWRATE, ray_get_rate),
+	IW_HANDLER(SIOCSIWRTS, ray_set_rts),
+	IW_HANDLER(SIOCGIWRTS, ray_get_rts),
+	IW_HANDLER(SIOCSIWFRAG, ray_set_frag),
+	IW_HANDLER(SIOCGIWFRAG, ray_get_frag),
+};
+
+#define SIOCSIPFRAMING	SIOCIWFIRSTPRIV	/* Set framing mode */
+#define SIOCGIPFRAMING	SIOCIWFIRSTPRIV + 1	/* Get framing mode */
+#define SIOCGIPCOUNTRY	SIOCIWFIRSTPRIV + 3	/* Get country code */
+
+static const iw_handler ray_private_handler[] = {
+	[0] = ray_set_framing,
+	[1] = ray_get_framing,
+	[3] = ray_get_country,
+};
+
+static const struct iw_priv_args ray_private_args[] = {
+/* cmd,		set_args,	get_args,	name */
+	{SIOCSIPFRAMING, IW_PRIV_TYPE_BYTE | IW_PRIV_SIZE_FIXED | 1, 0,
+	 "set_framing"},
+	{SIOCGIPFRAMING, 0, IW_PRIV_TYPE_BYTE | IW_PRIV_SIZE_FIXED | 1,
+	 "get_framing"},
+	{SIOCGIPCOUNTRY, 0, IW_PRIV_TYPE_BYTE | IW_PRIV_SIZE_FIXED | 1,
+	 "get_country"},
+};
+
+static const struct iw_handler_def ray_handler_def = {
+	.num_standard = ARRAY_SIZE(ray_handler),
+	.num_private = ARRAY_SIZE(ray_private_handler),
+	.num_private_args = ARRAY_SIZE(ray_private_args),
+	.standard = ray_handler,
+	.private = ray_private_handler,
+	.private_args = ray_private_args,
+	.get_wireless_stats = ray_get_wireless_stats,
+};
+
+/*===========================================================================*/
+static int ray_open(struct net_device *dev)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link;
+	link = local->finder;
+
+	dev_dbg(&link->dev, "ray_open('%s')\n", dev->name);
+
+	if (link->open == 0)
+		local->num_multi = 0;
+	link->open++;
+
+	/* If the card is not started, time to start it ! - Jean II */
+	if (local->card_status == CARD_AWAITING_PARAM) {
+		int i;
+
+		dev_dbg(&link->dev, "ray_open: doing init now !\n");
+
+		/* Download startup parameters */
+		if ((i = dl_startup_params(dev)) < 0) {
+			printk(KERN_INFO
+			       "ray_dev_init dl_startup_params failed - "
+			       "returns 0x%x\n", i);
+			return -1;
+		}
+	}
+
+	if (sniffer)
+		netif_stop_queue(dev);
+	else
+		netif_start_queue(dev);
+
+	dev_dbg(&link->dev, "ray_open ending\n");
+	return 0;
+} /* end ray_open */
+
+/*===========================================================================*/
+static int ray_dev_close(struct net_device *dev)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link;
+	link = local->finder;
+
+	dev_dbg(&link->dev, "ray_dev_close('%s')\n", dev->name);
+
+	link->open--;
+	netif_stop_queue(dev);
+
+	/* In here, we should stop the hardware (stop card from beeing active)
+	 * and set local->card_status to CARD_AWAITING_PARAM, so that while the
+	 * card is closed we can chage its configuration.
+	 * Probably also need a COR reset to get sane state - Jean II */
+
+	return 0;
+} /* end ray_dev_close */
+
+/*===========================================================================*/
+static void ray_reset(struct net_device *dev)
+{
+	pr_debug("ray_reset entered\n");
+}
+
+/*===========================================================================*/
+/* Cause a firmware interrupt if it is ready for one                         */
+/* Return nonzero if not ready                                               */
+static int interrupt_ecf(ray_dev_t *local, int ccs)
+{
+	int i = 50;
+	struct pcmcia_device *link = local->finder;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs interrupt_ecf - device not present\n");
+		return -1;
+	}
+	dev_dbg(&link->dev, "interrupt_ecf(local=%p, ccs = 0x%x\n", local, ccs);
+
+	while (i &&
+	       (readb(local->amem + CIS_OFFSET + ECF_INTR_OFFSET) &
+		ECF_INTR_SET))
+		i--;
+	if (i == 0) {
+		dev_dbg(&link->dev, "ray_cs interrupt_ecf card not ready for interrupt\n");
+		return -1;
+	}
+	/* Fill the mailbox, then kick the card */
+	writeb(ccs, local->sram + SCB_BASE);
+	writeb(ECF_INTR_SET, local->amem + CIS_OFFSET + ECF_INTR_OFFSET);
+	return 0;
+} /* interrupt_ecf */
+
+/*===========================================================================*/
+/* Get next free transmit CCS                                                */
+/* Return - index of current tx ccs                                          */
+static int get_free_tx_ccs(ray_dev_t *local)
+{
+	int i;
+	struct ccs __iomem *pccs = ccs_base(local);
+	struct pcmcia_device *link = local->finder;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs get_free_tx_ccs - device not present\n");
+		return ECARDGONE;
+	}
+
+	if (test_and_set_bit(0, &local->tx_ccs_lock)) {
+		dev_dbg(&link->dev, "ray_cs tx_ccs_lock busy\n");
+		return ECCSBUSY;
+	}
+
+	for (i = 0; i < NUMBER_OF_TX_CCS; i++) {
+		if (readb(&(pccs + i)->buffer_status) == CCS_BUFFER_FREE) {
+			writeb(CCS_BUFFER_BUSY, &(pccs + i)->buffer_status);
+			writeb(CCS_END_LIST, &(pccs + i)->link);
+			local->tx_ccs_lock = 0;
+			return i;
+		}
+	}
+	local->tx_ccs_lock = 0;
+	dev_dbg(&link->dev, "ray_cs ERROR no free tx CCS for raylink card\n");
+	return ECCSFULL;
+} /* get_free_tx_ccs */
+
+/*===========================================================================*/
+/* Get next free CCS                                                         */
+/* Return - index of current ccs                                             */
+static int get_free_ccs(ray_dev_t *local)
+{
+	int i;
+	struct ccs __iomem *pccs = ccs_base(local);
+	struct pcmcia_device *link = local->finder;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs get_free_ccs - device not present\n");
+		return ECARDGONE;
+	}
+	if (test_and_set_bit(0, &local->ccs_lock)) {
+		dev_dbg(&link->dev, "ray_cs ccs_lock busy\n");
+		return ECCSBUSY;
+	}
+
+	for (i = NUMBER_OF_TX_CCS; i < NUMBER_OF_CCS; i++) {
+		if (readb(&(pccs + i)->buffer_status) == CCS_BUFFER_FREE) {
+			writeb(CCS_BUFFER_BUSY, &(pccs + i)->buffer_status);
+			writeb(CCS_END_LIST, &(pccs + i)->link);
+			local->ccs_lock = 0;
+			return i;
+		}
+	}
+	local->ccs_lock = 0;
+	dev_dbg(&link->dev, "ray_cs ERROR no free CCS for raylink card\n");
+	return ECCSFULL;
+} /* get_free_ccs */
+
+/*===========================================================================*/
+static void authenticate_timeout(struct timer_list *t)
+{
+	ray_dev_t *local = from_timer(local, t, timer);
+	del_timer(&local->timer);
+	printk(KERN_INFO "ray_cs Authentication with access point failed"
+	       " - timeout\n");
+	join_net(&local->timer);
+}
+
+/*===========================================================================*/
+static int parse_addr(char *in_str, UCHAR *out)
+{
+	int i, k;
+	int len;
+
+	if (in_str == NULL)
+		return 0;
+	len = strnlen(in_str, ADDRLEN * 2 + 1) - 1;
+	if (len < 1)
+		return 0;
+	memset(out, 0, ADDRLEN);
+
+	i = 5;
+
+	while (len > 0) {
+		if ((k = hex_to_bin(in_str[len--])) != -1)
+			out[i] = k;
+		else
+			return 0;
+
+		if (len == 0)
+			break;
+		if ((k = hex_to_bin(in_str[len--])) != -1)
+			out[i] += k << 4;
+		else
+			return 0;
+		if (!i--)
+			break;
+	}
+	return 1;
+}
+
+/*===========================================================================*/
+static struct net_device_stats *ray_get_stats(struct net_device *dev)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	struct status __iomem *p = local->sram + STATUS_BASE;
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs net_device_stats - device not present\n");
+		return &local->stats;
+	}
+	if (readb(&p->mrx_overflow_for_host)) {
+		local->stats.rx_over_errors += swab16(readw(&p->mrx_overflow));
+		writeb(0, &p->mrx_overflow);
+		writeb(0, &p->mrx_overflow_for_host);
+	}
+	if (readb(&p->mrx_checksum_error_for_host)) {
+		local->stats.rx_crc_errors +=
+		    swab16(readw(&p->mrx_checksum_error));
+		writeb(0, &p->mrx_checksum_error);
+		writeb(0, &p->mrx_checksum_error_for_host);
+	}
+	if (readb(&p->rx_hec_error_for_host)) {
+		local->stats.rx_frame_errors += swab16(readw(&p->rx_hec_error));
+		writeb(0, &p->rx_hec_error);
+		writeb(0, &p->rx_hec_error_for_host);
+	}
+	return &local->stats;
+}
+
+/*===========================================================================*/
+static void ray_update_parm(struct net_device *dev, UCHAR objid, UCHAR *value,
+			    int len)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	int ccsindex;
+	int i;
+	struct ccs __iomem *pccs;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_update_parm - device not present\n");
+		return;
+	}
+
+	if ((ccsindex = get_free_ccs(local)) < 0) {
+		dev_dbg(&link->dev, "ray_update_parm - No free ccs\n");
+		return;
+	}
+	pccs = ccs_base(local) + ccsindex;
+	writeb(CCS_UPDATE_PARAMS, &pccs->cmd);
+	writeb(objid, &pccs->var.update_param.object_id);
+	writeb(1, &pccs->var.update_param.number_objects);
+	writeb(0, &pccs->var.update_param.failure_cause);
+	for (i = 0; i < len; i++) {
+		writeb(value[i], local->sram + HOST_TO_ECF_BASE);
+	}
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		dev_dbg(&link->dev, "ray_cs associate failed - ECF not ready for intr\n");
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+	}
+}
+
+/*===========================================================================*/
+static void ray_update_multi_list(struct net_device *dev, int all)
+{
+	int ccsindex;
+	struct ccs __iomem *pccs;
+	ray_dev_t *local = netdev_priv(dev);
+	struct pcmcia_device *link = local->finder;
+	void __iomem *p = local->sram + HOST_TO_ECF_BASE;
+
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_update_multi_list - device not present\n");
+		return;
+	} else
+		dev_dbg(&link->dev, "ray_update_multi_list(%p)\n", dev);
+	if ((ccsindex = get_free_ccs(local)) < 0) {
+		dev_dbg(&link->dev, "ray_update_multi - No free ccs\n");
+		return;
+	}
+	pccs = ccs_base(local) + ccsindex;
+	writeb(CCS_UPDATE_MULTICAST_LIST, &pccs->cmd);
+
+	if (all) {
+		writeb(0xff, &pccs->var);
+		local->num_multi = 0xff;
+	} else {
+		struct netdev_hw_addr *ha;
+		int i = 0;
+
+		/* Copy the kernel's list of MC addresses to card */
+		netdev_for_each_mc_addr(ha, dev) {
+			memcpy_toio(p, ha->addr, ETH_ALEN);
+			dev_dbg(&link->dev, "ray_update_multi add addr %pm\n",
+				ha->addr);
+			p += ETH_ALEN;
+			i++;
+		}
+		if (i > 256 / ADDRLEN)
+			i = 256 / ADDRLEN;
+		writeb((UCHAR) i, &pccs->var);
+		dev_dbg(&link->dev, "ray_cs update_multi %d addresses in list\n", i);
+		/* Interrupt the firmware to process the command */
+		local->num_multi = i;
+	}
+	if (interrupt_ecf(local, ccsindex)) {
+		dev_dbg(&link->dev,
+		      "ray_cs update_multi failed - ECF not ready for intr\n");
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+	}
+} /* end ray_update_multi_list */
+
+/*===========================================================================*/
+static void set_multicast_list(struct net_device *dev)
+{
+	ray_dev_t *local = netdev_priv(dev);
+	UCHAR promisc;
+
+	pr_debug("ray_cs set_multicast_list(%p)\n", dev);
+
+	if (dev->flags & IFF_PROMISC) {
+		if (local->sparm.b5.a_promiscuous_mode == 0) {
+			pr_debug("ray_cs set_multicast_list promisc on\n");
+			local->sparm.b5.a_promiscuous_mode = 1;
+			promisc = 1;
+			ray_update_parm(dev, OBJID_promiscuous_mode,
+					&promisc, sizeof(promisc));
+		}
+	} else {
+		if (local->sparm.b5.a_promiscuous_mode == 1) {
+			pr_debug("ray_cs set_multicast_list promisc off\n");
+			local->sparm.b5.a_promiscuous_mode = 0;
+			promisc = 0;
+			ray_update_parm(dev, OBJID_promiscuous_mode,
+					&promisc, sizeof(promisc));
+		}
+	}
+
+	if (dev->flags & IFF_ALLMULTI)
+		ray_update_multi_list(dev, 1);
+	else {
+		if (local->num_multi != netdev_mc_count(dev))
+			ray_update_multi_list(dev, 0);
+	}
+} /* end set_multicast_list */
+
+/*=============================================================================
+ * All routines below here are run at interrupt time.
+=============================================================================*/
+static irqreturn_t ray_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = (struct net_device *)dev_id;
+	struct pcmcia_device *link;
+	ray_dev_t *local;
+	struct ccs __iomem *pccs;
+	struct rcs __iomem *prcs;
+	UCHAR rcsindex;
+	UCHAR tmp;
+	UCHAR cmd;
+	UCHAR status;
+	UCHAR memtmp[ESSID_SIZE + 1];
+
+
+	if (dev == NULL)	/* Note that we want interrupts with dev->start == 0 */
+		return IRQ_NONE;
+
+	pr_debug("ray_cs: interrupt for *dev=%p\n", dev);
+
+	local = netdev_priv(dev);
+	link = local->finder;
+	if (!pcmcia_dev_present(link)) {
+		pr_debug(
+			"ray_cs interrupt from device not present or suspended.\n");
+		return IRQ_NONE;
+	}
+	rcsindex = readb(&((struct scb __iomem *)(local->sram))->rcs_index);
+
+	if (rcsindex >= (NUMBER_OF_CCS + NUMBER_OF_RCS)) {
+		dev_dbg(&link->dev, "ray_cs interrupt bad rcsindex = 0x%x\n", rcsindex);
+		clear_interrupt(local);
+		return IRQ_HANDLED;
+	}
+	if (rcsindex < NUMBER_OF_CCS) { /* If it's a returned CCS */
+		pccs = ccs_base(local) + rcsindex;
+		cmd = readb(&pccs->cmd);
+		status = readb(&pccs->buffer_status);
+		switch (cmd) {
+		case CCS_DOWNLOAD_STARTUP_PARAMS:	/* Happens in firmware someday */
+			del_timer(&local->timer);
+			if (status == CCS_COMMAND_COMPLETE) {
+				dev_dbg(&link->dev,
+				      "ray_cs interrupt download_startup_parameters OK\n");
+			} else {
+				dev_dbg(&link->dev,
+				      "ray_cs interrupt download_startup_parameters fail\n");
+			}
+			break;
+		case CCS_UPDATE_PARAMS:
+			dev_dbg(&link->dev, "ray_cs interrupt update params done\n");
+			if (status != CCS_COMMAND_COMPLETE) {
+				tmp =
+				    readb(&pccs->var.update_param.
+					  failure_cause);
+				dev_dbg(&link->dev,
+				      "ray_cs interrupt update params failed - reason %d\n",
+				      tmp);
+			}
+			break;
+		case CCS_REPORT_PARAMS:
+			dev_dbg(&link->dev, "ray_cs interrupt report params done\n");
+			break;
+		case CCS_UPDATE_MULTICAST_LIST:	/* Note that this CCS isn't returned */
+			dev_dbg(&link->dev,
+			      "ray_cs interrupt CCS Update Multicast List done\n");
+			break;
+		case CCS_UPDATE_POWER_SAVINGS_MODE:
+			dev_dbg(&link->dev,
+			      "ray_cs interrupt update power save mode done\n");
+			break;
+		case CCS_START_NETWORK:
+		case CCS_JOIN_NETWORK:
+			memcpy(memtmp, local->sparm.b4.a_current_ess_id,
+								ESSID_SIZE);
+			memtmp[ESSID_SIZE] = '\0';
+
+			if (status == CCS_COMMAND_COMPLETE) {
+				if (readb
+				    (&pccs->var.start_network.net_initiated) ==
+				    1) {
+					dev_dbg(&link->dev,
+					      "ray_cs interrupt network \"%s\" started\n",
+					      memtmp);
+				} else {
+					dev_dbg(&link->dev,
+					      "ray_cs interrupt network \"%s\" joined\n",
+					      memtmp);
+				}
+				memcpy_fromio(&local->bss_id,
+					      pccs->var.start_network.bssid,
+					      ADDRLEN);
+
+				if (local->fw_ver == 0x55)
+					local->net_default_tx_rate = 3;
+				else
+					local->net_default_tx_rate =
+					    readb(&pccs->var.start_network.
+						  net_default_tx_rate);
+				local->encryption =
+				    readb(&pccs->var.start_network.encryption);
+				if (!sniffer && (local->net_type == INFRA)
+				    && !(local->sparm.b4.a_acting_as_ap_status)) {
+					authenticate(local);
+				}
+				local->card_status = CARD_ACQ_COMPLETE;
+			} else {
+				local->card_status = CARD_ACQ_FAILED;
+
+				del_timer(&local->timer);
+				local->timer.expires = jiffies + HZ * 5;
+				if (status == CCS_START_NETWORK) {
+					dev_dbg(&link->dev,
+					      "ray_cs interrupt network \"%s\" start failed\n",
+					      memtmp);
+					local->timer.function = start_net;
+				} else {
+					dev_dbg(&link->dev,
+					      "ray_cs interrupt network \"%s\" join failed\n",
+					      memtmp);
+					local->timer.function = join_net;
+				}
+				add_timer(&local->timer);
+			}
+			break;
+		case CCS_START_ASSOCIATION:
+			if (status == CCS_COMMAND_COMPLETE) {
+				local->card_status = CARD_ASSOC_COMPLETE;
+				dev_dbg(&link->dev, "ray_cs association successful\n");
+			} else {
+				dev_dbg(&link->dev, "ray_cs association failed,\n");
+				local->card_status = CARD_ASSOC_FAILED;
+				join_net(&local->timer);
+			}
+			break;
+		case CCS_TX_REQUEST:
+			if (status == CCS_COMMAND_COMPLETE) {
+				dev_dbg(&link->dev,
+				      "ray_cs interrupt tx request complete\n");
+			} else {
+				dev_dbg(&link->dev,
+				      "ray_cs interrupt tx request failed\n");
+			}
+			if (!sniffer)
+				netif_start_queue(dev);
+			netif_wake_queue(dev);
+			break;
+		case CCS_TEST_MEMORY:
+			dev_dbg(&link->dev, "ray_cs interrupt mem test done\n");
+			break;
+		case CCS_SHUTDOWN:
+			dev_dbg(&link->dev,
+			      "ray_cs interrupt Unexpected CCS returned - Shutdown\n");
+			break;
+		case CCS_DUMP_MEMORY:
+			dev_dbg(&link->dev, "ray_cs interrupt dump memory done\n");
+			break;
+		case CCS_START_TIMER:
+			dev_dbg(&link->dev,
+			      "ray_cs interrupt DING - raylink timer expired\n");
+			break;
+		default:
+			dev_dbg(&link->dev,
+			      "ray_cs interrupt Unexpected CCS 0x%x returned 0x%x\n",
+			      rcsindex, cmd);
+		}
+		writeb(CCS_BUFFER_FREE, &pccs->buffer_status);
+	} else { /* It's an RCS */
+
+		prcs = rcs_base(local) + rcsindex;
+
+		switch (readb(&prcs->interrupt_id)) {
+		case PROCESS_RX_PACKET:
+			ray_rx(dev, local, prcs);
+			break;
+		case REJOIN_NET_COMPLETE:
+			dev_dbg(&link->dev, "ray_cs interrupt rejoin net complete\n");
+			local->card_status = CARD_ACQ_COMPLETE;
+			/* do we need to clear tx buffers CCS's? */
+			if (local->sparm.b4.a_network_type == ADHOC) {
+				if (!sniffer)
+					netif_start_queue(dev);
+			} else {
+				memcpy_fromio(&local->bss_id,
+					      prcs->var.rejoin_net_complete.
+					      bssid, ADDRLEN);
+				dev_dbg(&link->dev, "ray_cs new BSSID = %pm\n",
+					local->bss_id);
+				if (!sniffer)
+					authenticate(local);
+			}
+			break;
+		case ROAMING_INITIATED:
+			dev_dbg(&link->dev, "ray_cs interrupt roaming initiated\n");
+			netif_stop_queue(dev);
+			local->card_status = CARD_DOING_ACQ;
+			break;
+		case JAPAN_CALL_SIGN_RXD:
+			dev_dbg(&link->dev, "ray_cs interrupt japan call sign rx\n");
+			break;
+		default:
+			dev_dbg(&link->dev,
+			      "ray_cs Unexpected interrupt for RCS 0x%x cmd = 0x%x\n",
+			      rcsindex,
+			      (unsigned int)readb(&prcs->interrupt_id));
+			break;
+		}
+		writeb(CCS_BUFFER_FREE, &prcs->buffer_status);
+	}
+	clear_interrupt(local);
+	return IRQ_HANDLED;
+} /* ray_interrupt */
+
+/*===========================================================================*/
+static void ray_rx(struct net_device *dev, ray_dev_t *local,
+		   struct rcs __iomem *prcs)
+{
+	int rx_len;
+	unsigned int pkt_addr;
+	void __iomem *pmsg;
+	pr_debug("ray_rx process rx packet\n");
+
+	/* Calculate address of packet within Rx buffer */
+	pkt_addr = ((readb(&prcs->var.rx_packet.rx_data_ptr[0]) << 8)
+		    + readb(&prcs->var.rx_packet.rx_data_ptr[1])) & RX_BUFF_END;
+	/* Length of first packet fragment */
+	rx_len = (readb(&prcs->var.rx_packet.rx_data_length[0]) << 8)
+	    + readb(&prcs->var.rx_packet.rx_data_length[1]);
+
+	local->last_rsl = readb(&prcs->var.rx_packet.rx_sig_lev);
+	pmsg = local->rmem + pkt_addr;
+	switch (readb(pmsg)) {
+	case DATA_TYPE:
+		pr_debug("ray_rx data type\n");
+		rx_data(dev, prcs, pkt_addr, rx_len);
+		break;
+	case AUTHENTIC_TYPE:
+		pr_debug("ray_rx authentic type\n");
+		if (sniffer)
+			rx_data(dev, prcs, pkt_addr, rx_len);
+		else
+			rx_authenticate(local, prcs, pkt_addr, rx_len);
+		break;
+	case DEAUTHENTIC_TYPE:
+		pr_debug("ray_rx deauth type\n");
+		if (sniffer)
+			rx_data(dev, prcs, pkt_addr, rx_len);
+		else
+			rx_deauthenticate(local, prcs, pkt_addr, rx_len);
+		break;
+	case NULL_MSG_TYPE:
+		pr_debug("ray_cs rx NULL msg\n");
+		break;
+	case BEACON_TYPE:
+		pr_debug("ray_rx beacon type\n");
+		if (sniffer)
+			rx_data(dev, prcs, pkt_addr, rx_len);
+
+		copy_from_rx_buff(local, (UCHAR *) &local->last_bcn, pkt_addr,
+				  rx_len < sizeof(struct beacon_rx) ?
+				  rx_len : sizeof(struct beacon_rx));
+
+		local->beacon_rxed = 1;
+		/* Get the statistics so the card counters never overflow */
+		ray_get_stats(dev);
+		break;
+	default:
+		pr_debug("ray_cs unknown pkt type %2x\n",
+		      (unsigned int)readb(pmsg));
+		break;
+	}
+
+} /* end ray_rx */
+
+/*===========================================================================*/
+static void rx_data(struct net_device *dev, struct rcs __iomem *prcs,
+		    unsigned int pkt_addr, int rx_len)
+{
+	struct sk_buff *skb = NULL;
+	struct rcs __iomem *prcslink = prcs;
+	ray_dev_t *local = netdev_priv(dev);
+	UCHAR *rx_ptr;
+	int total_len;
+	int tmp;
+#ifdef WIRELESS_SPY
+	int siglev = local->last_rsl;
+	u_char linksrcaddr[ETH_ALEN];	/* Other end of the wireless link */
+#endif
+
+	if (!sniffer) {
+		if (translate) {
+/* TBD length needs fixing for translated header */
+			if (rx_len < (ETH_HLEN + RX_MAC_HEADER_LENGTH) ||
+			    rx_len >
+			    (dev->mtu + RX_MAC_HEADER_LENGTH + ETH_HLEN +
+			     FCS_LEN)) {
+				pr_debug(
+				      "ray_cs invalid packet length %d received\n",
+				      rx_len);
+				return;
+			}
+		} else { /* encapsulated ethernet */
+
+			if (rx_len < (ETH_HLEN + RX_MAC_HEADER_LENGTH) ||
+			    rx_len >
+			    (dev->mtu + RX_MAC_HEADER_LENGTH + ETH_HLEN +
+			     FCS_LEN)) {
+				pr_debug(
+				      "ray_cs invalid packet length %d received\n",
+				      rx_len);
+				return;
+			}
+		}
+	}
+	pr_debug("ray_cs rx_data packet\n");
+	/* If fragmented packet, verify sizes of fragments add up */
+	if (readb(&prcs->var.rx_packet.next_frag_rcs_index) != 0xFF) {
+		pr_debug("ray_cs rx'ed fragment\n");
+		tmp = (readb(&prcs->var.rx_packet.totalpacketlength[0]) << 8)
+		    + readb(&prcs->var.rx_packet.totalpacketlength[1]);
+		total_len = tmp;
+		prcslink = prcs;
+		do {
+			tmp -=
+			    (readb(&prcslink->var.rx_packet.rx_data_length[0])
+			     << 8)
+			    + readb(&prcslink->var.rx_packet.rx_data_length[1]);
+			if (readb(&prcslink->var.rx_packet.next_frag_rcs_index)
+			    == 0xFF || tmp < 0)
+				break;
+			prcslink = rcs_base(local)
+			    + readb(&prcslink->link_field);
+		} while (1);
+
+		if (tmp < 0) {
+			pr_debug(
+			      "ray_cs rx_data fragment lengths don't add up\n");
+			local->stats.rx_dropped++;
+			release_frag_chain(local, prcs);
+			return;
+		}
+	} else { /* Single unfragmented packet */
+		total_len = rx_len;
+	}
+
+	skb = dev_alloc_skb(total_len + 5);
+	if (skb == NULL) {
+		pr_debug("ray_cs rx_data could not allocate skb\n");
+		local->stats.rx_dropped++;
+		if (readb(&prcs->var.rx_packet.next_frag_rcs_index) != 0xFF)
+			release_frag_chain(local, prcs);
+		return;
+	}
+	skb_reserve(skb, 2);	/* Align IP on 16 byte (TBD check this) */
+
+	pr_debug("ray_cs rx_data total_len = %x, rx_len = %x\n", total_len,
+	      rx_len);
+
+/************************/
+	/* Reserve enough room for the whole damn packet. */
+	rx_ptr = skb_put(skb, total_len);
+	/* Copy the whole packet to sk_buff */
+	rx_ptr +=
+	    copy_from_rx_buff(local, rx_ptr, pkt_addr & RX_BUFF_END, rx_len);
+	/* Get source address */
+#ifdef WIRELESS_SPY
+	skb_copy_from_linear_data_offset(skb,
+					 offsetof(struct mac_header, addr_2),
+					 linksrcaddr, ETH_ALEN);
+#endif
+	/* Now, deal with encapsulation/translation/sniffer */
+	if (!sniffer) {
+		if (!translate) {
+			/* Encapsulated ethernet, so just lop off 802.11 MAC header */
+/* TBD reserve            skb_reserve( skb, RX_MAC_HEADER_LENGTH); */
+			skb_pull(skb, RX_MAC_HEADER_LENGTH);
+		} else {
+			/* Do translation */
+			untranslate(local, skb, total_len);
+		}
+	} else { /* sniffer mode, so just pass whole packet */
+	}
+
+/************************/
+	/* Now pick up the rest of the fragments if any */
+	tmp = 17;
+	if (readb(&prcs->var.rx_packet.next_frag_rcs_index) != 0xFF) {
+		prcslink = prcs;
+		pr_debug("ray_cs rx_data in fragment loop\n");
+		do {
+			prcslink = rcs_base(local)
+			    +
+			    readb(&prcslink->var.rx_packet.next_frag_rcs_index);
+			rx_len =
+			    ((readb(&prcslink->var.rx_packet.rx_data_length[0])
+			      << 8)
+			     +
+			     readb(&prcslink->var.rx_packet.rx_data_length[1]))
+			    & RX_BUFF_END;
+			pkt_addr =
+			    ((readb(&prcslink->var.rx_packet.rx_data_ptr[0]) <<
+			      8)
+			     + readb(&prcslink->var.rx_packet.rx_data_ptr[1]))
+			    & RX_BUFF_END;
+
+			rx_ptr +=
+			    copy_from_rx_buff(local, rx_ptr, pkt_addr, rx_len);
+
+		} while (tmp-- &&
+			 readb(&prcslink->var.rx_packet.next_frag_rcs_index) !=
+			 0xFF);
+		release_frag_chain(local, prcs);
+	}
+
+	skb->protocol = eth_type_trans(skb, dev);
+	netif_rx(skb);
+	local->stats.rx_packets++;
+	local->stats.rx_bytes += total_len;
+
+	/* Gather signal strength per address */
+#ifdef WIRELESS_SPY
+	/* For the Access Point or the node having started the ad-hoc net
+	 * note : ad-hoc work only in some specific configurations, but we
+	 * kludge in ray_get_wireless_stats... */
+	if (!memcmp(linksrcaddr, local->bss_id, ETH_ALEN)) {
+		/* Update statistics */
+		/*local->wstats.qual.qual = none ? */
+		local->wstats.qual.level = siglev;
+		/*local->wstats.qual.noise = none ? */
+		local->wstats.qual.updated = 0x2;
+	}
+	/* Now, update the spy stuff */
+	{
+		struct iw_quality wstats;
+		wstats.level = siglev;
+		/* wstats.noise = none ? */
+		/* wstats.qual = none ? */
+		wstats.updated = 0x2;
+		/* Update spy records */
+		wireless_spy_update(dev, linksrcaddr, &wstats);
+	}
+#endif /* WIRELESS_SPY */
+} /* end rx_data */
+
+/*===========================================================================*/
+static void untranslate(ray_dev_t *local, struct sk_buff *skb, int len)
+{
+	snaphdr_t *psnap = (snaphdr_t *) (skb->data + RX_MAC_HEADER_LENGTH);
+	struct ieee80211_hdr *pmac = (struct ieee80211_hdr *)skb->data;
+	__be16 type = *(__be16 *) psnap->ethertype;
+	int delta;
+	struct ethhdr *peth;
+	UCHAR srcaddr[ADDRLEN];
+	UCHAR destaddr[ADDRLEN];
+	static const UCHAR org_bridge[3] = { 0, 0, 0xf8 };
+	static const UCHAR org_1042[3] = { 0, 0, 0 };
+
+	memcpy(destaddr, ieee80211_get_DA(pmac), ADDRLEN);
+	memcpy(srcaddr, ieee80211_get_SA(pmac), ADDRLEN);
+
+#if 0
+	if {
+		print_hex_dump(KERN_DEBUG, "skb->data before untranslate: ",
+			       DUMP_PREFIX_NONE, 16, 1,
+			       skb->data, 64, true);
+		printk(KERN_DEBUG
+		       "type = %08x, xsap = %02x%02x%02x, org = %02x02x02x\n",
+		       ntohs(type), psnap->dsap, psnap->ssap, psnap->ctrl,
+		       psnap->org[0], psnap->org[1], psnap->org[2]);
+		printk(KERN_DEBUG "untranslate skb->data = %p\n", skb->data);
+	}
+#endif
+
+	if (psnap->dsap != 0xaa || psnap->ssap != 0xaa || psnap->ctrl != 3) {
+		/* not a snap type so leave it alone */
+		pr_debug("ray_cs untranslate NOT SNAP %02x %02x %02x\n",
+		      psnap->dsap, psnap->ssap, psnap->ctrl);
+
+		delta = RX_MAC_HEADER_LENGTH - ETH_HLEN;
+		peth = (struct ethhdr *)(skb->data + delta);
+		peth->h_proto = htons(len - RX_MAC_HEADER_LENGTH);
+	} else { /* Its a SNAP */
+		if (memcmp(psnap->org, org_bridge, 3) == 0) {
+		/* EtherII and nuke the LLC */
+			pr_debug("ray_cs untranslate Bridge encap\n");
+			delta = RX_MAC_HEADER_LENGTH
+			    + sizeof(struct snaphdr_t) - ETH_HLEN;
+			peth = (struct ethhdr *)(skb->data + delta);
+			peth->h_proto = type;
+		} else if (memcmp(psnap->org, org_1042, 3) == 0) {
+			switch (ntohs(type)) {
+			case ETH_P_IPX:
+			case ETH_P_AARP:
+				pr_debug("ray_cs untranslate RFC IPX/AARP\n");
+				delta = RX_MAC_HEADER_LENGTH - ETH_HLEN;
+				peth = (struct ethhdr *)(skb->data + delta);
+				peth->h_proto =
+				    htons(len - RX_MAC_HEADER_LENGTH);
+				break;
+			default:
+				pr_debug("ray_cs untranslate RFC default\n");
+				delta = RX_MAC_HEADER_LENGTH +
+				    sizeof(struct snaphdr_t) - ETH_HLEN;
+				peth = (struct ethhdr *)(skb->data + delta);
+				peth->h_proto = type;
+				break;
+			}
+		} else {
+			printk("ray_cs untranslate very confused by packet\n");
+			delta = RX_MAC_HEADER_LENGTH - ETH_HLEN;
+			peth = (struct ethhdr *)(skb->data + delta);
+			peth->h_proto = type;
+		}
+	}
+/* TBD reserve  skb_reserve(skb, delta); */
+	skb_pull(skb, delta);
+	pr_debug("untranslate after skb_pull(%d), skb->data = %p\n", delta,
+	      skb->data);
+	memcpy(peth->h_dest, destaddr, ADDRLEN);
+	memcpy(peth->h_source, srcaddr, ADDRLEN);
+#if 0
+	{
+		int i;
+		printk(KERN_DEBUG "skb->data after untranslate:");
+		for (i = 0; i < 64; i++)
+			printk("%02x ", skb->data[i]);
+		printk("\n");
+	}
+#endif
+} /* end untranslate */
+
+/*===========================================================================*/
+/* Copy data from circular receive buffer to PC memory.
+ * dest     = destination address in PC memory
+ * pkt_addr = source address in receive buffer
+ * len      = length of packet to copy
+ */
+static int copy_from_rx_buff(ray_dev_t *local, UCHAR *dest, int pkt_addr,
+			     int length)
+{
+	int wrap_bytes = (pkt_addr + length) - (RX_BUFF_END + 1);
+	if (wrap_bytes <= 0) {
+		memcpy_fromio(dest, local->rmem + pkt_addr, length);
+	} else { /* Packet wrapped in circular buffer */
+
+		memcpy_fromio(dest, local->rmem + pkt_addr,
+			      length - wrap_bytes);
+		memcpy_fromio(dest + length - wrap_bytes, local->rmem,
+			      wrap_bytes);
+	}
+	return length;
+}
+
+/*===========================================================================*/
+static void release_frag_chain(ray_dev_t *local, struct rcs __iomem *prcs)
+{
+	struct rcs __iomem *prcslink = prcs;
+	int tmp = 17;
+	unsigned rcsindex = readb(&prcs->var.rx_packet.next_frag_rcs_index);
+
+	while (tmp--) {
+		writeb(CCS_BUFFER_FREE, &prcslink->buffer_status);
+		if (rcsindex >= (NUMBER_OF_CCS + NUMBER_OF_RCS)) {
+			pr_debug("ray_cs interrupt bad rcsindex = 0x%x\n",
+			      rcsindex);
+			break;
+		}
+		prcslink = rcs_base(local) + rcsindex;
+		rcsindex = readb(&prcslink->var.rx_packet.next_frag_rcs_index);
+	}
+	writeb(CCS_BUFFER_FREE, &prcslink->buffer_status);
+}
+
+/*===========================================================================*/
+static void authenticate(ray_dev_t *local)
+{
+	struct pcmcia_device *link = local->finder;
+	dev_dbg(&link->dev, "ray_cs Starting authentication.\n");
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs authenticate - device not present\n");
+		return;
+	}
+
+	del_timer(&local->timer);
+	if (build_auth_frame(local, local->bss_id, OPEN_AUTH_REQUEST)) {
+		local->timer.function = join_net;
+	} else {
+		local->timer.function = authenticate_timeout;
+	}
+	local->timer.expires = jiffies + HZ * 2;
+	add_timer(&local->timer);
+	local->authentication_state = AWAITING_RESPONSE;
+} /* end authenticate */
+
+/*===========================================================================*/
+static void rx_authenticate(ray_dev_t *local, struct rcs __iomem *prcs,
+			    unsigned int pkt_addr, int rx_len)
+{
+	UCHAR buff[256];
+	struct ray_rx_msg *msg = (struct ray_rx_msg *) buff;
+
+	del_timer(&local->timer);
+
+	copy_from_rx_buff(local, buff, pkt_addr, rx_len & 0xff);
+	/* if we are trying to get authenticated */
+	if (local->sparm.b4.a_network_type == ADHOC) {
+		pr_debug("ray_cs rx_auth var= %6ph\n", msg->var);
+		if (msg->var[2] == 1) {
+			pr_debug("ray_cs Sending authentication response.\n");
+			if (!build_auth_frame
+			    (local, msg->mac.addr_2, OPEN_AUTH_RESPONSE)) {
+				local->authentication_state = NEED_TO_AUTH;
+				memcpy(local->auth_id, msg->mac.addr_2,
+				       ADDRLEN);
+			}
+		}
+	} else { /* Infrastructure network */
+
+		if (local->authentication_state == AWAITING_RESPONSE) {
+			/* Verify authentication sequence #2 and success */
+			if (msg->var[2] == 2) {
+				if ((msg->var[3] | msg->var[4]) == 0) {
+					pr_debug("Authentication successful\n");
+					local->card_status = CARD_AUTH_COMPLETE;
+					associate(local);
+					local->authentication_state =
+					    AUTHENTICATED;
+				} else {
+					pr_debug("Authentication refused\n");
+					local->card_status = CARD_AUTH_REFUSED;
+					join_net(&local->timer);
+					local->authentication_state =
+					    UNAUTHENTICATED;
+				}
+			}
+		}
+	}
+
+} /* end rx_authenticate */
+
+/*===========================================================================*/
+static void associate(ray_dev_t *local)
+{
+	struct ccs __iomem *pccs;
+	struct pcmcia_device *link = local->finder;
+	struct net_device *dev = link->priv;
+	int ccsindex;
+	if (!(pcmcia_dev_present(link))) {
+		dev_dbg(&link->dev, "ray_cs associate - device not present\n");
+		return;
+	}
+	/* If no tx buffers available, return */
+	if ((ccsindex = get_free_ccs(local)) < 0) {
+/* TBD should never be here but... what if we are? */
+		dev_dbg(&link->dev, "ray_cs associate - No free ccs\n");
+		return;
+	}
+	dev_dbg(&link->dev, "ray_cs Starting association with access point\n");
+	pccs = ccs_base(local) + ccsindex;
+	/* fill in the CCS */
+	writeb(CCS_START_ASSOCIATION, &pccs->cmd);
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		dev_dbg(&link->dev, "ray_cs associate failed - ECF not ready for intr\n");
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+
+		del_timer(&local->timer);
+		local->timer.expires = jiffies + HZ * 2;
+		local->timer.function = join_net;
+		add_timer(&local->timer);
+		local->card_status = CARD_ASSOC_FAILED;
+		return;
+	}
+	if (!sniffer)
+		netif_start_queue(dev);
+
+} /* end associate */
+
+/*===========================================================================*/
+static void rx_deauthenticate(ray_dev_t *local, struct rcs __iomem *prcs,
+			      unsigned int pkt_addr, int rx_len)
+{
+/*  UCHAR buff[256];
+    struct ray_rx_msg *msg = (struct ray_rx_msg *) buff;
+*/
+	pr_debug("Deauthentication frame received\n");
+	local->authentication_state = UNAUTHENTICATED;
+	/* Need to reauthenticate or rejoin depending on reason code */
+/*  copy_from_rx_buff(local, buff, pkt_addr, rx_len & 0xff);
+ */
+}
+
+/*===========================================================================*/
+static void clear_interrupt(ray_dev_t *local)
+{
+	writeb(0, local->amem + CIS_OFFSET + HCS_INTR_OFFSET);
+}
+
+/*===========================================================================*/
+#ifdef CONFIG_PROC_FS
+#define MAXDATA (PAGE_SIZE - 80)
+
+static const char *card_status[] = {
+	"Card inserted - uninitialized",	/* 0 */
+	"Card not downloaded",			/* 1 */
+	"Waiting for download parameters",	/* 2 */
+	"Card doing acquisition",		/* 3 */
+	"Acquisition complete",			/* 4 */
+	"Authentication complete",		/* 5 */
+	"Association complete",			/* 6 */
+	"???", "???", "???", "???",		/* 7 8 9 10 undefined */
+	"Card init error",			/* 11 */
+	"Download parameters error",		/* 12 */
+	"???",					/* 13 */
+	"Acquisition failed",			/* 14 */
+	"Authentication refused",		/* 15 */
+	"Association failed"			/* 16 */
+};
+
+static const char *nettype[] = { "Adhoc", "Infra " };
+static const char *framing[] = { "Encapsulation", "Translation" }
+
+;
+/*===========================================================================*/
+static int ray_cs_proc_show(struct seq_file *m, void *v)
+{
+/* Print current values which are not available via other means
+ * eg ifconfig
+ */
+	int i;
+	struct pcmcia_device *link;
+	struct net_device *dev;
+	ray_dev_t *local;
+	UCHAR *p;
+	struct freq_hop_element *pfh;
+	UCHAR c[33];
+
+	link = this_device;
+	if (!link)
+		return 0;
+	dev = (struct net_device *)link->priv;
+	if (!dev)
+		return 0;
+	local = netdev_priv(dev);
+	if (!local)
+		return 0;
+
+	seq_puts(m, "Raylink Wireless LAN driver status\n");
+	seq_printf(m, "%s\n", rcsid);
+	/* build 4 does not report version, and field is 0x55 after memtest */
+	seq_puts(m, "Firmware version     = ");
+	if (local->fw_ver == 0x55)
+		seq_puts(m, "4 - Use dump_cis for more details\n");
+	else
+		seq_printf(m, "%2d.%02d.%02d\n",
+			   local->fw_ver, local->fw_bld, local->fw_var);
+
+	for (i = 0; i < 32; i++)
+		c[i] = local->sparm.b5.a_current_ess_id[i];
+	c[32] = 0;
+	seq_printf(m, "%s network ESSID = \"%s\"\n",
+		   nettype[local->sparm.b5.a_network_type], c);
+
+	p = local->bss_id;
+	seq_printf(m, "BSSID                = %pM\n", p);
+
+	seq_printf(m, "Country code         = %d\n",
+		   local->sparm.b5.a_curr_country_code);
+
+	i = local->card_status;
+	if (i < 0)
+		i = 10;
+	if (i > 16)
+		i = 10;
+	seq_printf(m, "Card status          = %s\n", card_status[i]);
+
+	seq_printf(m, "Framing mode         = %s\n", framing[translate]);
+
+	seq_printf(m, "Last pkt signal lvl  = %d\n", local->last_rsl);
+
+	if (local->beacon_rxed) {
+		/* Pull some fields out of last beacon received */
+		seq_printf(m, "Beacon Interval      = %d Kus\n",
+			   local->last_bcn.beacon_intvl[0]
+			   + 256 * local->last_bcn.beacon_intvl[1]);
+
+		p = local->last_bcn.elements;
+		if (p[0] == C_ESSID_ELEMENT_ID)
+			p += p[1] + 2;
+		else {
+			seq_printf(m,
+				   "Parse beacon failed at essid element id = %d\n",
+				   p[0]);
+			return 0;
+		}
+
+		if (p[0] == C_SUPPORTED_RATES_ELEMENT_ID) {
+			seq_puts(m, "Supported rate codes = ");
+			for (i = 2; i < p[1] + 2; i++)
+				seq_printf(m, "0x%02x ", p[i]);
+			seq_putc(m, '\n');
+			p += p[1] + 2;
+		} else {
+			seq_puts(m, "Parse beacon failed at rates element\n");
+			return 0;
+		}
+
+		if (p[0] == C_FH_PARAM_SET_ELEMENT_ID) {
+			pfh = (struct freq_hop_element *)p;
+			seq_printf(m, "Hop dwell            = %d Kus\n",
+				   pfh->dwell_time[0] +
+				   256 * pfh->dwell_time[1]);
+			seq_printf(m, "Hop set              = %d\n",
+				   pfh->hop_set);
+			seq_printf(m, "Hop pattern          = %d\n",
+				   pfh->hop_pattern);
+			seq_printf(m, "Hop index            = %d\n",
+				   pfh->hop_index);
+			p += p[1] + 2;
+		} else {
+			seq_puts(m,
+				 "Parse beacon failed at FH param element\n");
+			return 0;
+		}
+	} else {
+		seq_puts(m, "No beacons received\n");
+	}
+	return 0;
+}
+#endif
+/*===========================================================================*/
+static int build_auth_frame(ray_dev_t *local, UCHAR *dest, int auth_type)
+{
+	int addr;
+	struct ccs __iomem *pccs;
+	struct tx_msg __iomem *ptx;
+	int ccsindex;
+
+	/* If no tx buffers available, return */
+	if ((ccsindex = get_free_tx_ccs(local)) < 0) {
+		pr_debug("ray_cs send authenticate - No free tx ccs\n");
+		return -1;
+	}
+
+	pccs = ccs_base(local) + ccsindex;
+
+	/* Address in card space */
+	addr = TX_BUF_BASE + (ccsindex << 11);
+	/* fill in the CCS */
+	writeb(CCS_TX_REQUEST, &pccs->cmd);
+	writeb(addr >> 8, pccs->var.tx_request.tx_data_ptr);
+	writeb(0x20, pccs->var.tx_request.tx_data_ptr + 1);
+	writeb(TX_AUTHENTICATE_LENGTH_MSB, pccs->var.tx_request.tx_data_length);
+	writeb(TX_AUTHENTICATE_LENGTH_LSB,
+	       pccs->var.tx_request.tx_data_length + 1);
+	writeb(0, &pccs->var.tx_request.pow_sav_mode);
+
+	ptx = local->sram + addr;
+	/* fill in the mac header */
+	writeb(PROTOCOL_VER | AUTHENTIC_TYPE, &ptx->mac.frame_ctl_1);
+	writeb(0, &ptx->mac.frame_ctl_2);
+
+	memcpy_toio(ptx->mac.addr_1, dest, ADDRLEN);
+	memcpy_toio(ptx->mac.addr_2, local->sparm.b4.a_mac_addr, ADDRLEN);
+	memcpy_toio(ptx->mac.addr_3, local->bss_id, ADDRLEN);
+
+	/* Fill in msg body with protocol 00 00, sequence 01 00 ,status 00 00 */
+	memset_io(ptx->var, 0, 6);
+	writeb(auth_type & 0xff, ptx->var + 2);
+
+	/* Interrupt the firmware to process the command */
+	if (interrupt_ecf(local, ccsindex)) {
+		pr_debug(
+		      "ray_cs send authentication request failed - ECF not ready for intr\n");
+		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
+		return -1;
+	}
+	return 0;
+} /* End build_auth_frame */
+
+/*===========================================================================*/
+#ifdef CONFIG_PROC_FS
+static ssize_t ray_cs_essid_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *pos)
+{
+	static char proc_essid[33];
+	unsigned int len = count;
+
+	if (len > 32)
+		len = 32;
+	memset(proc_essid, 0, 33);
+	if (copy_from_user(proc_essid, buffer, len))
+		return -EFAULT;
+	essid = proc_essid;
+	return count;
+}
+
+static const struct proc_ops ray_cs_essid_proc_ops = {
+	.proc_write	= ray_cs_essid_proc_write,
+	.proc_lseek	= noop_llseek,
+};
+
+static ssize_t int_proc_write(struct file *file, const char __user *buffer,
+			      size_t count, loff_t *pos)
+{
+	static char proc_number[10];
+	char *p;
+	int nr, len;
+
+	if (!count)
+		return 0;
+
+	if (count > 9)
+		return -EINVAL;
+	if (copy_from_user(proc_number, buffer, count))
+		return -EFAULT;
+	p = proc_number;
+	nr = 0;
+	len = count;
+	do {
+		unsigned int c = *p - '0';
+		if (c > 9)
+			return -EINVAL;
+		nr = nr * 10 + c;
+		p++;
+	} while (--len);
+	*(int *)pde_data(file_inode(file)) = nr;
+	return count;
+}
+
+static const struct proc_ops int_proc_ops = {
+	.proc_write	= int_proc_write,
+	.proc_lseek	= noop_llseek,
+};
+#endif
+
+static const struct pcmcia_device_id ray_ids[] = {
+	PCMCIA_DEVICE_MANF_CARD(0x01a6, 0x0000),
+	PCMCIA_DEVICE_NULL,
+};
+
+MODULE_DEVICE_TABLE(pcmcia, ray_ids);
+
+static struct pcmcia_driver ray_driver = {
+	.owner = THIS_MODULE,
+	.name = "ray_cs",
+	.probe = ray_probe,
+	.remove = ray_detach,
+	.id_table = ray_ids,
+	.suspend = ray_suspend,
+	.resume = ray_resume,
+};
+
+static int __init init_ray_cs(void)
+{
+	int rc;
+
+	pr_debug("%s\n", rcsid);
+	rc = pcmcia_register_driver(&ray_driver);
+	pr_debug("raylink init_module register_pcmcia_driver returns 0x%x\n",
+	      rc);
+	if (rc)
+		return rc;
+
+#ifdef CONFIG_PROC_FS
+	proc_mkdir("driver/ray_cs", NULL);
+
+	proc_create_single("driver/ray_cs/ray_cs", 0, NULL, ray_cs_proc_show);
+	proc_create("driver/ray_cs/essid", 0200, NULL, &ray_cs_essid_proc_ops);
+	proc_create_data("driver/ray_cs/net_type", 0200, NULL, &int_proc_ops,
+			 &net_type);
+	proc_create_data("driver/ray_cs/translate", 0200, NULL, &int_proc_ops,
+			 &translate);
+#endif
+	translate = !!translate;
+	return 0;
+} /* init_ray_cs */
+
+/*===========================================================================*/
+
+static void __exit exit_ray_cs(void)
+{
+	pr_debug("ray_cs: cleanup_module\n");
+
+#ifdef CONFIG_PROC_FS
+	remove_proc_subtree("driver/ray_cs", NULL);
+#endif
+
+	pcmcia_unregister_driver(&ray_driver);
+} /* exit_ray_cs */
+
+module_init(init_ray_cs);
+module_exit(exit_ray_cs);
+
+/*===========================================================================*/
diff --git a/drivers/net/wireless/legacy/ray_cs.h b/drivers/net/wireless/legacy/ray_cs.h
new file mode 100644
index 000000000000..0609d8625019
--- /dev/null
+++ b/drivers/net/wireless/legacy/ray_cs.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Raytheon wireless LAN PCMCIA card driver for Linux 
+   A  PCMCIA client driver for the Raylink wireless network card
+   Written by Corey Thomas
+*/
+
+#ifndef _RAY_CS_H_
+#define _RAY_CS_H_
+
+struct beacon_rx {
+    struct mac_header mac;
+    UCHAR timestamp[8];
+    UCHAR beacon_intvl[2];
+    UCHAR capability[2];
+    UCHAR elements[sizeof(struct essid_element) 
+                  + sizeof(struct rates_element)
+                  + sizeof(struct freq_hop_element) 
+                  + sizeof(struct japan_call_sign_element)
+                  + sizeof(struct tim_element)];
+};
+
+/* Return values for get_free{,_tx}_ccs */
+#define ECCSFULL  (-1)
+#define ECCSBUSY  (-2)
+#define ECARDGONE (-3)
+
+typedef struct ray_dev_t {
+    int card_status;
+    int authentication_state;
+    void __iomem *sram;            /* pointer to beginning of shared RAM     */
+    void __iomem *amem;            /* pointer to attribute mem window        */
+    void __iomem *rmem;            /* pointer to receive buffer window       */
+    struct pcmcia_device *finder;            /* pointer back to struct pcmcia_device for card    */
+    struct timer_list timer;
+    unsigned long tx_ccs_lock;
+    unsigned long ccs_lock;
+    int   dl_param_ccs;
+    union {
+        struct b4_startup_params b4;
+        struct b5_startup_params b5;
+    } sparm;
+    int timeout_flag;
+    UCHAR supported_rates[8];
+    UCHAR japan_call_sign[12];
+    struct startup_res_6 startup_res;
+    int num_multi;
+    /* Network parameters from start/join */
+    UCHAR bss_id[6];
+    UCHAR auth_id[6];
+    UCHAR net_default_tx_rate;
+    UCHAR encryption;
+    struct net_device_stats stats;
+
+    UCHAR net_type;
+    UCHAR sta_type;
+    UCHAR fw_ver;
+    UCHAR fw_bld;
+    UCHAR fw_var;
+    UCHAR ASIC_version;
+    UCHAR assoc_id[2];
+    UCHAR tib_length;
+    UCHAR last_rsl;
+    int beacon_rxed;
+    struct beacon_rx last_bcn;
+    iw_stats	wstats;		/* Wireless specific stats */
+#ifdef WIRELESS_SPY
+    struct iw_spy_data		spy_data;
+    struct iw_public_data	wireless_data;
+#endif	/* WIRELESS_SPY */
+
+} ray_dev_t;
+/*****************************************************************************/
+
+#endif /* _RAY_CS_H_ */
diff --git a/drivers/net/wireless/legacy/rayctl.h b/drivers/net/wireless/legacy/rayctl.h
new file mode 100644
index 000000000000..2b0f332043d7
--- /dev/null
+++ b/drivers/net/wireless/legacy/rayctl.h
@@ -0,0 +1,734 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _RAYCTL_H_
+#define _RAYCTL_H_
+
+typedef unsigned char UCHAR;
+
+/****** IEEE 802.11 constants ************************************************/
+#define ADDRLEN           6
+/* Frame control 1 bit fields */
+#define PROTOCOL_VER      0x00
+#define DATA_TYPE         0x08
+#define ASSOC_REQ_TYPE    0x00
+#define ASSOC_RESP_TYPE   0x10
+#define REASSOC_REQ_TYPE  0x20
+#define REASSOC_RESP_TYPE 0x30
+#define NULL_MSG_TYPE     0x48
+#define BEACON_TYPE       0x80
+#define DISASSOC_TYPE     0xA0
+#define PSPOLL_TYPE       0xA4
+#define AUTHENTIC_TYPE    0xB0
+#define DEAUTHENTIC_TYPE  0xC0
+/* Frame control 2 bit fields */
+#define FC2_TO_DS         0x01
+#define FC2_FROM_DS       0x02
+#define FC2_MORE_FRAG     0x04
+#define FC2_RETRY         0x08
+#define FC2_PSM           0x10
+#define FC2_MORE_DATA     0x20
+#define FC2_WEP           0x40
+#define FC2_ORDER         0x80
+/*****************************************************************************/
+/* 802.11 element ID's and lengths */
+#define C_BP_CAPABILITY_ESS             0x01
+#define C_BP_CAPABILITY_IBSS            0x02
+#define C_BP_CAPABILITY_CF_POLLABLE     0x04
+#define C_BP_CAPABILITY_CF_POLL_REQUEST 0x08
+#define C_BP_CAPABILITY_PRIVACY         0x10
+
+#define C_ESSID_ELEMENT_ID               0
+#define C_ESSID_ELEMENT_MAX_LENGTH       32
+
+#define C_SUPPORTED_RATES_ELEMENT_ID     1
+#define C_SUPPORTED_RATES_ELEMENT_LENGTH 2
+
+#define C_FH_PARAM_SET_ELEMENT_ID        2
+#define C_FH_PARAM_SET_ELEMENT_LNGTH     5
+
+#define C_CF_PARAM_SET_ELEMENT_ID        4
+#define C_CF_PARAM_SET_ELEMENT_LNGTH     6
+
+#define C_TIM_ELEMENT_ID                 5
+#define C_TIM_BITMAP_LENGTH            251
+#define C_TIM_BMCAST_BIT              0x01
+
+#define C_IBSS_ELEMENT_ID                6
+#define C_IBSS_ELEMENT_LENGTH            2
+
+#define C_JAPAN_CALL_SIGN_ELEMENT_ID    51
+#define C_JAPAN_CALL_SIGN_ELEMENT_LNGTH 12
+
+#define C_DISASSOC_REASON_CODE_LEN       2
+#define C_DISASSOC_REASON_CODE_DEFAULT   8
+
+#define C_CRC_LEN                        4
+#define C_NUM_SUPPORTED_RATES            8 
+/****** IEEE 802.11 mac header for type data packets *************************/
+struct mac_header {
+  UCHAR frame_ctl_1;                          
+  UCHAR frame_ctl_2;
+  UCHAR duration_lsb;
+  UCHAR duration_msb;
+  UCHAR addr_1[ADDRLEN];
+  UCHAR addr_2[ADDRLEN];
+  UCHAR addr_3[ADDRLEN];
+  UCHAR seq_frag_num[2];
+/*  UCHAR addr_4[ADDRLEN]; *//* only present for AP to AP (TO DS and FROM DS */
+};
+/****** IEEE 802.11 frame element structures *********************************/
+struct essid_element
+{
+  UCHAR id;
+  UCHAR length;
+  UCHAR text[C_ESSID_ELEMENT_MAX_LENGTH];
+};
+struct rates_element
+{
+  UCHAR id;
+  UCHAR length;
+  UCHAR value[8];
+};
+struct freq_hop_element
+{
+  UCHAR id;
+  UCHAR length;
+  UCHAR dwell_time[2];
+  UCHAR hop_set;
+  UCHAR hop_pattern;
+  UCHAR hop_index;
+};
+struct tim_element
+{
+  UCHAR id;
+  UCHAR length;
+  UCHAR dtim_count;
+  UCHAR dtim_period;    
+  UCHAR bitmap_control;
+  UCHAR tim[C_TIM_BITMAP_LENGTH];
+};
+struct ibss_element
+{
+  UCHAR id;
+  UCHAR length;
+  UCHAR atim_window[2];
+};
+struct japan_call_sign_element
+{
+  UCHAR id;
+  UCHAR length;
+  UCHAR call_sign[12];
+};
+/****** Beacon message structures ********************************************/
+/* .elements is a large lump of max size because elements are variable size  */
+struct infra_beacon
+{
+    UCHAR timestamp[8];
+    UCHAR beacon_intvl[2];
+    UCHAR capability[2];
+    UCHAR elements[sizeof(struct essid_element) 
+                  + sizeof(struct rates_element)
+                  + sizeof(struct freq_hop_element) 
+                  + sizeof(struct japan_call_sign_element)
+                  + sizeof(struct tim_element)];
+};
+struct adhoc_beacon
+{
+    UCHAR timestamp[8];
+    UCHAR beacon_intvl[2];
+    UCHAR capability[2];
+    UCHAR elements[sizeof(struct essid_element) 
+                  + sizeof(struct rates_element)
+                  + sizeof(struct freq_hop_element) 
+                  + sizeof(struct japan_call_sign_element)
+                  + sizeof(struct ibss_element)];
+};
+/*****************************************************************************/
+/*****************************************************************************/
+/* #define C_MAC_HDR_2_WEP 0x40 */
+/* TX/RX CCS constants */
+#define TX_HEADER_LENGTH 0x1C
+#define RX_MAC_HEADER_LENGTH 0x18
+#define TX_AUTHENTICATE_LENGTH (TX_HEADER_LENGTH + 6)
+#define TX_AUTHENTICATE_LENGTH_MSB (TX_AUTHENTICATE_LENGTH >> 8)
+#define TX_AUTHENTICATE_LENGTH_LSB (TX_AUTHENTICATE_LENGTH & 0xff)
+#define TX_DEAUTHENTICATE_LENGTH (TX_HEADER_LENGTH + 2)
+#define TX_DEAUTHENTICATE_LENGTH_MSB (TX_AUTHENTICATE_LENGTH >> 8)
+#define TX_DEAUTHENTICATE_LENGTH_LSB (TX_AUTHENTICATE_LENGTH & 0xff)
+#define FCS_LEN           4
+
+#define ADHOC                 0
+#define INFRA                 1
+
+#define TYPE_STA              0
+#define TYPE_AP               1
+
+#define PASSIVE_SCAN          1
+#define ACTIVE_SCAN           1
+
+#define PSM_CAM               0
+
+/* Country codes */
+#define USA                   1
+#define EUROPE                2
+#define JAPAN                 3
+#define KOREA                 4
+#define SPAIN                 5
+#define FRANCE                6
+#define ISRAEL                7
+#define AUSTRALIA             8
+#define JAPAN_TEST            9
+
+/* Hop pattern lengths */
+#define USA_HOP_MOD          79 
+#define EUROPE_HOP_MOD       79 
+#define JAPAN_HOP_MOD        23
+#define KOREA_HOP_MOD        23
+#define SPAIN_HOP_MOD        27
+#define FRANCE_HOP_MOD       35
+#define ISRAEL_HOP_MOD       35
+#define AUSTRALIA_HOP_MOD    47
+#define JAPAN_TEST_HOP_MOD   23
+
+#define ESSID_SIZE           32
+/**********************************************************************/
+/* CIS Register Constants */
+#define CIS_OFFSET             0x0f00
+/* Configuration Option Register (0x0F00) */
+#define COR_OFFSET             0x00
+#define COR_SOFT_RESET         0x80
+#define COR_LEVEL_IRQ          0x40
+#define COR_CONFIG_NUM         0x01
+#define COR_DEFAULT            (COR_LEVEL_IRQ | COR_CONFIG_NUM)
+
+/* Card Configuration and Status Register (0x0F01) */
+#define CCSR_OFFSET            0x01
+#define CCSR_HOST_INTR_PENDING 0x01
+#define CCSR_POWER_DOWN        0x04
+
+/* HCS Interrupt Register (0x0F05) */
+#define HCS_INTR_OFFSET        0x05
+/* #define HCS_INTR_OFFSET        0x0A */
+#define HCS_INTR_CLEAR         0x00
+
+/* ECF Interrupt Register (0x0F06) */
+#define ECF_INTR_OFFSET        0x06
+/* #define ECF_INTR_OFFSET        0x0C */
+#define ECF_INTR_SET           0x01
+
+/* Authorization Register 0 (0x0F08) */
+#define AUTH_0_ON              0x57
+
+/* Authorization Register 1 (0x0F09) */
+#define AUTH_1_ON              0x82
+
+/* Program Mode Register (0x0F0A) */
+#define PC2PM                  0x02
+#define PC2CAL                 0x10
+#define PC2MLSE                0x20
+
+/* PC Test Mode Register (0x0F0B) */
+#define PC_TEST_MODE           0x08
+
+/* Frequency Control Word (0x0F10) */
+/* Range 0x02 - 0xA6 */
+
+/* Test Mode Control 1-4 (0x0F14 - 0x0F17) */
+
+/**********************************************************************/
+
+/* Shared RAM Area */
+#define SCB_BASE               0x0000
+#define STATUS_BASE            0x0100
+#define HOST_TO_ECF_BASE       0x0200
+#define ECF_TO_HOST_BASE       0x0300
+#define CCS_BASE               0x0400
+#define RCS_BASE               0x0800
+#define INFRA_TIM_BASE         0x0C00
+#define SSID_LIST_BASE         0x0D00
+#define TX_BUF_BASE            0x1000
+#define RX_BUF_BASE            0x8000
+
+#define NUMBER_OF_CCS    64
+#define NUMBER_OF_RCS    64
+/*#define NUMBER_OF_TX_CCS 14 */
+#define NUMBER_OF_TX_CCS 14
+
+#define TX_BUF_SIZE      (2048 - sizeof(struct tx_msg))
+#define RX_BUFF_END      0x3FFF
+/* Values for buffer_status */
+#define CCS_BUFFER_FREE       0
+#define CCS_BUFFER_BUSY       1
+#define CCS_COMMAND_COMPLETE  2
+#define CCS_COMMAND_FAILED    3
+
+/* Values for cmd */
+#define CCS_DOWNLOAD_STARTUP_PARAMS    1
+#define CCS_UPDATE_PARAMS              2
+#define CCS_REPORT_PARAMS              3
+#define CCS_UPDATE_MULTICAST_LIST      4
+#define CCS_UPDATE_POWER_SAVINGS_MODE  5
+#define CCS_START_NETWORK              6
+#define CCS_JOIN_NETWORK               7
+#define CCS_START_ASSOCIATION          8
+#define CCS_TX_REQUEST                 9
+#define CCS_TEST_MEMORY              0xa
+#define CCS_SHUTDOWN                 0xb
+#define CCS_DUMP_MEMORY              0xc
+#define CCS_START_TIMER              0xe
+#define CCS_LAST_CMD                 CCS_START_TIMER
+
+/* Values for link field */
+#define CCS_END_LIST                 0xff
+
+/* values for buffer_status field */
+#define RCS_BUFFER_FREE       0
+#define RCS_BUFFER_BUSY       1
+#define RCS_COMPLETE          2
+#define RCS_FAILED            3
+#define RCS_BUFFER_RELEASE    0xFF
+
+/* values for interrupt_id field */
+#define PROCESS_RX_PACKET           0x80 /* */
+#define REJOIN_NET_COMPLETE         0x81 /* RCS ID: Rejoin Net Complete */
+#define ROAMING_INITIATED           0x82 /* RCS ID: Roaming Initiated   */
+#define JAPAN_CALL_SIGN_RXD         0x83 /* RCS ID: New Japan Call Sign */
+
+/*****************************************************************************/
+/* Memory types for dump memory command */
+#define C_MEM_PROG  0
+#define C_MEM_XDATA 1
+#define C_MEM_SFR   2
+#define C_MEM_IDATA 3
+
+/*** Return values for hw_xmit **********/
+#define XMIT_OK        (0)
+#define XMIT_MSG_BAD   (-1)
+#define XMIT_NO_CCS    (-2)
+#define XMIT_NO_INTR   (-3)
+#define XMIT_NEED_AUTH (-4)
+
+/*** Values for card status */
+#define CARD_INSERTED       (0)
+
+#define CARD_AWAITING_PARAM (1)
+#define CARD_INIT_ERROR     (11)
+
+#define CARD_DL_PARAM       (2)
+#define CARD_DL_PARAM_ERROR (12)
+
+#define CARD_DOING_ACQ      (3)
+
+#define CARD_ACQ_COMPLETE   (4)
+#define CARD_ACQ_FAILED     (14)
+
+#define CARD_AUTH_COMPLETE  (5)
+#define CARD_AUTH_REFUSED   (15)
+
+#define CARD_ASSOC_COMPLETE (6)
+#define CARD_ASSOC_FAILED   (16)
+
+/*** Values for authentication_state ***********************************/
+#define UNAUTHENTICATED     (0)
+#define AWAITING_RESPONSE   (1)
+#define AUTHENTICATED       (2)
+#define NEED_TO_AUTH        (3)
+
+/*** Values for authentication type ************************************/
+#define OPEN_AUTH_REQUEST   (1)
+#define OPEN_AUTH_RESPONSE  (2)
+#define BROADCAST_DEAUTH    (0xc0)
+/*** Values for timer functions ****************************************/
+#define TODO_NOTHING              (0)
+#define TODO_VERIFY_DL_START      (-1)
+#define TODO_START_NET            (-2)
+#define TODO_JOIN_NET             (-3)
+#define TODO_AUTHENTICATE_TIMEOUT (-4)
+#define TODO_SEND_CCS             (-5)
+/***********************************************************************/
+/* Parameter passing structure for update/report parameter CCS's */
+struct object_id {
+    void          *object_addr;
+    unsigned char object_length;
+};
+
+#define OBJID_network_type            0
+#define OBJID_acting_as_ap_status     1
+#define OBJID_current_ess_id          2
+#define OBJID_scanning_mode           3
+#define OBJID_power_mgt_state         4
+#define OBJID_mac_address             5
+#define OBJID_frag_threshold          6
+#define OBJID_hop_time                7
+#define OBJID_beacon_period           8
+#define OBJID_dtim_period             9
+#define OBJID_retry_max              10
+#define OBJID_ack_timeout            11
+#define OBJID_sifs                   12
+#define OBJID_difs                   13
+#define OBJID_pifs                   14
+#define OBJID_rts_threshold          15
+#define OBJID_scan_dwell_time        16
+#define OBJID_max_scan_dwell_time    17
+#define OBJID_assoc_resp_timeout     18
+#define OBJID_adhoc_scan_cycle_max   19
+#define OBJID_infra_scan_cycle_max   20
+#define OBJID_infra_super_cycle_max  21
+#define OBJID_promiscuous_mode       22
+#define OBJID_unique_word            23
+#define OBJID_slot_time              24
+#define OBJID_roaming_low_snr        25
+#define OBJID_low_snr_count_thresh   26
+#define OBJID_infra_missed_bcn       27
+#define OBJID_adhoc_missed_bcn       28
+#define OBJID_curr_country_code      29
+#define OBJID_hop_pattern            30
+#define OBJID_reserved               31
+#define OBJID_cw_max_msb             32
+#define OBJID_cw_min_msb             33
+#define OBJID_noise_filter_gain      34
+#define OBJID_noise_limit_offset     35
+#define OBJID_det_rssi_thresh_offset 36
+#define OBJID_med_busy_thresh_offset 37
+#define OBJID_det_sync_thresh        38
+#define OBJID_test_mode              39
+#define OBJID_test_min_chan_num      40
+#define OBJID_test_max_chan_num      41
+#define OBJID_allow_bcast_ID_prbrsp  42
+#define OBJID_privacy_must_start     43
+#define OBJID_privacy_can_join       44
+#define OBJID_basic_rate_set         45
+
+/**** Configuration/Status/Control Area ***************************/
+/*    System Control Block (SCB) Area
+ *    Located at Shared RAM offset 0
+ */
+struct scb {
+    UCHAR ccs_index;
+    UCHAR rcs_index;
+};
+
+/****** Status area at Shared RAM offset 0x0100 ******************************/
+struct status {
+    UCHAR mrx_overflow_for_host;         /* 0=ECF may write, 1=host may write*/
+    UCHAR mrx_checksum_error_for_host;   /* 0=ECF may write, 1=host may write*/
+    UCHAR rx_hec_error_for_host;         /* 0=ECF may write, 1=host may write*/
+    UCHAR reserved1;
+    short mrx_overflow;                  /* ECF increments on rx overflow    */
+    short mrx_checksum_error;            /* ECF increments on rx CRC error   */
+    short rx_hec_error;                  /* ECF incs on mac header CRC error */
+    UCHAR rxnoise;                       /* Average RSL measurement          */
+};
+
+/****** Host-to-ECF Data Area at Shared RAM offset 0x200 *********************/
+struct host_to_ecf_area {
+    
+};
+
+/****** ECF-to-Host Data Area at Shared RAM offset 0x0300 ********************/
+struct startup_res_518 {
+    UCHAR startup_word;
+    UCHAR station_addr[ADDRLEN];
+    UCHAR calc_prog_chksum;
+    UCHAR calc_cis_chksum;
+    UCHAR ecf_spare[7];
+    UCHAR japan_call_sign[12];
+};
+
+struct startup_res_6 {
+    UCHAR startup_word;
+    UCHAR station_addr[ADDRLEN];
+    UCHAR reserved;
+    UCHAR supp_rates[8];
+    UCHAR japan_call_sign[12];
+    UCHAR calc_prog_chksum;
+    UCHAR calc_cis_chksum;
+    UCHAR firmware_version[3];
+    UCHAR asic_version;
+    UCHAR tib_length;
+};
+
+struct start_join_net_params {
+    UCHAR net_type;
+    UCHAR ssid[ESSID_SIZE];
+    UCHAR reserved;
+    UCHAR privacy_can_join;
+};
+
+/****** Command Control Structure area at Shared ram offset 0x0400 ***********/
+/* Structures for command specific parameters (ccs.var) */
+struct update_param_cmd {
+    UCHAR object_id;
+    UCHAR number_objects;
+    UCHAR failure_cause;
+};
+struct report_param_cmd {
+    UCHAR object_id;
+    UCHAR number_objects;
+    UCHAR failure_cause;
+    UCHAR length;
+};
+struct start_network_cmd {
+    UCHAR update_param;
+    UCHAR bssid[ADDRLEN];
+    UCHAR net_initiated;
+    UCHAR net_default_tx_rate;
+    UCHAR encryption;
+};
+struct join_network_cmd {
+    UCHAR update_param;
+    UCHAR bssid[ADDRLEN];
+    UCHAR net_initiated;
+    UCHAR net_default_tx_rate;
+    UCHAR encryption;
+};
+struct tx_requested_cmd {
+ 
+    UCHAR tx_data_ptr[2];
+    UCHAR tx_data_length[2];
+    UCHAR host_reserved[2];
+    UCHAR reserved[3];
+    UCHAR tx_rate;
+    UCHAR pow_sav_mode;
+    UCHAR retries;
+    UCHAR antenna;
+};
+struct tx_requested_cmd_4 {
+ 
+    UCHAR tx_data_ptr[2];
+    UCHAR tx_data_length[2];
+    UCHAR dest_addr[ADDRLEN];
+    UCHAR pow_sav_mode;
+    UCHAR retries;
+    UCHAR station_id;
+};
+struct memory_dump_cmd {
+    UCHAR memory_type;
+    UCHAR memory_ptr[2];
+    UCHAR length;
+};
+struct update_association_cmd {
+    UCHAR status;
+    UCHAR aid[2];
+};
+struct start_timer_cmd {
+    UCHAR duration[2];
+};
+
+struct ccs {
+    UCHAR buffer_status;                 /* 0 = buffer free, 1 = buffer busy */
+                                         /* 2 = command complete, 3 = failed */
+    UCHAR cmd;                           /* command to ECF                   */
+    UCHAR link;                          /* link to next CCS, FF=end of list */
+    /* command specific parameters      */
+    union {
+        char reserved[13];
+        struct update_param_cmd update_param;
+        struct report_param_cmd report_param;
+        UCHAR nummulticast;
+        UCHAR mode;
+        struct start_network_cmd start_network;
+        struct join_network_cmd join_network;
+        struct tx_requested_cmd tx_request;
+        struct memory_dump_cmd memory_dump;
+        struct update_association_cmd update_assoc;
+        struct start_timer_cmd start_timer;
+    } var;
+};
+
+/*****************************************************************************/
+/* Transmit buffer structures */
+struct tib_structure {
+    UCHAR ccs_index;
+    UCHAR psm;
+    UCHAR pass_fail;
+    UCHAR retry_count;
+    UCHAR max_retries;
+    UCHAR frags_remaining;
+    UCHAR no_rb;
+    UCHAR rts_reqd;
+    UCHAR csma_tx_cntrl_2;
+    UCHAR sifs_tx_cntrl_2;
+    UCHAR tx_dma_addr_1[2];
+    UCHAR tx_dma_addr_2[2];
+    UCHAR var_dur_2mhz[2];
+    UCHAR var_dur_1mhz[2];
+    UCHAR max_dur_2mhz[2];
+    UCHAR max_dur_1mhz[2];
+    UCHAR hdr_len;
+    UCHAR max_frag_len[2];
+    UCHAR var_len[2];
+    UCHAR phy_hdr_4;
+    UCHAR mac_hdr_1;
+    UCHAR mac_hdr_2;
+    UCHAR sid[2];
+};
+
+struct phy_header {
+    UCHAR sfd[2];
+    UCHAR hdr_3;
+    UCHAR hdr_4;
+};
+struct ray_rx_msg {
+    struct mac_header mac;
+	UCHAR	var[];
+};
+
+struct tx_msg {
+    struct tib_structure tib;
+    struct phy_header phy;
+    struct mac_header mac;
+    UCHAR  var[1];
+};
+
+/****** ECF Receive Control Structure (RCS) Area at Shared RAM offset 0x0800  */
+/* Structures for command specific parameters (rcs.var) */
+struct rx_packet_cmd {
+    UCHAR rx_data_ptr[2];
+    UCHAR rx_data_length[2];
+    UCHAR rx_sig_lev;
+    UCHAR next_frag_rcs_index;
+    UCHAR totalpacketlength[2];
+};
+struct rejoin_net_cmplt_cmd {
+    UCHAR reserved;
+    UCHAR bssid[ADDRLEN];
+};
+struct japan_call_sign_rxd {
+    UCHAR rxd_call_sign[8];
+    UCHAR reserved[5];
+};
+
+struct rcs {
+    UCHAR buffer_status;
+    UCHAR interrupt_id;
+    UCHAR link_field;
+    /* command specific parameters      */
+    union {
+        UCHAR reserved[13]; 
+        struct rx_packet_cmd rx_packet;
+        struct rejoin_net_cmplt_cmd rejoin_net_complete;
+        struct japan_call_sign_rxd japan_call_sign;
+    } var;
+};
+
+/****** Startup parameter structures for both versions of firmware ***********/
+struct b4_startup_params {
+    UCHAR a_network_type;                /* C_ADHOC, C_INFRA                 */
+    UCHAR a_acting_as_ap_status;         /* C_TYPE_STA, C_TYPE_AP            */
+    UCHAR a_current_ess_id[ESSID_SIZE];  /* Null terminated unless 32 long   */
+    UCHAR a_scanning_mode;               /* passive 0, active 1              */
+    UCHAR a_power_mgt_state;             /* CAM 0,                           */
+    UCHAR a_mac_addr[ADDRLEN];           /*                                  */
+    UCHAR a_frag_threshold[2];           /* 512                              */
+    UCHAR a_hop_time[2];                 /* 16k * 2**n, n=0-4 in Kus         */
+    UCHAR a_beacon_period[2];            /* n * a_hop_time  in Kus           */
+    UCHAR a_dtim_period;                 /* in beacons                       */
+    UCHAR a_retry_max;                   /*                                  */
+    UCHAR a_ack_timeout;                 /*                                  */
+    UCHAR a_sifs;                        /*                                  */
+    UCHAR a_difs;                        /*                                  */
+    UCHAR a_pifs;                        /*                                  */
+    UCHAR a_rts_threshold[2];            /*                                  */
+    UCHAR a_scan_dwell_time[2];          /*                                  */
+    UCHAR a_max_scan_dwell_time[2];      /*                                  */
+    UCHAR a_assoc_resp_timeout_thresh;   /*                                  */
+    UCHAR a_adhoc_scan_cycle_max;        /*                                  */
+    UCHAR a_infra_scan_cycle_max;        /*                                  */
+    UCHAR a_infra_super_scan_cycle_max;  /*                                  */
+    UCHAR a_promiscuous_mode;            /*                                  */
+    UCHAR a_unique_word[2];              /*                                  */
+    UCHAR a_slot_time;                   /*                                  */
+    UCHAR a_roaming_low_snr_thresh;      /*                                  */
+    UCHAR a_low_snr_count_thresh;        /*                                  */
+    UCHAR a_infra_missed_bcn_thresh;     /*                                  */
+    UCHAR a_adhoc_missed_bcn_thresh;     /*                                  */
+    UCHAR a_curr_country_code;           /* C_USA                            */
+    UCHAR a_hop_pattern;                 /*                                  */
+    UCHAR a_hop_pattern_length;          /*                                  */
+/* b4 - b5 differences start here */
+    UCHAR a_cw_max;                      /*                                  */
+    UCHAR a_cw_min;                      /*                                  */
+    UCHAR a_noise_filter_gain;           /*                                  */
+    UCHAR a_noise_limit_offset;          /*                                  */
+    UCHAR a_det_rssi_thresh_offset;      /*                                  */
+    UCHAR a_med_busy_thresh_offset;      /*                                  */
+    UCHAR a_det_sync_thresh;             /*                                  */
+    UCHAR a_test_mode;                   /*                                  */
+    UCHAR a_test_min_chan_num;           /*                                  */
+    UCHAR a_test_max_chan_num;           /*                                  */
+    UCHAR a_rx_tx_delay;                 /*                                  */
+    UCHAR a_current_bss_id[ADDRLEN];     /*                                  */
+    UCHAR a_hop_set;                     /*                                  */
+};
+struct b5_startup_params {
+    UCHAR a_network_type;                /* C_ADHOC, C_INFRA                 */
+    UCHAR a_acting_as_ap_status;         /* C_TYPE_STA, C_TYPE_AP            */
+    UCHAR a_current_ess_id[ESSID_SIZE];  /* Null terminated unless 32 long   */
+    UCHAR a_scanning_mode;               /* passive 0, active 1              */
+    UCHAR a_power_mgt_state;             /* CAM 0,                           */
+    UCHAR a_mac_addr[ADDRLEN];           /*                                  */
+    UCHAR a_frag_threshold[2];           /* 512                              */
+    UCHAR a_hop_time[2];                 /* 16k * 2**n, n=0-4 in Kus         */
+    UCHAR a_beacon_period[2];            /* n * a_hop_time  in Kus           */
+    UCHAR a_dtim_period;                 /* in beacons                       */
+    UCHAR a_retry_max;                   /* 4                                */
+    UCHAR a_ack_timeout;                 /*                                  */
+    UCHAR a_sifs;                        /*                                  */
+    UCHAR a_difs;                        /*                                  */
+    UCHAR a_pifs;                        /*                                  */
+    UCHAR a_rts_threshold[2];            /*                                  */
+    UCHAR a_scan_dwell_time[2];          /*                                  */
+    UCHAR a_max_scan_dwell_time[2];      /*                                  */
+    UCHAR a_assoc_resp_timeout_thresh;   /*                                  */
+    UCHAR a_adhoc_scan_cycle_max;        /*                                  */
+    UCHAR a_infra_scan_cycle_max;        /*                                  */
+    UCHAR a_infra_super_scan_cycle_max;  /*                                  */
+    UCHAR a_promiscuous_mode;            /*                                  */
+    UCHAR a_unique_word[2];              /*                                  */
+    UCHAR a_slot_time;                   /*                                  */
+    UCHAR a_roaming_low_snr_thresh;      /*                                  */
+    UCHAR a_low_snr_count_thresh;        /*                                  */
+    UCHAR a_infra_missed_bcn_thresh;     /*                                  */
+    UCHAR a_adhoc_missed_bcn_thresh;     /*                                  */
+    UCHAR a_curr_country_code;           /* C_USA                            */
+    UCHAR a_hop_pattern;                 /*                                  */
+    UCHAR a_hop_pattern_length;          /*                                  */
+/* b4 - b5 differences start here */
+    UCHAR a_cw_max[2];                   /*                                  */
+    UCHAR a_cw_min[2];                   /*                                  */
+    UCHAR a_noise_filter_gain;           /*                                  */
+    UCHAR a_noise_limit_offset;          /*                                  */
+    UCHAR a_det_rssi_thresh_offset;      /*                                  */
+    UCHAR a_med_busy_thresh_offset;      /*                                  */
+    UCHAR a_det_sync_thresh;             /*                                  */
+    UCHAR a_test_mode;                   /*                                  */
+    UCHAR a_test_min_chan_num;           /*                                  */
+    UCHAR a_test_max_chan_num;           /*                                  */
+    UCHAR a_allow_bcast_SSID_probe_rsp;
+    UCHAR a_privacy_must_start;
+    UCHAR a_privacy_can_join;
+    UCHAR a_basic_rate_set[8];
+};
+
+/*****************************************************************************/
+#define RAY_IOCG_PARMS (SIOCDEVPRIVATE)
+#define RAY_IOCS_PARMS (SIOCDEVPRIVATE + 1)
+#define RAY_DO_CMD     (SIOCDEVPRIVATE + 2)
+
+/****** ethernet <-> 802.11 translation **************************************/
+typedef struct snaphdr_t
+{
+  UCHAR   dsap;
+  UCHAR   ssap;
+  UCHAR   ctrl;
+  UCHAR   org[3];
+  UCHAR   ethertype[2];
+} snaphdr_t;
+
+#define BRIDGE_ENCAP  0xf80000
+#define RFC1042_ENCAP 0
+#define SNAP_ID       0x0003aaaa
+#define RAY_IPX_TYPE  0x8137
+#define APPLEARP_TYPE 0x80f3
+/*****************************************************************************/
+#endif /* _RAYCTL_H_ */
diff --git a/drivers/net/wireless/legacy/rndis_wlan.c b/drivers/net/wireless/legacy/rndis_wlan.c
new file mode 100644
index 000000000000..bf72e5fd39cf
--- /dev/null
+++ b/drivers/net/wireless/legacy/rndis_wlan.c
@@ -0,0 +1,3760 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for RNDIS based wireless USB devices.
+ *
+ * Copyright (C) 2007 by Bjorge Dijkstra <bjd@jooz.net>
+ * Copyright (C) 2008-2009 by Jussi Kivilinna <jussi.kivilinna@iki.fi>
+ *
+ *  Portions of this file are based on NDISwrapper project,
+ *  Copyright (C) 2003-2005 Pontus Fuchs, Giridhar Pemmasani
+ *  http://ndiswrapper.sourceforge.net/
+ */
+
+// #define	DEBUG			// error path messages, extra info
+// #define	VERBOSE			// more; success messages
+
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/workqueue.h>
+#include <linux/mutex.h>
+#include <linux/mii.h>
+#include <linux/usb.h>
+#include <linux/usb/cdc.h>
+#include <linux/ieee80211.h>
+#include <linux/if_arp.h>
+#include <linux/ctype.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <net/cfg80211.h>
+#include <linux/usb/usbnet.h>
+#include <linux/usb/rndis_host.h>
+
+
+/* NOTE: All these are settings for Broadcom chipset */
+static char modparam_country[4] = "EU";
+module_param_string(country, modparam_country, 4, 0444);
+MODULE_PARM_DESC(country, "Country code (ISO 3166-1 alpha-2), default: EU");
+
+static int modparam_frameburst = 1;
+module_param_named(frameburst, modparam_frameburst, int, 0444);
+MODULE_PARM_DESC(frameburst, "enable frame bursting (default: on)");
+
+static int modparam_afterburner = 0;
+module_param_named(afterburner, modparam_afterburner, int, 0444);
+MODULE_PARM_DESC(afterburner,
+	"enable afterburner aka '125 High Speed Mode' (default: off)");
+
+static int modparam_power_save = 0;
+module_param_named(power_save, modparam_power_save, int, 0444);
+MODULE_PARM_DESC(power_save,
+	"set power save mode: 0=off, 1=on, 2=fast (default: off)");
+
+static int modparam_power_output = 3;
+module_param_named(power_output, modparam_power_output, int, 0444);
+MODULE_PARM_DESC(power_output,
+	"set power output: 0=25%, 1=50%, 2=75%, 3=100% (default: 100%)");
+
+static int modparam_roamtrigger = -70;
+module_param_named(roamtrigger, modparam_roamtrigger, int, 0444);
+MODULE_PARM_DESC(roamtrigger,
+	"set roaming dBm trigger: -80=optimize for distance, "
+				"-60=bandwidth (default: -70)");
+
+static int modparam_roamdelta = 1;
+module_param_named(roamdelta, modparam_roamdelta, int, 0444);
+MODULE_PARM_DESC(roamdelta,
+	"set roaming tendency: 0=aggressive, 1=moderate, "
+				"2=conservative (default: moderate)");
+
+static int modparam_workaround_interval;
+module_param_named(workaround_interval, modparam_workaround_interval,
+							int, 0444);
+MODULE_PARM_DESC(workaround_interval,
+	"set stall workaround interval in msecs (0=disabled) (default: 0)");
+
+/* Typical noise/maximum signal level values taken from ndiswrapper iw_ndis.h */
+#define	WL_NOISE	-96	/* typical noise level in dBm */
+#define	WL_SIGMAX	-32	/* typical maximum signal level in dBm */
+
+
+/* Assume that Broadcom 4320 (only chipset at time of writing known to be
+ * based on wireless rndis) has default txpower of 13dBm.
+ * This value is from Linksys WUSB54GSC User Guide, Appendix F: Specifications.
+ *  100% : 20 mW ~ 13dBm
+ *   75% : 15 mW ~ 12dBm
+ *   50% : 10 mW ~ 10dBm
+ *   25% :  5 mW ~  7dBm
+ */
+#define BCM4320_DEFAULT_TXPOWER_DBM_100 13
+#define BCM4320_DEFAULT_TXPOWER_DBM_75  12
+#define BCM4320_DEFAULT_TXPOWER_DBM_50  10
+#define BCM4320_DEFAULT_TXPOWER_DBM_25  7
+
+/* Known device types */
+#define RNDIS_UNKNOWN	0
+#define RNDIS_BCM4320A	1
+#define RNDIS_BCM4320B	2
+
+
+/* NDIS data structures. Taken from wpa_supplicant driver_ndis.c
+ * slightly modified for datatype endianess, etc
+ */
+#define NDIS_802_11_LENGTH_SSID 32
+#define NDIS_802_11_LENGTH_RATES 8
+#define NDIS_802_11_LENGTH_RATES_EX 16
+
+enum ndis_80211_net_type {
+	NDIS_80211_TYPE_FREQ_HOP,
+	NDIS_80211_TYPE_DIRECT_SEQ,
+	NDIS_80211_TYPE_OFDM_A,
+	NDIS_80211_TYPE_OFDM_G
+};
+
+enum ndis_80211_net_infra {
+	NDIS_80211_INFRA_ADHOC,
+	NDIS_80211_INFRA_INFRA,
+	NDIS_80211_INFRA_AUTO_UNKNOWN
+};
+
+enum ndis_80211_auth_mode {
+	NDIS_80211_AUTH_OPEN,
+	NDIS_80211_AUTH_SHARED,
+	NDIS_80211_AUTH_AUTO_SWITCH,
+	NDIS_80211_AUTH_WPA,
+	NDIS_80211_AUTH_WPA_PSK,
+	NDIS_80211_AUTH_WPA_NONE,
+	NDIS_80211_AUTH_WPA2,
+	NDIS_80211_AUTH_WPA2_PSK
+};
+
+enum ndis_80211_encr_status {
+	NDIS_80211_ENCR_WEP_ENABLED,
+	NDIS_80211_ENCR_DISABLED,
+	NDIS_80211_ENCR_WEP_KEY_ABSENT,
+	NDIS_80211_ENCR_NOT_SUPPORTED,
+	NDIS_80211_ENCR_TKIP_ENABLED,
+	NDIS_80211_ENCR_TKIP_KEY_ABSENT,
+	NDIS_80211_ENCR_CCMP_ENABLED,
+	NDIS_80211_ENCR_CCMP_KEY_ABSENT
+};
+
+enum ndis_80211_priv_filter {
+	NDIS_80211_PRIV_ACCEPT_ALL,
+	NDIS_80211_PRIV_8021X_WEP
+};
+
+enum ndis_80211_status_type {
+	NDIS_80211_STATUSTYPE_AUTHENTICATION,
+	NDIS_80211_STATUSTYPE_MEDIASTREAMMODE,
+	NDIS_80211_STATUSTYPE_PMKID_CANDIDATELIST,
+	NDIS_80211_STATUSTYPE_RADIOSTATE,
+};
+
+enum ndis_80211_media_stream_mode {
+	NDIS_80211_MEDIA_STREAM_OFF,
+	NDIS_80211_MEDIA_STREAM_ON
+};
+
+enum ndis_80211_radio_status {
+	NDIS_80211_RADIO_STATUS_ON,
+	NDIS_80211_RADIO_STATUS_HARDWARE_OFF,
+	NDIS_80211_RADIO_STATUS_SOFTWARE_OFF,
+};
+
+enum ndis_80211_addkey_bits {
+	NDIS_80211_ADDKEY_8021X_AUTH = cpu_to_le32(1 << 28),
+	NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ = cpu_to_le32(1 << 29),
+	NDIS_80211_ADDKEY_PAIRWISE_KEY = cpu_to_le32(1 << 30),
+	NDIS_80211_ADDKEY_TRANSMIT_KEY = cpu_to_le32(1 << 31)
+};
+
+enum ndis_80211_addwep_bits {
+	NDIS_80211_ADDWEP_PERCLIENT_KEY = cpu_to_le32(1 << 30),
+	NDIS_80211_ADDWEP_TRANSMIT_KEY = cpu_to_le32(1 << 31)
+};
+
+enum ndis_80211_power_mode {
+	NDIS_80211_POWER_MODE_CAM,
+	NDIS_80211_POWER_MODE_MAX_PSP,
+	NDIS_80211_POWER_MODE_FAST_PSP,
+};
+
+enum ndis_80211_pmkid_cand_list_flag_bits {
+	NDIS_80211_PMKID_CAND_PREAUTH = cpu_to_le32(1 << 0)
+};
+
+struct ndis_80211_auth_request {
+	__le32 length;
+	u8 bssid[ETH_ALEN];
+	u8 padding[2];
+	__le32 flags;
+} __packed;
+
+struct ndis_80211_pmkid_candidate {
+	u8 bssid[ETH_ALEN];
+	u8 padding[2];
+	__le32 flags;
+} __packed;
+
+struct ndis_80211_pmkid_cand_list {
+	__le32 version;
+	__le32 num_candidates;
+	struct ndis_80211_pmkid_candidate candidate_list[];
+} __packed;
+
+struct ndis_80211_status_indication {
+	__le32 status_type;
+	union {
+		__le32					media_stream_mode;
+		__le32					radio_status;
+		struct ndis_80211_auth_request		auth_request[0];
+		struct ndis_80211_pmkid_cand_list	cand_list;
+	} u;
+} __packed;
+
+struct ndis_80211_ssid {
+	__le32 length;
+	u8 essid[NDIS_802_11_LENGTH_SSID];
+} __packed;
+
+struct ndis_80211_conf_freq_hop {
+	__le32 length;
+	__le32 hop_pattern;
+	__le32 hop_set;
+	__le32 dwell_time;
+} __packed;
+
+struct ndis_80211_conf {
+	__le32 length;
+	__le32 beacon_period;
+	__le32 atim_window;
+	__le32 ds_config;
+	struct ndis_80211_conf_freq_hop fh_config;
+} __packed;
+
+struct ndis_80211_bssid_ex {
+	__le32 length;
+	u8 mac[ETH_ALEN];
+	u8 padding[2];
+	struct ndis_80211_ssid ssid;
+	__le32 privacy;
+	__le32 rssi;
+	__le32 net_type;
+	struct ndis_80211_conf config;
+	__le32 net_infra;
+	u8 rates[NDIS_802_11_LENGTH_RATES_EX];
+	__le32 ie_length;
+	u8 ies[];
+} __packed;
+
+struct ndis_80211_bssid_list_ex {
+	__le32 num_items;
+	u8 bssid_data[];
+} __packed;
+
+struct ndis_80211_fixed_ies {
+	u8 timestamp[8];
+	__le16 beacon_interval;
+	__le16 capabilities;
+} __packed;
+
+struct ndis_80211_wep_key {
+	__le32 size;
+	__le32 index;
+	__le32 length;
+	u8 material[32];
+} __packed;
+
+struct ndis_80211_key {
+	__le32 size;
+	__le32 index;
+	__le32 length;
+	u8 bssid[ETH_ALEN];
+	u8 padding[6];
+	u8 rsc[8];
+	u8 material[32];
+} __packed;
+
+struct ndis_80211_remove_key {
+	__le32 size;
+	__le32 index;
+	u8 bssid[ETH_ALEN];
+	u8 padding[2];
+} __packed;
+
+struct ndis_config_param {
+	__le32 name_offs;
+	__le32 name_length;
+	__le32 type;
+	__le32 value_offs;
+	__le32 value_length;
+} __packed;
+
+struct ndis_80211_assoc_info {
+	__le32 length;
+	__le16 req_ies;
+	struct req_ie {
+		__le16 capa;
+		__le16 listen_interval;
+		u8 cur_ap_address[ETH_ALEN];
+	} req_ie;
+	__le32 req_ie_length;
+	__le32 offset_req_ies;
+	__le16 resp_ies;
+	struct resp_ie {
+		__le16 capa;
+		__le16 status_code;
+		__le16 assoc_id;
+	} resp_ie;
+	__le32 resp_ie_length;
+	__le32 offset_resp_ies;
+} __packed;
+
+struct ndis_80211_capability {
+	__le32 length;
+	__le32 version;
+	__le32 num_pmkids;
+	__le32 num_auth_encr_pair;
+} __packed;
+
+struct ndis_80211_bssid_info {
+	u8 bssid[ETH_ALEN];
+	u8 pmkid[16];
+} __packed;
+
+struct ndis_80211_pmkid {
+	__le32 length;
+	__le32 bssid_info_count;
+	struct ndis_80211_bssid_info bssid_info[];
+} __packed;
+
+/*
+ *  private data
+ */
+#define CAP_MODE_80211A		1
+#define CAP_MODE_80211B		2
+#define CAP_MODE_80211G		4
+#define CAP_MODE_MASK		7
+
+#define WORK_LINK_UP		0
+#define WORK_LINK_DOWN		1
+#define WORK_SET_MULTICAST_LIST	2
+
+#define RNDIS_WLAN_ALG_NONE	0
+#define RNDIS_WLAN_ALG_WEP	(1<<0)
+#define RNDIS_WLAN_ALG_TKIP	(1<<1)
+#define RNDIS_WLAN_ALG_CCMP	(1<<2)
+
+#define RNDIS_WLAN_NUM_KEYS		4
+#define RNDIS_WLAN_KEY_MGMT_NONE	0
+#define RNDIS_WLAN_KEY_MGMT_802_1X	(1<<0)
+#define RNDIS_WLAN_KEY_MGMT_PSK		(1<<1)
+
+#define COMMAND_BUFFER_SIZE	(CONTROL_BUFFER_SIZE + sizeof(struct rndis_set))
+
+static const struct ieee80211_channel rndis_channels[] = {
+	{ .center_freq = 2412 },
+	{ .center_freq = 2417 },
+	{ .center_freq = 2422 },
+	{ .center_freq = 2427 },
+	{ .center_freq = 2432 },
+	{ .center_freq = 2437 },
+	{ .center_freq = 2442 },
+	{ .center_freq = 2447 },
+	{ .center_freq = 2452 },
+	{ .center_freq = 2457 },
+	{ .center_freq = 2462 },
+	{ .center_freq = 2467 },
+	{ .center_freq = 2472 },
+	{ .center_freq = 2484 },
+};
+
+static const struct ieee80211_rate rndis_rates[] = {
+	{ .bitrate = 10 },
+	{ .bitrate = 20, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 55, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 110, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 60 },
+	{ .bitrate = 90 },
+	{ .bitrate = 120 },
+	{ .bitrate = 180 },
+	{ .bitrate = 240 },
+	{ .bitrate = 360 },
+	{ .bitrate = 480 },
+	{ .bitrate = 540 }
+};
+
+static const u32 rndis_cipher_suites[] = {
+	WLAN_CIPHER_SUITE_WEP40,
+	WLAN_CIPHER_SUITE_WEP104,
+	WLAN_CIPHER_SUITE_TKIP,
+	WLAN_CIPHER_SUITE_CCMP,
+};
+
+struct rndis_wlan_encr_key {
+	int len;
+	u32 cipher;
+	u8 material[32];
+	u8 bssid[ETH_ALEN];
+	bool pairwise;
+	bool tx_key;
+};
+
+/* RNDIS device private data */
+struct rndis_wlan_private {
+	struct usbnet *usbdev;
+
+	struct wireless_dev wdev;
+
+	struct cfg80211_scan_request *scan_request;
+
+	struct workqueue_struct *workqueue;
+	struct delayed_work dev_poller_work;
+	struct delayed_work scan_work;
+	struct work_struct work;
+	struct mutex command_lock;
+	unsigned long work_pending;
+	int last_qual;
+	s32 cqm_rssi_thold;
+	u32 cqm_rssi_hyst;
+	int last_cqm_event_rssi;
+
+	struct ieee80211_supported_band band;
+	struct ieee80211_channel channels[ARRAY_SIZE(rndis_channels)];
+	struct ieee80211_rate rates[ARRAY_SIZE(rndis_rates)];
+	u32 cipher_suites[ARRAY_SIZE(rndis_cipher_suites)];
+
+	int device_type;
+	int caps;
+	int multicast_size;
+
+	/* module parameters */
+	char param_country[4];
+	int  param_frameburst;
+	int  param_afterburner;
+	int  param_power_save;
+	int  param_power_output;
+	int  param_roamtrigger;
+	int  param_roamdelta;
+	u32  param_workaround_interval;
+
+	/* hardware state */
+	bool radio_on;
+	int power_mode;
+	int infra_mode;
+	bool connected;
+	u8 bssid[ETH_ALEN];
+	u32 current_command_oid;
+
+	/* encryption stuff */
+	u8 encr_tx_key_index;
+	struct rndis_wlan_encr_key encr_keys[RNDIS_WLAN_NUM_KEYS];
+	int  wpa_version;
+
+	u8 command_buffer[COMMAND_BUFFER_SIZE];
+};
+
+/*
+ * cfg80211 ops
+ */
+static int rndis_change_virtual_intf(struct wiphy *wiphy,
+					struct net_device *dev,
+					enum nl80211_iftype type,
+					struct vif_params *params);
+
+static int rndis_scan(struct wiphy *wiphy,
+			struct cfg80211_scan_request *request);
+
+static int rndis_set_wiphy_params(struct wiphy *wiphy, u32 changed);
+
+static int rndis_set_tx_power(struct wiphy *wiphy,
+			      struct wireless_dev *wdev,
+			      enum nl80211_tx_power_setting type,
+			      int mbm);
+static int rndis_get_tx_power(struct wiphy *wiphy,
+			      struct wireless_dev *wdev,
+			      int *dbm);
+
+static int rndis_connect(struct wiphy *wiphy, struct net_device *dev,
+				struct cfg80211_connect_params *sme);
+
+static int rndis_disconnect(struct wiphy *wiphy, struct net_device *dev,
+				u16 reason_code);
+
+static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev,
+					struct cfg80211_ibss_params *params);
+
+static int rndis_leave_ibss(struct wiphy *wiphy, struct net_device *dev);
+
+static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
+			 int link_id,  u8 key_index, bool pairwise,
+			 const u8 *mac_addr, struct key_params *params);
+
+static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev,
+			 int link_id, u8 key_index, bool pairwise,
+			 const u8 *mac_addr);
+
+static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev,
+				 int link_id, u8 key_index, bool unicast,
+				 bool multicast);
+
+static int rndis_get_station(struct wiphy *wiphy, struct net_device *dev,
+			     const u8 *mac, struct station_info *sinfo);
+
+static int rndis_dump_station(struct wiphy *wiphy, struct net_device *dev,
+			       int idx, u8 *mac, struct station_info *sinfo);
+
+static int rndis_set_pmksa(struct wiphy *wiphy, struct net_device *netdev,
+				struct cfg80211_pmksa *pmksa);
+
+static int rndis_del_pmksa(struct wiphy *wiphy, struct net_device *netdev,
+				struct cfg80211_pmksa *pmksa);
+
+static int rndis_flush_pmksa(struct wiphy *wiphy, struct net_device *netdev);
+
+static int rndis_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev,
+				bool enabled, int timeout);
+
+static int rndis_set_cqm_rssi_config(struct wiphy *wiphy,
+					struct net_device *dev,
+					s32 rssi_thold, u32 rssi_hyst);
+
+static const struct cfg80211_ops rndis_config_ops = {
+	.change_virtual_intf = rndis_change_virtual_intf,
+	.scan = rndis_scan,
+	.set_wiphy_params = rndis_set_wiphy_params,
+	.set_tx_power = rndis_set_tx_power,
+	.get_tx_power = rndis_get_tx_power,
+	.connect = rndis_connect,
+	.disconnect = rndis_disconnect,
+	.join_ibss = rndis_join_ibss,
+	.leave_ibss = rndis_leave_ibss,
+	.add_key = rndis_add_key,
+	.del_key = rndis_del_key,
+	.set_default_key = rndis_set_default_key,
+	.get_station = rndis_get_station,
+	.dump_station = rndis_dump_station,
+	.set_pmksa = rndis_set_pmksa,
+	.del_pmksa = rndis_del_pmksa,
+	.flush_pmksa = rndis_flush_pmksa,
+	.set_power_mgmt = rndis_set_power_mgmt,
+	.set_cqm_rssi_config = rndis_set_cqm_rssi_config,
+};
+
+static void *rndis_wiphy_privid = &rndis_wiphy_privid;
+
+
+static struct rndis_wlan_private *get_rndis_wlan_priv(struct usbnet *dev)
+{
+	return (struct rndis_wlan_private *)dev->driver_priv;
+}
+
+static u32 get_bcm4320_power_dbm(struct rndis_wlan_private *priv)
+{
+	switch (priv->param_power_output) {
+	default:
+	case 3:
+		return BCM4320_DEFAULT_TXPOWER_DBM_100;
+	case 2:
+		return BCM4320_DEFAULT_TXPOWER_DBM_75;
+	case 1:
+		return BCM4320_DEFAULT_TXPOWER_DBM_50;
+	case 0:
+		return BCM4320_DEFAULT_TXPOWER_DBM_25;
+	}
+}
+
+static bool is_wpa_key(struct rndis_wlan_private *priv, u8 idx)
+{
+	int cipher = priv->encr_keys[idx].cipher;
+
+	return (cipher == WLAN_CIPHER_SUITE_CCMP ||
+		cipher == WLAN_CIPHER_SUITE_TKIP);
+}
+
+static int rndis_cipher_to_alg(u32 cipher)
+{
+	switch (cipher) {
+	default:
+		return RNDIS_WLAN_ALG_NONE;
+	case WLAN_CIPHER_SUITE_WEP40:
+	case WLAN_CIPHER_SUITE_WEP104:
+		return RNDIS_WLAN_ALG_WEP;
+	case WLAN_CIPHER_SUITE_TKIP:
+		return RNDIS_WLAN_ALG_TKIP;
+	case WLAN_CIPHER_SUITE_CCMP:
+		return RNDIS_WLAN_ALG_CCMP;
+	}
+}
+
+static int rndis_akm_suite_to_key_mgmt(u32 akm_suite)
+{
+	switch (akm_suite) {
+	default:
+		return RNDIS_WLAN_KEY_MGMT_NONE;
+	case WLAN_AKM_SUITE_8021X:
+		return RNDIS_WLAN_KEY_MGMT_802_1X;
+	case WLAN_AKM_SUITE_PSK:
+		return RNDIS_WLAN_KEY_MGMT_PSK;
+	}
+}
+
+#ifdef DEBUG
+static const char *oid_to_string(u32 oid)
+{
+	switch (oid) {
+#define OID_STR(oid) case oid: return(#oid)
+		/* from rndis_host.h */
+		OID_STR(RNDIS_OID_802_3_PERMANENT_ADDRESS);
+		OID_STR(RNDIS_OID_GEN_MAXIMUM_FRAME_SIZE);
+		OID_STR(RNDIS_OID_GEN_CURRENT_PACKET_FILTER);
+		OID_STR(RNDIS_OID_GEN_PHYSICAL_MEDIUM);
+
+		/* from rndis_wlan.c */
+		OID_STR(RNDIS_OID_GEN_LINK_SPEED);
+		OID_STR(RNDIS_OID_GEN_RNDIS_CONFIG_PARAMETER);
+
+		OID_STR(RNDIS_OID_GEN_XMIT_OK);
+		OID_STR(RNDIS_OID_GEN_RCV_OK);
+		OID_STR(RNDIS_OID_GEN_XMIT_ERROR);
+		OID_STR(RNDIS_OID_GEN_RCV_ERROR);
+		OID_STR(RNDIS_OID_GEN_RCV_NO_BUFFER);
+
+		OID_STR(RNDIS_OID_802_3_CURRENT_ADDRESS);
+		OID_STR(RNDIS_OID_802_3_MULTICAST_LIST);
+		OID_STR(RNDIS_OID_802_3_MAXIMUM_LIST_SIZE);
+
+		OID_STR(RNDIS_OID_802_11_BSSID);
+		OID_STR(RNDIS_OID_802_11_SSID);
+		OID_STR(RNDIS_OID_802_11_INFRASTRUCTURE_MODE);
+		OID_STR(RNDIS_OID_802_11_ADD_WEP);
+		OID_STR(RNDIS_OID_802_11_REMOVE_WEP);
+		OID_STR(RNDIS_OID_802_11_DISASSOCIATE);
+		OID_STR(RNDIS_OID_802_11_AUTHENTICATION_MODE);
+		OID_STR(RNDIS_OID_802_11_PRIVACY_FILTER);
+		OID_STR(RNDIS_OID_802_11_BSSID_LIST_SCAN);
+		OID_STR(RNDIS_OID_802_11_ENCRYPTION_STATUS);
+		OID_STR(RNDIS_OID_802_11_ADD_KEY);
+		OID_STR(RNDIS_OID_802_11_REMOVE_KEY);
+		OID_STR(RNDIS_OID_802_11_ASSOCIATION_INFORMATION);
+		OID_STR(RNDIS_OID_802_11_CAPABILITY);
+		OID_STR(RNDIS_OID_802_11_PMKID);
+		OID_STR(RNDIS_OID_802_11_NETWORK_TYPES_SUPPORTED);
+		OID_STR(RNDIS_OID_802_11_NETWORK_TYPE_IN_USE);
+		OID_STR(RNDIS_OID_802_11_TX_POWER_LEVEL);
+		OID_STR(RNDIS_OID_802_11_RSSI);
+		OID_STR(RNDIS_OID_802_11_RSSI_TRIGGER);
+		OID_STR(RNDIS_OID_802_11_FRAGMENTATION_THRESHOLD);
+		OID_STR(RNDIS_OID_802_11_RTS_THRESHOLD);
+		OID_STR(RNDIS_OID_802_11_SUPPORTED_RATES);
+		OID_STR(RNDIS_OID_802_11_CONFIGURATION);
+		OID_STR(RNDIS_OID_802_11_POWER_MODE);
+		OID_STR(RNDIS_OID_802_11_BSSID_LIST);
+#undef OID_STR
+	}
+
+	return "?";
+}
+#else
+static const char *oid_to_string(u32 oid)
+{
+	return "?";
+}
+#endif
+
+/* translate error code */
+static int rndis_error_status(__le32 rndis_status)
+{
+	int ret = -EINVAL;
+	switch (le32_to_cpu(rndis_status)) {
+	case RNDIS_STATUS_SUCCESS:
+		ret = 0;
+		break;
+	case RNDIS_STATUS_FAILURE:
+	case RNDIS_STATUS_INVALID_DATA:
+		ret = -EINVAL;
+		break;
+	case RNDIS_STATUS_NOT_SUPPORTED:
+		ret = -EOPNOTSUPP;
+		break;
+	case RNDIS_STATUS_ADAPTER_NOT_READY:
+	case RNDIS_STATUS_ADAPTER_NOT_OPEN:
+		ret = -EBUSY;
+		break;
+	}
+	return ret;
+}
+
+static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(dev);
+	union {
+		void			*buf;
+		struct rndis_msg_hdr	*header;
+		struct rndis_query	*get;
+		struct rndis_query_c	*get_c;
+	} u;
+	int ret;
+	size_t buflen, resplen, respoffs, copylen;
+
+	buflen = *len + sizeof(*u.get);
+	if (buflen < CONTROL_BUFFER_SIZE)
+		buflen = CONTROL_BUFFER_SIZE;
+
+	if (buflen > COMMAND_BUFFER_SIZE) {
+		u.buf = kmalloc(buflen, GFP_KERNEL);
+		if (!u.buf)
+			return -ENOMEM;
+	} else {
+		u.buf = priv->command_buffer;
+	}
+
+	mutex_lock(&priv->command_lock);
+
+	memset(u.get, 0, sizeof *u.get);
+	u.get->msg_type = cpu_to_le32(RNDIS_MSG_QUERY);
+	u.get->msg_len = cpu_to_le32(sizeof *u.get);
+	u.get->oid = cpu_to_le32(oid);
+
+	priv->current_command_oid = oid;
+	ret = rndis_command(dev, u.header, buflen);
+	priv->current_command_oid = 0;
+	if (ret < 0)
+		netdev_dbg(dev->net, "%s(%s): rndis_command() failed, %d (%08x)\n",
+			   __func__, oid_to_string(oid), ret,
+			   le32_to_cpu(u.get_c->status));
+
+	if (ret == 0) {
+		resplen = le32_to_cpu(u.get_c->len);
+		respoffs = le32_to_cpu(u.get_c->offset) + 8;
+
+		if (respoffs > buflen) {
+			/* Device returned data offset outside buffer, error. */
+			netdev_dbg(dev->net,
+				   "%s(%s): received invalid data offset: %zu > %zu\n",
+				   __func__, oid_to_string(oid), respoffs, buflen);
+
+			ret = -EINVAL;
+			goto exit_unlock;
+		}
+
+		copylen = min(resplen, buflen - respoffs);
+
+		if (copylen > *len)
+			copylen = *len;
+
+		memcpy(data, u.buf + respoffs, copylen);
+
+		*len = resplen;
+
+		ret = rndis_error_status(u.get_c->status);
+		if (ret < 0)
+			netdev_dbg(dev->net, "%s(%s): device returned error,  0x%08x (%d)\n",
+				   __func__, oid_to_string(oid),
+				   le32_to_cpu(u.get_c->status), ret);
+	}
+
+exit_unlock:
+	mutex_unlock(&priv->command_lock);
+
+	if (u.buf != priv->command_buffer)
+		kfree(u.buf);
+	return ret;
+}
+
+static int rndis_set_oid(struct usbnet *dev, u32 oid, const void *data,
+			 int len)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(dev);
+	union {
+		void			*buf;
+		struct rndis_msg_hdr	*header;
+		struct rndis_set	*set;
+		struct rndis_set_c	*set_c;
+	} u;
+	int ret, buflen;
+
+	buflen = len + sizeof(*u.set);
+	if (buflen < CONTROL_BUFFER_SIZE)
+		buflen = CONTROL_BUFFER_SIZE;
+
+	if (buflen > COMMAND_BUFFER_SIZE) {
+		u.buf = kmalloc(buflen, GFP_KERNEL);
+		if (!u.buf)
+			return -ENOMEM;
+	} else {
+		u.buf = priv->command_buffer;
+	}
+
+	mutex_lock(&priv->command_lock);
+
+	memset(u.set, 0, sizeof *u.set);
+	u.set->msg_type = cpu_to_le32(RNDIS_MSG_SET);
+	u.set->msg_len = cpu_to_le32(sizeof(*u.set) + len);
+	u.set->oid = cpu_to_le32(oid);
+	u.set->len = cpu_to_le32(len);
+	u.set->offset = cpu_to_le32(sizeof(*u.set) - 8);
+	u.set->handle = cpu_to_le32(0);
+	memcpy(u.buf + sizeof(*u.set), data, len);
+
+	priv->current_command_oid = oid;
+	ret = rndis_command(dev, u.header, buflen);
+	priv->current_command_oid = 0;
+	if (ret < 0)
+		netdev_dbg(dev->net, "%s(%s): rndis_command() failed, %d (%08x)\n",
+			   __func__, oid_to_string(oid), ret,
+			   le32_to_cpu(u.set_c->status));
+
+	if (ret == 0) {
+		ret = rndis_error_status(u.set_c->status);
+
+		if (ret < 0)
+			netdev_dbg(dev->net, "%s(%s): device returned error, 0x%08x (%d)\n",
+				   __func__, oid_to_string(oid),
+				   le32_to_cpu(u.set_c->status), ret);
+	}
+
+	mutex_unlock(&priv->command_lock);
+
+	if (u.buf != priv->command_buffer)
+		kfree(u.buf);
+	return ret;
+}
+
+static int rndis_reset(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct rndis_reset *reset;
+	int ret;
+
+	mutex_lock(&priv->command_lock);
+
+	reset = (void *)priv->command_buffer;
+	memset(reset, 0, sizeof(*reset));
+	reset->msg_type = cpu_to_le32(RNDIS_MSG_RESET);
+	reset->msg_len = cpu_to_le32(sizeof(*reset));
+	priv->current_command_oid = 0;
+	ret = rndis_command(usbdev, (void *)reset, CONTROL_BUFFER_SIZE);
+
+	mutex_unlock(&priv->command_lock);
+
+	if (ret < 0)
+		return ret;
+	return 0;
+}
+
+/*
+ * Specs say that we can only set config parameters only soon after device
+ * initialization.
+ *   value_type: 0 = u32, 2 = unicode string
+ */
+static int rndis_set_config_parameter(struct usbnet *dev, char *param,
+						int value_type, void *value)
+{
+	struct ndis_config_param *infobuf;
+	int value_len, info_len, param_len, ret, i;
+	__le16 *unibuf;
+	__le32 *dst_value;
+
+	if (value_type == 0)
+		value_len = sizeof(__le32);
+	else if (value_type == 2)
+		value_len = strlen(value) * sizeof(__le16);
+	else
+		return -EINVAL;
+
+	param_len = strlen(param) * sizeof(__le16);
+	info_len = sizeof(*infobuf) + param_len + value_len;
+
+#ifdef DEBUG
+	info_len += 12;
+#endif
+	infobuf = kmalloc(info_len, GFP_KERNEL);
+	if (!infobuf)
+		return -ENOMEM;
+
+#ifdef DEBUG
+	info_len -= 12;
+	/* extra 12 bytes are for padding (debug output) */
+	memset(infobuf, 0xCC, info_len + 12);
+#endif
+
+	if (value_type == 2)
+		netdev_dbg(dev->net, "setting config parameter: %s, value: %s\n",
+			   param, (u8 *)value);
+	else
+		netdev_dbg(dev->net, "setting config parameter: %s, value: %d\n",
+			   param, *(u32 *)value);
+
+	infobuf->name_offs = cpu_to_le32(sizeof(*infobuf));
+	infobuf->name_length = cpu_to_le32(param_len);
+	infobuf->type = cpu_to_le32(value_type);
+	infobuf->value_offs = cpu_to_le32(sizeof(*infobuf) + param_len);
+	infobuf->value_length = cpu_to_le32(value_len);
+
+	/* simple string to unicode string conversion */
+	unibuf = (void *)infobuf + sizeof(*infobuf);
+	for (i = 0; i < param_len / sizeof(__le16); i++)
+		unibuf[i] = cpu_to_le16(param[i]);
+
+	if (value_type == 2) {
+		unibuf = (void *)infobuf + sizeof(*infobuf) + param_len;
+		for (i = 0; i < value_len / sizeof(__le16); i++)
+			unibuf[i] = cpu_to_le16(((u8 *)value)[i]);
+	} else {
+		dst_value = (void *)infobuf + sizeof(*infobuf) + param_len;
+		*dst_value = cpu_to_le32(*(u32 *)value);
+	}
+
+#ifdef DEBUG
+	netdev_dbg(dev->net, "info buffer (len: %d)\n", info_len);
+	for (i = 0; i < info_len; i += 12) {
+		u32 *tmp = (u32 *)((u8 *)infobuf + i);
+		netdev_dbg(dev->net, "%08X:%08X:%08X\n",
+			   cpu_to_be32(tmp[0]),
+			   cpu_to_be32(tmp[1]),
+			   cpu_to_be32(tmp[2]));
+	}
+#endif
+
+	ret = rndis_set_oid(dev, RNDIS_OID_GEN_RNDIS_CONFIG_PARAMETER,
+							infobuf, info_len);
+	if (ret != 0)
+		netdev_dbg(dev->net, "setting rndis config parameter failed, %d\n",
+			   ret);
+
+	kfree(infobuf);
+	return ret;
+}
+
+static int rndis_set_config_parameter_str(struct usbnet *dev,
+						char *param, char *value)
+{
+	return rndis_set_config_parameter(dev, param, 2, value);
+}
+
+/*
+ * data conversion functions
+ */
+static int level_to_qual(int level)
+{
+	int qual = 100 * (level - WL_NOISE) / (WL_SIGMAX - WL_NOISE);
+	return qual >= 0 ? (qual <= 100 ? qual : 100) : 0;
+}
+
+/*
+ * common functions
+ */
+static int set_infra_mode(struct usbnet *usbdev, int mode);
+static void restore_keys(struct usbnet *usbdev);
+static int rndis_check_bssid_list(struct usbnet *usbdev, u8 *match_bssid,
+					bool *matched);
+
+static int rndis_start_bssid_list_scan(struct usbnet *usbdev)
+{
+	__le32 tmp;
+
+	/* Note: RNDIS_OID_802_11_BSSID_LIST_SCAN clears internal BSS list. */
+	tmp = cpu_to_le32(1);
+	return rndis_set_oid(usbdev, RNDIS_OID_802_11_BSSID_LIST_SCAN, &tmp,
+							sizeof(tmp));
+}
+
+static int set_essid(struct usbnet *usbdev, struct ndis_80211_ssid *ssid)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	int ret;
+
+	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_SSID,
+			    ssid, sizeof(*ssid));
+	if (ret < 0) {
+		netdev_warn(usbdev->net, "setting SSID failed (%08X)\n", ret);
+		return ret;
+	}
+	if (ret == 0) {
+		priv->radio_on = true;
+		netdev_dbg(usbdev->net, "%s(): radio_on = true\n", __func__);
+	}
+
+	return ret;
+}
+
+static int set_bssid(struct usbnet *usbdev, const u8 *bssid)
+{
+	int ret;
+
+	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_BSSID,
+			    bssid, ETH_ALEN);
+	if (ret < 0) {
+		netdev_warn(usbdev->net, "setting BSSID[%pM] failed (%08X)\n",
+			    bssid, ret);
+		return ret;
+	}
+
+	return ret;
+}
+
+static int clear_bssid(struct usbnet *usbdev)
+{
+	static const u8 broadcast_mac[ETH_ALEN] = {
+		0xff, 0xff, 0xff, 0xff, 0xff, 0xff
+	};
+
+	return set_bssid(usbdev, broadcast_mac);
+}
+
+static int get_bssid(struct usbnet *usbdev, u8 bssid[ETH_ALEN])
+{
+	int ret, len;
+
+	len = ETH_ALEN;
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_BSSID,
+			      bssid, &len);
+
+	if (ret != 0)
+		eth_zero_addr(bssid);
+
+	return ret;
+}
+
+static int get_association_info(struct usbnet *usbdev,
+			struct ndis_80211_assoc_info *info, int len)
+{
+	return rndis_query_oid(usbdev,
+			RNDIS_OID_802_11_ASSOCIATION_INFORMATION,
+			info, &len);
+}
+
+static bool is_associated(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	u8 bssid[ETH_ALEN];
+
+	if (!priv->radio_on)
+		return false;
+
+	return (get_bssid(usbdev, bssid) == 0 && !is_zero_ether_addr(bssid));
+}
+
+static int disassociate(struct usbnet *usbdev, bool reset_ssid)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ndis_80211_ssid ssid;
+	int i, ret = 0;
+
+	if (priv->radio_on) {
+		ret = rndis_set_oid(usbdev,
+				RNDIS_OID_802_11_DISASSOCIATE,
+				NULL, 0);
+		if (ret == 0) {
+			priv->radio_on = false;
+			netdev_dbg(usbdev->net, "%s(): radio_on = false\n",
+				   __func__);
+
+			if (reset_ssid)
+				msleep(100);
+		}
+	}
+
+	/* disassociate causes radio to be turned off; if reset_ssid
+	 * is given, set random ssid to enable radio */
+	if (reset_ssid) {
+		/* Set device to infrastructure mode so we don't get ad-hoc
+		 * 'media connect' indications with the random ssid.
+		 */
+		set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA);
+
+		ssid.length = cpu_to_le32(sizeof(ssid.essid));
+		get_random_bytes(&ssid.essid[2], sizeof(ssid.essid)-2);
+		ssid.essid[0] = 0x1;
+		ssid.essid[1] = 0xff;
+		for (i = 2; i < sizeof(ssid.essid); i++)
+			ssid.essid[i] = 0x1 + (ssid.essid[i] * 0xfe / 0xff);
+		ret = set_essid(usbdev, &ssid);
+	}
+	return ret;
+}
+
+static int set_auth_mode(struct usbnet *usbdev, u32 wpa_version,
+				enum nl80211_auth_type auth_type, int keymgmt)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	__le32 tmp;
+	int auth_mode, ret;
+
+	netdev_dbg(usbdev->net, "%s(): wpa_version=0x%x authalg=0x%x keymgmt=0x%x\n",
+		   __func__, wpa_version, auth_type, keymgmt);
+
+	if (wpa_version & NL80211_WPA_VERSION_2) {
+		if (keymgmt & RNDIS_WLAN_KEY_MGMT_802_1X)
+			auth_mode = NDIS_80211_AUTH_WPA2;
+		else
+			auth_mode = NDIS_80211_AUTH_WPA2_PSK;
+	} else if (wpa_version & NL80211_WPA_VERSION_1) {
+		if (keymgmt & RNDIS_WLAN_KEY_MGMT_802_1X)
+			auth_mode = NDIS_80211_AUTH_WPA;
+		else if (keymgmt & RNDIS_WLAN_KEY_MGMT_PSK)
+			auth_mode = NDIS_80211_AUTH_WPA_PSK;
+		else
+			auth_mode = NDIS_80211_AUTH_WPA_NONE;
+	} else if (auth_type == NL80211_AUTHTYPE_SHARED_KEY)
+		auth_mode = NDIS_80211_AUTH_SHARED;
+	else if (auth_type == NL80211_AUTHTYPE_OPEN_SYSTEM)
+		auth_mode = NDIS_80211_AUTH_OPEN;
+	else if (auth_type == NL80211_AUTHTYPE_AUTOMATIC)
+		auth_mode = NDIS_80211_AUTH_AUTO_SWITCH;
+	else
+		return -ENOTSUPP;
+
+	tmp = cpu_to_le32(auth_mode);
+	ret = rndis_set_oid(usbdev,
+			    RNDIS_OID_802_11_AUTHENTICATION_MODE,
+			    &tmp, sizeof(tmp));
+	if (ret != 0) {
+		netdev_warn(usbdev->net, "setting auth mode failed (%08X)\n",
+			    ret);
+		return ret;
+	}
+
+	priv->wpa_version = wpa_version;
+
+	return 0;
+}
+
+static int set_priv_filter(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	__le32 tmp;
+
+	netdev_dbg(usbdev->net, "%s(): wpa_version=0x%x\n",
+		   __func__, priv->wpa_version);
+
+	if (priv->wpa_version & NL80211_WPA_VERSION_2 ||
+	    priv->wpa_version & NL80211_WPA_VERSION_1)
+		tmp = cpu_to_le32(NDIS_80211_PRIV_8021X_WEP);
+	else
+		tmp = cpu_to_le32(NDIS_80211_PRIV_ACCEPT_ALL);
+
+	return rndis_set_oid(usbdev,
+			     RNDIS_OID_802_11_PRIVACY_FILTER, &tmp,
+			     sizeof(tmp));
+}
+
+static int set_encr_mode(struct usbnet *usbdev, int pairwise, int groupwise)
+{
+	__le32 tmp;
+	int encr_mode, ret;
+
+	netdev_dbg(usbdev->net, "%s(): cipher_pair=0x%x cipher_group=0x%x\n",
+		   __func__, pairwise, groupwise);
+
+	if (pairwise & RNDIS_WLAN_ALG_CCMP)
+		encr_mode = NDIS_80211_ENCR_CCMP_ENABLED;
+	else if (pairwise & RNDIS_WLAN_ALG_TKIP)
+		encr_mode = NDIS_80211_ENCR_TKIP_ENABLED;
+	else if (pairwise & RNDIS_WLAN_ALG_WEP)
+		encr_mode = NDIS_80211_ENCR_WEP_ENABLED;
+	else if (groupwise & RNDIS_WLAN_ALG_CCMP)
+		encr_mode = NDIS_80211_ENCR_CCMP_ENABLED;
+	else if (groupwise & RNDIS_WLAN_ALG_TKIP)
+		encr_mode = NDIS_80211_ENCR_TKIP_ENABLED;
+	else
+		encr_mode = NDIS_80211_ENCR_DISABLED;
+
+	tmp = cpu_to_le32(encr_mode);
+	ret = rndis_set_oid(usbdev,
+			RNDIS_OID_802_11_ENCRYPTION_STATUS, &tmp,
+			sizeof(tmp));
+	if (ret != 0) {
+		netdev_warn(usbdev->net, "setting encr mode failed (%08X)\n",
+			    ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int set_infra_mode(struct usbnet *usbdev, int mode)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	__le32 tmp;
+	int ret;
+
+	netdev_dbg(usbdev->net, "%s(): infra_mode=0x%x\n",
+		   __func__, priv->infra_mode);
+
+	tmp = cpu_to_le32(mode);
+	ret = rndis_set_oid(usbdev,
+			    RNDIS_OID_802_11_INFRASTRUCTURE_MODE,
+			    &tmp, sizeof(tmp));
+	if (ret != 0) {
+		netdev_warn(usbdev->net, "setting infra mode failed (%08X)\n",
+			    ret);
+		return ret;
+	}
+
+	/* NDIS drivers clear keys when infrastructure mode is
+	 * changed. But Linux tools assume otherwise. So set the
+	 * keys */
+	restore_keys(usbdev);
+
+	priv->infra_mode = mode;
+	return 0;
+}
+
+static int set_rts_threshold(struct usbnet *usbdev, u32 rts_threshold)
+{
+	__le32 tmp;
+
+	netdev_dbg(usbdev->net, "%s(): %i\n", __func__, rts_threshold);
+
+	if (rts_threshold == -1 || rts_threshold > 2347)
+		rts_threshold = 2347;
+
+	tmp = cpu_to_le32(rts_threshold);
+	return rndis_set_oid(usbdev,
+			     RNDIS_OID_802_11_RTS_THRESHOLD,
+			     &tmp, sizeof(tmp));
+}
+
+static int set_frag_threshold(struct usbnet *usbdev, u32 frag_threshold)
+{
+	__le32 tmp;
+
+	netdev_dbg(usbdev->net, "%s(): %i\n", __func__, frag_threshold);
+
+	if (frag_threshold < 256 || frag_threshold > 2346)
+		frag_threshold = 2346;
+
+	tmp = cpu_to_le32(frag_threshold);
+	return rndis_set_oid(usbdev,
+			RNDIS_OID_802_11_FRAGMENTATION_THRESHOLD,
+			&tmp, sizeof(tmp));
+}
+
+static void set_default_iw_params(struct usbnet *usbdev)
+{
+	set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA);
+	set_auth_mode(usbdev, 0, NL80211_AUTHTYPE_OPEN_SYSTEM,
+						RNDIS_WLAN_KEY_MGMT_NONE);
+	set_priv_filter(usbdev);
+	set_encr_mode(usbdev, RNDIS_WLAN_ALG_NONE, RNDIS_WLAN_ALG_NONE);
+}
+
+static int deauthenticate(struct usbnet *usbdev)
+{
+	int ret;
+
+	ret = disassociate(usbdev, true);
+	set_default_iw_params(usbdev);
+	return ret;
+}
+
+static int set_channel(struct usbnet *usbdev, int channel)
+{
+	struct ndis_80211_conf config;
+	unsigned int dsconfig;
+	int len, ret;
+
+	netdev_dbg(usbdev->net, "%s(%d)\n", __func__, channel);
+
+	/* this OID is valid only when not associated */
+	if (is_associated(usbdev))
+		return 0;
+
+	dsconfig = 1000 *
+		ieee80211_channel_to_frequency(channel, NL80211_BAND_2GHZ);
+
+	len = sizeof(config);
+	ret = rndis_query_oid(usbdev,
+			RNDIS_OID_802_11_CONFIGURATION,
+			&config, &len);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "%s(): querying configuration failed\n",
+			   __func__);
+		return ret;
+	}
+
+	config.ds_config = cpu_to_le32(dsconfig);
+	ret = rndis_set_oid(usbdev,
+			RNDIS_OID_802_11_CONFIGURATION,
+			&config, sizeof(config));
+
+	netdev_dbg(usbdev->net, "%s(): %d -> %d\n", __func__, channel, ret);
+
+	return ret;
+}
+
+static struct ieee80211_channel *get_current_channel(struct usbnet *usbdev,
+						     u32 *beacon_period)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ieee80211_channel *channel;
+	struct ndis_80211_conf config;
+	int len, ret;
+
+	/* Get channel and beacon interval */
+	len = sizeof(config);
+	ret = rndis_query_oid(usbdev,
+			RNDIS_OID_802_11_CONFIGURATION,
+			&config, &len);
+	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_CONFIGURATION -> %d\n",
+				__func__, ret);
+	if (ret < 0)
+		return NULL;
+
+	channel = ieee80211_get_channel(priv->wdev.wiphy,
+				KHZ_TO_MHZ(le32_to_cpu(config.ds_config)));
+	if (!channel)
+		return NULL;
+
+	if (beacon_period)
+		*beacon_period = le32_to_cpu(config.beacon_period);
+	return channel;
+}
+
+/* index must be 0 - N, as per NDIS  */
+static int add_wep_key(struct usbnet *usbdev, const u8 *key, int key_len,
+								u8 index)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ndis_80211_wep_key ndis_key;
+	u32 cipher;
+	int ret;
+
+	netdev_dbg(usbdev->net, "%s(idx: %d, len: %d)\n",
+		   __func__, index, key_len);
+
+	if (index >= RNDIS_WLAN_NUM_KEYS)
+		return -EINVAL;
+
+	if (key_len == 5)
+		cipher = WLAN_CIPHER_SUITE_WEP40;
+	else if (key_len == 13)
+		cipher = WLAN_CIPHER_SUITE_WEP104;
+	else
+		return -EINVAL;
+
+	memset(&ndis_key, 0, sizeof(ndis_key));
+
+	ndis_key.size = cpu_to_le32(sizeof(ndis_key));
+	ndis_key.length = cpu_to_le32(key_len);
+	ndis_key.index = cpu_to_le32(index);
+	memcpy(&ndis_key.material, key, key_len);
+
+	if (index == priv->encr_tx_key_index) {
+		ndis_key.index |= NDIS_80211_ADDWEP_TRANSMIT_KEY;
+		ret = set_encr_mode(usbdev, RNDIS_WLAN_ALG_WEP,
+							RNDIS_WLAN_ALG_NONE);
+		if (ret)
+			netdev_warn(usbdev->net, "encryption couldn't be enabled (%08X)\n",
+				    ret);
+	}
+
+	ret = rndis_set_oid(usbdev,
+			RNDIS_OID_802_11_ADD_WEP, &ndis_key,
+			sizeof(ndis_key));
+	if (ret != 0) {
+		netdev_warn(usbdev->net, "adding encryption key %d failed (%08X)\n",
+			    index + 1, ret);
+		return ret;
+	}
+
+	priv->encr_keys[index].len = key_len;
+	priv->encr_keys[index].cipher = cipher;
+	memcpy(&priv->encr_keys[index].material, key, key_len);
+	eth_broadcast_addr(priv->encr_keys[index].bssid);
+
+	return 0;
+}
+
+static int add_wpa_key(struct usbnet *usbdev, const u8 *key, int key_len,
+			u8 index, const u8 *addr, const u8 *rx_seq,
+			int seq_len, u32 cipher, __le32 flags)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ndis_80211_key ndis_key;
+	bool is_addr_ok;
+	int ret;
+
+	if (index >= RNDIS_WLAN_NUM_KEYS) {
+		netdev_dbg(usbdev->net, "%s(): index out of range (%i)\n",
+			   __func__, index);
+		return -EINVAL;
+	}
+	if (key_len > sizeof(ndis_key.material) || key_len < 0) {
+		netdev_dbg(usbdev->net, "%s(): key length out of range (%i)\n",
+			   __func__, key_len);
+		return -EINVAL;
+	}
+	if (flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ) {
+		if (!rx_seq || seq_len <= 0) {
+			netdev_dbg(usbdev->net, "%s(): recv seq flag without buffer\n",
+				   __func__);
+			return -EINVAL;
+		}
+		if (rx_seq && seq_len > sizeof(ndis_key.rsc)) {
+			netdev_dbg(usbdev->net, "%s(): too big recv seq buffer\n", __func__);
+			return -EINVAL;
+		}
+	}
+
+	is_addr_ok = addr && !is_zero_ether_addr(addr) &&
+					!is_broadcast_ether_addr(addr);
+	if ((flags & NDIS_80211_ADDKEY_PAIRWISE_KEY) && !is_addr_ok) {
+		netdev_dbg(usbdev->net, "%s(): pairwise but bssid invalid (%pM)\n",
+			   __func__, addr);
+		return -EINVAL;
+	}
+
+	netdev_dbg(usbdev->net, "%s(%i): flags:%i%i%i\n",
+		   __func__, index,
+		   !!(flags & NDIS_80211_ADDKEY_TRANSMIT_KEY),
+		   !!(flags & NDIS_80211_ADDKEY_PAIRWISE_KEY),
+		   !!(flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ));
+
+	memset(&ndis_key, 0, sizeof(ndis_key));
+
+	ndis_key.size = cpu_to_le32(sizeof(ndis_key) -
+				sizeof(ndis_key.material) + key_len);
+	ndis_key.length = cpu_to_le32(key_len);
+	ndis_key.index = cpu_to_le32(index) | flags;
+
+	if (cipher == WLAN_CIPHER_SUITE_TKIP && key_len == 32) {
+		/* wpa_supplicant gives us the Michael MIC RX/TX keys in
+		 * different order than NDIS spec, so swap the order here. */
+		memcpy(ndis_key.material, key, 16);
+		memcpy(ndis_key.material + 16, key + 24, 8);
+		memcpy(ndis_key.material + 24, key + 16, 8);
+	} else
+		memcpy(ndis_key.material, key, key_len);
+
+	if (flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ)
+		memcpy(ndis_key.rsc, rx_seq, seq_len);
+
+	if (flags & NDIS_80211_ADDKEY_PAIRWISE_KEY) {
+		/* pairwise key */
+		memcpy(ndis_key.bssid, addr, ETH_ALEN);
+	} else {
+		/* group key */
+		if (priv->infra_mode == NDIS_80211_INFRA_ADHOC)
+			eth_broadcast_addr(ndis_key.bssid);
+		else
+			get_bssid(usbdev, ndis_key.bssid);
+	}
+
+	ret = rndis_set_oid(usbdev,
+			RNDIS_OID_802_11_ADD_KEY, &ndis_key,
+			le32_to_cpu(ndis_key.size));
+	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_ADD_KEY -> %08X\n",
+		   __func__, ret);
+	if (ret != 0)
+		return ret;
+
+	memset(&priv->encr_keys[index], 0, sizeof(priv->encr_keys[index]));
+	priv->encr_keys[index].len = key_len;
+	priv->encr_keys[index].cipher = cipher;
+	memcpy(&priv->encr_keys[index].material, key, key_len);
+	if (flags & NDIS_80211_ADDKEY_PAIRWISE_KEY)
+		memcpy(&priv->encr_keys[index].bssid, ndis_key.bssid, ETH_ALEN);
+	else
+		eth_broadcast_addr(priv->encr_keys[index].bssid);
+
+	if (flags & NDIS_80211_ADDKEY_TRANSMIT_KEY)
+		priv->encr_tx_key_index = index;
+
+	return 0;
+}
+
+static int restore_key(struct usbnet *usbdev, u8 key_idx)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct rndis_wlan_encr_key key;
+
+	if (is_wpa_key(priv, key_idx))
+		return 0;
+
+	key = priv->encr_keys[key_idx];
+
+	netdev_dbg(usbdev->net, "%s(): %i:%i\n", __func__, key_idx, key.len);
+
+	if (key.len == 0)
+		return 0;
+
+	return add_wep_key(usbdev, key.material, key.len, key_idx);
+}
+
+static void restore_keys(struct usbnet *usbdev)
+{
+	int i;
+
+	for (i = 0; i < 4; i++)
+		restore_key(usbdev, i);
+}
+
+static void clear_key(struct rndis_wlan_private *priv, u8 idx)
+{
+	memset(&priv->encr_keys[idx], 0, sizeof(priv->encr_keys[idx]));
+}
+
+/* remove_key is for both wep and wpa */
+static int remove_key(struct usbnet *usbdev, u8 index, const u8 *bssid)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ndis_80211_remove_key remove_key;
+	__le32 keyindex;
+	bool is_wpa;
+	int ret;
+
+	if (index >= RNDIS_WLAN_NUM_KEYS)
+		return -ENOENT;
+
+	if (priv->encr_keys[index].len == 0)
+		return 0;
+
+	is_wpa = is_wpa_key(priv, index);
+
+	netdev_dbg(usbdev->net, "%s(): %i:%s:%i\n",
+		   __func__, index, is_wpa ? "wpa" : "wep",
+		   priv->encr_keys[index].len);
+
+	clear_key(priv, index);
+
+	if (is_wpa) {
+		remove_key.size = cpu_to_le32(sizeof(remove_key));
+		remove_key.index = cpu_to_le32(index);
+		if (bssid) {
+			/* pairwise key */
+			if (!is_broadcast_ether_addr(bssid))
+				remove_key.index |=
+					NDIS_80211_ADDKEY_PAIRWISE_KEY;
+			memcpy(remove_key.bssid, bssid,
+					sizeof(remove_key.bssid));
+		} else
+			memset(remove_key.bssid, 0xff,
+						sizeof(remove_key.bssid));
+
+		ret = rndis_set_oid(usbdev,
+				RNDIS_OID_802_11_REMOVE_KEY,
+				&remove_key, sizeof(remove_key));
+		if (ret != 0)
+			return ret;
+	} else {
+		keyindex = cpu_to_le32(index);
+		ret = rndis_set_oid(usbdev,
+				RNDIS_OID_802_11_REMOVE_WEP,
+				&keyindex, sizeof(keyindex));
+		if (ret != 0) {
+			netdev_warn(usbdev->net,
+				    "removing encryption key %d failed (%08X)\n",
+				    index, ret);
+			return ret;
+		}
+	}
+
+	/* if it is transmit key, disable encryption */
+	if (index == priv->encr_tx_key_index)
+		set_encr_mode(usbdev, RNDIS_WLAN_ALG_NONE, RNDIS_WLAN_ALG_NONE);
+
+	return 0;
+}
+
+static void set_multicast_list(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct netdev_hw_addr *ha;
+	__le32 filter, basefilter;
+	int ret;
+	char *mc_addrs = NULL;
+	int mc_count;
+
+	basefilter = filter = cpu_to_le32(RNDIS_PACKET_TYPE_DIRECTED |
+					  RNDIS_PACKET_TYPE_BROADCAST);
+
+	if (usbdev->net->flags & IFF_PROMISC) {
+		filter |= cpu_to_le32(RNDIS_PACKET_TYPE_PROMISCUOUS |
+				      RNDIS_PACKET_TYPE_ALL_LOCAL);
+	} else if (usbdev->net->flags & IFF_ALLMULTI) {
+		filter |= cpu_to_le32(RNDIS_PACKET_TYPE_ALL_MULTICAST);
+	}
+
+	if (filter != basefilter)
+		goto set_filter;
+
+	/*
+	 * mc_list should be accessed holding the lock, so copy addresses to
+	 * local buffer first.
+	 */
+	netif_addr_lock_bh(usbdev->net);
+	mc_count = netdev_mc_count(usbdev->net);
+	if (mc_count > priv->multicast_size) {
+		filter |= cpu_to_le32(RNDIS_PACKET_TYPE_ALL_MULTICAST);
+	} else if (mc_count) {
+		int i = 0;
+
+		mc_addrs = kmalloc_array(mc_count, ETH_ALEN, GFP_ATOMIC);
+		if (!mc_addrs) {
+			netif_addr_unlock_bh(usbdev->net);
+			return;
+		}
+
+		netdev_for_each_mc_addr(ha, usbdev->net)
+			memcpy(mc_addrs + i++ * ETH_ALEN,
+			       ha->addr, ETH_ALEN);
+	}
+	netif_addr_unlock_bh(usbdev->net);
+
+	if (filter != basefilter)
+		goto set_filter;
+
+	if (mc_count) {
+		ret = rndis_set_oid(usbdev,
+				RNDIS_OID_802_3_MULTICAST_LIST,
+				mc_addrs, mc_count * ETH_ALEN);
+		kfree(mc_addrs);
+		if (ret == 0)
+			filter |= cpu_to_le32(RNDIS_PACKET_TYPE_MULTICAST);
+		else
+			filter |= cpu_to_le32(RNDIS_PACKET_TYPE_ALL_MULTICAST);
+
+		netdev_dbg(usbdev->net, "RNDIS_OID_802_3_MULTICAST_LIST(%d, max: %d) -> %d\n",
+			   mc_count, priv->multicast_size, ret);
+	}
+
+set_filter:
+	ret = rndis_set_oid(usbdev, RNDIS_OID_GEN_CURRENT_PACKET_FILTER, &filter,
+							sizeof(filter));
+	if (ret < 0) {
+		netdev_warn(usbdev->net, "couldn't set packet filter: %08x\n",
+			    le32_to_cpu(filter));
+	}
+
+	netdev_dbg(usbdev->net, "RNDIS_OID_GEN_CURRENT_PACKET_FILTER(%08x) -> %d\n",
+		   le32_to_cpu(filter), ret);
+}
+
+#ifdef DEBUG
+static void debug_print_pmkids(struct usbnet *usbdev,
+				struct ndis_80211_pmkid *pmkids,
+				const char *func_str)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	int i, len, count, max_pmkids, entry_len;
+
+	max_pmkids = priv->wdev.wiphy->max_num_pmkids;
+	len = le32_to_cpu(pmkids->length);
+	count = le32_to_cpu(pmkids->bssid_info_count);
+
+	entry_len = (count > 0) ? (len - sizeof(*pmkids)) / count : -1;
+
+	netdev_dbg(usbdev->net, "%s(): %d PMKIDs (data len: %d, entry len: "
+				"%d)\n", func_str, count, len, entry_len);
+
+	if (count > max_pmkids)
+		count = max_pmkids;
+
+	for (i = 0; i < count; i++) {
+		u32 *tmp = (u32 *)pmkids->bssid_info[i].pmkid;
+
+		netdev_dbg(usbdev->net, "%s():  bssid: %pM, "
+				"pmkid: %08X:%08X:%08X:%08X\n",
+				func_str, pmkids->bssid_info[i].bssid,
+				cpu_to_be32(tmp[0]), cpu_to_be32(tmp[1]),
+				cpu_to_be32(tmp[2]), cpu_to_be32(tmp[3]));
+	}
+}
+#else
+static void debug_print_pmkids(struct usbnet *usbdev,
+				struct ndis_80211_pmkid *pmkids,
+				const char *func_str)
+{
+	return;
+}
+#endif
+
+static struct ndis_80211_pmkid *get_device_pmkids(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ndis_80211_pmkid *pmkids;
+	int len, ret, max_pmkids;
+
+	max_pmkids = priv->wdev.wiphy->max_num_pmkids;
+	len = struct_size(pmkids, bssid_info, max_pmkids);
+
+	pmkids = kzalloc(len, GFP_KERNEL);
+	if (!pmkids)
+		return ERR_PTR(-ENOMEM);
+
+	pmkids->length = cpu_to_le32(len);
+	pmkids->bssid_info_count = cpu_to_le32(max_pmkids);
+
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_PMKID,
+			pmkids, &len);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_PMKID(%d, %d)"
+				" -> %d\n", __func__, len, max_pmkids, ret);
+
+		kfree(pmkids);
+		return ERR_PTR(ret);
+	}
+
+	if (le32_to_cpu(pmkids->bssid_info_count) > max_pmkids)
+		pmkids->bssid_info_count = cpu_to_le32(max_pmkids);
+
+	debug_print_pmkids(usbdev, pmkids, __func__);
+
+	return pmkids;
+}
+
+static int set_device_pmkids(struct usbnet *usbdev,
+				struct ndis_80211_pmkid *pmkids)
+{
+	int ret, len, num_pmkids;
+
+	num_pmkids = le32_to_cpu(pmkids->bssid_info_count);
+	len = struct_size(pmkids, bssid_info, num_pmkids);
+	pmkids->length = cpu_to_le32(len);
+
+	debug_print_pmkids(usbdev, pmkids, __func__);
+
+	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_PMKID, pmkids,
+			    le32_to_cpu(pmkids->length));
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_PMKID(%d, %d) -> %d"
+				"\n", __func__, len, num_pmkids, ret);
+	}
+
+	kfree(pmkids);
+	return ret;
+}
+
+static struct ndis_80211_pmkid *remove_pmkid(struct usbnet *usbdev,
+						struct ndis_80211_pmkid *pmkids,
+						struct cfg80211_pmksa *pmksa,
+						int max_pmkids)
+{
+	int i, err;
+	unsigned int count;
+
+	count = le32_to_cpu(pmkids->bssid_info_count);
+
+	if (count > max_pmkids)
+		count = max_pmkids;
+
+	for (i = 0; i < count; i++)
+		if (ether_addr_equal(pmkids->bssid_info[i].bssid,
+				     pmksa->bssid))
+			break;
+
+	/* pmkid not found */
+	if (i == count) {
+		netdev_dbg(usbdev->net, "%s(): bssid not found (%pM)\n",
+					__func__, pmksa->bssid);
+		err = -ENOENT;
+		goto error;
+	}
+
+	for (; i + 1 < count; i++)
+		pmkids->bssid_info[i] = pmkids->bssid_info[i + 1];
+
+	count--;
+	pmkids->length = cpu_to_le32(struct_size(pmkids, bssid_info, count));
+	pmkids->bssid_info_count = cpu_to_le32(count);
+
+	return pmkids;
+error:
+	kfree(pmkids);
+	return ERR_PTR(err);
+}
+
+static struct ndis_80211_pmkid *update_pmkid(struct usbnet *usbdev,
+						struct ndis_80211_pmkid *pmkids,
+						struct cfg80211_pmksa *pmksa,
+						int max_pmkids)
+{
+	struct ndis_80211_pmkid *new_pmkids;
+	int i, err, newlen;
+	unsigned int count;
+
+	count = le32_to_cpu(pmkids->bssid_info_count);
+
+	if (count > max_pmkids)
+		count = max_pmkids;
+
+	/* update with new pmkid */
+	for (i = 0; i < count; i++) {
+		if (!ether_addr_equal(pmkids->bssid_info[i].bssid,
+				      pmksa->bssid))
+			continue;
+
+		memcpy(pmkids->bssid_info[i].pmkid, pmksa->pmkid,
+								WLAN_PMKID_LEN);
+
+		return pmkids;
+	}
+
+	/* out of space, return error */
+	if (i == max_pmkids) {
+		netdev_dbg(usbdev->net, "%s(): out of space\n", __func__);
+		err = -ENOSPC;
+		goto error;
+	}
+
+	/* add new pmkid */
+	newlen = struct_size(pmkids, bssid_info, count + 1);
+
+	new_pmkids = krealloc(pmkids, newlen, GFP_KERNEL);
+	if (!new_pmkids) {
+		err = -ENOMEM;
+		goto error;
+	}
+	pmkids = new_pmkids;
+
+	pmkids->length = cpu_to_le32(newlen);
+	pmkids->bssid_info_count = cpu_to_le32(count + 1);
+
+	memcpy(pmkids->bssid_info[count].bssid, pmksa->bssid, ETH_ALEN);
+	memcpy(pmkids->bssid_info[count].pmkid, pmksa->pmkid, WLAN_PMKID_LEN);
+
+	return pmkids;
+error:
+	kfree(pmkids);
+	return ERR_PTR(err);
+}
+
+/*
+ * cfg80211 ops
+ */
+static int rndis_change_virtual_intf(struct wiphy *wiphy,
+					struct net_device *dev,
+					enum nl80211_iftype type,
+					struct vif_params *params)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	int mode;
+
+	switch (type) {
+	case NL80211_IFTYPE_ADHOC:
+		mode = NDIS_80211_INFRA_ADHOC;
+		break;
+	case NL80211_IFTYPE_STATION:
+		mode = NDIS_80211_INFRA_INFRA;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	priv->wdev.iftype = type;
+
+	return set_infra_mode(usbdev, mode);
+}
+
+static int rndis_set_wiphy_params(struct wiphy *wiphy, u32 changed)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	int err;
+
+	if (changed & WIPHY_PARAM_FRAG_THRESHOLD) {
+		err = set_frag_threshold(usbdev, wiphy->frag_threshold);
+		if (err < 0)
+			return err;
+	}
+
+	if (changed & WIPHY_PARAM_RTS_THRESHOLD) {
+		err = set_rts_threshold(usbdev, wiphy->rts_threshold);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
+static int rndis_set_tx_power(struct wiphy *wiphy,
+			      struct wireless_dev *wdev,
+			      enum nl80211_tx_power_setting type,
+			      int mbm)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	netdev_dbg(usbdev->net, "%s(): type:0x%x mbm:%i\n",
+		   __func__, type, mbm);
+
+	if (mbm < 0 || (mbm % 100))
+		return -ENOTSUPP;
+
+	/* Device doesn't support changing txpower after initialization, only
+	 * turn off/on radio. Support 'auto' mode and setting same dBm that is
+	 * currently used.
+	 */
+	if (type == NL80211_TX_POWER_AUTOMATIC ||
+	    MBM_TO_DBM(mbm) == get_bcm4320_power_dbm(priv)) {
+		if (!priv->radio_on)
+			disassociate(usbdev, true); /* turn on radio */
+
+		return 0;
+	}
+
+	return -ENOTSUPP;
+}
+
+static int rndis_get_tx_power(struct wiphy *wiphy,
+			      struct wireless_dev *wdev,
+			      int *dbm)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	*dbm = get_bcm4320_power_dbm(priv);
+
+	netdev_dbg(usbdev->net, "%s(): dbm:%i\n", __func__, *dbm);
+
+	return 0;
+}
+
+#define SCAN_DELAY_JIFFIES (6 * HZ)
+static int rndis_scan(struct wiphy *wiphy,
+			struct cfg80211_scan_request *request)
+{
+	struct net_device *dev = request->wdev->netdev;
+	struct usbnet *usbdev = netdev_priv(dev);
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	int ret;
+	int delay = SCAN_DELAY_JIFFIES;
+
+	netdev_dbg(usbdev->net, "cfg80211.scan\n");
+
+	/* Get current bssid list from device before new scan, as new scan
+	 * clears internal bssid list.
+	 */
+	rndis_check_bssid_list(usbdev, NULL, NULL);
+
+	if (priv->scan_request && priv->scan_request != request)
+		return -EBUSY;
+
+	priv->scan_request = request;
+
+	ret = rndis_start_bssid_list_scan(usbdev);
+	if (ret == 0) {
+		if (priv->device_type == RNDIS_BCM4320A)
+			delay = HZ;
+
+		/* Wait before retrieving scan results from device */
+		queue_delayed_work(priv->workqueue, &priv->scan_work, delay);
+	}
+
+	return ret;
+}
+
+static bool rndis_bss_info_update(struct usbnet *usbdev,
+				  struct ndis_80211_bssid_ex *bssid)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ieee80211_channel *channel;
+	struct cfg80211_bss *bss;
+	s32 signal;
+	u64 timestamp;
+	u16 capability;
+	u16 beacon_interval;
+	struct ndis_80211_fixed_ies *fixed;
+	int ie_len, bssid_len;
+	u8 *ie;
+
+	netdev_dbg(usbdev->net, " found bssid: '%.32s' [%pM], len: %d\n",
+		   bssid->ssid.essid, bssid->mac, le32_to_cpu(bssid->length));
+
+	/* parse bssid structure */
+	bssid_len = le32_to_cpu(bssid->length);
+
+	if (bssid_len < sizeof(struct ndis_80211_bssid_ex) +
+			sizeof(struct ndis_80211_fixed_ies))
+		return NULL;
+
+	fixed = (struct ndis_80211_fixed_ies *)bssid->ies;
+
+	ie = (void *)(bssid->ies + sizeof(struct ndis_80211_fixed_ies));
+	ie_len = min(bssid_len - (int)sizeof(*bssid),
+					(int)le32_to_cpu(bssid->ie_length));
+	ie_len -= sizeof(struct ndis_80211_fixed_ies);
+	if (ie_len < 0)
+		return NULL;
+
+	/* extract data for cfg80211_inform_bss */
+	channel = ieee80211_get_channel(priv->wdev.wiphy,
+			KHZ_TO_MHZ(le32_to_cpu(bssid->config.ds_config)));
+	if (!channel)
+		return NULL;
+
+	signal = level_to_qual(le32_to_cpu(bssid->rssi));
+	timestamp = le64_to_cpu(*(__le64 *)fixed->timestamp);
+	capability = le16_to_cpu(fixed->capabilities);
+	beacon_interval = le16_to_cpu(fixed->beacon_interval);
+
+	bss = cfg80211_inform_bss(priv->wdev.wiphy, channel,
+				  CFG80211_BSS_FTYPE_UNKNOWN, bssid->mac,
+				  timestamp, capability, beacon_interval,
+				  ie, ie_len, signal, GFP_KERNEL);
+	cfg80211_put_bss(priv->wdev.wiphy, bss);
+
+	return (bss != NULL);
+}
+
+static struct ndis_80211_bssid_ex *next_bssid_list_item(
+					struct ndis_80211_bssid_ex *bssid,
+					int *bssid_len, void *buf, int len)
+{
+	void *buf_end, *bssid_end;
+
+	buf_end = (char *)buf + len;
+	bssid_end = (char *)bssid + *bssid_len;
+
+	if ((int)(buf_end - bssid_end) < sizeof(bssid->length)) {
+		*bssid_len = 0;
+		return NULL;
+	} else {
+		bssid = (void *)((char *)bssid + *bssid_len);
+		*bssid_len = le32_to_cpu(bssid->length);
+		return bssid;
+	}
+}
+
+static bool check_bssid_list_item(struct ndis_80211_bssid_ex *bssid,
+				  int bssid_len, void *buf, int len)
+{
+	void *buf_end, *bssid_end;
+
+	if (!bssid || bssid_len <= 0 || bssid_len > len)
+		return false;
+
+	buf_end = (char *)buf + len;
+	bssid_end = (char *)bssid + bssid_len;
+
+	return (int)(buf_end - bssid_end) >= 0 && (int)(bssid_end - buf) >= 0;
+}
+
+static int rndis_check_bssid_list(struct usbnet *usbdev, u8 *match_bssid,
+					bool *matched)
+{
+	void *buf = NULL;
+	struct ndis_80211_bssid_list_ex *bssid_list;
+	struct ndis_80211_bssid_ex *bssid;
+	int ret = -EINVAL, len, count, bssid_len, real_count, new_len;
+
+	netdev_dbg(usbdev->net, "%s()\n", __func__);
+
+	len = CONTROL_BUFFER_SIZE;
+resize_buf:
+	buf = kzalloc(len, GFP_KERNEL);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	/* BSSID-list might have got bigger last time we checked, keep
+	 * resizing until it won't get any bigger.
+	 */
+	new_len = len;
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_BSSID_LIST,
+			      buf, &new_len);
+	if (ret != 0 || new_len < sizeof(struct ndis_80211_bssid_list_ex))
+		goto out;
+
+	if (new_len > len) {
+		len = new_len;
+		kfree(buf);
+		goto resize_buf;
+	}
+
+	len = new_len;
+
+	bssid_list = buf;
+	count = le32_to_cpu(bssid_list->num_items);
+	real_count = 0;
+	netdev_dbg(usbdev->net, "%s(): buflen: %d\n", __func__, len);
+
+	bssid_len = 0;
+	bssid = next_bssid_list_item((void *)bssid_list->bssid_data,
+				     &bssid_len, buf, len);
+
+	/* Device returns incorrect 'num_items'. Workaround by ignoring the
+	 * received 'num_items' and walking through full bssid buffer instead.
+	 */
+	while (check_bssid_list_item(bssid, bssid_len, buf, len)) {
+		if (rndis_bss_info_update(usbdev, bssid) && match_bssid &&
+		    matched) {
+			if (ether_addr_equal(bssid->mac, match_bssid))
+				*matched = true;
+		}
+
+		real_count++;
+		bssid = next_bssid_list_item(bssid, &bssid_len, buf, len);
+	}
+
+	netdev_dbg(usbdev->net, "%s(): num_items from device: %d, really found:"
+				" %d\n", __func__, count, real_count);
+
+out:
+	kfree(buf);
+	return ret;
+}
+
+static void rndis_get_scan_results(struct work_struct *work)
+{
+	struct rndis_wlan_private *priv =
+		container_of(work, struct rndis_wlan_private, scan_work.work);
+	struct usbnet *usbdev = priv->usbdev;
+	struct cfg80211_scan_info info = {};
+	int ret;
+
+	netdev_dbg(usbdev->net, "get_scan_results\n");
+
+	if (!priv->scan_request)
+		return;
+
+	ret = rndis_check_bssid_list(usbdev, NULL, NULL);
+
+	info.aborted = ret < 0;
+	cfg80211_scan_done(priv->scan_request, &info);
+
+	priv->scan_request = NULL;
+}
+
+static int rndis_connect(struct wiphy *wiphy, struct net_device *dev,
+					struct cfg80211_connect_params *sme)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	struct ieee80211_channel *channel = sme->channel;
+	struct ndis_80211_ssid ssid;
+	int pairwise = RNDIS_WLAN_ALG_NONE;
+	int groupwise = RNDIS_WLAN_ALG_NONE;
+	int keymgmt = RNDIS_WLAN_KEY_MGMT_NONE;
+	int length, i, ret, chan = -1;
+
+	if (channel)
+		chan = ieee80211_frequency_to_channel(channel->center_freq);
+
+	groupwise = rndis_cipher_to_alg(sme->crypto.cipher_group);
+	for (i = 0; i < sme->crypto.n_ciphers_pairwise; i++)
+		pairwise |=
+			rndis_cipher_to_alg(sme->crypto.ciphers_pairwise[i]);
+
+	if (sme->crypto.n_ciphers_pairwise > 0 &&
+			pairwise == RNDIS_WLAN_ALG_NONE) {
+		netdev_err(usbdev->net, "Unsupported pairwise cipher\n");
+		return -ENOTSUPP;
+	}
+
+	for (i = 0; i < sme->crypto.n_akm_suites; i++)
+		keymgmt |=
+			rndis_akm_suite_to_key_mgmt(sme->crypto.akm_suites[i]);
+
+	if (sme->crypto.n_akm_suites > 0 &&
+			keymgmt == RNDIS_WLAN_KEY_MGMT_NONE) {
+		netdev_err(usbdev->net, "Invalid keymgmt\n");
+		return -ENOTSUPP;
+	}
+
+	netdev_dbg(usbdev->net, "cfg80211.connect('%.32s':[%pM]:%d:[%d,0x%x:0x%x]:[0x%x:0x%x]:0x%x)\n",
+		   sme->ssid, sme->bssid, chan,
+		   sme->privacy, sme->crypto.wpa_versions, sme->auth_type,
+		   groupwise, pairwise, keymgmt);
+
+	if (is_associated(usbdev))
+		disassociate(usbdev, false);
+
+	ret = set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "connect: set_infra_mode failed, %d\n",
+			   ret);
+		goto err_turn_radio_on;
+	}
+
+	ret = set_auth_mode(usbdev, sme->crypto.wpa_versions, sme->auth_type,
+								keymgmt);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "connect: set_auth_mode failed, %d\n",
+			   ret);
+		goto err_turn_radio_on;
+	}
+
+	set_priv_filter(usbdev);
+
+	ret = set_encr_mode(usbdev, pairwise, groupwise);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "connect: set_encr_mode failed, %d\n",
+			   ret);
+		goto err_turn_radio_on;
+	}
+
+	if (channel) {
+		ret = set_channel(usbdev, chan);
+		if (ret < 0) {
+			netdev_dbg(usbdev->net, "connect: set_channel failed, %d\n",
+				   ret);
+			goto err_turn_radio_on;
+		}
+	}
+
+	if (sme->key && ((groupwise | pairwise) & RNDIS_WLAN_ALG_WEP)) {
+		priv->encr_tx_key_index = sme->key_idx;
+		ret = add_wep_key(usbdev, sme->key, sme->key_len, sme->key_idx);
+		if (ret < 0) {
+			netdev_dbg(usbdev->net, "connect: add_wep_key failed, %d (%d, %d)\n",
+				   ret, sme->key_len, sme->key_idx);
+			goto err_turn_radio_on;
+		}
+	}
+
+	if (sme->bssid && !is_zero_ether_addr(sme->bssid) &&
+				!is_broadcast_ether_addr(sme->bssid)) {
+		ret = set_bssid(usbdev, sme->bssid);
+		if (ret < 0) {
+			netdev_dbg(usbdev->net, "connect: set_bssid failed, %d\n",
+				   ret);
+			goto err_turn_radio_on;
+		}
+	} else
+		clear_bssid(usbdev);
+
+	length = sme->ssid_len;
+	if (length > NDIS_802_11_LENGTH_SSID)
+		length = NDIS_802_11_LENGTH_SSID;
+
+	memset(&ssid, 0, sizeof(ssid));
+	ssid.length = cpu_to_le32(length);
+	memcpy(ssid.essid, sme->ssid, length);
+
+	/* Pause and purge rx queue, so we don't pass packets before
+	 * 'media connect'-indication.
+	 */
+	usbnet_pause_rx(usbdev);
+	usbnet_purge_paused_rxq(usbdev);
+
+	ret = set_essid(usbdev, &ssid);
+	if (ret < 0)
+		netdev_dbg(usbdev->net, "connect: set_essid failed, %d\n", ret);
+	return ret;
+
+err_turn_radio_on:
+	disassociate(usbdev, true);
+
+	return ret;
+}
+
+static int rndis_disconnect(struct wiphy *wiphy, struct net_device *dev,
+								u16 reason_code)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	netdev_dbg(usbdev->net, "cfg80211.disconnect(%d)\n", reason_code);
+
+	priv->connected = false;
+	eth_zero_addr(priv->bssid);
+
+	return deauthenticate(usbdev);
+}
+
+static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev,
+					struct cfg80211_ibss_params *params)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	struct ieee80211_channel *channel = params->chandef.chan;
+	struct ndis_80211_ssid ssid;
+	enum nl80211_auth_type auth_type;
+	int ret, alg, length, chan = -1;
+
+	if (channel)
+		chan = ieee80211_frequency_to_channel(channel->center_freq);
+
+	/* TODO: How to handle ad-hoc encryption?
+	 * connect() has *key, join_ibss() doesn't. RNDIS requires key to be
+	 * pre-shared for encryption (open/shared/wpa), is key set before
+	 * join_ibss? Which auth_type to use (not in params)? What about WPA?
+	 */
+	if (params->privacy) {
+		auth_type = NL80211_AUTHTYPE_SHARED_KEY;
+		alg = RNDIS_WLAN_ALG_WEP;
+	} else {
+		auth_type = NL80211_AUTHTYPE_OPEN_SYSTEM;
+		alg = RNDIS_WLAN_ALG_NONE;
+	}
+
+	netdev_dbg(usbdev->net, "cfg80211.join_ibss('%.32s':[%pM]:%d:%d)\n",
+		   params->ssid, params->bssid, chan, params->privacy);
+
+	if (is_associated(usbdev))
+		disassociate(usbdev, false);
+
+	ret = set_infra_mode(usbdev, NDIS_80211_INFRA_ADHOC);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "join_ibss: set_infra_mode failed, %d\n",
+			   ret);
+		goto err_turn_radio_on;
+	}
+
+	ret = set_auth_mode(usbdev, 0, auth_type, RNDIS_WLAN_KEY_MGMT_NONE);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "join_ibss: set_auth_mode failed, %d\n",
+			   ret);
+		goto err_turn_radio_on;
+	}
+
+	set_priv_filter(usbdev);
+
+	ret = set_encr_mode(usbdev, alg, RNDIS_WLAN_ALG_NONE);
+	if (ret < 0) {
+		netdev_dbg(usbdev->net, "join_ibss: set_encr_mode failed, %d\n",
+			   ret);
+		goto err_turn_radio_on;
+	}
+
+	if (channel) {
+		ret = set_channel(usbdev, chan);
+		if (ret < 0) {
+			netdev_dbg(usbdev->net, "join_ibss: set_channel failed, %d\n",
+				   ret);
+			goto err_turn_radio_on;
+		}
+	}
+
+	if (params->bssid && !is_zero_ether_addr(params->bssid) &&
+				!is_broadcast_ether_addr(params->bssid)) {
+		ret = set_bssid(usbdev, params->bssid);
+		if (ret < 0) {
+			netdev_dbg(usbdev->net, "join_ibss: set_bssid failed, %d\n",
+				   ret);
+			goto err_turn_radio_on;
+		}
+	} else
+		clear_bssid(usbdev);
+
+	length = params->ssid_len;
+	if (length > NDIS_802_11_LENGTH_SSID)
+		length = NDIS_802_11_LENGTH_SSID;
+
+	memset(&ssid, 0, sizeof(ssid));
+	ssid.length = cpu_to_le32(length);
+	memcpy(ssid.essid, params->ssid, length);
+
+	/* Don't need to pause rx queue for ad-hoc. */
+	usbnet_purge_paused_rxq(usbdev);
+	usbnet_resume_rx(usbdev);
+
+	ret = set_essid(usbdev, &ssid);
+	if (ret < 0)
+		netdev_dbg(usbdev->net, "join_ibss: set_essid failed, %d\n",
+			   ret);
+	return ret;
+
+err_turn_radio_on:
+	disassociate(usbdev, true);
+
+	return ret;
+}
+
+static int rndis_leave_ibss(struct wiphy *wiphy, struct net_device *dev)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	netdev_dbg(usbdev->net, "cfg80211.leave_ibss()\n");
+
+	priv->connected = false;
+	eth_zero_addr(priv->bssid);
+
+	return deauthenticate(usbdev);
+}
+
+static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
+			 int link_id,  u8 key_index, bool pairwise,
+			 const u8 *mac_addr, struct key_params *params)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	__le32 flags;
+
+	netdev_dbg(usbdev->net, "%s(%i, %pM, %08x)\n",
+		   __func__, key_index, mac_addr, params->cipher);
+
+	switch (params->cipher) {
+	case WLAN_CIPHER_SUITE_WEP40:
+	case WLAN_CIPHER_SUITE_WEP104:
+		return add_wep_key(usbdev, params->key, params->key_len,
+								key_index);
+	case WLAN_CIPHER_SUITE_TKIP:
+	case WLAN_CIPHER_SUITE_CCMP:
+		flags = 0;
+
+		if (params->seq && params->seq_len > 0)
+			flags |= NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ;
+		if (mac_addr)
+			flags |= NDIS_80211_ADDKEY_PAIRWISE_KEY |
+					NDIS_80211_ADDKEY_TRANSMIT_KEY;
+
+		return add_wpa_key(usbdev, params->key, params->key_len,
+				key_index, mac_addr, params->seq,
+				params->seq_len, params->cipher, flags);
+	default:
+		netdev_dbg(usbdev->net, "%s(): unsupported cipher %08x\n",
+			   __func__, params->cipher);
+		return -ENOTSUPP;
+	}
+}
+
+static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev,
+			 int link_id, u8 key_index, bool pairwise,
+			 const u8 *mac_addr)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	netdev_dbg(usbdev->net, "%s(%i, %pM)\n", __func__, key_index, mac_addr);
+
+	return remove_key(usbdev, key_index, mac_addr);
+}
+
+static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev,
+				 int link_id, u8 key_index, bool unicast,
+				 bool multicast)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	struct rndis_wlan_encr_key key;
+
+	netdev_dbg(usbdev->net, "%s(%i)\n", __func__, key_index);
+
+	if (key_index >= RNDIS_WLAN_NUM_KEYS)
+		return -ENOENT;
+
+	priv->encr_tx_key_index = key_index;
+
+	if (is_wpa_key(priv, key_index))
+		return 0;
+
+	key = priv->encr_keys[key_index];
+
+	return add_wep_key(usbdev, key.material, key.len, key_index);
+}
+
+static void rndis_fill_station_info(struct usbnet *usbdev,
+						struct station_info *sinfo)
+{
+	__le32 linkspeed, rssi;
+	int ret, len;
+
+	memset(sinfo, 0, sizeof(*sinfo));
+
+	len = sizeof(linkspeed);
+	ret = rndis_query_oid(usbdev, RNDIS_OID_GEN_LINK_SPEED, &linkspeed, &len);
+	if (ret == 0) {
+		sinfo->txrate.legacy = le32_to_cpu(linkspeed) / 1000;
+		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
+	}
+
+	len = sizeof(rssi);
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_RSSI,
+			      &rssi, &len);
+	if (ret == 0) {
+		sinfo->signal = level_to_qual(le32_to_cpu(rssi));
+		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_SIGNAL);
+	}
+}
+
+static int rndis_get_station(struct wiphy *wiphy, struct net_device *dev,
+			     const u8 *mac, struct station_info *sinfo)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	if (!ether_addr_equal(priv->bssid, mac))
+		return -ENOENT;
+
+	rndis_fill_station_info(usbdev, sinfo);
+
+	return 0;
+}
+
+static int rndis_dump_station(struct wiphy *wiphy, struct net_device *dev,
+			       int idx, u8 *mac, struct station_info *sinfo)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+
+	if (idx != 0)
+		return -ENOENT;
+
+	memcpy(mac, priv->bssid, ETH_ALEN);
+
+	rndis_fill_station_info(usbdev, sinfo);
+
+	return 0;
+}
+
+static int rndis_set_pmksa(struct wiphy *wiphy, struct net_device *netdev,
+				struct cfg80211_pmksa *pmksa)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	struct ndis_80211_pmkid *pmkids;
+	u32 *tmp = (u32 *)pmksa->pmkid;
+
+	netdev_dbg(usbdev->net, "%s(%pM, %08X:%08X:%08X:%08X)\n", __func__,
+			pmksa->bssid,
+			cpu_to_be32(tmp[0]), cpu_to_be32(tmp[1]),
+			cpu_to_be32(tmp[2]), cpu_to_be32(tmp[3]));
+
+	pmkids = get_device_pmkids(usbdev);
+	if (IS_ERR(pmkids)) {
+		/* couldn't read PMKID cache from device */
+		return PTR_ERR(pmkids);
+	}
+
+	pmkids = update_pmkid(usbdev, pmkids, pmksa, wiphy->max_num_pmkids);
+	if (IS_ERR(pmkids)) {
+		/* not found, list full, etc */
+		return PTR_ERR(pmkids);
+	}
+
+	return set_device_pmkids(usbdev, pmkids);
+}
+
+static int rndis_del_pmksa(struct wiphy *wiphy, struct net_device *netdev,
+				struct cfg80211_pmksa *pmksa)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	struct ndis_80211_pmkid *pmkids;
+	u32 *tmp = (u32 *)pmksa->pmkid;
+
+	netdev_dbg(usbdev->net, "%s(%pM, %08X:%08X:%08X:%08X)\n", __func__,
+			pmksa->bssid,
+			cpu_to_be32(tmp[0]), cpu_to_be32(tmp[1]),
+			cpu_to_be32(tmp[2]), cpu_to_be32(tmp[3]));
+
+	pmkids = get_device_pmkids(usbdev);
+	if (IS_ERR(pmkids)) {
+		/* Couldn't read PMKID cache from device */
+		return PTR_ERR(pmkids);
+	}
+
+	pmkids = remove_pmkid(usbdev, pmkids, pmksa, wiphy->max_num_pmkids);
+	if (IS_ERR(pmkids)) {
+		/* not found, etc */
+		return PTR_ERR(pmkids);
+	}
+
+	return set_device_pmkids(usbdev, pmkids);
+}
+
+static int rndis_flush_pmksa(struct wiphy *wiphy, struct net_device *netdev)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	struct ndis_80211_pmkid pmkid;
+
+	netdev_dbg(usbdev->net, "%s()\n", __func__);
+
+	memset(&pmkid, 0, sizeof(pmkid));
+
+	pmkid.length = cpu_to_le32(sizeof(pmkid));
+	pmkid.bssid_info_count = cpu_to_le32(0);
+
+	return rndis_set_oid(usbdev, RNDIS_OID_802_11_PMKID,
+			     &pmkid, sizeof(pmkid));
+}
+
+static int rndis_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev,
+				bool enabled, int timeout)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+	struct usbnet *usbdev = priv->usbdev;
+	int power_mode;
+	__le32 mode;
+	int ret;
+
+	if (priv->device_type != RNDIS_BCM4320B)
+		return -ENOTSUPP;
+
+	netdev_dbg(usbdev->net, "%s(): %s, %d\n", __func__,
+				enabled ? "enabled" : "disabled",
+				timeout);
+
+	if (enabled)
+		power_mode = NDIS_80211_POWER_MODE_FAST_PSP;
+	else
+		power_mode = NDIS_80211_POWER_MODE_CAM;
+
+	if (power_mode == priv->power_mode)
+		return 0;
+
+	priv->power_mode = power_mode;
+
+	mode = cpu_to_le32(power_mode);
+	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_POWER_MODE,
+			    &mode, sizeof(mode));
+
+	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_POWER_MODE -> %d\n",
+				__func__, ret);
+
+	return ret;
+}
+
+static int rndis_set_cqm_rssi_config(struct wiphy *wiphy,
+					struct net_device *dev,
+					s32 rssi_thold, u32 rssi_hyst)
+{
+	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
+
+	priv->cqm_rssi_thold = rssi_thold;
+	priv->cqm_rssi_hyst = rssi_hyst;
+	priv->last_cqm_event_rssi = 0;
+
+	return 0;
+}
+
+static void rndis_wlan_craft_connected_bss(struct usbnet *usbdev, u8 *bssid,
+					   struct ndis_80211_assoc_info *info)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ieee80211_channel *channel;
+	struct ndis_80211_ssid ssid;
+	struct cfg80211_bss *bss;
+	s32 signal;
+	u64 timestamp;
+	u16 capability;
+	u32 beacon_period = 0;
+	__le32 rssi;
+	u8 ie_buf[34];
+	int len, ret, ie_len;
+
+	/* Get signal quality, in case of error use rssi=0 and ignore error. */
+	len = sizeof(rssi);
+	rssi = 0;
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_RSSI,
+			      &rssi, &len);
+	signal = level_to_qual(le32_to_cpu(rssi));
+
+	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_RSSI -> %d, "
+		   "rssi:%d, qual: %d\n", __func__, ret, le32_to_cpu(rssi),
+		   level_to_qual(le32_to_cpu(rssi)));
+
+	/* Get AP capabilities */
+	if (info) {
+		capability = le16_to_cpu(info->resp_ie.capa);
+	} else {
+		/* Set atleast ESS/IBSS capability */
+		capability = (priv->infra_mode == NDIS_80211_INFRA_INFRA) ?
+				WLAN_CAPABILITY_ESS : WLAN_CAPABILITY_IBSS;
+	}
+
+	/* Get channel and beacon interval */
+	channel = get_current_channel(usbdev, &beacon_period);
+	if (!channel) {
+		netdev_warn(usbdev->net, "%s(): could not get channel.\n",
+					__func__);
+		return;
+	}
+
+	/* Get SSID, in case of error, use zero length SSID and ignore error. */
+	len = sizeof(ssid);
+	memset(&ssid, 0, sizeof(ssid));
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_SSID,
+			      &ssid, &len);
+	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_SSID -> %d, len: %d, ssid: "
+				"'%.32s'\n", __func__, ret,
+				le32_to_cpu(ssid.length), ssid.essid);
+
+	if (le32_to_cpu(ssid.length) > 32)
+		ssid.length = cpu_to_le32(32);
+
+	ie_buf[0] = WLAN_EID_SSID;
+	ie_buf[1] = le32_to_cpu(ssid.length);
+	memcpy(&ie_buf[2], ssid.essid, le32_to_cpu(ssid.length));
+
+	ie_len = le32_to_cpu(ssid.length) + 2;
+
+	/* no tsf */
+	timestamp = 0;
+
+	netdev_dbg(usbdev->net, "%s(): channel:%d(freq), bssid:[%pM], tsf:%d, "
+		"capa:%x, beacon int:%d, resp_ie(len:%d, essid:'%.32s'), "
+		"signal:%d\n", __func__, (channel ? channel->center_freq : -1),
+		bssid, (u32)timestamp, capability, beacon_period, ie_len,
+		ssid.essid, signal);
+
+	bss = cfg80211_inform_bss(priv->wdev.wiphy, channel,
+				  CFG80211_BSS_FTYPE_UNKNOWN, bssid,
+				  timestamp, capability, beacon_period,
+				  ie_buf, ie_len, signal, GFP_KERNEL);
+	cfg80211_put_bss(priv->wdev.wiphy, bss);
+}
+
+/*
+ * workers, indication handlers, device poller
+ */
+static void rndis_wlan_do_link_up_work(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct ndis_80211_assoc_info *info = NULL;
+	u8 bssid[ETH_ALEN];
+	unsigned int resp_ie_len, req_ie_len;
+	unsigned int offset;
+	u8 *req_ie, *resp_ie;
+	int ret;
+	bool roamed = false;
+	bool match_bss;
+
+	if (priv->infra_mode == NDIS_80211_INFRA_INFRA && priv->connected) {
+		/* received media connect indication while connected, either
+		 * device reassociated with same AP or roamed to new. */
+		roamed = true;
+	}
+
+	req_ie_len = 0;
+	resp_ie_len = 0;
+	req_ie = NULL;
+	resp_ie = NULL;
+
+	if (priv->infra_mode == NDIS_80211_INFRA_INFRA) {
+		info = kzalloc(CONTROL_BUFFER_SIZE, GFP_KERNEL);
+		if (!info) {
+			/* No memory? Try resume work later */
+			set_bit(WORK_LINK_UP, &priv->work_pending);
+			queue_work(priv->workqueue, &priv->work);
+			return;
+		}
+
+		/* Get association info IEs from device. */
+		ret = get_association_info(usbdev, info, CONTROL_BUFFER_SIZE);
+		if (!ret) {
+			req_ie_len = le32_to_cpu(info->req_ie_length);
+			if (req_ie_len > CONTROL_BUFFER_SIZE)
+				req_ie_len = CONTROL_BUFFER_SIZE;
+			if (req_ie_len != 0) {
+				offset = le32_to_cpu(info->offset_req_ies);
+
+				if (offset > CONTROL_BUFFER_SIZE)
+					offset = CONTROL_BUFFER_SIZE;
+
+				req_ie = (u8 *)info + offset;
+
+				if (offset + req_ie_len > CONTROL_BUFFER_SIZE)
+					req_ie_len =
+						CONTROL_BUFFER_SIZE - offset;
+			}
+
+			resp_ie_len = le32_to_cpu(info->resp_ie_length);
+			if (resp_ie_len > CONTROL_BUFFER_SIZE)
+				resp_ie_len = CONTROL_BUFFER_SIZE;
+			if (resp_ie_len != 0) {
+				offset = le32_to_cpu(info->offset_resp_ies);
+
+				if (offset > CONTROL_BUFFER_SIZE)
+					offset = CONTROL_BUFFER_SIZE;
+
+				resp_ie = (u8 *)info + offset;
+
+				if (offset + resp_ie_len > CONTROL_BUFFER_SIZE)
+					resp_ie_len =
+						CONTROL_BUFFER_SIZE - offset;
+			}
+		} else {
+			/* Since rndis_wlan_craft_connected_bss() might use info
+			 * later and expects info to contain valid data if
+			 * non-null, free info and set NULL here.
+			 */
+			kfree(info);
+			info = NULL;
+		}
+	} else if (WARN_ON(priv->infra_mode != NDIS_80211_INFRA_ADHOC))
+		return;
+
+	ret = get_bssid(usbdev, bssid);
+	if (ret < 0)
+		memset(bssid, 0, sizeof(bssid));
+
+	netdev_dbg(usbdev->net, "link up work: [%pM]%s\n",
+		   bssid, roamed ? " roamed" : "");
+
+	/* Internal bss list in device should contain at least the currently
+	 * connected bss and we can get it to cfg80211 with
+	 * rndis_check_bssid_list().
+	 *
+	 * NDIS spec says: "If the device is associated, but the associated
+	 *  BSSID is not in its BSSID scan list, then the driver must add an
+	 *  entry for the BSSID at the end of the data that it returns in
+	 *  response to query of RNDIS_OID_802_11_BSSID_LIST."
+	 *
+	 * NOTE: Seems to be true for BCM4320b variant, but not BCM4320a.
+	 */
+	match_bss = false;
+	rndis_check_bssid_list(usbdev, bssid, &match_bss);
+
+	if (!is_zero_ether_addr(bssid) && !match_bss) {
+		/* Couldn't get bss from device, we need to manually craft bss
+		 * for cfg80211.
+		 */
+		rndis_wlan_craft_connected_bss(usbdev, bssid, info);
+	}
+
+	if (priv->infra_mode == NDIS_80211_INFRA_INFRA) {
+		if (!roamed) {
+			cfg80211_connect_result(usbdev->net, bssid, req_ie,
+						req_ie_len, resp_ie,
+						resp_ie_len, 0, GFP_KERNEL);
+		} else {
+			struct cfg80211_roam_info roam_info = {
+				.links[0].channel =
+					get_current_channel(usbdev, NULL),
+				.links[0].bssid = bssid,
+				.req_ie = req_ie,
+				.req_ie_len = req_ie_len,
+				.resp_ie = resp_ie,
+				.resp_ie_len = resp_ie_len,
+			};
+
+			cfg80211_roamed(usbdev->net, &roam_info, GFP_KERNEL);
+		}
+	} else if (priv->infra_mode == NDIS_80211_INFRA_ADHOC)
+		cfg80211_ibss_joined(usbdev->net, bssid,
+				     get_current_channel(usbdev, NULL),
+				     GFP_KERNEL);
+
+	kfree(info);
+
+	priv->connected = true;
+	memcpy(priv->bssid, bssid, ETH_ALEN);
+
+	usbnet_resume_rx(usbdev);
+	netif_carrier_on(usbdev->net);
+}
+
+static void rndis_wlan_do_link_down_work(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+
+	if (priv->connected) {
+		priv->connected = false;
+		eth_zero_addr(priv->bssid);
+
+		deauthenticate(usbdev);
+
+		cfg80211_disconnected(usbdev->net, 0, NULL, 0, true, GFP_KERNEL);
+	}
+
+	netif_carrier_off(usbdev->net);
+}
+
+static void rndis_wlan_worker(struct work_struct *work)
+{
+	struct rndis_wlan_private *priv =
+		container_of(work, struct rndis_wlan_private, work);
+	struct usbnet *usbdev = priv->usbdev;
+
+	if (test_and_clear_bit(WORK_LINK_UP, &priv->work_pending))
+		rndis_wlan_do_link_up_work(usbdev);
+
+	if (test_and_clear_bit(WORK_LINK_DOWN, &priv->work_pending))
+		rndis_wlan_do_link_down_work(usbdev);
+
+	if (test_and_clear_bit(WORK_SET_MULTICAST_LIST, &priv->work_pending))
+		set_multicast_list(usbdev);
+}
+
+static void rndis_wlan_set_multicast_list(struct net_device *dev)
+{
+	struct usbnet *usbdev = netdev_priv(dev);
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+
+	if (test_bit(WORK_SET_MULTICAST_LIST, &priv->work_pending))
+		return;
+
+	set_bit(WORK_SET_MULTICAST_LIST, &priv->work_pending);
+	queue_work(priv->workqueue, &priv->work);
+}
+
+static void rndis_wlan_auth_indication(struct usbnet *usbdev,
+				struct ndis_80211_status_indication *indication,
+				int len)
+{
+	u8 *buf;
+	const char *type;
+	int flags, buflen, key_id;
+	bool pairwise_error, group_error;
+	struct ndis_80211_auth_request *auth_req;
+	enum nl80211_key_type key_type;
+
+	/* must have at least one array entry */
+	if (len < offsetof(struct ndis_80211_status_indication, u) +
+				sizeof(struct ndis_80211_auth_request)) {
+		netdev_info(usbdev->net, "authentication indication: too short message (%i)\n",
+			    len);
+		return;
+	}
+
+	buf = (void *)&indication->u.auth_request[0];
+	buflen = len - offsetof(struct ndis_80211_status_indication, u);
+
+	while (buflen >= sizeof(*auth_req)) {
+		auth_req = (void *)buf;
+		if (buflen < le32_to_cpu(auth_req->length))
+			return;
+		type = "unknown";
+		flags = le32_to_cpu(auth_req->flags);
+		pairwise_error = false;
+		group_error = false;
+
+		if (flags & 0x1)
+			type = "reauth request";
+		if (flags & 0x2)
+			type = "key update request";
+		if (flags & 0x6) {
+			pairwise_error = true;
+			type = "pairwise_error";
+		}
+		if (flags & 0xe) {
+			group_error = true;
+			type = "group_error";
+		}
+
+		netdev_info(usbdev->net, "authentication indication: %s (0x%08x)\n",
+			    type, le32_to_cpu(auth_req->flags));
+
+		if (pairwise_error) {
+			key_type = NL80211_KEYTYPE_PAIRWISE;
+			key_id = -1;
+
+			cfg80211_michael_mic_failure(usbdev->net,
+							auth_req->bssid,
+							key_type, key_id, NULL,
+							GFP_KERNEL);
+		}
+
+		if (group_error) {
+			key_type = NL80211_KEYTYPE_GROUP;
+			key_id = -1;
+
+			cfg80211_michael_mic_failure(usbdev->net,
+							auth_req->bssid,
+							key_type, key_id, NULL,
+							GFP_KERNEL);
+		}
+
+		buflen -= le32_to_cpu(auth_req->length);
+		buf += le32_to_cpu(auth_req->length);
+	}
+}
+
+static void rndis_wlan_pmkid_cand_list_indication(struct usbnet *usbdev,
+				struct ndis_80211_status_indication *indication,
+				int len)
+{
+	struct ndis_80211_pmkid_cand_list *cand_list;
+	int list_len, expected_len, i;
+
+	if (len < offsetof(struct ndis_80211_status_indication, u) +
+				sizeof(struct ndis_80211_pmkid_cand_list)) {
+		netdev_info(usbdev->net, "pmkid candidate list indication: too short message (%i)\n",
+			    len);
+		return;
+	}
+
+	list_len = le32_to_cpu(indication->u.cand_list.num_candidates) *
+			sizeof(struct ndis_80211_pmkid_candidate);
+	expected_len = sizeof(struct ndis_80211_pmkid_cand_list) + list_len +
+			offsetof(struct ndis_80211_status_indication, u);
+
+	if (len < expected_len) {
+		netdev_info(usbdev->net, "pmkid candidate list indication: list larger than buffer (%i < %i)\n",
+			    len, expected_len);
+		return;
+	}
+
+	cand_list = &indication->u.cand_list;
+
+	netdev_info(usbdev->net, "pmkid candidate list indication: version %i, candidates %i\n",
+		    le32_to_cpu(cand_list->version),
+		    le32_to_cpu(cand_list->num_candidates));
+
+	if (le32_to_cpu(cand_list->version) != 1)
+		return;
+
+	for (i = 0; i < le32_to_cpu(cand_list->num_candidates); i++) {
+		struct ndis_80211_pmkid_candidate *cand =
+						&cand_list->candidate_list[i];
+		bool preauth = !!(cand->flags & NDIS_80211_PMKID_CAND_PREAUTH);
+
+		netdev_dbg(usbdev->net, "cand[%i]: flags: 0x%08x, preauth: %d, bssid: %pM\n",
+			   i, le32_to_cpu(cand->flags), preauth, cand->bssid);
+
+		cfg80211_pmksa_candidate_notify(usbdev->net, i, cand->bssid,
+						preauth, GFP_ATOMIC);
+	}
+}
+
+static void rndis_wlan_media_specific_indication(struct usbnet *usbdev,
+			struct rndis_indicate *msg, int buflen)
+{
+	struct ndis_80211_status_indication *indication;
+	unsigned int len, offset;
+
+	offset = offsetof(struct rndis_indicate, status) +
+			le32_to_cpu(msg->offset);
+	len = le32_to_cpu(msg->length);
+
+	if (len < 8) {
+		netdev_info(usbdev->net, "media specific indication, ignore too short message (%i < 8)\n",
+			    len);
+		return;
+	}
+
+	if (len > buflen || offset > buflen || offset + len > buflen) {
+		netdev_info(usbdev->net, "media specific indication, too large to fit to buffer (%i > %i)\n",
+			    offset + len, buflen);
+		return;
+	}
+
+	indication = (void *)((u8 *)msg + offset);
+
+	switch (le32_to_cpu(indication->status_type)) {
+	case NDIS_80211_STATUSTYPE_RADIOSTATE:
+		netdev_info(usbdev->net, "radio state indication: %i\n",
+			    le32_to_cpu(indication->u.radio_status));
+		return;
+
+	case NDIS_80211_STATUSTYPE_MEDIASTREAMMODE:
+		netdev_info(usbdev->net, "media stream mode indication: %i\n",
+			    le32_to_cpu(indication->u.media_stream_mode));
+		return;
+
+	case NDIS_80211_STATUSTYPE_AUTHENTICATION:
+		rndis_wlan_auth_indication(usbdev, indication, len);
+		return;
+
+	case NDIS_80211_STATUSTYPE_PMKID_CANDIDATELIST:
+		rndis_wlan_pmkid_cand_list_indication(usbdev, indication, len);
+		return;
+
+	default:
+		netdev_info(usbdev->net, "media specific indication: unknown status type 0x%08x\n",
+			    le32_to_cpu(indication->status_type));
+	}
+}
+
+static void rndis_wlan_indication(struct usbnet *usbdev, void *ind, int buflen)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	struct rndis_indicate *msg = ind;
+
+	switch (le32_to_cpu(msg->status)) {
+	case RNDIS_STATUS_MEDIA_CONNECT:
+		if (priv->current_command_oid == RNDIS_OID_802_11_ADD_KEY) {
+			/* RNDIS_OID_802_11_ADD_KEY causes sometimes extra
+			 * "media connect" indications which confuses driver
+			 * and userspace to think that device is
+			 * roaming/reassociating when it isn't.
+			 */
+			netdev_dbg(usbdev->net, "ignored RNDIS_OID_802_11_ADD_KEY triggered 'media connect'\n");
+			return;
+		}
+
+		usbnet_pause_rx(usbdev);
+
+		netdev_info(usbdev->net, "media connect\n");
+
+		/* queue work to avoid recursive calls into rndis_command */
+		set_bit(WORK_LINK_UP, &priv->work_pending);
+		queue_work(priv->workqueue, &priv->work);
+		break;
+
+	case RNDIS_STATUS_MEDIA_DISCONNECT:
+		netdev_info(usbdev->net, "media disconnect\n");
+
+		/* queue work to avoid recursive calls into rndis_command */
+		set_bit(WORK_LINK_DOWN, &priv->work_pending);
+		queue_work(priv->workqueue, &priv->work);
+		break;
+
+	case RNDIS_STATUS_MEDIA_SPECIFIC_INDICATION:
+		rndis_wlan_media_specific_indication(usbdev, msg, buflen);
+		break;
+
+	default:
+		netdev_info(usbdev->net, "indication: 0x%08x\n",
+			    le32_to_cpu(msg->status));
+		break;
+	}
+}
+
+static int rndis_wlan_get_caps(struct usbnet *usbdev, struct wiphy *wiphy)
+{
+	struct {
+		__le32	num_items;
+		__le32	items[8];
+	} networks_supported;
+	struct ndis_80211_capability caps;
+	int len, retval, i, n;
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+
+	/* determine supported modes */
+	len = sizeof(networks_supported);
+	retval = rndis_query_oid(usbdev,
+				 RNDIS_OID_802_11_NETWORK_TYPES_SUPPORTED,
+				 &networks_supported, &len);
+	if (!retval) {
+		n = le32_to_cpu(networks_supported.num_items);
+		if (n > 8)
+			n = 8;
+		for (i = 0; i < n; i++) {
+			switch (le32_to_cpu(networks_supported.items[i])) {
+			case NDIS_80211_TYPE_FREQ_HOP:
+			case NDIS_80211_TYPE_DIRECT_SEQ:
+				priv->caps |= CAP_MODE_80211B;
+				break;
+			case NDIS_80211_TYPE_OFDM_A:
+				priv->caps |= CAP_MODE_80211A;
+				break;
+			case NDIS_80211_TYPE_OFDM_G:
+				priv->caps |= CAP_MODE_80211G;
+				break;
+			}
+		}
+	}
+
+	/* get device 802.11 capabilities, number of PMKIDs */
+	len = sizeof(caps);
+	retval = rndis_query_oid(usbdev,
+				 RNDIS_OID_802_11_CAPABILITY,
+				 &caps, &len);
+	if (!retval) {
+		netdev_dbg(usbdev->net, "RNDIS_OID_802_11_CAPABILITY -> len %d, "
+				"ver %d, pmkids %d, auth-encr-pairs %d\n",
+				le32_to_cpu(caps.length),
+				le32_to_cpu(caps.version),
+				le32_to_cpu(caps.num_pmkids),
+				le32_to_cpu(caps.num_auth_encr_pair));
+		wiphy->max_num_pmkids = le32_to_cpu(caps.num_pmkids);
+	} else
+		wiphy->max_num_pmkids = 0;
+
+	return retval;
+}
+
+static void rndis_do_cqm(struct usbnet *usbdev, s32 rssi)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	enum nl80211_cqm_rssi_threshold_event event;
+	int thold, hyst, last_event;
+
+	if (priv->cqm_rssi_thold >= 0 || rssi >= 0)
+		return;
+	if (priv->infra_mode != NDIS_80211_INFRA_INFRA)
+		return;
+
+	last_event = priv->last_cqm_event_rssi;
+	thold = priv->cqm_rssi_thold;
+	hyst = priv->cqm_rssi_hyst;
+
+	if (rssi < thold && (last_event == 0 || rssi < last_event - hyst))
+		event = NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW;
+	else if (rssi > thold && (last_event == 0 || rssi > last_event + hyst))
+		event = NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH;
+	else
+		return;
+
+	priv->last_cqm_event_rssi = rssi;
+	cfg80211_cqm_rssi_notify(usbdev->net, event, rssi, GFP_KERNEL);
+}
+
+#define DEVICE_POLLER_JIFFIES (HZ)
+static void rndis_device_poller(struct work_struct *work)
+{
+	struct rndis_wlan_private *priv =
+		container_of(work, struct rndis_wlan_private,
+							dev_poller_work.work);
+	struct usbnet *usbdev = priv->usbdev;
+	__le32 rssi, tmp;
+	int len, ret, j;
+	int update_jiffies = DEVICE_POLLER_JIFFIES;
+	void *buf;
+
+	/* Only check/do workaround when connected. Calling is_associated()
+	 * also polls device with rndis_command() and catches for media link
+	 * indications.
+	 */
+	if (!is_associated(usbdev)) {
+		/* Workaround bad scanning in BCM4320a devices with active
+		 * background scanning when not associated.
+		 */
+		if (priv->device_type == RNDIS_BCM4320A && priv->radio_on &&
+		    !priv->scan_request) {
+			/* Get previous scan results */
+			rndis_check_bssid_list(usbdev, NULL, NULL);
+
+			/* Initiate new scan */
+			rndis_start_bssid_list_scan(usbdev);
+		}
+
+		goto end;
+	}
+
+	len = sizeof(rssi);
+	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_RSSI,
+			      &rssi, &len);
+	if (ret == 0) {
+		priv->last_qual = level_to_qual(le32_to_cpu(rssi));
+		rndis_do_cqm(usbdev, le32_to_cpu(rssi));
+	}
+
+	netdev_dbg(usbdev->net, "dev-poller: RNDIS_OID_802_11_RSSI -> %d, rssi:%d, qual: %d\n",
+		   ret, le32_to_cpu(rssi), level_to_qual(le32_to_cpu(rssi)));
+
+	/* Workaround transfer stalls on poor quality links.
+	 * TODO: find right way to fix these stalls (as stalls do not happen
+	 * with ndiswrapper/windows driver). */
+	if (priv->param_workaround_interval > 0 && priv->last_qual <= 25) {
+		/* Decrease stats worker interval to catch stalls.
+		 * faster. Faster than 400-500ms causes packet loss,
+		 * Slower doesn't catch stalls fast enough.
+		 */
+		j = msecs_to_jiffies(priv->param_workaround_interval);
+		if (j > DEVICE_POLLER_JIFFIES)
+			j = DEVICE_POLLER_JIFFIES;
+		else if (j <= 0)
+			j = 1;
+		update_jiffies = j;
+
+		/* Send scan OID. Use of both OIDs is required to get device
+		 * working.
+		 */
+		tmp = cpu_to_le32(1);
+		rndis_set_oid(usbdev,
+			      RNDIS_OID_802_11_BSSID_LIST_SCAN,
+			      &tmp, sizeof(tmp));
+
+		len = CONTROL_BUFFER_SIZE;
+		buf = kmalloc(len, GFP_KERNEL);
+		if (!buf)
+			goto end;
+
+		rndis_query_oid(usbdev,
+				RNDIS_OID_802_11_BSSID_LIST,
+				buf, &len);
+		kfree(buf);
+	}
+
+end:
+	if (update_jiffies >= HZ)
+		update_jiffies = round_jiffies_relative(update_jiffies);
+	else {
+		j = round_jiffies_relative(update_jiffies);
+		if (abs(j - update_jiffies) <= 10)
+			update_jiffies = j;
+	}
+
+	queue_delayed_work(priv->workqueue, &priv->dev_poller_work,
+								update_jiffies);
+}
+
+/*
+ * driver/device initialization
+ */
+static void rndis_copy_module_params(struct usbnet *usbdev, int device_type)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+
+	priv->device_type = device_type;
+
+	priv->param_country[0] = modparam_country[0];
+	priv->param_country[1] = modparam_country[1];
+	priv->param_country[2] = 0;
+	priv->param_frameburst   = modparam_frameburst;
+	priv->param_afterburner  = modparam_afterburner;
+	priv->param_power_save   = modparam_power_save;
+	priv->param_power_output = modparam_power_output;
+	priv->param_roamtrigger  = modparam_roamtrigger;
+	priv->param_roamdelta    = modparam_roamdelta;
+
+	priv->param_country[0] = toupper(priv->param_country[0]);
+	priv->param_country[1] = toupper(priv->param_country[1]);
+	/* doesn't support EU as country code, use FI instead */
+	if (!strcmp(priv->param_country, "EU"))
+		strcpy(priv->param_country, "FI");
+
+	if (priv->param_power_save < 0)
+		priv->param_power_save = 0;
+	else if (priv->param_power_save > 2)
+		priv->param_power_save = 2;
+
+	if (priv->param_power_output < 0)
+		priv->param_power_output = 0;
+	else if (priv->param_power_output > 3)
+		priv->param_power_output = 3;
+
+	if (priv->param_roamtrigger < -80)
+		priv->param_roamtrigger = -80;
+	else if (priv->param_roamtrigger > -60)
+		priv->param_roamtrigger = -60;
+
+	if (priv->param_roamdelta < 0)
+		priv->param_roamdelta = 0;
+	else if (priv->param_roamdelta > 2)
+		priv->param_roamdelta = 2;
+
+	if (modparam_workaround_interval < 0)
+		priv->param_workaround_interval = 500;
+	else
+		priv->param_workaround_interval = modparam_workaround_interval;
+}
+
+static int unknown_early_init(struct usbnet *usbdev)
+{
+	/* copy module parameters for unknown so that iwconfig reports txpower
+	 * and workaround parameter is copied to private structure correctly.
+	 */
+	rndis_copy_module_params(usbdev, RNDIS_UNKNOWN);
+
+	/* This is unknown device, so do not try set configuration parameters.
+	 */
+
+	return 0;
+}
+
+static int bcm4320a_early_init(struct usbnet *usbdev)
+{
+	/* copy module parameters for bcm4320a so that iwconfig reports txpower
+	 * and workaround parameter is copied to private structure correctly.
+	 */
+	rndis_copy_module_params(usbdev, RNDIS_BCM4320A);
+
+	/* bcm4320a doesn't handle configuration parameters well. Try
+	 * set any and you get partially zeroed mac and broken device.
+	 */
+
+	return 0;
+}
+
+static int bcm4320b_early_init(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	char buf[8];
+
+	rndis_copy_module_params(usbdev, RNDIS_BCM4320B);
+
+	/* Early initialization settings, setting these won't have effect
+	 * if called after generic_rndis_bind().
+	 */
+
+	rndis_set_config_parameter_str(usbdev, "Country", priv->param_country);
+	rndis_set_config_parameter_str(usbdev, "FrameBursting",
+					priv->param_frameburst ? "1" : "0");
+	rndis_set_config_parameter_str(usbdev, "Afterburner",
+					priv->param_afterburner ? "1" : "0");
+	sprintf(buf, "%d", priv->param_power_save);
+	rndis_set_config_parameter_str(usbdev, "PowerSaveMode", buf);
+	sprintf(buf, "%d", priv->param_power_output);
+	rndis_set_config_parameter_str(usbdev, "PwrOut", buf);
+	sprintf(buf, "%d", priv->param_roamtrigger);
+	rndis_set_config_parameter_str(usbdev, "RoamTrigger", buf);
+	sprintf(buf, "%d", priv->param_roamdelta);
+	rndis_set_config_parameter_str(usbdev, "RoamDelta", buf);
+
+	return 0;
+}
+
+/* same as rndis_netdev_ops but with local multicast handler */
+static const struct net_device_ops rndis_wlan_netdev_ops = {
+	.ndo_open		= usbnet_open,
+	.ndo_stop		= usbnet_stop,
+	.ndo_start_xmit		= usbnet_start_xmit,
+	.ndo_tx_timeout		= usbnet_tx_timeout,
+	.ndo_get_stats64	= dev_get_tstats64,
+	.ndo_set_mac_address 	= eth_mac_addr,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_set_rx_mode	= rndis_wlan_set_multicast_list,
+};
+
+static int rndis_wlan_bind(struct usbnet *usbdev, struct usb_interface *intf)
+{
+	struct wiphy *wiphy;
+	struct rndis_wlan_private *priv;
+	int retval, len;
+	__le32 tmp;
+
+	/* allocate wiphy and rndis private data
+	 * NOTE: We only support a single virtual interface, so wiphy
+	 * and wireless_dev are somewhat synonymous for this device.
+	 */
+	wiphy = wiphy_new(&rndis_config_ops, sizeof(struct rndis_wlan_private));
+	if (!wiphy)
+		return -ENOMEM;
+
+	priv = wiphy_priv(wiphy);
+	usbdev->net->ieee80211_ptr = &priv->wdev;
+	priv->wdev.wiphy = wiphy;
+	priv->wdev.iftype = NL80211_IFTYPE_STATION;
+
+	/* These have to be initialized before calling generic_rndis_bind().
+	 * Otherwise we'll be in big trouble in rndis_wlan_early_init().
+	 */
+	usbdev->driver_priv = priv;
+	priv->usbdev = usbdev;
+
+	mutex_init(&priv->command_lock);
+
+	/* because rndis_command() sleeps we need to use workqueue */
+	priv->workqueue = create_singlethread_workqueue("rndis_wlan");
+	if (!priv->workqueue) {
+		wiphy_free(wiphy);
+		return -ENOMEM;
+	}
+	INIT_WORK(&priv->work, rndis_wlan_worker);
+	INIT_DELAYED_WORK(&priv->dev_poller_work, rndis_device_poller);
+	INIT_DELAYED_WORK(&priv->scan_work, rndis_get_scan_results);
+
+	/* try bind rndis_host */
+	retval = generic_rndis_bind(usbdev, intf, FLAG_RNDIS_PHYM_WIRELESS);
+	if (retval < 0)
+		goto fail;
+
+	/* generic_rndis_bind set packet filter to multicast_all+
+	 * promisc mode which doesn't work well for our devices (device
+	 * picks up rssi to closest station instead of to access point).
+	 *
+	 * rndis_host wants to avoid all OID as much as possible
+	 * so do promisc/multicast handling in rndis_wlan.
+	 */
+	usbdev->net->netdev_ops = &rndis_wlan_netdev_ops;
+
+	tmp = cpu_to_le32(RNDIS_PACKET_TYPE_DIRECTED | RNDIS_PACKET_TYPE_BROADCAST);
+	retval = rndis_set_oid(usbdev,
+			       RNDIS_OID_GEN_CURRENT_PACKET_FILTER,
+			       &tmp, sizeof(tmp));
+
+	len = sizeof(tmp);
+	retval = rndis_query_oid(usbdev,
+				 RNDIS_OID_802_3_MAXIMUM_LIST_SIZE,
+				 &tmp, &len);
+	priv->multicast_size = le32_to_cpu(tmp);
+	if (retval < 0 || priv->multicast_size < 0)
+		priv->multicast_size = 0;
+	if (priv->multicast_size > 0)
+		usbdev->net->flags |= IFF_MULTICAST;
+	else
+		usbdev->net->flags &= ~IFF_MULTICAST;
+
+	/* fill-out wiphy structure and register w/ cfg80211 */
+	memcpy(wiphy->perm_addr, usbdev->net->dev_addr, ETH_ALEN);
+	wiphy->privid = rndis_wiphy_privid;
+	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION)
+					| BIT(NL80211_IFTYPE_ADHOC);
+	wiphy->max_scan_ssids = 1;
+
+	/* TODO: fill-out band/encr information based on priv->caps */
+	rndis_wlan_get_caps(usbdev, wiphy);
+
+	memcpy(priv->channels, rndis_channels, sizeof(rndis_channels));
+	memcpy(priv->rates, rndis_rates, sizeof(rndis_rates));
+	priv->band.channels = priv->channels;
+	priv->band.n_channels = ARRAY_SIZE(rndis_channels);
+	priv->band.bitrates = priv->rates;
+	priv->band.n_bitrates = ARRAY_SIZE(rndis_rates);
+	wiphy->bands[NL80211_BAND_2GHZ] = &priv->band;
+	wiphy->signal_type = CFG80211_SIGNAL_TYPE_UNSPEC;
+
+	memcpy(priv->cipher_suites, rndis_cipher_suites,
+						sizeof(rndis_cipher_suites));
+	wiphy->cipher_suites = priv->cipher_suites;
+	wiphy->n_cipher_suites = ARRAY_SIZE(rndis_cipher_suites);
+
+	set_wiphy_dev(wiphy, &usbdev->udev->dev);
+
+	if (wiphy_register(wiphy)) {
+		retval = -ENODEV;
+		goto fail;
+	}
+
+	set_default_iw_params(usbdev);
+
+	priv->power_mode = -1;
+
+	/* set default rts/frag */
+	rndis_set_wiphy_params(wiphy,
+			WIPHY_PARAM_FRAG_THRESHOLD | WIPHY_PARAM_RTS_THRESHOLD);
+
+	/* turn radio off on init */
+	priv->radio_on = false;
+	disassociate(usbdev, false);
+	netif_carrier_off(usbdev->net);
+
+	return 0;
+
+fail:
+	cancel_delayed_work_sync(&priv->dev_poller_work);
+	cancel_delayed_work_sync(&priv->scan_work);
+	cancel_work_sync(&priv->work);
+	destroy_workqueue(priv->workqueue);
+
+	wiphy_free(wiphy);
+	return retval;
+}
+
+static void rndis_wlan_unbind(struct usbnet *usbdev, struct usb_interface *intf)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+
+	/* turn radio off */
+	disassociate(usbdev, false);
+
+	cancel_delayed_work_sync(&priv->dev_poller_work);
+	cancel_delayed_work_sync(&priv->scan_work);
+	cancel_work_sync(&priv->work);
+	destroy_workqueue(priv->workqueue);
+
+	rndis_unbind(usbdev, intf);
+
+	wiphy_unregister(priv->wdev.wiphy);
+	wiphy_free(priv->wdev.wiphy);
+}
+
+static int rndis_wlan_reset(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	int retval;
+
+	netdev_dbg(usbdev->net, "%s()\n", __func__);
+
+	retval = rndis_reset(usbdev);
+	if (retval)
+		netdev_warn(usbdev->net, "rndis_reset failed: %d\n", retval);
+
+	/* rndis_reset cleared multicast list, so restore here.
+	   (set_multicast_list() also turns on current packet filter) */
+	set_multicast_list(usbdev);
+
+	queue_delayed_work(priv->workqueue, &priv->dev_poller_work,
+		round_jiffies_relative(DEVICE_POLLER_JIFFIES));
+
+	return deauthenticate(usbdev);
+}
+
+static int rndis_wlan_stop(struct usbnet *usbdev)
+{
+	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
+	int retval;
+	__le32 filter;
+
+	netdev_dbg(usbdev->net, "%s()\n", __func__);
+
+	retval = disassociate(usbdev, false);
+
+	priv->work_pending = 0;
+	cancel_delayed_work_sync(&priv->dev_poller_work);
+	cancel_delayed_work_sync(&priv->scan_work);
+	cancel_work_sync(&priv->work);
+	flush_workqueue(priv->workqueue);
+
+	if (priv->scan_request) {
+		struct cfg80211_scan_info info = {
+			.aborted = true,
+		};
+
+		cfg80211_scan_done(priv->scan_request, &info);
+		priv->scan_request = NULL;
+	}
+
+	/* Set current packet filter zero to block receiving data packets from
+	   device. */
+	filter = 0;
+	rndis_set_oid(usbdev, RNDIS_OID_GEN_CURRENT_PACKET_FILTER, &filter,
+								sizeof(filter));
+
+	return retval;
+}
+
+static const struct driver_info	bcm4320b_info = {
+	.description =	"Wireless RNDIS device, BCM4320b based",
+	.flags =	FLAG_WLAN | FLAG_FRAMING_RN | FLAG_NO_SETINT |
+				FLAG_AVOID_UNLINK_URBS,
+	.bind =		rndis_wlan_bind,
+	.unbind =	rndis_wlan_unbind,
+	.status =	rndis_status,
+	.rx_fixup =	rndis_rx_fixup,
+	.tx_fixup =	rndis_tx_fixup,
+	.reset =	rndis_wlan_reset,
+	.stop =		rndis_wlan_stop,
+	.early_init =	bcm4320b_early_init,
+	.indication =	rndis_wlan_indication,
+};
+
+static const struct driver_info	bcm4320a_info = {
+	.description =	"Wireless RNDIS device, BCM4320a based",
+	.flags =	FLAG_WLAN | FLAG_FRAMING_RN | FLAG_NO_SETINT |
+				FLAG_AVOID_UNLINK_URBS,
+	.bind =		rndis_wlan_bind,
+	.unbind =	rndis_wlan_unbind,
+	.status =	rndis_status,
+	.rx_fixup =	rndis_rx_fixup,
+	.tx_fixup =	rndis_tx_fixup,
+	.reset =	rndis_wlan_reset,
+	.stop =		rndis_wlan_stop,
+	.early_init =	bcm4320a_early_init,
+	.indication =	rndis_wlan_indication,
+};
+
+static const struct driver_info rndis_wlan_info = {
+	.description =	"Wireless RNDIS device",
+	.flags =	FLAG_WLAN | FLAG_FRAMING_RN | FLAG_NO_SETINT |
+				FLAG_AVOID_UNLINK_URBS,
+	.bind =		rndis_wlan_bind,
+	.unbind =	rndis_wlan_unbind,
+	.status =	rndis_status,
+	.rx_fixup =	rndis_rx_fixup,
+	.tx_fixup =	rndis_tx_fixup,
+	.reset =	rndis_wlan_reset,
+	.stop =		rndis_wlan_stop,
+	.early_init =	unknown_early_init,
+	.indication =	rndis_wlan_indication,
+};
+
+/*-------------------------------------------------------------------------*/
+
+static const struct usb_device_id products [] = {
+#define	RNDIS_MASTER_INTERFACE \
+	.bInterfaceClass	= USB_CLASS_COMM, \
+	.bInterfaceSubClass	= 2 /* ACM */, \
+	.bInterfaceProtocol	= 0x0ff
+
+/* INF driver for these devices have DriverVer >= 4.xx.xx.xx and many custom
+ * parameters available. Chipset marked as 'BCM4320SKFBG' in NDISwrapper-wiki.
+ */
+{
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x0411,
+	.idProduct		= 0x00bc,	/* Buffalo WLI-U2-KG125S */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x0baf,
+	.idProduct		= 0x011b,	/* U.S. Robotics USR5421 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x050d,
+	.idProduct		= 0x011b,	/* Belkin F5D7051 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x1799,	/* Belkin has two vendor ids */
+	.idProduct		= 0x011b,	/* Belkin F5D7051 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x13b1,
+	.idProduct		= 0x0014,	/* Linksys WUSB54GSv2 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x13b1,
+	.idProduct		= 0x0026,	/* Linksys WUSB54GSC */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x0b05,
+	.idProduct		= 0x1717,	/* Asus WL169gE */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x0a5c,
+	.idProduct		= 0xd11b,	/* Eminent EM4045 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x1690,
+	.idProduct		= 0x0715,	/* BT Voyager 1055 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320b_info,
+},
+/* These devices have DriverVer < 4.xx.xx.xx and do not have any custom
+ * parameters available, hardware probably contain older firmware version with
+ * no way of updating. Chipset marked as 'BCM4320????' in NDISwrapper-wiki.
+ */
+{
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x13b1,
+	.idProduct		= 0x000e,	/* Linksys WUSB54GSv1 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320a_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x0baf,
+	.idProduct		= 0x0111,	/* U.S. Robotics USR5420 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320a_info,
+}, {
+	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
+			  | USB_DEVICE_ID_MATCH_DEVICE,
+	.idVendor		= 0x0411,
+	.idProduct		= 0x004b,	/* BUFFALO WLI-USB-G54 */
+	RNDIS_MASTER_INTERFACE,
+	.driver_info		= (unsigned long) &bcm4320a_info,
+},
+/* Generic Wireless RNDIS devices that we don't have exact
+ * idVendor/idProduct/chip yet.
+ */
+{
+	/* RNDIS is MSFT's un-official variant of CDC ACM */
+	USB_INTERFACE_INFO(USB_CLASS_COMM, 2 /* ACM */, 0x0ff),
+	.driver_info = (unsigned long) &rndis_wlan_info,
+}, {
+	/* "ActiveSync" is an undocumented variant of RNDIS, used in WM5 */
+	USB_INTERFACE_INFO(USB_CLASS_MISC, 1, 1),
+	.driver_info = (unsigned long) &rndis_wlan_info,
+},
+	{ },		// END
+};
+MODULE_DEVICE_TABLE(usb, products);
+
+static struct usb_driver rndis_wlan_driver = {
+	.name =		"rndis_wlan",
+	.id_table =	products,
+	.probe =	usbnet_probe,
+	.disconnect =	usbnet_disconnect,
+	.suspend =	usbnet_suspend,
+	.resume =	usbnet_resume,
+	.disable_hub_initiated_lpm = 1,
+};
+
+module_usb_driver(rndis_wlan_driver);
+
+MODULE_AUTHOR("Bjorge Dijkstra");
+MODULE_AUTHOR("Jussi Kivilinna");
+MODULE_DESCRIPTION("Driver for RNDIS based USB Wireless adapters");
+MODULE_LICENSE("GPL");
+
diff --git a/drivers/net/wireless/legacy/wl3501.h b/drivers/net/wireless/legacy/wl3501.h
new file mode 100644
index 000000000000..91f276dd22a1
--- /dev/null
+++ b/drivers/net/wireless/legacy/wl3501.h
@@ -0,0 +1,615 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __WL3501_H__
+#define __WL3501_H__
+
+#include <linux/spinlock.h>
+#include <linux/ieee80211.h>
+
+/* define for WLA 2.0 */
+#define WL3501_BLKSZ 256
+/*
+ * ID for input Signals of DRIVER block
+ * bit[7-5] is block ID: 000
+ * bit[4-0] is signal ID
+*/
+enum wl3501_signals {
+	WL3501_SIG_ALARM,
+	WL3501_SIG_MD_CONFIRM,
+	WL3501_SIG_MD_IND,
+	WL3501_SIG_ASSOC_CONFIRM,
+	WL3501_SIG_ASSOC_IND,
+	WL3501_SIG_AUTH_CONFIRM,
+	WL3501_SIG_AUTH_IND,
+	WL3501_SIG_DEAUTH_CONFIRM,
+	WL3501_SIG_DEAUTH_IND,
+	WL3501_SIG_DISASSOC_CONFIRM,
+	WL3501_SIG_DISASSOC_IND,
+	WL3501_SIG_GET_CONFIRM,
+	WL3501_SIG_JOIN_CONFIRM,
+	WL3501_SIG_PWR_MGMT_CONFIRM,
+	WL3501_SIG_REASSOC_CONFIRM,
+	WL3501_SIG_REASSOC_IND,
+	WL3501_SIG_SCAN_CONFIRM,
+	WL3501_SIG_SET_CONFIRM,
+	WL3501_SIG_START_CONFIRM,
+	WL3501_SIG_RESYNC_CONFIRM,
+	WL3501_SIG_SITE_CONFIRM,
+	WL3501_SIG_SAVE_CONFIRM,
+	WL3501_SIG_RFTEST_CONFIRM,
+/*
+ * ID for input Signals of MLME block
+ * bit[7-5] is block ID: 010
+ * bit[4-0] is signal ID
+ */
+	WL3501_SIG_ASSOC_REQ = 0x20,
+	WL3501_SIG_AUTH_REQ,
+	WL3501_SIG_DEAUTH_REQ,
+	WL3501_SIG_DISASSOC_REQ,
+	WL3501_SIG_GET_REQ,
+	WL3501_SIG_JOIN_REQ,
+	WL3501_SIG_PWR_MGMT_REQ,
+	WL3501_SIG_REASSOC_REQ,
+	WL3501_SIG_SCAN_REQ,
+	WL3501_SIG_SET_REQ,
+	WL3501_SIG_START_REQ,
+	WL3501_SIG_MD_REQ,
+	WL3501_SIG_RESYNC_REQ,
+	WL3501_SIG_SITE_REQ,
+	WL3501_SIG_SAVE_REQ,
+	WL3501_SIG_RF_TEST_REQ,
+	WL3501_SIG_MM_CONFIRM = 0x60,
+	WL3501_SIG_MM_IND,
+};
+
+enum wl3501_mib_attribs {
+	WL3501_MIB_ATTR_STATION_ID,
+	WL3501_MIB_ATTR_AUTH_ALGORITHMS,
+	WL3501_MIB_ATTR_AUTH_TYPE,
+	WL3501_MIB_ATTR_MEDIUM_OCCUPANCY_LIMIT,
+	WL3501_MIB_ATTR_CF_POLLABLE,
+	WL3501_MIB_ATTR_CFP_PERIOD,
+	WL3501_MIB_ATTR_CFPMAX_DURATION,
+	WL3501_MIB_ATTR_AUTH_RESP_TMOUT,
+	WL3501_MIB_ATTR_RX_DTIMS,
+	WL3501_MIB_ATTR_PRIV_OPT_IMPLEMENTED,
+	WL3501_MIB_ATTR_PRIV_INVOKED,
+	WL3501_MIB_ATTR_WEP_DEFAULT_KEYS,
+	WL3501_MIB_ATTR_WEP_DEFAULT_KEY_ID,
+	WL3501_MIB_ATTR_WEP_KEY_MAPPINGS,
+	WL3501_MIB_ATTR_WEP_KEY_MAPPINGS_LEN,
+	WL3501_MIB_ATTR_EXCLUDE_UNENCRYPTED,
+	WL3501_MIB_ATTR_WEP_ICV_ERROR_COUNT,
+	WL3501_MIB_ATTR_WEP_UNDECRYPTABLE_COUNT,
+	WL3501_MIB_ATTR_WEP_EXCLUDED_COUNT,
+	WL3501_MIB_ATTR_MAC_ADDR,
+	WL3501_MIB_ATTR_GROUP_ADDRS,
+	WL3501_MIB_ATTR_RTS_THRESHOLD,
+	WL3501_MIB_ATTR_SHORT_RETRY_LIMIT,
+	WL3501_MIB_ATTR_LONG_RETRY_LIMIT,
+	WL3501_MIB_ATTR_FRAG_THRESHOLD,
+	WL3501_MIB_ATTR_MAX_TX_MSDU_LIFETIME,
+	WL3501_MIB_ATTR_MAX_RX_LIFETIME,
+	WL3501_MIB_ATTR_MANUFACTURER_ID,
+	WL3501_MIB_ATTR_PRODUCT_ID,
+	WL3501_MIB_ATTR_TX_FRAG_COUNT,
+	WL3501_MIB_ATTR_MULTICAST_TX_FRAME_COUNT,
+	WL3501_MIB_ATTR_FAILED_COUNT,
+	WL3501_MIB_ATTR_RX_FRAG_COUNT,
+	WL3501_MIB_ATTR_MULTICAST_RX_COUNT,
+	WL3501_MIB_ATTR_FCS_ERROR_COUNT,
+	WL3501_MIB_ATTR_RETRY_COUNT,
+	WL3501_MIB_ATTR_MULTIPLE_RETRY_COUNT,
+	WL3501_MIB_ATTR_RTS_SUCCESS_COUNT,
+	WL3501_MIB_ATTR_RTS_FAILURE_COUNT,
+	WL3501_MIB_ATTR_ACK_FAILURE_COUNT,
+	WL3501_MIB_ATTR_FRAME_DUPLICATE_COUNT,
+	WL3501_MIB_ATTR_PHY_TYPE,
+	WL3501_MIB_ATTR_REG_DOMAINS_SUPPORT,
+	WL3501_MIB_ATTR_CURRENT_REG_DOMAIN,
+	WL3501_MIB_ATTR_SLOT_TIME,
+	WL3501_MIB_ATTR_CCA_TIME,
+	WL3501_MIB_ATTR_RX_TX_TURNAROUND_TIME,
+	WL3501_MIB_ATTR_TX_PLCP_DELAY,
+	WL3501_MIB_ATTR_RX_TX_SWITCH_TIME,
+	WL3501_MIB_ATTR_TX_RAMP_ON_TIME,
+	WL3501_MIB_ATTR_TX_RF_DELAY,
+	WL3501_MIB_ATTR_SIFS_TIME,
+	WL3501_MIB_ATTR_RX_RF_DELAY,
+	WL3501_MIB_ATTR_RX_PLCP_DELAY,
+	WL3501_MIB_ATTR_MAC_PROCESSING_DELAY,
+	WL3501_MIB_ATTR_TX_RAMP_OFF_TIME,
+	WL3501_MIB_ATTR_PREAMBLE_LEN,
+	WL3501_MIB_ATTR_PLCP_HEADER_LEN,
+	WL3501_MIB_ATTR_MPDU_DURATION_FACTOR,
+	WL3501_MIB_ATTR_AIR_PROPAGATION_TIME,
+	WL3501_MIB_ATTR_TEMP_TYPE,
+	WL3501_MIB_ATTR_CW_MIN,
+	WL3501_MIB_ATTR_CW_MAX,
+	WL3501_MIB_ATTR_SUPPORT_DATA_RATES_TX,
+	WL3501_MIB_ATTR_SUPPORT_DATA_RATES_RX,
+	WL3501_MIB_ATTR_MPDU_MAX_LEN,
+	WL3501_MIB_ATTR_SUPPORT_TX_ANTENNAS,
+	WL3501_MIB_ATTR_CURRENT_TX_ANTENNA,
+	WL3501_MIB_ATTR_SUPPORT_RX_ANTENNAS,
+	WL3501_MIB_ATTR_DIVERSITY_SUPPORT,
+	WL3501_MIB_ATTR_DIVERSITY_SELECTION_RS,
+	WL3501_MIB_ATTR_NR_SUPPORTED_PWR_LEVELS,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL1,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL2,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL3,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL4,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL5,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL6,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL7,
+	WL3501_MIB_ATTR_TX_PWR_LEVEL8,
+	WL3501_MIB_ATTR_CURRENT_TX_PWR_LEVEL,
+	WL3501_MIB_ATTR_CURRENT_CHAN,
+	WL3501_MIB_ATTR_CCA_MODE_SUPPORTED,
+	WL3501_MIB_ATTR_CURRENT_CCA_MODE,
+	WL3501_MIB_ATTR_ED_THRESHOLD,
+	WL3501_MIB_ATTR_SINTHESIZER_LOCKED,
+	WL3501_MIB_ATTR_CURRENT_PWR_STATE,
+	WL3501_MIB_ATTR_DOZE_TURNON_TIME,
+	WL3501_MIB_ATTR_RCR33,
+	WL3501_MIB_ATTR_DEFAULT_CHAN,
+	WL3501_MIB_ATTR_SSID,
+	WL3501_MIB_ATTR_PWR_MGMT_ENABLE,
+	WL3501_MIB_ATTR_NET_CAPABILITY,
+	WL3501_MIB_ATTR_ROUTING,
+};
+
+enum wl3501_net_type {
+	WL3501_NET_TYPE_INFRA,
+	WL3501_NET_TYPE_ADHOC,
+	WL3501_NET_TYPE_ANY_BSS,
+};
+
+enum wl3501_scan_type {
+	WL3501_SCAN_TYPE_ACTIVE,
+	WL3501_SCAN_TYPE_PASSIVE,
+};
+
+enum wl3501_tx_result {
+	WL3501_TX_RESULT_SUCCESS,
+	WL3501_TX_RESULT_NO_BSS,
+	WL3501_TX_RESULT_RETRY_LIMIT,
+};
+
+enum wl3501_sys_type {
+	WL3501_SYS_TYPE_OPEN,
+	WL3501_SYS_TYPE_SHARE_KEY,
+};
+
+enum wl3501_status {
+	WL3501_STATUS_SUCCESS,
+	WL3501_STATUS_INVALID,
+	WL3501_STATUS_TIMEOUT,
+	WL3501_STATUS_REFUSED,
+	WL3501_STATUS_MANY_REQ,
+	WL3501_STATUS_ALREADY_BSS,
+};
+
+#define WL3501_MGMT_CAPABILITY_ESS		0x0001  /* see 802.11 p.58 */
+#define WL3501_MGMT_CAPABILITY_IBSS		0x0002  /*      - " -	   */
+#define WL3501_MGMT_CAPABILITY_CF_POLLABLE	0x0004  /*      - " -	   */
+#define WL3501_MGMT_CAPABILITY_CF_POLL_REQUEST	0x0008  /*      - " -	   */
+#define WL3501_MGMT_CAPABILITY_PRIVACY		0x0010  /*      - " -	   */
+
+#define IW_REG_DOMAIN_FCC	0x10	/* Channel 1 to 11	USA    */
+#define IW_REG_DOMAIN_DOC	0x20	/* Channel 1 to 11	Canada */
+#define IW_REG_DOMAIN_ETSI	0x30	/* Channel 1 to 13	Europe */
+#define IW_REG_DOMAIN_SPAIN	0x31	/* Channel 10 to 11	Spain  */
+#define IW_REG_DOMAIN_FRANCE	0x32	/* Channel 10 to 13	France */
+#define IW_REG_DOMAIN_MKK	0x40	/* Channel 14		Japan  */
+#define IW_REG_DOMAIN_MKK1	0x41	/* Channel 1-14		Japan  */
+#define IW_REG_DOMAIN_ISRAEL	0x50	/* Channel 3 - 9	Israel */
+
+#define IW_MGMT_RATE_LABEL_MANDATORY 128 /* MSB */
+
+enum iw_mgmt_rate_labels {
+	IW_MGMT_RATE_LABEL_1MBIT   = 2,
+	IW_MGMT_RATE_LABEL_2MBIT   = 4,
+	IW_MGMT_RATE_LABEL_5_5MBIT = 11,
+	IW_MGMT_RATE_LABEL_11MBIT  = 22,
+};
+
+enum iw_mgmt_info_element_ids {
+	IW_MGMT_INFO_ELEMENT_SSID,		  /* Service Set Identity */
+	IW_MGMT_INFO_ELEMENT_SUPPORTED_RATES,
+	IW_MGMT_INFO_ELEMENT_FH_PARAMETER_SET,
+	IW_MGMT_INFO_ELEMENT_DS_PARAMETER_SET,
+	IW_MGMT_INFO_ELEMENT_CS_PARAMETER_SET,
+	IW_MGMT_INFO_ELEMENT_CS_TIM,		  /* Traffic Information Map */
+	IW_MGMT_INFO_ELEMENT_IBSS_PARAMETER_SET,
+	/* 7-15: Reserved, unused */
+	IW_MGMT_INFO_ELEMENT_CHALLENGE_TEXT = 16,
+	/* 17-31 Reserved for challenge text extension */
+	/* 32-255 Reserved, unused */
+};
+
+struct iw_mgmt_info_element {
+	u8 id; /* one of enum iw_mgmt_info_element_ids,
+		  but sizeof(enum) > sizeof(u8) :-( */
+	u8 len;
+	u8 data[];
+} __packed;
+
+struct iw_mgmt_essid_pset {
+	struct iw_mgmt_info_element el;
+	u8 			    essid[IW_ESSID_MAX_SIZE];
+} __packed;
+
+/*
+ * According to 802.11 Wireless Networks, the definitive guide - O'Reilly
+ * Pg 75
+ */ 
+#define IW_DATA_RATE_MAX_LABELS 8
+
+struct iw_mgmt_data_rset {
+	struct iw_mgmt_info_element el;
+	u8 			    data_rate_labels[IW_DATA_RATE_MAX_LABELS];
+} __packed;
+
+struct iw_mgmt_ds_pset {
+	struct iw_mgmt_info_element el;
+	u8 			    chan;
+} __packed;
+
+struct iw_mgmt_cf_pset {
+	struct iw_mgmt_info_element el;
+	u8 			    cfp_count;
+	u8 			    cfp_period;
+	u16 			    cfp_max_duration;
+	u16 			    cfp_dur_remaining;
+} __packed;
+
+struct iw_mgmt_ibss_pset {
+	struct iw_mgmt_info_element el;
+	u16 			    atim_window;
+} __packed;
+
+struct wl3501_tx_hdr {
+	u16	tx_cnt;
+	u8	sync[16];
+	u16	sfd;
+	u8	signal;
+	u8	service;
+	u16	len;
+	u16	crc16;
+	u16	frame_ctrl;
+	u16	duration_id;
+	u8	addr1[ETH_ALEN];
+	u8	addr2[ETH_ALEN];
+	u8	addr3[ETH_ALEN];
+	u16	seq_ctrl;
+	u8	addr4[ETH_ALEN];
+};
+
+struct wl3501_rx_hdr {
+	u16	rx_next_blk;
+	u16	rc_next_frame_blk;
+	u8	rx_blk_ctrl;
+	u8	rx_next_frame;
+	u8	rx_next_frame1;
+	u8	rssi;
+	char	time[8];
+	u8	signal;
+	u8	service;
+	u16	len;
+	u16	crc16;
+	u16	frame_ctrl;
+	u16	duration;
+	u8	addr1[ETH_ALEN];
+	u8	addr2[ETH_ALEN];
+	u8	addr3[ETH_ALEN];
+	u16	seq;
+	u8	addr4[ETH_ALEN];
+};
+
+struct wl3501_start_req {
+	u16			    next_blk;
+	u8			    sig_id;
+	u8			    bss_type;
+	u16			    beacon_period;
+	u16			    dtim_period;
+	u16			    probe_delay;
+	u16			    cap_info;
+	struct iw_mgmt_essid_pset   ssid;
+	struct iw_mgmt_data_rset    bss_basic_rset;
+	struct iw_mgmt_data_rset    operational_rset;
+	struct iw_mgmt_cf_pset	    cf_pset;
+	struct iw_mgmt_ds_pset	    ds_pset;
+	struct iw_mgmt_ibss_pset    ibss_pset;
+};
+
+struct wl3501_assoc_req {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	timeout;
+	u16	cap_info;
+	u16	listen_interval;
+	u8	mac_addr[ETH_ALEN];
+};
+
+struct wl3501_assoc_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	status;
+};
+
+struct wl3501_assoc_ind {
+	u16	next_blk;
+	u8	sig_id;
+	u8	mac_addr[ETH_ALEN];
+};
+
+struct wl3501_auth_req {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	type;
+	u16	timeout;
+	u8	mac_addr[ETH_ALEN];
+};
+
+struct wl3501_auth_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	type;
+	u16	status;
+	u8	mac_addr[ETH_ALEN];
+};
+
+struct wl3501_get_req {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	mib_attrib;
+};
+
+struct wl3501_get_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	mib_status;
+	u16	mib_attrib;
+	u8	mib_value[100];
+};
+
+struct wl3501_req {
+	u16			    beacon_period;
+	u16			    dtim_period;
+	u16			    cap_info;
+	u8			    bss_type;
+	u8			    bssid[ETH_ALEN];
+	struct iw_mgmt_essid_pset   ssid;
+	struct iw_mgmt_ds_pset	    ds_pset;
+	struct iw_mgmt_cf_pset	    cf_pset;
+	struct iw_mgmt_ibss_pset    ibss_pset;
+	struct iw_mgmt_data_rset    bss_basic_rset;
+};
+
+struct wl3501_join_req {
+	u16			    next_blk;
+	u8			    sig_id;
+	u8			    reserved;
+	struct iw_mgmt_data_rset    operational_rset;
+	u16			    reserved2;
+	u16			    timeout;
+	u16			    probe_delay;
+	u8			    timestamp[8];
+	u8			    local_time[8];
+	struct wl3501_req	    req;
+};
+
+struct wl3501_join_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	status;
+};
+
+struct wl3501_pwr_mgmt_req {
+	u16	next_blk;
+	u8	sig_id;
+	u8	pwr_save;
+	u8	wake_up;
+	u8	receive_dtims;
+};
+
+struct wl3501_pwr_mgmt_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	status;
+};
+
+struct wl3501_scan_req {
+	u16			    next_blk;
+	u8			    sig_id;
+	u8			    bss_type;
+	u16			    probe_delay;
+	u16			    min_chan_time;
+	u16			    max_chan_time;
+	u8			    chan_list[14];
+	u8			    bssid[ETH_ALEN];
+	struct iw_mgmt_essid_pset   ssid;
+	enum wl3501_scan_type	    scan_type;
+};
+
+struct wl3501_scan_confirm {
+	u16			    next_blk;
+	u8			    sig_id;
+	u8			    reserved;
+	u16			    status;
+	char			    timestamp[8];
+	char			    localtime[8];
+	struct wl3501_req	    req;
+	u8			    rssi;
+};
+
+struct wl3501_start_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	status;
+};
+
+struct wl3501_md_req {
+	u16	next_blk;
+	u8	sig_id;
+	u8	routing;
+	u16	data;
+	u16	size;
+	u8	pri;
+	u8	service_class;
+	struct {
+		u8	daddr[ETH_ALEN];
+		u8	saddr[ETH_ALEN];
+	} addr;
+};
+
+struct wl3501_md_ind {
+	u16	next_blk;
+	u8	sig_id;
+	u8	routing;
+	u16	data;
+	u16	size;
+	u8	reception;
+	u8	pri;
+	u8	service_class;
+	struct {
+		u8	daddr[ETH_ALEN];
+		u8	saddr[ETH_ALEN];
+	} addr;
+};
+
+struct wl3501_md_confirm {
+	u16	next_blk;
+	u8	sig_id;
+	u8	reserved;
+	u16	data;
+	u8	status;
+	u8	pri;
+	u8	service_class;
+};
+
+struct wl3501_resync_req {
+	u16	next_blk;
+	u8	sig_id;
+};
+
+/* Definitions for supporting clone adapters. */
+/* System Interface Registers (SIR space) */
+#define WL3501_NIC_GCR ((u8)0x00)	/* SIR0 - General Conf Register */
+#define WL3501_NIC_BSS ((u8)0x01)	/* SIR1 - Bank Switching Select Reg */
+#define WL3501_NIC_LMAL ((u8)0x02)	/* SIR2 - Local Mem addr Reg [7:0] */
+#define WL3501_NIC_LMAH ((u8)0x03)	/* SIR3 - Local Mem addr Reg [14:8] */
+#define WL3501_NIC_IODPA ((u8)0x04)	/* SIR4 - I/O Data Port A */
+#define WL3501_NIC_IODPB ((u8)0x05)	/* SIR5 - I/O Data Port B */
+#define WL3501_NIC_IODPC ((u8)0x06)	/* SIR6 - I/O Data Port C */
+#define WL3501_NIC_IODPD ((u8)0x07)	/* SIR7 - I/O Data Port D */
+
+/* Bits in GCR */
+#define WL3501_GCR_SWRESET ((u8)0x80)
+#define WL3501_GCR_CORESET ((u8)0x40)
+#define WL3501_GCR_DISPWDN ((u8)0x20)
+#define WL3501_GCR_ECWAIT  ((u8)0x10)
+#define WL3501_GCR_ECINT   ((u8)0x08)
+#define WL3501_GCR_INT2EC  ((u8)0x04)
+#define WL3501_GCR_ENECINT ((u8)0x02)
+#define WL3501_GCR_DAM     ((u8)0x01)
+
+/* Bits in BSS (Bank Switching Select Register) */
+#define WL3501_BSS_FPAGE0 ((u8)0x20)	/* Flash memory page0 */
+#define WL3501_BSS_FPAGE1 ((u8)0x28)
+#define WL3501_BSS_FPAGE2 ((u8)0x30)
+#define WL3501_BSS_FPAGE3 ((u8)0x38)
+#define WL3501_BSS_SPAGE0 ((u8)0x00)	/* SRAM page0 */
+#define WL3501_BSS_SPAGE1 ((u8)0x08)
+#define WL3501_BSS_SPAGE2 ((u8)0x10)
+#define WL3501_BSS_SPAGE3 ((u8)0x18)
+
+/* Define Driver Interface */
+/* Refer IEEE 802.11 */
+/* Tx packet header, include PLCP and MPDU */
+/* Tx PLCP Header */
+struct wl3501_80211_tx_plcp_hdr {
+	u8	sync[16];
+	u16	sfd;
+	u8	signal;
+	u8	service;
+	u16	len;
+	u16	crc16;
+} __packed;
+
+struct wl3501_80211_tx_hdr {
+	struct wl3501_80211_tx_plcp_hdr	pclp_hdr;
+	struct ieee80211_hdr		mac_hdr;
+} __packed __aligned(2);
+
+/*
+   Reserve the beginning Tx space for descriptor use.
+
+   TxBlockOffset -->	*----*----*----*----* \
+	(TxFreeDesc)	|  0 |  1 |  2 |  3 |  \
+			|  4 |  5 |  6 |  7 |   |
+			|  8 |  9 | 10 | 11 |   TX_DESC * 20
+			| 12 | 13 | 14 | 15 |   |
+			| 16 | 17 | 18 | 19 |  /
+   TxBufferBegin -->	*----*----*----*----* /
+   (TxBufferHead)	| 		    |
+   (TxBufferTail)	| 		    |
+			|    Send Buffer    |
+			| 		    |
+			|		    |
+			*-------------------*
+   TxBufferEnd    -------------------------/
+
+*/
+
+struct wl3501_card {
+	int				base_addr;
+	u8				mac_addr[ETH_ALEN];
+	spinlock_t			lock;
+	wait_queue_head_t		wait;
+	struct wl3501_get_confirm	sig_get_confirm;
+	struct wl3501_pwr_mgmt_confirm	sig_pwr_mgmt_confirm;
+	u16				tx_buffer_size;
+	u16				tx_buffer_head;
+	u16				tx_buffer_tail;
+	u16				tx_buffer_cnt;
+	u16				esbq_req_start;
+	u16				esbq_req_end;
+	u16				esbq_req_head;
+	u16				esbq_req_tail;
+	u16				esbq_confirm_start;
+	u16				esbq_confirm_end;
+	u16				esbq_confirm;
+	struct iw_mgmt_essid_pset  	essid;
+	struct iw_mgmt_essid_pset  	keep_essid;
+	u8				bssid[ETH_ALEN];
+	int				net_type;
+	char				nick[32];
+	char				card_name[32];
+	char				firmware_date[32];
+	u8				chan;
+	u8				cap_info;
+	u16				start_seg;
+	u16				bss_cnt;
+	u16				join_sta_bss;
+	u8				rssi;
+	u8				adhoc_times;
+	u8				reg_domain;
+	u8				version[2];
+	struct wl3501_scan_confirm	bss_set[20];
+
+	struct iw_statistics		wstats;
+	struct iw_spy_data		spy_data;
+	struct iw_public_data		wireless_data;
+	struct pcmcia_device		*p_dev;
+};
+#endif
diff --git a/drivers/net/wireless/legacy/wl3501_cs.c b/drivers/net/wireless/legacy/wl3501_cs.c
new file mode 100644
index 000000000000..7fb2f9513476
--- /dev/null
+++ b/drivers/net/wireless/legacy/wl3501_cs.c
@@ -0,0 +1,2030 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * WL3501 Wireless LAN PCMCIA Card Driver for Linux
+ * Written originally for Linux 2.0.30 by Fox Chen, mhchen@golf.ccl.itri.org.tw
+ * Ported to 2.2, 2.4 & 2.5 by Arnaldo Carvalho de Melo <acme@conectiva.com.br>
+ * Wireless extensions in 2.4 by Gustavo Niemeyer <niemeyer@conectiva.com>
+ *
+ * References used by Fox Chen while writing the original driver for 2.0.30:
+ *
+ *   1. WL24xx packet drivers (tooasm.asm)
+ *   2. Access Point Firmware Interface Specification for IEEE 802.11 SUTRO
+ *   3. IEEE 802.11
+ *   4. Linux network driver (/usr/src/linux/drivers/net)
+ *   5. ISA card driver - wl24.c
+ *   6. Linux PCMCIA skeleton driver - skeleton.c
+ *   7. Linux PCMCIA 3c589 network driver - 3c589_cs.c
+ *
+ * Tested with WL2400 firmware 1.2, Linux 2.0.30, and pcmcia-cs-2.9.12
+ *   1. Performance: about 165 Kbytes/sec in TCP/IP with Ad-Hoc mode.
+ *      rsh 192.168.1.3 "dd if=/dev/zero bs=1k count=1000" > /dev/null
+ *      (Specification 2M bits/sec. is about 250 Kbytes/sec., but we must deduct
+ *       ETHER/IP/UDP/TCP header, and acknowledgement overhead)
+ *
+ * Tested with Planet AP in 2.4.17, 184 Kbytes/s in UDP in Infrastructure mode,
+ * 173 Kbytes/s in TCP.
+ *
+ * Tested with Planet AP in 2.5.73-bk, 216 Kbytes/s in Infrastructure mode
+ * with a SMP machine (dual pentium 100), using pktgen, 432 pps (pkt_size = 60)
+ */
+
+#include <linux/delay.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/in.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/fcntl.h>
+#include <linux/if_arp.h>
+#include <linux/ioport.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/wireless.h>
+#include <net/cfg80211.h>
+
+#include <net/iw_handler.h>
+
+#include <pcmcia/cistpl.h>
+#include <pcmcia/cisreg.h>
+#include <pcmcia/ds.h>
+
+#include <asm/io.h>
+#include <linux/uaccess.h>
+
+#include "wl3501.h"
+
+#ifndef __i386__
+#define slow_down_io()
+#endif
+
+/* For rough constant delay */
+#define WL3501_NOPLOOP(n) { int x = 0; while (x++ < n) slow_down_io(); }
+
+
+
+#define wl3501_outb(a, b) { outb(a, b); slow_down_io(); }
+#define wl3501_outb_p(a, b) { outb_p(a, b); slow_down_io(); }
+#define wl3501_outsb(a, b, c) { outsb(a, b, c); slow_down_io(); }
+
+#define WL3501_RELEASE_TIMEOUT (25 * HZ)
+#define WL3501_MAX_ADHOC_TRIES 16
+
+#define WL3501_RESUME	0
+#define WL3501_SUSPEND	1
+
+static int wl3501_config(struct pcmcia_device *link);
+static void wl3501_release(struct pcmcia_device *link);
+
+static const struct {
+	int reg_domain;
+	int min, max, deflt;
+} iw_channel_table[] = {
+	{
+		.reg_domain = IW_REG_DOMAIN_FCC,
+		.min	    = 1,
+		.max	    = 11,
+		.deflt	    = 1,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_DOC,
+		.min	    = 1,
+		.max	    = 11,
+		.deflt	    = 1,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_ETSI,
+		.min	    = 1,
+		.max	    = 13,
+		.deflt	    = 1,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_SPAIN,
+		.min	    = 10,
+		.max	    = 11,
+		.deflt	    = 10,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_FRANCE,
+		.min	    = 10,
+		.max	    = 13,
+		.deflt	    = 10,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_MKK,
+		.min	    = 14,
+		.max	    = 14,
+		.deflt	    = 14,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_MKK1,
+		.min	    = 1,
+		.max	    = 14,
+		.deflt	    = 1,
+	},
+	{
+		.reg_domain = IW_REG_DOMAIN_ISRAEL,
+		.min	    = 3,
+		.max	    = 9,
+		.deflt	    = 9,
+	},
+};
+
+/**
+ * iw_valid_channel - validate channel in regulatory domain
+ * @reg_domain: regulatory domain
+ * @channel: channel to validate
+ *
+ * Returns 0 if invalid in the specified regulatory domain, non-zero if valid.
+ */
+static int iw_valid_channel(int reg_domain, int channel)
+{
+	int i, rc = 0;
+
+	for (i = 0; i < ARRAY_SIZE(iw_channel_table); i++)
+		if (reg_domain == iw_channel_table[i].reg_domain) {
+			rc = channel >= iw_channel_table[i].min &&
+			     channel <= iw_channel_table[i].max;
+			break;
+		}
+	return rc;
+}
+
+/**
+ * iw_default_channel - get default channel for a regulatory domain
+ * @reg_domain: regulatory domain
+ *
+ * Returns the default channel for a regulatory domain
+ */
+static int iw_default_channel(int reg_domain)
+{
+	int i, rc = 1;
+
+	for (i = 0; i < ARRAY_SIZE(iw_channel_table); i++)
+		if (reg_domain == iw_channel_table[i].reg_domain) {
+			rc = iw_channel_table[i].deflt;
+			break;
+		}
+	return rc;
+}
+
+static void iw_set_mgmt_info_element(enum iw_mgmt_info_element_ids id,
+				     struct iw_mgmt_info_element *el,
+				     void *value, int len)
+{
+	el->id  = id;
+	el->len = len;
+	memcpy(el->data, value, len);
+}
+
+static void iw_copy_mgmt_info_element(struct iw_mgmt_info_element *to,
+				      struct iw_mgmt_info_element *from)
+{
+	iw_set_mgmt_info_element(from->id, to, from->data, from->len);
+}
+
+static inline void wl3501_switch_page(struct wl3501_card *this, u8 page)
+{
+	wl3501_outb(page, this->base_addr + WL3501_NIC_BSS);
+}
+
+/*
+ * Get Ethernet MAC address.
+ *
+ * WARNING: We switch to FPAGE0 and switc back again.
+ *          Making sure there is no other WL function beening called by ISR.
+ */
+static int wl3501_get_flash_mac_addr(struct wl3501_card *this)
+{
+	int base_addr = this->base_addr;
+
+	/* get MAC addr */
+	wl3501_outb(WL3501_BSS_FPAGE3, base_addr + WL3501_NIC_BSS); /* BSS */
+	wl3501_outb(0x00, base_addr + WL3501_NIC_LMAL);	/* LMAL */
+	wl3501_outb(0x40, base_addr + WL3501_NIC_LMAH);	/* LMAH */
+
+	/* wait for reading EEPROM */
+	WL3501_NOPLOOP(100);
+	this->mac_addr[0] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->mac_addr[1] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->mac_addr[2] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->mac_addr[3] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->mac_addr[4] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->mac_addr[5] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->reg_domain = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	wl3501_outb(WL3501_BSS_FPAGE0, base_addr + WL3501_NIC_BSS);
+	wl3501_outb(0x04, base_addr + WL3501_NIC_LMAL);
+	wl3501_outb(0x40, base_addr + WL3501_NIC_LMAH);
+	WL3501_NOPLOOP(100);
+	this->version[0] = inb(base_addr + WL3501_NIC_IODPA);
+	WL3501_NOPLOOP(100);
+	this->version[1] = inb(base_addr + WL3501_NIC_IODPA);
+	/* switch to SRAM Page 0 (for safety) */
+	wl3501_switch_page(this, WL3501_BSS_SPAGE0);
+
+	/* The MAC addr should be 00:60:... */
+	return this->mac_addr[0] == 0x00 && this->mac_addr[1] == 0x60;
+}
+
+/**
+ * wl3501_set_to_wla - Move 'size' bytes from PC to card
+ * @this: Card
+ * @dest: Card addressing space
+ * @src: PC addressing space
+ * @size: Bytes to move
+ *
+ * Move 'size' bytes from PC to card. (Shouldn't be interrupted)
+ */
+static void wl3501_set_to_wla(struct wl3501_card *this, u16 dest, void *src,
+			      int size)
+{
+	/* switch to SRAM Page 0 */
+	wl3501_switch_page(this, (dest & 0x8000) ? WL3501_BSS_SPAGE1 :
+						   WL3501_BSS_SPAGE0);
+	/* set LMAL and LMAH */
+	wl3501_outb(dest & 0xff, this->base_addr + WL3501_NIC_LMAL);
+	wl3501_outb(((dest >> 8) & 0x7f), this->base_addr + WL3501_NIC_LMAH);
+
+	/* rep out to Port A */
+	wl3501_outsb(this->base_addr + WL3501_NIC_IODPA, src, size);
+}
+
+/**
+ * wl3501_get_from_wla - Move 'size' bytes from card to PC
+ * @this: Card
+ * @src: Card addressing space
+ * @dest: PC addressing space
+ * @size: Bytes to move
+ *
+ * Move 'size' bytes from card to PC. (Shouldn't be interrupted)
+ */
+static void wl3501_get_from_wla(struct wl3501_card *this, u16 src, void *dest,
+				int size)
+{
+	/* switch to SRAM Page 0 */
+	wl3501_switch_page(this, (src & 0x8000) ? WL3501_BSS_SPAGE1 :
+						  WL3501_BSS_SPAGE0);
+	/* set LMAL and LMAH */
+	wl3501_outb(src & 0xff, this->base_addr + WL3501_NIC_LMAL);
+	wl3501_outb((src >> 8) & 0x7f, this->base_addr + WL3501_NIC_LMAH);
+
+	/* rep get from Port A */
+	insb(this->base_addr + WL3501_NIC_IODPA, dest, size);
+}
+
+/*
+ * Get/Allocate a free Tx Data Buffer
+ *
+ *  *--------------*-----------------*----------------------------------*
+ *  |    PLCP      |    MAC Header   |  DST  SRC         Data ...       |
+ *  |  (24 bytes)  |    (30 bytes)   |  (6)  (6)  (Ethernet Row Data)   |
+ *  *--------------*-----------------*----------------------------------*
+ *  \               \- IEEE 802.11 -/ \-------------- len --------------/
+ *   \-struct wl3501_80211_tx_hdr--/   \-------- Ethernet Frame -------/
+ *
+ * Return = Position in Card
+ */
+static u16 wl3501_get_tx_buffer(struct wl3501_card *this, u16 len)
+{
+	u16 next, blk_cnt = 0, zero = 0;
+	u16 full_len = sizeof(struct wl3501_80211_tx_hdr) + len;
+	u16 ret = 0;
+
+	if (full_len > this->tx_buffer_cnt * 254)
+		goto out;
+	ret = this->tx_buffer_head;
+	while (full_len) {
+		if (full_len < 254)
+			full_len = 0;
+		else
+			full_len -= 254;
+		wl3501_get_from_wla(this, this->tx_buffer_head, &next,
+				    sizeof(next));
+		if (!full_len)
+			wl3501_set_to_wla(this, this->tx_buffer_head, &zero,
+					  sizeof(zero));
+		this->tx_buffer_head = next;
+		blk_cnt++;
+		/* if buffer is not enough */
+		if (!next && full_len) {
+			this->tx_buffer_head = ret;
+			ret = 0;
+			goto out;
+		}
+	}
+	this->tx_buffer_cnt -= blk_cnt;
+out:
+	return ret;
+}
+
+/*
+ * Free an allocated Tx Buffer. ptr must be correct position.
+ */
+static void wl3501_free_tx_buffer(struct wl3501_card *this, u16 ptr)
+{
+	/* check if all space is not free */
+	if (!this->tx_buffer_head)
+		this->tx_buffer_head = ptr;
+	else
+		wl3501_set_to_wla(this, this->tx_buffer_tail,
+				  &ptr, sizeof(ptr));
+	while (ptr) {
+		u16 next;
+
+		this->tx_buffer_cnt++;
+		wl3501_get_from_wla(this, ptr, &next, sizeof(next));
+		this->tx_buffer_tail = ptr;
+		ptr = next;
+	}
+}
+
+static int wl3501_esbq_req_test(struct wl3501_card *this)
+{
+	u8 tmp = 0;
+
+	wl3501_get_from_wla(this, this->esbq_req_head + 3, &tmp, sizeof(tmp));
+	return tmp & 0x80;
+}
+
+static void wl3501_esbq_req(struct wl3501_card *this, u16 *ptr)
+{
+	u16 tmp = 0;
+
+	wl3501_set_to_wla(this, this->esbq_req_head, ptr, 2);
+	wl3501_set_to_wla(this, this->esbq_req_head + 2, &tmp, sizeof(tmp));
+	this->esbq_req_head += 4;
+	if (this->esbq_req_head >= this->esbq_req_end)
+		this->esbq_req_head = this->esbq_req_start;
+}
+
+static int wl3501_esbq_exec(struct wl3501_card *this, void *sig, int sig_size)
+{
+	int rc = -EIO;
+
+	if (wl3501_esbq_req_test(this)) {
+		u16 ptr = wl3501_get_tx_buffer(this, sig_size);
+		if (ptr) {
+			wl3501_set_to_wla(this, ptr, sig, sig_size);
+			wl3501_esbq_req(this, &ptr);
+			rc = 0;
+		}
+	}
+	return rc;
+}
+
+static int wl3501_request_mib(struct wl3501_card *this, u8 index, void *bf)
+{
+	struct wl3501_get_req sig = {
+		.sig_id	    = WL3501_SIG_GET_REQ,
+		.mib_attrib = index,
+	};
+	unsigned long flags;
+	int rc = -EIO;
+
+	spin_lock_irqsave(&this->lock, flags);
+	if (wl3501_esbq_req_test(this)) {
+		u16 ptr = wl3501_get_tx_buffer(this, sizeof(sig));
+		if (ptr) {
+			wl3501_set_to_wla(this, ptr, &sig, sizeof(sig));
+			wl3501_esbq_req(this, &ptr);
+			this->sig_get_confirm.mib_status = 255;
+			rc = 0;
+		}
+	}
+	spin_unlock_irqrestore(&this->lock, flags);
+
+	return rc;
+}
+
+static int wl3501_get_mib_value(struct wl3501_card *this, u8 index,
+				void *bf, int size)
+{
+	int rc;
+
+	rc = wl3501_request_mib(this, index, bf);
+	if (rc)
+		return rc;
+
+	rc = wait_event_interruptible(this->wait,
+		this->sig_get_confirm.mib_status != 255);
+	if (rc)
+		return rc;
+
+	memcpy(bf, this->sig_get_confirm.mib_value, size);
+	return 0;
+}
+
+static int wl3501_pwr_mgmt(struct wl3501_card *this, int suspend)
+{
+	struct wl3501_pwr_mgmt_req sig = {
+		.sig_id		= WL3501_SIG_PWR_MGMT_REQ,
+		.pwr_save	= suspend,
+		.wake_up	= !suspend,
+		.receive_dtims	= 10,
+	};
+	unsigned long flags;
+	int rc = -EIO;
+
+	spin_lock_irqsave(&this->lock, flags);
+	if (wl3501_esbq_req_test(this)) {
+		u16 ptr = wl3501_get_tx_buffer(this, sizeof(sig));
+		if (ptr) {
+			wl3501_set_to_wla(this, ptr, &sig, sizeof(sig));
+			wl3501_esbq_req(this, &ptr);
+			this->sig_pwr_mgmt_confirm.status = 255;
+			spin_unlock_irqrestore(&this->lock, flags);
+			rc = wait_event_interruptible(this->wait,
+				this->sig_pwr_mgmt_confirm.status != 255);
+			printk(KERN_INFO "%s: %s status=%d\n", __func__,
+			       suspend ? "suspend" : "resume",
+			       this->sig_pwr_mgmt_confirm.status);
+			goto out;
+		}
+	}
+	spin_unlock_irqrestore(&this->lock, flags);
+out:
+	return rc;
+}
+
+/**
+ * wl3501_send_pkt - Send a packet.
+ * @this: Card
+ * @data: Ethernet raw frame.  (e.g. data[0] - data[5] is Dest MAC Addr,
+ *                                   data[6] - data[11] is Src MAC Addr)
+ * @len: Packet length
+ * Ref: IEEE 802.11
+ */
+static int wl3501_send_pkt(struct wl3501_card *this, u8 *data, u16 len)
+{
+	u16 bf, sig_bf, next, tmplen, pktlen;
+	struct wl3501_md_req sig = {
+		.sig_id = WL3501_SIG_MD_REQ,
+	};
+	size_t sig_addr_len = sizeof(sig.addr);
+	u8 *pdata = (char *)data;
+	int rc = -EIO;
+
+	if (wl3501_esbq_req_test(this)) {
+		sig_bf = wl3501_get_tx_buffer(this, sizeof(sig));
+		rc = -ENOMEM;
+		if (!sig_bf)	/* No free buffer available */
+			goto out;
+		bf = wl3501_get_tx_buffer(this, len + 26 + 24);
+		if (!bf) {
+			/* No free buffer available */
+			wl3501_free_tx_buffer(this, sig_bf);
+			goto out;
+		}
+		rc = 0;
+		memcpy(&sig.addr, pdata, sig_addr_len);
+		pktlen = len - sig_addr_len;
+		pdata += sig_addr_len;
+		sig.data = bf;
+		if (((*pdata) * 256 + (*(pdata + 1))) > 1500) {
+			u8 addr4[ETH_ALEN] = {
+				[0] = 0xAA, [1] = 0xAA, [2] = 0x03, [4] = 0x00,
+			};
+
+			wl3501_set_to_wla(this, bf + 2 +
+					  offsetof(struct wl3501_tx_hdr, addr4),
+					  addr4, sizeof(addr4));
+			sig.size = pktlen + 24 + 4 + 6;
+			if (pktlen > (254 - sizeof(struct wl3501_tx_hdr))) {
+				tmplen = 254 - sizeof(struct wl3501_tx_hdr);
+				pktlen -= tmplen;
+			} else {
+				tmplen = pktlen;
+				pktlen = 0;
+			}
+			wl3501_set_to_wla(this,
+					  bf + 2 + sizeof(struct wl3501_tx_hdr),
+					  pdata, tmplen);
+			pdata += tmplen;
+			wl3501_get_from_wla(this, bf, &next, sizeof(next));
+			bf = next;
+		} else {
+			sig.size = pktlen + 24 + 4 - 2;
+			pdata += 2;
+			pktlen -= 2;
+			if (pktlen > (254 - sizeof(struct wl3501_tx_hdr) + 6)) {
+				tmplen = 254 - sizeof(struct wl3501_tx_hdr) + 6;
+				pktlen -= tmplen;
+			} else {
+				tmplen = pktlen;
+				pktlen = 0;
+			}
+			wl3501_set_to_wla(this, bf + 2 +
+					  offsetof(struct wl3501_tx_hdr, addr4),
+					  pdata, tmplen);
+			pdata += tmplen;
+			wl3501_get_from_wla(this, bf, &next, sizeof(next));
+			bf = next;
+		}
+		while (pktlen > 0) {
+			if (pktlen > 254) {
+				tmplen = 254;
+				pktlen -= 254;
+			} else {
+				tmplen = pktlen;
+				pktlen = 0;
+			}
+			wl3501_set_to_wla(this, bf + 2, pdata, tmplen);
+			pdata += tmplen;
+			wl3501_get_from_wla(this, bf, &next, sizeof(next));
+			bf = next;
+		}
+		wl3501_set_to_wla(this, sig_bf, &sig, sizeof(sig));
+		wl3501_esbq_req(this, &sig_bf);
+	}
+out:
+	return rc;
+}
+
+static int wl3501_mgmt_resync(struct wl3501_card *this)
+{
+	struct wl3501_resync_req sig = {
+		.sig_id = WL3501_SIG_RESYNC_REQ,
+	};
+
+	return wl3501_esbq_exec(this, &sig, sizeof(sig));
+}
+
+static inline int wl3501_fw_bss_type(struct wl3501_card *this)
+{
+	return this->net_type == IW_MODE_INFRA ? WL3501_NET_TYPE_INFRA :
+						 WL3501_NET_TYPE_ADHOC;
+}
+
+static inline int wl3501_fw_cap_info(struct wl3501_card *this)
+{
+	return this->net_type == IW_MODE_INFRA ? WL3501_MGMT_CAPABILITY_ESS :
+						 WL3501_MGMT_CAPABILITY_IBSS;
+}
+
+static int wl3501_mgmt_scan(struct wl3501_card *this, u16 chan_time)
+{
+	struct wl3501_scan_req sig = {
+		.sig_id		= WL3501_SIG_SCAN_REQ,
+		.scan_type	= WL3501_SCAN_TYPE_ACTIVE,
+		.probe_delay	= 0x10,
+		.min_chan_time	= chan_time,
+		.max_chan_time	= chan_time,
+		.bss_type	= wl3501_fw_bss_type(this),
+	};
+
+	this->bss_cnt = this->join_sta_bss = 0;
+	return wl3501_esbq_exec(this, &sig, sizeof(sig));
+}
+
+static int wl3501_mgmt_join(struct wl3501_card *this, u16 stas)
+{
+	struct wl3501_join_req sig = {
+		.sig_id		  = WL3501_SIG_JOIN_REQ,
+		.timeout	  = 10,
+		.req.ds_pset = {
+			.el = {
+				.id  = IW_MGMT_INFO_ELEMENT_DS_PARAMETER_SET,
+				.len = 1,
+			},
+			.chan	= this->chan,
+		},
+	};
+
+	memcpy(&sig.req, &this->bss_set[stas].req, sizeof(sig.req));
+	return wl3501_esbq_exec(this, &sig, sizeof(sig));
+}
+
+static int wl3501_mgmt_start(struct wl3501_card *this)
+{
+	struct wl3501_start_req sig = {
+		.sig_id			= WL3501_SIG_START_REQ,
+		.beacon_period		= 400,
+		.dtim_period		= 1,
+		.ds_pset = {
+			.el = {
+				.id  = IW_MGMT_INFO_ELEMENT_DS_PARAMETER_SET,
+				.len = 1,
+			},
+			.chan	= this->chan,
+		},
+		.bss_basic_rset	= {
+			.el = {
+				.id	= IW_MGMT_INFO_ELEMENT_SUPPORTED_RATES,
+				.len = 2,
+			},
+			.data_rate_labels = {
+				[0] = IW_MGMT_RATE_LABEL_MANDATORY |
+				      IW_MGMT_RATE_LABEL_1MBIT,
+				[1] = IW_MGMT_RATE_LABEL_MANDATORY |
+				      IW_MGMT_RATE_LABEL_2MBIT,
+			},
+		},
+		.operational_rset	= {
+			.el = {
+				.id	= IW_MGMT_INFO_ELEMENT_SUPPORTED_RATES,
+				.len = 2,
+			},
+			.data_rate_labels = {
+				[0] = IW_MGMT_RATE_LABEL_MANDATORY |
+				      IW_MGMT_RATE_LABEL_1MBIT,
+				[1] = IW_MGMT_RATE_LABEL_MANDATORY |
+				      IW_MGMT_RATE_LABEL_2MBIT,
+			},
+		},
+		.ibss_pset		= {
+			.el = {
+				.id	 = IW_MGMT_INFO_ELEMENT_IBSS_PARAMETER_SET,
+				.len     = 2,
+			},
+			.atim_window = 10,
+		},
+		.bss_type		= wl3501_fw_bss_type(this),
+		.cap_info		= wl3501_fw_cap_info(this),
+	};
+
+	iw_copy_mgmt_info_element(&sig.ssid.el, &this->essid.el);
+	iw_copy_mgmt_info_element(&this->keep_essid.el, &this->essid.el);
+	return wl3501_esbq_exec(this, &sig, sizeof(sig));
+}
+
+static void wl3501_mgmt_scan_confirm(struct wl3501_card *this, u16 addr)
+{
+	u16 i = 0;
+	int matchflag = 0;
+	struct wl3501_scan_confirm sig;
+
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+	if (sig.status == WL3501_STATUS_SUCCESS) {
+		pr_debug("success");
+		if ((this->net_type == IW_MODE_INFRA &&
+		     (sig.req.cap_info & WL3501_MGMT_CAPABILITY_ESS)) ||
+		    (this->net_type == IW_MODE_ADHOC &&
+		     (sig.req.cap_info & WL3501_MGMT_CAPABILITY_IBSS)) ||
+		    this->net_type == IW_MODE_AUTO) {
+			if (!this->essid.el.len)
+				matchflag = 1;
+			else if (this->essid.el.len == 3 &&
+				 !memcmp(this->essid.essid, "ANY", 3))
+				matchflag = 1;
+			else if (this->essid.el.len != sig.req.ssid.el.len)
+				matchflag = 0;
+			else if (memcmp(this->essid.essid, sig.req.ssid.essid,
+					this->essid.el.len))
+				matchflag = 0;
+			else
+				matchflag = 1;
+			if (matchflag) {
+				for (i = 0; i < this->bss_cnt; i++) {
+					if (ether_addr_equal_unaligned(this->bss_set[i].req.bssid,
+								       sig.req.bssid)) {
+						matchflag = 0;
+						break;
+					}
+				}
+			}
+			if (matchflag && (i < 20)) {
+				memcpy(&this->bss_set[i].req,
+				       &sig.req, sizeof(sig.req));
+				this->bss_cnt++;
+				this->rssi = sig.rssi;
+				this->bss_set[i].rssi = sig.rssi;
+			}
+		}
+	} else if (sig.status == WL3501_STATUS_TIMEOUT) {
+		pr_debug("timeout");
+		this->join_sta_bss = 0;
+		for (i = this->join_sta_bss; i < this->bss_cnt; i++)
+			if (!wl3501_mgmt_join(this, i))
+				break;
+		this->join_sta_bss = i;
+		if (this->join_sta_bss == this->bss_cnt) {
+			if (this->net_type == IW_MODE_INFRA)
+				wl3501_mgmt_scan(this, 100);
+			else {
+				this->adhoc_times++;
+				if (this->adhoc_times > WL3501_MAX_ADHOC_TRIES)
+					wl3501_mgmt_start(this);
+				else
+					wl3501_mgmt_scan(this, 100);
+			}
+		}
+	}
+}
+
+/**
+ * wl3501_block_interrupt - Mask interrupt from SUTRO
+ * @this: Card
+ *
+ * Mask interrupt from SUTRO. (i.e. SUTRO cannot interrupt the HOST)
+ * Return: 1 if interrupt is originally enabled
+ */
+static int wl3501_block_interrupt(struct wl3501_card *this)
+{
+	u8 old = inb(this->base_addr + WL3501_NIC_GCR);
+	u8 new = old & (~(WL3501_GCR_ECINT | WL3501_GCR_INT2EC |
+			WL3501_GCR_ENECINT));
+
+	wl3501_outb(new, this->base_addr + WL3501_NIC_GCR);
+	return old & WL3501_GCR_ENECINT;
+}
+
+/**
+ * wl3501_unblock_interrupt - Enable interrupt from SUTRO
+ * @this: Card
+ *
+ * Enable interrupt from SUTRO. (i.e. SUTRO can interrupt the HOST)
+ * Return: 1 if interrupt is originally enabled
+ */
+static int wl3501_unblock_interrupt(struct wl3501_card *this)
+{
+	u8 old = inb(this->base_addr + WL3501_NIC_GCR);
+	u8 new = (old & ~(WL3501_GCR_ECINT | WL3501_GCR_INT2EC)) |
+		  WL3501_GCR_ENECINT;
+
+	wl3501_outb(new, this->base_addr + WL3501_NIC_GCR);
+	return old & WL3501_GCR_ENECINT;
+}
+
+/**
+ * wl3501_receive - Receive data from Receive Queue.
+ *
+ * Receive data from Receive Queue.
+ *
+ * @this: card
+ * @bf: address of host
+ * @size: size of buffer.
+ */
+static u16 wl3501_receive(struct wl3501_card *this, u8 *bf, u16 size)
+{
+	u16 next_addr, next_addr1;
+	u8 *data = bf + 12;
+
+	size -= 12;
+	wl3501_get_from_wla(this, this->start_seg + 2,
+			    &next_addr, sizeof(next_addr));
+	if (size > WL3501_BLKSZ - sizeof(struct wl3501_rx_hdr)) {
+		wl3501_get_from_wla(this,
+				    this->start_seg +
+					sizeof(struct wl3501_rx_hdr), data,
+				    WL3501_BLKSZ -
+					sizeof(struct wl3501_rx_hdr));
+		size -= WL3501_BLKSZ - sizeof(struct wl3501_rx_hdr);
+		data += WL3501_BLKSZ - sizeof(struct wl3501_rx_hdr);
+	} else {
+		wl3501_get_from_wla(this,
+				    this->start_seg +
+					sizeof(struct wl3501_rx_hdr),
+				    data, size);
+		size = 0;
+	}
+	while (size > 0) {
+		if (size > WL3501_BLKSZ - 5) {
+			wl3501_get_from_wla(this, next_addr + 5, data,
+					    WL3501_BLKSZ - 5);
+			size -= WL3501_BLKSZ - 5;
+			data += WL3501_BLKSZ - 5;
+			wl3501_get_from_wla(this, next_addr + 2, &next_addr1,
+					    sizeof(next_addr1));
+			next_addr = next_addr1;
+		} else {
+			wl3501_get_from_wla(this, next_addr + 5, data, size);
+			size = 0;
+		}
+	}
+	return 0;
+}
+
+static void wl3501_esbq_req_free(struct wl3501_card *this)
+{
+	u8 tmp;
+	u16 addr;
+
+	if (this->esbq_req_head == this->esbq_req_tail)
+		goto out;
+	wl3501_get_from_wla(this, this->esbq_req_tail + 3, &tmp, sizeof(tmp));
+	if (!(tmp & 0x80))
+		goto out;
+	wl3501_get_from_wla(this, this->esbq_req_tail, &addr, sizeof(addr));
+	wl3501_free_tx_buffer(this, addr);
+	this->esbq_req_tail += 4;
+	if (this->esbq_req_tail >= this->esbq_req_end)
+		this->esbq_req_tail = this->esbq_req_start;
+out:
+	return;
+}
+
+static int wl3501_esbq_confirm(struct wl3501_card *this)
+{
+	u8 tmp;
+
+	wl3501_get_from_wla(this, this->esbq_confirm + 3, &tmp, sizeof(tmp));
+	return tmp & 0x80;
+}
+
+static void wl3501_online(struct net_device *dev)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	printk(KERN_INFO "%s: Wireless LAN online. BSSID: %pM\n",
+	       dev->name, this->bssid);
+	netif_wake_queue(dev);
+}
+
+static void wl3501_esbq_confirm_done(struct wl3501_card *this)
+{
+	u8 tmp = 0;
+
+	wl3501_set_to_wla(this, this->esbq_confirm + 3, &tmp, sizeof(tmp));
+	this->esbq_confirm += 4;
+	if (this->esbq_confirm >= this->esbq_confirm_end)
+		this->esbq_confirm = this->esbq_confirm_start;
+}
+
+static int wl3501_mgmt_auth(struct wl3501_card *this)
+{
+	struct wl3501_auth_req sig = {
+		.sig_id	 = WL3501_SIG_AUTH_REQ,
+		.type	 = WL3501_SYS_TYPE_OPEN,
+		.timeout = 1000,
+	};
+
+	pr_debug("entry");
+	memcpy(sig.mac_addr, this->bssid, ETH_ALEN);
+	return wl3501_esbq_exec(this, &sig, sizeof(sig));
+}
+
+static int wl3501_mgmt_association(struct wl3501_card *this)
+{
+	struct wl3501_assoc_req sig = {
+		.sig_id		 = WL3501_SIG_ASSOC_REQ,
+		.timeout	 = 1000,
+		.listen_interval = 5,
+		.cap_info	 = this->cap_info,
+	};
+
+	pr_debug("entry");
+	memcpy(sig.mac_addr, this->bssid, ETH_ALEN);
+	return wl3501_esbq_exec(this, &sig, sizeof(sig));
+}
+
+static void wl3501_mgmt_join_confirm(struct net_device *dev, u16 addr)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	struct wl3501_join_confirm sig;
+
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+	if (sig.status == WL3501_STATUS_SUCCESS) {
+		if (this->net_type == IW_MODE_INFRA) {
+			if (this->join_sta_bss < this->bss_cnt) {
+				const int i = this->join_sta_bss;
+				memcpy(this->bssid,
+				       this->bss_set[i].req.bssid, ETH_ALEN);
+				this->chan = this->bss_set[i].req.ds_pset.chan;
+				iw_copy_mgmt_info_element(&this->keep_essid.el,
+						     &this->bss_set[i].req.ssid.el);
+				wl3501_mgmt_auth(this);
+			}
+		} else {
+			const int i = this->join_sta_bss;
+
+			memcpy(&this->bssid, &this->bss_set[i].req.bssid, ETH_ALEN);
+			this->chan = this->bss_set[i].req.ds_pset.chan;
+			iw_copy_mgmt_info_element(&this->keep_essid.el,
+						  &this->bss_set[i].req.ssid.el);
+			wl3501_online(dev);
+		}
+	} else {
+		int i;
+		this->join_sta_bss++;
+		for (i = this->join_sta_bss; i < this->bss_cnt; i++)
+			if (!wl3501_mgmt_join(this, i))
+				break;
+		this->join_sta_bss = i;
+		if (this->join_sta_bss == this->bss_cnt) {
+			if (this->net_type == IW_MODE_INFRA)
+				wl3501_mgmt_scan(this, 100);
+			else {
+				this->adhoc_times++;
+				if (this->adhoc_times > WL3501_MAX_ADHOC_TRIES)
+					wl3501_mgmt_start(this);
+				else
+					wl3501_mgmt_scan(this, 100);
+			}
+		}
+	}
+}
+
+static inline void wl3501_alarm_interrupt(struct net_device *dev,
+					  struct wl3501_card *this)
+{
+	if (this->net_type == IW_MODE_INFRA) {
+		printk(KERN_INFO "Wireless LAN offline\n");
+		netif_stop_queue(dev);
+		wl3501_mgmt_resync(this);
+	}
+}
+
+static inline void wl3501_md_confirm_interrupt(struct net_device *dev,
+					       struct wl3501_card *this,
+					       u16 addr)
+{
+	struct wl3501_md_confirm sig;
+
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+	wl3501_free_tx_buffer(this, sig.data);
+	if (netif_queue_stopped(dev))
+		netif_wake_queue(dev);
+}
+
+static inline void wl3501_md_ind_interrupt(struct net_device *dev,
+					   struct wl3501_card *this, u16 addr)
+{
+	struct wl3501_md_ind sig;
+	struct sk_buff *skb;
+	u8 rssi, addr4[ETH_ALEN];
+	u16 pkt_len;
+
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+	this->start_seg = sig.data;
+	wl3501_get_from_wla(this,
+			    sig.data + offsetof(struct wl3501_rx_hdr, rssi),
+			    &rssi, sizeof(rssi));
+	this->rssi = rssi <= 63 ? (rssi * 100) / 64 : 255;
+
+	wl3501_get_from_wla(this,
+			    sig.data +
+				offsetof(struct wl3501_rx_hdr, addr4),
+			    &addr4, sizeof(addr4));
+	if (!(addr4[0] == 0xAA && addr4[1] == 0xAA &&
+	      addr4[2] == 0x03 && addr4[4] == 0x00)) {
+		printk(KERN_INFO "Unsupported packet type!\n");
+		return;
+	}
+	pkt_len = sig.size + 12 - 24 - 4 - 6;
+
+	skb = dev_alloc_skb(pkt_len + 5);
+
+	if (!skb) {
+		printk(KERN_WARNING "%s: Can't alloc a sk_buff of size %d.\n",
+		       dev->name, pkt_len);
+		dev->stats.rx_dropped++;
+	} else {
+		skb->dev = dev;
+		skb_reserve(skb, 2); /* IP headers on 16 bytes boundaries */
+		skb_copy_to_linear_data(skb, (unsigned char *)&sig.addr,
+					sizeof(sig.addr));
+		wl3501_receive(this, skb->data, pkt_len);
+		skb_put(skb, pkt_len);
+		skb->protocol	= eth_type_trans(skb, dev);
+		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += skb->len;
+		netif_rx(skb);
+	}
+}
+
+static inline void wl3501_get_confirm_interrupt(struct wl3501_card *this,
+						u16 addr, void *sig, int size)
+{
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &this->sig_get_confirm,
+			    sizeof(this->sig_get_confirm));
+	wake_up(&this->wait);
+}
+
+static inline void wl3501_start_confirm_interrupt(struct net_device *dev,
+						  struct wl3501_card *this,
+						  u16 addr)
+{
+	struct wl3501_start_confirm sig;
+
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+	if (sig.status == WL3501_STATUS_SUCCESS)
+		netif_wake_queue(dev);
+}
+
+static inline void wl3501_assoc_confirm_interrupt(struct net_device *dev,
+						  u16 addr)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	struct wl3501_assoc_confirm sig;
+
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+
+	if (sig.status == WL3501_STATUS_SUCCESS)
+		wl3501_online(dev);
+}
+
+static inline void wl3501_auth_confirm_interrupt(struct wl3501_card *this,
+						 u16 addr)
+{
+	struct wl3501_auth_confirm sig;
+
+	pr_debug("entry");
+	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
+
+	if (sig.status == WL3501_STATUS_SUCCESS)
+		wl3501_mgmt_association(this);
+	else
+		wl3501_mgmt_resync(this);
+}
+
+static inline void wl3501_rx_interrupt(struct net_device *dev)
+{
+	int morepkts;
+	u16 addr;
+	u8 sig_id;
+	struct wl3501_card *this = netdev_priv(dev);
+
+	pr_debug("entry");
+loop:
+	morepkts = 0;
+	if (!wl3501_esbq_confirm(this))
+		goto free;
+	wl3501_get_from_wla(this, this->esbq_confirm, &addr, sizeof(addr));
+	wl3501_get_from_wla(this, addr + 2, &sig_id, sizeof(sig_id));
+
+	switch (sig_id) {
+	case WL3501_SIG_DEAUTH_IND:
+	case WL3501_SIG_DISASSOC_IND:
+	case WL3501_SIG_ALARM:
+		wl3501_alarm_interrupt(dev, this);
+		break;
+	case WL3501_SIG_MD_CONFIRM:
+		wl3501_md_confirm_interrupt(dev, this, addr);
+		break;
+	case WL3501_SIG_MD_IND:
+		wl3501_md_ind_interrupt(dev, this, addr);
+		break;
+	case WL3501_SIG_GET_CONFIRM:
+		wl3501_get_confirm_interrupt(this, addr,
+					     &this->sig_get_confirm,
+					     sizeof(this->sig_get_confirm));
+		break;
+	case WL3501_SIG_PWR_MGMT_CONFIRM:
+		wl3501_get_confirm_interrupt(this, addr,
+					     &this->sig_pwr_mgmt_confirm,
+					    sizeof(this->sig_pwr_mgmt_confirm));
+		break;
+	case WL3501_SIG_START_CONFIRM:
+		wl3501_start_confirm_interrupt(dev, this, addr);
+		break;
+	case WL3501_SIG_SCAN_CONFIRM:
+		wl3501_mgmt_scan_confirm(this, addr);
+		break;
+	case WL3501_SIG_JOIN_CONFIRM:
+		wl3501_mgmt_join_confirm(dev, addr);
+		break;
+	case WL3501_SIG_ASSOC_CONFIRM:
+		wl3501_assoc_confirm_interrupt(dev, addr);
+		break;
+	case WL3501_SIG_AUTH_CONFIRM:
+		wl3501_auth_confirm_interrupt(this, addr);
+		break;
+	case WL3501_SIG_RESYNC_CONFIRM:
+		wl3501_mgmt_resync(this); /* FIXME: should be resync_confirm */
+		break;
+	}
+	wl3501_esbq_confirm_done(this);
+	morepkts = 1;
+	/* free request if necessary */
+free:
+	wl3501_esbq_req_free(this);
+	if (morepkts)
+		goto loop;
+}
+
+static inline void wl3501_ack_interrupt(struct wl3501_card *this)
+{
+	wl3501_outb(WL3501_GCR_ECINT, this->base_addr + WL3501_NIC_GCR);
+}
+
+/**
+ * wl3501_interrupt - Hardware interrupt from card.
+ * @irq: Interrupt number
+ * @dev_id: net_device
+ *
+ * We must acknowledge the interrupt as soon as possible, and block the
+ * interrupt from the same card immediately to prevent re-entry.
+ *
+ * Before accessing the Control_Status_Block, we must lock SUTRO first.
+ * On the other hand, to prevent SUTRO from malfunctioning, we must
+ * unlock the SUTRO as soon as possible.
+ */
+static irqreturn_t wl3501_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct wl3501_card *this;
+
+	this = netdev_priv(dev);
+	spin_lock(&this->lock);
+	wl3501_ack_interrupt(this);
+	wl3501_block_interrupt(this);
+	wl3501_rx_interrupt(dev);
+	wl3501_unblock_interrupt(this);
+	spin_unlock(&this->lock);
+
+	return IRQ_HANDLED;
+}
+
+static int wl3501_reset_board(struct wl3501_card *this)
+{
+	u8 tmp = 0;
+	int i, rc = 0;
+
+	/* Coreset */
+	wl3501_outb_p(WL3501_GCR_CORESET, this->base_addr + WL3501_NIC_GCR);
+	wl3501_outb_p(0, this->base_addr + WL3501_NIC_GCR);
+	wl3501_outb_p(WL3501_GCR_CORESET, this->base_addr + WL3501_NIC_GCR);
+
+	/* Reset SRAM 0x480 to zero */
+	wl3501_set_to_wla(this, 0x480, &tmp, sizeof(tmp));
+
+	/* Start up */
+	wl3501_outb_p(0, this->base_addr + WL3501_NIC_GCR);
+
+	WL3501_NOPLOOP(1024 * 50);
+
+	wl3501_unblock_interrupt(this);	/* acme: was commented */
+
+	/* Polling Self_Test_Status */
+	for (i = 0; i < 10000; i++) {
+		wl3501_get_from_wla(this, 0x480, &tmp, sizeof(tmp));
+
+		if (tmp == 'W') {
+			/* firmware complete all test successfully */
+			tmp = 'A';
+			wl3501_set_to_wla(this, 0x480, &tmp, sizeof(tmp));
+			goto out;
+		}
+		WL3501_NOPLOOP(10);
+	}
+	printk(KERN_WARNING "%s: failed to reset the board!\n", __func__);
+	rc = -ENODEV;
+out:
+	return rc;
+}
+
+static int wl3501_init_firmware(struct wl3501_card *this)
+{
+	u16 ptr, next;
+	int rc = wl3501_reset_board(this);
+
+	if (rc)
+		goto fail;
+	this->card_name[0] = '\0';
+	wl3501_get_from_wla(this, 0x1a00,
+			    this->card_name, sizeof(this->card_name));
+	this->card_name[sizeof(this->card_name) - 1] = '\0';
+	this->firmware_date[0] = '\0';
+	wl3501_get_from_wla(this, 0x1a40,
+			    this->firmware_date, sizeof(this->firmware_date));
+	this->firmware_date[sizeof(this->firmware_date) - 1] = '\0';
+	/* Switch to SRAM Page 0 */
+	wl3501_switch_page(this, WL3501_BSS_SPAGE0);
+	/* Read parameter from card */
+	wl3501_get_from_wla(this, 0x482, &this->esbq_req_start, 2);
+	wl3501_get_from_wla(this, 0x486, &this->esbq_req_end, 2);
+	wl3501_get_from_wla(this, 0x488, &this->esbq_confirm_start, 2);
+	wl3501_get_from_wla(this, 0x48c, &this->esbq_confirm_end, 2);
+	wl3501_get_from_wla(this, 0x48e, &this->tx_buffer_head, 2);
+	wl3501_get_from_wla(this, 0x492, &this->tx_buffer_size, 2);
+	this->esbq_req_tail	= this->esbq_req_head = this->esbq_req_start;
+	this->esbq_req_end     += this->esbq_req_start;
+	this->esbq_confirm	= this->esbq_confirm_start;
+	this->esbq_confirm_end += this->esbq_confirm_start;
+	/* Initial Tx Buffer */
+	this->tx_buffer_cnt = 1;
+	ptr = this->tx_buffer_head;
+	next = ptr + WL3501_BLKSZ;
+	while ((next - this->tx_buffer_head) < this->tx_buffer_size) {
+		this->tx_buffer_cnt++;
+		wl3501_set_to_wla(this, ptr, &next, sizeof(next));
+		ptr = next;
+		next = ptr + WL3501_BLKSZ;
+	}
+	rc = 0;
+	next = 0;
+	wl3501_set_to_wla(this, ptr, &next, sizeof(next));
+	this->tx_buffer_tail = ptr;
+out:
+	return rc;
+fail:
+	printk(KERN_WARNING "%s: failed!\n", __func__);
+	goto out;
+}
+
+static int wl3501_close(struct net_device *dev)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	unsigned long flags;
+	struct pcmcia_device *link;
+	link = this->p_dev;
+
+	spin_lock_irqsave(&this->lock, flags);
+	link->open--;
+
+	/* Stop wl3501_hard_start_xmit() from now on */
+	netif_stop_queue(dev);
+	wl3501_ack_interrupt(this);
+
+	/* Mask interrupts from the SUTRO */
+	wl3501_block_interrupt(this);
+
+	printk(KERN_INFO "%s: WL3501 closed\n", dev->name);
+	spin_unlock_irqrestore(&this->lock, flags);
+	return 0;
+}
+
+/**
+ * wl3501_reset - Reset the SUTRO.
+ * @dev: network device
+ *
+ * It is almost the same as wl3501_open(). In fact, we may just wl3501_close()
+ * and wl3501_open() again, but I wouldn't like to free_irq() when the driver
+ * is running. It seems to be dangerous.
+ */
+static int wl3501_reset(struct net_device *dev)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = -ENODEV;
+	unsigned long flags;
+
+	spin_lock_irqsave(&this->lock, flags);
+	wl3501_block_interrupt(this);
+
+	if (wl3501_init_firmware(this)) {
+		printk(KERN_WARNING "%s: Can't initialize Firmware!\n",
+		       dev->name);
+		/* Free IRQ, and mark IRQ as unused */
+		free_irq(dev->irq, dev);
+		goto out;
+	}
+
+	/*
+	 * Queue has to be started only when the Card is Started
+	 */
+	netif_stop_queue(dev);
+	this->adhoc_times = 0;
+	wl3501_ack_interrupt(this);
+	wl3501_unblock_interrupt(this);
+	wl3501_mgmt_scan(this, 100);
+	pr_debug("%s: device reset", dev->name);
+	rc = 0;
+out:
+	spin_unlock_irqrestore(&this->lock, flags);
+	return rc;
+}
+
+static void wl3501_tx_timeout(struct net_device *dev, unsigned int txqueue)
+{
+	struct net_device_stats *stats = &dev->stats;
+	int rc;
+
+	stats->tx_errors++;
+	rc = wl3501_reset(dev);
+	if (rc)
+		printk(KERN_ERR "%s: Error %d resetting card on Tx timeout!\n",
+		       dev->name, rc);
+	else {
+		netif_trans_update(dev); /* prevent tx timeout */
+		netif_wake_queue(dev);
+	}
+}
+
+/*
+ * Return : 0 - OK
+ *	    1 - Could not transmit (dev_queue_xmit will queue it)
+ *		and try to sent it later
+ */
+static netdev_tx_t wl3501_hard_start_xmit(struct sk_buff *skb,
+						struct net_device *dev)
+{
+	int enabled, rc;
+	struct wl3501_card *this = netdev_priv(dev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&this->lock, flags);
+	enabled = wl3501_block_interrupt(this);
+	rc = wl3501_send_pkt(this, skb->data, skb->len);
+	if (enabled)
+		wl3501_unblock_interrupt(this);
+	if (rc) {
+		++dev->stats.tx_dropped;
+		netif_stop_queue(dev);
+	} else {
+		++dev->stats.tx_packets;
+		dev->stats.tx_bytes += skb->len;
+		dev_kfree_skb_irq(skb);
+
+		if (this->tx_buffer_cnt < 2)
+			netif_stop_queue(dev);
+	}
+	spin_unlock_irqrestore(&this->lock, flags);
+	return NETDEV_TX_OK;
+}
+
+static int wl3501_open(struct net_device *dev)
+{
+	int rc = -ENODEV;
+	struct wl3501_card *this = netdev_priv(dev);
+	unsigned long flags;
+	struct pcmcia_device *link;
+	link = this->p_dev;
+
+	spin_lock_irqsave(&this->lock, flags);
+	if (!pcmcia_dev_present(link))
+		goto out;
+	netif_device_attach(dev);
+	link->open++;
+
+	/* Initial WL3501 firmware */
+	pr_debug("%s: Initialize WL3501 firmware...", dev->name);
+	if (wl3501_init_firmware(this))
+		goto fail;
+	/* Initial device variables */
+	this->adhoc_times = 0;
+	/* Acknowledge Interrupt, for cleaning last state */
+	wl3501_ack_interrupt(this);
+
+	/* Enable interrupt from card after all */
+	wl3501_unblock_interrupt(this);
+	wl3501_mgmt_scan(this, 100);
+	rc = 0;
+	pr_debug("%s: WL3501 opened", dev->name);
+	printk(KERN_INFO "%s: Card Name: %s\n"
+			 "%s: Firmware Date: %s\n",
+			 dev->name, this->card_name,
+			 dev->name, this->firmware_date);
+out:
+	spin_unlock_irqrestore(&this->lock, flags);
+	return rc;
+fail:
+	printk(KERN_WARNING "%s: Can't initialize firmware!\n", dev->name);
+	goto out;
+}
+
+static struct iw_statistics *wl3501_get_wireless_stats(struct net_device *dev)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	struct iw_statistics *wstats = &this->wstats;
+	u32 value; /* size checked: it is u32 */
+
+	memset(wstats, 0, sizeof(*wstats));
+	wstats->status = netif_running(dev);
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_ICV_ERROR_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.code += value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_UNDECRYPTABLE_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.code += value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_EXCLUDED_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.code += value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_RETRY_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.retries	= value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_FAILED_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.misc += value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_RTS_FAILURE_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.misc += value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_ACK_FAILURE_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.misc += value;
+	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_FRAME_DUPLICATE_COUNT,
+				  &value, sizeof(value)))
+		wstats->discard.misc += value;
+	return wstats;
+}
+
+/**
+ * wl3501_detach - deletes a driver "instance"
+ * @link: FILL_IN
+ *
+ * This deletes a driver "instance". The device is de-registered with Card
+ * Services. If it has been released, all local data structures are freed.
+ * Otherwise, the structures will be freed when the device is released.
+ */
+static void wl3501_detach(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+
+	/* If the device is currently configured and active, we won't actually
+	 * delete it yet.  Instead, it is marked so that when the release()
+	 * function is called, that will trigger a proper detach(). */
+
+	while (link->open > 0)
+		wl3501_close(dev);
+
+	netif_device_detach(dev);
+	wl3501_release(link);
+
+	unregister_netdev(dev);
+	free_netdev(dev);
+}
+
+static int wl3501_get_name(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	strscpy(wrqu->name, "IEEE 802.11-DS", sizeof(wrqu->name));
+	return 0;
+}
+
+static int wl3501_set_freq(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	int channel = wrqu->freq.m;
+	int rc = -EINVAL;
+
+	if (iw_valid_channel(this->reg_domain, channel)) {
+		this->chan = channel;
+		rc = wl3501_reset(dev);
+	}
+	return rc;
+}
+
+static int wl3501_get_freq(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	wrqu->freq.m = 100000 *
+		ieee80211_channel_to_frequency(this->chan, NL80211_BAND_2GHZ);
+	wrqu->freq.e = 1;
+	return 0;
+}
+
+static int wl3501_set_mode(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	int rc = -EINVAL;
+
+	if (wrqu->mode == IW_MODE_INFRA ||
+	    wrqu->mode == IW_MODE_ADHOC ||
+	    wrqu->mode == IW_MODE_AUTO) {
+		struct wl3501_card *this = netdev_priv(dev);
+
+		this->net_type = wrqu->mode;
+		rc = wl3501_reset(dev);
+	}
+	return rc;
+}
+
+static int wl3501_get_mode(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	wrqu->mode = this->net_type;
+	return 0;
+}
+
+static int wl3501_get_sens(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	wrqu->sens.value = this->rssi;
+	wrqu->sens.disabled = !wrqu->sens.value;
+	wrqu->sens.fixed = 1;
+	return 0;
+}
+
+static int wl3501_get_range(struct net_device *dev,
+			    struct iw_request_info *info,
+			    union iwreq_data *wrqu, char *extra)
+{
+	struct iw_range *range = (struct iw_range *)extra;
+
+	/* Set the length (very important for backward compatibility) */
+	wrqu->data.length = sizeof(*range);
+
+	/* Set all the info we don't care or don't know about to zero */
+	memset(range, 0, sizeof(*range));
+
+	/* Set the Wireless Extension versions */
+	range->we_version_compiled	= WIRELESS_EXT;
+	range->we_version_source	= 1;
+	range->throughput		= 2 * 1000 * 1000;     /* ~2 Mb/s */
+	/* FIXME: study the code to fill in more fields... */
+	return 0;
+}
+
+static int wl3501_set_wap(struct net_device *dev, struct iw_request_info *info,
+			  union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = -EINVAL;
+
+	/* FIXME: we support other ARPHRDs...*/
+	if (wrqu->ap_addr.sa_family != ARPHRD_ETHER)
+		goto out;
+	if (is_broadcast_ether_addr(wrqu->ap_addr.sa_data)) {
+		/* FIXME: rescan? */
+	} else
+		memcpy(this->bssid, wrqu->ap_addr.sa_data, ETH_ALEN);
+		/* FIXME: rescan? deassoc & scan? */
+	rc = 0;
+out:
+	return rc;
+}
+
+static int wl3501_get_wap(struct net_device *dev, struct iw_request_info *info,
+			  union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	wrqu->ap_addr.sa_family = ARPHRD_ETHER;
+	memcpy(wrqu->ap_addr.sa_data, this->bssid, ETH_ALEN);
+	return 0;
+}
+
+static int wl3501_set_scan(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	/*
+	 * FIXME: trigger scanning with a reset, yes, I'm lazy
+	 */
+	return wl3501_reset(dev);
+}
+
+static int wl3501_get_scan(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	int i;
+	char *current_ev = extra;
+	struct iw_event iwe;
+
+	for (i = 0; i < this->bss_cnt; ++i) {
+		iwe.cmd			= SIOCGIWAP;
+		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
+		memcpy(iwe.u.ap_addr.sa_data, this->bss_set[i].req.bssid, ETH_ALEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_ADDR_LEN);
+		iwe.cmd		  = SIOCGIWESSID;
+		iwe.u.data.flags  = 1;
+		iwe.u.data.length = this->bss_set[i].req.ssid.el.len;
+		current_ev = iwe_stream_add_point(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe,
+						  this->bss_set[i].req.ssid.essid);
+		iwe.cmd	   = SIOCGIWMODE;
+		iwe.u.mode = this->bss_set[i].req.bss_type;
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_UINT_LEN);
+		iwe.cmd = SIOCGIWFREQ;
+		iwe.u.freq.m = this->bss_set[i].req.ds_pset.chan;
+		iwe.u.freq.e = 0;
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_FREQ_LEN);
+		iwe.cmd = SIOCGIWENCODE;
+		if (this->bss_set[i].req.cap_info & WL3501_MGMT_CAPABILITY_PRIVACY)
+			iwe.u.data.flags = IW_ENCODE_ENABLED | IW_ENCODE_NOKEY;
+		else
+			iwe.u.data.flags = IW_ENCODE_DISABLED;
+		iwe.u.data.length = 0;
+		current_ev = iwe_stream_add_point(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, NULL);
+	}
+	/* Length of data */
+	wrqu->data.length = (current_ev - extra);
+	wrqu->data.flags = 0; /* FIXME: set properly these flags */
+	return 0;
+}
+
+static int wl3501_set_essid(struct net_device *dev,
+			    struct iw_request_info *info,
+			    union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	if (wrqu->data.flags) {
+		iw_set_mgmt_info_element(IW_MGMT_INFO_ELEMENT_SSID,
+					 &this->essid.el,
+					 extra, wrqu->data.length);
+	} else { /* We accept any ESSID */
+		iw_set_mgmt_info_element(IW_MGMT_INFO_ELEMENT_SSID,
+					 &this->essid.el, "ANY", 3);
+	}
+	return wl3501_reset(dev);
+}
+
+static int wl3501_get_essid(struct net_device *dev,
+			    struct iw_request_info *info,
+			    union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&this->lock, flags);
+	wrqu->essid.flags  = 1;
+	wrqu->essid.length = this->essid.el.len;
+	memcpy(extra, this->essid.essid, this->essid.el.len);
+	spin_unlock_irqrestore(&this->lock, flags);
+	return 0;
+}
+
+static int wl3501_set_nick(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	if (wrqu->data.length > sizeof(this->nick))
+		return -E2BIG;
+	strscpy(this->nick, extra, wrqu->data.length);
+	return 0;
+}
+
+static int wl3501_get_nick(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	struct wl3501_card *this = netdev_priv(dev);
+
+	strscpy(extra, this->nick, 32);
+	wrqu->data.length = strlen(extra);
+	return 0;
+}
+
+static int wl3501_get_rate(struct net_device *dev, struct iw_request_info *info,
+			   union iwreq_data *wrqu, char *extra)
+{
+	/*
+	 * FIXME: have to see from where to get this info, perhaps this card
+	 * works at 1 Mbit/s too... for now leave at 2 Mbit/s that is the most
+	 * common with the Planet Access Points. -acme
+	 */
+	wrqu->bitrate.value = 2000000;
+	wrqu->bitrate.fixed = 1;
+	return 0;
+}
+
+static int wl3501_get_rts_threshold(struct net_device *dev,
+				    struct iw_request_info *info,
+				    union iwreq_data *wrqu, char *extra)
+{
+	u16 threshold; /* size checked: it is u16 */
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_RTS_THRESHOLD,
+				      &threshold, sizeof(threshold));
+	if (!rc) {
+		wrqu->rts.value = threshold;
+		wrqu->rts.disabled = threshold >= 2347;
+		wrqu->rts.fixed = 1;
+	}
+	return rc;
+}
+
+static int wl3501_get_frag_threshold(struct net_device *dev,
+				     struct iw_request_info *info,
+				     union iwreq_data *wrqu, char *extra)
+{
+	u16 threshold; /* size checked: it is u16 */
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_FRAG_THRESHOLD,
+				      &threshold, sizeof(threshold));
+	if (!rc) {
+		wrqu->frag.value = threshold;
+		wrqu->frag.disabled = threshold >= 2346;
+		wrqu->frag.fixed = 1;
+	}
+	return rc;
+}
+
+static int wl3501_get_txpow(struct net_device *dev,
+			    struct iw_request_info *info,
+			    union iwreq_data *wrqu, char *extra)
+{
+	u16 txpow;
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = wl3501_get_mib_value(this,
+				      WL3501_MIB_ATTR_CURRENT_TX_PWR_LEVEL,
+				      &txpow, sizeof(txpow));
+	if (!rc) {
+		wrqu->txpower.value = txpow;
+		wrqu->txpower.disabled = 0;
+		/*
+		 * From the MIB values I think this can be configurable,
+		 * as it lists several tx power levels -acme
+		 */
+		wrqu->txpower.fixed = 0;
+		wrqu->txpower.flags = IW_TXPOW_MWATT;
+	}
+	return rc;
+}
+
+static int wl3501_get_retry(struct net_device *dev,
+			    struct iw_request_info *info,
+			    union iwreq_data *wrqu, char *extra)
+{
+	u8 retry; /* size checked: it is u8 */
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = wl3501_get_mib_value(this,
+				      WL3501_MIB_ATTR_LONG_RETRY_LIMIT,
+				      &retry, sizeof(retry));
+	if (rc)
+		goto out;
+	if (wrqu->retry.flags & IW_RETRY_LONG) {
+		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
+		goto set_value;
+	}
+	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_SHORT_RETRY_LIMIT,
+				  &retry, sizeof(retry));
+	if (rc)
+		goto out;
+	wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_SHORT;
+set_value:
+	wrqu->retry.value = retry;
+	wrqu->retry.disabled = 0;
+out:
+	return rc;
+}
+
+static int wl3501_get_encode(struct net_device *dev,
+			     struct iw_request_info *info,
+			     union iwreq_data *wrqu, char *extra)
+{
+	u8 implemented, restricted, keys[100], len_keys, tocopy;
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = wl3501_get_mib_value(this,
+				      WL3501_MIB_ATTR_PRIV_OPT_IMPLEMENTED,
+				      &implemented, sizeof(implemented));
+	if (rc)
+		goto out;
+	if (!implemented) {
+		wrqu->encoding.flags = IW_ENCODE_DISABLED;
+		goto out;
+	}
+	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_EXCLUDE_UNENCRYPTED,
+				  &restricted, sizeof(restricted));
+	if (rc)
+		goto out;
+	wrqu->encoding.flags = restricted ? IW_ENCODE_RESTRICTED :
+					    IW_ENCODE_OPEN;
+	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_KEY_MAPPINGS_LEN,
+				  &len_keys, sizeof(len_keys));
+	if (rc)
+		goto out;
+	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_KEY_MAPPINGS,
+				  keys, len_keys);
+	if (rc)
+		goto out;
+	tocopy = min_t(u16, len_keys, wrqu->encoding.length);
+	tocopy = min_t(u8, tocopy, 100);
+	wrqu->encoding.length = tocopy;
+	memcpy(extra, keys, tocopy);
+out:
+	return rc;
+}
+
+static int wl3501_get_power(struct net_device *dev,
+			    struct iw_request_info *info,
+			    union iwreq_data *wrqu, char *extra)
+{
+	u8 pwr_state;
+	struct wl3501_card *this = netdev_priv(dev);
+	int rc = wl3501_get_mib_value(this,
+				      WL3501_MIB_ATTR_CURRENT_PWR_STATE,
+				      &pwr_state, sizeof(pwr_state));
+	if (rc)
+		goto out;
+	wrqu->power.disabled = !pwr_state;
+	wrqu->power.flags = IW_POWER_ON;
+out:
+	return rc;
+}
+
+static const iw_handler	wl3501_handler[] = {
+	IW_HANDLER(SIOCGIWNAME, wl3501_get_name),
+	IW_HANDLER(SIOCSIWFREQ, wl3501_set_freq),
+	IW_HANDLER(SIOCGIWFREQ, wl3501_get_freq),
+	IW_HANDLER(SIOCSIWMODE, wl3501_set_mode),
+	IW_HANDLER(SIOCGIWMODE, wl3501_get_mode),
+	IW_HANDLER(SIOCGIWSENS, wl3501_get_sens),
+	IW_HANDLER(SIOCGIWRANGE, wl3501_get_range),
+	IW_HANDLER(SIOCSIWSPY, iw_handler_set_spy),
+	IW_HANDLER(SIOCGIWSPY, iw_handler_get_spy),
+	IW_HANDLER(SIOCSIWTHRSPY, iw_handler_set_thrspy),
+	IW_HANDLER(SIOCGIWTHRSPY, iw_handler_get_thrspy),
+	IW_HANDLER(SIOCSIWAP, wl3501_set_wap),
+	IW_HANDLER(SIOCGIWAP, wl3501_get_wap),
+	IW_HANDLER(SIOCSIWSCAN, wl3501_set_scan),
+	IW_HANDLER(SIOCGIWSCAN, wl3501_get_scan),
+	IW_HANDLER(SIOCSIWESSID, wl3501_set_essid),
+	IW_HANDLER(SIOCGIWESSID, wl3501_get_essid),
+	IW_HANDLER(SIOCSIWNICKN, wl3501_set_nick),
+	IW_HANDLER(SIOCGIWNICKN, wl3501_get_nick),
+	IW_HANDLER(SIOCGIWRATE, wl3501_get_rate),
+	IW_HANDLER(SIOCGIWRTS, wl3501_get_rts_threshold),
+	IW_HANDLER(SIOCGIWFRAG, wl3501_get_frag_threshold),
+	IW_HANDLER(SIOCGIWTXPOW, wl3501_get_txpow),
+	IW_HANDLER(SIOCGIWRETRY, wl3501_get_retry),
+	IW_HANDLER(SIOCGIWENCODE, wl3501_get_encode),
+	IW_HANDLER(SIOCGIWPOWER, wl3501_get_power),
+};
+
+static const struct iw_handler_def wl3501_handler_def = {
+	.num_standard	= ARRAY_SIZE(wl3501_handler),
+	.standard	= (iw_handler *)wl3501_handler,
+	.get_wireless_stats = wl3501_get_wireless_stats,
+};
+
+static const struct net_device_ops wl3501_netdev_ops = {
+	.ndo_open		= wl3501_open,
+	.ndo_stop		= wl3501_close,
+	.ndo_start_xmit		= wl3501_hard_start_xmit,
+	.ndo_tx_timeout		= wl3501_tx_timeout,
+	.ndo_set_mac_address 	= eth_mac_addr,
+	.ndo_validate_addr	= eth_validate_addr,
+};
+
+static int wl3501_probe(struct pcmcia_device *p_dev)
+{
+	struct net_device *dev;
+	struct wl3501_card *this;
+
+	/* The io structure describes IO port mapping */
+	p_dev->resource[0]->end	= 16;
+	p_dev->resource[0]->flags	= IO_DATA_PATH_WIDTH_8;
+
+	/* General socket configuration */
+	p_dev->config_flags	= CONF_ENABLE_IRQ;
+	p_dev->config_index	= 1;
+
+	dev = alloc_etherdev(sizeof(struct wl3501_card));
+	if (!dev)
+		goto out_link;
+
+
+	dev->netdev_ops		= &wl3501_netdev_ops;
+	dev->watchdog_timeo	= 5 * HZ;
+
+	this = netdev_priv(dev);
+	this->wireless_data.spy_data = &this->spy_data;
+	this->p_dev = p_dev;
+	dev->wireless_data	= &this->wireless_data;
+	dev->wireless_handlers	= &wl3501_handler_def;
+	netif_stop_queue(dev);
+	p_dev->priv = dev;
+
+	return wl3501_config(p_dev);
+out_link:
+	return -ENOMEM;
+}
+
+static int wl3501_config(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+	int i = 0, j, ret;
+	struct wl3501_card *this;
+
+	/* Try allocating IO ports.  This tries a few fixed addresses.  If you
+	 * want, you can also read the card's config table to pick addresses --
+	 * see the serial driver for an example. */
+	link->io_lines = 5;
+
+	for (j = 0x280; j < 0x400; j += 0x20) {
+		/* The '^0x300' is so that we probe 0x300-0x3ff first, then
+		 * 0x200-0x2ff, and so on, because this seems safer */
+		link->resource[0]->start = j;
+		link->resource[1]->start = link->resource[0]->start + 0x10;
+		i = pcmcia_request_io(link);
+		if (i == 0)
+			break;
+	}
+	if (i != 0)
+		goto failed;
+
+	/* Now allocate an interrupt line. Note that this does not actually
+	 * assign a handler to the interrupt. */
+
+	ret = pcmcia_request_irq(link, wl3501_interrupt);
+	if (ret)
+		goto failed;
+
+	ret = pcmcia_enable_device(link);
+	if (ret)
+		goto failed;
+
+	dev->irq = link->irq;
+	dev->base_addr = link->resource[0]->start;
+	SET_NETDEV_DEV(dev, &link->dev);
+	if (register_netdev(dev)) {
+		printk(KERN_NOTICE "wl3501_cs: register_netdev() failed\n");
+		goto failed;
+	}
+
+	this = netdev_priv(dev);
+
+	this->base_addr = dev->base_addr;
+
+	if (!wl3501_get_flash_mac_addr(this)) {
+		printk(KERN_WARNING "%s: Can't read MAC addr in flash ROM?\n",
+		       dev->name);
+		unregister_netdev(dev);
+		goto failed;
+	}
+
+	eth_hw_addr_set(dev, this->mac_addr);
+
+	/* print probe information */
+	printk(KERN_INFO "%s: wl3501 @ 0x%3.3x, IRQ %d, "
+	       "MAC addr in flash ROM:%pM\n",
+	       dev->name, this->base_addr, (int)dev->irq,
+	       dev->dev_addr);
+	/*
+	 * Initialize card parameters - added by jss
+	 */
+	this->net_type		= IW_MODE_INFRA;
+	this->bss_cnt		= 0;
+	this->join_sta_bss	= 0;
+	this->adhoc_times	= 0;
+	iw_set_mgmt_info_element(IW_MGMT_INFO_ELEMENT_SSID, &this->essid.el,
+				 "ANY", 3);
+	this->card_name[0]	= '\0';
+	this->firmware_date[0]	= '\0';
+	this->rssi		= 255;
+	this->chan		= iw_default_channel(this->reg_domain);
+	strscpy(this->nick, "Planet WL3501", sizeof(this->nick));
+	spin_lock_init(&this->lock);
+	init_waitqueue_head(&this->wait);
+	netif_start_queue(dev);
+	return 0;
+
+failed:
+	wl3501_release(link);
+	return -ENODEV;
+}
+
+static void wl3501_release(struct pcmcia_device *link)
+{
+	pcmcia_disable_device(link);
+}
+
+static int wl3501_suspend(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+
+	wl3501_pwr_mgmt(netdev_priv(dev), WL3501_SUSPEND);
+	if (link->open)
+		netif_device_detach(dev);
+
+	return 0;
+}
+
+static int wl3501_resume(struct pcmcia_device *link)
+{
+	struct net_device *dev = link->priv;
+
+	wl3501_pwr_mgmt(netdev_priv(dev), WL3501_RESUME);
+	if (link->open) {
+		wl3501_reset(dev);
+		netif_device_attach(dev);
+	}
+
+	return 0;
+}
+
+
+static const struct pcmcia_device_id wl3501_ids[] = {
+	PCMCIA_DEVICE_MANF_CARD(0xd601, 0x0001),
+	PCMCIA_DEVICE_NULL
+};
+MODULE_DEVICE_TABLE(pcmcia, wl3501_ids);
+
+static struct pcmcia_driver wl3501_driver = {
+	.owner		= THIS_MODULE,
+	.name		= "wl3501_cs",
+	.probe		= wl3501_probe,
+	.remove		= wl3501_detach,
+	.id_table	= wl3501_ids,
+	.suspend	= wl3501_suspend,
+	.resume		= wl3501_resume,
+};
+module_pcmcia_driver(wl3501_driver);
+
+MODULE_AUTHOR("Fox Chen <mhchen@golf.ccl.itri.org.tw>, "
+	      "Arnaldo Carvalho de Melo <acme@conectiva.com.br>,"
+	      "Gustavo Niemeyer <niemeyer@conectiva.com>");
+MODULE_DESCRIPTION("Planet wl3501 wireless driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c
deleted file mode 100644
index 1f57a0055bbd..000000000000
--- a/drivers/net/wireless/ray_cs.c
+++ /dev/null
@@ -1,2819 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*=============================================================================
- *
- * A  PCMCIA client driver for the Raylink wireless LAN card.
- * The starting point for this module was the skeleton.c in the
- * PCMCIA 2.9.12 package written by David Hinds, dahinds@users.sourceforge.net
- *
- * Copyright (c) 1998  Corey Thomas (corey@world.std.com)
- *
- * Changes:
- * Arnaldo Carvalho de Melo <acme@conectiva.com.br> - 08/08/2000
- * - reorganize kmallocs in ray_attach, checking all for failure
- *   and releasing the previous allocations if one fails
- *
- * Daniele Bellucci <bellucda@tiscali.it> - 07/10/2003
- * - Audit copy_to_user in ioctl(SIOCGIWESSID)
- *
-=============================================================================*/
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/proc_fs.h>
-#include <linux/ptrace.h>
-#include <linux/seq_file.h>
-#include <linux/string.h>
-#include <linux/timer.h>
-#include <linux/init.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/if_arp.h>
-#include <linux/ioport.h>
-#include <linux/skbuff.h>
-#include <linux/ieee80211.h>
-
-#include <pcmcia/cistpl.h>
-#include <pcmcia/cisreg.h>
-#include <pcmcia/ds.h>
-
-#include <linux/wireless.h>
-#include <net/iw_handler.h>
-
-#include <asm/io.h>
-#include <asm/byteorder.h>
-#include <linux/uaccess.h>
-
-/* Warning : these stuff will slow down the driver... */
-#define WIRELESS_SPY		/* Enable spying addresses */
-/* Definitions we need for spy */
-typedef struct iw_statistics iw_stats;
-typedef u_char mac_addr[ETH_ALEN];	/* Hardware address */
-
-#include "rayctl.h"
-#include "ray_cs.h"
-
-
-/** Prototypes based on PCMCIA skeleton driver *******************************/
-static int ray_config(struct pcmcia_device *link);
-static void ray_release(struct pcmcia_device *link);
-static void ray_detach(struct pcmcia_device *p_dev);
-
-/***** Prototypes indicated by device structure ******************************/
-static int ray_dev_close(struct net_device *dev);
-static int ray_dev_config(struct net_device *dev, struct ifmap *map);
-static struct net_device_stats *ray_get_stats(struct net_device *dev);
-static int ray_dev_init(struct net_device *dev);
-
-static int ray_open(struct net_device *dev);
-static netdev_tx_t ray_dev_start_xmit(struct sk_buff *skb,
-					    struct net_device *dev);
-static void set_multicast_list(struct net_device *dev);
-static void ray_update_multi_list(struct net_device *dev, int all);
-static int translate_frame(ray_dev_t *local, struct tx_msg __iomem *ptx,
-			   unsigned char *data, int len);
-static void ray_build_header(ray_dev_t *local, struct tx_msg __iomem *ptx,
-			     UCHAR msg_type, unsigned char *data);
-static void untranslate(ray_dev_t *local, struct sk_buff *skb, int len);
-static iw_stats *ray_get_wireless_stats(struct net_device *dev);
-static const struct iw_handler_def ray_handler_def;
-
-/***** Prototypes for raylink functions **************************************/
-static void authenticate(ray_dev_t *local);
-static int build_auth_frame(ray_dev_t *local, UCHAR *dest, int auth_type);
-static void authenticate_timeout(struct timer_list *t);
-static int get_free_ccs(ray_dev_t *local);
-static int get_free_tx_ccs(ray_dev_t *local);
-static void init_startup_params(ray_dev_t *local);
-static int parse_addr(char *in_str, UCHAR *out);
-static int ray_hw_xmit(unsigned char *data, int len, struct net_device *dev, UCHAR type);
-static int ray_init(struct net_device *dev);
-static int interrupt_ecf(ray_dev_t *local, int ccs);
-static void ray_reset(struct net_device *dev);
-static void ray_update_parm(struct net_device *dev, UCHAR objid, UCHAR *value, int len);
-static void verify_dl_startup(struct timer_list *t);
-
-/* Prototypes for interrpt time functions **********************************/
-static irqreturn_t ray_interrupt(int reg, void *dev_id);
-static void clear_interrupt(ray_dev_t *local);
-static void rx_deauthenticate(ray_dev_t *local, struct rcs __iomem *prcs,
-			      unsigned int pkt_addr, int rx_len);
-static int copy_from_rx_buff(ray_dev_t *local, UCHAR *dest, int pkt_addr, int len);
-static void ray_rx(struct net_device *dev, ray_dev_t *local, struct rcs __iomem *prcs);
-static void release_frag_chain(ray_dev_t *local, struct rcs __iomem *prcs);
-static void rx_authenticate(ray_dev_t *local, struct rcs __iomem *prcs,
-			    unsigned int pkt_addr, int rx_len);
-static void rx_data(struct net_device *dev, struct rcs __iomem *prcs,
-		    unsigned int pkt_addr, int rx_len);
-static void associate(ray_dev_t *local);
-
-/* Card command functions */
-static int dl_startup_params(struct net_device *dev);
-static void join_net(struct timer_list *t);
-static void start_net(struct timer_list *t);
-
-/*===========================================================================*/
-/* Parameters that can be set with 'insmod' */
-
-/* ADHOC=0, Infrastructure=1 */
-static int net_type = ADHOC;
-
-/* Hop dwell time in Kus (1024 us units defined by 802.11) */
-static int hop_dwell = 128;
-
-/* Beacon period in Kus */
-static int beacon_period = 256;
-
-/* power save mode (0 = off, 1 = save power) */
-static int psm;
-
-/* String for network's Extended Service Set ID. 32 Characters max */
-static char *essid;
-
-/* Default to encapsulation unless translation requested */
-static bool translate = true;
-
-static int country = USA;
-
-static int sniffer;
-
-static int bc;
-
-/* 48 bit physical card address if overriding card's real physical
- * address is required.  Since IEEE 802.11 addresses are 48 bits
- * like ethernet, an int can't be used, so a string is used. To
- * allow use of addresses starting with a decimal digit, the first
- * character must be a letter and will be ignored. This letter is
- * followed by up to 12 hex digits which are the address.  If less
- * than 12 digits are used, the address will be left filled with 0's.
- * Note that bit 0 of the first byte is the broadcast bit, and evil
- * things will happen if it is not 0 in a card address.
- */
-static char *phy_addr = NULL;
-
-static unsigned int ray_mem_speed = 500;
-
-/* WARNING: THIS DRIVER IS NOT CAPABLE OF HANDLING MULTIPLE DEVICES! */
-static struct pcmcia_device *this_device = NULL;
-
-MODULE_AUTHOR("Corey Thomas <corey@world.std.com>");
-MODULE_DESCRIPTION("Raylink/WebGear wireless LAN driver");
-MODULE_LICENSE("GPL");
-
-module_param(net_type, int, 0);
-module_param(hop_dwell, int, 0);
-module_param(beacon_period, int, 0);
-module_param(psm, int, 0);
-module_param(essid, charp, 0);
-module_param(translate, bool, 0);
-module_param(country, int, 0);
-module_param(sniffer, int, 0);
-module_param(bc, int, 0);
-module_param(phy_addr, charp, 0);
-module_param(ray_mem_speed, int, 0);
-
-static const UCHAR b5_default_startup_parms[] = {
-	0, 0,			/* Adhoc station */
-	'L', 'I', 'N', 'U', 'X', 0, 0, 0,	/* 32 char ESSID */
-	0, 0, 0, 0, 0, 0, 0, 0,
-	0, 0, 0, 0, 0, 0, 0, 0,
-	0, 0, 0, 0, 0, 0, 0, 0,
-	1, 0,			/* Active scan, CA Mode */
-	0, 0, 0, 0, 0, 0,	/* No default MAC addr  */
-	0x7f, 0xff,		/* Frag threshold */
-	0x00, 0x80,		/* Hop time 128 Kus */
-	0x01, 0x00,		/* Beacon period 256 Kus */
-	0x01, 0x07, 0xa3,	/* DTIM, retries, ack timeout */
-	0x1d, 0x82, 0x4e,	/* SIFS, DIFS, PIFS */
-	0x7f, 0xff,		/* RTS threshold */
-	0x04, 0xe2, 0x38, 0xA4,	/* scan_dwell, max_scan_dwell */
-	0x05,			/* assoc resp timeout thresh */
-	0x08, 0x02, 0x08,	/* adhoc, infra, super cycle max */
-	0,			/* Promiscuous mode */
-	0x0c, 0x0bd,		/* Unique word */
-	0x32,			/* Slot time */
-	0xff, 0xff,		/* roam-low snr, low snr count */
-	0x05, 0xff,		/* Infra, adhoc missed bcn thresh */
-	0x01, 0x0b, 0x4f,	/* USA, hop pattern, hop pat length */
-/* b4 - b5 differences start here */
-	0x00, 0x3f,		/* CW max */
-	0x00, 0x0f,		/* CW min */
-	0x04, 0x08,		/* Noise gain, limit offset */
-	0x28, 0x28,		/* det rssi, med busy offsets */
-	7,			/* det sync thresh */
-	0, 2, 2,		/* test mode, min, max */
-	0,			/* allow broadcast SSID probe resp */
-	0, 0,			/* privacy must start, can join */
-	2, 0, 0, 0, 0, 0, 0, 0	/* basic rate set */
-};
-
-static const UCHAR b4_default_startup_parms[] = {
-	0, 0,			/* Adhoc station */
-	'L', 'I', 'N', 'U', 'X', 0, 0, 0,	/* 32 char ESSID */
-	0, 0, 0, 0, 0, 0, 0, 0,
-	0, 0, 0, 0, 0, 0, 0, 0,
-	0, 0, 0, 0, 0, 0, 0, 0,
-	1, 0,			/* Active scan, CA Mode */
-	0, 0, 0, 0, 0, 0,	/* No default MAC addr  */
-	0x7f, 0xff,		/* Frag threshold */
-	0x02, 0x00,		/* Hop time */
-	0x00, 0x01,		/* Beacon period */
-	0x01, 0x07, 0xa3,	/* DTIM, retries, ack timeout */
-	0x1d, 0x82, 0xce,	/* SIFS, DIFS, PIFS */
-	0x7f, 0xff,		/* RTS threshold */
-	0xfb, 0x1e, 0xc7, 0x5c,	/* scan_dwell, max_scan_dwell */
-	0x05,			/* assoc resp timeout thresh */
-	0x04, 0x02, 0x4,	/* adhoc, infra, super cycle max */
-	0,			/* Promiscuous mode */
-	0x0c, 0x0bd,		/* Unique word */
-	0x4e,			/* Slot time (TBD seems wrong) */
-	0xff, 0xff,		/* roam-low snr, low snr count */
-	0x05, 0xff,		/* Infra, adhoc missed bcn thresh */
-	0x01, 0x0b, 0x4e,	/* USA, hop pattern, hop pat length */
-/* b4 - b5 differences start here */
-	0x3f, 0x0f,		/* CW max, min */
-	0x04, 0x08,		/* Noise gain, limit offset */
-	0x28, 0x28,		/* det rssi, med busy offsets */
-	7,			/* det sync thresh */
-	0, 2, 2,		/* test mode, min, max */
-	0,			/* rx/tx delay */
-	0, 0, 0, 0, 0, 0,	/* current BSS id */
-	0			/* hop set */
-};
-
-/*===========================================================================*/
-static const u8 eth2_llc[] = { 0xaa, 0xaa, 3, 0, 0, 0 };
-
-static const char hop_pattern_length[] = { 1,
-	USA_HOP_MOD, EUROPE_HOP_MOD,
-	JAPAN_HOP_MOD, KOREA_HOP_MOD,
-	SPAIN_HOP_MOD, FRANCE_HOP_MOD,
-	ISRAEL_HOP_MOD, AUSTRALIA_HOP_MOD,
-	JAPAN_TEST_HOP_MOD
-};
-
-static const char rcsid[] =
-    "Raylink/WebGear wireless LAN - Corey <Thomas corey@world.std.com>";
-
-static const struct net_device_ops ray_netdev_ops = {
-	.ndo_init 		= ray_dev_init,
-	.ndo_open 		= ray_open,
-	.ndo_stop 		= ray_dev_close,
-	.ndo_start_xmit		= ray_dev_start_xmit,
-	.ndo_set_config		= ray_dev_config,
-	.ndo_get_stats		= ray_get_stats,
-	.ndo_set_rx_mode	= set_multicast_list,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-static int ray_probe(struct pcmcia_device *p_dev)
-{
-	ray_dev_t *local;
-	struct net_device *dev;
-
-	dev_dbg(&p_dev->dev, "ray_attach()\n");
-
-	/* Allocate space for private device-specific data */
-	dev = alloc_etherdev(sizeof(ray_dev_t));
-	if (!dev)
-		goto fail_alloc_dev;
-
-	local = netdev_priv(dev);
-	local->finder = p_dev;
-
-	/* The io structure describes IO port mapping. None used here */
-	p_dev->resource[0]->end = 0;
-	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-
-	/* General socket configuration */
-	p_dev->config_flags |= CONF_ENABLE_IRQ;
-	p_dev->config_index = 1;
-
-	p_dev->priv = dev;
-
-	local->finder = p_dev;
-	local->card_status = CARD_INSERTED;
-	local->authentication_state = UNAUTHENTICATED;
-	local->num_multi = 0;
-	dev_dbg(&p_dev->dev, "ray_attach p_dev = %p,  dev = %p,  local = %p, intr = %p\n",
-	      p_dev, dev, local, &ray_interrupt);
-
-	/* Raylink entries in the device structure */
-	dev->netdev_ops = &ray_netdev_ops;
-	dev->wireless_handlers = &ray_handler_def;
-#ifdef WIRELESS_SPY
-	local->wireless_data.spy_data = &local->spy_data;
-	dev->wireless_data = &local->wireless_data;
-#endif /* WIRELESS_SPY */
-
-
-	dev_dbg(&p_dev->dev, "ray_cs ray_attach calling ether_setup.)\n");
-	netif_stop_queue(dev);
-
-	timer_setup(&local->timer, NULL, 0);
-
-	this_device = p_dev;
-	return ray_config(p_dev);
-
-fail_alloc_dev:
-	return -ENOMEM;
-} /* ray_attach */
-
-static void ray_detach(struct pcmcia_device *link)
-{
-	struct net_device *dev;
-	ray_dev_t *local;
-
-	dev_dbg(&link->dev, "ray_detach\n");
-
-	this_device = NULL;
-	dev = link->priv;
-
-	ray_release(link);
-
-	local = netdev_priv(dev);
-	del_timer_sync(&local->timer);
-
-	if (link->priv) {
-		unregister_netdev(dev);
-		free_netdev(dev);
-	}
-	dev_dbg(&link->dev, "ray_cs ray_detach ending\n");
-} /* ray_detach */
-
-#define MAX_TUPLE_SIZE 128
-static int ray_config(struct pcmcia_device *link)
-{
-	int ret = 0;
-	int i;
-	struct net_device *dev = (struct net_device *)link->priv;
-	ray_dev_t *local = netdev_priv(dev);
-
-	dev_dbg(&link->dev, "ray_config\n");
-
-	/* Determine card type and firmware version */
-	printk(KERN_INFO "ray_cs Detected: %s%s%s%s\n",
-	       link->prod_id[0] ? link->prod_id[0] : " ",
-	       link->prod_id[1] ? link->prod_id[1] : " ",
-	       link->prod_id[2] ? link->prod_id[2] : " ",
-	       link->prod_id[3] ? link->prod_id[3] : " ");
-
-	/* Now allocate an interrupt line.  Note that this does not
-	   actually assign a handler to the interrupt.
-	 */
-	ret = pcmcia_request_irq(link, ray_interrupt);
-	if (ret)
-		goto failed;
-	dev->irq = link->irq;
-
-	ret = pcmcia_enable_device(link);
-	if (ret)
-		goto failed;
-
-/*** Set up 32k window for shared memory (transmit and control) ************/
-	link->resource[2]->flags |= WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
-	link->resource[2]->start = 0;
-	link->resource[2]->end = 0x8000;
-	ret = pcmcia_request_window(link, link->resource[2], ray_mem_speed);
-	if (ret)
-		goto failed;
-	ret = pcmcia_map_mem_page(link, link->resource[2], 0);
-	if (ret)
-		goto failed;
-	local->sram = ioremap(link->resource[2]->start,
-			resource_size(link->resource[2]));
-	if (!local->sram)
-		goto failed;
-
-/*** Set up 16k window for shared memory (receive buffer) ***************/
-	link->resource[3]->flags |=
-	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
-	link->resource[3]->start = 0;
-	link->resource[3]->end = 0x4000;
-	ret = pcmcia_request_window(link, link->resource[3], ray_mem_speed);
-	if (ret)
-		goto failed;
-	ret = pcmcia_map_mem_page(link, link->resource[3], 0x8000);
-	if (ret)
-		goto failed;
-	local->rmem = ioremap(link->resource[3]->start,
-			resource_size(link->resource[3]));
-	if (!local->rmem)
-		goto failed;
-
-/*** Set up window for attribute memory ***********************************/
-	link->resource[4]->flags |=
-	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_AM | WIN_ENABLE | WIN_USE_WAIT;
-	link->resource[4]->start = 0;
-	link->resource[4]->end = 0x1000;
-	ret = pcmcia_request_window(link, link->resource[4], ray_mem_speed);
-	if (ret)
-		goto failed;
-	ret = pcmcia_map_mem_page(link, link->resource[4], 0);
-	if (ret)
-		goto failed;
-	local->amem = ioremap(link->resource[4]->start,
-			resource_size(link->resource[4]));
-	if (!local->amem)
-		goto failed;
-
-	dev_dbg(&link->dev, "ray_config sram=%p\n", local->sram);
-	dev_dbg(&link->dev, "ray_config rmem=%p\n", local->rmem);
-	dev_dbg(&link->dev, "ray_config amem=%p\n", local->amem);
-	if (ray_init(dev) < 0) {
-		ray_release(link);
-		return -ENODEV;
-	}
-
-	SET_NETDEV_DEV(dev, &link->dev);
-	i = register_netdev(dev);
-	if (i != 0) {
-		printk("ray_config register_netdev() failed\n");
-		ray_release(link);
-		return i;
-	}
-
-	printk(KERN_INFO "%s: RayLink, irq %d, hw_addr %pM\n",
-	       dev->name, dev->irq, dev->dev_addr);
-
-	return 0;
-
-failed:
-	ray_release(link);
-	return -ENODEV;
-} /* ray_config */
-
-static inline struct ccs __iomem *ccs_base(ray_dev_t *dev)
-{
-	return dev->sram + CCS_BASE;
-}
-
-static inline struct rcs __iomem *rcs_base(ray_dev_t *dev)
-{
-	/*
-	 * This looks nonsensical, since there is a separate
-	 * RCS_BASE. But the difference between a "struct rcs"
-	 * and a "struct ccs" ends up being in the _index_ off
-	 * the base, so the base pointer is the same for both
-	 * ccs/rcs.
-	 */
-	return dev->sram + CCS_BASE;
-}
-
-/*===========================================================================*/
-static int ray_init(struct net_device *dev)
-{
-	int i;
-	struct ccs __iomem *pccs;
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	dev_dbg(&link->dev, "ray_init(0x%p)\n", dev);
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_init - device not present\n");
-		return -1;
-	}
-
-	local->net_type = net_type;
-	local->sta_type = TYPE_STA;
-
-	/* Copy the startup results to local memory */
-	memcpy_fromio(&local->startup_res, local->sram + ECF_TO_HOST_BASE,
-		      sizeof(struct startup_res_6));
-
-	/* Check Power up test status and get mac address from card */
-	if (local->startup_res.startup_word != 0x80) {
-		printk(KERN_INFO "ray_init ERROR card status = %2x\n",
-		       local->startup_res.startup_word);
-		local->card_status = CARD_INIT_ERROR;
-		return -1;
-	}
-
-	local->fw_ver = local->startup_res.firmware_version[0];
-	local->fw_bld = local->startup_res.firmware_version[1];
-	local->fw_var = local->startup_res.firmware_version[2];
-	dev_dbg(&link->dev, "ray_init firmware version %d.%d\n", local->fw_ver,
-	      local->fw_bld);
-
-	local->tib_length = 0x20;
-	if ((local->fw_ver == 5) && (local->fw_bld >= 30))
-		local->tib_length = local->startup_res.tib_length;
-	dev_dbg(&link->dev, "ray_init tib_length = 0x%02x\n", local->tib_length);
-	/* Initialize CCS's to buffer free state */
-	pccs = ccs_base(local);
-	for (i = 0; i < NUMBER_OF_CCS; i++) {
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-	}
-	init_startup_params(local);
-
-	/* copy mac address to startup parameters */
-	if (!parse_addr(phy_addr, local->sparm.b4.a_mac_addr)) {
-		memcpy(&local->sparm.b4.a_mac_addr,
-		       &local->startup_res.station_addr, ADDRLEN);
-	}
-
-	clear_interrupt(local);	/* Clear any interrupt from the card */
-	local->card_status = CARD_AWAITING_PARAM;
-	dev_dbg(&link->dev, "ray_init ending\n");
-	return 0;
-} /* ray_init */
-
-/*===========================================================================*/
-/* Download startup parameters to the card and command it to read them       */
-static int dl_startup_params(struct net_device *dev)
-{
-	int ccsindex;
-	ray_dev_t *local = netdev_priv(dev);
-	struct ccs __iomem *pccs;
-	struct pcmcia_device *link = local->finder;
-
-	dev_dbg(&link->dev, "dl_startup_params entered\n");
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs dl_startup_params - device not present\n");
-		return -1;
-	}
-
-	/* Copy parameters to host to ECF area */
-	if (local->fw_ver == 0x55)
-		memcpy_toio(local->sram + HOST_TO_ECF_BASE, &local->sparm.b4,
-			    sizeof(struct b4_startup_params));
-	else
-		memcpy_toio(local->sram + HOST_TO_ECF_BASE, &local->sparm.b5,
-			    sizeof(struct b5_startup_params));
-
-	/* Fill in the CCS fields for the ECF */
-	if ((ccsindex = get_free_ccs(local)) < 0)
-		return -1;
-	local->dl_param_ccs = ccsindex;
-	pccs = ccs_base(local) + ccsindex;
-	writeb(CCS_DOWNLOAD_STARTUP_PARAMS, &pccs->cmd);
-	dev_dbg(&link->dev, "dl_startup_params start ccsindex = %d\n",
-	      local->dl_param_ccs);
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		printk(KERN_INFO "ray dl_startup_params failed - "
-		       "ECF not ready for intr\n");
-		local->card_status = CARD_DL_PARAM_ERROR;
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-		return -2;
-	}
-	local->card_status = CARD_DL_PARAM;
-	/* Start kernel timer to wait for dl startup to complete. */
-	local->timer.expires = jiffies + HZ / 2;
-	local->timer.function = verify_dl_startup;
-	add_timer(&local->timer);
-	dev_dbg(&link->dev,
-	      "ray_cs dl_startup_params started timer for verify_dl_startup\n");
-	return 0;
-} /* dl_startup_params */
-
-/*===========================================================================*/
-static void init_startup_params(ray_dev_t *local)
-{
-	int i;
-
-	if (country > JAPAN_TEST)
-		country = USA;
-	else if (country < USA)
-		country = USA;
-	/* structure for hop time and beacon period is defined here using
-	 * New 802.11D6.1 format.  Card firmware is still using old format
-	 * until version 6.
-	 *    Before                    After
-	 *    a_hop_time ms byte        a_hop_time ms byte
-	 *    a_hop_time 2s byte        a_hop_time ls byte
-	 *    a_hop_time ls byte        a_beacon_period ms byte
-	 *    a_beacon_period           a_beacon_period ls byte
-	 *
-	 *    a_hop_time = uS           a_hop_time = KuS
-	 *    a_beacon_period = hops    a_beacon_period = KuS
-	 *//* 64ms = 010000 */
-	if (local->fw_ver == 0x55) {
-		memcpy(&local->sparm.b4, b4_default_startup_parms,
-		       sizeof(struct b4_startup_params));
-		/* Translate sane kus input values to old build 4/5 format */
-		/* i = hop time in uS truncated to 3 bytes */
-		i = (hop_dwell * 1024) & 0xffffff;
-		local->sparm.b4.a_hop_time[0] = (i >> 16) & 0xff;
-		local->sparm.b4.a_hop_time[1] = (i >> 8) & 0xff;
-		local->sparm.b4.a_beacon_period[0] = 0;
-		local->sparm.b4.a_beacon_period[1] =
-		    ((beacon_period / hop_dwell) - 1) & 0xff;
-		local->sparm.b4.a_curr_country_code = country;
-		local->sparm.b4.a_hop_pattern_length =
-		    hop_pattern_length[(int)country] - 1;
-		if (bc) {
-			local->sparm.b4.a_ack_timeout = 0x50;
-			local->sparm.b4.a_sifs = 0x3f;
-		}
-	} else { /* Version 5 uses real kus values */
-		memcpy((UCHAR *) &local->sparm.b5, b5_default_startup_parms,
-		       sizeof(struct b5_startup_params));
-
-		local->sparm.b5.a_hop_time[0] = (hop_dwell >> 8) & 0xff;
-		local->sparm.b5.a_hop_time[1] = hop_dwell & 0xff;
-		local->sparm.b5.a_beacon_period[0] =
-		    (beacon_period >> 8) & 0xff;
-		local->sparm.b5.a_beacon_period[1] = beacon_period & 0xff;
-		if (psm)
-			local->sparm.b5.a_power_mgt_state = 1;
-		local->sparm.b5.a_curr_country_code = country;
-		local->sparm.b5.a_hop_pattern_length =
-		    hop_pattern_length[(int)country];
-	}
-
-	local->sparm.b4.a_network_type = net_type & 0x01;
-	local->sparm.b4.a_acting_as_ap_status = TYPE_STA;
-
-	if (essid != NULL)
-		strncpy(local->sparm.b4.a_current_ess_id, essid, ESSID_SIZE);
-} /* init_startup_params */
-
-/*===========================================================================*/
-static void verify_dl_startup(struct timer_list *t)
-{
-	ray_dev_t *local = from_timer(local, t, timer);
-	struct ccs __iomem *pccs = ccs_base(local) + local->dl_param_ccs;
-	UCHAR status;
-	struct pcmcia_device *link = local->finder;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs verify_dl_startup - device not present\n");
-		return;
-	}
-#if 0
-	{
-		int i;
-		printk(KERN_DEBUG
-		       "verify_dl_startup parameters sent via ccs %d:\n",
-		       local->dl_param_ccs);
-		for (i = 0; i < sizeof(struct b5_startup_params); i++) {
-			printk(" %2x",
-			       (unsigned int)readb(local->sram +
-						   HOST_TO_ECF_BASE + i));
-		}
-		printk("\n");
-	}
-#endif
-
-	status = readb(&pccs->buffer_status);
-	if (status != CCS_BUFFER_FREE) {
-		printk(KERN_INFO
-		       "Download startup params failed.  Status = %d\n",
-		       status);
-		local->card_status = CARD_DL_PARAM_ERROR;
-		return;
-	}
-	if (local->sparm.b4.a_network_type == ADHOC)
-		start_net(&local->timer);
-	else
-		join_net(&local->timer);
-} /* end verify_dl_startup */
-
-/*===========================================================================*/
-/* Command card to start a network */
-static void start_net(struct timer_list *t)
-{
-	ray_dev_t *local = from_timer(local, t, timer);
-	struct ccs __iomem *pccs;
-	int ccsindex;
-	struct pcmcia_device *link = local->finder;
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs start_net - device not present\n");
-		return;
-	}
-	/* Fill in the CCS fields for the ECF */
-	if ((ccsindex = get_free_ccs(local)) < 0)
-		return;
-	pccs = ccs_base(local) + ccsindex;
-	writeb(CCS_START_NETWORK, &pccs->cmd);
-	writeb(0, &pccs->var.start_network.update_param);
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		dev_dbg(&link->dev, "ray start net failed - card not ready for intr\n");
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-		return;
-	}
-	local->card_status = CARD_DOING_ACQ;
-} /* end start_net */
-
-/*===========================================================================*/
-/* Command card to join a network */
-static void join_net(struct timer_list *t)
-{
-	ray_dev_t *local = from_timer(local, t, timer);
-
-	struct ccs __iomem *pccs;
-	int ccsindex;
-	struct pcmcia_device *link = local->finder;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs join_net - device not present\n");
-		return;
-	}
-	/* Fill in the CCS fields for the ECF */
-	if ((ccsindex = get_free_ccs(local)) < 0)
-		return;
-	pccs = ccs_base(local) + ccsindex;
-	writeb(CCS_JOIN_NETWORK, &pccs->cmd);
-	writeb(0, &pccs->var.join_network.update_param);
-	writeb(0, &pccs->var.join_network.net_initiated);
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		dev_dbg(&link->dev, "ray join net failed - card not ready for intr\n");
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-		return;
-	}
-	local->card_status = CARD_DOING_ACQ;
-}
-
-
-static void ray_release(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-	ray_dev_t *local = netdev_priv(dev);
-
-	dev_dbg(&link->dev, "ray_release\n");
-
-	del_timer(&local->timer);
-
-	iounmap(local->sram);
-	iounmap(local->rmem);
-	iounmap(local->amem);
-	pcmcia_disable_device(link);
-
-	dev_dbg(&link->dev, "ray_release ending\n");
-}
-
-static int ray_suspend(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-
-	if (link->open)
-		netif_device_detach(dev);
-
-	return 0;
-}
-
-static int ray_resume(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-
-	if (link->open) {
-		ray_reset(dev);
-		netif_device_attach(dev);
-	}
-
-	return 0;
-}
-
-/*===========================================================================*/
-static int ray_dev_init(struct net_device *dev)
-{
-#ifdef RAY_IMMEDIATE_INIT
-	int i;
-#endif /* RAY_IMMEDIATE_INIT */
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-
-	dev_dbg(&link->dev, "ray_dev_init(dev=%p)\n", dev);
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_dev_init - device not present\n");
-		return -1;
-	}
-#ifdef RAY_IMMEDIATE_INIT
-	/* Download startup parameters */
-	if ((i = dl_startup_params(dev)) < 0) {
-		printk(KERN_INFO "ray_dev_init dl_startup_params failed - "
-		       "returns 0x%x\n", i);
-		return -1;
-	}
-#else /* RAY_IMMEDIATE_INIT */
-	/* Postpone the card init so that we can still configure the card,
-	 * for example using the Wireless Extensions. The init will happen
-	 * in ray_open() - Jean II */
-	dev_dbg(&link->dev,
-	      "ray_dev_init: postponing card init to ray_open() ; Status = %d\n",
-	      local->card_status);
-#endif /* RAY_IMMEDIATE_INIT */
-
-	/* copy mac and broadcast addresses to linux device */
-	eth_hw_addr_set(dev, local->sparm.b4.a_mac_addr);
-	eth_broadcast_addr(dev->broadcast);
-
-	dev_dbg(&link->dev, "ray_dev_init ending\n");
-	return 0;
-}
-
-/*===========================================================================*/
-static int ray_dev_config(struct net_device *dev, struct ifmap *map)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	/* Dummy routine to satisfy device structure */
-	dev_dbg(&link->dev, "ray_dev_config(dev=%p,ifmap=%p)\n", dev, map);
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_dev_config - device not present\n");
-		return -1;
-	}
-
-	return 0;
-}
-
-/*===========================================================================*/
-static netdev_tx_t ray_dev_start_xmit(struct sk_buff *skb,
-					    struct net_device *dev)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	short length = skb->len;
-
-	if (!pcmcia_dev_present(link)) {
-		dev_dbg(&link->dev, "ray_dev_start_xmit - device not present\n");
-		dev_kfree_skb(skb);
-		return NETDEV_TX_OK;
-	}
-
-	dev_dbg(&link->dev, "ray_dev_start_xmit(skb=%p, dev=%p)\n", skb, dev);
-	if (local->authentication_state == NEED_TO_AUTH) {
-		dev_dbg(&link->dev, "ray_cs Sending authentication request.\n");
-		if (!build_auth_frame(local, local->auth_id, OPEN_AUTH_REQUEST)) {
-			local->authentication_state = AUTHENTICATED;
-			netif_stop_queue(dev);
-			return NETDEV_TX_BUSY;
-		}
-	}
-
-	if (length < ETH_ZLEN) {
-		if (skb_padto(skb, ETH_ZLEN))
-			return NETDEV_TX_OK;
-		length = ETH_ZLEN;
-	}
-	switch (ray_hw_xmit(skb->data, length, dev, DATA_TYPE)) {
-	case XMIT_NO_CCS:
-	case XMIT_NEED_AUTH:
-		netif_stop_queue(dev);
-		return NETDEV_TX_BUSY;
-	case XMIT_NO_INTR:
-	case XMIT_MSG_BAD:
-	case XMIT_OK:
-	default:
-		dev_kfree_skb(skb);
-	}
-
-	return NETDEV_TX_OK;
-} /* ray_dev_start_xmit */
-
-/*===========================================================================*/
-static int ray_hw_xmit(unsigned char *data, int len, struct net_device *dev,
-		       UCHAR msg_type)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct ccs __iomem *pccs;
-	int ccsindex;
-	int offset;
-	struct tx_msg __iomem *ptx;	/* Address of xmit buffer in PC space */
-	short int addr;		/* Address of xmit buffer in card space */
-
-	pr_debug("ray_hw_xmit(data=%p, len=%d, dev=%p)\n", data, len, dev);
-	if (len + TX_HEADER_LENGTH > TX_BUF_SIZE) {
-		printk(KERN_INFO "ray_hw_xmit packet too large: %d bytes\n",
-		       len);
-		return XMIT_MSG_BAD;
-	}
-	switch (ccsindex = get_free_tx_ccs(local)) {
-	case ECCSBUSY:
-		pr_debug("ray_hw_xmit tx_ccs table busy\n");
-		fallthrough;
-	case ECCSFULL:
-		pr_debug("ray_hw_xmit No free tx ccs\n");
-		fallthrough;
-	case ECARDGONE:
-		netif_stop_queue(dev);
-		return XMIT_NO_CCS;
-	default:
-		break;
-	}
-	addr = TX_BUF_BASE + (ccsindex << 11);
-
-	if (msg_type == DATA_TYPE) {
-		local->stats.tx_bytes += len;
-		local->stats.tx_packets++;
-	}
-
-	ptx = local->sram + addr;
-
-	ray_build_header(local, ptx, msg_type, data);
-	if (translate) {
-		offset = translate_frame(local, ptx, data, len);
-	} else { /* Encapsulate frame */
-		/* TBD TIB length will move address of ptx->var */
-		memcpy_toio(&ptx->var, data, len);
-		offset = 0;
-	}
-
-	/* fill in the CCS */
-	pccs = ccs_base(local) + ccsindex;
-	len += TX_HEADER_LENGTH + offset;
-	writeb(CCS_TX_REQUEST, &pccs->cmd);
-	writeb(addr >> 8, &pccs->var.tx_request.tx_data_ptr[0]);
-	writeb(local->tib_length, &pccs->var.tx_request.tx_data_ptr[1]);
-	writeb(len >> 8, &pccs->var.tx_request.tx_data_length[0]);
-	writeb(len & 0xff, &pccs->var.tx_request.tx_data_length[1]);
-/* TBD still need psm_cam? */
-	writeb(PSM_CAM, &pccs->var.tx_request.pow_sav_mode);
-	writeb(local->net_default_tx_rate, &pccs->var.tx_request.tx_rate);
-	writeb(0, &pccs->var.tx_request.antenna);
-	pr_debug("ray_hw_xmit default_tx_rate = 0x%x\n",
-	      local->net_default_tx_rate);
-
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		pr_debug("ray_hw_xmit failed - ECF not ready for intr\n");
-/* TBD very inefficient to copy packet to buffer, and then not
-   send it, but the alternative is to queue the messages and that
-   won't be done for a while.  Maybe set tbusy until a CCS is free?
-*/
-		writeb(CCS_BUFFER_FREE, &pccs->buffer_status);
-		return XMIT_NO_INTR;
-	}
-	return XMIT_OK;
-} /* end ray_hw_xmit */
-
-/*===========================================================================*/
-static int translate_frame(ray_dev_t *local, struct tx_msg __iomem *ptx,
-			   unsigned char *data, int len)
-{
-	__be16 proto = ((struct ethhdr *)data)->h_proto;
-	if (ntohs(proto) >= ETH_P_802_3_MIN) { /* DIX II ethernet frame */
-		pr_debug("ray_cs translate_frame DIX II\n");
-		/* Copy LLC header to card buffer */
-		memcpy_toio(&ptx->var, eth2_llc, sizeof(eth2_llc));
-		memcpy_toio(((void __iomem *)&ptx->var) + sizeof(eth2_llc),
-			    (UCHAR *) &proto, 2);
-		if (proto == htons(ETH_P_AARP) || proto == htons(ETH_P_IPX)) {
-			/* This is the selective translation table, only 2 entries */
-			writeb(0xf8,
-			       &((struct snaphdr_t __iomem *)ptx->var)->org[2]);
-		}
-		/* Copy body of ethernet packet without ethernet header */
-		memcpy_toio((void __iomem *)&ptx->var +
-			    sizeof(struct snaphdr_t), data + ETH_HLEN,
-			    len - ETH_HLEN);
-		return (int)sizeof(struct snaphdr_t) - ETH_HLEN;
-	} else { /* already  802 type, and proto is length */
-		pr_debug("ray_cs translate_frame 802\n");
-		if (proto == htons(0xffff)) { /* evil netware IPX 802.3 without LLC */
-			pr_debug("ray_cs translate_frame evil IPX\n");
-			memcpy_toio(&ptx->var, data + ETH_HLEN, len - ETH_HLEN);
-			return 0 - ETH_HLEN;
-		}
-		memcpy_toio(&ptx->var, data + ETH_HLEN, len - ETH_HLEN);
-		return 0 - ETH_HLEN;
-	}
-	/* TBD do other frame types */
-} /* end translate_frame */
-
-/*===========================================================================*/
-static void ray_build_header(ray_dev_t *local, struct tx_msg __iomem *ptx,
-			     UCHAR msg_type, unsigned char *data)
-{
-	writeb(PROTOCOL_VER | msg_type, &ptx->mac.frame_ctl_1);
-/*** IEEE 802.11 Address field assignments *************
-		TODS	FROMDS	addr_1		addr_2		addr_3	addr_4
-Adhoc		0	0	dest		src (terminal)	BSSID	N/A
-AP to Terminal	0	1	dest		AP(BSSID)	source	N/A
-Terminal to AP	1	0	AP(BSSID)	src (terminal)	dest	N/A
-AP to AP	1	1	dest AP		src AP		dest	source
-*******************************************************/
-	if (local->net_type == ADHOC) {
-		writeb(0, &ptx->mac.frame_ctl_2);
-		memcpy_toio(ptx->mac.addr_1, ((struct ethhdr *)data)->h_dest,
-			    ADDRLEN);
-		memcpy_toio(ptx->mac.addr_2, ((struct ethhdr *)data)->h_source,
-			    ADDRLEN);
-		memcpy_toio(ptx->mac.addr_3, local->bss_id, ADDRLEN);
-	} else { /* infrastructure */
-
-		if (local->sparm.b4.a_acting_as_ap_status) {
-			writeb(FC2_FROM_DS, &ptx->mac.frame_ctl_2);
-			memcpy_toio(ptx->mac.addr_1,
-				    ((struct ethhdr *)data)->h_dest, ADDRLEN);
-			memcpy_toio(ptx->mac.addr_2, local->bss_id, 6);
-			memcpy_toio(ptx->mac.addr_3,
-				    ((struct ethhdr *)data)->h_source, ADDRLEN);
-		} else { /* Terminal */
-
-			writeb(FC2_TO_DS, &ptx->mac.frame_ctl_2);
-			memcpy_toio(ptx->mac.addr_1, local->bss_id, ADDRLEN);
-			memcpy_toio(ptx->mac.addr_2,
-				    ((struct ethhdr *)data)->h_source, ADDRLEN);
-			memcpy_toio(ptx->mac.addr_3,
-				    ((struct ethhdr *)data)->h_dest, ADDRLEN);
-		}
-	}
-} /* end encapsulate_frame */
-
-/*====================================================================*/
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get protocol name
- */
-static int ray_get_name(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	strcpy(wrqu->name, "IEEE 802.11-FH");
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : set frequency
- */
-static int ray_set_freq(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	int err = -EINPROGRESS;	/* Call commit handler */
-
-	/* Reject if card is already initialised */
-	if (local->card_status != CARD_AWAITING_PARAM)
-		return -EBUSY;
-
-	/* Setting by channel number */
-	if ((wrqu->freq.m > USA_HOP_MOD) || (wrqu->freq.e > 0))
-		err = -EOPNOTSUPP;
-	else
-		local->sparm.b5.a_hop_pattern = wrqu->freq.m;
-
-	return err;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get frequency
- */
-static int ray_get_freq(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	wrqu->freq.m = local->sparm.b5.a_hop_pattern;
-	wrqu->freq.e = 0;
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : set ESSID
- */
-static int ray_set_essid(struct net_device *dev, struct iw_request_info *info,
-			 union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	/* Reject if card is already initialised */
-	if (local->card_status != CARD_AWAITING_PARAM)
-		return -EBUSY;
-
-	/* Check if we asked for `any' */
-	if (wrqu->essid.flags == 0)
-		/* Corey : can you do that ? */
-		return -EOPNOTSUPP;
-
-	/* Check the size of the string */
-	if (wrqu->essid.length > IW_ESSID_MAX_SIZE)
-		return -E2BIG;
-
-	/* Set the ESSID in the card */
-	memset(local->sparm.b5.a_current_ess_id, 0, IW_ESSID_MAX_SIZE);
-	memcpy(local->sparm.b5.a_current_ess_id, extra, wrqu->essid.length);
-
-	return -EINPROGRESS;	/* Call commit handler */
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get ESSID
- */
-static int ray_get_essid(struct net_device *dev, struct iw_request_info *info,
-			 union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	UCHAR tmp[IW_ESSID_MAX_SIZE + 1];
-
-	/* Get the essid that was set */
-	memcpy(extra, local->sparm.b5.a_current_ess_id, IW_ESSID_MAX_SIZE);
-	memcpy(tmp, local->sparm.b5.a_current_ess_id, IW_ESSID_MAX_SIZE);
-	tmp[IW_ESSID_MAX_SIZE] = '\0';
-
-	/* Push it out ! */
-	wrqu->essid.length = strlen(tmp);
-	wrqu->essid.flags = 1;	/* active */
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get AP address
- */
-static int ray_get_wap(struct net_device *dev, struct iw_request_info *info,
-		       union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	memcpy(wrqu->ap_addr.sa_data, local->bss_id, ETH_ALEN);
-	wrqu->ap_addr.sa_family = ARPHRD_ETHER;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : set Bit-Rate
- */
-static int ray_set_rate(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	/* Reject if card is already initialised */
-	if (local->card_status != CARD_AWAITING_PARAM)
-		return -EBUSY;
-
-	/* Check if rate is in range */
-	if ((wrqu->bitrate.value != 1000000) && (wrqu->bitrate.value != 2000000))
-		return -EINVAL;
-
-	/* Hack for 1.5 Mb/s instead of 2 Mb/s */
-	if ((local->fw_ver == 0x55) &&	/* Please check */
-	    (wrqu->bitrate.value == 2000000))
-		local->net_default_tx_rate = 3;
-	else
-		local->net_default_tx_rate = wrqu->bitrate.value / 500000;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get Bit-Rate
- */
-static int ray_get_rate(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	if (local->net_default_tx_rate == 3)
-		wrqu->bitrate.value = 2000000;	/* Hum... */
-	else
-		wrqu->bitrate.value = local->net_default_tx_rate * 500000;
-	wrqu->bitrate.fixed = 0;	/* We are in auto mode */
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : set RTS threshold
- */
-static int ray_set_rts(struct net_device *dev, struct iw_request_info *info,
-		       union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	int rthr = wrqu->rts.value;
-
-	/* Reject if card is already initialised */
-	if (local->card_status != CARD_AWAITING_PARAM)
-		return -EBUSY;
-
-	/* if(wrq->u.rts.fixed == 0) we should complain */
-	if (wrqu->rts.disabled)
-		rthr = 32767;
-	else {
-		if ((rthr < 0) || (rthr > 2347))   /* What's the max packet size ??? */
-			return -EINVAL;
-	}
-	local->sparm.b5.a_rts_threshold[0] = (rthr >> 8) & 0xFF;
-	local->sparm.b5.a_rts_threshold[1] = rthr & 0xFF;
-
-	return -EINPROGRESS;	/* Call commit handler */
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get RTS threshold
- */
-static int ray_get_rts(struct net_device *dev, struct iw_request_info *info,
-		       union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	wrqu->rts.value = (local->sparm.b5.a_rts_threshold[0] << 8)
-	    + local->sparm.b5.a_rts_threshold[1];
-	wrqu->rts.disabled = (wrqu->rts.value == 32767);
-	wrqu->rts.fixed = 1;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : set Fragmentation threshold
- */
-static int ray_set_frag(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	int fthr = wrqu->frag.value;
-
-	/* Reject if card is already initialised */
-	if (local->card_status != CARD_AWAITING_PARAM)
-		return -EBUSY;
-
-	/* if(wrq->u.frag.fixed == 0) should complain */
-	if (wrqu->frag.disabled)
-		fthr = 32767;
-	else {
-		if ((fthr < 256) || (fthr > 2347))	/* To check out ! */
-			return -EINVAL;
-	}
-	local->sparm.b5.a_frag_threshold[0] = (fthr >> 8) & 0xFF;
-	local->sparm.b5.a_frag_threshold[1] = fthr & 0xFF;
-
-	return -EINPROGRESS;	/* Call commit handler */
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get Fragmentation threshold
- */
-static int ray_get_frag(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	wrqu->frag.value = (local->sparm.b5.a_frag_threshold[0] << 8)
-	    + local->sparm.b5.a_frag_threshold[1];
-	wrqu->frag.disabled = (wrqu->frag.value == 32767);
-	wrqu->frag.fixed = 1;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : set Mode of Operation
- */
-static int ray_set_mode(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	int err = -EINPROGRESS;	/* Call commit handler */
-	char card_mode = 1;
-
-	/* Reject if card is already initialised */
-	if (local->card_status != CARD_AWAITING_PARAM)
-		return -EBUSY;
-
-	switch (wrqu->mode) {
-	case IW_MODE_ADHOC:
-		card_mode = 0;
-		fallthrough;
-	case IW_MODE_INFRA:
-		local->sparm.b5.a_network_type = card_mode;
-		break;
-	default:
-		err = -EINVAL;
-	}
-
-	return err;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get Mode of Operation
- */
-static int ray_get_mode(struct net_device *dev, struct iw_request_info *info,
-			union iwreq_data *wrqu, char *extra)
-{
-	ray_dev_t *local = netdev_priv(dev);
-
-	if (local->sparm.b5.a_network_type)
-		wrqu->mode = IW_MODE_INFRA;
-	else
-		wrqu->mode = IW_MODE_ADHOC;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Handler : get range info
- */
-static int ray_get_range(struct net_device *dev, struct iw_request_info *info,
-			 union iwreq_data *wrqu, char *extra)
-{
-	struct iw_range *range = (struct iw_range *)extra;
-
-	memset(range, 0, sizeof(struct iw_range));
-
-	/* Set the length (very important for backward compatibility) */
-	wrqu->data.length = sizeof(struct iw_range);
-
-	/* Set the Wireless Extension versions */
-	range->we_version_compiled = WIRELESS_EXT;
-	range->we_version_source = 9;
-
-	/* Set information in the range struct */
-	range->throughput = 1.1 * 1000 * 1000;	/* Put the right number here */
-	range->num_channels = hop_pattern_length[(int)country];
-	range->num_frequency = 0;
-	range->max_qual.qual = 0;
-	range->max_qual.level = 255;	/* What's the correct value ? */
-	range->max_qual.noise = 255;	/* Idem */
-	range->num_bitrates = 2;
-	range->bitrate[0] = 1000000;	/* 1 Mb/s */
-	range->bitrate[1] = 2000000;	/* 2 Mb/s */
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Private Handler : set framing mode
- */
-static int ray_set_framing(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	translate = !!*(extra);	/* Set framing mode */
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Private Handler : get framing mode
- */
-static int ray_get_framing(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	*(extra) = translate;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Wireless Private Handler : get country
- */
-static int ray_get_country(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	*(extra) = country;
-
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Commit handler : called after a bunch of SET operations
- */
-static int ray_commit(struct net_device *dev, struct iw_request_info *info,
-		      union iwreq_data *wrqu, char *extra)
-{
-	return 0;
-}
-
-/*------------------------------------------------------------------*/
-/*
- * Stats handler : return Wireless Stats
- */
-static iw_stats *ray_get_wireless_stats(struct net_device *dev)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	struct status __iomem *p = local->sram + STATUS_BASE;
-
-	local->wstats.status = local->card_status;
-#ifdef WIRELESS_SPY
-	if ((local->spy_data.spy_number > 0)
-	    && (local->sparm.b5.a_network_type == 0)) {
-		/* Get it from the first node in spy list */
-		local->wstats.qual.qual = local->spy_data.spy_stat[0].qual;
-		local->wstats.qual.level = local->spy_data.spy_stat[0].level;
-		local->wstats.qual.noise = local->spy_data.spy_stat[0].noise;
-		local->wstats.qual.updated =
-		    local->spy_data.spy_stat[0].updated;
-	}
-#endif /* WIRELESS_SPY */
-
-	if (pcmcia_dev_present(link)) {
-		local->wstats.qual.noise = readb(&p->rxnoise);
-		local->wstats.qual.updated |= 4;
-	}
-
-	return &local->wstats;
-} /* end ray_get_wireless_stats */
-
-/*------------------------------------------------------------------*/
-/*
- * Structures to export the Wireless Handlers
- */
-
-static const iw_handler ray_handler[] = {
-	IW_HANDLER(SIOCSIWCOMMIT, ray_commit),
-	IW_HANDLER(SIOCGIWNAME, ray_get_name),
-	IW_HANDLER(SIOCSIWFREQ, ray_set_freq),
-	IW_HANDLER(SIOCGIWFREQ, ray_get_freq),
-	IW_HANDLER(SIOCSIWMODE, ray_set_mode),
-	IW_HANDLER(SIOCGIWMODE, ray_get_mode),
-	IW_HANDLER(SIOCGIWRANGE, ray_get_range),
-#ifdef WIRELESS_SPY
-	IW_HANDLER(SIOCSIWSPY, iw_handler_set_spy),
-	IW_HANDLER(SIOCGIWSPY, iw_handler_get_spy),
-	IW_HANDLER(SIOCSIWTHRSPY, iw_handler_set_thrspy),
-	IW_HANDLER(SIOCGIWTHRSPY, iw_handler_get_thrspy),
-#endif /* WIRELESS_SPY */
-	IW_HANDLER(SIOCGIWAP, ray_get_wap),
-	IW_HANDLER(SIOCSIWESSID, ray_set_essid),
-	IW_HANDLER(SIOCGIWESSID, ray_get_essid),
-	IW_HANDLER(SIOCSIWRATE, ray_set_rate),
-	IW_HANDLER(SIOCGIWRATE, ray_get_rate),
-	IW_HANDLER(SIOCSIWRTS, ray_set_rts),
-	IW_HANDLER(SIOCGIWRTS, ray_get_rts),
-	IW_HANDLER(SIOCSIWFRAG, ray_set_frag),
-	IW_HANDLER(SIOCGIWFRAG, ray_get_frag),
-};
-
-#define SIOCSIPFRAMING	SIOCIWFIRSTPRIV	/* Set framing mode */
-#define SIOCGIPFRAMING	SIOCIWFIRSTPRIV + 1	/* Get framing mode */
-#define SIOCGIPCOUNTRY	SIOCIWFIRSTPRIV + 3	/* Get country code */
-
-static const iw_handler ray_private_handler[] = {
-	[0] = ray_set_framing,
-	[1] = ray_get_framing,
-	[3] = ray_get_country,
-};
-
-static const struct iw_priv_args ray_private_args[] = {
-/* cmd,		set_args,	get_args,	name */
-	{SIOCSIPFRAMING, IW_PRIV_TYPE_BYTE | IW_PRIV_SIZE_FIXED | 1, 0,
-	 "set_framing"},
-	{SIOCGIPFRAMING, 0, IW_PRIV_TYPE_BYTE | IW_PRIV_SIZE_FIXED | 1,
-	 "get_framing"},
-	{SIOCGIPCOUNTRY, 0, IW_PRIV_TYPE_BYTE | IW_PRIV_SIZE_FIXED | 1,
-	 "get_country"},
-};
-
-static const struct iw_handler_def ray_handler_def = {
-	.num_standard = ARRAY_SIZE(ray_handler),
-	.num_private = ARRAY_SIZE(ray_private_handler),
-	.num_private_args = ARRAY_SIZE(ray_private_args),
-	.standard = ray_handler,
-	.private = ray_private_handler,
-	.private_args = ray_private_args,
-	.get_wireless_stats = ray_get_wireless_stats,
-};
-
-/*===========================================================================*/
-static int ray_open(struct net_device *dev)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link;
-	link = local->finder;
-
-	dev_dbg(&link->dev, "ray_open('%s')\n", dev->name);
-
-	if (link->open == 0)
-		local->num_multi = 0;
-	link->open++;
-
-	/* If the card is not started, time to start it ! - Jean II */
-	if (local->card_status == CARD_AWAITING_PARAM) {
-		int i;
-
-		dev_dbg(&link->dev, "ray_open: doing init now !\n");
-
-		/* Download startup parameters */
-		if ((i = dl_startup_params(dev)) < 0) {
-			printk(KERN_INFO
-			       "ray_dev_init dl_startup_params failed - "
-			       "returns 0x%x\n", i);
-			return -1;
-		}
-	}
-
-	if (sniffer)
-		netif_stop_queue(dev);
-	else
-		netif_start_queue(dev);
-
-	dev_dbg(&link->dev, "ray_open ending\n");
-	return 0;
-} /* end ray_open */
-
-/*===========================================================================*/
-static int ray_dev_close(struct net_device *dev)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link;
-	link = local->finder;
-
-	dev_dbg(&link->dev, "ray_dev_close('%s')\n", dev->name);
-
-	link->open--;
-	netif_stop_queue(dev);
-
-	/* In here, we should stop the hardware (stop card from beeing active)
-	 * and set local->card_status to CARD_AWAITING_PARAM, so that while the
-	 * card is closed we can chage its configuration.
-	 * Probably also need a COR reset to get sane state - Jean II */
-
-	return 0;
-} /* end ray_dev_close */
-
-/*===========================================================================*/
-static void ray_reset(struct net_device *dev)
-{
-	pr_debug("ray_reset entered\n");
-}
-
-/*===========================================================================*/
-/* Cause a firmware interrupt if it is ready for one                         */
-/* Return nonzero if not ready                                               */
-static int interrupt_ecf(ray_dev_t *local, int ccs)
-{
-	int i = 50;
-	struct pcmcia_device *link = local->finder;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs interrupt_ecf - device not present\n");
-		return -1;
-	}
-	dev_dbg(&link->dev, "interrupt_ecf(local=%p, ccs = 0x%x\n", local, ccs);
-
-	while (i &&
-	       (readb(local->amem + CIS_OFFSET + ECF_INTR_OFFSET) &
-		ECF_INTR_SET))
-		i--;
-	if (i == 0) {
-		dev_dbg(&link->dev, "ray_cs interrupt_ecf card not ready for interrupt\n");
-		return -1;
-	}
-	/* Fill the mailbox, then kick the card */
-	writeb(ccs, local->sram + SCB_BASE);
-	writeb(ECF_INTR_SET, local->amem + CIS_OFFSET + ECF_INTR_OFFSET);
-	return 0;
-} /* interrupt_ecf */
-
-/*===========================================================================*/
-/* Get next free transmit CCS                                                */
-/* Return - index of current tx ccs                                          */
-static int get_free_tx_ccs(ray_dev_t *local)
-{
-	int i;
-	struct ccs __iomem *pccs = ccs_base(local);
-	struct pcmcia_device *link = local->finder;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs get_free_tx_ccs - device not present\n");
-		return ECARDGONE;
-	}
-
-	if (test_and_set_bit(0, &local->tx_ccs_lock)) {
-		dev_dbg(&link->dev, "ray_cs tx_ccs_lock busy\n");
-		return ECCSBUSY;
-	}
-
-	for (i = 0; i < NUMBER_OF_TX_CCS; i++) {
-		if (readb(&(pccs + i)->buffer_status) == CCS_BUFFER_FREE) {
-			writeb(CCS_BUFFER_BUSY, &(pccs + i)->buffer_status);
-			writeb(CCS_END_LIST, &(pccs + i)->link);
-			local->tx_ccs_lock = 0;
-			return i;
-		}
-	}
-	local->tx_ccs_lock = 0;
-	dev_dbg(&link->dev, "ray_cs ERROR no free tx CCS for raylink card\n");
-	return ECCSFULL;
-} /* get_free_tx_ccs */
-
-/*===========================================================================*/
-/* Get next free CCS                                                         */
-/* Return - index of current ccs                                             */
-static int get_free_ccs(ray_dev_t *local)
-{
-	int i;
-	struct ccs __iomem *pccs = ccs_base(local);
-	struct pcmcia_device *link = local->finder;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs get_free_ccs - device not present\n");
-		return ECARDGONE;
-	}
-	if (test_and_set_bit(0, &local->ccs_lock)) {
-		dev_dbg(&link->dev, "ray_cs ccs_lock busy\n");
-		return ECCSBUSY;
-	}
-
-	for (i = NUMBER_OF_TX_CCS; i < NUMBER_OF_CCS; i++) {
-		if (readb(&(pccs + i)->buffer_status) == CCS_BUFFER_FREE) {
-			writeb(CCS_BUFFER_BUSY, &(pccs + i)->buffer_status);
-			writeb(CCS_END_LIST, &(pccs + i)->link);
-			local->ccs_lock = 0;
-			return i;
-		}
-	}
-	local->ccs_lock = 0;
-	dev_dbg(&link->dev, "ray_cs ERROR no free CCS for raylink card\n");
-	return ECCSFULL;
-} /* get_free_ccs */
-
-/*===========================================================================*/
-static void authenticate_timeout(struct timer_list *t)
-{
-	ray_dev_t *local = from_timer(local, t, timer);
-	del_timer(&local->timer);
-	printk(KERN_INFO "ray_cs Authentication with access point failed"
-	       " - timeout\n");
-	join_net(&local->timer);
-}
-
-/*===========================================================================*/
-static int parse_addr(char *in_str, UCHAR *out)
-{
-	int i, k;
-	int len;
-
-	if (in_str == NULL)
-		return 0;
-	len = strnlen(in_str, ADDRLEN * 2 + 1) - 1;
-	if (len < 1)
-		return 0;
-	memset(out, 0, ADDRLEN);
-
-	i = 5;
-
-	while (len > 0) {
-		if ((k = hex_to_bin(in_str[len--])) != -1)
-			out[i] = k;
-		else
-			return 0;
-
-		if (len == 0)
-			break;
-		if ((k = hex_to_bin(in_str[len--])) != -1)
-			out[i] += k << 4;
-		else
-			return 0;
-		if (!i--)
-			break;
-	}
-	return 1;
-}
-
-/*===========================================================================*/
-static struct net_device_stats *ray_get_stats(struct net_device *dev)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	struct status __iomem *p = local->sram + STATUS_BASE;
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs net_device_stats - device not present\n");
-		return &local->stats;
-	}
-	if (readb(&p->mrx_overflow_for_host)) {
-		local->stats.rx_over_errors += swab16(readw(&p->mrx_overflow));
-		writeb(0, &p->mrx_overflow);
-		writeb(0, &p->mrx_overflow_for_host);
-	}
-	if (readb(&p->mrx_checksum_error_for_host)) {
-		local->stats.rx_crc_errors +=
-		    swab16(readw(&p->mrx_checksum_error));
-		writeb(0, &p->mrx_checksum_error);
-		writeb(0, &p->mrx_checksum_error_for_host);
-	}
-	if (readb(&p->rx_hec_error_for_host)) {
-		local->stats.rx_frame_errors += swab16(readw(&p->rx_hec_error));
-		writeb(0, &p->rx_hec_error);
-		writeb(0, &p->rx_hec_error_for_host);
-	}
-	return &local->stats;
-}
-
-/*===========================================================================*/
-static void ray_update_parm(struct net_device *dev, UCHAR objid, UCHAR *value,
-			    int len)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	int ccsindex;
-	int i;
-	struct ccs __iomem *pccs;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_update_parm - device not present\n");
-		return;
-	}
-
-	if ((ccsindex = get_free_ccs(local)) < 0) {
-		dev_dbg(&link->dev, "ray_update_parm - No free ccs\n");
-		return;
-	}
-	pccs = ccs_base(local) + ccsindex;
-	writeb(CCS_UPDATE_PARAMS, &pccs->cmd);
-	writeb(objid, &pccs->var.update_param.object_id);
-	writeb(1, &pccs->var.update_param.number_objects);
-	writeb(0, &pccs->var.update_param.failure_cause);
-	for (i = 0; i < len; i++) {
-		writeb(value[i], local->sram + HOST_TO_ECF_BASE);
-	}
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		dev_dbg(&link->dev, "ray_cs associate failed - ECF not ready for intr\n");
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-	}
-}
-
-/*===========================================================================*/
-static void ray_update_multi_list(struct net_device *dev, int all)
-{
-	int ccsindex;
-	struct ccs __iomem *pccs;
-	ray_dev_t *local = netdev_priv(dev);
-	struct pcmcia_device *link = local->finder;
-	void __iomem *p = local->sram + HOST_TO_ECF_BASE;
-
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_update_multi_list - device not present\n");
-		return;
-	} else
-		dev_dbg(&link->dev, "ray_update_multi_list(%p)\n", dev);
-	if ((ccsindex = get_free_ccs(local)) < 0) {
-		dev_dbg(&link->dev, "ray_update_multi - No free ccs\n");
-		return;
-	}
-	pccs = ccs_base(local) + ccsindex;
-	writeb(CCS_UPDATE_MULTICAST_LIST, &pccs->cmd);
-
-	if (all) {
-		writeb(0xff, &pccs->var);
-		local->num_multi = 0xff;
-	} else {
-		struct netdev_hw_addr *ha;
-		int i = 0;
-
-		/* Copy the kernel's list of MC addresses to card */
-		netdev_for_each_mc_addr(ha, dev) {
-			memcpy_toio(p, ha->addr, ETH_ALEN);
-			dev_dbg(&link->dev, "ray_update_multi add addr %pm\n",
-				ha->addr);
-			p += ETH_ALEN;
-			i++;
-		}
-		if (i > 256 / ADDRLEN)
-			i = 256 / ADDRLEN;
-		writeb((UCHAR) i, &pccs->var);
-		dev_dbg(&link->dev, "ray_cs update_multi %d addresses in list\n", i);
-		/* Interrupt the firmware to process the command */
-		local->num_multi = i;
-	}
-	if (interrupt_ecf(local, ccsindex)) {
-		dev_dbg(&link->dev,
-		      "ray_cs update_multi failed - ECF not ready for intr\n");
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-	}
-} /* end ray_update_multi_list */
-
-/*===========================================================================*/
-static void set_multicast_list(struct net_device *dev)
-{
-	ray_dev_t *local = netdev_priv(dev);
-	UCHAR promisc;
-
-	pr_debug("ray_cs set_multicast_list(%p)\n", dev);
-
-	if (dev->flags & IFF_PROMISC) {
-		if (local->sparm.b5.a_promiscuous_mode == 0) {
-			pr_debug("ray_cs set_multicast_list promisc on\n");
-			local->sparm.b5.a_promiscuous_mode = 1;
-			promisc = 1;
-			ray_update_parm(dev, OBJID_promiscuous_mode,
-					&promisc, sizeof(promisc));
-		}
-	} else {
-		if (local->sparm.b5.a_promiscuous_mode == 1) {
-			pr_debug("ray_cs set_multicast_list promisc off\n");
-			local->sparm.b5.a_promiscuous_mode = 0;
-			promisc = 0;
-			ray_update_parm(dev, OBJID_promiscuous_mode,
-					&promisc, sizeof(promisc));
-		}
-	}
-
-	if (dev->flags & IFF_ALLMULTI)
-		ray_update_multi_list(dev, 1);
-	else {
-		if (local->num_multi != netdev_mc_count(dev))
-			ray_update_multi_list(dev, 0);
-	}
-} /* end set_multicast_list */
-
-/*=============================================================================
- * All routines below here are run at interrupt time.
-=============================================================================*/
-static irqreturn_t ray_interrupt(int irq, void *dev_id)
-{
-	struct net_device *dev = (struct net_device *)dev_id;
-	struct pcmcia_device *link;
-	ray_dev_t *local;
-	struct ccs __iomem *pccs;
-	struct rcs __iomem *prcs;
-	UCHAR rcsindex;
-	UCHAR tmp;
-	UCHAR cmd;
-	UCHAR status;
-	UCHAR memtmp[ESSID_SIZE + 1];
-
-
-	if (dev == NULL)	/* Note that we want interrupts with dev->start == 0 */
-		return IRQ_NONE;
-
-	pr_debug("ray_cs: interrupt for *dev=%p\n", dev);
-
-	local = netdev_priv(dev);
-	link = local->finder;
-	if (!pcmcia_dev_present(link)) {
-		pr_debug(
-			"ray_cs interrupt from device not present or suspended.\n");
-		return IRQ_NONE;
-	}
-	rcsindex = readb(&((struct scb __iomem *)(local->sram))->rcs_index);
-
-	if (rcsindex >= (NUMBER_OF_CCS + NUMBER_OF_RCS)) {
-		dev_dbg(&link->dev, "ray_cs interrupt bad rcsindex = 0x%x\n", rcsindex);
-		clear_interrupt(local);
-		return IRQ_HANDLED;
-	}
-	if (rcsindex < NUMBER_OF_CCS) { /* If it's a returned CCS */
-		pccs = ccs_base(local) + rcsindex;
-		cmd = readb(&pccs->cmd);
-		status = readb(&pccs->buffer_status);
-		switch (cmd) {
-		case CCS_DOWNLOAD_STARTUP_PARAMS:	/* Happens in firmware someday */
-			del_timer(&local->timer);
-			if (status == CCS_COMMAND_COMPLETE) {
-				dev_dbg(&link->dev,
-				      "ray_cs interrupt download_startup_parameters OK\n");
-			} else {
-				dev_dbg(&link->dev,
-				      "ray_cs interrupt download_startup_parameters fail\n");
-			}
-			break;
-		case CCS_UPDATE_PARAMS:
-			dev_dbg(&link->dev, "ray_cs interrupt update params done\n");
-			if (status != CCS_COMMAND_COMPLETE) {
-				tmp =
-				    readb(&pccs->var.update_param.
-					  failure_cause);
-				dev_dbg(&link->dev,
-				      "ray_cs interrupt update params failed - reason %d\n",
-				      tmp);
-			}
-			break;
-		case CCS_REPORT_PARAMS:
-			dev_dbg(&link->dev, "ray_cs interrupt report params done\n");
-			break;
-		case CCS_UPDATE_MULTICAST_LIST:	/* Note that this CCS isn't returned */
-			dev_dbg(&link->dev,
-			      "ray_cs interrupt CCS Update Multicast List done\n");
-			break;
-		case CCS_UPDATE_POWER_SAVINGS_MODE:
-			dev_dbg(&link->dev,
-			      "ray_cs interrupt update power save mode done\n");
-			break;
-		case CCS_START_NETWORK:
-		case CCS_JOIN_NETWORK:
-			memcpy(memtmp, local->sparm.b4.a_current_ess_id,
-								ESSID_SIZE);
-			memtmp[ESSID_SIZE] = '\0';
-
-			if (status == CCS_COMMAND_COMPLETE) {
-				if (readb
-				    (&pccs->var.start_network.net_initiated) ==
-				    1) {
-					dev_dbg(&link->dev,
-					      "ray_cs interrupt network \"%s\" started\n",
-					      memtmp);
-				} else {
-					dev_dbg(&link->dev,
-					      "ray_cs interrupt network \"%s\" joined\n",
-					      memtmp);
-				}
-				memcpy_fromio(&local->bss_id,
-					      pccs->var.start_network.bssid,
-					      ADDRLEN);
-
-				if (local->fw_ver == 0x55)
-					local->net_default_tx_rate = 3;
-				else
-					local->net_default_tx_rate =
-					    readb(&pccs->var.start_network.
-						  net_default_tx_rate);
-				local->encryption =
-				    readb(&pccs->var.start_network.encryption);
-				if (!sniffer && (local->net_type == INFRA)
-				    && !(local->sparm.b4.a_acting_as_ap_status)) {
-					authenticate(local);
-				}
-				local->card_status = CARD_ACQ_COMPLETE;
-			} else {
-				local->card_status = CARD_ACQ_FAILED;
-
-				del_timer(&local->timer);
-				local->timer.expires = jiffies + HZ * 5;
-				if (status == CCS_START_NETWORK) {
-					dev_dbg(&link->dev,
-					      "ray_cs interrupt network \"%s\" start failed\n",
-					      memtmp);
-					local->timer.function = start_net;
-				} else {
-					dev_dbg(&link->dev,
-					      "ray_cs interrupt network \"%s\" join failed\n",
-					      memtmp);
-					local->timer.function = join_net;
-				}
-				add_timer(&local->timer);
-			}
-			break;
-		case CCS_START_ASSOCIATION:
-			if (status == CCS_COMMAND_COMPLETE) {
-				local->card_status = CARD_ASSOC_COMPLETE;
-				dev_dbg(&link->dev, "ray_cs association successful\n");
-			} else {
-				dev_dbg(&link->dev, "ray_cs association failed,\n");
-				local->card_status = CARD_ASSOC_FAILED;
-				join_net(&local->timer);
-			}
-			break;
-		case CCS_TX_REQUEST:
-			if (status == CCS_COMMAND_COMPLETE) {
-				dev_dbg(&link->dev,
-				      "ray_cs interrupt tx request complete\n");
-			} else {
-				dev_dbg(&link->dev,
-				      "ray_cs interrupt tx request failed\n");
-			}
-			if (!sniffer)
-				netif_start_queue(dev);
-			netif_wake_queue(dev);
-			break;
-		case CCS_TEST_MEMORY:
-			dev_dbg(&link->dev, "ray_cs interrupt mem test done\n");
-			break;
-		case CCS_SHUTDOWN:
-			dev_dbg(&link->dev,
-			      "ray_cs interrupt Unexpected CCS returned - Shutdown\n");
-			break;
-		case CCS_DUMP_MEMORY:
-			dev_dbg(&link->dev, "ray_cs interrupt dump memory done\n");
-			break;
-		case CCS_START_TIMER:
-			dev_dbg(&link->dev,
-			      "ray_cs interrupt DING - raylink timer expired\n");
-			break;
-		default:
-			dev_dbg(&link->dev,
-			      "ray_cs interrupt Unexpected CCS 0x%x returned 0x%x\n",
-			      rcsindex, cmd);
-		}
-		writeb(CCS_BUFFER_FREE, &pccs->buffer_status);
-	} else { /* It's an RCS */
-
-		prcs = rcs_base(local) + rcsindex;
-
-		switch (readb(&prcs->interrupt_id)) {
-		case PROCESS_RX_PACKET:
-			ray_rx(dev, local, prcs);
-			break;
-		case REJOIN_NET_COMPLETE:
-			dev_dbg(&link->dev, "ray_cs interrupt rejoin net complete\n");
-			local->card_status = CARD_ACQ_COMPLETE;
-			/* do we need to clear tx buffers CCS's? */
-			if (local->sparm.b4.a_network_type == ADHOC) {
-				if (!sniffer)
-					netif_start_queue(dev);
-			} else {
-				memcpy_fromio(&local->bss_id,
-					      prcs->var.rejoin_net_complete.
-					      bssid, ADDRLEN);
-				dev_dbg(&link->dev, "ray_cs new BSSID = %pm\n",
-					local->bss_id);
-				if (!sniffer)
-					authenticate(local);
-			}
-			break;
-		case ROAMING_INITIATED:
-			dev_dbg(&link->dev, "ray_cs interrupt roaming initiated\n");
-			netif_stop_queue(dev);
-			local->card_status = CARD_DOING_ACQ;
-			break;
-		case JAPAN_CALL_SIGN_RXD:
-			dev_dbg(&link->dev, "ray_cs interrupt japan call sign rx\n");
-			break;
-		default:
-			dev_dbg(&link->dev,
-			      "ray_cs Unexpected interrupt for RCS 0x%x cmd = 0x%x\n",
-			      rcsindex,
-			      (unsigned int)readb(&prcs->interrupt_id));
-			break;
-		}
-		writeb(CCS_BUFFER_FREE, &prcs->buffer_status);
-	}
-	clear_interrupt(local);
-	return IRQ_HANDLED;
-} /* ray_interrupt */
-
-/*===========================================================================*/
-static void ray_rx(struct net_device *dev, ray_dev_t *local,
-		   struct rcs __iomem *prcs)
-{
-	int rx_len;
-	unsigned int pkt_addr;
-	void __iomem *pmsg;
-	pr_debug("ray_rx process rx packet\n");
-
-	/* Calculate address of packet within Rx buffer */
-	pkt_addr = ((readb(&prcs->var.rx_packet.rx_data_ptr[0]) << 8)
-		    + readb(&prcs->var.rx_packet.rx_data_ptr[1])) & RX_BUFF_END;
-	/* Length of first packet fragment */
-	rx_len = (readb(&prcs->var.rx_packet.rx_data_length[0]) << 8)
-	    + readb(&prcs->var.rx_packet.rx_data_length[1]);
-
-	local->last_rsl = readb(&prcs->var.rx_packet.rx_sig_lev);
-	pmsg = local->rmem + pkt_addr;
-	switch (readb(pmsg)) {
-	case DATA_TYPE:
-		pr_debug("ray_rx data type\n");
-		rx_data(dev, prcs, pkt_addr, rx_len);
-		break;
-	case AUTHENTIC_TYPE:
-		pr_debug("ray_rx authentic type\n");
-		if (sniffer)
-			rx_data(dev, prcs, pkt_addr, rx_len);
-		else
-			rx_authenticate(local, prcs, pkt_addr, rx_len);
-		break;
-	case DEAUTHENTIC_TYPE:
-		pr_debug("ray_rx deauth type\n");
-		if (sniffer)
-			rx_data(dev, prcs, pkt_addr, rx_len);
-		else
-			rx_deauthenticate(local, prcs, pkt_addr, rx_len);
-		break;
-	case NULL_MSG_TYPE:
-		pr_debug("ray_cs rx NULL msg\n");
-		break;
-	case BEACON_TYPE:
-		pr_debug("ray_rx beacon type\n");
-		if (sniffer)
-			rx_data(dev, prcs, pkt_addr, rx_len);
-
-		copy_from_rx_buff(local, (UCHAR *) &local->last_bcn, pkt_addr,
-				  rx_len < sizeof(struct beacon_rx) ?
-				  rx_len : sizeof(struct beacon_rx));
-
-		local->beacon_rxed = 1;
-		/* Get the statistics so the card counters never overflow */
-		ray_get_stats(dev);
-		break;
-	default:
-		pr_debug("ray_cs unknown pkt type %2x\n",
-		      (unsigned int)readb(pmsg));
-		break;
-	}
-
-} /* end ray_rx */
-
-/*===========================================================================*/
-static void rx_data(struct net_device *dev, struct rcs __iomem *prcs,
-		    unsigned int pkt_addr, int rx_len)
-{
-	struct sk_buff *skb = NULL;
-	struct rcs __iomem *prcslink = prcs;
-	ray_dev_t *local = netdev_priv(dev);
-	UCHAR *rx_ptr;
-	int total_len;
-	int tmp;
-#ifdef WIRELESS_SPY
-	int siglev = local->last_rsl;
-	u_char linksrcaddr[ETH_ALEN];	/* Other end of the wireless link */
-#endif
-
-	if (!sniffer) {
-		if (translate) {
-/* TBD length needs fixing for translated header */
-			if (rx_len < (ETH_HLEN + RX_MAC_HEADER_LENGTH) ||
-			    rx_len >
-			    (dev->mtu + RX_MAC_HEADER_LENGTH + ETH_HLEN +
-			     FCS_LEN)) {
-				pr_debug(
-				      "ray_cs invalid packet length %d received\n",
-				      rx_len);
-				return;
-			}
-		} else { /* encapsulated ethernet */
-
-			if (rx_len < (ETH_HLEN + RX_MAC_HEADER_LENGTH) ||
-			    rx_len >
-			    (dev->mtu + RX_MAC_HEADER_LENGTH + ETH_HLEN +
-			     FCS_LEN)) {
-				pr_debug(
-				      "ray_cs invalid packet length %d received\n",
-				      rx_len);
-				return;
-			}
-		}
-	}
-	pr_debug("ray_cs rx_data packet\n");
-	/* If fragmented packet, verify sizes of fragments add up */
-	if (readb(&prcs->var.rx_packet.next_frag_rcs_index) != 0xFF) {
-		pr_debug("ray_cs rx'ed fragment\n");
-		tmp = (readb(&prcs->var.rx_packet.totalpacketlength[0]) << 8)
-		    + readb(&prcs->var.rx_packet.totalpacketlength[1]);
-		total_len = tmp;
-		prcslink = prcs;
-		do {
-			tmp -=
-			    (readb(&prcslink->var.rx_packet.rx_data_length[0])
-			     << 8)
-			    + readb(&prcslink->var.rx_packet.rx_data_length[1]);
-			if (readb(&prcslink->var.rx_packet.next_frag_rcs_index)
-			    == 0xFF || tmp < 0)
-				break;
-			prcslink = rcs_base(local)
-			    + readb(&prcslink->link_field);
-		} while (1);
-
-		if (tmp < 0) {
-			pr_debug(
-			      "ray_cs rx_data fragment lengths don't add up\n");
-			local->stats.rx_dropped++;
-			release_frag_chain(local, prcs);
-			return;
-		}
-	} else { /* Single unfragmented packet */
-		total_len = rx_len;
-	}
-
-	skb = dev_alloc_skb(total_len + 5);
-	if (skb == NULL) {
-		pr_debug("ray_cs rx_data could not allocate skb\n");
-		local->stats.rx_dropped++;
-		if (readb(&prcs->var.rx_packet.next_frag_rcs_index) != 0xFF)
-			release_frag_chain(local, prcs);
-		return;
-	}
-	skb_reserve(skb, 2);	/* Align IP on 16 byte (TBD check this) */
-
-	pr_debug("ray_cs rx_data total_len = %x, rx_len = %x\n", total_len,
-	      rx_len);
-
-/************************/
-	/* Reserve enough room for the whole damn packet. */
-	rx_ptr = skb_put(skb, total_len);
-	/* Copy the whole packet to sk_buff */
-	rx_ptr +=
-	    copy_from_rx_buff(local, rx_ptr, pkt_addr & RX_BUFF_END, rx_len);
-	/* Get source address */
-#ifdef WIRELESS_SPY
-	skb_copy_from_linear_data_offset(skb,
-					 offsetof(struct mac_header, addr_2),
-					 linksrcaddr, ETH_ALEN);
-#endif
-	/* Now, deal with encapsulation/translation/sniffer */
-	if (!sniffer) {
-		if (!translate) {
-			/* Encapsulated ethernet, so just lop off 802.11 MAC header */
-/* TBD reserve            skb_reserve( skb, RX_MAC_HEADER_LENGTH); */
-			skb_pull(skb, RX_MAC_HEADER_LENGTH);
-		} else {
-			/* Do translation */
-			untranslate(local, skb, total_len);
-		}
-	} else { /* sniffer mode, so just pass whole packet */
-	}
-
-/************************/
-	/* Now pick up the rest of the fragments if any */
-	tmp = 17;
-	if (readb(&prcs->var.rx_packet.next_frag_rcs_index) != 0xFF) {
-		prcslink = prcs;
-		pr_debug("ray_cs rx_data in fragment loop\n");
-		do {
-			prcslink = rcs_base(local)
-			    +
-			    readb(&prcslink->var.rx_packet.next_frag_rcs_index);
-			rx_len =
-			    ((readb(&prcslink->var.rx_packet.rx_data_length[0])
-			      << 8)
-			     +
-			     readb(&prcslink->var.rx_packet.rx_data_length[1]))
-			    & RX_BUFF_END;
-			pkt_addr =
-			    ((readb(&prcslink->var.rx_packet.rx_data_ptr[0]) <<
-			      8)
-			     + readb(&prcslink->var.rx_packet.rx_data_ptr[1]))
-			    & RX_BUFF_END;
-
-			rx_ptr +=
-			    copy_from_rx_buff(local, rx_ptr, pkt_addr, rx_len);
-
-		} while (tmp-- &&
-			 readb(&prcslink->var.rx_packet.next_frag_rcs_index) !=
-			 0xFF);
-		release_frag_chain(local, prcs);
-	}
-
-	skb->protocol = eth_type_trans(skb, dev);
-	netif_rx(skb);
-	local->stats.rx_packets++;
-	local->stats.rx_bytes += total_len;
-
-	/* Gather signal strength per address */
-#ifdef WIRELESS_SPY
-	/* For the Access Point or the node having started the ad-hoc net
-	 * note : ad-hoc work only in some specific configurations, but we
-	 * kludge in ray_get_wireless_stats... */
-	if (!memcmp(linksrcaddr, local->bss_id, ETH_ALEN)) {
-		/* Update statistics */
-		/*local->wstats.qual.qual = none ? */
-		local->wstats.qual.level = siglev;
-		/*local->wstats.qual.noise = none ? */
-		local->wstats.qual.updated = 0x2;
-	}
-	/* Now, update the spy stuff */
-	{
-		struct iw_quality wstats;
-		wstats.level = siglev;
-		/* wstats.noise = none ? */
-		/* wstats.qual = none ? */
-		wstats.updated = 0x2;
-		/* Update spy records */
-		wireless_spy_update(dev, linksrcaddr, &wstats);
-	}
-#endif /* WIRELESS_SPY */
-} /* end rx_data */
-
-/*===========================================================================*/
-static void untranslate(ray_dev_t *local, struct sk_buff *skb, int len)
-{
-	snaphdr_t *psnap = (snaphdr_t *) (skb->data + RX_MAC_HEADER_LENGTH);
-	struct ieee80211_hdr *pmac = (struct ieee80211_hdr *)skb->data;
-	__be16 type = *(__be16 *) psnap->ethertype;
-	int delta;
-	struct ethhdr *peth;
-	UCHAR srcaddr[ADDRLEN];
-	UCHAR destaddr[ADDRLEN];
-	static const UCHAR org_bridge[3] = { 0, 0, 0xf8 };
-	static const UCHAR org_1042[3] = { 0, 0, 0 };
-
-	memcpy(destaddr, ieee80211_get_DA(pmac), ADDRLEN);
-	memcpy(srcaddr, ieee80211_get_SA(pmac), ADDRLEN);
-
-#if 0
-	if {
-		print_hex_dump(KERN_DEBUG, "skb->data before untranslate: ",
-			       DUMP_PREFIX_NONE, 16, 1,
-			       skb->data, 64, true);
-		printk(KERN_DEBUG
-		       "type = %08x, xsap = %02x%02x%02x, org = %02x02x02x\n",
-		       ntohs(type), psnap->dsap, psnap->ssap, psnap->ctrl,
-		       psnap->org[0], psnap->org[1], psnap->org[2]);
-		printk(KERN_DEBUG "untranslate skb->data = %p\n", skb->data);
-	}
-#endif
-
-	if (psnap->dsap != 0xaa || psnap->ssap != 0xaa || psnap->ctrl != 3) {
-		/* not a snap type so leave it alone */
-		pr_debug("ray_cs untranslate NOT SNAP %02x %02x %02x\n",
-		      psnap->dsap, psnap->ssap, psnap->ctrl);
-
-		delta = RX_MAC_HEADER_LENGTH - ETH_HLEN;
-		peth = (struct ethhdr *)(skb->data + delta);
-		peth->h_proto = htons(len - RX_MAC_HEADER_LENGTH);
-	} else { /* Its a SNAP */
-		if (memcmp(psnap->org, org_bridge, 3) == 0) {
-		/* EtherII and nuke the LLC */
-			pr_debug("ray_cs untranslate Bridge encap\n");
-			delta = RX_MAC_HEADER_LENGTH
-			    + sizeof(struct snaphdr_t) - ETH_HLEN;
-			peth = (struct ethhdr *)(skb->data + delta);
-			peth->h_proto = type;
-		} else if (memcmp(psnap->org, org_1042, 3) == 0) {
-			switch (ntohs(type)) {
-			case ETH_P_IPX:
-			case ETH_P_AARP:
-				pr_debug("ray_cs untranslate RFC IPX/AARP\n");
-				delta = RX_MAC_HEADER_LENGTH - ETH_HLEN;
-				peth = (struct ethhdr *)(skb->data + delta);
-				peth->h_proto =
-				    htons(len - RX_MAC_HEADER_LENGTH);
-				break;
-			default:
-				pr_debug("ray_cs untranslate RFC default\n");
-				delta = RX_MAC_HEADER_LENGTH +
-				    sizeof(struct snaphdr_t) - ETH_HLEN;
-				peth = (struct ethhdr *)(skb->data + delta);
-				peth->h_proto = type;
-				break;
-			}
-		} else {
-			printk("ray_cs untranslate very confused by packet\n");
-			delta = RX_MAC_HEADER_LENGTH - ETH_HLEN;
-			peth = (struct ethhdr *)(skb->data + delta);
-			peth->h_proto = type;
-		}
-	}
-/* TBD reserve  skb_reserve(skb, delta); */
-	skb_pull(skb, delta);
-	pr_debug("untranslate after skb_pull(%d), skb->data = %p\n", delta,
-	      skb->data);
-	memcpy(peth->h_dest, destaddr, ADDRLEN);
-	memcpy(peth->h_source, srcaddr, ADDRLEN);
-#if 0
-	{
-		int i;
-		printk(KERN_DEBUG "skb->data after untranslate:");
-		for (i = 0; i < 64; i++)
-			printk("%02x ", skb->data[i]);
-		printk("\n");
-	}
-#endif
-} /* end untranslate */
-
-/*===========================================================================*/
-/* Copy data from circular receive buffer to PC memory.
- * dest     = destination address in PC memory
- * pkt_addr = source address in receive buffer
- * len      = length of packet to copy
- */
-static int copy_from_rx_buff(ray_dev_t *local, UCHAR *dest, int pkt_addr,
-			     int length)
-{
-	int wrap_bytes = (pkt_addr + length) - (RX_BUFF_END + 1);
-	if (wrap_bytes <= 0) {
-		memcpy_fromio(dest, local->rmem + pkt_addr, length);
-	} else { /* Packet wrapped in circular buffer */
-
-		memcpy_fromio(dest, local->rmem + pkt_addr,
-			      length - wrap_bytes);
-		memcpy_fromio(dest + length - wrap_bytes, local->rmem,
-			      wrap_bytes);
-	}
-	return length;
-}
-
-/*===========================================================================*/
-static void release_frag_chain(ray_dev_t *local, struct rcs __iomem *prcs)
-{
-	struct rcs __iomem *prcslink = prcs;
-	int tmp = 17;
-	unsigned rcsindex = readb(&prcs->var.rx_packet.next_frag_rcs_index);
-
-	while (tmp--) {
-		writeb(CCS_BUFFER_FREE, &prcslink->buffer_status);
-		if (rcsindex >= (NUMBER_OF_CCS + NUMBER_OF_RCS)) {
-			pr_debug("ray_cs interrupt bad rcsindex = 0x%x\n",
-			      rcsindex);
-			break;
-		}
-		prcslink = rcs_base(local) + rcsindex;
-		rcsindex = readb(&prcslink->var.rx_packet.next_frag_rcs_index);
-	}
-	writeb(CCS_BUFFER_FREE, &prcslink->buffer_status);
-}
-
-/*===========================================================================*/
-static void authenticate(ray_dev_t *local)
-{
-	struct pcmcia_device *link = local->finder;
-	dev_dbg(&link->dev, "ray_cs Starting authentication.\n");
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs authenticate - device not present\n");
-		return;
-	}
-
-	del_timer(&local->timer);
-	if (build_auth_frame(local, local->bss_id, OPEN_AUTH_REQUEST)) {
-		local->timer.function = join_net;
-	} else {
-		local->timer.function = authenticate_timeout;
-	}
-	local->timer.expires = jiffies + HZ * 2;
-	add_timer(&local->timer);
-	local->authentication_state = AWAITING_RESPONSE;
-} /* end authenticate */
-
-/*===========================================================================*/
-static void rx_authenticate(ray_dev_t *local, struct rcs __iomem *prcs,
-			    unsigned int pkt_addr, int rx_len)
-{
-	UCHAR buff[256];
-	struct ray_rx_msg *msg = (struct ray_rx_msg *) buff;
-
-	del_timer(&local->timer);
-
-	copy_from_rx_buff(local, buff, pkt_addr, rx_len & 0xff);
-	/* if we are trying to get authenticated */
-	if (local->sparm.b4.a_network_type == ADHOC) {
-		pr_debug("ray_cs rx_auth var= %6ph\n", msg->var);
-		if (msg->var[2] == 1) {
-			pr_debug("ray_cs Sending authentication response.\n");
-			if (!build_auth_frame
-			    (local, msg->mac.addr_2, OPEN_AUTH_RESPONSE)) {
-				local->authentication_state = NEED_TO_AUTH;
-				memcpy(local->auth_id, msg->mac.addr_2,
-				       ADDRLEN);
-			}
-		}
-	} else { /* Infrastructure network */
-
-		if (local->authentication_state == AWAITING_RESPONSE) {
-			/* Verify authentication sequence #2 and success */
-			if (msg->var[2] == 2) {
-				if ((msg->var[3] | msg->var[4]) == 0) {
-					pr_debug("Authentication successful\n");
-					local->card_status = CARD_AUTH_COMPLETE;
-					associate(local);
-					local->authentication_state =
-					    AUTHENTICATED;
-				} else {
-					pr_debug("Authentication refused\n");
-					local->card_status = CARD_AUTH_REFUSED;
-					join_net(&local->timer);
-					local->authentication_state =
-					    UNAUTHENTICATED;
-				}
-			}
-		}
-	}
-
-} /* end rx_authenticate */
-
-/*===========================================================================*/
-static void associate(ray_dev_t *local)
-{
-	struct ccs __iomem *pccs;
-	struct pcmcia_device *link = local->finder;
-	struct net_device *dev = link->priv;
-	int ccsindex;
-	if (!(pcmcia_dev_present(link))) {
-		dev_dbg(&link->dev, "ray_cs associate - device not present\n");
-		return;
-	}
-	/* If no tx buffers available, return */
-	if ((ccsindex = get_free_ccs(local)) < 0) {
-/* TBD should never be here but... what if we are? */
-		dev_dbg(&link->dev, "ray_cs associate - No free ccs\n");
-		return;
-	}
-	dev_dbg(&link->dev, "ray_cs Starting association with access point\n");
-	pccs = ccs_base(local) + ccsindex;
-	/* fill in the CCS */
-	writeb(CCS_START_ASSOCIATION, &pccs->cmd);
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		dev_dbg(&link->dev, "ray_cs associate failed - ECF not ready for intr\n");
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-
-		del_timer(&local->timer);
-		local->timer.expires = jiffies + HZ * 2;
-		local->timer.function = join_net;
-		add_timer(&local->timer);
-		local->card_status = CARD_ASSOC_FAILED;
-		return;
-	}
-	if (!sniffer)
-		netif_start_queue(dev);
-
-} /* end associate */
-
-/*===========================================================================*/
-static void rx_deauthenticate(ray_dev_t *local, struct rcs __iomem *prcs,
-			      unsigned int pkt_addr, int rx_len)
-{
-/*  UCHAR buff[256];
-    struct ray_rx_msg *msg = (struct ray_rx_msg *) buff;
-*/
-	pr_debug("Deauthentication frame received\n");
-	local->authentication_state = UNAUTHENTICATED;
-	/* Need to reauthenticate or rejoin depending on reason code */
-/*  copy_from_rx_buff(local, buff, pkt_addr, rx_len & 0xff);
- */
-}
-
-/*===========================================================================*/
-static void clear_interrupt(ray_dev_t *local)
-{
-	writeb(0, local->amem + CIS_OFFSET + HCS_INTR_OFFSET);
-}
-
-/*===========================================================================*/
-#ifdef CONFIG_PROC_FS
-#define MAXDATA (PAGE_SIZE - 80)
-
-static const char *card_status[] = {
-	"Card inserted - uninitialized",	/* 0 */
-	"Card not downloaded",			/* 1 */
-	"Waiting for download parameters",	/* 2 */
-	"Card doing acquisition",		/* 3 */
-	"Acquisition complete",			/* 4 */
-	"Authentication complete",		/* 5 */
-	"Association complete",			/* 6 */
-	"???", "???", "???", "???",		/* 7 8 9 10 undefined */
-	"Card init error",			/* 11 */
-	"Download parameters error",		/* 12 */
-	"???",					/* 13 */
-	"Acquisition failed",			/* 14 */
-	"Authentication refused",		/* 15 */
-	"Association failed"			/* 16 */
-};
-
-static const char *nettype[] = { "Adhoc", "Infra " };
-static const char *framing[] = { "Encapsulation", "Translation" }
-
-;
-/*===========================================================================*/
-static int ray_cs_proc_show(struct seq_file *m, void *v)
-{
-/* Print current values which are not available via other means
- * eg ifconfig
- */
-	int i;
-	struct pcmcia_device *link;
-	struct net_device *dev;
-	ray_dev_t *local;
-	UCHAR *p;
-	struct freq_hop_element *pfh;
-	UCHAR c[33];
-
-	link = this_device;
-	if (!link)
-		return 0;
-	dev = (struct net_device *)link->priv;
-	if (!dev)
-		return 0;
-	local = netdev_priv(dev);
-	if (!local)
-		return 0;
-
-	seq_puts(m, "Raylink Wireless LAN driver status\n");
-	seq_printf(m, "%s\n", rcsid);
-	/* build 4 does not report version, and field is 0x55 after memtest */
-	seq_puts(m, "Firmware version     = ");
-	if (local->fw_ver == 0x55)
-		seq_puts(m, "4 - Use dump_cis for more details\n");
-	else
-		seq_printf(m, "%2d.%02d.%02d\n",
-			   local->fw_ver, local->fw_bld, local->fw_var);
-
-	for (i = 0; i < 32; i++)
-		c[i] = local->sparm.b5.a_current_ess_id[i];
-	c[32] = 0;
-	seq_printf(m, "%s network ESSID = \"%s\"\n",
-		   nettype[local->sparm.b5.a_network_type], c);
-
-	p = local->bss_id;
-	seq_printf(m, "BSSID                = %pM\n", p);
-
-	seq_printf(m, "Country code         = %d\n",
-		   local->sparm.b5.a_curr_country_code);
-
-	i = local->card_status;
-	if (i < 0)
-		i = 10;
-	if (i > 16)
-		i = 10;
-	seq_printf(m, "Card status          = %s\n", card_status[i]);
-
-	seq_printf(m, "Framing mode         = %s\n", framing[translate]);
-
-	seq_printf(m, "Last pkt signal lvl  = %d\n", local->last_rsl);
-
-	if (local->beacon_rxed) {
-		/* Pull some fields out of last beacon received */
-		seq_printf(m, "Beacon Interval      = %d Kus\n",
-			   local->last_bcn.beacon_intvl[0]
-			   + 256 * local->last_bcn.beacon_intvl[1]);
-
-		p = local->last_bcn.elements;
-		if (p[0] == C_ESSID_ELEMENT_ID)
-			p += p[1] + 2;
-		else {
-			seq_printf(m,
-				   "Parse beacon failed at essid element id = %d\n",
-				   p[0]);
-			return 0;
-		}
-
-		if (p[0] == C_SUPPORTED_RATES_ELEMENT_ID) {
-			seq_puts(m, "Supported rate codes = ");
-			for (i = 2; i < p[1] + 2; i++)
-				seq_printf(m, "0x%02x ", p[i]);
-			seq_putc(m, '\n');
-			p += p[1] + 2;
-		} else {
-			seq_puts(m, "Parse beacon failed at rates element\n");
-			return 0;
-		}
-
-		if (p[0] == C_FH_PARAM_SET_ELEMENT_ID) {
-			pfh = (struct freq_hop_element *)p;
-			seq_printf(m, "Hop dwell            = %d Kus\n",
-				   pfh->dwell_time[0] +
-				   256 * pfh->dwell_time[1]);
-			seq_printf(m, "Hop set              = %d\n",
-				   pfh->hop_set);
-			seq_printf(m, "Hop pattern          = %d\n",
-				   pfh->hop_pattern);
-			seq_printf(m, "Hop index            = %d\n",
-				   pfh->hop_index);
-			p += p[1] + 2;
-		} else {
-			seq_puts(m,
-				 "Parse beacon failed at FH param element\n");
-			return 0;
-		}
-	} else {
-		seq_puts(m, "No beacons received\n");
-	}
-	return 0;
-}
-#endif
-/*===========================================================================*/
-static int build_auth_frame(ray_dev_t *local, UCHAR *dest, int auth_type)
-{
-	int addr;
-	struct ccs __iomem *pccs;
-	struct tx_msg __iomem *ptx;
-	int ccsindex;
-
-	/* If no tx buffers available, return */
-	if ((ccsindex = get_free_tx_ccs(local)) < 0) {
-		pr_debug("ray_cs send authenticate - No free tx ccs\n");
-		return -1;
-	}
-
-	pccs = ccs_base(local) + ccsindex;
-
-	/* Address in card space */
-	addr = TX_BUF_BASE + (ccsindex << 11);
-	/* fill in the CCS */
-	writeb(CCS_TX_REQUEST, &pccs->cmd);
-	writeb(addr >> 8, pccs->var.tx_request.tx_data_ptr);
-	writeb(0x20, pccs->var.tx_request.tx_data_ptr + 1);
-	writeb(TX_AUTHENTICATE_LENGTH_MSB, pccs->var.tx_request.tx_data_length);
-	writeb(TX_AUTHENTICATE_LENGTH_LSB,
-	       pccs->var.tx_request.tx_data_length + 1);
-	writeb(0, &pccs->var.tx_request.pow_sav_mode);
-
-	ptx = local->sram + addr;
-	/* fill in the mac header */
-	writeb(PROTOCOL_VER | AUTHENTIC_TYPE, &ptx->mac.frame_ctl_1);
-	writeb(0, &ptx->mac.frame_ctl_2);
-
-	memcpy_toio(ptx->mac.addr_1, dest, ADDRLEN);
-	memcpy_toio(ptx->mac.addr_2, local->sparm.b4.a_mac_addr, ADDRLEN);
-	memcpy_toio(ptx->mac.addr_3, local->bss_id, ADDRLEN);
-
-	/* Fill in msg body with protocol 00 00, sequence 01 00 ,status 00 00 */
-	memset_io(ptx->var, 0, 6);
-	writeb(auth_type & 0xff, ptx->var + 2);
-
-	/* Interrupt the firmware to process the command */
-	if (interrupt_ecf(local, ccsindex)) {
-		pr_debug(
-		      "ray_cs send authentication request failed - ECF not ready for intr\n");
-		writeb(CCS_BUFFER_FREE, &(pccs++)->buffer_status);
-		return -1;
-	}
-	return 0;
-} /* End build_auth_frame */
-
-/*===========================================================================*/
-#ifdef CONFIG_PROC_FS
-static ssize_t ray_cs_essid_proc_write(struct file *file,
-		const char __user *buffer, size_t count, loff_t *pos)
-{
-	static char proc_essid[33];
-	unsigned int len = count;
-
-	if (len > 32)
-		len = 32;
-	memset(proc_essid, 0, 33);
-	if (copy_from_user(proc_essid, buffer, len))
-		return -EFAULT;
-	essid = proc_essid;
-	return count;
-}
-
-static const struct proc_ops ray_cs_essid_proc_ops = {
-	.proc_write	= ray_cs_essid_proc_write,
-	.proc_lseek	= noop_llseek,
-};
-
-static ssize_t int_proc_write(struct file *file, const char __user *buffer,
-			      size_t count, loff_t *pos)
-{
-	static char proc_number[10];
-	char *p;
-	int nr, len;
-
-	if (!count)
-		return 0;
-
-	if (count > 9)
-		return -EINVAL;
-	if (copy_from_user(proc_number, buffer, count))
-		return -EFAULT;
-	p = proc_number;
-	nr = 0;
-	len = count;
-	do {
-		unsigned int c = *p - '0';
-		if (c > 9)
-			return -EINVAL;
-		nr = nr * 10 + c;
-		p++;
-	} while (--len);
-	*(int *)pde_data(file_inode(file)) = nr;
-	return count;
-}
-
-static const struct proc_ops int_proc_ops = {
-	.proc_write	= int_proc_write,
-	.proc_lseek	= noop_llseek,
-};
-#endif
-
-static const struct pcmcia_device_id ray_ids[] = {
-	PCMCIA_DEVICE_MANF_CARD(0x01a6, 0x0000),
-	PCMCIA_DEVICE_NULL,
-};
-
-MODULE_DEVICE_TABLE(pcmcia, ray_ids);
-
-static struct pcmcia_driver ray_driver = {
-	.owner = THIS_MODULE,
-	.name = "ray_cs",
-	.probe = ray_probe,
-	.remove = ray_detach,
-	.id_table = ray_ids,
-	.suspend = ray_suspend,
-	.resume = ray_resume,
-};
-
-static int __init init_ray_cs(void)
-{
-	int rc;
-
-	pr_debug("%s\n", rcsid);
-	rc = pcmcia_register_driver(&ray_driver);
-	pr_debug("raylink init_module register_pcmcia_driver returns 0x%x\n",
-	      rc);
-	if (rc)
-		return rc;
-
-#ifdef CONFIG_PROC_FS
-	proc_mkdir("driver/ray_cs", NULL);
-
-	proc_create_single("driver/ray_cs/ray_cs", 0, NULL, ray_cs_proc_show);
-	proc_create("driver/ray_cs/essid", 0200, NULL, &ray_cs_essid_proc_ops);
-	proc_create_data("driver/ray_cs/net_type", 0200, NULL, &int_proc_ops,
-			 &net_type);
-	proc_create_data("driver/ray_cs/translate", 0200, NULL, &int_proc_ops,
-			 &translate);
-#endif
-	translate = !!translate;
-	return 0;
-} /* init_ray_cs */
-
-/*===========================================================================*/
-
-static void __exit exit_ray_cs(void)
-{
-	pr_debug("ray_cs: cleanup_module\n");
-
-#ifdef CONFIG_PROC_FS
-	remove_proc_subtree("driver/ray_cs", NULL);
-#endif
-
-	pcmcia_unregister_driver(&ray_driver);
-} /* exit_ray_cs */
-
-module_init(init_ray_cs);
-module_exit(exit_ray_cs);
-
-/*===========================================================================*/
diff --git a/drivers/net/wireless/ray_cs.h b/drivers/net/wireless/ray_cs.h
deleted file mode 100644
index 0609d8625019..000000000000
--- a/drivers/net/wireless/ray_cs.h
+++ /dev/null
@@ -1,74 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Raytheon wireless LAN PCMCIA card driver for Linux 
-   A  PCMCIA client driver for the Raylink wireless network card
-   Written by Corey Thomas
-*/
-
-#ifndef _RAY_CS_H_
-#define _RAY_CS_H_
-
-struct beacon_rx {
-    struct mac_header mac;
-    UCHAR timestamp[8];
-    UCHAR beacon_intvl[2];
-    UCHAR capability[2];
-    UCHAR elements[sizeof(struct essid_element) 
-                  + sizeof(struct rates_element)
-                  + sizeof(struct freq_hop_element) 
-                  + sizeof(struct japan_call_sign_element)
-                  + sizeof(struct tim_element)];
-};
-
-/* Return values for get_free{,_tx}_ccs */
-#define ECCSFULL  (-1)
-#define ECCSBUSY  (-2)
-#define ECARDGONE (-3)
-
-typedef struct ray_dev_t {
-    int card_status;
-    int authentication_state;
-    void __iomem *sram;            /* pointer to beginning of shared RAM     */
-    void __iomem *amem;            /* pointer to attribute mem window        */
-    void __iomem *rmem;            /* pointer to receive buffer window       */
-    struct pcmcia_device *finder;            /* pointer back to struct pcmcia_device for card    */
-    struct timer_list timer;
-    unsigned long tx_ccs_lock;
-    unsigned long ccs_lock;
-    int   dl_param_ccs;
-    union {
-        struct b4_startup_params b4;
-        struct b5_startup_params b5;
-    } sparm;
-    int timeout_flag;
-    UCHAR supported_rates[8];
-    UCHAR japan_call_sign[12];
-    struct startup_res_6 startup_res;
-    int num_multi;
-    /* Network parameters from start/join */
-    UCHAR bss_id[6];
-    UCHAR auth_id[6];
-    UCHAR net_default_tx_rate;
-    UCHAR encryption;
-    struct net_device_stats stats;
-
-    UCHAR net_type;
-    UCHAR sta_type;
-    UCHAR fw_ver;
-    UCHAR fw_bld;
-    UCHAR fw_var;
-    UCHAR ASIC_version;
-    UCHAR assoc_id[2];
-    UCHAR tib_length;
-    UCHAR last_rsl;
-    int beacon_rxed;
-    struct beacon_rx last_bcn;
-    iw_stats	wstats;		/* Wireless specific stats */
-#ifdef WIRELESS_SPY
-    struct iw_spy_data		spy_data;
-    struct iw_public_data	wireless_data;
-#endif	/* WIRELESS_SPY */
-
-} ray_dev_t;
-/*****************************************************************************/
-
-#endif /* _RAY_CS_H_ */
diff --git a/drivers/net/wireless/rayctl.h b/drivers/net/wireless/rayctl.h
deleted file mode 100644
index 2b0f332043d7..000000000000
--- a/drivers/net/wireless/rayctl.h
+++ /dev/null
@@ -1,734 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _RAYCTL_H_
-#define _RAYCTL_H_
-
-typedef unsigned char UCHAR;
-
-/****** IEEE 802.11 constants ************************************************/
-#define ADDRLEN           6
-/* Frame control 1 bit fields */
-#define PROTOCOL_VER      0x00
-#define DATA_TYPE         0x08
-#define ASSOC_REQ_TYPE    0x00
-#define ASSOC_RESP_TYPE   0x10
-#define REASSOC_REQ_TYPE  0x20
-#define REASSOC_RESP_TYPE 0x30
-#define NULL_MSG_TYPE     0x48
-#define BEACON_TYPE       0x80
-#define DISASSOC_TYPE     0xA0
-#define PSPOLL_TYPE       0xA4
-#define AUTHENTIC_TYPE    0xB0
-#define DEAUTHENTIC_TYPE  0xC0
-/* Frame control 2 bit fields */
-#define FC2_TO_DS         0x01
-#define FC2_FROM_DS       0x02
-#define FC2_MORE_FRAG     0x04
-#define FC2_RETRY         0x08
-#define FC2_PSM           0x10
-#define FC2_MORE_DATA     0x20
-#define FC2_WEP           0x40
-#define FC2_ORDER         0x80
-/*****************************************************************************/
-/* 802.11 element ID's and lengths */
-#define C_BP_CAPABILITY_ESS             0x01
-#define C_BP_CAPABILITY_IBSS            0x02
-#define C_BP_CAPABILITY_CF_POLLABLE     0x04
-#define C_BP_CAPABILITY_CF_POLL_REQUEST 0x08
-#define C_BP_CAPABILITY_PRIVACY         0x10
-
-#define C_ESSID_ELEMENT_ID               0
-#define C_ESSID_ELEMENT_MAX_LENGTH       32
-
-#define C_SUPPORTED_RATES_ELEMENT_ID     1
-#define C_SUPPORTED_RATES_ELEMENT_LENGTH 2
-
-#define C_FH_PARAM_SET_ELEMENT_ID        2
-#define C_FH_PARAM_SET_ELEMENT_LNGTH     5
-
-#define C_CF_PARAM_SET_ELEMENT_ID        4
-#define C_CF_PARAM_SET_ELEMENT_LNGTH     6
-
-#define C_TIM_ELEMENT_ID                 5
-#define C_TIM_BITMAP_LENGTH            251
-#define C_TIM_BMCAST_BIT              0x01
-
-#define C_IBSS_ELEMENT_ID                6
-#define C_IBSS_ELEMENT_LENGTH            2
-
-#define C_JAPAN_CALL_SIGN_ELEMENT_ID    51
-#define C_JAPAN_CALL_SIGN_ELEMENT_LNGTH 12
-
-#define C_DISASSOC_REASON_CODE_LEN       2
-#define C_DISASSOC_REASON_CODE_DEFAULT   8
-
-#define C_CRC_LEN                        4
-#define C_NUM_SUPPORTED_RATES            8 
-/****** IEEE 802.11 mac header for type data packets *************************/
-struct mac_header {
-  UCHAR frame_ctl_1;                          
-  UCHAR frame_ctl_2;
-  UCHAR duration_lsb;
-  UCHAR duration_msb;
-  UCHAR addr_1[ADDRLEN];
-  UCHAR addr_2[ADDRLEN];
-  UCHAR addr_3[ADDRLEN];
-  UCHAR seq_frag_num[2];
-/*  UCHAR addr_4[ADDRLEN]; *//* only present for AP to AP (TO DS and FROM DS */
-};
-/****** IEEE 802.11 frame element structures *********************************/
-struct essid_element
-{
-  UCHAR id;
-  UCHAR length;
-  UCHAR text[C_ESSID_ELEMENT_MAX_LENGTH];
-};
-struct rates_element
-{
-  UCHAR id;
-  UCHAR length;
-  UCHAR value[8];
-};
-struct freq_hop_element
-{
-  UCHAR id;
-  UCHAR length;
-  UCHAR dwell_time[2];
-  UCHAR hop_set;
-  UCHAR hop_pattern;
-  UCHAR hop_index;
-};
-struct tim_element
-{
-  UCHAR id;
-  UCHAR length;
-  UCHAR dtim_count;
-  UCHAR dtim_period;    
-  UCHAR bitmap_control;
-  UCHAR tim[C_TIM_BITMAP_LENGTH];
-};
-struct ibss_element
-{
-  UCHAR id;
-  UCHAR length;
-  UCHAR atim_window[2];
-};
-struct japan_call_sign_element
-{
-  UCHAR id;
-  UCHAR length;
-  UCHAR call_sign[12];
-};
-/****** Beacon message structures ********************************************/
-/* .elements is a large lump of max size because elements are variable size  */
-struct infra_beacon
-{
-    UCHAR timestamp[8];
-    UCHAR beacon_intvl[2];
-    UCHAR capability[2];
-    UCHAR elements[sizeof(struct essid_element) 
-                  + sizeof(struct rates_element)
-                  + sizeof(struct freq_hop_element) 
-                  + sizeof(struct japan_call_sign_element)
-                  + sizeof(struct tim_element)];
-};
-struct adhoc_beacon
-{
-    UCHAR timestamp[8];
-    UCHAR beacon_intvl[2];
-    UCHAR capability[2];
-    UCHAR elements[sizeof(struct essid_element) 
-                  + sizeof(struct rates_element)
-                  + sizeof(struct freq_hop_element) 
-                  + sizeof(struct japan_call_sign_element)
-                  + sizeof(struct ibss_element)];
-};
-/*****************************************************************************/
-/*****************************************************************************/
-/* #define C_MAC_HDR_2_WEP 0x40 */
-/* TX/RX CCS constants */
-#define TX_HEADER_LENGTH 0x1C
-#define RX_MAC_HEADER_LENGTH 0x18
-#define TX_AUTHENTICATE_LENGTH (TX_HEADER_LENGTH + 6)
-#define TX_AUTHENTICATE_LENGTH_MSB (TX_AUTHENTICATE_LENGTH >> 8)
-#define TX_AUTHENTICATE_LENGTH_LSB (TX_AUTHENTICATE_LENGTH & 0xff)
-#define TX_DEAUTHENTICATE_LENGTH (TX_HEADER_LENGTH + 2)
-#define TX_DEAUTHENTICATE_LENGTH_MSB (TX_AUTHENTICATE_LENGTH >> 8)
-#define TX_DEAUTHENTICATE_LENGTH_LSB (TX_AUTHENTICATE_LENGTH & 0xff)
-#define FCS_LEN           4
-
-#define ADHOC                 0
-#define INFRA                 1
-
-#define TYPE_STA              0
-#define TYPE_AP               1
-
-#define PASSIVE_SCAN          1
-#define ACTIVE_SCAN           1
-
-#define PSM_CAM               0
-
-/* Country codes */
-#define USA                   1
-#define EUROPE                2
-#define JAPAN                 3
-#define KOREA                 4
-#define SPAIN                 5
-#define FRANCE                6
-#define ISRAEL                7
-#define AUSTRALIA             8
-#define JAPAN_TEST            9
-
-/* Hop pattern lengths */
-#define USA_HOP_MOD          79 
-#define EUROPE_HOP_MOD       79 
-#define JAPAN_HOP_MOD        23
-#define KOREA_HOP_MOD        23
-#define SPAIN_HOP_MOD        27
-#define FRANCE_HOP_MOD       35
-#define ISRAEL_HOP_MOD       35
-#define AUSTRALIA_HOP_MOD    47
-#define JAPAN_TEST_HOP_MOD   23
-
-#define ESSID_SIZE           32
-/**********************************************************************/
-/* CIS Register Constants */
-#define CIS_OFFSET             0x0f00
-/* Configuration Option Register (0x0F00) */
-#define COR_OFFSET             0x00
-#define COR_SOFT_RESET         0x80
-#define COR_LEVEL_IRQ          0x40
-#define COR_CONFIG_NUM         0x01
-#define COR_DEFAULT            (COR_LEVEL_IRQ | COR_CONFIG_NUM)
-
-/* Card Configuration and Status Register (0x0F01) */
-#define CCSR_OFFSET            0x01
-#define CCSR_HOST_INTR_PENDING 0x01
-#define CCSR_POWER_DOWN        0x04
-
-/* HCS Interrupt Register (0x0F05) */
-#define HCS_INTR_OFFSET        0x05
-/* #define HCS_INTR_OFFSET        0x0A */
-#define HCS_INTR_CLEAR         0x00
-
-/* ECF Interrupt Register (0x0F06) */
-#define ECF_INTR_OFFSET        0x06
-/* #define ECF_INTR_OFFSET        0x0C */
-#define ECF_INTR_SET           0x01
-
-/* Authorization Register 0 (0x0F08) */
-#define AUTH_0_ON              0x57
-
-/* Authorization Register 1 (0x0F09) */
-#define AUTH_1_ON              0x82
-
-/* Program Mode Register (0x0F0A) */
-#define PC2PM                  0x02
-#define PC2CAL                 0x10
-#define PC2MLSE                0x20
-
-/* PC Test Mode Register (0x0F0B) */
-#define PC_TEST_MODE           0x08
-
-/* Frequency Control Word (0x0F10) */
-/* Range 0x02 - 0xA6 */
-
-/* Test Mode Control 1-4 (0x0F14 - 0x0F17) */
-
-/**********************************************************************/
-
-/* Shared RAM Area */
-#define SCB_BASE               0x0000
-#define STATUS_BASE            0x0100
-#define HOST_TO_ECF_BASE       0x0200
-#define ECF_TO_HOST_BASE       0x0300
-#define CCS_BASE               0x0400
-#define RCS_BASE               0x0800
-#define INFRA_TIM_BASE         0x0C00
-#define SSID_LIST_BASE         0x0D00
-#define TX_BUF_BASE            0x1000
-#define RX_BUF_BASE            0x8000
-
-#define NUMBER_OF_CCS    64
-#define NUMBER_OF_RCS    64
-/*#define NUMBER_OF_TX_CCS 14 */
-#define NUMBER_OF_TX_CCS 14
-
-#define TX_BUF_SIZE      (2048 - sizeof(struct tx_msg))
-#define RX_BUFF_END      0x3FFF
-/* Values for buffer_status */
-#define CCS_BUFFER_FREE       0
-#define CCS_BUFFER_BUSY       1
-#define CCS_COMMAND_COMPLETE  2
-#define CCS_COMMAND_FAILED    3
-
-/* Values for cmd */
-#define CCS_DOWNLOAD_STARTUP_PARAMS    1
-#define CCS_UPDATE_PARAMS              2
-#define CCS_REPORT_PARAMS              3
-#define CCS_UPDATE_MULTICAST_LIST      4
-#define CCS_UPDATE_POWER_SAVINGS_MODE  5
-#define CCS_START_NETWORK              6
-#define CCS_JOIN_NETWORK               7
-#define CCS_START_ASSOCIATION          8
-#define CCS_TX_REQUEST                 9
-#define CCS_TEST_MEMORY              0xa
-#define CCS_SHUTDOWN                 0xb
-#define CCS_DUMP_MEMORY              0xc
-#define CCS_START_TIMER              0xe
-#define CCS_LAST_CMD                 CCS_START_TIMER
-
-/* Values for link field */
-#define CCS_END_LIST                 0xff
-
-/* values for buffer_status field */
-#define RCS_BUFFER_FREE       0
-#define RCS_BUFFER_BUSY       1
-#define RCS_COMPLETE          2
-#define RCS_FAILED            3
-#define RCS_BUFFER_RELEASE    0xFF
-
-/* values for interrupt_id field */
-#define PROCESS_RX_PACKET           0x80 /* */
-#define REJOIN_NET_COMPLETE         0x81 /* RCS ID: Rejoin Net Complete */
-#define ROAMING_INITIATED           0x82 /* RCS ID: Roaming Initiated   */
-#define JAPAN_CALL_SIGN_RXD         0x83 /* RCS ID: New Japan Call Sign */
-
-/*****************************************************************************/
-/* Memory types for dump memory command */
-#define C_MEM_PROG  0
-#define C_MEM_XDATA 1
-#define C_MEM_SFR   2
-#define C_MEM_IDATA 3
-
-/*** Return values for hw_xmit **********/
-#define XMIT_OK        (0)
-#define XMIT_MSG_BAD   (-1)
-#define XMIT_NO_CCS    (-2)
-#define XMIT_NO_INTR   (-3)
-#define XMIT_NEED_AUTH (-4)
-
-/*** Values for card status */
-#define CARD_INSERTED       (0)
-
-#define CARD_AWAITING_PARAM (1)
-#define CARD_INIT_ERROR     (11)
-
-#define CARD_DL_PARAM       (2)
-#define CARD_DL_PARAM_ERROR (12)
-
-#define CARD_DOING_ACQ      (3)
-
-#define CARD_ACQ_COMPLETE   (4)
-#define CARD_ACQ_FAILED     (14)
-
-#define CARD_AUTH_COMPLETE  (5)
-#define CARD_AUTH_REFUSED   (15)
-
-#define CARD_ASSOC_COMPLETE (6)
-#define CARD_ASSOC_FAILED   (16)
-
-/*** Values for authentication_state ***********************************/
-#define UNAUTHENTICATED     (0)
-#define AWAITING_RESPONSE   (1)
-#define AUTHENTICATED       (2)
-#define NEED_TO_AUTH        (3)
-
-/*** Values for authentication type ************************************/
-#define OPEN_AUTH_REQUEST   (1)
-#define OPEN_AUTH_RESPONSE  (2)
-#define BROADCAST_DEAUTH    (0xc0)
-/*** Values for timer functions ****************************************/
-#define TODO_NOTHING              (0)
-#define TODO_VERIFY_DL_START      (-1)
-#define TODO_START_NET            (-2)
-#define TODO_JOIN_NET             (-3)
-#define TODO_AUTHENTICATE_TIMEOUT (-4)
-#define TODO_SEND_CCS             (-5)
-/***********************************************************************/
-/* Parameter passing structure for update/report parameter CCS's */
-struct object_id {
-    void          *object_addr;
-    unsigned char object_length;
-};
-
-#define OBJID_network_type            0
-#define OBJID_acting_as_ap_status     1
-#define OBJID_current_ess_id          2
-#define OBJID_scanning_mode           3
-#define OBJID_power_mgt_state         4
-#define OBJID_mac_address             5
-#define OBJID_frag_threshold          6
-#define OBJID_hop_time                7
-#define OBJID_beacon_period           8
-#define OBJID_dtim_period             9
-#define OBJID_retry_max              10
-#define OBJID_ack_timeout            11
-#define OBJID_sifs                   12
-#define OBJID_difs                   13
-#define OBJID_pifs                   14
-#define OBJID_rts_threshold          15
-#define OBJID_scan_dwell_time        16
-#define OBJID_max_scan_dwell_time    17
-#define OBJID_assoc_resp_timeout     18
-#define OBJID_adhoc_scan_cycle_max   19
-#define OBJID_infra_scan_cycle_max   20
-#define OBJID_infra_super_cycle_max  21
-#define OBJID_promiscuous_mode       22
-#define OBJID_unique_word            23
-#define OBJID_slot_time              24
-#define OBJID_roaming_low_snr        25
-#define OBJID_low_snr_count_thresh   26
-#define OBJID_infra_missed_bcn       27
-#define OBJID_adhoc_missed_bcn       28
-#define OBJID_curr_country_code      29
-#define OBJID_hop_pattern            30
-#define OBJID_reserved               31
-#define OBJID_cw_max_msb             32
-#define OBJID_cw_min_msb             33
-#define OBJID_noise_filter_gain      34
-#define OBJID_noise_limit_offset     35
-#define OBJID_det_rssi_thresh_offset 36
-#define OBJID_med_busy_thresh_offset 37
-#define OBJID_det_sync_thresh        38
-#define OBJID_test_mode              39
-#define OBJID_test_min_chan_num      40
-#define OBJID_test_max_chan_num      41
-#define OBJID_allow_bcast_ID_prbrsp  42
-#define OBJID_privacy_must_start     43
-#define OBJID_privacy_can_join       44
-#define OBJID_basic_rate_set         45
-
-/**** Configuration/Status/Control Area ***************************/
-/*    System Control Block (SCB) Area
- *    Located at Shared RAM offset 0
- */
-struct scb {
-    UCHAR ccs_index;
-    UCHAR rcs_index;
-};
-
-/****** Status area at Shared RAM offset 0x0100 ******************************/
-struct status {
-    UCHAR mrx_overflow_for_host;         /* 0=ECF may write, 1=host may write*/
-    UCHAR mrx_checksum_error_for_host;   /* 0=ECF may write, 1=host may write*/
-    UCHAR rx_hec_error_for_host;         /* 0=ECF may write, 1=host may write*/
-    UCHAR reserved1;
-    short mrx_overflow;                  /* ECF increments on rx overflow    */
-    short mrx_checksum_error;            /* ECF increments on rx CRC error   */
-    short rx_hec_error;                  /* ECF incs on mac header CRC error */
-    UCHAR rxnoise;                       /* Average RSL measurement          */
-};
-
-/****** Host-to-ECF Data Area at Shared RAM offset 0x200 *********************/
-struct host_to_ecf_area {
-    
-};
-
-/****** ECF-to-Host Data Area at Shared RAM offset 0x0300 ********************/
-struct startup_res_518 {
-    UCHAR startup_word;
-    UCHAR station_addr[ADDRLEN];
-    UCHAR calc_prog_chksum;
-    UCHAR calc_cis_chksum;
-    UCHAR ecf_spare[7];
-    UCHAR japan_call_sign[12];
-};
-
-struct startup_res_6 {
-    UCHAR startup_word;
-    UCHAR station_addr[ADDRLEN];
-    UCHAR reserved;
-    UCHAR supp_rates[8];
-    UCHAR japan_call_sign[12];
-    UCHAR calc_prog_chksum;
-    UCHAR calc_cis_chksum;
-    UCHAR firmware_version[3];
-    UCHAR asic_version;
-    UCHAR tib_length;
-};
-
-struct start_join_net_params {
-    UCHAR net_type;
-    UCHAR ssid[ESSID_SIZE];
-    UCHAR reserved;
-    UCHAR privacy_can_join;
-};
-
-/****** Command Control Structure area at Shared ram offset 0x0400 ***********/
-/* Structures for command specific parameters (ccs.var) */
-struct update_param_cmd {
-    UCHAR object_id;
-    UCHAR number_objects;
-    UCHAR failure_cause;
-};
-struct report_param_cmd {
-    UCHAR object_id;
-    UCHAR number_objects;
-    UCHAR failure_cause;
-    UCHAR length;
-};
-struct start_network_cmd {
-    UCHAR update_param;
-    UCHAR bssid[ADDRLEN];
-    UCHAR net_initiated;
-    UCHAR net_default_tx_rate;
-    UCHAR encryption;
-};
-struct join_network_cmd {
-    UCHAR update_param;
-    UCHAR bssid[ADDRLEN];
-    UCHAR net_initiated;
-    UCHAR net_default_tx_rate;
-    UCHAR encryption;
-};
-struct tx_requested_cmd {
- 
-    UCHAR tx_data_ptr[2];
-    UCHAR tx_data_length[2];
-    UCHAR host_reserved[2];
-    UCHAR reserved[3];
-    UCHAR tx_rate;
-    UCHAR pow_sav_mode;
-    UCHAR retries;
-    UCHAR antenna;
-};
-struct tx_requested_cmd_4 {
- 
-    UCHAR tx_data_ptr[2];
-    UCHAR tx_data_length[2];
-    UCHAR dest_addr[ADDRLEN];
-    UCHAR pow_sav_mode;
-    UCHAR retries;
-    UCHAR station_id;
-};
-struct memory_dump_cmd {
-    UCHAR memory_type;
-    UCHAR memory_ptr[2];
-    UCHAR length;
-};
-struct update_association_cmd {
-    UCHAR status;
-    UCHAR aid[2];
-};
-struct start_timer_cmd {
-    UCHAR duration[2];
-};
-
-struct ccs {
-    UCHAR buffer_status;                 /* 0 = buffer free, 1 = buffer busy */
-                                         /* 2 = command complete, 3 = failed */
-    UCHAR cmd;                           /* command to ECF                   */
-    UCHAR link;                          /* link to next CCS, FF=end of list */
-    /* command specific parameters      */
-    union {
-        char reserved[13];
-        struct update_param_cmd update_param;
-        struct report_param_cmd report_param;
-        UCHAR nummulticast;
-        UCHAR mode;
-        struct start_network_cmd start_network;
-        struct join_network_cmd join_network;
-        struct tx_requested_cmd tx_request;
-        struct memory_dump_cmd memory_dump;
-        struct update_association_cmd update_assoc;
-        struct start_timer_cmd start_timer;
-    } var;
-};
-
-/*****************************************************************************/
-/* Transmit buffer structures */
-struct tib_structure {
-    UCHAR ccs_index;
-    UCHAR psm;
-    UCHAR pass_fail;
-    UCHAR retry_count;
-    UCHAR max_retries;
-    UCHAR frags_remaining;
-    UCHAR no_rb;
-    UCHAR rts_reqd;
-    UCHAR csma_tx_cntrl_2;
-    UCHAR sifs_tx_cntrl_2;
-    UCHAR tx_dma_addr_1[2];
-    UCHAR tx_dma_addr_2[2];
-    UCHAR var_dur_2mhz[2];
-    UCHAR var_dur_1mhz[2];
-    UCHAR max_dur_2mhz[2];
-    UCHAR max_dur_1mhz[2];
-    UCHAR hdr_len;
-    UCHAR max_frag_len[2];
-    UCHAR var_len[2];
-    UCHAR phy_hdr_4;
-    UCHAR mac_hdr_1;
-    UCHAR mac_hdr_2;
-    UCHAR sid[2];
-};
-
-struct phy_header {
-    UCHAR sfd[2];
-    UCHAR hdr_3;
-    UCHAR hdr_4;
-};
-struct ray_rx_msg {
-    struct mac_header mac;
-	UCHAR	var[];
-};
-
-struct tx_msg {
-    struct tib_structure tib;
-    struct phy_header phy;
-    struct mac_header mac;
-    UCHAR  var[1];
-};
-
-/****** ECF Receive Control Structure (RCS) Area at Shared RAM offset 0x0800  */
-/* Structures for command specific parameters (rcs.var) */
-struct rx_packet_cmd {
-    UCHAR rx_data_ptr[2];
-    UCHAR rx_data_length[2];
-    UCHAR rx_sig_lev;
-    UCHAR next_frag_rcs_index;
-    UCHAR totalpacketlength[2];
-};
-struct rejoin_net_cmplt_cmd {
-    UCHAR reserved;
-    UCHAR bssid[ADDRLEN];
-};
-struct japan_call_sign_rxd {
-    UCHAR rxd_call_sign[8];
-    UCHAR reserved[5];
-};
-
-struct rcs {
-    UCHAR buffer_status;
-    UCHAR interrupt_id;
-    UCHAR link_field;
-    /* command specific parameters      */
-    union {
-        UCHAR reserved[13]; 
-        struct rx_packet_cmd rx_packet;
-        struct rejoin_net_cmplt_cmd rejoin_net_complete;
-        struct japan_call_sign_rxd japan_call_sign;
-    } var;
-};
-
-/****** Startup parameter structures for both versions of firmware ***********/
-struct b4_startup_params {
-    UCHAR a_network_type;                /* C_ADHOC, C_INFRA                 */
-    UCHAR a_acting_as_ap_status;         /* C_TYPE_STA, C_TYPE_AP            */
-    UCHAR a_current_ess_id[ESSID_SIZE];  /* Null terminated unless 32 long   */
-    UCHAR a_scanning_mode;               /* passive 0, active 1              */
-    UCHAR a_power_mgt_state;             /* CAM 0,                           */
-    UCHAR a_mac_addr[ADDRLEN];           /*                                  */
-    UCHAR a_frag_threshold[2];           /* 512                              */
-    UCHAR a_hop_time[2];                 /* 16k * 2**n, n=0-4 in Kus         */
-    UCHAR a_beacon_period[2];            /* n * a_hop_time  in Kus           */
-    UCHAR a_dtim_period;                 /* in beacons                       */
-    UCHAR a_retry_max;                   /*                                  */
-    UCHAR a_ack_timeout;                 /*                                  */
-    UCHAR a_sifs;                        /*                                  */
-    UCHAR a_difs;                        /*                                  */
-    UCHAR a_pifs;                        /*                                  */
-    UCHAR a_rts_threshold[2];            /*                                  */
-    UCHAR a_scan_dwell_time[2];          /*                                  */
-    UCHAR a_max_scan_dwell_time[2];      /*                                  */
-    UCHAR a_assoc_resp_timeout_thresh;   /*                                  */
-    UCHAR a_adhoc_scan_cycle_max;        /*                                  */
-    UCHAR a_infra_scan_cycle_max;        /*                                  */
-    UCHAR a_infra_super_scan_cycle_max;  /*                                  */
-    UCHAR a_promiscuous_mode;            /*                                  */
-    UCHAR a_unique_word[2];              /*                                  */
-    UCHAR a_slot_time;                   /*                                  */
-    UCHAR a_roaming_low_snr_thresh;      /*                                  */
-    UCHAR a_low_snr_count_thresh;        /*                                  */
-    UCHAR a_infra_missed_bcn_thresh;     /*                                  */
-    UCHAR a_adhoc_missed_bcn_thresh;     /*                                  */
-    UCHAR a_curr_country_code;           /* C_USA                            */
-    UCHAR a_hop_pattern;                 /*                                  */
-    UCHAR a_hop_pattern_length;          /*                                  */
-/* b4 - b5 differences start here */
-    UCHAR a_cw_max;                      /*                                  */
-    UCHAR a_cw_min;                      /*                                  */
-    UCHAR a_noise_filter_gain;           /*                                  */
-    UCHAR a_noise_limit_offset;          /*                                  */
-    UCHAR a_det_rssi_thresh_offset;      /*                                  */
-    UCHAR a_med_busy_thresh_offset;      /*                                  */
-    UCHAR a_det_sync_thresh;             /*                                  */
-    UCHAR a_test_mode;                   /*                                  */
-    UCHAR a_test_min_chan_num;           /*                                  */
-    UCHAR a_test_max_chan_num;           /*                                  */
-    UCHAR a_rx_tx_delay;                 /*                                  */
-    UCHAR a_current_bss_id[ADDRLEN];     /*                                  */
-    UCHAR a_hop_set;                     /*                                  */
-};
-struct b5_startup_params {
-    UCHAR a_network_type;                /* C_ADHOC, C_INFRA                 */
-    UCHAR a_acting_as_ap_status;         /* C_TYPE_STA, C_TYPE_AP            */
-    UCHAR a_current_ess_id[ESSID_SIZE];  /* Null terminated unless 32 long   */
-    UCHAR a_scanning_mode;               /* passive 0, active 1              */
-    UCHAR a_power_mgt_state;             /* CAM 0,                           */
-    UCHAR a_mac_addr[ADDRLEN];           /*                                  */
-    UCHAR a_frag_threshold[2];           /* 512                              */
-    UCHAR a_hop_time[2];                 /* 16k * 2**n, n=0-4 in Kus         */
-    UCHAR a_beacon_period[2];            /* n * a_hop_time  in Kus           */
-    UCHAR a_dtim_period;                 /* in beacons                       */
-    UCHAR a_retry_max;                   /* 4                                */
-    UCHAR a_ack_timeout;                 /*                                  */
-    UCHAR a_sifs;                        /*                                  */
-    UCHAR a_difs;                        /*                                  */
-    UCHAR a_pifs;                        /*                                  */
-    UCHAR a_rts_threshold[2];            /*                                  */
-    UCHAR a_scan_dwell_time[2];          /*                                  */
-    UCHAR a_max_scan_dwell_time[2];      /*                                  */
-    UCHAR a_assoc_resp_timeout_thresh;   /*                                  */
-    UCHAR a_adhoc_scan_cycle_max;        /*                                  */
-    UCHAR a_infra_scan_cycle_max;        /*                                  */
-    UCHAR a_infra_super_scan_cycle_max;  /*                                  */
-    UCHAR a_promiscuous_mode;            /*                                  */
-    UCHAR a_unique_word[2];              /*                                  */
-    UCHAR a_slot_time;                   /*                                  */
-    UCHAR a_roaming_low_snr_thresh;      /*                                  */
-    UCHAR a_low_snr_count_thresh;        /*                                  */
-    UCHAR a_infra_missed_bcn_thresh;     /*                                  */
-    UCHAR a_adhoc_missed_bcn_thresh;     /*                                  */
-    UCHAR a_curr_country_code;           /* C_USA                            */
-    UCHAR a_hop_pattern;                 /*                                  */
-    UCHAR a_hop_pattern_length;          /*                                  */
-/* b4 - b5 differences start here */
-    UCHAR a_cw_max[2];                   /*                                  */
-    UCHAR a_cw_min[2];                   /*                                  */
-    UCHAR a_noise_filter_gain;           /*                                  */
-    UCHAR a_noise_limit_offset;          /*                                  */
-    UCHAR a_det_rssi_thresh_offset;      /*                                  */
-    UCHAR a_med_busy_thresh_offset;      /*                                  */
-    UCHAR a_det_sync_thresh;             /*                                  */
-    UCHAR a_test_mode;                   /*                                  */
-    UCHAR a_test_min_chan_num;           /*                                  */
-    UCHAR a_test_max_chan_num;           /*                                  */
-    UCHAR a_allow_bcast_SSID_probe_rsp;
-    UCHAR a_privacy_must_start;
-    UCHAR a_privacy_can_join;
-    UCHAR a_basic_rate_set[8];
-};
-
-/*****************************************************************************/
-#define RAY_IOCG_PARMS (SIOCDEVPRIVATE)
-#define RAY_IOCS_PARMS (SIOCDEVPRIVATE + 1)
-#define RAY_DO_CMD     (SIOCDEVPRIVATE + 2)
-
-/****** ethernet <-> 802.11 translation **************************************/
-typedef struct snaphdr_t
-{
-  UCHAR   dsap;
-  UCHAR   ssap;
-  UCHAR   ctrl;
-  UCHAR   org[3];
-  UCHAR   ethertype[2];
-} snaphdr_t;
-
-#define BRIDGE_ENCAP  0xf80000
-#define RFC1042_ENCAP 0
-#define SNAP_ID       0x0003aaaa
-#define RAY_IPX_TYPE  0x8137
-#define APPLEARP_TYPE 0x80f3
-/*****************************************************************************/
-#endif /* _RAYCTL_H_ */
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
deleted file mode 100644
index bf72e5fd39cf..000000000000
--- a/drivers/net/wireless/rndis_wlan.c
+++ /dev/null
@@ -1,3760 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * Driver for RNDIS based wireless USB devices.
- *
- * Copyright (C) 2007 by Bjorge Dijkstra <bjd@jooz.net>
- * Copyright (C) 2008-2009 by Jussi Kivilinna <jussi.kivilinna@iki.fi>
- *
- *  Portions of this file are based on NDISwrapper project,
- *  Copyright (C) 2003-2005 Pontus Fuchs, Giridhar Pemmasani
- *  http://ndiswrapper.sourceforge.net/
- */
-
-// #define	DEBUG			// error path messages, extra info
-// #define	VERBOSE			// more; success messages
-
-#include <linux/module.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/ethtool.h>
-#include <linux/workqueue.h>
-#include <linux/mutex.h>
-#include <linux/mii.h>
-#include <linux/usb.h>
-#include <linux/usb/cdc.h>
-#include <linux/ieee80211.h>
-#include <linux/if_arp.h>
-#include <linux/ctype.h>
-#include <linux/spinlock.h>
-#include <linux/slab.h>
-#include <net/cfg80211.h>
-#include <linux/usb/usbnet.h>
-#include <linux/usb/rndis_host.h>
-
-
-/* NOTE: All these are settings for Broadcom chipset */
-static char modparam_country[4] = "EU";
-module_param_string(country, modparam_country, 4, 0444);
-MODULE_PARM_DESC(country, "Country code (ISO 3166-1 alpha-2), default: EU");
-
-static int modparam_frameburst = 1;
-module_param_named(frameburst, modparam_frameburst, int, 0444);
-MODULE_PARM_DESC(frameburst, "enable frame bursting (default: on)");
-
-static int modparam_afterburner = 0;
-module_param_named(afterburner, modparam_afterburner, int, 0444);
-MODULE_PARM_DESC(afterburner,
-	"enable afterburner aka '125 High Speed Mode' (default: off)");
-
-static int modparam_power_save = 0;
-module_param_named(power_save, modparam_power_save, int, 0444);
-MODULE_PARM_DESC(power_save,
-	"set power save mode: 0=off, 1=on, 2=fast (default: off)");
-
-static int modparam_power_output = 3;
-module_param_named(power_output, modparam_power_output, int, 0444);
-MODULE_PARM_DESC(power_output,
-	"set power output: 0=25%, 1=50%, 2=75%, 3=100% (default: 100%)");
-
-static int modparam_roamtrigger = -70;
-module_param_named(roamtrigger, modparam_roamtrigger, int, 0444);
-MODULE_PARM_DESC(roamtrigger,
-	"set roaming dBm trigger: -80=optimize for distance, "
-				"-60=bandwidth (default: -70)");
-
-static int modparam_roamdelta = 1;
-module_param_named(roamdelta, modparam_roamdelta, int, 0444);
-MODULE_PARM_DESC(roamdelta,
-	"set roaming tendency: 0=aggressive, 1=moderate, "
-				"2=conservative (default: moderate)");
-
-static int modparam_workaround_interval;
-module_param_named(workaround_interval, modparam_workaround_interval,
-							int, 0444);
-MODULE_PARM_DESC(workaround_interval,
-	"set stall workaround interval in msecs (0=disabled) (default: 0)");
-
-/* Typical noise/maximum signal level values taken from ndiswrapper iw_ndis.h */
-#define	WL_NOISE	-96	/* typical noise level in dBm */
-#define	WL_SIGMAX	-32	/* typical maximum signal level in dBm */
-
-
-/* Assume that Broadcom 4320 (only chipset at time of writing known to be
- * based on wireless rndis) has default txpower of 13dBm.
- * This value is from Linksys WUSB54GSC User Guide, Appendix F: Specifications.
- *  100% : 20 mW ~ 13dBm
- *   75% : 15 mW ~ 12dBm
- *   50% : 10 mW ~ 10dBm
- *   25% :  5 mW ~  7dBm
- */
-#define BCM4320_DEFAULT_TXPOWER_DBM_100 13
-#define BCM4320_DEFAULT_TXPOWER_DBM_75  12
-#define BCM4320_DEFAULT_TXPOWER_DBM_50  10
-#define BCM4320_DEFAULT_TXPOWER_DBM_25  7
-
-/* Known device types */
-#define RNDIS_UNKNOWN	0
-#define RNDIS_BCM4320A	1
-#define RNDIS_BCM4320B	2
-
-
-/* NDIS data structures. Taken from wpa_supplicant driver_ndis.c
- * slightly modified for datatype endianess, etc
- */
-#define NDIS_802_11_LENGTH_SSID 32
-#define NDIS_802_11_LENGTH_RATES 8
-#define NDIS_802_11_LENGTH_RATES_EX 16
-
-enum ndis_80211_net_type {
-	NDIS_80211_TYPE_FREQ_HOP,
-	NDIS_80211_TYPE_DIRECT_SEQ,
-	NDIS_80211_TYPE_OFDM_A,
-	NDIS_80211_TYPE_OFDM_G
-};
-
-enum ndis_80211_net_infra {
-	NDIS_80211_INFRA_ADHOC,
-	NDIS_80211_INFRA_INFRA,
-	NDIS_80211_INFRA_AUTO_UNKNOWN
-};
-
-enum ndis_80211_auth_mode {
-	NDIS_80211_AUTH_OPEN,
-	NDIS_80211_AUTH_SHARED,
-	NDIS_80211_AUTH_AUTO_SWITCH,
-	NDIS_80211_AUTH_WPA,
-	NDIS_80211_AUTH_WPA_PSK,
-	NDIS_80211_AUTH_WPA_NONE,
-	NDIS_80211_AUTH_WPA2,
-	NDIS_80211_AUTH_WPA2_PSK
-};
-
-enum ndis_80211_encr_status {
-	NDIS_80211_ENCR_WEP_ENABLED,
-	NDIS_80211_ENCR_DISABLED,
-	NDIS_80211_ENCR_WEP_KEY_ABSENT,
-	NDIS_80211_ENCR_NOT_SUPPORTED,
-	NDIS_80211_ENCR_TKIP_ENABLED,
-	NDIS_80211_ENCR_TKIP_KEY_ABSENT,
-	NDIS_80211_ENCR_CCMP_ENABLED,
-	NDIS_80211_ENCR_CCMP_KEY_ABSENT
-};
-
-enum ndis_80211_priv_filter {
-	NDIS_80211_PRIV_ACCEPT_ALL,
-	NDIS_80211_PRIV_8021X_WEP
-};
-
-enum ndis_80211_status_type {
-	NDIS_80211_STATUSTYPE_AUTHENTICATION,
-	NDIS_80211_STATUSTYPE_MEDIASTREAMMODE,
-	NDIS_80211_STATUSTYPE_PMKID_CANDIDATELIST,
-	NDIS_80211_STATUSTYPE_RADIOSTATE,
-};
-
-enum ndis_80211_media_stream_mode {
-	NDIS_80211_MEDIA_STREAM_OFF,
-	NDIS_80211_MEDIA_STREAM_ON
-};
-
-enum ndis_80211_radio_status {
-	NDIS_80211_RADIO_STATUS_ON,
-	NDIS_80211_RADIO_STATUS_HARDWARE_OFF,
-	NDIS_80211_RADIO_STATUS_SOFTWARE_OFF,
-};
-
-enum ndis_80211_addkey_bits {
-	NDIS_80211_ADDKEY_8021X_AUTH = cpu_to_le32(1 << 28),
-	NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ = cpu_to_le32(1 << 29),
-	NDIS_80211_ADDKEY_PAIRWISE_KEY = cpu_to_le32(1 << 30),
-	NDIS_80211_ADDKEY_TRANSMIT_KEY = cpu_to_le32(1 << 31)
-};
-
-enum ndis_80211_addwep_bits {
-	NDIS_80211_ADDWEP_PERCLIENT_KEY = cpu_to_le32(1 << 30),
-	NDIS_80211_ADDWEP_TRANSMIT_KEY = cpu_to_le32(1 << 31)
-};
-
-enum ndis_80211_power_mode {
-	NDIS_80211_POWER_MODE_CAM,
-	NDIS_80211_POWER_MODE_MAX_PSP,
-	NDIS_80211_POWER_MODE_FAST_PSP,
-};
-
-enum ndis_80211_pmkid_cand_list_flag_bits {
-	NDIS_80211_PMKID_CAND_PREAUTH = cpu_to_le32(1 << 0)
-};
-
-struct ndis_80211_auth_request {
-	__le32 length;
-	u8 bssid[ETH_ALEN];
-	u8 padding[2];
-	__le32 flags;
-} __packed;
-
-struct ndis_80211_pmkid_candidate {
-	u8 bssid[ETH_ALEN];
-	u8 padding[2];
-	__le32 flags;
-} __packed;
-
-struct ndis_80211_pmkid_cand_list {
-	__le32 version;
-	__le32 num_candidates;
-	struct ndis_80211_pmkid_candidate candidate_list[];
-} __packed;
-
-struct ndis_80211_status_indication {
-	__le32 status_type;
-	union {
-		__le32					media_stream_mode;
-		__le32					radio_status;
-		struct ndis_80211_auth_request		auth_request[0];
-		struct ndis_80211_pmkid_cand_list	cand_list;
-	} u;
-} __packed;
-
-struct ndis_80211_ssid {
-	__le32 length;
-	u8 essid[NDIS_802_11_LENGTH_SSID];
-} __packed;
-
-struct ndis_80211_conf_freq_hop {
-	__le32 length;
-	__le32 hop_pattern;
-	__le32 hop_set;
-	__le32 dwell_time;
-} __packed;
-
-struct ndis_80211_conf {
-	__le32 length;
-	__le32 beacon_period;
-	__le32 atim_window;
-	__le32 ds_config;
-	struct ndis_80211_conf_freq_hop fh_config;
-} __packed;
-
-struct ndis_80211_bssid_ex {
-	__le32 length;
-	u8 mac[ETH_ALEN];
-	u8 padding[2];
-	struct ndis_80211_ssid ssid;
-	__le32 privacy;
-	__le32 rssi;
-	__le32 net_type;
-	struct ndis_80211_conf config;
-	__le32 net_infra;
-	u8 rates[NDIS_802_11_LENGTH_RATES_EX];
-	__le32 ie_length;
-	u8 ies[];
-} __packed;
-
-struct ndis_80211_bssid_list_ex {
-	__le32 num_items;
-	u8 bssid_data[];
-} __packed;
-
-struct ndis_80211_fixed_ies {
-	u8 timestamp[8];
-	__le16 beacon_interval;
-	__le16 capabilities;
-} __packed;
-
-struct ndis_80211_wep_key {
-	__le32 size;
-	__le32 index;
-	__le32 length;
-	u8 material[32];
-} __packed;
-
-struct ndis_80211_key {
-	__le32 size;
-	__le32 index;
-	__le32 length;
-	u8 bssid[ETH_ALEN];
-	u8 padding[6];
-	u8 rsc[8];
-	u8 material[32];
-} __packed;
-
-struct ndis_80211_remove_key {
-	__le32 size;
-	__le32 index;
-	u8 bssid[ETH_ALEN];
-	u8 padding[2];
-} __packed;
-
-struct ndis_config_param {
-	__le32 name_offs;
-	__le32 name_length;
-	__le32 type;
-	__le32 value_offs;
-	__le32 value_length;
-} __packed;
-
-struct ndis_80211_assoc_info {
-	__le32 length;
-	__le16 req_ies;
-	struct req_ie {
-		__le16 capa;
-		__le16 listen_interval;
-		u8 cur_ap_address[ETH_ALEN];
-	} req_ie;
-	__le32 req_ie_length;
-	__le32 offset_req_ies;
-	__le16 resp_ies;
-	struct resp_ie {
-		__le16 capa;
-		__le16 status_code;
-		__le16 assoc_id;
-	} resp_ie;
-	__le32 resp_ie_length;
-	__le32 offset_resp_ies;
-} __packed;
-
-struct ndis_80211_capability {
-	__le32 length;
-	__le32 version;
-	__le32 num_pmkids;
-	__le32 num_auth_encr_pair;
-} __packed;
-
-struct ndis_80211_bssid_info {
-	u8 bssid[ETH_ALEN];
-	u8 pmkid[16];
-} __packed;
-
-struct ndis_80211_pmkid {
-	__le32 length;
-	__le32 bssid_info_count;
-	struct ndis_80211_bssid_info bssid_info[];
-} __packed;
-
-/*
- *  private data
- */
-#define CAP_MODE_80211A		1
-#define CAP_MODE_80211B		2
-#define CAP_MODE_80211G		4
-#define CAP_MODE_MASK		7
-
-#define WORK_LINK_UP		0
-#define WORK_LINK_DOWN		1
-#define WORK_SET_MULTICAST_LIST	2
-
-#define RNDIS_WLAN_ALG_NONE	0
-#define RNDIS_WLAN_ALG_WEP	(1<<0)
-#define RNDIS_WLAN_ALG_TKIP	(1<<1)
-#define RNDIS_WLAN_ALG_CCMP	(1<<2)
-
-#define RNDIS_WLAN_NUM_KEYS		4
-#define RNDIS_WLAN_KEY_MGMT_NONE	0
-#define RNDIS_WLAN_KEY_MGMT_802_1X	(1<<0)
-#define RNDIS_WLAN_KEY_MGMT_PSK		(1<<1)
-
-#define COMMAND_BUFFER_SIZE	(CONTROL_BUFFER_SIZE + sizeof(struct rndis_set))
-
-static const struct ieee80211_channel rndis_channels[] = {
-	{ .center_freq = 2412 },
-	{ .center_freq = 2417 },
-	{ .center_freq = 2422 },
-	{ .center_freq = 2427 },
-	{ .center_freq = 2432 },
-	{ .center_freq = 2437 },
-	{ .center_freq = 2442 },
-	{ .center_freq = 2447 },
-	{ .center_freq = 2452 },
-	{ .center_freq = 2457 },
-	{ .center_freq = 2462 },
-	{ .center_freq = 2467 },
-	{ .center_freq = 2472 },
-	{ .center_freq = 2484 },
-};
-
-static const struct ieee80211_rate rndis_rates[] = {
-	{ .bitrate = 10 },
-	{ .bitrate = 20, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
-	{ .bitrate = 55, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
-	{ .bitrate = 110, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
-	{ .bitrate = 60 },
-	{ .bitrate = 90 },
-	{ .bitrate = 120 },
-	{ .bitrate = 180 },
-	{ .bitrate = 240 },
-	{ .bitrate = 360 },
-	{ .bitrate = 480 },
-	{ .bitrate = 540 }
-};
-
-static const u32 rndis_cipher_suites[] = {
-	WLAN_CIPHER_SUITE_WEP40,
-	WLAN_CIPHER_SUITE_WEP104,
-	WLAN_CIPHER_SUITE_TKIP,
-	WLAN_CIPHER_SUITE_CCMP,
-};
-
-struct rndis_wlan_encr_key {
-	int len;
-	u32 cipher;
-	u8 material[32];
-	u8 bssid[ETH_ALEN];
-	bool pairwise;
-	bool tx_key;
-};
-
-/* RNDIS device private data */
-struct rndis_wlan_private {
-	struct usbnet *usbdev;
-
-	struct wireless_dev wdev;
-
-	struct cfg80211_scan_request *scan_request;
-
-	struct workqueue_struct *workqueue;
-	struct delayed_work dev_poller_work;
-	struct delayed_work scan_work;
-	struct work_struct work;
-	struct mutex command_lock;
-	unsigned long work_pending;
-	int last_qual;
-	s32 cqm_rssi_thold;
-	u32 cqm_rssi_hyst;
-	int last_cqm_event_rssi;
-
-	struct ieee80211_supported_band band;
-	struct ieee80211_channel channels[ARRAY_SIZE(rndis_channels)];
-	struct ieee80211_rate rates[ARRAY_SIZE(rndis_rates)];
-	u32 cipher_suites[ARRAY_SIZE(rndis_cipher_suites)];
-
-	int device_type;
-	int caps;
-	int multicast_size;
-
-	/* module parameters */
-	char param_country[4];
-	int  param_frameburst;
-	int  param_afterburner;
-	int  param_power_save;
-	int  param_power_output;
-	int  param_roamtrigger;
-	int  param_roamdelta;
-	u32  param_workaround_interval;
-
-	/* hardware state */
-	bool radio_on;
-	int power_mode;
-	int infra_mode;
-	bool connected;
-	u8 bssid[ETH_ALEN];
-	u32 current_command_oid;
-
-	/* encryption stuff */
-	u8 encr_tx_key_index;
-	struct rndis_wlan_encr_key encr_keys[RNDIS_WLAN_NUM_KEYS];
-	int  wpa_version;
-
-	u8 command_buffer[COMMAND_BUFFER_SIZE];
-};
-
-/*
- * cfg80211 ops
- */
-static int rndis_change_virtual_intf(struct wiphy *wiphy,
-					struct net_device *dev,
-					enum nl80211_iftype type,
-					struct vif_params *params);
-
-static int rndis_scan(struct wiphy *wiphy,
-			struct cfg80211_scan_request *request);
-
-static int rndis_set_wiphy_params(struct wiphy *wiphy, u32 changed);
-
-static int rndis_set_tx_power(struct wiphy *wiphy,
-			      struct wireless_dev *wdev,
-			      enum nl80211_tx_power_setting type,
-			      int mbm);
-static int rndis_get_tx_power(struct wiphy *wiphy,
-			      struct wireless_dev *wdev,
-			      int *dbm);
-
-static int rndis_connect(struct wiphy *wiphy, struct net_device *dev,
-				struct cfg80211_connect_params *sme);
-
-static int rndis_disconnect(struct wiphy *wiphy, struct net_device *dev,
-				u16 reason_code);
-
-static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev,
-					struct cfg80211_ibss_params *params);
-
-static int rndis_leave_ibss(struct wiphy *wiphy, struct net_device *dev);
-
-static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
-			 int link_id,  u8 key_index, bool pairwise,
-			 const u8 *mac_addr, struct key_params *params);
-
-static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev,
-			 int link_id, u8 key_index, bool pairwise,
-			 const u8 *mac_addr);
-
-static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev,
-				 int link_id, u8 key_index, bool unicast,
-				 bool multicast);
-
-static int rndis_get_station(struct wiphy *wiphy, struct net_device *dev,
-			     const u8 *mac, struct station_info *sinfo);
-
-static int rndis_dump_station(struct wiphy *wiphy, struct net_device *dev,
-			       int idx, u8 *mac, struct station_info *sinfo);
-
-static int rndis_set_pmksa(struct wiphy *wiphy, struct net_device *netdev,
-				struct cfg80211_pmksa *pmksa);
-
-static int rndis_del_pmksa(struct wiphy *wiphy, struct net_device *netdev,
-				struct cfg80211_pmksa *pmksa);
-
-static int rndis_flush_pmksa(struct wiphy *wiphy, struct net_device *netdev);
-
-static int rndis_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev,
-				bool enabled, int timeout);
-
-static int rndis_set_cqm_rssi_config(struct wiphy *wiphy,
-					struct net_device *dev,
-					s32 rssi_thold, u32 rssi_hyst);
-
-static const struct cfg80211_ops rndis_config_ops = {
-	.change_virtual_intf = rndis_change_virtual_intf,
-	.scan = rndis_scan,
-	.set_wiphy_params = rndis_set_wiphy_params,
-	.set_tx_power = rndis_set_tx_power,
-	.get_tx_power = rndis_get_tx_power,
-	.connect = rndis_connect,
-	.disconnect = rndis_disconnect,
-	.join_ibss = rndis_join_ibss,
-	.leave_ibss = rndis_leave_ibss,
-	.add_key = rndis_add_key,
-	.del_key = rndis_del_key,
-	.set_default_key = rndis_set_default_key,
-	.get_station = rndis_get_station,
-	.dump_station = rndis_dump_station,
-	.set_pmksa = rndis_set_pmksa,
-	.del_pmksa = rndis_del_pmksa,
-	.flush_pmksa = rndis_flush_pmksa,
-	.set_power_mgmt = rndis_set_power_mgmt,
-	.set_cqm_rssi_config = rndis_set_cqm_rssi_config,
-};
-
-static void *rndis_wiphy_privid = &rndis_wiphy_privid;
-
-
-static struct rndis_wlan_private *get_rndis_wlan_priv(struct usbnet *dev)
-{
-	return (struct rndis_wlan_private *)dev->driver_priv;
-}
-
-static u32 get_bcm4320_power_dbm(struct rndis_wlan_private *priv)
-{
-	switch (priv->param_power_output) {
-	default:
-	case 3:
-		return BCM4320_DEFAULT_TXPOWER_DBM_100;
-	case 2:
-		return BCM4320_DEFAULT_TXPOWER_DBM_75;
-	case 1:
-		return BCM4320_DEFAULT_TXPOWER_DBM_50;
-	case 0:
-		return BCM4320_DEFAULT_TXPOWER_DBM_25;
-	}
-}
-
-static bool is_wpa_key(struct rndis_wlan_private *priv, u8 idx)
-{
-	int cipher = priv->encr_keys[idx].cipher;
-
-	return (cipher == WLAN_CIPHER_SUITE_CCMP ||
-		cipher == WLAN_CIPHER_SUITE_TKIP);
-}
-
-static int rndis_cipher_to_alg(u32 cipher)
-{
-	switch (cipher) {
-	default:
-		return RNDIS_WLAN_ALG_NONE;
-	case WLAN_CIPHER_SUITE_WEP40:
-	case WLAN_CIPHER_SUITE_WEP104:
-		return RNDIS_WLAN_ALG_WEP;
-	case WLAN_CIPHER_SUITE_TKIP:
-		return RNDIS_WLAN_ALG_TKIP;
-	case WLAN_CIPHER_SUITE_CCMP:
-		return RNDIS_WLAN_ALG_CCMP;
-	}
-}
-
-static int rndis_akm_suite_to_key_mgmt(u32 akm_suite)
-{
-	switch (akm_suite) {
-	default:
-		return RNDIS_WLAN_KEY_MGMT_NONE;
-	case WLAN_AKM_SUITE_8021X:
-		return RNDIS_WLAN_KEY_MGMT_802_1X;
-	case WLAN_AKM_SUITE_PSK:
-		return RNDIS_WLAN_KEY_MGMT_PSK;
-	}
-}
-
-#ifdef DEBUG
-static const char *oid_to_string(u32 oid)
-{
-	switch (oid) {
-#define OID_STR(oid) case oid: return(#oid)
-		/* from rndis_host.h */
-		OID_STR(RNDIS_OID_802_3_PERMANENT_ADDRESS);
-		OID_STR(RNDIS_OID_GEN_MAXIMUM_FRAME_SIZE);
-		OID_STR(RNDIS_OID_GEN_CURRENT_PACKET_FILTER);
-		OID_STR(RNDIS_OID_GEN_PHYSICAL_MEDIUM);
-
-		/* from rndis_wlan.c */
-		OID_STR(RNDIS_OID_GEN_LINK_SPEED);
-		OID_STR(RNDIS_OID_GEN_RNDIS_CONFIG_PARAMETER);
-
-		OID_STR(RNDIS_OID_GEN_XMIT_OK);
-		OID_STR(RNDIS_OID_GEN_RCV_OK);
-		OID_STR(RNDIS_OID_GEN_XMIT_ERROR);
-		OID_STR(RNDIS_OID_GEN_RCV_ERROR);
-		OID_STR(RNDIS_OID_GEN_RCV_NO_BUFFER);
-
-		OID_STR(RNDIS_OID_802_3_CURRENT_ADDRESS);
-		OID_STR(RNDIS_OID_802_3_MULTICAST_LIST);
-		OID_STR(RNDIS_OID_802_3_MAXIMUM_LIST_SIZE);
-
-		OID_STR(RNDIS_OID_802_11_BSSID);
-		OID_STR(RNDIS_OID_802_11_SSID);
-		OID_STR(RNDIS_OID_802_11_INFRASTRUCTURE_MODE);
-		OID_STR(RNDIS_OID_802_11_ADD_WEP);
-		OID_STR(RNDIS_OID_802_11_REMOVE_WEP);
-		OID_STR(RNDIS_OID_802_11_DISASSOCIATE);
-		OID_STR(RNDIS_OID_802_11_AUTHENTICATION_MODE);
-		OID_STR(RNDIS_OID_802_11_PRIVACY_FILTER);
-		OID_STR(RNDIS_OID_802_11_BSSID_LIST_SCAN);
-		OID_STR(RNDIS_OID_802_11_ENCRYPTION_STATUS);
-		OID_STR(RNDIS_OID_802_11_ADD_KEY);
-		OID_STR(RNDIS_OID_802_11_REMOVE_KEY);
-		OID_STR(RNDIS_OID_802_11_ASSOCIATION_INFORMATION);
-		OID_STR(RNDIS_OID_802_11_CAPABILITY);
-		OID_STR(RNDIS_OID_802_11_PMKID);
-		OID_STR(RNDIS_OID_802_11_NETWORK_TYPES_SUPPORTED);
-		OID_STR(RNDIS_OID_802_11_NETWORK_TYPE_IN_USE);
-		OID_STR(RNDIS_OID_802_11_TX_POWER_LEVEL);
-		OID_STR(RNDIS_OID_802_11_RSSI);
-		OID_STR(RNDIS_OID_802_11_RSSI_TRIGGER);
-		OID_STR(RNDIS_OID_802_11_FRAGMENTATION_THRESHOLD);
-		OID_STR(RNDIS_OID_802_11_RTS_THRESHOLD);
-		OID_STR(RNDIS_OID_802_11_SUPPORTED_RATES);
-		OID_STR(RNDIS_OID_802_11_CONFIGURATION);
-		OID_STR(RNDIS_OID_802_11_POWER_MODE);
-		OID_STR(RNDIS_OID_802_11_BSSID_LIST);
-#undef OID_STR
-	}
-
-	return "?";
-}
-#else
-static const char *oid_to_string(u32 oid)
-{
-	return "?";
-}
-#endif
-
-/* translate error code */
-static int rndis_error_status(__le32 rndis_status)
-{
-	int ret = -EINVAL;
-	switch (le32_to_cpu(rndis_status)) {
-	case RNDIS_STATUS_SUCCESS:
-		ret = 0;
-		break;
-	case RNDIS_STATUS_FAILURE:
-	case RNDIS_STATUS_INVALID_DATA:
-		ret = -EINVAL;
-		break;
-	case RNDIS_STATUS_NOT_SUPPORTED:
-		ret = -EOPNOTSUPP;
-		break;
-	case RNDIS_STATUS_ADAPTER_NOT_READY:
-	case RNDIS_STATUS_ADAPTER_NOT_OPEN:
-		ret = -EBUSY;
-		break;
-	}
-	return ret;
-}
-
-static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(dev);
-	union {
-		void			*buf;
-		struct rndis_msg_hdr	*header;
-		struct rndis_query	*get;
-		struct rndis_query_c	*get_c;
-	} u;
-	int ret;
-	size_t buflen, resplen, respoffs, copylen;
-
-	buflen = *len + sizeof(*u.get);
-	if (buflen < CONTROL_BUFFER_SIZE)
-		buflen = CONTROL_BUFFER_SIZE;
-
-	if (buflen > COMMAND_BUFFER_SIZE) {
-		u.buf = kmalloc(buflen, GFP_KERNEL);
-		if (!u.buf)
-			return -ENOMEM;
-	} else {
-		u.buf = priv->command_buffer;
-	}
-
-	mutex_lock(&priv->command_lock);
-
-	memset(u.get, 0, sizeof *u.get);
-	u.get->msg_type = cpu_to_le32(RNDIS_MSG_QUERY);
-	u.get->msg_len = cpu_to_le32(sizeof *u.get);
-	u.get->oid = cpu_to_le32(oid);
-
-	priv->current_command_oid = oid;
-	ret = rndis_command(dev, u.header, buflen);
-	priv->current_command_oid = 0;
-	if (ret < 0)
-		netdev_dbg(dev->net, "%s(%s): rndis_command() failed, %d (%08x)\n",
-			   __func__, oid_to_string(oid), ret,
-			   le32_to_cpu(u.get_c->status));
-
-	if (ret == 0) {
-		resplen = le32_to_cpu(u.get_c->len);
-		respoffs = le32_to_cpu(u.get_c->offset) + 8;
-
-		if (respoffs > buflen) {
-			/* Device returned data offset outside buffer, error. */
-			netdev_dbg(dev->net,
-				   "%s(%s): received invalid data offset: %zu > %zu\n",
-				   __func__, oid_to_string(oid), respoffs, buflen);
-
-			ret = -EINVAL;
-			goto exit_unlock;
-		}
-
-		copylen = min(resplen, buflen - respoffs);
-
-		if (copylen > *len)
-			copylen = *len;
-
-		memcpy(data, u.buf + respoffs, copylen);
-
-		*len = resplen;
-
-		ret = rndis_error_status(u.get_c->status);
-		if (ret < 0)
-			netdev_dbg(dev->net, "%s(%s): device returned error,  0x%08x (%d)\n",
-				   __func__, oid_to_string(oid),
-				   le32_to_cpu(u.get_c->status), ret);
-	}
-
-exit_unlock:
-	mutex_unlock(&priv->command_lock);
-
-	if (u.buf != priv->command_buffer)
-		kfree(u.buf);
-	return ret;
-}
-
-static int rndis_set_oid(struct usbnet *dev, u32 oid, const void *data,
-			 int len)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(dev);
-	union {
-		void			*buf;
-		struct rndis_msg_hdr	*header;
-		struct rndis_set	*set;
-		struct rndis_set_c	*set_c;
-	} u;
-	int ret, buflen;
-
-	buflen = len + sizeof(*u.set);
-	if (buflen < CONTROL_BUFFER_SIZE)
-		buflen = CONTROL_BUFFER_SIZE;
-
-	if (buflen > COMMAND_BUFFER_SIZE) {
-		u.buf = kmalloc(buflen, GFP_KERNEL);
-		if (!u.buf)
-			return -ENOMEM;
-	} else {
-		u.buf = priv->command_buffer;
-	}
-
-	mutex_lock(&priv->command_lock);
-
-	memset(u.set, 0, sizeof *u.set);
-	u.set->msg_type = cpu_to_le32(RNDIS_MSG_SET);
-	u.set->msg_len = cpu_to_le32(sizeof(*u.set) + len);
-	u.set->oid = cpu_to_le32(oid);
-	u.set->len = cpu_to_le32(len);
-	u.set->offset = cpu_to_le32(sizeof(*u.set) - 8);
-	u.set->handle = cpu_to_le32(0);
-	memcpy(u.buf + sizeof(*u.set), data, len);
-
-	priv->current_command_oid = oid;
-	ret = rndis_command(dev, u.header, buflen);
-	priv->current_command_oid = 0;
-	if (ret < 0)
-		netdev_dbg(dev->net, "%s(%s): rndis_command() failed, %d (%08x)\n",
-			   __func__, oid_to_string(oid), ret,
-			   le32_to_cpu(u.set_c->status));
-
-	if (ret == 0) {
-		ret = rndis_error_status(u.set_c->status);
-
-		if (ret < 0)
-			netdev_dbg(dev->net, "%s(%s): device returned error, 0x%08x (%d)\n",
-				   __func__, oid_to_string(oid),
-				   le32_to_cpu(u.set_c->status), ret);
-	}
-
-	mutex_unlock(&priv->command_lock);
-
-	if (u.buf != priv->command_buffer)
-		kfree(u.buf);
-	return ret;
-}
-
-static int rndis_reset(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct rndis_reset *reset;
-	int ret;
-
-	mutex_lock(&priv->command_lock);
-
-	reset = (void *)priv->command_buffer;
-	memset(reset, 0, sizeof(*reset));
-	reset->msg_type = cpu_to_le32(RNDIS_MSG_RESET);
-	reset->msg_len = cpu_to_le32(sizeof(*reset));
-	priv->current_command_oid = 0;
-	ret = rndis_command(usbdev, (void *)reset, CONTROL_BUFFER_SIZE);
-
-	mutex_unlock(&priv->command_lock);
-
-	if (ret < 0)
-		return ret;
-	return 0;
-}
-
-/*
- * Specs say that we can only set config parameters only soon after device
- * initialization.
- *   value_type: 0 = u32, 2 = unicode string
- */
-static int rndis_set_config_parameter(struct usbnet *dev, char *param,
-						int value_type, void *value)
-{
-	struct ndis_config_param *infobuf;
-	int value_len, info_len, param_len, ret, i;
-	__le16 *unibuf;
-	__le32 *dst_value;
-
-	if (value_type == 0)
-		value_len = sizeof(__le32);
-	else if (value_type == 2)
-		value_len = strlen(value) * sizeof(__le16);
-	else
-		return -EINVAL;
-
-	param_len = strlen(param) * sizeof(__le16);
-	info_len = sizeof(*infobuf) + param_len + value_len;
-
-#ifdef DEBUG
-	info_len += 12;
-#endif
-	infobuf = kmalloc(info_len, GFP_KERNEL);
-	if (!infobuf)
-		return -ENOMEM;
-
-#ifdef DEBUG
-	info_len -= 12;
-	/* extra 12 bytes are for padding (debug output) */
-	memset(infobuf, 0xCC, info_len + 12);
-#endif
-
-	if (value_type == 2)
-		netdev_dbg(dev->net, "setting config parameter: %s, value: %s\n",
-			   param, (u8 *)value);
-	else
-		netdev_dbg(dev->net, "setting config parameter: %s, value: %d\n",
-			   param, *(u32 *)value);
-
-	infobuf->name_offs = cpu_to_le32(sizeof(*infobuf));
-	infobuf->name_length = cpu_to_le32(param_len);
-	infobuf->type = cpu_to_le32(value_type);
-	infobuf->value_offs = cpu_to_le32(sizeof(*infobuf) + param_len);
-	infobuf->value_length = cpu_to_le32(value_len);
-
-	/* simple string to unicode string conversion */
-	unibuf = (void *)infobuf + sizeof(*infobuf);
-	for (i = 0; i < param_len / sizeof(__le16); i++)
-		unibuf[i] = cpu_to_le16(param[i]);
-
-	if (value_type == 2) {
-		unibuf = (void *)infobuf + sizeof(*infobuf) + param_len;
-		for (i = 0; i < value_len / sizeof(__le16); i++)
-			unibuf[i] = cpu_to_le16(((u8 *)value)[i]);
-	} else {
-		dst_value = (void *)infobuf + sizeof(*infobuf) + param_len;
-		*dst_value = cpu_to_le32(*(u32 *)value);
-	}
-
-#ifdef DEBUG
-	netdev_dbg(dev->net, "info buffer (len: %d)\n", info_len);
-	for (i = 0; i < info_len; i += 12) {
-		u32 *tmp = (u32 *)((u8 *)infobuf + i);
-		netdev_dbg(dev->net, "%08X:%08X:%08X\n",
-			   cpu_to_be32(tmp[0]),
-			   cpu_to_be32(tmp[1]),
-			   cpu_to_be32(tmp[2]));
-	}
-#endif
-
-	ret = rndis_set_oid(dev, RNDIS_OID_GEN_RNDIS_CONFIG_PARAMETER,
-							infobuf, info_len);
-	if (ret != 0)
-		netdev_dbg(dev->net, "setting rndis config parameter failed, %d\n",
-			   ret);
-
-	kfree(infobuf);
-	return ret;
-}
-
-static int rndis_set_config_parameter_str(struct usbnet *dev,
-						char *param, char *value)
-{
-	return rndis_set_config_parameter(dev, param, 2, value);
-}
-
-/*
- * data conversion functions
- */
-static int level_to_qual(int level)
-{
-	int qual = 100 * (level - WL_NOISE) / (WL_SIGMAX - WL_NOISE);
-	return qual >= 0 ? (qual <= 100 ? qual : 100) : 0;
-}
-
-/*
- * common functions
- */
-static int set_infra_mode(struct usbnet *usbdev, int mode);
-static void restore_keys(struct usbnet *usbdev);
-static int rndis_check_bssid_list(struct usbnet *usbdev, u8 *match_bssid,
-					bool *matched);
-
-static int rndis_start_bssid_list_scan(struct usbnet *usbdev)
-{
-	__le32 tmp;
-
-	/* Note: RNDIS_OID_802_11_BSSID_LIST_SCAN clears internal BSS list. */
-	tmp = cpu_to_le32(1);
-	return rndis_set_oid(usbdev, RNDIS_OID_802_11_BSSID_LIST_SCAN, &tmp,
-							sizeof(tmp));
-}
-
-static int set_essid(struct usbnet *usbdev, struct ndis_80211_ssid *ssid)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	int ret;
-
-	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_SSID,
-			    ssid, sizeof(*ssid));
-	if (ret < 0) {
-		netdev_warn(usbdev->net, "setting SSID failed (%08X)\n", ret);
-		return ret;
-	}
-	if (ret == 0) {
-		priv->radio_on = true;
-		netdev_dbg(usbdev->net, "%s(): radio_on = true\n", __func__);
-	}
-
-	return ret;
-}
-
-static int set_bssid(struct usbnet *usbdev, const u8 *bssid)
-{
-	int ret;
-
-	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_BSSID,
-			    bssid, ETH_ALEN);
-	if (ret < 0) {
-		netdev_warn(usbdev->net, "setting BSSID[%pM] failed (%08X)\n",
-			    bssid, ret);
-		return ret;
-	}
-
-	return ret;
-}
-
-static int clear_bssid(struct usbnet *usbdev)
-{
-	static const u8 broadcast_mac[ETH_ALEN] = {
-		0xff, 0xff, 0xff, 0xff, 0xff, 0xff
-	};
-
-	return set_bssid(usbdev, broadcast_mac);
-}
-
-static int get_bssid(struct usbnet *usbdev, u8 bssid[ETH_ALEN])
-{
-	int ret, len;
-
-	len = ETH_ALEN;
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_BSSID,
-			      bssid, &len);
-
-	if (ret != 0)
-		eth_zero_addr(bssid);
-
-	return ret;
-}
-
-static int get_association_info(struct usbnet *usbdev,
-			struct ndis_80211_assoc_info *info, int len)
-{
-	return rndis_query_oid(usbdev,
-			RNDIS_OID_802_11_ASSOCIATION_INFORMATION,
-			info, &len);
-}
-
-static bool is_associated(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	u8 bssid[ETH_ALEN];
-
-	if (!priv->radio_on)
-		return false;
-
-	return (get_bssid(usbdev, bssid) == 0 && !is_zero_ether_addr(bssid));
-}
-
-static int disassociate(struct usbnet *usbdev, bool reset_ssid)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ndis_80211_ssid ssid;
-	int i, ret = 0;
-
-	if (priv->radio_on) {
-		ret = rndis_set_oid(usbdev,
-				RNDIS_OID_802_11_DISASSOCIATE,
-				NULL, 0);
-		if (ret == 0) {
-			priv->radio_on = false;
-			netdev_dbg(usbdev->net, "%s(): radio_on = false\n",
-				   __func__);
-
-			if (reset_ssid)
-				msleep(100);
-		}
-	}
-
-	/* disassociate causes radio to be turned off; if reset_ssid
-	 * is given, set random ssid to enable radio */
-	if (reset_ssid) {
-		/* Set device to infrastructure mode so we don't get ad-hoc
-		 * 'media connect' indications with the random ssid.
-		 */
-		set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA);
-
-		ssid.length = cpu_to_le32(sizeof(ssid.essid));
-		get_random_bytes(&ssid.essid[2], sizeof(ssid.essid)-2);
-		ssid.essid[0] = 0x1;
-		ssid.essid[1] = 0xff;
-		for (i = 2; i < sizeof(ssid.essid); i++)
-			ssid.essid[i] = 0x1 + (ssid.essid[i] * 0xfe / 0xff);
-		ret = set_essid(usbdev, &ssid);
-	}
-	return ret;
-}
-
-static int set_auth_mode(struct usbnet *usbdev, u32 wpa_version,
-				enum nl80211_auth_type auth_type, int keymgmt)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	__le32 tmp;
-	int auth_mode, ret;
-
-	netdev_dbg(usbdev->net, "%s(): wpa_version=0x%x authalg=0x%x keymgmt=0x%x\n",
-		   __func__, wpa_version, auth_type, keymgmt);
-
-	if (wpa_version & NL80211_WPA_VERSION_2) {
-		if (keymgmt & RNDIS_WLAN_KEY_MGMT_802_1X)
-			auth_mode = NDIS_80211_AUTH_WPA2;
-		else
-			auth_mode = NDIS_80211_AUTH_WPA2_PSK;
-	} else if (wpa_version & NL80211_WPA_VERSION_1) {
-		if (keymgmt & RNDIS_WLAN_KEY_MGMT_802_1X)
-			auth_mode = NDIS_80211_AUTH_WPA;
-		else if (keymgmt & RNDIS_WLAN_KEY_MGMT_PSK)
-			auth_mode = NDIS_80211_AUTH_WPA_PSK;
-		else
-			auth_mode = NDIS_80211_AUTH_WPA_NONE;
-	} else if (auth_type == NL80211_AUTHTYPE_SHARED_KEY)
-		auth_mode = NDIS_80211_AUTH_SHARED;
-	else if (auth_type == NL80211_AUTHTYPE_OPEN_SYSTEM)
-		auth_mode = NDIS_80211_AUTH_OPEN;
-	else if (auth_type == NL80211_AUTHTYPE_AUTOMATIC)
-		auth_mode = NDIS_80211_AUTH_AUTO_SWITCH;
-	else
-		return -ENOTSUPP;
-
-	tmp = cpu_to_le32(auth_mode);
-	ret = rndis_set_oid(usbdev,
-			    RNDIS_OID_802_11_AUTHENTICATION_MODE,
-			    &tmp, sizeof(tmp));
-	if (ret != 0) {
-		netdev_warn(usbdev->net, "setting auth mode failed (%08X)\n",
-			    ret);
-		return ret;
-	}
-
-	priv->wpa_version = wpa_version;
-
-	return 0;
-}
-
-static int set_priv_filter(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	__le32 tmp;
-
-	netdev_dbg(usbdev->net, "%s(): wpa_version=0x%x\n",
-		   __func__, priv->wpa_version);
-
-	if (priv->wpa_version & NL80211_WPA_VERSION_2 ||
-	    priv->wpa_version & NL80211_WPA_VERSION_1)
-		tmp = cpu_to_le32(NDIS_80211_PRIV_8021X_WEP);
-	else
-		tmp = cpu_to_le32(NDIS_80211_PRIV_ACCEPT_ALL);
-
-	return rndis_set_oid(usbdev,
-			     RNDIS_OID_802_11_PRIVACY_FILTER, &tmp,
-			     sizeof(tmp));
-}
-
-static int set_encr_mode(struct usbnet *usbdev, int pairwise, int groupwise)
-{
-	__le32 tmp;
-	int encr_mode, ret;
-
-	netdev_dbg(usbdev->net, "%s(): cipher_pair=0x%x cipher_group=0x%x\n",
-		   __func__, pairwise, groupwise);
-
-	if (pairwise & RNDIS_WLAN_ALG_CCMP)
-		encr_mode = NDIS_80211_ENCR_CCMP_ENABLED;
-	else if (pairwise & RNDIS_WLAN_ALG_TKIP)
-		encr_mode = NDIS_80211_ENCR_TKIP_ENABLED;
-	else if (pairwise & RNDIS_WLAN_ALG_WEP)
-		encr_mode = NDIS_80211_ENCR_WEP_ENABLED;
-	else if (groupwise & RNDIS_WLAN_ALG_CCMP)
-		encr_mode = NDIS_80211_ENCR_CCMP_ENABLED;
-	else if (groupwise & RNDIS_WLAN_ALG_TKIP)
-		encr_mode = NDIS_80211_ENCR_TKIP_ENABLED;
-	else
-		encr_mode = NDIS_80211_ENCR_DISABLED;
-
-	tmp = cpu_to_le32(encr_mode);
-	ret = rndis_set_oid(usbdev,
-			RNDIS_OID_802_11_ENCRYPTION_STATUS, &tmp,
-			sizeof(tmp));
-	if (ret != 0) {
-		netdev_warn(usbdev->net, "setting encr mode failed (%08X)\n",
-			    ret);
-		return ret;
-	}
-
-	return 0;
-}
-
-static int set_infra_mode(struct usbnet *usbdev, int mode)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	__le32 tmp;
-	int ret;
-
-	netdev_dbg(usbdev->net, "%s(): infra_mode=0x%x\n",
-		   __func__, priv->infra_mode);
-
-	tmp = cpu_to_le32(mode);
-	ret = rndis_set_oid(usbdev,
-			    RNDIS_OID_802_11_INFRASTRUCTURE_MODE,
-			    &tmp, sizeof(tmp));
-	if (ret != 0) {
-		netdev_warn(usbdev->net, "setting infra mode failed (%08X)\n",
-			    ret);
-		return ret;
-	}
-
-	/* NDIS drivers clear keys when infrastructure mode is
-	 * changed. But Linux tools assume otherwise. So set the
-	 * keys */
-	restore_keys(usbdev);
-
-	priv->infra_mode = mode;
-	return 0;
-}
-
-static int set_rts_threshold(struct usbnet *usbdev, u32 rts_threshold)
-{
-	__le32 tmp;
-
-	netdev_dbg(usbdev->net, "%s(): %i\n", __func__, rts_threshold);
-
-	if (rts_threshold == -1 || rts_threshold > 2347)
-		rts_threshold = 2347;
-
-	tmp = cpu_to_le32(rts_threshold);
-	return rndis_set_oid(usbdev,
-			     RNDIS_OID_802_11_RTS_THRESHOLD,
-			     &tmp, sizeof(tmp));
-}
-
-static int set_frag_threshold(struct usbnet *usbdev, u32 frag_threshold)
-{
-	__le32 tmp;
-
-	netdev_dbg(usbdev->net, "%s(): %i\n", __func__, frag_threshold);
-
-	if (frag_threshold < 256 || frag_threshold > 2346)
-		frag_threshold = 2346;
-
-	tmp = cpu_to_le32(frag_threshold);
-	return rndis_set_oid(usbdev,
-			RNDIS_OID_802_11_FRAGMENTATION_THRESHOLD,
-			&tmp, sizeof(tmp));
-}
-
-static void set_default_iw_params(struct usbnet *usbdev)
-{
-	set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA);
-	set_auth_mode(usbdev, 0, NL80211_AUTHTYPE_OPEN_SYSTEM,
-						RNDIS_WLAN_KEY_MGMT_NONE);
-	set_priv_filter(usbdev);
-	set_encr_mode(usbdev, RNDIS_WLAN_ALG_NONE, RNDIS_WLAN_ALG_NONE);
-}
-
-static int deauthenticate(struct usbnet *usbdev)
-{
-	int ret;
-
-	ret = disassociate(usbdev, true);
-	set_default_iw_params(usbdev);
-	return ret;
-}
-
-static int set_channel(struct usbnet *usbdev, int channel)
-{
-	struct ndis_80211_conf config;
-	unsigned int dsconfig;
-	int len, ret;
-
-	netdev_dbg(usbdev->net, "%s(%d)\n", __func__, channel);
-
-	/* this OID is valid only when not associated */
-	if (is_associated(usbdev))
-		return 0;
-
-	dsconfig = 1000 *
-		ieee80211_channel_to_frequency(channel, NL80211_BAND_2GHZ);
-
-	len = sizeof(config);
-	ret = rndis_query_oid(usbdev,
-			RNDIS_OID_802_11_CONFIGURATION,
-			&config, &len);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "%s(): querying configuration failed\n",
-			   __func__);
-		return ret;
-	}
-
-	config.ds_config = cpu_to_le32(dsconfig);
-	ret = rndis_set_oid(usbdev,
-			RNDIS_OID_802_11_CONFIGURATION,
-			&config, sizeof(config));
-
-	netdev_dbg(usbdev->net, "%s(): %d -> %d\n", __func__, channel, ret);
-
-	return ret;
-}
-
-static struct ieee80211_channel *get_current_channel(struct usbnet *usbdev,
-						     u32 *beacon_period)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ieee80211_channel *channel;
-	struct ndis_80211_conf config;
-	int len, ret;
-
-	/* Get channel and beacon interval */
-	len = sizeof(config);
-	ret = rndis_query_oid(usbdev,
-			RNDIS_OID_802_11_CONFIGURATION,
-			&config, &len);
-	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_CONFIGURATION -> %d\n",
-				__func__, ret);
-	if (ret < 0)
-		return NULL;
-
-	channel = ieee80211_get_channel(priv->wdev.wiphy,
-				KHZ_TO_MHZ(le32_to_cpu(config.ds_config)));
-	if (!channel)
-		return NULL;
-
-	if (beacon_period)
-		*beacon_period = le32_to_cpu(config.beacon_period);
-	return channel;
-}
-
-/* index must be 0 - N, as per NDIS  */
-static int add_wep_key(struct usbnet *usbdev, const u8 *key, int key_len,
-								u8 index)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ndis_80211_wep_key ndis_key;
-	u32 cipher;
-	int ret;
-
-	netdev_dbg(usbdev->net, "%s(idx: %d, len: %d)\n",
-		   __func__, index, key_len);
-
-	if (index >= RNDIS_WLAN_NUM_KEYS)
-		return -EINVAL;
-
-	if (key_len == 5)
-		cipher = WLAN_CIPHER_SUITE_WEP40;
-	else if (key_len == 13)
-		cipher = WLAN_CIPHER_SUITE_WEP104;
-	else
-		return -EINVAL;
-
-	memset(&ndis_key, 0, sizeof(ndis_key));
-
-	ndis_key.size = cpu_to_le32(sizeof(ndis_key));
-	ndis_key.length = cpu_to_le32(key_len);
-	ndis_key.index = cpu_to_le32(index);
-	memcpy(&ndis_key.material, key, key_len);
-
-	if (index == priv->encr_tx_key_index) {
-		ndis_key.index |= NDIS_80211_ADDWEP_TRANSMIT_KEY;
-		ret = set_encr_mode(usbdev, RNDIS_WLAN_ALG_WEP,
-							RNDIS_WLAN_ALG_NONE);
-		if (ret)
-			netdev_warn(usbdev->net, "encryption couldn't be enabled (%08X)\n",
-				    ret);
-	}
-
-	ret = rndis_set_oid(usbdev,
-			RNDIS_OID_802_11_ADD_WEP, &ndis_key,
-			sizeof(ndis_key));
-	if (ret != 0) {
-		netdev_warn(usbdev->net, "adding encryption key %d failed (%08X)\n",
-			    index + 1, ret);
-		return ret;
-	}
-
-	priv->encr_keys[index].len = key_len;
-	priv->encr_keys[index].cipher = cipher;
-	memcpy(&priv->encr_keys[index].material, key, key_len);
-	eth_broadcast_addr(priv->encr_keys[index].bssid);
-
-	return 0;
-}
-
-static int add_wpa_key(struct usbnet *usbdev, const u8 *key, int key_len,
-			u8 index, const u8 *addr, const u8 *rx_seq,
-			int seq_len, u32 cipher, __le32 flags)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ndis_80211_key ndis_key;
-	bool is_addr_ok;
-	int ret;
-
-	if (index >= RNDIS_WLAN_NUM_KEYS) {
-		netdev_dbg(usbdev->net, "%s(): index out of range (%i)\n",
-			   __func__, index);
-		return -EINVAL;
-	}
-	if (key_len > sizeof(ndis_key.material) || key_len < 0) {
-		netdev_dbg(usbdev->net, "%s(): key length out of range (%i)\n",
-			   __func__, key_len);
-		return -EINVAL;
-	}
-	if (flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ) {
-		if (!rx_seq || seq_len <= 0) {
-			netdev_dbg(usbdev->net, "%s(): recv seq flag without buffer\n",
-				   __func__);
-			return -EINVAL;
-		}
-		if (rx_seq && seq_len > sizeof(ndis_key.rsc)) {
-			netdev_dbg(usbdev->net, "%s(): too big recv seq buffer\n", __func__);
-			return -EINVAL;
-		}
-	}
-
-	is_addr_ok = addr && !is_zero_ether_addr(addr) &&
-					!is_broadcast_ether_addr(addr);
-	if ((flags & NDIS_80211_ADDKEY_PAIRWISE_KEY) && !is_addr_ok) {
-		netdev_dbg(usbdev->net, "%s(): pairwise but bssid invalid (%pM)\n",
-			   __func__, addr);
-		return -EINVAL;
-	}
-
-	netdev_dbg(usbdev->net, "%s(%i): flags:%i%i%i\n",
-		   __func__, index,
-		   !!(flags & NDIS_80211_ADDKEY_TRANSMIT_KEY),
-		   !!(flags & NDIS_80211_ADDKEY_PAIRWISE_KEY),
-		   !!(flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ));
-
-	memset(&ndis_key, 0, sizeof(ndis_key));
-
-	ndis_key.size = cpu_to_le32(sizeof(ndis_key) -
-				sizeof(ndis_key.material) + key_len);
-	ndis_key.length = cpu_to_le32(key_len);
-	ndis_key.index = cpu_to_le32(index) | flags;
-
-	if (cipher == WLAN_CIPHER_SUITE_TKIP && key_len == 32) {
-		/* wpa_supplicant gives us the Michael MIC RX/TX keys in
-		 * different order than NDIS spec, so swap the order here. */
-		memcpy(ndis_key.material, key, 16);
-		memcpy(ndis_key.material + 16, key + 24, 8);
-		memcpy(ndis_key.material + 24, key + 16, 8);
-	} else
-		memcpy(ndis_key.material, key, key_len);
-
-	if (flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ)
-		memcpy(ndis_key.rsc, rx_seq, seq_len);
-
-	if (flags & NDIS_80211_ADDKEY_PAIRWISE_KEY) {
-		/* pairwise key */
-		memcpy(ndis_key.bssid, addr, ETH_ALEN);
-	} else {
-		/* group key */
-		if (priv->infra_mode == NDIS_80211_INFRA_ADHOC)
-			eth_broadcast_addr(ndis_key.bssid);
-		else
-			get_bssid(usbdev, ndis_key.bssid);
-	}
-
-	ret = rndis_set_oid(usbdev,
-			RNDIS_OID_802_11_ADD_KEY, &ndis_key,
-			le32_to_cpu(ndis_key.size));
-	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_ADD_KEY -> %08X\n",
-		   __func__, ret);
-	if (ret != 0)
-		return ret;
-
-	memset(&priv->encr_keys[index], 0, sizeof(priv->encr_keys[index]));
-	priv->encr_keys[index].len = key_len;
-	priv->encr_keys[index].cipher = cipher;
-	memcpy(&priv->encr_keys[index].material, key, key_len);
-	if (flags & NDIS_80211_ADDKEY_PAIRWISE_KEY)
-		memcpy(&priv->encr_keys[index].bssid, ndis_key.bssid, ETH_ALEN);
-	else
-		eth_broadcast_addr(priv->encr_keys[index].bssid);
-
-	if (flags & NDIS_80211_ADDKEY_TRANSMIT_KEY)
-		priv->encr_tx_key_index = index;
-
-	return 0;
-}
-
-static int restore_key(struct usbnet *usbdev, u8 key_idx)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct rndis_wlan_encr_key key;
-
-	if (is_wpa_key(priv, key_idx))
-		return 0;
-
-	key = priv->encr_keys[key_idx];
-
-	netdev_dbg(usbdev->net, "%s(): %i:%i\n", __func__, key_idx, key.len);
-
-	if (key.len == 0)
-		return 0;
-
-	return add_wep_key(usbdev, key.material, key.len, key_idx);
-}
-
-static void restore_keys(struct usbnet *usbdev)
-{
-	int i;
-
-	for (i = 0; i < 4; i++)
-		restore_key(usbdev, i);
-}
-
-static void clear_key(struct rndis_wlan_private *priv, u8 idx)
-{
-	memset(&priv->encr_keys[idx], 0, sizeof(priv->encr_keys[idx]));
-}
-
-/* remove_key is for both wep and wpa */
-static int remove_key(struct usbnet *usbdev, u8 index, const u8 *bssid)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ndis_80211_remove_key remove_key;
-	__le32 keyindex;
-	bool is_wpa;
-	int ret;
-
-	if (index >= RNDIS_WLAN_NUM_KEYS)
-		return -ENOENT;
-
-	if (priv->encr_keys[index].len == 0)
-		return 0;
-
-	is_wpa = is_wpa_key(priv, index);
-
-	netdev_dbg(usbdev->net, "%s(): %i:%s:%i\n",
-		   __func__, index, is_wpa ? "wpa" : "wep",
-		   priv->encr_keys[index].len);
-
-	clear_key(priv, index);
-
-	if (is_wpa) {
-		remove_key.size = cpu_to_le32(sizeof(remove_key));
-		remove_key.index = cpu_to_le32(index);
-		if (bssid) {
-			/* pairwise key */
-			if (!is_broadcast_ether_addr(bssid))
-				remove_key.index |=
-					NDIS_80211_ADDKEY_PAIRWISE_KEY;
-			memcpy(remove_key.bssid, bssid,
-					sizeof(remove_key.bssid));
-		} else
-			memset(remove_key.bssid, 0xff,
-						sizeof(remove_key.bssid));
-
-		ret = rndis_set_oid(usbdev,
-				RNDIS_OID_802_11_REMOVE_KEY,
-				&remove_key, sizeof(remove_key));
-		if (ret != 0)
-			return ret;
-	} else {
-		keyindex = cpu_to_le32(index);
-		ret = rndis_set_oid(usbdev,
-				RNDIS_OID_802_11_REMOVE_WEP,
-				&keyindex, sizeof(keyindex));
-		if (ret != 0) {
-			netdev_warn(usbdev->net,
-				    "removing encryption key %d failed (%08X)\n",
-				    index, ret);
-			return ret;
-		}
-	}
-
-	/* if it is transmit key, disable encryption */
-	if (index == priv->encr_tx_key_index)
-		set_encr_mode(usbdev, RNDIS_WLAN_ALG_NONE, RNDIS_WLAN_ALG_NONE);
-
-	return 0;
-}
-
-static void set_multicast_list(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct netdev_hw_addr *ha;
-	__le32 filter, basefilter;
-	int ret;
-	char *mc_addrs = NULL;
-	int mc_count;
-
-	basefilter = filter = cpu_to_le32(RNDIS_PACKET_TYPE_DIRECTED |
-					  RNDIS_PACKET_TYPE_BROADCAST);
-
-	if (usbdev->net->flags & IFF_PROMISC) {
-		filter |= cpu_to_le32(RNDIS_PACKET_TYPE_PROMISCUOUS |
-				      RNDIS_PACKET_TYPE_ALL_LOCAL);
-	} else if (usbdev->net->flags & IFF_ALLMULTI) {
-		filter |= cpu_to_le32(RNDIS_PACKET_TYPE_ALL_MULTICAST);
-	}
-
-	if (filter != basefilter)
-		goto set_filter;
-
-	/*
-	 * mc_list should be accessed holding the lock, so copy addresses to
-	 * local buffer first.
-	 */
-	netif_addr_lock_bh(usbdev->net);
-	mc_count = netdev_mc_count(usbdev->net);
-	if (mc_count > priv->multicast_size) {
-		filter |= cpu_to_le32(RNDIS_PACKET_TYPE_ALL_MULTICAST);
-	} else if (mc_count) {
-		int i = 0;
-
-		mc_addrs = kmalloc_array(mc_count, ETH_ALEN, GFP_ATOMIC);
-		if (!mc_addrs) {
-			netif_addr_unlock_bh(usbdev->net);
-			return;
-		}
-
-		netdev_for_each_mc_addr(ha, usbdev->net)
-			memcpy(mc_addrs + i++ * ETH_ALEN,
-			       ha->addr, ETH_ALEN);
-	}
-	netif_addr_unlock_bh(usbdev->net);
-
-	if (filter != basefilter)
-		goto set_filter;
-
-	if (mc_count) {
-		ret = rndis_set_oid(usbdev,
-				RNDIS_OID_802_3_MULTICAST_LIST,
-				mc_addrs, mc_count * ETH_ALEN);
-		kfree(mc_addrs);
-		if (ret == 0)
-			filter |= cpu_to_le32(RNDIS_PACKET_TYPE_MULTICAST);
-		else
-			filter |= cpu_to_le32(RNDIS_PACKET_TYPE_ALL_MULTICAST);
-
-		netdev_dbg(usbdev->net, "RNDIS_OID_802_3_MULTICAST_LIST(%d, max: %d) -> %d\n",
-			   mc_count, priv->multicast_size, ret);
-	}
-
-set_filter:
-	ret = rndis_set_oid(usbdev, RNDIS_OID_GEN_CURRENT_PACKET_FILTER, &filter,
-							sizeof(filter));
-	if (ret < 0) {
-		netdev_warn(usbdev->net, "couldn't set packet filter: %08x\n",
-			    le32_to_cpu(filter));
-	}
-
-	netdev_dbg(usbdev->net, "RNDIS_OID_GEN_CURRENT_PACKET_FILTER(%08x) -> %d\n",
-		   le32_to_cpu(filter), ret);
-}
-
-#ifdef DEBUG
-static void debug_print_pmkids(struct usbnet *usbdev,
-				struct ndis_80211_pmkid *pmkids,
-				const char *func_str)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	int i, len, count, max_pmkids, entry_len;
-
-	max_pmkids = priv->wdev.wiphy->max_num_pmkids;
-	len = le32_to_cpu(pmkids->length);
-	count = le32_to_cpu(pmkids->bssid_info_count);
-
-	entry_len = (count > 0) ? (len - sizeof(*pmkids)) / count : -1;
-
-	netdev_dbg(usbdev->net, "%s(): %d PMKIDs (data len: %d, entry len: "
-				"%d)\n", func_str, count, len, entry_len);
-
-	if (count > max_pmkids)
-		count = max_pmkids;
-
-	for (i = 0; i < count; i++) {
-		u32 *tmp = (u32 *)pmkids->bssid_info[i].pmkid;
-
-		netdev_dbg(usbdev->net, "%s():  bssid: %pM, "
-				"pmkid: %08X:%08X:%08X:%08X\n",
-				func_str, pmkids->bssid_info[i].bssid,
-				cpu_to_be32(tmp[0]), cpu_to_be32(tmp[1]),
-				cpu_to_be32(tmp[2]), cpu_to_be32(tmp[3]));
-	}
-}
-#else
-static void debug_print_pmkids(struct usbnet *usbdev,
-				struct ndis_80211_pmkid *pmkids,
-				const char *func_str)
-{
-	return;
-}
-#endif
-
-static struct ndis_80211_pmkid *get_device_pmkids(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ndis_80211_pmkid *pmkids;
-	int len, ret, max_pmkids;
-
-	max_pmkids = priv->wdev.wiphy->max_num_pmkids;
-	len = struct_size(pmkids, bssid_info, max_pmkids);
-
-	pmkids = kzalloc(len, GFP_KERNEL);
-	if (!pmkids)
-		return ERR_PTR(-ENOMEM);
-
-	pmkids->length = cpu_to_le32(len);
-	pmkids->bssid_info_count = cpu_to_le32(max_pmkids);
-
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_PMKID,
-			pmkids, &len);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_PMKID(%d, %d)"
-				" -> %d\n", __func__, len, max_pmkids, ret);
-
-		kfree(pmkids);
-		return ERR_PTR(ret);
-	}
-
-	if (le32_to_cpu(pmkids->bssid_info_count) > max_pmkids)
-		pmkids->bssid_info_count = cpu_to_le32(max_pmkids);
-
-	debug_print_pmkids(usbdev, pmkids, __func__);
-
-	return pmkids;
-}
-
-static int set_device_pmkids(struct usbnet *usbdev,
-				struct ndis_80211_pmkid *pmkids)
-{
-	int ret, len, num_pmkids;
-
-	num_pmkids = le32_to_cpu(pmkids->bssid_info_count);
-	len = struct_size(pmkids, bssid_info, num_pmkids);
-	pmkids->length = cpu_to_le32(len);
-
-	debug_print_pmkids(usbdev, pmkids, __func__);
-
-	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_PMKID, pmkids,
-			    le32_to_cpu(pmkids->length));
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_PMKID(%d, %d) -> %d"
-				"\n", __func__, len, num_pmkids, ret);
-	}
-
-	kfree(pmkids);
-	return ret;
-}
-
-static struct ndis_80211_pmkid *remove_pmkid(struct usbnet *usbdev,
-						struct ndis_80211_pmkid *pmkids,
-						struct cfg80211_pmksa *pmksa,
-						int max_pmkids)
-{
-	int i, err;
-	unsigned int count;
-
-	count = le32_to_cpu(pmkids->bssid_info_count);
-
-	if (count > max_pmkids)
-		count = max_pmkids;
-
-	for (i = 0; i < count; i++)
-		if (ether_addr_equal(pmkids->bssid_info[i].bssid,
-				     pmksa->bssid))
-			break;
-
-	/* pmkid not found */
-	if (i == count) {
-		netdev_dbg(usbdev->net, "%s(): bssid not found (%pM)\n",
-					__func__, pmksa->bssid);
-		err = -ENOENT;
-		goto error;
-	}
-
-	for (; i + 1 < count; i++)
-		pmkids->bssid_info[i] = pmkids->bssid_info[i + 1];
-
-	count--;
-	pmkids->length = cpu_to_le32(struct_size(pmkids, bssid_info, count));
-	pmkids->bssid_info_count = cpu_to_le32(count);
-
-	return pmkids;
-error:
-	kfree(pmkids);
-	return ERR_PTR(err);
-}
-
-static struct ndis_80211_pmkid *update_pmkid(struct usbnet *usbdev,
-						struct ndis_80211_pmkid *pmkids,
-						struct cfg80211_pmksa *pmksa,
-						int max_pmkids)
-{
-	struct ndis_80211_pmkid *new_pmkids;
-	int i, err, newlen;
-	unsigned int count;
-
-	count = le32_to_cpu(pmkids->bssid_info_count);
-
-	if (count > max_pmkids)
-		count = max_pmkids;
-
-	/* update with new pmkid */
-	for (i = 0; i < count; i++) {
-		if (!ether_addr_equal(pmkids->bssid_info[i].bssid,
-				      pmksa->bssid))
-			continue;
-
-		memcpy(pmkids->bssid_info[i].pmkid, pmksa->pmkid,
-								WLAN_PMKID_LEN);
-
-		return pmkids;
-	}
-
-	/* out of space, return error */
-	if (i == max_pmkids) {
-		netdev_dbg(usbdev->net, "%s(): out of space\n", __func__);
-		err = -ENOSPC;
-		goto error;
-	}
-
-	/* add new pmkid */
-	newlen = struct_size(pmkids, bssid_info, count + 1);
-
-	new_pmkids = krealloc(pmkids, newlen, GFP_KERNEL);
-	if (!new_pmkids) {
-		err = -ENOMEM;
-		goto error;
-	}
-	pmkids = new_pmkids;
-
-	pmkids->length = cpu_to_le32(newlen);
-	pmkids->bssid_info_count = cpu_to_le32(count + 1);
-
-	memcpy(pmkids->bssid_info[count].bssid, pmksa->bssid, ETH_ALEN);
-	memcpy(pmkids->bssid_info[count].pmkid, pmksa->pmkid, WLAN_PMKID_LEN);
-
-	return pmkids;
-error:
-	kfree(pmkids);
-	return ERR_PTR(err);
-}
-
-/*
- * cfg80211 ops
- */
-static int rndis_change_virtual_intf(struct wiphy *wiphy,
-					struct net_device *dev,
-					enum nl80211_iftype type,
-					struct vif_params *params)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	int mode;
-
-	switch (type) {
-	case NL80211_IFTYPE_ADHOC:
-		mode = NDIS_80211_INFRA_ADHOC;
-		break;
-	case NL80211_IFTYPE_STATION:
-		mode = NDIS_80211_INFRA_INFRA;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	priv->wdev.iftype = type;
-
-	return set_infra_mode(usbdev, mode);
-}
-
-static int rndis_set_wiphy_params(struct wiphy *wiphy, u32 changed)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	int err;
-
-	if (changed & WIPHY_PARAM_FRAG_THRESHOLD) {
-		err = set_frag_threshold(usbdev, wiphy->frag_threshold);
-		if (err < 0)
-			return err;
-	}
-
-	if (changed & WIPHY_PARAM_RTS_THRESHOLD) {
-		err = set_rts_threshold(usbdev, wiphy->rts_threshold);
-		if (err < 0)
-			return err;
-	}
-
-	return 0;
-}
-
-static int rndis_set_tx_power(struct wiphy *wiphy,
-			      struct wireless_dev *wdev,
-			      enum nl80211_tx_power_setting type,
-			      int mbm)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	netdev_dbg(usbdev->net, "%s(): type:0x%x mbm:%i\n",
-		   __func__, type, mbm);
-
-	if (mbm < 0 || (mbm % 100))
-		return -ENOTSUPP;
-
-	/* Device doesn't support changing txpower after initialization, only
-	 * turn off/on radio. Support 'auto' mode and setting same dBm that is
-	 * currently used.
-	 */
-	if (type == NL80211_TX_POWER_AUTOMATIC ||
-	    MBM_TO_DBM(mbm) == get_bcm4320_power_dbm(priv)) {
-		if (!priv->radio_on)
-			disassociate(usbdev, true); /* turn on radio */
-
-		return 0;
-	}
-
-	return -ENOTSUPP;
-}
-
-static int rndis_get_tx_power(struct wiphy *wiphy,
-			      struct wireless_dev *wdev,
-			      int *dbm)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	*dbm = get_bcm4320_power_dbm(priv);
-
-	netdev_dbg(usbdev->net, "%s(): dbm:%i\n", __func__, *dbm);
-
-	return 0;
-}
-
-#define SCAN_DELAY_JIFFIES (6 * HZ)
-static int rndis_scan(struct wiphy *wiphy,
-			struct cfg80211_scan_request *request)
-{
-	struct net_device *dev = request->wdev->netdev;
-	struct usbnet *usbdev = netdev_priv(dev);
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	int ret;
-	int delay = SCAN_DELAY_JIFFIES;
-
-	netdev_dbg(usbdev->net, "cfg80211.scan\n");
-
-	/* Get current bssid list from device before new scan, as new scan
-	 * clears internal bssid list.
-	 */
-	rndis_check_bssid_list(usbdev, NULL, NULL);
-
-	if (priv->scan_request && priv->scan_request != request)
-		return -EBUSY;
-
-	priv->scan_request = request;
-
-	ret = rndis_start_bssid_list_scan(usbdev);
-	if (ret == 0) {
-		if (priv->device_type == RNDIS_BCM4320A)
-			delay = HZ;
-
-		/* Wait before retrieving scan results from device */
-		queue_delayed_work(priv->workqueue, &priv->scan_work, delay);
-	}
-
-	return ret;
-}
-
-static bool rndis_bss_info_update(struct usbnet *usbdev,
-				  struct ndis_80211_bssid_ex *bssid)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ieee80211_channel *channel;
-	struct cfg80211_bss *bss;
-	s32 signal;
-	u64 timestamp;
-	u16 capability;
-	u16 beacon_interval;
-	struct ndis_80211_fixed_ies *fixed;
-	int ie_len, bssid_len;
-	u8 *ie;
-
-	netdev_dbg(usbdev->net, " found bssid: '%.32s' [%pM], len: %d\n",
-		   bssid->ssid.essid, bssid->mac, le32_to_cpu(bssid->length));
-
-	/* parse bssid structure */
-	bssid_len = le32_to_cpu(bssid->length);
-
-	if (bssid_len < sizeof(struct ndis_80211_bssid_ex) +
-			sizeof(struct ndis_80211_fixed_ies))
-		return NULL;
-
-	fixed = (struct ndis_80211_fixed_ies *)bssid->ies;
-
-	ie = (void *)(bssid->ies + sizeof(struct ndis_80211_fixed_ies));
-	ie_len = min(bssid_len - (int)sizeof(*bssid),
-					(int)le32_to_cpu(bssid->ie_length));
-	ie_len -= sizeof(struct ndis_80211_fixed_ies);
-	if (ie_len < 0)
-		return NULL;
-
-	/* extract data for cfg80211_inform_bss */
-	channel = ieee80211_get_channel(priv->wdev.wiphy,
-			KHZ_TO_MHZ(le32_to_cpu(bssid->config.ds_config)));
-	if (!channel)
-		return NULL;
-
-	signal = level_to_qual(le32_to_cpu(bssid->rssi));
-	timestamp = le64_to_cpu(*(__le64 *)fixed->timestamp);
-	capability = le16_to_cpu(fixed->capabilities);
-	beacon_interval = le16_to_cpu(fixed->beacon_interval);
-
-	bss = cfg80211_inform_bss(priv->wdev.wiphy, channel,
-				  CFG80211_BSS_FTYPE_UNKNOWN, bssid->mac,
-				  timestamp, capability, beacon_interval,
-				  ie, ie_len, signal, GFP_KERNEL);
-	cfg80211_put_bss(priv->wdev.wiphy, bss);
-
-	return (bss != NULL);
-}
-
-static struct ndis_80211_bssid_ex *next_bssid_list_item(
-					struct ndis_80211_bssid_ex *bssid,
-					int *bssid_len, void *buf, int len)
-{
-	void *buf_end, *bssid_end;
-
-	buf_end = (char *)buf + len;
-	bssid_end = (char *)bssid + *bssid_len;
-
-	if ((int)(buf_end - bssid_end) < sizeof(bssid->length)) {
-		*bssid_len = 0;
-		return NULL;
-	} else {
-		bssid = (void *)((char *)bssid + *bssid_len);
-		*bssid_len = le32_to_cpu(bssid->length);
-		return bssid;
-	}
-}
-
-static bool check_bssid_list_item(struct ndis_80211_bssid_ex *bssid,
-				  int bssid_len, void *buf, int len)
-{
-	void *buf_end, *bssid_end;
-
-	if (!bssid || bssid_len <= 0 || bssid_len > len)
-		return false;
-
-	buf_end = (char *)buf + len;
-	bssid_end = (char *)bssid + bssid_len;
-
-	return (int)(buf_end - bssid_end) >= 0 && (int)(bssid_end - buf) >= 0;
-}
-
-static int rndis_check_bssid_list(struct usbnet *usbdev, u8 *match_bssid,
-					bool *matched)
-{
-	void *buf = NULL;
-	struct ndis_80211_bssid_list_ex *bssid_list;
-	struct ndis_80211_bssid_ex *bssid;
-	int ret = -EINVAL, len, count, bssid_len, real_count, new_len;
-
-	netdev_dbg(usbdev->net, "%s()\n", __func__);
-
-	len = CONTROL_BUFFER_SIZE;
-resize_buf:
-	buf = kzalloc(len, GFP_KERNEL);
-	if (!buf) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	/* BSSID-list might have got bigger last time we checked, keep
-	 * resizing until it won't get any bigger.
-	 */
-	new_len = len;
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_BSSID_LIST,
-			      buf, &new_len);
-	if (ret != 0 || new_len < sizeof(struct ndis_80211_bssid_list_ex))
-		goto out;
-
-	if (new_len > len) {
-		len = new_len;
-		kfree(buf);
-		goto resize_buf;
-	}
-
-	len = new_len;
-
-	bssid_list = buf;
-	count = le32_to_cpu(bssid_list->num_items);
-	real_count = 0;
-	netdev_dbg(usbdev->net, "%s(): buflen: %d\n", __func__, len);
-
-	bssid_len = 0;
-	bssid = next_bssid_list_item((void *)bssid_list->bssid_data,
-				     &bssid_len, buf, len);
-
-	/* Device returns incorrect 'num_items'. Workaround by ignoring the
-	 * received 'num_items' and walking through full bssid buffer instead.
-	 */
-	while (check_bssid_list_item(bssid, bssid_len, buf, len)) {
-		if (rndis_bss_info_update(usbdev, bssid) && match_bssid &&
-		    matched) {
-			if (ether_addr_equal(bssid->mac, match_bssid))
-				*matched = true;
-		}
-
-		real_count++;
-		bssid = next_bssid_list_item(bssid, &bssid_len, buf, len);
-	}
-
-	netdev_dbg(usbdev->net, "%s(): num_items from device: %d, really found:"
-				" %d\n", __func__, count, real_count);
-
-out:
-	kfree(buf);
-	return ret;
-}
-
-static void rndis_get_scan_results(struct work_struct *work)
-{
-	struct rndis_wlan_private *priv =
-		container_of(work, struct rndis_wlan_private, scan_work.work);
-	struct usbnet *usbdev = priv->usbdev;
-	struct cfg80211_scan_info info = {};
-	int ret;
-
-	netdev_dbg(usbdev->net, "get_scan_results\n");
-
-	if (!priv->scan_request)
-		return;
-
-	ret = rndis_check_bssid_list(usbdev, NULL, NULL);
-
-	info.aborted = ret < 0;
-	cfg80211_scan_done(priv->scan_request, &info);
-
-	priv->scan_request = NULL;
-}
-
-static int rndis_connect(struct wiphy *wiphy, struct net_device *dev,
-					struct cfg80211_connect_params *sme)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	struct ieee80211_channel *channel = sme->channel;
-	struct ndis_80211_ssid ssid;
-	int pairwise = RNDIS_WLAN_ALG_NONE;
-	int groupwise = RNDIS_WLAN_ALG_NONE;
-	int keymgmt = RNDIS_WLAN_KEY_MGMT_NONE;
-	int length, i, ret, chan = -1;
-
-	if (channel)
-		chan = ieee80211_frequency_to_channel(channel->center_freq);
-
-	groupwise = rndis_cipher_to_alg(sme->crypto.cipher_group);
-	for (i = 0; i < sme->crypto.n_ciphers_pairwise; i++)
-		pairwise |=
-			rndis_cipher_to_alg(sme->crypto.ciphers_pairwise[i]);
-
-	if (sme->crypto.n_ciphers_pairwise > 0 &&
-			pairwise == RNDIS_WLAN_ALG_NONE) {
-		netdev_err(usbdev->net, "Unsupported pairwise cipher\n");
-		return -ENOTSUPP;
-	}
-
-	for (i = 0; i < sme->crypto.n_akm_suites; i++)
-		keymgmt |=
-			rndis_akm_suite_to_key_mgmt(sme->crypto.akm_suites[i]);
-
-	if (sme->crypto.n_akm_suites > 0 &&
-			keymgmt == RNDIS_WLAN_KEY_MGMT_NONE) {
-		netdev_err(usbdev->net, "Invalid keymgmt\n");
-		return -ENOTSUPP;
-	}
-
-	netdev_dbg(usbdev->net, "cfg80211.connect('%.32s':[%pM]:%d:[%d,0x%x:0x%x]:[0x%x:0x%x]:0x%x)\n",
-		   sme->ssid, sme->bssid, chan,
-		   sme->privacy, sme->crypto.wpa_versions, sme->auth_type,
-		   groupwise, pairwise, keymgmt);
-
-	if (is_associated(usbdev))
-		disassociate(usbdev, false);
-
-	ret = set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "connect: set_infra_mode failed, %d\n",
-			   ret);
-		goto err_turn_radio_on;
-	}
-
-	ret = set_auth_mode(usbdev, sme->crypto.wpa_versions, sme->auth_type,
-								keymgmt);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "connect: set_auth_mode failed, %d\n",
-			   ret);
-		goto err_turn_radio_on;
-	}
-
-	set_priv_filter(usbdev);
-
-	ret = set_encr_mode(usbdev, pairwise, groupwise);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "connect: set_encr_mode failed, %d\n",
-			   ret);
-		goto err_turn_radio_on;
-	}
-
-	if (channel) {
-		ret = set_channel(usbdev, chan);
-		if (ret < 0) {
-			netdev_dbg(usbdev->net, "connect: set_channel failed, %d\n",
-				   ret);
-			goto err_turn_radio_on;
-		}
-	}
-
-	if (sme->key && ((groupwise | pairwise) & RNDIS_WLAN_ALG_WEP)) {
-		priv->encr_tx_key_index = sme->key_idx;
-		ret = add_wep_key(usbdev, sme->key, sme->key_len, sme->key_idx);
-		if (ret < 0) {
-			netdev_dbg(usbdev->net, "connect: add_wep_key failed, %d (%d, %d)\n",
-				   ret, sme->key_len, sme->key_idx);
-			goto err_turn_radio_on;
-		}
-	}
-
-	if (sme->bssid && !is_zero_ether_addr(sme->bssid) &&
-				!is_broadcast_ether_addr(sme->bssid)) {
-		ret = set_bssid(usbdev, sme->bssid);
-		if (ret < 0) {
-			netdev_dbg(usbdev->net, "connect: set_bssid failed, %d\n",
-				   ret);
-			goto err_turn_radio_on;
-		}
-	} else
-		clear_bssid(usbdev);
-
-	length = sme->ssid_len;
-	if (length > NDIS_802_11_LENGTH_SSID)
-		length = NDIS_802_11_LENGTH_SSID;
-
-	memset(&ssid, 0, sizeof(ssid));
-	ssid.length = cpu_to_le32(length);
-	memcpy(ssid.essid, sme->ssid, length);
-
-	/* Pause and purge rx queue, so we don't pass packets before
-	 * 'media connect'-indication.
-	 */
-	usbnet_pause_rx(usbdev);
-	usbnet_purge_paused_rxq(usbdev);
-
-	ret = set_essid(usbdev, &ssid);
-	if (ret < 0)
-		netdev_dbg(usbdev->net, "connect: set_essid failed, %d\n", ret);
-	return ret;
-
-err_turn_radio_on:
-	disassociate(usbdev, true);
-
-	return ret;
-}
-
-static int rndis_disconnect(struct wiphy *wiphy, struct net_device *dev,
-								u16 reason_code)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	netdev_dbg(usbdev->net, "cfg80211.disconnect(%d)\n", reason_code);
-
-	priv->connected = false;
-	eth_zero_addr(priv->bssid);
-
-	return deauthenticate(usbdev);
-}
-
-static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev,
-					struct cfg80211_ibss_params *params)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	struct ieee80211_channel *channel = params->chandef.chan;
-	struct ndis_80211_ssid ssid;
-	enum nl80211_auth_type auth_type;
-	int ret, alg, length, chan = -1;
-
-	if (channel)
-		chan = ieee80211_frequency_to_channel(channel->center_freq);
-
-	/* TODO: How to handle ad-hoc encryption?
-	 * connect() has *key, join_ibss() doesn't. RNDIS requires key to be
-	 * pre-shared for encryption (open/shared/wpa), is key set before
-	 * join_ibss? Which auth_type to use (not in params)? What about WPA?
-	 */
-	if (params->privacy) {
-		auth_type = NL80211_AUTHTYPE_SHARED_KEY;
-		alg = RNDIS_WLAN_ALG_WEP;
-	} else {
-		auth_type = NL80211_AUTHTYPE_OPEN_SYSTEM;
-		alg = RNDIS_WLAN_ALG_NONE;
-	}
-
-	netdev_dbg(usbdev->net, "cfg80211.join_ibss('%.32s':[%pM]:%d:%d)\n",
-		   params->ssid, params->bssid, chan, params->privacy);
-
-	if (is_associated(usbdev))
-		disassociate(usbdev, false);
-
-	ret = set_infra_mode(usbdev, NDIS_80211_INFRA_ADHOC);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "join_ibss: set_infra_mode failed, %d\n",
-			   ret);
-		goto err_turn_radio_on;
-	}
-
-	ret = set_auth_mode(usbdev, 0, auth_type, RNDIS_WLAN_KEY_MGMT_NONE);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "join_ibss: set_auth_mode failed, %d\n",
-			   ret);
-		goto err_turn_radio_on;
-	}
-
-	set_priv_filter(usbdev);
-
-	ret = set_encr_mode(usbdev, alg, RNDIS_WLAN_ALG_NONE);
-	if (ret < 0) {
-		netdev_dbg(usbdev->net, "join_ibss: set_encr_mode failed, %d\n",
-			   ret);
-		goto err_turn_radio_on;
-	}
-
-	if (channel) {
-		ret = set_channel(usbdev, chan);
-		if (ret < 0) {
-			netdev_dbg(usbdev->net, "join_ibss: set_channel failed, %d\n",
-				   ret);
-			goto err_turn_radio_on;
-		}
-	}
-
-	if (params->bssid && !is_zero_ether_addr(params->bssid) &&
-				!is_broadcast_ether_addr(params->bssid)) {
-		ret = set_bssid(usbdev, params->bssid);
-		if (ret < 0) {
-			netdev_dbg(usbdev->net, "join_ibss: set_bssid failed, %d\n",
-				   ret);
-			goto err_turn_radio_on;
-		}
-	} else
-		clear_bssid(usbdev);
-
-	length = params->ssid_len;
-	if (length > NDIS_802_11_LENGTH_SSID)
-		length = NDIS_802_11_LENGTH_SSID;
-
-	memset(&ssid, 0, sizeof(ssid));
-	ssid.length = cpu_to_le32(length);
-	memcpy(ssid.essid, params->ssid, length);
-
-	/* Don't need to pause rx queue for ad-hoc. */
-	usbnet_purge_paused_rxq(usbdev);
-	usbnet_resume_rx(usbdev);
-
-	ret = set_essid(usbdev, &ssid);
-	if (ret < 0)
-		netdev_dbg(usbdev->net, "join_ibss: set_essid failed, %d\n",
-			   ret);
-	return ret;
-
-err_turn_radio_on:
-	disassociate(usbdev, true);
-
-	return ret;
-}
-
-static int rndis_leave_ibss(struct wiphy *wiphy, struct net_device *dev)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	netdev_dbg(usbdev->net, "cfg80211.leave_ibss()\n");
-
-	priv->connected = false;
-	eth_zero_addr(priv->bssid);
-
-	return deauthenticate(usbdev);
-}
-
-static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
-			 int link_id,  u8 key_index, bool pairwise,
-			 const u8 *mac_addr, struct key_params *params)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	__le32 flags;
-
-	netdev_dbg(usbdev->net, "%s(%i, %pM, %08x)\n",
-		   __func__, key_index, mac_addr, params->cipher);
-
-	switch (params->cipher) {
-	case WLAN_CIPHER_SUITE_WEP40:
-	case WLAN_CIPHER_SUITE_WEP104:
-		return add_wep_key(usbdev, params->key, params->key_len,
-								key_index);
-	case WLAN_CIPHER_SUITE_TKIP:
-	case WLAN_CIPHER_SUITE_CCMP:
-		flags = 0;
-
-		if (params->seq && params->seq_len > 0)
-			flags |= NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ;
-		if (mac_addr)
-			flags |= NDIS_80211_ADDKEY_PAIRWISE_KEY |
-					NDIS_80211_ADDKEY_TRANSMIT_KEY;
-
-		return add_wpa_key(usbdev, params->key, params->key_len,
-				key_index, mac_addr, params->seq,
-				params->seq_len, params->cipher, flags);
-	default:
-		netdev_dbg(usbdev->net, "%s(): unsupported cipher %08x\n",
-			   __func__, params->cipher);
-		return -ENOTSUPP;
-	}
-}
-
-static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev,
-			 int link_id, u8 key_index, bool pairwise,
-			 const u8 *mac_addr)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	netdev_dbg(usbdev->net, "%s(%i, %pM)\n", __func__, key_index, mac_addr);
-
-	return remove_key(usbdev, key_index, mac_addr);
-}
-
-static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev,
-				 int link_id, u8 key_index, bool unicast,
-				 bool multicast)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	struct rndis_wlan_encr_key key;
-
-	netdev_dbg(usbdev->net, "%s(%i)\n", __func__, key_index);
-
-	if (key_index >= RNDIS_WLAN_NUM_KEYS)
-		return -ENOENT;
-
-	priv->encr_tx_key_index = key_index;
-
-	if (is_wpa_key(priv, key_index))
-		return 0;
-
-	key = priv->encr_keys[key_index];
-
-	return add_wep_key(usbdev, key.material, key.len, key_index);
-}
-
-static void rndis_fill_station_info(struct usbnet *usbdev,
-						struct station_info *sinfo)
-{
-	__le32 linkspeed, rssi;
-	int ret, len;
-
-	memset(sinfo, 0, sizeof(*sinfo));
-
-	len = sizeof(linkspeed);
-	ret = rndis_query_oid(usbdev, RNDIS_OID_GEN_LINK_SPEED, &linkspeed, &len);
-	if (ret == 0) {
-		sinfo->txrate.legacy = le32_to_cpu(linkspeed) / 1000;
-		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
-	}
-
-	len = sizeof(rssi);
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_RSSI,
-			      &rssi, &len);
-	if (ret == 0) {
-		sinfo->signal = level_to_qual(le32_to_cpu(rssi));
-		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_SIGNAL);
-	}
-}
-
-static int rndis_get_station(struct wiphy *wiphy, struct net_device *dev,
-			     const u8 *mac, struct station_info *sinfo)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	if (!ether_addr_equal(priv->bssid, mac))
-		return -ENOENT;
-
-	rndis_fill_station_info(usbdev, sinfo);
-
-	return 0;
-}
-
-static int rndis_dump_station(struct wiphy *wiphy, struct net_device *dev,
-			       int idx, u8 *mac, struct station_info *sinfo)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-
-	if (idx != 0)
-		return -ENOENT;
-
-	memcpy(mac, priv->bssid, ETH_ALEN);
-
-	rndis_fill_station_info(usbdev, sinfo);
-
-	return 0;
-}
-
-static int rndis_set_pmksa(struct wiphy *wiphy, struct net_device *netdev,
-				struct cfg80211_pmksa *pmksa)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	struct ndis_80211_pmkid *pmkids;
-	u32 *tmp = (u32 *)pmksa->pmkid;
-
-	netdev_dbg(usbdev->net, "%s(%pM, %08X:%08X:%08X:%08X)\n", __func__,
-			pmksa->bssid,
-			cpu_to_be32(tmp[0]), cpu_to_be32(tmp[1]),
-			cpu_to_be32(tmp[2]), cpu_to_be32(tmp[3]));
-
-	pmkids = get_device_pmkids(usbdev);
-	if (IS_ERR(pmkids)) {
-		/* couldn't read PMKID cache from device */
-		return PTR_ERR(pmkids);
-	}
-
-	pmkids = update_pmkid(usbdev, pmkids, pmksa, wiphy->max_num_pmkids);
-	if (IS_ERR(pmkids)) {
-		/* not found, list full, etc */
-		return PTR_ERR(pmkids);
-	}
-
-	return set_device_pmkids(usbdev, pmkids);
-}
-
-static int rndis_del_pmksa(struct wiphy *wiphy, struct net_device *netdev,
-				struct cfg80211_pmksa *pmksa)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	struct ndis_80211_pmkid *pmkids;
-	u32 *tmp = (u32 *)pmksa->pmkid;
-
-	netdev_dbg(usbdev->net, "%s(%pM, %08X:%08X:%08X:%08X)\n", __func__,
-			pmksa->bssid,
-			cpu_to_be32(tmp[0]), cpu_to_be32(tmp[1]),
-			cpu_to_be32(tmp[2]), cpu_to_be32(tmp[3]));
-
-	pmkids = get_device_pmkids(usbdev);
-	if (IS_ERR(pmkids)) {
-		/* Couldn't read PMKID cache from device */
-		return PTR_ERR(pmkids);
-	}
-
-	pmkids = remove_pmkid(usbdev, pmkids, pmksa, wiphy->max_num_pmkids);
-	if (IS_ERR(pmkids)) {
-		/* not found, etc */
-		return PTR_ERR(pmkids);
-	}
-
-	return set_device_pmkids(usbdev, pmkids);
-}
-
-static int rndis_flush_pmksa(struct wiphy *wiphy, struct net_device *netdev)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	struct ndis_80211_pmkid pmkid;
-
-	netdev_dbg(usbdev->net, "%s()\n", __func__);
-
-	memset(&pmkid, 0, sizeof(pmkid));
-
-	pmkid.length = cpu_to_le32(sizeof(pmkid));
-	pmkid.bssid_info_count = cpu_to_le32(0);
-
-	return rndis_set_oid(usbdev, RNDIS_OID_802_11_PMKID,
-			     &pmkid, sizeof(pmkid));
-}
-
-static int rndis_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev,
-				bool enabled, int timeout)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-	struct usbnet *usbdev = priv->usbdev;
-	int power_mode;
-	__le32 mode;
-	int ret;
-
-	if (priv->device_type != RNDIS_BCM4320B)
-		return -ENOTSUPP;
-
-	netdev_dbg(usbdev->net, "%s(): %s, %d\n", __func__,
-				enabled ? "enabled" : "disabled",
-				timeout);
-
-	if (enabled)
-		power_mode = NDIS_80211_POWER_MODE_FAST_PSP;
-	else
-		power_mode = NDIS_80211_POWER_MODE_CAM;
-
-	if (power_mode == priv->power_mode)
-		return 0;
-
-	priv->power_mode = power_mode;
-
-	mode = cpu_to_le32(power_mode);
-	ret = rndis_set_oid(usbdev, RNDIS_OID_802_11_POWER_MODE,
-			    &mode, sizeof(mode));
-
-	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_POWER_MODE -> %d\n",
-				__func__, ret);
-
-	return ret;
-}
-
-static int rndis_set_cqm_rssi_config(struct wiphy *wiphy,
-					struct net_device *dev,
-					s32 rssi_thold, u32 rssi_hyst)
-{
-	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
-
-	priv->cqm_rssi_thold = rssi_thold;
-	priv->cqm_rssi_hyst = rssi_hyst;
-	priv->last_cqm_event_rssi = 0;
-
-	return 0;
-}
-
-static void rndis_wlan_craft_connected_bss(struct usbnet *usbdev, u8 *bssid,
-					   struct ndis_80211_assoc_info *info)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ieee80211_channel *channel;
-	struct ndis_80211_ssid ssid;
-	struct cfg80211_bss *bss;
-	s32 signal;
-	u64 timestamp;
-	u16 capability;
-	u32 beacon_period = 0;
-	__le32 rssi;
-	u8 ie_buf[34];
-	int len, ret, ie_len;
-
-	/* Get signal quality, in case of error use rssi=0 and ignore error. */
-	len = sizeof(rssi);
-	rssi = 0;
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_RSSI,
-			      &rssi, &len);
-	signal = level_to_qual(le32_to_cpu(rssi));
-
-	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_RSSI -> %d, "
-		   "rssi:%d, qual: %d\n", __func__, ret, le32_to_cpu(rssi),
-		   level_to_qual(le32_to_cpu(rssi)));
-
-	/* Get AP capabilities */
-	if (info) {
-		capability = le16_to_cpu(info->resp_ie.capa);
-	} else {
-		/* Set atleast ESS/IBSS capability */
-		capability = (priv->infra_mode == NDIS_80211_INFRA_INFRA) ?
-				WLAN_CAPABILITY_ESS : WLAN_CAPABILITY_IBSS;
-	}
-
-	/* Get channel and beacon interval */
-	channel = get_current_channel(usbdev, &beacon_period);
-	if (!channel) {
-		netdev_warn(usbdev->net, "%s(): could not get channel.\n",
-					__func__);
-		return;
-	}
-
-	/* Get SSID, in case of error, use zero length SSID and ignore error. */
-	len = sizeof(ssid);
-	memset(&ssid, 0, sizeof(ssid));
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_SSID,
-			      &ssid, &len);
-	netdev_dbg(usbdev->net, "%s(): RNDIS_OID_802_11_SSID -> %d, len: %d, ssid: "
-				"'%.32s'\n", __func__, ret,
-				le32_to_cpu(ssid.length), ssid.essid);
-
-	if (le32_to_cpu(ssid.length) > 32)
-		ssid.length = cpu_to_le32(32);
-
-	ie_buf[0] = WLAN_EID_SSID;
-	ie_buf[1] = le32_to_cpu(ssid.length);
-	memcpy(&ie_buf[2], ssid.essid, le32_to_cpu(ssid.length));
-
-	ie_len = le32_to_cpu(ssid.length) + 2;
-
-	/* no tsf */
-	timestamp = 0;
-
-	netdev_dbg(usbdev->net, "%s(): channel:%d(freq), bssid:[%pM], tsf:%d, "
-		"capa:%x, beacon int:%d, resp_ie(len:%d, essid:'%.32s'), "
-		"signal:%d\n", __func__, (channel ? channel->center_freq : -1),
-		bssid, (u32)timestamp, capability, beacon_period, ie_len,
-		ssid.essid, signal);
-
-	bss = cfg80211_inform_bss(priv->wdev.wiphy, channel,
-				  CFG80211_BSS_FTYPE_UNKNOWN, bssid,
-				  timestamp, capability, beacon_period,
-				  ie_buf, ie_len, signal, GFP_KERNEL);
-	cfg80211_put_bss(priv->wdev.wiphy, bss);
-}
-
-/*
- * workers, indication handlers, device poller
- */
-static void rndis_wlan_do_link_up_work(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct ndis_80211_assoc_info *info = NULL;
-	u8 bssid[ETH_ALEN];
-	unsigned int resp_ie_len, req_ie_len;
-	unsigned int offset;
-	u8 *req_ie, *resp_ie;
-	int ret;
-	bool roamed = false;
-	bool match_bss;
-
-	if (priv->infra_mode == NDIS_80211_INFRA_INFRA && priv->connected) {
-		/* received media connect indication while connected, either
-		 * device reassociated with same AP or roamed to new. */
-		roamed = true;
-	}
-
-	req_ie_len = 0;
-	resp_ie_len = 0;
-	req_ie = NULL;
-	resp_ie = NULL;
-
-	if (priv->infra_mode == NDIS_80211_INFRA_INFRA) {
-		info = kzalloc(CONTROL_BUFFER_SIZE, GFP_KERNEL);
-		if (!info) {
-			/* No memory? Try resume work later */
-			set_bit(WORK_LINK_UP, &priv->work_pending);
-			queue_work(priv->workqueue, &priv->work);
-			return;
-		}
-
-		/* Get association info IEs from device. */
-		ret = get_association_info(usbdev, info, CONTROL_BUFFER_SIZE);
-		if (!ret) {
-			req_ie_len = le32_to_cpu(info->req_ie_length);
-			if (req_ie_len > CONTROL_BUFFER_SIZE)
-				req_ie_len = CONTROL_BUFFER_SIZE;
-			if (req_ie_len != 0) {
-				offset = le32_to_cpu(info->offset_req_ies);
-
-				if (offset > CONTROL_BUFFER_SIZE)
-					offset = CONTROL_BUFFER_SIZE;
-
-				req_ie = (u8 *)info + offset;
-
-				if (offset + req_ie_len > CONTROL_BUFFER_SIZE)
-					req_ie_len =
-						CONTROL_BUFFER_SIZE - offset;
-			}
-
-			resp_ie_len = le32_to_cpu(info->resp_ie_length);
-			if (resp_ie_len > CONTROL_BUFFER_SIZE)
-				resp_ie_len = CONTROL_BUFFER_SIZE;
-			if (resp_ie_len != 0) {
-				offset = le32_to_cpu(info->offset_resp_ies);
-
-				if (offset > CONTROL_BUFFER_SIZE)
-					offset = CONTROL_BUFFER_SIZE;
-
-				resp_ie = (u8 *)info + offset;
-
-				if (offset + resp_ie_len > CONTROL_BUFFER_SIZE)
-					resp_ie_len =
-						CONTROL_BUFFER_SIZE - offset;
-			}
-		} else {
-			/* Since rndis_wlan_craft_connected_bss() might use info
-			 * later and expects info to contain valid data if
-			 * non-null, free info and set NULL here.
-			 */
-			kfree(info);
-			info = NULL;
-		}
-	} else if (WARN_ON(priv->infra_mode != NDIS_80211_INFRA_ADHOC))
-		return;
-
-	ret = get_bssid(usbdev, bssid);
-	if (ret < 0)
-		memset(bssid, 0, sizeof(bssid));
-
-	netdev_dbg(usbdev->net, "link up work: [%pM]%s\n",
-		   bssid, roamed ? " roamed" : "");
-
-	/* Internal bss list in device should contain at least the currently
-	 * connected bss and we can get it to cfg80211 with
-	 * rndis_check_bssid_list().
-	 *
-	 * NDIS spec says: "If the device is associated, but the associated
-	 *  BSSID is not in its BSSID scan list, then the driver must add an
-	 *  entry for the BSSID at the end of the data that it returns in
-	 *  response to query of RNDIS_OID_802_11_BSSID_LIST."
-	 *
-	 * NOTE: Seems to be true for BCM4320b variant, but not BCM4320a.
-	 */
-	match_bss = false;
-	rndis_check_bssid_list(usbdev, bssid, &match_bss);
-
-	if (!is_zero_ether_addr(bssid) && !match_bss) {
-		/* Couldn't get bss from device, we need to manually craft bss
-		 * for cfg80211.
-		 */
-		rndis_wlan_craft_connected_bss(usbdev, bssid, info);
-	}
-
-	if (priv->infra_mode == NDIS_80211_INFRA_INFRA) {
-		if (!roamed) {
-			cfg80211_connect_result(usbdev->net, bssid, req_ie,
-						req_ie_len, resp_ie,
-						resp_ie_len, 0, GFP_KERNEL);
-		} else {
-			struct cfg80211_roam_info roam_info = {
-				.links[0].channel =
-					get_current_channel(usbdev, NULL),
-				.links[0].bssid = bssid,
-				.req_ie = req_ie,
-				.req_ie_len = req_ie_len,
-				.resp_ie = resp_ie,
-				.resp_ie_len = resp_ie_len,
-			};
-
-			cfg80211_roamed(usbdev->net, &roam_info, GFP_KERNEL);
-		}
-	} else if (priv->infra_mode == NDIS_80211_INFRA_ADHOC)
-		cfg80211_ibss_joined(usbdev->net, bssid,
-				     get_current_channel(usbdev, NULL),
-				     GFP_KERNEL);
-
-	kfree(info);
-
-	priv->connected = true;
-	memcpy(priv->bssid, bssid, ETH_ALEN);
-
-	usbnet_resume_rx(usbdev);
-	netif_carrier_on(usbdev->net);
-}
-
-static void rndis_wlan_do_link_down_work(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-
-	if (priv->connected) {
-		priv->connected = false;
-		eth_zero_addr(priv->bssid);
-
-		deauthenticate(usbdev);
-
-		cfg80211_disconnected(usbdev->net, 0, NULL, 0, true, GFP_KERNEL);
-	}
-
-	netif_carrier_off(usbdev->net);
-}
-
-static void rndis_wlan_worker(struct work_struct *work)
-{
-	struct rndis_wlan_private *priv =
-		container_of(work, struct rndis_wlan_private, work);
-	struct usbnet *usbdev = priv->usbdev;
-
-	if (test_and_clear_bit(WORK_LINK_UP, &priv->work_pending))
-		rndis_wlan_do_link_up_work(usbdev);
-
-	if (test_and_clear_bit(WORK_LINK_DOWN, &priv->work_pending))
-		rndis_wlan_do_link_down_work(usbdev);
-
-	if (test_and_clear_bit(WORK_SET_MULTICAST_LIST, &priv->work_pending))
-		set_multicast_list(usbdev);
-}
-
-static void rndis_wlan_set_multicast_list(struct net_device *dev)
-{
-	struct usbnet *usbdev = netdev_priv(dev);
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-
-	if (test_bit(WORK_SET_MULTICAST_LIST, &priv->work_pending))
-		return;
-
-	set_bit(WORK_SET_MULTICAST_LIST, &priv->work_pending);
-	queue_work(priv->workqueue, &priv->work);
-}
-
-static void rndis_wlan_auth_indication(struct usbnet *usbdev,
-				struct ndis_80211_status_indication *indication,
-				int len)
-{
-	u8 *buf;
-	const char *type;
-	int flags, buflen, key_id;
-	bool pairwise_error, group_error;
-	struct ndis_80211_auth_request *auth_req;
-	enum nl80211_key_type key_type;
-
-	/* must have at least one array entry */
-	if (len < offsetof(struct ndis_80211_status_indication, u) +
-				sizeof(struct ndis_80211_auth_request)) {
-		netdev_info(usbdev->net, "authentication indication: too short message (%i)\n",
-			    len);
-		return;
-	}
-
-	buf = (void *)&indication->u.auth_request[0];
-	buflen = len - offsetof(struct ndis_80211_status_indication, u);
-
-	while (buflen >= sizeof(*auth_req)) {
-		auth_req = (void *)buf;
-		if (buflen < le32_to_cpu(auth_req->length))
-			return;
-		type = "unknown";
-		flags = le32_to_cpu(auth_req->flags);
-		pairwise_error = false;
-		group_error = false;
-
-		if (flags & 0x1)
-			type = "reauth request";
-		if (flags & 0x2)
-			type = "key update request";
-		if (flags & 0x6) {
-			pairwise_error = true;
-			type = "pairwise_error";
-		}
-		if (flags & 0xe) {
-			group_error = true;
-			type = "group_error";
-		}
-
-		netdev_info(usbdev->net, "authentication indication: %s (0x%08x)\n",
-			    type, le32_to_cpu(auth_req->flags));
-
-		if (pairwise_error) {
-			key_type = NL80211_KEYTYPE_PAIRWISE;
-			key_id = -1;
-
-			cfg80211_michael_mic_failure(usbdev->net,
-							auth_req->bssid,
-							key_type, key_id, NULL,
-							GFP_KERNEL);
-		}
-
-		if (group_error) {
-			key_type = NL80211_KEYTYPE_GROUP;
-			key_id = -1;
-
-			cfg80211_michael_mic_failure(usbdev->net,
-							auth_req->bssid,
-							key_type, key_id, NULL,
-							GFP_KERNEL);
-		}
-
-		buflen -= le32_to_cpu(auth_req->length);
-		buf += le32_to_cpu(auth_req->length);
-	}
-}
-
-static void rndis_wlan_pmkid_cand_list_indication(struct usbnet *usbdev,
-				struct ndis_80211_status_indication *indication,
-				int len)
-{
-	struct ndis_80211_pmkid_cand_list *cand_list;
-	int list_len, expected_len, i;
-
-	if (len < offsetof(struct ndis_80211_status_indication, u) +
-				sizeof(struct ndis_80211_pmkid_cand_list)) {
-		netdev_info(usbdev->net, "pmkid candidate list indication: too short message (%i)\n",
-			    len);
-		return;
-	}
-
-	list_len = le32_to_cpu(indication->u.cand_list.num_candidates) *
-			sizeof(struct ndis_80211_pmkid_candidate);
-	expected_len = sizeof(struct ndis_80211_pmkid_cand_list) + list_len +
-			offsetof(struct ndis_80211_status_indication, u);
-
-	if (len < expected_len) {
-		netdev_info(usbdev->net, "pmkid candidate list indication: list larger than buffer (%i < %i)\n",
-			    len, expected_len);
-		return;
-	}
-
-	cand_list = &indication->u.cand_list;
-
-	netdev_info(usbdev->net, "pmkid candidate list indication: version %i, candidates %i\n",
-		    le32_to_cpu(cand_list->version),
-		    le32_to_cpu(cand_list->num_candidates));
-
-	if (le32_to_cpu(cand_list->version) != 1)
-		return;
-
-	for (i = 0; i < le32_to_cpu(cand_list->num_candidates); i++) {
-		struct ndis_80211_pmkid_candidate *cand =
-						&cand_list->candidate_list[i];
-		bool preauth = !!(cand->flags & NDIS_80211_PMKID_CAND_PREAUTH);
-
-		netdev_dbg(usbdev->net, "cand[%i]: flags: 0x%08x, preauth: %d, bssid: %pM\n",
-			   i, le32_to_cpu(cand->flags), preauth, cand->bssid);
-
-		cfg80211_pmksa_candidate_notify(usbdev->net, i, cand->bssid,
-						preauth, GFP_ATOMIC);
-	}
-}
-
-static void rndis_wlan_media_specific_indication(struct usbnet *usbdev,
-			struct rndis_indicate *msg, int buflen)
-{
-	struct ndis_80211_status_indication *indication;
-	unsigned int len, offset;
-
-	offset = offsetof(struct rndis_indicate, status) +
-			le32_to_cpu(msg->offset);
-	len = le32_to_cpu(msg->length);
-
-	if (len < 8) {
-		netdev_info(usbdev->net, "media specific indication, ignore too short message (%i < 8)\n",
-			    len);
-		return;
-	}
-
-	if (len > buflen || offset > buflen || offset + len > buflen) {
-		netdev_info(usbdev->net, "media specific indication, too large to fit to buffer (%i > %i)\n",
-			    offset + len, buflen);
-		return;
-	}
-
-	indication = (void *)((u8 *)msg + offset);
-
-	switch (le32_to_cpu(indication->status_type)) {
-	case NDIS_80211_STATUSTYPE_RADIOSTATE:
-		netdev_info(usbdev->net, "radio state indication: %i\n",
-			    le32_to_cpu(indication->u.radio_status));
-		return;
-
-	case NDIS_80211_STATUSTYPE_MEDIASTREAMMODE:
-		netdev_info(usbdev->net, "media stream mode indication: %i\n",
-			    le32_to_cpu(indication->u.media_stream_mode));
-		return;
-
-	case NDIS_80211_STATUSTYPE_AUTHENTICATION:
-		rndis_wlan_auth_indication(usbdev, indication, len);
-		return;
-
-	case NDIS_80211_STATUSTYPE_PMKID_CANDIDATELIST:
-		rndis_wlan_pmkid_cand_list_indication(usbdev, indication, len);
-		return;
-
-	default:
-		netdev_info(usbdev->net, "media specific indication: unknown status type 0x%08x\n",
-			    le32_to_cpu(indication->status_type));
-	}
-}
-
-static void rndis_wlan_indication(struct usbnet *usbdev, void *ind, int buflen)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	struct rndis_indicate *msg = ind;
-
-	switch (le32_to_cpu(msg->status)) {
-	case RNDIS_STATUS_MEDIA_CONNECT:
-		if (priv->current_command_oid == RNDIS_OID_802_11_ADD_KEY) {
-			/* RNDIS_OID_802_11_ADD_KEY causes sometimes extra
-			 * "media connect" indications which confuses driver
-			 * and userspace to think that device is
-			 * roaming/reassociating when it isn't.
-			 */
-			netdev_dbg(usbdev->net, "ignored RNDIS_OID_802_11_ADD_KEY triggered 'media connect'\n");
-			return;
-		}
-
-		usbnet_pause_rx(usbdev);
-
-		netdev_info(usbdev->net, "media connect\n");
-
-		/* queue work to avoid recursive calls into rndis_command */
-		set_bit(WORK_LINK_UP, &priv->work_pending);
-		queue_work(priv->workqueue, &priv->work);
-		break;
-
-	case RNDIS_STATUS_MEDIA_DISCONNECT:
-		netdev_info(usbdev->net, "media disconnect\n");
-
-		/* queue work to avoid recursive calls into rndis_command */
-		set_bit(WORK_LINK_DOWN, &priv->work_pending);
-		queue_work(priv->workqueue, &priv->work);
-		break;
-
-	case RNDIS_STATUS_MEDIA_SPECIFIC_INDICATION:
-		rndis_wlan_media_specific_indication(usbdev, msg, buflen);
-		break;
-
-	default:
-		netdev_info(usbdev->net, "indication: 0x%08x\n",
-			    le32_to_cpu(msg->status));
-		break;
-	}
-}
-
-static int rndis_wlan_get_caps(struct usbnet *usbdev, struct wiphy *wiphy)
-{
-	struct {
-		__le32	num_items;
-		__le32	items[8];
-	} networks_supported;
-	struct ndis_80211_capability caps;
-	int len, retval, i, n;
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-
-	/* determine supported modes */
-	len = sizeof(networks_supported);
-	retval = rndis_query_oid(usbdev,
-				 RNDIS_OID_802_11_NETWORK_TYPES_SUPPORTED,
-				 &networks_supported, &len);
-	if (!retval) {
-		n = le32_to_cpu(networks_supported.num_items);
-		if (n > 8)
-			n = 8;
-		for (i = 0; i < n; i++) {
-			switch (le32_to_cpu(networks_supported.items[i])) {
-			case NDIS_80211_TYPE_FREQ_HOP:
-			case NDIS_80211_TYPE_DIRECT_SEQ:
-				priv->caps |= CAP_MODE_80211B;
-				break;
-			case NDIS_80211_TYPE_OFDM_A:
-				priv->caps |= CAP_MODE_80211A;
-				break;
-			case NDIS_80211_TYPE_OFDM_G:
-				priv->caps |= CAP_MODE_80211G;
-				break;
-			}
-		}
-	}
-
-	/* get device 802.11 capabilities, number of PMKIDs */
-	len = sizeof(caps);
-	retval = rndis_query_oid(usbdev,
-				 RNDIS_OID_802_11_CAPABILITY,
-				 &caps, &len);
-	if (!retval) {
-		netdev_dbg(usbdev->net, "RNDIS_OID_802_11_CAPABILITY -> len %d, "
-				"ver %d, pmkids %d, auth-encr-pairs %d\n",
-				le32_to_cpu(caps.length),
-				le32_to_cpu(caps.version),
-				le32_to_cpu(caps.num_pmkids),
-				le32_to_cpu(caps.num_auth_encr_pair));
-		wiphy->max_num_pmkids = le32_to_cpu(caps.num_pmkids);
-	} else
-		wiphy->max_num_pmkids = 0;
-
-	return retval;
-}
-
-static void rndis_do_cqm(struct usbnet *usbdev, s32 rssi)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	enum nl80211_cqm_rssi_threshold_event event;
-	int thold, hyst, last_event;
-
-	if (priv->cqm_rssi_thold >= 0 || rssi >= 0)
-		return;
-	if (priv->infra_mode != NDIS_80211_INFRA_INFRA)
-		return;
-
-	last_event = priv->last_cqm_event_rssi;
-	thold = priv->cqm_rssi_thold;
-	hyst = priv->cqm_rssi_hyst;
-
-	if (rssi < thold && (last_event == 0 || rssi < last_event - hyst))
-		event = NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW;
-	else if (rssi > thold && (last_event == 0 || rssi > last_event + hyst))
-		event = NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH;
-	else
-		return;
-
-	priv->last_cqm_event_rssi = rssi;
-	cfg80211_cqm_rssi_notify(usbdev->net, event, rssi, GFP_KERNEL);
-}
-
-#define DEVICE_POLLER_JIFFIES (HZ)
-static void rndis_device_poller(struct work_struct *work)
-{
-	struct rndis_wlan_private *priv =
-		container_of(work, struct rndis_wlan_private,
-							dev_poller_work.work);
-	struct usbnet *usbdev = priv->usbdev;
-	__le32 rssi, tmp;
-	int len, ret, j;
-	int update_jiffies = DEVICE_POLLER_JIFFIES;
-	void *buf;
-
-	/* Only check/do workaround when connected. Calling is_associated()
-	 * also polls device with rndis_command() and catches for media link
-	 * indications.
-	 */
-	if (!is_associated(usbdev)) {
-		/* Workaround bad scanning in BCM4320a devices with active
-		 * background scanning when not associated.
-		 */
-		if (priv->device_type == RNDIS_BCM4320A && priv->radio_on &&
-		    !priv->scan_request) {
-			/* Get previous scan results */
-			rndis_check_bssid_list(usbdev, NULL, NULL);
-
-			/* Initiate new scan */
-			rndis_start_bssid_list_scan(usbdev);
-		}
-
-		goto end;
-	}
-
-	len = sizeof(rssi);
-	ret = rndis_query_oid(usbdev, RNDIS_OID_802_11_RSSI,
-			      &rssi, &len);
-	if (ret == 0) {
-		priv->last_qual = level_to_qual(le32_to_cpu(rssi));
-		rndis_do_cqm(usbdev, le32_to_cpu(rssi));
-	}
-
-	netdev_dbg(usbdev->net, "dev-poller: RNDIS_OID_802_11_RSSI -> %d, rssi:%d, qual: %d\n",
-		   ret, le32_to_cpu(rssi), level_to_qual(le32_to_cpu(rssi)));
-
-	/* Workaround transfer stalls on poor quality links.
-	 * TODO: find right way to fix these stalls (as stalls do not happen
-	 * with ndiswrapper/windows driver). */
-	if (priv->param_workaround_interval > 0 && priv->last_qual <= 25) {
-		/* Decrease stats worker interval to catch stalls.
-		 * faster. Faster than 400-500ms causes packet loss,
-		 * Slower doesn't catch stalls fast enough.
-		 */
-		j = msecs_to_jiffies(priv->param_workaround_interval);
-		if (j > DEVICE_POLLER_JIFFIES)
-			j = DEVICE_POLLER_JIFFIES;
-		else if (j <= 0)
-			j = 1;
-		update_jiffies = j;
-
-		/* Send scan OID. Use of both OIDs is required to get device
-		 * working.
-		 */
-		tmp = cpu_to_le32(1);
-		rndis_set_oid(usbdev,
-			      RNDIS_OID_802_11_BSSID_LIST_SCAN,
-			      &tmp, sizeof(tmp));
-
-		len = CONTROL_BUFFER_SIZE;
-		buf = kmalloc(len, GFP_KERNEL);
-		if (!buf)
-			goto end;
-
-		rndis_query_oid(usbdev,
-				RNDIS_OID_802_11_BSSID_LIST,
-				buf, &len);
-		kfree(buf);
-	}
-
-end:
-	if (update_jiffies >= HZ)
-		update_jiffies = round_jiffies_relative(update_jiffies);
-	else {
-		j = round_jiffies_relative(update_jiffies);
-		if (abs(j - update_jiffies) <= 10)
-			update_jiffies = j;
-	}
-
-	queue_delayed_work(priv->workqueue, &priv->dev_poller_work,
-								update_jiffies);
-}
-
-/*
- * driver/device initialization
- */
-static void rndis_copy_module_params(struct usbnet *usbdev, int device_type)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-
-	priv->device_type = device_type;
-
-	priv->param_country[0] = modparam_country[0];
-	priv->param_country[1] = modparam_country[1];
-	priv->param_country[2] = 0;
-	priv->param_frameburst   = modparam_frameburst;
-	priv->param_afterburner  = modparam_afterburner;
-	priv->param_power_save   = modparam_power_save;
-	priv->param_power_output = modparam_power_output;
-	priv->param_roamtrigger  = modparam_roamtrigger;
-	priv->param_roamdelta    = modparam_roamdelta;
-
-	priv->param_country[0] = toupper(priv->param_country[0]);
-	priv->param_country[1] = toupper(priv->param_country[1]);
-	/* doesn't support EU as country code, use FI instead */
-	if (!strcmp(priv->param_country, "EU"))
-		strcpy(priv->param_country, "FI");
-
-	if (priv->param_power_save < 0)
-		priv->param_power_save = 0;
-	else if (priv->param_power_save > 2)
-		priv->param_power_save = 2;
-
-	if (priv->param_power_output < 0)
-		priv->param_power_output = 0;
-	else if (priv->param_power_output > 3)
-		priv->param_power_output = 3;
-
-	if (priv->param_roamtrigger < -80)
-		priv->param_roamtrigger = -80;
-	else if (priv->param_roamtrigger > -60)
-		priv->param_roamtrigger = -60;
-
-	if (priv->param_roamdelta < 0)
-		priv->param_roamdelta = 0;
-	else if (priv->param_roamdelta > 2)
-		priv->param_roamdelta = 2;
-
-	if (modparam_workaround_interval < 0)
-		priv->param_workaround_interval = 500;
-	else
-		priv->param_workaround_interval = modparam_workaround_interval;
-}
-
-static int unknown_early_init(struct usbnet *usbdev)
-{
-	/* copy module parameters for unknown so that iwconfig reports txpower
-	 * and workaround parameter is copied to private structure correctly.
-	 */
-	rndis_copy_module_params(usbdev, RNDIS_UNKNOWN);
-
-	/* This is unknown device, so do not try set configuration parameters.
-	 */
-
-	return 0;
-}
-
-static int bcm4320a_early_init(struct usbnet *usbdev)
-{
-	/* copy module parameters for bcm4320a so that iwconfig reports txpower
-	 * and workaround parameter is copied to private structure correctly.
-	 */
-	rndis_copy_module_params(usbdev, RNDIS_BCM4320A);
-
-	/* bcm4320a doesn't handle configuration parameters well. Try
-	 * set any and you get partially zeroed mac and broken device.
-	 */
-
-	return 0;
-}
-
-static int bcm4320b_early_init(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	char buf[8];
-
-	rndis_copy_module_params(usbdev, RNDIS_BCM4320B);
-
-	/* Early initialization settings, setting these won't have effect
-	 * if called after generic_rndis_bind().
-	 */
-
-	rndis_set_config_parameter_str(usbdev, "Country", priv->param_country);
-	rndis_set_config_parameter_str(usbdev, "FrameBursting",
-					priv->param_frameburst ? "1" : "0");
-	rndis_set_config_parameter_str(usbdev, "Afterburner",
-					priv->param_afterburner ? "1" : "0");
-	sprintf(buf, "%d", priv->param_power_save);
-	rndis_set_config_parameter_str(usbdev, "PowerSaveMode", buf);
-	sprintf(buf, "%d", priv->param_power_output);
-	rndis_set_config_parameter_str(usbdev, "PwrOut", buf);
-	sprintf(buf, "%d", priv->param_roamtrigger);
-	rndis_set_config_parameter_str(usbdev, "RoamTrigger", buf);
-	sprintf(buf, "%d", priv->param_roamdelta);
-	rndis_set_config_parameter_str(usbdev, "RoamDelta", buf);
-
-	return 0;
-}
-
-/* same as rndis_netdev_ops but with local multicast handler */
-static const struct net_device_ops rndis_wlan_netdev_ops = {
-	.ndo_open		= usbnet_open,
-	.ndo_stop		= usbnet_stop,
-	.ndo_start_xmit		= usbnet_start_xmit,
-	.ndo_tx_timeout		= usbnet_tx_timeout,
-	.ndo_get_stats64	= dev_get_tstats64,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_set_rx_mode	= rndis_wlan_set_multicast_list,
-};
-
-static int rndis_wlan_bind(struct usbnet *usbdev, struct usb_interface *intf)
-{
-	struct wiphy *wiphy;
-	struct rndis_wlan_private *priv;
-	int retval, len;
-	__le32 tmp;
-
-	/* allocate wiphy and rndis private data
-	 * NOTE: We only support a single virtual interface, so wiphy
-	 * and wireless_dev are somewhat synonymous for this device.
-	 */
-	wiphy = wiphy_new(&rndis_config_ops, sizeof(struct rndis_wlan_private));
-	if (!wiphy)
-		return -ENOMEM;
-
-	priv = wiphy_priv(wiphy);
-	usbdev->net->ieee80211_ptr = &priv->wdev;
-	priv->wdev.wiphy = wiphy;
-	priv->wdev.iftype = NL80211_IFTYPE_STATION;
-
-	/* These have to be initialized before calling generic_rndis_bind().
-	 * Otherwise we'll be in big trouble in rndis_wlan_early_init().
-	 */
-	usbdev->driver_priv = priv;
-	priv->usbdev = usbdev;
-
-	mutex_init(&priv->command_lock);
-
-	/* because rndis_command() sleeps we need to use workqueue */
-	priv->workqueue = create_singlethread_workqueue("rndis_wlan");
-	if (!priv->workqueue) {
-		wiphy_free(wiphy);
-		return -ENOMEM;
-	}
-	INIT_WORK(&priv->work, rndis_wlan_worker);
-	INIT_DELAYED_WORK(&priv->dev_poller_work, rndis_device_poller);
-	INIT_DELAYED_WORK(&priv->scan_work, rndis_get_scan_results);
-
-	/* try bind rndis_host */
-	retval = generic_rndis_bind(usbdev, intf, FLAG_RNDIS_PHYM_WIRELESS);
-	if (retval < 0)
-		goto fail;
-
-	/* generic_rndis_bind set packet filter to multicast_all+
-	 * promisc mode which doesn't work well for our devices (device
-	 * picks up rssi to closest station instead of to access point).
-	 *
-	 * rndis_host wants to avoid all OID as much as possible
-	 * so do promisc/multicast handling in rndis_wlan.
-	 */
-	usbdev->net->netdev_ops = &rndis_wlan_netdev_ops;
-
-	tmp = cpu_to_le32(RNDIS_PACKET_TYPE_DIRECTED | RNDIS_PACKET_TYPE_BROADCAST);
-	retval = rndis_set_oid(usbdev,
-			       RNDIS_OID_GEN_CURRENT_PACKET_FILTER,
-			       &tmp, sizeof(tmp));
-
-	len = sizeof(tmp);
-	retval = rndis_query_oid(usbdev,
-				 RNDIS_OID_802_3_MAXIMUM_LIST_SIZE,
-				 &tmp, &len);
-	priv->multicast_size = le32_to_cpu(tmp);
-	if (retval < 0 || priv->multicast_size < 0)
-		priv->multicast_size = 0;
-	if (priv->multicast_size > 0)
-		usbdev->net->flags |= IFF_MULTICAST;
-	else
-		usbdev->net->flags &= ~IFF_MULTICAST;
-
-	/* fill-out wiphy structure and register w/ cfg80211 */
-	memcpy(wiphy->perm_addr, usbdev->net->dev_addr, ETH_ALEN);
-	wiphy->privid = rndis_wiphy_privid;
-	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION)
-					| BIT(NL80211_IFTYPE_ADHOC);
-	wiphy->max_scan_ssids = 1;
-
-	/* TODO: fill-out band/encr information based on priv->caps */
-	rndis_wlan_get_caps(usbdev, wiphy);
-
-	memcpy(priv->channels, rndis_channels, sizeof(rndis_channels));
-	memcpy(priv->rates, rndis_rates, sizeof(rndis_rates));
-	priv->band.channels = priv->channels;
-	priv->band.n_channels = ARRAY_SIZE(rndis_channels);
-	priv->band.bitrates = priv->rates;
-	priv->band.n_bitrates = ARRAY_SIZE(rndis_rates);
-	wiphy->bands[NL80211_BAND_2GHZ] = &priv->band;
-	wiphy->signal_type = CFG80211_SIGNAL_TYPE_UNSPEC;
-
-	memcpy(priv->cipher_suites, rndis_cipher_suites,
-						sizeof(rndis_cipher_suites));
-	wiphy->cipher_suites = priv->cipher_suites;
-	wiphy->n_cipher_suites = ARRAY_SIZE(rndis_cipher_suites);
-
-	set_wiphy_dev(wiphy, &usbdev->udev->dev);
-
-	if (wiphy_register(wiphy)) {
-		retval = -ENODEV;
-		goto fail;
-	}
-
-	set_default_iw_params(usbdev);
-
-	priv->power_mode = -1;
-
-	/* set default rts/frag */
-	rndis_set_wiphy_params(wiphy,
-			WIPHY_PARAM_FRAG_THRESHOLD | WIPHY_PARAM_RTS_THRESHOLD);
-
-	/* turn radio off on init */
-	priv->radio_on = false;
-	disassociate(usbdev, false);
-	netif_carrier_off(usbdev->net);
-
-	return 0;
-
-fail:
-	cancel_delayed_work_sync(&priv->dev_poller_work);
-	cancel_delayed_work_sync(&priv->scan_work);
-	cancel_work_sync(&priv->work);
-	destroy_workqueue(priv->workqueue);
-
-	wiphy_free(wiphy);
-	return retval;
-}
-
-static void rndis_wlan_unbind(struct usbnet *usbdev, struct usb_interface *intf)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-
-	/* turn radio off */
-	disassociate(usbdev, false);
-
-	cancel_delayed_work_sync(&priv->dev_poller_work);
-	cancel_delayed_work_sync(&priv->scan_work);
-	cancel_work_sync(&priv->work);
-	destroy_workqueue(priv->workqueue);
-
-	rndis_unbind(usbdev, intf);
-
-	wiphy_unregister(priv->wdev.wiphy);
-	wiphy_free(priv->wdev.wiphy);
-}
-
-static int rndis_wlan_reset(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	int retval;
-
-	netdev_dbg(usbdev->net, "%s()\n", __func__);
-
-	retval = rndis_reset(usbdev);
-	if (retval)
-		netdev_warn(usbdev->net, "rndis_reset failed: %d\n", retval);
-
-	/* rndis_reset cleared multicast list, so restore here.
-	   (set_multicast_list() also turns on current packet filter) */
-	set_multicast_list(usbdev);
-
-	queue_delayed_work(priv->workqueue, &priv->dev_poller_work,
-		round_jiffies_relative(DEVICE_POLLER_JIFFIES));
-
-	return deauthenticate(usbdev);
-}
-
-static int rndis_wlan_stop(struct usbnet *usbdev)
-{
-	struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev);
-	int retval;
-	__le32 filter;
-
-	netdev_dbg(usbdev->net, "%s()\n", __func__);
-
-	retval = disassociate(usbdev, false);
-
-	priv->work_pending = 0;
-	cancel_delayed_work_sync(&priv->dev_poller_work);
-	cancel_delayed_work_sync(&priv->scan_work);
-	cancel_work_sync(&priv->work);
-	flush_workqueue(priv->workqueue);
-
-	if (priv->scan_request) {
-		struct cfg80211_scan_info info = {
-			.aborted = true,
-		};
-
-		cfg80211_scan_done(priv->scan_request, &info);
-		priv->scan_request = NULL;
-	}
-
-	/* Set current packet filter zero to block receiving data packets from
-	   device. */
-	filter = 0;
-	rndis_set_oid(usbdev, RNDIS_OID_GEN_CURRENT_PACKET_FILTER, &filter,
-								sizeof(filter));
-
-	return retval;
-}
-
-static const struct driver_info	bcm4320b_info = {
-	.description =	"Wireless RNDIS device, BCM4320b based",
-	.flags =	FLAG_WLAN | FLAG_FRAMING_RN | FLAG_NO_SETINT |
-				FLAG_AVOID_UNLINK_URBS,
-	.bind =		rndis_wlan_bind,
-	.unbind =	rndis_wlan_unbind,
-	.status =	rndis_status,
-	.rx_fixup =	rndis_rx_fixup,
-	.tx_fixup =	rndis_tx_fixup,
-	.reset =	rndis_wlan_reset,
-	.stop =		rndis_wlan_stop,
-	.early_init =	bcm4320b_early_init,
-	.indication =	rndis_wlan_indication,
-};
-
-static const struct driver_info	bcm4320a_info = {
-	.description =	"Wireless RNDIS device, BCM4320a based",
-	.flags =	FLAG_WLAN | FLAG_FRAMING_RN | FLAG_NO_SETINT |
-				FLAG_AVOID_UNLINK_URBS,
-	.bind =		rndis_wlan_bind,
-	.unbind =	rndis_wlan_unbind,
-	.status =	rndis_status,
-	.rx_fixup =	rndis_rx_fixup,
-	.tx_fixup =	rndis_tx_fixup,
-	.reset =	rndis_wlan_reset,
-	.stop =		rndis_wlan_stop,
-	.early_init =	bcm4320a_early_init,
-	.indication =	rndis_wlan_indication,
-};
-
-static const struct driver_info rndis_wlan_info = {
-	.description =	"Wireless RNDIS device",
-	.flags =	FLAG_WLAN | FLAG_FRAMING_RN | FLAG_NO_SETINT |
-				FLAG_AVOID_UNLINK_URBS,
-	.bind =		rndis_wlan_bind,
-	.unbind =	rndis_wlan_unbind,
-	.status =	rndis_status,
-	.rx_fixup =	rndis_rx_fixup,
-	.tx_fixup =	rndis_tx_fixup,
-	.reset =	rndis_wlan_reset,
-	.stop =		rndis_wlan_stop,
-	.early_init =	unknown_early_init,
-	.indication =	rndis_wlan_indication,
-};
-
-/*-------------------------------------------------------------------------*/
-
-static const struct usb_device_id products [] = {
-#define	RNDIS_MASTER_INTERFACE \
-	.bInterfaceClass	= USB_CLASS_COMM, \
-	.bInterfaceSubClass	= 2 /* ACM */, \
-	.bInterfaceProtocol	= 0x0ff
-
-/* INF driver for these devices have DriverVer >= 4.xx.xx.xx and many custom
- * parameters available. Chipset marked as 'BCM4320SKFBG' in NDISwrapper-wiki.
- */
-{
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x0411,
-	.idProduct		= 0x00bc,	/* Buffalo WLI-U2-KG125S */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x0baf,
-	.idProduct		= 0x011b,	/* U.S. Robotics USR5421 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x050d,
-	.idProduct		= 0x011b,	/* Belkin F5D7051 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x1799,	/* Belkin has two vendor ids */
-	.idProduct		= 0x011b,	/* Belkin F5D7051 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x13b1,
-	.idProduct		= 0x0014,	/* Linksys WUSB54GSv2 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x13b1,
-	.idProduct		= 0x0026,	/* Linksys WUSB54GSC */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x0b05,
-	.idProduct		= 0x1717,	/* Asus WL169gE */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x0a5c,
-	.idProduct		= 0xd11b,	/* Eminent EM4045 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x1690,
-	.idProduct		= 0x0715,	/* BT Voyager 1055 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320b_info,
-},
-/* These devices have DriverVer < 4.xx.xx.xx and do not have any custom
- * parameters available, hardware probably contain older firmware version with
- * no way of updating. Chipset marked as 'BCM4320????' in NDISwrapper-wiki.
- */
-{
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x13b1,
-	.idProduct		= 0x000e,	/* Linksys WUSB54GSv1 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320a_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x0baf,
-	.idProduct		= 0x0111,	/* U.S. Robotics USR5420 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320a_info,
-}, {
-	.match_flags	=   USB_DEVICE_ID_MATCH_INT_INFO
-			  | USB_DEVICE_ID_MATCH_DEVICE,
-	.idVendor		= 0x0411,
-	.idProduct		= 0x004b,	/* BUFFALO WLI-USB-G54 */
-	RNDIS_MASTER_INTERFACE,
-	.driver_info		= (unsigned long) &bcm4320a_info,
-},
-/* Generic Wireless RNDIS devices that we don't have exact
- * idVendor/idProduct/chip yet.
- */
-{
-	/* RNDIS is MSFT's un-official variant of CDC ACM */
-	USB_INTERFACE_INFO(USB_CLASS_COMM, 2 /* ACM */, 0x0ff),
-	.driver_info = (unsigned long) &rndis_wlan_info,
-}, {
-	/* "ActiveSync" is an undocumented variant of RNDIS, used in WM5 */
-	USB_INTERFACE_INFO(USB_CLASS_MISC, 1, 1),
-	.driver_info = (unsigned long) &rndis_wlan_info,
-},
-	{ },		// END
-};
-MODULE_DEVICE_TABLE(usb, products);
-
-static struct usb_driver rndis_wlan_driver = {
-	.name =		"rndis_wlan",
-	.id_table =	products,
-	.probe =	usbnet_probe,
-	.disconnect =	usbnet_disconnect,
-	.suspend =	usbnet_suspend,
-	.resume =	usbnet_resume,
-	.disable_hub_initiated_lpm = 1,
-};
-
-module_usb_driver(rndis_wlan_driver);
-
-MODULE_AUTHOR("Bjorge Dijkstra");
-MODULE_AUTHOR("Jussi Kivilinna");
-MODULE_DESCRIPTION("Driver for RNDIS based USB Wireless adapters");
-MODULE_LICENSE("GPL");
-
diff --git a/drivers/net/wireless/wl3501.h b/drivers/net/wireless/wl3501.h
deleted file mode 100644
index 91f276dd22a1..000000000000
--- a/drivers/net/wireless/wl3501.h
+++ /dev/null
@@ -1,615 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef __WL3501_H__
-#define __WL3501_H__
-
-#include <linux/spinlock.h>
-#include <linux/ieee80211.h>
-
-/* define for WLA 2.0 */
-#define WL3501_BLKSZ 256
-/*
- * ID for input Signals of DRIVER block
- * bit[7-5] is block ID: 000
- * bit[4-0] is signal ID
-*/
-enum wl3501_signals {
-	WL3501_SIG_ALARM,
-	WL3501_SIG_MD_CONFIRM,
-	WL3501_SIG_MD_IND,
-	WL3501_SIG_ASSOC_CONFIRM,
-	WL3501_SIG_ASSOC_IND,
-	WL3501_SIG_AUTH_CONFIRM,
-	WL3501_SIG_AUTH_IND,
-	WL3501_SIG_DEAUTH_CONFIRM,
-	WL3501_SIG_DEAUTH_IND,
-	WL3501_SIG_DISASSOC_CONFIRM,
-	WL3501_SIG_DISASSOC_IND,
-	WL3501_SIG_GET_CONFIRM,
-	WL3501_SIG_JOIN_CONFIRM,
-	WL3501_SIG_PWR_MGMT_CONFIRM,
-	WL3501_SIG_REASSOC_CONFIRM,
-	WL3501_SIG_REASSOC_IND,
-	WL3501_SIG_SCAN_CONFIRM,
-	WL3501_SIG_SET_CONFIRM,
-	WL3501_SIG_START_CONFIRM,
-	WL3501_SIG_RESYNC_CONFIRM,
-	WL3501_SIG_SITE_CONFIRM,
-	WL3501_SIG_SAVE_CONFIRM,
-	WL3501_SIG_RFTEST_CONFIRM,
-/*
- * ID for input Signals of MLME block
- * bit[7-5] is block ID: 010
- * bit[4-0] is signal ID
- */
-	WL3501_SIG_ASSOC_REQ = 0x20,
-	WL3501_SIG_AUTH_REQ,
-	WL3501_SIG_DEAUTH_REQ,
-	WL3501_SIG_DISASSOC_REQ,
-	WL3501_SIG_GET_REQ,
-	WL3501_SIG_JOIN_REQ,
-	WL3501_SIG_PWR_MGMT_REQ,
-	WL3501_SIG_REASSOC_REQ,
-	WL3501_SIG_SCAN_REQ,
-	WL3501_SIG_SET_REQ,
-	WL3501_SIG_START_REQ,
-	WL3501_SIG_MD_REQ,
-	WL3501_SIG_RESYNC_REQ,
-	WL3501_SIG_SITE_REQ,
-	WL3501_SIG_SAVE_REQ,
-	WL3501_SIG_RF_TEST_REQ,
-	WL3501_SIG_MM_CONFIRM = 0x60,
-	WL3501_SIG_MM_IND,
-};
-
-enum wl3501_mib_attribs {
-	WL3501_MIB_ATTR_STATION_ID,
-	WL3501_MIB_ATTR_AUTH_ALGORITHMS,
-	WL3501_MIB_ATTR_AUTH_TYPE,
-	WL3501_MIB_ATTR_MEDIUM_OCCUPANCY_LIMIT,
-	WL3501_MIB_ATTR_CF_POLLABLE,
-	WL3501_MIB_ATTR_CFP_PERIOD,
-	WL3501_MIB_ATTR_CFPMAX_DURATION,
-	WL3501_MIB_ATTR_AUTH_RESP_TMOUT,
-	WL3501_MIB_ATTR_RX_DTIMS,
-	WL3501_MIB_ATTR_PRIV_OPT_IMPLEMENTED,
-	WL3501_MIB_ATTR_PRIV_INVOKED,
-	WL3501_MIB_ATTR_WEP_DEFAULT_KEYS,
-	WL3501_MIB_ATTR_WEP_DEFAULT_KEY_ID,
-	WL3501_MIB_ATTR_WEP_KEY_MAPPINGS,
-	WL3501_MIB_ATTR_WEP_KEY_MAPPINGS_LEN,
-	WL3501_MIB_ATTR_EXCLUDE_UNENCRYPTED,
-	WL3501_MIB_ATTR_WEP_ICV_ERROR_COUNT,
-	WL3501_MIB_ATTR_WEP_UNDECRYPTABLE_COUNT,
-	WL3501_MIB_ATTR_WEP_EXCLUDED_COUNT,
-	WL3501_MIB_ATTR_MAC_ADDR,
-	WL3501_MIB_ATTR_GROUP_ADDRS,
-	WL3501_MIB_ATTR_RTS_THRESHOLD,
-	WL3501_MIB_ATTR_SHORT_RETRY_LIMIT,
-	WL3501_MIB_ATTR_LONG_RETRY_LIMIT,
-	WL3501_MIB_ATTR_FRAG_THRESHOLD,
-	WL3501_MIB_ATTR_MAX_TX_MSDU_LIFETIME,
-	WL3501_MIB_ATTR_MAX_RX_LIFETIME,
-	WL3501_MIB_ATTR_MANUFACTURER_ID,
-	WL3501_MIB_ATTR_PRODUCT_ID,
-	WL3501_MIB_ATTR_TX_FRAG_COUNT,
-	WL3501_MIB_ATTR_MULTICAST_TX_FRAME_COUNT,
-	WL3501_MIB_ATTR_FAILED_COUNT,
-	WL3501_MIB_ATTR_RX_FRAG_COUNT,
-	WL3501_MIB_ATTR_MULTICAST_RX_COUNT,
-	WL3501_MIB_ATTR_FCS_ERROR_COUNT,
-	WL3501_MIB_ATTR_RETRY_COUNT,
-	WL3501_MIB_ATTR_MULTIPLE_RETRY_COUNT,
-	WL3501_MIB_ATTR_RTS_SUCCESS_COUNT,
-	WL3501_MIB_ATTR_RTS_FAILURE_COUNT,
-	WL3501_MIB_ATTR_ACK_FAILURE_COUNT,
-	WL3501_MIB_ATTR_FRAME_DUPLICATE_COUNT,
-	WL3501_MIB_ATTR_PHY_TYPE,
-	WL3501_MIB_ATTR_REG_DOMAINS_SUPPORT,
-	WL3501_MIB_ATTR_CURRENT_REG_DOMAIN,
-	WL3501_MIB_ATTR_SLOT_TIME,
-	WL3501_MIB_ATTR_CCA_TIME,
-	WL3501_MIB_ATTR_RX_TX_TURNAROUND_TIME,
-	WL3501_MIB_ATTR_TX_PLCP_DELAY,
-	WL3501_MIB_ATTR_RX_TX_SWITCH_TIME,
-	WL3501_MIB_ATTR_TX_RAMP_ON_TIME,
-	WL3501_MIB_ATTR_TX_RF_DELAY,
-	WL3501_MIB_ATTR_SIFS_TIME,
-	WL3501_MIB_ATTR_RX_RF_DELAY,
-	WL3501_MIB_ATTR_RX_PLCP_DELAY,
-	WL3501_MIB_ATTR_MAC_PROCESSING_DELAY,
-	WL3501_MIB_ATTR_TX_RAMP_OFF_TIME,
-	WL3501_MIB_ATTR_PREAMBLE_LEN,
-	WL3501_MIB_ATTR_PLCP_HEADER_LEN,
-	WL3501_MIB_ATTR_MPDU_DURATION_FACTOR,
-	WL3501_MIB_ATTR_AIR_PROPAGATION_TIME,
-	WL3501_MIB_ATTR_TEMP_TYPE,
-	WL3501_MIB_ATTR_CW_MIN,
-	WL3501_MIB_ATTR_CW_MAX,
-	WL3501_MIB_ATTR_SUPPORT_DATA_RATES_TX,
-	WL3501_MIB_ATTR_SUPPORT_DATA_RATES_RX,
-	WL3501_MIB_ATTR_MPDU_MAX_LEN,
-	WL3501_MIB_ATTR_SUPPORT_TX_ANTENNAS,
-	WL3501_MIB_ATTR_CURRENT_TX_ANTENNA,
-	WL3501_MIB_ATTR_SUPPORT_RX_ANTENNAS,
-	WL3501_MIB_ATTR_DIVERSITY_SUPPORT,
-	WL3501_MIB_ATTR_DIVERSITY_SELECTION_RS,
-	WL3501_MIB_ATTR_NR_SUPPORTED_PWR_LEVELS,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL1,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL2,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL3,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL4,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL5,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL6,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL7,
-	WL3501_MIB_ATTR_TX_PWR_LEVEL8,
-	WL3501_MIB_ATTR_CURRENT_TX_PWR_LEVEL,
-	WL3501_MIB_ATTR_CURRENT_CHAN,
-	WL3501_MIB_ATTR_CCA_MODE_SUPPORTED,
-	WL3501_MIB_ATTR_CURRENT_CCA_MODE,
-	WL3501_MIB_ATTR_ED_THRESHOLD,
-	WL3501_MIB_ATTR_SINTHESIZER_LOCKED,
-	WL3501_MIB_ATTR_CURRENT_PWR_STATE,
-	WL3501_MIB_ATTR_DOZE_TURNON_TIME,
-	WL3501_MIB_ATTR_RCR33,
-	WL3501_MIB_ATTR_DEFAULT_CHAN,
-	WL3501_MIB_ATTR_SSID,
-	WL3501_MIB_ATTR_PWR_MGMT_ENABLE,
-	WL3501_MIB_ATTR_NET_CAPABILITY,
-	WL3501_MIB_ATTR_ROUTING,
-};
-
-enum wl3501_net_type {
-	WL3501_NET_TYPE_INFRA,
-	WL3501_NET_TYPE_ADHOC,
-	WL3501_NET_TYPE_ANY_BSS,
-};
-
-enum wl3501_scan_type {
-	WL3501_SCAN_TYPE_ACTIVE,
-	WL3501_SCAN_TYPE_PASSIVE,
-};
-
-enum wl3501_tx_result {
-	WL3501_TX_RESULT_SUCCESS,
-	WL3501_TX_RESULT_NO_BSS,
-	WL3501_TX_RESULT_RETRY_LIMIT,
-};
-
-enum wl3501_sys_type {
-	WL3501_SYS_TYPE_OPEN,
-	WL3501_SYS_TYPE_SHARE_KEY,
-};
-
-enum wl3501_status {
-	WL3501_STATUS_SUCCESS,
-	WL3501_STATUS_INVALID,
-	WL3501_STATUS_TIMEOUT,
-	WL3501_STATUS_REFUSED,
-	WL3501_STATUS_MANY_REQ,
-	WL3501_STATUS_ALREADY_BSS,
-};
-
-#define WL3501_MGMT_CAPABILITY_ESS		0x0001  /* see 802.11 p.58 */
-#define WL3501_MGMT_CAPABILITY_IBSS		0x0002  /*      - " -	   */
-#define WL3501_MGMT_CAPABILITY_CF_POLLABLE	0x0004  /*      - " -	   */
-#define WL3501_MGMT_CAPABILITY_CF_POLL_REQUEST	0x0008  /*      - " -	   */
-#define WL3501_MGMT_CAPABILITY_PRIVACY		0x0010  /*      - " -	   */
-
-#define IW_REG_DOMAIN_FCC	0x10	/* Channel 1 to 11	USA    */
-#define IW_REG_DOMAIN_DOC	0x20	/* Channel 1 to 11	Canada */
-#define IW_REG_DOMAIN_ETSI	0x30	/* Channel 1 to 13	Europe */
-#define IW_REG_DOMAIN_SPAIN	0x31	/* Channel 10 to 11	Spain  */
-#define IW_REG_DOMAIN_FRANCE	0x32	/* Channel 10 to 13	France */
-#define IW_REG_DOMAIN_MKK	0x40	/* Channel 14		Japan  */
-#define IW_REG_DOMAIN_MKK1	0x41	/* Channel 1-14		Japan  */
-#define IW_REG_DOMAIN_ISRAEL	0x50	/* Channel 3 - 9	Israel */
-
-#define IW_MGMT_RATE_LABEL_MANDATORY 128 /* MSB */
-
-enum iw_mgmt_rate_labels {
-	IW_MGMT_RATE_LABEL_1MBIT   = 2,
-	IW_MGMT_RATE_LABEL_2MBIT   = 4,
-	IW_MGMT_RATE_LABEL_5_5MBIT = 11,
-	IW_MGMT_RATE_LABEL_11MBIT  = 22,
-};
-
-enum iw_mgmt_info_element_ids {
-	IW_MGMT_INFO_ELEMENT_SSID,		  /* Service Set Identity */
-	IW_MGMT_INFO_ELEMENT_SUPPORTED_RATES,
-	IW_MGMT_INFO_ELEMENT_FH_PARAMETER_SET,
-	IW_MGMT_INFO_ELEMENT_DS_PARAMETER_SET,
-	IW_MGMT_INFO_ELEMENT_CS_PARAMETER_SET,
-	IW_MGMT_INFO_ELEMENT_CS_TIM,		  /* Traffic Information Map */
-	IW_MGMT_INFO_ELEMENT_IBSS_PARAMETER_SET,
-	/* 7-15: Reserved, unused */
-	IW_MGMT_INFO_ELEMENT_CHALLENGE_TEXT = 16,
-	/* 17-31 Reserved for challenge text extension */
-	/* 32-255 Reserved, unused */
-};
-
-struct iw_mgmt_info_element {
-	u8 id; /* one of enum iw_mgmt_info_element_ids,
-		  but sizeof(enum) > sizeof(u8) :-( */
-	u8 len;
-	u8 data[];
-} __packed;
-
-struct iw_mgmt_essid_pset {
-	struct iw_mgmt_info_element el;
-	u8 			    essid[IW_ESSID_MAX_SIZE];
-} __packed;
-
-/*
- * According to 802.11 Wireless Networks, the definitive guide - O'Reilly
- * Pg 75
- */ 
-#define IW_DATA_RATE_MAX_LABELS 8
-
-struct iw_mgmt_data_rset {
-	struct iw_mgmt_info_element el;
-	u8 			    data_rate_labels[IW_DATA_RATE_MAX_LABELS];
-} __packed;
-
-struct iw_mgmt_ds_pset {
-	struct iw_mgmt_info_element el;
-	u8 			    chan;
-} __packed;
-
-struct iw_mgmt_cf_pset {
-	struct iw_mgmt_info_element el;
-	u8 			    cfp_count;
-	u8 			    cfp_period;
-	u16 			    cfp_max_duration;
-	u16 			    cfp_dur_remaining;
-} __packed;
-
-struct iw_mgmt_ibss_pset {
-	struct iw_mgmt_info_element el;
-	u16 			    atim_window;
-} __packed;
-
-struct wl3501_tx_hdr {
-	u16	tx_cnt;
-	u8	sync[16];
-	u16	sfd;
-	u8	signal;
-	u8	service;
-	u16	len;
-	u16	crc16;
-	u16	frame_ctrl;
-	u16	duration_id;
-	u8	addr1[ETH_ALEN];
-	u8	addr2[ETH_ALEN];
-	u8	addr3[ETH_ALEN];
-	u16	seq_ctrl;
-	u8	addr4[ETH_ALEN];
-};
-
-struct wl3501_rx_hdr {
-	u16	rx_next_blk;
-	u16	rc_next_frame_blk;
-	u8	rx_blk_ctrl;
-	u8	rx_next_frame;
-	u8	rx_next_frame1;
-	u8	rssi;
-	char	time[8];
-	u8	signal;
-	u8	service;
-	u16	len;
-	u16	crc16;
-	u16	frame_ctrl;
-	u16	duration;
-	u8	addr1[ETH_ALEN];
-	u8	addr2[ETH_ALEN];
-	u8	addr3[ETH_ALEN];
-	u16	seq;
-	u8	addr4[ETH_ALEN];
-};
-
-struct wl3501_start_req {
-	u16			    next_blk;
-	u8			    sig_id;
-	u8			    bss_type;
-	u16			    beacon_period;
-	u16			    dtim_period;
-	u16			    probe_delay;
-	u16			    cap_info;
-	struct iw_mgmt_essid_pset   ssid;
-	struct iw_mgmt_data_rset    bss_basic_rset;
-	struct iw_mgmt_data_rset    operational_rset;
-	struct iw_mgmt_cf_pset	    cf_pset;
-	struct iw_mgmt_ds_pset	    ds_pset;
-	struct iw_mgmt_ibss_pset    ibss_pset;
-};
-
-struct wl3501_assoc_req {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	timeout;
-	u16	cap_info;
-	u16	listen_interval;
-	u8	mac_addr[ETH_ALEN];
-};
-
-struct wl3501_assoc_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	status;
-};
-
-struct wl3501_assoc_ind {
-	u16	next_blk;
-	u8	sig_id;
-	u8	mac_addr[ETH_ALEN];
-};
-
-struct wl3501_auth_req {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	type;
-	u16	timeout;
-	u8	mac_addr[ETH_ALEN];
-};
-
-struct wl3501_auth_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	type;
-	u16	status;
-	u8	mac_addr[ETH_ALEN];
-};
-
-struct wl3501_get_req {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	mib_attrib;
-};
-
-struct wl3501_get_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	mib_status;
-	u16	mib_attrib;
-	u8	mib_value[100];
-};
-
-struct wl3501_req {
-	u16			    beacon_period;
-	u16			    dtim_period;
-	u16			    cap_info;
-	u8			    bss_type;
-	u8			    bssid[ETH_ALEN];
-	struct iw_mgmt_essid_pset   ssid;
-	struct iw_mgmt_ds_pset	    ds_pset;
-	struct iw_mgmt_cf_pset	    cf_pset;
-	struct iw_mgmt_ibss_pset    ibss_pset;
-	struct iw_mgmt_data_rset    bss_basic_rset;
-};
-
-struct wl3501_join_req {
-	u16			    next_blk;
-	u8			    sig_id;
-	u8			    reserved;
-	struct iw_mgmt_data_rset    operational_rset;
-	u16			    reserved2;
-	u16			    timeout;
-	u16			    probe_delay;
-	u8			    timestamp[8];
-	u8			    local_time[8];
-	struct wl3501_req	    req;
-};
-
-struct wl3501_join_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	status;
-};
-
-struct wl3501_pwr_mgmt_req {
-	u16	next_blk;
-	u8	sig_id;
-	u8	pwr_save;
-	u8	wake_up;
-	u8	receive_dtims;
-};
-
-struct wl3501_pwr_mgmt_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	status;
-};
-
-struct wl3501_scan_req {
-	u16			    next_blk;
-	u8			    sig_id;
-	u8			    bss_type;
-	u16			    probe_delay;
-	u16			    min_chan_time;
-	u16			    max_chan_time;
-	u8			    chan_list[14];
-	u8			    bssid[ETH_ALEN];
-	struct iw_mgmt_essid_pset   ssid;
-	enum wl3501_scan_type	    scan_type;
-};
-
-struct wl3501_scan_confirm {
-	u16			    next_blk;
-	u8			    sig_id;
-	u8			    reserved;
-	u16			    status;
-	char			    timestamp[8];
-	char			    localtime[8];
-	struct wl3501_req	    req;
-	u8			    rssi;
-};
-
-struct wl3501_start_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	status;
-};
-
-struct wl3501_md_req {
-	u16	next_blk;
-	u8	sig_id;
-	u8	routing;
-	u16	data;
-	u16	size;
-	u8	pri;
-	u8	service_class;
-	struct {
-		u8	daddr[ETH_ALEN];
-		u8	saddr[ETH_ALEN];
-	} addr;
-};
-
-struct wl3501_md_ind {
-	u16	next_blk;
-	u8	sig_id;
-	u8	routing;
-	u16	data;
-	u16	size;
-	u8	reception;
-	u8	pri;
-	u8	service_class;
-	struct {
-		u8	daddr[ETH_ALEN];
-		u8	saddr[ETH_ALEN];
-	} addr;
-};
-
-struct wl3501_md_confirm {
-	u16	next_blk;
-	u8	sig_id;
-	u8	reserved;
-	u16	data;
-	u8	status;
-	u8	pri;
-	u8	service_class;
-};
-
-struct wl3501_resync_req {
-	u16	next_blk;
-	u8	sig_id;
-};
-
-/* Definitions for supporting clone adapters. */
-/* System Interface Registers (SIR space) */
-#define WL3501_NIC_GCR ((u8)0x00)	/* SIR0 - General Conf Register */
-#define WL3501_NIC_BSS ((u8)0x01)	/* SIR1 - Bank Switching Select Reg */
-#define WL3501_NIC_LMAL ((u8)0x02)	/* SIR2 - Local Mem addr Reg [7:0] */
-#define WL3501_NIC_LMAH ((u8)0x03)	/* SIR3 - Local Mem addr Reg [14:8] */
-#define WL3501_NIC_IODPA ((u8)0x04)	/* SIR4 - I/O Data Port A */
-#define WL3501_NIC_IODPB ((u8)0x05)	/* SIR5 - I/O Data Port B */
-#define WL3501_NIC_IODPC ((u8)0x06)	/* SIR6 - I/O Data Port C */
-#define WL3501_NIC_IODPD ((u8)0x07)	/* SIR7 - I/O Data Port D */
-
-/* Bits in GCR */
-#define WL3501_GCR_SWRESET ((u8)0x80)
-#define WL3501_GCR_CORESET ((u8)0x40)
-#define WL3501_GCR_DISPWDN ((u8)0x20)
-#define WL3501_GCR_ECWAIT  ((u8)0x10)
-#define WL3501_GCR_ECINT   ((u8)0x08)
-#define WL3501_GCR_INT2EC  ((u8)0x04)
-#define WL3501_GCR_ENECINT ((u8)0x02)
-#define WL3501_GCR_DAM     ((u8)0x01)
-
-/* Bits in BSS (Bank Switching Select Register) */
-#define WL3501_BSS_FPAGE0 ((u8)0x20)	/* Flash memory page0 */
-#define WL3501_BSS_FPAGE1 ((u8)0x28)
-#define WL3501_BSS_FPAGE2 ((u8)0x30)
-#define WL3501_BSS_FPAGE3 ((u8)0x38)
-#define WL3501_BSS_SPAGE0 ((u8)0x00)	/* SRAM page0 */
-#define WL3501_BSS_SPAGE1 ((u8)0x08)
-#define WL3501_BSS_SPAGE2 ((u8)0x10)
-#define WL3501_BSS_SPAGE3 ((u8)0x18)
-
-/* Define Driver Interface */
-/* Refer IEEE 802.11 */
-/* Tx packet header, include PLCP and MPDU */
-/* Tx PLCP Header */
-struct wl3501_80211_tx_plcp_hdr {
-	u8	sync[16];
-	u16	sfd;
-	u8	signal;
-	u8	service;
-	u16	len;
-	u16	crc16;
-} __packed;
-
-struct wl3501_80211_tx_hdr {
-	struct wl3501_80211_tx_plcp_hdr	pclp_hdr;
-	struct ieee80211_hdr		mac_hdr;
-} __packed __aligned(2);
-
-/*
-   Reserve the beginning Tx space for descriptor use.
-
-   TxBlockOffset -->	*----*----*----*----* \
-	(TxFreeDesc)	|  0 |  1 |  2 |  3 |  \
-			|  4 |  5 |  6 |  7 |   |
-			|  8 |  9 | 10 | 11 |   TX_DESC * 20
-			| 12 | 13 | 14 | 15 |   |
-			| 16 | 17 | 18 | 19 |  /
-   TxBufferBegin -->	*----*----*----*----* /
-   (TxBufferHead)	| 		    |
-   (TxBufferTail)	| 		    |
-			|    Send Buffer    |
-			| 		    |
-			|		    |
-			*-------------------*
-   TxBufferEnd    -------------------------/
-
-*/
-
-struct wl3501_card {
-	int				base_addr;
-	u8				mac_addr[ETH_ALEN];
-	spinlock_t			lock;
-	wait_queue_head_t		wait;
-	struct wl3501_get_confirm	sig_get_confirm;
-	struct wl3501_pwr_mgmt_confirm	sig_pwr_mgmt_confirm;
-	u16				tx_buffer_size;
-	u16				tx_buffer_head;
-	u16				tx_buffer_tail;
-	u16				tx_buffer_cnt;
-	u16				esbq_req_start;
-	u16				esbq_req_end;
-	u16				esbq_req_head;
-	u16				esbq_req_tail;
-	u16				esbq_confirm_start;
-	u16				esbq_confirm_end;
-	u16				esbq_confirm;
-	struct iw_mgmt_essid_pset  	essid;
-	struct iw_mgmt_essid_pset  	keep_essid;
-	u8				bssid[ETH_ALEN];
-	int				net_type;
-	char				nick[32];
-	char				card_name[32];
-	char				firmware_date[32];
-	u8				chan;
-	u8				cap_info;
-	u16				start_seg;
-	u16				bss_cnt;
-	u16				join_sta_bss;
-	u8				rssi;
-	u8				adhoc_times;
-	u8				reg_domain;
-	u8				version[2];
-	struct wl3501_scan_confirm	bss_set[20];
-
-	struct iw_statistics		wstats;
-	struct iw_spy_data		spy_data;
-	struct iw_public_data		wireless_data;
-	struct pcmcia_device		*p_dev;
-};
-#endif
diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c
deleted file mode 100644
index 7fb2f9513476..000000000000
--- a/drivers/net/wireless/wl3501_cs.c
+++ /dev/null
@@ -1,2030 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * WL3501 Wireless LAN PCMCIA Card Driver for Linux
- * Written originally for Linux 2.0.30 by Fox Chen, mhchen@golf.ccl.itri.org.tw
- * Ported to 2.2, 2.4 & 2.5 by Arnaldo Carvalho de Melo <acme@conectiva.com.br>
- * Wireless extensions in 2.4 by Gustavo Niemeyer <niemeyer@conectiva.com>
- *
- * References used by Fox Chen while writing the original driver for 2.0.30:
- *
- *   1. WL24xx packet drivers (tooasm.asm)
- *   2. Access Point Firmware Interface Specification for IEEE 802.11 SUTRO
- *   3. IEEE 802.11
- *   4. Linux network driver (/usr/src/linux/drivers/net)
- *   5. ISA card driver - wl24.c
- *   6. Linux PCMCIA skeleton driver - skeleton.c
- *   7. Linux PCMCIA 3c589 network driver - 3c589_cs.c
- *
- * Tested with WL2400 firmware 1.2, Linux 2.0.30, and pcmcia-cs-2.9.12
- *   1. Performance: about 165 Kbytes/sec in TCP/IP with Ad-Hoc mode.
- *      rsh 192.168.1.3 "dd if=/dev/zero bs=1k count=1000" > /dev/null
- *      (Specification 2M bits/sec. is about 250 Kbytes/sec., but we must deduct
- *       ETHER/IP/UDP/TCP header, and acknowledgement overhead)
- *
- * Tested with Planet AP in 2.4.17, 184 Kbytes/s in UDP in Infrastructure mode,
- * 173 Kbytes/s in TCP.
- *
- * Tested with Planet AP in 2.5.73-bk, 216 Kbytes/s in Infrastructure mode
- * with a SMP machine (dual pentium 100), using pktgen, 432 pps (pkt_size = 60)
- */
-
-#include <linux/delay.h>
-#include <linux/types.h>
-#include <linux/interrupt.h>
-#include <linux/in.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/fcntl.h>
-#include <linux/if_arp.h>
-#include <linux/ioport.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/wireless.h>
-#include <net/cfg80211.h>
-
-#include <net/iw_handler.h>
-
-#include <pcmcia/cistpl.h>
-#include <pcmcia/cisreg.h>
-#include <pcmcia/ds.h>
-
-#include <asm/io.h>
-#include <linux/uaccess.h>
-
-#include "wl3501.h"
-
-#ifndef __i386__
-#define slow_down_io()
-#endif
-
-/* For rough constant delay */
-#define WL3501_NOPLOOP(n) { int x = 0; while (x++ < n) slow_down_io(); }
-
-
-
-#define wl3501_outb(a, b) { outb(a, b); slow_down_io(); }
-#define wl3501_outb_p(a, b) { outb_p(a, b); slow_down_io(); }
-#define wl3501_outsb(a, b, c) { outsb(a, b, c); slow_down_io(); }
-
-#define WL3501_RELEASE_TIMEOUT (25 * HZ)
-#define WL3501_MAX_ADHOC_TRIES 16
-
-#define WL3501_RESUME	0
-#define WL3501_SUSPEND	1
-
-static int wl3501_config(struct pcmcia_device *link);
-static void wl3501_release(struct pcmcia_device *link);
-
-static const struct {
-	int reg_domain;
-	int min, max, deflt;
-} iw_channel_table[] = {
-	{
-		.reg_domain = IW_REG_DOMAIN_FCC,
-		.min	    = 1,
-		.max	    = 11,
-		.deflt	    = 1,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_DOC,
-		.min	    = 1,
-		.max	    = 11,
-		.deflt	    = 1,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_ETSI,
-		.min	    = 1,
-		.max	    = 13,
-		.deflt	    = 1,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_SPAIN,
-		.min	    = 10,
-		.max	    = 11,
-		.deflt	    = 10,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_FRANCE,
-		.min	    = 10,
-		.max	    = 13,
-		.deflt	    = 10,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_MKK,
-		.min	    = 14,
-		.max	    = 14,
-		.deflt	    = 14,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_MKK1,
-		.min	    = 1,
-		.max	    = 14,
-		.deflt	    = 1,
-	},
-	{
-		.reg_domain = IW_REG_DOMAIN_ISRAEL,
-		.min	    = 3,
-		.max	    = 9,
-		.deflt	    = 9,
-	},
-};
-
-/**
- * iw_valid_channel - validate channel in regulatory domain
- * @reg_domain: regulatory domain
- * @channel: channel to validate
- *
- * Returns 0 if invalid in the specified regulatory domain, non-zero if valid.
- */
-static int iw_valid_channel(int reg_domain, int channel)
-{
-	int i, rc = 0;
-
-	for (i = 0; i < ARRAY_SIZE(iw_channel_table); i++)
-		if (reg_domain == iw_channel_table[i].reg_domain) {
-			rc = channel >= iw_channel_table[i].min &&
-			     channel <= iw_channel_table[i].max;
-			break;
-		}
-	return rc;
-}
-
-/**
- * iw_default_channel - get default channel for a regulatory domain
- * @reg_domain: regulatory domain
- *
- * Returns the default channel for a regulatory domain
- */
-static int iw_default_channel(int reg_domain)
-{
-	int i, rc = 1;
-
-	for (i = 0; i < ARRAY_SIZE(iw_channel_table); i++)
-		if (reg_domain == iw_channel_table[i].reg_domain) {
-			rc = iw_channel_table[i].deflt;
-			break;
-		}
-	return rc;
-}
-
-static void iw_set_mgmt_info_element(enum iw_mgmt_info_element_ids id,
-				     struct iw_mgmt_info_element *el,
-				     void *value, int len)
-{
-	el->id  = id;
-	el->len = len;
-	memcpy(el->data, value, len);
-}
-
-static void iw_copy_mgmt_info_element(struct iw_mgmt_info_element *to,
-				      struct iw_mgmt_info_element *from)
-{
-	iw_set_mgmt_info_element(from->id, to, from->data, from->len);
-}
-
-static inline void wl3501_switch_page(struct wl3501_card *this, u8 page)
-{
-	wl3501_outb(page, this->base_addr + WL3501_NIC_BSS);
-}
-
-/*
- * Get Ethernet MAC address.
- *
- * WARNING: We switch to FPAGE0 and switc back again.
- *          Making sure there is no other WL function beening called by ISR.
- */
-static int wl3501_get_flash_mac_addr(struct wl3501_card *this)
-{
-	int base_addr = this->base_addr;
-
-	/* get MAC addr */
-	wl3501_outb(WL3501_BSS_FPAGE3, base_addr + WL3501_NIC_BSS); /* BSS */
-	wl3501_outb(0x00, base_addr + WL3501_NIC_LMAL);	/* LMAL */
-	wl3501_outb(0x40, base_addr + WL3501_NIC_LMAH);	/* LMAH */
-
-	/* wait for reading EEPROM */
-	WL3501_NOPLOOP(100);
-	this->mac_addr[0] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->mac_addr[1] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->mac_addr[2] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->mac_addr[3] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->mac_addr[4] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->mac_addr[5] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->reg_domain = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	wl3501_outb(WL3501_BSS_FPAGE0, base_addr + WL3501_NIC_BSS);
-	wl3501_outb(0x04, base_addr + WL3501_NIC_LMAL);
-	wl3501_outb(0x40, base_addr + WL3501_NIC_LMAH);
-	WL3501_NOPLOOP(100);
-	this->version[0] = inb(base_addr + WL3501_NIC_IODPA);
-	WL3501_NOPLOOP(100);
-	this->version[1] = inb(base_addr + WL3501_NIC_IODPA);
-	/* switch to SRAM Page 0 (for safety) */
-	wl3501_switch_page(this, WL3501_BSS_SPAGE0);
-
-	/* The MAC addr should be 00:60:... */
-	return this->mac_addr[0] == 0x00 && this->mac_addr[1] == 0x60;
-}
-
-/**
- * wl3501_set_to_wla - Move 'size' bytes from PC to card
- * @this: Card
- * @dest: Card addressing space
- * @src: PC addressing space
- * @size: Bytes to move
- *
- * Move 'size' bytes from PC to card. (Shouldn't be interrupted)
- */
-static void wl3501_set_to_wla(struct wl3501_card *this, u16 dest, void *src,
-			      int size)
-{
-	/* switch to SRAM Page 0 */
-	wl3501_switch_page(this, (dest & 0x8000) ? WL3501_BSS_SPAGE1 :
-						   WL3501_BSS_SPAGE0);
-	/* set LMAL and LMAH */
-	wl3501_outb(dest & 0xff, this->base_addr + WL3501_NIC_LMAL);
-	wl3501_outb(((dest >> 8) & 0x7f), this->base_addr + WL3501_NIC_LMAH);
-
-	/* rep out to Port A */
-	wl3501_outsb(this->base_addr + WL3501_NIC_IODPA, src, size);
-}
-
-/**
- * wl3501_get_from_wla - Move 'size' bytes from card to PC
- * @this: Card
- * @src: Card addressing space
- * @dest: PC addressing space
- * @size: Bytes to move
- *
- * Move 'size' bytes from card to PC. (Shouldn't be interrupted)
- */
-static void wl3501_get_from_wla(struct wl3501_card *this, u16 src, void *dest,
-				int size)
-{
-	/* switch to SRAM Page 0 */
-	wl3501_switch_page(this, (src & 0x8000) ? WL3501_BSS_SPAGE1 :
-						  WL3501_BSS_SPAGE0);
-	/* set LMAL and LMAH */
-	wl3501_outb(src & 0xff, this->base_addr + WL3501_NIC_LMAL);
-	wl3501_outb((src >> 8) & 0x7f, this->base_addr + WL3501_NIC_LMAH);
-
-	/* rep get from Port A */
-	insb(this->base_addr + WL3501_NIC_IODPA, dest, size);
-}
-
-/*
- * Get/Allocate a free Tx Data Buffer
- *
- *  *--------------*-----------------*----------------------------------*
- *  |    PLCP      |    MAC Header   |  DST  SRC         Data ...       |
- *  |  (24 bytes)  |    (30 bytes)   |  (6)  (6)  (Ethernet Row Data)   |
- *  *--------------*-----------------*----------------------------------*
- *  \               \- IEEE 802.11 -/ \-------------- len --------------/
- *   \-struct wl3501_80211_tx_hdr--/   \-------- Ethernet Frame -------/
- *
- * Return = Position in Card
- */
-static u16 wl3501_get_tx_buffer(struct wl3501_card *this, u16 len)
-{
-	u16 next, blk_cnt = 0, zero = 0;
-	u16 full_len = sizeof(struct wl3501_80211_tx_hdr) + len;
-	u16 ret = 0;
-
-	if (full_len > this->tx_buffer_cnt * 254)
-		goto out;
-	ret = this->tx_buffer_head;
-	while (full_len) {
-		if (full_len < 254)
-			full_len = 0;
-		else
-			full_len -= 254;
-		wl3501_get_from_wla(this, this->tx_buffer_head, &next,
-				    sizeof(next));
-		if (!full_len)
-			wl3501_set_to_wla(this, this->tx_buffer_head, &zero,
-					  sizeof(zero));
-		this->tx_buffer_head = next;
-		blk_cnt++;
-		/* if buffer is not enough */
-		if (!next && full_len) {
-			this->tx_buffer_head = ret;
-			ret = 0;
-			goto out;
-		}
-	}
-	this->tx_buffer_cnt -= blk_cnt;
-out:
-	return ret;
-}
-
-/*
- * Free an allocated Tx Buffer. ptr must be correct position.
- */
-static void wl3501_free_tx_buffer(struct wl3501_card *this, u16 ptr)
-{
-	/* check if all space is not free */
-	if (!this->tx_buffer_head)
-		this->tx_buffer_head = ptr;
-	else
-		wl3501_set_to_wla(this, this->tx_buffer_tail,
-				  &ptr, sizeof(ptr));
-	while (ptr) {
-		u16 next;
-
-		this->tx_buffer_cnt++;
-		wl3501_get_from_wla(this, ptr, &next, sizeof(next));
-		this->tx_buffer_tail = ptr;
-		ptr = next;
-	}
-}
-
-static int wl3501_esbq_req_test(struct wl3501_card *this)
-{
-	u8 tmp = 0;
-
-	wl3501_get_from_wla(this, this->esbq_req_head + 3, &tmp, sizeof(tmp));
-	return tmp & 0x80;
-}
-
-static void wl3501_esbq_req(struct wl3501_card *this, u16 *ptr)
-{
-	u16 tmp = 0;
-
-	wl3501_set_to_wla(this, this->esbq_req_head, ptr, 2);
-	wl3501_set_to_wla(this, this->esbq_req_head + 2, &tmp, sizeof(tmp));
-	this->esbq_req_head += 4;
-	if (this->esbq_req_head >= this->esbq_req_end)
-		this->esbq_req_head = this->esbq_req_start;
-}
-
-static int wl3501_esbq_exec(struct wl3501_card *this, void *sig, int sig_size)
-{
-	int rc = -EIO;
-
-	if (wl3501_esbq_req_test(this)) {
-		u16 ptr = wl3501_get_tx_buffer(this, sig_size);
-		if (ptr) {
-			wl3501_set_to_wla(this, ptr, sig, sig_size);
-			wl3501_esbq_req(this, &ptr);
-			rc = 0;
-		}
-	}
-	return rc;
-}
-
-static int wl3501_request_mib(struct wl3501_card *this, u8 index, void *bf)
-{
-	struct wl3501_get_req sig = {
-		.sig_id	    = WL3501_SIG_GET_REQ,
-		.mib_attrib = index,
-	};
-	unsigned long flags;
-	int rc = -EIO;
-
-	spin_lock_irqsave(&this->lock, flags);
-	if (wl3501_esbq_req_test(this)) {
-		u16 ptr = wl3501_get_tx_buffer(this, sizeof(sig));
-		if (ptr) {
-			wl3501_set_to_wla(this, ptr, &sig, sizeof(sig));
-			wl3501_esbq_req(this, &ptr);
-			this->sig_get_confirm.mib_status = 255;
-			rc = 0;
-		}
-	}
-	spin_unlock_irqrestore(&this->lock, flags);
-
-	return rc;
-}
-
-static int wl3501_get_mib_value(struct wl3501_card *this, u8 index,
-				void *bf, int size)
-{
-	int rc;
-
-	rc = wl3501_request_mib(this, index, bf);
-	if (rc)
-		return rc;
-
-	rc = wait_event_interruptible(this->wait,
-		this->sig_get_confirm.mib_status != 255);
-	if (rc)
-		return rc;
-
-	memcpy(bf, this->sig_get_confirm.mib_value, size);
-	return 0;
-}
-
-static int wl3501_pwr_mgmt(struct wl3501_card *this, int suspend)
-{
-	struct wl3501_pwr_mgmt_req sig = {
-		.sig_id		= WL3501_SIG_PWR_MGMT_REQ,
-		.pwr_save	= suspend,
-		.wake_up	= !suspend,
-		.receive_dtims	= 10,
-	};
-	unsigned long flags;
-	int rc = -EIO;
-
-	spin_lock_irqsave(&this->lock, flags);
-	if (wl3501_esbq_req_test(this)) {
-		u16 ptr = wl3501_get_tx_buffer(this, sizeof(sig));
-		if (ptr) {
-			wl3501_set_to_wla(this, ptr, &sig, sizeof(sig));
-			wl3501_esbq_req(this, &ptr);
-			this->sig_pwr_mgmt_confirm.status = 255;
-			spin_unlock_irqrestore(&this->lock, flags);
-			rc = wait_event_interruptible(this->wait,
-				this->sig_pwr_mgmt_confirm.status != 255);
-			printk(KERN_INFO "%s: %s status=%d\n", __func__,
-			       suspend ? "suspend" : "resume",
-			       this->sig_pwr_mgmt_confirm.status);
-			goto out;
-		}
-	}
-	spin_unlock_irqrestore(&this->lock, flags);
-out:
-	return rc;
-}
-
-/**
- * wl3501_send_pkt - Send a packet.
- * @this: Card
- * @data: Ethernet raw frame.  (e.g. data[0] - data[5] is Dest MAC Addr,
- *                                   data[6] - data[11] is Src MAC Addr)
- * @len: Packet length
- * Ref: IEEE 802.11
- */
-static int wl3501_send_pkt(struct wl3501_card *this, u8 *data, u16 len)
-{
-	u16 bf, sig_bf, next, tmplen, pktlen;
-	struct wl3501_md_req sig = {
-		.sig_id = WL3501_SIG_MD_REQ,
-	};
-	size_t sig_addr_len = sizeof(sig.addr);
-	u8 *pdata = (char *)data;
-	int rc = -EIO;
-
-	if (wl3501_esbq_req_test(this)) {
-		sig_bf = wl3501_get_tx_buffer(this, sizeof(sig));
-		rc = -ENOMEM;
-		if (!sig_bf)	/* No free buffer available */
-			goto out;
-		bf = wl3501_get_tx_buffer(this, len + 26 + 24);
-		if (!bf) {
-			/* No free buffer available */
-			wl3501_free_tx_buffer(this, sig_bf);
-			goto out;
-		}
-		rc = 0;
-		memcpy(&sig.addr, pdata, sig_addr_len);
-		pktlen = len - sig_addr_len;
-		pdata += sig_addr_len;
-		sig.data = bf;
-		if (((*pdata) * 256 + (*(pdata + 1))) > 1500) {
-			u8 addr4[ETH_ALEN] = {
-				[0] = 0xAA, [1] = 0xAA, [2] = 0x03, [4] = 0x00,
-			};
-
-			wl3501_set_to_wla(this, bf + 2 +
-					  offsetof(struct wl3501_tx_hdr, addr4),
-					  addr4, sizeof(addr4));
-			sig.size = pktlen + 24 + 4 + 6;
-			if (pktlen > (254 - sizeof(struct wl3501_tx_hdr))) {
-				tmplen = 254 - sizeof(struct wl3501_tx_hdr);
-				pktlen -= tmplen;
-			} else {
-				tmplen = pktlen;
-				pktlen = 0;
-			}
-			wl3501_set_to_wla(this,
-					  bf + 2 + sizeof(struct wl3501_tx_hdr),
-					  pdata, tmplen);
-			pdata += tmplen;
-			wl3501_get_from_wla(this, bf, &next, sizeof(next));
-			bf = next;
-		} else {
-			sig.size = pktlen + 24 + 4 - 2;
-			pdata += 2;
-			pktlen -= 2;
-			if (pktlen > (254 - sizeof(struct wl3501_tx_hdr) + 6)) {
-				tmplen = 254 - sizeof(struct wl3501_tx_hdr) + 6;
-				pktlen -= tmplen;
-			} else {
-				tmplen = pktlen;
-				pktlen = 0;
-			}
-			wl3501_set_to_wla(this, bf + 2 +
-					  offsetof(struct wl3501_tx_hdr, addr4),
-					  pdata, tmplen);
-			pdata += tmplen;
-			wl3501_get_from_wla(this, bf, &next, sizeof(next));
-			bf = next;
-		}
-		while (pktlen > 0) {
-			if (pktlen > 254) {
-				tmplen = 254;
-				pktlen -= 254;
-			} else {
-				tmplen = pktlen;
-				pktlen = 0;
-			}
-			wl3501_set_to_wla(this, bf + 2, pdata, tmplen);
-			pdata += tmplen;
-			wl3501_get_from_wla(this, bf, &next, sizeof(next));
-			bf = next;
-		}
-		wl3501_set_to_wla(this, sig_bf, &sig, sizeof(sig));
-		wl3501_esbq_req(this, &sig_bf);
-	}
-out:
-	return rc;
-}
-
-static int wl3501_mgmt_resync(struct wl3501_card *this)
-{
-	struct wl3501_resync_req sig = {
-		.sig_id = WL3501_SIG_RESYNC_REQ,
-	};
-
-	return wl3501_esbq_exec(this, &sig, sizeof(sig));
-}
-
-static inline int wl3501_fw_bss_type(struct wl3501_card *this)
-{
-	return this->net_type == IW_MODE_INFRA ? WL3501_NET_TYPE_INFRA :
-						 WL3501_NET_TYPE_ADHOC;
-}
-
-static inline int wl3501_fw_cap_info(struct wl3501_card *this)
-{
-	return this->net_type == IW_MODE_INFRA ? WL3501_MGMT_CAPABILITY_ESS :
-						 WL3501_MGMT_CAPABILITY_IBSS;
-}
-
-static int wl3501_mgmt_scan(struct wl3501_card *this, u16 chan_time)
-{
-	struct wl3501_scan_req sig = {
-		.sig_id		= WL3501_SIG_SCAN_REQ,
-		.scan_type	= WL3501_SCAN_TYPE_ACTIVE,
-		.probe_delay	= 0x10,
-		.min_chan_time	= chan_time,
-		.max_chan_time	= chan_time,
-		.bss_type	= wl3501_fw_bss_type(this),
-	};
-
-	this->bss_cnt = this->join_sta_bss = 0;
-	return wl3501_esbq_exec(this, &sig, sizeof(sig));
-}
-
-static int wl3501_mgmt_join(struct wl3501_card *this, u16 stas)
-{
-	struct wl3501_join_req sig = {
-		.sig_id		  = WL3501_SIG_JOIN_REQ,
-		.timeout	  = 10,
-		.req.ds_pset = {
-			.el = {
-				.id  = IW_MGMT_INFO_ELEMENT_DS_PARAMETER_SET,
-				.len = 1,
-			},
-			.chan	= this->chan,
-		},
-	};
-
-	memcpy(&sig.req, &this->bss_set[stas].req, sizeof(sig.req));
-	return wl3501_esbq_exec(this, &sig, sizeof(sig));
-}
-
-static int wl3501_mgmt_start(struct wl3501_card *this)
-{
-	struct wl3501_start_req sig = {
-		.sig_id			= WL3501_SIG_START_REQ,
-		.beacon_period		= 400,
-		.dtim_period		= 1,
-		.ds_pset = {
-			.el = {
-				.id  = IW_MGMT_INFO_ELEMENT_DS_PARAMETER_SET,
-				.len = 1,
-			},
-			.chan	= this->chan,
-		},
-		.bss_basic_rset	= {
-			.el = {
-				.id	= IW_MGMT_INFO_ELEMENT_SUPPORTED_RATES,
-				.len = 2,
-			},
-			.data_rate_labels = {
-				[0] = IW_MGMT_RATE_LABEL_MANDATORY |
-				      IW_MGMT_RATE_LABEL_1MBIT,
-				[1] = IW_MGMT_RATE_LABEL_MANDATORY |
-				      IW_MGMT_RATE_LABEL_2MBIT,
-			},
-		},
-		.operational_rset	= {
-			.el = {
-				.id	= IW_MGMT_INFO_ELEMENT_SUPPORTED_RATES,
-				.len = 2,
-			},
-			.data_rate_labels = {
-				[0] = IW_MGMT_RATE_LABEL_MANDATORY |
-				      IW_MGMT_RATE_LABEL_1MBIT,
-				[1] = IW_MGMT_RATE_LABEL_MANDATORY |
-				      IW_MGMT_RATE_LABEL_2MBIT,
-			},
-		},
-		.ibss_pset		= {
-			.el = {
-				.id	 = IW_MGMT_INFO_ELEMENT_IBSS_PARAMETER_SET,
-				.len     = 2,
-			},
-			.atim_window = 10,
-		},
-		.bss_type		= wl3501_fw_bss_type(this),
-		.cap_info		= wl3501_fw_cap_info(this),
-	};
-
-	iw_copy_mgmt_info_element(&sig.ssid.el, &this->essid.el);
-	iw_copy_mgmt_info_element(&this->keep_essid.el, &this->essid.el);
-	return wl3501_esbq_exec(this, &sig, sizeof(sig));
-}
-
-static void wl3501_mgmt_scan_confirm(struct wl3501_card *this, u16 addr)
-{
-	u16 i = 0;
-	int matchflag = 0;
-	struct wl3501_scan_confirm sig;
-
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-	if (sig.status == WL3501_STATUS_SUCCESS) {
-		pr_debug("success");
-		if ((this->net_type == IW_MODE_INFRA &&
-		     (sig.req.cap_info & WL3501_MGMT_CAPABILITY_ESS)) ||
-		    (this->net_type == IW_MODE_ADHOC &&
-		     (sig.req.cap_info & WL3501_MGMT_CAPABILITY_IBSS)) ||
-		    this->net_type == IW_MODE_AUTO) {
-			if (!this->essid.el.len)
-				matchflag = 1;
-			else if (this->essid.el.len == 3 &&
-				 !memcmp(this->essid.essid, "ANY", 3))
-				matchflag = 1;
-			else if (this->essid.el.len != sig.req.ssid.el.len)
-				matchflag = 0;
-			else if (memcmp(this->essid.essid, sig.req.ssid.essid,
-					this->essid.el.len))
-				matchflag = 0;
-			else
-				matchflag = 1;
-			if (matchflag) {
-				for (i = 0; i < this->bss_cnt; i++) {
-					if (ether_addr_equal_unaligned(this->bss_set[i].req.bssid,
-								       sig.req.bssid)) {
-						matchflag = 0;
-						break;
-					}
-				}
-			}
-			if (matchflag && (i < 20)) {
-				memcpy(&this->bss_set[i].req,
-				       &sig.req, sizeof(sig.req));
-				this->bss_cnt++;
-				this->rssi = sig.rssi;
-				this->bss_set[i].rssi = sig.rssi;
-			}
-		}
-	} else if (sig.status == WL3501_STATUS_TIMEOUT) {
-		pr_debug("timeout");
-		this->join_sta_bss = 0;
-		for (i = this->join_sta_bss; i < this->bss_cnt; i++)
-			if (!wl3501_mgmt_join(this, i))
-				break;
-		this->join_sta_bss = i;
-		if (this->join_sta_bss == this->bss_cnt) {
-			if (this->net_type == IW_MODE_INFRA)
-				wl3501_mgmt_scan(this, 100);
-			else {
-				this->adhoc_times++;
-				if (this->adhoc_times > WL3501_MAX_ADHOC_TRIES)
-					wl3501_mgmt_start(this);
-				else
-					wl3501_mgmt_scan(this, 100);
-			}
-		}
-	}
-}
-
-/**
- * wl3501_block_interrupt - Mask interrupt from SUTRO
- * @this: Card
- *
- * Mask interrupt from SUTRO. (i.e. SUTRO cannot interrupt the HOST)
- * Return: 1 if interrupt is originally enabled
- */
-static int wl3501_block_interrupt(struct wl3501_card *this)
-{
-	u8 old = inb(this->base_addr + WL3501_NIC_GCR);
-	u8 new = old & (~(WL3501_GCR_ECINT | WL3501_GCR_INT2EC |
-			WL3501_GCR_ENECINT));
-
-	wl3501_outb(new, this->base_addr + WL3501_NIC_GCR);
-	return old & WL3501_GCR_ENECINT;
-}
-
-/**
- * wl3501_unblock_interrupt - Enable interrupt from SUTRO
- * @this: Card
- *
- * Enable interrupt from SUTRO. (i.e. SUTRO can interrupt the HOST)
- * Return: 1 if interrupt is originally enabled
- */
-static int wl3501_unblock_interrupt(struct wl3501_card *this)
-{
-	u8 old = inb(this->base_addr + WL3501_NIC_GCR);
-	u8 new = (old & ~(WL3501_GCR_ECINT | WL3501_GCR_INT2EC)) |
-		  WL3501_GCR_ENECINT;
-
-	wl3501_outb(new, this->base_addr + WL3501_NIC_GCR);
-	return old & WL3501_GCR_ENECINT;
-}
-
-/**
- * wl3501_receive - Receive data from Receive Queue.
- *
- * Receive data from Receive Queue.
- *
- * @this: card
- * @bf: address of host
- * @size: size of buffer.
- */
-static u16 wl3501_receive(struct wl3501_card *this, u8 *bf, u16 size)
-{
-	u16 next_addr, next_addr1;
-	u8 *data = bf + 12;
-
-	size -= 12;
-	wl3501_get_from_wla(this, this->start_seg + 2,
-			    &next_addr, sizeof(next_addr));
-	if (size > WL3501_BLKSZ - sizeof(struct wl3501_rx_hdr)) {
-		wl3501_get_from_wla(this,
-				    this->start_seg +
-					sizeof(struct wl3501_rx_hdr), data,
-				    WL3501_BLKSZ -
-					sizeof(struct wl3501_rx_hdr));
-		size -= WL3501_BLKSZ - sizeof(struct wl3501_rx_hdr);
-		data += WL3501_BLKSZ - sizeof(struct wl3501_rx_hdr);
-	} else {
-		wl3501_get_from_wla(this,
-				    this->start_seg +
-					sizeof(struct wl3501_rx_hdr),
-				    data, size);
-		size = 0;
-	}
-	while (size > 0) {
-		if (size > WL3501_BLKSZ - 5) {
-			wl3501_get_from_wla(this, next_addr + 5, data,
-					    WL3501_BLKSZ - 5);
-			size -= WL3501_BLKSZ - 5;
-			data += WL3501_BLKSZ - 5;
-			wl3501_get_from_wla(this, next_addr + 2, &next_addr1,
-					    sizeof(next_addr1));
-			next_addr = next_addr1;
-		} else {
-			wl3501_get_from_wla(this, next_addr + 5, data, size);
-			size = 0;
-		}
-	}
-	return 0;
-}
-
-static void wl3501_esbq_req_free(struct wl3501_card *this)
-{
-	u8 tmp;
-	u16 addr;
-
-	if (this->esbq_req_head == this->esbq_req_tail)
-		goto out;
-	wl3501_get_from_wla(this, this->esbq_req_tail + 3, &tmp, sizeof(tmp));
-	if (!(tmp & 0x80))
-		goto out;
-	wl3501_get_from_wla(this, this->esbq_req_tail, &addr, sizeof(addr));
-	wl3501_free_tx_buffer(this, addr);
-	this->esbq_req_tail += 4;
-	if (this->esbq_req_tail >= this->esbq_req_end)
-		this->esbq_req_tail = this->esbq_req_start;
-out:
-	return;
-}
-
-static int wl3501_esbq_confirm(struct wl3501_card *this)
-{
-	u8 tmp;
-
-	wl3501_get_from_wla(this, this->esbq_confirm + 3, &tmp, sizeof(tmp));
-	return tmp & 0x80;
-}
-
-static void wl3501_online(struct net_device *dev)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	printk(KERN_INFO "%s: Wireless LAN online. BSSID: %pM\n",
-	       dev->name, this->bssid);
-	netif_wake_queue(dev);
-}
-
-static void wl3501_esbq_confirm_done(struct wl3501_card *this)
-{
-	u8 tmp = 0;
-
-	wl3501_set_to_wla(this, this->esbq_confirm + 3, &tmp, sizeof(tmp));
-	this->esbq_confirm += 4;
-	if (this->esbq_confirm >= this->esbq_confirm_end)
-		this->esbq_confirm = this->esbq_confirm_start;
-}
-
-static int wl3501_mgmt_auth(struct wl3501_card *this)
-{
-	struct wl3501_auth_req sig = {
-		.sig_id	 = WL3501_SIG_AUTH_REQ,
-		.type	 = WL3501_SYS_TYPE_OPEN,
-		.timeout = 1000,
-	};
-
-	pr_debug("entry");
-	memcpy(sig.mac_addr, this->bssid, ETH_ALEN);
-	return wl3501_esbq_exec(this, &sig, sizeof(sig));
-}
-
-static int wl3501_mgmt_association(struct wl3501_card *this)
-{
-	struct wl3501_assoc_req sig = {
-		.sig_id		 = WL3501_SIG_ASSOC_REQ,
-		.timeout	 = 1000,
-		.listen_interval = 5,
-		.cap_info	 = this->cap_info,
-	};
-
-	pr_debug("entry");
-	memcpy(sig.mac_addr, this->bssid, ETH_ALEN);
-	return wl3501_esbq_exec(this, &sig, sizeof(sig));
-}
-
-static void wl3501_mgmt_join_confirm(struct net_device *dev, u16 addr)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	struct wl3501_join_confirm sig;
-
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-	if (sig.status == WL3501_STATUS_SUCCESS) {
-		if (this->net_type == IW_MODE_INFRA) {
-			if (this->join_sta_bss < this->bss_cnt) {
-				const int i = this->join_sta_bss;
-				memcpy(this->bssid,
-				       this->bss_set[i].req.bssid, ETH_ALEN);
-				this->chan = this->bss_set[i].req.ds_pset.chan;
-				iw_copy_mgmt_info_element(&this->keep_essid.el,
-						     &this->bss_set[i].req.ssid.el);
-				wl3501_mgmt_auth(this);
-			}
-		} else {
-			const int i = this->join_sta_bss;
-
-			memcpy(&this->bssid, &this->bss_set[i].req.bssid, ETH_ALEN);
-			this->chan = this->bss_set[i].req.ds_pset.chan;
-			iw_copy_mgmt_info_element(&this->keep_essid.el,
-						  &this->bss_set[i].req.ssid.el);
-			wl3501_online(dev);
-		}
-	} else {
-		int i;
-		this->join_sta_bss++;
-		for (i = this->join_sta_bss; i < this->bss_cnt; i++)
-			if (!wl3501_mgmt_join(this, i))
-				break;
-		this->join_sta_bss = i;
-		if (this->join_sta_bss == this->bss_cnt) {
-			if (this->net_type == IW_MODE_INFRA)
-				wl3501_mgmt_scan(this, 100);
-			else {
-				this->adhoc_times++;
-				if (this->adhoc_times > WL3501_MAX_ADHOC_TRIES)
-					wl3501_mgmt_start(this);
-				else
-					wl3501_mgmt_scan(this, 100);
-			}
-		}
-	}
-}
-
-static inline void wl3501_alarm_interrupt(struct net_device *dev,
-					  struct wl3501_card *this)
-{
-	if (this->net_type == IW_MODE_INFRA) {
-		printk(KERN_INFO "Wireless LAN offline\n");
-		netif_stop_queue(dev);
-		wl3501_mgmt_resync(this);
-	}
-}
-
-static inline void wl3501_md_confirm_interrupt(struct net_device *dev,
-					       struct wl3501_card *this,
-					       u16 addr)
-{
-	struct wl3501_md_confirm sig;
-
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-	wl3501_free_tx_buffer(this, sig.data);
-	if (netif_queue_stopped(dev))
-		netif_wake_queue(dev);
-}
-
-static inline void wl3501_md_ind_interrupt(struct net_device *dev,
-					   struct wl3501_card *this, u16 addr)
-{
-	struct wl3501_md_ind sig;
-	struct sk_buff *skb;
-	u8 rssi, addr4[ETH_ALEN];
-	u16 pkt_len;
-
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-	this->start_seg = sig.data;
-	wl3501_get_from_wla(this,
-			    sig.data + offsetof(struct wl3501_rx_hdr, rssi),
-			    &rssi, sizeof(rssi));
-	this->rssi = rssi <= 63 ? (rssi * 100) / 64 : 255;
-
-	wl3501_get_from_wla(this,
-			    sig.data +
-				offsetof(struct wl3501_rx_hdr, addr4),
-			    &addr4, sizeof(addr4));
-	if (!(addr4[0] == 0xAA && addr4[1] == 0xAA &&
-	      addr4[2] == 0x03 && addr4[4] == 0x00)) {
-		printk(KERN_INFO "Unsupported packet type!\n");
-		return;
-	}
-	pkt_len = sig.size + 12 - 24 - 4 - 6;
-
-	skb = dev_alloc_skb(pkt_len + 5);
-
-	if (!skb) {
-		printk(KERN_WARNING "%s: Can't alloc a sk_buff of size %d.\n",
-		       dev->name, pkt_len);
-		dev->stats.rx_dropped++;
-	} else {
-		skb->dev = dev;
-		skb_reserve(skb, 2); /* IP headers on 16 bytes boundaries */
-		skb_copy_to_linear_data(skb, (unsigned char *)&sig.addr,
-					sizeof(sig.addr));
-		wl3501_receive(this, skb->data, pkt_len);
-		skb_put(skb, pkt_len);
-		skb->protocol	= eth_type_trans(skb, dev);
-		dev->stats.rx_packets++;
-		dev->stats.rx_bytes += skb->len;
-		netif_rx(skb);
-	}
-}
-
-static inline void wl3501_get_confirm_interrupt(struct wl3501_card *this,
-						u16 addr, void *sig, int size)
-{
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &this->sig_get_confirm,
-			    sizeof(this->sig_get_confirm));
-	wake_up(&this->wait);
-}
-
-static inline void wl3501_start_confirm_interrupt(struct net_device *dev,
-						  struct wl3501_card *this,
-						  u16 addr)
-{
-	struct wl3501_start_confirm sig;
-
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-	if (sig.status == WL3501_STATUS_SUCCESS)
-		netif_wake_queue(dev);
-}
-
-static inline void wl3501_assoc_confirm_interrupt(struct net_device *dev,
-						  u16 addr)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	struct wl3501_assoc_confirm sig;
-
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-
-	if (sig.status == WL3501_STATUS_SUCCESS)
-		wl3501_online(dev);
-}
-
-static inline void wl3501_auth_confirm_interrupt(struct wl3501_card *this,
-						 u16 addr)
-{
-	struct wl3501_auth_confirm sig;
-
-	pr_debug("entry");
-	wl3501_get_from_wla(this, addr, &sig, sizeof(sig));
-
-	if (sig.status == WL3501_STATUS_SUCCESS)
-		wl3501_mgmt_association(this);
-	else
-		wl3501_mgmt_resync(this);
-}
-
-static inline void wl3501_rx_interrupt(struct net_device *dev)
-{
-	int morepkts;
-	u16 addr;
-	u8 sig_id;
-	struct wl3501_card *this = netdev_priv(dev);
-
-	pr_debug("entry");
-loop:
-	morepkts = 0;
-	if (!wl3501_esbq_confirm(this))
-		goto free;
-	wl3501_get_from_wla(this, this->esbq_confirm, &addr, sizeof(addr));
-	wl3501_get_from_wla(this, addr + 2, &sig_id, sizeof(sig_id));
-
-	switch (sig_id) {
-	case WL3501_SIG_DEAUTH_IND:
-	case WL3501_SIG_DISASSOC_IND:
-	case WL3501_SIG_ALARM:
-		wl3501_alarm_interrupt(dev, this);
-		break;
-	case WL3501_SIG_MD_CONFIRM:
-		wl3501_md_confirm_interrupt(dev, this, addr);
-		break;
-	case WL3501_SIG_MD_IND:
-		wl3501_md_ind_interrupt(dev, this, addr);
-		break;
-	case WL3501_SIG_GET_CONFIRM:
-		wl3501_get_confirm_interrupt(this, addr,
-					     &this->sig_get_confirm,
-					     sizeof(this->sig_get_confirm));
-		break;
-	case WL3501_SIG_PWR_MGMT_CONFIRM:
-		wl3501_get_confirm_interrupt(this, addr,
-					     &this->sig_pwr_mgmt_confirm,
-					    sizeof(this->sig_pwr_mgmt_confirm));
-		break;
-	case WL3501_SIG_START_CONFIRM:
-		wl3501_start_confirm_interrupt(dev, this, addr);
-		break;
-	case WL3501_SIG_SCAN_CONFIRM:
-		wl3501_mgmt_scan_confirm(this, addr);
-		break;
-	case WL3501_SIG_JOIN_CONFIRM:
-		wl3501_mgmt_join_confirm(dev, addr);
-		break;
-	case WL3501_SIG_ASSOC_CONFIRM:
-		wl3501_assoc_confirm_interrupt(dev, addr);
-		break;
-	case WL3501_SIG_AUTH_CONFIRM:
-		wl3501_auth_confirm_interrupt(this, addr);
-		break;
-	case WL3501_SIG_RESYNC_CONFIRM:
-		wl3501_mgmt_resync(this); /* FIXME: should be resync_confirm */
-		break;
-	}
-	wl3501_esbq_confirm_done(this);
-	morepkts = 1;
-	/* free request if necessary */
-free:
-	wl3501_esbq_req_free(this);
-	if (morepkts)
-		goto loop;
-}
-
-static inline void wl3501_ack_interrupt(struct wl3501_card *this)
-{
-	wl3501_outb(WL3501_GCR_ECINT, this->base_addr + WL3501_NIC_GCR);
-}
-
-/**
- * wl3501_interrupt - Hardware interrupt from card.
- * @irq: Interrupt number
- * @dev_id: net_device
- *
- * We must acknowledge the interrupt as soon as possible, and block the
- * interrupt from the same card immediately to prevent re-entry.
- *
- * Before accessing the Control_Status_Block, we must lock SUTRO first.
- * On the other hand, to prevent SUTRO from malfunctioning, we must
- * unlock the SUTRO as soon as possible.
- */
-static irqreturn_t wl3501_interrupt(int irq, void *dev_id)
-{
-	struct net_device *dev = dev_id;
-	struct wl3501_card *this;
-
-	this = netdev_priv(dev);
-	spin_lock(&this->lock);
-	wl3501_ack_interrupt(this);
-	wl3501_block_interrupt(this);
-	wl3501_rx_interrupt(dev);
-	wl3501_unblock_interrupt(this);
-	spin_unlock(&this->lock);
-
-	return IRQ_HANDLED;
-}
-
-static int wl3501_reset_board(struct wl3501_card *this)
-{
-	u8 tmp = 0;
-	int i, rc = 0;
-
-	/* Coreset */
-	wl3501_outb_p(WL3501_GCR_CORESET, this->base_addr + WL3501_NIC_GCR);
-	wl3501_outb_p(0, this->base_addr + WL3501_NIC_GCR);
-	wl3501_outb_p(WL3501_GCR_CORESET, this->base_addr + WL3501_NIC_GCR);
-
-	/* Reset SRAM 0x480 to zero */
-	wl3501_set_to_wla(this, 0x480, &tmp, sizeof(tmp));
-
-	/* Start up */
-	wl3501_outb_p(0, this->base_addr + WL3501_NIC_GCR);
-
-	WL3501_NOPLOOP(1024 * 50);
-
-	wl3501_unblock_interrupt(this);	/* acme: was commented */
-
-	/* Polling Self_Test_Status */
-	for (i = 0; i < 10000; i++) {
-		wl3501_get_from_wla(this, 0x480, &tmp, sizeof(tmp));
-
-		if (tmp == 'W') {
-			/* firmware complete all test successfully */
-			tmp = 'A';
-			wl3501_set_to_wla(this, 0x480, &tmp, sizeof(tmp));
-			goto out;
-		}
-		WL3501_NOPLOOP(10);
-	}
-	printk(KERN_WARNING "%s: failed to reset the board!\n", __func__);
-	rc = -ENODEV;
-out:
-	return rc;
-}
-
-static int wl3501_init_firmware(struct wl3501_card *this)
-{
-	u16 ptr, next;
-	int rc = wl3501_reset_board(this);
-
-	if (rc)
-		goto fail;
-	this->card_name[0] = '\0';
-	wl3501_get_from_wla(this, 0x1a00,
-			    this->card_name, sizeof(this->card_name));
-	this->card_name[sizeof(this->card_name) - 1] = '\0';
-	this->firmware_date[0] = '\0';
-	wl3501_get_from_wla(this, 0x1a40,
-			    this->firmware_date, sizeof(this->firmware_date));
-	this->firmware_date[sizeof(this->firmware_date) - 1] = '\0';
-	/* Switch to SRAM Page 0 */
-	wl3501_switch_page(this, WL3501_BSS_SPAGE0);
-	/* Read parameter from card */
-	wl3501_get_from_wla(this, 0x482, &this->esbq_req_start, 2);
-	wl3501_get_from_wla(this, 0x486, &this->esbq_req_end, 2);
-	wl3501_get_from_wla(this, 0x488, &this->esbq_confirm_start, 2);
-	wl3501_get_from_wla(this, 0x48c, &this->esbq_confirm_end, 2);
-	wl3501_get_from_wla(this, 0x48e, &this->tx_buffer_head, 2);
-	wl3501_get_from_wla(this, 0x492, &this->tx_buffer_size, 2);
-	this->esbq_req_tail	= this->esbq_req_head = this->esbq_req_start;
-	this->esbq_req_end     += this->esbq_req_start;
-	this->esbq_confirm	= this->esbq_confirm_start;
-	this->esbq_confirm_end += this->esbq_confirm_start;
-	/* Initial Tx Buffer */
-	this->tx_buffer_cnt = 1;
-	ptr = this->tx_buffer_head;
-	next = ptr + WL3501_BLKSZ;
-	while ((next - this->tx_buffer_head) < this->tx_buffer_size) {
-		this->tx_buffer_cnt++;
-		wl3501_set_to_wla(this, ptr, &next, sizeof(next));
-		ptr = next;
-		next = ptr + WL3501_BLKSZ;
-	}
-	rc = 0;
-	next = 0;
-	wl3501_set_to_wla(this, ptr, &next, sizeof(next));
-	this->tx_buffer_tail = ptr;
-out:
-	return rc;
-fail:
-	printk(KERN_WARNING "%s: failed!\n", __func__);
-	goto out;
-}
-
-static int wl3501_close(struct net_device *dev)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	unsigned long flags;
-	struct pcmcia_device *link;
-	link = this->p_dev;
-
-	spin_lock_irqsave(&this->lock, flags);
-	link->open--;
-
-	/* Stop wl3501_hard_start_xmit() from now on */
-	netif_stop_queue(dev);
-	wl3501_ack_interrupt(this);
-
-	/* Mask interrupts from the SUTRO */
-	wl3501_block_interrupt(this);
-
-	printk(KERN_INFO "%s: WL3501 closed\n", dev->name);
-	spin_unlock_irqrestore(&this->lock, flags);
-	return 0;
-}
-
-/**
- * wl3501_reset - Reset the SUTRO.
- * @dev: network device
- *
- * It is almost the same as wl3501_open(). In fact, we may just wl3501_close()
- * and wl3501_open() again, but I wouldn't like to free_irq() when the driver
- * is running. It seems to be dangerous.
- */
-static int wl3501_reset(struct net_device *dev)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = -ENODEV;
-	unsigned long flags;
-
-	spin_lock_irqsave(&this->lock, flags);
-	wl3501_block_interrupt(this);
-
-	if (wl3501_init_firmware(this)) {
-		printk(KERN_WARNING "%s: Can't initialize Firmware!\n",
-		       dev->name);
-		/* Free IRQ, and mark IRQ as unused */
-		free_irq(dev->irq, dev);
-		goto out;
-	}
-
-	/*
-	 * Queue has to be started only when the Card is Started
-	 */
-	netif_stop_queue(dev);
-	this->adhoc_times = 0;
-	wl3501_ack_interrupt(this);
-	wl3501_unblock_interrupt(this);
-	wl3501_mgmt_scan(this, 100);
-	pr_debug("%s: device reset", dev->name);
-	rc = 0;
-out:
-	spin_unlock_irqrestore(&this->lock, flags);
-	return rc;
-}
-
-static void wl3501_tx_timeout(struct net_device *dev, unsigned int txqueue)
-{
-	struct net_device_stats *stats = &dev->stats;
-	int rc;
-
-	stats->tx_errors++;
-	rc = wl3501_reset(dev);
-	if (rc)
-		printk(KERN_ERR "%s: Error %d resetting card on Tx timeout!\n",
-		       dev->name, rc);
-	else {
-		netif_trans_update(dev); /* prevent tx timeout */
-		netif_wake_queue(dev);
-	}
-}
-
-/*
- * Return : 0 - OK
- *	    1 - Could not transmit (dev_queue_xmit will queue it)
- *		and try to sent it later
- */
-static netdev_tx_t wl3501_hard_start_xmit(struct sk_buff *skb,
-						struct net_device *dev)
-{
-	int enabled, rc;
-	struct wl3501_card *this = netdev_priv(dev);
-	unsigned long flags;
-
-	spin_lock_irqsave(&this->lock, flags);
-	enabled = wl3501_block_interrupt(this);
-	rc = wl3501_send_pkt(this, skb->data, skb->len);
-	if (enabled)
-		wl3501_unblock_interrupt(this);
-	if (rc) {
-		++dev->stats.tx_dropped;
-		netif_stop_queue(dev);
-	} else {
-		++dev->stats.tx_packets;
-		dev->stats.tx_bytes += skb->len;
-		dev_kfree_skb_irq(skb);
-
-		if (this->tx_buffer_cnt < 2)
-			netif_stop_queue(dev);
-	}
-	spin_unlock_irqrestore(&this->lock, flags);
-	return NETDEV_TX_OK;
-}
-
-static int wl3501_open(struct net_device *dev)
-{
-	int rc = -ENODEV;
-	struct wl3501_card *this = netdev_priv(dev);
-	unsigned long flags;
-	struct pcmcia_device *link;
-	link = this->p_dev;
-
-	spin_lock_irqsave(&this->lock, flags);
-	if (!pcmcia_dev_present(link))
-		goto out;
-	netif_device_attach(dev);
-	link->open++;
-
-	/* Initial WL3501 firmware */
-	pr_debug("%s: Initialize WL3501 firmware...", dev->name);
-	if (wl3501_init_firmware(this))
-		goto fail;
-	/* Initial device variables */
-	this->adhoc_times = 0;
-	/* Acknowledge Interrupt, for cleaning last state */
-	wl3501_ack_interrupt(this);
-
-	/* Enable interrupt from card after all */
-	wl3501_unblock_interrupt(this);
-	wl3501_mgmt_scan(this, 100);
-	rc = 0;
-	pr_debug("%s: WL3501 opened", dev->name);
-	printk(KERN_INFO "%s: Card Name: %s\n"
-			 "%s: Firmware Date: %s\n",
-			 dev->name, this->card_name,
-			 dev->name, this->firmware_date);
-out:
-	spin_unlock_irqrestore(&this->lock, flags);
-	return rc;
-fail:
-	printk(KERN_WARNING "%s: Can't initialize firmware!\n", dev->name);
-	goto out;
-}
-
-static struct iw_statistics *wl3501_get_wireless_stats(struct net_device *dev)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	struct iw_statistics *wstats = &this->wstats;
-	u32 value; /* size checked: it is u32 */
-
-	memset(wstats, 0, sizeof(*wstats));
-	wstats->status = netif_running(dev);
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_ICV_ERROR_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.code += value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_UNDECRYPTABLE_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.code += value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_EXCLUDED_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.code += value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_RETRY_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.retries	= value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_FAILED_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.misc += value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_RTS_FAILURE_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.misc += value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_ACK_FAILURE_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.misc += value;
-	if (!wl3501_get_mib_value(this, WL3501_MIB_ATTR_FRAME_DUPLICATE_COUNT,
-				  &value, sizeof(value)))
-		wstats->discard.misc += value;
-	return wstats;
-}
-
-/**
- * wl3501_detach - deletes a driver "instance"
- * @link: FILL_IN
- *
- * This deletes a driver "instance". The device is de-registered with Card
- * Services. If it has been released, all local data structures are freed.
- * Otherwise, the structures will be freed when the device is released.
- */
-static void wl3501_detach(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-
-	/* If the device is currently configured and active, we won't actually
-	 * delete it yet.  Instead, it is marked so that when the release()
-	 * function is called, that will trigger a proper detach(). */
-
-	while (link->open > 0)
-		wl3501_close(dev);
-
-	netif_device_detach(dev);
-	wl3501_release(link);
-
-	unregister_netdev(dev);
-	free_netdev(dev);
-}
-
-static int wl3501_get_name(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	strscpy(wrqu->name, "IEEE 802.11-DS", sizeof(wrqu->name));
-	return 0;
-}
-
-static int wl3501_set_freq(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	int channel = wrqu->freq.m;
-	int rc = -EINVAL;
-
-	if (iw_valid_channel(this->reg_domain, channel)) {
-		this->chan = channel;
-		rc = wl3501_reset(dev);
-	}
-	return rc;
-}
-
-static int wl3501_get_freq(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	wrqu->freq.m = 100000 *
-		ieee80211_channel_to_frequency(this->chan, NL80211_BAND_2GHZ);
-	wrqu->freq.e = 1;
-	return 0;
-}
-
-static int wl3501_set_mode(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	int rc = -EINVAL;
-
-	if (wrqu->mode == IW_MODE_INFRA ||
-	    wrqu->mode == IW_MODE_ADHOC ||
-	    wrqu->mode == IW_MODE_AUTO) {
-		struct wl3501_card *this = netdev_priv(dev);
-
-		this->net_type = wrqu->mode;
-		rc = wl3501_reset(dev);
-	}
-	return rc;
-}
-
-static int wl3501_get_mode(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	wrqu->mode = this->net_type;
-	return 0;
-}
-
-static int wl3501_get_sens(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	wrqu->sens.value = this->rssi;
-	wrqu->sens.disabled = !wrqu->sens.value;
-	wrqu->sens.fixed = 1;
-	return 0;
-}
-
-static int wl3501_get_range(struct net_device *dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *wrqu, char *extra)
-{
-	struct iw_range *range = (struct iw_range *)extra;
-
-	/* Set the length (very important for backward compatibility) */
-	wrqu->data.length = sizeof(*range);
-
-	/* Set all the info we don't care or don't know about to zero */
-	memset(range, 0, sizeof(*range));
-
-	/* Set the Wireless Extension versions */
-	range->we_version_compiled	= WIRELESS_EXT;
-	range->we_version_source	= 1;
-	range->throughput		= 2 * 1000 * 1000;     /* ~2 Mb/s */
-	/* FIXME: study the code to fill in more fields... */
-	return 0;
-}
-
-static int wl3501_set_wap(struct net_device *dev, struct iw_request_info *info,
-			  union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = -EINVAL;
-
-	/* FIXME: we support other ARPHRDs...*/
-	if (wrqu->ap_addr.sa_family != ARPHRD_ETHER)
-		goto out;
-	if (is_broadcast_ether_addr(wrqu->ap_addr.sa_data)) {
-		/* FIXME: rescan? */
-	} else
-		memcpy(this->bssid, wrqu->ap_addr.sa_data, ETH_ALEN);
-		/* FIXME: rescan? deassoc & scan? */
-	rc = 0;
-out:
-	return rc;
-}
-
-static int wl3501_get_wap(struct net_device *dev, struct iw_request_info *info,
-			  union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	wrqu->ap_addr.sa_family = ARPHRD_ETHER;
-	memcpy(wrqu->ap_addr.sa_data, this->bssid, ETH_ALEN);
-	return 0;
-}
-
-static int wl3501_set_scan(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	/*
-	 * FIXME: trigger scanning with a reset, yes, I'm lazy
-	 */
-	return wl3501_reset(dev);
-}
-
-static int wl3501_get_scan(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	int i;
-	char *current_ev = extra;
-	struct iw_event iwe;
-
-	for (i = 0; i < this->bss_cnt; ++i) {
-		iwe.cmd			= SIOCGIWAP;
-		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
-		memcpy(iwe.u.ap_addr.sa_data, this->bss_set[i].req.bssid, ETH_ALEN);
-		current_ev = iwe_stream_add_event(info, current_ev,
-						  extra + IW_SCAN_MAX_DATA,
-						  &iwe, IW_EV_ADDR_LEN);
-		iwe.cmd		  = SIOCGIWESSID;
-		iwe.u.data.flags  = 1;
-		iwe.u.data.length = this->bss_set[i].req.ssid.el.len;
-		current_ev = iwe_stream_add_point(info, current_ev,
-						  extra + IW_SCAN_MAX_DATA,
-						  &iwe,
-						  this->bss_set[i].req.ssid.essid);
-		iwe.cmd	   = SIOCGIWMODE;
-		iwe.u.mode = this->bss_set[i].req.bss_type;
-		current_ev = iwe_stream_add_event(info, current_ev,
-						  extra + IW_SCAN_MAX_DATA,
-						  &iwe, IW_EV_UINT_LEN);
-		iwe.cmd = SIOCGIWFREQ;
-		iwe.u.freq.m = this->bss_set[i].req.ds_pset.chan;
-		iwe.u.freq.e = 0;
-		current_ev = iwe_stream_add_event(info, current_ev,
-						  extra + IW_SCAN_MAX_DATA,
-						  &iwe, IW_EV_FREQ_LEN);
-		iwe.cmd = SIOCGIWENCODE;
-		if (this->bss_set[i].req.cap_info & WL3501_MGMT_CAPABILITY_PRIVACY)
-			iwe.u.data.flags = IW_ENCODE_ENABLED | IW_ENCODE_NOKEY;
-		else
-			iwe.u.data.flags = IW_ENCODE_DISABLED;
-		iwe.u.data.length = 0;
-		current_ev = iwe_stream_add_point(info, current_ev,
-						  extra + IW_SCAN_MAX_DATA,
-						  &iwe, NULL);
-	}
-	/* Length of data */
-	wrqu->data.length = (current_ev - extra);
-	wrqu->data.flags = 0; /* FIXME: set properly these flags */
-	return 0;
-}
-
-static int wl3501_set_essid(struct net_device *dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	if (wrqu->data.flags) {
-		iw_set_mgmt_info_element(IW_MGMT_INFO_ELEMENT_SSID,
-					 &this->essid.el,
-					 extra, wrqu->data.length);
-	} else { /* We accept any ESSID */
-		iw_set_mgmt_info_element(IW_MGMT_INFO_ELEMENT_SSID,
-					 &this->essid.el, "ANY", 3);
-	}
-	return wl3501_reset(dev);
-}
-
-static int wl3501_get_essid(struct net_device *dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-	unsigned long flags;
-
-	spin_lock_irqsave(&this->lock, flags);
-	wrqu->essid.flags  = 1;
-	wrqu->essid.length = this->essid.el.len;
-	memcpy(extra, this->essid.essid, this->essid.el.len);
-	spin_unlock_irqrestore(&this->lock, flags);
-	return 0;
-}
-
-static int wl3501_set_nick(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	if (wrqu->data.length > sizeof(this->nick))
-		return -E2BIG;
-	strscpy(this->nick, extra, wrqu->data.length);
-	return 0;
-}
-
-static int wl3501_get_nick(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	struct wl3501_card *this = netdev_priv(dev);
-
-	strscpy(extra, this->nick, 32);
-	wrqu->data.length = strlen(extra);
-	return 0;
-}
-
-static int wl3501_get_rate(struct net_device *dev, struct iw_request_info *info,
-			   union iwreq_data *wrqu, char *extra)
-{
-	/*
-	 * FIXME: have to see from where to get this info, perhaps this card
-	 * works at 1 Mbit/s too... for now leave at 2 Mbit/s that is the most
-	 * common with the Planet Access Points. -acme
-	 */
-	wrqu->bitrate.value = 2000000;
-	wrqu->bitrate.fixed = 1;
-	return 0;
-}
-
-static int wl3501_get_rts_threshold(struct net_device *dev,
-				    struct iw_request_info *info,
-				    union iwreq_data *wrqu, char *extra)
-{
-	u16 threshold; /* size checked: it is u16 */
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_RTS_THRESHOLD,
-				      &threshold, sizeof(threshold));
-	if (!rc) {
-		wrqu->rts.value = threshold;
-		wrqu->rts.disabled = threshold >= 2347;
-		wrqu->rts.fixed = 1;
-	}
-	return rc;
-}
-
-static int wl3501_get_frag_threshold(struct net_device *dev,
-				     struct iw_request_info *info,
-				     union iwreq_data *wrqu, char *extra)
-{
-	u16 threshold; /* size checked: it is u16 */
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_FRAG_THRESHOLD,
-				      &threshold, sizeof(threshold));
-	if (!rc) {
-		wrqu->frag.value = threshold;
-		wrqu->frag.disabled = threshold >= 2346;
-		wrqu->frag.fixed = 1;
-	}
-	return rc;
-}
-
-static int wl3501_get_txpow(struct net_device *dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *wrqu, char *extra)
-{
-	u16 txpow;
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = wl3501_get_mib_value(this,
-				      WL3501_MIB_ATTR_CURRENT_TX_PWR_LEVEL,
-				      &txpow, sizeof(txpow));
-	if (!rc) {
-		wrqu->txpower.value = txpow;
-		wrqu->txpower.disabled = 0;
-		/*
-		 * From the MIB values I think this can be configurable,
-		 * as it lists several tx power levels -acme
-		 */
-		wrqu->txpower.fixed = 0;
-		wrqu->txpower.flags = IW_TXPOW_MWATT;
-	}
-	return rc;
-}
-
-static int wl3501_get_retry(struct net_device *dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *wrqu, char *extra)
-{
-	u8 retry; /* size checked: it is u8 */
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = wl3501_get_mib_value(this,
-				      WL3501_MIB_ATTR_LONG_RETRY_LIMIT,
-				      &retry, sizeof(retry));
-	if (rc)
-		goto out;
-	if (wrqu->retry.flags & IW_RETRY_LONG) {
-		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
-		goto set_value;
-	}
-	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_SHORT_RETRY_LIMIT,
-				  &retry, sizeof(retry));
-	if (rc)
-		goto out;
-	wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_SHORT;
-set_value:
-	wrqu->retry.value = retry;
-	wrqu->retry.disabled = 0;
-out:
-	return rc;
-}
-
-static int wl3501_get_encode(struct net_device *dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *wrqu, char *extra)
-{
-	u8 implemented, restricted, keys[100], len_keys, tocopy;
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = wl3501_get_mib_value(this,
-				      WL3501_MIB_ATTR_PRIV_OPT_IMPLEMENTED,
-				      &implemented, sizeof(implemented));
-	if (rc)
-		goto out;
-	if (!implemented) {
-		wrqu->encoding.flags = IW_ENCODE_DISABLED;
-		goto out;
-	}
-	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_EXCLUDE_UNENCRYPTED,
-				  &restricted, sizeof(restricted));
-	if (rc)
-		goto out;
-	wrqu->encoding.flags = restricted ? IW_ENCODE_RESTRICTED :
-					    IW_ENCODE_OPEN;
-	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_KEY_MAPPINGS_LEN,
-				  &len_keys, sizeof(len_keys));
-	if (rc)
-		goto out;
-	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_WEP_KEY_MAPPINGS,
-				  keys, len_keys);
-	if (rc)
-		goto out;
-	tocopy = min_t(u16, len_keys, wrqu->encoding.length);
-	tocopy = min_t(u8, tocopy, 100);
-	wrqu->encoding.length = tocopy;
-	memcpy(extra, keys, tocopy);
-out:
-	return rc;
-}
-
-static int wl3501_get_power(struct net_device *dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *wrqu, char *extra)
-{
-	u8 pwr_state;
-	struct wl3501_card *this = netdev_priv(dev);
-	int rc = wl3501_get_mib_value(this,
-				      WL3501_MIB_ATTR_CURRENT_PWR_STATE,
-				      &pwr_state, sizeof(pwr_state));
-	if (rc)
-		goto out;
-	wrqu->power.disabled = !pwr_state;
-	wrqu->power.flags = IW_POWER_ON;
-out:
-	return rc;
-}
-
-static const iw_handler	wl3501_handler[] = {
-	IW_HANDLER(SIOCGIWNAME, wl3501_get_name),
-	IW_HANDLER(SIOCSIWFREQ, wl3501_set_freq),
-	IW_HANDLER(SIOCGIWFREQ, wl3501_get_freq),
-	IW_HANDLER(SIOCSIWMODE, wl3501_set_mode),
-	IW_HANDLER(SIOCGIWMODE, wl3501_get_mode),
-	IW_HANDLER(SIOCGIWSENS, wl3501_get_sens),
-	IW_HANDLER(SIOCGIWRANGE, wl3501_get_range),
-	IW_HANDLER(SIOCSIWSPY, iw_handler_set_spy),
-	IW_HANDLER(SIOCGIWSPY, iw_handler_get_spy),
-	IW_HANDLER(SIOCSIWTHRSPY, iw_handler_set_thrspy),
-	IW_HANDLER(SIOCGIWTHRSPY, iw_handler_get_thrspy),
-	IW_HANDLER(SIOCSIWAP, wl3501_set_wap),
-	IW_HANDLER(SIOCGIWAP, wl3501_get_wap),
-	IW_HANDLER(SIOCSIWSCAN, wl3501_set_scan),
-	IW_HANDLER(SIOCGIWSCAN, wl3501_get_scan),
-	IW_HANDLER(SIOCSIWESSID, wl3501_set_essid),
-	IW_HANDLER(SIOCGIWESSID, wl3501_get_essid),
-	IW_HANDLER(SIOCSIWNICKN, wl3501_set_nick),
-	IW_HANDLER(SIOCGIWNICKN, wl3501_get_nick),
-	IW_HANDLER(SIOCGIWRATE, wl3501_get_rate),
-	IW_HANDLER(SIOCGIWRTS, wl3501_get_rts_threshold),
-	IW_HANDLER(SIOCGIWFRAG, wl3501_get_frag_threshold),
-	IW_HANDLER(SIOCGIWTXPOW, wl3501_get_txpow),
-	IW_HANDLER(SIOCGIWRETRY, wl3501_get_retry),
-	IW_HANDLER(SIOCGIWENCODE, wl3501_get_encode),
-	IW_HANDLER(SIOCGIWPOWER, wl3501_get_power),
-};
-
-static const struct iw_handler_def wl3501_handler_def = {
-	.num_standard	= ARRAY_SIZE(wl3501_handler),
-	.standard	= (iw_handler *)wl3501_handler,
-	.get_wireless_stats = wl3501_get_wireless_stats,
-};
-
-static const struct net_device_ops wl3501_netdev_ops = {
-	.ndo_open		= wl3501_open,
-	.ndo_stop		= wl3501_close,
-	.ndo_start_xmit		= wl3501_hard_start_xmit,
-	.ndo_tx_timeout		= wl3501_tx_timeout,
-	.ndo_set_mac_address 	= eth_mac_addr,
-	.ndo_validate_addr	= eth_validate_addr,
-};
-
-static int wl3501_probe(struct pcmcia_device *p_dev)
-{
-	struct net_device *dev;
-	struct wl3501_card *this;
-
-	/* The io structure describes IO port mapping */
-	p_dev->resource[0]->end	= 16;
-	p_dev->resource[0]->flags	= IO_DATA_PATH_WIDTH_8;
-
-	/* General socket configuration */
-	p_dev->config_flags	= CONF_ENABLE_IRQ;
-	p_dev->config_index	= 1;
-
-	dev = alloc_etherdev(sizeof(struct wl3501_card));
-	if (!dev)
-		goto out_link;
-
-
-	dev->netdev_ops		= &wl3501_netdev_ops;
-	dev->watchdog_timeo	= 5 * HZ;
-
-	this = netdev_priv(dev);
-	this->wireless_data.spy_data = &this->spy_data;
-	this->p_dev = p_dev;
-	dev->wireless_data	= &this->wireless_data;
-	dev->wireless_handlers	= &wl3501_handler_def;
-	netif_stop_queue(dev);
-	p_dev->priv = dev;
-
-	return wl3501_config(p_dev);
-out_link:
-	return -ENOMEM;
-}
-
-static int wl3501_config(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-	int i = 0, j, ret;
-	struct wl3501_card *this;
-
-	/* Try allocating IO ports.  This tries a few fixed addresses.  If you
-	 * want, you can also read the card's config table to pick addresses --
-	 * see the serial driver for an example. */
-	link->io_lines = 5;
-
-	for (j = 0x280; j < 0x400; j += 0x20) {
-		/* The '^0x300' is so that we probe 0x300-0x3ff first, then
-		 * 0x200-0x2ff, and so on, because this seems safer */
-		link->resource[0]->start = j;
-		link->resource[1]->start = link->resource[0]->start + 0x10;
-		i = pcmcia_request_io(link);
-		if (i == 0)
-			break;
-	}
-	if (i != 0)
-		goto failed;
-
-	/* Now allocate an interrupt line. Note that this does not actually
-	 * assign a handler to the interrupt. */
-
-	ret = pcmcia_request_irq(link, wl3501_interrupt);
-	if (ret)
-		goto failed;
-
-	ret = pcmcia_enable_device(link);
-	if (ret)
-		goto failed;
-
-	dev->irq = link->irq;
-	dev->base_addr = link->resource[0]->start;
-	SET_NETDEV_DEV(dev, &link->dev);
-	if (register_netdev(dev)) {
-		printk(KERN_NOTICE "wl3501_cs: register_netdev() failed\n");
-		goto failed;
-	}
-
-	this = netdev_priv(dev);
-
-	this->base_addr = dev->base_addr;
-
-	if (!wl3501_get_flash_mac_addr(this)) {
-		printk(KERN_WARNING "%s: Can't read MAC addr in flash ROM?\n",
-		       dev->name);
-		unregister_netdev(dev);
-		goto failed;
-	}
-
-	eth_hw_addr_set(dev, this->mac_addr);
-
-	/* print probe information */
-	printk(KERN_INFO "%s: wl3501 @ 0x%3.3x, IRQ %d, "
-	       "MAC addr in flash ROM:%pM\n",
-	       dev->name, this->base_addr, (int)dev->irq,
-	       dev->dev_addr);
-	/*
-	 * Initialize card parameters - added by jss
-	 */
-	this->net_type		= IW_MODE_INFRA;
-	this->bss_cnt		= 0;
-	this->join_sta_bss	= 0;
-	this->adhoc_times	= 0;
-	iw_set_mgmt_info_element(IW_MGMT_INFO_ELEMENT_SSID, &this->essid.el,
-				 "ANY", 3);
-	this->card_name[0]	= '\0';
-	this->firmware_date[0]	= '\0';
-	this->rssi		= 255;
-	this->chan		= iw_default_channel(this->reg_domain);
-	strscpy(this->nick, "Planet WL3501", sizeof(this->nick));
-	spin_lock_init(&this->lock);
-	init_waitqueue_head(&this->wait);
-	netif_start_queue(dev);
-	return 0;
-
-failed:
-	wl3501_release(link);
-	return -ENODEV;
-}
-
-static void wl3501_release(struct pcmcia_device *link)
-{
-	pcmcia_disable_device(link);
-}
-
-static int wl3501_suspend(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-
-	wl3501_pwr_mgmt(netdev_priv(dev), WL3501_SUSPEND);
-	if (link->open)
-		netif_device_detach(dev);
-
-	return 0;
-}
-
-static int wl3501_resume(struct pcmcia_device *link)
-{
-	struct net_device *dev = link->priv;
-
-	wl3501_pwr_mgmt(netdev_priv(dev), WL3501_RESUME);
-	if (link->open) {
-		wl3501_reset(dev);
-		netif_device_attach(dev);
-	}
-
-	return 0;
-}
-
-
-static const struct pcmcia_device_id wl3501_ids[] = {
-	PCMCIA_DEVICE_MANF_CARD(0xd601, 0x0001),
-	PCMCIA_DEVICE_NULL
-};
-MODULE_DEVICE_TABLE(pcmcia, wl3501_ids);
-
-static struct pcmcia_driver wl3501_driver = {
-	.owner		= THIS_MODULE,
-	.name		= "wl3501_cs",
-	.probe		= wl3501_probe,
-	.remove		= wl3501_detach,
-	.id_table	= wl3501_ids,
-	.suspend	= wl3501_suspend,
-	.resume		= wl3501_resume,
-};
-module_pcmcia_driver(wl3501_driver);
-
-MODULE_AUTHOR("Fox Chen <mhchen@golf.ccl.itri.org.tw>, "
-	      "Arnaldo Carvalho de Melo <acme@conectiva.com.br>,"
-	      "Gustavo Niemeyer <niemeyer@conectiva.com>");
-MODULE_DESCRIPTION("Planet wl3501 wireless driver");
-MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-70-g09d2


From 659fda7f35f0d11126edfb6c6cd9ace99a2525c0 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 3 Mar 2023 23:23:31 +0100
Subject: wifi: brcmfmac: Use ISO3166 country code and rev 0 as fallback on
 4356

Many devices ship with a nvram ccode value of X2/XT/XU/XV/ALL which are
all special world-wide compatibility ccode-s. Most of these world-wide
ccode-s allow passive scan mode only for 2.4GHz channels 12-14,
only enabling them when an AP is seen on them.

Since linux-firmware has moved to the new cyfmac4356-pci.bin +
cyfmac4356-pci.clm_blob firmware files this no longer works and
4356 devices using e.g. an X2 ccode fail to connect to an AP on
channel 13.

Add the 4356 chip-id to the list of chips for which to use the ISO3166
country code + rev 0 as fallback in brcmf_translate_country_code() to
fix this.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230303222331.285663-1-hdegoede@redhat.com
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index e0a70a671550..548799fefb4b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -7881,6 +7881,7 @@ static bool brmcf_use_iso3166_ccode_fallback(struct brcmf_pub *drvr)
 	switch (drvr->bus_if->chip) {
 	case BRCM_CC_43430_CHIP_ID:
 	case BRCM_CC_4345_CHIP_ID:
+	case BRCM_CC_4356_CHIP_ID:
 	case BRCM_CC_43602_CHIP_ID:
 		return true;
 	default:
-- 
cgit v1.2.3-70-g09d2


From df259fc12b366097fc6b7c3ed1a2f9f77b4ba355 Mon Sep 17 00:00:00 2001
From: Martin Kaiser <martin@kaiser.cx>
Date: Sun, 5 Mar 2023 18:59:32 +0100
Subject: wifi: rtl8xxxu: mark Edimax EW-7811Un V2 as tested

The Edimax V2 (vid 0x7392, pid 0xb811) works well with the rtl8xxxu driver
since rtl8188eu support has been added. Remove the untested flag for this
device.

Signed-off-by: Martin Kaiser <martin@kaiser.cx>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230305175932.719103-1-martin@kaiser.cx
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 54ca6f2ced3f..47dcca842d7c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -7012,7 +7012,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
 		}
 		break;
 	case 0x7392:
-		if (id->idProduct == 0x7811 || id->idProduct == 0xa611)
+		if (id->idProduct == 0x7811 || id->idProduct == 0xa611 || id->idProduct == 0xb811)
 			untested = 0;
 		break;
 	case 0x050d:
-- 
cgit v1.2.3-70-g09d2


From 8a66293e73a520a42a7653d2ca32074ba323ff56 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Tue, 7 Mar 2023 22:18:48 +0800
Subject: wifi: rtw89: release RX standby timer of beamformee CSI to save power

Originally, we keep RX standby timer to handle beamformee CSI, but this
spends power and causes system not entering power save mode. To improve
power consumption, release the timer if throughput becomes low.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230307141848.26403-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h |  1 +
 drivers/net/wireless/realtek/rtw89/mac.c  | 35 +++++++++++++++++++++++++++++--
 drivers/net/wireless/realtek/rtw89/reg.h  |  2 ++
 3 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index b1a886898c5a..9073f526b5a3 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3168,6 +3168,7 @@ enum rtw89_flags {
 	RTW89_FLAG_RUNNING,
 	RTW89_FLAG_BFEE_MON,
 	RTW89_FLAG_BFEE_EN,
+	RTW89_FLAG_BFEE_TIMER_KEEP,
 	RTW89_FLAG_NAPI_RUNNING,
 	RTW89_FLAG_LEISURE_PS,
 	RTW89_FLAG_LOW_POWER_MODE,
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 3d1e4ffef1b1..7866fe925d24 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4931,6 +4931,24 @@ u16 rtw89_mac_get_plt_cnt(struct rtw89_dev *rtwdev, u8 band)
 	return cnt;
 }
 
+static void rtw89_mac_bfee_standby_timer(struct rtw89_dev *rtwdev, u8 mac_idx,
+					 bool keep)
+{
+	u32 reg;
+
+	rtw89_debug(rtwdev, RTW89_DBG_BF, "set bfee standby_timer to %d\n", keep);
+	reg = rtw89_mac_reg_by_idx(R_AX_BFMEE_RESP_OPTION, mac_idx);
+	if (keep) {
+		set_bit(RTW89_FLAG_BFEE_TIMER_KEEP, rtwdev->flags);
+		rtw89_write32_mask(rtwdev, reg, B_AX_BFMEE_BFRP_RX_STANDBY_TIMER_MASK,
+				   BFRP_RX_STANDBY_TIMER_KEEP);
+	} else {
+		clear_bit(RTW89_FLAG_BFEE_TIMER_KEEP, rtwdev->flags);
+		rtw89_write32_mask(rtwdev, reg, B_AX_BFMEE_BFRP_RX_STANDBY_TIMER_MASK,
+				   BFRP_RX_STANDBY_TIMER_RELEASE);
+	}
+}
+
 static void rtw89_mac_bfee_ctrl(struct rtw89_dev *rtwdev, u8 mac_idx, bool en)
 {
 	u32 reg;
@@ -4967,9 +4985,9 @@ static int rtw89_mac_init_bfee(struct rtw89_dev *rtwdev, u8 mac_idx)
 	rtw89_write32(rtwdev, reg, CSI_RRSC_BMAP);
 
 	reg = rtw89_mac_reg_by_idx(R_AX_BFMEE_RESP_OPTION, mac_idx);
-	val32 = FIELD_PREP(B_AX_BFMEE_BFRP_RX_STANDBY_TIMER_MASK, BFRP_RX_STANDBY_TIMER);
-	val32 |= FIELD_PREP(B_AX_BFMEE_NDP_RX_STANDBY_TIMER_MASK, NDP_RX_STANDBY_TIMER);
+	val32 = FIELD_PREP(B_AX_BFMEE_NDP_RX_STANDBY_TIMER_MASK, NDP_RX_STANDBY_TIMER);
 	rtw89_write32(rtwdev, reg, val32);
+	rtw89_mac_bfee_standby_timer(rtwdev, mac_idx, true);
 	rtw89_mac_bfee_ctrl(rtwdev, mac_idx, true);
 
 	reg = rtw89_mac_reg_by_idx(R_AX_TRXPTCL_RESP_CSI_CTRL_0, mac_idx);
@@ -5181,6 +5199,19 @@ void _rtw89_mac_bf_monitor_track(struct rtw89_dev *rtwdev)
 	struct rtw89_vif *rtwvif;
 	bool en = stats->tx_tfc_lv <= stats->rx_tfc_lv;
 	bool old = test_bit(RTW89_FLAG_BFEE_EN, rtwdev->flags);
+	bool keep_timer = true;
+	bool old_keep_timer;
+
+	old_keep_timer = test_bit(RTW89_FLAG_BFEE_TIMER_KEEP, rtwdev->flags);
+
+	if (stats->tx_tfc_lv <= RTW89_TFC_LOW && stats->rx_tfc_lv <= RTW89_TFC_LOW)
+		keep_timer = false;
+
+	if (keep_timer != old_keep_timer) {
+		rtw89_for_each_rtwvif(rtwdev, rtwvif)
+			rtw89_mac_bfee_standby_timer(rtwdev, rtwvif->mac_idx,
+						     keep_timer);
+	}
 
 	if (en == old)
 		return;
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 600257909df2..e5c0ab43ab7a 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -3056,6 +3056,8 @@
 #define R_AX_BFMEE_RESP_OPTION_C1 0xED80
 #define B_AX_BFMEE_NDP_RX_STANDBY_TIMER_MASK GENMASK(31, 24)
 #define B_AX_BFMEE_BFRP_RX_STANDBY_TIMER_MASK GENMASK(23, 20)
+#define BFRP_RX_STANDBY_TIMER_KEEP 0x0
+#define BFRP_RX_STANDBY_TIMER_RELEASE 0x1
 #define B_AX_MU_BFRPTSEG_SEL_MASK GENMASK(18, 17)
 #define B_AX_BFMEE_NDP_RXSTDBY_SEL BIT(16)
 #define BFRP_RX_STANDBY_TIMER		0x0
-- 
cgit v1.2.3-70-g09d2


From 0606b344021a37140ef467ddf30d4b8182e7bbe1 Mon Sep 17 00:00:00 2001
From: Martin Kaiser <martin@kaiser.cx>
Date: Tue, 7 Mar 2023 20:57:17 +0100
Subject: wifi: rtl8xxxu: use module_usb_driver

We can use the module_usb_driver macro instead of open-coding the driver's
init and exit functions. This is simpler and saves some lines of code.
Other realtek wireless drivers use module_usb_driver as well.

Signed-off-by: Martin Kaiser <martin@kaiser.cx>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Tested-by: Philipp Hortmann <philipp.g.hortmann@gmail.com> # Edimax N150
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230307195718.168021-1-martin@kaiser.cx
---
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c    | 20 +-------------------
 1 file changed, 1 insertion(+), 19 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 47dcca842d7c..280dc3e26ca7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -7451,24 +7451,6 @@ static struct usb_driver rtl8xxxu_driver = {
 	.disable_hub_initiated_lpm = 1,
 };
 
-static int __init rtl8xxxu_module_init(void)
-{
-	int res;
-
-	res = usb_register(&rtl8xxxu_driver);
-	if (res < 0)
-		pr_err(DRIVER_NAME ": usb_register() failed (%i)\n", res);
-
-	return res;
-}
-
-static void __exit rtl8xxxu_module_exit(void)
-{
-	usb_deregister(&rtl8xxxu_driver);
-}
-
-
 MODULE_DEVICE_TABLE(usb, dev_table);
 
-module_init(rtl8xxxu_module_init);
-module_exit(rtl8xxxu_module_exit);
+module_usb_driver(rtl8xxxu_driver);
-- 
cgit v1.2.3-70-g09d2


From a23c82e006db2f073fb65da8762f63735a510409 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 7 Mar 2023 15:01:48 -0800
Subject: wifi: ipw2x00: convert ipw_fw_error->elem to flexible array[]

The ipw_fw_error structure contains a payload[] flexible array as well as
two pointers to this array area, ->elem, and ->log. The total size of the
allocated structure is computed without use of the <linux/overflow.h>
macros.

There's no reason to keep both a payload[] and an extra pointer to both the
elem and log members. Convert the elem pointer member into the flexible
array member, removing payload.

Fix the allocation of the ipw_fw_error structure to use size_add(),
struct_size(), and array_size() to compute the allocation. This ensures
that any overflow saturates at SIZE_MAX rather than overflowing and
potentially allowing an undersized allocation.

Before the structure change, the layout of ipw_fw_error was:

struct ipw_fw_error {
        long unsigned int          jiffies;              /*     0     8 */
        u32                        status;               /*     8     4 */
        u32                        config;               /*    12     4 */
        u32                        elem_len;             /*    16     4 */
        u32                        log_len;              /*    20     4 */
        struct ipw_error_elem *    elem;                 /*    24     8 */
        struct ipw_event *         log;                  /*    32     8 */
        u8                         payload[];            /*    40     0 */

        /* size: 40, cachelines: 1, members: 8 */
        /* last cacheline: 40 bytes */
};

After this change, the layout is now:

struct ipw_fw_error {
        long unsigned int          jiffies;              /*     0     8 */
        u32                        status;               /*     8     4 */
        u32                        config;               /*    12     4 */
        u32                        elem_len;             /*    16     4 */
        u32                        log_len;              /*    20     4 */
        struct ipw_event *         log;                  /*    24     8 */
        struct ipw_error_elem      elem[];               /*    32     0 */

        /* size: 32, cachelines: 1, members: 7 */
        /* last cacheline: 32 bytes */
};

This saves a total of 8 bytes for every ipw_fw_error allocation, and
removes the risk of a potential overflow on the allocation.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Stanislav Yakovlev <stas.yakovlev@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230307230148.3735684-1-jacob.e.keller@intel.com
---
 drivers/net/wireless/intel/ipw2x00/ipw2200.c | 7 +++----
 drivers/net/wireless/intel/ipw2x00/ipw2200.h | 3 +--
 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2200.c b/drivers/net/wireless/intel/ipw2x00/ipw2200.c
index d382f2017325..b91b1a2d0be7 100644
--- a/drivers/net/wireless/intel/ipw2x00/ipw2200.c
+++ b/drivers/net/wireless/intel/ipw2x00/ipw2200.c
@@ -1234,9 +1234,9 @@ static struct ipw_fw_error *ipw_alloc_error_log(struct ipw_priv *priv)
 	u32 base = ipw_read32(priv, IPW_ERROR_LOG);
 	u32 elem_len = ipw_read_reg32(priv, base);
 
-	error = kmalloc(sizeof(*error) +
-			sizeof(*error->elem) * elem_len +
-			sizeof(*error->log) * log_len, GFP_ATOMIC);
+	error = kmalloc(size_add(struct_size(error, elem, elem_len),
+				 array_size(sizeof(*error->log), log_len)),
+			GFP_ATOMIC);
 	if (!error) {
 		IPW_ERROR("Memory allocation for firmware error log "
 			  "failed.\n");
@@ -1247,7 +1247,6 @@ static struct ipw_fw_error *ipw_alloc_error_log(struct ipw_priv *priv)
 	error->config = priv->config;
 	error->elem_len = elem_len;
 	error->log_len = log_len;
-	error->elem = (struct ipw_error_elem *)error->payload;
 	error->log = (struct ipw_event *)(error->elem + elem_len);
 
 	ipw_capture_event_log(priv, log_len, error->log);
diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2200.h b/drivers/net/wireless/intel/ipw2x00/ipw2200.h
index 09ddd21608d4..8ebf09121e17 100644
--- a/drivers/net/wireless/intel/ipw2x00/ipw2200.h
+++ b/drivers/net/wireless/intel/ipw2x00/ipw2200.h
@@ -1106,9 +1106,8 @@ struct ipw_fw_error {	 /* XXX */
 	u32 config;
 	u32 elem_len;
 	u32 log_len;
-	struct ipw_error_elem *elem;
 	struct ipw_event *log;
-	u8 payload[];
+	struct ipw_error_elem elem[];
 } __packed;
 
 #ifdef CONFIG_IPW2200_PROMISCUOUS
-- 
cgit v1.2.3-70-g09d2


From 84e9e2102bdc4f60ce131dcb74046219dc4425ae Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 7 Mar 2023 15:02:12 -0800
Subject: wifi: qtnfmac: use struct_size and size_sub for payload length

Replace the calculations for the payload length in
qtnf_cmd_band_fill_iftype with struct_size() and size_sub(). While
the payload length does not get directly passed to an allocation function,
the performed calculation is still calculating the size of a flexible array
structure (minus the size of a header structure).

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Igor Mitsyanko <imitsyanko@quantenna.com>
Cc: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230307230212.3735818-1-jacob.e.keller@intel.com
---
 drivers/net/wireless/quantenna/qtnfmac/commands.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/quantenna/qtnfmac/commands.c b/drivers/net/wireless/quantenna/qtnfmac/commands.c
index b1b73478d89b..68ae9c7ea95a 100644
--- a/drivers/net/wireless/quantenna/qtnfmac/commands.c
+++ b/drivers/net/wireless/quantenna/qtnfmac/commands.c
@@ -1325,9 +1325,10 @@ static int qtnf_cmd_band_fill_iftype(const u8 *data,
 	struct ieee80211_sband_iftype_data *iftype_data;
 	const struct qlink_tlv_iftype_data *tlv =
 		(const struct qlink_tlv_iftype_data *)data;
-	size_t payload_len = tlv->n_iftype_data * sizeof(*tlv->iftype_data) +
-		sizeof(*tlv) -
-		sizeof(struct qlink_tlv_hdr);
+	size_t payload_len;
+
+	payload_len = struct_size(tlv, iftype_data, tlv->n_iftype_data);
+	payload_len = size_sub(payload_len, sizeof(struct qlink_tlv_hdr));
 
 	if (tlv->hdr.len != cpu_to_le16(payload_len)) {
 		pr_err("bad IFTYPE_DATA TLV len %u\n", tlv->hdr.len);
-- 
cgit v1.2.3-70-g09d2


From e49bdd85c92dacb12151aa1b9cf48b81c81a6f98 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:19 +0800
Subject: wifi: rtw89: coex: Add more error_map and counter to log

The error map and counter can help to analyze is coexistence mechanism
going well or not. For example, if there is E2G (External control Wi-Fi
slot for Wi-Fi 2.4 GHz) hang counter, it means Wi-Fi firmware didn't cut
a slot for Wi-Fi 2.4 GHz. Maybe something wrong with firmware timer.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 112 ++++++++++++++++++------------
 drivers/net/wireless/realtek/rtw89/core.h |  31 ++++++---
 2 files changed, 88 insertions(+), 55 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index bcf483cafd20..f80952e6e11b 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -734,6 +734,7 @@ static void _reset_btc_var(struct rtw89_dev *rtwdev, u8 type)
 
 #define BTC_RPT_HDR_SIZE 3
 #define BTC_CHK_WLSLOT_DRIFT_MAX 15
+#define BTC_CHK_BTSLOT_DRIFT_MAX 15
 #define BTC_CHK_HANG_MAX 3
 
 static void _chk_btc_err(struct rtw89_dev *rtwdev, u8 type, u32 cnt)
@@ -748,62 +749,76 @@ static void _chk_btc_err(struct rtw89_dev *rtwdev, u8 type, u32 cnt)
 		    __func__, type, cnt);
 
 	switch (type) {
-	case BTC_DCNT_RPT_FREEZE:
+	case BTC_DCNT_RPT_HANG:
 		if (dm->cnt_dm[BTC_DCNT_RPT] == cnt && btc->fwinfo.rpt_en_map)
-			dm->cnt_dm[BTC_DCNT_RPT_FREEZE]++;
+			dm->cnt_dm[BTC_DCNT_RPT_HANG]++;
 		else
-			dm->cnt_dm[BTC_DCNT_RPT_FREEZE] = 0;
+			dm->cnt_dm[BTC_DCNT_RPT_HANG] = 0;
 
-		if (dm->cnt_dm[BTC_DCNT_RPT_FREEZE] >= BTC_CHK_HANG_MAX)
+		if (dm->cnt_dm[BTC_DCNT_RPT_HANG] >= BTC_CHK_HANG_MAX)
 			dm->error.map.wl_fw_hang = true;
 		else
 			dm->error.map.wl_fw_hang = false;
 
 		dm->cnt_dm[BTC_DCNT_RPT] = cnt;
 		break;
-	case BTC_DCNT_CYCLE_FREEZE:
+	case BTC_DCNT_CYCLE_HANG:
 		if (dm->cnt_dm[BTC_DCNT_CYCLE] == cnt &&
 		    (dm->tdma_now.type != CXTDMA_OFF ||
 		     dm->tdma_now.ext_ctrl == CXECTL_EXT))
-			dm->cnt_dm[BTC_DCNT_CYCLE_FREEZE]++;
+			dm->cnt_dm[BTC_DCNT_CYCLE_HANG]++;
 		else
-			dm->cnt_dm[BTC_DCNT_CYCLE_FREEZE] = 0;
+			dm->cnt_dm[BTC_DCNT_CYCLE_HANG] = 0;
 
-		if (dm->cnt_dm[BTC_DCNT_CYCLE_FREEZE] >= BTC_CHK_HANG_MAX)
+		if (dm->cnt_dm[BTC_DCNT_CYCLE_HANG] >= BTC_CHK_HANG_MAX)
 			dm->error.map.cycle_hang = true;
 		else
 			dm->error.map.cycle_hang = false;
 
 		dm->cnt_dm[BTC_DCNT_CYCLE] = cnt;
 		break;
-	case BTC_DCNT_W1_FREEZE:
+	case BTC_DCNT_W1_HANG:
 		if (dm->cnt_dm[BTC_DCNT_W1] == cnt &&
 		    dm->tdma_now.type != CXTDMA_OFF)
-			dm->cnt_dm[BTC_DCNT_W1_FREEZE]++;
+			dm->cnt_dm[BTC_DCNT_W1_HANG]++;
 		else
-			dm->cnt_dm[BTC_DCNT_W1_FREEZE] = 0;
+			dm->cnt_dm[BTC_DCNT_W1_HANG] = 0;
 
-		if (dm->cnt_dm[BTC_DCNT_W1_FREEZE] >= BTC_CHK_HANG_MAX)
+		if (dm->cnt_dm[BTC_DCNT_W1_HANG] >= BTC_CHK_HANG_MAX)
 			dm->error.map.w1_hang = true;
 		else
 			dm->error.map.w1_hang = false;
 
 		dm->cnt_dm[BTC_DCNT_W1] = cnt;
 		break;
-	case BTC_DCNT_B1_FREEZE:
+	case BTC_DCNT_B1_HANG:
 		if (dm->cnt_dm[BTC_DCNT_B1] == cnt &&
 		    dm->tdma_now.type != CXTDMA_OFF)
-			dm->cnt_dm[BTC_DCNT_B1_FREEZE]++;
+			dm->cnt_dm[BTC_DCNT_B1_HANG]++;
 		else
-			dm->cnt_dm[BTC_DCNT_B1_FREEZE] = 0;
+			dm->cnt_dm[BTC_DCNT_B1_HANG] = 0;
 
-		if (dm->cnt_dm[BTC_DCNT_B1_FREEZE] >= BTC_CHK_HANG_MAX)
+		if (dm->cnt_dm[BTC_DCNT_B1_HANG] >= BTC_CHK_HANG_MAX)
 			dm->error.map.b1_hang = true;
 		else
 			dm->error.map.b1_hang = false;
 
 		dm->cnt_dm[BTC_DCNT_B1] = cnt;
 		break;
+	case BTC_DCNT_E2G_HANG:
+		if (dm->cnt_dm[BTC_DCNT_E2G] == cnt &&
+		    dm->tdma_now.ext_ctrl == CXECTL_EXT)
+			dm->cnt_dm[BTC_DCNT_E2G_HANG]++;
+		else
+			dm->cnt_dm[BTC_DCNT_E2G_HANG] = 0;
+
+		if (dm->cnt_dm[BTC_DCNT_E2G_HANG] >= BTC_CHK_HANG_MAX)
+			dm->error.map.wl_e2g_hang = true;
+		else
+			dm->error.map.wl_e2g_hang = false;
+
+		dm->cnt_dm[BTC_DCNT_E2G] = cnt;
+		break;
 	case BTC_DCNT_TDMA_NONSYNC:
 		if (cnt != 0) /* if tdma not sync between drv/fw  */
 			dm->cnt_dm[BTC_DCNT_TDMA_NONSYNC]++;
@@ -822,23 +837,23 @@ static void _chk_btc_err(struct rtw89_dev *rtwdev, u8 type, u32 cnt)
 			dm->cnt_dm[BTC_DCNT_SLOT_NONSYNC] = 0;
 
 		if (dm->cnt_dm[BTC_DCNT_SLOT_NONSYNC] >= BTC_CHK_HANG_MAX)
-			dm->error.map.tdma_no_sync = true;
+			dm->error.map.slot_no_sync = true;
 		else
-			dm->error.map.tdma_no_sync = false;
+			dm->error.map.slot_no_sync = false;
 		break;
-	case BTC_DCNT_BTCNT_FREEZE:
+	case BTC_DCNT_BTCNT_HANG:
 		cnt = cx->cnt_bt[BTC_BCNT_HIPRI_RX] +
 		      cx->cnt_bt[BTC_BCNT_HIPRI_TX] +
 		      cx->cnt_bt[BTC_BCNT_LOPRI_RX] +
 		      cx->cnt_bt[BTC_BCNT_LOPRI_TX];
 
 		if (cnt == 0)
-			dm->cnt_dm[BTC_DCNT_BTCNT_FREEZE]++;
+			dm->cnt_dm[BTC_DCNT_BTCNT_HANG]++;
 		else
-			dm->cnt_dm[BTC_DCNT_BTCNT_FREEZE] = 0;
+			dm->cnt_dm[BTC_DCNT_BTCNT_HANG] = 0;
 
-		if ((dm->cnt_dm[BTC_DCNT_BTCNT_FREEZE] >= BTC_CHK_HANG_MAX &&
-		     bt->enable.now) || (!dm->cnt_dm[BTC_DCNT_BTCNT_FREEZE] &&
+		if ((dm->cnt_dm[BTC_DCNT_BTCNT_HANG] >= BTC_CHK_HANG_MAX &&
+		     bt->enable.now) || (!dm->cnt_dm[BTC_DCNT_BTCNT_HANG] &&
 		     !bt->enable.now))
 			_update_bt_scbd(rtwdev, false);
 		break;
@@ -853,6 +868,18 @@ static void _chk_btc_err(struct rtw89_dev *rtwdev, u8 type, u32 cnt)
 		else
 			dm->error.map.wl_slot_drift = false;
 		break;
+	case BTC_DCNT_BT_SLOT_DRIFT:
+		if (cnt >= BTC_CHK_BTSLOT_DRIFT_MAX)
+			dm->cnt_dm[BTC_DCNT_BT_SLOT_DRIFT]++;
+		else
+			dm->cnt_dm[BTC_DCNT_BT_SLOT_DRIFT] = 0;
+
+		if (dm->cnt_dm[BTC_DCNT_BT_SLOT_DRIFT] >= BTC_CHK_HANG_MAX)
+			dm->error.map.bt_slot_drift = true;
+		else
+			dm->error.map.bt_slot_drift = false;
+
+		break;
 	}
 }
 
@@ -1129,14 +1156,14 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 			wl->ver_info.fw = prpt->v1.wl_fw_ver;
 			dm->wl_fw_cx_offload = !!prpt->v1.wl_fw_cx_offload;
 
-			_chk_btc_err(rtwdev, BTC_DCNT_RPT_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_RPT_HANG,
 				     pfwinfo->event[BTF_EVNT_RPT]);
 
 			/* To avoid I/O if WL LPS or power-off */
 			if (wl->status.map.lps != BTC_LPS_RF_OFF &&
 			    !wl->status.map.rf_off) {
 				rtwdev->chip->ops->btc_update_bt_cnt(rtwdev);
-				_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_FREEZE, 0);
+				_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_HANG, 0);
 
 				btc->cx.cnt_bt[BTC_BCNT_POLUT] =
 					rtw89_mac_get_plt_cnt(rtwdev,
@@ -1164,8 +1191,8 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 			btc->cx.cnt_bt[BTC_BCNT_POLUT] =
 				le32_to_cpu(prpt->v4.bt_cnt[BTC_BCNT_POLLUTED]);
 
-			_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_FREEZE, 0);
-			_chk_btc_err(rtwdev, BTC_DCNT_RPT_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_HANG, 0);
+			_chk_btc_err(rtwdev, BTC_DCNT_RPT_HANG,
 				     pfwinfo->event[BTF_EVNT_RPT]);
 
 			if (le32_to_cpu(prpt->v4.bt_cnt[BTC_BCNT_RFK_TIMEOUT]) > 0)
@@ -1196,8 +1223,8 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 			btc->cx.cnt_bt[BTC_BCNT_POLUT] =
 				le16_to_cpu(prpt->v5.bt_cnt[BTC_BCNT_POLLUTED]);
 
-			_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_FREEZE, 0);
-			_chk_btc_err(rtwdev, BTC_DCNT_RPT_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_HANG, 0);
+			_chk_btc_err(rtwdev, BTC_DCNT_RPT_HANG,
 				     pfwinfo->event[BTF_EVNT_RPT]);
 
 			dm->error.map.bt_rfk_timeout = bt->rfk_info.map.timeout;
@@ -1258,11 +1285,11 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 					     BTC_DCNT_WL_SLOT_DRIFT, diff_t);
 			}
 
-			_chk_btc_err(rtwdev, BTC_DCNT_W1_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_W1_HANG,
 				     le32_to_cpu(pcysta->v2.slot_cnt[CXST_W1]));
-			_chk_btc_err(rtwdev, BTC_DCNT_W1_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_W1_HANG,
 				     le32_to_cpu(pcysta->v2.slot_cnt[CXST_B1]));
-			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_HANG,
 				     le16_to_cpu(pcysta->v2.cycles));
 		} else if (ver->fcxcysta == 3) {
 			if (le16_to_cpu(pcysta->v3.cycles) < BTC_CYSTA_CHK_PERIOD)
@@ -1299,11 +1326,11 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 				}
 			}
 
-			_chk_btc_err(rtwdev, BTC_DCNT_W1_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_W1_HANG,
 				     le32_to_cpu(pcysta->v3.slot_cnt[CXST_W1]));
-			_chk_btc_err(rtwdev, BTC_DCNT_B1_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_B1_HANG,
 				     le32_to_cpu(pcysta->v3.slot_cnt[CXST_B1]));
-			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_HANG,
 				     le16_to_cpu(pcysta->v3.cycles));
 		} else if (ver->fcxcysta == 4) {
 			if (le16_to_cpu(pcysta->v4.cycles) < BTC_CYSTA_CHK_PERIOD)
@@ -1341,11 +1368,11 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 				}
 			}
 
-			_chk_btc_err(rtwdev, BTC_DCNT_W1_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_W1_HANG,
 				     le16_to_cpu(pcysta->v4.slot_cnt[CXST_W1]));
-			_chk_btc_err(rtwdev, BTC_DCNT_B1_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_B1_HANG,
 				     le16_to_cpu(pcysta->v4.slot_cnt[CXST_B1]));
-			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_FREEZE,
+			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_HANG,
 				     le16_to_cpu(pcysta->v4.cycles));
 		} else {
 			goto err;
@@ -4578,7 +4605,7 @@ static void _update_bt_scbd(struct rtw89_dev *rtwdev, bool only_update)
 	}
 
 	if (!(val & BTC_BSCB_ON) ||
-	    btc->dm.cnt_dm[BTC_DCNT_BTCNT_FREEZE] >= BTC_CHK_HANG_MAX)
+	    btc->dm.cnt_dm[BTC_DCNT_BTCNT_HANG] >= BTC_CHK_HANG_MAX)
 		bt->enable.now = 0;
 	else
 		bt->enable.now = 1;
@@ -5349,7 +5376,7 @@ void rtw89_btc_ntfy_radio_state(struct rtw89_dev *rtwdev, enum btc_rfctrl rf_sta
 			_write_scbd(rtwdev, BTC_WSCB_ALL, false);
 	}
 
-	btc->dm.cnt_dm[BTC_DCNT_BTCNT_FREEZE] = 0;
+	btc->dm.cnt_dm[BTC_DCNT_BTCNT_HANG] = 0;
 	if (wl->status.map.lps_pre == BTC_LPS_OFF &&
 	    wl->status.map.lps_pre != wl->status.map.lps)
 		btc->dm.tdma_instant_excute = 1;
@@ -5700,11 +5727,6 @@ static void _show_cx_info(struct rtw89_dev *rtwdev, struct seq_file *m)
 	seq_printf(m, " %-15s : Coex:%d.%d.%d(branch:%d), ",
 		   "[coex_version]", ver_main, ver_sub, ver_hotfix, id_branch);
 
-	if (dm->wl_fw_cx_offload != BTC_CX_FW_OFFLOAD)
-		dm->error.map.offload_mismatch = true;
-	else
-		dm->error.map.offload_mismatch = false;
-
 	ver_main = FIELD_GET(GENMASK(31, 24), wl->ver_info.fw_coex);
 	ver_sub = FIELD_GET(GENMASK(23, 16), wl->ver_info.fw_coex);
 	ver_hotfix = FIELD_GET(GENMASK(15, 8), wl->ver_info.fw_coex);
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 9073f526b5a3..f998f8a3f277 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -883,20 +883,24 @@ enum rtw89_btc_dcnt {
 	BTC_DCNT_RUN = 0x0,
 	BTC_DCNT_CX_RUNINFO,
 	BTC_DCNT_RPT,
-	BTC_DCNT_RPT_FREEZE,
+	BTC_DCNT_RPT_HANG,
 	BTC_DCNT_CYCLE,
-	BTC_DCNT_CYCLE_FREEZE,
+	BTC_DCNT_CYCLE_HANG,
 	BTC_DCNT_W1,
-	BTC_DCNT_W1_FREEZE,
+	BTC_DCNT_W1_HANG,
 	BTC_DCNT_B1,
-	BTC_DCNT_B1_FREEZE,
+	BTC_DCNT_B1_HANG,
 	BTC_DCNT_TDMA_NONSYNC,
 	BTC_DCNT_SLOT_NONSYNC,
-	BTC_DCNT_BTCNT_FREEZE,
+	BTC_DCNT_BTCNT_HANG,
 	BTC_DCNT_WL_SLOT_DRIFT,
-	BTC_DCNT_BT_SLOT_DRIFT,
 	BTC_DCNT_WL_STA_LAST,
-	BTC_DCNT_NUM,
+	BTC_DCNT_BT_SLOT_DRIFT,
+	BTC_DCNT_BT_SLOT_FLOOD,
+	BTC_DCNT_FDDT_TRIG,
+	BTC_DCNT_E2G,
+	BTC_DCNT_E2G_HANG,
+	BTC_DCNT_NUM
 };
 
 enum rtw89_btc_wl_state_cnt {
@@ -1302,15 +1306,22 @@ struct rtw89_btc_dm_emap {
 	u32 pta_owner: 1;
 	u32 wl_rfk_timeout: 1;
 	u32 bt_rfk_timeout: 1;
-
 	u32 wl_fw_hang: 1;
-	u32 offload_mismatch: 1;
 	u32 cycle_hang: 1;
 	u32 w1_hang: 1;
-
 	u32 b1_hang: 1;
 	u32 tdma_no_sync: 1;
+	u32 slot_no_sync: 1;
 	u32 wl_slot_drift: 1;
+	u32 bt_slot_drift: 1;
+	u32 role_num_mismatch: 1;
+	u32 null1_tx_late: 1;
+	u32 bt_afh_conflict: 1;
+	u32 bt_leafh_conflict: 1;
+	u32 bt_slot_flood: 1;
+	u32 wl_e2g_hang: 1;
+	u32 wl_ver_mismatch: 1;
+	u32 bt_ver_mismatch: 1;
 };
 
 union rtw89_btc_dm_error_map {
-- 
cgit v1.2.3-70-g09d2


From 5049964c4af86865153c553fc6a138df02685ffb Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:20 +0800
Subject: wifi: rtw89: coex: Add WiFi role info v2

Remove WiFi traffic busy level & traffic rate from active role information.
This information will move to v5 version TDMA cycle info.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 277 +++++++++++++++++++++++++++++-
 drivers/net/wireless/realtek/rtw89/core.h |  32 ++++
 drivers/net/wireless/realtek/rtw89/fw.c   |  86 ++++++++++
 drivers/net/wireless/realtek/rtw89/fw.h   |  51 ++++++
 4 files changed, 439 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index f80952e6e11b..b56897529504 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -1819,6 +1819,8 @@ static void _fw_set_drv_info(struct rtw89_dev *rtwdev, u8 type)
 			rtw89_fw_h2c_cxdrv_role(rtwdev);
 		else if (ver->fwlrole == 1)
 			rtw89_fw_h2c_cxdrv_role_v1(rtwdev);
+		else if (ver->fwlrole == 2)
+			rtw89_fw_h2c_cxdrv_role_v2(rtwdev);
 		break;
 	case CXDRVINFO_CTRL:
 		rtw89_fw_h2c_cxdrv_ctrl(rtwdev);
@@ -2113,8 +2115,10 @@ static void _set_bt_afh_info(struct rtw89_dev *rtwdev)
 	struct rtw89_btc_bt_link_info *b = &bt->link_info;
 	struct rtw89_btc_wl_role_info *wl_rinfo = &wl->role_info;
 	struct rtw89_btc_wl_role_info_v1 *wl_rinfo_v1 = &wl->role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo_v2 = &wl->role_info_v2;
 	struct rtw89_btc_wl_active_role *r;
 	struct rtw89_btc_wl_active_role_v1 *r1;
+	struct rtw89_btc_wl_active_role_v2 *r2;
 	u8 en = 0, i, ch = 0, bw = 0;
 	u8 mode, connect_cnt;
 
@@ -2124,9 +2128,14 @@ static void _set_bt_afh_info(struct rtw89_dev *rtwdev)
 	if (ver->fwlrole == 0) {
 		mode = wl_rinfo->link_mode;
 		connect_cnt = wl_rinfo->connect_cnt;
-	} else {
+	} else if (ver->fwlrole == 1) {
 		mode = wl_rinfo_v1->link_mode;
 		connect_cnt = wl_rinfo_v1->connect_cnt;
+	} else if (ver->fwlrole == 2) {
+		mode = wl_rinfo_v2->link_mode;
+		connect_cnt = wl_rinfo_v2->connect_cnt;
+	} else {
+		return;
 	}
 
 	if (wl->status.map.rf_off || bt->whql_test ||
@@ -2139,6 +2148,7 @@ static void _set_bt_afh_info(struct rtw89_dev *rtwdev)
 		for (i = 0; i < RTW89_PORT_NUM; i++) {
 			r = &wl_rinfo->active_role[i];
 			r1 = &wl_rinfo_v1->active_role_v1[i];
+			r2 = &wl_rinfo_v2->active_role_v2[i];
 
 			if (ver->fwlrole == 0 &&
 			    (r->role == RTW89_WIFI_ROLE_P2P_GO ||
@@ -2152,6 +2162,12 @@ static void _set_bt_afh_info(struct rtw89_dev *rtwdev)
 				ch = r1->ch;
 				bw = r1->bw;
 				break;
+			} else if (ver->fwlrole == 2 &&
+				   (r2->role == RTW89_WIFI_ROLE_P2P_GO ||
+				    r2->role == RTW89_WIFI_ROLE_P2P_CLIENT)) {
+				ch = r2->ch;
+				bw = r2->bw;
+				break;
 			}
 		}
 	} else {
@@ -2160,6 +2176,7 @@ static void _set_bt_afh_info(struct rtw89_dev *rtwdev)
 		for (i = 0; i < RTW89_PORT_NUM; i++) {
 			r = &wl_rinfo->active_role[i];
 			r1 = &wl_rinfo_v1->active_role_v1[i];
+			r2 = &wl_rinfo_v2->active_role_v2[i];
 
 			if (ver->fwlrole == 0 &&
 			    r->connected && r->band == RTW89_BAND_2G) {
@@ -2171,6 +2188,11 @@ static void _set_bt_afh_info(struct rtw89_dev *rtwdev)
 				ch = r1->ch;
 				bw = r1->bw;
 				break;
+			} else if (ver->fwlrole == 2 &&
+				   r2->connected && r2->band == RTW89_BAND_2G) {
+				ch = r2->ch;
+				bw = r2->bw;
+				break;
 			}
 		}
 	}
@@ -3625,6 +3647,7 @@ static void _set_btg_ctrl(struct rtw89_dev *rtwdev)
 	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
 	struct rtw89_btc_wl_role_info *wl_rinfo = &wl->role_info;
 	struct rtw89_btc_wl_role_info_v1 *wl_rinfo_v1 = &wl->role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo_v2 = &wl->role_info_v2;
 	struct rtw89_btc_wl_dbcc_info *wl_dinfo = &wl->dbcc_info;
 	bool is_btg;
 	u8 mode;
@@ -3634,8 +3657,12 @@ static void _set_btg_ctrl(struct rtw89_dev *rtwdev)
 
 	if (ver->fwlrole == 0)
 		mode = wl_rinfo->link_mode;
-	else
+	else if (ver->fwlrole == 1)
 		mode = wl_rinfo_v1->link_mode;
+	else if (ver->fwlrole == 2)
+		mode = wl_rinfo_v2->link_mode;
+	else
+		return;
 
 	/* notify halbb ignore GNT_BT or not for WL BB Rx-AGC control */
 	if (mode == BTC_WLINK_5G) /* always 0 if 5G */
@@ -3736,6 +3763,7 @@ static void _set_wl_tx_limit(struct rtw89_dev *rtwdev)
 	struct rtw89_btc_bt_hid_desc *hid = &b->hid_desc;
 	struct rtw89_btc_wl_role_info *wl_rinfo = &wl->role_info;
 	struct rtw89_btc_wl_role_info_v1 *wl_rinfo_v1 = &wl->role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo_v2 = &wl->role_info_v2;
 	struct rtw89_txtime_data data = {.rtwdev = rtwdev};
 	u8 mode;
 	u8 tx_retry;
@@ -3748,8 +3776,12 @@ static void _set_wl_tx_limit(struct rtw89_dev *rtwdev)
 
 	if (ver->fwlrole == 0)
 		mode = wl_rinfo->link_mode;
-	else
+	else if (ver->fwlrole == 1)
 		mode = wl_rinfo_v1->link_mode;
+	else if (ver->fwlrole == 2)
+		mode = wl_rinfo_v2->link_mode;
+	else
+		return;
 
 	if (btc->dm.freerun || btc->ctrl.igno_bt || b->profile_cnt.now == 0 ||
 	    mode == BTC_WLINK_5G || mode == BTC_WLINK_NOLINK) {
@@ -3799,14 +3831,19 @@ static void _set_bt_rx_agc(struct rtw89_dev *rtwdev)
 	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
 	struct rtw89_btc_wl_role_info *wl_rinfo = &wl->role_info;
 	struct rtw89_btc_wl_role_info_v1 *wl_rinfo_v1 = &wl->role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo_v2 = &wl->role_info_v2;
 	struct rtw89_btc_bt_info *bt = &btc->cx.bt;
 	bool bt_hi_lna_rx = false;
 	u8 mode;
 
 	if (ver->fwlrole == 0)
 		mode = wl_rinfo->link_mode;
-	else
+	else if (ver->fwlrole == 1)
 		mode = wl_rinfo_v1->link_mode;
+	else if (ver->fwlrole == 2)
+		mode = wl_rinfo_v2->link_mode;
+	else
+		return;
 
 	if (mode != BTC_WLINK_NOLINK && btc->dm.wl_btg_rx)
 		bt_hi_lna_rx = true;
@@ -4079,6 +4116,68 @@ static void _action_wl_2g_scc_v1(struct rtw89_dev *rtwdev)
 	_set_policy(rtwdev, policy_type, BTC_ACT_WL_2G_SCC);
 }
 
+static void _action_wl_2g_scc_v2(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
+	struct rtw89_btc_bt_info *bt = &btc->cx.bt;
+	struct rtw89_btc_dm *dm = &btc->dm;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo = &wl->role_info_v2;
+	u16 policy_type = BTC_CXP_OFF_BT;
+	u32 dur;
+
+	if (btc->mdinfo.ant.type == BTC_ANT_DEDICATED) {
+		policy_type = BTC_CXP_OFF_EQ0;
+	} else {
+		/* shared-antenna */
+		switch (wl_rinfo->mrole_type) {
+		case BTC_WLMROLE_STA_GC:
+			dm->wl_scc.null_role1 = RTW89_WIFI_ROLE_STATION;
+			dm->wl_scc.null_role2 = RTW89_WIFI_ROLE_P2P_CLIENT;
+			dm->wl_scc.ebt_null = 0; /* no ext-slot-control */
+			_action_by_bt(rtwdev);
+			return;
+		case BTC_WLMROLE_STA_STA:
+			dm->wl_scc.null_role1 = RTW89_WIFI_ROLE_STATION;
+			dm->wl_scc.null_role2 = RTW89_WIFI_ROLE_STATION;
+			dm->wl_scc.ebt_null = 0; /* no ext-slot-control */
+			_action_by_bt(rtwdev);
+			return;
+		case BTC_WLMROLE_STA_GC_NOA:
+		case BTC_WLMROLE_STA_GO:
+		case BTC_WLMROLE_STA_GO_NOA:
+			dm->wl_scc.null_role1 = RTW89_WIFI_ROLE_STATION;
+			dm->wl_scc.null_role2 = RTW89_WIFI_ROLE_NONE;
+			dur = wl_rinfo->mrole_noa_duration;
+
+			if (wl->status.map._4way) {
+				dm->wl_scc.ebt_null = 0;
+				policy_type = BTC_CXP_OFFE_WL;
+			} else if (bt->link_info.status.map.connect == 0) {
+				dm->wl_scc.ebt_null = 0;
+				policy_type = BTC_CXP_OFFE_2GISOB;
+			} else if (bt->link_info.a2dp_desc.exist &&
+				   dur < btc->bt_req_len) {
+				dm->wl_scc.ebt_null = 1; /* tx null at EBT */
+				policy_type = BTC_CXP_OFFE_2GBWMIXB2;
+			} else if (bt->link_info.a2dp_desc.exist ||
+				   bt->link_info.pan_desc.exist) {
+				dm->wl_scc.ebt_null = 1; /* tx null at EBT */
+				policy_type = BTC_CXP_OFFE_2GBWISOB;
+			} else {
+				dm->wl_scc.ebt_null = 0;
+				policy_type = BTC_CXP_OFFE_2GBWISOB;
+			}
+			break;
+		default:
+			break;
+		}
+	}
+
+	_set_ant(rtwdev, NM_EXEC, BTC_PHY_ALL, BTC_ANT_W2G);
+	_set_policy(rtwdev, policy_type, BTC_ACT_WL_2G_SCC);
+}
+
 static void _action_wl_2g_ap(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
@@ -4518,6 +4617,156 @@ static void _update_wl_info_v1(struct rtw89_dev *rtwdev)
 	_fw_set_drv_info(rtwdev, CXDRVINFO_ROLE);
 }
 
+static void _update_wl_info_v2(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
+	struct rtw89_btc_wl_link_info *wl_linfo = wl->link_info;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo = &wl->role_info_v2;
+	struct rtw89_btc_wl_dbcc_info *wl_dinfo = &wl->dbcc_info;
+	u8 cnt_connect = 0, cnt_connecting = 0, cnt_active = 0;
+	u8 cnt_2g = 0, cnt_5g = 0, phy;
+	u32 wl_2g_ch[2] = {}, wl_5g_ch[2] = {};
+	bool b2g = false, b5g = false, client_joined = false;
+	u8 i;
+
+	memset(wl_rinfo, 0, sizeof(*wl_rinfo));
+
+	for (i = 0; i < RTW89_PORT_NUM; i++) {
+		if (!wl_linfo[i].active)
+			continue;
+
+		cnt_active++;
+		wl_rinfo->active_role_v2[cnt_active - 1].role = wl_linfo[i].role;
+		wl_rinfo->active_role_v2[cnt_active - 1].pid = wl_linfo[i].pid;
+		wl_rinfo->active_role_v2[cnt_active - 1].phy = wl_linfo[i].phy;
+		wl_rinfo->active_role_v2[cnt_active - 1].band = wl_linfo[i].band;
+		wl_rinfo->active_role_v2[cnt_active - 1].noa = (u8)wl_linfo[i].noa;
+		wl_rinfo->active_role_v2[cnt_active - 1].connected = 0;
+
+		wl->port_id[wl_linfo[i].role] = wl_linfo[i].pid;
+
+		phy = wl_linfo[i].phy;
+
+		if (rtwdev->dbcc_en && phy < RTW89_PHY_MAX) {
+			wl_dinfo->role[phy] = wl_linfo[i].role;
+			wl_dinfo->op_band[phy] = wl_linfo[i].band;
+			_update_dbcc_band(rtwdev, phy);
+			_fw_set_drv_info(rtwdev, CXDRVINFO_DBCC);
+		}
+
+		if (wl_linfo[i].connected == MLME_NO_LINK) {
+			continue;
+		} else if (wl_linfo[i].connected == MLME_LINKING) {
+			cnt_connecting++;
+		} else {
+			cnt_connect++;
+			if ((wl_linfo[i].role == RTW89_WIFI_ROLE_P2P_GO ||
+			     wl_linfo[i].role == RTW89_WIFI_ROLE_AP) &&
+			     wl_linfo[i].client_cnt > 1)
+				client_joined = true;
+		}
+
+		wl_rinfo->role_map.val |= BIT(wl_linfo[i].role);
+		wl_rinfo->active_role_v2[cnt_active - 1].ch = wl_linfo[i].ch;
+		wl_rinfo->active_role_v2[cnt_active - 1].bw = wl_linfo[i].bw;
+		wl_rinfo->active_role_v2[cnt_active - 1].connected = 1;
+
+		/* only care 2 roles + BT coex */
+		if (wl_linfo[i].band != RTW89_BAND_2G) {
+			if (cnt_5g <= ARRAY_SIZE(wl_5g_ch) - 1)
+				wl_5g_ch[cnt_5g] = wl_linfo[i].ch;
+			cnt_5g++;
+			b5g = true;
+		} else {
+			if (cnt_2g <= ARRAY_SIZE(wl_2g_ch) - 1)
+				wl_2g_ch[cnt_2g] = wl_linfo[i].ch;
+			cnt_2g++;
+			b2g = true;
+		}
+	}
+
+	wl_rinfo->connect_cnt = cnt_connect;
+
+	/* Be careful to change the following sequence!! */
+	if (cnt_connect == 0) {
+		wl_rinfo->link_mode = BTC_WLINK_NOLINK;
+		wl_rinfo->role_map.role.none = 1;
+	} else if (!b2g && b5g) {
+		wl_rinfo->link_mode = BTC_WLINK_5G;
+	} else if (wl_rinfo->role_map.role.nan) {
+		wl_rinfo->link_mode = BTC_WLINK_2G_NAN;
+	} else if (cnt_connect > BTC_TDMA_WLROLE_MAX) {
+		wl_rinfo->link_mode = BTC_WLINK_OTHER;
+	} else  if (b2g && b5g && cnt_connect == 2) {
+		if (rtwdev->dbcc_en) {
+			switch (wl_dinfo->role[RTW89_PHY_0]) {
+			case RTW89_WIFI_ROLE_STATION:
+				wl_rinfo->link_mode = BTC_WLINK_2G_STA;
+				break;
+			case RTW89_WIFI_ROLE_P2P_GO:
+				wl_rinfo->link_mode = BTC_WLINK_2G_GO;
+				break;
+			case RTW89_WIFI_ROLE_P2P_CLIENT:
+				wl_rinfo->link_mode = BTC_WLINK_2G_GC;
+				break;
+			case RTW89_WIFI_ROLE_AP:
+				wl_rinfo->link_mode = BTC_WLINK_2G_AP;
+				break;
+			default:
+				wl_rinfo->link_mode = BTC_WLINK_OTHER;
+				break;
+			}
+		} else {
+			wl_rinfo->link_mode = BTC_WLINK_25G_MCC;
+		}
+	} else if (!b5g && cnt_connect == 2) {
+		if (wl_rinfo->role_map.role.station &&
+		    (wl_rinfo->role_map.role.p2p_go ||
+		    wl_rinfo->role_map.role.p2p_gc ||
+		    wl_rinfo->role_map.role.ap)) {
+			if (wl_2g_ch[0] == wl_2g_ch[1])
+				wl_rinfo->link_mode = BTC_WLINK_2G_SCC;
+			else
+				wl_rinfo->link_mode = BTC_WLINK_2G_MCC;
+		} else {
+			wl_rinfo->link_mode = BTC_WLINK_2G_MCC;
+		}
+	} else if (!b5g && cnt_connect == 1) {
+		if (wl_rinfo->role_map.role.station)
+			wl_rinfo->link_mode = BTC_WLINK_2G_STA;
+		else if (wl_rinfo->role_map.role.ap)
+			wl_rinfo->link_mode = BTC_WLINK_2G_AP;
+		else if (wl_rinfo->role_map.role.p2p_go)
+			wl_rinfo->link_mode = BTC_WLINK_2G_GO;
+		else if (wl_rinfo->role_map.role.p2p_gc)
+			wl_rinfo->link_mode = BTC_WLINK_2G_GC;
+		else
+			wl_rinfo->link_mode = BTC_WLINK_OTHER;
+	}
+
+	/* if no client_joined, don't care P2P-GO/AP role */
+	if (wl_rinfo->role_map.role.p2p_go || wl_rinfo->role_map.role.ap) {
+		if (!client_joined) {
+			if (wl_rinfo->link_mode == BTC_WLINK_2G_SCC ||
+			    wl_rinfo->link_mode == BTC_WLINK_2G_MCC) {
+				wl_rinfo->link_mode = BTC_WLINK_2G_STA;
+				wl_rinfo->connect_cnt = 1;
+			} else if (wl_rinfo->link_mode == BTC_WLINK_2G_GO ||
+				 wl_rinfo->link_mode == BTC_WLINK_2G_AP) {
+				wl_rinfo->link_mode = BTC_WLINK_NOLINK;
+				wl_rinfo->connect_cnt = 0;
+			}
+		}
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_BTC,
+		    "[BTC], cnt_connect = %d, connecting = %d, link_mode = %d\n",
+		    cnt_connect, cnt_connecting, wl_rinfo->link_mode);
+
+	_fw_set_drv_info(rtwdev, CXDRVINFO_ROLE);
+}
+
 #define BTC_CHK_HANG_MAX 3
 #define BTC_SCB_INV_VALUE GENMASK(31, 0)
 
@@ -4676,6 +4925,7 @@ void _run_coex(struct rtw89_dev *rtwdev, enum btc_reason_and_action reason)
 	struct rtw89_btc_bt_info *bt = &btc->cx.bt;
 	struct rtw89_btc_wl_role_info *wl_rinfo = &wl->role_info;
 	struct rtw89_btc_wl_role_info_v1 *wl_rinfo_v1 = &wl->role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo_v2 = &wl->role_info_v2;
 	u8 mode;
 
 	lockdep_assert_held(&rtwdev->mutex);
@@ -4686,8 +4936,12 @@ void _run_coex(struct rtw89_dev *rtwdev, enum btc_reason_and_action reason)
 
 	if (ver->fwlrole == 0)
 		mode = wl_rinfo->link_mode;
-	else
+	else if (ver->fwlrole == 1)
 		mode = wl_rinfo_v1->link_mode;
+	else if (ver->fwlrole == 2)
+		mode = wl_rinfo_v2->link_mode;
+	else
+		return;
 
 	rtw89_debug(rtwdev, RTW89_DBG_BTC, "[BTC], %s(): reason=%d, mode=%d\n",
 		    __func__, reason, mode);
@@ -4812,6 +5066,8 @@ void _run_coex(struct rtw89_dev *rtwdev, enum btc_reason_and_action reason)
 			_action_wl_2g_scc(rtwdev);
 		else if (ver->fwlrole == 1)
 			_action_wl_2g_scc_v1(rtwdev);
+		else if (ver->fwlrole == 2)
+			_action_wl_2g_scc_v2(rtwdev);
 		break;
 	case BTC_WLINK_2G_MCC:
 		bt->scan_rx_low_pri = true;
@@ -5317,8 +5573,10 @@ void rtw89_btc_ntfy_role_info(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif
 	memcpy(wlinfo, &r, sizeof(*wlinfo));
 	if (ver->fwlrole == 0)
 		_update_wl_info(rtwdev);
-	else
+	else if (ver->fwlrole == 1)
 		_update_wl_info_v1(rtwdev);
+	else if (ver->fwlrole == 2)
+		_update_wl_info_v2(rtwdev);
 
 	if (wlinfo->role == RTW89_WIFI_ROLE_STATION &&
 	    wlinfo->connected == MLME_NO_LINK)
@@ -5838,6 +6096,7 @@ static void _show_wl_info(struct rtw89_dev *rtwdev, struct seq_file *m)
 	struct rtw89_btc_wl_info *wl = &cx->wl;
 	struct rtw89_btc_wl_role_info *wl_rinfo = &wl->role_info;
 	struct rtw89_btc_wl_role_info_v1 *wl_rinfo_v1 = &wl->role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 *wl_rinfo_v2 = &wl->role_info_v2;
 	u8 mode;
 
 	if (!(btc->dm.coex_info_map & BTC_COEX_INFO_WL))
@@ -5847,8 +6106,12 @@ static void _show_wl_info(struct rtw89_dev *rtwdev, struct seq_file *m)
 
 	if (ver->fwlrole == 0)
 		mode = wl_rinfo->link_mode;
-	else
+	else if (ver->fwlrole == 1)
 		mode = wl_rinfo_v1->link_mode;
+	else if (ver->fwlrole == 2)
+		mode = wl_rinfo_v2->link_mode;
+	else
+		return;
 
 	seq_printf(m, " %-15s : link_mode:%d, ", "[status]", mode);
 
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index f998f8a3f277..f51cd8665192 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1180,6 +1180,22 @@ struct rtw89_btc_wl_active_role_v1 {
 	u32 noa_duration; /* ms */
 };
 
+struct rtw89_btc_wl_active_role_v2 {
+	u8 connected: 1;
+	u8 pid: 3;
+	u8 phy: 1;
+	u8 noa: 1;
+	u8 band: 2;
+
+	u8 client_ps: 1;
+	u8 bw: 7;
+
+	u8 role;
+	u8 ch;
+
+	u32 noa_duration; /* ms */
+};
+
 struct rtw89_btc_wl_role_info_bpos {
 	u16 none: 1;
 	u16 station: 1;
@@ -1228,6 +1244,21 @@ struct rtw89_btc_wl_role_info_v1 { /* struct size must be n*4 bytes */
 	u32 rsvd: 27;
 };
 
+struct rtw89_btc_wl_role_info_v2 { /* struct size must be n*4 bytes */
+	u8 connect_cnt;
+	u8 link_mode;
+	union rtw89_btc_wl_role_info_map role_map;
+	struct rtw89_btc_wl_active_role_v2 active_role_v2[RTW89_PORT_NUM];
+	u32 mrole_type; /* btc_wl_mrole_type */
+	u32 mrole_noa_duration; /* ms */
+
+	u32 dbcc_en: 1;
+	u32 dbcc_chg: 1;
+	u32 dbcc_2g_phy: 2; /* which phy operate in 2G, HW_PHY_0 or HW_PHY_1 */
+	u32 link_mode_chg: 1;
+	u32 rsvd: 27;
+};
+
 struct rtw89_btc_wl_ver_info {
 	u32 fw_coex; /* match with which coex_ver */
 	u32 fw;
@@ -1343,6 +1374,7 @@ struct rtw89_btc_wl_info {
 	struct rtw89_btc_wl_afh_info afh_info;
 	struct rtw89_btc_wl_role_info role_info;
 	struct rtw89_btc_wl_role_info_v1 role_info_v1;
+	struct rtw89_btc_wl_role_info_v2 role_info_v2;
 	struct rtw89_btc_wl_scan_info scan_info;
 	struct rtw89_btc_wl_dbcc_info dbcc_info;
 	struct rtw89_btc_rf_para rf_para;
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 1a4ff24078fb..46d9cda2c617 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2038,6 +2038,92 @@ fail:
 	return ret;
 }
 
+#define H2C_LEN_CXDRVINFO_ROLE_SIZE_V2(max_role_num) \
+	(4 + 8 * (max_role_num) + H2C_LEN_CXDRVINFO_ROLE_DBCC_LEN + H2C_LEN_CXDRVHDR)
+
+int rtw89_fw_h2c_cxdrv_role_v2(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	const struct rtw89_btc_ver *ver = btc->ver;
+	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
+	struct rtw89_btc_wl_role_info_v2 *role_info = &wl->role_info_v2;
+	struct rtw89_btc_wl_role_info_bpos *bpos = &role_info->role_map.role;
+	struct rtw89_btc_wl_active_role_v2 *active = role_info->active_role_v2;
+	struct sk_buff *skb;
+	u32 len;
+	u8 *cmd, offset;
+	int ret;
+	int i;
+
+	len = H2C_LEN_CXDRVINFO_ROLE_SIZE_V2(ver->max_role_num);
+
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len);
+	if (!skb) {
+		rtw89_err(rtwdev, "failed to alloc skb for h2c cxdrv_role\n");
+		return -ENOMEM;
+	}
+	skb_put(skb, len);
+	cmd = skb->data;
+
+	RTW89_SET_FWCMD_CXHDR_TYPE(cmd, CXDRVINFO_ROLE);
+	RTW89_SET_FWCMD_CXHDR_LEN(cmd, len - H2C_LEN_CXDRVHDR);
+
+	RTW89_SET_FWCMD_CXROLE_CONNECT_CNT(cmd, role_info->connect_cnt);
+	RTW89_SET_FWCMD_CXROLE_LINK_MODE(cmd, role_info->link_mode);
+
+	RTW89_SET_FWCMD_CXROLE_ROLE_NONE(cmd, bpos->none);
+	RTW89_SET_FWCMD_CXROLE_ROLE_STA(cmd, bpos->station);
+	RTW89_SET_FWCMD_CXROLE_ROLE_AP(cmd, bpos->ap);
+	RTW89_SET_FWCMD_CXROLE_ROLE_VAP(cmd, bpos->vap);
+	RTW89_SET_FWCMD_CXROLE_ROLE_ADHOC(cmd, bpos->adhoc);
+	RTW89_SET_FWCMD_CXROLE_ROLE_ADHOC_MASTER(cmd, bpos->adhoc_master);
+	RTW89_SET_FWCMD_CXROLE_ROLE_MESH(cmd, bpos->mesh);
+	RTW89_SET_FWCMD_CXROLE_ROLE_MONITOR(cmd, bpos->moniter);
+	RTW89_SET_FWCMD_CXROLE_ROLE_P2P_DEV(cmd, bpos->p2p_device);
+	RTW89_SET_FWCMD_CXROLE_ROLE_P2P_GC(cmd, bpos->p2p_gc);
+	RTW89_SET_FWCMD_CXROLE_ROLE_P2P_GO(cmd, bpos->p2p_go);
+	RTW89_SET_FWCMD_CXROLE_ROLE_NAN(cmd, bpos->nan);
+
+	offset = PORT_DATA_OFFSET;
+	for (i = 0; i < RTW89_PORT_NUM; i++, active++) {
+		RTW89_SET_FWCMD_CXROLE_ACT_CONNECTED_V2(cmd, active->connected, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_PID_V2(cmd, active->pid, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_PHY_V2(cmd, active->phy, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_NOA_V2(cmd, active->noa, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_BAND_V2(cmd, active->band, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_CLIENT_PS_V2(cmd, active->client_ps, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_BW_V2(cmd, active->bw, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_ROLE_V2(cmd, active->role, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_CH_V2(cmd, active->ch, i, offset);
+		RTW89_SET_FWCMD_CXROLE_ACT_NOA_DUR_V2(cmd, active->noa_duration, i, offset);
+	}
+
+	offset = len - H2C_LEN_CXDRVINFO_ROLE_DBCC_LEN;
+	RTW89_SET_FWCMD_CXROLE_MROLE_TYPE(cmd, role_info->mrole_type, offset);
+	RTW89_SET_FWCMD_CXROLE_MROLE_NOA(cmd, role_info->mrole_noa_duration, offset);
+	RTW89_SET_FWCMD_CXROLE_DBCC_EN(cmd, role_info->dbcc_en, offset);
+	RTW89_SET_FWCMD_CXROLE_DBCC_CHG(cmd, role_info->dbcc_chg, offset);
+	RTW89_SET_FWCMD_CXROLE_DBCC_2G_PHY(cmd, role_info->dbcc_2g_phy, offset);
+	RTW89_SET_FWCMD_CXROLE_LINK_MODE_CHG(cmd, role_info->link_mode_chg, offset);
+
+	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
+			      H2C_CAT_OUTSRC, BTFC_SET,
+			      SET_DRV_INFO, 0, 0,
+			      len);
+
+	ret = rtw89_h2c_tx(rtwdev, skb, false);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to send h2c\n");
+		goto fail;
+	}
+
+	return 0;
+fail:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
 #define H2C_LEN_CXDRVINFO_CTRL (4 + H2C_LEN_CXDRVHDR)
 int rtw89_fw_h2c_cxdrv_ctrl(struct rtw89_dev *rtwdev)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index 3f6e0871381d..fa9e786ad7c9 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -2393,6 +2393,56 @@ static inline void RTW89_SET_FWCMD_CXROLE_ACT_NOA_DUR(void *cmd, u32 val, int n,
 	le32p_replace_bits((__le32 *)((u8 *)cmd + (20 + (12 + offset) * n)), val, GENMASK(31, 0));
 }
 
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_CONNECTED_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (6 + (12 + offset) * n), val, BIT(0));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_PID_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (6 + (12 + offset) * n), val, GENMASK(3, 1));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_PHY_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (6 + (12 + offset) * n), val, BIT(4));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_NOA_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (6 + (12 + offset) * n), val, BIT(5));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_BAND_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (6 + (12 + offset) * n), val, GENMASK(7, 6));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_CLIENT_PS_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (7 + (12 + offset) * n), val, BIT(0));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_BW_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (7 + (12 + offset) * n), val, GENMASK(7, 1));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_ROLE_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (8 + (12 + offset) * n), val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_CH_V2(void *cmd, u8 val, int n, u8 offset)
+{
+	u8p_replace_bits((u8 *)cmd + (9 + (12 + offset) * n), val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXROLE_ACT_NOA_DUR_V2(void *cmd, u32 val, int n, u8 offset)
+{
+	le32p_replace_bits((__le32 *)((u8 *)cmd + (10 + (12 + offset) * n)), val, GENMASK(31, 0));
+}
+
 static inline void RTW89_SET_FWCMD_CXROLE_MROLE_TYPE(void *cmd, u32 val, u8 offset)
 {
 	le32p_replace_bits((__le32 *)((u8 *)cmd + offset), val, GENMASK(31, 0));
@@ -3505,6 +3555,7 @@ int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi
 int rtw89_fw_h2c_cxdrv_init(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_role(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_role_v1(struct rtw89_dev *rtwdev);
+int rtw89_fw_h2c_cxdrv_role_v2(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_ctrl(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_rfk(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_del_pkt_offload(struct rtw89_dev *rtwdev, u8 id);
-- 
cgit v1.2.3-70-g09d2


From a2c0ce5d01a2218af4756d311ae91845b67ac5b9 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:21 +0800
Subject: wifi: rtw89: coex: Add traffic TX/RX info and its H2C

There is a new mechanism which can do some real time performance
tuning for WiFi and BT. This TX/RX info is a condition provide to
firmware to do traffic analysis.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 59 +++++++++++++++++++--
 drivers/net/wireless/realtek/rtw89/core.h | 43 +++++++++++++++
 drivers/net/wireless/realtek/rtw89/fw.c   | 56 ++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/fw.h   | 87 +++++++++++++++++++++++++++++++
 4 files changed, 241 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index b56897529504..908ef8bb149a 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -1809,6 +1809,9 @@ static void _fw_set_drv_info(struct rtw89_dev *rtwdev, u8 type)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
 	const struct rtw89_btc_ver *ver = btc->ver;
+	struct rtw89_btc_dm *dm = &btc->dm;
+	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
+	struct rtw89_btc_rf_trx_para rf_para = dm->rf_trx_para;
 
 	switch (type) {
 	case CXDRVINFO_INIT:
@@ -1825,6 +1828,19 @@ static void _fw_set_drv_info(struct rtw89_dev *rtwdev, u8 type)
 	case CXDRVINFO_CTRL:
 		rtw89_fw_h2c_cxdrv_ctrl(rtwdev);
 		break;
+	case CXDRVINFO_TRX:
+		dm->trx_info.tx_power = u32_get_bits(rf_para.wl_tx_power,
+						     RTW89_BTC_WL_DEF_TX_PWR);
+		dm->trx_info.rx_gain = u32_get_bits(rf_para.wl_rx_gain,
+						    RTW89_BTC_WL_DEF_TX_PWR);
+		dm->trx_info.bt_tx_power = u32_get_bits(rf_para.bt_tx_power,
+							RTW89_BTC_WL_DEF_TX_PWR);
+		dm->trx_info.bt_rx_gain = u32_get_bits(rf_para.bt_rx_gain,
+						       RTW89_BTC_WL_DEF_TX_PWR);
+		dm->trx_info.cn = wl->cn_report;
+		dm->trx_info.nhm = wl->nhm.pwr;
+		rtw89_fw_h2c_cxdrv_trx(rtwdev);
+		break;
 	case CXDRVINFO_RFK:
 		rtw89_fw_h2c_cxdrv_rfk(rtwdev);
 		break;
@@ -5361,6 +5377,8 @@ void rtw89_btc_ntfy_icmp_packet_work(struct work_struct *work)
 	mutex_unlock(&rtwdev->mutex);
 }
 
+#define BT_PROFILE_PROTOCOL_MASK GENMASK(7, 4)
+
 static void _update_bt_info(struct rtw89_dev *rtwdev, u8 *buf, u32 len)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
@@ -5417,6 +5435,7 @@ static void _update_bt_info(struct rtw89_dev *rtwdev, u8 *buf, u32 len)
 	a2dp->exist = btinfo.lb2.a2dp;
 	b->profile_cnt.now += (u8)a2dp->exist;
 	pan->active = btinfo.lb2.pan;
+	btc->dm.trx_info.bt_profile = u32_get_bits(btinfo.val, BT_PROFILE_PROTOCOL_MASK);
 
 	/* parse raw info low-Byte3 */
 	btinfo.val = bt->raw_info[BTC_BTINFO_L3];
@@ -5433,6 +5452,7 @@ static void _update_bt_info(struct rtw89_dev *rtwdev, u8 *buf, u32 len)
 	btinfo.val = bt->raw_info[BTC_BTINFO_H0];
 	/* raw val is dBm unit, translate from -100~ 0dBm to 0~100%*/
 	b->rssi = chip->ops->btc_get_bt_rssi(rtwdev, btinfo.hb0.rssi);
+	btc->dm.trx_info.bt_rssi = b->rssi;
 
 	/* parse raw info high-Byte1 */
 	btinfo.val = bt->raw_info[BTC_BTINFO_H1];
@@ -5766,6 +5786,8 @@ static void rtw89_btc_ntfy_wl_sta_iter(void *data, struct ieee80211_sta *sta)
 				(struct rtw89_btc_wl_sta_iter_data *)data;
 	struct rtw89_dev *rtwdev = iter_data->rtwdev;
 	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_dm *dm = &btc->dm;
+	const struct rtw89_btc_ver *ver = btc->ver;
 	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
 	struct rtw89_btc_wl_link_info *link_info = NULL;
 	struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
@@ -5773,6 +5795,8 @@ static void rtw89_btc_ntfy_wl_sta_iter(void *data, struct ieee80211_sta *sta)
 	struct rtw89_vif *rtwvif = rtwsta->rtwvif;
 	struct rtw89_traffic_stats *stats = &rtwvif->stats;
 	const struct rtw89_chip_info *chip = rtwdev->chip;
+	struct rtw89_btc_wl_role_info *r;
+	struct rtw89_btc_wl_role_info_v1 *r1;
 	u32 last_tx_rate, last_rx_rate;
 	u16 last_tx_lvl, last_rx_lvl;
 	u8 port = rtwvif->port;
@@ -5849,10 +5873,33 @@ static void rtw89_btc_ntfy_wl_sta_iter(void *data, struct ieee80211_sta *sta)
 	link_info_t->tx_rate = rtwsta->ra_report.hw_rate;
 	link_info_t->rx_rate = rtwsta->rx_hw_rate;
 
-	wl->role_info.active_role[port].tx_lvl = (u16)stats->tx_tfc_lv;
-	wl->role_info.active_role[port].rx_lvl = (u16)stats->rx_tfc_lv;
-	wl->role_info.active_role[port].tx_rate = rtwsta->ra_report.hw_rate;
-	wl->role_info.active_role[port].rx_rate = rtwsta->rx_hw_rate;
+	if (link_info->role == RTW89_WIFI_ROLE_STATION ||
+	    link_info->role == RTW89_WIFI_ROLE_P2P_CLIENT) {
+		dm->trx_info.tx_rate = link_info_t->tx_rate;
+		dm->trx_info.rx_rate = link_info_t->rx_rate;
+	}
+
+	if (ver->fwlrole == 0) {
+		r = &wl->role_info;
+		r->active_role[port].tx_lvl = stats->tx_tfc_lv;
+		r->active_role[port].rx_lvl = stats->rx_tfc_lv;
+		r->active_role[port].tx_rate = rtwsta->ra_report.hw_rate;
+		r->active_role[port].rx_rate = rtwsta->rx_hw_rate;
+	} else if (ver->fwlrole == 1) {
+		r1 = &wl->role_info_v1;
+		r1->active_role_v1[port].tx_lvl = stats->tx_tfc_lv;
+		r1->active_role_v1[port].rx_lvl = stats->rx_tfc_lv;
+		r1->active_role_v1[port].tx_rate = rtwsta->ra_report.hw_rate;
+		r1->active_role_v1[port].rx_rate = rtwsta->rx_hw_rate;
+	} else if (ver->fwlrole == 2) {
+		dm->trx_info.tx_lvl = stats->tx_tfc_lv;
+		dm->trx_info.rx_lvl = stats->rx_tfc_lv;
+		dm->trx_info.tx_rate = rtwsta->ra_report.hw_rate;
+		dm->trx_info.rx_rate = rtwsta->rx_hw_rate;
+	}
+
+	dm->trx_info.tx_tp = link_info_t->tx_throughput;
+	dm->trx_info.rx_tp = link_info_t->rx_throughput;
 
 	if (is_sta_change)
 		iter_data->is_sta_change = true;
@@ -5866,6 +5913,7 @@ static void rtw89_btc_ntfy_wl_sta_iter(void *data, struct ieee80211_sta *sta)
 void rtw89_btc_ntfy_wl_sta(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_dm *dm = &btc->dm;
 	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
 	struct rtw89_btc_wl_sta_iter_data data = {.rtwdev = rtwdev};
 	u8 i;
@@ -5884,6 +5932,9 @@ void rtw89_btc_ntfy_wl_sta(struct rtw89_dev *rtwdev)
 		}
 	}
 
+	if (dm->trx_info.wl_rssi != wl->rssi_level)
+		dm->trx_info.wl_rssi = wl->rssi_level;
+
 	rtw89_debug(rtwdev, RTW89_DBG_BTC, "[BTC], %s(): busy=%d\n",
 		    __func__, !!wl->status.map.busy);
 
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index f51cd8665192..024441d45a66 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1367,6 +1367,22 @@ struct rtw89_btc_rf_para {
 	u32 rx_gain_perpkt;
 };
 
+struct rtw89_btc_wl_nhm {
+	u8 instant_wl_nhm_dbm;
+	u8 instant_wl_nhm_per_mhz;
+	u16 valid_record_times;
+	s8 record_pwr[16];
+	u8 record_ratio[16];
+	s8 pwr; /* dbm_per_MHz  */
+	u8 ratio;
+	u8 current_status;
+	u8 refresh;
+	bool start_flag;
+	u8 last_ccx_rpt_stamp;
+	s8 pwr_max;
+	s8 pwr_min;
+};
+
 struct rtw89_btc_wl_info {
 	struct rtw89_btc_wl_link_info link_info[RTW89_PORT_NUM];
 	struct rtw89_btc_wl_rfk_info rfk_info;
@@ -1378,10 +1394,12 @@ struct rtw89_btc_wl_info {
 	struct rtw89_btc_wl_scan_info scan_info;
 	struct rtw89_btc_wl_dbcc_info dbcc_info;
 	struct rtw89_btc_rf_para rf_para;
+	struct rtw89_btc_wl_nhm nhm;
 	union rtw89_btc_wl_state_map status;
 
 	u8 port_id[RTW89_WIFI_ROLE_MLME_MAX];
 	u8 rssi_level;
+	u8 cn_report;
 
 	bool scbd_change;
 	u32 scbd;
@@ -2019,6 +2037,30 @@ struct rtw89_btc_rf_trx_para {
 	u8 bt_rx_gain;  /* LNA constrain level */
 };
 
+struct rtw89_btc_trx_info {
+	u8 tx_lvl;
+	u8 rx_lvl;
+	u8 wl_rssi;
+	u8 bt_rssi;
+
+	s8 tx_power; /* absolute Tx power (dBm), 0xff-> no BTC control */
+	s8 rx_gain;  /* rx gain table index (TBD.) */
+	s8 bt_tx_power; /* decrease Tx power (dB) */
+	s8 bt_rx_gain;  /* LNA constrain level */
+
+	u8 cn; /* condition_num */
+	s8 nhm;
+	u8 bt_profile;
+	u8 rsvd2;
+
+	u16 tx_rate;
+	u16 rx_rate;
+
+	u32 tx_tp;
+	u32 rx_tp;
+	u32 rx_err_ratio;
+};
+
 struct rtw89_btc_dm {
 	struct rtw89_btc_fbtc_slot slot[CXST_MAX];
 	struct rtw89_btc_fbtc_slot slot_now[CXST_MAX];
@@ -2030,6 +2072,7 @@ struct rtw89_btc_dm {
 	struct rtw89_btc_wl_tx_limit_para wl_tx_limit;
 	struct rtw89_btc_dm_step dm_step;
 	struct rtw89_btc_wl_scc_ctrl wl_scc;
+	struct rtw89_btc_trx_info trx_info;
 	union rtw89_btc_dm_error_map error;
 	u32 cnt_dm[BTC_DCNT_NUM];
 	u32 cnt_notify[BTC_NCNT_NUM];
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 46d9cda2c617..713aefd0cf8d 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2169,6 +2169,62 @@ fail:
 	return ret;
 }
 
+#define H2C_LEN_CXDRVINFO_TRX (28 + H2C_LEN_CXDRVHDR)
+int rtw89_fw_h2c_cxdrv_trx(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_trx_info *trx = &btc->dm.trx_info;
+	struct sk_buff *skb;
+	u8 *cmd;
+	int ret;
+
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, H2C_LEN_CXDRVINFO_TRX);
+	if (!skb) {
+		rtw89_err(rtwdev, "failed to alloc skb for h2c cxdrv_trx\n");
+		return -ENOMEM;
+	}
+	skb_put(skb, H2C_LEN_CXDRVINFO_TRX);
+	cmd = skb->data;
+
+	RTW89_SET_FWCMD_CXHDR_TYPE(cmd, CXDRVINFO_TRX);
+	RTW89_SET_FWCMD_CXHDR_LEN(cmd, H2C_LEN_CXDRVINFO_TRX - H2C_LEN_CXDRVHDR);
+
+	RTW89_SET_FWCMD_CXTRX_TXLV(cmd, trx->tx_lvl);
+	RTW89_SET_FWCMD_CXTRX_RXLV(cmd, trx->rx_lvl);
+	RTW89_SET_FWCMD_CXTRX_WLRSSI(cmd, trx->wl_rssi);
+	RTW89_SET_FWCMD_CXTRX_BTRSSI(cmd, trx->bt_rssi);
+	RTW89_SET_FWCMD_CXTRX_TXPWR(cmd, trx->tx_power);
+	RTW89_SET_FWCMD_CXTRX_RXGAIN(cmd, trx->rx_gain);
+	RTW89_SET_FWCMD_CXTRX_BTTXPWR(cmd, trx->bt_tx_power);
+	RTW89_SET_FWCMD_CXTRX_BTRXGAIN(cmd, trx->bt_rx_gain);
+	RTW89_SET_FWCMD_CXTRX_CN(cmd, trx->cn);
+	RTW89_SET_FWCMD_CXTRX_NHM(cmd, trx->nhm);
+	RTW89_SET_FWCMD_CXTRX_BTPROFILE(cmd, trx->bt_profile);
+	RTW89_SET_FWCMD_CXTRX_RSVD2(cmd, trx->rsvd2);
+	RTW89_SET_FWCMD_CXTRX_TXRATE(cmd, trx->tx_rate);
+	RTW89_SET_FWCMD_CXTRX_RXRATE(cmd, trx->rx_rate);
+	RTW89_SET_FWCMD_CXTRX_TXTP(cmd, trx->tx_tp);
+	RTW89_SET_FWCMD_CXTRX_RXTP(cmd, trx->rx_tp);
+	RTW89_SET_FWCMD_CXTRX_RXERRRA(cmd, trx->rx_err_ratio);
+
+	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
+			      H2C_CAT_OUTSRC, BTFC_SET,
+			      SET_DRV_INFO, 0, 0,
+			      H2C_LEN_CXDRVINFO_TRX);
+
+	ret = rtw89_h2c_tx(rtwdev, skb, false);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to send h2c\n");
+		goto fail;
+	}
+
+	return 0;
+fail:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
 #define H2C_LEN_CXDRVINFO_RFK (4 + H2C_LEN_CXDRVHDR)
 int rtw89_fw_h2c_cxdrv_rfk(struct rtw89_dev *rtwdev)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index fa9e786ad7c9..c3c67ddf61a2 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -2152,6 +2152,7 @@ enum rtw89_btc_cxdrvinfo {
 	CXDRVINFO_RUN,
 	CXDRVINFO_CTRL,
 	CXDRVINFO_SCAN,
+	CXDRVINFO_TRX,  /* WL traffic to WL fw */
 	CXDRVINFO_MAX,
 };
 
@@ -2493,6 +2494,91 @@ static inline void RTW89_SET_FWCMD_CXCTRL_TRACE_STEP(void *cmd, u32 val)
 	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 2), val, GENMASK(18, 3));
 }
 
+static inline void RTW89_SET_FWCMD_CXTRX_TXLV(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 2, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_RXLV(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 3, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_WLRSSI(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 4, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_BTRSSI(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 5, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_TXPWR(void *cmd, s8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 6, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_RXGAIN(void *cmd, s8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 7, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_BTTXPWR(void *cmd, s8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 8, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_BTRXGAIN(void *cmd, s8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 9, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_CN(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 10, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_NHM(void *cmd, s8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 11, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_BTPROFILE(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 12, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_RSVD2(void *cmd, u8 val)
+{
+	u8p_replace_bits((u8 *)cmd + 13, val, GENMASK(7, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_TXRATE(void *cmd, u16 val)
+{
+	le16p_replace_bits((__le16 *)((u8 *)cmd + 14), val, GENMASK(15, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_RXRATE(void *cmd, u16 val)
+{
+	le16p_replace_bits((__le16 *)((u8 *)cmd + 16), val, GENMASK(15, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_TXTP(void *cmd, u32 val)
+{
+	le32p_replace_bits((__le32 *)((u8 *)cmd + 18), val, GENMASK(31, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_RXTP(void *cmd, u32 val)
+{
+	le32p_replace_bits((__le32 *)((u8 *)cmd + 22), val, GENMASK(31, 0));
+}
+
+static inline void RTW89_SET_FWCMD_CXTRX_RXERRRA(void *cmd, u32 val)
+{
+	le32p_replace_bits((__le32 *)((u8 *)cmd + 26), val, GENMASK(31, 0));
+}
+
 static inline void RTW89_SET_FWCMD_CXRFK_STATE(void *cmd, u32 val)
 {
 	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 2), val, GENMASK(1, 0));
@@ -3557,6 +3643,7 @@ int rtw89_fw_h2c_cxdrv_role(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_role_v1(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_role_v2(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_ctrl(struct rtw89_dev *rtwdev);
+int rtw89_fw_h2c_cxdrv_trx(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_rfk(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_del_pkt_offload(struct rtw89_dev *rtwdev, u8 id);
 int rtw89_fw_h2c_add_pkt_offload(struct rtw89_dev *rtwdev, u8 *id,
-- 
cgit v1.2.3-70-g09d2


From e5e52feb5053a537180cf928428deb8bc697a42e Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:22 +0800
Subject: wifi: rtw89: coex: Add register monitor report v2 format

The v2 firmware report reduce its maximum register numbers from 30 to 20,
it can help to save firmware code size.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 148 +++++++++++++++++++++++++++---
 drivers/net/wireless/realtek/rtw89/core.h |  17 +++-
 2 files changed, 151 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index 908ef8bb149a..436d5e35ae49 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -226,7 +226,6 @@ struct rtw89_btc_btf_set_slot_table {
 	u8 buf[];
 } __packed;
 
-#define BTF_SET_MON_REG_VER 1
 struct rtw89_btc_btf_set_mon_reg {
 	u8 fver;
 	u8 reg_num;
@@ -1078,8 +1077,15 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 		break;
 	case BTC_RPT_TYPE_MREG:
 		pcinfo = &pfwinfo->rpt_fbtc_mregval.cinfo;
-		pfinfo = &pfwinfo->rpt_fbtc_mregval.finfo;
-		pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_mregval.finfo);
+		if (ver->fcxmreg == 1) {
+			pfinfo = &pfwinfo->rpt_fbtc_mregval.finfo.v1;
+			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_mregval.finfo.v1);
+		} else if (ver->fcxmreg == 2) {
+			pfinfo = &pfwinfo->rpt_fbtc_mregval.finfo.v2;
+			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_mregval.finfo.v2);
+		} else {
+			goto err;
+		}
 		pcinfo->req_fver = ver->fcxmreg;
 		break;
 	case BTC_RPT_TYPE_GPIO_DBG:
@@ -1709,18 +1715,26 @@ static void rtw89_btc_fw_set_slots(struct rtw89_dev *rtwdev, u8 num,
 static void btc_fw_set_monreg(struct rtw89_dev *rtwdev)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
+	const struct rtw89_btc_ver *ver = rtwdev->btc.ver;
 	struct rtw89_btc_btf_set_mon_reg *monreg = NULL;
-	u8 n, *ptr = NULL, ulen;
+	u8 n, *ptr = NULL, ulen, cxmreg_max;
 	u16 sz = 0;
 
 	n = chip->mon_reg_num;
-
 	rtw89_debug(rtwdev, RTW89_DBG_BTC,
 		    "[BTC], %s(): mon_reg_num=%d\n", __func__, n);
-	if (n > CXMREG_MAX) {
+
+	if (ver->fcxmreg == 1)
+		cxmreg_max = CXMREG_MAX;
+	else if (ver->fcxmreg == 2)
+		cxmreg_max = CXMREG_MAX_V2;
+	else
+		return;
+
+	if (n > cxmreg_max) {
 		rtw89_debug(rtwdev, RTW89_DBG_BTC,
 			    "[BTC], %s(): mon reg count %d > %d\n",
-			    __func__, n, CXMREG_MAX);
+			    __func__, n, cxmreg_max);
 		return;
 	}
 
@@ -1730,7 +1744,7 @@ static void btc_fw_set_monreg(struct rtw89_dev *rtwdev)
 	if (!monreg)
 		return;
 
-	monreg->fver = BTF_SET_MON_REG_VER;
+	monreg->fver = ver->fcxmreg;
 	monreg->reg_num = n;
 	ptr = &monreg->buf[0];
 	memcpy(ptr, chip->mon_reg, n * ulen);
@@ -7401,13 +7415,13 @@ static void _get_gnt(struct rtw89_dev *rtwdev, struct rtw89_mac_ax_coex_gnt *gnt
 	}
 }
 
-static void _show_mreg(struct rtw89_dev *rtwdev, struct seq_file *m)
+static void _show_mreg_v1(struct rtw89_dev *rtwdev, struct seq_file *m)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
 	struct rtw89_btc *btc = &rtwdev->btc;
 	struct rtw89_btc_btf_fwinfo *pfwinfo = &btc->fwinfo;
 	struct rtw89_btc_rpt_cmn_info *pcinfo = NULL;
-	struct rtw89_btc_fbtc_mreg_val *pmreg = NULL;
+	struct rtw89_btc_fbtc_mreg_val_v1 *pmreg = NULL;
 	struct rtw89_btc_fbtc_gpio_dbg *gdbg = NULL;
 	struct rtw89_btc_cx *cx = &btc->cx;
 	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
@@ -7457,7 +7471,7 @@ static void _show_mreg(struct rtw89_dev *rtwdev, struct seq_file *m)
 		return;
 	}
 
-	pmreg = &pfwinfo->rpt_fbtc_mregval.finfo;
+	pmreg = &pfwinfo->rpt_fbtc_mregval.finfo.v1;
 	rtw89_debug(rtwdev, RTW89_DBG_BTC,
 		    "[BTC], %s(): rpt_fbtc_mregval reg_num = %d\n",
 		    __func__, pmreg->reg_num);
@@ -7486,6 +7500,111 @@ static void _show_mreg(struct rtw89_dev *rtwdev, struct seq_file *m)
 		rtw89_debug(rtwdev, RTW89_DBG_BTC,
 			    "[BTC], %s(): stop due rpt_fbtc_gpio_dbg.cinfo\n",
 			    __func__);
+		seq_puts(m, "\n");
+		return;
+	}
+
+	gdbg = &pfwinfo->rpt_fbtc_gpio_dbg.finfo;
+	if (!gdbg->en_map)
+		return;
+
+	seq_printf(m, " %-15s : enable_map:0x%08x",
+		   "[gpio_dbg]", gdbg->en_map);
+
+	for (i = 0; i < BTC_DBG_MAX1; i++) {
+		if (!(gdbg->en_map & BIT(i)))
+			continue;
+		seq_printf(m, ", %d->GPIO%d", (u32)i, gdbg->gpio_map[i]);
+	}
+	seq_puts(m, "\n");
+}
+
+static void _show_mreg_v2(struct rtw89_dev *rtwdev, struct seq_file *m)
+{
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_btf_fwinfo *pfwinfo = &btc->fwinfo;
+	struct rtw89_btc_rpt_cmn_info *pcinfo = NULL;
+	struct rtw89_btc_fbtc_mreg_val_v2 *pmreg = NULL;
+	struct rtw89_btc_fbtc_gpio_dbg *gdbg = NULL;
+	struct rtw89_btc_cx *cx = &btc->cx;
+	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
+	struct rtw89_btc_bt_info *bt = &btc->cx.bt;
+	struct rtw89_mac_ax_coex_gnt gnt_cfg = {};
+	struct rtw89_mac_ax_gnt gnt;
+	u8 i = 0, type = 0, cnt = 0;
+	u32 val, offset;
+
+	if (!(btc->dm.coex_info_map & BTC_COEX_INFO_MREG))
+		return;
+
+	seq_puts(m, "========== [HW Status] ==========\n");
+
+	seq_printf(m,
+		   " %-15s : WL->BT:0x%08x(cnt:%d), BT->WL:0x%08x(total:%d, bt_update:%d)\n",
+		   "[scoreboard]", wl->scbd, cx->cnt_wl[BTC_WCNT_SCBDUPDATE],
+		   bt->scbd, cx->cnt_bt[BTC_BCNT_SCBDREAD],
+		   cx->cnt_bt[BTC_BCNT_SCBDUPDATE]);
+
+	/* To avoid I/O if WL LPS or power-off  */
+	if (!wl->status.map.lps && !wl->status.map.rf_off) {
+		btc->dm.pta_owner = rtw89_mac_get_ctrl_path(rtwdev);
+
+		_get_gnt(rtwdev, &gnt_cfg);
+		gnt = gnt_cfg.band[0];
+		seq_printf(m,
+			   " %-15s : pta_owner:%s, phy-0[gnt_wl:%s-%d/gnt_bt:%s-%d], ",
+			   "[gnt_status]",
+			   chip->chip_id == RTL8852C ? "HW" :
+			   btc->dm.pta_owner == BTC_CTRL_BY_WL ? "WL" : "BT",
+			   gnt.gnt_wl_sw_en ? "SW" : "HW", gnt.gnt_wl,
+			   gnt.gnt_bt_sw_en ? "SW" : "HW", gnt.gnt_bt);
+
+		gnt = gnt_cfg.band[1];
+		seq_printf(m, "phy-1[gnt_wl:%s-%d/gnt_bt:%s-%d]\n",
+			   gnt.gnt_wl_sw_en ? "SW" : "HW",
+			   gnt.gnt_wl,
+			   gnt.gnt_bt_sw_en ? "SW" : "HW",
+			   gnt.gnt_bt);
+	}
+	pcinfo = &pfwinfo->rpt_fbtc_mregval.cinfo;
+	if (!pcinfo->valid) {
+		rtw89_debug(rtwdev, RTW89_DBG_BTC,
+			    "[BTC], %s(): stop due rpt_fbtc_mregval.cinfo\n",
+			    __func__);
+		return;
+	}
+
+	pmreg = &pfwinfo->rpt_fbtc_mregval.finfo.v2;
+	rtw89_debug(rtwdev, RTW89_DBG_BTC,
+		    "[BTC], %s(): rpt_fbtc_mregval reg_num = %d\n",
+		    __func__, pmreg->reg_num);
+
+	for (i = 0; i < pmreg->reg_num; i++) {
+		type = (u8)le16_to_cpu(chip->mon_reg[i].type);
+		offset = le32_to_cpu(chip->mon_reg[i].offset);
+		val = le32_to_cpu(pmreg->mreg_val[i]);
+
+		if (cnt % 6 == 0)
+			seq_printf(m, " %-15s : %d_0x%04x=0x%08x",
+				   "[reg]", (u32)type, offset, val);
+		else
+			seq_printf(m, ", %d_0x%04x=0x%08x", (u32)type,
+				   offset, val);
+		if (cnt % 6 == 5)
+			seq_puts(m, "\n");
+		cnt++;
+
+		if (i >= pmreg->reg_num)
+			seq_puts(m, "\n");
+	}
+
+	pcinfo = &pfwinfo->rpt_fbtc_gpio_dbg.cinfo;
+	if (!pcinfo->valid) {
+		rtw89_debug(rtwdev, RTW89_DBG_BTC,
+			    "[BTC], %s(): stop due rpt_fbtc_gpio_dbg.cinfo\n",
+			    __func__);
+		seq_puts(m, "\n");
 		return;
 	}
 
@@ -7868,7 +7987,12 @@ void rtw89_btc_dump_info(struct rtw89_dev *rtwdev, struct seq_file *m)
 	_show_bt_info(rtwdev, m);
 	_show_dm_info(rtwdev, m);
 	_show_fw_dm_msg(rtwdev, m);
-	_show_mreg(rtwdev, m);
+
+	if (ver->fcxmreg == 1)
+		_show_mreg_v1(rtwdev, m);
+	else if (ver->fcxmreg == 2)
+		_show_mreg_v2(rtwdev, m);
+
 	if (ver->fcxbtcrpt == 1)
 		_show_summary_v1(rtwdev, m);
 	else if (ver->fcxbtcrpt == 4)
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 024441d45a66..b0525f258d3c 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1524,6 +1524,7 @@ union rtw89_btc_fbtc_tdma_le32 {
 };
 
 #define CXMREG_MAX 30
+#define CXMREG_MAX_V2 20
 #define FCXMAX_STEP 255 /*STEP trace record cnt, Max:65535, default:255*/
 #define BTC_CYCLE_SLOT_MAX 48 /* must be even number, non-zero */
 
@@ -1750,13 +1751,25 @@ struct rtw89_btc_fbtc_gpio_dbg {
 	u8 gpio_map[BTC_DBG_MAX1]; /*the debug signals to GPIO-Position */
 } __packed;
 
-struct rtw89_btc_fbtc_mreg_val {
+struct rtw89_btc_fbtc_mreg_val_v1 {
 	u8 fver; /* btc_ver::fcxmreg */
 	u8 reg_num;
 	__le16 rsvd;
 	__le32 mreg_val[CXMREG_MAX];
 } __packed;
 
+struct rtw89_btc_fbtc_mreg_val_v2 {
+	u8 fver; /* btc_ver::fcxmreg */
+	u8 reg_num;
+	__le16 rsvd;
+	__le32 mreg_val[CXMREG_MAX_V2];
+} __packed;
+
+union rtw89_btc_fbtc_mreg_val {
+	struct rtw89_btc_fbtc_mreg_val_v1 v1;
+	struct rtw89_btc_fbtc_mreg_val_v2 v2;
+};
+
 #define RTW89_DEF_FBTC_MREG(__type, __bytes, __offset) \
 	{ .type = cpu_to_le16(__type), .bytes = cpu_to_le16(__bytes), \
 	  .offset = cpu_to_le32(__offset), }
@@ -2203,7 +2216,7 @@ struct rtw89_btc_rpt_fbtc_nullsta {
 
 struct rtw89_btc_rpt_fbtc_mreg {
 	struct rtw89_btc_rpt_cmn_info cinfo; /* common info, by driver */
-	struct rtw89_btc_fbtc_mreg_val finfo; /* info from fw */
+	union rtw89_btc_fbtc_mreg_val finfo; /* info from fw */
 };
 
 struct rtw89_btc_rpt_fbtc_gpio_dbg {
-- 
cgit v1.2.3-70-g09d2


From 9dfa09e0628d2024ce4574f645344c00fe88a535 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:23 +0800
Subject: wifi: rtw89: coex: Fix wrong structure assignment at null data report

Correct pointer assignment of v1 null data report. It doesn't really
change logic at all, but it looks more readable.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-6-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index 436d5e35ae49..0ca621a9f80a 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -1065,7 +1065,7 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 	case BTC_RPT_TYPE_NULLSTA:
 		pcinfo = &pfwinfo->rpt_fbtc_nullsta.cinfo;
 		if (ver->fcxnullsta == 1) {
-			pfinfo = &pfwinfo->rpt_fbtc_nullsta.finfo;
+			pfinfo = &pfwinfo->rpt_fbtc_nullsta.finfo.v1;
 			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_nullsta.finfo.v1);
 		} else if (ver->fcxnullsta == 2) {
 			pfinfo = &pfwinfo->rpt_fbtc_nullsta.finfo.v2;
-- 
cgit v1.2.3-70-g09d2


From 262cc19ea902d2280e2e2a56b153f37466e3349e Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:24 +0800
Subject: wifi: rtw89: coex: Add v2 Bluetooth scan info

Compare to v1 and v2 removed some not usable parameters. Save firmware
code size. The information can show how frequent and how long the
Bluetooth scan do. It will help to debug coexistence issue.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-7-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 67 ++++++++++++++++++++++++++++---
 drivers/net/wireless/realtek/rtw89/core.h | 63 ++++++++++++++++++++---------
 2 files changed, 106 insertions(+), 24 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index 0ca621a9f80a..b867e54feb89 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -890,13 +890,15 @@ static void _update_bt_report(struct rtw89_dev *rtwdev, u8 rpt_type, u8 *pfinfo)
 	struct rtw89_btc_bt_link_info *bt_linfo = &bt->link_info;
 	struct rtw89_btc_bt_a2dp_desc *a2dp = &bt_linfo->a2dp_desc;
 	struct rtw89_btc_fbtc_btver *pver = NULL;
-	struct rtw89_btc_fbtc_btscan *pscan = NULL;
+	struct rtw89_btc_fbtc_btscan_v1 *pscan_v1;
+	struct rtw89_btc_fbtc_btscan_v2 *pscan_v2;
 	struct rtw89_btc_fbtc_btafh *pafh_v1 = NULL;
 	struct rtw89_btc_fbtc_btafh_v2 *pafh_v2 = NULL;
 	struct rtw89_btc_fbtc_btdevinfo *pdev = NULL;
+	bool scan_update = true;
+	int i;
 
 	pver = (struct rtw89_btc_fbtc_btver *)pfinfo;
-	pscan = (struct rtw89_btc_fbtc_btscan *)pfinfo;
 	pdev = (struct rtw89_btc_fbtc_btdevinfo *)pfinfo;
 
 	rtw89_debug(rtwdev, RTW89_DBG_BTC,
@@ -910,7 +912,26 @@ static void _update_bt_report(struct rtw89_dev *rtwdev, u8 rpt_type, u8 *pfinfo)
 		bt->feature = le32_to_cpu(pver->feature);
 		break;
 	case BTC_RPT_TYPE_BT_SCAN:
-		memcpy(bt->scan_info, pscan->scan, BTC_SCAN_MAX1);
+		if (ver->fcxbtscan == 1) {
+			pscan_v1 = (struct rtw89_btc_fbtc_btscan_v1 *)pfinfo;
+			for (i = 0; i < BTC_SCAN_MAX1; i++) {
+				bt->scan_info_v1[i] = pscan_v1->scan[i];
+				if (bt->scan_info_v1[i].win == 0 &&
+				    bt->scan_info_v1[i].intvl == 0)
+					scan_update = false;
+			}
+		} else if (ver->fcxbtscan == 2) {
+			pscan_v2 = (struct rtw89_btc_fbtc_btscan_v2 *)pfinfo;
+			for (i = 0; i < CXSCAN_MAX; i++) {
+				bt->scan_info_v2[i] = pscan_v2->para[i];
+				if ((pscan_v2->type & BIT(i)) &&
+				    pscan_v2->para[i].win == 0 &&
+				    pscan_v2->para[i].intvl == 0)
+					scan_update = false;
+			}
+		}
+		if (scan_update)
+			bt->scan_info_update = 1;
 		break;
 	case BTC_RPT_TYPE_BT_AFH:
 		if (ver->fcxbtafh == 2) {
@@ -1102,8 +1123,13 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 		break;
 	case BTC_RPT_TYPE_BT_SCAN:
 		pcinfo = &pfwinfo->rpt_fbtc_btscan.cinfo;
-		pfinfo = &pfwinfo->rpt_fbtc_btscan.finfo;
-		pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_btscan.finfo);
+		if (ver->fcxbtscan == 1) {
+			pfinfo = &pfwinfo->rpt_fbtc_btscan.finfo.v1;
+			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_btscan.finfo.v1);
+		} else if (ver->fcxbtscan == 2) {
+			pfinfo = &pfwinfo->rpt_fbtc_btscan.finfo.v2;
+			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_btscan.finfo.v2);
+		}
 		pcinfo->req_fver = ver->fcxbtscan;
 		break;
 	case BTC_RPT_TYPE_BT_AFH:
@@ -6346,11 +6372,40 @@ static void _show_bt_info(struct rtw89_dev *rtwdev, struct seq_file *m)
 		   cx->cnt_bt[BTC_BCNT_INFOSAME]);
 
 	seq_printf(m,
-		   " %-15s : Hi-rx = %d, Hi-tx = %d, Lo-rx = %d, Lo-tx = %d (bt_polut_wl_tx = %d)\n",
+		   " %-15s : Hi-rx = %d, Hi-tx = %d, Lo-rx = %d, Lo-tx = %d (bt_polut_wl_tx = %d)",
 		   "[trx_req_cnt]", cx->cnt_bt[BTC_BCNT_HIPRI_RX],
 		   cx->cnt_bt[BTC_BCNT_HIPRI_TX], cx->cnt_bt[BTC_BCNT_LOPRI_RX],
 		   cx->cnt_bt[BTC_BCNT_LOPRI_TX], cx->cnt_bt[BTC_BCNT_POLUT]);
 
+	if (!bt->scan_info_update) {
+		rtw89_btc_fw_en_rpt(rtwdev, RPT_EN_BT_SCAN_INFO, true);
+		seq_puts(m, "\n");
+	} else {
+		rtw89_btc_fw_en_rpt(rtwdev, RPT_EN_BT_SCAN_INFO, false);
+		if (ver->fcxbtscan == 1) {
+			seq_printf(m,
+				   "(INQ:%d-%d/PAGE:%d-%d/LE:%d-%d/INIT:%d-%d)",
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_INQ].win),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_INQ].intvl),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_PAGE].win),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_PAGE].intvl),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_BLE].win),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_BLE].intvl),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_INIT].win),
+				   le16_to_cpu(bt->scan_info_v1[BTC_SCAN_INIT].intvl));
+		} else if (ver->fcxbtscan == 2) {
+			seq_printf(m,
+				   "(BG:%d-%d/INIT:%d-%d/LE:%d-%d)",
+				   le16_to_cpu(bt->scan_info_v2[CXSCAN_BG].win),
+				   le16_to_cpu(bt->scan_info_v2[CXSCAN_BG].intvl),
+				   le16_to_cpu(bt->scan_info_v2[CXSCAN_INIT].win),
+				   le16_to_cpu(bt->scan_info_v2[CXSCAN_INIT].intvl),
+				   le16_to_cpu(bt->scan_info_v2[CXSCAN_LE].win),
+				   le16_to_cpu(bt->scan_info_v2[CXSCAN_LE].intvl));
+		}
+		seq_puts(m, "\n");
+	}
+
 	if (bt->enable.now && bt->ver_info.fw == 0)
 		rtw89_btc_fw_en_rpt(rtwdev, RPT_EN_BT_VER_INFO, true);
 	else
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index b0525f258d3c..02fb3c73b137 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1445,14 +1445,6 @@ struct rtw89_btc_wl_tx_limit_para {
 	u16 tx_retry;
 };
 
-struct rtw89_btc_bt_scan_info {
-	u16 win;
-	u16 intvl;
-	u32 enable: 1;
-	u32 interlace: 1;
-	u32 rsvd: 30;
-};
-
 enum rtw89_btc_bt_scan_type {
 	BTC_SCAN_INQ	= 0,
 	BTC_SCAN_PAGE,
@@ -1463,9 +1455,50 @@ enum rtw89_btc_bt_scan_type {
 	BTC_SCAN_MAX1,
 };
 
+enum rtw89_btc_ble_scan_type {
+	CXSCAN_BG = 0,
+	CXSCAN_INIT,
+	CXSCAN_LE,
+	CXSCAN_MAX
+};
+
+#define RTW89_BTC_BTC_SCAN_V1_FLAG_ENABLE BIT(0)
+#define RTW89_BTC_BTC_SCAN_V1_FLAG_INTERLACE BIT(1)
+
+struct rtw89_btc_bt_scan_info_v1 {
+	__le16 win;
+	__le16 intvl;
+	__le32 flags;
+} __packed;
+
+struct rtw89_btc_bt_scan_info_v2 {
+	__le16 win;
+	__le16 intvl;
+} __packed;
+
+struct rtw89_btc_fbtc_btscan_v1 {
+	u8 fver; /* btc_ver::fcxbtscan */
+	u8 rsvd;
+	__le16 rsvd2;
+	struct rtw89_btc_bt_scan_info_v1 scan[BTC_SCAN_MAX1];
+} __packed;
+
+struct rtw89_btc_fbtc_btscan_v2 {
+	u8 fver; /* btc_ver::fcxbtscan */
+	u8 type;
+	__le16 rsvd2;
+	struct rtw89_btc_bt_scan_info_v2 para[CXSCAN_MAX];
+} __packed;
+
+union rtw89_btc_fbtc_btscan {
+	struct rtw89_btc_fbtc_btscan_v1 v1;
+	struct rtw89_btc_fbtc_btscan_v2 v2;
+};
+
 struct rtw89_btc_bt_info {
 	struct rtw89_btc_bt_link_info link_info;
-	struct rtw89_btc_bt_scan_info scan_info[BTC_SCAN_MAX1];
+	struct rtw89_btc_bt_scan_info_v1 scan_info_v1[BTC_SCAN_MAX1];
+	struct rtw89_btc_bt_scan_info_v2 scan_info_v2[CXSCAN_MAX];
 	struct rtw89_btc_bt_ver_info ver_info;
 	struct rtw89_btc_bool_sta_chg enable;
 	struct rtw89_btc_bool_sta_chg inq_pag;
@@ -1488,7 +1521,8 @@ struct rtw89_btc_bt_info {
 	u32 run_patch_code: 1;
 	u32 hi_lna_rx: 1;
 	u32 scan_rx_low_pri: 1;
-	u32 rsvd: 21;
+	u32 scan_info_update: 1;
+	u32 rsvd: 20;
 };
 
 struct rtw89_btc_cx {
@@ -2006,13 +2040,6 @@ struct rtw89_btc_fbtc_btver {
 	__le32 feature;
 } __packed;
 
-struct rtw89_btc_fbtc_btscan {
-	u8 fver; /* btc_ver::fcxbtscan */
-	u8 rsvd;
-	__le16 rsvd2;
-	u8 scan[6];
-} __packed;
-
 struct rtw89_btc_fbtc_btafh {
 	u8 fver; /* btc_ver::fcxbtafh */
 	u8 rsvd;
@@ -2231,7 +2258,7 @@ struct rtw89_btc_rpt_fbtc_btver {
 
 struct rtw89_btc_rpt_fbtc_btscan {
 	struct rtw89_btc_rpt_cmn_info cinfo; /* common info, by driver */
-	struct rtw89_btc_fbtc_btscan finfo; /* info from fw */
+	union rtw89_btc_fbtc_btscan finfo; /* info from fw */
 };
 
 struct rtw89_btc_rpt_fbtc_btafh {
-- 
cgit v1.2.3-70-g09d2


From 3ab7f9b90cc0a737e0bd8a312dc48814c4682867 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 8 Mar 2023 13:32:25 +0800
Subject: wifi: rtw89: coex: Add v5 firmware cycle status report

To support v5 version firmware cycle report, apply the related structure
and functions. v5 cycle report add a group of status to show how the
free-run/TDMA training goes to. It is a firmware mechanism that can auto
adjust coexistence mode between TDMA and free run mechanism at 3 antenna
solution. v5 version provide more reference data to let the mechanism
make decision.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230308053225.24377-8-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 194 +++++++++++++++++++++++++++++-
 drivers/net/wireless/realtek/rtw89/coex.h |   5 +
 drivers/net/wireless/realtek/rtw89/core.h |  52 ++++++++
 3 files changed, 249 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index b867e54feb89..3dbd4ee14c70 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -987,8 +987,8 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 	void *rpt_content = NULL, *pfinfo = NULL;
 	u8 rpt_type = 0;
 	u16 wl_slot_set = 0, wl_slot_real = 0;
-	u32 trace_step = btc->ctrl.trace_step, rpt_len = 0, diff_t;
-	u32 cnt_leak_slot = 0, bt_slot_real = 0, cnt_rx_imr = 0;
+	u32 trace_step = btc->ctrl.trace_step, rpt_len = 0, diff_t = 0;
+	u32 cnt_leak_slot, bt_slot_real, bt_slot_set, cnt_rx_imr;
 	u8 i;
 
 	rtw89_debug(rtwdev, RTW89_DBG_BTC,
@@ -1061,6 +1061,10 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 			pfinfo = &pfwinfo->rpt_fbtc_cysta.finfo.v4;
 			pcysta->v4 = pfwinfo->rpt_fbtc_cysta.finfo.v4;
 			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_cysta.finfo.v4);
+		} else if (ver->fcxcysta == 5) {
+			pfinfo = &pfwinfo->rpt_fbtc_cysta.finfo.v5;
+			pcysta->v5 = pfwinfo->rpt_fbtc_cysta.finfo.v5;
+			pcinfo->req_len = sizeof(pfwinfo->rpt_fbtc_cysta.finfo.v5);
 		} else {
 			goto err;
 		}
@@ -1406,6 +1410,54 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 				     le16_to_cpu(pcysta->v4.slot_cnt[CXST_B1]));
 			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_HANG,
 				     le16_to_cpu(pcysta->v4.cycles));
+		} else if (ver->fcxcysta == 5) {
+			if (dm->fddt_train == BTC_FDDT_ENABLE)
+				break;
+			cnt_leak_slot = le16_to_cpu(pcysta->v5.slot_cnt[CXST_LK]);
+			cnt_rx_imr = le32_to_cpu(pcysta->v5.leak_slot.cnt_rximr);
+
+			/* Check Leak-AP */
+			if (cnt_leak_slot != 0 && cnt_rx_imr != 0 &&
+			    dm->tdma_now.rxflctrl) {
+				if (le16_to_cpu(pcysta->v5.cycles) >= BTC_CYSTA_CHK_PERIOD &&
+				    cnt_leak_slot < BTC_LEAK_AP_TH * cnt_rx_imr)
+					dm->leak_ap = 1;
+			}
+
+			/* Check diff time between real WL slot and W1 slot */
+			if (dm->tdma_now.type == CXTDMA_OFF) {
+				wl_slot_set = le16_to_cpu(dm->slot_now[CXST_W1].dur);
+				wl_slot_real = le16_to_cpu(pcysta->v5.cycle_time.tavg[CXT_WL]);
+
+				if (wl_slot_real > wl_slot_set)
+					diff_t = wl_slot_real - wl_slot_set;
+				else
+					diff_t = wl_slot_set - wl_slot_real;
+			}
+			_chk_btc_err(rtwdev, BTC_DCNT_WL_SLOT_DRIFT, diff_t);
+
+			/* Check diff time between real BT slot and EBT/E5G slot */
+			bt_slot_set = btc->bt_req_len;
+			bt_slot_real = le16_to_cpu(pcysta->v5.cycle_time.tavg[CXT_BT]);
+			diff_t = 0;
+			if (dm->tdma_now.type == CXTDMA_OFF &&
+			    dm->tdma_now.ext_ctrl == CXECTL_EXT &&
+			    bt_slot_set != 0) {
+				if (bt_slot_set > bt_slot_real)
+					diff_t = bt_slot_set - bt_slot_real;
+				else
+					diff_t = bt_slot_real - bt_slot_set;
+			}
+
+			_chk_btc_err(rtwdev, BTC_DCNT_BT_SLOT_DRIFT, diff_t);
+			_chk_btc_err(rtwdev, BTC_DCNT_E2G_HANG,
+				     le16_to_cpu(pcysta->v5.slot_cnt[CXST_E2G]));
+			_chk_btc_err(rtwdev, BTC_DCNT_W1_HANG,
+				     le16_to_cpu(pcysta->v5.slot_cnt[CXST_W1]));
+			_chk_btc_err(rtwdev, BTC_DCNT_B1_HANG,
+				     le16_to_cpu(pcysta->v5.slot_cnt[CXST_B1]));
+			_chk_btc_err(rtwdev, BTC_DCNT_CYCLE_HANG,
+				     le16_to_cpu(pcysta->v5.cycles));
 		} else {
 			goto err;
 		}
@@ -5039,6 +5091,7 @@ void _run_coex(struct rtw89_dev *rtwdev, enum btc_reason_and_action reason)
 	}
 
 	dm->cnt_dm[BTC_DCNT_RUN]++;
+	dm->fddt_train = BTC_FDDT_DISABLE;
 
 	if (btc->ctrl.always_freerun) {
 		_action_freerun(rtwdev);
@@ -6727,6 +6780,10 @@ static void _show_error(struct rtw89_dev *rtwdev, struct seq_file *m)
 		pcysta->v4 = pfwinfo->rpt_fbtc_cysta.finfo.v4;
 		except_cnt = pcysta->v4.except_cnt;
 		exception_map = le32_to_cpu(pcysta->v4.except_map);
+	} else if (ver->fcxcysta == 5) {
+		pcysta->v5 = pfwinfo->rpt_fbtc_cysta.finfo.v5;
+		except_cnt = pcysta->v5.except_cnt;
+		exception_map = le32_to_cpu(pcysta->v5.except_map);
 	} else {
 		return;
 	}
@@ -7215,6 +7272,137 @@ static void _show_fbtc_cysta_v4(struct rtw89_dev *rtwdev, struct seq_file *m)
 	}
 }
 
+static void _show_fbtc_cysta_v5(struct rtw89_dev *rtwdev, struct seq_file *m)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_bt_a2dp_desc *a2dp = &btc->cx.bt.link_info.a2dp_desc;
+	struct rtw89_btc_btf_fwinfo *pfwinfo = &btc->fwinfo;
+	struct rtw89_btc_dm *dm = &btc->dm;
+	struct rtw89_btc_fbtc_a2dp_trx_stat_v4 *a2dp_trx;
+	struct rtw89_btc_fbtc_cysta_v5 *pcysta;
+	struct rtw89_btc_rpt_cmn_info *pcinfo;
+	u8 i, cnt = 0, slot_pair, divide_cnt;
+	u16 cycle, c_begin, c_end, store_index;
+
+	pcinfo = &pfwinfo->rpt_fbtc_cysta.cinfo;
+	if (!pcinfo->valid)
+		return;
+
+	pcysta = &pfwinfo->rpt_fbtc_cysta.finfo.v5;
+	seq_printf(m,
+		   " %-15s : cycle:%d, bcn[all:%d/all_ok:%d/bt:%d/bt_ok:%d]",
+		   "[cycle_cnt]",
+		   le16_to_cpu(pcysta->cycles),
+		   le16_to_cpu(pcysta->bcn_cnt[CXBCN_ALL]),
+		   le16_to_cpu(pcysta->bcn_cnt[CXBCN_ALL_OK]),
+		   le16_to_cpu(pcysta->bcn_cnt[CXBCN_BT_SLOT]),
+		   le16_to_cpu(pcysta->bcn_cnt[CXBCN_BT_OK]));
+
+	for (i = 0; i < CXST_MAX; i++) {
+		if (!le16_to_cpu(pcysta->slot_cnt[i]))
+			continue;
+
+		seq_printf(m, ", %s:%d", id_to_slot(i),
+			   le16_to_cpu(pcysta->slot_cnt[i]));
+	}
+
+	if (dm->tdma_now.rxflctrl)
+		seq_printf(m, ", leak_rx:%d",
+			   le32_to_cpu(pcysta->leak_slot.cnt_rximr));
+
+	if (pcysta->collision_cnt)
+		seq_printf(m, ", collision:%d", pcysta->collision_cnt);
+
+	if (le16_to_cpu(pcysta->skip_cnt))
+		seq_printf(m, ", skip:%d",
+			   le16_to_cpu(pcysta->skip_cnt));
+
+	seq_puts(m, "\n");
+
+	seq_printf(m, " %-15s : avg_t[wl:%d/bt:%d/lk:%d.%03d]",
+		   "[cycle_time]",
+		   le16_to_cpu(pcysta->cycle_time.tavg[CXT_WL]),
+		   le16_to_cpu(pcysta->cycle_time.tavg[CXT_BT]),
+		   le16_to_cpu(pcysta->leak_slot.tavg) / 1000,
+		   le16_to_cpu(pcysta->leak_slot.tavg) % 1000);
+	seq_printf(m,
+		   ", max_t[wl:%d/bt:%d/lk:%d.%03d]\n",
+		   le16_to_cpu(pcysta->cycle_time.tmax[CXT_WL]),
+		   le16_to_cpu(pcysta->cycle_time.tmax[CXT_BT]),
+		   le16_to_cpu(pcysta->leak_slot.tmax) / 1000,
+		   le16_to_cpu(pcysta->leak_slot.tmax) % 1000);
+
+	cycle = le16_to_cpu(pcysta->cycles);
+	if (cycle <= 1)
+		return;
+
+	/* 1 cycle record 1 wl-slot and 1 bt-slot */
+	slot_pair = BTC_CYCLE_SLOT_MAX / 2;
+
+	if (cycle <= slot_pair)
+		c_begin = 1;
+	else
+		c_begin = cycle - slot_pair + 1;
+
+	c_end = cycle;
+
+	if (a2dp->exist)
+		divide_cnt = 3;
+	else
+		divide_cnt = BTC_CYCLE_SLOT_MAX / 4;
+
+	if (c_begin > c_end)
+		return;
+
+	for (cycle = c_begin; cycle <= c_end; cycle++) {
+		cnt++;
+		store_index = ((cycle - 1) % slot_pair) * 2;
+
+		if (cnt % divide_cnt == 1)
+			seq_printf(m, " %-15s : ", "[cycle_step]");
+
+		seq_printf(m, "->b%02d",
+			   le16_to_cpu(pcysta->slot_step_time[store_index]));
+		if (a2dp->exist) {
+			a2dp_trx = &pcysta->a2dp_trx[store_index];
+			seq_printf(m, "(%d/%d/%dM/%d/%d/%d)",
+				   a2dp_trx->empty_cnt,
+				   a2dp_trx->retry_cnt,
+				   a2dp_trx->tx_rate ? 3 : 2,
+				   a2dp_trx->tx_cnt,
+				   a2dp_trx->ack_cnt,
+				   a2dp_trx->nack_cnt);
+		}
+		seq_printf(m, "->w%02d",
+			   le16_to_cpu(pcysta->slot_step_time[store_index + 1]));
+		if (a2dp->exist) {
+			a2dp_trx = &pcysta->a2dp_trx[store_index + 1];
+			seq_printf(m, "(%d/%d/%dM/%d/%d/%d)",
+				   a2dp_trx->empty_cnt,
+				   a2dp_trx->retry_cnt,
+				   a2dp_trx->tx_rate ? 3 : 2,
+				   a2dp_trx->tx_cnt,
+				   a2dp_trx->ack_cnt,
+				   a2dp_trx->nack_cnt);
+		}
+		if (cnt % divide_cnt == 0 || cnt == c_end)
+			seq_puts(m, "\n");
+	}
+
+	if (a2dp->exist) {
+		seq_printf(m, " %-15s : a2dp_ept:%d, a2dp_late:%d",
+			   "[a2dp_t_sta]",
+			   le16_to_cpu(pcysta->a2dp_ept.cnt),
+			   le16_to_cpu(pcysta->a2dp_ept.cnt_timeout));
+
+		seq_printf(m, ", avg_t:%d, max_t:%d",
+			   le16_to_cpu(pcysta->a2dp_ept.tavg),
+			   le16_to_cpu(pcysta->a2dp_ept.tmax));
+
+		seq_puts(m, "\n");
+	}
+}
+
 static void _show_fbtc_nullsta(struct rtw89_dev *rtwdev, struct seq_file *m)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
@@ -7419,6 +7607,8 @@ static void _show_fw_dm_msg(struct rtw89_dev *rtwdev, struct seq_file *m)
 		_show_fbtc_cysta_v3(rtwdev, m);
 	else if (ver->fcxcysta == 4)
 		_show_fbtc_cysta_v4(rtwdev, m);
+	else if (ver->fcxcysta == 5)
+		_show_fbtc_cysta_v5(rtwdev, m);
 
 	_show_fbtc_nullsta(rtwdev, m);
 
diff --git a/drivers/net/wireless/realtek/rtw89/coex.h b/drivers/net/wireless/realtek/rtw89/coex.h
index 401fb55df82b..38cc53a505c3 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.h
+++ b/drivers/net/wireless/realtek/rtw89/coex.h
@@ -66,6 +66,11 @@ enum btc_rssi_st {
 	BTC_RSSI_ST_MAX
 };
 
+enum btc_fddt_en {
+	BTC_FDDT_DISABLE,
+	BTC_FDDT_ENABLE,
+};
+
 #define	BTC_RSSI_HIGH(_rssi_) \
 	({typeof(_rssi_) __rssi = (_rssi_); \
 	  ((__rssi == BTC_RSSI_ST_HIGH || \
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 02fb3c73b137..e447bfec508b 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1894,6 +1894,11 @@ struct rtw89_btc_fbtc_cycle_time_info {
 	__le16 tmaxdiff[CXT_MAX]; /* max wl-wl bt-bt cycle diff time */
 } __packed;
 
+struct rtw89_btc_fbtc_cycle_time_info_v5 {
+	__le16 tavg[CXT_MAX]; /* avg wl/bt cycle time */
+	__le16 tmax[CXT_MAX]; /* max wl/bt cycle time */
+} __packed;
+
 struct rtw89_btc_fbtc_a2dp_trx_stat {
 	u8 empty_cnt;
 	u8 retry_cnt;
@@ -1950,6 +1955,21 @@ struct rtw89_btc_fbtc_cycle_fddt_info {
 #define RTW89_BTC_FDDT_CELL_TRAIN_STATE GENMASK(3, 0)
 #define RTW89_BTC_FDDT_CELL_TRAIN_PHASE GENMASK(7, 4)
 
+struct rtw89_btc_fbtc_cycle_fddt_info_v5 {
+	__le16 train_cycle;
+	__le16 tp;
+
+	s8 tx_power; /* absolute Tx power (dBm), 0xff-> no BTC control */
+	s8 bt_tx_power; /* decrease Tx power (dB) */
+	s8 bt_rx_gain;  /* LNA constrain level */
+	u8 no_empty_cnt;
+
+	u8 rssi; /* [7:4] -> bt_rssi_level, [3:0]-> wl_rssi_level */
+	u8 cn; /* condition_num */
+	u8 train_status; /* [7:4]-> train-state, [3:0]-> train-phase */
+	u8 train_result; /* refer to enum btc_fddt_check_map */
+} __packed;
+
 struct rtw89_btc_fbtc_fddt_cell_status {
 	s8 wl_tx_pwr;
 	s8 bt_tx_pwr;
@@ -1957,6 +1977,12 @@ struct rtw89_btc_fbtc_fddt_cell_status {
 	u8 state_phase; /* [0:3] train state, [4:7] train phase */
 } __packed;
 
+struct rtw89_btc_fbtc_fddt_cell_status_v5 {
+	s8 wl_tx_pwr;
+	s8 bt_tx_pwr;
+	s8 bt_rx_gain;
+} __packed;
+
 struct rtw89_btc_fbtc_cysta_v3 { /* statistics for cycles */
 	u8 fver;
 	u8 rsvd;
@@ -2002,10 +2028,35 @@ struct rtw89_btc_fbtc_cysta_v4 { /* statistics for cycles */
 	__le32 except_map;
 } __packed;
 
+struct rtw89_btc_fbtc_cysta_v5 { /* statistics for cycles */
+	u8 fver;
+	u8 rsvd;
+	u8 collision_cnt; /* counter for event/timer occur at the same time */
+	u8 except_cnt;
+	u8 wl_rx_err_ratio[BTC_CYCLE_SLOT_MAX];
+
+	__le16 skip_cnt;
+	__le16 cycles; /* total cycle number */
+
+	__le16 slot_step_time[BTC_CYCLE_SLOT_MAX]; /* record the wl/bt slot time */
+	__le16 slot_cnt[CXST_MAX]; /* slot count */
+	__le16 bcn_cnt[CXBCN_MAX];
+	struct rtw89_btc_fbtc_cycle_time_info_v5 cycle_time;
+	struct rtw89_btc_fbtc_cycle_leak_info leak_slot;
+	struct rtw89_btc_fbtc_cycle_a2dp_empty_info a2dp_ept;
+	struct rtw89_btc_fbtc_a2dp_trx_stat_v4 a2dp_trx[BTC_CYCLE_SLOT_MAX];
+	struct rtw89_btc_fbtc_cycle_fddt_info_v5 fddt_trx[BTC_CYCLE_SLOT_MAX];
+	struct rtw89_btc_fbtc_fddt_cell_status_v5 fddt_cells[FDD_TRAIN_WL_DIRECTION]
+							    [FDD_TRAIN_WL_RSSI_LEVEL]
+							    [FDD_TRAIN_BT_RSSI_LEVEL];
+	__le32 except_map;
+} __packed;
+
 union rtw89_btc_fbtc_cysta_info {
 	struct rtw89_btc_fbtc_cysta_v2 v2;
 	struct rtw89_btc_fbtc_cysta_v3 v3;
 	struct rtw89_btc_fbtc_cysta_v4 v4;
+	struct rtw89_btc_fbtc_cysta_v5 v5;
 };
 
 struct rtw89_btc_fbtc_cynullsta_v1 { /* cycle null statistics */
@@ -2123,6 +2174,7 @@ struct rtw89_btc_dm {
 	u32 wl_only: 1;
 	u32 wl_fw_cx_offload: 1;
 	u32 freerun: 1;
+	u32 fddt_train: 1;
 	u32 wl_ps_ctrl: 2;
 	u32 wl_mimo_ps: 1;
 	u32 leak_ap: 1;
-- 
cgit v1.2.3-70-g09d2


From 48181d285623198c33bb9698992502687b258efa Mon Sep 17 00:00:00 2001
From: Dongliang Mu <dzm91@hust.edu.cn>
Date: Thu, 9 Mar 2023 10:16:36 +0800
Subject: wifi: rtw88: fix memory leak in rtw_usb_probe()

drivers/net/wireless/realtek/rtw88/usb.c:876 rtw_usb_probe()
warn: 'hw' from ieee80211_alloc_hw() not released on lines: 811

Fix this by modifying return to a goto statement.

Signed-off-by: Dongliang Mu <dzm91@hust.edu.cn>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230309021636.528601-1-dzm91@hust.edu.cn
---
 drivers/net/wireless/realtek/rtw88/usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index 2a8336b1847a..68e1b782d199 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -808,7 +808,7 @@ int rtw_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 
 	ret = rtw_usb_alloc_rx_bufs(rtwusb);
 	if (ret)
-		return ret;
+		goto err_release_hw;
 
 	ret = rtw_core_init(rtwdev);
 	if (ret)
-- 
cgit v1.2.3-70-g09d2


From 0da40e018fd034d87c9460123fa7f897b69fdee7 Mon Sep 17 00:00:00 2001
From: Jisoo Jang <jisoo.jang@yonsei.ac.kr>
Date: Thu, 9 Mar 2023 19:44:57 +0900
Subject: wifi: brcmfmac: slab-out-of-bounds read in brcmf_get_assoc_ies()

Fix a slab-out-of-bounds read that occurs in kmemdup() called from
brcmf_get_assoc_ies().
The bug could occur when assoc_info->req_len, data from a URB provided
by a USB device, is bigger than the size of buffer which is defined as
WL_EXTRA_BUF_MAX.

Add the size check for req_len/resp_len of assoc_info.

Found by a modified version of syzkaller.

[   46.592467][    T7] ==================================================================
[   46.594687][    T7] BUG: KASAN: slab-out-of-bounds in kmemdup+0x3e/0x50
[   46.596572][    T7] Read of size 3014656 at addr ffff888019442000 by task kworker/0:1/7
[   46.598575][    T7]
[   46.599157][    T7] CPU: 0 PID: 7 Comm: kworker/0:1 Tainted: G           O      5.14.0+ #145
[   46.601333][    T7] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
[   46.604360][    T7] Workqueue: events brcmf_fweh_event_worker
[   46.605943][    T7] Call Trace:
[   46.606584][    T7]  dump_stack_lvl+0x8e/0xd1
[   46.607446][    T7]  print_address_description.constprop.0.cold+0x93/0x334
[   46.608610][    T7]  ? kmemdup+0x3e/0x50
[   46.609341][    T7]  kasan_report.cold+0x79/0xd5
[   46.610151][    T7]  ? kmemdup+0x3e/0x50
[   46.610796][    T7]  kasan_check_range+0x14e/0x1b0
[   46.611691][    T7]  memcpy+0x20/0x60
[   46.612323][    T7]  kmemdup+0x3e/0x50
[   46.612987][    T7]  brcmf_get_assoc_ies+0x967/0xf60
[   46.613904][    T7]  ? brcmf_notify_vif_event+0x3d0/0x3d0
[   46.614831][    T7]  ? lock_chain_count+0x20/0x20
[   46.615683][    T7]  ? mark_lock.part.0+0xfc/0x2770
[   46.616552][    T7]  ? lock_chain_count+0x20/0x20
[   46.617409][    T7]  ? mark_lock.part.0+0xfc/0x2770
[   46.618244][    T7]  ? lock_chain_count+0x20/0x20
[   46.619024][    T7]  brcmf_bss_connect_done.constprop.0+0x241/0x2e0
[   46.620019][    T7]  ? brcmf_parse_configure_security.isra.0+0x2a0/0x2a0
[   46.620818][    T7]  ? __lock_acquire+0x181f/0x5790
[   46.621462][    T7]  brcmf_notify_connect_status+0x448/0x1950
[   46.622134][    T7]  ? rcu_read_lock_bh_held+0xb0/0xb0
[   46.622736][    T7]  ? brcmf_cfg80211_join_ibss+0x7b0/0x7b0
[   46.623390][    T7]  ? find_held_lock+0x2d/0x110
[   46.623962][    T7]  ? brcmf_fweh_event_worker+0x19f/0xc60
[   46.624603][    T7]  ? mark_held_locks+0x9f/0xe0
[   46.625145][    T7]  ? lockdep_hardirqs_on_prepare+0x3e0/0x3e0
[   46.625871][    T7]  ? brcmf_cfg80211_join_ibss+0x7b0/0x7b0
[   46.626545][    T7]  brcmf_fweh_call_event_handler.isra.0+0x90/0x100
[   46.627338][    T7]  brcmf_fweh_event_worker+0x557/0xc60
[   46.627962][    T7]  ? brcmf_fweh_call_event_handler.isra.0+0x100/0x100
[   46.628736][    T7]  ? rcu_read_lock_sched_held+0xa1/0xd0
[   46.629396][    T7]  ? rcu_read_lock_bh_held+0xb0/0xb0
[   46.629970][    T7]  ? lockdep_hardirqs_on_prepare+0x273/0x3e0
[   46.630649][    T7]  process_one_work+0x92b/0x1460
[   46.631205][    T7]  ? pwq_dec_nr_in_flight+0x330/0x330
[   46.631821][    T7]  ? rwlock_bug.part.0+0x90/0x90
[   46.632347][    T7]  worker_thread+0x95/0xe00
[   46.632832][    T7]  ? __kthread_parkme+0x115/0x1e0
[   46.633393][    T7]  ? process_one_work+0x1460/0x1460
[   46.633957][    T7]  kthread+0x3a1/0x480
[   46.634369][    T7]  ? set_kthread_struct+0x120/0x120
[   46.634933][    T7]  ret_from_fork+0x1f/0x30
[   46.635431][    T7]
[   46.635687][    T7] Allocated by task 7:
[   46.636151][    T7]  kasan_save_stack+0x1b/0x40
[   46.636628][    T7]  __kasan_kmalloc+0x7c/0x90
[   46.637108][    T7]  kmem_cache_alloc_trace+0x19e/0x330
[   46.637696][    T7]  brcmf_cfg80211_attach+0x4a0/0x4040
[   46.638275][    T7]  brcmf_attach+0x389/0xd40
[   46.638739][    T7]  brcmf_usb_probe+0x12de/0x1690
[   46.639279][    T7]  usb_probe_interface+0x2aa/0x760
[   46.639820][    T7]  really_probe+0x205/0xb70
[   46.640342][    T7]  __driver_probe_device+0x311/0x4b0
[   46.640876][    T7]  driver_probe_device+0x4e/0x150
[   46.641445][    T7]  __device_attach_driver+0x1cc/0x2a0
[   46.642000][    T7]  bus_for_each_drv+0x156/0x1d0
[   46.642543][    T7]  __device_attach+0x23f/0x3a0
[   46.643065][    T7]  bus_probe_device+0x1da/0x290
[   46.643644][    T7]  device_add+0xb7b/0x1eb0
[   46.644130][    T7]  usb_set_configuration+0xf59/0x16f0
[   46.644720][    T7]  usb_generic_driver_probe+0x82/0xa0
[   46.645295][    T7]  usb_probe_device+0xbb/0x250
[   46.645786][    T7]  really_probe+0x205/0xb70
[   46.646258][    T7]  __driver_probe_device+0x311/0x4b0
[   46.646804][    T7]  driver_probe_device+0x4e/0x150
[   46.647387][    T7]  __device_attach_driver+0x1cc/0x2a0
[   46.647926][    T7]  bus_for_each_drv+0x156/0x1d0
[   46.648454][    T7]  __device_attach+0x23f/0x3a0
[   46.648939][    T7]  bus_probe_device+0x1da/0x290
[   46.649478][    T7]  device_add+0xb7b/0x1eb0
[   46.649936][    T7]  usb_new_device.cold+0x49c/0x1029
[   46.650526][    T7]  hub_event+0x1c98/0x3950
[   46.650975][    T7]  process_one_work+0x92b/0x1460
[   46.651535][    T7]  worker_thread+0x95/0xe00
[   46.651991][    T7]  kthread+0x3a1/0x480
[   46.652413][    T7]  ret_from_fork+0x1f/0x30
[   46.652885][    T7]
[   46.653131][    T7] The buggy address belongs to the object at ffff888019442000
[   46.653131][    T7]  which belongs to the cache kmalloc-2k of size 2048
[   46.654669][    T7] The buggy address is located 0 bytes inside of
[   46.654669][    T7]  2048-byte region [ffff888019442000, ffff888019442800)
[   46.656137][    T7] The buggy address belongs to the page:
[   46.656720][    T7] page:ffffea0000651000 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x19440
[   46.657792][    T7] head:ffffea0000651000 order:3 compound_mapcount:0 compound_pincount:0
[   46.658673][    T7] flags: 0x100000000010200(slab|head|node=0|zone=1)
[   46.659422][    T7] raw: 0100000000010200 0000000000000000 dead000000000122 ffff888100042000
[   46.660363][    T7] raw: 0000000000000000 0000000000080008 00000001ffffffff 0000000000000000
[   46.661236][    T7] page dumped because: kasan: bad access detected
[   46.661956][    T7] page_owner tracks the page as allocated
[   46.662588][    T7] page last allocated via order 3, migratetype Unmovable, gfp_mask 0x52a20(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP), pid 7, ts 31136961085, free_ts 0
[   46.664271][    T7]  prep_new_page+0x1aa/0x240
[   46.664763][    T7]  get_page_from_freelist+0x159a/0x27c0
[   46.665340][    T7]  __alloc_pages+0x2da/0x6a0
[   46.665847][    T7]  alloc_pages+0xec/0x1e0
[   46.666308][    T7]  allocate_slab+0x380/0x4e0
[   46.666770][    T7]  ___slab_alloc+0x5bc/0x940
[   46.667264][    T7]  __slab_alloc+0x6d/0x80
[   46.667712][    T7]  kmem_cache_alloc_trace+0x30a/0x330
[   46.668299][    T7]  brcmf_usbdev_qinit.constprop.0+0x50/0x470
[   46.668885][    T7]  brcmf_usb_probe+0xc97/0x1690
[   46.669438][    T7]  usb_probe_interface+0x2aa/0x760
[   46.669988][    T7]  really_probe+0x205/0xb70
[   46.670487][    T7]  __driver_probe_device+0x311/0x4b0
[   46.671031][    T7]  driver_probe_device+0x4e/0x150
[   46.671604][    T7]  __device_attach_driver+0x1cc/0x2a0
[   46.672192][    T7]  bus_for_each_drv+0x156/0x1d0
[   46.672739][    T7] page_owner free stack trace missing
[   46.673335][    T7]
[   46.673620][    T7] Memory state around the buggy address:
[   46.674213][    T7]  ffff888019442700: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[   46.675083][    T7]  ffff888019442780: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[   46.675994][    T7] >ffff888019442800: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[   46.676875][    T7]                    ^
[   46.677323][    T7]  ffff888019442880: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[   46.678190][    T7]  ffff888019442900: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[   46.679052][    T7] ==================================================================
[   46.679945][    T7] Disabling lock debugging due to kernel taint
[   46.680725][    T7] Kernel panic - not syncing:

Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Jisoo Jang <jisoo.jang@yonsei.ac.kr>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230309104457.22628-1-jisoo.jang@yonsei.ac.kr
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 548799fefb4b..de8a2e27f49c 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6280,6 +6280,11 @@ static s32 brcmf_get_assoc_ies(struct brcmf_cfg80211_info *cfg,
 		(struct brcmf_cfg80211_assoc_ielen_le *)cfg->extra_buf;
 	req_len = le32_to_cpu(assoc_info->req_len);
 	resp_len = le32_to_cpu(assoc_info->resp_len);
+	if (req_len > WL_EXTRA_BUF_MAX || resp_len > WL_EXTRA_BUF_MAX) {
+		bphy_err(drvr, "invalid lengths in assoc info: req %u resp %u\n",
+			 req_len, resp_len);
+		return -EINVAL;
+	}
 	if (req_len) {
 		err = brcmf_fil_iovar_data_get(ifp, "assoc_req_ies",
 					       cfg->extra_buf,
-- 
cgit v1.2.3-70-g09d2


From 504ce971f260c178fa625f1278d9a762bc366504 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:07 -0800
Subject: ice: re-order ice_mbx_reset_snapshot function

A future change is going to refactor the VF mailbox overflow detection
logic, including modifying ice_mbx_reset_snapshot and its callers. To make
this change easier to review, first move the ice_mbx_reset_snapshot
function higher in the ice_vf_mbx.c file.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_vf_mbx.c | 48 ++++++++++++++---------------
 1 file changed, 24 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
index f56fa94ff3d0..2fe9a9504914 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
@@ -130,6 +130,30 @@ u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed)
  */
 #define ICE_IGNORE_MAX_MSG_CNT	0xFFFF
 
+/**
+ * ice_mbx_reset_snapshot - Reset mailbox snapshot structure
+ * @snap: pointer to mailbox snapshot structure in the ice_hw struct
+ *
+ * Reset the mailbox snapshot structure and clear VF counter array.
+ */
+static void ice_mbx_reset_snapshot(struct ice_mbx_snapshot *snap)
+{
+	u32 vfcntr_len;
+
+	if (!snap || !snap->mbx_vf.vf_cntr)
+		return;
+
+	/* Clear VF counters. */
+	vfcntr_len = snap->mbx_vf.vfcntr_len;
+	if (vfcntr_len)
+		memset(snap->mbx_vf.vf_cntr, 0,
+		       (vfcntr_len * sizeof(*snap->mbx_vf.vf_cntr)));
+
+	/* Reset mailbox snapshot for a new capture. */
+	memset(&snap->mbx_buf, 0, sizeof(snap->mbx_buf));
+	snap->mbx_buf.state = ICE_MAL_VF_DETECT_STATE_NEW_SNAPSHOT;
+}
+
 /**
  * ice_mbx_traverse - Pass through mailbox snapshot
  * @hw: pointer to the HW struct
@@ -201,30 +225,6 @@ ice_mbx_detect_malvf(struct ice_hw *hw, u16 vf_id,
 	return 0;
 }
 
-/**
- * ice_mbx_reset_snapshot - Reset mailbox snapshot structure
- * @snap: pointer to mailbox snapshot structure in the ice_hw struct
- *
- * Reset the mailbox snapshot structure and clear VF counter array.
- */
-static void ice_mbx_reset_snapshot(struct ice_mbx_snapshot *snap)
-{
-	u32 vfcntr_len;
-
-	if (!snap || !snap->mbx_vf.vf_cntr)
-		return;
-
-	/* Clear VF counters. */
-	vfcntr_len = snap->mbx_vf.vfcntr_len;
-	if (vfcntr_len)
-		memset(snap->mbx_vf.vf_cntr, 0,
-		       (vfcntr_len * sizeof(*snap->mbx_vf.vf_cntr)));
-
-	/* Reset mailbox snapshot for a new capture. */
-	memset(&snap->mbx_buf, 0, sizeof(snap->mbx_buf));
-	snap->mbx_buf.state = ICE_MAL_VF_DETECT_STATE_NEW_SNAPSHOT;
-}
-
 /**
  * ice_mbx_vf_state_handler - Handle states of the overflow algorithm
  * @hw: pointer to the HW struct
-- 
cgit v1.2.3-70-g09d2


From 28756d9ec93e6588b1c3a00cc9123e238a71c709 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:08 -0800
Subject: ice: convert ice_mbx_clear_malvf to void and use WARN

The ice_mbx_clear_malvf function checks for a few error conditions before
clearing the appropriate data. These error conditions are really warnings
that should never occur in a properly initialized driver. Every caller of
ice_mbx_clear_malvf just prints a dev_dbg message on failure which will
generally be ignored.

Convert this function to void and switch the error return values to
WARN_ON. This will make any potentially misconfiguration more visible and
makes future refactors that involve changing how we store the malicious VF
data easier.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c  |  6 ++----
 drivers/net/ethernet/intel/ice/ice_vf_lib.c | 12 ++++--------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.c | 16 +++++++---------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.h |  2 +-
 4 files changed, 14 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 96a64c25e2ef..7107c279752a 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -204,10 +204,8 @@ void ice_free_vfs(struct ice_pf *pf)
 		}
 
 		/* clear malicious info since the VF is getting released */
-		if (ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
-					ICE_MAX_SRIOV_VFS, vf->vf_id))
-			dev_dbg(dev, "failed to clear malicious VF state for VF %u\n",
-				vf->vf_id);
+		ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
+				    ICE_MAX_SRIOV_VFS, vf->vf_id);
 
 		mutex_unlock(&vf->cfg_lock);
 	}
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
index 0e57bd1b85fd..116b43588389 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
@@ -496,10 +496,8 @@ void ice_reset_all_vfs(struct ice_pf *pf)
 
 	/* clear all malicious info if the VFs are getting reset */
 	ice_for_each_vf(pf, bkt, vf)
-		if (ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
-					ICE_MAX_SRIOV_VFS, vf->vf_id))
-			dev_dbg(dev, "failed to clear malicious VF state for VF %u\n",
-				vf->vf_id);
+		ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
+				    ICE_MAX_SRIOV_VFS, vf->vf_id);
 
 	/* If VFs have been disabled, there is no need to reset */
 	if (test_and_set_bit(ICE_VF_DIS, pf->state)) {
@@ -705,10 +703,8 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 	ice_eswitch_replay_vf_mac_rule(vf);
 
 	/* if the VF has been reset allow it to come up again */
-	if (ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
-				ICE_MAX_SRIOV_VFS, vf->vf_id))
-		dev_dbg(dev, "failed to clear malicious VF state for VF %u\n",
-			vf->vf_id);
+	ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
+			    ICE_MAX_SRIOV_VFS, vf->vf_id);
 
 out_unlock:
 	if (flags & ICE_VF_RESET_LOCK)
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
index 2fe9a9504914..9f6acfeb0fc6 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
@@ -392,19 +392,19 @@ ice_mbx_report_malvf(struct ice_hw *hw, unsigned long *all_malvfs,
  * that the new VF loaded is not considered malicious before going
  * through the overflow detection algorithm.
  */
-int
+void
 ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, unsigned long *all_malvfs,
 		    u16 bitmap_len, u16 vf_id)
 {
-	if (!snap || !all_malvfs)
-		return -EINVAL;
+	if (WARN_ON(!snap || !all_malvfs))
+		return;
 
-	if (bitmap_len < snap->mbx_vf.vfcntr_len)
-		return -EINVAL;
+	if (WARN_ON(bitmap_len < snap->mbx_vf.vfcntr_len))
+		return;
 
 	/* Ensure VF ID value is not larger than bitmap or VF counter length */
-	if (vf_id >= bitmap_len || vf_id >= snap->mbx_vf.vfcntr_len)
-		return -EIO;
+	if (WARN_ON(vf_id >= bitmap_len || vf_id >= snap->mbx_vf.vfcntr_len))
+		return;
 
 	/* Clear VF ID bit in the bitmap tracking malicious VFs attached to PF */
 	clear_bit(vf_id, all_malvfs);
@@ -416,8 +416,6 @@ ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, unsigned long *all_malvfs,
 	 * values in the mailbox overflow detection algorithm.
 	 */
 	snap->mbx_vf.vf_cntr[vf_id] = 0;
-
-	return 0;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
index 582716e6d5f9..be593b951642 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
@@ -22,7 +22,7 @@ u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed);
 int
 ice_mbx_vf_state_handler(struct ice_hw *hw, struct ice_mbx_data *mbx_data,
 			 u16 vf_id, bool *is_mal_vf);
-int
+void
 ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, unsigned long *all_malvfs,
 		    u16 bitmap_len, u16 vf_id);
 int ice_mbx_init_snapshot(struct ice_hw *hw, u16 vf_count);
-- 
cgit v1.2.3-70-g09d2


From e4eaf8938852d092fa447b32adb8ec233621d86a Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:09 -0800
Subject: ice: track malicious VFs in new ice_mbx_vf_info structure

Currently the PF tracks malicious VFs in a malvfs bitmap which is used by
the ice_mbx_clear_malvf and ice_mbx_report_malvf functions. This bitmap is
used to ensure that we only report a VF as malicious once rather than
continuously spamming the event log.

This mechanism of storage for the malicious indication works well enough
for SR-IOV. However, it will not work with Scalable IOV. This is because
Scalable IOV VFs can be allocated dynamically and might change VF ID when
their underlying VSI changes.

To support this, the mailbox overflow logic will need to be refactored.
First, introduce a new ice_mbx_vf_info structure which will be used to
store data about a VF. Embed this structure in the struct ice_vf, and
ensure it gets initialized when a new VF is created.

For now this only stores the malicious indicator bit. Pass a pointer to the
VF's mbx_info structure instead of using a bitmap to keep track of these
bits.

A future change will extend this structure and the rest of the logic
associated with the overflow detection.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c  |  7 ++-
 drivers/net/ethernet/intel/ice/ice_type.h   |  7 +++
 drivers/net/ethernet/intel/ice/ice_vf_lib.c | 10 ++--
 drivers/net/ethernet/intel/ice/ice_vf_lib.h |  2 +-
 drivers/net/ethernet/intel/ice/ice_vf_mbx.c | 71 +++++++++++++----------------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.h |  9 ++--
 6 files changed, 53 insertions(+), 53 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 7107c279752a..44b94276df91 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -204,8 +204,8 @@ void ice_free_vfs(struct ice_pf *pf)
 		}
 
 		/* clear malicious info since the VF is getting released */
-		ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
-				    ICE_MAX_SRIOV_VFS, vf->vf_id);
+		ice_mbx_clear_malvf(&hw->mbx_snapshot, vf->vf_id,
+				    &vf->mbx_info);
 
 		mutex_unlock(&vf->cfg_lock);
 	}
@@ -1828,8 +1828,7 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 		/* if the VF is malicious and we haven't let the user
 		 * know about it, then let them know now
 		 */
-		status = ice_mbx_report_malvf(&pf->hw, pf->vfs.malvfs,
-					      ICE_MAX_SRIOV_VFS, vf_id,
+		status = ice_mbx_report_malvf(&pf->hw, &vf->mbx_info,
 					      &report_vf);
 		if (status)
 			dev_dbg(dev, "Error reporting malicious VF\n");
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index e3f622cad425..d243a0c59ea4 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -794,6 +794,13 @@ struct ice_mbx_vf_counter {
 	u32 vfcntr_len;
 };
 
+/* Structure used to track a single VF's messages on the mailbox:
+ * 1. malicious: whether this VF has been detected as malicious before
+ */
+struct ice_mbx_vf_info {
+	u8 malicious : 1;
+};
+
 /* Structure to hold data relevant to the captured static snapshot
  * of the PF-VF mailbox.
  */
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
index 116b43588389..69e89e960950 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
@@ -496,8 +496,8 @@ void ice_reset_all_vfs(struct ice_pf *pf)
 
 	/* clear all malicious info if the VFs are getting reset */
 	ice_for_each_vf(pf, bkt, vf)
-		ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
-				    ICE_MAX_SRIOV_VFS, vf->vf_id);
+		ice_mbx_clear_malvf(&hw->mbx_snapshot, vf->vf_id,
+				    &vf->mbx_info);
 
 	/* If VFs have been disabled, there is no need to reset */
 	if (test_and_set_bit(ICE_VF_DIS, pf->state)) {
@@ -703,8 +703,7 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 	ice_eswitch_replay_vf_mac_rule(vf);
 
 	/* if the VF has been reset allow it to come up again */
-	ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->vfs.malvfs,
-			    ICE_MAX_SRIOV_VFS, vf->vf_id);
+	ice_mbx_clear_malvf(&hw->mbx_snapshot, vf->vf_id, &vf->mbx_info);
 
 out_unlock:
 	if (flags & ICE_VF_RESET_LOCK)
@@ -760,6 +759,9 @@ void ice_initialize_vf_entry(struct ice_vf *vf)
 	ice_vf_ctrl_invalidate_vsi(vf);
 	ice_vf_fdir_init(vf);
 
+	/* Initialize mailbox info for this VF */
+	ice_mbx_init_vf_info(&pf->hw, &vf->mbx_info);
+
 	mutex_init(&vf->cfg_lock);
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.h b/drivers/net/ethernet/intel/ice/ice_vf_lib.h
index ef30f05b5d02..e3cda6fb71ab 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.h
@@ -74,7 +74,6 @@ struct ice_vfs {
 	u16 num_qps_per;		/* number of queue pairs per VF */
 	u16 num_msix_per;		/* number of MSI-X vectors per VF */
 	unsigned long last_printed_mdd_jiffies;	/* MDD message rate limit */
-	DECLARE_BITMAP(malvfs, ICE_MAX_SRIOV_VFS); /* malicious VF indicator */
 };
 
 /* VF information structure */
@@ -105,6 +104,7 @@ struct ice_vf {
 	DECLARE_BITMAP(rxq_ena, ICE_MAX_RSS_QS_PER_VF);
 	struct ice_vlan port_vlan_info;	/* Port VLAN ID, QoS, and TPID */
 	struct virtchnl_vlan_caps vlan_v2_caps;
+	struct ice_mbx_vf_info mbx_info;
 	u8 pf_set_mac:1;		/* VF MAC address set by VMM admin */
 	u8 trusted:1;
 	u8 spoofchk:1;
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
index 9f6acfeb0fc6..2e769bd0bf7e 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
@@ -345,35 +345,23 @@ ice_mbx_vf_state_handler(struct ice_hw *hw,
 /**
  * ice_mbx_report_malvf - Track and note malicious VF
  * @hw: pointer to the HW struct
- * @all_malvfs: all malicious VFs tracked by PF
- * @bitmap_len: length of bitmap in bits
- * @vf_id: relative virtual function ID of the malicious VF
+ * @vf_info: the mailbox tracking info structure for a VF
  * @report_malvf: boolean to indicate if malicious VF must be reported
  *
- * This function will update a bitmap that keeps track of the malicious
- * VFs attached to the PF. A malicious VF must be reported only once if
- * discovered between VF resets or loading so the function checks
- * the input vf_id against the bitmap to verify if the VF has been
- * detected in any previous mailbox iterations.
+ * This function updates the malicious indicator bit in the VF mailbox
+ * tracking structure. A malicious VF must be reported only once if discovered
+ * between VF resets or loading so the function first checks if the VF has
+ * already been detected in any previous mailbox iterations.
  */
 int
-ice_mbx_report_malvf(struct ice_hw *hw, unsigned long *all_malvfs,
-		     u16 bitmap_len, u16 vf_id, bool *report_malvf)
+ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
+		     bool *report_malvf)
 {
-	if (!all_malvfs || !report_malvf)
-		return -EINVAL;
-
-	*report_malvf = false;
-
-	if (bitmap_len < hw->mbx_snapshot.mbx_vf.vfcntr_len)
+	if (!report_malvf)
 		return -EINVAL;
 
-	if (vf_id >= bitmap_len)
-		return -EIO;
-
-	/* If the vf_id is found in the bitmap set bit and boolean to true */
-	if (!test_and_set_bit(vf_id, all_malvfs))
-		*report_malvf = true;
+	*report_malvf = !vf_info->malicious;
+	vf_info->malicious = 1;
 
 	return 0;
 }
@@ -381,33 +369,24 @@ ice_mbx_report_malvf(struct ice_hw *hw, unsigned long *all_malvfs,
 /**
  * ice_mbx_clear_malvf - Clear VF bitmap and counter for VF ID
  * @snap: pointer to the mailbox snapshot structure
- * @all_malvfs: all malicious VFs tracked by PF
- * @bitmap_len: length of bitmap in bits
  * @vf_id: relative virtual function ID of the malicious VF
+ * @vf_info: mailbox tracking structure for this VF
  *
- * In case of a VF reset, this function can be called to clear
- * the bit corresponding to the VF ID in the bitmap tracking all
- * malicious VFs attached to the PF. The function also clears the
- * VF counter array at the index of the VF ID. This is to ensure
- * that the new VF loaded is not considered malicious before going
- * through the overflow detection algorithm.
- */
+* In case of a VF reset, this function shall be called to clear the VF's
+* current mailbox tracking state.
+*/
 void
-ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, unsigned long *all_malvfs,
-		    u16 bitmap_len, u16 vf_id)
+ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, u16 vf_id,
+		    struct ice_mbx_vf_info *vf_info)
 {
-	if (WARN_ON(!snap || !all_malvfs))
-		return;
-
-	if (WARN_ON(bitmap_len < snap->mbx_vf.vfcntr_len))
+	if (WARN_ON(!snap))
 		return;
 
 	/* Ensure VF ID value is not larger than bitmap or VF counter length */
-	if (WARN_ON(vf_id >= bitmap_len || vf_id >= snap->mbx_vf.vfcntr_len))
+	if (WARN_ON(vf_id >= snap->mbx_vf.vfcntr_len))
 		return;
 
-	/* Clear VF ID bit in the bitmap tracking malicious VFs attached to PF */
-	clear_bit(vf_id, all_malvfs);
+	vf_info->malicious = 0;
 
 	/* Clear the VF counter in the mailbox snapshot structure for that VF ID.
 	 * This is to ensure that if a VF is unloaded and a new one brought back
@@ -418,6 +397,18 @@ ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, unsigned long *all_malvfs,
 	snap->mbx_vf.vf_cntr[vf_id] = 0;
 }
 
+/**
+ * ice_mbx_init_vf_info - Initialize a new VF mailbox tracking info
+ * @hw: pointer to the hardware structure
+ * @vf_info: the mailbox tracking info structure for a VF
+ *
+ * Initialize a VF mailbox tracking info structure.
+ */
+void ice_mbx_init_vf_info(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info)
+{
+	vf_info->malicious = 0;
+}
+
 /**
  * ice_mbx_init_snapshot - Initialize mailbox snapshot structure
  * @hw: pointer to the hardware structure
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
index be593b951642..2613cba61ac7 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
@@ -23,13 +23,14 @@ int
 ice_mbx_vf_state_handler(struct ice_hw *hw, struct ice_mbx_data *mbx_data,
 			 u16 vf_id, bool *is_mal_vf);
 void
-ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, unsigned long *all_malvfs,
-		    u16 bitmap_len, u16 vf_id);
+ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, u16 vf_id,
+		    struct ice_mbx_vf_info *vf_info);
+void ice_mbx_init_vf_info(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info);
 int ice_mbx_init_snapshot(struct ice_hw *hw, u16 vf_count);
 void ice_mbx_deinit_snapshot(struct ice_hw *hw);
 int
-ice_mbx_report_malvf(struct ice_hw *hw, unsigned long *all_malvfs,
-		     u16 bitmap_len, u16 vf_id, bool *report_malvf);
+ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
+		     bool *report_malvf);
 #else /* CONFIG_PCI_IOV */
 static inline int
 ice_aq_send_msg_to_vf(struct ice_hw __always_unused *hw,
-- 
cgit v1.2.3-70-g09d2


From 8cd8a6b17d275a45e3722d0215f6115b687c8c3e Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:10 -0800
Subject: ice: move VF overflow message count into struct ice_mbx_vf_info

The ice driver has some logic in ice_vf_mbx.c used to detect potentially
malicious VF behavior with regards to overflowing the PF mailbox. This
logic currently stores message counts in struct ice_mbx_vf_counter.vf_cntr
as an array. This array is allocated during initialization with
ice_mbx_init_snapshot.

This logic makes sense for SR-IOV where all VFs are allocated at once up
front. However, in the future with Scalable IOV this logic will not work.
VFs can be added and removed dynamically. We could try to keep the vf_cntr
array for the maximum possible number of VFs, but this is a waste of
memory.

Use the recently introduced struct ice_mbx_vf_info structure to store the
message count. Pass a pointer to the mbx_info for a VF instead of using its
VF ID. Replace the array of VF message counts with a linked list that
tracks all currently active mailbox tracking info structures.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c  |   9 +-
 drivers/net/ethernet/intel/ice/ice_type.h   |  18 +--
 drivers/net/ethernet/intel/ice/ice_vf_lib.c |   7 +-
 drivers/net/ethernet/intel/ice/ice_vf_mbx.c | 167 +++++++++-------------------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.h |   8 +-
 5 files changed, 69 insertions(+), 140 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 44b94276df91..8820f269bfdf 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -204,8 +204,7 @@ void ice_free_vfs(struct ice_pf *pf)
 		}
 
 		/* clear malicious info since the VF is getting released */
-		ice_mbx_clear_malvf(&hw->mbx_snapshot, vf->vf_id,
-				    &vf->mbx_info);
+		list_del(&vf->mbx_info.list_entry);
 
 		mutex_unlock(&vf->cfg_lock);
 	}
@@ -1025,9 +1024,7 @@ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs)
 		return -EBUSY;
 	}
 
-	err = ice_mbx_init_snapshot(&pf->hw, num_vfs);
-	if (err)
-		return err;
+	ice_mbx_init_snapshot(&pf->hw);
 
 	err = ice_pci_sriov_ena(pf, num_vfs);
 	if (err) {
@@ -1818,7 +1815,7 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 	mbxdata.async_watermark_val = ICE_MBX_OVERFLOW_WATERMARK;
 
 	/* check to see if we have a malicious VF */
-	status = ice_mbx_vf_state_handler(&pf->hw, &mbxdata, vf_id, &malvf);
+	status = ice_mbx_vf_state_handler(&pf->hw, &mbxdata, &vf->mbx_info, &malvf);
 	if (status)
 		goto out_put_vf;
 
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index d243a0c59ea4..a09556e57803 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -784,20 +784,14 @@ struct ice_mbx_snap_buffer_data {
 	u16 max_num_msgs_mbx;
 };
 
-/* Structure to track messages sent by VFs on mailbox:
- * 1. vf_cntr: a counter array of VFs to track the number of
- * asynchronous messages sent by each VF
- * 2. vfcntr_len: number of entries in VF counter array
- */
-struct ice_mbx_vf_counter {
-	u32 *vf_cntr;
-	u32 vfcntr_len;
-};
-
 /* Structure used to track a single VF's messages on the mailbox:
- * 1. malicious: whether this VF has been detected as malicious before
+ * 1. list_entry: linked list entry node
+ * 2. msg_count: the number of asynchronous messages sent by this VF
+ * 3. malicious: whether this VF has been detected as malicious before
  */
 struct ice_mbx_vf_info {
+	struct list_head list_entry;
+	u32 msg_count;
 	u8 malicious : 1;
 };
 
@@ -806,7 +800,7 @@ struct ice_mbx_vf_info {
  */
 struct ice_mbx_snapshot {
 	struct ice_mbx_snap_buffer_data mbx_buf;
-	struct ice_mbx_vf_counter mbx_vf;
+	struct list_head mbx_vf;
 };
 
 /* Structure to hold data to be used for capturing or updating a
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
index 69e89e960950..89fd6982df09 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
@@ -496,8 +496,7 @@ void ice_reset_all_vfs(struct ice_pf *pf)
 
 	/* clear all malicious info if the VFs are getting reset */
 	ice_for_each_vf(pf, bkt, vf)
-		ice_mbx_clear_malvf(&hw->mbx_snapshot, vf->vf_id,
-				    &vf->mbx_info);
+		ice_mbx_clear_malvf(&vf->mbx_info);
 
 	/* If VFs have been disabled, there is no need to reset */
 	if (test_and_set_bit(ICE_VF_DIS, pf->state)) {
@@ -599,12 +598,10 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 	struct ice_pf *pf = vf->pf;
 	struct ice_vsi *vsi;
 	struct device *dev;
-	struct ice_hw *hw;
 	int err = 0;
 	bool rsd;
 
 	dev = ice_pf_to_dev(pf);
-	hw = &pf->hw;
 
 	if (flags & ICE_VF_RESET_NOTIFY)
 		ice_notify_vf_reset(vf);
@@ -703,7 +700,7 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 	ice_eswitch_replay_vf_mac_rule(vf);
 
 	/* if the VF has been reset allow it to come up again */
-	ice_mbx_clear_malvf(&hw->mbx_snapshot, vf->vf_id, &vf->mbx_info);
+	ice_mbx_clear_malvf(&vf->mbx_info);
 
 out_unlock:
 	if (flags & ICE_VF_RESET_LOCK)
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
index 2e769bd0bf7e..4bfed5fb3a88 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
@@ -93,36 +93,31 @@ u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed)
  *
  * 2. When the caller starts processing its mailbox queue in response to an
  * interrupt, the structure ice_mbx_snapshot is expected to be cleared before
- * the algorithm can be run for the first time for that interrupt. This can be
- * done via ice_mbx_reset_snapshot().
+ * the algorithm can be run for the first time for that interrupt. This
+ * requires calling ice_mbx_reset_snapshot() as well as calling
+ * ice_mbx_reset_vf_info() for each VF tracking structure.
  *
  * 3. For every message read by the caller from the MBX Queue, the caller must
  * call the detection algorithm's entry function ice_mbx_vf_state_handler().
  * Before every call to ice_mbx_vf_state_handler() the struct ice_mbx_data is
  * filled as it is required to be passed to the algorithm.
  *
- * 4. Every time a message is read from the MBX queue, a VFId is received which
- * is passed to the state handler. The boolean output is_malvf of the state
- * handler ice_mbx_vf_state_handler() serves as an indicator to the caller
- * whether this VF is malicious or not.
+ * 4. Every time a message is read from the MBX queue, a tracking structure
+ * for the VF must be passed to the state handler. The boolean output
+ * report_malvf from ice_mbx_vf_state_handler() serves as an indicator to the
+ * caller whether it must report this VF as malicious or not.
  *
  * 5. When a VF is identified to be malicious, the caller can send a message
- * to the system administrator. The caller can invoke ice_mbx_report_malvf()
- * to help determine if a malicious VF is to be reported or not. This function
- * requires the caller to maintain a global bitmap to track all malicious VFs
- * and pass that to ice_mbx_report_malvf() along with the VFID which was identified
- * to be malicious by ice_mbx_vf_state_handler().
+ * to the system administrator.
  *
- * 6. The global bitmap maintained by PF can be cleared completely if PF is in
- * reset or the bit corresponding to a VF can be cleared if that VF is in reset.
- * When a VF is shut down and brought back up, we assume that the new VF
- * brought up is not malicious and hence report it if found malicious.
+ * 6. The PF is responsible for maintaining the struct ice_mbx_vf_info
+ * structure for each VF. The PF should clear the VF tracking structure if the
+ * VF is reset. When a VF is shut down and brought back up, we will then
+ * assume that the new VF is not malicious and may report it again if we
+ * detect it again.
  *
  * 7. The function ice_mbx_reset_snapshot() is called to reset the information
  * in ice_mbx_snapshot for every new mailbox interrupt handled.
- *
- * 8. The memory allocated for variables in ice_mbx_snapshot is de-allocated
- * when driver is unloaded.
  */
 #define ICE_RQ_DATA_MASK(rq_data) ((rq_data) & PF_MBX_ARQH_ARQH_M)
 /* Using the highest value for an unsigned 16-bit value 0xFFFF to indicate that
@@ -132,26 +127,21 @@ u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed)
 
 /**
  * ice_mbx_reset_snapshot - Reset mailbox snapshot structure
- * @snap: pointer to mailbox snapshot structure in the ice_hw struct
- *
- * Reset the mailbox snapshot structure and clear VF counter array.
+ * @snap: pointer to the mailbox snapshot
  */
 static void ice_mbx_reset_snapshot(struct ice_mbx_snapshot *snap)
 {
-	u32 vfcntr_len;
-
-	if (!snap || !snap->mbx_vf.vf_cntr)
-		return;
+	struct ice_mbx_vf_info *vf_info;
 
-	/* Clear VF counters. */
-	vfcntr_len = snap->mbx_vf.vfcntr_len;
-	if (vfcntr_len)
-		memset(snap->mbx_vf.vf_cntr, 0,
-		       (vfcntr_len * sizeof(*snap->mbx_vf.vf_cntr)));
-
-	/* Reset mailbox snapshot for a new capture. */
+	/* Clear mbx_buf in the mailbox snaphot structure and setting the
+	 * mailbox snapshot state to a new capture.
+	 */
 	memset(&snap->mbx_buf, 0, sizeof(snap->mbx_buf));
 	snap->mbx_buf.state = ICE_MAL_VF_DETECT_STATE_NEW_SNAPSHOT;
+
+	/* Reset message counts for all VFs to zero */
+	list_for_each_entry(vf_info, &snap->mbx_vf, list_entry)
+		vf_info->msg_count = 0;
 }
 
 /**
@@ -195,7 +185,7 @@ ice_mbx_traverse(struct ice_hw *hw,
 /**
  * ice_mbx_detect_malvf - Detect malicious VF in snapshot
  * @hw: pointer to the HW struct
- * @vf_id: relative virtual function ID
+ * @vf_info: mailbox tracking structure for a VF
  * @new_state: new algorithm state
  * @is_malvf: boolean output to indicate if VF is malicious
  *
@@ -204,19 +194,14 @@ ice_mbx_traverse(struct ice_hw *hw,
  * the permissible number of messages to send.
  */
 static int
-ice_mbx_detect_malvf(struct ice_hw *hw, u16 vf_id,
+ice_mbx_detect_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
 		     enum ice_mbx_snapshot_state *new_state,
 		     bool *is_malvf)
 {
-	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
+	/* increment the message count for this VF */
+	vf_info->msg_count++;
 
-	if (vf_id >= snap->mbx_vf.vfcntr_len)
-		return -EIO;
-
-	/* increment the message count in the VF array */
-	snap->mbx_vf.vf_cntr[vf_id]++;
-
-	if (snap->mbx_vf.vf_cntr[vf_id] >= ICE_ASYNC_VF_MSG_THRESHOLD)
+	if (vf_info->msg_count >= ICE_ASYNC_VF_MSG_THRESHOLD)
 		*is_malvf = true;
 
 	/* continue to iterate through the mailbox snapshot */
@@ -229,7 +214,7 @@ ice_mbx_detect_malvf(struct ice_hw *hw, u16 vf_id,
  * ice_mbx_vf_state_handler - Handle states of the overflow algorithm
  * @hw: pointer to the HW struct
  * @mbx_data: pointer to structure containing mailbox data
- * @vf_id: relative virtual function (VF) ID
+ * @vf_info: mailbox tracking structure for the VF in question
  * @is_malvf: boolean output to indicate if VF is malicious
  *
  * The function serves as an entry point for the malicious VF
@@ -250,7 +235,8 @@ ice_mbx_detect_malvf(struct ice_hw *hw, u16 vf_id,
  */
 int
 ice_mbx_vf_state_handler(struct ice_hw *hw,
-			 struct ice_mbx_data *mbx_data, u16 vf_id,
+			 struct ice_mbx_data *mbx_data,
+			 struct ice_mbx_vf_info *vf_info,
 			 bool *is_malvf)
 {
 	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
@@ -315,7 +301,8 @@ ice_mbx_vf_state_handler(struct ice_hw *hw,
 		if (snap_buf->num_pending_arq >=
 		    mbx_data->async_watermark_val) {
 			new_state = ICE_MAL_VF_DETECT_STATE_DETECT;
-			status = ice_mbx_detect_malvf(hw, vf_id, &new_state, is_malvf);
+			status = ice_mbx_detect_malvf(hw, vf_info, &new_state,
+						      is_malvf);
 		} else {
 			new_state = ICE_MAL_VF_DETECT_STATE_TRAVERSE;
 			ice_mbx_traverse(hw, &new_state);
@@ -329,7 +316,8 @@ ice_mbx_vf_state_handler(struct ice_hw *hw,
 
 	case ICE_MAL_VF_DETECT_STATE_DETECT:
 		new_state = ICE_MAL_VF_DETECT_STATE_DETECT;
-		status = ice_mbx_detect_malvf(hw, vf_id, &new_state, is_malvf);
+		status = ice_mbx_detect_malvf(hw, vf_info, &new_state,
+					      is_malvf);
 		break;
 
 	default:
@@ -367,34 +355,16 @@ ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
 }
 
 /**
- * ice_mbx_clear_malvf - Clear VF bitmap and counter for VF ID
- * @snap: pointer to the mailbox snapshot structure
- * @vf_id: relative virtual function ID of the malicious VF
- * @vf_info: mailbox tracking structure for this VF
+ * ice_mbx_clear_malvf - Clear VF mailbox info
+ * @vf_info: the mailbox tracking structure for a VF
  *
-* In case of a VF reset, this function shall be called to clear the VF's
-* current mailbox tracking state.
-*/
-void
-ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, u16 vf_id,
-		    struct ice_mbx_vf_info *vf_info)
+ * In case of a VF reset, this function shall be called to clear the VF's
+ * current mailbox tracking state.
+ */
+void ice_mbx_clear_malvf(struct ice_mbx_vf_info *vf_info)
 {
-	if (WARN_ON(!snap))
-		return;
-
-	/* Ensure VF ID value is not larger than bitmap or VF counter length */
-	if (WARN_ON(vf_id >= snap->mbx_vf.vfcntr_len))
-		return;
-
 	vf_info->malicious = 0;
-
-	/* Clear the VF counter in the mailbox snapshot structure for that VF ID.
-	 * This is to ensure that if a VF is unloaded and a new one brought back
-	 * up with the same VF ID for a snapshot currently in traversal or detect
-	 * state the counter for that VF ID does not increment on top of existing
-	 * values in the mailbox overflow detection algorithm.
-	 */
-	snap->mbx_vf.vf_cntr[vf_id] = 0;
+	vf_info->msg_count = 0;
 }
 
 /**
@@ -402,55 +372,32 @@ ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, u16 vf_id,
  * @hw: pointer to the hardware structure
  * @vf_info: the mailbox tracking info structure for a VF
  *
- * Initialize a VF mailbox tracking info structure.
+ * Initialize a VF mailbox tracking info structure and insert it into the
+ * snapshot list.
+ *
+ * If you remove the VF, you must also delete the associated VF info structure
+ * from the linked list.
  */
 void ice_mbx_init_vf_info(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info)
 {
-	vf_info->malicious = 0;
+	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
+
+	ice_mbx_clear_malvf(vf_info);
+	list_add(&vf_info->list_entry, &snap->mbx_vf);
 }
 
 /**
- * ice_mbx_init_snapshot - Initialize mailbox snapshot structure
+ * ice_mbx_init_snapshot - Initialize mailbox snapshot data
  * @hw: pointer to the hardware structure
- * @vf_count: number of VFs allocated on a PF
- *
- * Clear the mailbox snapshot structure and allocate memory
- * for the VF counter array based on the number of VFs allocated
- * on that PF.
  *
- * Assumption: This function will assume ice_get_caps() has already been
- * called to ensure that the vf_count can be compared against the number
- * of VFs supported as defined in the functional capabilities of the device.
+ * Clear the mailbox snapshot structure and initialize the VF mailbox list.
  */
-int ice_mbx_init_snapshot(struct ice_hw *hw, u16 vf_count)
+void ice_mbx_init_snapshot(struct ice_hw *hw)
 {
 	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
 
-	/* Ensure that the number of VFs allocated is non-zero and
-	 * is not greater than the number of supported VFs defined in
-	 * the functional capabilities of the PF.
-	 */
-	if (!vf_count || vf_count > hw->func_caps.num_allocd_vfs)
-		return -EINVAL;
-
-	snap->mbx_vf.vf_cntr = devm_kcalloc(ice_hw_to_dev(hw), vf_count,
-					    sizeof(*snap->mbx_vf.vf_cntr),
-					    GFP_KERNEL);
-	if (!snap->mbx_vf.vf_cntr)
-		return -ENOMEM;
-
-	/* Setting the VF counter length to the number of allocated
-	 * VFs for given PF's functional capabilities.
-	 */
-	snap->mbx_vf.vfcntr_len = vf_count;
-
-	/* Clear mbx_buf in the mailbox snaphot structure and setting the
-	 * mailbox snapshot state to a new capture.
-	 */
-	memset(&snap->mbx_buf, 0, sizeof(snap->mbx_buf));
-	snap->mbx_buf.state = ICE_MAL_VF_DETECT_STATE_NEW_SNAPSHOT;
-
-	return 0;
+	INIT_LIST_HEAD(&snap->mbx_vf);
+	ice_mbx_reset_snapshot(snap);
 }
 
 /**
@@ -463,10 +410,6 @@ void ice_mbx_deinit_snapshot(struct ice_hw *hw)
 {
 	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
 
-	/* Free VF counter array and reset VF counter length */
-	devm_kfree(ice_hw_to_dev(hw), snap->mbx_vf.vf_cntr);
-	snap->mbx_vf.vfcntr_len = 0;
-
 	/* Clear mbx_buf in the mailbox snaphot structure */
 	memset(&snap->mbx_buf, 0, sizeof(snap->mbx_buf));
 }
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
index 2613cba61ac7..a6d42f467dc5 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
@@ -21,12 +21,10 @@ ice_aq_send_msg_to_vf(struct ice_hw *hw, u16 vfid, u32 v_opcode, u32 v_retval,
 u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed);
 int
 ice_mbx_vf_state_handler(struct ice_hw *hw, struct ice_mbx_data *mbx_data,
-			 u16 vf_id, bool *is_mal_vf);
-void
-ice_mbx_clear_malvf(struct ice_mbx_snapshot *snap, u16 vf_id,
-		    struct ice_mbx_vf_info *vf_info);
+			 struct ice_mbx_vf_info *vf_info, bool *is_mal_vf);
+void ice_mbx_clear_malvf(struct ice_mbx_vf_info *vf_info);
 void ice_mbx_init_vf_info(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info);
-int ice_mbx_init_snapshot(struct ice_hw *hw, u16 vf_count);
+void ice_mbx_init_snapshot(struct ice_hw *hw);
 void ice_mbx_deinit_snapshot(struct ice_hw *hw);
 int
 ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
-- 
cgit v1.2.3-70-g09d2


From 4bdf5f258331f049bbff2d770cfcb62f6b789dfe Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:11 -0800
Subject: ice: remove ice_mbx_deinit_snapshot

The ice_mbx_deinit_snapshot function's only remaining job is to clear the
previous snapshot data. This snapshot data is initialized when SR-IOV adds
VFs, so it is not necessary to clear this data when removing VFs. Since no
allocation occurs we no longer need to free anything and we can safely
remove this function.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c  |  5 +----
 drivers/net/ethernet/intel/ice/ice_vf_mbx.c | 14 --------------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.h |  1 -
 3 files changed, 1 insertion(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 8820f269bfdf..b65025b51526 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1014,7 +1014,6 @@ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs)
 	if (!num_vfs) {
 		if (!pci_vfs_assigned(pdev)) {
 			ice_free_vfs(pf);
-			ice_mbx_deinit_snapshot(&pf->hw);
 			if (pf->lag)
 				ice_enable_lag(pf->lag);
 			return 0;
@@ -1027,10 +1026,8 @@ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs)
 	ice_mbx_init_snapshot(&pf->hw);
 
 	err = ice_pci_sriov_ena(pf, num_vfs);
-	if (err) {
-		ice_mbx_deinit_snapshot(&pf->hw);
+	if (err)
 		return err;
-	}
 
 	if (pf->lag)
 		ice_disable_lag(pf->lag);
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
index 4bfed5fb3a88..1f332ab43b00 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
@@ -399,17 +399,3 @@ void ice_mbx_init_snapshot(struct ice_hw *hw)
 	INIT_LIST_HEAD(&snap->mbx_vf);
 	ice_mbx_reset_snapshot(snap);
 }
-
-/**
- * ice_mbx_deinit_snapshot - Free mailbox snapshot structure
- * @hw: pointer to the hardware structure
- *
- * Clear the mailbox snapshot structure and free the VF counter array.
- */
-void ice_mbx_deinit_snapshot(struct ice_hw *hw)
-{
-	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
-
-	/* Clear mbx_buf in the mailbox snaphot structure */
-	memset(&snap->mbx_buf, 0, sizeof(snap->mbx_buf));
-}
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
index a6d42f467dc5..e4bdd93ccef1 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
@@ -25,7 +25,6 @@ ice_mbx_vf_state_handler(struct ice_hw *hw, struct ice_mbx_data *mbx_data,
 void ice_mbx_clear_malvf(struct ice_mbx_vf_info *vf_info);
 void ice_mbx_init_vf_info(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info);
 void ice_mbx_init_snapshot(struct ice_hw *hw);
-void ice_mbx_deinit_snapshot(struct ice_hw *hw);
 int
 ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
 		     bool *report_malvf);
-- 
cgit v1.2.3-70-g09d2


From 07cc1a942216d1f211f1c641af8b6f810bb16699 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:12 -0800
Subject: ice: merge ice_mbx_report_malvf with ice_mbx_vf_state_handler

The ice_mbx_report_malvf function is used to update the
ice_mbx_vf_info.malicious member after we detect a malicious VF. This is
done by calling ice_mbx_report_malvf after ice_mbx_vf_state_handler sets
its "is_malvf" return parameter true.

Instead of requiring two steps, directly update the malicious bit in the
state handler, and remove the need for separately calling
ice_mbx_report_malvf.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c  | 34 +++++++------------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.c | 51 +++++++++--------------------
 drivers/net/ethernet/intel/ice/ice_vf_mbx.h |  5 +--
 3 files changed, 28 insertions(+), 62 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index b65025b51526..6152c90d7286 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1794,7 +1794,7 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 	s16 vf_id = le16_to_cpu(event->desc.retval);
 	struct device *dev = ice_pf_to_dev(pf);
 	struct ice_mbx_data mbxdata;
-	bool malvf = false;
+	bool report_malvf = false;
 	struct ice_vf *vf;
 	int status;
 
@@ -1811,33 +1811,23 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 #define ICE_MBX_OVERFLOW_WATERMARK 64
 	mbxdata.async_watermark_val = ICE_MBX_OVERFLOW_WATERMARK;
 
-	/* check to see if we have a malicious VF */
-	status = ice_mbx_vf_state_handler(&pf->hw, &mbxdata, &vf->mbx_info, &malvf);
+	/* check to see if we have a newly malicious VF */
+	status = ice_mbx_vf_state_handler(&pf->hw, &mbxdata, &vf->mbx_info,
+					  &report_malvf);
 	if (status)
 		goto out_put_vf;
 
-	if (malvf) {
-		bool report_vf = false;
+	if (report_malvf) {
+		struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
 
-		/* if the VF is malicious and we haven't let the user
-		 * know about it, then let them know now
-		 */
-		status = ice_mbx_report_malvf(&pf->hw, &vf->mbx_info,
-					      &report_vf);
-		if (status)
-			dev_dbg(dev, "Error reporting malicious VF\n");
-
-		if (report_vf) {
-			struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
-
-			if (pf_vsi)
-				dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
-					 &vf->dev_lan_addr[0],
-					 pf_vsi->netdev->dev_addr);
-		}
+		if (pf_vsi)
+			dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
+				 &vf->dev_lan_addr[0],
+				 pf_vsi->netdev->dev_addr);
 	}
 
 out_put_vf:
 	ice_put_vf(vf);
-	return malvf;
+
+	return vf->mbx_info.malicious;
 }
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
index 1f332ab43b00..40cb4ba0789c 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.c
@@ -215,7 +215,7 @@ ice_mbx_detect_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
  * @hw: pointer to the HW struct
  * @mbx_data: pointer to structure containing mailbox data
  * @vf_info: mailbox tracking structure for the VF in question
- * @is_malvf: boolean output to indicate if VF is malicious
+ * @report_malvf: boolean output to indicate whether VF should be reported
  *
  * The function serves as an entry point for the malicious VF
  * detection algorithm by handling the different states and state
@@ -234,25 +234,24 @@ ice_mbx_detect_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
  * the static snapshot and look for a malicious VF.
  */
 int
-ice_mbx_vf_state_handler(struct ice_hw *hw,
-			 struct ice_mbx_data *mbx_data,
-			 struct ice_mbx_vf_info *vf_info,
-			 bool *is_malvf)
+ice_mbx_vf_state_handler(struct ice_hw *hw, struct ice_mbx_data *mbx_data,
+			 struct ice_mbx_vf_info *vf_info, bool *report_malvf)
 {
 	struct ice_mbx_snapshot *snap = &hw->mbx_snapshot;
 	struct ice_mbx_snap_buffer_data *snap_buf;
 	struct ice_ctl_q_info *cq = &hw->mailboxq;
 	enum ice_mbx_snapshot_state new_state;
+	bool is_malvf = false;
 	int status = 0;
 
-	if (!is_malvf || !mbx_data)
+	if (!report_malvf || !mbx_data || !vf_info)
 		return -EINVAL;
 
+	*report_malvf = false;
+
 	/* When entering the mailbox state machine assume that the VF
 	 * is not malicious until detected.
 	 */
-	*is_malvf = false;
-
 	 /* Checking if max messages allowed to be processed while servicing current
 	  * interrupt is not less than the defined AVF message threshold.
 	  */
@@ -301,8 +300,7 @@ ice_mbx_vf_state_handler(struct ice_hw *hw,
 		if (snap_buf->num_pending_arq >=
 		    mbx_data->async_watermark_val) {
 			new_state = ICE_MAL_VF_DETECT_STATE_DETECT;
-			status = ice_mbx_detect_malvf(hw, vf_info, &new_state,
-						      is_malvf);
+			status = ice_mbx_detect_malvf(hw, vf_info, &new_state, &is_malvf);
 		} else {
 			new_state = ICE_MAL_VF_DETECT_STATE_TRAVERSE;
 			ice_mbx_traverse(hw, &new_state);
@@ -316,8 +314,7 @@ ice_mbx_vf_state_handler(struct ice_hw *hw,
 
 	case ICE_MAL_VF_DETECT_STATE_DETECT:
 		new_state = ICE_MAL_VF_DETECT_STATE_DETECT;
-		status = ice_mbx_detect_malvf(hw, vf_info, &new_state,
-					      is_malvf);
+		status = ice_mbx_detect_malvf(hw, vf_info, &new_state, &is_malvf);
 		break;
 
 	default:
@@ -327,31 +324,13 @@ ice_mbx_vf_state_handler(struct ice_hw *hw,
 
 	snap_buf->state = new_state;
 
-	return status;
-}
-
-/**
- * ice_mbx_report_malvf - Track and note malicious VF
- * @hw: pointer to the HW struct
- * @vf_info: the mailbox tracking info structure for a VF
- * @report_malvf: boolean to indicate if malicious VF must be reported
- *
- * This function updates the malicious indicator bit in the VF mailbox
- * tracking structure. A malicious VF must be reported only once if discovered
- * between VF resets or loading so the function first checks if the VF has
- * already been detected in any previous mailbox iterations.
- */
-int
-ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
-		     bool *report_malvf)
-{
-	if (!report_malvf)
-		return -EINVAL;
-
-	*report_malvf = !vf_info->malicious;
-	vf_info->malicious = 1;
+	/* Only report VFs as malicious the first time we detect it */
+	if (is_malvf && !vf_info->malicious) {
+		vf_info->malicious = 1;
+		*report_malvf = true;
+	}
 
-	return 0;
+	return status;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
index e4bdd93ccef1..41250519bc56 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
@@ -21,13 +21,10 @@ ice_aq_send_msg_to_vf(struct ice_hw *hw, u16 vfid, u32 v_opcode, u32 v_retval,
 u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed);
 int
 ice_mbx_vf_state_handler(struct ice_hw *hw, struct ice_mbx_data *mbx_data,
-			 struct ice_mbx_vf_info *vf_info, bool *is_mal_vf);
+			 struct ice_mbx_vf_info *vf_info, bool *report_malvf);
 void ice_mbx_clear_malvf(struct ice_mbx_vf_info *vf_info);
 void ice_mbx_init_vf_info(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info);
 void ice_mbx_init_snapshot(struct ice_hw *hw);
-int
-ice_mbx_report_malvf(struct ice_hw *hw, struct ice_mbx_vf_info *vf_info,
-		     bool *report_malvf);
 #else /* CONFIG_PCI_IOV */
 static inline int
 ice_aq_send_msg_to_vf(struct ice_hw __always_unused *hw,
-- 
cgit v1.2.3-70-g09d2


From dde7db637d9981b47da0da575661d0ec83f8b25a Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:13 -0800
Subject: ice: initialize mailbox snapshot earlier in PF init

Now that we no longer depend on the number of VFs being allocated, we can
move the ice_mbx_init_snapshot function earlier. This will be required by
Scalable IOV as we will not be calling ice_sriov_configure for Scalable
VFs.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c   | 1 +
 drivers/net/ethernet/intel/ice/ice_sriov.c  | 2 --
 drivers/net/ethernet/intel/ice/ice_vf_mbx.h | 4 ++++
 3 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 567694bf098b..615a731d7afe 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3891,6 +3891,7 @@ static int ice_init_pf(struct ice_pf *pf)
 
 	mutex_init(&pf->vfs.table_lock);
 	hash_init(pf->vfs.table);
+	ice_mbx_init_snapshot(&pf->hw);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 6152c90d7286..79159cbb66ec 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1023,8 +1023,6 @@ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs)
 		return -EBUSY;
 	}
 
-	ice_mbx_init_snapshot(&pf->hw);
-
 	err = ice_pci_sriov_ena(pf, num_vfs);
 	if (err)
 		return err;
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
index 41250519bc56..44bc030d17e0 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
+++ b/drivers/net/ethernet/intel/ice/ice_vf_mbx.h
@@ -43,5 +43,9 @@ ice_conv_link_speed_to_virtchnl(bool __always_unused adv_link_support,
 	return 0;
 }
 
+static inline void ice_mbx_init_snapshot(struct ice_hw *hw)
+{
+}
+
 #endif /* CONFIG_PCI_IOV */
 #endif /* _ICE_VF_MBX_H_ */
-- 
cgit v1.2.3-70-g09d2


From 33b035e70611c10c5aa3864a7517570b25a46ebb Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:14 -0800
Subject: ice: declare ice_vc_process_vf_msg in ice_virtchnl.h

The ice_vc_process_vf_msg function is the main entry point for handling
virtchnl messages. This function is defined in ice_virtchnl.c but its
declaration is still in ice_sriov.c

The ice_sriov.c file used to contain all of the virtualization logic until
commit bf93bf791cec ("ice: introduce ice_virtchnl.c and ice_virtchnl.h")
moved the virtchnl logic to its own file.

The ice_vc_process_vf_msg function should have had its declaration moved to
ice_virtchnl.h then. Fix this.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.h    | 3 ---
 drivers/net/ethernet/intel/ice/ice_virtchnl.h | 6 ++++++
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.h b/drivers/net/ethernet/intel/ice/ice_sriov.h
index 955ab810a198..1082b0691a3f 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.h
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.h
@@ -33,7 +33,6 @@ int
 ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi);
 
 void ice_free_vfs(struct ice_pf *pf);
-void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event);
 void ice_restore_all_vfs_msi_state(struct pci_dev *pdev);
 bool
 ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
@@ -68,8 +67,6 @@ ice_vc_validate_pattern(struct ice_vf *vf, struct virtchnl_proto_hdrs *proto);
 static inline void ice_process_vflr_event(struct ice_pf *pf) { }
 static inline void ice_free_vfs(struct ice_pf *pf) { }
 static inline
-void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event) { }
-static inline
 void ice_vf_lan_overflow_event(struct ice_pf *pf, struct ice_rq_event_info *event) { }
 static inline void ice_print_vfs_mdd_events(struct ice_pf *pf) { }
 static inline void ice_print_vf_rx_mdd_event(struct ice_vf *vf) { }
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.h b/drivers/net/ethernet/intel/ice/ice_virtchnl.h
index b454654d7b0c..6d5af29c855e 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl.h
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.h
@@ -63,6 +63,7 @@ int
 ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode,
 		      enum virtchnl_status_code v_retval, u8 *msg, u16 msglen);
 bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id);
+void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event);
 #else /* CONFIG_PCI_IOV */
 static inline void ice_virtchnl_set_dflt_ops(struct ice_vf *vf) { }
 static inline void ice_virtchnl_set_repr_ops(struct ice_vf *vf) { }
@@ -81,6 +82,11 @@ static inline bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id)
 {
 	return false;
 }
+
+static inline void
+ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
+{
+}
 #endif /* !CONFIG_PCI_IOV */
 
 #endif /* _ICE_VIRTCHNL_H_ */
-- 
cgit v1.2.3-70-g09d2


From 4f0636fef61ad0b22fed4fb06f369d6ba38f807d Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:15 -0800
Subject: ice: always report VF overflowing mailbox even without PF VSI

In ice_is_malicious_vf we report a message warning the system administrator
when a VF is potentially spamming the PF with asynchronous messages that
could overflow the PF mailbox.

The specific message was requested by our customer support team to include
the VF and PF MAC address. In some cases we may not be able to locate the
PF VSI to obtain the MAC address for the PF. The current implementation
discards the message entirely in this case. Fix this to instead print a
zero address in that case so that we always print something here. Note that
dev_warn will also include the PCI device information allowing another
mechanism for determining on which PF the potentially malicious VF belongs.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 79159cbb66ec..185673afb781 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1817,11 +1817,11 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 
 	if (report_malvf) {
 		struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
+		u8 zero_addr[ETH_ALEN] = {};
 
-		if (pf_vsi)
-			dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
-				 &vf->dev_lan_addr[0],
-				 pf_vsi->netdev->dev_addr);
+		dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
+			 &vf->dev_lan_addr[0],
+			 pf_vsi ? pf_vsi->netdev->dev_addr : zero_addr);
 	}
 
 out_put_vf:
-- 
cgit v1.2.3-70-g09d2


From 3f22fc3131b814f0d5d1ce6d94fb8239e6df1754 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:16 -0800
Subject: ice: remove unnecessary &array[0] and just use array

In ice_is_malicious_vf we print the VF MAC address using %pM by passing the
address of the first element of vf->dev_lan_addr. This is equivalent to
just passing vf->dev_lan_addr, so do that.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 185673afb781..938be486721e 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1820,7 +1820,7 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 		u8 zero_addr[ETH_ALEN] = {};
 
 		dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
-			 &vf->dev_lan_addr[0],
+			 vf->dev_lan_addr,
 			 pf_vsi ? pf_vsi->netdev->dev_addr : zero_addr);
 	}
 
-- 
cgit v1.2.3-70-g09d2


From afc24d6584fbd246d98c0feb464b94da67661e3e Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:17 -0800
Subject: ice: pass mbxdata to ice_is_malicious_vf()

The ice_is_malicious_vf() function takes information about the current
state of the mailbox during a single interrupt. This information includes
the number of messages processed so far, as well as the number of pending
messages not yet processed.

A future refactor is going to make ice_vc_process_vf_msg() call
ice_is_malicious_vf() instead of having it called separately in ice_main.c
This change will require passing all the necessary arguments into
ice_vc_process_vf_msg().

To make this simpler, have the main loop fill in the struct ice_mbx_data
and pass that rather than passing in the num_msg_proc and num_msg_pending.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c  | 10 +++++++++-
 drivers/net/ethernet/intel/ice/ice_sriov.c | 14 +++-----------
 drivers/net/ethernet/intel/ice/ice_sriov.h |  5 ++---
 3 files changed, 14 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 615a731d7afe..a7e7a186009e 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1393,6 +1393,8 @@ static void ice_aq_cancel_waiting_tasks(struct ice_pf *pf)
 	wake_up(&pf->aq_wait_queue);
 }
 
+#define ICE_MBX_OVERFLOW_WATERMARK 64
+
 /**
  * __ice_clean_ctrlq - helper function to clean controlq rings
  * @pf: ptr to struct ice_pf
@@ -1483,6 +1485,7 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type)
 		return 0;
 
 	do {
+		struct ice_mbx_data data = {};
 		u16 opcode;
 		int ret;
 
@@ -1509,7 +1512,12 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type)
 			ice_vf_lan_overflow_event(pf, &event);
 			break;
 		case ice_mbx_opc_send_msg_to_pf:
-			if (!ice_is_malicious_vf(pf, &event, i, pending))
+			data.num_msg_proc = i;
+			data.num_pending_arq = pending;
+			data.max_num_msgs_mbx = hw->mailboxq.num_rq_entries;
+			data.async_watermark_val = ICE_MBX_OVERFLOW_WATERMARK;
+
+			if (!ice_is_malicious_vf(pf, &event, &data))
 				ice_vc_process_vf_msg(pf, &event);
 			break;
 		case ice_aqc_opc_fw_logging:
diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 938be486721e..5ae923ea979c 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1782,16 +1782,14 @@ void ice_restore_all_vfs_msi_state(struct pci_dev *pdev)
  * ice_is_malicious_vf - helper function to detect a malicious VF
  * @pf: ptr to struct ice_pf
  * @event: pointer to the AQ event
- * @num_msg_proc: the number of messages processed so far
- * @num_msg_pending: the number of messages peinding in admin queue
+ * @mbxdata: data about the state of the mailbox
  */
 bool
 ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
-		    u16 num_msg_proc, u16 num_msg_pending)
+		    struct ice_mbx_data *mbxdata)
 {
 	s16 vf_id = le16_to_cpu(event->desc.retval);
 	struct device *dev = ice_pf_to_dev(pf);
-	struct ice_mbx_data mbxdata;
 	bool report_malvf = false;
 	struct ice_vf *vf;
 	int status;
@@ -1803,14 +1801,8 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 	if (test_bit(ICE_VF_STATE_DIS, vf->vf_states))
 		goto out_put_vf;
 
-	mbxdata.num_msg_proc = num_msg_proc;
-	mbxdata.num_pending_arq = num_msg_pending;
-	mbxdata.max_num_msgs_mbx = pf->hw.mailboxq.num_rq_entries;
-#define ICE_MBX_OVERFLOW_WATERMARK 64
-	mbxdata.async_watermark_val = ICE_MBX_OVERFLOW_WATERMARK;
-
 	/* check to see if we have a newly malicious VF */
-	status = ice_mbx_vf_state_handler(&pf->hw, &mbxdata, &vf->mbx_info,
+	status = ice_mbx_vf_state_handler(&pf->hw, mbxdata, &vf->mbx_info,
 					  &report_malvf);
 	if (status)
 		goto out_put_vf;
diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.h b/drivers/net/ethernet/intel/ice/ice_sriov.h
index 1082b0691a3f..8fa61d954fae 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.h
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.h
@@ -36,7 +36,7 @@ void ice_free_vfs(struct ice_pf *pf);
 void ice_restore_all_vfs_msi_state(struct pci_dev *pdev);
 bool
 ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
-		    u16 num_msg_proc, u16 num_msg_pending);
+		    struct ice_mbx_data *mbxdata);
 
 int
 ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos,
@@ -75,8 +75,7 @@ static inline void ice_restore_all_vfs_msi_state(struct pci_dev *pdev) { }
 static inline bool
 ice_is_malicious_vf(struct ice_pf __always_unused *pf,
 		    struct ice_rq_event_info __always_unused *event,
-		    u16 __always_unused num_msg_proc,
-		    u16 __always_unused num_msg_pending)
+		    struct ice_mbx_data *mbxdata)
 {
 	return false;
 }
-- 
cgit v1.2.3-70-g09d2


From 4508bf02bf8a3de8fb65869f40dfdef74dc1b339 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:18 -0800
Subject: ice: print message if ice_mbx_vf_state_handler returns an error

If ice_mbx_vf_state_handler() returns an error, the ice_is_malicious_vf()
function just exits without printing anything.

Instead, use dev_warn_ratelimited to print a warning that we were unable to
check the status for this VF. The _ratelimited variant is used to avoid
potentially spamming the log if this function is failing consistently for
every single mailbox message.

Also we can drop the "goto" as it simply skips over a report_malvf check.
That variable should always be false if ice_mbx_vf_state_handler returns
non-zero.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 5ae923ea979c..f0daeda236de 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1805,7 +1805,8 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 	status = ice_mbx_vf_state_handler(&pf->hw, mbxdata, &vf->mbx_info,
 					  &report_malvf);
 	if (status)
-		goto out_put_vf;
+		dev_warn_ratelimited(dev, "Unable to check status of mailbox overflow for VF %u MAC %pM, status %d\n",
+				     vf->vf_id, vf->dev_lan_addr, status);
 
 	if (report_malvf) {
 		struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
-- 
cgit v1.2.3-70-g09d2


From c414463ab1bb098e67f4c1a4ef64f3e97780f087 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:19 -0800
Subject: ice: move ice_is_malicious_vf() to ice_virtchnl.c

The ice_is_malicious_vf() function is currently implemented in ice_sriov.c
This function is not Single Root specific, and a future change is going to
refactor the ice_vc_process_vf_msg() function to call this instead of
calling it before ice_vc_process_vf_msg() in the main loop of
__ice_clean_ctrlq.

To make that change easier to review, first move this function into
ice_virtchnl.c but leave the call in __ice_clean_ctrlq() alone.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sriov.c    | 45 ---------------------------
 drivers/net/ethernet/intel/ice/ice_sriov.h    | 11 -------
 drivers/net/ethernet/intel/ice/ice_virtchnl.c | 45 +++++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_virtchnl.h | 11 +++++++
 4 files changed, 56 insertions(+), 56 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index f0daeda236de..6fa62c3cedb0 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1777,48 +1777,3 @@ void ice_restore_all_vfs_msi_state(struct pci_dev *pdev)
 		}
 	}
 }
-
-/**
- * ice_is_malicious_vf - helper function to detect a malicious VF
- * @pf: ptr to struct ice_pf
- * @event: pointer to the AQ event
- * @mbxdata: data about the state of the mailbox
- */
-bool
-ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
-		    struct ice_mbx_data *mbxdata)
-{
-	s16 vf_id = le16_to_cpu(event->desc.retval);
-	struct device *dev = ice_pf_to_dev(pf);
-	bool report_malvf = false;
-	struct ice_vf *vf;
-	int status;
-
-	vf = ice_get_vf_by_id(pf, vf_id);
-	if (!vf)
-		return false;
-
-	if (test_bit(ICE_VF_STATE_DIS, vf->vf_states))
-		goto out_put_vf;
-
-	/* check to see if we have a newly malicious VF */
-	status = ice_mbx_vf_state_handler(&pf->hw, mbxdata, &vf->mbx_info,
-					  &report_malvf);
-	if (status)
-		dev_warn_ratelimited(dev, "Unable to check status of mailbox overflow for VF %u MAC %pM, status %d\n",
-				     vf->vf_id, vf->dev_lan_addr, status);
-
-	if (report_malvf) {
-		struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
-		u8 zero_addr[ETH_ALEN] = {};
-
-		dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
-			 vf->dev_lan_addr,
-			 pf_vsi ? pf_vsi->netdev->dev_addr : zero_addr);
-	}
-
-out_put_vf:
-	ice_put_vf(vf);
-
-	return vf->mbx_info.malicious;
-}
diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.h b/drivers/net/ethernet/intel/ice/ice_sriov.h
index 8fa61d954fae..346cb2666f3a 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.h
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.h
@@ -34,9 +34,6 @@ ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi);
 
 void ice_free_vfs(struct ice_pf *pf);
 void ice_restore_all_vfs_msi_state(struct pci_dev *pdev);
-bool
-ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
-		    struct ice_mbx_data *mbxdata);
 
 int
 ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos,
@@ -72,14 +69,6 @@ static inline void ice_print_vfs_mdd_events(struct ice_pf *pf) { }
 static inline void ice_print_vf_rx_mdd_event(struct ice_vf *vf) { }
 static inline void ice_restore_all_vfs_msi_state(struct pci_dev *pdev) { }
 
-static inline bool
-ice_is_malicious_vf(struct ice_pf __always_unused *pf,
-		    struct ice_rq_event_info __always_unused *event,
-		    struct ice_mbx_data *mbxdata)
-{
-	return false;
-}
-
 static inline int
 ice_sriov_configure(struct pci_dev __always_unused *pdev,
 		    int __always_unused num_vfs)
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.c b/drivers/net/ethernet/intel/ice/ice_virtchnl.c
index e24e3f5017ca..e0c573d9d1b9 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.c
@@ -3833,6 +3833,51 @@ void ice_virtchnl_set_repr_ops(struct ice_vf *vf)
 	vf->virtchnl_ops = &ice_virtchnl_repr_ops;
 }
 
+/**
+ * ice_is_malicious_vf - helper function to detect a malicious VF
+ * @pf: ptr to struct ice_pf
+ * @event: pointer to the AQ event
+ * @mbxdata: data about the state of the mailbox
+ */
+bool
+ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
+		    struct ice_mbx_data *mbxdata)
+{
+	s16 vf_id = le16_to_cpu(event->desc.retval);
+	struct device *dev = ice_pf_to_dev(pf);
+	bool report_malvf = false;
+	struct ice_vf *vf;
+	int status;
+
+	vf = ice_get_vf_by_id(pf, vf_id);
+	if (!vf)
+		return false;
+
+	if (test_bit(ICE_VF_STATE_DIS, vf->vf_states))
+		goto out_put_vf;
+
+	/* check to see if we have a newly malicious VF */
+	status = ice_mbx_vf_state_handler(&pf->hw, mbxdata, &vf->mbx_info,
+					  &report_malvf);
+	if (status)
+		dev_warn_ratelimited(dev, "Unable to check status of mailbox overflow for VF %u MAC %pM, status %d\n",
+				     vf->vf_id, vf->dev_lan_addr, status);
+
+	if (report_malvf) {
+		struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
+		u8 zero_addr[ETH_ALEN] = {};
+
+		dev_warn(dev, "VF MAC %pM on PF MAC %pM is generating asynchronous messages and may be overflowing the PF message queue. Please see the Adapter User Guide for more information\n",
+			 vf->dev_lan_addr,
+			 pf_vsi ? pf_vsi->netdev->dev_addr : zero_addr);
+	}
+
+out_put_vf:
+	ice_put_vf(vf);
+
+	return vf->mbx_info.malicious;
+}
+
 /**
  * ice_vc_process_vf_msg - Process request from VF
  * @pf: pointer to the PF structure
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.h b/drivers/net/ethernet/intel/ice/ice_virtchnl.h
index 6d5af29c855e..648a383fad85 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl.h
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.h
@@ -63,6 +63,9 @@ int
 ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode,
 		      enum virtchnl_status_code v_retval, u8 *msg, u16 msglen);
 bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id);
+bool
+ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
+		    struct ice_mbx_data *mbxdata);
 void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event);
 #else /* CONFIG_PCI_IOV */
 static inline void ice_virtchnl_set_dflt_ops(struct ice_vf *vf) { }
@@ -83,6 +86,14 @@ static inline bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id)
 	return false;
 }
 
+static inline bool
+ice_is_malicious_vf(struct ice_pf __always_unused *pf,
+		    struct ice_rq_event_info __always_unused *event,
+		    struct ice_mbx_data *mbxdata)
+{
+	return false;
+}
+
 static inline void
 ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
 {
-- 
cgit v1.2.3-70-g09d2


From be96815c616822d3800405b8fbebe3e069d6eed2 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Wed, 22 Feb 2023 09:09:20 -0800
Subject: ice: call ice_is_malicious_vf() from ice_vc_process_vf_msg()

The main loop in __ice_clean_ctrlq first checks if a VF might be malicious
before calling ice_vc_process_vf_msg(). This results in duplicate code in
both functions to obtain a reference to the VF, and exports the
ice_is_malicious_vf() from ice_virtchnl.c unnecessarily.

Refactor ice_is_malicious_vf() to be a static function that takes a pointer
to the VF. Call this in ice_vc_process_vf_msg() just after we obtain a
reference to the VF by calling ice_get_vf_by_id.

Pass the mailbox data from the __ice_clean_ctrlq function into
ice_vc_process_vf_msg() instead of calling ice_is_malicious_vf().

This reduces the number of exported functions and avoids the need to obtain
the VF reference twice for every mailbox message.

Note that the state check for ICE_VF_STATE_DIS is kept in
ice_is_malicious_vf() and we call this before checking that state in
ice_vc_process_vf_msg. This is intentional, as we stop responding to VF
messages from a VF once we detect that it may be overflowing the mailbox.
This ensures that we continue to silently ignore the message as before
without responding via ice_vc_send_msg_to_vf().

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c     |  3 +--
 drivers/net/ethernet/intel/ice/ice_virtchnl.c | 36 ++++++++++++++-------------
 drivers/net/ethernet/intel/ice/ice_virtchnl.h | 17 +++----------
 3 files changed, 24 insertions(+), 32 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index a7e7a186009e..20b3f3e6eda1 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1517,8 +1517,7 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type)
 			data.max_num_msgs_mbx = hw->mailboxq.num_rq_entries;
 			data.async_watermark_val = ICE_MBX_OVERFLOW_WATERMARK;
 
-			if (!ice_is_malicious_vf(pf, &event, &data))
-				ice_vc_process_vf_msg(pf, &event);
+			ice_vc_process_vf_msg(pf, &event, &data);
 			break;
 		case ice_aqc_opc_fw_logging:
 			ice_output_fw_log(hw, &event.desc, event.msg_buf);
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.c b/drivers/net/ethernet/intel/ice/ice_virtchnl.c
index e0c573d9d1b9..97243c616d5d 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.c
@@ -3834,27 +3834,26 @@ void ice_virtchnl_set_repr_ops(struct ice_vf *vf)
 }
 
 /**
- * ice_is_malicious_vf - helper function to detect a malicious VF
- * @pf: ptr to struct ice_pf
- * @event: pointer to the AQ event
+ * ice_is_malicious_vf - check if this vf might be overflowing mailbox
+ * @vf: the VF to check
  * @mbxdata: data about the state of the mailbox
+ *
+ * Detect if a given VF might be malicious and attempting to overflow the PF
+ * mailbox. If so, log a warning message and ignore this event.
  */
-bool
-ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
-		    struct ice_mbx_data *mbxdata)
+static bool
+ice_is_malicious_vf(struct ice_vf *vf, struct ice_mbx_data *mbxdata)
 {
-	s16 vf_id = le16_to_cpu(event->desc.retval);
-	struct device *dev = ice_pf_to_dev(pf);
 	bool report_malvf = false;
-	struct ice_vf *vf;
+	struct device *dev;
+	struct ice_pf *pf;
 	int status;
 
-	vf = ice_get_vf_by_id(pf, vf_id);
-	if (!vf)
-		return false;
+	pf = vf->pf;
+	dev = ice_pf_to_dev(pf);
 
 	if (test_bit(ICE_VF_STATE_DIS, vf->vf_states))
-		goto out_put_vf;
+		return vf->mbx_info.malicious;
 
 	/* check to see if we have a newly malicious VF */
 	status = ice_mbx_vf_state_handler(&pf->hw, mbxdata, &vf->mbx_info,
@@ -3872,9 +3871,6 @@ ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
 			 pf_vsi ? pf_vsi->netdev->dev_addr : zero_addr);
 	}
 
-out_put_vf:
-	ice_put_vf(vf);
-
 	return vf->mbx_info.malicious;
 }
 
@@ -3882,11 +3878,13 @@ out_put_vf:
  * ice_vc_process_vf_msg - Process request from VF
  * @pf: pointer to the PF structure
  * @event: pointer to the AQ event
+ * @mbxdata: information used to detect VF attempting mailbox overflow
  *
  * called from the common asq/arq handler to
  * process request from VF
  */
-void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
+void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event,
+			   struct ice_mbx_data *mbxdata)
 {
 	u32 v_opcode = le32_to_cpu(event->desc.cookie_high);
 	s16 vf_id = le16_to_cpu(event->desc.retval);
@@ -3908,6 +3906,10 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
 
 	mutex_lock(&vf->cfg_lock);
 
+	/* Check if the VF is trying to overflow the mailbox */
+	if (ice_is_malicious_vf(vf, mbxdata))
+		goto finish;
+
 	/* Check if VF is disabled. */
 	if (test_bit(ICE_VF_STATE_DIS, vf->vf_states)) {
 		err = -EPERM;
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.h b/drivers/net/ethernet/intel/ice/ice_virtchnl.h
index 648a383fad85..cd747718de73 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl.h
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.h
@@ -63,10 +63,8 @@ int
 ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode,
 		      enum virtchnl_status_code v_retval, u8 *msg, u16 msglen);
 bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id);
-bool
-ice_is_malicious_vf(struct ice_pf *pf, struct ice_rq_event_info *event,
-		    struct ice_mbx_data *mbxdata);
-void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event);
+void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event,
+			   struct ice_mbx_data *mbxdata);
 #else /* CONFIG_PCI_IOV */
 static inline void ice_virtchnl_set_dflt_ops(struct ice_vf *vf) { }
 static inline void ice_virtchnl_set_repr_ops(struct ice_vf *vf) { }
@@ -86,16 +84,9 @@ static inline bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id)
 	return false;
 }
 
-static inline bool
-ice_is_malicious_vf(struct ice_pf __always_unused *pf,
-		    struct ice_rq_event_info __always_unused *event,
-		    struct ice_mbx_data *mbxdata)
-{
-	return false;
-}
-
 static inline void
-ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
+ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event,
+		      struct ice_mbx_data *mbxdata)
 {
 }
 #endif /* !CONFIG_PCI_IOV */
-- 
cgit v1.2.3-70-g09d2


From 34f0677e7afd3a292bc1aadda7ce8e35faedb204 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 13 Mar 2023 11:40:17 -0700
Subject: bpf: fix precision propagation verbose logging

Fix wrong order of frame index vs register/slot index in precision
propagation verbose (level 2) output. It's wrong and very confusing as is.

Fixes: 529409ea92d5 ("bpf: propagate precision across all frames, not just the last one")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230313184017.4083374-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 519d465407ef..883d4ff2e288 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -15068,7 +15068,7 @@ static int propagate_precision(struct bpf_verifier_env *env,
 			    !(state_reg->live & REG_LIVE_READ))
 				continue;
 			if (env->log.level & BPF_LOG_LEVEL2)
-				verbose(env, "frame %d: propagating r%d\n", i, fr);
+				verbose(env, "frame %d: propagating r%d\n", fr, i);
 			err = mark_chain_precision_frame(env, fr, i);
 			if (err < 0)
 				return err;
@@ -15084,7 +15084,7 @@ static int propagate_precision(struct bpf_verifier_env *env,
 				continue;
 			if (env->log.level & BPF_LOG_LEVEL2)
 				verbose(env, "frame %d: propagating fp%d\n",
-					(-i - 1) * BPF_REG_SIZE, fr);
+					fr, (-i - 1) * BPF_REG_SIZE);
 			err = mark_chain_precision_stack_frame(env, fr, i);
 			if (err < 0)
 				return err;
-- 
cgit v1.2.3-70-g09d2


From 22df776a9a866713d9decfb92b633bcfdb571954 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Wed, 15 Feb 2023 17:30:33 -0600
Subject: tasks: Extract rcu_users out of union

In commit 3fbd7ee285b2b ("tasks: Add a count of task RCU users"), a
count on the number of RCU users was added to struct task_struct. This
was done so as to enable the removal of task_rcu_dereference(), and
allow tasks to be protected by RCU even after exiting and being removed
from the runqueue. In this commit, the 'refcount_t rcu_users' field that
keeps track of this refcount was put into a union co-located with
'struct rcu_head rcu', so as to avoid taking up any extra space in
task_struct. This was possible to do safely, because the field was only
ever decremented by a static set of specific callers, and then never
incremented again.

While this restriction of there only being a small, static set of users
of this field has worked fine, it prevents us from leveraging the field
to use RCU to protect tasks in other contexts.

During tracing, for example, it would be useful to be able to collect
some tasks that performed a certain operation, put them in a map, and
then periodically summarize who they are, which cgroup they're in, how
much CPU time they've utilized, etc. While this can currently be done
with 'usage', it becomes tricky when a task is already in a map, or if a
reference should only be taken if a task is valid and will not soon be
reaped. Ideally, we could do something like pass a reference to a map
value, and then try to acquire a reference to the task in an RCU read
region by using refcount_inc_not_zero().

Similarly, in sched_ext, schedulers are using integer pids to remember
tasks, and then looking them up with find_task_by_pid_ns(). This is
slow, error prone, and adds complexity. It would be more convenient and
performant if BPF schedulers could instead store tasks directly in maps,
and then leverage RCU to ensure they can be safely accessed with low
overhead.

Finally, overloading fields like this is error prone. Someone that wants
to use 'rcu_users' could easily overlook the fact that once the rcu
callback is scheduled, the refcount will go back to being nonzero, thus
precluding the use of refcount_inc_not_zero(). Furthermore, as described
below, it's possible to extract the fields of the union without changing
the size of task_struct.

There are several possible ways to enable this:

1. The lightest touch approach is likely the one proposed in this patch,
   which is to simply extract 'rcu_users' and 'rcu' from the union, so
   that scheduling the 'rcu' callback doesn't overwrite the 'rcu_users'
   refcount. If we have a trusted task pointer, this would allow us to
   use refcnt_inc_not_zero() inside of an RCU region to determine if we
   can safely acquire a reference to the task and store it in a map. As
   mentioned below, this can be done without changing the size of
   task_struct, by moving the location of the union to another location
   that has padding gaps we can fill in.

2. Removing 'refcount_t rcu_users', and instead having the entire task
   be freed in an rcu callback. This is likely the most sound overall
   design, though it changes the behavioral semantics exposed to
   callers, who currently expect that a task that's successfully looked
   up in e.g. the pid_list with find_task_by_pid_ns(), can always have a
   'usage' reference acquired on them, as it's guaranteed to be >
   0 until after the next gp. In order for this approach to work, we'd
   have to audit all callers. This approach also slightly changes
   behavior observed by user space by not invoking
   trace_sched_process_free() until the whole task_struct is actually being
   freed, rather than just after it's exited. It also may change
   timings, as memory will be freed in an RCU callback rather than
   immediately when the final 'usage' refcount drops to 0. This also is
   arguably a benefit, as it provides more predictable performance to
   callers who are refcounting tasks.

3. There may be other solutions as well that don't require changing the
   layout of task_struct. For example, we could possibly do something
   complex from the BPF side, such as listen for task exit and remove a
   task from a map when the task is exiting. This would likely require
   significant custom handling for task_struct in the verifier, so a
   more generalizable solution is likely warranted.

As mentioned above, this patch proposes the lightest-touch approach
which allows callers elsewhere in the kernel to use 'rcu_users' to
ensure the lifetime of a task, by extracting 'rcu_users' and 'rcu' from
the union. There is no size change in task_struct with this patch.

Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: David Vernet <void@manifault.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20230215233033.889644-1-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/sched.h | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 63d242164b1a..b11b4517760f 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1318,11 +1318,6 @@ struct task_struct {
 
 	struct tlbflush_unmap_batch	tlb_ubc;
 
-	union {
-		refcount_t		rcu_users;
-		struct rcu_head		rcu;
-	};
-
 	/* Cache last used pipe for splice(): */
 	struct pipe_inode_info		*splice_pipe;
 
@@ -1459,6 +1454,8 @@ struct task_struct {
 	unsigned long			saved_state_change;
 # endif
 #endif
+	struct rcu_head			rcu;
+	refcount_t			rcu_users;
 	int				pagefault_disabled;
 #ifdef CONFIG_MMU
 	struct task_struct		*oom_reaper_list;
-- 
cgit v1.2.3-70-g09d2


From e4ed8ba08e3f1ef24771eb95f87da129ad752063 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 7 Mar 2023 22:44:02 +0100
Subject: net: phy: micrel: Add support for PTP_PF_PEROUT for lan8841

Lan8841 has 10 GPIOs and it has 2 events(EVENT_A and EVENT_B). It is
possible to assigned the 2 events to any of the GPIOs, but a GPIO can
have only 1 event at a time.
These events are used to generate periodic signals. It is possible to
configure the length, the start time and the period of the signal by
configuring the event.
Currently the SW uses only EVENT_A to generate the perout.

These events are generated by comparing the target time with the PHC
time. In case the PHC time is changed to a value bigger than the target
time + reload time, then it would generate only 1 event and then it
would stop because target time + reload time is small than PHC time.
Therefore it is required to change also the target time every time when
the PHC is changed. The same will apply also when the PHC time is
changed to a smaller value.

This was tested using:
testptp -L 6,2
testptp -p 1000000000 -w 200000000

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20230307214402.793057-1-horatiu.vultur@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/micrel.c | 391 ++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 389 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 2c84fccef4f6..e5b2af69ac03 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -318,6 +318,7 @@ struct kszphy_ptp_priv {
 	struct ptp_clock_info ptp_clock_info;
 	/* Lock for ptp_clock */
 	struct mutex ptp_lock;
+	struct ptp_pin_desc *pin_config;
 };
 
 struct kszphy_priv {
@@ -3658,6 +3659,77 @@ static int lan8841_hwtstamp(struct mii_timestamper *mii_ts, struct ifreq *ifr)
 	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? -EFAULT : 0;
 }
 
+#define LAN8841_EVENT_A		0
+#define LAN8841_EVENT_B		1
+#define LAN8841_PTP_LTC_TARGET_SEC_HI(event)	((event) == LAN8841_EVENT_A ? 278 : 288)
+#define LAN8841_PTP_LTC_TARGET_SEC_LO(event)	((event) == LAN8841_EVENT_A ? 279 : 289)
+#define LAN8841_PTP_LTC_TARGET_NS_HI(event)	((event) == LAN8841_EVENT_A ? 280 : 290)
+#define LAN8841_PTP_LTC_TARGET_NS_LO(event)	((event) == LAN8841_EVENT_A ? 281 : 291)
+
+static int lan8841_ptp_set_target(struct kszphy_ptp_priv *ptp_priv, u8 event,
+				  s64 sec, u32 nsec)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	int ret;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_SEC_HI(event),
+			    upper_16_bits(sec));
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_SEC_LO(event),
+			    lower_16_bits(sec));
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_NS_HI(event) & 0x3fff,
+			    upper_16_bits(nsec));
+	if (ret)
+		return ret;
+
+	return phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_NS_LO(event),
+			    lower_16_bits(nsec));
+}
+
+#define LAN8841_BUFFER_TIME	2
+
+static int lan8841_ptp_update_target(struct kszphy_ptp_priv *ptp_priv,
+				     const struct timespec64 *ts)
+{
+	return lan8841_ptp_set_target(ptp_priv, LAN8841_EVENT_A,
+				      ts->tv_sec + LAN8841_BUFFER_TIME, ts->tv_nsec);
+}
+
+#define LAN8841_PTP_LTC_TARGET_RELOAD_SEC_HI(event)	((event) == LAN8841_EVENT_A ? 282 : 292)
+#define LAN8841_PTP_LTC_TARGET_RELOAD_SEC_LO(event)	((event) == LAN8841_EVENT_A ? 283 : 293)
+#define LAN8841_PTP_LTC_TARGET_RELOAD_NS_HI(event)	((event) == LAN8841_EVENT_A ? 284 : 294)
+#define LAN8841_PTP_LTC_TARGET_RELOAD_NS_LO(event)	((event) == LAN8841_EVENT_A ? 285 : 295)
+
+static int lan8841_ptp_set_reload(struct kszphy_ptp_priv *ptp_priv, u8 event,
+				  s64 sec, u32 nsec)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	int ret;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_RELOAD_SEC_HI(event),
+			    upper_16_bits(sec));
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_RELOAD_SEC_LO(event),
+			    lower_16_bits(sec));
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_RELOAD_NS_HI(event) & 0x3fff,
+			    upper_16_bits(nsec));
+	if (ret)
+		return ret;
+
+	return phy_write_mmd(phydev, 2, LAN8841_PTP_LTC_TARGET_RELOAD_NS_LO(event),
+			     lower_16_bits(nsec));
+}
+
 #define LAN8841_PTP_LTC_SET_SEC_HI	262
 #define LAN8841_PTP_LTC_SET_SEC_MID	263
 #define LAN8841_PTP_LTC_SET_SEC_LO	264
@@ -3671,6 +3743,7 @@ static int lan8841_ptp_settime64(struct ptp_clock_info *ptp,
 	struct kszphy_ptp_priv *ptp_priv = container_of(ptp, struct kszphy_ptp_priv,
 							ptp_clock_info);
 	struct phy_device *phydev = ptp_priv->phydev;
+	int ret;
 
 	/* Set the value to be stored */
 	mutex_lock(&ptp_priv->ptp_lock);
@@ -3683,9 +3756,10 @@ static int lan8841_ptp_settime64(struct ptp_clock_info *ptp,
 	/* Set the command to load the LTC */
 	phy_write_mmd(phydev, 2, LAN8841_PTP_CMD_CTL,
 		      LAN8841_PTP_CMD_CTL_PTP_LTC_LOAD);
+	ret = lan8841_ptp_update_target(ptp_priv, ts);
 	mutex_unlock(&ptp_priv->ptp_lock);
 
-	return 0;
+	return ret;
 }
 
 #define LAN8841_PTP_LTC_RD_SEC_HI	358
@@ -3740,6 +3814,7 @@ static int lan8841_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 	bool add = true;
 	u32 nsec;
 	s32 sec;
+	int ret;
 
 	/* The HW allows up to 15 sec to adjust the time, but here we limit to
 	 * 10 sec the adjustment. The reason is, in case the adjustment is 14
@@ -3803,7 +3878,13 @@ static int lan8841_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 	}
 	mutex_unlock(&ptp_priv->ptp_lock);
 
-	return 0;
+	/* Update the target clock */
+	ptp->gettime64(ptp, &ts);
+	mutex_lock(&ptp_priv->ptp_lock);
+	ret = lan8841_ptp_update_target(ptp_priv, &ts);
+	mutex_unlock(&ptp_priv->ptp_lock);
+
+	return ret;
 }
 
 #define LAN8841_PTP_LTC_RATE_ADJ_HI		269
@@ -3839,6 +3920,292 @@ static int lan8841_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 	return 0;
 }
 
+static int lan8841_ptp_verify(struct ptp_clock_info *ptp, unsigned int pin,
+			      enum ptp_pin_function func, unsigned int chan)
+{
+	switch (func) {
+	case PTP_PF_NONE:
+	case PTP_PF_PEROUT:
+		break;
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+#define LAN8841_PTP_GPIO_NUM	10
+#define LAN8841_GPIO_EN		128
+#define LAN8841_GPIO_DIR	129
+#define LAN8841_GPIO_BUF	130
+
+static int lan8841_ptp_perout_off(struct kszphy_ptp_priv *ptp_priv, int pin)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	int ret;
+
+	ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_EN, BIT(pin));
+	if (ret)
+		return ret;
+
+	ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_DIR, BIT(pin));
+	if (ret)
+		return ret;
+
+	return phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_BUF, BIT(pin));
+}
+
+static int lan8841_ptp_perout_on(struct kszphy_ptp_priv *ptp_priv, int pin)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	int ret;
+
+	ret = phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_EN, BIT(pin));
+	if (ret)
+		return ret;
+
+	ret = phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_DIR, BIT(pin));
+	if (ret)
+		return ret;
+
+	return phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_BUF, BIT(pin));
+}
+
+#define LAN8841_GPIO_DATA_SEL1				131
+#define LAN8841_GPIO_DATA_SEL2				132
+#define LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_MASK	GENMASK(2, 0)
+#define LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_A	1
+#define LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_B	2
+#define LAN8841_PTP_GENERAL_CONFIG			257
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_A	BIT(1)
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_B	BIT(3)
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_A_MASK	GENMASK(7, 4)
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_B_MASK	GENMASK(11, 8)
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_A		4
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_B		7
+
+static int lan8841_ptp_remove_event(struct kszphy_ptp_priv *ptp_priv, int pin,
+				    u8 event)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	u16 tmp;
+	int ret;
+
+	/* Now remove pin from the event. GPIO_DATA_SEL1 contains the GPIO
+	 * pins 0-4 while GPIO_DATA_SEL2 contains GPIO pins 5-9, therefore
+	 * depending on the pin, it requires to read a different register
+	 */
+	if (pin < 5) {
+		tmp = LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_MASK << (3 * pin);
+		ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_DATA_SEL1, tmp);
+	} else {
+		tmp = LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_MASK << (3 * (pin - 5));
+		ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_DATA_SEL2, tmp);
+	}
+	if (ret)
+		return ret;
+
+	/* Disable the event */
+	if (event == LAN8841_EVENT_A)
+		tmp = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_A |
+		      LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_A_MASK;
+	else
+		tmp = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_B |
+		      LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_B_MASK;
+	return phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_EN, tmp);
+}
+
+static int lan8841_ptp_enable_event(struct kszphy_ptp_priv *ptp_priv, int pin,
+				    u8 event, int pulse_width)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	u16 tmp;
+	int ret;
+
+	/* Enable the event */
+	if (event == LAN8841_EVENT_A)
+		ret = phy_modify_mmd(phydev, 2, LAN8841_PTP_GENERAL_CONFIG,
+				     LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_A |
+				     LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_A_MASK,
+				     LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_A |
+				     pulse_width << LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_A);
+	else
+		ret = phy_modify_mmd(phydev, 2, LAN8841_PTP_GENERAL_CONFIG,
+				     LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_B |
+				     LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_B_MASK,
+				     LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_POL_B |
+				     pulse_width << LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_B);
+	if (ret)
+		return ret;
+
+	/* Now connect the pin to the event. GPIO_DATA_SEL1 contains the GPIO
+	 * pins 0-4 while GPIO_DATA_SEL2 contains GPIO pins 5-9, therefore
+	 * depending on the pin, it requires to read a different register
+	 */
+	if (event == LAN8841_EVENT_A)
+		tmp = LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_A;
+	else
+		tmp = LAN8841_GPIO_DATA_SEL_GPIO_DATA_SEL_EVENT_B;
+
+	if (pin < 5)
+		ret = phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_DATA_SEL1,
+				       tmp << (3 * pin));
+	else
+		ret = phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_DATA_SEL2,
+				       tmp << (3 * (pin - 5)));
+
+	return ret;
+}
+
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_200MS	13
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100MS	12
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_50MS	11
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_10MS	10
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_5MS	9
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_1MS	8
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_500US	7
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100US	6
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_50US	5
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_10US	4
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_5US	3
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_1US	2
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_500NS	1
+#define LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100NS	0
+
+static int lan8841_ptp_perout(struct ptp_clock_info *ptp,
+			      struct ptp_clock_request *rq, int on)
+{
+	struct kszphy_ptp_priv *ptp_priv = container_of(ptp, struct kszphy_ptp_priv,
+							ptp_clock_info);
+	struct phy_device *phydev = ptp_priv->phydev;
+	struct timespec64 ts_on, ts_period;
+	s64 on_nsec, period_nsec;
+	int pulse_width;
+	int pin;
+	int ret;
+
+	if (rq->perout.flags & ~PTP_PEROUT_DUTY_CYCLE)
+		return -EOPNOTSUPP;
+
+	pin = ptp_find_pin(ptp_priv->ptp_clock, PTP_PF_PEROUT, rq->perout.index);
+	if (pin == -1 || pin >= LAN8841_PTP_GPIO_NUM)
+		return -EINVAL;
+
+	if (!on) {
+		ret = lan8841_ptp_perout_off(ptp_priv, pin);
+		if (ret)
+			return ret;
+
+		return lan8841_ptp_remove_event(ptp_priv, LAN8841_EVENT_A, pin);
+	}
+
+	ts_on.tv_sec = rq->perout.on.sec;
+	ts_on.tv_nsec = rq->perout.on.nsec;
+	on_nsec = timespec64_to_ns(&ts_on);
+
+	ts_period.tv_sec = rq->perout.period.sec;
+	ts_period.tv_nsec = rq->perout.period.nsec;
+	period_nsec = timespec64_to_ns(&ts_period);
+
+	if (period_nsec < 200) {
+		pr_warn_ratelimited("%s: perout period too small, minimim is 200 nsec\n",
+				    phydev_name(phydev));
+		return -EOPNOTSUPP;
+	}
+
+	if (on_nsec >= period_nsec) {
+		pr_warn_ratelimited("%s: pulse width must be smaller than period\n",
+				    phydev_name(phydev));
+		return -EINVAL;
+	}
+
+	switch (on_nsec) {
+	case 200000000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_200MS;
+		break;
+	case 100000000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100MS;
+		break;
+	case 50000000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_50MS;
+		break;
+	case 10000000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_10MS;
+		break;
+	case 5000000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_5MS;
+		break;
+	case 1000000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_1MS;
+		break;
+	case 500000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_500US;
+		break;
+	case 100000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100US;
+		break;
+	case 50000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_50US;
+		break;
+	case 10000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_10US;
+		break;
+	case 5000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_5US;
+		break;
+	case 1000:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_1US;
+		break;
+	case 500:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_500NS;
+		break;
+	case 100:
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100NS;
+		break;
+	default:
+		pr_warn_ratelimited("%s: Use default duty cycle of 100ns\n",
+				    phydev_name(phydev));
+		pulse_width = LAN8841_PTP_GENERAL_CONFIG_LTC_EVENT_100NS;
+		break;
+	}
+
+	mutex_lock(&ptp_priv->ptp_lock);
+	ret = lan8841_ptp_set_target(ptp_priv, LAN8841_EVENT_A, rq->perout.start.sec,
+				     rq->perout.start.nsec);
+	mutex_unlock(&ptp_priv->ptp_lock);
+	if (ret)
+		return ret;
+
+	ret = lan8841_ptp_set_reload(ptp_priv, LAN8841_EVENT_A, rq->perout.period.sec,
+				     rq->perout.period.nsec);
+	if (ret)
+		return ret;
+
+	ret = lan8841_ptp_enable_event(ptp_priv, pin, LAN8841_EVENT_A,
+				       pulse_width);
+	if (ret)
+		return ret;
+
+	ret = lan8841_ptp_perout_on(ptp_priv, pin);
+	if (ret)
+		lan8841_ptp_remove_event(ptp_priv, pin, LAN8841_EVENT_A);
+
+	return ret;
+}
+
+static int lan8841_ptp_enable(struct ptp_clock_info *ptp,
+			      struct ptp_clock_request *rq, int on)
+{
+	switch (rq->type) {
+	case PTP_CLK_REQ_PEROUT:
+		return lan8841_ptp_perout(ptp, rq, on);
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
 static struct ptp_clock_info lan8841_ptp_clock_info = {
 	.owner		= THIS_MODULE,
 	.name		= "lan8841 ptp",
@@ -3847,6 +4214,10 @@ static struct ptp_clock_info lan8841_ptp_clock_info = {
 	.settime64	= lan8841_ptp_settime64,
 	.adjtime	= lan8841_ptp_adjtime,
 	.adjfine	= lan8841_ptp_adjfine,
+	.verify         = lan8841_ptp_verify,
+	.enable         = lan8841_ptp_enable,
+	.n_per_out      = LAN8841_PTP_GPIO_NUM,
+	.n_pins         = LAN8841_PTP_GPIO_NUM,
 };
 
 #define LAN8841_OPERATION_MODE_STRAP_LOW_REGISTER 3
@@ -3874,7 +4245,23 @@ static int lan8841_probe(struct phy_device *phydev)
 	priv = phydev->priv;
 	ptp_priv = &priv->ptp_priv;
 
+	ptp_priv->pin_config = devm_kcalloc(&phydev->mdio.dev,
+					    LAN8841_PTP_GPIO_NUM,
+					    sizeof(*ptp_priv->pin_config),
+					    GFP_KERNEL);
+	if (!ptp_priv->pin_config)
+		return -ENOMEM;
+
+	for (int i = 0; i < LAN8841_PTP_GPIO_NUM; ++i) {
+		struct ptp_pin_desc *p = &ptp_priv->pin_config[i];
+
+		snprintf(p->name, sizeof(p->name), "pin%d", i);
+		p->index = i;
+		p->func = PTP_PF_NONE;
+	}
+
 	ptp_priv->ptp_clock_info = lan8841_ptp_clock_info;
+	ptp_priv->ptp_clock_info.pin_config = ptp_priv->pin_config;
 	ptp_priv->ptp_clock = ptp_clock_register(&ptp_priv->ptp_clock_info,
 						 &phydev->mdio.dev);
 	if (IS_ERR(ptp_priv->ptp_clock)) {
-- 
cgit v1.2.3-70-g09d2


From c5a8027de26e1b7c0ba4f8e430165b6a4a29a869 Mon Sep 17 00:00:00 2001
From: Alexander Stein <alexander.stein@ew.tq-group.com>
Date: Fri, 10 Mar 2023 08:45:00 +0100
Subject: net: phy: dp83867: Disable IRQs on suspend

Before putting the PHY into IEEE power down mode, disable IRQs to
prevent accessing the PHY once MDIO has already been shutdown.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230310074500.3472858-1-alexander.stein@ew.tq-group.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/dp83867.c | 28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c
index 89cd821f1f46..5821f04c69dc 100644
--- a/drivers/net/phy/dp83867.c
+++ b/drivers/net/phy/dp83867.c
@@ -693,6 +693,30 @@ static int dp83867_of_init(struct phy_device *phydev)
 }
 #endif /* CONFIG_OF_MDIO */
 
+static int dp83867_suspend(struct phy_device *phydev)
+{
+	/* Disable PHY Interrupts */
+	if (phy_interrupt_is_valid(phydev)) {
+		phydev->interrupts = PHY_INTERRUPT_DISABLED;
+		dp83867_config_intr(phydev);
+	}
+
+	return genphy_suspend(phydev);
+}
+
+static int dp83867_resume(struct phy_device *phydev)
+{
+	/* Enable PHY Interrupts */
+	if (phy_interrupt_is_valid(phydev)) {
+		phydev->interrupts = PHY_INTERRUPT_ENABLED;
+		dp83867_config_intr(phydev);
+	}
+
+	genphy_resume(phydev);
+
+	return 0;
+}
+
 static int dp83867_probe(struct phy_device *phydev)
 {
 	struct dp83867_private *dp83867;
@@ -968,8 +992,8 @@ static struct phy_driver dp83867_driver[] = {
 		.config_intr	= dp83867_config_intr,
 		.handle_interrupt = dp83867_handle_interrupt,
 
-		.suspend	= genphy_suspend,
-		.resume		= genphy_resume,
+		.suspend	= dp83867_suspend,
+		.resume		= dp83867_resume,
 
 		.link_change_notify = dp83867_link_change_notify,
 		.set_loopback	= dp83867_loopback,
-- 
cgit v1.2.3-70-g09d2


From ad4bf5f2406f6a2e29266bbad74e18f0d955ac4c Mon Sep 17 00:00:00 2001
From: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
Date: Fri, 10 Mar 2023 23:18:51 +0100
Subject: net: socket: suppress unused warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

suppress unused warnings and fix the error that there is
with the W=1 enabled.

Warning generated

net/socket.c: In function ‘__sys_getsockopt’:
net/socket.c:2300:13: error: variable ‘max_optlen’ set but not used [-Werror=unused-but-set-variable]
 2300 |         int max_optlen;

Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230310221851.304657-1-vincenzopalazzodev@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/socket.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/socket.c b/net/socket.c
index 9c92c0e6c4da..73e493da4589 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2292,9 +2292,9 @@ INDIRECT_CALLABLE_DECLARE(bool tcp_bpf_bypass_getsockopt(int level,
 int __sys_getsockopt(int fd, int level, int optname, char __user *optval,
 		int __user *optlen)
 {
+	int max_optlen __maybe_unused;
 	int err, fput_needed;
 	struct socket *sock;
-	int max_optlen;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (!sock)
-- 
cgit v1.2.3-70-g09d2


From 90c7dd32652bcfcc7716c762bc095a7d49ad85dd Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Wed, 8 Mar 2023 21:34:13 +0100
Subject: net: phy: smsc: use device_property_present in smsc_phy_probe

Use unified device property API.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/a969f012-1d3b-7a36-51cf-89a5f8f15a9b@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/smsc.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index ac951d67edcc..9cfaccce1241 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -269,7 +269,6 @@ static void smsc_get_stats(struct phy_device *phydev,
 static int smsc_phy_probe(struct phy_device *phydev)
 {
 	struct device *dev = &phydev->mdio.dev;
-	struct device_node *of_node = dev->of_node;
 	struct smsc_phy_priv *priv;
 	struct clk *refclk;
 
@@ -279,7 +278,7 @@ static int smsc_phy_probe(struct phy_device *phydev)
 
 	priv->energy_enable = true;
 
-	if (of_property_read_bool(of_node, "smsc,disable-energy-detect"))
+	if (device_property_present(dev, "smsc,disable-energy-detect"))
 		priv->energy_enable = false;
 
 	phydev->priv = priv;
-- 
cgit v1.2.3-70-g09d2


From 4821c186b9c343760a293b455b603735291a7865 Mon Sep 17 00:00:00 2001
From: Kristian Overskeid <koverskeid@gmail.com>
Date: Thu, 9 Mar 2023 10:23:02 +0100
Subject: net: hsr: Don't log netdev_err message on unknown prp dst node

If no frames has been exchanged with a node for HSR_NODE_FORGET_TIME, the
node will be deleted from the node_db list. If a frame is sent to the node
after it is deleted, a netdev_err message for each slave interface is
produced. This should not happen with dan nodes because of supervision
frames, but can happen often with san nodes, which clutters the kernel
log. Since the hsr protocol does not support sans, this is only relevant
for the prp protocol.

Signed-off-by: Kristian Overskeid <koverskeid@gmail.com>
Link: https://lore.kernel.org/r/20230309092302.179586-1-koverskeid@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/hsr/hsr_framereg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c
index 00db74d96583..b77f1189d19d 100644
--- a/net/hsr/hsr_framereg.c
+++ b/net/hsr/hsr_framereg.c
@@ -415,7 +415,7 @@ void hsr_addr_subst_dest(struct hsr_node *node_src, struct sk_buff *skb,
 	node_dst = find_node_by_addr_A(&port->hsr->node_db,
 				       eth_hdr(skb)->h_dest);
 	if (!node_dst) {
-		if (net_ratelimit())
+		if (port->hsr->prot_version != PRP_V1 && net_ratelimit())
 			netdev_err(skb->dev, "%s: Unknown node\n", __func__);
 		return;
 	}
-- 
cgit v1.2.3-70-g09d2


From be50da3e9d4ad1958f7b11322d44d94d5c25a4c1 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@nvidia.com>
Date: Thu, 9 Mar 2023 10:45:59 +0100
Subject: net: virtio_net: implement exact header length guest feature

Virtio spec introduced a feature VIRTIO_NET_F_GUEST_HDRLEN which when
set implicates that device benefits from knowing the exact size
of the header. For compatibility, to signal to the device that
the header is reliable driver also needs to set this feature.
Without this feature set by driver, device has to figure
out the header size itself.

Quoting the original virtio spec:
"hdr_len is a hint to the device as to how much of the header needs to
 be kept to copy into each packet"

"a hint" might not be clear for the reader what does it mean, if it is
"maybe like that" of "exactly like that". This feature just makes it
crystal clear and let the device count on the hdr_len being filled up
by the exact length of header.

Also note the spec already has following note about hdr_len:
"Due to various bugs in implementations, this field is not useful
 as a guarantee of the transport header size."

Without this feature the device needs to parse the header in core
data path handling. Accurate information helps the device to eliminate
such header parsing and directly use the hardware accelerators
for GSO operation.

virtio_net_hdr_from_skb() fills up hdr_len to skb_headlen(skb).
The driver already complies to fill the correct value. Introduce the
feature and advertise it.

Note that virtio spec also includes following note for device
implementation:
"Caution should be taken by the implementation so as to prevent
 a malicious driver from attacking the device by setting
 an incorrect hdr_len."

There is a plan to support this feature in our emulated device.
A device of SolidRun offers this feature bit. They claim this feature
will save the device a few cycles for every GSO packet.

Link: https://docs.oasis-open.org/virtio/virtio/v1.2/cs01/virtio-v1.2-cs01.html#x1-230006x3
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Reviewed-by: Alvaro Karsz <alvaro.karsz@solid-run.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20230309094559.917857-1-jiri@resnulli.us
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c        | 6 ++++--
 include/uapi/linux/virtio_net.h | 1 +
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index fb5e68ed3ec2..e85b03988733 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -62,7 +62,8 @@ static const unsigned long guest_offloads[] = {
 	VIRTIO_NET_F_GUEST_UFO,
 	VIRTIO_NET_F_GUEST_CSUM,
 	VIRTIO_NET_F_GUEST_USO4,
-	VIRTIO_NET_F_GUEST_USO6
+	VIRTIO_NET_F_GUEST_USO6,
+	VIRTIO_NET_F_GUEST_HDRLEN
 };
 
 #define GUEST_OFFLOAD_GRO_HW_MASK ((1ULL << VIRTIO_NET_F_GUEST_TSO4) | \
@@ -4213,7 +4214,8 @@ static struct virtio_device_id id_table[] = {
 	VIRTIO_NET_F_CTRL_MAC_ADDR, \
 	VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS, \
 	VIRTIO_NET_F_SPEED_DUPLEX, VIRTIO_NET_F_STANDBY, \
-	VIRTIO_NET_F_RSS, VIRTIO_NET_F_HASH_REPORT, VIRTIO_NET_F_NOTF_COAL
+	VIRTIO_NET_F_RSS, VIRTIO_NET_F_HASH_REPORT, VIRTIO_NET_F_NOTF_COAL, \
+	VIRTIO_NET_F_GUEST_HDRLEN
 
 static unsigned int features[] = {
 	VIRTNET_FEATURES,
diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h
index b4062bed186a..12c1c9699935 100644
--- a/include/uapi/linux/virtio_net.h
+++ b/include/uapi/linux/virtio_net.h
@@ -61,6 +61,7 @@
 #define VIRTIO_NET_F_GUEST_USO6	55	/* Guest can handle USOv6 in. */
 #define VIRTIO_NET_F_HOST_USO	56	/* Host can handle USO in. */
 #define VIRTIO_NET_F_HASH_REPORT  57	/* Supports hash report */
+#define VIRTIO_NET_F_GUEST_HDRLEN  59	/* Guest provides the exact hdr_len value. */
 #define VIRTIO_NET_F_RSS	  60	/* Supports RSS RX steering */
 #define VIRTIO_NET_F_RSC_EXT	  61	/* extended coalescing info */
 #define VIRTIO_NET_F_STANDBY	  62	/* Act as standby for another device
-- 
cgit v1.2.3-70-g09d2


From 7c6dddc239abe660598c49ec95ea0ed6399a4b2a Mon Sep 17 00:00:00 2001
From: Maxim Korotkov <korotkov.maxim.s@gmail.com>
Date: Thu, 9 Mar 2023 20:43:47 +0300
Subject: bnxt: avoid overflow in bnxt_get_nvram_directory()

The value of an arithmetic expression is subject
of possible overflow due to a failure to cast operands to a larger data
type before performing arithmetic. Used macro for multiplication instead
operator for avoiding overflow.

Found by Security Code and Linux Verification
Center (linuxtesting.org) with SVACE.

Signed-off-by: Maxim Korotkov <korotkov.maxim.s@gmail.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://lore.kernel.org/r/20230309174347.3515-1-korotkov.maxim.s@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index ec573127b707..696f32dfe41f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -2862,7 +2862,7 @@ static int bnxt_get_nvram_directory(struct net_device *dev, u32 len, u8 *data)
 	if (rc)
 		return rc;
 
-	buflen = dir_entries * entry_length;
+	buflen = mul_u32_u32(dir_entries, entry_length);
 	buf = hwrm_req_dma_slice(bp, req, buflen, &dma_handle);
 	if (!buf) {
 		hwrm_req_drop(bp, req);
-- 
cgit v1.2.3-70-g09d2


From c66b2111c9c952f3bbf454a755768e80308cc6e2 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Thu, 9 Mar 2023 14:55:54 -0300
Subject: selftests: tc-testing: add tests for action binding

Add tests that check if filters can bind actions, that is create an
action independently and then bind to a filter.

tdc-tests under category 'infra':
1..18
ok 1 abdc - Reference pedit action object in filter
ok 2 7a70 - Reference mpls action object in filter
ok 3 d241 - Reference bpf action object in filter
ok 4 383a - Reference connmark action object in filter
ok 5 c619 - Reference csum action object in filter
ok 6 a93d - Reference ct action object in filter
ok 7 8bb5 - Reference ctinfo action object in filter
ok 8 2241 - Reference gact action object in filter
ok 9 35e9 - Reference gate action object in filter
ok 10 b22e - Reference ife action object in filter
ok 11 ef74 - Reference mirred action object in filter
ok 12 2c81 - Reference nat action object in filter
ok 13 ac9d - Reference police action object in filter
ok 14 68be - Reference sample action object in filter
ok 15 cf01 - Reference skbedit action object in filter
ok 16 c109 - Reference skbmod action object in filter
ok 17 4abc - Reference tunnel_key action object in filter
ok 18 dadd - Reference vlan action object in filter

Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Link: https://lore.kernel.org/r/20230309175554.304824-1-pctammela@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../tc-testing/tc-tests/infra/actions.json         | 416 +++++++++++++++++++++
 1 file changed, 416 insertions(+)
 create mode 100644 tools/testing/selftests/tc-testing/tc-tests/infra/actions.json

diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json b/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json
new file mode 100644
index 000000000000..16f3a83605e4
--- /dev/null
+++ b/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json
@@ -0,0 +1,416 @@
+[
+    {
+        "id": "abdc",
+        "name": "Reference pedit action object in filter",
+        "category": [
+            "infra",
+            "pedit"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC action add action pedit munge offset 0 u8 clear index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action pedit index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action pedit"
+        ]
+    },
+    {
+        "id": "7a70",
+        "name": "Reference mpls action object in filter",
+        "category": [
+            "infra",
+            "mpls"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC action add action mpls pop protocol ipv4 index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action mpls index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action mpls"
+        ]
+    },
+    {
+        "id": "d241",
+        "name": "Reference bpf action object in filter",
+        "category": [
+            "infra",
+            "bpf"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC action add action bpf bytecode '4,40 0 0 12,21 0 1 2048,6 0 0 262144,6 0 0 0' index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action bpf index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action bpf"
+        ]
+    },
+    {
+        "id": "383a",
+        "name": "Reference connmark action object in filter",
+        "category": [
+            "infra",
+            "connmark"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action connmark"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action connmark index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action connmark"
+        ]
+    },
+    {
+        "id": "c619",
+        "name": "Reference csum action object in filter",
+        "category": [
+            "infra",
+            "csum"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action csum ip4h index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action csum index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action csum"
+        ]
+    },
+    {
+        "id": "a93d",
+        "name": "Reference ct action object in filter",
+        "category": [
+            "infra",
+            "ct"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action ct index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action ct index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action ct"
+        ]
+    },
+    {
+        "id": "8bb5",
+        "name": "Reference ctinfo action object in filter",
+        "category": [
+            "infra",
+            "ctinfo"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC action add action ctinfo index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action ctinfo index 10",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action ctinfo"
+        ]
+    },
+    {
+        "id": "2241",
+        "name": "Reference gact action object in filter",
+        "category": [
+            "infra",
+            "gact"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action pass index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action gact index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action gact"
+        ]
+    },
+    {
+        "id": "35e9",
+        "name": "Reference gate action object in filter",
+        "category": [
+            "infra",
+            "gate"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC action add action gate priority 1 sched-entry close 100000000ns index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action gate index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action gate"
+        ]
+    },
+    {
+        "id": "b22e",
+        "name": "Reference ife action object in filter",
+        "category": [
+            "infra",
+            "ife"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action ife encode allow mark pass index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action ife index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action ife"
+        ]
+    },
+    {
+        "id": "ef74",
+        "name": "Reference mirred action object in filter",
+        "category": [
+            "infra",
+            "mirred"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action mirred egress mirror index 1 dev lo"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action mirred index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action mirred"
+        ]
+    },
+    {
+        "id": "2c81",
+        "name": "Reference nat action object in filter",
+        "category": [
+            "infra",
+            "nat"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action nat ingress 192.168.1.1 200.200.200.1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action nat index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action nat"
+        ]
+    },
+    {
+        "id": "ac9d",
+        "name": "Reference police action object in filter",
+        "category": [
+            "infra",
+            "police"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action police rate 1kbit burst 10k index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action police index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action police"
+        ]
+    },
+    {
+        "id": "68be",
+        "name": "Reference sample action object in filter",
+        "category": [
+            "infra",
+            "sample"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action sample rate 10 group 1 index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action sample index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action sample"
+        ]
+    },
+    {
+        "id": "cf01",
+        "name": "Reference skbedit action object in filter",
+        "category": [
+            "infra",
+            "skbedit"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action skbedit mark 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action skbedit index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action skbedit"
+        ]
+    },
+    {
+        "id": "c109",
+        "name": "Reference skbmod action object in filter",
+        "category": [
+            "infra",
+            "skbmod"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action skbmod set dmac 11:22:33:44:55:66 index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action skbmod index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action skbmod"
+        ]
+    },
+    {
+        "id": "4abc",
+        "name": "Reference tunnel_key action object in filter",
+        "category": [
+            "infra",
+            "tunnel_key"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action tunnel_key set src_ip 10.10.10.1 dst_ip 20.20.20.2 id 1 index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action tunnel_key index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action tunnel_key"
+        ]
+    },
+    {
+        "id": "dadd",
+        "name": "Reference vlan action object in filter",
+        "category": [
+            "infra",
+            "tunnel_key"
+        ],
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY ingress",
+            "$TC actions add action vlan pop pipe index 1"
+        ],
+        "cmdUnderTest": "$TC filter add dev $DUMMY parent ffff: handle 0x1 prio 1 protocol ip matchall action vlan index 1",
+        "expExitCode": "0",
+        "verifyCmd": "$TC filter get dev $DUMMY parent ffff: handle 1 prio 1 protocol ip matchall",
+        "matchPattern": "^filter parent ffff: protocol ip pref 1 matchall.*handle 0x1.*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY ingress",
+            "$IP link del dev $DUMMY type dummy",
+            "$TC actions flush action vlan"
+        ]
+    }
+]
-- 
cgit v1.2.3-70-g09d2


From 9e36a204bd43553a9cd4bd574612cd9a5df791ea Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Mon, 13 Mar 2023 14:46:41 -0700
Subject: bpf: Disable migration when freeing stashed local kptr using obj drop

When a local kptr is stashed in a map and freed when the map goes away,
currently an error like the below appears:

[   39.195695] BUG: using smp_processor_id() in preemptible [00000000] code: kworker/u32:15/2875
[   39.196549] caller is bpf_mem_free+0x56/0xc0
[   39.196958] CPU: 15 PID: 2875 Comm: kworker/u32:15 Tainted: G           O       6.2.0-13016-g22df776a9a86 #4477
[   39.197897] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
[   39.198949] Workqueue: events_unbound bpf_map_free_deferred
[   39.199470] Call Trace:
[   39.199703]  <TASK>
[   39.199911]  dump_stack_lvl+0x60/0x70
[   39.200267]  check_preemption_disabled+0xbf/0xe0
[   39.200704]  bpf_mem_free+0x56/0xc0
[   39.201032]  ? bpf_obj_new_impl+0xa0/0xa0
[   39.201430]  bpf_obj_free_fields+0x1cd/0x200
[   39.201838]  array_map_free+0xad/0x220
[   39.202193]  ? finish_task_switch+0xe5/0x3c0
[   39.202614]  bpf_map_free_deferred+0xea/0x210
[   39.203006]  ? lockdep_hardirqs_on_prepare+0xe/0x220
[   39.203460]  process_one_work+0x64f/0xbe0
[   39.203822]  ? pwq_dec_nr_in_flight+0x110/0x110
[   39.204264]  ? do_raw_spin_lock+0x107/0x1c0
[   39.204662]  ? lockdep_hardirqs_on_prepare+0xe/0x220
[   39.205107]  worker_thread+0x74/0x7a0
[   39.205451]  ? process_one_work+0xbe0/0xbe0
[   39.205818]  kthread+0x171/0x1a0
[   39.206111]  ? kthread_complete_and_exit+0x20/0x20
[   39.206552]  ret_from_fork+0x1f/0x30
[   39.206886]  </TASK>

This happens because the call to __bpf_obj_drop_impl I added in the patch
adding support for stashing local kptrs doesn't disable migration. Prior
to that patch, __bpf_obj_drop_impl logic only ran when called by a BPF
progarm, whereas now it can be called from map free path, so it's
necessary to explicitly disable migration.

Also, refactor a bit to just call __bpf_obj_drop_impl directly instead
of bothering w/ dtor union and setting pointer-to-obj_drop.

Fixes: c8e187540914 ("bpf: Support __kptr to local kptrs")
Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230313214641.3731908-1-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  | 12 ++++--------
 kernel/bpf/btf.c     |  4 +---
 kernel/bpf/syscall.c | 10 +++++++---
 3 files changed, 12 insertions(+), 14 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 756b85f0d0d3..71cc92a4ba48 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -190,18 +190,14 @@ enum btf_field_type {
 };
 
 typedef void (*btf_dtor_kfunc_t)(void *);
-typedef void (*btf_dtor_obj_drop)(void *, const struct btf_record *);
 
 struct btf_field_kptr {
 	struct btf *btf;
 	struct module *module;
-	union {
-		/* dtor used if btf_is_kernel(btf), otherwise the type
-		 * is program-allocated and obj_drop is used
-		 */
-		btf_dtor_kfunc_t dtor;
-		btf_dtor_obj_drop obj_drop;
-	};
+	/* dtor used if btf_is_kernel(btf), otherwise the type is
+	 * program-allocated, dtor is NULL,  and __bpf_obj_drop_impl is used
+	 */
+	btf_dtor_kfunc_t dtor;
 	u32 btf_id;
 };
 
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 66fad7a16b6c..b7e5a5510b91 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3551,8 +3551,6 @@ static int btf_find_field(const struct btf *btf, const struct btf_type *t,
 	return -EINVAL;
 }
 
-extern void __bpf_obj_drop_impl(void *p, const struct btf_record *rec);
-
 static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 			  struct btf_field_info *info)
 {
@@ -3578,7 +3576,7 @@ static int btf_parse_kptr(const struct btf *btf, struct btf_field *field,
 		/* Type exists only in program BTF. Assume that it's a MEM_ALLOC
 		 * kptr allocated via bpf_obj_new
 		 */
-		field->kptr.dtor = (void *)&__bpf_obj_drop_impl;
+		field->kptr.dtor = NULL;
 		id = info->kptr.type_id;
 		kptr_btf = (struct btf *)btf;
 		btf_get(kptr_btf);
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 0684febc447a..5b88301a2ae0 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -650,6 +650,8 @@ void bpf_obj_free_timer(const struct btf_record *rec, void *obj)
 	bpf_timer_cancel_and_free(obj + rec->timer_off);
 }
 
+extern void __bpf_obj_drop_impl(void *p, const struct btf_record *rec);
+
 void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 {
 	const struct btf_field *fields;
@@ -679,9 +681,11 @@ void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 				pointee_struct_meta = btf_find_struct_meta(field->kptr.btf,
 									   field->kptr.btf_id);
 				WARN_ON_ONCE(!pointee_struct_meta);
-				field->kptr.obj_drop(xchgd_field, pointee_struct_meta ?
-								  pointee_struct_meta->record :
-								  NULL);
+				migrate_disable();
+				__bpf_obj_drop_impl(xchgd_field, pointee_struct_meta ?
+								 pointee_struct_meta->record :
+								 NULL);
+				migrate_enable();
 			} else {
 				field->kptr.dtor(xchgd_field);
 			}
-- 
cgit v1.2.3-70-g09d2


From bcc858689db5f2e5a8d4d6e8bc5bb9736cd80626 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Fri, 10 Mar 2023 08:47:16 -0600
Subject: net: Use of_property_present() for testing DT property presence

It is preferred to use typed property access functions (i.e.
of_property_read_<type> functions) rather than low-level
of_get_property/of_find_property functions for reading properties. As
part of this, convert of_get_property/of_find_property calls to the
recently added of_property_present() helper when we just want to test
for presence of a property and nothing more.

Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230310144716.1544083-1-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/mdio/of_mdio.c                            | 4 ++--
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/mdio/of_mdio.c b/drivers/net/mdio/of_mdio.c
index 510822d6d0d9..bf10d0688eea 100644
--- a/drivers/net/mdio/of_mdio.c
+++ b/drivers/net/mdio/of_mdio.c
@@ -131,7 +131,7 @@ bool of_mdiobus_child_is_phy(struct device_node *child)
 		return true;
 	}
 
-	if (!of_find_property(child, "compatible", NULL))
+	if (!of_property_present(child, "compatible"))
 		return true;
 
 	return false;
@@ -203,7 +203,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 	/* auto scan for PHYs with empty reg property */
 	for_each_available_child_of_node(np, child) {
 		/* Skip PHYs with reg property set */
-		if (of_find_property(child, "reg", NULL))
+		if (of_property_present(child, "reg"))
 			continue;
 
 		for (addr = 0; addr < PHY_MAX_ADDR; addr++) {
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index 52527b61341e..e406e11481a6 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -129,7 +129,7 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
 		sdio->drive_strength = val;
 
 	/* make sure there are interrupts defined in the node */
-	if (!of_find_property(np, "interrupts", NULL))
+	if (!of_property_present(np, "interrupts"))
 		return;
 
 	irq = irq_of_parse_and_map(np, 0);
-- 
cgit v1.2.3-70-g09d2


From 27d7fdf06fdb84455ff585b58c8034e2fab42583 Mon Sep 17 00:00:00 2001
From: Ross Zwisler <zwisler@google.com>
Date: Mon, 13 Mar 2023 14:56:27 -0600
Subject: bpf: use canonical ftrace path

The canonical location for the tracefs filesystem is at /sys/kernel/tracing.

But, from Documentation/trace/ftrace.rst:

  Before 4.1, all ftrace tracing control files were within the debugfs
  file system, which is typically located at /sys/kernel/debug/tracing.
  For backward compatibility, when mounting the debugfs file system,
  the tracefs file system will be automatically mounted at:

  /sys/kernel/debug/tracing

Many comments and samples in the bpf code still refer to this older
debugfs path, so let's update them to avoid confusion.  There are a few
spots where the bpf code explicitly checks both tracefs and debugfs
(tools/bpf/bpftool/tracelog.c and tools/lib/api/fs/fs.c) and I've left
those alone so that the tools can continue to work with both paths.

Signed-off-by: Ross Zwisler <zwisler@google.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20230313205628.1058720-2-zwisler@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/uapi/linux/bpf.h           | 8 ++++----
 samples/bpf/cpustat_kern.c         | 4 ++--
 samples/bpf/hbm.c                  | 4 ++--
 samples/bpf/ibumad_kern.c          | 4 ++--
 samples/bpf/lwt_len_hist.sh        | 2 +-
 samples/bpf/offwaketime_kern.c     | 2 +-
 samples/bpf/task_fd_query_user.c   | 4 ++--
 samples/bpf/test_lwt_bpf.sh        | 2 +-
 samples/bpf/test_overhead_tp.bpf.c | 4 ++--
 tools/include/uapi/linux/bpf.h     | 8 ++++----
 10 files changed, 21 insertions(+), 21 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index d8c534e05b0a..13129df937cd 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1647,17 +1647,17 @@ union bpf_attr {
  * 	Description
  * 		This helper is a "printk()-like" facility for debugging. It
  * 		prints a message defined by format *fmt* (of size *fmt_size*)
- * 		to file *\/sys/kernel/debug/tracing/trace* from DebugFS, if
+ * 		to file *\/sys/kernel/tracing/trace* from TraceFS, if
  * 		available. It can take up to three additional **u64**
  * 		arguments (as an eBPF helpers, the total number of arguments is
  * 		limited to five).
  *
  * 		Each time the helper is called, it appends a line to the trace.
- * 		Lines are discarded while *\/sys/kernel/debug/tracing/trace* is
- * 		open, use *\/sys/kernel/debug/tracing/trace_pipe* to avoid this.
+ * 		Lines are discarded while *\/sys/kernel/tracing/trace* is
+ * 		open, use *\/sys/kernel/tracing/trace_pipe* to avoid this.
  * 		The format of the trace is customizable, and the exact output
  * 		one will get depends on the options set in
- * 		*\/sys/kernel/debug/tracing/trace_options* (see also the
+ * 		*\/sys/kernel/tracing/trace_options* (see also the
  * 		*README* file under the same directory). However, it usually
  * 		defaults to something like:
  *
diff --git a/samples/bpf/cpustat_kern.c b/samples/bpf/cpustat_kern.c
index 5aefd19cdfa1..944f13fe164a 100644
--- a/samples/bpf/cpustat_kern.c
+++ b/samples/bpf/cpustat_kern.c
@@ -76,8 +76,8 @@ struct {
 
 /*
  * The trace events for cpu_idle and cpu_frequency are taken from:
- * /sys/kernel/debug/tracing/events/power/cpu_idle/format
- * /sys/kernel/debug/tracing/events/power/cpu_frequency/format
+ * /sys/kernel/tracing/events/power/cpu_idle/format
+ * /sys/kernel/tracing/events/power/cpu_frequency/format
  *
  * These two events have same format, so define one common structure.
  */
diff --git a/samples/bpf/hbm.c b/samples/bpf/hbm.c
index 516fbac28b71..ff58ec43f56a 100644
--- a/samples/bpf/hbm.c
+++ b/samples/bpf/hbm.c
@@ -65,7 +65,7 @@ static void Usage(void);
 static void read_trace_pipe2(void);
 static void do_error(char *msg, bool errno_flag);
 
-#define DEBUGFS "/sys/kernel/debug/tracing/"
+#define TRACEFS "/sys/kernel/tracing/"
 
 static struct bpf_program *bpf_prog;
 static struct bpf_object *obj;
@@ -77,7 +77,7 @@ static void read_trace_pipe2(void)
 	FILE *outf;
 	char *outFname = "hbm_out.log";
 
-	trace_fd = open(DEBUGFS "trace_pipe", O_RDONLY, 0);
+	trace_fd = open(TRACEFS "trace_pipe", O_RDONLY, 0);
 	if (trace_fd < 0) {
 		printf("Error opening trace_pipe\n");
 		return;
diff --git a/samples/bpf/ibumad_kern.c b/samples/bpf/ibumad_kern.c
index 9b193231024a..f07474c72525 100644
--- a/samples/bpf/ibumad_kern.c
+++ b/samples/bpf/ibumad_kern.c
@@ -39,8 +39,8 @@ struct {
 /* Taken from the current format defined in
  * include/trace/events/ib_umad.h
  * and
- * /sys/kernel/debug/tracing/events/ib_umad/ib_umad_read/format
- * /sys/kernel/debug/tracing/events/ib_umad/ib_umad_write/format
+ * /sys/kernel/tracing/events/ib_umad/ib_umad_read/format
+ * /sys/kernel/tracing/events/ib_umad/ib_umad_write/format
  */
 struct ib_umad_rw_args {
 	u64 pad;
diff --git a/samples/bpf/lwt_len_hist.sh b/samples/bpf/lwt_len_hist.sh
index 7078bfcc4f4d..381b2c634784 100755
--- a/samples/bpf/lwt_len_hist.sh
+++ b/samples/bpf/lwt_len_hist.sh
@@ -5,7 +5,7 @@ NS1=lwt_ns1
 VETH0=tst_lwt1a
 VETH1=tst_lwt1b
 BPF_PROG=lwt_len_hist.bpf.o
-TRACE_ROOT=/sys/kernel/debug/tracing
+TRACE_ROOT=/sys/kernel/tracing
 
 function cleanup {
 	# To reset saved histogram, remove pinned map
diff --git a/samples/bpf/offwaketime_kern.c b/samples/bpf/offwaketime_kern.c
index eb4d94742e6b..23f12b47e9e5 100644
--- a/samples/bpf/offwaketime_kern.c
+++ b/samples/bpf/offwaketime_kern.c
@@ -110,7 +110,7 @@ static inline int update_counts(void *ctx, u32 pid, u64 delta)
 }
 
 #if 1
-/* taken from /sys/kernel/debug/tracing/events/sched/sched_switch/format */
+/* taken from /sys/kernel/tracing/events/sched/sched_switch/format */
 struct sched_switch_args {
 	unsigned long long pad;
 	char prev_comm[TASK_COMM_LEN];
diff --git a/samples/bpf/task_fd_query_user.c b/samples/bpf/task_fd_query_user.c
index a33d74bd3a4b..1e61f2180470 100644
--- a/samples/bpf/task_fd_query_user.c
+++ b/samples/bpf/task_fd_query_user.c
@@ -235,7 +235,7 @@ static int test_debug_fs_uprobe(char *binary_path, long offset, bool is_return)
 	struct bpf_link *link;
 	ssize_t bytes;
 
-	snprintf(buf, sizeof(buf), "/sys/kernel/debug/tracing/%s_events",
+	snprintf(buf, sizeof(buf), "/sys/kernel/tracing/%s_events",
 		 event_type);
 	kfd = open(buf, O_WRONLY | O_TRUNC, 0);
 	CHECK_PERROR_RET(kfd < 0);
@@ -252,7 +252,7 @@ static int test_debug_fs_uprobe(char *binary_path, long offset, bool is_return)
 	close(kfd);
 	kfd = -1;
 
-	snprintf(buf, sizeof(buf), "/sys/kernel/debug/tracing/events/%ss/%s/id",
+	snprintf(buf, sizeof(buf), "/sys/kernel/tracing/events/%ss/%s/id",
 		 event_type, event_alias);
 	efd = open(buf, O_RDONLY, 0);
 	CHECK_PERROR_RET(efd < 0);
diff --git a/samples/bpf/test_lwt_bpf.sh b/samples/bpf/test_lwt_bpf.sh
index 2e9f5126963b..0bf2d0f6bf4b 100755
--- a/samples/bpf/test_lwt_bpf.sh
+++ b/samples/bpf/test_lwt_bpf.sh
@@ -21,7 +21,7 @@ IP_LOCAL="192.168.99.1"
 
 PROG_SRC="test_lwt_bpf.c"
 BPF_PROG="test_lwt_bpf.o"
-TRACE_ROOT=/sys/kernel/debug/tracing
+TRACE_ROOT=/sys/kernel/tracing
 CONTEXT_INFO=$(cat ${TRACE_ROOT}/trace_options | grep context)
 
 function lookup_mac()
diff --git a/samples/bpf/test_overhead_tp.bpf.c b/samples/bpf/test_overhead_tp.bpf.c
index 67cab3881969..8b498328e961 100644
--- a/samples/bpf/test_overhead_tp.bpf.c
+++ b/samples/bpf/test_overhead_tp.bpf.c
@@ -7,7 +7,7 @@
 #include "vmlinux.h"
 #include <bpf/bpf_helpers.h>
 
-/* from /sys/kernel/debug/tracing/events/task/task_rename/format */
+/* from /sys/kernel/tracing/events/task/task_rename/format */
 struct task_rename {
 	__u64 pad;
 	__u32 pid;
@@ -21,7 +21,7 @@ int prog(struct task_rename *ctx)
 	return 0;
 }
 
-/* from /sys/kernel/debug/tracing/events/fib/fib_table_lookup/format */
+/* from /sys/kernel/tracing/events/fib/fib_table_lookup/format */
 struct fib_table_lookup {
 	__u64 pad;
 	__u32 tb_id;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index d8c534e05b0a..13129df937cd 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1647,17 +1647,17 @@ union bpf_attr {
  * 	Description
  * 		This helper is a "printk()-like" facility for debugging. It
  * 		prints a message defined by format *fmt* (of size *fmt_size*)
- * 		to file *\/sys/kernel/debug/tracing/trace* from DebugFS, if
+ * 		to file *\/sys/kernel/tracing/trace* from TraceFS, if
  * 		available. It can take up to three additional **u64**
  * 		arguments (as an eBPF helpers, the total number of arguments is
  * 		limited to five).
  *
  * 		Each time the helper is called, it appends a line to the trace.
- * 		Lines are discarded while *\/sys/kernel/debug/tracing/trace* is
- * 		open, use *\/sys/kernel/debug/tracing/trace_pipe* to avoid this.
+ * 		Lines are discarded while *\/sys/kernel/tracing/trace* is
+ * 		open, use *\/sys/kernel/tracing/trace_pipe* to avoid this.
  * 		The format of the trace is customizable, and the exact output
  * 		one will get depends on the options set in
- * 		*\/sys/kernel/debug/tracing/trace_options* (see also the
+ * 		*\/sys/kernel/tracing/trace_options* (see also the
  * 		*README* file under the same directory). However, it usually
  * 		defaults to something like:
  *
-- 
cgit v1.2.3-70-g09d2


From ab4c15feb2ebcf9f4abe31457d7cbc8f3de9c2ab Mon Sep 17 00:00:00 2001
From: Ross Zwisler <zwisler@google.com>
Date: Mon, 13 Mar 2023 14:56:28 -0600
Subject: selftests/bpf: use canonical ftrace path

The canonical location for the tracefs filesystem is at
/sys/kernel/tracing.

But, from Documentation/trace/ftrace.rst:

  Before 4.1, all ftrace tracing control files were within the debugfs
  file system, which is typically located at /sys/kernel/debug/tracing.
  For backward compatibility, when mounting the debugfs file system,
  the tracefs file system will be automatically mounted at:

  /sys/kernel/debug/tracing

Many tests in the bpf selftest code still refer to this older debugfs
path, so let's update them to avoid confusion.

Signed-off-by: Ross Zwisler <zwisler@google.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20230313205628.1058720-3-zwisler@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/get_cgroup_id_user.c           |  9 +++++++--
 tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c |  7 ++++++-
 tools/testing/selftests/bpf/prog_tests/task_fd_query_tp.c  |  9 +++++++--
 tools/testing/selftests/bpf/prog_tests/tp_attach_query.c   |  9 +++++++--
 tools/testing/selftests/bpf/prog_tests/trace_printk.c      | 10 +++++++---
 tools/testing/selftests/bpf/prog_tests/trace_vprintk.c     | 10 +++++++---
 tools/testing/selftests/bpf/progs/test_stacktrace_map.c    |  2 +-
 tools/testing/selftests/bpf/progs/test_tracepoint.c        |  2 +-
 tools/testing/selftests/bpf/test_ftrace.sh                 |  7 ++++++-
 tools/testing/selftests/bpf/test_tunnel.sh                 | 13 +++++++++----
 tools/testing/selftests/bpf/trace_helpers.c                |  8 ++++++--
 11 files changed, 64 insertions(+), 22 deletions(-)

diff --git a/tools/testing/selftests/bpf/get_cgroup_id_user.c b/tools/testing/selftests/bpf/get_cgroup_id_user.c
index 156743cf5870..aefd83ebdcd7 100644
--- a/tools/testing/selftests/bpf/get_cgroup_id_user.c
+++ b/tools/testing/selftests/bpf/get_cgroup_id_user.c
@@ -86,8 +86,13 @@ int main(int argc, char **argv)
 	pid = getpid();
 	bpf_map_update_elem(pidmap_fd, &key, &pid, 0);
 
-	snprintf(buf, sizeof(buf),
-		 "/sys/kernel/debug/tracing/events/%s/id", probe_name);
+	if (access("/sys/kernel/tracing/trace", F_OK) == 0) {
+		snprintf(buf, sizeof(buf),
+			 "/sys/kernel/tracing/events/%s/id", probe_name);
+	} else {
+		snprintf(buf, sizeof(buf),
+			 "/sys/kernel/debug/tracing/events/%s/id", probe_name);
+	}
 	efd = open(buf, O_RDONLY, 0);
 	if (CHECK(efd < 0, "open", "err %d errno %d\n", efd, errno))
 		goto close_prog;
diff --git a/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c b/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c
index 113dba349a57..22be0a9a5a0a 100644
--- a/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c
@@ -338,7 +338,12 @@ static int get_syms(char ***symsp, size_t *cntp, bool kernel)
 	 * Filtering out duplicates by using hashmap__add, which won't
 	 * add existing entry.
 	 */
-	f = fopen("/sys/kernel/debug/tracing/available_filter_functions", "r");
+
+	if (access("/sys/kernel/tracing/trace", F_OK) == 0)
+		f = fopen("/sys/kernel/tracing/available_filter_functions", "r");
+	else
+		f = fopen("/sys/kernel/debug/tracing/available_filter_functions", "r");
+
 	if (!f)
 		return -EINVAL;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/task_fd_query_tp.c b/tools/testing/selftests/bpf/prog_tests/task_fd_query_tp.c
index c717741bf8b6..c91eda624657 100644
--- a/tools/testing/selftests/bpf/prog_tests/task_fd_query_tp.c
+++ b/tools/testing/selftests/bpf/prog_tests/task_fd_query_tp.c
@@ -17,8 +17,13 @@ static void test_task_fd_query_tp_core(const char *probe_name,
 	if (CHECK(err, "bpf_prog_test_load", "err %d errno %d\n", err, errno))
 		goto close_prog;
 
-	snprintf(buf, sizeof(buf),
-		 "/sys/kernel/debug/tracing/events/%s/id", probe_name);
+	if (access("/sys/kernel/tracing/trace", F_OK) == 0) {
+		snprintf(buf, sizeof(buf),
+			 "/sys/kernel/tracing/events/%s/id", probe_name);
+	} else {
+		snprintf(buf, sizeof(buf),
+			 "/sys/kernel/debug/tracing/events/%s/id", probe_name);
+	}
 	efd = open(buf, O_RDONLY, 0);
 	if (CHECK(efd < 0, "open", "err %d errno %d\n", efd, errno))
 		goto close_prog;
diff --git a/tools/testing/selftests/bpf/prog_tests/tp_attach_query.c b/tools/testing/selftests/bpf/prog_tests/tp_attach_query.c
index 770fcc3bb1ba..655d69f0ff0b 100644
--- a/tools/testing/selftests/bpf/prog_tests/tp_attach_query.c
+++ b/tools/testing/selftests/bpf/prog_tests/tp_attach_query.c
@@ -16,8 +16,13 @@ void serial_test_tp_attach_query(void)
 	for (i = 0; i < num_progs; i++)
 		obj[i] = NULL;
 
-	snprintf(buf, sizeof(buf),
-		 "/sys/kernel/debug/tracing/events/sched/sched_switch/id");
+	if (access("/sys/kernel/tracing/trace", F_OK) == 0) {
+		snprintf(buf, sizeof(buf),
+			 "/sys/kernel/tracing/events/sched/sched_switch/id");
+	} else {
+		snprintf(buf, sizeof(buf),
+			 "/sys/kernel/debug/tracing/events/sched/sched_switch/id");
+	}
 	efd = open(buf, O_RDONLY, 0);
 	if (CHECK(efd < 0, "open", "err %d errno %d\n", efd, errno))
 		return;
diff --git a/tools/testing/selftests/bpf/prog_tests/trace_printk.c b/tools/testing/selftests/bpf/prog_tests/trace_printk.c
index cade7f12315f..7b9124d506a5 100644
--- a/tools/testing/selftests/bpf/prog_tests/trace_printk.c
+++ b/tools/testing/selftests/bpf/prog_tests/trace_printk.c
@@ -5,7 +5,8 @@
 
 #include "trace_printk.lskel.h"
 
-#define TRACEBUF	"/sys/kernel/debug/tracing/trace_pipe"
+#define TRACEFS_PIPE	"/sys/kernel/tracing/trace_pipe"
+#define DEBUGFS_PIPE	"/sys/kernel/debug/tracing/trace_pipe"
 #define SEARCHMSG	"testing,testing"
 
 void serial_test_trace_printk(void)
@@ -34,8 +35,11 @@ void serial_test_trace_printk(void)
 	if (!ASSERT_OK(err, "trace_printk__attach"))
 		goto cleanup;
 
-	fp = fopen(TRACEBUF, "r");
-	if (!ASSERT_OK_PTR(fp, "fopen(TRACEBUF)"))
+	if (access(TRACEFS_PIPE, F_OK) == 0)
+		fp = fopen(TRACEFS_PIPE, "r");
+	else
+		fp = fopen(DEBUGFS_PIPE, "r");
+	if (!ASSERT_OK_PTR(fp, "fopen(TRACE_PIPE)"))
 		goto cleanup;
 
 	/* We do not want to wait forever if this test fails... */
diff --git a/tools/testing/selftests/bpf/prog_tests/trace_vprintk.c b/tools/testing/selftests/bpf/prog_tests/trace_vprintk.c
index 7a4e313e8558..44ea2fd88f4c 100644
--- a/tools/testing/selftests/bpf/prog_tests/trace_vprintk.c
+++ b/tools/testing/selftests/bpf/prog_tests/trace_vprintk.c
@@ -5,7 +5,8 @@
 
 #include "trace_vprintk.lskel.h"
 
-#define TRACEBUF	"/sys/kernel/debug/tracing/trace_pipe"
+#define TRACEFS_PIPE	"/sys/kernel/tracing/trace_pipe"
+#define DEBUGFS_PIPE	"/sys/kernel/debug/tracing/trace_pipe"
 #define SEARCHMSG	"1,2,3,4,5,6,7,8,9,10"
 
 void serial_test_trace_vprintk(void)
@@ -27,8 +28,11 @@ void serial_test_trace_vprintk(void)
 	if (!ASSERT_OK(err, "trace_vprintk__attach"))
 		goto cleanup;
 
-	fp = fopen(TRACEBUF, "r");
-	if (!ASSERT_OK_PTR(fp, "fopen(TRACEBUF)"))
+	if (access(TRACEFS_PIPE, F_OK) == 0)
+		fp = fopen(TRACEFS_PIPE, "r");
+	else
+		fp = fopen(DEBUGFS_PIPE, "r");
+	if (!ASSERT_OK_PTR(fp, "fopen(TRACE_PIPE)"))
 		goto cleanup;
 
 	/* We do not want to wait forever if this test fails... */
diff --git a/tools/testing/selftests/bpf/progs/test_stacktrace_map.c b/tools/testing/selftests/bpf/progs/test_stacktrace_map.c
index 728dbd39eff0..47568007b668 100644
--- a/tools/testing/selftests/bpf/progs/test_stacktrace_map.c
+++ b/tools/testing/selftests/bpf/progs/test_stacktrace_map.c
@@ -38,7 +38,7 @@ struct {
 	__type(value, stack_trace_t);
 } stack_amap SEC(".maps");
 
-/* taken from /sys/kernel/debug/tracing/events/sched/sched_switch/format */
+/* taken from /sys/kernel/tracing/events/sched/sched_switch/format */
 struct sched_switch_args {
 	unsigned long long pad;
 	char prev_comm[TASK_COMM_LEN];
diff --git a/tools/testing/selftests/bpf/progs/test_tracepoint.c b/tools/testing/selftests/bpf/progs/test_tracepoint.c
index 43bd7a20cc50..4cb8bbb6a320 100644
--- a/tools/testing/selftests/bpf/progs/test_tracepoint.c
+++ b/tools/testing/selftests/bpf/progs/test_tracepoint.c
@@ -4,7 +4,7 @@
 #include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
 
-/* taken from /sys/kernel/debug/tracing/events/sched/sched_switch/format */
+/* taken from /sys/kernel/tracing/events/sched/sched_switch/format */
 struct sched_switch_args {
 	unsigned long long pad;
 	char prev_comm[TASK_COMM_LEN];
diff --git a/tools/testing/selftests/bpf/test_ftrace.sh b/tools/testing/selftests/bpf/test_ftrace.sh
index 20de7bb873bc..f5109eb0e951 100755
--- a/tools/testing/selftests/bpf/test_ftrace.sh
+++ b/tools/testing/selftests/bpf/test_ftrace.sh
@@ -1,6 +1,11 @@
 #!/bin/bash
 
-TR=/sys/kernel/debug/tracing/
+if [[ -e /sys/kernel/tracing/trace ]]; then
+    TR=/sys/kernel/tracing/
+else
+    TR=/sys/kernel/debug/tracing/
+fi
+
 clear_trace() { # reset trace output
     echo > $TR/trace
 }
diff --git a/tools/testing/selftests/bpf/test_tunnel.sh b/tools/testing/selftests/bpf/test_tunnel.sh
index 06857b689c11..2dec7dbf29a2 100755
--- a/tools/testing/selftests/bpf/test_tunnel.sh
+++ b/tools/testing/selftests/bpf/test_tunnel.sh
@@ -571,8 +571,13 @@ setup_xfrm_tunnel()
 
 test_xfrm_tunnel()
 {
+	if [[ -e /sys/kernel/tracing/trace ]]; then
+		TRACE=/sys/kernel/tracing/trace
+	else
+		TRACE=/sys/kernel/debug/tracing/trace
+	fi
 	config_device
-	> /sys/kernel/debug/tracing/trace
+	> ${TRACE}
 	setup_xfrm_tunnel
 	mkdir -p ${BPF_PIN_TUNNEL_DIR}
 	bpftool prog loadall ${BPF_FILE} ${BPF_PIN_TUNNEL_DIR}
@@ -581,11 +586,11 @@ test_xfrm_tunnel()
 		${BPF_PIN_TUNNEL_DIR}/xfrm_get_state
 	ip netns exec at_ns0 ping $PING_ARG 10.1.1.200
 	sleep 1
-	grep "reqid 1" /sys/kernel/debug/tracing/trace
+	grep "reqid 1" ${TRACE}
 	check_err $?
-	grep "spi 0x1" /sys/kernel/debug/tracing/trace
+	grep "spi 0x1" ${TRACE}
 	check_err $?
-	grep "remote ip 0xac100164" /sys/kernel/debug/tracing/trace
+	grep "remote ip 0xac100164" ${TRACE}
 	check_err $?
 	cleanup
 
diff --git a/tools/testing/selftests/bpf/trace_helpers.c b/tools/testing/selftests/bpf/trace_helpers.c
index 09a16a77bae4..934bf28fc888 100644
--- a/tools/testing/selftests/bpf/trace_helpers.c
+++ b/tools/testing/selftests/bpf/trace_helpers.c
@@ -12,7 +12,8 @@
 #include <sys/mman.h>
 #include "trace_helpers.h"
 
-#define DEBUGFS "/sys/kernel/debug/tracing/"
+#define TRACEFS_PIPE	"/sys/kernel/tracing/trace_pipe"
+#define DEBUGFS_PIPE	"/sys/kernel/debug/tracing/trace_pipe"
 
 #define MAX_SYMS 300000
 static struct ksym syms[MAX_SYMS];
@@ -136,7 +137,10 @@ void read_trace_pipe(void)
 {
 	int trace_fd;
 
-	trace_fd = open(DEBUGFS "trace_pipe", O_RDONLY, 0);
+	if (access(TRACEFS_PIPE, F_OK) == 0)
+		trace_fd = open(TRACEFS_PIPE, O_RDONLY, 0);
+	else
+		trace_fd = open(DEBUGFS_PIPE, O_RDONLY, 0);
 	if (trace_fd < 0)
 		return;
 
-- 
cgit v1.2.3-70-g09d2


From b9fe8e8d03d0df28b2431e3aaf8e115cf7bf2f65 Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler@microsoft.com>
Date: Fri, 10 Mar 2023 23:38:14 +0000
Subject: bpf, docs: Add signed comparison example

Improve clarity by adding an example of a signed comparison instruction

Signed-off-by: Dave Thaler <dthaler@microsoft.com>
Acked-by: David Vernet <void@manifault.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/r/20230310233814.4641-1-dthaler1968@googlemail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/instruction-set.rst | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/Documentation/bpf/instruction-set.rst b/Documentation/bpf/instruction-set.rst
index 5e43e14abe80..b44640589055 100644
--- a/Documentation/bpf/instruction-set.rst
+++ b/Documentation/bpf/instruction-set.rst
@@ -11,7 +11,8 @@ Documentation conventions
 =========================
 
 For brevity, this document uses the type notion "u64", "u32", etc.
-to mean an unsigned integer whose width is the specified number of bits.
+to mean an unsigned integer whose width is the specified number of bits,
+and "s32", etc. to mean a signed integer of the specified number of bits.
 
 Registers and calling convention
 ================================
@@ -264,6 +265,14 @@ BPF_JSLE  0xd0   PC += off if dst <= src    signed
 The eBPF program needs to store the return value into register R0 before doing a
 BPF_EXIT.
 
+Example:
+
+``BPF_JSGE | BPF_X | BPF_JMP32`` (0x7e) means::
+
+  if (s32)dst s>= (s32)src goto +offset
+
+where 's>=' indicates a signed '>=' comparison.
+
 Helper functions
 ~~~~~~~~~~~~~~~~
 
-- 
cgit v1.2.3-70-g09d2


From c9267aa8b794c2188d49c7d7bd2990e98b2d6b84 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 13 Mar 2023 16:58:43 -0700
Subject: bpf: Fix bpf_strncmp proto.

bpf_strncmp() doesn't write into its first argument.
Make sure that the verifier knows about it.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230313235845.61029-2-alexei.starovoitov@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 kernel/bpf/helpers.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 77d64b6951b9..f753676ef652 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -571,7 +571,7 @@ static const struct bpf_func_proto bpf_strncmp_proto = {
 	.func		= bpf_strncmp,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
-	.arg1_type	= ARG_PTR_TO_MEM,
+	.arg1_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
 	.arg2_type	= ARG_CONST_SIZE,
 	.arg3_type	= ARG_PTR_TO_CONST_STR,
 };
-- 
cgit v1.2.3-70-g09d2


From 3e30be4288b31702d4898487a74e80ba14150a9f Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 13 Mar 2023 16:58:44 -0700
Subject: bpf: Allow helpers access trusted PTR_TO_BTF_ID.

The verifier rejects the code:
  bpf_strncmp(task->comm, 16, "my_task");
with the message:
  16: (85) call bpf_strncmp#182
  R1 type=trusted_ptr_ expected=fp, pkt, pkt_meta, map_key, map_value, mem, ringbuf_mem, buf

Teach the verifier that such access pattern is safe.
Do not allow untrusted and legacy ptr_to_btf_id to be passed into helpers.

Reported-by: David Vernet <void@manifault.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230313235845.61029-3-alexei.starovoitov@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 kernel/bpf/verifier.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 883d4ff2e288..2bbd89279070 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -6303,6 +6303,9 @@ static int check_helper_mem_access(struct bpf_verifier_env *env, int regno,
 				env,
 				regno, reg->off, access_size,
 				zero_size_allowed, ACCESS_HELPER, meta);
+	case PTR_TO_BTF_ID:
+		return check_ptr_to_btf_access(env, regs, regno, reg->off,
+					       access_size, BPF_READ, -1);
 	case PTR_TO_CTX:
 		/* in case the function doesn't know how to access the context,
 		 * (because we are in a program of type SYSCALL for example), we
@@ -7014,6 +7017,7 @@ static const struct bpf_reg_types mem_types = {
 		PTR_TO_MEM,
 		PTR_TO_MEM | MEM_RINGBUF,
 		PTR_TO_BUF,
+		PTR_TO_BTF_ID | PTR_TRUSTED,
 	},
 };
 
@@ -7145,6 +7149,17 @@ found:
 	if (base_type(reg->type) != PTR_TO_BTF_ID)
 		return 0;
 
+	if (compatible == &mem_types) {
+		if (!(arg_type & MEM_RDONLY)) {
+			verbose(env,
+				"%s() may write into memory pointed by R%d type=%s\n",
+				func_id_name(meta->func_id),
+				regno, reg_type_str(env, reg->type));
+			return -EACCES;
+		}
+		return 0;
+	}
+
 	switch ((int)reg->type) {
 	case PTR_TO_BTF_ID:
 	case PTR_TO_BTF_ID | PTR_TRUSTED:
-- 
cgit v1.2.3-70-g09d2


From f25fd6088216bd257902e5c212177cddcb291218 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 13 Mar 2023 16:58:45 -0700
Subject: selftests/bpf: Add various tests to check helper access into
 ptr_to_btf_id.

Add various tests to check helper access into ptr_to_btf_id.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230313235845.61029-4-alexei.starovoitov@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 .../selftests/bpf/progs/task_kfunc_failure.c       | 36 ++++++++++++++++++++++
 .../selftests/bpf/progs/task_kfunc_success.c       |  4 +++
 2 files changed, 40 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
index 002c7f69e47f..27994d6b2914 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
@@ -301,3 +301,39 @@ int BPF_PROG(task_kfunc_from_lsm_task_free, struct task_struct *task)
 	bpf_task_release(acquired);
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+__failure __msg("access beyond the end of member comm")
+int BPF_PROG(task_access_comm1, struct task_struct *task, u64 clone_flags)
+{
+	bpf_strncmp(task->comm, 17, "foo");
+	return 0;
+}
+
+SEC("tp_btf/task_newtask")
+__failure __msg("access beyond the end of member comm")
+int BPF_PROG(task_access_comm2, struct task_struct *task, u64 clone_flags)
+{
+	bpf_strncmp(task->comm + 1, 16, "foo");
+	return 0;
+}
+
+SEC("tp_btf/task_newtask")
+__failure __msg("write into memory")
+int BPF_PROG(task_access_comm3, struct task_struct *task, u64 clone_flags)
+{
+	bpf_probe_read_kernel(task->comm, 16, task->comm);
+	return 0;
+}
+
+SEC("fentry/__set_task_comm")
+__failure __msg("R1 type=ptr_ expected")
+int BPF_PROG(task_access_comm4, struct task_struct *task, const char *buf, bool exec)
+{
+	/*
+	 * task->comm is a legacy ptr_to_btf_id. The verifier cannot guarantee
+	 * its safety. Hence it cannot be accessed with normal load insns.
+	 */
+	bpf_strncmp(task->comm, 16, "foo");
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index aebc4bb14e7d..4f61596b0242 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -207,6 +207,10 @@ int BPF_PROG(test_task_from_pid_invalid, struct task_struct *task, u64 clone_fla
 	if (!is_test_kfunc_task())
 		return 0;
 
+	bpf_strncmp(task->comm, 12, "foo");
+	bpf_strncmp(task->comm, 16, "foo");
+	bpf_strncmp(&task->comm[8], 4, "foo");
+
 	if (is_pid_lookup_valid(-1)) {
 		err = 1;
 		return 0;
-- 
cgit v1.2.3-70-g09d2


From 487deb3e3393cccff0f148c4703efb185d46e314 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Mon, 13 Mar 2023 22:55:50 +0100
Subject: selftests/bpf: robustify test_xdp_do_redirect with more payload
 magics

Currently, the test relies on that only dropped ("xmitted") frames will
be recycled and if a frame became an skb, it will be freed later by the
stack and never come back to its page_pool.
So, it easily gets broken by trying to recycle skbs[0]:

  test_xdp_do_redirect:PASS:pkt_count_xdp 0 nsec
  test_xdp_do_redirect:FAIL:pkt_count_zero unexpected pkt_count_zero:
actual 9936 != expected 2
  test_xdp_do_redirect:PASS:pkt_count_tc 0 nsec

That huge mismatch happened because after the TC ingress hook zeroes the
magic, the page gets recycled when skb is freed, not returned to the MM
layer. "Live frames" mode initializes only new pages and keeps the
recycled ones as is by design, so they appear with zeroed magic on the
Rx path again.
Expand the possible magic values from two: 0 (was "xmitted"/dropped or
did hit the TC hook) and 0x42 (hit the input XDP prog) to three: the new
one will mark frames hit the TC hook, so that they will elide both
@pkt_count_zero and @pkt_count_xdp. They can then be recycled to their
page_pool or returned to the page allocator, this won't affect the
counters anyhow. Just make sure to mark them as "input" (0x42) when they
appear on the Rx path again.
Also make an enum from those magics, so that they will be always visible
and can be changed in just one place anytime. This also eases adding any
new marks later on.

Link: https://github.com/kernel-patches/bpf/actions/runs/4386538411/jobs/7681081789
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230313215553.1045175-2-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/progs/test_xdp_do_redirect.c     | 36 ++++++++++++++++------
 1 file changed, 26 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c b/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c
index 77a123071940..cd2d4e3258b8 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c
@@ -4,6 +4,19 @@
 
 #define ETH_ALEN 6
 #define HDR_SZ (sizeof(struct ethhdr) + sizeof(struct ipv6hdr) + sizeof(struct udphdr))
+
+/**
+ * enum frame_mark - magics to distinguish page/packet paths
+ * @MARK_XMIT: page was recycled due to the frame being "xmitted" by the NIC.
+ * @MARK_IN: frame is being processed by the input XDP prog.
+ * @MARK_SKB: frame did hit the TC ingress hook as an skb.
+ */
+enum frame_mark {
+	MARK_XMIT	= 0U,
+	MARK_IN		= 0x42,
+	MARK_SKB	= 0x45,
+};
+
 const volatile int ifindex_out;
 const volatile int ifindex_in;
 const volatile __u8 expect_dst[ETH_ALEN];
@@ -34,10 +47,10 @@ int xdp_redirect(struct xdp_md *xdp)
 	if (*metadata != 0x42)
 		return XDP_ABORTED;
 
-	if (*payload == 0) {
-		*payload = 0x42;
+	if (*payload == MARK_XMIT)
 		pkts_seen_zero++;
-	}
+
+	*payload = MARK_IN;
 
 	if (bpf_xdp_adjust_meta(xdp, 4))
 		return XDP_ABORTED;
@@ -51,7 +64,7 @@ int xdp_redirect(struct xdp_md *xdp)
 	return ret;
 }
 
-static bool check_pkt(void *data, void *data_end)
+static bool check_pkt(void *data, void *data_end, const __u32 mark)
 {
 	struct ipv6hdr *iph = data + sizeof(struct ethhdr);
 	__u8 *payload = data + HDR_SZ;
@@ -59,13 +72,13 @@ static bool check_pkt(void *data, void *data_end)
 	if (payload + 1 > data_end)
 		return false;
 
-	if (iph->nexthdr != IPPROTO_UDP || *payload != 0x42)
+	if (iph->nexthdr != IPPROTO_UDP || *payload != MARK_IN)
 		return false;
 
 	/* reset the payload so the same packet doesn't get counted twice when
 	 * it cycles back through the kernel path and out the dst veth
 	 */
-	*payload = 0;
+	*payload = mark;
 	return true;
 }
 
@@ -75,11 +88,11 @@ int xdp_count_pkts(struct xdp_md *xdp)
 	void *data = (void *)(long)xdp->data;
 	void *data_end = (void *)(long)xdp->data_end;
 
-	if (check_pkt(data, data_end))
+	if (check_pkt(data, data_end, MARK_XMIT))
 		pkts_seen_xdp++;
 
-	/* Return XDP_DROP to make sure the data page is recycled, like when it
-	 * exits a physical NIC. Recycled pages will be counted in the
+	/* Return %XDP_DROP to recycle the data page with %MARK_XMIT, like
+	 * it exited a physical NIC. Those pages will be counted in the
 	 * pkts_seen_zero counter above.
 	 */
 	return XDP_DROP;
@@ -91,9 +104,12 @@ int tc_count_pkts(struct __sk_buff *skb)
 	void *data = (void *)(long)skb->data;
 	void *data_end = (void *)(long)skb->data_end;
 
-	if (check_pkt(data, data_end))
+	if (check_pkt(data, data_end, MARK_SKB))
 		pkts_seen_tc++;
 
+	/* Will be either recycled or freed, %MARK_SKB makes sure it won't
+	 * hit any of the counters above.
+	 */
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 2c854e5fcd7e243f5a7cf6a6afa0ef83060c903c Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Mon, 13 Mar 2023 22:55:51 +0100
Subject: net: page_pool, skbuff: make skb_mark_for_recycle() always available

skb_mark_for_recycle() is guarded with CONFIG_PAGE_POOL, this creates
unneeded complication when using it in the generic code. For now, it's
only used in the drivers always selecting Page Pool, so this works.
Move the guards so that preprocessor will cut out only the operation
itself and the function will still be a noop on !PAGE_POOL systems,
but available there as well.
No functional changes.

Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202303020342.Wi2PRFFH-lkp@intel.com
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230313215553.1045175-3-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/skbuff.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index fe661011644b..3f3a2a82a86b 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -5069,12 +5069,12 @@ static inline u64 skb_get_kcov_handle(struct sk_buff *skb)
 #endif
 }
 
-#ifdef CONFIG_PAGE_POOL
 static inline void skb_mark_for_recycle(struct sk_buff *skb)
 {
+#ifdef CONFIG_PAGE_POOL
 	skb->pp_recycle = 1;
-}
 #endif
+}
 
 #endif	/* __KERNEL__ */
 #endif	/* _LINUX_SKBUFF_H */
-- 
cgit v1.2.3-70-g09d2


From 9c94bbf9a87b264294f42e6cc0f76d87854733ec Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Mon, 13 Mar 2023 22:55:52 +0100
Subject: xdp: recycle Page Pool backed skbs built from XDP frames

__xdp_build_skb_from_frame() state(d):

/* Until page_pool get SKB return path, release DMA here */

Page Pool got skb pages recycling in April 2021, but missed this
function.

xdp_release_frame() is relevant only for Page Pool backed frames and it
detaches the page from the corresponding page_pool in order to make it
freeable via page_frag_free(). It can instead just mark the output skb
as eligible for recycling if the frame is backed by a pp. No change for
other memory model types (the same condition check as before).
cpumap redirect and veth on Page Pool drivers now become zero-alloc (or
almost).

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230313215553.1045175-4-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/core/xdp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/core/xdp.c b/net/core/xdp.c
index 8c92fc553317..a2237cfca8e9 100644
--- a/net/core/xdp.c
+++ b/net/core/xdp.c
@@ -658,8 +658,8 @@ struct sk_buff *__xdp_build_skb_from_frame(struct xdp_frame *xdpf,
 	 * - RX ring dev queue index	(skb_record_rx_queue)
 	 */
 
-	/* Until page_pool get SKB return path, release DMA here */
-	xdp_release_frame(xdpf);
+	if (xdpf->mem.type == MEM_TYPE_PAGE_POOL)
+		skb_mark_for_recycle(skb);
 
 	/* Allow SKB to reuse area used by xdp_frame */
 	xdp_scrub_frame(xdpf);
-- 
cgit v1.2.3-70-g09d2


From d4e492338d11937c55841b1279287280d6e35894 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Mon, 13 Mar 2023 22:55:53 +0100
Subject: xdp: remove unused {__,}xdp_release_frame()

__xdp_build_skb_from_frame() was the last user of
{__,}xdp_release_frame(), which detaches pages from the page_pool.
All the consumers now recycle Page Pool skbs and page, except mlx5,
stmmac and tsnep drivers, which use page_pool_release_page() directly
(might change one day). It's safe to assume this functionality is not
needed anymore and can be removed (in favor of recycling).

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230313215553.1045175-5-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/net/xdp.h | 29 -----------------------------
 net/core/xdp.c    | 15 ---------------
 2 files changed, 44 deletions(-)

diff --git a/include/net/xdp.h b/include/net/xdp.h
index d517bfac937b..5393b3ebe56e 100644
--- a/include/net/xdp.h
+++ b/include/net/xdp.h
@@ -317,35 +317,6 @@ void xdp_flush_frame_bulk(struct xdp_frame_bulk *bq);
 void xdp_return_frame_bulk(struct xdp_frame *xdpf,
 			   struct xdp_frame_bulk *bq);
 
-/* When sending xdp_frame into the network stack, then there is no
- * return point callback, which is needed to release e.g. DMA-mapping
- * resources with page_pool.  Thus, have explicit function to release
- * frame resources.
- */
-void __xdp_release_frame(void *data, struct xdp_mem_info *mem);
-static inline void xdp_release_frame(struct xdp_frame *xdpf)
-{
-	struct xdp_mem_info *mem = &xdpf->mem;
-	struct skb_shared_info *sinfo;
-	int i;
-
-	/* Curr only page_pool needs this */
-	if (mem->type != MEM_TYPE_PAGE_POOL)
-		return;
-
-	if (likely(!xdp_frame_has_frags(xdpf)))
-		goto out;
-
-	sinfo = xdp_get_shared_info_from_frame(xdpf);
-	for (i = 0; i < sinfo->nr_frags; i++) {
-		struct page *page = skb_frag_page(&sinfo->frags[i]);
-
-		__xdp_release_frame(page_address(page), mem);
-	}
-out:
-	__xdp_release_frame(xdpf->data, mem);
-}
-
 static __always_inline unsigned int xdp_get_frame_len(struct xdp_frame *xdpf)
 {
 	struct skb_shared_info *sinfo;
diff --git a/net/core/xdp.c b/net/core/xdp.c
index a2237cfca8e9..8d3ad315f18d 100644
--- a/net/core/xdp.c
+++ b/net/core/xdp.c
@@ -531,21 +531,6 @@ out:
 }
 EXPORT_SYMBOL_GPL(xdp_return_buff);
 
-/* Only called for MEM_TYPE_PAGE_POOL see xdp.h */
-void __xdp_release_frame(void *data, struct xdp_mem_info *mem)
-{
-	struct xdp_mem_allocator *xa;
-	struct page *page;
-
-	rcu_read_lock();
-	xa = rhashtable_lookup(mem_id_ht, &mem->id, mem_id_rht_params);
-	page = virt_to_head_page(data);
-	if (xa)
-		page_pool_release_page(xa->page_pool, page);
-	rcu_read_unlock();
-}
-EXPORT_SYMBOL_GPL(__xdp_release_frame);
-
 void xdp_attachment_setup(struct xdp_attachment_info *info,
 			  struct netdev_bpf *bpf)
 {
-- 
cgit v1.2.3-70-g09d2


From 3c2611bac08a834697be918ac357eaff2e47d5b3 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 14 Mar 2023 15:28:11 -0700
Subject: selftests/bpf: Fix trace_virtqueue_add_sgs test issue with LLVM 17.

LLVM commit https://reviews.llvm.org/D143726 introduced hoistMinMax optimization
that transformed
  (i < VIRTIO_MAX_SGS) && (i < out_sgs)
into
  i < MIN(VIRTIO_MAX_SGS, out_sgs)

and caused the verifier to stop recognizing such loop as bounded.

Which resulted in the following test failure:

libbpf: prog 'trace_virtqueue_add_sgs': BPF program load failed: Bad address
libbpf: prog 'trace_virtqueue_add_sgs': -- BEGIN PROG LOAD LOG --
The sequence of 8193 jumps is too complex.
verification time 789206 usec
stack depth 56
processed 156446 insns (limit 1000000) max_states_per_insn 7 total_states 1746 peak_states 1701 mark_read 12
-- END PROG LOAD LOG --
libbpf: prog 'trace_virtqueue_add_sgs': failed to load: -14
libbpf: failed to load object 'loop6.bpf.o'

Workaround the verifier limitation for now with inline asm that
prevents this particular optimization.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/loop6.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/loop6.c b/tools/testing/selftests/bpf/progs/loop6.c
index 38de0331e6b4..e4ff97fbcce1 100644
--- a/tools/testing/selftests/bpf/progs/loop6.c
+++ b/tools/testing/selftests/bpf/progs/loop6.c
@@ -5,6 +5,7 @@
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
+#include "bpf_misc.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -76,6 +77,7 @@ int BPF_KPROBE(trace_virtqueue_add_sgs, void *unused, struct scatterlist **sgs,
 		return 0;
 
 	for (i = 0; (i < VIRTIO_MAX_SGS) && (i < out_sgs); i++) {
+		__sink(out_sgs);
 		for (n = 0, sgp = get_sgp(sgs, i); sgp && (n < SG_MAX);
 		     sgp = __sg_next(sgp)) {
 			bpf_probe_read_kernel(&len, sizeof(len), &sgp->length);
@@ -85,6 +87,7 @@ int BPF_KPROBE(trace_virtqueue_add_sgs, void *unused, struct scatterlist **sgs,
 	}
 
 	for (i = 0; (i < VIRTIO_MAX_SGS) && (i < in_sgs); i++) {
+		__sink(in_sgs);
 		for (n = 0, sgp = get_sgp(sgs, i); sgp && (n < SG_MAX);
 		     sgp = __sg_next(sgp)) {
 			bpf_probe_read_kernel(&len, sizeof(len), &sgp->length);
-- 
cgit v1.2.3-70-g09d2


From b8a2e3f93d412114a1539ea97b59b3e6ed6e1f9a Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Tue, 14 Mar 2023 11:59:49 -1000
Subject: cgroup: Make current_cgns_cgroup_dfl() safe to call after
 exit_task_namespace()

The commit 332ea1f697be ("bpf: Add bpf_cgroup_from_id() kfunc") added
bpf_cgroup_from_id() which calls current_cgns_cgroup_dfl() through
cgroup_get_from_id(). However, BPF programs may be attached to a point where
current->nsproxy has already been cleared to NULL by exit_task_namespace()
and calling bpf_cgroup_from_id() would cause an oops.

Just return the system-wide root if nsproxy has been cleared. This allows
all cgroups to be looked up after the task passed through
exit_task_namespace(), which semantically makes sense. Given that the only
way to get this behavior is through BPF programs, it seems safe but let's
see what others think.

Fixes: 332ea1f697be ("bpf: Add bpf_cgroup_from_id() kfunc")
Signed-off-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/ZBDuVWiFj2jiz3i8@slm.duckdns.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/cgroup/cgroup.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 935e8121b21e..8a5294f4ce72 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -1465,8 +1465,18 @@ static struct cgroup *current_cgns_cgroup_dfl(void)
 {
 	struct css_set *cset;
 
-	cset = current->nsproxy->cgroup_ns->root_cset;
-	return __cset_cgroup_from_root(cset, &cgrp_dfl_root);
+	if (current->nsproxy) {
+		cset = current->nsproxy->cgroup_ns->root_cset;
+		return __cset_cgroup_from_root(cset, &cgrp_dfl_root);
+	} else {
+		/*
+		 * NOTE: This function may be called from bpf_cgroup_from_id()
+		 * on a task which has already passed exit_task_namespaces() and
+		 * nsproxy == NULL. Fall back to cgrp_dfl_root which will make all
+		 * cgroups visible for lookups.
+		 */
+		return &cgrp_dfl_root.cgrp;
+	}
 }
 
 /* look up cgroup associated with given css_set on the specified hierarchy */
-- 
cgit v1.2.3-70-g09d2


From 68a84a127bb07d2a47e95c808520f742c54781c7 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Sun, 12 Mar 2023 20:51:55 +0100
Subject: net: lan966x: Change lan966x_police_del return type
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As the function always returns 0 change the return type to be
void instead of int. In this way also remove a wrong message
in case of error which would never happen.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Alvin Šipraga <alsi@bang-olufsen.dk>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230312195155.1492881-1-horatiu.vultur@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microchip/lan966x/lan966x_police.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_police.c b/drivers/net/ethernet/microchip/lan966x/lan966x_police.c
index 7d66fe75cd3b..7302df2300fd 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_police.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_police.c
@@ -49,8 +49,7 @@ static int lan966x_police_add(struct lan966x_port *port,
 	return 0;
 }
 
-static int lan966x_police_del(struct lan966x_port *port,
-			      u16 pol_idx)
+static void lan966x_police_del(struct lan966x_port *port, u16 pol_idx)
 {
 	struct lan966x *lan966x = port->lan966x;
 
@@ -67,8 +66,6 @@ static int lan966x_police_del(struct lan966x_port *port,
 	lan_wr(ANA_POL_PIR_CFG_PIR_RATE_SET(GENMASK(14, 0)) |
 	       ANA_POL_PIR_CFG_PIR_BURST_SET(0),
 	       lan966x, ANA_POL_PIR_CFG(pol_idx));
-
-	return 0;
 }
 
 static int lan966x_police_validate(struct lan966x_port *port,
@@ -186,7 +183,6 @@ int lan966x_police_port_del(struct lan966x_port *port,
 			    struct netlink_ext_ack *extack)
 {
 	struct lan966x *lan966x = port->lan966x;
-	int err;
 
 	if (port->tc.police_id != police_id) {
 		NL_SET_ERR_MSG_MOD(extack,
@@ -194,12 +190,7 @@ int lan966x_police_port_del(struct lan966x_port *port,
 		return -EINVAL;
 	}
 
-	err = lan966x_police_del(port, POL_IDX_PORT + port->chip_port);
-	if (err) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "Failed to add policer to port");
-		return err;
-	}
+	lan966x_police_del(port, POL_IDX_PORT + port->chip_port);
 
 	lan_rmw(ANA_POL_CFG_PORT_POL_ENA_SET(0) |
 		ANA_POL_CFG_POL_ORDER_SET(POL_ORDER),
-- 
cgit v1.2.3-70-g09d2


From b071af523579df7341cabf0f16fc661125e9a13f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 13 Mar 2023 20:17:31 +0000
Subject: neighbour: annotate lockless accesses to n->nud_state

We have many lockless accesses to n->nud_state.

Before adding another one in the following patch,
add annotations to readers and writers.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vxlan/vxlan_core.c  |  4 ++--
 include/net/neighbour.h         |  2 +-
 net/bridge/br_arp_nd_proxy.c    |  4 ++--
 net/bridge/br_netfilter_hooks.c |  3 ++-
 net/core/filter.c               |  4 ++--
 net/core/neighbour.c            | 28 ++++++++++++++--------------
 net/ipv4/arp.c                  |  8 ++++----
 net/ipv4/fib_semantics.c        |  4 ++--
 net/ipv4/nexthop.c              |  4 ++--
 net/ipv4/route.c                |  2 +-
 net/ipv6/ip6_output.c           |  2 +-
 net/ipv6/ndisc.c                |  4 ++--
 net/ipv6/route.c                |  2 +-
 13 files changed, 36 insertions(+), 35 deletions(-)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index b1b179effe2a..f2c30214cae8 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -1863,7 +1863,7 @@ static int arp_reduce(struct net_device *dev, struct sk_buff *skb, __be32 vni)
 		struct vxlan_fdb *f;
 		struct sk_buff	*reply;
 
-		if (!(n->nud_state & NUD_CONNECTED)) {
+		if (!(READ_ONCE(n->nud_state) & NUD_CONNECTED)) {
 			neigh_release(n);
 			goto out;
 		}
@@ -2027,7 +2027,7 @@ static int neigh_reduce(struct net_device *dev, struct sk_buff *skb, __be32 vni)
 		struct vxlan_fdb *f;
 		struct sk_buff *reply;
 
-		if (!(n->nud_state & NUD_CONNECTED)) {
+		if (!(READ_ONCE(n->nud_state) & NUD_CONNECTED)) {
 			neigh_release(n);
 			goto out;
 		}
diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 234799ca527e..c8d39bba2a0d 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -464,7 +464,7 @@ static __always_inline int neigh_event_send_probe(struct neighbour *neigh,
 
 	if (READ_ONCE(neigh->used) != now)
 		WRITE_ONCE(neigh->used, now);
-	if (!(neigh->nud_state & (NUD_CONNECTED | NUD_DELAY | NUD_PROBE)))
+	if (!(READ_ONCE(neigh->nud_state) & (NUD_CONNECTED | NUD_DELAY | NUD_PROBE)))
 		return __neigh_event_send(neigh, skb, immediate_ok);
 	return 0;
 }
diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c
index e5e48c6e35d7..b45c00c01dea 100644
--- a/net/bridge/br_arp_nd_proxy.c
+++ b/net/bridge/br_arp_nd_proxy.c
@@ -192,7 +192,7 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br,
 	if (n) {
 		struct net_bridge_fdb_entry *f;
 
-		if (!(n->nud_state & NUD_VALID)) {
+		if (!(READ_ONCE(n->nud_state) & NUD_VALID)) {
 			neigh_release(n);
 			return;
 		}
@@ -452,7 +452,7 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 	if (n) {
 		struct net_bridge_fdb_entry *f;
 
-		if (!(n->nud_state & NUD_VALID)) {
+		if (!(READ_ONCE(n->nud_state) & NUD_VALID)) {
 			neigh_release(n);
 			return;
 		}
diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c
index 638a4d5359db..3e3065bc0465 100644
--- a/net/bridge/br_netfilter_hooks.c
+++ b/net/bridge/br_netfilter_hooks.c
@@ -277,7 +277,8 @@ int br_nf_pre_routing_finish_bridge(struct net *net, struct sock *sk, struct sk_
 		struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 		int ret;
 
-		if ((neigh->nud_state & NUD_CONNECTED) && neigh->hh.hh_len) {
+		if ((READ_ONCE(neigh->nud_state) & NUD_CONNECTED) &&
+		    READ_ONCE(neigh->hh.hh_len)) {
 			neigh_hh_bridge(&neigh->hh, skb);
 			skb->dev = nf_bridge->physindev;
 			ret = br_handle_frame_finish(net, sk, skb);
diff --git a/net/core/filter.c b/net/core/filter.c
index 50f649f1b4a9..d052fac28d02 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -5871,7 +5871,7 @@ static int bpf_ipv4_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
 	else
 		neigh = __ipv6_neigh_lookup_noref_stub(dev, params->ipv6_dst);
 
-	if (!neigh || !(neigh->nud_state & NUD_VALID))
+	if (!neigh || !(READ_ONCE(neigh->nud_state) & NUD_VALID))
 		return BPF_FIB_LKUP_RET_NO_NEIGH;
 	memcpy(params->dmac, neigh->ha, ETH_ALEN);
 	memcpy(params->smac, dev->dev_addr, ETH_ALEN);
@@ -5992,7 +5992,7 @@ static int bpf_ipv6_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
 	 * not needed here.
 	 */
 	neigh = __ipv6_neigh_lookup_noref_stub(dev, dst);
-	if (!neigh || !(neigh->nud_state & NUD_VALID))
+	if (!neigh || !(READ_ONCE(neigh->nud_state) & NUD_VALID))
 		return BPF_FIB_LKUP_RET_NO_NEIGH;
 	memcpy(params->dmac, neigh->ha, ETH_ALEN);
 	memcpy(params->smac, dev->dev_addr, ETH_ALEN);
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 0116b0ff91a7..90d399b3f980 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -1093,13 +1093,13 @@ static void neigh_timer_handler(struct timer_list *t)
 					  neigh->used +
 					  NEIGH_VAR(neigh->parms, DELAY_PROBE_TIME))) {
 			neigh_dbg(2, "neigh %p is delayed\n", neigh);
-			neigh->nud_state = NUD_DELAY;
+			WRITE_ONCE(neigh->nud_state, NUD_DELAY);
 			neigh->updated = jiffies;
 			neigh_suspect(neigh);
 			next = now + NEIGH_VAR(neigh->parms, DELAY_PROBE_TIME);
 		} else {
 			neigh_dbg(2, "neigh %p is suspected\n", neigh);
-			neigh->nud_state = NUD_STALE;
+			WRITE_ONCE(neigh->nud_state, NUD_STALE);
 			neigh->updated = jiffies;
 			neigh_suspect(neigh);
 			notify = 1;
@@ -1109,14 +1109,14 @@ static void neigh_timer_handler(struct timer_list *t)
 				   neigh->confirmed +
 				   NEIGH_VAR(neigh->parms, DELAY_PROBE_TIME))) {
 			neigh_dbg(2, "neigh %p is now reachable\n", neigh);
-			neigh->nud_state = NUD_REACHABLE;
+			WRITE_ONCE(neigh->nud_state, NUD_REACHABLE);
 			neigh->updated = jiffies;
 			neigh_connect(neigh);
 			notify = 1;
 			next = neigh->confirmed + neigh->parms->reachable_time;
 		} else {
 			neigh_dbg(2, "neigh %p is probed\n", neigh);
-			neigh->nud_state = NUD_PROBE;
+			WRITE_ONCE(neigh->nud_state, NUD_PROBE);
 			neigh->updated = jiffies;
 			atomic_set(&neigh->probes, 0);
 			notify = 1;
@@ -1130,7 +1130,7 @@ static void neigh_timer_handler(struct timer_list *t)
 
 	if ((neigh->nud_state & (NUD_INCOMPLETE | NUD_PROBE)) &&
 	    atomic_read(&neigh->probes) >= neigh_max_probes(neigh)) {
-		neigh->nud_state = NUD_FAILED;
+		WRITE_ONCE(neigh->nud_state, NUD_FAILED);
 		notify = 1;
 		neigh_invalidate(neigh);
 		goto out;
@@ -1179,7 +1179,7 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb,
 			atomic_set(&neigh->probes,
 				   NEIGH_VAR(neigh->parms, UCAST_PROBES));
 			neigh_del_timer(neigh);
-			neigh->nud_state = NUD_INCOMPLETE;
+			WRITE_ONCE(neigh->nud_state, NUD_INCOMPLETE);
 			neigh->updated = now;
 			if (!immediate_ok) {
 				next = now + 1;
@@ -1191,7 +1191,7 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb,
 			}
 			neigh_add_timer(neigh, next);
 		} else {
-			neigh->nud_state = NUD_FAILED;
+			WRITE_ONCE(neigh->nud_state, NUD_FAILED);
 			neigh->updated = jiffies;
 			write_unlock_bh(&neigh->lock);
 
@@ -1201,7 +1201,7 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb,
 	} else if (neigh->nud_state & NUD_STALE) {
 		neigh_dbg(2, "neigh %p is delayed\n", neigh);
 		neigh_del_timer(neigh);
-		neigh->nud_state = NUD_DELAY;
+		WRITE_ONCE(neigh->nud_state, NUD_DELAY);
 		neigh->updated = jiffies;
 		neigh_add_timer(neigh, jiffies +
 				NEIGH_VAR(neigh->parms, DELAY_PROBE_TIME));
@@ -1313,7 +1313,7 @@ static int __neigh_update(struct neighbour *neigh, const u8 *lladdr,
 	neigh_update_flags(neigh, flags, &notify, &gc_update, &managed_update);
 	if (flags & (NEIGH_UPDATE_F_USE | NEIGH_UPDATE_F_MANAGED)) {
 		new = old & ~NUD_PERMANENT;
-		neigh->nud_state = new;
+		WRITE_ONCE(neigh->nud_state, new);
 		err = 0;
 		goto out;
 	}
@@ -1322,7 +1322,7 @@ static int __neigh_update(struct neighbour *neigh, const u8 *lladdr,
 		neigh_del_timer(neigh);
 		if (old & NUD_CONNECTED)
 			neigh_suspect(neigh);
-		neigh->nud_state = new;
+		WRITE_ONCE(neigh->nud_state, new);
 		err = 0;
 		notify = old & NUD_VALID;
 		if ((old & (NUD_INCOMPLETE | NUD_PROBE)) &&
@@ -1401,7 +1401,7 @@ static int __neigh_update(struct neighbour *neigh, const u8 *lladdr,
 						((new & NUD_REACHABLE) ?
 						 neigh->parms->reachable_time :
 						 0)));
-		neigh->nud_state = new;
+		WRITE_ONCE(neigh->nud_state, new);
 		notify = 1;
 	}
 
@@ -1488,7 +1488,7 @@ void __neigh_set_probe_once(struct neighbour *neigh)
 	neigh->updated = jiffies;
 	if (!(neigh->nud_state & NUD_FAILED))
 		return;
-	neigh->nud_state = NUD_INCOMPLETE;
+	WRITE_ONCE(neigh->nud_state, NUD_INCOMPLETE);
 	atomic_set(&neigh->probes, neigh_max_probes(neigh));
 	neigh_add_timer(neigh,
 			jiffies + max(NEIGH_VAR(neigh->parms, RETRANS_TIME),
@@ -3198,7 +3198,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
 			}
 			if (!(state->flags & NEIGH_SEQ_SKIP_NOARP))
 				break;
-			if (n->nud_state & ~NUD_NOARP)
+			if (READ_ONCE(n->nud_state) & ~NUD_NOARP)
 				break;
 next:
 			n = rcu_dereference_bh(n->next);
@@ -3240,7 +3240,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 			if (!(state->flags & NEIGH_SEQ_SKIP_NOARP))
 				break;
 
-			if (n->nud_state & ~NUD_NOARP)
+			if (READ_ONCE(n->nud_state) & ~NUD_NOARP)
 				break;
 next:
 			n = rcu_dereference_bh(n->next);
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 4f7237661afb..9456f5bb35e5 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -375,7 +375,7 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
 
 	probes -= NEIGH_VAR(neigh->parms, UCAST_PROBES);
 	if (probes < 0) {
-		if (!(neigh->nud_state & NUD_VALID))
+		if (!(READ_ONCE(neigh->nud_state) & NUD_VALID))
 			pr_debug("trying to ucast probe in NUD_INVALID\n");
 		neigh_ha_snapshot(dst_ha, neigh, dev);
 		dst_hw = dst_ha;
@@ -1123,7 +1123,7 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev)
 
 	neigh = neigh_lookup(&arp_tbl, &ip, dev);
 	if (neigh) {
-		if (!(neigh->nud_state & NUD_NOARP)) {
+		if (!(READ_ONCE(neigh->nud_state) & NUD_NOARP)) {
 			read_lock_bh(&neigh->lock);
 			memcpy(r->arp_ha.sa_data, neigh->ha, dev->addr_len);
 			r->arp_flags = arp_state_to_flags(neigh);
@@ -1144,12 +1144,12 @@ int arp_invalidate(struct net_device *dev, __be32 ip, bool force)
 	struct neigh_table *tbl = &arp_tbl;
 
 	if (neigh) {
-		if ((neigh->nud_state & NUD_VALID) && !force) {
+		if ((READ_ONCE(neigh->nud_state) & NUD_VALID) && !force) {
 			neigh_release(neigh);
 			return 0;
 		}
 
-		if (neigh->nud_state & ~NUD_NOARP)
+		if (READ_ONCE(neigh->nud_state) & ~NUD_NOARP)
 			err = neigh_update(neigh, NULL, NUD_FAILED,
 					   NEIGH_UPDATE_F_OVERRIDE|
 					   NEIGH_UPDATE_F_ADMIN, 0);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 3bb890a40ed7..574ff450c4d2 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -563,7 +563,7 @@ static int fib_detect_death(struct fib_info *fi, int order,
 		n = NULL;
 
 	if (n) {
-		state = n->nud_state;
+		state = READ_ONCE(n->nud_state);
 		neigh_release(n);
 	} else {
 		return 0;
@@ -2202,7 +2202,7 @@ static bool fib_good_nh(const struct fib_nh *nh)
 		else
 			n = NULL;
 		if (n)
-			state = n->nud_state;
+			state = READ_ONCE(n->nud_state);
 
 		rcu_read_unlock_bh();
 	}
diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index d8ef05347fd9..e28a99f1996b 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -1128,7 +1128,7 @@ static bool ipv6_good_nh(const struct fib6_nh *nh)
 
 	n = __ipv6_neigh_lookup_noref_stub(nh->fib_nh_dev, &nh->fib_nh_gw6);
 	if (n)
-		state = n->nud_state;
+		state = READ_ONCE(n->nud_state);
 
 	rcu_read_unlock_bh();
 
@@ -1145,7 +1145,7 @@ static bool ipv4_good_nh(const struct fib_nh *nh)
 	n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev,
 				      (__force u32)nh->fib_nh_gw4);
 	if (n)
-		state = n->nud_state;
+		state = READ_ONCE(n->nud_state);
 
 	rcu_read_unlock_bh();
 
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index de6e3515ab4f..232009d216c4 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -784,7 +784,7 @@ static void __ip_do_redirect(struct rtable *rt, struct sk_buff *skb, struct flow
 	if (!n)
 		n = neigh_create(&arp_tbl, &new_gw, rt->dst.dev);
 	if (!IS_ERR(n)) {
-		if (!(n->nud_state & NUD_VALID)) {
+		if (!(READ_ONCE(n->nud_state) & NUD_VALID)) {
 			neigh_event_send(n, NULL);
 		} else {
 			if (fib_lookup(net, fl4, &res, 0) == 0) {
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 4ce3f9d3bc8a..e5ed39a3c65f 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1153,7 +1153,7 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk,
 	rcu_read_lock_bh();
 	n = __ipv6_neigh_lookup_noref(rt->dst.dev,
 				      rt6_nexthop(rt, &fl6->daddr));
-	err = n && !(n->nud_state & NUD_VALID) ? -EINVAL : 0;
+	err = n && !(READ_ONCE(n->nud_state) & NUD_VALID) ? -EINVAL : 0;
 	rcu_read_unlock_bh();
 
 	if (err) {
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index c4be62c99f73..18634ebd20a4 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -745,7 +745,7 @@ static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb)
 		saddr = &ipv6_hdr(skb)->saddr;
 	probes -= NEIGH_VAR(neigh->parms, UCAST_PROBES);
 	if (probes < 0) {
-		if (!(neigh->nud_state & NUD_VALID)) {
+		if (!(READ_ONCE(neigh->nud_state) & NUD_VALID)) {
 			ND_PRINTK(1, dbg,
 				  "%s: trying to ucast probe in NUD_INVALID: %pI6\n",
 				  __func__, target);
@@ -1090,7 +1090,7 @@ static enum skb_drop_reason ndisc_recv_na(struct sk_buff *skb)
 		u8 old_flags = neigh->flags;
 		struct net *net = dev_net(dev);
 
-		if (neigh->nud_state & NUD_FAILED)
+		if (READ_ONCE(neigh->nud_state) & NUD_FAILED)
 			goto out;
 
 		/*
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 0fdb03df2287..25c00c6f5131 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -638,7 +638,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh)
 	idev = __in6_dev_get(dev);
 	neigh = __ipv6_neigh_lookup_noref(dev, nh_gw);
 	if (neigh) {
-		if (neigh->nud_state & NUD_VALID)
+		if (READ_ONCE(neigh->nud_state) & NUD_VALID)
 			goto out;
 
 		write_lock(&neigh->lock);
-- 
cgit v1.2.3-70-g09d2


From c486640aa710ddd06c13a7f7162126e1552e8842 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 13 Mar 2023 20:17:32 +0000
Subject: ipv6: remove one read_lock()/read_unlock() pair in rt6_check_neigh()

rt6_check_neigh() uses read_lock() to protect n->nud_state reading.

This seems overkill and causes false sharing.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv6/route.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 25c00c6f5131..e829bd880384 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -687,16 +687,16 @@ static enum rt6_nud_state rt6_check_neigh(const struct fib6_nh *fib6_nh)
 	neigh = __ipv6_neigh_lookup_noref(fib6_nh->fib_nh_dev,
 					  &fib6_nh->fib_nh_gw6);
 	if (neigh) {
-		read_lock(&neigh->lock);
-		if (neigh->nud_state & NUD_VALID)
+		u8 nud_state = READ_ONCE(neigh->nud_state);
+
+		if (nud_state & NUD_VALID)
 			ret = RT6_NUD_SUCCEED;
 #ifdef CONFIG_IPV6_ROUTER_PREF
-		else if (!(neigh->nud_state & NUD_FAILED))
+		else if (!(nud_state & NUD_FAILED))
 			ret = RT6_NUD_SUCCEED;
 		else
 			ret = RT6_NUD_FAIL_PROBE;
 #endif
-		read_unlock(&neigh->lock);
 	} else {
 		ret = IS_ENABLED(CONFIG_IPV6_ROUTER_PREF) ?
 		      RT6_NUD_SUCCEED : RT6_NUD_FAIL_DO_RR;
-- 
cgit v1.2.3-70-g09d2


From 69444581d0022b8afced2c90c441b7b4d9b8eba9 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Fri, 10 Mar 2023 10:08:08 +0100
Subject: net: dsa: microchip: add ksz_setup_tc_mode() function

Add ksz_setup_tc_mode() to make queue scheduling and shaping
configuration more visible.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz_common.c | 20 ++++++++++++--------
 drivers/net/dsa/microchip/ksz_common.h |  6 ++----
 2 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index 729b36eeb2c4..11f20501524f 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -32,10 +32,6 @@
 #include "ksz9477.h"
 #include "lan937x.h"
 
-#define KSZ_CBS_ENABLE ((MTI_SCHEDULE_STRICT_PRIO << MTI_SCHEDULE_MODE_S) | \
-			(MTI_SHAPING_SRP << MTI_SHAPING_S))
-#define KSZ_CBS_DISABLE ((MTI_SCHEDULE_WRR << MTI_SCHEDULE_MODE_S) |\
-			 (MTI_SHAPING_OFF << MTI_SHAPING_S))
 #define MIB_COUNTER_NUM 0x20
 
 struct ksz_stats_raw {
@@ -3091,6 +3087,14 @@ static int cinc_cal(s32 idle_slope, s32 send_slope, u32 *bw)
 	return 0;
 }
 
+static int ksz_setup_tc_mode(struct ksz_device *dev, int port, u8 scheduler,
+			     u8 shaper)
+{
+	return ksz_pwrite8(dev, port, REG_PORT_MTI_QUEUE_CTRL_0,
+			   FIELD_PREP(MTI_SCHEDULE_MODE_M, scheduler) |
+			   FIELD_PREP(MTI_SHAPING_M, shaper));
+}
+
 static int ksz_setup_tc_cbs(struct dsa_switch *ds, int port,
 			    struct tc_cbs_qopt_offload *qopt)
 {
@@ -3110,8 +3114,8 @@ static int ksz_setup_tc_cbs(struct dsa_switch *ds, int port,
 		return ret;
 
 	if (!qopt->enable)
-		return ksz_pwrite8(dev, port, REG_PORT_MTI_QUEUE_CTRL_0,
-				   KSZ_CBS_DISABLE);
+		return ksz_setup_tc_mode(dev, port, MTI_SCHEDULE_WRR,
+					 MTI_SHAPING_OFF);
 
 	/* High Credit */
 	ret = ksz_pwrite16(dev, port, REG_PORT_MTI_HI_WATER_MARK,
@@ -3136,8 +3140,8 @@ static int ksz_setup_tc_cbs(struct dsa_switch *ds, int port,
 			return ret;
 	}
 
-	return ksz_pwrite8(dev, port, REG_PORT_MTI_QUEUE_CTRL_0,
-			   KSZ_CBS_ENABLE);
+	return ksz_setup_tc_mode(dev, port, MTI_SCHEDULE_STRICT_PRIO,
+				 MTI_SHAPING_SRP);
 }
 
 static int ksz_setup_tc(struct dsa_switch *ds, int port,
diff --git a/drivers/net/dsa/microchip/ksz_common.h b/drivers/net/dsa/microchip/ksz_common.h
index d2d5761d58e9..7d87c9a79fb4 100644
--- a/drivers/net/dsa/microchip/ksz_common.h
+++ b/drivers/net/dsa/microchip/ksz_common.h
@@ -654,12 +654,10 @@ static inline int is_lan937x(struct ksz_device *dev)
 
 #define REG_PORT_MTI_QUEUE_CTRL_0	0x0914
 
-#define MTI_SCHEDULE_MODE_M		0x3
-#define MTI_SCHEDULE_MODE_S		6
+#define MTI_SCHEDULE_MODE_M		GENMASK(7, 6)
 #define MTI_SCHEDULE_STRICT_PRIO	0
 #define MTI_SCHEDULE_WRR		2
-#define MTI_SHAPING_M			0x3
-#define MTI_SHAPING_S			4
+#define MTI_SHAPING_M			GENMASK(5, 4)
 #define MTI_SHAPING_OFF			0
 #define MTI_SHAPING_SRP			1
 #define MTI_SHAPING_TIME_AWARE		2
-- 
cgit v1.2.3-70-g09d2


From c570f861fa059ea653599415a7c8cc1dfaf16763 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Fri, 10 Mar 2023 10:08:09 +0100
Subject: net: dsa: microchip: add ETS Qdisc support for KSZ9477 series

Add ETS Qdisc support for KSZ9477 of switches. Current implementation is
limited to strict priority mode.

Tested on KSZ8563R with following configuration:
tc qdisc replace dev lan2 root handle 1: ets strict 4 \
  priomap 3 3 2 2 1 1 0 0
ip link add link lan2 name v1 type vlan id 1 \
  egress-qos-map 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7

and patched iperf3 version:
https://github.com/esnet/iperf/pull/1476
iperf3 -c 172.17.0.1 -b100M  -l1472 -t100 -u -R --sock-prio 2

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz_common.c | 218 +++++++++++++++++++++++++++++++++
 drivers/net/dsa/microchip/ksz_common.h |  12 ++
 2 files changed, 230 insertions(+)

diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index 11f20501524f..5a7ce2aede68 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -1085,6 +1085,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 3,
 		.num_tx_queues = 4,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &ksz9477_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -1224,6 +1225,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 4,
 		.num_tx_queues = 4,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &ksz9477_dev_ops,
 		.phy_errata_9477 = true,
 		.mib_names = ksz9477_mib_names,
@@ -1348,6 +1350,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 3,
 		.num_tx_queues = 4,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &ksz9477_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -1375,6 +1378,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 3,
 		.num_tx_queues = 4,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &ksz9477_dev_ops,
 		.phy_errata_9477 = true,
 		.mib_names = ksz9477_mib_names,
@@ -1407,6 +1411,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 6,
 		.num_tx_queues = 8,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &lan937x_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -1433,6 +1438,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 6,
 		.num_tx_queues = 8,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &lan937x_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -1459,6 +1465,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 6,
 		.num_tx_queues = 8,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &lan937x_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -1489,6 +1496,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 6,
 		.num_tx_queues = 8,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &lan937x_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -1519,6 +1527,7 @@ const struct ksz_chip_data ksz_switch_chips[] = {
 		.port_nirqs = 6,
 		.num_tx_queues = 8,
 		.tc_cbs_supported = true,
+		.tc_ets_supported = true,
 		.ops = &lan937x_dev_ops,
 		.mib_names = ksz9477_mib_names,
 		.mib_cnt = ARRAY_SIZE(ksz9477_mib_names),
@@ -3144,12 +3153,221 @@ static int ksz_setup_tc_cbs(struct dsa_switch *ds, int port,
 				 MTI_SHAPING_SRP);
 }
 
+static int ksz_disable_egress_rate_limit(struct ksz_device *dev, int port)
+{
+	int queue, ret;
+
+	/* Configuration will not take effect until the last Port Queue X
+	 * Egress Limit Control Register is written.
+	 */
+	for (queue = 0; queue < dev->info->num_tx_queues; queue++) {
+		ret = ksz_pwrite8(dev, port, KSZ9477_REG_PORT_OUT_RATE_0 + queue,
+				  KSZ9477_OUT_RATE_NO_LIMIT);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int ksz_ets_band_to_queue(struct tc_ets_qopt_offload_replace_params *p,
+				 int band)
+{
+	/* Compared to queues, bands prioritize packets differently. In strict
+	 * priority mode, the lowest priority is assigned to Queue 0 while the
+	 * highest priority is given to Band 0.
+	 */
+	return p->bands - 1 - band;
+}
+
+static int ksz_queue_set_strict(struct ksz_device *dev, int port, int queue)
+{
+	int ret;
+
+	ret = ksz_pwrite32(dev, port, REG_PORT_MTI_QUEUE_INDEX__4, queue);
+	if (ret)
+		return ret;
+
+	return ksz_setup_tc_mode(dev, port, MTI_SCHEDULE_STRICT_PRIO,
+				 MTI_SHAPING_OFF);
+}
+
+static int ksz_queue_set_wrr(struct ksz_device *dev, int port, int queue,
+			     int weight)
+{
+	int ret;
+
+	ret = ksz_pwrite32(dev, port, REG_PORT_MTI_QUEUE_INDEX__4, queue);
+	if (ret)
+		return ret;
+
+	ret = ksz_setup_tc_mode(dev, port, MTI_SCHEDULE_WRR,
+				MTI_SHAPING_OFF);
+	if (ret)
+		return ret;
+
+	return ksz_pwrite8(dev, port, KSZ9477_PORT_MTI_QUEUE_CTRL_1, weight);
+}
+
+static int ksz_tc_ets_add(struct ksz_device *dev, int port,
+			  struct tc_ets_qopt_offload_replace_params *p)
+{
+	int ret, band, tc_prio;
+	u32 queue_map = 0;
+
+	/* In order to ensure proper prioritization, it is necessary to set the
+	 * rate limit for the related queue to zero. Otherwise strict priority
+	 * or WRR mode will not work. This is a hardware limitation.
+	 */
+	ret = ksz_disable_egress_rate_limit(dev, port);
+	if (ret)
+		return ret;
+
+	/* Configure queue scheduling mode for all bands. Currently only strict
+	 * prio mode is supported.
+	 */
+	for (band = 0; band < p->bands; band++) {
+		int queue = ksz_ets_band_to_queue(p, band);
+
+		ret = ksz_queue_set_strict(dev, port, queue);
+		if (ret)
+			return ret;
+	}
+
+	/* Configure the mapping between traffic classes and queues. Note:
+	 * priomap variable support 16 traffic classes, but the chip can handle
+	 * only 8 classes.
+	 */
+	for (tc_prio = 0; tc_prio < ARRAY_SIZE(p->priomap); tc_prio++) {
+		int queue;
+
+		if (tc_prio > KSZ9477_MAX_TC_PRIO)
+			break;
+
+		queue = ksz_ets_band_to_queue(p, p->priomap[tc_prio]);
+		queue_map |= queue << (tc_prio * KSZ9477_PORT_TC_MAP_S);
+	}
+
+	return ksz_pwrite32(dev, port, KSZ9477_PORT_MRI_TC_MAP__4, queue_map);
+}
+
+static int ksz_tc_ets_del(struct ksz_device *dev, int port)
+{
+	int ret, queue, tc_prio, s;
+	u32 queue_map = 0;
+
+	/* To restore the default chip configuration, set all queues to use the
+	 * WRR scheduler with a weight of 1.
+	 */
+	for (queue = 0; queue < dev->info->num_tx_queues; queue++) {
+		ret = ksz_queue_set_wrr(dev, port, queue,
+					KSZ9477_DEFAULT_WRR_WEIGHT);
+		if (ret)
+			return ret;
+	}
+
+	switch (dev->info->num_tx_queues) {
+	case 2:
+		s = 2;
+		break;
+	case 4:
+		s = 1;
+		break;
+	case 8:
+		s = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Revert the queue mapping for TC-priority to its default setting on
+	 * the chip.
+	 */
+	for (tc_prio = 0; tc_prio <= KSZ9477_MAX_TC_PRIO; tc_prio++) {
+		int queue;
+
+		queue = tc_prio >> s;
+		queue_map |= queue << (tc_prio * KSZ9477_PORT_TC_MAP_S);
+	}
+
+	return ksz_pwrite32(dev, port, KSZ9477_PORT_MRI_TC_MAP__4, queue_map);
+}
+
+static int ksz_tc_ets_validate(struct ksz_device *dev, int port,
+			       struct tc_ets_qopt_offload_replace_params *p)
+{
+	int band;
+
+	/* Since it is not feasible to share one port among multiple qdisc,
+	 * the user must configure all available queues appropriately.
+	 */
+	if (p->bands != dev->info->num_tx_queues) {
+		dev_err(dev->dev, "Not supported amount of bands. It should be %d\n",
+			dev->info->num_tx_queues);
+		return -EOPNOTSUPP;
+	}
+
+	for (band = 0; band < p->bands; ++band) {
+		/* The KSZ switches utilize a weighted round robin configuration
+		 * where a certain number of packets can be transmitted from a
+		 * queue before the next queue is serviced. For more information
+		 * on this, refer to section 5.2.8.4 of the KSZ8565R
+		 * documentation on the Port Transmit Queue Control 1 Register.
+		 * However, the current ETS Qdisc implementation (as of February
+		 * 2023) assigns a weight to each queue based on the number of
+		 * bytes or extrapolated bandwidth in percentages. Since this
+		 * differs from the KSZ switches' method and we don't want to
+		 * fake support by converting bytes to packets, it is better to
+		 * return an error instead.
+		 */
+		if (p->quanta[band]) {
+			dev_err(dev->dev, "Quanta/weights configuration is not supported.\n");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	return 0;
+}
+
+static int ksz_tc_setup_qdisc_ets(struct dsa_switch *ds, int port,
+				  struct tc_ets_qopt_offload *qopt)
+{
+	struct ksz_device *dev = ds->priv;
+	int ret;
+
+	if (!dev->info->tc_ets_supported)
+		return -EOPNOTSUPP;
+
+	if (qopt->parent != TC_H_ROOT) {
+		dev_err(dev->dev, "Parent should be \"root\"\n");
+		return -EOPNOTSUPP;
+	}
+
+	switch (qopt->command) {
+	case TC_ETS_REPLACE:
+		ret = ksz_tc_ets_validate(dev, port, &qopt->replace_params);
+		if (ret)
+			return ret;
+
+		return ksz_tc_ets_add(dev, port, &qopt->replace_params);
+	case TC_ETS_DESTROY:
+		return ksz_tc_ets_del(dev, port);
+	case TC_ETS_STATS:
+	case TC_ETS_GRAFT:
+		return -EOPNOTSUPP;
+	}
+
+	return -EOPNOTSUPP;
+}
+
 static int ksz_setup_tc(struct dsa_switch *ds, int port,
 			enum tc_setup_type type, void *type_data)
 {
 	switch (type) {
 	case TC_SETUP_QDISC_CBS:
 		return ksz_setup_tc_cbs(ds, port, type_data);
+	case TC_SETUP_QDISC_ETS:
+		return ksz_tc_setup_qdisc_ets(ds, port, type_data);
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/dsa/microchip/ksz_common.h b/drivers/net/dsa/microchip/ksz_common.h
index 7d87c9a79fb4..8abecaf6089e 100644
--- a/drivers/net/dsa/microchip/ksz_common.h
+++ b/drivers/net/dsa/microchip/ksz_common.h
@@ -51,6 +51,7 @@ struct ksz_chip_data {
 	u8 port_nirqs;
 	u8 num_tx_queues;
 	bool tc_cbs_supported;
+	bool tc_ets_supported;
 	const struct ksz_dev_ops *ops;
 	bool phy_errata_9477;
 	bool ksz87xx_eee_link_erratum;
@@ -649,6 +650,14 @@ static inline int is_lan937x(struct ksz_device *dev)
 #define KSZ8_LEGAL_PACKET_SIZE		1518
 #define KSZ9477_MAX_FRAME_SIZE		9000
 
+#define KSZ9477_REG_PORT_OUT_RATE_0	0x0420
+#define KSZ9477_OUT_RATE_NO_LIMIT	0
+
+#define KSZ9477_PORT_MRI_TC_MAP__4	0x0808
+
+#define KSZ9477_PORT_TC_MAP_S		4
+#define KSZ9477_MAX_TC_PRIO		7
+
 /* CBS related registers */
 #define REG_PORT_MTI_QUEUE_INDEX__4	0x0900
 
@@ -662,6 +671,9 @@ static inline int is_lan937x(struct ksz_device *dev)
 #define MTI_SHAPING_SRP			1
 #define MTI_SHAPING_TIME_AWARE		2
 
+#define KSZ9477_PORT_MTI_QUEUE_CTRL_1	0x0915
+#define KSZ9477_DEFAULT_WRR_WEIGHT	1
+
 #define REG_PORT_MTI_HI_WATER_MARK	0x0916
 #define REG_PORT_MTI_LO_WATER_MARK	0x0918
 
-- 
cgit v1.2.3-70-g09d2


From dc54e450a5dd0fd9147dc9fea0684293569d3609 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 10 Mar 2023 22:46:28 +0100
Subject: net: stmmac: qcom: drop of_match_ptr for ID table

The driver is specific to ARCH_QCOM which depends on OF thus the driver
is OF-only.  Its of_device_id table is built unconditionally, thus
of_match_ptr() for ID table does not make sense.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index 732774645c1a..32763566c214 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -606,7 +606,7 @@ static struct platform_driver qcom_ethqos_driver = {
 	.driver = {
 		.name           = "qcom-ethqos",
 		.pm		= &stmmac_pltfr_pm_ops,
-		.of_match_table = of_match_ptr(qcom_ethqos_match),
+		.of_match_table = qcom_ethqos_match,
 	},
 };
 module_platform_driver(qcom_ethqos_driver);
-- 
cgit v1.2.3-70-g09d2


From e6512465838bf634d147399c8c96612d3cb419b9 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 10 Mar 2023 22:46:29 +0100
Subject: net: stmmac: generic: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).

  drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c:72:34: error: ‘dwmac_generic_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c
index 5e731a72cce8..ef8f3a940938 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c
@@ -91,7 +91,7 @@ static struct platform_driver dwmac_generic_driver = {
 	.driver = {
 		.name           = STMMAC_RESOURCE_NAME,
 		.pm		= &stmmac_pltfr_pm_ops,
-		.of_match_table = of_match_ptr(dwmac_generic_match),
+		.of_match_table = dwmac_generic_match,
 	},
 };
 module_platform_driver(dwmac_generic_driver);
-- 
cgit v1.2.3-70-g09d2


From 69df36d524dbb6f0692252c5f9d41bed9a064ff5 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 10 Mar 2023 22:46:30 +0100
Subject: net: marvell: pxa168_eth: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).

  drivers/net/ethernet/marvell/pxa168_eth.c:1575:34: error: ‘pxa168_eth_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/pxa168_eth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c
index 87fff539d39d..d5691b6a2bc5 100644
--- a/drivers/net/ethernet/marvell/pxa168_eth.c
+++ b/drivers/net/ethernet/marvell/pxa168_eth.c
@@ -1586,7 +1586,7 @@ static struct platform_driver pxa168_eth_driver = {
 	.suspend = pxa168_eth_suspend,
 	.driver = {
 		.name		= DRIVER_NAME,
-		.of_match_table	= of_match_ptr(pxa168_eth_of_match),
+		.of_match_table	= pxa168_eth_of_match,
 	},
 };
 
-- 
cgit v1.2.3-70-g09d2


From 7f319fe4363c613f8b00b4bfab66c3659c38378b Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 10 Mar 2023 22:46:31 +0100
Subject: net: samsung: sxgbe: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).

  drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c:220:34: error: ‘sxgbe_dt_ids’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c
index 926532466691..4e5526303f07 100644
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c
+++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c
@@ -229,7 +229,7 @@ static struct platform_driver sxgbe_platform_driver = {
 	.driver	= {
 		.name		= SXGBE_RESOURCE_NAME,
 		.pm		= &sxgbe_platform_pm_ops,
-		.of_match_table	= of_match_ptr(sxgbe_dt_ids),
+		.of_match_table	= sxgbe_dt_ids,
 	},
 };
 
-- 
cgit v1.2.3-70-g09d2


From 7e9aa8cad0844b85a6877cd072c95ac2844eb666 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 10 Mar 2023 22:46:32 +0100
Subject: net: ni: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).

  drivers/net/ethernet/ni/nixge.c:1253:34: error: ‘nixge_dt_ids’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ni/nixge.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c
index 56e02cba0b8a..0fd156286d4d 100644
--- a/drivers/net/ethernet/ni/nixge.c
+++ b/drivers/net/ethernet/ni/nixge.c
@@ -1422,7 +1422,7 @@ static struct platform_driver nixge_driver = {
 	.remove		= nixge_remove,
 	.driver		= {
 		.name		= "nixge",
-		.of_match_table	= of_match_ptr(nixge_dt_ids),
+		.of_match_table	= nixge_dt_ids,
 	},
 };
 module_platform_driver(nixge_driver);
-- 
cgit v1.2.3-70-g09d2


From a52ed50a04de179cc866c18bdd1b9a2f59d6bdfc Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 12:13:28 +0100
Subject: nfc: trf7970a: mark OF related data as maybe unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can be compile tested with !CONFIG_OF making certain data
unused:

  drivers/nfc/trf7970a.c:2232:34: error: ‘trf7970a_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Mark Greer <mgreer@animalcreek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/nfc/trf7970a.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/nfc/trf7970a.c b/drivers/nfc/trf7970a.c
index 21d68664fe08..7eb17f46a815 100644
--- a/drivers/nfc/trf7970a.c
+++ b/drivers/nfc/trf7970a.c
@@ -2229,7 +2229,7 @@ static const struct dev_pm_ops trf7970a_pm_ops = {
 			   trf7970a_pm_runtime_resume, NULL)
 };
 
-static const struct of_device_id trf7970a_of_match[] = {
+static const struct of_device_id trf7970a_of_match[] __maybe_unused = {
 	{.compatible = "ti,trf7970a",},
 	{},
 };
-- 
cgit v1.2.3-70-g09d2


From 6ea1e67788f30710d1991de42802cbdeb67afec4 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:52 +0100
Subject: net: dsa: lantiq_gswip: mark OF related data as maybe unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can be compile tested with !CONFIG_OF making certain data
unused:

  drivers/net/dsa/lantiq_gswip.c:1888:34: error: ‘xway_gphy_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/lantiq_gswip.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 05ecaa007ab1..3c76a1a14aee 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -1885,7 +1885,7 @@ static const struct xway_gphy_match_data xrx300_gphy_data = {
 	.ge_firmware_name = "lantiq/xrx300_phy11g_a21.bin",
 };
 
-static const struct of_device_id xway_gphy_match[] = {
+static const struct of_device_id xway_gphy_match[] __maybe_unused = {
 	{ .compatible = "lantiq,xrx200-gphy-fw", .data = NULL },
 	{ .compatible = "lantiq,xrx200a1x-gphy-fw", .data = &xrx200a1x_gphy_data },
 	{ .compatible = "lantiq,xrx200a2x-gphy-fw", .data = &xrx200a2x_gphy_data },
-- 
cgit v1.2.3-70-g09d2


From ced5c5a0a2ea5eec6ed6cf1fcdde719fdafed82c Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:53 +0100
Subject: net: dsa: lan9303: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver will match mostly or only by DT table (even thought there is
regular ID table) so there is little benefit in of_match_ptr (this also
allows ACPI matching via PRP0001, even though it might not be relevant
here).

  drivers/net/dsa/lan9303_i2c.c:97:34: error: ‘lan9303_i2c_of_match’ defined but not used [-Werror=unused-const-variable=]
  drivers/net/dsa/lan9303_mdio.c:157:34: error: ‘lan9303_mdio_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/lan9303_i2c.c  | 2 +-
 drivers/net/dsa/lan9303_mdio.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/lan9303_i2c.c b/drivers/net/dsa/lan9303_i2c.c
index 1cb41c36bd47..e8844820c3a9 100644
--- a/drivers/net/dsa/lan9303_i2c.c
+++ b/drivers/net/dsa/lan9303_i2c.c
@@ -103,7 +103,7 @@ MODULE_DEVICE_TABLE(of, lan9303_i2c_of_match);
 static struct i2c_driver lan9303_i2c_driver = {
 	.driver = {
 		.name = "LAN9303_I2C",
-		.of_match_table = of_match_ptr(lan9303_i2c_of_match),
+		.of_match_table = lan9303_i2c_of_match,
 	},
 	.probe_new = lan9303_i2c_probe,
 	.remove = lan9303_i2c_remove,
diff --git a/drivers/net/dsa/lan9303_mdio.c b/drivers/net/dsa/lan9303_mdio.c
index 4f33369a2de5..d8ab2b77d201 100644
--- a/drivers/net/dsa/lan9303_mdio.c
+++ b/drivers/net/dsa/lan9303_mdio.c
@@ -164,7 +164,7 @@ MODULE_DEVICE_TABLE(of, lan9303_mdio_of_match);
 static struct mdio_driver lan9303_mdio_driver = {
 	.mdiodrv.driver = {
 		.name = "LAN9303_MDIO",
-		.of_match_table = of_match_ptr(lan9303_mdio_of_match),
+		.of_match_table = lan9303_mdio_of_match,
 	},
 	.probe  = lan9303_mdio_probe,
 	.remove = lan9303_mdio_remove,
-- 
cgit v1.2.3-70-g09d2


From 1eb8566dd08db9c7402a26b027086eaabc20ff2a Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:54 +0100
Subject: net: dsa: seville_vsc9953: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it might not be relevant here).

  drivers/net/dsa/ocelot/seville_vsc9953.c:1070:34: error: ‘seville_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/seville_vsc9953.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/ocelot/seville_vsc9953.c b/drivers/net/dsa/ocelot/seville_vsc9953.c
index 563ad338da25..96d4972a62f0 100644
--- a/drivers/net/dsa/ocelot/seville_vsc9953.c
+++ b/drivers/net/dsa/ocelot/seville_vsc9953.c
@@ -1079,7 +1079,7 @@ static struct platform_driver seville_vsc9953_driver = {
 	.shutdown	= seville_shutdown,
 	.driver = {
 		.name		= "mscc_seville",
-		.of_match_table	= of_match_ptr(seville_of_match),
+		.of_match_table	= seville_of_match,
 	},
 };
 module_platform_driver(seville_vsc9953_driver);
-- 
cgit v1.2.3-70-g09d2


From 00923ff2e1baf6bc9832f12e44f73439de5fcd72 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:55 +0100
Subject: net: dsa: ksz9477: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver will match mostly by DT table (even thought there is regular
ID table) so there is little benefit in of_match_ptr (this also allows
ACPI matching via PRP0001, even though it might not be relevant here).

  drivers/net/dsa/microchip/ksz9477_i2c.c:84:34: error: ‘ksz9477_dt_ids’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz9477_i2c.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/microchip/ksz9477_i2c.c b/drivers/net/dsa/microchip/ksz9477_i2c.c
index e315f669ec06..97a317263a2f 100644
--- a/drivers/net/dsa/microchip/ksz9477_i2c.c
+++ b/drivers/net/dsa/microchip/ksz9477_i2c.c
@@ -117,7 +117,7 @@ MODULE_DEVICE_TABLE(of, ksz9477_dt_ids);
 static struct i2c_driver ksz9477_i2c_driver = {
 	.driver = {
 		.name	= "ksz9477-switch",
-		.of_match_table = of_match_ptr(ksz9477_dt_ids),
+		.of_match_table = ksz9477_dt_ids,
 	},
 	.probe_new = ksz9477_i2c_probe,
 	.remove	= ksz9477_i2c_remove,
-- 
cgit v1.2.3-70-g09d2


From 0f17b42827ae63aaf08cad875d9575b1bca1e066 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:56 +0100
Subject: net: dsa: ocelot: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it might not be relevant here).

  drivers/net/dsa/ocelot/ocelot_ext.c:143:34: error: ‘ocelot_ext_switch_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/ocelot_ext.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/ocelot/ocelot_ext.c b/drivers/net/dsa/ocelot/ocelot_ext.c
index 063150659816..228737a32080 100644
--- a/drivers/net/dsa/ocelot/ocelot_ext.c
+++ b/drivers/net/dsa/ocelot/ocelot_ext.c
@@ -149,7 +149,7 @@ MODULE_DEVICE_TABLE(of, ocelot_ext_switch_of_match);
 static struct platform_driver ocelot_ext_switch_driver = {
 	.driver = {
 		.name = "ocelot-ext-switch",
-		.of_match_table = of_match_ptr(ocelot_ext_switch_of_match),
+		.of_match_table = ocelot_ext_switch_of_match,
 	},
 	.probe = ocelot_ext_probe,
 	.remove = ocelot_ext_remove,
-- 
cgit v1.2.3-70-g09d2


From b0b7d1b6260b3acb9f3f3bc7f3ca88a8898e95b1 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:57 +0100
Subject: net: phy: ks8995: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver will match mostly by DT table (even thought there is regular
ID table) so there is little benefit in of_match_ptr (this also allows
ACPI matching via PRP0001, even though it might not be relevant here).

  drivers/net/phy/spi_ks8995.c:156:34: error: ‘ks8895_spi_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/spi_ks8995.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index d4202d40d47a..7196e927c2cd 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -491,7 +491,7 @@ static void ks8995_remove(struct spi_device *spi)
 static struct spi_driver ks8995_driver = {
 	.driver = {
 		.name	    = "spi-ks8995",
-		.of_match_table = of_match_ptr(ks8895_spi_of_match),
+		.of_match_table = ks8895_spi_of_match,
 	},
 	.probe	  = ks8995_probe,
 	.remove	  = ks8995_remove,
-- 
cgit v1.2.3-70-g09d2


From 3df09beef650af667be5ec7eee88629e6e479a51 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:58 +0100
Subject: net: ieee802154: adf7242: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver will match mostly by DT table (even thought there is regular
ID table) so there is little benefit in of_match_ptr (this also allows
ACPI matching via PRP0001, even though it might not be relevant here).

  drivers/net/ieee802154/adf7242.c:1322:34: error: ‘adf7242_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ieee802154/adf7242.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ieee802154/adf7242.c b/drivers/net/ieee802154/adf7242.c
index 5cf218c674a5..509acc86001c 100644
--- a/drivers/net/ieee802154/adf7242.c
+++ b/drivers/net/ieee802154/adf7242.c
@@ -1336,7 +1336,7 @@ MODULE_DEVICE_TABLE(spi, adf7242_device_id);
 static struct spi_driver adf7242_driver = {
 	.id_table = adf7242_device_id,
 	.driver = {
-		   .of_match_table = of_match_ptr(adf7242_of_match),
+		   .of_match_table = adf7242_of_match,
 		   .name = "adf7242",
 		   .owner = THIS_MODULE,
 		   },
-- 
cgit v1.2.3-70-g09d2


From 3896c40b7824c8874963e257afc0f464200d2d2c Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:32:59 +0100
Subject: net: ieee802154: mcr20a: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver will match mostly by DT table (even thought there is regular
ID table) so there is little benefit in of_match_ptr (this also allows
ACPI matching via PRP0001, even though it might not be relevant here).

  drivers/net/ieee802154/mcr20a.c:1340:34: error: ‘mcr20a_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Stefan Schmidt <stefan@datenfreihafen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ieee802154/mcr20a.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ieee802154/mcr20a.c b/drivers/net/ieee802154/mcr20a.c
index f53d185e0568..87abe3b46316 100644
--- a/drivers/net/ieee802154/mcr20a.c
+++ b/drivers/net/ieee802154/mcr20a.c
@@ -1352,7 +1352,7 @@ MODULE_DEVICE_TABLE(spi, mcr20a_device_id);
 static struct spi_driver mcr20a_driver = {
 	.id_table = mcr20a_device_id,
 	.driver = {
-		.of_match_table = of_match_ptr(mcr20a_of_match),
+		.of_match_table = mcr20a_of_match,
 		.name	= "mcr20a",
 	},
 	.probe      = mcr20a_probe,
-- 
cgit v1.2.3-70-g09d2


From 32b7030681a488186bf1753b29a170f5d4644721 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:33:00 +0100
Subject: net: ieee802154: at86rf230: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver will match mostly by DT table (even thought there is regular
ID table) so there is little benefit in of_match_ptr (this also allows
ACPI matching via PRP0001, even though it might not be relevant here).

  drivers/net/ieee802154/at86rf230.c:1644:34: error: ‘at86rf230_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Stefan Schmidt <stefan@datenfreihafen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ieee802154/at86rf230.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ieee802154/at86rf230.c b/drivers/net/ieee802154/at86rf230.c
index 62b984f84d9f..164c7f605af5 100644
--- a/drivers/net/ieee802154/at86rf230.c
+++ b/drivers/net/ieee802154/at86rf230.c
@@ -1662,7 +1662,7 @@ MODULE_DEVICE_TABLE(spi, at86rf230_device_id);
 static struct spi_driver at86rf230_driver = {
 	.id_table = at86rf230_device_id,
 	.driver = {
-		.of_match_table = of_match_ptr(at86rf230_of_match),
+		.of_match_table = at86rf230_of_match,
 		.name	= "at86rf230",
 	},
 	.probe      = at86rf230_probe,
-- 
cgit v1.2.3-70-g09d2


From cdfe4fc4d946e2a5d589f73985043674cf298ff6 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:33:01 +0100
Subject: net: ieee802154: ca8210: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it might not be relevant here).

  drivers/net/ieee802154/ca8210.c:3174:34: error: ‘ca8210_of_ids’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Stefan Schmidt <stefan@datenfreihafen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ieee802154/ca8210.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ieee802154/ca8210.c b/drivers/net/ieee802154/ca8210.c
index 0b0c6c0764fe..b5def77e37d7 100644
--- a/drivers/net/ieee802154/ca8210.c
+++ b/drivers/net/ieee802154/ca8210.c
@@ -3180,7 +3180,7 @@ static struct spi_driver ca8210_spi_driver = {
 	.driver = {
 		.name =                 DRIVER_NAME,
 		.owner =                THIS_MODULE,
-		.of_match_table =       of_match_ptr(ca8210_of_ids),
+		.of_match_table =       ca8210_of_ids,
 	},
 	.probe  =                       ca8210_probe,
 	.remove =                       ca8210_remove
-- 
cgit v1.2.3-70-g09d2


From 059fa99723405b47833c7e14e062c7ba7066ca3c Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:33:02 +0100
Subject: net: ieee802154: adf7242: drop owner from driver

Core already sets owner in spi_driver.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ieee802154/adf7242.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ieee802154/adf7242.c b/drivers/net/ieee802154/adf7242.c
index 509acc86001c..f9972b8140f9 100644
--- a/drivers/net/ieee802154/adf7242.c
+++ b/drivers/net/ieee802154/adf7242.c
@@ -1338,7 +1338,6 @@ static struct spi_driver adf7242_driver = {
 	.driver = {
 		   .of_match_table = adf7242_of_match,
 		   .name = "adf7242",
-		   .owner = THIS_MODULE,
 		   },
 	.probe = adf7242_probe,
 	.remove = adf7242_remove,
-- 
cgit v1.2.3-70-g09d2


From 613a3c44a3737c9a2dfd9f70f9b79427459bb745 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 18:33:03 +0100
Subject: net: ieee802154: ca8210: drop owner from driver

Core already sets owner in spi_driver.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Stefan Schmidt <stefan@datenfreihafen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ieee802154/ca8210.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ieee802154/ca8210.c b/drivers/net/ieee802154/ca8210.c
index b5def77e37d7..1eada1db8dcf 100644
--- a/drivers/net/ieee802154/ca8210.c
+++ b/drivers/net/ieee802154/ca8210.c
@@ -3179,7 +3179,6 @@ MODULE_DEVICE_TABLE(of, ca8210_of_ids);
 static struct spi_driver ca8210_spi_driver = {
 	.driver = {
 		.name =                 DRIVER_NAME,
-		.owner =                THIS_MODULE,
 		.of_match_table =       ca8210_of_ids,
 	},
 	.probe  =                       ca8210_probe,
-- 
cgit v1.2.3-70-g09d2


From 45ef71d108e6f1545e2981d8af03be13f2e39411 Mon Sep 17 00:00:00 2001
From: Josef Miegl <josef@miegl.cz>
Date: Sun, 12 Mar 2023 17:45:57 +0100
Subject: net: geneve: set IFF_POINTOPOINT with IFLA_GENEVE_INNER_PROTO_INHERIT

The GENEVE tunnel used with IFLA_GENEVE_INNER_PROTO_INHERIT is
point-to-point, so set IFF_POINTOPOINT to reflect that.

Signed-off-by: Josef Miegl <josef@miegl.cz>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/geneve.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 89ff7f8e8c7e..5c3c7169c0a2 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -1426,7 +1426,7 @@ static int geneve_configure(struct net *net, struct net_device *dev,
 		dev->type = ARPHRD_NONE;
 		dev->hard_header_len = 0;
 		dev->addr_len = 0;
-		dev->flags = IFF_NOARP;
+		dev->flags = IFF_POINTOPOINT | IFF_NOARP;
 	}
 
 	err = register_netdevice(dev);
-- 
cgit v1.2.3-70-g09d2


From f947568e258038d3c2f8f38a9a7dabaca36643ec Mon Sep 17 00:00:00 2001
From: Stefan Raspl <raspl@linux.ibm.com>
Date: Mon, 13 Mar 2023 11:10:31 +0100
Subject: net/smc: Introduce explicit check for v2 support

Previously, v2 support was derived from a very specific format of the SEID
as part of the SMC-D codebase. Make this part of the SMC-D device API, so
implementers do not need to adhere to a specific SEID format.

Signed-off-by: Stefan Raspl <raspl@linux.ibm.com>
Reviewed-and-tested-by: Jan Karcher <jaka@linux.ibm.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/ism_drv.c | 7 +++++++
 include/net/smc.h          | 1 +
 net/smc/smc_ism.c          | 2 +-
 3 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/s390/net/ism_drv.c b/drivers/s390/net/ism_drv.c
index eb7e13486087..1c73d32966f1 100644
--- a/drivers/s390/net/ism_drv.c
+++ b/drivers/s390/net/ism_drv.c
@@ -842,6 +842,12 @@ static int smcd_move(struct smcd_dev *smcd, u64 dmb_tok, unsigned int idx,
 	return ism_move(smcd->priv, dmb_tok, idx, sf, offset, data, size);
 }
 
+static int smcd_supports_v2(void)
+{
+	return SYSTEM_EID.serial_number[0] != '0' ||
+		SYSTEM_EID.type[0] != '0';
+}
+
 static u64 smcd_get_local_gid(struct smcd_dev *smcd)
 {
 	return ism_get_local_gid(smcd->priv);
@@ -869,6 +875,7 @@ static const struct smcd_ops ism_ops = {
 	.reset_vlan_required = smcd_reset_vlan_required,
 	.signal_event = smcd_signal_ieq,
 	.move_data = smcd_move,
+	.supports_v2 = smcd_supports_v2,
 	.get_system_eid = ism_get_seid,
 	.get_local_gid = smcd_get_local_gid,
 	.get_chid = smcd_get_chid,
diff --git a/include/net/smc.h b/include/net/smc.h
index 597cb9381182..a002552be29c 100644
--- a/include/net/smc.h
+++ b/include/net/smc.h
@@ -67,6 +67,7 @@ struct smcd_ops {
 	int (*move_data)(struct smcd_dev *dev, u64 dmb_tok, unsigned int idx,
 			 bool sf, unsigned int offset, void *data,
 			 unsigned int size);
+	int (*supports_v2)(void);
 	u8* (*get_system_eid)(void);
 	u64 (*get_local_gid)(struct smcd_dev *dev);
 	u16 (*get_chid)(struct smcd_dev *dev);
diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c
index 3b0b7710c6b0..fbee2493091f 100644
--- a/net/smc/smc_ism.c
+++ b/net/smc/smc_ism.c
@@ -429,7 +429,7 @@ static void smcd_register_dev(struct ism_dev *ism)
 		u8 *system_eid = NULL;
 
 		system_eid = smcd->ops->get_system_eid();
-		if (system_eid[24] != '0' || system_eid[28] != '0') {
+		if (smcd->ops->supports_v2()) {
 			smc_ism_v2_capable = true;
 			memcpy(smc_ism_v2_system_eid, system_eid,
 			       SMC_MAX_EID_LEN);
-- 
cgit v1.2.3-70-g09d2


From 298c91dc40e51395198533433716d0521ed10995 Mon Sep 17 00:00:00 2001
From: Stefan Raspl <raspl@linux.ibm.com>
Date: Mon, 13 Mar 2023 11:10:32 +0100
Subject: net/ism: Remove extra include

Signed-off-by: Stefan Raspl <raspl@linux.ibm.com>
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/ism_drv.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/s390/net/ism_drv.c b/drivers/s390/net/ism_drv.c
index 1c73d32966f1..05749c877990 100644
--- a/drivers/s390/net/ism_drv.c
+++ b/drivers/s390/net/ism_drv.c
@@ -11,7 +11,6 @@
 #include <linux/types.h>
 #include <linux/interrupt.h>
 #include <linux/device.h>
-#include <linux/pci.h>
 #include <linux/err.h>
 #include <linux/ctype.h>
 #include <linux/processor.h>
-- 
cgit v1.2.3-70-g09d2


From a02d83f9947d8f71904eda4de046630c3eb6802c Mon Sep 17 00:00:00 2001
From: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Date: Mon, 13 Mar 2023 12:32:11 +0100
Subject: scm: fix MSG_CTRUNC setting condition for SO_PASSSEC

Currently, kernel would set MSG_CTRUNC flag if msg_control buffer
wasn't provided and SO_PASSCRED was set or if there was pending SCM_RIGHTS.

For some reason we have no corresponding check for SO_PASSSEC.

In the recvmsg(2) doc we have:
       MSG_CTRUNC
              indicates that some control data was discarded due to lack
              of space in the buffer for ancillary data.

So, we need to set MSG_CTRUNC flag for all types of SCM.

This change can break applications those don't check MSG_CTRUNC flag.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: Leon Romanovsky <leon@kernel.org>
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>

v2:
- commit message was rewritten according to Eric's suggestion
Acked-by: Paul Moore <paul@paul-moore.com>

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/scm.h | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/include/net/scm.h b/include/net/scm.h
index 1ce365f4c256..585adc1346bd 100644
--- a/include/net/scm.h
+++ b/include/net/scm.h
@@ -105,16 +105,27 @@ static inline void scm_passec(struct socket *sock, struct msghdr *msg, struct sc
 		}
 	}
 }
+
+static inline bool scm_has_secdata(struct socket *sock)
+{
+	return test_bit(SOCK_PASSSEC, &sock->flags);
+}
 #else
 static inline void scm_passec(struct socket *sock, struct msghdr *msg, struct scm_cookie *scm)
 { }
+
+static inline bool scm_has_secdata(struct socket *sock)
+{
+	return false;
+}
 #endif /* CONFIG_SECURITY_NETWORK */
 
 static __inline__ void scm_recv(struct socket *sock, struct msghdr *msg,
 				struct scm_cookie *scm, int flags)
 {
 	if (!msg->msg_control) {
-		if (test_bit(SOCK_PASSCRED, &sock->flags) || scm->fp)
+		if (test_bit(SOCK_PASSCRED, &sock->flags) || scm->fp ||
+		    scm_has_secdata(sock))
 			msg->msg_flags |= MSG_CTRUNC;
 		scm_destroy(scm);
 		return;
-- 
cgit v1.2.3-70-g09d2


From 3b50d9a174575473cc5ed8eba3570e02ea3d0733 Mon Sep 17 00:00:00 2001
From: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Date: Tue, 14 Mar 2023 05:18:48 +0100
Subject: MAINTAINERS: adjust file entries after wifi driver movement

Commit f79cbc77abde ("wifi: move mac80211_hwsim and virt_wifi to virtual
directory") and commit 298e50ad8eb8 ("wifi: move raycs, wl3501 and
rndis_wlan to legacy directory") move remaining wireless drivers into
subdirectories, but does not adjust the entries in MAINTAINERS.

Hence, ./scripts/get_maintainer.pl --self-test=patterns complains about
broken references.

Repair these file references in those wireless driver sections.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230314041848.5120-1-lukas.bulwahn@gmail.com
---
 MAINTAINERS | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 8d5bc223f305..0dab503c88ac 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -12275,7 +12275,7 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless.git
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git
 F:	Documentation/networking/mac80211-injection.rst
 F:	Documentation/networking/mac80211_hwsim/mac80211_hwsim.rst
-F:	drivers/net/wireless/mac80211_hwsim.[ch]
+F:	drivers/net/wireless/virtual/mac80211_hwsim.[ch]
 F:	include/net/mac80211.h
 F:	net/mac80211/
 
@@ -17542,7 +17542,7 @@ F:	include/ras/ras_event.h
 RAYLINK/WEBGEAR 802.11 WIRELESS LAN DRIVER
 L:	linux-wireless@vger.kernel.org
 S:	Orphan
-F:	drivers/net/wireless/ray*
+F:	drivers/net/wireless/legacy/ray*
 
 RC-CORE / LIRC FRAMEWORK
 M:	Sean Young <sean@mess.org>
@@ -21797,7 +21797,7 @@ USB WIRELESS RNDIS DRIVER (rndis_wlan)
 M:	Jussi Kivilinna <jussi.kivilinna@iki.fi>
 L:	linux-wireless@vger.kernel.org
 S:	Maintained
-F:	drivers/net/wireless/rndis_wlan.c
+F:	drivers/net/wireless/legacy/rndis_wlan.c
 
 USB XHCI DRIVER
 M:	Mathias Nyman <mathias.nyman@intel.com>
@@ -22551,7 +22551,7 @@ F:	drivers/input/misc/wistron_btns.c
 WL3501 WIRELESS PCMCIA CARD DRIVER
 L:	linux-wireless@vger.kernel.org
 S:	Odd fixes
-F:	drivers/net/wireless/wl3501*
+F:	drivers/net/wireless/legacy/wl3501*
 
 WOLFSON MICROELECTRONICS DRIVERS
 L:	patches@opensource.cirrus.com
-- 
cgit v1.2.3-70-g09d2


From f94d7a3a51070d185e9203b53a3d669e27931e3d Mon Sep 17 00:00:00 2001
From: Bastian Germann <bage@debian.org>
Date: Mon, 6 Mar 2023 13:50:40 +0100
Subject: wifi: ath9k: Remove Qwest/Actiontec 802AIN ID
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The USB device 1668:1200 is Qwest/Actiontec 802AIN which is also
correctly claimed to be supported by carl9170.

Supposedly, the successor 802AIN2 has an ath9k compatible chip
whose USB ID (unknown) could be inserted instead.

Drop the ID from the wrong driver. I happened to find this by chance while
packaging the ath9k and carl9170 firmware for Debian.

Signed-off-by: Bastian Germann <bage@debian.org>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230306125041.2221-1-bage@debian.org
---
 drivers/net/wireless/ath/ath9k/hif_usb.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index e0130beb304d..27ff1ca2631f 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -42,8 +42,6 @@ static const struct usb_device_id ath9k_hif_usb_ids[] = {
 
 	{ USB_DEVICE(0x0cf3, 0x7015),
 	  .driver_info = AR9287_USB },  /* Atheros */
-	{ USB_DEVICE(0x1668, 0x1200),
-	  .driver_info = AR9287_USB },  /* Verizon */
 
 	{ USB_DEVICE(0x0cf3, 0x7010),
 	  .driver_info = AR9280_USB },  /* Atheros */
-- 
cgit v1.2.3-70-g09d2


From f94557154d9fc77c392844523388edd4661a27a3 Mon Sep 17 00:00:00 2001
From: Vladimir Lypak <vladimir.lypak@gmail.com>
Date: Sat, 11 Mar 2023 20:36:47 +0530
Subject: wifi: wcn36xx: add support for pronto-v3

Pronto v3 has a different DXE address than prior Pronto versions. This
patch changes the macro to return the correct register address based on
the pronto version.

Signed-off-by: Vladimir Lypak <vladimir.lypak@gmail.com>
Signed-off-by: Sireesh Kodali <sireeshkodali1@gmail.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230311150647.22935-2-sireeshkodali1@gmail.com
---
 drivers/net/wireless/ath/wcn36xx/dxe.c     | 23 ++++++++++++-----------
 drivers/net/wireless/ath/wcn36xx/dxe.h     |  4 ++--
 drivers/net/wireless/ath/wcn36xx/main.c    |  1 +
 drivers/net/wireless/ath/wcn36xx/wcn36xx.h |  1 +
 4 files changed, 16 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/ath/wcn36xx/dxe.c b/drivers/net/wireless/ath/wcn36xx/dxe.c
index 4e9e13941c8f..9013f056eecb 100644
--- a/drivers/net/wireless/ath/wcn36xx/dxe.c
+++ b/drivers/net/wireless/ath/wcn36xx/dxe.c
@@ -112,8 +112,8 @@ int wcn36xx_dxe_alloc_ctl_blks(struct wcn36xx *wcn)
 	wcn->dxe_rx_l_ch.desc_num = WCN36XX_DXE_CH_DESC_NUMB_RX_L;
 	wcn->dxe_rx_h_ch.desc_num = WCN36XX_DXE_CH_DESC_NUMB_RX_H;
 
-	wcn->dxe_tx_l_ch.dxe_wq =  WCN36XX_DXE_WQ_TX_L;
-	wcn->dxe_tx_h_ch.dxe_wq =  WCN36XX_DXE_WQ_TX_H;
+	wcn->dxe_tx_l_ch.dxe_wq =  WCN36XX_DXE_WQ_TX_L(wcn);
+	wcn->dxe_tx_h_ch.dxe_wq =  WCN36XX_DXE_WQ_TX_H(wcn);
 
 	wcn->dxe_tx_l_ch.ctrl_bd = WCN36XX_DXE_CTRL_TX_L_BD;
 	wcn->dxe_tx_h_ch.ctrl_bd = WCN36XX_DXE_CTRL_TX_H_BD;
@@ -165,8 +165,9 @@ void wcn36xx_dxe_free_ctl_blks(struct wcn36xx *wcn)
 	wcn36xx_dxe_free_ctl_block(&wcn->dxe_rx_h_ch);
 }
 
-static int wcn36xx_dxe_init_descs(struct device *dev, struct wcn36xx_dxe_ch *wcn_ch)
+static int wcn36xx_dxe_init_descs(struct wcn36xx *wcn, struct wcn36xx_dxe_ch *wcn_ch)
 {
+	struct device *dev = wcn->dev;
 	struct wcn36xx_dxe_desc *cur_dxe = NULL;
 	struct wcn36xx_dxe_desc *prev_dxe = NULL;
 	struct wcn36xx_dxe_ctl *cur_ctl = NULL;
@@ -190,11 +191,11 @@ static int wcn36xx_dxe_init_descs(struct device *dev, struct wcn36xx_dxe_ch *wcn
 		switch (wcn_ch->ch_type) {
 		case WCN36XX_DXE_CH_TX_L:
 			cur_dxe->ctrl = WCN36XX_DXE_CTRL_TX_L;
-			cur_dxe->dst_addr_l = WCN36XX_DXE_WQ_TX_L;
+			cur_dxe->dst_addr_l = WCN36XX_DXE_WQ_TX_L(wcn);
 			break;
 		case WCN36XX_DXE_CH_TX_H:
 			cur_dxe->ctrl = WCN36XX_DXE_CTRL_TX_H;
-			cur_dxe->dst_addr_l = WCN36XX_DXE_WQ_TX_H;
+			cur_dxe->dst_addr_l = WCN36XX_DXE_WQ_TX_H(wcn);
 			break;
 		case WCN36XX_DXE_CH_RX_L:
 			cur_dxe->ctrl = WCN36XX_DXE_CTRL_RX_L;
@@ -914,7 +915,7 @@ int wcn36xx_dxe_init(struct wcn36xx *wcn)
 	/***************************************/
 	/* Init descriptors for TX LOW channel */
 	/***************************************/
-	ret = wcn36xx_dxe_init_descs(wcn->dev, &wcn->dxe_tx_l_ch);
+	ret = wcn36xx_dxe_init_descs(wcn, &wcn->dxe_tx_l_ch);
 	if (ret) {
 		dev_err(wcn->dev, "Error allocating descriptor\n");
 		return ret;
@@ -928,14 +929,14 @@ int wcn36xx_dxe_init(struct wcn36xx *wcn)
 	/* Program DMA destination addr for TX LOW */
 	wcn36xx_dxe_write_register(wcn,
 		WCN36XX_DXE_CH_DEST_ADDR_TX_L,
-		WCN36XX_DXE_WQ_TX_L);
+		WCN36XX_DXE_WQ_TX_L(wcn));
 
 	wcn36xx_dxe_read_register(wcn, WCN36XX_DXE_REG_CH_EN, &reg_data);
 
 	/***************************************/
 	/* Init descriptors for TX HIGH channel */
 	/***************************************/
-	ret = wcn36xx_dxe_init_descs(wcn->dev, &wcn->dxe_tx_h_ch);
+	ret = wcn36xx_dxe_init_descs(wcn, &wcn->dxe_tx_h_ch);
 	if (ret) {
 		dev_err(wcn->dev, "Error allocating descriptor\n");
 		goto out_err_txh_ch;
@@ -950,14 +951,14 @@ int wcn36xx_dxe_init(struct wcn36xx *wcn)
 	/* Program DMA destination addr for TX HIGH */
 	wcn36xx_dxe_write_register(wcn,
 		WCN36XX_DXE_CH_DEST_ADDR_TX_H,
-		WCN36XX_DXE_WQ_TX_H);
+		WCN36XX_DXE_WQ_TX_H(wcn));
 
 	wcn36xx_dxe_read_register(wcn, WCN36XX_DXE_REG_CH_EN, &reg_data);
 
 	/***************************************/
 	/* Init descriptors for RX LOW channel */
 	/***************************************/
-	ret = wcn36xx_dxe_init_descs(wcn->dev, &wcn->dxe_rx_l_ch);
+	ret = wcn36xx_dxe_init_descs(wcn, &wcn->dxe_rx_l_ch);
 	if (ret) {
 		dev_err(wcn->dev, "Error allocating descriptor\n");
 		goto out_err_rxl_ch;
@@ -988,7 +989,7 @@ int wcn36xx_dxe_init(struct wcn36xx *wcn)
 	/***************************************/
 	/* Init descriptors for RX HIGH channel */
 	/***************************************/
-	ret = wcn36xx_dxe_init_descs(wcn->dev, &wcn->dxe_rx_h_ch);
+	ret = wcn36xx_dxe_init_descs(wcn, &wcn->dxe_rx_h_ch);
 	if (ret) {
 		dev_err(wcn->dev, "Error allocating descriptor\n");
 		goto out_err_rxh_ch;
diff --git a/drivers/net/wireless/ath/wcn36xx/dxe.h b/drivers/net/wireless/ath/wcn36xx/dxe.h
index 26a31edf52e9..dd8c684a3ba7 100644
--- a/drivers/net/wireless/ath/wcn36xx/dxe.h
+++ b/drivers/net/wireless/ath/wcn36xx/dxe.h
@@ -135,8 +135,8 @@ H2H_TEST_RX_TX = DMA2
 	WCN36xx_DXE_CTRL_ENDIANNESS)
 
 /* TODO This must calculated properly but not hardcoded */
-#define WCN36XX_DXE_WQ_TX_L			0x17
-#define WCN36XX_DXE_WQ_TX_H			0x17
+#define WCN36XX_DXE_WQ_TX_L(wcn)    ((wcn)->is_pronto_v3 ? 0x6 : 0x17)
+#define WCN36XX_DXE_WQ_TX_H(wcn)    ((wcn)->is_pronto_v3 ? 0x6 : 0x17)
 #define WCN36XX_DXE_WQ_RX_L			0xB
 #define WCN36XX_DXE_WQ_RX_H			0x4
 
diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c
index 3b79cc1c7c5b..8dbd115a393c 100644
--- a/drivers/net/wireless/ath/wcn36xx/main.c
+++ b/drivers/net/wireless/ath/wcn36xx/main.c
@@ -1508,6 +1508,7 @@ static int wcn36xx_platform_get_resources(struct wcn36xx *wcn,
 	}
 
 	wcn->is_pronto = !!of_device_is_compatible(mmio_node, "qcom,pronto");
+	wcn->is_pronto_v3 = !!of_device_is_compatible(mmio_node, "qcom,pronto-v3-pil");
 
 	/* Map the CCU memory */
 	index = of_property_match_string(mmio_node, "reg-names", "ccu");
diff --git a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h
index 9aa08b636d08..ff4a8e5d7209 100644
--- a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h
+++ b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h
@@ -217,6 +217,7 @@ struct wcn36xx {
 	u8			fw_major;
 	u32			fw_feat_caps[WCN36XX_HAL_CAPS_SIZE];
 	bool			is_pronto;
+	bool			is_pronto_v3;
 
 	/* extra byte for the NULL termination */
 	u8			crm_version[WCN36XX_HAL_VERSION_LENGTH + 1];
-- 
cgit v1.2.3-70-g09d2


From 488d9a484f96eee4f0e8e108aed42a057a1c7295 Mon Sep 17 00:00:00 2001
From: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Date: Tue, 7 Mar 2023 02:47:06 -0800
Subject: wifi: ath12k: Add missing unwind goto in ath12k_pci_probe()

Smatch Warns:
	drivers/net/wireless/ath/ath12k/pci.c:1198 ath12k_pci_probe()
	warn: missing unwind goto?

Store the error value in ret and use correct label with a goto.

Only Compile tested, found with Smatch.

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/all/Y+426q6cfkEdb5Bv@kili/
Suggested-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230307104706.240119-1-harshit.m.mogalapalli@oracle.com
---
 drivers/net/wireless/ath/ath12k/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c
index d32637b0113d..00b0080dbac3 100644
--- a/drivers/net/wireless/ath/ath12k/pci.c
+++ b/drivers/net/wireless/ath/ath12k/pci.c
@@ -1223,7 +1223,8 @@ static int ath12k_pci_probe(struct pci_dev *pdev,
 			dev_err(&pdev->dev,
 				"Unknown hardware version found for QCN9274: 0x%x\n",
 				soc_hw_version_major);
-			return -EOPNOTSUPP;
+			ret = -EOPNOTSUPP;
+			goto err_pci_free_region;
 		}
 		break;
 	case WCN7850_DEVICE_ID:
-- 
cgit v1.2.3-70-g09d2


From 3b1088a09ec9438523c251d8435e78988824bc0d Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Tue, 7 Mar 2023 16:22:39 -0600
Subject: wifi: ath11k: Replace fake flex-array with flexible-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Address 25 of the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
drivers/net/wireless/ath/ath11k/debugfs_htt_stats.c:30:51: warning: array subscript <unknown> is outside array bounds of ‘const u32[0]’ {aka ‘const unsigned int[]’} [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/266
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/ZAe5L5DtmsQxzqRH@work
---
 .../net/wireless/ath/ath11k/debugfs_htt_stats.h    | 73 +++++++++++++---------
 1 file changed, 43 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath11k/debugfs_htt_stats.h
index 2b97cbbd28cb..0bbd58a380de 100644
--- a/drivers/net/wireless/ath/ath11k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath11k/debugfs_htt_stats.h
@@ -143,7 +143,8 @@ enum htt_tx_pdev_underrun_enum {
 /* Bytes stored in little endian order */
 /* Length should be multiple of DWORD */
 struct htt_stats_string_tlv {
-	u32 data[0]; /* Can be variable length */
+	 /* Can be variable length */
+	DECLARE_FLEX_ARRAY(u32, data);
 } __packed;
 
 #define HTT_STATS_MAC_ID	GENMASK(7, 0)
@@ -205,27 +206,32 @@ struct htt_tx_pdev_stats_cmn_tlv {
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_pdev_stats_urrn_tlv_v {
-	u32 urrn_stats[0]; /* HTT_TX_PDEV_MAX_URRN_STATS */
+	/* HTT_TX_PDEV_MAX_URRN_STATS */
+	DECLARE_FLEX_ARRAY(u32, urrn_stats);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_pdev_stats_flush_tlv_v {
-	u32 flush_errs[0]; /* HTT_TX_PDEV_MAX_FLUSH_REASON_STATS */
+	/* HTT_TX_PDEV_MAX_FLUSH_REASON_STATS */
+	DECLARE_FLEX_ARRAY(u32, flush_errs);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_pdev_stats_sifs_tlv_v {
-	u32 sifs_status[0]; /* HTT_TX_PDEV_MAX_SIFS_BURST_STATS */
+	/* HTT_TX_PDEV_MAX_SIFS_BURST_STATS */
+	DECLARE_FLEX_ARRAY(u32, sifs_status);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_pdev_stats_phy_err_tlv_v {
-	u32  phy_errs[0]; /* HTT_TX_PDEV_MAX_PHY_ERR_STATS */
+	/* HTT_TX_PDEV_MAX_PHY_ERR_STATS */
+	DECLARE_FLEX_ARRAY(u32, phy_errs);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_pdev_stats_sifs_hist_tlv_v {
-	u32 sifs_hist_status[0]; /* HTT_TX_PDEV_SIFS_BURST_HIST_STATS */
+	/* HTT_TX_PDEV_SIFS_BURST_HIST_STATS */
+	DECLARE_FLEX_ARRAY(u32, sifs_hist_status);
 };
 
 struct htt_tx_pdev_stats_tx_ppdu_stats_tlv_v {
@@ -590,20 +596,20 @@ struct htt_tx_hwq_difs_latency_stats_tlv_v {
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_hwq_cmd_result_stats_tlv_v {
-	/* Histogram of sched cmd result */
-	u32 cmd_result[0]; /* HTT_TX_HWQ_MAX_CMD_RESULT_STATS */
+	/* Histogram of sched cmd result, HTT_TX_HWQ_MAX_CMD_RESULT_STATS */
+	DECLARE_FLEX_ARRAY(u32, cmd_result);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_hwq_cmd_stall_stats_tlv_v {
-	/* Histogram of various pause conitions */
-	u32 cmd_stall_status[0]; /* HTT_TX_HWQ_MAX_CMD_STALL_STATS */
+	/* Histogram of various pause conitions, HTT_TX_HWQ_MAX_CMD_STALL_STATS */
+	DECLARE_FLEX_ARRAY(u32, cmd_stall_status);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_hwq_fes_result_stats_tlv_v {
-	/* Histogram of number of user fes result */
-	u32 fes_result[0]; /* HTT_TX_HWQ_MAX_FES_RESULT_STATS */
+	/* Histogram of number of user fes result, HTT_TX_HWQ_MAX_FES_RESULT_STATS */
+	DECLARE_FLEX_ARRAY(u32, fes_result);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size
@@ -635,8 +641,8 @@ struct htt_tx_hwq_tried_mpdu_cnt_hist_tlv_v {
  * #define WAL_TXOP_USED_HISTOGRAM_INTERVAL 1000 ( 1 ms )
  */
 struct htt_tx_hwq_txop_used_cnt_hist_tlv_v {
-	/* Histogram of txop used cnt */
-	u32 txop_used_cnt_hist[0]; /* HTT_TX_HWQ_TXOP_USED_CNT_HIST */
+	/* Histogram of txop used cnt,  HTT_TX_HWQ_TXOP_USED_CNT_HIST */
+	DECLARE_FLEX_ARRAY(u32, txop_used_cnt_hist);
 };
 
 /* == TX SELFGEN STATS == */
@@ -804,17 +810,20 @@ struct htt_tx_pdev_mpdu_stats_tlv {
 /* == TX SCHED STATS == */
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_sched_txq_cmd_posted_tlv_v {
-	u32 sched_cmd_posted[0]; /* HTT_TX_PDEV_SCHED_TX_MODE_MAX */
+	/* HTT_TX_PDEV_SCHED_TX_MODE_MAX */
+	DECLARE_FLEX_ARRAY(u32, sched_cmd_posted);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_sched_txq_cmd_reaped_tlv_v {
-	u32 sched_cmd_reaped[0]; /* HTT_TX_PDEV_SCHED_TX_MODE_MAX */
+	/* HTT_TX_PDEV_SCHED_TX_MODE_MAX */
+	DECLARE_FLEX_ARRAY(u32, sched_cmd_reaped);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_sched_txq_sched_order_su_tlv_v {
-	u32 sched_order_su[0]; /* HTT_TX_PDEV_NUM_SCHED_ORDER_LOG */
+	/* HTT_TX_PDEV_NUM_SCHED_ORDER_LOG */
+	DECLARE_FLEX_ARRAY(u32, sched_order_su);
 };
 
 enum htt_sched_txq_sched_ineligibility_tlv_enum {
@@ -842,7 +851,7 @@ enum htt_sched_txq_sched_ineligibility_tlv_enum {
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_sched_txq_sched_ineligibility_tlv_v {
 	/* indexed by htt_sched_txq_sched_ineligibility_tlv_enum */
-	u32 sched_ineligibility[0];
+	DECLARE_FLEX_ARRAY(u32, sched_ineligibility);
 };
 
 #define	HTT_TX_PDEV_STATS_SCHED_PER_TXQ_MAC_ID	GENMASK(7, 0)
@@ -888,18 +897,20 @@ struct htt_stats_tx_sched_cmn_tlv {
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_tqm_gen_mpdu_stats_tlv_v {
-	u32 gen_mpdu_end_reason[0]; /* HTT_TX_TQM_MAX_GEN_MPDU_END_REASON */
+	/* HTT_TX_TQM_MAX_GEN_MPDU_END_REASON */
+	DECLARE_FLEX_ARRAY(u32, gen_mpdu_end_reason);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_tqm_list_mpdu_stats_tlv_v {
-	u32 list_mpdu_end_reason[0]; /* HTT_TX_TQM_MAX_LIST_MPDU_END_REASON */
+	 /* HTT_TX_TQM_MAX_LIST_MPDU_END_REASON */
+	DECLARE_FLEX_ARRAY(u32, list_mpdu_end_reason);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_tx_tqm_list_mpdu_cnt_tlv_v {
-	u32 list_mpdu_cnt_hist[0];
-			/* HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS */
+	/* HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS */
+	DECLARE_FLEX_ARRAY(u32, list_mpdu_cnt_hist);
 };
 
 struct htt_tx_tqm_pdev_stats_tlv_v {
@@ -1098,7 +1109,7 @@ struct htt_tx_de_compl_stats_tlv {
  *                               ENTRIES_PER_BIN_COUNT)
  */
 struct htt_tx_de_fw2wbm_ring_full_hist_tlv {
-	u32 fw2wbm_ring_full_hist[0];
+	DECLARE_FLEX_ARRAY(u32, fw2wbm_ring_full_hist);
 };
 
 struct htt_tx_de_cmn_stats_tlv {
@@ -1151,7 +1162,7 @@ struct htt_ring_if_cmn_tlv {
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_sfm_client_user_tlv_v {
 	/* Number of DWORDS used per user and per client */
-	u32 dwords_used_by_user_n[0];
+	DECLARE_FLEX_ARRAY(u32, dwords_used_by_user_n);
 };
 
 struct htt_sfm_client_tlv {
@@ -1436,12 +1447,14 @@ struct htt_rx_soc_fw_stats_tlv {
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_rx_soc_fw_refill_ring_empty_tlv_v {
-	u32 refill_ring_empty_cnt[0]; /* HTT_RX_STATS_REFILL_MAX_RING */
+	/* HTT_RX_STATS_REFILL_MAX_RING */
+	DECLARE_FLEX_ARRAY(u32, refill_ring_empty_cnt);
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_rx_soc_fw_refill_ring_num_refill_tlv_v {
-	u32 refill_ring_num_refill[0]; /* HTT_RX_STATS_REFILL_MAX_RING */
+	/* HTT_RX_STATS_REFILL_MAX_RING */
+	DECLARE_FLEX_ARRAY(u32, refill_ring_num_refill);
 };
 
 /* RXDMA error code from WBM released packets */
@@ -1473,7 +1486,7 @@ enum htt_rx_rxdma_error_code_enum {
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_rx_soc_fw_refill_ring_num_rxdma_err_tlv_v {
-	u32 rxdma_err[0]; /* HTT_RX_RXDMA_MAX_ERR_CODE */
+	DECLARE_FLEX_ARRAY(u32, rxdma_err); /* HTT_RX_RXDMA_MAX_ERR_CODE */
 };
 
 /* REO error code from WBM released packets */
@@ -1505,7 +1518,7 @@ enum htt_rx_reo_error_code_enum {
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_rx_soc_fw_refill_ring_num_reo_err_tlv_v {
-	u32 reo_err[0]; /* HTT_RX_REO_MAX_ERR_CODE */
+	DECLARE_FLEX_ARRAY(u32, reo_err); /* HTT_RX_REO_MAX_ERR_CODE */
 };
 
 /* == RX PDEV STATS == */
@@ -1622,13 +1635,13 @@ struct htt_rx_pdev_fw_stats_phy_err_tlv {
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_rx_pdev_fw_ring_mpdu_err_tlv_v {
 	/* Num error MPDU for each RxDMA error type  */
-	u32 fw_ring_mpdu_err[0]; /* HTT_RX_STATS_RXDMA_MAX_ERR */
+	DECLARE_FLEX_ARRAY(u32, fw_ring_mpdu_err); /* HTT_RX_STATS_RXDMA_MAX_ERR */
 };
 
 /* NOTE: Variable length TLV, use length spec to infer array size */
 struct htt_rx_pdev_fw_mpdu_drop_tlv_v {
 	/* Num MPDU dropped  */
-	u32 fw_mpdu_drop[0]; /* HTT_RX_STATS_FW_DROP_REASON_MAX */
+	DECLARE_FLEX_ARRAY(u32, fw_mpdu_drop); /* HTT_RX_STATS_FW_DROP_REASON_MAX */
 };
 
 #define HTT_PDEV_CCA_STATS_TX_FRAME_INFO_PRESENT               (0x1)
-- 
cgit v1.2.3-70-g09d2


From 5a78ac33e3cb8822da64dd1af196e83664b332b0 Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Thu, 9 Mar 2023 15:23:08 +0530
Subject: wifi: ath11k: fix deinitialization of firmware resources

Currently, in ath11k_ahb_fw_resources_init(), iommu domain
mapping is done only for the chipsets having fixed firmware
memory. Also, for such chipsets, mapping is done only if it
does not have TrustZone support.

During deinitialization, only if TrustZone support is not there,
iommu is unmapped back. However, for non fixed firmware memory
chipsets, TrustZone support is not there and this makes the
condition check to true and it tries to unmap the memory which
was not mapped during initialization.

This leads to the following trace -

[   83.198790] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000008
[   83.259537] Modules linked in: ath11k_ahb ath11k qmi_helpers
.. snip ..
[   83.280286] pstate: 20000005 (nzCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[   83.287228] pc : __iommu_unmap+0x30/0x140
[   83.293907] lr : iommu_unmap+0x5c/0xa4
[   83.298072] sp : ffff80000b3abad0
.. snip ..
[   83.369175] Call trace:
[   83.376282]  __iommu_unmap+0x30/0x140
[   83.378541]  iommu_unmap+0x5c/0xa4
[   83.382360]  ath11k_ahb_fw_resource_deinit.part.12+0x2c/0xac [ath11k_ahb]
[   83.385666]  ath11k_ahb_free_resources+0x140/0x17c [ath11k_ahb]
[   83.392521]  ath11k_ahb_shutdown+0x34/0x40 [ath11k_ahb]
[   83.398248]  platform_shutdown+0x20/0x2c
[   83.403455]  device_shutdown+0x16c/0x1c4
[   83.407621]  kernel_restart_prepare+0x34/0x3c
[   83.411529]  kernel_restart+0x14/0x74
[   83.415781]  __do_sys_reboot+0x1c4/0x22c
[   83.419427]  __arm64_sys_reboot+0x1c/0x24
[   83.423420]  invoke_syscall+0x44/0xfc
[   83.427326]  el0_svc_common.constprop.3+0xac/0xe8
[   83.430974]  do_el0_svc+0xa0/0xa8
[   83.435659]  el0_svc+0x1c/0x44
[   83.438957]  el0t_64_sync_handler+0x60/0x144
[   83.441910]  el0t_64_sync+0x15c/0x160
[   83.446343] Code: aa0103f4 f9400001 f90027a1 d2800001 (f94006a0)
[   83.449903] ---[ end trace 0000000000000000 ]---

This can be reproduced by probing an AHB chipset which is not
having a fixed memory region. During reboot (or rmmod) trace
can be seen.

Fix this issue by adding a condition check on firmware fixed memory
hw_param as done in the counter initialization function.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Fixes: f9eec4947add ("ath11k: Add support for targets without trustzone")
Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230309095308.24937-1-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath11k/ahb.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
index b549576d0b51..5cbba9a8b6ba 100644
--- a/drivers/net/wireless/ath/ath11k/ahb.c
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -1078,6 +1078,12 @@ static int ath11k_ahb_fw_resource_deinit(struct ath11k_base *ab)
 	struct iommu_domain *iommu;
 	size_t unmapped_size;
 
+	/* Chipsets not requiring MSA would have not initialized
+	 * MSA resources, return success in such cases.
+	 */
+	if (!ab->hw_params.fixed_fw_mem)
+		return 0;
+
 	if (ab_ahb->fw.use_tz)
 		return 0;
 
-- 
cgit v1.2.3-70-g09d2


From 0936998393c7dffd68ba5d73eabdabf8dddbbc41 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 14 Mar 2023 19:49:13 +0200
Subject: wifi: iwlwifi: mvm: avoid sta lookup in queue alloc

In FW restart scenarios, we allocate the queues from the
iwl_mvm_realloc_queues_after_restart() function, but that
is called before we insert the station ID into our map
(mvm->fw_id_to_mac_id).

However, in all cases where we're actually allocating a
queue for a "real" (not bcast, aux, ...) station we have
the sta pointer already, so just pass it along to use it
instead of looking it up.

This fixes an issue where after restart we only allocated
a queue of size 16 (due to the ordering issue described
above), and thus never got good throughput again since no
aggregates could be formed on transmit.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.4d70868003e8.I3476fee5c12f5b1af2be5e2f38a9df7d66d02b62@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 50 +++++++++++++++-------------
 1 file changed, 27 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 69634fb82a9b..46af2b422849 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -766,7 +766,25 @@ static int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 sta_id,
 	return -ENOSPC;
 }
 
+static int iwl_mvm_get_queue_size(struct ieee80211_sta *sta)
+{
+	/* this queue isn't used for traffic (cab_queue) */
+	if (!sta)
+		return IWL_MGMT_QUEUE_SIZE;
+
+	/* support for 1k ba size */
+	if (sta->deflink.eht_cap.has_eht)
+		return IWL_DEFAULT_QUEUE_SIZE_EHT;
+
+	/* support for 256 ba size */
+	if (sta->deflink.he_cap.has_he)
+		return IWL_DEFAULT_QUEUE_SIZE_HE;
+
+	return IWL_DEFAULT_QUEUE_SIZE;
+}
+
 static int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
+				   struct ieee80211_sta *sta,
 				   u8 sta_id, u8 tid, unsigned int timeout)
 {
 	int queue, size;
@@ -776,22 +794,7 @@ static int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 		size = max_t(u32, IWL_MGMT_QUEUE_SIZE,
 			     mvm->trans->cfg->min_txq_size);
 	} else {
-		struct ieee80211_sta *sta;
-
-		rcu_read_lock();
-		sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]);
-
-		/* this queue isn't used for traffic (cab_queue) */
-		if (IS_ERR_OR_NULL(sta)) {
-			size = IWL_MGMT_QUEUE_SIZE;
-		} else if (sta->deflink.he_cap.has_he) {
-			/* support for 256 ba size */
-			size = IWL_DEFAULT_QUEUE_SIZE_HE;
-		} else {
-			size = IWL_DEFAULT_QUEUE_SIZE;
-		}
-
-		rcu_read_unlock();
+		size = iwl_mvm_get_queue_size(sta);
 	}
 
 	/* take the min with bc tbl entries allowed */
@@ -836,7 +839,8 @@ static int iwl_mvm_sta_alloc_queue_tvqm(struct iwl_mvm *mvm,
 	IWL_DEBUG_TX_QUEUES(mvm,
 			    "Allocating queue for sta %d on tid %d\n",
 			    mvmsta->sta_id, tid);
-	queue = iwl_mvm_tvqm_enable_txq(mvm, mvmsta->sta_id, tid, wdg_timeout);
+	queue = iwl_mvm_tvqm_enable_txq(mvm, sta, mvmsta->sta_id,
+					tid, wdg_timeout);
 	if (queue < 0)
 		return queue;
 
@@ -1537,7 +1541,8 @@ static void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
 			IWL_DEBUG_TX_QUEUES(mvm,
 					    "Re-mapping sta %d tid %d\n",
 					    mvm_sta->sta_id, i);
-			txq_id = iwl_mvm_tvqm_enable_txq(mvm, mvm_sta->sta_id,
+			txq_id = iwl_mvm_tvqm_enable_txq(mvm, sta,
+							 mvm_sta->sta_id,
 							 i, wdg);
 			/*
 			 * on failures, just set it to IWL_MVM_INVALID_QUEUE
@@ -2049,7 +2054,7 @@ static int iwl_mvm_enable_aux_snif_queue_tvqm(struct iwl_mvm *mvm, u8 sta_id)
 
 	WARN_ON(!iwl_mvm_has_new_tx_api(mvm));
 
-	return iwl_mvm_tvqm_enable_txq(mvm, sta_id, IWL_MAX_TID_COUNT,
+	return iwl_mvm_tvqm_enable_txq(mvm, NULL, sta_id, IWL_MAX_TID_COUNT,
 				       wdg_timeout);
 }
 
@@ -2233,7 +2238,7 @@ int iwl_mvm_send_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 * to firmware so enable queue here - after the station was added
 	 */
 	if (iwl_mvm_has_new_tx_api(mvm)) {
-		queue = iwl_mvm_tvqm_enable_txq(mvm, bsta->sta_id,
+		queue = iwl_mvm_tvqm_enable_txq(mvm, NULL, bsta->sta_id,
 						IWL_MAX_TID_COUNT,
 						wdg_timeout);
 		if (queue < 0) {
@@ -2427,9 +2432,8 @@ int iwl_mvm_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 * tfd_queue_mask.
 	 */
 	if (iwl_mvm_has_new_tx_api(mvm)) {
-		int queue = iwl_mvm_tvqm_enable_txq(mvm, msta->sta_id,
-						    0,
-						    timeout);
+		int queue = iwl_mvm_tvqm_enable_txq(mvm, NULL, msta->sta_id,
+						    0, timeout);
 		if (queue < 0) {
 			ret = queue;
 			goto err;
-- 
cgit v1.2.3-70-g09d2


From 51fa8c026e072dac49ca638d66a64a66b29e916f Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Tue, 14 Mar 2023 19:49:14 +0200
Subject: wifi: iwlwifi: yoyo: Add new tlv for dump file name extension

Add tlv in dump file for dump file name extension.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Greenman, Gregory <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.ffc28212994e.Ie5f10709548497061f95c1634d942dd2facf72ec@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c        | 32 +++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/fw/error-dump.h | 17 +++++++++++-
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h     |  4 +++
 3 files changed, 51 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index abf49022edbe..ca97f2fcb693 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -2320,6 +2320,34 @@ static u32 iwl_dump_ini_info(struct iwl_fw_runtime *fwrt,
 	return entry->size;
 }
 
+static u32 iwl_dump_ini_file_name_info(struct iwl_fw_runtime *fwrt,
+				       struct list_head *list)
+{
+	struct iwl_fw_ini_dump_entry *entry;
+	struct iwl_dump_file_name_info *tlv;
+	u32 len = strnlen(fwrt->trans->dbg.dump_file_name_ext,
+			  IWL_FW_INI_MAX_NAME);
+
+	if (!fwrt->trans->dbg.dump_file_name_ext_valid)
+		return 0;
+
+	entry = vzalloc(sizeof(*entry) + sizeof(*tlv) + len);
+	if (!entry)
+		return 0;
+
+	entry->size = sizeof(*tlv) + len;
+
+	tlv = (void *)entry->data;
+	tlv->type = cpu_to_le32(IWL_INI_DUMP_NAME_TYPE);
+	tlv->len = cpu_to_le32(len);
+	memcpy(tlv->data, fwrt->trans->dbg.dump_file_name_ext, len);
+
+	/* add the dump file name extension tlv to the list */
+	list_add_tail(&entry->list, list);
+
+	return entry->size;
+}
+
 static const struct iwl_dump_ini_mem_ops iwl_dump_ini_region_ops[] = {
 	[IWL_FW_INI_REGION_INVALID] = {},
 	[IWL_FW_INI_REGION_INTERNAL_BUFFER] = {
@@ -2495,8 +2523,10 @@ static u32 iwl_dump_ini_trigger(struct iwl_fw_runtime *fwrt,
 		size += iwl_dump_ini_mem(fwrt, list, &reg_data,
 					 &iwl_dump_ini_region_ops[IWL_FW_INI_REGION_DRAM_IMR]);
 
-	if (size)
+	if (size) {
+		size += iwl_dump_ini_file_name_info(fwrt, list);
 		size += iwl_dump_ini_info(fwrt, trigger, list);
+	}
 
 	return size;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h b/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h
index c62576e442bd..f5e08988dc7b 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2014, 2018-2021 Intel Corporation
+ * Copyright (C) 2014, 2018-2022 Intel Corporation
  * Copyright (C) 2014-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -75,6 +75,18 @@ struct iwl_fw_error_dump_data {
 	__u8 data[];
 } __packed;
 
+/**
+ * struct iwl_dump_file_name_info - data for dump file name addition
+ * @type: region type with reserved bits
+ * @len: the length of file name string to be added to dump file
+ * @data: the string need to be added to dump file
+ */
+struct iwl_dump_file_name_info {
+	__le32 type;
+	__le32 len;
+	__u8 data[];
+} __packed;
+
 /**
  * struct iwl_fw_error_dump_file - the layout of the header of the file
  * @barker: must be %IWL_FW_ERROR_DUMP_BARKER
@@ -231,6 +243,9 @@ struct iwl_fw_error_dump_mem {
 /* Use bit 31 as dump info type to avoid colliding with region types */
 #define IWL_INI_DUMP_INFO_TYPE BIT(31)
 
+/* Use bit 31 and bit 24 as dump name type to avoid colliding with region types */
+#define IWL_INI_DUMP_NAME_TYPE (BIT(31) | BIT(24))
+
 /**
  * struct iwl_fw_error_dump_data - data for one type
  * @type: &enum iwl_fw_ini_region_type
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 9aced3e44bc2..dd277a4fa8dd 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -775,6 +775,8 @@ struct iwl_imr_data {
  * @periodic_trig_list: periodic triggers list
  * @domains_bitmap: bitmap of active domains other than &IWL_FW_INI_DOMAIN_ALWAYS_ON
  * @ucode_preset: preset based on ucode
+ * @dump_file_name_ext: dump file name extension
+ * @dump_file_name_ext_valid: dump file name extension if valid or not
  */
 struct iwl_trans_debug {
 	u8 n_dest_reg;
@@ -813,6 +815,8 @@ struct iwl_trans_debug {
 	bool restart_required;
 	u32 last_tp_resetfw;
 	struct iwl_imr_data imr_data;
+	u8 dump_file_name_ext[IWL_FW_INI_MAX_NAME];
+	bool dump_file_name_ext_valid;
 };
 
 struct iwl_dma_ptr {
-- 
cgit v1.2.3-70-g09d2


From 834f920ef34b57a51fa892fae6de86830316c353 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Tue, 14 Mar 2023 19:49:15 +0200
Subject: wifi: iwlwifi: yoyo: Add driver defined dump file name

Add driver defined dump file name extension for beacon loss
and FW Assert case.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.2a2ee92995e9.I38fff588e32276796cd757309fc811241f827c7a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c    |  2 +-
 drivers/net/wireless/intel/iwlwifi/fw/dump.c      | 58 ++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h   |  4 ++
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c       |  5 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 61 +++++++++++++++--------
 5 files changed, 101 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 05720352e49f..43fb5cf85f05 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -281,7 +281,7 @@ static const struct iwl_ht_params iwl_gl_a_ht_params = {
 	.trans.gen2 = true,						\
 	.nvm_type = IWL_NVM_EXT,					\
 	.dbgc_supported = true,						\
-	.min_umac_error_event_table = 0x400000,				\
+	.min_umac_error_event_table = 0xD0000,				\
 	.d3_debug_data_base_addr = 0x401000,				\
 	.d3_debug_data_length = 60 * 1024,				\
 	.mon_smem_regs = {						\
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dump.c b/drivers/net/wireless/intel/iwlwifi/fw/dump.c
index 792f7fee1840..59ed321bcc27 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dump.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dump.c
@@ -14,6 +14,13 @@
 #include "iwl-csr.h"
 #include "pnvm.h"
 
+#define FW_ASSERT_LMAC_FATAL			0x70
+#define FW_ASSERT_LMAC2_FATAL			0x72
+#define FW_ASSERT_UMAC_FATAL			0x71
+#define UMAC_RT_NMI_LMAC2_FATAL			0x72
+#define RT_NMI_INTERRUPT_OTHER_LMAC_FATAL	0x73
+#define FW_ASSERT_NMI_UNKNOWN			0x84
+
 /*
  * Note: This structure is read from the device with IO accesses,
  * and the reading already does the endian conversion. As it is
@@ -96,6 +103,17 @@ struct iwl_umac_error_event_table {
 #define ERROR_START_OFFSET  (1 * sizeof(u32))
 #define ERROR_ELEM_SIZE     (7 * sizeof(u32))
 
+static bool iwl_fwrt_if_errorid_other_cpu(u32 err_id)
+{
+	err_id &= 0xFF;
+
+	if ((err_id >= FW_ASSERT_LMAC_FATAL &&
+	     err_id <= RT_NMI_INTERRUPT_OTHER_LMAC_FATAL) ||
+	    err_id == FW_ASSERT_NMI_UNKNOWN)
+		return  true;
+	return false;
+}
+
 static void iwl_fwrt_dump_umac_error_log(struct iwl_fw_runtime *fwrt)
 {
 	struct iwl_trans *trans = fwrt->trans;
@@ -113,6 +131,13 @@ static void iwl_fwrt_dump_umac_error_log(struct iwl_fw_runtime *fwrt)
 	if (table.valid)
 		fwrt->dump.umac_err_id = table.error_id;
 
+	if (!iwl_fwrt_if_errorid_other_cpu(fwrt->dump.umac_err_id) &&
+	    !fwrt->trans->dbg.dump_file_name_ext_valid) {
+		fwrt->trans->dbg.dump_file_name_ext_valid = true;
+		snprintf(fwrt->trans->dbg.dump_file_name_ext, IWL_FW_INI_MAX_NAME,
+			 "0x%x", fwrt->dump.umac_err_id);
+	}
+
 	if (ERROR_START_OFFSET <= table.valid * ERROR_ELEM_SIZE) {
 		IWL_ERR(trans, "Start IWL Error Log Dump:\n");
 		IWL_ERR(trans, "Transport status: 0x%08lX, valid: %d\n",
@@ -189,6 +214,13 @@ static void iwl_fwrt_dump_lmac_error_log(struct iwl_fw_runtime *fwrt, u8 lmac_nu
 	if (table.valid)
 		fwrt->dump.lmac_err_id[lmac_num] = table.error_id;
 
+	if (!iwl_fwrt_if_errorid_other_cpu(fwrt->dump.lmac_err_id[lmac_num]) &&
+	    !fwrt->trans->dbg.dump_file_name_ext_valid) {
+		fwrt->trans->dbg.dump_file_name_ext_valid = true;
+		snprintf(fwrt->trans->dbg.dump_file_name_ext, IWL_FW_INI_MAX_NAME,
+			 "0x%x", fwrt->dump.lmac_err_id[lmac_num]);
+	}
+
 	if (ERROR_START_OFFSET <= table.valid * ERROR_ELEM_SIZE) {
 		IWL_ERR(trans, "Start IWL Error Log Dump:\n");
 		IWL_ERR(trans, "Transport status: 0x%08lX, valid: %d\n",
@@ -274,6 +306,16 @@ static void iwl_fwrt_dump_tcm_error_log(struct iwl_fw_runtime *fwrt, int idx)
 
 	iwl_trans_read_mem_bytes(trans, base, &table, sizeof(table));
 
+	if (table.valid)
+		fwrt->dump.tcm_err_id[idx] = table.error_id;
+
+	if (!iwl_fwrt_if_errorid_other_cpu(fwrt->dump.tcm_err_id[idx]) &&
+	    !fwrt->trans->dbg.dump_file_name_ext_valid) {
+		fwrt->trans->dbg.dump_file_name_ext_valid = true;
+		snprintf(fwrt->trans->dbg.dump_file_name_ext, IWL_FW_INI_MAX_NAME,
+			 "0x%x", fwrt->dump.tcm_err_id[idx]);
+	}
+
 	IWL_ERR(fwrt, "TCM%d status:\n", idx + 1);
 	IWL_ERR(fwrt, "0x%08X | error ID\n", table.error_id);
 	IWL_ERR(fwrt, "0x%08X | tcm branchlink2\n", table.blink2);
@@ -337,6 +379,16 @@ static void iwl_fwrt_dump_rcm_error_log(struct iwl_fw_runtime *fwrt, int idx)
 
 	iwl_trans_read_mem_bytes(trans, base, &table, sizeof(table));
 
+	if (table.valid)
+		fwrt->dump.rcm_err_id[idx] = table.error_id;
+
+	if (!iwl_fwrt_if_errorid_other_cpu(fwrt->dump.rcm_err_id[idx]) &&
+	    !fwrt->trans->dbg.dump_file_name_ext_valid) {
+		fwrt->trans->dbg.dump_file_name_ext_valid = true;
+		snprintf(fwrt->trans->dbg.dump_file_name_ext, IWL_FW_INI_MAX_NAME,
+			 "0x%x", fwrt->dump.rcm_err_id[idx]);
+	}
+
 	IWL_ERR(fwrt, "RCM%d status:\n", idx + 1);
 	IWL_ERR(fwrt, "0x%08X | error ID\n", table.error_id);
 	IWL_ERR(fwrt, "0x%08X | rcm branchlink2\n", table.blink2);
@@ -444,8 +496,10 @@ void iwl_fwrt_dump_error_logs(struct iwl_fw_runtime *fwrt)
 	iwl_fwrt_dump_umac_error_log(fwrt);
 	iwl_fwrt_dump_tcm_error_log(fwrt, 0);
 	iwl_fwrt_dump_rcm_error_log(fwrt, 0);
-	iwl_fwrt_dump_tcm_error_log(fwrt, 1);
-	iwl_fwrt_dump_rcm_error_log(fwrt, 1);
+	if (fwrt->trans->dbg.tcm_error_event_table[1])
+		iwl_fwrt_dump_tcm_error_log(fwrt, 1);
+	if (fwrt->trans->dbg.rcm_error_event_table[1])
+		iwl_fwrt_dump_rcm_error_log(fwrt, 1);
 	iwl_fwrt_dump_iml_error_log(fwrt);
 	iwl_fwrt_dump_fseq_regs(fwrt);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index d3cb1ae68a96..a59cf4d9567c 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -24,6 +24,8 @@ struct iwl_fw_runtime_ops {
 };
 
 #define MAX_NUM_LMAC 2
+#define MAX_NUM_TCM 2
+#define MAX_NUM_RCM 2
 struct iwl_fwrt_shared_mem_cfg {
 	int num_lmacs;
 	int num_txfifo_entries;
@@ -129,6 +131,8 @@ struct iwl_fw_runtime {
 		unsigned long non_collect_ts_start[IWL_FW_INI_TIME_POINT_NUM];
 		u32 *d3_debug_data;
 		u32 lmac_err_id[MAX_NUM_LMAC];
+		u32 tcm_err_id[MAX_NUM_TCM];
+		u32 rcm_err_id[MAX_NUM_RCM];
 		u32 umac_err_id;
 
 		struct iwl_txf_iter_data txf_iter_data;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 45981e22b2db..3e9e9f13506b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -122,8 +122,6 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait,
 	u32 version = iwl_fw_lookup_notif_ver(mvm->fw, LEGACY_GROUP,
 					      UCODE_ALIVE_NTFY, 0);
 	u32 i;
-	struct iwl_trans *trans = mvm->trans;
-	enum iwl_device_family device_family = trans->trans_cfg->device_family;
 
 
 	if (version == 6) {
@@ -233,8 +231,7 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait,
 
 	if (umac_error_table) {
 		if (umac_error_table >=
-		    mvm->trans->cfg->min_umac_error_event_table ||
-		    device_family >= IWL_DEVICE_FAMILY_BZ) {
+		    mvm->trans->cfg->min_umac_error_event_table) {
 			iwl_fw_umac_set_alive_err_table(mvm->trans,
 							umac_error_table);
 		} else {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 114c96ba39ee..422550e31bc6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -430,47 +430,55 @@ static void iwl_mvm_mac_ctxt_set_ht_flags(struct iwl_mvm *mvm,
 	}
 }
 
-static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
-					struct ieee80211_vif *vif,
-					struct iwl_mac_ctx_cmd *cmd,
-					const u8 *bssid_override,
-					u32 action)
+static int iwl_mvm_get_mac_type(struct ieee80211_vif *vif)
 {
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct ieee80211_chanctx_conf *chanctx;
-	bool ht_enabled = !!(vif->bss_conf.ht_operation_mode &
-			     IEEE80211_HT_OP_MODE_PROTECTION);
-	u8 cck_ack_rates, ofdm_ack_rates;
-	const u8 *bssid = bssid_override ?: vif->bss_conf.bssid;
-	int i;
-
-	cmd->id_and_color = cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->id,
-							    mvmvif->color));
-	cmd->action = cpu_to_le32(action);
+	u32 mac_type = FW_MAC_TYPE_BSS_STA;
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
 		if (vif->p2p)
-			cmd->mac_type = cpu_to_le32(FW_MAC_TYPE_P2P_STA);
+			mac_type = FW_MAC_TYPE_P2P_STA;
 		else
-			cmd->mac_type = cpu_to_le32(FW_MAC_TYPE_BSS_STA);
+			mac_type = FW_MAC_TYPE_BSS_STA;
 		break;
 	case NL80211_IFTYPE_AP:
-		cmd->mac_type = cpu_to_le32(FW_MAC_TYPE_GO);
+		mac_type = FW_MAC_TYPE_GO;
 		break;
 	case NL80211_IFTYPE_MONITOR:
-		cmd->mac_type = cpu_to_le32(FW_MAC_TYPE_LISTENER);
+		mac_type = FW_MAC_TYPE_LISTENER;
 		break;
 	case NL80211_IFTYPE_P2P_DEVICE:
-		cmd->mac_type = cpu_to_le32(FW_MAC_TYPE_P2P_DEVICE);
+		mac_type = FW_MAC_TYPE_P2P_DEVICE;
 		break;
 	case NL80211_IFTYPE_ADHOC:
-		cmd->mac_type = cpu_to_le32(FW_MAC_TYPE_IBSS);
+		mac_type = FW_MAC_TYPE_IBSS;
 		break;
 	default:
 		WARN_ON_ONCE(1);
 	}
 
+	return mac_type;
+}
+
+static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					struct iwl_mac_ctx_cmd *cmd,
+					const u8 *bssid_override,
+					u32 action)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct ieee80211_chanctx_conf *chanctx;
+	bool ht_enabled = !!(vif->bss_conf.ht_operation_mode &
+			     IEEE80211_HT_OP_MODE_PROTECTION);
+	u8 cck_ack_rates, ofdm_ack_rates;
+	const u8 *bssid = bssid_override ?: vif->bss_conf.bssid;
+	int i;
+
+	cmd->id_and_color = cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->id,
+							    mvmvif->color));
+	cmd->action = cpu_to_le32(action);
+	cmd->mac_type = cpu_to_le32(iwl_mvm_get_mac_type(vif));
+
 	cmd->tsf_id = cpu_to_le32(mvmvif->tsf_id);
 
 	memcpy(cmd->node_addr, vif->addr, ETH_ALEN);
@@ -1428,6 +1436,7 @@ void iwl_mvm_rx_missed_beacons_notif(struct iwl_mvm *mvm,
 	struct ieee80211_vif *vif;
 	u32 id = le32_to_cpu(mb->mac_id);
 	union iwl_dbg_tlv_tp_data tp_data = { .fw_pkt = pkt };
+	u32 mac_type;
 
 	IWL_DEBUG_INFO(mvm,
 		       "missed bcn mac_id=%u, consecutive=%u (%u, %u, %u)\n",
@@ -1443,6 +1452,14 @@ void iwl_mvm_rx_missed_beacons_notif(struct iwl_mvm *mvm,
 	if (!vif)
 		goto out;
 
+	mac_type = iwl_mvm_get_mac_type(vif);
+
+	IWL_DEBUG_INFO(mvm, "missed beacon mac_type=%u,\n", mac_type);
+
+	mvm->trans->dbg.dump_file_name_ext_valid = true;
+	snprintf(mvm->trans->dbg.dump_file_name_ext, IWL_FW_INI_MAX_NAME,
+		 "MacId_%d_MacType_%d", id, mac_type);
+
 	rx_missed_bcon = le32_to_cpu(mb->consec_missed_beacons);
 	rx_missed_bcon_since_rx =
 		le32_to_cpu(mb->consec_missed_beacons_since_last_rx);
-- 
cgit v1.2.3-70-g09d2


From 9c4f15cadcd74becbb7e05f7fccd27180e433b8c Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:16 +0200
Subject: wifi: iwlwifi: mvm: Refactor STA_HE_CTXT_CMD sending flow

Refactor STA_HE_CTXT_CMD sending flow:

1. As the new MLD API is introduced, there are some common fields in
both the old and new APIs. The STA_HE_CTXT_CMD of the non-MLD API
has common fields with the link and mac commands of the new MLD API.
Put this common parts in functions so it can be used later by the
new MLD API.

2. The HE capability which indicates whether the NIC is ack-enabled or
not is the same for all bands. No need to take it from the specific
band which is currently in use. Take it from the low band - this
simplifies the code and doesn't require a phy_ctxt.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.7ca960596953.Ifc3e816461abbd69c6fd87752342afcedfebc293@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 94 ++++++++++++++---------
 1 file changed, 57 insertions(+), 37 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index ab02c6076276..b1f638af4b2c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1996,6 +1996,59 @@ static void iwl_mvm_get_optimal_ppe_info(struct iwl_he_pkt_ext_v2 *pkt_ext,
 	}
 }
 
+/*
+ * This function sets the MU EDCA parameters ans returns whether MU EDCA
+ * is enabled or not
+ */
+static bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
+					  struct iwl_mvm_vif *mvmvif,
+					  struct iwl_he_backoff_conf
+					  *trig_based_txf)
+{
+	int i;
+	/* Mark MU EDCA as enabled, unless none detected on some AC */
+	bool mu_edca_enabled = true;
+
+	for (i = 0; i < IEEE80211_NUM_ACS; i++) {
+		struct ieee80211_he_mu_edca_param_ac_rec *mu_edca =
+			&mvmvif->queue_params[i].mu_edca_param_rec;
+		u8 ac = iwl_mvm_mac80211_ac_to_ucode_ac(i);
+
+		if (!mvmvif->queue_params[i].mu_edca) {
+			mu_edca_enabled = false;
+			break;
+		}
+
+		trig_based_txf[ac].cwmin =
+			cpu_to_le16(mu_edca->ecw_min_max & 0xf);
+		trig_based_txf[ac].cwmax =
+			cpu_to_le16((mu_edca->ecw_min_max & 0xf0) >> 4);
+		trig_based_txf[ac].aifsn =
+			cpu_to_le16(mu_edca->aifsn & 0xf);
+		trig_based_txf[ac].mu_time =
+			cpu_to_le16(mu_edca->mu_edca_timer);
+	}
+
+	return mu_edca_enabled;
+}
+
+static bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif)
+{
+	const struct ieee80211_supported_band *sband;
+	const struct ieee80211_sta_he_cap *own_he_cap = NULL;
+
+	/* This capability is the same for all bands,
+	 * so take it from one of them.
+	 */
+	sband = mvm->hw->wiphy->bands[NL80211_BAND_2GHZ];
+	own_he_cap = ieee80211_get_he_iftype_cap(sband,
+						 ieee80211_vif_type_p2p(vif));
+
+	return (own_he_cap && (own_he_cap->he_cap_elem.mac_cap_info[2] &
+			       IEEE80211_HE_MAC_CAP2_ACK_EN));
+}
+
 static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 			       struct ieee80211_vif *vif, u8 sta_id)
 {
@@ -2015,9 +2068,6 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 	struct ieee80211_sta *sta;
 	u32 flags;
 	int i;
-	const struct ieee80211_sta_he_cap *own_he_cap = NULL;
-	struct ieee80211_chanctx_conf *chanctx_conf;
-	const struct ieee80211_supported_band *sband;
 	void *cmd;
 	u8 nominal_padding;
 
@@ -2045,16 +2095,6 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 
 	rcu_read_lock();
 
-	chanctx_conf = rcu_dereference(vif->bss_conf.chanctx_conf);
-	if (WARN_ON(!chanctx_conf)) {
-		rcu_read_unlock();
-		return;
-	}
-
-	sband = mvm->hw->wiphy->bands[chanctx_conf->def.chan->band];
-	own_he_cap = ieee80211_get_he_iftype_cap(sband,
-						 ieee80211_vif_type_p2p(vif));
-
 	sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_ctxt_cmd.sta_id]);
 	if (IS_ERR_OR_NULL(sta)) {
 		rcu_read_unlock();
@@ -2211,28 +2251,9 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 
 	rcu_read_unlock();
 
-	/* Mark MU EDCA as enabled, unless none detected on some AC */
-	flags |= STA_CTXT_HE_MU_EDCA_CW;
-	for (i = 0; i < IEEE80211_NUM_ACS; i++) {
-		struct ieee80211_he_mu_edca_param_ac_rec *mu_edca =
-			&mvmvif->queue_params[i].mu_edca_param_rec;
-		u8 ac = iwl_mvm_mac80211_ac_to_ucode_ac(i);
-
-		if (!mvmvif->queue_params[i].mu_edca) {
-			flags &= ~STA_CTXT_HE_MU_EDCA_CW;
-			break;
-		}
-
-		sta_ctxt_cmd.trig_based_txf[ac].cwmin =
-			cpu_to_le16(mu_edca->ecw_min_max & 0xf);
-		sta_ctxt_cmd.trig_based_txf[ac].cwmax =
-			cpu_to_le16((mu_edca->ecw_min_max & 0xf0) >> 4);
-		sta_ctxt_cmd.trig_based_txf[ac].aifsn =
-			cpu_to_le16(mu_edca->aifsn);
-		sta_ctxt_cmd.trig_based_txf[ac].mu_time =
-			cpu_to_le16(mu_edca->mu_edca_timer);
-	}
-
+	if (iwl_mvm_set_fw_mu_edca_params(mvm, mvmvif,
+					  &sta_ctxt_cmd.trig_based_txf[0]))
+		flags |= STA_CTXT_HE_MU_EDCA_CW;
 
 	if (vif->bss_conf.uora_exists) {
 		flags |= STA_CTXT_HE_TRIG_RND_ALLOC;
@@ -2243,8 +2264,7 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 			(vif->bss_conf.uora_ocw_range >> 3) & 0x7;
 	}
 
-	if (own_he_cap && !(own_he_cap->he_cap_elem.mac_cap_info[2] &
-			    IEEE80211_HE_MAC_CAP2_ACK_EN))
+	if (!iwl_mvm_is_nic_ack_enabled(mvm, vif))
 		flags |= STA_CTXT_HE_NIC_NOT_ACK_ENABLED;
 
 	if (vif->bss_conf.nontransmitted) {
-- 
cgit v1.2.3-70-g09d2


From af6d168f0ec613d55cdd9c4e0443108172577318 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:17 +0200
Subject: wifi: iwlwifi: mvm: Refactor MAC_CONTEXT_CMD sending flow

Refactor MAC_CONTEXT_CMD sending flow:

1. As the new MLD API is introduced, there are some common fields in
both the old and new APIs. The MAC_CONTEXT_CMD of the non-MLD API
has common fields with the link and mac commands of the new MLD API.
Put this common parts in functions so it can be used later by the
new MLD API.

2. Use iwl_mvm_mac_ctxt_send_cmd when removing a mac instead of
implementing the same functionality once again.

3. Change the debug print when sending the command to be more specific,
so it will be easy to distinguish later if the old or new mac command
was sent.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.3ab62700db4e.I2e353b308667c215aa456c160e0d90de2b9b85cc@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 320 +++++++++++++---------
 1 file changed, 197 insertions(+), 123 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 422550e31bc6..a30385b85a60 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -396,15 +396,45 @@ static void iwl_mvm_ack_rates(struct iwl_mvm *mvm,
 	*ofdm_rates = ofdm;
 }
 
-static void iwl_mvm_mac_ctxt_set_ht_flags(struct iwl_mvm *mvm,
-					 struct ieee80211_vif *vif,
-					 struct iwl_mac_ctx_cmd *cmd)
+static void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif,
+				       __le32 *cck_rates, __le32 *ofdm_rates)
+{
+	struct ieee80211_chanctx_conf *chanctx;
+	u8 cck_ack_rates, ofdm_ack_rates;
+
+	rcu_read_lock();
+	chanctx = rcu_dereference(vif->bss_conf.chanctx_conf);
+	iwl_mvm_ack_rates(mvm, vif, chanctx ? chanctx->def.chan->band
+					    : NL80211_BAND_2GHZ,
+			  &cck_ack_rates, &ofdm_ack_rates);
+	rcu_read_unlock();
+
+	*cck_rates = cpu_to_le32((u32)cck_ack_rates);
+	*ofdm_rates = cpu_to_le32((u32)ofdm_ack_rates);
+}
+
+static void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
+					    __le32 *protection_flags,
+					    u32 ht_flag,
+					    u32 tgg_flag)
 {
 	/* for both sta and ap, ht_operation_mode hold the protection_mode */
 	u8 protection_mode = vif->bss_conf.ht_operation_mode &
 				 IEEE80211_HT_OP_MODE_PROTECTION;
-	/* The fw does not distinguish between ht and fat */
-	u32 ht_flag = MAC_PROT_FLG_HT_PROT | MAC_PROT_FLG_FAT_PROT;
+	bool ht_enabled = !!(vif->bss_conf.ht_operation_mode &
+			     IEEE80211_HT_OP_MODE_PROTECTION);
+
+	if (vif->bss_conf.use_cts_prot)
+		*protection_flags |= cpu_to_le32(tgg_flag);
+
+	IWL_DEBUG_RATE(mvm, "use_cts_prot %d, ht_operation_mode %d\n",
+		       vif->bss_conf.use_cts_prot,
+		       vif->bss_conf.ht_operation_mode);
+
+	if (!ht_enabled)
+		return;
 
 	IWL_DEBUG_RATE(mvm, "protection mode set to %d\n", protection_mode);
 	/*
@@ -416,12 +446,12 @@ static void iwl_mvm_mac_ctxt_set_ht_flags(struct iwl_mvm *mvm,
 		break;
 	case IEEE80211_HT_OP_MODE_PROTECTION_NONMEMBER:
 	case IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED:
-		cmd->protection_flags |= cpu_to_le32(ht_flag);
+		*protection_flags |= cpu_to_le32(ht_flag);
 		break;
 	case IEEE80211_HT_OP_MODE_PROTECTION_20MHZ:
 		/* Protect when channel wider than 20MHz */
 		if (vif->bss_conf.chandef.width > NL80211_CHAN_WIDTH_20)
-			cmd->protection_flags |= cpu_to_le32(ht_flag);
+			*protection_flags |= cpu_to_le32(ht_flag);
 		break;
 	default:
 		IWL_ERR(mvm, "Illegal protection mode %d\n",
@@ -430,6 +460,34 @@ static void iwl_mvm_mac_ctxt_set_ht_flags(struct iwl_mvm *mvm,
 	}
 }
 
+static void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct iwl_ac_qos *ac, __le32 *qos_flags)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int i;
+
+	for (i = 0; i < IEEE80211_NUM_ACS; i++) {
+		u8 txf = iwl_mvm_mac_ac_to_tx_fifo(mvm, i);
+		u8 ucode_ac = iwl_mvm_mac80211_ac_to_ucode_ac(i);
+
+		ac[ucode_ac].cw_min =
+			cpu_to_le16(mvmvif->queue_params[i].cw_min);
+		ac[ucode_ac].cw_max =
+			cpu_to_le16(mvmvif->queue_params[i].cw_max);
+		ac[ucode_ac].edca_txop =
+			cpu_to_le16(mvmvif->queue_params[i].txop * 32);
+		ac[ucode_ac].aifsn = mvmvif->queue_params[i].aifs;
+		ac[ucode_ac].fifos_mask = BIT(txf);
+	}
+
+	if (vif->bss_conf.qos)
+		*qos_flags |= cpu_to_le32(MAC_QOS_FLG_UPDATE_EDCA);
+
+	if (vif->bss_conf.chandef.width != NL80211_CHAN_WIDTH_20_NOHT)
+		*qos_flags |= cpu_to_le32(MAC_QOS_FLG_TGN);
+}
+
 static int iwl_mvm_get_mac_type(struct ieee80211_vif *vif)
 {
 	u32 mac_type = FW_MAC_TYPE_BSS_STA;
@@ -456,7 +514,6 @@ static int iwl_mvm_get_mac_type(struct ieee80211_vif *vif)
 	default:
 		WARN_ON_ONCE(1);
 	}
-
 	return mac_type;
 }
 
@@ -467,12 +524,8 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 					u32 action)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct ieee80211_chanctx_conf *chanctx;
-	bool ht_enabled = !!(vif->bss_conf.ht_operation_mode &
-			     IEEE80211_HT_OP_MODE_PROTECTION);
-	u8 cck_ack_rates, ofdm_ack_rates;
 	const u8 *bssid = bssid_override ?: vif->bss_conf.bssid;
-	int i;
+	u32 ht_flag;
 
 	cmd->id_and_color = cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->id,
 							    mvmvif->color));
@@ -488,15 +541,8 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 	else
 		eth_broadcast_addr(cmd->bssid_addr);
 
-	rcu_read_lock();
-	chanctx = rcu_dereference(vif->bss_conf.chanctx_conf);
-	iwl_mvm_ack_rates(mvm, vif, chanctx ? chanctx->def.chan->band
-					    : NL80211_BAND_2GHZ,
-			  &cck_ack_rates, &ofdm_ack_rates);
-	rcu_read_unlock();
-
-	cmd->cck_rates = cpu_to_le32((u32)cck_ack_rates);
-	cmd->ofdm_rates = cpu_to_le32((u32)ofdm_ack_rates);
+	iwl_mvm_set_fw_basic_rates(mvm, vif, &cmd->cck_rates,
+				   &cmd->ofdm_rates);
 
 	cmd->cck_short_preamble =
 		cpu_to_le32(vif->bss_conf.use_short_preamble ?
@@ -507,33 +553,12 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 
 	cmd->filter_flags = 0;
 
-	for (i = 0; i < IEEE80211_NUM_ACS; i++) {
-		u8 txf = iwl_mvm_mac_ac_to_tx_fifo(mvm, i);
-		u8 ucode_ac = iwl_mvm_mac80211_ac_to_ucode_ac(i);
-
-		cmd->ac[ucode_ac].cw_min =
-			cpu_to_le16(mvmvif->queue_params[i].cw_min);
-		cmd->ac[ucode_ac].cw_max =
-			cpu_to_le16(mvmvif->queue_params[i].cw_max);
-		cmd->ac[ucode_ac].edca_txop =
-			cpu_to_le16(mvmvif->queue_params[i].txop * 32);
-		cmd->ac[ucode_ac].aifsn = mvmvif->queue_params[i].aifs;
-		cmd->ac[ucode_ac].fifos_mask = BIT(txf);
-	}
-
-	if (vif->bss_conf.qos)
-		cmd->qos_flags |= cpu_to_le32(MAC_QOS_FLG_UPDATE_EDCA);
+	iwl_mvm_set_fw_qos_params(mvm, vif, &cmd->ac[0], &cmd->qos_flags);
 
-	if (vif->bss_conf.use_cts_prot)
-		cmd->protection_flags |= cpu_to_le32(MAC_PROT_FLG_TGG_PROTECT);
-
-	IWL_DEBUG_RATE(mvm, "use_cts_prot %d, ht_operation_mode %d\n",
-		       vif->bss_conf.use_cts_prot,
-		       vif->bss_conf.ht_operation_mode);
-	if (vif->bss_conf.chandef.width != NL80211_CHAN_WIDTH_20_NOHT)
-		cmd->qos_flags |= cpu_to_le32(MAC_QOS_FLG_TGN);
-	if (ht_enabled)
-		iwl_mvm_mac_ctxt_set_ht_flags(mvm, vif, cmd);
+	/* The fw does not distinguish between ht and fat */
+	ht_flag = MAC_PROT_FLG_HT_PROT | MAC_PROT_FLG_FAT_PROT;
+	iwl_mvm_set_fw_protection_flags(mvm, vif, &cmd->protection_flags,
+					ht_flag, MAC_PROT_FLG_TGG_PROTECT);
 }
 
 static int iwl_mvm_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
@@ -542,11 +567,76 @@ static int iwl_mvm_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
 	int ret = iwl_mvm_send_cmd_pdu(mvm, MAC_CONTEXT_CMD, 0,
 				       sizeof(*cmd), cmd);
 	if (ret)
-		IWL_ERR(mvm, "Failed to send MAC context (action:%d): %d\n",
+		IWL_ERR(mvm, "Failed to send MAC_CONTEXT_CMD (action:%d): %d\n",
 			le32_to_cpu(cmd->action), ret);
 	return ret;
 }
 
+static void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm,
+				     struct ieee80211_vif *vif,
+				     __le64 *dtim_tsf, __le32 *dtim_time,
+				     __le32 *assoc_beacon_arrive_time)
+{
+	u32 dtim_offs;
+
+	/*
+	 * The DTIM count counts down, so when it is N that means N
+	 * more beacon intervals happen until the DTIM TBTT. Therefore
+	 * add this to the current time. If that ends up being in the
+	 * future, the firmware will handle it.
+	 *
+	 * Also note that the system_timestamp (which we get here as
+	 * "sync_device_ts") and TSF timestamp aren't at exactly the
+	 * same offset in the frame -- the TSF is at the first symbol
+	 * of the TSF, the system timestamp is at signal acquisition
+	 * time. This means there's an offset between them of at most
+	 * a few hundred microseconds (24 * 8 bits + PLCP time gives
+	 * 384us in the longest case), this is currently not relevant
+	 * as the firmware wakes up around 2ms before the TBTT.
+	 */
+	dtim_offs = vif->bss_conf.sync_dtim_count *
+			vif->bss_conf.beacon_int;
+	/* convert TU to usecs */
+	dtim_offs *= 1024;
+
+	*dtim_tsf =
+		cpu_to_le64(vif->bss_conf.sync_tsf + dtim_offs);
+	*dtim_time =
+		cpu_to_le32(vif->bss_conf.sync_device_ts + dtim_offs);
+	*assoc_beacon_arrive_time =
+		cpu_to_le32(vif->bss_conf.sync_device_ts);
+
+	IWL_DEBUG_INFO(mvm, "DTIM TBTT is 0x%llx/0x%x, offset %d\n",
+		       le64_to_cpu(*dtim_tsf),
+		       le32_to_cpu(*dtim_time),
+		       dtim_offs);
+}
+
+static __le32 iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(struct iwl_mvm *mvm,
+							   struct ieee80211_vif *vif)
+{
+	struct ieee80211_p2p_noa_attr *noa =
+		&vif->bss_conf.p2p_noa_attr;
+
+	return cpu_to_le32(noa->oppps_ctwindow &
+			IEEE80211_P2P_OPPPS_CTWINDOW_MASK);
+}
+
+static __le32 iwl_mvm_mac_ctxt_cmd_sta_get_twt_policy(struct iwl_mvm *mvm,
+						      struct ieee80211_vif *vif)
+{
+	__le32 twt_policy = cpu_to_le32(0);
+
+	if (vif->bss_conf.twt_requester && IWL_MVM_USE_TWT)
+		twt_policy |= cpu_to_le32(TWT_SUPPORTED);
+	if (vif->bss_conf.twt_protected)
+		twt_policy |= cpu_to_le32(PROTECTED_TWT_SUPPORTED);
+	if (vif->bss_conf.twt_broadcast)
+		twt_policy |= cpu_to_le32(BROADCAST_TWT_SUPPORTED);
+
+	return twt_policy;
+}
+
 static int iwl_mvm_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif,
 				    u32 action, bool force_assoc_off,
@@ -567,11 +657,9 @@ static int iwl_mvm_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 	cmd.filter_flags |= cpu_to_le32(MAC_FILTER_ACCEPT_GRP);
 
 	if (vif->p2p) {
-		struct ieee80211_p2p_noa_attr *noa =
-			&vif->bss_conf.p2p_noa_attr;
+		cmd.p2p_sta.ctwin =
+			iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(mvm, vif);
 
-		cmd.p2p_sta.ctwin = cpu_to_le32(noa->oppps_ctwindow &
-					IEEE80211_P2P_OPPPS_CTWINDOW_MASK);
 		ctxt_sta = &cmd.p2p_sta.sta;
 	} else {
 		ctxt_sta = &cmd.sta;
@@ -581,39 +669,10 @@ static int iwl_mvm_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 	if (vif->cfg.assoc && vif->bss_conf.dtim_period &&
 	    !force_assoc_off) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-		u32 dtim_offs;
 
-		/*
-		 * The DTIM count counts down, so when it is N that means N
-		 * more beacon intervals happen until the DTIM TBTT. Therefore
-		 * add this to the current time. If that ends up being in the
-		 * future, the firmware will handle it.
-		 *
-		 * Also note that the system_timestamp (which we get here as
-		 * "sync_device_ts") and TSF timestamp aren't at exactly the
-		 * same offset in the frame -- the TSF is at the first symbol
-		 * of the TSF, the system timestamp is at signal acquisition
-		 * time. This means there's an offset between them of at most
-		 * a few hundred microseconds (24 * 8 bits + PLCP time gives
-		 * 384us in the longest case), this is currently not relevant
-		 * as the firmware wakes up around 2ms before the TBTT.
-		 */
-		dtim_offs = vif->bss_conf.sync_dtim_count *
-				vif->bss_conf.beacon_int;
-		/* convert TU to usecs */
-		dtim_offs *= 1024;
-
-		ctxt_sta->dtim_tsf =
-			cpu_to_le64(vif->bss_conf.sync_tsf + dtim_offs);
-		ctxt_sta->dtim_time =
-			cpu_to_le32(vif->bss_conf.sync_device_ts + dtim_offs);
-		ctxt_sta->assoc_beacon_arrive_time =
-			cpu_to_le32(vif->bss_conf.sync_device_ts);
-
-		IWL_DEBUG_INFO(mvm, "DTIM TBTT is 0x%llx/0x%x, offset %d\n",
-			       le64_to_cpu(ctxt_sta->dtim_tsf),
-			       le32_to_cpu(ctxt_sta->dtim_time),
-			       dtim_offs);
+		iwl_mvm_set_fw_dtim_tbtt(mvm, vif, &ctxt_sta->dtim_tsf,
+					 &ctxt_sta->dtim_time,
+					 &ctxt_sta->assoc_beacon_arrive_time);
 
 		ctxt_sta->is_assoc = cpu_to_le32(1);
 
@@ -643,14 +702,8 @@ static int iwl_mvm_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 
 	if (vif->bss_conf.he_support && !iwlwifi_mod_params.disable_11ax) {
 		cmd.filter_flags |= cpu_to_le32(MAC_FILTER_IN_11AX);
-		if (vif->bss_conf.twt_requester && IWL_MVM_USE_TWT)
-			ctxt_sta->data_policy |= cpu_to_le32(TWT_SUPPORTED);
-		if (vif->bss_conf.twt_protected)
-			ctxt_sta->data_policy |=
-				cpu_to_le32(PROTECTED_TWT_SUPPORTED);
-		if (vif->bss_conf.twt_broadcast)
-			ctxt_sta->data_policy |=
-				cpu_to_le32(BROADCAST_TWT_SUPPORTED);
+		ctxt_sta->data_policy |=
+			iwl_mvm_mac_ctxt_cmd_sta_get_twt_policy(mvm, vif);
 	}
 
 
@@ -732,20 +785,11 @@ static void iwl_mvm_go_iterator(void *_data, u8 *mac, struct ieee80211_vif *vif)
 		data->go_active = true;
 }
 
-static int iwl_mvm_mac_ctxt_cmd_p2p_device(struct iwl_mvm *mvm,
-					   struct ieee80211_vif *vif,
-					   u32 action)
+static __le32 iwl_mac_ctxt_p2p_dev_has_extended_disc(struct iwl_mvm *mvm,
+						     struct ieee80211_vif *vif)
 {
-	struct iwl_mac_ctx_cmd cmd = {};
 	struct iwl_mvm_go_iterator_data data = {};
 
-	WARN_ON(vif->type != NL80211_IFTYPE_P2P_DEVICE);
-
-	iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action);
-
-	/* Override the filter flags to accept only probe requests */
-	cmd.filter_flags = cpu_to_le32(MAC_FILTER_IN_PROBE_REQUEST);
-
 	/*
 	 * This flag should be set to true when the P2P Device is
 	 * discoverable and there is at least another active P2P GO. Settings
@@ -758,7 +802,25 @@ static int iwl_mvm_mac_ctxt_cmd_p2p_device(struct iwl_mvm *mvm,
 		mvm->hw, IEEE80211_IFACE_ITER_RESUME_ALL,
 		iwl_mvm_go_iterator, &data);
 
-	cmd.p2p_dev.is_disc_extended = cpu_to_le32(data.go_active ? 1 : 0);
+	return cpu_to_le32(data.go_active ? 1 : 0);
+}
+
+static int iwl_mvm_mac_ctxt_cmd_p2p_device(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   u32 action)
+{
+	struct iwl_mac_ctx_cmd cmd = {};
+
+	WARN_ON(vif->type != NL80211_IFTYPE_P2P_DEVICE);
+
+	iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action);
+
+	cmd.p2p_dev.is_disc_extended =
+		iwl_mac_ctxt_p2p_dev_has_extended_disc(mvm, vif);
+
+	/* Override the filter flags to accept only probe requests */
+	cmd.filter_flags = cpu_to_le32(MAC_FILTER_IN_PROBE_REQUEST);
+
 	return iwl_mvm_mac_ctxt_send_cmd(mvm, &cmd);
 }
 
@@ -1102,6 +1164,30 @@ static void iwl_mvm_mac_ap_iterator(void *_data, u8 *mac,
 	data->beacon_int = vif->bss_conf.beacon_int;
 }
 
+/*
+ * Fill the filter flags for mac context of type AP or P2P GO.
+ */
+static void iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(struct iwl_mvm *mvm,
+						     struct iwl_mvm_vif *mvmvif,
+						     __le32 *filter_flags,
+						     int accept_probe_req_flag,
+						     int accept_beacon_flag)
+{
+	/*
+	 * in AP mode, pass probe requests and beacons from other APs
+	 * (needed for ht protection); when there're no any associated
+	 * station don't ask FW to pass beacons to prevent unnecessary
+	 * wake-ups.
+	 */
+	*filter_flags |= cpu_to_le32(accept_probe_req_flag);
+	if (mvmvif->ap_assoc_sta_count || !mvm->drop_bcn_ap_mode) {
+		*filter_flags |= cpu_to_le32(accept_beacon_flag);
+		IWL_DEBUG_HC(mvm, "Asking FW to pass beacons\n");
+	} else {
+		IWL_DEBUG_HC(mvm, "No need to receive beacons\n");
+	}
+}
+
 /*
  * Fill the specific data for mac context of type AP of P2P GO
  */
@@ -1121,19 +1207,10 @@ static void iwl_mvm_mac_ctxt_cmd_fill_ap(struct iwl_mvm *mvm,
 	/* in AP mode, the MCAST FIFO takes the EDCA params from VO */
 	cmd->ac[IWL_MVM_TX_FIFO_VO].fifos_mask |= BIT(IWL_MVM_TX_FIFO_MCAST);
 
-	/*
-	 * in AP mode, pass probe requests and beacons from other APs
-	 * (needed for ht protection); when there're no any associated
-	 * station don't ask FW to pass beacons to prevent unnecessary
-	 * wake-ups.
-	 */
-	cmd->filter_flags |= cpu_to_le32(MAC_FILTER_IN_PROBE_REQUEST);
-	if (mvmvif->ap_assoc_sta_count || !mvm->drop_bcn_ap_mode) {
-		cmd->filter_flags |= cpu_to_le32(MAC_FILTER_IN_BEACON);
-		IWL_DEBUG_HC(mvm, "Asking FW to pass beacons\n");
-	} else {
-		IWL_DEBUG_HC(mvm, "No need to receive beacons\n");
-	}
+	iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(mvm, mvmvif,
+						 &cmd->filter_flags,
+						 MAC_FILTER_IN_PROBE_REQUEST,
+						 MAC_FILTER_IN_BEACON);
 
 	ctxt_ap->bi = cpu_to_le32(vif->bss_conf.beacon_int);
 	ctxt_ap->dtim_interval = cpu_to_le32(vif->bss_conf.beacon_int *
@@ -1295,12 +1372,9 @@ int iwl_mvm_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 							   mvmvif->color));
 	cmd.action = cpu_to_le32(FW_CTXT_ACTION_REMOVE);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, MAC_CONTEXT_CMD, 0,
-				   sizeof(cmd), &cmd);
-	if (ret) {
-		IWL_ERR(mvm, "Failed to remove MAC context: %d\n", ret);
+	ret = iwl_mvm_mac_ctxt_send_cmd(mvm, &cmd);
+	if (ret)
 		return ret;
-	}
 
 	mvmvif->uploaded = false;
 
-- 
cgit v1.2.3-70-g09d2


From 9be162a7b670b7e8dbada3c139d77961f457a34d Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:18 +0200
Subject: wifi: iwlwifi: mvm: add support for the new MAC CTXT command

As a part of the new MLD FW API changes, we have a new MAC CTXT command.
Add structures and enum definitions, along with the functions that
sends this command (i.e. add, remove and change mac ctxt).
This functions will be in used in the next patches.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.99a41a5bc55f.I310312c829f5f2f69b64fcddce487b1eab80165b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/mac-cfg.h    | 109 ++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/Makefile    |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  |  24 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |   3 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c   | 271 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  18 ++
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |   1 +
 7 files changed, 411 insertions(+), 17 deletions(-)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index 712532f17630..f12b2a25cd3d 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -1,12 +1,14 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2019, 2021 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2019, 2021-2022 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
 #ifndef __iwl_fw_api_mac_cfg_h__
 #define __iwl_fw_api_mac_cfg_h__
 
+#include "mac.h"
+
 /**
  * enum iwl_mac_conf_subcmd_ids - mac configuration command IDs
  */
@@ -31,7 +33,10 @@ enum iwl_mac_conf_subcmd_ids {
 	 * @CANCEL_CHANNEL_SWITCH_CMD: &struct iwl_cancel_channel_switch_cmd
 	 */
 	CANCEL_CHANNEL_SWITCH_CMD = 0x6,
-
+	/**
+	 * @MAC_CONFIG_CMD: &struct iwl_mac_config_cmd
+	 */
+	MAC_CONFIG_CMD = 0x8,
 	/**
 	 * @SESSION_PROTECTION_NOTIF: &struct iwl_mvm_session_prot_notif
 	 */
@@ -182,4 +187,104 @@ struct iwl_mac_low_latency_cmd {
 	__le16 reserved;
 } __packed; /* MAC_LOW_LATENCY_API_S_VER_1 */
 
+/**
+ * struct iwl_mac_client_data - configuration data for client MAC context
+ *
+ * @is_assoc: 1 for associated state, 0 otherwise
+ * @assoc_id: unique ID assigned by the AP during association
+ * @data_policy: see &enum iwl_mac_data_policy
+ * @ctwin: client traffic window in TU (period after TBTT when GO is present).
+ *	0 indicates that there is no CT window.
+ */
+struct iwl_mac_client_data {
+	__le32 is_assoc;
+	__le32 assoc_id;
+	__le32 data_policy;
+	__le32 ctwin;
+} __packed; /* MAC_CONTEXT_CONFIG_CLIENT_DATA_API_S_VER_1 */
+
+/**
+ * struct iwl_mac_go_ibss_data - configuration data for GO and IBSS MAC context
+ *
+ * @beacon_template: beacon template ID
+ */
+struct iwl_mac_go_ibss_data {
+	__le32 beacon_template;
+} __packed; /* MAC_CONTEXT_CONFIG_GO_IBSS_DATA_API_S_VER_1 */
+
+/**
+ * struct iwl_mac_p2p_dev_data  - configuration data for P2P device MAC context
+ *
+ * @is_disc_extended: if set to true, P2P Device discoverability is enabled on
+ *	other channels as well. This should be to true only in case that the
+ *	device is discoverable and there is an active GO. Note that setting this
+ *	field when not needed, will increase the number of interrupts and have
+ *	effect on the platform power, as this setting opens the Rx filters on
+ *	all macs.
+ */
+struct iwl_mac_p2p_dev_data {
+	__le32 is_disc_extended;
+} __packed; /* MAC_CONTEXT_CONFIG_P2P_DEV_DATA_API_S_VER_1 */
+
+/**
+ * enum iwl_mac_config_filter_flags - MAC context configuration filter flags
+ *
+ * @MAC_CFG_FILTER_PROMISC: accept all data frames
+ * @MAC_CFG_FILTER_ACCEPT_CONTROL_AND_MGMT: pass all management and
+ *	control frames to the host
+ * @MAC_CFG_FILTER_ACCEPT_GRP: accept multicast frames
+ * @MAC_CFG_FILTER_ACCEPT_BEACON: accept beacon frames
+ * @MAC_CFG_FILTER_ACCEPT_BCAST_PROBE_RESP: accept broadcast probe response
+ * @MAC_CFG_FILTER_ACCEPT_PROBE_REQ: accept probe requests
+ */
+enum iwl_mac_config_filter_flags {
+	MAC_CFG_FILTER_PROMISC			= BIT(0),
+	MAC_CFG_FILTER_ACCEPT_CONTROL_AND_MGMT	= BIT(1),
+	MAC_CFG_FILTER_ACCEPT_GRP		= BIT(2),
+	MAC_CFG_FILTER_ACCEPT_BEACON		= BIT(3),
+	MAC_CFG_FILTER_ACCEPT_BCAST_PROBE_RESP	= BIT(4),
+	MAC_CFG_FILTER_ACCEPT_PROBE_REQ		= BIT(5),
+}; /* MAC_FILTER_FLAGS_MASK_E_VER_1 */
+
+/**
+ * struct iwl_mac_config_cmd - command structure to configure MAC contexts in
+ *	MLD API
+ * ( MAC_CONTEXT_CONFIG_CMD = 0x8 )
+ *
+ * @id_and_color: ID and color of the MAC
+ * @action: action to perform, one of FW_CTXT_ACTION_*
+ * @mac_type: one of &enum iwl_mac_types
+ * @local_mld_addr: mld address
+ * @reserved_for_local_mld_addr: reserved
+ * @filter_flags: combination of &enum iwl_mac_config_filter_flags
+ * @he_support: does this MAC support HE
+ * @eht_support: does this MAC support EHT. Requires he_support
+ * @nic_not_ack_enabled: mark that the NIC doesn't support receiving
+ *	ACK-enabled AGG, (i.e. both BACK and non-BACK frames in single AGG).
+ *	If the NIC is not ACK_ENABLED it may use the EOF-bit in first non-0
+ *	len delim to determine if AGG or single.
+ * @client: client mac data
+ * @go_ibss: mac data for go or ibss
+ * @p2p_dev: mac data for p2p device
+ */
+struct iwl_mac_config_cmd {
+	/* COMMON_INDEX_HDR_API_S_VER_1 */
+	__le32 id_and_color;
+	__le32 action;
+	/* MAC_CONTEXT_TYPE_API_E */
+	__le32 mac_type;
+	u8 local_mld_addr[6];
+	__le16 reserved_for_local_mld_addr;
+	__le32 filter_flags;
+	__le32 he_support;
+	__le32 eht_support;
+	__le32 nic_not_ack_enabled;
+	/* MAC_CONTEXT_CONFIG_SPECIFIC_DATA_API_U_VER_1 */
+	union {
+		struct iwl_mac_client_data client;
+		struct iwl_mac_go_ibss_data go_ibss;
+		struct iwl_mac_p2p_dev_data p2p_dev;
+	};
+} __packed; /* MAC_CONTEXT_CONFIG_CMD_API_S_VER_1 */
+
 #endif /* __iwl_fw_api_mac_cfg_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
index b28fcf0cf9cf..d71f0a99b7c4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
@@ -7,7 +7,7 @@ iwlmvm-y += power.o coex.o
 iwlmvm-y += tt.o offloading.o tdls.o
 iwlmvm-y += ftm-responder.o ftm-initiator.o
 iwlmvm-y += rfi.o
-iwlmvm-y += mld-key.o
+iwlmvm-y += mld-key.o mld-mac.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
 iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM) += d3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index a30385b85a60..b21327ec4401 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -488,7 +488,7 @@ static void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm,
 		*qos_flags |= cpu_to_le32(MAC_QOS_FLG_TGN);
 }
 
-static int iwl_mvm_get_mac_type(struct ieee80211_vif *vif)
+int iwl_mvm_get_mac_type(struct ieee80211_vif *vif)
 {
 	u32 mac_type = FW_MAC_TYPE_BSS_STA;
 
@@ -612,8 +612,8 @@ static void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm,
 		       dtim_offs);
 }
 
-static __le32 iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(struct iwl_mvm *mvm,
-							   struct ieee80211_vif *vif)
+__le32 iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(struct iwl_mvm *mvm,
+						    struct ieee80211_vif *vif)
 {
 	struct ieee80211_p2p_noa_attr *noa =
 		&vif->bss_conf.p2p_noa_attr;
@@ -622,8 +622,8 @@ static __le32 iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(struct iwl_mvm *mvm,
 			IEEE80211_P2P_OPPPS_CTWINDOW_MASK);
 }
 
-static __le32 iwl_mvm_mac_ctxt_cmd_sta_get_twt_policy(struct iwl_mvm *mvm,
-						      struct ieee80211_vif *vif)
+__le32 iwl_mvm_mac_ctxt_cmd_sta_get_twt_policy(struct iwl_mvm *mvm,
+					       struct ieee80211_vif *vif)
 {
 	__le32 twt_policy = cpu_to_le32(0);
 
@@ -785,8 +785,8 @@ static void iwl_mvm_go_iterator(void *_data, u8 *mac, struct ieee80211_vif *vif)
 		data->go_active = true;
 }
 
-static __le32 iwl_mac_ctxt_p2p_dev_has_extended_disc(struct iwl_mvm *mvm,
-						     struct ieee80211_vif *vif)
+__le32 iwl_mac_ctxt_p2p_dev_has_extended_disc(struct iwl_mvm *mvm,
+					      struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_go_iterator_data data = {};
 
@@ -1167,11 +1167,11 @@ static void iwl_mvm_mac_ap_iterator(void *_data, u8 *mac,
 /*
  * Fill the filter flags for mac context of type AP or P2P GO.
  */
-static void iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(struct iwl_mvm *mvm,
-						     struct iwl_mvm_vif *mvmvif,
-						     __le32 *filter_flags,
-						     int accept_probe_req_flag,
-						     int accept_beacon_flag)
+void iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(struct iwl_mvm *mvm,
+					      struct iwl_mvm_vif *mvmvif,
+					      __le32 *filter_flags,
+					      int accept_probe_req_flag,
+					      int accept_beacon_flag)
 {
 	/*
 	 * in AP mode, pass probe requests and beacons from other APs
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index b1f638af4b2c..210b134c735f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2032,8 +2032,7 @@ static bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
 	return mu_edca_enabled;
 }
 
-static bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm,
-				       struct ieee80211_vif *vif)
+bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	const struct ieee80211_supported_band *sband;
 	const struct ieee80211_sta_he_cap *own_he_cap = NULL;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
new file mode 100644
index 000000000000..240f51d2fd53
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
@@ -0,0 +1,271 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2022 Intel Corporation
+ */
+#include "mvm.h"
+
+static void iwl_mvm_mld_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
+					    struct iwl_mac_config_cmd *cmd,
+					    u32 action)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	cmd->id_and_color = cpu_to_le32(mvmvif->id);
+	cmd->action = cpu_to_le32(action);
+
+	cmd->mac_type = cpu_to_le32(iwl_mvm_get_mac_type(vif));
+
+	memcpy(cmd->local_mld_addr, vif->addr, ETH_ALEN);
+
+	cmd->filter_flags = cpu_to_le32(0);
+	cmd->he_support = cpu_to_le32(0);
+	cmd->eht_support = cpu_to_le32(0);
+
+	cmd->nic_not_ack_enabled =
+		cpu_to_le32(!iwl_mvm_is_nic_ack_enabled(mvm, vif));
+
+	if (iwlwifi_mod_params.disable_11ax)
+		return;
+	cmd->he_support = cpu_to_le32(vif->bss_conf.he_support);
+
+	if (!iwlwifi_mod_params.disable_11be && cmd->he_support)
+		cmd->eht_support = cpu_to_le32(vif->bss_conf.eht_support);
+}
+
+static int iwl_mvm_mld_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
+					 struct iwl_mac_config_cmd *cmd)
+{
+	int ret = iwl_mvm_send_cmd_pdu(mvm,
+				       WIDE_ID(MAC_CONF_GROUP, MAC_CONFIG_CMD),
+				       0, sizeof(*cmd), cmd);
+	if (ret)
+		IWL_ERR(mvm, "Failed to send MAC_CONFIG_CMD (action:%d): %d\n",
+			le32_to_cpu(cmd->action), ret);
+	return ret;
+}
+
+static int iwl_mvm_mld_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					u32 action, bool force_assoc_off)
+{
+	struct iwl_mac_config_cmd cmd = {};
+
+	WARN_ON(vif->type != NL80211_IFTYPE_STATION);
+
+	/* Fill the common data for all mac context types */
+	iwl_mvm_mld_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
+
+	/*
+	 * We always want to hear MCAST frames, if we're not authorized yet,
+	 * we'll drop them.
+	 */
+	cmd.filter_flags |= cpu_to_le32(MAC_CFG_FILTER_ACCEPT_GRP);
+
+	if (vif->p2p)
+		cmd.client.ctwin =
+			iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(mvm, vif);
+
+	if (vif->cfg.assoc && vif->bss_conf.dtim_period &&
+	    !force_assoc_off) {
+		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+		cmd.client.is_assoc = cpu_to_le32(1);
+
+		if (!mvmvif->authorized &&
+		    fw_has_capa(&mvm->fw->ucode_capa,
+				IWL_UCODE_TLV_CAPA_COEX_HIGH_PRIO))
+			cmd.client.data_policy |=
+				cpu_to_le32(COEX_HIGH_PRIORITY_ENABLE);
+
+	} else {
+		cmd.client.is_assoc = cpu_to_le32(0);
+
+		/* Allow beacons to pass through as long as we are not
+		 * associated, or we do not have dtim period information.
+		 */
+		cmd.filter_flags |= cpu_to_le32(MAC_CFG_FILTER_ACCEPT_BEACON);
+	}
+
+	cmd.client.assoc_id = cpu_to_le32(vif->cfg.aid);
+
+	if (vif->probe_req_reg && vif->cfg.assoc && vif->p2p)
+		cmd.filter_flags |= cpu_to_le32(MAC_CFG_FILTER_ACCEPT_PROBE_REQ);
+
+	if (vif->bss_conf.he_support && !iwlwifi_mod_params.disable_11ax)
+		cmd.client.data_policy |=
+			iwl_mvm_mac_ctxt_cmd_sta_get_twt_policy(mvm, vif);
+
+	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
+}
+
+static int iwl_mvm_mld_mac_ctxt_cmd_listener(struct iwl_mvm *mvm,
+					     struct ieee80211_vif *vif,
+					     u32 action)
+{
+	struct iwl_mac_config_cmd cmd = {};
+
+	WARN_ON(vif->type != NL80211_IFTYPE_MONITOR);
+
+	iwl_mvm_mld_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
+
+	cmd.filter_flags = cpu_to_le32(MAC_CFG_FILTER_PROMISC |
+				       MAC_FILTER_IN_CONTROL_AND_MGMT |
+				       MAC_CFG_FILTER_ACCEPT_BEACON |
+				       MAC_CFG_FILTER_ACCEPT_PROBE_REQ |
+				       MAC_CFG_FILTER_ACCEPT_GRP);
+
+	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
+}
+
+static int iwl_mvm_mld_mac_ctxt_cmd_ibss(struct iwl_mvm *mvm,
+					 struct ieee80211_vif *vif,
+					 u32 action)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mac_config_cmd cmd = {};
+
+	WARN_ON(vif->type != NL80211_IFTYPE_ADHOC);
+
+	iwl_mvm_mld_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
+
+	cmd.filter_flags = cpu_to_le32(MAC_CFG_FILTER_ACCEPT_BEACON |
+				       MAC_CFG_FILTER_ACCEPT_PROBE_REQ |
+				       MAC_CFG_FILTER_ACCEPT_GRP);
+
+	/* TODO: Assumes that the beacon id == mac context id */
+	cmd.go_ibss.beacon_template = cpu_to_le32(mvmvif->id);
+
+	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
+}
+
+static int iwl_mvm_mld_mac_ctxt_cmd_p2p_device(struct iwl_mvm *mvm,
+					       struct ieee80211_vif *vif,
+					       u32 action)
+{
+	struct iwl_mac_config_cmd cmd = {};
+
+	WARN_ON(vif->type != NL80211_IFTYPE_P2P_DEVICE);
+
+	iwl_mvm_mld_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
+
+	cmd.p2p_dev.is_disc_extended =
+		iwl_mac_ctxt_p2p_dev_has_extended_disc(mvm, vif);
+
+	/* Override the filter flags to accept only probe requests */
+	cmd.filter_flags = cpu_to_le32(MAC_CFG_FILTER_ACCEPT_PROBE_REQ);
+
+	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
+}
+
+static int iwl_mvm_mld_mac_ctxt_cmd_ap_go(struct iwl_mvm *mvm,
+					  struct ieee80211_vif *vif,
+					  u32 action)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mac_config_cmd cmd = {};
+
+	WARN_ON(vif->type != NL80211_IFTYPE_AP);
+
+	/* Fill the common data for all mac context types */
+	iwl_mvm_mld_mac_ctxt_cmd_common(mvm, vif, &cmd, action);
+
+	iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(mvm, mvmvif,
+						 &cmd.filter_flags,
+						 MAC_CFG_FILTER_ACCEPT_PROBE_REQ,
+						 MAC_CFG_FILTER_ACCEPT_BEACON);
+
+	/* TODO: Assume that the beacon id == mac context id */
+	cmd.go_ibss.beacon_template = cpu_to_le32(mvmvif->id);
+
+	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
+}
+
+static int iwl_mvm_mld_mac_ctx_send(struct iwl_mvm *mvm,
+				    struct ieee80211_vif *vif,
+				    u32 action, bool force_assoc_off)
+{
+	switch (vif->type) {
+	case NL80211_IFTYPE_STATION:
+		return iwl_mvm_mld_mac_ctxt_cmd_sta(mvm, vif, action,
+						    force_assoc_off);
+	case NL80211_IFTYPE_AP:
+		return iwl_mvm_mld_mac_ctxt_cmd_ap_go(mvm, vif, action);
+	case NL80211_IFTYPE_MONITOR:
+		return iwl_mvm_mld_mac_ctxt_cmd_listener(mvm, vif, action);
+	case NL80211_IFTYPE_P2P_DEVICE:
+		return iwl_mvm_mld_mac_ctxt_cmd_p2p_device(mvm, vif, action);
+	case NL80211_IFTYPE_ADHOC:
+		return iwl_mvm_mld_mac_ctxt_cmd_ibss(mvm, vif, action);
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+int iwl_mvm_mld_mac_ctxt_add(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	if (WARN_ON_ONCE(vif->type == NL80211_IFTYPE_NAN))
+		return -EOPNOTSUPP;
+
+	if (WARN_ONCE(mvmvif->uploaded, "Adding active MAC %pM/%d\n",
+		      vif->addr, ieee80211_vif_type_p2p(vif)))
+		return -EIO;
+
+	ret = iwl_mvm_mld_mac_ctx_send(mvm, vif, FW_CTXT_ACTION_ADD,
+				       true);
+	if (ret)
+		return ret;
+
+	/* will only do anything at resume from D3 time */
+	iwl_mvm_set_last_nonqos_seq(mvm, vif);
+
+	mvmvif->uploaded = true;
+	return 0;
+}
+
+int iwl_mvm_mld_mac_ctxt_changed(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif,
+				 bool force_assoc_off)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	if (WARN_ON_ONCE(vif->type == NL80211_IFTYPE_NAN))
+		return -EOPNOTSUPP;
+
+	if (WARN_ONCE(!mvmvif->uploaded, "Changing inactive MAC %pM/%d\n",
+		      vif->addr, ieee80211_vif_type_p2p(vif)))
+		return -EIO;
+
+	return iwl_mvm_mld_mac_ctx_send(mvm, vif, FW_CTXT_ACTION_MODIFY,
+					force_assoc_off);
+}
+
+int iwl_mvm_mld_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mac_config_cmd cmd = {
+		.action = cpu_to_le32(FW_CTXT_ACTION_REMOVE),
+		.id_and_color = cpu_to_le32(mvmvif->id),
+	};
+	int ret;
+
+	if (WARN_ON_ONCE(vif->type == NL80211_IFTYPE_NAN))
+		return -EOPNOTSUPP;
+
+	if (WARN_ONCE(!mvmvif->uploaded, "Removing inactive MAC %pM/%d\n",
+		      vif->addr, ieee80211_vif_type_p2p(vif)))
+		return -EIO;
+
+	ret = iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
+	if (ret)
+		return ret;
+
+	mvmvif->uploaded = false;
+
+	return 0;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 6bd1a4c72a12..1287207a34b3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1477,6 +1477,7 @@ void iwl_mvm_hwrate_to_tx_rate_v1(u32 rate_n_flags,
 				  struct ieee80211_tx_rate *r);
 u8 iwl_mvm_mac80211_idx_to_hwrate(const struct iwl_fw *fw, int rate_idx);
 u8 iwl_mvm_mac80211_ac_to_ucode_ac(enum ieee80211_ac_numbers ac);
+bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 
 static inline void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
 {
@@ -1638,6 +1639,23 @@ u8 iwl_mvm_get_channel_width(struct cfg80211_chan_def *chandef);
 u8 iwl_mvm_get_ctrl_pos(struct cfg80211_chan_def *chandef);
 
 /* MAC (virtual interface) programming */
+
+__le32 iwl_mac_ctxt_p2p_dev_has_extended_disc(struct iwl_mvm *mvm,
+					      struct ieee80211_vif *vif);
+void iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(struct iwl_mvm *mvm,
+					      struct iwl_mvm_vif *mvmvif,
+					      __le32 *filter_flags,
+					      int accept_probe_req_flag,
+					      int accept_beacon_flag);
+int iwl_mvm_get_mac_type(struct ieee80211_vif *vif);
+__le32 iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(struct iwl_mvm *mvm,
+						    struct ieee80211_vif *vif);
+__le32 iwl_mvm_mac_ctxt_cmd_sta_get_twt_policy(struct iwl_mvm *mvm,
+					       struct ieee80211_vif *vif);
+int iwl_mvm_mld_mac_ctxt_add(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				 bool force_assoc_off);
+int iwl_mvm_mld_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mac_ctxt_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mac_ctxt_add(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index f4e9446d9dc2..69c390ff2745 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -521,6 +521,7 @@ static const struct iwl_hcmd_names iwl_mvm_system_names[] = {
 static const struct iwl_hcmd_names iwl_mvm_mac_conf_names[] = {
 	HCMD_NAME(CHANNEL_SWITCH_TIME_EVENT_CMD),
 	HCMD_NAME(SESSION_PROTECTION_CMD),
+	HCMD_NAME(MAC_CONFIG_CMD),
 	HCMD_NAME(SESSION_PROTECTION_NOTIF),
 	HCMD_NAME(CHANNEL_SWITCH_START_NOTIF),
 };
-- 
cgit v1.2.3-70-g09d2


From 55eb1c5fa4b260491d8be3299d4546d0b34465f4 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:19 +0200
Subject: wifi: iwlwifi: mvm: add support for the new LINK command

As a part of the new MLD FW API changes, we have a new LINK command
to add/remove/configure a link.
Add structures and enum definitions, along with the functions that
sends this command (i.e. add, remove and change mac ctxt).
These functions will be in used in the next patches.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.d7808329effb.I13bea2db206b78540bc866bc3ab755ad5be78c53@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/mac-cfg.h    | 179 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/Makefile    |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      | 175 ++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  |  26 ++-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |   7 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  20 +++
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |   1 +
 7 files changed, 390 insertions(+), 20 deletions(-)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/link.c

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index f12b2a25cd3d..301f34f13523 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -37,6 +37,10 @@ enum iwl_mac_conf_subcmd_ids {
 	 * @MAC_CONFIG_CMD: &struct iwl_mac_config_cmd
 	 */
 	MAC_CONFIG_CMD = 0x8,
+	/**
+	 * @LINK_CONFIG_CMD: &struct iwl_link_config_cmd
+	 */
+	LINK_CONFIG_CMD = 0x9,
 	/**
 	 * @SESSION_PROTECTION_NOTIF: &struct iwl_mvm_session_prot_notif
 	 */
@@ -287,4 +291,179 @@ struct iwl_mac_config_cmd {
 	};
 } __packed; /* MAC_CONTEXT_CONFIG_CMD_API_S_VER_1 */
 
+/**
+ * enum iwl_link_ctx_modify_flags - indicate to the fw what fields are being
+ *	modified in &iwl_link_ctx_cfg_cmd
+ *
+ * @LINK_CONTEXT_MODIFY_ACTIVE: covers iwl_link_ctx_cfg_cmd::active
+ * @LINK_CONTEXT_MODIFY_RATES_INFO: covers iwl_link_ctx_cfg_cmd::cck_rates,
+ *	iwl_link_ctx_cfg_cmd::ofdm_rates,
+ *	iwl_link_ctx_cfg_cmd::cck_short_preamble,
+ *	iwl_link_ctx_cfg_cmd::short_slot
+ * @LINK_CONTEXT_MODIFY_PROTECT_FLAGS: covers
+ *	iwl_link_ctx_cfg_cmd::protection_flags
+ * @LINK_CONTEXT_MODIFY_QOS_PARAMS: covers iwl_link_ctx_cfg_cmd::qos_flags,
+ *	iwl_link_ctx_cfg_cmd::ac,
+ * @LINK_CONTEXT_MODIFY_BEACON_TIMING: covers iwl_link_ctx_cfg_cmd::bi,
+ *	iwl_link_ctx_cfg_cmd::dtim_interval,
+ *	iwl_link_ctx_cfg_cmd::dtim_time,
+ *	iwl_link_ctx_cfg_cmd::dtim_tsf,
+ *	iwl_link_ctx_cfg_cmd::assoc_beacon_arrive_time.
+ *	This flag can be set only once after assoc.
+ * @LINK_CONTEXT_MODIFY_HE_PARAMS: covers
+ *	iwl_link_ctx_cfg_cmd::htc_trig_based_pkt_ext
+ *	iwl_link_ctx_cfg_cmd::rand_alloc_ecwmin,
+ *	iwl_link_ctx_cfg_cmd::rand_alloc_ecwmax,
+ *	iwl_link_ctx_cfg_cmd::trig_based_txf,
+ *	iwl_link_ctx_cfg_cmd::bss_color,
+ *	iwl_link_ctx_cfg_cmd::ndp_fdbk_buff_th_exp,
+ *	iwl_link_ctx_cfg_cmd::ref_bssid_addr
+ *	iwl_link_ctx_cfg_cmd::bssid_index,
+ *	iwl_link_ctx_cfg_cmd::frame_time_rts_th.
+ *	This flag can be set any time.
+ * @LINK_CONTEXT_MODIFY_BSS_COLOR_DISABLE: covers
+ *	iwl_link_ctx_cfg_cmd::bss_color_disable
+ * @LINK_CONTEXT_MODIFY_EHT_PARAMS: covers iwl_link_ctx_cfg_cmd::puncture_mask.
+ *	This flag can be set only if the MAC that this link relates to has
+ *	eht_support set to true.
+ * @LINK_CONTEXT_MODIFY_ALL: set all above flags
+ */
+enum iwl_link_ctx_modify_flags {
+	LINK_CONTEXT_MODIFY_ACTIVE		= BIT(0),
+	LINK_CONTEXT_MODIFY_RATES_INFO		= BIT(1),
+	LINK_CONTEXT_MODIFY_PROTECT_FLAGS	= BIT(2),
+	LINK_CONTEXT_MODIFY_QOS_PARAMS		= BIT(3),
+	LINK_CONTEXT_MODIFY_BEACON_TIMING	= BIT(4),
+	LINK_CONTEXT_MODIFY_HE_PARAMS		= BIT(5),
+	LINK_CONTEXT_MODIFY_BSS_COLOR_DISABLE	= BIT(6),
+	LINK_CONTEXT_MODIFY_EHT_PARAMS		= BIT(7),
+	LINK_CONTEXT_MODIFY_ALL			= 0xff,
+}; /* LINK_CONTEXT_MODIFY_MASK_E_VER_1 */
+
+/**
+ * enum iwl_link_ctx_protection_flags - link protection flags
+ * @LINK_PROT_FLG_TGG_PROTECT: 11g protection when transmitting OFDM frames,
+ *	this will require CCK RTS/CTS2self.
+ *	RTS/CTS will protect full burst time.
+ * @LINK_PROT_FLG_HT_PROT: enable HT protection
+ * @LINK_PROT_FLG_FAT_PROT: protect 40 MHz transmissions
+ * @LINK_PROT_FLG_SELF_CTS_EN: allow CTS2self
+ */
+enum iwl_link_ctx_protection_flags {
+	LINK_PROT_FLG_TGG_PROTECT	= BIT(0),
+	LINK_PROT_FLG_HT_PROT		= BIT(1),
+	LINK_PROT_FLG_FAT_PROT		= BIT(2),
+	LINK_PROT_FLG_SELF_CTS_EN	= BIT(3),
+}; /* LINK_PROTECT_FLAGS_E_VER_1 */
+
+/**
+ * enum iwl_link_ctx_flags - link context flags
+ *
+ * @LINK_FLG_BSS_COLOR_DIS: BSS color disable, don't use the BSS
+ *	color for RX filter but use MAC header
+ *	enabled AGG, i.e. both BACK and non-BACK frames in a single AGG
+ * @LINK_FLG_MU_EDCA_CW: indicates that there is an element of MU EDCA
+ *	parameter set, i.e. the backoff counters for trig-based ACs
+ * @LINK_FLG_RU_2MHZ_BLOCK: indicates that 26-tone RU OFDMA transmission are
+ *      not allowed (as there are OBSS that might classify such transmissions as
+ *      radar pulses).
+ */
+enum iwl_link_ctx_flags {
+	LINK_FLG_BSS_COLOR_DIS		= BIT(0),
+	LINK_FLG_MU_EDCA_CW		= BIT(1),
+	LINK_FLG_RU_2MHZ_BLOCK		= BIT(2),
+}; /* LINK_CONTEXT_FLAG_E_VER_1 */
+
+/**
+ * struct iwl_link_config_cmd - command structure to configure the LINK context
+ *	in MLD API
+ * ( LINK_CONFIG_CMD =0x9 )
+ *
+ * @action: action to perform, one of FW_CTXT_ACTION_*
+ * @link_id: the id of the link that this cmd configures
+ * @mac_id: interface ID. Relevant only if action is FW_CTXT_ACTION_ADD
+ * @phy_id: PHY index. Can be changed only if the link was inactive
+ *	(and stays inactive). If the link is active (or becomes active),
+ *	this field is ignored.
+ * @local_link_addr: the links MAC address. Can be changed only if the link was
+ *	inactive (and stays inactive). If the link is active
+ *	(or becomes active), this field is ignored.
+ * @reserved_for_local_link_addr: reserved
+ * @modify_mask: from &enum iwl_link_ctx_modify_flags, selects what to change.
+ *	Relevant only if action is FW_CTXT_ACTION_MODIFY
+ * @active: indicates whether the link is active or not
+ * @listen_lmac: indicates whether the link should be allocated on the Listen
+ *	Lmac or on the Main Lmac. Cannot be changed on an active Link.
+ *	Relevant only for eSR.
+ * @cck_rates: basic rates available for CCK
+ * @ofdm_rates: basic rates available for OFDM
+ * @cck_short_preamble: 1 for enabling short preamble, 0 otherwise
+ * @short_slot: 1 for enabling short slots, 0 otherwise
+ * @protection_flags: combination of &enum iwl_link_ctx_protection_flags
+ * @qos_flags: from &enum iwl_mac_qos_flags
+ * @ac: one iwl_mac_qos configuration for each AC
+ * @htc_trig_based_pkt_ext: default PE in 4us units
+ * @rand_alloc_ecwmin: random CWmin = 2**ECWmin-1
+ * @rand_alloc_ecwmax: random CWmax = 2**ECWmax-1
+ * @ndp_fdbk_buff_th_exp: set exponent for the NDP feedback buffered threshold
+ * @trig_based_txf: MU EDCA Parameter set for the trigger based traffic queues
+ * @dtim_time: DTIM arrival time in system time
+ * @dtim_tsf: DTIM arrival time in TSF
+ * @assoc_beacon_arrive_time: TSF of first beacon after association
+ * @bi: beacon interval in TU, applicable only when associated
+ * @dtim_interval: DTIM interval in TU.
+ *	Relevant only for GO, otherwise this is offloaded.
+ * @beacon_template: beacon template ID. For GO only
+ * @puncture_mask: puncture mask for EHT
+ * @frame_time_rts_th: HE duration RTS threshold, in units of 32us
+ * @flags: a combination from &enum iwl_link_ctx_flags
+ * @flags_mask: what of %flags have changed. Also &enum iwl_link_ctx_flags
+ * Below fields are for multi-bssid:
+ * @ref_bssid_addr: reference BSSID used by the AP
+ * @reserved_for_ref_bssid_addr: reserved
+ * @bssid_index: index of the associated VAP
+ * @bss_color: 11ax AP ID that is used in the HE SIG-A to mark inter BSS frame
+ * @reserved: alignment
+ */
+struct iwl_link_config_cmd {
+	__le32 action;
+	__le32 link_id;
+	__le32 mac_id;
+	__le32 phy_id;
+	u8 local_link_addr[6];
+	__le16 reserved_for_local_link_addr;
+	__le32 modify_mask;
+	__le32 active;
+	__le32 listen_lmac;
+	__le32 cck_rates;
+	__le32 ofdm_rates;
+	__le32 cck_short_preamble;
+	__le32 short_slot;
+	__le32 protection_flags;
+	/* MAC_QOS_PARAM_API_S_VER_1 */
+	__le32 qos_flags;
+	struct iwl_ac_qos ac[AC_NUM + 1];
+	u8 htc_trig_based_pkt_ext;
+	u8 rand_alloc_ecwmin;
+	u8 rand_alloc_ecwmax;
+	u8 ndp_fdbk_buff_th_exp;
+	struct iwl_he_backoff_conf trig_based_txf[AC_NUM];
+	__le32 dtim_time;
+	__le64 dtim_tsf;
+	__le32 assoc_beacon_arrive_time;
+	__le32 bi;
+	__le32 dtim_interval;
+	__le32 beacon_template;
+	__le16 puncture_mask;
+	__le16 frame_time_rts_th;
+	__le32 flags;
+	__le32 flags_mask;
+	/* The below fields are for multi-bssid */
+	u8 ref_bssid_addr[6];
+	__le16 reserved_for_ref_bssid_addr;
+	u8 bssid_index;
+	u8 bss_color;
+	u8 reserved[2];
+} __packed; /* LINK_CONTEXT_CONFIG_CMD_API_S_VER_1 */
+
 #endif /* __iwl_fw_api_mac_cfg_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
index d71f0a99b7c4..ec6346c6c8e4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
@@ -7,7 +7,7 @@ iwlmvm-y += power.o coex.o
 iwlmvm-y += tt.o offloading.o tdls.o
 iwlmvm-y += ftm-responder.o ftm-initiator.o
 iwlmvm-y += rfi.o
-iwlmvm-y += mld-key.o mld-mac.o
+iwlmvm-y += mld-key.o mld-mac.o link.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
 iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM) += d3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
new file mode 100644
index 000000000000..2688cb49c951
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -0,0 +1,175 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2022 Intel Corporation
+ */
+#include "mvm.h"
+
+static int iwl_mvm_link_cmd_send(struct iwl_mvm *mvm,
+				 struct iwl_link_config_cmd *cmd,
+				 enum iwl_ctxt_action action)
+{
+	int ret;
+
+	cmd->action = cpu_to_le32(action);
+	ret = iwl_mvm_send_cmd_pdu(mvm,
+				   WIDE_ID(MAC_CONF_GROUP, LINK_CONFIG_CMD), 0,
+				   sizeof(*cmd), cmd);
+	if (ret)
+		IWL_ERR(mvm, "Failed to send LINK_CONFIG_CMD (action:%d): %d\n",
+			action, ret);
+	return ret;
+}
+
+int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_link_config_cmd cmd = {};
+
+	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+		return -EINVAL;
+
+	/* Update SF - Disable if needed. if this fails, SF might still be on
+	 * while many macs are bound, which is forbidden - so fail the binding.
+	 */
+	if (iwl_mvm_sf_update(mvm, vif, false))
+		return -EINVAL;
+
+	cmd.link_id = cpu_to_le32(mvmvif->phy_ctxt->id);
+	cmd.mac_id = cpu_to_le32(mvmvif->id);
+	cmd.phy_id = cpu_to_le32(mvmvif->phy_ctxt->id);
+
+	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
+
+	return iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_ADD);
+}
+
+int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 u32 changes, bool active)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_phy_ctxt *phyctxt = mvmvif->phy_ctxt;
+	struct iwl_link_config_cmd cmd = {};
+	u32 ht_flag, flags = 0, flags_mask = 0;
+
+	if (!phyctxt)
+		return -EINVAL;
+
+	cmd.link_id = cpu_to_le32(phyctxt->id);
+
+	/* The phy_id, link address and listen_lmac can be modified only until
+	 * the link becomes active, otherwise they will be ignored.
+	 */
+	cmd.phy_id = cpu_to_le32(phyctxt->id);
+	cmd.mac_id = cpu_to_le32(mvmvif->id);
+
+	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
+
+	cmd.active = cpu_to_le32(active);
+
+	/* TODO: set a value to cmd.listen_lmac when system requiremens
+	 * will define it
+	 */
+
+	iwl_mvm_set_fw_basic_rates(mvm, vif, &cmd.cck_rates, &cmd.ofdm_rates);
+
+	cmd.cck_short_preamble = cpu_to_le32(vif->bss_conf.use_short_preamble);
+	cmd.short_slot = cpu_to_le32(vif->bss_conf.use_short_slot);
+
+	/* The fw does not distinguish between ht and fat */
+	ht_flag = LINK_PROT_FLG_HT_PROT | LINK_PROT_FLG_FAT_PROT;
+	iwl_mvm_set_fw_protection_flags(mvm, vif, &cmd.protection_flags,
+					ht_flag, LINK_PROT_FLG_TGG_PROTECT);
+
+	iwl_mvm_set_fw_qos_params(mvm, vif, &cmd.ac[0], &cmd.qos_flags);
+
+	/* We need the dtim_period to set the MAC as associated */
+	if (vif->cfg.assoc && vif->bss_conf.dtim_period)
+		iwl_mvm_set_fw_dtim_tbtt(mvm, vif, &cmd.dtim_tsf,
+					 &cmd.dtim_time,
+					 &cmd.assoc_beacon_arrive_time);
+	else
+		changes &= ~LINK_CONTEXT_MODIFY_BEACON_TIMING;
+
+	cmd.bi = cpu_to_le32(vif->bss_conf.beacon_int);
+	cmd.dtim_interval = cpu_to_le32(vif->bss_conf.beacon_int *
+					vif->bss_conf.dtim_period);
+
+	/* TODO: Assumes that the beacon id == mac context id */
+	cmd.beacon_template = cpu_to_le32(mvmvif->id);
+
+	if (!vif->bss_conf.he_support || iwlwifi_mod_params.disable_11ax ||
+	    !vif->cfg.assoc) {
+		changes &= ~LINK_CONTEXT_MODIFY_HE_PARAMS;
+		goto send_cmd;
+	}
+
+	cmd.htc_trig_based_pkt_ext = vif->bss_conf.htc_trig_based_pkt_ext;
+
+	if (vif->bss_conf.uora_exists) {
+		cmd.rand_alloc_ecwmin =
+			vif->bss_conf.uora_ocw_range & 0x7;
+		cmd.rand_alloc_ecwmax =
+			(vif->bss_conf.uora_ocw_range >> 3) & 0x7;
+	}
+
+	/* TODO  how to set ndp_fdbk_buff_th_exp? */
+
+	if (iwl_mvm_set_fw_mu_edca_params(mvm, mvmvif,
+					  &cmd.trig_based_txf[0])) {
+		flags |= LINK_FLG_MU_EDCA_CW;
+		flags_mask |= LINK_FLG_MU_EDCA_CW;
+	}
+
+	if (vif->bss_conf.eht_puncturing && !iwlwifi_mod_params.disable_11be)
+		cmd.puncture_mask = cpu_to_le16(vif->bss_conf.eht_puncturing);
+	else
+		/* This flag can be set only if the MAC has eht support */
+		changes &= ~LINK_CONTEXT_MODIFY_EHT_PARAMS;
+
+	cmd.bss_color = vif->bss_conf.he_bss_color.color;
+
+	if (!vif->bss_conf.he_bss_color.enabled) {
+		flags |= LINK_FLG_BSS_COLOR_DIS;
+		flags_mask |= LINK_FLG_BSS_COLOR_DIS;
+	}
+
+	cmd.frame_time_rts_th = cpu_to_le16(vif->bss_conf.frame_time_rts_th);
+
+	/* Block 26-tone RU OFDMA transmissions */
+	if (mvmvif->he_ru_2mhz_block) {
+		flags |= LINK_FLG_RU_2MHZ_BLOCK;
+		flags_mask |= LINK_FLG_RU_2MHZ_BLOCK;
+	}
+
+	if (vif->bss_conf.nontransmitted) {
+		ether_addr_copy(cmd.ref_bssid_addr,
+				vif->bss_conf.transmitter_bssid);
+		cmd.bssid_index = vif->bss_conf.bssid_index;
+	}
+
+send_cmd:
+	cmd.modify_mask = cpu_to_le32(changes);
+	cmd.flags = cpu_to_le32(flags);
+	cmd.flags_mask = cpu_to_le32(flags_mask);
+
+	return iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_MODIFY);
+}
+
+int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_link_config_cmd cmd = {};
+	int ret;
+
+	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+		return -EINVAL;
+
+	cmd.link_id = cpu_to_le32(mvmvif->phy_ctxt->id);
+	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_REMOVE);
+
+	if (!ret)
+		if (iwl_mvm_sf_update(mvm, vif, true))
+			IWL_ERR(mvm, "Failed to update SF state\n");
+
+	return ret;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index b21327ec4401..4d7ad91c53d4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -396,9 +396,8 @@ static void iwl_mvm_ack_rates(struct iwl_mvm *mvm,
 	*ofdm_rates = ofdm;
 }
 
-static void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm,
-				       struct ieee80211_vif *vif,
-				       __le32 *cck_rates, __le32 *ofdm_rates)
+void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				__le32 *cck_rates, __le32 *ofdm_rates)
 {
 	struct ieee80211_chanctx_conf *chanctx;
 	u8 cck_ack_rates, ofdm_ack_rates;
@@ -414,11 +413,10 @@ static void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm,
 	*ofdm_rates = cpu_to_le32((u32)ofdm_ack_rates);
 }
 
-static void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
-					    struct ieee80211_vif *vif,
-					    __le32 *protection_flags,
-					    u32 ht_flag,
-					    u32 tgg_flag)
+void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
+				     struct ieee80211_vif *vif,
+				     __le32 *protection_flags, u32 ht_flag,
+				     u32 tgg_flag)
 {
 	/* for both sta and ap, ht_operation_mode hold the protection_mode */
 	u8 protection_mode = vif->bss_conf.ht_operation_mode &
@@ -460,9 +458,8 @@ static void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
 	}
 }
 
-static void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm,
-				      struct ieee80211_vif *vif,
-				      struct iwl_ac_qos *ac, __le32 *qos_flags)
+void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			       struct iwl_ac_qos *ac, __le32 *qos_flags)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int i;
@@ -572,10 +569,9 @@ static int iwl_mvm_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
 	return ret;
 }
 
-static void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm,
-				     struct ieee80211_vif *vif,
-				     __le64 *dtim_tsf, __le32 *dtim_time,
-				     __le32 *assoc_beacon_arrive_time)
+void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      __le64 *dtim_tsf, __le32 *dtim_time,
+			      __le32 *assoc_beacon_arrive_time)
 {
 	u32 dtim_offs;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 210b134c735f..cf08cb834cc4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2000,10 +2000,9 @@ static void iwl_mvm_get_optimal_ppe_info(struct iwl_he_pkt_ext_v2 *pkt_ext,
  * This function sets the MU EDCA parameters ans returns whether MU EDCA
  * is enabled or not
  */
-static bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
-					  struct iwl_mvm_vif *mvmvif,
-					  struct iwl_he_backoff_conf
-					  *trig_based_txf)
+bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
+				   struct iwl_mvm_vif *mvmvif,
+				   struct iwl_he_backoff_conf *trig_based_txf)
 {
 	int i;
 	/* Mark MU EDCA as enabled, unless none detected on some AC */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 1287207a34b3..4663cb7a0b10 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1640,6 +1640,20 @@ u8 iwl_mvm_get_ctrl_pos(struct cfg80211_chan_def *chandef);
 
 /* MAC (virtual interface) programming */
 
+void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				__le32 *cck_rates, __le32 *ofdm_rates);
+void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
+				     struct ieee80211_vif *vif,
+				     __le32 *protection_flags, u32 ht_flag,
+				  u32 tgg_flag);
+void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			       struct iwl_ac_qos *ac, __le32 *qos_flags);
+bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
+				   struct iwl_mvm_vif *mvmvif,
+				   struct iwl_he_backoff_conf *trig_based_txf);
+void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      __le64 *dtim_tsf, __le32 *dtim_time,
+			      __le32 *assoc_beacon_arrive_time);
 __le32 iwl_mac_ctxt_p2p_dev_has_extended_disc(struct iwl_mvm *mvm,
 					      struct ieee80211_vif *vif);
 void iwl_mvm_mac_ctxt_cmd_ap_set_filter_flags(struct iwl_mvm *mvm,
@@ -1702,6 +1716,12 @@ void iwl_mvm_channel_switch_error_notif(struct iwl_mvm *mvm,
 int iwl_mvm_binding_add_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_binding_remove_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 
+/* Links */
+int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 u32 changes, bool active);
+int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+
 /* Quota management */
 static inline size_t iwl_mvm_quota_cmd_size(struct iwl_mvm *mvm)
 {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 69c390ff2745..c3767f62026a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -522,6 +522,7 @@ static const struct iwl_hcmd_names iwl_mvm_mac_conf_names[] = {
 	HCMD_NAME(CHANNEL_SWITCH_TIME_EVENT_CMD),
 	HCMD_NAME(SESSION_PROTECTION_CMD),
 	HCMD_NAME(MAC_CONFIG_CMD),
+	HCMD_NAME(LINK_CONFIG_CMD),
 	HCMD_NAME(SESSION_PROTECTION_NOTIF),
 	HCMD_NAME(CHANNEL_SWITCH_START_NOTIF),
 };
-- 
cgit v1.2.3-70-g09d2


From 006c152ac9e56ac7871efa995854c3ff8cf6915a Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:20 +0200
Subject: wifi: iwlwifi: mvm: add support for the new STA related commands

As a part of the new MLD FW API changes, we have new commands for STA
related operations (add/remove/aux/disable tx).
Add structures and enum definitions, along with part of the functions
that sends this commands.
This functions will be in used and more will be added in the next patches.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.132873ce015c.I7b12a77e5be066730762e6ceeeaa7190293c3df1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/mac-cfg.h    | 128 ++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/Makefile    |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 281 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |   4 +
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       |   8 +-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  16 +-
 6 files changed, 431 insertions(+), 8 deletions(-)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index 301f34f13523..a111e7366d49 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -41,6 +41,22 @@ enum iwl_mac_conf_subcmd_ids {
 	 * @LINK_CONFIG_CMD: &struct iwl_link_config_cmd
 	 */
 	LINK_CONFIG_CMD = 0x9,
+	/**
+	 * @STA_CONFIG_CMD: &struct iwl_mvm_sta_cfg_cmd
+	 */
+	STA_CONFIG_CMD = 0xA,
+	/**
+	 * @AUX_STA_CMD: &struct iwl_mvm_aux_sta_cmd
+	 */
+	AUX_STA_CMD = 0xB,
+	/**
+	 * @STA_REMOVE_CMD: &struct iwl_mvm_remove_sta_cmd
+	 */
+	STA_REMOVE_CMD = 0xC,
+	/**
+	 * @STA_DISABLE_TX_CMD: &struct iwl_mvm_sta_disable_tx_cmd
+	 */
+	STA_DISABLE_TX_CMD = 0xD,
 	/**
 	 * @SESSION_PROTECTION_NOTIF: &struct iwl_mvm_session_prot_notif
 	 */
@@ -466,4 +482,116 @@ struct iwl_link_config_cmd {
 	u8 reserved[2];
 } __packed; /* LINK_CONTEXT_CONFIG_CMD_API_S_VER_1 */
 
+/**
+ * enum iwl_fw_sta_type - FW station types
+ * @STATION_TYPE_PEER: represents a peer - AP in BSS, a TDLS sta, a client in
+ *	P2P.
+ * @STATION_TYPE_BCAST_MGMT: The station used to send beacons and
+ *	probe responses. Also used for traffic injection in sniffer mode
+ * @STATION_TYPE_MCAST: the station used for BCAST / MCAST in GO. Will be
+ *	suspended / resumed at the right timing depending on the clients'
+ *	power save state and the DTIM timing
+ */
+enum iwl_fw_sta_type {
+	STATION_TYPE_PEER,
+	STATION_TYPE_BCAST_MGMT,
+	STATION_TYPE_MCAST,
+}; /* STATION_TYPE_E_VER_1 */
+
+/**
+ * struct iwl_mvm_sta_cfg_cmd - cmd structure to add a peer sta to the uCode's
+ *	station table
+ * ( STA_CONFIG_CMD = 0xA )
+ *
+ * @sta_id: index of station in uCode's station table
+ * @link_id: the id of the link that is used to communicate with this sta
+ * @peer_mld_address: the peers mld address
+ * @reserved_for_peer_mld_address: reserved
+ * @peer_link_address: the address of the link that is used to communicate
+ *	with this sta
+ * @reserved_for_peer_link_address: reserved
+ * @station_type: type of this station. See &enum iwl_fw_sta_type
+ * @assoc_id: for GO only
+ * @beamform_flags: beam forming controls
+ * @mfp: indicates whether the STA uses management frame protection or not.
+ * @mimo: indicates whether the sta uses mimo or not
+ * @mimo_protection: indicates whether the sta uses mimo protection or not
+ * @ack_enabled: indicates that the AP supports receiving ACK-
+ *	enabled AGG, i.e. both BACK and non-BACK frames in a single AGG
+ * @trig_rnd_alloc: indicates that trigger based random allocation
+ *	is enabled according to UORA element existence
+ * @tx_ampdu_spacing: minimum A-MPDU spacing:
+ *	4 - 2us density, 5 - 4us density, 6 - 8us density, 7 - 16us density
+ * @tx_ampdu_max_size: maximum A-MPDU length: 0 - 8K, 1 - 16K, 2 - 32K,
+ *	3 - 64K, 4 - 128K, 5 - 256K, 6 - 512K, 7 - 1024K.
+ * @sp_length: the size of the SP in actual number of frames
+ * @uapsd_acs:  4 LS bits are trigger enabled ACs, 4 MS bits are the deliver
+ *	enabled ACs.
+ * @pkt_ext: optional, exists according to PPE-present bit in the HE/EHT-PHY
+ *	capa
+ * @htc_flags: which features are supported in HTC
+ */
+struct iwl_mvm_sta_cfg_cmd {
+	__le32 sta_id;
+	__le32 link_id;
+	u8 peer_mld_address[ETH_ALEN];
+	__le16 reserved_for_peer_mld_address;
+	u8 peer_link_address[ETH_ALEN];
+	__le16 reserved_for_peer_link_address;
+	__le32 station_type;
+	__le32 assoc_id;
+	__le32 beamform_flags;
+	__le32 mfp;
+	__le32 mimo;
+	__le32 mimo_protection;
+	__le32 ack_enabled;
+	__le32 trig_rnd_alloc;
+	__le32 tx_ampdu_spacing;
+	__le32 tx_ampdu_max_size;
+	__le32 sp_length;
+	__le32 uapsd_acs;
+	struct iwl_he_pkt_ext_v2 pkt_ext;
+	__le32 htc_flags;
+} __packed; /* STA_CMD_API_S_VER_1 */
+
+/**
+ * struct iwl_mvm_aux_sta_cmd - command for AUX STA configuration
+ * ( AUX_STA_CMD = 0xB )
+ *
+ * @sta_id: index of aux sta to configure
+ * @lmac_id: ?
+ * @mac_addr: mac addr of the auxilary sta
+ * @reserved_for_mac_addr: reserved
+ */
+struct iwl_mvm_aux_sta_cmd {
+	__le32 sta_id;
+	__le32 lmac_id;
+	u8 mac_addr[ETH_ALEN];
+	__le16 reserved_for_mac_addr;
+
+} __packed; /* AUX_STA_CMD_API_S_VER_1 */
+
+/**
+ * struct iwl_mvm_remove_sta_cmd - a cmd structure to remove a sta added by
+ *	STA_CONFIG_CMD or AUX_STA_CONFIG_CMD
+ * ( STA_REMOVE_CMD = 0xC )
+ *
+ * @sta_id: index of station to remove
+ */
+struct iwl_mvm_remove_sta_cmd {
+	__le32 sta_id;
+} __packed; /* REMOVE_STA_API_S_VER_1 */
+
+/**
+ * struct iwl_mvm_sta_disable_tx_cmd - disable / re-enable tx to a sta
+ * ( STA_DISABLE_TX_CMD = 0xD )
+ *
+ * @sta_id: index of the station to disable tx to
+ * @disable: indicates if to disable or re-enable tx
+ */
+struct iwl_mvm_sta_disable_tx_cmd {
+	__le32 sta_id;
+	__le32 disable;
+} __packed; /* STA_DISABLE_TX_API_S_VER_1 */
+
 #endif /* __iwl_fw_api_mac_cfg_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
index ec6346c6c8e4..0e9b5381e265 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
@@ -7,7 +7,7 @@ iwlmvm-y += power.o coex.o
 iwlmvm-y += tt.o offloading.o tdls.o
 iwlmvm-y += ftm-responder.o ftm-initiator.o
 iwlmvm-y += rfi.o
-iwlmvm-y += mld-key.o mld-mac.o link.o
+iwlmvm-y += mld-key.o mld-mac.o link.o mld-sta.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
 iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM) += d3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
new file mode 100644
index 000000000000..cef4fc441afe
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -0,0 +1,281 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2022 Intel Corporation
+ */
+#include "mvm.h"
+
+static int iwl_mvm_mld_send_sta_cmd(struct iwl_mvm *mvm,
+				    struct iwl_mvm_sta_cfg_cmd *cmd)
+{
+	int ret = iwl_mvm_send_cmd_pdu(mvm,
+				       WIDE_ID(MAC_CONF_GROUP, STA_CONFIG_CMD),
+				       0, sizeof(*cmd), cmd);
+	if (ret)
+		IWL_ERR(mvm, "STA_CONFIG_CMD send failed, ret=0x%x\n", ret);
+	return ret;
+}
+
+/*
+ * Add an internal station to the FW table
+ */
+static int iwl_mvm_mld_add_int_sta_to_fw(struct iwl_mvm *mvm,
+					 struct iwl_mvm_int_sta *sta,
+					 const u8 *addr,
+					 u16 phy_id)
+{
+	struct iwl_mvm_sta_cfg_cmd cmd;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.sta_id = cpu_to_le32((u8)sta->sta_id);
+
+	cmd.link_id = cpu_to_le32(phy_id);
+
+	cmd.station_type = cpu_to_le32(sta->type);
+
+	if (addr) {
+		memcpy(cmd.peer_mld_address, addr, ETH_ALEN);
+		memcpy(cmd.peer_link_address, addr, ETH_ALEN);
+	}
+
+	return iwl_mvm_mld_send_sta_cmd(mvm, &cmd);
+}
+
+/*
+ * Remove a station from the FW table. Before sending the command to remove
+ * the station validate that the station is indeed known to the driver (sanity
+ * only).
+ */
+static int iwl_mvm_mld_rm_sta_from_fw(struct iwl_mvm *mvm, u32 sta_id)
+{
+	struct ieee80211_sta *sta;
+	struct iwl_mvm_remove_sta_cmd rm_sta_cmd = {
+		.sta_id = cpu_to_le32(sta_id),
+	};
+	int ret;
+
+	sta = rcu_dereference_protected(mvm->fw_id_to_mac_id[sta_id],
+					lockdep_is_held(&mvm->mutex));
+
+	/* Note: internal stations are marked as error values */
+	if (!sta) {
+		IWL_ERR(mvm, "Invalid station id\n");
+		return -EINVAL;
+	}
+
+	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(MAC_CONF_GROUP, STA_REMOVE_CMD),
+				   0, sizeof(rm_sta_cmd), &rm_sta_cmd);
+	if (ret) {
+		IWL_ERR(mvm, "Failed to remove station. Id=%d\n", sta_id);
+		return ret;
+	}
+
+	return 0;
+}
+
+/*
+ * Adds an internal sta to the FW table with its queues
+ */
+static int iwl_mvm_mld_add_int_sta_with_queue(struct iwl_mvm *mvm,
+					      struct iwl_mvm_int_sta *sta,
+					      const u8 *addr, int phy_id,
+					      u16 *queue, u8 tid,
+					      unsigned int *_wdg_timeout)
+{
+	int ret, txq;
+	unsigned int wdg_timeout = _wdg_timeout ? *_wdg_timeout :
+		mvm->trans->trans_cfg->base_params->wd_timeout;
+
+	if (WARN_ON_ONCE(sta->sta_id == IWL_MVM_INVALID_STA))
+		return -ENOSPC;
+
+	ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, phy_id);
+	if (ret)
+		return ret;
+
+	/*
+	 * For 22000 firmware and on we cannot add queue to a station unknown
+	 * to firmware so enable queue here - after the station was added
+	 */
+	txq = iwl_mvm_tvqm_enable_txq(mvm, NULL, sta->sta_id, tid,
+				      wdg_timeout);
+	if (txq < 0) {
+		iwl_mvm_mld_rm_sta_from_fw(mvm, sta->sta_id);
+		return txq;
+	}
+	*queue = txq;
+
+	return 0;
+}
+
+/*
+ * Adds a new int sta: allocate it in the driver, add it to the FW table,
+ * and add its queues.
+ */
+static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
+				   struct iwl_mvm_int_sta *int_sta, u16 *queue,
+				   enum nl80211_iftype iftype,
+				   enum iwl_fw_sta_type sta_type,
+				   int phy_id, const u8 *addr, u8 tid,
+				   unsigned int *wdg_timeout)
+{
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	/* qmask argument is not used in the new tx api, send a don't care */
+	ret = iwl_mvm_allocate_int_sta(mvm, int_sta, 0, iftype,
+				       sta_type);
+	if (ret)
+		return ret;
+
+	ret = iwl_mvm_mld_add_int_sta_with_queue(mvm, int_sta, addr, phy_id,
+						 queue, tid, wdg_timeout);
+	if (ret) {
+		iwl_mvm_dealloc_int_sta(mvm, int_sta);
+		return ret;
+	}
+
+	return 0;
+}
+
+/* Allocate a new station entry for the broadcast station to the given vif,
+ * and send it to the FW.
+ * Note that each P2P mac should have its own broadcast station.
+ */
+int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_int_sta *bsta = &mvmvif->bcast_sta;
+	static const u8 _baddr[] = {0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF};
+	const u8 *baddr = _baddr;
+	unsigned int wdg_timeout =
+		iwl_mvm_get_wd_timeout(mvm, vif, false, false);
+	u16 *queue;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (vif->type == NL80211_IFTYPE_ADHOC)
+		baddr = vif->bss_conf.bssid;
+
+	if (vif->type == NL80211_IFTYPE_AP ||
+	    vif->type == NL80211_IFTYPE_ADHOC) {
+		queue = &mvm->probe_queue;
+	} else if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
+		queue = &mvm->p2p_dev_queue;
+	} else {
+		WARN(1, "Missing required TXQ for adding bcast STA\n");
+		return -EINVAL;
+	}
+
+	return iwl_mvm_mld_add_int_sta(mvm, bsta, queue,
+				       ieee80211_vif_type_p2p(vif),
+				       STATION_TYPE_BCAST_MGMT,
+				       mvmvif->phy_ctxt->id, baddr,
+				       IWL_MAX_TID_COUNT, &wdg_timeout);
+}
+
+/* Allocate a new station entry for the sniffer station to the given vif,
+ * and send it to the FW.
+ */
+int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	lockdep_assert_held(&mvm->mutex);
+
+	return iwl_mvm_mld_add_int_sta(mvm, &mvm->snif_sta, &mvm->snif_queue,
+				       vif->type, STATION_TYPE_BCAST_MGMT,
+				       mvmvif->phy_ctxt->id, NULL,
+				       IWL_MAX_TID_COUNT, NULL);
+}
+
+static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm,
+				   struct ieee80211_sta *sta,
+				   u16 *queueptr, u8 tid)
+{
+	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
+	int queue = *queueptr;
+	int ret = 0;
+
+	if (mvm->sta_remove_requires_queue_remove) {
+		u32 cmd_id = WIDE_ID(DATA_PATH_GROUP,
+				     SCD_QUEUE_CONFIG_CMD);
+		struct iwl_scd_queue_cfg_cmd remove_cmd = {
+			.operation = cpu_to_le32(IWL_SCD_QUEUE_REMOVE),
+			.u.remove.tid = cpu_to_le32(tid),
+			.u.remove.sta_mask =
+				cpu_to_le32(BIT(mvmsta->sta_id)),
+		};
+
+		ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0,
+					   sizeof(remove_cmd),
+					   &remove_cmd);
+	}
+
+	iwl_trans_txq_free(mvm->trans, queue);
+	*queueptr = IWL_MVM_INVALID_QUEUE;
+
+	return ret;
+}
+
+/* Removes a sta from the FW table, disable its queues, and dealloc it
+ */
+static int iwl_mvm_mld_rm_int_sta(struct iwl_mvm *mvm,
+				  struct iwl_mvm_int_sta *int_sta,
+				  bool flush, u8 tid, u16 *queuptr)
+{
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (WARN_ON_ONCE(int_sta->sta_id == IWL_MVM_INVALID_STA))
+		return -EINVAL;
+
+	if (flush)
+		iwl_mvm_flush_sta(mvm, int_sta, true);
+
+	iwl_mvm_mld_disable_txq(mvm, NULL, queuptr, tid);
+
+	ret = iwl_mvm_mld_rm_sta_from_fw(mvm, int_sta->sta_id);
+	if (ret)
+		IWL_WARN(mvm, "Failed sending remove station\n");
+
+	iwl_mvm_dealloc_int_sta(mvm, int_sta);
+
+	return ret;
+}
+
+int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	u16 *queueptr;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	switch (vif->type) {
+	case NL80211_IFTYPE_AP:
+	case NL80211_IFTYPE_ADHOC:
+		queueptr = &mvm->probe_queue;
+		break;
+	case NL80211_IFTYPE_P2P_DEVICE:
+		queueptr = &mvm->p2p_dev_queue;
+		break;
+	default:
+		WARN(1, "Can't free bcast queue on vif type %d\n",
+		     vif->type);
+		return -EINVAL;
+	}
+
+	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->bcast_sta, true,
+				      IWL_MAX_TID_COUNT, queueptr);
+}
+
+int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	lockdep_assert_held(&mvm->mutex);
+
+	return iwl_mvm_mld_rm_int_sta(mvm, &mvm->snif_sta, false,
+				      IWL_MAX_TID_COUNT, &mvm->snif_queue);
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index c3767f62026a..03da4e53de67 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -523,6 +523,10 @@ static const struct iwl_hcmd_names iwl_mvm_mac_conf_names[] = {
 	HCMD_NAME(SESSION_PROTECTION_CMD),
 	HCMD_NAME(MAC_CONFIG_CMD),
 	HCMD_NAME(LINK_CONFIG_CMD),
+	HCMD_NAME(STA_CONFIG_CMD),
+	HCMD_NAME(AUX_STA_CMD),
+	HCMD_NAME(STA_REMOVE_CMD),
+	HCMD_NAME(STA_DISABLE_TX_CMD),
 	HCMD_NAME(SESSION_PROTECTION_NOTIF),
 	HCMD_NAME(CHANNEL_SWITCH_START_NOTIF),
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 46af2b422849..313c5416ae5e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -783,9 +783,9 @@ static int iwl_mvm_get_queue_size(struct ieee80211_sta *sta)
 	return IWL_DEFAULT_QUEUE_SIZE;
 }
 
-static int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
-				   struct ieee80211_sta *sta,
-				   u8 sta_id, u8 tid, unsigned int timeout)
+int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
+			    struct ieee80211_sta *sta,
+			    u8 sta_id, u8 tid, unsigned int timeout)
 {
 	int queue, size;
 
@@ -2005,7 +2005,7 @@ int iwl_mvm_rm_sta_id(struct iwl_mvm *mvm,
 int iwl_mvm_allocate_int_sta(struct iwl_mvm *mvm,
 			     struct iwl_mvm_int_sta *sta,
 			     u32 qmask, enum nl80211_iftype iftype,
-			     enum iwl_sta_type type)
+			     u8 type)
 {
 	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) ||
 	    sta->sta_id == IWL_MVM_INVALID_STA) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index f1a4fc3e4038..d11851b47684 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2021 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2022 Intel Corporation
  * Copyright (C) 2013-2014 Intel Mobile Communications GmbH
  * Copyright (C) 2015-2016 Intel Deutschland GmbH
  */
@@ -436,7 +436,7 @@ iwl_mvm_sta_from_mac80211(struct ieee80211_sta *sta)
  */
 struct iwl_mvm_int_sta {
 	u32 sta_id;
-	enum iwl_sta_type type;
+	u8 type;
 	u32 tfd_queue_msk;
 };
 
@@ -519,7 +519,7 @@ int iwl_mvm_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_allocate_int_sta(struct iwl_mvm *mvm,
 			     struct iwl_mvm_int_sta *sta,
 				    u32 qmask, enum nl80211_iftype iftype,
-				    enum iwl_sta_type type);
+				    u8 type);
 void iwl_mvm_dealloc_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 void iwl_mvm_dealloc_int_sta(struct iwl_mvm *mvm, struct iwl_mvm_int_sta *sta);
 int iwl_mvm_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
@@ -551,4 +551,14 @@ int iwl_mvm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 void iwl_mvm_cancel_channel_switch(struct iwl_mvm *mvm,
 				   struct ieee80211_vif *vif,
 				   u32 mac_id);
+/* Queues */
+int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
+			    struct ieee80211_sta *sta,
+			    u8 sta_id, u8 tid, unsigned int timeout);
+
+/* New MLD STA related APIs */
+int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 #endif /* __sta_h__ */
-- 
cgit v1.2.3-70-g09d2


From 1ab26632332eac61cc24464a74fd9bcf5ec5167b Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:21 +0200
Subject: wifi: iwlwifi: mvm: Add an add_interface() callback for mld mode

As the MLD mode and its new APIs are introduced,
we've decided to add a new ieee80211_ops dedicated for
MLD callbacks. Add the MLD add_interface() callback
which uses the new MLD APIs added the previous patches.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.6adc29bff39b.I97ed469028009be9392dcc6f7b5ffbe45f4b2c43@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/Makefile    |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  85 ++++++++++-----
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 118 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   8 ++
 4 files changed, 184 insertions(+), 29 deletions(-)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
index 0e9b5381e265..b82f79ac5378 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
@@ -7,7 +7,7 @@ iwlmvm-y += power.o coex.o
 iwlmvm-y += tt.o offloading.o tdls.o
 iwlmvm-y += ftm-responder.o ftm-initiator.o
 iwlmvm-y += rfi.o
-iwlmvm-y += mld-key.o mld-mac.o link.o mld-sta.o
+iwlmvm-y += mld-key.o mld-mac.o link.o mld-sta.o mld-mac80211.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
 iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM) += d3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index cf08cb834cc4..73b164aad86d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1216,7 +1216,7 @@ static void iwl_mvm_mac_stop(struct ieee80211_hw *hw)
 	cancel_work_sync(&mvm->async_handlers_wk);
 }
 
-static struct iwl_mvm_phy_ctxt *iwl_mvm_get_free_phy_ctxt(struct iwl_mvm *mvm)
+struct iwl_mvm_phy_ctxt *iwl_mvm_get_free_phy_ctxt(struct iwl_mvm *mvm)
 {
 	u16 i;
 
@@ -1350,7 +1350,7 @@ static void iwl_mvm_abort_channel_switch(struct ieee80211_hw *hw,
 	iwl_mvm_post_channel_switch(hw, vif);
 }
 
-static void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk)
+void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk)
 {
 	struct iwl_mvm_vif *mvmvif;
 	struct ieee80211_vif *vif;
@@ -1384,12 +1384,17 @@ iwl_mvm_chandef_get_primary_80(struct cfg80211_chan_def *chandef)
 	return (control_start - data_start) / 80;
 }
 
-static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
-				     struct ieee80211_vif *vif)
+/*
+ * Returns true if addding the interface is done
+ * (either with success or failure)
+ */
+bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
+				      struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif, int *ret)
 {
-	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
 
 	mvmvif->mvm = mvm;
 	RCU_INIT_POINTER(mvmvif->probe_resp_data, NULL);
@@ -1400,17 +1405,15 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 	 * don't really have to check the types.
 	 */
 
-	mutex_lock(&mvm->mutex);
-
 	/* make sure that beacon statistics don't go backwards with FW reset */
 	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
 		mvmvif->beacon_stats.accu_num_beacons +=
 			mvmvif->beacon_stats.num_beacons;
 
 	/* Allocate resources for the MAC context, and add it to the fw  */
-	ret = iwl_mvm_mac_ctxt_init(mvm, vif);
-	if (ret)
-		goto out_unlock;
+	*ret = iwl_mvm_mac_ctxt_init(mvm, vif);
+	if (*ret)
+		return true;
 
 	rcu_assign_pointer(mvm->vif_id_to_mac[mvmvif->id], vif);
 
@@ -1427,27 +1430,48 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 	 */
 	if (vif->type == NL80211_IFTYPE_AP ||
 	    vif->type == NL80211_IFTYPE_ADHOC) {
-		ret = iwl_mvm_alloc_bcast_sta(mvm, vif);
-		if (ret) {
-			IWL_ERR(mvm, "Failed to allocate bcast sta\n");
-			goto out_unlock;
-		}
-
-		/*
-		 * Only queue for this station is the mcast queue,
-		 * which shouldn't be in TFD mask anyway
-		 */
-		ret = iwl_mvm_allocate_int_sta(mvm, &mvmvif->mcast_sta,
-					       0, vif->type,
-					       IWL_STA_MULTICAST);
-		if (ret)
-			goto out_unlock;
-
 		iwl_mvm_vif_dbgfs_register(mvm, vif);
-		goto out_unlock;
+		return true;
 	}
 
 	mvmvif->features |= hw->netdev_features;
+	return false;
+}
+
+static int iwl_mvm_alloc_bcast_mcast_sta(struct iwl_mvm *mvm,
+					 struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	ret = iwl_mvm_alloc_bcast_sta(mvm, vif);
+	if (ret) {
+		IWL_ERR(mvm, "Failed to allocate bcast sta\n");
+		return ret;
+	}
+
+	/*
+	 * Only queue for this station is the mcast queue,
+	 * which shouldn't be in TFD mask anyway
+	 */
+	return iwl_mvm_allocate_int_sta(mvm, &mvmvif->mcast_sta, 0, vif->type,
+					IWL_STA_MULTICAST);
+}
+
+static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	mutex_lock(&mvm->mutex);
+
+	/* Common for MLD and non-MLD API */
+	if (iwl_mvm_mac_add_interface_common(mvm, hw, vif, &ret))
+		goto out;
 
 	ret = iwl_mvm_mac_ctxt_add(mvm, vif);
 	if (ret)
@@ -1516,6 +1540,11 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 		mvm->csme_vif = vif;
 	}
 
+out:
+	if (!ret && (vif->type == NL80211_IFTYPE_AP ||
+		     vif->type == NL80211_IFTYPE_ADHOC))
+		ret = iwl_mvm_alloc_bcast_mcast_sta(mvm, vif);
+
 	goto out_unlock;
 
  out_unbind:
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
new file mode 100644
index 000000000000..b233bdd68b37
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2022 Intel Corporation
+ */
+#include "mvm.h"
+
+static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	mutex_lock(&mvm->mutex);
+
+	/* Common for MLD and non-MLD API */
+	if (iwl_mvm_mac_add_interface_common(mvm, hw, vif, &ret))
+		goto out_unlock;
+
+	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
+	if (ret)
+		goto out_unlock;
+
+	ret = iwl_mvm_power_update_mac(mvm);
+	if (ret)
+		goto out_remove_mac;
+
+	/* beacon filtering */
+	ret = iwl_mvm_disable_beacon_filter(mvm, vif, 0);
+	if (ret)
+		goto out_remove_mac;
+
+	if (!mvm->bf_allowed_vif &&
+	    vif->type == NL80211_IFTYPE_STATION && !vif->p2p) {
+		mvm->bf_allowed_vif = mvmvif;
+		vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER |
+				     IEEE80211_VIF_SUPPORTS_CQM_RSSI;
+	}
+
+	/*
+	 * P2P_DEVICE interface does not have a channel context assigned to it,
+	 * so a dedicated PHY context is allocated to it and the corresponding
+	 * MAC context is bound to it at this stage.
+	 */
+	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
+		mvmvif->phy_ctxt = iwl_mvm_get_free_phy_ctxt(mvm);
+		if (!mvmvif->phy_ctxt) {
+			ret = -ENOSPC;
+			goto out_free_bf;
+		}
+
+		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->phy_ctxt);
+		ret = iwl_mvm_add_link(mvm, vif);
+		if (ret)
+			goto out_unref_phy;
+
+		ret = iwl_mvm_link_changed(mvm, vif,
+					   LINK_CONTEXT_MODIFY_ACTIVE |
+					   LINK_CONTEXT_MODIFY_RATES_INFO,
+					   true);
+		if (ret)
+			goto out_remove_link;
+
+		ret = iwl_mvm_mld_add_bcast_sta(mvm, vif);
+		if (ret)
+			goto out_remove_link;
+
+		/* Save a pointer to p2p device vif, so it can later be used to
+		 * update the p2p device MAC when a GO is started/stopped
+		 */
+		mvm->p2p_device_vif = vif;
+	}
+
+	iwl_mvm_tcm_add_vif(mvm, vif);
+	INIT_DELAYED_WORK(&mvmvif->csa_work,
+			  iwl_mvm_channel_switch_disconnect_wk);
+
+	if (vif->type == NL80211_IFTYPE_MONITOR) {
+		mvm->monitor_on = true;
+		ieee80211_hw_set(mvm->hw, RX_INCLUDES_FCS);
+	}
+
+	iwl_mvm_vif_dbgfs_register(mvm, vif);
+
+	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
+	    vif->type == NL80211_IFTYPE_STATION && !vif->p2p &&
+	    !mvm->csme_vif && mvm->mei_registered) {
+		iwl_mei_set_nic_info(vif->addr, mvm->nvm_data->hw_addr);
+		iwl_mei_set_netdev(ieee80211_vif_to_wdev(vif)->netdev);
+		mvm->csme_vif = vif;
+	}
+
+	goto out_unlock;
+
+ out_remove_link:
+	/* Link needs to be deactivated before removal */
+	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_remove_link(mvm, vif);
+ out_unref_phy:
+	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
+ out_free_bf:
+	if (mvm->bf_allowed_vif == mvmvif) {
+		mvm->bf_allowed_vif = NULL;
+		vif->driver_flags &= ~(IEEE80211_VIF_BEACON_FILTER |
+				       IEEE80211_VIF_SUPPORTS_CQM_RSSI);
+	}
+ out_remove_mac:
+	mvmvif->phy_ctxt = NULL;
+	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
+ out_unlock:
+	mutex_unlock(&mvm->mutex);
+
+	return ret;
+}
+
+const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
+	.add_interface = iwl_mvm_mld_mac_add_interface,
+};
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 4663cb7a0b10..50230a9dbcc8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -72,6 +72,7 @@
 #define IWL_MVM_OFFCHANNEL_QUEUE 0
 
 extern const struct ieee80211_ops iwl_mvm_hw_ops;
+extern const struct ieee80211_ops iwl_mvm_mld_hw_ops;
 
 /**
  * struct iwl_mvm_mod_params - module parameters for iwlmvm
@@ -1624,6 +1625,7 @@ void iwl_mvm_rx_shared_mem_cfg_notif(struct iwl_mvm *mvm,
 				     struct iwl_rx_cmd_buffer *rxb);
 
 /* MVM PHY */
+struct iwl_mvm_phy_ctxt *iwl_mvm_get_free_phy_ctxt(struct iwl_mvm *mvm);
 int iwl_mvm_phy_ctxt_add(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt,
 			 struct cfg80211_chan_def *chandef,
 			 u8 chains_static, u8 chains_dynamic);
@@ -1640,6 +1642,9 @@ u8 iwl_mvm_get_ctrl_pos(struct cfg80211_chan_def *chandef);
 
 /* MAC (virtual interface) programming */
 
+bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
+				      struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif, int *ret);
 void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				__le32 *cck_rates, __le32 *ofdm_rates);
 void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
@@ -2158,6 +2163,9 @@ static inline u8 iwl_mvm_phy_band_from_nl80211(enum nl80211_band band)
 	}
 }
 
+/* Channel Switch */
+void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk);
+
 /* Channel info utils */
 static inline bool iwl_mvm_has_ultra_hb_channel(struct iwl_mvm *mvm)
 {
-- 
cgit v1.2.3-70-g09d2


From 60efeca1c6a75d7f518c963fa4f3694dcfc1147e Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:22 +0200
Subject: wifi: iwlwifi: mvm: Add a remove_interface() callback for mld mode

As the MLD mode and its new APIs are introduced,
we've decided to add a new ieee80211_ops dedicated for
MLD callbacks. Add the MLD remove_interface() callback
which uses the new MLD APIs.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.b87c5c0a4b6b.I631173a73d6ffd7232aa539ea8b356a222fac398@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 32 ++++++++++++++++----
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 34 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  2 ++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       |  4 +--
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  2 ++
 5 files changed, 66 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 73b164aad86d..bdec5ae5e83a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1579,7 +1579,11 @@ static void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
 	}
 }
 
-static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
+/* This function is doing the common part of removing the interface for
+ * both - MLD and non-MLD modes. Returns true if removing the interface
+ * is done
+ */
+bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
 					 struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
@@ -1628,11 +1632,22 @@ static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
 			mvm->noa_duration = 0;
 		}
 #endif
-		iwl_mvm_dealloc_int_sta(mvm, &mvmvif->mcast_sta);
-		iwl_mvm_dealloc_bcast_sta(mvm, vif);
-		goto out_release;
+		return true;
 	}
 
+	iwl_mvm_power_update_mac(mvm);
+	return false;
+}
+
+static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	if (iwl_mvm_mac_remove_interface_common(hw, vif))
+		goto out;
+
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 		mvm->p2p_device_vif = NULL;
 		iwl_mvm_rm_p2p_bcast_sta(mvm, vif);
@@ -1641,7 +1656,6 @@ static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
 		mvmvif->phy_ctxt = NULL;
 	}
 
-	iwl_mvm_power_update_mac(mvm);
 	iwl_mvm_mac_ctxt_remove(mvm, vif);
 
 	RCU_INIT_POINTER(mvm->vif_id_to_mac[mvmvif->id], NULL);
@@ -1649,7 +1663,13 @@ static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		mvm->monitor_on = false;
 
-out_release:
+out:
+	if (vif->type == NL80211_IFTYPE_AP ||
+	    vif->type == NL80211_IFTYPE_ADHOC) {
+		iwl_mvm_dealloc_int_sta(mvm, &mvmvif->mcast_sta);
+		iwl_mvm_dealloc_bcast_sta(mvm, vif);
+	}
+
 	mutex_unlock(&mvm->mutex);
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index b233bdd68b37..8dca72ec55cb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -113,6 +113,40 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	return ret;
 }
 
+static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
+					     struct ieee80211_vif *vif)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	if (iwl_mvm_mac_remove_interface_common(hw, vif))
+		goto out;
+
+	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
+		mvm->p2p_device_vif = NULL;
+		iwl_mvm_mld_rm_bcast_sta(mvm, vif);
+		/* Link needs to be deactivated before removal */
+		iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
+				     false);
+		iwl_mvm_remove_link(mvm, vif);
+		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
+		mvmvif->phy_ctxt = NULL;
+	}
+
+	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
+
+	RCU_INIT_POINTER(mvm->vif_id_to_mac[mvmvif->id], NULL);
+
+	if (vif->type == NL80211_IFTYPE_MONITOR) {
+		mvm->monitor_on = false;
+		__clear_bit(IEEE80211_HW_RX_INCLUDES_FCS, mvm->hw->flags);
+	}
+
+out:
+	mutex_unlock(&mvm->mutex);
+}
+
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
+	.remove_interface = iwl_mvm_mld_mac_remove_interface,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 50230a9dbcc8..32e0af87b528 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1645,6 +1645,8 @@ u8 iwl_mvm_get_ctrl_pos(struct cfg80211_chan_def *chandef);
 bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
 				      struct ieee80211_hw *hw,
 				      struct ieee80211_vif *vif, int *ret);
+bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif);
 void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				__le32 *cck_rates, __le32 *ofdm_rates);
 void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 313c5416ae5e..bbcc7459f3a0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2256,8 +2256,8 @@ int iwl_mvm_send_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	return 0;
 }
 
-static void iwl_mvm_free_bcast_sta_queues(struct iwl_mvm *mvm,
-					  struct ieee80211_vif *vif)
+void iwl_mvm_free_bcast_sta_queues(struct iwl_mvm *mvm,
+				   struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	u16 *queueptr, queue;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index d11851b47684..cad6e879c999 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -510,6 +510,8 @@ int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id);
 int iwl_mvm_rm_aux_sta(struct iwl_mvm *mvm);
 
 int iwl_mvm_alloc_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+void iwl_mvm_free_bcast_sta_queues(struct iwl_mvm *mvm,
+				   struct ieee80211_vif *vif);
 int iwl_mvm_send_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_add_p2p_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_send_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
-- 
cgit v1.2.3-70-g09d2


From 8a919a78a469bb4e01ced9766bdbb7a03c8a4edf Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:23 +0200
Subject: wifi: iwlwifi: mvm: refactor __iwl_mvm_assign_vif_chanctx()

Since parts of the functionality of this function is going to be used
also by the MLD version of it, put in a separate function the parts
that are common for both MLD and non-MLD modes.
The common function will later be used in the MLD ops.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.844755701cac.I1c650718ad2381eabc38f4103c1aac67936a1ffc@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 47 +++++++++++++++++------
 1 file changed, 36 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index bdec5ae5e83a..8123dad3b710 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4405,15 +4405,21 @@ out_unlock:
 	mutex_unlock(&mvm->mutex);
 }
 
-static int __iwl_mvm_assign_vif_chanctx(struct iwl_mvm *mvm,
-					struct ieee80211_vif *vif,
-					struct ieee80211_chanctx_conf *ctx,
-					bool switching_chanctx)
+/*
+ * This function executes the common part for MLD and non-MLD modes.
+ *
+ * Returns true if we're done assigning the chanctx
+ * (either on failure or success)
+ */
+static bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
+						struct ieee80211_vif *vif,
+						struct ieee80211_chanctx_conf *ctx,
+						bool switching_chanctx,
+						int *ret)
 {
 	u16 *phy_ctxt_id = (u16 *)ctx->drv_priv;
 	struct iwl_mvm_phy_ctxt *phy_ctxt = &mvm->phy_ctxts[*phy_ctxt_id];
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
 
@@ -4432,19 +4438,32 @@ static int __iwl_mvm_assign_vif_chanctx(struct iwl_mvm *mvm,
 		 * The AP binding flow is handled as part of the start_ap flow
 		 * (in bss_info_changed), similarly for IBSS.
 		 */
-		ret = 0;
-		goto out;
+		*ret = 0;
+		return true;
 	case NL80211_IFTYPE_STATION:
-		mvmvif->csa_bcn_pending = false;
 		break;
 	case NL80211_IFTYPE_MONITOR:
 		/* always disable PS when a monitor interface is active */
 		mvmvif->ps_disabled = true;
 		break;
 	default:
-		ret = -EINVAL;
-		goto out;
+		*ret = -EINVAL;
+		return true;
 	}
+	return false;
+}
+
+static int __iwl_mvm_assign_vif_chanctx(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					struct ieee80211_chanctx_conf *ctx,
+					bool switching_chanctx)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	if (__iwl_mvm_assign_vif_chanctx_common(mvm, vif, ctx,
+						switching_chanctx, &ret))
+		goto out;
 
 	ret = iwl_mvm_binding_add_vif(mvm, vif);
 	if (ret)
@@ -4478,7 +4497,12 @@ static int __iwl_mvm_assign_vif_chanctx(struct iwl_mvm *mvm,
 		iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
 	}
 
-	if (switching_chanctx && vif->type == NL80211_IFTYPE_STATION) {
+	if (vif->type == NL80211_IFTYPE_STATION) {
+		if (!switching_chanctx) {
+			mvmvif->csa_bcn_pending = false;
+			goto out;
+		}
+
 		mvmvif->csa_bcn_pending = true;
 
 		if (!fw_has_capa(&mvm->fw->ucode_capa,
@@ -4506,6 +4530,7 @@ out:
 		mvmvif->phy_ctxt = NULL;
 	return ret;
 }
+
 static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 				      struct ieee80211_vif *vif,
 				      struct ieee80211_bss_conf *link_conf,
-- 
cgit v1.2.3-70-g09d2


From 50e81437a4831e32b0d04a828aa18c02356ee23e Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:24 +0200
Subject: wifi: iwlwifi: mvm: add an assign_vif_chanctx() callback for MLD mode

This is another patch in the series adding all the ops
for the new MLD ieee80211_ops.
The callback added here uses the new MLD FW API.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.d3753975e720.I45f89cc81370d2cf8d4f51748ccb3ec675eff1bd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  9 ++--
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 60 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  6 +++
 3 files changed, 70 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 8123dad3b710..d4a3d9259150 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4411,11 +4411,10 @@ out_unlock:
  * Returns true if we're done assigning the chanctx
  * (either on failure or success)
  */
-static bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
-						struct ieee80211_vif *vif,
-						struct ieee80211_chanctx_conf *ctx,
-						bool switching_chanctx,
-						int *ret)
+bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
+					 struct ieee80211_vif *vif,
+					 struct ieee80211_chanctx_conf *ctx,
+					 bool switching_chanctx, int *ret)
 {
 	u16 *phy_ctxt_id = (u16 *)ctx->drv_priv;
 	struct iwl_mvm_phy_ctxt *phy_ctxt = &mvm->phy_ctxts[*phy_ctxt_id];
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 8dca72ec55cb..81f88d1b0feb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -146,7 +146,67 @@ out:
 	mutex_unlock(&mvm->mutex);
 }
 
+static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
+					    struct ieee80211_chanctx_conf *ctx,
+					    bool switching_chanctx)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	if (__iwl_mvm_assign_vif_chanctx_common(mvm, vif, ctx,
+						switching_chanctx, &ret))
+		goto out;
+
+	ret = iwl_mvm_add_link(mvm, vif);
+	if (ret)
+		goto out;
+	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
+				   true);
+	if (ret)
+		goto out_remove_link;
+
+	/*
+	 * Power state must be updated before quotas,
+	 * otherwise fw will complain.
+	 */
+	iwl_mvm_power_update_mac(mvm);
+
+	if (vif->type == NL80211_IFTYPE_MONITOR) {
+		ret = iwl_mvm_mld_add_snif_sta(mvm, vif);
+		if (ret)
+			goto out_remove_link;
+	}
+
+	goto out;
+
+out_remove_link:
+	/* Link needs to be deactivated before removal */
+	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_remove_link(mvm, vif);
+	iwl_mvm_power_update_mac(mvm);
+out:
+	if (ret)
+		mvmvif->phy_ctxt = NULL;
+	return ret;
+}
+
+static int iwl_mvm_mld_assign_vif_chanctx(struct ieee80211_hw *hw,
+					  struct ieee80211_vif *vif,
+					  struct ieee80211_bss_conf *link_conf,
+					  struct ieee80211_chanctx_conf *ctx)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	int ret;
+
+	mutex_lock(&mvm->mutex);
+	ret = __iwl_mvm_mld_assign_vif_chanctx(mvm, vif, ctx, false);
+	mutex_unlock(&mvm->mutex);
+
+	return ret;
+}
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
+	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 32e0af87b528..1221a0717e61 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2168,6 +2168,12 @@ static inline u8 iwl_mvm_phy_band_from_nl80211(enum nl80211_band band)
 /* Channel Switch */
 void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk);
 
+/* Channel Context */
+bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
+					 struct ieee80211_vif *vif,
+					 struct ieee80211_chanctx_conf *ctx,
+					 bool switching_chanctx, int *ret);
+
 /* Channel info utils */
 static inline bool iwl_mvm_has_ultra_hb_channel(struct iwl_mvm *mvm)
 {
-- 
cgit v1.2.3-70-g09d2


From daddfae5527164edee09bcca0848c10ab72c5c5b Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:25 +0200
Subject: wifi: iwlwifi: mvm: refactor __iwl_mvm_unassign_vif_chanctx()

Since parts of the functionality of this function is going to be used
also by the MLD version of it, put in a separate function the parts
that are common for both MLD and non-MLD modes.
The common function will later be used in the MLD ops.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.fa05929badb9.I2222dc86cf7d7a7bb58c6a2f2529c8089bfe58b4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 53 +++++++++++++++--------
 1 file changed, 35 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d4a3d9259150..c006d68a785e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4545,52 +4545,69 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 	return ret;
 }
 
-static void __iwl_mvm_unassign_vif_chanctx(struct iwl_mvm *mvm,
-					   struct ieee80211_vif *vif,
-					   struct ieee80211_chanctx_conf *ctx,
-					   bool switching_chanctx)
+/*
+ * This function executes the common part for MLD and non-MLD modes.
+ *
+ * Returns if chanctx unassign chanctx is done
+ * (either on failure or success)
+ */
+static bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
+						  struct ieee80211_vif *vif,
+						  bool switching_chanctx)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct ieee80211_vif *disabled_vif = NULL;
 
 	lockdep_assert_held(&mvm->mutex);
 	iwl_mvm_remove_time_event(mvm, mvmvif, &mvmvif->time_event_data);
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_ADHOC:
-		goto out;
+		return true;
 	case NL80211_IFTYPE_MONITOR:
 		mvmvif->monitor_active = false;
 		mvmvif->ps_disabled = false;
-		iwl_mvm_rm_snif_sta(mvm, vif);
 		break;
 	case NL80211_IFTYPE_AP:
 		/* This part is triggered only during CSA */
 		if (!switching_chanctx || !mvmvif->ap_ibss_active)
-			goto out;
+			return true;
 
 		mvmvif->csa_countdown = false;
 
-		/* Set CS bit on all the stations */
-		iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, true);
-
 		/* Save blocked iface, the timeout is set on the next beacon */
 		rcu_assign_pointer(mvm->csa_tx_blocked_vif, vif);
 
 		mvmvif->ap_ibss_active = false;
 		break;
-	case NL80211_IFTYPE_STATION:
-		if (!switching_chanctx)
-			break;
+	default:
+		break;
+	}
+	return false;
+}
 
-		disabled_vif = vif;
+static void __iwl_mvm_unassign_vif_chanctx(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   struct ieee80211_chanctx_conf *ctx,
+					   bool switching_chanctx)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct ieee80211_vif *disabled_vif = NULL;
+
+	if (__iwl_mvm_unassign_vif_chanctx_common(mvm, vif, switching_chanctx))
+		goto out;
 
+	if (vif->type == NL80211_IFTYPE_MONITOR)
+		iwl_mvm_rm_snif_sta(mvm, vif);
+
+	if (vif->type == NL80211_IFTYPE_AP)
+		/* Set CS bit on all the stations */
+		iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, true);
+
+	if (vif->type == NL80211_IFTYPE_STATION && switching_chanctx) {
+		disabled_vif = vif;
 		if (!fw_has_capa(&mvm->fw->ucode_capa,
 				 IWL_UCODE_TLV_CAPA_CHANNEL_SWITCH_CMD))
 			iwl_mvm_mac_ctxt_changed(mvm, vif, true, NULL);
-		break;
-	default:
-		break;
 	}
 
 	iwl_mvm_update_quotas(mvm, false, disabled_vif);
-- 
cgit v1.2.3-70-g09d2


From 6f71e90e65e9bb5d5e12199e19670bb4973a6e39 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 14 Mar 2023 19:49:26 +0200
Subject: wifi: iwlwifi: mvm: add an unassign_vif_chanctx() callback for MLD
 mode

This is another patch in the series adding all the ops
for the new MLD ieee80211_ops.
The callback added here uses the new MLD FW API

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.5d4bcd384425.I263eef3aad8efe23a597843fe7c56924038c8fdc@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  6 +--
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 41 +++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 58 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  3 ++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  7 +++
 5 files changed, 112 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index c006d68a785e..93ca92e8eb98 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4551,9 +4551,9 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
  * Returns if chanctx unassign chanctx is done
  * (either on failure or success)
  */
-static bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
-						  struct ieee80211_vif *vif,
-						  bool switching_chanctx)
+bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   bool switching_chanctx)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 81f88d1b0feb..f402e235525c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -205,8 +205,49 @@ static int iwl_mvm_mld_assign_vif_chanctx(struct ieee80211_hw *hw,
 
 	return ret;
 }
+
+static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
+					       struct ieee80211_vif *vif,
+					       struct ieee80211_chanctx_conf *ctx,
+					       bool switching_chanctx)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	if (__iwl_mvm_unassign_vif_chanctx_common(mvm, vif, switching_chanctx))
+		goto out;
+
+	if (vif->type == NL80211_IFTYPE_MONITOR)
+		iwl_mvm_mld_rm_snif_sta(mvm, vif);
+
+	if (vif->type == NL80211_IFTYPE_AP)
+		/* Set CS bit on all the stations */
+		iwl_mvm_mld_modify_all_sta_disable_tx(mvm, mvmvif, true);
+
+	/* Link needs to be deactivated before removal */
+	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_remove_link(mvm, vif);
+
+out:
+	if (switching_chanctx)
+		return;
+	mvmvif->phy_ctxt = NULL;
+	iwl_mvm_power_update_mac(mvm);
+}
+
+static void iwl_mvm_mld_unassign_vif_chanctx(struct ieee80211_hw *hw,
+					     struct ieee80211_vif *vif,
+					     struct ieee80211_bss_conf *link_conf,
+					     struct ieee80211_chanctx_conf *ctx)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	mutex_lock(&mvm->mutex);
+	__iwl_mvm_mld_unassign_vif_chanctx(mvm, vif, ctx, false);
+	mutex_unlock(&mvm->mutex);
+}
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
 	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
+	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index cef4fc441afe..4ee6f9b250d3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -279,3 +279,61 @@ int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	return iwl_mvm_mld_rm_int_sta(mvm, &mvm->snif_sta, false,
 				      IWL_MAX_TID_COUNT, &mvm->snif_queue);
 }
+
+static void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
+					      struct ieee80211_sta *sta,
+					      bool disable)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_sta_disable_tx_cmd cmd;
+	int ret;
+
+	spin_lock_bh(&mvm_sta->lock);
+
+	if (mvm_sta->disable_tx == disable) {
+		spin_unlock_bh(&mvm_sta->lock);
+		return;
+	}
+
+	mvm_sta->disable_tx = disable;
+
+	cmd.sta_id = cpu_to_le32(mvm_sta->sta_id);
+	cmd.disable = cpu_to_le32(disable);
+
+	ret = iwl_mvm_send_cmd_pdu(mvm,
+				   WIDE_ID(MAC_CONF_GROUP, STA_DISABLE_TX_CMD),
+				   CMD_ASYNC, sizeof(cmd), &cmd);
+	if (ret)
+		IWL_ERR(mvm,
+			"Failed to send STA_DISABLE_TX_CMD command (%d)\n",
+			ret);
+
+	spin_unlock_bh(&mvm_sta->lock);
+}
+
+void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
+					   struct iwl_mvm_vif *mvmvif,
+					   bool disable)
+{
+	struct ieee80211_sta *sta;
+	struct iwl_mvm_sta *mvm_sta;
+	int i;
+
+	rcu_read_lock();
+
+	/* Block/unblock all the stations of the given mvmvif */
+	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++) {
+		sta = rcu_dereference(mvm->fw_id_to_mac_id[i]);
+		if (IS_ERR_OR_NULL(sta))
+			continue;
+
+		mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+		if (mvm_sta->mac_id_n_color !=
+		    FW_CMD_ID_AND_COLOR(mvmvif->id, mvmvif->color))
+			continue;
+
+		iwl_mvm_mld_sta_modify_disable_tx(mvm, sta, disable);
+	}
+
+	rcu_read_unlock();
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 1221a0717e61..7f73179e765e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2173,6 +2173,9 @@ bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
 					 struct ieee80211_vif *vif,
 					 struct ieee80211_chanctx_conf *ctx,
 					 bool switching_chanctx, int *ret);
+bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   bool switching_chanctx);
 
 /* Channel info utils */
 static inline bool iwl_mvm_has_ultra_hb_channel(struct iwl_mvm *mvm)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index cad6e879c999..f0d22e9a117b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -545,6 +545,7 @@ void iwl_mvm_sta_modify_disable_tx_ap(struct iwl_mvm *mvm,
 void iwl_mvm_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 				       struct iwl_mvm_vif *mvmvif,
 				       bool disable);
+
 void iwl_mvm_csa_client_absent(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 void iwl_mvm_add_new_dqa_stream_wk(struct work_struct *wk);
 int iwl_mvm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
@@ -559,8 +560,14 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 			    u8 sta_id, u8 tid, unsigned int timeout);
 
 /* New MLD STA related APIs */
+/* STA */
 int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+
+/* Queues */
+void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
+					   struct iwl_mvm_vif *mvmvif,
+					   bool disable);
 #endif /* __sta_h__ */
-- 
cgit v1.2.3-70-g09d2


From 093e71e26d67c2b0d194b34a7b56432bab049e90 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Tue, 14 Mar 2023 19:49:27 +0200
Subject: wifi: iwlwifi: mvm: remove setting of 'sta' parameter

cppcheck reports
[drivers/net/wireless/intel/iwlwifi/mvm/rs.c:2686]: (warning) Assignment of function parameter has no effect outside the function. Did you forget dereferencing it?

The setting of the 'sta' parameter is not needed.  In the if-check that sets it
to NULL, mvm_sta is also set to NULL.  Then the next statement checks if
mvm_sta is NULL and does an early return. So remove setting sta.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.711fe28cfdd6.I2f723f9d44f65720baaf3e84b72109759350a8f5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 1f81dff71bc4..7c976b7f1cd7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -2682,7 +2682,6 @@ static void rs_drv_get_rate(void *mvm_r, struct ieee80211_sta *sta,
 		/* if vif isn't initialized mvm doesn't know about
 		 * this station, so don't do anything with the it
 		 */
-		sta = NULL;
 		mvm_sta = NULL;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From e2e76bdcd3fd663bfbddbc09e167e7c1069659cd Mon Sep 17 00:00:00 2001
From: Julia Lawall <Julia.Lawall@inria.fr>
Date: Tue, 14 Mar 2023 19:49:28 +0200
Subject: wifi: iwlwifi: fix typos in comment

Spelling mistakes (triple letters) in comment.
Detected with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.fb2fc470e949.I1f3a347b533bcdb6bcd310f752ab3349800efb49@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/img.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/img.h b/drivers/net/wireless/intel/iwlwifi/fw/img.h
index f878ac508801..f5c4d93d1033 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/img.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/img.h
@@ -182,10 +182,10 @@ struct iwl_dump_exclude {
  * @enhance_sensitivity_table: device can do enhanced sensitivity.
  * @init_evtlog_ptr: event log offset for init ucode.
  * @init_evtlog_size: event log size for init ucode.
- * @init_errlog_ptr: error log offfset for init ucode.
+ * @init_errlog_ptr: error log offset for init ucode.
  * @inst_evtlog_ptr: event log offset for runtime ucode.
  * @inst_evtlog_size: event log size for runtime ucode.
- * @inst_errlog_ptr: error log offfset for runtime ucode.
+ * @inst_errlog_ptr: error log offset for runtime ucode.
  * @type: firmware type (&enum iwl_fw_type)
  * @human_readable: human readable version
  *	we get the ALIVE from the uCode
-- 
cgit v1.2.3-70-g09d2


From d2abe692fe6bea753aaa38c081f322fbed5930e9 Mon Sep 17 00:00:00 2001
From: Solomon Tan <solomonbstoner@protonmail.ch>
Date: Tue, 14 Mar 2023 19:49:29 +0200
Subject: wifi: iwlwifi: Remove prohibited spaces

This patch addresses the error from checkpatch.pl regarding the presence
of prohibited spaces.

Signed-off-by: Solomon Tan <solomonbstoner@protonmail.ch>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.557d6841a166.I0957923fe9ea20c1ec9140477033548ccfe93e9a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c   | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index aa8e08487b52..923bbfc151dd 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -81,7 +81,7 @@ static const u16 iwl_nvm_channels[] = {
 	/* 2.4 GHz */
 	1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
 	/* 5 GHz */
-	36, 40, 44 , 48, 52, 56, 60, 64,
+	36, 40, 44, 48, 52, 56, 60, 64,
 	100, 104, 108, 112, 116, 120, 124, 128, 132, 136, 140, 144,
 	149, 153, 157, 161, 165
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 85b99316d029..5790435dc69e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -214,9 +214,9 @@ static ssize_t iwl_dbgfs_set_nic_temperature_read(struct file *file,
 	int pos;
 
 	if (!mvm->temperature_test)
-		pos = scnprintf(buf , sizeof(buf), "disabled\n");
+		pos = scnprintf(buf, sizeof(buf), "disabled\n");
 	else
-		pos = scnprintf(buf , sizeof(buf), "%d\n", mvm->temperature);
+		pos = scnprintf(buf, sizeof(buf), "%d\n", mvm->temperature);
 
 	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
 }
@@ -261,7 +261,7 @@ static ssize_t iwl_dbgfs_set_nic_temperature_write(struct iwl_mvm *mvm,
 		mvm->temperature = temperature;
 	}
 	IWL_DEBUG_TEMP(mvm, "%sabling debug set temperature (temp = %d)\n",
-		       mvm->temperature_test ? "En" : "Dis" ,
+		       mvm->temperature_test ? "En" : "Dis",
 		       mvm->temperature);
 	/* handle the temperature change */
 	iwl_mvm_tt_handler(mvm);
@@ -291,7 +291,7 @@ static ssize_t iwl_dbgfs_nic_temp_read(struct file *file,
 	if (ret)
 		return -EIO;
 
-	pos = scnprintf(buf , sizeof(buf), "%d\n", temp);
+	pos = scnprintf(buf, sizeof(buf), "%d\n", temp);
 
 	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
 }
-- 
cgit v1.2.3-70-g09d2


From 473bc264356297b9df7ee3183b08b954be4a1152 Mon Sep 17 00:00:00 2001
From: Solomon Tan <solomonbstoner@protonmail.ch>
Date: Tue, 14 Mar 2023 19:49:30 +0200
Subject: wifi: iwlwifi: Add required space before open '('

This patch addresses the error from checkpatch.pl that a space is
required before an open parenthesis.

Signed-off-by: Solomon Tan <solomonbstoner@protonmail.ch>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.a21ba1967d94.Iaa52fc6517ea9efb3efc7b1b98f4df2a288de1c3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 3e9e9f13506b..fe7dfdcd531b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1037,7 +1037,7 @@ int iwl_mvm_ppag_send_cmd(struct iwl_mvm *mvm)
 
 	ret = iwl_read_ppag_table(&mvm->fwrt, &cmd, &cmd_size);
 	/* Not supporting PPAG table is a valid scenario */
-	if(ret < 0)
+	if (ret < 0)
 		return 0;
 
 	IWL_DEBUG_RADIO(mvm, "Sending PER_PLATFORM_ANT_GAIN_CMD\n");
-- 
cgit v1.2.3-70-g09d2


From f311d0113c7326ec411c9f81dbf77a2cc5ed00ff Mon Sep 17 00:00:00 2001
From: Solomon Tan <solomonbstoner@protonmail.ch>
Date: Tue, 14 Mar 2023 19:49:31 +0200
Subject: wifi: iwlwifi: Replace space with tabs as code indent

This patch addresses the checkpatch.pl warning that code indent should
use tabs.

Signed-off-by: Solomon Tan <solomonbstoner@protonmail.ch>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.eaca2a9b08f2.Ic81d60185c13cfc750bf93fbabac57314cb5de13@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 7f73179e765e..7cc3526e78ce 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1312,7 +1312,7 @@ static inline bool iwl_mvm_is_csum_supported(struct iwl_mvm *mvm)
 {
 	return fw_has_capa(&mvm->fw->ucode_capa,
 			   IWL_UCODE_TLV_CAPA_CSUM_SUPPORT) &&
-               !IWL_MVM_HW_CSUM_DISABLE;
+		!IWL_MVM_HW_CSUM_DISABLE;
 }
 
 static inline bool iwl_mvm_is_mplut_supported(struct iwl_mvm *mvm)
-- 
cgit v1.2.3-70-g09d2


From 71a54f7e74488f234fb56270253bcd08a3357d0e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 14 Mar 2023 19:49:32 +0200
Subject: wifi: iwlwifi: mvm: rs: print BAD_RATE for invalid HT/VHT index

If there's a rate->index that maps inside the range but
to an uninitialized value, then that's also a bad rate,
avoid printing "(nil)" in that case and rather print the
"BAD_RATE" string instead as in the else branch.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.06b38d160fc5.I45a9254d3658b1ce796aa4c427193d3cbf638d7e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 7c976b7f1cd7..bd135b7168cb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -512,10 +512,10 @@ static char *rs_pretty_rate(const struct rs_rate *rate)
 		 (rate->index <= IWL_RATE_MCS_9_INDEX))
 		rate_str = ht_vht_rates[rate->index];
 	else
-		rate_str = "BAD_RATE";
+		rate_str = NULL;
 
 	sprintf(buf, "(%s|%s|%s)", rs_pretty_lq_type(rate->type),
-		iwl_rs_pretty_ant(rate->ant), rate_str);
+		iwl_rs_pretty_ant(rate->ant), rate_str ?: "BAD_RATE");
 	return buf;
 }
 
-- 
cgit v1.2.3-70-g09d2


From c2db01752735b522ae150b8698fd4e526efbbc20 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Tue, 14 Mar 2023 19:49:33 +0200
Subject: wifi: iwlwifi: Update configurations for Bnj and Bz devices

Add missing rf support for Bz device and B-Step for BnJ device

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.14c373dcfe15.I96b96a8b149ac181e962f4b82e0d15b172823d0f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  | 36 +++++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  3 +++
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   | 15 +++++++++++
 3 files changed, 54 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 43fb5cf85f05..bc33d19ca08d 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -57,6 +57,8 @@
 #define IWL_BZ_A_MR_A_FW_PRE		"iwlwifi-bz-a0-mr-a0-"
 #define IWL_BZ_A_FM_A_FW_PRE		"iwlwifi-bz-a0-fm-a0-"
 #define IWL_BZ_A_FM4_A_FW_PRE		"iwlwifi-bz-a0-fm4-a0-"
+#define IWL_BZ_A_FM_B_FW_PRE		"iwlwifi-bz-a0-fm-b0-"
+#define IWL_BZ_A_FM4_B_FW_PRE		"iwlwifi-bz-a0-fm4-b0-"
 #define IWL_GL_A_FM_A_FW_PRE		"iwlwifi-gl-a0-fm-a0-"
 #define IWL_GL_B_FM_B_FW_PRE		"iwlwifi-gl-b0-fm-b0-"
 #define IWL_BZ_Z_GF_A_FW_PRE		"iwlwifi-bz-z0-gf-a0-"
@@ -64,6 +66,7 @@
 #define IWL_BNJ_A_FM4_A_FW_PRE		"iwlwifi-BzBnj-a0-fm4-a0-"
 #define IWL_BNJ_B_FM4_B_FW_PRE		"iwlwifi-BzBnj-b0-fm4-b0-"
 #define IWL_BNJ_A_GF_A_FW_PRE		"iwlwifi-BzBnj-a0-gf-a0-"
+#define IWL_BNJ_B_GF_A_FW_PRE		"iwlwifi-BzBnj-b0-gf-a0-"
 #define IWL_BNJ_A_GF4_A_FW_PRE		"iwlwifi-BzBnj-a0-gf4-a0-"
 #define IWL_BNJ_A_HR_B_FW_PRE		"iwlwifi-BzBnj-a0-hr-b0-"
 #define IWL_BNJ_B_FM_B_FW_PRE		"iwlwifi-BzBnj-b0-fm-b0-"
@@ -125,6 +128,10 @@
 		IWL_BZ_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_FM4_A_MODULE_FIRMWARE(api) \
 		IWL_BZ_A_FM4_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_BZ_A_FM_B_MODULE_FIRMWARE(api) \
+		IWL_BZ_A_FM_B_FW_PRE __stringify(api) ".ucode"
+#define IWL_BZ_A_FM4_B_MODULE_FIRMWARE(api) \
+		IWL_BZ_A_FM4_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_GL_A_FM_A_MODULE_FIRMWARE(api) \
 		IWL_GL_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_GL_B_FM_B_MODULE_FIRMWARE(api) \
@@ -137,6 +144,8 @@
 	IWL_BNJ_B_FM4_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_GF_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_GF_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_BNJ_B_GF_A_MODULE_FIRMWARE(api) \
+	IWL_BNJ_B_GF_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_GF4_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_HR_B_MODULE_FIRMWARE(api) \
@@ -961,6 +970,22 @@ const struct iwl_cfg iwl_cfg_bz_a0_fm4_a0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bz_a0_fm_b0 = {
+	.fw_name_pre = IWL_BZ_A_FM_B_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS_BZ | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
+const struct iwl_cfg iwl_cfg_bz_a0_fm4_b0 = {
+	.fw_name_pre = IWL_BZ_A_FM4_B_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS_BZ | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_gl_a0_fm_a0 = {
 	.fw_name_pre = IWL_GL_A_FM_A_FW_PRE,
 	.uhb_supported = true,
@@ -1017,6 +1042,14 @@ const struct iwl_cfg iwl_cfg_bnj_a0_gf_a0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bnj_b0_gf_a0 = {
+	.fw_name_pre = IWL_BNJ_B_GF_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bnj_a0_gf4_a0 = {
 	.fw_name_pre = IWL_BNJ_A_GF4_A_FW_PRE,
 	.uhb_supported = true,
@@ -1067,13 +1100,16 @@ MODULE_FIRMWARE(IWL_BZ_A_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_MR_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_FM_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BZ_A_FM_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_GL_A_FM_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_FM_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_FM4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_B_FM4_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BNJ_B_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_FM4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BZ_A_FM4_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_GL_B_FM_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_B_FM_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index eaa0ff2736c5..a4c7fd929cc2 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -650,12 +650,15 @@ extern const struct iwl_cfg iwl_cfg_bz_a0_gf4_a0;
 extern const struct iwl_cfg iwl_cfg_bz_a0_mr_a0;
 extern const struct iwl_cfg iwl_cfg_bz_a0_fm_a0;
 extern const struct iwl_cfg iwl_cfg_bz_a0_fm4_a0;
+extern const struct iwl_cfg iwl_cfg_bz_a0_fm_b0;
+extern const struct iwl_cfg iwl_cfg_bz_a0_fm4_b0;
 extern const struct iwl_cfg iwl_cfg_gl_a0_fm_a0;
 extern const struct iwl_cfg iwl_cfg_gl_b0_fm_b0;
 extern const struct iwl_cfg iwl_cfg_bz_z0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_fm_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_fm4_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_gf_a0;
+extern const struct iwl_cfg iwl_cfg_bnj_b0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_gf4_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_fm_b0;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 8aa8a678475c..077983913acd 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1159,6 +1159,16 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_NO_JACKET,
 		      iwl_cfg_bz_a0_fm4_a0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
+		      iwl_cfg_bz_a0_fm_b0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
+		      iwl_cfg_bz_a0_fm4_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
@@ -1203,6 +1213,11 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_gf_a0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
+		      iwl_cfg_bnj_b0_gf_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
-- 
cgit v1.2.3-70-g09d2


From 8f55564c05123b790f25562fa02ca2e1dbf9c4f5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 14 Mar 2023 19:49:34 +0200
Subject: wifi: iwlwifi: fw: pnvm: fix uefi reduced TX power loading

There are a number of issues here:
 * if trans->reduce_power_loaded is already true, we call
   iwl_trans_set_reduce_power() with an uninitialized len
   value
 * in this case we also clobber a previous load/setting
 * if iwl_uefi_get_reduced_power() returns an ERR_PTR() we
   try to kfree() it

Move the iwl_trans_set_reduce_power() call into the success
case only to fix these issues.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230314194113.aa2cf2281f5d.I33b4ab3427f1921c184c52fecd0f46781a89dc8a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/pnvm.c | 20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/pnvm.c b/drivers/net/wireless/intel/iwlwifi/fw/pnvm.c
index b6d3ac6ed440..c6f2672fdc73 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/pnvm.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/pnvm.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright(c) 2020-2021 Intel Corporation
+ * Copyright(c) 2020-2022 Intel Corporation
  */
 
 #include "iwl-drv.h"
@@ -318,7 +318,6 @@ parse:
 	kfree(data);
 
 skip_parse:
-	data = NULL;
 	/* now try to get the reduce power table, if not loaded yet */
 	if (!trans->reduce_power_loaded) {
 		data = iwl_uefi_get_reduced_power(trans, &len);
@@ -329,19 +328,16 @@ skip_parse:
 			 * trying again over and over.
 			 */
 			trans->reduce_power_loaded = true;
-
-			goto skip_reduce_power;
+		} else {
+			ret = iwl_trans_set_reduce_power(trans, data, len);
+			if (ret)
+				IWL_DEBUG_FW(trans,
+					     "Failed to set reduce power table %d\n",
+					     ret);
+			kfree(data);
 		}
 	}
 
-	ret = iwl_trans_set_reduce_power(trans, data, len);
-	if (ret)
-		IWL_DEBUG_FW(trans,
-			     "Failed to set reduce power table %d\n",
-			     ret);
-	kfree(data);
-
-skip_reduce_power:
 	iwl_init_notification_wait(notif_wait, &pnvm_wait,
 				   ntf_cmds, ARRAY_SIZE(ntf_cmds),
 				   iwl_pnvm_complete_fn, trans);
-- 
cgit v1.2.3-70-g09d2


From 4eca8cbf7ba83c3291b5841905ce64584036b1ff Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 Mar 2023 13:29:20 +0100
Subject: wifi: iwlwifi: suppress printf warnings in tracing

This can't really be fixed due to the macro layout of
tracepoints (you'd need a special tracepoint macro for
when this is needed), so just suppress the warnings.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c
index 999b7c652289..c190ec5effa1 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c
@@ -12,6 +12,7 @@
 #include "iwl-trans.h"
 
 #define CREATE_TRACE_POINTS
+#pragma GCC diagnostic ignored "-Wsuggest-attribute=format"
 #include "iwl-devtrace.h"
 
 EXPORT_TRACEPOINT_SYMBOL(iwlwifi_dev_ucode_event);
-- 
cgit v1.2.3-70-g09d2


From 31bf1dbccfb0a9861d4846755096b3fff5687f8a Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Fri, 10 Mar 2023 08:40:59 +0100
Subject: bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules

This resolves two problems with attachment of fentry/fexit/fmod_ret/lsm
to functions located in modules:

1. The verifier tries to find the address to attach to in kallsyms. This
   is always done by searching the entire kallsyms, not respecting the
   module in which the function is located. Such approach causes an
   incorrect attachment address to be computed if the function to attach
   to is shadowed by a function of the same name located earlier in
   kallsyms.

2. If the address to attach to is located in a module, the module
   reference is only acquired in register_fentry. If the module is
   unloaded between the place where the address is found
   (bpf_check_attach_target in the verifier) and register_fentry, it is
   possible that another module is loaded to the same address which may
   lead to potential errors.

Since the attachment must contain the BTF of the program to attach to,
we extract the module from it and search for the function address in the
correct module (resolving problem no. 1). Then, the module reference is
taken directly in bpf_check_attach_target and stored in the bpf program
(in bpf_prog_aux). The reference is only released when the program is
unloaded (resolving problem no. 2).

Signed-off-by: Viktor Malik <vmalik@redhat.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Link: https://lore.kernel.org/r/3f6a9d8ae850532b5ef864ef16327b0f7a669063.1678432753.git.vmalik@redhat.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h      |  2 ++
 kernel/bpf/syscall.c     |  6 ++++++
 kernel/bpf/trampoline.c  | 28 ----------------------------
 kernel/bpf/verifier.c    | 18 +++++++++++++++++-
 kernel/module/internal.h |  5 +++++
 5 files changed, 30 insertions(+), 29 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 71cc92a4ba48..3ef98fb92987 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1103,6 +1103,7 @@ struct bpf_trampoline {
 struct bpf_attach_target_info {
 	struct btf_func_model fmodel;
 	long tgt_addr;
+	struct module *tgt_mod;
 	const char *tgt_name;
 	const struct btf_type *tgt_type;
 };
@@ -1406,6 +1407,7 @@ struct bpf_prog_aux {
 	 * main prog always has linfo_idx == 0
 	 */
 	u32 linfo_idx;
+	struct module *mod;
 	u32 num_exentries;
 	struct exception_table_entry *extable;
 	union {
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 5b88301a2ae0..099e9068bcdd 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2067,6 +2067,7 @@ static void __bpf_prog_put_noref(struct bpf_prog *prog, bool deferred)
 {
 	bpf_prog_kallsyms_del_all(prog);
 	btf_put(prog->aux->btf);
+	module_put(prog->aux->mod);
 	kvfree(prog->aux->jited_linfo);
 	kvfree(prog->aux->linfo);
 	kfree(prog->aux->kfunc_tab);
@@ -3113,6 +3114,11 @@ static int bpf_tracing_prog_attach(struct bpf_prog *prog,
 		if (err)
 			goto out_unlock;
 
+		if (tgt_info.tgt_mod) {
+			module_put(prog->aux->mod);
+			prog->aux->mod = tgt_info.tgt_mod;
+		}
+
 		tr = bpf_trampoline_get(key, &tgt_info);
 		if (!tr) {
 			err = -ENOMEM;
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index d0ed7d6f5eec..f61d5138b12b 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -9,7 +9,6 @@
 #include <linux/btf.h>
 #include <linux/rcupdate_trace.h>
 #include <linux/rcupdate_wait.h>
-#include <linux/module.h>
 #include <linux/static_call.h>
 #include <linux/bpf_verifier.h>
 #include <linux/bpf_lsm.h>
@@ -172,26 +171,6 @@ out:
 	return tr;
 }
 
-static int bpf_trampoline_module_get(struct bpf_trampoline *tr)
-{
-	struct module *mod;
-	int err = 0;
-
-	preempt_disable();
-	mod = __module_text_address((unsigned long) tr->func.addr);
-	if (mod && !try_module_get(mod))
-		err = -ENOENT;
-	preempt_enable();
-	tr->mod = mod;
-	return err;
-}
-
-static void bpf_trampoline_module_put(struct bpf_trampoline *tr)
-{
-	module_put(tr->mod);
-	tr->mod = NULL;
-}
-
 static int unregister_fentry(struct bpf_trampoline *tr, void *old_addr)
 {
 	void *ip = tr->func.addr;
@@ -202,8 +181,6 @@ static int unregister_fentry(struct bpf_trampoline *tr, void *old_addr)
 	else
 		ret = bpf_arch_text_poke(ip, BPF_MOD_CALL, old_addr, NULL);
 
-	if (!ret)
-		bpf_trampoline_module_put(tr);
 	return ret;
 }
 
@@ -238,9 +215,6 @@ static int register_fentry(struct bpf_trampoline *tr, void *new_addr)
 		tr->func.ftrace_managed = true;
 	}
 
-	if (bpf_trampoline_module_get(tr))
-		return -ENOENT;
-
 	if (tr->func.ftrace_managed) {
 		ftrace_set_filter_ip(tr->fops, (unsigned long)ip, 0, 1);
 		ret = register_ftrace_direct_multi(tr->fops, (long)new_addr);
@@ -248,8 +222,6 @@ static int register_fentry(struct bpf_trampoline *tr, void *new_addr)
 		ret = bpf_arch_text_poke(ip, BPF_MOD_CALL, NULL, new_addr);
 	}
 
-	if (ret)
-		bpf_trampoline_module_put(tr);
 	return ret;
 }
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 2bbd89279070..60793f793ca6 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -24,6 +24,7 @@
 #include <linux/bpf_lsm.h>
 #include <linux/btf_ids.h>
 #include <linux/poison.h>
+#include "../module/internal.h"
 
 #include "disasm.h"
 
@@ -18307,6 +18308,7 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 	const char *tname;
 	struct btf *btf;
 	long addr = 0;
+	struct module *mod = NULL;
 
 	if (!btf_id) {
 		bpf_log(log, "Tracing programs must provide btf_id\n");
@@ -18480,8 +18482,17 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 			else
 				addr = (long) tgt_prog->aux->func[subprog]->bpf_func;
 		} else {
-			addr = kallsyms_lookup_name(tname);
+			if (btf_is_module(btf)) {
+				mod = btf_try_get_module(btf);
+				if (mod)
+					addr = find_kallsyms_symbol_value(mod, tname);
+				else
+					addr = 0;
+			} else {
+				addr = kallsyms_lookup_name(tname);
+			}
 			if (!addr) {
+				module_put(mod);
 				bpf_log(log,
 					"The address of function %s cannot be found\n",
 					tname);
@@ -18521,11 +18532,13 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 				break;
 			}
 			if (ret) {
+				module_put(mod);
 				bpf_log(log, "%s is not sleepable\n", tname);
 				return ret;
 			}
 		} else if (prog->expected_attach_type == BPF_MODIFY_RETURN) {
 			if (tgt_prog) {
+				module_put(mod);
 				bpf_log(log, "can't modify return codes of BPF programs\n");
 				return -EINVAL;
 			}
@@ -18534,6 +18547,7 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 			    !check_attach_modify_return(addr, tname))
 				ret = 0;
 			if (ret) {
+				module_put(mod);
 				bpf_log(log, "%s() is not modifiable\n", tname);
 				return ret;
 			}
@@ -18544,6 +18558,7 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 	tgt_info->tgt_addr = addr;
 	tgt_info->tgt_name = tname;
 	tgt_info->tgt_type = t;
+	tgt_info->tgt_mod = mod;
 	return 0;
 }
 
@@ -18623,6 +18638,7 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 	/* store info about the attachment target that will be used later */
 	prog->aux->attach_func_proto = tgt_info.tgt_type;
 	prog->aux->attach_func_name = tgt_info.tgt_name;
+	prog->aux->mod = tgt_info.tgt_mod;
 
 	if (tgt_prog) {
 		prog->aux->saved_dst_prog_type = tgt_prog->type;
diff --git a/kernel/module/internal.h b/kernel/module/internal.h
index 2e2bf236f558..5c9170f9135c 100644
--- a/kernel/module/internal.h
+++ b/kernel/module/internal.h
@@ -256,6 +256,11 @@ static inline bool sect_empty(const Elf_Shdr *sect)
 static inline void init_build_id(struct module *mod, const struct load_info *info) { }
 static inline void layout_symtab(struct module *mod, struct load_info *info) { }
 static inline void add_kallsyms(struct module *mod, const struct load_info *info) { }
+static inline unsigned long find_kallsyms_symbol_value(struct module *mod,
+						       const char *name)
+{
+	return 0;
+}
 #endif /* CONFIG_KALLSYMS */
 
 #ifdef CONFIG_SYSFS
-- 
cgit v1.2.3-70-g09d2


From aa3d65de4b9004d799f97700751a86d3ebd7d5f9 Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Fri, 10 Mar 2023 08:41:00 +0100
Subject: bpf/selftests: Test fentry attachment to shadowed functions

Adds a new test that tries to attach a program to fentry of two
functions of the same name, one located in vmlinux and the other in
bpf_testmod.

To avoid conflicts with existing tests, a new function
"bpf_fentry_shadow_test" was created both in vmlinux and in bpf_testmod.

The previous commit fixed a bug which caused this test to fail. The
verifier would always use the vmlinux function's address as the target
trampoline address, hence trying to create two trampolines for a single
address, which is forbidden.

The test (similarly to other fentry/fexit tests) is not working on arm64
at the moment.

Signed-off-by: Viktor Malik <vmalik@redhat.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/5fe2f364190b6f79b085066ed7c5989c5bc475fa.1678432753.git.vmalik@redhat.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/bpf/test_run.c                                 |   5 +
 tools/testing/selftests/bpf/DENYLIST.aarch64       |   1 +
 .../selftests/bpf/bpf_testmod/bpf_testmod.c        |   6 +
 .../bpf/prog_tests/module_fentry_shadow.c          | 128 +++++++++++++++++++++
 4 files changed, 140 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 6a8b33a103a4..71226f68270d 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -560,6 +560,11 @@ long noinline bpf_kfunc_call_test4(signed char a, short b, int c, long d)
 	return (long)a + (long)b + (long)c + d;
 }
 
+int noinline bpf_fentry_shadow_test(int a)
+{
+	return a + 1;
+}
+
 struct prog_test_member1 {
 	int a;
 };
diff --git a/tools/testing/selftests/bpf/DENYLIST.aarch64 b/tools/testing/selftests/bpf/DENYLIST.aarch64
index 99cc33c51eaa..0a6837f97c32 100644
--- a/tools/testing/selftests/bpf/DENYLIST.aarch64
+++ b/tools/testing/selftests/bpf/DENYLIST.aarch64
@@ -44,6 +44,7 @@ lookup_key                                       # test_lookup_key__attach unexp
 lru_bug                                          # lru_bug__attach unexpected error: -524 (errno 524)
 modify_return                                    # modify_return__attach failed unexpected error: -524 (errno 524)
 module_attach                                    # skel_attach skeleton attach failed: -524
+module_fentry_shadow                             # bpf_link_create unexpected bpf_link_create: actual -524 < expected 0
 mptcp/base                                       # run_test mptcp unexpected error: -524 (errno 524)
 netcnt                                           # packets unexpected packets: actual 10001 != expected 10000
 rcu_read_lock                                    # failed to attach: ERROR: strerror_r(-524)=22
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 5e6e85c8d77d..7999476b9446 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -268,6 +268,12 @@ static const struct btf_kfunc_id_set bpf_testmod_kfunc_set = {
 	.set   = &bpf_testmod_check_kfunc_ids,
 };
 
+noinline int bpf_fentry_shadow_test(int a)
+{
+	return a + 2;
+}
+EXPORT_SYMBOL_GPL(bpf_fentry_shadow_test);
+
 extern int bpf_fentry_test1(int a);
 
 static int bpf_testmod_init(void)
diff --git a/tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c b/tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c
new file mode 100644
index 000000000000..c7636e18b1eb
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c
@@ -0,0 +1,128 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2022 Red Hat */
+#include <test_progs.h>
+#include <bpf/btf.h>
+#include "bpf/libbpf_internal.h"
+#include "cgroup_helpers.h"
+
+static const char *module_name = "bpf_testmod";
+static const char *symbol_name = "bpf_fentry_shadow_test";
+
+static int get_bpf_testmod_btf_fd(void)
+{
+	struct bpf_btf_info info;
+	char name[64];
+	__u32 id = 0, len;
+	int err, fd;
+
+	while (true) {
+		err = bpf_btf_get_next_id(id, &id);
+		if (err) {
+			log_err("failed to iterate BTF objects");
+			return err;
+		}
+
+		fd = bpf_btf_get_fd_by_id(id);
+		if (fd < 0) {
+			if (errno == ENOENT)
+				continue; /* expected race: BTF was unloaded */
+			err = -errno;
+			log_err("failed to get FD for BTF object #%d", id);
+			return err;
+		}
+
+		len = sizeof(info);
+		memset(&info, 0, sizeof(info));
+		info.name = ptr_to_u64(name);
+		info.name_len = sizeof(name);
+
+		err = bpf_obj_get_info_by_fd(fd, &info, &len);
+		if (err) {
+			err = -errno;
+			log_err("failed to get info for BTF object #%d", id);
+			close(fd);
+			return err;
+		}
+
+		if (strcmp(name, module_name) == 0)
+			return fd;
+
+		close(fd);
+	}
+	return -ENOENT;
+}
+
+void test_module_fentry_shadow(void)
+{
+	struct btf *vmlinux_btf = NULL, *mod_btf = NULL;
+	int err, i;
+	int btf_fd[2] = {};
+	int prog_fd[2] = {};
+	int link_fd[2] = {};
+	__s32 btf_id[2] = {};
+
+	LIBBPF_OPTS(bpf_prog_load_opts, load_opts,
+		.expected_attach_type = BPF_TRACE_FENTRY,
+	);
+
+	const struct bpf_insn trace_program[] = {
+		BPF_MOV64_IMM(BPF_REG_0, 0),
+		BPF_EXIT_INSN(),
+	};
+
+	vmlinux_btf = btf__load_vmlinux_btf();
+	if (!ASSERT_OK_PTR(vmlinux_btf, "load_vmlinux_btf"))
+		return;
+
+	btf_fd[1] = get_bpf_testmod_btf_fd();
+	if (!ASSERT_GE(btf_fd[1], 0, "get_bpf_testmod_btf_fd"))
+		goto out;
+
+	mod_btf = btf_get_from_fd(btf_fd[1], vmlinux_btf);
+	if (!ASSERT_OK_PTR(mod_btf, "btf_get_from_fd"))
+		goto out;
+
+	btf_id[0] = btf__find_by_name_kind(vmlinux_btf, symbol_name, BTF_KIND_FUNC);
+	if (!ASSERT_GT(btf_id[0], 0, "btf_find_by_name"))
+		goto out;
+
+	btf_id[1] = btf__find_by_name_kind(mod_btf, symbol_name, BTF_KIND_FUNC);
+	if (!ASSERT_GT(btf_id[1], 0, "btf_find_by_name"))
+		goto out;
+
+	for (i = 0; i < 2; i++) {
+		load_opts.attach_btf_id = btf_id[i];
+		load_opts.attach_btf_obj_fd = btf_fd[i];
+		prog_fd[i] = bpf_prog_load(BPF_PROG_TYPE_TRACING, NULL, "GPL",
+					   trace_program,
+					   sizeof(trace_program) / sizeof(struct bpf_insn),
+					   &load_opts);
+		if (!ASSERT_GE(prog_fd[i], 0, "bpf_prog_load"))
+			goto out;
+
+		/* If the verifier incorrectly resolves addresses of the
+		 * shadowed functions and uses the same address for both the
+		 * vmlinux and the bpf_testmod functions, this will fail on
+		 * attempting to create two trampolines for the same address,
+		 * which is forbidden.
+		 */
+		link_fd[i] = bpf_link_create(prog_fd[i], 0, BPF_TRACE_FENTRY, NULL);
+		if (!ASSERT_GE(link_fd[i], 0, "bpf_link_create"))
+			goto out;
+	}
+
+	err = bpf_prog_test_run_opts(prog_fd[0], NULL);
+	ASSERT_OK(err, "running test");
+
+out:
+	btf__free(vmlinux_btf);
+	btf__free(mod_btf);
+	for (i = 0; i < 2; i++) {
+		if (btf_fd[i])
+			close(btf_fd[i]);
+		if (prog_fd[i] > 0)
+			close(prog_fd[i]);
+		if (link_fd[i] > 0)
+			close(link_fd[i]);
+	}
+}
-- 
cgit v1.2.3-70-g09d2


From 543c143dac5d64a8de8db3ccadea6d92c34fd1e6 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sun, 12 Mar 2023 14:26:37 +0100
Subject: ptp: ines: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it might not be relevant here).  This
also fixes !CONFIG_OF error:

  drivers/ptp/ptp_ines.c:783:34: error: ‘ines_ptp_ctrl_of_match’ defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Link: https://lore.kernel.org/r/20230312132637.352755-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/ptp/ptp_ines.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ptp/ptp_ines.c b/drivers/ptp/ptp_ines.c
index 61f47fb9d997..ed215b458183 100644
--- a/drivers/ptp/ptp_ines.c
+++ b/drivers/ptp/ptp_ines.c
@@ -792,7 +792,7 @@ static struct platform_driver ines_ptp_ctrl_driver = {
 	.remove = ines_ptp_ctrl_remove,
 	.driver = {
 		.name = "ines_ptp_ctrl",
-		.of_match_table = of_match_ptr(ines_ptp_ctrl_of_match),
+		.of_match_table = ines_ptp_ctrl_of_match,
 	},
 };
 module_platform_driver(ines_ptp_ctrl_driver);
-- 
cgit v1.2.3-70-g09d2


From 83456576a42050cc0675aece14277a3cbdee9cbe Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 14 Mar 2023 13:48:56 +0100
Subject: net: phy: update obsolete comment about PHY_STARTING

Commit 899a3cbbf77a ("net: phy: remove states PHY_STARTING and
PHY_PENDING") missed to update a comment in phy_probe. Remove
superfluous "Description:" prefix while we are here.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/20230314124856.44878-1-wsa+renesas@sang-engineering.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phy_device.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 1785f1cead97..c0760cbf534b 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -3076,9 +3076,7 @@ EXPORT_SYMBOL_GPL(fwnode_get_phy_node);
  * phy_probe - probe and init a PHY device
  * @dev: device to probe and init
  *
- * Description: Take care of setting up the phy_device structure,
- *   set the state to READY (the driver's init function should
- *   set it to STARTING if needed).
+ * Take care of setting up the phy_device structure, set the state to READY.
  */
 static int phy_probe(struct device *dev)
 {
-- 
cgit v1.2.3-70-g09d2


From a57cc54d69d601282c4451010b12f53630834203 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 14 Mar 2023 13:49:27 +0100
Subject: net: phy: micrel: drop superfluous use of temp variable

'temp' was used before commit c0c99d0cd107 ("net: phy: micrel: remove
the use of .ack_interrupt()") refactored the code. Now, we can simplify
it a little.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20230314124928.44948-1-wsa+renesas@sang-engineering.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/micrel.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index e5b2af69ac03..0cf0ef3a96a3 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -436,11 +436,9 @@ static int kszphy_config_intr(struct phy_device *phydev)
 		if (err)
 			return err;
 
-		temp = KSZPHY_INTCS_ALL;
-		err = phy_write(phydev, MII_KSZPHY_INTCS, temp);
+		err = phy_write(phydev, MII_KSZPHY_INTCS, KSZPHY_INTCS_ALL);
 	} else {
-		temp = 0;
-		err = phy_write(phydev, MII_KSZPHY_INTCS, temp);
+		err = phy_write(phydev, MII_KSZPHY_INTCS, 0);
 		if (err)
 			return err;
 
-- 
cgit v1.2.3-70-g09d2


From c05d145abea138dbba213ce775820af73e893d92 Mon Sep 17 00:00:00 2001
From: Moshe Shemesh <moshe@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:20 -0700
Subject: net/mlx5: remove redundant clear_bit

When shutdown or remove callbacks are called the driver sets the flag
MLX5_BREAK_FW_WAIT, to stop waiting for FW as teardown was called. There
is no need to clear the bit as once shutdown or remove were called as
there is no way back, the driver is going down. Furthermore, if not
cleared the flag can be used also in other loops where we may wait while
teardown was already called.

Use test_bit() instead of test_and_clear_bit() as there is no need to
clear the flag.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-2-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 540840e80493..0ff0eb660495 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -191,7 +191,7 @@ static int wait_fw_init(struct mlx5_core_dev *dev, u32 max_wait_mili,
 		if (!(fw_initializing >> 31))
 			break;
 		if (time_after(jiffies, end) ||
-		    test_and_clear_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state)) {
+		    test_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state)) {
 			err = -EBUSY;
 			break;
 		}
-- 
cgit v1.2.3-70-g09d2


From 8ff38e730c3f5ee717f25365ef8aa4739562d567 Mon Sep 17 00:00:00 2001
From: Moshe Shemesh <moshe@mellanox.com>
Date: Mon, 13 Mar 2023 22:42:21 -0700
Subject: net/mlx5: Stop waiting for PCI up if teardown was triggered

If driver teardown is called while PCI is turned off, there is a race
between health recovery and teardown. If health recovery already started
it will wait 60 sec trying to see if PCI gets back and it can recover,
but actually there is no need to wait anymore once teardown was called.

Use the MLX5_BREAK_FW_WAIT flag which is set on driver teardown to break
waiting for PCI up.

Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-3-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/health.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index f9438d4e43ca..016c5f99c470 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -325,6 +325,10 @@ int mlx5_health_wait_pci_up(struct mlx5_core_dev *dev)
 	while (sensor_pci_not_working(dev)) {
 		if (time_after(jiffies, end))
 			return -ETIMEDOUT;
+		if (test_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state)) {
+			mlx5_core_warn(dev, "device is being removed, stop waiting for PCI\n");
+			return -ENODEV;
+		}
 		msleep(100);
 	}
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From ceefcfb8a37510f2c2b99b9e8d6c0b3941687b2b Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:22 -0700
Subject: net/mlx5: Add comment to mlx5_devlink_params_register()

Add comment to mlx5_devlink_params_register() functions so it is clear
that only driver init params should be registered here.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-4-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/devlink.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index c5d2fdcabd56..b7784e02c2dd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -805,6 +805,11 @@ int mlx5_devlink_params_register(struct devlink *devlink)
 {
 	int err;
 
+	/* Here only the driver init params should be registered.
+	 * Runtime params should be registered by the code which
+	 * behaviour they configure.
+	 */
+
 	err = devl_params_register(devlink, mlx5_devlink_params,
 				   ARRAY_SIZE(mlx5_devlink_params));
 	if (err)
-- 
cgit v1.2.3-70-g09d2


From c1fef618d611b31964ab397aa0bf0611da94bade Mon Sep 17 00:00:00 2001
From: Sandipan Patra <spatra@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:23 -0700
Subject: net/mlx5: Implement thermal zone

Implement thermal zone support for mlx5 based HW. The NIC
uses temperature sensor provided by ASIC to report current temperature
to thermal core.

Signed-off-by: Sandipan Patra <spatra@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-5-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile  |   1 +
 drivers/net/ethernet/mellanox/mlx5/core/main.c    |   6 ++
 drivers/net/ethernet/mellanox/mlx5/core/thermal.c | 108 ++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlx5/core/thermal.h |  20 ++++
 include/linux/mlx5/driver.h                       |   3 +
 include/linux/mlx5/mlx5_ifc.h                     |  26 ++++++
 6 files changed, 164 insertions(+)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/thermal.c
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/thermal.h

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 8d4e25cc54ea..6c2f1d4a58ab 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -77,6 +77,7 @@ mlx5_core-$(CONFIG_MLX5_ESWITCH)   += esw/acl/helper.o \
 
 mlx5_core-$(CONFIG_MLX5_BRIDGE)    += esw/bridge.o en/rep/bridge.o
 
+mlx5_core-$(CONFIG_THERMAL)        += thermal.o
 mlx5_core-$(CONFIG_MLX5_MPFS)      += lib/mpfs.o
 mlx5_core-$(CONFIG_VXLAN)          += lib/vxlan.o
 mlx5_core-$(CONFIG_PTP_1588_CLOCK) += lib/clock.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 0ff0eb660495..644c889f9a32 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -52,6 +52,7 @@
 #include <linux/version.h>
 #include <net/devlink.h>
 #include "mlx5_core.h"
+#include "thermal.h"
 #include "lib/eq.h"
 #include "fs_core.h"
 #include "lib/mpfs.h"
@@ -1768,6 +1769,10 @@ static int probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		dev_err(&pdev->dev, "mlx5_crdump_enable failed with error code %d\n", err);
 
+	err = mlx5_thermal_init(dev);
+	if (err)
+		dev_err(&pdev->dev, "mlx5_thermal_init failed with error code %d\n", err);
+
 	pci_save_state(pdev);
 	devlink_register(devlink);
 	return 0;
@@ -1796,6 +1801,7 @@ static void remove_one(struct pci_dev *pdev)
 	set_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state);
 	devlink_unregister(devlink);
 	mlx5_sriov_disable(pdev);
+	mlx5_thermal_uninit(dev);
 	mlx5_crdump_disable(dev);
 	mlx5_drain_health_wq(dev);
 	mlx5_uninit_one(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/thermal.c b/drivers/net/ethernet/mellanox/mlx5/core/thermal.c
new file mode 100644
index 000000000000..e47fa6fb836f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/thermal.c
@@ -0,0 +1,108 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+// Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES.
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/thermal.h>
+#include <linux/err.h>
+#include <linux/mlx5/driver.h>
+#include "mlx5_core.h"
+#include "thermal.h"
+
+#define MLX5_THERMAL_POLL_INT_MSEC	1000
+#define MLX5_THERMAL_NUM_TRIPS		0
+#define MLX5_THERMAL_ASIC_SENSOR_INDEX	0
+
+/* Bit string indicating the writeablility of trip points if any */
+#define MLX5_THERMAL_TRIP_MASK	(BIT(MLX5_THERMAL_NUM_TRIPS) - 1)
+
+struct mlx5_thermal {
+	struct mlx5_core_dev *mdev;
+	struct thermal_zone_device *tzdev;
+};
+
+static int mlx5_thermal_get_mtmp_temp(struct mlx5_core_dev *mdev, u32 id, int *p_temp)
+{
+	u32 mtmp_out[MLX5_ST_SZ_DW(mtmp_reg)] = {};
+	u32 mtmp_in[MLX5_ST_SZ_DW(mtmp_reg)] = {};
+	int err;
+
+	MLX5_SET(mtmp_reg, mtmp_in, sensor_index, id);
+
+	err = mlx5_core_access_reg(mdev, mtmp_in,  sizeof(mtmp_in),
+				   mtmp_out, sizeof(mtmp_out),
+				   MLX5_REG_MTMP, 0, 0);
+
+	if (err)
+		return err;
+
+	*p_temp = MLX5_GET(mtmp_reg, mtmp_out, temperature);
+
+	return 0;
+}
+
+static int mlx5_thermal_get_temp(struct thermal_zone_device *tzdev,
+				 int *p_temp)
+{
+	struct mlx5_thermal *thermal = tzdev->devdata;
+	struct mlx5_core_dev *mdev = thermal->mdev;
+	int err;
+
+	err = mlx5_thermal_get_mtmp_temp(mdev, MLX5_THERMAL_ASIC_SENSOR_INDEX, p_temp);
+
+	if (err)
+		return err;
+
+	/* The unit of temp returned is in 0.125 C. The thermal
+	 * framework expects the value in 0.001 C.
+	 */
+	*p_temp *= 125;
+
+	return 0;
+}
+
+static struct thermal_zone_device_ops mlx5_thermal_ops = {
+	.get_temp = mlx5_thermal_get_temp,
+};
+
+int mlx5_thermal_init(struct mlx5_core_dev *mdev)
+{
+	struct mlx5_thermal *thermal;
+	struct thermal_zone_device *tzd;
+	const char *data = "mlx5";
+
+	tzd = thermal_zone_get_zone_by_name(data);
+	if (!IS_ERR(tzd))
+		return 0;
+
+	thermal = kzalloc(sizeof(*thermal), GFP_KERNEL);
+	if (!thermal)
+		return -ENOMEM;
+
+	thermal->mdev = mdev;
+	thermal->tzdev = thermal_zone_device_register(data,
+						      MLX5_THERMAL_NUM_TRIPS,
+						      MLX5_THERMAL_TRIP_MASK,
+						      thermal,
+						      &mlx5_thermal_ops,
+						      NULL, 0, MLX5_THERMAL_POLL_INT_MSEC);
+	if (IS_ERR(thermal->tzdev)) {
+		dev_err(mdev->device, "Failed to register thermal zone device (%s) %ld\n",
+			data, PTR_ERR(thermal->tzdev));
+		kfree(thermal);
+		return -EINVAL;
+	}
+
+	mdev->thermal = thermal;
+	return 0;
+}
+
+void mlx5_thermal_uninit(struct mlx5_core_dev *mdev)
+{
+	if (!mdev->thermal)
+		return;
+
+	thermal_zone_device_unregister(mdev->thermal->tzdev);
+	kfree(mdev->thermal);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/thermal.h b/drivers/net/ethernet/mellanox/mlx5/core/thermal.h
new file mode 100644
index 000000000000..7d752c122192
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/thermal.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+ * Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES.
+ */
+#ifndef __MLX5_THERMAL_DRIVER_H
+#define __MLX5_THERMAL_DRIVER_H
+
+#if IS_ENABLED(CONFIG_THERMAL)
+int mlx5_thermal_init(struct mlx5_core_dev *mdev);
+void mlx5_thermal_uninit(struct mlx5_core_dev *mdev);
+#else
+static inline int mlx5_thermal_init(struct mlx5_core_dev *mdev)
+{
+	mdev->thermal = NULL;
+	return 0;
+}
+
+static inline void mlx5_thermal_uninit(struct mlx5_core_dev *mdev) { }
+#endif
+
+#endif /* __MLX5_THERMAL_DRIVER_H */
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index f33389b42209..7a898113b6b7 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -134,6 +134,7 @@ enum {
 	MLX5_REG_PCAM		 = 0x507f,
 	MLX5_REG_NODE_DESC	 = 0x6001,
 	MLX5_REG_HOST_ENDIANNESS = 0x7004,
+	MLX5_REG_MTMP		 = 0x900A,
 	MLX5_REG_MCIA		 = 0x9014,
 	MLX5_REG_MFRL		 = 0x9028,
 	MLX5_REG_MLCR		 = 0x902b,
@@ -731,6 +732,7 @@ struct mlx5_fw_tracer;
 struct mlx5_vxlan;
 struct mlx5_geneve;
 struct mlx5_hv_vhca;
+struct mlx5_thermal;
 
 #define MLX5_LOG_SW_ICM_BLOCK_SIZE(dev) (MLX5_CAP_DEV_MEM(dev, log_sw_icm_alloc_granularity))
 #define MLX5_SW_ICM_BLOCK_SIZE(dev) (1 << MLX5_LOG_SW_ICM_BLOCK_SIZE(dev))
@@ -808,6 +810,7 @@ struct mlx5_core_dev {
 	struct mlx5_rsc_dump    *rsc_dump;
 	u32                      vsc_addr;
 	struct mlx5_hv_vhca	*hv_vhca;
+	struct mlx5_thermal	*thermal;
 };
 
 struct mlx5_db {
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 66d76e97a087..d2c164f0778c 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -10869,6 +10869,31 @@ struct mlx5_ifc_mrtc_reg_bits {
 	u8         time_l[0x20];
 };
 
+struct mlx5_ifc_mtmp_reg_bits {
+	u8         reserved_at_0[0x14];
+	u8         sensor_index[0xc];
+
+	u8         reserved_at_20[0x10];
+	u8         temperature[0x10];
+
+	u8         mte[0x1];
+	u8         mtr[0x1];
+	u8         reserved_at_42[0xe];
+	u8         max_temperature[0x10];
+
+	u8         tee[0x2];
+	u8         reserved_at_62[0xe];
+	u8         temp_threshold_hi[0x10];
+
+	u8         reserved_at_80[0x10];
+	u8         temp_threshold_lo[0x10];
+
+	u8         reserved_at_a0[0x20];
+
+	u8         sensor_name_hi[0x20];
+	u8         sensor_name_lo[0x20];
+};
+
 union mlx5_ifc_ports_control_registers_document_bits {
 	struct mlx5_ifc_bufferx_reg_bits bufferx_reg;
 	struct mlx5_ifc_eth_2819_cntrs_grp_data_layout_bits eth_2819_cntrs_grp_data_layout;
@@ -10931,6 +10956,7 @@ union mlx5_ifc_ports_control_registers_document_bits {
 	struct mlx5_ifc_mfrl_reg_bits mfrl_reg;
 	struct mlx5_ifc_mtutc_reg_bits mtutc_reg;
 	struct mlx5_ifc_mrtc_reg_bits mrtc_reg;
+	struct mlx5_ifc_mtmp_reg_bits mtmp_reg;
 	u8         reserved_at_0[0x60e0];
 };
 
-- 
cgit v1.2.3-70-g09d2


From aa98d15ea40b9d370871df7882b6d56d08aed692 Mon Sep 17 00:00:00 2001
From: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:24 -0700
Subject: net/mlx5e: Utilize the entire fifo

Previous check was comparing against the fifo mask. The mask is size of the
fifo (power of two) minus one, so a less than or equal comparator should be
used for checking if the fifo has room for the SKB.

Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-6-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index b9c2f67d3794..816ea83e6413 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -86,7 +86,7 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq);
 static inline bool
 mlx5e_skb_fifo_has_room(struct mlx5e_skb_fifo *fifo)
 {
-	return (u16)(*fifo->pc - *fifo->cc) < fifo->mask;
+	return (u16)(*fifo->pc - *fifo->cc) <= fifo->mask;
 }
 
 static inline bool
-- 
cgit v1.2.3-70-g09d2


From 2b5bd5b1611b817970f13b1c4dfdcda0b35e3690 Mon Sep 17 00:00:00 2001
From: Adham Faris <afaris@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:25 -0700
Subject: net/mlx5e: Rename RQ/SQ adaptive moderation state flag

Dynamic interrupt moderation RQ and SQ feature represented by
MLX5E_RQ_STATE_AM and MLX5E_SQ_STATE_AM enums respectively, is not
consistent with the feature naming in the driver, and with the formal
feature and library names.

Hence, change MLX5E_RQ_STATE_AM and MLX5E_SQ_STATE_AM enum type names in
core/en.h to MLX5E_RQ_STATE_DIM and MLX5E_SQ_STATE_DIM respectively.

Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-7-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      | 4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 88460b7796e5..66bca3a6a057 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -339,7 +339,7 @@ static inline u8 mlx5e_get_dcb_num_tc(struct mlx5e_params *params)
 enum {
 	MLX5E_RQ_STATE_ENABLED,
 	MLX5E_RQ_STATE_RECOVERING,
-	MLX5E_RQ_STATE_AM,
+	MLX5E_RQ_STATE_DIM,
 	MLX5E_RQ_STATE_NO_CSUM_COMPLETE,
 	MLX5E_RQ_STATE_CSUM_FULL, /* cqe_csum_full hw bit is set */
 	MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, /* set when mini_cqe_resp_stride_index cap is used */
@@ -390,7 +390,7 @@ enum {
 	MLX5E_SQ_STATE_MPWQE,
 	MLX5E_SQ_STATE_RECOVERING,
 	MLX5E_SQ_STATE_IPSEC,
-	MLX5E_SQ_STATE_AM,
+	MLX5E_SQ_STATE_DIM,
 	MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE,
 	MLX5E_SQ_STATE_PENDING_XSK_TX,
 	MLX5E_SQ_STATE_PENDING_TLS_RX_RESYNC,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 76a9c5194a70..5ca9fcf84586 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1188,7 +1188,7 @@ int mlx5e_open_rq(struct mlx5e_params *params, struct mlx5e_rq_param *param,
 		__set_bit(MLX5E_RQ_STATE_CSUM_FULL, &rq->state);
 
 	if (params->rx_dim_enabled)
-		__set_bit(MLX5E_RQ_STATE_AM, &rq->state);
+		__set_bit(MLX5E_RQ_STATE_DIM, &rq->state);
 
 	/* We disable csum_complete when XDP is enabled since
 	 * XDP programs might manipulate packets which will render
@@ -1664,7 +1664,7 @@ int mlx5e_open_txqsq(struct mlx5e_channel *c, u32 tisn, int txq_ix,
 		mlx5e_set_sq_maxrate(c->netdev, sq, tx_rate);
 
 	if (params->tx_dim_enabled)
-		sq->state |= BIT(MLX5E_SQ_STATE_AM);
+		sq->state |= BIT(MLX5E_SQ_STATE_DIM);
 
 	return 0;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
index 9a458a5d9853..a50bfda18e96 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
@@ -51,7 +51,7 @@ static void mlx5e_handle_tx_dim(struct mlx5e_txqsq *sq)
 	struct mlx5e_sq_stats *stats = sq->stats;
 	struct dim_sample dim_sample = {};
 
-	if (unlikely(!test_bit(MLX5E_SQ_STATE_AM, &sq->state)))
+	if (unlikely(!test_bit(MLX5E_SQ_STATE_DIM, &sq->state)))
 		return;
 
 	dim_update_sample(sq->cq.event_ctr, stats->packets, stats->bytes, &dim_sample);
@@ -63,7 +63,7 @@ static void mlx5e_handle_rx_dim(struct mlx5e_rq *rq)
 	struct mlx5e_rq_stats *stats = rq->stats;
 	struct dim_sample dim_sample = {};
 
-	if (unlikely(!test_bit(MLX5E_RQ_STATE_AM, &rq->state)))
+	if (unlikely(!test_bit(MLX5E_RQ_STATE_DIM, &rq->state)))
 		return;
 
 	dim_update_sample(rq->cq.event_ctr, stats->packets, stats->bytes, &dim_sample);
-- 
cgit v1.2.3-70-g09d2


From 1fe7bc109e3e18f7776ba71953cbc922b7ee9165 Mon Sep 17 00:00:00 2001
From: Adham Faris <afaris@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:26 -0700
Subject: net/mlx5e: Stringify RQ SW state in RQ devlink health diagnostics

One of the parameters that is retrieved/printed as a response to
devlink health diagnostics for rx reporter is the RQ SW state.

It's printed as a bitmap decimal number. Printing it as bitmap is
problematic and non informative.

In addition User can't count on SW state without accessing the kernel
sources (mlx5e rq state enum in en.h).

This patch prints RQ SW state in a textual representation, as a key:
value pairs, where disabled rq states will appear as '0' and enabled
ones will appear as '1'.

See below the generated output for rx health diagnostics devlink
command:

$ devlink health diagnose auxiliary/mlx5_core.eth.0/65535 reporter rx

Before:
=======================================================================
 Common config:
    RQ:
      type: 2 stride size: 2048 size: 8 ts_format: FRC
      CQ:
        stride size: 64 size: 1024
  RQs:
      channel ix: 0 rqn: 4172 HW state: 1 SW state: 37 WQE counter: 7 posted WQEs: 7 cc: 7
      CQ:
        cqn: 1033 HW status: 0 ci: 0 size: 1024
      EQ:
        eqn: 7 irqn: 32 vecidx: 0 ci: 2 size: 2048
      ICOSQ:
        sqn: 4169 HW state: 1 cc: 74 pc: 74 WQE size: 128
        CQ:
          cqn: 1030 cc: 1 size: 128
      channel ix: 1 ...
        .
        .

After:
=======================================================================
 Common config:
    RQ:
      type: 2 stride size: 2048 size: 8 ts_format: FRC
      CQ:
        stride size: 64 size: 1024
  RQs:
      channel ix: 0 rqn: 4172 HW state: 1 WQE counter: 7 posted WQEs: 7 cc: 7
        SW State:
          enabled: 1 recovering: 0 am: 1 no_csum_complete: 0 csum_full: 0 mini_cqe_hw_stridx: 1 shampo: 0 mini_cqe_enhanced: 0
      CQ:
        cqn: 1033 HW status: 0 ci: 0 size: 1024
      EQ:
        eqn: 7 irqn: 32 vecidx: 0 ci: 2 size: 2048
      ICOSQ:
        sqn: 4169 HW state: 1 cc: 74 pc: 74 WQE size: 128
        CQ:
          cqn: 1030 cc: 1 size: 128
       channel: ix: 1 ...
        .
        .

Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-8-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  6 ++-
 .../ethernet/mellanox/mlx5/core/en/reporter_rx.c   | 49 ++++++++++++++++++++--
 2 files changed, 50 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 66bca3a6a057..6c01da3bad74 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -336,8 +336,11 @@ static inline u8 mlx5e_get_dcb_num_tc(struct mlx5e_params *params)
 		params->mqprio.num_tc : 1;
 }
 
+/* Keep this enum consistent with the corresponding strings array
+ * declared in en/reporter_rx.c
+ */
 enum {
-	MLX5E_RQ_STATE_ENABLED,
+	MLX5E_RQ_STATE_ENABLED = 0,
 	MLX5E_RQ_STATE_RECOVERING,
 	MLX5E_RQ_STATE_DIM,
 	MLX5E_RQ_STATE_NO_CSUM_COMPLETE,
@@ -345,6 +348,7 @@ enum {
 	MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, /* set when mini_cqe_resp_stride_index cap is used */
 	MLX5E_RQ_STATE_SHAMPO, /* set when SHAMPO cap is used */
 	MLX5E_RQ_STATE_MINI_CQE_ENHANCED,  /* set when enhanced mini_cqe_cap is used */
+	MLX5E_NUM_RQ_STATES, /* Must be kept last */
 };
 
 struct mlx5e_cq {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
index c462fe76495b..98c87b3df806 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
@@ -8,6 +8,18 @@
 #include "ptp.h"
 #include "lib/tout.h"
 
+/* Keep this string array consistent with the MLX5E_RQ_STATE_* enums in en.h */
+static const char * const rq_sw_state_type_name[] = {
+	[MLX5E_RQ_STATE_ENABLED] = "enabled",
+	[MLX5E_RQ_STATE_RECOVERING] = "recovering",
+	[MLX5E_RQ_STATE_DIM] = "dim",
+	[MLX5E_RQ_STATE_NO_CSUM_COMPLETE] = "no_csum_complete",
+	[MLX5E_RQ_STATE_CSUM_FULL] = "csum_full",
+	[MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX] = "mini_cqe_hw_stridx",
+	[MLX5E_RQ_STATE_SHAMPO] = "shampo",
+	[MLX5E_RQ_STATE_MINI_CQE_ENHANCED] = "mini_cqe_enhanced",
+};
+
 static int mlx5e_query_rq_state(struct mlx5_core_dev *dev, u32 rqn, u8 *state)
 {
 	int outlen = MLX5_ST_SZ_BYTES(query_rq_out);
@@ -239,6 +251,35 @@ static int mlx5e_reporter_icosq_diagnose(struct mlx5e_icosq *icosq, u8 hw_state,
 	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
+static int mlx5e_health_rq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_rq *rq)
+{
+	int err;
+	int i;
+
+	BUILD_BUG_ON_MSG(ARRAY_SIZE(rq_sw_state_type_name) != MLX5E_NUM_RQ_STATES,
+			 "rq_sw_state_type_name string array must be consistent with MLX5E_RQ_STATE_* enum in en.h");
+	err = devlink_fmsg_obj_nest_start(fmsg);
+	if (err)
+		return err;
+
+	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(rq_sw_state_type_name); ++i) {
+		err = devlink_fmsg_u32_pair_put(fmsg, rq_sw_state_type_name[i],
+						test_bit(i, &rq->state));
+		if (err)
+			return err;
+	}
+
+	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	if (err)
+		return err;
+
+	return devlink_fmsg_obj_nest_end(fmsg);
+}
+
 static int
 mlx5e_rx_reporter_build_diagnose_output_rq_common(struct mlx5e_rq *rq,
 						  struct devlink_fmsg *fmsg)
@@ -265,10 +306,6 @@ mlx5e_rx_reporter_build_diagnose_output_rq_common(struct mlx5e_rq *rq,
 	if (err)
 		return err;
 
-	err = devlink_fmsg_u8_pair_put(fmsg, "SW state", rq->state);
-	if (err)
-		return err;
-
 	err = devlink_fmsg_u32_pair_put(fmsg, "WQE counter", wqe_counter);
 	if (err)
 		return err;
@@ -281,6 +318,10 @@ mlx5e_rx_reporter_build_diagnose_output_rq_common(struct mlx5e_rq *rq,
 	if (err)
 		return err;
 
+	err = mlx5e_health_rq_put_sw_state(fmsg, rq);
+	if (err)
+		return err;
+
 	err = mlx5e_health_cq_diag_fmsg(&rq->cq, fmsg);
 	if (err)
 		return err;
-- 
cgit v1.2.3-70-g09d2


From fc9d982a2512b294607b2922de3809e7c92810da Mon Sep 17 00:00:00 2001
From: Adham Faris <afaris@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:27 -0700
Subject: net/mlx5e: Expose SQ SW state as part of SQ health diagnostics

Add SQ SW state textual representation to devlink health diagnostics
for tx reporter.

SQ SW state can be retrieved by issuing the devlink command below:

$ devlink health diagnose auxiliary/mlx5_core.eth.0/65535 reporter tx

Output
=======================================================================
 Common Config:
    SQ:
      stride size: 64 size: 1024 ts_format: FRC
      CQ:
        stride size: 64 size: 1024
  SQs:
      channel ix: 0 tc: 0 txq ix: 0 sqn: 4170 HW state: 1 stopped: false cc: 0 pc: 0
        SW State:
          enabled: 1 mpwqe: 1 recovering: 0 ipsec: 0 am: 1 vlan_need_l2_inline: 1 pending_xsk_tx: 0 pending_tls_rx_resync: 0 xdp_multibuf: 0
      CQ:
        cqn: 1031 HW status: 0 ci: 0 size: 1024
      EQ:
        eqn: 7 irqn: 32 vecidx: 0 ci: 2 size: 2048

Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-9-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  6 ++-
 .../ethernet/mellanox/mlx5/core/en/reporter_tx.c   | 46 ++++++++++++++++++++++
 2 files changed, 51 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 6c01da3bad74..67f7e24d1f36 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -389,8 +389,11 @@ struct mlx5e_sq_dma {
 	enum mlx5e_dma_map_type type;
 };
 
+/* Keep this enum consistent with with the corresponding strings array
+ * declared in en/reporter_tx.c
+ */
 enum {
-	MLX5E_SQ_STATE_ENABLED,
+	MLX5E_SQ_STATE_ENABLED = 0,
 	MLX5E_SQ_STATE_MPWQE,
 	MLX5E_SQ_STATE_RECOVERING,
 	MLX5E_SQ_STATE_IPSEC,
@@ -399,6 +402,7 @@ enum {
 	MLX5E_SQ_STATE_PENDING_XSK_TX,
 	MLX5E_SQ_STATE_PENDING_TLS_RX_RESYNC,
 	MLX5E_SQ_STATE_XDP_MULTIBUF,
+	MLX5E_NUM_SQ_STATES, /* Must be kept last */
 };
 
 struct mlx5e_tx_mpwqe {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
index 34666e2b3871..44c1926843a1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
@@ -6,6 +6,19 @@
 #include "en/devlink.h"
 #include "lib/tout.h"
 
+/* Keep this string array consistent with the MLX5E_SQ_STATE_* enums in en.h */
+static const char * const sq_sw_state_type_name[] = {
+	[MLX5E_SQ_STATE_ENABLED] = "enabled",
+	[MLX5E_SQ_STATE_MPWQE] = "mpwqe",
+	[MLX5E_SQ_STATE_RECOVERING] = "recovering",
+	[MLX5E_SQ_STATE_IPSEC] = "ipsec",
+	[MLX5E_SQ_STATE_DIM] = "dim",
+	[MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE] = "vlan_need_l2_inline",
+	[MLX5E_SQ_STATE_PENDING_XSK_TX] = "pending_xsk_tx",
+	[MLX5E_SQ_STATE_PENDING_TLS_RX_RESYNC] = "pending_tls_rx_resync",
+	[MLX5E_SQ_STATE_XDP_MULTIBUF] = "xdp_multibuf",
+};
+
 static int mlx5e_wait_for_sq_flush(struct mlx5e_txqsq *sq)
 {
 	struct mlx5_core_dev *dev = sq->mdev;
@@ -37,6 +50,35 @@ static void mlx5e_reset_txqsq_cc_pc(struct mlx5e_txqsq *sq)
 	sq->pc = 0;
 }
 
+static int mlx5e_health_sq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_txqsq *sq)
+{
+	int err;
+	int i;
+
+	BUILD_BUG_ON_MSG(ARRAY_SIZE(sq_sw_state_type_name) != MLX5E_NUM_SQ_STATES,
+			 "sq_sw_state_type_name string array must be consistent with MLX5E_SQ_STATE_* enum in en.h");
+	err = devlink_fmsg_obj_nest_start(fmsg);
+	if (err)
+		return err;
+
+	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(sq_sw_state_type_name); ++i) {
+		err = devlink_fmsg_u32_pair_put(fmsg, sq_sw_state_type_name[i],
+						test_bit(i, &sq->state));
+		if (err)
+			return err;
+	}
+
+	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
+	if (err)
+		return err;
+
+	return devlink_fmsg_obj_nest_end(fmsg);
+}
+
 static int mlx5e_tx_reporter_err_cqe_recover(void *ctx)
 {
 	struct mlx5_core_dev *mdev;
@@ -190,6 +232,10 @@ mlx5e_tx_reporter_build_diagnose_output_sq_common(struct devlink_fmsg *fmsg,
 	if (err)
 		return err;
 
+	err = mlx5e_health_sq_put_sw_state(fmsg, sq);
+	if (err)
+		return err;
+
 	err = mlx5e_health_cq_diag_fmsg(&sq->cq, fmsg);
 	if (err)
 		return err;
-- 
cgit v1.2.3-70-g09d2


From bb76d250e55c72975b77f5d610fd3733306b3031 Mon Sep 17 00:00:00 2001
From: Adham Faris <afaris@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:28 -0700
Subject: net/mlx5e: Add XSK RQ state flag for RQ devlink health diagnostics

Currently RQ health diagnostics doesn't inform the user whether an RQ
is an XSK RQ or not.

Address this, by adding XSK state flag to RQ SW state enum in core/en.h.
XSK will be '1' if current RQ is an XSK RQ, and it will be '0' if it's
not.

In this example below, it can be seen that XSK field value is '1' since
xdpsock program have been attached to channel 0 before issuing the
devlink query command:

$ devlink health diagnose auxiliary/mlx5_core.eth.0/65535 reporter rx

Output:
=======================================================================
 Common config:
    RQ:
      type: 2 stride size: 4096 size: 16 ts_format: FRC
      CQ:
        stride size: 64 size: 1024
  RQs:
      channel ix: 0 rqn: 4236 HW state: 1 WQE counter: 15 posted WQEs: 15 cc: 15
        SW State:
          enabled: 1 recovering: 0 am: 1 no_csum_complete: 1 csum_full: 0 mini_cqe_hw_stridx: 1 shampo: 0 mini_cqe_enhanced: 0 xsk: 1
      CQ:
        cqn: 1085 HW status: 0 ci: 0 size: 1024
      EQ:
        eqn: 7 irqn: 32 vecidx: 0 ci: 5 size: 2048
      ICOSQ:
        sqn: 4229 HW state: 1 cc: 158 pc: 158 WQE size: 2048
        CQ:
          cqn: 1080 cc: 1 size: 2048

Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-10-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h             |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c   | 10 ++++++++--
 3 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 67f7e24d1f36..e4db252d406c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -348,6 +348,7 @@ enum {
 	MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, /* set when mini_cqe_resp_stride_index cap is used */
 	MLX5E_RQ_STATE_SHAMPO, /* set when SHAMPO cap is used */
 	MLX5E_RQ_STATE_MINI_CQE_ENHANCED,  /* set when enhanced mini_cqe_cap is used */
+	MLX5E_RQ_STATE_XSK, /* set to indicate an xsk rq */
 	MLX5E_NUM_RQ_STATES, /* Must be kept last */
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
index 98c87b3df806..b621f735cdc3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
@@ -18,6 +18,7 @@ static const char * const rq_sw_state_type_name[] = {
 	[MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX] = "mini_cqe_hw_stridx",
 	[MLX5E_RQ_STATE_SHAMPO] = "shampo",
 	[MLX5E_RQ_STATE_MINI_CQE_ENHANCED] = "mini_cqe_enhanced",
+	[MLX5E_RQ_STATE_XSK] = "xsk",
 };
 
 static int mlx5e_query_rq_state(struct mlx5_core_dev *dev, u32 rqn, u8 *state)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c
index 81a567e17264..ed279f450976 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c
@@ -93,13 +93,19 @@ static int mlx5e_open_xsk_rq(struct mlx5e_channel *c, struct mlx5e_params *param
 			     struct mlx5e_rq_param *rq_params, struct xsk_buff_pool *pool,
 			     struct mlx5e_xsk_param *xsk)
 {
+	struct mlx5e_rq *xskrq = &c->xskrq;
 	int err;
 
-	err = mlx5e_init_xsk_rq(c, params, pool, xsk, &c->xskrq);
+	err = mlx5e_init_xsk_rq(c, params, pool, xsk, xskrq);
 	if (err)
 		return err;
 
-	return mlx5e_open_rq(params, rq_params, xsk, cpu_to_node(c->cpu), &c->xskrq);
+	err = mlx5e_open_rq(params, rq_params, xsk, cpu_to_node(c->cpu), xskrq);
+	if (err)
+		return err;
+
+	__set_bit(MLX5E_RQ_STATE_XSK, &xskrq->state);
+	return 0;
 }
 
 int mlx5e_open_xsk(struct mlx5e_priv *priv, struct mlx5e_params *params,
-- 
cgit v1.2.3-70-g09d2


From 028522e2844393abc44f7bd7477eb4a455f01579 Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:29 -0700
Subject: net/mlx5: Move needed PTYS functions to core layer

Downstream patches require devlink params to access the PTYS register,
move the needed functions from mlx5e to the core layer.

Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-11-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/mellanox/mlx5/core/en/params.c    |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/port.c  | 157 +--------------------
 drivers/net/ethernet/mellanox/mlx5/core/en/port.h  |  14 --
 .../net/ethernet/mellanox/mlx5/core/en_ethtool.c   |  12 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c  |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/port.c     | 151 ++++++++++++++++++++
 include/linux/mlx5/port.h                          |  16 +++
 8 files changed, 179 insertions(+), 177 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index a21bd1179477..561da78d3b5c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -553,7 +553,7 @@ bool slow_pci_heuristic(struct mlx5_core_dev *mdev)
 	u32 link_speed = 0;
 	u32 pci_bw = 0;
 
-	mlx5e_port_max_linkspeed(mdev, &link_speed);
+	mlx5_port_max_linkspeed(mdev, &link_speed);
 	pci_bw = pcie_bandwidth_available(mdev->pdev, NULL, NULL, NULL);
 	mlx5_core_dbg_once(mdev, "Max link speed = %d, PCI BW = %d\n",
 			   link_speed, pci_bw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
index 505ba41195b9..dbe2b19a9570 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
@@ -32,101 +32,6 @@
 
 #include "port.h"
 
-/* speed in units of 1Mb */
-static const u32 mlx5e_link_speed[MLX5E_LINK_MODES_NUMBER] = {
-	[MLX5E_1000BASE_CX_SGMII] = 1000,
-	[MLX5E_1000BASE_KX]       = 1000,
-	[MLX5E_10GBASE_CX4]       = 10000,
-	[MLX5E_10GBASE_KX4]       = 10000,
-	[MLX5E_10GBASE_KR]        = 10000,
-	[MLX5E_20GBASE_KR2]       = 20000,
-	[MLX5E_40GBASE_CR4]       = 40000,
-	[MLX5E_40GBASE_KR4]       = 40000,
-	[MLX5E_56GBASE_R4]        = 56000,
-	[MLX5E_10GBASE_CR]        = 10000,
-	[MLX5E_10GBASE_SR]        = 10000,
-	[MLX5E_10GBASE_ER]        = 10000,
-	[MLX5E_40GBASE_SR4]       = 40000,
-	[MLX5E_40GBASE_LR4]       = 40000,
-	[MLX5E_50GBASE_SR2]       = 50000,
-	[MLX5E_100GBASE_CR4]      = 100000,
-	[MLX5E_100GBASE_SR4]      = 100000,
-	[MLX5E_100GBASE_KR4]      = 100000,
-	[MLX5E_100GBASE_LR4]      = 100000,
-	[MLX5E_100BASE_TX]        = 100,
-	[MLX5E_1000BASE_T]        = 1000,
-	[MLX5E_10GBASE_T]         = 10000,
-	[MLX5E_25GBASE_CR]        = 25000,
-	[MLX5E_25GBASE_KR]        = 25000,
-	[MLX5E_25GBASE_SR]        = 25000,
-	[MLX5E_50GBASE_CR2]       = 50000,
-	[MLX5E_50GBASE_KR2]       = 50000,
-};
-
-static const u32 mlx5e_ext_link_speed[MLX5E_EXT_LINK_MODES_NUMBER] = {
-	[MLX5E_SGMII_100M]			= 100,
-	[MLX5E_1000BASE_X_SGMII]		= 1000,
-	[MLX5E_5GBASE_R]			= 5000,
-	[MLX5E_10GBASE_XFI_XAUI_1]		= 10000,
-	[MLX5E_40GBASE_XLAUI_4_XLPPI_4]		= 40000,
-	[MLX5E_25GAUI_1_25GBASE_CR_KR]		= 25000,
-	[MLX5E_50GAUI_2_LAUI_2_50GBASE_CR2_KR2]	= 50000,
-	[MLX5E_50GAUI_1_LAUI_1_50GBASE_CR_KR]	= 50000,
-	[MLX5E_CAUI_4_100GBASE_CR4_KR4]		= 100000,
-	[MLX5E_100GAUI_2_100GBASE_CR2_KR2]	= 100000,
-	[MLX5E_200GAUI_4_200GBASE_CR4_KR4]	= 200000,
-	[MLX5E_400GAUI_8]			= 400000,
-	[MLX5E_100GAUI_1_100GBASE_CR_KR]	= 100000,
-	[MLX5E_200GAUI_2_200GBASE_CR2_KR2]	= 200000,
-	[MLX5E_400GAUI_4_400GBASE_CR4_KR4]	= 400000,
-};
-
-bool mlx5e_ptys_ext_supported(struct mlx5_core_dev *mdev)
-{
-	struct mlx5e_port_eth_proto eproto;
-	int err;
-
-	if (MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet))
-		return true;
-
-	err = mlx5_port_query_eth_proto(mdev, 1, true, &eproto);
-	if (err)
-		return false;
-
-	return !!eproto.cap;
-}
-
-static void mlx5e_port_get_speed_arr(struct mlx5_core_dev *mdev,
-				     const u32 **arr, u32 *size,
-				     bool force_legacy)
-{
-	bool ext = force_legacy ? false : mlx5e_ptys_ext_supported(mdev);
-
-	*size = ext ? ARRAY_SIZE(mlx5e_ext_link_speed) :
-		      ARRAY_SIZE(mlx5e_link_speed);
-	*arr  = ext ? mlx5e_ext_link_speed : mlx5e_link_speed;
-}
-
-int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
-			      struct mlx5e_port_eth_proto *eproto)
-{
-	u32 out[MLX5_ST_SZ_DW(ptys_reg)];
-	int err;
-
-	if (!eproto)
-		return -EINVAL;
-
-	err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN, port);
-	if (err)
-		return err;
-
-	eproto->cap   = MLX5_GET_ETH_PROTO(ptys_reg, out, ext,
-					   eth_proto_capability);
-	eproto->admin = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_admin);
-	eproto->oper  = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_oper);
-	return 0;
-}
-
 void mlx5_port_query_eth_autoneg(struct mlx5_core_dev *dev, u8 *an_status,
 				 u8 *an_disable_cap, u8 *an_disable_admin)
 {
@@ -172,30 +77,14 @@ int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
 			    sizeof(out), MLX5_REG_PTYS, 0, 1);
 }
 
-u32 mlx5e_port_ptys2speed(struct mlx5_core_dev *mdev, u32 eth_proto_oper,
-			  bool force_legacy)
-{
-	unsigned long temp = eth_proto_oper;
-	const u32 *table;
-	u32 speed = 0;
-	u32 max_size;
-	int i;
-
-	mlx5e_port_get_speed_arr(mdev, &table, &max_size, force_legacy);
-	i = find_first_bit(&temp, max_size);
-	if (i < max_size)
-		speed = table[i];
-	return speed;
-}
-
 int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
 {
-	struct mlx5e_port_eth_proto eproto;
+	struct mlx5_port_eth_proto eproto;
 	bool force_legacy = false;
 	bool ext;
 	int err;
 
-	ext = mlx5e_ptys_ext_supported(mdev);
+	ext = mlx5_ptys_ext_supported(mdev);
 	err = mlx5_port_query_eth_proto(mdev, 1, ext, &eproto);
 	if (err)
 		goto out;
@@ -205,7 +94,7 @@ int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
 		if (err)
 			goto out;
 	}
-	*speed = mlx5e_port_ptys2speed(mdev, eproto.oper, force_legacy);
+	*speed = mlx5_port_ptys2speed(mdev, eproto.oper, force_legacy);
 	if (!(*speed))
 		err = -EINVAL;
 
@@ -213,46 +102,6 @@ out:
 	return err;
 }
 
-int mlx5e_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
-{
-	struct mlx5e_port_eth_proto eproto;
-	u32 max_speed = 0;
-	const u32 *table;
-	u32 max_size;
-	bool ext;
-	int err;
-	int i;
-
-	ext = mlx5e_ptys_ext_supported(mdev);
-	err = mlx5_port_query_eth_proto(mdev, 1, ext, &eproto);
-	if (err)
-		return err;
-
-	mlx5e_port_get_speed_arr(mdev, &table, &max_size, false);
-	for (i = 0; i < max_size; ++i)
-		if (eproto.cap & MLX5E_PROT_MASK(i))
-			max_speed = max(max_speed, table[i]);
-
-	*speed = max_speed;
-	return 0;
-}
-
-u32 mlx5e_port_speed2linkmodes(struct mlx5_core_dev *mdev, u32 speed,
-			       bool force_legacy)
-{
-	u32 link_modes = 0;
-	const u32 *table;
-	u32 max_size;
-	int i;
-
-	mlx5e_port_get_speed_arr(mdev, &table, &max_size, force_legacy);
-	for (i = 0; i < max_size; ++i) {
-		if (table[i] == speed)
-			link_modes |= MLX5E_PROT_MASK(i);
-	}
-	return link_modes;
-}
-
 int mlx5e_port_query_pbmc(struct mlx5_core_dev *mdev, void *out)
 {
 	int sz = MLX5_ST_SZ_BYTES(pbmc_reg);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
index 3f474e370828..d1da225f35da 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
@@ -36,25 +36,11 @@
 #include <linux/mlx5/driver.h>
 #include "en.h"
 
-struct mlx5e_port_eth_proto {
-	u32 cap;
-	u32 admin;
-	u32 oper;
-};
-
-int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
-			      struct mlx5e_port_eth_proto *eproto);
 void mlx5_port_query_eth_autoneg(struct mlx5_core_dev *dev, u8 *an_status,
 				 u8 *an_disable_cap, u8 *an_disable_admin);
 int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
 			   u32 proto_admin, bool ext);
-u32 mlx5e_port_ptys2speed(struct mlx5_core_dev *mdev, u32 eth_proto_oper,
-			  bool force_legacy);
 int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
-int mlx5e_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
-u32 mlx5e_port_speed2linkmodes(struct mlx5_core_dev *mdev, u32 speed,
-			       bool force_legacy);
-bool mlx5e_ptys_ext_supported(struct mlx5_core_dev *mdev);
 int mlx5e_port_query_pbmc(struct mlx5_core_dev *mdev, void *out);
 int mlx5e_port_set_pbmc(struct mlx5_core_dev *mdev, void *in);
 int mlx5e_port_query_sbpr(struct mlx5_core_dev *mdev, u32 desc, u8 dir,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 7708acc9b2ab..53c35147f29b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -220,7 +220,7 @@ static void mlx5e_ethtool_get_speed_arr(struct mlx5_core_dev *mdev,
 					struct ptys2ethtool_config **arr,
 					u32 *size)
 {
-	bool ext = mlx5e_ptys_ext_supported(mdev);
+	bool ext = mlx5_ptys_ext_supported(mdev);
 
 	*arr = ext ? ptys2ext_ethtool_table : ptys2legacy_ethtool_table;
 	*size = ext ? ARRAY_SIZE(ptys2ext_ethtool_table) :
@@ -895,7 +895,7 @@ static void get_speed_duplex(struct net_device *netdev,
 	if (!netif_carrier_ok(netdev))
 		goto out;
 
-	speed = mlx5e_port_ptys2speed(priv->mdev, eth_proto_oper, force_legacy);
+	speed = mlx5_port_ptys2speed(priv->mdev, eth_proto_oper, force_legacy);
 	if (!speed) {
 		if (data_rate_oper)
 			speed = 100 * data_rate_oper;
@@ -980,7 +980,7 @@ static void get_lp_advertising(struct mlx5_core_dev *mdev, u32 eth_proto_lp,
 			       struct ethtool_link_ksettings *link_ksettings)
 {
 	unsigned long *lp_advertising = link_ksettings->link_modes.lp_advertising;
-	bool ext = mlx5e_ptys_ext_supported(mdev);
+	bool ext = mlx5_ptys_ext_supported(mdev);
 
 	ptys2ethtool_adver_link(lp_advertising, eth_proto_lp, ext);
 }
@@ -1160,7 +1160,7 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
 				     const struct ethtool_link_ksettings *link_ksettings)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
-	struct mlx5e_port_eth_proto eproto;
+	struct mlx5_port_eth_proto eproto;
 	const unsigned long *adver;
 	bool an_changes = false;
 	u8 an_disable_admin;
@@ -1180,7 +1180,7 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
 	autoneg = link_ksettings->base.autoneg;
 	speed = link_ksettings->base.speed;
 
-	ext_supported = mlx5e_ptys_ext_supported(mdev);
+	ext_supported = mlx5_ptys_ext_supported(mdev);
 	ext = ext_requested(autoneg, adver, ext_supported);
 	if (!ext_supported && ext)
 		return -EOPNOTSUPP;
@@ -1194,7 +1194,7 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
 		goto out;
 	}
 	link_modes = autoneg == AUTONEG_ENABLE ? ethtool2ptys_adver_func(adver) :
-		mlx5e_port_speed2linkmodes(mdev, speed, !ext);
+		mlx5_port_speed2linkmodes(mdev, speed, !ext);
 
 	err = mlx5e_speed_validate(priv->netdev, ext, link_modes, autoneg);
 	if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 70b8d2dfa751..79dd8ad5ede7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1108,7 +1108,7 @@ mlx5e_hairpin_params_init(struct mlx5e_hairpin_params *hairpin_params,
 
 	hairpin_params->mdev = mdev;
 	/* set hairpin pair per each 50Gbs share of the link */
-	mlx5e_port_max_linkspeed(mdev, &link_speed);
+	mlx5_port_max_linkspeed(mdev, &link_speed);
 	link_speed = max_t(u32, link_speed, 50000);
 	link_speed64 = link_speed;
 	do_div(link_speed64, 50000);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
index 75015d370922..7c79476cc5f9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
@@ -744,7 +744,7 @@ static int esw_qos_devlink_rate_to_mbps(struct mlx5_core_dev *mdev, const char *
 	u64 value;
 	int err;
 
-	err = mlx5e_port_max_linkspeed(mdev, &link_speed_max);
+	err = mlx5_port_max_linkspeed(mdev, &link_speed_max);
 	if (err) {
 		NL_SET_ERR_MSG_MOD(extack, "Failed to get link maximum speed");
 		return err;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index a1548e6bfb35..0daeb4b72cca 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -1054,3 +1054,154 @@ out:
 	kfree(out);
 	return err;
 }
+
+/* speed in units of 1Mb */
+static const u32 mlx5e_link_speed[MLX5E_LINK_MODES_NUMBER] = {
+	[MLX5E_1000BASE_CX_SGMII] = 1000,
+	[MLX5E_1000BASE_KX]       = 1000,
+	[MLX5E_10GBASE_CX4]       = 10000,
+	[MLX5E_10GBASE_KX4]       = 10000,
+	[MLX5E_10GBASE_KR]        = 10000,
+	[MLX5E_20GBASE_KR2]       = 20000,
+	[MLX5E_40GBASE_CR4]       = 40000,
+	[MLX5E_40GBASE_KR4]       = 40000,
+	[MLX5E_56GBASE_R4]        = 56000,
+	[MLX5E_10GBASE_CR]        = 10000,
+	[MLX5E_10GBASE_SR]        = 10000,
+	[MLX5E_10GBASE_ER]        = 10000,
+	[MLX5E_40GBASE_SR4]       = 40000,
+	[MLX5E_40GBASE_LR4]       = 40000,
+	[MLX5E_50GBASE_SR2]       = 50000,
+	[MLX5E_100GBASE_CR4]      = 100000,
+	[MLX5E_100GBASE_SR4]      = 100000,
+	[MLX5E_100GBASE_KR4]      = 100000,
+	[MLX5E_100GBASE_LR4]      = 100000,
+	[MLX5E_100BASE_TX]        = 100,
+	[MLX5E_1000BASE_T]        = 1000,
+	[MLX5E_10GBASE_T]         = 10000,
+	[MLX5E_25GBASE_CR]        = 25000,
+	[MLX5E_25GBASE_KR]        = 25000,
+	[MLX5E_25GBASE_SR]        = 25000,
+	[MLX5E_50GBASE_CR2]       = 50000,
+	[MLX5E_50GBASE_KR2]       = 50000,
+};
+
+static const u32 mlx5e_ext_link_speed[MLX5E_EXT_LINK_MODES_NUMBER] = {
+	[MLX5E_SGMII_100M] = 100,
+	[MLX5E_1000BASE_X_SGMII] = 1000,
+	[MLX5E_5GBASE_R] = 5000,
+	[MLX5E_10GBASE_XFI_XAUI_1] = 10000,
+	[MLX5E_40GBASE_XLAUI_4_XLPPI_4] = 40000,
+	[MLX5E_25GAUI_1_25GBASE_CR_KR] = 25000,
+	[MLX5E_50GAUI_2_LAUI_2_50GBASE_CR2_KR2] = 50000,
+	[MLX5E_50GAUI_1_LAUI_1_50GBASE_CR_KR] = 50000,
+	[MLX5E_CAUI_4_100GBASE_CR4_KR4] = 100000,
+	[MLX5E_100GAUI_2_100GBASE_CR2_KR2] = 100000,
+	[MLX5E_200GAUI_4_200GBASE_CR4_KR4] = 200000,
+	[MLX5E_400GAUI_8] = 400000,
+	[MLX5E_100GAUI_1_100GBASE_CR_KR] = 100000,
+	[MLX5E_200GAUI_2_200GBASE_CR2_KR2] = 200000,
+	[MLX5E_400GAUI_4_400GBASE_CR4_KR4] = 400000,
+};
+
+int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
+			      struct mlx5_port_eth_proto *eproto)
+{
+	u32 out[MLX5_ST_SZ_DW(ptys_reg)];
+	int err;
+
+	if (!eproto)
+		return -EINVAL;
+
+	err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN, port);
+	if (err)
+		return err;
+
+	eproto->cap   = MLX5_GET_ETH_PROTO(ptys_reg, out, ext,
+					   eth_proto_capability);
+	eproto->admin = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_admin);
+	eproto->oper  = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_oper);
+	return 0;
+}
+
+bool mlx5_ptys_ext_supported(struct mlx5_core_dev *mdev)
+{
+	struct mlx5_port_eth_proto eproto;
+	int err;
+
+	if (MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet))
+		return true;
+
+	err = mlx5_port_query_eth_proto(mdev, 1, true, &eproto);
+	if (err)
+		return false;
+
+	return !!eproto.cap;
+}
+
+static void mlx5e_port_get_speed_arr(struct mlx5_core_dev *mdev,
+				     const u32 **arr, u32 *size,
+				     bool force_legacy)
+{
+	bool ext = force_legacy ? false : mlx5_ptys_ext_supported(mdev);
+
+	*size = ext ? ARRAY_SIZE(mlx5e_ext_link_speed) :
+		      ARRAY_SIZE(mlx5e_link_speed);
+	*arr  = ext ? mlx5e_ext_link_speed : mlx5e_link_speed;
+}
+
+u32 mlx5_port_ptys2speed(struct mlx5_core_dev *mdev, u32 eth_proto_oper,
+			 bool force_legacy)
+{
+	unsigned long temp = eth_proto_oper;
+	const u32 *table;
+	u32 speed = 0;
+	u32 max_size;
+	int i;
+
+	mlx5e_port_get_speed_arr(mdev, &table, &max_size, force_legacy);
+	i = find_first_bit(&temp, max_size);
+	if (i < max_size)
+		speed = table[i];
+	return speed;
+}
+
+u32 mlx5_port_speed2linkmodes(struct mlx5_core_dev *mdev, u32 speed,
+			      bool force_legacy)
+{
+	u32 link_modes = 0;
+	const u32 *table;
+	u32 max_size;
+	int i;
+
+	mlx5e_port_get_speed_arr(mdev, &table, &max_size, force_legacy);
+	for (i = 0; i < max_size; ++i) {
+		if (table[i] == speed)
+			link_modes |= MLX5E_PROT_MASK(i);
+	}
+	return link_modes;
+}
+
+int mlx5_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
+{
+	struct mlx5_port_eth_proto eproto;
+	u32 max_speed = 0;
+	const u32 *table;
+	u32 max_size;
+	bool ext;
+	int err;
+	int i;
+
+	ext = mlx5_ptys_ext_supported(mdev);
+	err = mlx5_port_query_eth_proto(mdev, 1, ext, &eproto);
+	if (err)
+		return err;
+
+	mlx5e_port_get_speed_arr(mdev, &table, &max_size, false);
+	for (i = 0; i < max_size; ++i)
+		if (eproto.cap & MLX5E_PROT_MASK(i))
+			max_speed = max(max_speed, table[i]);
+
+	*speed = max_speed;
+	return 0;
+}
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index e96ee1e348cb..98b2e1e149f9 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -141,6 +141,12 @@ enum mlx5_ptys_width {
 	MLX5_PTYS_WIDTH_12X	= 1 << 4,
 };
 
+struct mlx5_port_eth_proto {
+	u32 cap;
+	u32 admin;
+	u32 oper;
+};
+
 #define MLX5E_PROT_MASK(link_mode) (1U << link_mode)
 #define MLX5_GET_ETH_PROTO(reg, out, ext, field)	\
 	(ext ? MLX5_GET(reg, out, ext_##field) :	\
@@ -218,4 +224,14 @@ int mlx5_set_trust_state(struct mlx5_core_dev *mdev, u8 trust_state);
 int mlx5_query_trust_state(struct mlx5_core_dev *mdev, u8 *trust_state);
 int mlx5_set_dscp2prio(struct mlx5_core_dev *mdev, u8 dscp, u8 prio);
 int mlx5_query_dscp2prio(struct mlx5_core_dev *mdev, u8 *dscp2prio);
+
+int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
+			      struct mlx5_port_eth_proto *eproto);
+bool mlx5_ptys_ext_supported(struct mlx5_core_dev *mdev);
+u32 mlx5_port_ptys2speed(struct mlx5_core_dev *mdev, u32 eth_proto_oper,
+			 bool force_legacy);
+u32 mlx5_port_speed2linkmodes(struct mlx5_core_dev *mdev, u32 speed,
+			      bool force_legacy);
+int mlx5_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
+
 #endif /* __MLX5_PORT_H__ */
-- 
cgit v1.2.3-70-g09d2


From 1bffcea42926b26e092045ac398850e80d950bb2 Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:30 -0700
Subject: net/mlx5e: Add devlink hairpin queues parameters

We refer to a TC NIC rule that involves forwarding as "hairpin".
Hairpin queues are mlx5 hardware specific implementation for hardware
forwarding of such packets.

Per the discussion in [1], move the hairpin queues control (number and
size) from debugfs to devlink.

Expose two devlink params:
- hairpin_num_queues: control the number of hairpin queues
- hairpin_queue_size: control the size (in packets) of the hairpin queues

[1] https://lore.kernel.org/all/20230111194608.7f15b9a1@kernel.org/

Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-12-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/devlink.rst             | 35 ++++++++++++
 Documentation/networking/devlink/mlx5.rst          | 12 ++++
 drivers/net/ethernet/mellanox/mlx5/core/devlink.c  | 66 ++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlx5/core/devlink.h  |  2 +
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 50 +++++++---------
 5 files changed, 134 insertions(+), 31 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
index 9b5c40ba7f0d..0995e4e5acd7 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
@@ -122,6 +122,41 @@ users try to enable them.
 
     $ devlink dev eswitch set pci/0000:06:00.0 mode switchdev
 
+hairpin_num_queues: Number of hairpin queues
+--------------------------------------------
+We refer to a TC NIC rule that involves forwarding as "hairpin".
+
+Hairpin queues are mlx5 hardware specific implementation for hardware
+forwarding of such packets.
+
+- Show the number of hairpin queues::
+
+    $ devlink dev param show pci/0000:06:00.0 name hairpin_num_queues
+      pci/0000:06:00.0:
+        name hairpin_num_queues type driver-specific
+          values:
+            cmode driverinit value 2
+
+- Change the number of hairpin queues::
+
+    $ devlink dev param set pci/0000:06:00.0 name hairpin_num_queues value 4 cmode driverinit
+
+hairpin_queue_size: Size of the hairpin queues
+----------------------------------------------
+Control the size of the hairpin queues.
+
+- Show the size of the hairpin queues::
+
+    $ devlink dev param show pci/0000:06:00.0 name hairpin_queue_size
+      pci/0000:06:00.0:
+        name hairpin_queue_size type driver-specific
+          values:
+            cmode driverinit value 1024
+
+- Change the size (in packets) of the hairpin queues::
+
+    $ devlink dev param set pci/0000:06:00.0 name hairpin_queue_size value 512 cmode driverinit
+
 Health reporters
 ================
 
diff --git a/Documentation/networking/devlink/mlx5.rst b/Documentation/networking/devlink/mlx5.rst
index 3321117cf605..202798d6501e 100644
--- a/Documentation/networking/devlink/mlx5.rst
+++ b/Documentation/networking/devlink/mlx5.rst
@@ -72,6 +72,18 @@ parameters.
 
        Default: disabled
 
+   * - ``hairpin_num_queues``
+     - u32
+     - driverinit
+     - We refer to a TC NIC rule that involves forwarding as "hairpin".
+       Hairpin queues are mlx5 hardware specific implementation for hardware
+       forwarding of such packets.
+
+       Control the number of hairpin queues.
+   * - ``hairpin_queue_size``
+     - u32
+     - driverinit
+     - Control the size (in packets) of the hairpin queues.
 
 The ``mlx5`` driver supports reloading via ``DEVLINK_CMD_RELOAD``
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index b7784e02c2dd..1ee2a472e1d2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -494,6 +494,61 @@ static int mlx5_devlink_eq_depth_validate(struct devlink *devlink, u32 id,
 	return (val.vu32 >= 64 && val.vu32 <= 4096) ? 0 : -EINVAL;
 }
 
+static int
+mlx5_devlink_hairpin_num_queues_validate(struct devlink *devlink, u32 id,
+					 union devlink_param_value val,
+					 struct netlink_ext_ack *extack)
+{
+	return val.vu32 ? 0 : -EINVAL;
+}
+
+static int
+mlx5_devlink_hairpin_queue_size_validate(struct devlink *devlink, u32 id,
+					 union devlink_param_value val,
+					 struct netlink_ext_ack *extack)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+	u32 val32 = val.vu32;
+
+	if (!is_power_of_2(val32)) {
+		NL_SET_ERR_MSG_MOD(extack, "Value is not power of two");
+		return -EINVAL;
+	}
+
+	if (val32 > BIT(MLX5_CAP_GEN(dev, log_max_hairpin_num_packets))) {
+		NL_SET_ERR_MSG_FMT_MOD(
+			extack, "Maximum hairpin queue size is %lu",
+			BIT(MLX5_CAP_GEN(dev, log_max_hairpin_num_packets)));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void mlx5_devlink_hairpin_params_init_values(struct devlink *devlink)
+{
+	struct mlx5_core_dev *dev = devlink_priv(devlink);
+	union devlink_param_value value;
+	u64 link_speed64;
+	u32 link_speed;
+
+	/* set hairpin pair per each 50Gbs share of the link */
+	mlx5_port_max_linkspeed(dev, &link_speed);
+	link_speed = max_t(u32, link_speed, 50000);
+	link_speed64 = link_speed;
+	do_div(link_speed64, 50000);
+
+	value.vu32 = link_speed64;
+	devl_param_driverinit_value_set(
+		devlink, MLX5_DEVLINK_PARAM_ID_HAIRPIN_NUM_QUEUES, value);
+
+	value.vu32 =
+		BIT(min_t(u32, 16 - MLX5_MPWRQ_MIN_LOG_STRIDE_SZ(dev),
+			  MLX5_CAP_GEN(dev, log_max_hairpin_num_packets)));
+	devl_param_driverinit_value_set(
+		devlink, MLX5_DEVLINK_PARAM_ID_HAIRPIN_QUEUE_SIZE, value);
+}
+
 static const struct devlink_param mlx5_devlink_params[] = {
 	DEVLINK_PARAM_GENERIC(ENABLE_ROCE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
 			      NULL, NULL, mlx5_devlink_enable_roce_validate),
@@ -547,6 +602,14 @@ static void mlx5_devlink_set_params_init_values(struct devlink *devlink)
 static const struct devlink_param mlx5_devlink_eth_params[] = {
 	DEVLINK_PARAM_GENERIC(ENABLE_ETH, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT),
 			      NULL, NULL, NULL),
+	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_HAIRPIN_NUM_QUEUES,
+			     "hairpin_num_queues", DEVLINK_PARAM_TYPE_U32,
+			     BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), NULL, NULL,
+			     mlx5_devlink_hairpin_num_queues_validate),
+	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_HAIRPIN_QUEUE_SIZE,
+			     "hairpin_queue_size", DEVLINK_PARAM_TYPE_U32,
+			     BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), NULL, NULL,
+			     mlx5_devlink_hairpin_queue_size_validate),
 };
 
 static int mlx5_devlink_eth_params_register(struct devlink *devlink)
@@ -567,6 +630,9 @@ static int mlx5_devlink_eth_params_register(struct devlink *devlink)
 	devl_param_driverinit_value_set(devlink,
 					DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH,
 					value);
+
+	mlx5_devlink_hairpin_params_init_values(devlink);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
index 212b12424146..5dcfb4d86d8a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
@@ -12,6 +12,8 @@ enum mlx5_devlink_param_id {
 	MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM,
 	MLX5_DEVLINK_PARAM_ID_ESW_PORT_METADATA,
 	MLX5_DEVLINK_PARAM_ID_ESW_MULTIPORT,
+	MLX5_DEVLINK_PARAM_ID_HAIRPIN_NUM_QUEUES,
+	MLX5_DEVLINK_PARAM_ID_HAIRPIN_QUEUE_SIZE,
 };
 
 struct mlx5_trap_ctx {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 79dd8ad5ede7..2e6351ef4d9c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -44,6 +44,7 @@
 #include <net/bareudp.h>
 #include <net/bonding.h>
 #include <net/dst_metadata.h>
+#include "devlink.h"
 #include "en.h"
 #include "en/tc/post_act.h"
 #include "en/tc/act_stats.h"
@@ -73,12 +74,6 @@
 #define MLX5E_TC_TABLE_NUM_GROUPS 4
 #define MLX5E_TC_TABLE_MAX_GROUP_SIZE BIT(18)
 
-struct mlx5e_hairpin_params {
-	struct mlx5_core_dev *mdev;
-	u32 num_queues;
-	u32 queue_size;
-};
-
 struct mlx5e_tc_table {
 	/* Protects the dynamic assignment of the t parameter
 	 * which is the nic tc root table.
@@ -101,7 +96,6 @@ struct mlx5e_tc_table {
 
 	struct mlx5_tc_ct_priv         *ct;
 	struct mapping_ctx             *mapping;
-	struct mlx5e_hairpin_params    hairpin_params;
 	struct dentry                  *dfs_root;
 
 	/* tc action stats */
@@ -1099,33 +1093,15 @@ static void mlx5e_tc_debugfs_init(struct mlx5e_tc_table *tc,
 			    &debugfs_hairpin_table_dump_fops);
 }
 
-static void
-mlx5e_hairpin_params_init(struct mlx5e_hairpin_params *hairpin_params,
-			  struct mlx5_core_dev *mdev)
-{
-	u64 link_speed64;
-	u32 link_speed;
-
-	hairpin_params->mdev = mdev;
-	/* set hairpin pair per each 50Gbs share of the link */
-	mlx5_port_max_linkspeed(mdev, &link_speed);
-	link_speed = max_t(u32, link_speed, 50000);
-	link_speed64 = link_speed;
-	do_div(link_speed64, 50000);
-	hairpin_params->num_queues = link_speed64;
-
-	hairpin_params->queue_size =
-		BIT(min_t(u32, 16 - MLX5_MPWRQ_MIN_LOG_STRIDE_SZ(mdev),
-			  MLX5_CAP_GEN(mdev, log_max_hairpin_num_packets)));
-}
-
 static int mlx5e_hairpin_flow_add(struct mlx5e_priv *priv,
 				  struct mlx5e_tc_flow *flow,
 				  struct mlx5e_tc_flow_parse_attr *parse_attr,
 				  struct netlink_ext_ack *extack)
 {
 	struct mlx5e_tc_table *tc = mlx5e_fs_get_tc(priv->fs);
+	struct devlink *devlink = priv_to_devlink(priv->mdev);
 	int peer_ifindex = parse_attr->mirred_ifindex[0];
+	union devlink_param_value val = {};
 	struct mlx5_hairpin_params params;
 	struct mlx5_core_dev *peer_mdev;
 	struct mlx5e_hairpin_entry *hpe;
@@ -1182,7 +1158,14 @@ static int mlx5e_hairpin_flow_add(struct mlx5e_priv *priv,
 		 hash_hairpin_info(peer_id, match_prio));
 	mutex_unlock(&tc->hairpin_tbl_lock);
 
-	params.log_num_packets = ilog2(tc->hairpin_params.queue_size);
+	err = devl_param_driverinit_value_get(
+		devlink, MLX5_DEVLINK_PARAM_ID_HAIRPIN_QUEUE_SIZE, &val);
+	if (err) {
+		err = -ENOMEM;
+		goto out_err;
+	}
+
+	params.log_num_packets = ilog2(val.vu32);
 	params.log_data_size =
 		clamp_t(u32,
 			params.log_num_packets +
@@ -1191,7 +1174,14 @@ static int mlx5e_hairpin_flow_add(struct mlx5e_priv *priv,
 			MLX5_CAP_GEN(priv->mdev, log_max_hairpin_wq_data_sz));
 
 	params.q_counter = priv->q_counter;
-	params.num_channels = tc->hairpin_params.num_queues;
+	err = devl_param_driverinit_value_get(
+		devlink, MLX5_DEVLINK_PARAM_ID_HAIRPIN_NUM_QUEUES, &val);
+	if (err) {
+		err = -ENOMEM;
+		goto out_err;
+	}
+
+	params.num_channels = val.vu32;
 
 	hp = mlx5e_hairpin_create(priv, &params, peer_ifindex);
 	hpe->hp = hp;
@@ -5289,8 +5279,6 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 	tc->ct = mlx5_tc_ct_init(priv, tc->chains, &tc->mod_hdr,
 				 MLX5_FLOW_NAMESPACE_KERNEL, tc->post_act);
 
-	mlx5e_hairpin_params_init(&tc->hairpin_params, dev);
-
 	tc->netdevice_nb.notifier_call = mlx5e_tc_netdev_event;
 	err = register_netdevice_notifier_dev_net(priv->netdev,
 						  &tc->netdevice_nb,
-- 
cgit v1.2.3-70-g09d2


From 8a0594c0961012ac07112f41800da29dd5632c03 Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:31 -0700
Subject: net/mlx5e: Add more information to hairpin table dump

Print the number of hairpin queues and size as part of the hairpin table
dump.

Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-13-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 2e6351ef4d9c..a139b5e88e2a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -583,6 +583,7 @@ struct mlx5e_hairpin {
 	struct mlx5e_tir direct_tir;
 
 	int num_channels;
+	u8 log_num_packets;
 	struct mlx5e_rqt indir_rqt;
 	struct mlx5e_tir indir_tir[MLX5E_NUM_INDIR_TIRS];
 	struct mlx5_ttc_table *ttc;
@@ -929,6 +930,7 @@ mlx5e_hairpin_create(struct mlx5e_priv *priv, struct mlx5_hairpin_params *params
 	hp->func_mdev = func_mdev;
 	hp->func_priv = priv;
 	hp->num_channels = params->num_channels;
+	hp->log_num_packets = params->log_num_packets;
 
 	err = mlx5e_hairpin_create_transport(hp);
 	if (err)
@@ -1070,9 +1072,11 @@ static int debugfs_hairpin_table_dump_show(struct seq_file *file, void *priv)
 
 	mutex_lock(&tc->hairpin_tbl_lock);
 	hash_for_each(tc->hairpin_tbl, bkt, hpe, hairpin_hlist)
-		seq_printf(file, "Hairpin peer_vhca_id %u prio %u refcnt %u\n",
+		seq_printf(file,
+			   "Hairpin peer_vhca_id %u prio %u refcnt %u num_channels %u num_packets %lu\n",
 			   hpe->peer_vhca_id, hpe->prio,
-			   refcount_read(&hpe->refcnt));
+			   refcount_read(&hpe->refcnt), hpe->hp->num_channels,
+			   BIT(hpe->hp->log_num_packets));
 	mutex_unlock(&tc->hairpin_tbl_lock);
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 244fd698207f0baec062da4737e31c76b1be2371 Mon Sep 17 00:00:00 2001
From: Maor Dickman <maord@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:32 -0700
Subject: net/mlx5e: TC, Extract indr setup block checks to function

In preparation for next patch which will add new check
if device block can be setup, extract all existing checks
to function to make it more readable and maintainable.

Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-14-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/mellanox/mlx5/core/en/rep/tc.c    | 58 ++++++++++++++--------
 1 file changed, 36 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
index 8f7452dc00ee..b4af006dc494 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
@@ -426,39 +426,53 @@ static bool mlx5e_rep_macvlan_mode_supported(const struct net_device *dev)
 	return macvlan->mode == MACVLAN_MODE_PASSTHRU;
 }
 
-static int
-mlx5e_rep_indr_setup_block(struct net_device *netdev, struct Qdisc *sch,
-			   struct mlx5e_rep_priv *rpriv,
-			   struct flow_block_offload *f,
-			   flow_setup_cb_t *setup_cb,
-			   void *data,
-			   void (*cleanup)(struct flow_block_cb *block_cb))
+static bool
+mlx5e_rep_check_indr_block_supported(struct mlx5e_rep_priv *rpriv,
+				     struct net_device *netdev,
+				     struct flow_block_offload *f)
 {
 	struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
-	bool is_ovs_int_port = netif_is_ovs_master(netdev);
-	struct mlx5e_rep_indr_block_priv *indr_priv;
-	struct flow_block_cb *block_cb;
 
-	if (!mlx5e_tc_tun_device_to_offload(priv, netdev) &&
-	    !(is_vlan_dev(netdev) && vlan_dev_real_dev(netdev) == rpriv->netdev) &&
-	    !is_ovs_int_port) {
-		if (!(netif_is_macvlan(netdev) && macvlan_dev_real_dev(netdev) == rpriv->netdev))
-			return -EOPNOTSUPP;
+	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS &&
+	    f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS)
+		return false;
+
+	if (mlx5e_tc_tun_device_to_offload(priv, netdev))
+		return true;
+
+	if (is_vlan_dev(netdev) && vlan_dev_real_dev(netdev) == rpriv->netdev)
+		return true;
+
+	if (netif_is_macvlan(netdev)) {
 		if (!mlx5e_rep_macvlan_mode_supported(netdev)) {
 			netdev_warn(netdev, "Offloading ingress filter is supported only with macvlan passthru mode");
-			return -EOPNOTSUPP;
+			return false;
 		}
+
+		if (macvlan_dev_real_dev(netdev) == rpriv->netdev)
+			return true;
 	}
 
-	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS &&
-	    f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS)
-		return -EOPNOTSUPP;
+	if (netif_is_ovs_master(netdev) && f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS &&
+	    mlx5e_tc_int_port_supported(esw))
+		return true;
 
-	if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS && !is_ovs_int_port)
-		return -EOPNOTSUPP;
+	return false;
+}
+
+static int
+mlx5e_rep_indr_setup_block(struct net_device *netdev, struct Qdisc *sch,
+			   struct mlx5e_rep_priv *rpriv,
+			   struct flow_block_offload *f,
+			   flow_setup_cb_t *setup_cb,
+			   void *data,
+			   void (*cleanup)(struct flow_block_cb *block_cb))
+{
+	struct mlx5e_rep_indr_block_priv *indr_priv;
+	struct flow_block_cb *block_cb;
 
-	if (is_ovs_int_port && !mlx5e_tc_int_port_supported(esw))
+	if (!mlx5e_rep_check_indr_block_supported(rpriv, netdev, f))
 		return -EOPNOTSUPP;
 
 	f->unlocked_driver_cb = true;
-- 
cgit v1.2.3-70-g09d2


From d5d006bb27ad9d408074eef7e091fbd87335f913 Mon Sep 17 00:00:00 2001
From: Maor Dickman <maord@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:33 -0700
Subject: net/mlx5e: Enable TC offload for ingress MACVLAN over bond

Support offloading of TC rules that filter ingress traffic from a MACVLAN
device, which is attached to bond device.

Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-15-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
index b4af006dc494..19c4a83982ca 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
@@ -433,6 +433,7 @@ mlx5e_rep_check_indr_block_supported(struct mlx5e_rep_priv *rpriv,
 {
 	struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
+	struct net_device *macvlan_real_dev;
 
 	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS &&
 	    f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS)
@@ -450,7 +451,11 @@ mlx5e_rep_check_indr_block_supported(struct mlx5e_rep_priv *rpriv,
 			return false;
 		}
 
-		if (macvlan_dev_real_dev(netdev) == rpriv->netdev)
+		macvlan_real_dev = macvlan_dev_real_dev(netdev);
+
+		if (macvlan_real_dev == rpriv->netdev)
+			return true;
+		if (netif_is_bond_master(macvlan_real_dev))
 			return true;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 63b02048f9a781e01e59743cd2d3e0a1f10c971a Mon Sep 17 00:00:00 2001
From: Maor Dickman <maord@nvidia.com>
Date: Mon, 13 Mar 2023 22:42:34 -0700
Subject: net/mlx5e: Enable TC offload for egress MACVLAN over bond

Support offloading of TC rules that mirror/redirect egress traffic to a
MACVLAN device, which is attached to bond device which master mlx5 devices.

Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230314054234.267365-16-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c
index 07cc65596f89..291193f7120d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c
@@ -234,6 +234,9 @@ parse_mirred(struct mlx5e_tc_act_parse_state *parse_state,
 	if (mlx5_lag_mpesw_do_mirred(priv->mdev, out_dev, extack))
 		return -EOPNOTSUPP;
 
+	if (netif_is_macvlan(out_dev))
+		out_dev = macvlan_dev_real_dev(out_dev);
+
 	out_dev = get_fdb_out_dev(uplink_dev, out_dev);
 	if (!out_dev)
 		return -ENODEV;
@@ -250,9 +253,6 @@ parse_mirred(struct mlx5e_tc_act_parse_state *parse_state,
 			return err;
 	}
 
-	if (netif_is_macvlan(out_dev))
-		out_dev = macvlan_dev_real_dev(out_dev);
-
 	err = verify_uplink_forwarding(priv, attr, out_dev, extack);
 	if (err)
 		return err;
-- 
cgit v1.2.3-70-g09d2


From 8a8db7aeaa6dae4ddae033858bf5b49aecfbbe42 Mon Sep 17 00:00:00 2001
From: Wentao Jia <wentao.jia@corigine.com>
Date: Tue, 14 Mar 2023 08:36:05 +0200
Subject: nfp: flower: add get_flow_act_ct() for ct action

CT action is a special case different from other actions, CT clear action
is not required when get ct action, but this case is not considered.
If CT clear action in the flow rule, skip the CT clear action when get ct
action, return the first ct action that is not a CT clear action

Signed-off-by: Wentao Jia <wentao.jia@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/netronome/nfp/flower/conntrack.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index d23830b5bcb8..a54d374788e1 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -1656,6 +1656,22 @@ void nfp_fl_ct_clean_flow_entry(struct nfp_fl_ct_flow_entry *entry)
 	kfree(entry);
 }
 
+static struct flow_action_entry *get_flow_act_ct(struct flow_rule *rule)
+{
+	struct flow_action_entry *act;
+	int i;
+
+	/* More than one ct action may be present in a flow rule,
+	 * Return the first one that is not a CT clear action
+	 */
+	flow_action_for_each(i, act, &rule->action) {
+		if (act->id == FLOW_ACTION_CT && act->ct.action != TCA_CT_ACT_CLEAR)
+			return act;
+	}
+
+	return NULL;
+}
+
 static struct flow_action_entry *get_flow_act(struct flow_rule *rule,
 					      enum flow_action_id act_id)
 {
@@ -1720,7 +1736,7 @@ int nfp_fl_ct_handle_pre_ct(struct nfp_flower_priv *priv,
 	struct nfp_fl_ct_zone_entry *zt;
 	int err;
 
-	ct_act = get_flow_act(flow->rule, FLOW_ACTION_CT);
+	ct_act = get_flow_act_ct(flow->rule);
 	if (!ct_act) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "unsupported offload: Conntrack action empty in conntrack offload");
-- 
cgit v1.2.3-70-g09d2


From cee7b339d806f71df611087724ddfbaad2d8f8b7 Mon Sep 17 00:00:00 2001
From: Wentao Jia <wentao.jia@corigine.com>
Date: Tue, 14 Mar 2023 08:36:06 +0200
Subject: nfp: flower: refactor function "is_pre_ct_flow"

In the scenario of multiple ct zones, ct state key match and ct action
is present in one flow rule, the flow rule is classified to post_ct_flow
in design.

There is no ct state key match for pre ct flow, the judging condition
is added to function "is_pre_ct_flow".

Chain_index is another field for judging which flows are pre ct flow
If chain_index not 0, the flow is not pre ct flow.

Signed-off-by: Wentao Jia <wentao.jia@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/netronome/nfp/flower/conntrack.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index a54d374788e1..e0d6c8754272 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -55,9 +55,21 @@ static void *get_hashentry(struct rhashtable *ht, void *key,
 
 bool is_pre_ct_flow(struct flow_cls_offload *flow)
 {
+	struct flow_rule *rule = flow_cls_offload_flow_rule(flow);
+	struct flow_dissector *dissector = rule->match.dissector;
 	struct flow_action_entry *act;
+	struct flow_match_ct ct;
 	int i;
 
+	if (dissector->used_keys & BIT(FLOW_DISSECTOR_KEY_CT)) {
+		flow_rule_match_ct(rule, &ct);
+		if (ct.key->ct_state)
+			return false;
+	}
+
+	if (flow->common.chain_index)
+		return false;
+
 	flow_action_for_each(i, act, &flow->rule->action) {
 		if (act->id == FLOW_ACTION_CT) {
 			/* The pre_ct rule only have the ct or ct nat action, cannot
-- 
cgit v1.2.3-70-g09d2


From 0b8d953cce26287a7f5eb78ff49496e9836c88ed Mon Sep 17 00:00:00 2001
From: Wentao Jia <wentao.jia@corigine.com>
Date: Tue, 14 Mar 2023 08:36:07 +0200
Subject: nfp: flower: refactor function "is_post_ct_flow"

'ct_clear' action only or no ct action is supported for 'post_ct_flow'.
But in scenario of multiple ct zones, one non 'ct_clear' ct action or
more ct actions, including 'ct_clear action', may be present in one flow
rule. If ct state match key is 'ct_established', the flow rule is still
expected to be classified as 'post_ct_flow'. Check ct status first in
function "is_post_ct_flow" to achieve this.

Signed-off-by: Wentao Jia <wentao.jia@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/netronome/nfp/flower/conntrack.c  | 25 +++++++++++-----------
 1 file changed, 12 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index e0d6c8754272..6b90b922bac0 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -94,24 +94,23 @@ bool is_post_ct_flow(struct flow_cls_offload *flow)
 	struct flow_match_ct ct;
 	int i;
 
-	/* post ct entry cannot contains any ct action except ct_clear. */
-	flow_action_for_each(i, act, &flow->rule->action) {
-		if (act->id == FLOW_ACTION_CT) {
-			/* ignore ct clear action. */
-			if (act->ct.action == TCA_CT_ACT_CLEAR) {
-				exist_ct_clear = true;
-				continue;
-			}
-
-			return false;
-		}
-	}
-
 	if (dissector->used_keys & BIT(FLOW_DISSECTOR_KEY_CT)) {
 		flow_rule_match_ct(rule, &ct);
 		if (ct.key->ct_state & TCA_FLOWER_KEY_CT_FLAGS_ESTABLISHED)
 			return true;
 	} else {
+		/* post ct entry cannot contains any ct action except ct_clear. */
+		flow_action_for_each(i, act, &flow->rule->action) {
+			if (act->id == FLOW_ACTION_CT) {
+				/* ignore ct clear action. */
+				if (act->ct.action == TCA_CT_ACT_CLEAR) {
+					exist_ct_clear = true;
+					continue;
+				}
+
+				return false;
+			}
+		}
 		/* when do nat with ct, the post ct entry ignore the ct status,
 		 * will match the nat field(sip/dip) instead. In this situation,
 		 * the flow chain index is not zero and contains ct clear action.
-- 
cgit v1.2.3-70-g09d2


From 3e44d19934b92398785b3ffc2353b9eba264140e Mon Sep 17 00:00:00 2001
From: Wentao Jia <wentao.jia@corigine.com>
Date: Tue, 14 Mar 2023 08:36:08 +0200
Subject: nfp: flower: add goto_chain_index for ct entry

The chain_index has different means in pre ct entry and post ct entry.
In pre ct entry, it means chain index, but in post ct entry, it means
goto chain index, it is confused.

chain_index and goto_chain_index may be present in one flow rule, It
cannot be distinguished by one field chain_index, both chain_index
and goto_chain_index are required in the follow-up patch to support
multiple ct zones

Another field goto_chain_index is added to record the goto chain index.
If no goto action in post ct entry, goto_chain_index is 0.

Signed-off-by: Wentao Jia <wentao.jia@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/netronome/nfp/flower/conntrack.c | 8 ++++++--
 drivers/net/ethernet/netronome/nfp/flower/conntrack.h | 2 ++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index 6b90b922bac0..86ea8cbc67a2 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -1254,7 +1254,7 @@ static int nfp_ct_do_tc_merge(struct nfp_fl_ct_zone_entry *zt,
 	/* Checks that the chain_index of the filter matches the
 	 * chain_index of the GOTO action.
 	 */
-	if (post_ct_entry->chain_index != pre_ct_entry->chain_index)
+	if (post_ct_entry->chain_index != pre_ct_entry->goto_chain_index)
 		return -EINVAL;
 
 	err = nfp_ct_merge_check(pre_ct_entry, post_ct_entry);
@@ -1783,7 +1783,8 @@ int nfp_fl_ct_handle_pre_ct(struct nfp_flower_priv *priv,
 	if (IS_ERR(ct_entry))
 		return PTR_ERR(ct_entry);
 	ct_entry->type = CT_TYPE_PRE_CT;
-	ct_entry->chain_index = ct_goto->chain_index;
+	ct_entry->chain_index = flow->common.chain_index;
+	ct_entry->goto_chain_index = ct_goto->chain_index;
 	list_add(&ct_entry->list_node, &zt->pre_ct_list);
 	zt->pre_ct_count++;
 
@@ -1806,6 +1807,7 @@ int nfp_fl_ct_handle_post_ct(struct nfp_flower_priv *priv,
 	struct nfp_fl_ct_zone_entry *zt;
 	bool wildcarded = false;
 	struct flow_match_ct ct;
+	struct flow_action_entry *ct_goto;
 
 	flow_rule_match_ct(rule, &ct);
 	if (!ct.mask->ct_zone) {
@@ -1830,6 +1832,8 @@ int nfp_fl_ct_handle_post_ct(struct nfp_flower_priv *priv,
 
 	ct_entry->type = CT_TYPE_POST_CT;
 	ct_entry->chain_index = flow->common.chain_index;
+	ct_goto = get_flow_act(flow->rule, FLOW_ACTION_GOTO);
+	ct_entry->goto_chain_index = ct_goto ? ct_goto->chain_index : 0;
 	list_add(&ct_entry->list_node, &zt->post_ct_list);
 	zt->post_ct_count++;
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.h b/drivers/net/ethernet/netronome/nfp/flower/conntrack.h
index 762c0b36e269..9440ab776ece 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.h
@@ -112,6 +112,7 @@ enum nfp_nfp_layer_name {
  * @cookie:	Flow cookie, same as original TC flow, used as key
  * @list_node:	Used by the list
  * @chain_index:	Chain index of the original flow
+ * @goto_chain_index:	goto chain index of the flow
  * @netdev:	netdev structure.
  * @type:	Type of pre-entry from enum ct_entry_type
  * @zt:		Reference to the zone table this belongs to
@@ -125,6 +126,7 @@ struct nfp_fl_ct_flow_entry {
 	unsigned long cookie;
 	struct list_head list_node;
 	u32 chain_index;
+	u32 goto_chain_index;
 	enum ct_entry_type type;
 	struct net_device *netdev;
 	struct nfp_fl_ct_zone_entry *zt;
-- 
cgit v1.2.3-70-g09d2


From 46a83c85b683b647d86e679e2b095494e87f4d7c Mon Sep 17 00:00:00 2001
From: Wentao Jia <wentao.jia@corigine.com>
Date: Tue, 14 Mar 2023 08:36:09 +0200
Subject: nfp: flower: prepare for parameterisation of number of offload rules

The fixed number of offload flow rule is only supported scenario of one
ct zone, in the scenario of multiple ct zones, dynamic number and more
number of offload flow rules are required. In order to support scenario
of multiple ct zones, parameter num_rules is added for to offload flow
rules

Signed-off-by: Wentao Jia <wentao.jia@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/netronome/nfp/flower/conntrack.c  | 54 +++++++++++-----------
 1 file changed, 28 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index 86ea8cbc67a2..ecffb6b0f3a1 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -693,34 +693,34 @@ static void nfp_fl_get_csum_flag(struct flow_action_entry *a_in, u8 ip_proto, u3
 static int nfp_fl_merge_actions_offload(struct flow_rule **rules,
 					struct nfp_flower_priv *priv,
 					struct net_device *netdev,
-					struct nfp_fl_payload *flow_pay)
+					struct nfp_fl_payload *flow_pay,
+					int num_rules)
 {
 	enum flow_action_hw_stats tmp_stats = FLOW_ACTION_HW_STATS_DONT_CARE;
 	struct flow_action_entry *a_in;
-	int i, j, num_actions, id;
+	int i, j, id, num_actions = 0;
 	struct flow_rule *a_rule;
 	int err = 0, offset = 0;
 
-	num_actions = rules[CT_TYPE_PRE_CT]->action.num_entries +
-		      rules[CT_TYPE_NFT]->action.num_entries +
-		      rules[CT_TYPE_POST_CT]->action.num_entries;
+	for (i = 0; i < num_rules; i++)
+		num_actions += rules[i]->action.num_entries;
 
 	/* Add one action to make sure there is enough room to add an checksum action
 	 * when do nat.
 	 */
-	a_rule = flow_rule_alloc(num_actions + 1);
+	a_rule = flow_rule_alloc(num_actions + (num_rules / 2));
 	if (!a_rule)
 		return -ENOMEM;
 
-	/* Actions need a BASIC dissector. */
-	a_rule->match = rules[CT_TYPE_PRE_CT]->match;
 	/* post_ct entry have one action at least. */
-	if (rules[CT_TYPE_POST_CT]->action.num_entries != 0) {
-		tmp_stats = rules[CT_TYPE_POST_CT]->action.entries[0].hw_stats;
-	}
+	if (rules[num_rules - 1]->action.num_entries != 0)
+		tmp_stats = rules[num_rules - 1]->action.entries[0].hw_stats;
+
+	/* Actions need a BASIC dissector. */
+	a_rule->match = rules[0]->match;
 
 	/* Copy actions */
-	for (j = 0; j < _CT_TYPE_MAX; j++) {
+	for (j = 0; j < num_rules; j++) {
 		u32 csum_updated = 0;
 		u8 ip_proto = 0;
 
@@ -758,8 +758,9 @@ static int nfp_fl_merge_actions_offload(struct flow_rule **rules,
 				/* nft entry is generated by tc ct, which mangle action do not care
 				 * the stats, inherit the post entry stats to meet the
 				 * flow_action_hw_stats_check.
+				 * nft entry flow rules are at odd array index.
 				 */
-				if (j == CT_TYPE_NFT) {
+				if (j & 0x01) {
 					if (a_in->hw_stats == FLOW_ACTION_HW_STATS_DONT_CARE)
 						a_in->hw_stats = tmp_stats;
 					nfp_fl_get_csum_flag(a_in, ip_proto, &csum_updated);
@@ -801,6 +802,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 	struct nfp_fl_payload *flow_pay;
 
 	struct flow_rule *rules[_CT_TYPE_MAX];
+	int num_rules = _CT_TYPE_MAX;
 	u8 *key, *msk, *kdata, *mdata;
 	struct nfp_port *port = NULL;
 	struct net_device *netdev;
@@ -820,7 +822,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 	memset(&key_map, 0, sizeof(key_map));
 
 	/* Calculate the resultant key layer and size for offload */
-	for (i = 0; i < _CT_TYPE_MAX; i++) {
+	for (i = 0; i < num_rules; i++) {
 		err = nfp_flower_calculate_key_layers(priv->app,
 						      m_entry->netdev,
 						      &tmp_layer, rules[i],
@@ -886,7 +888,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 	 * that the layer is not present.
 	 */
 	if (!qinq_sup) {
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
+		for (i = 0; i < num_rules; i++) {
 			offset = key_map[FLOW_PAY_META_TCI];
 			key = kdata + offset;
 			msk = mdata + offset;
@@ -900,7 +902,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		offset = key_map[FLOW_PAY_MAC_MPLS];
 		key = kdata + offset;
 		msk = mdata + offset;
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
+		for (i = 0; i < num_rules; i++) {
 			nfp_flower_compile_mac((struct nfp_flower_mac_mpls *)key,
 					       (struct nfp_flower_mac_mpls *)msk,
 					       rules[i]);
@@ -916,7 +918,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		offset = key_map[FLOW_PAY_IPV4];
 		key = kdata + offset;
 		msk = mdata + offset;
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
+		for (i = 0; i < num_rules; i++) {
 			nfp_flower_compile_ipv4((struct nfp_flower_ipv4 *)key,
 						(struct nfp_flower_ipv4 *)msk,
 						rules[i]);
@@ -927,7 +929,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		offset = key_map[FLOW_PAY_IPV6];
 		key = kdata + offset;
 		msk = mdata + offset;
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
+		for (i = 0; i < num_rules; i++) {
 			nfp_flower_compile_ipv6((struct nfp_flower_ipv6 *)key,
 						(struct nfp_flower_ipv6 *)msk,
 						rules[i]);
@@ -938,7 +940,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		offset = key_map[FLOW_PAY_L4];
 		key = kdata + offset;
 		msk = mdata + offset;
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
+		for (i = 0; i < num_rules; i++) {
 			nfp_flower_compile_tport((struct nfp_flower_tp_ports *)key,
 						 (struct nfp_flower_tp_ports *)msk,
 						 rules[i]);
@@ -949,7 +951,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		offset = key_map[FLOW_PAY_QINQ];
 		key = kdata + offset;
 		msk = mdata + offset;
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
+		for (i = 0; i < num_rules; i++) {
 			nfp_flower_compile_vlan((struct nfp_flower_vlan *)key,
 						(struct nfp_flower_vlan *)msk,
 						rules[i]);
@@ -965,7 +967,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 			struct nfp_ipv6_addr_entry *entry;
 			struct in6_addr *dst;
 
-			for (i = 0; i < _CT_TYPE_MAX; i++) {
+			for (i = 0; i < num_rules; i++) {
 				nfp_flower_compile_ipv6_gre_tun((void *)key,
 								(void *)msk, rules[i]);
 			}
@@ -982,7 +984,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		} else {
 			__be32 dst;
 
-			for (i = 0; i < _CT_TYPE_MAX; i++) {
+			for (i = 0; i < num_rules; i++) {
 				nfp_flower_compile_ipv4_gre_tun((void *)key,
 								(void *)msk, rules[i]);
 			}
@@ -1006,7 +1008,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 			struct nfp_ipv6_addr_entry *entry;
 			struct in6_addr *dst;
 
-			for (i = 0; i < _CT_TYPE_MAX; i++) {
+			for (i = 0; i < num_rules; i++) {
 				nfp_flower_compile_ipv6_udp_tun((void *)key,
 								(void *)msk, rules[i]);
 			}
@@ -1023,7 +1025,7 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		} else {
 			__be32 dst;
 
-			for (i = 0; i < _CT_TYPE_MAX; i++) {
+			for (i = 0; i < num_rules; i++) {
 				nfp_flower_compile_ipv4_udp_tun((void *)key,
 								(void *)msk, rules[i]);
 			}
@@ -1040,13 +1042,13 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 			offset = key_map[FLOW_PAY_GENEVE_OPT];
 			key = kdata + offset;
 			msk = mdata + offset;
-			for (i = 0; i < _CT_TYPE_MAX; i++)
+			for (i = 0; i < num_rules; i++)
 				nfp_flower_compile_geneve_opt(key, msk, rules[i]);
 		}
 	}
 
 	/* Merge actions into flow_pay */
-	err = nfp_fl_merge_actions_offload(rules, priv, netdev, flow_pay);
+	err = nfp_fl_merge_actions_offload(rules, priv, netdev, flow_pay, num_rules);
 	if (err)
 		goto ct_offload_err;
 
-- 
cgit v1.2.3-70-g09d2


From a87ceb3d42afebd86191054314ef1a2f98c4556b Mon Sep 17 00:00:00 2001
From: Wentao Jia <wentao.jia@corigine.com>
Date: Tue, 14 Mar 2023 08:36:10 +0200
Subject: nfp: flower: offload tc flows of multiple conntrack zones

If goto_chain action present in the post ct flow rule, merge flow rules
in this ct-zone, create a new pre_ct entry as the pre ct flow rule of
next ct-zone, but do not offload merged flow rules to firmware. Repeat
the process in the next ct-zone until no goto_chain action present in
the post ct flow rule in a certain ct-zone, merged all the flow rules.
Offload to firmware finally.

Signed-off-by: Wentao Jia <wentao.jia@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/netronome/nfp/flower/conntrack.c  | 145 ++++++++++++++++++---
 .../net/ethernet/netronome/nfp/flower/conntrack.h  |  30 ++++-
 .../net/ethernet/netronome/nfp/flower/offload.c    |   2 +-
 3 files changed, 154 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index ecffb6b0f3a1..73032173ac4e 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -522,6 +522,21 @@ static int nfp_ct_check_vlan_merge(struct flow_action_entry *a_in,
 	return 0;
 }
 
+/* Extra check for multiple ct-zones merge
+ * currently surpport nft entries merge check in different zones
+ */
+static int nfp_ct_merge_extra_check(struct nfp_fl_ct_flow_entry *nft_entry,
+				    struct nfp_fl_ct_tc_merge *tc_m_entry)
+{
+	struct nfp_fl_nft_tc_merge *prev_nft_m_entry;
+	struct nfp_fl_ct_flow_entry *pre_ct_entry;
+
+	pre_ct_entry = tc_m_entry->pre_ct_parent;
+	prev_nft_m_entry = pre_ct_entry->prev_m_entries[pre_ct_entry->num_prev_m_entries - 1];
+
+	return nfp_ct_merge_check(prev_nft_m_entry->nft_parent, nft_entry);
+}
+
 static int nfp_ct_merge_act_check(struct nfp_fl_ct_flow_entry *pre_ct_entry,
 				  struct nfp_fl_ct_flow_entry *post_ct_entry,
 				  struct nfp_fl_ct_flow_entry *nft_entry)
@@ -796,27 +811,34 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 {
 	enum nfp_flower_tun_type tun_type = NFP_FL_TUNNEL_NONE;
 	struct nfp_fl_ct_zone_entry *zt = m_entry->zt;
+	struct flow_rule *rules[NFP_MAX_ENTRY_RULES];
+	struct nfp_fl_ct_flow_entry *pre_ct_entry;
 	struct nfp_fl_key_ls key_layer, tmp_layer;
 	struct nfp_flower_priv *priv = zt->priv;
 	u16 key_map[_FLOW_PAY_LAYERS_MAX];
 	struct nfp_fl_payload *flow_pay;
-
-	struct flow_rule *rules[_CT_TYPE_MAX];
-	int num_rules = _CT_TYPE_MAX;
 	u8 *key, *msk, *kdata, *mdata;
 	struct nfp_port *port = NULL;
+	int num_rules, err, i, j = 0;
 	struct net_device *netdev;
 	bool qinq_sup;
 	u32 port_id;
 	u16 offset;
-	int i, err;
 
 	netdev = m_entry->netdev;
 	qinq_sup = !!(priv->flower_ext_feats & NFP_FL_FEATS_VLAN_QINQ);
 
-	rules[CT_TYPE_PRE_CT] = m_entry->tc_m_parent->pre_ct_parent->rule;
-	rules[CT_TYPE_NFT] = m_entry->nft_parent->rule;
-	rules[CT_TYPE_POST_CT] = m_entry->tc_m_parent->post_ct_parent->rule;
+	pre_ct_entry = m_entry->tc_m_parent->pre_ct_parent;
+	num_rules = pre_ct_entry->num_prev_m_entries * 2 + _CT_TYPE_MAX;
+
+	for (i = 0; i < pre_ct_entry->num_prev_m_entries; i++) {
+		rules[j++] = pre_ct_entry->prev_m_entries[i]->tc_m_parent->pre_ct_parent->rule;
+		rules[j++] = pre_ct_entry->prev_m_entries[i]->nft_parent->rule;
+	}
+
+	rules[j++] = m_entry->tc_m_parent->pre_ct_parent->rule;
+	rules[j++] = m_entry->nft_parent->rule;
+	rules[j++] = m_entry->tc_m_parent->post_ct_parent->rule;
 
 	memset(&key_layer, 0, sizeof(struct nfp_fl_key_ls));
 	memset(&key_map, 0, sizeof(key_map));
@@ -1181,6 +1203,12 @@ static int nfp_ct_do_nft_merge(struct nfp_fl_ct_zone_entry *zt,
 	if (err)
 		return err;
 
+	if (pre_ct_entry->num_prev_m_entries > 0) {
+		err = nfp_ct_merge_extra_check(nft_entry, tc_m_entry);
+		if (err)
+			return err;
+	}
+
 	/* Combine tc_merge and nft cookies for this cookie. */
 	new_cookie[0] = tc_m_entry->cookie[0];
 	new_cookie[1] = tc_m_entry->cookie[1];
@@ -1211,11 +1239,6 @@ static int nfp_ct_do_nft_merge(struct nfp_fl_ct_zone_entry *zt,
 	list_add(&nft_m_entry->tc_merge_list, &tc_m_entry->children);
 	list_add(&nft_m_entry->nft_flow_list, &nft_entry->children);
 
-	/* Generate offload structure and send to nfp */
-	err = nfp_fl_ct_add_offload(nft_m_entry);
-	if (err)
-		goto err_nft_ct_offload;
-
 	err = rhashtable_insert_fast(&zt->nft_merge_tb, &nft_m_entry->hash_node,
 				     nfp_nft_ct_merge_params);
 	if (err)
@@ -1223,12 +1246,20 @@ static int nfp_ct_do_nft_merge(struct nfp_fl_ct_zone_entry *zt,
 
 	zt->nft_merge_count++;
 
+	if (post_ct_entry->goto_chain_index > 0)
+		return nfp_fl_create_new_pre_ct(nft_m_entry);
+
+	/* Generate offload structure and send to nfp */
+	err = nfp_fl_ct_add_offload(nft_m_entry);
+	if (err)
+		goto err_nft_ct_offload;
+
 	return err;
 
-err_nft_ct_merge_insert:
+err_nft_ct_offload:
 	nfp_fl_ct_del_offload(zt->priv->app, nft_m_entry->tc_flower_cookie,
 			      nft_m_entry->netdev);
-err_nft_ct_offload:
+err_nft_ct_merge_insert:
 	list_del(&nft_m_entry->tc_merge_list);
 	list_del(&nft_m_entry->nft_flow_list);
 	kfree(nft_m_entry);
@@ -1474,7 +1505,7 @@ nfp_fl_ct_flow_entry *nfp_fl_ct_add_flow(struct nfp_fl_ct_zone_entry *zt,
 
 	entry->zt = zt;
 	entry->netdev = netdev;
-	entry->cookie = flow->cookie;
+	entry->cookie = flow->cookie > 0 ? flow->cookie : (unsigned long)entry;
 	entry->chain_index = flow->common.chain_index;
 	entry->tun_offset = NFP_FL_CT_NO_TUN;
 
@@ -1514,6 +1545,9 @@ nfp_fl_ct_flow_entry *nfp_fl_ct_add_flow(struct nfp_fl_ct_zone_entry *zt,
 
 	INIT_LIST_HEAD(&entry->children);
 
+	if (flow->cookie == 0)
+		return entry;
+
 	/* Now add a ct map entry to flower-priv */
 	map = get_hashentry(&zt->priv->ct_map_table, &flow->cookie,
 			    nfp_ct_map_params, sizeof(*map));
@@ -1572,6 +1606,14 @@ static void cleanup_nft_merge_entry(struct nfp_fl_nft_tc_merge *m_entry)
 	list_del(&m_entry->tc_merge_list);
 	list_del(&m_entry->nft_flow_list);
 
+	if (m_entry->next_pre_ct_entry) {
+		struct nfp_fl_ct_map_entry pre_ct_map_ent;
+
+		pre_ct_map_ent.ct_entry = m_entry->next_pre_ct_entry;
+		pre_ct_map_ent.cookie = 0;
+		nfp_fl_ct_del_flow(&pre_ct_map_ent);
+	}
+
 	kfree(m_entry);
 }
 
@@ -1742,7 +1784,8 @@ nfp_ct_merge_nft_with_tc(struct nfp_fl_ct_flow_entry *nft_entry,
 int nfp_fl_ct_handle_pre_ct(struct nfp_flower_priv *priv,
 			    struct net_device *netdev,
 			    struct flow_cls_offload *flow,
-			    struct netlink_ext_ack *extack)
+			    struct netlink_ext_ack *extack,
+			    struct nfp_fl_nft_tc_merge *m_entry)
 {
 	struct flow_action_entry *ct_act, *ct_goto;
 	struct nfp_fl_ct_flow_entry *ct_entry;
@@ -1787,6 +1830,20 @@ int nfp_fl_ct_handle_pre_ct(struct nfp_flower_priv *priv,
 	ct_entry->type = CT_TYPE_PRE_CT;
 	ct_entry->chain_index = flow->common.chain_index;
 	ct_entry->goto_chain_index = ct_goto->chain_index;
+
+	if (m_entry) {
+		struct nfp_fl_ct_flow_entry *pre_ct_entry;
+		int i;
+
+		pre_ct_entry = m_entry->tc_m_parent->pre_ct_parent;
+		for (i = 0; i < pre_ct_entry->num_prev_m_entries; i++)
+			ct_entry->prev_m_entries[i] = pre_ct_entry->prev_m_entries[i];
+		ct_entry->prev_m_entries[i++] = m_entry;
+		ct_entry->num_prev_m_entries = i;
+
+		m_entry->next_pre_ct_entry = ct_entry;
+	}
+
 	list_add(&ct_entry->list_node, &zt->pre_ct_list);
 	zt->pre_ct_count++;
 
@@ -1864,6 +1921,28 @@ int nfp_fl_ct_handle_post_ct(struct nfp_flower_priv *priv,
 	return 0;
 }
 
+int nfp_fl_create_new_pre_ct(struct nfp_fl_nft_tc_merge *m_entry)
+{
+	struct nfp_fl_ct_flow_entry *pre_ct_entry, *post_ct_entry;
+	struct flow_cls_offload new_pre_ct_flow;
+	int err;
+
+	pre_ct_entry = m_entry->tc_m_parent->pre_ct_parent;
+	if (pre_ct_entry->num_prev_m_entries >= NFP_MAX_RECIRC_CT_ZONES - 1)
+		return -1;
+
+	post_ct_entry = m_entry->tc_m_parent->post_ct_parent;
+	memset(&new_pre_ct_flow, 0, sizeof(struct flow_cls_offload));
+	new_pre_ct_flow.rule = post_ct_entry->rule;
+	new_pre_ct_flow.common.chain_index = post_ct_entry->chain_index;
+
+	err = nfp_fl_ct_handle_pre_ct(pre_ct_entry->zt->priv,
+				      pre_ct_entry->netdev,
+				      &new_pre_ct_flow, NULL,
+				      m_entry);
+	return err;
+}
+
 static void
 nfp_fl_ct_sub_stats(struct nfp_fl_nft_tc_merge *nft_merge,
 		    enum ct_entry_type type, u64 *m_pkts,
@@ -1909,6 +1988,32 @@ nfp_fl_ct_sub_stats(struct nfp_fl_nft_tc_merge *nft_merge,
 				  0, priv->stats[ctx_id].used,
 				  FLOW_ACTION_HW_STATS_DELAYED);
 	}
+
+	/* Update previous pre_ct/post_ct/nft flow stats */
+	if (nft_merge->tc_m_parent->pre_ct_parent->num_prev_m_entries > 0) {
+		struct nfp_fl_nft_tc_merge *tmp_nft_merge;
+		int i;
+
+		for (i = 0; i < nft_merge->tc_m_parent->pre_ct_parent->num_prev_m_entries; i++) {
+			tmp_nft_merge = nft_merge->tc_m_parent->pre_ct_parent->prev_m_entries[i];
+			flow_stats_update(&tmp_nft_merge->tc_m_parent->pre_ct_parent->stats,
+					  priv->stats[ctx_id].bytes,
+					  priv->stats[ctx_id].pkts,
+					  0, priv->stats[ctx_id].used,
+					  FLOW_ACTION_HW_STATS_DELAYED);
+			flow_stats_update(&tmp_nft_merge->tc_m_parent->post_ct_parent->stats,
+					  priv->stats[ctx_id].bytes,
+					  priv->stats[ctx_id].pkts,
+					  0, priv->stats[ctx_id].used,
+					  FLOW_ACTION_HW_STATS_DELAYED);
+			flow_stats_update(&tmp_nft_merge->nft_parent->stats,
+					  priv->stats[ctx_id].bytes,
+					  priv->stats[ctx_id].pkts,
+					  0, priv->stats[ctx_id].used,
+					  FLOW_ACTION_HW_STATS_DELAYED);
+		}
+	}
+
 	/* Reset stats from the nfp */
 	priv->stats[ctx_id].pkts = 0;
 	priv->stats[ctx_id].bytes = 0;
@@ -2113,10 +2218,12 @@ int nfp_fl_ct_del_flow(struct nfp_fl_ct_map_entry *ct_map_ent)
 	switch (ct_entry->type) {
 	case CT_TYPE_PRE_CT:
 		zt->pre_ct_count--;
-		rhashtable_remove_fast(m_table, &ct_map_ent->hash_node,
-				       nfp_ct_map_params);
+		if (ct_map_ent->cookie > 0)
+			rhashtable_remove_fast(m_table, &ct_map_ent->hash_node,
+					       nfp_ct_map_params);
 		nfp_fl_ct_clean_flow_entry(ct_entry);
-		kfree(ct_map_ent);
+		if (ct_map_ent->cookie > 0)
+			kfree(ct_map_ent);
 
 		if (!zt->pre_ct_count) {
 			zt->nft = NULL;
diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.h b/drivers/net/ethernet/netronome/nfp/flower/conntrack.h
index 9440ab776ece..c4ec78358033 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.h
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.h
@@ -86,6 +86,9 @@ enum ct_entry_type {
 	_CT_TYPE_MAX,
 };
 
+#define NFP_MAX_RECIRC_CT_ZONES 4
+#define NFP_MAX_ENTRY_RULES  (NFP_MAX_RECIRC_CT_ZONES * 2 + 1)
+
 enum nfp_nfp_layer_name {
 	FLOW_PAY_META_TCI =    0,
 	FLOW_PAY_INPORT,
@@ -114,27 +117,31 @@ enum nfp_nfp_layer_name {
  * @chain_index:	Chain index of the original flow
  * @goto_chain_index:	goto chain index of the flow
  * @netdev:	netdev structure.
- * @type:	Type of pre-entry from enum ct_entry_type
  * @zt:		Reference to the zone table this belongs to
  * @children:	List of tc_merge flows this flow forms part of
  * @rule:	Reference to the original TC flow rule
  * @stats:	Used to cache stats for updating
+ * @prev_m_entries:	Array of all previous nft_tc_merge entries
+ * @num_prev_m_entries:	The number of all previous nft_tc_merge entries
  * @tun_offset: Used to indicate tunnel action offset in action list
  * @flags:	Used to indicate flow flag like NAT which used by merge.
+ * @type:	Type of ct-entry from enum ct_entry_type
  */
 struct nfp_fl_ct_flow_entry {
 	unsigned long cookie;
 	struct list_head list_node;
 	u32 chain_index;
 	u32 goto_chain_index;
-	enum ct_entry_type type;
 	struct net_device *netdev;
 	struct nfp_fl_ct_zone_entry *zt;
 	struct list_head children;
 	struct flow_rule *rule;
 	struct flow_stats stats;
+	struct nfp_fl_nft_tc_merge *prev_m_entries[NFP_MAX_RECIRC_CT_ZONES - 1];
+	u8 num_prev_m_entries;
 	u8 tun_offset;		// Set to NFP_FL_CT_NO_TUN if no tun
 	u8 flags;
+	u8 type;
 };
 
 /**
@@ -171,6 +178,7 @@ struct nfp_fl_ct_tc_merge {
  * @nft_parent:	The nft_entry parent
  * @tc_flower_cookie:	The cookie of the flow offloaded to the nfp
  * @flow_pay:	Reference to the offloaded flow struct
+ * @next_pre_ct_entry:	Reference to the next ct zone pre ct entry
  */
 struct nfp_fl_nft_tc_merge {
 	struct net_device *netdev;
@@ -183,6 +191,7 @@ struct nfp_fl_nft_tc_merge {
 	struct nfp_fl_ct_flow_entry *nft_parent;
 	unsigned long tc_flower_cookie;
 	struct nfp_fl_payload *flow_pay;
+	struct nfp_fl_ct_flow_entry *next_pre_ct_entry;
 };
 
 /**
@@ -206,6 +215,7 @@ bool is_post_ct_flow(struct flow_cls_offload *flow);
  * @netdev:	netdev structure.
  * @flow:	TC flower classifier offload structure.
  * @extack:	Extack pointer for errors
+ * @m_entry:previous nfp_fl_nft_tc_merge entry
  *
  * Adds a new entry to the relevant zone table and tries to
  * merge with other +trk+est entries and offload if possible.
@@ -215,7 +225,8 @@ bool is_post_ct_flow(struct flow_cls_offload *flow);
 int nfp_fl_ct_handle_pre_ct(struct nfp_flower_priv *priv,
 			    struct net_device *netdev,
 			    struct flow_cls_offload *flow,
-			    struct netlink_ext_ack *extack);
+			    struct netlink_ext_ack *extack,
+			    struct nfp_fl_nft_tc_merge *m_entry);
 /**
  * nfp_fl_ct_handle_post_ct() - Handles +trk+est conntrack rules
  * @priv:	Pointer to app priv
@@ -233,6 +244,19 @@ int nfp_fl_ct_handle_post_ct(struct nfp_flower_priv *priv,
 			     struct flow_cls_offload *flow,
 			     struct netlink_ext_ack *extack);
 
+/**
+ * nfp_fl_create_new_pre_ct() - create next ct_zone -trk conntrack rules
+ * @m_entry:previous nfp_fl_nft_tc_merge entry
+ *
+ * Create a new pre_ct entry from previous nfp_fl_nft_tc_merge entry
+ * to the next relevant zone table. Try to merge with other +trk+est
+ * entries and offload if possible. The created new pre_ct entry is
+ * linked to the previous nfp_fl_nft_tc_merge entry.
+ *
+ * Return: negative value on error, 0 if configured successfully.
+ */
+int nfp_fl_create_new_pre_ct(struct nfp_fl_nft_tc_merge *m_entry);
+
 /**
  * nfp_fl_ct_clean_flow_entry() - Free a nfp_fl_ct_flow_entry
  * @entry:	Flow entry to cleanup
diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index 8593cafa6368..18328eb7f5c3 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -1344,7 +1344,7 @@ nfp_flower_add_offload(struct nfp_app *app, struct net_device *netdev,
 		port = nfp_port_from_netdev(netdev);
 
 	if (is_pre_ct_flow(flow))
-		return nfp_fl_ct_handle_pre_ct(priv, netdev, flow, extack);
+		return nfp_fl_ct_handle_pre_ct(priv, netdev, flow, extack, NULL);
 
 	if (is_post_ct_flow(flow))
 		return nfp_fl_ct_handle_post_ct(priv, netdev, flow, extack);
-- 
cgit v1.2.3-70-g09d2


From 9bdf4489a3950335af088604876fdc3209a9dd8e Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Tue, 14 Mar 2023 08:23:15 +0000
Subject: net: phy: micrel: Fix spelling mistake "minimim" -> "minimum"

There is a spelling mistake in a pr_warn_ratelimited message. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20230314082315.26532-1-colin.i.king@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/micrel.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 0cf0ef3a96a3..e26c6723caa4 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -4106,7 +4106,7 @@ static int lan8841_ptp_perout(struct ptp_clock_info *ptp,
 	period_nsec = timespec64_to_ns(&ts_period);
 
 	if (period_nsec < 200) {
-		pr_warn_ratelimited("%s: perout period too small, minimim is 200 nsec\n",
+		pr_warn_ratelimited("%s: perout period too small, minimum is 200 nsec\n",
 				    phydev_name(phydev));
 		return -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3-70-g09d2


From 499183cc3b52613f06cf4ce70809546971c96ed8 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Wed, 15 Mar 2023 20:28:29 -0700
Subject: wifi: iwlwifi: Avoid disabling GCC specific flag with clang

Clang errors:

  drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c:15:32: error: unknown warning group '-Wsuggest-attribute=format', ignored [-Werror,-Wunknown-warning-option]
  #pragma GCC diagnostic ignored "-Wsuggest-attribute=format"
                                 ^
  1 error generated.

The warning being disabled by this pragma is GCC specific. Guard its use
with CONFIG_CC_IS_GCC so that it is not used with clang to clear up the
error.

Fixes: 4eca8cbf7ba8 ("wifi: iwlwifi: suppress printf warnings in tracing")
Link: https://github.com/ClangBuiltLinux/linux/issues/1818
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20230315-iwlwifi-fix-pragma-v1-1-ad23f92c4739@kernel.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c
index c190ec5effa1..e46639b097f4 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.c
@@ -12,7 +12,9 @@
 #include "iwl-trans.h"
 
 #define CREATE_TRACE_POINTS
+#ifdef CONFIG_CC_IS_GCC
 #pragma GCC diagnostic ignored "-Wsuggest-attribute=format"
+#endif
 #include "iwl-devtrace.h"
 
 EXPORT_TRACEPOINT_SYMBOL(iwlwifi_dev_ucode_event);
-- 
cgit v1.2.3-70-g09d2


From ed01385c0d78a025bdc72128b7aa7c3309cd5852 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 15 Mar 2023 17:07:25 -0700
Subject: selftests/bpf: Use ASSERT_EQ instead ASSERT_OK for testing memcmp
 result

In tcp_hdr_options test, it ensures the received tcp hdr option
and the sk local storage have the expected values. It uses memcmp
to check that. Testing the memcmp result with ASSERT_OK is confusing
because ASSERT_OK will print out the errno which is not set.
This patch uses ASSERT_EQ to check for 0 instead.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20230316000726.1016773-1-martin.lau@linux.dev
---
 tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c b/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c
index 5cf85d0f9827..13bcaeb028b8 100644
--- a/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c
+++ b/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c
@@ -151,7 +151,7 @@ static int check_hdr_opt(const struct bpf_test_option *exp,
 			 const struct bpf_test_option *act,
 			 const char *hdr_desc)
 {
-	if (!ASSERT_OK(memcmp(exp, act, sizeof(*exp)), hdr_desc)) {
+	if (!ASSERT_EQ(memcmp(exp, act, sizeof(*exp)), 0, hdr_desc)) {
 		print_option(exp, "expected: ");
 		print_option(act, "  actual: ");
 		return -1;
@@ -169,7 +169,7 @@ static int check_hdr_stg(const struct hdr_stg *exp, int fd,
 		  "map_lookup(hdr_stg_map_fd)"))
 		return -1;
 
-	if (!ASSERT_OK(memcmp(exp, &act, sizeof(*exp)), stg_desc)) {
+	if (!ASSERT_EQ(memcmp(exp, &act, sizeof(*exp)), 0, stg_desc)) {
 		print_hdr_stg(exp, "expected: ");
 		print_hdr_stg(&act, "  actual: ");
 		return -1;
-- 
cgit v1.2.3-70-g09d2


From 226efec2b0efad60d4a6c4b2c3a8710dafc4dc21 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 15 Mar 2023 17:07:26 -0700
Subject: selftests/bpf: Fix a fd leak in an error path in network_helpers.c

In __start_server, it leaks a fd when setsockopt(SO_REUSEPORT) fails.
This patch fixes it.

Fixes: eed92afdd14c ("bpf: selftest: Test batching and bpf_(get|set)sockopt in bpf tcp iter")
Reported-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20230316000726.1016773-2-martin.lau@linux.dev
---
 tools/testing/selftests/bpf/network_helpers.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 01de33191226..596caa176582 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -95,7 +95,7 @@ static int __start_server(int type, int protocol, const struct sockaddr *addr,
 	if (reuseport &&
 	    setsockopt(fd, SOL_SOCKET, SO_REUSEPORT, &on, sizeof(on))) {
 		log_err("Failed to set SO_REUSEPORT");
-		return -1;
+		goto error_close;
 	}
 
 	if (bind(fd, addr, addrlen) < 0) {
-- 
cgit v1.2.3-70-g09d2


From 6cb9430be1471d631e1b6b138e6d26657a9caa81 Mon Sep 17 00:00:00 2001
From: Daniel Müller <deso@posteo.net>
Date: Wed, 15 Mar 2023 17:15:50 +0000
Subject: libbpf: Ignore warnings about "inefficient alignment"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some consumers of libbpf compile the code base with different warnings
enabled. In a report for perf, for example, -Wpacked was set which
caused warnings about "inefficient alignment" to be emitted on a subset
of supported architectures.

With this change we silence specifically those warnings, as we intentionally
worked with packed structs.

This is a similar resolution as in b2f10cd4e805 ("perf cpumap: Fix alignment
for masks in event encoding").

Fixes: 1eebcb60633f ("libbpf: Implement basic zip archive parsing support")
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/bpf/CA+G9fYtBnwxAWXi2+GyNByApxnf_DtP1-6+_zOKAdJKnJBexjg@mail.gmail.com/
Link: https://lore.kernel.org/bpf/20230315171550.1551603-1-deso@posteo.net
---
 tools/lib/bpf/zip.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/lib/bpf/zip.c b/tools/lib/bpf/zip.c
index f561aa07438f..3f26d629b2b4 100644
--- a/tools/lib/bpf/zip.c
+++ b/tools/lib/bpf/zip.c
@@ -16,6 +16,10 @@
 #include "libbpf_internal.h"
 #include "zip.h"
 
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wpacked"
+#pragma GCC diagnostic ignored "-Wattributes"
+
 /* Specification of ZIP file format can be found here:
  * https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT
  * For a high level overview of the structure of a ZIP file see
@@ -119,6 +123,8 @@ struct local_file_header {
 	__u16 extra_field_length;
 } __attribute__((packed));
 
+#pragma GCC diagnostic pop
+
 struct zip_archive {
 	void *data;
 	__u32 size;
-- 
cgit v1.2.3-70-g09d2


From 0ba13995be9b416ea1d3daaf3ba871a67f45899b Mon Sep 17 00:00:00 2001
From: Xu Liang <lxu@maxlinear.com>
Date: Wed, 15 Mar 2023 00:30:23 +0800
Subject: net: phy: mxl-gpy: enhance delay time required by loopback disable
 function

GPY2xx devices need 3 seconds to fully switch out of loopback mode
before it can safely re-enter loopback mode. Implement timeout mechanism
to guarantee 3 seconds waited before re-enter loopback mode.

Signed-off-by: Xu Liang <lxu@maxlinear.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/mxl-gpy.c | 35 +++++++++++++++++++++++++++++------
 1 file changed, 29 insertions(+), 6 deletions(-)

diff --git a/drivers/net/phy/mxl-gpy.c b/drivers/net/phy/mxl-gpy.c
index e5972b4ef6e8..8e6bb97b5f85 100644
--- a/drivers/net/phy/mxl-gpy.c
+++ b/drivers/net/phy/mxl-gpy.c
@@ -107,6 +107,13 @@ struct gpy_priv {
 
 	u8 fw_major;
 	u8 fw_minor;
+
+	/* It takes 3 seconds to fully switch out of loopback mode before
+	 * it can safely re-enter loopback mode. Record the time when
+	 * loopback is disabled. Check and wait if necessary before loopback
+	 * is enabled.
+	 */
+	u64 lb_dis_to;
 };
 
 static const struct {
@@ -769,18 +776,34 @@ static void gpy_get_wol(struct phy_device *phydev,
 
 static int gpy_loopback(struct phy_device *phydev, bool enable)
 {
+	struct gpy_priv *priv = phydev->priv;
+	u16 set = 0;
 	int ret;
 
-	ret = phy_modify(phydev, MII_BMCR, BMCR_LOOPBACK,
-			 enable ? BMCR_LOOPBACK : 0);
-	if (!ret) {
-		/* It takes some time for PHY device to switch
-		 * into/out-of loopback mode.
+	if (enable) {
+		u64 now = get_jiffies_64();
+
+		/* wait until 3 seconds from last disable */
+		if (time_before64(now, priv->lb_dis_to))
+			msleep(jiffies64_to_msecs(priv->lb_dis_to - now));
+
+		set = BMCR_LOOPBACK;
+	}
+
+	ret = phy_modify(phydev, MII_BMCR, BMCR_LOOPBACK, set);
+	if (ret <= 0)
+		return ret;
+
+	if (enable) {
+		/* It takes some time for PHY device to switch into
+		 * loopback mode.
 		 */
 		msleep(100);
+	} else {
+		priv->lb_dis_to = get_jiffies_64() + HZ * 3;
 	}
 
-	return ret;
+	return 0;
 }
 
 static int gpy115_loopback(struct phy_device *phydev, bool enable)
-- 
cgit v1.2.3-70-g09d2


From 053fdaa841bd1af9fe9c2c30bba81119059aac95 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Tue, 14 Mar 2023 15:13:08 -0500
Subject: nfc: mrvl: Move platform_data struct into driver

There are no users of nfcmrvl platform_data struct outside of the
driver and none will be added, so move it into the driver.

Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/nfc/nfcmrvl/nfcmrvl.h         | 30 ++++++++++++++++++++--
 include/linux/platform_data/nfcmrvl.h | 48 -----------------------------------
 2 files changed, 28 insertions(+), 50 deletions(-)
 delete mode 100644 include/linux/platform_data/nfcmrvl.h

diff --git a/drivers/nfc/nfcmrvl/nfcmrvl.h b/drivers/nfc/nfcmrvl/nfcmrvl.h
index 165bd0a95190..0f22b3233f73 100644
--- a/drivers/nfc/nfcmrvl/nfcmrvl.h
+++ b/drivers/nfc/nfcmrvl/nfcmrvl.h
@@ -8,8 +8,6 @@
 #ifndef _NFCMRVL_H_
 #define _NFCMRVL_H_
 
-#include <linux/platform_data/nfcmrvl.h>
-
 #include "fw_dnld.h"
 
 /* Define private flags: */
@@ -50,6 +48,34 @@ enum nfcmrvl_phy {
 	NFCMRVL_PHY_SPI		= 3,
 };
 
+struct nfcmrvl_platform_data {
+	/*
+	 * Generic
+	 */
+
+	/* GPIO that is wired to RESET_N signal */
+	int reset_n_io;
+	/* Tell if transport is muxed in HCI one */
+	unsigned int hci_muxed;
+
+	/*
+	 * UART specific
+	 */
+
+	/* Tell if UART needs flow control at init */
+	unsigned int flow_control;
+	/* Tell if firmware supports break control for power management */
+	unsigned int break_control;
+
+
+	/*
+	 * I2C specific
+	 */
+
+	unsigned int irq;
+	unsigned int irq_polarity;
+};
+
 struct nfcmrvl_private {
 
 	unsigned long flags;
diff --git a/include/linux/platform_data/nfcmrvl.h b/include/linux/platform_data/nfcmrvl.h
deleted file mode 100644
index 9e75ac8d19be..000000000000
--- a/include/linux/platform_data/nfcmrvl.h
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * Copyright (C) 2015, Marvell International Ltd.
- *
- * This software file (the "File") is distributed by Marvell International
- * Ltd. under the terms of the GNU General Public License Version 2, June 1991
- * (the "License").  You may use, redistribute and/or modify this File in
- * accordance with the terms and conditions of the License, a copy of which
- * is available on the worldwide web at
- * http://www.gnu.org/licenses/old-licenses/gpl-2.0.txt.
- *
- * THE FILE IS DISTRIBUTED AS-IS, WITHOUT WARRANTY OF ANY KIND, AND THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE
- * ARE EXPRESSLY DISCLAIMED.  The License provides additional details about
- * this warranty disclaimer.
- */
-
-#ifndef _NFCMRVL_PTF_H_
-#define _NFCMRVL_PTF_H_
-
-struct nfcmrvl_platform_data {
-	/*
-	 * Generic
-	 */
-
-	/* GPIO that is wired to RESET_N signal */
-	int reset_n_io;
-	/* Tell if transport is muxed in HCI one */
-	unsigned int hci_muxed;
-
-	/*
-	 * UART specific
-	 */
-
-	/* Tell if UART needs flow control at init */
-	unsigned int flow_control;
-	/* Tell if firmware supports break control for power management */
-	unsigned int break_control;
-
-
-	/*
-	 * I2C specific
-	 */
-
-	unsigned int irq;
-	unsigned int irq_polarity;
-};
-
-#endif /* _NFCMRVL_PTF_H_ */
-- 
cgit v1.2.3-70-g09d2


From cc6d85c1cb5ccc04c594392339c81227abd1667d Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Tue, 14 Mar 2023 15:13:09 -0500
Subject: nfc: mrvl: Use of_property_read_bool() for boolean properties

It is preferred to use typed property access functions (i.e.
of_property_read_<type> functions) rather than low-level
of_get_property/of_find_property functions for reading properties.
Convert reading boolean properties to of_property_read_bool().

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/nfc/nfcmrvl/i2c.c     |  2 +-
 drivers/nfc/nfcmrvl/main.c    |  6 +-----
 drivers/nfc/nfcmrvl/nfcmrvl.h |  6 +++---
 drivers/nfc/nfcmrvl/uart.c    | 11 ++---------
 4 files changed, 7 insertions(+), 18 deletions(-)

diff --git a/drivers/nfc/nfcmrvl/i2c.c b/drivers/nfc/nfcmrvl/i2c.c
index e74342b0b728..164e2ab859fd 100644
--- a/drivers/nfc/nfcmrvl/i2c.c
+++ b/drivers/nfc/nfcmrvl/i2c.c
@@ -168,7 +168,7 @@ static int nfcmrvl_i2c_parse_dt(struct device_node *node,
 		return ret;
 	}
 
-	if (of_find_property(node, "i2c-int-falling", NULL))
+	if (of_property_read_bool(node, "i2c-int-falling"))
 		pdata->irq_polarity = IRQF_TRIGGER_FALLING;
 	else
 		pdata->irq_polarity = IRQF_TRIGGER_RISING;
diff --git a/drivers/nfc/nfcmrvl/main.c b/drivers/nfc/nfcmrvl/main.c
index 1a5284de4341..141bc4b66dcb 100644
--- a/drivers/nfc/nfcmrvl/main.c
+++ b/drivers/nfc/nfcmrvl/main.c
@@ -261,11 +261,7 @@ int nfcmrvl_parse_dt(struct device_node *node,
 		return reset_n_io;
 	}
 	pdata->reset_n_io = reset_n_io;
-
-	if (of_find_property(node, "hci-muxed", NULL))
-		pdata->hci_muxed = 1;
-	else
-		pdata->hci_muxed = 0;
+	pdata->hci_muxed = of_property_read_bool(node, "hci-muxed");
 
 	return 0;
 }
diff --git a/drivers/nfc/nfcmrvl/nfcmrvl.h b/drivers/nfc/nfcmrvl/nfcmrvl.h
index 0f22b3233f73..f61a99e553db 100644
--- a/drivers/nfc/nfcmrvl/nfcmrvl.h
+++ b/drivers/nfc/nfcmrvl/nfcmrvl.h
@@ -56,16 +56,16 @@ struct nfcmrvl_platform_data {
 	/* GPIO that is wired to RESET_N signal */
 	int reset_n_io;
 	/* Tell if transport is muxed in HCI one */
-	unsigned int hci_muxed;
+	bool hci_muxed;
 
 	/*
 	 * UART specific
 	 */
 
 	/* Tell if UART needs flow control at init */
-	unsigned int flow_control;
+	bool flow_control;
 	/* Tell if firmware supports break control for power management */
-	unsigned int break_control;
+	bool break_control;
 
 
 	/*
diff --git a/drivers/nfc/nfcmrvl/uart.c b/drivers/nfc/nfcmrvl/uart.c
index 9c92cbdc42f0..956ae92f7573 100644
--- a/drivers/nfc/nfcmrvl/uart.c
+++ b/drivers/nfc/nfcmrvl/uart.c
@@ -76,15 +76,8 @@ static int nfcmrvl_uart_parse_dt(struct device_node *node,
 		return ret;
 	}
 
-	if (of_find_property(matched_node, "flow-control", NULL))
-		pdata->flow_control = 1;
-	else
-		pdata->flow_control = 0;
-
-	if (of_find_property(matched_node, "break-control", NULL))
-		pdata->break_control = 1;
-	else
-		pdata->break_control = 0;
+	pdata->flow_control = of_property_read_bool(matched_node, "flow-control");
+	pdata->break_control = of_property_read_bool(matched_node, "break-control");
 
 	of_node_put(matched_node);
 
-- 
cgit v1.2.3-70-g09d2


From 77473d1a962f3d4f7ba48324502b6d27b8ef2591 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 16 Mar 2023 00:40:24 -0500
Subject: bpf: Free struct bpf_cpumask in call_rcu handler

The struct bpf_cpumask type uses the bpf_mem_cache_{alloc,free}() APIs
to allocate and free its cpumasks. The bpf_mem allocator may currently
immediately reuse some memory when its freed, without waiting for an RCU
read cycle to elapse. We want to be able to treat struct bpf_cpumask
objects as completely RCU safe.

This is necessary for two reasons:

1. bpf_cpumask_kptr_get() currently does an RCU-protected
   refcnt_inc_not_zero(). This of course assumes that the underlying
   memory is not reused, and is therefore unsafe in its current form.

2. We want to be able to get rid of bpf_cpumask_kptr_get() entirely, and
   intead use the superior kptr RCU semantics now afforded by the
   verifier.

This patch fixes (1), and enables (2), by making struct bpf_cpumask RCU
safe. A subsequent patch will update the verifier to allow struct
bpf_cpumask * pointers to be passed to KF_RCU kfuncs, and then a latter
patch will remove bpf_cpumask_kptr_get().

Fixes: 516f4d3397c9 ("bpf: Enable cpumasks to be queried and used as kptrs")
Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230316054028.88924-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/cpumask.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c
index b6587ec40f1b..98eea62b6b7b 100644
--- a/kernel/bpf/cpumask.c
+++ b/kernel/bpf/cpumask.c
@@ -9,6 +9,7 @@
 /**
  * struct bpf_cpumask - refcounted BPF cpumask wrapper structure
  * @cpumask:	The actual cpumask embedded in the struct.
+ * @rcu:	The RCU head used to free the cpumask with RCU safety.
  * @usage:	Object reference counter. When the refcount goes to 0, the
  *		memory is released back to the BPF allocator, which provides
  *		RCU safety.
@@ -24,6 +25,7 @@
  */
 struct bpf_cpumask {
 	cpumask_t cpumask;
+	struct rcu_head rcu;
 	refcount_t usage;
 };
 
@@ -108,6 +110,16 @@ __bpf_kfunc struct bpf_cpumask *bpf_cpumask_kptr_get(struct bpf_cpumask **cpumas
 	return cpumask;
 }
 
+static void cpumask_free_cb(struct rcu_head *head)
+{
+	struct bpf_cpumask *cpumask;
+
+	cpumask = container_of(head, struct bpf_cpumask, rcu);
+	migrate_disable();
+	bpf_mem_cache_free(&bpf_cpumask_ma, cpumask);
+	migrate_enable();
+}
+
 /**
  * bpf_cpumask_release() - Release a previously acquired BPF cpumask.
  * @cpumask: The cpumask being released.
@@ -121,11 +133,8 @@ __bpf_kfunc void bpf_cpumask_release(struct bpf_cpumask *cpumask)
 	if (!cpumask)
 		return;
 
-	if (refcount_dec_and_test(&cpumask->usage)) {
-		migrate_disable();
-		bpf_mem_cache_free(&bpf_cpumask_ma, cpumask);
-		migrate_enable();
-	}
+	if (refcount_dec_and_test(&cpumask->usage))
+		call_rcu(&cpumask->rcu, cpumask_free_cb);
 }
 
 /**
-- 
cgit v1.2.3-70-g09d2


From 63d2d83d21a6e2c6f019da5b2d5cdabe6d1cb951 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 16 Mar 2023 00:40:25 -0500
Subject: bpf: Mark struct bpf_cpumask as rcu protected

struct bpf_cpumask is a BPF-wrapper around the struct cpumask type which
can be instantiated by a BPF program, and then queried as a cpumask in
similar fashion to normal kernel code. The previous patch in this series
makes the type fully RCU safe, so the type can be included in the
rcu_protected_type BTF ID list.

A subsequent patch will remove bpf_cpumask_kptr_get(), as it's no longer
useful now that we can just treat the type as RCU safe by default and do
our own if check.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230316054028.88924-3-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 60793f793ca6..15b5c5c729f9 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4599,6 +4599,7 @@ static bool in_rcu_cs(struct bpf_verifier_env *env)
 BTF_SET_START(rcu_protected_types)
 BTF_ID(struct, prog_test_ref_kfunc)
 BTF_ID(struct, cgroup)
+BTF_ID(struct, bpf_cpumask)
 BTF_SET_END(rcu_protected_types)
 
 static bool rcu_protected_object(const struct btf *btf, u32 btf_id)
-- 
cgit v1.2.3-70-g09d2


From a5a197df58c44ce32a86b57e970da4bd7b71b399 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 16 Mar 2023 00:40:26 -0500
Subject: bpf/selftests: Test using global cpumask kptr with RCU

Now that struct bpf_cpumask * is considered an RCU-safe type according
to the verifier, we should add tests that validate its common usages.
This patch adds those tests to the cpumask test suite. A subsequent
changes will remove bpf_cpumask_kptr_get(), and will adjust the selftest
and BPF documentation accordingly.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230316054028.88924-4-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/cpumask.c   |  1 +
 tools/testing/selftests/bpf/progs/cpumask_common.h |  6 +++
 .../testing/selftests/bpf/progs/cpumask_failure.c  | 62 ++++++++++++++++++++++
 .../testing/selftests/bpf/progs/cpumask_success.c  | 33 ++++++++++++
 4 files changed, 102 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/cpumask.c b/tools/testing/selftests/bpf/prog_tests/cpumask.c
index 5fbe457c4ebe..6c0fe23498c7 100644
--- a/tools/testing/selftests/bpf/prog_tests/cpumask.c
+++ b/tools/testing/selftests/bpf/prog_tests/cpumask.c
@@ -17,6 +17,7 @@ static const char * const cpumask_success_testcases[] = {
 	"test_insert_leave",
 	"test_insert_remove_release",
 	"test_insert_kptr_get_release",
+	"test_global_mask_rcu",
 };
 
 static void verify_success(const char *prog_name)
diff --git a/tools/testing/selftests/bpf/progs/cpumask_common.h b/tools/testing/selftests/bpf/progs/cpumask_common.h
index 65e5496ca1b2..7623782fbd62 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_common.h
+++ b/tools/testing/selftests/bpf/progs/cpumask_common.h
@@ -9,6 +9,9 @@
 
 int err;
 
+#define private(name) SEC(".bss." #name) __hidden __attribute__((aligned(8)))
+private(MASK) static struct bpf_cpumask __kptr * global_mask;
+
 struct __cpumask_map_value {
 	struct bpf_cpumask __kptr * cpumask;
 };
@@ -51,6 +54,9 @@ void bpf_cpumask_copy(struct bpf_cpumask *dst, const struct cpumask *src) __ksym
 u32 bpf_cpumask_any(const struct cpumask *src) __ksym;
 u32 bpf_cpumask_any_and(const struct cpumask *src1, const struct cpumask *src2) __ksym;
 
+void bpf_rcu_read_lock(void) __ksym;
+void bpf_rcu_read_unlock(void) __ksym;
+
 static inline const struct cpumask *cast(struct bpf_cpumask *cpumask)
 {
 	return (const struct cpumask *)cpumask;
diff --git a/tools/testing/selftests/bpf/progs/cpumask_failure.c b/tools/testing/selftests/bpf/progs/cpumask_failure.c
index cfe83f0ef9e2..9f726d55f747 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_failure.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_failure.c
@@ -127,3 +127,65 @@ int BPF_PROG(test_cpumask_null, struct task_struct *task, u64 clone_flags)
 
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+__failure __msg("R2 must be a rcu pointer")
+int BPF_PROG(test_global_mask_out_of_rcu, struct task_struct *task, u64 clone_flags)
+{
+	struct bpf_cpumask *local, *prev;
+
+	local = create_cpumask();
+	if (!local)
+		return 0;
+
+	prev = bpf_kptr_xchg(&global_mask, local);
+	if (prev) {
+		bpf_cpumask_release(prev);
+		err = 3;
+		return 0;
+	}
+
+	bpf_rcu_read_lock();
+	local = global_mask;
+	if (!local) {
+		err = 4;
+		bpf_rcu_read_unlock();
+		return 0;
+	}
+
+	bpf_rcu_read_unlock();
+
+	/* RCU region is exited before calling KF_RCU kfunc. */
+
+	bpf_cpumask_test_cpu(0, (const struct cpumask *)local);
+
+	return 0;
+}
+
+SEC("tp_btf/task_newtask")
+__failure __msg("NULL pointer passed to trusted arg1")
+int BPF_PROG(test_global_mask_no_null_check, struct task_struct *task, u64 clone_flags)
+{
+	struct bpf_cpumask *local, *prev;
+
+	local = create_cpumask();
+	if (!local)
+		return 0;
+
+	prev = bpf_kptr_xchg(&global_mask, local);
+	if (prev) {
+		bpf_cpumask_release(prev);
+		err = 3;
+		return 0;
+	}
+
+	bpf_rcu_read_lock();
+	local = global_mask;
+
+	/* No NULL check is performed on global cpumask kptr. */
+	bpf_cpumask_test_cpu(0, (const struct cpumask *)local);
+
+	bpf_rcu_read_unlock();
+
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/cpumask_success.c b/tools/testing/selftests/bpf/progs/cpumask_success.c
index 97ed08c4ff03..fe928ff72a06 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_success.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_success.c
@@ -423,3 +423,36 @@ int BPF_PROG(test_insert_kptr_get_release, struct task_struct *task, u64 clone_f
 
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+int BPF_PROG(test_global_mask_rcu, struct task_struct *task, u64 clone_flags)
+{
+	struct bpf_cpumask *local, *prev;
+
+	if (!is_test_task())
+		return 0;
+
+	local = create_cpumask();
+	if (!local)
+		return 0;
+
+	prev = bpf_kptr_xchg(&global_mask, local);
+	if (prev) {
+		bpf_cpumask_release(prev);
+		err = 3;
+		return 0;
+	}
+
+	bpf_rcu_read_lock();
+	local = global_mask;
+	if (!local) {
+		err = 4;
+		bpf_rcu_read_unlock();
+		return 0;
+	}
+
+	bpf_cpumask_test_cpu(0, (const struct cpumask *)local);
+	bpf_rcu_read_unlock();
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From 1b403ce77dfbf234723a91bc411dfb03a0499d6e Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 16 Mar 2023 00:40:27 -0500
Subject: bpf: Remove bpf_cpumask_kptr_get() kfunc

Now that struct bpf_cpumask is RCU safe, there's no need for this kfunc.
Rather than doing the following:

private(MASK) static struct bpf_cpumask __kptr *global;

int BPF_PROG(prog, s32 cpu, ...)
{
	struct bpf_cpumask *cpumask;

	bpf_rcu_read_lock();
	cpumask = bpf_cpumask_kptr_get(&global);
	if (!cpumask) {
		bpf_rcu_read_unlock();
		return -1;
	}
	bpf_cpumask_setall(cpumask);
	...
	bpf_cpumask_release(cpumask);
	bpf_rcu_read_unlock();
}

Programs can instead simply do (assume same global cpumask):

int BPF_PROG(prog, ...)
{
	struct bpf_cpumask *cpumask;

	bpf_rcu_read_lock();
	cpumask = global;
	if (!cpumask) {
		bpf_rcu_read_unlock();
		return -1;
	}
	bpf_cpumask_setall(cpumask);
	...
	bpf_rcu_read_unlock();
}

In other words, no extra atomic acquire / release, and less boilerplate
code.

This patch removes both the kfunc, as well as its selftests and
documentation.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230316054028.88924-5-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/cpumask.c                               | 29 ---------------------
 tools/testing/selftests/bpf/prog_tests/cpumask.c   |  1 -
 tools/testing/selftests/bpf/progs/cpumask_common.h |  1 -
 .../testing/selftests/bpf/progs/cpumask_failure.c  | 24 -----------------
 .../testing/selftests/bpf/progs/cpumask_success.c  | 30 ----------------------
 5 files changed, 85 deletions(-)

diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c
index 98eea62b6b7b..db9da2194c1a 100644
--- a/kernel/bpf/cpumask.c
+++ b/kernel/bpf/cpumask.c
@@ -82,34 +82,6 @@ __bpf_kfunc struct bpf_cpumask *bpf_cpumask_acquire(struct bpf_cpumask *cpumask)
 	return cpumask;
 }
 
-/**
- * bpf_cpumask_kptr_get() - Attempt to acquire a reference to a BPF cpumask
- *			    stored in a map.
- * @cpumaskp: A pointer to a BPF cpumask map value.
- *
- * Attempts to acquire a reference to a BPF cpumask stored in a map value. The
- * cpumask returned by this function must either be embedded in a map as a
- * kptr, or freed with bpf_cpumask_release(). This function may return NULL if
- * no BPF cpumask was found in the specified map value.
- */
-__bpf_kfunc struct bpf_cpumask *bpf_cpumask_kptr_get(struct bpf_cpumask **cpumaskp)
-{
-	struct bpf_cpumask *cpumask;
-
-	/* The BPF memory allocator frees memory backing its caches in an RCU
-	 * callback. Thus, we can safely use RCU to ensure that the cpumask is
-	 * safe to read.
-	 */
-	rcu_read_lock();
-
-	cpumask = READ_ONCE(*cpumaskp);
-	if (cpumask && !refcount_inc_not_zero(&cpumask->usage))
-		cpumask = NULL;
-
-	rcu_read_unlock();
-	return cpumask;
-}
-
 static void cpumask_free_cb(struct rcu_head *head)
 {
 	struct bpf_cpumask *cpumask;
@@ -435,7 +407,6 @@ BTF_SET8_START(cpumask_kfunc_btf_ids)
 BTF_ID_FLAGS(func, bpf_cpumask_create, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cpumask_release, KF_RELEASE | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_cpumask_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
-BTF_ID_FLAGS(func, bpf_cpumask_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cpumask_first, KF_RCU)
 BTF_ID_FLAGS(func, bpf_cpumask_first_zero, KF_RCU)
 BTF_ID_FLAGS(func, bpf_cpumask_set_cpu, KF_RCU)
diff --git a/tools/testing/selftests/bpf/prog_tests/cpumask.c b/tools/testing/selftests/bpf/prog_tests/cpumask.c
index 6c0fe23498c7..cdf4acc18e4c 100644
--- a/tools/testing/selftests/bpf/prog_tests/cpumask.c
+++ b/tools/testing/selftests/bpf/prog_tests/cpumask.c
@@ -16,7 +16,6 @@ static const char * const cpumask_success_testcases[] = {
 	"test_copy_any_anyand",
 	"test_insert_leave",
 	"test_insert_remove_release",
-	"test_insert_kptr_get_release",
 	"test_global_mask_rcu",
 };
 
diff --git a/tools/testing/selftests/bpf/progs/cpumask_common.h b/tools/testing/selftests/bpf/progs/cpumask_common.h
index 7623782fbd62..0c5b785a93e4 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_common.h
+++ b/tools/testing/selftests/bpf/progs/cpumask_common.h
@@ -26,7 +26,6 @@ struct array_map {
 struct bpf_cpumask *bpf_cpumask_create(void) __ksym;
 void bpf_cpumask_release(struct bpf_cpumask *cpumask) __ksym;
 struct bpf_cpumask *bpf_cpumask_acquire(struct bpf_cpumask *cpumask) __ksym;
-struct bpf_cpumask *bpf_cpumask_kptr_get(struct bpf_cpumask **cpumask) __ksym;
 u32 bpf_cpumask_first(const struct cpumask *cpumask) __ksym;
 u32 bpf_cpumask_first_zero(const struct cpumask *cpumask) __ksym;
 void bpf_cpumask_set_cpu(u32 cpu, struct bpf_cpumask *cpumask) __ksym;
diff --git a/tools/testing/selftests/bpf/progs/cpumask_failure.c b/tools/testing/selftests/bpf/progs/cpumask_failure.c
index 9f726d55f747..db4f94e72b61 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_failure.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_failure.c
@@ -94,30 +94,6 @@ int BPF_PROG(test_insert_remove_no_release, struct task_struct *task, u64 clone_
 	return 0;
 }
 
-SEC("tp_btf/task_newtask")
-__failure __msg("Unreleased reference")
-int BPF_PROG(test_kptr_get_no_release, struct task_struct *task, u64 clone_flags)
-{
-	struct bpf_cpumask *cpumask;
-	struct __cpumask_map_value *v;
-
-	cpumask = create_cpumask();
-	if (!cpumask)
-		return 0;
-
-	if (cpumask_map_insert(cpumask))
-		return 0;
-
-	v = cpumask_map_value_lookup();
-	if (!v)
-		return 0;
-
-	cpumask = bpf_cpumask_kptr_get(&v->cpumask);
-
-	/* cpumask is never released. */
-	return 0;
-}
-
 SEC("tp_btf/task_newtask")
 __failure __msg("NULL pointer passed to trusted arg0")
 int BPF_PROG(test_cpumask_null, struct task_struct *task, u64 clone_flags)
diff --git a/tools/testing/selftests/bpf/progs/cpumask_success.c b/tools/testing/selftests/bpf/progs/cpumask_success.c
index fe928ff72a06..2fcdd7f68ac7 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_success.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_success.c
@@ -394,36 +394,6 @@ int BPF_PROG(test_insert_remove_release, struct task_struct *task, u64 clone_fla
 	return 0;
 }
 
-SEC("tp_btf/task_newtask")
-int BPF_PROG(test_insert_kptr_get_release, struct task_struct *task, u64 clone_flags)
-{
-	struct bpf_cpumask *cpumask;
-	struct __cpumask_map_value *v;
-
-	cpumask = create_cpumask();
-	if (!cpumask)
-		return 0;
-
-	if (cpumask_map_insert(cpumask)) {
-		err = 3;
-		return 0;
-	}
-
-	v = cpumask_map_value_lookup();
-	if (!v) {
-		err = 4;
-		return 0;
-	}
-
-	cpumask = bpf_cpumask_kptr_get(&v->cpumask);
-	if (cpumask)
-		bpf_cpumask_release(cpumask);
-	else
-		err = 5;
-
-	return 0;
-}
-
 SEC("tp_btf/task_newtask")
 int BPF_PROG(test_global_mask_rcu, struct task_struct *task, u64 clone_flags)
 {
-- 
cgit v1.2.3-70-g09d2


From fec2c6d14fd5001e7d24a2ae44f0e9aea82a6149 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 16 Mar 2023 00:40:28 -0500
Subject: bpf,docs: Remove bpf_cpumask_kptr_get() from documentation

Now that the kfunc no longer exists, we can remove it and instead
describe how RCU can be used to get a struct bpf_cpumask from a map
value. This patch updates the BPF documentation accordingly.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230316054028.88924-6-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/cpumasks.rst | 30 ++++++++++--------------------
 1 file changed, 10 insertions(+), 20 deletions(-)

diff --git a/Documentation/bpf/cpumasks.rst b/Documentation/bpf/cpumasks.rst
index 75344cd230e5..41efd8874eeb 100644
--- a/Documentation/bpf/cpumasks.rst
+++ b/Documentation/bpf/cpumasks.rst
@@ -117,12 +117,7 @@ For example:
 As mentioned and illustrated above, these ``struct bpf_cpumask *`` objects can
 also be stored in a map and used as kptrs. If a ``struct bpf_cpumask *`` is in
 a map, the reference can be removed from the map with bpf_kptr_xchg(), or
-opportunistically acquired with bpf_cpumask_kptr_get():
-
-.. kernel-doc:: kernel/bpf/cpumask.c
-  :identifiers: bpf_cpumask_kptr_get
-
-Here is an example of a ``struct bpf_cpumask *`` being retrieved from a map:
+opportunistically acquired using RCU:
 
 .. code-block:: c
 
@@ -144,7 +139,7 @@ Here is an example of a ``struct bpf_cpumask *`` being retrieved from a map:
 	/**
 	 * A simple example tracepoint program showing how a
 	 * struct bpf_cpumask * kptr that is stored in a map can
-	 * be acquired using the bpf_cpumask_kptr_get() kfunc.
+	 * be passed to kfuncs using RCU protection.
 	 */
 	SEC("tp_btf/cgroup_mkdir")
 	int BPF_PROG(cgrp_ancestor_example, struct cgroup *cgrp, const char *path)
@@ -158,26 +153,21 @@ Here is an example of a ``struct bpf_cpumask *`` being retrieved from a map:
 		if (!v)
 			return -ENOENT;
 
+		bpf_rcu_read_lock();
 		/* Acquire a reference to the bpf_cpumask * kptr that's already stored in the map. */
-		kptr = bpf_cpumask_kptr_get(&v->cpumask);
-		if (!kptr)
+		kptr = v->cpumask;
+		if (!kptr) {
 			/* If no bpf_cpumask was present in the map, it's because
 			 * we're racing with another CPU that removed it with
 			 * bpf_kptr_xchg() between the bpf_map_lookup_elem()
-			 * above, and our call to bpf_cpumask_kptr_get().
-			 * bpf_cpumask_kptr_get() internally safely handles this
-			 * race, and will return NULL if the cpumask is no longer
-			 * present in the map by the time we invoke the kfunc.
+			 * above, and our load of the pointer from the map.
 			 */
+			bpf_rcu_read_unlock();
 			return -EBUSY;
+		}
 
-		/* Free the reference we just took above. Note that the
-		 * original struct bpf_cpumask * kptr is still in the map. It will
-		 * be freed either at a later time if another context deletes
-		 * it from the map, or automatically by the BPF subsystem if
-		 * it's still present when the map is destroyed.
-		 */
-		bpf_cpumask_release(kptr);
+		bpf_cpumask_setall(kptr);
+		bpf_rcu_read_unlock();
 
 		return 0;
 	}
-- 
cgit v1.2.3-70-g09d2


From 082cdc69a4651dd2a77539d69416a359ed1214f5 Mon Sep 17 00:00:00 2001
From: Luis Gerhorst <gerhorst@cs.fau.de>
Date: Wed, 15 Mar 2023 17:54:00 +0100
Subject: bpf: Remove misleading spec_v1 check on var-offset stack read

For every BPF_ADD/SUB involving a pointer, adjust_ptr_min_max_vals()
ensures that the resulting pointer has a constant offset if
bypass_spec_v1 is false. This is ensured by calling sanitize_check_bounds()
which in turn calls check_stack_access_for_ptr_arithmetic(). There,
-EACCESS is returned if the register's offset is not constant, thereby
rejecting the program.

In summary, an unprivileged user must never be able to create stack
pointers with a variable offset. That is also the case, because a
respective check in check_stack_write() is missing. If they were able
to create a variable-offset pointer, users could still use it in a
stack-write operation to trigger unsafe speculative behavior [1].

Because unprivileged users must already be prevented from creating
variable-offset stack pointers, viable options are to either remove
this check (replacing it with a clarifying comment), or to turn it
into a "verifier BUG"-message, also adding a similar check in
check_stack_write() (for consistency, as a second-level defense).
This patch implements the first option to reduce verifier bloat.

This check was introduced by commit 01f810ace9ed ("bpf: Allow
variable-offset stack access") which correctly notes that
"variable-offset reads and writes are disallowed (they were already
disallowed for the indirect access case) because the speculative
execution checking code doesn't support them". However, it does not
further discuss why the check in check_stack_read() is necessary.
The code which made this check obsolete was also introduced in this
commit.

I have compiled ~650 programs from the Linux selftests, Linux samples,
Cilium, and libbpf/examples projects and confirmed that none of these
trigger the check in check_stack_read() [2]. Instead, all of these
programs are, as expected, already rejected when constructing the
variable-offset pointers. Note that the check in
check_stack_access_for_ptr_arithmetic() also prints "off=%d" while the
code removed by this patch does not (the error removed does not appear
in the "verification_error" values). For reproducibility, the
repository linked includes the raw data and scripts used to create
the plot.

  [1] https://arxiv.org/pdf/1807.03757.pdf
  [2] https://gitlab.cs.fau.de/un65esoq/bpf-spectre/-/raw/53dc19fcf459c186613b1156a81504b39c8d49db/data/plots/23-02-26_23-56_bpftool/bpftool/0004-errors.pdf?inline=false

Fixes: 01f810ace9ed ("bpf: Allow variable-offset stack access")
Signed-off-by: Luis Gerhorst <gerhorst@cs.fau.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230315165358.23701-1-gerhorst@cs.fau.de
---
 kernel/bpf/verifier.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 15b5c5c729f9..d62b7127ff2a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4303,17 +4303,13 @@ static int check_stack_read(struct bpf_verifier_env *env,
 	}
 	/* Variable offset is prohibited for unprivileged mode for simplicity
 	 * since it requires corresponding support in Spectre masking for stack
-	 * ALU. See also retrieve_ptr_limit().
+	 * ALU. See also retrieve_ptr_limit(). The check in
+	 * check_stack_access_for_ptr_arithmetic() called by
+	 * adjust_ptr_min_max_vals() prevents users from creating stack pointers
+	 * with variable offsets, therefore no check is required here. Further,
+	 * just checking it here would be insufficient as speculative stack
+	 * writes could still lead to unsafe speculative behaviour.
 	 */
-	if (!env->bypass_spec_v1 && var_off) {
-		char tn_buf[48];
-
-		tnum_strn(tn_buf, sizeof(tn_buf), reg->var_off);
-		verbose(env, "R%d variable offset stack access prohibited for !root, var_off=%s\n",
-				ptr_regno, tn_buf);
-		return -EACCES;
-	}
-
 	if (!var_off) {
 		off += reg->var_off.value;
 		err = check_stack_read_fixed_off(env, state, off, size,
-- 
cgit v1.2.3-70-g09d2


From d565263b7d83371e64ef315bdc3428909808b46f Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Tue, 14 Mar 2023 20:18:24 +0200
Subject: net: dsa: hellcreek: Get rid of custom led_init_default_state_get()

LED core provides a helper to parse default state from firmware node.
Use it instead of custom implementation.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Link: https://lore.kernel.org/r/20230314181824.56881-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/hirschmann/hellcreek_ptp.c | 45 ++++++++++++++++--------------
 1 file changed, 24 insertions(+), 21 deletions(-)

diff --git a/drivers/net/dsa/hirschmann/hellcreek_ptp.c b/drivers/net/dsa/hirschmann/hellcreek_ptp.c
index b28baab6d56a..3e44ccb7db84 100644
--- a/drivers/net/dsa/hirschmann/hellcreek_ptp.c
+++ b/drivers/net/dsa/hirschmann/hellcreek_ptp.c
@@ -297,7 +297,8 @@ static enum led_brightness hellcreek_led_is_gm_get(struct led_classdev *ldev)
 static int hellcreek_led_setup(struct hellcreek *hellcreek)
 {
 	struct device_node *leds, *led = NULL;
-	const char *label, *state;
+	enum led_default_state state;
+	const char *label;
 	int ret = -EINVAL;
 
 	of_node_get(hellcreek->dev->of_node);
@@ -318,16 +319,17 @@ static int hellcreek_led_setup(struct hellcreek *hellcreek)
 	ret = of_property_read_string(led, "label", &label);
 	hellcreek->led_sync_good.name = ret ? "sync_good" : label;
 
-	ret = of_property_read_string(led, "default-state", &state);
-	if (!ret) {
-		if (!strcmp(state, "on"))
-			hellcreek->led_sync_good.brightness = 1;
-		else if (!strcmp(state, "off"))
-			hellcreek->led_sync_good.brightness = 0;
-		else if (!strcmp(state, "keep"))
-			hellcreek->led_sync_good.brightness =
-				hellcreek_get_brightness(hellcreek,
-							 STATUS_OUT_SYNC_GOOD);
+	state = led_init_default_state_get(of_fwnode_handle(led));
+	switch (state) {
+	case LEDS_DEFSTATE_ON:
+		hellcreek->led_sync_good.brightness = 1;
+		break;
+	case LEDS_DEFSTATE_KEEP:
+		hellcreek->led_sync_good.brightness =
+			hellcreek_get_brightness(hellcreek, STATUS_OUT_SYNC_GOOD);
+		break;
+	default:
+		hellcreek->led_sync_good.brightness = 0;
 	}
 
 	hellcreek->led_sync_good.max_brightness = 1;
@@ -344,16 +346,17 @@ static int hellcreek_led_setup(struct hellcreek *hellcreek)
 	ret = of_property_read_string(led, "label", &label);
 	hellcreek->led_is_gm.name = ret ? "is_gm" : label;
 
-	ret = of_property_read_string(led, "default-state", &state);
-	if (!ret) {
-		if (!strcmp(state, "on"))
-			hellcreek->led_is_gm.brightness = 1;
-		else if (!strcmp(state, "off"))
-			hellcreek->led_is_gm.brightness = 0;
-		else if (!strcmp(state, "keep"))
-			hellcreek->led_is_gm.brightness =
-				hellcreek_get_brightness(hellcreek,
-							 STATUS_OUT_IS_GM);
+	state = led_init_default_state_get(of_fwnode_handle(led));
+	switch (state) {
+	case LEDS_DEFSTATE_ON:
+		hellcreek->led_is_gm.brightness = 1;
+		break;
+	case LEDS_DEFSTATE_KEEP:
+		hellcreek->led_is_gm.brightness =
+			hellcreek_get_brightness(hellcreek, STATUS_OUT_IS_GM);
+		break;
+	default:
+		hellcreek->led_is_gm.brightness = 0;
 	}
 
 	hellcreek->led_is_gm.max_brightness = 1;
-- 
cgit v1.2.3-70-g09d2


From abc783a7b0ff34d20a799d2fbf0ed0b2b06b72ed Mon Sep 17 00:00:00 2001
From: Durai Manickam KR <durai.manickamkr@microchip.com>
Date: Wed, 15 Mar 2023 15:20:52 +0530
Subject: net: macb: Add PTP support to GEM for sama7g5

Add PTP capability to the Gigabit Ethernet MAC.

Signed-off-by: Durai Manickam KR <durai.manickamkr@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cadence/macb_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 6e141a8bbf43..27fc6c903d25 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -4844,7 +4844,7 @@ static const struct macb_config mpfs_config = {
 
 static const struct macb_config sama7g5_gem_config = {
 	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_CLK_HW_CHG |
-		MACB_CAPS_MIIONRGMII,
+		MACB_CAPS_MIIONRGMII | MACB_CAPS_GEM_HAS_PTP,
 	.dma_burst_length = 16,
 	.clk_init = macb_clk_init,
 	.init = macb_init,
-- 
cgit v1.2.3-70-g09d2


From 9bae0dd05e61009bf45f4a6a8fbde87ed9567166 Mon Sep 17 00:00:00 2001
From: Durai Manickam KR <durai.manickamkr@microchip.com>
Date: Wed, 15 Mar 2023 15:20:53 +0530
Subject: net: macb: Add PTP support to EMAC for sama7g5

Add PTP capability to the Ethernet MAC.

Signed-off-by: Durai Manickam KR <durai.manickamkr@microchip.com>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cadence/macb_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 27fc6c903d25..1dbee16fe90a 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -4853,7 +4853,8 @@ static const struct macb_config sama7g5_gem_config = {
 
 static const struct macb_config sama7g5_emac_config = {
 	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII |
-		MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_MIIONRGMII,
+		MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_MIIONRGMII |
+		MACB_CAPS_GEM_HAS_PTP,
 	.dma_burst_length = 16,
 	.clk_init = macb_clk_init,
 	.init = macb_init,
-- 
cgit v1.2.3-70-g09d2


From e5995bc7e2ba1a0d444f806016d2e4ea91c032d0 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 16 Mar 2023 18:50:50 +0100
Subject: bpf, test_run: fix crashes due to XDP frame overwriting/corruption
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

syzbot and Ilya faced the splats when %XDP_PASS happens for bpf_test_run
after skb PP recycling was enabled for {__,}xdp_build_skb_from_frame():

BUG: kernel NULL pointer dereference, address: 0000000000000d28
RIP: 0010:memset_erms+0xd/0x20 arch/x86/lib/memset_64.S:66
[...]
Call Trace:
 <TASK>
 __finalize_skb_around net/core/skbuff.c:321 [inline]
 __build_skb_around+0x232/0x3a0 net/core/skbuff.c:379
 build_skb_around+0x32/0x290 net/core/skbuff.c:444
 __xdp_build_skb_from_frame+0x121/0x760 net/core/xdp.c:622
 xdp_recv_frames net/bpf/test_run.c:248 [inline]
 xdp_test_run_batch net/bpf/test_run.c:334 [inline]
 bpf_test_run_xdp_live+0x1289/0x1930 net/bpf/test_run.c:362
 bpf_prog_test_run_xdp+0xa05/0x14e0 net/bpf/test_run.c:1418
[...]

This happens due to that it calls xdp_scrub_frame(), which nullifies
xdpf->data. bpf_test_run code doesn't reinit the frame when the XDP
program doesn't adjust head or tail. Previously, %XDP_PASS meant the
page will be released from the pool and returned to the MM layer, but
now it does return to the Pool with the nullified xdpf->data, which
doesn't get reinitialized then.
So, in addition to checking whether the head and/or tail have been
adjusted, check also for a potential XDP frame corruption. xdpf->data
is 100% affected and also xdpf->flags is the field closest to the
metadata / frame start. Checking for these two should be enough for
non-extreme cases.

Fixes: 9c94bbf9a87b ("xdp: recycle Page Pool backed skbs built from XDP frames")
Reported-by: syzbot+e1d1b65f7c32f2a86a9f@syzkaller.appspotmail.com
Link: https://lore.kernel.org/bpf/000000000000f1985705f6ef2243@google.com
Reported-by: Ilya Leoshkevich <iii@linux.ibm.com>
Link: https://lore.kernel.org/bpf/e07dd94022ad5731705891b9487cc9ed66328b94.camel@linux.ibm.com
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Tested-by: Ilya Leoshkevich <iii@linux.ibm.com>
Link: https://lore.kernel.org/r/20230316175051.922550-2-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/bpf/test_run.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 71226f68270d..8d6b31209bd6 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -208,6 +208,16 @@ static void xdp_test_run_teardown(struct xdp_test_data *xdp)
 	kfree(xdp->skbs);
 }
 
+static bool frame_was_changed(const struct xdp_page_head *head)
+{
+	/* xdp_scrub_frame() zeroes the data pointer, flags is the last field,
+	 * i.e. has the highest chances to be overwritten. If those two are
+	 * untouched, it's most likely safe to skip the context reset.
+	 */
+	return head->frm.data != head->orig_ctx.data ||
+	       head->frm.flags != head->orig_ctx.flags;
+}
+
 static bool ctx_was_changed(struct xdp_page_head *head)
 {
 	return head->orig_ctx.data != head->ctx.data ||
@@ -217,7 +227,7 @@ static bool ctx_was_changed(struct xdp_page_head *head)
 
 static void reset_ctx(struct xdp_page_head *head)
 {
-	if (likely(!ctx_was_changed(head)))
+	if (likely(!frame_was_changed(head) && !ctx_was_changed(head)))
 		return;
 
 	head->ctx.data = head->orig_ctx.data;
-- 
cgit v1.2.3-70-g09d2


From 5640b6d894342d153b719644681b0345fd28ee96 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 16 Mar 2023 18:50:51 +0100
Subject: selftests/bpf: fix "metadata marker" getting overwritten by the
 netstack
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Alexei noticed xdp_do_redirect test on BPF CI started failing on
BE systems after skb PP recycling was enabled:

test_xdp_do_redirect:PASS:prog_run 0 nsec
test_xdp_do_redirect:PASS:pkt_count_xdp 0 nsec
test_xdp_do_redirect:PASS:pkt_count_zero 0 nsec
test_xdp_do_redirect:FAIL:pkt_count_tc unexpected pkt_count_tc: actual
220 != expected 9998
test_max_pkt_size:PASS:prog_run_max_size 0 nsec
test_max_pkt_size:PASS:prog_run_too_big 0 nsec
close_netns:PASS:setns 0 nsec
 #289 xdp_do_redirect:FAIL
Summary: 270/1674 PASSED, 30 SKIPPED, 1 FAILED

and it doesn't happen on LE systems.
Ilya then hunted it down to:

 #0  0x0000000000aaeee6 in neigh_hh_output (hh=0x83258df0,
skb=0x88142200) at linux/include/net/neighbour.h:503
 #1  0x0000000000ab2cda in neigh_output (skip_cache=false,
skb=0x88142200, n=<optimized out>) at linux/include/net/neighbour.h:544
 #2  ip6_finish_output2 (net=net@entry=0x88edba00, sk=sk@entry=0x0,
skb=skb@entry=0x88142200) at linux/net/ipv6/ip6_output.c:134
 #3  0x0000000000ab4cbc in __ip6_finish_output (skb=0x88142200, sk=0x0,
net=0x88edba00) at linux/net/ipv6/ip6_output.c:195
 #4  ip6_finish_output (net=0x88edba00, sk=0x0, skb=0x88142200) at
linux/net/ipv6/ip6_output.c:206

xdp_do_redirect test places a u32 marker (0x42) right before the Ethernet
header to check it then in the XDP program and return %XDP_ABORTED if it's
not there. Neigh xmit code likes to round up hard header length to speed
up copying the header, so it overwrites two bytes in front of the Eth
header. On LE systems, 0x42 is one byte at `data - 4`, while on BE it's
`data - 1`, what explains why it happens only there.
It didn't happen previously due to that %XDP_PASS meant the page will be
discarded and replaced by a new one, but now it can be recycled as well,
while bpf_test_run code doesn't reinitialize the content of recycled
pages. This mark is limited to this particular test and its setup though,
so there's no need to predict 1000 different possible cases. Just move
it 4 bytes to the left, still keeping it 32 bit to match on more bytes.

Fixes: 9c94bbf9a87b ("xdp: recycle Page Pool backed skbs built from XDP frames")
Reported-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/CAADnVQ+B_JOU+EpP=DKhbY9yXdN6GiRPnpTTXfEZ9sNkUeb-yQ@mail.gmail.com
Reported-by: Ilya Leoshkevich <iii@linux.ibm.com> # + debugging
Link: https://lore.kernel.org/bpf/8341c1d9f935f410438e79d3bd8a9cc50aefe105.camel@linux.ibm.com
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Tested-by: Ilya Leoshkevich <iii@linux.ibm.com>
Link: https://lore.kernel.org/r/20230316175051.922550-3-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c | 7 ++++---
 tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c | 2 +-
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
index 856cbc29e6a1..4eaa3dcaebc8 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
@@ -86,12 +86,12 @@ static void test_max_pkt_size(int fd)
 void test_xdp_do_redirect(void)
 {
 	int err, xdp_prog_fd, tc_prog_fd, ifindex_src, ifindex_dst;
-	char data[sizeof(pkt_udp) + sizeof(__u32)];
+	char data[sizeof(pkt_udp) + sizeof(__u64)];
 	struct test_xdp_do_redirect *skel = NULL;
 	struct nstoken *nstoken = NULL;
 	struct bpf_link *link;
 	LIBBPF_OPTS(bpf_xdp_query_opts, query_opts);
-	struct xdp_md ctx_in = { .data = sizeof(__u32),
+	struct xdp_md ctx_in = { .data = sizeof(__u64),
 				 .data_end = sizeof(data) };
 	DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts,
 			    .data_in = &data,
@@ -105,8 +105,9 @@ void test_xdp_do_redirect(void)
 	DECLARE_LIBBPF_OPTS(bpf_tc_hook, tc_hook,
 			    .attach_point = BPF_TC_INGRESS);
 
-	memcpy(&data[sizeof(__u32)], &pkt_udp, sizeof(pkt_udp));
+	memcpy(&data[sizeof(__u64)], &pkt_udp, sizeof(pkt_udp));
 	*((__u32 *)data) = 0x42; /* metadata test value */
+	*((__u32 *)data + 4) = 0;
 
 	skel = test_xdp_do_redirect__open();
 	if (!ASSERT_OK_PTR(skel, "skel"))
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c b/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c
index cd2d4e3258b8..5baaafed0d2d 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_do_redirect.c
@@ -52,7 +52,7 @@ int xdp_redirect(struct xdp_md *xdp)
 
 	*payload = MARK_IN;
 
-	if (bpf_xdp_adjust_meta(xdp, 4))
+	if (bpf_xdp_adjust_meta(xdp, sizeof(__u64)))
 		return XDP_ABORTED;
 
 	if (retcode > XDP_PASS)
-- 
cgit v1.2.3-70-g09d2


From c36a77c33db36277ca33b6a2fa5f802ba93694f7 Mon Sep 17 00:00:00 2001
From: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Date: Wed, 15 Mar 2023 00:49:22 -0300
Subject: net: dsa: realtek: rtl8365mb: add change_mtu
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The rtl8365mb was using a fixed MTU size of 1536, which was probably
inspired by the rtl8366rb's initial frame size. However, unlike that
family, the rtl8365mb family can specify the max frame size in bytes,
rather than in fixed steps.

DSA calls change_mtu for the CPU port once the max MTU value among the
ports changes. As the max frame size is defined globally, the switch
is configured only when the call affects the CPU port.

The available specifications do not directly define the max supported
frame size, but it mentions a 16k limit. This driver will use the 0x3FFF
limit as it is used in the vendor API code. However, the switch sets the
max frame size to 16368 bytes (0x3FF0) after it resets.

change_mtu uses MTU size, or ethernet payload size, while the switch
works with frame size. The frame size is calculated considering the
ethernet header (14 bytes), a possible 802.1Q tag (4 bytes), the payload
size (MTU), and the Ethernet FCS (4 bytes). The CPU tag (8 bytes) is
consumed before the switch enforces the limit.

During setup, the driver will use the default 1500-byte MTU of DSA to
set the maximum frame size. The current sum will be
VLAN_ETH_HLEN+1500+ETH_FCS_LEN, which results in 1522 bytes.  Although
it is lower than the previous initial value of 1536 bytes, the driver
will increase the frame size for a larger MTU. However, if something
requires more space without increasing the MTU, such as QinQ, we would
need to add the extra length to the rtl8365mb_port_change_mtu() formula.

MTU was tested up to 2018 (with 802.1Q) as that is as far as mt7620
(where rtl8367s is stacked) can go. The register was manually
manipulated byte-by-byte to ensure the MTU to frame size conversion was
correct. For frames without 802.1Q tag, the frame size limit will be 4
bytes over the required size.

There is a jumbo register, enabled by default at 6k frame size.
However, the jumbo settings do not seem to limit nor expand the maximum
tested MTU (2018), even when jumbo is disabled. More tests are needed
with a device that can handle larger frames.

Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Reviewed-by: Alvin Šipraga <alsi@bang-olufsen.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/realtek/rtl8365mb.c | 40 +++++++++++++++++++++++++++++++++----
 1 file changed, 36 insertions(+), 4 deletions(-)

diff --git a/drivers/net/dsa/realtek/rtl8365mb.c b/drivers/net/dsa/realtek/rtl8365mb.c
index da31d8b839ac..41ea3b5a42b1 100644
--- a/drivers/net/dsa/realtek/rtl8365mb.c
+++ b/drivers/net/dsa/realtek/rtl8365mb.c
@@ -98,6 +98,7 @@
 #include <linux/of_irq.h>
 #include <linux/regmap.h>
 #include <linux/if_bridge.h>
+#include <linux/if_vlan.h>
 
 #include "realtek.h"
 
@@ -267,6 +268,7 @@
 /* Maximum packet length register */
 #define RTL8365MB_CFG0_MAX_LEN_REG	0x088C
 #define   RTL8365MB_CFG0_MAX_LEN_MASK	0x3FFF
+#define RTL8365MB_CFG0_MAX_LEN_MAX	0x3FFF
 
 /* Port learning limit registers */
 #define RTL8365MB_LUT_PORT_LEARN_LIMIT_BASE		0x0A20
@@ -1135,6 +1137,35 @@ static void rtl8365mb_phylink_mac_link_up(struct dsa_switch *ds, int port,
 	}
 }
 
+static int rtl8365mb_port_change_mtu(struct dsa_switch *ds, int port,
+				     int new_mtu)
+{
+	struct realtek_priv *priv = ds->priv;
+	int frame_size;
+
+	/* When a new MTU is set, DSA always sets the CPU port's MTU to the
+	 * largest MTU of the slave ports. Because the switch only has a global
+	 * RX length register, only allowing CPU port here is enough.
+	 */
+	if (!dsa_is_cpu_port(ds, port))
+		return 0;
+
+	frame_size = new_mtu + VLAN_ETH_HLEN + ETH_FCS_LEN;
+
+	dev_dbg(priv->dev, "changing mtu to %d (frame size: %d)\n",
+		new_mtu, frame_size);
+
+	return regmap_update_bits(priv->map, RTL8365MB_CFG0_MAX_LEN_REG,
+				  RTL8365MB_CFG0_MAX_LEN_MASK,
+				  FIELD_PREP(RTL8365MB_CFG0_MAX_LEN_MASK,
+					     frame_size));
+}
+
+static int rtl8365mb_port_max_mtu(struct dsa_switch *ds, int port)
+{
+	return RTL8365MB_CFG0_MAX_LEN_MAX - VLAN_ETH_HLEN - ETH_FCS_LEN;
+}
+
 static void rtl8365mb_port_stp_state_set(struct dsa_switch *ds, int port,
 					 u8 state)
 {
@@ -1980,10 +2011,7 @@ static int rtl8365mb_setup(struct dsa_switch *ds)
 		p->index = i;
 	}
 
-	/* Set maximum packet length to 1536 bytes */
-	ret = regmap_update_bits(priv->map, RTL8365MB_CFG0_MAX_LEN_REG,
-				 RTL8365MB_CFG0_MAX_LEN_MASK,
-				 FIELD_PREP(RTL8365MB_CFG0_MAX_LEN_MASK, 1536));
+	ret = rtl8365mb_port_change_mtu(ds, cpu->trap_port, ETH_DATA_LEN);
 	if (ret)
 		goto out_teardown_irq;
 
@@ -2103,6 +2131,8 @@ static const struct dsa_switch_ops rtl8365mb_switch_ops_smi = {
 	.get_eth_mac_stats = rtl8365mb_get_mac_stats,
 	.get_eth_ctrl_stats = rtl8365mb_get_ctrl_stats,
 	.get_stats64 = rtl8365mb_get_stats64,
+	.port_change_mtu = rtl8365mb_port_change_mtu,
+	.port_max_mtu = rtl8365mb_port_max_mtu,
 };
 
 static const struct dsa_switch_ops rtl8365mb_switch_ops_mdio = {
@@ -2124,6 +2154,8 @@ static const struct dsa_switch_ops rtl8365mb_switch_ops_mdio = {
 	.get_eth_mac_stats = rtl8365mb_get_mac_stats,
 	.get_eth_ctrl_stats = rtl8365mb_get_ctrl_stats,
 	.get_stats64 = rtl8365mb_get_stats64,
+	.port_change_mtu = rtl8365mb_port_change_mtu,
+	.port_max_mtu = rtl8365mb_port_max_mtu,
 };
 
 static const struct realtek_ops rtl8365mb_ops = {
-- 
cgit v1.2.3-70-g09d2


From 81dc07417f0f82cd8f9733a1a3ebc2a473423e86 Mon Sep 17 00:00:00 2001
From: Mengyuan Lou <mengyuanlou@net-swift.com>
Date: Wed, 15 Mar 2023 15:43:04 +0800
Subject: net: wangxun: Implement the ndo change mtu interface

Add ngbe and txgbe ndo_change_mtu support.

Signed-off-by: Mengyuan Lou <mengyuanlou@net-swift.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/wangxun/libwx/wx_hw.c      | 21 ++++++++++++++++++++-
 drivers/net/ethernet/wangxun/libwx/wx_hw.h      |  1 +
 drivers/net/ethernet/wangxun/libwx/wx_type.h    |  2 ++
 drivers/net/ethernet/wangxun/ngbe/ngbe_main.c   |  5 ++++-
 drivers/net/ethernet/wangxun/ngbe/ngbe_type.h   |  1 -
 drivers/net/ethernet/wangxun/txgbe/txgbe_main.c |  5 ++++-
 drivers/net/ethernet/wangxun/txgbe/txgbe_type.h |  1 -
 7 files changed, 31 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c
index 7db57f934a91..ca409b4054d0 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c
@@ -4,6 +4,7 @@
 #include <linux/etherdevice.h>
 #include <linux/netdevice.h>
 #include <linux/if_ether.h>
+#include <linux/if_vlan.h>
 #include <linux/iopoll.h>
 #include <linux/pci.h>
 
@@ -1261,7 +1262,7 @@ static void wx_set_rx_buffer_len(struct wx *wx)
 	struct net_device *netdev = wx->netdev;
 	u32 mhadd, max_frame;
 
-	max_frame = netdev->mtu + ETH_HLEN + ETH_FCS_LEN;
+	max_frame = netdev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
 	/* adjust max frame to be at least the size of a standard frame */
 	if (max_frame < (ETH_FRAME_LEN + ETH_FCS_LEN))
 		max_frame = (ETH_FRAME_LEN + ETH_FCS_LEN);
@@ -1271,6 +1272,24 @@ static void wx_set_rx_buffer_len(struct wx *wx)
 		wr32(wx, WX_PSR_MAX_SZ, max_frame);
 }
 
+/**
+ * wx_change_mtu - Change the Maximum Transfer Unit
+ * @netdev: network interface device structure
+ * @new_mtu: new value for maximum frame size
+ *
+ * Returns 0 on success, negative on failure
+ **/
+int wx_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct wx *wx = netdev_priv(netdev);
+
+	netdev->mtu = new_mtu;
+	wx_set_rx_buffer_len(wx);
+
+	return 0;
+}
+EXPORT_SYMBOL(wx_change_mtu);
+
 /* Disable the specified rx queue */
 void wx_disable_rx_queue(struct wx *wx, struct wx_ring *ring)
 {
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.h b/drivers/net/ethernet/wangxun/libwx/wx_hw.h
index 44dfd6ea442a..c173c56f0ab5 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_hw.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.h
@@ -23,6 +23,7 @@ void wx_flush_sw_mac_table(struct wx *wx);
 int wx_set_mac(struct net_device *netdev, void *p);
 void wx_disable_rx(struct wx *wx);
 void wx_set_rx_mode(struct net_device *netdev);
+int wx_change_mtu(struct net_device *netdev, int new_mtu);
 void wx_disable_rx_queue(struct wx *wx, struct wx_ring *ring);
 void wx_configure(struct wx *wx);
 int wx_disable_pcie_master(struct wx *wx);
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h
index 77d8d7f1707e..2b9efd13c500 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_type.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h
@@ -300,6 +300,8 @@
 #define WX_MAX_RXD                   8192
 #define WX_MAX_TXD                   8192
 
+#define WX_MAX_JUMBO_FRAME_SIZE      9432 /* max payload 9414 */
+
 /* Supported Rx Buffer Sizes */
 #define WX_RXBUFFER_256      256    /* Used for skb receive header */
 #define WX_RXBUFFER_2K       2048
diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
index 0e4163e1106f..1a004aa2adcb 100644
--- a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
+++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
@@ -9,6 +9,7 @@
 #include <linux/etherdevice.h>
 #include <net/ip.h>
 #include <linux/phy.h>
+#include <linux/if_vlan.h>
 
 #include "../libwx/wx_type.h"
 #include "../libwx/wx_hw.h"
@@ -469,6 +470,7 @@ static void ngbe_shutdown(struct pci_dev *pdev)
 static const struct net_device_ops ngbe_netdev_ops = {
 	.ndo_open               = ngbe_open,
 	.ndo_stop               = ngbe_close,
+	.ndo_change_mtu         = wx_change_mtu,
 	.ndo_start_xmit         = wx_xmit_frame,
 	.ndo_set_rx_mode        = wx_set_rx_mode,
 	.ndo_validate_addr      = eth_validate_addr,
@@ -560,7 +562,8 @@ static int ngbe_probe(struct pci_dev *pdev,
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
 
 	netdev->min_mtu = ETH_MIN_MTU;
-	netdev->max_mtu = NGBE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN);
+	netdev->max_mtu = WX_MAX_JUMBO_FRAME_SIZE -
+			  (ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN);
 
 	wx->bd_number = func_nums;
 	/* setup the private structure */
diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h b/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h
index a2351349785e..373d5af628cd 100644
--- a/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h
+++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h
@@ -137,7 +137,6 @@ enum NGBE_MSCA_CMD_value {
 #define NGBE_RX_PB_SIZE				42
 #define NGBE_MC_TBL_SIZE			128
 #define NGBE_TDB_PB_SZ				(20 * 1024) /* 160KB Packet Buffer */
-#define NGBE_MAX_JUMBO_FRAME_SIZE		9432 /* max payload 9414 */
 
 /* TX/RX descriptor defines */
 #define NGBE_DEFAULT_TXD			512 /* default ring size */
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
index 859feaafd350..843a88bc416f 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
@@ -8,6 +8,7 @@
 #include <linux/string.h>
 #include <linux/etherdevice.h>
 #include <net/ip.h>
+#include <linux/if_vlan.h>
 
 #include "../libwx/wx_type.h"
 #include "../libwx/wx_lib.h"
@@ -486,6 +487,7 @@ static void txgbe_shutdown(struct pci_dev *pdev)
 static const struct net_device_ops txgbe_netdev_ops = {
 	.ndo_open               = txgbe_open,
 	.ndo_stop               = txgbe_close,
+	.ndo_change_mtu         = wx_change_mtu,
 	.ndo_start_xmit         = wx_xmit_frame,
 	.ndo_set_rx_mode        = wx_set_rx_mode,
 	.ndo_validate_addr      = eth_validate_addr,
@@ -603,7 +605,8 @@ static int txgbe_probe(struct pci_dev *pdev,
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
 
 	netdev->min_mtu = ETH_MIN_MTU;
-	netdev->max_mtu = TXGBE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN);
+	netdev->max_mtu = WX_MAX_JUMBO_FRAME_SIZE -
+			  (ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN);
 
 	/* make sure the EEPROM is good */
 	err = txgbe_validate_eeprom_checksum(wx, NULL);
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
index 563ea51deca6..63a1c733718d 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
@@ -79,7 +79,6 @@
 #define TXGBE_SP_MC_TBL_SIZE    128
 #define TXGBE_SP_RX_PB_SIZE     512
 #define TXGBE_SP_TDB_PB_SZ      (160 * 1024) /* 160KB Packet Buffer */
-#define TXGBE_MAX_JUMBO_FRAME_SIZE      9432 /* max payload 9414 */
 
 /* TX/RX descriptor defines */
 #define TXGBE_DEFAULT_TXD               512
-- 
cgit v1.2.3-70-g09d2


From bd7fc6e1957c2102866f9e464c1f2302e891b7e9 Mon Sep 17 00:00:00 2001
From: Shradha Gupta <shradhagupta@linux.microsoft.com>
Date: Wed, 15 Mar 2023 04:55:13 -0700
Subject: net: mana: Add new MANA VF performance counters for easier
 troubleshooting

Extended performance counter stats in 'ethtool -S <interface>' output
for MANA VF to facilitate troubleshooting.

Tested-on: Ubuntu22
Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c      | 62 ++++++++++++++++++++--
 drivers/net/ethernet/microsoft/mana/mana_ethtool.c | 52 +++++++++++++++++-
 include/net/mana/mana.h                            | 18 +++++++
 3 files changed, 128 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 6120f2b6684f..492474b4d8aa 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -156,6 +156,7 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	struct mana_txq *txq;
 	struct mana_cq *cq;
 	int err, len;
+	u16 ihs;
 
 	if (unlikely(!apc->port_is_up))
 		goto tx_drop;
@@ -166,6 +167,7 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	txq = &apc->tx_qp[txq_idx].txq;
 	gdma_sq = txq->gdma_sq;
 	cq = &apc->tx_qp[txq_idx].tx_cq;
+	tx_stats = &txq->stats;
 
 	pkg.tx_oob.s_oob.vcq_num = cq->gdma_id;
 	pkg.tx_oob.s_oob.vsq_frame = txq->vsq_frame;
@@ -179,10 +181,17 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 
 	pkg.tx_oob.s_oob.pkt_fmt = pkt_fmt;
 
-	if (pkt_fmt == MANA_SHORT_PKT_FMT)
+	if (pkt_fmt == MANA_SHORT_PKT_FMT) {
 		pkg.wqe_req.inline_oob_size = sizeof(struct mana_tx_short_oob);
-	else
+		u64_stats_update_begin(&tx_stats->syncp);
+		tx_stats->short_pkt_fmt++;
+		u64_stats_update_end(&tx_stats->syncp);
+	} else {
 		pkg.wqe_req.inline_oob_size = sizeof(struct mana_tx_oob);
+		u64_stats_update_begin(&tx_stats->syncp);
+		tx_stats->long_pkt_fmt++;
+		u64_stats_update_end(&tx_stats->syncp);
+	}
 
 	pkg.wqe_req.inline_oob_data = &pkg.tx_oob;
 	pkg.wqe_req.flags = 0;
@@ -232,9 +241,35 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 						 &ipv6_hdr(skb)->daddr, 0,
 						 IPPROTO_TCP, 0);
 		}
+
+		if (skb->encapsulation) {
+			ihs = skb_inner_tcp_all_headers(skb);
+			u64_stats_update_begin(&tx_stats->syncp);
+			tx_stats->tso_inner_packets++;
+			tx_stats->tso_inner_bytes += skb->len - ihs;
+			u64_stats_update_end(&tx_stats->syncp);
+		} else {
+			if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+				ihs = skb_transport_offset(skb) + sizeof(struct udphdr);
+			} else {
+				ihs = skb_tcp_all_headers(skb);
+				if (ipv6_has_hopopt_jumbo(skb))
+					ihs -= sizeof(struct hop_jumbo_hdr);
+			}
+
+			u64_stats_update_begin(&tx_stats->syncp);
+			tx_stats->tso_packets++;
+			tx_stats->tso_bytes += skb->len - ihs;
+			u64_stats_update_end(&tx_stats->syncp);
+		}
+
 	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		csum_type = mana_checksum_info(skb);
 
+		u64_stats_update_begin(&tx_stats->syncp);
+		tx_stats->csum_partial++;
+		u64_stats_update_end(&tx_stats->syncp);
+
 		if (csum_type == IPPROTO_TCP) {
 			pkg.tx_oob.s_oob.is_outer_ipv4 = ipv4;
 			pkg.tx_oob.s_oob.is_outer_ipv6 = ipv6;
@@ -254,8 +289,12 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 		}
 	}
 
-	if (mana_map_skb(skb, apc, &pkg))
+	if (mana_map_skb(skb, apc, &pkg)) {
+		u64_stats_update_begin(&tx_stats->syncp);
+		tx_stats->mana_map_err++;
+		u64_stats_update_end(&tx_stats->syncp);
 		goto free_sgl_ptr;
+	}
 
 	skb_queue_tail(&txq->pending_skbs, skb);
 
@@ -1038,6 +1077,8 @@ static void mana_poll_tx_cq(struct mana_cq *cq)
 	if (comp_read < 1)
 		return;
 
+	apc->eth_stats.tx_cqes = comp_read;
+
 	for (i = 0; i < comp_read; i++) {
 		struct mana_tx_comp_oob *cqe_oob;
 
@@ -1064,6 +1105,7 @@ static void mana_poll_tx_cq(struct mana_cq *cq)
 		case CQE_TX_VLAN_TAGGING_VIOLATION:
 			WARN_ONCE(1, "TX: CQE error %d: ignored.\n",
 				  cqe_oob->cqe_hdr.cqe_type);
+			apc->eth_stats.tx_cqe_err++;
 			break;
 
 		default:
@@ -1072,6 +1114,7 @@ static void mana_poll_tx_cq(struct mana_cq *cq)
 			 */
 			WARN_ONCE(1, "TX: Unexpected CQE type %d: HW BUG?\n",
 				  cqe_oob->cqe_hdr.cqe_type);
+			apc->eth_stats.tx_cqe_unknown_type++;
 			return;
 		}
 
@@ -1118,6 +1161,8 @@ static void mana_poll_tx_cq(struct mana_cq *cq)
 		WARN_ON_ONCE(1);
 
 	cq->work_done = pkt_transmitted;
+
+	apc->eth_stats.tx_cqes -= pkt_transmitted;
 }
 
 static void mana_post_pkt_rxq(struct mana_rxq *rxq)
@@ -1252,12 +1297,15 @@ static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	struct gdma_context *gc = rxq->gdma_rq->gdma_dev->gdma_context;
 	struct net_device *ndev = rxq->ndev;
 	struct mana_recv_buf_oob *rxbuf_oob;
+	struct mana_port_context *apc;
 	struct device *dev = gc->dev;
 	void *new_buf, *old_buf;
 	struct page *new_page;
 	u32 curr, pktlen;
 	dma_addr_t da;
 
+	apc = netdev_priv(ndev);
+
 	switch (oob->cqe_hdr.cqe_type) {
 	case CQE_RX_OKAY:
 		break;
@@ -1270,6 +1318,7 @@ static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 
 	case CQE_RX_COALESCED_4:
 		netdev_err(ndev, "RX coalescing is unsupported\n");
+		apc->eth_stats.rx_coalesced_err++;
 		return;
 
 	case CQE_RX_OBJECT_FENCE:
@@ -1279,6 +1328,7 @@ static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	default:
 		netdev_err(ndev, "Unknown RX CQE type = %d\n",
 			   oob->cqe_hdr.cqe_type);
+		apc->eth_stats.rx_cqe_unknown_type++;
 		return;
 	}
 
@@ -1341,11 +1391,15 @@ static void mana_poll_rx_cq(struct mana_cq *cq)
 {
 	struct gdma_comp *comp = cq->gdma_comp_buf;
 	struct mana_rxq *rxq = cq->rxq;
+	struct mana_port_context *apc;
 	int comp_read, i;
 
+	apc = netdev_priv(rxq->ndev);
+
 	comp_read = mana_gd_poll_cq(cq->gdma_cq, comp, CQE_POLLING_BUFFER);
 	WARN_ON_ONCE(comp_read > CQE_POLLING_BUFFER);
 
+	apc->eth_stats.rx_cqes = comp_read;
 	rxq->xdp_flush = false;
 
 	for (i = 0; i < comp_read; i++) {
@@ -1357,6 +1411,8 @@ static void mana_poll_rx_cq(struct mana_cq *cq)
 			return;
 
 		mana_process_rx_cqe(rxq, cq, &comp[i]);
+
+		apc->eth_stats.rx_cqes--;
 	}
 
 	if (rxq->xdp_flush)
diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
index 5b776a33a817..a64c81410dc1 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
@@ -13,6 +13,15 @@ static const struct {
 } mana_eth_stats[] = {
 	{"stop_queue", offsetof(struct mana_ethtool_stats, stop_queue)},
 	{"wake_queue", offsetof(struct mana_ethtool_stats, wake_queue)},
+	{"tx_cqes", offsetof(struct mana_ethtool_stats, tx_cqes)},
+	{"tx_cq_err", offsetof(struct mana_ethtool_stats, tx_cqe_err)},
+	{"tx_cqe_unknown_type", offsetof(struct mana_ethtool_stats,
+					tx_cqe_unknown_type)},
+	{"rx_cqes", offsetof(struct mana_ethtool_stats, rx_cqes)},
+	{"rx_coalesced_err", offsetof(struct mana_ethtool_stats,
+					rx_coalesced_err)},
+	{"rx_cqe_unknown_type", offsetof(struct mana_ethtool_stats,
+					rx_cqe_unknown_type)},
 };
 
 static int mana_get_sset_count(struct net_device *ndev, int stringset)
@@ -23,7 +32,8 @@ static int mana_get_sset_count(struct net_device *ndev, int stringset)
 	if (stringset != ETH_SS_STATS)
 		return -EINVAL;
 
-	return ARRAY_SIZE(mana_eth_stats) + num_queues * 8;
+	return ARRAY_SIZE(mana_eth_stats) + num_queues *
+				(MANA_STATS_RX_COUNT + MANA_STATS_TX_COUNT);
 }
 
 static void mana_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
@@ -61,6 +71,22 @@ static void mana_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
 		p += ETH_GSTRING_LEN;
 		sprintf(p, "tx_%d_xdp_xmit", i);
 		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_tso_packets", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_tso_bytes", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_tso_inner_packets", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_tso_inner_bytes", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_long_pkt_fmt", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_short_pkt_fmt", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_csum_partial", i);
+		p += ETH_GSTRING_LEN;
+		sprintf(p, "tx_%d_mana_map_err", i);
+		p += ETH_GSTRING_LEN;
 	}
 }
 
@@ -78,6 +104,14 @@ static void mana_get_ethtool_stats(struct net_device *ndev,
 	u64 xdp_xmit;
 	u64 xdp_drop;
 	u64 xdp_tx;
+	u64 tso_packets;
+	u64 tso_bytes;
+	u64 tso_inner_packets;
+	u64 tso_inner_bytes;
+	u64 long_pkt_fmt;
+	u64 short_pkt_fmt;
+	u64 csum_partial;
+	u64 mana_map_err;
 	int q, i = 0;
 
 	if (!apc->port_is_up)
@@ -113,11 +147,27 @@ static void mana_get_ethtool_stats(struct net_device *ndev,
 			packets = tx_stats->packets;
 			bytes = tx_stats->bytes;
 			xdp_xmit = tx_stats->xdp_xmit;
+			tso_packets = tx_stats->tso_packets;
+			tso_bytes = tx_stats->tso_bytes;
+			tso_inner_packets = tx_stats->tso_inner_packets;
+			tso_inner_bytes = tx_stats->tso_inner_bytes;
+			long_pkt_fmt = tx_stats->long_pkt_fmt;
+			short_pkt_fmt = tx_stats->short_pkt_fmt;
+			csum_partial = tx_stats->csum_partial;
+			mana_map_err = tx_stats->mana_map_err;
 		} while (u64_stats_fetch_retry(&tx_stats->syncp, start));
 
 		data[i++] = packets;
 		data[i++] = bytes;
 		data[i++] = xdp_xmit;
+		data[i++] = tso_packets;
+		data[i++] = tso_bytes;
+		data[i++] = tso_inner_packets;
+		data[i++] = tso_inner_bytes;
+		data[i++] = long_pkt_fmt;
+		data[i++] = short_pkt_fmt;
+		data[i++] = csum_partial;
+		data[i++] = mana_map_err;
 	}
 }
 
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index 3bb579962a14..bb11a6535d80 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -48,6 +48,10 @@ enum TRI_STATE {
 
 #define MAX_PORTS_IN_MANA_DEV 256
 
+/* Update this count whenever the respective structures are changed */
+#define MANA_STATS_RX_COUNT 5
+#define MANA_STATS_TX_COUNT 11
+
 struct mana_stats_rx {
 	u64 packets;
 	u64 bytes;
@@ -61,6 +65,14 @@ struct mana_stats_tx {
 	u64 packets;
 	u64 bytes;
 	u64 xdp_xmit;
+	u64 tso_packets;
+	u64 tso_bytes;
+	u64 tso_inner_packets;
+	u64 tso_inner_bytes;
+	u64 short_pkt_fmt;
+	u64 long_pkt_fmt;
+	u64 csum_partial;
+	u64 mana_map_err;
 	struct u64_stats_sync syncp;
 };
 
@@ -331,6 +343,12 @@ struct mana_tx_qp {
 struct mana_ethtool_stats {
 	u64 stop_queue;
 	u64 wake_queue;
+	u64 tx_cqes;
+	u64 tx_cqe_err;
+	u64 tx_cqe_unknown_type;
+	u64 rx_cqes;
+	u64 rx_coalesced_err;
+	u64 rx_cqe_unknown_type;
 };
 
 struct mana_context {
-- 
cgit v1.2.3-70-g09d2


From 40235edeadf58e4232bfcf8bf15be453cfe233b7 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Wed, 15 Mar 2023 13:29:47 +0530
Subject: dt-bindings: net: ti: k3-am654-cpsw-nuss: Fix compatible order

Reorder compatibles to follow alphanumeric order.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml b/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
index 628d63e1eb1f..6f56add1919b 100644
--- a/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
+++ b/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
@@ -54,11 +54,11 @@ properties:
 
   compatible:
     enum:
+      - ti,am642-cpsw-nuss
       - ti,am654-cpsw-nuss
       - ti,j7200-cpswxg-nuss
       - ti,j721e-cpsw-nuss
       - ti,j721e-cpswxg-nuss
-      - ti,am642-cpsw-nuss
 
   reg:
     maxItems: 1
@@ -215,8 +215,8 @@ allOf:
           compatible:
             contains:
               enum:
-                - ti,j721e-cpswxg-nuss
                 - ti,j7200-cpswxg-nuss
+                - ti,j721e-cpswxg-nuss
     then:
       properties:
         ethernet-ports:
-- 
cgit v1.2.3-70-g09d2


From e0c9c2a7dd738120c2fbc155c6fba1066f109be0 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Wed, 15 Mar 2023 13:29:48 +0530
Subject: dt-bindings: net: ti: k3-am654-cpsw-nuss: Add J784S4 CPSW9G support

Update bindings for TI K3 J784S4 SoC which contains 9 ports (8 external
ports) CPSW9G module and add compatible for it.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml b/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
index 6f56add1919b..306709bcc9e9 100644
--- a/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
+++ b/Documentation/devicetree/bindings/net/ti,k3-am654-cpsw-nuss.yaml
@@ -59,6 +59,7 @@ properties:
       - ti,j7200-cpswxg-nuss
       - ti,j721e-cpsw-nuss
       - ti,j721e-cpswxg-nuss
+      - ti,j784s4-cpswxg-nuss
 
   reg:
     maxItems: 1
@@ -197,7 +198,9 @@ allOf:
         properties:
           compatible:
             contains:
-              const: ti,j721e-cpswxg-nuss
+              enum:
+                - ti,j721e-cpswxg-nuss
+                - ti,j784s4-cpswxg-nuss
     then:
       properties:
         ethernet-ports:
@@ -217,6 +220,7 @@ allOf:
               enum:
                 - ti,j7200-cpswxg-nuss
                 - ti,j721e-cpswxg-nuss
+                - ti,j784s4-cpswxg-nuss
     then:
       properties:
         ethernet-ports:
-- 
cgit v1.2.3-70-g09d2


From 8c44fa12c8fa09c6c12f0dc25129a6d13ee0a1ea Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:45 +0200
Subject: net: Add MDB net device operations

Add MDB net device operations that will be invoked by rtnetlink code in
response to received RTM_{NEW,DEL,GET}MDB messages. Subsequent patches
will implement these operations in the bridge and VXLAN drivers.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ee483071cf59..23b0d7eaaadd 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1307,6 +1307,17 @@ struct netdev_net_notifier {
  *	Used to add FDB entries to dump requests. Implementers should add
  *	entries to skb and update idx with the number of entries.
  *
+ * int (*ndo_mdb_add)(struct net_device *dev, struct nlattr *tb[],
+ *		      u16 nlmsg_flags, struct netlink_ext_ack *extack);
+ *	Adds an MDB entry to dev.
+ * int (*ndo_mdb_del)(struct net_device *dev, struct nlattr *tb[],
+ *		      struct netlink_ext_ack *extack);
+ *	Deletes the MDB entry from dev.
+ * int (*ndo_mdb_dump)(struct net_device *dev, struct sk_buff *skb,
+ *		       struct netlink_callback *cb);
+ *	Dumps MDB entries from dev. The first argument (marker) in the netlink
+ *	callback is used by core rtnetlink code.
+ *
  * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh,
  *			     u16 flags, struct netlink_ext_ack *extack)
  * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
@@ -1569,6 +1580,16 @@ struct net_device_ops {
 					       const unsigned char *addr,
 					       u16 vid, u32 portid, u32 seq,
 					       struct netlink_ext_ack *extack);
+	int			(*ndo_mdb_add)(struct net_device *dev,
+					       struct nlattr *tb[],
+					       u16 nlmsg_flags,
+					       struct netlink_ext_ack *extack);
+	int			(*ndo_mdb_del)(struct net_device *dev,
+					       struct nlattr *tb[],
+					       struct netlink_ext_ack *extack);
+	int			(*ndo_mdb_dump)(struct net_device *dev,
+						struct sk_buff *skb,
+						struct netlink_callback *cb);
 	int			(*ndo_bridge_setlink)(struct net_device *dev,
 						      struct nlmsghdr *nlh,
 						      u16 flags,
-- 
cgit v1.2.3-70-g09d2


From c009de1061b57162fed206998dcdf8001416a481 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:46 +0200
Subject: bridge: mcast: Implement MDB net device operations

Implement the previously added MDB net device operations in the bridge
driver so that they could be invoked by core rtnetlink code in the next
patch.

The operations are identical to the existing br_mdb_{dump,add,del}
functions. The '_new' suffix will be removed in the next patch. The
functions are re-implemented in this patch to make the conversion in the
next patch easier to review.

Add dummy implementations when 'CONFIG_BRIDGE_IGMP_SNOOPING' is
disabled, so that an error will be returned to user space when it is
trying to add or delete an MDB entry. This is consistent with existing
behavior where the bridge driver does not even register rtnetlink
handlers for RTM_{NEW,DEL,GET}MDB messages when this Kconfig option is
disabled.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_device.c  |   3 ++
 net/bridge/br_mdb.c     | 124 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/bridge/br_private.h |  25 ++++++++++
 3 files changed, 152 insertions(+)

diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index b82906fc999a..85fa4d73bb53 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -468,6 +468,9 @@ static const struct net_device_ops br_netdev_ops = {
 	.ndo_fdb_del_bulk	 = br_fdb_delete_bulk,
 	.ndo_fdb_dump		 = br_fdb_dump,
 	.ndo_fdb_get		 = br_fdb_get,
+	.ndo_mdb_add		 = br_mdb_add_new,
+	.ndo_mdb_del		 = br_mdb_del_new,
+	.ndo_mdb_dump		 = br_mdb_dump_new,
 	.ndo_bridge_getlink	 = br_getlink,
 	.ndo_bridge_setlink	 = br_setlink,
 	.ndo_bridge_dellink	 = br_dellink,
diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 25c48d81a597..cb8270a5480b 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -458,6 +458,39 @@ out:
 	return skb->len;
 }
 
+int br_mdb_dump_new(struct net_device *dev, struct sk_buff *skb,
+		    struct netlink_callback *cb)
+{
+	struct net_bridge *br = netdev_priv(dev);
+	struct br_port_msg *bpm;
+	struct nlmsghdr *nlh;
+	int err;
+
+	nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid,
+			cb->nlh->nlmsg_seq, RTM_GETMDB, sizeof(*bpm),
+			NLM_F_MULTI);
+	if (!nlh)
+		return -EMSGSIZE;
+
+	bpm = nlmsg_data(nlh);
+	memset(bpm, 0, sizeof(*bpm));
+	bpm->ifindex = dev->ifindex;
+
+	rcu_read_lock();
+
+	err = br_mdb_fill_info(skb, cb, dev);
+	if (err)
+		goto out;
+	err = br_rports_fill_info(skb, &br->multicast_ctx);
+	if (err)
+		goto out;
+
+out:
+	rcu_read_unlock();
+	nlmsg_end(skb, nlh);
+	return err;
+}
+
 static int nlmsg_populate_mdb_fill(struct sk_buff *skb,
 				   struct net_device *dev,
 				   struct net_bridge_mdb_entry *mp,
@@ -1459,6 +1492,65 @@ out:
 	return err;
 }
 
+int br_mdb_add_new(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
+		   struct netlink_ext_ack *extack)
+{
+	struct net_bridge_vlan_group *vg;
+	struct br_mdb_config cfg = {};
+	struct net_bridge_vlan *v;
+	int err;
+
+	/* Configuration structure will be initialized here. */
+
+	err = -EINVAL;
+	/* host join errors which can happen before creating the group */
+	if (!cfg.p && !br_group_is_l2(&cfg.group)) {
+		/* don't allow any flags for host-joined IP groups */
+		if (cfg.entry->state) {
+			NL_SET_ERR_MSG_MOD(extack, "Flags are not allowed for host groups");
+			goto out;
+		}
+		if (!br_multicast_is_star_g(&cfg.group)) {
+			NL_SET_ERR_MSG_MOD(extack, "Groups with sources cannot be manually host joined");
+			goto out;
+		}
+	}
+
+	if (br_group_is_l2(&cfg.group) && cfg.entry->state != MDB_PERMANENT) {
+		NL_SET_ERR_MSG_MOD(extack, "Only permanent L2 entries allowed");
+		goto out;
+	}
+
+	if (cfg.p) {
+		if (cfg.p->state == BR_STATE_DISABLED && cfg.entry->state != MDB_PERMANENT) {
+			NL_SET_ERR_MSG_MOD(extack, "Port is in disabled state and entry is not permanent");
+			goto out;
+		}
+		vg = nbp_vlan_group(cfg.p);
+	} else {
+		vg = br_vlan_group(cfg.br);
+	}
+
+	/* If vlan filtering is enabled and VLAN is not specified
+	 * install mdb entry on all vlans configured on the port.
+	 */
+	if (br_vlan_enabled(cfg.br->dev) && vg && cfg.entry->vid == 0) {
+		list_for_each_entry(v, &vg->vlan_list, vlist) {
+			cfg.entry->vid = v->vid;
+			cfg.group.vid = v->vid;
+			err = __br_mdb_add(&cfg, extack);
+			if (err)
+				break;
+		}
+	} else {
+		err = __br_mdb_add(&cfg, extack);
+	}
+
+out:
+	br_mdb_config_fini(&cfg);
+	return err;
+}
+
 static int __br_mdb_del(const struct br_mdb_config *cfg)
 {
 	struct br_mdb_entry *entry = cfg->entry;
@@ -1535,6 +1627,38 @@ static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
 	return err;
 }
 
+int br_mdb_del_new(struct net_device *dev, struct nlattr *tb[],
+		   struct netlink_ext_ack *extack)
+{
+	struct net_bridge_vlan_group *vg;
+	struct br_mdb_config cfg = {};
+	struct net_bridge_vlan *v;
+	int err = 0;
+
+	/* Configuration structure will be initialized here. */
+
+	if (cfg.p)
+		vg = nbp_vlan_group(cfg.p);
+	else
+		vg = br_vlan_group(cfg.br);
+
+	/* If vlan filtering is enabled and VLAN is not specified
+	 * delete mdb entry on all vlans configured on the port.
+	 */
+	if (br_vlan_enabled(cfg.br->dev) && vg && cfg.entry->vid == 0) {
+		list_for_each_entry(v, &vg->vlan_list, vlist) {
+			cfg.entry->vid = v->vid;
+			cfg.group.vid = v->vid;
+			err = __br_mdb_del(&cfg);
+		}
+	} else {
+		err = __br_mdb_del(&cfg);
+	}
+
+	br_mdb_config_fini(&cfg);
+	return err;
+}
+
 void br_mdb_init(void)
 {
 	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_GETMDB, NULL, br_mdb_dump, 0);
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index cef5f6ea850c..a72847c1dc9f 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -981,6 +981,12 @@ void br_multicast_get_stats(const struct net_bridge *br,
 u32 br_multicast_ngroups_get(const struct net_bridge_mcast_port *pmctx);
 void br_multicast_ngroups_set_max(struct net_bridge_mcast_port *pmctx, u32 max);
 u32 br_multicast_ngroups_get_max(const struct net_bridge_mcast_port *pmctx);
+int br_mdb_add_new(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
+		   struct netlink_ext_ack *extack);
+int br_mdb_del_new(struct net_device *dev, struct nlattr *tb[],
+		   struct netlink_ext_ack *extack);
+int br_mdb_dump_new(struct net_device *dev, struct sk_buff *skb,
+		    struct netlink_callback *cb);
 void br_mdb_init(void);
 void br_mdb_uninit(void);
 void br_multicast_host_join(const struct net_bridge_mcast *brmctx,
@@ -1374,6 +1380,25 @@ static inline bool br_multicast_querier_exists(struct net_bridge_mcast *brmctx,
 	return false;
 }
 
+static inline int br_mdb_add_new(struct net_device *dev, struct nlattr *tb[],
+				 u16 nlmsg_flags,
+				 struct netlink_ext_ack *extack)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int br_mdb_del_new(struct net_device *dev, struct nlattr *tb[],
+				 struct netlink_ext_ack *extack)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int br_mdb_dump_new(struct net_device *dev, struct sk_buff *skb,
+				  struct netlink_callback *cb)
+{
+	return 0;
+}
+
 static inline void br_mdb_init(void)
 {
 }
-- 
cgit v1.2.3-70-g09d2


From cc7f5022f810cf7bf4f1826dd620656c35942a13 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:47 +0200
Subject: rtnetlink: bridge: mcast: Move MDB handlers out of bridge driver

Currently, the bridge driver registers handlers for MDB netlink
messages, making it impossible for other drivers to implement MDB
support.

As a preparation for VXLAN MDB support, move the MDB handlers out of the
bridge driver to the core rtnetlink code. The rtnetlink code will call
into individual drivers by invoking their previously added MDB net
device operations.

Note that while the diffstat is large, the change is mechanical. It
moves code out of the bridge driver to rtnetlink code. Also note that a
similar change was made in 2012 with commit 77162022ab26 ("net: add
generic PF_BRIDGE:RTM_ FDB hooks") that moved FDB handlers out of the
bridge driver to the core rtnetlink code.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_device.c  |   6 +-
 net/bridge/br_mdb.c     | 301 ++----------------------------------------------
 net/bridge/br_netlink.c |   3 -
 net/bridge/br_private.h |  35 ++----
 net/core/rtnetlink.c    | 217 ++++++++++++++++++++++++++++++++++
 5 files changed, 244 insertions(+), 318 deletions(-)

diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 85fa4d73bb53..df47c876230e 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -468,9 +468,9 @@ static const struct net_device_ops br_netdev_ops = {
 	.ndo_fdb_del_bulk	 = br_fdb_delete_bulk,
 	.ndo_fdb_dump		 = br_fdb_dump,
 	.ndo_fdb_get		 = br_fdb_get,
-	.ndo_mdb_add		 = br_mdb_add_new,
-	.ndo_mdb_del		 = br_mdb_del_new,
-	.ndo_mdb_dump		 = br_mdb_dump_new,
+	.ndo_mdb_add		 = br_mdb_add,
+	.ndo_mdb_del		 = br_mdb_del,
+	.ndo_mdb_dump		 = br_mdb_dump,
 	.ndo_bridge_getlink	 = br_getlink,
 	.ndo_bridge_setlink	 = br_setlink,
 	.ndo_bridge_dellink	 = br_dellink,
diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index cb8270a5480b..76636c61db21 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -380,86 +380,8 @@ out:
 	return err;
 }
 
-static int br_mdb_valid_dump_req(const struct nlmsghdr *nlh,
-				 struct netlink_ext_ack *extack)
-{
-	struct br_port_msg *bpm;
-
-	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*bpm))) {
-		NL_SET_ERR_MSG_MOD(extack, "Invalid header for mdb dump request");
-		return -EINVAL;
-	}
-
-	bpm = nlmsg_data(nlh);
-	if (bpm->ifindex) {
-		NL_SET_ERR_MSG_MOD(extack, "Filtering by device index is not supported for mdb dump request");
-		return -EINVAL;
-	}
-	if (nlmsg_attrlen(nlh, sizeof(*bpm))) {
-		NL_SET_ERR_MSG(extack, "Invalid data after header in mdb dump request");
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int br_mdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
-{
-	struct net_device *dev;
-	struct net *net = sock_net(skb->sk);
-	struct nlmsghdr *nlh = NULL;
-	int idx = 0, s_idx;
-
-	if (cb->strict_check) {
-		int err = br_mdb_valid_dump_req(cb->nlh, cb->extack);
-
-		if (err < 0)
-			return err;
-	}
-
-	s_idx = cb->args[0];
-
-	rcu_read_lock();
-
-	for_each_netdev_rcu(net, dev) {
-		if (netif_is_bridge_master(dev)) {
-			struct net_bridge *br = netdev_priv(dev);
-			struct br_port_msg *bpm;
-
-			if (idx < s_idx)
-				goto skip;
-
-			nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid,
-					cb->nlh->nlmsg_seq, RTM_GETMDB,
-					sizeof(*bpm), NLM_F_MULTI);
-			if (nlh == NULL)
-				break;
-
-			bpm = nlmsg_data(nlh);
-			memset(bpm, 0, sizeof(*bpm));
-			bpm->ifindex = dev->ifindex;
-			if (br_mdb_fill_info(skb, cb, dev) < 0)
-				goto out;
-			if (br_rports_fill_info(skb, &br->multicast_ctx) < 0)
-				goto out;
-
-			cb->args[1] = 0;
-			nlmsg_end(skb, nlh);
-		skip:
-			idx++;
-		}
-	}
-
-out:
-	if (nlh)
-		nlmsg_end(skb, nlh);
-	rcu_read_unlock();
-	cb->args[0] = idx;
-	return skb->len;
-}
-
-int br_mdb_dump_new(struct net_device *dev, struct sk_buff *skb,
-		    struct netlink_callback *cb)
+int br_mdb_dump(struct net_device *dev, struct sk_buff *skb,
+		struct netlink_callback *cb)
 {
 	struct net_bridge *br = netdev_priv(dev);
 	struct br_port_msg *bpm;
@@ -716,60 +638,6 @@ static const struct nla_policy br_mdbe_attrs_pol[MDBE_ATTR_MAX + 1] = {
 	[MDBE_ATTR_RTPROT] = NLA_POLICY_MIN(NLA_U8, RTPROT_STATIC),
 };
 
-static int validate_mdb_entry(const struct nlattr *attr,
-			      struct netlink_ext_ack *extack)
-{
-	struct br_mdb_entry *entry = nla_data(attr);
-
-	if (nla_len(attr) != sizeof(struct br_mdb_entry)) {
-		NL_SET_ERR_MSG_MOD(extack, "Invalid MDBA_SET_ENTRY attribute length");
-		return -EINVAL;
-	}
-
-	if (entry->ifindex == 0) {
-		NL_SET_ERR_MSG_MOD(extack, "Zero entry ifindex is not allowed");
-		return -EINVAL;
-	}
-
-	if (entry->addr.proto == htons(ETH_P_IP)) {
-		if (!ipv4_is_multicast(entry->addr.u.ip4)) {
-			NL_SET_ERR_MSG_MOD(extack, "IPv4 entry group address is not multicast");
-			return -EINVAL;
-		}
-		if (ipv4_is_local_multicast(entry->addr.u.ip4)) {
-			NL_SET_ERR_MSG_MOD(extack, "IPv4 entry group address is local multicast");
-			return -EINVAL;
-		}
-#if IS_ENABLED(CONFIG_IPV6)
-	} else if (entry->addr.proto == htons(ETH_P_IPV6)) {
-		if (ipv6_addr_is_ll_all_nodes(&entry->addr.u.ip6)) {
-			NL_SET_ERR_MSG_MOD(extack, "IPv6 entry group address is link-local all nodes");
-			return -EINVAL;
-		}
-#endif
-	} else if (entry->addr.proto == 0) {
-		/* L2 mdb */
-		if (!is_multicast_ether_addr(entry->addr.u.mac_addr)) {
-			NL_SET_ERR_MSG_MOD(extack, "L2 entry group is not multicast");
-			return -EINVAL;
-		}
-	} else {
-		NL_SET_ERR_MSG_MOD(extack, "Unknown entry protocol");
-		return -EINVAL;
-	}
-
-	if (entry->state != MDB_PERMANENT && entry->state != MDB_TEMPORARY) {
-		NL_SET_ERR_MSG_MOD(extack, "Unknown entry state");
-		return -EINVAL;
-	}
-	if (entry->vid >= VLAN_VID_MASK) {
-		NL_SET_ERR_MSG_MOD(extack, "Invalid entry VLAN id");
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static bool is_valid_mdb_source(struct nlattr *attr, __be16 proto,
 				struct netlink_ext_ack *extack)
 {
@@ -1332,49 +1200,16 @@ static int br_mdb_config_attrs_init(struct nlattr *set_attrs,
 	return 0;
 }
 
-static const struct nla_policy mdba_policy[MDBA_SET_ENTRY_MAX + 1] = {
-	[MDBA_SET_ENTRY_UNSPEC] = { .strict_start_type = MDBA_SET_ENTRY_ATTRS + 1 },
-	[MDBA_SET_ENTRY] = NLA_POLICY_VALIDATE_FN(NLA_BINARY,
-						  validate_mdb_entry,
-						  sizeof(struct br_mdb_entry)),
-	[MDBA_SET_ENTRY_ATTRS] = { .type = NLA_NESTED },
-};
-
-static int br_mdb_config_init(struct net *net, const struct nlmsghdr *nlh,
-			      struct br_mdb_config *cfg,
+static int br_mdb_config_init(struct br_mdb_config *cfg, struct net_device *dev,
+			      struct nlattr *tb[], u16 nlmsg_flags,
 			      struct netlink_ext_ack *extack)
 {
-	struct nlattr *tb[MDBA_SET_ENTRY_MAX + 1];
-	struct br_port_msg *bpm;
-	struct net_device *dev;
-	int err;
-
-	err = nlmsg_parse_deprecated(nlh, sizeof(*bpm), tb,
-				     MDBA_SET_ENTRY_MAX, mdba_policy, extack);
-	if (err)
-		return err;
+	struct net *net = dev_net(dev);
 
 	memset(cfg, 0, sizeof(*cfg));
 	cfg->filter_mode = MCAST_EXCLUDE;
 	cfg->rt_protocol = RTPROT_STATIC;
-	cfg->nlflags = nlh->nlmsg_flags;
-
-	bpm = nlmsg_data(nlh);
-	if (!bpm->ifindex) {
-		NL_SET_ERR_MSG_MOD(extack, "Invalid bridge ifindex");
-		return -EINVAL;
-	}
-
-	dev = __dev_get_by_index(net, bpm->ifindex);
-	if (!dev) {
-		NL_SET_ERR_MSG_MOD(extack, "Bridge device doesn't exist");
-		return -ENODEV;
-	}
-
-	if (!netif_is_bridge_master(dev)) {
-		NL_SET_ERR_MSG_MOD(extack, "Device is not a bridge");
-		return -EOPNOTSUPP;
-	}
+	cfg->nlflags = nlmsg_flags;
 
 	cfg->br = netdev_priv(dev);
 
@@ -1388,11 +1223,6 @@ static int br_mdb_config_init(struct net *net, const struct nlmsghdr *nlh,
 		return -EINVAL;
 	}
 
-	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, MDBA_SET_ENTRY)) {
-		NL_SET_ERR_MSG_MOD(extack, "Missing MDBA_SET_ENTRY attribute");
-		return -EINVAL;
-	}
-
 	cfg->entry = nla_data(tb[MDBA_SET_ENTRY]);
 
 	if (cfg->entry->ifindex != cfg->br->dev->ifindex) {
@@ -1430,16 +1260,15 @@ static void br_mdb_config_fini(struct br_mdb_config *cfg)
 	br_mdb_config_src_list_fini(cfg);
 }
 
-static int br_mdb_add(struct sk_buff *skb, struct nlmsghdr *nlh,
-		      struct netlink_ext_ack *extack)
+int br_mdb_add(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
+	       struct netlink_ext_ack *extack)
 {
-	struct net *net = sock_net(skb->sk);
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge_vlan *v;
 	struct br_mdb_config cfg;
 	int err;
 
-	err = br_mdb_config_init(net, nlh, &cfg, extack);
+	err = br_mdb_config_init(&cfg, dev, tb, nlmsg_flags, extack);
 	if (err)
 		return err;
 
@@ -1492,65 +1321,6 @@ out:
 	return err;
 }
 
-int br_mdb_add_new(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
-		   struct netlink_ext_ack *extack)
-{
-	struct net_bridge_vlan_group *vg;
-	struct br_mdb_config cfg = {};
-	struct net_bridge_vlan *v;
-	int err;
-
-	/* Configuration structure will be initialized here. */
-
-	err = -EINVAL;
-	/* host join errors which can happen before creating the group */
-	if (!cfg.p && !br_group_is_l2(&cfg.group)) {
-		/* don't allow any flags for host-joined IP groups */
-		if (cfg.entry->state) {
-			NL_SET_ERR_MSG_MOD(extack, "Flags are not allowed for host groups");
-			goto out;
-		}
-		if (!br_multicast_is_star_g(&cfg.group)) {
-			NL_SET_ERR_MSG_MOD(extack, "Groups with sources cannot be manually host joined");
-			goto out;
-		}
-	}
-
-	if (br_group_is_l2(&cfg.group) && cfg.entry->state != MDB_PERMANENT) {
-		NL_SET_ERR_MSG_MOD(extack, "Only permanent L2 entries allowed");
-		goto out;
-	}
-
-	if (cfg.p) {
-		if (cfg.p->state == BR_STATE_DISABLED && cfg.entry->state != MDB_PERMANENT) {
-			NL_SET_ERR_MSG_MOD(extack, "Port is in disabled state and entry is not permanent");
-			goto out;
-		}
-		vg = nbp_vlan_group(cfg.p);
-	} else {
-		vg = br_vlan_group(cfg.br);
-	}
-
-	/* If vlan filtering is enabled and VLAN is not specified
-	 * install mdb entry on all vlans configured on the port.
-	 */
-	if (br_vlan_enabled(cfg.br->dev) && vg && cfg.entry->vid == 0) {
-		list_for_each_entry(v, &vg->vlan_list, vlist) {
-			cfg.entry->vid = v->vid;
-			cfg.group.vid = v->vid;
-			err = __br_mdb_add(&cfg, extack);
-			if (err)
-				break;
-		}
-	} else {
-		err = __br_mdb_add(&cfg, extack);
-	}
-
-out:
-	br_mdb_config_fini(&cfg);
-	return err;
-}
-
 static int __br_mdb_del(const struct br_mdb_config *cfg)
 {
 	struct br_mdb_entry *entry = cfg->entry;
@@ -1592,16 +1362,15 @@ unlock:
 	return err;
 }
 
-static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
-		      struct netlink_ext_ack *extack)
+int br_mdb_del(struct net_device *dev, struct nlattr *tb[],
+	       struct netlink_ext_ack *extack)
 {
-	struct net *net = sock_net(skb->sk);
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge_vlan *v;
 	struct br_mdb_config cfg;
 	int err;
 
-	err = br_mdb_config_init(net, nlh, &cfg, extack);
+	err = br_mdb_config_init(&cfg, dev, tb, 0, extack);
 	if (err)
 		return err;
 
@@ -1626,49 +1395,3 @@ static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
 	br_mdb_config_fini(&cfg);
 	return err;
 }
-
-int br_mdb_del_new(struct net_device *dev, struct nlattr *tb[],
-		   struct netlink_ext_ack *extack)
-{
-	struct net_bridge_vlan_group *vg;
-	struct br_mdb_config cfg = {};
-	struct net_bridge_vlan *v;
-	int err = 0;
-
-	/* Configuration structure will be initialized here. */
-
-	if (cfg.p)
-		vg = nbp_vlan_group(cfg.p);
-	else
-		vg = br_vlan_group(cfg.br);
-
-	/* If vlan filtering is enabled and VLAN is not specified
-	 * delete mdb entry on all vlans configured on the port.
-	 */
-	if (br_vlan_enabled(cfg.br->dev) && vg && cfg.entry->vid == 0) {
-		list_for_each_entry(v, &vg->vlan_list, vlist) {
-			cfg.entry->vid = v->vid;
-			cfg.group.vid = v->vid;
-			err = __br_mdb_del(&cfg);
-		}
-	} else {
-		err = __br_mdb_del(&cfg);
-	}
-
-	br_mdb_config_fini(&cfg);
-	return err;
-}
-
-void br_mdb_init(void)
-{
-	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_GETMDB, NULL, br_mdb_dump, 0);
-	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_NEWMDB, br_mdb_add, NULL, 0);
-	rtnl_register_module(THIS_MODULE, PF_BRIDGE, RTM_DELMDB, br_mdb_del, NULL, 0);
-}
-
-void br_mdb_uninit(void)
-{
-	rtnl_unregister(PF_BRIDGE, RTM_GETMDB);
-	rtnl_unregister(PF_BRIDGE, RTM_NEWMDB);
-	rtnl_unregister(PF_BRIDGE, RTM_DELMDB);
-}
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 9173e52b89e2..fefb1c0e248b 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -1886,7 +1886,6 @@ int __init br_netlink_init(void)
 {
 	int err;
 
-	br_mdb_init();
 	br_vlan_rtnl_init();
 	rtnl_af_register(&br_af_ops);
 
@@ -1898,13 +1897,11 @@ int __init br_netlink_init(void)
 
 out_af:
 	rtnl_af_unregister(&br_af_ops);
-	br_mdb_uninit();
 	return err;
 }
 
 void br_netlink_fini(void)
 {
-	br_mdb_uninit();
 	br_vlan_rtnl_uninit();
 	rtnl_af_unregister(&br_af_ops);
 	rtnl_link_unregister(&br_link_ops);
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index a72847c1dc9f..7264fd40f82f 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -981,14 +981,12 @@ void br_multicast_get_stats(const struct net_bridge *br,
 u32 br_multicast_ngroups_get(const struct net_bridge_mcast_port *pmctx);
 void br_multicast_ngroups_set_max(struct net_bridge_mcast_port *pmctx, u32 max);
 u32 br_multicast_ngroups_get_max(const struct net_bridge_mcast_port *pmctx);
-int br_mdb_add_new(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
-		   struct netlink_ext_ack *extack);
-int br_mdb_del_new(struct net_device *dev, struct nlattr *tb[],
-		   struct netlink_ext_ack *extack);
-int br_mdb_dump_new(struct net_device *dev, struct sk_buff *skb,
-		    struct netlink_callback *cb);
-void br_mdb_init(void);
-void br_mdb_uninit(void);
+int br_mdb_add(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
+	       struct netlink_ext_ack *extack);
+int br_mdb_del(struct net_device *dev, struct nlattr *tb[],
+	       struct netlink_ext_ack *extack);
+int br_mdb_dump(struct net_device *dev, struct sk_buff *skb,
+		struct netlink_callback *cb);
 void br_multicast_host_join(const struct net_bridge_mcast *brmctx,
 			    struct net_bridge_mdb_entry *mp, bool notify);
 void br_multicast_host_leave(struct net_bridge_mdb_entry *mp, bool notify);
@@ -1380,33 +1378,24 @@ static inline bool br_multicast_querier_exists(struct net_bridge_mcast *brmctx,
 	return false;
 }
 
-static inline int br_mdb_add_new(struct net_device *dev, struct nlattr *tb[],
-				 u16 nlmsg_flags,
-				 struct netlink_ext_ack *extack)
+static inline int br_mdb_add(struct net_device *dev, struct nlattr *tb[],
+			     u16 nlmsg_flags, struct netlink_ext_ack *extack)
 {
 	return -EOPNOTSUPP;
 }
 
-static inline int br_mdb_del_new(struct net_device *dev, struct nlattr *tb[],
-				 struct netlink_ext_ack *extack)
+static inline int br_mdb_del(struct net_device *dev, struct nlattr *tb[],
+			     struct netlink_ext_ack *extack)
 {
 	return -EOPNOTSUPP;
 }
 
-static inline int br_mdb_dump_new(struct net_device *dev, struct sk_buff *skb,
-				  struct netlink_callback *cb)
+static inline int br_mdb_dump(struct net_device *dev, struct sk_buff *skb,
+			      struct netlink_callback *cb)
 {
 	return 0;
 }
 
-static inline void br_mdb_init(void)
-{
-}
-
-static inline void br_mdb_uninit(void)
-{
-}
-
 static inline int br_mdb_hash_init(struct net_bridge *br)
 {
 	return 0;
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 5d8eb57867a9..f347d9fa78c7 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -54,6 +54,9 @@
 #include <net/rtnetlink.h>
 #include <net/net_namespace.h>
 #include <net/devlink.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/addrconf.h>
+#endif
 
 #include "dev.h"
 
@@ -6063,6 +6066,216 @@ static int rtnl_stats_set(struct sk_buff *skb, struct nlmsghdr *nlh,
 	return 0;
 }
 
+static int rtnl_mdb_valid_dump_req(const struct nlmsghdr *nlh,
+				   struct netlink_ext_ack *extack)
+{
+	struct br_port_msg *bpm;
+
+	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*bpm))) {
+		NL_SET_ERR_MSG(extack, "Invalid header for mdb dump request");
+		return -EINVAL;
+	}
+
+	bpm = nlmsg_data(nlh);
+	if (bpm->ifindex) {
+		NL_SET_ERR_MSG(extack, "Filtering by device index is not supported for mdb dump request");
+		return -EINVAL;
+	}
+	if (nlmsg_attrlen(nlh, sizeof(*bpm))) {
+		NL_SET_ERR_MSG(extack, "Invalid data after header in mdb dump request");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+struct rtnl_mdb_dump_ctx {
+	long idx;
+};
+
+static int rtnl_mdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct rtnl_mdb_dump_ctx *ctx = (void *)cb->ctx;
+	struct net *net = sock_net(skb->sk);
+	struct net_device *dev;
+	int idx, s_idx;
+	int err;
+
+	NL_ASSERT_DUMP_CTX_FITS(struct rtnl_mdb_dump_ctx);
+
+	if (cb->strict_check) {
+		err = rtnl_mdb_valid_dump_req(cb->nlh, cb->extack);
+		if (err)
+			return err;
+	}
+
+	s_idx = ctx->idx;
+	idx = 0;
+
+	for_each_netdev(net, dev) {
+		if (idx < s_idx)
+			goto skip;
+		if (!dev->netdev_ops->ndo_mdb_dump)
+			goto skip;
+
+		err = dev->netdev_ops->ndo_mdb_dump(dev, skb, cb);
+		if (err == -EMSGSIZE)
+			goto out;
+		/* Moving on to next device, reset markers and sequence
+		 * counters since they are all maintained per-device.
+		 */
+		memset(cb->ctx, 0, sizeof(cb->ctx));
+		cb->prev_seq = 0;
+		cb->seq = 0;
+skip:
+		idx++;
+	}
+
+out:
+	ctx->idx = idx;
+	return skb->len;
+}
+
+static int rtnl_validate_mdb_entry(const struct nlattr *attr,
+				   struct netlink_ext_ack *extack)
+{
+	struct br_mdb_entry *entry = nla_data(attr);
+
+	if (nla_len(attr) != sizeof(struct br_mdb_entry)) {
+		NL_SET_ERR_MSG_ATTR(extack, attr, "Invalid attribute length");
+		return -EINVAL;
+	}
+
+	if (entry->ifindex == 0) {
+		NL_SET_ERR_MSG(extack, "Zero entry ifindex is not allowed");
+		return -EINVAL;
+	}
+
+	if (entry->addr.proto == htons(ETH_P_IP)) {
+		if (!ipv4_is_multicast(entry->addr.u.ip4)) {
+			NL_SET_ERR_MSG(extack, "IPv4 entry group address is not multicast");
+			return -EINVAL;
+		}
+		if (ipv4_is_local_multicast(entry->addr.u.ip4)) {
+			NL_SET_ERR_MSG(extack, "IPv4 entry group address is local multicast");
+			return -EINVAL;
+		}
+#if IS_ENABLED(CONFIG_IPV6)
+	} else if (entry->addr.proto == htons(ETH_P_IPV6)) {
+		if (ipv6_addr_is_ll_all_nodes(&entry->addr.u.ip6)) {
+			NL_SET_ERR_MSG(extack, "IPv6 entry group address is link-local all nodes");
+			return -EINVAL;
+		}
+#endif
+	} else if (entry->addr.proto == 0) {
+		/* L2 mdb */
+		if (!is_multicast_ether_addr(entry->addr.u.mac_addr)) {
+			NL_SET_ERR_MSG(extack, "L2 entry group is not multicast");
+			return -EINVAL;
+		}
+	} else {
+		NL_SET_ERR_MSG(extack, "Unknown entry protocol");
+		return -EINVAL;
+	}
+
+	if (entry->state != MDB_PERMANENT && entry->state != MDB_TEMPORARY) {
+		NL_SET_ERR_MSG(extack, "Unknown entry state");
+		return -EINVAL;
+	}
+	if (entry->vid >= VLAN_VID_MASK) {
+		NL_SET_ERR_MSG(extack, "Invalid entry VLAN id");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const struct nla_policy mdba_policy[MDBA_SET_ENTRY_MAX + 1] = {
+	[MDBA_SET_ENTRY_UNSPEC] = { .strict_start_type = MDBA_SET_ENTRY_ATTRS + 1 },
+	[MDBA_SET_ENTRY] = NLA_POLICY_VALIDATE_FN(NLA_BINARY,
+						  rtnl_validate_mdb_entry,
+						  sizeof(struct br_mdb_entry)),
+	[MDBA_SET_ENTRY_ATTRS] = { .type = NLA_NESTED },
+};
+
+static int rtnl_mdb_add(struct sk_buff *skb, struct nlmsghdr *nlh,
+			struct netlink_ext_ack *extack)
+{
+	struct nlattr *tb[MDBA_SET_ENTRY_MAX + 1];
+	struct net *net = sock_net(skb->sk);
+	struct br_port_msg *bpm;
+	struct net_device *dev;
+	int err;
+
+	err = nlmsg_parse_deprecated(nlh, sizeof(*bpm), tb,
+				     MDBA_SET_ENTRY_MAX, mdba_policy, extack);
+	if (err)
+		return err;
+
+	bpm = nlmsg_data(nlh);
+	if (!bpm->ifindex) {
+		NL_SET_ERR_MSG(extack, "Invalid ifindex");
+		return -EINVAL;
+	}
+
+	dev = __dev_get_by_index(net, bpm->ifindex);
+	if (!dev) {
+		NL_SET_ERR_MSG(extack, "Device doesn't exist");
+		return -ENODEV;
+	}
+
+	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, MDBA_SET_ENTRY)) {
+		NL_SET_ERR_MSG(extack, "Missing MDBA_SET_ENTRY attribute");
+		return -EINVAL;
+	}
+
+	if (!dev->netdev_ops->ndo_mdb_add) {
+		NL_SET_ERR_MSG(extack, "Device does not support MDB operations");
+		return -EOPNOTSUPP;
+	}
+
+	return dev->netdev_ops->ndo_mdb_add(dev, tb, nlh->nlmsg_flags, extack);
+}
+
+static int rtnl_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
+			struct netlink_ext_ack *extack)
+{
+	struct nlattr *tb[MDBA_SET_ENTRY_MAX + 1];
+	struct net *net = sock_net(skb->sk);
+	struct br_port_msg *bpm;
+	struct net_device *dev;
+	int err;
+
+	err = nlmsg_parse_deprecated(nlh, sizeof(*bpm), tb,
+				     MDBA_SET_ENTRY_MAX, mdba_policy, extack);
+	if (err)
+		return err;
+
+	bpm = nlmsg_data(nlh);
+	if (!bpm->ifindex) {
+		NL_SET_ERR_MSG(extack, "Invalid ifindex");
+		return -EINVAL;
+	}
+
+	dev = __dev_get_by_index(net, bpm->ifindex);
+	if (!dev) {
+		NL_SET_ERR_MSG(extack, "Device doesn't exist");
+		return -ENODEV;
+	}
+
+	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, MDBA_SET_ENTRY)) {
+		NL_SET_ERR_MSG(extack, "Missing MDBA_SET_ENTRY attribute");
+		return -EINVAL;
+	}
+
+	if (!dev->netdev_ops->ndo_mdb_del) {
+		NL_SET_ERR_MSG(extack, "Device does not support MDB operations");
+		return -EOPNOTSUPP;
+	}
+
+	return dev->netdev_ops->ndo_mdb_del(dev, tb, extack);
+}
+
 /* Process one rtnetlink message. */
 
 static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
@@ -6297,4 +6510,8 @@ void __init rtnetlink_init(void)
 	rtnl_register(PF_UNSPEC, RTM_GETSTATS, rtnl_stats_get, rtnl_stats_dump,
 		      0);
 	rtnl_register(PF_UNSPEC, RTM_SETSTATS, rtnl_stats_set, NULL, 0);
+
+	rtnl_register(PF_BRIDGE, RTM_GETMDB, NULL, rtnl_mdb_dump, 0);
+	rtnl_register(PF_BRIDGE, RTM_NEWMDB, rtnl_mdb_add, NULL, 0);
+	rtnl_register(PF_BRIDGE, RTM_DELMDB, rtnl_mdb_del, NULL, 0);
 }
-- 
cgit v1.2.3-70-g09d2


From da654c80a0ebba2e2a0614e017c9bbe57f643fe2 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:48 +0200
Subject: rtnetlink: bridge: mcast: Relax group address validation in common
 code

In the upcoming VXLAN MDB implementation, the 0.0.0.0 and :: MDB entries
will act as catchall entries for unregistered IP multicast traffic in a
similar fashion to the 00:00:00:00:00:00 VXLAN FDB entry that is used to
transmit BUM traffic.

In deployments where inter-subnet multicast forwarding is used, not all
the VTEPs in a tenant domain are members in all the broadcast domains.
It is therefore advantageous to transmit BULL (broadcast, unknown
unicast and link-local multicast) and unregistered IP multicast traffic
on different tunnels. If the same tunnel was used, a VTEP only
interested in IP multicast traffic would also pull all the BULL traffic
and drop it as it is not a member in the originating broadcast domain
[1].

Prepare for this change by allowing the 0.0.0.0 group address in the
common rtnetlink MDB code and forbid it in the bridge driver. A similar
change is not needed for IPv6 because the common code only validates
that the group address is not the all-nodes address.

[1] https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-irb-mcast#section-2.6

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_mdb.c  | 6 ++++++
 net/core/rtnetlink.c | 5 +++--
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 76636c61db21..7305f5f8215c 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -1246,6 +1246,12 @@ static int br_mdb_config_init(struct br_mdb_config *cfg, struct net_device *dev,
 		}
 	}
 
+	if (cfg->entry->addr.proto == htons(ETH_P_IP) &&
+	    ipv4_is_zeronet(cfg->entry->addr.u.ip4)) {
+		NL_SET_ERR_MSG_MOD(extack, "IPv4 entry group address 0.0.0.0 is not allowed");
+		return -EINVAL;
+	}
+
 	if (tb[MDBA_SET_ENTRY_ATTRS])
 		return br_mdb_config_attrs_init(tb[MDBA_SET_ENTRY_ATTRS], cfg,
 						extack);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index f347d9fa78c7..b7b1661d0d56 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -6152,8 +6152,9 @@ static int rtnl_validate_mdb_entry(const struct nlattr *attr,
 	}
 
 	if (entry->addr.proto == htons(ETH_P_IP)) {
-		if (!ipv4_is_multicast(entry->addr.u.ip4)) {
-			NL_SET_ERR_MSG(extack, "IPv4 entry group address is not multicast");
+		if (!ipv4_is_multicast(entry->addr.u.ip4) &&
+		    !ipv4_is_zeronet(entry->addr.u.ip4)) {
+			NL_SET_ERR_MSG(extack, "IPv4 entry group address is not multicast or 0.0.0.0");
 			return -EINVAL;
 		}
 		if (ipv4_is_local_multicast(entry->addr.u.ip4)) {
-- 
cgit v1.2.3-70-g09d2


From f307c8bf37a346ed3e8b6090b64b4ca8d61e1bcd Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:49 +0200
Subject: vxlan: Move address helpers to private headers

Move the helpers out of the core C file to the private header so that
they could be used by the upcoming MDB code.

While at it, constify the second argument of vxlan_nla_get_addr().

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan/vxlan_core.c    | 47 ---------------------------------------
 drivers/net/vxlan/vxlan_private.h | 45 +++++++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+), 47 deletions(-)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index f2c30214cae8..2c65cc5dd55d 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -71,53 +71,6 @@ static inline bool vxlan_collect_metadata(struct vxlan_sock *vs)
 	       ip_tunnel_collect_metadata();
 }
 
-#if IS_ENABLED(CONFIG_IPV6)
-static int vxlan_nla_get_addr(union vxlan_addr *ip, struct nlattr *nla)
-{
-	if (nla_len(nla) >= sizeof(struct in6_addr)) {
-		ip->sin6.sin6_addr = nla_get_in6_addr(nla);
-		ip->sa.sa_family = AF_INET6;
-		return 0;
-	} else if (nla_len(nla) >= sizeof(__be32)) {
-		ip->sin.sin_addr.s_addr = nla_get_in_addr(nla);
-		ip->sa.sa_family = AF_INET;
-		return 0;
-	} else {
-		return -EAFNOSUPPORT;
-	}
-}
-
-static int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
-			      const union vxlan_addr *ip)
-{
-	if (ip->sa.sa_family == AF_INET6)
-		return nla_put_in6_addr(skb, attr, &ip->sin6.sin6_addr);
-	else
-		return nla_put_in_addr(skb, attr, ip->sin.sin_addr.s_addr);
-}
-
-#else /* !CONFIG_IPV6 */
-
-static int vxlan_nla_get_addr(union vxlan_addr *ip, struct nlattr *nla)
-{
-	if (nla_len(nla) >= sizeof(struct in6_addr)) {
-		return -EAFNOSUPPORT;
-	} else if (nla_len(nla) >= sizeof(__be32)) {
-		ip->sin.sin_addr.s_addr = nla_get_in_addr(nla);
-		ip->sa.sa_family = AF_INET;
-		return 0;
-	} else {
-		return -EAFNOSUPPORT;
-	}
-}
-
-static int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
-			      const union vxlan_addr *ip)
-{
-	return nla_put_in_addr(skb, attr, ip->sin.sin_addr.s_addr);
-}
-#endif
-
 /* Find VXLAN socket based on network namespace, address family, UDP port,
  * enabled unshareable flags and socket device binding (see l3mdev with
  * non-default VRF).
diff --git a/drivers/net/vxlan/vxlan_private.h b/drivers/net/vxlan/vxlan_private.h
index 599c3b4fdd5e..038528f9684a 100644
--- a/drivers/net/vxlan/vxlan_private.h
+++ b/drivers/net/vxlan/vxlan_private.h
@@ -85,6 +85,31 @@ bool vxlan_addr_equal(const union vxlan_addr *a, const union vxlan_addr *b)
 		return a->sin.sin_addr.s_addr == b->sin.sin_addr.s_addr;
 }
 
+static inline int vxlan_nla_get_addr(union vxlan_addr *ip,
+				     const struct nlattr *nla)
+{
+	if (nla_len(nla) >= sizeof(struct in6_addr)) {
+		ip->sin6.sin6_addr = nla_get_in6_addr(nla);
+		ip->sa.sa_family = AF_INET6;
+		return 0;
+	} else if (nla_len(nla) >= sizeof(__be32)) {
+		ip->sin.sin_addr.s_addr = nla_get_in_addr(nla);
+		ip->sa.sa_family = AF_INET;
+		return 0;
+	} else {
+		return -EAFNOSUPPORT;
+	}
+}
+
+static inline int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
+				     const union vxlan_addr *ip)
+{
+	if (ip->sa.sa_family == AF_INET6)
+		return nla_put_in6_addr(skb, attr, &ip->sin6.sin6_addr);
+	else
+		return nla_put_in_addr(skb, attr, ip->sin.sin_addr.s_addr);
+}
+
 #else /* !CONFIG_IPV6 */
 
 static inline
@@ -93,6 +118,26 @@ bool vxlan_addr_equal(const union vxlan_addr *a, const union vxlan_addr *b)
 	return a->sin.sin_addr.s_addr == b->sin.sin_addr.s_addr;
 }
 
+static inline int vxlan_nla_get_addr(union vxlan_addr *ip,
+				     const struct nlattr *nla)
+{
+	if (nla_len(nla) >= sizeof(struct in6_addr)) {
+		return -EAFNOSUPPORT;
+	} else if (nla_len(nla) >= sizeof(__be32)) {
+		ip->sin.sin_addr.s_addr = nla_get_in_addr(nla);
+		ip->sa.sa_family = AF_INET;
+		return 0;
+	} else {
+		return -EAFNOSUPPORT;
+	}
+}
+
+static inline int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
+				     const union vxlan_addr *ip)
+{
+	return nla_put_in_addr(skb, attr, ip->sin.sin_addr.s_addr);
+}
+
 #endif
 
 static inline struct vxlan_vni_node *
-- 
cgit v1.2.3-70-g09d2


From 6ab271aaad25351ea8587d67c6837678b875eb2c Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:50 +0200
Subject: vxlan: Expose vxlan_xmit_one()

Given a packet and a remote destination, the function will take care of
encapsulating the packet and transmitting it to the destination.

Expose it so that it could be used in subsequent patches by the MDB code
to transmit a packet to the remote destination(s) stored in the MDB
entry.

It will allow us to keep the MDB code self-contained, not exposing its
data structures to the rest of the VXLAN driver.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan/vxlan_core.c    | 5 ++---
 drivers/net/vxlan/vxlan_private.h | 2 ++
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index 2c65cc5dd55d..5de1a20497a6 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -2395,9 +2395,8 @@ static int encap_bypass_if_local(struct sk_buff *skb, struct net_device *dev,
 	return 0;
 }
 
-static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
-			   __be32 default_vni, struct vxlan_rdst *rdst,
-			   bool did_rsc)
+void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
+		    __be32 default_vni, struct vxlan_rdst *rdst, bool did_rsc)
 {
 	struct dst_cache *dst_cache;
 	struct ip_tunnel_info *info;
diff --git a/drivers/net/vxlan/vxlan_private.h b/drivers/net/vxlan/vxlan_private.h
index 038528f9684a..f4977925cb8a 100644
--- a/drivers/net/vxlan/vxlan_private.h
+++ b/drivers/net/vxlan/vxlan_private.h
@@ -172,6 +172,8 @@ int vxlan_fdb_update(struct vxlan_dev *vxlan,
 		     __be16 port, __be32 src_vni, __be32 vni,
 		     __u32 ifindex, __u16 ndm_flags, u32 nhid,
 		     bool swdev_notify, struct netlink_ext_ack *extack);
+void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
+		    __be32 default_vni, struct vxlan_rdst *rdst, bool did_rsc);
 int vxlan_vni_in_use(struct net *src_net, struct vxlan_dev *vxlan,
 		     struct vxlan_config *conf, __be32 vni);
 
-- 
cgit v1.2.3-70-g09d2


From a3a48de5eade770e911d35291217bdd69ce04ef1 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:51 +0200
Subject: vxlan: mdb: Add MDB control path support

Implement MDB control path support, enabling the creation, deletion,
replacement and dumping of MDB entries in a similar fashion to the
bridge driver. Unlike the bridge driver, each entry stores a list of
remote VTEPs to which matched packets need to be replicated to and not a
list of bridge ports.

The motivating use case is the installation of MDB entries by a user
space control plane in response to received EVPN routes. As such, only
allow permanent MDB entries to be installed and do not implement
snooping functionality, avoiding a lot of unnecessary complexity.

Since entries can only be modified by user space under RTNL, use RTNL as
the write lock. Use RCU to ensure that MDB entries and remotes are not
freed while being accessed from the data path during transmission.

In terms of uAPI, reuse the existing MDB netlink interface, but add a
few new attributes to request and response messages:

* IP address of the destination VXLAN tunnel endpoint where the
  multicast receivers reside.

* UDP destination port number to use to connect to the remote VXLAN
  tunnel endpoint.

* VXLAN VNI Network Identifier to use to connect to the remote VXLAN
  tunnel endpoint. Required when Ingress Replication (IR) is used and
  the remote VTEP is not a member of originating broadcast domain
  (VLAN/VNI) [1].

* Source VNI Network Identifier the MDB entry belongs to. Used only when
  the VXLAN device is in external mode.

* Interface index of the outgoing interface to reach the remote VXLAN
  tunnel endpoint. This is required when the underlay destination IP is
  multicast (P2MP), as the multicast routing tables are not consulted.

All the new attributes are added under the 'MDBA_SET_ENTRY_ATTRS' nest
which is strictly validated by the bridge driver, thereby automatically
rejecting the new attributes.

[1] https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-irb-mcast#section-3.2.2

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan/Makefile        |    2 +-
 drivers/net/vxlan/vxlan_core.c    |    8 +
 drivers/net/vxlan/vxlan_mdb.c     | 1341 +++++++++++++++++++++++++++++++++++++
 drivers/net/vxlan/vxlan_private.h |   31 +
 include/net/vxlan.h               |    5 +
 include/uapi/linux/if_bridge.h    |   10 +
 6 files changed, 1396 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/vxlan/vxlan_mdb.c

diff --git a/drivers/net/vxlan/Makefile b/drivers/net/vxlan/Makefile
index d4c255499b72..91b8fec8b6cf 100644
--- a/drivers/net/vxlan/Makefile
+++ b/drivers/net/vxlan/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_VXLAN) += vxlan.o
 
-vxlan-objs := vxlan_core.o vxlan_multicast.o vxlan_vnifilter.o
+vxlan-objs := vxlan_core.o vxlan_multicast.o vxlan_vnifilter.o vxlan_mdb.o
diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index 5de1a20497a6..a8b26d4f76de 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -2878,8 +2878,14 @@ static int vxlan_init(struct net_device *dev)
 	if (err)
 		goto err_free_percpu;
 
+	err = vxlan_mdb_init(vxlan);
+	if (err)
+		goto err_gro_cells_destroy;
+
 	return 0;
 
+err_gro_cells_destroy:
+	gro_cells_destroy(&vxlan->gro_cells);
 err_free_percpu:
 	free_percpu(dev->tstats);
 err_vnigroup_uninit:
@@ -2904,6 +2910,8 @@ static void vxlan_uninit(struct net_device *dev)
 {
 	struct vxlan_dev *vxlan = netdev_priv(dev);
 
+	vxlan_mdb_fini(vxlan);
+
 	if (vxlan->cfg.flags & VXLAN_F_VNIFILTER)
 		vxlan_vnigroup_uninit(vxlan);
 
diff --git a/drivers/net/vxlan/vxlan_mdb.c b/drivers/net/vxlan/vxlan_mdb.c
new file mode 100644
index 000000000000..129692b3663f
--- /dev/null
+++ b/drivers/net/vxlan/vxlan_mdb.c
@@ -0,0 +1,1341 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/if_bridge.h>
+#include <linux/in.h>
+#include <linux/list.h>
+#include <linux/netdevice.h>
+#include <linux/netlink.h>
+#include <linux/rhashtable.h>
+#include <linux/rhashtable-types.h>
+#include <linux/rtnetlink.h>
+#include <linux/skbuff.h>
+#include <linux/types.h>
+#include <net/netlink.h>
+#include <net/vxlan.h>
+
+#include "vxlan_private.h"
+
+struct vxlan_mdb_entry_key {
+	union vxlan_addr src;
+	union vxlan_addr dst;
+	__be32 vni;
+};
+
+struct vxlan_mdb_entry {
+	struct rhash_head rhnode;
+	struct list_head remotes;
+	struct vxlan_mdb_entry_key key;
+	struct hlist_node mdb_node;
+	struct rcu_head rcu;
+};
+
+#define VXLAN_MDB_REMOTE_F_BLOCKED	BIT(0)
+
+struct vxlan_mdb_remote {
+	struct list_head list;
+	struct vxlan_rdst __rcu *rd;
+	u8 flags;
+	u8 filter_mode;
+	u8 rt_protocol;
+	struct hlist_head src_list;
+	struct rcu_head rcu;
+};
+
+#define VXLAN_SGRP_F_DELETE	BIT(0)
+
+struct vxlan_mdb_src_entry {
+	struct hlist_node node;
+	union vxlan_addr addr;
+	u8 flags;
+};
+
+struct vxlan_mdb_dump_ctx {
+	long reserved;
+	long entry_idx;
+	long remote_idx;
+};
+
+struct vxlan_mdb_config_src_entry {
+	union vxlan_addr addr;
+	struct list_head node;
+};
+
+struct vxlan_mdb_config {
+	struct vxlan_dev *vxlan;
+	struct vxlan_mdb_entry_key group;
+	struct list_head src_list;
+	union vxlan_addr remote_ip;
+	u32 remote_ifindex;
+	__be32 remote_vni;
+	__be16 remote_port;
+	u16 nlflags;
+	u8 flags;
+	u8 filter_mode;
+	u8 rt_protocol;
+};
+
+static const struct rhashtable_params vxlan_mdb_rht_params = {
+	.head_offset = offsetof(struct vxlan_mdb_entry, rhnode),
+	.key_offset = offsetof(struct vxlan_mdb_entry, key),
+	.key_len = sizeof(struct vxlan_mdb_entry_key),
+	.automatic_shrinking = true,
+};
+
+static int __vxlan_mdb_add(const struct vxlan_mdb_config *cfg,
+			   struct netlink_ext_ack *extack);
+static int __vxlan_mdb_del(const struct vxlan_mdb_config *cfg,
+			   struct netlink_ext_ack *extack);
+
+static void vxlan_br_mdb_entry_fill(const struct vxlan_dev *vxlan,
+				    const struct vxlan_mdb_entry *mdb_entry,
+				    const struct vxlan_mdb_remote *remote,
+				    struct br_mdb_entry *e)
+{
+	const union vxlan_addr *dst = &mdb_entry->key.dst;
+
+	memset(e, 0, sizeof(*e));
+	e->ifindex = vxlan->dev->ifindex;
+	e->state = MDB_PERMANENT;
+
+	if (remote->flags & VXLAN_MDB_REMOTE_F_BLOCKED)
+		e->flags |= MDB_FLAGS_BLOCKED;
+
+	switch (dst->sa.sa_family) {
+	case AF_INET:
+		e->addr.u.ip4 = dst->sin.sin_addr.s_addr;
+		e->addr.proto = htons(ETH_P_IP);
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case AF_INET6:
+		e->addr.u.ip6 = dst->sin6.sin6_addr;
+		e->addr.proto = htons(ETH_P_IPV6);
+		break;
+#endif
+	}
+}
+
+static int vxlan_mdb_entry_info_fill_srcs(struct sk_buff *skb,
+					  const struct vxlan_mdb_remote *remote)
+{
+	struct vxlan_mdb_src_entry *ent;
+	struct nlattr *nest;
+
+	if (hlist_empty(&remote->src_list))
+		return 0;
+
+	nest = nla_nest_start(skb, MDBA_MDB_EATTR_SRC_LIST);
+	if (!nest)
+		return -EMSGSIZE;
+
+	hlist_for_each_entry(ent, &remote->src_list, node) {
+		struct nlattr *nest_ent;
+
+		nest_ent = nla_nest_start(skb, MDBA_MDB_SRCLIST_ENTRY);
+		if (!nest_ent)
+			goto out_cancel_err;
+
+		if (vxlan_nla_put_addr(skb, MDBA_MDB_SRCATTR_ADDRESS,
+				       &ent->addr) ||
+		    nla_put_u32(skb, MDBA_MDB_SRCATTR_TIMER, 0))
+			goto out_cancel_err;
+
+		nla_nest_end(skb, nest_ent);
+	}
+
+	nla_nest_end(skb, nest);
+
+	return 0;
+
+out_cancel_err:
+	nla_nest_cancel(skb, nest);
+	return -EMSGSIZE;
+}
+
+static int vxlan_mdb_entry_info_fill(const struct vxlan_dev *vxlan,
+				     struct sk_buff *skb,
+				     const struct vxlan_mdb_entry *mdb_entry,
+				     const struct vxlan_mdb_remote *remote)
+{
+	struct vxlan_rdst *rd = rtnl_dereference(remote->rd);
+	struct br_mdb_entry e;
+	struct nlattr *nest;
+
+	nest = nla_nest_start_noflag(skb, MDBA_MDB_ENTRY_INFO);
+	if (!nest)
+		return -EMSGSIZE;
+
+	vxlan_br_mdb_entry_fill(vxlan, mdb_entry, remote, &e);
+
+	if (nla_put_nohdr(skb, sizeof(e), &e) ||
+	    nla_put_u32(skb, MDBA_MDB_EATTR_TIMER, 0))
+		goto nest_err;
+
+	if (!vxlan_addr_any(&mdb_entry->key.src) &&
+	    vxlan_nla_put_addr(skb, MDBA_MDB_EATTR_SOURCE, &mdb_entry->key.src))
+		goto nest_err;
+
+	if (nla_put_u8(skb, MDBA_MDB_EATTR_RTPROT, remote->rt_protocol) ||
+	    nla_put_u8(skb, MDBA_MDB_EATTR_GROUP_MODE, remote->filter_mode) ||
+	    vxlan_mdb_entry_info_fill_srcs(skb, remote) ||
+	    vxlan_nla_put_addr(skb, MDBA_MDB_EATTR_DST, &rd->remote_ip))
+		goto nest_err;
+
+	if (rd->remote_port && rd->remote_port != vxlan->cfg.dst_port &&
+	    nla_put_u16(skb, MDBA_MDB_EATTR_DST_PORT,
+			be16_to_cpu(rd->remote_port)))
+		goto nest_err;
+
+	if (rd->remote_vni != vxlan->default_dst.remote_vni &&
+	    nla_put_u32(skb, MDBA_MDB_EATTR_VNI, be32_to_cpu(rd->remote_vni)))
+		goto nest_err;
+
+	if (rd->remote_ifindex &&
+	    nla_put_u32(skb, MDBA_MDB_EATTR_IFINDEX, rd->remote_ifindex))
+		goto nest_err;
+
+	if ((vxlan->cfg.flags & VXLAN_F_COLLECT_METADATA) &&
+	    mdb_entry->key.vni && nla_put_u32(skb, MDBA_MDB_EATTR_SRC_VNI,
+					      be32_to_cpu(mdb_entry->key.vni)))
+		goto nest_err;
+
+	nla_nest_end(skb, nest);
+
+	return 0;
+
+nest_err:
+	nla_nest_cancel(skb, nest);
+	return -EMSGSIZE;
+}
+
+static int vxlan_mdb_entry_fill(const struct vxlan_dev *vxlan,
+				struct sk_buff *skb,
+				struct vxlan_mdb_dump_ctx *ctx,
+				const struct vxlan_mdb_entry *mdb_entry)
+{
+	int remote_idx = 0, s_remote_idx = ctx->remote_idx;
+	struct vxlan_mdb_remote *remote;
+	struct nlattr *nest;
+	int err = 0;
+
+	nest = nla_nest_start_noflag(skb, MDBA_MDB_ENTRY);
+	if (!nest)
+		return -EMSGSIZE;
+
+	list_for_each_entry(remote, &mdb_entry->remotes, list) {
+		if (remote_idx < s_remote_idx)
+			goto skip;
+
+		err = vxlan_mdb_entry_info_fill(vxlan, skb, mdb_entry, remote);
+		if (err)
+			break;
+skip:
+		remote_idx++;
+	}
+
+	ctx->remote_idx = err ? remote_idx : 0;
+	nla_nest_end(skb, nest);
+	return err;
+}
+
+static int vxlan_mdb_fill(const struct vxlan_dev *vxlan, struct sk_buff *skb,
+			  struct vxlan_mdb_dump_ctx *ctx)
+{
+	int entry_idx = 0, s_entry_idx = ctx->entry_idx;
+	struct vxlan_mdb_entry *mdb_entry;
+	struct nlattr *nest;
+	int err = 0;
+
+	nest = nla_nest_start_noflag(skb, MDBA_MDB);
+	if (!nest)
+		return -EMSGSIZE;
+
+	hlist_for_each_entry(mdb_entry, &vxlan->mdb_list, mdb_node) {
+		if (entry_idx < s_entry_idx)
+			goto skip;
+
+		err = vxlan_mdb_entry_fill(vxlan, skb, ctx, mdb_entry);
+		if (err)
+			break;
+skip:
+		entry_idx++;
+	}
+
+	ctx->entry_idx = err ? entry_idx : 0;
+	nla_nest_end(skb, nest);
+	return err;
+}
+
+int vxlan_mdb_dump(struct net_device *dev, struct sk_buff *skb,
+		   struct netlink_callback *cb)
+{
+	struct vxlan_mdb_dump_ctx *ctx = (void *)cb->ctx;
+	struct vxlan_dev *vxlan = netdev_priv(dev);
+	struct br_port_msg *bpm;
+	struct nlmsghdr *nlh;
+	int err;
+
+	ASSERT_RTNL();
+
+	NL_ASSERT_DUMP_CTX_FITS(struct vxlan_mdb_dump_ctx);
+
+	nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid,
+			cb->nlh->nlmsg_seq, RTM_NEWMDB, sizeof(*bpm),
+			NLM_F_MULTI);
+	if (!nlh)
+		return -EMSGSIZE;
+
+	bpm = nlmsg_data(nlh);
+	memset(bpm, 0, sizeof(*bpm));
+	bpm->family = AF_BRIDGE;
+	bpm->ifindex = dev->ifindex;
+
+	err = vxlan_mdb_fill(vxlan, skb, ctx);
+
+	nlmsg_end(skb, nlh);
+
+	cb->seq = vxlan->mdb_seq;
+	nl_dump_check_consistent(cb, nlh);
+
+	return err;
+}
+
+static const struct nla_policy
+vxlan_mdbe_src_list_entry_pol[MDBE_SRCATTR_MAX + 1] = {
+	[MDBE_SRCATTR_ADDRESS] = NLA_POLICY_RANGE(NLA_BINARY,
+						  sizeof(struct in_addr),
+						  sizeof(struct in6_addr)),
+};
+
+static const struct nla_policy
+vxlan_mdbe_src_list_pol[MDBE_SRC_LIST_MAX + 1] = {
+	[MDBE_SRC_LIST_ENTRY] = NLA_POLICY_NESTED(vxlan_mdbe_src_list_entry_pol),
+};
+
+static struct netlink_range_validation vni_range = {
+	.max = VXLAN_N_VID - 1,
+};
+
+static const struct nla_policy vxlan_mdbe_attrs_pol[MDBE_ATTR_MAX + 1] = {
+	[MDBE_ATTR_SOURCE] = NLA_POLICY_RANGE(NLA_BINARY,
+					      sizeof(struct in_addr),
+					      sizeof(struct in6_addr)),
+	[MDBE_ATTR_GROUP_MODE] = NLA_POLICY_RANGE(NLA_U8, MCAST_EXCLUDE,
+						  MCAST_INCLUDE),
+	[MDBE_ATTR_SRC_LIST] = NLA_POLICY_NESTED(vxlan_mdbe_src_list_pol),
+	[MDBE_ATTR_RTPROT] = NLA_POLICY_MIN(NLA_U8, RTPROT_STATIC),
+	[MDBE_ATTR_DST] = NLA_POLICY_RANGE(NLA_BINARY,
+					   sizeof(struct in_addr),
+					   sizeof(struct in6_addr)),
+	[MDBE_ATTR_DST_PORT] = { .type = NLA_U16 },
+	[MDBE_ATTR_VNI] = NLA_POLICY_FULL_RANGE(NLA_U32, &vni_range),
+	[MDBE_ATTR_IFINDEX] = NLA_POLICY_MIN(NLA_S32, 1),
+	[MDBE_ATTR_SRC_VNI] = NLA_POLICY_FULL_RANGE(NLA_U32, &vni_range),
+};
+
+static bool vxlan_mdb_is_valid_source(const struct nlattr *attr, __be16 proto,
+				      struct netlink_ext_ack *extack)
+{
+	switch (proto) {
+	case htons(ETH_P_IP):
+		if (nla_len(attr) != sizeof(struct in_addr)) {
+			NL_SET_ERR_MSG_MOD(extack, "IPv4 invalid source address length");
+			return false;
+		}
+		if (ipv4_is_multicast(nla_get_in_addr(attr))) {
+			NL_SET_ERR_MSG_MOD(extack, "IPv4 multicast source address is not allowed");
+			return false;
+		}
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case htons(ETH_P_IPV6): {
+		struct in6_addr src;
+
+		if (nla_len(attr) != sizeof(struct in6_addr)) {
+			NL_SET_ERR_MSG_MOD(extack, "IPv6 invalid source address length");
+			return false;
+		}
+		src = nla_get_in6_addr(attr);
+		if (ipv6_addr_is_multicast(&src)) {
+			NL_SET_ERR_MSG_MOD(extack, "IPv6 multicast source address is not allowed");
+			return false;
+		}
+		break;
+	}
+#endif
+	default:
+		NL_SET_ERR_MSG_MOD(extack, "Invalid protocol used with source address");
+		return false;
+	}
+
+	return true;
+}
+
+static void vxlan_mdb_config_group_set(struct vxlan_mdb_config *cfg,
+				       const struct br_mdb_entry *entry,
+				       const struct nlattr *source_attr)
+{
+	struct vxlan_mdb_entry_key *group = &cfg->group;
+
+	switch (entry->addr.proto) {
+	case htons(ETH_P_IP):
+		group->dst.sa.sa_family = AF_INET;
+		group->dst.sin.sin_addr.s_addr = entry->addr.u.ip4;
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case htons(ETH_P_IPV6):
+		group->dst.sa.sa_family = AF_INET6;
+		group->dst.sin6.sin6_addr = entry->addr.u.ip6;
+		break;
+#endif
+	}
+
+	if (source_attr)
+		vxlan_nla_get_addr(&group->src, source_attr);
+}
+
+static bool vxlan_mdb_is_star_g(const struct vxlan_mdb_entry_key *group)
+{
+	return !vxlan_addr_any(&group->dst) && vxlan_addr_any(&group->src);
+}
+
+static bool vxlan_mdb_is_sg(const struct vxlan_mdb_entry_key *group)
+{
+	return !vxlan_addr_any(&group->dst) && !vxlan_addr_any(&group->src);
+}
+
+static int vxlan_mdb_config_src_entry_init(struct vxlan_mdb_config *cfg,
+					   __be16 proto,
+					   const struct nlattr *src_entry,
+					   struct netlink_ext_ack *extack)
+{
+	struct nlattr *tb[MDBE_SRCATTR_MAX + 1];
+	struct vxlan_mdb_config_src_entry *src;
+	int err;
+
+	err = nla_parse_nested(tb, MDBE_SRCATTR_MAX, src_entry,
+			       vxlan_mdbe_src_list_entry_pol, extack);
+	if (err)
+		return err;
+
+	if (NL_REQ_ATTR_CHECK(extack, src_entry, tb, MDBE_SRCATTR_ADDRESS))
+		return -EINVAL;
+
+	if (!vxlan_mdb_is_valid_source(tb[MDBE_SRCATTR_ADDRESS], proto,
+				       extack))
+		return -EINVAL;
+
+	src = kzalloc(sizeof(*src), GFP_KERNEL);
+	if (!src)
+		return -ENOMEM;
+
+	err = vxlan_nla_get_addr(&src->addr, tb[MDBE_SRCATTR_ADDRESS]);
+	if (err)
+		goto err_free_src;
+
+	list_add_tail(&src->node, &cfg->src_list);
+
+	return 0;
+
+err_free_src:
+	kfree(src);
+	return err;
+}
+
+static void
+vxlan_mdb_config_src_entry_fini(struct vxlan_mdb_config_src_entry *src)
+{
+	list_del(&src->node);
+	kfree(src);
+}
+
+static int vxlan_mdb_config_src_list_init(struct vxlan_mdb_config *cfg,
+					  __be16 proto,
+					  const struct nlattr *src_list,
+					  struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_config_src_entry *src, *tmp;
+	struct nlattr *src_entry;
+	int rem, err;
+
+	nla_for_each_nested(src_entry, src_list, rem) {
+		err = vxlan_mdb_config_src_entry_init(cfg, proto, src_entry,
+						      extack);
+		if (err)
+			goto err_src_entry_init;
+	}
+
+	return 0;
+
+err_src_entry_init:
+	list_for_each_entry_safe_reverse(src, tmp, &cfg->src_list, node)
+		vxlan_mdb_config_src_entry_fini(src);
+	return err;
+}
+
+static void vxlan_mdb_config_src_list_fini(struct vxlan_mdb_config *cfg)
+{
+	struct vxlan_mdb_config_src_entry *src, *tmp;
+
+	list_for_each_entry_safe_reverse(src, tmp, &cfg->src_list, node)
+		vxlan_mdb_config_src_entry_fini(src);
+}
+
+static int vxlan_mdb_config_attrs_init(struct vxlan_mdb_config *cfg,
+				       const struct br_mdb_entry *entry,
+				       const struct nlattr *set_attrs,
+				       struct netlink_ext_ack *extack)
+{
+	struct nlattr *mdbe_attrs[MDBE_ATTR_MAX + 1];
+	int err;
+
+	err = nla_parse_nested(mdbe_attrs, MDBE_ATTR_MAX, set_attrs,
+			       vxlan_mdbe_attrs_pol, extack);
+	if (err)
+		return err;
+
+	if (NL_REQ_ATTR_CHECK(extack, set_attrs, mdbe_attrs, MDBE_ATTR_DST)) {
+		NL_SET_ERR_MSG_MOD(extack, "Missing remote destination IP address");
+		return -EINVAL;
+	}
+
+	if (mdbe_attrs[MDBE_ATTR_SOURCE] &&
+	    !vxlan_mdb_is_valid_source(mdbe_attrs[MDBE_ATTR_SOURCE],
+				       entry->addr.proto, extack))
+		return -EINVAL;
+
+	vxlan_mdb_config_group_set(cfg, entry, mdbe_attrs[MDBE_ATTR_SOURCE]);
+
+	/* rtnetlink code only validates that IPv4 group address is
+	 * multicast.
+	 */
+	if (!vxlan_addr_is_multicast(&cfg->group.dst) &&
+	    !vxlan_addr_any(&cfg->group.dst)) {
+		NL_SET_ERR_MSG_MOD(extack, "Group address is not multicast");
+		return -EINVAL;
+	}
+
+	if (vxlan_addr_any(&cfg->group.dst) &&
+	    mdbe_attrs[MDBE_ATTR_SOURCE]) {
+		NL_SET_ERR_MSG_MOD(extack, "Source cannot be specified for the all-zeros entry");
+		return -EINVAL;
+	}
+
+	if (vxlan_mdb_is_sg(&cfg->group))
+		cfg->filter_mode = MCAST_INCLUDE;
+
+	if (mdbe_attrs[MDBE_ATTR_GROUP_MODE]) {
+		if (!vxlan_mdb_is_star_g(&cfg->group)) {
+			NL_SET_ERR_MSG_MOD(extack, "Filter mode can only be set for (*, G) entries");
+			return -EINVAL;
+		}
+		cfg->filter_mode = nla_get_u8(mdbe_attrs[MDBE_ATTR_GROUP_MODE]);
+	}
+
+	if (mdbe_attrs[MDBE_ATTR_SRC_LIST]) {
+		if (!vxlan_mdb_is_star_g(&cfg->group)) {
+			NL_SET_ERR_MSG_MOD(extack, "Source list can only be set for (*, G) entries");
+			return -EINVAL;
+		}
+		if (!mdbe_attrs[MDBE_ATTR_GROUP_MODE]) {
+			NL_SET_ERR_MSG_MOD(extack, "Source list cannot be set without filter mode");
+			return -EINVAL;
+		}
+		err = vxlan_mdb_config_src_list_init(cfg, entry->addr.proto,
+						     mdbe_attrs[MDBE_ATTR_SRC_LIST],
+						     extack);
+		if (err)
+			return err;
+	}
+
+	if (vxlan_mdb_is_star_g(&cfg->group) && list_empty(&cfg->src_list) &&
+	    cfg->filter_mode == MCAST_INCLUDE) {
+		NL_SET_ERR_MSG_MOD(extack, "Cannot add (*, G) INCLUDE with an empty source list");
+		return -EINVAL;
+	}
+
+	if (mdbe_attrs[MDBE_ATTR_RTPROT])
+		cfg->rt_protocol = nla_get_u8(mdbe_attrs[MDBE_ATTR_RTPROT]);
+
+	err = vxlan_nla_get_addr(&cfg->remote_ip, mdbe_attrs[MDBE_ATTR_DST]);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Invalid remote destination address");
+		goto err_src_list_fini;
+	}
+
+	if (mdbe_attrs[MDBE_ATTR_DST_PORT])
+		cfg->remote_port =
+			cpu_to_be16(nla_get_u16(mdbe_attrs[MDBE_ATTR_DST_PORT]));
+
+	if (mdbe_attrs[MDBE_ATTR_VNI])
+		cfg->remote_vni =
+			cpu_to_be32(nla_get_u32(mdbe_attrs[MDBE_ATTR_VNI]));
+
+	if (mdbe_attrs[MDBE_ATTR_IFINDEX]) {
+		cfg->remote_ifindex =
+			nla_get_s32(mdbe_attrs[MDBE_ATTR_IFINDEX]);
+		if (!__dev_get_by_index(cfg->vxlan->net, cfg->remote_ifindex)) {
+			NL_SET_ERR_MSG_MOD(extack, "Outgoing interface not found");
+			err = -EINVAL;
+			goto err_src_list_fini;
+		}
+	}
+
+	if (mdbe_attrs[MDBE_ATTR_SRC_VNI])
+		cfg->group.vni =
+			cpu_to_be32(nla_get_u32(mdbe_attrs[MDBE_ATTR_SRC_VNI]));
+
+	return 0;
+
+err_src_list_fini:
+	vxlan_mdb_config_src_list_fini(cfg);
+	return err;
+}
+
+static int vxlan_mdb_config_init(struct vxlan_mdb_config *cfg,
+				 struct net_device *dev, struct nlattr *tb[],
+				 u16 nlmsg_flags,
+				 struct netlink_ext_ack *extack)
+{
+	struct br_mdb_entry *entry = nla_data(tb[MDBA_SET_ENTRY]);
+	struct vxlan_dev *vxlan = netdev_priv(dev);
+
+	memset(cfg, 0, sizeof(*cfg));
+	cfg->vxlan = vxlan;
+	cfg->group.vni = vxlan->default_dst.remote_vni;
+	INIT_LIST_HEAD(&cfg->src_list);
+	cfg->nlflags = nlmsg_flags;
+	cfg->filter_mode = MCAST_EXCLUDE;
+	cfg->rt_protocol = RTPROT_STATIC;
+	cfg->remote_vni = vxlan->default_dst.remote_vni;
+	cfg->remote_port = vxlan->cfg.dst_port;
+
+	if (entry->ifindex != dev->ifindex) {
+		NL_SET_ERR_MSG_MOD(extack, "Port net device must be the VXLAN net device");
+		return -EINVAL;
+	}
+
+	/* State is not part of the entry key and can be ignored on deletion
+	 * requests.
+	 */
+	if ((nlmsg_flags & (NLM_F_CREATE | NLM_F_REPLACE)) &&
+	    entry->state != MDB_PERMANENT) {
+		NL_SET_ERR_MSG_MOD(extack, "MDB entry must be permanent");
+		return -EINVAL;
+	}
+
+	if (entry->flags) {
+		NL_SET_ERR_MSG_MOD(extack, "Invalid MDB entry flags");
+		return -EINVAL;
+	}
+
+	if (entry->vid) {
+		NL_SET_ERR_MSG_MOD(extack, "VID must not be specified");
+		return -EINVAL;
+	}
+
+	if (entry->addr.proto != htons(ETH_P_IP) &&
+	    entry->addr.proto != htons(ETH_P_IPV6)) {
+		NL_SET_ERR_MSG_MOD(extack, "Group address must be an IPv4 / IPv6 address");
+		return -EINVAL;
+	}
+
+	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, MDBA_SET_ENTRY_ATTRS)) {
+		NL_SET_ERR_MSG_MOD(extack, "Missing MDBA_SET_ENTRY_ATTRS attribute");
+		return -EINVAL;
+	}
+
+	return vxlan_mdb_config_attrs_init(cfg, entry, tb[MDBA_SET_ENTRY_ATTRS],
+					   extack);
+}
+
+static void vxlan_mdb_config_fini(struct vxlan_mdb_config *cfg)
+{
+	vxlan_mdb_config_src_list_fini(cfg);
+}
+
+static struct vxlan_mdb_entry *
+vxlan_mdb_entry_lookup(struct vxlan_dev *vxlan,
+		       const struct vxlan_mdb_entry_key *group)
+{
+	return rhashtable_lookup_fast(&vxlan->mdb_tbl, group,
+				      vxlan_mdb_rht_params);
+}
+
+static struct vxlan_mdb_remote *
+vxlan_mdb_remote_lookup(const struct vxlan_mdb_entry *mdb_entry,
+			const union vxlan_addr *addr)
+{
+	struct vxlan_mdb_remote *remote;
+
+	list_for_each_entry(remote, &mdb_entry->remotes, list) {
+		struct vxlan_rdst *rd = rtnl_dereference(remote->rd);
+
+		if (vxlan_addr_equal(addr, &rd->remote_ip))
+			return remote;
+	}
+
+	return NULL;
+}
+
+static void vxlan_mdb_rdst_free(struct rcu_head *head)
+{
+	struct vxlan_rdst *rd = container_of(head, struct vxlan_rdst, rcu);
+
+	dst_cache_destroy(&rd->dst_cache);
+	kfree(rd);
+}
+
+static int vxlan_mdb_remote_rdst_init(const struct vxlan_mdb_config *cfg,
+				      struct vxlan_mdb_remote *remote)
+{
+	struct vxlan_rdst *rd;
+	int err;
+
+	rd = kzalloc(sizeof(*rd), GFP_KERNEL);
+	if (!rd)
+		return -ENOMEM;
+
+	err = dst_cache_init(&rd->dst_cache, GFP_KERNEL);
+	if (err)
+		goto err_free_rdst;
+
+	rd->remote_ip = cfg->remote_ip;
+	rd->remote_port = cfg->remote_port;
+	rd->remote_vni = cfg->remote_vni;
+	rd->remote_ifindex = cfg->remote_ifindex;
+	rcu_assign_pointer(remote->rd, rd);
+
+	return 0;
+
+err_free_rdst:
+	kfree(rd);
+	return err;
+}
+
+static void vxlan_mdb_remote_rdst_fini(struct vxlan_rdst *rd)
+{
+	call_rcu(&rd->rcu, vxlan_mdb_rdst_free);
+}
+
+static int vxlan_mdb_remote_init(const struct vxlan_mdb_config *cfg,
+				 struct vxlan_mdb_remote *remote)
+{
+	int err;
+
+	err = vxlan_mdb_remote_rdst_init(cfg, remote);
+	if (err)
+		return err;
+
+	remote->flags = cfg->flags;
+	remote->filter_mode = cfg->filter_mode;
+	remote->rt_protocol = cfg->rt_protocol;
+	INIT_HLIST_HEAD(&remote->src_list);
+
+	return 0;
+}
+
+static void vxlan_mdb_remote_fini(struct vxlan_dev *vxlan,
+				  struct vxlan_mdb_remote *remote)
+{
+	WARN_ON_ONCE(!hlist_empty(&remote->src_list));
+	vxlan_mdb_remote_rdst_fini(rtnl_dereference(remote->rd));
+}
+
+static struct vxlan_mdb_src_entry *
+vxlan_mdb_remote_src_entry_lookup(const struct vxlan_mdb_remote *remote,
+				  const union vxlan_addr *addr)
+{
+	struct vxlan_mdb_src_entry *ent;
+
+	hlist_for_each_entry(ent, &remote->src_list, node) {
+		if (vxlan_addr_equal(&ent->addr, addr))
+			return ent;
+	}
+
+	return NULL;
+}
+
+static struct vxlan_mdb_src_entry *
+vxlan_mdb_remote_src_entry_add(struct vxlan_mdb_remote *remote,
+			       const union vxlan_addr *addr)
+{
+	struct vxlan_mdb_src_entry *ent;
+
+	ent = kzalloc(sizeof(*ent), GFP_KERNEL);
+	if (!ent)
+		return NULL;
+
+	ent->addr = *addr;
+	hlist_add_head(&ent->node, &remote->src_list);
+
+	return ent;
+}
+
+static void
+vxlan_mdb_remote_src_entry_del(struct vxlan_mdb_src_entry *ent)
+{
+	hlist_del(&ent->node);
+	kfree(ent);
+}
+
+static int
+vxlan_mdb_remote_src_fwd_add(const struct vxlan_mdb_config *cfg,
+			     const union vxlan_addr *addr,
+			     struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_config sg_cfg;
+
+	memset(&sg_cfg, 0, sizeof(sg_cfg));
+	sg_cfg.vxlan = cfg->vxlan;
+	sg_cfg.group.src = *addr;
+	sg_cfg.group.dst = cfg->group.dst;
+	sg_cfg.group.vni = cfg->group.vni;
+	INIT_LIST_HEAD(&sg_cfg.src_list);
+	sg_cfg.remote_ip = cfg->remote_ip;
+	sg_cfg.remote_ifindex = cfg->remote_ifindex;
+	sg_cfg.remote_vni = cfg->remote_vni;
+	sg_cfg.remote_port = cfg->remote_port;
+	sg_cfg.nlflags = cfg->nlflags;
+	sg_cfg.filter_mode = MCAST_INCLUDE;
+	if (cfg->filter_mode == MCAST_EXCLUDE)
+		sg_cfg.flags = VXLAN_MDB_REMOTE_F_BLOCKED;
+	sg_cfg.rt_protocol = cfg->rt_protocol;
+
+	return __vxlan_mdb_add(&sg_cfg, extack);
+}
+
+static void
+vxlan_mdb_remote_src_fwd_del(struct vxlan_dev *vxlan,
+			     const struct vxlan_mdb_entry_key *group,
+			     const struct vxlan_mdb_remote *remote,
+			     const union vxlan_addr *addr)
+{
+	struct vxlan_rdst *rd = rtnl_dereference(remote->rd);
+	struct vxlan_mdb_config sg_cfg;
+
+	memset(&sg_cfg, 0, sizeof(sg_cfg));
+	sg_cfg.vxlan = vxlan;
+	sg_cfg.group.src = *addr;
+	sg_cfg.group.dst = group->dst;
+	sg_cfg.group.vni = group->vni;
+	INIT_LIST_HEAD(&sg_cfg.src_list);
+	sg_cfg.remote_ip = rd->remote_ip;
+
+	__vxlan_mdb_del(&sg_cfg, NULL);
+}
+
+static int
+vxlan_mdb_remote_src_add(const struct vxlan_mdb_config *cfg,
+			 struct vxlan_mdb_remote *remote,
+			 const struct vxlan_mdb_config_src_entry *src,
+			 struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_src_entry *ent;
+	int err;
+
+	ent = vxlan_mdb_remote_src_entry_lookup(remote, &src->addr);
+	if (!ent) {
+		ent = vxlan_mdb_remote_src_entry_add(remote, &src->addr);
+		if (!ent)
+			return -ENOMEM;
+	} else if (!(cfg->nlflags & NLM_F_REPLACE)) {
+		NL_SET_ERR_MSG_MOD(extack, "Source entry already exists");
+		return -EEXIST;
+	}
+
+	err = vxlan_mdb_remote_src_fwd_add(cfg, &ent->addr, extack);
+	if (err)
+		goto err_src_del;
+
+	/* Clear flags in case source entry was marked for deletion as part of
+	 * replace flow.
+	 */
+	ent->flags = 0;
+
+	return 0;
+
+err_src_del:
+	vxlan_mdb_remote_src_entry_del(ent);
+	return err;
+}
+
+static void vxlan_mdb_remote_src_del(struct vxlan_dev *vxlan,
+				     const struct vxlan_mdb_entry_key *group,
+				     const struct vxlan_mdb_remote *remote,
+				     struct vxlan_mdb_src_entry *ent)
+{
+	vxlan_mdb_remote_src_fwd_del(vxlan, group, remote, &ent->addr);
+	vxlan_mdb_remote_src_entry_del(ent);
+}
+
+static int vxlan_mdb_remote_srcs_add(const struct vxlan_mdb_config *cfg,
+				     struct vxlan_mdb_remote *remote,
+				     struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_config_src_entry *src;
+	struct vxlan_mdb_src_entry *ent;
+	struct hlist_node *tmp;
+	int err;
+
+	list_for_each_entry(src, &cfg->src_list, node) {
+		err = vxlan_mdb_remote_src_add(cfg, remote, src, extack);
+		if (err)
+			goto err_src_del;
+	}
+
+	return 0;
+
+err_src_del:
+	hlist_for_each_entry_safe(ent, tmp, &remote->src_list, node)
+		vxlan_mdb_remote_src_del(cfg->vxlan, &cfg->group, remote, ent);
+	return err;
+}
+
+static void vxlan_mdb_remote_srcs_del(struct vxlan_dev *vxlan,
+				      const struct vxlan_mdb_entry_key *group,
+				      struct vxlan_mdb_remote *remote)
+{
+	struct vxlan_mdb_src_entry *ent;
+	struct hlist_node *tmp;
+
+	hlist_for_each_entry_safe(ent, tmp, &remote->src_list, node)
+		vxlan_mdb_remote_src_del(vxlan, group, remote, ent);
+}
+
+static size_t
+vxlan_mdb_nlmsg_src_list_size(const struct vxlan_mdb_entry_key *group,
+			      const struct vxlan_mdb_remote *remote)
+{
+	struct vxlan_mdb_src_entry *ent;
+	size_t nlmsg_size;
+
+	if (hlist_empty(&remote->src_list))
+		return 0;
+
+	/* MDBA_MDB_EATTR_SRC_LIST */
+	nlmsg_size = nla_total_size(0);
+
+	hlist_for_each_entry(ent, &remote->src_list, node) {
+			      /* MDBA_MDB_SRCLIST_ENTRY */
+		nlmsg_size += nla_total_size(0) +
+			      /* MDBA_MDB_SRCATTR_ADDRESS */
+			      nla_total_size(vxlan_addr_size(&group->dst)) +
+			      /* MDBA_MDB_SRCATTR_TIMER */
+			      nla_total_size(sizeof(u8));
+	}
+
+	return nlmsg_size;
+}
+
+static size_t vxlan_mdb_nlmsg_size(const struct vxlan_dev *vxlan,
+				   const struct vxlan_mdb_entry *mdb_entry,
+				   const struct vxlan_mdb_remote *remote)
+{
+	const struct vxlan_mdb_entry_key *group = &mdb_entry->key;
+	struct vxlan_rdst *rd = rtnl_dereference(remote->rd);
+	size_t nlmsg_size;
+
+	nlmsg_size = NLMSG_ALIGN(sizeof(struct br_port_msg)) +
+		     /* MDBA_MDB */
+		     nla_total_size(0) +
+		     /* MDBA_MDB_ENTRY */
+		     nla_total_size(0) +
+		     /* MDBA_MDB_ENTRY_INFO */
+		     nla_total_size(sizeof(struct br_mdb_entry)) +
+		     /* MDBA_MDB_EATTR_TIMER */
+		     nla_total_size(sizeof(u32));
+	/* MDBA_MDB_EATTR_SOURCE */
+	if (vxlan_mdb_is_sg(group))
+		nlmsg_size += nla_total_size(vxlan_addr_size(&group->dst));
+	/* MDBA_MDB_EATTR_RTPROT */
+	nlmsg_size += nla_total_size(sizeof(u8));
+	/* MDBA_MDB_EATTR_SRC_LIST */
+	nlmsg_size += vxlan_mdb_nlmsg_src_list_size(group, remote);
+	/* MDBA_MDB_EATTR_GROUP_MODE */
+	nlmsg_size += nla_total_size(sizeof(u8));
+	/* MDBA_MDB_EATTR_DST */
+	nlmsg_size += nla_total_size(vxlan_addr_size(&rd->remote_ip));
+	/* MDBA_MDB_EATTR_DST_PORT */
+	if (rd->remote_port && rd->remote_port != vxlan->cfg.dst_port)
+		nlmsg_size += nla_total_size(sizeof(u16));
+	/* MDBA_MDB_EATTR_VNI */
+	if (rd->remote_vni != vxlan->default_dst.remote_vni)
+		nlmsg_size += nla_total_size(sizeof(u32));
+	/* MDBA_MDB_EATTR_IFINDEX */
+	if (rd->remote_ifindex)
+		nlmsg_size += nla_total_size(sizeof(u32));
+	/* MDBA_MDB_EATTR_SRC_VNI */
+	if ((vxlan->cfg.flags & VXLAN_F_COLLECT_METADATA) && group->vni)
+		nlmsg_size += nla_total_size(sizeof(u32));
+
+	return nlmsg_size;
+}
+
+static int vxlan_mdb_nlmsg_fill(const struct vxlan_dev *vxlan,
+				struct sk_buff *skb,
+				const struct vxlan_mdb_entry *mdb_entry,
+				const struct vxlan_mdb_remote *remote,
+				int type)
+{
+	struct nlattr *mdb_nest, *mdb_entry_nest;
+	struct br_port_msg *bpm;
+	struct nlmsghdr *nlh;
+
+	nlh = nlmsg_put(skb, 0, 0, type, sizeof(*bpm), 0);
+	if (!nlh)
+		return -EMSGSIZE;
+
+	bpm = nlmsg_data(nlh);
+	memset(bpm, 0, sizeof(*bpm));
+	bpm->family  = AF_BRIDGE;
+	bpm->ifindex = vxlan->dev->ifindex;
+
+	mdb_nest = nla_nest_start_noflag(skb, MDBA_MDB);
+	if (!mdb_nest)
+		goto cancel;
+	mdb_entry_nest = nla_nest_start_noflag(skb, MDBA_MDB_ENTRY);
+	if (!mdb_entry_nest)
+		goto cancel;
+
+	if (vxlan_mdb_entry_info_fill(vxlan, skb, mdb_entry, remote))
+		goto cancel;
+
+	nla_nest_end(skb, mdb_entry_nest);
+	nla_nest_end(skb, mdb_nest);
+	nlmsg_end(skb, nlh);
+
+	return 0;
+
+cancel:
+	nlmsg_cancel(skb, nlh);
+	return -EMSGSIZE;
+}
+
+static void vxlan_mdb_remote_notify(const struct vxlan_dev *vxlan,
+				    const struct vxlan_mdb_entry *mdb_entry,
+				    const struct vxlan_mdb_remote *remote,
+				    int type)
+{
+	struct net *net = dev_net(vxlan->dev);
+	struct sk_buff *skb;
+	int err = -ENOBUFS;
+
+	skb = nlmsg_new(vxlan_mdb_nlmsg_size(vxlan, mdb_entry, remote),
+			GFP_KERNEL);
+	if (!skb)
+		goto errout;
+
+	err = vxlan_mdb_nlmsg_fill(vxlan, skb, mdb_entry, remote, type);
+	if (err) {
+		kfree_skb(skb);
+		goto errout;
+	}
+
+	rtnl_notify(skb, net, 0, RTNLGRP_MDB, NULL, GFP_KERNEL);
+	return;
+errout:
+	rtnl_set_sk_err(net, RTNLGRP_MDB, err);
+}
+
+static int
+vxlan_mdb_remote_srcs_replace(const struct vxlan_mdb_config *cfg,
+			      const struct vxlan_mdb_entry *mdb_entry,
+			      struct vxlan_mdb_remote *remote,
+			      struct netlink_ext_ack *extack)
+{
+	struct vxlan_dev *vxlan = cfg->vxlan;
+	struct vxlan_mdb_src_entry *ent;
+	struct hlist_node *tmp;
+	int err;
+
+	hlist_for_each_entry(ent, &remote->src_list, node)
+		ent->flags |= VXLAN_SGRP_F_DELETE;
+
+	err = vxlan_mdb_remote_srcs_add(cfg, remote, extack);
+	if (err)
+		goto err_clear_delete;
+
+	hlist_for_each_entry_safe(ent, tmp, &remote->src_list, node) {
+		if (ent->flags & VXLAN_SGRP_F_DELETE)
+			vxlan_mdb_remote_src_del(vxlan, &mdb_entry->key, remote,
+						 ent);
+	}
+
+	return 0;
+
+err_clear_delete:
+	hlist_for_each_entry(ent, &remote->src_list, node)
+		ent->flags &= ~VXLAN_SGRP_F_DELETE;
+	return err;
+}
+
+static int vxlan_mdb_remote_replace(const struct vxlan_mdb_config *cfg,
+				    const struct vxlan_mdb_entry *mdb_entry,
+				    struct vxlan_mdb_remote *remote,
+				    struct netlink_ext_ack *extack)
+{
+	struct vxlan_rdst *new_rd, *old_rd = rtnl_dereference(remote->rd);
+	struct vxlan_dev *vxlan = cfg->vxlan;
+	int err;
+
+	err = vxlan_mdb_remote_rdst_init(cfg, remote);
+	if (err)
+		return err;
+	new_rd = rtnl_dereference(remote->rd);
+
+	err = vxlan_mdb_remote_srcs_replace(cfg, mdb_entry, remote, extack);
+	if (err)
+		goto err_rdst_reset;
+
+	WRITE_ONCE(remote->flags, cfg->flags);
+	WRITE_ONCE(remote->filter_mode, cfg->filter_mode);
+	remote->rt_protocol = cfg->rt_protocol;
+	vxlan_mdb_remote_notify(vxlan, mdb_entry, remote, RTM_NEWMDB);
+
+	vxlan_mdb_remote_rdst_fini(old_rd);
+
+	return 0;
+
+err_rdst_reset:
+	rcu_assign_pointer(remote->rd, old_rd);
+	vxlan_mdb_remote_rdst_fini(new_rd);
+	return err;
+}
+
+static int vxlan_mdb_remote_add(const struct vxlan_mdb_config *cfg,
+				struct vxlan_mdb_entry *mdb_entry,
+				struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_remote *remote;
+	int err;
+
+	remote = vxlan_mdb_remote_lookup(mdb_entry, &cfg->remote_ip);
+	if (remote) {
+		if (!(cfg->nlflags & NLM_F_REPLACE)) {
+			NL_SET_ERR_MSG_MOD(extack, "Replace not specified and MDB remote entry already exists");
+			return -EEXIST;
+		}
+		return vxlan_mdb_remote_replace(cfg, mdb_entry, remote, extack);
+	}
+
+	if (!(cfg->nlflags & NLM_F_CREATE)) {
+		NL_SET_ERR_MSG_MOD(extack, "Create not specified and entry does not exist");
+		return -ENOENT;
+	}
+
+	remote = kzalloc(sizeof(*remote), GFP_KERNEL);
+	if (!remote)
+		return -ENOMEM;
+
+	err = vxlan_mdb_remote_init(cfg, remote);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to initialize remote MDB entry");
+		goto err_free_remote;
+	}
+
+	err = vxlan_mdb_remote_srcs_add(cfg, remote, extack);
+	if (err)
+		goto err_remote_fini;
+
+	list_add_rcu(&remote->list, &mdb_entry->remotes);
+	vxlan_mdb_remote_notify(cfg->vxlan, mdb_entry, remote, RTM_NEWMDB);
+
+	return 0;
+
+err_remote_fini:
+	vxlan_mdb_remote_fini(cfg->vxlan, remote);
+err_free_remote:
+	kfree(remote);
+	return err;
+}
+
+static void vxlan_mdb_remote_del(struct vxlan_dev *vxlan,
+				 struct vxlan_mdb_entry *mdb_entry,
+				 struct vxlan_mdb_remote *remote)
+{
+	vxlan_mdb_remote_notify(vxlan, mdb_entry, remote, RTM_DELMDB);
+	list_del_rcu(&remote->list);
+	vxlan_mdb_remote_srcs_del(vxlan, &mdb_entry->key, remote);
+	vxlan_mdb_remote_fini(vxlan, remote);
+	kfree_rcu(remote, rcu);
+}
+
+static struct vxlan_mdb_entry *
+vxlan_mdb_entry_get(struct vxlan_dev *vxlan,
+		    const struct vxlan_mdb_entry_key *group)
+{
+	struct vxlan_mdb_entry *mdb_entry;
+	int err;
+
+	mdb_entry = vxlan_mdb_entry_lookup(vxlan, group);
+	if (mdb_entry)
+		return mdb_entry;
+
+	mdb_entry = kzalloc(sizeof(*mdb_entry), GFP_KERNEL);
+	if (!mdb_entry)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&mdb_entry->remotes);
+	memcpy(&mdb_entry->key, group, sizeof(mdb_entry->key));
+	hlist_add_head(&mdb_entry->mdb_node, &vxlan->mdb_list);
+
+	err = rhashtable_lookup_insert_fast(&vxlan->mdb_tbl,
+					    &mdb_entry->rhnode,
+					    vxlan_mdb_rht_params);
+	if (err)
+		goto err_free_entry;
+
+	return mdb_entry;
+
+err_free_entry:
+	hlist_del(&mdb_entry->mdb_node);
+	kfree(mdb_entry);
+	return ERR_PTR(err);
+}
+
+static void vxlan_mdb_entry_put(struct vxlan_dev *vxlan,
+				struct vxlan_mdb_entry *mdb_entry)
+{
+	if (!list_empty(&mdb_entry->remotes))
+		return;
+
+	rhashtable_remove_fast(&vxlan->mdb_tbl, &mdb_entry->rhnode,
+			       vxlan_mdb_rht_params);
+	hlist_del(&mdb_entry->mdb_node);
+	kfree_rcu(mdb_entry, rcu);
+}
+
+static int __vxlan_mdb_add(const struct vxlan_mdb_config *cfg,
+			   struct netlink_ext_ack *extack)
+{
+	struct vxlan_dev *vxlan = cfg->vxlan;
+	struct vxlan_mdb_entry *mdb_entry;
+	int err;
+
+	mdb_entry = vxlan_mdb_entry_get(vxlan, &cfg->group);
+	if (IS_ERR(mdb_entry))
+		return PTR_ERR(mdb_entry);
+
+	err = vxlan_mdb_remote_add(cfg, mdb_entry, extack);
+	if (err)
+		goto err_entry_put;
+
+	vxlan->mdb_seq++;
+
+	return 0;
+
+err_entry_put:
+	vxlan_mdb_entry_put(vxlan, mdb_entry);
+	return err;
+}
+
+static int __vxlan_mdb_del(const struct vxlan_mdb_config *cfg,
+			   struct netlink_ext_ack *extack)
+{
+	struct vxlan_dev *vxlan = cfg->vxlan;
+	struct vxlan_mdb_entry *mdb_entry;
+	struct vxlan_mdb_remote *remote;
+
+	mdb_entry = vxlan_mdb_entry_lookup(vxlan, &cfg->group);
+	if (!mdb_entry) {
+		NL_SET_ERR_MSG_MOD(extack, "Did not find MDB entry");
+		return -ENOENT;
+	}
+
+	remote = vxlan_mdb_remote_lookup(mdb_entry, &cfg->remote_ip);
+	if (!remote) {
+		NL_SET_ERR_MSG_MOD(extack, "Did not find MDB remote entry");
+		return -ENOENT;
+	}
+
+	vxlan_mdb_remote_del(vxlan, mdb_entry, remote);
+	vxlan_mdb_entry_put(vxlan, mdb_entry);
+
+	vxlan->mdb_seq++;
+
+	return 0;
+}
+
+int vxlan_mdb_add(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
+		  struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_config cfg;
+	int err;
+
+	ASSERT_RTNL();
+
+	err = vxlan_mdb_config_init(&cfg, dev, tb, nlmsg_flags, extack);
+	if (err)
+		return err;
+
+	err = __vxlan_mdb_add(&cfg, extack);
+
+	vxlan_mdb_config_fini(&cfg);
+	return err;
+}
+
+int vxlan_mdb_del(struct net_device *dev, struct nlattr *tb[],
+		  struct netlink_ext_ack *extack)
+{
+	struct vxlan_mdb_config cfg;
+	int err;
+
+	ASSERT_RTNL();
+
+	err = vxlan_mdb_config_init(&cfg, dev, tb, 0, extack);
+	if (err)
+		return err;
+
+	err = __vxlan_mdb_del(&cfg, extack);
+
+	vxlan_mdb_config_fini(&cfg);
+	return err;
+}
+
+static void vxlan_mdb_check_empty(void *ptr, void *arg)
+{
+	WARN_ON_ONCE(1);
+}
+
+static void vxlan_mdb_remotes_flush(struct vxlan_dev *vxlan,
+				    struct vxlan_mdb_entry *mdb_entry)
+{
+	struct vxlan_mdb_remote *remote, *tmp;
+
+	list_for_each_entry_safe(remote, tmp, &mdb_entry->remotes, list)
+		vxlan_mdb_remote_del(vxlan, mdb_entry, remote);
+}
+
+static void vxlan_mdb_entries_flush(struct vxlan_dev *vxlan)
+{
+	struct vxlan_mdb_entry *mdb_entry;
+	struct hlist_node *tmp;
+
+	/* The removal of an entry cannot trigger the removal of another entry
+	 * since entries are always added to the head of the list.
+	 */
+	hlist_for_each_entry_safe(mdb_entry, tmp, &vxlan->mdb_list, mdb_node) {
+		vxlan_mdb_remotes_flush(vxlan, mdb_entry);
+		vxlan_mdb_entry_put(vxlan, mdb_entry);
+	}
+}
+
+int vxlan_mdb_init(struct vxlan_dev *vxlan)
+{
+	int err;
+
+	err = rhashtable_init(&vxlan->mdb_tbl, &vxlan_mdb_rht_params);
+	if (err)
+		return err;
+
+	INIT_HLIST_HEAD(&vxlan->mdb_list);
+
+	return 0;
+}
+
+void vxlan_mdb_fini(struct vxlan_dev *vxlan)
+{
+	vxlan_mdb_entries_flush(vxlan);
+	rhashtable_free_and_destroy(&vxlan->mdb_tbl, vxlan_mdb_check_empty,
+				    NULL);
+}
diff --git a/drivers/net/vxlan/vxlan_private.h b/drivers/net/vxlan/vxlan_private.h
index f4977925cb8a..7bcc38faae27 100644
--- a/drivers/net/vxlan/vxlan_private.h
+++ b/drivers/net/vxlan/vxlan_private.h
@@ -110,6 +110,14 @@ static inline int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
 		return nla_put_in_addr(skb, attr, ip->sin.sin_addr.s_addr);
 }
 
+static inline bool vxlan_addr_is_multicast(const union vxlan_addr *ip)
+{
+	if (ip->sa.sa_family == AF_INET6)
+		return ipv6_addr_is_multicast(&ip->sin6.sin6_addr);
+	else
+		return ipv4_is_multicast(ip->sin.sin_addr.s_addr);
+}
+
 #else /* !CONFIG_IPV6 */
 
 static inline
@@ -138,8 +146,21 @@ static inline int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
 	return nla_put_in_addr(skb, attr, ip->sin.sin_addr.s_addr);
 }
 
+static inline bool vxlan_addr_is_multicast(const union vxlan_addr *ip)
+{
+	return ipv4_is_multicast(ip->sin.sin_addr.s_addr);
+}
+
 #endif
 
+static inline size_t vxlan_addr_size(const union vxlan_addr *ip)
+{
+	if (ip->sa.sa_family == AF_INET6)
+		return sizeof(struct in6_addr);
+	else
+		return sizeof(__be32);
+}
+
 static inline struct vxlan_vni_node *
 vxlan_vnifilter_lookup(struct vxlan_dev *vxlan, __be32 vni)
 {
@@ -206,4 +227,14 @@ int vxlan_igmp_join(struct vxlan_dev *vxlan, union vxlan_addr *rip,
 		    int rifindex);
 int vxlan_igmp_leave(struct vxlan_dev *vxlan, union vxlan_addr *rip,
 		     int rifindex);
+
+/* vxlan_mdb.c */
+int vxlan_mdb_dump(struct net_device *dev, struct sk_buff *skb,
+		   struct netlink_callback *cb);
+int vxlan_mdb_add(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
+		  struct netlink_ext_ack *extack);
+int vxlan_mdb_del(struct net_device *dev, struct nlattr *tb[],
+		  struct netlink_ext_ack *extack);
+int vxlan_mdb_init(struct vxlan_dev *vxlan);
+void vxlan_mdb_fini(struct vxlan_dev *vxlan);
 #endif
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index bca5b01af247..110b703d8978 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -3,6 +3,7 @@
 #define __NET_VXLAN_H 1
 
 #include <linux/if_vlan.h>
+#include <linux/rhashtable-types.h>
 #include <net/udp_tunnel.h>
 #include <net/dst_metadata.h>
 #include <net/rtnetlink.h>
@@ -302,6 +303,10 @@ struct vxlan_dev {
 	struct vxlan_vni_group  __rcu *vnigrp;
 
 	struct hlist_head fdb_head[FDB_HASH_SIZE];
+
+	struct rhashtable mdb_tbl;
+	struct hlist_head mdb_list;
+	unsigned int mdb_seq;
 };
 
 #define VXLAN_F_LEARN			0x01
diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index d60c456710b3..c9d624f528c5 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -633,6 +633,11 @@ enum {
 	MDBA_MDB_EATTR_GROUP_MODE,
 	MDBA_MDB_EATTR_SOURCE,
 	MDBA_MDB_EATTR_RTPROT,
+	MDBA_MDB_EATTR_DST,
+	MDBA_MDB_EATTR_DST_PORT,
+	MDBA_MDB_EATTR_VNI,
+	MDBA_MDB_EATTR_IFINDEX,
+	MDBA_MDB_EATTR_SRC_VNI,
 	__MDBA_MDB_EATTR_MAX
 };
 #define MDBA_MDB_EATTR_MAX (__MDBA_MDB_EATTR_MAX - 1)
@@ -728,6 +733,11 @@ enum {
 	MDBE_ATTR_SRC_LIST,
 	MDBE_ATTR_GROUP_MODE,
 	MDBE_ATTR_RTPROT,
+	MDBE_ATTR_DST,
+	MDBE_ATTR_DST_PORT,
+	MDBE_ATTR_VNI,
+	MDBE_ATTR_IFINDEX,
+	MDBE_ATTR_SRC_VNI,
 	__MDBE_ATTR_MAX,
 };
 #define MDBE_ATTR_MAX (__MDBE_ATTR_MAX - 1)
-- 
cgit v1.2.3-70-g09d2


From bc6c6b013ffee36eb555cc0a68aa3d9608e1fad2 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:52 +0200
Subject: vxlan: mdb: Add an internal flag to indicate MDB usage

Add an internal flag to indicate whether MDB entries are configured or
not. Set the flag after installing the first MDB entry and clear it
before deleting the last one.

The flag will be consulted by the data path which will only perform an
MDB lookup if the flag is set, thereby keeping the MDB overhead to a
minimum when the MDB is not used.

Another option would have been to use a static key, but it is global and
not per-device, unlike the current approach.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan/vxlan_mdb.c | 7 +++++++
 include/net/vxlan.h           | 1 +
 2 files changed, 8 insertions(+)

diff --git a/drivers/net/vxlan/vxlan_mdb.c b/drivers/net/vxlan/vxlan_mdb.c
index 129692b3663f..b32b1fb4a74a 100644
--- a/drivers/net/vxlan/vxlan_mdb.c
+++ b/drivers/net/vxlan/vxlan_mdb.c
@@ -1185,6 +1185,9 @@ vxlan_mdb_entry_get(struct vxlan_dev *vxlan,
 	if (err)
 		goto err_free_entry;
 
+	if (hlist_is_singular_node(&mdb_entry->mdb_node, &vxlan->mdb_list))
+		vxlan->cfg.flags |= VXLAN_F_MDB;
+
 	return mdb_entry;
 
 err_free_entry:
@@ -1199,6 +1202,9 @@ static void vxlan_mdb_entry_put(struct vxlan_dev *vxlan,
 	if (!list_empty(&mdb_entry->remotes))
 		return;
 
+	if (hlist_is_singular_node(&mdb_entry->mdb_node, &vxlan->mdb_list))
+		vxlan->cfg.flags &= ~VXLAN_F_MDB;
+
 	rhashtable_remove_fast(&vxlan->mdb_tbl, &mdb_entry->rhnode,
 			       vxlan_mdb_rht_params);
 	hlist_del(&mdb_entry->mdb_node);
@@ -1336,6 +1342,7 @@ int vxlan_mdb_init(struct vxlan_dev *vxlan)
 void vxlan_mdb_fini(struct vxlan_dev *vxlan)
 {
 	vxlan_mdb_entries_flush(vxlan);
+	WARN_ON_ONCE(vxlan->cfg.flags & VXLAN_F_MDB);
 	rhashtable_free_and_destroy(&vxlan->mdb_tbl, vxlan_mdb_check_empty,
 				    NULL);
 }
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 110b703d8978..b7b2e9abfb37 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -327,6 +327,7 @@ struct vxlan_dev {
 #define VXLAN_F_IPV6_LINKLOCAL		0x8000
 #define VXLAN_F_TTL_INHERIT		0x10000
 #define VXLAN_F_VNIFILTER               0x20000
+#define VXLAN_F_MDB			0x40000
 
 /* Flags that are used in the receive path. These flags must match in
  * order for a socket to be shareable
-- 
cgit v1.2.3-70-g09d2


From 0f83e69f44bf8dc8ab48ff0196b3475c1f0f6c07 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:53 +0200
Subject: vxlan: Add MDB data path support

Integrate MDB support into the Tx path of the VXLAN driver, allowing it
to selectively forward IP multicast traffic according to the matched MDB
entry.

If MDB entries are configured (i.e., 'VXLAN_F_MDB' is set) and the
packet is an IP multicast packet, perform up to three different lookups
according to the following priority:

1. For an (S, G) entry, using {Source VNI, Source IP, Destination IP}.
2. For a (*, G) entry, using {Source VNI, Destination IP}.
3. For the catchall MDB entry (0.0.0.0 or ::), using the source VNI.

The catchall MDB entry is similar to the catchall FDB entry
(00:00:00:00:00:00) that is currently used to transmit BUM (broadcast,
unknown unicast and multicast) traffic. However, unlike the catchall FDB
entry, this entry is only used to transmit unregistered IP multicast
traffic that is not link-local. Therefore, when configured, the catchall
FDB entry will only transmit BULL (broadcast, unknown unicast,
link-local multicast) traffic.

The catchall MDB entry is useful in deployments where inter-subnet
multicast forwarding is used and not all the VTEPs in a tenant domain
are members in all the broadcast domains. In such deployments it is
advantageous to transmit BULL (broadcast, unknown unicast and link-local
multicast) and unregistered IP multicast traffic on different tunnels.
If the same tunnel was used, a VTEP only interested in IP multicast
traffic would also pull all the BULL traffic and drop it as it is not a
member in the originating broadcast domain [1].

If the packet did not match an MDB entry (or if the packet is not an IP
multicast packet), return it to the Tx path, allowing it to be forwarded
according to the FDB.

If the packet did match an MDB entry, forward it to the associated
remote VTEPs. However, if the entry is a (*, G) entry and the associated
remote is in INCLUDE mode, then skip over it as the source IP is not in
its source list (otherwise the packet would have matched on an (S, G)
entry). Similarly, if the associated remote is marked as BLOCKED (can
only be set on (S, G) entries), then skip over it as well as the remote
is in EXCLUDE mode and the source IP is in its source list.

[1] https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-irb-mcast#section-2.6

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan/vxlan_core.c    |  15 +++++
 drivers/net/vxlan/vxlan_mdb.c     | 114 ++++++++++++++++++++++++++++++++++++++
 drivers/net/vxlan/vxlan_private.h |   6 ++
 3 files changed, 135 insertions(+)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index a8b26d4f76de..8450768d2300 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -2743,6 +2743,21 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
 #endif
 	}
 
+	if (vxlan->cfg.flags & VXLAN_F_MDB) {
+		struct vxlan_mdb_entry *mdb_entry;
+
+		rcu_read_lock();
+		mdb_entry = vxlan_mdb_entry_skb_get(vxlan, skb, vni);
+		if (mdb_entry) {
+			netdev_tx_t ret;
+
+			ret = vxlan_mdb_xmit(vxlan, mdb_entry, skb);
+			rcu_read_unlock();
+			return ret;
+		}
+		rcu_read_unlock();
+	}
+
 	eth = eth_hdr(skb);
 	f = vxlan_find_mac(vxlan, eth->h_dest, vni);
 	did_rsc = false;
diff --git a/drivers/net/vxlan/vxlan_mdb.c b/drivers/net/vxlan/vxlan_mdb.c
index b32b1fb4a74a..5e041622261a 100644
--- a/drivers/net/vxlan/vxlan_mdb.c
+++ b/drivers/net/vxlan/vxlan_mdb.c
@@ -1298,6 +1298,120 @@ int vxlan_mdb_del(struct net_device *dev, struct nlattr *tb[],
 	return err;
 }
 
+struct vxlan_mdb_entry *vxlan_mdb_entry_skb_get(struct vxlan_dev *vxlan,
+						struct sk_buff *skb,
+						__be32 src_vni)
+{
+	struct vxlan_mdb_entry *mdb_entry;
+	struct vxlan_mdb_entry_key group;
+
+	if (!is_multicast_ether_addr(eth_hdr(skb)->h_dest) ||
+	    is_broadcast_ether_addr(eth_hdr(skb)->h_dest))
+		return NULL;
+
+	/* When not in collect metadata mode, 'src_vni' is zero, but MDB
+	 * entries are stored with the VNI of the VXLAN device.
+	 */
+	if (!(vxlan->cfg.flags & VXLAN_F_COLLECT_METADATA))
+		src_vni = vxlan->default_dst.remote_vni;
+
+	memset(&group, 0, sizeof(group));
+	group.vni = src_vni;
+
+	switch (skb->protocol) {
+	case htons(ETH_P_IP):
+		if (!pskb_may_pull(skb, sizeof(struct iphdr)))
+			return NULL;
+		group.dst.sa.sa_family = AF_INET;
+		group.dst.sin.sin_addr.s_addr = ip_hdr(skb)->daddr;
+		group.src.sa.sa_family = AF_INET;
+		group.src.sin.sin_addr.s_addr = ip_hdr(skb)->saddr;
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case htons(ETH_P_IPV6):
+		if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
+			return NULL;
+		group.dst.sa.sa_family = AF_INET6;
+		group.dst.sin6.sin6_addr = ipv6_hdr(skb)->daddr;
+		group.src.sa.sa_family = AF_INET6;
+		group.src.sin6.sin6_addr = ipv6_hdr(skb)->saddr;
+		break;
+#endif
+	default:
+		return NULL;
+	}
+
+	mdb_entry = vxlan_mdb_entry_lookup(vxlan, &group);
+	if (mdb_entry)
+		return mdb_entry;
+
+	memset(&group.src, 0, sizeof(group.src));
+	mdb_entry = vxlan_mdb_entry_lookup(vxlan, &group);
+	if (mdb_entry)
+		return mdb_entry;
+
+	/* No (S, G) or (*, G) found. Look up the all-zeros entry, but only if
+	 * the destination IP address is not link-local multicast since we want
+	 * to transmit such traffic together with broadcast and unknown unicast
+	 * traffic.
+	 */
+	switch (skb->protocol) {
+	case htons(ETH_P_IP):
+		if (ipv4_is_local_multicast(group.dst.sin.sin_addr.s_addr))
+			return NULL;
+		group.dst.sin.sin_addr.s_addr = 0;
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case htons(ETH_P_IPV6):
+		if (ipv6_addr_type(&group.dst.sin6.sin6_addr) &
+		    IPV6_ADDR_LINKLOCAL)
+			return NULL;
+		memset(&group.dst.sin6.sin6_addr, 0,
+		       sizeof(group.dst.sin6.sin6_addr));
+		break;
+#endif
+	default:
+		return NULL;
+	}
+
+	return vxlan_mdb_entry_lookup(vxlan, &group);
+}
+
+netdev_tx_t vxlan_mdb_xmit(struct vxlan_dev *vxlan,
+			   const struct vxlan_mdb_entry *mdb_entry,
+			   struct sk_buff *skb)
+{
+	struct vxlan_mdb_remote *remote, *fremote = NULL;
+	__be32 src_vni = mdb_entry->key.vni;
+
+	list_for_each_entry_rcu(remote, &mdb_entry->remotes, list) {
+		struct sk_buff *skb1;
+
+		if ((vxlan_mdb_is_star_g(&mdb_entry->key) &&
+		     READ_ONCE(remote->filter_mode) == MCAST_INCLUDE) ||
+		    (READ_ONCE(remote->flags) & VXLAN_MDB_REMOTE_F_BLOCKED))
+			continue;
+
+		if (!fremote) {
+			fremote = remote;
+			continue;
+		}
+
+		skb1 = skb_clone(skb, GFP_ATOMIC);
+		if (skb1)
+			vxlan_xmit_one(skb1, vxlan->dev, src_vni,
+				       rcu_dereference(remote->rd), false);
+	}
+
+	if (fremote)
+		vxlan_xmit_one(skb, vxlan->dev, src_vni,
+			       rcu_dereference(fremote->rd), false);
+	else
+		kfree_skb(skb);
+
+	return NETDEV_TX_OK;
+}
+
 static void vxlan_mdb_check_empty(void *ptr, void *arg)
 {
 	WARN_ON_ONCE(1);
diff --git a/drivers/net/vxlan/vxlan_private.h b/drivers/net/vxlan/vxlan_private.h
index 7bcc38faae27..817fa3075842 100644
--- a/drivers/net/vxlan/vxlan_private.h
+++ b/drivers/net/vxlan/vxlan_private.h
@@ -235,6 +235,12 @@ int vxlan_mdb_add(struct net_device *dev, struct nlattr *tb[], u16 nlmsg_flags,
 		  struct netlink_ext_ack *extack);
 int vxlan_mdb_del(struct net_device *dev, struct nlattr *tb[],
 		  struct netlink_ext_ack *extack);
+struct vxlan_mdb_entry *vxlan_mdb_entry_skb_get(struct vxlan_dev *vxlan,
+						struct sk_buff *skb,
+						__be32 src_vni);
+netdev_tx_t vxlan_mdb_xmit(struct vxlan_dev *vxlan,
+			   const struct vxlan_mdb_entry *mdb_entry,
+			   struct sk_buff *skb);
 int vxlan_mdb_init(struct vxlan_dev *vxlan);
 void vxlan_mdb_fini(struct vxlan_dev *vxlan);
 #endif
-- 
cgit v1.2.3-70-g09d2


From 08f876a7d79ed235f90af0373d1e548a71c1f4f6 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:54 +0200
Subject: vxlan: Enable MDB support

Now that the VXLAN MDB control and data paths are in place we can expose
the VXLAN MDB functionality to user space.

Set the VXLAN MDB net device operations to the appropriate functions,
thereby allowing the rtnetlink code to reach the VXLAN driver.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/vxlan/vxlan_core.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index 8450768d2300..e2e5f5dac7e6 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -3083,6 +3083,9 @@ static const struct net_device_ops vxlan_netdev_ether_ops = {
 	.ndo_fdb_del		= vxlan_fdb_delete,
 	.ndo_fdb_dump		= vxlan_fdb_dump,
 	.ndo_fdb_get		= vxlan_fdb_get,
+	.ndo_mdb_add		= vxlan_mdb_add,
+	.ndo_mdb_del		= vxlan_mdb_del,
+	.ndo_mdb_dump		= vxlan_mdb_dump,
 	.ndo_fill_metadata_dst	= vxlan_fill_metadata_dst,
 };
 
-- 
cgit v1.2.3-70-g09d2


From 62199e3f16583e766f46d1767deca109fd8ea408 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 15 Mar 2023 15:11:55 +0200
Subject: selftests: net: Add VXLAN MDB test

Add test cases for VXLAN MDB, testing the control and data paths. Two
different sets of namespaces (i.e., ns{1,2}_v4 and ns{1,2}_v6) are used
in order to test VXLAN MDB with both IPv4 and IPv6 underlays,
respectively.

Example truncated output:

 # ./test_vxlan_mdb.sh
 [...]
 Tests passed: 620
 Tests failed:   0

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/Makefile          |    1 +
 tools/testing/selftests/net/config            |    1 +
 tools/testing/selftests/net/test_vxlan_mdb.sh | 2318 +++++++++++++++++++++++++
 3 files changed, 2320 insertions(+)
 create mode 100755 tools/testing/selftests/net/test_vxlan_mdb.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 099741290184..a179fbd6f972 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -81,6 +81,7 @@ TEST_GEN_FILES += sctp_hello
 TEST_GEN_FILES += csum
 TEST_GEN_FILES += nat6to4.o
 TEST_GEN_FILES += ip_local_port_range
+TEST_PROGS += test_vxlan_mdb.sh
 
 TEST_FILES := settings
 
diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config
index cc9fd55ab869..4c7ce07afa2f 100644
--- a/tools/testing/selftests/net/config
+++ b/tools/testing/selftests/net/config
@@ -48,3 +48,4 @@ CONFIG_BAREUDP=m
 CONFIG_IPV6_IOAM6_LWTUNNEL=y
 CONFIG_CRYPTO_SM4_GENERIC=y
 CONFIG_AMT=m
+CONFIG_VXLAN=m
diff --git a/tools/testing/selftests/net/test_vxlan_mdb.sh b/tools/testing/selftests/net/test_vxlan_mdb.sh
new file mode 100755
index 000000000000..31e5f0f8859d
--- /dev/null
+++ b/tools/testing/selftests/net/test_vxlan_mdb.sh
@@ -0,0 +1,2318 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# This test is for checking VXLAN MDB functionality. The topology consists of
+# two sets of namespaces: One for the testing of IPv4 underlay and another for
+# IPv6. In both cases, both IPv4 and IPv6 overlay traffic are tested.
+#
+# Data path functionality is tested by sending traffic from one of the upper
+# namespaces and checking using ingress tc filters that the expected traffic
+# was received by one of the lower namespaces.
+#
+# +------------------------------------+ +------------------------------------+
+# | ns1_v4                             | | ns1_v6                             |
+# |                                    | |                                    |
+# |    br0.10    br0.4000  br0.20      | |    br0.10    br0.4000  br0.20      |
+# |       +         +         +        | |       +         +         +        |
+# |       |         |         |        | |       |         |         |        |
+# |       |         |         |        | |       |         |         |        |
+# |       +---------+---------+        | |       +---------+---------+        |
+# |                 |                  | |                 |                  |
+# |                 |                  | |                 |                  |
+# |                 +                  | |                 +                  |
+# |                br0                 | |                br0                 |
+# |                 +                  | |                 +                  |
+# |                 |                  | |                 |                  |
+# |                 |                  | |                 |                  |
+# |                 +                  | |                 +                  |
+# |                vx0                 | |                vx0                 |
+# |                                    | |                                    |
+# |                                    | |                                    |
+# |               veth0                | |               veth0                |
+# |                 +                  | |                 +                  |
+# +-----------------|------------------+ +-----------------|------------------+
+#                   |                                      |
+# +-----------------|------------------+ +-----------------|------------------+
+# |                 +                  | |                 +                  |
+# |               veth0                | |               veth0                |
+# |                                    | |                                    |
+# |                                    | |                                    |
+# |                vx0                 | |                vx0                 |
+# |                 +                  | |                 +                  |
+# |                 |                  | |                 |                  |
+# |                 |                  | |                 |                  |
+# |                 +                  | |                 +                  |
+# |                br0                 | |                br0                 |
+# |                 +                  | |                 +                  |
+# |                 |                  | |                 |                  |
+# |                 |                  | |                 |                  |
+# |       +---------+---------+        | |       +---------+---------+        |
+# |       |         |         |        | |       |         |         |        |
+# |       |         |         |        | |       |         |         |        |
+# |       +         +         +        | |       +         +         +        |
+# |    br0.10    br0.4000  br0.10      | |    br0.10    br0.4000  br0.20      |
+# |                                    | |                                    |
+# | ns2_v4                             | | ns2_v6                             |
+# +------------------------------------+ +------------------------------------+
+
+ret=0
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+CONTROL_PATH_TESTS="
+	basic_star_g_ipv4_ipv4
+	basic_star_g_ipv6_ipv4
+	basic_star_g_ipv4_ipv6
+	basic_star_g_ipv6_ipv6
+	basic_sg_ipv4_ipv4
+	basic_sg_ipv6_ipv4
+	basic_sg_ipv4_ipv6
+	basic_sg_ipv6_ipv6
+	star_g_ipv4_ipv4
+	star_g_ipv6_ipv4
+	star_g_ipv4_ipv6
+	star_g_ipv6_ipv6
+	sg_ipv4_ipv4
+	sg_ipv6_ipv4
+	sg_ipv4_ipv6
+	sg_ipv6_ipv6
+	dump_ipv4_ipv4
+	dump_ipv6_ipv4
+	dump_ipv4_ipv6
+	dump_ipv6_ipv6
+"
+
+DATA_PATH_TESTS="
+	encap_params_ipv4_ipv4
+	encap_params_ipv6_ipv4
+	encap_params_ipv4_ipv6
+	encap_params_ipv6_ipv6
+	starg_exclude_ir_ipv4_ipv4
+	starg_exclude_ir_ipv6_ipv4
+	starg_exclude_ir_ipv4_ipv6
+	starg_exclude_ir_ipv6_ipv6
+	starg_include_ir_ipv4_ipv4
+	starg_include_ir_ipv6_ipv4
+	starg_include_ir_ipv4_ipv6
+	starg_include_ir_ipv6_ipv6
+	starg_exclude_p2mp_ipv4_ipv4
+	starg_exclude_p2mp_ipv6_ipv4
+	starg_exclude_p2mp_ipv4_ipv6
+	starg_exclude_p2mp_ipv6_ipv6
+	starg_include_p2mp_ipv4_ipv4
+	starg_include_p2mp_ipv6_ipv4
+	starg_include_p2mp_ipv4_ipv6
+	starg_include_p2mp_ipv6_ipv6
+	egress_vni_translation_ipv4_ipv4
+	egress_vni_translation_ipv6_ipv4
+	egress_vni_translation_ipv4_ipv6
+	egress_vni_translation_ipv6_ipv6
+	all_zeros_mdb_ipv4
+	all_zeros_mdb_ipv6
+	mdb_fdb_ipv4_ipv4
+	mdb_fdb_ipv6_ipv4
+	mdb_fdb_ipv4_ipv6
+	mdb_fdb_ipv6_ipv6
+	mdb_torture_ipv4_ipv4
+	mdb_torture_ipv6_ipv4
+	mdb_torture_ipv4_ipv6
+	mdb_torture_ipv6_ipv6
+"
+
+# All tests in this script. Can be overridden with -t option.
+TESTS="
+	$CONTROL_PATH_TESTS
+	$DATA_PATH_TESTS
+"
+VERBOSE=0
+PAUSE_ON_FAIL=no
+PAUSE=no
+
+################################################################################
+# Utilities
+
+log_test()
+{
+	local rc=$1
+	local expected=$2
+	local msg="$3"
+
+	if [ ${rc} -eq ${expected} ]; then
+		printf "TEST: %-60s  [ OK ]\n" "${msg}"
+		nsuccess=$((nsuccess+1))
+	else
+		ret=1
+		nfail=$((nfail+1))
+		printf "TEST: %-60s  [FAIL]\n" "${msg}"
+		if [ "$VERBOSE" = "1" ]; then
+			echo "    rc=$rc, expected $expected"
+		fi
+
+		if [ "${PAUSE_ON_FAIL}" = "yes" ]; then
+		echo
+			echo "hit enter to continue, 'q' to quit"
+			read a
+			[ "$a" = "q" ] && exit 1
+		fi
+	fi
+
+	if [ "${PAUSE}" = "yes" ]; then
+		echo
+		echo "hit enter to continue, 'q' to quit"
+		read a
+		[ "$a" = "q" ] && exit 1
+	fi
+
+	[ "$VERBOSE" = "1" ] && echo
+}
+
+run_cmd()
+{
+	local cmd="$1"
+	local out
+	local stderr="2>/dev/null"
+
+	if [ "$VERBOSE" = "1" ]; then
+		printf "COMMAND: $cmd\n"
+		stderr=
+	fi
+
+	out=$(eval $cmd $stderr)
+	rc=$?
+	if [ "$VERBOSE" = "1" -a -n "$out" ]; then
+		echo "    $out"
+	fi
+
+	return $rc
+}
+
+tc_check_packets()
+{
+	local ns=$1; shift
+	local id=$1; shift
+	local handle=$1; shift
+	local count=$1; shift
+	local pkts
+
+	sleep 0.1
+	pkts=$(tc -n $ns -j -s filter show $id \
+		| jq ".[] | select(.options.handle == $handle) | \
+		.options.actions[0].stats.packets")
+	[[ $pkts == $count ]]
+}
+
+################################################################################
+# Setup
+
+setup_common_ns()
+{
+	local ns=$1; shift
+	local local_addr=$1; shift
+
+	ip netns exec $ns sysctl -qw net.ipv4.ip_forward=1
+	ip netns exec $ns sysctl -qw net.ipv4.fib_multipath_use_neigh=1
+	ip netns exec $ns sysctl -qw net.ipv4.conf.default.ignore_routes_with_linkdown=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.all.keep_addr_on_down=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.all.forwarding=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.default.forwarding=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.default.ignore_routes_with_linkdown=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.all.accept_dad=0
+	ip netns exec $ns sysctl -qw net.ipv6.conf.default.accept_dad=0
+
+	ip -n $ns link set dev lo up
+	ip -n $ns address add $local_addr dev lo
+
+	ip -n $ns link set dev veth0 up
+
+	ip -n $ns link add name br0 up type bridge vlan_filtering 1 \
+		vlan_default_pvid 0 mcast_snooping 0
+
+	ip -n $ns link add link br0 name br0.10 up type vlan id 10
+	bridge -n $ns vlan add vid 10 dev br0 self
+
+	ip -n $ns link add link br0 name br0.20 up type vlan id 20
+	bridge -n $ns vlan add vid 20 dev br0 self
+
+	ip -n $ns link add link br0 name br0.4000 up type vlan id 4000
+	bridge -n $ns vlan add vid 4000 dev br0 self
+
+	ip -n $ns link add name vx0 up master br0 type vxlan \
+		local $local_addr dstport 4789 external vnifilter
+	bridge -n $ns link set dev vx0 vlan_tunnel on
+
+	bridge -n $ns vlan add vid 10 dev vx0
+	bridge -n $ns vlan add vid 10 dev vx0 tunnel_info id 10010
+	bridge -n $ns vni add vni 10010 dev vx0
+
+	bridge -n $ns vlan add vid 20 dev vx0
+	bridge -n $ns vlan add vid 20 dev vx0 tunnel_info id 10020
+	bridge -n $ns vni add vni 10020 dev vx0
+
+	bridge -n $ns vlan add vid 4000 dev vx0 pvid
+	bridge -n $ns vlan add vid 4000 dev vx0 tunnel_info id 14000
+	bridge -n $ns vni add vni 14000 dev vx0
+}
+
+setup_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local local_addr1=$1; shift
+	local local_addr2=$1; shift
+
+	ip netns add $ns1
+	ip netns add $ns2
+
+	ip link add name veth0 type veth peer name veth1
+	ip link set dev veth0 netns $ns1 name veth0
+	ip link set dev veth1 netns $ns2 name veth0
+
+	setup_common_ns $ns1 $local_addr1
+	setup_common_ns $ns2 $local_addr2
+}
+
+setup_v4()
+{
+	setup_common ns1_v4 ns2_v4 192.0.2.1 192.0.2.2
+
+	ip -n ns1_v4 address add 192.0.2.17/28 dev veth0
+	ip -n ns2_v4 address add 192.0.2.18/28 dev veth0
+
+	ip -n ns1_v4 route add default via 192.0.2.18
+	ip -n ns2_v4 route add default via 192.0.2.17
+}
+
+cleanup_v4()
+{
+	ip netns del ns2_v4
+	ip netns del ns1_v4
+}
+
+setup_v6()
+{
+	setup_common ns1_v6 ns2_v6 2001:db8:1::1 2001:db8:1::2
+
+	ip -n ns1_v6 address add 2001:db8:2::1/64 dev veth0 nodad
+	ip -n ns2_v6 address add 2001:db8:2::2/64 dev veth0 nodad
+
+	ip -n ns1_v6 route add default via 2001:db8:2::2
+	ip -n ns2_v6 route add default via 2001:db8:2::1
+}
+
+cleanup_v6()
+{
+	ip netns del ns2_v6
+	ip netns del ns1_v6
+}
+
+setup()
+{
+	set -e
+
+	setup_v4
+	setup_v6
+
+	sleep 5
+
+	set +e
+}
+
+cleanup()
+{
+	cleanup_v6 &> /dev/null
+	cleanup_v4 &> /dev/null
+}
+
+################################################################################
+# Tests - Control path
+
+basic_common()
+{
+	local ns1=$1; shift
+	local grp_key=$1; shift
+	local vtep_ip=$1; shift
+
+	# Test basic control path operations common to all MDB entry types.
+
+	# Basic add, replace and delete behavior.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	log_test $? 0 "MDB entry addition"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\""
+	log_test $? 0 "MDB entry presence after addition"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	log_test $? 0 "MDB entry replacement"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\""
+	log_test $? 0 "MDB entry presence after replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+	log_test $? 0 "MDB entry deletion"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\""
+	log_test $? 1 "MDB entry presence after deletion"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+	log_test $? 255 "Non-existent MDB entry deletion"
+
+	# Default protocol and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \"proto static\""
+	log_test $? 0 "MDB entry default protocol"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 $grp_key permanent proto 123 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \"proto 123\""
+	log_test $? 0 "MDB entry protocol replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+
+	# Default destination port and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \" dst_port \""
+	log_test $? 1 "MDB entry default destination port"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 $grp_key permanent dst $vtep_ip dst_port 1234 src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \"dst_port 1234\""
+	log_test $? 0 "MDB entry destination port replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+
+	# Default destination VNI and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \" vni \""
+	log_test $? 1 "MDB entry default destination VNI"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 $grp_key permanent dst $vtep_ip vni 1234 src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \"vni 1234\""
+	log_test $? 0 "MDB entry destination VNI replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+
+	# Default outgoing interface and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \" via \""
+	log_test $? 1 "MDB entry default outgoing interface"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010 via veth0"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep \"$grp_key\" | grep \"via veth0\""
+	log_test $? 0 "MDB entry outgoing interface replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+
+	# Common error cases.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port veth0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	log_test $? 255 "MDB entry with mismatch between device and port"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key temp dst $vtep_ip src_vni 10010"
+	log_test $? 255 "MDB entry with temp state"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent vid 10 dst $vtep_ip src_vni 10010"
+	log_test $? 255 "MDB entry with VLAN"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp 01:02:03:04:05:06 permanent dst $vtep_ip src_vni 10010"
+	log_test $? 255 "MDB entry MAC address"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent"
+	log_test $? 255 "MDB entry without extended parameters"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent proto 3 dst $vtep_ip src_vni 10010"
+	log_test $? 255 "MDB entry with an invalid protocol"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip vni $((2 ** 24)) src_vni 10010"
+	log_test $? 255 "MDB entry with an invalid destination VNI"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni $((2 ** 24))"
+	log_test $? 255 "MDB entry with an invalid source VNI"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent src_vni 10010"
+	log_test $? 255 "MDB entry without a remote destination IP"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 $grp_key permanent dst $vtep_ip src_vni 10010"
+	log_test $? 255 "Duplicate MDB entries"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 $grp_key dst $vtep_ip src_vni 10010"
+}
+
+basic_star_g_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local grp_key="grp 239.1.1.1"
+	local vtep_ip=198.51.100.100
+
+	echo
+	echo "Control path: Basic (*, G) operations - IPv4 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_star_g_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local grp_key="grp ff0e::1"
+	local vtep_ip=198.51.100.100
+
+	echo
+	echo "Control path: Basic (*, G) operations - IPv6 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_star_g_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local grp_key="grp 239.1.1.1"
+	local vtep_ip=2001:db8:1000::1
+
+	echo
+	echo "Control path: Basic (*, G) operations - IPv4 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_star_g_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local grp_key="grp ff0e::1"
+	local vtep_ip=2001:db8:1000::1
+
+	echo
+	echo "Control path: Basic (*, G) operations - IPv6 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_sg_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local grp_key="grp 239.1.1.1 src 192.0.2.129"
+	local vtep_ip=198.51.100.100
+
+	echo
+	echo "Control path: Basic (S, G) operations - IPv4 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_sg_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local grp_key="grp ff0e::1 src 2001:db8:100::1"
+	local vtep_ip=198.51.100.100
+
+	echo
+	echo "Control path: Basic (S, G) operations - IPv6 overlay / IPv4 underlay"
+	echo "---------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_sg_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local grp_key="grp 239.1.1.1 src 192.0.2.129"
+	local vtep_ip=2001:db8:1000::1
+
+	echo
+	echo "Control path: Basic (S, G) operations - IPv4 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+basic_sg_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local grp_key="grp ff0e::1 src 2001:db8:100::1"
+	local vtep_ip=2001:db8:1000::1
+
+	echo
+	echo "Control path: Basic (S, G) operations - IPv6 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------------"
+
+	basic_common $ns1 "$grp_key" $vtep_ip
+}
+
+star_g_common()
+{
+	local ns1=$1; shift
+	local grp=$1; shift
+	local src1=$1; shift
+	local src2=$1; shift
+	local src3=$1; shift
+	local vtep_ip=$1; shift
+	local all_zeros_grp=$1; shift
+
+	# Test control path operations specific to (*, G) entries.
+
+	# Basic add, replace and delete behavior.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	log_test $? 0 "(*, G) MDB entry addition with source list"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \""
+	log_test $? 0 "(*, G) MDB entry presence after addition"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 0 "(S, G) MDB entry presence after addition"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	log_test $? 0 "(*, G) MDB entry replacement with source list"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \""
+	log_test $? 0 "(*, G) MDB entry presence after replacement"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 0 "(S, G) MDB entry presence after replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+	log_test $? 0 "(*, G) MDB entry deletion"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \""
+	log_test $? 1 "(*, G) MDB entry presence after deletion"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 1 "(S, G) MDB entry presence after deletion"
+
+	# Default filter mode and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep exclude"
+	log_test $? 0 "(*, G) MDB entry default filter mode"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode include source_list $src1 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep include"
+	log_test $? 0 "(*, G) MDB entry after replacing filter mode to \"include\""
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 0 "(S, G) MDB entry after replacing filter mode to \"include\""
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\" | grep blocked"
+	log_test $? 1 "\"blocked\" flag after replacing filter mode to \"include\""
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep exclude"
+	log_test $? 0 "(*, G) MDB entry after replacing filter mode to \"exclude\""
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 0 "(S, G) MDB entry after replacing filter mode to \"exclude\""
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\" | grep blocked"
+	log_test $? 0 "\"blocked\" flag after replacing filter mode to \"exclude\""
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+
+	# Default source list and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep source_list"
+	log_test $? 1 "(*, G) MDB entry default source list"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1,$src2,$src3 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 0 "(S, G) MDB entry of 1st source after replacing source list"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src2\""
+	log_test $? 0 "(S, G) MDB entry of 2nd source after replacing source list"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src3\""
+	log_test $? 0 "(S, G) MDB entry of 3rd source after replacing source list"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1,$src3 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src1\""
+	log_test $? 0 "(S, G) MDB entry of 1st source after removing source"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src2\""
+	log_test $? 1 "(S, G) MDB entry of 2nd source after removing source"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \"src $src3\""
+	log_test $? 0 "(S, G) MDB entry of 3rd source after removing source"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+
+	# Default protocol and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \"proto static\""
+	log_test $? 0 "(*, G) MDB entry default protocol"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \"proto static\""
+	log_test $? 0 "(S, G) MDB entry default protocol"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 proto bgp dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \"proto bgp\""
+	log_test $? 0 "(*, G) MDB entry protocol after replacement"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \"proto bgp\""
+	log_test $? 0 "(S, G) MDB entry protocol after replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+
+	# Default destination port and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \" dst_port \""
+	log_test $? 1 "(*, G) MDB entry default destination port"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \" dst_port \""
+	log_test $? 1 "(S, G) MDB entry default destination port"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip dst_port 1234 src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \" dst_port 1234 \""
+	log_test $? 0 "(*, G) MDB entry destination port after replacement"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \" dst_port 1234 \""
+	log_test $? 0 "(S, G) MDB entry destination port after replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+
+	# Default destination VNI and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \" vni \""
+	log_test $? 1 "(*, G) MDB entry default destination VNI"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \" vni \""
+	log_test $? 1 "(S, G) MDB entry default destination VNI"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip vni 1234 src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \" vni 1234 \""
+	log_test $? 0 "(*, G) MDB entry destination VNI after replacement"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \" vni 1234 \""
+	log_test $? 0 "(S, G) MDB entry destination VNI after replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+
+	# Default outgoing interface and replacement.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \" via \""
+	log_test $? 1 "(*, G) MDB entry default outgoing interface"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \" via \""
+	log_test $? 1 "(S, G) MDB entry default outgoing interface"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $src1 dst $vtep_ip src_vni 10010 via veth0"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep -v \" src \" | grep \" via veth0 \""
+	log_test $? 0 "(*, G) MDB entry outgoing interface after replacement"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep \" src \" | grep \" via veth0 \""
+	log_test $? 0 "(S, G) MDB entry outgoing interface after replacement"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep_ip src_vni 10010"
+
+	# Error cases.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $all_zeros_grp permanent filter_mode exclude dst $vtep_ip src_vni 10010"
+	log_test $? 255 "All-zeros group with filter mode"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $all_zeros_grp permanent source_list $src1 dst $vtep_ip src_vni 10010"
+	log_test $? 255 "All-zeros group with source list"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode include dst $vtep_ip src_vni 10010"
+	log_test $? 255 "(*, G) INCLUDE with an empty source list"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $grp dst $vtep_ip src_vni 10010"
+	log_test $? 255 "Invalid source in source list"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp permanent source_list $src1 dst $vtep_ip src_vni 10010"
+	log_test $? 255 "Source list without filter mode"
+}
+
+star_g_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local grp=239.1.1.1
+	local src1=192.0.2.129
+	local src2=192.0.2.130
+	local src3=192.0.2.131
+	local vtep_ip=198.51.100.100
+	local all_zeros_grp=0.0.0.0
+
+	echo
+	echo "Control path: (*, G) operations - IPv4 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------"
+
+	star_g_common $ns1 $grp $src1 $src2 $src3 $vtep_ip $all_zeros_grp
+}
+
+star_g_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local grp=ff0e::1
+	local src1=2001:db8:100::1
+	local src2=2001:db8:100::2
+	local src3=2001:db8:100::3
+	local vtep_ip=198.51.100.100
+	local all_zeros_grp=::
+
+	echo
+	echo "Control path: (*, G) operations - IPv6 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------"
+
+	star_g_common $ns1 $grp $src1 $src2 $src3 $vtep_ip $all_zeros_grp
+}
+
+star_g_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local grp=239.1.1.1
+	local src1=192.0.2.129
+	local src2=192.0.2.130
+	local src3=192.0.2.131
+	local vtep_ip=2001:db8:1000::1
+	local all_zeros_grp=0.0.0.0
+
+	echo
+	echo "Control path: (*, G) operations - IPv4 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------"
+
+	star_g_common $ns1 $grp $src1 $src2 $src3 $vtep_ip $all_zeros_grp
+}
+
+star_g_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local grp=ff0e::1
+	local src1=2001:db8:100::1
+	local src2=2001:db8:100::2
+	local src3=2001:db8:100::3
+	local vtep_ip=2001:db8:1000::1
+	local all_zeros_grp=::
+
+	echo
+	echo "Control path: (*, G) operations - IPv6 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------"
+
+	star_g_common $ns1 $grp $src1 $src2 $src3 $vtep_ip $all_zeros_grp
+}
+
+sg_common()
+{
+	local ns1=$1; shift
+	local grp=$1; shift
+	local src=$1; shift
+	local vtep_ip=$1; shift
+	local all_zeros_grp=$1; shift
+
+	# Test control path operations specific to (S, G) entries.
+
+	# Default filter mode.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp src $src permanent dst $vtep_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 -d -s mdb show dev vx0 | grep $grp | grep include"
+	log_test $? 0 "(S, G) MDB entry default filter mode"
+
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp src $src permanent dst $vtep_ip src_vni 10010"
+
+	# Error cases.
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp src $src permanent filter_mode include dst $vtep_ip src_vni 10010"
+	log_test $? 255 "(S, G) with filter mode"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp src $src permanent source_list $src dst $vtep_ip src_vni 10010"
+	log_test $? 255 "(S, G) with source list"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp src $grp permanent dst $vtep_ip src_vni 10010"
+	log_test $? 255 "(S, G) with an invalid source list"
+
+	run_cmd "bridge -n $ns1 mdb add dev vx0 port vx0 grp $all_zeros_grp src $src permanent dst $vtep_ip src_vni 10010"
+	log_test $? 255 "All-zeros group with source"
+}
+
+sg_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local grp=239.1.1.1
+	local src=192.0.2.129
+	local vtep_ip=198.51.100.100
+	local all_zeros_grp=0.0.0.0
+
+	echo
+	echo "Control path: (S, G) operations - IPv4 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------"
+
+	sg_common $ns1 $grp $src $vtep_ip $all_zeros_grp
+}
+
+sg_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+	local vtep_ip=198.51.100.100
+	local all_zeros_grp=::
+
+	echo
+	echo "Control path: (S, G) operations - IPv6 overlay / IPv4 underlay"
+	echo "--------------------------------------------------------------"
+
+	sg_common $ns1 $grp $src $vtep_ip $all_zeros_grp
+}
+
+sg_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local grp=239.1.1.1
+	local src=192.0.2.129
+	local vtep_ip=2001:db8:1000::1
+	local all_zeros_grp=0.0.0.0
+
+	echo
+	echo "Control path: (S, G) operations - IPv4 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------"
+
+	sg_common $ns1 $grp $src $vtep_ip $all_zeros_grp
+}
+
+sg_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+	local vtep_ip=2001:db8:1000::1
+	local all_zeros_grp=::
+
+	echo
+	echo "Control path: (S, G) operations - IPv6 overlay / IPv6 underlay"
+	echo "--------------------------------------------------------------"
+
+	sg_common $ns1 $grp $src $vtep_ip $all_zeros_grp
+}
+
+ipv4_grps_get()
+{
+	local max_grps=$1; shift
+	local i
+
+	for i in $(seq 0 $((max_grps - 1))); do
+		echo "239.1.1.$i"
+	done
+}
+
+ipv6_grps_get()
+{
+	local max_grps=$1; shift
+	local i
+
+	for i in $(seq 0 $((max_grps - 1))); do
+		echo "ff0e::$(printf %x $i)"
+	done
+}
+
+dump_common()
+{
+	local ns1=$1; shift
+	local local_addr=$1; shift
+	local remote_prefix=$1; shift
+	local fn=$1; shift
+	local max_vxlan_devs=2
+	local max_remotes=64
+	local max_grps=256
+	local num_entries
+	local batch_file
+	local grp
+	local i j
+
+	# The kernel maintains various markers for the MDB dump. Add a test for
+	# large scale MDB dump to make sure that all the configured entries are
+	# dumped and that the markers are used correctly.
+
+	# Create net devices.
+	for i in $(seq 1 $max_vxlan_devs); do
+		ip -n $ns1 link add name vx-test${i} up type vxlan \
+			local $local_addr dstport 4789 external vnifilter
+	done
+
+	# Create batch file with MDB entries.
+	batch_file=$(mktemp)
+	for i in $(seq 1 $max_vxlan_devs); do
+		for j in $(seq 1 $max_remotes); do
+			for grp in $($fn $max_grps); do
+				echo "mdb add dev vx-test${i} port vx-test${i} grp $grp permanent dst ${remote_prefix}${j}" >> $batch_file
+			done
+		done
+	done
+
+	# Program the batch file and check for expected number of entries.
+	bridge -n $ns1 -b $batch_file
+	for i in $(seq 1 $max_vxlan_devs); do
+		num_entries=$(bridge -n $ns1 mdb show dev vx-test${i} | grep "permanent" | wc -l)
+		[[ $num_entries -eq $((max_grps * max_remotes)) ]]
+		log_test $? 0 "Large scale dump - VXLAN device #$i"
+	done
+
+	rm -rf $batch_file
+}
+
+dump_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local local_addr=192.0.2.1
+	local remote_prefix=198.51.100.
+	local fn=ipv4_grps_get
+
+	echo
+	echo "Control path: Large scale MDB dump - IPv4 overlay / IPv4 underlay"
+	echo "-----------------------------------------------------------------"
+
+	dump_common $ns1 $local_addr $remote_prefix $fn
+}
+
+dump_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local local_addr=192.0.2.1
+	local remote_prefix=198.51.100.
+	local fn=ipv6_grps_get
+
+	echo
+	echo "Control path: Large scale MDB dump - IPv6 overlay / IPv4 underlay"
+	echo "-----------------------------------------------------------------"
+
+	dump_common $ns1 $local_addr $remote_prefix $fn
+}
+
+dump_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local local_addr=2001:db8:1::1
+	local remote_prefix=2001:db8:1000::
+	local fn=ipv4_grps_get
+
+	echo
+	echo "Control path: Large scale MDB dump - IPv4 overlay / IPv6 underlay"
+	echo "-----------------------------------------------------------------"
+
+	dump_common $ns1 $local_addr $remote_prefix $fn
+}
+
+dump_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local local_addr=2001:db8:1::1
+	local remote_prefix=2001:db8:1000::
+	local fn=ipv6_grps_get
+
+	echo
+	echo "Control path: Large scale MDB dump - IPv6 overlay / IPv6 underlay"
+	echo "-----------------------------------------------------------------"
+
+	dump_common $ns1 $local_addr $remote_prefix $fn
+}
+
+################################################################################
+# Tests - Data path
+
+encap_params_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local plen=$1; shift
+	local enc_ethtype=$1; shift
+	local grp=$1; shift
+	local src=$1; shift
+	local mz=$1; shift
+
+	# Test that packets forwarded by the VXLAN MDB are encapsulated with
+	# the correct parameters. Transmit packets from the first namespace and
+	# check that they hit the corresponding filters on the ingress of the
+	# second namespace.
+
+	run_cmd "tc -n $ns2 qdisc replace dev veth0 clsact"
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "ip -n $ns2 address replace $vtep1_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep2_ip/$plen dev lo"
+
+	# Check destination IP.
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep2_ip src_vni 10020"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $vtep1_ip action pass"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Destination IP - match"
+
+	run_cmd "ip netns exec $ns1 $mz br0.20 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Destination IP - no match"
+
+	run_cmd "tc -n $ns2 filter del dev vx0 ingress pref 1 handle 101 flower"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep2_ip src_vni 10020"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep1_ip src_vni 10010"
+
+	# Check destination port.
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip dst_port 1111 src_vni 10020"
+
+	run_cmd "tc -n $ns2 filter replace dev veth0 ingress pref 1 handle 101 proto $enc_ethtype flower ip_proto udp dst_port 4789 action pass"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev veth0 ingress" 101 1
+	log_test $? 0 "Default destination port - match"
+
+	run_cmd "ip netns exec $ns1 $mz br0.20 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev veth0 ingress" 101 1
+	log_test $? 0 "Default destination port - no match"
+
+	run_cmd "tc -n $ns2 filter replace dev veth0 ingress pref 1 handle 101 proto $enc_ethtype flower ip_proto udp dst_port 1111 action pass"
+	run_cmd "ip netns exec $ns1 $mz br0.20 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev veth0 ingress" 101 1
+	log_test $? 0 "Non-default destination port - match"
+
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev veth0 ingress" 101 1
+	log_test $? 0 "Non-default destination port - no match"
+
+	run_cmd "tc -n $ns2 filter del dev veth0 ingress pref 1 handle 101 flower"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep1_ip src_vni 10020"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep1_ip src_vni 10010"
+
+	# Check default VNI.
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip src_vni 10020"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_key_id 10010 action pass"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Default destination VNI - match"
+
+	run_cmd "ip netns exec $ns1 $mz br0.20 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Default destination VNI - no match"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip vni 10020 src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip vni 10010 src_vni 10020"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_key_id 10020 action pass"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Non-default destination VNI - match"
+
+	run_cmd "ip netns exec $ns1 $mz br0.20 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Non-default destination VNI - no match"
+
+	run_cmd "tc -n $ns2 filter del dev vx0 ingress pref 1 handle 101 flower"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep1_ip src_vni 10020"
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep1_ip src_vni 10010"
+}
+
+encap_params_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local enc_ethtype="ip"
+	local grp=239.1.1.1
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: Encapsulation parameters - IPv4 overlay / IPv4 underlay"
+	echo "------------------------------------------------------------------"
+
+	encap_params_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $enc_ethtype \
+		$grp $src "mausezahn"
+}
+
+encap_params_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local enc_ethtype="ip"
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: Encapsulation parameters - IPv6 overlay / IPv4 underlay"
+	echo "------------------------------------------------------------------"
+
+	encap_params_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $enc_ethtype \
+		$grp $src "mausezahn -6"
+}
+
+encap_params_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local enc_ethtype="ipv6"
+	local grp=239.1.1.1
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: Encapsulation parameters - IPv4 overlay / IPv6 underlay"
+	echo "------------------------------------------------------------------"
+
+	encap_params_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $enc_ethtype \
+		$grp $src "mausezahn"
+}
+
+encap_params_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local enc_ethtype="ipv6"
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: Encapsulation parameters - IPv6 overlay / IPv6 underlay"
+	echo "------------------------------------------------------------------"
+
+	encap_params_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $enc_ethtype \
+		$grp $src "mausezahn -6"
+}
+
+starg_exclude_ir_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local plen=$1; shift
+	local grp=$1; shift
+	local valid_src=$1; shift
+	local invalid_src=$1; shift
+	local mz=$1; shift
+
+	# Install a (*, G) EXCLUDE MDB entry with one source and two remote
+	# VTEPs. Make sure that the source in the source list is not forwarded
+	# and that a source not in the list is forwarded. Remove one of the
+	# VTEPs from the entry and make sure that packets are only forwarded to
+	# the remaining VTEP.
+
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "ip -n $ns2 address replace $vtep1_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep2_ip/$plen dev lo"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $vtep1_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 102 proto all flower enc_dst_ip $vtep2_ip action pass"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $invalid_src dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $invalid_src dst $vtep2_ip src_vni 10010"
+
+	# Check that invalid source is not forwarded to any VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $invalid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 0
+	log_test $? 0 "Block excluded source - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 0
+	log_test $? 0 "Block excluded source - second VTEP"
+
+	# Check that valid source is forwarded to both VTEPs.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Forward valid source - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Forward valid source - second VTEP"
+
+	# Remove second VTEP.
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep2_ip src_vni 10010"
+
+	# Check that invalid source is not forwarded to any VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $invalid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Block excluded source after removal - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Block excluded source after removal - second VTEP"
+
+	# Check that valid source is forwarded to the remaining VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 2
+	log_test $? 0 "Forward valid source after removal - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Forward valid source after removal - second VTEP"
+}
+
+starg_exclude_ir_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - IR - IPv4 overlay / IPv4 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_exclude_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_exclude_ir_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - IR - IPv6 overlay / IPv4 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_exclude_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_exclude_ir_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - IR - IPv4 overlay / IPv6 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_exclude_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_exclude_ir_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - IR - IPv6 overlay / IPv6 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_exclude_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_include_ir_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local plen=$1; shift
+	local grp=$1; shift
+	local valid_src=$1; shift
+	local invalid_src=$1; shift
+	local mz=$1; shift
+
+	# Install a (*, G) INCLUDE MDB entry with one source and two remote
+	# VTEPs. Make sure that the source in the source list is forwarded and
+	# that a source not in the list is not forwarded. Remove one of the
+	# VTEPs from the entry and make sure that packets are only forwarded to
+	# the remaining VTEP.
+
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "ip -n $ns2 address replace $vtep1_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep2_ip/$plen dev lo"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $vtep1_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 102 proto all flower enc_dst_ip $vtep2_ip action pass"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode include source_list $valid_src dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode include source_list $valid_src dst $vtep2_ip src_vni 10010"
+
+	# Check that invalid source is not forwarded to any VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $invalid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 0
+	log_test $? 0 "Block excluded source - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 0
+	log_test $? 0 "Block excluded source - second VTEP"
+
+	# Check that valid source is forwarded to both VTEPs.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Forward valid source - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Forward valid source - second VTEP"
+
+	# Remove second VTEP.
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep2_ip src_vni 10010"
+
+	# Check that invalid source is not forwarded to any VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $invalid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Block excluded source after removal - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Block excluded source after removal - second VTEP"
+
+	# Check that valid source is forwarded to the remaining VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 2
+	log_test $? 0 "Forward valid source after removal - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Forward valid source after removal - second VTEP"
+}
+
+starg_include_ir_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) INCLUDE - IR - IPv4 overlay / IPv4 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_include_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_include_ir_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) INCLUDE - IR - IPv6 overlay / IPv4 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_include_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_include_ir_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) INCLUDE - IR - IPv4 overlay / IPv6 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_include_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_include_ir_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) INCLUDE - IR - IPv6 overlay / IPv6 underlay"
+	echo "-------------------------------------------------------------"
+
+	starg_include_ir_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_exclude_p2mp_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local mcast_grp=$1; shift
+	local plen=$1; shift
+	local grp=$1; shift
+	local valid_src=$1; shift
+	local invalid_src=$1; shift
+	local mz=$1; shift
+
+	# Install a (*, G) EXCLUDE MDB entry with one source and one multicast
+	# group to which packets are sent. Make sure that the source in the
+	# source list is not forwarded and that a source not in the list is
+	# forwarded.
+
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "ip -n $ns2 address replace $mcast_grp/$plen dev veth0 autojoin"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $mcast_grp action pass"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode exclude source_list $invalid_src dst $mcast_grp src_vni 10010 via veth0"
+
+	# Check that invalid source is not forwarded.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $invalid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 0
+	log_test $? 0 "Block excluded source"
+
+	# Check that valid source is forwarded.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Forward valid source"
+
+	# Remove the VTEP from the multicast group.
+	run_cmd "ip -n $ns2 address del $mcast_grp/$plen dev veth0"
+
+	# Check that valid source is not received anymore.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Receive of valid source after removal from group"
+}
+
+starg_exclude_p2mp_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local mcast_grp=238.1.1.1
+	local plen=32
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - P2MP - IPv4 overlay / IPv4 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_exclude_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_exclude_p2mp_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local mcast_grp=238.1.1.1
+	local plen=32
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - P2MP - IPv6 overlay / IPv4 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_exclude_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_exclude_p2mp_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local mcast_grp=ff0e::2
+	local plen=128
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - P2MP - IPv4 overlay / IPv6 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_exclude_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_exclude_p2mp_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local mcast_grp=ff0e::2
+	local plen=128
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) EXCLUDE - P2MP - IPv6 overlay / IPv6 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_exclude_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_include_p2mp_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local mcast_grp=$1; shift
+	local plen=$1; shift
+	local grp=$1; shift
+	local valid_src=$1; shift
+	local invalid_src=$1; shift
+	local mz=$1; shift
+
+	# Install a (*, G) INCLUDE MDB entry with one source and one multicast
+	# group to which packets are sent. Make sure that the source in the
+	# source list is forwarded and that a source not in the list is not
+	# forwarded.
+
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "ip -n $ns2 address replace $mcast_grp/$plen dev veth0 autojoin"
+
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $mcast_grp action pass"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent filter_mode include source_list $valid_src dst $mcast_grp src_vni 10010 via veth0"
+
+	# Check that invalid source is not forwarded.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $invalid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 0
+	log_test $? 0 "Block excluded source"
+
+	# Check that valid source is forwarded.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Forward valid source"
+
+	# Remove the VTEP from the multicast group.
+	run_cmd "ip -n $ns2 address del $mcast_grp/$plen dev veth0"
+
+	# Check that valid source is not received anymore.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $valid_src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Receive of valid source after removal from group"
+}
+
+starg_include_p2mp_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local mcast_grp=238.1.1.1
+	local plen=32
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) INCLUDE - P2MP - IPv4 overlay / IPv4 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_include_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_include_p2mp_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local mcast_grp=238.1.1.1
+	local plen=32
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) INCLUDE - P2MP - IPv6 overlay / IPv4 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_include_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+starg_include_p2mp_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local mcast_grp=ff0e::2
+	local plen=128
+	local grp=239.1.1.1
+	local valid_src=192.0.2.129
+	local invalid_src=192.0.2.145
+
+	echo
+	echo "Data path: (*, G) INCLUDE - P2MP - IPv4 overlay / IPv6 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_include_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn"
+}
+
+starg_include_p2mp_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local mcast_grp=ff0e::2
+	local plen=128
+	local grp=ff0e::1
+	local valid_src=2001:db8:100::1
+	local invalid_src=2001:db8:200::1
+
+	echo
+	echo "Data path: (*, G) INCLUDE - P2MP - IPv6 overlay / IPv6 underlay"
+	echo "---------------------------------------------------------------"
+
+	starg_include_p2mp_common $ns1 $ns2 $mcast_grp $plen $grp \
+		$valid_src $invalid_src "mausezahn -6"
+}
+
+egress_vni_translation_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local mcast_grp=$1; shift
+	local plen=$1; shift
+	local proto=$1; shift
+	local grp=$1; shift
+	local src=$1; shift
+	local mz=$1; shift
+
+	# When P2MP tunnels are used with optimized inter-subnet multicast
+	# (OISM) [1], the ingress VTEP does not perform VNI translation and
+	# uses the VNI of the source broadcast domain (BD). If the egress VTEP
+	# is a member in the source BD, then no VNI translation is needed.
+	# Otherwise, the egress VTEP needs to translate the VNI to the
+	# supplementary broadcast domain (SBD) VNI, which is usually the L3VNI.
+	#
+	# In this test, remove the VTEP in the second namespace from VLAN 10
+	# (VNI 10010) and make sure that a packet sent from this VLAN on the
+	# first VTEP is received by the SVI corresponding to the L3VNI (14000 /
+	# VLAN 4000) on the second VTEP.
+	#
+	# The second VTEP will be able to decapsulate the packet with VNI 10010
+	# because this VNI is configured on its shared VXLAN device. Later,
+	# when ingressing the bridge, the VNI to VLAN lookup will fail because
+	# the VTEP is not a member in VLAN 10, which will cause the packet to
+	# be tagged with VLAN 4000 since it is configured as PVID.
+	#
+	# [1] https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-irb-mcast
+
+	run_cmd "tc -n $ns2 qdisc replace dev br0.4000 clsact"
+	run_cmd "ip -n $ns2 address replace $mcast_grp/$plen dev veth0 autojoin"
+	run_cmd "tc -n $ns2 filter replace dev br0.4000 ingress pref 1 handle 101 proto $proto flower src_ip $src dst_ip $grp action pass"
+
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp src $src permanent dst $mcast_grp src_vni 10010 via veth0"
+
+	# Remove the second VTEP from VLAN 10.
+	run_cmd "bridge -n $ns2 vlan del vid 10 dev vx0"
+
+	# Make sure that packets sent from the first VTEP over VLAN 10 are
+	# received by the SVI corresponding to the L3VNI (14000 / VLAN 4000) on
+	# the second VTEP, since it is configured as PVID.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev br0.4000 ingress" 101 1
+	log_test $? 0 "Egress VNI translation - PVID configured"
+
+	# Remove PVID flag from VLAN 4000 on the second VTEP and make sure
+	# packets are no longer received by the SVI interface.
+	run_cmd "bridge -n $ns2 vlan add vid 4000 dev vx0"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev br0.4000 ingress" 101 1
+	log_test $? 0 "Egress VNI translation - no PVID configured"
+
+	# Reconfigure the PVID and make sure packets are received again.
+	run_cmd "bridge -n $ns2 vlan add vid 4000 dev vx0 pvid"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev br0.4000 ingress" 101 2
+	log_test $? 0 "Egress VNI translation - PVID reconfigured"
+}
+
+egress_vni_translation_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local mcast_grp=238.1.1.1
+	local plen=32
+	local proto="ipv4"
+	local grp=239.1.1.1
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: Egress VNI translation - IPv4 overlay / IPv4 underlay"
+	echo "----------------------------------------------------------------"
+
+	egress_vni_translation_common $ns1 $ns2 $mcast_grp $plen $proto $grp \
+		$src "mausezahn"
+}
+
+egress_vni_translation_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local mcast_grp=238.1.1.1
+	local plen=32
+	local proto="ipv6"
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: Egress VNI translation - IPv6 overlay / IPv4 underlay"
+	echo "----------------------------------------------------------------"
+
+	egress_vni_translation_common $ns1 $ns2 $mcast_grp $plen $proto $grp \
+		$src "mausezahn -6"
+}
+
+egress_vni_translation_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local mcast_grp=ff0e::2
+	local plen=128
+	local proto="ipv4"
+	local grp=239.1.1.1
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: Egress VNI translation - IPv4 overlay / IPv6 underlay"
+	echo "----------------------------------------------------------------"
+
+	egress_vni_translation_common $ns1 $ns2 $mcast_grp $plen $proto $grp \
+		$src "mausezahn"
+}
+
+egress_vni_translation_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local mcast_grp=ff0e::2
+	local plen=128
+	local proto="ipv6"
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: Egress VNI translation - IPv6 overlay / IPv6 underlay"
+	echo "----------------------------------------------------------------"
+
+	egress_vni_translation_common $ns1 $ns2 $mcast_grp $plen $proto $grp \
+		$src "mausezahn -6"
+}
+
+all_zeros_mdb_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local vtep3_ip=$1; shift
+	local vtep4_ip=$1; shift
+	local plen=$1; shift
+	local ipv4_grp=239.1.1.1
+	local ipv4_unreg_grp=239.2.2.2
+	local ipv4_ll_grp=224.0.0.100
+	local ipv4_src=192.0.2.129
+	local ipv6_grp=ff0e::1
+	local ipv6_unreg_grp=ff0e::2
+	local ipv6_ll_grp=ff02::1
+	local ipv6_src=2001:db8:100::1
+
+	# Install all-zeros (catchall) MDB entries for IPv4 and IPv6 traffic
+	# and make sure they only forward unregistered IP multicast traffic
+	# which is not link-local. Also make sure that each entry only forwards
+	# traffic from the matching address family.
+
+	# Associate two different VTEPs with one all-zeros MDB entry: Two with
+	# the IPv4 entry (0.0.0.0) and another two with the IPv6 one (::).
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp 0.0.0.0 permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp 0.0.0.0 permanent dst $vtep2_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp :: permanent dst $vtep3_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp :: permanent dst $vtep4_ip src_vni 10010"
+
+	# Associate one VTEP from each set with a regular MDB entry: One with
+	# an IPv4 entry and another with an IPv6 one.
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $ipv4_grp permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $ipv6_grp permanent dst $vtep3_ip src_vni 10010"
+
+	# Add filters to match on decapsulated traffic in the second namespace.
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $vtep1_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 102 proto all flower enc_dst_ip $vtep2_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 103 proto all flower enc_dst_ip $vtep3_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 104 proto all flower enc_dst_ip $vtep4_ip action pass"
+
+	# Configure the VTEP addresses in the second namespace to enable
+	# decapsulation.
+	run_cmd "ip -n $ns2 address replace $vtep1_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep2_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep3_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep4_ip/$plen dev lo"
+
+	# Send registered IPv4 multicast and make sure it only arrives to the
+	# first VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn br0.10 -A $ipv4_src -B $ipv4_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Registered IPv4 multicast - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 0
+	log_test $? 0 "Registered IPv4 multicast - second VTEP"
+
+	# Send unregistered IPv4 multicast that is not link-local and make sure
+	# it arrives to the first and second VTEPs.
+	run_cmd "ip netns exec $ns1 mausezahn br0.10 -A $ipv4_src -B $ipv4_unreg_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 2
+	log_test $? 0 "Unregistered IPv4 multicast - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Unregistered IPv4 multicast - second VTEP"
+
+	# Send IPv4 link-local multicast traffic and make sure it does not
+	# arrive to any VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn br0.10 -A $ipv4_src -B $ipv4_ll_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 2
+	log_test $? 0 "Link-local IPv4 multicast - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Link-local IPv4 multicast - second VTEP"
+
+	# Send registered IPv4 multicast using a unicast MAC address and make
+	# sure it does not arrive to any VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn br0.10 -a own -b 00:11:22:33:44:55 -A $ipv4_src -B $ipv4_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 2
+	log_test $? 0 "Registered IPv4 multicast with a unicast MAC - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Registered IPv4 multicast with a unicast MAC - second VTEP"
+
+	# Send registered IPv4 multicast using a broadcast MAC address and make
+	# sure it does not arrive to any VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn br0.10 -a own -b bcast -A $ipv4_src -B $ipv4_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 2
+	log_test $? 0 "Registered IPv4 multicast with a broadcast MAC - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Registered IPv4 multicast with a broadcast MAC - second VTEP"
+
+	# Make sure IPv4 traffic did not reach the VTEPs associated with
+	# IPv6 entries.
+	tc_check_packets "$ns2" "dev vx0 ingress" 103 0
+	log_test $? 0 "IPv4 traffic - third VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 104 0
+	log_test $? 0 "IPv4 traffic - fourth VTEP"
+
+	# Reset IPv4 filters before testing IPv6 traffic.
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto all flower enc_dst_ip $vtep1_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 102 proto all flower enc_dst_ip $vtep2_ip action pass"
+
+	# Send registered IPv6 multicast and make sure it only arrives to the
+	# third VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn -6 br0.10 -A $ipv6_src -B $ipv6_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 103 1
+	log_test $? 0 "Registered IPv6 multicast - third VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 104 0
+	log_test $? 0 "Registered IPv6 multicast - fourth VTEP"
+
+	# Send unregistered IPv6 multicast that is not link-local and make sure
+	# it arrives to the third and fourth VTEPs.
+	run_cmd "ip netns exec $ns1 mausezahn -6 br0.10 -A $ipv6_src -B $ipv6_unreg_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 103 2
+	log_test $? 0 "Unregistered IPv6 multicast - third VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 104 1
+	log_test $? 0 "Unregistered IPv6 multicast - fourth VTEP"
+
+	# Send IPv6 link-local multicast traffic and make sure it does not
+	# arrive to any VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn -6 br0.10 -A $ipv6_src -B $ipv6_ll_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 103 2
+	log_test $? 0 "Link-local IPv6 multicast - third VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 104 1
+	log_test $? 0 "Link-local IPv6 multicast - fourth VTEP"
+
+	# Send registered IPv6 multicast using a unicast MAC address and make
+	# sure it does not arrive to any VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn -6 br0.10 -a own -b 00:11:22:33:44:55 -A $ipv6_src -B $ipv6_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 103 2
+	log_test $? 0 "Registered IPv6 multicast with a unicast MAC - third VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 104 1
+	log_test $? 0 "Registered IPv6 multicast with a unicast MAC - fourth VTEP"
+
+	# Send registered IPv6 multicast using a broadcast MAC address and make
+	# sure it does not arrive to any VTEP.
+	run_cmd "ip netns exec $ns1 mausezahn -6 br0.10 -a own -b bcast -A $ipv6_src -B $ipv6_grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 103 2
+	log_test $? 0 "Registered IPv6 multicast with a broadcast MAC - third VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 104 1
+	log_test $? 0 "Registered IPv6 multicast with a broadcast MAC - fourth VTEP"
+
+	# Make sure IPv6 traffic did not reach the VTEPs associated with
+	# IPv4 entries.
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 0
+	log_test $? 0 "IPv6 traffic - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 0
+	log_test $? 0 "IPv6 traffic - second VTEP"
+}
+
+all_zeros_mdb_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.101
+	local vtep2_ip=198.51.100.102
+	local vtep3_ip=198.51.100.103
+	local vtep4_ip=198.51.100.104
+	local plen=32
+
+	echo
+	echo "Data path: All-zeros MDB entry - IPv4 underlay"
+	echo "----------------------------------------------"
+
+	all_zeros_mdb_common $ns1 $ns2 $vtep1_ip $vtep2_ip $vtep3_ip \
+		$vtep4_ip $plen
+}
+
+all_zeros_mdb_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local vtep3_ip=2001:db8:3000::1
+	local vtep4_ip=2001:db8:4000::1
+	local plen=128
+
+	echo
+	echo "Data path: All-zeros MDB entry - IPv6 underlay"
+	echo "----------------------------------------------"
+
+	all_zeros_mdb_common $ns1 $ns2 $vtep1_ip $vtep2_ip $vtep3_ip \
+		$vtep4_ip $plen
+}
+
+mdb_fdb_common()
+{
+	local ns1=$1; shift
+	local ns2=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local plen=$1; shift
+	local proto=$1; shift
+	local grp=$1; shift
+	local src=$1; shift
+	local mz=$1; shift
+
+	# Install an MDB entry and an FDB entry and make sure that the FDB
+	# entry only forwards traffic that was not forwarded by the MDB.
+
+	# Associate the MDB entry with one VTEP and the FDB entry with another
+	# VTEP.
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 fdb add 00:00:00:00:00:00 dev vx0 self static dst $vtep2_ip src_vni 10010"
+
+	# Add filters to match on decapsulated traffic in the second namespace.
+	run_cmd "tc -n $ns2 qdisc replace dev vx0 clsact"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 101 proto $proto flower ip_proto udp dst_port 54321 enc_dst_ip $vtep1_ip action pass"
+	run_cmd "tc -n $ns2 filter replace dev vx0 ingress pref 1 handle 102 proto $proto flower ip_proto udp dst_port 54321 enc_dst_ip $vtep2_ip action pass"
+
+	# Configure the VTEP addresses in the second namespace to enable
+	# decapsulation.
+	run_cmd "ip -n $ns2 address replace $vtep1_ip/$plen dev lo"
+	run_cmd "ip -n $ns2 address replace $vtep2_ip/$plen dev lo"
+
+	# Send IP multicast traffic and make sure it is forwarded by the MDB
+	# and only arrives to the first VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "IP multicast - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 0
+	log_test $? 0 "IP multicast - second VTEP"
+
+	# Send broadcast traffic and make sure it is forwarded by the FDB and
+	# only arrives to the second VTEP.
+	run_cmd "ip netns exec $ns1 $mz br0.10 -a own -b bcast -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "Broadcast - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 1
+	log_test $? 0 "Broadcast - second VTEP"
+
+	# Remove the MDB entry and make sure that IP multicast is now forwarded
+	# by the FDB to the second VTEP.
+	run_cmd "bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp dst $vtep1_ip src_vni 10010"
+	run_cmd "ip netns exec $ns1 $mz br0.10 -A $src -B $grp -t udp sp=12345,dp=54321 -p 100 -c 1 -q"
+	tc_check_packets "$ns2" "dev vx0 ingress" 101 1
+	log_test $? 0 "IP multicast after removal - first VTEP"
+	tc_check_packets "$ns2" "dev vx0 ingress" 102 2
+	log_test $? 0 "IP multicast after removal - second VTEP"
+}
+
+mdb_fdb_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local proto="ipv4"
+	local grp=239.1.1.1
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: MDB with FDB - IPv4 overlay / IPv4 underlay"
+	echo "------------------------------------------------------"
+
+	mdb_fdb_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $proto $grp $src \
+		"mausezahn"
+}
+
+mdb_fdb_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local ns2=ns2_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local plen=32
+	local proto="ipv6"
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: MDB with FDB - IPv6 overlay / IPv4 underlay"
+	echo "------------------------------------------------------"
+
+	mdb_fdb_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $proto $grp $src \
+		"mausezahn -6"
+}
+
+mdb_fdb_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local proto="ipv4"
+	local grp=239.1.1.1
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: MDB with FDB - IPv4 overlay / IPv6 underlay"
+	echo "------------------------------------------------------"
+
+	mdb_fdb_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $proto $grp $src \
+		"mausezahn"
+}
+
+mdb_fdb_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local ns2=ns2_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local plen=128
+	local proto="ipv6"
+	local grp=ff0e::1
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: MDB with FDB - IPv6 overlay / IPv6 underlay"
+	echo "------------------------------------------------------"
+
+	mdb_fdb_common $ns1 $ns2 $vtep1_ip $vtep2_ip $plen $proto $grp $src \
+		"mausezahn -6"
+}
+
+mdb_grp1_loop()
+{
+	local ns1=$1; shift
+	local vtep1_ip=$1; shift
+	local grp1=$1; shift
+
+	while true; do
+		bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp1 dst $vtep1_ip src_vni 10010
+		bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp1 permanent dst $vtep1_ip src_vni 10010
+	done >/dev/null 2>&1
+}
+
+mdb_grp2_loop()
+{
+	local ns1=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local grp2=$1; shift
+
+	while true; do
+		bridge -n $ns1 mdb del dev vx0 port vx0 grp $grp2 dst $vtep1_ip src_vni 10010
+		bridge -n $ns1 mdb add dev vx0 port vx0 grp $grp2 permanent dst $vtep1_ip src_vni 10010
+		bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp2 permanent dst $vtep2_ip src_vni 10010
+	done >/dev/null 2>&1
+}
+
+mdb_torture_common()
+{
+	local ns1=$1; shift
+	local vtep1_ip=$1; shift
+	local vtep2_ip=$1; shift
+	local grp1=$1; shift
+	local grp2=$1; shift
+	local src=$1; shift
+	local mz=$1; shift
+	local pid1
+	local pid2
+	local pid3
+	local pid4
+
+	# Continuously send two streams that are forwarded by two different MDB
+	# entries. The first entry will be added and deleted in a loop. This
+	# allows us to test that the data path does not use freed MDB entry
+	# memory. The second entry will have two remotes, one that is added and
+	# deleted in a loop and another that is replaced in a loop. This allows
+	# us to test that the data path does not use freed remote entry memory.
+	# The test is considered successful if nothing crashed.
+
+	# Create the MDB entries that will be continuously deleted / replaced.
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp1 permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp2 permanent dst $vtep1_ip src_vni 10010"
+	run_cmd "bridge -n $ns1 mdb replace dev vx0 port vx0 grp $grp2 permanent dst $vtep2_ip src_vni 10010"
+
+	mdb_grp1_loop $ns1 $vtep1_ip $grp1 &
+	pid1=$!
+	mdb_grp2_loop $ns1 $vtep1_ip $vtep2_ip $grp2 &
+	pid2=$!
+	ip netns exec $ns1 $mz br0.10 -A $src -B $grp1 -t udp sp=12345,dp=54321 -p 100 -c 0 -q &
+	pid3=$!
+	ip netns exec $ns1 $mz br0.10 -A $src -B $grp2 -t udp sp=12345,dp=54321 -p 100 -c 0 -q &
+	pid4=$!
+
+	sleep 30
+	kill -9 $pid1 $pid2 $pid3 $pid4
+	wait $pid1 $pid2 $pid3 $pid4 2>/dev/null
+
+	log_test 0 0 "Torture test"
+}
+
+mdb_torture_ipv4_ipv4()
+{
+	local ns1=ns1_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local grp1=239.1.1.1
+	local grp2=239.2.2.2
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: MDB torture test - IPv4 overlay / IPv4 underlay"
+	echo "----------------------------------------------------------"
+
+	mdb_torture_common $ns1 $vtep1_ip $vtep2_ip $grp1 $grp2 $src \
+		"mausezahn"
+}
+
+mdb_torture_ipv6_ipv4()
+{
+	local ns1=ns1_v4
+	local vtep1_ip=198.51.100.100
+	local vtep2_ip=198.51.100.200
+	local grp1=ff0e::1
+	local grp2=ff0e::2
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: MDB torture test - IPv6 overlay / IPv4 underlay"
+	echo "----------------------------------------------------------"
+
+	mdb_torture_common $ns1 $vtep1_ip $vtep2_ip $grp1 $grp2 $src \
+		"mausezahn -6"
+}
+
+mdb_torture_ipv4_ipv6()
+{
+	local ns1=ns1_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local grp1=239.1.1.1
+	local grp2=239.2.2.2
+	local src=192.0.2.129
+
+	echo
+	echo "Data path: MDB torture test - IPv4 overlay / IPv6 underlay"
+	echo "----------------------------------------------------------"
+
+	mdb_torture_common $ns1 $vtep1_ip $vtep2_ip $grp1 $grp2 $src \
+		"mausezahn"
+}
+
+mdb_torture_ipv6_ipv6()
+{
+	local ns1=ns1_v6
+	local vtep1_ip=2001:db8:1000::1
+	local vtep2_ip=2001:db8:2000::1
+	local grp1=ff0e::1
+	local grp2=ff0e::2
+	local src=2001:db8:100::1
+
+	echo
+	echo "Data path: MDB torture test - IPv6 overlay / IPv6 underlay"
+	echo "----------------------------------------------------------"
+
+	mdb_torture_common $ns1 $vtep1_ip $vtep2_ip $grp1 $grp2 $src \
+		"mausezahn -6"
+}
+
+################################################################################
+# Usage
+
+usage()
+{
+	cat <<EOF
+usage: ${0##*/} OPTS
+
+        -t <test>   Test(s) to run (default: all)
+                    (options: $TESTS)
+        -c          Control path tests only
+        -d          Data path tests only
+        -p          Pause on fail
+        -P          Pause after each test before cleanup
+        -v          Verbose mode (show commands and output)
+EOF
+}
+
+################################################################################
+# Main
+
+trap cleanup EXIT
+
+while getopts ":t:cdpPvh" opt; do
+	case $opt in
+		t) TESTS=$OPTARG;;
+		c) TESTS=${CONTROL_PATH_TESTS};;
+		d) TESTS=${DATA_PATH_TESTS};;
+		p) PAUSE_ON_FAIL=yes;;
+		P) PAUSE=yes;;
+		v) VERBOSE=$(($VERBOSE + 1));;
+		h) usage; exit 0;;
+		*) usage; exit 1;;
+	esac
+done
+
+# Make sure we don't pause twice.
+[ "${PAUSE}" = "yes" ] && PAUSE_ON_FAIL=no
+
+if [ "$(id -u)" -ne 0 ];then
+	echo "SKIP: Need root privileges"
+	exit $ksft_skip;
+fi
+
+if [ ! -x "$(command -v ip)" ]; then
+	echo "SKIP: Could not run test without ip tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v bridge)" ]; then
+	echo "SKIP: Could not run test without bridge tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v mausezahn)" ]; then
+	echo "SKIP: Could not run test without mausezahn tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v jq)" ]; then
+	echo "SKIP: Could not run test without jq tool"
+	exit $ksft_skip
+fi
+
+bridge mdb help 2>&1 | grep -q "src_vni"
+if [ $? -ne 0 ]; then
+   echo "SKIP: iproute2 bridge too old, missing VXLAN MDB support"
+   exit $ksft_skip
+fi
+
+# Start clean.
+cleanup
+
+for t in $TESTS
+do
+	setup; $t; cleanup;
+done
+
+if [ "$TESTS" != "none" ]; then
+	printf "\nTests passed: %3d\n" ${nsuccess}
+	printf "Tests failed: %3d\n"   ${nfail}
+fi
+
+exit $ret
-- 
cgit v1.2.3-70-g09d2


From ef63461caf427a77a04620d74ba90035a712af9c Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 15 Mar 2023 14:46:43 +0000
Subject: net: pcs: xpcs: remove double-read of link state when using AN

Phylink does not want the current state of the link when reading the
PCS link state - it wants the latched state. Don't double-read the
MII status register. Phylink will re-read as necessary to capture
transient link-down events as of dbae3388ea9c ("net: phylink: Force
retrigger in case of latched link-fail indicator").

The above referenced commit is a dependency for this change, and thus
this change should not be backported to any kernel that does not
contain the above referenced commit.

Fixes: fcb26bd2b6ca ("net: phy: Add Synopsys DesignWare XPCS MDIO module")
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index bc428a816719..04a685353041 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -321,7 +321,7 @@ static int xpcs_read_fault_c73(struct dw_xpcs *xpcs,
 	return 0;
 }
 
-static int xpcs_read_link_c73(struct dw_xpcs *xpcs, bool an)
+static int xpcs_read_link_c73(struct dw_xpcs *xpcs)
 {
 	bool link = true;
 	int ret;
@@ -333,15 +333,6 @@ static int xpcs_read_link_c73(struct dw_xpcs *xpcs, bool an)
 	if (!(ret & MDIO_STAT1_LSTATUS))
 		link = false;
 
-	if (an) {
-		ret = xpcs_read(xpcs, MDIO_MMD_AN, MDIO_STAT1);
-		if (ret < 0)
-			return ret;
-
-		if (!(ret & MDIO_STAT1_LSTATUS))
-			link = false;
-	}
-
 	return link;
 }
 
@@ -935,7 +926,7 @@ static int xpcs_get_state_c73(struct dw_xpcs *xpcs,
 	int ret;
 
 	/* Link needs to be read first ... */
-	state->link = xpcs_read_link_c73(xpcs, state->an_enabled) > 0 ? 1 : 0;
+	state->link = xpcs_read_link_c73(xpcs) > 0 ? 1 : 0;
 
 	/* ... and then we check the faults. */
 	ret = xpcs_read_fault_c73(xpcs, state);
-- 
cgit v1.2.3-70-g09d2


From ecec0ebbc6381a5a375f1cf10c4858f24e91e2ef Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 15 Mar 2023 14:46:49 +0000
Subject: net: pcs: lynx: don't print an_enabled in pcs_get_state()

an_enabled will be going away, and in any case, pcs_get_state() should
not be updating this member. Remove the print.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-lynx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/pcs/pcs-lynx.c b/drivers/net/pcs/pcs-lynx.c
index 3903f3baba2b..622c3de3f3a8 100644
--- a/drivers/net/pcs/pcs-lynx.c
+++ b/drivers/net/pcs/pcs-lynx.c
@@ -112,11 +112,11 @@ static void lynx_pcs_get_state(struct phylink_pcs *pcs,
 	}
 
 	dev_dbg(&lynx->mdio->dev,
-		"mode=%s/%s/%s link=%u an_enabled=%u an_complete=%u\n",
+		"mode=%s/%s/%s link=%u an_complete=%u\n",
 		phy_modes(state->interface),
 		phy_speed_to_str(state->speed),
 		phy_duplex_to_str(state->duplex),
-		state->link, state->an_enabled, state->an_complete);
+		state->link, state->an_complete);
 }
 
 static int lynx_pcs_config_giga(struct mdio_device *pcs, unsigned int mode,
-- 
cgit v1.2.3-70-g09d2


From 731b73dba359e3ff00517c13aa0daa82b34ff466 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Date: Wed, 15 Mar 2023 08:33:02 -0700
Subject: vlan: partially enable SIOCSHWTSTAMP in container

Setting timestamp filter was explicitly disabled on vlan devices in
containers because it might affect other processes on the host. But it's
absolutely legit in case when real device is in the same namespace.

Fixes: 873017af7784 ("vlan: disable SIOCSHWTSTAMP in container")
Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/8021q/vlan_dev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 296d0145932f..5920544e93e8 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -365,7 +365,7 @@ static int vlan_dev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 	switch (cmd) {
 	case SIOCSHWTSTAMP:
-		if (!net_eq(dev_net(dev), &init_net))
+		if (!net_eq(dev_net(dev), dev_net(real_dev)))
 			break;
 		fallthrough;
 	case SIOCGMIIPHY:
-- 
cgit v1.2.3-70-g09d2


From cee1af825d65b8122627fc2efbc36c1bd51ee103 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Mar 2023 20:57:41 +0000
Subject: tcp: annotate lockless accesses to sk->sk_err_soft

This field can be read/written without lock synchronization.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_input.c |  2 +-
 net/ipv4/tcp_ipv4.c  |  6 +++---
 net/ipv4/tcp_timer.c |  6 +++---
 net/ipv6/tcp_ipv6.c  | 11 ++++++-----
 4 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index cc072d2cfcd8..8b5b6ca6617d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3874,7 +3874,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	/* We passed data and got it acked, remove any soft error
 	 * log. Something worked...
 	 */
-	sk->sk_err_soft = 0;
+	WRITE_ONCE(sk->sk_err_soft, 0);
 	icsk->icsk_probes_out = 0;
 	tp->rcv_tstamp = tcp_jiffies32;
 	if (!prior_packets)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index ea370afa70ed..4f6894469b62 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -361,7 +361,7 @@ void tcp_v4_mtu_reduced(struct sock *sk)
 	 * for the case, if this connection will not able to recover.
 	 */
 	if (mtu < dst_mtu(dst) && ip_dont_fragment(sk, dst))
-		sk->sk_err_soft = EMSGSIZE;
+		WRITE_ONCE(sk->sk_err_soft, EMSGSIZE);
 
 	mtu = dst_mtu(dst);
 
@@ -602,7 +602,7 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
 
 			tcp_done(sk);
 		} else {
-			sk->sk_err_soft = err;
+			WRITE_ONCE(sk->sk_err_soft, err);
 		}
 		goto out;
 	}
@@ -628,7 +628,7 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
 		sk->sk_err = err;
 		sk_error_report(sk);
 	} else	{ /* Only an error on timeout */
-		sk->sk_err_soft = err;
+		WRITE_ONCE(sk->sk_err_soft, err);
 	}
 
 out:
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index cb79127f45c3..8823e2182713 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -67,7 +67,7 @@ u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when)
 
 static void tcp_write_err(struct sock *sk)
 {
-	sk->sk_err = sk->sk_err_soft ? : ETIMEDOUT;
+	sk->sk_err = READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT;
 	sk_error_report(sk);
 
 	tcp_write_queue_purge(sk);
@@ -110,7 +110,7 @@ static int tcp_out_of_resources(struct sock *sk, bool do_reset)
 		shift++;
 
 	/* If some dubious ICMP arrived, penalize even more. */
-	if (sk->sk_err_soft)
+	if (READ_ONCE(sk->sk_err_soft))
 		shift++;
 
 	if (tcp_check_oom(sk, shift)) {
@@ -146,7 +146,7 @@ static int tcp_orphan_retries(struct sock *sk, bool alive)
 	int retries = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_orphan_retries); /* May be zero. */
 
 	/* We know from an ICMP that something is wrong. */
-	if (sk->sk_err_soft && !alive)
+	if (READ_ONCE(sk->sk_err_soft) && !alive)
 		retries = 0;
 
 	/* However, if socket sent something recently, select some safe
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 1bf93b61aa06..dc963eebc668 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -497,8 +497,9 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			sk_error_report(sk);		/* Wake people up to see the error (see connect in sock.c) */
 
 			tcp_done(sk);
-		} else
-			sk->sk_err_soft = err;
+		} else {
+			WRITE_ONCE(sk->sk_err_soft, err);
+		}
 		goto out;
 	case TCP_LISTEN:
 		break;
@@ -514,9 +515,9 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	if (!sock_owned_by_user(sk) && np->recverr) {
 		sk->sk_err = err;
 		sk_error_report(sk);
-	} else
-		sk->sk_err_soft = err;
-
+	} else {
+		WRITE_ONCE(sk->sk_err_soft, err);
+	}
 out:
 	bh_unlock_sock(sk);
 	sock_put(sk);
-- 
cgit v1.2.3-70-g09d2


From 9a25f0cb0d7ee689f54f38890e66bc78520b0c62 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Mar 2023 20:57:42 +0000
Subject: dccp: annotate lockless accesses to sk->sk_err_soft

This field can be read/written without lock synchronization.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dccp/ipv4.c  | 12 +++++++-----
 net/dccp/ipv6.c  | 11 ++++++-----
 net/dccp/timer.c |  2 +-
 3 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index b780827f5e0a..3ab68415d121 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -177,7 +177,7 @@ static inline void dccp_do_pmtu_discovery(struct sock *sk,
 	 * for the case, if this connection will not able to recover.
 	 */
 	if (mtu < dst_mtu(dst) && ip_dont_fragment(sk, dst))
-		sk->sk_err_soft = EMSGSIZE;
+		WRITE_ONCE(sk->sk_err_soft, EMSGSIZE);
 
 	mtu = dst_mtu(dst);
 
@@ -339,8 +339,9 @@ static int dccp_v4_err(struct sk_buff *skb, u32 info)
 			sk_error_report(sk);
 
 			dccp_done(sk);
-		} else
-			sk->sk_err_soft = err;
+		} else {
+			WRITE_ONCE(sk->sk_err_soft, err);
+		}
 		goto out;
 	}
 
@@ -364,8 +365,9 @@ static int dccp_v4_err(struct sk_buff *skb, u32 info)
 	if (!sock_owned_by_user(sk) && inet->recverr) {
 		sk->sk_err = err;
 		sk_error_report(sk);
-	} else /* Only an error on timeout */
-		sk->sk_err_soft = err;
+	} else { /* Only an error on timeout */
+		WRITE_ONCE(sk->sk_err_soft, err);
+	}
 out:
 	bh_unlock_sock(sk);
 	sock_put(sk);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index b9d7c3dd1cb3..47fb10834223 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -174,17 +174,18 @@ static int dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			 */
 			sk_error_report(sk);
 			dccp_done(sk);
-		} else
-			sk->sk_err_soft = err;
+		} else {
+			WRITE_ONCE(sk->sk_err_soft, err);
+		}
 		goto out;
 	}
 
 	if (!sock_owned_by_user(sk) && np->recverr) {
 		sk->sk_err = err;
 		sk_error_report(sk);
-	} else
-		sk->sk_err_soft = err;
-
+	} else {
+		WRITE_ONCE(sk->sk_err_soft, err);
+	}
 out:
 	bh_unlock_sock(sk);
 	sock_put(sk);
diff --git a/net/dccp/timer.c b/net/dccp/timer.c
index 27a3b37acd2e..b3255e87cc7e 100644
--- a/net/dccp/timer.c
+++ b/net/dccp/timer.c
@@ -19,7 +19,7 @@ int  sysctl_dccp_retries2		__read_mostly = TCP_RETR2;
 
 static void dccp_write_err(struct sock *sk)
 {
-	sk->sk_err = sk->sk_err_soft ? : ETIMEDOUT;
+	sk->sk_err = READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT;
 	sk_error_report(sk);
 
 	dccp_send_reset(sk, DCCP_RESET_CODE_ABORTED);
-- 
cgit v1.2.3-70-g09d2


From 2f2d9972affae1f3282eff97c794cf843aedf61e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Mar 2023 20:57:43 +0000
Subject: net: annotate lockless accesses to sk->sk_err_soft

This field can be read/written without lock synchronization.

tcp and dccp have been handled in different patches.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/dlm/lowcomms.c                | 7 ++++---
 net/atm/signaling.c              | 2 +-
 net/ipv4/af_inet.c               | 2 +-
 net/ipv6/af_inet6.c              | 2 +-
 net/ipv6/inet6_connection_sock.c | 2 +-
 net/sctp/input.c                 | 2 +-
 net/sctp/ipv6.c                  | 2 +-
 7 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index a9b14f81d655..bd786b3be5ec 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -601,7 +601,7 @@ static void lowcomms_error_report(struct sock *sk)
 				   "sk_err=%d/%d\n", dlm_our_nodeid(),
 				   con->nodeid, &inet->inet_daddr,
 				   ntohs(inet->inet_dport), sk->sk_err,
-				   sk->sk_err_soft);
+				   READ_ONCE(sk->sk_err_soft));
 		break;
 #if IS_ENABLED(CONFIG_IPV6)
 	case AF_INET6:
@@ -610,14 +610,15 @@ static void lowcomms_error_report(struct sock *sk)
 				   "dport %d, sk_err=%d/%d\n", dlm_our_nodeid(),
 				   con->nodeid, &sk->sk_v6_daddr,
 				   ntohs(inet->inet_dport), sk->sk_err,
-				   sk->sk_err_soft);
+				   READ_ONCE(sk->sk_err_soft));
 		break;
 #endif
 	default:
 		printk_ratelimited(KERN_ERR "dlm: node %d: socket error "
 				   "invalid socket family %d set, "
 				   "sk_err=%d/%d\n", dlm_our_nodeid(),
-				   sk->sk_family, sk->sk_err, sk->sk_err_soft);
+				   sk->sk_family, sk->sk_err,
+				   READ_ONCE(sk->sk_err_soft));
 		break;
 	}
 
diff --git a/net/atm/signaling.c b/net/atm/signaling.c
index 5de06ab8ed75..e70ae2c113f9 100644
--- a/net/atm/signaling.c
+++ b/net/atm/signaling.c
@@ -125,7 +125,7 @@ as_indicate_complete:
 		break;
 	case as_addparty:
 	case as_dropparty:
-		sk->sk_err_soft = -msg->reply;
+		WRITE_ONCE(sk->sk_err_soft, -msg->reply);
 					/* < 0 failure, otherwise ep_ref */
 		clear_bit(ATM_VF_WAITING, &vcc->flags);
 		break;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 8db6747f892f..940062e08f57 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1322,7 +1322,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 		    sk->sk_state != TCP_SYN_SENT ||
 		    (sk->sk_userlocks & SOCK_BINDADDR_LOCK) ||
 		    (err = inet_sk_reselect_saddr(sk)) != 0)
-			sk->sk_err_soft = -err;
+			WRITE_ONCE(sk->sk_err_soft, -err);
 	}
 
 	return err;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 38689bedfce7..e1b679a590c9 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -845,7 +845,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 		dst = ip6_dst_lookup_flow(sock_net(sk), sk, &fl6, final_p);
 		if (IS_ERR(dst)) {
 			sk->sk_route_caps = 0;
-			sk->sk_err_soft = -PTR_ERR(dst);
+			WRITE_ONCE(sk->sk_err_soft, -PTR_ERR(dst));
 			return PTR_ERR(dst);
 		}
 
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 5a9f4d722f35..0c50dcd35fe8 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -120,7 +120,7 @@ int inet6_csk_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl_unused
 
 	dst = inet6_csk_route_socket(sk, &fl6);
 	if (IS_ERR(dst)) {
-		sk->sk_err_soft = -PTR_ERR(dst);
+		WRITE_ONCE(sk->sk_err_soft, -PTR_ERR(dst));
 		sk->sk_route_caps = 0;
 		kfree_skb(skb);
 		return PTR_ERR(dst);
diff --git a/net/sctp/input.c b/net/sctp/input.c
index bf70371301ff..127bf28a6033 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -585,7 +585,7 @@ static void sctp_v4_err_handle(struct sctp_transport *t, struct sk_buff *skb,
 		sk->sk_err = err;
 		sk_error_report(sk);
 	} else {  /* Only an error on timeout */
-		sk->sk_err_soft = err;
+		WRITE_ONCE(sk->sk_err_soft, err);
 	}
 }
 
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 62b436a2c8fe..43f2731bf590 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -155,7 +155,7 @@ static void sctp_v6_err_handle(struct sctp_transport *t, struct sk_buff *skb,
 		sk->sk_err = err;
 		sk_error_report(sk);
 	} else {
-		sk->sk_err_soft = err;
+		WRITE_ONCE(sk->sk_err_soft, err);
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From e13ec3da05d130f0d10da8e1fbe1be26dcdb0e27 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Mar 2023 20:57:44 +0000
Subject: tcp: annotate lockless access to sk->sk_err

tcp_poll() reads sk->sk_err without socket lock held/owned.

We should used READ_ONCE() here, and update writers
to use WRITE_ONCE().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp.c        | 11 ++++++-----
 net/ipv4/tcp_input.c  |  6 +++---
 net/ipv4/tcp_ipv4.c   |  4 ++--
 net/ipv4/tcp_output.c |  2 +-
 net/ipv4/tcp_timer.c  |  2 +-
 net/ipv6/tcp_ipv6.c   |  4 ++--
 6 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 288693981b00..01569de651b6 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -589,7 +589,8 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 	}
 	/* This barrier is coupled with smp_wmb() in tcp_reset() */
 	smp_rmb();
-	if (sk->sk_err || !skb_queue_empty_lockless(&sk->sk_error_queue))
+	if (READ_ONCE(sk->sk_err) ||
+	    !skb_queue_empty_lockless(&sk->sk_error_queue))
 		mask |= EPOLLERR;
 
 	return mask;
@@ -3094,7 +3095,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	if (old_state == TCP_LISTEN) {
 		inet_csk_listen_stop(sk);
 	} else if (unlikely(tp->repair)) {
-		sk->sk_err = ECONNABORTED;
+		WRITE_ONCE(sk->sk_err, ECONNABORTED);
 	} else if (tcp_need_reset(old_state) ||
 		   (tp->snd_nxt != tp->write_seq &&
 		    (1 << old_state) & (TCPF_CLOSING | TCPF_LAST_ACK))) {
@@ -3102,9 +3103,9 @@ int tcp_disconnect(struct sock *sk, int flags)
 		 * states
 		 */
 		tcp_send_active_reset(sk, gfp_any());
-		sk->sk_err = ECONNRESET;
+		WRITE_ONCE(sk->sk_err, ECONNRESET);
 	} else if (old_state == TCP_SYN_SENT)
-		sk->sk_err = ECONNRESET;
+		WRITE_ONCE(sk->sk_err, ECONNRESET);
 
 	tcp_clear_xmit_timers(sk);
 	__skb_queue_purge(&sk->sk_receive_queue);
@@ -4692,7 +4693,7 @@ int tcp_abort(struct sock *sk, int err)
 	bh_lock_sock(sk);
 
 	if (!sock_flag(sk, SOCK_DEAD)) {
-		sk->sk_err = err;
+		WRITE_ONCE(sk->sk_err, err);
 		/* This barrier is coupled with smp_rmb() in tcp_poll() */
 		smp_wmb();
 		sk_error_report(sk);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 8b5b6ca6617d..754ddbe0577f 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4322,15 +4322,15 @@ void tcp_reset(struct sock *sk, struct sk_buff *skb)
 	/* We want the right error as BSD sees it (and indeed as we do). */
 	switch (sk->sk_state) {
 	case TCP_SYN_SENT:
-		sk->sk_err = ECONNREFUSED;
+		WRITE_ONCE(sk->sk_err, ECONNREFUSED);
 		break;
 	case TCP_CLOSE_WAIT:
-		sk->sk_err = EPIPE;
+		WRITE_ONCE(sk->sk_err, EPIPE);
 		break;
 	case TCP_CLOSE:
 		return;
 	default:
-		sk->sk_err = ECONNRESET;
+		WRITE_ONCE(sk->sk_err, ECONNRESET);
 	}
 	/* This barrier is coupled with smp_rmb() in tcp_poll() */
 	smp_wmb();
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 4f6894469b62..89daa6b953ff 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -596,7 +596,7 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
 		ip_icmp_error(sk, skb, err, th->dest, info, (u8 *)th);
 
 		if (!sock_owned_by_user(sk)) {
-			sk->sk_err = err;
+			WRITE_ONCE(sk->sk_err, err);
 
 			sk_error_report(sk);
 
@@ -625,7 +625,7 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
 
 	inet = inet_sk(sk);
 	if (!sock_owned_by_user(sk) && inet->recverr) {
-		sk->sk_err = err;
+		WRITE_ONCE(sk->sk_err, err);
 		sk_error_report(sk);
 	} else	{ /* Only an error on timeout */
 		WRITE_ONCE(sk->sk_err_soft, err);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 71d01cf3c13e..f7e00d90a730 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3699,7 +3699,7 @@ static void tcp_connect_init(struct sock *sk)
 	tp->rx_opt.rcv_wscale = rcv_wscale;
 	tp->rcv_ssthresh = tp->rcv_wnd;
 
-	sk->sk_err = 0;
+	WRITE_ONCE(sk->sk_err, 0);
 	sock_reset_flag(sk, SOCK_DONE);
 	tp->snd_wnd = 0;
 	tcp_init_wl(tp, 0);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 8823e2182713..b839c2f91292 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -67,7 +67,7 @@ u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when)
 
 static void tcp_write_err(struct sock *sk)
 {
-	sk->sk_err = READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT;
+	WRITE_ONCE(sk->sk_err, READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT);
 	sk_error_report(sk);
 
 	tcp_write_queue_purge(sk);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index dc963eebc668..35cf523c9efd 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -493,7 +493,7 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		ipv6_icmp_error(sk, skb, err, th->dest, ntohl(info), (u8 *)th);
 
 		if (!sock_owned_by_user(sk)) {
-			sk->sk_err = err;
+			WRITE_ONCE(sk->sk_err, err);
 			sk_error_report(sk);		/* Wake people up to see the error (see connect in sock.c) */
 
 			tcp_done(sk);
@@ -513,7 +513,7 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	}
 
 	if (!sock_owned_by_user(sk) && np->recverr) {
-		sk->sk_err = err;
+		WRITE_ONCE(sk->sk_err, err);
 		sk_error_report(sk);
 	} else {
 		WRITE_ONCE(sk->sk_err_soft, err);
-- 
cgit v1.2.3-70-g09d2


From 9ae8e5ad99b8ebcd3d3dd46075f3825e6f08f063 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Mar 2023 20:57:45 +0000
Subject: mptcp: annotate lockless accesses to sk->sk_err

mptcp_poll() reads sk->sk_err without socket lock held/owned.

Add READ_ONCE() and WRITE_ONCE() to avoid load/store tearing.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/pm_netlink.c | 2 +-
 net/mptcp/protocol.c   | 8 ++++----
 net/mptcp/subflow.c    | 4 ++--
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 56628b52d100..cbaa1b49f7fe 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -2019,7 +2019,7 @@ static int mptcp_event_put_token_and_ssk(struct sk_buff *skb,
 	    nla_put_s32(skb, MPTCP_ATTR_IF_IDX, ssk->sk_bound_dev_if))
 		return -EMSGSIZE;
 
-	sk_err = ssk->sk_err;
+	sk_err = READ_ONCE(ssk->sk_err);
 	if (sk_err && sk->sk_state == TCP_ESTABLISHED &&
 	    nla_put_u8(skb, MPTCP_ATTR_ERROR, sk_err))
 		return -EMSGSIZE;
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 3ad9c46202fc..3005a5adf715 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2463,15 +2463,15 @@ static void mptcp_check_fastclose(struct mptcp_sock *msk)
 	/* Mirror the tcp_reset() error propagation */
 	switch (sk->sk_state) {
 	case TCP_SYN_SENT:
-		sk->sk_err = ECONNREFUSED;
+		WRITE_ONCE(sk->sk_err, ECONNREFUSED);
 		break;
 	case TCP_CLOSE_WAIT:
-		sk->sk_err = EPIPE;
+		WRITE_ONCE(sk->sk_err, EPIPE);
 		break;
 	case TCP_CLOSE:
 		return;
 	default:
-		sk->sk_err = ECONNRESET;
+		WRITE_ONCE(sk->sk_err, ECONNRESET);
 	}
 
 	inet_sk_state_store(sk, TCP_CLOSE);
@@ -3791,7 +3791,7 @@ static __poll_t mptcp_poll(struct file *file, struct socket *sock,
 
 	/* This barrier is coupled with smp_wmb() in __mptcp_error_report() */
 	smp_rmb();
-	if (sk->sk_err)
+	if (READ_ONCE(sk->sk_err))
 		mask |= EPOLLERR;
 
 	return mask;
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 4ae1a7304cf0..01874059a168 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1335,7 +1335,7 @@ fallback:
 			subflow->reset_reason = MPTCP_RST_EMPTCP;
 
 reset:
-			ssk->sk_err = EBADMSG;
+			WRITE_ONCE(ssk->sk_err, EBADMSG);
 			tcp_set_state(ssk, TCP_CLOSE);
 			while ((skb = skb_peek(&ssk->sk_receive_queue)))
 				sk_eat_skb(ssk, skb);
@@ -1419,7 +1419,7 @@ void __mptcp_error_report(struct sock *sk)
 		ssk_state = inet_sk_state_load(ssk);
 		if (ssk_state == TCP_CLOSE && !sock_flag(sk, SOCK_DEAD))
 			inet_sk_state_store(sk, ssk_state);
-		sk->sk_err = -err;
+		WRITE_ONCE(sk->sk_err, -err);
 
 		/* This barrier is coupled with smp_rmb() in mptcp_poll() */
 		smp_wmb();
-- 
cgit v1.2.3-70-g09d2


From cc04410af7de348234ac36a5f50c4ce416efdb4b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 15 Mar 2023 20:57:46 +0000
Subject: af_unix: annotate lockless accesses to sk->sk_err

unix_poll() and unix_dgram_poll() read sk->sk_err
without any lock held.

Add relevant READ_ONCE()/WRITE_ONCE() annotations.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/unix/af_unix.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 0b0f18ecce44..fb31e8a4409e 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -557,7 +557,7 @@ static void unix_dgram_disconnected(struct sock *sk, struct sock *other)
 		 * when peer was not connected to us.
 		 */
 		if (!sock_flag(other, SOCK_DEAD) && unix_peer(other) == sk) {
-			other->sk_err = ECONNRESET;
+			WRITE_ONCE(other->sk_err, ECONNRESET);
 			sk_error_report(other);
 		}
 	}
@@ -630,7 +630,7 @@ static void unix_release_sock(struct sock *sk, int embrion)
 			/* No more writes */
 			skpair->sk_shutdown = SHUTDOWN_MASK;
 			if (!skb_queue_empty(&sk->sk_receive_queue) || embrion)
-				skpair->sk_err = ECONNRESET;
+				WRITE_ONCE(skpair->sk_err, ECONNRESET);
 			unix_state_unlock(skpair);
 			skpair->sk_state_change(skpair);
 			sk_wake_async(skpair, SOCK_WAKE_WAITD, POLL_HUP);
@@ -3165,7 +3165,7 @@ static __poll_t unix_poll(struct file *file, struct socket *sock, poll_table *wa
 	mask = 0;
 
 	/* exceptional events? */
-	if (sk->sk_err)
+	if (READ_ONCE(sk->sk_err))
 		mask |= EPOLLERR;
 	if (sk->sk_shutdown == SHUTDOWN_MASK)
 		mask |= EPOLLHUP;
@@ -3208,7 +3208,8 @@ static __poll_t unix_dgram_poll(struct file *file, struct socket *sock,
 	mask = 0;
 
 	/* exceptional events? */
-	if (sk->sk_err || !skb_queue_empty_lockless(&sk->sk_error_queue))
+	if (READ_ONCE(sk->sk_err) ||
+	    !skb_queue_empty_lockless(&sk->sk_error_queue))
 		mask |= EPOLLERR |
 			(sock_flag(sk, SOCK_SELECT_ERR_QUEUE) ? EPOLLPRI : 0);
 
-- 
cgit v1.2.3-70-g09d2


From 2e80aeae9f807ac7e967dea7633abea5829c6531 Mon Sep 17 00:00:00 2001
From: Praveen Kaligineedi <pkaligineedi@google.com>
Date: Wed, 15 Mar 2023 16:33:08 -0700
Subject: gve: XDP support GQI-QPL: helper function changes

This patch adds/modifies helper functions needed to add XDP
support.

Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/google/gve/gve.h         |  5 ++++
 drivers/net/ethernet/google/gve/gve_ethtool.c | 26 ++++++++++------
 drivers/net/ethernet/google/gve/gve_main.c    | 27 ++++++++++-------
 drivers/net/ethernet/google/gve/gve_rx.c      |  2 +-
 drivers/net/ethernet/google/gve/gve_rx_dqo.c  |  2 +-
 drivers/net/ethernet/google/gve/gve_tx.c      | 43 ++++++++++++++++-----------
 drivers/net/ethernet/google/gve/gve_utils.c   |  6 ++--
 drivers/net/ethernet/google/gve/gve_utils.h   |  3 +-
 8 files changed, 70 insertions(+), 44 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index 64eb0442c82f..f52f23198278 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -855,6 +855,11 @@ static inline bool gve_is_gqi(struct gve_priv *priv)
 		priv->queue_format == GVE_GQI_QPL_FORMAT;
 }
 
+static inline u32 gve_num_tx_queues(struct gve_priv *priv)
+{
+	return priv->tx_cfg.num_queues;
+}
+
 /* buffers */
 int gve_alloc_page(struct gve_priv *priv, struct device *dev,
 		   struct page **page, dma_addr_t *dma,
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
index ce574d097e28..5b6e31812fae 100644
--- a/drivers/net/ethernet/google/gve/gve_ethtool.c
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -81,8 +81,10 @@ static void gve_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 {
 	struct gve_priv *priv = netdev_priv(netdev);
 	char *s = (char *)data;
+	int num_tx_queues;
 	int i, j;
 
+	num_tx_queues = gve_num_tx_queues(priv);
 	switch (stringset) {
 	case ETH_SS_STATS:
 		memcpy(s, *gve_gstrings_main_stats,
@@ -97,7 +99,7 @@ static void gve_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 			}
 		}
 
-		for (i = 0; i < priv->tx_cfg.num_queues; i++) {
+		for (i = 0; i < num_tx_queues; i++) {
 			for (j = 0; j < NUM_GVE_TX_CNTS; j++) {
 				snprintf(s, ETH_GSTRING_LEN,
 					 gve_gstrings_tx_stats[j], i);
@@ -124,12 +126,14 @@ static void gve_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 static int gve_get_sset_count(struct net_device *netdev, int sset)
 {
 	struct gve_priv *priv = netdev_priv(netdev);
+	int num_tx_queues;
 
+	num_tx_queues = gve_num_tx_queues(priv);
 	switch (sset) {
 	case ETH_SS_STATS:
 		return GVE_MAIN_STATS_LEN + GVE_ADMINQ_STATS_LEN +
 		       (priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS) +
-		       (priv->tx_cfg.num_queues * NUM_GVE_TX_CNTS);
+		       (num_tx_queues * NUM_GVE_TX_CNTS);
 	case ETH_SS_PRIV_FLAGS:
 		return GVE_PRIV_FLAGS_STR_LEN;
 	default:
@@ -153,18 +157,20 @@ gve_get_ethtool_stats(struct net_device *netdev,
 	struct gve_priv *priv;
 	bool skip_nic_stats;
 	unsigned int start;
+	int num_tx_queues;
 	int ring;
 	int i, j;
 
 	ASSERT_RTNL();
 
 	priv = netdev_priv(netdev);
+	num_tx_queues = gve_num_tx_queues(priv);
 	report_stats = priv->stats_report->stats;
 	rx_qid_to_stats_idx = kmalloc_array(priv->rx_cfg.num_queues,
 					    sizeof(int), GFP_KERNEL);
 	if (!rx_qid_to_stats_idx)
 		return;
-	tx_qid_to_stats_idx = kmalloc_array(priv->tx_cfg.num_queues,
+	tx_qid_to_stats_idx = kmalloc_array(num_tx_queues,
 					    sizeof(int), GFP_KERNEL);
 	if (!tx_qid_to_stats_idx) {
 		kfree(rx_qid_to_stats_idx);
@@ -195,7 +201,7 @@ gve_get_ethtool_stats(struct net_device *netdev,
 		}
 	}
 	for (tx_pkts = 0, tx_bytes = 0, tx_dropped = 0, ring = 0;
-	     ring < priv->tx_cfg.num_queues; ring++) {
+	     ring < num_tx_queues; ring++) {
 		if (priv->tx) {
 			do {
 				start =
@@ -232,7 +238,7 @@ gve_get_ethtool_stats(struct net_device *netdev,
 	i = GVE_MAIN_STATS_LEN;
 
 	/* For rx cross-reporting stats, start from nic rx stats in report */
-	base_stats_idx = GVE_TX_STATS_REPORT_NUM * priv->tx_cfg.num_queues +
+	base_stats_idx = GVE_TX_STATS_REPORT_NUM * num_tx_queues +
 		GVE_RX_STATS_REPORT_NUM * priv->rx_cfg.num_queues;
 	max_stats_idx = NIC_RX_STATS_REPORT_NUM * priv->rx_cfg.num_queues +
 		base_stats_idx;
@@ -298,7 +304,7 @@ gve_get_ethtool_stats(struct net_device *netdev,
 
 	/* For tx cross-reporting stats, start from nic tx stats in report */
 	base_stats_idx = max_stats_idx;
-	max_stats_idx = NIC_TX_STATS_REPORT_NUM * priv->tx_cfg.num_queues +
+	max_stats_idx = NIC_TX_STATS_REPORT_NUM * num_tx_queues +
 		max_stats_idx;
 	/* Preprocess the stats report for tx, map queue id to start index */
 	skip_nic_stats = false;
@@ -316,7 +322,7 @@ gve_get_ethtool_stats(struct net_device *netdev,
 	}
 	/* walk TX rings */
 	if (priv->tx) {
-		for (ring = 0; ring < priv->tx_cfg.num_queues; ring++) {
+		for (ring = 0; ring < num_tx_queues; ring++) {
 			struct gve_tx_ring *tx = &priv->tx[ring];
 
 			if (gve_is_gqi(priv)) {
@@ -355,7 +361,7 @@ gve_get_ethtool_stats(struct net_device *netdev,
 			}
 		}
 	} else {
-		i += priv->tx_cfg.num_queues * NUM_GVE_TX_CNTS;
+		i += num_tx_queues * NUM_GVE_TX_CNTS;
 	}
 
 	kfree(rx_qid_to_stats_idx);
@@ -502,7 +508,9 @@ static int gve_set_priv_flags(struct net_device *netdev, u32 flags)
 {
 	struct gve_priv *priv = netdev_priv(netdev);
 	u64 ori_flags, new_flags;
+	int num_tx_queues;
 
+	num_tx_queues = gve_num_tx_queues(priv);
 	ori_flags = READ_ONCE(priv->ethtool_flags);
 	new_flags = ori_flags;
 
@@ -522,7 +530,7 @@ static int gve_set_priv_flags(struct net_device *netdev, u32 flags)
 	/* delete report stats timer. */
 	if (!(flags & BIT(0)) && (ori_flags & BIT(0))) {
 		int tx_stats_num = GVE_TX_STATS_REPORT_NUM *
-			priv->tx_cfg.num_queues;
+			num_tx_queues;
 		int rx_stats_num = GVE_RX_STATS_REPORT_NUM *
 			priv->rx_cfg.num_queues;
 
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index 07111c241e0e..3cfdeeb74f60 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -90,8 +90,10 @@ static void gve_get_stats(struct net_device *dev, struct rtnl_link_stats64 *s)
 	struct gve_priv *priv = netdev_priv(dev);
 	unsigned int start;
 	u64 packets, bytes;
+	int num_tx_queues;
 	int ring;
 
+	num_tx_queues = gve_num_tx_queues(priv);
 	if (priv->rx) {
 		for (ring = 0; ring < priv->rx_cfg.num_queues; ring++) {
 			do {
@@ -106,7 +108,7 @@ static void gve_get_stats(struct net_device *dev, struct rtnl_link_stats64 *s)
 		}
 	}
 	if (priv->tx) {
-		for (ring = 0; ring < priv->tx_cfg.num_queues; ring++) {
+		for (ring = 0; ring < num_tx_queues; ring++) {
 			do {
 				start =
 				  u64_stats_fetch_begin(&priv->tx[ring].statss);
@@ -180,7 +182,7 @@ static int gve_alloc_stats_report(struct gve_priv *priv)
 	int tx_stats_num, rx_stats_num;
 
 	tx_stats_num = (GVE_TX_STATS_REPORT_NUM + NIC_TX_STATS_REPORT_NUM) *
-		       priv->tx_cfg.num_queues;
+		       gve_num_tx_queues(priv);
 	rx_stats_num = (GVE_RX_STATS_REPORT_NUM + NIC_RX_STATS_REPORT_NUM) *
 		       priv->rx_cfg.num_queues;
 	priv->stats_report_len = struct_size(priv->stats_report, stats,
@@ -622,20 +624,21 @@ static int gve_unregister_qpls(struct gve_priv *priv)
 
 static int gve_create_rings(struct gve_priv *priv)
 {
+	int num_tx_queues = gve_num_tx_queues(priv);
 	int err;
 	int i;
 
-	err = gve_adminq_create_tx_queues(priv, priv->tx_cfg.num_queues);
+	err = gve_adminq_create_tx_queues(priv, num_tx_queues);
 	if (err) {
 		netif_err(priv, drv, priv->dev, "failed to create %d tx queues\n",
-			  priv->tx_cfg.num_queues);
+			  num_tx_queues);
 		/* This failure will trigger a reset - no need to clean
 		 * up
 		 */
 		return err;
 	}
 	netif_dbg(priv, drv, priv->dev, "created %d tx queues\n",
-		  priv->tx_cfg.num_queues);
+		  num_tx_queues);
 
 	err = gve_adminq_create_rx_queues(priv, priv->rx_cfg.num_queues);
 	if (err) {
@@ -675,7 +678,7 @@ static void add_napi_init_sync_stats(struct gve_priv *priv,
 	int i;
 
 	/* Add tx napi & init sync stats*/
-	for (i = 0; i < priv->tx_cfg.num_queues; i++) {
+	for (i = 0; i < gve_num_tx_queues(priv); i++) {
 		int ntfy_idx = gve_tx_idx_to_ntfy(priv, i);
 
 		u64_stats_init(&priv->tx[i].statss);
@@ -753,9 +756,10 @@ free_tx:
 
 static int gve_destroy_rings(struct gve_priv *priv)
 {
+	int num_tx_queues = gve_num_tx_queues(priv);
 	int err;
 
-	err = gve_adminq_destroy_tx_queues(priv, priv->tx_cfg.num_queues);
+	err = gve_adminq_destroy_tx_queues(priv, num_tx_queues);
 	if (err) {
 		netif_err(priv, drv, priv->dev,
 			  "failed to destroy tx queues\n");
@@ -784,11 +788,12 @@ static void gve_rx_free_rings(struct gve_priv *priv)
 
 static void gve_free_rings(struct gve_priv *priv)
 {
+	int num_tx_queues = gve_num_tx_queues(priv);
 	int ntfy_idx;
 	int i;
 
 	if (priv->tx) {
-		for (i = 0; i < priv->tx_cfg.num_queues; i++) {
+		for (i = 0; i < num_tx_queues; i++) {
 			ntfy_idx = gve_tx_idx_to_ntfy(priv, i);
 			gve_remove_napi(priv, ntfy_idx);
 		}
@@ -1118,7 +1123,7 @@ static void gve_turndown(struct gve_priv *priv)
 		return;
 
 	/* Disable napi to prevent more work from coming in */
-	for (idx = 0; idx < priv->tx_cfg.num_queues; idx++) {
+	for (idx = 0; idx < gve_num_tx_queues(priv); idx++) {
 		int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
 		struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
 
@@ -1146,7 +1151,7 @@ static void gve_turnup(struct gve_priv *priv)
 	netif_tx_start_all_queues(priv->dev);
 
 	/* Enable napi and unmask interrupts for all queues */
-	for (idx = 0; idx < priv->tx_cfg.num_queues; idx++) {
+	for (idx = 0; idx < gve_num_tx_queues(priv); idx++) {
 		int ntfy_idx = gve_tx_idx_to_ntfy(priv, idx);
 		struct gve_notify_block *block = &priv->ntfy_blocks[ntfy_idx];
 
@@ -1306,7 +1311,7 @@ void gve_handle_report_stats(struct gve_priv *priv)
 	be64_add_cpu(&priv->stats_report->written_count, 1);
 	/* tx stats */
 	if (priv->tx) {
-		for (idx = 0; idx < priv->tx_cfg.num_queues; idx++) {
+		for (idx = 0; idx < gve_num_tx_queues(priv); idx++) {
 			u32 last_completion = 0;
 			u32 tx_frames = 0;
 
diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c
index 1f55137722b0..db1c74b1d7d3 100644
--- a/drivers/net/ethernet/google/gve/gve_rx.c
+++ b/drivers/net/ethernet/google/gve/gve_rx.c
@@ -556,7 +556,7 @@ static struct sk_buff *gve_rx_skb(struct gve_priv *priv, struct gve_rx_ring *rx,
 
 	if (len <= priv->rx_copybreak && is_only_frag)  {
 		/* Just copy small packets */
-		skb = gve_rx_copy(netdev, napi, page_info, len, GVE_RX_PAD);
+		skb = gve_rx_copy(netdev, napi, page_info, len);
 		if (skb) {
 			u64_stats_update_begin(&rx->statss);
 			rx->rx_copied_pkt++;
diff --git a/drivers/net/ethernet/google/gve/gve_rx_dqo.c b/drivers/net/ethernet/google/gve/gve_rx_dqo.c
index 630f42a3037b..e57b73eb70f6 100644
--- a/drivers/net/ethernet/google/gve/gve_rx_dqo.c
+++ b/drivers/net/ethernet/google/gve/gve_rx_dqo.c
@@ -568,7 +568,7 @@ static int gve_rx_dqo(struct napi_struct *napi, struct gve_rx_ring *rx,
 
 	if (eop && buf_len <= priv->rx_copybreak) {
 		rx->ctx.skb_head = gve_rx_copy(priv->dev, napi,
-					       &buf_state->page_info, buf_len, 0);
+					       &buf_state->page_info, buf_len);
 		if (unlikely(!rx->ctx.skb_head))
 			goto error;
 		rx->ctx.skb_tail = rx->ctx.skb_head;
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
index 4888bf05fbed..0fb052ce9e0b 100644
--- a/drivers/net/ethernet/google/gve/gve_tx.c
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -374,18 +374,18 @@ static int gve_maybe_stop_tx(struct gve_priv *priv, struct gve_tx_ring *tx,
 }
 
 static void gve_tx_fill_pkt_desc(union gve_tx_desc *pkt_desc,
-				 struct sk_buff *skb, bool is_gso,
+				 u16 csum_offset, u8 ip_summed, bool is_gso,
 				 int l4_hdr_offset, u32 desc_cnt,
-				 u16 hlen, u64 addr)
+				 u16 hlen, u64 addr, u16 pkt_len)
 {
 	/* l4_hdr_offset and csum_offset are in units of 16-bit words */
 	if (is_gso) {
 		pkt_desc->pkt.type_flags = GVE_TXD_TSO | GVE_TXF_L4CSUM;
-		pkt_desc->pkt.l4_csum_offset = skb->csum_offset >> 1;
+		pkt_desc->pkt.l4_csum_offset = csum_offset >> 1;
 		pkt_desc->pkt.l4_hdr_offset = l4_hdr_offset >> 1;
-	} else if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
+	} else if (likely(ip_summed == CHECKSUM_PARTIAL)) {
 		pkt_desc->pkt.type_flags = GVE_TXD_STD | GVE_TXF_L4CSUM;
-		pkt_desc->pkt.l4_csum_offset = skb->csum_offset >> 1;
+		pkt_desc->pkt.l4_csum_offset = csum_offset >> 1;
 		pkt_desc->pkt.l4_hdr_offset = l4_hdr_offset >> 1;
 	} else {
 		pkt_desc->pkt.type_flags = GVE_TXD_STD;
@@ -393,7 +393,7 @@ static void gve_tx_fill_pkt_desc(union gve_tx_desc *pkt_desc,
 		pkt_desc->pkt.l4_hdr_offset = 0;
 	}
 	pkt_desc->pkt.desc_cnt = desc_cnt;
-	pkt_desc->pkt.len = cpu_to_be16(skb->len);
+	pkt_desc->pkt.len = cpu_to_be16(pkt_len);
 	pkt_desc->pkt.seg_len = cpu_to_be16(hlen);
 	pkt_desc->pkt.seg_addr = cpu_to_be64(addr);
 }
@@ -412,15 +412,16 @@ static void gve_tx_fill_mtd_desc(union gve_tx_desc *mtd_desc,
 }
 
 static void gve_tx_fill_seg_desc(union gve_tx_desc *seg_desc,
-				 struct sk_buff *skb, bool is_gso,
+				 u16 l3_offset, u16 gso_size,
+				 bool is_gso_v6, bool is_gso,
 				 u16 len, u64 addr)
 {
 	seg_desc->seg.type_flags = GVE_TXD_SEG;
 	if (is_gso) {
-		if (skb_is_gso_v6(skb))
+		if (is_gso_v6)
 			seg_desc->seg.type_flags |= GVE_TXSF_IPV6;
-		seg_desc->seg.l3_offset = skb_network_offset(skb) >> 1;
-		seg_desc->seg.mss = cpu_to_be16(skb_shinfo(skb)->gso_size);
+		seg_desc->seg.l3_offset = l3_offset >> 1;
+		seg_desc->seg.mss = cpu_to_be16(gso_size);
 	}
 	seg_desc->seg.seg_len = cpu_to_be16(len);
 	seg_desc->seg.seg_addr = cpu_to_be64(addr);
@@ -473,9 +474,10 @@ static int gve_tx_add_skb_copy(struct gve_priv *priv, struct gve_tx_ring *tx, st
 	payload_nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, skb->len - hlen,
 					   &info->iov[payload_iov]);
 
-	gve_tx_fill_pkt_desc(pkt_desc, skb, is_gso, l4_hdr_offset,
+	gve_tx_fill_pkt_desc(pkt_desc, skb->csum_offset, skb->ip_summed,
+			     is_gso, l4_hdr_offset,
 			     1 + mtd_desc_nr + payload_nfrags, hlen,
-			     info->iov[hdr_nfrags - 1].iov_offset);
+			     info->iov[hdr_nfrags - 1].iov_offset, skb->len);
 
 	skb_copy_bits(skb, 0,
 		      tx->tx_fifo.base + info->iov[hdr_nfrags - 1].iov_offset,
@@ -494,7 +496,9 @@ static int gve_tx_add_skb_copy(struct gve_priv *priv, struct gve_tx_ring *tx, st
 		next_idx = (tx->req + 1 + mtd_desc_nr + i - payload_iov) & tx->mask;
 		seg_desc = &tx->desc[next_idx];
 
-		gve_tx_fill_seg_desc(seg_desc, skb, is_gso,
+		gve_tx_fill_seg_desc(seg_desc, skb_network_offset(skb),
+				     skb_shinfo(skb)->gso_size,
+				     skb_is_gso_v6(skb), is_gso,
 				     info->iov[i].iov_len,
 				     info->iov[i].iov_offset);
 
@@ -552,8 +556,9 @@ static int gve_tx_add_skb_no_copy(struct gve_priv *priv, struct gve_tx_ring *tx,
 	if (mtd_desc_nr)
 		num_descriptors++;
 
-	gve_tx_fill_pkt_desc(pkt_desc, skb, is_gso, l4_hdr_offset,
-			     num_descriptors, hlen, addr);
+	gve_tx_fill_pkt_desc(pkt_desc, skb->csum_offset, skb->ip_summed,
+			     is_gso, l4_hdr_offset,
+			     num_descriptors, hlen, addr, skb->len);
 
 	if (mtd_desc_nr) {
 		idx = (idx + 1) & tx->mask;
@@ -569,7 +574,9 @@ static int gve_tx_add_skb_no_copy(struct gve_priv *priv, struct gve_tx_ring *tx,
 		addr += hlen;
 		idx = (idx + 1) & tx->mask;
 		seg_desc = &tx->desc[idx];
-		gve_tx_fill_seg_desc(seg_desc, skb, is_gso, len, addr);
+		gve_tx_fill_seg_desc(seg_desc, skb_network_offset(skb),
+				     skb_shinfo(skb)->gso_size,
+				     skb_is_gso_v6(skb), is_gso, len, addr);
 	}
 
 	for (i = 0; i < shinfo->nr_frags; i++) {
@@ -587,7 +594,9 @@ static int gve_tx_add_skb_no_copy(struct gve_priv *priv, struct gve_tx_ring *tx,
 		dma_unmap_len_set(&tx->info[idx], len, len);
 		dma_unmap_addr_set(&tx->info[idx], dma, addr);
 
-		gve_tx_fill_seg_desc(seg_desc, skb, is_gso, len, addr);
+		gve_tx_fill_seg_desc(seg_desc, skb_network_offset(skb),
+				     skb_shinfo(skb)->gso_size,
+				     skb_is_gso_v6(skb), is_gso, len, addr);
 	}
 
 	return num_descriptors;
diff --git a/drivers/net/ethernet/google/gve/gve_utils.c b/drivers/net/ethernet/google/gve/gve_utils.c
index 6ba46adaaee3..26e08d753270 100644
--- a/drivers/net/ethernet/google/gve/gve_utils.c
+++ b/drivers/net/ethernet/google/gve/gve_utils.c
@@ -49,10 +49,10 @@ void gve_rx_add_to_block(struct gve_priv *priv, int queue_idx)
 }
 
 struct sk_buff *gve_rx_copy(struct net_device *dev, struct napi_struct *napi,
-			    struct gve_rx_slot_page_info *page_info, u16 len,
-			    u16 padding)
+			    struct gve_rx_slot_page_info *page_info, u16 len)
 {
-	void *va = page_info->page_address + padding + page_info->page_offset;
+	void *va = page_info->page_address + page_info->page_offset +
+		page_info->pad;
 	struct sk_buff *skb;
 
 	skb = napi_alloc_skb(napi, len);
diff --git a/drivers/net/ethernet/google/gve/gve_utils.h b/drivers/net/ethernet/google/gve/gve_utils.h
index 79595940b351..324fd98a6112 100644
--- a/drivers/net/ethernet/google/gve/gve_utils.h
+++ b/drivers/net/ethernet/google/gve/gve_utils.h
@@ -18,8 +18,7 @@ void gve_rx_remove_from_block(struct gve_priv *priv, int queue_idx);
 void gve_rx_add_to_block(struct gve_priv *priv, int queue_idx);
 
 struct sk_buff *gve_rx_copy(struct net_device *dev, struct napi_struct *napi,
-			    struct gve_rx_slot_page_info *page_info, u16 len,
-			    u16 pad);
+			    struct gve_rx_slot_page_info *page_info, u16 len);
 
 /* Decrement pagecnt_bias. Set it back to INT_MAX if it reached zero. */
 void gve_dec_pagecnt_bias(struct gve_rx_slot_page_info *page_info);
-- 
cgit v1.2.3-70-g09d2


From 7fc2bf78a430f975e53f2b1e87e99d8f83cafd11 Mon Sep 17 00:00:00 2001
From: Praveen Kaligineedi <pkaligineedi@google.com>
Date: Wed, 15 Mar 2023 16:33:09 -0700
Subject: gve: Changes to add new TX queues

Changes to enable adding and removing TX queues without calling
gve_close() and gve_open().

Made the following changes:
1) priv->tx, priv->rx and priv->qpls arrays are allocated based on
   max tx queues and max rx queues
2) Changed gve_adminq_create_tx_queues(), gve_adminq_destroy_tx_queues(),
gve_tx_alloc_rings() and gve_tx_free_rings() functions to add/remove a
subset of TX queues rather than all the TX queues.

Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/google/gve/gve.h        | 45 ++++++++++-----
 drivers/net/ethernet/google/gve/gve_adminq.c |  8 +--
 drivers/net/ethernet/google/gve/gve_adminq.h |  4 +-
 drivers/net/ethernet/google/gve/gve_main.c   | 83 ++++++++++++++++++++--------
 drivers/net/ethernet/google/gve/gve_rx.c     |  2 +-
 drivers/net/ethernet/google/gve/gve_tx.c     | 12 ++--
 6 files changed, 104 insertions(+), 50 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index f52f23198278..f354a6448c25 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -798,16 +798,35 @@ static inline u32 gve_num_rx_qpls(struct gve_priv *priv)
 	return priv->rx_cfg.num_queues;
 }
 
+static inline u32 gve_tx_qpl_id(struct gve_priv *priv, int tx_qid)
+{
+	return tx_qid;
+}
+
+static inline u32 gve_rx_qpl_id(struct gve_priv *priv, int rx_qid)
+{
+	return priv->tx_cfg.max_queues + rx_qid;
+}
+
+static inline u32 gve_tx_start_qpl_id(struct gve_priv *priv)
+{
+	return gve_tx_qpl_id(priv, 0);
+}
+
+static inline u32 gve_rx_start_qpl_id(struct gve_priv *priv)
+{
+	return gve_rx_qpl_id(priv, 0);
+}
+
 /* Returns a pointer to the next available tx qpl in the list of qpls
  */
 static inline
-struct gve_queue_page_list *gve_assign_tx_qpl(struct gve_priv *priv)
+struct gve_queue_page_list *gve_assign_tx_qpl(struct gve_priv *priv, int tx_qid)
 {
-	int id = find_first_zero_bit(priv->qpl_cfg.qpl_id_map,
-				     priv->qpl_cfg.qpl_map_size);
+	int id = gve_tx_qpl_id(priv, tx_qid);
 
-	/* we are out of tx qpls */
-	if (id >= gve_num_tx_qpls(priv))
+	/* QPL already in use */
+	if (test_bit(id, priv->qpl_cfg.qpl_id_map))
 		return NULL;
 
 	set_bit(id, priv->qpl_cfg.qpl_id_map);
@@ -817,14 +836,12 @@ struct gve_queue_page_list *gve_assign_tx_qpl(struct gve_priv *priv)
 /* Returns a pointer to the next available rx qpl in the list of qpls
  */
 static inline
-struct gve_queue_page_list *gve_assign_rx_qpl(struct gve_priv *priv)
+struct gve_queue_page_list *gve_assign_rx_qpl(struct gve_priv *priv, int rx_qid)
 {
-	int id = find_next_zero_bit(priv->qpl_cfg.qpl_id_map,
-				    priv->qpl_cfg.qpl_map_size,
-				    gve_num_tx_qpls(priv));
+	int id = gve_rx_qpl_id(priv, rx_qid);
 
-	/* we are out of rx qpls */
-	if (id == gve_num_tx_qpls(priv) + gve_num_rx_qpls(priv))
+	/* QPL already in use */
+	if (test_bit(id, priv->qpl_cfg.qpl_id_map))
 		return NULL;
 
 	set_bit(id, priv->qpl_cfg.qpl_id_map);
@@ -843,7 +860,7 @@ static inline void gve_unassign_qpl(struct gve_priv *priv, int id)
 static inline enum dma_data_direction gve_qpl_dma_dir(struct gve_priv *priv,
 						      int id)
 {
-	if (id < gve_num_tx_qpls(priv))
+	if (id < gve_rx_start_qpl_id(priv))
 		return DMA_TO_DEVICE;
 	else
 		return DMA_FROM_DEVICE;
@@ -869,8 +886,8 @@ void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma,
 /* tx handling */
 netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev);
 bool gve_tx_poll(struct gve_notify_block *block, int budget);
-int gve_tx_alloc_rings(struct gve_priv *priv);
-void gve_tx_free_rings_gqi(struct gve_priv *priv);
+int gve_tx_alloc_rings(struct gve_priv *priv, int start_id, int num_rings);
+void gve_tx_free_rings_gqi(struct gve_priv *priv, int start_id, int num_rings);
 u32 gve_tx_load_event_counter(struct gve_priv *priv,
 			      struct gve_tx_ring *tx);
 bool gve_tx_clean_pending(struct gve_priv *priv, struct gve_tx_ring *tx);
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c
index 60061288ad9d..252974202a3f 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.c
+++ b/drivers/net/ethernet/google/gve/gve_adminq.c
@@ -516,12 +516,12 @@ static int gve_adminq_create_tx_queue(struct gve_priv *priv, u32 queue_index)
 	return gve_adminq_issue_cmd(priv, &cmd);
 }
 
-int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 num_queues)
+int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues)
 {
 	int err;
 	int i;
 
-	for (i = 0; i < num_queues; i++) {
+	for (i = start_id; i < start_id + num_queues; i++) {
 		err = gve_adminq_create_tx_queue(priv, i);
 		if (err)
 			return err;
@@ -604,12 +604,12 @@ static int gve_adminq_destroy_tx_queue(struct gve_priv *priv, u32 queue_index)
 	return 0;
 }
 
-int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 num_queues)
+int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues)
 {
 	int err;
 	int i;
 
-	for (i = 0; i < num_queues; i++) {
+	for (i = start_id; i < start_id + num_queues; i++) {
 		err = gve_adminq_destroy_tx_queue(priv, i);
 		if (err)
 			return err;
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h
index cf29662e6ad1..f894beb3deaf 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.h
+++ b/drivers/net/ethernet/google/gve/gve_adminq.h
@@ -410,8 +410,8 @@ int gve_adminq_configure_device_resources(struct gve_priv *priv,
 					  dma_addr_t db_array_bus_addr,
 					  u32 num_ntfy_blks);
 int gve_adminq_deconfigure_device_resources(struct gve_priv *priv);
-int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 num_queues);
-int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 queue_id);
+int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues);
+int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 start_id, u32 num_queues);
 int gve_adminq_create_rx_queues(struct gve_priv *priv, u32 num_queues);
 int gve_adminq_destroy_rx_queues(struct gve_priv *priv, u32 queue_id);
 int gve_adminq_register_page_list(struct gve_priv *priv,
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index 3cfdeeb74f60..160ca77c2751 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -584,11 +584,26 @@ static void gve_remove_napi(struct gve_priv *priv, int ntfy_idx)
 
 static int gve_register_qpls(struct gve_priv *priv)
 {
-	int num_qpls = gve_num_tx_qpls(priv) + gve_num_rx_qpls(priv);
+	int start_id;
 	int err;
 	int i;
 
-	for (i = 0; i < num_qpls; i++) {
+	start_id = gve_tx_start_qpl_id(priv);
+	for (i = start_id; i < start_id + gve_num_tx_qpls(priv); i++) {
+		err = gve_adminq_register_page_list(priv, &priv->qpls[i]);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "failed to register queue page list %d\n",
+				  priv->qpls[i].id);
+			/* This failure will trigger a reset - no need to clean
+			 * up
+			 */
+			return err;
+		}
+	}
+
+	start_id = gve_rx_start_qpl_id(priv);
+	for (i = start_id; i < start_id + gve_num_rx_qpls(priv); i++) {
 		err = gve_adminq_register_page_list(priv, &priv->qpls[i]);
 		if (err) {
 			netif_err(priv, drv, priv->dev,
@@ -605,11 +620,24 @@ static int gve_register_qpls(struct gve_priv *priv)
 
 static int gve_unregister_qpls(struct gve_priv *priv)
 {
-	int num_qpls = gve_num_tx_qpls(priv) + gve_num_rx_qpls(priv);
+	int start_id;
 	int err;
 	int i;
 
-	for (i = 0; i < num_qpls; i++) {
+	start_id = gve_tx_start_qpl_id(priv);
+	for (i = start_id; i < start_id + gve_num_tx_qpls(priv); i++) {
+		err = gve_adminq_unregister_page_list(priv, priv->qpls[i].id);
+		/* This failure will trigger a reset - no need to clean up */
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to unregister queue page list %d\n",
+				  priv->qpls[i].id);
+			return err;
+		}
+	}
+
+	start_id = gve_rx_start_qpl_id(priv);
+	for (i = start_id; i < start_id + gve_num_rx_qpls(priv); i++) {
 		err = gve_adminq_unregister_page_list(priv, priv->qpls[i].id);
 		/* This failure will trigger a reset - no need to clean up */
 		if (err) {
@@ -628,7 +656,7 @@ static int gve_create_rings(struct gve_priv *priv)
 	int err;
 	int i;
 
-	err = gve_adminq_create_tx_queues(priv, num_tx_queues);
+	err = gve_adminq_create_tx_queues(priv, 0, num_tx_queues);
 	if (err) {
 		netif_err(priv, drv, priv->dev, "failed to create %d tx queues\n",
 			  num_tx_queues);
@@ -695,10 +723,10 @@ static void add_napi_init_sync_stats(struct gve_priv *priv,
 	}
 }
 
-static void gve_tx_free_rings(struct gve_priv *priv)
+static void gve_tx_free_rings(struct gve_priv *priv, int start_id, int num_rings)
 {
 	if (gve_is_gqi(priv)) {
-		gve_tx_free_rings_gqi(priv);
+		gve_tx_free_rings_gqi(priv, start_id, num_rings);
 	} else {
 		gve_tx_free_rings_dqo(priv);
 	}
@@ -709,20 +737,20 @@ static int gve_alloc_rings(struct gve_priv *priv)
 	int err;
 
 	/* Setup tx rings */
-	priv->tx = kvcalloc(priv->tx_cfg.num_queues, sizeof(*priv->tx),
+	priv->tx = kvcalloc(priv->tx_cfg.max_queues, sizeof(*priv->tx),
 			    GFP_KERNEL);
 	if (!priv->tx)
 		return -ENOMEM;
 
 	if (gve_is_gqi(priv))
-		err = gve_tx_alloc_rings(priv);
+		err = gve_tx_alloc_rings(priv, 0, gve_num_tx_queues(priv));
 	else
 		err = gve_tx_alloc_rings_dqo(priv);
 	if (err)
 		goto free_tx;
 
 	/* Setup rx rings */
-	priv->rx = kvcalloc(priv->rx_cfg.num_queues, sizeof(*priv->rx),
+	priv->rx = kvcalloc(priv->rx_cfg.max_queues, sizeof(*priv->rx),
 			    GFP_KERNEL);
 	if (!priv->rx) {
 		err = -ENOMEM;
@@ -747,7 +775,7 @@ free_rx:
 	kvfree(priv->rx);
 	priv->rx = NULL;
 free_tx_queue:
-	gve_tx_free_rings(priv);
+	gve_tx_free_rings(priv, 0, gve_num_tx_queues(priv));
 free_tx:
 	kvfree(priv->tx);
 	priv->tx = NULL;
@@ -759,7 +787,7 @@ static int gve_destroy_rings(struct gve_priv *priv)
 	int num_tx_queues = gve_num_tx_queues(priv);
 	int err;
 
-	err = gve_adminq_destroy_tx_queues(priv, num_tx_queues);
+	err = gve_adminq_destroy_tx_queues(priv, 0, num_tx_queues);
 	if (err) {
 		netif_err(priv, drv, priv->dev,
 			  "failed to destroy tx queues\n");
@@ -797,7 +825,7 @@ static void gve_free_rings(struct gve_priv *priv)
 			ntfy_idx = gve_tx_idx_to_ntfy(priv, i);
 			gve_remove_napi(priv, ntfy_idx);
 		}
-		gve_tx_free_rings(priv);
+		gve_tx_free_rings(priv, 0, num_tx_queues);
 		kvfree(priv->tx);
 		priv->tx = NULL;
 	}
@@ -894,40 +922,46 @@ static void gve_free_queue_page_list(struct gve_priv *priv, u32 id)
 			      qpl->page_buses[i], gve_qpl_dma_dir(priv, id));
 
 	kvfree(qpl->page_buses);
+	qpl->page_buses = NULL;
 free_pages:
 	kvfree(qpl->pages);
+	qpl->pages = NULL;
 	priv->num_registered_pages -= qpl->num_entries;
 }
 
 static int gve_alloc_qpls(struct gve_priv *priv)
 {
-	int num_qpls = gve_num_tx_qpls(priv) + gve_num_rx_qpls(priv);
+	int max_queues = priv->tx_cfg.max_queues + priv->rx_cfg.max_queues;
+	int start_id;
 	int i, j;
 	int err;
 
-	if (num_qpls == 0)
+	if (priv->queue_format != GVE_GQI_QPL_FORMAT)
 		return 0;
 
-	priv->qpls = kvcalloc(num_qpls, sizeof(*priv->qpls), GFP_KERNEL);
+	priv->qpls = kvcalloc(max_queues, sizeof(*priv->qpls), GFP_KERNEL);
 	if (!priv->qpls)
 		return -ENOMEM;
 
-	for (i = 0; i < gve_num_tx_qpls(priv); i++) {
+	start_id = gve_tx_start_qpl_id(priv);
+	for (i = start_id; i < start_id + gve_num_tx_qpls(priv); i++) {
 		err = gve_alloc_queue_page_list(priv, i,
 						priv->tx_pages_per_qpl);
 		if (err)
 			goto free_qpls;
 	}
-	for (; i < num_qpls; i++) {
+
+	start_id = gve_rx_start_qpl_id(priv);
+	for (i = start_id; i < start_id + gve_num_rx_qpls(priv); i++) {
 		err = gve_alloc_queue_page_list(priv, i,
 						priv->rx_data_slot_cnt);
 		if (err)
 			goto free_qpls;
 	}
 
-	priv->qpl_cfg.qpl_map_size = BITS_TO_LONGS(num_qpls) *
+	priv->qpl_cfg.qpl_map_size = BITS_TO_LONGS(max_queues) *
 				     sizeof(unsigned long) * BITS_PER_BYTE;
-	priv->qpl_cfg.qpl_id_map = kvcalloc(BITS_TO_LONGS(num_qpls),
+	priv->qpl_cfg.qpl_id_map = kvcalloc(BITS_TO_LONGS(max_queues),
 					    sizeof(unsigned long), GFP_KERNEL);
 	if (!priv->qpl_cfg.qpl_id_map) {
 		err = -ENOMEM;
@@ -940,23 +974,26 @@ free_qpls:
 	for (j = 0; j <= i; j++)
 		gve_free_queue_page_list(priv, j);
 	kvfree(priv->qpls);
+	priv->qpls = NULL;
 	return err;
 }
 
 static void gve_free_qpls(struct gve_priv *priv)
 {
-	int num_qpls = gve_num_tx_qpls(priv) + gve_num_rx_qpls(priv);
+	int max_queues = priv->tx_cfg.max_queues + priv->rx_cfg.max_queues;
 	int i;
 
-	if (num_qpls == 0)
+	if (!priv->qpls)
 		return;
 
 	kvfree(priv->qpl_cfg.qpl_id_map);
+	priv->qpl_cfg.qpl_id_map = NULL;
 
-	for (i = 0; i < num_qpls; i++)
+	for (i = 0; i < max_queues; i++)
 		gve_free_queue_page_list(priv, i);
 
 	kvfree(priv->qpls);
+	priv->qpls = NULL;
 }
 
 /* Use this to schedule a reset when the device is capable of continuing
diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c
index db1c74b1d7d3..051a15e4f1af 100644
--- a/drivers/net/ethernet/google/gve/gve_rx.c
+++ b/drivers/net/ethernet/google/gve/gve_rx.c
@@ -124,7 +124,7 @@ static int gve_prefill_rx_pages(struct gve_rx_ring *rx)
 		return -ENOMEM;
 
 	if (!rx->data.raw_addressing) {
-		rx->data.qpl = gve_assign_rx_qpl(priv);
+		rx->data.qpl = gve_assign_rx_qpl(priv, rx->q_num);
 		if (!rx->data.qpl) {
 			kvfree(rx->data.page_info);
 			rx->data.page_info = NULL;
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
index 0fb052ce9e0b..e24e73e74e33 100644
--- a/drivers/net/ethernet/google/gve/gve_tx.c
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -195,7 +195,7 @@ static int gve_tx_alloc_ring(struct gve_priv *priv, int idx)
 	tx->raw_addressing = priv->queue_format == GVE_GQI_RDA_FORMAT;
 	tx->dev = &priv->pdev->dev;
 	if (!tx->raw_addressing) {
-		tx->tx_fifo.qpl = gve_assign_tx_qpl(priv);
+		tx->tx_fifo.qpl = gve_assign_tx_qpl(priv, idx);
 		if (!tx->tx_fifo.qpl)
 			goto abort_with_desc;
 		/* map Tx FIFO */
@@ -233,12 +233,12 @@ abort_with_info:
 	return -ENOMEM;
 }
 
-int gve_tx_alloc_rings(struct gve_priv *priv)
+int gve_tx_alloc_rings(struct gve_priv *priv, int start_id, int num_rings)
 {
 	int err = 0;
 	int i;
 
-	for (i = 0; i < priv->tx_cfg.num_queues; i++) {
+	for (i = start_id; i < start_id + num_rings; i++) {
 		err = gve_tx_alloc_ring(priv, i);
 		if (err) {
 			netif_err(priv, drv, priv->dev,
@@ -251,17 +251,17 @@ int gve_tx_alloc_rings(struct gve_priv *priv)
 	if (err) {
 		int j;
 
-		for (j = 0; j < i; j++)
+		for (j = start_id; j < i; j++)
 			gve_tx_free_ring(priv, j);
 	}
 	return err;
 }
 
-void gve_tx_free_rings_gqi(struct gve_priv *priv)
+void gve_tx_free_rings_gqi(struct gve_priv *priv, int start_id, int num_rings)
 {
 	int i;
 
-	for (i = 0; i < priv->tx_cfg.num_queues; i++)
+	for (i = start_id; i < start_id + num_rings; i++)
 		gve_tx_free_ring(priv, i);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 75eaae158b1b7d8d5bde2bafc0bcf778423071d3 Mon Sep 17 00:00:00 2001
From: Praveen Kaligineedi <pkaligineedi@google.com>
Date: Wed, 15 Mar 2023 16:33:10 -0700
Subject: gve: Add XDP DROP and TX support for GQI-QPL format

Add support for XDP PASS, DROP and TX actions.

This patch contains the following changes:
1) Support installing/uninstalling XDP program
2) Add dedicated XDP TX queues
3) Add support for XDP DROP action
4) Add support for XDP TX action

Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/google/gve/gve.h         |  44 ++-
 drivers/net/ethernet/google/gve/gve_ethtool.c |  37 ++-
 drivers/net/ethernet/google/gve/gve_main.c    | 422 ++++++++++++++++++++++++--
 drivers/net/ethernet/google/gve/gve_rx.c      |  74 ++++-
 drivers/net/ethernet/google/gve/gve_tx.c      | 149 ++++++++-
 5 files changed, 687 insertions(+), 39 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index f354a6448c25..8d5234d4ba67 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -47,6 +47,10 @@
 
 #define GVE_RX_BUFFER_SIZE_DQO 2048
 
+#define GVE_XDP_ACTIONS 5
+
+#define GVE_TX_MAX_HEADER_SIZE 182
+
 /* Each slot in the desc ring has a 1:1 mapping to a slot in the data ring */
 struct gve_rx_desc_queue {
 	struct gve_rx_desc *desc_ring; /* the descriptor ring */
@@ -230,7 +234,9 @@ struct gve_rx_ring {
 	u64 rx_frag_flip_cnt; /* free-running count of rx segments where page_flip was used */
 	u64 rx_frag_copy_cnt; /* free-running count of rx segments copied */
 	u64 rx_frag_alloc_cnt; /* free-running count of rx page allocations */
-
+	u64 xdp_tx_errors;
+	u64 xdp_redirect_errors;
+	u64 xdp_actions[GVE_XDP_ACTIONS];
 	u32 q_num; /* queue index */
 	u32 ntfy_id; /* notification block index */
 	struct gve_queue_resources *q_resources; /* head and tail pointer idx */
@@ -238,6 +244,9 @@ struct gve_rx_ring {
 	struct u64_stats_sync statss; /* sync stats for 32bit archs */
 
 	struct gve_rx_ctx ctx; /* Info for packet currently being processed in this ring. */
+
+	/* XDP stuff */
+	struct xdp_rxq_info xdp_rxq;
 };
 
 /* A TX desc ring entry */
@@ -259,6 +268,9 @@ struct gve_tx_iovec {
  */
 struct gve_tx_buffer_state {
 	struct sk_buff *skb; /* skb for this pkt */
+	struct {
+		u16 size; /* size of xmitted xdp pkt */
+	} xdp;
 	union {
 		struct gve_tx_iovec iov[GVE_TX_MAX_IOVEC]; /* segments of this pkt */
 		struct {
@@ -526,9 +538,11 @@ struct gve_priv {
 	u16 rx_data_slot_cnt; /* rx buffer length */
 	u64 max_registered_pages;
 	u64 num_registered_pages; /* num pages registered with NIC */
+	struct bpf_prog *xdp_prog; /* XDP BPF program */
 	u32 rx_copybreak; /* copy packets smaller than this */
 	u16 default_num_queues; /* default num queues to set up */
 
+	u16 num_xdp_queues;
 	struct gve_queue_config tx_cfg;
 	struct gve_queue_config rx_cfg;
 	struct gve_qpl_config qpl_cfg; /* map used QPL ids */
@@ -785,7 +799,17 @@ static inline u32 gve_num_tx_qpls(struct gve_priv *priv)
 	if (priv->queue_format != GVE_GQI_QPL_FORMAT)
 		return 0;
 
-	return priv->tx_cfg.num_queues;
+	return priv->tx_cfg.num_queues + priv->num_xdp_queues;
+}
+
+/* Returns the number of XDP tx queue page lists
+ */
+static inline u32 gve_num_xdp_qpls(struct gve_priv *priv)
+{
+	if (priv->queue_format != GVE_GQI_QPL_FORMAT)
+		return 0;
+
+	return priv->num_xdp_queues;
 }
 
 /* Returns the number of rx queue page lists
@@ -874,7 +898,17 @@ static inline bool gve_is_gqi(struct gve_priv *priv)
 
 static inline u32 gve_num_tx_queues(struct gve_priv *priv)
 {
-	return priv->tx_cfg.num_queues;
+	return priv->tx_cfg.num_queues + priv->num_xdp_queues;
+}
+
+static inline u32 gve_xdp_tx_queue_id(struct gve_priv *priv, u32 queue_id)
+{
+	return priv->tx_cfg.num_queues + queue_id;
+}
+
+static inline u32 gve_xdp_tx_start_queue_id(struct gve_priv *priv)
+{
+	return gve_xdp_tx_queue_id(priv, 0);
 }
 
 /* buffers */
@@ -885,7 +919,11 @@ void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma,
 		   enum dma_data_direction);
 /* tx handling */
 netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev);
+int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
+		     void *data, int len);
+void gve_xdp_tx_flush(struct gve_priv *priv, u32 xdp_qid);
 bool gve_tx_poll(struct gve_notify_block *block, int budget);
+bool gve_xdp_poll(struct gve_notify_block *block, int budget);
 int gve_tx_alloc_rings(struct gve_priv *priv, int start_id, int num_rings);
 void gve_tx_free_rings_gqi(struct gve_priv *priv, int start_id, int num_rings);
 u32 gve_tx_load_event_counter(struct gve_priv *priv,
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
index 5b6e31812fae..067b393ccf9d 100644
--- a/drivers/net/ethernet/google/gve/gve_ethtool.c
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -34,6 +34,11 @@ static u32 gve_get_msglevel(struct net_device *netdev)
 	return priv->msg_enable;
 }
 
+/* For the following stats column string names, make sure the order
+ * matches how it is filled in the code. For xdp_aborted, xdp_drop,
+ * xdp_pass, xdp_tx, xdp_redirect, make sure it also matches the order
+ * as declared in enum xdp_action inside file uapi/linux/bpf.h .
+ */
 static const char gve_gstrings_main_stats[][ETH_GSTRING_LEN] = {
 	"rx_packets", "tx_packets", "rx_bytes", "tx_bytes",
 	"rx_dropped", "tx_dropped", "tx_timeouts",
@@ -49,6 +54,9 @@ static const char gve_gstrings_rx_stats[][ETH_GSTRING_LEN] = {
 	"rx_dropped_pkt[%u]", "rx_copybreak_pkt[%u]", "rx_copied_pkt[%u]",
 	"rx_queue_drop_cnt[%u]", "rx_no_buffers_posted[%u]",
 	"rx_drops_packet_over_mru[%u]", "rx_drops_invalid_checksum[%u]",
+	"rx_xdp_aborted[%u]", "rx_xdp_drop[%u]", "rx_xdp_pass[%u]",
+	"rx_xdp_tx[%u]", "rx_xdp_redirect[%u]",
+	"rx_xdp_tx_errors[%u]", "rx_xdp_redirect_errors[%u]",
 };
 
 static const char gve_gstrings_tx_stats[][ETH_GSTRING_LEN] = {
@@ -289,14 +297,25 @@ gve_get_ethtool_stats(struct net_device *netdev,
 			if (skip_nic_stats) {
 				/* skip NIC rx stats */
 				i += NIC_RX_STATS_REPORT_NUM;
-				continue;
-			}
-			for (j = 0; j < NIC_RX_STATS_REPORT_NUM; j++) {
-				u64 value =
-				be64_to_cpu(report_stats[rx_qid_to_stats_idx[ring] + j].value);
+			} else {
+				stats_idx = rx_qid_to_stats_idx[ring];
+				for (j = 0; j < NIC_RX_STATS_REPORT_NUM; j++) {
+					u64 value =
+						be64_to_cpu(report_stats[stats_idx + j].value);
 
-				data[i++] = value;
+					data[i++] = value;
+				}
 			}
+			/* XDP rx counters */
+			do {
+				start =	u64_stats_fetch_begin(&priv->rx[ring].statss);
+				for (j = 0; j < GVE_XDP_ACTIONS; j++)
+					data[i + j] = rx->xdp_actions[j];
+				data[i + j++] = rx->xdp_tx_errors;
+				data[i + j++] = rx->xdp_redirect_errors;
+			} while (u64_stats_fetch_retry(&priv->rx[ring].statss,
+						       start));
+			i += GVE_XDP_ACTIONS + 2; /* XDP rx counters */
 		}
 	} else {
 		i += priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS;
@@ -418,6 +437,12 @@ static int gve_set_channels(struct net_device *netdev,
 	if (!new_rx || !new_tx)
 		return -EINVAL;
 
+	if (priv->num_xdp_queues &&
+	    (new_tx != new_rx || (2 * new_tx > priv->tx_cfg.max_queues))) {
+		dev_err(&priv->pdev->dev, "XDP load failed: The number of configured RX queues should be equal to the number of configured TX queues and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues");
+		return -EINVAL;
+	}
+
 	if (!netif_carrier_ok(netdev)) {
 		priv->tx_cfg.num_queues = new_tx;
 		priv->rx_cfg.num_queues = new_rx;
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index 160ca77c2751..f49398857921 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -4,8 +4,10 @@
  * Copyright (C) 2015-2021 Google, Inc.
  */
 
+#include <linux/bpf.h>
 #include <linux/cpumask.h>
 #include <linux/etherdevice.h>
+#include <linux/filter.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <linux/pci.h>
@@ -247,8 +249,13 @@ static int gve_napi_poll(struct napi_struct *napi, int budget)
 	block = container_of(napi, struct gve_notify_block, napi);
 	priv = block->priv;
 
-	if (block->tx)
-		reschedule |= gve_tx_poll(block, budget);
+	if (block->tx) {
+		if (block->tx->q_num < priv->tx_cfg.num_queues)
+			reschedule |= gve_tx_poll(block, budget);
+		else
+			reschedule |= gve_xdp_poll(block, budget);
+	}
+
 	if (block->rx) {
 		work_done = gve_rx_poll(block, budget);
 		reschedule |= work_done == budget;
@@ -582,6 +589,28 @@ static void gve_remove_napi(struct gve_priv *priv, int ntfy_idx)
 	netif_napi_del(&block->napi);
 }
 
+static int gve_register_xdp_qpls(struct gve_priv *priv)
+{
+	int start_id;
+	int err;
+	int i;
+
+	start_id = gve_tx_qpl_id(priv, gve_xdp_tx_start_queue_id(priv));
+	for (i = start_id; i < start_id + gve_num_xdp_qpls(priv); i++) {
+		err = gve_adminq_register_page_list(priv, &priv->qpls[i]);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "failed to register queue page list %d\n",
+				  priv->qpls[i].id);
+			/* This failure will trigger a reset - no need to clean
+			 * up
+			 */
+			return err;
+		}
+	}
+	return 0;
+}
+
 static int gve_register_qpls(struct gve_priv *priv)
 {
 	int start_id;
@@ -618,6 +647,26 @@ static int gve_register_qpls(struct gve_priv *priv)
 	return 0;
 }
 
+static int gve_unregister_xdp_qpls(struct gve_priv *priv)
+{
+	int start_id;
+	int err;
+	int i;
+
+	start_id = gve_tx_qpl_id(priv, gve_xdp_tx_start_queue_id(priv));
+	for (i = start_id; i < start_id + gve_num_xdp_qpls(priv); i++) {
+		err = gve_adminq_unregister_page_list(priv, priv->qpls[i].id);
+		/* This failure will trigger a reset - no need to clean up */
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to unregister queue page list %d\n",
+				  priv->qpls[i].id);
+			return err;
+		}
+	}
+	return 0;
+}
+
 static int gve_unregister_qpls(struct gve_priv *priv)
 {
 	int start_id;
@@ -650,6 +699,27 @@ static int gve_unregister_qpls(struct gve_priv *priv)
 	return 0;
 }
 
+static int gve_create_xdp_rings(struct gve_priv *priv)
+{
+	int err;
+
+	err = gve_adminq_create_tx_queues(priv,
+					  gve_xdp_tx_start_queue_id(priv),
+					  priv->num_xdp_queues);
+	if (err) {
+		netif_err(priv, drv, priv->dev, "failed to create %d XDP tx queues\n",
+			  priv->num_xdp_queues);
+		/* This failure will trigger a reset - no need to clean
+		 * up
+		 */
+		return err;
+	}
+	netif_dbg(priv, drv, priv->dev, "created %d XDP tx queues\n",
+		  priv->num_xdp_queues);
+
+	return 0;
+}
+
 static int gve_create_rings(struct gve_priv *priv)
 {
 	int num_tx_queues = gve_num_tx_queues(priv);
@@ -699,6 +769,23 @@ static int gve_create_rings(struct gve_priv *priv)
 	return 0;
 }
 
+static void add_napi_init_xdp_sync_stats(struct gve_priv *priv,
+					 int (*napi_poll)(struct napi_struct *napi,
+							  int budget))
+{
+	int start_id = gve_xdp_tx_start_queue_id(priv);
+	int i;
+
+	/* Add xdp tx napi & init sync stats*/
+	for (i = start_id; i < start_id + priv->num_xdp_queues; i++) {
+		int ntfy_idx = gve_tx_idx_to_ntfy(priv, i);
+
+		u64_stats_init(&priv->tx[i].statss);
+		priv->tx[i].ntfy_id = ntfy_idx;
+		gve_add_napi(priv, ntfy_idx, napi_poll);
+	}
+}
+
 static void add_napi_init_sync_stats(struct gve_priv *priv,
 				     int (*napi_poll)(struct napi_struct *napi,
 						      int budget))
@@ -732,6 +819,23 @@ static void gve_tx_free_rings(struct gve_priv *priv, int start_id, int num_rings
 	}
 }
 
+static int gve_alloc_xdp_rings(struct gve_priv *priv)
+{
+	int start_id;
+	int err = 0;
+
+	if (!priv->num_xdp_queues)
+		return 0;
+
+	start_id = gve_xdp_tx_start_queue_id(priv);
+	err = gve_tx_alloc_rings(priv, start_id, priv->num_xdp_queues);
+	if (err)
+		return err;
+	add_napi_init_xdp_sync_stats(priv, gve_napi_poll);
+
+	return 0;
+}
+
 static int gve_alloc_rings(struct gve_priv *priv)
 {
 	int err;
@@ -782,6 +886,26 @@ free_tx:
 	return err;
 }
 
+static int gve_destroy_xdp_rings(struct gve_priv *priv)
+{
+	int start_id;
+	int err;
+
+	start_id = gve_xdp_tx_start_queue_id(priv);
+	err = gve_adminq_destroy_tx_queues(priv,
+					   start_id,
+					   priv->num_xdp_queues);
+	if (err) {
+		netif_err(priv, drv, priv->dev,
+			  "failed to destroy XDP queues\n");
+		/* This failure will trigger a reset - no need to clean up */
+		return err;
+	}
+	netif_dbg(priv, drv, priv->dev, "destroyed XDP queues\n");
+
+	return 0;
+}
+
 static int gve_destroy_rings(struct gve_priv *priv)
 {
 	int num_tx_queues = gve_num_tx_queues(priv);
@@ -814,6 +938,21 @@ static void gve_rx_free_rings(struct gve_priv *priv)
 		gve_rx_free_rings_dqo(priv);
 }
 
+static void gve_free_xdp_rings(struct gve_priv *priv)
+{
+	int ntfy_idx, start_id;
+	int i;
+
+	start_id = gve_xdp_tx_start_queue_id(priv);
+	if (priv->tx) {
+		for (i = start_id; i <  start_id + priv->num_xdp_queues; i++) {
+			ntfy_idx = gve_tx_idx_to_ntfy(priv, i);
+			gve_remove_napi(priv, ntfy_idx);
+		}
+		gve_tx_free_rings(priv, start_id, priv->num_xdp_queues);
+	}
+}
+
 static void gve_free_rings(struct gve_priv *priv)
 {
 	int num_tx_queues = gve_num_tx_queues(priv);
@@ -929,6 +1068,28 @@ free_pages:
 	priv->num_registered_pages -= qpl->num_entries;
 }
 
+static int gve_alloc_xdp_qpls(struct gve_priv *priv)
+{
+	int start_id;
+	int i, j;
+	int err;
+
+	start_id = gve_tx_qpl_id(priv, gve_xdp_tx_start_queue_id(priv));
+	for (i = start_id; i < start_id + gve_num_xdp_qpls(priv); i++) {
+		err = gve_alloc_queue_page_list(priv, i,
+						priv->tx_pages_per_qpl);
+		if (err)
+			goto free_qpls;
+	}
+
+	return 0;
+
+free_qpls:
+	for (j = start_id; j <= i; j++)
+		gve_free_queue_page_list(priv, j);
+	return err;
+}
+
 static int gve_alloc_qpls(struct gve_priv *priv)
 {
 	int max_queues = priv->tx_cfg.max_queues + priv->rx_cfg.max_queues;
@@ -978,6 +1139,16 @@ free_qpls:
 	return err;
 }
 
+static void gve_free_xdp_qpls(struct gve_priv *priv)
+{
+	int start_id;
+	int i;
+
+	start_id = gve_tx_qpl_id(priv, gve_xdp_tx_start_queue_id(priv));
+	for (i = start_id; i < start_id + gve_num_xdp_qpls(priv); i++)
+		gve_free_queue_page_list(priv, i);
+}
+
 static void gve_free_qpls(struct gve_priv *priv)
 {
 	int max_queues = priv->tx_cfg.max_queues + priv->rx_cfg.max_queues;
@@ -1011,11 +1182,64 @@ static int gve_reset_recovery(struct gve_priv *priv, bool was_up);
 static void gve_turndown(struct gve_priv *priv);
 static void gve_turnup(struct gve_priv *priv);
 
+static int gve_reg_xdp_info(struct gve_priv *priv, struct net_device *dev)
+{
+	struct napi_struct *napi;
+	struct gve_rx_ring *rx;
+	int err = 0;
+	int i, j;
+
+	if (!priv->num_xdp_queues)
+		return 0;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+		rx = &priv->rx[i];
+		napi = &priv->ntfy_blocks[rx->ntfy_id].napi;
+
+		err = xdp_rxq_info_reg(&rx->xdp_rxq, dev, i,
+				       napi->napi_id);
+		if (err)
+			goto err;
+		err = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq,
+						 MEM_TYPE_PAGE_SHARED, NULL);
+		if (err)
+			goto err;
+	}
+	return 0;
+
+err:
+	for (j = i; j >= 0; j--) {
+		rx = &priv->rx[j];
+		if (xdp_rxq_info_is_reg(&rx->xdp_rxq))
+			xdp_rxq_info_unreg(&rx->xdp_rxq);
+	}
+	return err;
+}
+
+static void gve_unreg_xdp_info(struct gve_priv *priv)
+{
+	int i;
+
+	if (!priv->num_xdp_queues)
+		return;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+		struct gve_rx_ring *rx = &priv->rx[i];
+
+		xdp_rxq_info_unreg(&rx->xdp_rxq);
+	}
+}
+
 static int gve_open(struct net_device *dev)
 {
 	struct gve_priv *priv = netdev_priv(dev);
 	int err;
 
+	if (priv->xdp_prog)
+		priv->num_xdp_queues = priv->rx_cfg.num_queues;
+	else
+		priv->num_xdp_queues = 0;
+
 	err = gve_alloc_qpls(priv);
 	if (err)
 		return err;
@@ -1031,6 +1255,10 @@ static int gve_open(struct net_device *dev)
 	if (err)
 		goto free_rings;
 
+	err = gve_reg_xdp_info(priv, dev);
+	if (err)
+		goto free_rings;
+
 	err = gve_register_qpls(priv);
 	if (err)
 		goto reset;
@@ -1095,6 +1323,7 @@ static int gve_close(struct net_device *dev)
 	}
 	del_timer_sync(&priv->stats_report_timer);
 
+	gve_unreg_xdp_info(priv);
 	gve_free_rings(priv);
 	gve_free_qpls(priv);
 	priv->interface_down_cnt++;
@@ -1111,6 +1340,167 @@ err:
 	return gve_reset_recovery(priv, false);
 }
 
+static int gve_remove_xdp_queues(struct gve_priv *priv)
+{
+	int err;
+
+	err = gve_destroy_xdp_rings(priv);
+	if (err)
+		return err;
+
+	err = gve_unregister_xdp_qpls(priv);
+	if (err)
+		return err;
+
+	gve_unreg_xdp_info(priv);
+	gve_free_xdp_rings(priv);
+	gve_free_xdp_qpls(priv);
+	priv->num_xdp_queues = 0;
+	return 0;
+}
+
+static int gve_add_xdp_queues(struct gve_priv *priv)
+{
+	int err;
+
+	priv->num_xdp_queues = priv->tx_cfg.num_queues;
+
+	err = gve_alloc_xdp_qpls(priv);
+	if (err)
+		goto err;
+
+	err = gve_alloc_xdp_rings(priv);
+	if (err)
+		goto free_xdp_qpls;
+
+	err = gve_reg_xdp_info(priv, priv->dev);
+	if (err)
+		goto free_xdp_rings;
+
+	err = gve_register_xdp_qpls(priv);
+	if (err)
+		goto free_xdp_rings;
+
+	err = gve_create_xdp_rings(priv);
+	if (err)
+		goto free_xdp_rings;
+
+	return 0;
+
+free_xdp_rings:
+	gve_free_xdp_rings(priv);
+free_xdp_qpls:
+	gve_free_xdp_qpls(priv);
+err:
+	priv->num_xdp_queues = 0;
+	return err;
+}
+
+static void gve_handle_link_status(struct gve_priv *priv, bool link_status)
+{
+	if (!gve_get_napi_enabled(priv))
+		return;
+
+	if (link_status == netif_carrier_ok(priv->dev))
+		return;
+
+	if (link_status) {
+		netdev_info(priv->dev, "Device link is up.\n");
+		netif_carrier_on(priv->dev);
+	} else {
+		netdev_info(priv->dev, "Device link is down.\n");
+		netif_carrier_off(priv->dev);
+	}
+}
+
+static int gve_set_xdp(struct gve_priv *priv, struct bpf_prog *prog,
+		       struct netlink_ext_ack *extack)
+{
+	struct bpf_prog *old_prog;
+	int err = 0;
+	u32 status;
+
+	old_prog = READ_ONCE(priv->xdp_prog);
+	if (!netif_carrier_ok(priv->dev)) {
+		WRITE_ONCE(priv->xdp_prog, prog);
+		if (old_prog)
+			bpf_prog_put(old_prog);
+		return 0;
+	}
+
+	gve_turndown(priv);
+	if (!old_prog && prog) {
+		// Allocate XDP TX queues if an XDP program is
+		// being installed
+		err = gve_add_xdp_queues(priv);
+		if (err)
+			goto out;
+	} else if (old_prog && !prog) {
+		// Remove XDP TX queues if an XDP program is
+		// being uninstalled
+		err = gve_remove_xdp_queues(priv);
+		if (err)
+			goto out;
+	}
+	WRITE_ONCE(priv->xdp_prog, prog);
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+out:
+	gve_turnup(priv);
+	status = ioread32be(&priv->reg_bar0->device_status);
+	gve_handle_link_status(priv, GVE_DEVICE_STATUS_LINK_STATUS_MASK & status);
+	return err;
+}
+
+static int verify_xdp_configuration(struct net_device *dev)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+
+	if (dev->features & NETIF_F_LRO) {
+		netdev_warn(dev, "XDP is not supported when LRO is on.\n");
+		return -EOPNOTSUPP;
+	}
+
+	if (priv->queue_format != GVE_GQI_QPL_FORMAT) {
+		netdev_warn(dev, "XDP is not supported in mode %d.\n",
+			    priv->queue_format);
+		return -EOPNOTSUPP;
+	}
+
+	if (dev->mtu > (PAGE_SIZE / 2) - sizeof(struct ethhdr) - GVE_RX_PAD) {
+		netdev_warn(dev, "XDP is not supported for mtu %d.\n",
+			    dev->mtu);
+		return -EOPNOTSUPP;
+	}
+
+	if (priv->rx_cfg.num_queues != priv->tx_cfg.num_queues ||
+	    (2 * priv->tx_cfg.num_queues > priv->tx_cfg.max_queues)) {
+		netdev_warn(dev, "XDP load failed: The number of configured RX queues %d should be equal to the number of configured TX queues %d and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues %d",
+			    priv->rx_cfg.num_queues,
+			    priv->tx_cfg.num_queues,
+			    priv->tx_cfg.max_queues);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int gve_xdp(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	int err;
+
+	err = verify_xdp_configuration(dev);
+	if (err)
+		return err;
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return gve_set_xdp(priv, xdp->prog, xdp->extack);
+	default:
+		return -EINVAL;
+	}
+}
+
 int gve_adjust_queues(struct gve_priv *priv,
 		      struct gve_queue_config new_rx_config,
 		      struct gve_queue_config new_tx_config)
@@ -1305,6 +1695,7 @@ static const struct net_device_ops gve_netdev_ops = {
 	.ndo_get_stats64	=	gve_get_stats,
 	.ndo_tx_timeout         =       gve_tx_timeout,
 	.ndo_set_features	=	gve_set_features,
+	.ndo_bpf		=	gve_xdp,
 };
 
 static void gve_handle_status(struct gve_priv *priv, u32 status)
@@ -1411,23 +1802,6 @@ void gve_handle_report_stats(struct gve_priv *priv)
 	}
 }
 
-static void gve_handle_link_status(struct gve_priv *priv, bool link_status)
-{
-	if (!gve_get_napi_enabled(priv))
-		return;
-
-	if (link_status == netif_carrier_ok(priv->dev))
-		return;
-
-	if (link_status) {
-		netdev_info(priv->dev, "Device link is up.\n");
-		netif_carrier_on(priv->dev);
-	} else {
-		netdev_info(priv->dev, "Device link is down.\n");
-		netif_carrier_off(priv->dev);
-	}
-}
-
 /* Handle NIC status register changes, reset requests and report stats */
 static void gve_service_task(struct work_struct *work)
 {
@@ -1441,6 +1815,15 @@ static void gve_service_task(struct work_struct *work)
 	gve_handle_link_status(priv, GVE_DEVICE_STATUS_LINK_STATUS_MASK & status);
 }
 
+static void gve_set_netdev_xdp_features(struct gve_priv *priv)
+{
+	if (priv->queue_format == GVE_GQI_QPL_FORMAT) {
+		priv->dev->xdp_features = NETDEV_XDP_ACT_BASIC;
+	} else {
+		priv->dev->xdp_features = 0;
+	}
+}
+
 static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device)
 {
 	int num_ntfy;
@@ -1519,6 +1902,7 @@ static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device)
 	}
 
 setup_device:
+	gve_set_netdev_xdp_features(priv);
 	err = gve_setup_device_resources(priv);
 	if (!err)
 		return 0;
diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c
index 051a15e4f1af..3241f6ea29be 100644
--- a/drivers/net/ethernet/google/gve/gve_rx.c
+++ b/drivers/net/ethernet/google/gve/gve_rx.c
@@ -8,6 +8,8 @@
 #include "gve_adminq.h"
 #include "gve_utils.h"
 #include <linux/etherdevice.h>
+#include <linux/filter.h>
+#include <net/xdp.h>
 
 static void gve_rx_free_buffer(struct device *dev,
 			       struct gve_rx_slot_page_info *page_info,
@@ -591,6 +593,43 @@ static struct sk_buff *gve_rx_skb(struct gve_priv *priv, struct gve_rx_ring *rx,
 	return skb;
 }
 
+static void gve_xdp_done(struct gve_priv *priv, struct gve_rx_ring *rx,
+			 struct xdp_buff *xdp, struct bpf_prog *xprog,
+			 int xdp_act)
+{
+	struct gve_tx_ring *tx;
+	int tx_qid;
+	int err;
+
+	switch (xdp_act) {
+	case XDP_ABORTED:
+	case XDP_DROP:
+	default:
+		break;
+	case XDP_TX:
+		tx_qid = gve_xdp_tx_queue_id(priv, rx->q_num);
+		tx = &priv->tx[tx_qid];
+		err = gve_xdp_xmit_one(priv, tx, xdp->data,
+				       xdp->data_end - xdp->data);
+
+		if (unlikely(err)) {
+			u64_stats_update_begin(&rx->statss);
+			rx->xdp_tx_errors++;
+			u64_stats_update_end(&rx->statss);
+		}
+		break;
+	case XDP_REDIRECT:
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_redirect_errors++;
+		u64_stats_update_end(&rx->statss);
+		break;
+	}
+	u64_stats_update_begin(&rx->statss);
+	if ((u32)xdp_act < GVE_XDP_ACTIONS)
+		rx->xdp_actions[xdp_act]++;
+	u64_stats_update_end(&rx->statss);
+}
+
 #define GVE_PKTCONT_BIT_IS_SET(x) (GVE_RXF_PKT_CONT & (x))
 static void gve_rx(struct gve_rx_ring *rx, netdev_features_t feat,
 		   struct gve_rx_desc *desc, u32 idx,
@@ -603,9 +642,12 @@ static void gve_rx(struct gve_rx_ring *rx, netdev_features_t feat,
 	union gve_rx_data_slot *data_slot;
 	struct gve_priv *priv = rx->gve;
 	struct sk_buff *skb = NULL;
+	struct bpf_prog *xprog;
+	struct xdp_buff xdp;
 	dma_addr_t page_bus;
 	void *va;
 
+	u16 len = frag_size;
 	struct napi_struct *napi = &priv->ntfy_blocks[rx->ntfy_id].napi;
 	bool is_first_frag = ctx->frag_cnt == 0;
 
@@ -645,9 +687,35 @@ static void gve_rx(struct gve_rx_ring *rx, netdev_features_t feat,
 	dma_sync_single_for_cpu(&priv->pdev->dev, page_bus,
 				PAGE_SIZE, DMA_FROM_DEVICE);
 	page_info->pad = is_first_frag ? GVE_RX_PAD : 0;
+	len -= page_info->pad;
 	frag_size -= page_info->pad;
 
-	skb = gve_rx_skb(priv, rx, page_info, napi, frag_size,
+	xprog = READ_ONCE(priv->xdp_prog);
+	if (xprog && is_only_frag) {
+		void *old_data;
+		int xdp_act;
+
+		xdp_init_buff(&xdp, rx->packet_buffer_size, &rx->xdp_rxq);
+		xdp_prepare_buff(&xdp, page_info->page_address +
+				 page_info->page_offset, GVE_RX_PAD,
+				 len, false);
+		old_data = xdp.data;
+		xdp_act = bpf_prog_run_xdp(xprog, &xdp);
+		if (xdp_act != XDP_PASS) {
+			gve_xdp_done(priv, rx, &xdp, xprog, xdp_act);
+			ctx->total_size += frag_size;
+			goto finish_ok_pkt;
+		}
+
+		page_info->pad += xdp.data - old_data;
+		len = xdp.data_end - xdp.data;
+
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_actions[XDP_PASS]++;
+		u64_stats_update_end(&rx->statss);
+	}
+
+	skb = gve_rx_skb(priv, rx, page_info, napi, len,
 			 data_slot, is_only_frag);
 	if (!skb) {
 		u64_stats_update_begin(&rx->statss);
@@ -773,6 +841,7 @@ static bool gve_rx_refill_buffers(struct gve_priv *priv, struct gve_rx_ring *rx)
 static int gve_clean_rx_done(struct gve_rx_ring *rx, int budget,
 			     netdev_features_t feat)
 {
+	u64 xdp_txs = rx->xdp_actions[XDP_TX];
 	struct gve_rx_ctx *ctx = &rx->ctx;
 	struct gve_priv *priv = rx->gve;
 	struct gve_rx_cnts cnts = {0};
@@ -820,6 +889,9 @@ static int gve_clean_rx_done(struct gve_rx_ring *rx, int budget,
 		u64_stats_update_end(&rx->statss);
 	}
 
+	if (xdp_txs != rx->xdp_actions[XDP_TX])
+		gve_xdp_tx_flush(priv, rx->q_num);
+
 	/* restock ring slots */
 	if (!rx->data.raw_addressing) {
 		/* In QPL mode buffs are refilled as the desc are processed */
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
index e24e73e74e33..3e96ee7537ce 100644
--- a/drivers/net/ethernet/google/gve/gve_tx.c
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -19,6 +19,14 @@ static inline void gve_tx_put_doorbell(struct gve_priv *priv,
 	iowrite32be(val, &priv->db_bar2[be32_to_cpu(q_resources->db_index)]);
 }
 
+void gve_xdp_tx_flush(struct gve_priv *priv, u32 xdp_qid)
+{
+	u32 tx_qid = gve_xdp_tx_queue_id(priv, xdp_qid);
+	struct gve_tx_ring *tx = &priv->tx[tx_qid];
+
+	gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+}
+
 /* gvnic can only transmit from a Registered Segment.
  * We copy skb payloads into the registered segment before writing Tx
  * descriptors and ringing the Tx doorbell.
@@ -132,6 +140,50 @@ static void gve_tx_free_fifo(struct gve_tx_fifo *fifo, size_t bytes)
 	atomic_add(bytes, &fifo->available);
 }
 
+static size_t gve_tx_clear_buffer_state(struct gve_tx_buffer_state *info)
+{
+	size_t space_freed = 0;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(info->iov); i++) {
+		space_freed += info->iov[i].iov_len + info->iov[i].iov_padding;
+		info->iov[i].iov_len = 0;
+		info->iov[i].iov_padding = 0;
+	}
+	return space_freed;
+}
+
+static int gve_clean_xdp_done(struct gve_priv *priv, struct gve_tx_ring *tx,
+			      u32 to_do)
+{
+	struct gve_tx_buffer_state *info;
+	u32 clean_end = tx->done + to_do;
+	u64 pkts = 0, bytes = 0;
+	size_t space_freed = 0;
+	u32 idx;
+
+	for (; tx->done < clean_end; tx->done++) {
+		idx = tx->done & tx->mask;
+		info = &tx->info[idx];
+
+		if (unlikely(!info->xdp.size))
+			continue;
+
+		bytes += info->xdp.size;
+		pkts++;
+
+		info->xdp.size = 0;
+		space_freed += gve_tx_clear_buffer_state(info);
+	}
+
+	gve_tx_free_fifo(&tx->tx_fifo, space_freed);
+	u64_stats_update_begin(&tx->statss);
+	tx->bytes_done += bytes;
+	tx->pkt_done += pkts;
+	u64_stats_update_end(&tx->statss);
+	return pkts;
+}
+
 static int gve_clean_tx_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 			     u32 to_do, bool try_to_wake);
 
@@ -144,8 +196,12 @@ static void gve_tx_free_ring(struct gve_priv *priv, int idx)
 
 	gve_tx_remove_from_block(priv, idx);
 	slots = tx->mask + 1;
-	gve_clean_tx_done(priv, tx, priv->tx_desc_cnt, false);
-	netdev_tx_reset_queue(tx->netdev_txq);
+	if (tx->q_num < priv->tx_cfg.num_queues) {
+		gve_clean_tx_done(priv, tx, priv->tx_desc_cnt, false);
+		netdev_tx_reset_queue(tx->netdev_txq);
+	} else {
+		gve_clean_xdp_done(priv, tx, priv->tx_desc_cnt);
+	}
 
 	dma_free_coherent(hdev, sizeof(*tx->q_resources),
 			  tx->q_resources, tx->q_resources_bus);
@@ -213,7 +269,8 @@ static int gve_tx_alloc_ring(struct gve_priv *priv, int idx)
 
 	netif_dbg(priv, drv, priv->dev, "tx[%d]->bus=%lx\n", idx,
 		  (unsigned long)tx->bus);
-	tx->netdev_txq = netdev_get_tx_queue(priv->dev, idx);
+	if (idx < priv->tx_cfg.num_queues)
+		tx->netdev_txq = netdev_get_tx_queue(priv->dev, idx);
 	gve_tx_add_to_block(priv, idx);
 
 	return 0;
@@ -657,6 +714,65 @@ netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 }
 
+static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
+			   void *data, int len)
+{
+	int pad, nfrags, ndescs, iovi, offset;
+	struct gve_tx_buffer_state *info;
+	u32 reqi = tx->req;
+
+	pad = gve_tx_fifo_pad_alloc_one_frag(&tx->tx_fifo, len);
+	if (pad >= GVE_TX_MAX_HEADER_SIZE)
+		pad = 0;
+	info = &tx->info[reqi & tx->mask];
+	info->xdp.size = len;
+
+	nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, pad + len,
+				   &info->iov[0]);
+	iovi = pad > 0;
+	ndescs = nfrags - iovi;
+	offset = 0;
+
+	while (iovi < nfrags) {
+		if (!offset)
+			gve_tx_fill_pkt_desc(&tx->desc[reqi & tx->mask], 0,
+					     CHECKSUM_NONE, false, 0, ndescs,
+					     info->iov[iovi].iov_len,
+					     info->iov[iovi].iov_offset, len);
+		else
+			gve_tx_fill_seg_desc(&tx->desc[reqi & tx->mask],
+					     0, 0, false, false,
+					     info->iov[iovi].iov_len,
+					     info->iov[iovi].iov_offset);
+
+		memcpy(tx->tx_fifo.base + info->iov[iovi].iov_offset,
+		       data + offset, info->iov[iovi].iov_len);
+		gve_dma_sync_for_device(&priv->pdev->dev,
+					tx->tx_fifo.qpl->page_buses,
+					info->iov[iovi].iov_offset,
+					info->iov[iovi].iov_len);
+		offset += info->iov[iovi].iov_len;
+		iovi++;
+		reqi++;
+	}
+
+	return ndescs;
+}
+
+int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
+		     void *data, int len)
+{
+	int nsegs;
+
+	if (!gve_can_tx(tx, len + GVE_TX_MAX_HEADER_SIZE - 1))
+		return -EBUSY;
+
+	nsegs = gve_tx_fill_xdp(priv, tx, data, len);
+	tx->req += nsegs;
+
+	return 0;
+}
+
 #define GVE_TX_START_THRESH	PAGE_SIZE
 
 static int gve_clean_tx_done(struct gve_priv *priv, struct gve_tx_ring *tx,
@@ -666,8 +782,8 @@ static int gve_clean_tx_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 	u64 pkts = 0, bytes = 0;
 	size_t space_freed = 0;
 	struct sk_buff *skb;
-	int i, j;
 	u32 idx;
+	int j;
 
 	for (j = 0; j < to_do; j++) {
 		idx = tx->done & tx->mask;
@@ -689,12 +805,7 @@ static int gve_clean_tx_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 			dev_consume_skb_any(skb);
 			if (tx->raw_addressing)
 				continue;
-			/* FIFO free */
-			for (i = 0; i < ARRAY_SIZE(info->iov); i++) {
-				space_freed += info->iov[i].iov_len + info->iov[i].iov_padding;
-				info->iov[i].iov_len = 0;
-				info->iov[i].iov_padding = 0;
-			}
+			space_freed += gve_tx_clear_buffer_state(info);
 		}
 	}
 
@@ -729,6 +840,24 @@ u32 gve_tx_load_event_counter(struct gve_priv *priv,
 	return be32_to_cpu(counter);
 }
 
+bool gve_xdp_poll(struct gve_notify_block *block, int budget)
+{
+	struct gve_priv *priv = block->priv;
+	struct gve_tx_ring *tx = block->tx;
+	u32 nic_done;
+	u32 to_do;
+
+	/* If budget is 0, do all the work */
+	if (budget == 0)
+		budget = INT_MAX;
+
+	/* Find out how much work there is to be done */
+	nic_done = gve_tx_load_event_counter(priv, tx);
+	to_do = min_t(u32, (nic_done - tx->done), budget);
+	gve_clean_xdp_done(priv, tx, to_do);
+	return nic_done != tx->done;
+}
+
 bool gve_tx_poll(struct gve_notify_block *block, int budget)
 {
 	struct gve_priv *priv = block->priv;
-- 
cgit v1.2.3-70-g09d2


From 39a7f4aa3e4a7947614cf1d5c27abba3300adb1e Mon Sep 17 00:00:00 2001
From: Praveen Kaligineedi <pkaligineedi@google.com>
Date: Wed, 15 Mar 2023 16:33:11 -0700
Subject: gve: Add XDP REDIRECT support for GQI-QPL format

This patch contains the following changes:
1) Support for XDP REDIRECT action on rx
2) ndo_xdp_xmit callback support

In GQI-QPL queue format, the driver needs to allocate a fixed size
memory, the size specified by vNIC device, for RX/TX and register this
memory as a bounce buffer with the vNIC device when a queue is created.
The number of pages in the bounce buffer is limited and the pages need to
be made available to the vNIC by copying the RX data out to prevent
head-of-line blocking. The XDP_REDIRECT packets are therefore immediately
copied to a newly allocated page.

Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/google/gve/gve.h         | 15 +++++++--
 drivers/net/ethernet/google/gve/gve_ethtool.c | 26 ++++++++++-----
 drivers/net/ethernet/google/gve/gve_main.c    | 19 +++++++++++
 drivers/net/ethernet/google/gve/gve_rx.c      | 47 +++++++++++++++++++++++---
 drivers/net/ethernet/google/gve/gve_tx.c      | 48 +++++++++++++++++++++++++--
 5 files changed, 138 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index 8d5234d4ba67..a3b2aec2c575 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -236,6 +236,7 @@ struct gve_rx_ring {
 	u64 rx_frag_alloc_cnt; /* free-running count of rx page allocations */
 	u64 xdp_tx_errors;
 	u64 xdp_redirect_errors;
+	u64 xdp_alloc_fails;
 	u64 xdp_actions[GVE_XDP_ACTIONS];
 	u32 q_num; /* queue index */
 	u32 ntfy_id; /* notification block index */
@@ -247,6 +248,7 @@ struct gve_rx_ring {
 
 	/* XDP stuff */
 	struct xdp_rxq_info xdp_rxq;
+	struct page_frag_cache page_cache; /* Page cache to allocate XDP frames */
 };
 
 /* A TX desc ring entry */
@@ -267,7 +269,10 @@ struct gve_tx_iovec {
  * ring entry but only used for a pkt_desc not a seg_desc
  */
 struct gve_tx_buffer_state {
-	struct sk_buff *skb; /* skb for this pkt */
+	union {
+		struct sk_buff *skb; /* skb for this pkt */
+		struct xdp_frame *xdp_frame; /* xdp_frame */
+	};
 	struct {
 		u16 size; /* size of xmitted xdp pkt */
 	} xdp;
@@ -385,6 +390,8 @@ struct gve_tx_ring {
 		struct {
 			/* Spinlock for when cleanup in progress */
 			spinlock_t clean_lock;
+			/* Spinlock for XDP tx traffic */
+			spinlock_t xdp_lock;
 		};
 
 		/* DQO fields. */
@@ -462,6 +469,8 @@ struct gve_tx_ring {
 	dma_addr_t q_resources_bus; /* dma address of the queue resources */
 	dma_addr_t complq_bus_dqo; /* dma address of the dqo.compl_ring */
 	struct u64_stats_sync statss; /* sync stats for 32bit archs */
+	u64 xdp_xmit;
+	u64 xdp_xmit_errors;
 } ____cacheline_aligned;
 
 /* Wraps the info for one irq including the napi struct and the queues
@@ -919,8 +928,10 @@ void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma,
 		   enum dma_data_direction);
 /* tx handling */
 netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev);
+int gve_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+		 u32 flags);
 int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
-		     void *data, int len);
+		     void *data, int len, void *frame_p);
 void gve_xdp_tx_flush(struct gve_priv *priv, u32 xdp_qid);
 bool gve_tx_poll(struct gve_notify_block *block, int budget);
 bool gve_xdp_poll(struct gve_notify_block *block, int budget);
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
index 067b393ccf9d..23db0f3534a8 100644
--- a/drivers/net/ethernet/google/gve/gve_ethtool.c
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -56,13 +56,14 @@ static const char gve_gstrings_rx_stats[][ETH_GSTRING_LEN] = {
 	"rx_drops_packet_over_mru[%u]", "rx_drops_invalid_checksum[%u]",
 	"rx_xdp_aborted[%u]", "rx_xdp_drop[%u]", "rx_xdp_pass[%u]",
 	"rx_xdp_tx[%u]", "rx_xdp_redirect[%u]",
-	"rx_xdp_tx_errors[%u]", "rx_xdp_redirect_errors[%u]",
+	"rx_xdp_tx_errors[%u]", "rx_xdp_redirect_errors[%u]", "rx_xdp_alloc_fails[%u]",
 };
 
 static const char gve_gstrings_tx_stats[][ETH_GSTRING_LEN] = {
 	"tx_posted_desc[%u]", "tx_completed_desc[%u]", "tx_consumed_desc[%u]", "tx_bytes[%u]",
 	"tx_wake[%u]", "tx_stop[%u]", "tx_event_counter[%u]",
 	"tx_dma_mapping_error[%u]",
+	"tx_xdp_xmit[%u]", "tx_xdp_xmit_errors[%u]"
 };
 
 static const char gve_gstrings_adminq_stats[][ETH_GSTRING_LEN] = {
@@ -313,9 +314,10 @@ gve_get_ethtool_stats(struct net_device *netdev,
 					data[i + j] = rx->xdp_actions[j];
 				data[i + j++] = rx->xdp_tx_errors;
 				data[i + j++] = rx->xdp_redirect_errors;
+				data[i + j++] = rx->xdp_alloc_fails;
 			} while (u64_stats_fetch_retry(&priv->rx[ring].statss,
 						       start));
-			i += GVE_XDP_ACTIONS + 2; /* XDP rx counters */
+			i += GVE_XDP_ACTIONS + 3; /* XDP rx counters */
 		}
 	} else {
 		i += priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS;
@@ -371,13 +373,21 @@ gve_get_ethtool_stats(struct net_device *netdev,
 			if (skip_nic_stats) {
 				/* skip NIC tx stats */
 				i += NIC_TX_STATS_REPORT_NUM;
-				continue;
-			}
-			for (j = 0; j < NIC_TX_STATS_REPORT_NUM; j++) {
-				u64 value =
-				be64_to_cpu(report_stats[tx_qid_to_stats_idx[ring] + j].value);
-				data[i++] = value;
+			} else {
+				stats_idx = tx_qid_to_stats_idx[ring];
+				for (j = 0; j < NIC_TX_STATS_REPORT_NUM; j++) {
+					u64 value =
+						be64_to_cpu(report_stats[stats_idx + j].value);
+					data[i++] = value;
+				}
 			}
+			do {
+				start = u64_stats_fetch_begin(&priv->tx[ring].statss);
+				data[i] = tx->xdp_xmit;
+				data[i + 1] = tx->xdp_xmit_errors;
+			} while (u64_stats_fetch_retry(&priv->tx[ring].statss,
+						       start));
+			i += 2; /* XDP tx counters */
 		}
 	} else {
 		i += num_tx_queues * NUM_GVE_TX_CNTS;
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index f49398857921..2e8ea4dd71e8 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -1230,6 +1230,21 @@ static void gve_unreg_xdp_info(struct gve_priv *priv)
 	}
 }
 
+static void gve_drain_page_cache(struct gve_priv *priv)
+{
+	struct page_frag_cache *nc;
+	int i;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+		nc = &priv->rx[i].page_cache;
+		if (nc->va) {
+			__page_frag_cache_drain(virt_to_page(nc->va),
+						nc->pagecnt_bias);
+			nc->va = NULL;
+		}
+	}
+}
+
 static int gve_open(struct net_device *dev)
 {
 	struct gve_priv *priv = netdev_priv(dev);
@@ -1313,6 +1328,7 @@ static int gve_close(struct net_device *dev)
 	netif_carrier_off(dev);
 	if (gve_get_device_rings_ok(priv)) {
 		gve_turndown(priv);
+		gve_drain_page_cache(priv);
 		err = gve_destroy_rings(priv);
 		if (err)
 			goto err;
@@ -1696,6 +1712,7 @@ static const struct net_device_ops gve_netdev_ops = {
 	.ndo_tx_timeout         =       gve_tx_timeout,
 	.ndo_set_features	=	gve_set_features,
 	.ndo_bpf		=	gve_xdp,
+	.ndo_xdp_xmit		=	gve_xdp_xmit,
 };
 
 static void gve_handle_status(struct gve_priv *priv, u32 status)
@@ -1819,6 +1836,8 @@ static void gve_set_netdev_xdp_features(struct gve_priv *priv)
 {
 	if (priv->queue_format == GVE_GQI_QPL_FORMAT) {
 		priv->dev->xdp_features = NETDEV_XDP_ACT_BASIC;
+		priv->dev->xdp_features |= NETDEV_XDP_ACT_REDIRECT;
+		priv->dev->xdp_features |= NETDEV_XDP_ACT_NDO_XMIT;
 	} else {
 		priv->dev->xdp_features = 0;
 	}
diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c
index 3241f6ea29be..ed4b5a540e6d 100644
--- a/drivers/net/ethernet/google/gve/gve_rx.c
+++ b/drivers/net/ethernet/google/gve/gve_rx.c
@@ -593,6 +593,35 @@ static struct sk_buff *gve_rx_skb(struct gve_priv *priv, struct gve_rx_ring *rx,
 	return skb;
 }
 
+static int gve_xdp_redirect(struct net_device *dev, struct gve_rx_ring *rx,
+			    struct xdp_buff *orig, struct bpf_prog *xdp_prog)
+{
+	int total_len, len = orig->data_end - orig->data;
+	int headroom = XDP_PACKET_HEADROOM;
+	struct xdp_buff new;
+	void *frame;
+	int err;
+
+	total_len = headroom + SKB_DATA_ALIGN(len) +
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	frame = page_frag_alloc(&rx->page_cache, total_len, GFP_ATOMIC);
+	if (!frame) {
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_alloc_fails++;
+		u64_stats_update_end(&rx->statss);
+		return -ENOMEM;
+	}
+	xdp_init_buff(&new, total_len, &rx->xdp_rxq);
+	xdp_prepare_buff(&new, frame, headroom, len, false);
+	memcpy(new.data, orig->data, len);
+
+	err = xdp_do_redirect(dev, &new, xdp_prog);
+	if (err)
+		page_frag_free(frame);
+
+	return err;
+}
+
 static void gve_xdp_done(struct gve_priv *priv, struct gve_rx_ring *rx,
 			 struct xdp_buff *xdp, struct bpf_prog *xprog,
 			 int xdp_act)
@@ -609,8 +638,10 @@ static void gve_xdp_done(struct gve_priv *priv, struct gve_rx_ring *rx,
 	case XDP_TX:
 		tx_qid = gve_xdp_tx_queue_id(priv, rx->q_num);
 		tx = &priv->tx[tx_qid];
+		spin_lock(&tx->xdp_lock);
 		err = gve_xdp_xmit_one(priv, tx, xdp->data,
-				       xdp->data_end - xdp->data);
+				       xdp->data_end - xdp->data, NULL);
+		spin_unlock(&tx->xdp_lock);
 
 		if (unlikely(err)) {
 			u64_stats_update_begin(&rx->statss);
@@ -619,9 +650,13 @@ static void gve_xdp_done(struct gve_priv *priv, struct gve_rx_ring *rx,
 		}
 		break;
 	case XDP_REDIRECT:
-		u64_stats_update_begin(&rx->statss);
-		rx->xdp_redirect_errors++;
-		u64_stats_update_end(&rx->statss);
+		err = gve_xdp_redirect(priv->dev, rx, xdp, xprog);
+
+		if (unlikely(err)) {
+			u64_stats_update_begin(&rx->statss);
+			rx->xdp_redirect_errors++;
+			u64_stats_update_end(&rx->statss);
+		}
 		break;
 	}
 	u64_stats_update_begin(&rx->statss);
@@ -841,6 +876,7 @@ static bool gve_rx_refill_buffers(struct gve_priv *priv, struct gve_rx_ring *rx)
 static int gve_clean_rx_done(struct gve_rx_ring *rx, int budget,
 			     netdev_features_t feat)
 {
+	u64 xdp_redirects = rx->xdp_actions[XDP_REDIRECT];
 	u64 xdp_txs = rx->xdp_actions[XDP_TX];
 	struct gve_rx_ctx *ctx = &rx->ctx;
 	struct gve_priv *priv = rx->gve;
@@ -892,6 +928,9 @@ static int gve_clean_rx_done(struct gve_rx_ring *rx, int budget,
 	if (xdp_txs != rx->xdp_actions[XDP_TX])
 		gve_xdp_tx_flush(priv, rx->q_num);
 
+	if (xdp_redirects != rx->xdp_actions[XDP_REDIRECT])
+		xdp_do_flush();
+
 	/* restock ring slots */
 	if (!rx->data.raw_addressing) {
 		/* In QPL mode buffs are refilled as the desc are processed */
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
index 3e96ee7537ce..d928c3c79618 100644
--- a/drivers/net/ethernet/google/gve/gve_tx.c
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -173,6 +173,10 @@ static int gve_clean_xdp_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 		pkts++;
 
 		info->xdp.size = 0;
+		if (info->xdp_frame) {
+			xdp_return_frame(info->xdp_frame);
+			info->xdp_frame = NULL;
+		}
 		space_freed += gve_tx_clear_buffer_state(info);
 	}
 
@@ -233,6 +237,7 @@ static int gve_tx_alloc_ring(struct gve_priv *priv, int idx)
 	/* Make sure everything is zeroed to start */
 	memset(tx, 0, sizeof(*tx));
 	spin_lock_init(&tx->clean_lock);
+	spin_lock_init(&tx->xdp_lock);
 	tx->q_num = idx;
 
 	tx->mask = slots - 1;
@@ -715,7 +720,7 @@ netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev)
 }
 
 static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
-			   void *data, int len)
+			   void *data, int len, void *frame_p)
 {
 	int pad, nfrags, ndescs, iovi, offset;
 	struct gve_tx_buffer_state *info;
@@ -725,6 +730,7 @@ static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
 	if (pad >= GVE_TX_MAX_HEADER_SIZE)
 		pad = 0;
 	info = &tx->info[reqi & tx->mask];
+	info->xdp_frame = frame_p;
 	info->xdp.size = len;
 
 	nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, pad + len,
@@ -759,15 +765,51 @@ static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
 	return ndescs;
 }
 
+int gve_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+		 u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct gve_tx_ring *tx;
+	int i, err = 0, qid;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	qid = gve_xdp_tx_queue_id(priv,
+				  smp_processor_id() % priv->num_xdp_queues);
+
+	tx = &priv->tx[qid];
+
+	spin_lock(&tx->xdp_lock);
+	for (i = 0; i < n; i++) {
+		err = gve_xdp_xmit_one(priv, tx, frames[i]->data,
+				       frames[i]->len, frames[i]);
+		if (err)
+			break;
+	}
+
+	if (flags & XDP_XMIT_FLUSH)
+		gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+
+	spin_unlock(&tx->xdp_lock);
+
+	u64_stats_update_begin(&tx->statss);
+	tx->xdp_xmit += n;
+	tx->xdp_xmit_errors += n - i;
+	u64_stats_update_end(&tx->statss);
+
+	return i ? i : err;
+}
+
 int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
-		     void *data, int len)
+		     void *data, int len, void *frame_p)
 {
 	int nsegs;
 
 	if (!gve_can_tx(tx, len + GVE_TX_MAX_HEADER_SIZE - 1))
 		return -EBUSY;
 
-	nsegs = gve_tx_fill_xdp(priv, tx, data, len);
+	nsegs = gve_tx_fill_xdp(priv, tx, data, len, frame_p);
 	tx->req += nsegs;
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From fd8e40321a12391e6f554cc637d0c4b6109682a9 Mon Sep 17 00:00:00 2001
From: Praveen Kaligineedi <pkaligineedi@google.com>
Date: Wed, 15 Mar 2023 16:33:12 -0700
Subject: gve: Add AF_XDP zero-copy support for GQI-QPL format

Adding AF_XDP zero-copy support.

Note: Although these changes support AF_XDP socket in zero-copy
mode, there is still a copy happening within the driver between
XSK buffer pool and QPL bounce buffers in GQI-QPL format.
In GQI-QPL queue format, the driver needs to allocate a fixed size
memory, the size specified by vNIC device, for RX/TX and register this
memory as a bounce buffer with the vNIC device when a queue is
created. The number of pages in the bounce buffer is limited and the
pages need to be made available to the vNIC by copying the RX data out
to prevent head-of-line blocking. Therefore, we cannot pass the XSK
buffer pool to the vNIC.

The number of copies on RX path from the bounce buffer to XSK buffer is 2
for AF_XDP copy mode (bounce buffer -> allocated page frag -> XSK buffer)
and 1 for AF_XDP zero-copy mode (bounce buffer -> XSK buffer).

This patch contains the following changes:
1) Enable and disable XSK buffer pool
2) Copy XDP packets from QPL bounce buffers to XSK buffer on rx
3) Copy XDP packets from XSK buffer to QPL bounce buffers and
   ring the doorbell as part of XDP TX napi poll
4) ndo_xsk_wakeup callback support

Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/google/gve/gve.h         |   7 ++
 drivers/net/ethernet/google/gve/gve_ethtool.c |  14 ++-
 drivers/net/ethernet/google/gve/gve_main.c    | 174 +++++++++++++++++++++++++-
 drivers/net/ethernet/google/gve/gve_rx.c      |  30 +++++
 drivers/net/ethernet/google/gve/gve_tx.c      |  58 ++++++++-
 5 files changed, 274 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index a3b2aec2c575..e214b51d3c8b 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -248,6 +248,8 @@ struct gve_rx_ring {
 
 	/* XDP stuff */
 	struct xdp_rxq_info xdp_rxq;
+	struct xdp_rxq_info xsk_rxq;
+	struct xsk_buff_pool *xsk_pool;
 	struct page_frag_cache page_cache; /* Page cache to allocate XDP frames */
 };
 
@@ -275,6 +277,7 @@ struct gve_tx_buffer_state {
 	};
 	struct {
 		u16 size; /* size of xmitted xdp pkt */
+		u8 is_xsk; /* xsk buff */
 	} xdp;
 	union {
 		struct gve_tx_iovec iov[GVE_TX_MAX_IOVEC]; /* segments of this pkt */
@@ -469,6 +472,10 @@ struct gve_tx_ring {
 	dma_addr_t q_resources_bus; /* dma address of the queue resources */
 	dma_addr_t complq_bus_dqo; /* dma address of the dqo.compl_ring */
 	struct u64_stats_sync statss; /* sync stats for 32bit archs */
+	struct xsk_buff_pool *xsk_pool;
+	u32 xdp_xsk_wakeup;
+	u32 xdp_xsk_done;
+	u64 xdp_xsk_sent;
 	u64 xdp_xmit;
 	u64 xdp_xmit_errors;
 } ____cacheline_aligned;
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
index 23db0f3534a8..b18804e934d3 100644
--- a/drivers/net/ethernet/google/gve/gve_ethtool.c
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -62,8 +62,8 @@ static const char gve_gstrings_rx_stats[][ETH_GSTRING_LEN] = {
 static const char gve_gstrings_tx_stats[][ETH_GSTRING_LEN] = {
 	"tx_posted_desc[%u]", "tx_completed_desc[%u]", "tx_consumed_desc[%u]", "tx_bytes[%u]",
 	"tx_wake[%u]", "tx_stop[%u]", "tx_event_counter[%u]",
-	"tx_dma_mapping_error[%u]",
-	"tx_xdp_xmit[%u]", "tx_xdp_xmit_errors[%u]"
+	"tx_dma_mapping_error[%u]", "tx_xsk_wakeup[%u]",
+	"tx_xsk_done[%u]", "tx_xsk_sent[%u]", "tx_xdp_xmit[%u]", "tx_xdp_xmit_errors[%u]"
 };
 
 static const char gve_gstrings_adminq_stats[][ETH_GSTRING_LEN] = {
@@ -381,13 +381,17 @@ gve_get_ethtool_stats(struct net_device *netdev,
 					data[i++] = value;
 				}
 			}
+			/* XDP xsk counters */
+			data[i++] = tx->xdp_xsk_wakeup;
+			data[i++] = tx->xdp_xsk_done;
 			do {
 				start = u64_stats_fetch_begin(&priv->tx[ring].statss);
-				data[i] = tx->xdp_xmit;
-				data[i + 1] = tx->xdp_xmit_errors;
+				data[i] = tx->xdp_xsk_sent;
+				data[i + 1] = tx->xdp_xmit;
+				data[i + 2] = tx->xdp_xmit_errors;
 			} while (u64_stats_fetch_retry(&priv->tx[ring].statss,
 						       start));
-			i += 2; /* XDP tx counters */
+			i += 3; /* XDP tx counters */
 		}
 	} else {
 		i += num_tx_queues * NUM_GVE_TX_CNTS;
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index 2e8ea4dd71e8..57ce74315eba 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -17,6 +17,7 @@
 #include <linux/utsname.h>
 #include <linux/version.h>
 #include <net/sch_generic.h>
+#include <net/xdp_sock_drv.h>
 #include "gve.h"
 #include "gve_dqo.h"
 #include "gve_adminq.h"
@@ -1188,6 +1189,7 @@ static int gve_reg_xdp_info(struct gve_priv *priv, struct net_device *dev)
 	struct gve_rx_ring *rx;
 	int err = 0;
 	int i, j;
+	u32 tx_qid;
 
 	if (!priv->num_xdp_queues)
 		return 0;
@@ -1204,6 +1206,24 @@ static int gve_reg_xdp_info(struct gve_priv *priv, struct net_device *dev)
 						 MEM_TYPE_PAGE_SHARED, NULL);
 		if (err)
 			goto err;
+		rx->xsk_pool = xsk_get_pool_from_qid(dev, i);
+		if (rx->xsk_pool) {
+			err = xdp_rxq_info_reg(&rx->xsk_rxq, dev, i,
+					       napi->napi_id);
+			if (err)
+				goto err;
+			err = xdp_rxq_info_reg_mem_model(&rx->xsk_rxq,
+							 MEM_TYPE_XSK_BUFF_POOL, NULL);
+			if (err)
+				goto err;
+			xsk_pool_set_rxq_info(rx->xsk_pool,
+					      &rx->xsk_rxq);
+		}
+	}
+
+	for (i = 0; i < priv->num_xdp_queues; i++) {
+		tx_qid = gve_xdp_tx_queue_id(priv, i);
+		priv->tx[tx_qid].xsk_pool = xsk_get_pool_from_qid(dev, i);
 	}
 	return 0;
 
@@ -1212,13 +1232,15 @@ err:
 		rx = &priv->rx[j];
 		if (xdp_rxq_info_is_reg(&rx->xdp_rxq))
 			xdp_rxq_info_unreg(&rx->xdp_rxq);
+		if (xdp_rxq_info_is_reg(&rx->xsk_rxq))
+			xdp_rxq_info_unreg(&rx->xsk_rxq);
 	}
 	return err;
 }
 
 static void gve_unreg_xdp_info(struct gve_priv *priv)
 {
-	int i;
+	int i, tx_qid;
 
 	if (!priv->num_xdp_queues)
 		return;
@@ -1227,6 +1249,15 @@ static void gve_unreg_xdp_info(struct gve_priv *priv)
 		struct gve_rx_ring *rx = &priv->rx[i];
 
 		xdp_rxq_info_unreg(&rx->xdp_rxq);
+		if (rx->xsk_pool) {
+			xdp_rxq_info_unreg(&rx->xsk_rxq);
+			rx->xsk_pool = NULL;
+		}
+	}
+
+	for (i = 0; i < priv->num_xdp_queues; i++) {
+		tx_qid = gve_xdp_tx_queue_id(priv, i);
+		priv->tx[tx_qid].xsk_pool = NULL;
 	}
 }
 
@@ -1469,6 +1500,140 @@ out:
 	return err;
 }
 
+static int gve_xsk_pool_enable(struct net_device *dev,
+			       struct xsk_buff_pool *pool,
+			       u16 qid)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct napi_struct *napi;
+	struct gve_rx_ring *rx;
+	int tx_qid;
+	int err;
+
+	if (qid >= priv->rx_cfg.num_queues) {
+		dev_err(&priv->pdev->dev, "xsk pool invalid qid %d", qid);
+		return -EINVAL;
+	}
+	if (xsk_pool_get_rx_frame_size(pool) <
+	     priv->dev->max_mtu + sizeof(struct ethhdr)) {
+		dev_err(&priv->pdev->dev, "xsk pool frame_len too small");
+		return -EINVAL;
+	}
+
+	err = xsk_pool_dma_map(pool, &priv->pdev->dev,
+			       DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
+	if (err)
+		return err;
+
+	/* If XDP prog is not installed, return */
+	if (!priv->xdp_prog)
+		return 0;
+
+	rx = &priv->rx[qid];
+	napi = &priv->ntfy_blocks[rx->ntfy_id].napi;
+	err = xdp_rxq_info_reg(&rx->xsk_rxq, dev, qid, napi->napi_id);
+	if (err)
+		goto err;
+
+	err = xdp_rxq_info_reg_mem_model(&rx->xsk_rxq,
+					 MEM_TYPE_XSK_BUFF_POOL, NULL);
+	if (err)
+		goto err;
+
+	xsk_pool_set_rxq_info(pool, &rx->xsk_rxq);
+	rx->xsk_pool = pool;
+
+	tx_qid = gve_xdp_tx_queue_id(priv, qid);
+	priv->tx[tx_qid].xsk_pool = pool;
+
+	return 0;
+err:
+	if (xdp_rxq_info_is_reg(&rx->xsk_rxq))
+		xdp_rxq_info_unreg(&rx->xsk_rxq);
+
+	xsk_pool_dma_unmap(pool,
+			   DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
+	return err;
+}
+
+static int gve_xsk_pool_disable(struct net_device *dev,
+				u16 qid)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	struct napi_struct *napi_rx;
+	struct napi_struct *napi_tx;
+	struct xsk_buff_pool *pool;
+	int tx_qid;
+
+	pool = xsk_get_pool_from_qid(dev, qid);
+	if (!pool)
+		return -EINVAL;
+	if (qid >= priv->rx_cfg.num_queues)
+		return -EINVAL;
+
+	/* If XDP prog is not installed, unmap DMA and return */
+	if (!priv->xdp_prog)
+		goto done;
+
+	tx_qid = gve_xdp_tx_queue_id(priv, qid);
+	if (!netif_running(dev)) {
+		priv->rx[qid].xsk_pool = NULL;
+		xdp_rxq_info_unreg(&priv->rx[qid].xsk_rxq);
+		priv->tx[tx_qid].xsk_pool = NULL;
+		goto done;
+	}
+
+	napi_rx = &priv->ntfy_blocks[priv->rx[qid].ntfy_id].napi;
+	napi_disable(napi_rx); /* make sure current rx poll is done */
+
+	napi_tx = &priv->ntfy_blocks[priv->tx[tx_qid].ntfy_id].napi;
+	napi_disable(napi_tx); /* make sure current tx poll is done */
+
+	priv->rx[qid].xsk_pool = NULL;
+	xdp_rxq_info_unreg(&priv->rx[qid].xsk_rxq);
+	priv->tx[tx_qid].xsk_pool = NULL;
+	smp_mb(); /* Make sure it is visible to the workers on datapath */
+
+	napi_enable(napi_rx);
+	if (gve_rx_work_pending(&priv->rx[qid]))
+		napi_schedule(napi_rx);
+
+	napi_enable(napi_tx);
+	if (gve_tx_clean_pending(priv, &priv->tx[tx_qid]))
+		napi_schedule(napi_tx);
+
+done:
+	xsk_pool_dma_unmap(pool,
+			   DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
+	return 0;
+}
+
+static int gve_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags)
+{
+	struct gve_priv *priv = netdev_priv(dev);
+	int tx_queue_id = gve_xdp_tx_queue_id(priv, queue_id);
+
+	if (queue_id >= priv->rx_cfg.num_queues || !priv->xdp_prog)
+		return -EINVAL;
+
+	if (flags & XDP_WAKEUP_TX) {
+		struct gve_tx_ring *tx = &priv->tx[tx_queue_id];
+		struct napi_struct *napi =
+			&priv->ntfy_blocks[tx->ntfy_id].napi;
+
+		if (!napi_if_scheduled_mark_missed(napi)) {
+			/* Call local_bh_enable to trigger SoftIRQ processing */
+			local_bh_disable();
+			napi_schedule(napi);
+			local_bh_enable();
+		}
+
+		tx->xdp_xsk_wakeup++;
+	}
+
+	return 0;
+}
+
 static int verify_xdp_configuration(struct net_device *dev)
 {
 	struct gve_priv *priv = netdev_priv(dev);
@@ -1512,6 +1677,11 @@ static int gve_xdp(struct net_device *dev, struct netdev_bpf *xdp)
 	switch (xdp->command) {
 	case XDP_SETUP_PROG:
 		return gve_set_xdp(priv, xdp->prog, xdp->extack);
+	case XDP_SETUP_XSK_POOL:
+		if (xdp->xsk.pool)
+			return gve_xsk_pool_enable(dev, xdp->xsk.pool, xdp->xsk.queue_id);
+		else
+			return gve_xsk_pool_disable(dev, xdp->xsk.queue_id);
 	default:
 		return -EINVAL;
 	}
@@ -1713,6 +1883,7 @@ static const struct net_device_ops gve_netdev_ops = {
 	.ndo_set_features	=	gve_set_features,
 	.ndo_bpf		=	gve_xdp,
 	.ndo_xdp_xmit		=	gve_xdp_xmit,
+	.ndo_xsk_wakeup		=	gve_xsk_wakeup,
 };
 
 static void gve_handle_status(struct gve_priv *priv, u32 status)
@@ -1838,6 +2009,7 @@ static void gve_set_netdev_xdp_features(struct gve_priv *priv)
 		priv->dev->xdp_features = NETDEV_XDP_ACT_BASIC;
 		priv->dev->xdp_features |= NETDEV_XDP_ACT_REDIRECT;
 		priv->dev->xdp_features |= NETDEV_XDP_ACT_NDO_XMIT;
+		priv->dev->xdp_features |= NETDEV_XDP_ACT_XSK_ZEROCOPY;
 	} else {
 		priv->dev->xdp_features = 0;
 	}
diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c
index ed4b5a540e6d..d1da7413dc4d 100644
--- a/drivers/net/ethernet/google/gve/gve_rx.c
+++ b/drivers/net/ethernet/google/gve/gve_rx.c
@@ -10,6 +10,7 @@
 #include <linux/etherdevice.h>
 #include <linux/filter.h>
 #include <net/xdp.h>
+#include <net/xdp_sock_drv.h>
 
 static void gve_rx_free_buffer(struct device *dev,
 			       struct gve_rx_slot_page_info *page_info,
@@ -593,6 +594,31 @@ static struct sk_buff *gve_rx_skb(struct gve_priv *priv, struct gve_rx_ring *rx,
 	return skb;
 }
 
+static int gve_xsk_pool_redirect(struct net_device *dev,
+				 struct gve_rx_ring *rx,
+				 void *data, int len,
+				 struct bpf_prog *xdp_prog)
+{
+	struct xdp_buff *xdp;
+	int err;
+
+	if (rx->xsk_pool->frame_len < len)
+		return -E2BIG;
+	xdp = xsk_buff_alloc(rx->xsk_pool);
+	if (!xdp) {
+		u64_stats_update_begin(&rx->statss);
+		rx->xdp_alloc_fails++;
+		u64_stats_update_end(&rx->statss);
+		return -ENOMEM;
+	}
+	xdp->data_end = xdp->data + len;
+	memcpy(xdp->data, data, len);
+	err = xdp_do_redirect(dev, xdp, xdp_prog);
+	if (err)
+		xsk_buff_free(xdp);
+	return err;
+}
+
 static int gve_xdp_redirect(struct net_device *dev, struct gve_rx_ring *rx,
 			    struct xdp_buff *orig, struct bpf_prog *xdp_prog)
 {
@@ -602,6 +628,10 @@ static int gve_xdp_redirect(struct net_device *dev, struct gve_rx_ring *rx,
 	void *frame;
 	int err;
 
+	if (rx->xsk_pool)
+		return gve_xsk_pool_redirect(dev, rx, orig->data,
+					     len, xdp_prog);
+
 	total_len = headroom + SKB_DATA_ALIGN(len) +
 		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 	frame = page_frag_alloc(&rx->page_cache, total_len, GFP_ATOMIC);
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
index d928c3c79618..e50510b8e784 100644
--- a/drivers/net/ethernet/google/gve/gve_tx.c
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -11,6 +11,7 @@
 #include <linux/tcp.h>
 #include <linux/vmalloc.h>
 #include <linux/skbuff.h>
+#include <net/xdp_sock_drv.h>
 
 static inline void gve_tx_put_doorbell(struct gve_priv *priv,
 				       struct gve_queue_resources *q_resources,
@@ -160,6 +161,7 @@ static int gve_clean_xdp_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 	u32 clean_end = tx->done + to_do;
 	u64 pkts = 0, bytes = 0;
 	size_t space_freed = 0;
+	u32 xsk_complete = 0;
 	u32 idx;
 
 	for (; tx->done < clean_end; tx->done++) {
@@ -171,6 +173,7 @@ static int gve_clean_xdp_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 
 		bytes += info->xdp.size;
 		pkts++;
+		xsk_complete += info->xdp.is_xsk;
 
 		info->xdp.size = 0;
 		if (info->xdp_frame) {
@@ -181,6 +184,8 @@ static int gve_clean_xdp_done(struct gve_priv *priv, struct gve_tx_ring *tx,
 	}
 
 	gve_tx_free_fifo(&tx->tx_fifo, space_freed);
+	if (xsk_complete > 0 && tx->xsk_pool)
+		xsk_tx_completed(tx->xsk_pool, xsk_complete);
 	u64_stats_update_begin(&tx->statss);
 	tx->bytes_done += bytes;
 	tx->pkt_done += pkts;
@@ -720,7 +725,7 @@ netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev)
 }
 
 static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
-			   void *data, int len, void *frame_p)
+			   void *data, int len, void *frame_p, bool is_xsk)
 {
 	int pad, nfrags, ndescs, iovi, offset;
 	struct gve_tx_buffer_state *info;
@@ -732,6 +737,7 @@ static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
 	info = &tx->info[reqi & tx->mask];
 	info->xdp_frame = frame_p;
 	info->xdp.size = len;
+	info->xdp.is_xsk = is_xsk;
 
 	nfrags = gve_tx_alloc_fifo(&tx->tx_fifo, pad + len,
 				   &info->iov[0]);
@@ -809,7 +815,7 @@ int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
 	if (!gve_can_tx(tx, len + GVE_TX_MAX_HEADER_SIZE - 1))
 		return -EBUSY;
 
-	nsegs = gve_tx_fill_xdp(priv, tx, data, len, frame_p);
+	nsegs = gve_tx_fill_xdp(priv, tx, data, len, frame_p, false);
 	tx->req += nsegs;
 
 	return 0;
@@ -882,11 +888,43 @@ u32 gve_tx_load_event_counter(struct gve_priv *priv,
 	return be32_to_cpu(counter);
 }
 
+static int gve_xsk_tx(struct gve_priv *priv, struct gve_tx_ring *tx,
+		      int budget)
+{
+	struct xdp_desc desc;
+	int sent = 0, nsegs;
+	void *data;
+
+	spin_lock(&tx->xdp_lock);
+	while (sent < budget) {
+		if (!gve_can_tx(tx, GVE_TX_START_THRESH))
+			goto out;
+
+		if (!xsk_tx_peek_desc(tx->xsk_pool, &desc)) {
+			tx->xdp_xsk_done = tx->xdp_xsk_wakeup;
+			goto out;
+		}
+
+		data = xsk_buff_raw_get_data(tx->xsk_pool, desc.addr);
+		nsegs = gve_tx_fill_xdp(priv, tx, data, desc.len, NULL, true);
+		tx->req += nsegs;
+		sent++;
+	}
+out:
+	if (sent > 0) {
+		gve_tx_put_doorbell(priv, tx->q_resources, tx->req);
+		xsk_tx_release(tx->xsk_pool);
+	}
+	spin_unlock(&tx->xdp_lock);
+	return sent;
+}
+
 bool gve_xdp_poll(struct gve_notify_block *block, int budget)
 {
 	struct gve_priv *priv = block->priv;
 	struct gve_tx_ring *tx = block->tx;
 	u32 nic_done;
+	bool repoll;
 	u32 to_do;
 
 	/* If budget is 0, do all the work */
@@ -897,7 +935,21 @@ bool gve_xdp_poll(struct gve_notify_block *block, int budget)
 	nic_done = gve_tx_load_event_counter(priv, tx);
 	to_do = min_t(u32, (nic_done - tx->done), budget);
 	gve_clean_xdp_done(priv, tx, to_do);
-	return nic_done != tx->done;
+	repoll = nic_done != tx->done;
+
+	if (tx->xsk_pool) {
+		int sent = gve_xsk_tx(priv, tx, budget);
+
+		u64_stats_update_begin(&tx->statss);
+		tx->xdp_xsk_sent += sent;
+		u64_stats_update_end(&tx->statss);
+		repoll |= (sent == budget);
+		if (xsk_uses_need_wakeup(tx->xsk_pool))
+			xsk_set_tx_need_wakeup(tx->xsk_pool);
+	}
+
+	/* If we still have work we want to repoll */
+	return repoll;
 }
 
 bool gve_tx_poll(struct gve_notify_block *block, int budget)
-- 
cgit v1.2.3-70-g09d2


From b9d83ab8a708f23a4001d60e9d8d0b3be3d9f607 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:06 +0000
Subject: net/packet: annotate accesses to po->xmit

po->xmit can be set from setsockopt(PACKET_QDISC_BYPASS),
while read locklessly.

Use READ_ONCE()/WRITE_ONCE() to avoid potential load/store
tearing issues.

Fixes: d346a3fae3ff ("packet: introduce PACKET_QDISC_BYPASS socket option")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index d4e76e2ae153..d25dd9f63cc4 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -307,7 +307,8 @@ static void packet_cached_dev_reset(struct packet_sock *po)
 
 static bool packet_use_direct_xmit(const struct packet_sock *po)
 {
-	return po->xmit == packet_direct_xmit;
+	/* Paired with WRITE_ONCE() in packet_setsockopt() */
+	return READ_ONCE(po->xmit) == packet_direct_xmit;
 }
 
 static u16 packet_pick_tx_queue(struct sk_buff *skb)
@@ -2867,7 +2868,8 @@ tpacket_error:
 		packet_inc_pending(&po->tx_ring);
 
 		status = TP_STATUS_SEND_REQUEST;
-		err = po->xmit(skb);
+		/* Paired with WRITE_ONCE() in packet_setsockopt() */
+		err = READ_ONCE(po->xmit)(skb);
 		if (unlikely(err != 0)) {
 			if (err > 0)
 				err = net_xmit_errno(err);
@@ -3070,7 +3072,8 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 		virtio_net_hdr_set_proto(skb, &vnet_hdr);
 	}
 
-	err = po->xmit(skb);
+	/* Paired with WRITE_ONCE() in packet_setsockopt() */
+	err = READ_ONCE(po->xmit)(skb);
 	if (unlikely(err != 0)) {
 		if (err > 0)
 			err = net_xmit_errno(err);
@@ -4007,7 +4010,8 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (copy_from_sockptr(&val, optval, sizeof(val)))
 			return -EFAULT;
 
-		po->xmit = val ? packet_direct_xmit : dev_queue_xmit;
+		/* Paired with all lockless reads of po->xmit */
+		WRITE_ONCE(po->xmit, val ? packet_direct_xmit : dev_queue_xmit);
 		return 0;
 	}
 	default:
-- 
cgit v1.2.3-70-g09d2


From ee5675ecdf7a4e713ed21d98a70c2871d6ebed01 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:07 +0000
Subject: net/packet: convert po->origdev to an atomic flag

syzbot/KCAN reported that po->origdev can be read
while another thread is changing its value.

We can avoid this splat by converting this field
to an actual bit.

Following patches will convert remaining 1bit fields.

Fixes: 80feaacb8a64 ("[AF_PACKET]: Add option to return orig_dev to userspace.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 10 ++++------
 net/packet/diag.c      |  2 +-
 net/packet/internal.h  | 22 +++++++++++++++++++++-
 3 files changed, 26 insertions(+), 8 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index d25dd9f63cc4..af7c44169b86 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2184,7 +2184,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev,
 	sll = &PACKET_SKB_CB(skb)->sa.ll;
 	sll->sll_hatype = dev->type;
 	sll->sll_pkttype = skb->pkt_type;
-	if (unlikely(po->origdev))
+	if (unlikely(packet_sock_flag(po, PACKET_SOCK_ORIGDEV)))
 		sll->sll_ifindex = orig_dev->ifindex;
 	else
 		sll->sll_ifindex = dev->ifindex;
@@ -2461,7 +2461,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 	sll->sll_hatype = dev->type;
 	sll->sll_protocol = skb->protocol;
 	sll->sll_pkttype = skb->pkt_type;
-	if (unlikely(po->origdev))
+	if (unlikely(packet_sock_flag(po, PACKET_SOCK_ORIGDEV)))
 		sll->sll_ifindex = orig_dev->ifindex;
 	else
 		sll->sll_ifindex = dev->ifindex;
@@ -3914,9 +3914,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (copy_from_sockptr(&val, optval, sizeof(val)))
 			return -EFAULT;
 
-		lock_sock(sk);
-		po->origdev = !!val;
-		release_sock(sk);
+		packet_sock_flag_set(po, PACKET_SOCK_ORIGDEV, val);
 		return 0;
 	}
 	case PACKET_VNET_HDR:
@@ -4065,7 +4063,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		val = po->auxdata;
 		break;
 	case PACKET_ORIGDEV:
-		val = po->origdev;
+		val = packet_sock_flag(po, PACKET_SOCK_ORIGDEV);
 		break;
 	case PACKET_VNET_HDR:
 		val = po->has_vnet_hdr;
diff --git a/net/packet/diag.c b/net/packet/diag.c
index 07812ae5ca07..e1ac9bb375b3 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -25,7 +25,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 		pinfo.pdi_flags |= PDI_RUNNING;
 	if (po->auxdata)
 		pinfo.pdi_flags |= PDI_AUXDATA;
-	if (po->origdev)
+	if (packet_sock_flag(po, PACKET_SOCK_ORIGDEV))
 		pinfo.pdi_flags |= PDI_ORIGDEV;
 	if (po->has_vnet_hdr)
 		pinfo.pdi_flags |= PDI_VNETHDR;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 48af35b1aed2..178cd1852238 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -116,9 +116,9 @@ struct packet_sock {
 	int			copy_thresh;
 	spinlock_t		bind_lock;
 	struct mutex		pg_vec_lock;
+	unsigned long		flags;
 	unsigned int		running;	/* bind_lock must be held */
 	unsigned int		auxdata:1,	/* writer must hold sock lock */
-				origdev:1,
 				has_vnet_hdr:1,
 				tp_loss:1,
 				tp_tx_has_off:1;
@@ -144,4 +144,24 @@ static inline struct packet_sock *pkt_sk(struct sock *sk)
 	return (struct packet_sock *)sk;
 }
 
+enum packet_sock_flags {
+	PACKET_SOCK_ORIGDEV,
+};
+
+static inline void packet_sock_flag_set(struct packet_sock *po,
+					enum packet_sock_flags flag,
+					bool val)
+{
+	if (val)
+		set_bit(flag, &po->flags);
+	else
+		clear_bit(flag, &po->flags);
+}
+
+static inline bool packet_sock_flag(const struct packet_sock *po,
+				    enum packet_sock_flags flag)
+{
+	return test_bit(flag, &po->flags);
+}
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From fd53c297aa7b077ae98a3d3d2d3aa278a1686ba6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:08 +0000
Subject: net/packet: convert po->auxdata to an atomic flag

po->auxdata can be read while another thread
is changing its value, potentially raising KCSAN splat.

Convert it to PACKET_SOCK_AUXDATA flag.

Fixes: 8dc419447415 ("[PACKET]: Add optional checksum computation for recvmsg")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 8 +++-----
 net/packet/diag.c      | 2 +-
 net/packet/internal.h  | 4 ++--
 3 files changed, 6 insertions(+), 8 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index af7c44169b86..ecd9fc27e360 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -3514,7 +3514,7 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 		memcpy(msg->msg_name, &PACKET_SKB_CB(skb)->sa, copy_len);
 	}
 
-	if (pkt_sk(sk)->auxdata) {
+	if (packet_sock_flag(pkt_sk(sk), PACKET_SOCK_AUXDATA)) {
 		struct tpacket_auxdata aux;
 
 		aux.tp_status = TP_STATUS_USER;
@@ -3900,9 +3900,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (copy_from_sockptr(&val, optval, sizeof(val)))
 			return -EFAULT;
 
-		lock_sock(sk);
-		po->auxdata = !!val;
-		release_sock(sk);
+		packet_sock_flag_set(po, PACKET_SOCK_AUXDATA, val);
 		return 0;
 	}
 	case PACKET_ORIGDEV:
@@ -4060,7 +4058,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 
 		break;
 	case PACKET_AUXDATA:
-		val = po->auxdata;
+		val = packet_sock_flag(po, PACKET_SOCK_AUXDATA);
 		break;
 	case PACKET_ORIGDEV:
 		val = packet_sock_flag(po, PACKET_SOCK_ORIGDEV);
diff --git a/net/packet/diag.c b/net/packet/diag.c
index e1ac9bb375b3..d704c7bf51b2 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -23,7 +23,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 	pinfo.pdi_flags = 0;
 	if (po->running)
 		pinfo.pdi_flags |= PDI_RUNNING;
-	if (po->auxdata)
+	if (packet_sock_flag(po, PACKET_SOCK_AUXDATA))
 		pinfo.pdi_flags |= PDI_AUXDATA;
 	if (packet_sock_flag(po, PACKET_SOCK_ORIGDEV))
 		pinfo.pdi_flags |= PDI_ORIGDEV;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 178cd1852238..3bae8ea7a36f 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -118,8 +118,7 @@ struct packet_sock {
 	struct mutex		pg_vec_lock;
 	unsigned long		flags;
 	unsigned int		running;	/* bind_lock must be held */
-	unsigned int		auxdata:1,	/* writer must hold sock lock */
-				has_vnet_hdr:1,
+	unsigned int		has_vnet_hdr:1, /* writer must hold sock lock */
 				tp_loss:1,
 				tp_tx_has_off:1;
 	int			pressure;
@@ -146,6 +145,7 @@ static inline struct packet_sock *pkt_sk(struct sock *sk)
 
 enum packet_sock_flags {
 	PACKET_SOCK_ORIGDEV,
+	PACKET_SOCK_AUXDATA,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From 1051ce4ab64db91f7b62369ddc321ba8747f8c84 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:09 +0000
Subject: net/packet: annotate accesses to po->tp_tstamp

tp_tstamp is read locklessly.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 9 +++++----
 net/packet/diag.c      | 2 +-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ecd9fc27e360..a27a811fa2b0 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -474,7 +474,7 @@ static __u32 __packet_set_timestamp(struct packet_sock *po, void *frame,
 	struct timespec64 ts;
 	__u32 ts_status;
 
-	if (!(ts_status = tpacket_get_timestamp(skb, &ts, po->tp_tstamp)))
+	if (!(ts_status = tpacket_get_timestamp(skb, &ts, READ_ONCE(po->tp_tstamp))))
 		return 0;
 
 	h.raw = frame;
@@ -2403,7 +2403,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 	 * closer to the time of capture.
 	 */
 	ts_status = tpacket_get_timestamp(skb, &ts,
-					  po->tp_tstamp | SOF_TIMESTAMPING_SOFTWARE);
+					  READ_ONCE(po->tp_tstamp) |
+					  SOF_TIMESTAMPING_SOFTWARE);
 	if (!ts_status)
 		ktime_get_real_ts64(&ts);
 
@@ -3945,7 +3946,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (copy_from_sockptr(&val, optval, sizeof(val)))
 			return -EFAULT;
 
-		po->tp_tstamp = val;
+		WRITE_ONCE(po->tp_tstamp, val);
 		return 0;
 	}
 	case PACKET_FANOUT:
@@ -4097,7 +4098,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		val = po->tp_loss;
 		break;
 	case PACKET_TIMESTAMP:
-		val = po->tp_tstamp;
+		val = READ_ONCE(po->tp_tstamp);
 		break;
 	case PACKET_FANOUT:
 		val = (po->fanout ?
diff --git a/net/packet/diag.c b/net/packet/diag.c
index d704c7bf51b2..0abca32e2f87 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -18,7 +18,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 	pinfo.pdi_version = po->tp_version;
 	pinfo.pdi_reserve = po->tp_reserve;
 	pinfo.pdi_copy_thresh = po->copy_thresh;
-	pinfo.pdi_tstamp = po->tp_tstamp;
+	pinfo.pdi_tstamp = READ_ONCE(po->tp_tstamp);
 
 	pinfo.pdi_flags = 0;
 	if (po->running)
-- 
cgit v1.2.3-70-g09d2


From 7438344660fa55b33b8234c1797c886eb73667a7 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:10 +0000
Subject: net/packet: convert po->tp_tx_has_off to an atomic flag

This is to use existing space in po->flags, and reclaim
the storage used by the non atomic bit fields.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 6 +++---
 net/packet/internal.h  | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index a27a811fa2b0..7800dc622ff3 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2672,7 +2672,7 @@ static int tpacket_parse_header(struct packet_sock *po, void *frame,
 		return -EMSGSIZE;
 	}
 
-	if (unlikely(po->tp_tx_has_off)) {
+	if (unlikely(packet_sock_flag(po, PACKET_SOCK_TX_HAS_OFF))) {
 		int off_min, off_max;
 
 		off_min = po->tp_hdrlen - sizeof(struct sockaddr_ll);
@@ -3993,7 +3993,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 
 		lock_sock(sk);
 		if (!po->rx_ring.pg_vec && !po->tx_ring.pg_vec)
-			po->tp_tx_has_off = !!val;
+			packet_sock_flag_set(po, PACKET_SOCK_TX_HAS_OFF, val);
 
 		release_sock(sk);
 		return 0;
@@ -4120,7 +4120,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		lv = sizeof(rstats);
 		break;
 	case PACKET_TX_HAS_OFF:
-		val = po->tp_tx_has_off;
+		val = packet_sock_flag(po, PACKET_SOCK_TX_HAS_OFF);
 		break;
 	case PACKET_QDISC_BYPASS:
 		val = packet_use_direct_xmit(po);
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 3bae8ea7a36f..0d16a581e271 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -119,8 +119,7 @@ struct packet_sock {
 	unsigned long		flags;
 	unsigned int		running;	/* bind_lock must be held */
 	unsigned int		has_vnet_hdr:1, /* writer must hold sock lock */
-				tp_loss:1,
-				tp_tx_has_off:1;
+				tp_loss:1;
 	int			pressure;
 	int			ifindex;	/* bound device		*/
 	__be16			num;
@@ -146,6 +145,7 @@ static inline struct packet_sock *pkt_sk(struct sock *sk)
 enum packet_sock_flags {
 	PACKET_SOCK_ORIGDEV,
 	PACKET_SOCK_AUXDATA,
+	PACKET_SOCK_TX_HAS_OFF,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From 164bddace2e03f6005e650cb88f101a66ebdc05a Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:11 +0000
Subject: net/packet: convert po->tp_loss to an atomic flag

tp_loss can be read locklessly.

Convert it to an atomic flag to avoid races.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 6 +++---
 net/packet/diag.c      | 2 +-
 net/packet/internal.h  | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 7800dc622ff3..119063c8a1c5 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2843,7 +2843,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 
 		if (unlikely(tp_len < 0)) {
 tpacket_error:
-			if (po->tp_loss) {
+			if (packet_sock_flag(po, PACKET_SOCK_TP_LOSS)) {
 				__packet_set_status(po, ph,
 						TP_STATUS_AVAILABLE);
 				packet_increment_head(&po->tx_ring);
@@ -3886,7 +3886,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) {
 			ret = -EBUSY;
 		} else {
-			po->tp_loss = !!val;
+			packet_sock_flag_set(po, PACKET_SOCK_TP_LOSS, val);
 			ret = 0;
 		}
 		release_sock(sk);
@@ -4095,7 +4095,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		val = po->tp_reserve;
 		break;
 	case PACKET_LOSS:
-		val = po->tp_loss;
+		val = packet_sock_flag(po, PACKET_SOCK_TP_LOSS);
 		break;
 	case PACKET_TIMESTAMP:
 		val = READ_ONCE(po->tp_tstamp);
diff --git a/net/packet/diag.c b/net/packet/diag.c
index 0abca32e2f87..8bb4ce6a8e61 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -29,7 +29,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 		pinfo.pdi_flags |= PDI_ORIGDEV;
 	if (po->has_vnet_hdr)
 		pinfo.pdi_flags |= PDI_VNETHDR;
-	if (po->tp_loss)
+	if (packet_sock_flag(po, PACKET_SOCK_TP_LOSS))
 		pinfo.pdi_flags |= PDI_LOSS;
 
 	return nla_put(nlskb, PACKET_DIAG_INFO, sizeof(pinfo), &pinfo);
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 0d16a581e271..9d406a92ede8 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -118,8 +118,7 @@ struct packet_sock {
 	struct mutex		pg_vec_lock;
 	unsigned long		flags;
 	unsigned int		running;	/* bind_lock must be held */
-	unsigned int		has_vnet_hdr:1, /* writer must hold sock lock */
-				tp_loss:1;
+	unsigned int		has_vnet_hdr:1; /* writer must hold sock lock */
 	int			pressure;
 	int			ifindex;	/* bound device		*/
 	__be16			num;
@@ -146,6 +145,7 @@ enum packet_sock_flags {
 	PACKET_SOCK_ORIGDEV,
 	PACKET_SOCK_AUXDATA,
 	PACKET_SOCK_TX_HAS_OFF,
+	PACKET_SOCK_TP_LOSS,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From 50d935eafee292fc432d5ac8c8715a6492961abc Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:12 +0000
Subject: net/packet: convert po->has_vnet_hdr to an atomic flag

po->has_vnet_hdr can be read locklessly.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 19 ++++++++++---------
 net/packet/diag.c      |  2 +-
 net/packet/internal.h  |  2 +-
 3 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 119063c8a1c5..5a6b05e17ca2 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2309,7 +2309,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		netoff = TPACKET_ALIGN(po->tp_hdrlen +
 				       (maclen < 16 ? 16 : maclen)) +
 				       po->tp_reserve;
-		if (po->has_vnet_hdr) {
+		if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
 			netoff += sizeof(struct virtio_net_hdr);
 			do_vnet = true;
 		}
@@ -2780,7 +2780,8 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	size_max = po->tx_ring.frame_size
 		- (po->tp_hdrlen - sizeof(struct sockaddr_ll));
 
-	if ((size_max > dev->mtu + reserve + VLAN_HLEN) && !po->has_vnet_hdr)
+	if ((size_max > dev->mtu + reserve + VLAN_HLEN) &&
+	    !packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR))
 		size_max = dev->mtu + reserve + VLAN_HLEN;
 
 	reinit_completion(&po->skb_completion);
@@ -2809,7 +2810,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 		status = TP_STATUS_SEND_REQUEST;
 		hlen = LL_RESERVED_SPACE(dev);
 		tlen = dev->needed_tailroom;
-		if (po->has_vnet_hdr) {
+		if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
 			vnet_hdr = data;
 			data += sizeof(*vnet_hdr);
 			tp_len -= sizeof(*vnet_hdr);
@@ -2837,7 +2838,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 					  addr, hlen, copylen, &sockc);
 		if (likely(tp_len >= 0) &&
 		    tp_len > dev->mtu + reserve &&
-		    !po->has_vnet_hdr &&
+		    !packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR) &&
 		    !packet_extra_vlan_len_allowed(dev, skb))
 			tp_len = -EMSGSIZE;
 
@@ -2856,7 +2857,7 @@ tpacket_error:
 			}
 		}
 
-		if (po->has_vnet_hdr) {
+		if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
 			if (virtio_net_hdr_to_skb(skb, vnet_hdr, vio_le())) {
 				tp_len = -EINVAL;
 				goto tpacket_error;
@@ -2991,7 +2992,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 
 	if (sock->type == SOCK_RAW)
 		reserve = dev->hard_header_len;
-	if (po->has_vnet_hdr) {
+	if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
 		err = packet_snd_vnet_parse(msg, &len, &vnet_hdr);
 		if (err)
 			goto out_unlock;
@@ -3451,7 +3452,7 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 
 	packet_rcv_try_clear_pressure(pkt_sk(sk));
 
-	if (pkt_sk(sk)->has_vnet_hdr) {
+	if (packet_sock_flag(pkt_sk(sk), PACKET_SOCK_HAS_VNET_HDR)) {
 		err = packet_rcv_vnet(msg, skb, &len);
 		if (err)
 			goto out_free;
@@ -3931,7 +3932,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) {
 			ret = -EBUSY;
 		} else {
-			po->has_vnet_hdr = !!val;
+			packet_sock_flag_set(po, PACKET_SOCK_HAS_VNET_HDR, val);
 			ret = 0;
 		}
 		release_sock(sk);
@@ -4065,7 +4066,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		val = packet_sock_flag(po, PACKET_SOCK_ORIGDEV);
 		break;
 	case PACKET_VNET_HDR:
-		val = po->has_vnet_hdr;
+		val = packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR);
 		break;
 	case PACKET_VERSION:
 		val = po->tp_version;
diff --git a/net/packet/diag.c b/net/packet/diag.c
index 8bb4ce6a8e61..56240aaf032b 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -27,7 +27,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 		pinfo.pdi_flags |= PDI_AUXDATA;
 	if (packet_sock_flag(po, PACKET_SOCK_ORIGDEV))
 		pinfo.pdi_flags |= PDI_ORIGDEV;
-	if (po->has_vnet_hdr)
+	if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR))
 		pinfo.pdi_flags |= PDI_VNETHDR;
 	if (packet_sock_flag(po, PACKET_SOCK_TP_LOSS))
 		pinfo.pdi_flags |= PDI_LOSS;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 9d406a92ede8..2521176807f4 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -118,7 +118,6 @@ struct packet_sock {
 	struct mutex		pg_vec_lock;
 	unsigned long		flags;
 	unsigned int		running;	/* bind_lock must be held */
-	unsigned int		has_vnet_hdr:1; /* writer must hold sock lock */
 	int			pressure;
 	int			ifindex;	/* bound device		*/
 	__be16			num;
@@ -146,6 +145,7 @@ enum packet_sock_flags {
 	PACKET_SOCK_AUXDATA,
 	PACKET_SOCK_TX_HAS_OFF,
 	PACKET_SOCK_TP_LOSS,
+	PACKET_SOCK_HAS_VNET_HDR,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From 61edf479818e63978cabd243b82ca80f8948a313 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:13 +0000
Subject: net/packet: convert po->running to an atomic flag

Instead of consuming 32 bits for po->running, use
one available bit in po->flags.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 20 ++++++++++----------
 net/packet/diag.c      |  2 +-
 net/packet/internal.h  |  2 +-
 3 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 5a6b05e17ca2..ec446452bbe8 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -340,14 +340,14 @@ static void __register_prot_hook(struct sock *sk)
 {
 	struct packet_sock *po = pkt_sk(sk);
 
-	if (!po->running) {
+	if (!packet_sock_flag(po, PACKET_SOCK_RUNNING)) {
 		if (po->fanout)
 			__fanout_link(sk, po);
 		else
 			dev_add_pack(&po->prot_hook);
 
 		sock_hold(sk);
-		po->running = 1;
+		packet_sock_flag_set(po, PACKET_SOCK_RUNNING, 1);
 	}
 }
 
@@ -369,7 +369,7 @@ static void __unregister_prot_hook(struct sock *sk, bool sync)
 
 	lockdep_assert_held_once(&po->bind_lock);
 
-	po->running = 0;
+	packet_sock_flag_set(po, PACKET_SOCK_RUNNING, 0);
 
 	if (po->fanout)
 		__fanout_unlink(sk, po);
@@ -389,7 +389,7 @@ static void unregister_prot_hook(struct sock *sk, bool sync)
 {
 	struct packet_sock *po = pkt_sk(sk);
 
-	if (po->running)
+	if (packet_sock_flag(po, PACKET_SOCK_RUNNING))
 		__unregister_prot_hook(sk, sync);
 }
 
@@ -1782,7 +1782,7 @@ static int fanout_add(struct sock *sk, struct fanout_args *args)
 	err = -EINVAL;
 
 	spin_lock(&po->bind_lock);
-	if (po->running &&
+	if (packet_sock_flag(po, PACKET_SOCK_RUNNING) &&
 	    match->type == type &&
 	    match->prot_hook.type == po->prot_hook.type &&
 	    match->prot_hook.dev == po->prot_hook.dev) {
@@ -3222,7 +3222,7 @@ static int packet_do_bind(struct sock *sk, const char *name, int ifindex,
 
 	if (need_rehook) {
 		dev_hold(dev);
-		if (po->running) {
+		if (packet_sock_flag(po, PACKET_SOCK_RUNNING)) {
 			rcu_read_unlock();
 			/* prevents packet_notifier() from calling
 			 * register_prot_hook()
@@ -3235,7 +3235,7 @@ static int packet_do_bind(struct sock *sk, const char *name, int ifindex,
 								 dev->ifindex);
 		}
 
-		BUG_ON(po->running);
+		BUG_ON(packet_sock_flag(po, PACKET_SOCK_RUNNING));
 		WRITE_ONCE(po->num, proto);
 		po->prot_hook.type = proto;
 
@@ -4159,7 +4159,7 @@ static int packet_notifier(struct notifier_block *this,
 		case NETDEV_DOWN:
 			if (dev->ifindex == po->ifindex) {
 				spin_lock(&po->bind_lock);
-				if (po->running) {
+				if (packet_sock_flag(po, PACKET_SOCK_RUNNING)) {
 					__unregister_prot_hook(sk, false);
 					sk->sk_err = ENETDOWN;
 					if (!sock_flag(sk, SOCK_DEAD))
@@ -4470,7 +4470,7 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
 
 	/* Detach socket from network */
 	spin_lock(&po->bind_lock);
-	was_running = po->running;
+	was_running = packet_sock_flag(po, PACKET_SOCK_RUNNING);
 	num = po->num;
 	if (was_running) {
 		WRITE_ONCE(po->num, 0);
@@ -4681,7 +4681,7 @@ static int packet_seq_show(struct seq_file *seq, void *v)
 			   s->sk_type,
 			   ntohs(READ_ONCE(po->num)),
 			   READ_ONCE(po->ifindex),
-			   po->running,
+			   packet_sock_flag(po, PACKET_SOCK_RUNNING),
 			   atomic_read(&s->sk_rmem_alloc),
 			   from_kuid_munged(seq_user_ns(seq), sock_i_uid(s)),
 			   sock_i_ino(s));
diff --git a/net/packet/diag.c b/net/packet/diag.c
index 56240aaf032b..de4ced5cf3e8 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -21,7 +21,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 	pinfo.pdi_tstamp = READ_ONCE(po->tp_tstamp);
 
 	pinfo.pdi_flags = 0;
-	if (po->running)
+	if (packet_sock_flag(po, PACKET_SOCK_RUNNING))
 		pinfo.pdi_flags |= PDI_RUNNING;
 	if (packet_sock_flag(po, PACKET_SOCK_AUXDATA))
 		pinfo.pdi_flags |= PDI_AUXDATA;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 2521176807f4..58f042c63172 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -117,7 +117,6 @@ struct packet_sock {
 	spinlock_t		bind_lock;
 	struct mutex		pg_vec_lock;
 	unsigned long		flags;
-	unsigned int		running;	/* bind_lock must be held */
 	int			pressure;
 	int			ifindex;	/* bound device		*/
 	__be16			num;
@@ -146,6 +145,7 @@ enum packet_sock_flags {
 	PACKET_SOCK_TX_HAS_OFF,
 	PACKET_SOCK_TP_LOSS,
 	PACKET_SOCK_HAS_VNET_HDR,
+	PACKET_SOCK_RUNNING,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From 791a3e9f1a86fe8eb09173c9788493b8b5c957f4 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 01:10:14 +0000
Subject: net/packet: convert po->pressure to an atomic flag

Not only this removes some READ_ONCE()/WRITE_ONCE(),
this also removes one integer.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 14 ++++++++------
 net/packet/internal.h  |  2 +-
 2 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ec446452bbe8..7b9367b233d3 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -1307,22 +1307,23 @@ static int __packet_rcv_has_room(const struct packet_sock *po,
 
 static int packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
 {
-	int pressure, ret;
+	bool pressure;
+	int ret;
 
 	ret = __packet_rcv_has_room(po, skb);
 	pressure = ret != ROOM_NORMAL;
 
-	if (READ_ONCE(po->pressure) != pressure)
-		WRITE_ONCE(po->pressure, pressure);
+	if (packet_sock_flag(po, PACKET_SOCK_PRESSURE) != pressure)
+		packet_sock_flag_set(po, PACKET_SOCK_PRESSURE, pressure);
 
 	return ret;
 }
 
 static void packet_rcv_try_clear_pressure(struct packet_sock *po)
 {
-	if (READ_ONCE(po->pressure) &&
+	if (packet_sock_flag(po, PACKET_SOCK_PRESSURE) &&
 	    __packet_rcv_has_room(po, NULL) == ROOM_NORMAL)
-		WRITE_ONCE(po->pressure,  0);
+		packet_sock_flag_set(po, PACKET_SOCK_PRESSURE, false);
 }
 
 static void packet_sock_destruct(struct sock *sk)
@@ -1409,7 +1410,8 @@ static unsigned int fanout_demux_rollover(struct packet_fanout *f,
 	i = j = min_t(int, po->rollover->sock, num - 1);
 	do {
 		po_next = pkt_sk(rcu_dereference(f->arr[i]));
-		if (po_next != po_skip && !READ_ONCE(po_next->pressure) &&
+		if (po_next != po_skip &&
+		    !packet_sock_flag(po_next, PACKET_SOCK_PRESSURE) &&
 		    packet_rcv_has_room(po_next, skb) == ROOM_NORMAL) {
 			if (i != j)
 				po->rollover->sock = i;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 58f042c63172..680703dbce5e 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -117,7 +117,6 @@ struct packet_sock {
 	spinlock_t		bind_lock;
 	struct mutex		pg_vec_lock;
 	unsigned long		flags;
-	int			pressure;
 	int			ifindex;	/* bound device		*/
 	__be16			num;
 	struct packet_rollover	*rollover;
@@ -146,6 +145,7 @@ enum packet_sock_flags {
 	PACKET_SOCK_TP_LOSS,
 	PACKET_SOCK_HAS_VNET_HDR,
 	PACKET_SOCK_RUNNING,
+	PACKET_SOCK_PRESSURE,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From 74bf6477c18b2904936763132e9224a41b8da13a Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 15 Mar 2023 21:49:13 -0700
Subject: netlink-specs: add partial specification for devlink

Devlink is quite complex but put in the very basics so we can
incrementally fill in the commands as needed.

$ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/devlink.yaml \
    --dump get

[{'bus-name': 'netdevsim',
  'dev-name': 'netdevsim1',
  'dev-stats': {'reload-stats': {'reload-action-info': {'reload-action': 1,
                                                        'reload-action-stats': {'reload-stats-entry': [{'reload-stats-limit': 0,
                                                                                                        'reload-stats-value': 0}]}}},
                'remote-reload-stats': {'reload-action-info': {'reload-action': 2,
                                                               'reload-action-stats': {'reload-stats-entry': [{'reload-stats-limit': 0,
                                                                                                               'reload-stats-value': 0},
                                                                                                              {'reload-stats-limit': 1,
                                                                                                               'reload-stats-value': 0}]}}}},
  'reload-failed': 0}]

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/netlink/specs/devlink.yaml | 198 +++++++++++++++++++++++++++++++
 1 file changed, 198 insertions(+)
 create mode 100644 Documentation/netlink/specs/devlink.yaml

diff --git a/Documentation/netlink/specs/devlink.yaml b/Documentation/netlink/specs/devlink.yaml
new file mode 100644
index 000000000000..90641668232e
--- /dev/null
+++ b/Documentation/netlink/specs/devlink.yaml
@@ -0,0 +1,198 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: devlink
+
+protocol: genetlink-legacy
+
+doc: Partial family for Devlink.
+
+attribute-sets:
+  -
+    name: devlink
+    attributes:
+      -
+        name: bus-name
+        type: string
+        value: 1
+      -
+        name: dev-name
+        type: string
+      -
+        name: port-index
+        type: u32
+
+      # TODO: fill in the attributes in between
+
+      -
+        name: info-driver-name
+        type: string
+        value: 98
+      -
+        name: info-serial-number
+        type: string
+      -
+        name: info-version-fixed
+        type: nest
+        multi-attr: true
+        nested-attributes: dl-info-version
+      -
+        name: info-version-running
+        type: nest
+        multi-attr: true
+        nested-attributes: dl-info-version
+      -
+        name: info-version-stored
+        type: nest
+        multi-attr: true
+        nested-attributes: dl-info-version
+      -
+        name: info-version-name
+        type: string
+      -
+        name: info-version-value
+        type: string
+
+      # TODO: fill in the attributes in between
+
+      -
+        name: reload-failed
+        type: u8
+        value: 136
+
+      # TODO: fill in the attributes in between
+
+      -
+        name: reload-action
+        type: u8
+        value: 153
+
+      # TODO: fill in the attributes in between
+
+      -
+        name: dev-stats
+        type: nest
+        value: 156
+        nested-attributes: dl-dev-stats
+      -
+        name: reload-stats
+        type: nest
+        nested-attributes: dl-reload-stats
+      -
+        name: reload-stats-entry
+        type: nest
+        multi-attr: true
+        nested-attributes: dl-reload-stats-entry
+      -
+        name: reload-stats-limit
+        type: u8
+      -
+        name: reload-stats-value
+        type: u32
+      -
+        name: remote-reload-stats
+        type: nest
+        nested-attributes: dl-reload-stats
+      -
+        name: reload-action-info
+        type: nest
+        nested-attributes: dl-reload-act-info
+      -
+        name: reload-action-stats
+        type: nest
+        nested-attributes: dl-reload-act-stats
+  -
+    name: dl-dev-stats
+    subset-of: devlink
+    attributes:
+      -
+        name: reload-stats
+        type: nest
+      -
+        name: remote-reload-stats
+        type: nest
+  -
+    name: dl-reload-stats
+    subset-of: devlink
+    attributes:
+      -
+        name: reload-action-info
+        type: nest
+  -
+    name: dl-reload-act-info
+    subset-of: devlink
+    attributes:
+      -
+        name: reload-action
+        type: u8
+      -
+        name: reload-action-stats
+        type: nest
+  -
+    name: dl-reload-act-stats
+    subset-of: devlink
+    attributes:
+      -
+        name: reload-stats-entry
+        type: nest
+  -
+    name: dl-reload-stats-entry
+    subset-of: devlink
+    attributes:
+      -
+        name: reload-stats-limit
+        type: u8
+      -
+        name: reload-stats-value
+        type: u32
+  -
+    name: dl-info-version
+    subset-of: devlink
+    attributes:
+      -
+        name: info-version-name
+        type: string
+      -
+        name: info-version-value
+        type: string
+
+operations:
+  enum-model: directional
+  list:
+    -
+      name: get
+      doc: Get devlink instances.
+      attribute-set: devlink
+
+      do:
+        request:
+          value: 1
+          attributes: &dev-id-attrs
+            - bus-name
+            - dev-name
+        reply:  &get-reply
+          value: 3
+          attributes:
+            - bus-name
+            - dev-name
+            - reload-failed
+            - reload-action
+            - dev-stats
+      dump:
+        reply: *get-reply
+
+      # TODO: fill in the operations in between
+
+    -
+      name: info-get
+      doc: Get device information, like driver name, hardware and firmware versions etc.
+      attribute-set: devlink
+
+      do:
+        request:
+          value: 51
+          attributes: *dev-id-attrs
+        reply:
+          value: 51
+          attributes:
+            - bus-name
+            - dev-name
-- 
cgit v1.2.3-70-g09d2


From 82b3297009b6831dfe47f0f38ed4043e39f58c9f Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 15 Mar 2023 21:50:27 -0700
Subject: netlink: specs: allow uapi-header in genetlink

Chuck wanted to put the UAPI header in linux/net/ which seems
reasonable, allow genetlink families to choose the location.
It doesn't really matter for non-C-like languages.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/netlink/genetlink-c.yaml      | 2 +-
 Documentation/netlink/genetlink-legacy.yaml | 2 +-
 Documentation/netlink/genetlink.yaml        | 3 +++
 3 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/Documentation/netlink/genetlink-c.yaml b/Documentation/netlink/genetlink-c.yaml
index f082a5ad7cf1..c83643d403b7 100644
--- a/Documentation/netlink/genetlink-c.yaml
+++ b/Documentation/netlink/genetlink-c.yaml
@@ -33,10 +33,10 @@ properties:
   protocol:
     description: Schema compatibility level. Default is "genetlink".
     enum: [ genetlink, genetlink-c ]
-  # Start genetlink-c
   uapi-header:
     description: Path to the uAPI header, default is linux/${family-name}.h
     type: string
+  # Start genetlink-c
   c-family-name:
     description: Name of the define for the family name.
     type: string
diff --git a/Documentation/netlink/genetlink-legacy.yaml b/Documentation/netlink/genetlink-legacy.yaml
index c6b8c77f7d12..792875dd7ed1 100644
--- a/Documentation/netlink/genetlink-legacy.yaml
+++ b/Documentation/netlink/genetlink-legacy.yaml
@@ -33,10 +33,10 @@ properties:
   protocol:
     description: Schema compatibility level. Default is "genetlink".
     enum: [ genetlink, genetlink-c, genetlink-legacy ] # Trim
-  # Start genetlink-c
   uapi-header:
     description: Path to the uAPI header, default is linux/${family-name}.h
     type: string
+  # Start genetlink-c
   c-family-name:
     description: Name of the define for the family name.
     type: string
diff --git a/Documentation/netlink/genetlink.yaml b/Documentation/netlink/genetlink.yaml
index b2d56ab9e615..8952e84ff207 100644
--- a/Documentation/netlink/genetlink.yaml
+++ b/Documentation/netlink/genetlink.yaml
@@ -33,6 +33,9 @@ properties:
   protocol:
     description: Schema compatibility level. Default is "genetlink".
     enum: [ genetlink ]
+  uapi-header:
+    description: Path to the uAPI header, default is linux/${family-name}.h
+    type: string
 
   definitions:
     description: List of type and constant definitions (enums, flags, defines).
-- 
cgit v1.2.3-70-g09d2


From abc17a11ed29b0471e428d86189acca8d1a213c6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:31:55 +0000
Subject: inet: preserve const qualifier in inet_sk()

We can change inet_sk() to propagate const qualifier of its argument.

This should avoid some potential errors caused by accidental
(const -> not_const) promotion.

Other helpers like tcp_sk(), udp_sk(), raw_sk() will be handled
in separate patch series.

v2: use container_of_const() as advised by Jakub and Linus

Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/netdev/20230315142841.3a2ac99a@kernel.org/
Link: https://lore.kernel.org/netdev/CAHk-=wiOf12nrYEF2vJMcucKjWPN-Ns_SW9fA7LwST_2Dzp7rw@mail.gmail.com/
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h     | 5 +----
 include/trace/events/sock.h | 4 ++--
 include/trace/events/tcp.h  | 2 +-
 net/ipv4/ip_output.c        | 5 +++--
 net/ipv6/ping.c             | 2 +-
 net/ipv6/udp.c              | 2 +-
 net/mptcp/sockopt.c         | 2 +-
 security/lsm_audit.c        | 4 ++--
 8 files changed, 12 insertions(+), 14 deletions(-)

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 51857117ac09..caa20a905531 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -305,10 +305,7 @@ static inline struct sock *skb_to_full_sk(const struct sk_buff *skb)
 	return sk_to_full_sk(skb->sk);
 }
 
-static inline struct inet_sock *inet_sk(const struct sock *sk)
-{
-	return (struct inet_sock *)sk;
-}
+#define inet_sk(ptr) container_of_const(ptr, struct inet_sock, sk)
 
 static inline void __inet_sk_copy_descendant(struct sock *sk_to,
 					     const struct sock *sk_from,
diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h
index 03d19fc562f8..fd206a6ab5b8 100644
--- a/include/trace/events/sock.h
+++ b/include/trace/events/sock.h
@@ -158,7 +158,7 @@ TRACE_EVENT(inet_sock_set_state,
 	),
 
 	TP_fast_assign(
-		struct inet_sock *inet = inet_sk(sk);
+		const struct inet_sock *inet = inet_sk(sk);
 		struct in6_addr *pin6;
 		__be32 *p32;
 
@@ -222,7 +222,7 @@ TRACE_EVENT(inet_sk_error_report,
 	),
 
 	TP_fast_assign(
-		struct inet_sock *inet = inet_sk(sk);
+		const struct inet_sock *inet = inet_sk(sk);
 		struct in6_addr *pin6;
 		__be32 *p32;
 
diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 901b440238d5..bf06db8d2046 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -67,7 +67,7 @@ DECLARE_EVENT_CLASS(tcp_event_sk_skb,
 	),
 
 	TP_fast_assign(
-		struct inet_sock *inet = inet_sk(sk);
+		const struct inet_sock *inet = inet_sk(sk);
 		__be32 *p32;
 
 		__entry->skbaddr = skb;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index e7bef36ce26f..cb04dbad9ea4 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -129,7 +129,8 @@ int ip_local_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(ip_local_out);
 
-static inline int ip_select_ttl(struct inet_sock *inet, struct dst_entry *dst)
+static inline int ip_select_ttl(const struct inet_sock *inet,
+				const struct dst_entry *dst)
 {
 	int ttl = inet->uc_ttl;
 
@@ -146,7 +147,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, const struct sock *sk,
 			  __be32 saddr, __be32 daddr, struct ip_options_rcu *opt,
 			  u8 tos)
 {
-	struct inet_sock *inet = inet_sk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
 	struct rtable *rt = skb_rtable(skb);
 	struct net *net = sock_net(sk);
 	struct iphdr *iph;
diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c
index 808983bc2ec9..c4835dbdfcff 100644
--- a/net/ipv6/ping.c
+++ b/net/ipv6/ping.c
@@ -237,7 +237,7 @@ static int ping_v6_seq_show(struct seq_file *seq, void *v)
 		seq_puts(seq, IPV6_SEQ_DGRAM_HEADER);
 	} else {
 		int bucket = ((struct ping_iter_state *) seq->private)->bucket;
-		struct inet_sock *inet = inet_sk(v);
+		struct inet_sock *inet = inet_sk((struct sock *)v);
 		__u16 srcp = ntohs(inet->inet_sport);
 		__u16 destp = ntohs(inet->inet_dport);
 		ip6_dgram_sock_seq_show(seq, v, srcp, destp, bucket);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 9fb2f33ee3a7..ab4ae886235a 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1708,7 +1708,7 @@ int udp6_seq_show(struct seq_file *seq, void *v)
 		seq_puts(seq, IPV6_SEQ_DGRAM_HEADER);
 	} else {
 		int bucket = ((struct udp_iter_state *)seq->private)->bucket;
-		struct inet_sock *inet = inet_sk(v);
+		const struct inet_sock *inet = inet_sk((const struct sock *)v);
 		__u16 srcp = ntohs(inet->inet_sport);
 		__u16 destp = ntohs(inet->inet_dport);
 		__ip6_dgram_sock_seq_show(seq, v, srcp, destp,
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c
index 8a9656248b0f..5cef4d3d21ac 100644
--- a/net/mptcp/sockopt.c
+++ b/net/mptcp/sockopt.c
@@ -1046,7 +1046,7 @@ static int mptcp_getsockopt_tcpinfo(struct mptcp_sock *msk, char __user *optval,
 
 static void mptcp_get_sub_addrs(const struct sock *sk, struct mptcp_subflow_addrs *a)
 {
-	struct inet_sock *inet = inet_sk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
 
 	memset(a, 0, sizeof(*a));
 
diff --git a/security/lsm_audit.c b/security/lsm_audit.c
index a7355b4b9bb8..00d3bdd386e2 100644
--- a/security/lsm_audit.c
+++ b/security/lsm_audit.c
@@ -317,7 +317,7 @@ static void dump_common_audit_data(struct audit_buffer *ab,
 
 			switch (sk->sk_family) {
 			case AF_INET: {
-				struct inet_sock *inet = inet_sk(sk);
+				const struct inet_sock *inet = inet_sk(sk);
 
 				print_ipv4_addr(ab, inet->inet_rcv_saddr,
 						inet->inet_sport,
@@ -329,7 +329,7 @@ static void dump_common_audit_data(struct audit_buffer *ab,
 			}
 #if IS_ENABLED(CONFIG_IPV6)
 			case AF_INET6: {
-				struct inet_sock *inet = inet_sk(sk);
+				const struct inet_sock *inet = inet_sk(sk);
 
 				print_ipv6_addr(ab, &sk->sk_v6_rcv_saddr,
 						inet->inet_sport,
-- 
cgit v1.2.3-70-g09d2


From 33e972bdf0b0aa208b67164c64eef3c307e4b303 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:31:56 +0000
Subject: ipv4: constify ip_mc_sf_allow() socket argument

This clarifies ip_mc_sf_allow() intent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h | 2 +-
 net/ipv4/igmp.c      | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index b19d3284551f..ebf4349a53af 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -122,7 +122,7 @@ extern int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf,
 			sockptr_t optval, sockptr_t optlen);
 extern int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 			sockptr_t optval, size_t offset);
-extern int ip_mc_sf_allow(struct sock *sk, __be32 local, __be32 rmt,
+extern int ip_mc_sf_allow(const struct sock *sk, __be32 local, __be32 rmt,
 			  int dif, int sdif);
 extern void ip_mc_init_dev(struct in_device *);
 extern void ip_mc_destroy_dev(struct in_device *);
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index c920aa9a62a9..48ff5f13e797 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -2638,10 +2638,10 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 /*
  * check if a multicast source filter allows delivery for a given <src,dst,intf>
  */
-int ip_mc_sf_allow(struct sock *sk, __be32 loc_addr, __be32 rmt_addr,
+int ip_mc_sf_allow(const struct sock *sk, __be32 loc_addr, __be32 rmt_addr,
 		   int dif, int sdif)
 {
-	struct inet_sock *inet = inet_sk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
 	struct ip_mc_socklist *pmc;
 	struct ip_sf_socklist *psl;
 	int i;
-- 
cgit v1.2.3-70-g09d2


From a0a989d3007538cd7dae9431d8ac47850ced3100 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:31:57 +0000
Subject: udp: constify __udp_is_mcast_sock() socket argument

This clarifies __udp_is_mcast_sock() intent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/udp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index dc8feb54d835..aa32afd871ee 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -578,12 +578,12 @@ struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport,
 EXPORT_SYMBOL_GPL(udp4_lib_lookup);
 #endif
 
-static inline bool __udp_is_mcast_sock(struct net *net, struct sock *sk,
+static inline bool __udp_is_mcast_sock(struct net *net, const struct sock *sk,
 				       __be16 loc_port, __be32 loc_addr,
 				       __be16 rmt_port, __be32 rmt_addr,
 				       int dif, int sdif, unsigned short hnum)
 {
-	struct inet_sock *inet = inet_sk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
 
 	if (!net_eq(sock_net(sk), net) ||
 	    udp_sk(sk)->udp_port_hash != hnum ||
-- 
cgit v1.2.3-70-g09d2


From 66eb554c6449cef8c1e1b814f74d13f264582591 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:31:58 +0000
Subject: ipv6: constify inet6_mc_check()

inet6_mc_check() is essentially a read-only function.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 2 +-
 net/ipv6/mcast.c       | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index c04f359655b8..82da55101b5a 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -223,7 +223,7 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex,
 		      const struct in6_addr *addr);
 void __ipv6_sock_mc_close(struct sock *sk);
 void ipv6_sock_mc_close(struct sock *sk);
-bool inet6_mc_check(struct sock *sk, const struct in6_addr *mc_addr,
+bool inet6_mc_check(const struct sock *sk, const struct in6_addr *mc_addr,
 		    const struct in6_addr *src_addr);
 
 int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr);
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 1c02160cf7a4..714cdc9e2b8e 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -627,12 +627,12 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 	return 0;
 }
 
-bool inet6_mc_check(struct sock *sk, const struct in6_addr *mc_addr,
+bool inet6_mc_check(const struct sock *sk, const struct in6_addr *mc_addr,
 		    const struct in6_addr *src_addr)
 {
-	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct ipv6_mc_socklist *mc;
-	struct ip6_sf_socklist *psl;
+	const struct ipv6_pinfo *np = inet6_sk(sk);
+	const struct ipv6_mc_socklist *mc;
+	const struct ip6_sf_socklist *psl;
 	bool rv = true;
 
 	rcu_read_lock();
-- 
cgit v1.2.3-70-g09d2


From dc3731bad8e133ca1893ae414fdebdf85a333495 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:31:59 +0000
Subject: udp6: constify __udp_v6_is_mcast_sock() socket argument

This clarifies __udp_v6_is_mcast_sock() intent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/udp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index ab4ae886235a..d350e57c4792 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -805,12 +805,12 @@ static int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	return 0;
 }
 
-static bool __udp_v6_is_mcast_sock(struct net *net, struct sock *sk,
+static bool __udp_v6_is_mcast_sock(struct net *net, const struct sock *sk,
 				   __be16 loc_port, const struct in6_addr *loc_addr,
 				   __be16 rmt_port, const struct in6_addr *rmt_addr,
 				   int dif, int sdif, unsigned short hnum)
 {
-	struct inet_sock *inet = inet_sk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
 
 	if (!net_eq(sock_net(sk), net))
 		return false;
-- 
cgit v1.2.3-70-g09d2


From db6af4fdb150b45e1ba6b295ccfd3df482e022d2 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:32:00 +0000
Subject: ipv6: raw: constify raw_v6_match() socket argument

This clarifies raw_v6_match() intent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/rawv6.h | 2 +-
 net/ipv6/raw.c      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/net/rawv6.h b/include/net/rawv6.h
index bc70909625f6..82810cbe3798 100644
--- a/include/net/rawv6.h
+++ b/include/net/rawv6.h
@@ -6,7 +6,7 @@
 #include <net/raw.h>
 
 extern struct raw_hashinfo raw_v6_hashinfo;
-bool raw_v6_match(struct net *net, struct sock *sk, unsigned short num,
+bool raw_v6_match(struct net *net, const struct sock *sk, unsigned short num,
 		  const struct in6_addr *loc_addr,
 		  const struct in6_addr *rmt_addr, int dif, int sdif);
 
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index bac9ba747bde..6ac2f2690c44 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -64,7 +64,7 @@
 struct raw_hashinfo raw_v6_hashinfo;
 EXPORT_SYMBOL_GPL(raw_v6_hashinfo);
 
-bool raw_v6_match(struct net *net, struct sock *sk, unsigned short num,
+bool raw_v6_match(struct net *net, const struct sock *sk, unsigned short num,
 		  const struct in6_addr *loc_addr,
 		  const struct in6_addr *rmt_addr, int dif, int sdif)
 {
-- 
cgit v1.2.3-70-g09d2


From 0a8c2568209ee0c3392593c7c5c7fe41c625a383 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:32:01 +0000
Subject: ipv4: raw: constify raw_v4_match() socket argument

This clarifies raw_v4_match() intent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h | 2 +-
 net/ipv4/raw.c    | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/net/raw.h b/include/net/raw.h
index 2c004c20ed99..7ad15830cf38 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -22,7 +22,7 @@
 extern struct proto raw_prot;
 
 extern struct raw_hashinfo raw_v4_hashinfo;
-bool raw_v4_match(struct net *net, struct sock *sk, unsigned short num,
+bool raw_v4_match(struct net *net, const struct sock *sk, unsigned short num,
 		  __be32 raddr, __be32 laddr, int dif, int sdif);
 
 int raw_abort(struct sock *sk, int err);
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 94df935ee0c5..3cf68695b40d 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -116,10 +116,10 @@ void raw_unhash_sk(struct sock *sk)
 }
 EXPORT_SYMBOL_GPL(raw_unhash_sk);
 
-bool raw_v4_match(struct net *net, struct sock *sk, unsigned short num,
+bool raw_v4_match(struct net *net, const struct sock *sk, unsigned short num,
 		  __be32 raddr, __be32 laddr, int dif, int sdif)
 {
-	struct inet_sock *inet = inet_sk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
 
 	if (net_eq(sock_net(sk), net) && inet->inet_num == num	&&
 	    !(inet->inet_daddr && inet->inet_daddr != raddr) 	&&
-- 
cgit v1.2.3-70-g09d2


From 736c8b52c8ada4ca93a59f7d597c5650ebc45921 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Mar 2023 15:32:02 +0000
Subject: inet_diag: constify raw_lookup() socket argument

Now both raw_v4_match() and raw_v6_match() accept a const socket,
raw_lookup() can do the same to clarify its role.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/raw_diag.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/raw_diag.c b/net/ipv4/raw_diag.c
index 999321834b94..bca49a844f01 100644
--- a/net/ipv4/raw_diag.c
+++ b/net/ipv4/raw_diag.c
@@ -34,7 +34,7 @@ raw_get_hashinfo(const struct inet_diag_req_v2 *r)
  * use helper to figure it out.
  */
 
-static bool raw_lookup(struct net *net, struct sock *sk,
+static bool raw_lookup(struct net *net, const struct sock *sk,
 		       const struct inet_diag_req_v2 *req)
 {
 	struct inet_diag_req_raw *r = (void *)req;
-- 
cgit v1.2.3-70-g09d2


From 79a22238b4f22c45cadd3b4040d644f4de320d1b Mon Sep 17 00:00:00 2001
From: Kai Shen <KaiShen@linux.alibaba.com>
Date: Fri, 17 Mar 2023 03:21:32 +0000
Subject: net/smc: Use percpu ref for wr tx reference

The refcount wr_tx_refcnt may cause cache thrashing problems among
cores and we can use percpu ref to mitigate this issue here. We
gain some performance improvement with percpu ref here on our
customized smc-r verion. Applying cache alignment may also mitigate
this problem but it seem more reasonable to use percpu ref here.
We can also replace wr_reg_refcnt with one percpu reference like
wr_tx_refcnt.

redis-benchmark on smc-r with atomic wr_tx_refcnt:
SET: 525707.06 requests per second, p50=0.087 msec
GET: 554877.38 requests per second, p50=0.087 msec

redis-benchmark on the percpu_ref version:
SET: 540482.06 requests per second, p50=0.087 msec
GET: 570711.12 requests per second, p50=0.079 msec

Cases are like "redis-benchmark -h x.x.x.x -q -t set,get -P 1 -n
5000000 -c 50 -d 10 --threads 4".

Signed-off-by: Kai Shen <KaiShen@linux.alibaba.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/smc_core.h | 10 ++++++++--
 net/smc/smc_wr.c   | 35 ++++++++++++++++++++++++++++-------
 net/smc/smc_wr.h   |  5 ++---
 3 files changed, 38 insertions(+), 12 deletions(-)

diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h
index 08b457c2d294..1645fba0d2d3 100644
--- a/net/smc/smc_core.h
+++ b/net/smc/smc_core.h
@@ -106,7 +106,10 @@ struct smc_link {
 	unsigned long		*wr_tx_mask;	/* bit mask of used indexes */
 	u32			wr_tx_cnt;	/* number of WR send buffers */
 	wait_queue_head_t	wr_tx_wait;	/* wait for free WR send buf */
-	atomic_t		wr_tx_refcnt;	/* tx refs to link */
+	struct {
+		struct percpu_ref	wr_tx_refs;
+	} ____cacheline_aligned_in_smp;
+	struct completion	tx_ref_comp;
 
 	struct smc_wr_buf	*wr_rx_bufs;	/* WR recv payload buffers */
 	struct ib_recv_wr	*wr_rx_ibs;	/* WR recv meta data */
@@ -122,7 +125,10 @@ struct smc_link {
 
 	struct ib_reg_wr	wr_reg;		/* WR register memory region */
 	wait_queue_head_t	wr_reg_wait;	/* wait for wr_reg result */
-	atomic_t		wr_reg_refcnt;	/* reg refs to link */
+	struct {
+		struct percpu_ref	wr_reg_refs;
+	} ____cacheline_aligned_in_smp;
+	struct completion	reg_ref_comp;
 	enum smc_wr_reg_state	wr_reg_state;	/* state of wr_reg request */
 
 	u8			gid[SMC_GID_SIZE];/* gid matching used vlan id*/
diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c
index b0678a417e09..0021065a600a 100644
--- a/net/smc/smc_wr.c
+++ b/net/smc/smc_wr.c
@@ -377,12 +377,11 @@ int smc_wr_reg_send(struct smc_link *link, struct ib_mr *mr)
 	if (rc)
 		return rc;
 
-	atomic_inc(&link->wr_reg_refcnt);
+	percpu_ref_get(&link->wr_reg_refs);
 	rc = wait_event_interruptible_timeout(link->wr_reg_wait,
 					      (link->wr_reg_state != POSTED),
 					      SMC_WR_REG_MR_WAIT_TIME);
-	if (atomic_dec_and_test(&link->wr_reg_refcnt))
-		wake_up_all(&link->wr_reg_wait);
+	percpu_ref_put(&link->wr_reg_refs);
 	if (!rc) {
 		/* timeout - terminate link */
 		smcr_link_down_cond_sched(link);
@@ -647,8 +646,10 @@ void smc_wr_free_link(struct smc_link *lnk)
 	smc_wr_wakeup_tx_wait(lnk);
 
 	smc_wr_tx_wait_no_pending_sends(lnk);
-	wait_event(lnk->wr_reg_wait, (!atomic_read(&lnk->wr_reg_refcnt)));
-	wait_event(lnk->wr_tx_wait, (!atomic_read(&lnk->wr_tx_refcnt)));
+	percpu_ref_kill(&lnk->wr_reg_refs);
+	wait_for_completion(&lnk->reg_ref_comp);
+	percpu_ref_kill(&lnk->wr_tx_refs);
+	wait_for_completion(&lnk->tx_ref_comp);
 
 	if (lnk->wr_rx_dma_addr) {
 		ib_dma_unmap_single(ibdev, lnk->wr_rx_dma_addr,
@@ -847,6 +848,20 @@ void smc_wr_add_dev(struct smc_ib_device *smcibdev)
 	tasklet_setup(&smcibdev->send_tasklet, smc_wr_tx_tasklet_fn);
 }
 
+static void smcr_wr_tx_refs_free(struct percpu_ref *ref)
+{
+	struct smc_link *lnk = container_of(ref, struct smc_link, wr_tx_refs);
+
+	complete(&lnk->tx_ref_comp);
+}
+
+static void smcr_wr_reg_refs_free(struct percpu_ref *ref)
+{
+	struct smc_link *lnk = container_of(ref, struct smc_link, wr_reg_refs);
+
+	complete(&lnk->reg_ref_comp);
+}
+
 int smc_wr_create_link(struct smc_link *lnk)
 {
 	struct ib_device *ibdev = lnk->smcibdev->ibdev;
@@ -890,9 +905,15 @@ int smc_wr_create_link(struct smc_link *lnk)
 	smc_wr_init_sge(lnk);
 	bitmap_zero(lnk->wr_tx_mask, SMC_WR_BUF_CNT);
 	init_waitqueue_head(&lnk->wr_tx_wait);
-	atomic_set(&lnk->wr_tx_refcnt, 0);
+	rc = percpu_ref_init(&lnk->wr_tx_refs, smcr_wr_tx_refs_free, 0, GFP_KERNEL);
+	if (rc)
+		goto dma_unmap;
+	init_completion(&lnk->tx_ref_comp);
 	init_waitqueue_head(&lnk->wr_reg_wait);
-	atomic_set(&lnk->wr_reg_refcnt, 0);
+	rc = percpu_ref_init(&lnk->wr_reg_refs, smcr_wr_reg_refs_free, 0, GFP_KERNEL);
+	if (rc)
+		goto dma_unmap;
+	init_completion(&lnk->reg_ref_comp);
 	init_waitqueue_head(&lnk->wr_rx_empty_wait);
 	return rc;
 
diff --git a/net/smc/smc_wr.h b/net/smc/smc_wr.h
index 45e9b894d3f8..f3008dda222a 100644
--- a/net/smc/smc_wr.h
+++ b/net/smc/smc_wr.h
@@ -63,14 +63,13 @@ static inline bool smc_wr_tx_link_hold(struct smc_link *link)
 {
 	if (!smc_link_sendable(link))
 		return false;
-	atomic_inc(&link->wr_tx_refcnt);
+	percpu_ref_get(&link->wr_tx_refs);
 	return true;
 }
 
 static inline void smc_wr_tx_link_put(struct smc_link *link)
 {
-	if (atomic_dec_and_test(&link->wr_tx_refcnt))
-		wake_up_all(&link->wr_tx_wait);
+	percpu_ref_put(&link->wr_tx_refs);
 }
 
 static inline void smc_wr_drain_cq(struct smc_link *lnk)
-- 
cgit v1.2.3-70-g09d2


From bd5314f8dd2d41330eecb60f0490c3fcfe1fc99d Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Fri, 17 Mar 2023 10:56:01 +0100
Subject: kallsyms, bpf: Move find_kallsyms_symbol_value out of internal header

Moving find_kallsyms_symbol_value from kernel/module/internal.h to
include/linux/module.h. The reason is that internal.h is not prepared to
be included when CONFIG_MODULES=n. find_kallsyms_symbol_value is used by
kernel/bpf/verifier.c and including internal.h from it (without modules)
leads into a compilation error:

  In file included from ../include/linux/container_of.h:5,
                   from ../include/linux/list.h:5,
                   from ../include/linux/timer.h:5,
                   from ../include/linux/workqueue.h:9,
                   from ../include/linux/bpf.h:10,
                   from ../include/linux/bpf-cgroup.h:5,
                   from ../kernel/bpf/verifier.c:7:
  ../kernel/bpf/../module/internal.h: In function 'mod_find':
  ../include/linux/container_of.h:20:54: error: invalid use of undefined type 'struct module'
     20 |         static_assert(__same_type(*(ptr), ((type *)0)->member) ||       \
        |                                                      ^~
  [...]

This patch fixes the above error.

Fixes: 31bf1dbccfb0 ("bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Viktor Malik <vmalik@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/oe-kbuild-all/202303161404.OrmfCy09-lkp@intel.com/
Link: https://lore.kernel.org/bpf/20230317095601.386738-1-vmalik@redhat.com
---
 include/linux/module.h   | 8 ++++++++
 kernel/bpf/verifier.c    | 2 +-
 kernel/module/internal.h | 6 ------
 3 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/include/linux/module.h b/include/linux/module.h
index 4435ad9439ab..41cfd3be57e5 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -616,6 +616,8 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
 /* Look for this name: can be of form module:name. */
 unsigned long module_kallsyms_lookup_name(const char *name);
 
+unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name);
+
 extern void __noreturn __module_put_and_kthread_exit(struct module *mod,
 			long code);
 #define module_put_and_kthread_exit(code) __module_put_and_kthread_exit(THIS_MODULE, code)
@@ -796,6 +798,12 @@ static inline unsigned long module_kallsyms_lookup_name(const char *name)
 	return 0;
 }
 
+static inline unsigned long find_kallsyms_symbol_value(struct module *mod,
+						       const char *name)
+{
+	return 0;
+}
+
 static inline int register_module_notifier(struct notifier_block *nb)
 {
 	/* no events will happen anyway, so this can always succeed */
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d62b7127ff2a..99394a2f7ee4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -24,7 +24,7 @@
 #include <linux/bpf_lsm.h>
 #include <linux/btf_ids.h>
 #include <linux/poison.h>
-#include "../module/internal.h"
+#include <linux/module.h>
 
 #include "disasm.h"
 
diff --git a/kernel/module/internal.h b/kernel/module/internal.h
index 5c9170f9135c..1c877561a7d2 100644
--- a/kernel/module/internal.h
+++ b/kernel/module/internal.h
@@ -246,7 +246,6 @@ static inline void kmemleak_load_module(const struct module *mod,
 void init_build_id(struct module *mod, const struct load_info *info);
 void layout_symtab(struct module *mod, struct load_info *info);
 void add_kallsyms(struct module *mod, const struct load_info *info);
-unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name);
 
 static inline bool sect_empty(const Elf_Shdr *sect)
 {
@@ -256,11 +255,6 @@ static inline bool sect_empty(const Elf_Shdr *sect)
 static inline void init_build_id(struct module *mod, const struct load_info *info) { }
 static inline void layout_symtab(struct module *mod, struct load_info *info) { }
 static inline void add_kallsyms(struct module *mod, const struct load_info *info) { }
-static inline unsigned long find_kallsyms_symbol_value(struct module *mod,
-						       const char *name)
-{
-	return 0;
-}
 #endif /* CONFIG_KALLSYMS */
 
 #ifdef CONFIG_SYSFS
-- 
cgit v1.2.3-70-g09d2


From 0f10f647f45545004ea50b73a7a7c5c3309ff286 Mon Sep 17 00:00:00 2001
From: Bagas Sanjaya <bagasdotme@gmail.com>
Date: Tue, 14 Mar 2023 14:44:49 +0700
Subject: bpf, docs: Use internal linking for link to netdev subsystem doc

Commit d56b0c461d19da ("bpf, docs: Fix link to netdev-FAQ target")
attempts to fix linking problem to undefined "netdev-FAQ" label
introduced in 287f4fa99a5281 ("docs: Update references to netdev-FAQ")
by changing internal cross reference to netdev subsystem documentation
(Documentation/process/maintainer-netdev.rst) to external one at
docs.kernel.org. However, the linking problem is still not
resolved, as the generated link points to non-existent netdev-FAQ
section of the external doc, which when clicked, will instead going
to the top of the doc.

Revert back to internal linking by simply mention the doc path while
massaging the leading text to the link, since the netdev subsystem
doc contains no FAQs but rather general information about the subsystem.

Fixes: d56b0c461d19 ("bpf, docs: Fix link to netdev-FAQ target")
Fixes: 287f4fa99a52 ("docs: Update references to netdev-FAQ")
Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230314074449.23620-1-bagasdotme@gmail.com
---
 Documentation/bpf/bpf_devel_QA.rst | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/Documentation/bpf/bpf_devel_QA.rst b/Documentation/bpf/bpf_devel_QA.rst
index 5f5f9ccc3862..e151e61dff38 100644
--- a/Documentation/bpf/bpf_devel_QA.rst
+++ b/Documentation/bpf/bpf_devel_QA.rst
@@ -128,7 +128,8 @@ into the bpf-next tree will make their way into net-next tree. net and
 net-next are both run by David S. Miller. From there, they will go
 into the kernel mainline tree run by Linus Torvalds. To read up on the
 process of net and net-next being merged into the mainline tree, see
-the `netdev-FAQ`_.
+the documentation on netdev subsystem at
+Documentation/process/maintainer-netdev.rst.
 
 
 
@@ -147,7 +148,8 @@ request)::
 Q: How do I indicate which tree (bpf vs. bpf-next) my patch should be applied to?
 ---------------------------------------------------------------------------------
 
-A: The process is the very same as described in the `netdev-FAQ`_,
+A: The process is the very same as described in the netdev subsystem
+documentation at Documentation/process/maintainer-netdev.rst,
 so please read up on it. The subject line must indicate whether the
 patch is a fix or rather "next-like" content in order to let the
 maintainers know whether it is targeted at bpf or bpf-next.
@@ -206,8 +208,9 @@ ii) run extensive BPF test suite and
 Once the BPF pull request was accepted by David S. Miller, then
 the patches end up in net or net-next tree, respectively, and
 make their way from there further into mainline. Again, see the
-`netdev-FAQ`_ for additional information e.g. on how often they are
-merged to mainline.
+documentation for netdev subsystem at
+Documentation/process/maintainer-netdev.rst for additional information
+e.g. on how often they are merged to mainline.
 
 Q: How long do I need to wait for feedback on my BPF patches?
 -------------------------------------------------------------
@@ -230,7 +233,8 @@ Q: Are patches applied to bpf-next when the merge window is open?
 -----------------------------------------------------------------
 A: For the time when the merge window is open, bpf-next will not be
 processed. This is roughly analogous to net-next patch processing,
-so feel free to read up on the `netdev-FAQ`_ about further details.
+so feel free to read up on the netdev docs at
+Documentation/process/maintainer-netdev.rst about further details.
 
 During those two weeks of merge window, we might ask you to resend
 your patch series once bpf-next is open again. Once Linus released
@@ -394,7 +398,8 @@ netdev kernel mailing list in Cc and ask for the fix to be queued up:
   netdev@vger.kernel.org
 
 The process in general is the same as on netdev itself, see also the
-`netdev-FAQ`_.
+the documentation on networking subsystem at
+Documentation/process/maintainer-netdev.rst.
 
 Q: Do you also backport to kernels not currently maintained as stable?
 ----------------------------------------------------------------------
@@ -410,7 +415,7 @@ Q: The BPF patch I am about to submit needs to go to stable as well
 What should I do?
 
 A: The same rules apply as with netdev patch submissions in general, see
-the `netdev-FAQ`_.
+the netdev docs at Documentation/process/maintainer-netdev.rst.
 
 Never add "``Cc: stable@vger.kernel.org``" to the patch description, but
 ask the BPF maintainers to queue the patches instead. This can be done
@@ -685,7 +690,6 @@ when:
 
 .. Links
 .. _Documentation/process/: https://www.kernel.org/doc/html/latest/process/
-.. _netdev-FAQ: https://www.kernel.org/doc/html/latest/process/maintainer-netdev.html
 .. _selftests:
    https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/testing/selftests/bpf/
 .. _Documentation/dev-tools/kselftest.rst:
-- 
cgit v1.2.3-70-g09d2


From 58aa2afbb1e61fcf35bfcc819952a3c13d9f9203 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Fri, 17 Mar 2023 13:19:17 -0700
Subject: bpf: Allow ld_imm64 instruction to point to kfunc.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allow ld_imm64 insn with BPF_PSEUDO_BTF_ID to hold the address of kfunc. The
ld_imm64 pointing to a valid kfunc will be seen as non-null PTR_TO_MEM by
is_branch_taken() logic of the verifier, while libbpf will resolve address to
unknown kfunc as ld_imm64 reg, 0 which will also be recognized by
is_branch_taken() and the verifier will proceed dead code elimination. BPF
programs can use this logic to detect at load time whether kfunc is present in
the kernel with bpf_ksym_exists() macro that is introduced in the next patches.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20230317201920.62030-2-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 99394a2f7ee4..8bc44f5dc5b6 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -15952,8 +15952,8 @@ static int check_pseudo_btf_id(struct bpf_verifier_env *env,
 		goto err_put;
 	}
 
-	if (!btf_type_is_var(t)) {
-		verbose(env, "pseudo btf_id %d in ldimm64 isn't KIND_VAR.\n", id);
+	if (!btf_type_is_var(t) && !btf_type_is_func(t)) {
+		verbose(env, "pseudo btf_id %d in ldimm64 isn't KIND_VAR or KIND_FUNC\n", id);
 		err = -EINVAL;
 		goto err_put;
 	}
@@ -15966,6 +15966,14 @@ static int check_pseudo_btf_id(struct bpf_verifier_env *env,
 		err = -ENOENT;
 		goto err_put;
 	}
+	insn[0].imm = (u32)addr;
+	insn[1].imm = addr >> 32;
+
+	if (btf_type_is_func(t)) {
+		aux->btf_var.reg_type = PTR_TO_MEM | MEM_RDONLY;
+		aux->btf_var.mem_size = 0;
+		goto check_btf;
+	}
 
 	datasec_id = find_btf_percpu_datasec(btf);
 	if (datasec_id > 0) {
@@ -15978,9 +15986,6 @@ static int check_pseudo_btf_id(struct bpf_verifier_env *env,
 		}
 	}
 
-	insn[0].imm = (u32)addr;
-	insn[1].imm = addr >> 32;
-
 	type = t->type;
 	t = btf_type_skip_modifiers(btf, type, NULL);
 	if (percpu) {
@@ -16008,7 +16013,7 @@ static int check_pseudo_btf_id(struct bpf_verifier_env *env,
 		aux->btf_var.btf = btf;
 		aux->btf_var.btf_id = type;
 	}
-
+check_btf:
 	/* check whether we recorded this BTF (and maybe module) already */
 	for (i = 0; i < env->used_btf_cnt; i++) {
 		if (env->used_btfs[i].btf == btf) {
-- 
cgit v1.2.3-70-g09d2


From 5fc13ad59b60708e52932188854d4d5bf2b0e10e Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Fri, 17 Mar 2023 13:19:18 -0700
Subject: libbpf: Fix relocation of kfunc ksym in ld_imm64 insn.

void *p = kfunc; -> generates ld_imm64 insn.
kfunc() -> generates bpf_call insn.

libbpf patches bpf_call insn correctly while only btf_id part of ld_imm64 is
set in the former case. Which means that pointers to kfuncs in modules are not
patched correctly and the verifier rejects load of such programs due to btf_id
being out of range. Fix libbpf to patch ld_imm64 for kfunc.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230317201920.62030-3-alexei.starovoitov@gmail.com
---
 tools/lib/bpf/libbpf.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index a557718401e4..4c34fbd7b5be 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -7533,6 +7533,12 @@ static int bpf_object__resolve_ksym_func_btf_id(struct bpf_object *obj,
 	ext->is_set = true;
 	ext->ksym.kernel_btf_id = kfunc_id;
 	ext->ksym.btf_fd_idx = mod_btf ? mod_btf->fd_array_idx : 0;
+	/* Also set kernel_btf_obj_fd to make sure that bpf_object__relocate_data()
+	 * populates FD into ld_imm64 insn when it's used to point to kfunc.
+	 * {kernel_btf_id, btf_fd_idx} -> fixup bpf_call.
+	 * {kernel_btf_id, kernel_btf_obj_fd} -> fixup ld_imm64.
+	 */
+	ext->ksym.kernel_btf_obj_fd = mod_btf ? mod_btf->fd : 0;
 	pr_debug("extern (func ksym) '%s': resolved to kernel [%d]\n",
 		 ext->name, kfunc_id);
 
-- 
cgit v1.2.3-70-g09d2


From 5cbd3fe3a91df46ea201cc5d8ab4e390332ec26e Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Fri, 17 Mar 2023 13:19:19 -0700
Subject: libbpf: Introduce bpf_ksym_exists() macro.

Introduce bpf_ksym_exists() macro that can be used by BPF programs
to detect at load time whether particular ksym (either variable or kfunc)
is present in the kernel.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230317201920.62030-4-alexei.starovoitov@gmail.com
---
 tools/lib/bpf/bpf_helpers.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index 7d12d3e620cc..e7e1a8acc299 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -177,6 +177,11 @@ enum libbpf_tristate {
 #define __kptr_untrusted __attribute__((btf_type_tag("kptr_untrusted")))
 #define __kptr __attribute__((btf_type_tag("kptr")))
 
+#define bpf_ksym_exists(sym) ({									\
+	_Static_assert(!__builtin_constant_p(!!sym), #sym " should be marked as __weak");	\
+	!!sym;											\
+})
+
 #ifndef ___bpf_concat
 #define ___bpf_concat(a, b) a ## b
 #endif
-- 
cgit v1.2.3-70-g09d2


From 95fdf6e313a981b0729886f86916190cb418b04c Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Fri, 17 Mar 2023 13:19:20 -0700
Subject: selftests/bpf: Add test for bpf_ksym_exists().
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add load and run time test for bpf_ksym_exists() and check that the verifier
performs dead code elimination for non-existing kfunc.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20230317201920.62030-5-alexei.starovoitov@gmail.com
---
 .../testing/selftests/bpf/progs/task_kfunc_success.c | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index 4f61596b0242..cfa7f12b84e8 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -17,6 +17,10 @@ int err, pid;
  *         TP_PROTO(struct task_struct *p, u64 clone_flags)
  */
 
+struct task_struct *bpf_task_acquire(struct task_struct *p) __ksym __weak;
+void invalid_kfunc(void) __ksym __weak;
+void bpf_testmod_test_mod_kfunc(int i) __ksym __weak;
+
 static bool is_test_kfunc_task(void)
 {
 	int cur_pid = bpf_get_current_pid_tgid() >> 32;
@@ -26,7 +30,21 @@ static bool is_test_kfunc_task(void)
 
 static int test_acquire_release(struct task_struct *task)
 {
-	struct task_struct *acquired;
+	struct task_struct *acquired = NULL;
+
+	if (!bpf_ksym_exists(bpf_task_acquire)) {
+		err = 3;
+		return 0;
+	}
+	if (!bpf_ksym_exists(bpf_testmod_test_mod_kfunc)) {
+		err = 4;
+		return 0;
+	}
+	if (bpf_ksym_exists(invalid_kfunc)) {
+		/* the verifier's dead code elimination should remove this */
+		err = 5;
+		asm volatile ("goto -1"); /* for (;;); */
+	}
 
 	acquired = bpf_task_acquire(task);
 	bpf_task_release(acquired);
-- 
cgit v1.2.3-70-g09d2


From 2be7aa76cc69633930fb747e1d85d33a63a60c02 Mon Sep 17 00:00:00 2001
From: Manu Bretelle <chantr4@gmail.com>
Date: Fri, 17 Mar 2023 09:32:56 -0700
Subject: selftests/bpf: Add --json-summary option to test_progs

Currently, test_progs outputs all stdout/stderr as it runs, and when it
is done, prints a summary.

It is non-trivial for tooling to parse that output and extract meaningful
information from it.

This change adds a new option, `--json-summary`/`-J` that let the caller
specify a file where `test_progs{,-no_alu32}` can write a summary of the
run in a json format that can later be parsed by tooling.

Currently, it creates a summary section with successes/skipped/failures
followed by a list of failed tests and subtests.

A test contains the following fields:
- name: the name of the test
- number: the number of the test
- message: the log message that was printed by the test.
- failed: A boolean indicating whether the test failed or not. Currently
we only output failed tests, but in the future, successful tests could
be added.
- subtests: A list of subtests associated with this test.

A subtest contains the following fields:
- name: same as above
- number: sanme as above
- message: the log message that was printed by the subtest.
- failed: same as above but for the subtest

An example run and json content below:
```
$ sudo ./test_progs -a $(grep -v '^#' ./DENYLIST.aarch64 | awk '{print
$1","}' | tr -d '\n') -j -J /tmp/test_progs.json
$ jq < /tmp/test_progs.json | head -n 30
{
  "success": 29,
  "success_subtest": 23,
  "skipped": 3,
  "failed": 28,
  "results": [
    {
      "name": "bpf_cookie",
      "number": 10,
      "message": "test_bpf_cookie:PASS:skel_open 0 nsec\n",
      "failed": true,
      "subtests": [
        {
          "name": "multi_kprobe_link_api",
          "number": 2,
          "message": "kprobe_multi_link_api_subtest:PASS:load_kallsyms 0 nsec\nlibbpf: extern 'bpf_testmod_fentry_test1' (strong): not resolved\nlibbpf: failed to load object 'kprobe_multi'\nlibbpf: failed to load BPF skeleton 'kprobe_multi': -3\nkprobe_multi_link_api_subtest:FAIL:fentry_raw_skel_load unexpected error: -3\n",
          "failed": true
        },
        {
          "name": "multi_kprobe_attach_api",
          "number": 3,
          "message": "libbpf: extern 'bpf_testmod_fentry_test1' (strong): not resolved\nlibbpf: failed to load object 'kprobe_multi'\nlibbpf: failed to load BPF skeleton 'kprobe_multi': -3\nkprobe_multi_attach_api_subtest:FAIL:fentry_raw_skel_load unexpected error: -3\n",
          "failed": true
        },
        {
          "name": "lsm",
          "number": 8,
          "message": "lsm_subtest:PASS:lsm.link_create 0 nsec\nlsm_subtest:FAIL:stack_mprotect unexpected stack_mprotect: actual 0 != expected -1\n",
          "failed": true
        }
```

The file can then be used to print a summary of the test run and list of
failing tests/subtests:

```
$ jq -r < /tmp/test_progs.json '"Success: \(.success)/\(.success_subtest), Skipped: \(.skipped), Failed: \(.failed)"'

Success: 29/23, Skipped: 3, Failed: 28
$ jq -r < /tmp/test_progs.json '.results | map([
    if .failed then "#\(.number) \(.name)" else empty end,
    (
        . as {name: $tname, number: $tnum} | .subtests | map(
            if .failed then "#\($tnum)/\(.number) \($tname)/\(.name)" else empty end
        )
    )
]) | flatten | .[]' | head -n 20
 #10 bpf_cookie
 #10/2 bpf_cookie/multi_kprobe_link_api
 #10/3 bpf_cookie/multi_kprobe_attach_api
 #10/8 bpf_cookie/lsm
 #15 bpf_mod_race
 #15/1 bpf_mod_race/ksym (used_btfs UAF)
 #15/2 bpf_mod_race/kfunc (kfunc_btf_tab UAF)
 #36 cgroup_hierarchical_stats
 #61 deny_namespace
 #61/1 deny_namespace/unpriv_userns_create_no_bpf
 #73 fexit_stress
 #83 get_func_ip_test
 #99 kfunc_dynptr_param
 #99/1 kfunc_dynptr_param/dynptr_data_null
 #99/4 kfunc_dynptr_param/dynptr_data_null
 #100 kprobe_multi_bench_attach
 #100/1 kprobe_multi_bench_attach/kernel
 #100/2 kprobe_multi_bench_attach/modules
 #101 kprobe_multi_test
 #101/1 kprobe_multi_test/skel_api
```

Signed-off-by: Manu Bretelle <chantr4@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230317163256.3809328-1-chantr4@gmail.com
---
 tools/testing/selftests/bpf/Makefile      |  4 +-
 tools/testing/selftests/bpf/json_writer.c |  1 +
 tools/testing/selftests/bpf/json_writer.h |  1 +
 tools/testing/selftests/bpf/test_progs.c  | 83 +++++++++++++++++++++++++++++--
 tools/testing/selftests/bpf/test_progs.h  |  1 +
 5 files changed, 84 insertions(+), 6 deletions(-)
 create mode 120000 tools/testing/selftests/bpf/json_writer.c
 create mode 120000 tools/testing/selftests/bpf/json_writer.h

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 55811c448eb7..fc092582d16d 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -234,6 +234,7 @@ $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED): $(BPFOBJ)
 CGROUP_HELPERS	:= $(OUTPUT)/cgroup_helpers.o
 TESTING_HELPERS	:= $(OUTPUT)/testing_helpers.o
 TRACE_HELPERS	:= $(OUTPUT)/trace_helpers.o
+JSON_WRITER		:= $(OUTPUT)/json_writer.o
 CAP_HELPERS	:= $(OUTPUT)/cap_helpers.o
 
 $(OUTPUT)/test_dev_cgroup: $(CGROUP_HELPERS) $(TESTING_HELPERS)
@@ -559,7 +560,8 @@ TRUNNER_BPF_PROGS_DIR := progs
 TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c	\
 			 network_helpers.c testing_helpers.c		\
 			 btf_helpers.c flow_dissector_load.h		\
-			 cap_helpers.c test_loader.c xsk.c disasm.c
+			 cap_helpers.c test_loader.c xsk.c disasm.c \
+			 json_writer.c
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read $(OUTPUT)/bpf_testmod.ko	\
 		       $(OUTPUT)/liburandom_read.so			\
 		       $(OUTPUT)/xdp_synproxy				\
diff --git a/tools/testing/selftests/bpf/json_writer.c b/tools/testing/selftests/bpf/json_writer.c
new file mode 120000
index 000000000000..5effa31e2f39
--- /dev/null
+++ b/tools/testing/selftests/bpf/json_writer.c
@@ -0,0 +1 @@
+../../../bpf/bpftool/json_writer.c
\ No newline at end of file
diff --git a/tools/testing/selftests/bpf/json_writer.h b/tools/testing/selftests/bpf/json_writer.h
new file mode 120000
index 000000000000..e0a264c26752
--- /dev/null
+++ b/tools/testing/selftests/bpf/json_writer.h
@@ -0,0 +1 @@
+../../../bpf/bpftool/json_writer.h
\ No newline at end of file
diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index 6d5e3022c75f..d903e6a72a96 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -18,6 +18,7 @@
 #include <sys/socket.h>
 #include <sys/un.h>
 #include <bpf/btf.h>
+#include "json_writer.h"
 
 static bool verbose(void)
 {
@@ -269,10 +270,23 @@ static void print_subtest_name(int test_num, int subtest_num,
 	fprintf(env.stdout, "\n");
 }
 
+static void jsonw_write_log_message(json_writer_t *w, char *log_buf, size_t log_cnt)
+{
+	/* open_memstream (from stdio_hijack_init) ensures that log_bug is terminated by a
+	 * null byte. Yet in parallel mode, log_buf will be NULL if there is no message.
+	 */
+	if (log_cnt) {
+		jsonw_string_field(w, "message", log_buf);
+	} else {
+		jsonw_string_field(w, "message", "");
+	}
+}
+
 static void dump_test_log(const struct prog_test_def *test,
 			  const struct test_state *test_state,
 			  bool skip_ok_subtests,
-			  bool par_exec_result)
+			  bool par_exec_result,
+			  json_writer_t *w)
 {
 	bool test_failed = test_state->error_cnt > 0;
 	bool force_log = test_state->force_log;
@@ -296,6 +310,16 @@ static void dump_test_log(const struct prog_test_def *test,
 	if (test_state->log_cnt && print_test)
 		print_test_log(test_state->log_buf, test_state->log_cnt);
 
+	if (w && print_test) {
+		jsonw_start_object(w);
+		jsonw_string_field(w, "name", test->test_name);
+		jsonw_uint_field(w, "number", test->test_num);
+		jsonw_write_log_message(w, test_state->log_buf, test_state->log_cnt);
+		jsonw_bool_field(w, "failed", test_failed);
+		jsonw_name(w, "subtests");
+		jsonw_start_array(w);
+	}
+
 	for (i = 0; i < test_state->subtest_num; i++) {
 		subtest_state = &test_state->subtest_states[i];
 		subtest_failed = subtest_state->error_cnt;
@@ -314,6 +338,20 @@ static void dump_test_log(const struct prog_test_def *test,
 				   test->test_name, subtest_state->name,
 				   test_result(subtest_state->error_cnt,
 					       subtest_state->skipped));
+
+		if (w && print_subtest) {
+			jsonw_start_object(w);
+			jsonw_string_field(w, "name", subtest_state->name);
+			jsonw_uint_field(w, "number", i+1);
+			jsonw_write_log_message(w, subtest_state->log_buf, subtest_state->log_cnt);
+			jsonw_bool_field(w, "failed", subtest_failed);
+			jsonw_end_object(w);
+		}
+	}
+
+	if (w && print_test) {
+		jsonw_end_array(w);
+		jsonw_end_object(w);
 	}
 
 	print_test_result(test, test_state);
@@ -715,6 +753,7 @@ enum ARG_KEYS {
 	ARG_TEST_NAME_GLOB_DENYLIST = 'd',
 	ARG_NUM_WORKERS = 'j',
 	ARG_DEBUG = -1,
+	ARG_JSON_SUMMARY = 'J'
 };
 
 static const struct argp_option opts[] = {
@@ -740,6 +779,7 @@ static const struct argp_option opts[] = {
 	  "Number of workers to run in parallel, default to number of cpus." },
 	{ "debug", ARG_DEBUG, NULL, 0,
 	  "print extra debug information for test_progs." },
+	{ "json-summary", ARG_JSON_SUMMARY, "FILE", 0, "Write report in json format to this file."},
 	{},
 };
 
@@ -870,6 +910,13 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 	case ARG_DEBUG:
 		env->debug = true;
 		break;
+	case ARG_JSON_SUMMARY:
+		env->json = fopen(arg, "w");
+		if (env->json == NULL) {
+			perror("Failed to open json summary file");
+			return -errno;
+		}
+		break;
 	case ARGP_KEY_ARG:
 		argp_usage(state);
 		break;
@@ -1017,7 +1064,7 @@ void crash_handler(int signum)
 		stdio_restore();
 	if (env.test) {
 		env.test_state->error_cnt++;
-		dump_test_log(env.test, env.test_state, true, false);
+		dump_test_log(env.test, env.test_state, true, false, NULL);
 	}
 	if (env.worker_id != -1)
 		fprintf(stderr, "[%d]: ", env.worker_id);
@@ -1124,7 +1171,7 @@ static void run_one_test(int test_num)
 
 	stdio_restore();
 
-	dump_test_log(test, state, false, false);
+	dump_test_log(test, state, false, false, NULL);
 }
 
 struct dispatch_data {
@@ -1283,7 +1330,7 @@ static void *dispatch_thread(void *ctx)
 		} while (false);
 
 		pthread_mutex_lock(&stdout_output_lock);
-		dump_test_log(test, state, false, true);
+		dump_test_log(test, state, false, true, NULL);
 		pthread_mutex_unlock(&stdout_output_lock);
 	} /* while (true) */
 error:
@@ -1308,6 +1355,7 @@ static void calculate_summary_and_print_errors(struct test_env *env)
 {
 	int i;
 	int succ_cnt = 0, fail_cnt = 0, sub_succ_cnt = 0, skip_cnt = 0;
+	json_writer_t *w = NULL;
 
 	for (i = 0; i < prog_test_cnt; i++) {
 		struct test_state *state = &test_states[i];
@@ -1324,6 +1372,22 @@ static void calculate_summary_and_print_errors(struct test_env *env)
 			succ_cnt++;
 	}
 
+	if (env->json) {
+		w = jsonw_new(env->json);
+		if (!w)
+			fprintf(env->stderr, "Failed to create new JSON stream.");
+	}
+
+	if (w) {
+		jsonw_start_object(w);
+		jsonw_uint_field(w, "success", succ_cnt);
+		jsonw_uint_field(w, "success_subtest", sub_succ_cnt);
+		jsonw_uint_field(w, "skipped", skip_cnt);
+		jsonw_uint_field(w, "failed", fail_cnt);
+		jsonw_name(w, "results");
+		jsonw_start_array(w);
+	}
+
 	/*
 	 * We only print error logs summary when there are failed tests and
 	 * verbose mode is not enabled. Otherwise, results may be incosistent.
@@ -1340,10 +1404,19 @@ static void calculate_summary_and_print_errors(struct test_env *env)
 			if (!state->tested || !state->error_cnt)
 				continue;
 
-			dump_test_log(test, state, true, true);
+			dump_test_log(test, state, true, true, w);
 		}
 	}
 
+	if (w) {
+		jsonw_end_array(w);
+		jsonw_end_object(w);
+		jsonw_destroy(&w);
+	}
+
+	if (env->json)
+		fclose(env->json);
+
 	printf("Summary: %d/%d PASSED, %d SKIPPED, %d FAILED\n",
 	       succ_cnt, sub_succ_cnt, skip_cnt, fail_cnt);
 
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 3cbf005747ed..4b06b8347cd4 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -114,6 +114,7 @@ struct test_env {
 	FILE *stdout;
 	FILE *stderr;
 	int nr_cpus;
+	FILE *json;
 
 	int succ_cnt; /* successful tests */
 	int sub_succ_cnt; /* successful sub-tests */
-- 
cgit v1.2.3-70-g09d2


From 6365ba64b4dbe8b59ddaeaa724b281f3787715d5 Mon Sep 17 00:00:00 2001
From: Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>
Date: Wed, 8 Mar 2023 15:05:31 +0000
Subject: ptp: kvm: Use decrypted memory in confidential guest on x86

KVM_HC_CLOCK_PAIRING currently fails inside SEV-SNP guests because the
guest passes an address to static data to the host. In confidential
computing the host can't access arbitrary guest memory so handling the
hypercall runs into an "rmpfault". To make the hypercall work, the guest
needs to explicitly mark the memory as decrypted. Do that in
kvm_arch_ptp_init(), but retain the previous behavior for
non-confidential guests to save us from having to allocate memory.

Add a new arch-specific function (kvm_arch_ptp_exit()) to free the
allocation and mark the memory as encrypted again.

Signed-off-by: Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>
Link: https://lore.kernel.org/r/20230308150531.477741-1-jpiotrowski@linux.microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/ptp/ptp_kvm_arm.c    |  4 +++
 drivers/ptp/ptp_kvm_common.c |  1 +
 drivers/ptp/ptp_kvm_x86.c    | 59 +++++++++++++++++++++++++++++++++++---------
 include/linux/ptp_kvm.h      |  1 +
 4 files changed, 54 insertions(+), 11 deletions(-)

diff --git a/drivers/ptp/ptp_kvm_arm.c b/drivers/ptp/ptp_kvm_arm.c
index b7d28c8dfb84..e68e6943167b 100644
--- a/drivers/ptp/ptp_kvm_arm.c
+++ b/drivers/ptp/ptp_kvm_arm.c
@@ -22,6 +22,10 @@ int kvm_arch_ptp_init(void)
 	return 0;
 }
 
+void kvm_arch_ptp_exit(void)
+{
+}
+
 int kvm_arch_ptp_get_clock(struct timespec64 *ts)
 {
 	return kvm_arch_ptp_get_crosststamp(NULL, ts, NULL);
diff --git a/drivers/ptp/ptp_kvm_common.c b/drivers/ptp/ptp_kvm_common.c
index 9141162c4237..2418977989be 100644
--- a/drivers/ptp/ptp_kvm_common.c
+++ b/drivers/ptp/ptp_kvm_common.c
@@ -130,6 +130,7 @@ static struct kvm_ptp_clock kvm_ptp_clock;
 static void __exit ptp_kvm_exit(void)
 {
 	ptp_clock_unregister(kvm_ptp_clock.ptp_clock);
+	kvm_arch_ptp_exit();
 }
 
 static int __init ptp_kvm_init(void)
diff --git a/drivers/ptp/ptp_kvm_x86.c b/drivers/ptp/ptp_kvm_x86.c
index 4991054a2135..902844cc1a17 100644
--- a/drivers/ptp/ptp_kvm_x86.c
+++ b/drivers/ptp/ptp_kvm_x86.c
@@ -14,27 +14,64 @@
 #include <uapi/linux/kvm_para.h>
 #include <linux/ptp_clock_kernel.h>
 #include <linux/ptp_kvm.h>
+#include <linux/set_memory.h>
 
 static phys_addr_t clock_pair_gpa;
-static struct kvm_clock_pairing clock_pair;
+static struct kvm_clock_pairing clock_pair_glbl;
+static struct kvm_clock_pairing *clock_pair;
 
 int kvm_arch_ptp_init(void)
 {
+	struct page *p;
 	long ret;
 
 	if (!kvm_para_available())
 		return -ENODEV;
 
-	clock_pair_gpa = slow_virt_to_phys(&clock_pair);
-	if (!pvclock_get_pvti_cpu0_va())
-		return -ENODEV;
+	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) {
+		p = alloc_page(GFP_KERNEL | __GFP_ZERO);
+		if (!p)
+			return -ENOMEM;
+
+		clock_pair = page_address(p);
+		ret = set_memory_decrypted((unsigned long)clock_pair, 1);
+		if (ret) {
+			__free_page(p);
+			clock_pair = NULL;
+			goto nofree;
+		}
+	} else {
+		clock_pair = &clock_pair_glbl;
+	}
+
+	clock_pair_gpa = slow_virt_to_phys(clock_pair);
+	if (!pvclock_get_pvti_cpu0_va()) {
+		ret = -ENODEV;
+		goto err;
+	}
 
 	ret = kvm_hypercall2(KVM_HC_CLOCK_PAIRING, clock_pair_gpa,
 			     KVM_CLOCK_PAIRING_WALLCLOCK);
-	if (ret == -KVM_ENOSYS)
-		return -ENODEV;
+	if (ret == -KVM_ENOSYS) {
+		ret = -ENODEV;
+		goto err;
+	}
 
 	return ret;
+
+err:
+	kvm_arch_ptp_exit();
+nofree:
+	return ret;
+}
+
+void kvm_arch_ptp_exit(void)
+{
+	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) {
+		WARN_ON(set_memory_encrypted((unsigned long)clock_pair, 1));
+		free_page((unsigned long)clock_pair);
+		clock_pair = NULL;
+	}
 }
 
 int kvm_arch_ptp_get_clock(struct timespec64 *ts)
@@ -49,8 +86,8 @@ int kvm_arch_ptp_get_clock(struct timespec64 *ts)
 		return -EOPNOTSUPP;
 	}
 
-	ts->tv_sec = clock_pair.sec;
-	ts->tv_nsec = clock_pair.nsec;
+	ts->tv_sec = clock_pair->sec;
+	ts->tv_nsec = clock_pair->nsec;
 
 	return 0;
 }
@@ -81,9 +118,9 @@ int kvm_arch_ptp_get_crosststamp(u64 *cycle, struct timespec64 *tspec,
 			pr_err_ratelimited("clock pairing hypercall ret %lu\n", ret);
 			return -EOPNOTSUPP;
 		}
-		tspec->tv_sec = clock_pair.sec;
-		tspec->tv_nsec = clock_pair.nsec;
-		*cycle = __pvclock_read_cycles(src, clock_pair.tsc);
+		tspec->tv_sec = clock_pair->sec;
+		tspec->tv_nsec = clock_pair->nsec;
+		*cycle = __pvclock_read_cycles(src, clock_pair->tsc);
 	} while (pvclock_read_retry(src, version));
 
 	*cs = &kvm_clock;
diff --git a/include/linux/ptp_kvm.h b/include/linux/ptp_kvm.h
index c2e28deef33a..746fd67c3480 100644
--- a/include/linux/ptp_kvm.h
+++ b/include/linux/ptp_kvm.h
@@ -14,6 +14,7 @@ struct timespec64;
 struct clocksource;
 
 int kvm_arch_ptp_init(void);
+void kvm_arch_ptp_exit(void);
 int kvm_arch_ptp_get_clock(struct timespec64 *ts);
 int kvm_arch_ptp_get_crosststamp(u64 *cycle,
 		struct timespec64 *tspec, struct clocksource **cs);
-- 
cgit v1.2.3-70-g09d2


From 4dd2744fae6dc705c5d2298fe0f0812d34f5b64e Mon Sep 17 00:00:00 2001
From: mengyuanlou <mengyuanlou@net-swift.com>
Date: Wed, 15 Mar 2023 17:18:46 +0800
Subject: net: wangxun: Remove macro that is redefined

Remove PCI_VENDOR_ID_WANGXUN which is redefined in
drivers/pci/quirks.

Signed-off-by: mengyuanlou <mengyuanlou@net-swift.com>
Link: https://lore.kernel.org/r/20230315091846.17314-1-mengyuanlou@net-swift.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/wangxun/libwx/wx_type.h | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h
index 2b9efd13c500..2bec5b1bc196 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_type.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h
@@ -7,11 +7,6 @@
 #include <linux/bitfield.h>
 #include <linux/netdevice.h>
 
-/* Vendor ID */
-#ifndef PCI_VENDOR_ID_WANGXUN
-#define PCI_VENDOR_ID_WANGXUN                   0x8088
-#endif
-
 #define WX_NCSI_SUP                             0x8000
 #define WX_NCSI_MASK                            0x8000
 #define WX_WOL_SUP                              0x4000
-- 
cgit v1.2.3-70-g09d2


From b1a2de9ccfe63b50e493ebb57b85beb5785200c7 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 15 Mar 2023 17:38:43 +0100
Subject: net: dsa: mv88e6xxx: don't dispose of Global2 IRQ mappings from
 mdiobus code

irq_find_mapping() does not need irq_dispose_mapping(), only
irq_create_mapping() does.

Calling irq_dispose_mapping() from mv88e6xxx_g2_irq_mdio_free() and from
the error path of mv88e6xxx_g2_irq_mdio_setup() effectively means that
the mdiobus logic (for internal PHY interrupts) is disposing of a
hwirq->virq mapping which it is not responsible of (but instead, the
function pair mv88e6xxx_g2_irq_setup() + mv88e6xxx_g2_irq_free() is).

With the current code structure, this isn't such a huge problem, because
mv88e6xxx_g2_irq_mdio_free() is called relatively close to the real
owner of the IRQ mappings:

mv88e6xxx_remove()
-> mv88e6xxx_unregister_switch()
-> mv88e6xxx_mdios_unregister()
   -> mv88e6xxx_g2_irq_mdio_free()
-> mv88e6xxx_g2_irq_free()

and the switch isn't 'live' in any way such that it would be able of
generating interrupts at this point (mv88e6xxx_unregister_switch() has
been called).

However, there is a desire to split mv88e6xxx_mdios_unregister() and
mv88e6xxx_g2_irq_free() such that mv88e6xxx_mdios_unregister() only gets
called from mv88e6xxx_teardown(). This is much more problematic, as can
be seen below.

In a cross-chip scenario (say 3 switches d0032004.mdio-mii:10,
d0032004.mdio-mii:11 and d0032004.mdio-mii:12 which form a single DSA
tree), it is possible to unbind the device driver from a single switch
(say d0032004.mdio-mii:10).

When that happens, mv88e6xxx_remove() will be called for just that one
switch, and this will call mv88e6xxx_unregister_switch() which will tear
down the entire tree (calling mv88e6xxx_teardown() for all 3 switches).

Assuming mv88e6xxx_mdios_unregister() was moved to mv88e6xxx_teardown(),
at this stage, all 3 switches will have called irq_dispose_mapping() on
their mdiobus virqs.

When we bind again the device driver to d0032004.mdio-mii:10,
mv88e6xxx_probe() is called for it, which calls dsa_register_switch().
The DSA tree is now complete again, and mv88e6xxx_setup() is called for
all 3 switches.

Also assuming that mv88e6xxx_mdios_register() is moved to
mv88e6xxx_setup() (the 2 assumptions go together), at this point,
d0032004.mdio-mii:11 and d0032004.mdio-mii:12 don't have an IRQ mapping
for the internal PHYs anymore, as they've disposed of it in
mv88e6xxx_teardown(). Whereas switch d0032004.mdio-mii:10 has re-created
it, because its code path comes from mv88e6xxx_probe().

Simply put, this change prepares the driver to handle the movement of
mv88e6xxx_mdios_register() to mv88e6xxx_setup() for cross-chip DSA trees.

Also, the code being deleted was partially wrong anyway (in a way which
may have hidden this other issue). mv88e6xxx_g2_irq_mdio_setup()
populates bus->irq[] starting with offset chip->info->phy_base_addr, but
the teardown path doesn't apply that offset too. So it disposes of virq
0 for phy = [ 0, phy_base_addr ).

All switch families have phy_base_addr = 0, except for MV88E6141 and
MV88E6341 which have it as 0x10. I guess those families would have
happened to work by mistake in cross-chip scenarios too.

I'm deleting the body of mv88e6xxx_g2_irq_mdio_free() but leaving its
call sites and prototype in place. This is because, if we ever need to
add back some teardown procedure in the future, it will be perhaps
error-prone to deduce the proper call sites again. Whereas like this,
no extra code should get generated, it shouldn't bother anybody.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/mv88e6xxx/global2.c | 20 ++++----------------
 1 file changed, 4 insertions(+), 16 deletions(-)

diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c
index ed3b2f88e783..a26546d3d7b5 100644
--- a/drivers/net/dsa/mv88e6xxx/global2.c
+++ b/drivers/net/dsa/mv88e6xxx/global2.c
@@ -1176,31 +1176,19 @@ out:
 int mv88e6xxx_g2_irq_mdio_setup(struct mv88e6xxx_chip *chip,
 				struct mii_bus *bus)
 {
-	int phy, irq, err, err_phy;
+	int phy, irq;
 
 	for (phy = 0; phy < chip->info->num_internal_phys; phy++) {
 		irq = irq_find_mapping(chip->g2_irq.domain, phy);
-		if (irq < 0) {
-			err = irq;
-			goto out;
-		}
+		if (irq < 0)
+			return irq;
+
 		bus->irq[chip->info->phy_base_addr + phy] = irq;
 	}
 	return 0;
-out:
-	err_phy = phy;
-
-	for (phy = 0; phy < err_phy; phy++)
-		irq_dispose_mapping(bus->irq[phy]);
-
-	return err;
 }
 
 void mv88e6xxx_g2_irq_mdio_free(struct mv88e6xxx_chip *chip,
 				struct mii_bus *bus)
 {
-	int phy;
-
-	for (phy = 0; phy < chip->info->num_internal_phys; phy++)
-		irq_dispose_mapping(bus->irq[phy]);
 }
-- 
cgit v1.2.3-70-g09d2


From f1bee740fa821c28582bb17c42e2242041e598b8 Mon Sep 17 00:00:00 2001
From: Klaus Kudielka <klaus.kudielka@gmail.com>
Date: Wed, 15 Mar 2023 17:38:44 +0100
Subject: net: dsa: mv88e6xxx: re-order functions

Move mv88e6xxx_setup() below mv88e6xxx_mdios_register(), so that we are
able to call the latter one from here. Do the same thing for the
inverse functions.

Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 358 +++++++++++++++++++--------------------
 1 file changed, 179 insertions(+), 179 deletions(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 30383c4f8fd0..184c116dfc8f 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -3680,185 +3680,6 @@ static int mv88e6390_setup_errata(struct mv88e6xxx_chip *chip)
 	return mv88e6xxx_software_reset(chip);
 }
 
-static void mv88e6xxx_teardown(struct dsa_switch *ds)
-{
-	mv88e6xxx_teardown_devlink_params(ds);
-	dsa_devlink_resources_unregister(ds);
-	mv88e6xxx_teardown_devlink_regions_global(ds);
-}
-
-static int mv88e6xxx_setup(struct dsa_switch *ds)
-{
-	struct mv88e6xxx_chip *chip = ds->priv;
-	u8 cmode;
-	int err;
-	int i;
-
-	chip->ds = ds;
-	ds->slave_mii_bus = mv88e6xxx_default_mdio_bus(chip);
-
-	/* Since virtual bridges are mapped in the PVT, the number we support
-	 * depends on the physical switch topology. We need to let DSA figure
-	 * that out and therefore we cannot set this at dsa_register_switch()
-	 * time.
-	 */
-	if (mv88e6xxx_has_pvt(chip))
-		ds->max_num_bridges = MV88E6XXX_MAX_PVT_SWITCHES -
-				      ds->dst->last_switch - 1;
-
-	mv88e6xxx_reg_lock(chip);
-
-	if (chip->info->ops->setup_errata) {
-		err = chip->info->ops->setup_errata(chip);
-		if (err)
-			goto unlock;
-	}
-
-	/* Cache the cmode of each port. */
-	for (i = 0; i < mv88e6xxx_num_ports(chip); i++) {
-		if (chip->info->ops->port_get_cmode) {
-			err = chip->info->ops->port_get_cmode(chip, i, &cmode);
-			if (err)
-				goto unlock;
-
-			chip->ports[i].cmode = cmode;
-		}
-	}
-
-	err = mv88e6xxx_vtu_setup(chip);
-	if (err)
-		goto unlock;
-
-	/* Must be called after mv88e6xxx_vtu_setup (which flushes the
-	 * VTU, thereby also flushing the STU).
-	 */
-	err = mv88e6xxx_stu_setup(chip);
-	if (err)
-		goto unlock;
-
-	/* Setup Switch Port Registers */
-	for (i = 0; i < mv88e6xxx_num_ports(chip); i++) {
-		if (dsa_is_unused_port(ds, i))
-			continue;
-
-		/* Prevent the use of an invalid port. */
-		if (mv88e6xxx_is_invalid_port(chip, i)) {
-			dev_err(chip->dev, "port %d is invalid\n", i);
-			err = -EINVAL;
-			goto unlock;
-		}
-
-		err = mv88e6xxx_setup_port(chip, i);
-		if (err)
-			goto unlock;
-	}
-
-	err = mv88e6xxx_irl_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_mac_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_phy_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_pvt_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_atu_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_broadcast_setup(chip, 0);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_pot_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_rmu_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_rsvd2cpu_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_trunk_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_devmap_setup(chip);
-	if (err)
-		goto unlock;
-
-	err = mv88e6xxx_pri_setup(chip);
-	if (err)
-		goto unlock;
-
-	/* Setup PTP Hardware Clock and timestamping */
-	if (chip->info->ptp_support) {
-		err = mv88e6xxx_ptp_setup(chip);
-		if (err)
-			goto unlock;
-
-		err = mv88e6xxx_hwtstamp_setup(chip);
-		if (err)
-			goto unlock;
-	}
-
-	err = mv88e6xxx_stats_setup(chip);
-	if (err)
-		goto unlock;
-
-unlock:
-	mv88e6xxx_reg_unlock(chip);
-
-	if (err)
-		return err;
-
-	/* Have to be called without holding the register lock, since
-	 * they take the devlink lock, and we later take the locks in
-	 * the reverse order when getting/setting parameters or
-	 * resource occupancy.
-	 */
-	err = mv88e6xxx_setup_devlink_resources(ds);
-	if (err)
-		return err;
-
-	err = mv88e6xxx_setup_devlink_params(ds);
-	if (err)
-		goto out_resources;
-
-	err = mv88e6xxx_setup_devlink_regions_global(ds);
-	if (err)
-		goto out_params;
-
-	return 0;
-
-out_params:
-	mv88e6xxx_teardown_devlink_params(ds);
-out_resources:
-	dsa_devlink_resources_unregister(ds);
-
-	return err;
-}
-
-static int mv88e6xxx_port_setup(struct dsa_switch *ds, int port)
-{
-	return mv88e6xxx_setup_devlink_regions_port(ds, port);
-}
-
-static void mv88e6xxx_port_teardown(struct dsa_switch *ds, int port)
-{
-	mv88e6xxx_teardown_devlink_regions_port(ds, port);
-}
-
 /* prod_id for switch families which do not have a PHY model number */
 static const u16 family_prod_id_table[] = {
 	[MV88E6XXX_FAMILY_6341] = MV88E6XXX_PORT_SWITCH_ID_PROD_6341,
@@ -4062,6 +3883,185 @@ static int mv88e6xxx_mdios_register(struct mv88e6xxx_chip *chip,
 	return 0;
 }
 
+static void mv88e6xxx_teardown(struct dsa_switch *ds)
+{
+	mv88e6xxx_teardown_devlink_params(ds);
+	dsa_devlink_resources_unregister(ds);
+	mv88e6xxx_teardown_devlink_regions_global(ds);
+}
+
+static int mv88e6xxx_setup(struct dsa_switch *ds)
+{
+	struct mv88e6xxx_chip *chip = ds->priv;
+	u8 cmode;
+	int err;
+	int i;
+
+	chip->ds = ds;
+	ds->slave_mii_bus = mv88e6xxx_default_mdio_bus(chip);
+
+	/* Since virtual bridges are mapped in the PVT, the number we support
+	 * depends on the physical switch topology. We need to let DSA figure
+	 * that out and therefore we cannot set this at dsa_register_switch()
+	 * time.
+	 */
+	if (mv88e6xxx_has_pvt(chip))
+		ds->max_num_bridges = MV88E6XXX_MAX_PVT_SWITCHES -
+				      ds->dst->last_switch - 1;
+
+	mv88e6xxx_reg_lock(chip);
+
+	if (chip->info->ops->setup_errata) {
+		err = chip->info->ops->setup_errata(chip);
+		if (err)
+			goto unlock;
+	}
+
+	/* Cache the cmode of each port. */
+	for (i = 0; i < mv88e6xxx_num_ports(chip); i++) {
+		if (chip->info->ops->port_get_cmode) {
+			err = chip->info->ops->port_get_cmode(chip, i, &cmode);
+			if (err)
+				goto unlock;
+
+			chip->ports[i].cmode = cmode;
+		}
+	}
+
+	err = mv88e6xxx_vtu_setup(chip);
+	if (err)
+		goto unlock;
+
+	/* Must be called after mv88e6xxx_vtu_setup (which flushes the
+	 * VTU, thereby also flushing the STU).
+	 */
+	err = mv88e6xxx_stu_setup(chip);
+	if (err)
+		goto unlock;
+
+	/* Setup Switch Port Registers */
+	for (i = 0; i < mv88e6xxx_num_ports(chip); i++) {
+		if (dsa_is_unused_port(ds, i))
+			continue;
+
+		/* Prevent the use of an invalid port. */
+		if (mv88e6xxx_is_invalid_port(chip, i)) {
+			dev_err(chip->dev, "port %d is invalid\n", i);
+			err = -EINVAL;
+			goto unlock;
+		}
+
+		err = mv88e6xxx_setup_port(chip, i);
+		if (err)
+			goto unlock;
+	}
+
+	err = mv88e6xxx_irl_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_mac_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_phy_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_pvt_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_atu_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_broadcast_setup(chip, 0);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_pot_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_rmu_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_rsvd2cpu_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_trunk_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_devmap_setup(chip);
+	if (err)
+		goto unlock;
+
+	err = mv88e6xxx_pri_setup(chip);
+	if (err)
+		goto unlock;
+
+	/* Setup PTP Hardware Clock and timestamping */
+	if (chip->info->ptp_support) {
+		err = mv88e6xxx_ptp_setup(chip);
+		if (err)
+			goto unlock;
+
+		err = mv88e6xxx_hwtstamp_setup(chip);
+		if (err)
+			goto unlock;
+	}
+
+	err = mv88e6xxx_stats_setup(chip);
+	if (err)
+		goto unlock;
+
+unlock:
+	mv88e6xxx_reg_unlock(chip);
+
+	if (err)
+		return err;
+
+	/* Have to be called without holding the register lock, since
+	 * they take the devlink lock, and we later take the locks in
+	 * the reverse order when getting/setting parameters or
+	 * resource occupancy.
+	 */
+	err = mv88e6xxx_setup_devlink_resources(ds);
+	if (err)
+		return err;
+
+	err = mv88e6xxx_setup_devlink_params(ds);
+	if (err)
+		goto out_resources;
+
+	err = mv88e6xxx_setup_devlink_regions_global(ds);
+	if (err)
+		goto out_params;
+
+	return 0;
+
+out_params:
+	mv88e6xxx_teardown_devlink_params(ds);
+out_resources:
+	dsa_devlink_resources_unregister(ds);
+
+	return err;
+}
+
+static int mv88e6xxx_port_setup(struct dsa_switch *ds, int port)
+{
+	return mv88e6xxx_setup_devlink_regions_port(ds, port);
+}
+
+static void mv88e6xxx_port_teardown(struct dsa_switch *ds, int port)
+{
+	mv88e6xxx_teardown_devlink_regions_port(ds, port);
+}
+
 static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds)
 {
 	struct mv88e6xxx_chip *chip = ds->priv;
-- 
cgit v1.2.3-70-g09d2


From 2cb0658d4f88608a19b3833789ac0ac8a4214f0b Mon Sep 17 00:00:00 2001
From: Klaus Kudielka <klaus.kudielka@gmail.com>
Date: Wed, 15 Mar 2023 17:38:45 +0100
Subject: net: dsa: mv88e6xxx: move call to mv88e6xxx_mdios_register()

Call the rather expensive mv88e6xxx_mdios_register() at the beginning of
mv88e6xxx_setup(). This avoids the double call via mv88e6xxx_probe()
during boot.

For symmetry, call mv88e6xxx_mdios_unregister() at the end of
mv88e6xxx_teardown().

Link: https://lore.kernel.org/lkml/449bde236c08d5ab5e54abd73b645d8b29955894.camel@gmail.com/
Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 184c116dfc8f..de25f755725e 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -3848,9 +3848,9 @@ static void mv88e6xxx_mdios_unregister(struct mv88e6xxx_chip *chip)
 	}
 }
 
-static int mv88e6xxx_mdios_register(struct mv88e6xxx_chip *chip,
-				    struct device_node *np)
+static int mv88e6xxx_mdios_register(struct mv88e6xxx_chip *chip)
 {
+	struct device_node *np = chip->dev->of_node;
 	struct device_node *child;
 	int err;
 
@@ -3885,9 +3885,12 @@ static int mv88e6xxx_mdios_register(struct mv88e6xxx_chip *chip,
 
 static void mv88e6xxx_teardown(struct dsa_switch *ds)
 {
+	struct mv88e6xxx_chip *chip = ds->priv;
+
 	mv88e6xxx_teardown_devlink_params(ds);
 	dsa_devlink_resources_unregister(ds);
 	mv88e6xxx_teardown_devlink_regions_global(ds);
+	mv88e6xxx_mdios_unregister(chip);
 }
 
 static int mv88e6xxx_setup(struct dsa_switch *ds)
@@ -3897,6 +3900,10 @@ static int mv88e6xxx_setup(struct dsa_switch *ds)
 	int err;
 	int i;
 
+	err = mv88e6xxx_mdios_register(chip);
+	if (err)
+		return err;
+
 	chip->ds = ds;
 	ds->slave_mii_bus = mv88e6xxx_default_mdio_bus(chip);
 
@@ -4023,7 +4030,7 @@ unlock:
 	mv88e6xxx_reg_unlock(chip);
 
 	if (err)
-		return err;
+		goto out_mdios;
 
 	/* Have to be called without holding the register lock, since
 	 * they take the devlink lock, and we later take the locks in
@@ -4032,7 +4039,7 @@ unlock:
 	 */
 	err = mv88e6xxx_setup_devlink_resources(ds);
 	if (err)
-		return err;
+		goto out_mdios;
 
 	err = mv88e6xxx_setup_devlink_params(ds);
 	if (err)
@@ -4048,6 +4055,8 @@ out_params:
 	mv88e6xxx_teardown_devlink_params(ds);
 out_resources:
 	dsa_devlink_resources_unregister(ds);
+out_mdios:
+	mv88e6xxx_mdios_unregister(chip);
 
 	return err;
 }
@@ -7228,18 +7237,12 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev)
 	if (err)
 		goto out_g1_atu_prob_irq;
 
-	err = mv88e6xxx_mdios_register(chip, np);
-	if (err)
-		goto out_g1_vtu_prob_irq;
-
 	err = mv88e6xxx_register_switch(chip);
 	if (err)
-		goto out_mdio;
+		goto out_g1_vtu_prob_irq;
 
 	return 0;
 
-out_mdio:
-	mv88e6xxx_mdios_unregister(chip);
 out_g1_vtu_prob_irq:
 	mv88e6xxx_g1_vtu_prob_irq_free(chip);
 out_g1_atu_prob_irq:
@@ -7276,7 +7279,6 @@ static void mv88e6xxx_remove(struct mdio_device *mdiodev)
 
 	mv88e6xxx_phy_destroy(chip);
 	mv88e6xxx_unregister_switch(chip);
-	mv88e6xxx_mdios_unregister(chip);
 
 	mv88e6xxx_g1_vtu_prob_irq_free(chip);
 	mv88e6xxx_g1_atu_prob_irq_free(chip);
-- 
cgit v1.2.3-70-g09d2


From 2c7e46edbd0308c12b8232ff4270d0894cbdce44 Mon Sep 17 00:00:00 2001
From: Klaus Kudielka <klaus.kudielka@gmail.com>
Date: Wed, 15 Mar 2023 17:38:46 +0100
Subject: net: dsa: mv88e6xxx: mask apparently non-existing phys during probing

To avoid excessive mdio bus transactions during probing, mask all phy
addresses that do not exist (there is a 1:1 mapping between switch port
number and phy address).

Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index de25f755725e..260e38c5c6e6 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -3805,6 +3805,7 @@ static int mv88e6xxx_mdio_register(struct mv88e6xxx_chip *chip,
 	bus->read_c45 = mv88e6xxx_mdio_read_c45;
 	bus->write_c45 = mv88e6xxx_mdio_write_c45;
 	bus->parent = chip->dev;
+	bus->phy_mask = GENMASK(31, mv88e6xxx_num_ports(chip));
 
 	if (!external) {
 		err = mv88e6xxx_g2_irq_mdio_setup(chip, bus);
-- 
cgit v1.2.3-70-g09d2


From 0de10fd6eb94259a749d558ee0d34083ae010a1d Mon Sep 17 00:00:00 2001
From: Alex Elder <elder@linaro.org>
Date: Wed, 15 Mar 2023 14:43:05 -0500
Subject: dt-bindings: net: qcom,ipa: add SDX65 compatible

Add support for SDX65, which uses IPA v5.0.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Alex Elder <elder@linaro.org>
Link: https://lore.kernel.org/r/20230315194305.1647311-1-elder@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/qcom,ipa.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/qcom,ipa.yaml b/Documentation/devicetree/bindings/net/qcom,ipa.yaml
index 4aeda379726f..2d5e4ffb2f9e 100644
--- a/Documentation/devicetree/bindings/net/qcom,ipa.yaml
+++ b/Documentation/devicetree/bindings/net/qcom,ipa.yaml
@@ -49,6 +49,7 @@ properties:
       - qcom,sc7280-ipa
       - qcom,sdm845-ipa
       - qcom,sdx55-ipa
+      - qcom,sdx65-ipa
       - qcom,sm6350-ipa
       - qcom,sm8350-ipa
 
-- 
cgit v1.2.3-70-g09d2


From ed0578a46c5f08647f5ecb0e3c735a53267426f7 Mon Sep 17 00:00:00 2001
From: Harini Katakam <harini.katakam@xilinx.com>
Date: Thu, 16 Mar 2023 14:00:50 +0530
Subject: net: macb: Increase halt timeout to accommodate 10Mbps link

Increase halt timeout to accommodate for 16K SRAM at 10Mbps rounded.

Signed-off-by: Harini Katakam <harini.katakam@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20230316083050.2108-1-harini.katakam@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cadence/macb_main.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 26f41243743b..c4edd20c1c66 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -94,8 +94,7 @@ struct sifive_fu540_macb_mgmt {
 /* Graceful stop timeouts in us. We should allow up to
  * 1 frame time (10 Mbits/s, full-duplex, ignoring collisions)
  */
-#define MACB_HALT_TIMEOUT	1230
-
+#define MACB_HALT_TIMEOUT	14000
 #define MACB_PM_TIMEOUT  100 /* ms */
 
 #define MACB_MDIO_TIMEOUT	1000000 /* in usecs */
-- 
cgit v1.2.3-70-g09d2


From 36bd28c1cb0dbf48645cfe43159907fb3253b33a Mon Sep 17 00:00:00 2001
From: haozhe chang <haozhe.chang@mediatek.com>
Date: Thu, 16 Mar 2023 17:58:20 +0800
Subject: wwan: core: Support slicing in port TX flow of WWAN subsystem

wwan_port_fops_write inputs the SKB parameter to the TX callback of
the WWAN device driver. However, the WWAN device (e.g., t7xx) may
have an MTU less than the size of SKB, causing the TX buffer to be
sliced and copied once more in the WWAN device driver.

This patch implements the slicing in the WWAN subsystem and gives
the WWAN devices driver the option to slice(by frag_len) or not. By
doing so, the additional memory copy is reduced.

Meanwhile, this patch gives WWAN devices driver the option to reserve
headroom in fragments for the device-specific metadata.

Signed-off-by: haozhe chang <haozhe.chang@mediatek.com>
Reviewed-by: Loic Poulain <loic.poulain@linaro.org>
Link: https://lore.kernel.org/r/20230316095826.181904-1-haozhe.chang@mediatek.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/wwan/iosm/iosm_ipc_port.c  |  3 +-
 drivers/net/wwan/mhi_wwan_ctrl.c       |  2 +-
 drivers/net/wwan/rpmsg_wwan_ctrl.c     |  2 +-
 drivers/net/wwan/t7xx/t7xx_port_wwan.c | 36 +++++++++++----------
 drivers/net/wwan/wwan_core.c           | 58 ++++++++++++++++++++++++++--------
 drivers/net/wwan/wwan_hwsim.c          |  2 +-
 drivers/usb/class/cdc-wdm.c            |  3 +-
 include/linux/wwan.h                   | 11 +++++++
 8 files changed, 81 insertions(+), 36 deletions(-)

diff --git a/drivers/net/wwan/iosm/iosm_ipc_port.c b/drivers/net/wwan/iosm/iosm_ipc_port.c
index b6d81c627277..5d5b4183e14a 100644
--- a/drivers/net/wwan/iosm/iosm_ipc_port.c
+++ b/drivers/net/wwan/iosm/iosm_ipc_port.c
@@ -63,7 +63,8 @@ struct iosm_cdev *ipc_port_init(struct iosm_imem *ipc_imem,
 	ipc_port->ipc_imem = ipc_imem;
 
 	ipc_port->iosm_port = wwan_create_port(ipc_port->dev, port_type,
-					       &ipc_wwan_ctrl_ops, ipc_port);
+					       &ipc_wwan_ctrl_ops, NULL,
+					       ipc_port);
 
 	return ipc_port;
 }
diff --git a/drivers/net/wwan/mhi_wwan_ctrl.c b/drivers/net/wwan/mhi_wwan_ctrl.c
index f7ca52353f40..e9f979d2d851 100644
--- a/drivers/net/wwan/mhi_wwan_ctrl.c
+++ b/drivers/net/wwan/mhi_wwan_ctrl.c
@@ -237,7 +237,7 @@ static int mhi_wwan_ctrl_probe(struct mhi_device *mhi_dev,
 
 	/* Register as a wwan port, id->driver_data contains wwan port type */
 	port = wwan_create_port(&cntrl->mhi_dev->dev, id->driver_data,
-				&wwan_pops, mhiwwan);
+				&wwan_pops, NULL, mhiwwan);
 	if (IS_ERR(port)) {
 		kfree(mhiwwan);
 		return PTR_ERR(port);
diff --git a/drivers/net/wwan/rpmsg_wwan_ctrl.c b/drivers/net/wwan/rpmsg_wwan_ctrl.c
index 31c24420ab2e..06f4b02f1552 100644
--- a/drivers/net/wwan/rpmsg_wwan_ctrl.c
+++ b/drivers/net/wwan/rpmsg_wwan_ctrl.c
@@ -129,7 +129,7 @@ static int rpmsg_wwan_ctrl_probe(struct rpmsg_device *rpdev)
 
 	/* Register as a wwan port, id.driver_data contains wwan port type */
 	port = wwan_create_port(parent, rpdev->id.driver_data,
-				&rpmsg_wwan_pops, rpwwan);
+				&rpmsg_wwan_pops, NULL, rpwwan);
 	if (IS_ERR(port))
 		return PTR_ERR(port);
 
diff --git a/drivers/net/wwan/t7xx/t7xx_port_wwan.c b/drivers/net/wwan/t7xx/t7xx_port_wwan.c
index 24bd21942403..17389c8f6600 100644
--- a/drivers/net/wwan/t7xx/t7xx_port_wwan.c
+++ b/drivers/net/wwan/t7xx/t7xx_port_wwan.c
@@ -54,13 +54,13 @@ static void t7xx_port_ctrl_stop(struct wwan_port *port)
 static int t7xx_port_ctrl_tx(struct wwan_port *port, struct sk_buff *skb)
 {
 	struct t7xx_port *port_private = wwan_port_get_drvdata(port);
-	size_t len, offset, chunk_len = 0, txq_mtu = CLDMA_MTU;
 	const struct t7xx_port_conf *port_conf;
+	struct sk_buff *cur = skb, *cloned;
 	struct t7xx_fsm_ctl *ctl;
 	enum md_state md_state;
+	int cnt = 0, ret;
 
-	len = skb->len;
-	if (!len || !port_private->chan_enable)
+	if (!port_private->chan_enable)
 		return -EINVAL;
 
 	port_conf = port_private->port_conf;
@@ -72,23 +72,21 @@ static int t7xx_port_ctrl_tx(struct wwan_port *port, struct sk_buff *skb)
 		return -ENODEV;
 	}
 
-	for (offset = 0; offset < len; offset += chunk_len) {
-		struct sk_buff *skb_ccci;
-		int ret;
-
-		chunk_len = min(len - offset, txq_mtu - sizeof(struct ccci_header));
-		skb_ccci = t7xx_port_alloc_skb(chunk_len);
-		if (!skb_ccci)
-			return -ENOMEM;
-
-		skb_put_data(skb_ccci, skb->data + offset, chunk_len);
-		ret = t7xx_port_send_skb(port_private, skb_ccci, 0, 0);
+	while (cur) {
+		cloned = skb_clone(cur, GFP_KERNEL);
+		cloned->len = skb_headlen(cur);
+		ret = t7xx_port_send_skb(port_private, cloned, 0, 0);
 		if (ret) {
-			dev_kfree_skb_any(skb_ccci);
+			dev_kfree_skb(cloned);
 			dev_err(port_private->dev, "Write error on %s port, %d\n",
 				port_conf->name, ret);
-			return ret;
+			return cnt ? cnt + ret : ret;
 		}
+		cnt += cur->len;
+		if (cur == skb)
+			cur = skb_shinfo(skb)->frag_list;
+		else
+			cur = cur->next;
 	}
 
 	dev_kfree_skb(skb);
@@ -154,13 +152,17 @@ static int t7xx_port_wwan_disable_chl(struct t7xx_port *port)
 static void t7xx_port_wwan_md_state_notify(struct t7xx_port *port, unsigned int state)
 {
 	const struct t7xx_port_conf *port_conf = port->port_conf;
+	unsigned int header_len = sizeof(struct ccci_header);
+	struct wwan_port_caps caps;
 
 	if (state != MD_STATE_READY)
 		return;
 
 	if (!port->wwan.wwan_port) {
+		caps.frag_len = CLDMA_MTU - header_len;
+		caps.headroom_len = header_len;
 		port->wwan.wwan_port = wwan_create_port(port->dev, port_conf->port_type,
-							&wwan_ops, port);
+							&wwan_ops, &caps, port);
 		if (IS_ERR(port->wwan.wwan_port))
 			dev_err(port->dev, "Unable to create WWWAN port %s", port_conf->name);
 	}
diff --git a/drivers/net/wwan/wwan_core.c b/drivers/net/wwan/wwan_core.c
index 966d0ccd2276..2e1c01cf00a9 100644
--- a/drivers/net/wwan/wwan_core.c
+++ b/drivers/net/wwan/wwan_core.c
@@ -67,6 +67,8 @@ struct wwan_device {
  * @rxq: Buffer inbound queue
  * @waitqueue: The waitqueue for port fops (read/write/poll)
  * @data_lock: Port specific data access serialization
+ * @headroom_len: SKB reserved headroom size
+ * @frag_len: Length to fragment packet
  * @at_data: AT port specific data
  */
 struct wwan_port {
@@ -79,6 +81,8 @@ struct wwan_port {
 	struct sk_buff_head rxq;
 	wait_queue_head_t waitqueue;
 	struct mutex data_lock;	/* Port specific data access serialization */
+	size_t headroom_len;
+	size_t frag_len;
 	union {
 		struct {
 			struct ktermios termios;
@@ -426,6 +430,7 @@ static int __wwan_port_dev_assign_name(struct wwan_port *port, const char *fmt)
 struct wwan_port *wwan_create_port(struct device *parent,
 				   enum wwan_port_type type,
 				   const struct wwan_port_ops *ops,
+				   struct wwan_port_caps *caps,
 				   void *drvdata)
 {
 	struct wwan_device *wwandev;
@@ -459,6 +464,8 @@ struct wwan_port *wwan_create_port(struct device *parent,
 
 	port->type = type;
 	port->ops = ops;
+	port->frag_len = caps ? caps->frag_len : SIZE_MAX;
+	port->headroom_len = caps ? caps->headroom_len : 0;
 	mutex_init(&port->ops_lock);
 	skb_queue_head_init(&port->rxq);
 	init_waitqueue_head(&port->waitqueue);
@@ -702,30 +709,53 @@ static ssize_t wwan_port_fops_read(struct file *filp, char __user *buf,
 static ssize_t wwan_port_fops_write(struct file *filp, const char __user *buf,
 				    size_t count, loff_t *offp)
 {
+	struct sk_buff *skb, *head = NULL, *tail = NULL;
 	struct wwan_port *port = filp->private_data;
-	struct sk_buff *skb;
+	size_t frag_len, remain = count;
 	int ret;
 
 	ret = wwan_wait_tx(port, !!(filp->f_flags & O_NONBLOCK));
 	if (ret)
 		return ret;
 
-	skb = alloc_skb(count, GFP_KERNEL);
-	if (!skb)
-		return -ENOMEM;
+	do {
+		frag_len = min(remain, port->frag_len);
+		skb = alloc_skb(frag_len + port->headroom_len, GFP_KERNEL);
+		if (!skb) {
+			ret = -ENOMEM;
+			goto freeskb;
+		}
+		skb_reserve(skb, port->headroom_len);
+
+		if (!head) {
+			head = skb;
+		} else if (!tail) {
+			skb_shinfo(head)->frag_list = skb;
+			tail = skb;
+		} else {
+			tail->next = skb;
+			tail = skb;
+		}
 
-	if (copy_from_user(skb_put(skb, count), buf, count)) {
-		kfree_skb(skb);
-		return -EFAULT;
-	}
+		if (copy_from_user(skb_put(skb, frag_len), buf + count - remain, frag_len)) {
+			ret = -EFAULT;
+			goto freeskb;
+		}
 
-	ret = wwan_port_op_tx(port, skb, !!(filp->f_flags & O_NONBLOCK));
-	if (ret) {
-		kfree_skb(skb);
-		return ret;
-	}
+		if (skb != head) {
+			head->data_len += skb->len;
+			head->len += skb->len;
+			head->truesize += skb->truesize;
+		}
+	} while (remain -= frag_len);
+
+	ret = wwan_port_op_tx(port, head, !!(filp->f_flags & O_NONBLOCK));
+	if (!ret)
+		return count;
 
-	return count;
+freeskb:
+	kfree_skb(head);
+	return ret;
 }
 
 static __poll_t wwan_port_fops_poll(struct file *filp, poll_table *wait)
diff --git a/drivers/net/wwan/wwan_hwsim.c b/drivers/net/wwan/wwan_hwsim.c
index 2397a903d8f5..dfbdaa259a3f 100644
--- a/drivers/net/wwan/wwan_hwsim.c
+++ b/drivers/net/wwan/wwan_hwsim.c
@@ -205,7 +205,7 @@ static struct wwan_hwsim_port *wwan_hwsim_port_new(struct wwan_hwsim_dev *dev)
 
 	port->wwan = wwan_create_port(&dev->dev, WWAN_PORT_AT,
 				      &wwan_hwsim_port_ops,
-				      port);
+				      NULL, port);
 	if (IS_ERR(port->wwan)) {
 		err = PTR_ERR(port->wwan);
 		goto err_free_port;
diff --git a/drivers/usb/class/cdc-wdm.c b/drivers/usb/class/cdc-wdm.c
index 1f0951be15ab..c553decb5461 100644
--- a/drivers/usb/class/cdc-wdm.c
+++ b/drivers/usb/class/cdc-wdm.c
@@ -929,7 +929,8 @@ static void wdm_wwan_init(struct wdm_device *desc)
 		return;
 	}
 
-	port = wwan_create_port(&intf->dev, desc->wwanp_type, &wdm_wwan_port_ops, desc);
+	port = wwan_create_port(&intf->dev, desc->wwanp_type, &wdm_wwan_port_ops,
+				NULL, desc);
 	if (IS_ERR(port)) {
 		dev_err(&intf->dev, "%s: Unable to create WWAN port\n",
 			dev_name(intf->usb_dev));
diff --git a/include/linux/wwan.h b/include/linux/wwan.h
index 24d76500b1cc..01fa15506286 100644
--- a/include/linux/wwan.h
+++ b/include/linux/wwan.h
@@ -64,11 +64,21 @@ struct wwan_port_ops {
 			    poll_table *wait);
 };
 
+/** struct wwan_port_caps - The WWAN port capbilities
+ * @frag_len: WWAN port TX fragments length
+ * @headroom_len: WWAN port TX fragments reserved headroom length
+ */
+struct wwan_port_caps {
+	size_t frag_len;
+	unsigned int headroom_len;
+};
+
 /**
  * wwan_create_port - Add a new WWAN port
  * @parent: Device to use as parent and shared by all WWAN ports
  * @type: WWAN port type
  * @ops: WWAN port operations
+ * @caps: WWAN port capabilities
  * @drvdata: Pointer to caller driver data
  *
  * Allocate and register a new WWAN port. The port will be automatically exposed
@@ -86,6 +96,7 @@ struct wwan_port_ops {
 struct wwan_port *wwan_create_port(struct device *parent,
 				   enum wwan_port_type type,
 				   const struct wwan_port_ops *ops,
+				   struct wwan_port_caps *caps,
 				   void *drvdata);
 
 /**
-- 
cgit v1.2.3-70-g09d2


From df5e87f16c338aa4f62ed3353e73e39e68af965e Mon Sep 17 00:00:00 2001
From: Gavin Li <gavinl@nvidia.com>
Date: Thu, 16 Mar 2023 09:07:54 +0200
Subject: vxlan: Remove unused argument from vxlan_build_gbp_hdr( ) and
 vxlan_build_gpe_hdr( )

Remove unused argument (i.e. u32 vxflags) in vxlan_build_gbp_hdr( ) and
vxlan_build_gpe_hdr( ) function arguments.

Signed-off-by: Gavin Li <gavinl@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vxlan/vxlan_core.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index e2e5f5dac7e6..8cc147c0d069 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -2093,8 +2093,7 @@ static bool route_shortcircuit(struct net_device *dev, struct sk_buff *skb)
 	return false;
 }
 
-static void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, u32 vxflags,
-				struct vxlan_metadata *md)
+static void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, struct vxlan_metadata *md)
 {
 	struct vxlanhdr_gbp *gbp;
 
@@ -2113,8 +2112,7 @@ static void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, u32 vxflags,
 	gbp->policy_id = htons(md->gbp & VXLAN_GBP_ID_MASK);
 }
 
-static int vxlan_build_gpe_hdr(struct vxlanhdr *vxh, u32 vxflags,
-			       __be16 protocol)
+static int vxlan_build_gpe_hdr(struct vxlanhdr *vxh, __be16 protocol)
 {
 	struct vxlanhdr_gpe *gpe = (struct vxlanhdr_gpe *)vxh;
 
@@ -2177,9 +2175,9 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 	}
 
 	if (vxflags & VXLAN_F_GBP)
-		vxlan_build_gbp_hdr(vxh, vxflags, md);
+		vxlan_build_gbp_hdr(vxh, md);
 	if (vxflags & VXLAN_F_GPE) {
-		err = vxlan_build_gpe_hdr(vxh, vxflags, skb->protocol);
+		err = vxlan_build_gpe_hdr(vxh, skb->protocol);
 		if (err < 0)
 			return err;
 		inner_protocol = skb->protocol;
-- 
cgit v1.2.3-70-g09d2


From c641e9279f3530aa2fe4bcb250477b555b75104a Mon Sep 17 00:00:00 2001
From: Gavin Li <gavinl@nvidia.com>
Date: Thu, 16 Mar 2023 09:07:55 +0200
Subject: vxlan: Expose helper vxlan_build_gbp_hdr

The function vxlan_build_gbp_hdr will be used by other modules to build
gbp option in vxlan header according to gbp flags.

Signed-off-by: Gavin Li <gavinl@nvidia.com>
Reviewed-by: Gavi Teitz <gavi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Acked-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vxlan/vxlan_core.c | 19 -------------------
 include/net/vxlan.h            | 19 +++++++++++++++++++
 2 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index 8cc147c0d069..561fe1b314f5 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -2093,25 +2093,6 @@ static bool route_shortcircuit(struct net_device *dev, struct sk_buff *skb)
 	return false;
 }
 
-static void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, struct vxlan_metadata *md)
-{
-	struct vxlanhdr_gbp *gbp;
-
-	if (!md->gbp)
-		return;
-
-	gbp = (struct vxlanhdr_gbp *)vxh;
-	vxh->vx_flags |= VXLAN_HF_GBP;
-
-	if (md->gbp & VXLAN_GBP_DONT_LEARN)
-		gbp->dont_learn = 1;
-
-	if (md->gbp & VXLAN_GBP_POLICY_APPLIED)
-		gbp->policy_applied = 1;
-
-	gbp->policy_id = htons(md->gbp & VXLAN_GBP_ID_MASK);
-}
-
 static int vxlan_build_gpe_hdr(struct vxlanhdr *vxh, __be16 protocol)
 {
 	struct vxlanhdr_gpe *gpe = (struct vxlanhdr_gpe *)vxh;
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index b7b2e9abfb37..20bd7d893e10 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -572,4 +572,23 @@ static inline bool vxlan_fdb_nh_path_select(struct nexthop *nh,
 	return true;
 }
 
+static inline void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, const struct vxlan_metadata *md)
+{
+	struct vxlanhdr_gbp *gbp;
+
+	if (!md->gbp)
+		return;
+
+	gbp = (struct vxlanhdr_gbp *)vxh;
+	vxh->vx_flags |= VXLAN_HF_GBP;
+
+	if (md->gbp & VXLAN_GBP_DONT_LEARN)
+		gbp->dont_learn = 1;
+
+	if (md->gbp & VXLAN_GBP_POLICY_APPLIED)
+		gbp->policy_applied = 1;
+
+	gbp->policy_id = htons(md->gbp & VXLAN_GBP_ID_MASK);
+}
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 58de53c1025836529af9e22e96375d0874e2baca Mon Sep 17 00:00:00 2001
From: Gavin Li <gavinl@nvidia.com>
Date: Thu, 16 Mar 2023 09:07:56 +0200
Subject: net/mlx5e: Add helper for encap_info_equal for tunnels with options

For tunnels with options, eg, geneve and vxlan with gbp, they share the
same way to compare the headers and options. Extract the code as a common
function for them.

Signed-off-by: Gavin Li <gavinl@nvidia.com>
Reviewed-by: Gavi Teitz <gavi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Acked-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/mellanox/mlx5/core/en/tc_tun.h    |  3 ++
 .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c  | 32 ++++++++++++++++++++++
 .../ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c | 24 +---------------
 3 files changed, 36 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
index b38f693bbb52..92065568bb19 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
@@ -115,6 +115,9 @@ int mlx5e_tc_tun_parse_udp_ports(struct mlx5e_priv *priv,
 bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a,
 					   struct mlx5e_encap_key *b);
 
+bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a,
+					   struct mlx5e_encap_key *b,
+					   __be16 tun_flags);
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif //__MLX5_EN_TC_TUNNEL_H__
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
index 780224fd67a1..a108e73c9f66 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
@@ -3,6 +3,7 @@
 
 #include <net/fib_notifier.h>
 #include <net/nexthop.h>
+#include <net/ip_tunnels.h>
 #include "tc_tun_encap.h"
 #include "en_tc.h"
 #include "tc_tun.h"
@@ -571,6 +572,37 @@ bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a,
 		a->tc_tunnel->tunnel_type == b->tc_tunnel->tunnel_type;
 }
 
+bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a,
+					   struct mlx5e_encap_key *b,
+					   __be16 tun_flags)
+{
+	struct ip_tunnel_info *a_info;
+	struct ip_tunnel_info *b_info;
+	bool a_has_opts, b_has_opts;
+
+	if (!mlx5e_tc_tun_encap_info_equal_generic(a, b))
+		return false;
+
+	a_has_opts = !!(a->ip_tun_key->tun_flags & tun_flags);
+	b_has_opts = !!(b->ip_tun_key->tun_flags & tun_flags);
+
+	/* keys are equal when both don't have any options attached */
+	if (!a_has_opts && !b_has_opts)
+		return true;
+
+	if (a_has_opts != b_has_opts)
+		return false;
+
+	/* options stored in memory next to ip_tunnel_info struct */
+	a_info = container_of(a->ip_tun_key, struct ip_tunnel_info, key);
+	b_info = container_of(b->ip_tun_key, struct ip_tunnel_info, key);
+
+	return a_info->options_len == b_info->options_len &&
+	       !memcmp(ip_tunnel_info_opts(a_info),
+		       ip_tunnel_info_opts(b_info),
+		       a_info->options_len);
+}
+
 static int cmp_decap_info(struct mlx5e_decap_key *a,
 			  struct mlx5e_decap_key *b)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c
index 054d80c4e65c..2bcd10b6d653 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c
@@ -337,29 +337,7 @@ static int mlx5e_tc_tun_parse_geneve(struct mlx5e_priv *priv,
 static bool mlx5e_tc_tun_encap_info_equal_geneve(struct mlx5e_encap_key *a,
 						 struct mlx5e_encap_key *b)
 {
-	struct ip_tunnel_info *a_info;
-	struct ip_tunnel_info *b_info;
-	bool a_has_opts, b_has_opts;
-
-	if (!mlx5e_tc_tun_encap_info_equal_generic(a, b))
-		return false;
-
-	a_has_opts = !!(a->ip_tun_key->tun_flags & TUNNEL_GENEVE_OPT);
-	b_has_opts = !!(b->ip_tun_key->tun_flags & TUNNEL_GENEVE_OPT);
-
-	/* keys are equal when both don't have any options attached */
-	if (!a_has_opts && !b_has_opts)
-		return true;
-
-	if (a_has_opts != b_has_opts)
-		return false;
-
-	/* geneve options stored in memory next to ip_tunnel_info struct */
-	a_info = container_of(a->ip_tun_key, struct ip_tunnel_info, key);
-	b_info = container_of(b->ip_tun_key, struct ip_tunnel_info, key);
-
-	return a_info->options_len == b_info->options_len &&
-		memcmp(a_info + 1, b_info + 1, a_info->options_len) == 0;
+	return mlx5e_tc_tun_encap_info_equal_options(a, b, TUNNEL_GENEVE_OPT);
 }
 
 struct mlx5e_tc_tunnel geneve_tunnel = {
-- 
cgit v1.2.3-70-g09d2


From bc9d003dc48c381763c3a6309bfc5eecf9962b9c Mon Sep 17 00:00:00 2001
From: Gavin Li <gavinl@nvidia.com>
Date: Thu, 16 Mar 2023 09:07:57 +0200
Subject: ip_tunnel: Preserve pointer const in ip_tunnel_info_opts

Change ip_tunnel_info_opts( ) from static function to macro to cast return
value and preserve the const-ness of the pointer.

Signed-off-by: Gavin Li <gavinl@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/ip_tunnels.h | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index fca357679816..255b32a90850 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -67,6 +67,12 @@ struct ip_tunnel_key {
 	GENMASK((sizeof_field(struct ip_tunnel_info,		\
 			      options_len) * BITS_PER_BYTE) - 1, 0)
 
+#define ip_tunnel_info_opts(info)				\
+	_Generic(info,						\
+		 const struct ip_tunnel_info * : ((const void *)((info) + 1)),\
+		 struct ip_tunnel_info * : ((void *)((info) + 1))\
+	)
+
 struct ip_tunnel_info {
 	struct ip_tunnel_key	key;
 #ifdef CONFIG_DST_CACHE
@@ -485,11 +491,6 @@ static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len)
 	}
 }
 
-static inline void *ip_tunnel_info_opts(struct ip_tunnel_info *info)
-{
-	return info + 1;
-}
-
 static inline void ip_tunnel_info_opts_get(void *to,
 					   const struct ip_tunnel_info *info)
 {
-- 
cgit v1.2.3-70-g09d2


From 6ee44c518159c364e5d30ed85d357fe1d8e2c141 Mon Sep 17 00:00:00 2001
From: Gavin Li <gavinl@nvidia.com>
Date: Thu, 16 Mar 2023 09:07:58 +0200
Subject: net/mlx5e: TC, Add support for VxLAN GBP encap/decap flows offload

Add HW offloading support for TC flows with VxLAN GBP encap/decap.

Example of encap rule:
tc filter add dev eth0 protocol ip ingress flower \
    action tunnel_key set id 42 vxlan_opts 512 \
    action mirred egress redirect dev vxlan1

Example of decap rule:
tc filter add dev vxlan1 protocol ip ingress flower \
    enc_key_id 42 enc_dst_port 4789 vxlan_opts 1024 \
    action tunnel_key unset action mirred egress redirect dev eth0

Signed-off-by: Gavin Li <gavinl@nvidia.com>
Reviewed-by: Gavi Teitz <gavi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Acked-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c  | 72 +++++++++++++++++++++-
 include/linux/mlx5/device.h                        |  6 ++
 include/linux/mlx5/mlx5_ifc.h                      | 13 +++-
 3 files changed, 87 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
index 1f62c702b625..a184d739d5f8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /* Copyright (c) 2018 Mellanox Technologies. */
 
+#include <net/ip_tunnels.h>
 #include <net/vxlan.h>
 #include "lib/vxlan.h"
 #include "en/tc_tun.h"
@@ -86,9 +87,11 @@ static int mlx5e_gen_ip_tunnel_header_vxlan(char buf[],
 	const struct ip_tunnel_key *tun_key = &e->tun_info->key;
 	__be32 tun_id = tunnel_id_to_key32(tun_key->tun_id);
 	struct udphdr *udp = (struct udphdr *)(buf);
+	const struct vxlan_metadata *md;
 	struct vxlanhdr *vxh;
 
-	if (tun_key->tun_flags & TUNNEL_VXLAN_OPT)
+	if ((tun_key->tun_flags & TUNNEL_VXLAN_OPT) &&
+	    e->tun_info->options_len != sizeof(*md))
 		return -EOPNOTSUPP;
 	vxh = (struct vxlanhdr *)((char *)udp + sizeof(struct udphdr));
 	*ip_proto = IPPROTO_UDP;
@@ -96,6 +99,57 @@ static int mlx5e_gen_ip_tunnel_header_vxlan(char buf[],
 	udp->dest = tun_key->tp_dst;
 	vxh->vx_flags = VXLAN_HF_VNI;
 	vxh->vx_vni = vxlan_vni_field(tun_id);
+	if (tun_key->tun_flags & TUNNEL_VXLAN_OPT) {
+		md = ip_tunnel_info_opts(e->tun_info);
+		vxlan_build_gbp_hdr(vxh, md);
+	}
+
+	return 0;
+}
+
+static int mlx5e_tc_tun_parse_vxlan_gbp_option(struct mlx5e_priv *priv,
+					       struct mlx5_flow_spec *spec,
+					       struct flow_cls_offload *f)
+{
+	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	struct netlink_ext_ack *extack = f->common.extack;
+	struct flow_match_enc_opts enc_opts;
+	void *misc5_c, *misc5_v;
+	u32 *gbp, *gbp_mask;
+
+	flow_rule_match_enc_opts(rule, &enc_opts);
+
+	if (memchr_inv(&enc_opts.mask->data, 0, sizeof(enc_opts.mask->data)) &&
+	    !MLX5_CAP_ESW_FT_FIELD_SUPPORT_2(priv->mdev, tunnel_header_0_1)) {
+		NL_SET_ERR_MSG_MOD(extack, "Matching on VxLAN GBP is not supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (enc_opts.key->dst_opt_type != TUNNEL_VXLAN_OPT) {
+		NL_SET_ERR_MSG_MOD(extack, "Wrong VxLAN option type: not GBP");
+		return -EOPNOTSUPP;
+	}
+
+	if (enc_opts.key->len != sizeof(*gbp) ||
+	    enc_opts.mask->len != sizeof(*gbp_mask)) {
+		NL_SET_ERR_MSG_MOD(extack, "VxLAN GBP option/mask len is not 32 bits");
+		return -EINVAL;
+	}
+
+	gbp = (u32 *)&enc_opts.key->data[0];
+	gbp_mask = (u32 *)&enc_opts.mask->data[0];
+
+	if (*gbp_mask & ~VXLAN_GBP_MASK) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Wrong VxLAN GBP mask(0x%08X)\n", *gbp_mask);
+		return -EINVAL;
+	}
+
+	misc5_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_5);
+	misc5_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_5);
+	MLX5_SET(fte_match_set_misc5, misc5_c, tunnel_header_0, *gbp_mask);
+	MLX5_SET(fte_match_set_misc5, misc5_v, tunnel_header_0, *gbp);
+
+	spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_5;
 
 	return 0;
 }
@@ -122,6 +176,14 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv,
 	if (!enc_keyid.mask->keyid)
 		return 0;
 
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_OPTS)) {
+		int err;
+
+		err = mlx5e_tc_tun_parse_vxlan_gbp_option(priv, spec, f);
+		if (err)
+			return err;
+	}
+
 	/* match on VNI is required */
 
 	if (!MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev,
@@ -143,6 +205,12 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv,
 	return 0;
 }
 
+static bool mlx5e_tc_tun_encap_info_equal_vxlan(struct mlx5e_encap_key *a,
+						struct mlx5e_encap_key *b)
+{
+	return mlx5e_tc_tun_encap_info_equal_options(a, b, TUNNEL_VXLAN_OPT);
+}
+
 static int mlx5e_tc_tun_get_remote_ifindex(struct net_device *mirred_dev)
 {
 	const struct vxlan_dev *vxlan = netdev_priv(mirred_dev);
@@ -160,6 +228,6 @@ struct mlx5e_tc_tunnel vxlan_tunnel = {
 	.generate_ip_tun_hdr  = mlx5e_gen_ip_tunnel_header_vxlan,
 	.parse_udp_ports      = mlx5e_tc_tun_parse_udp_ports_vxlan,
 	.parse_tunnel         = mlx5e_tc_tun_parse_vxlan,
-	.encap_info_equal     = mlx5e_tc_tun_encap_info_equal_generic,
+	.encap_info_equal     = mlx5e_tc_tun_encap_info_equal_vxlan,
 	.get_remote_ifindex   = mlx5e_tc_tun_get_remote_ifindex,
 };
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 71b06ebad402..af4dd536a52c 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -1357,6 +1357,12 @@ enum mlx5_qcam_feature_groups {
 #define MLX5_CAP_ESW_INGRESS_ACL_MAX(mdev, cap) \
 	MLX5_CAP_ESW_FLOWTABLE_MAX(mdev, flow_table_properties_esw_acl_ingress.cap)
 
+#define MLX5_CAP_ESW_FT_FIELD_SUPPORT_2(mdev, cap) \
+	MLX5_CAP_ESW_FLOWTABLE(mdev, ft_field_support_2_esw_fdb.cap)
+
+#define MLX5_CAP_ESW_FT_FIELD_SUPPORT_2_MAX(mdev, cap) \
+	MLX5_CAP_ESW_FLOWTABLE_MAX(mdev, ft_field_support_2_esw_fdb.cap)
+
 #define MLX5_CAP_ESW(mdev, cap) \
 	MLX5_GET(e_switch_cap, \
 		 mdev->caps.hca[MLX5_CAP_ESWITCH]->cur, cap)
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index d2c164f0778c..e47d6c58da35 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -404,10 +404,13 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 	u8         metadata_reg_c_0[0x1];
 };
 
+/* Table 2170 - Flow Table Fields Supported 2 Format */
 struct mlx5_ifc_flow_table_fields_supported_2_bits {
 	u8         reserved_at_0[0xe];
 	u8         bth_opcode[0x1];
-	u8         reserved_at_f[0x11];
+	u8         reserved_at_f[0x1];
+	u8         tunnel_header_0_1[0x1];
+	u8         reserved_at_11[0xf];
 
 	u8         reserved_at_20[0x60];
 };
@@ -895,7 +898,13 @@ struct mlx5_ifc_flow_table_eswitch_cap_bits {
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_esw_acl_egress;
 
-	u8      reserved_at_800[0x1000];
+	u8      reserved_at_800[0xC00];
+
+	struct mlx5_ifc_flow_table_fields_supported_2_bits ft_field_support_2_esw_fdb;
+
+	struct mlx5_ifc_flow_table_fields_supported_2_bits ft_field_bitmask_support_2_esw_fdb;
+
+	u8      reserved_at_1500[0x300];
 
 	u8      sw_steering_fdb_action_drop_icm_address_rx[0x40];
 
-- 
cgit v1.2.3-70-g09d2


From 94c540fbfc80ef95ec398fc77da1920de2946edb Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:30 +0000
Subject: udp: preserve const qualifier in udp_sk()

We can change udp_sk() to propagate const qualifier of its argument,
thanks to container_of_const()

This should avoid some potential errors caused by accidental
(const -> not_const) promotion.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/include/linux/udp.h b/include/linux/udp.h
index a2892e151644..43c1fb2d2c21 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -97,10 +97,7 @@ struct udp_sock {
 
 #define UDP_MAX_SEGMENTS	(1 << 6UL)
 
-static inline struct udp_sock *udp_sk(const struct sock *sk)
-{
-	return (struct udp_sock *)sk;
-}
+#define udp_sk(ptr) container_of_const(ptr, struct udp_sock, inet.sk)
 
 static inline void udp_set_no_check6_tx(struct sock *sk, bool val)
 {
-- 
cgit v1.2.3-70-g09d2


From 68ac9a8b6e65c7cbbe96541353dab1b3f8de2043 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:31 +0000
Subject: af_packet: preserve const qualifier in pkt_sk()

We can change pkt_sk() to propagate const qualifier of its argument,
thanks to container_of_const()

This should avoid some potential errors caused by accidental
(const -> not_const) promotion.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/internal.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/net/packet/internal.h b/net/packet/internal.h
index 680703dbce5e..e793e99646f1 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -133,10 +133,7 @@ struct packet_sock {
 	atomic_t		tp_drops ____cacheline_aligned_in_smp;
 };
 
-static inline struct packet_sock *pkt_sk(struct sock *sk)
-{
-	return (struct packet_sock *)sk;
-}
+#define pkt_sk(ptr) container_of_const(ptr, struct packet_sock, sk)
 
 enum packet_sock_flags {
 	PACKET_SOCK_ORIGDEV,
-- 
cgit v1.2.3-70-g09d2


From 0a2db4630b72486ec2f207ae433c2156b7fd3837 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:32 +0000
Subject: raw: preserve const qualifier in raw_sk()

We can change raw_sk() to propagate const qualifier of its argument,
thanks to container_of_const()

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/include/net/raw.h b/include/net/raw.h
index 7ad15830cf38..c215af02f758 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -83,10 +83,7 @@ struct raw_sock {
 	u32		   ipmr_table;
 };
 
-static inline struct raw_sock *raw_sk(const struct sock *sk)
-{
-	return (struct raw_sock *)sk;
-}
+#define raw_sk(ptr) container_of_const(ptr, struct raw_sock, inet.sk)
 
 static inline bool raw_sk_bound_dev_eq(struct net *net, int bound_dev_if,
 				       int dif, int sdif)
-- 
cgit v1.2.3-70-g09d2


From 47fcae28b9ec409423ba7f67f93e8345acce8a36 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:33 +0000
Subject: ipv6: raw: preserve const qualifier in raw6_sk()

We can change raw6_sk() to propagate its argument const qualifier,
thanks to container_of_const().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 37dfdcfcdd54..839247a4f48e 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -336,10 +336,7 @@ static inline struct ipv6_pinfo *inet6_sk(const struct sock *__sk)
 	return sk_fullsock(__sk) ? inet_sk(__sk)->pinet6 : NULL;
 }
 
-static inline struct raw6_sock *raw6_sk(const struct sock *sk)
-{
-	return (struct raw6_sock *)sk;
-}
+#define raw6_sk(ptr) container_of_const(ptr, struct raw6_sock, inet.sk)
 
 #define ipv6_only_sock(sk)	(sk->sk_ipv6only)
 #define ipv6_sk_rxinfo(sk)	((sk)->sk_family == PF_INET6 && \
-- 
cgit v1.2.3-70-g09d2


From ae6084b739925d84ef4a481d8eaa2187455f2e2f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:34 +0000
Subject: dccp: preserve const qualifier in dccp_sk()

We can change dccp_sk() to propagate its argument const qualifier,
thanks to container_of_const().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 07e547c02fd8..325af611909f 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -305,10 +305,8 @@ struct dccp_sock {
 	struct timer_list		dccps_xmit_timer;
 };
 
-static inline struct dccp_sock *dccp_sk(const struct sock *sk)
-{
-	return (struct dccp_sock *)sk;
-}
+#define dccp_sk(ptr)	container_of_const(ptr, struct dccp_sock, \
+					   dccps_inet_connection.icsk_inet.sk)
 
 static inline const char *dccp_role(const struct sock *sk)
 {
-- 
cgit v1.2.3-70-g09d2


From b064ba9c3cfaf3d22d6153ec3c245eaa4d913674 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:35 +0000
Subject: af_unix: preserve const qualifier in unix_sk()

We can change unix_sk() to propagate its argument const qualifier,
thanks to container_of_const().

We need to change dump_common_audit_data() 'struct unix_sock *u'
local var to get a const attribute.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/af_unix.h | 5 +----
 security/lsm_audit.c  | 2 +-
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index 45ebde587138..824c258143a3 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -74,10 +74,7 @@ struct unix_sock {
 #endif
 };
 
-static inline struct unix_sock *unix_sk(const struct sock *sk)
-{
-	return (struct unix_sock *)sk;
-}
+#define unix_sk(ptr) container_of_const(ptr, struct unix_sock, sk)
 
 #define peer_wait peer_wq.wait
 
diff --git a/security/lsm_audit.c b/security/lsm_audit.c
index 00d3bdd386e2..368e77ca43c4 100644
--- a/security/lsm_audit.c
+++ b/security/lsm_audit.c
@@ -310,7 +310,7 @@ static void dump_common_audit_data(struct audit_buffer *ab,
 	case LSM_AUDIT_DATA_NET:
 		if (a->u.net->sk) {
 			const struct sock *sk = a->u.net->sk;
-			struct unix_sock *u;
+			const struct unix_sock *u;
 			struct unix_address *addr;
 			int len = 0;
 			char *p = NULL;
-- 
cgit v1.2.3-70-g09d2


From 407db475d505d5aa4c2352b80e429a02f5ccd1e4 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:36 +0000
Subject: smc: preserve const qualifier in smc_sk()

We can change smc_sk() to propagate its argument const qualifier,
thanks to container_of_const().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Karsten Graul <kgraul@linux.ibm.com>
Cc: Wenjia Zhang <wenjia@linux.ibm.com>
Cc: Jan Karcher <jaka@linux.ibm.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/smc.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/net/smc/smc.h b/net/smc/smc.h
index 5ed765ea0c73..2eeea4cdc718 100644
--- a/net/smc/smc.h
+++ b/net/smc/smc.h
@@ -283,10 +283,7 @@ struct smc_sock {				/* smc sock container */
 						 * */
 };
 
-static inline struct smc_sock *smc_sk(const struct sock *sk)
-{
-	return (struct smc_sock *)sk;
-}
+#define smc_sk(ptr) container_of_const(ptr, struct smc_sock, sk)
 
 static inline void smc_init_saved_callbacks(struct smc_sock *smc)
 {
-- 
cgit v1.2.3-70-g09d2


From c7154ca8e075cc456fe773879263b33ae307a59e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:37 +0000
Subject: x25: preserve const qualifier in [a]x25_sk()

We can change [a]x25_sk() to propagate their argument const qualifier,
thanks to container_of_const().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ax25.h | 5 +----
 include/net/x25.h  | 5 +----
 2 files changed, 2 insertions(+), 8 deletions(-)

diff --git a/include/net/ax25.h b/include/net/ax25.h
index f8cf3629a419..0d939e5aee4e 100644
--- a/include/net/ax25.h
+++ b/include/net/ax25.h
@@ -260,10 +260,7 @@ struct ax25_sock {
 	struct ax25_cb		*cb;
 };
 
-static inline struct ax25_sock *ax25_sk(const struct sock *sk)
-{
-	return (struct ax25_sock *) sk;
-}
+#define ax25_sk(ptr) container_of_const(ptr, struct ax25_sock, sk)
 
 static inline struct ax25_cb *sk_to_ax25(const struct sock *sk)
 {
diff --git a/include/net/x25.h b/include/net/x25.h
index d7d6c2b4ffa7..597eb53c471e 100644
--- a/include/net/x25.h
+++ b/include/net/x25.h
@@ -177,10 +177,7 @@ struct x25_forward {
 	atomic_t		refcnt;
 };
 
-static inline struct x25_sock *x25_sk(const struct sock *sk)
-{
-	return (struct x25_sock *)sk;
-}
+#define x25_sk(ptr) container_of_const(ptr, struct x25_sock, sk)
 
 /* af_x25.c */
 extern int  sysctl_x25_restart_request_timeout;
-- 
cgit v1.2.3-70-g09d2


From 403a40f2304d4730a780ab9d6a2b93d1e4ac39d2 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:38 +0000
Subject: mptcp: preserve const qualifier in mptcp_sk()

We can change mptcp_sk() to propagate its argument const qualifier,
thanks to container_of_const().

We need to change few things to avoid build errors:

mptcp_set_datafin_timeout() and mptcp_rtx_head() have to accept
non-const sk pointers.

@msk local variable in mptcp_pending_tail() must be const.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Matthieu Baerts <matthieu.baerts@tessares.net>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 2 +-
 net/mptcp/protocol.h | 9 +++------
 2 files changed, 4 insertions(+), 7 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 6738181b2977..2d26b9114373 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -459,7 +459,7 @@ static bool mptcp_pending_data_fin(struct sock *sk, u64 *seq)
 	return false;
 }
 
-static void mptcp_set_datafin_timeout(const struct sock *sk)
+static void mptcp_set_datafin_timeout(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	u32 retransmits;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 339a6f072989..e1310bc113be 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -334,10 +334,7 @@ static inline void msk_owned_by_me(const struct mptcp_sock *msk)
 	sock_owned_by_me((const struct sock *)msk);
 }
 
-static inline struct mptcp_sock *mptcp_sk(const struct sock *sk)
-{
-	return (struct mptcp_sock *)sk;
-}
+#define mptcp_sk(ptr) container_of_const(ptr, struct mptcp_sock, sk.icsk_inet.sk)
 
 /* the msk socket don't use the backlog, also account for the bulk
  * free memory
@@ -371,7 +368,7 @@ static inline struct mptcp_data_frag *mptcp_send_next(struct sock *sk)
 
 static inline struct mptcp_data_frag *mptcp_pending_tail(const struct sock *sk)
 {
-	struct mptcp_sock *msk = mptcp_sk(sk);
+	const struct mptcp_sock *msk = mptcp_sk(sk);
 
 	if (!msk->first_pending)
 		return NULL;
@@ -382,7 +379,7 @@ static inline struct mptcp_data_frag *mptcp_pending_tail(const struct sock *sk)
 	return list_last_entry(&msk->rtx_queue, struct mptcp_data_frag, list);
 }
 
-static inline struct mptcp_data_frag *mptcp_rtx_head(const struct sock *sk)
+static inline struct mptcp_data_frag *mptcp_rtx_head(struct sock *sk)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
-- 
cgit v1.2.3-70-g09d2


From e9d9da91548b21e189fcd0259a0f2d26d1afc509 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 15:55:39 +0000
Subject: tcp: preserve const qualifier in tcp_sk()

We can change tcp_sk() to propagate its argument const qualifier,
thanks to container_of_const().

We have two places where a const sock pointer has to be upgraded
to a write one. We have been using const qualifier for lockless
listeners to clearly identify points where writes could happen.

Add tcp_sk_rw() helper to better document these.

tcp_inbound_md5_hash(), __tcp_grow_window(), tcp_reset_check()
and tcp_rack_reo_wnd() get an additional const qualififer
for their @tp local variables.

smc_check_reset_syn_req() also needs a similar change.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h      | 10 ++++++----
 include/net/tcp.h        |  2 +-
 net/ipv4/tcp.c           |  2 +-
 net/ipv4/tcp_input.c     |  4 ++--
 net/ipv4/tcp_minisocks.c |  5 +++--
 net/ipv4/tcp_output.c    |  9 +++++++--
 net/ipv4/tcp_recovery.c  |  2 +-
 7 files changed, 21 insertions(+), 13 deletions(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index ca7f05a130d2..b4c08ac86983 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -472,10 +472,12 @@ enum tsq_flags {
 	TCPF_MTU_REDUCED_DEFERRED	= (1UL << TCP_MTU_REDUCED_DEFERRED),
 };
 
-static inline struct tcp_sock *tcp_sk(const struct sock *sk)
-{
-	return (struct tcp_sock *)sk;
-}
+#define tcp_sk(ptr) container_of_const(ptr, struct tcp_sock, inet_conn.icsk_inet.sk)
+
+/* Variant of tcp_sk() upgrading a const sock to a read/write tcp socket.
+ * Used in context of (lockless) tcp listeners.
+ */
+#define tcp_sk_rw(ptr) container_of(ptr, struct tcp_sock, inet_conn.icsk_inet.sk)
 
 struct tcp_timewait_sock {
 	struct inet_timewait_sock tw_sk;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index db9f828e9d1e..a0a91a988272 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -529,7 +529,7 @@ static inline void tcp_synq_overflow(const struct sock *sk)
 
 	last_overflow = READ_ONCE(tcp_sk(sk)->rx_opt.ts_recent_stamp);
 	if (!time_between32(now, last_overflow, last_overflow + HZ))
-		WRITE_ONCE(tcp_sk(sk)->rx_opt.ts_recent_stamp, now);
+		WRITE_ONCE(tcp_sk_rw(sk)->rx_opt.ts_recent_stamp, now);
 }
 
 /* syncookies: no recent synqueue overflow on this listening socket? */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 01569de651b6..fd68d49490f2 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4570,7 +4570,7 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 	const __u8 *hash_location = NULL;
 	struct tcp_md5sig_key *hash_expected;
 	const struct tcphdr *th = tcp_hdr(skb);
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 	int genhash, l3index;
 	u8 newhash[16];
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 754ddbe0577f..2b75cd9e2e92 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -458,7 +458,7 @@ static void tcp_sndbuf_expand(struct sock *sk)
 static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb,
 			     unsigned int skbtruesize)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 	/* Optimize this! */
 	int truesize = tcp_win_from_space(sk, skbtruesize) >> 1;
 	int window = tcp_win_from_space(sk, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])) >> 1;
@@ -5693,7 +5693,7 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *t
  */
 static bool tcp_reset_check(const struct sock *sk, const struct sk_buff *skb)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 
 	return unlikely(TCP_SKB_CB(skb)->seq == (tp->rcv_nxt - 1) &&
 			(1 << sk->sk_state) & (TCPF_CLOSE_WAIT | TCPF_LAST_ACK |
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 9a7ef7732c24..dac0d62120e6 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -463,7 +463,7 @@ void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst)
 }
 EXPORT_SYMBOL_GPL(tcp_ca_openreq_child);
 
-static void smc_check_reset_syn_req(struct tcp_sock *oldtp,
+static void smc_check_reset_syn_req(const struct tcp_sock *oldtp,
 				    struct request_sock *req,
 				    struct tcp_sock *newtp)
 {
@@ -492,7 +492,8 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
 	const struct inet_request_sock *ireq = inet_rsk(req);
 	struct tcp_request_sock *treq = tcp_rsk(req);
 	struct inet_connection_sock *newicsk;
-	struct tcp_sock *oldtp, *newtp;
+	const struct tcp_sock *oldtp;
+	struct tcp_sock *newtp;
 	u32 seq;
 
 	if (!newsk)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index b1e743b39a83..cfe128b81a01 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -4127,8 +4127,13 @@ int tcp_rtx_synack(const struct sock *sk, struct request_sock *req)
 	if (!res) {
 		TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS);
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNRETRANS);
-		if (unlikely(tcp_passive_fastopen(sk)))
-			tcp_sk(sk)->total_retrans++;
+		if (unlikely(tcp_passive_fastopen(sk))) {
+			/* sk has const attribute because listeners are lockless.
+			 * However in this case, we are dealing with a passive fastopen
+			 * socket thus we can change total_retrans value.
+			 */
+			tcp_sk_rw(sk)->total_retrans++;
+		}
 		trace_tcp_retransmit_synack(sk, req);
 	}
 	return res;
diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c
index 50abaa941387..acf4869c5d3b 100644
--- a/net/ipv4/tcp_recovery.c
+++ b/net/ipv4/tcp_recovery.c
@@ -4,7 +4,7 @@
 
 static u32 tcp_rack_reo_wnd(const struct sock *sk)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcp_sock *tp = tcp_sk(sk);
 
 	if (!tp->reord_seen) {
 		/* If reordering has not been observed, be aggressive during
-- 
cgit v1.2.3-70-g09d2


From 08ff1c9f3e927ba3701c113dda70953a6f4afffa Mon Sep 17 00:00:00 2001
From: Sreevani Sreejith <ssreevani@meta.com>
Date: Wed, 15 Mar 2023 12:54:05 -0700
Subject: bpf, docs: Libbpf overview documentation

This patch documents overview of libbpf, including its features for
developing BPF programs.

Signed-off-by: Sreevani Sreejith <ssreevani@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230315195405.2051559-1-ssreevani@meta.com
---
 Documentation/bpf/libbpf/index.rst           |  25 ++-
 Documentation/bpf/libbpf/libbpf_overview.rst | 228 +++++++++++++++++++++++++++
 2 files changed, 245 insertions(+), 8 deletions(-)
 create mode 100644 Documentation/bpf/libbpf/libbpf_overview.rst

diff --git a/Documentation/bpf/libbpf/index.rst b/Documentation/bpf/libbpf/index.rst
index f9b3b252e28f..7545a2049692 100644
--- a/Documentation/bpf/libbpf/index.rst
+++ b/Documentation/bpf/libbpf/index.rst
@@ -2,23 +2,32 @@
 
 .. _libbpf:
 
+======
 libbpf
 ======
 
+If you are looking to develop BPF applications using the libbpf library, this
+directory contains important documentation that you should read.
+
+To get started, it is recommended to begin with the :doc:`libbpf Overview
+<libbpf_overview>` document, which provides a high-level understanding of the
+libbpf APIs and their usage. This will give you a solid foundation to start
+exploring and utilizing the various features of libbpf to develop your BPF
+applications.
+
 .. toctree::
    :maxdepth: 1
 
+   libbpf_overview
    API Documentation <https://libbpf.readthedocs.io/en/latest/api.html>
    program_types
    libbpf_naming_convention
    libbpf_build
 
-This is documentation for libbpf, a userspace library for loading and
-interacting with bpf programs.
 
-All general BPF questions, including kernel functionality, libbpf APIs and
-their application, should be sent to bpf@vger.kernel.org mailing list.
-You can `subscribe <http://vger.kernel.org/vger-lists.html#bpf>`_ to the
-mailing list search its `archive <https://lore.kernel.org/bpf/>`_.
-Please search the archive before asking new questions. It very well might
-be that this was already addressed or answered before.
+All general BPF questions, including kernel functionality, libbpf APIs and their
+application, should be sent to bpf@vger.kernel.org mailing list.  You can
+`subscribe <http://vger.kernel.org/vger-lists.html#bpf>`_ to the mailing list
+search its `archive <https://lore.kernel.org/bpf/>`_.  Please search the archive
+before asking new questions. It may be that this was already addressed or
+answered before.
diff --git a/Documentation/bpf/libbpf/libbpf_overview.rst b/Documentation/bpf/libbpf/libbpf_overview.rst
new file mode 100644
index 000000000000..f36a2d4ffea2
--- /dev/null
+++ b/Documentation/bpf/libbpf/libbpf_overview.rst
@@ -0,0 +1,228 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+===============
+libbpf Overview
+===============
+
+libbpf is a C-based library containing a BPF loader that takes compiled BPF
+object files and prepares and loads them into the Linux kernel. libbpf takes the
+heavy lifting of loading, verifying, and attaching BPF programs to various
+kernel hooks, allowing BPF application developers to focus only on BPF program
+correctness and performance.
+
+The following are the high-level features supported by libbpf:
+
+* Provides high-level and low-level APIs for user space programs to interact
+  with BPF programs. The low-level APIs wrap all the bpf system call
+  functionality, which is useful when users need more fine-grained control
+  over the interactions between user space and BPF programs.
+* Provides overall support for the BPF object skeleton generated by bpftool.
+  The skeleton file simplifies the process for the user space programs to access
+  global variables and work with BPF programs.
+* Provides BPF-side APIS, including BPF helper definitions, BPF maps support,
+  and tracing helpers, allowing developers to simplify BPF code writing.
+* Supports BPF CO-RE mechanism, enabling BPF developers to write portable
+  BPF programs that can be compiled once and run across different kernel
+  versions.
+
+This document will delve into the above concepts in detail, providing a deeper
+understanding of the capabilities and advantages of libbpf and how it can help
+you develop BPF applications efficiently.
+
+BPF App Lifecycle and libbpf APIs
+==================================
+
+A BPF application consists of one or more BPF programs (either cooperating or
+completely independent), BPF maps, and global variables. The global
+variables are shared between all BPF programs, which allows them to cooperate on
+a common set of data. libbpf provides APIs that user space programs can use to
+manipulate the BPF programs by triggering different phases of a BPF application
+lifecycle.
+
+The following section provides a brief overview of each phase in the BPF life
+cycle:
+
+* **Open phase**: In this phase, libbpf parses the BPF
+  object file and discovers BPF maps, BPF programs, and global variables. After
+  a BPF app is opened, user space apps can make additional adjustments
+  (setting BPF program types, if necessary; pre-setting initial values for
+  global variables, etc.) before all the entities are created and loaded.
+
+* **Load phase**: In the load phase, libbpf creates BPF
+  maps, resolves various relocations, and verifies and loads BPF programs into
+  the kernel. At this point, libbpf validates all the parts of a BPF application
+  and loads the BPF program into the kernel, but no BPF program has yet been
+  executed. After the load phase, it’s possible to set up the initial BPF map
+  state without racing with the BPF program code execution.
+
+* **Attachment phase**: In this phase, libbpf
+  attaches BPF programs to various BPF hook points (e.g., tracepoints, kprobes,
+  cgroup hooks, network packet processing pipeline, etc.). During this
+  phase, BPF programs perform useful work such as processing
+  packets, or updating BPF maps and global variables that can be read from user
+  space.
+
+* **Tear down phase**: In the tear down phase,
+  libbpf detaches BPF programs and unloads them from the kernel. BPF maps are
+  destroyed, and all the resources used by the BPF app are freed.
+
+BPF Object Skeleton File
+========================
+
+BPF skeleton is an alternative interface to libbpf APIs for working with BPF
+objects. Skeleton code abstract away generic libbpf APIs to significantly
+simplify code for manipulating BPF programs from user space. Skeleton code
+includes a bytecode representation of the BPF object file, simplifying the
+process of distributing your BPF code. With BPF bytecode embedded, there are no
+extra files to deploy along with your application binary.
+
+You can generate the skeleton header file ``(.skel.h)`` for a specific object
+file by passing the BPF object to the bpftool. The generated BPF skeleton
+provides the following custom functions that correspond to the BPF lifecycle,
+each of them prefixed with the specific object name:
+
+* ``<name>__open()`` – creates and opens BPF application (``<name>`` stands for
+  the specific bpf object name)
+* ``<name>__load()`` – instantiates, loads,and verifies BPF application parts
+* ``<name>__attach()`` – attaches all auto-attachable BPF programs (it’s
+  optional, you can have more control by using libbpf APIs directly)
+* ``<name>__destroy()`` – detaches all BPF programs and
+  frees up all used resources
+
+Using the skeleton code is the recommended way to work with bpf programs. Keep
+in mind, BPF skeleton provides access to the underlying BPF object, so whatever
+was possible to do with generic libbpf APIs is still possible even when the BPF
+skeleton is used. It's an additive convenience feature, with no syscalls, and no
+cumbersome code.
+
+Other Advantages of Using Skeleton File
+---------------------------------------
+
+* BPF skeleton provides an interface for user space programs to work with BPF
+  global variables. The skeleton code memory maps global variables as a struct
+  into user space. The struct interface allows user space programs to initialize
+  BPF programs before the BPF load phase and fetch and update data from user
+  space afterward.
+
+* The ``skel.h`` file reflects the object file structure by listing out the
+  available maps, programs, etc. BPF skeleton provides direct access to all the
+  BPF maps and BPF programs as struct fields. This eliminates the need for
+  string-based lookups with ``bpf_object_find_map_by_name()`` and
+  ``bpf_object_find_program_by_name()`` APIs, reducing errors due to BPF source
+  code and user-space code getting out of sync.
+
+* The embedded bytecode representation of the object file ensures that the
+  skeleton and the BPF object file are always in sync.
+
+BPF Helpers
+===========
+
+libbpf provides BPF-side APIs that BPF programs can use to interact with the
+system. The BPF helpers definition allows developers to use them in BPF code as
+any other plain C function. For example, there are helper functions to print
+debugging messages, get the time since the system was booted, interact with BPF
+maps, manipulate network packets, etc.
+
+For a complete description of what the helpers do, the arguments they take, and
+the return value, see the `bpf-helpers
+<https://man7.org/linux/man-pages/man7/bpf-helpers.7.html>`_ man page.
+
+BPF CO-RE (Compile Once – Run Everywhere)
+=========================================
+
+BPF programs work in the kernel space and have access to kernel memory and data
+structures. One limitation that BPF applications come across is the lack of
+portability across different kernel versions and configurations. `BCC
+<https://github.com/iovisor/bcc/>`_ is one of the solutions for BPF
+portability. However, it comes with runtime overhead and a large binary size
+from embedding the compiler with the application.
+
+libbpf steps up the BPF program portability by supporting the BPF CO-RE concept.
+BPF CO-RE brings together BTF type information, libbpf, and the compiler to
+produce a single executable binary that you can run on multiple kernel versions
+and configurations.
+
+To make BPF programs portable libbpf relies on the BTF type information of the
+running kernel. Kernel also exposes this self-describing authoritative BTF
+information through ``sysfs`` at ``/sys/kernel/btf/vmlinux``.
+
+You can generate the BTF information for the running kernel with the following
+command:
+
+::
+
+  $ bpftool btf dump file /sys/kernel/btf/vmlinux format c > vmlinux.h
+
+The command generates a ``vmlinux.h`` header file with all kernel types
+(:doc:`BTF types <../btf>`) that the running kernel uses. Including
+``vmlinux.h`` in your BPF program eliminates dependency on system-wide kernel
+headers.
+
+libbpf enables portability of BPF programs by looking at the BPF program’s
+recorded BTF type and relocation information and matching them to BTF
+information (vmlinux) provided by the running kernel. libbpf then resolves and
+matches all the types and fields, and updates necessary offsets and other
+relocatable data to ensure that BPF program’s logic functions correctly for a
+specific kernel on the host. BPF CO-RE concept thus eliminates overhead
+associated with BPF development and allows developers to write portable BPF
+applications without modifications and runtime source code compilation on the
+target machine.
+
+The following code snippet shows how to read the parent field of a kernel
+``task_struct`` using BPF CO-RE and libbf. The basic helper to read a field in a
+CO-RE relocatable manner is ``bpf_core_read(dst, sz, src)``, which will read
+``sz`` bytes from the field referenced by ``src`` into the memory pointed to by
+``dst``.
+
+.. code-block:: C
+   :emphasize-lines: 6
+
+    //...
+    struct task_struct *task = (void *)bpf_get_current_task();
+    struct task_struct *parent_task;
+    int err;
+
+    err = bpf_core_read(&parent_task, sizeof(void *), &task->parent);
+    if (err) {
+      /* handle error */
+    }
+
+    /* parent_task contains the value of task->parent pointer */
+
+In the code snippet, we first get a pointer to the current ``task_struct`` using
+``bpf_get_current_task()``.  We then use ``bpf_core_read()`` to read the parent
+field of task struct into the ``parent_task`` variable. ``bpf_core_read()`` is
+just like ``bpf_probe_read_kernel()`` BPF helper, except it records information
+about the field that should be relocated on the target kernel. i.e, if the
+``parent`` field gets shifted to a different offset within
+``struct task_struct`` due to some new field added in front of it, libbpf will
+automatically adjust the actual offset to the proper value.
+
+Getting Started with libbpf
+===========================
+
+Check out the `libbpf-bootstrap <https://github.com/libbpf/libbpf-bootstrap>`_
+repository with simple examples of using libbpf to build various BPF
+applications.
+
+See also `libbpf API documentation
+<https://libbpf.readthedocs.io/en/latest/api.html>`_.
+
+libbpf and Rust
+===============
+
+If you are building BPF applications in Rust, it is recommended to use the
+`Libbpf-rs <https://github.com/libbpf/libbpf-rs>`_ library instead of bindgen
+bindings directly to libbpf. Libbpf-rs wraps libbpf functionality in
+Rust-idiomatic interfaces and provides libbpf-cargo plugin to handle BPF code
+compilation and skeleton generation. Using Libbpf-rs will make building user
+space part of the BPF application easier. Note that the BPF program themselves
+must still be written in plain C.
+
+Additional Documentation
+========================
+
+* `Program types and ELF Sections <https://libbpf.readthedocs.io/en/latest/program_types.html>`_
+* `API naming convention <https://libbpf.readthedocs.io/en/latest/libbpf_naming_convention.html>`_
+* `Building libbpf <https://libbpf.readthedocs.io/en/latest/libbpf_build.html>`_
+* `API documentation Convention <https://libbpf.readthedocs.io/en/latest/libbpf_naming_convention.html#api-documentation-convention>`_
-- 
cgit v1.2.3-70-g09d2


From b31587feaa0133b19a1900a26bce07a3c6d21af6 Mon Sep 17 00:00:00 2001
From: Bartosz Wawrzyniak <bwawrzyn@cisco.com>
Date: Thu, 16 Mar 2023 10:03:39 +0000
Subject: net: macb: Set MDIO clock divisor for pclk higher than 160MHz

Currently macb sets clock divisor for pclk up to 160 MHz.
Function gem_mdc_clk_div was updated to enable divisor
for higher values of pclk.

Signed-off-by: Bartosz Wawrzyniak <bwawrzyn@cisco.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/cadence/macb.h      | 2 ++
 drivers/net/ethernet/cadence/macb_main.c | 6 +++++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 14dfec4db8f9..c1fc91c97cee 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -692,6 +692,8 @@
 #define GEM_CLK_DIV48				3
 #define GEM_CLK_DIV64				4
 #define GEM_CLK_DIV96				5
+#define GEM_CLK_DIV128				6
+#define GEM_CLK_DIV224				7
 
 /* Constants for MAN register */
 #define MACB_MAN_C22_SOF			1
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index c4edd20c1c66..7698719b1909 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -2640,8 +2640,12 @@ static u32 gem_mdc_clk_div(struct macb *bp)
 		config = GEM_BF(CLK, GEM_CLK_DIV48);
 	else if (pclk_hz <= 160000000)
 		config = GEM_BF(CLK, GEM_CLK_DIV64);
-	else
+	else if (pclk_hz <= 240000000)
 		config = GEM_BF(CLK, GEM_CLK_DIV96);
+	else if (pclk_hz <= 320000000)
+		config = GEM_BF(CLK, GEM_CLK_DIV128);
+	else
+		config = GEM_BF(CLK, GEM_CLK_DIV224);
 
 	return config;
 }
-- 
cgit v1.2.3-70-g09d2


From c0e906a953f03c56d8029b0207195ee202fd646d Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Thu, 16 Mar 2023 14:04:19 +0200
Subject: net: smc91x: Replace of_gpio.h with what indeed is used

of_gpio.h in this driver is solely used as a proxy to other headers.
This is incorrect usage of the of_gpio.h. Replace it .h with what
indeed is used in the code.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/smsc/smc91x.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c
index 35e99bf0c401..032eccf8eb42 100644
--- a/drivers/net/ethernet/smsc/smc91x.c
+++ b/drivers/net/ethernet/smsc/smc91x.c
@@ -57,6 +57,7 @@ static const char version[] =
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/delay.h>
+#include <linux/gpio/consumer.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>
 #include <linux/errno.h>
@@ -69,7 +70,6 @@ static const char version[] =
 #include <linux/workqueue.h>
 #include <linux/of.h>
 #include <linux/of_device.h>
-#include <linux/of_gpio.h>
 
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
-- 
cgit v1.2.3-70-g09d2


From a593a2fcfdfb92cfd0ffc54bc81b07e6bfaaaf46 Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Thu, 16 Mar 2023 14:08:26 +0200
Subject: net: phy: at803x: Replace of_gpio.h with what indeed is used

of_gpio.h in this driver is solely used as a proxy to other headers.
This is incorrect usage of the of_gpio.h. Replace it .h with what
indeed is used in the code.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/at803x.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c
index 22f4458274aa..656136628ffd 100644
--- a/drivers/net/phy/at803x.c
+++ b/drivers/net/phy/at803x.c
@@ -13,12 +13,11 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/ethtool_netlink.h>
-#include <linux/of_gpio.h>
 #include <linux/bitfield.h>
-#include <linux/gpio/consumer.h>
 #include <linux/regulator/of_regulator.h>
 #include <linux/regulator/driver.h>
 #include <linux/regulator/consumer.h>
+#include <linux/of.h>
 #include <linux/phylink.h>
 #include <linux/sfp.h>
 #include <dt-bindings/net/qca-ar803x.h>
-- 
cgit v1.2.3-70-g09d2


From e485f3a6eae0849f83b94936778a2325f72a0c89 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Fri, 17 Mar 2023 13:09:03 -0700
Subject: ixgb: Remove ixgb driver

There are likely no users of this driver as the hardware has been
discontinued since 2010. Remove the driver and all references to it
in documentation.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/PCI/pci-error-recovery.rst           |    1 -
 .../networking/device_drivers/ethernet/index.rst   |    1 -
 .../device_drivers/ethernet/intel/ixgb.rst         |  468 ----
 arch/loongarch/configs/loongson3_defconfig         |    1 -
 arch/mips/configs/loongson2k_defconfig             |    1 -
 arch/mips/configs/loongson3_defconfig              |    1 -
 arch/mips/configs/mtx1_defconfig                   |    1 -
 arch/powerpc/configs/powernv_defconfig             |    1 -
 arch/powerpc/configs/ppc64_defconfig               |    1 -
 arch/powerpc/configs/ppc64e_defconfig              |    1 -
 arch/powerpc/configs/ppc6xx_defconfig              |    1 -
 arch/powerpc/configs/pseries_defconfig             |    1 -
 arch/powerpc/configs/skiroot_defconfig             |    1 -
 drivers/net/ethernet/intel/Kconfig                 |   17 -
 drivers/net/ethernet/intel/Makefile                |    1 -
 drivers/net/ethernet/intel/ixgb/Makefile           |    9 -
 drivers/net/ethernet/intel/ixgb/ixgb.h             |  179 --
 drivers/net/ethernet/intel/ixgb/ixgb_ee.c          |  580 -----
 drivers/net/ethernet/intel/ixgb/ixgb_ee.h          |   79 -
 drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c     |  642 ------
 drivers/net/ethernet/intel/ixgb/ixgb_hw.c          | 1229 -----------
 drivers/net/ethernet/intel/ixgb/ixgb_hw.h          |  767 -------
 drivers/net/ethernet/intel/ixgb/ixgb_ids.h         |   23 -
 drivers/net/ethernet/intel/ixgb/ixgb_main.c        | 2285 --------------------
 drivers/net/ethernet/intel/ixgb/ixgb_osdep.h       |   39 -
 drivers/net/ethernet/intel/ixgb/ixgb_param.c       |  442 ----
 26 files changed, 6772 deletions(-)
 delete mode 100644 Documentation/networking/device_drivers/ethernet/intel/ixgb.rst
 delete mode 100644 drivers/net/ethernet/intel/ixgb/Makefile
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb.h
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_ee.c
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_ee.h
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_hw.c
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_hw.h
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_ids.h
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_main.c
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_osdep.h
 delete mode 100644 drivers/net/ethernet/intel/ixgb/ixgb_param.c

diff --git a/Documentation/PCI/pci-error-recovery.rst b/Documentation/PCI/pci-error-recovery.rst
index bdafeb4b66dc..9981d330da8f 100644
--- a/Documentation/PCI/pci-error-recovery.rst
+++ b/Documentation/PCI/pci-error-recovery.rst
@@ -418,7 +418,6 @@ That is, the recovery API only requires that:
    - drivers/next/e100.c
    - drivers/net/e1000
    - drivers/net/e1000e
-   - drivers/net/ixgb
    - drivers/net/ixgbe
    - drivers/net/cxgb3
    - drivers/net/s2io.c
diff --git a/Documentation/networking/device_drivers/ethernet/index.rst b/Documentation/networking/device_drivers/ethernet/index.rst
index 392969ac88ad..6e9e7012d000 100644
--- a/Documentation/networking/device_drivers/ethernet/index.rst
+++ b/Documentation/networking/device_drivers/ethernet/index.rst
@@ -31,7 +31,6 @@ Contents:
    intel/fm10k
    intel/igb
    intel/igbvf
-   intel/ixgb
    intel/ixgbe
    intel/ixgbevf
    intel/i40e
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ixgb.rst b/Documentation/networking/device_drivers/ethernet/intel/ixgb.rst
deleted file mode 100644
index c6a233e68ad6..000000000000
--- a/Documentation/networking/device_drivers/ethernet/intel/ixgb.rst
+++ /dev/null
@@ -1,468 +0,0 @@
-.. SPDX-License-Identifier: GPL-2.0+
-
-=====================================================================
-Linux Base Driver for 10 Gigabit Intel(R) Ethernet Network Connection
-=====================================================================
-
-October 1, 2018
-
-
-Contents
-========
-
-- In This Release
-- Identifying Your Adapter
-- Command Line Parameters
-- Improving Performance
-- Additional Configurations
-- Known Issues/Troubleshooting
-- Support
-
-
-
-In This Release
-===============
-
-This file describes the ixgb Linux Base Driver for the 10 Gigabit Intel(R)
-Network Connection.  This driver includes support for Itanium(R)2-based
-systems.
-
-For questions related to hardware requirements, refer to the documentation
-supplied with your 10 Gigabit adapter.  All hardware requirements listed apply
-to use with Linux.
-
-The following features are available in this kernel:
- - Native VLANs
- - Channel Bonding (teaming)
- - SNMP
-
-Channel Bonding documentation can be found in the Linux kernel source:
-/Documentation/networking/bonding.rst
-
-The driver information previously displayed in the /proc filesystem is not
-supported in this release.  Alternatively, you can use ethtool (version 1.6
-or later), lspci, and iproute2 to obtain the same information.
-
-Instructions on updating ethtool can be found in the section "Additional
-Configurations" later in this document.
-
-
-Identifying Your Adapter
-========================
-
-The following Intel network adapters are compatible with the drivers in this
-release:
-
-+------------+------------------------------+----------------------------------+
-| Controller | Adapter Name                 | Physical Layer                   |
-+============+==============================+==================================+
-| 82597EX    | Intel(R) PRO/10GbE LR/SR/CX4 | - 10G Base-LR (fiber)            |
-|            | Server Adapters              | - 10G Base-SR (fiber)            |
-|            |                              | - 10G Base-CX4 (copper)          |
-+------------+------------------------------+----------------------------------+
-
-For more information on how to identify your adapter, go to the Adapter &
-Driver ID Guide at:
-
-    https://support.intel.com
-
-
-Command Line Parameters
-=======================
-
-If the driver is built as a module, the  following optional parameters are
-used by entering them on the command line with the modprobe command using
-this syntax::
-
-    modprobe ixgb [<option>=<VAL1>,<VAL2>,...]
-
-For example, with two 10GbE PCI adapters, entering::
-
-    modprobe ixgb TxDescriptors=80,128
-
-loads the ixgb driver with 80 TX resources for the first adapter and 128 TX
-resources for the second adapter.
-
-The default value for each parameter is generally the recommended setting,
-unless otherwise noted.
-
-Copybreak
----------
-:Valid Range: 0-XXXX
-:Default Value: 256
-
-    This is the maximum size of packet that is copied to a new buffer on
-    receive.
-
-Debug
------
-:Valid Range: 0-16 (0=none,...,16=all)
-:Default Value: 0
-
-    This parameter adjusts the level of debug messages displayed in the
-    system logs.
-
-FlowControl
------------
-:Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx)
-:Default Value: 1 if no EEPROM, otherwise read from EEPROM
-
-    This parameter controls the automatic generation(Tx) and response(Rx) to
-    Ethernet PAUSE frames.  There are hardware bugs associated with enabling
-    Tx flow control so beware.
-
-RxDescriptors
--------------
-:Valid Range: 64-4096
-:Default Value: 1024
-
-    This value is the number of receive descriptors allocated by the driver.
-    Increasing this value allows the driver to buffer more incoming packets.
-    Each descriptor is 16 bytes.  A receive buffer is also allocated for
-    each descriptor and can be either 2048, 4056, 8192, or 16384 bytes,
-    depending on the MTU setting.  When the MTU size is 1500 or less, the
-    receive buffer size is 2048 bytes. When the MTU is greater than 1500 the
-    receive buffer size will be either 4056, 8192, or 16384 bytes.  The
-    maximum MTU size is 16114.
-
-TxDescriptors
--------------
-:Valid Range: 64-4096
-:Default Value: 256
-
-    This value is the number of transmit descriptors allocated by the driver.
-    Increasing this value allows the driver to queue more transmits.  Each
-    descriptor is 16 bytes.
-
-RxIntDelay
-----------
-:Valid Range: 0-65535 (0=off)
-:Default Value: 72
-
-    This value delays the generation of receive interrupts in units of
-    0.8192 microseconds.  Receive interrupt reduction can improve CPU
-    efficiency if properly tuned for specific network traffic.  Increasing
-    this value adds extra latency to frame reception and can end up
-    decreasing the throughput of TCP traffic.  If the system is reporting
-    dropped receives, this value may be set too high, causing the driver to
-    run out of available receive descriptors.
-
-TxIntDelay
-----------
-:Valid Range: 0-65535 (0=off)
-:Default Value: 32
-
-    This value delays the generation of transmit interrupts in units of
-    0.8192 microseconds.  Transmit interrupt reduction can improve CPU
-    efficiency if properly tuned for specific network traffic.  Increasing
-    this value adds extra latency to frame transmission and can end up
-    decreasing the throughput of TCP traffic.  If this value is set too high,
-    it will cause the driver to run out of available transmit descriptors.
-
-XsumRX
-------
-:Valid Range: 0-1
-:Default Value: 1
-
-    A value of '1' indicates that the driver should enable IP checksum
-    offload for received packets (both UDP and TCP) to the adapter hardware.
-
-RxFCHighThresh
---------------
-:Valid Range: 1,536-262,136 (0x600 - 0x3FFF8, 8 byte granularity)
-:Default Value: 196,608 (0x30000)
-
-    Receive Flow control high threshold (when we send a pause frame)
-
-RxFCLowThresh
--------------
-:Valid Range: 64-262,136 (0x40 - 0x3FFF8, 8 byte granularity)
-:Default Value: 163,840 (0x28000)
-
-    Receive Flow control low threshold (when we send a resume frame)
-
-FCReqTimeout
-------------
-:Valid Range: 1-65535
-:Default Value: 65535
-
-    Flow control request timeout (how long to pause the link partner's tx)
-
-IntDelayEnable
---------------
-:Value Range: 0,1
-:Default Value: 1
-
-    Interrupt Delay, 0 disables transmit interrupt delay and 1 enables it.
-
-
-Improving Performance
-=====================
-
-With the 10 Gigabit server adapters, the default Linux configuration will
-very likely limit the total available throughput artificially.  There is a set
-of configuration changes that, when applied together, will increase the ability
-of Linux to transmit and receive data.  The following enhancements were
-originally acquired from settings published at https://www.spec.org/web99/ for
-various submitted results using Linux.
-
-NOTE:
-  These changes are only suggestions, and serve as a starting point for
-  tuning your network performance.
-
-The changes are made in three major ways, listed in order of greatest effect:
-
-- Use ip link to modify the mtu (maximum transmission unit) and the txqueuelen
-  parameter.
-- Use sysctl to modify /proc parameters (essentially kernel tuning)
-- Use setpci to modify the MMRBC field in PCI-X configuration space to increase
-  transmit burst lengths on the bus.
-
-NOTE:
-  setpci modifies the adapter's configuration registers to allow it to read
-  up to 4k bytes at a time (for transmits).  However, for some systems the
-  behavior after modifying this register may be undefined (possibly errors of
-  some kind).  A power-cycle, hard reset or explicitly setting the e6 register
-  back to 22 (setpci -d 8086:1a48 e6.b=22) may be required to get back to a
-  stable configuration.
-
-- COPY these lines and paste them into ixgb_perf.sh:
-
-::
-
-  #!/bin/bash
-  echo "configuring network performance , edit this file to change the interface
-  or device ID of 10GbE card"
-  # set mmrbc to 4k reads, modify only Intel 10GbE device IDs
-  # replace 1a48 with appropriate 10GbE device's ID installed on the system,
-  # if needed.
-  setpci -d 8086:1a48 e6.b=2e
-  # set the MTU (max transmission unit) - it requires your switch and clients
-  # to change as well.
-  # set the txqueuelen
-  # your ixgb adapter should be loaded as eth1 for this to work, change if needed
-  ip li set dev eth1 mtu 9000 txqueuelen 1000 up
-  # call the sysctl utility to modify /proc/sys entries
-  sysctl -p ./sysctl_ixgb.conf
-
-- COPY these lines and paste them into sysctl_ixgb.conf:
-
-::
-
-  # some of the defaults may be different for your kernel
-  # call this file with sysctl -p <this file>
-  # these are just suggested values that worked well to increase throughput in
-  # several network benchmark tests, your mileage may vary
-
-  ### IPV4 specific settings
-  # turn TCP timestamp support off, default 1, reduces CPU use
-  net.ipv4.tcp_timestamps = 0
-  # turn SACK support off, default on
-  # on systems with a VERY fast bus -> memory interface this is the big gainer
-  net.ipv4.tcp_sack = 0
-  # set min/default/max TCP read buffer, default 4096 87380 174760
-  net.ipv4.tcp_rmem = 10000000 10000000 10000000
-  # set min/pressure/max TCP write buffer, default 4096 16384 131072
-  net.ipv4.tcp_wmem = 10000000 10000000 10000000
-  # set min/pressure/max TCP buffer space, default 31744 32256 32768
-  net.ipv4.tcp_mem = 10000000 10000000 10000000
-
-  ### CORE settings (mostly for socket and UDP effect)
-  # set maximum receive socket buffer size, default 131071
-  net.core.rmem_max = 524287
-  # set maximum send socket buffer size, default 131071
-  net.core.wmem_max = 524287
-  # set default receive socket buffer size, default 65535
-  net.core.rmem_default = 524287
-  # set default send socket buffer size, default 65535
-  net.core.wmem_default = 524287
-  # set maximum amount of option memory buffers, default 10240
-  net.core.optmem_max = 524287
-  # set number of unprocessed input packets before kernel starts dropping them; default 300
-  net.core.netdev_max_backlog = 300000
-
-Edit the ixgb_perf.sh script if necessary to change eth1 to whatever interface
-your ixgb driver is using and/or replace '1a48' with appropriate 10GbE device's
-ID installed on the system.
-
-NOTE:
-  Unless these scripts are added to the boot process, these changes will
-  only last only until the next system reboot.
-
-
-Resolving Slow UDP Traffic
---------------------------
-If your server does not seem to be able to receive UDP traffic as fast as it
-can receive TCP traffic, it could be because Linux, by default, does not set
-the network stack buffers as large as they need to be to support high UDP
-transfer rates.  One way to alleviate this problem is to allow more memory to
-be used by the IP stack to store incoming data.
-
-For instance, use the commands::
-
-    sysctl -w net.core.rmem_max=262143
-
-and::
-
-    sysctl -w net.core.rmem_default=262143
-
-to increase the read buffer memory max and default to 262143 (256k - 1) from
-defaults of max=131071 (128k - 1) and default=65535 (64k - 1).  These variables
-will increase the amount of memory used by the network stack for receives, and
-can be increased significantly more if necessary for your application.
-
-
-Additional Configurations
-=========================
-
-Configuring the Driver on Different Distributions
--------------------------------------------------
-Configuring a network driver to load properly when the system is started is
-distribution dependent. Typically, the configuration process involves adding
-an alias line to /etc/modprobe.conf as well as editing other system startup
-scripts and/or configuration files.  Many popular Linux distributions ship
-with tools to make these changes for you.  To learn the proper way to
-configure a network device for your system, refer to your distribution
-documentation.  If during this process you are asked for the driver or module
-name, the name for the Linux Base Driver for the Intel 10GbE Family of
-Adapters is ixgb.
-
-Viewing Link Messages
----------------------
-Link messages will not be displayed to the console if the distribution is
-restricting system messages. In order to see network driver link messages on
-your console, set dmesg to eight by entering the following::
-
-    dmesg -n 8
-
-NOTE: This setting is not saved across reboots.
-
-Jumbo Frames
-------------
-The driver supports Jumbo Frames for all adapters. Jumbo Frames support is
-enabled by changing the MTU to a value larger than the default of 1500.
-The maximum value for the MTU is 16114.  Use the ip command to
-increase the MTU size.  For example::
-
-    ip li set dev ethx mtu 9000
-
-The maximum MTU setting for Jumbo Frames is 16114.  This value coincides
-with the maximum Jumbo Frames size of 16128.
-
-Ethtool
--------
-The driver utilizes the ethtool interface for driver configuration and
-diagnostics, as well as displaying statistical information.  The ethtool
-version 1.6 or later is required for this functionality.
-
-The latest release of ethtool can be found from
-https://www.kernel.org/pub/software/network/ethtool/
-
-NOTE:
-  The ethtool version 1.6 only supports a limited set of ethtool options.
-  Support for a more complete ethtool feature set can be enabled by
-  upgrading to the latest version.
-
-NAPI
-----
-NAPI (Rx polling mode) is supported in the ixgb driver.
-
-See https://wiki.linuxfoundation.org/networking/napi for more information on
-NAPI.
-
-
-Known Issues/Troubleshooting
-============================
-
-NOTE:
-  After installing the driver, if your Intel Network Connection is not
-  working, verify in the "In This Release" section of the readme that you have
-  installed the correct driver.
-
-Cable Interoperability Issue with Fujitsu XENPAK Module in SmartBits Chassis
-----------------------------------------------------------------------------
-Excessive CRC errors may be observed if the Intel(R) PRO/10GbE CX4
-Server adapter is connected to a Fujitsu XENPAK CX4 module in a SmartBits
-chassis using 15 m/24AWG cable assemblies manufactured by Fujitsu or Leoni.
-The CRC errors may be received either by the Intel(R) PRO/10GbE CX4
-Server adapter or the SmartBits. If this situation occurs using a different
-cable assembly may resolve the issue.
-
-Cable Interoperability Issues with HP Procurve 3400cl Switch Port
------------------------------------------------------------------
-Excessive CRC errors may be observed if the Intel(R) PRO/10GbE CX4 Server
-adapter is connected to an HP Procurve 3400cl switch port using short cables
-(1 m or shorter). If this situation occurs, using a longer cable may resolve
-the issue.
-
-Excessive CRC errors may be observed using Fujitsu 24AWG cable assemblies that
-Are 10 m or longer or where using a Leoni 15 m/24AWG cable assembly. The CRC
-errors may be received either by the CX4 Server adapter or at the switch. If
-this situation occurs, using a different cable assembly may resolve the issue.
-
-Jumbo Frames System Requirement
--------------------------------
-Memory allocation failures have been observed on Linux systems with 64 MB
-of RAM or less that are running Jumbo Frames.  If you are using Jumbo
-Frames, your system may require more than the advertised minimum
-requirement of 64 MB of system memory.
-
-Performance Degradation with Jumbo Frames
------------------------------------------
-Degradation in throughput performance may be observed in some Jumbo frames
-environments.  If this is observed, increasing the application's socket buffer
-size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values may help.
-See the specific application manual and /usr/src/linux*/Documentation/
-networking/ip-sysctl.txt for more details.
-
-Allocating Rx Buffers when Using Jumbo Frames
----------------------------------------------
-Allocating Rx buffers when using Jumbo Frames on 2.6.x kernels may fail if
-the available memory is heavily fragmented. This issue may be seen with PCI-X
-adapters or with packet split disabled. This can be reduced or eliminated
-by changing the amount of available memory for receive buffer allocation, by
-increasing /proc/sys/vm/min_free_kbytes.
-
-Multiple Interfaces on Same Ethernet Broadcast Network
-------------------------------------------------------
-Due to the default ARP behavior on Linux, it is not possible to have
-one system on two IP networks in the same Ethernet broadcast domain
-(non-partitioned switch) behave as expected.  All Ethernet interfaces
-will respond to IP traffic for any IP address assigned to the system.
-This results in unbalanced receive traffic.
-
-If you have multiple interfaces in a server, do either of the following:
-
-  - Turn on ARP filtering by entering::
-
-      echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter
-
-  - Install the interfaces in separate broadcast domains - either in
-    different switches or in a switch partitioned to VLANs.
-
-UDP Stress Test Dropped Packet Issue
---------------------------------------
-Under small packets UDP stress test with 10GbE driver, the Linux system
-may drop UDP packets due to the fullness of socket buffers. You may want
-to change the driver's Flow Control variables to the minimum value for
-controlling packet reception.
-
-Tx Hangs Possible Under Stress
-------------------------------
-Under stress conditions, if TX hangs occur, turning off TSO
-"ethtool -K eth0 tso off" may resolve the problem.
-
-
-Support
-=======
-For general information, go to the Intel support website at:
-
-https://www.intel.com/support/
-
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
-If an issue is identified with the released source code on a supported kernel
-with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net
diff --git a/arch/loongarch/configs/loongson3_defconfig b/arch/loongarch/configs/loongson3_defconfig
index e18213f01cc4..6cd26dd3c134 100644
--- a/arch/loongarch/configs/loongson3_defconfig
+++ b/arch/loongarch/configs/loongson3_defconfig
@@ -487,7 +487,6 @@ CONFIG_CHELSIO_T4=m
 CONFIG_E1000=y
 CONFIG_E1000E=y
 CONFIG_IGB=y
-CONFIG_IXGB=y
 CONFIG_IXGBE=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MELLANOX is not set
diff --git a/arch/mips/configs/loongson2k_defconfig b/arch/mips/configs/loongson2k_defconfig
index 728bef666f7a..0ab029ecad21 100644
--- a/arch/mips/configs/loongson2k_defconfig
+++ b/arch/mips/configs/loongson2k_defconfig
@@ -154,7 +154,6 @@ CONFIG_TUN=m
 CONFIG_E1000=y
 CONFIG_E1000E=y
 CONFIG_IGB=y
-CONFIG_IXGB=y
 CONFIG_IXGBE=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MELLANOX is not set
diff --git a/arch/mips/configs/loongson3_defconfig b/arch/mips/configs/loongson3_defconfig
index aca66a5f330d..6f4a52608ea4 100644
--- a/arch/mips/configs/loongson3_defconfig
+++ b/arch/mips/configs/loongson3_defconfig
@@ -207,7 +207,6 @@ CONFIG_VIRTIO_NET=m
 CONFIG_E1000=y
 CONFIG_E1000E=y
 CONFIG_IGB=y
-CONFIG_IXGB=y
 CONFIG_IXGBE=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MELLANOX is not set
diff --git a/arch/mips/configs/mtx1_defconfig b/arch/mips/configs/mtx1_defconfig
index edf9634aa8ee..e1b66aac7025 100644
--- a/arch/mips/configs/mtx1_defconfig
+++ b/arch/mips/configs/mtx1_defconfig
@@ -280,7 +280,6 @@ CONFIG_SUNDANCE=m
 CONFIG_PCMCIA_FMVJ18X=m
 CONFIG_E100=m
 CONFIG_E1000=m
-CONFIG_IXGB=m
 CONFIG_SKGE=m
 CONFIG_SKY2=m
 CONFIG_MYRI10GE=m
diff --git a/arch/powerpc/configs/powernv_defconfig b/arch/powerpc/configs/powernv_defconfig
index c92652575064..f2a9be02a8d2 100644
--- a/arch/powerpc/configs/powernv_defconfig
+++ b/arch/powerpc/configs/powernv_defconfig
@@ -170,7 +170,6 @@ CONFIG_S2IO=m
 CONFIG_E100=y
 CONFIG_E1000=y
 CONFIG_E1000E=y
-CONFIG_IXGB=m
 CONFIG_IXGBE=m
 CONFIG_I40E=m
 CONFIG_MLX4_EN=m
diff --git a/arch/powerpc/configs/ppc64_defconfig b/arch/powerpc/configs/ppc64_defconfig
index d6949a6c5b2b..6c46e5560d96 100644
--- a/arch/powerpc/configs/ppc64_defconfig
+++ b/arch/powerpc/configs/ppc64_defconfig
@@ -182,7 +182,6 @@ CONFIG_IBMVNIC=m
 CONFIG_E100=y
 CONFIG_E1000=y
 CONFIG_E1000E=y
-CONFIG_IXGB=m
 CONFIG_IXGBE=m
 CONFIG_I40E=m
 CONFIG_MLX4_EN=m
diff --git a/arch/powerpc/configs/ppc64e_defconfig b/arch/powerpc/configs/ppc64e_defconfig
index f97a2d31bbf7..776c32964e12 100644
--- a/arch/powerpc/configs/ppc64e_defconfig
+++ b/arch/powerpc/configs/ppc64e_defconfig
@@ -102,7 +102,6 @@ CONFIG_PCNET32=y
 CONFIG_TIGON3=y
 CONFIG_E100=y
 CONFIG_E1000=y
-CONFIG_IXGB=m
 CONFIG_SUNGEM=y
 CONFIG_BROADCOM_PHY=m
 CONFIG_MARVELL_PHY=y
diff --git a/arch/powerpc/configs/ppc6xx_defconfig b/arch/powerpc/configs/ppc6xx_defconfig
index f73c98be56c8..5927b2312936 100644
--- a/arch/powerpc/configs/ppc6xx_defconfig
+++ b/arch/powerpc/configs/ppc6xx_defconfig
@@ -455,7 +455,6 @@ CONFIG_E100=m
 CONFIG_E1000=m
 CONFIG_E1000E=m
 CONFIG_IGB=m
-CONFIG_IXGB=m
 CONFIG_IXGBE=m
 CONFIG_MV643XX_ETH=m
 CONFIG_SKGE=m
diff --git a/arch/powerpc/configs/pseries_defconfig b/arch/powerpc/configs/pseries_defconfig
index 7497e17ea657..49b3ff4e3b18 100644
--- a/arch/powerpc/configs/pseries_defconfig
+++ b/arch/powerpc/configs/pseries_defconfig
@@ -164,7 +164,6 @@ CONFIG_IBMVNIC=y
 CONFIG_E100=y
 CONFIG_E1000=y
 CONFIG_E1000E=y
-CONFIG_IXGB=m
 CONFIG_IXGBE=m
 CONFIG_I40E=m
 CONFIG_MLX4_EN=m
diff --git a/arch/powerpc/configs/skiroot_defconfig b/arch/powerpc/configs/skiroot_defconfig
index e0964210f259..71cfb990a74f 100644
--- a/arch/powerpc/configs/skiroot_defconfig
+++ b/arch/powerpc/configs/skiroot_defconfig
@@ -149,7 +149,6 @@ CONFIG_BE2NET=m
 CONFIG_E1000=m
 CONFIG_E1000E=m
 CONFIG_IGB=m
-CONFIG_IXGB=m
 CONFIG_IXGBE=m
 CONFIG_I40E=m
 # CONFIG_NET_VENDOR_MARVELL is not set
diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
index c18c3b373846..9bc0a9519899 100644
--- a/drivers/net/ethernet/intel/Kconfig
+++ b/drivers/net/ethernet/intel/Kconfig
@@ -139,23 +139,6 @@ config IGBVF
 	  To compile this driver as a module, choose M here. The module
 	  will be called igbvf.
 
-config IXGB
-	tristate "Intel(R) PRO/10GbE support"
-	depends on PCI
-	help
-	  This driver supports Intel(R) PRO/10GbE family of adapters for
-	  PCI-X type cards. For PCI-E type cards, use the "ixgbe" driver
-	  instead. For more information on how to identify your adapter, go
-	  to the Adapter & Driver ID Guide that can be located at:
-
-	  <http://support.intel.com>
-
-	  More specific information on configuring the driver is in
-	  <file:Documentation/networking/device_drivers/ethernet/intel/ixgb.rst>.
-
-	  To compile this driver as a module, choose M here. The module
-	  will be called ixgb.
-
 config IXGBE
 	tristate "Intel(R) 10GbE PCI Express adapters support"
 	depends on PCI
diff --git a/drivers/net/ethernet/intel/Makefile b/drivers/net/ethernet/intel/Makefile
index 3075290063f6..d80d04132073 100644
--- a/drivers/net/ethernet/intel/Makefile
+++ b/drivers/net/ethernet/intel/Makefile
@@ -12,7 +12,6 @@ obj-$(CONFIG_IGBVF) += igbvf/
 obj-$(CONFIG_IXGBE) += ixgbe/
 obj-$(CONFIG_IXGBEVF) += ixgbevf/
 obj-$(CONFIG_I40E) += i40e/
-obj-$(CONFIG_IXGB) += ixgb/
 obj-$(CONFIG_IAVF) += iavf/
 obj-$(CONFIG_FM10K) += fm10k/
 obj-$(CONFIG_ICE) += ice/
diff --git a/drivers/net/ethernet/intel/ixgb/Makefile b/drivers/net/ethernet/intel/ixgb/Makefile
deleted file mode 100644
index 2433e9300a33..000000000000
--- a/drivers/net/ethernet/intel/ixgb/Makefile
+++ /dev/null
@@ -1,9 +0,0 @@
-# SPDX-License-Identifier: GPL-2.0
-# Copyright(c) 1999 - 2008 Intel Corporation.
-#
-# Makefile for the Intel(R) PRO/10GbE ethernet driver
-#
-
-obj-$(CONFIG_IXGB) += ixgb.o
-
-ixgb-objs := ixgb_main.o ixgb_hw.o ixgb_ee.o ixgb_ethtool.o ixgb_param.o
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb.h b/drivers/net/ethernet/intel/ixgb/ixgb.h
deleted file mode 100644
index 81ac39576803..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb.h
+++ /dev/null
@@ -1,179 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#ifndef _IXGB_H_
-#define _IXGB_H_
-
-#include <linux/stddef.h>
-#include <linux/module.h>
-#include <linux/types.h>
-#include <asm/byteorder.h>
-#include <linux/mm.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/pci.h>
-#include <linux/kernel.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/skbuff.h>
-#include <linux/delay.h>
-#include <linux/timer.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <linux/interrupt.h>
-#include <linux/string.h>
-#include <linux/pagemap.h>
-#include <linux/dma-mapping.h>
-#include <linux/bitops.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <linux/capability.h>
-#include <linux/in.h>
-#include <linux/ip.h>
-#include <linux/tcp.h>
-#include <linux/udp.h>
-#include <net/pkt_sched.h>
-#include <linux/list.h>
-#include <linux/reboot.h>
-#include <net/checksum.h>
-
-#include <linux/ethtool.h>
-#include <linux/if_vlan.h>
-
-#define BAR_0		0
-#define BAR_1		1
-
-struct ixgb_adapter;
-#include "ixgb_hw.h"
-#include "ixgb_ee.h"
-#include "ixgb_ids.h"
-
-/* TX/RX descriptor defines */
-#define DEFAULT_TXD      256
-#define MAX_TXD         4096
-#define MIN_TXD           64
-
-/* hardware cannot reliably support more than 512 descriptors owned by
- * hardware descriptor cache otherwise an unreliable ring under heavy
- * receive load may result */
-#define DEFAULT_RXD      512
-#define MAX_RXD          512
-#define MIN_RXD           64
-
-/* Supported Rx Buffer Sizes */
-#define IXGB_RXBUFFER_2048  2048
-#define IXGB_RXBUFFER_4096  4096
-#define IXGB_RXBUFFER_8192  8192
-#define IXGB_RXBUFFER_16384 16384
-
-/* How many Rx Buffers do we bundle into one write to the hardware ? */
-#define IXGB_RX_BUFFER_WRITE	8	/* Must be power of 2 */
-
-/* wrapper around a pointer to a socket buffer,
- * so a DMA handle can be stored along with the buffer */
-struct ixgb_buffer {
-	struct sk_buff *skb;
-	dma_addr_t dma;
-	unsigned long time_stamp;
-	u16 length;
-	u16 next_to_watch;
-	u16 mapped_as_page;
-};
-
-struct ixgb_desc_ring {
-	/* pointer to the descriptor ring memory */
-	void *desc;
-	/* physical address of the descriptor ring */
-	dma_addr_t dma;
-	/* length of descriptor ring in bytes */
-	unsigned int size;
-	/* number of descriptors in the ring */
-	unsigned int count;
-	/* next descriptor to associate a buffer with */
-	unsigned int next_to_use;
-	/* next descriptor to check for DD status bit */
-	unsigned int next_to_clean;
-	/* array of buffer information structs */
-	struct ixgb_buffer *buffer_info;
-};
-
-#define IXGB_DESC_UNUSED(R) \
-	((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \
-	(R)->next_to_clean - (R)->next_to_use - 1)
-
-#define IXGB_GET_DESC(R, i, type)	(&(((struct type *)((R).desc))[i]))
-#define IXGB_RX_DESC(R, i)		IXGB_GET_DESC(R, i, ixgb_rx_desc)
-#define IXGB_TX_DESC(R, i)		IXGB_GET_DESC(R, i, ixgb_tx_desc)
-#define IXGB_CONTEXT_DESC(R, i)	IXGB_GET_DESC(R, i, ixgb_context_desc)
-
-/* board specific private data structure */
-
-struct ixgb_adapter {
-	struct timer_list watchdog_timer;
-	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
-	u32 bd_number;
-	u32 rx_buffer_len;
-	u32 part_num;
-	u16 link_speed;
-	u16 link_duplex;
-	struct work_struct tx_timeout_task;
-
-	/* TX */
-	struct ixgb_desc_ring tx_ring ____cacheline_aligned_in_smp;
-	unsigned int restart_queue;
-	unsigned long timeo_start;
-	u32 tx_cmd_type;
-	u64 hw_csum_tx_good;
-	u64 hw_csum_tx_error;
-	u32 tx_int_delay;
-	u32 tx_timeout_count;
-	bool tx_int_delay_enable;
-	bool detect_tx_hung;
-
-	/* RX */
-	struct ixgb_desc_ring rx_ring;
-	u64 hw_csum_rx_error;
-	u64 hw_csum_rx_good;
-	u32 rx_int_delay;
-	bool rx_csum;
-
-	/* OS defined structs */
-	struct napi_struct napi;
-	struct net_device *netdev;
-	struct pci_dev *pdev;
-
-	/* structs defined in ixgb_hw.h */
-	struct ixgb_hw hw;
-	u16 msg_enable;
-	struct ixgb_hw_stats stats;
-	u32 alloc_rx_buff_failed;
-	bool have_msi;
-	unsigned long flags;
-};
-
-enum ixgb_state_t {
-	/* TBD
-	__IXGB_TESTING,
-	__IXGB_RESETTING,
-	*/
-	__IXGB_DOWN
-};
-
-/* Exported from other modules */
-void ixgb_check_options(struct ixgb_adapter *adapter);
-void ixgb_set_ethtool_ops(struct net_device *netdev);
-extern char ixgb_driver_name[];
-
-void ixgb_set_speed_duplex(struct net_device *netdev);
-
-int ixgb_up(struct ixgb_adapter *adapter);
-void ixgb_down(struct ixgb_adapter *adapter, bool kill_watchdog);
-void ixgb_reset(struct ixgb_adapter *adapter);
-int ixgb_setup_rx_resources(struct ixgb_adapter *adapter);
-int ixgb_setup_tx_resources(struct ixgb_adapter *adapter);
-void ixgb_free_rx_resources(struct ixgb_adapter *adapter);
-void ixgb_free_tx_resources(struct ixgb_adapter *adapter);
-void ixgb_update_stats(struct ixgb_adapter *adapter);
-
-
-#endif /* _IXGB_H_ */
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_ee.c b/drivers/net/ethernet/intel/ixgb/ixgb_ee.c
deleted file mode 100644
index 129286fc1634..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_ee.c
+++ /dev/null
@@ -1,580 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include "ixgb_hw.h"
-#include "ixgb_ee.h"
-/* Local prototypes */
-static u16 ixgb_shift_in_bits(struct ixgb_hw *hw);
-
-static void ixgb_shift_out_bits(struct ixgb_hw *hw,
-				u16 data,
-				u16 count);
-static void ixgb_standby_eeprom(struct ixgb_hw *hw);
-
-static bool ixgb_wait_eeprom_command(struct ixgb_hw *hw);
-
-static void ixgb_cleanup_eeprom(struct ixgb_hw *hw);
-
-/******************************************************************************
- * Raises the EEPROM's clock input.
- *
- * hw - Struct containing variables accessed by shared code
- * eecd_reg - EECD's current value
- *****************************************************************************/
-static void
-ixgb_raise_clock(struct ixgb_hw *hw,
-		  u32 *eecd_reg)
-{
-	/* Raise the clock input to the EEPROM (by setting the SK bit), and then
-	 *  wait 50 microseconds.
-	 */
-	*eecd_reg = *eecd_reg | IXGB_EECD_SK;
-	IXGB_WRITE_REG(hw, EECD, *eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-}
-
-/******************************************************************************
- * Lowers the EEPROM's clock input.
- *
- * hw - Struct containing variables accessed by shared code
- * eecd_reg - EECD's current value
- *****************************************************************************/
-static void
-ixgb_lower_clock(struct ixgb_hw *hw,
-		  u32 *eecd_reg)
-{
-	/* Lower the clock input to the EEPROM (by clearing the SK bit), and then
-	 * wait 50 microseconds.
-	 */
-	*eecd_reg = *eecd_reg & ~IXGB_EECD_SK;
-	IXGB_WRITE_REG(hw, EECD, *eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-}
-
-/******************************************************************************
- * Shift data bits out to the EEPROM.
- *
- * hw - Struct containing variables accessed by shared code
- * data - data to send to the EEPROM
- * count - number of bits to shift out
- *****************************************************************************/
-static void
-ixgb_shift_out_bits(struct ixgb_hw *hw,
-					 u16 data,
-					 u16 count)
-{
-	u32 eecd_reg;
-	u32 mask;
-
-	/* We need to shift "count" bits out to the EEPROM. So, value in the
-	 * "data" parameter will be shifted out to the EEPROM one bit at a time.
-	 * In order to do this, "data" must be broken down into bits.
-	 */
-	mask = 0x01 << (count - 1);
-	eecd_reg = IXGB_READ_REG(hw, EECD);
-	eecd_reg &= ~(IXGB_EECD_DO | IXGB_EECD_DI);
-	do {
-		/* A "1" is shifted out to the EEPROM by setting bit "DI" to a "1",
-		 * and then raising and then lowering the clock (the SK bit controls
-		 * the clock input to the EEPROM).  A "0" is shifted out to the EEPROM
-		 * by setting "DI" to "0" and then raising and then lowering the clock.
-		 */
-		eecd_reg &= ~IXGB_EECD_DI;
-
-		if (data & mask)
-			eecd_reg |= IXGB_EECD_DI;
-
-		IXGB_WRITE_REG(hw, EECD, eecd_reg);
-		IXGB_WRITE_FLUSH(hw);
-
-		udelay(50);
-
-		ixgb_raise_clock(hw, &eecd_reg);
-		ixgb_lower_clock(hw, &eecd_reg);
-
-		mask = mask >> 1;
-
-	} while (mask);
-
-	/* We leave the "DI" bit set to "0" when we leave this routine. */
-	eecd_reg &= ~IXGB_EECD_DI;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-}
-
-/******************************************************************************
- * Shift data bits in from the EEPROM
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static u16
-ixgb_shift_in_bits(struct ixgb_hw *hw)
-{
-	u32 eecd_reg;
-	u32 i;
-	u16 data;
-
-	/* In order to read a register from the EEPROM, we need to shift 16 bits
-	 * in from the EEPROM. Bits are "shifted in" by raising the clock input to
-	 * the EEPROM (setting the SK bit), and then reading the value of the "DO"
-	 * bit.  During this "shifting in" process the "DI" bit should always be
-	 * clear..
-	 */
-
-	eecd_reg = IXGB_READ_REG(hw, EECD);
-
-	eecd_reg &= ~(IXGB_EECD_DO | IXGB_EECD_DI);
-	data = 0;
-
-	for (i = 0; i < 16; i++) {
-		data = data << 1;
-		ixgb_raise_clock(hw, &eecd_reg);
-
-		eecd_reg = IXGB_READ_REG(hw, EECD);
-
-		eecd_reg &= ~(IXGB_EECD_DI);
-		if (eecd_reg & IXGB_EECD_DO)
-			data |= 1;
-
-		ixgb_lower_clock(hw, &eecd_reg);
-	}
-
-	return data;
-}
-
-/******************************************************************************
- * Prepares EEPROM for access
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Lowers EEPROM clock. Clears input pin. Sets the chip select pin. This
- * function should be called before issuing a command to the EEPROM.
- *****************************************************************************/
-static void
-ixgb_setup_eeprom(struct ixgb_hw *hw)
-{
-	u32 eecd_reg;
-
-	eecd_reg = IXGB_READ_REG(hw, EECD);
-
-	/*  Clear SK and DI  */
-	eecd_reg &= ~(IXGB_EECD_SK | IXGB_EECD_DI);
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-
-	/*  Set CS  */
-	eecd_reg |= IXGB_EECD_CS;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-}
-
-/******************************************************************************
- * Returns EEPROM to a "standby" state
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_standby_eeprom(struct ixgb_hw *hw)
-{
-	u32 eecd_reg;
-
-	eecd_reg = IXGB_READ_REG(hw, EECD);
-
-	/*  Deselect EEPROM  */
-	eecd_reg &= ~(IXGB_EECD_CS | IXGB_EECD_SK);
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-
-	/*  Clock high  */
-	eecd_reg |= IXGB_EECD_SK;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-
-	/*  Select EEPROM  */
-	eecd_reg |= IXGB_EECD_CS;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-
-	/*  Clock low  */
-	eecd_reg &= ~IXGB_EECD_SK;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-}
-
-/******************************************************************************
- * Raises then lowers the EEPROM's clock pin
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_clock_eeprom(struct ixgb_hw *hw)
-{
-	u32 eecd_reg;
-
-	eecd_reg = IXGB_READ_REG(hw, EECD);
-
-	/*  Rising edge of clock  */
-	eecd_reg |= IXGB_EECD_SK;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-
-	/*  Falling edge of clock  */
-	eecd_reg &= ~IXGB_EECD_SK;
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-	IXGB_WRITE_FLUSH(hw);
-	udelay(50);
-}
-
-/******************************************************************************
- * Terminates a command by lowering the EEPROM's chip select pin
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_cleanup_eeprom(struct ixgb_hw *hw)
-{
-	u32 eecd_reg;
-
-	eecd_reg = IXGB_READ_REG(hw, EECD);
-
-	eecd_reg &= ~(IXGB_EECD_CS | IXGB_EECD_DI);
-
-	IXGB_WRITE_REG(hw, EECD, eecd_reg);
-
-	ixgb_clock_eeprom(hw);
-}
-
-/******************************************************************************
- * Waits for the EEPROM to finish the current command.
- *
- * hw - Struct containing variables accessed by shared code
- *
- * The command is done when the EEPROM's data out pin goes high.
- *
- * Returns:
- *      true: EEPROM data pin is high before timeout.
- *      false:  Time expired.
- *****************************************************************************/
-static bool
-ixgb_wait_eeprom_command(struct ixgb_hw *hw)
-{
-	u32 eecd_reg;
-	u32 i;
-
-	/* Toggle the CS line.  This in effect tells to EEPROM to actually execute
-	 * the command in question.
-	 */
-	ixgb_standby_eeprom(hw);
-
-	/* Now read DO repeatedly until is high (equal to '1').  The EEPROM will
-	 * signal that the command has been completed by raising the DO signal.
-	 * If DO does not go high in 10 milliseconds, then error out.
-	 */
-	for (i = 0; i < 200; i++) {
-		eecd_reg = IXGB_READ_REG(hw, EECD);
-
-		if (eecd_reg & IXGB_EECD_DO)
-			return true;
-
-		udelay(50);
-	}
-	ASSERT(0);
-	return false;
-}
-
-/******************************************************************************
- * Verifies that the EEPROM has a valid checksum
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Reads the first 64 16 bit words of the EEPROM and sums the values read.
- * If the sum of the 64 16 bit words is 0xBABA, the EEPROM's checksum is
- * valid.
- *
- * Returns:
- *  true: Checksum is valid
- *  false: Checksum is not valid.
- *****************************************************************************/
-bool
-ixgb_validate_eeprom_checksum(struct ixgb_hw *hw)
-{
-	u16 checksum = 0;
-	u16 i;
-
-	for (i = 0; i < (EEPROM_CHECKSUM_REG + 1); i++)
-		checksum += ixgb_read_eeprom(hw, i);
-
-	if (checksum == (u16) EEPROM_SUM)
-		return true;
-	else
-		return false;
-}
-
-/******************************************************************************
- * Calculates the EEPROM checksum and writes it to the EEPROM
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Sums the first 63 16 bit words of the EEPROM. Subtracts the sum from 0xBABA.
- * Writes the difference to word offset 63 of the EEPROM.
- *****************************************************************************/
-void
-ixgb_update_eeprom_checksum(struct ixgb_hw *hw)
-{
-	u16 checksum = 0;
-	u16 i;
-
-	for (i = 0; i < EEPROM_CHECKSUM_REG; i++)
-		checksum += ixgb_read_eeprom(hw, i);
-
-	checksum = (u16) EEPROM_SUM - checksum;
-
-	ixgb_write_eeprom(hw, EEPROM_CHECKSUM_REG, checksum);
-}
-
-/******************************************************************************
- * Writes a 16 bit word to a given offset in the EEPROM.
- *
- * hw - Struct containing variables accessed by shared code
- * reg - offset within the EEPROM to be written to
- * data - 16 bit word to be written to the EEPROM
- *
- * If ixgb_update_eeprom_checksum is not called after this function, the
- * EEPROM will most likely contain an invalid checksum.
- *
- *****************************************************************************/
-void
-ixgb_write_eeprom(struct ixgb_hw *hw, u16 offset, u16 data)
-{
-	struct ixgb_ee_map_type *ee_map = (struct ixgb_ee_map_type *)hw->eeprom;
-
-	/* Prepare the EEPROM for writing */
-	ixgb_setup_eeprom(hw);
-
-	/*  Send the 9-bit EWEN (write enable) command to the EEPROM (5-bit opcode
-	 *  plus 4-bit dummy).  This puts the EEPROM into write/erase mode.
-	 */
-	ixgb_shift_out_bits(hw, EEPROM_EWEN_OPCODE, 5);
-	ixgb_shift_out_bits(hw, 0, 4);
-
-	/*  Prepare the EEPROM  */
-	ixgb_standby_eeprom(hw);
-
-	/*  Send the Write command (3-bit opcode + 6-bit addr)  */
-	ixgb_shift_out_bits(hw, EEPROM_WRITE_OPCODE, 3);
-	ixgb_shift_out_bits(hw, offset, 6);
-
-	/*  Send the data  */
-	ixgb_shift_out_bits(hw, data, 16);
-
-	ixgb_wait_eeprom_command(hw);
-
-	/*  Recover from write  */
-	ixgb_standby_eeprom(hw);
-
-	/* Send the 9-bit EWDS (write disable) command to the EEPROM (5-bit
-	 * opcode plus 4-bit dummy).  This takes the EEPROM out of write/erase
-	 * mode.
-	 */
-	ixgb_shift_out_bits(hw, EEPROM_EWDS_OPCODE, 5);
-	ixgb_shift_out_bits(hw, 0, 4);
-
-	/*  Done with writing  */
-	ixgb_cleanup_eeprom(hw);
-
-	/* clear the init_ctrl_reg_1 to signify that the cache is invalidated */
-	ee_map->init_ctrl_reg_1 = cpu_to_le16(EEPROM_ICW1_SIGNATURE_CLEAR);
-}
-
-/******************************************************************************
- * Reads a 16 bit word from the EEPROM.
- *
- * hw - Struct containing variables accessed by shared code
- * offset - offset of 16 bit word in the EEPROM to read
- *
- * Returns:
- *  The 16-bit value read from the eeprom
- *****************************************************************************/
-u16
-ixgb_read_eeprom(struct ixgb_hw *hw,
-		  u16 offset)
-{
-	u16 data;
-
-	/*  Prepare the EEPROM for reading  */
-	ixgb_setup_eeprom(hw);
-
-	/*  Send the READ command (opcode + addr)  */
-	ixgb_shift_out_bits(hw, EEPROM_READ_OPCODE, 3);
-	/*
-	 * We have a 64 word EEPROM, there are 6 address bits
-	 */
-	ixgb_shift_out_bits(hw, offset, 6);
-
-	/*  Read the data  */
-	data = ixgb_shift_in_bits(hw);
-
-	/*  End this read operation  */
-	ixgb_standby_eeprom(hw);
-
-	return data;
-}
-
-/******************************************************************************
- * Reads eeprom and stores data in shared structure.
- * Validates eeprom checksum and eeprom signature.
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Returns:
- *      true: if eeprom read is successful
- *      false: otherwise.
- *****************************************************************************/
-bool
-ixgb_get_eeprom_data(struct ixgb_hw *hw)
-{
-	u16 i;
-	u16 checksum = 0;
-	struct ixgb_ee_map_type *ee_map;
-
-	ENTER();
-
-	ee_map = (struct ixgb_ee_map_type *)hw->eeprom;
-
-	pr_debug("Reading eeprom data\n");
-	for (i = 0; i < IXGB_EEPROM_SIZE ; i++) {
-		u16 ee_data;
-		ee_data = ixgb_read_eeprom(hw, i);
-		checksum += ee_data;
-		hw->eeprom[i] = cpu_to_le16(ee_data);
-	}
-
-	if (checksum != (u16) EEPROM_SUM) {
-		pr_debug("Checksum invalid\n");
-		/* clear the init_ctrl_reg_1 to signify that the cache is
-		 * invalidated */
-		ee_map->init_ctrl_reg_1 = cpu_to_le16(EEPROM_ICW1_SIGNATURE_CLEAR);
-		return false;
-	}
-
-	if ((ee_map->init_ctrl_reg_1 & cpu_to_le16(EEPROM_ICW1_SIGNATURE_MASK))
-		 != cpu_to_le16(EEPROM_ICW1_SIGNATURE_VALID)) {
-		pr_debug("Signature invalid\n");
-		return false;
-	}
-
-	return true;
-}
-
-/******************************************************************************
- * Local function to check if the eeprom signature is good
- * If the eeprom signature is good, calls ixgb)get_eeprom_data.
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Returns:
- *      true: eeprom signature was good and the eeprom read was successful
- *      false: otherwise.
- ******************************************************************************/
-static bool
-ixgb_check_and_get_eeprom_data (struct ixgb_hw* hw)
-{
-	struct ixgb_ee_map_type *ee_map = (struct ixgb_ee_map_type *)hw->eeprom;
-
-	if ((ee_map->init_ctrl_reg_1 & cpu_to_le16(EEPROM_ICW1_SIGNATURE_MASK))
-	    == cpu_to_le16(EEPROM_ICW1_SIGNATURE_VALID)) {
-		return true;
-	} else {
-		return ixgb_get_eeprom_data(hw);
-	}
-}
-
-/******************************************************************************
- * return a word from the eeprom
- *
- * hw - Struct containing variables accessed by shared code
- * index - Offset of eeprom word
- *
- * Returns:
- *          Word at indexed offset in eeprom, if valid, 0 otherwise.
- ******************************************************************************/
-__le16
-ixgb_get_eeprom_word(struct ixgb_hw *hw, u16 index)
-{
-
-	if (index < IXGB_EEPROM_SIZE && ixgb_check_and_get_eeprom_data(hw))
-		return hw->eeprom[index];
-
-	return 0;
-}
-
-/******************************************************************************
- * return the mac address from EEPROM
- *
- * hw       - Struct containing variables accessed by shared code
- * mac_addr - Ethernet Address if EEPROM contents are valid, 0 otherwise
- *
- * Returns: None.
- ******************************************************************************/
-void
-ixgb_get_ee_mac_addr(struct ixgb_hw *hw,
-			u8 *mac_addr)
-{
-	int i;
-	struct ixgb_ee_map_type *ee_map = (struct ixgb_ee_map_type *)hw->eeprom;
-
-	ENTER();
-
-	if (ixgb_check_and_get_eeprom_data(hw)) {
-		for (i = 0; i < ETH_ALEN; i++) {
-			mac_addr[i] = ee_map->mac_addr[i];
-		}
-		pr_debug("eeprom mac address = %pM\n", mac_addr);
-	}
-}
-
-
-/******************************************************************************
- * return the Printed Board Assembly number from EEPROM
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Returns:
- *          PBA number if EEPROM contents are valid, 0 otherwise
- ******************************************************************************/
-u32
-ixgb_get_ee_pba_number(struct ixgb_hw *hw)
-{
-	if (ixgb_check_and_get_eeprom_data(hw))
-		return le16_to_cpu(hw->eeprom[EEPROM_PBA_1_2_REG])
-			| (le16_to_cpu(hw->eeprom[EEPROM_PBA_3_4_REG])<<16);
-
-	return 0;
-}
-
-
-/******************************************************************************
- * return the Device Id from EEPROM
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Returns:
- *          Device Id if EEPROM contents are valid, 0 otherwise
- ******************************************************************************/
-u16
-ixgb_get_ee_device_id(struct ixgb_hw *hw)
-{
-	struct ixgb_ee_map_type *ee_map = (struct ixgb_ee_map_type *)hw->eeprom;
-
-	if (ixgb_check_and_get_eeprom_data(hw))
-		return le16_to_cpu(ee_map->device_id);
-
-	return 0;
-}
-
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_ee.h b/drivers/net/ethernet/intel/ixgb/ixgb_ee.h
deleted file mode 100644
index 3ee0a09e5d0a..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_ee.h
+++ /dev/null
@@ -1,79 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#ifndef _IXGB_EE_H_
-#define _IXGB_EE_H_
-
-#define IXGB_EEPROM_SIZE    64	/* Size in words */
-
-/* EEPROM Commands */
-#define EEPROM_READ_OPCODE  0x6	/* EEPROM read opcode */
-#define EEPROM_WRITE_OPCODE 0x5	/* EEPROM write opcode */
-#define EEPROM_ERASE_OPCODE 0x7	/* EEPROM erase opcode */
-#define EEPROM_EWEN_OPCODE  0x13	/* EEPROM erase/write enable */
-#define EEPROM_EWDS_OPCODE  0x10	/* EEPROM erase/write disable */
-
-/* EEPROM MAP (Word Offsets) */
-#define EEPROM_IA_1_2_REG        0x0000
-#define EEPROM_IA_3_4_REG        0x0001
-#define EEPROM_IA_5_6_REG        0x0002
-#define EEPROM_COMPATIBILITY_REG 0x0003
-#define EEPROM_PBA_1_2_REG       0x0008
-#define EEPROM_PBA_3_4_REG       0x0009
-#define EEPROM_INIT_CONTROL1_REG 0x000A
-#define EEPROM_SUBSYS_ID_REG     0x000B
-#define EEPROM_SUBVEND_ID_REG    0x000C
-#define EEPROM_DEVICE_ID_REG     0x000D
-#define EEPROM_VENDOR_ID_REG     0x000E
-#define EEPROM_INIT_CONTROL2_REG 0x000F
-#define EEPROM_SWDPINS_REG       0x0020
-#define EEPROM_CIRCUIT_CTRL_REG  0x0021
-#define EEPROM_D0_D3_POWER_REG   0x0022
-#define EEPROM_FLASH_VERSION     0x0032
-#define EEPROM_CHECKSUM_REG      0x003F
-
-/* Mask bits for fields in Word 0x0a of the EEPROM */
-
-#define EEPROM_ICW1_SIGNATURE_MASK  0xC000
-#define EEPROM_ICW1_SIGNATURE_VALID 0x4000
-#define EEPROM_ICW1_SIGNATURE_CLEAR 0x0000
-
-/* For checksumming, the sum of all words in the EEPROM should equal 0xBABA. */
-#define EEPROM_SUM 0xBABA
-
-/* EEPROM Map Sizes (Byte Counts) */
-#define PBA_SIZE 4
-
-/* EEPROM Map defines (WORD OFFSETS)*/
-
-/* EEPROM structure */
-struct ixgb_ee_map_type {
-	u8 mac_addr[ETH_ALEN];
-	__le16 compatibility;
-	__le16 reserved1[4];
-	__le32 pba_number;
-	__le16 init_ctrl_reg_1;
-	__le16 subsystem_id;
-	__le16 subvendor_id;
-	__le16 device_id;
-	__le16 vendor_id;
-	__le16 init_ctrl_reg_2;
-	__le16 oem_reserved[16];
-	__le16 swdpins_reg;
-	__le16 circuit_ctrl_reg;
-	u8 d3_power;
-	u8 d0_power;
-	__le16 reserved2[28];
-	__le16 checksum;
-};
-
-/* EEPROM Functions */
-u16 ixgb_read_eeprom(struct ixgb_hw *hw, u16 reg);
-
-bool ixgb_validate_eeprom_checksum(struct ixgb_hw *hw);
-
-void ixgb_update_eeprom_checksum(struct ixgb_hw *hw);
-
-void ixgb_write_eeprom(struct ixgb_hw *hw, u16 reg, u16 data);
-
-#endif				/* IXGB_EE_H */
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c b/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c
deleted file mode 100644
index efa980514944..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c
+++ /dev/null
@@ -1,642 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-/* ethtool support for ixgb */
-
-#include "ixgb.h"
-
-#include <linux/uaccess.h>
-
-#define IXGB_ALL_RAR_ENTRIES 16
-
-enum {NETDEV_STATS, IXGB_STATS};
-
-struct ixgb_stats {
-	char stat_string[ETH_GSTRING_LEN];
-	int type;
-	int sizeof_stat;
-	int stat_offset;
-};
-
-#define IXGB_STAT(m)		IXGB_STATS, \
-				sizeof_field(struct ixgb_adapter, m), \
-				offsetof(struct ixgb_adapter, m)
-#define IXGB_NETDEV_STAT(m)	NETDEV_STATS, \
-				sizeof_field(struct net_device, m), \
-				offsetof(struct net_device, m)
-
-static struct ixgb_stats ixgb_gstrings_stats[] = {
-	{"rx_packets", IXGB_NETDEV_STAT(stats.rx_packets)},
-	{"tx_packets", IXGB_NETDEV_STAT(stats.tx_packets)},
-	{"rx_bytes", IXGB_NETDEV_STAT(stats.rx_bytes)},
-	{"tx_bytes", IXGB_NETDEV_STAT(stats.tx_bytes)},
-	{"rx_errors", IXGB_NETDEV_STAT(stats.rx_errors)},
-	{"tx_errors", IXGB_NETDEV_STAT(stats.tx_errors)},
-	{"rx_dropped", IXGB_NETDEV_STAT(stats.rx_dropped)},
-	{"tx_dropped", IXGB_NETDEV_STAT(stats.tx_dropped)},
-	{"multicast", IXGB_NETDEV_STAT(stats.multicast)},
-	{"collisions", IXGB_NETDEV_STAT(stats.collisions)},
-
-/*	{ "rx_length_errors", IXGB_NETDEV_STAT(stats.rx_length_errors) },	*/
-	{"rx_over_errors", IXGB_NETDEV_STAT(stats.rx_over_errors)},
-	{"rx_crc_errors", IXGB_NETDEV_STAT(stats.rx_crc_errors)},
-	{"rx_frame_errors", IXGB_NETDEV_STAT(stats.rx_frame_errors)},
-	{"rx_no_buffer_count", IXGB_STAT(stats.rnbc)},
-	{"rx_fifo_errors", IXGB_NETDEV_STAT(stats.rx_fifo_errors)},
-	{"rx_missed_errors", IXGB_NETDEV_STAT(stats.rx_missed_errors)},
-	{"tx_aborted_errors", IXGB_NETDEV_STAT(stats.tx_aborted_errors)},
-	{"tx_carrier_errors", IXGB_NETDEV_STAT(stats.tx_carrier_errors)},
-	{"tx_fifo_errors", IXGB_NETDEV_STAT(stats.tx_fifo_errors)},
-	{"tx_heartbeat_errors", IXGB_NETDEV_STAT(stats.tx_heartbeat_errors)},
-	{"tx_window_errors", IXGB_NETDEV_STAT(stats.tx_window_errors)},
-	{"tx_deferred_ok", IXGB_STAT(stats.dc)},
-	{"tx_timeout_count", IXGB_STAT(tx_timeout_count) },
-	{"tx_restart_queue", IXGB_STAT(restart_queue) },
-	{"rx_long_length_errors", IXGB_STAT(stats.roc)},
-	{"rx_short_length_errors", IXGB_STAT(stats.ruc)},
-	{"tx_tcp_seg_good", IXGB_STAT(stats.tsctc)},
-	{"tx_tcp_seg_failed", IXGB_STAT(stats.tsctfc)},
-	{"rx_flow_control_xon", IXGB_STAT(stats.xonrxc)},
-	{"rx_flow_control_xoff", IXGB_STAT(stats.xoffrxc)},
-	{"tx_flow_control_xon", IXGB_STAT(stats.xontxc)},
-	{"tx_flow_control_xoff", IXGB_STAT(stats.xofftxc)},
-	{"rx_csum_offload_good", IXGB_STAT(hw_csum_rx_good)},
-	{"rx_csum_offload_errors", IXGB_STAT(hw_csum_rx_error)},
-	{"tx_csum_offload_good", IXGB_STAT(hw_csum_tx_good)},
-	{"tx_csum_offload_errors", IXGB_STAT(hw_csum_tx_error)}
-};
-
-#define IXGB_STATS_LEN	ARRAY_SIZE(ixgb_gstrings_stats)
-
-static int
-ixgb_get_link_ksettings(struct net_device *netdev,
-			struct ethtool_link_ksettings *cmd)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	ethtool_link_ksettings_zero_link_mode(cmd, supported);
-	ethtool_link_ksettings_add_link_mode(cmd, supported, 10000baseT_Full);
-	ethtool_link_ksettings_add_link_mode(cmd, supported, FIBRE);
-
-	ethtool_link_ksettings_zero_link_mode(cmd, advertising);
-	ethtool_link_ksettings_add_link_mode(cmd, advertising, 10000baseT_Full);
-	ethtool_link_ksettings_add_link_mode(cmd, advertising, FIBRE);
-
-	cmd->base.port = PORT_FIBRE;
-
-	if (netif_carrier_ok(adapter->netdev)) {
-		cmd->base.speed = SPEED_10000;
-		cmd->base.duplex = DUPLEX_FULL;
-	} else {
-		cmd->base.speed = SPEED_UNKNOWN;
-		cmd->base.duplex = DUPLEX_UNKNOWN;
-	}
-
-	cmd->base.autoneg = AUTONEG_DISABLE;
-	return 0;
-}
-
-void ixgb_set_speed_duplex(struct net_device *netdev)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	/* be optimistic about our link, since we were up before */
-	adapter->link_speed = 10000;
-	adapter->link_duplex = FULL_DUPLEX;
-	netif_carrier_on(netdev);
-	netif_wake_queue(netdev);
-}
-
-static int
-ixgb_set_link_ksettings(struct net_device *netdev,
-			const struct ethtool_link_ksettings *cmd)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	u32 speed = cmd->base.speed;
-
-	if (cmd->base.autoneg == AUTONEG_ENABLE ||
-	    (speed + cmd->base.duplex != SPEED_10000 + DUPLEX_FULL))
-		return -EINVAL;
-
-	if (netif_running(adapter->netdev)) {
-		ixgb_down(adapter, true);
-		ixgb_reset(adapter);
-		ixgb_up(adapter);
-		ixgb_set_speed_duplex(netdev);
-	} else
-		ixgb_reset(adapter);
-
-	return 0;
-}
-
-static void
-ixgb_get_pauseparam(struct net_device *netdev,
-			 struct ethtool_pauseparam *pause)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-
-	pause->autoneg = AUTONEG_DISABLE;
-
-	if (hw->fc.type == ixgb_fc_rx_pause)
-		pause->rx_pause = 1;
-	else if (hw->fc.type == ixgb_fc_tx_pause)
-		pause->tx_pause = 1;
-	else if (hw->fc.type == ixgb_fc_full) {
-		pause->rx_pause = 1;
-		pause->tx_pause = 1;
-	}
-}
-
-static int
-ixgb_set_pauseparam(struct net_device *netdev,
-			 struct ethtool_pauseparam *pause)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-
-	if (pause->autoneg == AUTONEG_ENABLE)
-		return -EINVAL;
-
-	if (pause->rx_pause && pause->tx_pause)
-		hw->fc.type = ixgb_fc_full;
-	else if (pause->rx_pause && !pause->tx_pause)
-		hw->fc.type = ixgb_fc_rx_pause;
-	else if (!pause->rx_pause && pause->tx_pause)
-		hw->fc.type = ixgb_fc_tx_pause;
-	else if (!pause->rx_pause && !pause->tx_pause)
-		hw->fc.type = ixgb_fc_none;
-
-	if (netif_running(adapter->netdev)) {
-		ixgb_down(adapter, true);
-		ixgb_up(adapter);
-		ixgb_set_speed_duplex(netdev);
-	} else
-		ixgb_reset(adapter);
-
-	return 0;
-}
-
-static u32
-ixgb_get_msglevel(struct net_device *netdev)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	return adapter->msg_enable;
-}
-
-static void
-ixgb_set_msglevel(struct net_device *netdev, u32 data)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	adapter->msg_enable = data;
-}
-#define IXGB_GET_STAT(_A_, _R_) _A_->stats._R_
-
-static int
-ixgb_get_regs_len(struct net_device *netdev)
-{
-#define IXGB_REG_DUMP_LEN  136*sizeof(u32)
-	return IXGB_REG_DUMP_LEN;
-}
-
-static void
-ixgb_get_regs(struct net_device *netdev,
-		   struct ethtool_regs *regs, void *p)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-	u32 *reg = p;
-	u32 *reg_start = reg;
-	u8 i;
-
-	/* the 1 (one) below indicates an attempt at versioning, if the
-	 * interface in ethtool or the driver changes, this 1 should be
-	 * incremented */
-	regs->version = (1<<24) | hw->revision_id << 16 | hw->device_id;
-
-	/* General Registers */
-	*reg++ = IXGB_READ_REG(hw, CTRL0);	/*   0 */
-	*reg++ = IXGB_READ_REG(hw, CTRL1);	/*   1 */
-	*reg++ = IXGB_READ_REG(hw, STATUS);	/*   2 */
-	*reg++ = IXGB_READ_REG(hw, EECD);	/*   3 */
-	*reg++ = IXGB_READ_REG(hw, MFS);	/*   4 */
-
-	/* Interrupt */
-	*reg++ = IXGB_READ_REG(hw, ICR);	/*   5 */
-	*reg++ = IXGB_READ_REG(hw, ICS);	/*   6 */
-	*reg++ = IXGB_READ_REG(hw, IMS);	/*   7 */
-	*reg++ = IXGB_READ_REG(hw, IMC);	/*   8 */
-
-	/* Receive */
-	*reg++ = IXGB_READ_REG(hw, RCTL);	/*   9 */
-	*reg++ = IXGB_READ_REG(hw, FCRTL);	/*  10 */
-	*reg++ = IXGB_READ_REG(hw, FCRTH);	/*  11 */
-	*reg++ = IXGB_READ_REG(hw, RDBAL);	/*  12 */
-	*reg++ = IXGB_READ_REG(hw, RDBAH);	/*  13 */
-	*reg++ = IXGB_READ_REG(hw, RDLEN);	/*  14 */
-	*reg++ = IXGB_READ_REG(hw, RDH);	/*  15 */
-	*reg++ = IXGB_READ_REG(hw, RDT);	/*  16 */
-	*reg++ = IXGB_READ_REG(hw, RDTR);	/*  17 */
-	*reg++ = IXGB_READ_REG(hw, RXDCTL);	/*  18 */
-	*reg++ = IXGB_READ_REG(hw, RAIDC);	/*  19 */
-	*reg++ = IXGB_READ_REG(hw, RXCSUM);	/*  20 */
-
-	/* there are 16 RAR entries in hardware, we only use 3 */
-	for (i = 0; i < IXGB_ALL_RAR_ENTRIES; i++) {
-		*reg++ = IXGB_READ_REG_ARRAY(hw, RAL, (i << 1)); /*21,...,51 */
-		*reg++ = IXGB_READ_REG_ARRAY(hw, RAH, (i << 1)); /*22,...,52 */
-	}
-
-	/* Transmit */
-	*reg++ = IXGB_READ_REG(hw, TCTL);	/*  53 */
-	*reg++ = IXGB_READ_REG(hw, TDBAL);	/*  54 */
-	*reg++ = IXGB_READ_REG(hw, TDBAH);	/*  55 */
-	*reg++ = IXGB_READ_REG(hw, TDLEN);	/*  56 */
-	*reg++ = IXGB_READ_REG(hw, TDH);	/*  57 */
-	*reg++ = IXGB_READ_REG(hw, TDT);	/*  58 */
-	*reg++ = IXGB_READ_REG(hw, TIDV);	/*  59 */
-	*reg++ = IXGB_READ_REG(hw, TXDCTL);	/*  60 */
-	*reg++ = IXGB_READ_REG(hw, TSPMT);	/*  61 */
-	*reg++ = IXGB_READ_REG(hw, PAP);	/*  62 */
-
-	/* Physical */
-	*reg++ = IXGB_READ_REG(hw, PCSC1);	/*  63 */
-	*reg++ = IXGB_READ_REG(hw, PCSC2);	/*  64 */
-	*reg++ = IXGB_READ_REG(hw, PCSS1);	/*  65 */
-	*reg++ = IXGB_READ_REG(hw, PCSS2);	/*  66 */
-	*reg++ = IXGB_READ_REG(hw, XPCSS);	/*  67 */
-	*reg++ = IXGB_READ_REG(hw, UCCR);	/*  68 */
-	*reg++ = IXGB_READ_REG(hw, XPCSTC);	/*  69 */
-	*reg++ = IXGB_READ_REG(hw, MACA);	/*  70 */
-	*reg++ = IXGB_READ_REG(hw, APAE);	/*  71 */
-	*reg++ = IXGB_READ_REG(hw, ARD);	/*  72 */
-	*reg++ = IXGB_READ_REG(hw, AIS);	/*  73 */
-	*reg++ = IXGB_READ_REG(hw, MSCA);	/*  74 */
-	*reg++ = IXGB_READ_REG(hw, MSRWD);	/*  75 */
-
-	/* Statistics */
-	*reg++ = IXGB_GET_STAT(adapter, tprl);	/*  76 */
-	*reg++ = IXGB_GET_STAT(adapter, tprh);	/*  77 */
-	*reg++ = IXGB_GET_STAT(adapter, gprcl);	/*  78 */
-	*reg++ = IXGB_GET_STAT(adapter, gprch);	/*  79 */
-	*reg++ = IXGB_GET_STAT(adapter, bprcl);	/*  80 */
-	*reg++ = IXGB_GET_STAT(adapter, bprch);	/*  81 */
-	*reg++ = IXGB_GET_STAT(adapter, mprcl);	/*  82 */
-	*reg++ = IXGB_GET_STAT(adapter, mprch);	/*  83 */
-	*reg++ = IXGB_GET_STAT(adapter, uprcl);	/*  84 */
-	*reg++ = IXGB_GET_STAT(adapter, uprch);	/*  85 */
-	*reg++ = IXGB_GET_STAT(adapter, vprcl);	/*  86 */
-	*reg++ = IXGB_GET_STAT(adapter, vprch);	/*  87 */
-	*reg++ = IXGB_GET_STAT(adapter, jprcl);	/*  88 */
-	*reg++ = IXGB_GET_STAT(adapter, jprch);	/*  89 */
-	*reg++ = IXGB_GET_STAT(adapter, gorcl);	/*  90 */
-	*reg++ = IXGB_GET_STAT(adapter, gorch);	/*  91 */
-	*reg++ = IXGB_GET_STAT(adapter, torl);	/*  92 */
-	*reg++ = IXGB_GET_STAT(adapter, torh);	/*  93 */
-	*reg++ = IXGB_GET_STAT(adapter, rnbc);	/*  94 */
-	*reg++ = IXGB_GET_STAT(adapter, ruc);	/*  95 */
-	*reg++ = IXGB_GET_STAT(adapter, roc);	/*  96 */
-	*reg++ = IXGB_GET_STAT(adapter, rlec);	/*  97 */
-	*reg++ = IXGB_GET_STAT(adapter, crcerrs);	/*  98 */
-	*reg++ = IXGB_GET_STAT(adapter, icbc);	/*  99 */
-	*reg++ = IXGB_GET_STAT(adapter, ecbc);	/* 100 */
-	*reg++ = IXGB_GET_STAT(adapter, mpc);	/* 101 */
-	*reg++ = IXGB_GET_STAT(adapter, tptl);	/* 102 */
-	*reg++ = IXGB_GET_STAT(adapter, tpth);	/* 103 */
-	*reg++ = IXGB_GET_STAT(adapter, gptcl);	/* 104 */
-	*reg++ = IXGB_GET_STAT(adapter, gptch);	/* 105 */
-	*reg++ = IXGB_GET_STAT(adapter, bptcl);	/* 106 */
-	*reg++ = IXGB_GET_STAT(adapter, bptch);	/* 107 */
-	*reg++ = IXGB_GET_STAT(adapter, mptcl);	/* 108 */
-	*reg++ = IXGB_GET_STAT(adapter, mptch);	/* 109 */
-	*reg++ = IXGB_GET_STAT(adapter, uptcl);	/* 110 */
-	*reg++ = IXGB_GET_STAT(adapter, uptch);	/* 111 */
-	*reg++ = IXGB_GET_STAT(adapter, vptcl);	/* 112 */
-	*reg++ = IXGB_GET_STAT(adapter, vptch);	/* 113 */
-	*reg++ = IXGB_GET_STAT(adapter, jptcl);	/* 114 */
-	*reg++ = IXGB_GET_STAT(adapter, jptch);	/* 115 */
-	*reg++ = IXGB_GET_STAT(adapter, gotcl);	/* 116 */
-	*reg++ = IXGB_GET_STAT(adapter, gotch);	/* 117 */
-	*reg++ = IXGB_GET_STAT(adapter, totl);	/* 118 */
-	*reg++ = IXGB_GET_STAT(adapter, toth);	/* 119 */
-	*reg++ = IXGB_GET_STAT(adapter, dc);	/* 120 */
-	*reg++ = IXGB_GET_STAT(adapter, plt64c);	/* 121 */
-	*reg++ = IXGB_GET_STAT(adapter, tsctc);	/* 122 */
-	*reg++ = IXGB_GET_STAT(adapter, tsctfc);	/* 123 */
-	*reg++ = IXGB_GET_STAT(adapter, ibic);	/* 124 */
-	*reg++ = IXGB_GET_STAT(adapter, rfc);	/* 125 */
-	*reg++ = IXGB_GET_STAT(adapter, lfc);	/* 126 */
-	*reg++ = IXGB_GET_STAT(adapter, pfrc);	/* 127 */
-	*reg++ = IXGB_GET_STAT(adapter, pftc);	/* 128 */
-	*reg++ = IXGB_GET_STAT(adapter, mcfrc);	/* 129 */
-	*reg++ = IXGB_GET_STAT(adapter, mcftc);	/* 130 */
-	*reg++ = IXGB_GET_STAT(adapter, xonrxc);	/* 131 */
-	*reg++ = IXGB_GET_STAT(adapter, xontxc);	/* 132 */
-	*reg++ = IXGB_GET_STAT(adapter, xoffrxc);	/* 133 */
-	*reg++ = IXGB_GET_STAT(adapter, xofftxc);	/* 134 */
-	*reg++ = IXGB_GET_STAT(adapter, rjc);	/* 135 */
-
-	regs->len = (reg - reg_start) * sizeof(u32);
-}
-
-static int
-ixgb_get_eeprom_len(struct net_device *netdev)
-{
-	/* return size in bytes */
-	return IXGB_EEPROM_SIZE << 1;
-}
-
-static int
-ixgb_get_eeprom(struct net_device *netdev,
-		  struct ethtool_eeprom *eeprom, u8 *bytes)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-	__le16 *eeprom_buff;
-	int i, max_len, first_word, last_word;
-	int ret_val = 0;
-
-	if (eeprom->len == 0) {
-		ret_val = -EINVAL;
-		goto geeprom_error;
-	}
-
-	eeprom->magic = hw->vendor_id | (hw->device_id << 16);
-
-	max_len = ixgb_get_eeprom_len(netdev);
-
-	if (eeprom->offset > eeprom->offset + eeprom->len) {
-		ret_val = -EINVAL;
-		goto geeprom_error;
-	}
-
-	if ((eeprom->offset + eeprom->len) > max_len)
-		eeprom->len = (max_len - eeprom->offset);
-
-	first_word = eeprom->offset >> 1;
-	last_word = (eeprom->offset + eeprom->len - 1) >> 1;
-
-	eeprom_buff = kmalloc_array(last_word - first_word + 1,
-				    sizeof(__le16),
-				    GFP_KERNEL);
-	if (!eeprom_buff)
-		return -ENOMEM;
-
-	/* note the eeprom was good because the driver loaded */
-	for (i = 0; i <= (last_word - first_word); i++)
-		eeprom_buff[i] = ixgb_get_eeprom_word(hw, (first_word + i));
-
-	memcpy(bytes, (u8 *)eeprom_buff + (eeprom->offset & 1), eeprom->len);
-	kfree(eeprom_buff);
-
-geeprom_error:
-	return ret_val;
-}
-
-static int
-ixgb_set_eeprom(struct net_device *netdev,
-		  struct ethtool_eeprom *eeprom, u8 *bytes)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-	u16 *eeprom_buff;
-	void *ptr;
-	int max_len, first_word, last_word;
-	u16 i;
-
-	if (eeprom->len == 0)
-		return -EINVAL;
-
-	if (eeprom->magic != (hw->vendor_id | (hw->device_id << 16)))
-		return -EFAULT;
-
-	max_len = ixgb_get_eeprom_len(netdev);
-
-	if (eeprom->offset > eeprom->offset + eeprom->len)
-		return -EINVAL;
-
-	if ((eeprom->offset + eeprom->len) > max_len)
-		eeprom->len = (max_len - eeprom->offset);
-
-	first_word = eeprom->offset >> 1;
-	last_word = (eeprom->offset + eeprom->len - 1) >> 1;
-	eeprom_buff = kmalloc(max_len, GFP_KERNEL);
-	if (!eeprom_buff)
-		return -ENOMEM;
-
-	ptr = (void *)eeprom_buff;
-
-	if (eeprom->offset & 1) {
-		/* need read/modify/write of first changed EEPROM word */
-		/* only the second byte of the word is being modified */
-		eeprom_buff[0] = ixgb_read_eeprom(hw, first_word);
-		ptr++;
-	}
-	if ((eeprom->offset + eeprom->len) & 1) {
-		/* need read/modify/write of last changed EEPROM word */
-		/* only the first byte of the word is being modified */
-		eeprom_buff[last_word - first_word]
-			= ixgb_read_eeprom(hw, last_word);
-	}
-
-	memcpy(ptr, bytes, eeprom->len);
-	for (i = 0; i <= (last_word - first_word); i++)
-		ixgb_write_eeprom(hw, first_word + i, eeprom_buff[i]);
-
-	/* Update the checksum over the first part of the EEPROM if needed */
-	if (first_word <= EEPROM_CHECKSUM_REG)
-		ixgb_update_eeprom_checksum(hw);
-
-	kfree(eeprom_buff);
-	return 0;
-}
-
-static void
-ixgb_get_drvinfo(struct net_device *netdev,
-		   struct ethtool_drvinfo *drvinfo)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	strscpy(drvinfo->driver,  ixgb_driver_name,
-		sizeof(drvinfo->driver));
-	strscpy(drvinfo->bus_info, pci_name(adapter->pdev),
-		sizeof(drvinfo->bus_info));
-}
-
-static void
-ixgb_get_ringparam(struct net_device *netdev,
-		   struct ethtool_ringparam *ring,
-		   struct kernel_ethtool_ringparam *kernel_ring,
-		   struct netlink_ext_ack *extack)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_desc_ring *txdr = &adapter->tx_ring;
-	struct ixgb_desc_ring *rxdr = &adapter->rx_ring;
-
-	ring->rx_max_pending = MAX_RXD;
-	ring->tx_max_pending = MAX_TXD;
-	ring->rx_pending = rxdr->count;
-	ring->tx_pending = txdr->count;
-}
-
-static int
-ixgb_set_ringparam(struct net_device *netdev,
-		   struct ethtool_ringparam *ring,
-		   struct kernel_ethtool_ringparam *kernel_ring,
-		   struct netlink_ext_ack *extack)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_desc_ring *txdr = &adapter->tx_ring;
-	struct ixgb_desc_ring *rxdr = &adapter->rx_ring;
-	struct ixgb_desc_ring tx_old, tx_new, rx_old, rx_new;
-	int err;
-
-	tx_old = adapter->tx_ring;
-	rx_old = adapter->rx_ring;
-
-	if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))
-		return -EINVAL;
-
-	if (netif_running(adapter->netdev))
-		ixgb_down(adapter, true);
-
-	rxdr->count = max(ring->rx_pending,(u32)MIN_RXD);
-	rxdr->count = min(rxdr->count,(u32)MAX_RXD);
-	rxdr->count = ALIGN(rxdr->count, IXGB_REQ_RX_DESCRIPTOR_MULTIPLE);
-
-	txdr->count = max(ring->tx_pending,(u32)MIN_TXD);
-	txdr->count = min(txdr->count,(u32)MAX_TXD);
-	txdr->count = ALIGN(txdr->count, IXGB_REQ_TX_DESCRIPTOR_MULTIPLE);
-
-	if (netif_running(adapter->netdev)) {
-		/* Try to get new resources before deleting old */
-		if ((err = ixgb_setup_rx_resources(adapter)))
-			goto err_setup_rx;
-		if ((err = ixgb_setup_tx_resources(adapter)))
-			goto err_setup_tx;
-
-		/* save the new, restore the old in order to free it,
-		 * then restore the new back again */
-
-		rx_new = adapter->rx_ring;
-		tx_new = adapter->tx_ring;
-		adapter->rx_ring = rx_old;
-		adapter->tx_ring = tx_old;
-		ixgb_free_rx_resources(adapter);
-		ixgb_free_tx_resources(adapter);
-		adapter->rx_ring = rx_new;
-		adapter->tx_ring = tx_new;
-		if ((err = ixgb_up(adapter)))
-			return err;
-		ixgb_set_speed_duplex(netdev);
-	}
-
-	return 0;
-err_setup_tx:
-	ixgb_free_rx_resources(adapter);
-err_setup_rx:
-	adapter->rx_ring = rx_old;
-	adapter->tx_ring = tx_old;
-	ixgb_up(adapter);
-	return err;
-}
-
-static int
-ixgb_set_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	switch (state) {
-	case ETHTOOL_ID_ACTIVE:
-		return 2;
-
-	case ETHTOOL_ID_ON:
-		ixgb_led_on(&adapter->hw);
-		break;
-
-	case ETHTOOL_ID_OFF:
-	case ETHTOOL_ID_INACTIVE:
-		ixgb_led_off(&adapter->hw);
-	}
-
-	return 0;
-}
-
-static int
-ixgb_get_sset_count(struct net_device *netdev, int sset)
-{
-	switch (sset) {
-	case ETH_SS_STATS:
-		return IXGB_STATS_LEN;
-	default:
-		return -EOPNOTSUPP;
-	}
-}
-
-static void
-ixgb_get_ethtool_stats(struct net_device *netdev,
-		struct ethtool_stats *stats, u64 *data)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	int i;
-	char *p = NULL;
-
-	ixgb_update_stats(adapter);
-	for (i = 0; i < IXGB_STATS_LEN; i++) {
-		switch (ixgb_gstrings_stats[i].type) {
-		case NETDEV_STATS:
-			p = (char *) netdev +
-					ixgb_gstrings_stats[i].stat_offset;
-			break;
-		case IXGB_STATS:
-			p = (char *) adapter +
-					ixgb_gstrings_stats[i].stat_offset;
-			break;
-		}
-
-		data[i] = (ixgb_gstrings_stats[i].sizeof_stat ==
-			sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
-	}
-}
-
-static void
-ixgb_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
-{
-	int i;
-
-	switch(stringset) {
-	case ETH_SS_STATS:
-		for (i = 0; i < IXGB_STATS_LEN; i++) {
-			memcpy(data + i * ETH_GSTRING_LEN,
-			ixgb_gstrings_stats[i].stat_string,
-			ETH_GSTRING_LEN);
-		}
-		break;
-	}
-}
-
-static const struct ethtool_ops ixgb_ethtool_ops = {
-	.get_drvinfo = ixgb_get_drvinfo,
-	.get_regs_len = ixgb_get_regs_len,
-	.get_regs = ixgb_get_regs,
-	.get_link = ethtool_op_get_link,
-	.get_eeprom_len = ixgb_get_eeprom_len,
-	.get_eeprom = ixgb_get_eeprom,
-	.set_eeprom = ixgb_set_eeprom,
-	.get_ringparam = ixgb_get_ringparam,
-	.set_ringparam = ixgb_set_ringparam,
-	.get_pauseparam	= ixgb_get_pauseparam,
-	.set_pauseparam	= ixgb_set_pauseparam,
-	.get_msglevel = ixgb_get_msglevel,
-	.set_msglevel = ixgb_set_msglevel,
-	.get_strings = ixgb_get_strings,
-	.set_phys_id = ixgb_set_phys_id,
-	.get_sset_count = ixgb_get_sset_count,
-	.get_ethtool_stats = ixgb_get_ethtool_stats,
-	.get_link_ksettings = ixgb_get_link_ksettings,
-	.set_link_ksettings = ixgb_set_link_ksettings,
-};
-
-void ixgb_set_ethtool_ops(struct net_device *netdev)
-{
-	netdev->ethtool_ops = &ixgb_ethtool_ops;
-}
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_hw.c b/drivers/net/ethernet/intel/ixgb/ixgb_hw.c
deleted file mode 100644
index 98bd3267b99b..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_hw.c
+++ /dev/null
@@ -1,1229 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-/* ixgb_hw.c
- * Shared functions for accessing and configuring the adapter
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/pci_ids.h>
-#include "ixgb_hw.h"
-#include "ixgb_ids.h"
-
-#include <linux/etherdevice.h>
-
-/*  Local function prototypes */
-
-static u32 ixgb_hash_mc_addr(struct ixgb_hw *hw, u8 * mc_addr);
-
-static void ixgb_mta_set(struct ixgb_hw *hw, u32 hash_value);
-
-static void ixgb_get_bus_info(struct ixgb_hw *hw);
-
-static bool ixgb_link_reset(struct ixgb_hw *hw);
-
-static void ixgb_optics_reset(struct ixgb_hw *hw);
-
-static void ixgb_optics_reset_bcm(struct ixgb_hw *hw);
-
-static ixgb_phy_type ixgb_identify_phy(struct ixgb_hw *hw);
-
-static void ixgb_clear_hw_cntrs(struct ixgb_hw *hw);
-
-static void ixgb_clear_vfta(struct ixgb_hw *hw);
-
-static void ixgb_init_rx_addrs(struct ixgb_hw *hw);
-
-static u16 ixgb_read_phy_reg(struct ixgb_hw *hw,
-				  u32 reg_address,
-				  u32 phy_address,
-				  u32 device_type);
-
-static bool ixgb_setup_fc(struct ixgb_hw *hw);
-
-static bool mac_addr_valid(u8 *mac_addr);
-
-static u32 ixgb_mac_reset(struct ixgb_hw *hw)
-{
-	u32 ctrl_reg;
-
-	ctrl_reg =  IXGB_CTRL0_RST |
-				IXGB_CTRL0_SDP3_DIR |   /* All pins are Output=1 */
-				IXGB_CTRL0_SDP2_DIR |
-				IXGB_CTRL0_SDP1_DIR |
-				IXGB_CTRL0_SDP0_DIR |
-				IXGB_CTRL0_SDP3	 |   /* Initial value 1101   */
-				IXGB_CTRL0_SDP2	 |
-				IXGB_CTRL0_SDP0;
-
-#ifdef HP_ZX1
-	/* Workaround for 82597EX reset errata */
-	IXGB_WRITE_REG_IO(hw, CTRL0, ctrl_reg);
-#else
-	IXGB_WRITE_REG(hw, CTRL0, ctrl_reg);
-#endif
-
-	/* Delay a few ms just to allow the reset to complete */
-	msleep(IXGB_DELAY_AFTER_RESET);
-	ctrl_reg = IXGB_READ_REG(hw, CTRL0);
-#ifdef DBG
-	/* Make sure the self-clearing global reset bit did self clear */
-	ASSERT(!(ctrl_reg & IXGB_CTRL0_RST));
-#endif
-
-	if (hw->subsystem_vendor_id == PCI_VENDOR_ID_SUN) {
-		ctrl_reg =  /* Enable interrupt from XFP and SerDes */
-			   IXGB_CTRL1_GPI0_EN |
-			   IXGB_CTRL1_SDP6_DIR |
-			   IXGB_CTRL1_SDP7_DIR |
-			   IXGB_CTRL1_SDP6 |
-			   IXGB_CTRL1_SDP7;
-		IXGB_WRITE_REG(hw, CTRL1, ctrl_reg);
-		ixgb_optics_reset_bcm(hw);
-	}
-
-	if (hw->phy_type == ixgb_phy_type_txn17401)
-		ixgb_optics_reset(hw);
-
-	return ctrl_reg;
-}
-
-/******************************************************************************
- * Reset the transmit and receive units; mask and clear all interrupts.
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-bool
-ixgb_adapter_stop(struct ixgb_hw *hw)
-{
-	u32 ctrl_reg;
-
-	ENTER();
-
-	/* If we are stopped or resetting exit gracefully and wait to be
-	 * started again before accessing the hardware.
-	 */
-	if (hw->adapter_stopped) {
-		pr_debug("Exiting because the adapter is already stopped!!!\n");
-		return false;
-	}
-
-	/* Set the Adapter Stopped flag so other driver functions stop
-	 * touching the Hardware.
-	 */
-	hw->adapter_stopped = true;
-
-	/* Clear interrupt mask to stop board from generating interrupts */
-	pr_debug("Masking off all interrupts\n");
-	IXGB_WRITE_REG(hw, IMC, 0xFFFFFFFF);
-
-	/* Disable the Transmit and Receive units.  Then delay to allow
-	 * any pending transactions to complete before we hit the MAC with
-	 * the global reset.
-	 */
-	IXGB_WRITE_REG(hw, RCTL, IXGB_READ_REG(hw, RCTL) & ~IXGB_RCTL_RXEN);
-	IXGB_WRITE_REG(hw, TCTL, IXGB_READ_REG(hw, TCTL) & ~IXGB_TCTL_TXEN);
-	IXGB_WRITE_FLUSH(hw);
-	msleep(IXGB_DELAY_BEFORE_RESET);
-
-	/* Issue a global reset to the MAC.  This will reset the chip's
-	 * transmit, receive, DMA, and link units.  It will not effect
-	 * the current PCI configuration.  The global reset bit is self-
-	 * clearing, and should clear within a microsecond.
-	 */
-	pr_debug("Issuing a global reset to MAC\n");
-
-	ctrl_reg = ixgb_mac_reset(hw);
-
-	/* Clear interrupt mask to stop board from generating interrupts */
-	pr_debug("Masking off all interrupts\n");
-	IXGB_WRITE_REG(hw, IMC, 0xffffffff);
-
-	/* Clear any pending interrupt events. */
-	IXGB_READ_REG(hw, ICR);
-
-	return ctrl_reg & IXGB_CTRL0_RST;
-}
-
-
-/******************************************************************************
- * Identifies the vendor of the optics module on the adapter.  The SR adapters
- * support two different types of XPAK optics, so it is necessary to determine
- * which optics are present before applying any optics-specific workarounds.
- *
- * hw - Struct containing variables accessed by shared code.
- *
- * Returns: the vendor of the XPAK optics module.
- *****************************************************************************/
-static ixgb_xpak_vendor
-ixgb_identify_xpak_vendor(struct ixgb_hw *hw)
-{
-	u32 i;
-	u16 vendor_name[5];
-	ixgb_xpak_vendor xpak_vendor;
-
-	ENTER();
-
-	/* Read the first few bytes of the vendor string from the XPAK NVR
-	 * registers.  These are standard XENPAK/XPAK registers, so all XPAK
-	 * devices should implement them. */
-	for (i = 0; i < 5; i++) {
-		vendor_name[i] = ixgb_read_phy_reg(hw,
-						   MDIO_PMA_PMD_XPAK_VENDOR_NAME
-						   + i, IXGB_PHY_ADDRESS,
-						   MDIO_MMD_PMAPMD);
-	}
-
-	/* Determine the actual vendor */
-	if (vendor_name[0] == 'I' &&
-	    vendor_name[1] == 'N' &&
-	    vendor_name[2] == 'T' &&
-	    vendor_name[3] == 'E' && vendor_name[4] == 'L') {
-		xpak_vendor = ixgb_xpak_vendor_intel;
-	} else {
-		xpak_vendor = ixgb_xpak_vendor_infineon;
-	}
-
-	return xpak_vendor;
-}
-
-/******************************************************************************
- * Determine the physical layer module on the adapter.
- *
- * hw - Struct containing variables accessed by shared code.  The device_id
- *      field must be (correctly) populated before calling this routine.
- *
- * Returns: the phy type of the adapter.
- *****************************************************************************/
-static ixgb_phy_type
-ixgb_identify_phy(struct ixgb_hw *hw)
-{
-	ixgb_phy_type phy_type;
-	ixgb_xpak_vendor xpak_vendor;
-
-	ENTER();
-
-	/* Infer the transceiver/phy type from the device id */
-	switch (hw->device_id) {
-	case IXGB_DEVICE_ID_82597EX:
-		pr_debug("Identified TXN17401 optics\n");
-		phy_type = ixgb_phy_type_txn17401;
-		break;
-
-	case IXGB_DEVICE_ID_82597EX_SR:
-		/* The SR adapters carry two different types of XPAK optics
-		 * modules; read the vendor identifier to determine the exact
-		 * type of optics. */
-		xpak_vendor = ixgb_identify_xpak_vendor(hw);
-		if (xpak_vendor == ixgb_xpak_vendor_intel) {
-			pr_debug("Identified TXN17201 optics\n");
-			phy_type = ixgb_phy_type_txn17201;
-		} else {
-			pr_debug("Identified G6005 optics\n");
-			phy_type = ixgb_phy_type_g6005;
-		}
-		break;
-	case IXGB_DEVICE_ID_82597EX_LR:
-		pr_debug("Identified G6104 optics\n");
-		phy_type = ixgb_phy_type_g6104;
-		break;
-	case IXGB_DEVICE_ID_82597EX_CX4:
-		pr_debug("Identified CX4\n");
-		xpak_vendor = ixgb_identify_xpak_vendor(hw);
-		if (xpak_vendor == ixgb_xpak_vendor_intel) {
-			pr_debug("Identified TXN17201 optics\n");
-			phy_type = ixgb_phy_type_txn17201;
-		} else {
-			pr_debug("Identified G6005 optics\n");
-			phy_type = ixgb_phy_type_g6005;
-		}
-		break;
-	default:
-		pr_debug("Unknown physical layer module\n");
-		phy_type = ixgb_phy_type_unknown;
-		break;
-	}
-
-	/* update phy type for sun specific board */
-	if (hw->subsystem_vendor_id == PCI_VENDOR_ID_SUN)
-		phy_type = ixgb_phy_type_bcm;
-
-	return phy_type;
-}
-
-/******************************************************************************
- * Performs basic configuration of the adapter.
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Resets the controller.
- * Reads and validates the EEPROM.
- * Initializes the receive address registers.
- * Initializes the multicast table.
- * Clears all on-chip counters.
- * Calls routine to setup flow control settings.
- * Leaves the transmit and receive units disabled and uninitialized.
- *
- * Returns:
- *      true if successful,
- *      false if unrecoverable problems were encountered.
- *****************************************************************************/
-bool
-ixgb_init_hw(struct ixgb_hw *hw)
-{
-	u32 i;
-	bool status;
-
-	ENTER();
-
-	/* Issue a global reset to the MAC.  This will reset the chip's
-	 * transmit, receive, DMA, and link units.  It will not effect
-	 * the current PCI configuration.  The global reset bit is self-
-	 * clearing, and should clear within a microsecond.
-	 */
-	pr_debug("Issuing a global reset to MAC\n");
-
-	ixgb_mac_reset(hw);
-
-	pr_debug("Issuing an EE reset to MAC\n");
-#ifdef HP_ZX1
-	/* Workaround for 82597EX reset errata */
-	IXGB_WRITE_REG_IO(hw, CTRL1, IXGB_CTRL1_EE_RST);
-#else
-	IXGB_WRITE_REG(hw, CTRL1, IXGB_CTRL1_EE_RST);
-#endif
-
-	/* Delay a few ms just to allow the reset to complete */
-	msleep(IXGB_DELAY_AFTER_EE_RESET);
-
-	if (!ixgb_get_eeprom_data(hw))
-		return false;
-
-	/* Use the device id to determine the type of phy/transceiver. */
-	hw->device_id = ixgb_get_ee_device_id(hw);
-	hw->phy_type = ixgb_identify_phy(hw);
-
-	/* Setup the receive addresses.
-	 * Receive Address Registers (RARs 0 - 15).
-	 */
-	ixgb_init_rx_addrs(hw);
-
-	/*
-	 * Check that a valid MAC address has been set.
-	 * If it is not valid, we fail hardware init.
-	 */
-	if (!mac_addr_valid(hw->curr_mac_addr)) {
-		pr_debug("MAC address invalid after ixgb_init_rx_addrs\n");
-		return(false);
-	}
-
-	/* tell the routines in this file they can access hardware again */
-	hw->adapter_stopped = false;
-
-	/* Fill in the bus_info structure */
-	ixgb_get_bus_info(hw);
-
-	/* Zero out the Multicast HASH table */
-	pr_debug("Zeroing the MTA\n");
-	for (i = 0; i < IXGB_MC_TBL_SIZE; i++)
-		IXGB_WRITE_REG_ARRAY(hw, MTA, i, 0);
-
-	/* Zero out the VLAN Filter Table Array */
-	ixgb_clear_vfta(hw);
-
-	/* Zero all of the hardware counters */
-	ixgb_clear_hw_cntrs(hw);
-
-	/* Call a subroutine to setup flow control. */
-	status = ixgb_setup_fc(hw);
-
-	/* 82597EX errata: Call check-for-link in case lane deskew is locked */
-	ixgb_check_for_link(hw);
-
-	return status;
-}
-
-/******************************************************************************
- * Initializes receive address filters.
- *
- * hw - Struct containing variables accessed by shared code
- *
- * Places the MAC address in receive address register 0 and clears the rest
- * of the receive address registers. Clears the multicast table. Assumes
- * the receiver is in reset when the routine is called.
- *****************************************************************************/
-static void
-ixgb_init_rx_addrs(struct ixgb_hw *hw)
-{
-	u32 i;
-
-	ENTER();
-
-	/*
-	 * If the current mac address is valid, assume it is a software override
-	 * to the permanent address.
-	 * Otherwise, use the permanent address from the eeprom.
-	 */
-	if (!mac_addr_valid(hw->curr_mac_addr)) {
-
-		/* Get the MAC address from the eeprom for later reference */
-		ixgb_get_ee_mac_addr(hw, hw->curr_mac_addr);
-
-		pr_debug("Keeping Permanent MAC Addr = %pM\n",
-			 hw->curr_mac_addr);
-	} else {
-
-		/* Setup the receive address. */
-		pr_debug("Overriding MAC Address in RAR[0]\n");
-		pr_debug("New MAC Addr = %pM\n", hw->curr_mac_addr);
-
-		ixgb_rar_set(hw, hw->curr_mac_addr, 0);
-	}
-
-	/* Zero out the other 15 receive addresses. */
-	pr_debug("Clearing RAR[1-15]\n");
-	for (i = 1; i < IXGB_RAR_ENTRIES; i++) {
-		/* Write high reg first to disable the AV bit first */
-		IXGB_WRITE_REG_ARRAY(hw, RA, ((i << 1) + 1), 0);
-		IXGB_WRITE_REG_ARRAY(hw, RA, (i << 1), 0);
-	}
-}
-
-/******************************************************************************
- * Updates the MAC's list of multicast addresses.
- *
- * hw - Struct containing variables accessed by shared code
- * mc_addr_list - the list of new multicast addresses
- * mc_addr_count - number of addresses
- * pad - number of bytes between addresses in the list
- *
- * The given list replaces any existing list. Clears the last 15 receive
- * address registers and the multicast table. Uses receive address registers
- * for the first 15 multicast addresses, and hashes the rest into the
- * multicast table.
- *****************************************************************************/
-void
-ixgb_mc_addr_list_update(struct ixgb_hw *hw,
-			  u8 *mc_addr_list,
-			  u32 mc_addr_count,
-			  u32 pad)
-{
-	u32 hash_value;
-	u32 i;
-	u32 rar_used_count = 1;		/* RAR[0] is used for our MAC address */
-	u8 *mca;
-
-	ENTER();
-
-	/* Set the new number of MC addresses that we are being requested to use. */
-	hw->num_mc_addrs = mc_addr_count;
-
-	/* Clear RAR[1-15] */
-	pr_debug("Clearing RAR[1-15]\n");
-	for (i = rar_used_count; i < IXGB_RAR_ENTRIES; i++) {
-		IXGB_WRITE_REG_ARRAY(hw, RA, (i << 1), 0);
-		IXGB_WRITE_REG_ARRAY(hw, RA, ((i << 1) + 1), 0);
-	}
-
-	/* Clear the MTA */
-	pr_debug("Clearing MTA\n");
-	for (i = 0; i < IXGB_MC_TBL_SIZE; i++)
-		IXGB_WRITE_REG_ARRAY(hw, MTA, i, 0);
-
-	/* Add the new addresses */
-	mca = mc_addr_list;
-	for (i = 0; i < mc_addr_count; i++) {
-		pr_debug("Adding the multicast addresses:\n");
-		pr_debug("MC Addr #%d = %pM\n", i, mca);
-
-		/* Place this multicast address in the RAR if there is room, *
-		 * else put it in the MTA
-		 */
-		if (rar_used_count < IXGB_RAR_ENTRIES) {
-			ixgb_rar_set(hw, mca, rar_used_count);
-			pr_debug("Added a multicast address to RAR[%d]\n", i);
-			rar_used_count++;
-		} else {
-			hash_value = ixgb_hash_mc_addr(hw, mca);
-
-			pr_debug("Hash value = 0x%03X\n", hash_value);
-
-			ixgb_mta_set(hw, hash_value);
-		}
-
-		mca += ETH_ALEN + pad;
-	}
-
-	pr_debug("MC Update Complete\n");
-}
-
-/******************************************************************************
- * Hashes an address to determine its location in the multicast table
- *
- * hw - Struct containing variables accessed by shared code
- * mc_addr - the multicast address to hash
- *
- * Returns:
- *      The hash value
- *****************************************************************************/
-static u32
-ixgb_hash_mc_addr(struct ixgb_hw *hw,
-		   u8 *mc_addr)
-{
-	u32 hash_value = 0;
-
-	ENTER();
-
-	/* The portion of the address that is used for the hash table is
-	 * determined by the mc_filter_type setting.
-	 */
-	switch (hw->mc_filter_type) {
-		/* [0] [1] [2] [3] [4] [5]
-		 * 01  AA  00  12  34  56
-		 * LSB                 MSB - According to H/W docs */
-	case 0:
-		/* [47:36] i.e. 0x563 for above example address */
-		hash_value =
-		    ((mc_addr[4] >> 4) | (((u16) mc_addr[5]) << 4));
-		break;
-	case 1:		/* [46:35] i.e. 0xAC6 for above example address */
-		hash_value =
-		    ((mc_addr[4] >> 3) | (((u16) mc_addr[5]) << 5));
-		break;
-	case 2:		/* [45:34] i.e. 0x5D8 for above example address */
-		hash_value =
-		    ((mc_addr[4] >> 2) | (((u16) mc_addr[5]) << 6));
-		break;
-	case 3:		/* [43:32] i.e. 0x634 for above example address */
-		hash_value = ((mc_addr[4]) | (((u16) mc_addr[5]) << 8));
-		break;
-	default:
-		/* Invalid mc_filter_type, what should we do? */
-		pr_debug("MC filter type param set incorrectly\n");
-		ASSERT(0);
-		break;
-	}
-
-	hash_value &= 0xFFF;
-	return hash_value;
-}
-
-/******************************************************************************
- * Sets the bit in the multicast table corresponding to the hash value.
- *
- * hw - Struct containing variables accessed by shared code
- * hash_value - Multicast address hash value
- *****************************************************************************/
-static void
-ixgb_mta_set(struct ixgb_hw *hw,
-		  u32 hash_value)
-{
-	u32 hash_bit, hash_reg;
-	u32 mta_reg;
-
-	/* The MTA is a register array of 128 32-bit registers.
-	 * It is treated like an array of 4096 bits.  We want to set
-	 * bit BitArray[hash_value]. So we figure out what register
-	 * the bit is in, read it, OR in the new bit, then write
-	 * back the new value.  The register is determined by the
-	 * upper 7 bits of the hash value and the bit within that
-	 * register are determined by the lower 5 bits of the value.
-	 */
-	hash_reg = (hash_value >> 5) & 0x7F;
-	hash_bit = hash_value & 0x1F;
-
-	mta_reg = IXGB_READ_REG_ARRAY(hw, MTA, hash_reg);
-
-	mta_reg |= (1 << hash_bit);
-
-	IXGB_WRITE_REG_ARRAY(hw, MTA, hash_reg, mta_reg);
-}
-
-/******************************************************************************
- * Puts an ethernet address into a receive address register.
- *
- * hw - Struct containing variables accessed by shared code
- * addr - Address to put into receive address register
- * index - Receive address register to write
- *****************************************************************************/
-void
-ixgb_rar_set(struct ixgb_hw *hw,
-		  const u8 *addr,
-		  u32 index)
-{
-	u32 rar_low, rar_high;
-
-	ENTER();
-
-	/* HW expects these in little endian so we reverse the byte order
-	 * from network order (big endian) to little endian
-	 */
-	rar_low = ((u32) addr[0] |
-		   ((u32)addr[1] << 8) |
-		   ((u32)addr[2] << 16) |
-		   ((u32)addr[3] << 24));
-
-	rar_high = ((u32) addr[4] |
-			((u32)addr[5] << 8) |
-			IXGB_RAH_AV);
-
-	IXGB_WRITE_REG_ARRAY(hw, RA, (index << 1), rar_low);
-	IXGB_WRITE_REG_ARRAY(hw, RA, ((index << 1) + 1), rar_high);
-}
-
-/******************************************************************************
- * Writes a value to the specified offset in the VLAN filter table.
- *
- * hw - Struct containing variables accessed by shared code
- * offset - Offset in VLAN filter table to write
- * value - Value to write into VLAN filter table
- *****************************************************************************/
-void
-ixgb_write_vfta(struct ixgb_hw *hw,
-		 u32 offset,
-		 u32 value)
-{
-	IXGB_WRITE_REG_ARRAY(hw, VFTA, offset, value);
-}
-
-/******************************************************************************
- * Clears the VLAN filter table
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_clear_vfta(struct ixgb_hw *hw)
-{
-	u32 offset;
-
-	for (offset = 0; offset < IXGB_VLAN_FILTER_TBL_SIZE; offset++)
-		IXGB_WRITE_REG_ARRAY(hw, VFTA, offset, 0);
-}
-
-/******************************************************************************
- * Configures the flow control settings based on SW configuration.
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-
-static bool
-ixgb_setup_fc(struct ixgb_hw *hw)
-{
-	u32 ctrl_reg;
-	u32 pap_reg = 0;   /* by default, assume no pause time */
-	bool status = true;
-
-	ENTER();
-
-	/* Get the current control reg 0 settings */
-	ctrl_reg = IXGB_READ_REG(hw, CTRL0);
-
-	/* Clear the Receive Pause Enable and Transmit Pause Enable bits */
-	ctrl_reg &= ~(IXGB_CTRL0_RPE | IXGB_CTRL0_TPE);
-
-	/* The possible values of the "flow_control" parameter are:
-	 *      0:  Flow control is completely disabled
-	 *      1:  Rx flow control is enabled (we can receive pause frames
-	 *          but not send pause frames).
-	 *      2:  Tx flow control is enabled (we can send pause frames
-	 *          but we do not support receiving pause frames).
-	 *      3:  Both Rx and TX flow control (symmetric) are enabled.
-	 *  other:  Invalid.
-	 */
-	switch (hw->fc.type) {
-	case ixgb_fc_none:	/* 0 */
-		/* Set CMDC bit to disable Rx Flow control */
-		ctrl_reg |= (IXGB_CTRL0_CMDC);
-		break;
-	case ixgb_fc_rx_pause:	/* 1 */
-		/* RX Flow control is enabled, and TX Flow control is
-		 * disabled.
-		 */
-		ctrl_reg |= (IXGB_CTRL0_RPE);
-		break;
-	case ixgb_fc_tx_pause:	/* 2 */
-		/* TX Flow control is enabled, and RX Flow control is
-		 * disabled, by a software over-ride.
-		 */
-		ctrl_reg |= (IXGB_CTRL0_TPE);
-		pap_reg = hw->fc.pause_time;
-		break;
-	case ixgb_fc_full:	/* 3 */
-		/* Flow control (both RX and TX) is enabled by a software
-		 * over-ride.
-		 */
-		ctrl_reg |= (IXGB_CTRL0_RPE | IXGB_CTRL0_TPE);
-		pap_reg = hw->fc.pause_time;
-		break;
-	default:
-		/* We should never get here.  The value should be 0-3. */
-		pr_debug("Flow control param set incorrectly\n");
-		ASSERT(0);
-		break;
-	}
-
-	/* Write the new settings */
-	IXGB_WRITE_REG(hw, CTRL0, ctrl_reg);
-
-	if (pap_reg != 0)
-		IXGB_WRITE_REG(hw, PAP, pap_reg);
-
-	/* Set the flow control receive threshold registers.  Normally,
-	 * these registers will be set to a default threshold that may be
-	 * adjusted later by the driver's runtime code.  However, if the
-	 * ability to transmit pause frames in not enabled, then these
-	 * registers will be set to 0.
-	 */
-	if (!(hw->fc.type & ixgb_fc_tx_pause)) {
-		IXGB_WRITE_REG(hw, FCRTL, 0);
-		IXGB_WRITE_REG(hw, FCRTH, 0);
-	} else {
-	   /* We need to set up the Receive Threshold high and low water
-	    * marks as well as (optionally) enabling the transmission of XON
-	    * frames. */
-		if (hw->fc.send_xon) {
-			IXGB_WRITE_REG(hw, FCRTL,
-				(hw->fc.low_water | IXGB_FCRTL_XONE));
-		} else {
-			IXGB_WRITE_REG(hw, FCRTL, hw->fc.low_water);
-		}
-		IXGB_WRITE_REG(hw, FCRTH, hw->fc.high_water);
-	}
-	return status;
-}
-
-/******************************************************************************
- * Reads a word from a device over the Management Data Interface (MDI) bus.
- * This interface is used to manage Physical layer devices.
- *
- * hw          - Struct containing variables accessed by hw code
- * reg_address - Offset of device register being read.
- * phy_address - Address of device on MDI.
- *
- * Returns:  Data word (16 bits) from MDI device.
- *
- * The 82597EX has support for several MDI access methods.  This routine
- * uses the new protocol MDI Single Command and Address Operation.
- * This requires that first an address cycle command is sent, followed by a
- * read command.
- *****************************************************************************/
-static u16
-ixgb_read_phy_reg(struct ixgb_hw *hw,
-		u32 reg_address,
-		u32 phy_address,
-		u32 device_type)
-{
-	u32 i;
-	u32 data;
-	u32 command = 0;
-
-	ASSERT(reg_address <= IXGB_MAX_PHY_REG_ADDRESS);
-	ASSERT(phy_address <= IXGB_MAX_PHY_ADDRESS);
-	ASSERT(device_type <= IXGB_MAX_PHY_DEV_TYPE);
-
-	/* Setup and write the address cycle command */
-	command = ((reg_address << IXGB_MSCA_NP_ADDR_SHIFT) |
-		   (device_type << IXGB_MSCA_DEV_TYPE_SHIFT) |
-		   (phy_address << IXGB_MSCA_PHY_ADDR_SHIFT) |
-		   (IXGB_MSCA_ADDR_CYCLE | IXGB_MSCA_MDI_COMMAND));
-
-	IXGB_WRITE_REG(hw, MSCA, command);
-
-    /**************************************************************
-    ** Check every 10 usec to see if the address cycle completed
-    ** The COMMAND bit will clear when the operation is complete.
-    ** This may take as long as 64 usecs (we'll wait 100 usecs max)
-    ** from the CPU Write to the Ready bit assertion.
-    **************************************************************/
-
-	for (i = 0; i < 10; i++)
-	{
-		udelay(10);
-
-		command = IXGB_READ_REG(hw, MSCA);
-
-		if ((command & IXGB_MSCA_MDI_COMMAND) == 0)
-			break;
-	}
-
-	ASSERT((command & IXGB_MSCA_MDI_COMMAND) == 0);
-
-	/* Address cycle complete, setup and write the read command */
-	command = ((reg_address << IXGB_MSCA_NP_ADDR_SHIFT) |
-		   (device_type << IXGB_MSCA_DEV_TYPE_SHIFT) |
-		   (phy_address << IXGB_MSCA_PHY_ADDR_SHIFT) |
-		   (IXGB_MSCA_READ | IXGB_MSCA_MDI_COMMAND));
-
-	IXGB_WRITE_REG(hw, MSCA, command);
-
-    /**************************************************************
-    ** Check every 10 usec to see if the read command completed
-    ** The COMMAND bit will clear when the operation is complete.
-    ** The read may take as long as 64 usecs (we'll wait 100 usecs max)
-    ** from the CPU Write to the Ready bit assertion.
-    **************************************************************/
-
-	for (i = 0; i < 10; i++)
-	{
-		udelay(10);
-
-		command = IXGB_READ_REG(hw, MSCA);
-
-		if ((command & IXGB_MSCA_MDI_COMMAND) == 0)
-			break;
-	}
-
-	ASSERT((command & IXGB_MSCA_MDI_COMMAND) == 0);
-
-	/* Operation is complete, get the data from the MDIO Read/Write Data
-	 * register and return.
-	 */
-	data = IXGB_READ_REG(hw, MSRWD);
-	data >>= IXGB_MSRWD_READ_DATA_SHIFT;
-	return((u16) data);
-}
-
-/******************************************************************************
- * Writes a word to a device over the Management Data Interface (MDI) bus.
- * This interface is used to manage Physical layer devices.
- *
- * hw          - Struct containing variables accessed by hw code
- * reg_address - Offset of device register being read.
- * phy_address - Address of device on MDI.
- * device_type - Also known as the Device ID or DID.
- * data        - 16-bit value to be written
- *
- * Returns:  void.
- *
- * The 82597EX has support for several MDI access methods.  This routine
- * uses the new protocol MDI Single Command and Address Operation.
- * This requires that first an address cycle command is sent, followed by a
- * write command.
- *****************************************************************************/
-static void
-ixgb_write_phy_reg(struct ixgb_hw *hw,
-			u32 reg_address,
-			u32 phy_address,
-			u32 device_type,
-			u16 data)
-{
-	u32 i;
-	u32 command = 0;
-
-	ASSERT(reg_address <= IXGB_MAX_PHY_REG_ADDRESS);
-	ASSERT(phy_address <= IXGB_MAX_PHY_ADDRESS);
-	ASSERT(device_type <= IXGB_MAX_PHY_DEV_TYPE);
-
-	/* Put the data in the MDIO Read/Write Data register */
-	IXGB_WRITE_REG(hw, MSRWD, (u32)data);
-
-	/* Setup and write the address cycle command */
-	command = ((reg_address << IXGB_MSCA_NP_ADDR_SHIFT)  |
-			   (device_type << IXGB_MSCA_DEV_TYPE_SHIFT) |
-			   (phy_address << IXGB_MSCA_PHY_ADDR_SHIFT) |
-			   (IXGB_MSCA_ADDR_CYCLE | IXGB_MSCA_MDI_COMMAND));
-
-	IXGB_WRITE_REG(hw, MSCA, command);
-
-	/**************************************************************
-	** Check every 10 usec to see if the address cycle completed
-	** The COMMAND bit will clear when the operation is complete.
-	** This may take as long as 64 usecs (we'll wait 100 usecs max)
-	** from the CPU Write to the Ready bit assertion.
-	**************************************************************/
-
-	for (i = 0; i < 10; i++)
-	{
-		udelay(10);
-
-		command = IXGB_READ_REG(hw, MSCA);
-
-		if ((command & IXGB_MSCA_MDI_COMMAND) == 0)
-			break;
-	}
-
-	ASSERT((command & IXGB_MSCA_MDI_COMMAND) == 0);
-
-	/* Address cycle complete, setup and write the write command */
-	command = ((reg_address << IXGB_MSCA_NP_ADDR_SHIFT)  |
-			   (device_type << IXGB_MSCA_DEV_TYPE_SHIFT) |
-			   (phy_address << IXGB_MSCA_PHY_ADDR_SHIFT) |
-			   (IXGB_MSCA_WRITE | IXGB_MSCA_MDI_COMMAND));
-
-	IXGB_WRITE_REG(hw, MSCA, command);
-
-	/**************************************************************
-	** Check every 10 usec to see if the read command completed
-	** The COMMAND bit will clear when the operation is complete.
-	** The write may take as long as 64 usecs (we'll wait 100 usecs max)
-	** from the CPU Write to the Ready bit assertion.
-	**************************************************************/
-
-	for (i = 0; i < 10; i++)
-	{
-		udelay(10);
-
-		command = IXGB_READ_REG(hw, MSCA);
-
-		if ((command & IXGB_MSCA_MDI_COMMAND) == 0)
-			break;
-	}
-
-	ASSERT((command & IXGB_MSCA_MDI_COMMAND) == 0);
-
-	/* Operation is complete, return. */
-}
-
-/******************************************************************************
- * Checks to see if the link status of the hardware has changed.
- *
- * hw - Struct containing variables accessed by hw code
- *
- * Called by any function that needs to check the link status of the adapter.
- *****************************************************************************/
-void
-ixgb_check_for_link(struct ixgb_hw *hw)
-{
-	u32 status_reg;
-	u32 xpcss_reg;
-
-	ENTER();
-
-	xpcss_reg = IXGB_READ_REG(hw, XPCSS);
-	status_reg = IXGB_READ_REG(hw, STATUS);
-
-	if ((xpcss_reg & IXGB_XPCSS_ALIGN_STATUS) &&
-	    (status_reg & IXGB_STATUS_LU)) {
-		hw->link_up = true;
-	} else if (!(xpcss_reg & IXGB_XPCSS_ALIGN_STATUS) &&
-		   (status_reg & IXGB_STATUS_LU)) {
-		pr_debug("XPCSS Not Aligned while Status:LU is set\n");
-		hw->link_up = ixgb_link_reset(hw);
-	} else {
-		/*
-		 * 82597EX errata.  Since the lane deskew problem may prevent
-		 * link, reset the link before reporting link down.
-		 */
-		hw->link_up = ixgb_link_reset(hw);
-	}
-	/*  Anything else for 10 Gig?? */
-}
-
-/******************************************************************************
- * Check for a bad link condition that may have occurred.
- * The indication is that the RFC / LFC registers may be incrementing
- * continually.  A full adapter reset is required to recover.
- *
- * hw - Struct containing variables accessed by hw code
- *
- * Called by any function that needs to check the link status of the adapter.
- *****************************************************************************/
-bool ixgb_check_for_bad_link(struct ixgb_hw *hw)
-{
-	u32 newLFC, newRFC;
-	bool bad_link_returncode = false;
-
-	if (hw->phy_type == ixgb_phy_type_txn17401) {
-		newLFC = IXGB_READ_REG(hw, LFC);
-		newRFC = IXGB_READ_REG(hw, RFC);
-		if ((hw->lastLFC + 250 < newLFC)
-		    || (hw->lastRFC + 250 < newRFC)) {
-			pr_debug("BAD LINK! too many LFC/RFC since last check\n");
-			bad_link_returncode = true;
-		}
-		hw->lastLFC = newLFC;
-		hw->lastRFC = newRFC;
-	}
-
-	return bad_link_returncode;
-}
-
-/******************************************************************************
- * Clears all hardware statistics counters.
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_clear_hw_cntrs(struct ixgb_hw *hw)
-{
-	ENTER();
-
-	/* if we are stopped or resetting exit gracefully */
-	if (hw->adapter_stopped) {
-		pr_debug("Exiting because the adapter is stopped!!!\n");
-		return;
-	}
-
-	IXGB_READ_REG(hw, TPRL);
-	IXGB_READ_REG(hw, TPRH);
-	IXGB_READ_REG(hw, GPRCL);
-	IXGB_READ_REG(hw, GPRCH);
-	IXGB_READ_REG(hw, BPRCL);
-	IXGB_READ_REG(hw, BPRCH);
-	IXGB_READ_REG(hw, MPRCL);
-	IXGB_READ_REG(hw, MPRCH);
-	IXGB_READ_REG(hw, UPRCL);
-	IXGB_READ_REG(hw, UPRCH);
-	IXGB_READ_REG(hw, VPRCL);
-	IXGB_READ_REG(hw, VPRCH);
-	IXGB_READ_REG(hw, JPRCL);
-	IXGB_READ_REG(hw, JPRCH);
-	IXGB_READ_REG(hw, GORCL);
-	IXGB_READ_REG(hw, GORCH);
-	IXGB_READ_REG(hw, TORL);
-	IXGB_READ_REG(hw, TORH);
-	IXGB_READ_REG(hw, RNBC);
-	IXGB_READ_REG(hw, RUC);
-	IXGB_READ_REG(hw, ROC);
-	IXGB_READ_REG(hw, RLEC);
-	IXGB_READ_REG(hw, CRCERRS);
-	IXGB_READ_REG(hw, ICBC);
-	IXGB_READ_REG(hw, ECBC);
-	IXGB_READ_REG(hw, MPC);
-	IXGB_READ_REG(hw, TPTL);
-	IXGB_READ_REG(hw, TPTH);
-	IXGB_READ_REG(hw, GPTCL);
-	IXGB_READ_REG(hw, GPTCH);
-	IXGB_READ_REG(hw, BPTCL);
-	IXGB_READ_REG(hw, BPTCH);
-	IXGB_READ_REG(hw, MPTCL);
-	IXGB_READ_REG(hw, MPTCH);
-	IXGB_READ_REG(hw, UPTCL);
-	IXGB_READ_REG(hw, UPTCH);
-	IXGB_READ_REG(hw, VPTCL);
-	IXGB_READ_REG(hw, VPTCH);
-	IXGB_READ_REG(hw, JPTCL);
-	IXGB_READ_REG(hw, JPTCH);
-	IXGB_READ_REG(hw, GOTCL);
-	IXGB_READ_REG(hw, GOTCH);
-	IXGB_READ_REG(hw, TOTL);
-	IXGB_READ_REG(hw, TOTH);
-	IXGB_READ_REG(hw, DC);
-	IXGB_READ_REG(hw, PLT64C);
-	IXGB_READ_REG(hw, TSCTC);
-	IXGB_READ_REG(hw, TSCTFC);
-	IXGB_READ_REG(hw, IBIC);
-	IXGB_READ_REG(hw, RFC);
-	IXGB_READ_REG(hw, LFC);
-	IXGB_READ_REG(hw, PFRC);
-	IXGB_READ_REG(hw, PFTC);
-	IXGB_READ_REG(hw, MCFRC);
-	IXGB_READ_REG(hw, MCFTC);
-	IXGB_READ_REG(hw, XONRXC);
-	IXGB_READ_REG(hw, XONTXC);
-	IXGB_READ_REG(hw, XOFFRXC);
-	IXGB_READ_REG(hw, XOFFTXC);
-	IXGB_READ_REG(hw, RJC);
-}
-
-/******************************************************************************
- * Turns on the software controllable LED
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-void
-ixgb_led_on(struct ixgb_hw *hw)
-{
-	u32 ctrl0_reg = IXGB_READ_REG(hw, CTRL0);
-
-	/* To turn on the LED, clear software-definable pin 0 (SDP0). */
-	ctrl0_reg &= ~IXGB_CTRL0_SDP0;
-	IXGB_WRITE_REG(hw, CTRL0, ctrl0_reg);
-}
-
-/******************************************************************************
- * Turns off the software controllable LED
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-void
-ixgb_led_off(struct ixgb_hw *hw)
-{
-	u32 ctrl0_reg = IXGB_READ_REG(hw, CTRL0);
-
-	/* To turn off the LED, set software-definable pin 0 (SDP0). */
-	ctrl0_reg |= IXGB_CTRL0_SDP0;
-	IXGB_WRITE_REG(hw, CTRL0, ctrl0_reg);
-}
-
-/******************************************************************************
- * Gets the current PCI bus type, speed, and width of the hardware
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_get_bus_info(struct ixgb_hw *hw)
-{
-	u32 status_reg;
-
-	status_reg = IXGB_READ_REG(hw, STATUS);
-
-	hw->bus.type = (status_reg & IXGB_STATUS_PCIX_MODE) ?
-		ixgb_bus_type_pcix : ixgb_bus_type_pci;
-
-	if (hw->bus.type == ixgb_bus_type_pci) {
-		hw->bus.speed = (status_reg & IXGB_STATUS_PCI_SPD) ?
-			ixgb_bus_speed_66 : ixgb_bus_speed_33;
-	} else {
-		switch (status_reg & IXGB_STATUS_PCIX_SPD_MASK) {
-		case IXGB_STATUS_PCIX_SPD_66:
-			hw->bus.speed = ixgb_bus_speed_66;
-			break;
-		case IXGB_STATUS_PCIX_SPD_100:
-			hw->bus.speed = ixgb_bus_speed_100;
-			break;
-		case IXGB_STATUS_PCIX_SPD_133:
-			hw->bus.speed = ixgb_bus_speed_133;
-			break;
-		default:
-			hw->bus.speed = ixgb_bus_speed_reserved;
-			break;
-		}
-	}
-
-	hw->bus.width = (status_reg & IXGB_STATUS_BUS64) ?
-		ixgb_bus_width_64 : ixgb_bus_width_32;
-}
-
-/******************************************************************************
- * Tests a MAC address to ensure it is a valid Individual Address
- *
- * mac_addr - pointer to MAC address.
- *
- *****************************************************************************/
-static bool
-mac_addr_valid(u8 *mac_addr)
-{
-	bool is_valid = true;
-	ENTER();
-
-	/* Make sure it is not a multicast address */
-	if (is_multicast_ether_addr(mac_addr)) {
-		pr_debug("MAC address is multicast\n");
-		is_valid = false;
-	}
-	/* Not a broadcast address */
-	else if (is_broadcast_ether_addr(mac_addr)) {
-		pr_debug("MAC address is broadcast\n");
-		is_valid = false;
-	}
-	/* Reject the zero address */
-	else if (is_zero_ether_addr(mac_addr)) {
-		pr_debug("MAC address is all zeros\n");
-		is_valid = false;
-	}
-	return is_valid;
-}
-
-/******************************************************************************
- * Resets the 10GbE link.  Waits the settle time and returns the state of
- * the link.
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static bool
-ixgb_link_reset(struct ixgb_hw *hw)
-{
-	bool link_status = false;
-	u8 wait_retries = MAX_RESET_ITERATIONS;
-	u8 lrst_retries = MAX_RESET_ITERATIONS;
-
-	do {
-		/* Reset the link */
-		IXGB_WRITE_REG(hw, CTRL0,
-			       IXGB_READ_REG(hw, CTRL0) | IXGB_CTRL0_LRST);
-
-		/* Wait for link-up and lane re-alignment */
-		do {
-			udelay(IXGB_DELAY_USECS_AFTER_LINK_RESET);
-			link_status =
-			    ((IXGB_READ_REG(hw, STATUS) & IXGB_STATUS_LU)
-			     && (IXGB_READ_REG(hw, XPCSS) &
-				 IXGB_XPCSS_ALIGN_STATUS)) ? true : false;
-		} while (!link_status && --wait_retries);
-
-	} while (!link_status && --lrst_retries);
-
-	return link_status;
-}
-
-/******************************************************************************
- * Resets the 10GbE optics module.
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-static void
-ixgb_optics_reset(struct ixgb_hw *hw)
-{
-	if (hw->phy_type == ixgb_phy_type_txn17401) {
-		ixgb_write_phy_reg(hw,
-				   MDIO_CTRL1,
-				   IXGB_PHY_ADDRESS,
-				   MDIO_MMD_PMAPMD,
-				   MDIO_CTRL1_RESET);
-
-		ixgb_read_phy_reg(hw, MDIO_CTRL1, IXGB_PHY_ADDRESS, MDIO_MMD_PMAPMD);
-	}
-}
-
-/******************************************************************************
- * Resets the 10GbE optics module for Sun variant NIC.
- *
- * hw - Struct containing variables accessed by shared code
- *****************************************************************************/
-
-#define   IXGB_BCM8704_USER_PMD_TX_CTRL_REG         0xC803
-#define   IXGB_BCM8704_USER_PMD_TX_CTRL_REG_VAL     0x0164
-#define   IXGB_BCM8704_USER_CTRL_REG                0xC800
-#define   IXGB_BCM8704_USER_CTRL_REG_VAL            0x7FBF
-#define   IXGB_BCM8704_USER_DEV3_ADDR               0x0003
-#define   IXGB_SUN_PHY_ADDRESS                      0x0000
-#define   IXGB_SUN_PHY_RESET_DELAY                     305
-
-static void
-ixgb_optics_reset_bcm(struct ixgb_hw *hw)
-{
-	u32 ctrl = IXGB_READ_REG(hw, CTRL0);
-	ctrl &= ~IXGB_CTRL0_SDP2;
-	ctrl |= IXGB_CTRL0_SDP3;
-	IXGB_WRITE_REG(hw, CTRL0, ctrl);
-	IXGB_WRITE_FLUSH(hw);
-
-	/* SerDes needs extra delay */
-	msleep(IXGB_SUN_PHY_RESET_DELAY);
-
-	/* Broadcom 7408L configuration */
-	/* Reference clock config */
-	ixgb_write_phy_reg(hw,
-			   IXGB_BCM8704_USER_PMD_TX_CTRL_REG,
-			   IXGB_SUN_PHY_ADDRESS,
-			   IXGB_BCM8704_USER_DEV3_ADDR,
-			   IXGB_BCM8704_USER_PMD_TX_CTRL_REG_VAL);
-	/*  we must read the registers twice */
-	ixgb_read_phy_reg(hw,
-			  IXGB_BCM8704_USER_PMD_TX_CTRL_REG,
-			  IXGB_SUN_PHY_ADDRESS,
-			  IXGB_BCM8704_USER_DEV3_ADDR);
-	ixgb_read_phy_reg(hw,
-			  IXGB_BCM8704_USER_PMD_TX_CTRL_REG,
-			  IXGB_SUN_PHY_ADDRESS,
-			  IXGB_BCM8704_USER_DEV3_ADDR);
-
-	ixgb_write_phy_reg(hw,
-			   IXGB_BCM8704_USER_CTRL_REG,
-			   IXGB_SUN_PHY_ADDRESS,
-			   IXGB_BCM8704_USER_DEV3_ADDR,
-			   IXGB_BCM8704_USER_CTRL_REG_VAL);
-	ixgb_read_phy_reg(hw,
-			  IXGB_BCM8704_USER_CTRL_REG,
-			  IXGB_SUN_PHY_ADDRESS,
-			  IXGB_BCM8704_USER_DEV3_ADDR);
-	ixgb_read_phy_reg(hw,
-			  IXGB_BCM8704_USER_CTRL_REG,
-			  IXGB_SUN_PHY_ADDRESS,
-			  IXGB_BCM8704_USER_DEV3_ADDR);
-
-	/* SerDes needs extra delay */
-	msleep(IXGB_SUN_PHY_RESET_DELAY);
-}
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_hw.h b/drivers/net/ethernet/intel/ixgb/ixgb_hw.h
deleted file mode 100644
index 70bcff5fb3db..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_hw.h
+++ /dev/null
@@ -1,767 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#ifndef _IXGB_HW_H_
-#define _IXGB_HW_H_
-
-#include <linux/mdio.h>
-
-#include "ixgb_osdep.h"
-
-/* Enums */
-typedef enum {
-	ixgb_mac_unknown = 0,
-	ixgb_82597,
-	ixgb_num_macs
-} ixgb_mac_type;
-
-/* Types of physical layer modules */
-typedef enum {
-	ixgb_phy_type_unknown = 0,
-	ixgb_phy_type_g6005,	/* 850nm, MM fiber, XPAK transceiver */
-	ixgb_phy_type_g6104,	/* 1310nm, SM fiber, XPAK transceiver */
-	ixgb_phy_type_txn17201,	/* 850nm, MM fiber, XPAK transceiver */
-	ixgb_phy_type_txn17401,	/* 1310nm, SM fiber, XENPAK transceiver */
-	ixgb_phy_type_bcm	/* SUN specific board */
-} ixgb_phy_type;
-
-/* XPAK transceiver vendors, for the SR adapters */
-typedef enum {
-	ixgb_xpak_vendor_intel,
-	ixgb_xpak_vendor_infineon
-} ixgb_xpak_vendor;
-
-/* Media Types */
-typedef enum {
-	ixgb_media_type_unknown = 0,
-	ixgb_media_type_fiber = 1,
-	ixgb_media_type_copper = 2,
-	ixgb_num_media_types
-} ixgb_media_type;
-
-/* Flow Control Settings */
-typedef enum {
-	ixgb_fc_none = 0,
-	ixgb_fc_rx_pause = 1,
-	ixgb_fc_tx_pause = 2,
-	ixgb_fc_full = 3,
-	ixgb_fc_default = 0xFF
-} ixgb_fc_type;
-
-/* PCI bus types */
-typedef enum {
-	ixgb_bus_type_unknown = 0,
-	ixgb_bus_type_pci,
-	ixgb_bus_type_pcix
-} ixgb_bus_type;
-
-/* PCI bus speeds */
-typedef enum {
-	ixgb_bus_speed_unknown = 0,
-	ixgb_bus_speed_33,
-	ixgb_bus_speed_66,
-	ixgb_bus_speed_100,
-	ixgb_bus_speed_133,
-	ixgb_bus_speed_reserved
-} ixgb_bus_speed;
-
-/* PCI bus widths */
-typedef enum {
-	ixgb_bus_width_unknown = 0,
-	ixgb_bus_width_32,
-	ixgb_bus_width_64
-} ixgb_bus_width;
-
-#define IXGB_EEPROM_SIZE    64	/* Size in words */
-
-#define SPEED_10000  10000
-#define FULL_DUPLEX  2
-
-#define MIN_NUMBER_OF_DESCRIPTORS       8
-#define MAX_NUMBER_OF_DESCRIPTORS  0xFFF8	/* 13 bits in RDLEN/TDLEN, 128B aligned     */
-
-#define IXGB_DELAY_BEFORE_RESET        10	/* allow 10ms after idling rx/tx units      */
-#define IXGB_DELAY_AFTER_RESET          1	/* allow 1ms after the reset                */
-#define IXGB_DELAY_AFTER_EE_RESET      10	/* allow 10ms after the EEPROM reset        */
-
-#define IXGB_DELAY_USECS_AFTER_LINK_RESET    13	/* allow 13 microseconds after the reset    */
-					   /* NOTE: this is MICROSECONDS               */
-#define MAX_RESET_ITERATIONS            8	/* number of iterations to get things right */
-
-/* General Registers */
-#define IXGB_CTRL0   0x00000	/* Device Control Register 0 - RW */
-#define IXGB_CTRL1   0x00008	/* Device Control Register 1 - RW */
-#define IXGB_STATUS  0x00010	/* Device Status Register - RO */
-#define IXGB_EECD    0x00018	/* EEPROM/Flash Control/Data Register - RW */
-#define IXGB_MFS     0x00020	/* Maximum Frame Size - RW */
-
-/* Interrupt */
-#define IXGB_ICR     0x00080	/* Interrupt Cause Read - R/clr */
-#define IXGB_ICS     0x00088	/* Interrupt Cause Set - RW */
-#define IXGB_IMS     0x00090	/* Interrupt Mask Set/Read - RW */
-#define IXGB_IMC     0x00098	/* Interrupt Mask Clear - WO */
-
-/* Receive */
-#define IXGB_RCTL    0x00100	/* RX Control - RW */
-#define IXGB_FCRTL   0x00108	/* Flow Control Receive Threshold Low - RW */
-#define IXGB_FCRTH   0x00110	/* Flow Control Receive Threshold High - RW */
-#define IXGB_RDBAL   0x00118	/* RX Descriptor Base Low - RW */
-#define IXGB_RDBAH   0x0011C	/* RX Descriptor Base High - RW */
-#define IXGB_RDLEN   0x00120	/* RX Descriptor Length - RW */
-#define IXGB_RDH     0x00128	/* RX Descriptor Head - RW */
-#define IXGB_RDT     0x00130	/* RX Descriptor Tail - RW */
-#define IXGB_RDTR    0x00138	/* RX Delay Timer Ring - RW */
-#define IXGB_RXDCTL  0x00140	/* Receive Descriptor Control - RW */
-#define IXGB_RAIDC   0x00148	/* Receive Adaptive Interrupt Delay Control - RW */
-#define IXGB_RXCSUM  0x00158	/* Receive Checksum Control - RW */
-#define IXGB_RA      0x00180	/* Receive Address Array Base - RW */
-#define IXGB_RAL     0x00180	/* Receive Address Low [0:15] - RW */
-#define IXGB_RAH     0x00184	/* Receive Address High [0:15] - RW */
-#define IXGB_MTA     0x00200	/* Multicast Table Array [0:127] - RW */
-#define IXGB_VFTA    0x00400	/* VLAN Filter Table Array [0:127] - RW */
-#define IXGB_REQ_RX_DESCRIPTOR_MULTIPLE 8
-
-/* Transmit */
-#define IXGB_TCTL    0x00600	/* TX Control - RW */
-#define IXGB_TDBAL   0x00608	/* TX Descriptor Base Low - RW */
-#define IXGB_TDBAH   0x0060C	/* TX Descriptor Base High - RW */
-#define IXGB_TDLEN   0x00610	/* TX Descriptor Length - RW */
-#define IXGB_TDH     0x00618	/* TX Descriptor Head - RW */
-#define IXGB_TDT     0x00620	/* TX Descriptor Tail - RW */
-#define IXGB_TIDV    0x00628	/* TX Interrupt Delay Value - RW */
-#define IXGB_TXDCTL  0x00630	/* Transmit Descriptor Control - RW */
-#define IXGB_TSPMT   0x00638	/* TCP Segmentation PAD & Min Threshold - RW */
-#define IXGB_PAP     0x00640	/* Pause and Pace - RW */
-#define IXGB_REQ_TX_DESCRIPTOR_MULTIPLE 8
-
-/* Physical */
-#define IXGB_PCSC1   0x00700	/* PCS Control 1 - RW */
-#define IXGB_PCSC2   0x00708	/* PCS Control 2 - RW */
-#define IXGB_PCSS1   0x00710	/* PCS Status 1 - RO */
-#define IXGB_PCSS2   0x00718	/* PCS Status 2 - RO */
-#define IXGB_XPCSS   0x00720	/* 10GBASE-X PCS Status (or XGXS Lane Status) - RO */
-#define IXGB_UCCR    0x00728	/* Unilink Circuit Control Register */
-#define IXGB_XPCSTC  0x00730	/* 10GBASE-X PCS Test Control */
-#define IXGB_MACA    0x00738	/* MDI Autoscan Command and Address - RW */
-#define IXGB_APAE    0x00740	/* Autoscan PHY Address Enable - RW */
-#define IXGB_ARD     0x00748	/* Autoscan Read Data - RO */
-#define IXGB_AIS     0x00750	/* Autoscan Interrupt Status - RO */
-#define IXGB_MSCA    0x00758	/* MDI Single Command and Address - RW */
-#define IXGB_MSRWD   0x00760	/* MDI Single Read and Write Data - RW, RO */
-
-/* Wake-up */
-#define IXGB_WUFC    0x00808	/* Wake Up Filter Control - RW */
-#define IXGB_WUS     0x00810	/* Wake Up Status - RO */
-#define IXGB_FFLT    0x01000	/* Flexible Filter Length Table - RW */
-#define IXGB_FFMT    0x01020	/* Flexible Filter Mask Table - RW */
-#define IXGB_FTVT    0x01420	/* Flexible Filter Value Table - RW */
-
-/* Statistics */
-#define IXGB_TPRL    0x02000	/* Total Packets Received (Low) */
-#define IXGB_TPRH    0x02004	/* Total Packets Received (High) */
-#define IXGB_GPRCL   0x02008	/* Good Packets Received Count (Low) */
-#define IXGB_GPRCH   0x0200C	/* Good Packets Received Count (High) */
-#define IXGB_BPRCL   0x02010	/* Broadcast Packets Received Count (Low) */
-#define IXGB_BPRCH   0x02014	/* Broadcast Packets Received Count (High) */
-#define IXGB_MPRCL   0x02018	/* Multicast Packets Received Count (Low) */
-#define IXGB_MPRCH   0x0201C	/* Multicast Packets Received Count (High) */
-#define IXGB_UPRCL   0x02020	/* Unicast Packets Received Count (Low) */
-#define IXGB_UPRCH   0x02024	/* Unicast Packets Received Count (High) */
-#define IXGB_VPRCL   0x02028	/* VLAN Packets Received Count (Low) */
-#define IXGB_VPRCH   0x0202C	/* VLAN Packets Received Count (High) */
-#define IXGB_JPRCL   0x02030	/* Jumbo Packets Received Count (Low) */
-#define IXGB_JPRCH   0x02034	/* Jumbo Packets Received Count (High) */
-#define IXGB_GORCL   0x02038	/* Good Octets Received Count (Low) */
-#define IXGB_GORCH   0x0203C	/* Good Octets Received Count (High) */
-#define IXGB_TORL    0x02040	/* Total Octets Received (Low) */
-#define IXGB_TORH    0x02044	/* Total Octets Received (High) */
-#define IXGB_RNBC    0x02048	/* Receive No Buffers Count */
-#define IXGB_RUC     0x02050	/* Receive Undersize Count */
-#define IXGB_ROC     0x02058	/* Receive Oversize Count */
-#define IXGB_RLEC    0x02060	/* Receive Length Error Count */
-#define IXGB_CRCERRS 0x02068	/* CRC Error Count */
-#define IXGB_ICBC    0x02070	/* Illegal control byte in mid-packet Count */
-#define IXGB_ECBC    0x02078	/* Error Control byte in mid-packet Count */
-#define IXGB_MPC     0x02080	/* Missed Packets Count */
-#define IXGB_TPTL    0x02100	/* Total Packets Transmitted (Low) */
-#define IXGB_TPTH    0x02104	/* Total Packets Transmitted (High) */
-#define IXGB_GPTCL   0x02108	/* Good Packets Transmitted Count (Low) */
-#define IXGB_GPTCH   0x0210C	/* Good Packets Transmitted Count (High) */
-#define IXGB_BPTCL   0x02110	/* Broadcast Packets Transmitted Count (Low) */
-#define IXGB_BPTCH   0x02114	/* Broadcast Packets Transmitted Count (High) */
-#define IXGB_MPTCL   0x02118	/* Multicast Packets Transmitted Count (Low) */
-#define IXGB_MPTCH   0x0211C	/* Multicast Packets Transmitted Count (High) */
-#define IXGB_UPTCL   0x02120	/* Unicast Packets Transmitted Count (Low) */
-#define IXGB_UPTCH   0x02124	/* Unicast Packets Transmitted Count (High) */
-#define IXGB_VPTCL   0x02128	/* VLAN Packets Transmitted Count (Low) */
-#define IXGB_VPTCH   0x0212C	/* VLAN Packets Transmitted Count (High) */
-#define IXGB_JPTCL   0x02130	/* Jumbo Packets Transmitted Count (Low) */
-#define IXGB_JPTCH   0x02134	/* Jumbo Packets Transmitted Count (High) */
-#define IXGB_GOTCL   0x02138	/* Good Octets Transmitted Count (Low) */
-#define IXGB_GOTCH   0x0213C	/* Good Octets Transmitted Count (High) */
-#define IXGB_TOTL    0x02140	/* Total Octets Transmitted Count (Low) */
-#define IXGB_TOTH    0x02144	/* Total Octets Transmitted Count (High) */
-#define IXGB_DC      0x02148	/* Defer Count */
-#define IXGB_PLT64C  0x02150	/* Packet Transmitted was less than 64 bytes Count */
-#define IXGB_TSCTC   0x02170	/* TCP Segmentation Context Transmitted Count */
-#define IXGB_TSCTFC  0x02178	/* TCP Segmentation Context Tx Fail Count */
-#define IXGB_IBIC    0x02180	/* Illegal byte during Idle stream count */
-#define IXGB_RFC     0x02188	/* Remote Fault Count */
-#define IXGB_LFC     0x02190	/* Local Fault Count */
-#define IXGB_PFRC    0x02198	/* Pause Frame Receive Count */
-#define IXGB_PFTC    0x021A0	/* Pause Frame Transmit Count */
-#define IXGB_MCFRC   0x021A8	/* MAC Control Frames (non-Pause) Received Count */
-#define IXGB_MCFTC   0x021B0	/* MAC Control Frames (non-Pause) Transmitted Count */
-#define IXGB_XONRXC  0x021B8	/* XON Received Count */
-#define IXGB_XONTXC  0x021C0	/* XON Transmitted Count */
-#define IXGB_XOFFRXC 0x021C8	/* XOFF Received Count */
-#define IXGB_XOFFTXC 0x021D0	/* XOFF Transmitted Count */
-#define IXGB_RJC     0x021D8	/* Receive Jabber Count */
-
-/* CTRL0 Bit Masks */
-#define IXGB_CTRL0_LRST     0x00000008
-#define IXGB_CTRL0_JFE      0x00000010
-#define IXGB_CTRL0_XLE      0x00000020
-#define IXGB_CTRL0_MDCS     0x00000040
-#define IXGB_CTRL0_CMDC     0x00000080
-#define IXGB_CTRL0_SDP0     0x00040000
-#define IXGB_CTRL0_SDP1     0x00080000
-#define IXGB_CTRL0_SDP2     0x00100000
-#define IXGB_CTRL0_SDP3     0x00200000
-#define IXGB_CTRL0_SDP0_DIR 0x00400000
-#define IXGB_CTRL0_SDP1_DIR 0x00800000
-#define IXGB_CTRL0_SDP2_DIR 0x01000000
-#define IXGB_CTRL0_SDP3_DIR 0x02000000
-#define IXGB_CTRL0_RST      0x04000000
-#define IXGB_CTRL0_RPE      0x08000000
-#define IXGB_CTRL0_TPE      0x10000000
-#define IXGB_CTRL0_VME      0x40000000
-
-/* CTRL1 Bit Masks */
-#define IXGB_CTRL1_GPI0_EN     0x00000001
-#define IXGB_CTRL1_GPI1_EN     0x00000002
-#define IXGB_CTRL1_GPI2_EN     0x00000004
-#define IXGB_CTRL1_GPI3_EN     0x00000008
-#define IXGB_CTRL1_SDP4        0x00000010
-#define IXGB_CTRL1_SDP5        0x00000020
-#define IXGB_CTRL1_SDP6        0x00000040
-#define IXGB_CTRL1_SDP7        0x00000080
-#define IXGB_CTRL1_SDP4_DIR    0x00000100
-#define IXGB_CTRL1_SDP5_DIR    0x00000200
-#define IXGB_CTRL1_SDP6_DIR    0x00000400
-#define IXGB_CTRL1_SDP7_DIR    0x00000800
-#define IXGB_CTRL1_EE_RST      0x00002000
-#define IXGB_CTRL1_RO_DIS      0x00020000
-#define IXGB_CTRL1_PCIXHM_MASK 0x00C00000
-#define IXGB_CTRL1_PCIXHM_1_2  0x00000000
-#define IXGB_CTRL1_PCIXHM_5_8  0x00400000
-#define IXGB_CTRL1_PCIXHM_3_4  0x00800000
-#define IXGB_CTRL1_PCIXHM_7_8  0x00C00000
-
-/* STATUS Bit Masks */
-#define IXGB_STATUS_LU            0x00000002
-#define IXGB_STATUS_AIP           0x00000004
-#define IXGB_STATUS_TXOFF         0x00000010
-#define IXGB_STATUS_XAUIME        0x00000020
-#define IXGB_STATUS_RES           0x00000040
-#define IXGB_STATUS_RIS           0x00000080
-#define IXGB_STATUS_RIE           0x00000100
-#define IXGB_STATUS_RLF           0x00000200
-#define IXGB_STATUS_RRF           0x00000400
-#define IXGB_STATUS_PCI_SPD       0x00000800
-#define IXGB_STATUS_BUS64         0x00001000
-#define IXGB_STATUS_PCIX_MODE     0x00002000
-#define IXGB_STATUS_PCIX_SPD_MASK 0x0000C000
-#define IXGB_STATUS_PCIX_SPD_66   0x00000000
-#define IXGB_STATUS_PCIX_SPD_100  0x00004000
-#define IXGB_STATUS_PCIX_SPD_133  0x00008000
-#define IXGB_STATUS_REV_ID_MASK   0x000F0000
-#define IXGB_STATUS_REV_ID_SHIFT  16
-
-/* EECD Bit Masks */
-#define IXGB_EECD_SK       0x00000001
-#define IXGB_EECD_CS       0x00000002
-#define IXGB_EECD_DI       0x00000004
-#define IXGB_EECD_DO       0x00000008
-#define IXGB_EECD_FWE_MASK 0x00000030
-#define IXGB_EECD_FWE_DIS  0x00000010
-#define IXGB_EECD_FWE_EN   0x00000020
-
-/* MFS */
-#define IXGB_MFS_SHIFT 16
-
-/* Interrupt Register Bit Masks (used for ICR, ICS, IMS, and IMC) */
-#define IXGB_INT_TXDW     0x00000001
-#define IXGB_INT_TXQE     0x00000002
-#define IXGB_INT_LSC      0x00000004
-#define IXGB_INT_RXSEQ    0x00000008
-#define IXGB_INT_RXDMT0   0x00000010
-#define IXGB_INT_RXO      0x00000040
-#define IXGB_INT_RXT0     0x00000080
-#define IXGB_INT_AUTOSCAN 0x00000200
-#define IXGB_INT_GPI0     0x00000800
-#define IXGB_INT_GPI1     0x00001000
-#define IXGB_INT_GPI2     0x00002000
-#define IXGB_INT_GPI3     0x00004000
-
-/* RCTL Bit Masks */
-#define IXGB_RCTL_RXEN        0x00000002
-#define IXGB_RCTL_SBP         0x00000004
-#define IXGB_RCTL_UPE         0x00000008
-#define IXGB_RCTL_MPE         0x00000010
-#define IXGB_RCTL_RDMTS_MASK  0x00000300
-#define IXGB_RCTL_RDMTS_1_2   0x00000000
-#define IXGB_RCTL_RDMTS_1_4   0x00000100
-#define IXGB_RCTL_RDMTS_1_8   0x00000200
-#define IXGB_RCTL_MO_MASK     0x00003000
-#define IXGB_RCTL_MO_47_36    0x00000000
-#define IXGB_RCTL_MO_46_35    0x00001000
-#define IXGB_RCTL_MO_45_34    0x00002000
-#define IXGB_RCTL_MO_43_32    0x00003000
-#define IXGB_RCTL_MO_SHIFT    12
-#define IXGB_RCTL_BAM         0x00008000
-#define IXGB_RCTL_BSIZE_MASK  0x00030000
-#define IXGB_RCTL_BSIZE_2048  0x00000000
-#define IXGB_RCTL_BSIZE_4096  0x00010000
-#define IXGB_RCTL_BSIZE_8192  0x00020000
-#define IXGB_RCTL_BSIZE_16384 0x00030000
-#define IXGB_RCTL_VFE         0x00040000
-#define IXGB_RCTL_CFIEN       0x00080000
-#define IXGB_RCTL_CFI         0x00100000
-#define IXGB_RCTL_RPDA_MASK   0x00600000
-#define IXGB_RCTL_RPDA_MC_MAC 0x00000000
-#define IXGB_RCTL_MC_ONLY     0x00400000
-#define IXGB_RCTL_CFF         0x00800000
-#define IXGB_RCTL_SECRC       0x04000000
-#define IXGB_RDT_FPDB         0x80000000
-
-#define IXGB_RCTL_IDLE_RX_UNIT 0
-
-/* FCRTL Bit Masks */
-#define IXGB_FCRTL_XONE       0x80000000
-
-/* RXDCTL Bit Masks */
-#define IXGB_RXDCTL_PTHRESH_MASK  0x000001FF
-#define IXGB_RXDCTL_PTHRESH_SHIFT 0
-#define IXGB_RXDCTL_HTHRESH_MASK  0x0003FE00
-#define IXGB_RXDCTL_HTHRESH_SHIFT 9
-#define IXGB_RXDCTL_WTHRESH_MASK  0x07FC0000
-#define IXGB_RXDCTL_WTHRESH_SHIFT 18
-
-/* RAIDC Bit Masks */
-#define IXGB_RAIDC_HIGHTHRS_MASK 0x0000003F
-#define IXGB_RAIDC_DELAY_MASK    0x000FF800
-#define IXGB_RAIDC_DELAY_SHIFT   11
-#define IXGB_RAIDC_POLL_MASK     0x1FF00000
-#define IXGB_RAIDC_POLL_SHIFT    20
-#define IXGB_RAIDC_RXT_GATE      0x40000000
-#define IXGB_RAIDC_EN            0x80000000
-
-#define IXGB_RAIDC_POLL_1000_INTERRUPTS_PER_SECOND      1220
-#define IXGB_RAIDC_POLL_5000_INTERRUPTS_PER_SECOND      244
-#define IXGB_RAIDC_POLL_10000_INTERRUPTS_PER_SECOND     122
-#define IXGB_RAIDC_POLL_20000_INTERRUPTS_PER_SECOND     61
-
-/* RXCSUM Bit Masks */
-#define IXGB_RXCSUM_IPOFL 0x00000100
-#define IXGB_RXCSUM_TUOFL 0x00000200
-
-/* RAH Bit Masks */
-#define IXGB_RAH_ASEL_MASK 0x00030000
-#define IXGB_RAH_ASEL_DEST 0x00000000
-#define IXGB_RAH_ASEL_SRC  0x00010000
-#define IXGB_RAH_AV        0x80000000
-
-/* TCTL Bit Masks */
-#define IXGB_TCTL_TCE  0x00000001
-#define IXGB_TCTL_TXEN 0x00000002
-#define IXGB_TCTL_TPDE 0x00000004
-
-#define IXGB_TCTL_IDLE_TX_UNIT  0
-
-/* TXDCTL Bit Masks */
-#define IXGB_TXDCTL_PTHRESH_MASK  0x0000007F
-#define IXGB_TXDCTL_HTHRESH_MASK  0x00007F00
-#define IXGB_TXDCTL_HTHRESH_SHIFT 8
-#define IXGB_TXDCTL_WTHRESH_MASK  0x007F0000
-#define IXGB_TXDCTL_WTHRESH_SHIFT 16
-
-/* TSPMT Bit Masks */
-#define IXGB_TSPMT_TSMT_MASK   0x0000FFFF
-#define IXGB_TSPMT_TSPBP_MASK  0xFFFF0000
-#define IXGB_TSPMT_TSPBP_SHIFT 16
-
-/* PAP Bit Masks */
-#define IXGB_PAP_TXPC_MASK 0x0000FFFF
-#define IXGB_PAP_TXPV_MASK 0x000F0000
-#define IXGB_PAP_TXPV_10G  0x00000000
-#define IXGB_PAP_TXPV_1G   0x00010000
-#define IXGB_PAP_TXPV_2G   0x00020000
-#define IXGB_PAP_TXPV_3G   0x00030000
-#define IXGB_PAP_TXPV_4G   0x00040000
-#define IXGB_PAP_TXPV_5G   0x00050000
-#define IXGB_PAP_TXPV_6G   0x00060000
-#define IXGB_PAP_TXPV_7G   0x00070000
-#define IXGB_PAP_TXPV_8G   0x00080000
-#define IXGB_PAP_TXPV_9G   0x00090000
-#define IXGB_PAP_TXPV_WAN  0x000F0000
-
-/* PCSC1 Bit Masks */
-#define IXGB_PCSC1_LOOPBACK 0x00004000
-
-/* PCSC2 Bit Masks */
-#define IXGB_PCSC2_PCS_TYPE_MASK  0x00000003
-#define IXGB_PCSC2_PCS_TYPE_10GBX 0x00000001
-
-/* PCSS1 Bit Masks */
-#define IXGB_PCSS1_LOCAL_FAULT    0x00000080
-#define IXGB_PCSS1_RX_LINK_STATUS 0x00000004
-
-/* PCSS2 Bit Masks */
-#define IXGB_PCSS2_DEV_PRES_MASK 0x0000C000
-#define IXGB_PCSS2_DEV_PRES      0x00004000
-#define IXGB_PCSS2_TX_LF         0x00000800
-#define IXGB_PCSS2_RX_LF         0x00000400
-#define IXGB_PCSS2_10GBW         0x00000004
-#define IXGB_PCSS2_10GBX         0x00000002
-#define IXGB_PCSS2_10GBR         0x00000001
-
-/* XPCSS Bit Masks */
-#define IXGB_XPCSS_ALIGN_STATUS 0x00001000
-#define IXGB_XPCSS_PATTERN_TEST 0x00000800
-#define IXGB_XPCSS_LANE_3_SYNC  0x00000008
-#define IXGB_XPCSS_LANE_2_SYNC  0x00000004
-#define IXGB_XPCSS_LANE_1_SYNC  0x00000002
-#define IXGB_XPCSS_LANE_0_SYNC  0x00000001
-
-/* XPCSTC Bit Masks */
-#define IXGB_XPCSTC_BERT_TRIG       0x00200000
-#define IXGB_XPCSTC_BERT_SST        0x00100000
-#define IXGB_XPCSTC_BERT_PSZ_MASK   0x000C0000
-#define IXGB_XPCSTC_BERT_PSZ_SHIFT  17
-#define IXGB_XPCSTC_BERT_PSZ_INF    0x00000003
-#define IXGB_XPCSTC_BERT_PSZ_68     0x00000001
-#define IXGB_XPCSTC_BERT_PSZ_1028   0x00000000
-
-/* MSCA bit Masks */
-/* New Protocol Address */
-#define IXGB_MSCA_NP_ADDR_MASK      0x0000FFFF
-#define IXGB_MSCA_NP_ADDR_SHIFT     0
-/* Either Device Type or Register Address,depending on ST_CODE */
-#define IXGB_MSCA_DEV_TYPE_MASK     0x001F0000
-#define IXGB_MSCA_DEV_TYPE_SHIFT    16
-#define IXGB_MSCA_PHY_ADDR_MASK     0x03E00000
-#define IXGB_MSCA_PHY_ADDR_SHIFT    21
-#define IXGB_MSCA_OP_CODE_MASK      0x0C000000
-/* OP_CODE == 00, Address cycle, New Protocol           */
-/* OP_CODE == 01, Write operation                       */
-/* OP_CODE == 10, Read operation                        */
-/* OP_CODE == 11, Read, auto increment, New Protocol    */
-#define IXGB_MSCA_ADDR_CYCLE        0x00000000
-#define IXGB_MSCA_WRITE             0x04000000
-#define IXGB_MSCA_READ              0x08000000
-#define IXGB_MSCA_READ_AUTOINC      0x0C000000
-#define IXGB_MSCA_OP_CODE_SHIFT     26
-#define IXGB_MSCA_ST_CODE_MASK      0x30000000
-/* ST_CODE == 00, New Protocol  */
-/* ST_CODE == 01, Old Protocol  */
-#define IXGB_MSCA_NEW_PROTOCOL      0x00000000
-#define IXGB_MSCA_OLD_PROTOCOL      0x10000000
-#define IXGB_MSCA_ST_CODE_SHIFT     28
-/* Initiate command, self-clearing when command completes */
-#define IXGB_MSCA_MDI_COMMAND       0x40000000
-/*MDI In Progress Enable. */
-#define IXGB_MSCA_MDI_IN_PROG_EN    0x80000000
-
-/* MSRWD bit masks */
-#define IXGB_MSRWD_WRITE_DATA_MASK  0x0000FFFF
-#define IXGB_MSRWD_WRITE_DATA_SHIFT 0
-#define IXGB_MSRWD_READ_DATA_MASK   0xFFFF0000
-#define IXGB_MSRWD_READ_DATA_SHIFT  16
-
-/* Definitions for the optics devices on the MDIO bus. */
-#define IXGB_PHY_ADDRESS             0x0	/* Single PHY, multiple "Devices" */
-
-#define MDIO_PMA_PMD_XPAK_VENDOR_NAME       0x803A	/* XPAK/XENPAK devices only */
-
-/* Vendor-specific MDIO registers */
-#define G6XXX_PMA_PMD_VS1                   0xC001	/* Vendor-specific register */
-#define G6XXX_XGXS_XAUI_VS2                 0x18	/* Vendor-specific register */
-
-#define G6XXX_PMA_PMD_VS1_PLL_RESET         0x80
-#define G6XXX_PMA_PMD_VS1_REMOVE_PLL_RESET  0x00
-#define G6XXX_XGXS_XAUI_VS2_INPUT_MASK      0x0F	/* XAUI lanes synchronized */
-
-/* Layout of a single receive descriptor.  The controller assumes that this
- * structure is packed into 16 bytes, which is a safe assumption with most
- * compilers.  However, some compilers may insert padding between the fields,
- * in which case the structure must be packed in some compiler-specific
- * manner. */
-struct ixgb_rx_desc {
-	__le64 buff_addr;
-	__le16 length;
-	__le16 reserved;
-	u8 status;
-	u8 errors;
-	__le16 special;
-};
-
-#define IXGB_RX_DESC_STATUS_DD    0x01
-#define IXGB_RX_DESC_STATUS_EOP   0x02
-#define IXGB_RX_DESC_STATUS_IXSM  0x04
-#define IXGB_RX_DESC_STATUS_VP    0x08
-#define IXGB_RX_DESC_STATUS_TCPCS 0x20
-#define IXGB_RX_DESC_STATUS_IPCS  0x40
-#define IXGB_RX_DESC_STATUS_PIF   0x80
-
-#define IXGB_RX_DESC_ERRORS_CE   0x01
-#define IXGB_RX_DESC_ERRORS_SE   0x02
-#define IXGB_RX_DESC_ERRORS_P    0x08
-#define IXGB_RX_DESC_ERRORS_TCPE 0x20
-#define IXGB_RX_DESC_ERRORS_IPE  0x40
-#define IXGB_RX_DESC_ERRORS_RXE  0x80
-
-#define IXGB_RX_DESC_SPECIAL_VLAN_MASK  0x0FFF	/* VLAN ID is in lower 12 bits */
-#define IXGB_RX_DESC_SPECIAL_PRI_MASK   0xE000	/* Priority is in upper 3 bits */
-#define IXGB_RX_DESC_SPECIAL_PRI_SHIFT  0x000D	/* Priority is in upper 3 of 16 */
-
-/* Layout of a single transmit descriptor.  The controller assumes that this
- * structure is packed into 16 bytes, which is a safe assumption with most
- * compilers.  However, some compilers may insert padding between the fields,
- * in which case the structure must be packed in some compiler-specific
- * manner. */
-struct ixgb_tx_desc {
-	__le64 buff_addr;
-	__le32 cmd_type_len;
-	u8 status;
-	u8 popts;
-	__le16 vlan;
-};
-
-#define IXGB_TX_DESC_LENGTH_MASK    0x000FFFFF
-#define IXGB_TX_DESC_TYPE_MASK      0x00F00000
-#define IXGB_TX_DESC_TYPE_SHIFT     20
-#define IXGB_TX_DESC_CMD_MASK       0xFF000000
-#define IXGB_TX_DESC_CMD_SHIFT      24
-#define IXGB_TX_DESC_CMD_EOP        0x01000000
-#define IXGB_TX_DESC_CMD_TSE        0x04000000
-#define IXGB_TX_DESC_CMD_RS         0x08000000
-#define IXGB_TX_DESC_CMD_VLE        0x40000000
-#define IXGB_TX_DESC_CMD_IDE        0x80000000
-
-#define IXGB_TX_DESC_TYPE           0x00100000
-
-#define IXGB_TX_DESC_STATUS_DD  0x01
-
-#define IXGB_TX_DESC_POPTS_IXSM 0x01
-#define IXGB_TX_DESC_POPTS_TXSM 0x02
-#define IXGB_TX_DESC_SPECIAL_PRI_SHIFT  IXGB_RX_DESC_SPECIAL_PRI_SHIFT	/* Priority is in upper 3 of 16 */
-
-struct ixgb_context_desc {
-	u8 ipcss;
-	u8 ipcso;
-	__le16 ipcse;
-	u8 tucss;
-	u8 tucso;
-	__le16 tucse;
-	__le32 cmd_type_len;
-	u8 status;
-	u8 hdr_len;
-	__le16 mss;
-};
-
-#define IXGB_CONTEXT_DESC_CMD_TCP 0x01000000
-#define IXGB_CONTEXT_DESC_CMD_IP  0x02000000
-#define IXGB_CONTEXT_DESC_CMD_TSE 0x04000000
-#define IXGB_CONTEXT_DESC_CMD_RS  0x08000000
-#define IXGB_CONTEXT_DESC_CMD_IDE 0x80000000
-
-#define IXGB_CONTEXT_DESC_TYPE 0x00000000
-
-#define IXGB_CONTEXT_DESC_STATUS_DD 0x01
-
-/* Filters */
-#define IXGB_MC_TBL_SIZE          128	/* Multicast Filter Table (4096 bits) */
-#define IXGB_VLAN_FILTER_TBL_SIZE 128	/* VLAN Filter Table (4096 bits) */
-#define IXGB_RAR_ENTRIES		  3	/* Number of entries in Rx Address array */
-
-#define IXGB_MEMORY_REGISTER_BASE_ADDRESS   0
-#define ENET_HEADER_SIZE			14
-#define ENET_FCS_LENGTH			 4
-#define IXGB_MAX_NUM_MULTICAST_ADDRESSES	128
-#define IXGB_MIN_ENET_FRAME_SIZE_WITHOUT_FCS	60
-#define IXGB_MAX_ENET_FRAME_SIZE_WITHOUT_FCS	1514
-#define IXGB_MAX_JUMBO_FRAME_SIZE		0x3F00
-
-/* Phy Addresses */
-#define IXGB_OPTICAL_PHY_ADDR 0x0	/* Optical Module phy address */
-#define IXGB_XAUII_PHY_ADDR   0x1	/* Xauii transceiver phy address */
-#define IXGB_DIAG_PHY_ADDR    0x1F	/* Diagnostic Device phy address */
-
-/* This structure takes a 64k flash and maps it for identification commands */
-struct ixgb_flash_buffer {
-	u8 manufacturer_id;
-	u8 device_id;
-	u8 filler1[0x2AA8];
-	u8 cmd2;
-	u8 filler2[0x2AAA];
-	u8 cmd1;
-	u8 filler3[0xAAAA];
-};
-
-/* Flow control parameters */
-struct ixgb_fc {
-	u32 high_water;	/* Flow Control High-water          */
-	u32 low_water;	/* Flow Control Low-water           */
-	u16 pause_time;	/* Flow Control Pause timer         */
-	bool send_xon;		/* Flow control send XON            */
-	ixgb_fc_type type;	/* Type of flow control             */
-};
-
-/* The historical defaults for the flow control values are given below. */
-#define FC_DEFAULT_HI_THRESH        (0x8000)	/* 32KB */
-#define FC_DEFAULT_LO_THRESH        (0x4000)	/* 16KB */
-#define FC_DEFAULT_TX_TIMER         (0x100)	/* ~130 us */
-
-/* Phy definitions */
-#define IXGB_MAX_PHY_REG_ADDRESS    0xFFFF
-#define IXGB_MAX_PHY_ADDRESS        31
-#define IXGB_MAX_PHY_DEV_TYPE       31
-
-/* Bus parameters */
-struct ixgb_bus {
-	ixgb_bus_speed speed;
-	ixgb_bus_width width;
-	ixgb_bus_type type;
-};
-
-struct ixgb_hw {
-	u8 __iomem *hw_addr;/* Base Address of the hardware     */
-	void *back;		/* Pointer to OS-dependent struct   */
-	struct ixgb_fc fc;	/* Flow control parameters          */
-	struct ixgb_bus bus;	/* Bus parameters                   */
-	u32 phy_id;	/* Phy Identifier                   */
-	u32 phy_addr;	/* XGMII address of Phy             */
-	ixgb_mac_type mac_type;	/* Identifier for MAC controller    */
-	ixgb_phy_type phy_type;	/* Transceiver/phy identifier       */
-	u32 max_frame_size;	/* Maximum frame size supported     */
-	u32 mc_filter_type;	/* Multicast filter hash type       */
-	u32 num_mc_addrs;	/* Number of current Multicast addrs */
-	u8 curr_mac_addr[ETH_ALEN];	/* Individual address currently programmed in MAC */
-	u32 num_tx_desc;	/* Number of Transmit descriptors   */
-	u32 num_rx_desc;	/* Number of Receive descriptors    */
-	u32 rx_buffer_size;	/* Size of Receive buffer           */
-	bool link_up;		/* true if link is valid            */
-	bool adapter_stopped;	/* State of adapter                 */
-	u16 device_id;	/* device id from PCI configuration space */
-	u16 vendor_id;	/* vendor id from PCI configuration space */
-	u8 revision_id;	/* revision id from PCI configuration space */
-	u16 subsystem_vendor_id;	/* subsystem vendor id from PCI configuration space */
-	u16 subsystem_id;	/* subsystem id from PCI configuration space */
-	u32 bar0;		/* Base Address registers           */
-	u32 bar1;
-	u32 bar2;
-	u32 bar3;
-	u16 pci_cmd_word;	/* PCI command register id from PCI configuration space */
-	__le16 eeprom[IXGB_EEPROM_SIZE];	/* EEPROM contents read at init time  */
-	unsigned long io_base;	/* Our I/O mapped location */
-	u32 lastLFC;
-	u32 lastRFC;
-};
-
-/* Statistics reported by the hardware */
-struct ixgb_hw_stats {
-	u64 tprl;
-	u64 tprh;
-	u64 gprcl;
-	u64 gprch;
-	u64 bprcl;
-	u64 bprch;
-	u64 mprcl;
-	u64 mprch;
-	u64 uprcl;
-	u64 uprch;
-	u64 vprcl;
-	u64 vprch;
-	u64 jprcl;
-	u64 jprch;
-	u64 gorcl;
-	u64 gorch;
-	u64 torl;
-	u64 torh;
-	u64 rnbc;
-	u64 ruc;
-	u64 roc;
-	u64 rlec;
-	u64 crcerrs;
-	u64 icbc;
-	u64 ecbc;
-	u64 mpc;
-	u64 tptl;
-	u64 tpth;
-	u64 gptcl;
-	u64 gptch;
-	u64 bptcl;
-	u64 bptch;
-	u64 mptcl;
-	u64 mptch;
-	u64 uptcl;
-	u64 uptch;
-	u64 vptcl;
-	u64 vptch;
-	u64 jptcl;
-	u64 jptch;
-	u64 gotcl;
-	u64 gotch;
-	u64 totl;
-	u64 toth;
-	u64 dc;
-	u64 plt64c;
-	u64 tsctc;
-	u64 tsctfc;
-	u64 ibic;
-	u64 rfc;
-	u64 lfc;
-	u64 pfrc;
-	u64 pftc;
-	u64 mcfrc;
-	u64 mcftc;
-	u64 xonrxc;
-	u64 xontxc;
-	u64 xoffrxc;
-	u64 xofftxc;
-	u64 rjc;
-};
-
-/* Function Prototypes */
-bool ixgb_adapter_stop(struct ixgb_hw *hw);
-bool ixgb_init_hw(struct ixgb_hw *hw);
-bool ixgb_adapter_start(struct ixgb_hw *hw);
-void ixgb_check_for_link(struct ixgb_hw *hw);
-bool ixgb_check_for_bad_link(struct ixgb_hw *hw);
-
-void ixgb_rar_set(struct ixgb_hw *hw, const u8 *addr, u32 index);
-
-/* Filters (multicast, vlan, receive) */
-void ixgb_mc_addr_list_update(struct ixgb_hw *hw, u8 *mc_addr_list,
-			      u32 mc_addr_count, u32 pad);
-
-/* Vfta functions */
-void ixgb_write_vfta(struct ixgb_hw *hw, u32 offset, u32 value);
-
-/* Access functions to eeprom data */
-void ixgb_get_ee_mac_addr(struct ixgb_hw *hw, u8 *mac_addr);
-u32 ixgb_get_ee_pba_number(struct ixgb_hw *hw);
-u16 ixgb_get_ee_device_id(struct ixgb_hw *hw);
-bool ixgb_get_eeprom_data(struct ixgb_hw *hw);
-__le16 ixgb_get_eeprom_word(struct ixgb_hw *hw, u16 index);
-
-/* Everything else */
-void ixgb_led_on(struct ixgb_hw *hw);
-void ixgb_led_off(struct ixgb_hw *hw);
-void ixgb_write_pci_cfg(struct ixgb_hw *hw,
-			 u32 reg,
-			 u16 * value);
-
-
-#endif /* _IXGB_HW_H_ */
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_ids.h b/drivers/net/ethernet/intel/ixgb/ixgb_ids.h
deleted file mode 100644
index 9695b8215f01..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_ids.h
+++ /dev/null
@@ -1,23 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#ifndef _IXGB_IDS_H_
-#define _IXGB_IDS_H_
-
-/**********************************************************************
-** The Device and Vendor IDs for 10 Gigabit MACs
-**********************************************************************/
-
-#define IXGB_DEVICE_ID_82597EX      0x1048
-#define IXGB_DEVICE_ID_82597EX_SR   0x1A48
-#define IXGB_DEVICE_ID_82597EX_LR   0x1B48
-#define IXGB_SUBDEVICE_ID_A11F      0xA11F
-#define IXGB_SUBDEVICE_ID_A01F      0xA01F
-
-#define IXGB_DEVICE_ID_82597EX_CX4   0x109E
-#define IXGB_SUBDEVICE_ID_A00C  0xA00C
-#define IXGB_SUBDEVICE_ID_A01C  0xA01C
-#define IXGB_SUBDEVICE_ID_7036  0x7036
-
-#endif /* #ifndef _IXGB_IDS_H_ */
-/* End of File */
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
deleted file mode 100644
index b4d47e7a76c8..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c
+++ /dev/null
@@ -1,2285 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/prefetch.h>
-#include "ixgb.h"
-
-char ixgb_driver_name[] = "ixgb";
-static char ixgb_driver_string[] = "Intel(R) PRO/10GbE Network Driver";
-
-static const char ixgb_copyright[] = "Copyright (c) 1999-2008 Intel Corporation.";
-
-#define IXGB_CB_LENGTH 256
-static unsigned int copybreak __read_mostly = IXGB_CB_LENGTH;
-module_param(copybreak, uint, 0644);
-MODULE_PARM_DESC(copybreak,
-	"Maximum size of packet that is copied to a new buffer on receive");
-
-/* ixgb_pci_tbl - PCI Device ID Table
- *
- * Wildcard entries (PCI_ANY_ID) should come last
- * Last entry must be all 0s
- *
- * { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
- *   Class, Class Mask, private data (not used) }
- */
-static const struct pci_device_id ixgb_pci_tbl[] = {
-	{PCI_VENDOR_ID_INTEL, IXGB_DEVICE_ID_82597EX,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_INTEL, IXGB_DEVICE_ID_82597EX_CX4,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_INTEL, IXGB_DEVICE_ID_82597EX_SR,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_INTEL, IXGB_DEVICE_ID_82597EX_LR,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-
-	/* required last entry */
-	{0,}
-};
-
-MODULE_DEVICE_TABLE(pci, ixgb_pci_tbl);
-
-/* Local Function Prototypes */
-static int ixgb_init_module(void);
-static void ixgb_exit_module(void);
-static int ixgb_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
-static void ixgb_remove(struct pci_dev *pdev);
-static int ixgb_sw_init(struct ixgb_adapter *adapter);
-static int ixgb_open(struct net_device *netdev);
-static int ixgb_close(struct net_device *netdev);
-static void ixgb_configure_tx(struct ixgb_adapter *adapter);
-static void ixgb_configure_rx(struct ixgb_adapter *adapter);
-static void ixgb_setup_rctl(struct ixgb_adapter *adapter);
-static void ixgb_clean_tx_ring(struct ixgb_adapter *adapter);
-static void ixgb_clean_rx_ring(struct ixgb_adapter *adapter);
-static void ixgb_set_multi(struct net_device *netdev);
-static void ixgb_watchdog(struct timer_list *t);
-static netdev_tx_t ixgb_xmit_frame(struct sk_buff *skb,
-				   struct net_device *netdev);
-static int ixgb_change_mtu(struct net_device *netdev, int new_mtu);
-static int ixgb_set_mac(struct net_device *netdev, void *p);
-static irqreturn_t ixgb_intr(int irq, void *data);
-static bool ixgb_clean_tx_irq(struct ixgb_adapter *adapter);
-
-static int ixgb_clean(struct napi_struct *, int);
-static bool ixgb_clean_rx_irq(struct ixgb_adapter *, int *, int);
-static void ixgb_alloc_rx_buffers(struct ixgb_adapter *, int);
-
-static void ixgb_tx_timeout(struct net_device *dev, unsigned int txqueue);
-static void ixgb_tx_timeout_task(struct work_struct *work);
-
-static void ixgb_vlan_strip_enable(struct ixgb_adapter *adapter);
-static void ixgb_vlan_strip_disable(struct ixgb_adapter *adapter);
-static int ixgb_vlan_rx_add_vid(struct net_device *netdev,
-				__be16 proto, u16 vid);
-static int ixgb_vlan_rx_kill_vid(struct net_device *netdev,
-				 __be16 proto, u16 vid);
-static void ixgb_restore_vlan(struct ixgb_adapter *adapter);
-
-static pci_ers_result_t ixgb_io_error_detected (struct pci_dev *pdev,
-                             pci_channel_state_t state);
-static pci_ers_result_t ixgb_io_slot_reset (struct pci_dev *pdev);
-static void ixgb_io_resume (struct pci_dev *pdev);
-
-static const struct pci_error_handlers ixgb_err_handler = {
-	.error_detected = ixgb_io_error_detected,
-	.slot_reset = ixgb_io_slot_reset,
-	.resume = ixgb_io_resume,
-};
-
-static struct pci_driver ixgb_driver = {
-	.name     = ixgb_driver_name,
-	.id_table = ixgb_pci_tbl,
-	.probe    = ixgb_probe,
-	.remove   = ixgb_remove,
-	.err_handler = &ixgb_err_handler
-};
-
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
-MODULE_DESCRIPTION("Intel(R) PRO/10GbE Network Driver");
-MODULE_LICENSE("GPL v2");
-
-#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK)
-static int debug = -1;
-module_param(debug, int, 0);
-MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
-
-/**
- * ixgb_init_module - Driver Registration Routine
- *
- * ixgb_init_module is the first routine called when the driver is
- * loaded. All it does is register with the PCI subsystem.
- **/
-
-static int __init
-ixgb_init_module(void)
-{
-	pr_info("%s\n", ixgb_driver_string);
-	pr_info("%s\n", ixgb_copyright);
-
-	return pci_register_driver(&ixgb_driver);
-}
-
-module_init(ixgb_init_module);
-
-/**
- * ixgb_exit_module - Driver Exit Cleanup Routine
- *
- * ixgb_exit_module is called just before the driver is removed
- * from memory.
- **/
-
-static void __exit
-ixgb_exit_module(void)
-{
-	pci_unregister_driver(&ixgb_driver);
-}
-
-module_exit(ixgb_exit_module);
-
-/**
- * ixgb_irq_disable - Mask off interrupt generation on the NIC
- * @adapter: board private structure
- **/
-
-static void
-ixgb_irq_disable(struct ixgb_adapter *adapter)
-{
-	IXGB_WRITE_REG(&adapter->hw, IMC, ~0);
-	IXGB_WRITE_FLUSH(&adapter->hw);
-	synchronize_irq(adapter->pdev->irq);
-}
-
-/**
- * ixgb_irq_enable - Enable default interrupt generation settings
- * @adapter: board private structure
- **/
-
-static void
-ixgb_irq_enable(struct ixgb_adapter *adapter)
-{
-	u32 val = IXGB_INT_RXT0 | IXGB_INT_RXDMT0 |
-		  IXGB_INT_TXDW | IXGB_INT_LSC;
-	if (adapter->hw.subsystem_vendor_id == PCI_VENDOR_ID_SUN)
-		val |= IXGB_INT_GPI0;
-	IXGB_WRITE_REG(&adapter->hw, IMS, val);
-	IXGB_WRITE_FLUSH(&adapter->hw);
-}
-
-int
-ixgb_up(struct ixgb_adapter *adapter)
-{
-	struct net_device *netdev = adapter->netdev;
-	int err, irq_flags = IRQF_SHARED;
-	int max_frame = netdev->mtu + ENET_HEADER_SIZE + ENET_FCS_LENGTH;
-	struct ixgb_hw *hw = &adapter->hw;
-
-	/* hardware has been reset, we need to reload some things */
-
-	ixgb_rar_set(hw, netdev->dev_addr, 0);
-	ixgb_set_multi(netdev);
-
-	ixgb_restore_vlan(adapter);
-
-	ixgb_configure_tx(adapter);
-	ixgb_setup_rctl(adapter);
-	ixgb_configure_rx(adapter);
-	ixgb_alloc_rx_buffers(adapter, IXGB_DESC_UNUSED(&adapter->rx_ring));
-
-	/* disable interrupts and get the hardware into a known state */
-	IXGB_WRITE_REG(&adapter->hw, IMC, 0xffffffff);
-
-	/* only enable MSI if bus is in PCI-X mode */
-	if (IXGB_READ_REG(&adapter->hw, STATUS) & IXGB_STATUS_PCIX_MODE) {
-		err = pci_enable_msi(adapter->pdev);
-		if (!err) {
-			adapter->have_msi = true;
-			irq_flags = 0;
-		}
-		/* proceed to try to request regular interrupt */
-	}
-
-	err = request_irq(adapter->pdev->irq, ixgb_intr, irq_flags,
-	                  netdev->name, netdev);
-	if (err) {
-		if (adapter->have_msi)
-			pci_disable_msi(adapter->pdev);
-		netif_err(adapter, probe, adapter->netdev,
-			  "Unable to allocate interrupt Error: %d\n", err);
-		return err;
-	}
-
-	if ((hw->max_frame_size != max_frame) ||
-		(hw->max_frame_size !=
-		(IXGB_READ_REG(hw, MFS) >> IXGB_MFS_SHIFT))) {
-
-		hw->max_frame_size = max_frame;
-
-		IXGB_WRITE_REG(hw, MFS, hw->max_frame_size << IXGB_MFS_SHIFT);
-
-		if (hw->max_frame_size >
-		   IXGB_MAX_ENET_FRAME_SIZE_WITHOUT_FCS + ENET_FCS_LENGTH) {
-			u32 ctrl0 = IXGB_READ_REG(hw, CTRL0);
-
-			if (!(ctrl0 & IXGB_CTRL0_JFE)) {
-				ctrl0 |= IXGB_CTRL0_JFE;
-				IXGB_WRITE_REG(hw, CTRL0, ctrl0);
-			}
-		}
-	}
-
-	clear_bit(__IXGB_DOWN, &adapter->flags);
-
-	napi_enable(&adapter->napi);
-	ixgb_irq_enable(adapter);
-
-	netif_wake_queue(netdev);
-
-	mod_timer(&adapter->watchdog_timer, jiffies);
-
-	return 0;
-}
-
-void
-ixgb_down(struct ixgb_adapter *adapter, bool kill_watchdog)
-{
-	struct net_device *netdev = adapter->netdev;
-
-	/* prevent the interrupt handler from restarting watchdog */
-	set_bit(__IXGB_DOWN, &adapter->flags);
-
-	netif_carrier_off(netdev);
-
-	napi_disable(&adapter->napi);
-	/* waiting for NAPI to complete can re-enable interrupts */
-	ixgb_irq_disable(adapter);
-	free_irq(adapter->pdev->irq, netdev);
-
-	if (adapter->have_msi)
-		pci_disable_msi(adapter->pdev);
-
-	if (kill_watchdog)
-		del_timer_sync(&adapter->watchdog_timer);
-
-	adapter->link_speed = 0;
-	adapter->link_duplex = 0;
-	netif_stop_queue(netdev);
-
-	ixgb_reset(adapter);
-	ixgb_clean_tx_ring(adapter);
-	ixgb_clean_rx_ring(adapter);
-}
-
-void
-ixgb_reset(struct ixgb_adapter *adapter)
-{
-	struct ixgb_hw *hw = &adapter->hw;
-
-	ixgb_adapter_stop(hw);
-	if (!ixgb_init_hw(hw))
-		netif_err(adapter, probe, adapter->netdev, "ixgb_init_hw failed\n");
-
-	/* restore frame size information */
-	IXGB_WRITE_REG(hw, MFS, hw->max_frame_size << IXGB_MFS_SHIFT);
-	if (hw->max_frame_size >
-	    IXGB_MAX_ENET_FRAME_SIZE_WITHOUT_FCS + ENET_FCS_LENGTH) {
-		u32 ctrl0 = IXGB_READ_REG(hw, CTRL0);
-		if (!(ctrl0 & IXGB_CTRL0_JFE)) {
-			ctrl0 |= IXGB_CTRL0_JFE;
-			IXGB_WRITE_REG(hw, CTRL0, ctrl0);
-		}
-	}
-}
-
-static netdev_features_t
-ixgb_fix_features(struct net_device *netdev, netdev_features_t features)
-{
-	/*
-	 * Tx VLAN insertion does not work per HW design when Rx stripping is
-	 * disabled.
-	 */
-	if (!(features & NETIF_F_HW_VLAN_CTAG_RX))
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
-}
-
-static int
-ixgb_set_features(struct net_device *netdev, netdev_features_t features)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	netdev_features_t changed = features ^ netdev->features;
-
-	if (!(changed & (NETIF_F_RXCSUM|NETIF_F_HW_VLAN_CTAG_RX)))
-		return 0;
-
-	adapter->rx_csum = !!(features & NETIF_F_RXCSUM);
-
-	if (netif_running(netdev)) {
-		ixgb_down(adapter, true);
-		ixgb_up(adapter);
-		ixgb_set_speed_duplex(netdev);
-	} else
-		ixgb_reset(adapter);
-
-	return 0;
-}
-
-
-static const struct net_device_ops ixgb_netdev_ops = {
-	.ndo_open 		= ixgb_open,
-	.ndo_stop		= ixgb_close,
-	.ndo_start_xmit		= ixgb_xmit_frame,
-	.ndo_set_rx_mode	= ixgb_set_multi,
-	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_set_mac_address	= ixgb_set_mac,
-	.ndo_change_mtu		= ixgb_change_mtu,
-	.ndo_tx_timeout		= ixgb_tx_timeout,
-	.ndo_vlan_rx_add_vid	= ixgb_vlan_rx_add_vid,
-	.ndo_vlan_rx_kill_vid	= ixgb_vlan_rx_kill_vid,
-	.ndo_fix_features       = ixgb_fix_features,
-	.ndo_set_features       = ixgb_set_features,
-};
-
-/**
- * ixgb_probe - Device Initialization Routine
- * @pdev: PCI device information struct
- * @ent: entry in ixgb_pci_tbl
- *
- * Returns 0 on success, negative on failure
- *
- * ixgb_probe initializes an adapter identified by a pci_dev structure.
- * The OS initialization, configuring of the adapter private structure,
- * and a hardware reset occur.
- **/
-
-static int
-ixgb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
-{
-	struct net_device *netdev = NULL;
-	struct ixgb_adapter *adapter;
-	static int cards_found = 0;
-	u8 addr[ETH_ALEN];
-	int i;
-	int err;
-
-	err = pci_enable_device(pdev);
-	if (err)
-		return err;
-
-	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
-	if (err) {
-		pr_err("No usable DMA configuration, aborting\n");
-		goto err_dma_mask;
-	}
-
-	err = pci_request_regions(pdev, ixgb_driver_name);
-	if (err)
-		goto err_request_regions;
-
-	pci_set_master(pdev);
-
-	netdev = alloc_etherdev(sizeof(struct ixgb_adapter));
-	if (!netdev) {
-		err = -ENOMEM;
-		goto err_alloc_etherdev;
-	}
-
-	SET_NETDEV_DEV(netdev, &pdev->dev);
-
-	pci_set_drvdata(pdev, netdev);
-	adapter = netdev_priv(netdev);
-	adapter->netdev = netdev;
-	adapter->pdev = pdev;
-	adapter->hw.back = adapter;
-	adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
-
-	adapter->hw.hw_addr = pci_ioremap_bar(pdev, BAR_0);
-	if (!adapter->hw.hw_addr) {
-		err = -EIO;
-		goto err_ioremap;
-	}
-
-	for (i = BAR_1; i < PCI_STD_NUM_BARS; i++) {
-		if (pci_resource_len(pdev, i) == 0)
-			continue;
-		if (pci_resource_flags(pdev, i) & IORESOURCE_IO) {
-			adapter->hw.io_base = pci_resource_start(pdev, i);
-			break;
-		}
-	}
-
-	netdev->netdev_ops = &ixgb_netdev_ops;
-	ixgb_set_ethtool_ops(netdev);
-	netdev->watchdog_timeo = 5 * HZ;
-	netif_napi_add(netdev, &adapter->napi, ixgb_clean);
-
-	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
-
-	adapter->bd_number = cards_found;
-	adapter->link_speed = 0;
-	adapter->link_duplex = 0;
-
-	/* setup the private structure */
-
-	err = ixgb_sw_init(adapter);
-	if (err)
-		goto err_sw_init;
-
-	netdev->hw_features = NETIF_F_SG |
-			   NETIF_F_TSO |
-			   NETIF_F_HW_CSUM |
-			   NETIF_F_HW_VLAN_CTAG_TX |
-			   NETIF_F_HW_VLAN_CTAG_RX;
-	netdev->features = netdev->hw_features |
-			   NETIF_F_HW_VLAN_CTAG_FILTER;
-	netdev->hw_features |= NETIF_F_RXCSUM;
-
-	netdev->features |= NETIF_F_HIGHDMA;
-	netdev->vlan_features |= NETIF_F_HIGHDMA;
-
-	/* MTU range: 68 - 16114 */
-	netdev->min_mtu = ETH_MIN_MTU;
-	netdev->max_mtu = IXGB_MAX_JUMBO_FRAME_SIZE - ETH_HLEN;
-
-	/* make sure the EEPROM is good */
-
-	if (!ixgb_validate_eeprom_checksum(&adapter->hw)) {
-		netif_err(adapter, probe, adapter->netdev,
-			  "The EEPROM Checksum Is Not Valid\n");
-		err = -EIO;
-		goto err_eeprom;
-	}
-
-	ixgb_get_ee_mac_addr(&adapter->hw, addr);
-	eth_hw_addr_set(netdev, addr);
-
-	if (!is_valid_ether_addr(netdev->dev_addr)) {
-		netif_err(adapter, probe, adapter->netdev, "Invalid MAC Address\n");
-		err = -EIO;
-		goto err_eeprom;
-	}
-
-	adapter->part_num = ixgb_get_ee_pba_number(&adapter->hw);
-
-	timer_setup(&adapter->watchdog_timer, ixgb_watchdog, 0);
-
-	INIT_WORK(&adapter->tx_timeout_task, ixgb_tx_timeout_task);
-
-	strcpy(netdev->name, "eth%d");
-	err = register_netdev(netdev);
-	if (err)
-		goto err_register;
-
-	/* carrier off reporting is important to ethtool even BEFORE open */
-	netif_carrier_off(netdev);
-
-	netif_info(adapter, probe, adapter->netdev,
-		   "Intel(R) PRO/10GbE Network Connection\n");
-	ixgb_check_options(adapter);
-	/* reset the hardware with the new settings */
-
-	ixgb_reset(adapter);
-
-	cards_found++;
-	return 0;
-
-err_register:
-err_sw_init:
-err_eeprom:
-	iounmap(adapter->hw.hw_addr);
-err_ioremap:
-	free_netdev(netdev);
-err_alloc_etherdev:
-	pci_release_regions(pdev);
-err_request_regions:
-err_dma_mask:
-	pci_disable_device(pdev);
-	return err;
-}
-
-/**
- * ixgb_remove - Device Removal Routine
- * @pdev: PCI device information struct
- *
- * ixgb_remove is called by the PCI subsystem to alert the driver
- * that it should release a PCI device.  The could be caused by a
- * Hot-Plug event, or because the driver is going to be removed from
- * memory.
- **/
-
-static void
-ixgb_remove(struct pci_dev *pdev)
-{
-	struct net_device *netdev = pci_get_drvdata(pdev);
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	cancel_work_sync(&adapter->tx_timeout_task);
-
-	unregister_netdev(netdev);
-
-	iounmap(adapter->hw.hw_addr);
-	pci_release_regions(pdev);
-
-	free_netdev(netdev);
-	pci_disable_device(pdev);
-}
-
-/**
- * ixgb_sw_init - Initialize general software structures (struct ixgb_adapter)
- * @adapter: board private structure to initialize
- *
- * ixgb_sw_init initializes the Adapter private data structure.
- * Fields are initialized based on PCI device information and
- * OS network device settings (MTU size).
- **/
-
-static int
-ixgb_sw_init(struct ixgb_adapter *adapter)
-{
-	struct ixgb_hw *hw = &adapter->hw;
-	struct net_device *netdev = adapter->netdev;
-	struct pci_dev *pdev = adapter->pdev;
-
-	/* PCI config space info */
-
-	hw->vendor_id = pdev->vendor;
-	hw->device_id = pdev->device;
-	hw->subsystem_vendor_id = pdev->subsystem_vendor;
-	hw->subsystem_id = pdev->subsystem_device;
-
-	hw->max_frame_size = netdev->mtu + ENET_HEADER_SIZE + ENET_FCS_LENGTH;
-	adapter->rx_buffer_len = hw->max_frame_size + 8; /* + 8 for errata */
-
-	if ((hw->device_id == IXGB_DEVICE_ID_82597EX) ||
-	    (hw->device_id == IXGB_DEVICE_ID_82597EX_CX4) ||
-	    (hw->device_id == IXGB_DEVICE_ID_82597EX_LR) ||
-	    (hw->device_id == IXGB_DEVICE_ID_82597EX_SR))
-		hw->mac_type = ixgb_82597;
-	else {
-		/* should never have loaded on this device */
-		netif_err(adapter, probe, adapter->netdev, "unsupported device id\n");
-	}
-
-	/* enable flow control to be programmed */
-	hw->fc.send_xon = 1;
-
-	set_bit(__IXGB_DOWN, &adapter->flags);
-	return 0;
-}
-
-/**
- * ixgb_open - Called when a network interface is made active
- * @netdev: network interface device structure
- *
- * Returns 0 on success, negative value on failure
- *
- * The open entry point is called when a network interface is made
- * active by the system (IFF_UP).  At this point all resources needed
- * for transmit and receive operations are allocated, the interrupt
- * handler is registered with the OS, the watchdog timer is started,
- * and the stack is notified that the interface is ready.
- **/
-
-static int
-ixgb_open(struct net_device *netdev)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	int err;
-
-	/* allocate transmit descriptors */
-	err = ixgb_setup_tx_resources(adapter);
-	if (err)
-		goto err_setup_tx;
-
-	netif_carrier_off(netdev);
-
-	/* allocate receive descriptors */
-
-	err = ixgb_setup_rx_resources(adapter);
-	if (err)
-		goto err_setup_rx;
-
-	err = ixgb_up(adapter);
-	if (err)
-		goto err_up;
-
-	netif_start_queue(netdev);
-
-	return 0;
-
-err_up:
-	ixgb_free_rx_resources(adapter);
-err_setup_rx:
-	ixgb_free_tx_resources(adapter);
-err_setup_tx:
-	ixgb_reset(adapter);
-
-	return err;
-}
-
-/**
- * ixgb_close - Disables a network interface
- * @netdev: network interface device structure
- *
- * Returns 0, this is not allowed to fail
- *
- * The close entry point is called when an interface is de-activated
- * by the OS.  The hardware is still under the drivers control, but
- * needs to be disabled.  A global MAC reset is issued to stop the
- * hardware, and all transmit and receive resources are freed.
- **/
-
-static int
-ixgb_close(struct net_device *netdev)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	ixgb_down(adapter, true);
-
-	ixgb_free_tx_resources(adapter);
-	ixgb_free_rx_resources(adapter);
-
-	return 0;
-}
-
-/**
- * ixgb_setup_tx_resources - allocate Tx resources (Descriptors)
- * @adapter: board private structure
- *
- * Return 0 on success, negative on failure
- **/
-
-int
-ixgb_setup_tx_resources(struct ixgb_adapter *adapter)
-{
-	struct ixgb_desc_ring *txdr = &adapter->tx_ring;
-	struct pci_dev *pdev = adapter->pdev;
-	int size;
-
-	size = sizeof(struct ixgb_buffer) * txdr->count;
-	txdr->buffer_info = vzalloc(size);
-	if (!txdr->buffer_info)
-		return -ENOMEM;
-
-	/* round up to nearest 4K */
-
-	txdr->size = txdr->count * sizeof(struct ixgb_tx_desc);
-	txdr->size = ALIGN(txdr->size, 4096);
-
-	txdr->desc = dma_alloc_coherent(&pdev->dev, txdr->size, &txdr->dma,
-					GFP_KERNEL);
-	if (!txdr->desc) {
-		vfree(txdr->buffer_info);
-		return -ENOMEM;
-	}
-
-	txdr->next_to_use = 0;
-	txdr->next_to_clean = 0;
-
-	return 0;
-}
-
-/**
- * ixgb_configure_tx - Configure 82597 Transmit Unit after Reset.
- * @adapter: board private structure
- *
- * Configure the Tx unit of the MAC after a reset.
- **/
-
-static void
-ixgb_configure_tx(struct ixgb_adapter *adapter)
-{
-	u64 tdba = adapter->tx_ring.dma;
-	u32 tdlen = adapter->tx_ring.count * sizeof(struct ixgb_tx_desc);
-	u32 tctl;
-	struct ixgb_hw *hw = &adapter->hw;
-
-	/* Setup the Base and Length of the Tx Descriptor Ring
-	 * tx_ring.dma can be either a 32 or 64 bit value
-	 */
-
-	IXGB_WRITE_REG(hw, TDBAL, (tdba & 0x00000000ffffffffULL));
-	IXGB_WRITE_REG(hw, TDBAH, (tdba >> 32));
-
-	IXGB_WRITE_REG(hw, TDLEN, tdlen);
-
-	/* Setup the HW Tx Head and Tail descriptor pointers */
-
-	IXGB_WRITE_REG(hw, TDH, 0);
-	IXGB_WRITE_REG(hw, TDT, 0);
-
-	/* don't set up txdctl, it induces performance problems if configured
-	 * incorrectly */
-	/* Set the Tx Interrupt Delay register */
-
-	IXGB_WRITE_REG(hw, TIDV, adapter->tx_int_delay);
-
-	/* Program the Transmit Control Register */
-
-	tctl = IXGB_TCTL_TCE | IXGB_TCTL_TXEN | IXGB_TCTL_TPDE;
-	IXGB_WRITE_REG(hw, TCTL, tctl);
-
-	/* Setup Transmit Descriptor Settings for this adapter */
-	adapter->tx_cmd_type =
-		IXGB_TX_DESC_TYPE |
-		(adapter->tx_int_delay_enable ? IXGB_TX_DESC_CMD_IDE : 0);
-}
-
-/**
- * ixgb_setup_rx_resources - allocate Rx resources (Descriptors)
- * @adapter: board private structure
- *
- * Returns 0 on success, negative on failure
- **/
-
-int
-ixgb_setup_rx_resources(struct ixgb_adapter *adapter)
-{
-	struct ixgb_desc_ring *rxdr = &adapter->rx_ring;
-	struct pci_dev *pdev = adapter->pdev;
-	int size;
-
-	size = sizeof(struct ixgb_buffer) * rxdr->count;
-	rxdr->buffer_info = vzalloc(size);
-	if (!rxdr->buffer_info)
-		return -ENOMEM;
-
-	/* Round up to nearest 4K */
-
-	rxdr->size = rxdr->count * sizeof(struct ixgb_rx_desc);
-	rxdr->size = ALIGN(rxdr->size, 4096);
-
-	rxdr->desc = dma_alloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
-					GFP_KERNEL);
-
-	if (!rxdr->desc) {
-		vfree(rxdr->buffer_info);
-		return -ENOMEM;
-	}
-
-	rxdr->next_to_clean = 0;
-	rxdr->next_to_use = 0;
-
-	return 0;
-}
-
-/**
- * ixgb_setup_rctl - configure the receive control register
- * @adapter: Board private structure
- **/
-
-static void
-ixgb_setup_rctl(struct ixgb_adapter *adapter)
-{
-	u32 rctl;
-
-	rctl = IXGB_READ_REG(&adapter->hw, RCTL);
-
-	rctl &= ~(3 << IXGB_RCTL_MO_SHIFT);
-
-	rctl |=
-		IXGB_RCTL_BAM | IXGB_RCTL_RDMTS_1_2 |
-		IXGB_RCTL_RXEN | IXGB_RCTL_CFF |
-		(adapter->hw.mc_filter_type << IXGB_RCTL_MO_SHIFT);
-
-	rctl |= IXGB_RCTL_SECRC;
-
-	if (adapter->rx_buffer_len <= IXGB_RXBUFFER_2048)
-		rctl |= IXGB_RCTL_BSIZE_2048;
-	else if (adapter->rx_buffer_len <= IXGB_RXBUFFER_4096)
-		rctl |= IXGB_RCTL_BSIZE_4096;
-	else if (adapter->rx_buffer_len <= IXGB_RXBUFFER_8192)
-		rctl |= IXGB_RCTL_BSIZE_8192;
-	else if (adapter->rx_buffer_len <= IXGB_RXBUFFER_16384)
-		rctl |= IXGB_RCTL_BSIZE_16384;
-
-	IXGB_WRITE_REG(&adapter->hw, RCTL, rctl);
-}
-
-/**
- * ixgb_configure_rx - Configure 82597 Receive Unit after Reset.
- * @adapter: board private structure
- *
- * Configure the Rx unit of the MAC after a reset.
- **/
-
-static void
-ixgb_configure_rx(struct ixgb_adapter *adapter)
-{
-	u64 rdba = adapter->rx_ring.dma;
-	u32 rdlen = adapter->rx_ring.count * sizeof(struct ixgb_rx_desc);
-	struct ixgb_hw *hw = &adapter->hw;
-	u32 rctl;
-	u32 rxcsum;
-
-	/* make sure receives are disabled while setting up the descriptors */
-
-	rctl = IXGB_READ_REG(hw, RCTL);
-	IXGB_WRITE_REG(hw, RCTL, rctl & ~IXGB_RCTL_RXEN);
-
-	/* set the Receive Delay Timer Register */
-
-	IXGB_WRITE_REG(hw, RDTR, adapter->rx_int_delay);
-
-	/* Setup the Base and Length of the Rx Descriptor Ring */
-
-	IXGB_WRITE_REG(hw, RDBAL, (rdba & 0x00000000ffffffffULL));
-	IXGB_WRITE_REG(hw, RDBAH, (rdba >> 32));
-
-	IXGB_WRITE_REG(hw, RDLEN, rdlen);
-
-	/* Setup the HW Rx Head and Tail Descriptor Pointers */
-	IXGB_WRITE_REG(hw, RDH, 0);
-	IXGB_WRITE_REG(hw, RDT, 0);
-
-	/* due to the hardware errata with RXDCTL, we are unable to use any of
-	 * the performance enhancing features of it without causing other
-	 * subtle bugs, some of the bugs could include receive length
-	 * corruption at high data rates (WTHRESH > 0) and/or receive
-	 * descriptor ring irregularites (particularly in hardware cache) */
-	IXGB_WRITE_REG(hw, RXDCTL, 0);
-
-	/* Enable Receive Checksum Offload for TCP and UDP */
-	if (adapter->rx_csum) {
-		rxcsum = IXGB_READ_REG(hw, RXCSUM);
-		rxcsum |= IXGB_RXCSUM_TUOFL;
-		IXGB_WRITE_REG(hw, RXCSUM, rxcsum);
-	}
-
-	/* Enable Receives */
-
-	IXGB_WRITE_REG(hw, RCTL, rctl);
-}
-
-/**
- * ixgb_free_tx_resources - Free Tx Resources
- * @adapter: board private structure
- *
- * Free all transmit software resources
- **/
-
-void
-ixgb_free_tx_resources(struct ixgb_adapter *adapter)
-{
-	struct pci_dev *pdev = adapter->pdev;
-
-	ixgb_clean_tx_ring(adapter);
-
-	vfree(adapter->tx_ring.buffer_info);
-	adapter->tx_ring.buffer_info = NULL;
-
-	dma_free_coherent(&pdev->dev, adapter->tx_ring.size,
-			  adapter->tx_ring.desc, adapter->tx_ring.dma);
-
-	adapter->tx_ring.desc = NULL;
-}
-
-static void
-ixgb_unmap_and_free_tx_resource(struct ixgb_adapter *adapter,
-                                struct ixgb_buffer *buffer_info)
-{
-	if (buffer_info->dma) {
-		if (buffer_info->mapped_as_page)
-			dma_unmap_page(&adapter->pdev->dev, buffer_info->dma,
-				       buffer_info->length, DMA_TO_DEVICE);
-		else
-			dma_unmap_single(&adapter->pdev->dev, buffer_info->dma,
-					 buffer_info->length, DMA_TO_DEVICE);
-		buffer_info->dma = 0;
-	}
-
-	if (buffer_info->skb) {
-		dev_kfree_skb_any(buffer_info->skb);
-		buffer_info->skb = NULL;
-	}
-	buffer_info->time_stamp = 0;
-	/* these fields must always be initialized in tx
-	 * buffer_info->length = 0;
-	 * buffer_info->next_to_watch = 0; */
-}
-
-/**
- * ixgb_clean_tx_ring - Free Tx Buffers
- * @adapter: board private structure
- **/
-
-static void
-ixgb_clean_tx_ring(struct ixgb_adapter *adapter)
-{
-	struct ixgb_desc_ring *tx_ring = &adapter->tx_ring;
-	struct ixgb_buffer *buffer_info;
-	unsigned long size;
-	unsigned int i;
-
-	/* Free all the Tx ring sk_buffs */
-
-	for (i = 0; i < tx_ring->count; i++) {
-		buffer_info = &tx_ring->buffer_info[i];
-		ixgb_unmap_and_free_tx_resource(adapter, buffer_info);
-	}
-
-	size = sizeof(struct ixgb_buffer) * tx_ring->count;
-	memset(tx_ring->buffer_info, 0, size);
-
-	/* Zero out the descriptor ring */
-
-	memset(tx_ring->desc, 0, tx_ring->size);
-
-	tx_ring->next_to_use = 0;
-	tx_ring->next_to_clean = 0;
-
-	IXGB_WRITE_REG(&adapter->hw, TDH, 0);
-	IXGB_WRITE_REG(&adapter->hw, TDT, 0);
-}
-
-/**
- * ixgb_free_rx_resources - Free Rx Resources
- * @adapter: board private structure
- *
- * Free all receive software resources
- **/
-
-void
-ixgb_free_rx_resources(struct ixgb_adapter *adapter)
-{
-	struct ixgb_desc_ring *rx_ring = &adapter->rx_ring;
-	struct pci_dev *pdev = adapter->pdev;
-
-	ixgb_clean_rx_ring(adapter);
-
-	vfree(rx_ring->buffer_info);
-	rx_ring->buffer_info = NULL;
-
-	dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc,
-			  rx_ring->dma);
-
-	rx_ring->desc = NULL;
-}
-
-/**
- * ixgb_clean_rx_ring - Free Rx Buffers
- * @adapter: board private structure
- **/
-
-static void
-ixgb_clean_rx_ring(struct ixgb_adapter *adapter)
-{
-	struct ixgb_desc_ring *rx_ring = &adapter->rx_ring;
-	struct ixgb_buffer *buffer_info;
-	struct pci_dev *pdev = adapter->pdev;
-	unsigned long size;
-	unsigned int i;
-
-	/* Free all the Rx ring sk_buffs */
-
-	for (i = 0; i < rx_ring->count; i++) {
-		buffer_info = &rx_ring->buffer_info[i];
-		if (buffer_info->dma) {
-			dma_unmap_single(&pdev->dev,
-					 buffer_info->dma,
-					 buffer_info->length,
-					 DMA_FROM_DEVICE);
-			buffer_info->dma = 0;
-			buffer_info->length = 0;
-		}
-
-		if (buffer_info->skb) {
-			dev_kfree_skb(buffer_info->skb);
-			buffer_info->skb = NULL;
-		}
-	}
-
-	size = sizeof(struct ixgb_buffer) * rx_ring->count;
-	memset(rx_ring->buffer_info, 0, size);
-
-	/* Zero out the descriptor ring */
-
-	memset(rx_ring->desc, 0, rx_ring->size);
-
-	rx_ring->next_to_clean = 0;
-	rx_ring->next_to_use = 0;
-
-	IXGB_WRITE_REG(&adapter->hw, RDH, 0);
-	IXGB_WRITE_REG(&adapter->hw, RDT, 0);
-}
-
-/**
- * ixgb_set_mac - Change the Ethernet Address of the NIC
- * @netdev: network interface device structure
- * @p: pointer to an address structure
- *
- * Returns 0 on success, negative on failure
- **/
-
-static int
-ixgb_set_mac(struct net_device *netdev, void *p)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct sockaddr *addr = p;
-
-	if (!is_valid_ether_addr(addr->sa_data))
-		return -EADDRNOTAVAIL;
-
-	eth_hw_addr_set(netdev, addr->sa_data);
-
-	ixgb_rar_set(&adapter->hw, addr->sa_data, 0);
-
-	return 0;
-}
-
-/**
- * ixgb_set_multi - Multicast and Promiscuous mode set
- * @netdev: network interface device structure
- *
- * The set_multi entry point is called whenever the multicast address
- * list or the network interface flags are updated.  This routine is
- * responsible for configuring the hardware for proper multicast,
- * promiscuous mode, and all-multi behavior.
- **/
-
-static void
-ixgb_set_multi(struct net_device *netdev)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-	struct netdev_hw_addr *ha;
-	u32 rctl;
-
-	/* Check for Promiscuous and All Multicast modes */
-
-	rctl = IXGB_READ_REG(hw, RCTL);
-
-	if (netdev->flags & IFF_PROMISC) {
-		rctl |= (IXGB_RCTL_UPE | IXGB_RCTL_MPE);
-		/* disable VLAN filtering */
-		rctl &= ~IXGB_RCTL_CFIEN;
-		rctl &= ~IXGB_RCTL_VFE;
-	} else {
-		if (netdev->flags & IFF_ALLMULTI) {
-			rctl |= IXGB_RCTL_MPE;
-			rctl &= ~IXGB_RCTL_UPE;
-		} else {
-			rctl &= ~(IXGB_RCTL_UPE | IXGB_RCTL_MPE);
-		}
-		/* enable VLAN filtering */
-		rctl |= IXGB_RCTL_VFE;
-		rctl &= ~IXGB_RCTL_CFIEN;
-	}
-
-	if (netdev_mc_count(netdev) > IXGB_MAX_NUM_MULTICAST_ADDRESSES) {
-		rctl |= IXGB_RCTL_MPE;
-		IXGB_WRITE_REG(hw, RCTL, rctl);
-	} else {
-		u8 *mta = kmalloc_array(ETH_ALEN,
-				        IXGB_MAX_NUM_MULTICAST_ADDRESSES,
-				        GFP_ATOMIC);
-		u8 *addr;
-		if (!mta)
-			goto alloc_failed;
-
-		IXGB_WRITE_REG(hw, RCTL, rctl);
-
-		addr = mta;
-		netdev_for_each_mc_addr(ha, netdev) {
-			memcpy(addr, ha->addr, ETH_ALEN);
-			addr += ETH_ALEN;
-		}
-
-		ixgb_mc_addr_list_update(hw, mta, netdev_mc_count(netdev), 0);
-		kfree(mta);
-	}
-
-alloc_failed:
-	if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
-		ixgb_vlan_strip_enable(adapter);
-	else
-		ixgb_vlan_strip_disable(adapter);
-
-}
-
-/**
- * ixgb_watchdog - Timer Call-back
- * @t: pointer to timer_list containing our private info pointer
- **/
-
-static void
-ixgb_watchdog(struct timer_list *t)
-{
-	struct ixgb_adapter *adapter = from_timer(adapter, t, watchdog_timer);
-	struct net_device *netdev = adapter->netdev;
-	struct ixgb_desc_ring *txdr = &adapter->tx_ring;
-
-	ixgb_check_for_link(&adapter->hw);
-
-	if (ixgb_check_for_bad_link(&adapter->hw)) {
-		/* force the reset path */
-		netif_stop_queue(netdev);
-	}
-
-	if (adapter->hw.link_up) {
-		if (!netif_carrier_ok(netdev)) {
-			netdev_info(netdev,
-				    "NIC Link is Up 10 Gbps Full Duplex, Flow Control: %s\n",
-				    (adapter->hw.fc.type == ixgb_fc_full) ?
-				    "RX/TX" :
-				    (adapter->hw.fc.type == ixgb_fc_rx_pause) ?
-				     "RX" :
-				    (adapter->hw.fc.type == ixgb_fc_tx_pause) ?
-				    "TX" : "None");
-			adapter->link_speed = 10000;
-			adapter->link_duplex = FULL_DUPLEX;
-			netif_carrier_on(netdev);
-		}
-	} else {
-		if (netif_carrier_ok(netdev)) {
-			adapter->link_speed = 0;
-			adapter->link_duplex = 0;
-			netdev_info(netdev, "NIC Link is Down\n");
-			netif_carrier_off(netdev);
-		}
-	}
-
-	ixgb_update_stats(adapter);
-
-	if (!netif_carrier_ok(netdev)) {
-		if (IXGB_DESC_UNUSED(txdr) + 1 < txdr->count) {
-			/* We've lost link, so the controller stops DMA,
-			 * but we've got queued Tx work that's never going
-			 * to get done, so reset controller to flush Tx.
-			 * (Do the reset outside of interrupt context). */
-			schedule_work(&adapter->tx_timeout_task);
-			/* return immediately since reset is imminent */
-			return;
-		}
-	}
-
-	/* Force detection of hung controller every watchdog period */
-	adapter->detect_tx_hung = true;
-
-	/* generate an interrupt to force clean up of any stragglers */
-	IXGB_WRITE_REG(&adapter->hw, ICS, IXGB_INT_TXDW);
-
-	/* Reset the timer */
-	mod_timer(&adapter->watchdog_timer, jiffies + 2 * HZ);
-}
-
-#define IXGB_TX_FLAGS_CSUM		0x00000001
-#define IXGB_TX_FLAGS_VLAN		0x00000002
-#define IXGB_TX_FLAGS_TSO		0x00000004
-
-static int
-ixgb_tso(struct ixgb_adapter *adapter, struct sk_buff *skb)
-{
-	struct ixgb_context_desc *context_desc;
-	unsigned int i;
-	u8 ipcss, ipcso, tucss, tucso, hdr_len;
-	u16 ipcse, tucse, mss;
-
-	if (likely(skb_is_gso(skb))) {
-		struct ixgb_buffer *buffer_info;
-		struct iphdr *iph;
-		int err;
-
-		err = skb_cow_head(skb, 0);
-		if (err < 0)
-			return err;
-
-		hdr_len = skb_tcp_all_headers(skb);
-		mss = skb_shinfo(skb)->gso_size;
-		iph = ip_hdr(skb);
-		iph->tot_len = 0;
-		iph->check = 0;
-		tcp_hdr(skb)->check = ~csum_tcpudp_magic(iph->saddr,
-							 iph->daddr, 0,
-							 IPPROTO_TCP, 0);
-		ipcss = skb_network_offset(skb);
-		ipcso = (void *)&(iph->check) - (void *)skb->data;
-		ipcse = skb_transport_offset(skb) - 1;
-		tucss = skb_transport_offset(skb);
-		tucso = (void *)&(tcp_hdr(skb)->check) - (void *)skb->data;
-		tucse = 0;
-
-		i = adapter->tx_ring.next_to_use;
-		context_desc = IXGB_CONTEXT_DESC(adapter->tx_ring, i);
-		buffer_info = &adapter->tx_ring.buffer_info[i];
-		WARN_ON(buffer_info->dma != 0);
-
-		context_desc->ipcss = ipcss;
-		context_desc->ipcso = ipcso;
-		context_desc->ipcse = cpu_to_le16(ipcse);
-		context_desc->tucss = tucss;
-		context_desc->tucso = tucso;
-		context_desc->tucse = cpu_to_le16(tucse);
-		context_desc->mss = cpu_to_le16(mss);
-		context_desc->hdr_len = hdr_len;
-		context_desc->status = 0;
-		context_desc->cmd_type_len = cpu_to_le32(
-						  IXGB_CONTEXT_DESC_TYPE
-						| IXGB_CONTEXT_DESC_CMD_TSE
-						| IXGB_CONTEXT_DESC_CMD_IP
-						| IXGB_CONTEXT_DESC_CMD_TCP
-						| IXGB_CONTEXT_DESC_CMD_IDE
-						| (skb->len - (hdr_len)));
-
-
-		if (++i == adapter->tx_ring.count) i = 0;
-		adapter->tx_ring.next_to_use = i;
-
-		return 1;
-	}
-
-	return 0;
-}
-
-static bool
-ixgb_tx_csum(struct ixgb_adapter *adapter, struct sk_buff *skb)
-{
-	struct ixgb_context_desc *context_desc;
-	unsigned int i;
-	u8 css, cso;
-
-	if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
-		struct ixgb_buffer *buffer_info;
-		css = skb_checksum_start_offset(skb);
-		cso = css + skb->csum_offset;
-
-		i = adapter->tx_ring.next_to_use;
-		context_desc = IXGB_CONTEXT_DESC(adapter->tx_ring, i);
-		buffer_info = &adapter->tx_ring.buffer_info[i];
-		WARN_ON(buffer_info->dma != 0);
-
-		context_desc->tucss = css;
-		context_desc->tucso = cso;
-		context_desc->tucse = 0;
-		/* zero out any previously existing data in one instruction */
-		*(u32 *)&(context_desc->ipcss) = 0;
-		context_desc->status = 0;
-		context_desc->hdr_len = 0;
-		context_desc->mss = 0;
-		context_desc->cmd_type_len =
-			cpu_to_le32(IXGB_CONTEXT_DESC_TYPE
-				    | IXGB_TX_DESC_CMD_IDE);
-
-		if (++i == adapter->tx_ring.count) i = 0;
-		adapter->tx_ring.next_to_use = i;
-
-		return true;
-	}
-
-	return false;
-}
-
-#define IXGB_MAX_TXD_PWR	14
-#define IXGB_MAX_DATA_PER_TXD	(1<<IXGB_MAX_TXD_PWR)
-
-static int
-ixgb_tx_map(struct ixgb_adapter *adapter, struct sk_buff *skb,
-	    unsigned int first)
-{
-	struct ixgb_desc_ring *tx_ring = &adapter->tx_ring;
-	struct pci_dev *pdev = adapter->pdev;
-	struct ixgb_buffer *buffer_info;
-	int len = skb_headlen(skb);
-	unsigned int offset = 0, size, count = 0, i;
-	unsigned int mss = skb_shinfo(skb)->gso_size;
-	unsigned int nr_frags = skb_shinfo(skb)->nr_frags;
-	unsigned int f;
-
-	i = tx_ring->next_to_use;
-
-	while (len) {
-		buffer_info = &tx_ring->buffer_info[i];
-		size = min(len, IXGB_MAX_DATA_PER_TXD);
-		/* Workaround for premature desc write-backs
-		 * in TSO mode.  Append 4-byte sentinel desc */
-		if (unlikely(mss && !nr_frags && size == len && size > 8))
-			size -= 4;
-
-		buffer_info->length = size;
-		WARN_ON(buffer_info->dma != 0);
-		buffer_info->time_stamp = jiffies;
-		buffer_info->mapped_as_page = false;
-		buffer_info->dma = dma_map_single(&pdev->dev,
-						  skb->data + offset,
-						  size, DMA_TO_DEVICE);
-		if (dma_mapping_error(&pdev->dev, buffer_info->dma))
-			goto dma_error;
-		buffer_info->next_to_watch = 0;
-
-		len -= size;
-		offset += size;
-		count++;
-		if (len) {
-			i++;
-			if (i == tx_ring->count)
-				i = 0;
-		}
-	}
-
-	for (f = 0; f < nr_frags; f++) {
-		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
-		len = skb_frag_size(frag);
-		offset = 0;
-
-		while (len) {
-			i++;
-			if (i == tx_ring->count)
-				i = 0;
-
-			buffer_info = &tx_ring->buffer_info[i];
-			size = min(len, IXGB_MAX_DATA_PER_TXD);
-
-			/* Workaround for premature desc write-backs
-			 * in TSO mode.  Append 4-byte sentinel desc */
-			if (unlikely(mss && (f == (nr_frags - 1))
-				     && size == len && size > 8))
-				size -= 4;
-
-			buffer_info->length = size;
-			buffer_info->time_stamp = jiffies;
-			buffer_info->mapped_as_page = true;
-			buffer_info->dma =
-				skb_frag_dma_map(&pdev->dev, frag, offset, size,
-						 DMA_TO_DEVICE);
-			if (dma_mapping_error(&pdev->dev, buffer_info->dma))
-				goto dma_error;
-			buffer_info->next_to_watch = 0;
-
-			len -= size;
-			offset += size;
-			count++;
-		}
-	}
-	tx_ring->buffer_info[i].skb = skb;
-	tx_ring->buffer_info[first].next_to_watch = i;
-
-	return count;
-
-dma_error:
-	dev_err(&pdev->dev, "TX DMA map failed\n");
-	buffer_info->dma = 0;
-	if (count)
-		count--;
-
-	while (count--) {
-		if (i==0)
-			i += tx_ring->count;
-		i--;
-		buffer_info = &tx_ring->buffer_info[i];
-		ixgb_unmap_and_free_tx_resource(adapter, buffer_info);
-	}
-
-	return 0;
-}
-
-static void
-ixgb_tx_queue(struct ixgb_adapter *adapter, int count, int vlan_id,int tx_flags)
-{
-	struct ixgb_desc_ring *tx_ring = &adapter->tx_ring;
-	struct ixgb_tx_desc *tx_desc = NULL;
-	struct ixgb_buffer *buffer_info;
-	u32 cmd_type_len = adapter->tx_cmd_type;
-	u8 status = 0;
-	u8 popts = 0;
-	unsigned int i;
-
-	if (tx_flags & IXGB_TX_FLAGS_TSO) {
-		cmd_type_len |= IXGB_TX_DESC_CMD_TSE;
-		popts |= (IXGB_TX_DESC_POPTS_IXSM | IXGB_TX_DESC_POPTS_TXSM);
-	}
-
-	if (tx_flags & IXGB_TX_FLAGS_CSUM)
-		popts |= IXGB_TX_DESC_POPTS_TXSM;
-
-	if (tx_flags & IXGB_TX_FLAGS_VLAN)
-		cmd_type_len |= IXGB_TX_DESC_CMD_VLE;
-
-	i = tx_ring->next_to_use;
-
-	while (count--) {
-		buffer_info = &tx_ring->buffer_info[i];
-		tx_desc = IXGB_TX_DESC(*tx_ring, i);
-		tx_desc->buff_addr = cpu_to_le64(buffer_info->dma);
-		tx_desc->cmd_type_len =
-			cpu_to_le32(cmd_type_len | buffer_info->length);
-		tx_desc->status = status;
-		tx_desc->popts = popts;
-		tx_desc->vlan = cpu_to_le16(vlan_id);
-
-		if (++i == tx_ring->count) i = 0;
-	}
-
-	tx_desc->cmd_type_len |=
-		cpu_to_le32(IXGB_TX_DESC_CMD_EOP | IXGB_TX_DESC_CMD_RS);
-
-	/* Force memory writes to complete before letting h/w
-	 * know there are new descriptors to fetch.  (Only
-	 * applicable for weak-ordered memory model archs,
-	 * such as IA-64). */
-	wmb();
-
-	tx_ring->next_to_use = i;
-	IXGB_WRITE_REG(&adapter->hw, TDT, i);
-}
-
-static int __ixgb_maybe_stop_tx(struct net_device *netdev, int size)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_desc_ring *tx_ring = &adapter->tx_ring;
-
-	netif_stop_queue(netdev);
-	/* Herbert's original patch had:
-	 *  smp_mb__after_netif_stop_queue();
-	 * but since that doesn't exist yet, just open code it. */
-	smp_mb();
-
-	/* We need to check again in a case another CPU has just
-	 * made room available. */
-	if (likely(IXGB_DESC_UNUSED(tx_ring) < size))
-		return -EBUSY;
-
-	/* A reprieve! */
-	netif_start_queue(netdev);
-	++adapter->restart_queue;
-	return 0;
-}
-
-static int ixgb_maybe_stop_tx(struct net_device *netdev,
-                              struct ixgb_desc_ring *tx_ring, int size)
-{
-	if (likely(IXGB_DESC_UNUSED(tx_ring) >= size))
-		return 0;
-	return __ixgb_maybe_stop_tx(netdev, size);
-}
-
-
-/* Tx Descriptors needed, worst case */
-#define TXD_USE_COUNT(S) (((S) >> IXGB_MAX_TXD_PWR) + \
-			 (((S) & (IXGB_MAX_DATA_PER_TXD - 1)) ? 1 : 0))
-#define DESC_NEEDED TXD_USE_COUNT(IXGB_MAX_DATA_PER_TXD) /* skb->date */ + \
-	MAX_SKB_FRAGS * TXD_USE_COUNT(PAGE_SIZE) + 1 /* for context */ \
-	+ 1 /* one more needed for sentinel TSO workaround */
-
-static netdev_tx_t
-ixgb_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	unsigned int first;
-	unsigned int tx_flags = 0;
-	int vlan_id = 0;
-	int count = 0;
-	int tso;
-
-	if (test_bit(__IXGB_DOWN, &adapter->flags)) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
-	if (skb->len <= 0) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
-	if (unlikely(ixgb_maybe_stop_tx(netdev, &adapter->tx_ring,
-                     DESC_NEEDED)))
-		return NETDEV_TX_BUSY;
-
-	if (skb_vlan_tag_present(skb)) {
-		tx_flags |= IXGB_TX_FLAGS_VLAN;
-		vlan_id = skb_vlan_tag_get(skb);
-	}
-
-	first = adapter->tx_ring.next_to_use;
-
-	tso = ixgb_tso(adapter, skb);
-	if (tso < 0) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
-	if (likely(tso))
-		tx_flags |= IXGB_TX_FLAGS_TSO;
-	else if (ixgb_tx_csum(adapter, skb))
-		tx_flags |= IXGB_TX_FLAGS_CSUM;
-
-	count = ixgb_tx_map(adapter, skb, first);
-
-	if (count) {
-		ixgb_tx_queue(adapter, count, vlan_id, tx_flags);
-		/* Make sure there is space in the ring for the next send. */
-		ixgb_maybe_stop_tx(netdev, &adapter->tx_ring, DESC_NEEDED);
-
-	} else {
-		dev_kfree_skb_any(skb);
-		adapter->tx_ring.buffer_info[first].time_stamp = 0;
-		adapter->tx_ring.next_to_use = first;
-	}
-
-	return NETDEV_TX_OK;
-}
-
-/**
- * ixgb_tx_timeout - Respond to a Tx Hang
- * @netdev: network interface device structure
- * @txqueue: queue hanging (unused)
- **/
-
-static void
-ixgb_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	/* Do the reset outside of interrupt context */
-	schedule_work(&adapter->tx_timeout_task);
-}
-
-static void
-ixgb_tx_timeout_task(struct work_struct *work)
-{
-	struct ixgb_adapter *adapter =
-		container_of(work, struct ixgb_adapter, tx_timeout_task);
-
-	adapter->tx_timeout_count++;
-	ixgb_down(adapter, true);
-	ixgb_up(adapter);
-}
-
-/**
- * ixgb_change_mtu - Change the Maximum Transfer Unit
- * @netdev: network interface device structure
- * @new_mtu: new value for maximum frame size
- *
- * Returns 0 on success, negative on failure
- **/
-
-static int
-ixgb_change_mtu(struct net_device *netdev, int new_mtu)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	int max_frame = new_mtu + ENET_HEADER_SIZE + ENET_FCS_LENGTH;
-
-	if (netif_running(netdev))
-		ixgb_down(adapter, true);
-
-	adapter->rx_buffer_len = max_frame + 8; /* + 8 for errata */
-
-	netdev->mtu = new_mtu;
-
-	if (netif_running(netdev))
-		ixgb_up(adapter);
-
-	return 0;
-}
-
-/**
- * ixgb_update_stats - Update the board statistics counters.
- * @adapter: board private structure
- **/
-
-void
-ixgb_update_stats(struct ixgb_adapter *adapter)
-{
-	struct net_device *netdev = adapter->netdev;
-	struct pci_dev *pdev = adapter->pdev;
-
-	/* Prevent stats update while adapter is being reset */
-	if (pci_channel_offline(pdev))
-		return;
-
-	if ((netdev->flags & IFF_PROMISC) || (netdev->flags & IFF_ALLMULTI) ||
-	   (netdev_mc_count(netdev) > IXGB_MAX_NUM_MULTICAST_ADDRESSES)) {
-		u64 multi = IXGB_READ_REG(&adapter->hw, MPRCL);
-		u32 bcast_l = IXGB_READ_REG(&adapter->hw, BPRCL);
-		u32 bcast_h = IXGB_READ_REG(&adapter->hw, BPRCH);
-		u64 bcast = ((u64)bcast_h << 32) | bcast_l;
-
-		multi |= ((u64)IXGB_READ_REG(&adapter->hw, MPRCH) << 32);
-		/* fix up multicast stats by removing broadcasts */
-		if (multi >= bcast)
-			multi -= bcast;
-
-		adapter->stats.mprcl += (multi & 0xFFFFFFFF);
-		adapter->stats.mprch += (multi >> 32);
-		adapter->stats.bprcl += bcast_l;
-		adapter->stats.bprch += bcast_h;
-	} else {
-		adapter->stats.mprcl += IXGB_READ_REG(&adapter->hw, MPRCL);
-		adapter->stats.mprch += IXGB_READ_REG(&adapter->hw, MPRCH);
-		adapter->stats.bprcl += IXGB_READ_REG(&adapter->hw, BPRCL);
-		adapter->stats.bprch += IXGB_READ_REG(&adapter->hw, BPRCH);
-	}
-	adapter->stats.tprl += IXGB_READ_REG(&adapter->hw, TPRL);
-	adapter->stats.tprh += IXGB_READ_REG(&adapter->hw, TPRH);
-	adapter->stats.gprcl += IXGB_READ_REG(&adapter->hw, GPRCL);
-	adapter->stats.gprch += IXGB_READ_REG(&adapter->hw, GPRCH);
-	adapter->stats.uprcl += IXGB_READ_REG(&adapter->hw, UPRCL);
-	adapter->stats.uprch += IXGB_READ_REG(&adapter->hw, UPRCH);
-	adapter->stats.vprcl += IXGB_READ_REG(&adapter->hw, VPRCL);
-	adapter->stats.vprch += IXGB_READ_REG(&adapter->hw, VPRCH);
-	adapter->stats.jprcl += IXGB_READ_REG(&adapter->hw, JPRCL);
-	adapter->stats.jprch += IXGB_READ_REG(&adapter->hw, JPRCH);
-	adapter->stats.gorcl += IXGB_READ_REG(&adapter->hw, GORCL);
-	adapter->stats.gorch += IXGB_READ_REG(&adapter->hw, GORCH);
-	adapter->stats.torl += IXGB_READ_REG(&adapter->hw, TORL);
-	adapter->stats.torh += IXGB_READ_REG(&adapter->hw, TORH);
-	adapter->stats.rnbc += IXGB_READ_REG(&adapter->hw, RNBC);
-	adapter->stats.ruc += IXGB_READ_REG(&adapter->hw, RUC);
-	adapter->stats.roc += IXGB_READ_REG(&adapter->hw, ROC);
-	adapter->stats.rlec += IXGB_READ_REG(&adapter->hw, RLEC);
-	adapter->stats.crcerrs += IXGB_READ_REG(&adapter->hw, CRCERRS);
-	adapter->stats.icbc += IXGB_READ_REG(&adapter->hw, ICBC);
-	adapter->stats.ecbc += IXGB_READ_REG(&adapter->hw, ECBC);
-	adapter->stats.mpc += IXGB_READ_REG(&adapter->hw, MPC);
-	adapter->stats.tptl += IXGB_READ_REG(&adapter->hw, TPTL);
-	adapter->stats.tpth += IXGB_READ_REG(&adapter->hw, TPTH);
-	adapter->stats.gptcl += IXGB_READ_REG(&adapter->hw, GPTCL);
-	adapter->stats.gptch += IXGB_READ_REG(&adapter->hw, GPTCH);
-	adapter->stats.bptcl += IXGB_READ_REG(&adapter->hw, BPTCL);
-	adapter->stats.bptch += IXGB_READ_REG(&adapter->hw, BPTCH);
-	adapter->stats.mptcl += IXGB_READ_REG(&adapter->hw, MPTCL);
-	adapter->stats.mptch += IXGB_READ_REG(&adapter->hw, MPTCH);
-	adapter->stats.uptcl += IXGB_READ_REG(&adapter->hw, UPTCL);
-	adapter->stats.uptch += IXGB_READ_REG(&adapter->hw, UPTCH);
-	adapter->stats.vptcl += IXGB_READ_REG(&adapter->hw, VPTCL);
-	adapter->stats.vptch += IXGB_READ_REG(&adapter->hw, VPTCH);
-	adapter->stats.jptcl += IXGB_READ_REG(&adapter->hw, JPTCL);
-	adapter->stats.jptch += IXGB_READ_REG(&adapter->hw, JPTCH);
-	adapter->stats.gotcl += IXGB_READ_REG(&adapter->hw, GOTCL);
-	adapter->stats.gotch += IXGB_READ_REG(&adapter->hw, GOTCH);
-	adapter->stats.totl += IXGB_READ_REG(&adapter->hw, TOTL);
-	adapter->stats.toth += IXGB_READ_REG(&adapter->hw, TOTH);
-	adapter->stats.dc += IXGB_READ_REG(&adapter->hw, DC);
-	adapter->stats.plt64c += IXGB_READ_REG(&adapter->hw, PLT64C);
-	adapter->stats.tsctc += IXGB_READ_REG(&adapter->hw, TSCTC);
-	adapter->stats.tsctfc += IXGB_READ_REG(&adapter->hw, TSCTFC);
-	adapter->stats.ibic += IXGB_READ_REG(&adapter->hw, IBIC);
-	adapter->stats.rfc += IXGB_READ_REG(&adapter->hw, RFC);
-	adapter->stats.lfc += IXGB_READ_REG(&adapter->hw, LFC);
-	adapter->stats.pfrc += IXGB_READ_REG(&adapter->hw, PFRC);
-	adapter->stats.pftc += IXGB_READ_REG(&adapter->hw, PFTC);
-	adapter->stats.mcfrc += IXGB_READ_REG(&adapter->hw, MCFRC);
-	adapter->stats.mcftc += IXGB_READ_REG(&adapter->hw, MCFTC);
-	adapter->stats.xonrxc += IXGB_READ_REG(&adapter->hw, XONRXC);
-	adapter->stats.xontxc += IXGB_READ_REG(&adapter->hw, XONTXC);
-	adapter->stats.xoffrxc += IXGB_READ_REG(&adapter->hw, XOFFRXC);
-	adapter->stats.xofftxc += IXGB_READ_REG(&adapter->hw, XOFFTXC);
-	adapter->stats.rjc += IXGB_READ_REG(&adapter->hw, RJC);
-
-	/* Fill out the OS statistics structure */
-
-	netdev->stats.rx_packets = adapter->stats.gprcl;
-	netdev->stats.tx_packets = adapter->stats.gptcl;
-	netdev->stats.rx_bytes = adapter->stats.gorcl;
-	netdev->stats.tx_bytes = adapter->stats.gotcl;
-	netdev->stats.multicast = adapter->stats.mprcl;
-	netdev->stats.collisions = 0;
-
-	/* ignore RLEC as it reports errors for padded (<64bytes) frames
-	 * with a length in the type/len field */
-	netdev->stats.rx_errors =
-	    /* adapter->stats.rnbc + */ adapter->stats.crcerrs +
-	    adapter->stats.ruc +
-	    adapter->stats.roc /*+ adapter->stats.rlec */  +
-	    adapter->stats.icbc +
-	    adapter->stats.ecbc + adapter->stats.mpc;
-
-	/* see above
-	 * netdev->stats.rx_length_errors = adapter->stats.rlec;
-	 */
-
-	netdev->stats.rx_crc_errors = adapter->stats.crcerrs;
-	netdev->stats.rx_fifo_errors = adapter->stats.mpc;
-	netdev->stats.rx_missed_errors = adapter->stats.mpc;
-	netdev->stats.rx_over_errors = adapter->stats.mpc;
-
-	netdev->stats.tx_errors = 0;
-	netdev->stats.rx_frame_errors = 0;
-	netdev->stats.tx_aborted_errors = 0;
-	netdev->stats.tx_carrier_errors = 0;
-	netdev->stats.tx_fifo_errors = 0;
-	netdev->stats.tx_heartbeat_errors = 0;
-	netdev->stats.tx_window_errors = 0;
-}
-
-/**
- * ixgb_intr - Interrupt Handler
- * @irq: interrupt number
- * @data: pointer to a network interface device structure
- **/
-
-static irqreturn_t
-ixgb_intr(int irq, void *data)
-{
-	struct net_device *netdev = data;
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	struct ixgb_hw *hw = &adapter->hw;
-	u32 icr = IXGB_READ_REG(hw, ICR);
-
-	if (unlikely(!icr))
-		return IRQ_NONE;  /* Not our interrupt */
-
-	if (unlikely(icr & (IXGB_INT_RXSEQ | IXGB_INT_LSC)))
-		if (!test_bit(__IXGB_DOWN, &adapter->flags))
-			mod_timer(&adapter->watchdog_timer, jiffies);
-
-	if (napi_schedule_prep(&adapter->napi)) {
-
-		/* Disable interrupts and register for poll. The flush
-		  of the posted write is intentionally left out.
-		*/
-
-		IXGB_WRITE_REG(&adapter->hw, IMC, ~0);
-		__napi_schedule(&adapter->napi);
-	}
-	return IRQ_HANDLED;
-}
-
-/**
- * ixgb_clean - NAPI Rx polling callback
- * @napi: napi struct pointer
- * @budget: max number of receives to clean
- **/
-
-static int
-ixgb_clean(struct napi_struct *napi, int budget)
-{
-	struct ixgb_adapter *adapter = container_of(napi, struct ixgb_adapter, napi);
-	int work_done = 0;
-
-	ixgb_clean_tx_irq(adapter);
-	ixgb_clean_rx_irq(adapter, &work_done, budget);
-
-	/* If budget not fully consumed, exit the polling mode */
-	if (work_done < budget) {
-		napi_complete_done(napi, work_done);
-		if (!test_bit(__IXGB_DOWN, &adapter->flags))
-			ixgb_irq_enable(adapter);
-	}
-
-	return work_done;
-}
-
-/**
- * ixgb_clean_tx_irq - Reclaim resources after transmit completes
- * @adapter: board private structure
- **/
-
-static bool
-ixgb_clean_tx_irq(struct ixgb_adapter *adapter)
-{
-	struct ixgb_desc_ring *tx_ring = &adapter->tx_ring;
-	struct net_device *netdev = adapter->netdev;
-	struct ixgb_tx_desc *tx_desc, *eop_desc;
-	struct ixgb_buffer *buffer_info;
-	unsigned int i, eop;
-	bool cleaned = false;
-
-	i = tx_ring->next_to_clean;
-	eop = tx_ring->buffer_info[i].next_to_watch;
-	eop_desc = IXGB_TX_DESC(*tx_ring, eop);
-
-	while (eop_desc->status & IXGB_TX_DESC_STATUS_DD) {
-
-		rmb(); /* read buffer_info after eop_desc */
-		for (cleaned = false; !cleaned; ) {
-			tx_desc = IXGB_TX_DESC(*tx_ring, i);
-			buffer_info = &tx_ring->buffer_info[i];
-
-			if (tx_desc->popts &
-			   (IXGB_TX_DESC_POPTS_TXSM |
-			    IXGB_TX_DESC_POPTS_IXSM))
-				adapter->hw_csum_tx_good++;
-
-			ixgb_unmap_and_free_tx_resource(adapter, buffer_info);
-
-			*(u32 *)&(tx_desc->status) = 0;
-
-			cleaned = (i == eop);
-			if (++i == tx_ring->count) i = 0;
-		}
-
-		eop = tx_ring->buffer_info[i].next_to_watch;
-		eop_desc = IXGB_TX_DESC(*tx_ring, eop);
-	}
-
-	tx_ring->next_to_clean = i;
-
-	if (unlikely(cleaned && netif_carrier_ok(netdev) &&
-		     IXGB_DESC_UNUSED(tx_ring) >= DESC_NEEDED)) {
-		/* Make sure that anybody stopping the queue after this
-		 * sees the new next_to_clean. */
-		smp_mb();
-
-		if (netif_queue_stopped(netdev) &&
-		    !(test_bit(__IXGB_DOWN, &adapter->flags))) {
-			netif_wake_queue(netdev);
-			++adapter->restart_queue;
-		}
-	}
-
-	if (adapter->detect_tx_hung) {
-		/* detect a transmit hang in hardware, this serializes the
-		 * check with the clearing of time_stamp and movement of i */
-		adapter->detect_tx_hung = false;
-		if (tx_ring->buffer_info[eop].time_stamp &&
-		   time_after(jiffies, tx_ring->buffer_info[eop].time_stamp + HZ)
-		   && !(IXGB_READ_REG(&adapter->hw, STATUS) &
-		        IXGB_STATUS_TXOFF)) {
-			/* detected Tx unit hang */
-			netif_err(adapter, drv, adapter->netdev,
-				  "Detected Tx Unit Hang\n"
-				  "  TDH                  <%x>\n"
-				  "  TDT                  <%x>\n"
-				  "  next_to_use          <%x>\n"
-				  "  next_to_clean        <%x>\n"
-				  "buffer_info[next_to_clean]\n"
-				  "  time_stamp           <%lx>\n"
-				  "  next_to_watch        <%x>\n"
-				  "  jiffies              <%lx>\n"
-				  "  next_to_watch.status <%x>\n",
-				  IXGB_READ_REG(&adapter->hw, TDH),
-				  IXGB_READ_REG(&adapter->hw, TDT),
-				  tx_ring->next_to_use,
-				  tx_ring->next_to_clean,
-				  tx_ring->buffer_info[eop].time_stamp,
-				  eop,
-				  jiffies,
-				  eop_desc->status);
-			netif_stop_queue(netdev);
-		}
-	}
-
-	return cleaned;
-}
-
-/**
- * ixgb_rx_checksum - Receive Checksum Offload for 82597.
- * @adapter: board private structure
- * @rx_desc: receive descriptor
- * @skb: socket buffer with received data
- **/
-
-static void
-ixgb_rx_checksum(struct ixgb_adapter *adapter,
-                 struct ixgb_rx_desc *rx_desc,
-                 struct sk_buff *skb)
-{
-	/* Ignore Checksum bit is set OR
-	 * TCP Checksum has not been calculated
-	 */
-	if ((rx_desc->status & IXGB_RX_DESC_STATUS_IXSM) ||
-	   (!(rx_desc->status & IXGB_RX_DESC_STATUS_TCPCS))) {
-		skb_checksum_none_assert(skb);
-		return;
-	}
-
-	/* At this point we know the hardware did the TCP checksum */
-	/* now look at the TCP checksum error bit */
-	if (rx_desc->errors & IXGB_RX_DESC_ERRORS_TCPE) {
-		/* let the stack verify checksum errors */
-		skb_checksum_none_assert(skb);
-		adapter->hw_csum_rx_error++;
-	} else {
-		/* TCP checksum is good */
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-		adapter->hw_csum_rx_good++;
-	}
-}
-
-/*
- * this should improve performance for small packets with large amounts
- * of reassembly being done in the stack
- */
-static void ixgb_check_copybreak(struct napi_struct *napi,
-				 struct ixgb_buffer *buffer_info,
-				 u32 length, struct sk_buff **skb)
-{
-	struct sk_buff *new_skb;
-
-	if (length > copybreak)
-		return;
-
-	new_skb = napi_alloc_skb(napi, length);
-	if (!new_skb)
-		return;
-
-	skb_copy_to_linear_data_offset(new_skb, -NET_IP_ALIGN,
-				       (*skb)->data - NET_IP_ALIGN,
-				       length + NET_IP_ALIGN);
-	/* save the skb in buffer_info as good */
-	buffer_info->skb = *skb;
-	*skb = new_skb;
-}
-
-/**
- * ixgb_clean_rx_irq - Send received data up the network stack,
- * @adapter: board private structure
- * @work_done: output pointer to amount of packets cleaned
- * @work_to_do: how much work we can complete
- **/
-
-static bool
-ixgb_clean_rx_irq(struct ixgb_adapter *adapter, int *work_done, int work_to_do)
-{
-	struct ixgb_desc_ring *rx_ring = &adapter->rx_ring;
-	struct net_device *netdev = adapter->netdev;
-	struct pci_dev *pdev = adapter->pdev;
-	struct ixgb_rx_desc *rx_desc, *next_rxd;
-	struct ixgb_buffer *buffer_info, *next_buffer, *next2_buffer;
-	u32 length;
-	unsigned int i, j;
-	int cleaned_count = 0;
-	bool cleaned = false;
-
-	i = rx_ring->next_to_clean;
-	rx_desc = IXGB_RX_DESC(*rx_ring, i);
-	buffer_info = &rx_ring->buffer_info[i];
-
-	while (rx_desc->status & IXGB_RX_DESC_STATUS_DD) {
-		struct sk_buff *skb;
-		u8 status;
-
-		if (*work_done >= work_to_do)
-			break;
-
-		(*work_done)++;
-		rmb();	/* read descriptor and rx_buffer_info after status DD */
-		status = rx_desc->status;
-		skb = buffer_info->skb;
-		buffer_info->skb = NULL;
-
-		prefetch(skb->data - NET_IP_ALIGN);
-
-		if (++i == rx_ring->count)
-			i = 0;
-		next_rxd = IXGB_RX_DESC(*rx_ring, i);
-		prefetch(next_rxd);
-
-		j = i + 1;
-		if (j == rx_ring->count)
-			j = 0;
-		next2_buffer = &rx_ring->buffer_info[j];
-		prefetch(next2_buffer);
-
-		next_buffer = &rx_ring->buffer_info[i];
-
-		cleaned = true;
-		cleaned_count++;
-
-		dma_unmap_single(&pdev->dev,
-				 buffer_info->dma,
-				 buffer_info->length,
-				 DMA_FROM_DEVICE);
-		buffer_info->dma = 0;
-
-		length = le16_to_cpu(rx_desc->length);
-		rx_desc->length = 0;
-
-		if (unlikely(!(status & IXGB_RX_DESC_STATUS_EOP))) {
-
-			/* All receives must fit into a single buffer */
-
-			pr_debug("Receive packet consumed multiple buffers length<%x>\n",
-				 length);
-
-			dev_kfree_skb_irq(skb);
-			goto rxdesc_done;
-		}
-
-		if (unlikely(rx_desc->errors &
-		    (IXGB_RX_DESC_ERRORS_CE | IXGB_RX_DESC_ERRORS_SE |
-		     IXGB_RX_DESC_ERRORS_P | IXGB_RX_DESC_ERRORS_RXE))) {
-			dev_kfree_skb_irq(skb);
-			goto rxdesc_done;
-		}
-
-		ixgb_check_copybreak(&adapter->napi, buffer_info, length, &skb);
-
-		/* Good Receive */
-		skb_put(skb, length);
-
-		/* Receive Checksum Offload */
-		ixgb_rx_checksum(adapter, rx_desc, skb);
-
-		skb->protocol = eth_type_trans(skb, netdev);
-		if (status & IXGB_RX_DESC_STATUS_VP)
-			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
-				       le16_to_cpu(rx_desc->special));
-
-		netif_receive_skb(skb);
-
-rxdesc_done:
-		/* clean up descriptor, might be written over by hw */
-		rx_desc->status = 0;
-
-		/* return some buffers to hardware, one at a time is too slow */
-		if (unlikely(cleaned_count >= IXGB_RX_BUFFER_WRITE)) {
-			ixgb_alloc_rx_buffers(adapter, cleaned_count);
-			cleaned_count = 0;
-		}
-
-		/* use prefetched values */
-		rx_desc = next_rxd;
-		buffer_info = next_buffer;
-	}
-
-	rx_ring->next_to_clean = i;
-
-	cleaned_count = IXGB_DESC_UNUSED(rx_ring);
-	if (cleaned_count)
-		ixgb_alloc_rx_buffers(adapter, cleaned_count);
-
-	return cleaned;
-}
-
-/**
- * ixgb_alloc_rx_buffers - Replace used receive buffers
- * @adapter: address of board private structure
- * @cleaned_count: how many buffers to allocate
- **/
-
-static void
-ixgb_alloc_rx_buffers(struct ixgb_adapter *adapter, int cleaned_count)
-{
-	struct ixgb_desc_ring *rx_ring = &adapter->rx_ring;
-	struct net_device *netdev = adapter->netdev;
-	struct pci_dev *pdev = adapter->pdev;
-	struct ixgb_rx_desc *rx_desc;
-	struct ixgb_buffer *buffer_info;
-	struct sk_buff *skb;
-	unsigned int i;
-	long cleancount;
-
-	i = rx_ring->next_to_use;
-	buffer_info = &rx_ring->buffer_info[i];
-	cleancount = IXGB_DESC_UNUSED(rx_ring);
-
-
-	/* leave three descriptors unused */
-	while (--cleancount > 2 && cleaned_count--) {
-		/* recycle! its good for you */
-		skb = buffer_info->skb;
-		if (skb) {
-			skb_trim(skb, 0);
-			goto map_skb;
-		}
-
-		skb = netdev_alloc_skb_ip_align(netdev, adapter->rx_buffer_len);
-		if (unlikely(!skb)) {
-			/* Better luck next round */
-			adapter->alloc_rx_buff_failed++;
-			break;
-		}
-
-		buffer_info->skb = skb;
-		buffer_info->length = adapter->rx_buffer_len;
-map_skb:
-		buffer_info->dma = dma_map_single(&pdev->dev,
-		                                  skb->data,
-		                                  adapter->rx_buffer_len,
-						  DMA_FROM_DEVICE);
-		if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
-			adapter->alloc_rx_buff_failed++;
-			break;
-		}
-
-		rx_desc = IXGB_RX_DESC(*rx_ring, i);
-		rx_desc->buff_addr = cpu_to_le64(buffer_info->dma);
-		/* guarantee DD bit not set now before h/w gets descriptor
-		 * this is the rest of the workaround for h/w double
-		 * writeback. */
-		rx_desc->status = 0;
-
-
-		if (++i == rx_ring->count)
-			i = 0;
-		buffer_info = &rx_ring->buffer_info[i];
-	}
-
-	if (likely(rx_ring->next_to_use != i)) {
-		rx_ring->next_to_use = i;
-		if (unlikely(i-- == 0))
-			i = (rx_ring->count - 1);
-
-		/* Force memory writes to complete before letting h/w
-		 * know there are new descriptors to fetch.  (Only
-		 * applicable for weak-ordered memory model archs, such
-		 * as IA-64). */
-		wmb();
-		IXGB_WRITE_REG(&adapter->hw, RDT, i);
-	}
-}
-
-static void
-ixgb_vlan_strip_enable(struct ixgb_adapter *adapter)
-{
-	u32 ctrl;
-
-	/* enable VLAN tag insert/strip */
-	ctrl = IXGB_READ_REG(&adapter->hw, CTRL0);
-	ctrl |= IXGB_CTRL0_VME;
-	IXGB_WRITE_REG(&adapter->hw, CTRL0, ctrl);
-}
-
-static void
-ixgb_vlan_strip_disable(struct ixgb_adapter *adapter)
-{
-	u32 ctrl;
-
-	/* disable VLAN tag insert/strip */
-	ctrl = IXGB_READ_REG(&adapter->hw, CTRL0);
-	ctrl &= ~IXGB_CTRL0_VME;
-	IXGB_WRITE_REG(&adapter->hw, CTRL0, ctrl);
-}
-
-static int
-ixgb_vlan_rx_add_vid(struct net_device *netdev, __be16 proto, u16 vid)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	u32 vfta, index;
-
-	/* add VID to filter table */
-
-	index = (vid >> 5) & 0x7F;
-	vfta = IXGB_READ_REG_ARRAY(&adapter->hw, VFTA, index);
-	vfta |= (1 << (vid & 0x1F));
-	ixgb_write_vfta(&adapter->hw, index, vfta);
-	set_bit(vid, adapter->active_vlans);
-
-	return 0;
-}
-
-static int
-ixgb_vlan_rx_kill_vid(struct net_device *netdev, __be16 proto, u16 vid)
-{
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	u32 vfta, index;
-
-	/* remove VID from filter table */
-
-	index = (vid >> 5) & 0x7F;
-	vfta = IXGB_READ_REG_ARRAY(&adapter->hw, VFTA, index);
-	vfta &= ~(1 << (vid & 0x1F));
-	ixgb_write_vfta(&adapter->hw, index, vfta);
-	clear_bit(vid, adapter->active_vlans);
-
-	return 0;
-}
-
-static void
-ixgb_restore_vlan(struct ixgb_adapter *adapter)
-{
-	u16 vid;
-
-	for_each_set_bit(vid, adapter->active_vlans, VLAN_N_VID)
-		ixgb_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), vid);
-}
-
-/**
- * ixgb_io_error_detected - called when PCI error is detected
- * @pdev:    pointer to pci device with error
- * @state:   pci channel state after error
- *
- * This callback is called by the PCI subsystem whenever
- * a PCI bus error is detected.
- */
-static pci_ers_result_t ixgb_io_error_detected(struct pci_dev *pdev,
-                                               pci_channel_state_t state)
-{
-	struct net_device *netdev = pci_get_drvdata(pdev);
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	netif_device_detach(netdev);
-
-	if (state == pci_channel_io_perm_failure)
-		return PCI_ERS_RESULT_DISCONNECT;
-
-	if (netif_running(netdev))
-		ixgb_down(adapter, true);
-
-	pci_disable_device(pdev);
-
-	/* Request a slot reset. */
-	return PCI_ERS_RESULT_NEED_RESET;
-}
-
-/**
- * ixgb_io_slot_reset - called after the pci bus has been reset.
- * @pdev: pointer to pci device with error
- *
- * This callback is called after the PCI bus has been reset.
- * Basically, this tries to restart the card from scratch.
- * This is a shortened version of the device probe/discovery code,
- * it resembles the first-half of the ixgb_probe() routine.
- */
-static pci_ers_result_t ixgb_io_slot_reset(struct pci_dev *pdev)
-{
-	struct net_device *netdev = pci_get_drvdata(pdev);
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-	u8 addr[ETH_ALEN];
-
-	if (pci_enable_device(pdev)) {
-		netif_err(adapter, probe, adapter->netdev,
-			  "Cannot re-enable PCI device after reset\n");
-		return PCI_ERS_RESULT_DISCONNECT;
-	}
-
-	/* Perform card reset only on one instance of the card */
-	if (0 != PCI_FUNC (pdev->devfn))
-		return PCI_ERS_RESULT_RECOVERED;
-
-	pci_set_master(pdev);
-
-	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
-	ixgb_reset(adapter);
-
-	/* Make sure the EEPROM is good */
-	if (!ixgb_validate_eeprom_checksum(&adapter->hw)) {
-		netif_err(adapter, probe, adapter->netdev,
-			  "After reset, the EEPROM checksum is not valid\n");
-		return PCI_ERS_RESULT_DISCONNECT;
-	}
-	ixgb_get_ee_mac_addr(&adapter->hw, addr);
-	eth_hw_addr_set(netdev, addr);
-	memcpy(netdev->perm_addr, netdev->dev_addr, netdev->addr_len);
-
-	if (!is_valid_ether_addr(netdev->perm_addr)) {
-		netif_err(adapter, probe, adapter->netdev,
-			  "After reset, invalid MAC address\n");
-		return PCI_ERS_RESULT_DISCONNECT;
-	}
-
-	return PCI_ERS_RESULT_RECOVERED;
-}
-
-/**
- * ixgb_io_resume - called when its OK to resume normal operations
- * @pdev: pointer to pci device with error
- *
- * The error recovery driver tells us that its OK to resume
- * normal operation. Implementation resembles the second-half
- * of the ixgb_probe() routine.
- */
-static void ixgb_io_resume(struct pci_dev *pdev)
-{
-	struct net_device *netdev = pci_get_drvdata(pdev);
-	struct ixgb_adapter *adapter = netdev_priv(netdev);
-
-	pci_set_master(pdev);
-
-	if (netif_running(netdev)) {
-		if (ixgb_up(adapter)) {
-			pr_err("can't bring device back up after reset\n");
-			return;
-		}
-	}
-
-	netif_device_attach(netdev);
-	mod_timer(&adapter->watchdog_timer, jiffies);
-}
-
-/* ixgb_main.c */
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_osdep.h b/drivers/net/ethernet/intel/ixgb/ixgb_osdep.h
deleted file mode 100644
index 7bd54efa698d..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_osdep.h
+++ /dev/null
@@ -1,39 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-/* glue for the OS independent part of ixgb
- * includes register access macros
- */
-
-#ifndef _IXGB_OSDEP_H_
-#define _IXGB_OSDEP_H_
-
-#include <linux/types.h>
-#include <linux/delay.h>
-#include <asm/io.h>
-#include <linux/interrupt.h>
-#include <linux/sched.h>
-#include <linux/if_ether.h>
-
-#undef ASSERT
-#define ASSERT(x)	BUG_ON(!(x))
-
-#define ENTER() pr_debug("%s\n", __func__);
-
-#define IXGB_WRITE_REG(a, reg, value) ( \
-	writel((value), ((a)->hw_addr + IXGB_##reg)))
-
-#define IXGB_READ_REG(a, reg) ( \
-	readl((a)->hw_addr + IXGB_##reg))
-
-#define IXGB_WRITE_REG_ARRAY(a, reg, offset, value) ( \
-	writel((value), ((a)->hw_addr + IXGB_##reg + ((offset) << 2))))
-
-#define IXGB_READ_REG_ARRAY(a, reg, offset) ( \
-	readl((a)->hw_addr + IXGB_##reg + ((offset) << 2)))
-
-#define IXGB_WRITE_FLUSH(a) IXGB_READ_REG(a, STATUS)
-
-#define IXGB_MEMCPY memcpy
-
-#endif /* _IXGB_OSDEP_H_ */
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_param.c b/drivers/net/ethernet/intel/ixgb/ixgb_param.c
deleted file mode 100644
index d40f96250691..000000000000
--- a/drivers/net/ethernet/intel/ixgb/ixgb_param.c
+++ /dev/null
@@ -1,442 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Copyright(c) 1999 - 2008 Intel Corporation. */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include "ixgb.h"
-
-/* This is the only thing that needs to be changed to adjust the
- * maximum number of ports that the driver can manage.
- */
-
-#define IXGB_MAX_NIC 8
-
-#define OPTION_UNSET	-1
-#define OPTION_DISABLED 0
-#define OPTION_ENABLED  1
-
-/* All parameters are treated the same, as an integer array of values.
- * This macro just reduces the need to repeat the same declaration code
- * over and over (plus this helps to avoid typo bugs).
- */
-
-#define IXGB_PARAM_INIT { [0 ... IXGB_MAX_NIC] = OPTION_UNSET }
-#define IXGB_PARAM(X, desc)					\
-	static int X[IXGB_MAX_NIC+1]		\
-		= IXGB_PARAM_INIT;				\
-	static unsigned int num_##X = 0;			\
-	module_param_array_named(X, X, int, &num_##X, 0);	\
-	MODULE_PARM_DESC(X, desc);
-
-/* Transmit Descriptor Count
- *
- * Valid Range: 64-4096
- *
- * Default Value: 256
- */
-
-IXGB_PARAM(TxDescriptors, "Number of transmit descriptors");
-
-/* Receive Descriptor Count
- *
- * Valid Range: 64-4096
- *
- * Default Value: 1024
- */
-
-IXGB_PARAM(RxDescriptors, "Number of receive descriptors");
-
-/* User Specified Flow Control Override
- *
- * Valid Range: 0-3
- *  - 0 - No Flow Control
- *  - 1 - Rx only, respond to PAUSE frames but do not generate them
- *  - 2 - Tx only, generate PAUSE frames but ignore them on receive
- *  - 3 - Full Flow Control Support
- *
- * Default Value: 2 - Tx only (silicon bug avoidance)
- */
-
-IXGB_PARAM(FlowControl, "Flow Control setting");
-
-/* XsumRX - Receive Checksum Offload Enable/Disable
- *
- * Valid Range: 0, 1
- *  - 0 - disables all checksum offload
- *  - 1 - enables receive IP/TCP/UDP checksum offload
- *        on 82597 based NICs
- *
- * Default Value: 1
- */
-
-IXGB_PARAM(XsumRX, "Disable or enable Receive Checksum offload");
-
-/* Transmit Interrupt Delay in units of 0.8192 microseconds
- *
- * Valid Range: 0-65535
- *
- * Default Value: 32
- */
-
-IXGB_PARAM(TxIntDelay, "Transmit Interrupt Delay");
-
-/* Receive Interrupt Delay in units of 0.8192 microseconds
- *
- * Valid Range: 0-65535
- *
- * Default Value: 72
- */
-
-IXGB_PARAM(RxIntDelay, "Receive Interrupt Delay");
-
-/* Receive Flow control high threshold (when we send a pause frame)
- * (FCRTH)
- *
- * Valid Range: 1,536 - 262,136 (0x600 - 0x3FFF8, 8 byte granularity)
- *
- * Default Value: 196,608 (0x30000)
- */
-
-IXGB_PARAM(RxFCHighThresh, "Receive Flow Control High Threshold");
-
-/* Receive Flow control low threshold (when we send a resume frame)
- * (FCRTL)
- *
- * Valid Range: 64 - 262,136 (0x40 - 0x3FFF8, 8 byte granularity)
- *              must be less than high threshold by at least 8 bytes
- *
- * Default Value:  163,840 (0x28000)
- */
-
-IXGB_PARAM(RxFCLowThresh, "Receive Flow Control Low Threshold");
-
-/* Flow control request timeout (how long to pause the link partner's tx)
- * (PAP 15:0)
- *
- * Valid Range: 1 - 65535
- *
- * Default Value:  65535 (0xffff) (we'll send an xon if we recover)
- */
-
-IXGB_PARAM(FCReqTimeout, "Flow Control Request Timeout");
-
-/* Interrupt Delay Enable
- *
- * Valid Range: 0, 1
- *
- *  - 0 - disables transmit interrupt delay
- *  - 1 - enables transmmit interrupt delay
- *
- * Default Value: 1
- */
-
-IXGB_PARAM(IntDelayEnable, "Transmit Interrupt Delay Enable");
-
-
-#define DEFAULT_TIDV	   		     32
-#define MAX_TIDV			 0xFFFF
-#define MIN_TIDV			      0
-
-#define DEFAULT_RDTR		   	     72
-#define MAX_RDTR			 0xFFFF
-#define MIN_RDTR			      0
-
-#define DEFAULT_FCRTL	  		0x28000
-#define DEFAULT_FCRTH			0x30000
-#define MIN_FCRTL			      0
-#define MAX_FCRTL			0x3FFE8
-#define MIN_FCRTH			      8
-#define MAX_FCRTH			0x3FFF0
-
-#define MIN_FCPAUSE			      1
-#define MAX_FCPAUSE			 0xffff
-#define DEFAULT_FCPAUSE		  	 0xFFFF /* this may be too long */
-
-struct ixgb_option {
-	enum { enable_option, range_option, list_option } type;
-	const char *name;
-	const char *err;
-	int def;
-	union {
-		struct {	/* range_option info */
-			int min;
-			int max;
-		} r;
-		struct {	/* list_option info */
-			int nr;
-			const struct ixgb_opt_list {
-				int i;
-				const char *str;
-			} *p;
-		} l;
-	} arg;
-};
-
-static int
-ixgb_validate_option(unsigned int *value, const struct ixgb_option *opt)
-{
-	if (*value == OPTION_UNSET) {
-		*value = opt->def;
-		return 0;
-	}
-
-	switch (opt->type) {
-	case enable_option:
-		switch (*value) {
-		case OPTION_ENABLED:
-			pr_info("%s Enabled\n", opt->name);
-			return 0;
-		case OPTION_DISABLED:
-			pr_info("%s Disabled\n", opt->name);
-			return 0;
-		}
-		break;
-	case range_option:
-		if (*value >= opt->arg.r.min && *value <= opt->arg.r.max) {
-			pr_info("%s set to %i\n", opt->name, *value);
-			return 0;
-		}
-		break;
-	case list_option: {
-		int i;
-		const struct ixgb_opt_list *ent;
-
-		for (i = 0; i < opt->arg.l.nr; i++) {
-			ent = &opt->arg.l.p[i];
-			if (*value == ent->i) {
-				if (ent->str[0] != '\0')
-					pr_info("%s\n", ent->str);
-				return 0;
-			}
-		}
-	}
-		break;
-	default:
-		BUG();
-	}
-
-	pr_info("Invalid %s specified (%i) %s\n", opt->name, *value, opt->err);
-	*value = opt->def;
-	return -1;
-}
-
-/**
- * ixgb_check_options - Range Checking for Command Line Parameters
- * @adapter: board private structure
- *
- * This routine checks all command line parameters for valid user
- * input.  If an invalid value is given, or if no user specified
- * value exists, a default value is used.  The final value is stored
- * in a variable in the adapter structure.
- **/
-
-void
-ixgb_check_options(struct ixgb_adapter *adapter)
-{
-	int bd = adapter->bd_number;
-	if (bd >= IXGB_MAX_NIC) {
-		pr_notice("Warning: no configuration for board #%i\n", bd);
-		pr_notice("Using defaults for all values\n");
-	}
-
-	{ /* Transmit Descriptor Count */
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Transmit Descriptors",
-			.err  = "using default of " __MODULE_STRING(DEFAULT_TXD),
-			.def  = DEFAULT_TXD,
-			.arg  = { .r = { .min = MIN_TXD,
-					 .max = MAX_TXD}}
-		};
-		struct ixgb_desc_ring *tx_ring = &adapter->tx_ring;
-
-		if (num_TxDescriptors > bd) {
-			tx_ring->count = TxDescriptors[bd];
-			ixgb_validate_option(&tx_ring->count, &opt);
-		} else {
-			tx_ring->count = opt.def;
-		}
-		tx_ring->count = ALIGN(tx_ring->count, IXGB_REQ_TX_DESCRIPTOR_MULTIPLE);
-	}
-	{ /* Receive Descriptor Count */
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Receive Descriptors",
-			.err  = "using default of " __MODULE_STRING(DEFAULT_RXD),
-			.def  = DEFAULT_RXD,
-			.arg  = { .r = { .min = MIN_RXD,
-					 .max = MAX_RXD}}
-		};
-		struct ixgb_desc_ring *rx_ring = &adapter->rx_ring;
-
-		if (num_RxDescriptors > bd) {
-			rx_ring->count = RxDescriptors[bd];
-			ixgb_validate_option(&rx_ring->count, &opt);
-		} else {
-			rx_ring->count = opt.def;
-		}
-		rx_ring->count = ALIGN(rx_ring->count, IXGB_REQ_RX_DESCRIPTOR_MULTIPLE);
-	}
-	{ /* Receive Checksum Offload Enable */
-		static const struct ixgb_option opt = {
-			.type = enable_option,
-			.name = "Receive Checksum Offload",
-			.err  = "defaulting to Enabled",
-			.def  = OPTION_ENABLED
-		};
-
-		if (num_XsumRX > bd) {
-			unsigned int rx_csum = XsumRX[bd];
-			ixgb_validate_option(&rx_csum, &opt);
-			adapter->rx_csum = rx_csum;
-		} else {
-			adapter->rx_csum = opt.def;
-		}
-	}
-	{ /* Flow Control */
-
-		static const struct ixgb_opt_list fc_list[] = {
-		       { ixgb_fc_none, "Flow Control Disabled" },
-		       { ixgb_fc_rx_pause, "Flow Control Receive Only" },
-		       { ixgb_fc_tx_pause, "Flow Control Transmit Only" },
-		       { ixgb_fc_full, "Flow Control Enabled" },
-		       { ixgb_fc_default, "Flow Control Hardware Default" }
-		};
-
-		static const struct ixgb_option opt = {
-			.type = list_option,
-			.name = "Flow Control",
-			.err  = "reading default settings from EEPROM",
-			.def  = ixgb_fc_tx_pause,
-			.arg  = { .l = { .nr = ARRAY_SIZE(fc_list),
-					 .p = fc_list }}
-		};
-
-		if (num_FlowControl > bd) {
-			unsigned int fc = FlowControl[bd];
-			ixgb_validate_option(&fc, &opt);
-			adapter->hw.fc.type = fc;
-		} else {
-			adapter->hw.fc.type = opt.def;
-		}
-	}
-	{ /* Receive Flow Control High Threshold */
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Rx Flow Control High Threshold",
-			.err  = "using default of " __MODULE_STRING(DEFAULT_FCRTH),
-			.def  = DEFAULT_FCRTH,
-			.arg  = { .r = { .min = MIN_FCRTH,
-					 .max = MAX_FCRTH}}
-		};
-
-		if (num_RxFCHighThresh > bd) {
-			adapter->hw.fc.high_water = RxFCHighThresh[bd];
-			ixgb_validate_option(&adapter->hw.fc.high_water, &opt);
-		} else {
-			adapter->hw.fc.high_water = opt.def;
-		}
-		if (!(adapter->hw.fc.type & ixgb_fc_tx_pause) )
-			pr_info("Ignoring RxFCHighThresh when no RxFC\n");
-	}
-	{ /* Receive Flow Control Low Threshold */
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Rx Flow Control Low Threshold",
-			.err  = "using default of " __MODULE_STRING(DEFAULT_FCRTL),
-			.def  = DEFAULT_FCRTL,
-			.arg  = { .r = { .min = MIN_FCRTL,
-					 .max = MAX_FCRTL}}
-		};
-
-		if (num_RxFCLowThresh > bd) {
-			adapter->hw.fc.low_water = RxFCLowThresh[bd];
-			ixgb_validate_option(&adapter->hw.fc.low_water, &opt);
-		} else {
-			adapter->hw.fc.low_water = opt.def;
-		}
-		if (!(adapter->hw.fc.type & ixgb_fc_tx_pause) )
-			pr_info("Ignoring RxFCLowThresh when no RxFC\n");
-	}
-	{ /* Flow Control Pause Time Request*/
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Flow Control Pause Time Request",
-			.err  = "using default of "__MODULE_STRING(DEFAULT_FCPAUSE),
-			.def  = DEFAULT_FCPAUSE,
-			.arg = { .r = { .min = MIN_FCPAUSE,
-					.max = MAX_FCPAUSE}}
-		};
-
-		if (num_FCReqTimeout > bd) {
-			unsigned int pause_time = FCReqTimeout[bd];
-			ixgb_validate_option(&pause_time, &opt);
-			adapter->hw.fc.pause_time = pause_time;
-		} else {
-			adapter->hw.fc.pause_time = opt.def;
-		}
-		if (!(adapter->hw.fc.type & ixgb_fc_tx_pause) )
-			pr_info("Ignoring FCReqTimeout when no RxFC\n");
-	}
-	/* high low and spacing check for rx flow control thresholds */
-	if (adapter->hw.fc.type & ixgb_fc_tx_pause) {
-		/* high must be greater than low */
-		if (adapter->hw.fc.high_water < (adapter->hw.fc.low_water + 8)) {
-			/* set defaults */
-			pr_info("RxFCHighThresh must be >= (RxFCLowThresh + 8), Using Defaults\n");
-			adapter->hw.fc.high_water = DEFAULT_FCRTH;
-			adapter->hw.fc.low_water  = DEFAULT_FCRTL;
-		}
-	}
-	{ /* Receive Interrupt Delay */
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Receive Interrupt Delay",
-			.err  = "using default of " __MODULE_STRING(DEFAULT_RDTR),
-			.def  = DEFAULT_RDTR,
-			.arg  = { .r = { .min = MIN_RDTR,
-					 .max = MAX_RDTR}}
-		};
-
-		if (num_RxIntDelay > bd) {
-			adapter->rx_int_delay = RxIntDelay[bd];
-			ixgb_validate_option(&adapter->rx_int_delay, &opt);
-		} else {
-			adapter->rx_int_delay = opt.def;
-		}
-	}
-	{ /* Transmit Interrupt Delay */
-		static const struct ixgb_option opt = {
-			.type = range_option,
-			.name = "Transmit Interrupt Delay",
-			.err  = "using default of " __MODULE_STRING(DEFAULT_TIDV),
-			.def  = DEFAULT_TIDV,
-			.arg  = { .r = { .min = MIN_TIDV,
-					 .max = MAX_TIDV}}
-		};
-
-		if (num_TxIntDelay > bd) {
-			adapter->tx_int_delay = TxIntDelay[bd];
-			ixgb_validate_option(&adapter->tx_int_delay, &opt);
-		} else {
-			adapter->tx_int_delay = opt.def;
-		}
-	}
-
-	{ /* Transmit Interrupt Delay Enable */
-		static const struct ixgb_option opt = {
-			.type = enable_option,
-			.name = "Tx Interrupt Delay Enable",
-			.err  = "defaulting to Enabled",
-			.def  = OPTION_ENABLED
-		};
-
-		if (num_IntDelayEnable > bd) {
-			unsigned int ide = IntDelayEnable[bd];
-			ixgb_validate_option(&ide, &opt);
-			adapter->tx_int_delay_enable = ide;
-		} else {
-			adapter->tx_int_delay_enable = opt.def;
-		}
-	}
-}
-- 
cgit v1.2.3-70-g09d2


From 72abf2179969a54c16c03b1649b922ec179b364a Mon Sep 17 00:00:00 2001
From: Harini Katakam <harini.katakam@xilinx.com>
Date: Fri, 17 Mar 2023 17:09:43 +0530
Subject: net: macb: Reset TX when TX halt times out

Reset TX when halt times out i.e. disable TX, clean up TX BDs,
interrupts (already done) and enable TX.
This addresses the issue observed when iperf is run at 10Mps Half
duplex where, after multiple collisions and retries, TX halts.

Signed-off-by: Harini Katakam <harini.katakam@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/cadence/macb_main.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 7698719b1909..f77bd1223c8f 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -1070,6 +1070,7 @@ static void macb_tx_error_task(struct work_struct *work)
 {
 	struct macb_queue	*queue = container_of(work, struct macb_queue,
 						      tx_error_task);
+	bool			halt_timeout = false;
 	struct macb		*bp = queue->bp;
 	struct macb_tx_skb	*tx_skb;
 	struct macb_dma_desc	*desc;
@@ -1097,9 +1098,11 @@ static void macb_tx_error_task(struct work_struct *work)
 	 * (in case we have just queued new packets)
 	 * macb/gem must be halted to write TBQP register
 	 */
-	if (macb_halt_tx(bp))
-		/* Just complain for now, reinitializing TX path can be good */
+	if (macb_halt_tx(bp)) {
 		netdev_err(bp->dev, "BUG: halt tx timed out\n");
+		macb_writel(bp, NCR, macb_readl(bp, NCR) & (~MACB_BIT(TE)));
+		halt_timeout = true;
+	}
 
 	/* Treat frames in TX queue including the ones that caused the error.
 	 * Free transmit buffers in upper layer.
@@ -1170,6 +1173,9 @@ static void macb_tx_error_task(struct work_struct *work)
 	macb_writel(bp, TSR, macb_readl(bp, TSR));
 	queue_writel(queue, IER, MACB_TX_INT_FLAGS);
 
+	if (halt_timeout)
+		macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TE));
+
 	/* Now we are ready to start transmission again */
 	netif_tx_start_all_queues(bp->dev);
 	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(TSTART));
-- 
cgit v1.2.3-70-g09d2


From 105a201ebf3312990b96c4fbaade22e31402f8cc Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 17 Mar 2023 16:20:02 +0000
Subject: net/packet: remove po->xmit

Use PACKET_SOCK_QDISC_BYPASS atomic bit instead of a pointer.

This removes one indirect call in fast path,
and READ_ONCE()/WRITE_ONCE() annotations as well.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Suggested-by: Willem de Bruijn <willemb@google.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 24 +++++++++---------------
 net/packet/internal.h  |  2 +-
 2 files changed, 10 insertions(+), 16 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 7b9367b233d3..497193f73030 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -270,8 +270,11 @@ static noinline struct sk_buff *nf_hook_direct_egress(struct sk_buff *skb)
 }
 #endif
 
-static int packet_direct_xmit(struct sk_buff *skb)
+static int packet_xmit(const struct packet_sock *po, struct sk_buff *skb)
 {
+	if (!packet_sock_flag(po, PACKET_SOCK_QDISC_BYPASS))
+		return dev_queue_xmit(skb);
+
 #ifdef CONFIG_NETFILTER_EGRESS
 	if (nf_hook_egress_active()) {
 		skb = nf_hook_direct_egress(skb);
@@ -305,12 +308,6 @@ static void packet_cached_dev_reset(struct packet_sock *po)
 	RCU_INIT_POINTER(po->cached_dev, NULL);
 }
 
-static bool packet_use_direct_xmit(const struct packet_sock *po)
-{
-	/* Paired with WRITE_ONCE() in packet_setsockopt() */
-	return READ_ONCE(po->xmit) == packet_direct_xmit;
-}
-
 static u16 packet_pick_tx_queue(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
@@ -2872,8 +2869,7 @@ tpacket_error:
 		packet_inc_pending(&po->tx_ring);
 
 		status = TP_STATUS_SEND_REQUEST;
-		/* Paired with WRITE_ONCE() in packet_setsockopt() */
-		err = READ_ONCE(po->xmit)(skb);
+		err = packet_xmit(po, skb);
 		if (unlikely(err != 0)) {
 			if (err > 0)
 				err = net_xmit_errno(err);
@@ -3076,8 +3072,8 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 		virtio_net_hdr_set_proto(skb, &vnet_hdr);
 	}
 
-	/* Paired with WRITE_ONCE() in packet_setsockopt() */
-	err = READ_ONCE(po->xmit)(skb);
+	err = packet_xmit(po, skb);
+
 	if (unlikely(err != 0)) {
 		if (err > 0)
 			err = net_xmit_errno(err);
@@ -3359,7 +3355,6 @@ static int packet_create(struct net *net, struct socket *sock, int protocol,
 	init_completion(&po->skb_completion);
 	sk->sk_family = PF_PACKET;
 	po->num = proto;
-	po->xmit = dev_queue_xmit;
 
 	err = packet_alloc_pending(po);
 	if (err)
@@ -4010,8 +4005,7 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (copy_from_sockptr(&val, optval, sizeof(val)))
 			return -EFAULT;
 
-		/* Paired with all lockless reads of po->xmit */
-		WRITE_ONCE(po->xmit, val ? packet_direct_xmit : dev_queue_xmit);
+		packet_sock_flag_set(po, PACKET_SOCK_QDISC_BYPASS, val);
 		return 0;
 	}
 	default:
@@ -4126,7 +4120,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		val = packet_sock_flag(po, PACKET_SOCK_TX_HAS_OFF);
 		break;
 	case PACKET_QDISC_BYPASS:
-		val = packet_use_direct_xmit(po);
+		val = packet_sock_flag(po, PACKET_SOCK_QDISC_BYPASS);
 		break;
 	default:
 		return -ENOPROTOOPT;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index e793e99646f1..27930f69f368 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -128,7 +128,6 @@ struct packet_sock {
 	unsigned int		tp_tstamp;
 	struct completion	skb_completion;
 	struct net_device __rcu	*cached_dev;
-	int			(*xmit)(struct sk_buff *skb);
 	struct packet_type	prot_hook ____cacheline_aligned_in_smp;
 	atomic_t		tp_drops ____cacheline_aligned_in_smp;
 };
@@ -143,6 +142,7 @@ enum packet_sock_flags {
 	PACKET_SOCK_HAS_VNET_HDR,
 	PACKET_SOCK_RUNNING,
 	PACKET_SOCK_PRESSURE,
+	PACKET_SOCK_QDISC_BYPASS,
 };
 
 static inline void packet_sock_flag_set(struct packet_sock *po,
-- 
cgit v1.2.3-70-g09d2


From ff89ac704e2c92edfdb636070a7e0b91c1e4ec2d Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Fri, 17 Mar 2023 16:27:12 +0100
Subject: net: lan966x: Don't read RX timestamp if not needed

Whenever a frame was received to the CPU, the HW is timestamping the
frame. In the IFH(Inter Frame Header) it is found the nanosecond part
of the timestamps the SW is required to read from HW the second part.
But reading the second part it seems to be a expensive operations, so
so change this such to read the second part only when rx filter is
enabled.
Doing this change gives the RX a performance boost of ~70mbit.

before:
[  5]   0.00-10.01  sec   546 MBytes   457 Mbits/sec    0 sender

now:
[  5]   0.00-10.01  sec   652 MBytes   530 Mbits/sec    0 sender

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/microchip/lan966x/lan966x_fdma.c    |  2 +-
 .../net/ethernet/microchip/lan966x/lan966x_main.c    |  2 +-
 .../net/ethernet/microchip/lan966x/lan966x_main.h    |  5 +++--
 drivers/net/ethernet/microchip/lan966x/lan966x_ptp.c | 20 ++++++++++----------
 4 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c b/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c
index 55b484b10562..2ed76bb61a73 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c
@@ -517,7 +517,7 @@ static struct sk_buff *lan966x_fdma_rx_get_frame(struct lan966x_rx *rx,
 	if (likely(!(skb->dev->features & NETIF_F_RXFCS)))
 		skb_trim(skb, skb->len - ETH_FCS_LEN);
 
-	lan966x_ptp_rxtstamp(lan966x, skb, timestamp);
+	lan966x_ptp_rxtstamp(lan966x, skb, src_port, timestamp);
 	skb->protocol = eth_type_trans(skb, skb->dev);
 
 	if (lan966x->bridge_mask & BIT(src_port)) {
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
index 685e8cd7658c..4584a78c6ecb 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
@@ -668,7 +668,7 @@ static irqreturn_t lan966x_xtr_irq_handler(int irq, void *args)
 			*buf = val;
 		}
 
-		lan966x_ptp_rxtstamp(lan966x, skb, timestamp);
+		lan966x_ptp_rxtstamp(lan966x, skb, src_port, timestamp);
 		skb->protocol = eth_type_trans(skb, dev);
 
 		if (lan966x->bridge_mask & BIT(src_port)) {
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
index cbdae0ab8bb6..851afb0166b1 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
@@ -407,7 +407,8 @@ struct lan966x_port {
 	struct phy *serdes;
 	struct fwnode_handle *fwnode;
 
-	u8 ptp_cmd;
+	u8 ptp_tx_cmd;
+	bool ptp_rx_cmd;
 	u16 ts_id;
 	struct sk_buff_head tx_skbs;
 
@@ -527,7 +528,7 @@ void lan966x_ptp_deinit(struct lan966x *lan966x);
 int lan966x_ptp_hwtstamp_set(struct lan966x_port *port, struct ifreq *ifr);
 int lan966x_ptp_hwtstamp_get(struct lan966x_port *port, struct ifreq *ifr);
 void lan966x_ptp_rxtstamp(struct lan966x *lan966x, struct sk_buff *skb,
-			  u64 timestamp);
+			  u64 src_port, u64 timestamp);
 int lan966x_ptp_txtstamp_request(struct lan966x_port *port,
 				 struct sk_buff *skb);
 void lan966x_ptp_txtstamp_release(struct lan966x_port *port,
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_ptp.c b/drivers/net/ethernet/microchip/lan966x/lan966x_ptp.c
index 931e37b9a0ad..266a21a2d124 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_ptp.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_ptp.c
@@ -272,13 +272,13 @@ int lan966x_ptp_hwtstamp_set(struct lan966x_port *port, struct ifreq *ifr)
 
 	switch (cfg.tx_type) {
 	case HWTSTAMP_TX_ON:
-		port->ptp_cmd = IFH_REW_OP_TWO_STEP_PTP;
+		port->ptp_tx_cmd = IFH_REW_OP_TWO_STEP_PTP;
 		break;
 	case HWTSTAMP_TX_ONESTEP_SYNC:
-		port->ptp_cmd = IFH_REW_OP_ONE_STEP_PTP;
+		port->ptp_tx_cmd = IFH_REW_OP_ONE_STEP_PTP;
 		break;
 	case HWTSTAMP_TX_OFF:
-		port->ptp_cmd = IFH_REW_OP_NOOP;
+		port->ptp_tx_cmd = IFH_REW_OP_NOOP;
 		break;
 	default:
 		return -ERANGE;
@@ -286,6 +286,7 @@ int lan966x_ptp_hwtstamp_set(struct lan966x_port *port, struct ifreq *ifr)
 
 	switch (cfg.rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
+		port->ptp_rx_cmd = false;
 		break;
 	case HWTSTAMP_FILTER_ALL:
 	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
@@ -301,6 +302,7 @@ int lan966x_ptp_hwtstamp_set(struct lan966x_port *port, struct ifreq *ifr)
 	case HWTSTAMP_FILTER_PTP_V2_SYNC:
 	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
 	case HWTSTAMP_FILTER_NTP_ALL:
+		port->ptp_rx_cmd = true;
 		cfg.rx_filter = HWTSTAMP_FILTER_ALL;
 		break;
 	default:
@@ -332,7 +334,7 @@ static int lan966x_ptp_classify(struct lan966x_port *port, struct sk_buff *skb)
 	u8 msgtype;
 	int type;
 
-	if (port->ptp_cmd == IFH_REW_OP_NOOP)
+	if (port->ptp_tx_cmd == IFH_REW_OP_NOOP)
 		return IFH_REW_OP_NOOP;
 
 	type = ptp_classify_raw(skb);
@@ -343,7 +345,7 @@ static int lan966x_ptp_classify(struct lan966x_port *port, struct sk_buff *skb)
 	if (!header)
 		return IFH_REW_OP_NOOP;
 
-	if (port->ptp_cmd == IFH_REW_OP_TWO_STEP_PTP)
+	if (port->ptp_tx_cmd == IFH_REW_OP_TWO_STEP_PTP)
 		return IFH_REW_OP_TWO_STEP_PTP;
 
 	/* If it is sync and run 1 step then set the correct operation,
@@ -1009,9 +1011,6 @@ static int lan966x_ptp_phc_init(struct lan966x *lan966x,
 	phc->index = index;
 	phc->lan966x = lan966x;
 
-	/* PTP Rx stamping is always enabled.  */
-	phc->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
-
 	return 0;
 }
 
@@ -1088,14 +1087,15 @@ void lan966x_ptp_deinit(struct lan966x *lan966x)
 }
 
 void lan966x_ptp_rxtstamp(struct lan966x *lan966x, struct sk_buff *skb,
-			  u64 timestamp)
+			  u64 src_port, u64 timestamp)
 {
 	struct skb_shared_hwtstamps *shhwtstamps;
 	struct lan966x_phc *phc;
 	struct timespec64 ts;
 	u64 full_ts_in_ns;
 
-	if (!lan966x->ptp)
+	if (!lan966x->ptp ||
+	    !lan966x->ports[src_port]->ptp_rx_cmd)
 		return;
 
 	phc = &lan966x->phc[LAN966X_PHC_PORT];
-- 
cgit v1.2.3-70-g09d2


From fd7627833ddf1d845dcc0edb14bff97ddbb1272e Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Fri, 17 Mar 2023 16:27:13 +0100
Subject: net: lan966x: Stop using packing library

When a frame is injected from CPU, it is required to create an IFH(Inter
frame header) which sits in front of the frame that is transmitted.
This IFH, contains different fields like destination port, to bypass the
analyzer, priotity, etc. Lan966x it is using packing library to set and
get the fields of this IFH. But this seems to be an expensive
operations.
If this is changed with a simpler implementation, the RX will be
improved with ~5Mbit while on the TX is a much bigger improvement as it
is required to set more fields. Below are the numbers for TX.

Before:
[  5]   0.00-10.02  sec   439 MBytes   367 Mbits/sec    0 sender

After:
[  5]   0.00-10.00  sec   578 MBytes   485 Mbits/sec    0 sender

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microchip/lan966x/Kconfig     |  1 -
 .../net/ethernet/microchip/lan966x/lan966x_main.c  | 74 +++++++++++++++-------
 2 files changed, 50 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/Kconfig b/drivers/net/ethernet/microchip/lan966x/Kconfig
index 8bcd60f17d6d..571e6d4da1e9 100644
--- a/drivers/net/ethernet/microchip/lan966x/Kconfig
+++ b/drivers/net/ethernet/microchip/lan966x/Kconfig
@@ -6,7 +6,6 @@ config LAN966X_SWITCH
 	depends on NET_SWITCHDEV
 	depends on BRIDGE || BRIDGE=n
 	select PHYLINK
-	select PACKING
 	select PAGE_POOL
 	select VCAP
 	help
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
index 4584a78c6ecb..9be6462f1cc5 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
@@ -7,7 +7,6 @@
 #include <linux/ip.h>
 #include <linux/of_platform.h>
 #include <linux/of_net.h>
-#include <linux/packing.h>
 #include <linux/phy/phy.h>
 #include <linux/reset.h>
 #include <net/addrconf.h>
@@ -305,46 +304,57 @@ err:
 	return NETDEV_TX_BUSY;
 }
 
+static void lan966x_ifh_set(u8 *ifh, size_t val, size_t pos, size_t length)
+{
+	int i = 0;
+
+	do {
+		u8 p = IFH_LEN_BYTES - (pos + i) / 8 - 1;
+		u8 v = val >> i & 0xff;
+
+		/* There is no need to check for limits of the array, as these
+		 * will never be written
+		 */
+		ifh[p] |= v << ((pos + i) % 8);
+		ifh[p - 1] |= v >> (8 - (pos + i) % 8);
+
+		i += 8;
+	} while (i < length);
+}
+
 void lan966x_ifh_set_bypass(void *ifh, u64 bypass)
 {
-	packing(ifh, &bypass, IFH_POS_BYPASS + IFH_WID_BYPASS - 1,
-		IFH_POS_BYPASS, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, bypass, IFH_POS_BYPASS, IFH_WID_BYPASS);
 }
 
-void lan966x_ifh_set_port(void *ifh, u64 bypass)
+void lan966x_ifh_set_port(void *ifh, u64 port)
 {
-	packing(ifh, &bypass, IFH_POS_DSTS + IFH_WID_DSTS - 1,
-		IFH_POS_DSTS, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, port, IFH_POS_DSTS, IFH_WID_DSTS);
 }
 
-static void lan966x_ifh_set_qos_class(void *ifh, u64 bypass)
+static void lan966x_ifh_set_qos_class(void *ifh, u64 qos)
 {
-	packing(ifh, &bypass, IFH_POS_QOS_CLASS + IFH_WID_QOS_CLASS - 1,
-		IFH_POS_QOS_CLASS, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, qos, IFH_POS_QOS_CLASS, IFH_WID_QOS_CLASS);
 }
 
-static void lan966x_ifh_set_ipv(void *ifh, u64 bypass)
+static void lan966x_ifh_set_ipv(void *ifh, u64 ipv)
 {
-	packing(ifh, &bypass, IFH_POS_IPV + IFH_WID_IPV - 1,
-		IFH_POS_IPV, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, ipv, IFH_POS_IPV, IFH_WID_IPV);
 }
 
 static void lan966x_ifh_set_vid(void *ifh, u64 vid)
 {
-	packing(ifh, &vid, IFH_POS_TCI + IFH_WID_TCI - 1,
-		IFH_POS_TCI, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, vid, IFH_POS_TCI, IFH_WID_TCI);
 }
 
 static void lan966x_ifh_set_rew_op(void *ifh, u64 rew_op)
 {
-	packing(ifh, &rew_op, IFH_POS_REW_CMD + IFH_WID_REW_CMD - 1,
-		IFH_POS_REW_CMD, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, rew_op, IFH_POS_REW_CMD, IFH_WID_REW_CMD);
 }
 
 static void lan966x_ifh_set_timestamp(void *ifh, u64 timestamp)
 {
-	packing(ifh, &timestamp, IFH_POS_TIMESTAMP + IFH_WID_TIMESTAMP - 1,
-		IFH_POS_TIMESTAMP, IFH_LEN * 4, PACK, 0);
+	lan966x_ifh_set(ifh, timestamp, IFH_POS_TIMESTAMP, IFH_WID_TIMESTAMP);
 }
 
 static netdev_tx_t lan966x_port_xmit(struct sk_buff *skb,
@@ -582,22 +592,38 @@ static int lan966x_rx_frame_word(struct lan966x *lan966x, u8 grp, u32 *rval)
 	}
 }
 
+static u64 lan966x_ifh_get(u8 *ifh, size_t pos, size_t length)
+{
+	u64 val = 0;
+	u8 v;
+
+	for (int i = 0; i < length ; i++) {
+		int j = pos + i;
+		int k = j % 8;
+
+		if (i == 0 || k == 0)
+			v = ifh[IFH_LEN_BYTES - (j / 8) - 1];
+
+		if (v & (1 << k))
+			val |= (1 << i);
+	}
+
+	return val;
+}
+
 void lan966x_ifh_get_src_port(void *ifh, u64 *src_port)
 {
-	packing(ifh, src_port, IFH_POS_SRCPORT + IFH_WID_SRCPORT - 1,
-		IFH_POS_SRCPORT, IFH_LEN * 4, UNPACK, 0);
+	*src_port = lan966x_ifh_get(ifh, IFH_POS_SRCPORT, IFH_WID_SRCPORT);
 }
 
 static void lan966x_ifh_get_len(void *ifh, u64 *len)
 {
-	packing(ifh, len, IFH_POS_LEN + IFH_WID_LEN - 1,
-		IFH_POS_LEN, IFH_LEN * 4, UNPACK, 0);
+	*len = lan966x_ifh_get(ifh, IFH_POS_LEN, IFH_WID_LEN);
 }
 
 void lan966x_ifh_get_timestamp(void *ifh, u64 *timestamp)
 {
-	packing(ifh, timestamp, IFH_POS_TIMESTAMP + IFH_WID_TIMESTAMP - 1,
-		IFH_POS_TIMESTAMP, IFH_LEN * 4, UNPACK, 0);
+	*timestamp = lan966x_ifh_get(ifh, IFH_POS_TIMESTAMP, IFH_WID_TIMESTAMP);
 }
 
 static irqreturn_t lan966x_xtr_irq_handler(int irq, void *args)
-- 
cgit v1.2.3-70-g09d2


From d46e04ccd40457a0119b76e11ab64a2ad403e138 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 13 Mar 2023 15:42:59 +0200
Subject: wifi: rtl8xxxu: RTL8192EU always needs full init

Always run the entire init sequence (rtl8xxxu_init_device()) for
RTL8192EU. It's what the vendor driver does too.

This fixes a bug where the device is unable to connect after
rebooting:

wlp3s0f3u2: send auth to ... (try 1/3)
wlp3s0f3u2: send auth to ... (try 2/3)
wlp3s0f3u2: send auth to ... (try 3/3)
wlp3s0f3u2: authentication with ... timed out

Rebooting leaves the device powered on (partially? at least the
firmware is still running), but not really in a working state.

Cc: stable@vger.kernel.org
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Acked-by: Jes Sorensen <jes@trained-monkey.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/4eb111a9-d4c4-37d0-b376-4e202de7153c@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
index 5cfc00237f42..b2f902bfae70 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
@@ -1817,6 +1817,7 @@ struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.rx_desc_size = sizeof(struct rtl8xxxu_rxdesc24),
 	.has_s0s1 = 0,
 	.gen2_thermal_meter = 1,
+	.needs_full_init = 1,
 	.adda_1t_init = 0x0fc01616,
 	.adda_1t_path_on = 0x0fc01616,
 	.adda_2t_path_on_a = 0x0fc01616,
-- 
cgit v1.2.3-70-g09d2


From 70664495e3d24803ec47ccaccee9a822220558a0 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 13 Mar 2023 15:58:14 +0200
Subject: wifi: rtl8xxxu: Support new chip RTL8710BU aka RTL8188GU

This chip is found in cheap "free driver" USB adapters from Aliexpress.
Initially they pretend to be a CD-ROM containing the driver for Windows.
"Ejecting" switches the device to wifi mode.

Features: 2.4 GHz, b/g/n mode, 1T1R, 150 Mbps.

This chip is more unique than other Realtek chips:

* The registers at addresses 0x0-0xff, which all the other chips use,
  can't be used here. New registers at 0x8000-0x80ff must be used
  instead. And it's not a simple matter of adding 0x8000: 0x2
  (REG_SYS_FUNC) became 0x8004, 0x80 (REG_MCU_FW_DL) became 0x8090,
  etc.

* Also there are a few new registers which must be accessed indirectly
  because their addresses don't fit in 16 bits. No other chips seem to
  have these.

* The vendor driver compiles to 8188gu.ko, but the code calls the chip
  RTL8710B(U) pretty much everywhere, including messages visible to the
  user.

Another difference compared to the other chips supported by rtl8xxxu is
that it has a new PHY status struct, or three of them actually, from
which we extract the RSSI, among other things. This is not unique,
though, just new. The chips supported by rtw88 also use it.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Tested-by: Philipp Hortmann <philipp.g.hortmann@gmail.com> # Edimax N150
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/4edbe29f-00b9-8eef-9789-20bed0b141e2@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/Kconfig      |    2 +-
 drivers/net/wireless/realtek/rtl8xxxu/Makefile     |    2 +-
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h   |  313 +++-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c |    5 +-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c |    7 +-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c |    2 +
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c |    5 +-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c | 1878 ++++++++++++++++++++
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c |    5 +-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c |    5 +-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c  |  275 ++-
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h  |   44 +
 12 files changed, 2481 insertions(+), 62 deletions(-)
 create mode 100644 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/Kconfig b/drivers/net/wireless/realtek/rtl8xxxu/Kconfig
index 2eed20b0988c..82bcaf44a65f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/Kconfig
+++ b/drivers/net/wireless/realtek/rtl8xxxu/Kconfig
@@ -11,7 +11,7 @@ config RTL8XXXU
 	  parts written to utilize the Linux mac80211 stack.
 	  The driver is known to work with a number of RTL8723AU,
 	  RL8188CU, RTL8188RU, RTL8191CU, RTL8192CU, RTL8723BU, RTL8192EU,
-	  RTL8188FU, and RTL8188EU devices.
+	  RTL8188FU, RTL8188EU, and RTL8710BU (aka RTL8188GU) devices.
 
 	  This driver is under development and has a limited feature
 	  set. In particular it does not yet support 40MHz channels
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/Makefile b/drivers/net/wireless/realtek/rtl8xxxu/Makefile
index 0cb58fb30228..1bf083c15dcd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/Makefile
+++ b/drivers/net/wireless/realtek/rtl8xxxu/Makefile
@@ -3,4 +3,4 @@ obj-$(CONFIG_RTL8XXXU)	+= rtl8xxxu.o
 
 rtl8xxxu-y	:= rtl8xxxu_core.o rtl8xxxu_8192e.o rtl8xxxu_8723b.o \
 		   rtl8xxxu_8723a.o rtl8xxxu_8192c.o rtl8xxxu_8188f.o \
-		   rtl8xxxu_8188e.o
+		   rtl8xxxu_8188e.o rtl8xxxu_8710b.o
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index c8cee4a24755..9d48c69ffece 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -103,7 +103,8 @@ enum rtl8xxxu_rtl_chip {
 	RTL8822B = 0x8822b,
 	RTL8703B = 0x8703b,
 	RTL8195A = 0x8195a,
-	RTL8188F = 0x8188f
+	RTL8188F = 0x8188f,
+	RTL8710B = 0x8710b,
 };
 
 enum rtl8xxxu_rx_type {
@@ -618,6 +619,265 @@ struct rtl8723au_phy_stats {
 #endif
 };
 
+struct jaguar2_phy_stats_type0 {
+	/* DW0 */
+	u8		page_num;
+	u8		pwdb;
+#ifdef __LITTLE_ENDIAN
+	u8		gain: 6;
+	u8		rsvd_0: 1;
+	u8		trsw: 1;
+#else
+	u8		trsw: 1;
+	u8		rsvd_0: 1;
+	u8		gain: 6;
+#endif
+	u8		rsvd_1;
+
+	/* DW1 */
+	u8		rsvd_2;
+#ifdef __LITTLE_ENDIAN
+	u8		rxsc: 4;
+	u8		agc_table: 4;
+#else
+	u8		agc_table: 4;
+	u8		rxsc: 4;
+#endif
+	u8		channel;
+	u8		band;
+
+	/* DW2 */
+	u16		length;
+#ifdef __LITTLE_ENDIAN
+	u8		antidx_a: 3;
+	u8		antidx_b: 3;
+	u8		rsvd_3: 2;
+	u8		antidx_c: 3;
+	u8		antidx_d: 3;
+	u8		rsvd_4:2;
+#else
+	u8		rsvd_3: 2;
+	u8		antidx_b: 3;
+	u8		antidx_a: 3;
+	u8		rsvd_4:2;
+	u8		antidx_d: 3;
+	u8		antidx_c: 3;
+#endif
+
+	/* DW3 */
+	u8		signal_quality;
+#ifdef __LITTLE_ENDIAN
+	u8		vga:5;
+	u8		lna_l:3;
+	u8		bb_power:6;
+	u8		rsvd_9:1;
+	u8		lna_h:1;
+#else
+	u8		lna_l:3;
+	u8		vga:5;
+	u8		lna_h:1;
+	u8		rsvd_9:1;
+	u8		bb_power:6;
+#endif
+	u8		rsvd_5;
+
+	/* DW4 */
+	u32		rsvd_6;
+
+	/* DW5 */
+	u32		rsvd_7;
+
+	/* DW6 */
+	u32		rsvd_8;
+} __packed;
+
+struct jaguar2_phy_stats_type1 {
+	/* DW0 and DW1 */
+	u8		page_num;
+	u8		pwdb[4];
+#ifdef __LITTLE_ENDIAN
+	u8		l_rxsc: 4;
+	u8		ht_rxsc: 4;
+#else
+	u8		ht_rxsc: 4;
+	u8		l_rxsc: 4;
+#endif
+	u8		channel;
+#ifdef __LITTLE_ENDIAN
+	u8		band: 2;
+	u8		rsvd_0: 1;
+	u8		hw_antsw_occu: 1;
+	u8		gnt_bt: 1;
+	u8		ldpc: 1;
+	u8		stbc: 1;
+	u8		beamformed: 1;
+#else
+	u8		beamformed: 1;
+	u8		stbc: 1;
+	u8		ldpc: 1;
+	u8		gnt_bt: 1;
+	u8		hw_antsw_occu: 1;
+	u8		rsvd_0: 1;
+	u8		band: 2;
+#endif
+
+	/* DW2 */
+	u16		lsig_length;
+#ifdef __LITTLE_ENDIAN
+	u8		antidx_a: 3;
+	u8		antidx_b: 3;
+	u8		rsvd_1: 2;
+	u8		antidx_c: 3;
+	u8		antidx_d: 3;
+	u8		rsvd_2: 2;
+#else
+	u8		rsvd_1: 2;
+	u8		antidx_b: 3;
+	u8		antidx_a: 3;
+	u8		rsvd_2: 2;
+	u8		antidx_d: 3;
+	u8		antidx_c: 3;
+#endif
+
+	/* DW3 */
+	u8		paid;
+#ifdef __LITTLE_ENDIAN
+	u8		paid_msb: 1;
+	u8		gid: 6;
+	u8		rsvd_3: 1;
+#else
+	u8		rsvd_3: 1;
+	u8		gid: 6;
+	u8		paid_msb: 1;
+#endif
+	u8		intf_pos;
+#ifdef __LITTLE_ENDIAN
+	u8		intf_pos_msb: 1;
+	u8		rsvd_4: 2;
+	u8		nb_intf_flag: 1;
+	u8		rf_mode: 2;
+	u8		rsvd_5: 2;
+#else
+	u8		rsvd_5: 2;
+	u8		rf_mode: 2;
+	u8		nb_intf_flag: 1;
+	u8		rsvd_4: 2;
+	u8		intf_pos_msb: 1;
+#endif
+
+	/* DW4 */
+	s8		rxevm[4];			/* s(8,1) */
+
+	/* DW5 */
+	s8		cfo_tail[4];			/* s(8,7) */
+
+	/* DW6 */
+	s8		rxsnr[4];			/* s(8,1) */
+} __packed;
+
+struct jaguar2_phy_stats_type2 {
+	/* DW0 ane DW1 */
+	u8		page_num;
+	u8		pwdb[4];
+#ifdef __LITTLE_ENDIAN
+	u8		l_rxsc: 4;
+	u8		ht_rxsc: 4;
+#else
+	u8		ht_rxsc: 4;
+	u8		l_rxsc: 4;
+#endif
+	u8		channel;
+#ifdef __LITTLE_ENDIAN
+	u8		band: 2;
+	u8		rsvd_0: 1;
+	u8		hw_antsw_occu: 1;
+	u8		gnt_bt: 1;
+	u8		ldpc: 1;
+	u8		stbc: 1;
+	u8		beamformed: 1;
+#else
+	u8		beamformed: 1;
+	u8		stbc: 1;
+	u8		ldpc: 1;
+	u8		gnt_bt: 1;
+	u8		hw_antsw_occu: 1;
+	u8		rsvd_0: 1;
+	u8		band: 2;
+#endif
+
+	/* DW2 */
+#ifdef __LITTLE_ENDIAN
+	u8		shift_l_map: 6;
+	u8		rsvd_1: 2;
+#else
+	u8		rsvd_1: 2;
+	u8		shift_l_map: 6;
+#endif
+	u8		cnt_pw2cca;
+#ifdef __LITTLE_ENDIAN
+	u8		agc_table_a: 4;
+	u8		agc_table_b: 4;
+	u8		agc_table_c: 4;
+	u8		agc_table_d: 4;
+#else
+	u8		agc_table_b: 4;
+	u8		agc_table_a: 4;
+	u8		agc_table_d: 4;
+	u8		agc_table_c: 4;
+#endif
+
+	/* DW3 ~ DW6*/
+	u8		cnt_cca2agc_rdy;
+#ifdef __LITTLE_ENDIAN
+	u8		gain_a: 6;
+	u8		rsvd_2: 1;
+	u8		trsw_a: 1;
+	u8		gain_b: 6;
+	u8		rsvd_3: 1;
+	u8		trsw_b: 1;
+	u8		gain_c: 6;
+	u8		rsvd_4: 1;
+	u8		trsw_c: 1;
+	u8		gain_d: 6;
+	u8		rsvd_5: 1;
+	u8		trsw_d: 1;
+	u8		aagc_step_a: 2;
+	u8		aagc_step_b: 2;
+	u8		aagc_step_c: 2;
+	u8		aagc_step_d: 2;
+#else
+	u8		trsw_a: 1;
+	u8		rsvd_2: 1;
+	u8		gain_a: 6;
+	u8		trsw_b: 1;
+	u8		rsvd_3: 1;
+	u8		gain_b: 6;
+	u8		trsw_c: 1;
+	u8		rsvd_4: 1;
+	u8		gain_c: 6;
+	u8		trsw_d: 1;
+	u8		rsvd_5: 1;
+	u8		gain_d: 6;
+	u8		aagc_step_d: 2;
+	u8		aagc_step_c: 2;
+	u8		aagc_step_b: 2;
+	u8		aagc_step_a: 2;
+#endif
+	u8		ht_aagc_gain[4];
+	u8		dagc_gain[4];
+#ifdef __LITTLE_ENDIAN
+	u8		counter: 6;
+	u8		rsvd_6: 2;
+	u8		syn_count: 5;
+	u8		rsvd_7:3;
+#else
+	u8		rsvd_6: 2;
+	u8		counter: 6;
+	u8		rsvd_7:3;
+	u8		syn_count: 5;
+#endif
+} __packed;
+
 /*
  * Regs to backup
  */
@@ -963,6 +1223,29 @@ struct rtl8188eu_efuse {
 	u8 res12[0xc3];
 } __packed;
 
+struct rtl8710bu_efuse {
+	__le16 rtl_id;
+	u8 res0[0x1e];
+	struct rtl8188fu_efuse_tx_power tx_power_index_A;	/* 0x20 */
+	u8 res1[0x9c];			/* 0x2c */
+	u8 channel_plan;		/* 0xc8 */
+	u8 xtal_k;			/* 0xc9 */
+	u8 thermal_meter;		/* 0xca */
+	u8 res2[0x4f];
+	u8 mac_addr[ETH_ALEN];		/* 0x11a */
+	u8 res3[0x11];
+	u8 rf_board_option;		/* 0x131 */
+	u8 res4[2];
+	u8 eeprom_version;		/* 0x134 */
+	u8 eeprom_customer_id;		/* 0x135 */
+	u8 res5[5];
+	u8 country_code;		/* 0x13b */
+	u8 res6[0x84];
+	u8 vid[2];			/* 0x1c0 */
+	u8 pid[2];			/* 0x1c2 */
+	u8 res7[0x3c];
+} __packed;
+
 struct rtl8xxxu_reg8val {
 	u16 reg;
 	u8 val;
@@ -1486,6 +1769,7 @@ struct rtl8xxxu_priv {
 	struct rtl8723au_idx ht20_tx_power_diff[RTL8723B_TX_COUNT];
 	struct rtl8723au_idx ht40_tx_power_diff[RTL8723B_TX_COUNT];
 	struct rtl8xxxu_power_base *power_base;
+	u8 package_type;
 	u32 chip_cut:4;
 	u32 rom_rev:4;
 	u32 is_multi_func:1;
@@ -1505,6 +1789,7 @@ struct rtl8xxxu_priv {
 	u32 ep_tx_low_queue:1;
 	u32 rx_buf_aggregation:1;
 	u32 cck_agc_report_type:1;
+	u32 cck_new_agc:1;
 	u8 default_crystal_cap;
 	unsigned int pipe_interrupt;
 	unsigned int pipe_in;
@@ -1522,6 +1807,8 @@ struct rtl8xxxu_priv {
 	int nr_out_eps;
 
 	struct mutex h2c_mutex;
+	/* Protect the indirect register accesses of RTL8710BU. */
+	struct mutex syson_indirect_access_mutex;
 
 	struct usb_anchor rx_anchor;
 	struct usb_anchor tx_anchor;
@@ -1542,6 +1829,7 @@ struct rtl8xxxu_priv {
 		struct rtl8192eu_efuse efuse8192eu;
 		struct rtl8188fu_efuse efuse8188fu;
 		struct rtl8188eu_efuse efuse8188eu;
+		struct rtl8710bu_efuse efuse8710bu;
 	} efuse_wifi;
 	u32 adda_backup[RTL8XXXU_ADDA_REGS];
 	u32 mac_backup[RTL8XXXU_MAC_REGS];
@@ -1586,6 +1874,7 @@ struct rtl8xxxu_tx_urb {
 
 struct rtl8xxxu_fileops {
 	int (*identify_chip) (struct rtl8xxxu_priv *priv);
+	int (*read_efuse) (struct rtl8xxxu_priv *priv);
 	int (*parse_efuse) (struct rtl8xxxu_priv *priv);
 	int (*load_firmware) (struct rtl8xxxu_priv *priv);
 	int (*power_on) (struct rtl8xxxu_priv *priv);
@@ -1599,6 +1888,11 @@ struct rtl8xxxu_fileops {
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
 	void (*config_channel) (struct ieee80211_hw *hw);
 	int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb);
+	void (*parse_phystats) (struct rtl8xxxu_priv *priv,
+				struct ieee80211_rx_status *rx_status,
+				struct rtl8723au_phy_stats *phy_stats,
+				u32 rxmcs, struct ieee80211_hdr *hdr,
+				bool crc_icv_err);
 	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
 	void (*init_statistics) (struct rtl8xxxu_priv *priv);
 	void (*init_burst) (struct rtl8xxxu_priv *priv);
@@ -1618,7 +1912,7 @@ struct rtl8xxxu_fileops {
 			     bool short_preamble, bool ampdu_enable,
 			     u32 rts_rate);
 	void (*set_crystal_cap) (struct rtl8xxxu_priv *priv, u8 crystal_cap);
-	s8 (*cck_rssi) (struct rtl8xxxu_priv *priv, u8 cck_agc_rpt);
+	s8 (*cck_rssi) (struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats);
 	int (*led_classdev_brightness_set) (struct led_classdev *led_cdev,
 					    enum led_brightness brightness);
 	int writeN_block_size;
@@ -1687,10 +1981,12 @@ void rtl8xxxu_identify_vendor_2bits(struct rtl8xxxu_priv *priv, u32 vendor);
 void rtl8xxxu_config_endpoints_sie(struct rtl8xxxu_priv *priv);
 int rtl8xxxu_config_endpoints_no_sie(struct rtl8xxxu_priv *priv);
 int rtl8xxxu_read_efuse8(struct rtl8xxxu_priv *priv, u16 offset, u8 *data);
+int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv);
 void rtl8xxxu_reset_8051(struct rtl8xxxu_priv *priv);
 int rtl8xxxu_auto_llt_table(struct rtl8xxxu_priv *priv);
 void rtl8xxxu_gen2_prepare_calibrate(struct rtl8xxxu_priv *priv, u8 start);
 void rtl8723a_phy_lc_calibrate(struct rtl8xxxu_priv *priv);
+void rtl8188f_phy_lc_calibrate(struct rtl8xxxu_priv *priv);
 int rtl8xxxu_flush_fifo(struct rtl8xxxu_priv *priv);
 int rtl8xxxu_gen2_h2c_cmd(struct rtl8xxxu_priv *priv,
 			  struct h2c_cmd *h2c, int len);
@@ -1724,6 +2020,16 @@ void rtl8xxxu_gen2_disable_rf(struct rtl8xxxu_priv *priv);
 void rtl8xxxu_init_burst(struct rtl8xxxu_priv *priv);
 int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb);
 int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb);
+void rtl8723au_rx_parse_phystats(struct rtl8xxxu_priv *priv,
+				 struct ieee80211_rx_status *rx_status,
+				 struct rtl8723au_phy_stats *phy_stats,
+				 u32 rxmcs, struct ieee80211_hdr *hdr,
+				 bool crc_icv_err);
+void jaguar2_rx_parse_phystats(struct rtl8xxxu_priv *priv,
+			       struct ieee80211_rx_status *rx_status,
+			       struct rtl8723au_phy_stats *phy_stats,
+			       u32 rxmcs, struct ieee80211_hdr *hdr,
+			       bool crc_icv_err);
 int rtl8xxxu_gen2_channel_to_group(int channel);
 bool rtl8xxxu_simularity_compare(struct rtl8xxxu_priv *priv,
 				 int result[][8], int c1, int c2);
@@ -1749,12 +2055,13 @@ void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv,
 void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv);
 void rtl8723a_set_crystal_cap(struct rtl8xxxu_priv *priv, u8 crystal_cap);
 void rtl8188f_set_crystal_cap(struct rtl8xxxu_priv *priv, u8 crystal_cap);
-s8 rtl8723a_cck_rssi(struct rtl8xxxu_priv *priv, u8 cck_agc_rpt);
+s8 rtl8723a_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats);
 void rtl8xxxu_update_ra_report(struct rtl8xxxu_ra_report *rarpt,
 			       u8 rate, u8 sgi, u8 bw);
 void rtl8188e_ra_info_init_all(struct rtl8xxxu_ra_info *ra);
 void rtl8188e_handle_ra_tx_report2(struct rtl8xxxu_priv *priv, struct sk_buff *skb);
 
+extern struct rtl8xxxu_fileops rtl8710bu_fops;
 extern struct rtl8xxxu_fileops rtl8188fu_fops;
 extern struct rtl8xxxu_fileops rtl8188eu_fops;
 extern struct rtl8xxxu_fileops rtl8192cu_fops;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
index f15b099899e5..6a82ec47568e 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
@@ -1326,13 +1326,14 @@ static void rtl8188e_usb_quirks(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_EARLY_MODE_CONTROL_8188E + 3, 0x01);
 }
 
-static s8 rtl8188e_cck_rssi(struct rtl8xxxu_priv *priv, u8 cck_agc_rpt)
+static s8 rtl8188e_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats)
 {
 	/* only use lna 0/1/2/3/7 */
 	static const s8 lna_gain_table_0[8] = {17, -1, -13, -29, -32, -35, -38, -41};
 	/* only use lna 3/7 */
 	static const s8 lna_gain_table_1[8] = {29, 20, 12, 3, -6, -15, -24, -33};
 
+	u8 cck_agc_rpt = phy_stats->cck_agc_rpt_ofdm_cfosho_a;
 	s8 rx_pwr_all = 0x00;
 	u8 vga_idx, lna_idx;
 	s8 lna_gain = 0;
@@ -1856,6 +1857,7 @@ struct rtl8xxxu_fileops rtl8188eu_fops = {
 	.load_firmware = rtl8188eu_load_firmware,
 	.power_on = rtl8188eu_power_on,
 	.power_off = rtl8188eu_power_off,
+	.read_efuse = rtl8xxxu_read_efuse,
 	.reset_8051 = rtl8188eu_reset_8051,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.init_phy_bb = rtl8188eu_init_phy_bb,
@@ -1864,6 +1866,7 @@ struct rtl8xxxu_fileops rtl8188eu_fops = {
 	.phy_iq_calibrate = rtl8188eu_phy_iq_calibrate,
 	.config_channel = rtl8188eu_config_channel,
 	.parse_rx_desc = rtl8xxxu_parse_rxdesc16,
+	.parse_phystats = rtl8723au_rx_parse_phystats,
 	.init_aggregation = rtl8188eu_init_aggregation,
 	.enable_rf = rtl8188e_enable_rf,
 	.disable_rf = rtl8188e_disable_rf,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
index af6e2c8a5025..82dee1fed477 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
@@ -791,7 +791,7 @@ static int rtl8188fu_init_phy_rf(struct rtl8xxxu_priv *priv)
 	return ret;
 }
 
-static void rtl8188f_phy_lc_calibrate(struct rtl8xxxu_priv *priv)
+void rtl8188f_phy_lc_calibrate(struct rtl8xxxu_priv *priv)
 {
 	u32 val32;
 	u32 rf_amode, lstf;
@@ -1677,8 +1677,9 @@ void rtl8188f_set_crystal_cap(struct rtl8xxxu_priv *priv, u8 crystal_cap)
 	cfo->crystal_cap = crystal_cap;
 }
 
-static s8 rtl8188f_cck_rssi(struct rtl8xxxu_priv *priv, u8 cck_agc_rpt)
+static s8 rtl8188f_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats)
 {
+	u8 cck_agc_rpt = phy_stats->cck_agc_rpt_ofdm_cfosho_a;
 	s8 rx_pwr_all = 0x00;
 	u8 vga_idx, lna_idx;
 
@@ -1714,6 +1715,7 @@ struct rtl8xxxu_fileops rtl8188fu_fops = {
 	.load_firmware = rtl8188fu_load_firmware,
 	.power_on = rtl8188fu_power_on,
 	.power_off = rtl8188fu_power_off,
+	.read_efuse = rtl8xxxu_read_efuse,
 	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.init_phy_bb = rtl8188fu_init_phy_bb,
@@ -1723,6 +1725,7 @@ struct rtl8xxxu_fileops rtl8188fu_fops = {
 	.phy_iq_calibrate = rtl8188fu_phy_iq_calibrate,
 	.config_channel = rtl8188fu_config_channel,
 	.parse_rx_desc = rtl8xxxu_parse_rxdesc24,
+	.parse_phystats = rtl8723au_rx_parse_phystats,
 	.init_aggregation = rtl8188fu_init_aggregation,
 	.init_statistics = rtl8188fu_init_statistics,
 	.init_burst = rtl8xxxu_init_burst,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
index e61d65c3579b..caeba56241fc 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
@@ -594,6 +594,7 @@ struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.load_firmware = rtl8192cu_load_firmware,
 	.power_on = rtl8192cu_power_on,
 	.power_off = rtl8xxxu_power_off,
+	.read_efuse = rtl8xxxu_read_efuse,
 	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.init_phy_bb = rtl8xxxu_gen1_init_phy_bb,
@@ -602,6 +603,7 @@ struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.phy_iq_calibrate = rtl8xxxu_gen1_phy_iq_calibrate,
 	.config_channel = rtl8xxxu_gen1_config_channel,
 	.parse_rx_desc = rtl8xxxu_parse_rxdesc16,
+	.parse_phystats = rtl8723au_rx_parse_phystats,
 	.init_aggregation = rtl8xxxu_gen1_init_aggregation,
 	.enable_rf = rtl8xxxu_gen1_enable_rf,
 	.disable_rf = rtl8xxxu_gen1_disable_rf,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
index b2f902bfae70..4498748164af 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
@@ -1742,11 +1742,12 @@ static void rtl8192e_enable_rf(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_TXPAUSE, 0x00);
 }
 
-static s8 rtl8192e_cck_rssi(struct rtl8xxxu_priv *priv, u8 cck_agc_rpt)
+static s8 rtl8192e_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats)
 {
 	static const s8 lna_gain_table_0[8] = {15, 9, -10, -21, -23, -27, -43, -44};
 	static const s8 lna_gain_table_1[8] = {24, 18, 13, -4, -11, -18, -31, -36};
 
+	u8 cck_agc_rpt = phy_stats->cck_agc_rpt_ofdm_cfosho_a;
 	s8 rx_pwr_all = 0x00;
 	u8 vga_idx, lna_idx;
 	s8 lna_gain = 0;
@@ -1793,6 +1794,7 @@ struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.load_firmware = rtl8192eu_load_firmware,
 	.power_on = rtl8192eu_power_on,
 	.power_off = rtl8192eu_power_off,
+	.read_efuse = rtl8xxxu_read_efuse,
 	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.init_phy_bb = rtl8192eu_init_phy_bb,
@@ -1801,6 +1803,7 @@ struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.phy_iq_calibrate = rtl8192eu_phy_iq_calibrate,
 	.config_channel = rtl8xxxu_gen2_config_channel,
 	.parse_rx_desc = rtl8xxxu_parse_rxdesc24,
+	.parse_phystats = rtl8723au_rx_parse_phystats,
 	.enable_rf = rtl8192e_enable_rf,
 	.disable_rf = rtl8xxxu_gen2_disable_rf,
 	.usb_quirks = rtl8xxxu_gen2_usb_quirks,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c
new file mode 100644
index 000000000000..920466e39604
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c
@@ -0,0 +1,1878 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * RTL8XXXU mac80211 USB driver - 8710bu aka 8188gu specific subdriver
+ *
+ * Copyright (c) 2023 Bitterblue Smith <rtl8821cerfe2@gmail.com>
+ *
+ * Portions copied from existing rtl8xxxu code:
+ * Copyright (c) 2014 - 2017 Jes Sorensen <Jes.Sorensen@gmail.com>
+ *
+ * Portions, notably calibration code:
+ * Copyright(c) 2007 - 2011 Realtek Corporation. All rights reserved.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <linux/usb.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/wireless.h>
+#include <linux/firmware.h>
+#include <linux/moduleparam.h>
+#include <net/mac80211.h>
+#include "rtl8xxxu.h"
+#include "rtl8xxxu_regs.h"
+
+static const struct rtl8xxxu_reg8val rtl8710b_mac_init_table[] = {
+	{0x421, 0x0F}, {0x428, 0x0A}, {0x429, 0x10}, {0x430, 0x00},
+	{0x431, 0x00}, {0x432, 0x00}, {0x433, 0x01}, {0x434, 0x04},
+	{0x435, 0x05}, {0x436, 0x07}, {0x437, 0x08}, {0x43C, 0x04},
+	{0x43D, 0x05}, {0x43E, 0x07}, {0x43F, 0x08}, {0x440, 0x5D},
+	{0x441, 0x01}, {0x442, 0x00}, {0x444, 0x10}, {0x445, 0x00},
+	{0x446, 0x00}, {0x447, 0x00}, {0x448, 0x00}, {0x449, 0xF0},
+	{0x44A, 0x0F}, {0x44B, 0x3E}, {0x44C, 0x10}, {0x44D, 0x00},
+	{0x44E, 0x00}, {0x44F, 0x00}, {0x450, 0x00}, {0x451, 0xF0},
+	{0x452, 0x0F}, {0x453, 0x00}, {0x456, 0x5E}, {0x460, 0x66},
+	{0x461, 0x66}, {0x4C8, 0xFF}, {0x4C9, 0x08}, {0x4CC, 0xFF},
+	{0x4CD, 0xFF}, {0x4CE, 0x01}, {0x500, 0x26}, {0x501, 0xA2},
+	{0x502, 0x2F}, {0x503, 0x00}, {0x504, 0x28}, {0x505, 0xA3},
+	{0x506, 0x5E}, {0x507, 0x00}, {0x508, 0x2B}, {0x509, 0xA4},
+	{0x50A, 0x5E}, {0x50B, 0x00}, {0x50C, 0x4F}, {0x50D, 0xA4},
+	{0x50E, 0x00}, {0x50F, 0x00}, {0x512, 0x1C}, {0x514, 0x0A},
+	{0x516, 0x0A}, {0x525, 0x4F}, {0x550, 0x10}, {0x551, 0x10},
+	{0x559, 0x02}, {0x55C, 0x28}, {0x55D, 0xFF}, {0x605, 0x30},
+	{0x608, 0x0E}, {0x609, 0x2A}, {0x620, 0xFF}, {0x621, 0xFF},
+	{0x622, 0xFF}, {0x623, 0xFF}, {0x624, 0xFF}, {0x625, 0xFF},
+	{0x626, 0xFF}, {0x627, 0xFF}, {0x638, 0x28}, {0x63C, 0x0A},
+	{0x63D, 0x0A}, {0x63E, 0x0C}, {0x63F, 0x0C}, {0x640, 0x40},
+	{0x642, 0x40}, {0x643, 0x00}, {0x652, 0xC8}, {0x66A, 0xB0},
+	{0x66E, 0x05}, {0x700, 0x21}, {0x701, 0x43}, {0x702, 0x65},
+	{0x703, 0x87}, {0x708, 0x21}, {0x709, 0x43}, {0x70A, 0x65},
+	{0x70B, 0x87},
+	{0xffff, 0xff},
+};
+
+/* If updating the phy init tables, also update rtl8710b_revise_cck_tx_psf(). */
+static const struct rtl8xxxu_reg32val rtl8710bu_qfn48m_u_phy_init_table[] = {
+	{0x800, 0x80045700}, {0x804, 0x00000001},
+	{0x808, 0x00FC8000}, {0x80C, 0x0000000A},
+	{0x810, 0x10001331}, {0x814, 0x020C3D10},
+	{0x818, 0x00200385}, {0x81C, 0x00000000},
+	{0x820, 0x01000100}, {0x824, 0x00390204},
+	{0x828, 0x00000000}, {0x82C, 0x00000000},
+	{0x830, 0x00000000}, {0x834, 0x00000000},
+	{0x838, 0x00000000}, {0x83C, 0x00000000},
+	{0x840, 0x00010000}, {0x844, 0x00000000},
+	{0x848, 0x00000000}, {0x84C, 0x00000000},
+	{0x850, 0x00030000}, {0x854, 0x00000000},
+	{0x858, 0x7E1A569A}, {0x85C, 0x569A569A},
+	{0x860, 0x00000130}, {0x864, 0x20000000},
+	{0x868, 0x00000000}, {0x86C, 0x27272700},
+	{0x870, 0x00050000}, {0x874, 0x25005000},
+	{0x878, 0x00000808}, {0x87C, 0x004F0201},
+	{0x880, 0xB0000B1E}, {0x884, 0x00000007},
+	{0x888, 0x00000000}, {0x88C, 0xCCC400C0},
+	{0x890, 0x00000800}, {0x894, 0xFFFFFFFE},
+	{0x898, 0x40302010}, {0x89C, 0x00706050},
+	{0x900, 0x00000000}, {0x904, 0x00000023},
+	{0x908, 0x00000000}, {0x90C, 0x81121111},
+	{0x910, 0x00000402}, {0x914, 0x00000201},
+	{0x920, 0x18C6318C}, {0x924, 0x0000018C},
+	{0x948, 0x99000000}, {0x94C, 0x00000010},
+	{0x950, 0x00003000}, {0x954, 0x5A880000},
+	{0x958, 0x4BC6D87A}, {0x95C, 0x04EB9B79},
+	{0x96C, 0x00000003}, {0x970, 0x00000000},
+	{0x974, 0x00000000}, {0x978, 0x00000000},
+	{0x97C, 0x13000000}, {0x980, 0x00000000},
+	{0xA00, 0x00D046C8}, {0xA04, 0x80FF800C},
+	{0xA08, 0x84838300}, {0xA0C, 0x2E20100F},
+	{0xA10, 0x9500BB78}, {0xA14, 0x1114D028},
+	{0xA18, 0x00881117}, {0xA1C, 0x89140F00},
+	{0xA20, 0xE82C0001}, {0xA24, 0x64B80C1C},
+	{0xA28, 0x00008810}, {0xA2C, 0x00D30000},
+	{0xA70, 0x101FBF00}, {0xA74, 0x00000007},
+	{0xA78, 0x00000900}, {0xA7C, 0x225B0606},
+	{0xA80, 0x218075B1}, {0xA84, 0x00200000},
+	{0xA88, 0x040C0000}, {0xA8C, 0x12345678},
+	{0xA90, 0xABCDEF00}, {0xA94, 0x001B1B89},
+	{0xA98, 0x00000000}, {0xA9C, 0x80020000},
+	{0xAA0, 0x00000000}, {0xAA4, 0x0000000C},
+	{0xAA8, 0xCA110058}, {0xAAC, 0x01235667},
+	{0xAB0, 0x00000000}, {0xAB4, 0x20201402},
+	{0xB2C, 0x00000000}, {0xC00, 0x48071D40},
+	{0xC04, 0x03A05611}, {0xC08, 0x000000E4},
+	{0xC0C, 0x6C6C6C6C}, {0xC10, 0x18800000},
+	{0xC14, 0x40000100}, {0xC18, 0x08800000},
+	{0xC1C, 0x40000100}, {0xC20, 0x00000000},
+	{0xC24, 0x00000000}, {0xC28, 0x00000000},
+	{0xC2C, 0x00000000}, {0xC30, 0x69E9AC4A},
+	{0xC34, 0x31000040}, {0xC38, 0x21688080},
+	{0xC3C, 0x0000170C}, {0xC40, 0x1F78403F},
+	{0xC44, 0x00010036}, {0xC48, 0xEC020107},
+	{0xC4C, 0x007F037F}, {0xC50, 0x69553420},
+	{0xC54, 0x43BC0094}, {0xC58, 0x00013169},
+	{0xC5C, 0x00250492}, {0xC60, 0x00280A00},
+	{0xC64, 0x7112848B}, {0xC68, 0x47C074FF},
+	{0xC6C, 0x00000036}, {0xC70, 0x2C7F000D},
+	{0xC74, 0x020600DB}, {0xC78, 0x0000001F},
+	{0xC7C, 0x00B91612}, {0xC80, 0x390000E4},
+	{0xC84, 0x11F60000}, {0xC88, 0x1051B75F},
+	{0xC8C, 0x20200109}, {0xC90, 0x00091521},
+	{0xC94, 0x00000000}, {0xC98, 0x00121820},
+	{0xC9C, 0x00007F7F}, {0xCA0, 0x00011000},
+	{0xCA4, 0x800000A0}, {0xCA8, 0x84E6C606},
+	{0xCAC, 0x00000060}, {0xCB0, 0x00000000},
+	{0xCB4, 0x00000000}, {0xCB8, 0x00000000},
+	{0xCBC, 0x28000000}, {0xCC0, 0x1051B75F},
+	{0xCC4, 0x00000109}, {0xCC8, 0x000442D6},
+	{0xCCC, 0x00000000}, {0xCD0, 0x000001C8},
+	{0xCD4, 0x001C8000}, {0xCD8, 0x00000100},
+	{0xCDC, 0x40100000}, {0xCE0, 0x00222220},
+	{0xCE4, 0x10000000}, {0xCE8, 0x37644302},
+	{0xCEC, 0x2F97D40C}, {0xD00, 0x04030740},
+	{0xD04, 0x40020401}, {0xD08, 0x0000907F},
+	{0xD0C, 0x20010201}, {0xD10, 0xA0633333},
+	{0xD14, 0x3333BC53}, {0xD18, 0x7A8F5B6F},
+	{0xD2C, 0xCB979975}, {0xD30, 0x00000000},
+	{0xD34, 0x40608000}, {0xD38, 0x88000000},
+	{0xD3C, 0xC0127353}, {0xD40, 0x00000000},
+	{0xD44, 0x00000000}, {0xD48, 0x00000000},
+	{0xD4C, 0x00000000}, {0xD50, 0x00006528},
+	{0xD54, 0x00000000}, {0xD58, 0x00000282},
+	{0xD5C, 0x30032064}, {0xD60, 0x4653DE68},
+	{0xD64, 0x04518A3C}, {0xD68, 0x00002101},
+	{0xE00, 0x2D2D2D2D}, {0xE04, 0x2D2D2D2D},
+	{0xE08, 0x0390272D}, {0xE10, 0x2D2D2D2D},
+	{0xE14, 0x2D2D2D2D}, {0xE18, 0x2D2D2D2D},
+	{0xE1C, 0x2D2D2D2D}, {0xE28, 0x00000000},
+	{0xE30, 0x1000DC1F}, {0xE34, 0x10008C1F},
+	{0xE38, 0x02140102}, {0xE3C, 0x681604C2},
+	{0xE40, 0x01007C00}, {0xE44, 0x01004800},
+	{0xE48, 0xFB000000}, {0xE4C, 0x000028D1},
+	{0xE50, 0x1000DC1F}, {0xE54, 0x10008C1F},
+	{0xE58, 0x02140102}, {0xE5C, 0x28160D05},
+	{0xE60, 0x0000C008}, {0xE68, 0x001B25A4},
+	{0xE64, 0x281600A0}, {0xE6C, 0x01C00010},
+	{0xE70, 0x01C00010}, {0xE74, 0x02000010},
+	{0xE78, 0x02000010}, {0xE7C, 0x02000010},
+	{0xE80, 0x02000010}, {0xE84, 0x01C00010},
+	{0xE88, 0x02000010}, {0xE8C, 0x01C00010},
+	{0xED0, 0x01C00010}, {0xED4, 0x01C00010},
+	{0xED8, 0x01C00010}, {0xEDC, 0x00000010},
+	{0xEE0, 0x00000010}, {0xEEC, 0x03C00010},
+	{0xF14, 0x00000003}, {0xF00, 0x00100300},
+	{0xF08, 0x0000800B}, {0xF0C, 0x0000F007},
+	{0xF10, 0x0000A487}, {0xF1C, 0x80000064},
+	{0xF38, 0x00030155}, {0xF3C, 0x0000003A},
+	{0xF4C, 0x13000000}, {0xF50, 0x00000000},
+	{0xF18, 0x00000000},
+	{0xffff, 0xffffffff},
+};
+
+/* If updating the phy init tables, also update rtl8710b_revise_cck_tx_psf(). */
+static const struct rtl8xxxu_reg32val rtl8710bu_qfn48m_s_phy_init_table[] = {
+	{0x800, 0x80045700}, {0x804, 0x00000001},
+	{0x808, 0x00FC8000}, {0x80C, 0x0000000A},
+	{0x810, 0x10001331}, {0x814, 0x020C3D10},
+	{0x818, 0x00200385}, {0x81C, 0x00000000},
+	{0x820, 0x01000100}, {0x824, 0x00390204},
+	{0x828, 0x00000000}, {0x82C, 0x00000000},
+	{0x830, 0x00000000}, {0x834, 0x00000000},
+	{0x838, 0x00000000}, {0x83C, 0x00000000},
+	{0x840, 0x00010000}, {0x844, 0x00000000},
+	{0x848, 0x00000000}, {0x84C, 0x00000000},
+	{0x850, 0x00030000}, {0x854, 0x00000000},
+	{0x858, 0x7E1A569A}, {0x85C, 0x569A569A},
+	{0x860, 0x00000130}, {0x864, 0x20000000},
+	{0x868, 0x00000000}, {0x86C, 0x27272700},
+	{0x870, 0x00050000}, {0x874, 0x25005000},
+	{0x878, 0x00000808}, {0x87C, 0x004F0201},
+	{0x880, 0xB0000B1E}, {0x884, 0x00000007},
+	{0x888, 0x00000000}, {0x88C, 0xCCC400C0},
+	{0x890, 0x00000800}, {0x894, 0xFFFFFFFE},
+	{0x898, 0x40302010}, {0x89C, 0x00706050},
+	{0x900, 0x00000000}, {0x904, 0x00000023},
+	{0x908, 0x00000000}, {0x90C, 0x81121111},
+	{0x910, 0x00000402}, {0x914, 0x00000201},
+	{0x920, 0x18C6318C}, {0x924, 0x0000018C},
+	{0x948, 0x99000000}, {0x94C, 0x00000010},
+	{0x950, 0x00003000}, {0x954, 0x5A880000},
+	{0x958, 0x4BC6D87A}, {0x95C, 0x04EB9B79},
+	{0x96C, 0x00000003}, {0x970, 0x00000000},
+	{0x974, 0x00000000}, {0x978, 0x00000000},
+	{0x97C, 0x13000000}, {0x980, 0x00000000},
+	{0xA00, 0x00D046C8}, {0xA04, 0x80FF800C},
+	{0xA08, 0x84838300}, {0xA0C, 0x2A20100F},
+	{0xA10, 0x9500BB78}, {0xA14, 0x1114D028},
+	{0xA18, 0x00881117}, {0xA1C, 0x89140F00},
+	{0xA20, 0xE82C0001}, {0xA24, 0x64B80C1C},
+	{0xA28, 0x00008810}, {0xA2C, 0x00D30000},
+	{0xA70, 0x101FBF00}, {0xA74, 0x00000007},
+	{0xA78, 0x00000900}, {0xA7C, 0x225B0606},
+	{0xA80, 0x218075B1}, {0xA84, 0x00200000},
+	{0xA88, 0x040C0000}, {0xA8C, 0x12345678},
+	{0xA90, 0xABCDEF00}, {0xA94, 0x001B1B89},
+	{0xA98, 0x00000000}, {0xA9C, 0x80020000},
+	{0xAA0, 0x00000000}, {0xAA4, 0x0000000C},
+	{0xAA8, 0xCA110058}, {0xAAC, 0x01235667},
+	{0xAB0, 0x00000000}, {0xAB4, 0x20201402},
+	{0xB2C, 0x00000000}, {0xC00, 0x48071D40},
+	{0xC04, 0x03A05611}, {0xC08, 0x000000E4},
+	{0xC0C, 0x6C6C6C6C}, {0xC10, 0x18800000},
+	{0xC14, 0x40000100}, {0xC18, 0x08800000},
+	{0xC1C, 0x40000100}, {0xC20, 0x00000000},
+	{0xC24, 0x00000000}, {0xC28, 0x00000000},
+	{0xC2C, 0x00000000}, {0xC30, 0x69E9AC4A},
+	{0xC34, 0x31000040}, {0xC38, 0x21688080},
+	{0xC3C, 0x0000170C}, {0xC40, 0x1F78403F},
+	{0xC44, 0x00010036}, {0xC48, 0xEC020107},
+	{0xC4C, 0x007F037F}, {0xC50, 0x69553420},
+	{0xC54, 0x43BC0094}, {0xC58, 0x00013169},
+	{0xC5C, 0x00250492}, {0xC60, 0x00280A00},
+	{0xC64, 0x7112848B}, {0xC68, 0x47C074FF},
+	{0xC6C, 0x00000036}, {0xC70, 0x2C7F000D},
+	{0xC74, 0x020600DB}, {0xC78, 0x0000001F},
+	{0xC7C, 0x00B91612}, {0xC80, 0x390000E4},
+	{0xC84, 0x11F60000}, {0xC88, 0x1051B75F},
+	{0xC8C, 0x20200109}, {0xC90, 0x00091521},
+	{0xC94, 0x00000000}, {0xC98, 0x00121820},
+	{0xC9C, 0x00007F7F}, {0xCA0, 0x00011000},
+	{0xCA4, 0x800000A0}, {0xCA8, 0x84E6C606},
+	{0xCAC, 0x00000060}, {0xCB0, 0x00000000},
+	{0xCB4, 0x00000000}, {0xCB8, 0x00000000},
+	{0xCBC, 0x28000000}, {0xCC0, 0x1051B75F},
+	{0xCC4, 0x00000109}, {0xCC8, 0x000442D6},
+	{0xCCC, 0x00000000}, {0xCD0, 0x000001C8},
+	{0xCD4, 0x001C8000}, {0xCD8, 0x00000100},
+	{0xCDC, 0x40100000}, {0xCE0, 0x00222220},
+	{0xCE4, 0x10000000}, {0xCE8, 0x37644302},
+	{0xCEC, 0x2F97D40C}, {0xD00, 0x04030740},
+	{0xD04, 0x40020401}, {0xD08, 0x0000907F},
+	{0xD0C, 0x20010201}, {0xD10, 0xA0633333},
+	{0xD14, 0x3333BC53}, {0xD18, 0x7A8F5B6F},
+	{0xD2C, 0xCB979975}, {0xD30, 0x00000000},
+	{0xD34, 0x40608000}, {0xD38, 0x88000000},
+	{0xD3C, 0xC0127353}, {0xD40, 0x00000000},
+	{0xD44, 0x00000000}, {0xD48, 0x00000000},
+	{0xD4C, 0x00000000}, {0xD50, 0x00006528},
+	{0xD54, 0x00000000}, {0xD58, 0x00000282},
+	{0xD5C, 0x30032064}, {0xD60, 0x4653DE68},
+	{0xD64, 0x04518A3C}, {0xD68, 0x00002101},
+	{0xE00, 0x2D2D2D2D}, {0xE04, 0x2D2D2D2D},
+	{0xE08, 0x0390272D}, {0xE10, 0x2D2D2D2D},
+	{0xE14, 0x2D2D2D2D}, {0xE18, 0x2D2D2D2D},
+	{0xE1C, 0x2D2D2D2D}, {0xE28, 0x00000000},
+	{0xE30, 0x1000DC1F}, {0xE34, 0x10008C1F},
+	{0xE38, 0x02140102}, {0xE3C, 0x681604C2},
+	{0xE40, 0x01007C00}, {0xE44, 0x01004800},
+	{0xE48, 0xFB000000}, {0xE4C, 0x000028D1},
+	{0xE50, 0x1000DC1F}, {0xE54, 0x10008C1F},
+	{0xE58, 0x02140102}, {0xE5C, 0x28160D05},
+	{0xE60, 0x0000C008}, {0xE68, 0x001B25A4},
+	{0xE64, 0x281600A0}, {0xE6C, 0x01C00010},
+	{0xE70, 0x01C00010}, {0xE74, 0x02000010},
+	{0xE78, 0x02000010}, {0xE7C, 0x02000010},
+	{0xE80, 0x02000010}, {0xE84, 0x01C00010},
+	{0xE88, 0x02000010}, {0xE8C, 0x01C00010},
+	{0xED0, 0x01C00010}, {0xED4, 0x01C00010},
+	{0xED8, 0x01C00010}, {0xEDC, 0x00000010},
+	{0xEE0, 0x00000010}, {0xEEC, 0x03C00010},
+	{0xF14, 0x00000003}, {0xF00, 0x00100300},
+	{0xF08, 0x0000800B}, {0xF0C, 0x0000F007},
+	{0xF10, 0x0000A487}, {0xF1C, 0x80000064},
+	{0xF38, 0x00030155}, {0xF3C, 0x0000003A},
+	{0xF4C, 0x13000000}, {0xF50, 0x00000000},
+	{0xF18, 0x00000000},
+	{0xffff, 0xffffffff},
+};
+
+static const struct rtl8xxxu_reg32val rtl8710b_agc_table[] = {
+	{0xC78, 0xFC000001}, {0xC78, 0xFB010001},
+	{0xC78, 0xFA020001}, {0xC78, 0xF9030001},
+	{0xC78, 0xF8040001}, {0xC78, 0xF7050001},
+	{0xC78, 0xF6060001}, {0xC78, 0xF5070001},
+	{0xC78, 0xF4080001}, {0xC78, 0xF3090001},
+	{0xC78, 0xF20A0001}, {0xC78, 0xF10B0001},
+	{0xC78, 0xF00C0001}, {0xC78, 0xEF0D0001},
+	{0xC78, 0xEE0E0001}, {0xC78, 0xED0F0001},
+	{0xC78, 0xEC100001}, {0xC78, 0xEB110001},
+	{0xC78, 0xEA120001}, {0xC78, 0xE9130001},
+	{0xC78, 0xE8140001}, {0xC78, 0xE7150001},
+	{0xC78, 0xE6160001}, {0xC78, 0xE5170001},
+	{0xC78, 0xE4180001}, {0xC78, 0xE3190001},
+	{0xC78, 0xE21A0001}, {0xC78, 0xE11B0001},
+	{0xC78, 0xE01C0001}, {0xC78, 0xC31D0001},
+	{0xC78, 0xC21E0001}, {0xC78, 0xC11F0001},
+	{0xC78, 0xC0200001}, {0xC78, 0xA3210001},
+	{0xC78, 0xA2220001}, {0xC78, 0xA1230001},
+	{0xC78, 0xA0240001}, {0xC78, 0x86250001},
+	{0xC78, 0x85260001}, {0xC78, 0x84270001},
+	{0xC78, 0x83280001}, {0xC78, 0x82290001},
+	{0xC78, 0x812A0001}, {0xC78, 0x802B0001},
+	{0xC78, 0x632C0001}, {0xC78, 0x622D0001},
+	{0xC78, 0x612E0001}, {0xC78, 0x602F0001},
+	{0xC78, 0x42300001}, {0xC78, 0x41310001},
+	{0xC78, 0x40320001}, {0xC78, 0x23330001},
+	{0xC78, 0x22340001}, {0xC78, 0x21350001},
+	{0xC78, 0x20360001}, {0xC78, 0x02370001},
+	{0xC78, 0x01380001}, {0xC78, 0x00390001},
+	{0xC78, 0x003A0001}, {0xC78, 0x003B0001},
+	{0xC78, 0x003C0001}, {0xC78, 0x003D0001},
+	{0xC78, 0x003E0001}, {0xC78, 0x003F0001},
+	{0xC78, 0xF7400001}, {0xC78, 0xF7410001},
+	{0xC78, 0xF7420001}, {0xC78, 0xF7430001},
+	{0xC78, 0xF7440001}, {0xC78, 0xF7450001},
+	{0xC78, 0xF7460001}, {0xC78, 0xF7470001},
+	{0xC78, 0xF7480001}, {0xC78, 0xF6490001},
+	{0xC78, 0xF34A0001}, {0xC78, 0xF24B0001},
+	{0xC78, 0xF14C0001}, {0xC78, 0xF04D0001},
+	{0xC78, 0xD14E0001}, {0xC78, 0xD04F0001},
+	{0xC78, 0xB5500001}, {0xC78, 0xB4510001},
+	{0xC78, 0xB3520001}, {0xC78, 0xB2530001},
+	{0xC78, 0xB1540001}, {0xC78, 0xB0550001},
+	{0xC78, 0xAF560001}, {0xC78, 0xAE570001},
+	{0xC78, 0xAD580001}, {0xC78, 0xAC590001},
+	{0xC78, 0xAB5A0001}, {0xC78, 0xAA5B0001},
+	{0xC78, 0xA95C0001}, {0xC78, 0xA85D0001},
+	{0xC78, 0xA75E0001}, {0xC78, 0xA65F0001},
+	{0xC78, 0xA5600001}, {0xC78, 0xA4610001},
+	{0xC78, 0xA3620001}, {0xC78, 0xA2630001},
+	{0xC78, 0xA1640001}, {0xC78, 0xA0650001},
+	{0xC78, 0x87660001}, {0xC78, 0x86670001},
+	{0xC78, 0x85680001}, {0xC78, 0x84690001},
+	{0xC78, 0x836A0001}, {0xC78, 0x826B0001},
+	{0xC78, 0x816C0001}, {0xC78, 0x806D0001},
+	{0xC78, 0x636E0001}, {0xC78, 0x626F0001},
+	{0xC78, 0x61700001}, {0xC78, 0x60710001},
+	{0xC78, 0x42720001}, {0xC78, 0x41730001},
+	{0xC78, 0x40740001}, {0xC78, 0x23750001},
+	{0xC78, 0x22760001}, {0xC78, 0x21770001},
+	{0xC78, 0x20780001}, {0xC78, 0x03790001},
+	{0xC78, 0x027A0001}, {0xC78, 0x017B0001},
+	{0xC78, 0x007C0001}, {0xC78, 0x007D0001},
+	{0xC78, 0x007E0001}, {0xC78, 0x007F0001},
+	{0xC50, 0x69553422}, {0xC50, 0x69553420},
+	{0xffff, 0xffffffff}
+};
+
+static const struct rtl8xxxu_rfregval rtl8710bu_qfn48m_u_radioa_init_table[] = {
+	{0x00, 0x00030000}, {0x08, 0x00008400},
+	{0x17, 0x00000000}, {0x18, 0x00000C01},
+	{0x19, 0x000739D2}, {0x1C, 0x00000C4C},
+	{0x1B, 0x00000C6C}, {0x1E, 0x00080009},
+	{0x1F, 0x00000880}, {0x2F, 0x0001A060},
+	{0x3F, 0x00015000}, {0x42, 0x000060C0},
+	{0x57, 0x000D0000}, {0x58, 0x000C0160},
+	{0x67, 0x00001552}, {0x83, 0x00000000},
+	{0xB0, 0x000FF9F0}, {0xB1, 0x00010018},
+	{0xB2, 0x00054C00}, {0xB4, 0x0004486B},
+	{0xB5, 0x0000112A}, {0xB6, 0x0000053E},
+	{0xB7, 0x00014408}, {0xB8, 0x00010200},
+	{0xB9, 0x00080801}, {0xBA, 0x00040001},
+	{0xBB, 0x00000400}, {0xBF, 0x000C0000},
+	{0xC2, 0x00002400}, {0xC3, 0x00000009},
+	{0xC4, 0x00040C91}, {0xC5, 0x00099999},
+	{0xC6, 0x000000A3}, {0xC7, 0x00088820},
+	{0xC8, 0x00076C06}, {0xC9, 0x00000000},
+	{0xCA, 0x00080000}, {0xDF, 0x00000180},
+	{0xEF, 0x000001A8}, {0x3D, 0x00000003},
+	{0x3D, 0x00080003}, {0x51, 0x000F1E69},
+	{0x52, 0x000FBF6C}, {0x53, 0x0000032F},
+	{0x54, 0x00055007}, {0x56, 0x000517F0},
+	{0x35, 0x000000F4}, {0x35, 0x00000179},
+	{0x35, 0x000002F4}, {0x36, 0x00000BF8},
+	{0x36, 0x00008BF8}, {0x36, 0x00010BF8},
+	{0x36, 0x00018BF8}, {0x18, 0x00000C01},
+	{0x5A, 0x00048000}, {0x5A, 0x00048000},
+	{0x34, 0x0000ADF5}, {0x34, 0x00009DF2},
+	{0x34, 0x00008DEF}, {0x34, 0x00007DEC},
+	{0x34, 0x00006DE9}, {0x34, 0x00005CEC},
+	{0x34, 0x00004CE9}, {0x34, 0x00003C6C},
+	{0x34, 0x00002C69}, {0x34, 0x0000106E},
+	{0x34, 0x0000006B}, {0x84, 0x00048000},
+	{0x87, 0x00000065}, {0x8E, 0x00065540},
+	{0xDF, 0x00000110}, {0x86, 0x0000002A},
+	{0x8F, 0x00088000}, {0x81, 0x0003FD80},
+	{0xEF, 0x00082000}, {0x3B, 0x000F0F00},
+	{0x3B, 0x000E0E00}, {0x3B, 0x000DFE00},
+	{0x3B, 0x000C0D00}, {0x3B, 0x000B0C00},
+	{0x3B, 0x000A0500}, {0x3B, 0x00090400},
+	{0x3B, 0x00080000}, {0x3B, 0x00070F00},
+	{0x3B, 0x00060E00}, {0x3B, 0x00050A00},
+	{0x3B, 0x00040D00}, {0x3B, 0x00030C00},
+	{0x3B, 0x00020500}, {0x3B, 0x00010400},
+	{0x3B, 0x00000000}, {0xEF, 0x00080000},
+	{0xEF, 0x00088000}, {0x3B, 0x00000170},
+	{0x3B, 0x000C0030}, {0xEF, 0x00080000},
+	{0xEF, 0x00080000}, {0x30, 0x00010000},
+	{0x31, 0x0000000F}, {0x32, 0x00047EFE},
+	{0xEF, 0x00000000}, {0x00, 0x00010159},
+	{0x18, 0x0000FC01}, {0xFE, 0x00000000},
+	{0x00, 0x00033D95},
+	{0xff, 0xffffffff}
+};
+
+static const struct rtl8xxxu_rfregval rtl8710bu_qfn48m_s_radioa_init_table[] = {
+	{0x00, 0x00030000}, {0x08, 0x00008400},
+	{0x17, 0x00000000}, {0x18, 0x00000C01},
+	{0x19, 0x000739D2}, {0x1C, 0x00000C4C},
+	{0x1B, 0x00000C6C}, {0x1E, 0x00080009},
+	{0x1F, 0x00000880}, {0x2F, 0x0001A060},
+	{0x3F, 0x00015000}, {0x42, 0x000060C0},
+	{0x57, 0x000D0000}, {0x58, 0x000C0160},
+	{0x67, 0x00001552}, {0x83, 0x00000000},
+	{0xB0, 0x000FF9F0}, {0xB1, 0x00010018},
+	{0xB2, 0x00054C00}, {0xB4, 0x0004486B},
+	{0xB5, 0x0000112A}, {0xB6, 0x0000053E},
+	{0xB7, 0x00014408}, {0xB8, 0x00010200},
+	{0xB9, 0x00080801}, {0xBA, 0x00040001},
+	{0xBB, 0x00000400}, {0xBF, 0x000C0000},
+	{0xC2, 0x00002400}, {0xC3, 0x00000009},
+	{0xC4, 0x00040C91}, {0xC5, 0x00099999},
+	{0xC6, 0x000000A3}, {0xC7, 0x00088820},
+	{0xC8, 0x00076C06}, {0xC9, 0x00000000},
+	{0xCA, 0x00080000}, {0xDF, 0x00000180},
+	{0xEF, 0x000001A8}, {0x3D, 0x00000003},
+	{0x3D, 0x00080003}, {0x51, 0x000F1E69},
+	{0x52, 0x000FBF6C}, {0x53, 0x0000032F},
+	{0x54, 0x00055007}, {0x56, 0x000517F0},
+	{0x35, 0x000000F4}, {0x35, 0x00000179},
+	{0x35, 0x000002F4}, {0x36, 0x00000BF8},
+	{0x36, 0x00008BF8}, {0x36, 0x00010BF8},
+	{0x36, 0x00018BF8}, {0x18, 0x00000C01},
+	{0x5A, 0x00048000}, {0x5A, 0x00048000},
+	{0x34, 0x0000ADF5}, {0x34, 0x00009DF2},
+	{0x34, 0x00008DEF}, {0x34, 0x00007DEC},
+	{0x34, 0x00006DE9}, {0x34, 0x00005CEC},
+	{0x34, 0x00004CE9}, {0x34, 0x00003C6C},
+	{0x34, 0x00002C69}, {0x34, 0x0000106E},
+	{0x34, 0x0000006B}, {0x84, 0x00048000},
+	{0x87, 0x00000065}, {0x8E, 0x00065540},
+	{0xDF, 0x00000110}, {0x86, 0x0000002A},
+	{0x8F, 0x00088000}, {0x81, 0x0003FD80},
+	{0xEF, 0x00082000}, {0x3B, 0x000F0F00},
+	{0x3B, 0x000E0E00}, {0x3B, 0x000DFE00},
+	{0x3B, 0x000C0D00}, {0x3B, 0x000B0C00},
+	{0x3B, 0x000A0500}, {0x3B, 0x00090400},
+	{0x3B, 0x00080000}, {0x3B, 0x00070F00},
+	{0x3B, 0x00060E00}, {0x3B, 0x00050A00},
+	{0x3B, 0x00040D00}, {0x3B, 0x00030C00},
+	{0x3B, 0x00020500}, {0x3B, 0x00010400},
+	{0x3B, 0x00000000}, {0xEF, 0x00080000},
+	{0xEF, 0x00088000}, {0x3B, 0x000000B0},
+	{0x3B, 0x000C0030}, {0xEF, 0x00080000},
+	{0xEF, 0x00080000}, {0x30, 0x00010000},
+	{0x31, 0x0000000F}, {0x32, 0x00047EFE},
+	{0xEF, 0x00000000}, {0x00, 0x00010159},
+	{0x18, 0x0000FC01}, {0xFE, 0x00000000},
+	{0x00, 0x00033D95},
+	{0xff, 0xffffffff}
+};
+
+static u32 rtl8710b_indirect_read32(struct rtl8xxxu_priv *priv, u32 addr)
+{
+	struct device *dev = &priv->udev->dev;
+	u32 val32, value = 0xffffffff;
+	u8 polling_count = 0xff;
+
+	if (!IS_ALIGNED(addr, 4)) {
+		dev_warn(dev, "%s: Aborting because 0x%x is not a multiple of 4.\n",
+			 __func__, addr);
+		return value;
+	}
+
+	mutex_lock(&priv->syson_indirect_access_mutex);
+
+	rtl8xxxu_write32(priv, REG_USB_HOST_INDIRECT_ADDR_8710B, addr);
+	rtl8xxxu_write32(priv, REG_EFUSE_INDIRECT_CTRL_8710B, NORMAL_REG_READ_OFFSET);
+
+	do
+		val32 = rtl8xxxu_read32(priv, REG_EFUSE_INDIRECT_CTRL_8710B);
+	while ((val32 & BIT(31)) && (--polling_count > 0));
+
+	if (polling_count == 0)
+		dev_warn(dev, "%s: Failed to read from 0x%x, 0x806c = 0x%x\n",
+			 __func__, addr, val32);
+	else
+		value = rtl8xxxu_read32(priv, REG_USB_HOST_INDIRECT_DATA_8710B);
+
+	mutex_unlock(&priv->syson_indirect_access_mutex);
+
+	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_REG_READ)
+		dev_info(dev, "%s(%04x) = 0x%08x\n", __func__, addr, value);
+
+	return value;
+}
+
+static void rtl8710b_indirect_write32(struct rtl8xxxu_priv *priv, u32 addr, u32 val)
+{
+	struct device *dev = &priv->udev->dev;
+	u8 polling_count = 0xff;
+	u32 val32;
+
+	if (!IS_ALIGNED(addr, 4)) {
+		dev_warn(dev, "%s: Aborting because 0x%x is not a multiple of 4.\n",
+			 __func__, addr);
+		return;
+	}
+
+	mutex_lock(&priv->syson_indirect_access_mutex);
+
+	rtl8xxxu_write32(priv, REG_USB_HOST_INDIRECT_ADDR_8710B, addr);
+	rtl8xxxu_write32(priv, REG_USB_HOST_INDIRECT_DATA_8710B, val);
+	rtl8xxxu_write32(priv, REG_EFUSE_INDIRECT_CTRL_8710B, NORMAL_REG_WRITE_OFFSET);
+
+	do
+		val32 = rtl8xxxu_read32(priv, REG_EFUSE_INDIRECT_CTRL_8710B);
+	while ((val32 & BIT(31)) && (--polling_count > 0));
+
+	if (polling_count == 0)
+		dev_warn(dev, "%s: Failed to write 0x%x to 0x%x, 0x806c = 0x%x\n",
+			 __func__, val, addr, val32);
+
+	mutex_unlock(&priv->syson_indirect_access_mutex);
+
+	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_REG_WRITE)
+		dev_info(dev, "%s(%04x) = 0x%08x\n", __func__, addr, val);
+}
+
+static u32 rtl8710b_read_syson_reg(struct rtl8xxxu_priv *priv, u32 addr)
+{
+	return rtl8710b_indirect_read32(priv, addr | SYSON_REG_BASE_ADDR_8710B);
+}
+
+static void rtl8710b_write_syson_reg(struct rtl8xxxu_priv *priv, u32 addr, u32 val)
+{
+	rtl8710b_indirect_write32(priv, addr | SYSON_REG_BASE_ADDR_8710B, val);
+}
+
+static int rtl8710b_read_efuse8(struct rtl8xxxu_priv *priv, u16 offset, u8 *data)
+{
+	u32 val32;
+	int i;
+
+	/* Write Address */
+	rtl8xxxu_write32(priv, REG_USB_HOST_INDIRECT_ADDR_8710B, offset);
+
+	rtl8xxxu_write32(priv, REG_EFUSE_INDIRECT_CTRL_8710B, EFUSE_READ_OFFSET);
+
+	/* Poll for data read */
+	val32 = rtl8xxxu_read32(priv, REG_EFUSE_INDIRECT_CTRL_8710B);
+	for (i = 0; i < RTL8XXXU_MAX_REG_POLL; i++) {
+		val32 = rtl8xxxu_read32(priv, REG_EFUSE_INDIRECT_CTRL_8710B);
+		if (!(val32 & BIT(31)))
+			break;
+	}
+
+	if (i == RTL8XXXU_MAX_REG_POLL)
+		return -EIO;
+
+	val32 = rtl8xxxu_read32(priv, REG_USB_HOST_INDIRECT_DATA_8710B);
+
+	*data = val32 & 0xff;
+	return 0;
+}
+
+#define EEPROM_PACKAGE_TYPE_8710B	0xF8
+#define PACKAGE_QFN48M_U		0xee
+#define PACKAGE_QFN48M_S		0xfe
+
+static int rtl8710bu_identify_chip(struct rtl8xxxu_priv *priv)
+{
+	struct device *dev = &priv->udev->dev;
+	u32 cfg0, cfg2, vendor;
+	u8 package_type = 0x7; /* a nonsense value */
+
+	sprintf(priv->chip_name, "8710BU");
+	priv->rtl_chip = RTL8710B;
+	priv->rf_paths = 1;
+	priv->rx_paths = 1;
+	priv->tx_paths = 1;
+	priv->has_wifi = 1;
+
+	cfg0 = rtl8710b_read_syson_reg(priv, REG_SYS_SYSTEM_CFG0_8710B);
+	priv->chip_cut = cfg0 & 0xf;
+
+	if (cfg0 & BIT(16)) {
+		dev_info(dev, "%s: Unsupported test chip\n", __func__);
+		return -EOPNOTSUPP;
+	}
+
+	vendor = u32_get_bits(cfg0, 0xc0);
+
+	/* SMIC and TSMC are swapped compared to rtl8xxxu_identify_vendor_2bits */
+	switch (vendor) {
+	case 0:
+		sprintf(priv->chip_vendor, "SMIC");
+		priv->vendor_smic = 1;
+		break;
+	case 1:
+		sprintf(priv->chip_vendor, "TSMC");
+		break;
+	case 2:
+		sprintf(priv->chip_vendor, "UMC");
+		priv->vendor_umc = 1;
+		break;
+	default:
+		sprintf(priv->chip_vendor, "unknown");
+		break;
+	}
+
+	rtl8710b_read_efuse8(priv, EEPROM_PACKAGE_TYPE_8710B, &package_type);
+
+	if (package_type == 0xff) {
+		dev_warn(dev, "Package type is undefined. Assuming it based on the vendor.\n");
+
+		if (priv->vendor_umc) {
+			package_type = PACKAGE_QFN48M_U;
+		} else if (priv->vendor_smic) {
+			package_type = PACKAGE_QFN48M_S;
+		} else {
+			dev_warn(dev, "The vendor is neither UMC nor SMIC. Assuming the package type is QFN48M_U.\n");
+
+			/*
+			 * In this case the vendor driver doesn't set
+			 * the package type to anything, which is the
+			 * same as setting it to PACKAGE_DEFAULT (0).
+			 */
+			package_type = PACKAGE_QFN48M_U;
+		}
+	} else if (package_type != PACKAGE_QFN48M_S &&
+		   package_type != PACKAGE_QFN48M_U) {
+		dev_warn(dev, "Failed to read the package type. Assuming it's the default QFN48M_U.\n");
+
+		/*
+		 * In this case the vendor driver actually sets it to
+		 * PACKAGE_DEFAULT, but that selects the same values
+		 * from the init tables as PACKAGE_QFN48M_U.
+		 */
+		package_type = PACKAGE_QFN48M_U;
+	}
+
+	priv->package_type = package_type;
+
+	dev_dbg(dev, "Package type: 0x%x\n", package_type);
+
+	cfg2 = rtl8710b_read_syson_reg(priv, REG_SYS_SYSTEM_CFG2_8710B);
+	priv->rom_rev = cfg2 & 0xf;
+
+	return rtl8xxxu_config_endpoints_no_sie(priv);
+}
+
+static void rtl8710b_revise_cck_tx_psf(struct rtl8xxxu_priv *priv, u8 channel)
+{
+	if (channel == 13) {
+		/* Normal values */
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER2, 0x64B80C1C);
+		rtl8xxxu_write32(priv, REG_CCK0_DEBUG_PORT, 0x00008810);
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER3, 0x01235667);
+		/* Special value for channel 13 */
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER1, 0xd1d80001);
+	} else if (channel == 14) {
+		/* Special values for channel 14 */
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER2, 0x0000B81C);
+		rtl8xxxu_write32(priv, REG_CCK0_DEBUG_PORT, 0x00000000);
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER3, 0x00003667);
+		/* Normal value */
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER1, 0xE82C0001);
+	} else {
+		/* Restore normal values from the phy init table */
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER2, 0x64B80C1C);
+		rtl8xxxu_write32(priv, REG_CCK0_DEBUG_PORT, 0x00008810);
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER3, 0x01235667);
+		rtl8xxxu_write32(priv, REG_CCK0_TX_FILTER1, 0xE82C0001);
+	}
+}
+
+static void rtl8710bu_config_channel(struct ieee80211_hw *hw)
+{
+	struct rtl8xxxu_priv *priv = hw->priv;
+	bool ht40 = conf_is_ht40(&hw->conf);
+	u8 channel, subchannel = 0;
+	bool sec_ch_above = 0;
+	u32 val32;
+	u16 val16;
+
+	channel = (u8)hw->conf.chandef.chan->hw_value;
+
+	if (conf_is_ht40_plus(&hw->conf)) {
+		sec_ch_above = 1;
+		channel += 2;
+		subchannel = 2;
+	} else if (conf_is_ht40_minus(&hw->conf)) {
+		sec_ch_above = 0;
+		channel -= 2;
+		subchannel = 1;
+	}
+
+	/* Set channel */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_MODE_AG);
+	u32p_replace_bits(&val32, channel, MODE_AG_CHANNEL_MASK);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_MODE_AG, val32);
+
+	rtl8710b_revise_cck_tx_psf(priv, channel);
+
+	/* Set bandwidth mode */
+	val16 = rtl8xxxu_read16(priv, REG_WMAC_TRXPTCL_CTL);
+	val16 &= ~WMAC_TRXPTCL_CTL_BW_MASK;
+	if (ht40)
+		val16 |= WMAC_TRXPTCL_CTL_BW_40;
+	rtl8xxxu_write16(priv, REG_WMAC_TRXPTCL_CTL, val16);
+
+	rtl8xxxu_write8(priv, REG_DATA_SUBCHANNEL, subchannel);
+
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
+	u32p_replace_bits(&val32, ht40, FPGA_RF_MODE);
+	rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_FPGA1_RF_MODE);
+	u32p_replace_bits(&val32, ht40, FPGA_RF_MODE);
+	rtl8xxxu_write32(priv, REG_FPGA1_RF_MODE, val32);
+
+	if (ht40) {
+		/* Set Control channel to upper or lower. */
+		val32 = rtl8xxxu_read32(priv, REG_CCK0_SYSTEM);
+		u32p_replace_bits(&val32, !sec_ch_above, CCK0_SIDEBAND);
+		rtl8xxxu_write32(priv, REG_CCK0_SYSTEM, val32);
+	}
+
+	/* RXADC CLK */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
+	val32 |= GENMASK(10, 8);
+	rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+
+	/* TXDAC CLK */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
+	val32 |= BIT(14) | BIT(12);
+	val32 &= ~BIT(13);
+	rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+
+	/* small BW */
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_TX_PSDO_NOISE_WEIGHT);
+	val32 &= ~GENMASK(31, 30);
+	rtl8xxxu_write32(priv, REG_OFDM0_TX_PSDO_NOISE_WEIGHT, val32);
+
+	/* adc buffer clk */
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_TX_PSDO_NOISE_WEIGHT);
+	val32 &= ~BIT(29);
+	val32 |= BIT(28);
+	rtl8xxxu_write32(priv, REG_OFDM0_TX_PSDO_NOISE_WEIGHT, val32);
+
+	/* adc buffer clk */
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_XA_RX_AFE);
+	val32 &= ~BIT(29);
+	val32 |= BIT(28);
+	rtl8xxxu_write32(priv, REG_OFDM0_XA_RX_AFE, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_XB_RF_INT_OE);
+	val32 &= ~BIT(30);
+	val32 |= BIT(29);
+	rtl8xxxu_write32(priv, REG_FPGA0_XB_RF_INT_OE, val32);
+
+	if (ht40) {
+		val32 = rtl8xxxu_read32(priv, REG_OFDM_RX_DFIR);
+		val32 &= ~BIT(19);
+		rtl8xxxu_write32(priv, REG_OFDM_RX_DFIR, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_OFDM_RX_DFIR);
+		val32 &= ~GENMASK(23, 20);
+		rtl8xxxu_write32(priv, REG_OFDM_RX_DFIR, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_OFDM_RX_DFIR);
+		val32 &= ~GENMASK(27, 24);
+		rtl8xxxu_write32(priv, REG_OFDM_RX_DFIR, val32);
+
+		/* RF TRX_BW */
+		val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_MODE_AG);
+		val32 &= ~MODE_AG_BW_MASK;
+		val32 |= MODE_AG_BW_40MHZ_8723B;
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_MODE_AG, val32);
+	} else {
+		val32 = rtl8xxxu_read32(priv, REG_OFDM_RX_DFIR);
+		val32 |= BIT(19);
+		rtl8xxxu_write32(priv, REG_OFDM_RX_DFIR, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_OFDM_RX_DFIR);
+		val32 &= ~GENMASK(23, 20);
+		val32 |= BIT(23);
+		rtl8xxxu_write32(priv, REG_OFDM_RX_DFIR, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_OFDM_RX_DFIR);
+		val32 &= ~GENMASK(27, 24);
+		val32 |= BIT(27) | BIT(25);
+		rtl8xxxu_write32(priv, REG_OFDM_RX_DFIR, val32);
+
+		/* RF TRX_BW */
+		val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_MODE_AG);
+		val32 &= ~MODE_AG_BW_MASK;
+		val32 |= MODE_AG_BW_20MHZ_8723B;
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_MODE_AG, val32);
+	}
+}
+
+static void rtl8710bu_init_aggregation(struct rtl8xxxu_priv *priv)
+{
+	u32 agg_rx;
+	u8 agg_ctrl;
+
+	/* RX aggregation */
+	agg_ctrl = rtl8xxxu_read8(priv, REG_TRXDMA_CTRL);
+	agg_ctrl &= ~TRXDMA_CTRL_RXDMA_AGG_EN;
+
+	agg_rx = rtl8xxxu_read32(priv, REG_RXDMA_AGG_PG_TH);
+	agg_rx &= ~RXDMA_USB_AGG_ENABLE;
+	agg_rx &= ~0xFF0F; /* reset agg size and timeout */
+
+	rtl8xxxu_write8(priv, REG_TRXDMA_CTRL, agg_ctrl);
+	rtl8xxxu_write32(priv, REG_RXDMA_AGG_PG_TH, agg_rx);
+}
+
+static void rtl8710bu_init_statistics(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+
+	/* Time duration for NHM unit: 4us, 0xc350=200ms */
+	rtl8xxxu_write16(priv, REG_NHM_TIMER_8723B + 2, 0xc350);
+	rtl8xxxu_write16(priv, REG_NHM_TH9_TH10_8723B + 2, 0xffff);
+	rtl8xxxu_write32(priv, REG_NHM_TH3_TO_TH0_8723B, 0xffffff50);
+	rtl8xxxu_write32(priv, REG_NHM_TH7_TO_TH4_8723B, 0xffffffff);
+
+	/* TH8 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 |= 0xff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* Enable CCK */
+	val32 = rtl8xxxu_read32(priv, REG_NHM_TH9_TH10_8723B);
+	val32 &= ~(BIT(8) | BIT(9) | BIT(10));
+	val32 |= BIT(8);
+	rtl8xxxu_write32(priv, REG_NHM_TH9_TH10_8723B, val32);
+
+	/* Max power amongst all RX antennas */
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_FA_RSTC);
+	val32 |= BIT(7);
+	rtl8xxxu_write32(priv, REG_OFDM0_FA_RSTC, val32);
+}
+
+static int rtl8710b_read_efuse(struct rtl8xxxu_priv *priv)
+{
+	struct device *dev = &priv->udev->dev;
+	u8 val8, word_mask, header, extheader;
+	u16 efuse_addr, offset;
+	int i, ret = 0;
+	u32 val32;
+
+	val32 = rtl8710b_read_syson_reg(priv, REG_SYS_EEPROM_CTRL0_8710B);
+	priv->boot_eeprom = u32_get_bits(val32, EEPROM_BOOT);
+	priv->has_eeprom = u32_get_bits(val32, EEPROM_ENABLE);
+
+	/* Default value is 0xff */
+	memset(priv->efuse_wifi.raw, 0xff, EFUSE_MAP_LEN);
+
+	efuse_addr = 0;
+	while (efuse_addr < EFUSE_REAL_CONTENT_LEN_8723A) {
+		u16 map_addr;
+
+		ret = rtl8710b_read_efuse8(priv, efuse_addr++, &header);
+		if (ret || header == 0xff)
+			goto exit;
+
+		if ((header & 0x1f) == 0x0f) {	/* extended header */
+			offset = (header & 0xe0) >> 5;
+
+			ret = rtl8710b_read_efuse8(priv, efuse_addr++, &extheader);
+			if (ret)
+				goto exit;
+
+			/* All words disabled */
+			if ((extheader & 0x0f) == 0x0f)
+				continue;
+
+			offset |= ((extheader & 0xf0) >> 1);
+			word_mask = extheader & 0x0f;
+		} else {
+			offset = (header >> 4) & 0x0f;
+			word_mask = header & 0x0f;
+		}
+
+		/* Get word enable value from PG header */
+
+		/* We have 8 bits to indicate validity */
+		map_addr = offset * 8;
+		for (i = 0; i < EFUSE_MAX_WORD_UNIT; i++) {
+			/* Check word enable condition in the section */
+			if (word_mask & BIT(i)) {
+				map_addr += 2;
+				continue;
+			}
+
+			ret = rtl8710b_read_efuse8(priv, efuse_addr++, &val8);
+			if (ret)
+				goto exit;
+			if (map_addr >= EFUSE_MAP_LEN - 1) {
+				dev_warn(dev, "%s: Illegal map_addr (%04x), efuse corrupt!\n",
+					 __func__, map_addr);
+				ret = -EINVAL;
+				goto exit;
+			}
+			priv->efuse_wifi.raw[map_addr++] = val8;
+
+			ret = rtl8710b_read_efuse8(priv, efuse_addr++, &val8);
+			if (ret)
+				goto exit;
+			priv->efuse_wifi.raw[map_addr++] = val8;
+		}
+	}
+
+exit:
+
+	return ret;
+}
+
+static int rtl8710bu_parse_efuse(struct rtl8xxxu_priv *priv)
+{
+	struct rtl8710bu_efuse *efuse = &priv->efuse_wifi.efuse8710bu;
+
+	if (efuse->rtl_id != cpu_to_le16(0x8195))
+		return -EINVAL;
+
+	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
+
+	memcpy(priv->cck_tx_power_index_A, efuse->tx_power_index_A.cck_base,
+	       sizeof(efuse->tx_power_index_A.cck_base));
+
+	memcpy(priv->ht40_1s_tx_power_index_A,
+	       efuse->tx_power_index_A.ht40_base,
+	       sizeof(efuse->tx_power_index_A.ht40_base));
+
+	priv->ofdm_tx_power_diff[0].a = efuse->tx_power_index_A.ht20_ofdm_1s_diff.a;
+	priv->ht20_tx_power_diff[0].a = efuse->tx_power_index_A.ht20_ofdm_1s_diff.b;
+
+	priv->default_crystal_cap = efuse->xtal_k & 0x3f;
+
+	return 0;
+}
+
+static int rtl8710bu_load_firmware(struct rtl8xxxu_priv *priv)
+{
+	if (priv->vendor_smic) {
+		return rtl8xxxu_load_firmware(priv, "rtlwifi/rtl8710bufw_SMIC.bin");
+	} else if (priv->vendor_umc) {
+		return rtl8xxxu_load_firmware(priv, "rtlwifi/rtl8710bufw_UMC.bin");
+	} else {
+		dev_err(&priv->udev->dev, "We have no suitable firmware for this chip.\n");
+		return -1;
+	}
+}
+
+static void rtl8710bu_init_phy_bb(struct rtl8xxxu_priv *priv)
+{
+	const struct rtl8xxxu_reg32val *phy_init_table;
+	u32 val32;
+
+	/* Enable BB and RF */
+	val32 = rtl8xxxu_read32(priv, REG_SYS_FUNC_8710B);
+	val32 |= GENMASK(17, 16) | GENMASK(26, 24);
+	rtl8xxxu_write32(priv, REG_SYS_FUNC_8710B, val32);
+
+	if (priv->package_type == PACKAGE_QFN48M_U)
+		phy_init_table = rtl8710bu_qfn48m_u_phy_init_table;
+	else
+		phy_init_table = rtl8710bu_qfn48m_s_phy_init_table;
+
+	rtl8xxxu_init_phy_regs(priv, phy_init_table);
+
+	rtl8xxxu_init_phy_regs(priv, rtl8710b_agc_table);
+}
+
+static int rtl8710bu_init_phy_rf(struct rtl8xxxu_priv *priv)
+{
+	const struct rtl8xxxu_rfregval *radioa_init_table;
+
+	if (priv->package_type == PACKAGE_QFN48M_U)
+		radioa_init_table = rtl8710bu_qfn48m_u_radioa_init_table;
+	else
+		radioa_init_table = rtl8710bu_qfn48m_s_radioa_init_table;
+
+	return rtl8xxxu_init_phy_rf(priv, radioa_init_table, RF_A);
+}
+
+static int rtl8710bu_iqk_path_a(struct rtl8xxxu_priv *priv, u32 *lok_result)
+{
+	u32 reg_eac, reg_e94, reg_e9c, val32, path_sel_bb;
+	int result = 0;
+
+	path_sel_bb = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x99000000);
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/*
+	 * Enable path A PA in TX IQK mode
+	 */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x20000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0000f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0x07ff7);
+
+	/* PA,PAD gain adjust */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF);
+	val32 |= BIT(11);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, val32);
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_56);
+	u32p_replace_bits(&val32, 0x1ed, 0x00fff);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_56, val32);
+
+	/* enter IQK mode */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0x808000, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* path-A IQK setting */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x18008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x38008c1c);
+
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_A, 0x821403ff);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_A, 0x28160c06);
+
+	/* LO calibration setting */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_RSP, 0x02002911);
+
+	/* One shot, path A LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xfa000000);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000);
+
+	mdelay(10);
+
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel_bb);
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF);
+	val32 &= ~BIT(11);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, val32);
+
+	/* save LOK result */
+	*lok_result = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_TXM_IDAC);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_e94 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_A);
+	reg_e9c = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_A);
+
+	if (!(reg_eac & BIT(28)) &&
+	    ((reg_e94 & 0x03ff0000) != 0x01420000) &&
+	    ((reg_e9c & 0x03ff0000) != 0x00420000))
+		result |= 0x01;
+
+	return result;
+}
+
+static int rtl8710bu_rx_iqk_path_a(struct rtl8xxxu_priv *priv, u32 lok_result)
+{
+	u32 reg_ea4, reg_eac, reg_e94, reg_e9c, val32, path_sel_bb, tmp;
+	int result = 0;
+
+	path_sel_bb = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x99000000);
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* modify RXIQK mode table */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x30000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0000f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xf1173);
+
+	/* PA,PAD gain adjust */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF);
+	val32 |= BIT(11);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, val32);
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_56);
+	u32p_replace_bits(&val32, 0xf, 0x003e0);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_56, val32);
+
+	/*
+	 * Enter IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0x808000, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* path-A IQK setting */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x18008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x38008c1c);
+
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_A, 0x8216129f);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_A, 0x28160c00);
+
+	/*
+	 * Tx IQK setting
+	 */
+	rtl8xxxu_write32(priv, REG_TX_IQK, 0x01007c00);
+	rtl8xxxu_write32(priv, REG_RX_IQK, 0x01004800);
+
+	/* LO calibration setting */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_RSP, 0x0046a911);
+
+	/* One shot, path A LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf9000000);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000);
+
+	mdelay(10);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_e94 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_A);
+	reg_e9c = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_A);
+
+	if (!(reg_eac & BIT(28)) &&
+	    ((reg_e94 & 0x03ff0000) != 0x01420000) &&
+	    ((reg_e9c & 0x03ff0000) != 0x00420000)) {
+		result |= 0x01;
+	} else { /* If TX not OK, ignore RX */
+
+		/* reload RF path */
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel_bb);
+
+		/*
+		 * Leave IQK mode
+		 */
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+		u32p_replace_bits(&val32, 0, 0xffffff00);
+		rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+		val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF);
+		val32 &= ~BIT(11);
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, val32);
+
+		return result;
+	}
+
+	val32 = 0x80007c00 | (reg_e94 & 0x3ff0000) | ((reg_e9c & 0x3ff0000) >> 16);
+	rtl8xxxu_write32(priv, REG_TX_IQK, val32);
+
+	/*
+	 * Modify RX IQK mode table
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x30000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0000f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xf7ff2);
+
+	/*
+	 * PA, PAD setting
+	 */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF);
+	val32 |= BIT(11);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, val32);
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_56);
+	u32p_replace_bits(&val32, 0x2a, 0x00fff);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_56, val32);
+
+	/*
+	 * Enter IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0x808000, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/*
+	 * RX IQK setting
+	 */
+	rtl8xxxu_write32(priv, REG_RX_IQK, 0x01004800);
+
+	/* path-A IQK setting */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x18008c1c);
+
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_A, 0x2816169f);
+
+	/* LO calibration setting */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_RSP, 0x0046a911);
+
+	/* One shot, path A LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf9000000);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000);
+
+	mdelay(10);
+
+	/* reload RF path */
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel_bb);
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF);
+	val32 &= ~BIT(11);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, val32);
+
+	/* reload LOK value */
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXM_IDAC, lok_result);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_ea4 = rtl8xxxu_read32(priv, REG_RX_POWER_BEFORE_IQK_A_2);
+
+	tmp = (reg_eac & 0x03ff0000) >> 16;
+	if ((tmp & 0x200) > 0)
+		tmp = 0x400 - tmp;
+
+	if (!(reg_eac & BIT(27)) &&
+	    ((reg_ea4 & 0x03ff0000) != 0x01320000) &&
+	    ((reg_eac & 0x03ff0000) != 0x00360000) &&
+	    (((reg_ea4 & 0x03ff0000) >> 16) < 0x11a) &&
+	    (((reg_ea4 & 0x03ff0000) >> 16) > 0xe6) &&
+	    (tmp < 0x1a))
+		result |= 0x02;
+
+	return result;
+}
+
+static void rtl8710bu_phy_iqcalibrate(struct rtl8xxxu_priv *priv,
+				      int result[][8], int t)
+{
+	struct device *dev = &priv->udev->dev;
+	u32 i, val32, rx_initial_gain, lok_result;
+	u32 path_sel_bb, path_sel_rf;
+	int path_a_ok;
+	int retry = 2;
+	static const u32 adda_regs[RTL8XXXU_ADDA_REGS] = {
+		REG_FPGA0_XCD_SWITCH_CTRL, REG_BLUETOOTH,
+		REG_RX_WAIT_CCA, REG_TX_CCK_RFON,
+		REG_TX_CCK_BBON, REG_TX_OFDM_RFON,
+		REG_TX_OFDM_BBON, REG_TX_TO_RX,
+		REG_TX_TO_TX, REG_RX_CCK,
+		REG_RX_OFDM, REG_RX_WAIT_RIFS,
+		REG_RX_TO_RX, REG_STANDBY,
+		REG_SLEEP, REG_PMPD_ANAEN
+	};
+	static const u32 iqk_mac_regs[RTL8XXXU_MAC_REGS] = {
+		REG_TXPAUSE, REG_BEACON_CTRL,
+		REG_BEACON_CTRL_1, REG_GPIO_MUXCFG
+	};
+	static const u32 iqk_bb_regs[RTL8XXXU_BB_REGS] = {
+		REG_OFDM0_TRX_PATH_ENABLE, REG_OFDM0_TR_MUX_PAR,
+		REG_FPGA0_XCD_RF_SW_CTRL, REG_CONFIG_ANT_A, REG_CONFIG_ANT_B,
+		REG_FPGA0_XAB_RF_SW_CTRL, REG_FPGA0_XA_RF_INT_OE,
+		REG_FPGA0_XB_RF_INT_OE, REG_CCK0_AFE_SETTING
+	};
+
+	/*
+	 * Note: IQ calibration must be performed after loading
+	 *       PHY_REG.txt , and radio_a, radio_b.txt
+	 */
+
+	rx_initial_gain = rtl8xxxu_read32(priv, REG_OFDM0_XA_AGC_CORE1);
+
+	if (t == 0) {
+		/* Save ADDA parameters, turn Path A ADDA on */
+		rtl8xxxu_save_regs(priv, adda_regs, priv->adda_backup,
+				   RTL8XXXU_ADDA_REGS);
+		rtl8xxxu_save_mac_regs(priv, iqk_mac_regs, priv->mac_backup);
+		rtl8xxxu_save_regs(priv, iqk_bb_regs,
+				   priv->bb_backup, RTL8XXXU_BB_REGS);
+	}
+
+	rtl8xxxu_path_adda_on(priv, adda_regs, true);
+
+	if (t == 0) {
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_XA_HSSI_PARM1);
+		priv->pi_enabled = u32_get_bits(val32, FPGA0_HSSI_PARM1_PI);
+	}
+
+	if (!priv->pi_enabled) {
+		/* Switch BB to PI mode to do IQ Calibration */
+		rtl8xxxu_write32(priv, REG_FPGA0_XA_HSSI_PARM1, 0x01000100);
+		rtl8xxxu_write32(priv, REG_FPGA0_XB_HSSI_PARM1, 0x01000100);
+	}
+
+	/* MAC settings */
+	val32 = rtl8xxxu_read32(priv, REG_TX_PTCL_CTRL);
+	val32 |= 0x00ff0000;
+	rtl8xxxu_write32(priv, REG_TX_PTCL_CTRL, val32);
+
+	/* save RF path */
+	path_sel_bb = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+	path_sel_rf = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_S0S1);
+
+	/* BB setting */
+	val32 = rtl8xxxu_read32(priv, REG_CCK0_AFE_SETTING);
+	val32 |= 0x0f000000;
+	rtl8xxxu_write32(priv, REG_CCK0_AFE_SETTING, val32);
+	rtl8xxxu_write32(priv, REG_RX_WAIT_CCA, 0x03c00010);
+	rtl8xxxu_write32(priv, REG_OFDM0_TRX_PATH_ENABLE, 0x03a05601);
+	rtl8xxxu_write32(priv, REG_OFDM0_TR_MUX_PAR, 0x000800e4);
+	rtl8xxxu_write32(priv, REG_FPGA0_XCD_RF_SW_CTRL, 0x25204000);
+
+	/* IQ calibration setting */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0x808000, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+	rtl8xxxu_write32(priv, REG_TX_IQK, 0x01007c00);
+	rtl8xxxu_write32(priv, REG_RX_IQK, 0x01004800);
+
+	for (i = 0; i < retry; i++) {
+		path_a_ok = rtl8710bu_iqk_path_a(priv, &lok_result);
+
+		if (path_a_ok == 0x01) {
+			val32 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_A);
+			result[t][0] = (val32 >> 16) & 0x3ff;
+
+			val32 = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_A);
+			result[t][1] = (val32 >> 16) & 0x3ff;
+			break;
+		} else {
+			result[t][0] = 0x100;
+			result[t][1] = 0x0;
+		}
+	}
+
+	for (i = 0; i < retry; i++) {
+		path_a_ok = rtl8710bu_rx_iqk_path_a(priv, lok_result);
+
+		if (path_a_ok == 0x03) {
+			val32 = rtl8xxxu_read32(priv, REG_RX_POWER_BEFORE_IQK_A_2);
+			result[t][2] = (val32 >> 16) & 0x3ff;
+
+			val32 = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+			result[t][3] = (val32 >> 16) & 0x3ff;
+			break;
+		} else {
+			result[t][2] = 0x100;
+			result[t][3] = 0x0;
+		}
+	}
+
+	if (!path_a_ok)
+		dev_warn(dev, "%s: Path A IQK failed!\n", __func__);
+
+	/* Back to BB mode, load original value */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	u32p_replace_bits(&val32, 0, 0xffffff00);
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	if (t == 0)
+		return;
+
+	/* Reload ADDA power saving parameters */
+	rtl8xxxu_restore_regs(priv, adda_regs, priv->adda_backup, RTL8XXXU_ADDA_REGS);
+
+	/* Reload MAC parameters */
+	rtl8xxxu_restore_mac_regs(priv, iqk_mac_regs, priv->mac_backup);
+
+	/* Reload BB parameters */
+	rtl8xxxu_restore_regs(priv, iqk_bb_regs, priv->bb_backup, RTL8XXXU_BB_REGS);
+
+	/* Reload RF path */
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel_bb);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_S0S1, path_sel_rf);
+
+	/* Restore RX initial gain */
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_XA_AGC_CORE1);
+	u32p_replace_bits(&val32, 0x50, 0x000000ff);
+	rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, val32);
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_XA_AGC_CORE1);
+	u32p_replace_bits(&val32, rx_initial_gain & 0xff, 0x000000ff);
+	rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, val32);
+
+	/* Load 0xe30 IQC default value */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x01008c00);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x01008c00);
+}
+
+static void rtl8710bu_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
+{
+	struct device *dev = &priv->udev->dev;
+	int result[4][8]; /* last is final result */
+	int i, candidate;
+	bool path_a_ok;
+	s32 reg_e94, reg_e9c, reg_ea4, reg_eac;
+	s32 reg_tmp = 0;
+	bool simu;
+	u32 path_sel_bb;
+
+	/* Save RF path */
+	path_sel_bb = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+
+	memset(result, 0, sizeof(result));
+	candidate = -1;
+
+	path_a_ok = false;
+
+	for (i = 0; i < 3; i++) {
+		rtl8710bu_phy_iqcalibrate(priv, result, i);
+
+		if (i == 1) {
+			simu = rtl8xxxu_gen2_simularity_compare(priv, result, 0, 1);
+			if (simu) {
+				candidate = 0;
+				break;
+			}
+		}
+
+		if (i == 2) {
+			simu = rtl8xxxu_gen2_simularity_compare(priv, result, 0, 2);
+			if (simu) {
+				candidate = 0;
+				break;
+			}
+
+			simu = rtl8xxxu_gen2_simularity_compare(priv, result, 1, 2);
+			if (simu) {
+				candidate = 1;
+			} else {
+				for (i = 0; i < 8; i++)
+					reg_tmp += result[3][i];
+
+				if (reg_tmp)
+					candidate = 3;
+				else
+					candidate = -1;
+			}
+		}
+	}
+
+	if (candidate >= 0) {
+		reg_e94 = result[candidate][0];
+		reg_e9c = result[candidate][1];
+		reg_ea4 = result[candidate][2];
+		reg_eac = result[candidate][3];
+
+		dev_dbg(dev, "%s: candidate is %x\n", __func__, candidate);
+		dev_dbg(dev, "%s: e94=%x e9c=%x ea4=%x eac=%x\n",
+			__func__, reg_e94, reg_e9c, reg_ea4, reg_eac);
+
+		path_a_ok = true;
+
+		if (reg_e94)
+			rtl8xxxu_fill_iqk_matrix_a(priv, path_a_ok, result,
+						   candidate, (reg_ea4 == 0));
+	}
+
+	rtl8xxxu_save_regs(priv, rtl8xxxu_iqk_phy_iq_bb_reg,
+			   priv->bb_recovery_backup, RTL8XXXU_BB_REGS);
+
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel_bb);
+}
+
+static int rtl8710b_emu_to_active(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	int count, ret = 0;
+
+	/* AFE power mode selection: 1: LDO mode, 0: Power-cut mode */
+	val8 = rtl8xxxu_read8(priv, 0x5d);
+	val8 &= ~BIT(0);
+	rtl8xxxu_write8(priv, 0x5d, val8);
+
+	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC_8710B);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, REG_SYS_FUNC_8710B, val8);
+
+	rtl8xxxu_write8(priv, 0x56, 0x0e);
+
+	val8 = rtl8xxxu_read8(priv, 0x20);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, 0x20, val8);
+
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val8 = rtl8xxxu_read8(priv, 0x20);
+		if (!(val8 & BIT(0)))
+			break;
+
+		udelay(10);
+	}
+
+	if (!count)
+		ret = -EBUSY;
+
+	return ret;
+}
+
+static int rtl8710bu_active_to_emu(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u32 val32;
+	int count, ret = 0;
+
+	/* Turn off RF */
+	val32 = rtl8xxxu_read32(priv, REG_SYS_FUNC_8710B);
+	val32 &= ~GENMASK(26, 24);
+	rtl8xxxu_write32(priv, REG_SYS_FUNC_8710B, val32);
+
+	/* BB reset */
+	val32 = rtl8xxxu_read32(priv, REG_SYS_FUNC_8710B);
+	val32 &= ~GENMASK(17, 16);
+	rtl8xxxu_write32(priv, REG_SYS_FUNC_8710B, val32);
+
+	/* Turn off MAC by HW state machine */
+	val8 = rtl8xxxu_read8(priv, 0x20);
+	val8 |= BIT(1);
+	rtl8xxxu_write8(priv, 0x20, val8);
+
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val8 = rtl8xxxu_read8(priv, 0x20);
+		if ((val8 & BIT(1)) == 0) {
+			ret = 0;
+			break;
+		}
+		udelay(10);
+	}
+
+	if (!count)
+		ret = -EBUSY;
+
+	return ret;
+}
+
+static int rtl8710bu_active_to_lps(struct rtl8xxxu_priv *priv)
+{
+	struct device *dev = &priv->udev->dev;
+	u8 val8;
+	u16 val16;
+	u32 val32;
+	int retry, retval;
+
+	/* Tx Pause */
+	rtl8xxxu_write8(priv, REG_TXPAUSE, 0xff);
+
+	retry = 100;
+	retval = -EBUSY;
+	/*
+	 * Poll 32 bit wide REG_SCH_TX_CMD for 0x00000000 to ensure no TX is pending.
+	 */
+	do {
+		val32 = rtl8xxxu_read32(priv, REG_SCH_TX_CMD);
+		if (!val32) {
+			retval = 0;
+			break;
+		}
+		udelay(10);
+	} while (retry--);
+
+	if (!retry) {
+		dev_warn(dev, "Failed to flush TX queue\n");
+		retval = -EBUSY;
+		return retval;
+	}
+
+	/* Disable CCK and OFDM, clock gated */
+	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC);
+	val8 &= ~SYS_FUNC_BBRSTB;
+	rtl8xxxu_write8(priv, REG_SYS_FUNC, val8);
+
+	udelay(2);
+
+	/* Whole BB is reset */
+	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC);
+	val8 &= ~SYS_FUNC_BB_GLB_RSTN;
+	rtl8xxxu_write8(priv, REG_SYS_FUNC, val8);
+
+	/* Reset MAC TRX */
+	val16 = rtl8xxxu_read16(priv, REG_CR);
+	val16 &= 0xff00;
+	val16 |= CR_HCI_RXDMA_ENABLE | CR_HCI_TXDMA_ENABLE;
+	val16 &= ~CR_SECURITY_ENABLE;
+	rtl8xxxu_write16(priv, REG_CR, val16);
+
+	/* Respond TxOK to scheduler */
+	val8 = rtl8xxxu_read8(priv, REG_DUAL_TSF_RST);
+	val8 |= DUAL_TSF_TX_OK;
+	rtl8xxxu_write8(priv, REG_DUAL_TSF_RST, val8);
+
+	return retval;
+}
+
+static int rtl8710bu_power_on(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+	u16 val16;
+	u8 val8;
+	int ret;
+
+	rtl8xxxu_write8(priv, REG_USB_ACCESS_TIMEOUT, 0x80);
+
+	val8 = rtl8xxxu_read8(priv, REG_SYS_ISO_CTRL);
+	val8 &= ~BIT(5);
+	rtl8xxxu_write8(priv, REG_SYS_ISO_CTRL, val8);
+
+	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC_8710B);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, REG_SYS_FUNC_8710B, val8);
+
+	val8 = rtl8xxxu_read8(priv, 0x20);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, 0x20, val8);
+
+	rtl8xxxu_write8(priv, REG_AFE_CTRL_8710B, 0);
+
+	val8 = rtl8xxxu_read8(priv, REG_WL_STATUS_8710B);
+	val8 |= BIT(1);
+	rtl8xxxu_write8(priv, REG_WL_STATUS_8710B, val8);
+
+	ret = rtl8710b_emu_to_active(priv);
+	if (ret)
+		return ret;
+
+	rtl8xxxu_write16(priv, REG_CR, 0);
+
+	val16 = rtl8xxxu_read16(priv, REG_CR);
+
+	val16 |= CR_HCI_TXDMA_ENABLE | CR_HCI_RXDMA_ENABLE |
+		 CR_TXDMA_ENABLE | CR_RXDMA_ENABLE |
+		 CR_PROTOCOL_ENABLE | CR_SCHEDULE_ENABLE |
+		 CR_SECURITY_ENABLE | CR_CALTIMER_ENABLE;
+	rtl8xxxu_write16(priv, REG_CR, val16);
+
+	/* Enable hardware sequence number. */
+	val8 = rtl8xxxu_read8(priv, REG_HWSEQ_CTRL);
+	val8 |= 0x7f;
+	rtl8xxxu_write8(priv, REG_HWSEQ_CTRL, val8);
+
+	udelay(2);
+
+	/*
+	 * Technically the rest was in the rtl8710bu_hal_init function,
+	 * not the power_on function, but it's fine because we only
+	 * call power_on from init_device.
+	 */
+
+	val8 = rtl8xxxu_read8(priv, 0xfef9);
+	val8 &= ~BIT(0);
+	rtl8xxxu_write8(priv, 0xfef9, val8);
+
+	/* Clear the 0x40000138[5] to prevent CM4 Suspend */
+	val32 = rtl8710b_read_syson_reg(priv, 0x138);
+	val32 &= ~BIT(5);
+	rtl8710b_write_syson_reg(priv, 0x138, val32);
+
+	return ret;
+}
+
+static void rtl8710bu_power_off(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+	u8 val8;
+
+	rtl8xxxu_flush_fifo(priv);
+
+	rtl8xxxu_write32(priv, REG_HISR0_8710B, 0xffffffff);
+	rtl8xxxu_write32(priv, REG_HIMR0_8710B, 0x0);
+
+	/* Set the 0x40000138[5] to allow CM4 Suspend */
+	val32 = rtl8710b_read_syson_reg(priv, 0x138);
+	val32 |= BIT(5);
+	rtl8710b_write_syson_reg(priv, 0x138, val32);
+
+	/* Stop rx */
+	rtl8xxxu_write8(priv, REG_CR, 0x00);
+
+	rtl8710bu_active_to_lps(priv);
+
+	/* Reset MCU ? */
+	val8 = rtl8xxxu_read8(priv, REG_8051FW_CTRL_V1_8710B + 3);
+	val8 &= ~BIT(0);
+	rtl8xxxu_write8(priv, REG_8051FW_CTRL_V1_8710B + 3, val8);
+
+	/* Reset MCU ready status */
+	rtl8xxxu_write8(priv, REG_8051FW_CTRL_V1_8710B, 0x00);
+
+	rtl8710bu_active_to_emu(priv);
+}
+
+static void rtl8710b_reset_8051(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+
+	val8 = rtl8xxxu_read8(priv, REG_8051FW_CTRL_V1_8710B + 3);
+	val8 &= ~BIT(0);
+	rtl8xxxu_write8(priv, REG_8051FW_CTRL_V1_8710B + 3, val8);
+
+	udelay(50);
+
+	val8 = rtl8xxxu_read8(priv, REG_8051FW_CTRL_V1_8710B + 3);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, REG_8051FW_CTRL_V1_8710B + 3, val8);
+}
+
+static void rtl8710b_enable_rf(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+
+	rtl8xxxu_write8(priv, REG_RF_CTRL, RF_ENABLE | RF_RSTB | RF_SDMRSTB);
+
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_TRX_PATH_ENABLE);
+	val32 &= ~(OFDM_RF_PATH_RX_MASK | OFDM_RF_PATH_TX_MASK);
+	val32 |= OFDM_RF_PATH_RX_A | OFDM_RF_PATH_TX_A;
+	rtl8xxxu_write32(priv, REG_OFDM0_TRX_PATH_ENABLE, val32);
+
+	rtl8xxxu_write8(priv, REG_TXPAUSE, 0x00);
+}
+
+static void rtl8710b_disable_rf(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_TRX_PATH_ENABLE);
+	val32 &= ~OFDM_RF_PATH_TX_MASK;
+	rtl8xxxu_write32(priv, REG_OFDM0_TRX_PATH_ENABLE, val32);
+
+	/* Power down RF module */
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_AC, 0);
+}
+
+static void rtl8710b_usb_quirks(struct rtl8xxxu_priv *priv)
+{
+	u16 val16;
+
+	rtl8xxxu_gen2_usb_quirks(priv);
+
+	val16 = rtl8xxxu_read16(priv, REG_CR);
+	val16 |= (CR_MAC_TX_ENABLE | CR_MAC_RX_ENABLE);
+	rtl8xxxu_write16(priv, REG_CR, val16);
+}
+
+#define XTAL1	GENMASK(29, 24)
+#define XTAL0	GENMASK(23, 18)
+
+static void rtl8710b_set_crystal_cap(struct rtl8xxxu_priv *priv, u8 crystal_cap)
+{
+	struct rtl8xxxu_cfo_tracking *cfo = &priv->cfo_tracking;
+	u32 val32;
+
+	if (crystal_cap == cfo->crystal_cap)
+		return;
+
+	val32 = rtl8710b_read_syson_reg(priv, REG_SYS_XTAL_CTRL0_8710B);
+
+	dev_dbg(&priv->udev->dev,
+		"%s: Adjusting crystal cap from 0x%x (actually 0x%x 0x%x) to 0x%x\n",
+		__func__,
+		cfo->crystal_cap,
+		u32_get_bits(val32, XTAL1),
+		u32_get_bits(val32, XTAL0),
+		crystal_cap);
+
+	u32p_replace_bits(&val32, crystal_cap, XTAL1);
+	u32p_replace_bits(&val32, crystal_cap, XTAL0);
+	rtl8710b_write_syson_reg(priv, REG_SYS_XTAL_CTRL0_8710B, val32);
+
+	cfo->crystal_cap = crystal_cap;
+}
+
+static s8 rtl8710b_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats)
+{
+	struct jaguar2_phy_stats_type0 *phy_stats0 = (struct jaguar2_phy_stats_type0 *)phy_stats;
+	u8 lna_idx = (phy_stats0->lna_h << 3) | phy_stats0->lna_l;
+	u8 vga_idx = phy_stats0->vga;
+	s8 rx_pwr_all = 0x00;
+
+	switch (lna_idx) {
+	case 7:
+		rx_pwr_all = -52 - (2 * vga_idx);
+		break;
+	case 6:
+		rx_pwr_all = -42 - (2 * vga_idx);
+		break;
+	case 5:
+		rx_pwr_all = -36 - (2 * vga_idx);
+		break;
+	case 3:
+		rx_pwr_all = -12 - (2 * vga_idx);
+		break;
+	case 2:
+		rx_pwr_all = 0 - (2 * vga_idx);
+		break;
+	default:
+		rx_pwr_all = 0;
+		break;
+	}
+
+	return rx_pwr_all;
+}
+
+struct rtl8xxxu_fileops rtl8710bu_fops = {
+	.identify_chip = rtl8710bu_identify_chip,
+	.parse_efuse = rtl8710bu_parse_efuse,
+	.load_firmware = rtl8710bu_load_firmware,
+	.power_on = rtl8710bu_power_on,
+	.power_off = rtl8710bu_power_off,
+	.read_efuse = rtl8710b_read_efuse,
+	.reset_8051 = rtl8710b_reset_8051,
+	.llt_init = rtl8xxxu_auto_llt_table,
+	.init_phy_bb = rtl8710bu_init_phy_bb,
+	.init_phy_rf = rtl8710bu_init_phy_rf,
+	.phy_lc_calibrate = rtl8188f_phy_lc_calibrate,
+	.phy_iq_calibrate = rtl8710bu_phy_iq_calibrate,
+	.config_channel = rtl8710bu_config_channel,
+	.parse_rx_desc = rtl8xxxu_parse_rxdesc24,
+	.parse_phystats = jaguar2_rx_parse_phystats,
+	.init_aggregation = rtl8710bu_init_aggregation,
+	.init_statistics = rtl8710bu_init_statistics,
+	.init_burst = rtl8xxxu_init_burst,
+	.enable_rf = rtl8710b_enable_rf,
+	.disable_rf = rtl8710b_disable_rf,
+	.usb_quirks = rtl8710b_usb_quirks,
+	.set_tx_power = rtl8188f_set_tx_power,
+	.update_rate_mask = rtl8xxxu_gen2_update_rate_mask,
+	.report_connect = rtl8xxxu_gen2_report_connect,
+	.report_rssi = rtl8xxxu_gen2_report_rssi,
+	.fill_txdesc = rtl8xxxu_fill_txdesc_v2,
+	.set_crystal_cap = rtl8710b_set_crystal_cap,
+	.cck_rssi = rtl8710b_cck_rssi,
+	.writeN_block_size = 4,
+	.rx_desc_size = sizeof(struct rtl8xxxu_rxdesc24),
+	.tx_desc_size = sizeof(struct rtl8xxxu_txdesc40),
+	.has_tx_report = 1,
+	.gen2_thermal_meter = 1,
+	.needs_full_init = 1,
+	.adda_1t_init = 0x03c00016,
+	.adda_1t_path_on = 0x03c00016,
+	.trxff_boundary = 0x3f7f,
+	.pbp_rx = PBP_PAGE_SIZE_256,
+	.pbp_tx = PBP_PAGE_SIZE_256,
+	.mactable = rtl8710b_mac_init_table,
+	.total_page_num = TX_TOTAL_PAGE_NUM_8723B,
+	.page_num_hi = TX_PAGE_NUM_HI_PQ_8723B,
+	.page_num_lo = TX_PAGE_NUM_LO_PQ_8723B,
+	.page_num_norm = TX_PAGE_NUM_NORM_PQ_8723B,
+};
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c
index 5e7b58d395ba..d219be19d07f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c
@@ -435,8 +435,9 @@ void rtl8723a_set_crystal_cap(struct rtl8xxxu_priv *priv, u8 crystal_cap)
 	cfo->crystal_cap = crystal_cap;
 }
 
-s8 rtl8723a_cck_rssi(struct rtl8xxxu_priv *priv, u8 cck_agc_rpt)
+s8 rtl8723a_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats)
 {
+	u8 cck_agc_rpt = phy_stats->cck_agc_rpt_ofdm_cfosho_a;
 	s8 rx_pwr_all = 0x00;
 
 	switch (cck_agc_rpt & 0xc0) {
@@ -487,6 +488,7 @@ struct rtl8xxxu_fileops rtl8723au_fops = {
 	.load_firmware = rtl8723au_load_firmware,
 	.power_on = rtl8723au_power_on,
 	.power_off = rtl8xxxu_power_off,
+	.read_efuse = rtl8xxxu_read_efuse,
 	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.init_phy_bb = rtl8xxxu_gen1_init_phy_bb,
@@ -495,6 +497,7 @@ struct rtl8xxxu_fileops rtl8723au_fops = {
 	.phy_iq_calibrate = rtl8xxxu_gen1_phy_iq_calibrate,
 	.config_channel = rtl8xxxu_gen1_config_channel,
 	.parse_rx_desc = rtl8xxxu_parse_rxdesc16,
+	.parse_phystats = rtl8723au_rx_parse_phystats,
 	.init_aggregation = rtl8xxxu_gen1_init_aggregation,
 	.enable_rf = rtl8xxxu_gen1_enable_rf,
 	.disable_rf = rtl8xxxu_gen1_disable_rf,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
index 21613d60dc22..d99538eb8398 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
@@ -1675,8 +1675,9 @@ static void rtl8723bu_init_statistics(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write32(priv, REG_OFDM0_FA_RSTC, val32);
 }
 
-static s8 rtl8723b_cck_rssi(struct rtl8xxxu_priv *priv, u8 cck_agc_rpt)
+static s8 rtl8723b_cck_rssi(struct rtl8xxxu_priv *priv, struct rtl8723au_phy_stats *phy_stats)
 {
+	u8 cck_agc_rpt = phy_stats->cck_agc_rpt_ofdm_cfosho_a;
 	s8 rx_pwr_all = 0x00;
 	u8 vga_idx, lna_idx;
 
@@ -1709,6 +1710,7 @@ struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.load_firmware = rtl8723bu_load_firmware,
 	.power_on = rtl8723bu_power_on,
 	.power_off = rtl8723bu_power_off,
+	.read_efuse = rtl8xxxu_read_efuse,
 	.reset_8051 = rtl8723bu_reset_8051,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.init_phy_bb = rtl8723bu_init_phy_bb,
@@ -1718,6 +1720,7 @@ struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8xxxu_gen2_config_channel,
 	.parse_rx_desc = rtl8xxxu_parse_rxdesc24,
+	.parse_phystats = rtl8723au_rx_parse_phystats,
 	.init_aggregation = rtl8723bu_init_aggregation,
 	.init_statistics = rtl8723bu_init_statistics,
 	.init_burst = rtl8xxxu_init_burst,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 280dc3e26ca7..c152b228606f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -54,6 +54,8 @@ MODULE_FIRMWARE("rtlwifi/rtl8192eu_nic.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8723bu_nic.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8723bu_bt.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8188fufw.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8710bufw_SMIC.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8710bufw_UMC.bin");
 
 module_param_named(debug, rtl8xxxu_debug, int, 0600);
 MODULE_PARM_DESC(debug, "Set debug mask");
@@ -654,6 +656,9 @@ u8 rtl8xxxu_read8(struct rtl8xxxu_priv *priv, u16 addr)
 	int len;
 	u8 data;
 
+	if (priv->rtl_chip == RTL8710B && addr <= 0xff)
+		addr |= 0x8000;
+
 	mutex_lock(&priv->usb_buf_mutex);
 	len = usb_control_msg(udev, usb_rcvctrlpipe(udev, 0),
 			      REALTEK_USB_CMD_REQ, REALTEK_USB_READ,
@@ -674,6 +679,9 @@ u16 rtl8xxxu_read16(struct rtl8xxxu_priv *priv, u16 addr)
 	int len;
 	u16 data;
 
+	if (priv->rtl_chip == RTL8710B && addr <= 0xff)
+		addr |= 0x8000;
+
 	mutex_lock(&priv->usb_buf_mutex);
 	len = usb_control_msg(udev, usb_rcvctrlpipe(udev, 0),
 			      REALTEK_USB_CMD_REQ, REALTEK_USB_READ,
@@ -694,6 +702,9 @@ u32 rtl8xxxu_read32(struct rtl8xxxu_priv *priv, u16 addr)
 	int len;
 	u32 data;
 
+	if (priv->rtl_chip == RTL8710B && addr <= 0xff)
+		addr |= 0x8000;
+
 	mutex_lock(&priv->usb_buf_mutex);
 	len = usb_control_msg(udev, usb_rcvctrlpipe(udev, 0),
 			      REALTEK_USB_CMD_REQ, REALTEK_USB_READ,
@@ -713,6 +724,9 @@ int rtl8xxxu_write8(struct rtl8xxxu_priv *priv, u16 addr, u8 val)
 	struct usb_device *udev = priv->udev;
 	int ret;
 
+	if (priv->rtl_chip == RTL8710B && addr <= 0xff)
+		addr |= 0x8000;
+
 	mutex_lock(&priv->usb_buf_mutex);
 	priv->usb_buf.val8 = val;
 	ret = usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
@@ -733,6 +747,9 @@ int rtl8xxxu_write16(struct rtl8xxxu_priv *priv, u16 addr, u16 val)
 	struct usb_device *udev = priv->udev;
 	int ret;
 
+	if (priv->rtl_chip == RTL8710B && addr <= 0xff)
+		addr |= 0x8000;
+
 	mutex_lock(&priv->usb_buf_mutex);
 	priv->usb_buf.val16 = cpu_to_le16(val);
 	ret = usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
@@ -752,6 +769,9 @@ int rtl8xxxu_write32(struct rtl8xxxu_priv *priv, u16 addr, u32 val)
 	struct usb_device *udev = priv->udev;
 	int ret;
 
+	if (priv->rtl_chip == RTL8710B && addr <= 0xff)
+		addr |= 0x8000;
+
 	mutex_lock(&priv->usb_buf_mutex);
 	priv->usb_buf.val32 = cpu_to_le32(val);
 	ret = usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
@@ -1699,7 +1719,7 @@ rtl8xxxu_read_efuse8(struct rtl8xxxu_priv *priv, u16 offset, u8 *data)
 	return 0;
 }
 
-static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
+int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 {
 	struct device *dev = &priv->udev->dev;
 	int i, ret = 0;
@@ -1845,12 +1865,18 @@ void rtl8xxxu_reset_8051(struct rtl8xxxu_priv *priv)
 static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 {
 	struct device *dev = &priv->udev->dev;
+	u16 reg_mcu_fw_dl;
 	int ret = 0, i;
 	u32 val32;
 
+	if (priv->rtl_chip == RTL8710B)
+		reg_mcu_fw_dl = REG_8051FW_CTRL_V1_8710B;
+	else
+		reg_mcu_fw_dl = REG_MCU_FW_DL;
+
 	/* Poll checksum report */
 	for (i = 0; i < RTL8XXXU_FIRMWARE_POLL_MAX; i++) {
-		val32 = rtl8xxxu_read32(priv, REG_MCU_FW_DL);
+		val32 = rtl8xxxu_read32(priv, reg_mcu_fw_dl);
 		if (val32 & MCU_FW_DL_CSUM_REPORT)
 			break;
 	}
@@ -1861,10 +1887,10 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 		goto exit;
 	}
 
-	val32 = rtl8xxxu_read32(priv, REG_MCU_FW_DL);
+	val32 = rtl8xxxu_read32(priv, reg_mcu_fw_dl);
 	val32 |= MCU_FW_DL_READY;
 	val32 &= ~MCU_WINT_INIT_READY;
-	rtl8xxxu_write32(priv, REG_MCU_FW_DL, val32);
+	rtl8xxxu_write32(priv, reg_mcu_fw_dl, val32);
 
 	/*
 	 * Reset the 8051 in order for the firmware to start running,
@@ -1874,7 +1900,7 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 
 	/* Wait for firmware to become ready */
 	for (i = 0; i < RTL8XXXU_FIRMWARE_POLL_MAX; i++) {
-		val32 = rtl8xxxu_read32(priv, REG_MCU_FW_DL);
+		val32 = rtl8xxxu_read32(priv, reg_mcu_fw_dl);
 		if (val32 & MCU_WINT_INIT_READY)
 			break;
 
@@ -1890,7 +1916,7 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 	/*
 	 * Init H2C command
 	 */
-	if (priv->rtl_chip == RTL8723B || priv->rtl_chip == RTL8188F)
+	if (priv->rtl_chip == RTL8723B || priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B)
 		rtl8xxxu_write8(priv, REG_HMTFR, 0x0f);
 exit:
 	return ret;
@@ -1899,42 +1925,56 @@ exit:
 static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 {
 	int pages, remainder, i, ret;
+	u16 reg_mcu_fw_dl;
 	u8 val8;
 	u16 val16;
 	u32 val32;
 	u8 *fwptr;
 
-	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC + 1);
-	val8 |= 4;
-	rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, val8);
+	if (priv->rtl_chip == RTL8710B) {
+		reg_mcu_fw_dl = REG_8051FW_CTRL_V1_8710B;
+	} else {
+		reg_mcu_fw_dl = REG_MCU_FW_DL;
 
-	/* 8051 enable */
-	val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
-	val16 |= SYS_FUNC_CPU_ENABLE;
-	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+		val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC + 1);
+		val8 |= 4;
+		rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, val8);
+
+		/* 8051 enable */
+		val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+		val16 |= SYS_FUNC_CPU_ENABLE;
+		rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+	}
 
-	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
+	val8 = rtl8xxxu_read8(priv, reg_mcu_fw_dl);
 	if (val8 & MCU_FW_RAM_SEL) {
 		dev_info(&priv->udev->dev,
 			 "Firmware is already running, resetting the MCU.\n");
-		rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00);
+		rtl8xxxu_write8(priv, reg_mcu_fw_dl, 0x00);
 		priv->fops->reset_8051(priv);
 	}
 
 	/* MCU firmware download enable */
-	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
+	val8 = rtl8xxxu_read8(priv, reg_mcu_fw_dl);
 	val8 |= MCU_FW_DL_ENABLE;
-	rtl8xxxu_write8(priv, REG_MCU_FW_DL, val8);
+	rtl8xxxu_write8(priv, reg_mcu_fw_dl, val8);
 
 	/* 8051 reset */
-	val32 = rtl8xxxu_read32(priv, REG_MCU_FW_DL);
+	val32 = rtl8xxxu_read32(priv, reg_mcu_fw_dl);
 	val32 &= ~BIT(19);
-	rtl8xxxu_write32(priv, REG_MCU_FW_DL, val32);
+	rtl8xxxu_write32(priv, reg_mcu_fw_dl, val32);
+
+	if (priv->rtl_chip == RTL8710B) {
+		/* We must set 0x8090[8]=1 before download FW. */
+		val8 = rtl8xxxu_read8(priv, reg_mcu_fw_dl + 1);
+		val8 |= BIT(0);
+		rtl8xxxu_write8(priv, reg_mcu_fw_dl + 1, val8);
+	}
 
 	/* Reset firmware download checksum */
-	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
+	val8 = rtl8xxxu_read8(priv, reg_mcu_fw_dl);
 	val8 |= MCU_FW_DL_CSUM_REPORT;
-	rtl8xxxu_write8(priv, REG_MCU_FW_DL, val8);
+	rtl8xxxu_write8(priv, reg_mcu_fw_dl, val8);
 
 	pages = priv->fw_size / RTL_FW_PAGE_SIZE;
 	remainder = priv->fw_size % RTL_FW_PAGE_SIZE;
@@ -1942,9 +1982,9 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 	fwptr = priv->fw_data->data;
 
 	for (i = 0; i < pages; i++) {
-		val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL + 2) & 0xF8;
+		val8 = rtl8xxxu_read8(priv, reg_mcu_fw_dl + 2) & 0xF8;
 		val8 |= i;
-		rtl8xxxu_write8(priv, REG_MCU_FW_DL + 2, val8);
+		rtl8xxxu_write8(priv, reg_mcu_fw_dl + 2, val8);
 
 		ret = rtl8xxxu_writeN(priv, REG_FW_START_ADDRESS,
 				      fwptr, RTL_FW_PAGE_SIZE);
@@ -1957,9 +1997,9 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 	}
 
 	if (remainder) {
-		val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL + 2) & 0xF8;
+		val8 = rtl8xxxu_read8(priv, reg_mcu_fw_dl + 2) & 0xF8;
 		val8 |= i;
-		rtl8xxxu_write8(priv, REG_MCU_FW_DL + 2, val8);
+		rtl8xxxu_write8(priv, reg_mcu_fw_dl + 2, val8);
 		ret = rtl8xxxu_writeN(priv, REG_FW_START_ADDRESS,
 				      fwptr, remainder);
 		if (ret != remainder) {
@@ -1971,9 +2011,9 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 	ret = 0;
 fw_abort:
 	/* MCU firmware download disable */
-	val16 = rtl8xxxu_read16(priv, REG_MCU_FW_DL);
+	val16 = rtl8xxxu_read16(priv, reg_mcu_fw_dl);
 	val16 &= ~MCU_FW_DL_ENABLE;
-	rtl8xxxu_write16(priv, REG_MCU_FW_DL, val16);
+	rtl8xxxu_write16(priv, reg_mcu_fw_dl, val16);
 
 	return ret;
 }
@@ -2013,6 +2053,7 @@ int rtl8xxxu_load_firmware(struct rtl8xxxu_priv *priv, const char *fw_name)
 	case 0x5300:
 	case 0x2300:
 	case 0x88f0:
+	case 0x10b0:
 		break;
 	default:
 		ret = -EINVAL;
@@ -3823,7 +3864,7 @@ void rtl8xxxu_init_burst(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_HT_SINGLE_AMPDU_8723B, val8);
 
 	rtl8xxxu_write16(priv, REG_MAX_AGGR_NUM, 0x0c14);
-	if (priv->rtl_chip == RTL8723B)
+	if (priv->rtl_chip == RTL8723B || priv->rtl_chip == RTL8710B)
 		val8 = 0x5e;
 	else if (priv->rtl_chip == RTL8188F)
 		val8 = 0x70; /* 0x5e would make it very slow */
@@ -3831,13 +3872,17 @@ void rtl8xxxu_init_burst(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write32(priv, REG_AGGLEN_LMT, 0xffffffff);
 	rtl8xxxu_write8(priv, REG_RX_PKT_LIMIT, 0x18);
 	rtl8xxxu_write8(priv, REG_PIFS, 0x00);
-	if (priv->rtl_chip == RTL8188F) {
+	if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B) {
 		rtl8xxxu_write8(priv, REG_FWHW_TXQ_CTRL, FWHW_TXQ_CTRL_AMPDU_RETRY);
 		rtl8xxxu_write32(priv, REG_FAST_EDCA_CTRL, 0x03086666);
 	}
+	/*
+	 * The RTL8710BU vendor driver uses 0x50 here and it works fine,
+	 * but in rtl8xxxu 0x50 causes slow upload and random packet loss. Why?
+	 */
 	if (priv->rtl_chip == RTL8723B)
 		val8 = 0x50;
-	else if (priv->rtl_chip == RTL8188F)
+	else if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B)
 		val8 = 0x28; /* 0x50 would make the upload slow */
 	rtl8xxxu_write8(priv, REG_USTIME_TSF_8723B, val8);
 	rtl8xxxu_write8(priv, REG_USTIME_EDCA, val8);
@@ -3923,7 +3968,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	/* RFSW Control - clear bit 14 ?? */
 	if (priv->rtl_chip != RTL8723B && priv->rtl_chip != RTL8192E &&
-	    priv->rtl_chip != RTL8188E)
+	    priv->rtl_chip != RTL8188E && priv->rtl_chip != RTL8710B)
 		rtl8xxxu_write32(priv, REG_FPGA0_TX_INFO, 0x00000003);
 
 	val32 = FPGA0_RF_TRSW | FPGA0_RF_TRSWB | FPGA0_RF_ANTSW |
@@ -3936,7 +3981,8 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	rtl8xxxu_write32(priv, REG_FPGA0_XAB_RF_SW_CTRL, val32);
 
 	/* 0x860[6:5]= 00 - why? - this sets antenna B */
-	if (priv->rtl_chip != RTL8192E && priv->rtl_chip != RTL8188E)
+	if (priv->rtl_chip != RTL8192E && priv->rtl_chip != RTL8188E &&
+	    priv->rtl_chip != RTL8710B)
 		rtl8xxxu_write32(priv, REG_FPGA0_XA_RF_INT_OE, 0x66f60210);
 
 	if (!macpower) {
@@ -4009,10 +4055,14 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 			val8 &= 0xf8;
 			rtl8xxxu_write8(priv, 0xa3, val8);
 		}
+
+		if (priv->rtl_chip == RTL8710B)
+			rtl8xxxu_write8(priv, REG_EARLY_MODE_CONTROL_8710B, 0);
 	}
 
 	/*
-	 * Unit in 8 bytes, not obvious what it is used for
+	 * Unit in 8 bytes.
+	 * Get Rx PHY status in order to report RSSI and others.
 	 */
 	rtl8xxxu_write8(priv, REG_RX_DRVINFO_SZ, 4);
 
@@ -4031,6 +4081,8 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		val8 = rtl8xxxu_read8(priv, REG_USB_SPECIAL_OPTION);
 		val8 |= USB_SPEC_INT_BULK_SELECT;
 		rtl8xxxu_write8(priv, REG_USB_SPECIAL_OPTION, val8);
+	} else if (priv->rtl_chip == RTL8710B) {
+		rtl8xxxu_write32(priv, REG_HIMR0_8710B, 0);
 	} else {
 		/*
 		 * Enable all interrupts - not obvious USB needs to do this
@@ -4050,7 +4102,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		RCR_APPEND_PHYSTAT | RCR_APPEND_ICV | RCR_APPEND_MIC;
 	rtl8xxxu_write32(priv, REG_RCR, val32);
 
-	if (priv->rtl_chip == RTL8188F) {
+	if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B) {
 		/* Accept all data frames */
 		rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0xffff);
 
@@ -4119,7 +4171,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	val16 = BEACON_DISABLE_TSF_UPDATE | (BEACON_DISABLE_TSF_UPDATE << 8);
 	rtl8xxxu_write16(priv, REG_BEACON_CTRL, val16);
 	rtl8xxxu_write16(priv, REG_TBTT_PROHIBIT, 0x6404);
-	if (priv->rtl_chip != RTL8188F)
+	if (priv->rtl_chip != RTL8188F && priv->rtl_chip != RTL8710B)
 		/* Firmware will control REG_DRVERLYINT when power saving is enable, */
 		/* so don't set this register on STA mode. */
 		rtl8xxxu_write8(priv, REG_DRIVER_EARLY_INT, DRIVER_EARLY_INT_TIME);
@@ -4129,14 +4181,14 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/*
 	 * Initialize burst parameters
 	 */
-
 	if (priv->fops->init_burst)
 		priv->fops->init_burst(priv);
 
 	if (fops->init_aggregation)
 		fops->init_aggregation(priv);
 
-	if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8188E) {
+	if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8188E ||
+	    priv->rtl_chip == RTL8710B) {
 		rtl8xxxu_write16(priv, REG_PKT_VO_VI_LIFE_TIME, 0x0400); /* unit: 256us. 256ms */
 		rtl8xxxu_write16(priv, REG_PKT_BE_BK_LIFE_TIME, 0x0400); /* unit: 256us. 256ms */
 	}
@@ -4159,7 +4211,8 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	fops->set_tx_power(priv, 1, false);
 
 	/* Let the 8051 take control of antenna setting */
-	if (priv->rtl_chip != RTL8192E && priv->rtl_chip != RTL8188F) {
+	if (priv->rtl_chip != RTL8192E && priv->rtl_chip != RTL8188F &&
+	    priv->rtl_chip != RTL8710B) {
 		val8 = rtl8xxxu_read8(priv, REG_LEDCFG2);
 		val8 |= LEDCFG2_DPDT_SELECT;
 		rtl8xxxu_write8(priv, REG_LEDCFG2, val8);
@@ -4170,7 +4223,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/* Disable BAR - not sure if this has any effect on USB */
 	rtl8xxxu_write32(priv, REG_BAR_MODE_CTRL, 0x0201ffff);
 
-	if (priv->rtl_chip != RTL8188F && priv->rtl_chip != RTL8188E)
+	if (priv->rtl_chip != RTL8188F && priv->rtl_chip != RTL8188E && priv->rtl_chip != RTL8710B)
 		rtl8xxxu_write16(priv, REG_FAST_EDCA_CTRL, 0);
 
 	if (fops->init_statistics)
@@ -4209,7 +4262,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	 * This should enable thermal meter
 	 */
 	if (fops->gen2_thermal_meter) {
-		if (priv->rtl_chip == RTL8188F) {
+		if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B) {
 			val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_T_METER_8723B);
 			val32 |= 0x30000;
 			rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_T_METER_8723B, val32);
@@ -4281,6 +4334,24 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write32(priv, REG_AGC_RPT, val32);
 	}
 
+	if (priv->rtl_chip == RTL8710B) {
+		/*
+		 * 0x76D[5:4] is Port0,Port1 Enable Bit.
+		 * This is only for 8710B, 2b'00 for MP and 2b'11 for Normal Driver
+		 */
+		val8 = rtl8xxxu_read8(priv, REG_PORT_CONTROL_8710B);
+		val8 |= BIT(5) | BIT(4);
+		rtl8xxxu_write8(priv, REG_PORT_CONTROL_8710B, val8);
+
+		/* Set 0x5c[8] and [2:0] = 1, LDO mode */
+		val32 = rtl8xxxu_read32(priv, REG_WL_RF_PSS_8710B);
+		val32 |= 0x107;
+		rtl8xxxu_write32(priv, REG_WL_RF_PSS_8710B, val32);
+	}
+
+	val32 = rtl8xxxu_read32(priv, 0xa9c);
+	priv->cck_new_agc = u32_get_bits(val32, BIT(17));
+
 	/* Initialise the center frequency offset tracking */
 	if (priv->fops->set_crystal_cap) {
 		val32 = rtl8xxxu_read32(priv, REG_OFDM1_CFO_TRACKING);
@@ -5370,6 +5441,10 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 
 	rtl8xxxu_calc_tx_desc_csum(tx_desc);
 
+	/* avoid zero checksum make tx hang */
+	if (priv->rtl_chip == RTL8710B)
+		tx_desc->csum = ~tx_desc->csum;
+
 	usb_fill_bulk_urb(&tx_urb->urb, priv->udev, priv->pipe_out[queue],
 			  skb->data, skb->len, rtl8xxxu_tx_complete, skb);
 
@@ -5385,11 +5460,11 @@ error:
 	dev_kfree_skb(skb);
 }
 
-static void rtl8xxxu_rx_parse_phystats(struct rtl8xxxu_priv *priv,
-				       struct ieee80211_rx_status *rx_status,
-				       struct rtl8723au_phy_stats *phy_stats,
-				       u32 rxmcs, struct ieee80211_hdr *hdr,
-				       bool crc_icv_err)
+void rtl8723au_rx_parse_phystats(struct rtl8xxxu_priv *priv,
+				 struct ieee80211_rx_status *rx_status,
+				 struct rtl8723au_phy_stats *phy_stats,
+				 u32 rxmcs, struct ieee80211_hdr *hdr,
+				 bool crc_icv_err)
 {
 	if (phy_stats->sgi_en)
 		rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
@@ -5398,9 +5473,7 @@ static void rtl8xxxu_rx_parse_phystats(struct rtl8xxxu_priv *priv,
 		/*
 		 * Handle PHY stats for CCK rates
 		 */
-		u8 cck_agc_rpt = phy_stats->cck_agc_rpt_ofdm_cfosho_a;
-
-		rx_status->signal = priv->fops->cck_rssi(priv, cck_agc_rpt);
+		rx_status->signal = priv->fops->cck_rssi(priv, phy_stats);
 	} else {
 		bool parse_cfo = priv->fops->set_crystal_cap &&
 				 priv->vif &&
@@ -5422,6 +5495,96 @@ static void rtl8xxxu_rx_parse_phystats(struct rtl8xxxu_priv *priv,
 	}
 }
 
+static void jaguar2_rx_parse_phystats_type0(struct rtl8xxxu_priv *priv,
+					    struct ieee80211_rx_status *rx_status,
+					    struct jaguar2_phy_stats_type0 *phy_stats0,
+					    u32 rxmcs, struct ieee80211_hdr *hdr,
+					    bool crc_icv_err)
+{
+	s8 rx_power = phy_stats0->pwdb - 110;
+
+	if (!priv->cck_new_agc)
+		rx_power = priv->fops->cck_rssi(priv, (struct rtl8723au_phy_stats *)phy_stats0);
+
+	rx_status->signal = rx_power;
+}
+
+static void jaguar2_rx_parse_phystats_type1(struct rtl8xxxu_priv *priv,
+					    struct ieee80211_rx_status *rx_status,
+					    struct jaguar2_phy_stats_type1 *phy_stats1,
+					    u32 rxmcs, struct ieee80211_hdr *hdr,
+					    bool crc_icv_err)
+{
+	bool parse_cfo = priv->fops->set_crystal_cap &&
+			 priv->vif &&
+			 priv->vif->type == NL80211_IFTYPE_STATION &&
+			 priv->vif->cfg.assoc &&
+			 !crc_icv_err &&
+			 !ieee80211_is_ctl(hdr->frame_control) &&
+			 ether_addr_equal(priv->vif->bss_conf.bssid, hdr->addr2);
+	u8 pwdb_max = 0;
+	int rx_path;
+
+	if (parse_cfo) {
+		/* Only path-A and path-B have CFO tail and short CFO */
+		priv->cfo_tracking.cfo_tail[RF_A] = phy_stats1->cfo_tail[RF_A];
+		priv->cfo_tracking.cfo_tail[RF_B] = phy_stats1->cfo_tail[RF_B];
+
+		priv->cfo_tracking.packet_count++;
+	}
+
+	for (rx_path = 0; rx_path < priv->rx_paths; rx_path++)
+		pwdb_max = max(pwdb_max, phy_stats1->pwdb[rx_path]);
+
+	rx_status->signal = pwdb_max - 110;
+}
+
+static void jaguar2_rx_parse_phystats_type2(struct rtl8xxxu_priv *priv,
+					    struct ieee80211_rx_status *rx_status,
+					    struct jaguar2_phy_stats_type2 *phy_stats2,
+					    u32 rxmcs, struct ieee80211_hdr *hdr,
+					    bool crc_icv_err)
+{
+	u8 pwdb_max = 0;
+	int rx_path;
+
+	for (rx_path = 0; rx_path < priv->rx_paths; rx_path++)
+		pwdb_max = max(pwdb_max, phy_stats2->pwdb[rx_path]);
+
+	rx_status->signal = pwdb_max - 110;
+}
+
+void jaguar2_rx_parse_phystats(struct rtl8xxxu_priv *priv,
+			       struct ieee80211_rx_status *rx_status,
+			       struct rtl8723au_phy_stats *phy_stats,
+			       u32 rxmcs, struct ieee80211_hdr *hdr,
+			       bool crc_icv_err)
+{
+	struct jaguar2_phy_stats_type0 *phy_stats0 = (struct jaguar2_phy_stats_type0 *)phy_stats;
+	struct jaguar2_phy_stats_type1 *phy_stats1 = (struct jaguar2_phy_stats_type1 *)phy_stats;
+	struct jaguar2_phy_stats_type2 *phy_stats2 = (struct jaguar2_phy_stats_type2 *)phy_stats;
+
+	switch (phy_stats0->page_num) {
+	case 0:
+		/* CCK */
+		jaguar2_rx_parse_phystats_type0(priv, rx_status, phy_stats0,
+						rxmcs, hdr, crc_icv_err);
+		break;
+	case 1:
+		/* OFDM */
+		jaguar2_rx_parse_phystats_type1(priv, rx_status, phy_stats1,
+						rxmcs, hdr, crc_icv_err);
+		break;
+	case 2:
+		/* Also OFDM but different (how?) */
+		jaguar2_rx_parse_phystats_type2(priv, rx_status, phy_stats2,
+						rxmcs, hdr, crc_icv_err);
+		break;
+	default:
+		return;
+	}
+}
+
 static void rtl8xxxu_free_rx_resources(struct rtl8xxxu_priv *priv)
 {
 	struct rtl8xxxu_rx_urb *rx_urb, *tmp;
@@ -5920,7 +6083,7 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb)
 			skb_trim(skb, pkt_len);
 
 			if (rx_desc->phy_stats)
-				rtl8xxxu_rx_parse_phystats(
+				priv->fops->parse_phystats(
 					priv, rx_status, phy_stats,
 					rx_desc->rxmcs,
 					(struct ieee80211_hdr *)skb->data,
@@ -5995,7 +6158,7 @@ int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb)
 	}
 
 	if (rx_desc->phy_stats)
-		rtl8xxxu_rx_parse_phystats(priv, rx_status, phy_stats,
+		priv->fops->parse_phystats(priv, rx_status, phy_stats,
 					   rx_desc->rxmcs, (struct ieee80211_hdr *)skb->data,
 					   rx_desc->crc32 || rx_desc->icverr);
 
@@ -7007,6 +7170,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
 		case 0x818b:
 		case 0xf179:
 		case 0x8179:
+		case 0xb711:
 			untested = 0;
 			break;
 		}
@@ -7055,6 +7219,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
 	priv->udev = udev;
 	priv->fops = (struct rtl8xxxu_fileops *)id->driver_info;
 	mutex_init(&priv->usb_buf_mutex);
+	mutex_init(&priv->syson_indirect_access_mutex);
 	mutex_init(&priv->h2c_mutex);
 	INIT_LIST_HEAD(&priv->tx_urb_free_list);
 	spin_lock_init(&priv->tx_urb_lock);
@@ -7084,7 +7249,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
 	else
 		INIT_WORK(&priv->c2hcmd_work, rtl8xxxu_c2hcmd_callback);
 
-	ret = rtl8xxxu_read_efuse(priv);
+	ret = priv->fops->read_efuse(priv);
 	if (ret) {
 		dev_err(&udev->dev, "Fatal - failed to read EFuse\n");
 		goto err_set_intfdata;
@@ -7174,6 +7339,7 @@ err_set_intfdata:
 
 	kfree(priv->fw_data);
 	mutex_destroy(&priv->usb_buf_mutex);
+	mutex_destroy(&priv->syson_indirect_access_mutex);
 	mutex_destroy(&priv->h2c_mutex);
 
 	ieee80211_free_hw(hw);
@@ -7203,6 +7369,7 @@ static void rtl8xxxu_disconnect(struct usb_interface *interface)
 
 	kfree(priv->fw_data);
 	mutex_destroy(&priv->usb_buf_mutex);
+	mutex_destroy(&priv->syson_indirect_access_mutex);
 	mutex_destroy(&priv->h2c_mutex);
 
 	if (priv->udev->state != USB_STATE_NOTATTACHED) {
@@ -7283,6 +7450,12 @@ static const struct usb_device_id dev_table[] = {
 /* Rosewill USB-N150 Nano */
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0xffef, 0xff, 0xff, 0xff),
 	.driver_info = (unsigned long)&rtl8188eu_fops},
+/* RTL8710BU aka RTL8188GU (not to be confused with RTL8188GTVU) */
+{USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0xb711, 0xff, 0xff, 0xff),
+	.driver_info = (unsigned long)&rtl8710bu_fops},
+/* TOTOLINK N150UA V5 / N150UA-B */
+{USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x2005, 0xff, 0xff, 0xff),
+	.driver_info = (unsigned long)&rtl8710bu_fops},
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 /* Still supported by rtlwifi */
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x8176, 0xff, 0xff, 0xff),
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 5849fa4e1566..4dffbab494c3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -490,6 +490,8 @@
 #define  RXDMA_PRO_DMA_BURST_CNT	GENMASK(3, 2)	/* Set to 0x3. */
 #define  RXDMA_PRO_DMA_BURST_SIZE	GENMASK(5, 4)	/* Set to 0x1. */
 
+#define REG_EARLY_MODE_CONTROL_8710B	0x02bc
+
 #define REG_RF_BB_CMD_ADDR		0x02c0
 #define REG_RF_BB_CMD_DATA		0x02c4
 
@@ -845,6 +847,7 @@
 #define REG_BT_CONTROL_8723BU		0x0764
 #define  BT_CONTROL_BT_GRANT		BIT(12)
 
+#define REG_PORT_CONTROL_8710B		0x076d
 #define REG_WLAN_ACT_CONTROL_8723B	0x076e
 
 #define REG_FPGA0_RF_MODE		0x0800
@@ -1004,8 +1007,12 @@
 #define  CCK_PD_TYPE1_LV3_TH		0xdd
 #define  CCK_PD_TYPE1_LV4_TH		0xed
 
+#define REG_CCK0_TX_FILTER1		0x0a20
+#define REG_CCK0_TX_FILTER2		0x0a24
+#define REG_CCK0_DEBUG_PORT		0x0a28	/* debug port and Tx filter3 */
 #define REG_AGC_RPT			0xa80
 #define  AGC_RPT_CCK			BIT(7)
+#define REG_CCK0_TX_FILTER3		0x0aac
 
 #define REG_CONFIG_ANT_A		0x0b68
 #define REG_CONFIG_ANT_B		0x0b6c
@@ -1179,6 +1186,8 @@
 						    Unavailable */
 #define  USB_HIMR_ROK			BIT(0)	/*  Receive DMA OK Interrupt */
 
+#define REG_USB_ACCESS_TIMEOUT		0xfe4c
+
 #define REG_USB_SPECIAL_OPTION		0xfe55
 #define  USB_SPEC_USB_AGG_ENABLE	BIT(3)	/* Enable USB aggregation */
 #define  USB_SPEC_INT_BULK_SELECT	BIT(4)	/* Use interrupt endpoint to
@@ -1204,6 +1213,41 @@
 #define REG_NORMAL_SIE_MAC_ADDR		0xfe70	/* 0xfe70 - 0xfe75 */
 #define REG_NORMAL_SIE_STRING		0xfe80	/* 0xfe80 - 0xfedf */
 
+/*
+ * 8710B register addresses between 0x00 and 0xff must have 0x8000
+ * added to them. We take care of that in the rtl8xxxu_read{8,16,32}
+ * and rtl8xxxu_write{8,16,32} functions.
+ */
+#define REG_SYS_FUNC_8710B		0x0004
+#define REG_AFE_CTRL_8710B		0x0050
+#define REG_WL_RF_PSS_8710B		0x005c
+#define REG_EFUSE_INDIRECT_CTRL_8710B	0x006c
+#define  NORMAL_REG_READ_OFFSET		0x83000000
+#define  NORMAL_REG_WRITE_OFFSET	0x84000000
+#define  EFUSE_READ_OFFSET		0x85000000
+#define  EFUSE_WRITE_OFFSET		0x86000000
+#define REG_HIMR0_8710B			0x0080
+#define REG_HISR0_8710B			0x0084
+/*
+ * 8710B uses this instead of REG_MCU_FW_DL, but at least bits
+ * 0-7 have the same meaning.
+ */
+#define REG_8051FW_CTRL_V1_8710B	0x0090
+#define REG_USB_HOST_INDIRECT_DATA_8710B	0x009c
+#define REG_WL_STATUS_8710B		0x00f0
+#define REG_USB_HOST_INDIRECT_ADDR_8710B	0x00f8
+
+/*
+ * 8710B registers which must be accessed through rtl8710b_read_syson_reg
+ * and rtl8710b_write_syson_reg.
+ */
+#define SYSON_REG_BASE_ADDR_8710B	0x40000000
+#define REG_SYS_XTAL_CTRL0_8710B	0x060
+#define REG_SYS_EEPROM_CTRL0_8710B	0x0e0
+#define REG_SYS_SYSTEM_CFG0_8710B	0x1f0
+#define REG_SYS_SYSTEM_CFG1_8710B	0x1f4
+#define REG_SYS_SYSTEM_CFG2_8710B	0x1f8
+
 /* RF6052 registers */
 #define RF6052_REG_AC			0x00
 #define RF6052_REG_IQADJ_G1		0x01
-- 
cgit v1.2.3-70-g09d2


From 672faa7bbf6014a301b6edc477918af445c92ce2 Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:07 -0700
Subject: phy: phy-ocelot-serdes: add ability to be used in a non-syscon
 configuration

The phy-ocelot-serdes module has exclusively been used in a syscon setup,
from an internal CPU. The addition of external control of ocelot switches
via an existing MFD implementation means that syscon is no longer the only
interface that phy-ocelot-serdes will see.

In the MFD configuration, an IORESOURCE_REG resource will exist for the
device. Utilize this resource to be able to function in both syscon and
non-syscon configurations.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/phy/mscc/phy-ocelot-serdes.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/phy/mscc/phy-ocelot-serdes.c b/drivers/phy/mscc/phy-ocelot-serdes.c
index 76f596365176..d9443e865a78 100644
--- a/drivers/phy/mscc/phy-ocelot-serdes.c
+++ b/drivers/phy/mscc/phy-ocelot-serdes.c
@@ -494,6 +494,7 @@ static int serdes_probe(struct platform_device *pdev)
 {
 	struct phy_provider *provider;
 	struct serdes_ctrl *ctrl;
+	struct resource *res;
 	unsigned int i;
 	int ret;
 
@@ -503,6 +504,14 @@ static int serdes_probe(struct platform_device *pdev)
 
 	ctrl->dev = &pdev->dev;
 	ctrl->regs = syscon_node_to_regmap(pdev->dev.parent->of_node);
+	if (IS_ERR(ctrl->regs)) {
+		/* Fall back to using IORESOURCE_REG, if possible */
+		res = platform_get_resource(pdev, IORESOURCE_REG, 0);
+		if (res)
+			ctrl->regs = dev_get_regmap(ctrl->dev->parent,
+						    res->name);
+	}
+
 	if (IS_ERR(ctrl->regs))
 		return PTR_ERR(ctrl->regs);
 
-- 
cgit v1.2.3-70-g09d2


From c21ff0939d1d9af1acdf0bf420f206ad3bad2c92 Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:08 -0700
Subject: mfd: ocelot: add ocelot-serdes capability

Add support for the Ocelot SERDES module to support functionality of all
non-internal phy ports.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/mfd/ocelot-core.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/mfd/ocelot-core.c b/drivers/mfd/ocelot-core.c
index e1772ff00cad..9cccf54fc9c8 100644
--- a/drivers/mfd/ocelot-core.c
+++ b/drivers/mfd/ocelot-core.c
@@ -45,6 +45,9 @@
 #define VSC7512_SIO_CTRL_RES_START	0x710700f8
 #define VSC7512_SIO_CTRL_RES_SIZE	0x00000100
 
+#define VSC7512_HSIO_RES_START		0x710d0000
+#define VSC7512_HSIO_RES_SIZE		0x00000128
+
 #define VSC7512_ANA_RES_START		0x71880000
 #define VSC7512_ANA_RES_SIZE		0x00010000
 
@@ -129,8 +132,13 @@ static const struct resource vsc7512_sgpio_resources[] = {
 	DEFINE_RES_REG_NAMED(VSC7512_SIO_CTRL_RES_START, VSC7512_SIO_CTRL_RES_SIZE, "gcb_sio"),
 };
 
+static const struct resource vsc7512_serdes_resources[] = {
+	DEFINE_RES_REG_NAMED(VSC7512_HSIO_RES_START, VSC7512_HSIO_RES_SIZE, "hsio"),
+};
+
 static const struct resource vsc7512_switch_resources[] = {
 	DEFINE_RES_REG_NAMED(VSC7512_ANA_RES_START, VSC7512_ANA_RES_SIZE, "ana"),
+	DEFINE_RES_REG_NAMED(VSC7512_HSIO_RES_START, VSC7512_HSIO_RES_SIZE, "hsio"),
 	DEFINE_RES_REG_NAMED(VSC7512_QS_RES_START, VSC7512_QS_RES_SIZE, "qs"),
 	DEFINE_RES_REG_NAMED(VSC7512_QSYS_RES_START, VSC7512_QSYS_RES_SIZE, "qsys"),
 	DEFINE_RES_REG_NAMED(VSC7512_REW_RES_START, VSC7512_REW_RES_SIZE, "rew"),
@@ -176,6 +184,11 @@ static const struct mfd_cell vsc7512_devs[] = {
 		.use_of_reg = true,
 		.num_resources = ARRAY_SIZE(vsc7512_miim1_resources),
 		.resources = vsc7512_miim1_resources,
+	}, {
+		.name = "ocelot-serdes",
+		.of_compatible = "mscc,vsc7514-serdes",
+		.num_resources = ARRAY_SIZE(vsc7512_serdes_resources),
+		.resources = vsc7512_serdes_resources,
 	}, {
 		.name = "ocelot-ext-switch",
 		.of_compatible = "mscc,vsc7512-switch",
-- 
cgit v1.2.3-70-g09d2


From fec53f44945877c8627da4d3ad70e3ac7e204f38 Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:09 -0700
Subject: net: mscc: ocelot: expose ocelot_pll5_init routine

Ocelot chips have an internal PLL that must be used when communicating
through external phys. Expose the init routine, so it can be used by other
drivers.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mscc/ocelot.c         | 31 ++++++++++++++++++++++++++++++
 drivers/net/ethernet/mscc/ocelot_vsc7514.c | 30 -----------------------------
 include/soc/mscc/ocelot.h                  |  2 ++
 3 files changed, 33 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 08acb7b89086..9b8403e29445 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -7,6 +7,7 @@
 #include <linux/dsa/ocelot.h>
 #include <linux/if_bridge.h>
 #include <linux/iopoll.h>
+#include <soc/mscc/ocelot_hsio.h>
 #include <soc/mscc/ocelot_vcap.h>
 #include "ocelot.h"
 #include "ocelot_vcap.h"
@@ -211,6 +212,36 @@ static void ocelot_mact_init(struct ocelot *ocelot)
 	ocelot_write(ocelot, MACACCESS_CMD_INIT, ANA_TABLES_MACACCESS);
 }
 
+void ocelot_pll5_init(struct ocelot *ocelot)
+{
+	/* Configure PLL5. This will need a proper CCF driver
+	 * The values are coming from the VTSS API for Ocelot
+	 */
+	regmap_write(ocelot->targets[HSIO], HSIO_PLL5G_CFG4,
+		     HSIO_PLL5G_CFG4_IB_CTRL(0x7600) |
+		     HSIO_PLL5G_CFG4_IB_BIAS_CTRL(0x8));
+	regmap_write(ocelot->targets[HSIO], HSIO_PLL5G_CFG0,
+		     HSIO_PLL5G_CFG0_CORE_CLK_DIV(0x11) |
+		     HSIO_PLL5G_CFG0_CPU_CLK_DIV(2) |
+		     HSIO_PLL5G_CFG0_ENA_BIAS |
+		     HSIO_PLL5G_CFG0_ENA_VCO_BUF |
+		     HSIO_PLL5G_CFG0_ENA_CP1 |
+		     HSIO_PLL5G_CFG0_SELCPI(2) |
+		     HSIO_PLL5G_CFG0_LOOP_BW_RES(0xe) |
+		     HSIO_PLL5G_CFG0_SELBGV820(4) |
+		     HSIO_PLL5G_CFG0_DIV4 |
+		     HSIO_PLL5G_CFG0_ENA_CLKTREE |
+		     HSIO_PLL5G_CFG0_ENA_LANE);
+	regmap_write(ocelot->targets[HSIO], HSIO_PLL5G_CFG2,
+		     HSIO_PLL5G_CFG2_EN_RESET_FRQ_DET |
+		     HSIO_PLL5G_CFG2_EN_RESET_OVERRUN |
+		     HSIO_PLL5G_CFG2_GAIN_TEST(0x8) |
+		     HSIO_PLL5G_CFG2_ENA_AMPCTRL |
+		     HSIO_PLL5G_CFG2_PWD_AMPCTRL_N |
+		     HSIO_PLL5G_CFG2_AMPC_SEL(0x10));
+}
+EXPORT_SYMBOL(ocelot_pll5_init);
+
 static void ocelot_vcap_enable(struct ocelot *ocelot, int port)
 {
 	ocelot_write_gix(ocelot, ANA_PORT_VCAP_S2_CFG_S2_ENA |
diff --git a/drivers/net/ethernet/mscc/ocelot_vsc7514.c b/drivers/net/ethernet/mscc/ocelot_vsc7514.c
index 7388c3b0535c..97e90e2869d4 100644
--- a/drivers/net/ethernet/mscc/ocelot_vsc7514.c
+++ b/drivers/net/ethernet/mscc/ocelot_vsc7514.c
@@ -18,7 +18,6 @@
 
 #include <soc/mscc/ocelot.h>
 #include <soc/mscc/ocelot_vcap.h>
-#include <soc/mscc/ocelot_hsio.h>
 #include <soc/mscc/vsc7514_regs.h>
 #include "ocelot_fdma.h"
 #include "ocelot.h"
@@ -26,35 +25,6 @@
 #define VSC7514_VCAP_POLICER_BASE			128
 #define VSC7514_VCAP_POLICER_MAX			191
 
-static void ocelot_pll5_init(struct ocelot *ocelot)
-{
-	/* Configure PLL5. This will need a proper CCF driver
-	 * The values are coming from the VTSS API for Ocelot
-	 */
-	regmap_write(ocelot->targets[HSIO], HSIO_PLL5G_CFG4,
-		     HSIO_PLL5G_CFG4_IB_CTRL(0x7600) |
-		     HSIO_PLL5G_CFG4_IB_BIAS_CTRL(0x8));
-	regmap_write(ocelot->targets[HSIO], HSIO_PLL5G_CFG0,
-		     HSIO_PLL5G_CFG0_CORE_CLK_DIV(0x11) |
-		     HSIO_PLL5G_CFG0_CPU_CLK_DIV(2) |
-		     HSIO_PLL5G_CFG0_ENA_BIAS |
-		     HSIO_PLL5G_CFG0_ENA_VCO_BUF |
-		     HSIO_PLL5G_CFG0_ENA_CP1 |
-		     HSIO_PLL5G_CFG0_SELCPI(2) |
-		     HSIO_PLL5G_CFG0_LOOP_BW_RES(0xe) |
-		     HSIO_PLL5G_CFG0_SELBGV820(4) |
-		     HSIO_PLL5G_CFG0_DIV4 |
-		     HSIO_PLL5G_CFG0_ENA_CLKTREE |
-		     HSIO_PLL5G_CFG0_ENA_LANE);
-	regmap_write(ocelot->targets[HSIO], HSIO_PLL5G_CFG2,
-		     HSIO_PLL5G_CFG2_EN_RESET_FRQ_DET |
-		     HSIO_PLL5G_CFG2_EN_RESET_OVERRUN |
-		     HSIO_PLL5G_CFG2_GAIN_TEST(0x8) |
-		     HSIO_PLL5G_CFG2_ENA_AMPCTRL |
-		     HSIO_PLL5G_CFG2_PWD_AMPCTRL_N |
-		     HSIO_PLL5G_CFG2_AMPC_SEL(0x10));
-}
-
 static int ocelot_chip_init(struct ocelot *ocelot, const struct ocelot_ops *ops)
 {
 	int ret;
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 2080879e4134..751d9b250615 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -1183,4 +1183,6 @@ ocelot_mrp_del_ring_role(struct ocelot *ocelot, int port,
 }
 #endif
 
+void ocelot_pll5_init(struct ocelot *ocelot);
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 69f7f89c0db52c5a3fe1bc9ba69d8248b5ee0bca Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:10 -0700
Subject: net: mscc: ocelot: expose generic phylink_mac_config routine

The ocelot-switch driver can utilize the phylink_mac_config routine. Move
this to the ocelot library location and export the symbol to make this
possible.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mscc/ocelot.c     | 26 ++++++++++++++++++++++++++
 drivers/net/ethernet/mscc/ocelot_net.c | 21 +++------------------
 include/soc/mscc/ocelot.h              |  3 +++
 3 files changed, 32 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 9b8403e29445..8292e93a3782 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -809,6 +809,32 @@ static int ocelot_port_flush(struct ocelot *ocelot, int port)
 	return err;
 }
 
+void ocelot_phylink_mac_config(struct ocelot *ocelot, int port,
+			       unsigned int link_an_mode,
+			       const struct phylink_link_state *state)
+{
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+
+	/* Disable HDX fast control */
+	ocelot_port_writel(ocelot_port, DEV_PORT_MISC_HDX_FAST_DIS,
+			   DEV_PORT_MISC);
+
+	/* SGMII only for now */
+	ocelot_port_writel(ocelot_port, PCS1G_MODE_CFG_SGMII_MODE_ENA,
+			   PCS1G_MODE_CFG);
+	ocelot_port_writel(ocelot_port, PCS1G_SD_CFG_SD_SEL, PCS1G_SD_CFG);
+
+	/* Enable PCS */
+	ocelot_port_writel(ocelot_port, PCS1G_CFG_PCS_ENA, PCS1G_CFG);
+
+	/* No aneg on SGMII */
+	ocelot_port_writel(ocelot_port, 0, PCS1G_ANEG_CFG);
+
+	/* No loopback */
+	ocelot_port_writel(ocelot_port, 0, PCS1G_LB_CFG);
+}
+EXPORT_SYMBOL_GPL(ocelot_phylink_mac_config);
+
 void ocelot_phylink_mac_link_down(struct ocelot *ocelot, int port,
 				  unsigned int link_an_mode,
 				  phy_interface_t interface,
diff --git a/drivers/net/ethernet/mscc/ocelot_net.c b/drivers/net/ethernet/mscc/ocelot_net.c
index ca4bde861397..590a2b2816ad 100644
--- a/drivers/net/ethernet/mscc/ocelot_net.c
+++ b/drivers/net/ethernet/mscc/ocelot_net.c
@@ -1675,25 +1675,10 @@ static void vsc7514_phylink_mac_config(struct phylink_config *config,
 {
 	struct net_device *ndev = to_net_dev(config->dev);
 	struct ocelot_port_private *priv = netdev_priv(ndev);
-	struct ocelot_port *ocelot_port = &priv->port;
-
-	/* Disable HDX fast control */
-	ocelot_port_writel(ocelot_port, DEV_PORT_MISC_HDX_FAST_DIS,
-			   DEV_PORT_MISC);
-
-	/* SGMII only for now */
-	ocelot_port_writel(ocelot_port, PCS1G_MODE_CFG_SGMII_MODE_ENA,
-			   PCS1G_MODE_CFG);
-	ocelot_port_writel(ocelot_port, PCS1G_SD_CFG_SD_SEL, PCS1G_SD_CFG);
-
-	/* Enable PCS */
-	ocelot_port_writel(ocelot_port, PCS1G_CFG_PCS_ENA, PCS1G_CFG);
-
-	/* No aneg on SGMII */
-	ocelot_port_writel(ocelot_port, 0, PCS1G_ANEG_CFG);
+	struct ocelot *ocelot = priv->port.ocelot;
+	int port = priv->port.index;
 
-	/* No loopback */
-	ocelot_port_writel(ocelot_port, 0, PCS1G_LB_CFG);
+	ocelot_phylink_mac_config(ocelot, port, link_an_mode, state);
 }
 
 static void vsc7514_phylink_mac_link_down(struct phylink_config *config,
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 751d9b250615..87ade87d3540 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -1111,6 +1111,9 @@ int ocelot_sb_occ_tc_port_bind_get(struct ocelot *ocelot, int port,
 				   enum devlink_sb_pool_type pool_type,
 				   u32 *p_cur, u32 *p_max);
 
+void ocelot_phylink_mac_config(struct ocelot *ocelot, int port,
+			       unsigned int link_an_mode,
+			       const struct phylink_link_state *state);
 void ocelot_phylink_mac_link_down(struct ocelot *ocelot, int port,
 				  unsigned int link_an_mode,
 				  phy_interface_t interface,
-- 
cgit v1.2.3-70-g09d2


From dfca93ed51a7cf8bfda876705816a5e55381ac4a Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:11 -0700
Subject: net: mscc: ocelot: expose serdes configuration function

During chip initialization, ports that use SGMII / QSGMII to interface to
external phys need to be configured on the VSC7513 and VSC7514. Expose this
configuration routine, so it can be used by DSA drivers.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mscc/ocelot.c     | 40 ++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/mscc/ocelot_net.c | 29 +++---------------------
 include/soc/mscc/ocelot.h              |  4 ++++
 3 files changed, 47 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 8292e93a3782..1502bb2c8ea7 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -7,6 +7,7 @@
 #include <linux/dsa/ocelot.h>
 #include <linux/if_bridge.h>
 #include <linux/iopoll.h>
+#include <linux/phy/phy.h>
 #include <soc/mscc/ocelot_hsio.h>
 #include <soc/mscc/ocelot_vcap.h>
 #include "ocelot.h"
@@ -809,6 +810,45 @@ static int ocelot_port_flush(struct ocelot *ocelot, int port)
 	return err;
 }
 
+int ocelot_port_configure_serdes(struct ocelot *ocelot, int port,
+				 struct device_node *portnp)
+{
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+	struct device *dev = ocelot->dev;
+	int err;
+
+	/* Ensure clock signals and speed are set on all QSGMII links */
+	if (ocelot_port->phy_mode == PHY_INTERFACE_MODE_QSGMII)
+		ocelot_port_rmwl(ocelot_port, 0,
+				 DEV_CLOCK_CFG_MAC_TX_RST |
+				 DEV_CLOCK_CFG_MAC_RX_RST,
+				 DEV_CLOCK_CFG);
+
+	if (ocelot_port->phy_mode != PHY_INTERFACE_MODE_INTERNAL) {
+		struct phy *serdes = of_phy_get(portnp, NULL);
+
+		if (IS_ERR(serdes)) {
+			err = PTR_ERR(serdes);
+			dev_err_probe(dev, err,
+				      "missing SerDes phys for port %d\n",
+				      port);
+			return err;
+		}
+
+		err = phy_set_mode_ext(serdes, PHY_MODE_ETHERNET,
+				       ocelot_port->phy_mode);
+		of_phy_put(serdes);
+		if (err) {
+			dev_err(dev, "Could not SerDes mode on port %d: %pe\n",
+				port, ERR_PTR(err));
+			return err;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ocelot_port_configure_serdes);
+
 void ocelot_phylink_mac_config(struct ocelot *ocelot, int port,
 			       unsigned int link_an_mode,
 			       const struct phylink_link_state *state)
diff --git a/drivers/net/ethernet/mscc/ocelot_net.c b/drivers/net/ethernet/mscc/ocelot_net.c
index 590a2b2816ad..21a87a3fc556 100644
--- a/drivers/net/ethernet/mscc/ocelot_net.c
+++ b/drivers/net/ethernet/mscc/ocelot_net.c
@@ -1742,34 +1742,11 @@ static int ocelot_port_phylink_create(struct ocelot *ocelot, int port,
 		return -EINVAL;
 	}
 
-	/* Ensure clock signals and speed are set on all QSGMII links */
-	if (phy_mode == PHY_INTERFACE_MODE_QSGMII)
-		ocelot_port_rmwl(ocelot_port, 0,
-				 DEV_CLOCK_CFG_MAC_TX_RST |
-				 DEV_CLOCK_CFG_MAC_RX_RST,
-				 DEV_CLOCK_CFG);
-
 	ocelot_port->phy_mode = phy_mode;
 
-	if (phy_mode != PHY_INTERFACE_MODE_INTERNAL) {
-		struct phy *serdes = of_phy_get(portnp, NULL);
-
-		if (IS_ERR(serdes)) {
-			err = PTR_ERR(serdes);
-			dev_err_probe(dev, err,
-				      "missing SerDes phys for port %d\n",
-				      port);
-			return err;
-		}
-
-		err = phy_set_mode_ext(serdes, PHY_MODE_ETHERNET, phy_mode);
-		of_phy_put(serdes);
-		if (err) {
-			dev_err(dev, "Could not SerDes mode on port %d: %pe\n",
-				port, ERR_PTR(err));
-			return err;
-		}
-	}
+	err = ocelot_port_configure_serdes(ocelot, port, portnp);
+	if (err)
+		return err;
 
 	priv = container_of(ocelot_port, struct ocelot_port_private, port);
 
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 87ade87d3540..d757b5e26d26 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -644,6 +644,7 @@ enum ocelot_tag_prefix {
 };
 
 struct ocelot;
+struct device_node;
 
 struct ocelot_ops {
 	struct net_device *(*port_to_netdev)(struct ocelot *ocelot, int port);
@@ -1111,6 +1112,9 @@ int ocelot_sb_occ_tc_port_bind_get(struct ocelot *ocelot, int port,
 				   enum devlink_sb_pool_type pool_type,
 				   u32 *p_cur, u32 *p_max);
 
+int ocelot_port_configure_serdes(struct ocelot *ocelot, int port,
+				 struct device_node *portnp);
+
 void ocelot_phylink_mac_config(struct ocelot *ocelot, int port,
 			       unsigned int link_an_mode,
 			       const struct phylink_link_state *state);
-- 
cgit v1.2.3-70-g09d2


From 3821fd0107b09966a0b8ef0b2de3e999ba534266 Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:12 -0700
Subject: net: dsa: felix: attempt to initialize internal hsio plls

The VSC7512 and VSC7514 have internal PLLs that can be used to control
different peripherals. Initialize these high speed I/O (HSIO) PLLs when
they exist, so that dependent peripherals like QSGMII can function.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index d4cc9e60f369..21dcb9cadc12 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1555,6 +1555,9 @@ static int felix_setup(struct dsa_switch *ds)
 	if (err)
 		return err;
 
+	if (ocelot->targets[HSIO])
+		ocelot_pll5_init(ocelot);
+
 	err = ocelot_init(ocelot);
 	if (err)
 		goto out_mdiobus_free;
-- 
cgit v1.2.3-70-g09d2


From 544435c9346a34f90a1ed9f19fc40126ba6d4aaa Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:13 -0700
Subject: net: dsa: felix: allow configurable phylink_mac_config

If a user of the Felix driver has a port running in SGMII / QSGMII mode, it
will need to utilize phylink_mac_config(). Add this configurability.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 12 ++++++++++++
 drivers/net/dsa/ocelot/felix.h |  3 +++
 2 files changed, 15 insertions(+)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 21dcb9cadc12..845068bcbeb4 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1056,6 +1056,17 @@ static void felix_phylink_get_caps(struct dsa_switch *ds, int port,
 		  config->supported_interfaces);
 }
 
+static void felix_phylink_mac_config(struct dsa_switch *ds, int port,
+				     unsigned int mode,
+				     const struct phylink_link_state *state)
+{
+	struct ocelot *ocelot = ds->priv;
+	struct felix *felix = ocelot_to_felix(ocelot);
+
+	if (felix->info->phylink_mac_config)
+		felix->info->phylink_mac_config(ocelot, port, mode, state);
+}
+
 static struct phylink_pcs *felix_phylink_mac_select_pcs(struct dsa_switch *ds,
 							int port,
 							phy_interface_t iface)
@@ -2088,6 +2099,7 @@ const struct dsa_switch_ops felix_switch_ops = {
 	.get_sset_count			= felix_get_sset_count,
 	.get_ts_info			= felix_get_ts_info,
 	.phylink_get_caps		= felix_phylink_get_caps,
+	.phylink_mac_config		= felix_phylink_mac_config,
 	.phylink_mac_select_pcs		= felix_phylink_mac_select_pcs,
 	.phylink_mac_link_down		= felix_phylink_mac_link_down,
 	.phylink_mac_link_up		= felix_phylink_mac_link_up,
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index d5d0b30c0b75..98771273512b 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -58,6 +58,9 @@ struct felix_info {
 	void	(*tas_guard_bands_update)(struct ocelot *ocelot, int port);
 	void	(*port_sched_speed_set)(struct ocelot *ocelot, int port,
 					u32 speed);
+	void	(*phylink_mac_config)(struct ocelot *ocelot, int port,
+				      unsigned int mode,
+				      const struct phylink_link_state *state);
 };
 
 /* Methods for initializing the hardware resources specific to a tagging
-- 
cgit v1.2.3-70-g09d2


From 6865ecee385bc91d47e6a1801fdf70732e33a6fd Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:14 -0700
Subject: net: dsa: felix: allow serdes configuration for dsa ports

Ports for Ocelot devices (VSC7511, VSC7512, VSC7513 and VSC7514) support
external phys. When external phys are used, additional configuration on
each port is required to enable QSGMII mode and set external phy modes.

Add a configurable hook into these routines, so the external ports can be
used.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 4 ++++
 drivers/net/dsa/ocelot/felix.h | 4 ++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 845068bcbeb4..6dcebcfd71e7 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1585,6 +1585,10 @@ static int felix_setup(struct dsa_switch *ds)
 	dsa_switch_for_each_available_port(dp, ds) {
 		ocelot_init_port(ocelot, dp->index);
 
+		if (felix->info->configure_serdes)
+			felix->info->configure_serdes(ocelot, dp->index,
+						      dp->dn);
+
 		/* Set the default QoS Classification based on PCP and DEI
 		 * bits of vlan tag.
 		 */
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index 98771273512b..96008c046da5 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -15,6 +15,8 @@
 #define OCELOT_PORT_MODE_USXGMII	BIT(4)
 #define OCELOT_PORT_MODE_1000BASEX	BIT(5)
 
+struct device_node;
+
 /* Platform-specific information */
 struct felix_info {
 	/* Hardcoded resources provided by the hardware instantiation. */
@@ -61,6 +63,8 @@ struct felix_info {
 	void	(*phylink_mac_config)(struct ocelot *ocelot, int port,
 				      unsigned int mode,
 				      const struct phylink_link_state *state);
+	int	(*configure_serdes)(struct ocelot *ocelot, int port,
+				    struct device_node *portnp);
 };
 
 /* Methods for initializing the hardware resources specific to a tagging
-- 
cgit v1.2.3-70-g09d2


From 4c05e5ceecbbcc0bd570eae218066e1b8bab6438 Mon Sep 17 00:00:00 2001
From: Colin Foster <colin.foster@in-advantage.com>
Date: Fri, 17 Mar 2023 11:54:15 -0700
Subject: net: dsa: ocelot: add support for external phys

The VSC7512 has four ports with internal phys that are already supported.
There are additional ports that can be configured to work with external
phys.

Add support for these additional ethernet ports.

Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/ocelot_ext.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/drivers/net/dsa/ocelot/ocelot_ext.c b/drivers/net/dsa/ocelot/ocelot_ext.c
index 228737a32080..c29bee5a5c48 100644
--- a/drivers/net/dsa/ocelot/ocelot_ext.c
+++ b/drivers/net/dsa/ocelot/ocelot_ext.c
@@ -20,13 +20,13 @@ static const u32 vsc7512_port_modes[VSC7514_NUM_PORTS] = {
 	OCELOT_PORT_MODE_INTERNAL,
 	OCELOT_PORT_MODE_INTERNAL,
 	OCELOT_PORT_MODE_INTERNAL,
-	OCELOT_PORT_MODE_NONE,
-	OCELOT_PORT_MODE_NONE,
-	OCELOT_PORT_MODE_NONE,
-	OCELOT_PORT_MODE_NONE,
-	OCELOT_PORT_MODE_NONE,
-	OCELOT_PORT_MODE_NONE,
-	OCELOT_PORT_MODE_NONE,
+	OCELOT_PORT_MODE_SERDES,
+	OCELOT_PORT_MODE_SERDES,
+	OCELOT_PORT_MODE_SERDES,
+	OCELOT_PORT_MODE_SERDES,
+	OCELOT_PORT_MODE_SERDES,
+	OCELOT_PORT_MODE_SGMII,
+	OCELOT_PORT_MODE_SERDES,
 };
 
 static const struct ocelot_ops ocelot_ext_ops = {
@@ -59,6 +59,8 @@ static const struct felix_info vsc7512_info = {
 	.num_ports			= VSC7514_NUM_PORTS,
 	.num_tx_queues			= OCELOT_NUM_TC,
 	.port_modes			= vsc7512_port_modes,
+	.phylink_mac_config		= ocelot_phylink_mac_config,
+	.configure_serdes		= ocelot_port_configure_serdes,
 };
 
 static int ocelot_ext_probe(struct platform_device *pdev)
-- 
cgit v1.2.3-70-g09d2


From 8e80e5648092f63b17c2665b2a36422a77143278 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:36 +0200
Subject: net/mlx5: fs_chains: Refactor to detach chains from tc usage

To support more generic chains that will be used on other
namespaces and without tc, refactor to remove the dependency
on tc terms.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/bb8570d532d569285b5bff981578507bd15350cb.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 20 +----
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |  6 +-
 .../ethernet/mellanox/mlx5/core/lib/fs_chains.c    | 89 +++++++++++-----------
 .../ethernet/mellanox/mlx5/core/lib/fs_chains.h    |  9 ++-
 4 files changed, 55 insertions(+), 69 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 70b8d2dfa751..98c9befb7803 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -5186,22 +5186,6 @@ static int mlx5e_tc_netdev_event(struct notifier_block *this,
 	return NOTIFY_DONE;
 }
 
-static int mlx5e_tc_nic_get_ft_size(struct mlx5_core_dev *dev)
-{
-	int tc_grp_size, tc_tbl_size;
-	u32 max_flow_counter;
-
-	max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) |
-			    MLX5_CAP_GEN(dev, max_flow_counter_15_0);
-
-	tc_grp_size = min_t(int, max_flow_counter, MLX5E_TC_TABLE_MAX_GROUP_SIZE);
-
-	tc_tbl_size = min_t(int, tc_grp_size * MLX5E_TC_TABLE_NUM_GROUPS,
-			    BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev, log_max_ft_size)));
-
-	return tc_tbl_size;
-}
-
 static int mlx5e_tc_nic_create_miss_table(struct mlx5e_priv *priv)
 {
 	struct mlx5e_tc_table *tc = mlx5e_fs_get_tc(priv->fs);
@@ -5274,10 +5258,10 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 		attr.flags = MLX5_CHAINS_AND_PRIOS_SUPPORTED |
 			MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED;
 	attr.ns = MLX5_FLOW_NAMESPACE_KERNEL;
-	attr.max_ft_sz = mlx5e_tc_nic_get_ft_size(dev);
 	attr.max_grp_num = MLX5E_TC_TABLE_NUM_GROUPS;
 	attr.default_ft = tc->miss_t;
 	attr.mapping = chains_mapping;
+	attr.fs_base_prio = MLX5E_TC_PRIO;
 
 	tc->chains = mlx5_chains_create(dev, &attr);
 	if (IS_ERR(tc->chains)) {
@@ -5285,6 +5269,8 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 		goto err_miss;
 	}
 
+	mlx5_chains_print_info(tc->chains);
+
 	tc->post_act = mlx5e_tc_post_act_init(priv, tc->chains, MLX5_FLOW_NAMESPACE_KERNEL);
 	tc->ct = mlx5_tc_ct_init(priv, tc->chains, &tc->mod_hdr,
 				 MLX5_FLOW_NAMESPACE_KERNEL, tc->post_act);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index d766a64b1823..5b02880dfa20 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1374,14 +1374,11 @@ esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb)
 	struct mlx5_flow_table *nf_ft, *ft;
 	struct mlx5_chains_attr attr = {};
 	struct mlx5_fs_chains *chains;
-	u32 fdb_max;
 	int err;
 
-	fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size);
-
 	esw_init_chains_offload_flags(esw, &attr.flags);
 	attr.ns = MLX5_FLOW_NAMESPACE_FDB;
-	attr.max_ft_sz = fdb_max;
+	attr.fs_base_prio = FDB_TC_OFFLOAD;
 	attr.max_grp_num = esw->params.large_group_num;
 	attr.default_ft = miss_fdb;
 	attr.mapping = esw->offloads.reg_c0_obj_pool;
@@ -1392,6 +1389,7 @@ esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb)
 		esw_warn(dev, "Failed to create fdb chains err(%d)\n", err);
 		return err;
 	}
+	mlx5_chains_print_info(chains);
 
 	esw->fdb_table.offloads.esw_chains_priv = chains;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c
index 81ed91fee59b..db9df9798ffa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c
@@ -14,10 +14,8 @@
 #define chains_lock(chains) ((chains)->lock)
 #define chains_ht(chains) ((chains)->chains_ht)
 #define prios_ht(chains) ((chains)->prios_ht)
-#define tc_default_ft(chains) ((chains)->tc_default_ft)
-#define tc_end_ft(chains) ((chains)->tc_end_ft)
-#define ns_to_chains_fs_prio(ns) ((ns) == MLX5_FLOW_NAMESPACE_FDB ? \
-				  FDB_TC_OFFLOAD : MLX5E_TC_PRIO)
+#define chains_default_ft(chains) ((chains)->chains_default_ft)
+#define chains_end_ft(chains) ((chains)->chains_end_ft)
 #define FT_TBL_SZ (64 * 1024)
 
 struct mlx5_fs_chains {
@@ -28,13 +26,15 @@ struct mlx5_fs_chains {
 	/* Protects above chains_ht and prios_ht */
 	struct mutex lock;
 
-	struct mlx5_flow_table *tc_default_ft;
-	struct mlx5_flow_table *tc_end_ft;
+	struct mlx5_flow_table *chains_default_ft;
+	struct mlx5_flow_table *chains_end_ft;
 	struct mapping_ctx *chains_mapping;
 
 	enum mlx5_flow_namespace_type ns;
 	u32 group_num;
 	u32 flags;
+	int fs_base_prio;
+	int fs_base_level;
 };
 
 struct fs_chain {
@@ -145,7 +145,7 @@ void
 mlx5_chains_set_end_ft(struct mlx5_fs_chains *chains,
 		       struct mlx5_flow_table *ft)
 {
-	tc_end_ft(chains) = ft;
+	chains_end_ft(chains) = ft;
 }
 
 static struct mlx5_flow_table *
@@ -164,11 +164,11 @@ mlx5_chains_create_table(struct mlx5_fs_chains *chains,
 	sz = (chain == mlx5_chains_get_nf_ft_chain(chains)) ? FT_TBL_SZ : POOL_NEXT_SIZE;
 	ft_attr.max_fte = sz;
 
-	/* We use tc_default_ft(chains) as the table's next_ft till
+	/* We use chains_default_ft(chains) as the table's next_ft till
 	 * ignore_flow_level is allowed on FT creation and not just for FTEs.
 	 * Instead caller should add an explicit miss rule if needed.
 	 */
-	ft_attr.next_ft = tc_default_ft(chains);
+	ft_attr.next_ft = chains_default_ft(chains);
 
 	/* The root table(chain 0, prio 1, level 0) is required to be
 	 * connected to the previous fs_core managed prio.
@@ -177,22 +177,22 @@ mlx5_chains_create_table(struct mlx5_fs_chains *chains,
 	 */
 	if (!mlx5_chains_ignore_flow_level_supported(chains) ||
 	    (chain == 0 && prio == 1 && level == 0)) {
-		ft_attr.level = level;
-		ft_attr.prio = prio - 1;
+		ft_attr.level = chains->fs_base_level;
+		ft_attr.prio = chains->fs_base_prio;
 		ns = (chains->ns == MLX5_FLOW_NAMESPACE_FDB) ?
 			mlx5_get_fdb_sub_ns(chains->dev, chain) :
 			mlx5_get_flow_namespace(chains->dev, chains->ns);
 	} else {
 		ft_attr.flags |= MLX5_FLOW_TABLE_UNMANAGED;
-		ft_attr.prio = ns_to_chains_fs_prio(chains->ns);
+		ft_attr.prio = chains->fs_base_prio;
 		/* Firmware doesn't allow us to create another level 0 table,
-		 * so we create all unmanaged tables as level 1.
+		 * so we create all unmanaged tables as level 1 (base + 1).
 		 *
 		 * To connect them, we use explicit miss rules with
 		 * ignore_flow_level. Caller is responsible to create
 		 * these rules (if needed).
 		 */
-		ft_attr.level = 1;
+		ft_attr.level = chains->fs_base_level + 1;
 		ns = mlx5_get_flow_namespace(chains->dev, chains->ns);
 	}
 
@@ -220,7 +220,8 @@ create_chain_restore(struct fs_chain *chain)
 	int err;
 
 	if (chain->chain == mlx5_chains_get_nf_ft_chain(chains) ||
-	    !mlx5_chains_prios_supported(chains))
+	    !mlx5_chains_prios_supported(chains) ||
+	    !chains->chains_mapping)
 		return 0;
 
 	err = mlx5_chains_get_chain_mapping(chains, chain->chain, &index);
@@ -380,7 +381,7 @@ mlx5_chains_add_miss_rule(struct fs_chain *chain,
 	dest.type  = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dest.ft = next_ft;
 
-	if (next_ft == tc_end_ft(chains) &&
+	if (chains->chains_mapping && next_ft == chains_end_ft(chains) &&
 	    chain->chain != mlx5_chains_get_nf_ft_chain(chains) &&
 	    mlx5_chains_prios_supported(chains)) {
 		act.modify_hdr = chain->miss_modify_hdr;
@@ -494,8 +495,8 @@ mlx5_chains_create_prio(struct mlx5_fs_chains *chains,
 
 	/* Default miss for each chain: */
 	next_ft = (chain == mlx5_chains_get_nf_ft_chain(chains)) ?
-		  tc_default_ft(chains) :
-		  tc_end_ft(chains);
+		  chains_default_ft(chains) :
+		  chains_end_ft(chains);
 	list_for_each(pos, &chain_s->prios_list) {
 		struct prio *p = list_entry(pos, struct prio, list);
 
@@ -681,7 +682,7 @@ err_get_prio:
 struct mlx5_flow_table *
 mlx5_chains_get_tc_end_ft(struct mlx5_fs_chains *chains)
 {
-	return tc_end_ft(chains);
+	return chains_end_ft(chains);
 }
 
 struct mlx5_flow_table *
@@ -718,48 +719,38 @@ mlx5_chains_destroy_global_table(struct mlx5_fs_chains *chains,
 static struct mlx5_fs_chains *
 mlx5_chains_init(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr)
 {
-	struct mlx5_fs_chains *chains_priv;
-	u32 max_flow_counter;
+	struct mlx5_fs_chains *chains;
 	int err;
 
-	chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL);
-	if (!chains_priv)
+	chains = kzalloc(sizeof(*chains), GFP_KERNEL);
+	if (!chains)
 		return ERR_PTR(-ENOMEM);
 
-	max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) |
-			    MLX5_CAP_GEN(dev, max_flow_counter_15_0);
-
-	mlx5_core_dbg(dev,
-		      "Init flow table chains, max counters(%d), groups(%d), max flow table size(%d)\n",
-		      max_flow_counter, attr->max_grp_num, attr->max_ft_sz);
-
-	chains_priv->dev = dev;
-	chains_priv->flags = attr->flags;
-	chains_priv->ns = attr->ns;
-	chains_priv->group_num = attr->max_grp_num;
-	chains_priv->chains_mapping = attr->mapping;
-	tc_default_ft(chains_priv) = tc_end_ft(chains_priv) = attr->default_ft;
+	chains->dev = dev;
+	chains->flags = attr->flags;
+	chains->ns = attr->ns;
+	chains->group_num = attr->max_grp_num;
+	chains->chains_mapping = attr->mapping;
+	chains->fs_base_prio = attr->fs_base_prio;
+	chains->fs_base_level = attr->fs_base_level;
+	chains_default_ft(chains) = chains_end_ft(chains) = attr->default_ft;
 
-	mlx5_core_info(dev, "Supported tc offload range - chains: %u, prios: %u\n",
-		       mlx5_chains_get_chain_range(chains_priv),
-		       mlx5_chains_get_prio_range(chains_priv));
-
-	err = rhashtable_init(&chains_ht(chains_priv), &chain_params);
+	err = rhashtable_init(&chains_ht(chains), &chain_params);
 	if (err)
 		goto init_chains_ht_err;
 
-	err = rhashtable_init(&prios_ht(chains_priv), &prio_params);
+	err = rhashtable_init(&prios_ht(chains), &prio_params);
 	if (err)
 		goto init_prios_ht_err;
 
-	mutex_init(&chains_lock(chains_priv));
+	mutex_init(&chains_lock(chains));
 
-	return chains_priv;
+	return chains;
 
 init_prios_ht_err:
-	rhashtable_destroy(&chains_ht(chains_priv));
+	rhashtable_destroy(&chains_ht(chains));
 init_chains_ht_err:
-	kfree(chains_priv);
+	kfree(chains);
 	return ERR_PTR(err);
 }
 
@@ -808,3 +799,9 @@ mlx5_chains_put_chain_mapping(struct mlx5_fs_chains *chains, u32 chain_mapping)
 
 	return mapping_remove(ctx, chain_mapping);
 }
+
+void
+mlx5_chains_print_info(struct mlx5_fs_chains *chains)
+{
+	mlx5_core_dbg(chains->dev, "Flow table chains groups(%d)\n", chains->group_num);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h
index d50bdb226cef..8972fe05723a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h
@@ -17,8 +17,9 @@ enum mlx5_chains_flags {
 
 struct mlx5_chains_attr {
 	enum mlx5_flow_namespace_type ns;
+	int fs_base_prio;
+	int fs_base_level;
 	u32 flags;
-	u32 max_ft_sz;
 	u32 max_grp_num;
 	struct mlx5_flow_table *default_ft;
 	struct mapping_ctx *mapping;
@@ -68,6 +69,8 @@ void mlx5_chains_destroy(struct mlx5_fs_chains *chains);
 void
 mlx5_chains_set_end_ft(struct mlx5_fs_chains *chains,
 		       struct mlx5_flow_table *ft);
+void
+mlx5_chains_print_info(struct mlx5_fs_chains *chains);
 
 #else /* CONFIG_MLX5_CLS_ACT */
 
@@ -89,7 +92,9 @@ static inline struct mlx5_fs_chains *
 mlx5_chains_create(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr)
 { return NULL; }
 static inline void
-mlx5_chains_destroy(struct mlx5_fs_chains *chains) {};
+mlx5_chains_destroy(struct mlx5_fs_chains *chains) {}
+static inline void
+mlx5_chains_print_info(struct mlx5_fs_chains *chains) {}
 
 #endif /* CONFIG_MLX5_CLS_ACT */
 
-- 
cgit v1.2.3-70-g09d2


From 664eab8a02d3331f2facf6c4359f4baaeefe6a3b Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:37 +0200
Subject: net/mlx5: fs_core: Allow ignore_flow_level on TX dest

ignore_flow_level is also supported by firmware on TX,
remove this limitation.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Raed Salem <raeds@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/d0025722bfac0a82da758eb540fbf1ff3cacdf74.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 731acbe22dc7..3ade166073fa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -1762,7 +1762,8 @@ static bool dest_is_valid(struct mlx5_flow_destination *dest,
 
 	if (ignore_level) {
 		if (ft->type != FS_FT_FDB &&
-		    ft->type != FS_FT_NIC_RX)
+		    ft->type != FS_FT_NIC_RX &&
+		    ft->type != FS_FT_NIC_TX)
 			return false;
 
 		if (dest->type == MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE &&
-- 
cgit v1.2.3-70-g09d2


From fa5aa2f890733407e6df28b878926ae66a351b8a Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:38 +0200
Subject: net/mlx5e: Use chains for IPsec policy priority offload

Currently, policy priority field is ignored and so order
of matching is unpredictable.

Use chains for RX/TX policy offload to support the
priority field.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/9ef3ef88858217932696ad413b1b147b799a11be.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   |  18 +-
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |   2 +
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 362 ++++++++++++++++-----
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    |  17 +-
 4 files changed, 315 insertions(+), 84 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 7b0d3de0ec6c..83012bece548 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -499,7 +499,8 @@ static void mlx5e_xfrm_update_curlft(struct xfrm_state *x)
 	mlx5e_ipsec_aso_update_curlft(sa_entry, &x->curlft.packets);
 }
 
-static int mlx5e_xfrm_validate_policy(struct xfrm_policy *x,
+static int mlx5e_xfrm_validate_policy(struct mlx5_core_dev *mdev,
+				      struct xfrm_policy *x,
 				      struct netlink_ext_ack *extack)
 {
 	if (x->type != XFRM_POLICY_TYPE_MAIN) {
@@ -535,6 +536,18 @@ static int mlx5e_xfrm_validate_policy(struct xfrm_policy *x,
 		return -EINVAL;
 	}
 
+	if (x->priority) {
+		if (!(mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PRIO)) {
+			NL_SET_ERR_MSG_MOD(extack, "Device does not support policy priority");
+			return -EINVAL;
+		}
+
+		if (x->priority == U32_MAX) {
+			NL_SET_ERR_MSG_MOD(extack, "Device does not support requested policy priority");
+			return -EINVAL;
+		}
+	}
+
 	return 0;
 }
 
@@ -560,6 +573,7 @@ mlx5e_ipsec_build_accel_pol_attrs(struct mlx5e_ipsec_pol_entry *pol_entry,
 	attrs->upspec.sport = ntohs(sel->sport);
 	attrs->upspec.sport_mask = ntohs(sel->sport_mask);
 	attrs->upspec.proto = sel->proto;
+	attrs->prio = x->priority;
 }
 
 static int mlx5e_xfrm_add_policy(struct xfrm_policy *x,
@@ -576,7 +590,7 @@ static int mlx5e_xfrm_add_policy(struct xfrm_policy *x,
 		return -EOPNOTSUPP;
 	}
 
-	err = mlx5e_xfrm_validate_policy(x, extack);
+	err = mlx5e_xfrm_validate_policy(priv->mdev, x, extack);
 	if (err)
 		return err;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 12f044330639..b36e047396da 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -94,6 +94,7 @@ enum mlx5_ipsec_cap {
 	MLX5_IPSEC_CAP_ESN		= 1 << 1,
 	MLX5_IPSEC_CAP_PACKET_OFFLOAD	= 1 << 2,
 	MLX5_IPSEC_CAP_ROCE             = 1 << 3,
+	MLX5_IPSEC_CAP_PRIO             = 1 << 4,
 };
 
 struct mlx5e_priv;
@@ -198,6 +199,7 @@ struct mlx5_accel_pol_xfrm_attrs {
 	u8 type : 2;
 	u8 dir : 2;
 	u32 reqid;
+	u32 prio;
 };
 
 struct mlx5e_ipsec_pol_entry {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 9871ba1b25ff..0c9640d575a7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -7,6 +7,7 @@
 #include "ipsec.h"
 #include "fs_core.h"
 #include "lib/ipsec_fs_roce.h"
+#include "lib/fs_chains.h"
 
 #define NUM_IPSEC_FTE BIT(15)
 
@@ -34,6 +35,7 @@ struct mlx5e_ipsec_rx {
 	struct mlx5e_ipsec_miss sa;
 	struct mlx5e_ipsec_rule status;
 	struct mlx5e_ipsec_fc *fc;
+	struct mlx5_fs_chains *chains;
 };
 
 struct mlx5e_ipsec_tx {
@@ -41,6 +43,7 @@ struct mlx5e_ipsec_tx {
 	struct mlx5e_ipsec_miss pol;
 	struct mlx5_flow_namespace *ns;
 	struct mlx5e_ipsec_fc *fc;
+	struct mlx5_fs_chains *chains;
 };
 
 /* IPsec RX flow steering */
@@ -51,6 +54,67 @@ static enum mlx5_traffic_types family2tt(u32 family)
 	return MLX5_TT_IPV6_IPSEC_ESP;
 }
 
+static struct mlx5e_ipsec_rx *ipsec_rx(struct mlx5e_ipsec *ipsec, u32 family)
+{
+	if (family == AF_INET)
+		return ipsec->rx_ipv4;
+
+	return ipsec->rx_ipv6;
+}
+
+static struct mlx5_fs_chains *
+ipsec_chains_create(struct mlx5_core_dev *mdev, struct mlx5_flow_table *miss_ft,
+		    enum mlx5_flow_namespace_type ns, int base_prio,
+		    int base_level, struct mlx5_flow_table **root_ft)
+{
+	struct mlx5_chains_attr attr = {};
+	struct mlx5_fs_chains *chains;
+	struct mlx5_flow_table *ft;
+	int err;
+
+	attr.flags = MLX5_CHAINS_AND_PRIOS_SUPPORTED |
+		     MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED;
+	attr.max_grp_num = 2;
+	attr.default_ft = miss_ft;
+	attr.ns = ns;
+	attr.fs_base_prio = base_prio;
+	attr.fs_base_level = base_level;
+	chains = mlx5_chains_create(mdev, &attr);
+	if (IS_ERR(chains))
+		return chains;
+
+	/* Create chain 0, prio 1, level 0 to connect chains to prev in fs_core */
+	ft = mlx5_chains_get_table(chains, 0, 1, 0);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		goto err_chains_get;
+	}
+
+	*root_ft = ft;
+	return chains;
+
+err_chains_get:
+	mlx5_chains_destroy(chains);
+	return ERR_PTR(err);
+}
+
+static void ipsec_chains_destroy(struct mlx5_fs_chains *chains)
+{
+	mlx5_chains_put_table(chains, 0, 1, 0);
+	mlx5_chains_destroy(chains);
+}
+
+static struct mlx5_flow_table *
+ipsec_chains_get_table(struct mlx5_fs_chains *chains, u32 prio)
+{
+	return mlx5_chains_get_table(chains, 0, prio + 1, 0);
+}
+
+static void ipsec_chains_put_table(struct mlx5_fs_chains *chains, u32 prio)
+{
+	mlx5_chains_put_table(chains, 0, prio + 1, 0);
+}
+
 static struct mlx5_flow_table *ipsec_ft_create(struct mlx5_flow_namespace *ns,
 					       int level, int prio,
 					       int max_num_groups)
@@ -170,9 +234,18 @@ out:
 static void rx_destroy(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 		       struct mlx5e_ipsec_rx *rx, u32 family)
 {
-	mlx5_del_flow_rules(rx->pol.rule);
-	mlx5_destroy_flow_group(rx->pol.group);
-	mlx5_destroy_flow_table(rx->ft.pol);
+	struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false);
+
+	/* disconnect */
+	mlx5_ttc_fwd_default_dest(ttc, family2tt(family));
+
+	if (rx->chains) {
+		ipsec_chains_destroy(rx->chains);
+	} else {
+		mlx5_del_flow_rules(rx->pol.rule);
+		mlx5_destroy_flow_group(rx->pol.group);
+		mlx5_destroy_flow_table(rx->ft.pol);
+	}
 
 	mlx5_del_flow_rules(rx->sa.rule);
 	mlx5_destroy_flow_group(rx->sa.group);
@@ -238,6 +311,20 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	if (err)
 		goto err_fs;
 
+	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PRIO) {
+		rx->chains = ipsec_chains_create(mdev, rx->ft.sa,
+						 MLX5_FLOW_NAMESPACE_KERNEL,
+						 MLX5E_NIC_PRIO,
+						 MLX5E_ACCEL_FS_POL_FT_LEVEL,
+						 &rx->ft.pol);
+		if (IS_ERR(rx->chains)) {
+			err = PTR_ERR(rx->chains);
+			goto err_pol_ft;
+		}
+
+		goto connect;
+	}
+
 	ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_POL_FT_LEVEL, MLX5E_NIC_PRIO,
 			     2);
 	if (IS_ERR(ft)) {
@@ -252,6 +339,12 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	if (err)
 		goto err_pol_miss;
 
+connect:
+	/* connect */
+	memset(dest, 0x00, sizeof(*dest));
+	dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest[0].ft = rx->ft.pol;
+	mlx5_ttc_fwd_dest(ttc, family2tt(family), &dest[0]);
 	return 0;
 
 err_pol_miss:
@@ -271,69 +364,112 @@ err_fs_ft_status:
 	return err;
 }
 
-static struct mlx5e_ipsec_rx *rx_ft_get(struct mlx5_core_dev *mdev,
-					struct mlx5e_ipsec *ipsec, u32 family)
+static int rx_get(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
+		  struct mlx5e_ipsec_rx *rx, u32 family)
 {
-	struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false);
-	struct mlx5_flow_destination dest = {};
-	struct mlx5e_ipsec_rx *rx;
-	int err = 0;
-
-	if (family == AF_INET)
-		rx = ipsec->rx_ipv4;
-	else
-		rx = ipsec->rx_ipv6;
+	int err;
 
-	mutex_lock(&rx->ft.mutex);
 	if (rx->ft.refcnt)
 		goto skip;
 
-	/* create FT */
 	err = rx_create(mdev, ipsec, rx, family);
 	if (err)
-		goto out;
-
-	/* connect */
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-	dest.ft = rx->ft.pol;
-	mlx5_ttc_fwd_dest(ttc, family2tt(family), &dest);
+		return err;
 
 skip:
 	rx->ft.refcnt++;
-out:
+	return 0;
+}
+
+static void rx_put(struct mlx5e_ipsec *ipsec, struct mlx5e_ipsec_rx *rx,
+		   u32 family)
+{
+	if (--rx->ft.refcnt)
+		return;
+
+	rx_destroy(ipsec->mdev, ipsec, rx, family);
+}
+
+static struct mlx5e_ipsec_rx *rx_ft_get(struct mlx5_core_dev *mdev,
+					struct mlx5e_ipsec *ipsec, u32 family)
+{
+	struct mlx5e_ipsec_rx *rx = ipsec_rx(ipsec, family);
+	int err;
+
+	mutex_lock(&rx->ft.mutex);
+	err = rx_get(mdev, ipsec, rx, family);
 	mutex_unlock(&rx->ft.mutex);
 	if (err)
 		return ERR_PTR(err);
+
 	return rx;
 }
 
-static void rx_ft_put(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
-		      u32 family)
+static struct mlx5_flow_table *rx_ft_get_policy(struct mlx5_core_dev *mdev,
+						struct mlx5e_ipsec *ipsec,
+						u32 family, u32 prio)
 {
-	struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false);
-	struct mlx5e_ipsec_rx *rx;
+	struct mlx5e_ipsec_rx *rx = ipsec_rx(ipsec, family);
+	struct mlx5_flow_table *ft;
+	int err;
 
-	if (family == AF_INET)
-		rx = ipsec->rx_ipv4;
-	else
-		rx = ipsec->rx_ipv6;
+	mutex_lock(&rx->ft.mutex);
+	err = rx_get(mdev, ipsec, rx, family);
+	if (err)
+		goto err_get;
+
+	ft = rx->chains ? ipsec_chains_get_table(rx->chains, prio) : rx->ft.pol;
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		goto err_get_ft;
+	}
+
+	mutex_unlock(&rx->ft.mutex);
+	return ft;
+
+err_get_ft:
+	rx_put(ipsec, rx, family);
+err_get:
+	mutex_unlock(&rx->ft.mutex);
+	return ERR_PTR(err);
+}
+
+static void rx_ft_put(struct mlx5e_ipsec *ipsec, u32 family)
+{
+	struct mlx5e_ipsec_rx *rx = ipsec_rx(ipsec, family);
 
 	mutex_lock(&rx->ft.mutex);
-	rx->ft.refcnt--;
-	if (rx->ft.refcnt)
-		goto out;
+	rx_put(ipsec, rx, family);
+	mutex_unlock(&rx->ft.mutex);
+}
 
-	/* disconnect */
-	mlx5_ttc_fwd_default_dest(ttc, family2tt(family));
+static void rx_ft_put_policy(struct mlx5e_ipsec *ipsec, u32 family, u32 prio)
+{
+	struct mlx5e_ipsec_rx *rx = ipsec_rx(ipsec, family);
 
-	/* remove FT */
-	rx_destroy(mdev, ipsec, rx, family);
+	mutex_lock(&rx->ft.mutex);
+	if (rx->chains)
+		ipsec_chains_put_table(rx->chains, prio);
 
-out:
+	rx_put(ipsec, rx, family);
 	mutex_unlock(&rx->ft.mutex);
 }
 
 /* IPsec TX flow steering */
+static void tx_destroy(struct mlx5e_ipsec_tx *tx, struct mlx5_ipsec_fs *roce)
+{
+	mlx5_ipsec_fs_roce_tx_destroy(roce);
+	if (tx->chains) {
+		ipsec_chains_destroy(tx->chains);
+	} else {
+		mlx5_del_flow_rules(tx->pol.rule);
+		mlx5_destroy_flow_group(tx->pol.group);
+		mlx5_destroy_flow_table(tx->ft.pol);
+	}
+
+	mlx5_destroy_flow_table(tx->ft.sa);
+}
+
 static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 		     struct mlx5_ipsec_fs *roce)
 {
@@ -347,6 +483,18 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 
 	tx->ft.sa = ft;
 
+	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PRIO) {
+		tx->chains = ipsec_chains_create(
+			mdev, tx->ft.sa, MLX5_FLOW_NAMESPACE_EGRESS_IPSEC, 0, 0,
+			&tx->ft.pol);
+		if (IS_ERR(tx->chains)) {
+			err = PTR_ERR(tx->chains);
+			goto err_pol_ft;
+		}
+
+		goto connect_roce;
+	}
+
 	ft = ipsec_ft_create(tx->ns, 0, 0, 2);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
@@ -356,44 +504,96 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dest.ft = tx->ft.sa;
 	err = ipsec_miss_create(mdev, tx->ft.pol, &tx->pol, &dest);
-	if (err)
-		goto err_pol_miss;
+	if (err) {
+		mlx5_destroy_flow_table(tx->ft.pol);
+		goto err_pol_ft;
+	}
 
+connect_roce:
 	err = mlx5_ipsec_fs_roce_tx_create(mdev, roce, tx->ft.pol);
 	if (err)
 		goto err_roce;
 	return 0;
 
 err_roce:
-	mlx5_del_flow_rules(tx->pol.rule);
-	mlx5_destroy_flow_group(tx->pol.group);
-err_pol_miss:
-	mlx5_destroy_flow_table(tx->ft.pol);
+	if (tx->chains) {
+		ipsec_chains_destroy(tx->chains);
+	} else {
+		mlx5_del_flow_rules(tx->pol.rule);
+		mlx5_destroy_flow_group(tx->pol.group);
+		mlx5_destroy_flow_table(tx->ft.pol);
+	}
 err_pol_ft:
 	mlx5_destroy_flow_table(tx->ft.sa);
 	return err;
 }
 
-static struct mlx5e_ipsec_tx *tx_ft_get(struct mlx5_core_dev *mdev,
-					struct mlx5e_ipsec *ipsec)
+static int tx_get(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
+		  struct mlx5e_ipsec_tx *tx)
 {
-	struct mlx5e_ipsec_tx *tx = ipsec->tx;
-	int err = 0;
+	int err;
 
-	mutex_lock(&tx->ft.mutex);
 	if (tx->ft.refcnt)
 		goto skip;
 
 	err = tx_create(mdev, tx, ipsec->roce);
 	if (err)
-		goto out;
+		return err;
 
 skip:
 	tx->ft.refcnt++;
-out:
+	return 0;
+}
+
+static void tx_put(struct mlx5e_ipsec *ipsec, struct mlx5e_ipsec_tx *tx)
+{
+	if (--tx->ft.refcnt)
+		return;
+
+	tx_destroy(tx, ipsec->roce);
+}
+
+static struct mlx5_flow_table *tx_ft_get_policy(struct mlx5_core_dev *mdev,
+						struct mlx5e_ipsec *ipsec,
+						u32 prio)
+{
+	struct mlx5e_ipsec_tx *tx = ipsec->tx;
+	struct mlx5_flow_table *ft;
+	int err;
+
+	mutex_lock(&tx->ft.mutex);
+	err = tx_get(mdev, ipsec, tx);
+	if (err)
+		goto err_get;
+
+	ft = tx->chains ? ipsec_chains_get_table(tx->chains, prio) : tx->ft.pol;
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		goto err_get_ft;
+	}
+
+	mutex_unlock(&tx->ft.mutex);
+	return ft;
+
+err_get_ft:
+	tx_put(ipsec, tx);
+err_get:
+	mutex_unlock(&tx->ft.mutex);
+	return ERR_PTR(err);
+}
+
+static struct mlx5e_ipsec_tx *tx_ft_get(struct mlx5_core_dev *mdev,
+					struct mlx5e_ipsec *ipsec)
+{
+	struct mlx5e_ipsec_tx *tx = ipsec->tx;
+	int err;
+
+	mutex_lock(&tx->ft.mutex);
+	err = tx_get(mdev, ipsec, tx);
 	mutex_unlock(&tx->ft.mutex);
 	if (err)
 		return ERR_PTR(err);
+
 	return tx;
 }
 
@@ -402,16 +602,19 @@ static void tx_ft_put(struct mlx5e_ipsec *ipsec)
 	struct mlx5e_ipsec_tx *tx = ipsec->tx;
 
 	mutex_lock(&tx->ft.mutex);
-	tx->ft.refcnt--;
-	if (tx->ft.refcnt)
-		goto out;
+	tx_put(ipsec, tx);
+	mutex_unlock(&tx->ft.mutex);
+}
 
-	mlx5_ipsec_fs_roce_tx_destroy(ipsec->roce);
-	mlx5_del_flow_rules(tx->pol.rule);
-	mlx5_destroy_flow_group(tx->pol.group);
-	mlx5_destroy_flow_table(tx->ft.pol);
-	mlx5_destroy_flow_table(tx->ft.sa);
-out:
+static void tx_ft_put_policy(struct mlx5e_ipsec *ipsec, u32 prio)
+{
+	struct mlx5e_ipsec_tx *tx = ipsec->tx;
+
+	mutex_lock(&tx->ft.mutex);
+	if (tx->chains)
+		ipsec_chains_put_table(tx->chains, prio);
+
+	tx_put(ipsec, tx);
 	mutex_unlock(&tx->ft.mutex);
 }
 
@@ -676,7 +879,7 @@ err_pkt_reformat:
 err_mod_header:
 	kvfree(spec);
 err_alloc:
-	rx_ft_put(mdev, ipsec, attrs->family);
+	rx_ft_put(ipsec, attrs->family);
 	return err;
 }
 
@@ -690,7 +893,7 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
 	struct mlx5e_ipsec_tx *tx;
-	int err = 0;
+	int err;
 
 	tx = tx_ft_get(mdev, ipsec);
 	if (IS_ERR(tx))
@@ -760,16 +963,17 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 {
 	struct mlx5_accel_pol_xfrm_attrs *attrs = &pol_entry->attrs;
 	struct mlx5_core_dev *mdev = mlx5e_ipsec_pol2dev(pol_entry);
+	struct mlx5e_ipsec_tx *tx = pol_entry->ipsec->tx;
 	struct mlx5_flow_destination dest[2] = {};
 	struct mlx5_flow_act flow_act = {};
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
-	struct mlx5e_ipsec_tx *tx;
+	struct mlx5_flow_table *ft;
 	int err, dstn = 0;
 
-	tx = tx_ft_get(mdev, pol_entry->ipsec);
-	if (IS_ERR(tx))
-		return PTR_ERR(tx);
+	ft = tx_ft_get_policy(mdev, pol_entry->ipsec, attrs->prio);
+	if (IS_ERR(ft))
+		return PTR_ERR(ft);
 
 	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec) {
@@ -811,7 +1015,7 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	dest[dstn].ft = tx->ft.sa;
 	dest[dstn].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dstn++;
-	rule = mlx5_add_flow_rules(tx->ft.pol, spec, &flow_act, dest, dstn);
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, dest, dstn);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		mlx5_core_err(mdev, "fail to add TX ipsec rule err=%d\n", err);
@@ -828,7 +1032,7 @@ err_action:
 err_mod_header:
 	kvfree(spec);
 err_alloc:
-	tx_ft_put(pol_entry->ipsec);
+	tx_ft_put_policy(pol_entry->ipsec, attrs->prio);
 	return err;
 }
 
@@ -840,12 +1044,15 @@ static int rx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	struct mlx5_flow_act flow_act = {};
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
+	struct mlx5_flow_table *ft;
 	struct mlx5e_ipsec_rx *rx;
 	int err, dstn = 0;
 
-	rx = rx_ft_get(mdev, pol_entry->ipsec, attrs->family);
-	if (IS_ERR(rx))
-		return PTR_ERR(rx);
+	ft = rx_ft_get_policy(mdev, pol_entry->ipsec, attrs->family, attrs->prio);
+	if (IS_ERR(ft))
+		return PTR_ERR(ft);
+
+	rx = ipsec_rx(pol_entry->ipsec, attrs->family);
 
 	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
 	if (!spec) {
@@ -880,7 +1087,7 @@ static int rx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	dest[dstn].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dest[dstn].ft = rx->ft.sa;
 	dstn++;
-	rule = mlx5_add_flow_rules(rx->ft.pol, spec, &flow_act, dest, dstn);
+	rule = mlx5_add_flow_rules(ft, spec, &flow_act, dest, dstn);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		mlx5_core_err(mdev, "Fail to add RX IPsec policy rule err=%d\n", err);
@@ -894,7 +1101,7 @@ static int rx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 err_action:
 	kvfree(spec);
 err_alloc:
-	rx_ft_put(mdev, pol_entry->ipsec, attrs->family);
+	rx_ft_put_policy(pol_entry->ipsec, attrs->family, attrs->prio);
 	return err;
 }
 
@@ -1032,7 +1239,7 @@ void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	}
 
 	mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
-	rx_ft_put(mdev, sa_entry->ipsec, sa_entry->attrs.family);
+	rx_ft_put(sa_entry->ipsec, sa_entry->attrs.family);
 }
 
 int mlx5e_accel_ipsec_fs_add_pol(struct mlx5e_ipsec_pol_entry *pol_entry)
@@ -1051,12 +1258,13 @@ void mlx5e_accel_ipsec_fs_del_pol(struct mlx5e_ipsec_pol_entry *pol_entry)
 	mlx5_del_flow_rules(ipsec_rule->rule);
 
 	if (pol_entry->attrs.dir == XFRM_DEV_OFFLOAD_IN) {
-		rx_ft_put(mdev, pol_entry->ipsec, pol_entry->attrs.family);
+		rx_ft_put_policy(pol_entry->ipsec, pol_entry->attrs.family,
+				 pol_entry->attrs.prio);
 		return;
 	}
 
 	mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
-	tx_ft_put(pol_entry->ipsec);
+	tx_ft_put_policy(pol_entry->ipsec, pol_entry->attrs.prio);
 }
 
 void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 5fa7a4c40429..67be8d36bb76 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -36,11 +36,18 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev)
 	    MLX5_CAP_ETH(mdev, insert_trailer) && MLX5_CAP_ETH(mdev, swp))
 		caps |= MLX5_IPSEC_CAP_CRYPTO;
 
-	if (MLX5_CAP_IPSEC(mdev, ipsec_full_offload) &&
-	    MLX5_CAP_FLOWTABLE_NIC_TX(mdev, reformat_add_esp_trasport) &&
-	    MLX5_CAP_FLOWTABLE_NIC_RX(mdev, reformat_del_esp_trasport) &&
-	    MLX5_CAP_FLOWTABLE_NIC_RX(mdev, decap))
-		caps |= MLX5_IPSEC_CAP_PACKET_OFFLOAD;
+	if (MLX5_CAP_IPSEC(mdev, ipsec_full_offload)) {
+		if (MLX5_CAP_FLOWTABLE_NIC_TX(mdev,
+					      reformat_add_esp_trasport) &&
+		    MLX5_CAP_FLOWTABLE_NIC_RX(mdev,
+					      reformat_del_esp_trasport) &&
+		    MLX5_CAP_FLOWTABLE_NIC_RX(mdev, decap))
+			caps |= MLX5_IPSEC_CAP_PACKET_OFFLOAD;
+
+		if (MLX5_CAP_FLOWTABLE_NIC_TX(mdev, ignore_flow_level) &&
+		    MLX5_CAP_FLOWTABLE_NIC_RX(mdev, ignore_flow_level))
+			caps |= MLX5_IPSEC_CAP_PRIO;
+	}
 
 	if (mlx5_get_roce_state(mdev) &&
 	    MLX5_CAP_GEN_2(mdev, flow_table_type_2_type) & MLX5_FT_NIC_RX_2_NIC_RX_RDMA &&
-- 
cgit v1.2.3-70-g09d2


From e0aeb9b90acf6ee7c2d11141522ffbb5481734d3 Mon Sep 17 00:00:00 2001
From: Raed Salem <raeds@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:39 +0200
Subject: xfrm: add new device offload acquire flag

During XFRM acquire flow, a default SA is created to be updated later,
once acquire netlink message is handled in user space. When the relevant
policy is offloaded this default SA is also offloaded to IPsec offload
supporting driver, however this SA does not have context suitable for
offloading in HW, nor is interesting to offload to HW, consequently needs
a special driver handling apart from other offloaded SA(s).
Add a special flag that marks such SA so driver can handle it correctly.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/f5da0834d8c6b82ab9ba38bd4a0c55e71f0e3dab.1678714336.git.leon@kernel.org
Acked-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 include/net/xfrm.h    | 5 +++++
 net/xfrm/xfrm_state.c | 1 +
 2 files changed, 6 insertions(+)

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 3e1f70e8e424..33ee3f5936e6 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -138,6 +138,10 @@ enum {
 	XFRM_DEV_OFFLOAD_PACKET,
 };
 
+enum {
+	XFRM_DEV_OFFLOAD_FLAG_ACQ = 1,
+};
+
 struct xfrm_dev_offload {
 	struct net_device	*dev;
 	netdevice_tracker	dev_tracker;
@@ -145,6 +149,7 @@ struct xfrm_dev_offload {
 	unsigned long		offload_handle;
 	u8			dir : 2;
 	u8			type : 2;
+	u8			flags : 2;
 };
 
 struct xfrm_mode {
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 2ab3e09e2227..7cca0a1fa5ff 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1272,6 +1272,7 @@ found:
 			xso->dir = xdo->dir;
 			xso->dev = xdo->dev;
 			xso->real_dev = xdo->real_dev;
+			xso->flags = XFRM_DEV_OFFLOAD_FLAG_ACQ;
 			netdev_tracker_alloc(xso->dev, &xso->dev_tracker,
 					     GFP_ATOMIC);
 			error = xso->dev->xfrmdev_ops->xdo_dev_state_add(x, NULL);
-- 
cgit v1.2.3-70-g09d2


From c9fa320b00cff04980b8514d497068e59a8ee131 Mon Sep 17 00:00:00 2001
From: Raed Salem <raeds@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:40 +0200
Subject: xfrm: copy_to_user_state fetch offloaded SA packets/bytes statistics

Both in RX and TX, the traffic that performs IPsec packet offload
transformation is accounted by HW only. Consequently, the HW should
be queried for packets/bytes statistics when user asks for such
transformation data.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/d90ec74186452b1509ee94875d942cb777b7181e.1678714336.git.leon@kernel.org
Acked-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 net/xfrm/xfrm_user.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index cf5172d4ce68..5eee905b2450 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -901,6 +901,8 @@ static void copy_to_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p)
 	memcpy(&p->id, &x->id, sizeof(p->id));
 	memcpy(&p->sel, &x->sel, sizeof(p->sel));
 	memcpy(&p->lft, &x->lft, sizeof(p->lft));
+	if (x->xso.dev)
+		xfrm_dev_state_update_curlft(x);
 	memcpy(&p->curlft, &x->curlft, sizeof(p->curlft));
 	put_unaligned(x->stats.replay_window, &p->stats.replay_window);
 	put_unaligned(x->stats.replay, &p->stats.replay);
-- 
cgit v1.2.3-70-g09d2


From b3beba1fb404dd713a325cb241fe780857172225 Mon Sep 17 00:00:00 2001
From: Raed Salem <raeds@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:41 +0200
Subject: net/mlx5e: Allow policies with reqid 0, to support IKE policy holes

IKE policies hole, is special policy that exists to allow for IKE
traffic to bypass IPsec encryption even though there is already a
policies and SA(s) configured on same endpoints, these policies
does not nessecarly have the reqid configured, so need to add
an exception for such policies. These kind of policies are allowed
under the condition that at least upper protocol and/or ips
are not 0.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/cbcadde312c24de74c47d9b0616f86a5818cc9bf.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 11 ++--
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  7 +++
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 68 ++++++++++++++--------
 3 files changed, 59 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 83012bece548..9cc59dc8b592 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -503,6 +503,8 @@ static int mlx5e_xfrm_validate_policy(struct mlx5_core_dev *mdev,
 				      struct xfrm_policy *x,
 				      struct netlink_ext_ack *extack)
 {
+	struct xfrm_selector *sel = &x->selector;
+
 	if (x->type != XFRM_POLICY_TYPE_MAIN) {
 		NL_SET_ERR_MSG_MOD(extack, "Cannot offload non-main policy types");
 		return -EINVAL;
@@ -520,8 +522,9 @@ static int mlx5e_xfrm_validate_policy(struct mlx5_core_dev *mdev,
 		return -EINVAL;
 	}
 
-	if (!x->xfrm_vec[0].reqid) {
-		NL_SET_ERR_MSG_MOD(extack, "Cannot offload policy without reqid");
+	if (!x->xfrm_vec[0].reqid && sel->proto == IPPROTO_IP &&
+	    addr6_all_zero(sel->saddr.a6) && addr6_all_zero(sel->daddr.a6)) {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported policy with reqid 0 without at least one of upper protocol or ip addr(s) different than 0");
 		return -EINVAL;
 	}
 
@@ -530,8 +533,8 @@ static int mlx5e_xfrm_validate_policy(struct mlx5_core_dev *mdev,
 		return -EINVAL;
 	}
 
-	if (x->selector.proto != IPPROTO_IP &&
-	    (x->selector.proto != IPPROTO_UDP || x->xdo.dir != XFRM_DEV_OFFLOAD_OUT)) {
+	if (sel->proto != IPPROTO_IP &&
+	    (sel->proto != IPPROTO_UDP || x->xdo.dir != XFRM_DEV_OFFLOAD_OUT)) {
 		NL_SET_ERR_MSG_MOD(extack, "Device does not support upper protocol other than UDP, and only Tx direction");
 		return -EINVAL;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index b36e047396da..f3e81c3383e5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -254,6 +254,13 @@ mlx5e_ipsec_pol2dev(struct mlx5e_ipsec_pol_entry *pol_entry)
 {
 	return pol_entry->ipsec->mdev;
 }
+
+static inline bool addr6_all_zero(__be32 *addr6)
+{
+	static const __be32 zaddr6[4] = {};
+
+	return !memcmp(addr6, zaddr6, sizeof(*zaddr6));
+}
 #else
 static inline void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 0c9640d575a7..9f694a8e21fd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -621,37 +621,53 @@ static void tx_ft_put_policy(struct mlx5e_ipsec *ipsec, u32 prio)
 static void setup_fte_addr4(struct mlx5_flow_spec *spec, __be32 *saddr,
 			    __be32 *daddr)
 {
+	if (!*saddr && !*daddr)
+		return;
+
 	spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS;
 
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_version);
 	MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_version, 4);
 
-	memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
-			    outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), saddr, 4);
-	memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
-			    outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), daddr, 4);
-	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
-			 outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4);
-	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
-			 outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4);
+	if (*saddr) {
+		memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
+				    outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), saddr, 4);
+		MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+				 outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4);
+	}
+
+	if (*daddr) {
+		memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
+				    outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), daddr, 4);
+		MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+				 outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4);
+	}
 }
 
 static void setup_fte_addr6(struct mlx5_flow_spec *spec, __be32 *saddr,
 			    __be32 *daddr)
 {
+	if (addr6_all_zero(saddr) && addr6_all_zero(daddr))
+		return;
+
 	spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS;
 
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_version);
 	MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_version, 6);
 
-	memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
-			    outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), saddr, 16);
-	memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
-			    outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), daddr, 16);
-	memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
-			    outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), 0xff, 16);
-	memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
-			    outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), 0xff, 16);
+	if (!addr6_all_zero(saddr)) {
+		memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
+				    outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), saddr, 16);
+		memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+				    outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), 0xff, 16);
+	}
+
+	if (!addr6_all_zero(daddr)) {
+		memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value,
+				    outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), daddr, 16);
+		memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+				    outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), 0xff, 16);
+	}
 }
 
 static void setup_fte_esp(struct mlx5_flow_spec *spec)
@@ -920,7 +936,8 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 		setup_fte_reg_a(spec);
 		break;
 	case XFRM_DEV_OFFLOAD_PACKET:
-		setup_fte_reg_c0(spec, attrs->reqid);
+		if (attrs->reqid)
+			setup_fte_reg_c0(spec, attrs->reqid);
 		err = setup_pkt_reformat(mdev, attrs, &flow_act);
 		if (err)
 			goto err_pkt_reformat;
@@ -989,10 +1006,12 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	setup_fte_no_frags(spec);
 	setup_fte_upper_proto_match(spec, &attrs->upspec);
 
-	err = setup_modify_header(mdev, attrs->reqid, XFRM_DEV_OFFLOAD_OUT,
-				  &flow_act);
-	if (err)
-		goto err_mod_header;
+	if (attrs->reqid) {
+		err = setup_modify_header(mdev, attrs->reqid,
+					  XFRM_DEV_OFFLOAD_OUT, &flow_act);
+		if (err)
+			goto err_mod_header;
+	}
 
 	switch (attrs->action) {
 	case XFRM_POLICY_ALLOW:
@@ -1028,7 +1047,8 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	return 0;
 
 err_action:
-	mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr);
+	if (attrs->reqid)
+		mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr);
 err_mod_header:
 	kvfree(spec);
 err_alloc:
@@ -1263,7 +1283,9 @@ void mlx5e_accel_ipsec_fs_del_pol(struct mlx5e_ipsec_pol_entry *pol_entry)
 		return;
 	}
 
-	mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
+	if (ipsec_rule->modify_hdr)
+		mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr);
+
 	tx_ft_put_policy(pol_entry->ipsec, pol_entry->attrs.prio);
 }
 
-- 
cgit v1.2.3-70-g09d2


From aa8bd0c9518cc5902d88510bf007e451699221f2 Mon Sep 17 00:00:00 2001
From: Raed Salem <raeds@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:42 +0200
Subject: net/mlx5e: Support IPsec acquire default SA

During XFRM stack acquire flow, a default SA is created to be updated
later, once acquire netlink message is handled in user space.

This SA is also passed to IPsec offload supporting driver, however this
SA acts only as placeholder and does not have context suitable for
offloading in HW yet. Identify this kind of SA by special offload flag
(XFRM_DEV_OFFLOAD_FLAG_ACQ), and create a SW only context.

In such cases with special mark so it won't be installed in HW in addition
flow and on remove/delete free this SW only context.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/8f36d6b61631dcd73fef0a0ac623456030bc9db0.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 26 +++++++++++++++++-----
 1 file changed, 21 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 9cc59dc8b592..20a6bd1c03a3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -308,6 +308,7 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 	struct net_device *netdev = x->xso.real_dev;
 	struct mlx5e_ipsec *ipsec;
 	struct mlx5e_priv *priv;
+	gfp_t gfp;
 	int err;
 
 	priv = netdev_priv(netdev);
@@ -315,16 +316,20 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 		return -EOPNOTSUPP;
 
 	ipsec = priv->ipsec;
-	err = mlx5e_xfrm_validate_state(priv->mdev, x, extack);
-	if (err)
-		return err;
-
-	sa_entry = kzalloc(sizeof(*sa_entry), GFP_KERNEL);
+	gfp = (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ) ? GFP_ATOMIC : GFP_KERNEL;
+	sa_entry = kzalloc(sizeof(*sa_entry), gfp);
 	if (!sa_entry)
 		return -ENOMEM;
 
 	sa_entry->x = x;
 	sa_entry->ipsec = ipsec;
+	/* Check if this SA is originated from acquire flow temporary SA */
+	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
+		goto out;
+
+	err = mlx5e_xfrm_validate_state(priv->mdev, x, extack);
+	if (err)
+		goto err_xfrm;
 
 	/* check esn */
 	mlx5e_ipsec_update_esn_state(sa_entry);
@@ -353,6 +358,7 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 				mlx5e_ipsec_set_iv_esn : mlx5e_ipsec_set_iv;
 
 	INIT_WORK(&sa_entry->modify_work.work, _update_xfrm_state);
+out:
 	x->xso.offload_handle = (unsigned long)sa_entry;
 	return 0;
 
@@ -372,6 +378,9 @@ static void mlx5e_xfrm_del_state(struct xfrm_state *x)
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
 	struct mlx5e_ipsec_sa_entry *old;
 
+	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
+		return;
+
 	old = xa_erase_bh(&ipsec->sadb, sa_entry->ipsec_obj_id);
 	WARN_ON(old != sa_entry);
 }
@@ -380,9 +389,13 @@ static void mlx5e_xfrm_free_state(struct xfrm_state *x)
 {
 	struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x);
 
+	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
+		goto sa_entry_free;
+
 	cancel_work_sync(&sa_entry->modify_work.work);
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 	mlx5_ipsec_free_sa_ctx(sa_entry);
+sa_entry_free:
 	kfree(sa_entry);
 }
 
@@ -486,6 +499,9 @@ static void mlx5e_xfrm_update_curlft(struct xfrm_state *x)
 
 	lockdep_assert_held(&x->lock);
 
+	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
+		return;
+
 	if (sa_entry->attrs.soft_packet_limit == XFRM_INF)
 		/* Limits are not configured, as soft limit
 		 * must be lowever than hard limit.
-- 
cgit v1.2.3-70-g09d2


From d0c19a310e77d25e860d1543b5967fe6be6f5fe0 Mon Sep 17 00:00:00 2001
From: Raed Salem <raeds@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:43 +0200
Subject: net/mlx5e: Use one rule to count all IPsec Tx offloaded traffic

Currently one counter is shared between all IPsec Tx offloaded
rules to count the total amount of packets/bytes that was IPsec
Tx offloaded, replace this scheme by adding a new flow table (ft)
with one rule that counts all flows that passes through this
table (like Rx status ft), this ft is pointed by all IPsec Tx
offloaded rules. The above allows to have a counter per tx flow
rule in while keeping a separate global counter that store the
aggregation outcome of all these per flow counters.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/09b9119d1deb6e482fd2d17e1f5760d7c5be1e48.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 58 ++++++++++++++++++++--
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  |  2 +-
 2 files changed, 55 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 9f694a8e21fd..d1e4fd1e21d5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -41,6 +41,7 @@ struct mlx5e_ipsec_rx {
 struct mlx5e_ipsec_tx {
 	struct mlx5e_ipsec_ft ft;
 	struct mlx5e_ipsec_miss pol;
+	struct mlx5e_ipsec_rule status;
 	struct mlx5_flow_namespace *ns;
 	struct mlx5e_ipsec_fc *fc;
 	struct mlx5_fs_chains *chains;
@@ -455,6 +456,39 @@ static void rx_ft_put_policy(struct mlx5e_ipsec *ipsec, u32 family, u32 prio)
 	mutex_unlock(&rx->ft.mutex);
 }
 
+static int ipsec_counter_rule_tx(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx)
+{
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act flow_act = {};
+	struct mlx5_flow_handle *fte;
+	struct mlx5_flow_spec *spec;
+	int err;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	if (!spec)
+		return -ENOMEM;
+
+	/* create fte */
+	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW |
+			  MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest.counter_id = mlx5_fc_id(tx->fc->cnt);
+	fte = mlx5_add_flow_rules(tx->ft.status, spec, &flow_act, &dest, 1);
+	if (IS_ERR(fte)) {
+		err = PTR_ERR(fte);
+		mlx5_core_err(mdev, "Fail to add ipsec tx counter rule err=%d\n", err);
+		goto err_rule;
+	}
+
+	kvfree(spec);
+	tx->status.rule = fte;
+	return 0;
+
+err_rule:
+	kvfree(spec);
+	return err;
+}
+
 /* IPsec TX flow steering */
 static void tx_destroy(struct mlx5e_ipsec_tx *tx, struct mlx5_ipsec_fs *roce)
 {
@@ -468,6 +502,8 @@ static void tx_destroy(struct mlx5e_ipsec_tx *tx, struct mlx5_ipsec_fs *roce)
 	}
 
 	mlx5_destroy_flow_table(tx->ft.sa);
+	mlx5_del_flow_rules(tx->status.rule);
+	mlx5_destroy_flow_table(tx->ft.status);
 }
 
 static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
@@ -477,10 +513,20 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 	struct mlx5_flow_table *ft;
 	int err;
 
-	ft = ipsec_ft_create(tx->ns, 1, 0, 4);
+	ft = ipsec_ft_create(tx->ns, 2, 0, 1);
 	if (IS_ERR(ft))
 		return PTR_ERR(ft);
+	tx->ft.status = ft;
 
+	err = ipsec_counter_rule_tx(mdev, tx);
+	if (err)
+		goto err_status_rule;
+
+	ft = ipsec_ft_create(tx->ns, 1, 0, 4);
+	if (IS_ERR(ft)) {
+		err = PTR_ERR(ft);
+		goto err_sa_ft;
+	}
 	tx->ft.sa = ft;
 
 	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PRIO) {
@@ -525,6 +571,10 @@ err_roce:
 	}
 err_pol_ft:
 	mlx5_destroy_flow_table(tx->ft.sa);
+err_sa_ft:
+	mlx5_del_flow_rules(tx->status.rule);
+err_status_rule:
+	mlx5_destroy_flow_table(tx->ft.status);
 	return err;
 }
 
@@ -949,11 +999,11 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC;
 	flow_act.crypto.obj_id = sa_entry->ipsec_obj_id;
 	flow_act.flags |= FLOW_ACT_NO_APPEND;
-	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_ALLOW |
+	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
 			   MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT |
 			   MLX5_FLOW_CONTEXT_ACTION_COUNT;
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
-	dest.counter_id = mlx5_fc_id(tx->fc->cnt);
+	dest.ft = tx->ft.status;
+	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	rule = mlx5_add_flow_rules(tx->ft.sa, spec, &flow_act, &dest, 1);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 3ade166073fa..8e3da9d4fe1c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -137,7 +137,7 @@
 #define LAG_MIN_LEVEL (OFFLOADS_MIN_LEVEL + KERNEL_RX_MACSEC_MIN_LEVEL + 1)
 
 #define KERNEL_TX_IPSEC_NUM_PRIOS  1
-#define KERNEL_TX_IPSEC_NUM_LEVELS 2
+#define KERNEL_TX_IPSEC_NUM_LEVELS 3
 #define KERNEL_TX_IPSEC_MIN_LEVEL        (KERNEL_TX_IPSEC_NUM_LEVELS)
 
 #define KERNEL_TX_MACSEC_NUM_PRIOS  1
-- 
cgit v1.2.3-70-g09d2


From 5a6cddb89b51d99a7702e63829644a5860dd9c41 Mon Sep 17 00:00:00 2001
From: Raed Salem <raeds@nvidia.com>
Date: Tue, 14 Mar 2023 10:58:44 +0200
Subject: net/mlx5e: Update IPsec per SA packets/bytes count

Providing per SA packets/bytes statistics mandates creating unique
counter per SA flow for Rx/Tx, whenever offloaded SA statistics is
desired query the specific SA counter to provide the stack with the
needed data.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Link: https://lore.kernel.org/r/7d5ce20ac495f3054afb633128700e7b7eeeb3cd.1678714336.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 20 ++++------
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  4 +-
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 44 +++++++++++++++++-----
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    | 15 --------
 4 files changed, 42 insertions(+), 41 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 20a6bd1c03a3..91fa0a366316 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -495,24 +495,18 @@ static void mlx5e_xfrm_advance_esn_state(struct xfrm_state *x)
 static void mlx5e_xfrm_update_curlft(struct xfrm_state *x)
 {
 	struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x);
-	int err;
+	struct mlx5e_ipsec_rule *ipsec_rule = &sa_entry->ipsec_rule;
+	u64 packets, bytes, lastuse;
 
-	lockdep_assert_held(&x->lock);
+	lockdep_assert(lockdep_is_held(&x->lock) ||
+		       lockdep_is_held(&dev_net(x->xso.real_dev)->xfrm.xfrm_cfg_mutex));
 
 	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
 		return;
 
-	if (sa_entry->attrs.soft_packet_limit == XFRM_INF)
-		/* Limits are not configured, as soft limit
-		 * must be lowever than hard limit.
-		 */
-		return;
-
-	err = mlx5e_ipsec_aso_query(sa_entry, NULL);
-	if (err)
-		return;
-
-	mlx5e_ipsec_aso_update_curlft(sa_entry, &x->curlft.packets);
+	mlx5_fc_query_cached(ipsec_rule->fc, &bytes, &packets, &lastuse);
+	x->curlft.packets += packets;
+	x->curlft.bytes += bytes;
 }
 
 static int mlx5e_xfrm_validate_policy(struct mlx5_core_dev *mdev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index f3e81c3383e5..68ae5230eb75 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -162,6 +162,7 @@ struct mlx5e_ipsec_rule {
 	struct mlx5_flow_handle *rule;
 	struct mlx5_modify_hdr *modify_hdr;
 	struct mlx5_pkt_reformat *pkt_reformat;
+	struct mlx5_fc *fc;
 };
 
 struct mlx5e_ipsec_modify_state_work {
@@ -235,9 +236,6 @@ void mlx5e_ipsec_aso_cleanup(struct mlx5e_ipsec *ipsec);
 
 int mlx5e_ipsec_aso_query(struct mlx5e_ipsec_sa_entry *sa_entry,
 			  struct mlx5_wqe_aso_ctrl_seg *data);
-void mlx5e_ipsec_aso_update_curlft(struct mlx5e_ipsec_sa_entry *sa_entry,
-				   u64 *packets);
-
 void mlx5e_accel_ipsec_fs_read_stats(struct mlx5e_priv *priv,
 				     void *ipsec_stats);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index d1e4fd1e21d5..0539640a4d88 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -876,11 +876,12 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs;
 	struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry);
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
-	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_destination dest[2];
 	struct mlx5_flow_act flow_act = {};
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
 	struct mlx5e_ipsec_rx *rx;
+	struct mlx5_fc *counter;
 	int err;
 
 	rx = rx_ft_get(mdev, ipsec, attrs->family);
@@ -917,14 +918,22 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 		break;
 	}
 
+	counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(counter)) {
+		err = PTR_ERR(counter);
+		goto err_add_cnt;
+	}
 	flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC;
 	flow_act.crypto.obj_id = sa_entry->ipsec_obj_id;
 	flow_act.flags |= FLOW_ACT_NO_APPEND;
 	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-			   MLX5_FLOW_CONTEXT_ACTION_CRYPTO_DECRYPT;
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-	dest.ft = rx->ft.status;
-	rule = mlx5_add_flow_rules(rx->ft.sa, spec, &flow_act, &dest, 1);
+			   MLX5_FLOW_CONTEXT_ACTION_CRYPTO_DECRYPT |
+			   MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest[0].ft = rx->ft.status;
+	dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest[1].counter_id = mlx5_fc_id(counter);
+	rule = mlx5_add_flow_rules(rx->ft.sa, spec, &flow_act, dest, 2);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		mlx5_core_err(mdev, "fail to add RX ipsec rule err=%d\n", err);
@@ -934,10 +943,13 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 
 	sa_entry->ipsec_rule.rule = rule;
 	sa_entry->ipsec_rule.modify_hdr = flow_act.modify_hdr;
+	sa_entry->ipsec_rule.fc = counter;
 	sa_entry->ipsec_rule.pkt_reformat = flow_act.pkt_reformat;
 	return 0;
 
 err_add_flow:
+	mlx5_fc_destroy(mdev, counter);
+err_add_cnt:
 	if (flow_act.pkt_reformat)
 		mlx5_packet_reformat_dealloc(mdev, flow_act.pkt_reformat);
 err_pkt_reformat:
@@ -954,11 +966,12 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs;
 	struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry);
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
-	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_destination dest[2];
 	struct mlx5_flow_act flow_act = {};
 	struct mlx5_flow_handle *rule;
 	struct mlx5_flow_spec *spec;
 	struct mlx5e_ipsec_tx *tx;
+	struct mlx5_fc *counter;
 	int err;
 
 	tx = tx_ft_get(mdev, ipsec);
@@ -996,15 +1009,23 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 		break;
 	}
 
+	counter = mlx5_fc_create(mdev, true);
+	if (IS_ERR(counter)) {
+		err = PTR_ERR(counter);
+		goto err_add_cnt;
+	}
+
 	flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC;
 	flow_act.crypto.obj_id = sa_entry->ipsec_obj_id;
 	flow_act.flags |= FLOW_ACT_NO_APPEND;
 	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
 			   MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT |
 			   MLX5_FLOW_CONTEXT_ACTION_COUNT;
-	dest.ft = tx->ft.status;
-	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-	rule = mlx5_add_flow_rules(tx->ft.sa, spec, &flow_act, &dest, 1);
+	dest[0].ft = tx->ft.status;
+	dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+	dest[1].counter_id = mlx5_fc_id(counter);
+	rule = mlx5_add_flow_rules(tx->ft.sa, spec, &flow_act, dest, 2);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
 		mlx5_core_err(mdev, "fail to add TX ipsec rule err=%d\n", err);
@@ -1013,10 +1034,13 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 
 	kvfree(spec);
 	sa_entry->ipsec_rule.rule = rule;
+	sa_entry->ipsec_rule.fc = counter;
 	sa_entry->ipsec_rule.pkt_reformat = flow_act.pkt_reformat;
 	return 0;
 
 err_add_flow:
+	mlx5_fc_destroy(mdev, counter);
+err_add_cnt:
 	if (flow_act.pkt_reformat)
 		mlx5_packet_reformat_dealloc(mdev, flow_act.pkt_reformat);
 err_pkt_reformat:
@@ -1299,7 +1323,7 @@ void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry);
 
 	mlx5_del_flow_rules(ipsec_rule->rule);
-
+	mlx5_fc_destroy(mdev, ipsec_rule->fc);
 	if (ipsec_rule->pkt_reformat)
 		mlx5_packet_reformat_dealloc(mdev, ipsec_rule->pkt_reformat);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 67be8d36bb76..5342b0b07681 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -489,18 +489,3 @@ int mlx5e_ipsec_aso_query(struct mlx5e_ipsec_sa_entry *sa_entry,
 	spin_unlock_bh(&aso->lock);
 	return ret;
 }
-
-void mlx5e_ipsec_aso_update_curlft(struct mlx5e_ipsec_sa_entry *sa_entry,
-				   u64 *packets)
-{
-	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
-	struct mlx5e_ipsec_aso *aso = ipsec->aso;
-	u64 hard_cnt;
-
-	hard_cnt = MLX5_GET(ipsec_aso, aso->ctx, remove_flow_pkt_cnt);
-	/* HW decresases the limit till it reaches zero to fire an avent.
-	 * We need to fix the calculations, so the returned count is a total
-	 * number of passed packets and not how much left.
-	 */
-	*packets = sa_entry->attrs.hard_packet_limit - hard_cnt;
-}
-- 
cgit v1.2.3-70-g09d2


From a69e332b4ef9f697a4558bab9c442a02b3659fcb Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 18 Mar 2023 21:32:41 +0100
Subject: net: phy: smsc: export functions for use by meson-gxl PHY driver

The Amlogic Meson internal PHY's have the same register layout as
certain SMSC PHY's (also for non-c22-standard registers). This seems
to be more than just coincidence. Apparently they also need the same
workaround for EDPD mode (energy detect power down). Therefore let's
export SMSC PHY driver functionality for use by the meson-gxl PHY
driver.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Chris Healy <healych@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c  | 20 +++++++++++++-------
 include/linux/smscphy.h |  6 ++++++
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 721871184205..730964b856ab 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -54,7 +54,7 @@ static int smsc_phy_ack_interrupt(struct phy_device *phydev)
 	return rc < 0 ? rc : 0;
 }
 
-static int smsc_phy_config_intr(struct phy_device *phydev)
+int smsc_phy_config_intr(struct phy_device *phydev)
 {
 	int rc;
 
@@ -75,8 +75,9 @@ static int smsc_phy_config_intr(struct phy_device *phydev)
 
 	return rc < 0 ? rc : 0;
 }
+EXPORT_SYMBOL_GPL(smsc_phy_config_intr);
 
-static irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev)
+irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev)
 {
 	int irq_status;
 
@@ -95,18 +96,20 @@ static irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev)
 
 	return IRQ_HANDLED;
 }
+EXPORT_SYMBOL_GPL(smsc_phy_handle_interrupt);
 
-static int smsc_phy_config_init(struct phy_device *phydev)
+int smsc_phy_config_init(struct phy_device *phydev)
 {
 	struct smsc_phy_priv *priv = phydev->priv;
 
-	if (!priv->energy_enable || phydev->irq != PHY_POLL)
+	if (!priv || !priv->energy_enable || phydev->irq != PHY_POLL)
 		return 0;
 
 	/* Enable energy detect power down mode */
 	return phy_set_bits(phydev, MII_LAN83C185_CTRL_STATUS,
 			    MII_LAN83C185_EDPWRDOWN);
 }
+EXPORT_SYMBOL_GPL(smsc_phy_config_init);
 
 static int smsc_phy_reset(struct phy_device *phydev)
 {
@@ -186,7 +189,7 @@ static int lan95xx_config_aneg_ext(struct phy_device *phydev)
  * The workaround is only applicable to poll mode. Energy Detect Power-Down may
  * not be used in interrupt mode lest link change detection becomes unreliable.
  */
-static int lan87xx_read_status(struct phy_device *phydev)
+int lan87xx_read_status(struct phy_device *phydev)
 {
 	struct smsc_phy_priv *priv = phydev->priv;
 	int err;
@@ -195,7 +198,8 @@ static int lan87xx_read_status(struct phy_device *phydev)
 	if (err)
 		return err;
 
-	if (!phydev->link && priv->energy_enable && phydev->irq == PHY_POLL) {
+	if (!phydev->link && priv && priv->energy_enable &&
+	    phydev->irq == PHY_POLL) {
 		/* Disable EDPD to wake up PHY */
 		int rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
 		if (rc < 0)
@@ -229,6 +233,7 @@ static int lan87xx_read_status(struct phy_device *phydev)
 
 	return err;
 }
+EXPORT_SYMBOL_GPL(lan87xx_read_status);
 
 static int smsc_get_sset_count(struct phy_device *phydev)
 {
@@ -269,7 +274,7 @@ static void smsc_get_stats(struct phy_device *phydev,
 		data[i] = smsc_get_stat(phydev, i);
 }
 
-static int smsc_phy_probe(struct phy_device *phydev)
+int smsc_phy_probe(struct phy_device *phydev)
 {
 	struct device *dev = &phydev->mdio.dev;
 	struct smsc_phy_priv *priv;
@@ -294,6 +299,7 @@ static int smsc_phy_probe(struct phy_device *phydev)
 
 	return clk_set_rate(refclk, 50 * 1000 * 1000);
 }
+EXPORT_SYMBOL_GPL(smsc_phy_probe);
 
 static struct phy_driver smsc_phy_driver[] = {
 {
diff --git a/include/linux/smscphy.h b/include/linux/smscphy.h
index 1a136271ba6a..80f37c1dba58 100644
--- a/include/linux/smscphy.h
+++ b/include/linux/smscphy.h
@@ -28,4 +28,10 @@
 #define MII_LAN83C185_MODE_POWERDOWN 0xC0 /* Power Down mode */
 #define MII_LAN83C185_MODE_ALL       0xE0 /* All capable mode */
 
+int smsc_phy_config_intr(struct phy_device *phydev);
+irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev);
+int smsc_phy_config_init(struct phy_device *phydev);
+int lan87xx_read_status(struct phy_device *phydev);
+int smsc_phy_probe(struct phy_device *phydev);
+
 #endif /* __LINUX_SMSCPHY_H__ */
-- 
cgit v1.2.3-70-g09d2


From be66fcc16ce677eaf4b119861d468dc1881a13aa Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 18 Mar 2023 21:36:04 +0100
Subject: net: phy: meson-gxl: reuse functionality of the SMSC PHY driver

The Amlogic Meson internal PHY's have the same register layout as
certain SMSC PHY's (also for non-c22-standard registers). This seems
to be more than just coincidence. Apparently they also need the same
workaround for EDPD mode (energy detect power down). Therefore let's
reuse SMSC PHY driver functionality in the meson-gxl PHY driver.

Tested with a G12A internal PHY. I don't have GXL test hw,
therefore I replace only the callbacks that are identical in
the SMSC PHY driver.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig     |  1 +
 drivers/net/phy/meson-gxl.c | 77 +++++----------------------------------------
 2 files changed, 9 insertions(+), 69 deletions(-)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 54874555c921..6b9525def973 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -70,6 +70,7 @@ config AMD_PHY
 config MESON_GXL_PHY
 	tristate "Amlogic Meson GXL Internal PHY"
 	depends on ARCH_MESON || COMPILE_TEST
+	select SMSC_PHY
 	help
 	  Currently has a driver for the Amlogic Meson GXL Internal PHY
 
diff --git a/drivers/net/phy/meson-gxl.c b/drivers/net/phy/meson-gxl.c
index a6015cd03bff..3dea7c752057 100644
--- a/drivers/net/phy/meson-gxl.c
+++ b/drivers/net/phy/meson-gxl.c
@@ -13,6 +13,7 @@
 #include <linux/phy.h>
 #include <linux/netdevice.h>
 #include <linux/bitfield.h>
+#include <linux/smscphy.h>
 
 #define TSTCNTL		20
 #define  TSTCNTL_READ		BIT(15)
@@ -23,18 +24,6 @@
 #define  TSTCNTL_WRITE_ADDRESS	GENMASK(4, 0)
 #define TSTREAD1	21
 #define TSTWRITE	23
-#define INTSRC_FLAG	29
-#define  INTSRC_ANEG_PR		BIT(1)
-#define  INTSRC_PARALLEL_FAULT	BIT(2)
-#define  INTSRC_ANEG_LP_ACK	BIT(3)
-#define  INTSRC_LINK_DOWN	BIT(4)
-#define  INTSRC_REMOTE_FAULT	BIT(5)
-#define  INTSRC_ANEG_COMPLETE	BIT(6)
-#define  INTSRC_ENERGY_DETECT	BIT(7)
-#define INTSRC_MASK	30
-
-#define INT_SOURCES (INTSRC_LINK_DOWN | INTSRC_ANEG_COMPLETE | \
-		     INTSRC_ENERGY_DETECT)
 
 #define BANK_ANALOG_DSP		0
 #define BANK_WOL		1
@@ -195,59 +184,6 @@ read_status_continue:
 	return genphy_read_status(phydev);
 }
 
-static int meson_gxl_ack_interrupt(struct phy_device *phydev)
-{
-	int ret = phy_read(phydev, INTSRC_FLAG);
-
-	return ret < 0 ? ret : 0;
-}
-
-static int meson_gxl_config_intr(struct phy_device *phydev)
-{
-	int ret;
-
-	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
-		/* Ack any pending IRQ */
-		ret = meson_gxl_ack_interrupt(phydev);
-		if (ret)
-			return ret;
-
-		ret = phy_write(phydev, INTSRC_MASK, INT_SOURCES);
-	} else {
-		ret = phy_write(phydev, INTSRC_MASK, 0);
-
-		/* Ack any pending IRQ */
-		ret = meson_gxl_ack_interrupt(phydev);
-	}
-
-	return ret;
-}
-
-static irqreturn_t meson_gxl_handle_interrupt(struct phy_device *phydev)
-{
-	int irq_status;
-
-	irq_status = phy_read(phydev, INTSRC_FLAG);
-	if (irq_status < 0) {
-		phy_error(phydev);
-		return IRQ_NONE;
-	}
-
-	irq_status &= INT_SOURCES;
-
-	if (irq_status == 0)
-		return IRQ_NONE;
-
-	/* Aneg-complete interrupt is used for link-up detection */
-	if (phydev->autoneg == AUTONEG_ENABLE &&
-	    irq_status == INTSRC_ENERGY_DETECT)
-		return IRQ_HANDLED;
-
-	phy_trigger_machine(phydev);
-
-	return IRQ_HANDLED;
-}
-
 static struct phy_driver meson_gxl_phy[] = {
 	{
 		PHY_ID_MATCH_EXACT(0x01814400),
@@ -257,8 +193,8 @@ static struct phy_driver meson_gxl_phy[] = {
 		.soft_reset     = genphy_soft_reset,
 		.config_init	= meson_gxl_config_init,
 		.read_status	= meson_gxl_read_status,
-		.config_intr	= meson_gxl_config_intr,
-		.handle_interrupt = meson_gxl_handle_interrupt,
+		.config_intr	= smsc_phy_config_intr,
+		.handle_interrupt = smsc_phy_handle_interrupt,
 		.suspend        = genphy_suspend,
 		.resume         = genphy_resume,
 		.read_mmd	= genphy_read_mmd_unsupported,
@@ -268,9 +204,12 @@ static struct phy_driver meson_gxl_phy[] = {
 		.name		= "Meson G12A Internal PHY",
 		/* PHY_BASIC_FEATURES */
 		.flags		= PHY_IS_INTERNAL,
+		.probe		= smsc_phy_probe,
+		.config_init	= smsc_phy_config_init,
 		.soft_reset     = genphy_soft_reset,
-		.config_intr	= meson_gxl_config_intr,
-		.handle_interrupt = meson_gxl_handle_interrupt,
+		.read_status	= lan87xx_read_status,
+		.config_intr	= smsc_phy_config_intr,
+		.handle_interrupt = smsc_phy_handle_interrupt,
 		.suspend        = genphy_suspend,
 		.resume         = genphy_resume,
 		.read_mmd	= genphy_read_mmd_unsupported,
-- 
cgit v1.2.3-70-g09d2


From 5fc3f6c90cca19e4b13433621d9c2dcae875f4d7 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sat, 18 Mar 2023 22:50:10 +0100
Subject: r8169: consolidate disabling ASPM before EPHY access

Now that rtl_hw_aspm_clkreq_enable() is a no-op for chip versions < 32,
we can consolidate disabling ASPM before EPHY access in rtl_hw_start().

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/realtek/r8169_main.c | 42 +++----------------------------
 1 file changed, 3 insertions(+), 39 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 6563e4c6a136..9f8357bbc8a4 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2962,8 +2962,6 @@ static void rtl_hw_start_8168e_2(struct rtl8169_private *tp)
 	RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) | PFM_EN);
 	RTL_W32(tp, MISC, RTL_R32(tp, MISC) | PWM_EN);
 	rtl_mod_config5(tp, Spi_en, 0);
-
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8168f(struct rtl8169_private *tp)
@@ -3054,11 +3052,7 @@ static void rtl_hw_start_8168g_1(struct rtl8169_private *tp)
 	};
 
 	rtl_hw_start_8168g(tp);
-
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8168g_1);
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8168g_2(struct rtl8169_private *tp)
@@ -3076,9 +3070,6 @@ static void rtl_hw_start_8168g_2(struct rtl8169_private *tp)
 	};
 
 	rtl_hw_start_8168g(tp);
-
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8168g_2);
 }
 
@@ -3099,8 +3090,6 @@ static void rtl_hw_start_8411_2(struct rtl8169_private *tp)
 
 	rtl_hw_start_8168g(tp);
 
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8411_2);
 
 	/* The following Realtek-provided magic fixes an issue with the RX unit
@@ -3238,8 +3227,6 @@ static void rtl_hw_start_8411_2(struct rtl8169_private *tp)
 	r8168_mac_ocp_write(tp, 0xFC32, 0x0C25);
 	r8168_mac_ocp_write(tp, 0xFC34, 0x00A9);
 	r8168_mac_ocp_write(tp, 0xFC36, 0x012D);
-
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
@@ -3254,8 +3241,6 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
 	};
 	int rg_saw_cnt;
 
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8168h_1);
 
 	rtl_set_fifo_size(tp, 0x08, 0x10, 0x02, 0x06);
@@ -3303,8 +3288,6 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
 	r8168_mac_ocp_write(tp, 0xe63e, 0x0000);
 	r8168_mac_ocp_write(tp, 0xc094, 0x0000);
 	r8168_mac_ocp_write(tp, 0xc09e, 0x0000);
-
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8168ep(struct rtl8169_private *tp)
@@ -3343,8 +3326,6 @@ static void rtl_hw_start_8168ep_3(struct rtl8169_private *tp)
 		{ 0x1e, 0x0000,	0x2000 },
 	};
 
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8168ep_3);
 
 	rtl_hw_start_8168ep(tp);
@@ -3355,8 +3336,6 @@ static void rtl_hw_start_8168ep_3(struct rtl8169_private *tp)
 	r8168_mac_ocp_modify(tp, 0xd3e2, 0x0fff, 0x0271);
 	r8168_mac_ocp_modify(tp, 0xd3e4, 0x00ff, 0x0000);
 	r8168_mac_ocp_modify(tp, 0xe860, 0x0000, 0x0080);
-
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8117(struct rtl8169_private *tp)
@@ -3368,9 +3347,6 @@ static void rtl_hw_start_8117(struct rtl8169_private *tp)
 	int rg_saw_cnt;
 
 	rtl8168ep_stop_cmac(tp);
-
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8117);
 
 	rtl_set_fifo_size(tp, 0x08, 0x10, 0x02, 0x06);
@@ -3420,8 +3396,6 @@ static void rtl_hw_start_8117(struct rtl8169_private *tp)
 
 	/* firmware is for MAC only */
 	r8169_apply_firmware(tp);
-
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8102e_1(struct rtl8169_private *tp)
@@ -3544,8 +3518,6 @@ static void rtl_hw_start_8402(struct rtl8169_private *tp)
 
 static void rtl_hw_start_8106(struct rtl8169_private *tp)
 {
-	rtl_hw_aspm_clkreq_enable(tp, false);
-
 	/* Force LAN exit from ASPM if Rx/Tx are not idle */
 	RTL_W32(tp, FuncEvent, RTL_R32(tp, FuncEvent) | 0x002800);
 
@@ -3562,7 +3534,6 @@ static void rtl_hw_start_8106(struct rtl8169_private *tp)
 	rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000);
 
 	rtl_pcie_state_l2l3_disable(tp);
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 DECLARE_RTL_COND(rtl_mac_ocp_e00e_cond)
@@ -3650,13 +3621,8 @@ static void rtl_hw_start_8125a_2(struct rtl8169_private *tp)
 	};
 
 	rtl_set_def_aspm_entry_latency(tp);
-
-	/* disable aspm and clock request before access ephy */
-	rtl_hw_aspm_clkreq_enable(tp, false);
 	rtl_ephy_init(tp, e_info_8125a_2);
-
 	rtl_hw_start_8125_common(tp);
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_start_8125b(struct rtl8169_private *tp)
@@ -3671,12 +3637,8 @@ static void rtl_hw_start_8125b(struct rtl8169_private *tp)
 	};
 
 	rtl_set_def_aspm_entry_latency(tp);
-	rtl_hw_aspm_clkreq_enable(tp, false);
-
 	rtl_ephy_init(tp, e_info_8125b);
 	rtl_hw_start_8125_common(tp);
-
-	rtl_hw_aspm_clkreq_enable(tp, true);
 }
 
 static void rtl_hw_config(struct rtl8169_private *tp)
@@ -3772,7 +3734,8 @@ static void rtl_hw_start_8169(struct rtl8169_private *tp)
 static void rtl_hw_start(struct  rtl8169_private *tp)
 {
 	rtl_unlock_config_regs(tp);
-
+	/* disable aspm and clock request before ephy access */
+	rtl_hw_aspm_clkreq_enable(tp, false);
 	RTL_W16(tp, CPlusCmd, tp->cp_cmd);
 
 	if (tp->mac_version <= RTL_GIGA_MAC_VER_06)
@@ -3783,6 +3746,7 @@ static void rtl_hw_start(struct  rtl8169_private *tp)
 		rtl_hw_start_8168(tp);
 
 	rtl_enable_exit_l1(tp);
+	rtl_hw_aspm_clkreq_enable(tp, true);
 	rtl_set_rx_max_size(tp);
 	rtl_set_rx_tx_desc_registers(tp);
 	rtl_lock_config_regs(tp);
-- 
cgit v1.2.3-70-g09d2


From 56aecc0a655d997342ca85a3dd1f59715f2b3349 Mon Sep 17 00:00:00 2001
From: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Date: Mon, 20 Mar 2023 08:32:01 +0100
Subject: MAINTAINERS: remove file entry in NFC SUBSYSTEM after platform_data
 movement

Commit 053fdaa841bd ("nfc: mrvl: Move platform_data struct into driver")
moves the nfcmrvl.h header file from include/linux/platform_data to the
driver's directory, but misses to adjust MAINTAINERS.

Hence, ./scripts/get_maintainer.pl --self-test=patterns complains about a
broken reference.

Just remove the file entry in NFC SUBSYSTEM, as the new location of the
code is already covered by another pattern in that section.

Fixes: 053fdaa841bd ("nfc: mrvl: Move platform_data struct into driver")
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 1 -
 1 file changed, 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 9faef5784c03..30ca644d704f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -14662,7 +14662,6 @@ S:	Maintained
 B:	mailto:linux-nfc@lists.01.org
 F:	Documentation/devicetree/bindings/net/nfc/
 F:	drivers/nfc/
-F:	include/linux/platform_data/nfcmrvl.h
 F:	include/net/nfc/
 F:	include/uapi/linux/nfc.h
 F:	net/nfc/
-- 
cgit v1.2.3-70-g09d2


From a506d6ce1dd184051037dc9d26c3eb187c9fe625 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Sun, 19 Mar 2023 13:30:13 -0700
Subject: libbpf: Fix ld_imm64 copy logic for ksym in light skeleton.

Unlike normal libbpf the light skeleton 'loader' program is doing
btf_find_by_name_kind() call at run-time to find ksym in the kernel and
populate its {btf_id, btf_obj_fd} pair in ld_imm64 insn. To avoid doing the
search multiple times for the same ksym it remembers the first patched ld_imm64
insn and copies {btf_id, btf_obj_fd} from it into subsequent ld_imm64 insn.
Fix a bug in copying logic, since it may incorrectly clear BPF_PSEUDO_BTF_ID flag.

Also replace always true if (btf_obj_fd >= 0) check with unconditional JMP_JA
to clarify the code.

Fixes: d995816b77eb ("libbpf: Avoid reload of imm for weak, unresolved, repeating ksym")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230319203014.55866-1-alexei.starovoitov@gmail.com
---
 tools/lib/bpf/gen_loader.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/tools/lib/bpf/gen_loader.c b/tools/lib/bpf/gen_loader.c
index 23f5c46708f8..b74c82bb831e 100644
--- a/tools/lib/bpf/gen_loader.c
+++ b/tools/lib/bpf/gen_loader.c
@@ -804,11 +804,13 @@ static void emit_relo_ksym_btf(struct bpf_gen *gen, struct ksym_relo_desc *relo,
 		return;
 	/* try to copy from existing ldimm64 insn */
 	if (kdesc->ref > 1) {
-		move_blob2blob(gen, insn + offsetof(struct bpf_insn, imm), 4,
-			       kdesc->insn + offsetof(struct bpf_insn, imm));
 		move_blob2blob(gen, insn + sizeof(struct bpf_insn) + offsetof(struct bpf_insn, imm), 4,
 			       kdesc->insn + sizeof(struct bpf_insn) + offsetof(struct bpf_insn, imm));
-		/* jump over src_reg adjustment if imm is not 0, reuse BPF_REG_0 from move_blob2blob */
+		move_blob2blob(gen, insn + offsetof(struct bpf_insn, imm), 4,
+			       kdesc->insn + offsetof(struct bpf_insn, imm));
+		/* jump over src_reg adjustment if imm (btf_id) is not 0, reuse BPF_REG_0 from move_blob2blob
+		 * If btf_id is zero, clear BPF_PSEUDO_BTF_ID flag in src_reg of ld_imm64 insn
+		 */
 		emit(gen, BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3));
 		goto clear_src_reg;
 	}
@@ -831,7 +833,7 @@ static void emit_relo_ksym_btf(struct bpf_gen *gen, struct ksym_relo_desc *relo,
 	emit(gen, BPF_STX_MEM(BPF_W, BPF_REG_8, BPF_REG_7,
 			      sizeof(struct bpf_insn) + offsetof(struct bpf_insn, imm)));
 	/* skip src_reg adjustment */
-	emit(gen, BPF_JMP_IMM(BPF_JSGE, BPF_REG_7, 0, 3));
+	emit(gen, BPF_JMP_IMM(BPF_JA, 0, 0, 3));
 clear_src_reg:
 	/* clear bpf_object__relocate_data's src_reg assignment, otherwise we get a verifier failure */
 	reg_mask = src_reg_mask();
-- 
cgit v1.2.3-70-g09d2


From bb4a6a9237293346cf1b3b7bc4ff4dfc1977a103 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Sun, 19 Mar 2023 13:30:14 -0700
Subject: selftest/bpf: Add a test case for ld_imm64 copy logic.

Add a test case to exercise {btf_id, btf_obj_fd} copy logic between ld_imm64 insns.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230319203014.55866-2-alexei.starovoitov@gmail.com
---
 tools/testing/selftests/bpf/progs/test_ksyms_weak.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/test_ksyms_weak.c b/tools/testing/selftests/bpf/progs/test_ksyms_weak.c
index 5f8379aadb29..7003eef0c192 100644
--- a/tools/testing/selftests/bpf/progs/test_ksyms_weak.c
+++ b/tools/testing/selftests/bpf/progs/test_ksyms_weak.c
@@ -37,7 +37,7 @@ int pass_handler(const void *ctx)
 
 	/* tests existing symbols. */
 	rq = (struct rq *)bpf_per_cpu_ptr(&runqueues, 0);
-	if (rq)
+	if (rq && bpf_ksym_exists(&runqueues))
 		out__existing_typed = rq->cpu;
 	out__existing_typeless = (__u64)&bpf_prog_active;
 
-- 
cgit v1.2.3-70-g09d2


From 01dc26c980b0a92d9ba94c28652b2675968727b6 Mon Sep 17 00:00:00 2001
From: Liu Pan <patteliu@gmail.com>
Date: Mon, 20 Mar 2023 11:07:20 +0800
Subject: libbpf: Explicitly call write to append content to file

Write data to fd by calling "vdprintf", in most implementations
of the standard library, the data is finally written by the writev syscall.
But "uprobe_events/kprobe_events" does not allow segmented writes,
so switch the "append_to_file" function to explicit write() call.

Signed-off-by: Liu Pan <patteliu@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230320030720.650-1-patteliu@gmail.com
---
 tools/lib/bpf/libbpf.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 4c34fbd7b5be..149864ea88d1 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -9918,16 +9918,20 @@ static int append_to_file(const char *file, const char *fmt, ...)
 {
 	int fd, n, err = 0;
 	va_list ap;
+	char buf[1024];
+
+	va_start(ap, fmt);
+	n = vsnprintf(buf, sizeof(buf), fmt, ap);
+	va_end(ap);
+
+	if (n < 0 || n >= sizeof(buf))
+		return -EINVAL;
 
 	fd = open(file, O_WRONLY | O_APPEND | O_CLOEXEC, 0);
 	if (fd < 0)
 		return -errno;
 
-	va_start(ap, fmt);
-	n = vdprintf(fd, fmt, ap);
-	va_end(ap);
-
-	if (n < 0)
+	if (write(fd, buf, n) < 0)
 		err = -errno;
 
 	close(fd);
-- 
cgit v1.2.3-70-g09d2


From f5d43ddd334b7c32fcaed9ba46afbd85cb467f1f Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:56:28 +0000
Subject: net: ethernet: mtk_eth_soc: add support for MT7981 SoC

The MediaTek MT7981 SoC comes with two 1G/2.5G SGMII ports, just like
MT7986.

In addition MT7981 is equipped with a built-in 1000Base-T PHY which can
be used with GMAC1.

As many MT7981 boards make use of inverting SGMII signal polarity, add
new device-tree attribute 'mediatek,pn_swap' to support them.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_eth_path.c | 14 ++++++++++---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c  | 21 +++++++++++++++++++
 drivers/net/ethernet/mediatek/mtk_eth_soc.h  | 31 ++++++++++++++++++++++++++++
 drivers/net/ethernet/mediatek/mtk_sgmii.c    | 10 +++++++++
 4 files changed, 73 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_path.c b/drivers/net/ethernet/mediatek/mtk_eth_path.c
index 72648535a14d..317e447f4991 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_path.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_path.c
@@ -96,12 +96,20 @@ static int set_mux_gmac2_gmac0_to_gephy(struct mtk_eth *eth, int path)
 
 static int set_mux_u3_gmac2_to_qphy(struct mtk_eth *eth, int path)
 {
-	unsigned int val = 0;
+	unsigned int val = 0, mask = 0, reg = 0;
 	bool updated = true;
 
 	switch (path) {
 	case MTK_ETH_PATH_GMAC2_SGMII:
-		val = CO_QPHY_SEL;
+		if (MTK_HAS_CAPS(eth->soc->caps, MTK_U3_COPHY_V2)) {
+			reg = USB_PHY_SWITCH_REG;
+			val = SGMII_QPHY_SEL;
+			mask = QPHY_SEL_MASK;
+		} else {
+			reg = INFRA_MISC2;
+			val = CO_QPHY_SEL;
+			mask = val;
+		}
 		break;
 	default:
 		updated = false;
@@ -109,7 +117,7 @@ static int set_mux_u3_gmac2_to_qphy(struct mtk_eth *eth, int path)
 	}
 
 	if (updated)
-		regmap_update_bits(eth->infra, INFRA_MISC2, CO_QPHY_SEL, val);
+		regmap_update_bits(eth->infra, reg, mask, val);
 
 	dev_dbg(eth->dev, "path %s in %s updated = %d\n",
 		mtk_eth_path_name(path), __func__, updated);
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 52aa71f0c499..40490197e124 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -4846,6 +4846,26 @@ static const struct mtk_soc_data mt7629_data = {
 	},
 };
 
+static const struct mtk_soc_data mt7981_data = {
+	.reg_map = &mt7986_reg_map,
+	.ana_rgc3 = 0x128,
+	.caps = MT7981_CAPS,
+	.hw_features = MTK_HW_FEATURES,
+	.required_clks = MT7981_CLKS_BITMAP,
+	.required_pctl = false,
+	.offload_version = 2,
+	.hash_offset = 4,
+	.foe_entry_size = sizeof(struct mtk_foe_entry),
+	.txrx = {
+		.txd_size = sizeof(struct mtk_tx_dma_v2),
+		.rxd_size = sizeof(struct mtk_rx_dma_v2),
+		.rx_irq_done_mask = MTK_RX_DONE_INT_V2,
+		.rx_dma_l4_valid = RX_DMA_L4_VALID_V2,
+		.dma_max_len = MTK_TX_DMA_BUF_LEN_V2,
+		.dma_len_offset = 8,
+	},
+};
+
 static const struct mtk_soc_data mt7986_data = {
 	.reg_map = &mt7986_reg_map,
 	.ana_rgc3 = 0x128,
@@ -4888,6 +4908,7 @@ const struct of_device_id of_mtk_match[] = {
 	{ .compatible = "mediatek,mt7622-eth", .data = &mt7622_data},
 	{ .compatible = "mediatek,mt7623-eth", .data = &mt7623_data},
 	{ .compatible = "mediatek,mt7629-eth", .data = &mt7629_data},
+	{ .compatible = "mediatek,mt7981-eth", .data = &mt7981_data},
 	{ .compatible = "mediatek,mt7986-eth", .data = &mt7986_data},
 	{ .compatible = "ralink,rt5350-eth", .data = &rt5350_data},
 	{},
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index 084a6badef6d..adf01f894b02 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -556,11 +556,22 @@
 #define SGMSYS_QPHY_PWR_STATE_CTRL 0xe8
 #define	SGMII_PHYA_PWD		BIT(4)
 
+/* Register to QPHY wrapper control */
+#define SGMSYS_QPHY_WRAP_CTRL	0xec
+#define SGMII_PN_SWAP_MASK	GENMASK(1, 0)
+#define SGMII_PN_SWAP_TX_RX	(BIT(0) | BIT(1))
+#define MTK_SGMII_FLAG_PN_SWAP	BIT(0)
+
 /* Infrasys subsystem config registers */
 #define INFRA_MISC2            0x70c
 #define CO_QPHY_SEL            BIT(0)
 #define GEPHY_MAC_SEL          BIT(1)
 
+/* Top misc registers */
+#define USB_PHY_SWITCH_REG	0x218
+#define QPHY_SEL_MASK		GENMASK(1, 0)
+#define SGMII_QPHY_SEL		0x2
+
 /* MT7628/88 specific stuff */
 #define MT7628_PDMA_OFFSET	0x0800
 #define MT7628_SDM_OFFSET	0x0c00
@@ -741,6 +752,17 @@ enum mtk_clks_map {
 				 BIT(MTK_CLK_SGMII2_CDR_FB) | \
 				 BIT(MTK_CLK_SGMII_CK) | \
 				 BIT(MTK_CLK_ETH2PLL) | BIT(MTK_CLK_SGMIITOP))
+#define MT7981_CLKS_BITMAP	(BIT(MTK_CLK_FE) | BIT(MTK_CLK_GP2) | BIT(MTK_CLK_GP1) | \
+				 BIT(MTK_CLK_WOCPU0) | \
+				 BIT(MTK_CLK_SGMII_TX_250M) | \
+				 BIT(MTK_CLK_SGMII_RX_250M) | \
+				 BIT(MTK_CLK_SGMII_CDR_REF) | \
+				 BIT(MTK_CLK_SGMII_CDR_FB) | \
+				 BIT(MTK_CLK_SGMII2_TX_250M) | \
+				 BIT(MTK_CLK_SGMII2_RX_250M) | \
+				 BIT(MTK_CLK_SGMII2_CDR_REF) | \
+				 BIT(MTK_CLK_SGMII2_CDR_FB) | \
+				 BIT(MTK_CLK_SGMII_CK))
 #define MT7986_CLKS_BITMAP	(BIT(MTK_CLK_FE) | BIT(MTK_CLK_GP2) | BIT(MTK_CLK_GP1) | \
 				 BIT(MTK_CLK_WOCPU1) | BIT(MTK_CLK_WOCPU0) | \
 				 BIT(MTK_CLK_SGMII_TX_250M) | \
@@ -854,6 +876,7 @@ enum mkt_eth_capabilities {
 	MTK_NETSYS_V2_BIT,
 	MTK_SOC_MT7628_BIT,
 	MTK_RSTCTRL_PPE1_BIT,
+	MTK_U3_COPHY_V2_BIT,
 
 	/* MUX BITS*/
 	MTK_ETH_MUX_GDM1_TO_GMAC1_ESW_BIT,
@@ -888,6 +911,7 @@ enum mkt_eth_capabilities {
 #define MTK_NETSYS_V2		BIT(MTK_NETSYS_V2_BIT)
 #define MTK_SOC_MT7628		BIT(MTK_SOC_MT7628_BIT)
 #define MTK_RSTCTRL_PPE1	BIT(MTK_RSTCTRL_PPE1_BIT)
+#define MTK_U3_COPHY_V2		BIT(MTK_U3_COPHY_V2_BIT)
 
 #define MTK_ETH_MUX_GDM1_TO_GMAC1_ESW		\
 	BIT(MTK_ETH_MUX_GDM1_TO_GMAC1_ESW_BIT)
@@ -960,6 +984,11 @@ enum mkt_eth_capabilities {
 		      MTK_MUX_U3_GMAC2_TO_QPHY | \
 		      MTK_MUX_GMAC12_TO_GEPHY_SGMII | MTK_QDMA)
 
+#define MT7981_CAPS  (MTK_GMAC1_SGMII | MTK_GMAC2_SGMII | MTK_GMAC2_GEPHY | \
+		      MTK_MUX_GMAC12_TO_GEPHY_SGMII | MTK_QDMA | \
+		      MTK_MUX_U3_GMAC2_TO_QPHY | MTK_U3_COPHY_V2 | \
+		      MTK_NETSYS_V2 | MTK_RSTCTRL_PPE1)
+
 #define MT7986_CAPS  (MTK_GMAC1_SGMII | MTK_GMAC2_SGMII | \
 		      MTK_MUX_GMAC12_TO_GEPHY_SGMII | MTK_QDMA | \
 		      MTK_NETSYS_V2 | MTK_RSTCTRL_PPE1)
@@ -1073,12 +1102,14 @@ struct mtk_soc_data {
  * @ana_rgc3:          The offset refers to register ANA_RGC3 related to regmap
  * @interface:         Currently configured interface mode
  * @pcs:               Phylink PCS structure
+ * @flags:             Flags indicating hardware properties
  */
 struct mtk_pcs {
 	struct regmap	*regmap;
 	u32             ana_rgc3;
 	phy_interface_t	interface;
 	struct phylink_pcs pcs;
+	u32		flags;
 };
 
 /* struct mtk_sgmii -  This is the structure holding sgmii regmap and its
diff --git a/drivers/net/ethernet/mediatek/mtk_sgmii.c b/drivers/net/ethernet/mediatek/mtk_sgmii.c
index 83976dc86887..61bd9986466a 100644
--- a/drivers/net/ethernet/mediatek/mtk_sgmii.c
+++ b/drivers/net/ethernet/mediatek/mtk_sgmii.c
@@ -91,6 +91,11 @@ static int mtk_pcs_config(struct phylink_pcs *pcs, unsigned int mode,
 		regmap_update_bits(mpcs->regmap, SGMII_RESERVED_0,
 				   SGMII_SW_RESET, SGMII_SW_RESET);
 
+		if (mpcs->flags & MTK_SGMII_FLAG_PN_SWAP)
+			regmap_update_bits(mpcs->regmap, SGMSYS_QPHY_WRAP_CTRL,
+					   SGMII_PN_SWAP_MASK,
+					   SGMII_PN_SWAP_TX_RX);
+
 		if (interface == PHY_INTERFACE_MODE_2500BASEX)
 			rgc3 = RG_PHY_SPEED_3_125G;
 		else
@@ -186,6 +191,11 @@ int mtk_sgmii_init(struct mtk_sgmii *ss, struct device_node *r, u32 ana_rgc3)
 
 		ss->pcs[i].ana_rgc3 = ana_rgc3;
 		ss->pcs[i].regmap = syscon_node_to_regmap(np);
+
+		ss->pcs[i].flags = 0;
+		if (of_property_read_bool(np, "mediatek,pnswap"))
+			ss->pcs[i].flags |= MTK_SGMII_FLAG_PN_SWAP;
+
 		of_node_put(np);
 		if (IS_ERR(ss->pcs[i].regmap))
 			return PTR_ERR(ss->pcs[i].regmap);
-- 
cgit v1.2.3-70-g09d2


From e3ac1c270466e4fe930010ca688f71b98282e0e6 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:56:40 +0000
Subject: dt-bindings: net: mediatek,net: add mt7981-eth binding

Introduce DT bindings for the MT7981 SoC to mediatek,net.yaml.

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/mediatek,net.yaml      | 53 ++++++++++++++++++++--
 1 file changed, 48 insertions(+), 5 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/mediatek,net.yaml b/Documentation/devicetree/bindings/net/mediatek,net.yaml
index 7ef696204c5a..b7f6474dc5ab 100644
--- a/Documentation/devicetree/bindings/net/mediatek,net.yaml
+++ b/Documentation/devicetree/bindings/net/mediatek,net.yaml
@@ -21,6 +21,7 @@ properties:
       - mediatek,mt7623-eth
       - mediatek,mt7622-eth
       - mediatek,mt7629-eth
+      - mediatek,mt7981-eth
       - mediatek,mt7986-eth
       - ralink,rt5350-eth
 
@@ -78,6 +79,11 @@ properties:
     description:
       List of phandles to wireless ethernet dispatch nodes.
 
+  mediatek,wed-pcie:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description:
+      Phandle to the mediatek wed-pcie controller.
+
   dma-coherent: true
 
   mdio-bus:
@@ -123,6 +129,8 @@ allOf:
 
         mediatek,wed: false
 
+        mediatek,wed-pcie: false
+
   - if:
       properties:
         compatible:
@@ -160,6 +168,8 @@ allOf:
           description:
             Phandle to the mediatek pcie-mirror controller.
 
+        mediatek,wed-pcie: false
+
   - if:
       properties:
         compatible:
@@ -206,6 +216,44 @@ allOf:
 
         mediatek,wed: false
 
+        mediatek,wed-pcie: false
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: mediatek,mt7981-eth
+    then:
+      properties:
+        interrupts:
+          minItems: 4
+
+        clocks:
+          minItems: 15
+          maxItems: 15
+
+        clock-names:
+          items:
+            - const: fe
+            - const: gp2
+            - const: gp1
+            - const: wocpu0
+            - const: sgmii_ck
+            - const: sgmii_tx250m
+            - const: sgmii_rx250m
+            - const: sgmii_cdr_ref
+            - const: sgmii_cdr_fb
+            - const: sgmii2_tx250m
+            - const: sgmii2_rx250m
+            - const: sgmii2_cdr_ref
+            - const: sgmii2_cdr_fb
+            - const: netsys0
+            - const: netsys1
+
+        mediatek,sgmiisys:
+          minItems: 2
+          maxItems: 2
+
   - if:
       properties:
         compatible:
@@ -242,11 +290,6 @@ allOf:
           minItems: 2
           maxItems: 2
 
-        mediatek,wed-pcie:
-          $ref: /schemas/types.yaml#/definitions/phandle
-          description:
-            Phandle to the mediatek wed-pcie controller.
-
 patternProperties:
   "^mac@[0-1]$":
     type: object
-- 
cgit v1.2.3-70-g09d2


From d4f08a703565abf47baa5a77d05365cf4598d55c Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:56:52 +0000
Subject: dt-bindings: arm: mediatek: sgmiisys: Convert to DT schema

Convert mediatek,sgmiiisys bindings to DT schema format.
Add maintainer Matthias Brugger, no maintainers were listed in the
original documentation.
As this node is also referenced by the Ethernet controller and used
as SGMII PCS add this fact to the description.
Move the file to Documentation/devicetree/bindings/net/pcs/ which seems
more appropriate given that the great majority of registers are related
to SGMII PCS functionality and only one register represents clock bits.

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../bindings/arm/mediatek/mediatek,sgmiisys.txt    | 27 ------------
 .../bindings/net/pcs/mediatek,sgmiisys.yaml        | 49 ++++++++++++++++++++++
 2 files changed, 49 insertions(+), 27 deletions(-)
 delete mode 100644 Documentation/devicetree/bindings/arm/mediatek/mediatek,sgmiisys.txt
 create mode 100644 Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,sgmiisys.txt b/Documentation/devicetree/bindings/arm/mediatek/mediatek,sgmiisys.txt
deleted file mode 100644
index d2c24c277514..000000000000
--- a/Documentation/devicetree/bindings/arm/mediatek/mediatek,sgmiisys.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-MediaTek SGMIISYS controller
-============================
-
-The MediaTek SGMIISYS controller provides various clocks to the system.
-
-Required Properties:
-
-- compatible: Should be:
-	- "mediatek,mt7622-sgmiisys", "syscon"
-	- "mediatek,mt7629-sgmiisys", "syscon"
-	- "mediatek,mt7981-sgmiisys_0", "syscon"
-	- "mediatek,mt7981-sgmiisys_1", "syscon"
-	- "mediatek,mt7986-sgmiisys_0", "syscon"
-	- "mediatek,mt7986-sgmiisys_1", "syscon"
-- #clock-cells: Must be 1
-
-The SGMIISYS controller uses the common clk binding from
-Documentation/devicetree/bindings/clock/clock-bindings.txt
-The available clocks are defined in dt-bindings/clock/mt*-clk.h.
-
-Example:
-
-sgmiisys: sgmiisys@1b128000 {
-	compatible = "mediatek,mt7622-sgmiisys", "syscon";
-	reg = <0 0x1b128000 0 0x1000>;
-	#clock-cells = <1>;
-};
diff --git a/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml b/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml
new file mode 100644
index 000000000000..7ce597011a32
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml
@@ -0,0 +1,49 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/pcs/mediatek,sgmiisys.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: MediaTek SGMIISYS Controller
+
+maintainers:
+  - Matthias Brugger <matthias.bgg@gmail.com>
+
+description:
+  The MediaTek SGMIISYS controller provides a SGMII PCS and some clocks
+  to the ethernet subsystem to which it is attached.
+
+properties:
+  compatible:
+    items:
+      - enum:
+          - mediatek,mt7622-sgmiisys
+          - mediatek,mt7629-sgmiisys
+          - mediatek,mt7986-sgmiisys_0
+          - mediatek,mt7986-sgmiisys_1
+      - const: syscon
+
+  reg:
+    maxItems: 1
+
+  '#clock-cells':
+    const: 1
+
+required:
+  - compatible
+  - reg
+  - '#clock-cells'
+
+additionalProperties: false
+
+examples:
+  - |
+    soc {
+      #address-cells = <2>;
+      #size-cells = <2>;
+      sgmiisys: syscon@1b128000 {
+        compatible = "mediatek,mt7622-sgmiisys", "syscon";
+        reg = <0 0x1b128000 0 0x1000>;
+        #clock-cells = <1>;
+      };
+    };
-- 
cgit v1.2.3-70-g09d2


From 4f7eb19c4f44078100659f6ba073b0cc7191bc91 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:57:04 +0000
Subject: dt-bindings: net: pcs: mediatek,sgmiisys: add MT7981 SoC

Add mediatek,pnswap boolean property needed on many boards using the
MediaTek MT7981 SoC.

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml b/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml
index 7ce597011a32..66a95191bd77 100644
--- a/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml
+++ b/Documentation/devicetree/bindings/net/pcs/mediatek,sgmiisys.yaml
@@ -19,6 +19,8 @@ properties:
       - enum:
           - mediatek,mt7622-sgmiisys
           - mediatek,mt7629-sgmiisys
+          - mediatek,mt7981-sgmiisys_0
+          - mediatek,mt7981-sgmiisys_1
           - mediatek,mt7986-sgmiisys_0
           - mediatek,mt7986-sgmiisys_1
       - const: syscon
@@ -29,6 +31,10 @@ properties:
   '#clock-cells':
     const: 1
 
+  mediatek,pnswap:
+    description: Invert polarity of the SGMII data lanes
+    type: boolean
+
 required:
   - compatible
   - reg
-- 
cgit v1.2.3-70-g09d2


From c0a440031d4314d1023c1b87f43a4233634eebdb Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:57:15 +0000
Subject: net: ethernet: mtk_eth_soc: set MDIO bus clock frequency
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set MDIO bus clock frequency and allow setting a custom maximum
frequency from device tree.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 21 +++++++++++++++++++++
 drivers/net/ethernet/mediatek/mtk_eth_soc.h |  7 +++++++
 2 files changed, 28 insertions(+)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 40490197e124..6fe0e9e843cc 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -765,8 +765,10 @@ static const struct phylink_mac_ops mtk_phylink_ops = {
 
 static int mtk_mdio_init(struct mtk_eth *eth)
 {
+	unsigned int max_clk = 2500000, divider;
 	struct device_node *mii_np;
 	int ret;
+	u32 val;
 
 	mii_np = of_get_child_by_name(eth->dev->of_node, "mdio-bus");
 	if (!mii_np) {
@@ -794,6 +796,25 @@ static int mtk_mdio_init(struct mtk_eth *eth)
 	eth->mii_bus->parent = eth->dev;
 
 	snprintf(eth->mii_bus->id, MII_BUS_ID_SIZE, "%pOFn", mii_np);
+
+	if (!of_property_read_u32(mii_np, "clock-frequency", &val)) {
+		if (val > MDC_MAX_FREQ || val < MDC_MAX_FREQ / MDC_MAX_DIVIDER) {
+			dev_err(eth->dev, "MDIO clock frequency out of range");
+			ret = -EINVAL;
+			goto err_put_node;
+		}
+		max_clk = val;
+	}
+	divider = min_t(unsigned int, DIV_ROUND_UP(MDC_MAX_FREQ, max_clk), 63);
+
+	/* Configure MDC Divider */
+	val = mtk_r32(eth, MTK_PPSC);
+	val &= ~PPSC_MDC_CFG;
+	val |= FIELD_PREP(PPSC_MDC_CFG, divider) | PPSC_MDC_TURBO;
+	mtk_w32(eth, val, MTK_PPSC);
+
+	dev_dbg(eth->dev, "MDC is running on %d Hz\n", MDC_MAX_FREQ / divider);
+
 	ret = of_mdiobus_register(eth->mii_bus, mii_np);
 
 err_put_node:
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index adf01f894b02..529c95c481b7 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -363,6 +363,13 @@
 #define RX_DMA_VTAG_V2		BIT(0)
 #define RX_DMA_L4_VALID_V2	BIT(2)
 
+/* PHY Polling and SMI Master Control registers */
+#define MTK_PPSC		0x10000
+#define PPSC_MDC_CFG		GENMASK(29, 24)
+#define PPSC_MDC_TURBO		BIT(20)
+#define MDC_MAX_FREQ		25000000
+#define MDC_MAX_DIVIDER		63
+
 /* PHY Indirect Access Control registers */
 #define MTK_PHY_IAC		0x10004
 #define PHY_IAC_ACCESS		BIT(31)
-- 
cgit v1.2.3-70-g09d2


From 3fbe4d8c0e53e40faf2e0c7505536f9483af6d23 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:57:35 +0000
Subject: net: ethernet: mtk_eth_soc: ppe: add support for flow accounting
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The PPE units found in MT7622 and newer support packet and byte
accounting of hw-offloaded flows. Add support for reading those counters
as found in MediaTek's SDK[1].

[1]: https://git01.mediatek.com/plugins/gitiles/openwrt/feeds/mtk-openwrt-feeds/+/bc6a6a375c800dc2b80e1a325a2c732d1737df92
Tested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c     |   8 +-
 drivers/net/ethernet/mediatek/mtk_eth_soc.h     |   3 +
 drivers/net/ethernet/mediatek/mtk_ppe.c         | 114 +++++++++++++++++++++++-
 drivers/net/ethernet/mediatek/mtk_ppe.h         |  25 +++++-
 drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c |   9 +-
 drivers/net/ethernet/mediatek/mtk_ppe_offload.c |   8 ++
 drivers/net/ethernet/mediatek/mtk_ppe_regs.h    |  14 +++
 7 files changed, 172 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 6fe0e9e843cc..fc08c89b3e51 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -4697,8 +4697,8 @@ static int mtk_probe(struct platform_device *pdev)
 		for (i = 0; i < num_ppe; i++) {
 			u32 ppe_addr = eth->soc->reg_map->ppe_base + i * 0x400;
 
-			eth->ppe[i] = mtk_ppe_init(eth, eth->base + ppe_addr,
-						   eth->soc->offload_version, i);
+			eth->ppe[i] = mtk_ppe_init(eth, eth->base + ppe_addr, i);
+
 			if (!eth->ppe[i]) {
 				err = -ENOMEM;
 				goto err_deinit_ppe;
@@ -4820,6 +4820,7 @@ static const struct mtk_soc_data mt7622_data = {
 	.required_pctl = false,
 	.offload_version = 2,
 	.hash_offset = 2,
+	.has_accounting = true,
 	.foe_entry_size = sizeof(struct mtk_foe_entry) - 16,
 	.txrx = {
 		.txd_size = sizeof(struct mtk_tx_dma),
@@ -4857,6 +4858,7 @@ static const struct mtk_soc_data mt7629_data = {
 	.hw_features = MTK_HW_FEATURES,
 	.required_clks = MT7629_CLKS_BITMAP,
 	.required_pctl = false,
+	.has_accounting = true,
 	.txrx = {
 		.txd_size = sizeof(struct mtk_tx_dma),
 		.rxd_size = sizeof(struct mtk_rx_dma),
@@ -4877,6 +4879,7 @@ static const struct mtk_soc_data mt7981_data = {
 	.offload_version = 2,
 	.hash_offset = 4,
 	.foe_entry_size = sizeof(struct mtk_foe_entry),
+	.has_accounting = true,
 	.txrx = {
 		.txd_size = sizeof(struct mtk_tx_dma_v2),
 		.rxd_size = sizeof(struct mtk_rx_dma_v2),
@@ -4897,6 +4900,7 @@ static const struct mtk_soc_data mt7986_data = {
 	.offload_version = 2,
 	.hash_offset = 4,
 	.foe_entry_size = sizeof(struct mtk_foe_entry),
+	.has_accounting = true,
 	.txrx = {
 		.txd_size = sizeof(struct mtk_tx_dma_v2),
 		.rxd_size = sizeof(struct mtk_rx_dma_v2),
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index 529c95c481b7..666a7a448c66 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -1070,6 +1070,8 @@ struct mtk_reg_map {
  *				the extra setup for those pins used by GMAC.
  * @hash_offset			Flow table hash offset.
  * @foe_entry_size		Foe table entry size.
+ * @has_accounting		Bool indicating support for accounting of
+ *				offloaded flows.
  * @txd_size			Tx DMA descriptor size.
  * @rxd_size			Rx DMA descriptor size.
  * @rx_irq_done_mask		Rx irq done register mask.
@@ -1087,6 +1089,7 @@ struct mtk_soc_data {
 	u8		hash_offset;
 	u16		foe_entry_size;
 	netdev_features_t hw_features;
+	bool		has_accounting;
 	struct {
 		u32	txd_size;
 		u32	rxd_size;
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index 6883eb34cd8b..c099e8736716 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -74,6 +74,48 @@ static int mtk_ppe_wait_busy(struct mtk_ppe *ppe)
 	return ret;
 }
 
+static int mtk_ppe_mib_wait_busy(struct mtk_ppe *ppe)
+{
+	int ret;
+	u32 val;
+
+	ret = readl_poll_timeout(ppe->base + MTK_PPE_MIB_SER_CR, val,
+				 !(val & MTK_PPE_MIB_SER_CR_ST),
+				 20, MTK_PPE_WAIT_TIMEOUT_US);
+
+	if (ret)
+		dev_err(ppe->dev, "MIB table busy");
+
+	return ret;
+}
+
+static int mtk_mib_entry_read(struct mtk_ppe *ppe, u16 index, u64 *bytes, u64 *packets)
+{
+	u32 byte_cnt_low, byte_cnt_high, pkt_cnt_low, pkt_cnt_high;
+	u32 val, cnt_r0, cnt_r1, cnt_r2;
+	int ret;
+
+	val = FIELD_PREP(MTK_PPE_MIB_SER_CR_ADDR, index) | MTK_PPE_MIB_SER_CR_ST;
+	ppe_w32(ppe, MTK_PPE_MIB_SER_CR, val);
+
+	ret = mtk_ppe_mib_wait_busy(ppe);
+	if (ret)
+		return ret;
+
+	cnt_r0 = readl(ppe->base + MTK_PPE_MIB_SER_R0);
+	cnt_r1 = readl(ppe->base + MTK_PPE_MIB_SER_R1);
+	cnt_r2 = readl(ppe->base + MTK_PPE_MIB_SER_R2);
+
+	byte_cnt_low = FIELD_GET(MTK_PPE_MIB_SER_R0_BYTE_CNT_LOW, cnt_r0);
+	byte_cnt_high = FIELD_GET(MTK_PPE_MIB_SER_R1_BYTE_CNT_HIGH, cnt_r1);
+	pkt_cnt_low = FIELD_GET(MTK_PPE_MIB_SER_R1_PKT_CNT_LOW, cnt_r1);
+	pkt_cnt_high = FIELD_GET(MTK_PPE_MIB_SER_R2_PKT_CNT_HIGH, cnt_r2);
+	*bytes = ((u64)byte_cnt_high << 32) | byte_cnt_low;
+	*packets = (pkt_cnt_high << 16) | pkt_cnt_low;
+
+	return 0;
+}
+
 static void mtk_ppe_cache_clear(struct mtk_ppe *ppe)
 {
 	ppe_set(ppe, MTK_PPE_CACHE_CTL, MTK_PPE_CACHE_CTL_CLEAR);
@@ -458,6 +500,13 @@ __mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 		hwe->ib1 &= ~MTK_FOE_IB1_STATE;
 		hwe->ib1 |= FIELD_PREP(MTK_FOE_IB1_STATE, MTK_FOE_STATE_INVALID);
 		dma_wmb();
+		if (ppe->accounting) {
+			struct mtk_foe_accounting *acct;
+
+			acct = ppe->acct_table + entry->hash * sizeof(*acct);
+			acct->packets = 0;
+			acct->bytes = 0;
+		}
 	}
 	entry->hash = 0xffff;
 
@@ -565,6 +614,9 @@ __mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 	wmb();
 	hwe->ib1 = entry->ib1;
 
+	if (ppe->accounting)
+		*mtk_foe_entry_ib2(eth, hwe) |= MTK_FOE_IB2_MIB_CNT;
+
 	dma_wmb();
 
 	mtk_ppe_cache_clear(ppe);
@@ -756,11 +808,39 @@ int mtk_ppe_prepare_reset(struct mtk_ppe *ppe)
 	return mtk_ppe_wait_busy(ppe);
 }
 
-struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
-			     int version, int index)
+struct mtk_foe_accounting *mtk_foe_entry_get_mib(struct mtk_ppe *ppe, u32 index,
+						 struct mtk_foe_accounting *diff)
+{
+	struct mtk_foe_accounting *acct;
+	int size = sizeof(struct mtk_foe_accounting);
+	u64 bytes, packets;
+
+	if (!ppe->accounting)
+		return NULL;
+
+	if (mtk_mib_entry_read(ppe, index, &bytes, &packets))
+		return NULL;
+
+	acct = ppe->acct_table + index * size;
+
+	acct->bytes += bytes;
+	acct->packets += packets;
+
+	if (diff) {
+		diff->bytes = bytes;
+		diff->packets = packets;
+	}
+
+	return acct;
+}
+
+struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base, int index)
 {
+	bool accounting = eth->soc->has_accounting;
 	const struct mtk_soc_data *soc = eth->soc;
+	struct mtk_foe_accounting *acct;
 	struct device *dev = eth->dev;
+	struct mtk_mib_entry *mib;
 	struct mtk_ppe *ppe;
 	u32 foe_flow_size;
 	void *foe;
@@ -777,7 +857,8 @@ struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
 	ppe->base = base;
 	ppe->eth = eth;
 	ppe->dev = dev;
-	ppe->version = version;
+	ppe->version = eth->soc->offload_version;
+	ppe->accounting = accounting;
 
 	foe = dmam_alloc_coherent(ppe->dev,
 				  MTK_PPE_ENTRIES * soc->foe_entry_size,
@@ -793,6 +874,23 @@ struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
 	if (!ppe->foe_flow)
 		goto err_free_l2_flows;
 
+	if (accounting) {
+		mib = dmam_alloc_coherent(ppe->dev, MTK_PPE_ENTRIES * sizeof(*mib),
+					  &ppe->mib_phys, GFP_KERNEL);
+		if (!mib)
+			return NULL;
+
+		ppe->mib_table = mib;
+
+		acct = devm_kzalloc(dev, MTK_PPE_ENTRIES * sizeof(*acct),
+				    GFP_KERNEL);
+
+		if (!acct)
+			return NULL;
+
+		ppe->acct_table = acct;
+	}
+
 	mtk_ppe_debugfs_init(ppe, index);
 
 	return ppe;
@@ -922,6 +1020,16 @@ void mtk_ppe_start(struct mtk_ppe *ppe)
 		ppe_w32(ppe, MTK_PPE_DEFAULT_CPU_PORT1, 0xcb777);
 		ppe_w32(ppe, MTK_PPE_SBW_CTRL, 0x7f);
 	}
+
+	if (ppe->accounting && ppe->mib_phys) {
+		ppe_w32(ppe, MTK_PPE_MIB_TB_BASE, ppe->mib_phys);
+		ppe_m32(ppe, MTK_PPE_MIB_CFG, MTK_PPE_MIB_CFG_EN,
+			MTK_PPE_MIB_CFG_EN);
+		ppe_m32(ppe, MTK_PPE_MIB_CFG, MTK_PPE_MIB_CFG_RD_CLR,
+			MTK_PPE_MIB_CFG_RD_CLR);
+		ppe_m32(ppe, MTK_PPE_MIB_CACHE_CTL, MTK_PPE_MIB_CACHE_CTL_EN,
+			MTK_PPE_MIB_CFG_RD_CLR);
+	}
 }
 
 int mtk_ppe_stop(struct mtk_ppe *ppe)
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index 5e8bc48252b1..e1aab2e8e262 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -57,6 +57,7 @@ enum {
 #define MTK_FOE_IB2_MULTICAST		BIT(8)
 
 #define MTK_FOE_IB2_WDMA_QID2		GENMASK(13, 12)
+#define MTK_FOE_IB2_MIB_CNT		BIT(15)
 #define MTK_FOE_IB2_WDMA_DEVIDX		BIT(16)
 #define MTK_FOE_IB2_WDMA_WINFO		BIT(17)
 
@@ -285,16 +286,34 @@ struct mtk_flow_entry {
 	unsigned long cookie;
 };
 
+struct mtk_mib_entry {
+	u32	byt_cnt_l;
+	u16	byt_cnt_h;
+	u32	pkt_cnt_l;
+	u8	pkt_cnt_h;
+	u8	_rsv0;
+	u32	_rsv1;
+} __packed;
+
+struct mtk_foe_accounting {
+	u64	bytes;
+	u64	packets;
+};
+
 struct mtk_ppe {
 	struct mtk_eth *eth;
 	struct device *dev;
 	void __iomem *base;
 	int version;
 	char dirname[5];
+	bool accounting;
 
 	void *foe_table;
 	dma_addr_t foe_phys;
 
+	struct mtk_mib_entry *mib_table;
+	dma_addr_t mib_phys;
+
 	u16 foe_check_time[MTK_PPE_ENTRIES];
 	struct hlist_head *foe_flow;
 
@@ -303,8 +322,8 @@ struct mtk_ppe {
 	void *acct_table;
 };
 
-struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
-			     int version, int index);
+struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base, int index);
+
 void mtk_ppe_deinit(struct mtk_eth *eth);
 void mtk_ppe_start(struct mtk_ppe *ppe);
 int mtk_ppe_stop(struct mtk_ppe *ppe);
@@ -359,5 +378,7 @@ int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_flow_entry *entry);
 void mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry);
 int mtk_foe_entry_idle_time(struct mtk_ppe *ppe, struct mtk_flow_entry *entry);
 int mtk_ppe_debugfs_init(struct mtk_ppe *ppe, int index);
+struct mtk_foe_accounting *mtk_foe_entry_get_mib(struct mtk_ppe *ppe, u32 index,
+						 struct mtk_foe_accounting *diff);
 
 #endif
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c b/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
index 391b071bcff3..53cf87e9acbb 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
@@ -82,6 +82,7 @@ mtk_ppe_debugfs_foe_show(struct seq_file *m, void *private, bool bind)
 		struct mtk_foe_entry *entry = mtk_foe_get_entry(ppe, i);
 		struct mtk_foe_mac_info *l2;
 		struct mtk_flow_addr_info ai = {};
+		struct mtk_foe_accounting *acct;
 		unsigned char h_source[ETH_ALEN];
 		unsigned char h_dest[ETH_ALEN];
 		int type, state;
@@ -95,6 +96,8 @@ mtk_ppe_debugfs_foe_show(struct seq_file *m, void *private, bool bind)
 		if (bind && state != MTK_FOE_STATE_BIND)
 			continue;
 
+		acct = mtk_foe_entry_get_mib(ppe, i, NULL);
+
 		type = FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, entry->ib1);
 		seq_printf(m, "%05x %s %7s", i,
 			   mtk_foe_entry_state_str(state),
@@ -153,9 +156,11 @@ mtk_ppe_debugfs_foe_show(struct seq_file *m, void *private, bool bind)
 		*((__be16 *)&h_dest[4]) = htons(l2->dest_mac_lo);
 
 		seq_printf(m, " eth=%pM->%pM etype=%04x"
-			      " vlan=%d,%d ib1=%08x ib2=%08x\n",
+			      " vlan=%d,%d ib1=%08x ib2=%08x"
+			      " packets=%llu bytes=%llu\n",
 			   h_source, h_dest, ntohs(l2->etype),
-			   l2->vlan1, l2->vlan2, entry->ib1, ib2);
+			   l2->vlan1, l2->vlan2, entry->ib1, ib2,
+			   acct ? acct->packets : 0, acct ? acct->bytes : 0);
 	}
 
 	return 0;
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index 81afd5ee3fbf..f02ccffb1e79 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -497,6 +497,7 @@ static int
 mtk_flow_offload_stats(struct mtk_eth *eth, struct flow_cls_offload *f)
 {
 	struct mtk_flow_entry *entry;
+	struct mtk_foe_accounting diff;
 	u32 idle;
 
 	entry = rhashtable_lookup(&eth->flow_table, &f->cookie,
@@ -507,6 +508,13 @@ mtk_flow_offload_stats(struct mtk_eth *eth, struct flow_cls_offload *f)
 	idle = mtk_foe_entry_idle_time(eth->ppe[entry->ppe_index], entry);
 	f->stats.lastused = jiffies - idle * HZ;
 
+	if (entry->hash != 0xFFFF &&
+	    mtk_foe_entry_get_mib(eth->ppe[entry->ppe_index], entry->hash,
+				  &diff)) {
+		f->stats.pkts += diff.packets;
+		f->stats.bytes += diff.bytes;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_regs.h b/drivers/net/ethernet/mediatek/mtk_ppe_regs.h
index 0fdb983b0a88..a2e61b3eb006 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_regs.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_regs.h
@@ -149,6 +149,20 @@ enum {
 
 #define MTK_PPE_MIB_TB_BASE			0x338
 
+#define MTK_PPE_MIB_SER_CR			0x33C
+#define MTK_PPE_MIB_SER_CR_ST			BIT(16)
+#define MTK_PPE_MIB_SER_CR_ADDR			GENMASK(13, 0)
+
+#define MTK_PPE_MIB_SER_R0			0x340
+#define MTK_PPE_MIB_SER_R0_BYTE_CNT_LOW		GENMASK(31, 0)
+
+#define MTK_PPE_MIB_SER_R1			0x344
+#define MTK_PPE_MIB_SER_R1_PKT_CNT_LOW		GENMASK(31, 16)
+#define MTK_PPE_MIB_SER_R1_BYTE_CNT_HIGH	GENMASK(15, 0)
+
+#define MTK_PPE_MIB_SER_R2			0x348
+#define MTK_PPE_MIB_SER_R2_PKT_CNT_HIGH		GENMASK(23, 0)
+
 #define MTK_PPE_MIB_CACHE_CTL			0x350
 #define MTK_PPE_MIB_CACHE_CTL_EN		BIT(0)
 #define MTK_PPE_MIB_CACHE_CTL_FLUSH		BIT(2)
-- 
cgit v1.2.3-70-g09d2


From 4765a9722e09765866e131ec31f7b9cf4c1f4854 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:57:50 +0000
Subject: net: pcs: add driver for MediaTek SGMII PCS

The SGMII core found in several MediaTek SoCs is identical to what can
also be found in MediaTek's MT7531 Ethernet switch IC.
As this has not always been clear, both drivers developed different
implementations to deal with the PCS.
Recently Alexander Couzens pointed out this fact which lead to the
development of this shared driver.

Add a dedicated driver, mostly by copying the code now found in the
Ethernet driver. The now redundant code will be removed by a follow-up
commit.

Suggested-by: Alexander Couzens <lynxis@fe80.eu>
Suggested-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS                       |   8 +
 drivers/net/pcs/Kconfig           |   7 +
 drivers/net/pcs/Makefile          |   1 +
 drivers/net/pcs/pcs-mtk-lynxi.c   | 305 ++++++++++++++++++++++++++++++++++++++
 include/linux/pcs/pcs-mtk-lynxi.h |  13 ++
 5 files changed, 334 insertions(+)
 create mode 100644 drivers/net/pcs/pcs-mtk-lynxi.c
 create mode 100644 include/linux/pcs/pcs-mtk-lynxi.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 30ca644d704f..54a2a8122a97 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -13042,6 +13042,14 @@ L:	netdev@vger.kernel.org
 S:	Maintained
 F:	drivers/net/ethernet/mediatek/
 
+MEDIATEK ETHERNET PCS DRIVER
+M:	Alexander Couzens <lynxis@fe80.eu>
+M:	Daniel Golle <daniel@makrotopia.org>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/net/pcs/pcs-mtk-lynxi.c
+F:	include/linux/pcs/pcs-mtk-lynxi.h
+
 MEDIATEK I2C CONTROLLER DRIVER
 M:	Qii Wang <qii.wang@mediatek.com>
 L:	linux-i2c@vger.kernel.org
diff --git a/drivers/net/pcs/Kconfig b/drivers/net/pcs/Kconfig
index 6e7e6c346a3e..7c34fb7cbf7b 100644
--- a/drivers/net/pcs/Kconfig
+++ b/drivers/net/pcs/Kconfig
@@ -18,6 +18,13 @@ config PCS_LYNX
 	  This module provides helpers to phylink for managing the Lynx PCS
 	  which is part of the Layerscape and QorIQ Ethernet SERDES.
 
+config PCS_MTK_LYNXI
+	tristate
+	select REGMAP
+	help
+	  This module provides helpers to phylink for managing the LynxI PCS
+	  which is part of MediaTek's SoC and Ethernet switch ICs.
+
 config PCS_RZN1_MIIC
 	tristate "Renesas RZ/N1 MII converter"
 	depends on OF && (ARCH_RZN1 || COMPILE_TEST)
diff --git a/drivers/net/pcs/Makefile b/drivers/net/pcs/Makefile
index 4c780d8f2e98..9b9afd6b1c22 100644
--- a/drivers/net/pcs/Makefile
+++ b/drivers/net/pcs/Makefile
@@ -5,5 +5,6 @@ pcs_xpcs-$(CONFIG_PCS_XPCS)	:= pcs-xpcs.o pcs-xpcs-nxp.o
 
 obj-$(CONFIG_PCS_XPCS)		+= pcs_xpcs.o
 obj-$(CONFIG_PCS_LYNX)		+= pcs-lynx.o
+obj-$(CONFIG_PCS_MTK_LYNXI)	+= pcs-mtk-lynxi.o
 obj-$(CONFIG_PCS_RZN1_MIIC)	+= pcs-rzn1-miic.o
 obj-$(CONFIG_PCS_ALTERA_TSE)	+= pcs-altera-tse.o
diff --git a/drivers/net/pcs/pcs-mtk-lynxi.c b/drivers/net/pcs/pcs-mtk-lynxi.c
new file mode 100644
index 000000000000..888452325edc
--- /dev/null
+++ b/drivers/net/pcs/pcs-mtk-lynxi.c
@@ -0,0 +1,305 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018-2019 MediaTek Inc.
+/* A library for MediaTek SGMII circuit
+ *
+ * Author: Sean Wang <sean.wang@mediatek.com>
+ * Author: Alexander Couzens <lynxis@fe80.eu>
+ * Author: Daniel Golle <daniel@makrotopia.org>
+ *
+ */
+
+#include <linux/mdio.h>
+#include <linux/of.h>
+#include <linux/pcs/pcs-mtk-lynxi.h>
+#include <linux/phylink.h>
+#include <linux/regmap.h>
+
+/* SGMII subsystem config registers */
+/* BMCR (low 16) BMSR (high 16) */
+#define SGMSYS_PCS_CONTROL_1		0x0
+#define SGMII_BMCR			GENMASK(15, 0)
+#define SGMII_BMSR			GENMASK(31, 16)
+
+#define SGMSYS_PCS_DEVICE_ID		0x4
+#define SGMII_LYNXI_DEV_ID		0x4d544950
+
+#define SGMSYS_PCS_ADVERTISE		0x8
+#define SGMII_ADVERTISE			GENMASK(15, 0)
+#define SGMII_LPA			GENMASK(31, 16)
+
+#define SGMSYS_PCS_SCRATCH		0x14
+#define SGMII_DEV_VERSION		GENMASK(31, 16)
+
+/* Register to programmable link timer, the unit in 2 * 8ns */
+#define SGMSYS_PCS_LINK_TIMER		0x18
+#define SGMII_LINK_TIMER_MASK		GENMASK(19, 0)
+#define SGMII_LINK_TIMER_VAL(ns)	FIELD_PREP(SGMII_LINK_TIMER_MASK, \
+						   ((ns) / 2 / 8))
+
+/* Register to control remote fault */
+#define SGMSYS_SGMII_MODE		0x20
+#define SGMII_IF_MODE_SGMII		BIT(0)
+#define SGMII_SPEED_DUPLEX_AN		BIT(1)
+#define SGMII_SPEED_MASK		GENMASK(3, 2)
+#define SGMII_SPEED_10			FIELD_PREP(SGMII_SPEED_MASK, 0)
+#define SGMII_SPEED_100			FIELD_PREP(SGMII_SPEED_MASK, 1)
+#define SGMII_SPEED_1000		FIELD_PREP(SGMII_SPEED_MASK, 2)
+#define SGMII_DUPLEX_HALF		BIT(4)
+#define SGMII_REMOTE_FAULT_DIS		BIT(8)
+
+/* Register to reset SGMII design */
+#define SGMSYS_RESERVED_0		0x34
+#define SGMII_SW_RESET			BIT(0)
+
+/* Register to set SGMII speed, ANA RG_ Control Signals III */
+#define SGMII_PHY_SPEED_MASK		GENMASK(3, 2)
+#define SGMII_PHY_SPEED_1_25G		FIELD_PREP(SGMII_PHY_SPEED_MASK, 0)
+#define SGMII_PHY_SPEED_3_125G		FIELD_PREP(SGMII_PHY_SPEED_MASK, 1)
+
+/* Register to power up QPHY */
+#define SGMSYS_QPHY_PWR_STATE_CTRL	0xe8
+#define	SGMII_PHYA_PWD			BIT(4)
+
+/* Register to QPHY wrapper control */
+#define SGMSYS_QPHY_WRAP_CTRL		0xec
+#define SGMII_PN_SWAP_MASK		GENMASK(1, 0)
+#define SGMII_PN_SWAP_TX_RX		(BIT(0) | BIT(1))
+
+/* struct mtk_pcs_lynxi -  This structure holds each sgmii regmap andassociated
+ *                         data
+ * @regmap:                The register map pointing at the range used to setup
+ *                         SGMII modes
+ * @dev:                   Pointer to device owning the PCS
+ * @ana_rgc3:              The offset of register ANA_RGC3 relative to regmap
+ * @interface:             Currently configured interface mode
+ * @pcs:                   Phylink PCS structure
+ * @flags:                 Flags indicating hardware properties
+ */
+struct mtk_pcs_lynxi {
+	struct regmap		*regmap;
+	u32			ana_rgc3;
+	phy_interface_t		interface;
+	struct			phylink_pcs pcs;
+	u32			flags;
+};
+
+static struct mtk_pcs_lynxi *pcs_to_mtk_pcs_lynxi(struct phylink_pcs *pcs)
+{
+	return container_of(pcs, struct mtk_pcs_lynxi, pcs);
+}
+
+static void mtk_pcs_lynxi_get_state(struct phylink_pcs *pcs,
+				    struct phylink_link_state *state)
+{
+	struct mtk_pcs_lynxi *mpcs = pcs_to_mtk_pcs_lynxi(pcs);
+	unsigned int bm, adv;
+
+	/* Read the BMSR and LPA */
+	regmap_read(mpcs->regmap, SGMSYS_PCS_CONTROL_1, &bm);
+	regmap_read(mpcs->regmap, SGMSYS_PCS_ADVERTISE, &adv);
+
+	phylink_mii_c22_pcs_decode_state(state, FIELD_GET(SGMII_BMSR, bm),
+					 FIELD_GET(SGMII_LPA, adv));
+}
+
+static int mtk_pcs_lynxi_config(struct phylink_pcs *pcs, unsigned int mode,
+				phy_interface_t interface,
+				const unsigned long *advertising,
+				bool permit_pause_to_mac)
+{
+	struct mtk_pcs_lynxi *mpcs = pcs_to_mtk_pcs_lynxi(pcs);
+	bool mode_changed = false, changed, use_an;
+	unsigned int rgc3, sgm_mode, bmcr;
+	int advertise, link_timer;
+
+	advertise = phylink_mii_c22_pcs_encode_advertisement(interface,
+							     advertising);
+	if (advertise < 0)
+		return advertise;
+
+	/* Clearing IF_MODE_BIT0 switches the PCS to BASE-X mode, and
+	 * we assume that fixes it's speed at bitrate = line rate (in
+	 * other words, 1000Mbps or 2500Mbps).
+	 */
+	if (interface == PHY_INTERFACE_MODE_SGMII) {
+		sgm_mode = SGMII_IF_MODE_SGMII;
+		if (phylink_autoneg_inband(mode)) {
+			sgm_mode |= SGMII_REMOTE_FAULT_DIS |
+				    SGMII_SPEED_DUPLEX_AN;
+			use_an = true;
+		} else {
+			use_an = false;
+		}
+	} else if (phylink_autoneg_inband(mode)) {
+		/* 1000base-X or 2500base-X autoneg */
+		sgm_mode = SGMII_REMOTE_FAULT_DIS;
+		use_an = linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+					   advertising);
+	} else {
+		/* 1000base-X or 2500base-X without autoneg */
+		sgm_mode = 0;
+		use_an = false;
+	}
+
+	if (use_an)
+		bmcr = BMCR_ANENABLE;
+	else
+		bmcr = 0;
+
+	if (mpcs->interface != interface) {
+		link_timer = phylink_get_link_timer_ns(interface);
+		if (link_timer < 0)
+			return link_timer;
+
+		/* PHYA power down */
+		regmap_set_bits(mpcs->regmap, SGMSYS_QPHY_PWR_STATE_CTRL,
+				SGMII_PHYA_PWD);
+
+		/* Reset SGMII PCS state */
+		regmap_set_bits(mpcs->regmap, SGMSYS_RESERVED_0,
+				SGMII_SW_RESET);
+
+		if (mpcs->flags & MTK_SGMII_FLAG_PN_SWAP)
+			regmap_update_bits(mpcs->regmap, SGMSYS_QPHY_WRAP_CTRL,
+					   SGMII_PN_SWAP_MASK,
+					   SGMII_PN_SWAP_TX_RX);
+
+		if (interface == PHY_INTERFACE_MODE_2500BASEX)
+			rgc3 = SGMII_PHY_SPEED_3_125G;
+		else
+			rgc3 = SGMII_PHY_SPEED_1_25G;
+
+		/* Configure the underlying interface speed */
+		regmap_update_bits(mpcs->regmap, mpcs->ana_rgc3,
+				   SGMII_PHY_SPEED_MASK, rgc3);
+
+		/* Setup the link timer */
+		regmap_write(mpcs->regmap, SGMSYS_PCS_LINK_TIMER,
+			     SGMII_LINK_TIMER_VAL(link_timer));
+
+		mpcs->interface = interface;
+		mode_changed = true;
+	}
+
+	/* Update the advertisement, noting whether it has changed */
+	regmap_update_bits_check(mpcs->regmap, SGMSYS_PCS_ADVERTISE,
+				 SGMII_ADVERTISE, advertise, &changed);
+
+	/* Update the sgmsys mode register */
+	regmap_update_bits(mpcs->regmap, SGMSYS_SGMII_MODE,
+			   SGMII_REMOTE_FAULT_DIS | SGMII_SPEED_DUPLEX_AN |
+			   SGMII_IF_MODE_SGMII, sgm_mode);
+
+	/* Update the BMCR */
+	regmap_update_bits(mpcs->regmap, SGMSYS_PCS_CONTROL_1,
+			   BMCR_ANENABLE, bmcr);
+
+	/* Release PHYA power down state
+	 * Only removing bit SGMII_PHYA_PWD isn't enough.
+	 * There are cases when the SGMII_PHYA_PWD register contains 0x9 which
+	 * prevents SGMII from working. The SGMII still shows link but no traffic
+	 * can flow. Writing 0x0 to the PHYA_PWD register fix the issue. 0x0 was
+	 * taken from a good working state of the SGMII interface.
+	 * Unknown how much the QPHY needs but it is racy without a sleep.
+	 * Tested on mt7622 & mt7986.
+	 */
+	usleep_range(50, 100);
+	regmap_write(mpcs->regmap, SGMSYS_QPHY_PWR_STATE_CTRL, 0);
+
+	return changed || mode_changed;
+}
+
+static void mtk_pcs_lynxi_restart_an(struct phylink_pcs *pcs)
+{
+	struct mtk_pcs_lynxi *mpcs = pcs_to_mtk_pcs_lynxi(pcs);
+
+	regmap_set_bits(mpcs->regmap, SGMSYS_PCS_CONTROL_1, BMCR_ANRESTART);
+}
+
+static void mtk_pcs_lynxi_link_up(struct phylink_pcs *pcs, unsigned int mode,
+				  phy_interface_t interface, int speed,
+				  int duplex)
+{
+	struct mtk_pcs_lynxi *mpcs = pcs_to_mtk_pcs_lynxi(pcs);
+	unsigned int sgm_mode;
+
+	if (!phylink_autoneg_inband(mode)) {
+		/* Force the speed and duplex setting */
+		if (speed == SPEED_10)
+			sgm_mode = SGMII_SPEED_10;
+		else if (speed == SPEED_100)
+			sgm_mode = SGMII_SPEED_100;
+		else
+			sgm_mode = SGMII_SPEED_1000;
+
+		if (duplex != DUPLEX_FULL)
+			sgm_mode |= SGMII_DUPLEX_HALF;
+
+		regmap_update_bits(mpcs->regmap, SGMSYS_SGMII_MODE,
+				   SGMII_DUPLEX_HALF | SGMII_SPEED_MASK,
+				   sgm_mode);
+	}
+}
+
+static const struct phylink_pcs_ops mtk_pcs_lynxi_ops = {
+	.pcs_get_state = mtk_pcs_lynxi_get_state,
+	.pcs_config = mtk_pcs_lynxi_config,
+	.pcs_an_restart = mtk_pcs_lynxi_restart_an,
+	.pcs_link_up = mtk_pcs_lynxi_link_up,
+};
+
+struct phylink_pcs *mtk_pcs_lynxi_create(struct device *dev,
+					 struct regmap *regmap, u32 ana_rgc3,
+					 u32 flags)
+{
+	struct mtk_pcs_lynxi *mpcs;
+	u32 id, ver;
+	int ret;
+
+	ret = regmap_read(regmap, SGMSYS_PCS_DEVICE_ID, &id);
+	if (ret < 0)
+		return NULL;
+
+	if (id != SGMII_LYNXI_DEV_ID) {
+		dev_err(dev, "unknown PCS device id %08x\n", id);
+		return NULL;
+	}
+
+	ret = regmap_read(regmap, SGMSYS_PCS_SCRATCH, &ver);
+	if (ret < 0)
+		return NULL;
+
+	ver = FIELD_GET(SGMII_DEV_VERSION, ver);
+	if (ver != 0x1) {
+		dev_err(dev, "unknown PCS device version %04x\n", ver);
+		return NULL;
+	}
+
+	dev_dbg(dev, "MediaTek LynxI SGMII PCS (id 0x%08x, ver 0x%04x)\n", id,
+		ver);
+
+	mpcs = kzalloc(sizeof(*mpcs), GFP_KERNEL);
+	if (!mpcs)
+		return NULL;
+
+	mpcs->ana_rgc3 = ana_rgc3;
+	mpcs->regmap = regmap;
+	mpcs->flags = flags;
+	mpcs->pcs.ops = &mtk_pcs_lynxi_ops;
+	mpcs->pcs.poll = true;
+	mpcs->interface = PHY_INTERFACE_MODE_NA;
+
+	return &mpcs->pcs;
+}
+EXPORT_SYMBOL(mtk_pcs_lynxi_create);
+
+void mtk_pcs_lynxi_destroy(struct phylink_pcs *pcs)
+{
+	if (!pcs)
+		return;
+
+	kfree(pcs_to_mtk_pcs_lynxi(pcs));
+}
+EXPORT_SYMBOL(mtk_pcs_lynxi_destroy);
+
+MODULE_LICENSE("GPL");
diff --git a/include/linux/pcs/pcs-mtk-lynxi.h b/include/linux/pcs/pcs-mtk-lynxi.h
new file mode 100644
index 000000000000..be3b4ab32f4a
--- /dev/null
+++ b/include/linux/pcs/pcs-mtk-lynxi.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __LINUX_PCS_MTK_LYNXI_H
+#define __LINUX_PCS_MTK_LYNXI_H
+
+#include <linux/phylink.h>
+#include <linux/regmap.h>
+
+#define MTK_SGMII_FLAG_PN_SWAP BIT(0)
+struct phylink_pcs *mtk_pcs_lynxi_create(struct device *dev,
+					 struct regmap *regmap,
+					 u32 ana_rgc3, u32 flags);
+void mtk_pcs_lynxi_destroy(struct phylink_pcs *pcs);
+#endif
-- 
cgit v1.2.3-70-g09d2


From 2a3ec7ae313310c1092e4256208cc04d1958e469 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:58:02 +0000
Subject: net: ethernet: mtk_eth_soc: switch to external PCS driver

Now that we got a PCS driver, use it and remove the now redundant
PCS code and it's header macros from the Ethernet driver.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/Kconfig       |   2 +
 drivers/net/ethernet/mediatek/Makefile      |   2 +-
 drivers/net/ethernet/mediatek/mtk_eth_soc.c |  61 ++++++--
 drivers/net/ethernet/mediatek/mtk_eth_soc.h |  93 +-----------
 drivers/net/ethernet/mediatek/mtk_sgmii.c   | 217 ----------------------------
 5 files changed, 56 insertions(+), 319 deletions(-)
 delete mode 100644 drivers/net/ethernet/mediatek/mtk_sgmii.c

diff --git a/drivers/net/ethernet/mediatek/Kconfig b/drivers/net/ethernet/mediatek/Kconfig
index 97374fb3ee79..da0db417ab69 100644
--- a/drivers/net/ethernet/mediatek/Kconfig
+++ b/drivers/net/ethernet/mediatek/Kconfig
@@ -19,6 +19,8 @@ config NET_MEDIATEK_SOC
 	select DIMLIB
 	select PAGE_POOL
 	select PAGE_POOL_STATS
+	select PCS_MTK_LYNXI
+	select REGMAP_MMIO
 	help
 	  This driver supports the gigabit ethernet MACs in the
 	  MediaTek SoC family.
diff --git a/drivers/net/ethernet/mediatek/Makefile b/drivers/net/ethernet/mediatek/Makefile
index 8e0c61c33ff8..03e008fbc859 100644
--- a/drivers/net/ethernet/mediatek/Makefile
+++ b/drivers/net/ethernet/mediatek/Makefile
@@ -4,7 +4,7 @@
 #
 
 obj-$(CONFIG_NET_MEDIATEK_SOC) += mtk_eth.o
-mtk_eth-y := mtk_eth_soc.o mtk_sgmii.o mtk_eth_path.o mtk_ppe.o mtk_ppe_debugfs.o mtk_ppe_offload.o
+mtk_eth-y := mtk_eth_soc.o mtk_eth_path.o mtk_ppe.o mtk_ppe_debugfs.o mtk_ppe_offload.o
 mtk_eth-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed.o mtk_wed_mcu.o mtk_wed_wo.o
 ifdef CONFIG_DEBUG_FS
 mtk_eth-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed_debugfs.o
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index fc08c89b3e51..4403d3bbfd4d 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -20,6 +20,7 @@
 #include <linux/interrupt.h>
 #include <linux/pinctrl/devinfo.h>
 #include <linux/phylink.h>
+#include <linux/pcs/pcs-mtk-lynxi.h>
 #include <linux/jhash.h>
 #include <linux/bitfield.h>
 #include <net/dsa.h>
@@ -437,7 +438,7 @@ static struct phylink_pcs *mtk_mac_select_pcs(struct phylink_config *config,
 		sid = (MTK_HAS_CAPS(eth->soc->caps, MTK_SHARED_SGMII)) ?
 		       0 : mac->id;
 
-		return mtk_sgmii_select_pcs(eth->sgmii, sid);
+		return eth->sgmii_pcs[sid];
 	}
 
 	return NULL;
@@ -4053,8 +4054,17 @@ static int mtk_unreg_dev(struct mtk_eth *eth)
 	return 0;
 }
 
+static void mtk_sgmii_destroy(struct mtk_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < MTK_MAX_DEVS; i++)
+		mtk_pcs_lynxi_destroy(eth->sgmii_pcs[i]);
+}
+
 static int mtk_cleanup(struct mtk_eth *eth)
 {
+	mtk_sgmii_destroy(eth);
 	mtk_unreg_dev(eth);
 	mtk_free_dev(eth);
 	cancel_work_sync(&eth->pending_work);
@@ -4500,6 +4510,36 @@ void mtk_eth_set_dma_device(struct mtk_eth *eth, struct device *dma_dev)
 	rtnl_unlock();
 }
 
+static int mtk_sgmii_init(struct mtk_eth *eth)
+{
+	struct device_node *np;
+	struct regmap *regmap;
+	u32 flags;
+	int i;
+
+	for (i = 0; i < MTK_MAX_DEVS; i++) {
+		np = of_parse_phandle(eth->dev->of_node, "mediatek,sgmiisys", i);
+		if (!np)
+			break;
+
+		regmap = syscon_node_to_regmap(np);
+		flags = 0;
+		if (of_property_read_bool(np, "mediatek,pnswap"))
+			flags |= MTK_SGMII_FLAG_PN_SWAP;
+
+		of_node_put(np);
+
+		if (IS_ERR(regmap))
+			return PTR_ERR(regmap);
+
+		eth->sgmii_pcs[i] = mtk_pcs_lynxi_create(eth->dev, regmap,
+							 eth->soc->ana_rgc3,
+							 flags);
+	}
+
+	return 0;
+}
+
 static int mtk_probe(struct platform_device *pdev)
 {
 	struct resource *res = NULL;
@@ -4563,13 +4603,7 @@ static int mtk_probe(struct platform_device *pdev)
 	}
 
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_SGMII)) {
-		eth->sgmii = devm_kzalloc(eth->dev, sizeof(*eth->sgmii),
-					  GFP_KERNEL);
-		if (!eth->sgmii)
-			return -ENOMEM;
-
-		err = mtk_sgmii_init(eth->sgmii, pdev->dev.of_node,
-				     eth->soc->ana_rgc3);
+		err = mtk_sgmii_init(eth);
 
 		if (err)
 			return err;
@@ -4580,14 +4614,17 @@ static int mtk_probe(struct platform_device *pdev)
 							    "mediatek,pctl");
 		if (IS_ERR(eth->pctl)) {
 			dev_err(&pdev->dev, "no pctl regmap found\n");
-			return PTR_ERR(eth->pctl);
+			err = PTR_ERR(eth->pctl);
+			goto err_destroy_sgmii;
 		}
 	}
 
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_NETSYS_V2)) {
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-		if (!res)
-			return -EINVAL;
+		if (!res) {
+			err = -EINVAL;
+			goto err_destroy_sgmii;
+		}
 	}
 
 	if (eth->soc->offload_version) {
@@ -4746,6 +4783,8 @@ err_deinit_hw:
 	mtk_hw_deinit(eth);
 err_wed_exit:
 	mtk_wed_exit();
+err_destroy_sgmii:
+	mtk_sgmii_destroy(eth);
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index 666a7a448c66..23c7abeb5c14 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -510,65 +510,6 @@
 #define ETHSYS_DMA_AG_MAP_QDMA	BIT(1)
 #define ETHSYS_DMA_AG_MAP_PPE	BIT(2)
 
-/* SGMII subsystem config registers */
-/* BMCR (low 16) BMSR (high 16) */
-#define SGMSYS_PCS_CONTROL_1	0x0
-#define SGMII_BMCR		GENMASK(15, 0)
-#define SGMII_BMSR		GENMASK(31, 16)
-#define SGMII_AN_RESTART	BIT(9)
-#define SGMII_ISOLATE		BIT(10)
-#define SGMII_AN_ENABLE		BIT(12)
-#define SGMII_LINK_STATYS	BIT(18)
-#define SGMII_AN_ABILITY	BIT(19)
-#define SGMII_AN_COMPLETE	BIT(21)
-#define SGMII_PCS_FAULT		BIT(23)
-#define SGMII_AN_EXPANSION_CLR	BIT(30)
-
-#define SGMSYS_PCS_ADVERTISE	0x8
-#define SGMII_ADVERTISE		GENMASK(15, 0)
-#define SGMII_LPA		GENMASK(31, 16)
-
-/* Register to programmable link timer, the unit in 2 * 8ns */
-#define SGMSYS_PCS_LINK_TIMER	0x18
-#define SGMII_LINK_TIMER_MASK	GENMASK(19, 0)
-#define SGMII_LINK_TIMER_DEFAULT	(0x186a0 & SGMII_LINK_TIMER_MASK)
-
-/* Register to control remote fault */
-#define SGMSYS_SGMII_MODE		0x20
-#define SGMII_IF_MODE_SGMII		BIT(0)
-#define SGMII_SPEED_DUPLEX_AN		BIT(1)
-#define SGMII_SPEED_MASK		GENMASK(3, 2)
-#define SGMII_SPEED_10			FIELD_PREP(SGMII_SPEED_MASK, 0)
-#define SGMII_SPEED_100			FIELD_PREP(SGMII_SPEED_MASK, 1)
-#define SGMII_SPEED_1000		FIELD_PREP(SGMII_SPEED_MASK, 2)
-#define SGMII_DUPLEX_HALF		BIT(4)
-#define SGMII_IF_MODE_BIT5		BIT(5)
-#define SGMII_REMOTE_FAULT_DIS		BIT(8)
-#define SGMII_CODE_SYNC_SET_VAL		BIT(9)
-#define SGMII_CODE_SYNC_SET_EN		BIT(10)
-#define SGMII_SEND_AN_ERROR_EN		BIT(11)
-#define SGMII_IF_MODE_MASK		GENMASK(5, 1)
-
-/* Register to reset SGMII design */
-#define SGMII_RESERVED_0	0x34
-#define SGMII_SW_RESET		BIT(0)
-
-/* Register to set SGMII speed, ANA RG_ Control Signals III*/
-#define SGMSYS_ANA_RG_CS3	0x2028
-#define RG_PHY_SPEED_MASK	(BIT(2) | BIT(3))
-#define RG_PHY_SPEED_1_25G	0x0
-#define RG_PHY_SPEED_3_125G	BIT(2)
-
-/* Register to power up QPHY */
-#define SGMSYS_QPHY_PWR_STATE_CTRL 0xe8
-#define	SGMII_PHYA_PWD		BIT(4)
-
-/* Register to QPHY wrapper control */
-#define SGMSYS_QPHY_WRAP_CTRL	0xec
-#define SGMII_PN_SWAP_MASK	GENMASK(1, 0)
-#define SGMII_PN_SWAP_TX_RX	(BIT(0) | BIT(1))
-#define MTK_SGMII_FLAG_PN_SWAP	BIT(0)
-
 /* Infrasys subsystem config registers */
 #define INFRA_MISC2            0x70c
 #define CO_QPHY_SEL            BIT(0)
@@ -1105,31 +1046,6 @@ struct mtk_soc_data {
 /* currently no SoC has more than 2 macs */
 #define MTK_MAX_DEVS			2
 
-/* struct mtk_pcs -    This structure holds each sgmii regmap and associated
- *                     data
- * @regmap:            The register map pointing at the range used to setup
- *                     SGMII modes
- * @ana_rgc3:          The offset refers to register ANA_RGC3 related to regmap
- * @interface:         Currently configured interface mode
- * @pcs:               Phylink PCS structure
- * @flags:             Flags indicating hardware properties
- */
-struct mtk_pcs {
-	struct regmap	*regmap;
-	u32             ana_rgc3;
-	phy_interface_t	interface;
-	struct phylink_pcs pcs;
-	u32		flags;
-};
-
-/* struct mtk_sgmii -  This is the structure holding sgmii regmap and its
- *                     characteristics
- * @pcs                Array of individual PCS structures
- */
-struct mtk_sgmii {
-	struct mtk_pcs	pcs[MTK_MAX_DEVS];
-};
-
 /* struct mtk_eth -	This is the main datasructure for holding the state
  *			of the driver
  * @dev:		The device pointer
@@ -1149,6 +1065,7 @@ struct mtk_sgmii {
  *			MII modes
  * @infra:              The register map pointing at the range used to setup
  *                      SGMII and GePHY path
+ * @sgmii_pcs:		Pointers to mtk-pcs-lynxi phylink_pcs instances
  * @pctl:		The register map pointing at the range used to setup
  *			GMAC port drive/slew values
  * @dma_refcnt:		track how many netdevs are using the DMA engine
@@ -1189,8 +1106,8 @@ struct mtk_eth {
 	u32				msg_enable;
 	unsigned long			sysclk;
 	struct regmap			*ethsys;
-	struct regmap                   *infra;
-	struct mtk_sgmii                *sgmii;
+	struct regmap			*infra;
+	struct phylink_pcs		*sgmii_pcs[MTK_MAX_DEVS];
 	struct regmap			*pctl;
 	bool				hwlro;
 	refcount_t			dma_refcnt;
@@ -1352,10 +1269,6 @@ void mtk_stats_update_mac(struct mtk_mac *mac);
 void mtk_w32(struct mtk_eth *eth, u32 val, unsigned reg);
 u32 mtk_r32(struct mtk_eth *eth, unsigned reg);
 
-struct phylink_pcs *mtk_sgmii_select_pcs(struct mtk_sgmii *ss, int id);
-int mtk_sgmii_init(struct mtk_sgmii *ss, struct device_node *np,
-		   u32 ana_rgc3);
-
 int mtk_gmac_sgmii_path_setup(struct mtk_eth *eth, int mac_id);
 int mtk_gmac_gephy_path_setup(struct mtk_eth *eth, int mac_id);
 int mtk_gmac_rgmii_path_setup(struct mtk_eth *eth, int mac_id);
diff --git a/drivers/net/ethernet/mediatek/mtk_sgmii.c b/drivers/net/ethernet/mediatek/mtk_sgmii.c
deleted file mode 100644
index 61bd9986466a..000000000000
--- a/drivers/net/ethernet/mediatek/mtk_sgmii.c
+++ /dev/null
@@ -1,217 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-// Copyright (c) 2018-2019 MediaTek Inc.
-
-/* A library for MediaTek SGMII circuit
- *
- * Author: Sean Wang <sean.wang@mediatek.com>
- *
- */
-
-#include <linux/mfd/syscon.h>
-#include <linux/of.h>
-#include <linux/phylink.h>
-#include <linux/regmap.h>
-
-#include "mtk_eth_soc.h"
-
-static struct mtk_pcs *pcs_to_mtk_pcs(struct phylink_pcs *pcs)
-{
-	return container_of(pcs, struct mtk_pcs, pcs);
-}
-
-static void mtk_pcs_get_state(struct phylink_pcs *pcs,
-			      struct phylink_link_state *state)
-{
-	struct mtk_pcs *mpcs = pcs_to_mtk_pcs(pcs);
-	unsigned int bm, adv;
-
-	/* Read the BMSR and LPA */
-	regmap_read(mpcs->regmap, SGMSYS_PCS_CONTROL_1, &bm);
-	regmap_read(mpcs->regmap, SGMSYS_PCS_ADVERTISE, &adv);
-
-	phylink_mii_c22_pcs_decode_state(state, FIELD_GET(SGMII_BMSR, bm),
-					 FIELD_GET(SGMII_LPA, adv));
-}
-
-static int mtk_pcs_config(struct phylink_pcs *pcs, unsigned int mode,
-			  phy_interface_t interface,
-			  const unsigned long *advertising,
-			  bool permit_pause_to_mac)
-{
-	bool mode_changed = false, changed, use_an;
-	struct mtk_pcs *mpcs = pcs_to_mtk_pcs(pcs);
-	unsigned int rgc3, sgm_mode, bmcr;
-	int advertise, link_timer;
-
-	advertise = phylink_mii_c22_pcs_encode_advertisement(interface,
-							     advertising);
-	if (advertise < 0)
-		return advertise;
-
-	/* Clearing IF_MODE_BIT0 switches the PCS to BASE-X mode, and
-	 * we assume that fixes it's speed at bitrate = line rate (in
-	 * other words, 1000Mbps or 2500Mbps).
-	 */
-	if (interface == PHY_INTERFACE_MODE_SGMII) {
-		sgm_mode = SGMII_IF_MODE_SGMII;
-		if (phylink_autoneg_inband(mode)) {
-			sgm_mode |= SGMII_REMOTE_FAULT_DIS |
-				    SGMII_SPEED_DUPLEX_AN;
-			use_an = true;
-		} else {
-			use_an = false;
-		}
-	} else if (phylink_autoneg_inband(mode)) {
-		/* 1000base-X or 2500base-X autoneg */
-		sgm_mode = SGMII_REMOTE_FAULT_DIS;
-		use_an = linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
-					   advertising);
-	} else {
-		/* 1000base-X or 2500base-X without autoneg */
-		sgm_mode = 0;
-		use_an = false;
-	}
-
-	if (use_an) {
-		bmcr = SGMII_AN_ENABLE;
-	} else {
-		bmcr = 0;
-	}
-
-	if (mpcs->interface != interface) {
-		link_timer = phylink_get_link_timer_ns(interface);
-		if (link_timer < 0)
-			return link_timer;
-
-		/* PHYA power down */
-		regmap_update_bits(mpcs->regmap, SGMSYS_QPHY_PWR_STATE_CTRL,
-				   SGMII_PHYA_PWD, SGMII_PHYA_PWD);
-
-		/* Reset SGMII PCS state */
-		regmap_update_bits(mpcs->regmap, SGMII_RESERVED_0,
-				   SGMII_SW_RESET, SGMII_SW_RESET);
-
-		if (mpcs->flags & MTK_SGMII_FLAG_PN_SWAP)
-			regmap_update_bits(mpcs->regmap, SGMSYS_QPHY_WRAP_CTRL,
-					   SGMII_PN_SWAP_MASK,
-					   SGMII_PN_SWAP_TX_RX);
-
-		if (interface == PHY_INTERFACE_MODE_2500BASEX)
-			rgc3 = RG_PHY_SPEED_3_125G;
-		else
-			rgc3 = 0;
-
-		/* Configure the underlying interface speed */
-		regmap_update_bits(mpcs->regmap, mpcs->ana_rgc3,
-				   RG_PHY_SPEED_3_125G, rgc3);
-
-		/* Setup the link timer */
-		regmap_write(mpcs->regmap, SGMSYS_PCS_LINK_TIMER, link_timer / 2 / 8);
-
-		mpcs->interface = interface;
-		mode_changed = true;
-	}
-
-	/* Update the advertisement, noting whether it has changed */
-	regmap_update_bits_check(mpcs->regmap, SGMSYS_PCS_ADVERTISE,
-				 SGMII_ADVERTISE, advertise, &changed);
-
-	/* Update the sgmsys mode register */
-	regmap_update_bits(mpcs->regmap, SGMSYS_SGMII_MODE,
-			   SGMII_REMOTE_FAULT_DIS | SGMII_SPEED_DUPLEX_AN |
-			   SGMII_IF_MODE_SGMII, sgm_mode);
-
-	/* Update the BMCR */
-	regmap_update_bits(mpcs->regmap, SGMSYS_PCS_CONTROL_1,
-			   SGMII_AN_ENABLE, bmcr);
-
-	/* Release PHYA power down state
-	 * Only removing bit SGMII_PHYA_PWD isn't enough.
-	 * There are cases when the SGMII_PHYA_PWD register contains 0x9 which
-	 * prevents SGMII from working. The SGMII still shows link but no traffic
-	 * can flow. Writing 0x0 to the PHYA_PWD register fix the issue. 0x0 was
-	 * taken from a good working state of the SGMII interface.
-	 * Unknown how much the QPHY needs but it is racy without a sleep.
-	 * Tested on mt7622 & mt7986.
-	 */
-	usleep_range(50, 100);
-	regmap_write(mpcs->regmap, SGMSYS_QPHY_PWR_STATE_CTRL, 0);
-
-	return changed || mode_changed;
-}
-
-static void mtk_pcs_restart_an(struct phylink_pcs *pcs)
-{
-	struct mtk_pcs *mpcs = pcs_to_mtk_pcs(pcs);
-
-	regmap_update_bits(mpcs->regmap, SGMSYS_PCS_CONTROL_1,
-			   SGMII_AN_RESTART, SGMII_AN_RESTART);
-}
-
-static void mtk_pcs_link_up(struct phylink_pcs *pcs, unsigned int mode,
-			    phy_interface_t interface, int speed, int duplex)
-{
-	struct mtk_pcs *mpcs = pcs_to_mtk_pcs(pcs);
-	unsigned int sgm_mode;
-
-	if (!phylink_autoneg_inband(mode)) {
-		/* Force the speed and duplex setting */
-		if (speed == SPEED_10)
-			sgm_mode = SGMII_SPEED_10;
-		else if (speed == SPEED_100)
-			sgm_mode = SGMII_SPEED_100;
-		else
-			sgm_mode = SGMII_SPEED_1000;
-
-		if (duplex != DUPLEX_FULL)
-			sgm_mode |= SGMII_DUPLEX_HALF;
-
-		regmap_update_bits(mpcs->regmap, SGMSYS_SGMII_MODE,
-				   SGMII_DUPLEX_HALF | SGMII_SPEED_MASK,
-				   sgm_mode);
-	}
-}
-
-static const struct phylink_pcs_ops mtk_pcs_ops = {
-	.pcs_get_state = mtk_pcs_get_state,
-	.pcs_config = mtk_pcs_config,
-	.pcs_an_restart = mtk_pcs_restart_an,
-	.pcs_link_up = mtk_pcs_link_up,
-};
-
-int mtk_sgmii_init(struct mtk_sgmii *ss, struct device_node *r, u32 ana_rgc3)
-{
-	struct device_node *np;
-	int i;
-
-	for (i = 0; i < MTK_MAX_DEVS; i++) {
-		np = of_parse_phandle(r, "mediatek,sgmiisys", i);
-		if (!np)
-			break;
-
-		ss->pcs[i].ana_rgc3 = ana_rgc3;
-		ss->pcs[i].regmap = syscon_node_to_regmap(np);
-
-		ss->pcs[i].flags = 0;
-		if (of_property_read_bool(np, "mediatek,pnswap"))
-			ss->pcs[i].flags |= MTK_SGMII_FLAG_PN_SWAP;
-
-		of_node_put(np);
-		if (IS_ERR(ss->pcs[i].regmap))
-			return PTR_ERR(ss->pcs[i].regmap);
-
-		ss->pcs[i].pcs.ops = &mtk_pcs_ops;
-		ss->pcs[i].pcs.poll = true;
-		ss->pcs[i].interface = PHY_INTERFACE_MODE_NA;
-	}
-
-	return 0;
-}
-
-struct phylink_pcs *mtk_sgmii_select_pcs(struct mtk_sgmii *ss, int id)
-{
-	if (!ss->pcs[id].regmap)
-		return NULL;
-
-	return &ss->pcs[id].pcs;
-}
-- 
cgit v1.2.3-70-g09d2


From 5b89aeae6e00696ed9f15f18adbbe47dd1d4c0b7 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 19 Mar 2023 12:58:43 +0000
Subject: net: dsa: mt7530: use external PCS driver
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement regmap access wrappers, for now only to be used by the
pcs-mtk-lynxi driver.
Make use of this external PCS driver and drop the now reduntant
implementation in mt7530.c.
As a nice side effect the SGMII registers can now also more easily be
inspected for debugging via /sys/kernel/debug/regmap.

Tested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/Kconfig  |   1 +
 drivers/net/dsa/mt7530.c | 277 +++++++++++------------------------------------
 drivers/net/dsa/mt7530.h |  47 +-------
 3 files changed, 71 insertions(+), 254 deletions(-)

diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index f6f3b43dfb06..6b45fa8b6907 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -38,6 +38,7 @@ config NET_DSA_MT7530
 	tristate "MediaTek MT7530 and MT7531 Ethernet switch support"
 	select NET_DSA_TAG_MTK
 	select MEDIATEK_GE_PHY
+	select PCS_MTK_LYNXI
 	help
 	  This enables support for the MediaTek MT7530 and MT7531 Ethernet
 	  switch chips. Multi-chip module MT7530 in MT7621AT, MT7621DAT,
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index c2d81b7a429d..e28d35d8cf68 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -14,6 +14,7 @@
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
 #include <linux/of_platform.h>
+#include <linux/pcs/pcs-mtk-lynxi.h>
 #include <linux/phylink.h>
 #include <linux/regmap.h>
 #include <linux/regulator/consumer.h>
@@ -2572,128 +2573,11 @@ static int mt7531_rgmii_setup(struct mt7530_priv *priv, u32 port,
 	return 0;
 }
 
-static void mt7531_pcs_link_up(struct phylink_pcs *pcs, unsigned int mode,
-			       phy_interface_t interface, int speed, int duplex)
-{
-	struct mt7530_priv *priv = pcs_to_mt753x_pcs(pcs)->priv;
-	int port = pcs_to_mt753x_pcs(pcs)->port;
-	unsigned int val;
-
-	/* For adjusting speed and duplex of SGMII force mode. */
-	if (interface != PHY_INTERFACE_MODE_SGMII ||
-	    phylink_autoneg_inband(mode))
-		return;
-
-	/* SGMII force mode setting */
-	val = mt7530_read(priv, MT7531_SGMII_MODE(port));
-	val &= ~MT7531_SGMII_IF_MODE_MASK;
-
-	switch (speed) {
-	case SPEED_10:
-		val |= MT7531_SGMII_FORCE_SPEED_10;
-		break;
-	case SPEED_100:
-		val |= MT7531_SGMII_FORCE_SPEED_100;
-		break;
-	case SPEED_1000:
-		val |= MT7531_SGMII_FORCE_SPEED_1000;
-		break;
-	}
-
-	/* MT7531 SGMII 1G force mode can only work in full duplex mode,
-	 * no matter MT7531_SGMII_FORCE_HALF_DUPLEX is set or not.
-	 *
-	 * The speed check is unnecessary as the MAC capabilities apply
-	 * this restriction. --rmk
-	 */
-	if ((speed == SPEED_10 || speed == SPEED_100) &&
-	    duplex != DUPLEX_FULL)
-		val |= MT7531_SGMII_FORCE_HALF_DUPLEX;
-
-	mt7530_write(priv, MT7531_SGMII_MODE(port), val);
-}
-
 static bool mt753x_is_mac_port(u32 port)
 {
 	return (port == 5 || port == 6);
 }
 
-static int mt7531_sgmii_setup_mode_force(struct mt7530_priv *priv, u32 port,
-					 phy_interface_t interface)
-{
-	u32 val;
-
-	if (!mt753x_is_mac_port(port))
-		return -EINVAL;
-
-	mt7530_set(priv, MT7531_QPHY_PWR_STATE_CTRL(port),
-		   MT7531_SGMII_PHYA_PWD);
-
-	val = mt7530_read(priv, MT7531_PHYA_CTRL_SIGNAL3(port));
-	val &= ~MT7531_RG_TPHY_SPEED_MASK;
-	/* Setup 2.5 times faster clock for 2.5Gbps data speeds with 10B/8B
-	 * encoding.
-	 */
-	val |= (interface == PHY_INTERFACE_MODE_2500BASEX) ?
-		MT7531_RG_TPHY_SPEED_3_125G : MT7531_RG_TPHY_SPEED_1_25G;
-	mt7530_write(priv, MT7531_PHYA_CTRL_SIGNAL3(port), val);
-
-	mt7530_clear(priv, MT7531_PCS_CONTROL_1(port), MT7531_SGMII_AN_ENABLE);
-
-	/* MT7531 SGMII 1G and 2.5G force mode can only work in full duplex
-	 * mode, no matter MT7531_SGMII_FORCE_HALF_DUPLEX is set or not.
-	 */
-	mt7530_rmw(priv, MT7531_SGMII_MODE(port),
-		   MT7531_SGMII_IF_MODE_MASK | MT7531_SGMII_REMOTE_FAULT_DIS,
-		   MT7531_SGMII_FORCE_SPEED_1000);
-
-	mt7530_write(priv, MT7531_QPHY_PWR_STATE_CTRL(port), 0);
-
-	return 0;
-}
-
-static int mt7531_sgmii_setup_mode_an(struct mt7530_priv *priv, int port,
-				      phy_interface_t interface)
-{
-	if (!mt753x_is_mac_port(port))
-		return -EINVAL;
-
-	mt7530_set(priv, MT7531_QPHY_PWR_STATE_CTRL(port),
-		   MT7531_SGMII_PHYA_PWD);
-
-	mt7530_rmw(priv, MT7531_PHYA_CTRL_SIGNAL3(port),
-		   MT7531_RG_TPHY_SPEED_MASK, MT7531_RG_TPHY_SPEED_1_25G);
-
-	mt7530_set(priv, MT7531_SGMII_MODE(port),
-		   MT7531_SGMII_REMOTE_FAULT_DIS |
-		   MT7531_SGMII_SPEED_DUPLEX_AN);
-
-	mt7530_rmw(priv, MT7531_PCS_SPEED_ABILITY(port),
-		   MT7531_SGMII_TX_CONFIG_MASK, 1);
-
-	mt7530_set(priv, MT7531_PCS_CONTROL_1(port), MT7531_SGMII_AN_ENABLE);
-
-	mt7530_set(priv, MT7531_PCS_CONTROL_1(port), MT7531_SGMII_AN_RESTART);
-
-	mt7530_write(priv, MT7531_QPHY_PWR_STATE_CTRL(port), 0);
-
-	return 0;
-}
-
-static void mt7531_pcs_an_restart(struct phylink_pcs *pcs)
-{
-	struct mt7530_priv *priv = pcs_to_mt753x_pcs(pcs)->priv;
-	int port = pcs_to_mt753x_pcs(pcs)->port;
-	u32 val;
-
-	/* Only restart AN when AN is enabled */
-	val = mt7530_read(priv, MT7531_PCS_CONTROL_1(port));
-	if (val & MT7531_SGMII_AN_ENABLE) {
-		val |= MT7531_SGMII_AN_RESTART;
-		mt7530_write(priv, MT7531_PCS_CONTROL_1(port), val);
-	}
-}
-
 static int
 mt7531_mac_config(struct dsa_switch *ds, int port, unsigned int mode,
 		  phy_interface_t interface)
@@ -2716,11 +2600,11 @@ mt7531_mac_config(struct dsa_switch *ds, int port, unsigned int mode,
 		phydev = dp->slave->phydev;
 		return mt7531_rgmii_setup(priv, port, interface, phydev);
 	case PHY_INTERFACE_MODE_SGMII:
-		return mt7531_sgmii_setup_mode_an(priv, port, interface);
 	case PHY_INTERFACE_MODE_NA:
 	case PHY_INTERFACE_MODE_1000BASEX:
 	case PHY_INTERFACE_MODE_2500BASEX:
-		return mt7531_sgmii_setup_mode_force(priv, port, interface);
+		/* handled in SGMII PCS driver */
+		return 0;
 	default:
 		return -EINVAL;
 	}
@@ -2745,11 +2629,11 @@ mt753x_phylink_mac_select_pcs(struct dsa_switch *ds, int port,
 
 	switch (interface) {
 	case PHY_INTERFACE_MODE_TRGMII:
+		return &priv->pcs[port].pcs;
 	case PHY_INTERFACE_MODE_SGMII:
 	case PHY_INTERFACE_MODE_1000BASEX:
 	case PHY_INTERFACE_MODE_2500BASEX:
-		return &priv->pcs[port].pcs;
-
+		return priv->ports[port].sgmii_pcs;
 	default:
 		return NULL;
 	}
@@ -2987,86 +2871,6 @@ static void mt7530_pcs_get_state(struct phylink_pcs *pcs,
 		state->pause |= MLO_PAUSE_TX;
 }
 
-static int
-mt7531_sgmii_pcs_get_state_an(struct mt7530_priv *priv, int port,
-			      struct phylink_link_state *state)
-{
-	u32 status, val;
-	u16 config_reg;
-
-	status = mt7530_read(priv, MT7531_PCS_CONTROL_1(port));
-	state->link = !!(status & MT7531_SGMII_LINK_STATUS);
-	state->an_complete = !!(status & MT7531_SGMII_AN_COMPLETE);
-	if (state->interface == PHY_INTERFACE_MODE_SGMII &&
-	    (status & MT7531_SGMII_AN_ENABLE)) {
-		val = mt7530_read(priv, MT7531_PCS_SPEED_ABILITY(port));
-		config_reg = val >> 16;
-
-		switch (config_reg & LPA_SGMII_SPD_MASK) {
-		case LPA_SGMII_1000:
-			state->speed = SPEED_1000;
-			break;
-		case LPA_SGMII_100:
-			state->speed = SPEED_100;
-			break;
-		case LPA_SGMII_10:
-			state->speed = SPEED_10;
-			break;
-		default:
-			dev_err(priv->dev, "invalid sgmii PHY speed\n");
-			state->link = false;
-			return -EINVAL;
-		}
-
-		if (config_reg & LPA_SGMII_FULL_DUPLEX)
-			state->duplex = DUPLEX_FULL;
-		else
-			state->duplex = DUPLEX_HALF;
-	}
-
-	return 0;
-}
-
-static void
-mt7531_sgmii_pcs_get_state_inband(struct mt7530_priv *priv, int port,
-				  struct phylink_link_state *state)
-{
-	unsigned int val;
-
-	val = mt7530_read(priv, MT7531_PCS_CONTROL_1(port));
-	state->link = !!(val & MT7531_SGMII_LINK_STATUS);
-	if (!state->link)
-		return;
-
-	state->an_complete = state->link;
-
-	if (state->interface == PHY_INTERFACE_MODE_2500BASEX)
-		state->speed = SPEED_2500;
-	else
-		state->speed = SPEED_1000;
-
-	state->duplex = DUPLEX_FULL;
-	state->pause = MLO_PAUSE_NONE;
-}
-
-static void mt7531_pcs_get_state(struct phylink_pcs *pcs,
-				 struct phylink_link_state *state)
-{
-	struct mt7530_priv *priv = pcs_to_mt753x_pcs(pcs)->priv;
-	int port = pcs_to_mt753x_pcs(pcs)->port;
-
-	if (state->interface == PHY_INTERFACE_MODE_SGMII) {
-		mt7531_sgmii_pcs_get_state_an(priv, port, state);
-		return;
-	} else if ((state->interface == PHY_INTERFACE_MODE_1000BASEX) ||
-		   (state->interface == PHY_INTERFACE_MODE_2500BASEX)) {
-		mt7531_sgmii_pcs_get_state_inband(priv, port, state);
-		return;
-	}
-
-	state->link = false;
-}
-
 static int mt753x_pcs_config(struct phylink_pcs *pcs, unsigned int mode,
 			     phy_interface_t interface,
 			     const unsigned long *advertising,
@@ -3086,18 +2890,57 @@ static const struct phylink_pcs_ops mt7530_pcs_ops = {
 	.pcs_an_restart = mt7530_pcs_an_restart,
 };
 
-static const struct phylink_pcs_ops mt7531_pcs_ops = {
-	.pcs_validate = mt753x_pcs_validate,
-	.pcs_get_state = mt7531_pcs_get_state,
-	.pcs_config = mt753x_pcs_config,
-	.pcs_an_restart = mt7531_pcs_an_restart,
-	.pcs_link_up = mt7531_pcs_link_up,
+static int mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val)
+{
+	struct mt7530_priv *priv = context;
+
+	*val = mt7530_read(priv, reg);
+	return 0;
+};
+
+static int mt7530_regmap_write(void *context, unsigned int reg, unsigned int val)
+{
+	struct mt7530_priv *priv = context;
+
+	mt7530_write(priv, reg, val);
+	return 0;
+};
+
+static int mt7530_regmap_update_bits(void *context, unsigned int reg,
+				     unsigned int mask, unsigned int val)
+{
+	struct mt7530_priv *priv = context;
+
+	mt7530_rmw(priv, reg, mask, val);
+	return 0;
+};
+
+static const struct regmap_bus mt7531_regmap_bus = {
+	.reg_write = mt7530_regmap_write,
+	.reg_read = mt7530_regmap_read,
+	.reg_update_bits = mt7530_regmap_update_bits,
+};
+
+#define MT7531_PCS_REGMAP_CONFIG(_name, _reg_base) \
+	{				\
+		.name = _name,		\
+		.reg_bits = 16,		\
+		.val_bits = 32,		\
+		.reg_stride = 4,	\
+		.reg_base = _reg_base,	\
+		.max_register = 0x17c,	\
+	}
+
+static const struct regmap_config mt7531_pcs_config[] = {
+	MT7531_PCS_REGMAP_CONFIG("port5", MT7531_SGMII_REG_BASE(5)),
+	MT7531_PCS_REGMAP_CONFIG("port6", MT7531_SGMII_REG_BASE(6)),
 };
 
 static int
 mt753x_setup(struct dsa_switch *ds)
 {
 	struct mt7530_priv *priv = ds->priv;
+	struct regmap *regmap;
 	int i, ret;
 
 	/* Initialise the PCS devices */
@@ -3105,8 +2948,6 @@ mt753x_setup(struct dsa_switch *ds)
 		priv->pcs[i].pcs.ops = priv->info->pcs_ops;
 		priv->pcs[i].priv = priv;
 		priv->pcs[i].port = i;
-		if (mt753x_is_mac_port(i))
-			priv->pcs[i].pcs.poll = 1;
 	}
 
 	ret = priv->info->sw_setup(ds);
@@ -3121,6 +2962,16 @@ mt753x_setup(struct dsa_switch *ds)
 	if (ret && priv->irq)
 		mt7530_free_irq_common(priv);
 
+	if (priv->id == ID_MT7531)
+		for (i = 0; i < 2; i++) {
+			regmap = devm_regmap_init(ds->dev,
+						  &mt7531_regmap_bus, priv,
+						  &mt7531_pcs_config[i]);
+			priv->ports[5 + i].sgmii_pcs =
+				mtk_pcs_lynxi_create(ds->dev, regmap,
+						     MT7531_PHYA_CTRL_SIGNAL3, 0);
+		}
+
 	return ret;
 }
 
@@ -3216,7 +3067,7 @@ static const struct mt753x_info mt753x_table[] = {
 	},
 	[ID_MT7531] = {
 		.id = ID_MT7531,
-		.pcs_ops = &mt7531_pcs_ops,
+		.pcs_ops = &mt7530_pcs_ops,
 		.sw_setup = mt7531_setup,
 		.phy_read_c22 = mt7531_ind_c22_phy_read,
 		.phy_write_c22 = mt7531_ind_c22_phy_write,
@@ -3326,7 +3177,7 @@ static void
 mt7530_remove(struct mdio_device *mdiodev)
 {
 	struct mt7530_priv *priv = dev_get_drvdata(&mdiodev->dev);
-	int ret = 0;
+	int ret = 0, i;
 
 	if (!priv)
 		return;
@@ -3345,6 +3196,10 @@ mt7530_remove(struct mdio_device *mdiodev)
 		mt7530_free_irq(priv);
 
 	dsa_unregister_switch(priv->ds);
+
+	for (i = 0; i < 2; ++i)
+		mtk_pcs_lynxi_destroy(priv->ports[5 + i].sgmii_pcs);
+
 	mutex_destroy(&priv->reg_mutex);
 }
 
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index 6b2fc6290ea8..c5d29f3fc1d8 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -364,47 +364,8 @@ enum mt7530_vlan_port_acc_frm {
 					 CCR_TX_OCT_CNT_BAD)
 
 /* MT7531 SGMII register group */
-#define MT7531_SGMII_REG_BASE		0x5000
-#define MT7531_SGMII_REG(p, r)		(MT7531_SGMII_REG_BASE + \
-					((p) - 5) * 0x1000 + (r))
-
-/* Register forSGMII PCS_CONTROL_1 */
-#define MT7531_PCS_CONTROL_1(p)		MT7531_SGMII_REG(p, 0x00)
-#define  MT7531_SGMII_LINK_STATUS	BIT(18)
-#define  MT7531_SGMII_AN_ENABLE		BIT(12)
-#define  MT7531_SGMII_AN_RESTART	BIT(9)
-#define  MT7531_SGMII_AN_COMPLETE	BIT(21)
-
-/* Register for SGMII PCS_SPPED_ABILITY */
-#define MT7531_PCS_SPEED_ABILITY(p)	MT7531_SGMII_REG(p, 0x08)
-#define  MT7531_SGMII_TX_CONFIG_MASK	GENMASK(15, 0)
-#define  MT7531_SGMII_TX_CONFIG		BIT(0)
-
-/* Register for SGMII_MODE */
-#define MT7531_SGMII_MODE(p)		MT7531_SGMII_REG(p, 0x20)
-#define  MT7531_SGMII_REMOTE_FAULT_DIS	BIT(8)
-#define  MT7531_SGMII_IF_MODE_MASK	GENMASK(5, 1)
-#define  MT7531_SGMII_FORCE_DUPLEX	BIT(4)
-#define  MT7531_SGMII_FORCE_SPEED_MASK	GENMASK(3, 2)
-#define  MT7531_SGMII_FORCE_SPEED_1000	BIT(3)
-#define  MT7531_SGMII_FORCE_SPEED_100	BIT(2)
-#define  MT7531_SGMII_FORCE_SPEED_10	0
-#define  MT7531_SGMII_SPEED_DUPLEX_AN	BIT(1)
-
-enum mt7531_sgmii_force_duplex {
-	MT7531_SGMII_FORCE_FULL_DUPLEX = 0,
-	MT7531_SGMII_FORCE_HALF_DUPLEX = 0x10,
-};
-
-/* Fields of QPHY_PWR_STATE_CTRL */
-#define MT7531_QPHY_PWR_STATE_CTRL(p)	MT7531_SGMII_REG(p, 0xe8)
-#define  MT7531_SGMII_PHYA_PWD		BIT(4)
-
-/* Values of SGMII SPEED */
-#define MT7531_PHYA_CTRL_SIGNAL3(p)	MT7531_SGMII_REG(p, 0x128)
-#define  MT7531_RG_TPHY_SPEED_MASK	(BIT(2) | BIT(3))
-#define  MT7531_RG_TPHY_SPEED_1_25G	0x0
-#define  MT7531_RG_TPHY_SPEED_3_125G	BIT(2)
+#define MT7531_SGMII_REG_BASE(p)	(0x5000 + ((p) - 5) * 0x1000)
+#define MT7531_PHYA_CTRL_SIGNAL3	0x128
 
 /* Register for system reset */
 #define MT7530_SYS_CTRL			0x7000
@@ -703,13 +664,13 @@ struct mt7530_fdb {
  * @pm:		The matrix used to show all connections with the port.
  * @pvid:	The VLAN specified is to be considered a PVID at ingress.  Any
  *		untagged frames will be assigned to the related VLAN.
- * @vlan_filtering: The flags indicating whether the port that can recognize
- *		    VLAN-tagged frames.
+ * @sgmii_pcs:	Pointer to PCS instance for SerDes ports
  */
 struct mt7530_port {
 	bool enable;
 	u32 pm;
 	u16 pvid;
+	struct phylink_pcs *sgmii_pcs;
 };
 
 /* Port 5 interface select definitions */
-- 
cgit v1.2.3-70-g09d2


From a08df15eab0c572bc1d98c2392bcfcd744f26eda Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Sun, 19 Mar 2023 13:24:33 -0400
Subject: net: cxgb3: remove unused fl_to_qset function

clang with W=1 reports
drivers/net/ethernet/chelsio/cxgb3/sge.c:169:32: error: unused function
  'fl_to_qset' [-Werror,-Wunused-function]
static inline struct sge_qset *fl_to_qset(const struct sge_fl *q, int qidx)
                               ^
This function is not used, so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20230319172433.1708161-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/chelsio/cxgb3/sge.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/ethernet/chelsio/cxgb3/sge.c b/drivers/net/ethernet/chelsio/cxgb3/sge.c
index 62dfbdd33365..efa7f401529e 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/sge.c
@@ -166,11 +166,6 @@ static u8 flit_desc_map[] = {
 #endif
 };
 
-static inline struct sge_qset *fl_to_qset(const struct sge_fl *q, int qidx)
-{
-	return container_of(q, struct sge_qset, fl[qidx]);
-}
-
 static inline struct sge_qset *rspq_to_qset(const struct sge_rspq *q)
 {
 	return container_of(q, struct sge_qset, rspq);
-- 
cgit v1.2.3-70-g09d2


From 04aae213e719ec2bb310158c4025316ace50589b Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 20 Mar 2023 18:41:13 -0700
Subject: net: skbuff: rename __pkt_vlan_present_offset to __mono_tc_offset

vlan_present is gone since
commit 354259fa73e2 ("net: remove skb->vlan_present")
rename the offset field to what BPF is currently looking
for in this byte - mono_delivery_time and tc_at_ingress.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230321014115.997841-2-kuba@kernel.org
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/skbuff.h                               | 4 ++--
 net/core/filter.c                                    | 8 ++++----
 tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c | 6 +++---
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 3f3a2a82a86b..5a63878a4550 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -955,7 +955,7 @@ struct sk_buff {
 	__u8			csum_valid:1;
 
 	/* private: */
-	__u8			__pkt_vlan_present_offset[0];
+	__u8			__mono_tc_offset[0];
 	/* public: */
 	__u8			remcsum_offload:1;
 	__u8			csum_complete_sw:1;
@@ -1078,7 +1078,7 @@ struct sk_buff {
 #define TC_AT_INGRESS_MASK		(1 << 7)
 #define SKB_MONO_DELIVERY_TIME_MASK	(1 << 5)
 #endif
-#define PKT_VLAN_PRESENT_OFFSET	offsetof(struct sk_buff, __pkt_vlan_present_offset)
+#define SKB_BF_MONO_TC_OFFSET		offsetof(struct sk_buff, __mono_tc_offset)
 
 #ifdef __KERNEL__
 /*
diff --git a/net/core/filter.c b/net/core/filter.c
index 50f649f1b4a9..3370efad1dda 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -9185,7 +9185,7 @@ static struct bpf_insn *bpf_convert_tstamp_type_read(const struct bpf_insn *si,
 	__u8 tmp_reg = BPF_REG_AX;
 
 	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg,
-			      PKT_VLAN_PRESENT_OFFSET);
+			      SKB_BF_MONO_TC_OFFSET);
 	*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg,
 				SKB_MONO_DELIVERY_TIME_MASK, 2);
 	*insn++ = BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_UNSPEC);
@@ -9232,7 +9232,7 @@ static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_prog *prog,
 		/* AX is needed because src_reg and dst_reg could be the same */
 		__u8 tmp_reg = BPF_REG_AX;
 
-		*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, PKT_VLAN_PRESENT_OFFSET);
+		*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET);
 		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg,
 					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK);
 		*insn++ = BPF_JMP32_IMM(BPF_JNE, tmp_reg,
@@ -9267,14 +9267,14 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog,
 	if (!prog->tstamp_type_access) {
 		__u8 tmp_reg = BPF_REG_AX;
 
-		*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, PKT_VLAN_PRESENT_OFFSET);
+		*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET);
 		/* Writing __sk_buff->tstamp as ingress, goto <clear> */
 		*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1);
 		/* goto <store> */
 		*insn++ = BPF_JMP_A(2);
 		/* <clear>: mono_delivery_time */
 		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK);
-		*insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, PKT_VLAN_PRESENT_OFFSET);
+		*insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET);
 	}
 #endif
 
diff --git a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
index d5fe3d4b936c..ae7b6e50e405 100644
--- a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
+++ b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
@@ -68,17 +68,17 @@ static struct test_case test_cases[] = {
 #if defined(__x86_64__) || defined(__aarch64__)
 	{
 		N(SCHED_CLS, struct __sk_buff, tstamp),
-		.read  = "r11 = *(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset);"
+		.read  = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);"
 			 "w11 &= 160;"
 			 "if w11 != 0xa0 goto pc+2;"
 			 "$dst = 0;"
 			 "goto pc+1;"
 			 "$dst = *(u64 *)($ctx + sk_buff::tstamp);",
-		.write = "r11 = *(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset);"
+		.write = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);"
 			 "if w11 & 0x80 goto pc+1;"
 			 "goto pc+2;"
 			 "w11 &= -33;"
-			 "*(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset) = r11;"
+			 "*(u8 *)($ctx + sk_buff::__mono_tc_offset) = r11;"
 			 "*(u64 *)($ctx + sk_buff::tstamp) = $src;",
 	},
 #endif
-- 
cgit v1.2.3-70-g09d2


From b94e032b7ad6318b36615f0e0cc3b0d61a5531e8 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 20 Mar 2023 18:41:14 -0700
Subject: net: skbuff: reorder bytes 2 and 3 of the bitfield

BPF needs to know the offsets of fields it tries to access.
Zero-length fields are added to make offsetof() work.
This unfortunately partitions the bitfield (fields across
the zero-length members can't be coalesced).

Reorder bytes 2 and 3, BPF needs to know the offset of fields
previously in byte 3 and some fields in byte 2 should really
be optional.

The two bytes are always in the same cacheline so it should
not matter.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230321014115.997841-3-kuba@kernel.org
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/skbuff.h | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 5a63878a4550..36d31e74db37 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -944,16 +944,6 @@ struct sk_buff {
 	__u8			ip_summed:2;
 	__u8			ooo_okay:1;
 
-	__u8			l4_hash:1;
-	__u8			sw_hash:1;
-	__u8			wifi_acked_valid:1;
-	__u8			wifi_acked:1;
-	__u8			no_fcs:1;
-	/* Indicates the inner headers are valid in the skbuff. */
-	__u8			encapsulation:1;
-	__u8			encap_hdr_csum:1;
-	__u8			csum_valid:1;
-
 	/* private: */
 	__u8			__mono_tc_offset[0];
 	/* public: */
@@ -966,6 +956,16 @@ struct sk_buff {
 	__u8			tc_skip_classify:1;
 	__u8			tc_at_ingress:1;	/* See TC_AT_INGRESS_MASK */
 #endif
+
+	__u8			l4_hash:1;
+	__u8			sw_hash:1;
+	__u8			wifi_acked_valid:1;
+	__u8			wifi_acked:1;
+	__u8			no_fcs:1;
+	/* Indicates the inner headers are valid in the skbuff. */
+	__u8			encapsulation:1;
+	__u8			encap_hdr_csum:1;
+	__u8			csum_valid:1;
 #ifdef CONFIG_IPV6_NDISC_NODETYPE
 	__u8			ndisc_nodetype:2;
 #endif
-- 
cgit v1.2.3-70-g09d2


From c0ba861117c3e8deb03855d7dc5a7717958bbb18 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 20 Mar 2023 18:41:15 -0700
Subject: net: skbuff: move the fields BPF cares about directly next to the
 offset marker

To avoid more possible BPF dependencies with moving bitfields
around keep the fields BPF cares about right next to the offset
marker.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230321014115.997841-4-kuba@kernel.org
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/skbuff.h                               | 18 +++++++++---------
 tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c |  8 ++++----
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 36d31e74db37..6aeb0e7b9511 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -947,15 +947,15 @@ struct sk_buff {
 	/* private: */
 	__u8			__mono_tc_offset[0];
 	/* public: */
-	__u8			remcsum_offload:1;
-	__u8			csum_complete_sw:1;
-	__u8			csum_level:2;
-	__u8			dst_pending_confirm:1;
 	__u8			mono_delivery_time:1;	/* See SKB_MONO_DELIVERY_TIME_MASK */
 #ifdef CONFIG_NET_CLS_ACT
-	__u8			tc_skip_classify:1;
 	__u8			tc_at_ingress:1;	/* See TC_AT_INGRESS_MASK */
+	__u8			tc_skip_classify:1;
 #endif
+	__u8			remcsum_offload:1;
+	__u8			csum_complete_sw:1;
+	__u8			csum_level:2;
+	__u8			dst_pending_confirm:1;
 
 	__u8			l4_hash:1;
 	__u8			sw_hash:1;
@@ -1072,11 +1072,11 @@ struct sk_buff {
  * around, you also must adapt these constants.
  */
 #ifdef __BIG_ENDIAN_BITFIELD
-#define TC_AT_INGRESS_MASK		(1 << 0)
-#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 2)
+#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 7)
+#define TC_AT_INGRESS_MASK		(1 << 6)
 #else
-#define TC_AT_INGRESS_MASK		(1 << 7)
-#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 5)
+#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 0)
+#define TC_AT_INGRESS_MASK		(1 << 1)
 #endif
 #define SKB_BF_MONO_TC_OFFSET		offsetof(struct sk_buff, __mono_tc_offset)
 
diff --git a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
index ae7b6e50e405..4951aa978f33 100644
--- a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
+++ b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
@@ -69,15 +69,15 @@ static struct test_case test_cases[] = {
 	{
 		N(SCHED_CLS, struct __sk_buff, tstamp),
 		.read  = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);"
-			 "w11 &= 160;"
-			 "if w11 != 0xa0 goto pc+2;"
+			 "w11 &= 3;"
+			 "if w11 != 0x3 goto pc+2;"
 			 "$dst = 0;"
 			 "goto pc+1;"
 			 "$dst = *(u64 *)($ctx + sk_buff::tstamp);",
 		.write = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);"
-			 "if w11 & 0x80 goto pc+1;"
+			 "if w11 & 0x2 goto pc+1;"
 			 "goto pc+2;"
-			 "w11 &= -33;"
+			 "w11 &= -2;"
 			 "*(u8 *)($ctx + sk_buff::__mono_tc_offset) = r11;"
 			 "*(u64 *)($ctx + sk_buff::tstamp) = $src;",
 	},
-- 
cgit v1.2.3-70-g09d2


From a4926c2943dd17f84b552ba6fb754df0dc6a2fa8 Mon Sep 17 00:00:00 2001
From: Marek Behún <kabel@kernel.org>
Date: Sun, 19 Mar 2023 15:02:38 +0100
Subject: net: dsa: mv88e6xxx: fix mdio bus' phy_mask member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit 2c7e46edbd03 ("net: dsa: mv88e6xxx: mask apparently non-existing
phys during probing") added non-trivial bus->phy_mask in
mv88e6xxx_mdio_register() in order to avoid excessive mdio bus
transactions during probing.

But the mask is incorrect for switches with non-zero phy_base_addr (such
as 88E6341).

Fix this.

Fixes: 2c7e46edbd03 ("net: dsa: mv88e6xxx: mask apparently non-existing phys during probing")
Signed-off-by: Marek Behún <kabel@kernel.org>
Tested-by: Klaus Kudielka <klaus.kudielka@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20230319140238.9470-1-kabel@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 260e38c5c6e6..b73d1d6747b7 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -3805,7 +3805,9 @@ static int mv88e6xxx_mdio_register(struct mv88e6xxx_chip *chip,
 	bus->read_c45 = mv88e6xxx_mdio_read_c45;
 	bus->write_c45 = mv88e6xxx_mdio_write_c45;
 	bus->parent = chip->dev;
-	bus->phy_mask = GENMASK(31, mv88e6xxx_num_ports(chip));
+	bus->phy_mask = ~GENMASK(chip->info->phy_base_addr +
+				 mv88e6xxx_num_ports(chip) - 1,
+				 chip->info->phy_base_addr);
 
 	if (!external) {
 		err = mv88e6xxx_g2_irq_mdio_setup(chip, bus);
-- 
cgit v1.2.3-70-g09d2


From 9ef70d0130f282638b28cfce24222f71ada00c9c Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Sun, 19 Mar 2023 12:33:29 +0000
Subject: net: dsa: qca8k: remove assignment of an_enabled in pcs_get_state()

pcs_get_state() implementations are not supposed to alter an_enabled.
Remove this assignment.

Fixes: b3591c2a3661 ("net: dsa: qca8k: Switch to PHYLINK instead of PHYLIB")
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/E1pdsE5-00Dl2l-8F@rmk-PC.armlinux.org.uk
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/qca/qca8k-8xxx.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c
index 55df4479ea30..62810903f1b3 100644
--- a/drivers/net/dsa/qca/qca8k-8xxx.c
+++ b/drivers/net/dsa/qca/qca8k-8xxx.c
@@ -1483,7 +1483,6 @@ static void qca8k_pcs_get_state(struct phylink_pcs *pcs,
 
 	state->link = !!(reg & QCA8K_PORT_STATUS_LINK_UP);
 	state->an_complete = state->link;
-	state->an_enabled = !!(reg & QCA8K_PORT_STATUS_LINK_AUTO);
 	state->duplex = (reg & QCA8K_PORT_STATUS_DUPLEX) ? DUPLEX_FULL :
 							   DUPLEX_HALF;
 
-- 
cgit v1.2.3-70-g09d2


From ce3bf94871f7aeb0167370206699872472f30220 Mon Sep 17 00:00:00 2001
From: Álvaro Fernández Rojas <noltari@gmail.com>
Date: Sun, 19 Mar 2023 23:08:05 +0100
Subject: net: dsa: b53: add support for BCM63xx RGMIIs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BCM63xx RGMII ports require additional configuration in order to work.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230319220805.124024-1-noltari@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/b53/b53_common.c | 46 ++++++++++++++++++++++++++++++++++++++++
 drivers/net/dsa/b53/b53_priv.h   |  1 +
 2 files changed, 47 insertions(+)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 59cdfc51ce06..6e212f6f1cb9 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -1209,6 +1209,46 @@ static void b53_force_port_config(struct b53_device *dev, int port,
 	b53_write8(dev, B53_CTRL_PAGE, off, reg);
 }
 
+static void b53_adjust_63xx_rgmii(struct dsa_switch *ds, int port,
+				  phy_interface_t interface)
+{
+	struct b53_device *dev = ds->priv;
+	u8 rgmii_ctrl = 0, off;
+
+	if (port == dev->imp_port)
+		off = B53_RGMII_CTRL_IMP;
+	else
+		off = B53_RGMII_CTRL_P(port);
+
+	b53_read8(dev, B53_CTRL_PAGE, off, &rgmii_ctrl);
+
+	switch (interface) {
+	case PHY_INTERFACE_MODE_RGMII_ID:
+		rgmii_ctrl |= (RGMII_CTRL_DLL_RXC | RGMII_CTRL_DLL_TXC);
+		break;
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+		rgmii_ctrl &= ~(RGMII_CTRL_DLL_TXC);
+		rgmii_ctrl |= RGMII_CTRL_DLL_RXC;
+		break;
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		rgmii_ctrl &= ~(RGMII_CTRL_DLL_RXC);
+		rgmii_ctrl |= RGMII_CTRL_DLL_TXC;
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	default:
+		rgmii_ctrl &= ~(RGMII_CTRL_DLL_RXC | RGMII_CTRL_DLL_TXC);
+		break;
+	}
+
+	if (port != dev->imp_port)
+		rgmii_ctrl |= RGMII_CTRL_ENABLE_GMII;
+
+	b53_write8(dev, B53_CTRL_PAGE, off, rgmii_ctrl);
+
+	dev_dbg(ds->dev, "Configured port %d for %s\n", port,
+		phy_modes(interface));
+}
+
 static void b53_adjust_link(struct dsa_switch *ds, int port,
 			    struct phy_device *phydev)
 {
@@ -1235,6 +1275,9 @@ static void b53_adjust_link(struct dsa_switch *ds, int port,
 			      tx_pause, rx_pause);
 	b53_force_link(dev, port, phydev->link);
 
+	if (is63xx(dev) && port >= B53_63XX_RGMII0)
+		b53_adjust_63xx_rgmii(ds, port, phydev->interface);
+
 	if (is531x5(dev) && phy_interface_is_rgmii(phydev)) {
 		if (port == dev->imp_port)
 			off = B53_RGMII_CTRL_IMP;
@@ -1402,6 +1445,9 @@ void b53_phylink_mac_link_up(struct dsa_switch *ds, int port,
 {
 	struct b53_device *dev = ds->priv;
 
+	if (is63xx(dev) && port >= B53_63XX_RGMII0)
+		b53_adjust_63xx_rgmii(ds, port, interface);
+
 	if (mode == MLO_AN_PHY)
 		return;
 
diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h
index 795cbffd5c2b..4cf9f540696e 100644
--- a/drivers/net/dsa/b53/b53_priv.h
+++ b/drivers/net/dsa/b53/b53_priv.h
@@ -211,6 +211,7 @@ static inline int is58xx(struct b53_device *dev)
 		dev->chip_id == BCM7278_DEVICE_ID;
 }
 
+#define B53_63XX_RGMII0	4
 #define B53_CPU_PORT_25	5
 #define B53_CPU_PORT	8
 
-- 
cgit v1.2.3-70-g09d2


From 251d5a2813f96a2f8d566575a9b97a42b32669f9 Mon Sep 17 00:00:00 2001
From: Josef Miegl <josef@miegl.cz>
Date: Sun, 19 Mar 2023 23:09:54 +0100
Subject: net: geneve: accept every ethertype

The Geneve encapsulation, as defined in RFC 8926, has a Protocol Type
field, which states the Ethertype of the payload appearing after the
Geneve header.

Commit 435fe1c0c1f7 ("net: geneve: support IPv4/IPv6 as inner protocol")
introduced a new IFLA_GENEVE_INNER_PROTO_INHERIT flag that allowed the
use of other Ethertypes than Ethernet. However, it did not get rid of a
restriction that prohibits receiving payloads other than Ethernet,
instead the commit white-listed additional Ethertypes, IPv4 and IPv6.

This patch removes this restriction, making it possible to receive any
Ethertype as a payload, if the IFLA_GENEVE_INNER_PROTO_INHERIT flag is
set.

The restriction was set in place back in commit 0b5e8b8eeae4
("net: Add Geneve tunneling protocol driver"), which implemented a
protocol layer driver for Geneve to be used with Open vSwitch. The
relevant discussion about introducing the Ethertype white-list can be
found here:
https://lore.kernel.org/netdev/CAEP_g=_1q3ACX5NTHxLDnysL+dTMUVzdLpgw1apLKEdDSWPztw@mail.gmail.com/

<quote>
>> +       if (unlikely(geneveh->proto_type != htons(ETH_P_TEB)))
>
> Why? I thought the point of geneve carrying protocol field was to
> allow protocols other than Ethernet... is this temporary maybe?

Yes, it is temporary. Currently OVS only handles Ethernet packets but
this restriction can be lifted once we have a consumer that is capable
of handling other protocols.
</quote>

This white-list was then ported to a generic Geneve netdevice in commit
371bd1061d29 ("geneve: Consolidate Geneve functionality in single
module."). Preserving the Ethertype white-list at this point made sense,
as the Geneve device could send out only Ethernet payloads anyways.

However, now that the Geneve netdevice supports encapsulating other
payloads with IFLA_GENEVE_INNER_PROTO_INHERIT and we have a consumer
capable of other protocols, it seems appropriate to lift the restriction
and allow any Geneve payload to be received.

Signed-off-by: Josef Miegl <josef@miegl.cz>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Eyal Birger <eyal.birger@gmail.com>
Link: https://lore.kernel.org/r/20230319220954.21834-1-josef@miegl.cz
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/geneve.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 5c3c7169c0a2..78f9d588f712 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -365,13 +365,6 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (unlikely(geneveh->ver != GENEVE_VER))
 		goto drop;
 
-	inner_proto = geneveh->proto_type;
-
-	if (unlikely((inner_proto != htons(ETH_P_TEB) &&
-		      inner_proto != htons(ETH_P_IP) &&
-		      inner_proto != htons(ETH_P_IPV6))))
-		goto drop;
-
 	gs = rcu_dereference_sk_user_data(sk);
 	if (!gs)
 		goto drop;
@@ -380,6 +373,8 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!geneve)
 		goto drop;
 
+	inner_proto = geneveh->proto_type;
+
 	if (unlikely((!geneve->cfg.inner_proto_inherit &&
 		      inner_proto != htons(ETH_P_TEB)))) {
 		geneve->dev->stats.rx_dropped++;
-- 
cgit v1.2.3-70-g09d2


From c8384d4a51e7cb0e6587f3143f29099f202c5de1 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Sun, 19 Mar 2023 16:41:08 -0700
Subject: net: pasemi: Fix return type of pasemi_mac_start_tx()

With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
indirect call targets are validated against the expected function
pointer prototype to make sure the call target is valid to help mitigate
ROP attacks. If they are not identical, there is a failure at run time,
which manifests as either a kernel panic or thread getting killed. A
warning in clang aims to catch these at compile time, which reveals:

  drivers/net/ethernet/pasemi/pasemi_mac.c:1665:21: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
          .ndo_start_xmit         = pasemi_mac_start_tx,
                                    ^~~~~~~~~~~~~~~~~~~
  1 error generated.

->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
'netdev_tx_t', not 'int'. Adjust the return type of
pasemi_mac_start_tx() to match the prototype's to resolve the warning.
While PowerPC does not currently implement support for kCFI, it could in
the future, which means this warning becomes a fatal CFI failure at run
time.

Link: https://github.com/ClangBuiltLinux/linux/issues/1750
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20230319-pasemi-incompatible-pointer-types-strict-v1-1-1b9459d8aef0@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/pasemi/pasemi_mac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.c b/drivers/net/ethernet/pasemi/pasemi_mac.c
index aaab590ef548..ed7dd0a04235 100644
--- a/drivers/net/ethernet/pasemi/pasemi_mac.c
+++ b/drivers/net/ethernet/pasemi/pasemi_mac.c
@@ -1423,7 +1423,7 @@ static void pasemi_mac_queue_csdesc(const struct sk_buff *skb,
 	write_dma_reg(PAS_DMA_TXCHAN_INCR(txring->chan.chno), 2);
 }
 
-static int pasemi_mac_start_tx(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t pasemi_mac_start_tx(struct sk_buff *skb, struct net_device *dev)
 {
 	struct pasemi_mac * const mac = netdev_priv(dev);
 	struct pasemi_mac_txring * const txring = tx_ring(mac);
-- 
cgit v1.2.3-70-g09d2


From d71980d47e270a7dffef4f3541d550cbde686d9a Mon Sep 17 00:00:00 2001
From: Andrii Staikov <andrii.staikov@intel.com>
Date: Wed, 8 Mar 2023 11:36:51 +0100
Subject: igb: refactor igb_ptp_adjfine_82580 to use diff_by_scaled_ppm

Driver's .adjfine interface functions use adjust_by_scaled_ppm and
diff_by_scaled_ppm introduced in commit 1060707e3809
("ptp: introduce helpers to adjust by scaled parts per million")
to calculate the required adjustment in a concise manner,
but not igb_ptp_adjfine_82580.
Fix it by introducing IGB_82580_BASE_PERIOD and changing function logic
to use diff_by_scaled_ppm.

Signed-off-by: Andrii Staikov <andrii.staikov@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/igb_ptp.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index 6f471b91f562..405886ee5261 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -67,6 +67,7 @@
 #define INCVALUE_82576_MASK		GENMASK(E1000_TIMINCA_16NS_SHIFT - 1, 0)
 #define INCVALUE_82576			(16u << IGB_82576_TSYNC_SHIFT)
 #define IGB_NBITS_82580			40
+#define IGB_82580_BASE_PERIOD		0x800000000
 
 static void igb_ptp_tx_hwtstamp(struct igb_adapter *adapter);
 static void igb_ptp_sdp_init(struct igb_adapter *adapter);
@@ -209,17 +210,11 @@ static int igb_ptp_adjfine_82580(struct ptp_clock_info *ptp, long scaled_ppm)
 	struct igb_adapter *igb = container_of(ptp, struct igb_adapter,
 					       ptp_caps);
 	struct e1000_hw *hw = &igb->hw;
-	int neg_adj = 0;
+	bool neg_adj;
 	u64 rate;
 	u32 inca;
 
-	if (scaled_ppm < 0) {
-		neg_adj = 1;
-		scaled_ppm = -scaled_ppm;
-	}
-	rate = scaled_ppm;
-	rate <<= 13;
-	rate = div_u64(rate, 15625);
+	neg_adj = diff_by_scaled_ppm(IGB_82580_BASE_PERIOD, scaled_ppm, &rate);
 
 	inca = rate & INCVALUE_MASK;
 	if (neg_adj)
-- 
cgit v1.2.3-70-g09d2


From 5a9b7bfb0d15e0877115b8fa6992ebc59855c5d7 Mon Sep 17 00:00:00 2001
From: Dawid Wesierski <dawidx.wesierski@intel.com>
Date: Wed, 1 Mar 2023 11:57:05 +0100
Subject: igbvf: add PCI reset handler functions

There was a problem with resuming ping after conducting a PCI reset.
This commit adds two functions, igbvf_io_prepare and igbvf_io_done,
which, after being added to the pci_error_handlers struct,
will prepare the drivers for a PCI reset and then bring the interface up
and reset it after. This will prevent the driver from ending up in
incorrect state. Test_and_set_bit is highly reliable in this context,
so we are not including a timeout in this commit

This introduces 900ms - 1100ms of overhead to this operation but it's in
non-time-critical flow. And also allows the driver to continue
functioning after the reset.

Functionality documented in ethernet-controller-i350-datasheet 4.2.1.3
https://www.intel.com/content/www/us/en/products/details/ethernet/gigabit-controllers/i350-controllers/docs.html

Signed-off-by: Dawid Wesierski <dawidx.wesierski@intel.com>
Signed-off-by: Kamil Maziarz <kamil.maziarz@intel.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igbvf/netdev.c | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index 3a32809510fc..790383078e0d 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -2589,6 +2589,33 @@ static void igbvf_io_resume(struct pci_dev *pdev)
 	netif_device_attach(netdev);
 }
 
+/**
+ * igbvf_io_prepare - prepare device driver for PCI reset
+ * @pdev: PCI device information struct
+ */
+static void igbvf_io_prepare(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct igbvf_adapter *adapter = netdev_priv(netdev);
+
+	while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state))
+		usleep_range(1000, 2000);
+	igbvf_down(adapter);
+}
+
+/**
+ * igbvf_io_reset_done - PCI reset done, device driver reset can begin
+ * @pdev: PCI device information struct
+ */
+static void igbvf_io_reset_done(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct igbvf_adapter *adapter = netdev_priv(netdev);
+
+	igbvf_up(adapter);
+	clear_bit(__IGBVF_RESETTING, &adapter->state);
+}
+
 static void igbvf_print_device_info(struct igbvf_adapter *adapter)
 {
 	struct e1000_hw *hw = &adapter->hw;
@@ -2916,6 +2943,8 @@ static const struct pci_error_handlers igbvf_err_handler = {
 	.error_detected = igbvf_io_error_detected,
 	.slot_reset = igbvf_io_slot_reset,
 	.resume = igbvf_io_resume,
+	.reset_prepare = igbvf_io_prepare,
+	.reset_done = igbvf_io_reset_done,
 };
 
 static const struct pci_device_id igbvf_pci_tbl[] = {
-- 
cgit v1.2.3-70-g09d2


From 65364bbe0b022f397142e2486d0296203c45f0bd Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Thu, 23 Feb 2023 09:03:17 +0200
Subject: igc: Remove obsolete DMA coalescing code

DMA coalescing is not applicable for i225 parts. This patch comes to tidy
up the driver code.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_defines.h |  3 ---
 drivers/net/ethernet/intel/igc/igc_i225.c    | 19 +++++--------------
 drivers/net/ethernet/intel/igc/igc_regs.h    |  1 -
 3 files changed, 5 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 9dec3563ce3a..44a507029946 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -662,9 +662,6 @@
  */
 #define IGC_TW_SYSTEM_100_MASK		0x0000FF00
 #define IGC_TW_SYSTEM_100_SHIFT		8
-#define IGC_DMACR_DMAC_EN		0x80000000 /* Enable DMA Coalescing */
-#define IGC_DMACR_DMACTHR_MASK		0x00FF0000
-#define IGC_DMACR_DMACTHR_SHIFT		16
 /* Reg val to set scale to 1024 nsec */
 #define IGC_LTRMINV_SCALE_1024		2
 /* Reg val to set scale to 32768 nsec */
diff --git a/drivers/net/ethernet/intel/igc/igc_i225.c b/drivers/net/ethernet/intel/igc/igc_i225.c
index 59d5c467ea6e..17546a035ab1 100644
--- a/drivers/net/ethernet/intel/igc/igc_i225.c
+++ b/drivers/net/ethernet/intel/igc/igc_i225.c
@@ -593,20 +593,11 @@ s32 igc_set_ltr_i225(struct igc_hw *hw, bool link)
 		size = rd32(IGC_RXPBS) &
 		       IGC_RXPBS_SIZE_I225_MASK;
 
-		/* Calculations vary based on DMAC settings. */
-		if (rd32(IGC_DMACR) & IGC_DMACR_DMAC_EN) {
-			size -= (rd32(IGC_DMACR) &
-				 IGC_DMACR_DMACTHR_MASK) >>
-				 IGC_DMACR_DMACTHR_SHIFT;
-			/* Convert size to bits. */
-			size *= 1024 * 8;
-		} else {
-			/* Convert size to bytes, subtract the MTU, and then
-			 * convert the size to bits.
-			 */
-			size *= 1024;
-			size *= 8;
-		}
+		/* Convert size to bytes, subtract the MTU, and then
+		 * convert the size to bits.
+		 */
+		size *= 1024;
+		size *= 8;
 
 		if (size < 0) {
 			hw_dbg("Invalid effective Rx buffer size %d\n",
diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h
index 01c86d36856d..dba5a5759b1c 100644
--- a/drivers/net/ethernet/intel/igc/igc_regs.h
+++ b/drivers/net/ethernet/intel/igc/igc_regs.h
@@ -292,7 +292,6 @@
 
 /* LTR registers */
 #define IGC_LTRC	0x01A0 /* Latency Tolerance Reporting Control */
-#define IGC_DMACR	0x02508 /* DMA Coalescing Control Register */
 #define IGC_LTRMINV	0x5BB0 /* LTR Minimum Value */
 #define IGC_LTRMAXV	0x5BB4 /* LTR Maximum Value */
 
-- 
cgit v1.2.3-70-g09d2


From f6f4e739b16487c89d29ef8b425d019bed64f541 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Mon, 20 Mar 2023 19:23:17 -0400
Subject: net: atheros: atl1c: remove unused atl1c_irq_reset function

clang with W=1 reports
drivers/net/ethernet/atheros/atl1c/atl1c_main.c:214:20: error:
  unused function 'atl1c_irq_reset' [-Werror,-Wunused-function]
static inline void atl1c_irq_reset(struct atl1c_adapter *adapter)
                   ^
This function is not used, so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://lore.kernel.org/r/20230320232317.1729464-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/atheros/atl1c/atl1c_main.c | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 40c781695d58..4a288799633f 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -207,16 +207,6 @@ static inline void atl1c_irq_disable(struct atl1c_adapter *adapter)
 	synchronize_irq(adapter->pdev->irq);
 }
 
-/**
- * atl1c_irq_reset - reset interrupt confiure on the NIC
- * @adapter: board private structure
- */
-static inline void atl1c_irq_reset(struct atl1c_adapter *adapter)
-{
-	atomic_set(&adapter->irq_sem, 1);
-	atl1c_irq_enable(adapter);
-}
-
 /*
  * atl1c_wait_until_idle - wait up to AT_HW_MAX_IDLE_DELAY reads
  * of the idle status register until the device is actually idle
-- 
cgit v1.2.3-70-g09d2


From 4c5c496a942fa3d2197f5be22762695170c95863 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 21 Mar 2023 04:01:13 +0000
Subject: ipv6: flowlabel: do not disable BH where not needed

struct ip6_flowlabel are rcu managed, and call_rcu() is used
to delay fl_free_rcu() after RCU grace period.

There is no point disabling BH for pure RCU lookups.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv6/ip6_flowlabel.c | 51 +++++++++++++++++++++++++-----------------------
 1 file changed, 27 insertions(+), 24 deletions(-)

diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 18481eb76a0a..b3ca4beb4405 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -58,18 +58,18 @@ DEFINE_STATIC_KEY_DEFERRED_FALSE(ipv6_flowlabel_exclusive, HZ);
 EXPORT_SYMBOL(ipv6_flowlabel_exclusive);
 
 #define for_each_fl_rcu(hash, fl)				\
-	for (fl = rcu_dereference_bh(fl_ht[(hash)]);		\
+	for (fl = rcu_dereference(fl_ht[(hash)]);		\
 	     fl != NULL;					\
-	     fl = rcu_dereference_bh(fl->next))
+	     fl = rcu_dereference(fl->next))
 #define for_each_fl_continue_rcu(fl)				\
-	for (fl = rcu_dereference_bh(fl->next);			\
+	for (fl = rcu_dereference(fl->next);			\
 	     fl != NULL;					\
-	     fl = rcu_dereference_bh(fl->next))
+	     fl = rcu_dereference(fl->next))
 
 #define for_each_sk_fl_rcu(np, sfl)				\
-	for (sfl = rcu_dereference_bh(np->ipv6_fl_list);	\
+	for (sfl = rcu_dereference(np->ipv6_fl_list);	\
 	     sfl != NULL;					\
-	     sfl = rcu_dereference_bh(sfl->next))
+	     sfl = rcu_dereference(sfl->next))
 
 static inline struct ip6_flowlabel *__fl_lookup(struct net *net, __be32 label)
 {
@@ -86,11 +86,11 @@ static struct ip6_flowlabel *fl_lookup(struct net *net, __be32 label)
 {
 	struct ip6_flowlabel *fl;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	fl = __fl_lookup(net, label);
 	if (fl && !atomic_inc_not_zero(&fl->users))
 		fl = NULL;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 	return fl;
 }
 
@@ -217,6 +217,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net,
 
 	fl->label = label & IPV6_FLOWLABEL_MASK;
 
+	rcu_read_lock();
 	spin_lock_bh(&ip6_fl_lock);
 	if (label == 0) {
 		for (;;) {
@@ -240,6 +241,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net,
 		if (lfl) {
 			atomic_inc(&lfl->users);
 			spin_unlock_bh(&ip6_fl_lock);
+			rcu_read_unlock();
 			return lfl;
 		}
 	}
@@ -249,6 +251,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net,
 	rcu_assign_pointer(fl_ht[FL_HASH(fl->label)], fl);
 	atomic_inc(&fl_size);
 	spin_unlock_bh(&ip6_fl_lock);
+	rcu_read_unlock();
 	return NULL;
 }
 
@@ -263,17 +266,17 @@ struct ip6_flowlabel *__fl6_sock_lookup(struct sock *sk, __be32 label)
 
 	label &= IPV6_FLOWLABEL_MASK;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	for_each_sk_fl_rcu(np, sfl) {
 		struct ip6_flowlabel *fl = sfl->fl;
 
 		if (fl->label == label && atomic_inc_not_zero(&fl->users)) {
 			fl->lastuse = jiffies;
-			rcu_read_unlock_bh();
+			rcu_read_unlock();
 			return fl;
 		}
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 	return NULL;
 }
 EXPORT_SYMBOL_GPL(__fl6_sock_lookup);
@@ -475,10 +478,10 @@ static int mem_check(struct sock *sk)
 	if (room > FL_MAX_SIZE - FL_MAX_PER_SOCK)
 		return 0;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	for_each_sk_fl_rcu(np, sfl)
 		count++;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	if (room <= 0 ||
 	    ((count >= FL_MAX_PER_SOCK ||
@@ -515,7 +518,7 @@ int ipv6_flowlabel_opt_get(struct sock *sk, struct in6_flowlabel_req *freq,
 		return 0;
 	}
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 
 	for_each_sk_fl_rcu(np, sfl) {
 		if (sfl->fl->label == (np->flow_label & IPV6_FLOWLABEL_MASK)) {
@@ -527,11 +530,11 @@ int ipv6_flowlabel_opt_get(struct sock *sk, struct in6_flowlabel_req *freq,
 			freq->flr_linger = sfl->fl->linger / HZ;
 
 			spin_unlock_bh(&ip6_fl_lock);
-			rcu_read_unlock_bh();
+			rcu_read_unlock();
 			return 0;
 		}
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return -ENOENT;
 }
@@ -581,16 +584,16 @@ static int ipv6_flowlabel_renew(struct sock *sk, struct in6_flowlabel_req *freq)
 	struct ipv6_fl_socklist *sfl;
 	int err;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	for_each_sk_fl_rcu(np, sfl) {
 		if (sfl->fl->label == freq->flr_label) {
 			err = fl6_renew(sfl->fl, freq->flr_linger,
 					freq->flr_expires);
-			rcu_read_unlock_bh();
+			rcu_read_unlock();
 			return err;
 		}
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	if (freq->flr_share == IPV6_FL_S_NONE &&
 	    ns_capable(net->user_ns, CAP_NET_ADMIN)) {
@@ -641,11 +644,11 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq,
 
 	if (freq->flr_label) {
 		err = -EEXIST;
-		rcu_read_lock_bh();
+		rcu_read_lock();
 		for_each_sk_fl_rcu(np, sfl) {
 			if (sfl->fl->label == freq->flr_label) {
 				if (freq->flr_flags & IPV6_FL_F_EXCL) {
-					rcu_read_unlock_bh();
+					rcu_read_unlock();
 					goto done;
 				}
 				fl1 = sfl->fl;
@@ -654,7 +657,7 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq,
 				break;
 			}
 		}
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 
 		if (!fl1)
 			fl1 = fl_lookup(net, freq->flr_label);
@@ -809,7 +812,7 @@ static void *ip6fl_seq_start(struct seq_file *seq, loff_t *pos)
 
 	state->pid_ns = proc_pid_ns(file_inode(seq->file)->i_sb);
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	return *pos ? ip6fl_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 }
 
@@ -828,7 +831,7 @@ static void *ip6fl_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 static void ip6fl_seq_stop(struct seq_file *seq, void *v)
 	__releases(RCU)
 {
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 }
 
 static int ip6fl_seq_show(struct seq_file *seq, void *v)
-- 
cgit v1.2.3-70-g09d2


From 09eed1192cec1755967f2af8394207acdde579a1 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 21 Mar 2023 04:01:14 +0000
Subject: neighbour: switch to standard rcu, instead of rcu_bh

rcu_bh is no longer a win, especially for objects freed
with standard call_rcu().

Switch neighbour code to no longer disable BH when not necessary.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/arp.h        |  8 +++---
 include/net/ndisc.h      | 12 ++++-----
 include/net/neighbour.h  |  6 ++---
 include/net/nexthop.h    |  6 ++---
 net/core/filter.c        | 16 +++++++-----
 net/core/neighbour.c     | 64 ++++++++++++++++++++++++------------------------
 net/ipv4/fib_semantics.c |  4 +--
 net/ipv4/ip_output.c     |  6 ++---
 net/ipv4/nexthop.c       |  8 +++---
 net/ipv4/route.c         |  4 +--
 net/ipv6/addrconf.c      | 14 +++++------
 net/ipv6/ip6_output.c    | 10 ++++----
 net/ipv6/route.c         | 12 ++++-----
 13 files changed, 87 insertions(+), 83 deletions(-)

diff --git a/include/net/arp.h b/include/net/arp.h
index d7ef4ec71dfe..e8747e0713c7 100644
--- a/include/net/arp.h
+++ b/include/net/arp.h
@@ -38,11 +38,11 @@ static inline struct neighbour *__ipv4_neigh_lookup(struct net_device *dev, u32
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv4_neigh_lookup_noref(dev, key);
 	if (n && !refcount_inc_not_zero(&n->refcnt))
 		n = NULL;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return n;
 }
@@ -51,10 +51,10 @@ static inline void __ipv4_confirm_neigh(struct net_device *dev, u32 key)
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv4_neigh_lookup_noref(dev, key);
 	neigh_confirm(n);
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 }
 
 void arp_init(void);
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 07e5168cdaf9..52eae0943433 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -395,11 +395,11 @@ static inline struct neighbour *__ipv6_neigh_lookup(struct net_device *dev, cons
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref(dev, pkey);
 	if (n && !refcount_inc_not_zero(&n->refcnt))
 		n = NULL;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return n;
 }
@@ -409,10 +409,10 @@ static inline void __ipv6_confirm_neigh(struct net_device *dev,
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref(dev, pkey);
 	neigh_confirm(n);
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 }
 
 static inline void __ipv6_confirm_neigh_stub(struct net_device *dev,
@@ -420,10 +420,10 @@ static inline void __ipv6_confirm_neigh_stub(struct net_device *dev,
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref_stub(dev, pkey);
 	neigh_confirm(n);
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 }
 
 /* uses ipv6_stub and is meant for use outside of IPv6 core */
diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index c8d39bba2a0d..3fa5774bddac 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -299,14 +299,14 @@ static inline struct neighbour *___neigh_lookup_noref(
 	const void *pkey,
 	struct net_device *dev)
 {
-	struct neigh_hash_table *nht = rcu_dereference_bh(tbl->nht);
+	struct neigh_hash_table *nht = rcu_dereference(tbl->nht);
 	struct neighbour *n;
 	u32 hash_val;
 
 	hash_val = hash(pkey, dev, nht->hash_rnd) >> (32 - nht->hash_shift);
-	for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]);
+	for (n = rcu_dereference(nht->hash_buckets[hash_val]);
 	     n != NULL;
-	     n = rcu_dereference_bh(n->next)) {
+	     n = rcu_dereference(n->next)) {
 		if (n->dev == dev && key_eq(n, pkey))
 			return n;
 	}
diff --git a/include/net/nexthop.h b/include/net/nexthop.h
index 28085b995ddc..9fa291a04621 100644
--- a/include/net/nexthop.h
+++ b/include/net/nexthop.h
@@ -498,7 +498,7 @@ static inline struct fib6_nh *nexthop_fib6_nh(struct nexthop *nh)
 }
 
 /* Variant of nexthop_fib6_nh().
- * Caller should either hold rcu_read_lock_bh(), or RTNL.
+ * Caller should either hold rcu_read_lock(), or RTNL.
  */
 static inline struct fib6_nh *nexthop_fib6_nh_bh(struct nexthop *nh)
 {
@@ -507,13 +507,13 @@ static inline struct fib6_nh *nexthop_fib6_nh_bh(struct nexthop *nh)
 	if (nh->is_group) {
 		struct nh_group *nh_grp;
 
-		nh_grp = rcu_dereference_bh_rtnl(nh->nh_grp);
+		nh_grp = rcu_dereference_rtnl(nh->nh_grp);
 		nh = nexthop_mpath_select(nh_grp, 0);
 		if (!nh)
 			return NULL;
 	}
 
-	nhi = rcu_dereference_bh_rtnl(nh->nh_info);
+	nhi = rcu_dereference_rtnl(nh->nh_info);
 	if (nhi->family == AF_INET6)
 		return &nhi->fib6_nh;
 
diff --git a/net/core/filter.c b/net/core/filter.c
index d052fac28d02..a8c8fd96c822 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2204,7 +2204,7 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb,
 			return -ENOMEM;
 	}
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	if (!nh) {
 		dst = skb_dst(skb);
 		nexthop = rt6_nexthop(container_of(dst, struct rt6_info, dst),
@@ -2217,10 +2217,12 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb,
 		int ret;
 
 		sock_confirm_neigh(skb, neigh);
+		local_bh_disable();
 		dev_xmit_recursion_inc();
 		ret = neigh_output(neigh, skb, false);
 		dev_xmit_recursion_dec();
-		rcu_read_unlock_bh();
+		local_bh_enable();
+		rcu_read_unlock();
 		return ret;
 	}
 	rcu_read_unlock_bh();
@@ -2302,7 +2304,7 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb,
 			return -ENOMEM;
 	}
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	if (!nh) {
 		struct dst_entry *dst = skb_dst(skb);
 		struct rtable *rt = container_of(dst, struct rtable, dst);
@@ -2314,7 +2316,7 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb,
 	} else if (nh->nh_family == AF_INET) {
 		neigh = ip_neigh_gw4(dev, nh->ipv4_nh);
 	} else {
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 		goto out_drop;
 	}
 
@@ -2322,13 +2324,15 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb,
 		int ret;
 
 		sock_confirm_neigh(skb, neigh);
+		local_bh_disable();
 		dev_xmit_recursion_inc();
 		ret = neigh_output(neigh, skb, is_v6gw);
 		dev_xmit_recursion_dec();
-		rcu_read_unlock_bh();
+		local_bh_enable();
+		rcu_read_unlock();
 		return ret;
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 out_drop:
 	kfree_skb(skb);
 	return -ENETDOWN;
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 90d399b3f980..ddd0f32de20e 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -614,7 +614,7 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
 
 	NEIGH_CACHE_STAT_INC(tbl, lookups);
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __neigh_lookup_noref(tbl, pkey, dev);
 	if (n) {
 		if (!refcount_inc_not_zero(&n->refcnt))
@@ -622,7 +622,7 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
 		NEIGH_CACHE_STAT_INC(tbl, hits);
 	}
 
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 	return n;
 }
 EXPORT_SYMBOL(neigh_lookup);
@@ -2184,11 +2184,11 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl,
 			.ndtc_proxy_qlen	= tbl->proxy_queue.qlen,
 		};
 
-		rcu_read_lock_bh();
-		nht = rcu_dereference_bh(tbl->nht);
+		rcu_read_lock();
+		nht = rcu_dereference(tbl->nht);
 		ndc.ndtc_hash_rnd = nht->hash_rnd[0];
 		ndc.ndtc_hash_mask = ((1 << nht->hash_shift) - 1);
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 
 		if (nla_put(skb, NDTA_CONFIG, sizeof(ndc), &ndc))
 			goto nla_put_failure;
@@ -2703,15 +2703,15 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
 	if (filter->dev_idx || filter->master_idx)
 		flags |= NLM_F_DUMP_FILTERED;
 
-	rcu_read_lock_bh();
-	nht = rcu_dereference_bh(tbl->nht);
+	rcu_read_lock();
+	nht = rcu_dereference(tbl->nht);
 
 	for (h = s_h; h < (1 << nht->hash_shift); h++) {
 		if (h > s_h)
 			s_idx = 0;
-		for (n = rcu_dereference_bh(nht->hash_buckets[h]), idx = 0;
+		for (n = rcu_dereference(nht->hash_buckets[h]), idx = 0;
 		     n != NULL;
-		     n = rcu_dereference_bh(n->next)) {
+		     n = rcu_dereference(n->next)) {
 			if (idx < s_idx || !net_eq(dev_net(n->dev), net))
 				goto next;
 			if (neigh_ifindex_filtered(n->dev, filter->dev_idx) ||
@@ -2730,7 +2730,7 @@ next:
 	}
 	rc = skb->len;
 out:
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 	cb->args[1] = h;
 	cb->args[2] = idx;
 	return rc;
@@ -3075,20 +3075,20 @@ void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void
 	int chain;
 	struct neigh_hash_table *nht;
 
-	rcu_read_lock_bh();
-	nht = rcu_dereference_bh(tbl->nht);
+	rcu_read_lock();
+	nht = rcu_dereference(tbl->nht);
 
-	read_lock(&tbl->lock); /* avoid resizes */
+	read_lock_bh(&tbl->lock); /* avoid resizes */
 	for (chain = 0; chain < (1 << nht->hash_shift); chain++) {
 		struct neighbour *n;
 
-		for (n = rcu_dereference_bh(nht->hash_buckets[chain]);
+		for (n = rcu_dereference(nht->hash_buckets[chain]);
 		     n != NULL;
-		     n = rcu_dereference_bh(n->next))
+		     n = rcu_dereference(n->next))
 			cb(n, cookie);
 	}
-	read_unlock(&tbl->lock);
-	rcu_read_unlock_bh();
+	read_unlock_bh(&tbl->lock);
+	rcu_read_unlock();
 }
 EXPORT_SYMBOL(neigh_for_each);
 
@@ -3138,7 +3138,7 @@ int neigh_xmit(int index, struct net_device *dev,
 		tbl = neigh_tables[index];
 		if (!tbl)
 			goto out;
-		rcu_read_lock_bh();
+		rcu_read_lock();
 		if (index == NEIGH_ARP_TABLE) {
 			u32 key = *((u32 *)addr);
 
@@ -3150,11 +3150,11 @@ int neigh_xmit(int index, struct net_device *dev,
 			neigh = __neigh_create(tbl, addr, dev, false);
 		err = PTR_ERR(neigh);
 		if (IS_ERR(neigh)) {
-			rcu_read_unlock_bh();
+			rcu_read_unlock();
 			goto out_kfree_skb;
 		}
 		err = neigh->output(neigh, skb);
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 	}
 	else if (index == NEIGH_LINK_TABLE) {
 		err = dev_hard_header(skb, dev, ntohs(skb->protocol),
@@ -3183,7 +3183,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
 
 	state->flags &= ~NEIGH_SEQ_IS_PNEIGH;
 	for (bucket = 0; bucket < (1 << nht->hash_shift); bucket++) {
-		n = rcu_dereference_bh(nht->hash_buckets[bucket]);
+		n = rcu_dereference(nht->hash_buckets[bucket]);
 
 		while (n) {
 			if (!net_eq(dev_net(n->dev), net))
@@ -3201,7 +3201,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
 			if (READ_ONCE(n->nud_state) & ~NUD_NOARP)
 				break;
 next:
-			n = rcu_dereference_bh(n->next);
+			n = rcu_dereference(n->next);
 		}
 
 		if (n)
@@ -3225,7 +3225,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 		if (v)
 			return n;
 	}
-	n = rcu_dereference_bh(n->next);
+	n = rcu_dereference(n->next);
 
 	while (1) {
 		while (n) {
@@ -3243,7 +3243,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 			if (READ_ONCE(n->nud_state) & ~NUD_NOARP)
 				break;
 next:
-			n = rcu_dereference_bh(n->next);
+			n = rcu_dereference(n->next);
 		}
 
 		if (n)
@@ -3252,7 +3252,7 @@ next:
 		if (++state->bucket >= (1 << nht->hash_shift))
 			break;
 
-		n = rcu_dereference_bh(nht->hash_buckets[state->bucket]);
+		n = rcu_dereference(nht->hash_buckets[state->bucket]);
 	}
 
 	if (n && pos)
@@ -3354,7 +3354,7 @@ static void *neigh_get_idx_any(struct seq_file *seq, loff_t *pos)
 
 void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl, unsigned int neigh_seq_flags)
 	__acquires(tbl->lock)
-	__acquires(rcu_bh)
+	__acquires(rcu)
 {
 	struct neigh_seq_state *state = seq->private;
 
@@ -3362,9 +3362,9 @@ void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl
 	state->bucket = 0;
 	state->flags = (neigh_seq_flags & ~NEIGH_SEQ_IS_PNEIGH);
 
-	rcu_read_lock_bh();
-	state->nht = rcu_dereference_bh(tbl->nht);
-	read_lock(&tbl->lock);
+	rcu_read_lock();
+	state->nht = rcu_dereference(tbl->nht);
+	read_lock_bh(&tbl->lock);
 
 	return *pos ? neigh_get_idx_any(seq, pos) : SEQ_START_TOKEN;
 }
@@ -3399,13 +3399,13 @@ EXPORT_SYMBOL(neigh_seq_next);
 
 void neigh_seq_stop(struct seq_file *seq, void *v)
 	__releases(tbl->lock)
-	__releases(rcu_bh)
+	__releases(rcu)
 {
 	struct neigh_seq_state *state = seq->private;
 	struct neigh_table *tbl = state->tbl;
 
-	read_unlock(&tbl->lock);
-	rcu_read_unlock_bh();
+	read_unlock_bh(&tbl->lock);
+	rcu_read_unlock();
 }
 EXPORT_SYMBOL(neigh_seq_stop);
 
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 574ff450c4d2..65ba18a91865 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -2191,7 +2191,7 @@ static bool fib_good_nh(const struct fib_nh *nh)
 	if (nh->fib_nh_scope == RT_SCOPE_LINK) {
 		struct neighbour *n;
 
-		rcu_read_lock_bh();
+		rcu_read_lock();
 
 		if (likely(nh->fib_nh_gw_family == AF_INET))
 			n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev,
@@ -2204,7 +2204,7 @@ static bool fib_good_nh(const struct fib_nh *nh)
 		if (n)
 			state = READ_ONCE(n->nud_state);
 
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 	}
 
 	return !!(state & NUD_VALID);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index cb04dbad9ea4..22a90a9392eb 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -219,7 +219,7 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s
 			return res;
 	}
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	neigh = ip_neigh_for_gw(rt, skb, &is_v6gw);
 	if (!IS_ERR(neigh)) {
 		int res;
@@ -227,10 +227,10 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s
 		sock_confirm_neigh(skb, neigh);
 		/* if crossing protocols, can not use the cached header */
 		res = neigh_output(neigh, skb, is_v6gw);
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 		return res;
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	net_dbg_ratelimited("%s: No header cache and no neighbour!\n",
 			    __func__);
diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index e28a99f1996b..f95142e56da0 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -1124,13 +1124,13 @@ static bool ipv6_good_nh(const struct fib6_nh *nh)
 	int state = NUD_REACHABLE;
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 
 	n = __ipv6_neigh_lookup_noref_stub(nh->fib_nh_dev, &nh->fib_nh_gw6);
 	if (n)
 		state = READ_ONCE(n->nud_state);
 
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return !!(state & NUD_VALID);
 }
@@ -1140,14 +1140,14 @@ static bool ipv4_good_nh(const struct fib_nh *nh)
 	int state = NUD_REACHABLE;
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 
 	n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev,
 				      (__force u32)nh->fib_nh_gw4);
 	if (n)
 		state = READ_ONCE(n->nud_state);
 
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return !!(state & NUD_VALID);
 }
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 232009d216c4..6a0a0bb452e9 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -408,7 +408,7 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
 	struct net_device *dev = dst->dev;
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 
 	if (likely(rt->rt_gw_family == AF_INET)) {
 		n = ip_neigh_gw4(dev, rt->rt_gw4);
@@ -424,7 +424,7 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
 	if (!IS_ERR(n) && !refcount_inc_not_zero(&n->refcnt))
 		n = NULL;
 
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return n;
 }
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index faa47f9ea73a..31e0097878c5 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1034,7 +1034,7 @@ static int ipv6_add_addr_hash(struct net_device *dev, struct inet6_ifaddr *ifa)
 	unsigned int hash = inet6_addr_hash(net, &ifa->addr);
 	int err = 0;
 
-	spin_lock(&net->ipv6.addrconf_hash_lock);
+	spin_lock_bh(&net->ipv6.addrconf_hash_lock);
 
 	/* Ignore adding duplicate addresses on an interface */
 	if (ipv6_chk_same_addr(net, &ifa->addr, dev, hash)) {
@@ -1044,7 +1044,7 @@ static int ipv6_add_addr_hash(struct net_device *dev, struct inet6_ifaddr *ifa)
 		hlist_add_head_rcu(&ifa->addr_lst, &net->ipv6.inet6_addr_lst[hash]);
 	}
 
-	spin_unlock(&net->ipv6.addrconf_hash_lock);
+	spin_unlock_bh(&net->ipv6.addrconf_hash_lock);
 
 	return err;
 }
@@ -1139,15 +1139,15 @@ ipv6_add_addr(struct inet6_dev *idev, struct ifa6_config *cfg,
 	/* For caller */
 	refcount_set(&ifa->refcnt, 1);
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 
 	err = ipv6_add_addr_hash(idev->dev, ifa);
 	if (err < 0) {
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 		goto out;
 	}
 
-	write_lock(&idev->lock);
+	write_lock_bh(&idev->lock);
 
 	/* Add to inet6_dev unicast addr list. */
 	ipv6_link_dev_addr(idev, ifa);
@@ -1158,9 +1158,9 @@ ipv6_add_addr(struct inet6_dev *idev, struct ifa6_config *cfg,
 	}
 
 	in6_ifa_hold(ifa);
-	write_unlock(&idev->lock);
+	write_unlock_bh(&idev->lock);
 
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	inet6addr_notifier_call_chain(NETDEV_UP, ifa);
 out:
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index e5ed39a3c65f..0b6140f0179d 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -116,7 +116,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *
 			return res;
 	}
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	nexthop = rt6_nexthop((struct rt6_info *)dst, daddr);
 	neigh = __ipv6_neigh_lookup_noref(dev, nexthop);
 
@@ -124,7 +124,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *
 		if (unlikely(!neigh))
 			neigh = __neigh_create(&nd_tbl, nexthop, dev, false);
 		if (IS_ERR(neigh)) {
-			rcu_read_unlock_bh();
+			rcu_read_unlock();
 			IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTNOROUTES);
 			kfree_skb_reason(skb, SKB_DROP_REASON_NEIGH_CREATEFAIL);
 			return -EINVAL;
@@ -132,7 +132,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *
 	}
 	sock_confirm_neigh(skb, neigh);
 	ret = neigh_output(neigh, skb, false);
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 	return ret;
 }
 
@@ -1150,11 +1150,11 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk,
 	 * dst entry of the nexthop router
 	 */
 	rt = (struct rt6_info *) *dst;
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref(rt->dst.dev,
 				      rt6_nexthop(rt, &fl6->daddr));
 	err = n && !(READ_ONCE(n->nud_state) & NUD_VALID) ? -EINVAL : 0;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	if (err) {
 		struct inet6_ifaddr *ifp;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index e829bd880384..244df77fac87 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -633,7 +633,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh)
 
 	nh_gw = &fib6_nh->fib_nh_gw6;
 	dev = fib6_nh->fib_nh_dev;
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	last_probe = READ_ONCE(fib6_nh->last_probe);
 	idev = __in6_dev_get(dev);
 	neigh = __ipv6_neigh_lookup_noref(dev, nh_gw);
@@ -641,7 +641,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh)
 		if (READ_ONCE(neigh->nud_state) & NUD_VALID)
 			goto out;
 
-		write_lock(&neigh->lock);
+		write_lock_bh(&neigh->lock);
 		if (!(neigh->nud_state & NUD_VALID) &&
 		    time_after(jiffies,
 			       neigh->updated + idev->cnf.rtr_probe_interval)) {
@@ -649,7 +649,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh)
 			if (work)
 				__neigh_set_probe_once(neigh);
 		}
-		write_unlock(&neigh->lock);
+		write_unlock_bh(&neigh->lock);
 	} else if (time_after(jiffies, last_probe +
 				       idev->cnf.rtr_probe_interval)) {
 		work = kmalloc(sizeof(*work), GFP_ATOMIC);
@@ -667,7 +667,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh)
 	}
 
 out:
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 }
 #else
 static inline void rt6_probe(struct fib6_nh *fib6_nh)
@@ -683,7 +683,7 @@ static enum rt6_nud_state rt6_check_neigh(const struct fib6_nh *fib6_nh)
 	enum rt6_nud_state ret = RT6_NUD_FAIL_HARD;
 	struct neighbour *neigh;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	neigh = __ipv6_neigh_lookup_noref(fib6_nh->fib_nh_dev,
 					  &fib6_nh->fib_nh_gw6);
 	if (neigh) {
@@ -701,7 +701,7 @@ static enum rt6_nud_state rt6_check_neigh(const struct fib6_nh *fib6_nh)
 		ret = IS_ENABLED(CONFIG_IPV6_ROUTER_PREF) ?
 		      RT6_NUD_SUCCEED : RT6_NUD_FAIL_DO_RR;
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return ret;
 }
-- 
cgit v1.2.3-70-g09d2


From fe602c87df1b6927562f4ee61edd851bb9578a49 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 21 Mar 2023 04:01:15 +0000
Subject: net: remove rcu_dereference_bh_rtnl()

This helper is no longer used in the tree.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/rtnetlink.h | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 92ad75549e9c..f0c87baaf6c0 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -61,16 +61,6 @@ static inline bool lockdep_rtnl_is_held(void)
 #define rcu_dereference_rtnl(p)					\
 	rcu_dereference_check(p, lockdep_rtnl_is_held())
 
-/**
- * rcu_dereference_bh_rtnl - rcu_dereference_bh with debug checking
- * @p: The pointer to read, prior to dereference
- *
- * Do an rcu_dereference_bh(p), but check caller either holds rcu_read_lock_bh()
- * or RTNL. Note : Please prefer rtnl_dereference() or rcu_dereference_bh()
- */
-#define rcu_dereference_bh_rtnl(p)				\
-	rcu_dereference_bh_check(p, lockdep_rtnl_is_held())
-
 /**
  * rtnl_dereference - fetch RCU pointer when updates are prevented by RTNL
  * @p: The pointer to read, prior to dereferencing
-- 
cgit v1.2.3-70-g09d2


From 56c874f7dbcab2ab5cf8055d46a2ef36dec3d664 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 20 Mar 2023 21:41:59 -0700
Subject: tools: ynl: skip the explicit op array size when not needed

Jiri suggests it reads more naturally to skip the explicit
array size when possible. When we export the symbol we want
to make sure that the size is right but for statics its
not needed.

Link: https://lore.kernel.org/r/20230321044159.1031040-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/netdev-genl-gen.c | 2 +-
 tools/net/ynl/ynl-gen-c.py | 4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/core/netdev-genl-gen.c b/net/core/netdev-genl-gen.c
index 3abab70d66dd..de17ca2f7dbf 100644
--- a/net/core/netdev-genl-gen.c
+++ b/net/core/netdev-genl-gen.c
@@ -16,7 +16,7 @@ static const struct nla_policy netdev_dev_get_nl_policy[NETDEV_A_DEV_IFINDEX + 1
 };
 
 /* Ops table for netdev */
-static const struct genl_split_ops netdev_nl_ops[2] = {
+static const struct genl_split_ops netdev_nl_ops[] = {
 	{
 		.cmd		= NETDEV_CMD_DEV_GET,
 		.doit		= netdev_nl_dev_get_doit,
diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index c16671a02621..972b87c7aaaf 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -1696,7 +1696,9 @@ def print_kernel_op_table_fwd(family, cw, terminate):
                          'split': 'genl_split_ops'}
         struct_type = pol_to_struct[family.kernel_policy]
 
-        if family.kernel_policy == 'split':
+        if not exported:
+            cnt = ""
+        elif family.kernel_policy == 'split':
             cnt = 0
             for op in family.ops.values():
                 if 'do' in op:
-- 
cgit v1.2.3-70-g09d2


From 68e93ac5a31d4975b25f819b2dfe914c72abc3bb Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Wed, 15 Mar 2023 12:24:43 +0200
Subject: wifi: ath11k: fix BUFFER_DONE read on monitor ring rx  buffer

Perform dma_sync_single_for_cpu() on monitor ring rx buffer before
reading BUFFER_DONE tag and do dma_unmap_single() only after device
had set BUFFER_DONE tag to the buffer.

Also when BUFFER_DONE tag is not set, allow the buffer to get read
next time without freeing skb.

This helps to fix AP+Monitor VAP with flood traffic scenario to see
monitor ring rx buffer overrun missing BUFFER_DONE tag to be set.

Also remove redundant rx dma buf free performed on DP
rx_mon_status_refill_ring.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Sathishkumar Muruganandam <quic_murugana@quicinc.com>
Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230309164434.32660-1-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 57 +++++++++++++--------------------
 1 file changed, 23 insertions(+), 34 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index b65a84a88264..99859b59138e 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -435,7 +435,6 @@ fail_free_skb:
 static int ath11k_dp_rxdma_buf_ring_free(struct ath11k *ar,
 					 struct dp_rxdma_ring *rx_ring)
 {
-	struct ath11k_pdev_dp *dp = &ar->dp;
 	struct sk_buff *skb;
 	int buf_id;
 
@@ -453,28 +452,6 @@ static int ath11k_dp_rxdma_buf_ring_free(struct ath11k *ar,
 	idr_destroy(&rx_ring->bufs_idr);
 	spin_unlock_bh(&rx_ring->idr_lock);
 
-	/* if rxdma1_enable is false, mon_status_refill_ring
-	 * isn't setup, so don't clean.
-	 */
-	if (!ar->ab->hw_params.rxdma1_enable)
-		return 0;
-
-	rx_ring = &dp->rx_mon_status_refill_ring[0];
-
-	spin_lock_bh(&rx_ring->idr_lock);
-	idr_for_each_entry(&rx_ring->bufs_idr, skb, buf_id) {
-		idr_remove(&rx_ring->bufs_idr, buf_id);
-		/* XXX: Understand where internal driver does this dma_unmap
-		 * of rxdma_buffer.
-		 */
-		dma_unmap_single(ar->ab->dev, ATH11K_SKB_RXCB(skb)->paddr,
-				 skb->len + skb_tailroom(skb), DMA_BIDIRECTIONAL);
-		dev_kfree_skb_any(skb);
-	}
-
-	idr_destroy(&rx_ring->bufs_idr);
-	spin_unlock_bh(&rx_ring->idr_lock);
-
 	return 0;
 }
 
@@ -3029,39 +3006,51 @@ static int ath11k_dp_rx_reap_mon_status_ring(struct ath11k_base *ab, int mac_id,
 
 			spin_lock_bh(&rx_ring->idr_lock);
 			skb = idr_find(&rx_ring->bufs_idr, buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
+
 			if (!skb) {
 				ath11k_warn(ab, "rx monitor status with invalid buf_id %d\n",
 					    buf_id);
-				spin_unlock_bh(&rx_ring->idr_lock);
 				pmon->buf_state = DP_MON_STATUS_REPLINISH;
 				goto move_next;
 			}
 
-			idr_remove(&rx_ring->bufs_idr, buf_id);
-			spin_unlock_bh(&rx_ring->idr_lock);
-
 			rxcb = ATH11K_SKB_RXCB(skb);
 
-			dma_unmap_single(ab->dev, rxcb->paddr,
-					 skb->len + skb_tailroom(skb),
-					 DMA_FROM_DEVICE);
+			dma_sync_single_for_cpu(ab->dev, rxcb->paddr,
+						skb->len + skb_tailroom(skb),
+						DMA_FROM_DEVICE);
 
 			tlv = (struct hal_tlv_hdr *)skb->data;
 			if (FIELD_GET(HAL_TLV_HDR_TAG, tlv->tl) !=
 					HAL_RX_STATUS_BUFFER_DONE) {
-				ath11k_warn(ab, "mon status DONE not set %lx\n",
+				ath11k_warn(ab, "mon status DONE not set %lx, buf_id %d\n",
 					    FIELD_GET(HAL_TLV_HDR_TAG,
-						      tlv->tl));
-				dev_kfree_skb_any(skb);
+						      tlv->tl), buf_id);
+				/* If done status is missing, hold onto status
+				 * ring until status is done for this status
+				 * ring buffer.
+				 * Keep HP in mon_status_ring unchanged,
+				 * and break from here.
+				 * Check status for same buffer for next time
+				 */
 				pmon->buf_state = DP_MON_STATUS_NO_DMA;
-				goto move_next;
+				break;
 			}
 
+			spin_lock_bh(&rx_ring->idr_lock);
+			idr_remove(&rx_ring->bufs_idr, buf_id);
+			spin_unlock_bh(&rx_ring->idr_lock);
 			if (ab->hw_params.full_monitor_mode) {
 				ath11k_dp_rx_mon_update_status_buf_state(pmon, tlv);
 				if (paddr == pmon->mon_status_paddr)
 					pmon->buf_state = DP_MON_STATUS_MATCH;
 			}
+
+			dma_unmap_single(ab->dev, rxcb->paddr,
+					 skb->len + skb_tailroom(skb),
+					 DMA_FROM_DEVICE);
+
 			__skb_queue_tail(skb_list, skb);
 		} else {
 			pmon->buf_state = DP_MON_STATUS_REPLINISH;
-- 
cgit v1.2.3-70-g09d2


From 4d1f51551777b4a8e6f5503f4d64625586de2c6d Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Mon, 20 Mar 2023 20:43:29 +0300
Subject: virtio/vsock: check transport before skb allocation

Pointer to transport could be checked before allocation of skbuff, thus
there is no need to free skbuff when this pointer is NULL.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Bobby Eshleman <bobby.eshleman@bytedance.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://lore.kernel.org/r/08d61bef-0c11-c7f9-9266-cb2109070314@sberdevices.ru
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/vmw_vsock/virtio_transport_common.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index 6564192e7f20..957cdc01c8e8 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -850,6 +850,9 @@ static int virtio_transport_reset_no_sock(const struct virtio_transport *t,
 	if (le16_to_cpu(hdr->op) == VIRTIO_VSOCK_OP_RST)
 		return 0;
 
+	if (!t)
+		return -ENOTCONN;
+
 	reply = virtio_transport_alloc_skb(&info, 0,
 					   le64_to_cpu(hdr->dst_cid),
 					   le32_to_cpu(hdr->dst_port),
@@ -858,11 +861,6 @@ static int virtio_transport_reset_no_sock(const struct virtio_transport *t,
 	if (!reply)
 		return -ENOMEM;
 
-	if (!t) {
-		kfree_skb(reply);
-		return -ENOTCONN;
-	}
-
 	return t->send_pkt(reply);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 960412bee0ea75f6b3c2dca4a3535795ee84c47a Mon Sep 17 00:00:00 2001
From: Rajat Soni <quic_rajson@quicinc.com>
Date: Wed, 15 Mar 2023 14:36:32 +0530
Subject: wifi: ath12k: fix memory leak in ath12k_qmi_driver_event_work()

Currently the buffer pointed by event is not freed in case
ATH12K_FLAG_UNREGISTERING bit is set, this causes memory leak.

Add a goto skip instead of return, to ensure event and all the
list entries are freed properly.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Rajat Soni <quic_rajson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230315090632.15065-1-quic_rajson@quicinc.com
---
 drivers/net/wireless/ath/ath12k/qmi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/qmi.c b/drivers/net/wireless/ath/ath12k/qmi.c
index 979a63f2e2ab..03ba245fbee9 100644
--- a/drivers/net/wireless/ath/ath12k/qmi.c
+++ b/drivers/net/wireless/ath/ath12k/qmi.c
@@ -2991,7 +2991,7 @@ static void ath12k_qmi_driver_event_work(struct work_struct *work)
 		spin_unlock(&qmi->event_lock);
 
 		if (test_bit(ATH12K_FLAG_UNREGISTERING, &ab->dev_flags))
-			return;
+			goto skip;
 
 		switch (event->type) {
 		case ATH12K_QMI_EVENT_SERVER_ARRIVE:
@@ -3032,6 +3032,8 @@ static void ath12k_qmi_driver_event_work(struct work_struct *work)
 			ath12k_warn(ab, "invalid event type: %d", event->type);
 			break;
 		}
+
+skip:
 		kfree(event);
 		spin_lock(&qmi->event_lock);
 	}
-- 
cgit v1.2.3-70-g09d2


From cbc0008c9b39ea95b5125ab77fb409d9a32a42e1 Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Wed, 15 Mar 2023 17:02:02 +0530
Subject: wifi: ath12k: fix firmware assert during channel switch for peer sta

Currently, during change in bandwidth for peer sta, host sends the
new value of channel width via WMI_PEER_CHWIDTH set peer param command
alone. This can lead to firmware assert in some scenario since before
the command, firmware was having value of channel width and its
corresponding phymode. After the command, host tries to set the new
value of channel width alone which can become incompatible when compared
with its phymode.

For example:

Bandwidth Upgrade
~~~~~~~~~~~~~~~~~~
After association, sta is in 40 MHz bandwidth in 11ax-HE40 phymode.
After bandwidth upgrades, sta moves to 80 MHz but as per phymode,
max bandwidth is still 40 MHz. Hence, firmware assert is seen.
So in this case first phymode should be moved to 11ax-HE80
followed by bandwidth change.

Bandwidth Downgrade
~~~~~~~~~~~~~~~~~~
Similarly, reverse of above is also possible when sta is in 40 MHz
bandwidth in 11ax-HE40 phymode. Bandwidth should be changed to 20 MHz
and if host sends phymode first then, phymode will become 11ax-HE20
and will be incompatible with bandwidth value and hence firmware
assert will be seen. Hence, in this case first channel width
should be set followed by phymode.

Fix this issue by sending WMI set peer param command for phymode as
well as bandwidth based on the type of bandwidth change i.e upgrade
or downgrade.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Signed-off-by: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230315113202.8774-1-quic_aarasahu@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h |   1 +
 drivers/net/wireless/ath/ath12k/mac.c  | 117 +++++++++++++++++++++++++--------
 2 files changed, 89 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index dffa687ee40e..9439052a652e 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -395,6 +395,7 @@ struct ath12k_sta {
 	u8 rssi_comb;
 	struct ath12k_rx_peer_stats *rx_stats;
 	struct ath12k_wbm_tx_stats *wbm_tx_stats;
+	u32 bw_prev;
 };
 
 #define ATH12K_MIN_5G_FREQ 4150
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index bf7e5b6977b2..ee792822b411 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -3220,10 +3220,11 @@ static void ath12k_sta_rc_update_wk(struct work_struct *wk)
 	enum nl80211_band band;
 	const u8 *ht_mcs_mask;
 	const u16 *vht_mcs_mask;
-	u32 changed, bw, nss, smps;
+	u32 changed, bw, nss, smps, bw_prev;
 	int err, num_vht_rates;
 	const struct cfg80211_bitrate_mask *mask;
 	struct ath12k_wmi_peer_assoc_arg peer_arg;
+	enum wmi_phy_mode peer_phymode;
 
 	arsta = container_of(wk, struct ath12k_sta, update_wk);
 	sta = container_of((void *)arsta, struct ieee80211_sta, drv_priv);
@@ -3243,6 +3244,7 @@ static void ath12k_sta_rc_update_wk(struct work_struct *wk)
 	arsta->changed = 0;
 
 	bw = arsta->bw;
+	bw_prev = arsta->bw_prev;
 	nss = arsta->nss;
 	smps = arsta->smps;
 
@@ -3255,11 +3257,53 @@ static void ath12k_sta_rc_update_wk(struct work_struct *wk)
 			   ath12k_mac_max_vht_nss(vht_mcs_mask)));
 
 	if (changed & IEEE80211_RC_BW_CHANGED) {
-		err = ath12k_wmi_set_peer_param(ar, sta->addr, arvif->vdev_id,
-						WMI_PEER_CHWIDTH, bw);
-		if (err)
-			ath12k_warn(ar->ab, "failed to update STA %pM peer bw %d: %d\n",
-				    sta->addr, bw, err);
+		ath12k_peer_assoc_h_phymode(ar, arvif->vif, sta, &peer_arg);
+		peer_phymode = peer_arg.peer_phymode;
+
+		if (bw > bw_prev) {
+			/* Phymode shows maximum supported channel width, if we
+			 * upgrade bandwidth then due to sanity check of firmware,
+			 * we have to send WMI_PEER_PHYMODE followed by
+			 * WMI_PEER_CHWIDTH
+			 */
+			ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "mac bandwidth upgrade for sta %pM new %d old %d\n",
+				   sta->addr, bw, bw_prev);
+			err = ath12k_wmi_set_peer_param(ar, sta->addr,
+							arvif->vdev_id, WMI_PEER_PHYMODE,
+							peer_phymode);
+			if (err) {
+				ath12k_warn(ar->ab, "failed to update STA %pM to peer phymode %d: %d\n",
+					    sta->addr, peer_phymode, err);
+				goto err_rc_bw_changed;
+			}
+			err = ath12k_wmi_set_peer_param(ar, sta->addr,
+							arvif->vdev_id, WMI_PEER_CHWIDTH,
+							bw);
+			if (err)
+				ath12k_warn(ar->ab, "failed to update STA %pM to peer bandwidth %d: %d\n",
+					    sta->addr, bw, err);
+		} else {
+			/* When we downgrade bandwidth this will conflict with phymode
+			 * and cause to trigger firmware crash. In this case we send
+			 * WMI_PEER_CHWIDTH followed by WMI_PEER_PHYMODE
+			 */
+			ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "mac bandwidth downgrade for sta %pM new %d old %d\n",
+				   sta->addr, bw, bw_prev);
+			err = ath12k_wmi_set_peer_param(ar, sta->addr,
+							arvif->vdev_id, WMI_PEER_CHWIDTH,
+							bw);
+			if (err) {
+				ath12k_warn(ar->ab, "failed to update STA %pM peer to bandwidth %d: %d\n",
+					    sta->addr, bw, err);
+				goto err_rc_bw_changed;
+			}
+			err = ath12k_wmi_set_peer_param(ar, sta->addr,
+							arvif->vdev_id, WMI_PEER_PHYMODE,
+							peer_phymode);
+			if (err)
+				ath12k_warn(ar->ab, "failed to update STA %pM to peer phymode %d: %d\n",
+					    sta->addr, peer_phymode, err);
+		}
 	}
 
 	if (changed & IEEE80211_RC_NSS_CHANGED) {
@@ -3321,7 +3365,7 @@ static void ath12k_sta_rc_update_wk(struct work_struct *wk)
 					    sta->addr, arvif->vdev_id);
 		}
 	}
-
+err_rc_bw_changed:
 	mutex_unlock(&ar->conf_mutex);
 }
 
@@ -3433,6 +3477,34 @@ exit:
 	return ret;
 }
 
+static u32 ath12k_mac_ieee80211_sta_bw_to_wmi(struct ath12k *ar,
+					      struct ieee80211_sta *sta)
+{
+	u32 bw = WMI_PEER_CHWIDTH_20MHZ;
+
+	switch (sta->deflink.bandwidth) {
+	case IEEE80211_STA_RX_BW_20:
+		bw = WMI_PEER_CHWIDTH_20MHZ;
+		break;
+	case IEEE80211_STA_RX_BW_40:
+		bw = WMI_PEER_CHWIDTH_40MHZ;
+		break;
+	case IEEE80211_STA_RX_BW_80:
+		bw = WMI_PEER_CHWIDTH_80MHZ;
+		break;
+	case IEEE80211_STA_RX_BW_160:
+		bw = WMI_PEER_CHWIDTH_160MHZ;
+		break;
+	default:
+		ath12k_warn(ar->ab, "Invalid bandwidth %d in rc update for %pM\n",
+			    sta->deflink.bandwidth, sta->addr);
+		bw = WMI_PEER_CHWIDTH_20MHZ;
+		break;
+	}
+
+	return bw;
+}
+
 static int ath12k_mac_op_sta_state(struct ieee80211_hw *hw,
 				   struct ieee80211_vif *vif,
 				   struct ieee80211_sta *sta,
@@ -3498,6 +3570,13 @@ static int ath12k_mac_op_sta_state(struct ieee80211_hw *hw,
 		if (ret)
 			ath12k_warn(ar->ab, "Failed to associate station: %pM\n",
 				    sta->addr);
+
+		spin_lock_bh(&ar->data_lock);
+
+		arsta->bw = ath12k_mac_ieee80211_sta_bw_to_wmi(ar, sta);
+		arsta->bw_prev = sta->deflink.bandwidth;
+
+		spin_unlock_bh(&ar->data_lock);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTHORIZED) {
 		spin_lock_bh(&ar->ab->base_lock);
@@ -3607,28 +3686,8 @@ static void ath12k_mac_op_sta_rc_update(struct ieee80211_hw *hw,
 	spin_lock_bh(&ar->data_lock);
 
 	if (changed & IEEE80211_RC_BW_CHANGED) {
-		bw = WMI_PEER_CHWIDTH_20MHZ;
-
-		switch (sta->deflink.bandwidth) {
-		case IEEE80211_STA_RX_BW_20:
-			bw = WMI_PEER_CHWIDTH_20MHZ;
-			break;
-		case IEEE80211_STA_RX_BW_40:
-			bw = WMI_PEER_CHWIDTH_40MHZ;
-			break;
-		case IEEE80211_STA_RX_BW_80:
-			bw = WMI_PEER_CHWIDTH_80MHZ;
-			break;
-		case IEEE80211_STA_RX_BW_160:
-			bw = WMI_PEER_CHWIDTH_160MHZ;
-			break;
-		default:
-			ath12k_warn(ar->ab, "Invalid bandwidth %d in rc update for %pM\n",
-				    sta->deflink.bandwidth, sta->addr);
-			bw = WMI_PEER_CHWIDTH_20MHZ;
-			break;
-		}
-
+		bw = ath12k_mac_ieee80211_sta_bw_to_wmi(ar, sta);
+		arsta->bw_prev = arsta->bw;
 		arsta->bw = bw;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 4c26033d75e414cf0c500055ec58511f93684d87 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Thu, 16 Mar 2023 12:11:44 +0800
Subject: wifi: ath12k: Identify DFS channel when sending scan channel list
 command

WMI_CHAN_INFO_DFS flag should be set when configuring a DFS channel
included in scan channel list. Without it, firmware will not send a
probe request frame which is needed in connection to an AP configured
with hidden SSID/network_id. So fix this to allow probe request frames
to be sent in cases where a beacon frame has been seen on the channel
first.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230316041144.7770-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wmi.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 3e6991120e53..af910296c41e 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -2438,6 +2438,9 @@ int ath12k_wmi_send_scan_chan_list_cmd(struct ath12k *ar,
 			if (channel_arg->psc_channel)
 				chan_info->info |= cpu_to_le32(WMI_CHAN_INFO_PSC);
 
+			if (channel_arg->dfs_set)
+				chan_info->info |= cpu_to_le32(WMI_CHAN_INFO_DFS);
+
 			chan_info->info |= le32_encode_bits(channel_arg->phy_mode,
 							    WMI_CHAN_INFO_MODE);
 			*reg1 |= le32_encode_bits(channel_arg->minpower,
-- 
cgit v1.2.3-70-g09d2


From 72d17c3e86d0f7fad8f3ae77d87d00b7b6788a8e Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Fri, 17 Mar 2023 14:25:33 +0800
Subject: wifi: ath12k: Enable IMPS for WCN7850

To reduce power consumption, change idle_ps to true to enable
Idle Mode Power Save (IMPS) for WCN7850.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230317062533.18240-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/hw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/hw.c b/drivers/net/wireless/ath/ath12k/hw.c
index 91d576fd4b0f..1ffac7e3deaa 100644
--- a/drivers/net/wireless/ath/ath12k/hw.c
+++ b/drivers/net/wireless/ath/ath12k/hw.c
@@ -944,7 +944,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.interface_modes = BIT(NL80211_IFTYPE_STATION),
 		.supports_monitor = false,
 
-		.idle_ps = false,
+		.idle_ps = true,
 		.download_calib = false,
 		.supports_suspend = false,
 		.tcl_ring_retry = false,
-- 
cgit v1.2.3-70-g09d2


From 59da2d7b0e99c58590a0d3d16905cda27b76da3a Mon Sep 17 00:00:00 2001
From: Jason Xing <kernelxing@tencent.com>
Date: Tue, 21 Mar 2023 09:57:46 +0800
Subject: net-sysfs: display two backlog queue len separately

Sometimes we need to know which one of backlog queue can be exactly
long enough to cause some latency when debugging this part is needed.
Thus, we can then separate the display of both.

Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230321015746.96994-1-kerneljasonxing@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/net-procfs.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/net/core/net-procfs.c b/net/core/net-procfs.c
index 1ec23bf8b05c..09f7ed1a04e8 100644
--- a/net/core/net-procfs.c
+++ b/net/core/net-procfs.c
@@ -115,10 +115,14 @@ static int dev_seq_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static u32 softnet_backlog_len(struct softnet_data *sd)
+static u32 softnet_input_pkt_queue_len(struct softnet_data *sd)
 {
-	return skb_queue_len_lockless(&sd->input_pkt_queue) +
-	       skb_queue_len_lockless(&sd->process_queue);
+	return skb_queue_len_lockless(&sd->input_pkt_queue);
+}
+
+static u32 softnet_process_queue_len(struct softnet_data *sd)
+{
+	return skb_queue_len_lockless(&sd->process_queue);
 }
 
 static struct softnet_data *softnet_get_online(loff_t *pos)
@@ -152,6 +156,8 @@ static void softnet_seq_stop(struct seq_file *seq, void *v)
 static int softnet_seq_show(struct seq_file *seq, void *v)
 {
 	struct softnet_data *sd = v;
+	u32 input_qlen = softnet_input_pkt_queue_len(sd);
+	u32 process_qlen = softnet_process_queue_len(sd);
 	unsigned int flow_limit_count = 0;
 
 #ifdef CONFIG_NET_FLOW_LIMIT
@@ -169,12 +175,14 @@ static int softnet_seq_show(struct seq_file *seq, void *v)
 	 * mapping the data a specific CPU
 	 */
 	seq_printf(seq,
-		   "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x\n",
+		   "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x "
+		   "%08x %08x\n",
 		   sd->processed, sd->dropped, sd->time_squeeze, 0,
 		   0, 0, 0, 0, /* was fastroute */
 		   0,	/* was cpu_collision */
 		   sd->received_rps, flow_limit_count,
-		   softnet_backlog_len(sd), (int)seq->index);
+		   input_qlen + process_qlen, (int)seq->index,
+		   input_qlen, process_qlen);
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 7849c42da2ca09f11fe04aeb8dbde63df756edaa Mon Sep 17 00:00:00 2001
From: Grygorii Strashko <grygorii.strashko@ti.com>
Date: Tue, 21 Mar 2023 11:56:00 +0530
Subject: net: ethernet: ti: am65-cpts: adjust estf following ptp changes

When the CPTS clock is synced/adjusted by running linuxptp (ptp4l/phc2sys),
it will cause the TSN EST schedule to drift away over time. This is because
the schedule is driven by the EstF periodic counter whose pulse length is
defined in ref_clk cycles and it does not automatically sync to CPTS clock.
   _______
 _|
  ^
  expected cycle start time boundary
   _______________
 _|_|___|_|
  ^
  EstF drifted away -> direction

To fix it, the same PPM adjustment has to be applied to EstF as done to the
PHC CPTS clock, in order to correct the TSN EST cycle length and keep them
in sync.

Drifted cycle:
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230373377017
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230373877017
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230374377017
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230374877017
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230375377017
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230375877023
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230376377018
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230376877018
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635968230377377018

Stable cycle:
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863193375473
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863193875473
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863194375473
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863194875473
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863195375473
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863195875473
AM65_CPTS_EVT: 7 e1:01770001 e2:000000ff t:1635966863196375473

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Link: https://lore.kernel.org/r/20230321062600.2539544-1-s-vadapalli@ti.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/ti/am65-cpts.c | 34 ++++++++++++++++++++++++----------
 1 file changed, 24 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpts.c b/drivers/net/ethernet/ti/am65-cpts.c
index 16ee9c29cb35..1d9a399c9661 100644
--- a/drivers/net/ethernet/ti/am65-cpts.c
+++ b/drivers/net/ethernet/ti/am65-cpts.c
@@ -175,6 +175,7 @@ struct am65_cpts {
 	u64 timestamp;
 	u32 genf_enable;
 	u32 hw_ts_enable;
+	u32 estf_enable;
 	struct sk_buff_head txq;
 	bool pps_enabled;
 	bool pps_present;
@@ -405,13 +406,13 @@ static irqreturn_t am65_cpts_interrupt(int irq, void *dev_id)
 static int am65_cpts_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 {
 	struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info);
-	u32 pps_ctrl_val = 0, pps_ppm_hi = 0, pps_ppm_low = 0;
+	u32 estf_ctrl_val = 0, estf_ppm_hi = 0, estf_ppm_low = 0;
 	s32 ppb = scaled_ppm_to_ppb(scaled_ppm);
 	int pps_index = cpts->pps_genf_idx;
 	u64 adj_period, pps_adj_period;
 	u32 ctrl_val, ppm_hi, ppm_low;
 	unsigned long flags;
-	int neg_adj = 0;
+	int neg_adj = 0, i;
 
 	if (ppb < 0) {
 		neg_adj = 1;
@@ -441,19 +442,19 @@ static int am65_cpts_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 	ppm_low = lower_32_bits(adj_period);
 
 	if (cpts->pps_enabled) {
-		pps_ctrl_val = am65_cpts_read32(cpts, genf[pps_index].control);
+		estf_ctrl_val = am65_cpts_read32(cpts, genf[pps_index].control);
 		if (neg_adj)
-			pps_ctrl_val &= ~BIT(1);
+			estf_ctrl_val &= ~BIT(1);
 		else
-			pps_ctrl_val |= BIT(1);
+			estf_ctrl_val |= BIT(1);
 
 		/* GenF PPM will do correction using cpts refclk tick which is
 		 * (cpts->ts_add_val + 1) ns, so GenF length PPM adj period
 		 * need to be corrected.
 		 */
 		pps_adj_period = adj_period * (cpts->ts_add_val + 1);
-		pps_ppm_hi = upper_32_bits(pps_adj_period) & 0x3FF;
-		pps_ppm_low = lower_32_bits(pps_adj_period);
+		estf_ppm_hi = upper_32_bits(pps_adj_period) & 0x3FF;
+		estf_ppm_low = lower_32_bits(pps_adj_period);
 	}
 
 	spin_lock_irqsave(&cpts->lock, flags);
@@ -471,11 +472,18 @@ static int am65_cpts_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
 	am65_cpts_write32(cpts, ppm_low, ts_ppm_low);
 
 	if (cpts->pps_enabled) {
-		am65_cpts_write32(cpts, pps_ctrl_val, genf[pps_index].control);
-		am65_cpts_write32(cpts, pps_ppm_hi, genf[pps_index].ppm_hi);
-		am65_cpts_write32(cpts, pps_ppm_low, genf[pps_index].ppm_low);
+		am65_cpts_write32(cpts, estf_ctrl_val, genf[pps_index].control);
+		am65_cpts_write32(cpts, estf_ppm_hi, genf[pps_index].ppm_hi);
+		am65_cpts_write32(cpts, estf_ppm_low, genf[pps_index].ppm_low);
 	}
 
+	for (i = 0; i < AM65_CPTS_ESTF_MAX_NUM; i++) {
+		if (cpts->estf_enable & BIT(i)) {
+			am65_cpts_write32(cpts, estf_ctrl_val, estf[i].control);
+			am65_cpts_write32(cpts, estf_ppm_hi, estf[i].ppm_hi);
+			am65_cpts_write32(cpts, estf_ppm_low, estf[i].ppm_low);
+		}
+	}
 	/* All GenF/EstF can be updated here the same way */
 	spin_unlock_irqrestore(&cpts->lock, flags);
 
@@ -596,6 +604,11 @@ int am65_cpts_estf_enable(struct am65_cpts *cpts, int idx,
 	am65_cpts_write32(cpts, val, estf[idx].comp_lo);
 	val = lower_32_bits(cycles);
 	am65_cpts_write32(cpts, val, estf[idx].length);
+	am65_cpts_write32(cpts, 0, estf[idx].control);
+	am65_cpts_write32(cpts, 0, estf[idx].ppm_hi);
+	am65_cpts_write32(cpts, 0, estf[idx].ppm_low);
+
+	cpts->estf_enable |= BIT(idx);
 
 	dev_dbg(cpts->dev, "%s: ESTF:%u enabled\n", __func__, idx);
 
@@ -606,6 +619,7 @@ EXPORT_SYMBOL_GPL(am65_cpts_estf_enable);
 void am65_cpts_estf_disable(struct am65_cpts *cpts, int idx)
 {
 	am65_cpts_write32(cpts, 0, estf[idx].length);
+	cpts->estf_enable &= ~BIT(idx);
 
 	dev_dbg(cpts->dev, "%s: ESTF:%u disabled\n", __func__, idx);
 }
-- 
cgit v1.2.3-70-g09d2


From 1595ecce1cf32688760281f40b58b7a1d4a69aa9 Mon Sep 17 00:00:00 2001
From: Krishnanand Prabhu <krishnanand.prabhu@intel.com>
Date: Mon, 20 Mar 2023 12:33:02 +0200
Subject: wifi: iwlwifi: mvm: add support for PTP HW clock (PHC)

Add support to enable/disable PHC clock. The PHC clock includes support
for fetching the cross timestamp i.e. a non-atomic snapshot of the current
time from the hardware (WiFi device) clock and system clock (wall-clock)
simultaneously.

Signed-off-by: Krishnanand Prabhu <krishnanand.prabhu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.ae1d64f513b9.Ib3b6ad61c9fa2fc5908f1e0d6f59f4af6eec1a77@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/Makefile |   1 +
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c     |   3 +
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h    |  15 +++
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c    |   2 +
 drivers/net/wireless/intel/iwlwifi/mvm/ptp.c    | 131 ++++++++++++++++++++++++
 5 files changed, 152 insertions(+)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/ptp.c

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
index b82f79ac5378..943e8de9bc98 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
@@ -8,6 +8,7 @@ iwlmvm-y += tt.o offloading.o tdls.o
 iwlmvm-y += ftm-responder.o ftm-initiator.o
 iwlmvm-y += rfi.o
 iwlmvm-y += mld-key.o mld-mac.o link.o mld-sta.o mld-mac80211.o
+iwlmvm-y += ptp.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
 iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM) += d3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index fe7dfdcd531b..4079c0d939c1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1669,6 +1669,9 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
 		iwl_mvm_send_recovery_cmd(mvm, ERROR_RECOVERY_UPDATE_DB);
 
+	if (!mvm->ptp_data.ptp_clock)
+		iwl_mvm_ptp_init(mvm);
+
 	if (iwl_acpi_get_eckv(mvm->dev, &mvm->ext_clock_valid))
 		IWL_DEBUG_INFO(mvm, "ECKV table doesn't exist in BIOS\n");
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 7cc3526e78ce..4d99e76811a3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -16,6 +16,8 @@
 #include <linux/thermal.h>
 #endif
 
+#include <linux/ptp_clock_kernel.h>
+
 #include <linux/ktime.h>
 
 #include "iwl-op-mode.h"
@@ -770,6 +772,15 @@ struct iwl_mvm_dqa_txq_info {
 	enum iwl_mvm_queue_status status;
 };
 
+struct ptp_data {
+	struct ptp_clock *ptp_clock;
+	struct ptp_clock_info ptp_clock_info;
+	/* keeps track of GP2 wrap-around */
+	u32 last_gp2;
+	u32 wrap_counter;
+	struct delayed_work dwork;
+};
+
 struct iwl_mvm {
 	/* for logger access */
 	struct device *dev;
@@ -1080,6 +1091,8 @@ struct iwl_mvm {
 
 	struct list_head resp_pasn_list;
 
+	struct ptp_data ptp_data;
+
 	struct {
 		u8 range_resp;
 	} cmd_ver;
@@ -2121,6 +2134,8 @@ void iwl_mvm_event_frame_timeout_callback(struct iwl_mvm *mvm,
 					  const struct ieee80211_sta *sta,
 					  u16 tid);
 
+void iwl_mvm_ptp_init(struct iwl_mvm *mvm);
+void iwl_mvm_ptp_remove(struct iwl_mvm *mvm);
 int iwl_mvm_sar_select_profile(struct iwl_mvm *mvm, int prof_a, int prof_b);
 int iwl_mvm_get_sar_geo_profile(struct iwl_mvm *mvm);
 int iwl_mvm_ppag_send_cmd(struct iwl_mvm *mvm);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 03da4e53de67..e7def1a0f767 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1435,6 +1435,8 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode)
 	kfree(mvm->error_recovery_buf);
 	mvm->error_recovery_buf = NULL;
 
+	iwl_mvm_ptp_remove(mvm);
+
 	iwl_trans_op_mode_leave(mvm->trans);
 
 	iwl_phy_db_free(mvm->phy_db);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
new file mode 100644
index 000000000000..f786f5efb497
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
@@ -0,0 +1,131 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2021 - 2023 Intel Corporation
+ */
+
+#include "mvm.h"
+#include "iwl-debug.h"
+#include <linux/timekeeping.h>
+
+#define IWL_PTP_GP2_WRAP	0x100000000ULL
+#define IWL_PTP_WRAP_TIME	(3600 * HZ)
+
+static void iwl_mvm_ptp_update_new_read(struct iwl_mvm *mvm, u32 gp2)
+{
+	if (gp2 < mvm->ptp_data.last_gp2) {
+		mvm->ptp_data.wrap_counter++;
+		IWL_DEBUG_INFO(mvm,
+			       "PTP: wraparound detected (new counter=%u)\n",
+			       mvm->ptp_data.wrap_counter);
+	}
+
+	mvm->ptp_data.last_gp2 = gp2;
+	schedule_delayed_work(&mvm->ptp_data.dwork, IWL_PTP_WRAP_TIME);
+}
+
+static int
+iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
+			       struct system_device_crosststamp *xtstamp)
+{
+	struct iwl_mvm *mvm = container_of(ptp, struct iwl_mvm,
+					   ptp_data.ptp_clock_info);
+	/* Raw value read from GP2 register in usec */
+	u32 gp2;
+	/* GP2 value in ns*/
+	s64 gp2_ns;
+	/* System (wall) time */
+	ktime_t sys_time;
+
+	memset(xtstamp, 0, sizeof(struct system_device_crosststamp));
+
+	if (!mvm->ptp_data.ptp_clock) {
+		IWL_ERR(mvm, "No PHC clock registered\n");
+		return -ENODEV;
+	}
+
+	iwl_mvm_get_sync_time(mvm, CLOCK_REALTIME, &gp2, NULL, &sys_time);
+
+	iwl_mvm_ptp_update_new_read(mvm, gp2);
+
+	gp2_ns = (gp2 + (mvm->ptp_data.wrap_counter * IWL_PTP_GP2_WRAP)) *
+		NSEC_PER_USEC;
+
+	IWL_INFO(mvm, "Got Sync Time: GP2:%u, last_GP2: %u, GP2_ns: %lld, sys_time: %lld\n",
+		 gp2, mvm->ptp_data.last_gp2, gp2_ns, (s64)sys_time);
+
+	/* System monotonic raw time is not used */
+	xtstamp->device = (ktime_t)gp2_ns;
+	xtstamp->sys_realtime = sys_time;
+
+	return 0;
+}
+
+static void iwl_mvm_ptp_work(struct work_struct *wk)
+{
+	struct iwl_mvm *mvm = container_of(wk, struct iwl_mvm,
+					   ptp_data.dwork.work);
+	u32 gp2;
+
+	mutex_lock(&mvm->mutex);
+	gp2 = iwl_mvm_get_systime(mvm);
+	iwl_mvm_ptp_update_new_read(mvm, gp2);
+	mutex_unlock(&mvm->mutex);
+}
+
+/* iwl_mvm_ptp_init - initialize PTP for devices which support it.
+ * @mvm: internal mvm structure, see &struct iwl_mvm.
+ *
+ * Performs the required steps for enabling PTP support.
+ */
+void iwl_mvm_ptp_init(struct iwl_mvm *mvm)
+{
+	/* Warn if the interface already has a ptp_clock defined */
+	if (WARN_ON(mvm->ptp_data.ptp_clock))
+		return;
+
+	mvm->ptp_data.ptp_clock_info.owner = THIS_MODULE;
+	mvm->ptp_data.ptp_clock_info.max_adj = 0x7fffffff;
+	mvm->ptp_data.ptp_clock_info.getcrosststamp =
+					iwl_mvm_phc_get_crosstimestamp;
+
+	/* Give a short 'friendly name' to identify the PHC clock */
+	snprintf(mvm->ptp_data.ptp_clock_info.name,
+		 sizeof(mvm->ptp_data.ptp_clock_info.name),
+		 "%s", "iwlwifi-PTP");
+
+	INIT_DELAYED_WORK(&mvm->ptp_data.dwork, iwl_mvm_ptp_work);
+
+	mvm->ptp_data.ptp_clock =
+		ptp_clock_register(&mvm->ptp_data.ptp_clock_info, mvm->dev);
+
+	if (IS_ERR(mvm->ptp_data.ptp_clock)) {
+		IWL_ERR(mvm, "Failed to register PHC clock (%ld)\n",
+			PTR_ERR(mvm->ptp_data.ptp_clock));
+		mvm->ptp_data.ptp_clock = NULL;
+	} else if (mvm->ptp_data.ptp_clock) {
+		IWL_INFO(mvm, "Registered PHC clock: %s, with index: %d\n",
+			 mvm->ptp_data.ptp_clock_info.name,
+			 ptp_clock_index(mvm->ptp_data.ptp_clock));
+	}
+}
+
+/* iwl_mvm_ptp_remove - disable PTP device.
+ * @mvm: internal mvm structure, see &struct iwl_mvm.
+ *
+ * Disable PTP support.
+ */
+void iwl_mvm_ptp_remove(struct iwl_mvm *mvm)
+{
+	if (mvm->ptp_data.ptp_clock) {
+		IWL_INFO(mvm, "Unregistering PHC clock: %s, with index: %d\n",
+			 mvm->ptp_data.ptp_clock_info.name,
+			 ptp_clock_index(mvm->ptp_data.ptp_clock));
+
+		ptp_clock_unregister(mvm->ptp_data.ptp_clock);
+		mvm->ptp_data.ptp_clock = NULL;
+		memset(&mvm->ptp_data.ptp_clock_info, 0,
+		       sizeof(mvm->ptp_data.ptp_clock_info));
+		mvm->ptp_data.last_gp2 = 0;
+		cancel_delayed_work_sync(&mvm->ptp_data.dwork);
+	}
+}
-- 
cgit v1.2.3-70-g09d2


From 21fb8da6ebe40ce83468d6198143b4b583b967f9 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 20 Mar 2023 12:33:03 +0200
Subject: wifi: iwlwifi: mvm: read synced time from firmware if supported

If the firmware supports reading synced GP2/ATRB timestamps,
read the synced timestamps from firmware instead of reading the
GP2 register and the system time separately. Reading the synced
time from firmware should be more accurate.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.a6be5f0b5580.Idedb496a5943fa496143066ffbed08185a8c4582@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/datapath.h   | 49 ++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/fw/file.h       |  1 +
 drivers/net/wireless/intel/iwlwifi/mvm/ptp.c       | 57 +++++++++++++++++++++-
 3 files changed, 104 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
index 8b38a0073077..59db52f32e43 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2020 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2022 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -26,6 +26,11 @@ enum iwl_data_path_subcmd_ids {
 	 */
 	TRIGGER_RX_QUEUES_NOTIF_CMD = 0x2,
 
+	/**
+	 * @WNM_PLATFORM_PTM_REQUEST_CMD: &struct iwl_time_sync_cfg_cmd
+	 */
+	WNM_PLATFORM_PTM_REQUEST_CMD = 0x3,
+
 	/**
 	 * @STA_HE_CTXT_CMD: &struct iwl_he_sta_context_cmd
 	 */
@@ -146,6 +151,48 @@ enum iwl_channel_estimation_flags {
 	IWL_CHANNEL_ESTIMATION_COUNTER	= BIT(2),
 };
 
+/**
+ * enum iwl_synced_time_operation - PTM request options
+ *
+ * @IWL_SYNCED_TIME_OPERATION_READ_ARTB: read only the ARTB time
+ * @IWL_SYNCED_TIME_OPERATION_READ_GP2: read only the GP2 time
+ * @IWL_SYNCED_TIME_OPERATION_READ_BOTH: latch the ARTB and GP2 clocks and
+ *	provide timestamps from both clocks for the same time point
+ */
+enum iwl_synced_time_operation {
+	IWL_SYNCED_TIME_OPERATION_READ_ARTB = 1,
+	IWL_SYNCED_TIME_OPERATION_READ_GP2,
+	IWL_SYNCED_TIME_OPERATION_READ_BOTH,
+};
+
+/**
+ * struct iwl_synced_time_cmd - request synced GP2/ARTB timestamps
+ *
+ * @operation: one of &enum iwl_synced_time_operation
+ */
+struct iwl_synced_time_cmd {
+	__le32 operation;
+} __packed; /* WNM_80211V_TIMING_CMD_API_S_VER_1 */
+
+/**
+ * struct iwl_synced_time_rsp - response to iwl_synced_time_cmd
+ *
+ * @operation: one of &enum iwl_synced_time_operation
+ * @platform_timestamp_hi: high DWORD of the ARTB clock timestamp in nanoseconds
+ * @platform_timestamp_lo: low DWORD of the ARTB clock timestamp in nanoseconds
+ * @gp2_timestamp_hi: high DWORD of the GP2 clock timestamp in 10's of
+ *	nanoseconds
+ * @gp2_timestamp_lo: low DWORD of the GP2 clock timestamp in 10's of
+ *	nanoseconds
+ */
+struct iwl_synced_time_rsp {
+	__le32 operation;
+	__le32 platform_timestamp_hi;
+	__le32 platform_timestamp_lo;
+	__le32 gp2_timestamp_hi;
+	__le32 gp2_timestamp_lo;
+} __packed; /* WNM_80211V_TIMING_RSP_API_S_VER_1 */
+
 /**
  * struct iwl_channel_estimation_cfg - channel estimation reporting config
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index a7817d952022..e4b2f1a87575 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -455,6 +455,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_BIGTK_SUPPORT		= (__force iwl_ucode_tlv_capa_t)100,
 	IWL_UCODE_TLV_CAPA_DRAM_FRAG_SUPPORT		= (__force iwl_ucode_tlv_capa_t)104,
 	IWL_UCODE_TLV_CAPA_DUMP_COMPLETE_SUPPORT	= (__force iwl_ucode_tlv_capa_t)105,
+	IWL_UCODE_TLV_CAPA_SYNCED_TIME			= (__force iwl_ucode_tlv_capa_t)106,
 
 #ifdef __CHECKER__
 	/* sparse says it cannot increment the previous enum member */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
index f786f5efb497..10f23c8b1dd4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
@@ -23,12 +23,54 @@ static void iwl_mvm_ptp_update_new_read(struct iwl_mvm *mvm, u32 gp2)
 	schedule_delayed_work(&mvm->ptp_data.dwork, IWL_PTP_WRAP_TIME);
 }
 
+static int
+iwl_mvm_get_crosstimestamp_fw(struct iwl_mvm *mvm, u32 *gp2, u64 *sys_time)
+{
+	struct iwl_synced_time_cmd synced_time_cmd = {
+		.operation = cpu_to_le32(IWL_SYNCED_TIME_OPERATION_READ_BOTH)
+	};
+	struct iwl_host_cmd cmd = {
+		.id = WIDE_ID(DATA_PATH_GROUP, WNM_PLATFORM_PTM_REQUEST_CMD),
+		.flags = CMD_WANT_SKB,
+		.data[0] = &synced_time_cmd,
+		.len[0] = sizeof(synced_time_cmd),
+	};
+	struct iwl_synced_time_rsp *resp;
+	struct iwl_rx_packet *pkt;
+	int ret;
+	u64 gp2_10ns;
+
+	ret = iwl_mvm_send_cmd(mvm, &cmd);
+	if (ret)
+		return ret;
+
+	pkt = cmd.resp_pkt;
+
+	if (iwl_rx_packet_payload_len(pkt) != sizeof(*resp)) {
+		IWL_ERR(mvm, "PTP: Invalid command response\n");
+		iwl_free_resp(&cmd);
+		return -EIO;
+	}
+
+	resp = (void *)pkt->data;
+
+	gp2_10ns = (u64)le32_to_cpu(resp->gp2_timestamp_hi) << 32 |
+		le32_to_cpu(resp->gp2_timestamp_lo);
+	*gp2 = gp2_10ns / 100;
+
+	*sys_time = (u64)le32_to_cpu(resp->platform_timestamp_hi) << 32 |
+		le32_to_cpu(resp->platform_timestamp_lo);
+
+	return ret;
+}
+
 static int
 iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
 			       struct system_device_crosststamp *xtstamp)
 {
 	struct iwl_mvm *mvm = container_of(ptp, struct iwl_mvm,
 					   ptp_data.ptp_clock_info);
+	int ret = 0;
 	/* Raw value read from GP2 register in usec */
 	u32 gp2;
 	/* GP2 value in ns*/
@@ -43,7 +85,16 @@ iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
 		return -ENODEV;
 	}
 
-	iwl_mvm_get_sync_time(mvm, CLOCK_REALTIME, &gp2, NULL, &sys_time);
+	mutex_lock(&mvm->mutex);
+	if (fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_SYNCED_TIME)) {
+		ret = iwl_mvm_get_crosstimestamp_fw(mvm, &gp2, &sys_time);
+
+		if (ret)
+			goto out;
+	} else {
+		iwl_mvm_get_sync_time(mvm, CLOCK_REALTIME, &gp2, NULL,
+				      &sys_time);
+	}
 
 	iwl_mvm_ptp_update_new_read(mvm, gp2);
 
@@ -57,7 +108,9 @@ iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
 	xtstamp->device = (ktime_t)gp2_ns;
 	xtstamp->sys_realtime = sys_time;
 
-	return 0;
+out:
+	mutex_unlock(&mvm->mutex);
+	return ret;
 }
 
 static void iwl_mvm_ptp_work(struct work_struct *wk)
-- 
cgit v1.2.3-70-g09d2


From c7eca79def44f1faf024d8442044287bef749818 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 20 Mar 2023 12:33:04 +0200
Subject: wifi: iwlwifi: mvm: report hardware timestamps in RX/TX status

For TM/FTM frames, report the hardware timestamps reported by the
fw as part of the RX/TX status. Since the fw reports the timestamps
in a dedicated notification (and not as part of the RX/TX status),
hold the frame until the fw timestamps notification is received.
Timestamping is enabled when a station is connected and disabled
when disconnected. For AP interface, only the first station will
have timestamping enabled since the fw only supports timestamping
for one peer.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.e0392d498101.I9bf12c8ecfb3f17253a13dc48a48647ddd6e7855@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/commands.h   |  18 +++
 .../net/wireless/intel/iwlwifi/fw/api/datapath.h   | 135 ++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/fw/file.h       |   1 +
 drivers/net/wireless/intel/iwlwifi/mvm/Makefile    |   1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   8 +
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |  14 ++
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c      |   7 +-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       |   8 +
 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c | 174 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h |  30 ++++
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c        |   4 +-
 11 files changed, 396 insertions(+), 4 deletions(-)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
 create mode 100644 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h b/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h
index 28c87a480246..111d96cbde6f 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h
@@ -265,6 +265,24 @@ enum iwl_legacy_cmds {
 	 */
 	HOT_SPOT_CMD = 0x53,
 
+	/**
+	 * @WNM_80211V_TIMING_MEASUREMENT_NOTIFICATION: Time Sync
+	 *	measurement notification for TM/FTM. Sent on receipt of
+	 *	respective WNM action frame for TM protocol or public action
+	 *	frame for FTM protocol from peer device along with additional
+	 *	meta data specified in &struct iwl_time_msmt_notify
+	 */
+	WNM_80211V_TIMING_MEASUREMENT_NOTIFICATION = 0x67,
+
+	/**
+	 * @WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION: Time Sync
+	 *	measurement confirmation notification for TM/FTM. Sent on
+	 *	receipt of Ack from peer for previously Tx'ed TM/FTM
+	 *	action frame along with additional meta data specified in
+	 *	&struct iwl_time_msmt_cfm_notify
+	 */
+	WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION = 0x68,
+
 	/**
 	 * @SCAN_OFFLOAD_COMPLETE:
 	 * notification, &struct iwl_periodic_scan_complete
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
index 59db52f32e43..6f59381b9f9a 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
@@ -31,6 +31,12 @@ enum iwl_data_path_subcmd_ids {
 	 */
 	WNM_PLATFORM_PTM_REQUEST_CMD = 0x3,
 
+	/**
+	 * @WNM_80211V_TIMING_MEASUREMENT_CONFIG_CMD:
+	 *	&struct iwl_time_sync_cfg_cmd
+	 */
+	WNM_80211V_TIMING_MEASUREMENT_CONFIG_CMD = 0x4,
+
 	/**
 	 * @STA_HE_CTXT_CMD: &struct iwl_he_sta_context_cmd
 	 */
@@ -151,6 +157,25 @@ enum iwl_channel_estimation_flags {
 	IWL_CHANNEL_ESTIMATION_COUNTER	= BIT(2),
 };
 
+enum iwl_time_sync_protocol_type {
+	IWL_TIME_SYNC_PROTOCOL_TM	= BIT(0),
+	IWL_TIME_SYNC_PROTOCOL_FTM	= BIT(1),
+}; /* WNM_TIMING_ENABLED_PROTOCOL_API_E_VER_1 */
+
+/**
+ * struct iwl_time_sync_cfg_cmd - TM/FTM time sync measurement configuration
+ *
+ * @protocols: The type of frames to raise notifications for. A bitmap
+ *	of @iwl_time_sync_protocol_type
+ * @peer_addr: peer address with which TM/FTM measurements are required
+ * @reserved: for alignment
+ */
+struct iwl_time_sync_cfg_cmd {
+	__le32 protocols;
+	u8 peer_addr[ETH_ALEN];
+	u8 reserved[2];
+} __packed; /* WNM_80211V_TIMING_MEASUREMENT_CONFIG_CMD_API_S_VER_1 */
+
 /**
  * enum iwl_synced_time_operation - PTM request options
  *
@@ -193,6 +218,116 @@ struct iwl_synced_time_rsp {
 	__le32 gp2_timestamp_lo;
 } __packed; /* WNM_80211V_TIMING_RSP_API_S_VER_1 */
 
+/* PTP_CTX_MAX_DATA_SIZE_IN_API_D_VER_1 */
+#define PTP_CTX_MAX_DATA_SIZE   128
+
+/**
+ * struct iwl_time_msmt_ptp_ctx - Vendor specific information element
+ * to allow a space for flexibility for the userspace App
+ *
+ * @element_id: element id of vendor specific ie
+ * @length: length of vendor specific ie
+ * @reserved: for alignment
+ * @data: vendor specific data blob
+ */
+struct iwl_time_msmt_ptp_ctx {
+	/* Differentiate between FTM and TM specific Vendor IEs */
+	union {
+		struct {
+			u8 element_id;
+			u8 length;
+			__le16 reserved;
+			u8 data[PTP_CTX_MAX_DATA_SIZE];
+		} ftm; /* FTM specific vendor IE */
+		struct {
+			u8 element_id;
+			u8 length;
+			u8 data[PTP_CTX_MAX_DATA_SIZE];
+		} tm; /* TM specific vendor IE */
+	};
+} __packed /* PTP_CTX_VER_1 */;
+
+/**
+ * struct iwl_time_msmt_notify - Time Sync measurement notification
+ * for TM/FTM, along with additional meta data.
+ *
+ * @peer_addr: peer address
+ * @reserved: for alignment
+ * @dialog_token: measurement flow dialog token number
+ * @followup_dialog_token: Measurement flow previous dialog token number
+ * @t1_hi: high dword of t1-time of the Tx'ed action frame departure on
+ *	sender side in units of 10 nano seconds
+ * @t1_lo: low dword of t1-time of the Tx'ed action frame departure on
+ *	sender side in units of 10 nano seconds
+ * @t1_max_err: maximum t1-time error in units of 10 nano seconds
+ * @t4_hi: high dword of t4-time of the Rx'ed action frame's Ack arrival on
+ *	sender side in units of 10 nano seconds
+ * @t4_lo: low dword of t4-time of the Rx'ed action frame's Ack arrival on
+ *	sender side in units of 10 nano seconds
+ * @t4_max_err: maximum t4-time error in units of 10 nano seconds
+ * @t2_hi: high dword of t2-time of the Rx'ed action frame arrival on
+ *	receiver side in units of 10 nano seconds
+ * @t2_lo: low dword of t2-time of the Rx'ed action frame arrival on
+ *	receiver side in units of 10 nano seconds
+ * @t2_max_err: maximum t2-time error in units of 10 nano seconds
+ * @t3_hi: high dword of t3-time of the Tx'ed action frame's Ack departure on
+ *	receiver side in units of 10 nano seconds
+ * @t3_lo: low dword of t3-time of the Tx'ed action frame's Ack departure on
+ *	receiver side in units of 10 nano seconds
+ * @t3_max_err: maximum t3-time error in units of 10 nano seconds
+ * @ptp: vendor specific information element
+ */
+struct iwl_time_msmt_notify {
+	u8 peer_addr[ETH_ALEN];
+	u8 reserved[2];
+	__le32 dialog_token;
+	__le32 followup_dialog_token;
+	__le32 t1_hi;
+	__le32 t1_lo;
+	__le32 t1_max_err;
+	__le32 t4_hi;
+	__le32 t4_lo;
+	__le32 t4_max_err;
+	__le32 t2_hi;
+	__le32 t2_lo;
+	__le32 t2_max_err;
+	__le32 t3_hi;
+	__le32 t3_lo;
+	__le32 t3_max_err;
+	struct iwl_time_msmt_ptp_ctx ptp;
+} __packed; /* WNM_80211V_TIMING_MEASUREMENT_NTFY_API_S_VER_1 */
+
+/**
+ * struct iwl_time_msmt_cfm_notify - Time Sync measurement confirmation
+ * notification for TM/FTM. Sent on receipt of 802.11 Ack from peer for the
+ * Tx'ed TM/FTM measurement action frame.
+ *
+ * @peer_addr: peer address
+ * @reserved: for alignment
+ * @dialog_token: measurement flow dialog token number
+ * @t1_hi: high dword of t1-time of the Tx'ed action frame departure on
+ *	sender side in units of 10 nano seconds
+ * @t1_lo: low dword of t1-time of the Tx'ed action frame departure on
+ *	sender side in units of 10 nano seconds
+ * @t1_max_err: maximum t1-time error in units of 10 nano seconds
+ * @t4_hi: high dword of t4-time of the Rx'ed action frame's Ack arrival on
+ *	sender side in units of 10 nano seconds
+ * @t4_lo: low dword of t4-time of the Rx'ed action frame's Ack arrival on
+ *	sender side in units of 10 nano seconds
+ * @t4_max_err: maximum t4-time error in units of 10 nano seconds
+ */
+struct iwl_time_msmt_cfm_notify {
+	u8 peer_addr[ETH_ALEN];
+	u8 reserved[2];
+	__le32 dialog_token;
+	__le32 t1_hi;
+	__le32 t1_lo;
+	__le32 t1_max_err;
+	__le32 t4_hi;
+	__le32 t4_lo;
+	__le32 t4_max_err;
+} __packed; /* WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NTFY_API_S_VER_1 */
+
 /**
  * struct iwl_channel_estimation_cfg - channel estimation reporting config
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index e4b2f1a87575..1c47046e8f69 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -456,6 +456,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_DRAM_FRAG_SUPPORT		= (__force iwl_ucode_tlv_capa_t)104,
 	IWL_UCODE_TLV_CAPA_DUMP_COMPLETE_SUPPORT	= (__force iwl_ucode_tlv_capa_t)105,
 	IWL_UCODE_TLV_CAPA_SYNCED_TIME			= (__force iwl_ucode_tlv_capa_t)106,
+	IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM        = (__force iwl_ucode_tlv_capa_t)108,
 
 #ifdef __CHECKER__
 	/* sparse says it cannot increment the previous enum member */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
index 943e8de9bc98..593fe28d89cf 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile
@@ -9,6 +9,7 @@ iwlmvm-y += ftm-responder.o ftm-initiator.o
 iwlmvm-y += rfi.o
 iwlmvm-y += mld-key.o mld-mac.o link.o mld-sta.o mld-mac80211.o
 iwlmvm-y += ptp.o
+iwlmvm-y += time-sync.o
 iwlmvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o debugfs-vif.o
 iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o
 iwlmvm-$(CONFIG_PM) += d3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 4d99e76811a3..b6b73164357d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -781,6 +781,12 @@ struct ptp_data {
 	struct delayed_work dwork;
 };
 
+struct iwl_time_sync_data {
+	struct sk_buff_head frame_list;
+	u8 peer_addr[ETH_ALEN];
+	bool active;
+};
+
 struct iwl_mvm {
 	/* for logger access */
 	struct device *dev;
@@ -1126,6 +1132,8 @@ struct iwl_mvm {
 	bool sta_remove_requires_queue_remove;
 
 	bool pldr_sync;
+
+	struct iwl_time_sync_data time_sync;
 };
 
 /* Extract MVM priv from op_mode and _hw */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index e7def1a0f767..70926bb1df6c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -29,6 +29,7 @@
 #include "fw-api.h"
 #include "fw/acpi.h"
 #include "fw/uefi.h"
+#include "time-sync.h"
 
 #define DRV_DESCRIPTION	"The new Intel(R) wireless AGN driver for Linux"
 MODULE_DESCRIPTION(DRV_DESCRIPTION);
@@ -404,6 +405,15 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 	RX_HANDLER_GRP(SYSTEM_GROUP, RFI_DEACTIVATE_NOTIF,
 		       iwl_rfi_deactivate_notif_handler, RX_HANDLER_ASYNC_UNLOCKED,
 		       struct iwl_rfi_deactivate_notif),
+
+	RX_HANDLER_GRP(LEGACY_GROUP,
+		       WNM_80211V_TIMING_MEASUREMENT_NOTIFICATION,
+		       iwl_mvm_time_sync_msmt_event, RX_HANDLER_SYNC,
+		       struct iwl_time_msmt_notify),
+	RX_HANDLER_GRP(LEGACY_GROUP,
+		       WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION,
+		       iwl_mvm_time_sync_msmt_confirm_event, RX_HANDLER_SYNC,
+		       struct iwl_time_msmt_cfm_notify),
 };
 #undef RX_HANDLER
 #undef RX_HANDLER_GRP
@@ -449,6 +459,8 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
 	HCMD_NAME(SCAN_OFFLOAD_PROFILES_QUERY_CMD),
 	HCMD_NAME(BT_COEX_UPDATE_REDUCED_TXP),
 	HCMD_NAME(BT_COEX_CI),
+	HCMD_NAME(WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION),
+	HCMD_NAME(WNM_80211V_TIMING_MEASUREMENT_NOTIFICATION),
 	HCMD_NAME(PHY_CONFIGURATION_CMD),
 	HCMD_NAME(CALIB_RES_NOTIF_PHY_DB),
 	HCMD_NAME(PHY_DB_CMD),
@@ -1338,6 +1350,8 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	else
 		memset(&mvm->rx_stats, 0, sizeof(struct mvm_statistics_rx));
 
+	iwl_mvm_init_time_sync(&mvm->time_sync);
+
 	mvm->debugfs_dir = dbgfs_dir;
 
 	mvm->mei_registered = !iwl_mei_register(mvm, &mei_ops);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 91556d43735a..6286132f165a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -9,6 +9,7 @@
 #include "iwl-trans.h"
 #include "mvm.h"
 #include "fw-api.h"
+#include "time-sync.h"
 
 static inline int iwl_mvm_check_pn(struct iwl_mvm *mvm, struct sk_buff *skb,
 				   int queue, struct ieee80211_sta *sta)
@@ -2585,9 +2586,9 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 		goto out;
 	}
 
-	if (!iwl_mvm_reorder(mvm, napi, queue, sta, skb, desc))
-		iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, queue,
-						sta);
+	if (!iwl_mvm_reorder(mvm, napi, queue, sta, skb, desc) &&
+	    (likely(!iwl_mvm_time_sync_frame(mvm, skb, hdr->addr2))))
+		iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, queue, sta);
 out:
 	rcu_read_unlock();
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index bbcc7459f3a0..1594221abdba 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -9,6 +9,7 @@
 #include "mvm.h"
 #include "sta.h"
 #include "rs.h"
+#include "time-sync.h"
 
 /*
  * New version of ADD_STA_sta command added new fields at the end of the
@@ -1770,6 +1771,11 @@ update_fw:
 		}
 	}
 
+	if (!sta->tdls)
+		iwl_mvm_time_sync_config(mvm, sta->addr,
+					 IWL_TIME_SYNC_PROTOCOL_TM |
+					 IWL_TIME_SYNC_PROTOCOL_FTM);
+
 	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
 
 	return 0;
@@ -1984,6 +1990,8 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 	spin_lock_bh(&mvm_sta->lock);
 	spin_unlock_bh(&mvm_sta->lock);
 
+	iwl_mvm_time_sync_sta_rm(mvm, sta);
+
 	ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->sta_id);
 	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->sta_id], NULL);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
new file mode 100644
index 000000000000..b42e58e5ff24
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2022 Intel Corporation
+ */
+
+#include "mvm.h"
+#include "time-sync.h"
+#include <linux/ieee80211.h>
+
+void iwl_mvm_init_time_sync(struct iwl_time_sync_data *data)
+{
+	skb_queue_head_init(&data->frame_list);
+}
+
+static bool iwl_mvm_is_skb_match(struct sk_buff *skb, u8 *addr, u8 dialog_token)
+{
+	struct ieee80211_mgmt *mgmt = (void *)skb->data;
+	u8 skb_dialog_token;
+
+	if (ieee80211_is_timing_measurement(skb))
+		skb_dialog_token = mgmt->u.action.u.wnm_timing_msr.dialog_token;
+	else
+		skb_dialog_token = mgmt->u.action.u.ftm.dialog_token;
+
+	if ((ether_addr_equal(mgmt->sa, addr) ||
+	     ether_addr_equal(mgmt->da, addr)) &&
+	    skb_dialog_token == dialog_token)
+		return true;
+
+	return false;
+}
+
+static struct sk_buff *iwl_mvm_time_sync_find_skb(struct iwl_mvm *mvm, u8 *addr,
+						  u8 dialog_token)
+{
+	struct sk_buff *skb;
+
+	/* The queue is expected to have only one SKB. If there are other SKBs
+	 * in the queue, they did not get a time sync notification and are
+	 * probably obsolete by now, so drop them.
+	 */
+	while ((skb = skb_dequeue(&mvm->time_sync.frame_list))) {
+		if (iwl_mvm_is_skb_match(skb, addr, dialog_token))
+			break;
+
+		kfree_skb(skb);
+		skb = NULL;
+	}
+
+	return skb;
+}
+
+static u64 iwl_mvm_get_64_bit(__le32 high, __le32 low)
+{
+	return ((u64)le32_to_cpu(high) << 32) | le32_to_cpu(low);
+}
+
+void iwl_mvm_time_sync_msmt_event(struct iwl_mvm *mvm,
+				  struct iwl_rx_cmd_buffer *rxb)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_time_msmt_notify *notif = (void *)pkt->data;
+	struct ieee80211_rx_status *rx_status;
+	struct skb_shared_hwtstamps *shwt;
+	u64 ts_10ns;
+	struct sk_buff *skb =
+		iwl_mvm_time_sync_find_skb(mvm, notif->peer_addr,
+					   le32_to_cpu(notif->dialog_token));
+
+	if (!skb) {
+		IWL_DEBUG_INFO(mvm, "Time sync event but no pending skb\n");
+		return;
+	}
+
+	ts_10ns = iwl_mvm_get_64_bit(notif->t3_hi, notif->t3_lo);
+	rx_status = IEEE80211_SKB_RXCB(skb);
+	rx_status->ack_tx_hwtstamp = ktime_set(0, ts_10ns * 10);
+
+	ts_10ns = iwl_mvm_get_64_bit(notif->t2_hi, notif->t2_lo);
+	shwt = skb_hwtstamps(skb);
+	shwt->hwtstamp = ktime_set(0, ts_10ns * 10);
+
+	IWL_DEBUG_INFO(mvm,
+		       "Time sync: RX event - report frame t2=%llu t3=%llu\n",
+		       ktime_to_ns(shwt->hwtstamp),
+		       ktime_to_ns(rx_status->ack_tx_hwtstamp));
+	ieee80211_rx_napi(mvm->hw, NULL, skb, NULL);
+}
+
+void iwl_mvm_time_sync_msmt_confirm_event(struct iwl_mvm *mvm,
+					  struct iwl_rx_cmd_buffer *rxb)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_time_msmt_cfm_notify *notif = (void *)pkt->data;
+	struct ieee80211_tx_status status = {};
+	struct skb_shared_hwtstamps *shwt;
+	u64 ts_10ns;
+
+	status.skb =
+		iwl_mvm_time_sync_find_skb(mvm, notif->peer_addr,
+					   le32_to_cpu(notif->dialog_token));
+
+	if (!status.skb) {
+		IWL_DEBUG_INFO(mvm, "Time sync confirm but no pending skb\n");
+		return;
+	}
+
+	status.info = IEEE80211_SKB_CB(status.skb);
+
+	ts_10ns = iwl_mvm_get_64_bit(notif->t4_hi, notif->t4_lo);
+	status.ack_hwtstamp = ktime_set(0, ts_10ns * 10);
+
+	ts_10ns = iwl_mvm_get_64_bit(notif->t1_hi, notif->t1_lo);
+	shwt = skb_hwtstamps(status.skb);
+	shwt->hwtstamp = ktime_set(0, ts_10ns * 10);
+
+	IWL_DEBUG_INFO(mvm,
+		       "Time sync: TX event - report frame t1=%llu t4=%llu\n",
+		       ktime_to_ns(shwt->hwtstamp),
+		       ktime_to_ns(status.ack_hwtstamp));
+	ieee80211_tx_status_ext(mvm->hw, &status);
+}
+
+int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, u8 *addr, u32 protocols)
+{
+	struct iwl_time_sync_cfg_cmd cmd = {};
+	int err;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (!fw_has_capa(&mvm->fw->ucode_capa,
+			 IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM))
+		return -EINVAL;
+
+	/* The fw only supports one peer. We do allow reconfiguration of the
+	 * same peer for cases of fw reset etc.
+	 */
+	if (mvm->time_sync.active &&
+	    !ether_addr_equal(addr, mvm->time_sync.peer_addr)) {
+		IWL_DEBUG_INFO(mvm, "Time sync: reject config for peer: %pM\n",
+			       addr);
+		return -ENOBUFS;
+	}
+
+	if (protocols & ~(IWL_TIME_SYNC_PROTOCOL_TM |
+			  IWL_TIME_SYNC_PROTOCOL_FTM))
+		return -EINVAL;
+
+	cmd.protocols = cpu_to_le32(protocols);
+
+	ether_addr_copy(cmd.peer_addr, addr);
+
+	err = iwl_mvm_send_cmd_pdu(mvm,
+				   WIDE_ID(DATA_PATH_GROUP,
+					   WNM_80211V_TIMING_MEASUREMENT_CONFIG_CMD),
+				   0, sizeof(cmd), &cmd);
+	if (err) {
+		IWL_ERR(mvm, "Failed to send time sync cfg cmd: %d\n", err);
+	} else {
+		mvm->time_sync.active = protocols != 0;
+		ether_addr_copy(mvm->time_sync.peer_addr, addr);
+		IWL_DEBUG_INFO(mvm, "Time sync: set peer addr=%pM\n", addr);
+	}
+
+	return err;
+}
+
+void iwl_mvm_time_sync_sta_rm(struct iwl_mvm *mvm, struct ieee80211_sta *sta)
+{
+	/* Disable time sync with this station */
+	iwl_mvm_time_sync_config(mvm, sta->addr, 0);
+
+	skb_queue_purge(&mvm->time_sync.frame_list);
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h
new file mode 100644
index 000000000000..c43d1d0b25cb
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/*
+ * Copyright (C) 2022 Intel Corporation
+ */
+#ifndef __TIME_SYNC_H__
+#define __TIME_SYNC_H__
+
+#include "mvm.h"
+#include <linux/ieee80211.h>
+
+void iwl_mvm_init_time_sync(struct iwl_time_sync_data *data);
+void iwl_mvm_time_sync_msmt_event(struct iwl_mvm *mvm,
+				  struct iwl_rx_cmd_buffer *rxb);
+void iwl_mvm_time_sync_msmt_confirm_event(struct iwl_mvm *mvm,
+					  struct iwl_rx_cmd_buffer *rxb);
+int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, u8 *addr, u32 protocols);
+void iwl_mvm_time_sync_sta_rm(struct iwl_mvm *mvm, struct ieee80211_sta *sta);
+
+static inline
+bool iwl_mvm_time_sync_frame(struct iwl_mvm *mvm, struct sk_buff *skb, u8 *addr)
+{
+	if (ether_addr_equal(mvm->time_sync.peer_addr, addr) &&
+	    (ieee80211_is_timing_measurement(skb) || ieee80211_is_ftm(skb))) {
+		skb_queue_tail(&mvm->time_sync.frame_list, skb);
+		return true;
+	}
+
+	return false;
+}
+#endif
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index a6d69885cd3f..6e2e0791d848 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -14,6 +14,7 @@
 #include "iwl-eeprom-parse.h"
 #include "mvm.h"
 #include "sta.h"
+#include "time-sync.h"
 
 static void
 iwl_mvm_bar_check_trigger(struct iwl_mvm *mvm, const u8 *addr,
@@ -1643,7 +1644,8 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 		info->status.status_driver_data[0] =
 			RS_DRV_DATA_PACK(lq_color, tx_resp->reduced_tpc);
 
-		ieee80211_tx_status(mvm->hw, skb);
+		if (likely(!iwl_mvm_time_sync_frame(mvm, skb, hdr->addr1)))
+			ieee80211_tx_status(mvm->hw, skb);
 	}
 
 	/* This is an aggregation queue or might become one, so we use
-- 
cgit v1.2.3-70-g09d2


From 8e33f046ff0756dd02a0c6da835f32b3a5daf35e Mon Sep 17 00:00:00 2001
From: Krishnanand Prabhu <krishnanand.prabhu@intel.com>
Date: Mon, 20 Mar 2023 12:33:05 +0200
Subject: wifi: iwlwifi: mvm: add support for timing measurement

Add support for timing measurement in extended capabilities, used for
time synchronization.

Signed-off-by: Krishnanand Prabhu <krishnanand.prabhu@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.001d2b459ece.I15ab8fc214edc35f1f362006a9e1a22b89e7ed8e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 44 +++++++++++++++++++++--
 1 file changed, 41 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 93ca92e8eb98..7ee8d24606a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -222,19 +222,39 @@ int iwl_mvm_init_fw_regd(struct iwl_mvm *mvm)
 	return ret;
 }
 
+/* Each capability added here should also be add to tm_if_types_ext_capa_sta */
 static const u8 he_if_types_ext_capa_sta[] = {
 	 [0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
 	 [2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT,
 	 [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
 };
 
-static const struct wiphy_iftype_ext_capab he_iftypes_ext_capa[] = {
+static const u8 tm_if_types_ext_capa_sta[] = {
+	 [0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	 [2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT |
+	       WLAN_EXT_CAPA3_TIMING_MEASUREMENT_SUPPORT,
+	 [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+	 [9] = WLAN_EXT_CAPA10_TWT_REQUESTER_SUPPORT,
+};
+
+/* Additional interface types for which extended capabilities are
+ * specified separately
+ */
+static const struct wiphy_iftype_ext_capab add_iftypes_ext_capa[] = {
 	{
 		.iftype = NL80211_IFTYPE_STATION,
 		.extended_capabilities = he_if_types_ext_capa_sta,
 		.extended_capabilities_mask = he_if_types_ext_capa_sta,
 		.extended_capabilities_len = sizeof(he_if_types_ext_capa_sta),
 	},
+	{
+		.iftype = NL80211_IFTYPE_STATION,
+		.extended_capabilities = tm_if_types_ext_capa_sta,
+		.extended_capabilities_mask = tm_if_types_ext_capa_sta,
+		.extended_capabilities_len = sizeof(tm_if_types_ext_capa_sta),
+		/* relevant only if EHT is supported */
+		.eml_capabilities = IEEE80211_EML_CAP_EMLSR_SUPP,
+	},
 };
 
 static int
@@ -562,16 +582,34 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 					      NL80211_EXT_FEATURE_OCE_PROBE_REQ_DEFERRAL_SUPPRESSION);
 	}
 
+	hw->wiphy->iftype_ext_capab = NULL;
+	hw->wiphy->num_iftype_ext_capab = 0;
+
 	if (mvm->nvm_data->sku_cap_11ax_enable &&
 	    !iwlwifi_mod_params.disable_11ax) {
-		hw->wiphy->iftype_ext_capab = he_iftypes_ext_capa;
+		hw->wiphy->iftype_ext_capab = add_iftypes_ext_capa;
 		hw->wiphy->num_iftype_ext_capab =
-			ARRAY_SIZE(he_iftypes_ext_capa);
+			ARRAY_SIZE(add_iftypes_ext_capa) - 1;
 
 		ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID);
 		ieee80211_hw_set(hw, SUPPORTS_ONLY_HE_MULTI_BSSID);
 	}
 
+	if (iwl_fw_lookup_cmd_ver(mvm->fw,
+				  WIDE_ID(DATA_PATH_GROUP,
+					  WNM_80211V_TIMING_MEASUREMENT_CONFIG_CMD),
+				  IWL_FW_CMD_VER_UNKNOWN) >= 1) {
+		IWL_DEBUG_INFO(mvm->trans, "Timing measurement supported\n");
+
+		if (!hw->wiphy->iftype_ext_capab) {
+			hw->wiphy->num_iftype_ext_capab = 1;
+			hw->wiphy->iftype_ext_capab = add_iftypes_ext_capa +
+				ARRAY_SIZE(add_iftypes_ext_capa) - 1;
+		} else {
+			hw->wiphy->iftype_ext_capab = add_iftypes_ext_capa + 1;
+		}
+	}
+
 	mvm->rts_threshold = IEEE80211_MAX_RTS_THRESHOLD;
 
 #ifdef CONFIG_PM_SLEEP
-- 
cgit v1.2.3-70-g09d2


From a5de7de7e78eddd3ae16aec01c5601ae36c1c2e8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 20 Mar 2023 12:33:06 +0200
Subject: wifi: iwlwifi: mvm: enable TX beacon protection

Check hardware/firmware support and enable TX beacon protection
as well if supported, programming the key into the firmware as
usual.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.6a90a4f4f469.Ia028dea75f9a8eed40786d876d51f97fb3142688@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/file.h      |  1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 30 +++++++++++++++++------
 2 files changed, 23 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index 1c47046e8f69..f2eefca6e260 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -457,6 +457,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_DUMP_COMPLETE_SUPPORT	= (__force iwl_ucode_tlv_capa_t)105,
 	IWL_UCODE_TLV_CAPA_SYNCED_TIME			= (__force iwl_ucode_tlv_capa_t)106,
 	IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM        = (__force iwl_ucode_tlv_capa_t)108,
+	IWL_UCODE_TLV_CAPA_BIGTK_TX_SUPPORT		= (__force iwl_ucode_tlv_capa_t)109,
 
 #ifdef __CHECKER__
 	/* sparse says it cannot increment the previous enum member */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 7ee8d24606a2..99a51d519214 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -280,6 +280,8 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	bool unified = fw_has_capa(&mvm->fw->ucode_capa,
 				   IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG);
 #endif
+	u32 sec_key_id = WIDE_ID(DATA_PATH_GROUP, SEC_KEY_CMD);
+	u8 sec_key_ver = iwl_fw_lookup_cmd_ver(mvm->fw, sec_key_id, 0);
 
 	/* Tell mac80211 our characteristics */
 	ieee80211_hw_set(hw, SIGNAL_DBM);
@@ -404,8 +406,13 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 		hw->wiphy->pmsr_capa = &iwl_mvm_pmsr_capa;
 	}
 
-	if (fw_has_capa(&mvm->fw->ucode_capa,
-			IWL_UCODE_TLV_CAPA_BIGTK_SUPPORT))
+	if (sec_key_ver &&
+	    fw_has_capa(&mvm->fw->ucode_capa,
+			IWL_UCODE_TLV_CAPA_BIGTK_TX_SUPPORT))
+		wiphy_ext_feature_set(hw->wiphy,
+				      NL80211_EXT_FEATURE_BEACON_PROTECTION);
+	else if (fw_has_capa(&mvm->fw->ucode_capa,
+			     IWL_UCODE_TLV_CAPA_BIGTK_SUPPORT))
 		wiphy_ext_feature_set(hw->wiphy,
 				      NL80211_EXT_FEATURE_BEACON_PROTECTION_CLIENT);
 
@@ -3778,7 +3785,8 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 
 	switch (cmd) {
 	case SET_KEY:
-		if (keyidx == 6 || keyidx == 7)
+		if (vif->type == NL80211_IFTYPE_STATION &&
+		    (keyidx == 6 || keyidx == 7))
 			rcu_assign_pointer(mvmvif->bcn_prot.keys[keyidx - 6],
 					   key);
 
@@ -3789,10 +3797,14 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 			 * on IBSS they're per-station and because we're lazy
 			 * we don't support them for RX, so do the same.
 			 * CMAC/GMAC in AP/IBSS modes must be done in software.
+			 *
+			 * Except, of course, beacon protection - it must be
+			 * offloaded since we just set a beacon template.
 			 */
-			if (key->cipher == WLAN_CIPHER_SUITE_AES_CMAC ||
-			    key->cipher == WLAN_CIPHER_SUITE_BIP_GMAC_128 ||
-			    key->cipher == WLAN_CIPHER_SUITE_BIP_GMAC_256) {
+			if (keyidx < 6 &&
+			    (key->cipher == WLAN_CIPHER_SUITE_AES_CMAC ||
+			     key->cipher == WLAN_CIPHER_SUITE_BIP_GMAC_128 ||
+			     key->cipher == WLAN_CIPHER_SUITE_BIP_GMAC_256)) {
 				ret = -EOPNOTSUPP;
 				break;
 			}
@@ -3873,7 +3885,8 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 		if (mvmsta && key->flags & IEEE80211_KEY_FLAG_PAIRWISE)
 			mvmsta->pairwise_cipher = key->cipher;
 
-		IWL_DEBUG_MAC80211(mvm, "set hwcrypto key\n");
+		IWL_DEBUG_MAC80211(mvm, "set hwcrypto key (sta:%pM, id:%d)\n",
+				   sta ? sta->addr : NULL, key->keyidx);
 
 		if (sec_key_ver)
 			ret = iwl_mvm_sec_key_add(mvm, vif, sta, key);
@@ -3897,7 +3910,8 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 
 		break;
 	case DISABLE_KEY:
-		if (keyidx == 6 || keyidx == 7)
+		if (vif->type == NL80211_IFTYPE_STATION &&
+		    (keyidx == 6 || keyidx == 7))
 			RCU_INIT_POINTER(mvmvif->bcn_prot.keys[keyidx - 6],
 					 NULL);
 
-- 
cgit v1.2.3-70-g09d2


From a2f49f7d52a9e7d0dce8f090541a28192b7abd59 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 20 Mar 2023 12:33:07 +0200
Subject: wifi: iwlwifi: mvm: implement PHC clock adjustments

Implement the following PHC operations:
1. adjtime - for adding an offset to the PHC clock time.
2. adjfine - for adjusting the PHC clock frequency.
3. gettime64 - for getting the PHC clock time. This function returns
   the time as adjusted by adjtime and adjfine.

The adjusted time will also be used for time sync frames timestamping.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.f59750deeee4.I110a7e3ac3c6d39d9dbe1fe77001e3f5bc3814eb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  21 +++-
 drivers/net/wireless/intel/iwlwifi/mvm/ptp.c       | 127 ++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c |  28 +++--
 3 files changed, 158 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index b6b73164357d..2cd86237bc14 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -775,10 +775,26 @@ struct iwl_mvm_dqa_txq_info {
 struct ptp_data {
 	struct ptp_clock *ptp_clock;
 	struct ptp_clock_info ptp_clock_info;
-	/* keeps track of GP2 wrap-around */
+
+	struct delayed_work dwork;
+
+	/* The last GP2 reading from the hw */
 	u32 last_gp2;
+
+	/* number of wraparounds since scale_update_adj_time_ns */
 	u32 wrap_counter;
-	struct delayed_work dwork;
+
+	/* GP2 time when the scale was last updated */
+	u32 scale_update_gp2;
+
+	/* Adjusted time when the scale was last updated in nanoseconds */
+	u64 scale_update_adj_time_ns;
+
+	/* clock frequency offset, scaled to 65536000000 */
+	u64 scaled_freq;
+
+	/* Delta between hardware clock and ptp clock in nanoseconds */
+	s64 delta;
 };
 
 struct iwl_time_sync_data {
@@ -2144,6 +2160,7 @@ void iwl_mvm_event_frame_timeout_callback(struct iwl_mvm *mvm,
 
 void iwl_mvm_ptp_init(struct iwl_mvm *mvm);
 void iwl_mvm_ptp_remove(struct iwl_mvm *mvm);
+u64 iwl_mvm_ptp_get_adj_time(struct iwl_mvm *mvm, u64 base_time);
 int iwl_mvm_sar_select_profile(struct iwl_mvm *mvm, int prof_a, int prof_b);
 int iwl_mvm_get_sar_geo_profile(struct iwl_mvm *mvm);
 int iwl_mvm_ppag_send_cmd(struct iwl_mvm *mvm);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
index 10f23c8b1dd4..1eafaaed415d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
@@ -6,12 +6,31 @@
 #include "mvm.h"
 #include "iwl-debug.h"
 #include <linux/timekeeping.h>
+#include <linux/math64.h>
 
 #define IWL_PTP_GP2_WRAP	0x100000000ULL
 #define IWL_PTP_WRAP_TIME	(3600 * HZ)
 
+/* The scaled_ppm parameter is ppm (parts per million) with a 16-bit fractional
+ * part, which means that a value of 1 in one of those fields actually means
+ * 2^-16 ppm, and 2^16=65536 is 1 ppm.
+ */
+#define SCALE_FACTOR	65536000000ULL
+#define IWL_PTP_WRAP_THRESHOLD_USEC	(5000)
+
 static void iwl_mvm_ptp_update_new_read(struct iwl_mvm *mvm, u32 gp2)
 {
+	/* If the difference is above the threshold, assume it's a wraparound.
+	 * Otherwise assume it's an old read and ignore it.
+	 */
+	if (gp2 < mvm->ptp_data.last_gp2 &&
+	    mvm->ptp_data.last_gp2 - gp2 < IWL_PTP_WRAP_THRESHOLD_USEC) {
+		IWL_DEBUG_INFO(mvm,
+			       "PTP: ignore old read (gp2=%u, last_gp2=%u)\n",
+			       gp2, mvm->ptp_data.last_gp2);
+		return;
+	}
+
 	if (gp2 < mvm->ptp_data.last_gp2) {
 		mvm->ptp_data.wrap_counter++;
 		IWL_DEBUG_INFO(mvm,
@@ -23,6 +42,45 @@ static void iwl_mvm_ptp_update_new_read(struct iwl_mvm *mvm, u32 gp2)
 	schedule_delayed_work(&mvm->ptp_data.dwork, IWL_PTP_WRAP_TIME);
 }
 
+u64 iwl_mvm_ptp_get_adj_time(struct iwl_mvm *mvm, u64 base_time_ns)
+{
+	struct ptp_data *data = &mvm->ptp_data;
+	u64 last_gp2_ns = mvm->ptp_data.scale_update_gp2 * NSEC_PER_USEC;
+	u64 res;
+	u64 diff;
+
+	iwl_mvm_ptp_update_new_read(mvm,
+				    div64_u64(base_time_ns, NSEC_PER_USEC));
+
+	IWL_DEBUG_INFO(mvm, "base_time_ns=%llu, wrap_counter=%u\n",
+		       (unsigned long long)base_time_ns, data->wrap_counter);
+
+	base_time_ns = base_time_ns +
+		(data->wrap_counter * IWL_PTP_GP2_WRAP * NSEC_PER_USEC);
+
+	/* It is possible that a GP2 timestamp was received from fw before the
+	 * last scale update. Since we don't know how to scale - ignore it.
+	 */
+	if (base_time_ns < last_gp2_ns) {
+		IWL_DEBUG_INFO(mvm, "Time before scale update - ignore\n");
+		return 0;
+	}
+
+	diff = base_time_ns - last_gp2_ns;
+	IWL_DEBUG_INFO(mvm, "diff ns=%llu\n", (unsigned long long)diff);
+
+	diff = mul_u64_u64_div_u64(diff, data->scaled_freq,
+				   SCALE_FACTOR);
+	IWL_DEBUG_INFO(mvm, "scaled diff ns=%llu\n", (unsigned long long)diff);
+
+	res = data->scale_update_adj_time_ns + data->delta + diff;
+
+	IWL_DEBUG_INFO(mvm, "base=%llu delta=%lld adj=%llu\n",
+		       (unsigned long long)base_time_ns, (long long)data->delta,
+		       (unsigned long long)res);
+	return res;
+}
+
 static int
 iwl_mvm_get_crosstimestamp_fw(struct iwl_mvm *mvm, u32 *gp2, u64 *sys_time)
 {
@@ -96,10 +154,7 @@ iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
 				      &sys_time);
 	}
 
-	iwl_mvm_ptp_update_new_read(mvm, gp2);
-
-	gp2_ns = (gp2 + (mvm->ptp_data.wrap_counter * IWL_PTP_GP2_WRAP)) *
-		NSEC_PER_USEC;
+	gp2_ns = iwl_mvm_ptp_get_adj_time(mvm, (u64)gp2 * NSEC_PER_USEC);
 
 	IWL_INFO(mvm, "Got Sync Time: GP2:%u, last_GP2: %u, GP2_ns: %lld, sys_time: %lld\n",
 		 gp2, mvm->ptp_data.last_gp2, gp2_ns, (s64)sys_time);
@@ -125,6 +180,66 @@ static void iwl_mvm_ptp_work(struct work_struct *wk)
 	mutex_unlock(&mvm->mutex);
 }
 
+static int iwl_mvm_ptp_gettime(struct ptp_clock_info *ptp,
+			       struct timespec64 *ts)
+{
+	struct iwl_mvm *mvm = container_of(ptp, struct iwl_mvm,
+					   ptp_data.ptp_clock_info);
+	u64 gp2;
+	u64 ns;
+
+	mutex_lock(&mvm->mutex);
+	gp2 = iwl_mvm_get_systime(mvm);
+	ns = iwl_mvm_ptp_get_adj_time(mvm, gp2 * NSEC_PER_USEC);
+	mutex_unlock(&mvm->mutex);
+
+	*ts = ns_to_timespec64(ns);
+	return 0;
+}
+
+static int iwl_mvm_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct iwl_mvm *mvm = container_of(ptp, struct iwl_mvm,
+					   ptp_data.ptp_clock_info);
+	struct ptp_data *data = container_of(ptp, struct ptp_data,
+					     ptp_clock_info);
+
+	mutex_lock(&mvm->mutex);
+	data->delta += delta;
+	IWL_DEBUG_INFO(mvm, "delta=%lld, new delta=%lld\n", (long long)delta,
+		       (long long)data->delta);
+	mutex_unlock(&mvm->mutex);
+	return 0;
+}
+
+static int iwl_mvm_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct iwl_mvm *mvm = container_of(ptp, struct iwl_mvm,
+					   ptp_data.ptp_clock_info);
+	struct ptp_data *data = &mvm->ptp_data;
+	u32 gp2;
+
+	mutex_lock(&mvm->mutex);
+
+	/* Must call _iwl_mvm_ptp_get_adj_time() before updating
+	 * data->scale_update_gp2 or data->scaled_freq since
+	 * scale_update_adj_time_ns should reflect the previous scaled_freq.
+	 */
+	gp2 = iwl_mvm_get_systime(mvm);
+	data->scale_update_adj_time_ns =
+		iwl_mvm_ptp_get_adj_time(mvm, gp2 * NSEC_PER_USEC);
+	data->scale_update_gp2 = gp2;
+	data->wrap_counter = 0;
+	data->delta = 0;
+
+	data->scaled_freq = SCALE_FACTOR + scaled_ppm;
+	IWL_DEBUG_INFO(mvm, "adjfine: scaled_ppm=%ld new=%llu\n",
+		       scaled_ppm, (unsigned long long)data->scaled_freq);
+
+	mutex_unlock(&mvm->mutex);
+	return 0;
+}
+
 /* iwl_mvm_ptp_init - initialize PTP for devices which support it.
  * @mvm: internal mvm structure, see &struct iwl_mvm.
  *
@@ -140,6 +255,10 @@ void iwl_mvm_ptp_init(struct iwl_mvm *mvm)
 	mvm->ptp_data.ptp_clock_info.max_adj = 0x7fffffff;
 	mvm->ptp_data.ptp_clock_info.getcrosststamp =
 					iwl_mvm_phc_get_crosstimestamp;
+	mvm->ptp_data.ptp_clock_info.adjfine = iwl_mvm_ptp_adjfine;
+	mvm->ptp_data.ptp_clock_info.adjtime = iwl_mvm_ptp_adjtime;
+	mvm->ptp_data.ptp_clock_info.gettime64 = iwl_mvm_ptp_gettime;
+	mvm->ptp_data.scaled_freq = SCALE_FACTOR;
 
 	/* Give a short 'friendly name' to identify the PHC clock */
 	snprintf(mvm->ptp_data.ptp_clock_info.name,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
index b42e58e5ff24..434d27a44dea 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
@@ -66,19 +66,22 @@ void iwl_mvm_time_sync_msmt_event(struct iwl_mvm *mvm,
 	struct sk_buff *skb =
 		iwl_mvm_time_sync_find_skb(mvm, notif->peer_addr,
 					   le32_to_cpu(notif->dialog_token));
+	u64 adj_time;
 
 	if (!skb) {
 		IWL_DEBUG_INFO(mvm, "Time sync event but no pending skb\n");
 		return;
 	}
 
-	ts_10ns = iwl_mvm_get_64_bit(notif->t3_hi, notif->t3_lo);
-	rx_status = IEEE80211_SKB_RXCB(skb);
-	rx_status->ack_tx_hwtstamp = ktime_set(0, ts_10ns * 10);
-
 	ts_10ns = iwl_mvm_get_64_bit(notif->t2_hi, notif->t2_lo);
+	adj_time = iwl_mvm_ptp_get_adj_time(mvm, ts_10ns * 10);
 	shwt = skb_hwtstamps(skb);
-	shwt->hwtstamp = ktime_set(0, ts_10ns * 10);
+	shwt->hwtstamp = ktime_set(0, adj_time);
+
+	ts_10ns = iwl_mvm_get_64_bit(notif->t3_hi, notif->t3_lo);
+	adj_time = iwl_mvm_ptp_get_adj_time(mvm, ts_10ns * 10);
+	rx_status = IEEE80211_SKB_RXCB(skb);
+	rx_status->ack_tx_hwtstamp = ktime_set(0, adj_time);
 
 	IWL_DEBUG_INFO(mvm,
 		       "Time sync: RX event - report frame t2=%llu t3=%llu\n",
@@ -94,7 +97,7 @@ void iwl_mvm_time_sync_msmt_confirm_event(struct iwl_mvm *mvm,
 	struct iwl_time_msmt_cfm_notify *notif = (void *)pkt->data;
 	struct ieee80211_tx_status status = {};
 	struct skb_shared_hwtstamps *shwt;
-	u64 ts_10ns;
+	u64 ts_10ns, adj_time;
 
 	status.skb =
 		iwl_mvm_time_sync_find_skb(mvm, notif->peer_addr,
@@ -105,14 +108,15 @@ void iwl_mvm_time_sync_msmt_confirm_event(struct iwl_mvm *mvm,
 		return;
 	}
 
-	status.info = IEEE80211_SKB_CB(status.skb);
-
-	ts_10ns = iwl_mvm_get_64_bit(notif->t4_hi, notif->t4_lo);
-	status.ack_hwtstamp = ktime_set(0, ts_10ns * 10);
-
 	ts_10ns = iwl_mvm_get_64_bit(notif->t1_hi, notif->t1_lo);
+	adj_time = iwl_mvm_ptp_get_adj_time(mvm, ts_10ns * 10);
 	shwt = skb_hwtstamps(status.skb);
-	shwt->hwtstamp = ktime_set(0, ts_10ns * 10);
+	shwt->hwtstamp = ktime_set(0, adj_time);
+
+	ts_10ns = iwl_mvm_get_64_bit(notif->t4_hi, notif->t4_lo);
+	adj_time = iwl_mvm_ptp_get_adj_time(mvm, ts_10ns * 10);
+	status.info = IEEE80211_SKB_CB(status.skb);
+	status.ack_hwtstamp = ktime_set(0, adj_time);
 
 	IWL_DEBUG_INFO(mvm,
 		       "Time sync: TX event - report frame t1=%llu t4=%llu\n",
-- 
cgit v1.2.3-70-g09d2


From e86103394474b0c15eca4bb9ed678f6e9ab5d004 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 20 Mar 2023 12:33:08 +0200
Subject: wifi: iwlwifi: mvm: select ptp cross timestamp from multiple reads

iwl_mvm_get_sync_time() reads the gp2 from the device and then
reads the system clock. Since the two reads are not done atomically,
unexpected delays may happen between the two reads (e.g. context
switch) which make it inaccurate.
In order to improve the accuracy of the cross timestamp, call
iwl_mvm_get_sync_time() multiple times in a loop and take the
result in which the difference between the two clock is the smallest.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.d9e6f8f8998a.I569939ec4ddf0c6c64c112e7d0c30583f5509d9a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/ptp.c | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
index 1eafaaed415d..5c2bfc8ed88d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
@@ -18,6 +18,8 @@
 #define SCALE_FACTOR	65536000000ULL
 #define IWL_PTP_WRAP_THRESHOLD_USEC	(5000)
 
+#define IWL_PTP_GET_CROSS_TS_NUM	5
+
 static void iwl_mvm_ptp_update_new_read(struct iwl_mvm *mvm, u32 gp2)
 {
 	/* If the difference is above the threshold, assume it's a wraparound.
@@ -122,6 +124,28 @@ iwl_mvm_get_crosstimestamp_fw(struct iwl_mvm *mvm, u32 *gp2, u64 *sys_time)
 	return ret;
 }
 
+static void iwl_mvm_phc_get_crosstimestamp_loop(struct iwl_mvm *mvm,
+						ktime_t *sys_time, u32 *gp2)
+{
+	u64 diff = 0, new_diff;
+	u64 tmp_sys_time;
+	u32 tmp_gp2;
+	int i;
+
+	for (i = 0; i < IWL_PTP_GET_CROSS_TS_NUM; i++) {
+		iwl_mvm_get_sync_time(mvm, CLOCK_REALTIME, &tmp_gp2, NULL,
+				      &tmp_sys_time);
+		new_diff = tmp_sys_time - ((u64)tmp_gp2 * NSEC_PER_USEC);
+		if (!diff || new_diff < diff) {
+			*sys_time = tmp_sys_time;
+			*gp2 = tmp_gp2;
+			diff = new_diff;
+			IWL_DEBUG_INFO(mvm, "PTP: new times: gp2=%u sys=%lld\n",
+				       *gp2, *sys_time);
+		}
+	}
+}
+
 static int
 iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
 			       struct system_device_crosststamp *xtstamp)
@@ -150,8 +174,7 @@ iwl_mvm_phc_get_crosstimestamp(struct ptp_clock_info *ptp,
 		if (ret)
 			goto out;
 	} else {
-		iwl_mvm_get_sync_time(mvm, CLOCK_REALTIME, &gp2, NULL,
-				      &sys_time);
+		iwl_mvm_phc_get_crosstimestamp_loop(mvm, &sys_time, &gp2);
 	}
 
 	gp2_ns = iwl_mvm_ptp_get_adj_time(mvm, (u64)gp2 * NSEC_PER_USEC);
-- 
cgit v1.2.3-70-g09d2


From f947b62c03b15d14349668384bb365aabe898650 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:09 +0200
Subject: wifi: iwlwifi: mvm: add start_ap() and join_ibss() callbacks for MLD
 mode

This is another patch in the series adding all the ops
for the new MLD ieee80211_ops.
The callbacks added here use the new MLD FW API

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.3eb485c359fc.I28be198ea9389083d5a01f68c92763722613ba9b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 76 ++++++++++++---------
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 78 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 45 +++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  4 ++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  2 +
 5 files changed, 175 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 99a51d519214..165358a4a97e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2669,13 +2669,54 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 		iwl_mvm_apply_fw_smps_request(vif);
 }
 
+bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif,
+				  int *ret)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int i;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	mvmvif->ap_assoc_sta_count = 0;
+
+	/* must be set before quota calculations */
+	mvmvif->ap_ibss_active = true;
+
+	/* send all the early keys to the device now */
+	for (i = 0; i < ARRAY_SIZE(mvmvif->ap_early_keys); i++) {
+		struct ieee80211_key_conf *key = mvmvif->ap_early_keys[i];
+
+		if (!key)
+			continue;
+
+		mvmvif->ap_early_keys[i] = NULL;
+
+		*ret = __iwl_mvm_mac_set_key(hw, SET_KEY, vif, NULL, key);
+		if (*ret)
+			return true;
+	}
+
+	if (vif->type == NL80211_IFTYPE_AP && !vif->p2p) {
+		iwl_mvm_vif_set_low_latency(mvmvif, true,
+					    LOW_LATENCY_VIF_TYPE);
+		iwl_mvm_send_low_latency_cmd(mvm, true, mvmvif->id);
+	}
+
+	/* power updated needs to be done before quotas */
+	iwl_mvm_power_update_mac(mvm);
+
+	return false;
+}
+
 static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 				 struct ieee80211_vif *vif,
 				 struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	int ret, i;
+	int ret;
 
 	mutex_lock(&mvm->mutex);
 
@@ -2692,8 +2733,6 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 	if (vif->type == NL80211_IFTYPE_AP)
 		iwl_mvm_mac_ctxt_recalc_tsf_id(mvm, vif);
 
-	mvmvif->ap_assoc_sta_count = 0;
-
 	/* Add the mac context */
 	ret = iwl_mvm_mac_ctxt_add(mvm, vif);
 	if (ret)
@@ -2739,35 +2778,12 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 		}
 	}
 
-	/* must be set before quota calculations */
-	mvmvif->ap_ibss_active = true;
-
-	/* send all the early keys to the device now */
-	for (i = 0; i < ARRAY_SIZE(mvmvif->ap_early_keys); i++) {
-		struct ieee80211_key_conf *key = mvmvif->ap_early_keys[i];
-
-		if (!key)
-			continue;
-
-		mvmvif->ap_early_keys[i] = NULL;
-
-		ret = __iwl_mvm_mac_set_key(hw, SET_KEY, vif, NULL, key);
-		if (ret)
-			goto out_quota_failed;
-	}
-
-	if (vif->type == NL80211_IFTYPE_AP && !vif->p2p) {
-		iwl_mvm_vif_set_low_latency(mvmvif, true,
-					    LOW_LATENCY_VIF_TYPE);
-		iwl_mvm_send_low_latency_cmd(mvm, true, mvmvif->id);
-	}
-
-	/* power updated needs to be done before quotas */
-	iwl_mvm_power_update_mac(mvm);
+	if (iwl_mvm_start_ap_ibss_common(hw, vif, &ret))
+		goto out_failed;
 
 	ret = iwl_mvm_update_quotas(mvm, false, NULL);
 	if (ret)
-		goto out_quota_failed;
+		goto out_failed;
 
 	/* Need to update the P2P Device MAC (only GO, IBSS is single vif) */
 	if (vif->p2p && mvm->p2p_device_vif)
@@ -2783,7 +2799,7 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 
 	goto out_unlock;
 
-out_quota_failed:
+out_failed:
 	iwl_mvm_power_update_mac(mvm);
 	mvmvif->ap_ibss_active = false;
 	iwl_mvm_send_rm_bcast_sta(mvm, vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index f402e235525c..866415dc40bc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -245,9 +245,87 @@ static void iwl_mvm_mld_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	__iwl_mvm_mld_unassign_vif_chanctx(mvm, vif, ctx, false);
 	mutex_unlock(&mvm->mutex);
 }
+
+static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	mutex_lock(&mvm->mutex);
+
+	/* Send the beacon template */
+	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif);
+	if (ret)
+		goto out_unlock;
+
+	/* No need to re-calculate the tsf_is, as it was offloaded */
+
+	/* Add the mac context */
+	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
+	if (ret)
+		goto out_unlock;
+
+	/* Add link and activate it */
+	ret = iwl_mvm_add_link(mvm, vif);
+	if (ret)
+		goto out_remove_mac;
+
+	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
+				   true);
+	if (ret)
+		goto out_remove_link;
+
+	ret = iwl_mvm_mld_add_mcast_sta(mvm, vif);
+	if (ret)
+		goto out_remove_link;
+
+	/* Send the bcast station. At this stage the TBTT and DTIM time
+	 * events are added and applied to the scheduler
+	 */
+	ret = iwl_mvm_mld_add_bcast_sta(mvm, vif);
+	if (ret)
+		goto out_rm_mcast;
+
+	if (iwl_mvm_start_ap_ibss_common(hw, vif, &ret))
+		goto out_failed;
+
+	/* Need to update the P2P Device MAC (only GO, IBSS is single vif) */
+	if (vif->p2p && mvm->p2p_device_vif)
+		iwl_mvm_mld_mac_ctxt_changed(mvm, mvm->p2p_device_vif, false);
+
+	iwl_mvm_bt_coex_vif_change(mvm);
+
+	/* we don't support TDLS during DCM */
+	if (iwl_mvm_phy_ctx_count(mvm) > 1)
+		iwl_mvm_teardown_tdls_peers(mvm);
+
+	iwl_mvm_ftm_restart_responder(mvm, vif);
+
+	goto out_unlock;
+
+out_failed:
+	iwl_mvm_power_update_mac(mvm);
+	mvmvif->ap_ibss_active = false;
+	iwl_mvm_mld_rm_bcast_sta(mvm, vif);
+out_rm_mcast:
+	iwl_mvm_mld_rm_mcast_sta(mvm, vif);
+out_remove_link:
+	/* Link needs to be deactivated before removal */
+	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_remove_link(mvm, vif);
+out_remove_mac:
+	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
+out_unlock:
+	mutex_unlock(&mvm->mutex);
+	return ret;
+}
+
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
 	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
 	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
+	.join_ibss = iwl_mvm_mld_start_ap_ibss,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 4ee6f9b250d3..d81fc9207ef0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -176,6 +176,38 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 				       IWL_MAX_TID_COUNT, &wdg_timeout);
 }
 
+/* Allocate a new station entry for the broadcast station to the given vif,
+ * and send it to the FW.
+ * Note that each AP/GO mac should have its own multicast station.
+ */
+int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_int_sta *msta = &mvmvif->mcast_sta;
+	static const u8 _maddr[] = {0x03, 0x00, 0x00, 0x00, 0x00, 0x00};
+	const u8 *maddr = _maddr;
+	unsigned int timeout = iwl_mvm_get_wd_timeout(mvm, vif, false, false);
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (WARN_ON(vif->type != NL80211_IFTYPE_AP &&
+		    vif->type != NL80211_IFTYPE_ADHOC))
+		return -EOPNOTSUPP;
+
+	/* In IBSS, ieee80211_check_queues() sets the cab_queue to be
+	 * invalid, so make sure we use the queue we want.
+	 * Note that this is done here as we want to avoid making DQA
+	 * changes in mac80211 layer.
+	 */
+	if (vif->type == NL80211_IFTYPE_ADHOC)
+		mvmvif->cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
+
+	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvmvif->cab_queue,
+				       vif->type, STATION_TYPE_MCAST,
+				       mvmvif->phy_ctxt->id, maddr, 0,
+				       &timeout);
+}
+
 /* Allocate a new station entry for the sniffer station to the given vif,
  * and send it to the FW.
  */
@@ -272,6 +304,19 @@ int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 				      IWL_MAX_TID_COUNT, queueptr);
 }
 
+/* Send the FW a request to remove the station from it's internal data
+ * structures, and in addition remove it from the local data structure.
+ */
+int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	lockdep_assert_held(&mvm->mutex);
+
+	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->mcast_sta, true, 0,
+				      &mvmvif->cab_queue);
+}
+
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	lockdep_assert_held(&mvm->mutex);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 2cd86237bc14..a21ccbfd017c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1766,6 +1766,10 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			 u32 changes, bool active);
 int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 
+/* AP and IBSS */
+bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif, int *ret);
+
 /* Quota management */
 static inline size_t iwl_mvm_quota_cmd_size(struct iwl_mvm *mvm)
 {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index f0d22e9a117b..432768bfc433 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -563,8 +563,10 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 /* STA */
 int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 
 /* Queues */
 void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From fd1a54c14bd93166784c2adfaf1f4f22c597e963 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:10 +0200
Subject: wifi: iwlwifi: mvm: add stop_ap() and leave_ibss() callbacks for MLD
 mode

This is another patch in the series adding all the ops
for the new MLD ieee80211_ops.
The callbacks added here use the new MLD FW API

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.42b0d4726b8d.I0755baace47c0ab1d9d70137448125d3140ef3af@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 22 +++++++++++-----
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 30 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  2 ++
 3 files changed, 48 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 165358a4a97e..c6eb4cb44a09 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2826,16 +2826,15 @@ static int iwl_mvm_start_ibss(struct ieee80211_hw *hw,
 	return iwl_mvm_start_ap_ibss(hw, vif, &vif->bss_conf);
 }
 
-static void iwl_mvm_stop_ap_ibss(struct ieee80211_hw *hw,
-				 struct ieee80211_vif *vif,
-				 struct ieee80211_bss_conf *link_conf)
+/* Common part for MLD and non-MLD ops */
+void iwl_mvm_stop_ap_ibss_common(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif)
 {
-	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	iwl_mvm_prepare_mac_removal(mvm, vif);
+	lockdep_assert_held(&mvm->mutex);
 
-	mutex_lock(&mvm->mutex);
+	iwl_mvm_prepare_mac_removal(mvm, vif);
 
 	/* Handle AP stop while in CSA */
 	if (rcu_access_pointer(mvm->csa_vif) == vif) {
@@ -2860,6 +2859,17 @@ static void iwl_mvm_stop_ap_ibss(struct ieee80211_hw *hw,
 	}
 
 	iwl_mvm_bt_coex_vif_change(mvm);
+}
+
+static void iwl_mvm_stop_ap_ibss(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_bss_conf *link_conf)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	mutex_lock(&mvm->mutex);
+
+	iwl_mvm_stop_ap_ibss_common(mvm, vif);
 
 	/* Need to update the P2P Device MAC (only GO, IBSS is single vif) */
 	if (vif->p2p && mvm->p2p_device_vif)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 866415dc40bc..f800bf0b6d90 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -322,10 +322,40 @@ out_unlock:
 	return ret;
 }
 
+static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	mutex_lock(&mvm->mutex);
+
+	iwl_mvm_stop_ap_ibss_common(mvm, vif);
+
+	/* Need to update the P2P Device MAC (only GO, IBSS is single vif) */
+	if (vif->p2p && mvm->p2p_device_vif)
+		iwl_mvm_mld_mac_ctxt_changed(mvm, mvm->p2p_device_vif, false);
+
+	iwl_mvm_ftm_responder_clear(mvm, vif);
+
+	iwl_mvm_mld_rm_bcast_sta(mvm, vif);
+	iwl_mvm_mld_rm_mcast_sta(mvm, vif);
+
+	/* Link needs to be deactivated before removal */
+	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_remove_link(mvm, vif);
+
+	iwl_mvm_power_update_mac(mvm);
+
+	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
+
+	mutex_unlock(&mvm->mutex);
+}
+
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
 	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
 	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
 	.join_ibss = iwl_mvm_mld_start_ap_ibss,
+	.leave_ibss = iwl_mvm_mld_stop_ap_ibss,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index a21ccbfd017c..a03c2e5bbc25 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1769,6 +1769,8 @@ int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 /* AP and IBSS */
 bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
 				  struct ieee80211_vif *vif, int *ret);
+void iwl_mvm_stop_ap_ibss_common(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif);
 
 /* Quota management */
 static inline size_t iwl_mvm_quota_cmd_size(struct iwl_mvm *mvm)
-- 
cgit v1.2.3-70-g09d2


From f5034bcf6868bf5c2de272819aa6de23f5d588f7 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:11 +0200
Subject: wifi: iwlwifi: mvm: Don't send MAC CTXT cmd after deauthorization

We used to send a MAC CTXT cmd to ask the FW to not pass MCAST frames
if we're associated but not authorized, because we don't have the
keys in that stage, and after authorization - we sent the cmd again
to ask the FW to pass MCAST, as we have the keys now.
The patch linked below was changing this strategy to always allow
MCAST frames, and if we're not authorized - the driver will drop them.
But we're still sending the MAC CTXT cmd after deaouthorization even
though we don't tell the FW to not pass MCAST frames anymore.
Basically we don't tell the FW anything new with this cmd.
Fix this by not sending MAC CTXT command after deauthorization.
For authorization we're sending the cmd to configure other changes too,
so keep it.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.11b3481bc497.I9672acff9cfc00e7e1a187e7178caa3a1911a1b5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index c6eb4cb44a09..d5708f9ebbe9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3547,11 +3547,6 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 
 			mvmvif->authorized = 1;
 
-			/*
-			 * Now that the station is authorized, i.e., keys were already
-			 * installed, need to indicate to the FW that
-			 * multicast data frames can be forwarded to the driver
-			 */
 			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
 			iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
 		}
@@ -3566,12 +3561,10 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 		iwl_mvm_rs_rate_init(mvm, sta, mvmvif->phy_ctxt->channel->band,
 				     false);
 		if (!sta->tdls) {
-			/* Multicast data frames are no longer allowed */
-			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
-
 			/*
-			 * Set this after the above iwl_mvm_mac_ctxt_changed()
-			 * to avoid sending high prio again for a little time.
+			 * Set this but don't call iwl_mvm_mac_ctxt_changed()
+			 * yet to avoid sending high prio again for a little
+			 * time.
 			 */
 			mvmvif->authorized = 0;
 
-- 
cgit v1.2.3-70-g09d2


From 4df6a07551c8a70460b590e534785d5e97d1d528 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:12 +0200
Subject: wifi: iwlwifi: mvm: refactor iwl_mvm_cfg_he_sta()

This function sends the STA_HE_CTXT_CMD, which won't be used in the
new MLD API. Instead, a part of the fields of this command
will be in the new STA cmd. Put the parts that are common to both
commands in functions, which will later be used in sending the new
STA cmd.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.19b9b6c346ab.I0b8512eb64d8f03ff83879bafe9707f897d5b3c6@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 287 +++++++++++-----------
 1 file changed, 147 insertions(+), 140 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d5708f9ebbe9..6e059027517f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2023,9 +2023,9 @@ static void iwl_mvm_set_pkt_ext_from_he_ppe(struct iwl_mvm *mvm,
 			  inheritance);
 }
 
-static void iwl_mvm_set_pkt_ext_from_nominal_padding(struct iwl_he_pkt_ext_v2 *pkt_ext,
-						     u8 nominal_padding,
-						     u32 *flags)
+static int
+iwl_mvm_set_pkt_ext_from_nominal_padding(struct iwl_he_pkt_ext_v2 *pkt_ext,
+					 u8 nominal_padding)
 {
 	int low_th = -1;
 	int high_th = -1;
@@ -2048,21 +2048,22 @@ static void iwl_mvm_set_pkt_ext_from_nominal_padding(struct iwl_he_pkt_ext_v2 *p
 		break;
 	}
 
+	if (low_th < 0 || high_th < 0)
+		return -EINVAL;
+
 	/* Set the PPE thresholds accordingly */
-	if (low_th >= 0 && high_th >= 0) {
-		for (i = 0; i < MAX_HE_SUPP_NSS; i++) {
-			u8 bw;
+	for (i = 0; i < MAX_HE_SUPP_NSS; i++) {
+		u8 bw;
 
-			for (bw = 0;
-			     bw < ARRAY_SIZE(pkt_ext->pkt_ext_qam_th[i]);
-			     bw++) {
-				pkt_ext->pkt_ext_qam_th[i][bw][0] = low_th;
-				pkt_ext->pkt_ext_qam_th[i][bw][1] = high_th;
-			}
+		for (bw = 0;
+			bw < ARRAY_SIZE(pkt_ext->pkt_ext_qam_th[i]);
+			bw++) {
+			pkt_ext->pkt_ext_qam_th[i][bw][0] = low_th;
+			pkt_ext->pkt_ext_qam_th[i][bw][1] = high_th;
 		}
-
-		*flags |= STA_CTXT_HE_PACKET_EXT;
 	}
+
+	return 0;
 }
 
 static void iwl_mvm_get_optimal_ppe_info(struct iwl_he_pkt_ext_v2 *pkt_ext,
@@ -2090,6 +2091,103 @@ static void iwl_mvm_get_optimal_ppe_info(struct iwl_he_pkt_ext_v2 *pkt_ext,
 	}
 }
 
+/* Set the pkt_ext field according to PPE Thresholds element */
+static int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm,
+				   struct ieee80211_sta *sta,
+				   struct iwl_he_pkt_ext_v2 *pkt_ext)
+{
+	u8 nominal_padding;
+	int i, ret = 0;
+
+	/* Initialize the PPE thresholds to "None" (7), as described in Table
+	 * 9-262ac of 80211.ax/D3.0.
+	 */
+	memset(pkt_ext, IWL_HE_PKT_EXT_NONE,
+	       sizeof(struct iwl_he_pkt_ext_v2));
+
+	if (sta->deflink.eht_cap.has_eht) {
+		nominal_padding =
+			u8_get_bits(sta->deflink.eht_cap.eht_cap_elem.phy_cap_info[5],
+				    IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK);
+
+		/* If PPE Thresholds exists, parse them into a FW-familiar
+		 * format.
+		 */
+		if (sta->deflink.eht_cap.eht_cap_elem.phy_cap_info[5] &
+		    IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT) {
+			u8 nss = (sta->deflink.eht_cap.eht_ppe_thres[0] &
+				IEEE80211_EHT_PPE_THRES_NSS_MASK) + 1;
+			u8 *ppe = &sta->deflink.eht_cap.eht_ppe_thres[0];
+			u8 ru_index_bitmap =
+				u16_get_bits(*ppe,
+					     IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK);
+			 /* Starting after PPE header */
+			u8 ppe_pos_bit = IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE;
+
+			iwl_mvm_parse_ppe(mvm, pkt_ext, nss, ru_index_bitmap,
+					  ppe, ppe_pos_bit, true);
+		/* EHT PPE Thresholds doesn't exist - set the API according
+		 * to HE PPE Tresholds
+		 */
+		} else if (sta->deflink.he_cap.he_cap_elem.phy_cap_info[6] &
+			   IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
+			/* Even though HE Capabilities IE doesn't contain PPE
+			 * Thresholds for BW 320Mhz, thresholds for this BW will
+			 * be filled in with the same values as 160Mhz, due to
+			 * the inheritance, as required.
+			 */
+			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, sta, pkt_ext,
+							true);
+
+			/* According to the requirements, for MCSs 12-13 the
+			 * maximum value between HE PPE Threshold and Common
+			 * Nominal Packet Padding needs to be taken
+			 */
+			iwl_mvm_get_optimal_ppe_info(pkt_ext, nominal_padding);
+
+		/* if PPE Thresholds doesn't present in both EHT IE and HE IE -
+		 * take the Thresholds from Common Nominal Packet Padding field
+		 */
+		} else {
+			ret = iwl_mvm_set_pkt_ext_from_nominal_padding(pkt_ext,
+								       nominal_padding);
+		}
+	} else if (sta->deflink.he_cap.has_he) {
+		/* If PPE Thresholds exist, parse them into a FW-familiar format. */
+		if (sta->deflink.he_cap.he_cap_elem.phy_cap_info[6] &
+			IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
+			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, sta, pkt_ext,
+							false);
+		/* PPE Thresholds doesn't exist - set the API PPE values
+		 * according to Common Nominal Packet Padding field.
+		 */
+		} else {
+			nominal_padding =
+				u8_get_bits(sta->deflink.he_cap.he_cap_elem.phy_cap_info[9],
+					    IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK);
+			if (nominal_padding != IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED)
+				ret = iwl_mvm_set_pkt_ext_from_nominal_padding(pkt_ext,
+									       nominal_padding);
+		}
+	}
+
+	for (i = 0; i < MAX_HE_SUPP_NSS; i++) {
+		int bw;
+
+		for (bw = 0;
+		     bw < ARRAY_SIZE(*pkt_ext->pkt_ext_qam_th[i]);
+		     bw++) {
+			u8 *qam_th =
+				&pkt_ext->pkt_ext_qam_th[i][bw][0];
+
+			IWL_DEBUG_HT(mvm,
+				     "PPE table: nss[%d] bw[%d] PPET8 = %d, PPET16 = %d\n",
+				     i, bw, qam_th[0], qam_th[1]);
+		}
+	}
+	return ret;
+}
+
 /*
  * This function sets the MU EDCA parameters ans returns whether MU EDCA
  * is enabled or not
@@ -2141,6 +2239,37 @@ bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 			       IEEE80211_HE_MAC_CAP2_ACK_EN));
 }
 
+static __le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta)
+{
+	u8 *mac_cap_info = &sta->deflink.he_cap.he_cap_elem.mac_cap_info[0];
+	__le32 htc_flags = 0;
+
+	if (mac_cap_info[0] & IEEE80211_HE_MAC_CAP0_HTC_HE)
+		htc_flags |= cpu_to_le32(IWL_HE_HTC_SUPPORT);
+	if ((mac_cap_info[1] & IEEE80211_HE_MAC_CAP1_LINK_ADAPTATION) ||
+	    (mac_cap_info[2] & IEEE80211_HE_MAC_CAP2_LINK_ADAPTATION)) {
+		u8 link_adap =
+			((mac_cap_info[2] &
+			  IEEE80211_HE_MAC_CAP2_LINK_ADAPTATION) << 1) +
+			 (mac_cap_info[1] &
+			  IEEE80211_HE_MAC_CAP1_LINK_ADAPTATION);
+
+		if (link_adap == 2)
+			htc_flags |=
+				cpu_to_le32(IWL_HE_HTC_LINK_ADAP_UNSOLICITED);
+		else if (link_adap == 3)
+			htc_flags |= cpu_to_le32(IWL_HE_HTC_LINK_ADAP_BOTH);
+	}
+	if (mac_cap_info[2] & IEEE80211_HE_MAC_CAP2_BSR)
+		htc_flags |= cpu_to_le32(IWL_HE_HTC_BSR_SUPP);
+	if (mac_cap_info[3] & IEEE80211_HE_MAC_CAP3_OMI_CONTROL)
+		htc_flags |= cpu_to_le32(IWL_HE_HTC_OMI_SUPP);
+	if (mac_cap_info[4] & IEEE80211_HE_MAC_CAP4_BQR)
+		htc_flags |= cpu_to_le32(IWL_HE_HTC_BQR_SUPP);
+
+	return htc_flags;
+}
+
 static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 			       struct ieee80211_vif *vif, u8 sta_id)
 {
@@ -2161,7 +2290,6 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 	u32 flags;
 	int i;
 	void *cmd;
-	u8 nominal_padding;
 
 	if (!fw_has_api(&mvm->fw->ucode_capa, IWL_UCODE_TLV_API_MBSSID_HE))
 		ver = 1;
@@ -2206,132 +2334,11 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 		flags |= STA_CTXT_HE_RU_2MHZ_BLOCK;
 
 	/* HTC flags */
-	if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[0] &
-	    IEEE80211_HE_MAC_CAP0_HTC_HE)
-		sta_ctxt_cmd.htc_flags |= cpu_to_le32(IWL_HE_HTC_SUPPORT);
-	if ((sta->deflink.he_cap.he_cap_elem.mac_cap_info[1] &
-	      IEEE80211_HE_MAC_CAP1_LINK_ADAPTATION) ||
-	    (sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] &
-	      IEEE80211_HE_MAC_CAP2_LINK_ADAPTATION)) {
-		u8 link_adap =
-			((sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] &
-			  IEEE80211_HE_MAC_CAP2_LINK_ADAPTATION) << 1) +
-			 (sta->deflink.he_cap.he_cap_elem.mac_cap_info[1] &
-			  IEEE80211_HE_MAC_CAP1_LINK_ADAPTATION);
-
-		if (link_adap == 2)
-			sta_ctxt_cmd.htc_flags |=
-				cpu_to_le32(IWL_HE_HTC_LINK_ADAP_UNSOLICITED);
-		else if (link_adap == 3)
-			sta_ctxt_cmd.htc_flags |=
-				cpu_to_le32(IWL_HE_HTC_LINK_ADAP_BOTH);
-	}
-	if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] & IEEE80211_HE_MAC_CAP2_BSR)
-		sta_ctxt_cmd.htc_flags |= cpu_to_le32(IWL_HE_HTC_BSR_SUPP);
-	if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[3] &
-	    IEEE80211_HE_MAC_CAP3_OMI_CONTROL)
-		sta_ctxt_cmd.htc_flags |= cpu_to_le32(IWL_HE_HTC_OMI_SUPP);
-	if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[4] & IEEE80211_HE_MAC_CAP4_BQR)
-		sta_ctxt_cmd.htc_flags |= cpu_to_le32(IWL_HE_HTC_BQR_SUPP);
+	sta_ctxt_cmd.htc_flags = iwl_mvm_get_sta_htc_flags(sta);
 
-	/*
-	 * Initialize the PPE thresholds to "None" (7), as described in Table
-	 * 9-262ac of 80211.ax/D3.0.
-	 */
-	memset(&sta_ctxt_cmd.pkt_ext, IWL_HE_PKT_EXT_NONE,
-	       sizeof(sta_ctxt_cmd.pkt_ext));
-
-	if (sta->deflink.eht_cap.has_eht) {
-		nominal_padding =
-			u8_get_bits(sta->deflink.eht_cap.eht_cap_elem.phy_cap_info[5],
-				    IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK);
-
-		/* If PPE Thresholds exists, parse them into a FW-familiar format. */
-		if (sta->deflink.eht_cap.eht_cap_elem.phy_cap_info[5] &
-		    IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT) {
-			u8 nss = (sta->deflink.eht_cap.eht_ppe_thres[0] &
-				IEEE80211_EHT_PPE_THRES_NSS_MASK) + 1;
-			u8 *ppe = &sta->deflink.eht_cap.eht_ppe_thres[0];
-			u8 ru_index_bitmap =
-				u16_get_bits(*ppe,
-					     IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK);
-			 /* Starting after PPE header */
-			u8 ppe_pos_bit = IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE;
-
-			iwl_mvm_parse_ppe(mvm,
-					  &sta_ctxt_cmd.pkt_ext,
-					  nss, ru_index_bitmap, ppe,
-					  ppe_pos_bit, true);
-			flags |= STA_CTXT_HE_PACKET_EXT;
-		/* EHT PPE Thresholds doesn't exist - set the API according to HE PPE Tresholds*/
-		} else if (sta->deflink.he_cap.he_cap_elem.phy_cap_info[6] &
-			   IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
-			struct iwl_he_pkt_ext_v2 *pkt_ext =
-				&sta_ctxt_cmd.pkt_ext;
-
-			/*
-			* Even though HE Capabilities IE doesn't contain PPE
-			* Thresholds for BW 320Mhz, thresholds for this BW will
-			* be filled in with the same values as 160Mhz, due to
-			* the inheritance, as required.
-			*/
-			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, sta, pkt_ext,
-							true);
-
-			/*
-			* According to the requirements, for MCSs 12-13 the maximum value between
-			* HE PPE Threshold and Common Nominal Packet Padding needs to be taken
-			*/
-			iwl_mvm_get_optimal_ppe_info(pkt_ext, nominal_padding);
-
-			flags |= STA_CTXT_HE_PACKET_EXT;
-
-		/*
-		* if PPE Thresholds doesn't present in both EHT IE and HE IE -
-		* take the Thresholds from Common Nominal Packet Padding field
-		*/
-		} else {
-			iwl_mvm_set_pkt_ext_from_nominal_padding(&sta_ctxt_cmd.pkt_ext,
-								 nominal_padding,
-								 &flags);
-		}
-	} else if (sta->deflink.he_cap.has_he) {
-		/* If PPE Thresholds exist, parse them into a FW-familiar format. */
-		if (sta->deflink.he_cap.he_cap_elem.phy_cap_info[6] &
-			IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
-			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, sta,
-							&sta_ctxt_cmd.pkt_ext,
-							false);
-			flags |= STA_CTXT_HE_PACKET_EXT;
-		/*
-		* PPE Thresholds doesn't exist - set the API PPE values
-		* according to Common Nominal Packet Padding field.
-		*/
-		} else {
-			nominal_padding =
-				u8_get_bits(sta->deflink.he_cap.he_cap_elem.phy_cap_info[9],
-					    IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK);
-			if (nominal_padding != IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED)
-				iwl_mvm_set_pkt_ext_from_nominal_padding(&sta_ctxt_cmd.pkt_ext,
-									 nominal_padding,
-									 &flags);
-		}
-	}
-
-	for (i = 0; i < MAX_HE_SUPP_NSS; i++) {
-		int bw;
-
-		for (bw = 0;
-		     bw < ARRAY_SIZE(sta_ctxt_cmd.pkt_ext.pkt_ext_qam_th[i]);
-		     bw++) {
-			u8 *qam_th =
-				&sta_ctxt_cmd.pkt_ext.pkt_ext_qam_th[i][bw][0];
-
-			IWL_DEBUG_HT(mvm,
-				     "PPE table: nss[%d] bw[%d] PPET8 = %d, PPET16 = %d\n",
-				     i, bw, qam_th[0], qam_th[1]);
-		}
-	}
+	/* PPE Thresholds */
+	if (!iwl_mvm_set_sta_pkt_ext(mvm, sta, &sta_ctxt_cmd.pkt_ext))
+		flags |= STA_CTXT_HE_PACKET_EXT;
 
 	if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] &
 	    IEEE80211_HE_MAC_CAP2_32BIT_BA_BITMAP)
-- 
cgit v1.2.3-70-g09d2


From 2ce1c07a8a54213d60acef267688b1c3a75f2f97 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:13 +0200
Subject: wifi: iwlwifi: mvm: refactor iwl_mvm_sta

As the new MLD API is introduced, we have a new enumeration
for the different station types. Since struct iwl_mvm_sta's
sta_type field will be in use for both new and old API,
change its type to a unified one, so it can hold both.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.c8b86d8ddd3e.I9581235860b91654b4f6c3a8797c777702690998@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 432768bfc433..a0c13c409ccd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -386,7 +386,7 @@ struct iwl_mvm_sta {
 	u32 mac_id_n_color;
 	u16 tid_disable_agg;
 	u16 max_agg_bufsize;
-	enum iwl_sta_type sta_type;
+	u8 sta_type;
 	enum ieee80211_sta_state sta_state;
 	bool bt_reduced_txpower;
 	bool next_status_eosp;
-- 
cgit v1.2.3-70-g09d2


From 77b6a2e5f49bb3a4756a4b82f24221e62fa6ea8c Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:14 +0200
Subject: wifi: iwlwifi: mvm: refactor iwl_mvm_sta_send_to_fw()

We have a new STA cmd as part of the new MLD API. There are some
parts of sending the STA cmd which are common to both the old and
the new one. Put this parts in functions which will later be used
to send the new STA cmd.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.4da940bd7384.I3a66990fbffe9611b5e41f3686c2aff37ba2eb56@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 116 ++++++++++++++++-----------
 1 file changed, 67 insertions(+), 49 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 1594221abdba..a40182e87f55 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -52,6 +52,68 @@ static int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm,
 	return IWL_MVM_INVALID_STA;
 }
 
+/* Calculate the ampdu density and max size */
+static u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta,
+				      u32 *_agg_size)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	u32 agg_size = 0, mpdu_dens = 0;
+
+	if (sta->deflink.ht_cap.ht_supported)
+		mpdu_dens = sta->deflink.ht_cap.ampdu_density;
+
+	if (mvm_sta->vif->bss_conf.chandef.chan->band == NL80211_BAND_6GHZ) {
+		mpdu_dens = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+					  IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START);
+		agg_size = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+					 IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP);
+	} else if (sta->deflink.vht_cap.vht_supported) {
+		agg_size = sta->deflink.vht_cap.cap &
+			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK;
+		agg_size >>=
+			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT;
+	} else if (sta->deflink.ht_cap.ht_supported) {
+		agg_size = sta->deflink.ht_cap.ampdu_factor;
+	}
+
+	/* D6.0 10.12.2 A-MPDU length limit rules
+	 * A STA indicates the maximum length of the A-MPDU preEOF padding
+	 * that it can receive in an HE PPDU in the Maximum A-MPDU Length
+	 * Exponent field in its HT Capabilities, VHT Capabilities,
+	 * and HE 6 GHz Band Capabilities elements (if present) and the
+	 * Maximum AMPDU Length Exponent Extension field in its HE
+	 * Capabilities element
+	 */
+	if (sta->deflink.he_cap.has_he)
+		agg_size +=
+		    u8_get_bits(sta->deflink.he_cap.he_cap_elem.mac_cap_info[3],
+				IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK);
+
+	/* Limit to max A-MPDU supported by FW */
+	if (agg_size > (STA_FLG_MAX_AGG_SIZE_4M >> STA_FLG_MAX_AGG_SIZE_SHIFT))
+		agg_size = (STA_FLG_MAX_AGG_SIZE_4M >>
+			    STA_FLG_MAX_AGG_SIZE_SHIFT);
+
+	*_agg_size = agg_size;
+	return mpdu_dens;
+}
+
+static u8 iwl_mvm_get_sta_uapsd_acs(struct ieee80211_sta *sta)
+{
+	u8 uapsd_acs = 0;
+
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BK)
+		uapsd_acs |= BIT(AC_BK);
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BE)
+		uapsd_acs |= BIT(AC_BE);
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VI)
+		uapsd_acs |= BIT(AC_VI);
+	if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VO)
+		uapsd_acs |= BIT(AC_VO);
+
+	return uapsd_acs | uapsd_acs << 4;
+}
+
 /* send station add/update command to firmware */
 int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			   bool update, unsigned int flags)
@@ -135,68 +197,24 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		break;
 	}
 
-	if (sta->deflink.ht_cap.ht_supported) {
-		add_sta_cmd.station_flags_msk |=
-			cpu_to_le32(STA_FLG_MAX_AGG_SIZE_MSK |
-				    STA_FLG_AGG_MPDU_DENS_MSK);
-
-		mpdu_dens = sta->deflink.ht_cap.ampdu_density;
-	}
-
-	if (mvm_sta->vif->bss_conf.chandef.chan->band == NL80211_BAND_6GHZ) {
+	if (sta->deflink.ht_cap.ht_supported ||
+	    mvm_sta->vif->bss_conf.chandef.chan->band == NL80211_BAND_6GHZ)
 		add_sta_cmd.station_flags_msk |=
 			cpu_to_le32(STA_FLG_MAX_AGG_SIZE_MSK |
 				    STA_FLG_AGG_MPDU_DENS_MSK);
 
-		mpdu_dens = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
-					  IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START);
-		agg_size = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
-					 IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP);
-	} else if (sta->deflink.vht_cap.vht_supported) {
-		agg_size = sta->deflink.vht_cap.cap &
-			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK;
-		agg_size >>=
-			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT;
-	} else if (sta->deflink.ht_cap.ht_supported) {
-		agg_size = sta->deflink.ht_cap.ampdu_factor;
-	}
-
-	/* D6.0 10.12.2 A-MPDU length limit rules
-	 * A STA indicates the maximum length of the A-MPDU preEOF padding
-	 * that it can receive in an HE PPDU in the Maximum A-MPDU Length
-	 * Exponent field in its HT Capabilities, VHT Capabilities,
-	 * and HE 6 GHz Band Capabilities elements (if present) and the
-	 * Maximum AMPDU Length Exponent Extension field in its HE
-	 * Capabilities element
-	 */
-	if (sta->deflink.he_cap.has_he)
-		agg_size += u8_get_bits(sta->deflink.he_cap.he_cap_elem.mac_cap_info[3],
-					IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK);
-
-	/* Limit to max A-MPDU supported by FW */
-	if (agg_size > (STA_FLG_MAX_AGG_SIZE_4M >> STA_FLG_MAX_AGG_SIZE_SHIFT))
-		agg_size = (STA_FLG_MAX_AGG_SIZE_4M >>
-			    STA_FLG_MAX_AGG_SIZE_SHIFT);
-
+	mpdu_dens = iwl_mvm_get_sta_ampdu_dens(sta, &agg_size);
 	add_sta_cmd.station_flags |=
 		cpu_to_le32(agg_size << STA_FLG_MAX_AGG_SIZE_SHIFT);
 	add_sta_cmd.station_flags |=
 		cpu_to_le32(mpdu_dens << STA_FLG_AGG_MPDU_DENS_SHIFT);
+
 	if (mvm_sta->sta_state >= IEEE80211_STA_ASSOC)
 		add_sta_cmd.assoc_id = cpu_to_le16(sta->aid);
 
 	if (sta->wme) {
 		add_sta_cmd.modify_mask |= STA_MODIFY_UAPSD_ACS;
-
-		if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BK)
-			add_sta_cmd.uapsd_acs |= BIT(AC_BK);
-		if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BE)
-			add_sta_cmd.uapsd_acs |= BIT(AC_BE);
-		if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VI)
-			add_sta_cmd.uapsd_acs |= BIT(AC_VI);
-		if (sta->uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VO)
-			add_sta_cmd.uapsd_acs |= BIT(AC_VO);
-		add_sta_cmd.uapsd_acs |= add_sta_cmd.uapsd_acs << 4;
+		add_sta_cmd.uapsd_acs = iwl_mvm_get_sta_uapsd_acs(sta);
 		add_sta_cmd.sp_length = sta->max_sp ? sta->max_sp * 2 : 128;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 6a8dee1391ff1d4942e8f09b5fb8c1087cd6a6c7 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:15 +0200
Subject: wifi: iwlwifi: mvm: remove not needed initializations

In iwl_mvm_add_sta() we're initializing the new allocated mvm_sta.
We are setting some fields to zero even though it was allocated with
kzalloc, for the case of HW restart. But in such a case we will never
get to this initializations due to the goto statement that we have in
this case. Remove these initializations.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.c0b4d1e986a5.I6959bf1aca74c865e3c1edbf711f5fe8fb8c0c9f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index a40182e87f55..c22ba1ea42f6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1702,20 +1702,12 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 		mvm_sta->max_agg_bufsize = LINK_QUAL_AGG_FRAME_LIMIT_DEF;
 	else
 		mvm_sta->max_agg_bufsize = LINK_QUAL_AGG_FRAME_LIMIT_GEN2_DEF;
-	mvm_sta->tx_protection = 0;
 	mvm_sta->tt_tx_protection = false;
 	mvm_sta->sta_type = sta->tdls ? IWL_STA_TDLS_LINK : IWL_STA_LINK;
 
-	/* HW restart, don't assume the memory has been zeroed */
 	mvm_sta->tid_disable_agg = 0xffff; /* No aggs at first */
-	mvm_sta->tfd_queue_msk = 0;
 
-	/* for HW restart - reset everything but the sequence number */
 	for (i = 0; i <= IWL_MAX_TID_COUNT; i++) {
-		u16 seq = mvm_sta->tid_data[i].seq_number;
-		memset(&mvm_sta->tid_data[i], 0, sizeof(mvm_sta->tid_data[i]));
-		mvm_sta->tid_data[i].seq_number = seq;
-
 		/*
 		 * Mark all queues for this STA as unallocated and defer TX
 		 * frames until the queue is allocated
@@ -1732,8 +1724,6 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 		atomic_set(&mvmtxq->tx_request, 0);
 	}
 
-	mvm_sta->agg_tids = 0;
-
 	if (iwl_mvm_has_new_rx_api(mvm) &&
 	    !test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
 		int q;
-- 
cgit v1.2.3-70-g09d2


From 69aef848052b8f8a5d8348606809cd3d92db9b69 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 20 Mar 2023 12:33:16 +0200
Subject: wifi: iwlwifi: mvm: refactor iwl_mvm_add_sta(), iwl_mvm_rm_sta()

As we have a new MLD STA cmd, there will be a different function to
add/remove a station in MLD mode. But both functions will share a
common part. Put this part in a separate function which will be used
later in adding/removing a sta in MLD mode.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.d01e0c9ccdc3.I5e9e27c3b363b36209a0ff960d2e59708e7ff0bf@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 182 ++++++++++++++++-----------
 1 file changed, 106 insertions(+), 76 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index c22ba1ea42f6..7d9d4e85b548 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1648,52 +1648,17 @@ static int iwl_mvm_add_int_sta_common(struct iwl_mvm *mvm,
 	return ret;
 }
 
-int iwl_mvm_add_sta(struct iwl_mvm *mvm,
-		    struct ieee80211_vif *vif,
-		    struct ieee80211_sta *sta)
+/* Initialize driver data of a new sta */
+static int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, int sta_id, u8 sta_type)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_rxq_dup_data *dup_data;
-	int i, ret, sta_id;
-	bool sta_update = false;
-	unsigned int sta_flags = 0;
+	int i, ret = 0;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
-		sta_id = iwl_mvm_find_free_sta_id(mvm,
-						  ieee80211_vif_type_p2p(vif));
-	else
-		sta_id = mvm_sta->sta_id;
-
-	if (sta_id == IWL_MVM_INVALID_STA)
-		return -ENOSPC;
-
-	spin_lock_init(&mvm_sta->lock);
-
-	/* if this is a HW restart re-alloc existing queues */
-	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
-		struct iwl_mvm_int_sta tmp_sta = {
-			.sta_id = sta_id,
-			.type = mvm_sta->sta_type,
-		};
-
-		/*
-		 * First add an empty station since allocating
-		 * a queue requires a valid station
-		 */
-		ret = iwl_mvm_add_int_sta_common(mvm, &tmp_sta, sta->addr,
-						 mvmvif->id, mvmvif->color);
-		if (ret)
-			goto err;
-
-		iwl_mvm_realloc_queues_after_restart(mvm, sta);
-		sta_update = true;
-		sta_flags = iwl_mvm_has_new_tx_api(mvm) ? 0 : STA_MODIFY_QUEUES;
-		goto update_fw;
-	}
-
 	mvm_sta->sta_id = sta_id;
 	mvm_sta->mac_id_n_color = FW_CMD_ID_AND_COLOR(mvmvif->id,
 						      mvmvif->color);
@@ -1703,7 +1668,7 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 	else
 		mvm_sta->max_agg_bufsize = LINK_QUAL_AGG_FRAME_LIMIT_GEN2_DEF;
 	mvm_sta->tt_tx_protection = false;
-	mvm_sta->sta_type = sta->tdls ? IWL_STA_TDLS_LINK : IWL_STA_LINK;
+	mvm_sta->sta_type = sta_type;
 
 	mvm_sta->tid_disable_agg = 0xffff; /* No aggs at first */
 
@@ -1724,8 +1689,7 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 		atomic_set(&mvmtxq->tx_request, 0);
 	}
 
-	if (iwl_mvm_has_new_rx_api(mvm) &&
-	    !test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
+	if (iwl_mvm_has_new_rx_api(mvm)) {
 		int q;
 
 		dup_data = kcalloc(mvm->trans->num_rx_queues,
@@ -1751,7 +1715,7 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 		ret = iwl_mvm_reserve_sta_stream(mvm, sta,
 						 ieee80211_vif_type_p2p(vif));
 		if (ret)
-			goto err;
+			return ret;
 	}
 
 	/*
@@ -1765,6 +1729,56 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 
 	iwl_mvm_toggle_tx_ant(mvm, &mvm_sta->tx_ant);
 
+	return 0;
+}
+
+int iwl_mvm_add_sta(struct iwl_mvm *mvm,
+		    struct ieee80211_vif *vif,
+		    struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	int ret, sta_id;
+	bool sta_update = false;
+	unsigned int sta_flags = 0;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+		sta_id = iwl_mvm_find_free_sta_id(mvm,
+						  ieee80211_vif_type_p2p(vif));
+	else
+		sta_id = mvm_sta->sta_id;
+
+	if (sta_id == IWL_MVM_INVALID_STA)
+		return -ENOSPC;
+
+	spin_lock_init(&mvm_sta->lock);
+
+	/* if this is a HW restart re-alloc existing queues */
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
+		struct iwl_mvm_int_sta tmp_sta = {
+			.sta_id = sta_id,
+			.type = mvm_sta->sta_type,
+		};
+
+		/* First add an empty station since allocating
+		 * a queue requires a valid station
+		 */
+		ret = iwl_mvm_add_int_sta_common(mvm, &tmp_sta, sta->addr,
+						 mvmvif->id, mvmvif->color);
+		if (ret)
+			goto err;
+
+		iwl_mvm_realloc_queues_after_restart(mvm, sta);
+		sta_update = true;
+		sta_flags = iwl_mvm_has_new_tx_api(mvm) ? 0 : STA_MODIFY_QUEUES;
+		goto update_fw;
+	}
+
+	ret = iwl_mvm_sta_init(mvm, vif, sta, sta_id,
+			       sta->tdls ? IWL_STA_TDLS_LINK : IWL_STA_LINK);
+
 update_fw:
 	ret = iwl_mvm_sta_send_to_fw(mvm, sta, sta_update, sta_flags);
 	if (ret)
@@ -1912,43 +1926,16 @@ int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
 	return 0;
 }
 
-int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
-		   struct ieee80211_vif *vif,
-		   struct ieee80211_sta *sta)
+/* Execute the common part for both MLD and non-MLD modes */
+static bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, int *ret)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	u8 sta_id = mvm_sta->sta_id;
-	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	if (iwl_mvm_has_new_rx_api(mvm))
-		kfree(mvm_sta->dup_data);
-
-	ret = iwl_mvm_drain_sta(mvm, mvm_sta, true);
-	if (ret)
-		return ret;
-
-	/* flush its queues here since we are freeing mvm_sta */
-	ret = iwl_mvm_flush_sta(mvm, mvm_sta, false);
-	if (ret)
-		return ret;
-	if (iwl_mvm_has_new_tx_api(mvm)) {
-		ret = iwl_mvm_wait_sta_queues_empty(mvm, mvm_sta);
-	} else {
-		u32 q_mask = mvm_sta->tfd_queue_msk;
-
-		ret = iwl_trans_wait_tx_queues_empty(mvm->trans,
-						     q_mask);
-	}
-	if (ret)
-		return ret;
-
-	ret = iwl_mvm_drain_sta(mvm, mvm_sta, false);
-
-	iwl_mvm_disable_sta_queues(mvm, vif, sta);
-
 	/* If there is a TXQ still marked as reserved - free it */
 	if (mvm_sta->reserved_queue != IEEE80211_INVAL_HW_QUEUE) {
 		u8 reserved_txq = mvm_sta->reserved_queue;
@@ -1963,8 +1950,10 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 		if (WARN((*status != IWL_MVM_QUEUE_RESERVED) &&
 			 (*status != IWL_MVM_QUEUE_FREE),
 			 "sta_id %d reserved txq %d status %d",
-			 sta_id, reserved_txq, *status))
-			return -EINVAL;
+			 sta_id, reserved_txq, *status)) {
+			*ret = -EINVAL;
+			return true;
+		}
 
 		*status = IWL_MVM_QUEUE_FREE;
 	}
@@ -1973,7 +1962,7 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 	    mvmvif->ap_sta_id == sta_id) {
 		/* if associated - we can't remove the AP STA now */
 		if (vif->cfg.assoc)
-			return ret;
+			return true;
 
 		/* first remove remaining keys */
 		iwl_mvm_sec_key_remove_ap(mvm, vif);
@@ -2000,6 +1989,47 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 
 	iwl_mvm_time_sync_sta_rm(mvm, sta);
 
+	return false;
+}
+
+int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
+		   struct ieee80211_vif *vif,
+		   struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (iwl_mvm_has_new_rx_api(mvm))
+		kfree(mvm_sta->dup_data);
+
+	ret = iwl_mvm_drain_sta(mvm, mvm_sta, true);
+	if (ret)
+		return ret;
+
+	/* flush its queues here since we are freeing mvm_sta */
+	ret = iwl_mvm_flush_sta(mvm, mvm_sta, false);
+	if (ret)
+		return ret;
+	if (iwl_mvm_has_new_tx_api(mvm)) {
+		ret = iwl_mvm_wait_sta_queues_empty(mvm, mvm_sta);
+	} else {
+		u32 q_mask = mvm_sta->tfd_queue_msk;
+
+		ret = iwl_trans_wait_tx_queues_empty(mvm->trans,
+						     q_mask);
+	}
+	if (ret)
+		return ret;
+
+	ret = iwl_mvm_drain_sta(mvm, mvm_sta, false);
+
+	iwl_mvm_disable_sta_queues(mvm, vif, sta);
+
+	if (iwl_mvm_sta_del(mvm, vif, sta, &ret))
+		return ret;
+
 	ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->sta_id);
 	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->sta_id], NULL);
 
-- 
cgit v1.2.3-70-g09d2


From 3068248b860c3c5163b0c6881cbaee5ebc020184 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Mon, 20 Mar 2023 12:33:17 +0200
Subject: wifi: iwlwifi: Update configurations for Bnj device

Add missing rf support for spacific rf for BnJ device

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.eeb7ebb9ed87.I30c3938221aeb6f6ba2c84b1a77eea54c9dd3ba1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  | 24 ++++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  2 ++
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   | 14 ++++++++++++--
 3 files changed, 38 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index bc33d19ca08d..e649daaf842c 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -68,7 +68,9 @@
 #define IWL_BNJ_A_GF_A_FW_PRE		"iwlwifi-BzBnj-a0-gf-a0-"
 #define IWL_BNJ_B_GF_A_FW_PRE		"iwlwifi-BzBnj-b0-gf-a0-"
 #define IWL_BNJ_A_GF4_A_FW_PRE		"iwlwifi-BzBnj-a0-gf4-a0-"
+#define IWL_BNJ_B_GF4_A_FW_PRE		"iwlwifi-BzBnj-b0-gf4-a0-"
 #define IWL_BNJ_A_HR_B_FW_PRE		"iwlwifi-BzBnj-a0-hr-b0-"
+#define IWL_BNJ_B_HR_B_FW_PRE		"iwlwifi-BzBnj-b0-hr-b0-"
 #define IWL_BNJ_B_FM_B_FW_PRE		"iwlwifi-BzBnj-b0-fm-b0-"
 
 
@@ -148,8 +150,12 @@
 	IWL_BNJ_B_GF_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_GF4_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_BNJ_B_GF4_A_MODULE_FIRMWARE(api) \
+	IWL_BNJ_B_GF4_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_HR_B_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_HR_B_FW_PRE __stringify(api) ".ucode"
+#define IWL_BNJ_B_HR_B_MODULE_FIRMWARE(api) \
+	IWL_BNJ_B_HR_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_B_FM_B_MODULE_FIRMWARE(api) \
 	IWL_BNJ_B_FM_B_FW_PRE __stringify(api) ".ucode"
 
@@ -1058,6 +1064,14 @@ const struct iwl_cfg iwl_cfg_bnj_a0_gf4_a0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bnj_b0_gf4_a0 = {
+	.fw_name_pre = IWL_BNJ_B_GF4_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0 = {
 	.fw_name_pre = IWL_BNJ_A_HR_B_FW_PRE,
 	.uhb_supported = true,
@@ -1066,6 +1080,14 @@ const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bnj_b0_hr_b0 = {
+	.fw_name_pre = IWL_BNJ_B_HR_B_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bnj_b0_fm_b0 = {
 	.fw_name_pre = IWL_BNJ_B_FM_B_FW_PRE,
 	.uhb_supported = true,
@@ -1108,7 +1130,9 @@ MODULE_FIRMWARE(IWL_BNJ_B_FM4_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_B_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BNJ_B_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BNJ_B_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_FM4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_FM4_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_GL_B_FM_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index a4c7fd929cc2..9b7b6fca2b1b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -660,7 +660,9 @@ extern const struct iwl_cfg iwl_cfg_bnj_a0_fm4_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_gf4_a0;
+extern const struct iwl_cfg iwl_cfg_bnj_b0_gf4_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0;
+extern const struct iwl_cfg iwl_cfg_bnj_b0_hr_b0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_fm_b0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_fm4_b0;
 #endif /* CONFIG_IWLMVM */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 077983913acd..6c935d73943a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1219,15 +1219,25 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_b0_gf_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_gf4_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
+		      iwl_cfg_bnj_b0_gf4_a0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_hr_b0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_NO_JACKET,
+		      iwl_cfg_bnj_b0_hr_b0, iwl_bz_name),
 
 /* SoF with JF2 */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-- 
cgit v1.2.3-70-g09d2


From cf85123a210fe99cdd972a5cc84857fec925f794 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 20 Mar 2023 12:33:18 +0200
Subject: wifi: iwlwifi: mvm: support enabling and disabling HW timestamping

Instead of enabling HW timestamping by default every time a station
is connected, disable it by default and enable it only upon request
for a specific station. HW timestamping can be enabled for only one
peer at a time.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.62b98fbf545b.I450c1017ada7900a71a63d879bb542a08e3166c8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c        |  7 +++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 29 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       |  8 ------
 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c | 13 +++-------
 drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h |  4 +--
 .../chromeOS/adjustments/024-time-sync.spatch      |  6 +++++
 6 files changed, 47 insertions(+), 20 deletions(-)
 create mode 100644 intc-scripts/chromeOS/adjustments/024-time-sync.spatch

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 4079c0d939c1..05236b09bcdd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -21,6 +21,7 @@
 #include "iwl-phy-db.h"
 #include "iwl-modparams.h"
 #include "iwl-nvm-parse.h"
+#include "time-sync.h"
 
 #define MVM_UCODE_ALIVE_TIMEOUT	(HZ)
 #define MVM_UCODE_CALIB_TIMEOUT	(2 * HZ)
@@ -1666,8 +1667,12 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 			goto error;
 	}
 
-	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
 		iwl_mvm_send_recovery_cmd(mvm, ERROR_RECOVERY_UPDATE_DB);
+		iwl_mvm_time_sync_config(mvm, mvm->time_sync.peer_addr,
+					 IWL_TIME_SYNC_PROTOCOL_TM |
+					 IWL_TIME_SYNC_PROTOCOL_FTM);
+	}
 
 	if (!mvm->ptp_data.ptp_clock)
 		iwl_mvm_ptp_init(mvm);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 6e059027517f..33582f4eea91 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -28,6 +28,7 @@
 #include "fw/error-dump.h"
 #include "iwl-prph.h"
 #include "iwl-nvm-parse.h"
+#include "time-sync.h"
 
 static const struct ieee80211_iface_limit iwl_mvm_limits[] = {
 	{
@@ -416,6 +417,10 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 		wiphy_ext_feature_set(hw->wiphy,
 				      NL80211_EXT_FEATURE_BEACON_PROTECTION_CLIENT);
 
+	if (fw_has_capa(&mvm->fw->ucode_capa,
+			IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM))
+		hw->wiphy->hw_timestamp_max_peers = 1;
+
 	ieee80211_hw_set(hw, SINGLE_SCAN_ON_ALL_BANDS);
 	hw->wiphy->features |=
 		NL80211_FEATURE_SCHED_SCAN_RANDOM_MAC_ADDR |
@@ -5765,6 +5770,29 @@ static bool iwl_mvm_mac_can_aggregate(struct ieee80211_hw *hw,
 	return iwl_mvm_can_hw_csum(skb) == iwl_mvm_can_hw_csum(head);
 }
 
+static int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
+				    struct ieee80211_vif *vif,
+				    struct cfg80211_set_hw_timestamp *hwts)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	u32 protocols = 0;
+	int ret;
+
+	/* HW timestamping is only supported for a specific station */
+	if (!hwts->macaddr)
+		return -EOPNOTSUPP;
+
+	if (hwts->enable)
+		protocols =
+			IWL_TIME_SYNC_PROTOCOL_TM | IWL_TIME_SYNC_PROTOCOL_FTM;
+
+	mutex_lock(&mvm->mutex);
+	ret = iwl_mvm_time_sync_config(mvm, hwts->macaddr, protocols);
+	mutex_unlock(&mvm->mutex);
+
+	return ret;
+}
+
 const struct ieee80211_ops iwl_mvm_hw_ops = {
 	.tx = iwl_mvm_mac_tx,
 	.wake_tx_queue = iwl_mvm_mac_wake_tx_queue,
@@ -5853,4 +5881,5 @@ const struct ieee80211_ops iwl_mvm_hw_ops = {
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	.sta_add_debugfs = iwl_mvm_sta_add_debugfs,
 #endif
+	.set_hw_timestamp = iwl_mvm_set_hw_timestamp,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 7d9d4e85b548..ec843f1b0165 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -9,7 +9,6 @@
 #include "mvm.h"
 #include "sta.h"
 #include "rs.h"
-#include "time-sync.h"
 
 /*
  * New version of ADD_STA_sta command added new fields at the end of the
@@ -1793,11 +1792,6 @@ update_fw:
 		}
 	}
 
-	if (!sta->tdls)
-		iwl_mvm_time_sync_config(mvm, sta->addr,
-					 IWL_TIME_SYNC_PROTOCOL_TM |
-					 IWL_TIME_SYNC_PROTOCOL_FTM);
-
 	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
 
 	return 0;
@@ -1987,8 +1981,6 @@ static bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	spin_lock_bh(&mvm_sta->lock);
 	spin_unlock_bh(&mvm_sta->lock);
 
-	iwl_mvm_time_sync_sta_rm(mvm, sta);
-
 	return false;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
index 434d27a44dea..edae3e24192b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.c
@@ -125,7 +125,7 @@ void iwl_mvm_time_sync_msmt_confirm_event(struct iwl_mvm *mvm,
 	ieee80211_tx_status_ext(mvm->hw, &status);
 }
 
-int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, u8 *addr, u32 protocols)
+int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, const u8 *addr, u32 protocols)
 {
 	struct iwl_time_sync_cfg_cmd cmd = {};
 	int err;
@@ -166,13 +166,8 @@ int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, u8 *addr, u32 protocols)
 		IWL_DEBUG_INFO(mvm, "Time sync: set peer addr=%pM\n", addr);
 	}
 
-	return err;
-}
+	if (!mvm->time_sync.active)
+		skb_queue_purge(&mvm->time_sync.frame_list);
 
-void iwl_mvm_time_sync_sta_rm(struct iwl_mvm *mvm, struct ieee80211_sta *sta)
-{
-	/* Disable time sync with this station */
-	iwl_mvm_time_sync_config(mvm, sta->addr, 0);
-
-	skb_queue_purge(&mvm->time_sync.frame_list);
+	return err;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h
index c43d1d0b25cb..2cfd0fb5e781 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-sync.h
@@ -13,8 +13,8 @@ void iwl_mvm_time_sync_msmt_event(struct iwl_mvm *mvm,
 				  struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_time_sync_msmt_confirm_event(struct iwl_mvm *mvm,
 					  struct iwl_rx_cmd_buffer *rxb);
-int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, u8 *addr, u32 protocols);
-void iwl_mvm_time_sync_sta_rm(struct iwl_mvm *mvm, struct ieee80211_sta *sta);
+int iwl_mvm_time_sync_config(struct iwl_mvm *mvm, const u8 *addr,
+			     u32 protocols);
 
 static inline
 bool iwl_mvm_time_sync_frame(struct iwl_mvm *mvm, struct sk_buff *skb, u8 *addr)
diff --git a/intc-scripts/chromeOS/adjustments/024-time-sync.spatch b/intc-scripts/chromeOS/adjustments/024-time-sync.spatch
new file mode 100644
index 000000000000..425e4700b343
--- /dev/null
+++ b/intc-scripts/chromeOS/adjustments/024-time-sync.spatch
@@ -0,0 +1,6 @@
+@@
+struct ieee80211_hw *hw;
+expression val;
+@@
+-hw->wiphy->hw_timestamp_max_peers = val
++hw_timestamp_max_peers(hw)
-- 
cgit v1.2.3-70-g09d2


From 9457077df49e2a637a1f1c3be42c14af604fa920 Mon Sep 17 00:00:00 2001
From: Abhishek Naik <abhishek.naik@intel.com>
Date: Mon, 20 Mar 2023 12:33:19 +0200
Subject: wifi: iwlwifi: mvm: Add debugfs to get TAS status

Add debugfs file in mvm to retrieve TAS status per LMAC,
TAS block list, current mcc, OEM name and OEM allowed list.

Signed-off-by: Abhishek Naik <abhishek.naik@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230320122330.8efc8c41efae.I94e1a6efb9c33e2cdbcf4bf3ed2384005397dee9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/debug.h |  96 +++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c  | 190 +++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c       |  10 ++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |   2 +-
 4 files changed, 296 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h b/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h
index 0c555089e05f..8fef38139bf6 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h
@@ -42,6 +42,12 @@ enum iwl_debug_cmds {
 	 * &struct iwl_buf_alloc_cmd
 	 */
 	BUFFER_ALLOCATION = 0x8,
+	/**
+	 * @GET_TAS_STATUS:
+	 * sends command to fw to get TAS status
+	 * the response is &struct iwl_mvm_tas_status_resp
+	 */
+	GET_TAS_STATUS = 0xA,
 	/**
 	 * @FW_DUMP_COMPLETE_CMD:
 	 * sends command to fw once dump collection completed
@@ -421,4 +427,94 @@ struct iwl_dbg_dump_complete_cmd {
 	__le32 tp_data;
 } __packed; /* FW_DUMP_COMPLETE_CMD_API_S_VER_1 */
 
+#define TAS_LMAC_BAND_HB       0
+#define TAS_LMAC_BAND_LB       1
+#define TAS_LMAC_BAND_UHB      2
+#define TAS_LMAC_BAND_INVALID  3
+
+/**
+ * struct iwl_mvm_tas_status_per_mac - tas status per lmac
+ * @static_status: tas statically enabled or disabled per lmac - TRUE/FALSE
+ * @static_dis_reason: TAS static disable reason, uses
+ *	&enum iwl_mvm_tas_statically_disabled_reason
+ * @dynamic_status: Current TAS  status. uses
+ *	&enum iwl_mvm_tas_dyna_status
+ * @near_disconnection: is TAS currently near disconnection per lmac? - TRUE/FALSE
+ * @max_reg_pwr_limit: Regulatory power limits in dBm
+ * @sar_limit: SAR limits per lmac in dBm
+ * @band: Band per lmac
+ * @reserved: reserved
+ */
+struct iwl_mvm_tas_status_per_mac {
+	u8 static_status;
+	u8 static_dis_reason;
+	u8 dynamic_status;
+	u8 near_disconnection;
+	__le16 max_reg_pwr_limit;
+	__le16 sar_limit;
+	u8 band;
+	u8 reserved[3];
+} __packed; /*DEBUG_GET_TAS_STATUS_PER_MAC_S_VER_1*/
+
+/**
+ * struct iwl_mvm_tas_status_resp - Response to GET_TAS_STATUS
+ * @tas_fw_version: TAS FW version
+ * @is_uhb_for_usa_enable: is UHB enabled in USA? - TRUE/FALSE
+ * @curr_mcc: current mcc
+ * @block_list: country block list
+ * @tas_status_mac: TAS status per lmac, uses
+ *	&struct iwl_mvm_tas_status_per_mac
+ * @in_dual_radio: is TAS in dual radio? - TRUE/FALSE
+ * @reserved: reserved
+ */
+struct iwl_mvm_tas_status_resp {
+	u8 tas_fw_version;
+	u8 is_uhb_for_usa_enable;
+	__le16 curr_mcc;
+	__le16 block_list[16];
+	struct iwl_mvm_tas_status_per_mac tas_status_mac[2];
+	u8 in_dual_radio;
+	u8 reserved[3];
+} __packed; /*DEBUG_GET_TAS_STATUS_RSP_API_S_VER_3*/
+
+/**
+ * enum iwl_mvm_tas_dyna_status - TAS current running status
+ * @TAS_DYNA_INACTIVE: TAS status is inactive
+ * @TAS_DYNA_INACTIVE_MVM_MODE: TAS is disabled due because FW is in MVM mode
+ *	or is in softap mode.
+ * @TAS_DYNA_INACTIVE_TRIGGER_MODE: TAS is disabled because FW is in
+ *	multi user trigger mode
+ * @TAS_DYNA_INACTIVE_BLOCK_LISTED: TAS is disabled because  current mcc
+ *	is blocklisted mcc
+ * @TAS_DYNA_INACTIVE_UHB_NON_US: TAS is disabled because current band is UHB
+ *	and current mcc is USA
+ * @TAS_DYNA_ACTIVE: TAS is currently active
+ * @TAS_DYNA_STATUS_MAX: TAS status max value
+ */
+enum iwl_mvm_tas_dyna_status {
+	TAS_DYNA_INACTIVE,
+	TAS_DYNA_INACTIVE_MVM_MODE,
+	TAS_DYNA_INACTIVE_TRIGGER_MODE,
+	TAS_DYNA_INACTIVE_BLOCK_LISTED,
+	TAS_DYNA_INACTIVE_UHB_NON_US,
+	TAS_DYNA_ACTIVE,
+
+	TAS_DYNA_STATUS_MAX,
+}; /*_TAS_DYNA_STATUS_E*/
+
+/**
+ * enum iwl_mvm_tas_statically_disabled_reason - TAS statically disabled reason
+ * @TAS_DISABLED_DUE_TO_BIOS: TAS is disabled because TAS is disabled in BIOS
+ * @TAS_DISABLED_DUE_TO_SAR_6DBM: TAS is disabled because SAR limit is less than 6 Dbm
+ * @TAS_DISABLED_REASON_INVALID: TAS disable reason is invalid
+ * @TAS_DISABLED_REASON_MAX: TAS disable reason max value
+ */
+enum iwl_mvm_tas_statically_disabled_reason {
+	TAS_DISABLED_DUE_TO_BIOS,
+	TAS_DISABLED_DUE_TO_SAR_6DBM,
+	TAS_DISABLED_REASON_INVALID,
+
+	TAS_DISABLED_REASON_MAX,
+}; /*_TAS_STATICALLY_DISABLED_REASON_E*/
+
 #endif /* __iwl_fw_api_debug_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 5790435dc69e..2a6f42f07c9e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2012-2014, 2018-2022 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2023 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -8,6 +8,7 @@
 #include <linux/err.h>
 #include <linux/ieee80211.h>
 #include <linux/netdevice.h>
+#include <linux/dmi.h>
 
 #include "mvm.h"
 #include "sta.h"
@@ -15,6 +16,7 @@
 #include "debugfs.h"
 #include "iwl-modparams.h"
 #include "fw/error-dump.h"
+#include "fw/api/phy-ctxt.h"
 
 static ssize_t iwl_dbgfs_ctdp_budget_read(struct file *file,
 					  char __user *user_buf,
@@ -714,6 +716,190 @@ static ssize_t iwl_dbgfs_fw_ver_read(struct file *file, char __user *user_buf,
 	return ret;
 }
 
+static ssize_t iwl_dbgfs_tas_get_status_read(struct file *file,
+					     char __user *user_buf,
+					     size_t count, loff_t *ppos)
+{
+	struct iwl_mvm *mvm = file->private_data;
+	struct iwl_mvm_tas_status_resp tas_rsp;
+	struct iwl_mvm_tas_status_resp *rsp = &tas_rsp;
+	static const size_t bufsz = 1024;
+	char *buff, *pos, *endpos;
+	const char * const tas_dis_reason[TAS_DISABLED_REASON_MAX] = {
+		[TAS_DISABLED_DUE_TO_BIOS] =
+			"Due To BIOS",
+		[TAS_DISABLED_DUE_TO_SAR_6DBM] =
+			"Due To SAR Limit Less Than 6 dBm",
+		[TAS_DISABLED_REASON_INVALID] =
+			"N/A",
+	};
+	const char * const tas_current_status[TAS_DYNA_STATUS_MAX] = {
+		[TAS_DYNA_INACTIVE] = "INACTIVE",
+		[TAS_DYNA_INACTIVE_MVM_MODE] =
+			"inactive due to mvm mode",
+		[TAS_DYNA_INACTIVE_TRIGGER_MODE] =
+			"inactive due to trigger mode",
+		[TAS_DYNA_INACTIVE_BLOCK_LISTED] =
+			"inactive due to block listed",
+		[TAS_DYNA_INACTIVE_UHB_NON_US] =
+			"inactive due to uhb non US",
+		[TAS_DYNA_ACTIVE] = "ACTIVE",
+	};
+	struct iwl_host_cmd hcmd = {
+		.id = WIDE_ID(DEBUG_GROUP, GET_TAS_STATUS),
+		.flags = CMD_WANT_SKB,
+		.len = { 0, },
+		.data = { NULL, },
+	};
+	int ret, i, tmp;
+	bool tas_enabled = false;
+	unsigned long dyn_status;
+
+	if (!iwl_mvm_firmware_running(mvm))
+		return -ENODEV;
+
+	mutex_lock(&mvm->mutex);
+	ret = iwl_mvm_send_cmd(mvm, &hcmd);
+	mutex_unlock(&mvm->mutex);
+	if (ret < 0)
+		return ret;
+
+	buff = kzalloc(bufsz, GFP_KERNEL);
+	if (!buff)
+		return -ENOMEM;
+	pos = buff;
+	endpos = pos + bufsz;
+
+	rsp = (void *)hcmd.resp_pkt->data;
+
+	pos += scnprintf(pos, endpos - pos, "TAS Conclusion:\n");
+	for (i = 0; i < rsp->in_dual_radio + 1; i++) {
+		if (rsp->tas_status_mac[i].band != TAS_LMAC_BAND_INVALID &&
+		    rsp->tas_status_mac[i].dynamic_status & BIT(TAS_DYNA_ACTIVE)) {
+			pos += scnprintf(pos, endpos - pos, "\tON for ");
+			switch (rsp->tas_status_mac[i].band) {
+			case TAS_LMAC_BAND_HB:
+				pos += scnprintf(pos, endpos - pos, "HB\n");
+				break;
+			case TAS_LMAC_BAND_LB:
+				pos += scnprintf(pos, endpos - pos, "LB\n");
+				break;
+			case TAS_LMAC_BAND_UHB:
+				pos += scnprintf(pos, endpos - pos, "UHB\n");
+				break;
+			case TAS_LMAC_BAND_INVALID:
+				pos += scnprintf(pos, endpos - pos,
+						 "INVALID BAND\n");
+				break;
+			default:
+				pos += scnprintf(pos, endpos - pos,
+						 "Unsupported band (%d)\n",
+						 rsp->tas_status_mac[i].band);
+				goto out;
+			}
+			tas_enabled = true;
+		}
+	}
+	if (!tas_enabled)
+		pos += scnprintf(pos, endpos - pos, "\tOFF\n");
+
+	pos += scnprintf(pos, endpos - pos, "TAS Report\n");
+	pos += scnprintf(pos, endpos - pos, "TAS FW version: %d\n",
+			 rsp->tas_fw_version);
+	pos += scnprintf(pos, endpos - pos, "Is UHB enabled for USA?: %s\n",
+			 rsp->is_uhb_for_usa_enable ? "True" : "False");
+	pos += scnprintf(pos, endpos - pos, "Current MCC: 0x%x\n",
+			 le16_to_cpu(rsp->curr_mcc));
+
+	pos += scnprintf(pos, endpos - pos, "Block list entries:");
+	for (i = 0; i < APCI_WTAS_BLACK_LIST_MAX; i++)
+		pos += scnprintf(pos, endpos - pos, " 0x%x",
+				 le16_to_cpu(rsp->block_list[i]));
+
+	pos += scnprintf(pos, endpos - pos, "\nOEM name: %s\n",
+			 dmi_get_system_info(DMI_SYS_VENDOR));
+	pos += scnprintf(pos, endpos - pos, "\tVendor In Approved List: %s\n",
+			 iwl_mvm_is_vendor_in_approved_list() ? "YES" : "NO");
+	pos += scnprintf(pos, endpos - pos,
+			 "\tDo TAS Support Dual Radio?: %s\n",
+			 rsp->in_dual_radio ? "TRUE" : "FALSE");
+
+	for (i = 0; i < rsp->in_dual_radio + 1; i++) {
+		if (rsp->tas_status_mac[i].static_status == 0) {
+			pos += scnprintf(pos, endpos - pos,
+					 "Static status: disabled\n");
+			pos += scnprintf(pos, endpos - pos,
+					 "Static disabled reason: %s (0)\n",
+					 tas_dis_reason[0]);
+			goto out;
+		}
+
+		pos += scnprintf(pos, endpos - pos, "TAS status for ");
+		switch (rsp->tas_status_mac[i].band) {
+		case TAS_LMAC_BAND_HB:
+			pos += scnprintf(pos, endpos - pos, "High band\n");
+			break;
+		case TAS_LMAC_BAND_LB:
+			pos += scnprintf(pos, endpos - pos, "Low band\n");
+			break;
+		case TAS_LMAC_BAND_UHB:
+			pos += scnprintf(pos, endpos - pos,
+					 "Ultra high band\n");
+			break;
+		case TAS_LMAC_BAND_INVALID:
+			pos += scnprintf(pos, endpos - pos,
+					 "INVALID band\n");
+			break;
+		default:
+			pos += scnprintf(pos, endpos - pos,
+					 "Unsupported band (%d)\n",
+					 rsp->tas_status_mac[i].band);
+			goto out;
+		}
+		pos += scnprintf(pos, endpos - pos, "Static status: %sabled\n",
+				 rsp->tas_status_mac[i].static_status ?
+				 "En" : "Dis");
+		pos += scnprintf(pos, endpos - pos,
+				 "\tStatic Disabled Reason: ");
+		if (rsp->tas_status_mac[i].static_dis_reason < TAS_DISABLED_REASON_MAX)
+			pos += scnprintf(pos, endpos - pos, "%s (%d)\n",
+					 tas_dis_reason[rsp->tas_status_mac[i].static_dis_reason],
+					 rsp->tas_status_mac[i].static_dis_reason);
+		else
+			pos += scnprintf(pos, endpos - pos,
+					 "unsupported value (%d)\n",
+					 rsp->tas_status_mac[i].static_dis_reason);
+
+		pos += scnprintf(pos, endpos - pos, "Dynamic status:\n");
+		dyn_status = (rsp->tas_status_mac[i].dynamic_status);
+		for_each_set_bit(tmp, &dyn_status, sizeof(dyn_status)) {
+			if (tmp >= 0 && tmp < TAS_DYNA_STATUS_MAX)
+				pos += scnprintf(pos, endpos - pos,
+						 "\t%s (%d)\n",
+						 tas_current_status[tmp], tmp);
+		}
+
+		pos += scnprintf(pos, endpos - pos,
+				 "Is near disconnection?: %s\n",
+				 rsp->tas_status_mac[i].near_disconnection ?
+				 "True" : "False");
+		tmp = le16_to_cpu(rsp->tas_status_mac[i].max_reg_pwr_limit);
+		pos += scnprintf(pos, endpos - pos,
+				 "Max. regulatory pwr limit (dBm): %d.%03d\n",
+				 tmp / 8, 125 * (tmp % 8));
+		tmp = le16_to_cpu(rsp->tas_status_mac[i].sar_limit);
+		pos += scnprintf(pos, endpos - pos,
+				 "SAR limit (dBm): %d.%03d\n",
+				 tmp / 8, 125 * (tmp % 8));
+	}
+
+out:
+	ret = simple_read_from_buffer(user_buf, count, ppos, buff, pos - buff);
+	kfree(buff);
+	iwl_free_resp(&hcmd);
+	return ret;
+}
+
 static ssize_t iwl_dbgfs_phy_integration_ver_read(struct file *file,
 						  char __user *user_buf,
 						  size_t count, loff_t *ppos)
@@ -1685,6 +1871,7 @@ MVM_DEBUGFS_READ_FILE_OPS(fw_rx_stats);
 MVM_DEBUGFS_READ_FILE_OPS(drv_rx_stats);
 MVM_DEBUGFS_READ_FILE_OPS(fw_ver);
 MVM_DEBUGFS_READ_FILE_OPS(phy_integration_ver);
+MVM_DEBUGFS_READ_FILE_OPS(tas_get_status);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_restart, 10);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_nmi, 10);
 MVM_DEBUGFS_WRITE_FILE_OPS(bt_tx_prio, 10);
@@ -1894,6 +2081,7 @@ void iwl_mvm_dbgfs_register(struct iwl_mvm *mvm)
 
 	if (mvm->fw->phy_integration_ver)
 		MVM_DEBUGFS_ADD_FILE(phy_integration_ver, mvm->debugfs_dir, 0400);
+	MVM_DEBUGFS_ADD_FILE(tas_get_status, mvm->debugfs_dir, 0400);
 #ifdef CONFIG_ACPI
 	MVM_DEBUGFS_ADD_FILE(sar_geo_profile, mvm->debugfs_dir, 0400);
 #endif
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 05236b09bcdd..6d6fa35c2b21 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1092,6 +1092,11 @@ static const struct dmi_system_id dmi_tas_approved_list[] = {
 	{}
 };
 
+bool iwl_mvm_is_vendor_in_approved_list(void)
+{
+	return dmi_check_system(dmi_tas_approved_list);
+}
+
 static bool iwl_mvm_add_to_tas_block_list(__le32 *list, __le32 *le_size, unsigned int mcc)
 {
 	int i;
@@ -1371,6 +1376,11 @@ static void iwl_mvm_lari_cfg(struct iwl_mvm *mvm)
 {
 }
 
+bool iwl_mvm_is_vendor_in_approved_list(void)
+{
+	return false;
+}
+
 static u8 iwl_mvm_eval_dsm_rfi(struct iwl_mvm *mvm)
 {
 	return DSM_VALUE_RFI_DISABLE;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index a03c2e5bbc25..cea94250d197 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2343,5 +2343,5 @@ static inline void iwl_mvm_mei_set_sw_rfkill_state(struct iwl_mvm *mvm)
 void iwl_mvm_send_roaming_forbidden_event(struct iwl_mvm *mvm,
 					  struct ieee80211_vif *vif,
 					  bool forbidden);
-
+bool iwl_mvm_is_vendor_in_approved_list(void);
 #endif /* __IWL_MVM_H__ */
-- 
cgit v1.2.3-70-g09d2


From 61587f1556fec39e8bafc40c8715f560639a4cf2 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 21 Mar 2023 10:12:48 +0100
Subject: wifi: mac80211: add support for letting drivers register tc offload
 support

On newer MediaTek SoCs (e.g. MT7986), WLAN->WLAN or WLAN->Ethernet flows can
be offloaded by the SoC. In order to support that, the .ndo_setup_tc op is
needed.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20230321091248.30947-1-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h     |  9 +++++++++
 net/mac80211/driver-ops.h  | 17 +++++++++++++++++
 net/mac80211/ieee80211_i.h |  3 ++-
 net/mac80211/iface.c       | 11 +++++++++++
 net/mac80211/trace.h       | 25 +++++++++++++++++++++++++
 5 files changed, 64 insertions(+), 1 deletion(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index f12edca660ba..fcfe3e9aff3d 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -4227,6 +4227,10 @@ struct ieee80211_prep_tx_info {
  * @set_hw_timestamp: Enable/disable HW timestamping of TM/FTM frames. This is
  *	not restored at HW reset by mac80211 so drivers need to take care of
  *	that.
+ * @net_setup_tc: Called from .ndo_setup_tc in order to prepare hardware
+ *	flow offloading for flows originating from the vif.
+ *	Note that the driver must not assume that the vif driver_data is valid
+ *	at this point, since the callback can be called during netdev teardown.
  */
 struct ieee80211_ops {
 	void (*tx)(struct ieee80211_hw *hw,
@@ -4593,6 +4597,11 @@ struct ieee80211_ops {
 	int (*set_hw_timestamp)(struct ieee80211_hw *hw,
 				struct ieee80211_vif *vif,
 				struct cfg80211_set_hw_timestamp *hwts);
+	int (*net_setup_tc)(struct ieee80211_hw *hw,
+			    struct ieee80211_vif *vif,
+			    struct net_device *dev,
+			    enum tc_setup_type type,
+			    void *type_data);
 };
 
 /**
diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h
index a68d606e6987..0bf208f5bbc5 100644
--- a/net/mac80211/driver-ops.h
+++ b/net/mac80211/driver-ops.h
@@ -1502,6 +1502,23 @@ static inline int drv_net_fill_forward_path(struct ieee80211_local *local,
 	return ret;
 }
 
+static inline int drv_net_setup_tc(struct ieee80211_local *local,
+				   struct ieee80211_sub_if_data *sdata,
+				   struct net_device *dev,
+				   enum tc_setup_type type, void *type_data)
+{
+	int ret = -EOPNOTSUPP;
+
+	sdata = get_bss_sdata(sdata);
+	trace_drv_net_setup_tc(local, sdata, type);
+	if (local->ops->net_setup_tc)
+		ret = local->ops->net_setup_tc(&local->hw, &sdata->vif, dev,
+					       type, type_data);
+	trace_drv_return_int(local, ret);
+
+	return ret;
+}
+
 int drv_change_vif_links(struct ieee80211_local *local,
 			 struct ieee80211_sub_if_data *sdata,
 			 u16 old_links, u16 new_links,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 3d4edc25a69e..b2535614483e 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1939,7 +1939,8 @@ void ieee80211_color_collision_detection_work(struct work_struct *work);
 /* interface handling */
 #define MAC80211_SUPPORTED_FEATURES_TX	(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | \
 					 NETIF_F_HW_CSUM | NETIF_F_SG | \
-					 NETIF_F_HIGHDMA | NETIF_F_GSO_SOFTWARE)
+					 NETIF_F_HIGHDMA | NETIF_F_GSO_SOFTWARE | \
+					 NETIF_F_HW_TC)
 #define MAC80211_SUPPORTED_FEATURES_RX	(NETIF_F_RXCSUM)
 #define MAC80211_SUPPORTED_FEATURES	(MAC80211_SUPPORTED_FEATURES_TX | \
 					 MAC80211_SUPPORTED_FEATURES_RX)
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 23ed13f15067..bd2c48870add 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -813,6 +813,15 @@ ieee80211_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	dev_fetch_sw_netstats(stats, dev->tstats);
 }
 
+static int ieee80211_netdev_setup_tc(struct net_device *dev,
+				     enum tc_setup_type type, void *type_data)
+{
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_local *local = sdata->local;
+
+	return drv_net_setup_tc(local, sdata, dev, type, type_data);
+}
+
 static const struct net_device_ops ieee80211_dataif_ops = {
 	.ndo_open		= ieee80211_open,
 	.ndo_stop		= ieee80211_stop,
@@ -821,6 +830,7 @@ static const struct net_device_ops ieee80211_dataif_ops = {
 	.ndo_set_rx_mode	= ieee80211_set_multicast_list,
 	.ndo_set_mac_address 	= ieee80211_change_mac,
 	.ndo_get_stats64	= ieee80211_get_stats64,
+	.ndo_setup_tc		= ieee80211_netdev_setup_tc,
 };
 
 static u16 ieee80211_monitor_select_queue(struct net_device *dev,
@@ -929,6 +939,7 @@ static const struct net_device_ops ieee80211_dataif_8023_ops = {
 	.ndo_set_mac_address	= ieee80211_change_mac,
 	.ndo_get_stats64	= ieee80211_get_stats64,
 	.ndo_fill_forward_path	= ieee80211_netdev_fill_forward_path,
+	.ndo_setup_tc		= ieee80211_netdev_setup_tc,
 };
 
 static bool ieee80211_iftype_supports_hdr_offload(enum nl80211_iftype iftype)
diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h
index 9f4377566c42..e0ccf5fe708a 100644
--- a/net/mac80211/trace.h
+++ b/net/mac80211/trace.h
@@ -2478,6 +2478,31 @@ DEFINE_EVENT(sta_event, drv_net_fill_forward_path,
 	TP_ARGS(local, sdata, sta)
 );
 
+TRACE_EVENT(drv_net_setup_tc,
+	TP_PROTO(struct ieee80211_local *local,
+		 struct ieee80211_sub_if_data *sdata,
+		 u8 type),
+
+	TP_ARGS(local, sdata, type),
+
+	TP_STRUCT__entry(
+		LOCAL_ENTRY
+		VIF_ENTRY
+		__field(u8, type)
+	),
+
+	TP_fast_assign(
+		LOCAL_ASSIGN;
+		VIF_ASSIGN;
+		__entry->type = type;
+	),
+
+	TP_printk(
+		LOCAL_PR_FMT VIF_PR_FMT " type:%d\n",
+		LOCAL_PR_ARG, VIF_PR_ARG, __entry->type
+	)
+);
+
 TRACE_EVENT(drv_change_vif_links,
 	TP_PROTO(struct ieee80211_local *local,
 		 struct ieee80211_sub_if_data *sdata,
-- 
cgit v1.2.3-70-g09d2


From ce1fdb065695f49ef6f126d35c1abbfe645d62d5 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 21 Mar 2023 07:58:26 +0100
Subject: sh_eth: remove open coded netif_running()

It had a purpose back in the days, but today we have a handy helper.

Reported-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230321065826.2044-1-wsa+renesas@sang-engineering.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/sh_eth.c | 6 +-----
 drivers/net/ethernet/renesas/sh_eth.h | 1 -
 2 files changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index d8ec729825be..2d9787231099 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -2441,8 +2441,6 @@ static int sh_eth_open(struct net_device *ndev)
 
 	netif_start_queue(ndev);
 
-	mdp->is_opened = 1;
-
 	return ret;
 
 out_free_irq:
@@ -2565,7 +2563,7 @@ static struct net_device_stats *sh_eth_get_stats(struct net_device *ndev)
 	if (mdp->cd->no_tx_cntrs)
 		return &ndev->stats;
 
-	if (!mdp->is_opened)
+	if (!netif_running(ndev))
 		return &ndev->stats;
 
 	sh_eth_update_stat(ndev, &ndev->stats.tx_dropped, TROCR);
@@ -2614,8 +2612,6 @@ static int sh_eth_close(struct net_device *ndev)
 	/* Free all the skbuffs in the Rx queue and the DMA buffer. */
 	sh_eth_ring_free(ndev);
 
-	mdp->is_opened = 0;
-
 	pm_runtime_put(&mdp->pdev->dev);
 
 	return 0;
diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h
index a5c07c6ff44a..f56dbc8a064a 100644
--- a/drivers/net/ethernet/renesas/sh_eth.h
+++ b/drivers/net/ethernet/renesas/sh_eth.h
@@ -560,7 +560,6 @@ struct sh_eth_private {
 
 	unsigned no_ether_link:1;
 	unsigned ether_link_active_low:1;
-	unsigned is_opened:1;
 	unsigned wol_enabled:1;
 };
 
-- 
cgit v1.2.3-70-g09d2


From e626dad92383ca16d1d71e66124a272a0cbfe7bd Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 14 Mar 2023 10:59:51 +0100
Subject: wifi: mac80211: fix race in mesh sequence number assignment

Since the sequence number is shared across different tx queues, it needs
to be atomic in order to avoid accidental duplicate assignment

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20230314095956.62085-2-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h | 2 +-
 net/mac80211/mesh.c        | 6 ++----
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index b2535614483e..04017d5c435a 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -696,7 +696,7 @@ struct ieee80211_if_mesh {
 	struct mesh_stats mshstats;
 	struct mesh_config mshcfg;
 	atomic_t estab_plinks;
-	u32 mesh_seqnum;
+	atomic_t mesh_seqnum;
 	bool accepting_plinks;
 	int num_gates;
 	struct beacon_data __rcu *beacon;
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index 5a99b8f6e465..0608ed415831 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -752,10 +752,8 @@ unsigned int ieee80211_new_mesh_header(struct ieee80211_sub_if_data *sdata,
 
 	meshhdr->ttl = sdata->u.mesh.mshcfg.dot11MeshTTL;
 
-	/* FIXME: racy -- TX on multiple queues can be concurrent */
-	put_unaligned(cpu_to_le32(sdata->u.mesh.mesh_seqnum), &meshhdr->seqnum);
-	sdata->u.mesh.mesh_seqnum++;
-
+	put_unaligned_le32(atomic_inc_return(&sdata->u.mesh.mesh_seqnum),
+			   &meshhdr->seqnum);
 	if (addr4or5 && !addr6) {
 		meshhdr->flags |= MESH_FLAGS_AE_A4;
 		memcpy(meshhdr->eaddr1, addr4or5, ETH_ALEN);
-- 
cgit v1.2.3-70-g09d2


From d5edb9ae8d568745f893c5c5fa3837d85311b131 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 14 Mar 2023 10:59:53 +0100
Subject: wifi: mac80211: mesh fast xmit support

Previously, fast xmit only worked on interface types where initially a
sta lookup is performed, and a cached header can be attached to the sta,
requiring only some fields to be updated at runtime.

This technique is not directly applicable for a mesh device type due
to the dynamic nature of the topology and protocol. There are more
addresses that need to be filled, and there is an extra header with a
dynamic length based on the addressing mode.

Change the code to cache entries contain a copy of the mesh subframe header +
bridge tunnel header, as well as an embedded struct ieee80211_fast_tx, which
contains the information for building the 802.11 header.

Add a mesh specific early fast xmit call, which looks up a cached entry and
adds only the mesh subframe header, before passing it over to the generic
fast xmit code.

To ensure the changes in network are reflected in these cached headers,
flush affected cached entries on path changes, as well as other conditions
that currently trigger a fast xmit check in other modes (key changes etc.)

This code is loosely based on a previous implementation by:
Sriram R <quic_srirrama@quicinc.com>

Cc: Sriram R <quic_srirrama@quicinc.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20230314095956.62085-4-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h  |  20 ++++
 net/mac80211/mesh.c         |  92 +++++++++++++++
 net/mac80211/mesh.h         |  44 +++++++
 net/mac80211/mesh_hwmp.c    |  37 ++++--
 net/mac80211/mesh_pathtbl.c | 282 ++++++++++++++++++++++++++++++++++++++++++++
 net/mac80211/rx.c           |  10 +-
 net/mac80211/tx.c           |  47 +++++---
 7 files changed, 507 insertions(+), 25 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 04017d5c435a..4fd4e241f8cd 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -37,6 +37,7 @@
 extern const struct cfg80211_ops mac80211_config_ops;
 
 struct ieee80211_local;
+struct ieee80211_mesh_fast_tx;
 
 /* Maximum number of broadcast/multicast frames to buffer when some of the
  * associated stations are using power saving. */
@@ -656,6 +657,19 @@ struct mesh_table {
 	atomic_t entries;		/* Up to MAX_MESH_NEIGHBOURS */
 };
 
+/**
+ * struct mesh_tx_cache - mesh fast xmit header cache
+ *
+ * @rht: hash table containing struct ieee80211_mesh_fast_tx, using skb DA as key
+ * @walk_head: linked list containing all ieee80211_mesh_fast_tx objects
+ * @walk_lock: lock protecting walk_head and rht
+ */
+struct mesh_tx_cache {
+	struct rhashtable rht;
+	struct hlist_head walk_head;
+	spinlock_t walk_lock;
+};
+
 struct ieee80211_if_mesh {
 	struct timer_list housekeeping_timer;
 	struct timer_list mesh_path_timer;
@@ -734,6 +748,7 @@ struct ieee80211_if_mesh {
 	struct mesh_table mpp_paths; /* Store paths for MPP&MAP */
 	int mesh_paths_generation;
 	int mpp_paths_generation;
+	struct mesh_tx_cache tx_cache;
 };
 
 #ifdef CONFIG_MAC80211_MESH
@@ -2018,6 +2033,11 @@ int ieee80211_tx_control_port(struct wiphy *wiphy, struct net_device *dev,
 			      int link_id, u64 *cookie);
 int ieee80211_probe_mesh_link(struct wiphy *wiphy, struct net_device *dev,
 			      const u8 *buf, size_t len);
+void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
+			   struct sta_info *sta,
+			   struct ieee80211_fast_tx *fast_tx,
+			   struct sk_buff *skb, bool ampdu,
+			   const u8 *da, const u8 *sa);
 
 /* HT */
 void ieee80211_apply_htcap_overrides(struct ieee80211_sub_if_data *sdata,
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index 0608ed415831..6b94cf2a4046 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -10,6 +10,7 @@
 #include <asm/unaligned.h>
 #include "ieee80211_i.h"
 #include "mesh.h"
+#include "wme.h"
 #include "driver-ops.h"
 
 static int mesh_allocated;
@@ -698,6 +699,95 @@ ieee80211_mesh_update_bss_params(struct ieee80211_sub_if_data *sdata,
 			__le32_to_cpu(he_oper->he_oper_params);
 }
 
+bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
+			      struct sk_buff *skb, u32 ctrl_flags)
+{
+	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
+	struct ieee80211_mesh_fast_tx *entry;
+	struct ieee80211s_hdr *meshhdr;
+	u8 sa[ETH_ALEN] __aligned(2);
+	struct tid_ampdu_tx *tid_tx;
+	struct sta_info *sta;
+	bool copy_sa = false;
+	u16 ethertype;
+	u8 tid;
+
+	if (ctrl_flags & IEEE80211_TX_CTRL_SKIP_MPATH_LOOKUP)
+		return false;
+
+	if (ifmsh->mshcfg.dot11MeshNolearn)
+		return false;
+
+	/* Add support for these cases later */
+	if (ifmsh->ps_peers_light_sleep || ifmsh->ps_peers_deep_sleep)
+		return false;
+
+	if (is_multicast_ether_addr(skb->data))
+		return false;
+
+	ethertype = (skb->data[12] << 8) | skb->data[13];
+	if (ethertype < ETH_P_802_3_MIN)
+		return false;
+
+	if (skb->sk && skb_shinfo(skb)->tx_flags & SKBTX_WIFI_STATUS)
+		return false;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		skb_set_transport_header(skb, skb_checksum_start_offset(skb));
+		if (skb_checksum_help(skb))
+			return false;
+	}
+
+	entry = mesh_fast_tx_get(sdata, skb->data);
+	if (!entry)
+		return false;
+
+	if (skb_headroom(skb) < entry->hdrlen + entry->fast_tx.hdr_len)
+		return false;
+
+	sta = rcu_dereference(entry->mpath->next_hop);
+	if (!sta)
+		return false;
+
+	tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
+	tid_tx = rcu_dereference(sta->ampdu_mlme.tid_tx[tid]);
+	if (tid_tx) {
+		if (!test_bit(HT_AGG_STATE_OPERATIONAL, &tid_tx->state))
+			return false;
+		if (tid_tx->timeout)
+			tid_tx->last_tx = jiffies;
+	}
+
+	skb = skb_share_check(skb, GFP_ATOMIC);
+	if (!skb)
+		return true;
+
+	skb_set_queue_mapping(skb, ieee80211_select_queue(sdata, sta, skb));
+
+	meshhdr = (struct ieee80211s_hdr *)entry->hdr;
+	if ((meshhdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) {
+		/* preserve SA from eth header for 6-addr frames */
+		ether_addr_copy(sa, skb->data + ETH_ALEN);
+		copy_sa = true;
+	}
+
+	memcpy(skb_push(skb, entry->hdrlen - 2 * ETH_ALEN), entry->hdr,
+	       entry->hdrlen);
+
+	meshhdr = (struct ieee80211s_hdr *)skb->data;
+	put_unaligned_le32(atomic_inc_return(&sdata->u.mesh.mesh_seqnum),
+			   &meshhdr->seqnum);
+	meshhdr->ttl = sdata->u.mesh.mshcfg.dot11MeshTTL;
+	if (copy_sa)
+	    ether_addr_copy(meshhdr->eaddr2, sa);
+
+	skb_push(skb, 2 * ETH_ALEN);
+	__ieee80211_xmit_fast(sdata, sta, &entry->fast_tx, skb, tid_tx,
+			      entry->mpath->dst, sdata->vif.addr);
+
+	return true;
+}
+
 /**
  * ieee80211_fill_mesh_addresses - fill addresses of a locally originated mesh frame
  * @hdr:	802.11 frame header
@@ -780,6 +870,8 @@ static void ieee80211_mesh_housekeeping(struct ieee80211_sub_if_data *sdata)
 	changed = mesh_accept_plinks_update(sdata);
 	ieee80211_mbss_info_change_notify(sdata, changed);
 
+	mesh_fast_tx_gc(sdata);
+
 	mod_timer(&ifmsh->housekeeping_timer,
 		  round_jiffies(jiffies +
 				IEEE80211_MESH_HOUSEKEEPING_INTERVAL));
diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h
index b2b717a78114..13f394e677ae 100644
--- a/net/mac80211/mesh.h
+++ b/net/mac80211/mesh.h
@@ -122,11 +122,41 @@ struct mesh_path {
 	u8 rann_snd_addr[ETH_ALEN];
 	u32 rann_metric;
 	unsigned long last_preq_to_root;
+	unsigned long fast_tx_check;
 	bool is_root;
 	bool is_gate;
 	u32 path_change_count;
 };
 
+#define MESH_FAST_TX_CACHE_MAX_SIZE		512
+#define MESH_FAST_TX_CACHE_THRESHOLD_SIZE	384
+#define MESH_FAST_TX_CACHE_TIMEOUT		8000 /* msecs */
+
+/**
+ * struct ieee80211_mesh_fast_tx - cached mesh fast tx entry
+ * @rhash: rhashtable pointer
+ * @addr_key: The Ethernet DA which is the key for this entry
+ * @fast_tx: base fast_tx data
+ * @hdr: cached mesh and rfc1042 headers
+ * @hdrlen: length of mesh + rfc1042
+ * @walk_list: list containing all the fast tx entries
+ * @mpath: mesh path corresponding to the Mesh DA
+ * @mppath: MPP entry corresponding to this DA
+ * @timestamp: Last used time of this entry
+ */
+struct ieee80211_mesh_fast_tx {
+	struct rhash_head rhash;
+	u8 addr_key[ETH_ALEN] __aligned(2);
+
+	struct ieee80211_fast_tx fast_tx;
+	u8 hdr[sizeof(struct ieee80211s_hdr) + sizeof(rfc1042_header)];
+	u16 hdrlen;
+
+	struct mesh_path *mpath, *mppath;
+	struct hlist_node walk_list;
+	unsigned long timestamp;
+};
+
 /* Recent multicast cache */
 /* RMC_BUCKETS must be a power of 2, maximum 256 */
 #define RMC_BUCKETS		256
@@ -298,6 +328,20 @@ void mesh_path_discard_frame(struct ieee80211_sub_if_data *sdata,
 void mesh_path_tx_root_frame(struct ieee80211_sub_if_data *sdata);
 
 bool mesh_action_is_path_sel(struct ieee80211_mgmt *mgmt);
+struct ieee80211_mesh_fast_tx *
+mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr);
+bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
+			      struct sk_buff *skb, u32 ctrl_flags);
+void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
+			struct sk_buff *skb, struct mesh_path *mpath);
+void mesh_fast_tx_gc(struct ieee80211_sub_if_data *sdata);
+void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata,
+			     const u8 *addr);
+void mesh_fast_tx_flush_mpath(struct mesh_path *mpath);
+void mesh_fast_tx_flush_sta(struct ieee80211_sub_if_data *sdata,
+			    struct sta_info *sta);
+void mesh_path_refresh(struct ieee80211_sub_if_data *sdata,
+		       struct mesh_path *mpath, const u8 *addr);
 
 #ifdef CONFIG_MAC80211_MESH
 static inline
diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c
index 9b1ce7c3925a..5217e1d97dd6 100644
--- a/net/mac80211/mesh_hwmp.c
+++ b/net/mac80211/mesh_hwmp.c
@@ -394,6 +394,7 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata,
 	u32 orig_sn, orig_metric;
 	unsigned long orig_lifetime, exp_time;
 	u32 last_hop_metric, new_metric;
+	bool flush_mpath = false;
 	bool process = true;
 	u8 hopcount;
 
@@ -491,8 +492,10 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata,
 		}
 
 		if (fresh_info) {
-			if (rcu_access_pointer(mpath->next_hop) != sta)
+			if (rcu_access_pointer(mpath->next_hop) != sta) {
 				mpath->path_change_count++;
+				flush_mpath = true;
+			}
 			mesh_path_assign_nexthop(mpath, sta);
 			mpath->flags |= MESH_PATH_SN_VALID;
 			mpath->metric = new_metric;
@@ -502,6 +505,8 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata,
 			mpath->hop_count = hopcount;
 			mesh_path_activate(mpath);
 			spin_unlock_bh(&mpath->state_lock);
+			if (flush_mpath)
+				mesh_fast_tx_flush_mpath(mpath);
 			ewma_mesh_fail_avg_init(&sta->mesh->fail_avg);
 			/* init it at a low value - 0 start is tricky */
 			ewma_mesh_fail_avg_add(&sta->mesh->fail_avg, 1);
@@ -539,8 +544,10 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata,
 		}
 
 		if (fresh_info) {
-			if (rcu_access_pointer(mpath->next_hop) != sta)
+			if (rcu_access_pointer(mpath->next_hop) != sta) {
 				mpath->path_change_count++;
+				flush_mpath = true;
+			}
 			mesh_path_assign_nexthop(mpath, sta);
 			mpath->metric = last_hop_metric;
 			mpath->exp_time = time_after(mpath->exp_time, exp_time)
@@ -548,6 +555,8 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata,
 			mpath->hop_count = 1;
 			mesh_path_activate(mpath);
 			spin_unlock_bh(&mpath->state_lock);
+			if (flush_mpath)
+				mesh_fast_tx_flush_mpath(mpath);
 			ewma_mesh_fail_avg_init(&sta->mesh->fail_avg);
 			/* init it at a low value - 0 start is tricky */
 			ewma_mesh_fail_avg_add(&sta->mesh->fail_avg, 1);
@@ -1215,6 +1224,20 @@ static int mesh_nexthop_lookup_nolearn(struct ieee80211_sub_if_data *sdata,
 	return 0;
 }
 
+void mesh_path_refresh(struct ieee80211_sub_if_data *sdata,
+		       struct mesh_path *mpath, const u8 *addr)
+{
+	if (mpath->flags & (MESH_PATH_REQ_QUEUED | MESH_PATH_FIXED |
+			    MESH_PATH_RESOLVING))
+		return;
+
+	if (time_after(jiffies,
+		       mpath->exp_time -
+		       msecs_to_jiffies(sdata->u.mesh.mshcfg.path_refresh_time)) &&
+	    (!addr || ether_addr_equal(sdata->vif.addr, addr)))
+		mesh_queue_preq(mpath, PREQ_Q_F_START | PREQ_Q_F_REFRESH);
+}
+
 /**
  * mesh_nexthop_lookup - put the appropriate next hop on a mesh frame. Calling
  * this function is considered "using" the associated mpath, so preempt a path
@@ -1242,19 +1265,15 @@ int mesh_nexthop_lookup(struct ieee80211_sub_if_data *sdata,
 	if (!mpath || !(mpath->flags & MESH_PATH_ACTIVE))
 		return -ENOENT;
 
-	if (time_after(jiffies,
-		       mpath->exp_time -
-		       msecs_to_jiffies(sdata->u.mesh.mshcfg.path_refresh_time)) &&
-	    ether_addr_equal(sdata->vif.addr, hdr->addr4) &&
-	    !(mpath->flags & MESH_PATH_RESOLVING) &&
-	    !(mpath->flags & MESH_PATH_FIXED))
-		mesh_queue_preq(mpath, PREQ_Q_F_START | PREQ_Q_F_REFRESH);
+	mesh_path_refresh(sdata, mpath, hdr->addr4);
 
 	next_hop = rcu_dereference(mpath->next_hop);
 	if (next_hop) {
 		memcpy(hdr->addr1, next_hop->sta.addr, ETH_ALEN);
 		memcpy(hdr->addr2, sdata->vif.addr, ETH_ALEN);
 		ieee80211_mps_set_frame_flags(sdata, next_hop, hdr);
+		if (ieee80211_hw_check(&sdata->local->hw, SUPPORT_FAST_XMIT))
+			mesh_fast_tx_cache(sdata, skb, mpath);
 		return 0;
 	}
 
diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c
index 3b81e6df3f34..d32e304eeb4b 100644
--- a/net/mac80211/mesh_pathtbl.c
+++ b/net/mac80211/mesh_pathtbl.c
@@ -14,6 +14,7 @@
 #include "wme.h"
 #include "ieee80211_i.h"
 #include "mesh.h"
+#include <linux/rhashtable.h>
 
 static void mesh_path_free_rcu(struct mesh_table *tbl, struct mesh_path *mpath);
 
@@ -32,6 +33,41 @@ static const struct rhashtable_params mesh_rht_params = {
 	.hashfn = mesh_table_hash,
 };
 
+static const struct rhashtable_params fast_tx_rht_params = {
+	.nelem_hint = 10,
+	.automatic_shrinking = true,
+	.key_len = ETH_ALEN,
+	.key_offset = offsetof(struct ieee80211_mesh_fast_tx, addr_key),
+	.head_offset = offsetof(struct ieee80211_mesh_fast_tx, rhash),
+	.hashfn = mesh_table_hash,
+};
+
+static void __mesh_fast_tx_entry_free(void *ptr, void *tblptr)
+{
+	struct ieee80211_mesh_fast_tx *entry = ptr;
+
+	kfree_rcu(entry, fast_tx.rcu_head);
+}
+
+static void mesh_fast_tx_deinit(struct ieee80211_sub_if_data *sdata)
+{
+	struct mesh_tx_cache *cache;
+
+	cache = &sdata->u.mesh.tx_cache;
+	rhashtable_free_and_destroy(&cache->rht,
+				    __mesh_fast_tx_entry_free, NULL);
+}
+
+static void mesh_fast_tx_init(struct ieee80211_sub_if_data *sdata)
+{
+	struct mesh_tx_cache *cache;
+
+	cache = &sdata->u.mesh.tx_cache;
+	rhashtable_init(&cache->rht, &fast_tx_rht_params);
+	INIT_HLIST_HEAD(&cache->walk_head);
+	spin_lock_init(&cache->walk_lock);
+}
+
 static inline bool mpath_expired(struct mesh_path *mpath)
 {
 	return (mpath->flags & MESH_PATH_ACTIVE) &&
@@ -381,6 +417,243 @@ struct mesh_path *mesh_path_new(struct ieee80211_sub_if_data *sdata,
 	return new_mpath;
 }
 
+static void mesh_fast_tx_entry_free(struct mesh_tx_cache *cache,
+				    struct ieee80211_mesh_fast_tx *entry)
+{
+	hlist_del_rcu(&entry->walk_list);
+	rhashtable_remove_fast(&cache->rht, &entry->rhash, fast_tx_rht_params);
+	kfree_rcu(entry, fast_tx.rcu_head);
+}
+
+struct ieee80211_mesh_fast_tx *
+mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr)
+{
+	struct ieee80211_mesh_fast_tx *entry;
+	struct mesh_tx_cache *cache;
+
+	cache = &sdata->u.mesh.tx_cache;
+	entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+	if (!entry)
+		return NULL;
+
+	if (!(entry->mpath->flags & MESH_PATH_ACTIVE) ||
+	    mpath_expired(entry->mpath)) {
+		spin_lock_bh(&cache->walk_lock);
+		entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+		if (entry)
+		    mesh_fast_tx_entry_free(cache, entry);
+		spin_unlock_bh(&cache->walk_lock);
+		return NULL;
+	}
+
+	mesh_path_refresh(sdata, entry->mpath, NULL);
+	if (entry->mppath)
+		entry->mppath->exp_time = jiffies;
+	entry->timestamp = jiffies;
+
+	return entry;
+}
+
+void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
+			struct sk_buff *skb, struct mesh_path *mpath)
+{
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_mesh_fast_tx *entry, *prev;
+	struct ieee80211_mesh_fast_tx build = {};
+	struct ieee80211s_hdr *meshhdr;
+	struct mesh_tx_cache *cache;
+	struct ieee80211_key *key;
+	struct mesh_path *mppath;
+	struct sta_info *sta;
+	u8 *qc;
+
+	if (sdata->noack_map ||
+	    !ieee80211_is_data_qos(hdr->frame_control))
+		return;
+
+	build.fast_tx.hdr_len = ieee80211_hdrlen(hdr->frame_control);
+	meshhdr = (struct ieee80211s_hdr *)(skb->data + build.fast_tx.hdr_len);
+	build.hdrlen = ieee80211_get_mesh_hdrlen(meshhdr);
+
+	cache = &sdata->u.mesh.tx_cache;
+	if (atomic_read(&cache->rht.nelems) >= MESH_FAST_TX_CACHE_MAX_SIZE)
+		return;
+
+	sta = rcu_dereference(mpath->next_hop);
+	if (!sta)
+		return;
+
+	if ((meshhdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) {
+		/* This is required to keep the mppath alive */
+		mppath = mpp_path_lookup(sdata, meshhdr->eaddr1);
+		if (!mppath)
+			return;
+		build.mppath = mppath;
+	} else if (ieee80211_has_a4(hdr->frame_control)) {
+		mppath = mpath;
+	} else {
+		return;
+	}
+
+	/* rate limit, in case fast xmit can't be enabled */
+	if (mppath->fast_tx_check == jiffies)
+		return;
+
+	mppath->fast_tx_check = jiffies;
+
+	/*
+	 * Same use of the sta lock as in ieee80211_check_fast_xmit, in order
+	 * to protect against concurrent sta key updates.
+	 */
+	spin_lock_bh(&sta->lock);
+	key = rcu_access_pointer(sta->ptk[sta->ptk_idx]);
+	if (!key)
+		key = rcu_access_pointer(sdata->default_unicast_key);
+	build.fast_tx.key = key;
+
+	if (key) {
+		bool gen_iv, iv_spc;
+
+		gen_iv = key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV;
+		iv_spc = key->conf.flags & IEEE80211_KEY_FLAG_PUT_IV_SPACE;
+
+		if (!(key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) ||
+		    (key->flags & KEY_FLAG_TAINTED))
+			goto unlock_sta;
+
+		switch (key->conf.cipher) {
+		case WLAN_CIPHER_SUITE_CCMP:
+		case WLAN_CIPHER_SUITE_CCMP_256:
+			if (gen_iv)
+				build.fast_tx.pn_offs = build.fast_tx.hdr_len;
+			if (gen_iv || iv_spc)
+				build.fast_tx.hdr_len += IEEE80211_CCMP_HDR_LEN;
+			break;
+		case WLAN_CIPHER_SUITE_GCMP:
+		case WLAN_CIPHER_SUITE_GCMP_256:
+			if (gen_iv)
+				build.fast_tx.pn_offs = build.fast_tx.hdr_len;
+			if (gen_iv || iv_spc)
+				build.fast_tx.hdr_len += IEEE80211_GCMP_HDR_LEN;
+			break;
+		default:
+			goto unlock_sta;
+		}
+	}
+
+	memcpy(build.addr_key, mppath->dst, ETH_ALEN);
+	build.timestamp = jiffies;
+	build.fast_tx.band = info->band;
+	build.fast_tx.da_offs = offsetof(struct ieee80211_hdr, addr3);
+	build.fast_tx.sa_offs = offsetof(struct ieee80211_hdr, addr4);
+	build.mpath = mpath;
+	memcpy(build.hdr, meshhdr, build.hdrlen);
+	memcpy(build.hdr + build.hdrlen, rfc1042_header, sizeof(rfc1042_header));
+	build.hdrlen += sizeof(rfc1042_header);
+	memcpy(build.fast_tx.hdr, hdr, build.fast_tx.hdr_len);
+
+	hdr = (struct ieee80211_hdr *)build.fast_tx.hdr;
+	if (build.fast_tx.key)
+		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PROTECTED);
+
+	qc = ieee80211_get_qos_ctl(hdr);
+	qc[1] |= IEEE80211_QOS_CTL_MESH_CONTROL_PRESENT >> 8;
+
+	entry = kmemdup(&build, sizeof(build), GFP_ATOMIC);
+	if (!entry)
+		goto unlock_sta;
+
+	spin_lock(&cache->walk_lock);
+	prev = rhashtable_lookup_get_insert_fast(&cache->rht,
+						 &entry->rhash,
+						 fast_tx_rht_params);
+	if (unlikely(IS_ERR(prev))) {
+		kfree(entry);
+		goto unlock_cache;
+	}
+
+	/*
+	 * replace any previous entry in the hash table, in case we're
+	 * replacing it with a different type (e.g. mpath -> mpp)
+	 */
+	if (unlikely(prev)) {
+		rhashtable_replace_fast(&cache->rht, &prev->rhash,
+					&entry->rhash, fast_tx_rht_params);
+		hlist_del_rcu(&prev->walk_list);
+		kfree_rcu(prev, fast_tx.rcu_head);
+	}
+
+	hlist_add_head(&entry->walk_list, &cache->walk_head);
+
+unlock_cache:
+	spin_unlock(&cache->walk_lock);
+unlock_sta:
+	spin_unlock_bh(&sta->lock);
+}
+
+void mesh_fast_tx_gc(struct ieee80211_sub_if_data *sdata)
+{
+	unsigned long timeout = msecs_to_jiffies(MESH_FAST_TX_CACHE_TIMEOUT);
+	struct mesh_tx_cache *cache;
+	struct ieee80211_mesh_fast_tx *entry;
+	struct hlist_node *n;
+
+	cache = &sdata->u.mesh.tx_cache;
+	if (atomic_read(&cache->rht.nelems) < MESH_FAST_TX_CACHE_THRESHOLD_SIZE)
+		return;
+
+	spin_lock_bh(&cache->walk_lock);
+	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
+		if (!time_is_after_jiffies(entry->timestamp + timeout))
+			mesh_fast_tx_entry_free(cache, entry);
+	spin_unlock_bh(&cache->walk_lock);
+}
+
+void mesh_fast_tx_flush_mpath(struct mesh_path *mpath)
+{
+	struct ieee80211_sub_if_data *sdata = mpath->sdata;
+	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
+	struct ieee80211_mesh_fast_tx *entry;
+	struct hlist_node *n;
+
+	cache = &sdata->u.mesh.tx_cache;
+	spin_lock_bh(&cache->walk_lock);
+	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
+		if (entry->mpath == mpath)
+			mesh_fast_tx_entry_free(cache, entry);
+	spin_unlock_bh(&cache->walk_lock);
+}
+
+void mesh_fast_tx_flush_sta(struct ieee80211_sub_if_data *sdata,
+			    struct sta_info *sta)
+{
+	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
+	struct ieee80211_mesh_fast_tx *entry;
+	struct hlist_node *n;
+
+	cache = &sdata->u.mesh.tx_cache;
+	spin_lock_bh(&cache->walk_lock);
+	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
+		if (rcu_access_pointer(entry->mpath->next_hop) == sta)
+			mesh_fast_tx_entry_free(cache, entry);
+	spin_unlock_bh(&cache->walk_lock);
+}
+
+void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata,
+			     const u8 *addr)
+{
+	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
+	struct ieee80211_mesh_fast_tx *entry;
+
+	cache = &sdata->u.mesh.tx_cache;
+	spin_lock_bh(&cache->walk_lock);
+	entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+	if (entry)
+		mesh_fast_tx_entry_free(cache, entry);
+	spin_unlock_bh(&cache->walk_lock);
+}
+
 /**
  * mesh_path_add - allocate and add a new path to the mesh path table
  * @dst: destination address of the path (ETH_ALEN length)
@@ -464,6 +737,8 @@ int mpp_path_add(struct ieee80211_sub_if_data *sdata,
 
 	if (ret)
 		kfree(new_mpath);
+	else
+		mesh_fast_tx_flush_addr(sdata, dst);
 
 	sdata->u.mesh.mpp_paths_generation++;
 	return ret;
@@ -523,6 +798,10 @@ static void __mesh_path_del(struct mesh_table *tbl, struct mesh_path *mpath)
 {
 	hlist_del_rcu(&mpath->walk_list);
 	rhashtable_remove_fast(&tbl->rhead, &mpath->rhash, mesh_rht_params);
+	if (tbl == &mpath->sdata->u.mesh.mpp_paths)
+		mesh_fast_tx_flush_addr(mpath->sdata, mpath->dst);
+	else
+		mesh_fast_tx_flush_mpath(mpath);
 	mesh_path_free_rcu(tbl, mpath);
 }
 
@@ -747,6 +1026,7 @@ void mesh_path_fix_nexthop(struct mesh_path *mpath, struct sta_info *next_hop)
 	mpath->exp_time = 0;
 	mpath->flags = MESH_PATH_FIXED | MESH_PATH_SN_VALID;
 	mesh_path_activate(mpath);
+	mesh_fast_tx_flush_mpath(mpath);
 	spin_unlock_bh(&mpath->state_lock);
 	ewma_mesh_fail_avg_init(&next_hop->mesh->fail_avg);
 	/* init it at a low value - 0 start is tricky */
@@ -758,6 +1038,7 @@ void mesh_pathtbl_init(struct ieee80211_sub_if_data *sdata)
 {
 	mesh_table_init(&sdata->u.mesh.mesh_paths);
 	mesh_table_init(&sdata->u.mesh.mpp_paths);
+	mesh_fast_tx_init(sdata);
 }
 
 static
@@ -785,6 +1066,7 @@ void mesh_path_expire(struct ieee80211_sub_if_data *sdata)
 
 void mesh_pathtbl_unregister(struct ieee80211_sub_if_data *sdata)
 {
+	mesh_fast_tx_deinit(sdata);
 	mesh_table_free(&sdata->u.mesh.mesh_paths);
 	mesh_table_free(&sdata->u.mesh.mpp_paths);
 }
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 0255c5745e1c..b22d736285bf 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2772,6 +2772,7 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta
 	if (mesh_hdr->flags & MESH_FLAGS_AE) {
 		struct mesh_path *mppath;
 		char *proxied_addr;
+		bool update = false;
 
 		if (multicast)
 			proxied_addr = mesh_hdr->eaddr1;
@@ -2787,11 +2788,18 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta
 			mpp_path_add(sdata, proxied_addr, eth->h_source);
 		} else {
 			spin_lock_bh(&mppath->state_lock);
-			if (!ether_addr_equal(mppath->mpp, eth->h_source))
+			if (!ether_addr_equal(mppath->mpp, eth->h_source)) {
 				memcpy(mppath->mpp, eth->h_source, ETH_ALEN);
+				update = true;
+			}
 			mppath->exp_time = jiffies;
 			spin_unlock_bh(&mppath->state_lock);
 		}
+
+		/* flush fast xmit cache if the address path changed */
+		if (update)
+			mesh_fast_tx_flush_addr(sdata, proxied_addr);
+
 		rcu_read_unlock();
 	}
 
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 628d60f3db2a..ad6193693363 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -3019,6 +3019,9 @@ void ieee80211_check_fast_xmit(struct sta_info *sta)
 	if (!ieee80211_hw_check(&local->hw, SUPPORT_FAST_XMIT))
 		return;
 
+	if (ieee80211_vif_is_mesh(&sdata->vif))
+		mesh_fast_tx_flush_sta(sdata, sta);
+
 	/* Locking here protects both the pointer itself, and against concurrent
 	 * invocations winning data access races to, e.g., the key pointer that
 	 * is used.
@@ -3371,7 +3374,8 @@ static bool ieee80211_amsdu_prepare_head(struct ieee80211_sub_if_data *sdata,
 static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,
 				      struct sta_info *sta,
 				      struct ieee80211_fast_tx *fast_tx,
-				      struct sk_buff *skb)
+				      struct sk_buff *skb,
+				      const u8 *da, const u8 *sa)
 {
 	struct ieee80211_local *local = sdata->local;
 	struct fq *fq = &local->fq;
@@ -3400,6 +3404,9 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,
 	if (sdata->vif.offload_flags & IEEE80211_OFFLOAD_ENCAP_ENABLED)
 		return false;
 
+	if (ieee80211_vif_is_mesh(&sdata->vif))
+		return false;
+
 	if (skb_is_gso(skb))
 		return false;
 
@@ -3484,7 +3491,8 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,
 
 	ret = true;
 	data = skb_push(skb, ETH_ALEN + 2);
-	memmove(data, data + ETH_ALEN + 2, 2 * ETH_ALEN);
+	ether_addr_copy(data, da);
+	ether_addr_copy(data + ETH_ALEN, sa);
 
 	data += 2 * ETH_ALEN;
 	len = cpu_to_be16(subframe_len);
@@ -3632,10 +3640,11 @@ free:
 	return NULL;
 }
 
-static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
-				  struct sta_info *sta,
-				  struct ieee80211_fast_tx *fast_tx,
-				  struct sk_buff *skb, u8 tid, bool ampdu)
+void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
+			   struct sta_info *sta,
+			   struct ieee80211_fast_tx *fast_tx,
+			   struct sk_buff *skb, bool ampdu,
+			   const u8 *da, const u8 *sa)
 {
 	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_hdr *hdr = (void *)fast_tx->hdr;
@@ -3644,14 +3653,13 @@ static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 	ieee80211_tx_result r;
 	int hw_headroom = sdata->local->hw.extra_tx_headroom;
 	int extra_head = fast_tx->hdr_len - (ETH_HLEN - 2);
-	struct ethhdr eth;
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (unlikely(!skb))
 		return;
 
 	if ((hdr->frame_control & cpu_to_le16(IEEE80211_STYPE_QOS_DATA)) &&
-	    ieee80211_amsdu_aggregate(sdata, sta, fast_tx, skb))
+	    ieee80211_amsdu_aggregate(sdata, sta, fast_tx, skb, da, sa))
 		return;
 
 	/* will not be crypto-handled beyond what we do here, so use false
@@ -3664,11 +3672,10 @@ static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 					  ENCRYPT_NO)))
 		goto free;
 
-	memcpy(&eth, skb->data, ETH_HLEN - 2);
 	hdr = skb_push(skb, extra_head);
 	memcpy(skb->data, fast_tx->hdr, fast_tx->hdr_len);
-	memcpy(skb->data + fast_tx->da_offs, eth.h_dest, ETH_ALEN);
-	memcpy(skb->data + fast_tx->sa_offs, eth.h_source, ETH_ALEN);
+	memcpy(skb->data + fast_tx->da_offs, da, ETH_ALEN);
+	memcpy(skb->data + fast_tx->sa_offs, sa, ETH_ALEN);
 
 	info = IEEE80211_SKB_CB(skb);
 	memset(info, 0, sizeof(*info));
@@ -3686,7 +3693,8 @@ static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 #endif
 
 	if (hdr->frame_control & cpu_to_le16(IEEE80211_STYPE_QOS_DATA)) {
-		tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
+		u8 tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
+
 		*ieee80211_get_qos_ctl(hdr) = tid;
 	}
 
@@ -3729,6 +3737,7 @@ static bool ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 	struct ieee80211_hdr *hdr = (void *)fast_tx->hdr;
 	struct tid_ampdu_tx *tid_tx = NULL;
 	struct sk_buff *next;
+	struct ethhdr eth;
 	u8 tid = IEEE80211_NUM_TIDS;
 
 	/* control port protocol needs a lot of special handling */
@@ -3754,6 +3763,8 @@ static bool ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 		}
 	}
 
+	memcpy(&eth, skb->data, ETH_HLEN - 2);
+
 	/* after this point (skb is modified) we cannot return false */
 	skb = ieee80211_tx_skb_fixup(skb, ieee80211_sdata_netdev_features(sdata));
 	if (!skb)
@@ -3761,7 +3772,8 @@ static bool ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 
 	skb_list_walk_safe(skb, skb, next) {
 		skb_mark_not_on_list(skb);
-		__ieee80211_xmit_fast(sdata, sta, fast_tx, skb, tid, tid_tx);
+		__ieee80211_xmit_fast(sdata, sta, fast_tx, skb, tid_tx,
+				      eth.h_dest, eth.h_source);
 	}
 
 	return true;
@@ -4245,8 +4257,15 @@ void __ieee80211_subif_start_xmit(struct sk_buff *skb,
 		return;
 	}
 
+	sk_pacing_shift_update(skb->sk, sdata->local->hw.tx_sk_pacing_shift);
+
 	rcu_read_lock();
 
+	if (ieee80211_vif_is_mesh(&sdata->vif) &&
+	    ieee80211_hw_check(&local->hw, SUPPORT_FAST_XMIT) &&
+	    ieee80211_mesh_xmit_fast(sdata, skb, ctrl_flags))
+		goto out;
+
 	if (ieee80211_lookup_ra_sta(sdata, skb, &sta))
 		goto out_free;
 
@@ -4256,8 +4275,6 @@ void __ieee80211_subif_start_xmit(struct sk_buff *skb,
 	skb_set_queue_mapping(skb, ieee80211_select_queue(sdata, sta, skb));
 	ieee80211_aggr_check(sdata, sta, skb);
 
-	sk_pacing_shift_update(skb->sk, sdata->local->hw.tx_sk_pacing_shift);
-
 	if (sta) {
 		struct ieee80211_fast_tx *fast_tx;
 
-- 
cgit v1.2.3-70-g09d2


From 8b0f5cb6bc7cbbee4d78b3221683dcb4d1ed23d0 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 14 Mar 2023 10:59:54 +0100
Subject: wifi: mac80211: use mesh header cache to speed up mesh forwarding

Significantly reduces mesh forwarding path CPU usage and enables the
direct use of iTXQ.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/20230314095956.62085-5-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  2 ++
 net/mac80211/rx.c          | 65 +++++++++++++++++++++++++++++++++++++++++++++-
 net/mac80211/tx.c          |  6 ++---
 3 files changed, 68 insertions(+), 5 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 4fd4e241f8cd..c8c037a9e4e2 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2038,6 +2038,8 @@ void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			   struct ieee80211_fast_tx *fast_tx,
 			   struct sk_buff *skb, bool ampdu,
 			   const u8 *da, const u8 *sa);
+void ieee80211_aggr_check(struct ieee80211_sub_if_data *sdata,
+			  struct sta_info *sta, struct sk_buff *skb);
 
 /* HT */
 void ieee80211_apply_htcap_overrides(struct ieee80211_sub_if_data *sdata,
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index b22d736285bf..f6e9a66046a6 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2701,6 +2701,65 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx)
 	}
 }
 
+#ifdef CONFIG_MAC80211_MESH
+static bool
+ieee80211_rx_mesh_fast_forward(struct ieee80211_sub_if_data *sdata,
+			       struct sk_buff *skb, int hdrlen)
+{
+	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
+	struct ieee80211_mesh_fast_tx *entry = NULL;
+	struct ieee80211s_hdr *mesh_hdr;
+	struct tid_ampdu_tx *tid_tx;
+	struct sta_info *sta;
+	struct ethhdr eth;
+	u8 tid;
+
+	mesh_hdr = (struct ieee80211s_hdr *)(skb->data + sizeof(eth));
+	if ((mesh_hdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6)
+		entry = mesh_fast_tx_get(sdata, mesh_hdr->eaddr1);
+	else if (!(mesh_hdr->flags & MESH_FLAGS_AE))
+		entry = mesh_fast_tx_get(sdata, skb->data);
+	if (!entry)
+		return false;
+
+	sta = rcu_dereference(entry->mpath->next_hop);
+	if (!sta)
+		return false;
+
+	if (skb_linearize(skb))
+		return false;
+
+	tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
+	tid_tx = rcu_dereference(sta->ampdu_mlme.tid_tx[tid]);
+	if (tid_tx) {
+		if (!test_bit(HT_AGG_STATE_OPERATIONAL, &tid_tx->state))
+			return false;
+
+		if (tid_tx->timeout)
+			tid_tx->last_tx = jiffies;
+	}
+
+	ieee80211_aggr_check(sdata, sta, skb);
+
+	if (ieee80211_get_8023_tunnel_proto(skb->data + hdrlen,
+					    &skb->protocol))
+		hdrlen += ETH_ALEN;
+	else
+		skb->protocol = htons(skb->len - hdrlen);
+	skb_set_network_header(skb, hdrlen + 2);
+
+	skb->dev = sdata->dev;
+	memcpy(&eth, skb->data, ETH_HLEN - 2);
+	skb_pull(skb, 2);
+	__ieee80211_xmit_fast(sdata, sta, &entry->fast_tx, skb, tid_tx,
+			      eth.h_dest, eth.h_source);
+	IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_unicast);
+	IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_frames);
+
+	return true;
+}
+#endif
+
 static ieee80211_rx_result
 ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta,
 		       struct sk_buff *skb)
@@ -2805,6 +2864,10 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta
 
 	skb_set_queue_mapping(skb, ieee802_1d_to_ac[skb->priority]);
 
+	if (!multicast &&
+	    ieee80211_rx_mesh_fast_forward(sdata, skb, mesh_hdrlen))
+		return RX_QUEUED;
+
 	ieee80211_fill_mesh_addresses(&hdr, &hdr.frame_control,
 				      eth->h_dest, eth->h_source);
 	hdrlen = ieee80211_hdrlen(hdr.frame_control);
@@ -2843,6 +2906,7 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta
 	info->control.flags |= IEEE80211_TX_INTCFL_NEED_TXPROCESSING;
 	info->control.vif = &sdata->vif;
 	info->control.jiffies = jiffies;
+	fwd_skb->dev = sdata->dev;
 	if (multicast) {
 		IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_mcast);
 		memcpy(fwd_hdr->addr2, sdata->vif.addr, ETH_ALEN);
@@ -2864,7 +2928,6 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta
 	}
 
 	IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_frames);
-	fwd_skb->dev = sdata->dev;
 	ieee80211_add_pending_skb(local, fwd_skb);
 
 rx_accept:
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index ad6193693363..de17926484bd 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -1189,10 +1189,8 @@ static bool ieee80211_tx_prep_agg(struct ieee80211_tx_data *tx,
 	return queued;
 }
 
-static void
-ieee80211_aggr_check(struct ieee80211_sub_if_data *sdata,
-		     struct sta_info *sta,
-		     struct sk_buff *skb)
+void ieee80211_aggr_check(struct ieee80211_sub_if_data *sdata,
+			  struct sta_info *sta, struct sk_buff *skb)
 {
 	struct rate_control_ref *ref = sdata->local->rate_ctrl;
 	u16 tid;
-- 
cgit v1.2.3-70-g09d2


From 3468e1e0c639032a603450f0830ccabfa76f5806 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 14 Mar 2023 10:59:55 +0100
Subject: wifi: mac80211: add mesh fast-rx support

This helps bring down rx CPU usage by avoiding calls to the rx handlers in
the slow path. Supports forwarding and local rx, including A-MSDU.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/20230314095956.62085-6-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/rx.c | 34 +++++++++++++++++++++++++++++-----
 1 file changed, 29 insertions(+), 5 deletions(-)

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index f6e9a66046a6..85fb1d3eeb2f 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -4542,6 +4542,12 @@ void ieee80211_check_fast_rx(struct sta_info *sta)
 			fastrx.internal_forward = 0;
 		}
 
+		break;
+	case NL80211_IFTYPE_MESH_POINT:
+		fastrx.expected_ds_bits = cpu_to_le16(IEEE80211_FCTL_FROMDS |
+						      IEEE80211_FCTL_TODS);
+		fastrx.da_offs = offsetof(struct ieee80211_hdr, addr3);
+		fastrx.sa_offs = offsetof(struct ieee80211_hdr, addr4);
 		break;
 	default:
 		goto clear;
@@ -4751,6 +4757,7 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx,
 	struct sk_buff *skb = rx->skb;
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb);
+	static ieee80211_rx_result res;
 	int orig_len = skb->len;
 	int hdrlen = ieee80211_hdrlen(hdr->frame_control);
 	int snap_offs = hdrlen;
@@ -4812,7 +4819,8 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx,
 		snap_offs += IEEE80211_CCMP_HDR_LEN;
 	}
 
-	if (!(status->rx_flags & IEEE80211_RX_AMSDU)) {
+	if (!ieee80211_vif_is_mesh(&rx->sdata->vif) &&
+	    !(status->rx_flags & IEEE80211_RX_AMSDU)) {
 		if (!pskb_may_pull(skb, snap_offs + sizeof(*payload)))
 			return false;
 
@@ -4851,13 +4859,29 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx,
 	/* do the header conversion - first grab the addresses */
 	ether_addr_copy(addrs.da, skb->data + fast_rx->da_offs);
 	ether_addr_copy(addrs.sa, skb->data + fast_rx->sa_offs);
-	skb_postpull_rcsum(skb, skb->data + snap_offs,
-			   sizeof(rfc1042_header) + 2);
-	/* remove the SNAP but leave the ethertype */
-	skb_pull(skb, snap_offs + sizeof(rfc1042_header));
+	if (ieee80211_vif_is_mesh(&rx->sdata->vif)) {
+	    skb_pull(skb, snap_offs - 2);
+	    put_unaligned_be16(skb->len - 2, skb->data);
+	} else {
+	    skb_postpull_rcsum(skb, skb->data + snap_offs,
+			       sizeof(rfc1042_header) + 2);
+
+	    /* remove the SNAP but leave the ethertype */
+	    skb_pull(skb, snap_offs + sizeof(rfc1042_header));
+	}
 	/* push the addresses in front */
 	memcpy(skb_push(skb, sizeof(addrs)), &addrs, sizeof(addrs));
 
+	res = ieee80211_rx_mesh_data(rx->sdata, rx->sta, rx->skb);
+	switch (res) {
+	case RX_QUEUED:
+		return true;
+	case RX_CONTINUE:
+		break;
+	default:
+		goto drop;
+	}
+
 	ieee80211_rx_8023(rx, fast_rx, orig_len);
 
 	return true;
-- 
cgit v1.2.3-70-g09d2


From fe4a6d2db3bad41e9f22c860596f355af8493ebb Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 14 Mar 2023 10:59:56 +0100
Subject: wifi: mac80211: implement support for yet another mesh A-MSDU format

MT7996 hardware supports mesh A-MSDU subframes in hardware, but uses a
big-endian length field

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/20230314095956.62085-7-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h  | 11 +++++++----
 net/mac80211/rx.c       | 22 ++++++++++++++++------
 net/mac80211/sta_info.h |  5 ++++-
 net/wireless/util.c     | 36 +++++++++++++++++++++++-------------
 4 files changed, 50 insertions(+), 24 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 7cebba1c4135..86cb048dc924 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -6274,10 +6274,13 @@ static inline int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
  * mesh control field.
  *
  * @skb: The input A-MSDU frame without any headers.
- * @mesh_hdr: use standard compliant mesh A-MSDU subframe header
+ * @mesh_hdr: the type of mesh header to test
+ *	0: non-mesh A-MSDU length field
+ *	1: big-endian mesh A-MSDU length field
+ *	2: little-endian mesh A-MSDU length field
  * Returns: true if subframe header lengths are valid for the @mesh_hdr mode
  */
-bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr);
+bool ieee80211_is_valid_amsdu(struct sk_buff *skb, u8 mesh_hdr);
 
 /**
  * ieee80211_amsdu_to_8023s - decode an IEEE 802.11n A-MSDU frame
@@ -6294,13 +6297,13 @@ bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr);
  * @extra_headroom: The hardware extra headroom for SKBs in the @list.
  * @check_da: DA to check in the inner ethernet header, or NULL
  * @check_sa: SA to check in the inner ethernet header, or NULL
- * @mesh_control: A-MSDU subframe header includes the mesh control field
+ * @mesh_control: see mesh_hdr in ieee80211_is_valid_amsdu
  */
 void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 			      const u8 *addr, enum nl80211_iftype iftype,
 			      const unsigned int extra_headroom,
 			      const u8 *check_da, const u8 *check_sa,
-			      bool mesh_control);
+			      u8 mesh_control);
 
 /**
  * ieee80211_get_8023_tunnel_proto - get RFC1042 or bridge tunnel encap protocol
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 85fb1d3eeb2f..1c957194554b 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2983,13 +2983,23 @@ __ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx, u8 data_offset)
 		return RX_DROP_UNUSABLE;
 
 	if (rx->sta && rx->sta->amsdu_mesh_control < 0) {
-		bool valid_std = ieee80211_is_valid_amsdu(skb, true);
-		bool valid_nonstd = ieee80211_is_valid_amsdu(skb, false);
+		s8 valid = -1;
+		int i;
+
+		for (i = 0; i <= 2; i++) {
+			if (!ieee80211_is_valid_amsdu(skb, i))
+				continue;
+
+			if (valid >= 0) {
+				/* ambiguous */
+				valid = -1;
+				break;
+			}
+
+			valid = i;
+		}
 
-		if (valid_std && !valid_nonstd)
-			rx->sta->amsdu_mesh_control = 1;
-		else if (valid_nonstd && !valid_std)
-			rx->sta->amsdu_mesh_control = 0;
+		rx->sta->amsdu_mesh_control = valid;
 	}
 
 	ieee80211_amsdu_to_8023s(skb, &frame_list, dev->dev_addr,
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index e8e482a82d77..f354d470e174 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -623,7 +623,10 @@ struct link_sta_info {
  * @cparams: CoDel parameters for this station.
  * @reserved_tid: reserved TID (if any, otherwise IEEE80211_TID_UNRESERVED)
  * @amsdu_mesh_control: track the mesh A-MSDU format used by the peer
- *	(-1: not yet known, 0: non-standard [without mesh header], 1: standard)
+ *	(-1: not yet known,
+ *	  0: non-mesh A-MSDU length field
+ *	  1: big-endian mesh A-MSDU length field
+ *	  2: little-endian mesh A-MSDU length field)
  * @fast_tx: TX fastpath information
  * @fast_rx: RX fastpath information
  * @tdls_chandef: a TDLS peer can have a wider chandef that is compatible to
diff --git a/net/wireless/util.c b/net/wireless/util.c
index d1a89e82ead0..3bc0c3072e78 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -776,7 +776,24 @@ __ieee80211_amsdu_copy(struct sk_buff *skb, unsigned int hlen,
 	return frame;
 }
 
-bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr)
+static u16
+ieee80211_amsdu_subframe_length(void *field, u8 mesh_flags, u8 hdr_type)
+{
+	__le16 *field_le = field;
+	__be16 *field_be = field;
+	u16 len;
+
+	if (hdr_type >= 2)
+		len = le16_to_cpu(*field_le);
+	else
+		len = be16_to_cpu(*field_be);
+	if (hdr_type)
+		len += __ieee80211_get_mesh_hdrlen(mesh_flags);
+
+	return len;
+}
+
+bool ieee80211_is_valid_amsdu(struct sk_buff *skb, u8 mesh_hdr)
 {
 	int offset = 0, remaining, subframe_len, padding;
 
@@ -790,12 +807,8 @@ bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr)
 		if (skb_copy_bits(skb, offset + 2 * ETH_ALEN, &hdr, sizeof(hdr)) < 0)
 			return false;
 
-		if (mesh_hdr)
-			len = le16_to_cpu(*(__le16 *)&hdr.len) +
-			      __ieee80211_get_mesh_hdrlen(hdr.mesh_flags);
-		else
-			len = ntohs(hdr.len);
-
+		len = ieee80211_amsdu_subframe_length(&hdr.len, hdr.mesh_flags,
+						      mesh_hdr);
 		subframe_len = sizeof(struct ethhdr) + len;
 		padding = (4 - subframe_len) & 0x3;
 		remaining = skb->len - offset;
@@ -812,7 +825,7 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 			      const u8 *addr, enum nl80211_iftype iftype,
 			      const unsigned int extra_headroom,
 			      const u8 *check_da, const u8 *check_sa,
-			      bool mesh_control)
+			      u8 mesh_control)
 {
 	unsigned int hlen = ALIGN(extra_headroom, 4);
 	struct sk_buff *frame = NULL;
@@ -837,11 +850,8 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 		skb_copy_bits(skb, offset, &hdr, copy_len);
 		if (iftype == NL80211_IFTYPE_MESH_POINT)
 			mesh_len = __ieee80211_get_mesh_hdrlen(hdr.flags);
-		if (mesh_control)
-			len = le16_to_cpu(*(__le16 *)&hdr.eth.h_proto) + mesh_len;
-		else
-			len = ntohs(hdr.eth.h_proto);
-
+		len = ieee80211_amsdu_subframe_length(&hdr.eth.h_proto, hdr.flags,
+						      mesh_control);
 		subframe_len = sizeof(struct ethhdr) + len;
 		padding = (4 - subframe_len) & 0x3;
 
-- 
cgit v1.2.3-70-g09d2


From 8e40c3b6e1538401d2cf8d43087ebe1db8026af9 Mon Sep 17 00:00:00 2001
From: Manikanta Pubbisetty <quic_mpubbise@quicinc.com>
Date: Wed, 8 Mar 2023 16:15:56 +0530
Subject: wifi: nl80211: Update the documentation of
 NL80211_SCAN_FLAG_COLOCATED_6GHZ

Currently when NL80211_SCAN_FLAG_COLOCATED_6GHZ is set in the scan flags,
in addition to the co-located APs, PSC channels in the 6 GHz band would
also be scanned if the user space has asked for it. In other words, the
scan would happen on PSC channels & co-located 6 GHz channels that were
reported in the RNR IE.

Update the documentation of NL80211_SCAN_FLAG_COLOCATED_6GHZ flag to
reflect the above said behavior.

Signed-off-by: Manikanta Pubbisetty <quic_mpubbise@quicinc.com>
Link: https://lore.kernel.org/r/20230308104556.9399-1-quic_mpubbise@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/uapi/linux/nl80211.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index 9a0ac0363f1f..14e958a32b84 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -6544,7 +6544,9 @@ enum nl80211_timeout_reason {
  *	channels on which APs are expected to be found. Note that when not set,
  *	the scan logic would scan all 6GHz channels, but since transmission of
  *	probe requests on non PSC channels is limited, it is highly likely that
- *	these channels would passively be scanned.
+ *	these channels would passively be scanned. Also note that when the flag
+ *	is set, in addition to the colocated APs, PSC channels would also be
+ *	scanned if the user space has asked for it.
  */
 enum nl80211_scan_flags {
 	NL80211_SCAN_FLAG_LOW_PRIORITY				= 1<<0,
-- 
cgit v1.2.3-70-g09d2


From bd54f3c29077f23dad92ef82a78061b40be30c65 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Mon, 5 Dec 2022 16:50:37 -0800
Subject: wifi: mac80211: generate EMA beacons in AP mode

Add APIs to generate an array of beacons for an EMA AP (enhanced
multiple BSSID advertisements), each including a single MBSSID element.
EMA profile periodicity equals the count of elements.

- ieee80211_beacon_get_template_ema_list() - Generate and return all
EMA beacon templates. Drivers must call ieee80211_beacon_free_ema_list()
to free the memory. No change in the prototype for the existing API,
ieee80211_beacon_get_template(), which should be used for non-EMA AP.

- ieee80211_beacon_get_template_ema_index() - Generate a beacon which
includes the multiple BSSID element at the given index. Drivers can use
this function in a loop until NULL is returned which indicates end of
available MBSSID elements.

- ieee80211_beacon_free_ema_list() - free the memory allocated for the
list of EMA beacon templates.

Modify existing functions ieee80211_beacon_get_ap(),
ieee80211_get_mbssid_beacon_len() and ieee80211_beacon_add_mbssid()
to accept a new parameter for EMA index.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Co-developed-by: John Crispin <john@phrozen.org>
Signed-off-by: John Crispin <john@phrozen.org>
Link: https://lore.kernel.org/r/20221206005040.3177-2-quic_alokad@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h     |  68 +++++++++++++++++++++++
 net/mac80211/cfg.c         |  11 ++--
 net/mac80211/ieee80211_i.h |  10 +++-
 net/mac80211/tx.c          | 134 +++++++++++++++++++++++++++++++++++++++++----
 4 files changed, 205 insertions(+), 18 deletions(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index fcfe3e9aff3d..679421d37a42 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -5285,6 +5285,74 @@ ieee80211_beacon_get_template(struct ieee80211_hw *hw,
 			      struct ieee80211_mutable_offsets *offs,
 			      unsigned int link_id);
 
+/**
+ * ieee80211_beacon_get_template_ema_index - EMA beacon template generation
+ * @hw: pointer obtained from ieee80211_alloc_hw().
+ * @vif: &struct ieee80211_vif pointer from the add_interface callback.
+ * @offs: &struct ieee80211_mutable_offsets pointer to struct that will
+ *	receive the offsets that may be updated by the driver.
+ * @link_id: the link id to which the beacon belongs (or 0 for a non-MLD AP).
+ * @ema_index: index of the beacon in the EMA set.
+ *
+ * This function follows the same rules as ieee80211_beacon_get_template()
+ * but returns a beacon template which includes multiple BSSID element at the
+ * requested index.
+ *
+ * Return: The beacon template. %NULL indicates the end of EMA templates.
+ */
+struct sk_buff *
+ieee80211_beacon_get_template_ema_index(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					struct ieee80211_mutable_offsets *offs,
+					unsigned int link_id, u8 ema_index);
+
+/**
+ * struct ieee80211_ema_beacons - List of EMA beacons
+ * @cnt: count of EMA beacons.
+ *
+ * @bcn: array of EMA beacons.
+ * @bcn.skb: the skb containing this specific beacon
+ * @bcn.offs: &struct ieee80211_mutable_offsets pointer to struct that will
+ *	receive the offsets that may be updated by the driver.
+ */
+struct ieee80211_ema_beacons {
+	u8 cnt;
+	struct {
+		struct sk_buff *skb;
+		struct ieee80211_mutable_offsets offs;
+	} bcn[];
+};
+
+/**
+ * ieee80211_beacon_get_template_ema_list - EMA beacon template generation
+ * @hw: pointer obtained from ieee80211_alloc_hw().
+ * @vif: &struct ieee80211_vif pointer from the add_interface callback.
+ * @link_id: the link id to which the beacon belongs (or 0 for a non-MLD AP)
+ *
+ * This function follows the same rules as ieee80211_beacon_get_template()
+ * but allocates and returns a pointer to list of all beacon templates required
+ * to cover all profiles in the multiple BSSID set. Each template includes only
+ * one multiple BSSID element.
+ *
+ * Driver must call ieee80211_beacon_free_ema_list() to free the memory.
+ *
+ * Return: EMA beacon templates of type struct ieee80211_ema_beacons *.
+ *	%NULL on error.
+ */
+struct ieee80211_ema_beacons *
+ieee80211_beacon_get_template_ema_list(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       unsigned int link_id);
+
+/**
+ * ieee80211_beacon_free_ema_list - free an EMA beacon template list
+ * @ema_beacons: list of EMA beacons of type &struct ieee80211_ema_beacons pointers.
+ *
+ * This function will free a list previously acquired by calling
+ * ieee80211_beacon_get_template_ema_list()
+ */
+void ieee80211_beacon_free_ema_list(struct ieee80211_ema_beacons *ema_beacons);
+
 /**
  * ieee80211_beacon_get_tim - beacon generation function
  * @hw: pointer obtained from ieee80211_alloc_hw().
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 760ad934f9e1..db5fa334b801 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1122,11 +1122,11 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata,
 	if (params->mbssid_ies) {
 		mbssid = params->mbssid_ies;
 		size += struct_size(new->mbssid_ies, elem, mbssid->cnt);
-		size += ieee80211_get_mbssid_beacon_len(mbssid);
+		size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt);
 	} else if (old && old->mbssid_ies) {
 		mbssid = old->mbssid_ies;
 		size += struct_size(new->mbssid_ies, elem, mbssid->cnt);
-		size += ieee80211_get_mbssid_beacon_len(mbssid);
+		size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt);
 	}
 
 	new = kzalloc(size, GFP_KERNEL);
@@ -3406,8 +3406,11 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon)
 
 	len = beacon->head_len + beacon->tail_len + beacon->beacon_ies_len +
 	      beacon->proberesp_ies_len + beacon->assocresp_ies_len +
-	      beacon->probe_resp_len + beacon->lci_len + beacon->civicloc_len +
-	      ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies);
+	      beacon->probe_resp_len + beacon->lci_len + beacon->civicloc_len;
+
+	if (beacon->mbssid_ies)
+		len += ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies,
+						       beacon->mbssid_ies->cnt);
 
 	new_beacon = kzalloc(sizeof(*new_beacon) + len, GFP_KERNEL);
 	if (!new_beacon)
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index c8c037a9e4e2..84d10e993eca 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1186,13 +1186,17 @@ ieee80211_vif_get_shift(struct ieee80211_vif *vif)
 }
 
 static inline int
-ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems)
+ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems, u8 i)
 {
-	int i, len = 0;
+	int len = 0;
 
-	if (!elems)
+	if (!elems || !elems->cnt || i > elems->cnt)
 		return 0;
 
+	if (i < elems->cnt)
+		return elems->elem[i].len;
+
+	/* i == elems->cnt, calculate total length of all MBSSID elements */
 	for (i = 0; i < elems->cnt; i++)
 		len += elems->elem[i].len;
 
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index de17926484bd..139eec6c64da 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -5212,13 +5212,20 @@ ieee80211_beacon_get_finish(struct ieee80211_hw *hw,
 }
 
 static void
-ieee80211_beacon_add_mbssid(struct sk_buff *skb, struct beacon_data *beacon)
+ieee80211_beacon_add_mbssid(struct sk_buff *skb, struct beacon_data *beacon,
+			    u8 i)
 {
-	int i;
+	if (!beacon->mbssid_ies || !beacon->mbssid_ies->cnt ||
+	    i > beacon->mbssid_ies->cnt)
+		return;
 
-	if (!beacon->mbssid_ies)
+	if (i < beacon->mbssid_ies->cnt) {
+		skb_put_data(skb, beacon->mbssid_ies->elem[i].data,
+			     beacon->mbssid_ies->elem[i].len);
 		return;
+	}
 
+	/* i == beacon->mbssid_ies->cnt, include all MBSSID elements */
 	for (i = 0; i < beacon->mbssid_ies->cnt; i++)
 		skb_put_data(skb, beacon->mbssid_ies->elem[i].data,
 			     beacon->mbssid_ies->elem[i].len);
@@ -5231,7 +5238,8 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw,
 			struct ieee80211_mutable_offsets *offs,
 			bool is_template,
 			struct beacon_data *beacon,
-			struct ieee80211_chanctx_conf *chanctx_conf)
+			struct ieee80211_chanctx_conf *chanctx_conf,
+			u8 ema_index)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
@@ -5250,7 +5258,9 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw,
 	/* headroom, head length,
 	 * tail length, maximum TIM length and multiple BSSID length
 	 */
-	mbssid_len = ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies);
+	mbssid_len = ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies,
+						     ema_index);
+
 	skb = dev_alloc_skb(local->tx_headroom + beacon->head_len +
 			    beacon->tail_len + 256 +
 			    local->hw.extra_beacon_tailroom + mbssid_len);
@@ -5268,7 +5278,7 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw,
 		offs->cntdwn_counter_offs[0] = beacon->cntdwn_counter_offsets[0];
 
 		if (mbssid_len) {
-			ieee80211_beacon_add_mbssid(skb, beacon);
+			ieee80211_beacon_add_mbssid(skb, beacon, ema_index);
 			offs->mbssid_off = skb->len - mbssid_len;
 		}
 
@@ -5287,12 +5297,51 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw,
 	return skb;
 }
 
+static struct ieee80211_ema_beacons *
+ieee80211_beacon_get_ap_ema_list(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_link_data *link,
+				 struct ieee80211_mutable_offsets *offs,
+				 bool is_template, struct beacon_data *beacon,
+				 struct ieee80211_chanctx_conf *chanctx_conf)
+{
+	struct ieee80211_ema_beacons *ema = NULL;
+
+	if (!beacon->mbssid_ies || !beacon->mbssid_ies->cnt)
+		return NULL;
+
+	ema = kzalloc(struct_size(ema, bcn, beacon->mbssid_ies->cnt),
+		      GFP_ATOMIC);
+	if (!ema)
+		return NULL;
+
+	for (ema->cnt = 0; ema->cnt < beacon->mbssid_ies->cnt; ema->cnt++) {
+		ema->bcn[ema->cnt].skb =
+			ieee80211_beacon_get_ap(hw, vif, link,
+						&ema->bcn[ema->cnt].offs,
+						is_template, beacon,
+						chanctx_conf, ema->cnt);
+		if (!ema->bcn[ema->cnt].skb)
+			break;
+	}
+
+	if (ema->cnt == beacon->mbssid_ies->cnt)
+		return ema;
+
+	ieee80211_beacon_free_ema_list(ema);
+	return NULL;
+}
+
+#define IEEE80211_INCLUDE_ALL_MBSSID_ELEMS -1
+
 static struct sk_buff *
 __ieee80211_beacon_get(struct ieee80211_hw *hw,
 		       struct ieee80211_vif *vif,
 		       struct ieee80211_mutable_offsets *offs,
 		       bool is_template,
-		       unsigned int link_id)
+		       unsigned int link_id,
+		       int ema_index,
+		       struct ieee80211_ema_beacons **ema_beacons)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct beacon_data *beacon = NULL;
@@ -5321,8 +5370,29 @@ __ieee80211_beacon_get(struct ieee80211_hw *hw,
 		if (!beacon)
 			goto out;
 
-		skb = ieee80211_beacon_get_ap(hw, vif, link, offs, is_template,
-					      beacon, chanctx_conf);
+		if (ema_beacons) {
+			*ema_beacons =
+				ieee80211_beacon_get_ap_ema_list(hw, vif, link,
+								 offs,
+								 is_template,
+								 beacon,
+								 chanctx_conf);
+		} else {
+			if (beacon->mbssid_ies && beacon->mbssid_ies->cnt) {
+				if (ema_index >= beacon->mbssid_ies->cnt)
+					goto out; /* End of MBSSID elements */
+
+				if (ema_index <= IEEE80211_INCLUDE_ALL_MBSSID_ELEMS)
+					ema_index = beacon->mbssid_ies->cnt;
+			} else {
+				ema_index = 0;
+			}
+
+			skb = ieee80211_beacon_get_ap(hw, vif, link, offs,
+						      is_template, beacon,
+						      chanctx_conf,
+						      ema_index);
+		}
 	} else if (sdata->vif.type == NL80211_IFTYPE_ADHOC) {
 		struct ieee80211_if_ibss *ifibss = &sdata->u.ibss;
 		struct ieee80211_hdr *hdr;
@@ -5410,10 +5480,50 @@ ieee80211_beacon_get_template(struct ieee80211_hw *hw,
 			      struct ieee80211_mutable_offsets *offs,
 			      unsigned int link_id)
 {
-	return __ieee80211_beacon_get(hw, vif, offs, true, link_id);
+	return __ieee80211_beacon_get(hw, vif, offs, true, link_id,
+				      IEEE80211_INCLUDE_ALL_MBSSID_ELEMS, NULL);
 }
 EXPORT_SYMBOL(ieee80211_beacon_get_template);
 
+struct sk_buff *
+ieee80211_beacon_get_template_ema_index(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					struct ieee80211_mutable_offsets *offs,
+					unsigned int link_id, u8 ema_index)
+{
+	return __ieee80211_beacon_get(hw, vif, offs, true, link_id, ema_index,
+				      NULL);
+}
+EXPORT_SYMBOL(ieee80211_beacon_get_template_ema_index);
+
+void ieee80211_beacon_free_ema_list(struct ieee80211_ema_beacons *ema_beacons)
+{
+	u8 i;
+
+	if (!ema_beacons)
+		return;
+
+	for (i = 0; i < ema_beacons->cnt; i++)
+		kfree_skb(ema_beacons->bcn[i].skb);
+
+	kfree(ema_beacons);
+}
+EXPORT_SYMBOL(ieee80211_beacon_free_ema_list);
+
+struct ieee80211_ema_beacons *
+ieee80211_beacon_get_template_ema_list(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       unsigned int link_id)
+{
+	struct ieee80211_ema_beacons *ema_beacons = NULL;
+
+	WARN_ON(__ieee80211_beacon_get(hw, vif, NULL, false, link_id, 0,
+				       &ema_beacons));
+
+	return ema_beacons;
+}
+EXPORT_SYMBOL(ieee80211_beacon_get_template_ema_list);
+
 struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw,
 					 struct ieee80211_vif *vif,
 					 u16 *tim_offset, u16 *tim_length,
@@ -5421,7 +5531,9 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw,
 {
 	struct ieee80211_mutable_offsets offs = {};
 	struct sk_buff *bcn = __ieee80211_beacon_get(hw, vif, &offs, false,
-						     link_id);
+						     link_id,
+						     IEEE80211_INCLUDE_ALL_MBSSID_ELEMS,
+						     NULL);
 	struct sk_buff *copy;
 	int shift;
 
-- 
cgit v1.2.3-70-g09d2


From b3a912e3e220257a25219cd81b9aa8a639cb1e18 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Mon, 5 Dec 2022 16:50:38 -0800
Subject: wifi: mac80211_hwsim: move beacon transmission to a separate function

Move the beacon transmission to a separate function to facilitate
addition of EMA beacon transmission.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Link: https://lore.kernel.org/r/20221206005040.3177-3-quic_alokad@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 61 ++++++++++++++++-----------
 1 file changed, 36 insertions(+), 25 deletions(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index f4bdc243ea0d..0b4f489c8036 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -2055,38 +2055,18 @@ static void mac80211_hwsim_tx_frame(struct ieee80211_hw *hw,
 	dev_kfree_skb(skb);
 }
 
-static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
-				     struct ieee80211_vif *vif)
+static void __mac80211_hwsim_beacon_tx(struct ieee80211_bss_conf *link_conf,
+				       struct mac80211_hwsim_data *data,
+				       struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       struct sk_buff *skb)
 {
-	struct mac80211_hwsim_link_data *link_data = arg;
-	u32 link_id = link_data->link_id;
-	struct ieee80211_bss_conf *link_conf;
-	struct mac80211_hwsim_data *data =
-		container_of(link_data, struct mac80211_hwsim_data,
-			     link_data[link_id]);
-	struct ieee80211_hw *hw = data->hw;
 	struct ieee80211_tx_info *info;
 	struct ieee80211_rate *txrate;
 	struct ieee80211_mgmt *mgmt;
-	struct sk_buff *skb;
 	/* TODO: get MCS */
 	int bitrate = 100;
 
-	hwsim_check_magic(vif);
-
-	link_conf = rcu_dereference(vif->link_conf[link_id]);
-	if (!link_conf)
-		return;
-
-	if (vif->type != NL80211_IFTYPE_AP &&
-	    vif->type != NL80211_IFTYPE_MESH_POINT &&
-	    vif->type != NL80211_IFTYPE_ADHOC &&
-	    vif->type != NL80211_IFTYPE_OCB)
-		return;
-
-	skb = ieee80211_beacon_get(hw, vif, link_data->link_id);
-	if (skb == NULL)
-		return;
 	info = IEEE80211_SKB_CB(skb);
 	if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE))
 		ieee80211_get_tx_rates(vif, NULL, skb,
@@ -2116,6 +2096,37 @@ static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
 
 	mac80211_hwsim_tx_frame(hw, skb,
 			rcu_dereference(link_conf->chanctx_conf)->def.chan);
+}
+
+static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
+				     struct ieee80211_vif *vif)
+{
+	struct mac80211_hwsim_link_data *link_data = arg;
+	u32 link_id = link_data->link_id;
+	struct ieee80211_bss_conf *link_conf;
+	struct mac80211_hwsim_data *data =
+		container_of(link_data, struct mac80211_hwsim_data,
+			     link_data[link_id]);
+	struct ieee80211_hw *hw = data->hw;
+	struct sk_buff *skb;
+
+	hwsim_check_magic(vif);
+
+	link_conf = rcu_dereference(vif->link_conf[link_id]);
+	if (!link_conf)
+		return;
+
+	if (vif->type != NL80211_IFTYPE_AP &&
+	    vif->type != NL80211_IFTYPE_MESH_POINT &&
+	    vif->type != NL80211_IFTYPE_ADHOC &&
+	    vif->type != NL80211_IFTYPE_OCB)
+		return;
+
+	skb = ieee80211_beacon_get(hw, vif, link_id);
+	if (!skb)
+		return;
+
+	__mac80211_hwsim_beacon_tx(link_conf, data, hw, vif, skb);
 
 	while ((skb = ieee80211_get_buffered_bc(hw, vif)) != NULL) {
 		mac80211_hwsim_tx_frame(hw, skb,
-- 
cgit v1.2.3-70-g09d2


From c4f4d9f7e7f0a7af82b086354c9277fc8446fe9c Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Mon, 5 Dec 2022 16:50:39 -0800
Subject: wifi: mac80211_hwsim: Multiple BSSID support

Advertise multiple BSSID support for up to 8 interfaces.
Do not send beacons from the non-transmitting interfaces.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Link: https://lore.kernel.org/r/20221206005040.3177-4-quic_alokad@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index 0b4f489c8036..58ceabbd1550 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -2122,6 +2122,9 @@ static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
 	    vif->type != NL80211_IFTYPE_OCB)
 		return;
 
+	if (vif->mbssid_tx_vif && vif->mbssid_tx_vif != vif)
+		return;
+
 	skb = ieee80211_beacon_get(hw, vif, link_id);
 	if (!skb)
 		return;
@@ -4405,6 +4408,8 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 		hw->wiphy->n_cipher_suites = param->n_ciphers;
 	}
 
+	hw->wiphy->mbssid_max_interfaces = 8;
+
 	data->rx_rssi = DEFAULT_RX_RSSI;
 
 	INIT_DELAYED_WORK(&data->roc_start, hw_roc_start);
-- 
cgit v1.2.3-70-g09d2


From 0dd45ebc08de2449efe1a0908147796856a5f824 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Mon, 5 Dec 2022 16:50:40 -0800
Subject: wifi: mac80211_hwsim: EMA support

Add support enhanced multi-BSS advertisements (EMA)
for profile periodicity up to 3 beacons.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Link: https://lore.kernel.org/r/20221206005040.3177-5-quic_alokad@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index 58ceabbd1550..2aa7b740b1ce 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -2125,11 +2125,27 @@ static void mac80211_hwsim_beacon_tx(void *arg, u8 *mac,
 	if (vif->mbssid_tx_vif && vif->mbssid_tx_vif != vif)
 		return;
 
-	skb = ieee80211_beacon_get(hw, vif, link_id);
-	if (!skb)
-		return;
+	if (vif->bss_conf.ema_ap) {
+		struct ieee80211_ema_beacons *ema;
+		u8 i = 0;
+
+		ema = ieee80211_beacon_get_template_ema_list(hw, vif, link_id);
+		if (!ema || !ema->cnt)
+			return;
+
+		for (i = 0; i < ema->cnt; i++) {
+			__mac80211_hwsim_beacon_tx(link_conf, data, hw, vif,
+						   ema->bcn[i].skb);
+			ema->bcn[i].skb = NULL; /* Already freed */
+		}
+		ieee80211_beacon_free_ema_list(ema);
+	} else {
+		skb = ieee80211_beacon_get(hw, vif, link_id);
+		if (!skb)
+			return;
 
-	__mac80211_hwsim_beacon_tx(link_conf, data, hw, vif, skb);
+		__mac80211_hwsim_beacon_tx(link_conf, data, hw, vif, skb);
+	}
 
 	while ((skb = ieee80211_get_buffered_bc(hw, vif)) != NULL) {
 		mac80211_hwsim_tx_frame(hw, skb,
@@ -4409,6 +4425,7 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 	}
 
 	hw->wiphy->mbssid_max_interfaces = 8;
+	hw->wiphy->ema_max_profile_periodicity = 3;
 
 	data->rx_rssi = DEFAULT_RX_RSSI;
 
-- 
cgit v1.2.3-70-g09d2


From 593ef60c74451ba05a87effb50edb5cd45312115 Mon Sep 17 00:00:00 2001
From: Xiaoyan Li <lixiaoyan@google.com>
Date: Tue, 21 Mar 2023 16:12:01 +0800
Subject: net-zerocopy: Reduce compound page head access

When compound pages are enabled, although the mm layer still
returns an array of page pointers, a subset (or all) of them
may have the same page head since a max 180kb skb can span 2
hugepages if it is on the boundary, be a mix of pages and 1 hugepage,
or fit completely in a hugepage. Instead of referencing page head
on all page pointers, use page length arithmetic to only call page
head when referencing a known different page head to avoid touching
a cold cacheline.

Tested:
See next patch with changes to tcp_mmap

Correntess:
On a pair of separate hosts as send with MSG_ZEROCOPY will
force a copy on tx if using loopback alone, check that the SHA
on the message sent is equivalent to checksum on the message received,
since the current program already checks for the length.

echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
./tcp_mmap -s -z
./tcp_mmap -H $DADDR -z

SHA256 is correct
received 2 MB (100 % mmap'ed) in 0.005914 s, 2.83686 Gbit
  cpu usage user:0.001984 sys:0.000963, 1473.5 usec per MB, 10 c-switches

Performance:
Run neper between adjacent hosts with the same config
tcp_stream -Z --skip-rx-copy -6 -T 20 -F 1000 --stime-use-proc --test-length=30

Before patch: stime_end=37.670000
After patch: stime_end=30.310000

Signed-off-by: Coco Li <lixiaoyan@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230321081202.2370275-1-lixiaoyan@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/datagram.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index e4ff2db40c98..5662dff3d381 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -622,12 +622,12 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
 	frag = skb_shinfo(skb)->nr_frags;
 
 	while (length && iov_iter_count(from)) {
+		struct page *head, *last_head = NULL;
 		struct page *pages[MAX_SKB_FRAGS];
-		struct page *last_head = NULL;
+		int refs, order, n = 0;
 		size_t start;
 		ssize_t copied;
 		unsigned long truesize;
-		int refs, n = 0;
 
 		if (frag == MAX_SKB_FRAGS)
 			return -EMSGSIZE;
@@ -650,9 +650,17 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
 		} else {
 			refcount_add(truesize, &skb->sk->sk_wmem_alloc);
 		}
+
+		head = compound_head(pages[n]);
+		order = compound_order(head);
+
 		for (refs = 0; copied != 0; start = 0) {
 			int size = min_t(int, copied, PAGE_SIZE - start);
-			struct page *head = compound_head(pages[n]);
+
+			if (pages[n] - head > (1UL << order) - 1) {
+				head = compound_head(pages[n]);
+				order = compound_order(head);
+			}
 
 			start += (pages[n] - head) << PAGE_SHIFT;
 			copied -= size;
-- 
cgit v1.2.3-70-g09d2


From 5c5945dc695c54f2b55a934a10b6c4e220f9c140 Mon Sep 17 00:00:00 2001
From: Xiaoyan Li <lixiaoyan@google.com>
Date: Tue, 21 Mar 2023 16:12:02 +0800
Subject: selftests/net: Add SHA256 computation over data sent in tcp_mmap

Add option to compute and send SHA256 over data sent (-i).

This is to ensure the correctness of data received.
Data is randomly populated from /dev/urandom.

Tested:
./tcp_mmap -s -z -i
./tcp_mmap -z -H $ADDR -i
SHA256 is correct

./tcp_mmap -s -i
./tcp_mmap -H $ADDR -i
SHA256 is correct

Signed-off-by: Coco Li <lixiaoyan@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230321081202.2370275-2-lixiaoyan@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 tools/testing/selftests/net/Makefile   |   2 +-
 tools/testing/selftests/net/tcp_mmap.c | 102 +++++++++++++++++++++++++++++----
 2 files changed, 92 insertions(+), 12 deletions(-)

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index e57750e44f71..1de34ec99290 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -89,7 +89,7 @@ TEST_FILES := settings
 include ../lib.mk
 
 $(OUTPUT)/reuseport_bpf_numa: LDLIBS += -lnuma
-$(OUTPUT)/tcp_mmap: LDLIBS += -lpthread
+$(OUTPUT)/tcp_mmap: LDLIBS += -lpthread -lcrypto
 $(OUTPUT)/tcp_inq: LDLIBS += -lpthread
 $(OUTPUT)/bind_bhash: LDLIBS += -lpthread
 
diff --git a/tools/testing/selftests/net/tcp_mmap.c b/tools/testing/selftests/net/tcp_mmap.c
index 46a02bbd31d0..607cc9ad8d1b 100644
--- a/tools/testing/selftests/net/tcp_mmap.c
+++ b/tools/testing/selftests/net/tcp_mmap.c
@@ -66,11 +66,16 @@
 #include <poll.h>
 #include <linux/tcp.h>
 #include <assert.h>
+#include <openssl/pem.h>
 
 #ifndef MSG_ZEROCOPY
 #define MSG_ZEROCOPY    0x4000000
 #endif
 
+#ifndef min
+#define min(a, b)  ((a) < (b) ? (a) : (b))
+#endif
+
 #define FILE_SZ (1ULL << 35)
 static int cfg_family = AF_INET6;
 static socklen_t cfg_alen = sizeof(struct sockaddr_in6);
@@ -81,12 +86,14 @@ static int sndbuf; /* Default: autotuning.  Can be set with -w <integer> option
 static int zflg; /* zero copy option. (MSG_ZEROCOPY for sender, mmap() for receiver */
 static int xflg; /* hash received data (simple xor) (-h option) */
 static int keepflag; /* -k option: receiver shall keep all received file in memory (no munmap() calls) */
+static int integrity; /* -i option: sender and receiver compute sha256 over the data.*/
 
 static size_t chunk_size  = 512*1024;
 
 static size_t map_align;
 
 unsigned long htotal;
+unsigned int digest_len;
 
 static inline void prefetch(const void *x)
 {
@@ -148,12 +155,14 @@ static void *mmap_large_buffer(size_t need, size_t *allocated)
 
 void *child_thread(void *arg)
 {
+	unsigned char digest[SHA256_DIGEST_LENGTH];
 	unsigned long total_mmap = 0, total = 0;
 	struct tcp_zerocopy_receive zc;
+	unsigned char *buffer = NULL;
 	unsigned long delta_usec;
+	EVP_MD_CTX *ctx = NULL;
 	int flags = MAP_SHARED;
 	struct timeval t0, t1;
-	char *buffer = NULL;
 	void *raddr = NULL;
 	void *addr = NULL;
 	double throughput;
@@ -180,6 +189,14 @@ void *child_thread(void *arg)
 			addr = ALIGN_PTR_UP(raddr, map_align);
 		}
 	}
+	if (integrity) {
+		ctx = EVP_MD_CTX_new();
+		if (!ctx) {
+			perror("cannot enable SHA computing");
+			goto error;
+		}
+		EVP_DigestInit_ex(ctx, EVP_sha256(), NULL);
+	}
 	while (1) {
 		struct pollfd pfd = { .fd = fd, .events = POLLIN, };
 		int sub;
@@ -191,7 +208,7 @@ void *child_thread(void *arg)
 
 			memset(&zc, 0, sizeof(zc));
 			zc.address = (__u64)((unsigned long)addr);
-			zc.length = chunk_size;
+			zc.length = min(chunk_size, FILE_SZ - lu);
 
 			res = getsockopt(fd, IPPROTO_TCP, TCP_ZEROCOPY_RECEIVE,
 					 &zc, &zc_len);
@@ -200,6 +217,8 @@ void *child_thread(void *arg)
 
 			if (zc.length) {
 				assert(zc.length <= chunk_size);
+				if (integrity)
+					EVP_DigestUpdate(ctx, addr, zc.length);
 				total_mmap += zc.length;
 				if (xflg)
 					hash_zone(addr, zc.length);
@@ -211,22 +230,30 @@ void *child_thread(void *arg)
 			}
 			if (zc.recv_skip_hint) {
 				assert(zc.recv_skip_hint <= chunk_size);
-				lu = read(fd, buffer, zc.recv_skip_hint);
+				lu = read(fd, buffer, min(zc.recv_skip_hint,
+							  FILE_SZ - total));
 				if (lu > 0) {
+					if (integrity)
+						EVP_DigestUpdate(ctx, buffer, lu);
 					if (xflg)
 						hash_zone(buffer, lu);
 					total += lu;
 				}
+				if (lu == 0)
+					goto end;
 			}
 			continue;
 		}
 		sub = 0;
 		while (sub < chunk_size) {
-			lu = read(fd, buffer + sub, chunk_size - sub);
+			lu = read(fd, buffer + sub, min(chunk_size - sub,
+							FILE_SZ - total));
 			if (lu == 0)
 				goto end;
 			if (lu < 0)
 				break;
+			if (integrity)
+				EVP_DigestUpdate(ctx, buffer + sub, lu);
 			if (xflg)
 				hash_zone(buffer + sub, lu);
 			total += lu;
@@ -237,6 +264,20 @@ end:
 	gettimeofday(&t1, NULL);
 	delta_usec = (t1.tv_sec - t0.tv_sec) * 1000000 + t1.tv_usec - t0.tv_usec;
 
+	if (integrity) {
+		fcntl(fd, F_SETFL, 0);
+		EVP_DigestFinal_ex(ctx, digest, &digest_len);
+		lu = read(fd, buffer, SHA256_DIGEST_LENGTH);
+		if (lu != SHA256_DIGEST_LENGTH)
+			perror("Error: Cannot read SHA256\n");
+
+		if (memcmp(digest, buffer,
+			   SHA256_DIGEST_LENGTH))
+			fprintf(stderr, "Error: SHA256 of the data is not right\n");
+		else
+			printf("\nSHA256 is correct\n");
+	}
+
 	throughput = 0;
 	if (delta_usec)
 		throughput = total * 8.0 / (double)delta_usec / 1000.0;
@@ -368,19 +409,38 @@ static unsigned long default_huge_page_size(void)
 	return hps;
 }
 
+static void randomize(void *target, size_t count)
+{
+	static int urandom = -1;
+	ssize_t got;
+
+	urandom = open("/dev/urandom", O_RDONLY);
+	if (urandom < 0) {
+		perror("open /dev/urandom");
+		exit(1);
+	}
+	got = read(urandom, target, count);
+	if (got != count) {
+		perror("read /dev/urandom");
+		exit(1);
+	}
+}
+
 int main(int argc, char *argv[])
 {
+	unsigned char digest[SHA256_DIGEST_LENGTH];
 	struct sockaddr_storage listenaddr, addr;
 	unsigned int max_pacing_rate = 0;
+	EVP_MD_CTX *ctx = NULL;
+	unsigned char *buffer;
 	uint64_t total = 0;
 	char *host = NULL;
 	int fd, c, on = 1;
 	size_t buffer_sz;
-	char *buffer;
 	int sflg = 0;
 	int mss = 0;
 
-	while ((c = getopt(argc, argv, "46p:svr:w:H:zxkP:M:C:a:")) != -1) {
+	while ((c = getopt(argc, argv, "46p:svr:w:H:zxkP:M:C:a:i")) != -1) {
 		switch (c) {
 		case '4':
 			cfg_family = PF_INET;
@@ -426,6 +486,9 @@ int main(int argc, char *argv[])
 		case 'a':
 			map_align = atol(optarg);
 			break;
+		case 'i':
+			integrity = 1;
+			break;
 		default:
 			exit(1);
 		}
@@ -468,7 +531,7 @@ int main(int argc, char *argv[])
 	}
 
 	buffer = mmap_large_buffer(chunk_size, &buffer_sz);
-	if (buffer == (char *)-1) {
+	if (buffer == (unsigned char *)-1) {
 		perror("mmap");
 		exit(1);
 	}
@@ -501,17 +564,34 @@ int main(int argc, char *argv[])
 		perror("setsockopt SO_ZEROCOPY, (-z option disabled)");
 		zflg = 0;
 	}
+	if (integrity) {
+		randomize(buffer, buffer_sz);
+		ctx = EVP_MD_CTX_new();
+		if (!ctx) {
+			perror("cannot enable SHA computing");
+			exit(1);
+		}
+		EVP_DigestInit_ex(ctx, EVP_sha256(), NULL);
+	}
 	while (total < FILE_SZ) {
+		size_t offset = total % chunk_size;
 		int64_t wr = FILE_SZ - total;
 
-		if (wr > chunk_size)
-			wr = chunk_size;
-		/* Note : we just want to fill the pipe with 0 bytes */
-		wr = send(fd, buffer, (size_t)wr, zflg ? MSG_ZEROCOPY : 0);
+		if (wr > chunk_size - offset)
+			wr = chunk_size - offset;
+		/* Note : we just want to fill the pipe with random bytes */
+		wr = send(fd, buffer + offset,
+			  (size_t)wr, zflg ? MSG_ZEROCOPY : 0);
 		if (wr <= 0)
 			break;
+		if (integrity)
+			EVP_DigestUpdate(ctx, buffer + offset, wr);
 		total += wr;
 	}
+	if (integrity && total == FILE_SZ) {
+		EVP_DigestFinal_ex(ctx, digest, &digest_len);
+		send(fd, digest, (size_t)SHA256_DIGEST_LENGTH, 0);
+	}
 	close(fd);
 	munmap(buffer, buffer_sz);
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 9a321fd3308e262f2a76761bea86dd0f311e3f86 Mon Sep 17 00:00:00 2001
From: Tushar Vyavahare <tushar.vyavahare@intel.com>
Date: Mon, 20 Mar 2023 15:57:05 +0530
Subject: selftests/xsk: add xdp populate metadata test

Add a new test in copy-mode for testing the copying of metadata from the
buffer in kernel-space to user-space. This is accomplished by adding a
new XDP program and using the bss map to store a counter that is written
to the metadata field. This counter is incremented for every packet so
that the number becomes unique and should be the same as the payload. It
is store in the bss so the value can be reset between runs.

The XDP program populates the metadata and the userspace program checks
the value stored in the metadata field against the payload using the new
is_metadata_correct() function. To turn this verification on or off, add
a new parameter (use_metadata) to the ifobject structure.

Signed-off-by: Tushar Vyavahare <tushar.vyavahare@intel.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20230320102705.306187-1-tushar.vyavahare@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/xsk_xdp_progs.c | 25 ++++++++++++
 tools/testing/selftests/bpf/xsk_xdp_metadata.h    |  5 +++
 tools/testing/selftests/bpf/xskxceiver.c          | 46 ++++++++++++++++++++++-
 tools/testing/selftests/bpf/xskxceiver.h          |  2 +
 4 files changed, 77 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/bpf/xsk_xdp_metadata.h

diff --git a/tools/testing/selftests/bpf/progs/xsk_xdp_progs.c b/tools/testing/selftests/bpf/progs/xsk_xdp_progs.c
index 744a01d0e57d..a630c95c7471 100644
--- a/tools/testing/selftests/bpf/progs/xsk_xdp_progs.c
+++ b/tools/testing/selftests/bpf/progs/xsk_xdp_progs.c
@@ -3,6 +3,7 @@
 
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
+#include "xsk_xdp_metadata.h"
 
 struct {
 	__uint(type, BPF_MAP_TYPE_XSKMAP);
@@ -12,6 +13,7 @@ struct {
 } xsk SEC(".maps");
 
 static unsigned int idx;
+int count = 0;
 
 SEC("xdp") int xsk_def_prog(struct xdp_md *xdp)
 {
@@ -27,4 +29,27 @@ SEC("xdp") int xsk_xdp_drop(struct xdp_md *xdp)
 	return bpf_redirect_map(&xsk, 0, XDP_DROP);
 }
 
+SEC("xdp") int xsk_xdp_populate_metadata(struct xdp_md *xdp)
+{
+	void *data, *data_meta;
+	struct xdp_info *meta;
+	int err;
+
+	/* Reserve enough for all custom metadata. */
+	err = bpf_xdp_adjust_meta(xdp, -(int)sizeof(struct xdp_info));
+	if (err)
+		return XDP_DROP;
+
+	data = (void *)(long)xdp->data;
+	data_meta = (void *)(long)xdp->data_meta;
+
+	if (data_meta + sizeof(struct xdp_info) > data)
+		return XDP_DROP;
+
+	meta = data_meta;
+	meta->count = count++;
+
+	return bpf_redirect_map(&xsk, 0, XDP_DROP);
+}
+
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/xsk_xdp_metadata.h b/tools/testing/selftests/bpf/xsk_xdp_metadata.h
new file mode 100644
index 000000000000..943133da378a
--- /dev/null
+++ b/tools/testing/selftests/bpf/xsk_xdp_metadata.h
@@ -0,0 +1,5 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+struct xdp_info {
+	__u64 count;
+} __attribute__((aligned(32)));
diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index a17655107a94..b65e0645b0cd 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -103,6 +103,7 @@
 #include <bpf/bpf.h>
 #include <linux/filter.h>
 #include "../kselftest.h"
+#include "xsk_xdp_metadata.h"
 
 static const char *MAC1 = "\x00\x0A\x56\x9E\xEE\x62";
 static const char *MAC2 = "\x00\x0A\x56\x9E\xEE\x61";
@@ -464,6 +465,7 @@ static void __test_spec_init(struct test_spec *test, struct ifobject *ifobj_tx,
 		ifobj->use_fill_ring = true;
 		ifobj->release_rx = true;
 		ifobj->validation_func = NULL;
+		ifobj->use_metadata = false;
 
 		if (i == 0) {
 			ifobj->rx_on = false;
@@ -798,6 +800,20 @@ static bool is_offset_correct(struct xsk_umem_info *umem, struct pkt_stream *pkt
 	return false;
 }
 
+static bool is_metadata_correct(struct pkt *pkt, void *buffer, u64 addr)
+{
+	void *data = xsk_umem__get_data(buffer, addr);
+	struct xdp_info *meta = data - sizeof(struct xdp_info);
+
+	if (meta->count != pkt->payload) {
+		ksft_print_msg("[%s] expected meta_count [%d], got meta_count [%d]\n",
+			       __func__, pkt->payload, meta->count);
+		return false;
+	}
+
+	return true;
+}
+
 static bool is_pkt_valid(struct pkt *pkt, void *buffer, u64 addr, u32 len)
 {
 	void *data = xsk_umem__get_data(buffer, addr);
@@ -959,7 +975,8 @@ static int receive_pkts(struct test_spec *test, struct pollfd *fds)
 			addr = xsk_umem__add_offset_to_addr(addr);
 
 			if (!is_pkt_valid(pkt, umem->buffer, addr, desc->len) ||
-			    !is_offset_correct(umem, pkt_stream, addr, pkt->addr))
+			    !is_offset_correct(umem, pkt_stream, addr, pkt->addr) ||
+			    (ifobj->use_metadata && !is_metadata_correct(pkt, umem->buffer, addr)))
 				return TEST_FAILURE;
 
 			if (ifobj->use_fill_ring)
@@ -1686,6 +1703,30 @@ static void testapp_xdp_drop(struct test_spec *test)
 	testapp_validate_traffic(test);
 }
 
+static void testapp_xdp_metadata_count(struct test_spec *test)
+{
+	struct xsk_xdp_progs *skel_rx = test->ifobj_rx->xdp_progs;
+	struct xsk_xdp_progs *skel_tx = test->ifobj_tx->xdp_progs;
+	struct bpf_map *data_map;
+	int count = 0;
+	int key = 0;
+
+	test_spec_set_name(test, "XDP_METADATA_COUNT");
+	test_spec_set_xdp_prog(test, skel_rx->progs.xsk_xdp_populate_metadata,
+			       skel_tx->progs.xsk_xdp_populate_metadata,
+			       skel_rx->maps.xsk, skel_tx->maps.xsk);
+	test->ifobj_rx->use_metadata = true;
+
+	data_map = bpf_object__find_map_by_name(skel_rx->obj, "xsk_xdp_.bss");
+	if (!data_map || !bpf_map__is_internal(data_map))
+		exit_with_error(ENOMEM);
+
+	if (bpf_map_update_elem(bpf_map__fd(data_map), &key, &count, BPF_ANY))
+		exit_with_error(errno);
+
+	testapp_validate_traffic(test);
+}
+
 static void testapp_poll_txq_tmout(struct test_spec *test)
 {
 	test_spec_set_name(test, "POLL_TXQ_FULL");
@@ -1835,6 +1876,9 @@ static void run_pkt_test(struct test_spec *test, enum test_mode mode, enum test_
 	case TEST_TYPE_XDP_DROP_HALF:
 		testapp_xdp_drop(test);
 		break;
+	case TEST_TYPE_XDP_METADATA_COUNT:
+		testapp_xdp_metadata_count(test);
+		break;
 	default:
 		break;
 	}
diff --git a/tools/testing/selftests/bpf/xskxceiver.h b/tools/testing/selftests/bpf/xskxceiver.h
index 3e8ec7d8ec32..bdb4efedf3a9 100644
--- a/tools/testing/selftests/bpf/xskxceiver.h
+++ b/tools/testing/selftests/bpf/xskxceiver.h
@@ -88,6 +88,7 @@ enum test_type {
 	TEST_TYPE_STATS_FILL_EMPTY,
 	TEST_TYPE_BPF_RES,
 	TEST_TYPE_XDP_DROP_HALF,
+	TEST_TYPE_XDP_METADATA_COUNT,
 	TEST_TYPE_MAX
 };
 
@@ -158,6 +159,7 @@ struct ifobject {
 	bool use_fill_ring;
 	bool release_rx;
 	bool shared_umem;
+	bool use_metadata;
 	u8 dst_mac[ETH_ALEN];
 	u8 src_mac[ETH_ALEN];
 };
-- 
cgit v1.2.3-70-g09d2


From a18f721415b4d4edea23cb8c7d6276330302af14 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 21 Mar 2023 13:38:51 -0700
Subject: libbpf: Rename RELO_EXTERN_VAR/FUNC.

RELO_EXTERN_VAR/FUNC names are not correct anymore. RELO_EXTERN_VAR represent
ksym symbol in ld_imm64 insn. It can point to kernel variable or kfunc.
Rename RELO_EXTERN_VAR->RELO_EXTERN_LD64 and RELO_EXTERN_FUNC->RELO_EXTERN_CALL
to match what they actually represent.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230321203854.3035-2-alexei.starovoitov@gmail.com
---
 tools/lib/bpf/libbpf.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 149864ea88d1..f8131f963803 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -315,8 +315,8 @@ enum reloc_type {
 	RELO_LD64,
 	RELO_CALL,
 	RELO_DATA,
-	RELO_EXTERN_VAR,
-	RELO_EXTERN_FUNC,
+	RELO_EXTERN_LD64,
+	RELO_EXTERN_CALL,
 	RELO_SUBPROG_ADDR,
 	RELO_CORE,
 };
@@ -4009,9 +4009,9 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 		pr_debug("prog '%s': found extern #%d '%s' (sym %d) for insn #%u\n",
 			 prog->name, i, ext->name, ext->sym_idx, insn_idx);
 		if (insn->code == (BPF_JMP | BPF_CALL))
-			reloc_desc->type = RELO_EXTERN_FUNC;
+			reloc_desc->type = RELO_EXTERN_CALL;
 		else
-			reloc_desc->type = RELO_EXTERN_VAR;
+			reloc_desc->type = RELO_EXTERN_LD64;
 		reloc_desc->insn_idx = insn_idx;
 		reloc_desc->sym_off = i; /* sym_off stores extern index */
 		return 0;
@@ -5855,7 +5855,7 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 						   relo->map_idx, map);
 			}
 			break;
-		case RELO_EXTERN_VAR:
+		case RELO_EXTERN_LD64:
 			ext = &obj->externs[relo->sym_off];
 			if (ext->type == EXT_KCFG) {
 				if (obj->gen_loader) {
@@ -5877,7 +5877,7 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 				}
 			}
 			break;
-		case RELO_EXTERN_FUNC:
+		case RELO_EXTERN_CALL:
 			ext = &obj->externs[relo->sym_off];
 			insn[0].src_reg = BPF_PSEUDO_KFUNC_CALL;
 			if (ext->is_set) {
@@ -6115,7 +6115,7 @@ bpf_object__reloc_code(struct bpf_object *obj, struct bpf_program *main_prog,
 			continue;
 
 		relo = find_prog_insn_relo(prog, insn_idx);
-		if (relo && relo->type == RELO_EXTERN_FUNC)
+		if (relo && relo->type == RELO_EXTERN_CALL)
 			/* kfunc relocations will be handled later
 			 * in bpf_object__relocate_data()
 			 */
@@ -7072,14 +7072,14 @@ static int bpf_program_record_relos(struct bpf_program *prog)
 		struct extern_desc *ext = &obj->externs[relo->sym_off];
 
 		switch (relo->type) {
-		case RELO_EXTERN_VAR:
+		case RELO_EXTERN_LD64:
 			if (ext->type != EXT_KSYM)
 				continue;
 			bpf_gen__record_extern(obj->gen_loader, ext->name,
 					       ext->is_weak, !ext->ksym.type_id,
 					       BTF_KIND_VAR, relo->insn_idx);
 			break;
-		case RELO_EXTERN_FUNC:
+		case RELO_EXTERN_CALL:
 			bpf_gen__record_extern(obj->gen_loader, ext->name,
 					       ext->is_weak, false, BTF_KIND_FUNC,
 					       relo->insn_idx);
-- 
cgit v1.2.3-70-g09d2


From 1057d299459657b85e593a4b6294a000f920672a Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 21 Mar 2023 13:38:52 -0700
Subject: bpf: Teach the verifier to recognize rdonly_mem as not null.

Teach the verifier to recognize PTR_TO_MEM | MEM_RDONLY as not NULL
otherwise if (!bpf_ksym_exists(known_kfunc)) doesn't go through
dead code elimination.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230321203854.3035-3-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 8bc44f5dc5b6..5693e4a92752 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -486,8 +486,17 @@ static bool type_is_sk_pointer(enum bpf_reg_type type)
 		type == PTR_TO_XDP_SOCK;
 }
 
+static bool type_may_be_null(u32 type)
+{
+	return type & PTR_MAYBE_NULL;
+}
+
 static bool reg_type_not_null(enum bpf_reg_type type)
 {
+	if (type_may_be_null(type))
+		return false;
+
+	type = base_type(type);
 	return type == PTR_TO_SOCKET ||
 		type == PTR_TO_TCP_SOCK ||
 		type == PTR_TO_MAP_VALUE ||
@@ -531,11 +540,6 @@ static bool type_is_rdonly_mem(u32 type)
 	return type & MEM_RDONLY;
 }
 
-static bool type_may_be_null(u32 type)
-{
-	return type & PTR_MAYBE_NULL;
-}
-
 static bool is_acquire_function(enum bpf_func_id func_id,
 				const struct bpf_map *map)
 {
-- 
cgit v1.2.3-70-g09d2


From 708cdc5706a4701be9e5f81cb2e2b60b57f34c42 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 21 Mar 2023 13:38:53 -0700
Subject: libbpf: Support kfunc detection in light skeleton.

Teach gen_loader to find {btf_id, btf_obj_fd} of kernel variables and kfuncs
and populate corresponding ld_imm64 and bpf_call insns.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230321203854.3035-4-alexei.starovoitov@gmail.com
---
 tools/lib/bpf/bpf_gen_internal.h |  4 +++-
 tools/lib/bpf/gen_loader.c       | 38 +++++++++++++++++++-------------------
 tools/lib/bpf/libbpf.c           |  7 +++++--
 3 files changed, 27 insertions(+), 22 deletions(-)

diff --git a/tools/lib/bpf/bpf_gen_internal.h b/tools/lib/bpf/bpf_gen_internal.h
index 223308931d55..fdf44403ff36 100644
--- a/tools/lib/bpf/bpf_gen_internal.h
+++ b/tools/lib/bpf/bpf_gen_internal.h
@@ -11,6 +11,7 @@ struct ksym_relo_desc {
 	int insn_idx;
 	bool is_weak;
 	bool is_typeless;
+	bool is_ld64;
 };
 
 struct ksym_desc {
@@ -24,6 +25,7 @@ struct ksym_desc {
 		bool typeless;
 	};
 	int insn;
+	bool is_ld64;
 };
 
 struct bpf_gen {
@@ -65,7 +67,7 @@ void bpf_gen__map_update_elem(struct bpf_gen *gen, int map_idx, void *value, __u
 void bpf_gen__map_freeze(struct bpf_gen *gen, int map_idx);
 void bpf_gen__record_attach_target(struct bpf_gen *gen, const char *name, enum bpf_attach_type type);
 void bpf_gen__record_extern(struct bpf_gen *gen, const char *name, bool is_weak,
-			    bool is_typeless, int kind, int insn_idx);
+			    bool is_typeless, bool is_ld64, int kind, int insn_idx);
 void bpf_gen__record_relo_core(struct bpf_gen *gen, const struct bpf_core_relo *core_relo);
 void bpf_gen__populate_outer_map(struct bpf_gen *gen, int outer_map_idx, int key, int inner_map_idx);
 
diff --git a/tools/lib/bpf/gen_loader.c b/tools/lib/bpf/gen_loader.c
index b74c82bb831e..83e8e3bfd8ff 100644
--- a/tools/lib/bpf/gen_loader.c
+++ b/tools/lib/bpf/gen_loader.c
@@ -560,7 +560,7 @@ static void emit_find_attach_target(struct bpf_gen *gen)
 }
 
 void bpf_gen__record_extern(struct bpf_gen *gen, const char *name, bool is_weak,
-			    bool is_typeless, int kind, int insn_idx)
+			    bool is_typeless, bool is_ld64, int kind, int insn_idx)
 {
 	struct ksym_relo_desc *relo;
 
@@ -574,6 +574,7 @@ void bpf_gen__record_extern(struct bpf_gen *gen, const char *name, bool is_weak,
 	relo->name = name;
 	relo->is_weak = is_weak;
 	relo->is_typeless = is_typeless;
+	relo->is_ld64 = is_ld64;
 	relo->kind = kind;
 	relo->insn_idx = insn_idx;
 	gen->relo_cnt++;
@@ -586,9 +587,11 @@ static struct ksym_desc *get_ksym_desc(struct bpf_gen *gen, struct ksym_relo_des
 	int i;
 
 	for (i = 0; i < gen->nr_ksyms; i++) {
-		if (!strcmp(gen->ksyms[i].name, relo->name)) {
-			gen->ksyms[i].ref++;
-			return &gen->ksyms[i];
+		kdesc = &gen->ksyms[i];
+		if (kdesc->kind == relo->kind && kdesc->is_ld64 == relo->is_ld64 &&
+		    !strcmp(kdesc->name, relo->name)) {
+			kdesc->ref++;
+			return kdesc;
 		}
 	}
 	kdesc = libbpf_reallocarray(gen->ksyms, gen->nr_ksyms + 1, sizeof(*kdesc));
@@ -603,6 +606,7 @@ static struct ksym_desc *get_ksym_desc(struct bpf_gen *gen, struct ksym_relo_des
 	kdesc->ref = 1;
 	kdesc->off = 0;
 	kdesc->insn = 0;
+	kdesc->is_ld64 = relo->is_ld64;
 	return kdesc;
 }
 
@@ -864,23 +868,17 @@ static void emit_relo(struct bpf_gen *gen, struct ksym_relo_desc *relo, int insn
 {
 	int insn;
 
-	pr_debug("gen: emit_relo (%d): %s at %d\n", relo->kind, relo->name, relo->insn_idx);
+	pr_debug("gen: emit_relo (%d): %s at %d %s\n",
+		 relo->kind, relo->name, relo->insn_idx, relo->is_ld64 ? "ld64" : "call");
 	insn = insns + sizeof(struct bpf_insn) * relo->insn_idx;
 	emit2(gen, BPF_LD_IMM64_RAW_FULL(BPF_REG_8, BPF_PSEUDO_MAP_IDX_VALUE, 0, 0, 0, insn));
-	switch (relo->kind) {
-	case BTF_KIND_VAR:
+	if (relo->is_ld64) {
 		if (relo->is_typeless)
 			emit_relo_ksym_typeless(gen, relo, insn);
 		else
 			emit_relo_ksym_btf(gen, relo, insn);
-		break;
-	case BTF_KIND_FUNC:
+	} else {
 		emit_relo_kfunc_btf(gen, relo, insn);
-		break;
-	default:
-		pr_warn("Unknown relocation kind '%d'\n", relo->kind);
-		gen->error = -EDOM;
-		return;
 	}
 }
 
@@ -903,18 +901,20 @@ static void cleanup_core_relo(struct bpf_gen *gen)
 
 static void cleanup_relos(struct bpf_gen *gen, int insns)
 {
+	struct ksym_desc *kdesc;
 	int i, insn;
 
 	for (i = 0; i < gen->nr_ksyms; i++) {
+		kdesc = &gen->ksyms[i];
 		/* only close fds for typed ksyms and kfuncs */
-		if (gen->ksyms[i].kind == BTF_KIND_VAR && !gen->ksyms[i].typeless) {
+		if (kdesc->is_ld64 && !kdesc->typeless) {
 			/* close fd recorded in insn[insn_idx + 1].imm */
-			insn = gen->ksyms[i].insn;
+			insn = kdesc->insn;
 			insn += sizeof(struct bpf_insn) + offsetof(struct bpf_insn, imm);
 			emit_sys_close_blob(gen, insn);
-		} else if (gen->ksyms[i].kind == BTF_KIND_FUNC) {
-			emit_sys_close_blob(gen, blob_fd_array_off(gen, gen->ksyms[i].off));
-			if (gen->ksyms[i].off < MAX_FD_ARRAY_SZ)
+		} else if (!kdesc->is_ld64) {
+			emit_sys_close_blob(gen, blob_fd_array_off(gen, kdesc->off));
+			if (kdesc->off < MAX_FD_ARRAY_SZ)
 				gen->nr_fd_array--;
 		}
 	}
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index f8131f963803..5d32aa8ea38a 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -7070,18 +7070,21 @@ static int bpf_program_record_relos(struct bpf_program *prog)
 	for (i = 0; i < prog->nr_reloc; i++) {
 		struct reloc_desc *relo = &prog->reloc_desc[i];
 		struct extern_desc *ext = &obj->externs[relo->sym_off];
+		int kind;
 
 		switch (relo->type) {
 		case RELO_EXTERN_LD64:
 			if (ext->type != EXT_KSYM)
 				continue;
+			kind = btf_is_var(btf__type_by_id(obj->btf, ext->btf_id)) ?
+				BTF_KIND_VAR : BTF_KIND_FUNC;
 			bpf_gen__record_extern(obj->gen_loader, ext->name,
 					       ext->is_weak, !ext->ksym.type_id,
-					       BTF_KIND_VAR, relo->insn_idx);
+					       true, kind, relo->insn_idx);
 			break;
 		case RELO_EXTERN_CALL:
 			bpf_gen__record_extern(obj->gen_loader, ext->name,
-					       ext->is_weak, false, BTF_KIND_FUNC,
+					       ext->is_weak, false, false, BTF_KIND_FUNC,
 					       relo->insn_idx);
 			break;
 		case RELO_CORE: {
-- 
cgit v1.2.3-70-g09d2


From 3b2ec2140fa27febb21034943d656898b659dc02 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 21 Mar 2023 13:38:54 -0700
Subject: selftests/bpf: Add light skeleton test for kfunc detection.

Add light skeleton test for kfunc detection and denylist it for s390.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230321203854.3035-5-alexei.starovoitov@gmail.com
---
 tools/testing/selftests/bpf/DENYLIST.s390x          |  1 +
 tools/testing/selftests/bpf/progs/test_ksyms_weak.c | 15 +++++++++++++++
 2 files changed, 16 insertions(+)

diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
index 34cb8b2de8ca..c7463f3ec3c0 100644
--- a/tools/testing/selftests/bpf/DENYLIST.s390x
+++ b/tools/testing/selftests/bpf/DENYLIST.s390x
@@ -11,6 +11,7 @@ get_stack_raw_tp                         # user_stack corrupted user stack
 iters/testmod_seq*                       # s390x doesn't support kfuncs in modules yet
 kprobe_multi_bench_attach                # bpf_program__attach_kprobe_multi_opts unexpected error: -95
 kprobe_multi_test                        # relies on fentry
+ksyms_btf/weak_ksyms*                    # test_ksyms_weak__open_and_load unexpected error: -22                        (kfunc)
 ksyms_module                             # test_ksyms_module__open_and_load unexpected error: -9                       (?)
 ksyms_module_libbpf                      # JIT does not support calling kernel function                                (kfunc)
 ksyms_module_lskel                       # test_ksyms_module_lskel__open_and_load unexpected error: -9                 (?)
diff --git a/tools/testing/selftests/bpf/progs/test_ksyms_weak.c b/tools/testing/selftests/bpf/progs/test_ksyms_weak.c
index 7003eef0c192..d00268c91e19 100644
--- a/tools/testing/selftests/bpf/progs/test_ksyms_weak.c
+++ b/tools/testing/selftests/bpf/progs/test_ksyms_weak.c
@@ -20,6 +20,8 @@ __u64 out__non_existent_typed = -1;
 /* test existing weak symbols can be resolved. */
 extern const struct rq runqueues __ksym __weak; /* typed */
 extern const void bpf_prog_active __ksym __weak; /* typeless */
+struct task_struct *bpf_task_acquire(struct task_struct *p) __ksym __weak;
+void bpf_testmod_test_mod_kfunc(int i) __ksym __weak;
 
 
 /* non-existent weak symbols. */
@@ -29,6 +31,7 @@ extern const void bpf_link_fops1 __ksym __weak;
 
 /* typed symbols, default to zero. */
 extern const int bpf_link_fops2 __ksym __weak;
+void invalid_kfunc(void) __ksym __weak;
 
 SEC("raw_tp/sys_enter")
 int pass_handler(const void *ctx)
@@ -50,6 +53,18 @@ int pass_handler(const void *ctx)
 	if (&bpf_link_fops2) /* can't happen */
 		out__non_existent_typed = (__u64)bpf_per_cpu_ptr(&bpf_link_fops2, 0);
 
+	if (!bpf_ksym_exists(bpf_task_acquire))
+		/* dead code won't be seen by the verifier */
+		bpf_task_acquire(0);
+
+	if (!bpf_ksym_exists(bpf_testmod_test_mod_kfunc))
+		/* dead code won't be seen by the verifier */
+		bpf_testmod_test_mod_kfunc(0);
+
+	if (bpf_ksym_exists(invalid_kfunc))
+		/* dead code won't be seen by the verifier */
+		invalid_kfunc();
+
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 6f56ad1b92328997e1b1792047099df6f8d7acb5 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Mar 2023 21:48:01 +0000
Subject: netfilter: nft_redir: use `struct nf_nat_range2` throughout and
 deduplicate eval call-backs

`nf_nat_redirect_ipv4` takes a `struct nf_nat_ipv4_multi_range_compat`,
but converts it internally to a `struct nf_nat_range2`.  Change the
function to take the latter, factor out the code now shared with
`nf_nat_redirect_ipv6`, move the conversion to the xt_REDIRECT module,
and update the ipv4 range initialization in the nft_redir module.

Replace a bare hex constant for 127.0.0.1 with a macro.

Remove `WARN_ON`.  `nf_nat_setup_info` calls `nf_ct_is_confirmed`:

	/* Can't setup nat info for confirmed ct. */
	if (nf_ct_is_confirmed(ct))
		return NF_ACCEPT;

This means that `ct` cannot be null or the kernel will crash, and
implies that `ctinfo` is `IP_CT_NEW` or `IP_CT_RELATED`.

nft_redir has separate ipv4 and ipv6 call-backs which share much of
their code, and an inet one switch containing a switch that calls one of
the others based on the family of the packet.  Merge the ipv4 and ipv6
ones into the inet one in order to get rid of the duplicate code.

Const-qualify the `priv` pointer since we don't need to write through
it.

Assign `priv->flags` to the range instead of OR-ing it in.

Set the `NF_NAT_RANGE_PROTO_SPECIFIED` flag once during init, rather
than on every eval.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/net/netfilter/nf_nat_redirect.h |  3 +-
 net/netfilter/nf_nat_redirect.c         | 71 +++++++++++++---------------
 net/netfilter/nft_redir.c               | 84 ++++++++++++---------------------
 net/netfilter/xt_REDIRECT.c             | 10 +++-
 4 files changed, 72 insertions(+), 96 deletions(-)

diff --git a/include/net/netfilter/nf_nat_redirect.h b/include/net/netfilter/nf_nat_redirect.h
index 2418653a66db..279380de904c 100644
--- a/include/net/netfilter/nf_nat_redirect.h
+++ b/include/net/netfilter/nf_nat_redirect.h
@@ -6,8 +6,7 @@
 #include <uapi/linux/netfilter/nf_nat.h>
 
 unsigned int
-nf_nat_redirect_ipv4(struct sk_buff *skb,
-		     const struct nf_nat_ipv4_multi_range_compat *mr,
+nf_nat_redirect_ipv4(struct sk_buff *skb, const struct nf_nat_range2 *range,
 		     unsigned int hooknum);
 unsigned int
 nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
diff --git a/net/netfilter/nf_nat_redirect.c b/net/netfilter/nf_nat_redirect.c
index f91579c821e9..6616ba5d0b04 100644
--- a/net/netfilter/nf_nat_redirect.c
+++ b/net/netfilter/nf_nat_redirect.c
@@ -10,6 +10,7 @@
 
 #include <linux/if.h>
 #include <linux/inetdevice.h>
+#include <linux/in.h>
 #include <linux/ip.h>
 #include <linux/kernel.h>
 #include <linux/netdevice.h>
@@ -24,54 +25,56 @@
 #include <net/netfilter/nf_nat.h>
 #include <net/netfilter/nf_nat_redirect.h>
 
+static unsigned int
+nf_nat_redirect(struct sk_buff *skb, const struct nf_nat_range2 *range,
+		const union nf_inet_addr *newdst)
+{
+	struct nf_nat_range2 newrange;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct;
+
+	ct = nf_ct_get(skb, &ctinfo);
+
+	memset(&newrange, 0, sizeof(newrange));
+
+	newrange.flags		= range->flags | NF_NAT_RANGE_MAP_IPS;
+	newrange.min_addr	= *newdst;
+	newrange.max_addr	= *newdst;
+	newrange.min_proto	= range->min_proto;
+	newrange.max_proto	= range->max_proto;
+
+	return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST);
+}
+
 unsigned int
-nf_nat_redirect_ipv4(struct sk_buff *skb,
-		     const struct nf_nat_ipv4_multi_range_compat *mr,
+nf_nat_redirect_ipv4(struct sk_buff *skb, const struct nf_nat_range2 *range,
 		     unsigned int hooknum)
 {
-	struct nf_conn *ct;
-	enum ip_conntrack_info ctinfo;
-	__be32 newdst;
-	struct nf_nat_range2 newrange;
+	union nf_inet_addr newdst = {};
 
 	WARN_ON(hooknum != NF_INET_PRE_ROUTING &&
 		hooknum != NF_INET_LOCAL_OUT);
 
-	ct = nf_ct_get(skb, &ctinfo);
-	WARN_ON(!(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED)));
-
 	/* Local packets: make them go to loopback */
 	if (hooknum == NF_INET_LOCAL_OUT) {
-		newdst = htonl(0x7F000001);
+		newdst.ip = htonl(INADDR_LOOPBACK);
 	} else {
 		const struct in_device *indev;
 
-		newdst = 0;
-
 		indev = __in_dev_get_rcu(skb->dev);
 		if (indev) {
 			const struct in_ifaddr *ifa;
 
 			ifa = rcu_dereference(indev->ifa_list);
 			if (ifa)
-				newdst = ifa->ifa_local;
+				newdst.ip = ifa->ifa_local;
 		}
 
-		if (!newdst)
+		if (!newdst.ip)
 			return NF_DROP;
 	}
 
-	/* Transfer from original range. */
-	memset(&newrange.min_addr, 0, sizeof(newrange.min_addr));
-	memset(&newrange.max_addr, 0, sizeof(newrange.max_addr));
-	newrange.flags	     = mr->range[0].flags | NF_NAT_RANGE_MAP_IPS;
-	newrange.min_addr.ip = newdst;
-	newrange.max_addr.ip = newdst;
-	newrange.min_proto   = mr->range[0].min;
-	newrange.max_proto   = mr->range[0].max;
-
-	/* Hand modified range to generic setup. */
-	return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST);
+	return nf_nat_redirect(skb, range, &newdst);
 }
 EXPORT_SYMBOL_GPL(nf_nat_redirect_ipv4);
 
@@ -81,14 +84,10 @@ unsigned int
 nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
 		     unsigned int hooknum)
 {
-	struct nf_nat_range2 newrange;
-	struct in6_addr newdst;
-	enum ip_conntrack_info ctinfo;
-	struct nf_conn *ct;
+	union nf_inet_addr newdst = {};
 
-	ct = nf_ct_get(skb, &ctinfo);
 	if (hooknum == NF_INET_LOCAL_OUT) {
-		newdst = loopback_addr;
+		newdst.in6 = loopback_addr;
 	} else {
 		struct inet6_dev *idev;
 		struct inet6_ifaddr *ifa;
@@ -98,7 +97,7 @@ nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
 		if (idev != NULL) {
 			read_lock_bh(&idev->lock);
 			list_for_each_entry(ifa, &idev->addr_list, if_list) {
-				newdst = ifa->addr;
+				newdst.in6 = ifa->addr;
 				addr = true;
 				break;
 			}
@@ -109,12 +108,6 @@ nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
 			return NF_DROP;
 	}
 
-	newrange.flags		= range->flags | NF_NAT_RANGE_MAP_IPS;
-	newrange.min_addr.in6	= newdst;
-	newrange.max_addr.in6	= newdst;
-	newrange.min_proto	= range->min_proto;
-	newrange.max_proto	= range->max_proto;
-
-	return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST);
+	return nf_nat_redirect(skb, range, &newdst);
 }
 EXPORT_SYMBOL_GPL(nf_nat_redirect_ipv6);
diff --git a/net/netfilter/nft_redir.c b/net/netfilter/nft_redir.c
index 67cec56bc84a..a70196ffcb1e 100644
--- a/net/netfilter/nft_redir.c
+++ b/net/netfilter/nft_redir.c
@@ -64,6 +64,8 @@ static int nft_redir_init(const struct nft_ctx *ctx,
 		} else {
 			priv->sreg_proto_max = priv->sreg_proto_min;
 		}
+
+		priv->flags |= NF_NAT_RANGE_PROTO_SPECIFIED;
 	}
 
 	if (tb[NFTA_REDIR_FLAGS]) {
@@ -99,25 +101,37 @@ nla_put_failure:
 	return -1;
 }
 
-static void nft_redir_ipv4_eval(const struct nft_expr *expr,
-				struct nft_regs *regs,
-				const struct nft_pktinfo *pkt)
+static void nft_redir_eval(const struct nft_expr *expr,
+			   struct nft_regs *regs,
+			   const struct nft_pktinfo *pkt)
 {
-	struct nft_redir *priv = nft_expr_priv(expr);
-	struct nf_nat_ipv4_multi_range_compat mr;
+	const struct nft_redir *priv = nft_expr_priv(expr);
+	struct nf_nat_range2 range;
 
-	memset(&mr, 0, sizeof(mr));
+	memset(&range, 0, sizeof(range));
+	range.flags = priv->flags;
 	if (priv->sreg_proto_min) {
-		mr.range[0].min.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_min]);
-		mr.range[0].max.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_max]);
-		mr.range[0].flags |= NF_NAT_RANGE_PROTO_SPECIFIED;
+		range.min_proto.all = (__force __be16)
+			nft_reg_load16(&regs->data[priv->sreg_proto_min]);
+		range.max_proto.all = (__force __be16)
+			nft_reg_load16(&regs->data[priv->sreg_proto_max]);
 	}
 
-	mr.range[0].flags |= priv->flags;
-
-	regs->verdict.code = nf_nat_redirect_ipv4(pkt->skb, &mr, nft_hook(pkt));
+	switch (nft_pf(pkt)) {
+	case NFPROTO_IPV4:
+		regs->verdict.code = nf_nat_redirect_ipv4(pkt->skb, &range,
+							  nft_hook(pkt));
+		break;
+#ifdef CONFIG_NF_TABLES_IPV6
+	case NFPROTO_IPV6:
+		regs->verdict.code = nf_nat_redirect_ipv6(pkt->skb, &range,
+							  nft_hook(pkt));
+		break;
+#endif
+	default:
+		WARN_ON_ONCE(1);
+		break;
+	}
 }
 
 static void
@@ -130,7 +144,7 @@ static struct nft_expr_type nft_redir_ipv4_type;
 static const struct nft_expr_ops nft_redir_ipv4_ops = {
 	.type		= &nft_redir_ipv4_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_redir)),
-	.eval		= nft_redir_ipv4_eval,
+	.eval		= nft_redir_eval,
 	.init		= nft_redir_init,
 	.destroy	= nft_redir_ipv4_destroy,
 	.dump		= nft_redir_dump,
@@ -148,28 +162,6 @@ static struct nft_expr_type nft_redir_ipv4_type __read_mostly = {
 };
 
 #ifdef CONFIG_NF_TABLES_IPV6
-static void nft_redir_ipv6_eval(const struct nft_expr *expr,
-				struct nft_regs *regs,
-				const struct nft_pktinfo *pkt)
-{
-	struct nft_redir *priv = nft_expr_priv(expr);
-	struct nf_nat_range2 range;
-
-	memset(&range, 0, sizeof(range));
-	if (priv->sreg_proto_min) {
-		range.min_proto.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_min]);
-		range.max_proto.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_max]);
-		range.flags |= NF_NAT_RANGE_PROTO_SPECIFIED;
-	}
-
-	range.flags |= priv->flags;
-
-	regs->verdict.code =
-		nf_nat_redirect_ipv6(pkt->skb, &range, nft_hook(pkt));
-}
-
 static void
 nft_redir_ipv6_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr)
 {
@@ -180,7 +172,7 @@ static struct nft_expr_type nft_redir_ipv6_type;
 static const struct nft_expr_ops nft_redir_ipv6_ops = {
 	.type		= &nft_redir_ipv6_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_redir)),
-	.eval		= nft_redir_ipv6_eval,
+	.eval		= nft_redir_eval,
 	.init		= nft_redir_init,
 	.destroy	= nft_redir_ipv6_destroy,
 	.dump		= nft_redir_dump,
@@ -199,20 +191,6 @@ static struct nft_expr_type nft_redir_ipv6_type __read_mostly = {
 #endif
 
 #ifdef CONFIG_NF_TABLES_INET
-static void nft_redir_inet_eval(const struct nft_expr *expr,
-				struct nft_regs *regs,
-				const struct nft_pktinfo *pkt)
-{
-	switch (nft_pf(pkt)) {
-	case NFPROTO_IPV4:
-		return nft_redir_ipv4_eval(expr, regs, pkt);
-	case NFPROTO_IPV6:
-		return nft_redir_ipv6_eval(expr, regs, pkt);
-	}
-
-	WARN_ON_ONCE(1);
-}
-
 static void
 nft_redir_inet_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr)
 {
@@ -223,7 +201,7 @@ static struct nft_expr_type nft_redir_inet_type;
 static const struct nft_expr_ops nft_redir_inet_ops = {
 	.type		= &nft_redir_inet_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_redir)),
-	.eval		= nft_redir_inet_eval,
+	.eval		= nft_redir_eval,
 	.init		= nft_redir_init,
 	.destroy	= nft_redir_inet_destroy,
 	.dump		= nft_redir_dump,
diff --git a/net/netfilter/xt_REDIRECT.c b/net/netfilter/xt_REDIRECT.c
index 353ca7801251..ff66b56a3f97 100644
--- a/net/netfilter/xt_REDIRECT.c
+++ b/net/netfilter/xt_REDIRECT.c
@@ -46,7 +46,6 @@ static void redirect_tg_destroy(const struct xt_tgdtor_param *par)
 	nf_ct_netns_put(par->net, par->family);
 }
 
-/* FIXME: Take multiple ranges --RR */
 static int redirect_tg4_check(const struct xt_tgchk_param *par)
 {
 	const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo;
@@ -65,7 +64,14 @@ static int redirect_tg4_check(const struct xt_tgchk_param *par)
 static unsigned int
 redirect_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 {
-	return nf_nat_redirect_ipv4(skb, par->targinfo, xt_hooknum(par));
+	const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo;
+	struct nf_nat_range2 range = {
+		.flags       = mr->range[0].flags,
+		.min_proto   = mr->range[0].min,
+		.max_proto   = mr->range[0].max,
+	};
+
+	return nf_nat_redirect_ipv4(skb, &range, xt_hooknum(par));
 }
 
 static struct xt_target redirect_tg_reg[] __read_mostly = {
-- 
cgit v1.2.3-70-g09d2


From f6ca5d5ed7ecee1ff4721565fb5a276884d1e785 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Wed, 15 Mar 2023 21:48:02 +0000
Subject: netfilter: nft_masq: deduplicate eval call-backs

nft_masq has separate ipv4 and ipv6 call-backs which share much of their
code, and an inet one switch containing a switch that calls one of the
others based on the family of the packet.  Merge the ipv4 and ipv6 ones
into the inet one in order to get rid of the duplicate code.

Const-qualify the `priv` pointer since we don't need to write through it.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nft_masq.c | 75 +++++++++++++++++++-----------------------------
 1 file changed, 29 insertions(+), 46 deletions(-)

diff --git a/net/netfilter/nft_masq.c b/net/netfilter/nft_masq.c
index 9544c2f16998..b115d77fbbc7 100644
--- a/net/netfilter/nft_masq.c
+++ b/net/netfilter/nft_masq.c
@@ -96,23 +96,39 @@ nla_put_failure:
 	return -1;
 }
 
-static void nft_masq_ipv4_eval(const struct nft_expr *expr,
-			       struct nft_regs *regs,
-			       const struct nft_pktinfo *pkt)
+static void nft_masq_eval(const struct nft_expr *expr,
+			  struct nft_regs *regs,
+			  const struct nft_pktinfo *pkt)
 {
-	struct nft_masq *priv = nft_expr_priv(expr);
+	const struct nft_masq *priv = nft_expr_priv(expr);
 	struct nf_nat_range2 range;
 
 	memset(&range, 0, sizeof(range));
 	range.flags = priv->flags;
 	if (priv->sreg_proto_min) {
-		range.min_proto.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_min]);
-		range.max_proto.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_max]);
+		range.min_proto.all = (__force __be16)
+			nft_reg_load16(&regs->data[priv->sreg_proto_min]);
+		range.max_proto.all = (__force __be16)
+			nft_reg_load16(&regs->data[priv->sreg_proto_max]);
+	}
+
+	switch (nft_pf(pkt)) {
+	case NFPROTO_IPV4:
+		regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb,
+							    nft_hook(pkt),
+							    &range,
+							    nft_out(pkt));
+		break;
+#ifdef CONFIG_NF_TABLES_IPV6
+	case NFPROTO_IPV6:
+		regs->verdict.code = nf_nat_masquerade_ipv6(pkt->skb, &range,
+							    nft_out(pkt));
+		break;
+#endif
+	default:
+		WARN_ON_ONCE(1);
+		break;
 	}
-	regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, nft_hook(pkt),
-						    &range, nft_out(pkt));
 }
 
 static void
@@ -125,7 +141,7 @@ static struct nft_expr_type nft_masq_ipv4_type;
 static const struct nft_expr_ops nft_masq_ipv4_ops = {
 	.type		= &nft_masq_ipv4_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_masq)),
-	.eval		= nft_masq_ipv4_eval,
+	.eval		= nft_masq_eval,
 	.init		= nft_masq_init,
 	.destroy	= nft_masq_ipv4_destroy,
 	.dump		= nft_masq_dump,
@@ -143,25 +159,6 @@ static struct nft_expr_type nft_masq_ipv4_type __read_mostly = {
 };
 
 #ifdef CONFIG_NF_TABLES_IPV6
-static void nft_masq_ipv6_eval(const struct nft_expr *expr,
-			       struct nft_regs *regs,
-			       const struct nft_pktinfo *pkt)
-{
-	struct nft_masq *priv = nft_expr_priv(expr);
-	struct nf_nat_range2 range;
-
-	memset(&range, 0, sizeof(range));
-	range.flags = priv->flags;
-	if (priv->sreg_proto_min) {
-		range.min_proto.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_min]);
-		range.max_proto.all = (__force __be16)nft_reg_load16(
-			&regs->data[priv->sreg_proto_max]);
-	}
-	regs->verdict.code = nf_nat_masquerade_ipv6(pkt->skb, &range,
-						    nft_out(pkt));
-}
-
 static void
 nft_masq_ipv6_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr)
 {
@@ -172,7 +169,7 @@ static struct nft_expr_type nft_masq_ipv6_type;
 static const struct nft_expr_ops nft_masq_ipv6_ops = {
 	.type		= &nft_masq_ipv6_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_masq)),
-	.eval		= nft_masq_ipv6_eval,
+	.eval		= nft_masq_eval,
 	.init		= nft_masq_init,
 	.destroy	= nft_masq_ipv6_destroy,
 	.dump		= nft_masq_dump,
@@ -204,20 +201,6 @@ static inline void nft_masq_module_exit_ipv6(void) {}
 #endif
 
 #ifdef CONFIG_NF_TABLES_INET
-static void nft_masq_inet_eval(const struct nft_expr *expr,
-			       struct nft_regs *regs,
-			       const struct nft_pktinfo *pkt)
-{
-	switch (nft_pf(pkt)) {
-	case NFPROTO_IPV4:
-		return nft_masq_ipv4_eval(expr, regs, pkt);
-	case NFPROTO_IPV6:
-		return nft_masq_ipv6_eval(expr, regs, pkt);
-	}
-
-	WARN_ON_ONCE(1);
-}
-
 static void
 nft_masq_inet_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr)
 {
@@ -228,7 +211,7 @@ static struct nft_expr_type nft_masq_inet_type;
 static const struct nft_expr_ops nft_masq_inet_ops = {
 	.type		= &nft_masq_inet_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_masq)),
-	.eval		= nft_masq_inet_eval,
+	.eval		= nft_masq_eval,
 	.init		= nft_masq_init,
 	.destroy	= nft_masq_inet_destroy,
 	.dump		= nft_masq_dump,
-- 
cgit v1.2.3-70-g09d2


From bde7170a04d69ab8013d0582e60e414d46d46dd6 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 16 Mar 2023 13:59:48 +0100
Subject: netfilter: xtables: disable 32bit compat interface by default

This defaulted to 'y' because before this knob existed the 32bit
compat layer was always compiled in if CONFIG_COMPAT was set.

32bit iptables on 64bit kernel isn't common anymore, so remove
the default-y now.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 4d6737160857..d0bf630482c1 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -753,7 +753,6 @@ if NETFILTER_XTABLES
 config NETFILTER_XTABLES_COMPAT
 	bool "Netfilter Xtables 32bit support"
 	depends on COMPAT
-	default y
 	help
 	   This option provides a translation layer to run 32bit arp,ip(6),ebtables
 	   binaries on 64bit kernels.
-- 
cgit v1.2.3-70-g09d2


From 36ce9982ef2fb63fdf39996900866965d71f5a5e Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 16 Mar 2023 15:44:08 +0100
Subject: xtables: move icmp/icmpv6 logic to xt_tcpudp

icmp/icmp6 matches are baked into ip(6)_tables.ko.

This means that even if iptables-nft is used, a rule like
"-p icmp --icmp-type 1" will load the ip(6)tables modules.

Move them to xt_tcpdudp.ko instead to avoid this.

This will also allow to eventually add kconfig knobs to build kernels
that support iptables-nft but not iptables-legacy (old set/getsockopt
interface).

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/ipv4/netfilter/ip_tables.c  |  68 +------------------------
 net/ipv6/netfilter/ip6_tables.c |  68 +------------------------
 net/netfilter/xt_tcpudp.c       | 110 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 112 insertions(+), 134 deletions(-)

diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index da5998011ab9..7da1df4997d0 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -14,7 +14,6 @@
 #include <linux/vmalloc.h>
 #include <linux/netdevice.h>
 #include <linux/module.h>
-#include <linux/icmp.h>
 #include <net/ip.h>
 #include <net/compat.h>
 #include <linux/uaccess.h>
@@ -31,7 +30,6 @@
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
 MODULE_DESCRIPTION("IPv4 packet filter");
-MODULE_ALIAS("ipt_icmp");
 
 void *ipt_alloc_initial_table(const struct xt_table *info)
 {
@@ -1799,52 +1797,6 @@ void ipt_unregister_table_exit(struct net *net, const char *name)
 		__ipt_unregister_table(net, table);
 }
 
-/* Returns 1 if the type and code is matched by the range, 0 otherwise */
-static inline bool
-icmp_type_code_match(u_int8_t test_type, u_int8_t min_code, u_int8_t max_code,
-		     u_int8_t type, u_int8_t code,
-		     bool invert)
-{
-	return ((test_type == 0xFF) ||
-		(type == test_type && code >= min_code && code <= max_code))
-		^ invert;
-}
-
-static bool
-icmp_match(const struct sk_buff *skb, struct xt_action_param *par)
-{
-	const struct icmphdr *ic;
-	struct icmphdr _icmph;
-	const struct ipt_icmp *icmpinfo = par->matchinfo;
-
-	/* Must not be a fragment. */
-	if (par->fragoff != 0)
-		return false;
-
-	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
-	if (ic == NULL) {
-		/* We've been asked to examine this packet, and we
-		 * can't.  Hence, no choice but to drop.
-		 */
-		par->hotdrop = true;
-		return false;
-	}
-
-	return icmp_type_code_match(icmpinfo->type,
-				    icmpinfo->code[0],
-				    icmpinfo->code[1],
-				    ic->type, ic->code,
-				    !!(icmpinfo->invflags&IPT_ICMP_INV));
-}
-
-static int icmp_checkentry(const struct xt_mtchk_param *par)
-{
-	const struct ipt_icmp *icmpinfo = par->matchinfo;
-
-	/* Must specify no unknown invflags */
-	return (icmpinfo->invflags & ~IPT_ICMP_INV) ? -EINVAL : 0;
-}
-
 static struct xt_target ipt_builtin_tg[] __read_mostly = {
 	{
 		.name             = XT_STANDARD_TARGET,
@@ -1875,18 +1827,6 @@ static struct nf_sockopt_ops ipt_sockopts = {
 	.owner		= THIS_MODULE,
 };
 
-static struct xt_match ipt_builtin_mt[] __read_mostly = {
-	{
-		.name       = "icmp",
-		.match      = icmp_match,
-		.matchsize  = sizeof(struct ipt_icmp),
-		.checkentry = icmp_checkentry,
-		.proto      = IPPROTO_ICMP,
-		.family     = NFPROTO_IPV4,
-		.me	    = THIS_MODULE,
-	},
-};
-
 static int __net_init ip_tables_net_init(struct net *net)
 {
 	return xt_proto_init(net, NFPROTO_IPV4);
@@ -1914,19 +1854,14 @@ static int __init ip_tables_init(void)
 	ret = xt_register_targets(ipt_builtin_tg, ARRAY_SIZE(ipt_builtin_tg));
 	if (ret < 0)
 		goto err2;
-	ret = xt_register_matches(ipt_builtin_mt, ARRAY_SIZE(ipt_builtin_mt));
-	if (ret < 0)
-		goto err4;
 
 	/* Register setsockopt */
 	ret = nf_register_sockopt(&ipt_sockopts);
 	if (ret < 0)
-		goto err5;
+		goto err4;
 
 	return 0;
 
-err5:
-	xt_unregister_matches(ipt_builtin_mt, ARRAY_SIZE(ipt_builtin_mt));
 err4:
 	xt_unregister_targets(ipt_builtin_tg, ARRAY_SIZE(ipt_builtin_tg));
 err2:
@@ -1939,7 +1874,6 @@ static void __exit ip_tables_fini(void)
 {
 	nf_unregister_sockopt(&ipt_sockopts);
 
-	xt_unregister_matches(ipt_builtin_mt, ARRAY_SIZE(ipt_builtin_mt));
 	xt_unregister_targets(ipt_builtin_tg, ARRAY_SIZE(ipt_builtin_tg));
 	unregister_pernet_subsys(&ip_tables_net_ops);
 }
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 0ce0ed17c758..fd9f049d6d41 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -18,7 +18,6 @@
 #include <linux/netdevice.h>
 #include <linux/module.h>
 #include <linux/poison.h>
-#include <linux/icmpv6.h>
 #include <net/ipv6.h>
 #include <net/compat.h>
 #include <linux/uaccess.h>
@@ -35,7 +34,6 @@
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>");
 MODULE_DESCRIPTION("IPv6 packet filter");
-MODULE_ALIAS("ip6t_icmp6");
 
 void *ip6t_alloc_initial_table(const struct xt_table *info)
 {
@@ -1805,52 +1803,6 @@ void ip6t_unregister_table_exit(struct net *net, const char *name)
 		__ip6t_unregister_table(net, table);
 }
 
-/* Returns 1 if the type and code is matched by the range, 0 otherwise */
-static inline bool
-icmp6_type_code_match(u_int8_t test_type, u_int8_t min_code, u_int8_t max_code,
-		     u_int8_t type, u_int8_t code,
-		     bool invert)
-{
-	return (type == test_type && code >= min_code && code <= max_code)
-		^ invert;
-}
-
-static bool
-icmp6_match(const struct sk_buff *skb, struct xt_action_param *par)
-{
-	const struct icmp6hdr *ic;
-	struct icmp6hdr _icmph;
-	const struct ip6t_icmp *icmpinfo = par->matchinfo;
-
-	/* Must not be a fragment. */
-	if (par->fragoff != 0)
-		return false;
-
-	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
-	if (ic == NULL) {
-		/* We've been asked to examine this packet, and we
-		 * can't.  Hence, no choice but to drop.
-		 */
-		par->hotdrop = true;
-		return false;
-	}
-
-	return icmp6_type_code_match(icmpinfo->type,
-				     icmpinfo->code[0],
-				     icmpinfo->code[1],
-				     ic->icmp6_type, ic->icmp6_code,
-				     !!(icmpinfo->invflags&IP6T_ICMP_INV));
-}
-
-/* Called when user tries to insert an entry of this type. */
-static int icmp6_checkentry(const struct xt_mtchk_param *par)
-{
-	const struct ip6t_icmp *icmpinfo = par->matchinfo;
-
-	/* Must specify no unknown invflags */
-	return (icmpinfo->invflags & ~IP6T_ICMP_INV) ? -EINVAL : 0;
-}
-
 /* The built-in targets: standard (NULL) and error. */
 static struct xt_target ip6t_builtin_tg[] __read_mostly = {
 	{
@@ -1882,18 +1834,6 @@ static struct nf_sockopt_ops ip6t_sockopts = {
 	.owner		= THIS_MODULE,
 };
 
-static struct xt_match ip6t_builtin_mt[] __read_mostly = {
-	{
-		.name       = "icmp6",
-		.match      = icmp6_match,
-		.matchsize  = sizeof(struct ip6t_icmp),
-		.checkentry = icmp6_checkentry,
-		.proto      = IPPROTO_ICMPV6,
-		.family     = NFPROTO_IPV6,
-		.me	    = THIS_MODULE,
-	},
-};
-
 static int __net_init ip6_tables_net_init(struct net *net)
 {
 	return xt_proto_init(net, NFPROTO_IPV6);
@@ -1921,19 +1861,14 @@ static int __init ip6_tables_init(void)
 	ret = xt_register_targets(ip6t_builtin_tg, ARRAY_SIZE(ip6t_builtin_tg));
 	if (ret < 0)
 		goto err2;
-	ret = xt_register_matches(ip6t_builtin_mt, ARRAY_SIZE(ip6t_builtin_mt));
-	if (ret < 0)
-		goto err4;
 
 	/* Register setsockopt */
 	ret = nf_register_sockopt(&ip6t_sockopts);
 	if (ret < 0)
-		goto err5;
+		goto err4;
 
 	return 0;
 
-err5:
-	xt_unregister_matches(ip6t_builtin_mt, ARRAY_SIZE(ip6t_builtin_mt));
 err4:
 	xt_unregister_targets(ip6t_builtin_tg, ARRAY_SIZE(ip6t_builtin_tg));
 err2:
@@ -1946,7 +1881,6 @@ static void __exit ip6_tables_fini(void)
 {
 	nf_unregister_sockopt(&ip6t_sockopts);
 
-	xt_unregister_matches(ip6t_builtin_mt, ARRAY_SIZE(ip6t_builtin_mt));
 	xt_unregister_targets(ip6t_builtin_tg, ARRAY_SIZE(ip6t_builtin_tg));
 	unregister_pernet_subsys(&ip6_tables_net_ops);
 }
diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c
index 11ec2abf0c72..e8991130a3de 100644
--- a/net/netfilter/xt_tcpudp.c
+++ b/net/netfilter/xt_tcpudp.c
@@ -4,6 +4,7 @@
 #include <linux/module.h>
 #include <net/ip.h>
 #include <linux/ipv6.h>
+#include <linux/icmp.h>
 #include <net/ipv6.h>
 #include <net/tcp.h>
 #include <net/udp.h>
@@ -20,6 +21,8 @@ MODULE_ALIAS("ipt_udp");
 MODULE_ALIAS("ipt_tcp");
 MODULE_ALIAS("ip6t_udp");
 MODULE_ALIAS("ip6t_tcp");
+MODULE_ALIAS("ipt_icmp");
+MODULE_ALIAS("ip6t_icmp6");
 
 /* Returns 1 if the port is matched by the range, 0 otherwise */
 static inline bool
@@ -161,6 +164,95 @@ static int udp_mt_check(const struct xt_mtchk_param *par)
 	return (udpinfo->invflags & ~XT_UDP_INV_MASK) ? -EINVAL : 0;
 }
 
+/* Returns 1 if the type and code is matched by the range, 0 otherwise */
+static bool type_code_in_range(u8 test_type, u8 min_code, u8 max_code,
+			       u8 type, u8 code)
+{
+	return type == test_type && code >= min_code && code <= max_code;
+}
+
+static bool icmp_type_code_match(u8 test_type, u8 min_code, u8 max_code,
+				 u8 type, u8 code, bool invert)
+{
+	return (test_type == 0xFF ||
+		type_code_in_range(test_type, min_code, max_code, type, code))
+		^ invert;
+}
+
+static bool icmp6_type_code_match(u8 test_type, u8 min_code, u8 max_code,
+				  u8 type, u8 code, bool invert)
+{
+	return type_code_in_range(test_type, min_code, max_code, type, code) ^ invert;
+}
+
+static bool
+icmp_match(const struct sk_buff *skb, struct xt_action_param *par)
+{
+	const struct icmphdr *ic;
+	struct icmphdr _icmph;
+	const struct ipt_icmp *icmpinfo = par->matchinfo;
+
+	/* Must not be a fragment. */
+	if (par->fragoff != 0)
+		return false;
+
+	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
+	if (!ic) {
+		/* We've been asked to examine this packet, and we
+		 * can't.  Hence, no choice but to drop.
+		 */
+		par->hotdrop = true;
+		return false;
+	}
+
+	return icmp_type_code_match(icmpinfo->type,
+				    icmpinfo->code[0],
+				    icmpinfo->code[1],
+				    ic->type, ic->code,
+				    !!(icmpinfo->invflags & IPT_ICMP_INV));
+}
+
+static bool
+icmp6_match(const struct sk_buff *skb, struct xt_action_param *par)
+{
+	const struct icmp6hdr *ic;
+	struct icmp6hdr _icmph;
+	const struct ip6t_icmp *icmpinfo = par->matchinfo;
+
+	/* Must not be a fragment. */
+	if (par->fragoff != 0)
+		return false;
+
+	ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph);
+	if (!ic) {
+		/* We've been asked to examine this packet, and we
+		 * can't.  Hence, no choice but to drop.
+		 */
+		par->hotdrop = true;
+		return false;
+	}
+
+	return icmp6_type_code_match(icmpinfo->type,
+				     icmpinfo->code[0],
+				     icmpinfo->code[1],
+				     ic->icmp6_type, ic->icmp6_code,
+				     !!(icmpinfo->invflags & IP6T_ICMP_INV));
+}
+
+static int icmp_checkentry(const struct xt_mtchk_param *par)
+{
+	const struct ipt_icmp *icmpinfo = par->matchinfo;
+
+	return (icmpinfo->invflags & ~IPT_ICMP_INV) ? -EINVAL : 0;
+}
+
+static int icmp6_checkentry(const struct xt_mtchk_param *par)
+{
+	const struct ip6t_icmp *icmpinfo = par->matchinfo;
+
+	return (icmpinfo->invflags & ~IP6T_ICMP_INV) ? -EINVAL : 0;
+}
+
 static struct xt_match tcpudp_mt_reg[] __read_mostly = {
 	{
 		.name		= "tcp",
@@ -216,6 +308,24 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = {
 		.proto		= IPPROTO_UDPLITE,
 		.me		= THIS_MODULE,
 	},
+	{
+		.name       = "icmp",
+		.match      = icmp_match,
+		.matchsize  = sizeof(struct ipt_icmp),
+		.checkentry = icmp_checkentry,
+		.proto      = IPPROTO_ICMP,
+		.family     = NFPROTO_IPV4,
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "icmp6",
+		.match      = icmp6_match,
+		.matchsize  = sizeof(struct ip6t_icmp),
+		.checkentry = icmp6_checkentry,
+		.proto      = IPPROTO_ICMPV6,
+		.family     = NFPROTO_IPV6,
+		.me	    = THIS_MODULE,
+	},
 };
 
 static int __init tcpudp_mt_init(void)
-- 
cgit v1.2.3-70-g09d2


From b0e214d212030fe497d4d150bb3474e50ad5d093 Mon Sep 17 00:00:00 2001
From: Madhu Koriginja <madhu.koriginja@nxp.com>
Date: Tue, 21 Mar 2023 21:28:44 +0530
Subject: netfilter: keep conntrack reference until IPsecv6 policy checks are
 done

Keep the conntrack reference until policy checks have been performed for
IPsec V6 NAT support, just like ipv4.

The reference needs to be dropped before a packet is
queued to avoid having the conntrack module unloadable.

Fixes: 58a317f1061c ("netfilter: ipv6: add IPv6 NAT support")
Signed-off-by: Madhu Koriginja <madhu.koriginja@nxp.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/dccp/ipv6.c      |  1 +
 net/ipv6/ip6_input.c | 14 ++++++--------
 net/ipv6/raw.c       |  5 ++---
 net/ipv6/tcp_ipv6.c  |  2 ++
 net/ipv6/udp.c       |  2 ++
 5 files changed, 13 insertions(+), 11 deletions(-)

diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 47fb10834223..93c98990d726 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -784,6 +784,7 @@ lookup:
 
 	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
 		goto discard_and_relse;
+	nf_reset_ct(skb);
 
 	return __sk_receive_skb(sk, skb, 1, dh->dccph_doff * 4,
 				refcounted) ? -1 : 0;
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index e1ebf5e42ebe..d94041bb4287 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -404,10 +404,6 @@ resubmit_final:
 			/* Only do this once for first final protocol */
 			have_final = true;
 
-			/* Free reference early: we don't need it any more,
-			   and it may hold ip_conntrack module loaded
-			   indefinitely. */
-			nf_reset_ct(skb);
 
 			skb_postpull_rcsum(skb, skb_network_header(skb),
 					   skb_network_header_len(skb));
@@ -430,10 +426,12 @@ resubmit_final:
 				goto discard;
 			}
 		}
-		if (!(ipprot->flags & INET6_PROTO_NOPOLICY) &&
-		    !xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) {
-			SKB_DR_SET(reason, XFRM_POLICY);
-			goto discard;
+		if (!(ipprot->flags & INET6_PROTO_NOPOLICY)) {
+			if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) {
+				SKB_DR_SET(reason, XFRM_POLICY);
+				goto discard;
+			}
+			nf_reset_ct(skb);
 		}
 
 		ret = INDIRECT_CALL_2(ipprot->handler, tcp_v6_rcv, udpv6_rcv,
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 6ac2f2690c44..4ab62a9c5c8e 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -194,10 +194,8 @@ static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
 			struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);
 
 			/* Not releasing hash table! */
-			if (clone) {
-				nf_reset_ct(clone);
+			if (clone)
 				rawv6_rcv(sk, clone);
-			}
 		}
 	}
 	rcu_read_unlock();
@@ -391,6 +389,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
 		kfree_skb_reason(skb, SKB_DROP_REASON_XFRM_POLICY);
 		return NET_RX_DROP;
 	}
+	nf_reset_ct(skb);
 
 	if (!rp->checksum)
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 35cf523c9efd..244cf86c4cbb 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1723,6 +1723,8 @@ process:
 	if (drop_reason)
 		goto discard_and_relse;
 
+	nf_reset_ct(skb);
+
 	if (tcp_filter(sk, skb)) {
 		drop_reason = SKB_DROP_REASON_SOCKET_FILTER;
 		goto discard_and_relse;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index d350e57c4792..4caa70a1b871 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -704,6 +704,7 @@ static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
 		drop_reason = SKB_DROP_REASON_XFRM_POLICY;
 		goto drop;
 	}
+	nf_reset_ct(skb);
 
 	if (static_branch_unlikely(&udpv6_encap_needed_key) && up->encap_type) {
 		int (*encap_rcv)(struct sock *sk, struct sk_buff *skb);
@@ -1027,6 +1028,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
 
 	if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
 		goto discard;
+	nf_reset_ct(skb);
 
 	if (udp_lib_checksum_complete(skb))
 		goto csum_error;
-- 
cgit v1.2.3-70-g09d2


From 830154cdc57971b06f81d4ffc39b868e3d7693de Mon Sep 17 00:00:00 2001
From: JP Kobryn <inwardvessel@gmail.com>
Date: Wed, 22 Mar 2023 12:47:53 -0700
Subject: bpf/selftests: coverage for bpf_map_ops errors

These tests expose the issue of being unable to properly check for errors
returned from inlined bpf map helpers that make calls to the bpf_map_ops
functions. At best, a check for zero or non-zero can be done but these
tests show it is not possible to check for a negative value or for a
specific error value.

Signed-off-by: JP Kobryn <inwardvessel@gmail.com>
Tested-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230322194754.185781-2-inwardvessel@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/map_ops.c | 162 +++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/test_map_ops.c | 138 +++++++++++++++++++
 2 files changed, 300 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/map_ops.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_map_ops.c

diff --git a/tools/testing/selftests/bpf/prog_tests/map_ops.c b/tools/testing/selftests/bpf/prog_tests/map_ops.c
new file mode 100644
index 000000000000..be5e42a413b4
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/map_ops.c
@@ -0,0 +1,162 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <errno.h>
+#include <sys/syscall.h>
+#include <unistd.h>
+
+#include "test_map_ops.skel.h"
+#include "test_progs.h"
+
+static void map_update(void)
+{
+	(void)syscall(__NR_getpid);
+}
+
+static void map_delete(void)
+{
+	(void)syscall(__NR_getppid);
+}
+
+static void map_push(void)
+{
+	(void)syscall(__NR_getuid);
+}
+
+static void map_pop(void)
+{
+	(void)syscall(__NR_geteuid);
+}
+
+static void map_peek(void)
+{
+	(void)syscall(__NR_getgid);
+}
+
+static void map_for_each_pass(void)
+{
+	(void)syscall(__NR_gettid);
+}
+
+static void map_for_each_fail(void)
+{
+	(void)syscall(__NR_getpgid);
+}
+
+static int setup(struct test_map_ops **skel)
+{
+	int err = 0;
+
+	if (!skel)
+		return -1;
+
+	*skel = test_map_ops__open();
+	if (!ASSERT_OK_PTR(*skel, "test_map_ops__open"))
+		return -1;
+
+	(*skel)->rodata->pid = getpid();
+
+	err = test_map_ops__load(*skel);
+	if (!ASSERT_OK(err, "test_map_ops__load"))
+		return err;
+
+	err = test_map_ops__attach(*skel);
+	if (!ASSERT_OK(err, "test_map_ops__attach"))
+		return err;
+
+	return err;
+}
+
+static void teardown(struct test_map_ops **skel)
+{
+	if (skel && *skel)
+		test_map_ops__destroy(*skel);
+}
+
+static void map_ops_update_delete_subtest(void)
+{
+	struct test_map_ops *skel;
+
+	if (setup(&skel))
+		goto teardown;
+
+	map_update();
+	ASSERT_OK(skel->bss->err, "map_update_initial");
+
+	map_update();
+	ASSERT_LT(skel->bss->err, 0, "map_update_existing");
+	ASSERT_EQ(skel->bss->err, -EEXIST, "map_update_existing");
+
+	map_delete();
+	ASSERT_OK(skel->bss->err, "map_delete_existing");
+
+	map_delete();
+	ASSERT_LT(skel->bss->err, 0, "map_delete_non_existing");
+	ASSERT_EQ(skel->bss->err, -ENOENT, "map_delete_non_existing");
+
+teardown:
+	teardown(&skel);
+}
+
+static void map_ops_push_peek_pop_subtest(void)
+{
+	struct test_map_ops *skel;
+
+	if (setup(&skel))
+		goto teardown;
+
+	map_push();
+	ASSERT_OK(skel->bss->err, "map_push_initial");
+
+	map_push();
+	ASSERT_LT(skel->bss->err, 0, "map_push_when_full");
+	ASSERT_EQ(skel->bss->err, -E2BIG, "map_push_when_full");
+
+	map_peek();
+	ASSERT_OK(skel->bss->err, "map_peek");
+
+	map_pop();
+	ASSERT_OK(skel->bss->err, "map_pop");
+
+	map_peek();
+	ASSERT_LT(skel->bss->err, 0, "map_peek_when_empty");
+	ASSERT_EQ(skel->bss->err, -ENOENT, "map_peek_when_empty");
+
+	map_pop();
+	ASSERT_LT(skel->bss->err, 0, "map_pop_when_empty");
+	ASSERT_EQ(skel->bss->err, -ENOENT, "map_pop_when_empty");
+
+teardown:
+	teardown(&skel);
+}
+
+static void map_ops_for_each_subtest(void)
+{
+	struct test_map_ops *skel;
+
+	if (setup(&skel))
+		goto teardown;
+
+	map_for_each_pass();
+	/* expect to iterate over 1 element */
+	ASSERT_EQ(skel->bss->err, 1, "map_for_each_no_flags");
+
+	map_for_each_fail();
+	ASSERT_LT(skel->bss->err, 0, "map_for_each_with_flags");
+	ASSERT_EQ(skel->bss->err, -EINVAL, "map_for_each_with_flags");
+
+teardown:
+	teardown(&skel);
+}
+
+void test_map_ops(void)
+{
+	if (test__start_subtest("map_ops_update_delete"))
+		map_ops_update_delete_subtest();
+
+	if (test__start_subtest("map_ops_push_peek_pop"))
+		map_ops_push_peek_pop_subtest();
+
+	if (test__start_subtest("map_ops_for_each"))
+		map_ops_for_each_subtest();
+}
diff --git a/tools/testing/selftests/bpf/progs/test_map_ops.c b/tools/testing/selftests/bpf/progs/test_map_ops.c
new file mode 100644
index 000000000000..b53b46a090c8
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_map_ops.c
@@ -0,0 +1,138 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+
+char _license[] SEC("license") = "GPL";
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+} hash_map SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_STACK);
+	__uint(max_entries, 1);
+	__type(value, int);
+} stack_map SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+} array_map SEC(".maps");
+
+const volatile pid_t pid;
+long err = 0;
+
+static u64 callback(u64 map, u64 key, u64 val, u64 ctx, u64 flags)
+{
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_getpid")
+int map_update(void *ctx)
+{
+	const int key = 0;
+	const int val = 1;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	err = bpf_map_update_elem(&hash_map, &key, &val, BPF_NOEXIST);
+
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_getppid")
+int map_delete(void *ctx)
+{
+	const int key = 0;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	err = bpf_map_delete_elem(&hash_map, &key);
+
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_getuid")
+int map_push(void *ctx)
+{
+	const int val = 1;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	err = bpf_map_push_elem(&stack_map, &val, 0);
+
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_geteuid")
+int map_pop(void *ctx)
+{
+	int val;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	err = bpf_map_pop_elem(&stack_map, &val);
+
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_getgid")
+int map_peek(void *ctx)
+{
+	int val;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	err = bpf_map_peek_elem(&stack_map, &val);
+
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_gettid")
+int map_for_each_pass(void *ctx)
+{
+	const int key = 0;
+	const int val = 1;
+	const u64 flags = 0;
+	int callback_ctx;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	bpf_map_update_elem(&array_map, &key, &val, flags);
+
+	err = bpf_for_each_map_elem(&array_map, callback, &callback_ctx, flags);
+
+	return 0;
+}
+
+SEC("tp/syscalls/sys_enter_getpgid")
+int map_for_each_fail(void *ctx)
+{
+	const int key = 0;
+	const int val = 1;
+	const u64 flags = BPF_NOEXIST;
+	int callback_ctx;
+
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+
+	bpf_map_update_elem(&array_map, &key, &val, flags);
+
+	/* calling for_each with non-zero flags will return error */
+	err = bpf_for_each_map_elem(&array_map, callback, &callback_ctx, flags);
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From d7ba4cc900bf1eea2d8c807c6b1fc6bd61f41237 Mon Sep 17 00:00:00 2001
From: JP Kobryn <inwardvessel@gmail.com>
Date: Wed, 22 Mar 2023 12:47:54 -0700
Subject: bpf: return long from bpf_map_ops funcs

This patch changes the return types of bpf_map_ops functions to long, where
previously int was returned. Using long allows for bpf programs to maintain
the sign bit in the absence of sign extension during situations where
inlined bpf helper funcs make calls to the bpf_map_ops funcs and a negative
error is returned.

The definitions of the helper funcs are generated from comments in the bpf
uapi header at `include/uapi/linux/bpf.h`. The return type of these
helpers was previously changed from int to long in commit bdb7b79b4ce8. For
any case where one of the map helpers call the bpf_map_ops funcs that are
still returning 32-bit int, a compiler might not include sign extension
instructions to properly convert the 32-bit negative value a 64-bit
negative value.

For example:
bpf assembly excerpt of an inlined helper calling a kernel function and
checking for a specific error:

; err = bpf_map_update_elem(&mymap, &key, &val, BPF_NOEXIST);
  ...
  46:	call   0xffffffffe103291c	; htab_map_update_elem
; if (err && err != -EEXIST) {
  4b:	cmp    $0xffffffffffffffef,%rax ; cmp -EEXIST,%rax

kernel function assembly excerpt of return value from
`htab_map_update_elem` returning 32-bit int:

movl $0xffffffef, %r9d
...
movl %r9d, %eax

...results in the comparison:
cmp $0xffffffffffffffef, $0x00000000ffffffef

Fixes: bdb7b79b4ce8 ("bpf: Switch most helper return values from 32-bit int to 64-bit long")
Tested-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: JP Kobryn <inwardvessel@gmail.com>
Link: https://lore.kernel.org/r/20230322194754.185781-3-inwardvessel@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h            | 14 +++++++-------
 include/linux/filter.h         |  6 +++---
 kernel/bpf/arraymap.c          | 12 ++++++------
 kernel/bpf/bloom_filter.c      | 12 ++++++------
 kernel/bpf/bpf_cgrp_storage.c  |  6 +++---
 kernel/bpf/bpf_inode_storage.c |  6 +++---
 kernel/bpf/bpf_struct_ops.c    |  6 +++---
 kernel/bpf/bpf_task_storage.c  |  6 +++---
 kernel/bpf/cpumap.c            |  8 ++++----
 kernel/bpf/devmap.c            | 24 ++++++++++++------------
 kernel/bpf/hashtab.c           | 36 ++++++++++++++++++------------------
 kernel/bpf/local_storage.c     |  6 +++---
 kernel/bpf/lpm_trie.c          |  6 +++---
 kernel/bpf/queue_stack_maps.c  | 22 +++++++++++-----------
 kernel/bpf/reuseport_array.c   |  2 +-
 kernel/bpf/ringbuf.c           |  6 +++---
 kernel/bpf/stackmap.c          |  6 +++---
 kernel/bpf/verifier.c          | 14 +++++++-------
 net/core/bpf_sk_storage.c      |  6 +++---
 net/core/sock_map.c            |  8 ++++----
 net/xdp/xskmap.c               |  8 ++++----
 21 files changed, 110 insertions(+), 110 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 3ef98fb92987..ec0df059f562 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -96,11 +96,11 @@ struct bpf_map_ops {
 
 	/* funcs callable from userspace and from eBPF programs */
 	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
-	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
-	int (*map_delete_elem)(struct bpf_map *map, void *key);
-	int (*map_push_elem)(struct bpf_map *map, void *value, u64 flags);
-	int (*map_pop_elem)(struct bpf_map *map, void *value);
-	int (*map_peek_elem)(struct bpf_map *map, void *value);
+	long (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
+	long (*map_delete_elem)(struct bpf_map *map, void *key);
+	long (*map_push_elem)(struct bpf_map *map, void *value, u64 flags);
+	long (*map_pop_elem)(struct bpf_map *map, void *value);
+	long (*map_peek_elem)(struct bpf_map *map, void *value);
 	void *(*map_lookup_percpu_elem)(struct bpf_map *map, void *key, u32 cpu);
 
 	/* funcs called by prog_array and perf_event_array map */
@@ -139,7 +139,7 @@ struct bpf_map_ops {
 	struct bpf_local_storage __rcu ** (*map_owner_storage_ptr)(void *owner);
 
 	/* Misc helpers.*/
-	int (*map_redirect)(struct bpf_map *map, u64 key, u64 flags);
+	long (*map_redirect)(struct bpf_map *map, u64 key, u64 flags);
 
 	/* map_meta_equal must be implemented for maps that can be
 	 * used as an inner map.  It is a runtime check to ensure
@@ -157,7 +157,7 @@ struct bpf_map_ops {
 	int (*map_set_for_each_callback_args)(struct bpf_verifier_env *env,
 					      struct bpf_func_state *caller,
 					      struct bpf_func_state *callee);
-	int (*map_for_each_callback)(struct bpf_map *map,
+	long (*map_for_each_callback)(struct bpf_map *map,
 				     bpf_callback_t callback_fn,
 				     void *callback_ctx, u64 flags);
 
diff --git a/include/linux/filter.h b/include/linux/filter.h
index efa5d4a1677e..23c08c31bea9 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1504,9 +1504,9 @@ static inline bool bpf_sk_lookup_run_v6(struct net *net, int protocol,
 }
 #endif /* IS_ENABLED(CONFIG_IPV6) */
 
-static __always_inline int __bpf_xdp_redirect_map(struct bpf_map *map, u64 index,
-						  u64 flags, const u64 flag_mask,
-						  void *lookup_elem(struct bpf_map *map, u32 key))
+static __always_inline long __bpf_xdp_redirect_map(struct bpf_map *map, u64 index,
+						   u64 flags, const u64 flag_mask,
+						   void *lookup_elem(struct bpf_map *map, u32 key))
 {
 	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
 	const u64 action_mask = XDP_ABORTED | XDP_DROP | XDP_PASS | XDP_TX;
diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 1588c793a715..2058e89b5ddd 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -307,8 +307,8 @@ static int array_map_get_next_key(struct bpf_map *map, void *key, void *next_key
 }
 
 /* Called from syscall or from eBPF program */
-static int array_map_update_elem(struct bpf_map *map, void *key, void *value,
-				 u64 map_flags)
+static long array_map_update_elem(struct bpf_map *map, void *key, void *value,
+				  u64 map_flags)
 {
 	struct bpf_array *array = container_of(map, struct bpf_array, map);
 	u32 index = *(u32 *)key;
@@ -386,7 +386,7 @@ int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value,
 }
 
 /* Called from syscall or from eBPF program */
-static int array_map_delete_elem(struct bpf_map *map, void *key)
+static long array_map_delete_elem(struct bpf_map *map, void *key)
 {
 	return -EINVAL;
 }
@@ -686,8 +686,8 @@ static const struct bpf_iter_seq_info iter_seq_info = {
 	.seq_priv_size		= sizeof(struct bpf_iter_seq_array_map_info),
 };
 
-static int bpf_for_each_array_elem(struct bpf_map *map, bpf_callback_t callback_fn,
-				   void *callback_ctx, u64 flags)
+static long bpf_for_each_array_elem(struct bpf_map *map, bpf_callback_t callback_fn,
+				    void *callback_ctx, u64 flags)
 {
 	u32 i, key, num_elems = 0;
 	struct bpf_array *array;
@@ -871,7 +871,7 @@ int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
 	return 0;
 }
 
-static int fd_array_map_delete_elem(struct bpf_map *map, void *key)
+static long fd_array_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_array *array = container_of(map, struct bpf_array, map);
 	void *old_ptr;
diff --git a/kernel/bpf/bloom_filter.c b/kernel/bpf/bloom_filter.c
index 6350c5d35a9b..db19784601a7 100644
--- a/kernel/bpf/bloom_filter.c
+++ b/kernel/bpf/bloom_filter.c
@@ -41,7 +41,7 @@ static u32 hash(struct bpf_bloom_filter *bloom, void *value,
 	return h & bloom->bitset_mask;
 }
 
-static int bloom_map_peek_elem(struct bpf_map *map, void *value)
+static long bloom_map_peek_elem(struct bpf_map *map, void *value)
 {
 	struct bpf_bloom_filter *bloom =
 		container_of(map, struct bpf_bloom_filter, map);
@@ -56,7 +56,7 @@ static int bloom_map_peek_elem(struct bpf_map *map, void *value)
 	return 0;
 }
 
-static int bloom_map_push_elem(struct bpf_map *map, void *value, u64 flags)
+static long bloom_map_push_elem(struct bpf_map *map, void *value, u64 flags)
 {
 	struct bpf_bloom_filter *bloom =
 		container_of(map, struct bpf_bloom_filter, map);
@@ -73,12 +73,12 @@ static int bloom_map_push_elem(struct bpf_map *map, void *value, u64 flags)
 	return 0;
 }
 
-static int bloom_map_pop_elem(struct bpf_map *map, void *value)
+static long bloom_map_pop_elem(struct bpf_map *map, void *value)
 {
 	return -EOPNOTSUPP;
 }
 
-static int bloom_map_delete_elem(struct bpf_map *map, void *value)
+static long bloom_map_delete_elem(struct bpf_map *map, void *value)
 {
 	return -EOPNOTSUPP;
 }
@@ -177,8 +177,8 @@ static void *bloom_map_lookup_elem(struct bpf_map *map, void *key)
 	return ERR_PTR(-EINVAL);
 }
 
-static int bloom_map_update_elem(struct bpf_map *map, void *key,
-				 void *value, u64 flags)
+static long bloom_map_update_elem(struct bpf_map *map, void *key,
+				  void *value, u64 flags)
 {
 	/* The eBPF program should use map_push_elem instead */
 	return -EINVAL;
diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c
index c975cacdd16b..f5b016a5484d 100644
--- a/kernel/bpf/bpf_cgrp_storage.c
+++ b/kernel/bpf/bpf_cgrp_storage.c
@@ -93,8 +93,8 @@ static void *bpf_cgrp_storage_lookup_elem(struct bpf_map *map, void *key)
 	return sdata ? sdata->data : NULL;
 }
 
-static int bpf_cgrp_storage_update_elem(struct bpf_map *map, void *key,
-					  void *value, u64 map_flags)
+static long bpf_cgrp_storage_update_elem(struct bpf_map *map, void *key,
+					 void *value, u64 map_flags)
 {
 	struct bpf_local_storage_data *sdata;
 	struct cgroup *cgroup;
@@ -125,7 +125,7 @@ static int cgroup_storage_delete(struct cgroup *cgroup, struct bpf_map *map)
 	return 0;
 }
 
-static int bpf_cgrp_storage_delete_elem(struct bpf_map *map, void *key)
+static long bpf_cgrp_storage_delete_elem(struct bpf_map *map, void *key)
 {
 	struct cgroup *cgroup;
 	int err, fd;
diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c
index ad2ab0187e45..9a5f05151898 100644
--- a/kernel/bpf/bpf_inode_storage.c
+++ b/kernel/bpf/bpf_inode_storage.c
@@ -91,8 +91,8 @@ static void *bpf_fd_inode_storage_lookup_elem(struct bpf_map *map, void *key)
 	return sdata ? sdata->data : NULL;
 }
 
-static int bpf_fd_inode_storage_update_elem(struct bpf_map *map, void *key,
-					 void *value, u64 map_flags)
+static long bpf_fd_inode_storage_update_elem(struct bpf_map *map, void *key,
+					     void *value, u64 map_flags)
 {
 	struct bpf_local_storage_data *sdata;
 	struct file *f;
@@ -127,7 +127,7 @@ static int inode_storage_delete(struct inode *inode, struct bpf_map *map)
 	return 0;
 }
 
-static int bpf_fd_inode_storage_delete_elem(struct bpf_map *map, void *key)
+static long bpf_fd_inode_storage_delete_elem(struct bpf_map *map, void *key)
 {
 	struct file *f;
 	int fd, err;
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 38903fb52f98..ba7a94276e3b 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -349,8 +349,8 @@ int bpf_struct_ops_prepare_trampoline(struct bpf_tramp_links *tlinks,
 					   model, flags, tlinks, NULL);
 }
 
-static int bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
-					  void *value, u64 flags)
+static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
+					   void *value, u64 flags)
 {
 	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
 	const struct bpf_struct_ops *st_ops = st_map->st_ops;
@@ -524,7 +524,7 @@ unlock:
 	return err;
 }
 
-static int bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key)
+static long bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key)
 {
 	enum bpf_struct_ops_state prev_state;
 	struct bpf_struct_ops_map *st_map;
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index c88cc04c17c1..ab5bd1ef58c4 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -120,8 +120,8 @@ out:
 	return ERR_PTR(err);
 }
 
-static int bpf_pid_task_storage_update_elem(struct bpf_map *map, void *key,
-					    void *value, u64 map_flags)
+static long bpf_pid_task_storage_update_elem(struct bpf_map *map, void *key,
+					     void *value, u64 map_flags)
 {
 	struct bpf_local_storage_data *sdata;
 	struct task_struct *task;
@@ -173,7 +173,7 @@ static int task_storage_delete(struct task_struct *task, struct bpf_map *map,
 	return 0;
 }
 
-static int bpf_pid_task_storage_delete_elem(struct bpf_map *map, void *key)
+static long bpf_pid_task_storage_delete_elem(struct bpf_map *map, void *key)
 {
 	struct task_struct *task;
 	unsigned int f_flags;
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 871809e71b4e..8ec18faa74ac 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -540,7 +540,7 @@ static void __cpu_map_entry_replace(struct bpf_cpu_map *cmap,
 	}
 }
 
-static int cpu_map_delete_elem(struct bpf_map *map, void *key)
+static long cpu_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
 	u32 key_cpu = *(u32 *)key;
@@ -553,8 +553,8 @@ static int cpu_map_delete_elem(struct bpf_map *map, void *key)
 	return 0;
 }
 
-static int cpu_map_update_elem(struct bpf_map *map, void *key, void *value,
-			       u64 map_flags)
+static long cpu_map_update_elem(struct bpf_map *map, void *key, void *value,
+				u64 map_flags)
 {
 	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
 	struct bpf_cpumap_val cpumap_value = {};
@@ -667,7 +667,7 @@ static int cpu_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 	return 0;
 }
 
-static int cpu_map_redirect(struct bpf_map *map, u64 index, u64 flags)
+static long cpu_map_redirect(struct bpf_map *map, u64 index, u64 flags)
 {
 	return __bpf_xdp_redirect_map(map, index, flags, 0,
 				      __cpu_map_lookup_elem);
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 19b036a228f7..802692fa3905 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -809,7 +809,7 @@ static void __dev_map_entry_free(struct rcu_head *rcu)
 	kfree(dev);
 }
 
-static int dev_map_delete_elem(struct bpf_map *map, void *key)
+static long dev_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *old_dev;
@@ -826,7 +826,7 @@ static int dev_map_delete_elem(struct bpf_map *map, void *key)
 	return 0;
 }
 
-static int dev_map_hash_delete_elem(struct bpf_map *map, void *key)
+static long dev_map_hash_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *old_dev;
@@ -897,8 +897,8 @@ err_out:
 	return ERR_PTR(-EINVAL);
 }
 
-static int __dev_map_update_elem(struct net *net, struct bpf_map *map,
-				 void *key, void *value, u64 map_flags)
+static long __dev_map_update_elem(struct net *net, struct bpf_map *map,
+				  void *key, void *value, u64 map_flags)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *dev, *old_dev;
@@ -939,15 +939,15 @@ static int __dev_map_update_elem(struct net *net, struct bpf_map *map,
 	return 0;
 }
 
-static int dev_map_update_elem(struct bpf_map *map, void *key, void *value,
-			       u64 map_flags)
+static long dev_map_update_elem(struct bpf_map *map, void *key, void *value,
+				u64 map_flags)
 {
 	return __dev_map_update_elem(current->nsproxy->net_ns,
 				     map, key, value, map_flags);
 }
 
-static int __dev_map_hash_update_elem(struct net *net, struct bpf_map *map,
-				     void *key, void *value, u64 map_flags)
+static long __dev_map_hash_update_elem(struct net *net, struct bpf_map *map,
+				       void *key, void *value, u64 map_flags)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *dev, *old_dev;
@@ -999,21 +999,21 @@ out_err:
 	return err;
 }
 
-static int dev_map_hash_update_elem(struct bpf_map *map, void *key, void *value,
-				   u64 map_flags)
+static long dev_map_hash_update_elem(struct bpf_map *map, void *key, void *value,
+				     u64 map_flags)
 {
 	return __dev_map_hash_update_elem(current->nsproxy->net_ns,
 					 map, key, value, map_flags);
 }
 
-static int dev_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags)
+static long dev_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags)
 {
 	return __bpf_xdp_redirect_map(map, ifindex, flags,
 				      BPF_F_BROADCAST | BPF_F_EXCLUDE_INGRESS,
 				      __dev_map_lookup_elem);
 }
 
-static int dev_hash_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags)
+static long dev_hash_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags)
 {
 	return __bpf_xdp_redirect_map(map, ifindex, flags,
 				      BPF_F_BROADCAST | BPF_F_EXCLUDE_INGRESS,
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 0df4b0c10f59..96b645bba3a4 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -1073,8 +1073,8 @@ static int check_flags(struct bpf_htab *htab, struct htab_elem *l_old,
 }
 
 /* Called from syscall or from eBPF program */
-static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
-				u64 map_flags)
+static long htab_map_update_elem(struct bpf_map *map, void *key, void *value,
+				 u64 map_flags)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
@@ -1175,8 +1175,8 @@ static void htab_lru_push_free(struct bpf_htab *htab, struct htab_elem *elem)
 	bpf_lru_push_free(&htab->lru, &elem->lru_node);
 }
 
-static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value,
-				    u64 map_flags)
+static long htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value,
+				     u64 map_flags)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new, *l_old = NULL;
@@ -1242,9 +1242,9 @@ err:
 	return ret;
 }
 
-static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key,
-					 void *value, u64 map_flags,
-					 bool onallcpus)
+static long __htab_percpu_map_update_elem(struct bpf_map *map, void *key,
+					  void *value, u64 map_flags,
+					  bool onallcpus)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
@@ -1297,9 +1297,9 @@ err:
 	return ret;
 }
 
-static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
-					     void *value, u64 map_flags,
-					     bool onallcpus)
+static long __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
+					      void *value, u64 map_flags,
+					      bool onallcpus)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
@@ -1364,21 +1364,21 @@ err:
 	return ret;
 }
 
-static int htab_percpu_map_update_elem(struct bpf_map *map, void *key,
-				       void *value, u64 map_flags)
+static long htab_percpu_map_update_elem(struct bpf_map *map, void *key,
+					void *value, u64 map_flags)
 {
 	return __htab_percpu_map_update_elem(map, key, value, map_flags, false);
 }
 
-static int htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
-					   void *value, u64 map_flags)
+static long htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
+					    void *value, u64 map_flags)
 {
 	return __htab_lru_percpu_map_update_elem(map, key, value, map_flags,
 						 false);
 }
 
 /* Called from syscall or from eBPF program */
-static int htab_map_delete_elem(struct bpf_map *map, void *key)
+static long htab_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct hlist_nulls_head *head;
@@ -1414,7 +1414,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key)
 	return ret;
 }
 
-static int htab_lru_map_delete_elem(struct bpf_map *map, void *key)
+static long htab_lru_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct hlist_nulls_head *head;
@@ -2134,8 +2134,8 @@ static const struct bpf_iter_seq_info iter_seq_info = {
 	.seq_priv_size		= sizeof(struct bpf_iter_seq_hash_map_info),
 };
 
-static int bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_fn,
-				  void *callback_ctx, u64 flags)
+static long bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_fn,
+				   void *callback_ctx, u64 flags)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct hlist_nulls_head *head;
diff --git a/kernel/bpf/local_storage.c b/kernel/bpf/local_storage.c
index a993560f200a..4c7bbec4a9e4 100644
--- a/kernel/bpf/local_storage.c
+++ b/kernel/bpf/local_storage.c
@@ -141,8 +141,8 @@ static void *cgroup_storage_lookup_elem(struct bpf_map *_map, void *key)
 	return &READ_ONCE(storage->buf)->data[0];
 }
 
-static int cgroup_storage_update_elem(struct bpf_map *map, void *key,
-				      void *value, u64 flags)
+static long cgroup_storage_update_elem(struct bpf_map *map, void *key,
+				       void *value, u64 flags)
 {
 	struct bpf_cgroup_storage *storage;
 	struct bpf_storage_buffer *new;
@@ -348,7 +348,7 @@ static void cgroup_storage_map_free(struct bpf_map *_map)
 	bpf_map_area_free(map);
 }
 
-static int cgroup_storage_delete_elem(struct bpf_map *map, void *key)
+static long cgroup_storage_delete_elem(struct bpf_map *map, void *key)
 {
 	return -EINVAL;
 }
diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c
index dc23f2ac9cde..e0d3ddf2037a 100644
--- a/kernel/bpf/lpm_trie.c
+++ b/kernel/bpf/lpm_trie.c
@@ -300,8 +300,8 @@ static struct lpm_trie_node *lpm_trie_node_alloc(const struct lpm_trie *trie,
 }
 
 /* Called from syscall or from eBPF program */
-static int trie_update_elem(struct bpf_map *map,
-			    void *_key, void *value, u64 flags)
+static long trie_update_elem(struct bpf_map *map,
+			     void *_key, void *value, u64 flags)
 {
 	struct lpm_trie *trie = container_of(map, struct lpm_trie, map);
 	struct lpm_trie_node *node, *im_node = NULL, *new_node = NULL;
@@ -431,7 +431,7 @@ out:
 }
 
 /* Called from syscall or from eBPF program */
-static int trie_delete_elem(struct bpf_map *map, void *_key)
+static long trie_delete_elem(struct bpf_map *map, void *_key)
 {
 	struct lpm_trie *trie = container_of(map, struct lpm_trie, map);
 	struct bpf_lpm_trie_key *key = _key;
diff --git a/kernel/bpf/queue_stack_maps.c b/kernel/bpf/queue_stack_maps.c
index 63ecbbcb349d..601609164ef3 100644
--- a/kernel/bpf/queue_stack_maps.c
+++ b/kernel/bpf/queue_stack_maps.c
@@ -95,7 +95,7 @@ static void queue_stack_map_free(struct bpf_map *map)
 	bpf_map_area_free(qs);
 }
 
-static int __queue_map_get(struct bpf_map *map, void *value, bool delete)
+static long __queue_map_get(struct bpf_map *map, void *value, bool delete)
 {
 	struct bpf_queue_stack *qs = bpf_queue_stack(map);
 	unsigned long flags;
@@ -124,7 +124,7 @@ out:
 }
 
 
-static int __stack_map_get(struct bpf_map *map, void *value, bool delete)
+static long __stack_map_get(struct bpf_map *map, void *value, bool delete)
 {
 	struct bpf_queue_stack *qs = bpf_queue_stack(map);
 	unsigned long flags;
@@ -156,32 +156,32 @@ out:
 }
 
 /* Called from syscall or from eBPF program */
-static int queue_map_peek_elem(struct bpf_map *map, void *value)
+static long queue_map_peek_elem(struct bpf_map *map, void *value)
 {
 	return __queue_map_get(map, value, false);
 }
 
 /* Called from syscall or from eBPF program */
-static int stack_map_peek_elem(struct bpf_map *map, void *value)
+static long stack_map_peek_elem(struct bpf_map *map, void *value)
 {
 	return __stack_map_get(map, value, false);
 }
 
 /* Called from syscall or from eBPF program */
-static int queue_map_pop_elem(struct bpf_map *map, void *value)
+static long queue_map_pop_elem(struct bpf_map *map, void *value)
 {
 	return __queue_map_get(map, value, true);
 }
 
 /* Called from syscall or from eBPF program */
-static int stack_map_pop_elem(struct bpf_map *map, void *value)
+static long stack_map_pop_elem(struct bpf_map *map, void *value)
 {
 	return __stack_map_get(map, value, true);
 }
 
 /* Called from syscall or from eBPF program */
-static int queue_stack_map_push_elem(struct bpf_map *map, void *value,
-				     u64 flags)
+static long queue_stack_map_push_elem(struct bpf_map *map, void *value,
+				      u64 flags)
 {
 	struct bpf_queue_stack *qs = bpf_queue_stack(map);
 	unsigned long irq_flags;
@@ -227,14 +227,14 @@ static void *queue_stack_map_lookup_elem(struct bpf_map *map, void *key)
 }
 
 /* Called from syscall or from eBPF program */
-static int queue_stack_map_update_elem(struct bpf_map *map, void *key,
-				       void *value, u64 flags)
+static long queue_stack_map_update_elem(struct bpf_map *map, void *key,
+					void *value, u64 flags)
 {
 	return -EINVAL;
 }
 
 /* Called from syscall or from eBPF program */
-static int queue_stack_map_delete_elem(struct bpf_map *map, void *key)
+static long queue_stack_map_delete_elem(struct bpf_map *map, void *key)
 {
 	return -EINVAL;
 }
diff --git a/kernel/bpf/reuseport_array.c b/kernel/bpf/reuseport_array.c
index 71cb72f5b733..cbf2d8d784b8 100644
--- a/kernel/bpf/reuseport_array.c
+++ b/kernel/bpf/reuseport_array.c
@@ -59,7 +59,7 @@ static void *reuseport_array_lookup_elem(struct bpf_map *map, void *key)
 }
 
 /* Called from syscall only */
-static int reuseport_array_delete_elem(struct bpf_map *map, void *key)
+static long reuseport_array_delete_elem(struct bpf_map *map, void *key)
 {
 	struct reuseport_array *array = reuseport_array(map);
 	u32 index = *(u32 *)key;
diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c
index 0d2a45ff83f1..875ac9b698d9 100644
--- a/kernel/bpf/ringbuf.c
+++ b/kernel/bpf/ringbuf.c
@@ -242,13 +242,13 @@ static void *ringbuf_map_lookup_elem(struct bpf_map *map, void *key)
 	return ERR_PTR(-ENOTSUPP);
 }
 
-static int ringbuf_map_update_elem(struct bpf_map *map, void *key, void *value,
-				   u64 flags)
+static long ringbuf_map_update_elem(struct bpf_map *map, void *key, void *value,
+				    u64 flags)
 {
 	return -ENOTSUPP;
 }
 
-static int ringbuf_map_delete_elem(struct bpf_map *map, void *key)
+static long ringbuf_map_delete_elem(struct bpf_map *map, void *key)
 {
 	return -ENOTSUPP;
 }
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 0f1d8dced933..b25fce425b2c 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -618,14 +618,14 @@ static int stack_map_get_next_key(struct bpf_map *map, void *key,
 	return 0;
 }
 
-static int stack_map_update_elem(struct bpf_map *map, void *key, void *value,
-				 u64 map_flags)
+static long stack_map_update_elem(struct bpf_map *map, void *key, void *value,
+				  u64 map_flags)
 {
 	return -EINVAL;
 }
 
 /* Called from syscall or from eBPF program */
-static int stack_map_delete_elem(struct bpf_map *map, void *key)
+static long stack_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
 	struct stack_map_bucket *old_bucket;
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5693e4a92752..50c995697f0e 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -17692,21 +17692,21 @@ static int do_misc_fixups(struct bpf_verifier_env *env)
 			BUILD_BUG_ON(!__same_type(ops->map_lookup_elem,
 				     (void *(*)(struct bpf_map *map, void *key))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_delete_elem,
-				     (int (*)(struct bpf_map *map, void *key))NULL));
+				     (long (*)(struct bpf_map *map, void *key))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_update_elem,
-				     (int (*)(struct bpf_map *map, void *key, void *value,
+				     (long (*)(struct bpf_map *map, void *key, void *value,
 					      u64 flags))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_push_elem,
-				     (int (*)(struct bpf_map *map, void *value,
+				     (long (*)(struct bpf_map *map, void *value,
 					      u64 flags))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_pop_elem,
-				     (int (*)(struct bpf_map *map, void *value))NULL));
+				     (long (*)(struct bpf_map *map, void *value))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_peek_elem,
-				     (int (*)(struct bpf_map *map, void *value))NULL));
+				     (long (*)(struct bpf_map *map, void *value))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_redirect,
-				     (int (*)(struct bpf_map *map, u64 index, u64 flags))NULL));
+				     (long (*)(struct bpf_map *map, u64 index, u64 flags))NULL));
 			BUILD_BUG_ON(!__same_type(ops->map_for_each_callback,
-				     (int (*)(struct bpf_map *map,
+				     (long (*)(struct bpf_map *map,
 					      bpf_callback_t callback_fn,
 					      void *callback_ctx,
 					      u64 flags))NULL));
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index 24c3dc0d62e5..cb0f5a105b89 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -94,8 +94,8 @@ static void *bpf_fd_sk_storage_lookup_elem(struct bpf_map *map, void *key)
 	return ERR_PTR(err);
 }
 
-static int bpf_fd_sk_storage_update_elem(struct bpf_map *map, void *key,
-					 void *value, u64 map_flags)
+static long bpf_fd_sk_storage_update_elem(struct bpf_map *map, void *key,
+					  void *value, u64 map_flags)
 {
 	struct bpf_local_storage_data *sdata;
 	struct socket *sock;
@@ -114,7 +114,7 @@ static int bpf_fd_sk_storage_update_elem(struct bpf_map *map, void *key,
 	return err;
 }
 
-static int bpf_fd_sk_storage_delete_elem(struct bpf_map *map, void *key)
+static long bpf_fd_sk_storage_delete_elem(struct bpf_map *map, void *key)
 {
 	struct socket *sock;
 	int fd, err;
diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index 9b854e236d23..7c189c2e2fbf 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -437,7 +437,7 @@ static void sock_map_delete_from_link(struct bpf_map *map, struct sock *sk,
 	__sock_map_delete(stab, sk, link_raw);
 }
 
-static int sock_map_delete_elem(struct bpf_map *map, void *key)
+static long sock_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
 	u32 i = *(u32 *)key;
@@ -587,8 +587,8 @@ out:
 	return ret;
 }
 
-static int sock_map_update_elem(struct bpf_map *map, void *key,
-				void *value, u64 flags)
+static long sock_map_update_elem(struct bpf_map *map, void *key,
+				 void *value, u64 flags)
 {
 	struct sock *sk = (struct sock *)value;
 	int ret;
@@ -925,7 +925,7 @@ static void sock_hash_delete_from_link(struct bpf_map *map, struct sock *sk,
 	raw_spin_unlock_bh(&bucket->lock);
 }
 
-static int sock_hash_delete_elem(struct bpf_map *map, void *key)
+static long sock_hash_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_shtab *htab = container_of(map, struct bpf_shtab, map);
 	u32 hash, key_size = map->key_size;
diff --git a/net/xdp/xskmap.c b/net/xdp/xskmap.c
index 0c38d7175922..2c1427074a3b 100644
--- a/net/xdp/xskmap.c
+++ b/net/xdp/xskmap.c
@@ -162,8 +162,8 @@ static void *xsk_map_lookup_elem_sys_only(struct bpf_map *map, void *key)
 	return ERR_PTR(-EOPNOTSUPP);
 }
 
-static int xsk_map_update_elem(struct bpf_map *map, void *key, void *value,
-			       u64 map_flags)
+static long xsk_map_update_elem(struct bpf_map *map, void *key, void *value,
+				u64 map_flags)
 {
 	struct xsk_map *m = container_of(map, struct xsk_map, map);
 	struct xdp_sock __rcu **map_entry;
@@ -223,7 +223,7 @@ out:
 	return err;
 }
 
-static int xsk_map_delete_elem(struct bpf_map *map, void *key)
+static long xsk_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct xsk_map *m = container_of(map, struct xsk_map, map);
 	struct xdp_sock __rcu **map_entry;
@@ -243,7 +243,7 @@ static int xsk_map_delete_elem(struct bpf_map *map, void *key)
 	return 0;
 }
 
-static int xsk_map_redirect(struct bpf_map *map, u64 index, u64 flags)
+static long xsk_map_redirect(struct bpf_map *map, u64 index, u64 flags)
 {
 	return __bpf_xdp_redirect_map(map, index, flags, 0,
 				      __xsk_map_lookup_elem);
-- 
cgit v1.2.3-70-g09d2


From 7be14c1c9030f73cc18b4ff23b78a0a081f16188 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 22 Mar 2023 22:30:55 +0100
Subject: bpf: Fix __reg_bound_offset 64->32 var_off subreg propagation

Xu reports that after commit 3f50f132d840 ("bpf: Verifier, do explicit ALU32
bounds tracking"), the following BPF program is rejected by the verifier:

   0: (61) r2 = *(u32 *)(r1 +0)          ; R2_w=pkt(off=0,r=0,imm=0)
   1: (61) r3 = *(u32 *)(r1 +4)          ; R3_w=pkt_end(off=0,imm=0)
   2: (bf) r1 = r2
   3: (07) r1 += 1
   4: (2d) if r1 > r3 goto pc+8
   5: (71) r1 = *(u8 *)(r2 +0)           ; R1_w=scalar(umax=255,var_off=(0x0; 0xff))
   6: (18) r0 = 0x7fffffffffffff10
   8: (0f) r1 += r0                      ; R1_w=scalar(umin=0x7fffffffffffff10,umax=0x800000000000000f)
   9: (18) r0 = 0x8000000000000000
  11: (07) r0 += 1
  12: (ad) if r0 < r1 goto pc-2
  13: (b7) r0 = 0
  14: (95) exit

And the verifier log says:

  func#0 @0
  0: R1=ctx(off=0,imm=0) R10=fp0
  0: (61) r2 = *(u32 *)(r1 +0)          ; R1=ctx(off=0,imm=0) R2_w=pkt(off=0,r=0,imm=0)
  1: (61) r3 = *(u32 *)(r1 +4)          ; R1=ctx(off=0,imm=0) R3_w=pkt_end(off=0,imm=0)
  2: (bf) r1 = r2                       ; R1_w=pkt(off=0,r=0,imm=0) R2_w=pkt(off=0,r=0,imm=0)
  3: (07) r1 += 1                       ; R1_w=pkt(off=1,r=0,imm=0)
  4: (2d) if r1 > r3 goto pc+8          ; R1_w=pkt(off=1,r=1,imm=0) R3_w=pkt_end(off=0,imm=0)
  5: (71) r1 = *(u8 *)(r2 +0)           ; R1_w=scalar(umax=255,var_off=(0x0; 0xff)) R2_w=pkt(off=0,r=1,imm=0)
  6: (18) r0 = 0x7fffffffffffff10       ; R0_w=9223372036854775568
  8: (0f) r1 += r0                      ; R0_w=9223372036854775568 R1_w=scalar(umin=9223372036854775568,umax=9223372036854775823,s32_min=-240,s32_max=15)
  9: (18) r0 = 0x8000000000000000       ; R0_w=-9223372036854775808
  11: (07) r0 += 1                      ; R0_w=-9223372036854775807
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775807 R1_w=scalar(umin=9223372036854775568,umax=9223372036854775809)
  13: (b7) r0 = 0                       ; R0_w=0
  14: (95) exit

  from 12 to 11: R0_w=-9223372036854775807 R1_w=scalar(umin=9223372036854775810,umax=9223372036854775823,var_off=(0x8000000000000000; 0xffffffff)) R2_w=pkt(off=0,r=1,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775806
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775806 R1_w=scalar(umin=9223372036854775810,umax=9223372036854775810,var_off=(0x8000000000000000; 0xffffffff))
  13: safe

  [...]

  from 12 to 11: R0_w=-9223372036854775795 R1=scalar(umin=9223372036854775822,umax=9223372036854775823,var_off=(0x8000000000000000; 0xffffffff)) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775794
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775794 R1=scalar(umin=9223372036854775822,umax=9223372036854775822,var_off=(0x8000000000000000; 0xffffffff))
  13: safe

  from 12 to 11: R0_w=-9223372036854775794 R1=scalar(umin=9223372036854775823,umax=9223372036854775823,var_off=(0x8000000000000000; 0xffffffff)) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775793
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775793 R1=scalar(umin=9223372036854775823,umax=9223372036854775823,var_off=(0x8000000000000000; 0xffffffff))
  13: safe

  from 12 to 11: R0_w=-9223372036854775793 R1=scalar(umin=9223372036854775824,umax=9223372036854775823,var_off=(0x8000000000000000; 0xffffffff)) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775792
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775792 R1=scalar(umin=9223372036854775824,umax=9223372036854775823,var_off=(0x8000000000000000; 0xffffffff))
  13: safe

  [...]

The 64bit umin=9223372036854775810 bound continuously bumps by +1 while
umax=9223372036854775823 stays as-is until the verifier complexity limit
is reached and the program gets finally rejected. During this simulation,
the umin also eventually surpasses umax. Looking at the first 'from 12
to 11' output line from the loop, R1 has the following state:

  R1_w=scalar(umin=0x8000000000000002 (9223372036854775810),
              umax=0x800000000000000f (9223372036854775823),
          var_off=(0x8000000000000000;
                           0xffffffff))

The var_off has technically not an inconsistent state but it's very
imprecise and far off surpassing 64bit umax bounds whereas the expected
output with refined known bits in var_off should have been like:

  R1_w=scalar(umin=0x8000000000000002 (9223372036854775810),
              umax=0x800000000000000f (9223372036854775823),
          var_off=(0x8000000000000000;
                                  0xf))

In the above log, var_off stays as var_off=(0x8000000000000000; 0xffffffff)
and does not converge into a narrower mask where more bits become known,
eventually transforming R1 into a constant upon umin=9223372036854775823,
umax=9223372036854775823 case where the verifier would have terminated and
let the program pass.

The __reg_combine_64_into_32() marks the subregister unknown and propagates
64bit {s,u}min/{s,u}max bounds to their 32bit equivalents iff they are within
the 32bit universe. The question came up whether __reg_combine_64_into_32()
should special case the situation that when 64bit {s,u}min bounds have
the same value as 64bit {s,u}max bounds to then assign the latter as
well to the 32bit reg->{s,u}32_{min,max}_value. As can be seen from the
above example however, that is just /one/ special case and not a /generic/
solution given above example would still not be addressed this way and
remain at an imprecise var_off=(0x8000000000000000; 0xffffffff).

The improvement is needed in __reg_bound_offset() to refine var32_off with
the updated var64_off instead of the prior reg->var_off. The reg_bounds_sync()
code first refines information about the register's min/max bounds via
__update_reg_bounds() from the current var_off, then in __reg_deduce_bounds()
from sign bit and with the potentially learned bits from bounds it'll
update the var_off tnum in __reg_bound_offset(). For example, intersecting
with the old var_off might have improved bounds slightly, e.g. if umax
was 0x7f...f and var_off was (0; 0xf...fc), then new var_off will then
result in (0; 0x7f...fc). The intersected var64_off holds then the
universe which is a superset of var32_off. The point for the latter is
not to broaden, but to further refine known bits based on the intersection
of var_off with 32 bit bounds, so that we later construct the final var_off
from upper and lower 32 bits. The final __update_reg_bounds() can then
potentially still slightly refine bounds if more bits became known from the
new var_off.

After the improvement, we can see R1 converging successively:

  func#0 @0
  0: R1=ctx(off=0,imm=0) R10=fp0
  0: (61) r2 = *(u32 *)(r1 +0)          ; R1=ctx(off=0,imm=0) R2_w=pkt(off=0,r=0,imm=0)
  1: (61) r3 = *(u32 *)(r1 +4)          ; R1=ctx(off=0,imm=0) R3_w=pkt_end(off=0,imm=0)
  2: (bf) r1 = r2                       ; R1_w=pkt(off=0,r=0,imm=0) R2_w=pkt(off=0,r=0,imm=0)
  3: (07) r1 += 1                       ; R1_w=pkt(off=1,r=0,imm=0)
  4: (2d) if r1 > r3 goto pc+8          ; R1_w=pkt(off=1,r=1,imm=0) R3_w=pkt_end(off=0,imm=0)
  5: (71) r1 = *(u8 *)(r2 +0)           ; R1_w=scalar(umax=255,var_off=(0x0; 0xff)) R2_w=pkt(off=0,r=1,imm=0)
  6: (18) r0 = 0x7fffffffffffff10       ; R0_w=9223372036854775568
  8: (0f) r1 += r0                      ; R0_w=9223372036854775568 R1_w=scalar(umin=9223372036854775568,umax=9223372036854775823,s32_min=-240,s32_max=15)
  9: (18) r0 = 0x8000000000000000       ; R0_w=-9223372036854775808
  11: (07) r0 += 1                      ; R0_w=-9223372036854775807
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775807 R1_w=scalar(umin=9223372036854775568,umax=9223372036854775809)
  13: (b7) r0 = 0                       ; R0_w=0
  14: (95) exit

  from 12 to 11: R0_w=-9223372036854775807 R1_w=scalar(umin=9223372036854775810,umax=9223372036854775823,var_off=(0x8000000000000000; 0xf),s32_min=0,s32_max=15,u32_max=15) R2_w=pkt(off=0,r=1,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775806
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775806 R1_w=-9223372036854775806
  13: safe

  from 12 to 11: R0_w=-9223372036854775806 R1_w=scalar(umin=9223372036854775811,umax=9223372036854775823,var_off=(0x8000000000000000; 0xf),s32_min=0,s32_max=15,u32_max=15) R2_w=pkt(off=0,r=1,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775805
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775805 R1_w=-9223372036854775805
  13: safe

  [...]

  from 12 to 11: R0_w=-9223372036854775798 R1=scalar(umin=9223372036854775819,umax=9223372036854775823,var_off=(0x8000000000000008; 0x7),s32_min=8,s32_max=15,u32_min=8,u32_max=15) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775797
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775797 R1=-9223372036854775797
  13: safe

  from 12 to 11: R0_w=-9223372036854775797 R1=scalar(umin=9223372036854775820,umax=9223372036854775823,var_off=(0x800000000000000c; 0x3),s32_min=12,s32_max=15,u32_min=12,u32_max=15) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775796
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775796 R1=-9223372036854775796
  13: safe

  from 12 to 11: R0_w=-9223372036854775796 R1=scalar(umin=9223372036854775821,umax=9223372036854775823,var_off=(0x800000000000000c; 0x3),s32_min=12,s32_max=15,u32_min=12,u32_max=15) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775795
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775795 R1=-9223372036854775795
  13: safe

  from 12 to 11: R0_w=-9223372036854775795 R1=scalar(umin=9223372036854775822,umax=9223372036854775823,var_off=(0x800000000000000e; 0x1),s32_min=14,s32_max=15,u32_min=14,u32_max=15) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775794
  12: (ad) if r0 < r1 goto pc-2         ; R0_w=-9223372036854775794 R1=-9223372036854775794
  13: safe

  from 12 to 11: R0_w=-9223372036854775794 R1=-9223372036854775793 R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  11: (07) r0 += 1                      ; R0_w=-9223372036854775793
  12: (ad) if r0 < r1 goto pc-2
  last_idx 12 first_idx 12
  parent didn't have regs=1 stack=0 marks: R0_rw=P-9223372036854775801 R1_r=scalar(umin=9223372036854775815,umax=9223372036854775823,var_off=(0x8000000000000000; 0xf),s32_min=0,s32_max=15,u32_max=15) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  last_idx 11 first_idx 11
  regs=1 stack=0 before 11: (07) r0 += 1
  parent didn't have regs=1 stack=0 marks: R0_rw=P-9223372036854775805 R1_rw=scalar(umin=9223372036854775812,umax=9223372036854775823,var_off=(0x8000000000000000; 0xf),s32_min=0,s32_max=15,u32_max=15) R2_w=pkt(off=0,r=1,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
  last_idx 12 first_idx 0
  regs=1 stack=0 before 12: (ad) if r0 < r1 goto pc-2
  regs=1 stack=0 before 11: (07) r0 += 1
  regs=1 stack=0 before 12: (ad) if r0 < r1 goto pc-2
  regs=1 stack=0 before 11: (07) r0 += 1
  regs=1 stack=0 before 12: (ad) if r0 < r1 goto pc-2
  regs=1 stack=0 before 11: (07) r0 += 1
  regs=1 stack=0 before 9: (18) r0 = 0x8000000000000000
  last_idx 12 first_idx 12
  parent didn't have regs=2 stack=0 marks: R0_rw=P-9223372036854775801 R1_r=Pscalar(umin=9223372036854775815,umax=9223372036854775823,var_off=(0x8000000000000000; 0xf),s32_min=0,s32_max=15,u32_max=15) R2=pkt(off=0,r=1,imm=0) R3=pkt_end(off=0,imm=0) R10=fp0
  last_idx 11 first_idx 11
  regs=2 stack=0 before 11: (07) r0 += 1
  parent didn't have regs=2 stack=0 marks: R0_rw=P-9223372036854775805 R1_rw=Pscalar(umin=9223372036854775812,umax=9223372036854775823,var_off=(0x8000000000000000; 0xf),s32_min=0,s32_max=15,u32_max=15) R2_w=pkt(off=0,r=1,imm=0) R3_w=pkt_end(off=0,imm=0) R10=fp0
  last_idx 12 first_idx 0
  regs=2 stack=0 before 12: (ad) if r0 < r1 goto pc-2
  regs=2 stack=0 before 11: (07) r0 += 1
  regs=2 stack=0 before 12: (ad) if r0 < r1 goto pc-2
  regs=2 stack=0 before 11: (07) r0 += 1
  regs=2 stack=0 before 12: (ad) if r0 < r1 goto pc-2
  regs=2 stack=0 before 11: (07) r0 += 1
  regs=2 stack=0 before 9: (18) r0 = 0x8000000000000000
  regs=2 stack=0 before 8: (0f) r1 += r0
  regs=3 stack=0 before 6: (18) r0 = 0x7fffffffffffff10
  regs=2 stack=0 before 5: (71) r1 = *(u8 *)(r2 +0)
  13: safe

  from 4 to 13: safe
  verification time 322 usec
  stack depth 0
  processed 56 insns (limit 1000000) max_states_per_insn 1 total_states 3 peak_states 3 mark_read 1

This also fixes up a test case along with this improvement where we match
on the verifier log. The updated log now has a refined var_off, too.

Fixes: 3f50f132d840 ("bpf: Verifier, do explicit ALU32 bounds tracking")
Reported-by: Xu Kuohai <xukuohai@huaweicloud.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20230314203424.4015351-2-xukuohai@huaweicloud.com
Link: https://lore.kernel.org/bpf/20230322213056.2470-1-daniel@iogearbox.net
---
 kernel/bpf/verifier.c                          | 6 +++---
 tools/testing/selftests/bpf/prog_tests/align.c | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 50c995697f0e..fd2f216de920 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -2149,9 +2149,9 @@ static void __reg_bound_offset(struct bpf_reg_state *reg)
 	struct tnum var64_off = tnum_intersect(reg->var_off,
 					       tnum_range(reg->umin_value,
 							  reg->umax_value));
-	struct tnum var32_off = tnum_intersect(tnum_subreg(reg->var_off),
-						tnum_range(reg->u32_min_value,
-							   reg->u32_max_value));
+	struct tnum var32_off = tnum_intersect(tnum_subreg(var64_off),
+					       tnum_range(reg->u32_min_value,
+							  reg->u32_max_value));
 
 	reg->var_off = tnum_or(tnum_clear_subreg(var64_off), var32_off);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/align.c b/tools/testing/selftests/bpf/prog_tests/align.c
index c94fa8d6c4f6..b92770592563 100644
--- a/tools/testing/selftests/bpf/prog_tests/align.c
+++ b/tools/testing/selftests/bpf/prog_tests/align.c
@@ -575,14 +575,14 @@ static struct bpf_align_test tests[] = {
 			/* New unknown value in R7 is (4n), >= 76 */
 			{14, "R7_w=scalar(umin=76,umax=1096,var_off=(0x0; 0x7fc))"},
 			/* Adding it to packet pointer gives nice bounds again */
-			{16, "R5_w=pkt(id=3,off=0,r=0,umin=2,umax=1082,var_off=(0x2; 0xfffffffc)"},
+			{16, "R5_w=pkt(id=3,off=0,r=0,umin=2,umax=1082,var_off=(0x2; 0x7fc)"},
 			/* At the time the word size load is performed from R5,
 			 * its total fixed offset is NET_IP_ALIGN + reg->off (0)
 			 * which is 2.  Then the variable offset is (4n+2), so
 			 * the total offset is 4-byte aligned and meets the
 			 * load's requirements.
 			 */
-			{20, "R5=pkt(id=3,off=0,r=4,umin=2,umax=1082,var_off=(0x2; 0xfffffffc)"},
+			{20, "R5=pkt(id=3,off=0,r=4,umin=2,umax=1082,var_off=(0x2; 0x7fc)"},
 		},
 	},
 };
-- 
cgit v1.2.3-70-g09d2


From 1a3148fc171f5cde11b4c24e808a953ff725a3e2 Mon Sep 17 00:00:00 2001
From: Xu Kuohai <xukuohai@huawei.com>
Date: Wed, 22 Mar 2023 22:30:56 +0100
Subject: selftests/bpf: Check when bounds are not in the 32-bit range

Add cases to check if bound is updated correctly when 64-bit value is
not in the 32-bit range.

Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20230322213056.2470-2-daniel@iogearbox.net
---
 tools/testing/selftests/bpf/verifier/bounds.c | 121 ++++++++++++++++++++++++++
 1 file changed, 121 insertions(+)

diff --git a/tools/testing/selftests/bpf/verifier/bounds.c b/tools/testing/selftests/bpf/verifier/bounds.c
index 33125d5f6772..74b1917d4208 100644
--- a/tools/testing/selftests/bpf/verifier/bounds.c
+++ b/tools/testing/selftests/bpf/verifier/bounds.c
@@ -753,3 +753,124 @@
 	.result_unpriv = REJECT,
 	.result = ACCEPT,
 },
+{
+	"bound check with JMP_JLT for crossing 64-bit signed boundary",
+	.insns = {
+	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
+	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 8),
+
+	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
+	BPF_LD_IMM64(BPF_REG_0, 0x7fffffffffffff10),
+	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
+
+	BPF_LD_IMM64(BPF_REG_0, 0x8000000000000000),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
+	/* r1 unsigned range is [0x7fffffffffffff10, 0x800000000000000f] */
+	BPF_JMP_REG(BPF_JLT, BPF_REG_0, BPF_REG_1, -2),
+
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = ACCEPT,
+	.prog_type = BPF_PROG_TYPE_XDP,
+},
+{
+	"bound check with JMP_JSLT for crossing 64-bit signed boundary",
+	.insns = {
+	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
+	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 8),
+
+	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
+	BPF_LD_IMM64(BPF_REG_0, 0x7fffffffffffff10),
+	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
+
+	BPF_LD_IMM64(BPF_REG_0, 0x8000000000000000),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
+	/* r1 signed range is [S64_MIN, S64_MAX] */
+	BPF_JMP_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -2),
+
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.errstr = "BPF program is too large",
+	.result = REJECT,
+	.prog_type = BPF_PROG_TYPE_XDP,
+},
+{
+	"bound check for loop upper bound greater than U32_MAX",
+	.insns = {
+	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
+	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 8),
+
+	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
+	BPF_LD_IMM64(BPF_REG_0, 0x100000000),
+	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
+
+	BPF_LD_IMM64(BPF_REG_0, 0x100000000),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
+	BPF_JMP_REG(BPF_JLT, BPF_REG_0, BPF_REG_1, -2),
+
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = ACCEPT,
+	.prog_type = BPF_PROG_TYPE_XDP,
+},
+{
+	"bound check with JMP32_JLT for crossing 32-bit signed boundary",
+	.insns = {
+	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
+	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 6),
+
+	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
+	BPF_MOV32_IMM(BPF_REG_0, 0x7fffff10),
+	BPF_ALU32_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
+
+	BPF_MOV32_IMM(BPF_REG_0, 0x80000000),
+	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 1),
+	/* r1 unsigned range is [0, 0x8000000f] */
+	BPF_JMP32_REG(BPF_JLT, BPF_REG_0, BPF_REG_1, -2),
+
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = ACCEPT,
+	.prog_type = BPF_PROG_TYPE_XDP,
+},
+{
+	"bound check with JMP32_JSLT for crossing 32-bit signed boundary",
+	.insns = {
+	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
+	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 6),
+
+	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
+	BPF_MOV32_IMM(BPF_REG_0, 0x7fffff10),
+	BPF_ALU32_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
+
+	BPF_MOV32_IMM(BPF_REG_0, 0x80000000),
+	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 1),
+	/* r1 signed range is [S32_MIN, S32_MAX] */
+	BPF_JMP32_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -2),
+
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.errstr = "BPF program is too large",
+	.result = REJECT,
+	.prog_type = BPF_PROG_TYPE_XDP,
+},
-- 
cgit v1.2.3-70-g09d2


From b63cbc490e18d893632929b8faa55bb28da3fcd4 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 22 Mar 2023 16:25:02 -0700
Subject: bpf: remember meta->iter info only for initialized iters

For iter_new() functions iterator state's slot might not be yet
initialized, in which case iter_get_spi() will return -ERANGE. This is
expected and is handled properly. But for iter_next() and iter_destroy()
cases iter slot is supposed to be initialized and correct, so -ERANGE is
not possible.

Move meta->iter.{spi,frameno} initialization into iter_next/iter_destroy
handling branch to make it more explicit that valid information will be
remembered in meta->iter block for subsequent use in process_iter_next_call(),
avoiding confusingly looking -ERANGE assignment for meta->iter.spi.

Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230322232502.836171-1-andrii@kernel.org
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 kernel/bpf/verifier.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index fd2f216de920..64f06f6e16bf 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -6778,13 +6778,6 @@ static int process_iter_arg(struct bpf_verifier_env *env, int regno, int insn_id
 	t = btf_type_skip_modifiers(meta->btf, t->type, &btf_id);	/* STRUCT */
 	nr_slots = t->size / BPF_REG_SIZE;
 
-	spi = iter_get_spi(env, reg, nr_slots);
-	if (spi < 0 && spi != -ERANGE)
-		return spi;
-
-	meta->iter.spi = spi;
-	meta->iter.frameno = reg->frameno;
-
 	if (is_iter_new_kfunc(meta)) {
 		/* bpf_iter_<type>_new() expects pointer to uninit iter state */
 		if (!is_iter_reg_valid_uninit(env, reg, nr_slots)) {
@@ -6811,10 +6804,17 @@ static int process_iter_arg(struct bpf_verifier_env *env, int regno, int insn_id
 			return -EINVAL;
 		}
 
+		spi = iter_get_spi(env, reg, nr_slots);
+		if (spi < 0)
+			return spi;
+
 		err = mark_iter_read(env, reg, spi, nr_slots);
 		if (err)
 			return err;
 
+		/* remember meta->iter info for process_iter_next_call() */
+		meta->iter.spi = spi;
+		meta->iter.frameno = reg->frameno;
 		meta->ref_obj_id = iter_ref_obj_id(env, reg, spi);
 
 		if (is_iter_destroy_kfunc(meta)) {
-- 
cgit v1.2.3-70-g09d2


From a3a4e300439bde3517edd61d982d747000a9f212 Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Tue, 21 Mar 2023 07:44:47 -0700
Subject: bnxt: Change fw_cap to u64 to accommodate more capability bits

The current fw_cap field (u32) has run out of bits to save any
new capability.

Change the field to u64.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Edwin Peer <edwin.peer@broadcom.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Acked-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.h | 56 +++++++++++++++----------------
 1 file changed, 28 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index c0628ac1b798..d7eb0d244f42 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1968,34 +1968,34 @@ struct bnxt {
 
 	u32			msg_enable;
 
-	u32			fw_cap;
-	#define BNXT_FW_CAP_SHORT_CMD			0x00000001
-	#define BNXT_FW_CAP_LLDP_AGENT			0x00000002
-	#define BNXT_FW_CAP_DCBX_AGENT			0x00000004
-	#define BNXT_FW_CAP_NEW_RM			0x00000008
-	#define BNXT_FW_CAP_IF_CHANGE			0x00000010
-	#define BNXT_FW_CAP_KONG_MB_CHNL		0x00000080
-	#define BNXT_FW_CAP_OVS_64BIT_HANDLE		0x00000400
-	#define BNXT_FW_CAP_TRUSTED_VF			0x00000800
-	#define BNXT_FW_CAP_ERROR_RECOVERY		0x00002000
-	#define BNXT_FW_CAP_PKG_VER			0x00004000
-	#define BNXT_FW_CAP_CFA_ADV_FLOW		0x00008000
-	#define BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2	0x00010000
-	#define BNXT_FW_CAP_PCIE_STATS_SUPPORTED	0x00020000
-	#define BNXT_FW_CAP_EXT_STATS_SUPPORTED		0x00040000
-	#define BNXT_FW_CAP_RSS_HASH_TYPE_DELTA		0x00080000
-	#define BNXT_FW_CAP_ERR_RECOVER_RELOAD		0x00100000
-	#define BNXT_FW_CAP_HOT_RESET			0x00200000
-	#define BNXT_FW_CAP_PTP_RTC			0x00400000
-	#define BNXT_FW_CAP_RX_ALL_PKT_TS		0x00800000
-	#define BNXT_FW_CAP_VLAN_RX_STRIP		0x01000000
-	#define BNXT_FW_CAP_VLAN_TX_INSERT		0x02000000
-	#define BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED	0x04000000
-	#define BNXT_FW_CAP_LIVEPATCH			0x08000000
-	#define BNXT_FW_CAP_PTP_PPS			0x10000000
-	#define BNXT_FW_CAP_HOT_RESET_IF		0x20000000
-	#define BNXT_FW_CAP_RING_MONITOR		0x40000000
-	#define BNXT_FW_CAP_DBG_QCAPS			0x80000000
+	u64			fw_cap;
+	#define BNXT_FW_CAP_SHORT_CMD			BIT_ULL(0)
+	#define BNXT_FW_CAP_LLDP_AGENT			BIT_ULL(1)
+	#define BNXT_FW_CAP_DCBX_AGENT			BIT_ULL(2)
+	#define BNXT_FW_CAP_NEW_RM			BIT_ULL(3)
+	#define BNXT_FW_CAP_IF_CHANGE			BIT_ULL(4)
+	#define BNXT_FW_CAP_KONG_MB_CHNL		BIT_ULL(7)
+	#define BNXT_FW_CAP_OVS_64BIT_HANDLE		BIT_ULL(10)
+	#define BNXT_FW_CAP_TRUSTED_VF			BIT_ULL(11)
+	#define BNXT_FW_CAP_ERROR_RECOVERY		BIT_ULL(13)
+	#define BNXT_FW_CAP_PKG_VER			BIT_ULL(14)
+	#define BNXT_FW_CAP_CFA_ADV_FLOW		BIT_ULL(15)
+	#define BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2	BIT_ULL(16)
+	#define BNXT_FW_CAP_PCIE_STATS_SUPPORTED	BIT_ULL(17)
+	#define BNXT_FW_CAP_EXT_STATS_SUPPORTED		BIT_ULL(18)
+	#define BNXT_FW_CAP_RSS_HASH_TYPE_DELTA		BIT_ULL(19)
+	#define BNXT_FW_CAP_ERR_RECOVER_RELOAD		BIT_ULL(20)
+	#define BNXT_FW_CAP_HOT_RESET			BIT_ULL(21)
+	#define BNXT_FW_CAP_PTP_RTC			BIT_ULL(22)
+	#define BNXT_FW_CAP_RX_ALL_PKT_TS		BIT_ULL(23)
+	#define BNXT_FW_CAP_VLAN_RX_STRIP		BIT_ULL(24)
+	#define BNXT_FW_CAP_VLAN_TX_INSERT		BIT_ULL(25)
+	#define BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED	BIT_ULL(26)
+	#define BNXT_FW_CAP_LIVEPATCH			BIT_ULL(27)
+	#define BNXT_FW_CAP_PTP_PPS			BIT_ULL(28)
+	#define BNXT_FW_CAP_HOT_RESET_IF		BIT_ULL(29)
+	#define BNXT_FW_CAP_RING_MONITOR		BIT_ULL(30)
+	#define BNXT_FW_CAP_DBG_QCAPS			BIT_ULL(31)
 
 	u32			fw_dbg_cap;
 
-- 
cgit v1.2.3-70-g09d2


From edc5287315489ccc6d049efe46227cedc9ede673 Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Tue, 21 Mar 2023 07:44:48 -0700
Subject: bnxt: Defer PTP initialization to after querying function caps

Driver uses the flag BNXT_FLAG_MULTI_HOST to determine whether
to use non-realtime mode PHC when running on a multi host NIC.
However when ptp initializes on a NIC with shared PHC, we still
don't have this flag set yet because HWRM_FUNC_QCFG is issued
much later.

Move the ptp initialization code after we have issued func_qcfg.
The next patch will use the BNXT_FLAG_MULTI_HOST flag during PTP
initialization.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Acked-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 +++-
 drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 +
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index f533a8f46217..656a28ac2ff0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -7769,7 +7769,7 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp)
 		if (flags & FUNC_QCAPS_RESP_FLAGS_WOL_MAGICPKT_SUPPORTED)
 			bp->flags |= BNXT_FLAG_WOL_CAP;
 		if (flags & FUNC_QCAPS_RESP_FLAGS_PTP_SUPPORTED) {
-			__bnxt_hwrm_ptp_qcfg(bp);
+			bp->fw_cap |= BNXT_FW_CAP_PTP;
 		} else {
 			bnxt_ptp_clear(bp);
 			kfree(bp->ptp_cfg);
@@ -12298,6 +12298,8 @@ static int bnxt_fw_init_one_p2(struct bnxt *bp)
 	bnxt_hwrm_vnic_qcaps(bp);
 	bnxt_hwrm_port_led_qcaps(bp);
 	bnxt_ethtool_init(bp);
+	if (bp->fw_cap & BNXT_FW_CAP_PTP)
+		__bnxt_hwrm_ptp_qcfg(bp);
 	bnxt_dcb_init(bp);
 	return 0;
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index d7eb0d244f42..0eeaed95a4ac 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1996,6 +1996,7 @@ struct bnxt {
 	#define BNXT_FW_CAP_HOT_RESET_IF		BIT_ULL(29)
 	#define BNXT_FW_CAP_RING_MONITOR		BIT_ULL(30)
 	#define BNXT_FW_CAP_DBG_QCAPS			BIT_ULL(31)
+	#define BNXT_FW_CAP_PTP				BIT_ULL(32)
 
 	u32			fw_dbg_cap;
 
-- 
cgit v1.2.3-70-g09d2


From a02c33130709736f5a770a55e5bfeda871a8c1cc Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Tue, 21 Mar 2023 07:44:49 -0700
Subject: bnxt: Enforce PTP software freq adjustments only when in non-RTC mode

Currently driver performs software based frequency adjustments
when RTC capability is not discovered or when in shared PHC mode.
But there may be some old firmware versions that still support
hardware freq adjustments without RTC capability being exposed.
In this situation driver will use non-realtime mode even on single
host NICs.

Hence enforce software frequency adjustments only when running in
shared PHC mode. Make suitable changes for cyclecounter for the
same.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Acked-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index a3a3978a4d1c..e46689128e32 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -230,7 +230,7 @@ static int bnxt_ptp_adjfine(struct ptp_clock_info *ptp_info, long scaled_ppm)
 						ptp_info);
 	struct bnxt *bp = ptp->bp;
 
-	if (BNXT_PTP_USE_RTC(bp))
+	if (!BNXT_MH(bp))
 		return bnxt_ptp_adjfine_rtc(bp, scaled_ppm);
 
 	spin_lock_bh(&ptp->ptp_lock);
@@ -861,9 +861,15 @@ static void bnxt_ptp_timecounter_init(struct bnxt *bp, bool init_tc)
 		memset(&ptp->cc, 0, sizeof(ptp->cc));
 		ptp->cc.read = bnxt_cc_read;
 		ptp->cc.mask = CYCLECOUNTER_MASK(48);
-		ptp->cc.shift = BNXT_CYCLES_SHIFT;
-		ptp->cc.mult = clocksource_khz2mult(BNXT_DEVCLK_FREQ, ptp->cc.shift);
-		ptp->cmult = ptp->cc.mult;
+		if (BNXT_MH(bp)) {
+			/* Use timecounter based non-real time mode */
+			ptp->cc.shift = BNXT_CYCLES_SHIFT;
+			ptp->cc.mult = clocksource_khz2mult(BNXT_DEVCLK_FREQ, ptp->cc.shift);
+			ptp->cmult = ptp->cc.mult;
+		} else {
+			ptp->cc.shift = 0;
+			ptp->cc.mult = 1;
+		}
 		ptp->next_overflow_check = jiffies + BNXT_PHC_OVERFLOW_PERIOD;
 	}
 	if (init_tc)
-- 
cgit v1.2.3-70-g09d2


From 22f5c234141d657a132f230aec08a670ee96232a Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 21 Mar 2023 14:17:45 +0100
Subject: ethernet: remove superfluous clearing of phydev

phy_disconnect() calls phy_detach() which already clears 'phydev' if it
is attached to a struct net_device.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230321131745.27688-1-wsa+renesas@sang-engineering.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep_main.c    | 1 -
 drivers/net/ethernet/microchip/lan743x_main.c | 1 -
 drivers/net/ethernet/renesas/rswitch.c        | 4 +---
 drivers/net/ethernet/smsc/smsc911x.c          | 2 --
 drivers/net/ethernet/sunplus/spl2sw_phy.c     | 4 +---
 5 files changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 6982aaa928b5..ed1b6102cfeb 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -246,7 +246,6 @@ static void tsnep_phy_close(struct tsnep_adapter *adapter)
 {
 	phy_stop(adapter->netdev->phydev);
 	phy_disconnect(adapter->netdev->phydev);
-	adapter->netdev->phydev = NULL;
 }
 
 static void tsnep_tx_ring_cleanup(struct tsnep_tx *tx)
diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c
index 7e0871b631e4..957d96a91a8a 100644
--- a/drivers/net/ethernet/microchip/lan743x_main.c
+++ b/drivers/net/ethernet/microchip/lan743x_main.c
@@ -1466,7 +1466,6 @@ static void lan743x_phy_close(struct lan743x_adapter *adapter)
 
 	phy_stop(netdev->phydev);
 	phy_disconnect(netdev->phydev);
-	netdev->phydev = NULL;
 }
 
 static void lan743x_phy_interface_select(struct lan743x_adapter *adapter)
diff --git a/drivers/net/ethernet/renesas/rswitch.c b/drivers/net/ethernet/renesas/rswitch.c
index c4f93d24c6a4..29afaddb598d 100644
--- a/drivers/net/ethernet/renesas/rswitch.c
+++ b/drivers/net/ethernet/renesas/rswitch.c
@@ -1324,10 +1324,8 @@ out:
 
 static void rswitch_phy_device_deinit(struct rswitch_device *rdev)
 {
-	if (rdev->ndev->phydev) {
+	if (rdev->ndev->phydev)
 		phy_disconnect(rdev->ndev->phydev);
-		rdev->ndev->phydev = NULL;
-	}
 }
 
 static int rswitch_serdes_set_params(struct rswitch_device *rdev)
diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c
index a2e511912e6a..39446d4e94b6 100644
--- a/drivers/net/ethernet/smsc/smsc911x.c
+++ b/drivers/net/ethernet/smsc/smsc911x.c
@@ -1741,7 +1741,6 @@ irq_stop_out:
 	free_irq(dev->irq, dev);
 mii_free_out:
 	phy_disconnect(dev->phydev);
-	dev->phydev = NULL;
 out:
 	pm_runtime_put(dev->dev.parent);
 	return retval;
@@ -1772,7 +1771,6 @@ static int smsc911x_stop(struct net_device *dev)
 	if (dev->phydev) {
 		phy_stop(dev->phydev);
 		phy_disconnect(dev->phydev);
-		dev->phydev = NULL;
 	}
 	netif_carrier_off(dev);
 	pm_runtime_put(dev->dev.parent);
diff --git a/drivers/net/ethernet/sunplus/spl2sw_phy.c b/drivers/net/ethernet/sunplus/spl2sw_phy.c
index 404f508a54d4..6f899e48f51d 100644
--- a/drivers/net/ethernet/sunplus/spl2sw_phy.c
+++ b/drivers/net/ethernet/sunplus/spl2sw_phy.c
@@ -84,9 +84,7 @@ void spl2sw_phy_remove(struct spl2sw_common *comm)
 	for (i = 0; i < MAX_NETDEV_NUM; i++)
 		if (comm->ndev[i]) {
 			ndev = comm->ndev[i];
-			if (ndev) {
+			if (ndev)
 				phy_disconnect(ndev->phydev);
-				ndev->phydev = NULL;
-			}
 		}
 }
-- 
cgit v1.2.3-70-g09d2


From 3ec5ac3133b54d5bcaaa9ddc95e15377178ef66e Mon Sep 17 00:00:00 2001
From: Álvaro Fernández Rojas <noltari@gmail.com>
Date: Tue, 21 Mar 2023 18:33:56 +0100
Subject: dt-bindings: net: dsa: b53: add more 63xx SoCs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BCM6318, BCM6362 and BCM63268 are SoCs with a B53 MMAP switch.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml b/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml
index 5bef4128d175..57e0ef93b134 100644
--- a/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml
@@ -57,8 +57,11 @@ properties:
       - items:
           - enum:
               - brcm,bcm3384-switch
+              - brcm,bcm6318-switch
               - brcm,bcm6328-switch
+              - brcm,bcm6362-switch
               - brcm,bcm6368-switch
+              - brcm,bcm63268-switch
           - const: brcm,bcm63xx-switch
 
 required:
-- 
cgit v1.2.3-70-g09d2


From a2b212fe5c32613e2beacc5ec3db4cdc0ab5a9e9 Mon Sep 17 00:00:00 2001
From: Álvaro Fernández Rojas <noltari@gmail.com>
Date: Tue, 21 Mar 2023 18:33:57 +0100
Subject: net: dsa: b53: mmap: add more 63xx SoCs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BCM6318, BCM6362 and BCM63268 are SoCs with a B53 MMAP switch.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/b53/b53_mmap.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c
index e968322dfbf0..f63aebd445e8 100644
--- a/drivers/net/dsa/b53/b53_mmap.c
+++ b/drivers/net/dsa/b53/b53_mmap.c
@@ -331,8 +331,11 @@ static void b53_mmap_shutdown(struct platform_device *pdev)
 
 static const struct of_device_id b53_mmap_of_table[] = {
 	{ .compatible = "brcm,bcm3384-switch" },
+	{ .compatible = "brcm,bcm6318-switch" },
 	{ .compatible = "brcm,bcm6328-switch" },
+	{ .compatible = "brcm,bcm6362-switch" },
 	{ .compatible = "brcm,bcm6368-switch" },
+	{ .compatible = "brcm,bcm63268-switch" },
 	{ .compatible = "brcm,bcm63xx-switch" },
 	{ /* sentinel */ },
 };
-- 
cgit v1.2.3-70-g09d2


From 260887c770eb60f57176bccf011fd0fe1bc62dc5 Mon Sep 17 00:00:00 2001
From: Álvaro Fernández Rojas <noltari@gmail.com>
Date: Tue, 21 Mar 2023 18:33:58 +0100
Subject: net: dsa: b53: mmap: allow passing a chip ID
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BCM6318 and BCM63268 SoCs require a special handling for their RGMIIs, so we
should be able to identify them as a special BCM63xx switch.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/b53/b53_common.c | 13 +++++++++++++
 drivers/net/dsa/b53/b53_mmap.c   | 32 +++++++++++++++++++++++---------
 drivers/net/dsa/b53/b53_priv.h   |  9 ++++++++-
 3 files changed, 44 insertions(+), 10 deletions(-)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 6e212f6f1cb9..97327d7a6760 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -2465,6 +2465,19 @@ static const struct b53_chip_data b53_switch_chips[] = {
 		.jumbo_pm_reg = B53_JUMBO_PORT_MASK_63XX,
 		.jumbo_size_reg = B53_JUMBO_MAX_SIZE_63XX,
 	},
+	{
+		.chip_id = BCM63268_DEVICE_ID,
+		.dev_name = "BCM63268",
+		.vlans = 4096,
+		.enabled_ports = 0, /* pdata must provide them */
+		.arl_bins = 4,
+		.arl_buckets = 1024,
+		.imp_port = 8,
+		.vta_regs = B53_VTA_REGS_63XX,
+		.duplex_reg = B53_DUPLEX_STAT_63XX,
+		.jumbo_pm_reg = B53_JUMBO_PORT_MASK_63XX,
+		.jumbo_size_reg = B53_JUMBO_MAX_SIZE_63XX,
+	},
 	{
 		.chip_id = BCM53010_DEVICE_ID,
 		.dev_name = "BCM53010",
diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c
index f63aebd445e8..a83a0c9d6939 100644
--- a/drivers/net/dsa/b53/b53_mmap.c
+++ b/drivers/net/dsa/b53/b53_mmap.c
@@ -248,7 +248,7 @@ static int b53_mmap_probe_of(struct platform_device *pdev,
 		return -ENOMEM;
 
 	pdata->regs = mem;
-	pdata->chip_id = BCM63XX_DEVICE_ID;
+	pdata->chip_id = (u32)(unsigned long)device_get_match_data(dev);
 	pdata->big_endian = of_property_read_bool(np, "big-endian");
 
 	of_ports = of_get_child_by_name(np, "ports");
@@ -330,14 +330,28 @@ static void b53_mmap_shutdown(struct platform_device *pdev)
 }
 
 static const struct of_device_id b53_mmap_of_table[] = {
-	{ .compatible = "brcm,bcm3384-switch" },
-	{ .compatible = "brcm,bcm6318-switch" },
-	{ .compatible = "brcm,bcm6328-switch" },
-	{ .compatible = "brcm,bcm6362-switch" },
-	{ .compatible = "brcm,bcm6368-switch" },
-	{ .compatible = "brcm,bcm63268-switch" },
-	{ .compatible = "brcm,bcm63xx-switch" },
-	{ /* sentinel */ },
+	{
+		.compatible = "brcm,bcm3384-switch",
+		.data = (void *)BCM63XX_DEVICE_ID,
+	}, {
+		.compatible = "brcm,bcm6318-switch",
+		.data = (void *)BCM63268_DEVICE_ID,
+	}, {
+		.compatible = "brcm,bcm6328-switch",
+		.data = (void *)BCM63XX_DEVICE_ID,
+	}, {
+		.compatible = "brcm,bcm6362-switch",
+		.data = (void *)BCM63XX_DEVICE_ID,
+	}, {
+		.compatible = "brcm,bcm6368-switch",
+		.data = (void *)BCM63XX_DEVICE_ID,
+	}, {
+		.compatible = "brcm,bcm63268-switch",
+		.data = (void *)BCM63268_DEVICE_ID,
+	}, {
+		.compatible = "brcm,bcm63xx-switch",
+		.data = (void *)BCM63XX_DEVICE_ID,
+	}, { /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, b53_mmap_of_table);
 
diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h
index 4cf9f540696e..a689a6950189 100644
--- a/drivers/net/dsa/b53/b53_priv.h
+++ b/drivers/net/dsa/b53/b53_priv.h
@@ -70,6 +70,7 @@ enum {
 	BCM53125_DEVICE_ID = 0x53125,
 	BCM53128_DEVICE_ID = 0x53128,
 	BCM63XX_DEVICE_ID = 0x6300,
+	BCM63268_DEVICE_ID = 0x63268,
 	BCM53010_DEVICE_ID = 0x53010,
 	BCM53011_DEVICE_ID = 0x53011,
 	BCM53012_DEVICE_ID = 0x53012,
@@ -191,7 +192,13 @@ static inline int is531x5(struct b53_device *dev)
 
 static inline int is63xx(struct b53_device *dev)
 {
-	return dev->chip_id == BCM63XX_DEVICE_ID;
+	return dev->chip_id == BCM63XX_DEVICE_ID ||
+		dev->chip_id == BCM63268_DEVICE_ID;
+}
+
+static inline int is63268(struct b53_device *dev)
+{
+	return dev->chip_id == BCM63268_DEVICE_ID;
 }
 
 static inline int is5301x(struct b53_device *dev)
-- 
cgit v1.2.3-70-g09d2


From 594c6c2e3ea2146d693c64ec3d0a26127d47aeec Mon Sep 17 00:00:00 2001
From: Álvaro Fernández Rojas <noltari@gmail.com>
Date: Tue, 21 Mar 2023 18:33:59 +0100
Subject: net: dsa: b53: add BCM63268 RGMII configuration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BCM63268 requires special RGMII configuration to work.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/b53/b53_common.c | 6 +++++-
 drivers/net/dsa/b53/b53_regs.h   | 1 +
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 97327d7a6760..1f9b251a5452 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -1240,8 +1240,12 @@ static void b53_adjust_63xx_rgmii(struct dsa_switch *ds, int port,
 		break;
 	}
 
-	if (port != dev->imp_port)
+	if (port != dev->imp_port) {
+		if (is63268(dev))
+			rgmii_ctrl |= RGMII_CTRL_MII_OVERRIDE;
+
 		rgmii_ctrl |= RGMII_CTRL_ENABLE_GMII;
+	}
 
 	b53_write8(dev, B53_CTRL_PAGE, off, rgmii_ctrl);
 
diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h
index b2c539a42154..bfbcb66bef66 100644
--- a/drivers/net/dsa/b53/b53_regs.h
+++ b/drivers/net/dsa/b53/b53_regs.h
@@ -138,6 +138,7 @@
 
 #define B53_RGMII_CTRL_IMP		0x60
 #define   RGMII_CTRL_ENABLE_GMII	BIT(7)
+#define   RGMII_CTRL_MII_OVERRIDE	BIT(6)
 #define   RGMII_CTRL_TIMING_SEL		BIT(2)
 #define   RGMII_CTRL_DLL_RXC		BIT(1)
 #define   RGMII_CTRL_DLL_TXC		BIT(0)
-- 
cgit v1.2.3-70-g09d2


From 3079bfdbda6cc776b4fba4556258966753a6d840 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Mon, 20 Mar 2023 18:37:54 -0500
Subject: dt-bindings: net: Drop unneeded quotes

Cleanup bindings dropping unneeded quotes. Once all these are fixed,
checking for this can be enabled in yamllint.

Acked-by: Marc Kleine-Budde <mkl@pengutronix.de> # for bindings/net/can
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com>
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Nobuhiro Iwamatsu <nobuhiro1.iwamatsu@toshiba.co.jp> # for bindings/net/toshiba,visconti-dwmac.yaml
Reviewed-by: Heiko Stuebner <heiko@sntech.de> #rockchip
Link: https://lore.kernel.org/r/20230320233758.2918972-1-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/actions,owl-emac.yaml      |  2 +-
 .../bindings/net/allwinner,sun4i-a10-emac.yaml         |  2 +-
 .../bindings/net/allwinner,sun4i-a10-mdio.yaml         |  2 +-
 Documentation/devicetree/bindings/net/altr,tse.yaml    |  2 +-
 .../devicetree/bindings/net/amlogic,meson-dwmac.yaml   |  4 ++--
 .../devicetree/bindings/net/aspeed,ast2600-mdio.yaml   |  2 +-
 Documentation/devicetree/bindings/net/brcm,amac.yaml   |  2 +-
 .../devicetree/bindings/net/brcm,systemport.yaml       |  2 +-
 .../devicetree/bindings/net/broadcom-bluetooth.yaml    |  2 +-
 .../devicetree/bindings/net/can/xilinx,can.yaml        |  6 +++---
 .../devicetree/bindings/net/dsa/brcm,sf2.yaml          |  2 +-
 Documentation/devicetree/bindings/net/dsa/qca8k.yaml   |  2 +-
 .../devicetree/bindings/net/engleder,tsnep.yaml        |  2 +-
 .../devicetree/bindings/net/ethernet-phy.yaml          |  2 +-
 .../devicetree/bindings/net/fsl,qoriq-mc-dpmac.yaml    |  2 +-
 .../bindings/net/intel,ixp46x-ptp-timer.yaml           |  4 ++--
 .../devicetree/bindings/net/intel,ixp4xx-ethernet.yaml | 12 ++++++------
 .../devicetree/bindings/net/intel,ixp4xx-hss.yaml      | 18 +++++++++---------
 .../devicetree/bindings/net/marvell,mvusb.yaml         |  2 +-
 .../devicetree/bindings/net/marvell-bluetooth.yaml     |  4 ++--
 Documentation/devicetree/bindings/net/mdio-gpio.yaml   |  2 +-
 .../devicetree/bindings/net/mediatek,net.yaml          |  2 +-
 .../devicetree/bindings/net/mediatek,star-emac.yaml    |  2 +-
 .../bindings/net/microchip,lan966x-switch.yaml         |  2 +-
 .../bindings/net/microchip,sparx5-switch.yaml          |  4 ++--
 Documentation/devicetree/bindings/net/mscc,miim.yaml   |  2 +-
 .../devicetree/bindings/net/nfc/marvell,nci.yaml       |  2 +-
 .../devicetree/bindings/net/nfc/nxp,pn532.yaml         |  2 +-
 .../bindings/net/pse-pd/podl-pse-regulator.yaml        |  2 +-
 .../devicetree/bindings/net/qcom,ipq4019-mdio.yaml     |  2 +-
 .../devicetree/bindings/net/qcom,ipq8064-mdio.yaml     |  2 +-
 .../devicetree/bindings/net/rockchip,emac.yaml         |  2 +-
 .../devicetree/bindings/net/rockchip-dwmac.yaml        |  4 ++--
 Documentation/devicetree/bindings/net/sff,sfp.yaml     |  4 ++--
 Documentation/devicetree/bindings/net/snps,dwmac.yaml  |  2 +-
 Documentation/devicetree/bindings/net/stm32-dwmac.yaml |  8 ++++----
 .../devicetree/bindings/net/ti,cpsw-switch.yaml        | 10 +++++-----
 .../devicetree/bindings/net/ti,davinci-mdio.yaml       |  2 +-
 Documentation/devicetree/bindings/net/ti,dp83822.yaml  |  6 +++---
 Documentation/devicetree/bindings/net/ti,dp83867.yaml  |  6 +++---
 Documentation/devicetree/bindings/net/ti,dp83869.yaml  |  6 +++---
 .../bindings/net/toshiba,visconti-dwmac.yaml           |  4 ++--
 .../devicetree/bindings/net/vertexcom-mse102x.yaml     |  4 ++--
 43 files changed, 79 insertions(+), 79 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/actions,owl-emac.yaml b/Documentation/devicetree/bindings/net/actions,owl-emac.yaml
index d30fada2ac39..5718ab4654b2 100644
--- a/Documentation/devicetree/bindings/net/actions,owl-emac.yaml
+++ b/Documentation/devicetree/bindings/net/actions,owl-emac.yaml
@@ -16,7 +16,7 @@ description: |
   operation modes at 10/100 Mb/s data transfer rates.
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-emac.yaml b/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-emac.yaml
index 987b91b9afe9..eb26623dab51 100644
--- a/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-emac.yaml
+++ b/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-emac.yaml
@@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Allwinner A10 EMAC Ethernet Controller
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 maintainers:
   - Chen-Yu Tsai <wens@csie.org>
diff --git a/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-mdio.yaml b/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-mdio.yaml
index ede977cdfb8d..85f552b907f3 100644
--- a/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-mdio.yaml
+++ b/Documentation/devicetree/bindings/net/allwinner,sun4i-a10-mdio.yaml
@@ -11,7 +11,7 @@ maintainers:
   - Maxime Ripard <mripard@kernel.org>
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 # Select every compatible, including the deprecated ones. This way, we
 # will be able to report a warning when we have that compatible, since
diff --git a/Documentation/devicetree/bindings/net/altr,tse.yaml b/Documentation/devicetree/bindings/net/altr,tse.yaml
index 8d1d94494349..9d02af468906 100644
--- a/Documentation/devicetree/bindings/net/altr,tse.yaml
+++ b/Documentation/devicetree/bindings/net/altr,tse.yaml
@@ -66,7 +66,7 @@ required:
   - tx-fifo-depth
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
   - if:
       properties:
         compatible:
diff --git a/Documentation/devicetree/bindings/net/amlogic,meson-dwmac.yaml b/Documentation/devicetree/bindings/net/amlogic,meson-dwmac.yaml
index ddd5a073c3a8..a2c51a84efa5 100644
--- a/Documentation/devicetree/bindings/net/amlogic,meson-dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/amlogic,meson-dwmac.yaml
@@ -2,8 +2,8 @@
 # Copyright 2019 BayLibre, SAS
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/amlogic,meson-dwmac.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/amlogic,meson-dwmac.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Amlogic Meson DWMAC Ethernet controller
 
diff --git a/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml b/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml
index f81eda8cb0a5..d6ef468495c5 100644
--- a/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml
+++ b/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml
@@ -15,7 +15,7 @@ description: |+
   MAC.
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/brcm,amac.yaml b/Documentation/devicetree/bindings/net/brcm,amac.yaml
index ee2eac8f5710..210fb29c4e7b 100644
--- a/Documentation/devicetree/bindings/net/brcm,amac.yaml
+++ b/Documentation/devicetree/bindings/net/brcm,amac.yaml
@@ -10,7 +10,7 @@ maintainers:
   - Florian Fainelli <f.fainelli@gmail.com>
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
   - if:
       properties:
         compatible:
diff --git a/Documentation/devicetree/bindings/net/brcm,systemport.yaml b/Documentation/devicetree/bindings/net/brcm,systemport.yaml
index 5fc9c9fafd85..b40006d44791 100644
--- a/Documentation/devicetree/bindings/net/brcm,systemport.yaml
+++ b/Documentation/devicetree/bindings/net/brcm,systemport.yaml
@@ -66,7 +66,7 @@ required:
   - phy-mode
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 unevaluatedProperties: false
 
diff --git a/Documentation/devicetree/bindings/net/broadcom-bluetooth.yaml b/Documentation/devicetree/bindings/net/broadcom-bluetooth.yaml
index b964c7dcec15..cc70b00c6ce5 100644
--- a/Documentation/devicetree/bindings/net/broadcom-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/broadcom-bluetooth.yaml
@@ -121,7 +121,7 @@ required:
   - compatible
 
 dependencies:
-  brcm,requires-autobaud-mode: [ 'shutdown-gpios' ]
+  brcm,requires-autobaud-mode: [ shutdown-gpios ]
 
 if:
   not:
diff --git a/Documentation/devicetree/bindings/net/can/xilinx,can.yaml b/Documentation/devicetree/bindings/net/can/xilinx,can.yaml
index 65af8183cb9c..897d2cbda45b 100644
--- a/Documentation/devicetree/bindings/net/can/xilinx,can.yaml
+++ b/Documentation/devicetree/bindings/net/can/xilinx,can.yaml
@@ -35,15 +35,15 @@ properties:
     maxItems: 1
 
   tx-fifo-depth:
-    $ref: "/schemas/types.yaml#/definitions/uint32"
+    $ref: /schemas/types.yaml#/definitions/uint32
     description: CAN Tx fifo depth (Zynq, Axi CAN).
 
   rx-fifo-depth:
-    $ref: "/schemas/types.yaml#/definitions/uint32"
+    $ref: /schemas/types.yaml#/definitions/uint32
     description: CAN Rx fifo depth (Zynq, Axi CAN, CAN FD in sequential Rx mode)
 
   tx-mailbox-count:
-    $ref: "/schemas/types.yaml#/definitions/uint32"
+    $ref: /schemas/types.yaml#/definitions/uint32
     description: CAN Tx mailbox buffer count (CAN FD)
 
 required:
diff --git a/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml b/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml
index eed16e216fb6..37bf33bd4670 100644
--- a/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml
@@ -103,7 +103,7 @@ required:
   - "#size-cells"
 
 allOf:
-  - $ref: "dsa.yaml#"
+  - $ref: dsa.yaml#
   - if:
       properties:
         compatible:
diff --git a/Documentation/devicetree/bindings/net/dsa/qca8k.yaml b/Documentation/devicetree/bindings/net/dsa/qca8k.yaml
index 389892592aac..fe9ebe285938 100644
--- a/Documentation/devicetree/bindings/net/dsa/qca8k.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/qca8k.yaml
@@ -66,7 +66,7 @@ properties:
                  With the legacy mapping the reg corresponding to the internal
                  mdio is the switch reg with an offset of -1.
 
-$ref: "dsa.yaml#"
+$ref: dsa.yaml#
 
 patternProperties:
   "^(ethernet-)?ports$":
diff --git a/Documentation/devicetree/bindings/net/engleder,tsnep.yaml b/Documentation/devicetree/bindings/net/engleder,tsnep.yaml
index 4116667133ce..82a5d7927ca4 100644
--- a/Documentation/devicetree/bindings/net/engleder,tsnep.yaml
+++ b/Documentation/devicetree/bindings/net/engleder,tsnep.yaml
@@ -62,7 +62,7 @@ properties:
 
   mdio:
     type: object
-    $ref: "mdio.yaml#"
+    $ref: mdio.yaml#
     description: optional node for embedded MDIO controller
 
 required:
diff --git a/Documentation/devicetree/bindings/net/ethernet-phy.yaml b/Documentation/devicetree/bindings/net/ethernet-phy.yaml
index 1327b81f15a2..ac04f8efa35c 100644
--- a/Documentation/devicetree/bindings/net/ethernet-phy.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-phy.yaml
@@ -83,7 +83,7 @@ properties:
       0: Disable 2.4 Vpp operating mode.
       1: Request 2.4 Vpp operating mode from link partner.
       Absence of this property will leave configuration to default values.
-    $ref: "/schemas/types.yaml#/definitions/uint32"
+    $ref: /schemas/types.yaml#/definitions/uint32
     enum: [0, 1]
 
   broken-turn-around:
diff --git a/Documentation/devicetree/bindings/net/fsl,qoriq-mc-dpmac.yaml b/Documentation/devicetree/bindings/net/fsl,qoriq-mc-dpmac.yaml
index 6e0763898d3a..a1b71b35319e 100644
--- a/Documentation/devicetree/bindings/net/fsl,qoriq-mc-dpmac.yaml
+++ b/Documentation/devicetree/bindings/net/fsl,qoriq-mc-dpmac.yaml
@@ -14,7 +14,7 @@ description:
   located under the 'dpmacs' node for the fsl-mc bus DTS node.
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/intel,ixp46x-ptp-timer.yaml b/Documentation/devicetree/bindings/net/intel,ixp46x-ptp-timer.yaml
index 8b9b3f915d92..f92730b1d2fa 100644
--- a/Documentation/devicetree/bindings/net/intel,ixp46x-ptp-timer.yaml
+++ b/Documentation/devicetree/bindings/net/intel,ixp46x-ptp-timer.yaml
@@ -2,8 +2,8 @@
 # Copyright 2018 Linaro Ltd.
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/intel,ixp46x-ptp-timer.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/intel,ixp46x-ptp-timer.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Intel IXP46x PTP Timer (TSYNC)
 
diff --git a/Documentation/devicetree/bindings/net/intel,ixp4xx-ethernet.yaml b/Documentation/devicetree/bindings/net/intel,ixp4xx-ethernet.yaml
index 4e1b79818aff..4fdc5328826c 100644
--- a/Documentation/devicetree/bindings/net/intel,ixp4xx-ethernet.yaml
+++ b/Documentation/devicetree/bindings/net/intel,ixp4xx-ethernet.yaml
@@ -2,13 +2,13 @@
 # Copyright 2018 Linaro Ltd.
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/intel,ixp4xx-ethernet.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/intel,ixp4xx-ethernet.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Intel IXP4xx ethernet
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 maintainers:
   - Linus Walleij <linus.walleij@linaro.org>
@@ -28,7 +28,7 @@ properties:
     description: Ethernet MMIO address range
 
   queue-rx:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the RX queue node
@@ -36,7 +36,7 @@ properties:
     description: phandle to the RX queue on the NPE
 
   queue-txready:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the TX READY queue node
@@ -48,7 +48,7 @@ properties:
   phy-handle: true
 
   intel,npe-handle:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the NPE this ethernet instance is using
diff --git a/Documentation/devicetree/bindings/net/intel,ixp4xx-hss.yaml b/Documentation/devicetree/bindings/net/intel,ixp4xx-hss.yaml
index e6329febb60c..7a405e9b37b2 100644
--- a/Documentation/devicetree/bindings/net/intel,ixp4xx-hss.yaml
+++ b/Documentation/devicetree/bindings/net/intel,ixp4xx-hss.yaml
@@ -2,8 +2,8 @@
 # Copyright 2021 Linaro Ltd.
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/intel,ixp4xx-hss.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/intel,ixp4xx-hss.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Intel IXP4xx V.35 WAN High Speed Serial Link (HSS)
 
@@ -24,7 +24,7 @@ properties:
     description: The HSS instance
 
   intel,npe-handle:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       items:
         - description: phandle to the NPE this HSS instance is using
@@ -33,7 +33,7 @@ properties:
       and the instance to use in the second cell
 
   intel,queue-chl-rxtrig:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the RX trigger queue on the NPE
@@ -41,7 +41,7 @@ properties:
     description: phandle to the RX trigger queue on the NPE
 
   intel,queue-chl-txready:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the TX ready queue on the NPE
@@ -49,7 +49,7 @@ properties:
     description: phandle to the TX ready queue on the NPE
 
   intel,queue-pkt-rx:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the RX queue on the NPE
@@ -57,7 +57,7 @@ properties:
     description: phandle to the packet RX queue on the NPE
 
   intel,queue-pkt-tx:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     maxItems: 4
     items:
       items:
@@ -66,7 +66,7 @@ properties:
     description: phandle to the packet TX0, TX1, TX2 and TX3 queues on the NPE
 
   intel,queue-pkt-rxfree:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     maxItems: 4
     items:
       items:
@@ -76,7 +76,7 @@ properties:
       RXFREE3 queues on the NPE
 
   intel,queue-pkt-txdone:
-    $ref: '/schemas/types.yaml#/definitions/phandle-array'
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the TXDONE queue on the NPE
diff --git a/Documentation/devicetree/bindings/net/marvell,mvusb.yaml b/Documentation/devicetree/bindings/net/marvell,mvusb.yaml
index 8e288ab38fd7..3a3325168048 100644
--- a/Documentation/devicetree/bindings/net/marvell,mvusb.yaml
+++ b/Documentation/devicetree/bindings/net/marvell,mvusb.yaml
@@ -20,7 +20,7 @@ description: |+
   definition.
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml b/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
index 309ef21a1e37..6aa7a078faa2 100644
--- a/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/marvell-bluetooth.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/marvell-bluetooth.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Marvell Bluetooth chips
 
diff --git a/Documentation/devicetree/bindings/net/mdio-gpio.yaml b/Documentation/devicetree/bindings/net/mdio-gpio.yaml
index 1d83b8dcce2c..dca1aec119e3 100644
--- a/Documentation/devicetree/bindings/net/mdio-gpio.yaml
+++ b/Documentation/devicetree/bindings/net/mdio-gpio.yaml
@@ -12,7 +12,7 @@ maintainers:
   - Russell King <linux@armlinux.org.uk>
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/mediatek,net.yaml b/Documentation/devicetree/bindings/net/mediatek,net.yaml
index b7f6474dc5ab..acb2b2ac4fe1 100644
--- a/Documentation/devicetree/bindings/net/mediatek,net.yaml
+++ b/Documentation/devicetree/bindings/net/mediatek,net.yaml
@@ -97,7 +97,7 @@ properties:
     const: 0
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
   - if:
       properties:
         compatible:
diff --git a/Documentation/devicetree/bindings/net/mediatek,star-emac.yaml b/Documentation/devicetree/bindings/net/mediatek,star-emac.yaml
index 64c893c98d80..2e889f9a563e 100644
--- a/Documentation/devicetree/bindings/net/mediatek,star-emac.yaml
+++ b/Documentation/devicetree/bindings/net/mediatek,star-emac.yaml
@@ -15,7 +15,7 @@ description:
   modes with flow-control as well as CRC offloading and VLAN tags.
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/microchip,lan966x-switch.yaml b/Documentation/devicetree/bindings/net/microchip,lan966x-switch.yaml
index dc116f14750e..306ef9ecf2b9 100644
--- a/Documentation/devicetree/bindings/net/microchip,lan966x-switch.yaml
+++ b/Documentation/devicetree/bindings/net/microchip,lan966x-switch.yaml
@@ -73,7 +73,7 @@ properties:
       "^port@[0-9a-f]+$":
         type: object
 
-        $ref: "/schemas/net/ethernet-controller.yaml#"
+        $ref: /schemas/net/ethernet-controller.yaml#
         unevaluatedProperties: false
 
         properties:
diff --git a/Documentation/devicetree/bindings/net/microchip,sparx5-switch.yaml b/Documentation/devicetree/bindings/net/microchip,sparx5-switch.yaml
index 57ffeb8fc876..fcafef8d5a33 100644
--- a/Documentation/devicetree/bindings/net/microchip,sparx5-switch.yaml
+++ b/Documentation/devicetree/bindings/net/microchip,sparx5-switch.yaml
@@ -99,7 +99,7 @@ properties:
 
           microchip,bandwidth:
             description: Specifies bandwidth in Mbit/s allocated to the port.
-            $ref: "/schemas/types.yaml#/definitions/uint32"
+            $ref: /schemas/types.yaml#/definitions/uint32
             maximum: 25000
 
           microchip,sd-sgpio:
@@ -107,7 +107,7 @@ properties:
               Index of the ports Signal Detect SGPIO in the set of 384 SGPIOs
               This is optional, and only needed if the default used index is
               is not correct.
-            $ref: "/schemas/types.yaml#/definitions/uint32"
+            $ref: /schemas/types.yaml#/definitions/uint32
             minimum: 0
             maximum: 383
 
diff --git a/Documentation/devicetree/bindings/net/mscc,miim.yaml b/Documentation/devicetree/bindings/net/mscc,miim.yaml
index 2c451cfa4e0b..5b292e7c9e46 100644
--- a/Documentation/devicetree/bindings/net/mscc,miim.yaml
+++ b/Documentation/devicetree/bindings/net/mscc,miim.yaml
@@ -10,7 +10,7 @@ maintainers:
   - Alexandre Belloni <alexandre.belloni@bootlin.com>
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/nfc/marvell,nci.yaml b/Documentation/devicetree/bindings/net/nfc/marvell,nci.yaml
index 308485a8ee6c..8e9a95f24c80 100644
--- a/Documentation/devicetree/bindings/net/nfc/marvell,nci.yaml
+++ b/Documentation/devicetree/bindings/net/nfc/marvell,nci.yaml
@@ -28,7 +28,7 @@ properties:
     maxItems: 1
 
   reset-n-io:
-    $ref: "/schemas/types.yaml#/definitions/phandle-array"
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     maxItems: 1
     description: |
       Output GPIO pin used to reset the chip (active low)
diff --git a/Documentation/devicetree/bindings/net/nfc/nxp,pn532.yaml b/Documentation/devicetree/bindings/net/nfc/nxp,pn532.yaml
index 0509e0166345..07c67c1e985f 100644
--- a/Documentation/devicetree/bindings/net/nfc/nxp,pn532.yaml
+++ b/Documentation/devicetree/bindings/net/nfc/nxp,pn532.yaml
@@ -31,7 +31,7 @@ required:
   - compatible
 
 dependencies:
-  interrupts: [ 'reg' ]
+  interrupts: [ reg ]
 
 additionalProperties: false
 
diff --git a/Documentation/devicetree/bindings/net/pse-pd/podl-pse-regulator.yaml b/Documentation/devicetree/bindings/net/pse-pd/podl-pse-regulator.yaml
index c6b1c188abf7..94a527e6aa1b 100644
--- a/Documentation/devicetree/bindings/net/pse-pd/podl-pse-regulator.yaml
+++ b/Documentation/devicetree/bindings/net/pse-pd/podl-pse-regulator.yaml
@@ -13,7 +13,7 @@ description: Regulator based PoDL PSE controller. The device must be referenced
   by the PHY node to control power injection to the Ethernet cable.
 
 allOf:
-  - $ref: "pse-controller.yaml#"
+  - $ref: pse-controller.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/qcom,ipq4019-mdio.yaml b/Documentation/devicetree/bindings/net/qcom,ipq4019-mdio.yaml
index 7631ecc8fd01..3407e909e8a7 100644
--- a/Documentation/devicetree/bindings/net/qcom,ipq4019-mdio.yaml
+++ b/Documentation/devicetree/bindings/net/qcom,ipq4019-mdio.yaml
@@ -51,7 +51,7 @@ required:
   - "#size-cells"
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
   - if:
       properties:
diff --git a/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml b/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml
index d7748dd33199..144001ff840c 100644
--- a/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml
+++ b/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml
@@ -14,7 +14,7 @@ description:
   used to communicate with the gmac phy connected.
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/rockchip,emac.yaml b/Documentation/devicetree/bindings/net/rockchip,emac.yaml
index a6d4f14df442..364028b3bba4 100644
--- a/Documentation/devicetree/bindings/net/rockchip,emac.yaml
+++ b/Documentation/devicetree/bindings/net/rockchip,emac.yaml
@@ -61,7 +61,7 @@ required:
   - mdio
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
   - if:
       properties:
         compatible:
diff --git a/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml b/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml
index 04936632fcbb..2a21bbe02892 100644
--- a/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/rockchip-dwmac.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/rockchip-dwmac.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Rockchip 10/100/1000 Ethernet driver(GMAC)
 
diff --git a/Documentation/devicetree/bindings/net/sff,sfp.yaml b/Documentation/devicetree/bindings/net/sff,sfp.yaml
index 231c4d75e4b1..973e478a399d 100644
--- a/Documentation/devicetree/bindings/net/sff,sfp.yaml
+++ b/Documentation/devicetree/bindings/net/sff,sfp.yaml
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/sff,sfp.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/sff,sfp.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Small Form Factor (SFF) Committee Small Form-factor Pluggable (SFP)
   Transceiver
diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 16b7d2904696..74f2ddc12018 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -555,7 +555,7 @@ dependencies:
   snps,reset-delays-us: ["snps,reset-gpio"]
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
   - if:
       properties:
         compatible:
diff --git a/Documentation/devicetree/bindings/net/stm32-dwmac.yaml b/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
index 5c93167b3b41..fc8c96b08d7d 100644
--- a/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
@@ -2,8 +2,8 @@
 # Copyright 2019 BayLibre, SAS
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/stm32-dwmac.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/stm32-dwmac.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: STMicroelectronics STM32 / MCU DWMAC glue layer controller
 
@@ -26,7 +26,7 @@ select:
     - compatible
 
 allOf:
-  - $ref: "snps,dwmac.yaml#"
+  - $ref: snps,dwmac.yaml#
 
 properties:
   compatible:
@@ -73,7 +73,7 @@ properties:
         - ptp_ref
 
   st,syscon:
-    $ref: "/schemas/types.yaml#/definitions/phandle-array"
+    $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
       - items:
           - description: phandle to the syscon node which encompases the glue register
diff --git a/Documentation/devicetree/bindings/net/ti,cpsw-switch.yaml b/Documentation/devicetree/bindings/net/ti,cpsw-switch.yaml
index e36c7817be69..b04ac4966608 100644
--- a/Documentation/devicetree/bindings/net/ti,cpsw-switch.yaml
+++ b/Documentation/devicetree/bindings/net/ti,cpsw-switch.yaml
@@ -62,10 +62,10 @@ properties:
 
   interrupt-names:
     items:
-      - const: "rx_thresh"
-      - const: "rx"
-      - const: "tx"
-      - const: "misc"
+      - const: rx_thresh
+      - const: rx
+      - const: tx
+      - const: misc
 
   pinctrl-names: true
 
@@ -154,7 +154,7 @@ patternProperties:
     type: object
     description:
       CPSW MDIO bus.
-    $ref: "ti,davinci-mdio.yaml#"
+    $ref: ti,davinci-mdio.yaml#
 
 
 required:
diff --git a/Documentation/devicetree/bindings/net/ti,davinci-mdio.yaml b/Documentation/devicetree/bindings/net/ti,davinci-mdio.yaml
index a339202c5e8e..53604fab0b73 100644
--- a/Documentation/devicetree/bindings/net/ti,davinci-mdio.yaml
+++ b/Documentation/devicetree/bindings/net/ti,davinci-mdio.yaml
@@ -13,7 +13,7 @@ description:
   TI SoC Davinci/Keystone2 MDIO Controller
 
 allOf:
-  - $ref: "mdio.yaml#"
+  - $ref: mdio.yaml#
 
 properties:
   compatible:
diff --git a/Documentation/devicetree/bindings/net/ti,dp83822.yaml b/Documentation/devicetree/bindings/net/ti,dp83822.yaml
index f2489a9c852f..db74474207ed 100644
--- a/Documentation/devicetree/bindings/net/ti,dp83822.yaml
+++ b/Documentation/devicetree/bindings/net/ti,dp83822.yaml
@@ -2,8 +2,8 @@
 # Copyright (C) 2020 Texas Instruments Incorporated
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/ti,dp83822.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/ti,dp83822.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: TI DP83822 ethernet PHY
 
@@ -21,7 +21,7 @@ description: |
     http://www.ti.com/lit/ds/symlink/dp83822i.pdf
 
 allOf:
-  - $ref: "ethernet-phy.yaml#"
+  - $ref: ethernet-phy.yaml#
 
 properties:
   reg:
diff --git a/Documentation/devicetree/bindings/net/ti,dp83867.yaml b/Documentation/devicetree/bindings/net/ti,dp83867.yaml
index b8c0e4b5b494..4bc1f98fd9fe 100644
--- a/Documentation/devicetree/bindings/net/ti,dp83867.yaml
+++ b/Documentation/devicetree/bindings/net/ti,dp83867.yaml
@@ -2,13 +2,13 @@
 # Copyright (C) 2019 Texas Instruments Incorporated
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/ti,dp83867.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/ti,dp83867.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: TI DP83867 ethernet PHY
 
 allOf:
-  - $ref: "ethernet-controller.yaml#"
+  - $ref: ethernet-controller.yaml#
 
 maintainers:
   - Andrew Davis <afd@ti.com>
diff --git a/Documentation/devicetree/bindings/net/ti,dp83869.yaml b/Documentation/devicetree/bindings/net/ti,dp83869.yaml
index b04ff0014a59..fb6725df4668 100644
--- a/Documentation/devicetree/bindings/net/ti,dp83869.yaml
+++ b/Documentation/devicetree/bindings/net/ti,dp83869.yaml
@@ -2,13 +2,13 @@
 # Copyright (C) 2019 Texas Instruments Incorporated
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/ti,dp83869.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/ti,dp83869.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: TI DP83869 ethernet PHY
 
 allOf:
-  - $ref: "ethernet-phy.yaml#"
+  - $ref: ethernet-phy.yaml#
 
 maintainers:
   - Andrew Davis <afd@ti.com>
diff --git a/Documentation/devicetree/bindings/net/toshiba,visconti-dwmac.yaml b/Documentation/devicetree/bindings/net/toshiba,visconti-dwmac.yaml
index 0988ed8d1c12..474fa8bcf302 100644
--- a/Documentation/devicetree/bindings/net/toshiba,visconti-dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/toshiba,visconti-dwmac.yaml
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/toshiba,visconti-dwmac.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/toshiba,visconti-dwmac.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Toshiba Visconti DWMAC Ethernet controller
 
diff --git a/Documentation/devicetree/bindings/net/vertexcom-mse102x.yaml b/Documentation/devicetree/bindings/net/vertexcom-mse102x.yaml
index 6a71f694cb55..4c4ced8cfa4b 100644
--- a/Documentation/devicetree/bindings/net/vertexcom-mse102x.yaml
+++ b/Documentation/devicetree/bindings/net/vertexcom-mse102x.yaml
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
 %YAML 1.2
 ---
-$id: "http://devicetree.org/schemas/net/vertexcom-mse102x.yaml#"
-$schema: "http://devicetree.org/meta-schemas/core.yaml#"
+$id: http://devicetree.org/schemas/net/vertexcom-mse102x.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: The Vertexcom MSE102x (SPI)
 
-- 
cgit v1.2.3-70-g09d2


From 603c3345589de01e26c93d3a74413af6ca1c93b9 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Tue, 21 Mar 2023 14:48:11 -0400
Subject: liquidio: remove unused IQ_INSTR_MODE_64B function

clang with W=1 reports
drivers/net/ethernet/cavium/liquidio/request_manager.c:43:19: error:
  unused function 'IQ_INSTR_MODE_64B' [-Werror,-Wunused-function]
static inline int IQ_INSTR_MODE_64B(struct octeon_device *oct, int iq_no)
                  ^
This function and its macro wrapper are not used, so remove them.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230321184811.1827306-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cavium/liquidio/request_manager.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c
index 8e59c2825533..32f854c0cd79 100644
--- a/drivers/net/ethernet/cavium/liquidio/request_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c
@@ -40,15 +40,6 @@ static void  __check_db_timeout(struct octeon_device *oct, u64 iq_no);
 
 static void (*reqtype_free_fn[MAX_OCTEON_DEVICES][REQTYPE_LAST + 1]) (void *);
 
-static inline int IQ_INSTR_MODE_64B(struct octeon_device *oct, int iq_no)
-{
-	struct octeon_instr_queue *iq =
-	    (struct octeon_instr_queue *)oct->instr_queue[iq_no];
-	return iq->iqcmd_64B;
-}
-
-#define IQ_INSTR_MODE_32B(oct, iq_no)  (!IQ_INSTR_MODE_64B(oct, iq_no))
-
 /* Define this to return the request status comaptible to old code */
 /*#define OCTEON_USE_OLD_REQ_STATUS*/
 
-- 
cgit v1.2.3-70-g09d2


From a2935a1cd85fc76198a1980abfeb6dc9a776dcf3 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 21 Mar 2023 16:49:55 +0530
Subject: net: ethernet: ti: am65-cpsw: Simplify setting supported interface

Convert the existing IF/ELSE statement based approach of setting the
supported_interfaces member of struct "phylink_config", to SWITCH
statements. This will help scale to newer PHY-MODES as well as newer
compatibles.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 4cfbc1c2b1c4..cba8db14e160 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2143,15 +2143,30 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 	port->slave.phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100 | MAC_1000FD;
 	port->slave.phylink_config.mac_managed_pm = true; /* MAC does PM */
 
-	if (phy_interface_mode_is_rgmii(port->slave.phy_if)) {
+	switch (port->slave.phy_if) {
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
 		phy_interface_set_rgmii(port->slave.phylink_config.supported_interfaces);
-	} else if (port->slave.phy_if == PHY_INTERFACE_MODE_RMII) {
+		break;
+
+	case PHY_INTERFACE_MODE_RMII:
 		__set_bit(PHY_INTERFACE_MODE_RMII,
 			  port->slave.phylink_config.supported_interfaces);
-	} else if (common->pdata.extra_modes & BIT(port->slave.phy_if)) {
-		__set_bit(PHY_INTERFACE_MODE_QSGMII,
-			  port->slave.phylink_config.supported_interfaces);
-	} else {
+		break;
+
+	case PHY_INTERFACE_MODE_QSGMII:
+		if (common->pdata.extra_modes & BIT(port->slave.phy_if)) {
+			__set_bit(port->slave.phy_if,
+				  port->slave.phylink_config.supported_interfaces);
+		} else {
+			dev_err(dev, "selected phy-mode is not supported\n");
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	default:
 		dev_err(dev, "selected phy-mode is not supported\n");
 		return -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3-70-g09d2


From e0f72db37547d605d66c344c4d01a39ad4b7c4be Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 21 Mar 2023 16:49:56 +0530
Subject: net: ethernet: ti: am65-cpsw: Add support for SGMII mode

Add support for configuring the CPSW Ethernet Switch in SGMII mode.

Depending on the SoC, allow selecting SGMII mode as a supported interface,
based on the compatible used.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index cba8db14e160..d2ca1f2035f4 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -76,6 +76,7 @@
 #define AM65_CPSW_PORTN_REG_TS_CTL_LTYPE2       0x31C
 
 #define AM65_CPSW_SGMII_CONTROL_REG		0x010
+#define AM65_CPSW_SGMII_MR_ADV_ABILITY_REG	0x018
 #define AM65_CPSW_SGMII_CONTROL_MR_AN_ENABLE	BIT(0)
 
 #define AM65_CPSW_CTL_VLAN_AWARE		BIT(1)
@@ -1496,9 +1497,14 @@ static void am65_cpsw_nuss_mac_config(struct phylink_config *config, unsigned in
 	struct am65_cpsw_port *port = container_of(slave, struct am65_cpsw_port, slave);
 	struct am65_cpsw_common *common = port->common;
 
-	if (common->pdata.extra_modes & BIT(state->interface))
+	if (common->pdata.extra_modes & BIT(state->interface)) {
+		if (state->interface == PHY_INTERFACE_MODE_SGMII)
+			writel(ADVERTISE_SGMII,
+			       port->sgmii_base + AM65_CPSW_SGMII_MR_ADV_ABILITY_REG);
+
 		writel(AM65_CPSW_SGMII_CONTROL_MR_AN_ENABLE,
 		       port->sgmii_base + AM65_CPSW_SGMII_CONTROL_REG);
+	}
 }
 
 static void am65_cpsw_nuss_mac_link_down(struct phylink_config *config, unsigned int mode,
@@ -1539,6 +1545,8 @@ static void am65_cpsw_nuss_mac_link_up(struct phylink_config *config, struct phy
 
 	if (speed == SPEED_1000)
 		mac_control |= CPSW_SL_CTL_GIG;
+	if (interface == PHY_INTERFACE_MODE_SGMII)
+		mac_control |= CPSW_SL_CTL_EXT_EN;
 	if (speed == SPEED_10 && phy_interface_mode_is_rgmii(interface))
 		/* Can be used with in band mode only */
 		mac_control |= CPSW_SL_CTL_EXT_EN;
@@ -2157,6 +2165,7 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 		break;
 
 	case PHY_INTERFACE_MODE_QSGMII:
+	case PHY_INTERFACE_MODE_SGMII:
 		if (common->pdata.extra_modes & BIT(port->slave.phy_if)) {
 			__set_bit(port->slave.phy_if,
 				  port->slave.phylink_config.supported_interfaces);
-- 
cgit v1.2.3-70-g09d2


From 2e20e764f24e12cf906f6fe61d79d1bcb69d6b65 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 21 Mar 2023 16:49:57 +0530
Subject: net: ethernet: ti: am65-cpsw: Enable SGMII mode for J7200

TI's J7200 SoC supports SGMII mode with the CPSW5G instance of the CPSW
Ethernet Switch. Thus, enable it by adding SGMII mode to the extra_modes
member of the "j7200_cpswxg_pdata" SoC data.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index d2ca1f2035f4..66e1fe58b895 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2777,7 +2777,7 @@ static const struct am65_cpsw_pdata j7200_cpswxg_pdata = {
 	.quirks = 0,
 	.ale_dev_id = "am64-cpswxg",
 	.fdqring_mode = K3_RINGACC_RING_MODE_RING,
-	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII),
+	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII) | BIT(PHY_INTERFACE_MODE_SGMII),
 };
 
 static const struct am65_cpsw_pdata j721e_cpswxg_pdata = {
-- 
cgit v1.2.3-70-g09d2


From 186016da9cca040ae535de09ab1be67e32373443 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 21 Mar 2023 16:49:58 +0530
Subject: net: ethernet: ti: am65-cpsw: Enable SGMII mode for J721E

TI's J721E SoC supports SGMII mode with the CPSW9G instance of the CPSW
Ethernet Switch. Thus, enable it by adding SGMII mode to the extra_modes
member of the "j721e_cpswxg_pdata" SoC data.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 66e1fe58b895..9ddb79776c88 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2784,7 +2784,7 @@ static const struct am65_cpsw_pdata j721e_cpswxg_pdata = {
 	.quirks = 0,
 	.ale_dev_id = "am64-cpswxg",
 	.fdqring_mode = K3_RINGACC_RING_MODE_MESSAGE,
-	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII),
+	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII) | BIT(PHY_INTERFACE_MODE_SGMII),
 };
 
 static const struct of_device_id am65_cpsw_nuss_of_mtable[] = {
-- 
cgit v1.2.3-70-g09d2


From 5dd0dfd55baec0742ba8f5625a0dd064aca7db16 Mon Sep 17 00:00:00 2001
From: Nick Child <nnac123@linux.ibm.com>
Date: Tue, 21 Mar 2023 10:07:24 -0500
Subject: net: Catch invalid index in XPS mapping

When setting the XPS value of a TX queue, warn the user once if the
index of the queue is greater than the number of allocated TX queues.

Previously, this scenario went uncaught. In the best case, it resulted
in unnecessary allocations. In the worst case, it resulted in
out-of-bounds memory references through calls to `netdev_get_tx_queue(
dev, index)`. Therefore, it is important to inform the user but not
worth returning an error and risk downing the netdevice.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Reviewed-by: Piotr Raczynski <piotr.raczynski@intel.com>
Link: https://lore.kernel.org/r/20230321150725.127229-1-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/dev.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/core/dev.c b/net/core/dev.c
index c7853192563d..c278beee6792 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2535,6 +2535,8 @@ int __netif_set_xps_queue(struct net_device *dev, const unsigned long *mask,
 	struct xps_map *map, *new_map;
 	unsigned int nr_ids;
 
+	WARN_ON_ONCE(index >= dev->num_tx_queues);
+
 	if (dev->num_tc) {
 		/* Do not allow XPS on subordinate device directly */
 		num_tc = dev->num_tc;
-- 
cgit v1.2.3-70-g09d2


From 1cc6571f562774f1d928dc8b3cff50829b86e970 Mon Sep 17 00:00:00 2001
From: Nick Child <nnac123@linux.ibm.com>
Date: Tue, 21 Mar 2023 10:07:25 -0500
Subject: netdev: Enforce index cap in netdev_get_tx_queue

When requesting a TX queue at a given index, warn on out-of-bounds
referencing if the index is greater than the allocated number of
queues.

Specifically, since this function is used heavily in the networking
stack use DEBUG_NET_WARN_ON_ONCE to avoid executing a new branch on
every packet.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://lore.kernel.org/r/20230321150725.127229-2-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 7621c512765f..674ee5daa7b1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2484,6 +2484,7 @@ static inline
 struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
 					 unsigned int index)
 {
+	DEBUG_NET_WARN_ON_ONCE(index >= dev->num_tx_queues);
 	return &dev->_tx[index];
 }
 
-- 
cgit v1.2.3-70-g09d2


From 99d0f3a1095f4c938b1665025c29411edafe8a01 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 21 Mar 2023 15:58:44 +0000
Subject: net: dpaa2-mac: use Autoneg bit rather than an_enabled

The Autoneg bit in the advertising bitmap and state->an_enabled are
always identical. Thus, we will be removing state->an_enabled.

Use the Autoneg bit in the advertising bitmap to indicate whether
autonegotiation should be used, rather than using the an_enabled
member which will be going away. This means we use the same condition
as phylink_mii_c22_pcs_config().

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c
index c886f33f8c6f..b1871e6c4006 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c
@@ -159,7 +159,8 @@ static void dpaa2_mac_config(struct phylink_config *config, unsigned int mode,
 	struct dpmac_link_state *dpmac_state = &mac->state;
 	int err;
 
-	if (state->an_enabled)
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			      state->advertising))
 		dpmac_state->options |= DPMAC_LINK_OPT_AUTONEG;
 	else
 		dpmac_state->options &= ~DPMAC_LINK_OPT_AUTONEG;
-- 
cgit v1.2.3-70-g09d2


From 459fd2f11204c962e3153020f4f56748e0e10afb Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 21 Mar 2023 15:58:49 +0000
Subject: net: pcs: xpcs: use Autoneg bit rather than an_enabled

The Autoneg bit in the advertising bitmap and state->an_enabled are
always identical. Thus, we will be removing state->an_enabled.

Use the Autoneg bit in the advertising bitmap to indicate whether
autonegotiation should be used, rather than using the an_enabled
member which will be going away.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/pcs/pcs-xpcs.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index 04a685353041..539cd43eae8d 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -923,6 +923,7 @@ static int xpcs_get_state_c73(struct dw_xpcs *xpcs,
 			      struct phylink_link_state *state,
 			      const struct xpcs_compat *compat)
 {
+	bool an_enabled;
 	int ret;
 
 	/* Link needs to be read first ... */
@@ -940,11 +941,13 @@ static int xpcs_get_state_c73(struct dw_xpcs *xpcs,
 		return xpcs_do_config(xpcs, state->interface, MLO_AN_INBAND, NULL);
 	}
 
-	if (state->an_enabled && xpcs_aneg_done_c73(xpcs, state, compat)) {
+	an_enabled = linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+				       state->advertising);
+	if (an_enabled && xpcs_aneg_done_c73(xpcs, state, compat)) {
 		state->an_complete = true;
 		xpcs_read_lpa_c73(xpcs, state);
 		xpcs_resolve_lpa_c73(xpcs, state);
-	} else if (state->an_enabled) {
+	} else if (an_enabled) {
 		state->link = 0;
 	} else if (state->link) {
 		xpcs_resolve_pma(xpcs, state);
@@ -999,7 +1002,8 @@ static int xpcs_get_state_c37_1000basex(struct dw_xpcs *xpcs,
 {
 	int lpa, bmsr;
 
-	if (state->an_enabled) {
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			      state->advertising)) {
 		/* Reset link state */
 		state->link = false;
 
-- 
cgit v1.2.3-70-g09d2


From 4ee9b0dcf09f426fbad7ed132d73ea2ba379d8ee Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 21 Mar 2023 15:58:54 +0000
Subject: net: phylink: remove an_enabled

The Autoneg bit in the advertising bitmap and state->an_enabled are
always identical. state->an_enabled is now no longer used by any
drivers, so lets kill this duplication.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phylink.c | 37 +++++++++++++++++--------------------
 include/linux/phylink.h   |  2 --
 2 files changed, 17 insertions(+), 22 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 1a2f074685fa..f7da96f0c75b 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -843,7 +843,6 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 		phylink_set(pl->supported, Autoneg);
 		phylink_set(pl->supported, Asym_Pause);
 		phylink_set(pl->supported, Pause);
-		pl->link_config.an_enabled = true;
 		pl->cfg_link_an_mode = MLO_AN_INBAND;
 
 		switch (pl->link_config.interface) {
@@ -945,9 +944,6 @@ static int phylink_parse_mode(struct phylink *pl, struct fwnode_handle *fwnode)
 				    "failed to validate link configuration for in-band status\n");
 			return -EINVAL;
 		}
-
-		/* Check if MAC/PCS also supports Autoneg. */
-		pl->link_config.an_enabled = phylink_test(pl->supported, Autoneg);
 	}
 
 	return 0;
@@ -957,7 +953,8 @@ static void phylink_apply_manual_flow(struct phylink *pl,
 				      struct phylink_link_state *state)
 {
 	/* If autoneg is disabled, pause AN is also disabled */
-	if (!state->an_enabled)
+	if (!linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			       state->advertising))
 		state->pause &= ~MLO_PAUSE_AN;
 
 	/* Manual configuration of pause modes */
@@ -997,21 +994,22 @@ static void phylink_mac_config(struct phylink *pl,
 			       const struct phylink_link_state *state)
 {
 	phylink_dbg(pl,
-		    "%s: mode=%s/%s/%s/%s/%s adv=%*pb pause=%02x link=%u an=%u\n",
+		    "%s: mode=%s/%s/%s/%s/%s adv=%*pb pause=%02x link=%u\n",
 		    __func__, phylink_an_mode_str(pl->cur_link_an_mode),
 		    phy_modes(state->interface),
 		    phy_speed_to_str(state->speed),
 		    phy_duplex_to_str(state->duplex),
 		    phy_rate_matching_to_str(state->rate_matching),
 		    __ETHTOOL_LINK_MODE_MASK_NBITS, state->advertising,
-		    state->pause, state->link, state->an_enabled);
+		    state->pause, state->link);
 
 	pl->mac_ops->mac_config(pl->config, pl->cur_link_an_mode, state);
 }
 
 static void phylink_mac_pcs_an_restart(struct phylink *pl)
 {
-	if (pl->link_config.an_enabled &&
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			      pl->link_config.advertising) &&
 	    phy_interface_mode_is_8023z(pl->link_config.interface) &&
 	    phylink_autoneg_inband(pl->cur_link_an_mode)) {
 		if (pl->pcs)
@@ -1138,9 +1136,9 @@ static void phylink_mac_pcs_get_state(struct phylink *pl,
 	linkmode_copy(state->advertising, pl->link_config.advertising);
 	linkmode_zero(state->lp_advertising);
 	state->interface = pl->link_config.interface;
-	state->an_enabled = pl->link_config.an_enabled;
 	state->rate_matching = pl->link_config.rate_matching;
-	if (state->an_enabled) {
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			      state->advertising)) {
 		state->speed = SPEED_UNKNOWN;
 		state->duplex = DUPLEX_UNKNOWN;
 		state->pause = MLO_PAUSE_NONE;
@@ -1531,7 +1529,6 @@ struct phylink *phylink_create(struct phylink_config *config,
 	pl->link_config.pause = MLO_PAUSE_AN;
 	pl->link_config.speed = SPEED_UNKNOWN;
 	pl->link_config.duplex = DUPLEX_UNKNOWN;
-	pl->link_config.an_enabled = true;
 	pl->mac_ops = mac_ops;
 	__set_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state);
 	timer_setup(&pl->link_poll, phylink_fixed_poll, 0);
@@ -2136,8 +2133,9 @@ static void phylink_get_ksettings(const struct phylink_link_state *state,
 		kset->base.speed = state->speed;
 		kset->base.duplex = state->duplex;
 	}
-	kset->base.autoneg = state->an_enabled ? AUTONEG_ENABLE :
-				AUTONEG_DISABLE;
+	kset->base.autoneg = linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+					       state->advertising) ?
+				AUTONEG_ENABLE : AUTONEG_DISABLE;
 }
 
 /**
@@ -2284,9 +2282,8 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 	/* We have ruled out the case with a PHY attached, and the
 	 * fixed-link cases.  All that is left are in-band links.
 	 */
-	config.an_enabled = kset->base.autoneg == AUTONEG_ENABLE;
 	linkmode_mod_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, config.advertising,
-			 config.an_enabled);
+			 kset->base.autoneg == AUTONEG_ENABLE);
 
 	/* If this link is with an SFP, ensure that changes to advertised modes
 	 * also cause the associated interface to be selected such that the
@@ -2320,13 +2317,14 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 	}
 
 	/* If autonegotiation is enabled, we must have an advertisement */
-	if (config.an_enabled && phylink_is_empty_linkmode(config.advertising))
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			      config.advertising) &&
+	    phylink_is_empty_linkmode(config.advertising))
 		return -EINVAL;
 
 	mutex_lock(&pl->state_mutex);
 	pl->link_config.speed = config.speed;
 	pl->link_config.duplex = config.duplex;
-	pl->link_config.an_enabled = config.an_enabled;
 
 	if (pl->link_config.interface != config.interface) {
 		/* The interface changed, e.g. 1000base-X <-> 2500base-X */
@@ -2932,7 +2930,6 @@ static int phylink_sfp_config_phy(struct phylink *pl, u8 mode,
 	config.speed = SPEED_UNKNOWN;
 	config.duplex = DUPLEX_UNKNOWN;
 	config.pause = MLO_PAUSE_AN;
-	config.an_enabled = pl->link_config.an_enabled;
 
 	/* Ignore errors if we're expecting a PHY to attach later */
 	ret = phylink_validate(pl, support, &config);
@@ -3001,7 +2998,6 @@ static int phylink_sfp_config_optical(struct phylink *pl)
 	config.speed = SPEED_UNKNOWN;
 	config.duplex = DUPLEX_UNKNOWN;
 	config.pause = MLO_PAUSE_AN;
-	config.an_enabled = true;
 
 	/* For all the interfaces that are supported, reduce the sfp_support
 	 * mask to only those link modes that can be supported.
@@ -3300,7 +3296,8 @@ void phylink_mii_c22_pcs_decode_state(struct phylink_link_state *state,
 	/* If there is no link or autonegotiation is disabled, the LP advertisement
 	 * data is not meaningful, so don't go any further.
 	 */
-	if (!state->link || !state->an_enabled)
+	if (!state->link || !linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+					       state->advertising))
 		return;
 
 	switch (state->interface) {
diff --git a/include/linux/phylink.h b/include/linux/phylink.h
index c492c26202b5..9ff56b050584 100644
--- a/include/linux/phylink.h
+++ b/include/linux/phylink.h
@@ -93,7 +93,6 @@ static inline bool phylink_autoneg_inband(unsigned int mode)
  *   the medium link mode (@speed and @duplex) and the speed/duplex of the phy
  *   interface mode (@interface) are different.
  * @link: true if the link is up.
- * @an_enabled: true if autonegotiation is enabled/desired.
  * @an_complete: true if autonegotiation has completed.
  */
 struct phylink_link_state {
@@ -105,7 +104,6 @@ struct phylink_link_state {
 	int pause;
 	int rate_matching;
 	unsigned int link:1;
-	unsigned int an_enabled:1;
 	unsigned int an_complete:1;
 };
 
-- 
cgit v1.2.3-70-g09d2


From f5fca219ad4548bc45f0221f9857ad22cb8136a1 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 21 Mar 2023 16:45:17 +0000
Subject: net: do not use skb_mac_header() in qdisc_pkt_len_init()

We want to remove our use of skb_mac_header() in tx paths,
eg remove skb_reset_mac_header() from __dev_queue_xmit().

Idea is that ndo_start_xmit() can get the mac header
simply looking at skb->data.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/dev.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index c278beee6792..7172334a418f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3735,25 +3735,25 @@ static void qdisc_pkt_len_init(struct sk_buff *skb)
 	 * we add to pkt_len the headers size of all segments
 	 */
 	if (shinfo->gso_size && skb_transport_header_was_set(skb)) {
-		unsigned int hdr_len;
 		u16 gso_segs = shinfo->gso_segs;
+		unsigned int hdr_len;
 
 		/* mac layer + network layer */
-		hdr_len = skb_transport_header(skb) - skb_mac_header(skb);
+		hdr_len = skb_transport_offset(skb);
 
 		/* + transport layer */
 		if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) {
 			const struct tcphdr *th;
 			struct tcphdr _tcphdr;
 
-			th = skb_header_pointer(skb, skb_transport_offset(skb),
+			th = skb_header_pointer(skb, hdr_len,
 						sizeof(_tcphdr), &_tcphdr);
 			if (likely(th))
 				hdr_len += __tcp_hdrlen(th);
 		} else {
 			struct udphdr _udphdr;
 
-			if (skb_header_pointer(skb, skb_transport_offset(skb),
+			if (skb_header_pointer(skb, hdr_len,
 					       sizeof(_udphdr), &_udphdr))
 				hdr_len += sizeof(struct udphdr);
 		}
-- 
cgit v1.2.3-70-g09d2


From e495a9673caf7619fc234b6808814bdf5072e0a6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 21 Mar 2023 16:45:18 +0000
Subject: sch_cake: do not use skb_mac_header() in cake_overhead()

We want to remove our use of skb_mac_header() in tx paths,
eg remove skb_reset_mac_header() from __dev_queue_xmit().

Idea is that ndo_start_xmit() can get the mac header
simply looking at skb->data.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/sch_cake.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index 7970217b565a..891e007d5c0b 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -1360,7 +1360,7 @@ static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb)
 		return cake_calc_overhead(q, len, off);
 
 	/* borrowed from qdisc_pkt_len_init() */
-	hdr_len = skb_transport_header(skb) - skb_mac_header(skb);
+	hdr_len = skb_transport_offset(skb);
 
 	/* + transport layer */
 	if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 |
@@ -1368,14 +1368,14 @@ static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb)
 		const struct tcphdr *th;
 		struct tcphdr _tcphdr;
 
-		th = skb_header_pointer(skb, skb_transport_offset(skb),
+		th = skb_header_pointer(skb, hdr_len,
 					sizeof(_tcphdr), &_tcphdr);
 		if (likely(th))
 			hdr_len += __tcp_hdrlen(th);
 	} else {
 		struct udphdr _udphdr;
 
-		if (skb_header_pointer(skb, skb_transport_offset(skb),
+		if (skb_header_pointer(skb, hdr_len,
 				       sizeof(_udphdr), &_udphdr))
 			hdr_len += sizeof(struct udphdr);
 	}
-- 
cgit v1.2.3-70-g09d2


From b3be94885af458f78bf92ceeaa37722d5c74c1e6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 21 Mar 2023 16:45:19 +0000
Subject: net/sched: remove two skb_mac_header() uses

tcf_mirred_act() and tcf_mpls_act() can use skb_network_offset()
instead of relying on skb_mac_header().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/act_mirred.c | 2 +-
 net/sched/act_mpls.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index 8037ec9b1d31..ec43764e92e7 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -295,7 +295,7 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb,
 	at_nh = skb->data == skb_network_header(skb);
 	if (at_nh != expects_nh) {
 		mac_len = skb_at_tc_ingress(skb) ? skb->mac_len :
-			  skb_network_header(skb) - skb_mac_header(skb);
+			  skb_network_offset(skb);
 		if (expects_nh) {
 			/* target device/action expect data at nh */
 			skb_pull_rcsum(skb2, mac_len);
diff --git a/net/sched/act_mpls.c b/net/sched/act_mpls.c
index 809f7928a1be..1010dc632874 100644
--- a/net/sched/act_mpls.c
+++ b/net/sched/act_mpls.c
@@ -69,7 +69,7 @@ TC_INDIRECT_SCOPE int tcf_mpls_act(struct sk_buff *skb,
 		skb_push_rcsum(skb, skb->mac_len);
 		mac_len = skb->mac_len;
 	} else {
-		mac_len = skb_network_header(skb) - skb_mac_header(skb);
+		mac_len = skb_network_offset(skb);
 	}
 
 	ret = READ_ONCE(m->tcf_action);
-- 
cgit v1.2.3-70-g09d2


From 8110633db49d7de2e4852f697322e1f3f8e61b04 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 21 Mar 2023 16:58:46 +0000
Subject: net: sfp-bus: allow SFP quirks to override Autoneg and pause bits

Allow SFP quirks to override the Autoneg, Pause and Asym_Pause bits in
the support mask.

Some modules have an inaccessible PHY on which is only accessible via
2500base-X without Autonegotiation. We therefore want to be able to
clear the Autoneg bit. Rearrange sfp_parse_support() to allow a SFP
modes quirk to override this bit.

Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/sfp-bus.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c
index daac293e8ede..1dd50f2ca05d 100644
--- a/drivers/net/phy/sfp-bus.c
+++ b/drivers/net/phy/sfp-bus.c
@@ -151,6 +151,10 @@ void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 	unsigned int br_min, br_nom, br_max;
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(modes) = { 0, };
 
+	phylink_set(modes, Autoneg);
+	phylink_set(modes, Pause);
+	phylink_set(modes, Asym_Pause);
+
 	/* Decode the bitrate information to MBd */
 	br_min = br_nom = br_max = 0;
 	if (id->base.br_nominal) {
@@ -329,10 +333,6 @@ void sfp_parse_support(struct sfp_bus *bus, const struct sfp_eeprom_id *id,
 		bus->sfp_quirk->modes(id, modes, interfaces);
 
 	linkmode_or(support, support, modes);
-
-	phylink_set(support, Autoneg);
-	phylink_set(support, Pause);
-	phylink_set(support, Asym_Pause);
 }
 EXPORT_SYMBOL_GPL(sfp_parse_support);
 
-- 
cgit v1.2.3-70-g09d2


From 50e96acbe11667b6fe9d99e1348c6c224b2f11dd Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 21 Mar 2023 16:58:51 +0000
Subject: net: sfp: add quirk for 2.5G copper SFP

Add a quirk for a copper SFP that identifies itself as "OEM"
"SFP-2.5G-T". This module's PHY is inaccessible, and can only run
at 2500base-X with the host without negotiation. Add a quirk to
enable the 2500base-X interface mode with 2500base-T support, and
disable autonegotiation.

Reported-by: Frank Wunderlich <frank-w@public-files.de>
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/sfp.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 39e3095796d0..9c1fa0b1737f 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -360,6 +360,23 @@ static void sfp_quirk_2500basex(const struct sfp_eeprom_id *id,
 	__set_bit(PHY_INTERFACE_MODE_2500BASEX, interfaces);
 }
 
+static void sfp_quirk_disable_autoneg(const struct sfp_eeprom_id *id,
+				      unsigned long *modes,
+				      unsigned long *interfaces)
+{
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, modes);
+}
+
+static void sfp_quirk_oem_2_5g(const struct sfp_eeprom_id *id,
+			       unsigned long *modes,
+			       unsigned long *interfaces)
+{
+	/* Copper 2.5G SFP */
+	linkmode_set_bit(ETHTOOL_LINK_MODE_2500baseT_Full_BIT, modes);
+	__set_bit(PHY_INTERFACE_MODE_2500BASEX, interfaces);
+	sfp_quirk_disable_autoneg(id, modes, interfaces);
+}
+
 static void sfp_quirk_ubnt_uf_instant(const struct sfp_eeprom_id *id,
 				      unsigned long *modes,
 				      unsigned long *interfaces)
@@ -401,6 +418,7 @@ static const struct sfp_quirk sfp_quirks[] = {
 	SFP_QUIRK_M("UBNT", "UF-INSTANT", sfp_quirk_ubnt_uf_instant),
 
 	SFP_QUIRK_F("OEM", "SFP-10G-T", sfp_fixup_rollball_cc),
+	SFP_QUIRK_M("OEM", "SFP-2.5G-T", sfp_quirk_oem_2_5g),
 	SFP_QUIRK_F("OEM", "RTSFP-10", sfp_fixup_rollball_cc),
 	SFP_QUIRK_F("OEM", "RTSFP-10G", sfp_fixup_rollball_cc),
 	SFP_QUIRK_F("Turris", "RTSFP-10", sfp_fixup_rollball),
-- 
cgit v1.2.3-70-g09d2


From ed4c7d6162895dcc9d32f1a16a42f13d67469383 Mon Sep 17 00:00:00 2001
From: Alex Elder <elder@linaro.org>
Date: Tue, 21 Mar 2023 13:26:42 -0500
Subject: net: ipa: add IPA v5.0 register definitions

Add the definitions of IPA register offsets and fields for IPA v5.0.
These are used for the SDX65 SoC.

In the Makefile, split IPA_VERSIONS to use IPA_REG_VERSIONS and
IPA_DATA_VERSIONS instead, to allow IPA register definitions for a
new version to be added separate from the IPA data.

Rename GSI_IPA_VERSIONS to be GSI_REG_VERSIONS for consistency.

Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ipa/Makefile           |  12 +-
 drivers/net/ipa/ipa_reg.c          |   2 +
 drivers/net/ipa/ipa_reg.h          |   1 +
 drivers/net/ipa/reg/ipa_reg-v5.0.c | 564 +++++++++++++++++++++++++++++++++++++
 4 files changed, 574 insertions(+), 5 deletions(-)
 create mode 100644 drivers/net/ipa/reg/ipa_reg-v5.0.c

diff --git a/drivers/net/ipa/Makefile b/drivers/net/ipa/Makefile
index cba199422f47..377e2c0fb89b 100644
--- a/drivers/net/ipa/Makefile
+++ b/drivers/net/ipa/Makefile
@@ -2,10 +2,12 @@
 #
 # Makefile for the Qualcomm IPA driver.
 
-IPA_VERSIONS		:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11
+IPA_REG_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11 5.0
 
 # Some IPA versions can reuse another set of GSI register definitions.
-GSI_IPA_VERSIONS	:=	3.1 3.5.1 4.0 4.5 4.9 4.11
+GSI_REG_VERSIONS	:=	3.1 3.5.1 4.0 4.5 4.9 4.11
+
+IPA_DATA_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11
 
 obj-$(CONFIG_QCOM_IPA)	+=	ipa.o
 
@@ -16,8 +18,8 @@ ipa-y			:=	ipa_main.o ipa_power.o ipa_reg.o ipa_mem.o \
 				ipa_resource.o ipa_qmi.o ipa_qmi_msg.o \
 				ipa_sysfs.o
 
-ipa-y			+=	$(GSI_IPA_VERSIONS:%=reg/gsi_reg-v%.o)
+ipa-y			+=	$(IPA_REG_VERSIONS:%=reg/ipa_reg-v%.o)
 
-ipa-y			+=	$(IPA_VERSIONS:%=reg/ipa_reg-v%.o)
+ipa-y			+=	$(GSI_REG_VERSIONS:%=reg/gsi_reg-v%.o)
 
-ipa-y			+=	$(IPA_VERSIONS:%=data/ipa_data-v%.o)
+ipa-y			+=	$(IPA_DATA_VERSIONS:%=data/ipa_data-v%.o)
diff --git a/drivers/net/ipa/ipa_reg.c b/drivers/net/ipa/ipa_reg.c
index 3f475428dddd..818a84f7c42d 100644
--- a/drivers/net/ipa/ipa_reg.c
+++ b/drivers/net/ipa/ipa_reg.c
@@ -123,6 +123,8 @@ static const struct regs *ipa_regs(enum ipa_version version)
 		return &ipa_regs_v4_9;
 	case IPA_VERSION_4_11:
 		return &ipa_regs_v4_11;
+	case IPA_VERSION_5_0:
+		return &ipa_regs_v5_0;
 	default:
 		return NULL;
 	}
diff --git a/drivers/net/ipa/ipa_reg.h b/drivers/net/ipa/ipa_reg.h
index 7dd65d39333d..3ac48dea865b 100644
--- a/drivers/net/ipa/ipa_reg.h
+++ b/drivers/net/ipa/ipa_reg.h
@@ -636,6 +636,7 @@ extern const struct regs ipa_regs_v4_5;
 extern const struct regs ipa_regs_v4_7;
 extern const struct regs ipa_regs_v4_9;
 extern const struct regs ipa_regs_v4_11;
+extern const struct regs ipa_regs_v5_0;
 
 const struct reg *ipa_reg(struct ipa *ipa, enum ipa_reg_id reg_id);
 
diff --git a/drivers/net/ipa/reg/ipa_reg-v5.0.c b/drivers/net/ipa/reg/ipa_reg-v5.0.c
new file mode 100644
index 000000000000..95e0edff4170
--- /dev/null
+++ b/drivers/net/ipa/reg/ipa_reg-v5.0.c
@@ -0,0 +1,564 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Copyright (C) 2023 Linaro Ltd. */
+
+#include <linux/types.h>
+
+#include "../ipa.h"
+#include "../ipa_reg.h"
+
+static const u32 reg_flavor_0_fmask[] = {
+	[MAX_PIPES]					= GENMASK(7, 0),
+	[MAX_CONS_PIPES]				= GENMASK(15, 8),
+	[MAX_PROD_PIPES]				= GENMASK(23, 16),
+	[PROD_LOWEST]					= GENMASK(31, 24),
+};
+
+REG_FIELDS(FLAVOR_0, flavor_0, 0x00000000);
+
+static const u32 reg_comp_cfg_fmask[] = {
+	[RAM_ARB_PRI_CLIENT_SAMP_FIX_DIS]		= BIT(0),
+	[GSI_SNOC_BYPASS_DIS]				= BIT(1),
+	[GEN_QMB_0_SNOC_BYPASS_DIS]			= BIT(2),
+	[GEN_QMB_1_SNOC_BYPASS_DIS]			= BIT(3),
+						/* Bit 4 reserved */
+	[IPA_QMB_SELECT_CONS_EN]			= BIT(5),
+	[IPA_QMB_SELECT_PROD_EN]			= BIT(6),
+	[GSI_MULTI_INORDER_RD_DIS]			= BIT(7),
+	[GSI_MULTI_INORDER_WR_DIS]			= BIT(8),
+	[GEN_QMB_0_MULTI_INORDER_RD_DIS]		= BIT(9),
+	[GEN_QMB_1_MULTI_INORDER_RD_DIS]		= BIT(10),
+	[GEN_QMB_0_MULTI_INORDER_WR_DIS]		= BIT(11),
+	[GEN_QMB_1_MULTI_INORDER_WR_DIS]		= BIT(12),
+	[GEN_QMB_0_SNOC_CNOC_LOOP_PROT_DIS]		= BIT(13),
+	[GSI_SNOC_CNOC_LOOP_PROT_DISABLE]		= BIT(14),
+	[GSI_MULTI_AXI_MASTERS_DIS]			= BIT(15),
+	[IPA_QMB_SELECT_GLOBAL_EN]			= BIT(16),
+	[FULL_FLUSH_WAIT_RS_CLOSURE_EN]			= BIT(17),
+						/* Bit 18 reserved */
+	[QMB_RAM_RD_CACHE_DISABLE]			= BIT(19),
+	[GENQMB_AOOOWR]					= BIT(20),
+	[IF_OUT_OF_BUF_STOP_RESET_MASK_EN]		= BIT(21),
+	[ATOMIC_FETCHER_ARB_LOCK_DIS]			= GENMASK(27, 22),
+						/* Bits 28-29 reserved */
+	[GEN_QMB_1_DYNAMIC_ASIZE]			= BIT(30),
+	[GEN_QMB_0_DYNAMIC_ASIZE]			= BIT(31),
+};
+
+REG_FIELDS(COMP_CFG, comp_cfg, 0x0000002c);
+
+static const u32 reg_clkon_cfg_fmask[] = {
+	[CLKON_RX]					= BIT(0),
+	[CLKON_PROC]					= BIT(1),
+	[TX_WRAPPER]					= BIT(2),
+	[CLKON_MISC]					= BIT(3),
+	[RAM_ARB]					= BIT(4),
+	[FTCH_HPS]					= BIT(5),
+	[FTCH_DPS]					= BIT(6),
+	[CLKON_HPS]					= BIT(7),
+	[CLKON_DPS]					= BIT(8),
+	[RX_HPS_CMDQS]					= BIT(9),
+	[HPS_DPS_CMDQS]					= BIT(10),
+	[DPS_TX_CMDQS]					= BIT(11),
+	[RSRC_MNGR]					= BIT(12),
+	[CTX_HANDLER]					= BIT(13),
+	[ACK_MNGR]					= BIT(14),
+	[D_DCPH]					= BIT(15),
+	[H_DCPH]					= BIT(16),
+						/* Bit 17 reserved */
+	[NTF_TX_CMDQS]					= BIT(18),
+	[CLKON_TX_0]					= BIT(19),
+	[CLKON_TX_1]					= BIT(20),
+	[CLKON_FNR]					= BIT(21),
+	[QSB2AXI_CMDQ_L]				= BIT(22),
+	[AGGR_WRAPPER]					= BIT(23),
+	[RAM_SLAVEWAY]					= BIT(24),
+	[CLKON_QMB]					= BIT(25),
+	[WEIGHT_ARB]					= BIT(26),
+	[GSI_IF]					= BIT(27),
+	[CLKON_GLOBAL]					= BIT(28),
+	[GLOBAL_2X_CLK]					= BIT(29),
+	[DPL_FIFO]					= BIT(30),
+	[DRBIP]						= BIT(31),
+};
+
+REG_FIELDS(CLKON_CFG, clkon_cfg, 0x00000034);
+
+static const u32 reg_route_fmask[] = {
+	[ROUTE_DEF_PIPE]				= GENMASK(7, 0),
+	[ROUTE_FRAG_DEF_PIPE]				= GENMASK(15, 8),
+	[ROUTE_DEF_HDR_OFST]				= GENMASK(25, 16),
+	[ROUTE_DEF_HDR_TABLE]				= BIT(26),
+	[ROUTE_DEF_RETAIN_HDR]				= BIT(27),
+	[ROUTE_DIS]					= BIT(28),
+						/* Bits 29-31 reserved */
+};
+
+REG_FIELDS(ROUTE, route, 0x00000038);
+
+static const u32 reg_shared_mem_size_fmask[] = {
+	[MEM_SIZE]					= GENMASK(15, 0),
+	[MEM_BADDR]					= GENMASK(31, 16),
+};
+
+REG_FIELDS(SHARED_MEM_SIZE, shared_mem_size, 0x00000040);
+
+static const u32 reg_qsb_max_writes_fmask[] = {
+	[GEN_QMB_0_MAX_WRITES]				= GENMASK(3, 0),
+	[GEN_QMB_1_MAX_WRITES]				= GENMASK(7, 4),
+						/* Bits 8-31 reserved */
+};
+
+REG_FIELDS(QSB_MAX_WRITES, qsb_max_writes, 0x00000054);
+
+static const u32 reg_qsb_max_reads_fmask[] = {
+	[GEN_QMB_0_MAX_READS]				= GENMASK(3, 0),
+	[GEN_QMB_1_MAX_READS]				= GENMASK(7, 4),
+						/* Bits 8-15 reserved */
+	[GEN_QMB_0_MAX_READS_BEATS]			= GENMASK(23, 16),
+	[GEN_QMB_1_MAX_READS_BEATS]			= GENMASK(31, 24),
+};
+
+REG_FIELDS(QSB_MAX_READS, qsb_max_reads, 0x00000058);
+
+/* Valid bits defined by ipa->available */
+
+REG_STRIDE(STATE_AGGR_ACTIVE, state_aggr_active, 0x00000100, 0x0004);
+
+static const u32 reg_filt_rout_cache_flush_fmask[] = {
+	[ROUTER_CACHE]					= BIT(0),
+						/* Bits 1-3 reserved */
+	[FILTER_CACHE]					= BIT(4),
+						/* Bits 5-31 reserved */
+};
+
+REG_FIELDS(FILT_ROUT_CACHE_FLUSH, filt_rout_cache_flush, 0x0000404);
+
+static const u32 reg_local_pkt_proc_cntxt_fmask[] = {
+	[IPA_BASE_ADDR]					= GENMASK(17, 0),
+						/* Bits 18-31 reserved */
+};
+
+/* Offset must be a multiple of 8 */
+REG_FIELDS(LOCAL_PKT_PROC_CNTXT, local_pkt_proc_cntxt, 0x00000478);
+
+static const u32 reg_ipa_tx_cfg_fmask[] = {
+						/* Bits 0-1 reserved */
+	[PREFETCH_ALMOST_EMPTY_SIZE_TX0]		= GENMASK(5, 2),
+	[DMAW_SCND_OUTSD_PRED_THRESHOLD]		= GENMASK(9, 6),
+	[DMAW_SCND_OUTSD_PRED_EN]			= BIT(10),
+	[DMAW_MAX_BEATS_256_DIS]			= BIT(11),
+	[PA_MASK_EN]					= BIT(12),
+	[PREFETCH_ALMOST_EMPTY_SIZE_TX1]		= GENMASK(16, 13),
+	[DUAL_TX_ENABLE]				= BIT(17),
+	[SSPND_PA_NO_START_STATE]			= BIT(18),
+						/* Bit 19 reserved */
+	[HOLB_STICKY_DROP_EN]				= BIT(20),
+						/* Bits 21-31 reserved */
+};
+
+REG_FIELDS(IPA_TX_CFG, ipa_tx_cfg, 0x00000488);
+
+static const u32 reg_idle_indication_cfg_fmask[] = {
+	[ENTER_IDLE_DEBOUNCE_THRESH]			= GENMASK(15, 0),
+	[CONST_NON_IDLE_ENABLE]				= BIT(16),
+						/* Bits 17-31 reserved */
+};
+
+REG_FIELDS(IDLE_INDICATION_CFG, idle_indication_cfg, 0x000004a8);
+
+static const u32 reg_qtime_timestamp_cfg_fmask[] = {
+	[DPL_TIMESTAMP_LSB]				= GENMASK(4, 0),
+						/* Bits 5-6 reserved */
+	[DPL_TIMESTAMP_SEL]				= BIT(7),
+	[TAG_TIMESTAMP_LSB]				= GENMASK(12, 8),
+						/* Bits 13-15 reserved */
+	[NAT_TIMESTAMP_LSB]				= GENMASK(20, 16),
+						/* Bits 21-31 reserved */
+};
+
+REG_FIELDS(QTIME_TIMESTAMP_CFG, qtime_timestamp_cfg, 0x000004ac);
+
+static const u32 reg_timers_xo_clk_div_cfg_fmask[] = {
+	[DIV_VALUE]					= GENMASK(8, 0),
+						/* Bits 9-30 reserved */
+	[DIV_ENABLE]					= BIT(31),
+};
+
+REG_FIELDS(TIMERS_XO_CLK_DIV_CFG, timers_xo_clk_div_cfg, 0x000004b0);
+
+static const u32 reg_timers_pulse_gran_cfg_fmask[] = {
+	[PULSE_GRAN_0]					= GENMASK(2, 0),
+	[PULSE_GRAN_1]					= GENMASK(5, 3),
+	[PULSE_GRAN_2]					= GENMASK(8, 6),
+	[PULSE_GRAN_3]					= GENMASK(11, 9),
+						/* Bits 12-31 reserved */
+};
+
+REG_FIELDS(TIMERS_PULSE_GRAN_CFG, timers_pulse_gran_cfg, 0x000004b4);
+
+static const u32 reg_src_rsrc_grp_01_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(SRC_RSRC_GRP_01_RSRC_TYPE, src_rsrc_grp_01_rsrc_type,
+		  0x00000500, 0x0020);
+
+static const u32 reg_src_rsrc_grp_23_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(SRC_RSRC_GRP_23_RSRC_TYPE, src_rsrc_grp_23_rsrc_type,
+		  0x00000504, 0x0020);
+
+static const u32 reg_src_rsrc_grp_45_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(SRC_RSRC_GRP_45_RSRC_TYPE, src_rsrc_grp_45_rsrc_type,
+		  0x00000508, 0x0020);
+
+static const u32 reg_src_rsrc_grp_67_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(SRC_RSRC_GRP_67_RSRC_TYPE, src_rsrc_grp_67_rsrc_type,
+		  0x0000050c, 0x0020);
+
+static const u32 reg_dst_rsrc_grp_01_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(DST_RSRC_GRP_01_RSRC_TYPE, dst_rsrc_grp_01_rsrc_type,
+		  0x00000600, 0x0020);
+
+static const u32 reg_dst_rsrc_grp_23_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(DST_RSRC_GRP_23_RSRC_TYPE, dst_rsrc_grp_23_rsrc_type,
+		  0x00000604, 0x0020);
+
+static const u32 reg_dst_rsrc_grp_45_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(DST_RSRC_GRP_45_RSRC_TYPE, dst_rsrc_grp_45_rsrc_type,
+		  0x00000608, 0x0020);
+
+static const u32 reg_dst_rsrc_grp_67_rsrc_type_fmask[] = {
+	[X_MIN_LIM]					= GENMASK(5, 0),
+						/* Bits 6-7 reserved */
+	[X_MAX_LIM]					= GENMASK(13, 8),
+						/* Bits 14-15 reserved */
+	[Y_MIN_LIM]					= GENMASK(21, 16),
+						/* Bits 22-23 reserved */
+	[Y_MAX_LIM]					= GENMASK(29, 24),
+						/* Bits 30-31 reserved */
+};
+
+REG_STRIDE_FIELDS(DST_RSRC_GRP_67_RSRC_TYPE, dst_rsrc_grp_67_rsrc_type,
+		  0x0000060c, 0x0020);
+
+/* Valid bits defined by ipa->available */
+
+REG_STRIDE(AGGR_FORCE_CLOSE, aggr_force_close, 0x000006b0, 0x0004);
+
+static const u32 reg_endp_init_cfg_fmask[] = {
+	[FRAG_OFFLOAD_EN]				= BIT(0),
+	[CS_OFFLOAD_EN]					= GENMASK(2, 1),
+	[CS_METADATA_HDR_OFFSET]			= GENMASK(6, 3),
+						/* Bit 7 reserved */
+	[CS_GEN_QMB_MASTER_SEL]				= BIT(8),
+						/* Bits 9-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_CFG, endp_init_cfg, 0x00001008, 0x0080);
+
+static const u32 reg_endp_init_nat_fmask[] = {
+	[NAT_EN]					= GENMASK(1, 0),
+						/* Bits 2-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_NAT, endp_init_nat, 0x0000100c, 0x0080);
+
+static const u32 reg_endp_init_hdr_fmask[] = {
+	[HDR_LEN]					= GENMASK(5, 0),
+	[HDR_OFST_METADATA_VALID]			= BIT(6),
+	[HDR_OFST_METADATA]				= GENMASK(12, 7),
+	[HDR_ADDITIONAL_CONST_LEN]			= GENMASK(18, 13),
+	[HDR_OFST_PKT_SIZE_VALID]			= BIT(19),
+	[HDR_OFST_PKT_SIZE]				= GENMASK(25, 20),
+						/* Bit 26 reserved */
+	[HDR_LEN_INC_DEAGG_HDR]				= BIT(27),
+	[HDR_LEN_MSB]					= GENMASK(29, 28),
+	[HDR_OFST_METADATA_MSB]				= GENMASK(31, 30),
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_HDR, endp_init_hdr, 0x00001010, 0x0080);
+
+static const u32 reg_endp_init_hdr_ext_fmask[] = {
+	[HDR_ENDIANNESS]				= BIT(0),
+	[HDR_TOTAL_LEN_OR_PAD_VALID]			= BIT(1),
+	[HDR_TOTAL_LEN_OR_PAD]				= BIT(2),
+	[HDR_PAYLOAD_LEN_INC_PADDING]			= BIT(3),
+	[HDR_TOTAL_LEN_OR_PAD_OFFSET]			= GENMASK(9, 4),
+	[HDR_PAD_TO_ALIGNMENT]				= GENMASK(13, 10),
+						/* Bits 14-15 reserved */
+	[HDR_TOTAL_LEN_OR_PAD_OFFSET_MSB]		= GENMASK(17, 16),
+	[HDR_OFST_PKT_SIZE_MSB]				= GENMASK(19, 18),
+	[HDR_ADDITIONAL_CONST_LEN_MSB]			= GENMASK(21, 20),
+	[HDR_BYTES_TO_REMOVE_VALID]			= BIT(22),
+						/* Bit 23 reserved */
+	[HDR_BYTES_TO_REMOVE]				= GENMASK(31, 24),
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_HDR_EXT, endp_init_hdr_ext, 0x00001014, 0x0080);
+
+REG_STRIDE(ENDP_INIT_HDR_METADATA_MASK, endp_init_hdr_metadata_mask,
+	   0x00001018, 0x0080);
+
+static const u32 reg_endp_init_mode_fmask[] = {
+	[ENDP_MODE]					= GENMASK(2, 0),
+	[DCPH_ENABLE]					= BIT(3),
+	[DEST_PIPE_INDEX]				= GENMASK(11, 4),
+	[BYTE_THRESHOLD]				= GENMASK(27, 12),
+	[PIPE_REPLICATION_EN]				= BIT(28),
+	[PAD_EN]					= BIT(29),
+	[DRBIP_ACL_ENABLE]				= BIT(30),
+						/* Bit 31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_MODE, endp_init_mode, 0x00001020, 0x0080);
+
+static const u32 reg_endp_init_aggr_fmask[] = {
+	[AGGR_EN]					= GENMASK(1, 0),
+	[AGGR_TYPE]					= GENMASK(4, 2),
+	[BYTE_LIMIT]					= GENMASK(10, 5),
+						/* Bit 11 reserved */
+	[TIME_LIMIT]					= GENMASK(16, 12),
+	[PKT_LIMIT]					= GENMASK(22, 17),
+	[SW_EOF_ACTIVE]					= BIT(23),
+	[FORCE_CLOSE]					= BIT(24),
+						/* Bit 25 reserved */
+	[HARD_BYTE_LIMIT_EN]				= BIT(26),
+	[AGGR_GRAN_SEL]					= BIT(27),
+						/* Bits 28-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_AGGR, endp_init_aggr, 0x00001024, 0x0080);
+
+static const u32 reg_endp_init_hol_block_en_fmask[] = {
+	[HOL_BLOCK_EN]					= BIT(0),
+						/* Bits 1-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_HOL_BLOCK_EN, endp_init_hol_block_en,
+		  0x0000102c, 0x0080);
+
+static const u32 reg_endp_init_hol_block_timer_fmask[] = {
+	[TIMER_LIMIT]					= GENMASK(4, 0),
+						/* Bits 5-7 reserved */
+	[TIMER_GRAN_SEL]				= GENMASK(9, 8),
+						/* Bits 10-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_HOL_BLOCK_TIMER, endp_init_hol_block_timer,
+		  0x00001030, 0x0080);
+
+static const u32 reg_endp_init_deaggr_fmask[] = {
+	[DEAGGR_HDR_LEN]				= GENMASK(5, 0),
+	[SYSPIPE_ERR_DETECTION]				= BIT(6),
+	[PACKET_OFFSET_VALID]				= BIT(7),
+	[PACKET_OFFSET_LOCATION]			= GENMASK(13, 8),
+	[IGNORE_MIN_PKT_ERR]				= BIT(14),
+						/* Bit 15 reserved */
+	[MAX_PACKET_LEN]				= GENMASK(31, 16),
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_DEAGGR, endp_init_deaggr, 0x00001034, 0x0080);
+
+static const u32 reg_endp_init_rsrc_grp_fmask[] = {
+	[ENDP_RSRC_GRP]					= GENMASK(2, 0),
+						/* Bits 3-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_RSRC_GRP, endp_init_rsrc_grp, 0x00001038, 0x0080);
+
+static const u32 reg_endp_init_seq_fmask[] = {
+	[SEQ_TYPE]					= GENMASK(7, 0),
+						/* Bits 8-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_INIT_SEQ, endp_init_seq, 0x0000103c, 0x0080);
+
+static const u32 reg_endp_status_fmask[] = {
+	[STATUS_EN]					= BIT(0),
+	[STATUS_ENDP]					= GENMASK(8, 1),
+	[STATUS_PKT_SUPPRESS]				= BIT(9),
+						/* Bits 10-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_STATUS, endp_status, 0x00001040, 0x0080);
+
+static const u32 reg_endp_filter_cache_cfg_fmask[] = {
+	[CACHE_MSK_SRC_ID]				= BIT(0),
+	[CACHE_MSK_SRC_IP]				= BIT(1),
+	[CACHE_MSK_DST_IP]				= BIT(2),
+	[CACHE_MSK_SRC_PORT]				= BIT(3),
+	[CACHE_MSK_DST_PORT]				= BIT(4),
+	[CACHE_MSK_PROTOCOL]				= BIT(5),
+	[CACHE_MSK_METADATA]				= BIT(6),
+						/* Bits 7-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_FILTER_CACHE_CFG, endp_filter_cache_cfg,
+		  0x0000105c, 0x0080);
+
+static const u32 reg_endp_router_cache_cfg_fmask[] = {
+	[CACHE_MSK_SRC_ID]				= BIT(0),
+	[CACHE_MSK_SRC_IP]				= BIT(1),
+	[CACHE_MSK_DST_IP]				= BIT(2),
+	[CACHE_MSK_SRC_PORT]				= BIT(3),
+	[CACHE_MSK_DST_PORT]				= BIT(4),
+	[CACHE_MSK_PROTOCOL]				= BIT(5),
+	[CACHE_MSK_METADATA]				= BIT(6),
+						/* Bits 7-31 reserved */
+};
+
+REG_STRIDE_FIELDS(ENDP_ROUTER_CACHE_CFG, endp_router_cache_cfg,
+		  0x00001070, 0x0080);
+
+/* Valid bits defined by enum ipa_irq_id; only used for GSI_EE_AP */
+REG(IPA_IRQ_STTS, ipa_irq_stts, 0x0000c008 + 0x1000 * GSI_EE_AP);
+
+/* Valid bits defined by enum ipa_irq_id; only used for GSI_EE_AP */
+REG(IPA_IRQ_EN, ipa_irq_en, 0x0000c00c + 0x1000 * GSI_EE_AP);
+
+/* Valid bits defined by enum ipa_irq_id; only used for GSI_EE_AP */
+REG(IPA_IRQ_CLR, ipa_irq_clr, 0x0000c010 + 0x1000 * GSI_EE_AP);
+
+static const u32 reg_ipa_irq_uc_fmask[] = {
+	[UC_INTR]					= BIT(0),
+						/* Bits 1-31 reserved */
+};
+
+REG_FIELDS(IPA_IRQ_UC, ipa_irq_uc, 0x0000c01c + 0x1000 * GSI_EE_AP);
+
+/* Valid bits defined by ipa->available */
+
+REG_STRIDE(IRQ_SUSPEND_INFO, irq_suspend_info,
+	   0x0000c030 + 0x1000 * GSI_EE_AP, 0x0004);
+
+/* Valid bits defined by ipa->available */
+
+REG_STRIDE(IRQ_SUSPEND_EN, irq_suspend_en,
+	   0x0000c050 + 0x1000 * GSI_EE_AP, 0x0004);
+
+/* Valid bits defined by ipa->available */
+
+REG_STRIDE(IRQ_SUSPEND_CLR, irq_suspend_clr,
+	   0x0000c070 + 0x1000 * GSI_EE_AP, 0x0004);
+
+static const struct reg *reg_array[] = {
+	[COMP_CFG]			= &reg_comp_cfg,
+	[CLKON_CFG]			= &reg_clkon_cfg,
+	[ROUTE]				= &reg_route,
+	[SHARED_MEM_SIZE]		= &reg_shared_mem_size,
+	[QSB_MAX_WRITES]		= &reg_qsb_max_writes,
+	[QSB_MAX_READS]			= &reg_qsb_max_reads,
+	[FILT_ROUT_CACHE_FLUSH]		= &reg_filt_rout_cache_flush,
+	[STATE_AGGR_ACTIVE]		= &reg_state_aggr_active,
+	[LOCAL_PKT_PROC_CNTXT]		= &reg_local_pkt_proc_cntxt,
+	[AGGR_FORCE_CLOSE]		= &reg_aggr_force_close,
+	[IPA_TX_CFG]			= &reg_ipa_tx_cfg,
+	[FLAVOR_0]			= &reg_flavor_0,
+	[IDLE_INDICATION_CFG]		= &reg_idle_indication_cfg,
+	[QTIME_TIMESTAMP_CFG]		= &reg_qtime_timestamp_cfg,
+	[TIMERS_XO_CLK_DIV_CFG]		= &reg_timers_xo_clk_div_cfg,
+	[TIMERS_PULSE_GRAN_CFG]		= &reg_timers_pulse_gran_cfg,
+	[SRC_RSRC_GRP_01_RSRC_TYPE]	= &reg_src_rsrc_grp_01_rsrc_type,
+	[SRC_RSRC_GRP_23_RSRC_TYPE]	= &reg_src_rsrc_grp_23_rsrc_type,
+	[SRC_RSRC_GRP_45_RSRC_TYPE]	= &reg_src_rsrc_grp_45_rsrc_type,
+	[SRC_RSRC_GRP_67_RSRC_TYPE]	= &reg_src_rsrc_grp_67_rsrc_type,
+	[DST_RSRC_GRP_01_RSRC_TYPE]	= &reg_dst_rsrc_grp_01_rsrc_type,
+	[DST_RSRC_GRP_23_RSRC_TYPE]	= &reg_dst_rsrc_grp_23_rsrc_type,
+	[DST_RSRC_GRP_45_RSRC_TYPE]	= &reg_dst_rsrc_grp_45_rsrc_type,
+	[DST_RSRC_GRP_67_RSRC_TYPE]	= &reg_dst_rsrc_grp_67_rsrc_type,
+	[ENDP_INIT_CFG]			= &reg_endp_init_cfg,
+	[ENDP_INIT_NAT]			= &reg_endp_init_nat,
+	[ENDP_INIT_HDR]			= &reg_endp_init_hdr,
+	[ENDP_INIT_HDR_EXT]		= &reg_endp_init_hdr_ext,
+	[ENDP_INIT_HDR_METADATA_MASK]	= &reg_endp_init_hdr_metadata_mask,
+	[ENDP_INIT_MODE]		= &reg_endp_init_mode,
+	[ENDP_INIT_AGGR]		= &reg_endp_init_aggr,
+	[ENDP_INIT_HOL_BLOCK_EN]	= &reg_endp_init_hol_block_en,
+	[ENDP_INIT_HOL_BLOCK_TIMER]	= &reg_endp_init_hol_block_timer,
+	[ENDP_INIT_DEAGGR]		= &reg_endp_init_deaggr,
+	[ENDP_INIT_RSRC_GRP]		= &reg_endp_init_rsrc_grp,
+	[ENDP_INIT_SEQ]			= &reg_endp_init_seq,
+	[ENDP_STATUS]			= &reg_endp_status,
+	[ENDP_FILTER_CACHE_CFG]		= &reg_endp_filter_cache_cfg,
+	[ENDP_ROUTER_CACHE_CFG]		= &reg_endp_router_cache_cfg,
+	[IPA_IRQ_STTS]			= &reg_ipa_irq_stts,
+	[IPA_IRQ_EN]			= &reg_ipa_irq_en,
+	[IPA_IRQ_CLR]			= &reg_ipa_irq_clr,
+	[IPA_IRQ_UC]			= &reg_ipa_irq_uc,
+	[IRQ_SUSPEND_INFO]		= &reg_irq_suspend_info,
+	[IRQ_SUSPEND_EN]		= &reg_irq_suspend_en,
+	[IRQ_SUSPEND_CLR]		= &reg_irq_suspend_clr,
+};
+
+const struct regs ipa_regs_v5_0 = {
+	.reg_count	= ARRAY_SIZE(reg_array),
+	.reg		= reg_array,
+};
-- 
cgit v1.2.3-70-g09d2


From faf0678ec8a0aa9039d8b188d012206abd67dd5c Mon Sep 17 00:00:00 2001
From: Alex Elder <elder@linaro.org>
Date: Tue, 21 Mar 2023 13:26:43 -0500
Subject: net: ipa: add IPA v5.0 GSI register definitions

Add the definitions of GSI register offsets and fields for IPA v5.0.
These are used for the SDX65 SoC.  Increase the maximum channel and
event ring counts supported by the driver, so those implemented by
the SDX65 are supported.

Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ipa/Makefile           |   2 +-
 drivers/net/ipa/gsi.h              |   4 +-
 drivers/net/ipa/gsi_reg.c          |   3 +
 drivers/net/ipa/gsi_reg.h          |   1 +
 drivers/net/ipa/reg/gsi_reg-v5.0.c | 317 +++++++++++++++++++++++++++++++++++++
 5 files changed, 324 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/ipa/reg/gsi_reg-v5.0.c

diff --git a/drivers/net/ipa/Makefile b/drivers/net/ipa/Makefile
index 377e2c0fb89b..514b7b6cd073 100644
--- a/drivers/net/ipa/Makefile
+++ b/drivers/net/ipa/Makefile
@@ -5,7 +5,7 @@
 IPA_REG_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11 5.0
 
 # Some IPA versions can reuse another set of GSI register definitions.
-GSI_REG_VERSIONS	:=	3.1 3.5.1 4.0 4.5 4.9 4.11
+GSI_REG_VERSIONS	:=	3.1 3.5.1 4.0 4.5 4.9 4.11 5.0
 
 IPA_DATA_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11
 
diff --git a/drivers/net/ipa/gsi.h b/drivers/net/ipa/gsi.h
index 50bc80cb167c..42063b227c18 100644
--- a/drivers/net/ipa/gsi.h
+++ b/drivers/net/ipa/gsi.h
@@ -16,8 +16,8 @@
 #include "ipa_version.h"
 
 /* Maximum number of channels and event rings supported by the driver */
-#define GSI_CHANNEL_COUNT_MAX	23
-#define GSI_EVT_RING_COUNT_MAX	24
+#define GSI_CHANNEL_COUNT_MAX	28
+#define GSI_EVT_RING_COUNT_MAX	28
 
 /* Maximum TLV FIFO size for a channel; 64 here is arbitrary (and high) */
 #define GSI_TLV_MAX		64
diff --git a/drivers/net/ipa/gsi_reg.c b/drivers/net/ipa/gsi_reg.c
index 1651fbad4bd5..c5458e28b12f 100644
--- a/drivers/net/ipa/gsi_reg.c
+++ b/drivers/net/ipa/gsi_reg.c
@@ -109,6 +109,9 @@ static const struct regs *gsi_regs(struct gsi *gsi)
 	case IPA_VERSION_4_11:
 		return &gsi_regs_v4_11;
 
+	case IPA_VERSION_5_0:
+		return &gsi_regs_v5_0;
+
 	default:
 		return NULL;
 	}
diff --git a/drivers/net/ipa/gsi_reg.h b/drivers/net/ipa/gsi_reg.h
index 48fde65fa2e8..cf046567f3fe 100644
--- a/drivers/net/ipa/gsi_reg.h
+++ b/drivers/net/ipa/gsi_reg.h
@@ -355,6 +355,7 @@ extern const struct regs gsi_regs_v4_0;
 extern const struct regs gsi_regs_v4_5;
 extern const struct regs gsi_regs_v4_9;
 extern const struct regs gsi_regs_v4_11;
+extern const struct regs gsi_regs_v5_0;
 
 /**
  * gsi_reg() - Return the structure describing a GSI register
diff --git a/drivers/net/ipa/reg/gsi_reg-v5.0.c b/drivers/net/ipa/reg/gsi_reg-v5.0.c
new file mode 100644
index 000000000000..d7b81a36d673
--- /dev/null
+++ b/drivers/net/ipa/reg/gsi_reg-v5.0.c
@@ -0,0 +1,317 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Copyright (C) 2023 Linaro Ltd. */
+
+#include <linux/types.h>
+
+#include "../gsi.h"
+#include "../reg.h"
+#include "../gsi_reg.h"
+
+REG(INTER_EE_SRC_CH_IRQ_MSK, inter_ee_src_ch_irq_msk,
+    0x0000c01c + 0x1000 * GSI_EE_AP);
+
+REG(INTER_EE_SRC_EV_CH_IRQ_MSK, inter_ee_src_ev_ch_irq_msk,
+    0x0000c028 + 0x1000 * GSI_EE_AP);
+
+static const u32 reg_ch_c_cntxt_0_fmask[] = {
+	[CHTYPE_PROTOCOL]				= GENMASK(6, 0),
+	[CHTYPE_DIR]					= BIT(7),
+	[CH_EE]						= GENMASK(11, 8),
+	[CHID]						= GENMASK(19, 12),
+	[CHSTATE]					= GENMASK(23, 20),
+	[ELEMENT_SIZE]					= GENMASK(31, 24),
+};
+
+REG_STRIDE_FIELDS(CH_C_CNTXT_0, ch_c_cntxt_0,
+		  0x00014000 + 0x12000 * GSI_EE_AP, 0x80);
+
+static const u32 reg_ch_c_cntxt_1_fmask[] = {
+	[CH_R_LENGTH]					= GENMASK(23, 0),
+	[ERINDEX]					= GENMASK(31, 24),
+};
+
+REG_STRIDE_FIELDS(CH_C_CNTXT_1, ch_c_cntxt_1,
+		  0x00014004 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_CNTXT_2, ch_c_cntxt_2, 0x00014008 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_CNTXT_3, ch_c_cntxt_3, 0x0001400c + 0x12000 * GSI_EE_AP, 0x80);
+
+static const u32 reg_ch_c_qos_fmask[] = {
+	[WRR_WEIGHT]					= GENMASK(3, 0),
+						/* Bits 4-7 reserved */
+	[MAX_PREFETCH]					= BIT(8),
+	[USE_DB_ENG]					= BIT(9),
+	[PREFETCH_MODE]					= GENMASK(13, 10),
+						/* Bits 14-15 reserved */
+	[EMPTY_LVL_THRSHOLD]				= GENMASK(23, 16),
+	[DB_IN_BYTES]					= BIT(24),
+	[LOW_LATENCY_EN]				= BIT(25),
+						/* Bits 26-31 reserved */
+};
+
+REG_STRIDE_FIELDS(CH_C_QOS, ch_c_qos, 0x00014048 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_SCRATCH_0, ch_c_scratch_0,
+	   0x0001404c + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_SCRATCH_1, ch_c_scratch_1,
+	   0x00014050 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_SCRATCH_2, ch_c_scratch_2,
+	   0x00014054 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_SCRATCH_3, ch_c_scratch_3,
+	   0x00014058 + 0x12000 * GSI_EE_AP, 0x80);
+
+static const u32 reg_ev_ch_e_cntxt_0_fmask[] = {
+	[EV_CHTYPE]					= GENMASK(6, 0),
+	[EV_INTYPE]					= BIT(7),
+	[EV_EVCHID]					= GENMASK(15, 8),
+	[EV_EE]						= GENMASK(19, 16),
+	[EV_CHSTATE]					= GENMASK(23, 20),
+	[EV_ELEMENT_SIZE]				= GENMASK(31, 24),
+};
+
+REG_STRIDE_FIELDS(EV_CH_E_CNTXT_0, ev_ch_e_cntxt_0,
+		  0x0001c000 + 0x12000 * GSI_EE_AP, 0x80);
+
+static const u32 reg_ev_ch_e_cntxt_1_fmask[] = {
+	[R_LENGTH]					= GENMASK(19, 0),
+};
+
+REG_STRIDE_FIELDS(EV_CH_E_CNTXT_1, ev_ch_e_cntxt_1,
+		  0x0001c004 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_2, ev_ch_e_cntxt_2,
+	   0x0001c008 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_3, ev_ch_e_cntxt_3,
+	   0x0001c00c + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_4, ev_ch_e_cntxt_4,
+	   0x0001c010 + 0x12000 * GSI_EE_AP, 0x80);
+
+static const u32 reg_ev_ch_e_cntxt_8_fmask[] = {
+	[EV_MODT]					= GENMASK(15, 0),
+	[EV_MODC]					= GENMASK(23, 16),
+	[EV_MOD_CNT]					= GENMASK(31, 24),
+};
+
+REG_STRIDE_FIELDS(EV_CH_E_CNTXT_8, ev_ch_e_cntxt_8,
+		  0x0001c020 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_9, ev_ch_e_cntxt_9,
+	   0x0001c024 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_10, ev_ch_e_cntxt_10,
+	   0x0001c028 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_11, ev_ch_e_cntxt_11,
+	   0x0001c02c + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_12, ev_ch_e_cntxt_12,
+	   0x0001c030 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_CNTXT_13, ev_ch_e_cntxt_13,
+	   0x0001c034 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_SCRATCH_0, ev_ch_e_scratch_0,
+	   0x0001c048 + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(EV_CH_E_SCRATCH_1, ev_ch_e_scratch_1,
+	   0x0001c04c + 0x12000 * GSI_EE_AP, 0x80);
+
+REG_STRIDE(CH_C_DOORBELL_0, ch_c_doorbell_0,
+	   0x00024000 + 0x12000 * GSI_EE_AP, 0x08);
+
+REG_STRIDE(EV_CH_E_DOORBELL_0, ev_ch_e_doorbell_0,
+	   0x00024800 + 0x12000 * GSI_EE_AP, 0x08);
+
+static const u32 reg_gsi_status_fmask[] = {
+	[ENABLED]					= BIT(0),
+						/* Bits 1-31 reserved */
+};
+
+REG_FIELDS(GSI_STATUS, gsi_status, 0x00025000 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_ch_cmd_fmask[] = {
+	[CH_CHID]					= GENMASK(7, 0),
+						/* Bits 8-23 reserved */
+	[CH_OPCODE]					= GENMASK(31, 24),
+};
+
+REG_FIELDS(CH_CMD, ch_cmd, 0x00025008 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_ev_ch_cmd_fmask[] = {
+	[EV_CHID]					= GENMASK(7, 0),
+						/* Bits 8-23 reserved */
+	[EV_OPCODE]					= GENMASK(31, 24),
+};
+
+REG_FIELDS(EV_CH_CMD, ev_ch_cmd, 0x00025010 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_generic_cmd_fmask[] = {
+	[GENERIC_OPCODE]				= GENMASK(4, 0),
+	[GENERIC_CHID]					= GENMASK(9, 5),
+	[GENERIC_EE]					= GENMASK(13, 10),
+						/* Bits 14-31 reserved */
+};
+
+REG_FIELDS(GENERIC_CMD, generic_cmd, 0x00025018 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_hw_param_2_fmask[] = {
+	[NUM_CH_PER_EE]					= GENMASK(7, 0),
+	[IRAM_SIZE]					= GENMASK(12, 8),
+	[GSI_CH_PEND_TRANSLATE]				= BIT(13),
+	[GSI_CH_FULL_LOGIC]				= BIT(14),
+	[GSI_USE_SDMA]					= BIT(15),
+	[GSI_SDMA_N_INT]				= GENMASK(18, 16),
+	[GSI_SDMA_MAX_BURST]				= GENMASK(26, 19),
+	[GSI_SDMA_N_IOVEC]				= GENMASK(29, 27),
+	[GSI_USE_RD_WR_ENG]				= BIT(30),
+	[GSI_USE_INTER_EE]				= BIT(31),
+};
+
+REG_FIELDS(HW_PARAM_2, hw_param_2, 0x00025040 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_hw_param_4_fmask[] = {
+	[EV_PER_EE]					= GENMASK(7, 0),
+	[IRAM_PROTOCOL_COUNT]				= GENMASK(15, 8),
+						/* Bits 16-31 reserved */
+};
+
+REG_FIELDS(HW_PARAM_4, hw_param_4, 0x00025050 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_TYPE_IRQ, cntxt_type_irq, 0x00025080 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_TYPE_IRQ_MSK, cntxt_type_irq_msk, 0x00025088 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_CH_IRQ, cntxt_src_ch_irq, 0x00025090 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_CH_IRQ_MSK, cntxt_src_ch_irq_msk,
+    0x00025094 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_CH_IRQ_CLR, cntxt_src_ch_irq_clr,
+    0x00025098 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_EV_CH_IRQ, cntxt_src_ev_ch_irq, 0x0002509c + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_EV_CH_IRQ_MSK, cntxt_src_ev_ch_irq_msk,
+    0x000250a0 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_EV_CH_IRQ_CLR, cntxt_src_ev_ch_irq_clr,
+    0x000250a4 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_IEOB_IRQ, cntxt_src_ieob_irq, 0x000250a8 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_IEOB_IRQ_MSK, cntxt_src_ieob_irq_msk,
+    0x000250ac + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_SRC_IEOB_IRQ_CLR, cntxt_src_ieob_irq_clr,
+    0x000250b0 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_GLOB_IRQ_STTS, cntxt_glob_irq_stts, 0x00025200 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_GLOB_IRQ_EN, cntxt_glob_irq_en, 0x00025204 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_GLOB_IRQ_CLR, cntxt_glob_irq_clr, 0x00025208 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_GSI_IRQ_STTS, cntxt_gsi_irq_stts, 0x0002520c + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_GSI_IRQ_EN, cntxt_gsi_irq_en, 0x00025210 + 0x12000 * GSI_EE_AP);
+
+REG(CNTXT_GSI_IRQ_CLR, cntxt_gsi_irq_clr, 0x00025214 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_cntxt_intset_fmask[] = {
+	[INTYPE]					= BIT(0)
+						/* Bits 1-31 reserved */
+};
+
+REG_FIELDS(CNTXT_INTSET, cntxt_intset, 0x00025220 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_error_log_fmask[] = {
+	[ERR_ARG3]					= GENMASK(3, 0),
+	[ERR_ARG2]					= GENMASK(7, 4),
+	[ERR_ARG1]					= GENMASK(11, 8),
+	[ERR_CODE]					= GENMASK(15, 12),
+						/* Bits 16-18 reserved */
+	[ERR_VIRT_IDX]					= GENMASK(23, 19),
+	[ERR_TYPE]					= GENMASK(27, 24),
+	[ERR_EE]					= GENMASK(31, 28),
+};
+
+REG_FIELDS(ERROR_LOG, error_log, 0x00025240 + 0x12000 * GSI_EE_AP);
+
+REG(ERROR_LOG_CLR, error_log_clr, 0x00025244 + 0x12000 * GSI_EE_AP);
+
+static const u32 reg_cntxt_scratch_0_fmask[] = {
+	[INTER_EE_RESULT]				= GENMASK(2, 0),
+						/* Bits 3-4 reserved */
+	[GENERIC_EE_RESULT]				= GENMASK(7, 5),
+						/* Bits 8-31 reserved */
+};
+
+REG_FIELDS(CNTXT_SCRATCH_0, cntxt_scratch_0, 0x00025400 + 0x12000 * GSI_EE_AP);
+
+static const struct reg *reg_array[] = {
+	[INTER_EE_SRC_CH_IRQ_MSK]	= &reg_inter_ee_src_ch_irq_msk,
+	[INTER_EE_SRC_EV_CH_IRQ_MSK]	= &reg_inter_ee_src_ev_ch_irq_msk,
+	[CH_C_CNTXT_0]			= &reg_ch_c_cntxt_0,
+	[CH_C_CNTXT_1]			= &reg_ch_c_cntxt_1,
+	[CH_C_CNTXT_2]			= &reg_ch_c_cntxt_2,
+	[CH_C_CNTXT_3]			= &reg_ch_c_cntxt_3,
+	[CH_C_QOS]			= &reg_ch_c_qos,
+	[CH_C_SCRATCH_0]		= &reg_ch_c_scratch_0,
+	[CH_C_SCRATCH_1]		= &reg_ch_c_scratch_1,
+	[CH_C_SCRATCH_2]		= &reg_ch_c_scratch_2,
+	[CH_C_SCRATCH_3]		= &reg_ch_c_scratch_3,
+	[EV_CH_E_CNTXT_0]		= &reg_ev_ch_e_cntxt_0,
+	[EV_CH_E_CNTXT_1]		= &reg_ev_ch_e_cntxt_1,
+	[EV_CH_E_CNTXT_2]		= &reg_ev_ch_e_cntxt_2,
+	[EV_CH_E_CNTXT_3]		= &reg_ev_ch_e_cntxt_3,
+	[EV_CH_E_CNTXT_4]		= &reg_ev_ch_e_cntxt_4,
+	[EV_CH_E_CNTXT_8]		= &reg_ev_ch_e_cntxt_8,
+	[EV_CH_E_CNTXT_9]		= &reg_ev_ch_e_cntxt_9,
+	[EV_CH_E_CNTXT_10]		= &reg_ev_ch_e_cntxt_10,
+	[EV_CH_E_CNTXT_11]		= &reg_ev_ch_e_cntxt_11,
+	[EV_CH_E_CNTXT_12]		= &reg_ev_ch_e_cntxt_12,
+	[EV_CH_E_CNTXT_13]		= &reg_ev_ch_e_cntxt_13,
+	[EV_CH_E_SCRATCH_0]		= &reg_ev_ch_e_scratch_0,
+	[EV_CH_E_SCRATCH_1]		= &reg_ev_ch_e_scratch_1,
+	[CH_C_DOORBELL_0]		= &reg_ch_c_doorbell_0,
+	[EV_CH_E_DOORBELL_0]		= &reg_ev_ch_e_doorbell_0,
+	[GSI_STATUS]			= &reg_gsi_status,
+	[CH_CMD]			= &reg_ch_cmd,
+	[EV_CH_CMD]			= &reg_ev_ch_cmd,
+	[GENERIC_CMD]			= &reg_generic_cmd,
+	[HW_PARAM_2]			= &reg_hw_param_2,
+	[HW_PARAM_4]			= &reg_hw_param_4,
+	[CNTXT_TYPE_IRQ]		= &reg_cntxt_type_irq,
+	[CNTXT_TYPE_IRQ_MSK]		= &reg_cntxt_type_irq_msk,
+	[CNTXT_SRC_CH_IRQ]		= &reg_cntxt_src_ch_irq,
+	[CNTXT_SRC_CH_IRQ_MSK]		= &reg_cntxt_src_ch_irq_msk,
+	[CNTXT_SRC_CH_IRQ_CLR]		= &reg_cntxt_src_ch_irq_clr,
+	[CNTXT_SRC_EV_CH_IRQ]		= &reg_cntxt_src_ev_ch_irq,
+	[CNTXT_SRC_EV_CH_IRQ_MSK]	= &reg_cntxt_src_ev_ch_irq_msk,
+	[CNTXT_SRC_EV_CH_IRQ_CLR]	= &reg_cntxt_src_ev_ch_irq_clr,
+	[CNTXT_SRC_IEOB_IRQ]		= &reg_cntxt_src_ieob_irq,
+	[CNTXT_SRC_IEOB_IRQ_MSK]	= &reg_cntxt_src_ieob_irq_msk,
+	[CNTXT_SRC_IEOB_IRQ_CLR]	= &reg_cntxt_src_ieob_irq_clr,
+	[CNTXT_GLOB_IRQ_STTS]		= &reg_cntxt_glob_irq_stts,
+	[CNTXT_GLOB_IRQ_EN]		= &reg_cntxt_glob_irq_en,
+	[CNTXT_GLOB_IRQ_CLR]		= &reg_cntxt_glob_irq_clr,
+	[CNTXT_GSI_IRQ_STTS]		= &reg_cntxt_gsi_irq_stts,
+	[CNTXT_GSI_IRQ_EN]		= &reg_cntxt_gsi_irq_en,
+	[CNTXT_GSI_IRQ_CLR]		= &reg_cntxt_gsi_irq_clr,
+	[CNTXT_INTSET]			= &reg_cntxt_intset,
+	[ERROR_LOG]			= &reg_error_log,
+	[ERROR_LOG_CLR]			= &reg_error_log_clr,
+	[CNTXT_SCRATCH_0]		= &reg_cntxt_scratch_0,
+};
+
+const struct regs gsi_regs_v5_0 = {
+	.reg_count	= ARRAY_SIZE(reg_array),
+	.reg		= reg_array,
+};
-- 
cgit v1.2.3-70-g09d2


From cb7550b44383195f8b5e9acacfb132b806875472 Mon Sep 17 00:00:00 2001
From: Alex Elder <elder@linaro.org>
Date: Tue, 21 Mar 2023 13:26:44 -0500
Subject: net: ipa: add IPA v5.0 configuration data

Add the configuration data required for IPA v5.0, which is used in
the SDX65 SoC.

Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ipa/Makefile             |   2 +-
 drivers/net/ipa/data/ipa_data-v5.0.c | 481 +++++++++++++++++++++++++++++++++++
 drivers/net/ipa/ipa_data.h           |   3 +-
 drivers/net/ipa/ipa_main.c           |   6 +-
 4 files changed, 489 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/ipa/data/ipa_data-v5.0.c

diff --git a/drivers/net/ipa/Makefile b/drivers/net/ipa/Makefile
index 514b7b6cd073..7293d5cc2b2b 100644
--- a/drivers/net/ipa/Makefile
+++ b/drivers/net/ipa/Makefile
@@ -7,7 +7,7 @@ IPA_REG_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11 5.0
 # Some IPA versions can reuse another set of GSI register definitions.
 GSI_REG_VERSIONS	:=	3.1 3.5.1 4.0 4.5 4.9 4.11 5.0
 
-IPA_DATA_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11
+IPA_DATA_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11 5.0
 
 obj-$(CONFIG_QCOM_IPA)	+=	ipa.o
 
diff --git a/drivers/net/ipa/data/ipa_data-v5.0.c b/drivers/net/ipa/data/ipa_data-v5.0.c
new file mode 100644
index 000000000000..4d8171dae4cd
--- /dev/null
+++ b/drivers/net/ipa/data/ipa_data-v5.0.c
@@ -0,0 +1,481 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Copyright (C) 2023 Linaro Ltd. */
+
+#include <linux/log2.h>
+
+#include "../gsi.h"
+#include "../ipa_data.h"
+#include "../ipa_endpoint.h"
+#include "../ipa_mem.h"
+
+/** enum ipa_resource_type - IPA resource types for an SoC having IPA v5.0 */
+enum ipa_resource_type {
+	/* Source resource types; first must have value 0 */
+	IPA_RESOURCE_TYPE_SRC_PKT_CONTEXTS		= 0,
+	IPA_RESOURCE_TYPE_SRC_DESCRIPTOR_LISTS,
+	IPA_RESOURCE_TYPE_SRC_DESCRIPTOR_BUFF,
+	IPA_RESOURCE_TYPE_SRC_HPS_DMARS,
+	IPA_RESOURCE_TYPE_SRC_ACK_ENTRIES,
+
+	/* Destination resource types; first must have value 0 */
+	IPA_RESOURCE_TYPE_DST_DATA_SECTORS		= 0,
+	IPA_RESOURCE_TYPE_DST_DPS_DMARS,
+	IPA_RESOURCE_TYPE_DST_ULSO_SEGMENTS,
+};
+
+/* Resource groups used for an SoC having IPA v5.0 */
+enum ipa_rsrc_group_id {
+	/* Source resource group identifiers */
+	IPA_RSRC_GROUP_SRC_UL				= 0,
+	IPA_RSRC_GROUP_SRC_DL,
+	IPA_RSRC_GROUP_SRC_UNUSED_2,
+	IPA_RSRC_GROUP_SRC_UNUSED_3,
+	IPA_RSRC_GROUP_SRC_URLLC,
+	IPA_RSRC_GROUP_SRC_U_RX_QC,
+	IPA_RSRC_GROUP_SRC_COUNT,	/* Last in set; not a source group */
+
+	/* Destination resource group identifiers */
+	IPA_RSRC_GROUP_DST_UL				= 0,
+	IPA_RSRC_GROUP_DST_DL,
+	IPA_RSRC_GROUP_DST_DMA,
+	IPA_RSRC_GROUP_DST_QDSS,
+	IPA_RSRC_GROUP_DST_CV2X,
+	IPA_RSRC_GROUP_DST_UC,
+	IPA_RSRC_GROUP_DST_DRB_IP,
+	IPA_RSRC_GROUP_DST_COUNT,	/* Last; not a destination group */
+};
+
+/* QSB configuration data for an SoC having IPA v5.0 */
+static const struct ipa_qsb_data ipa_qsb_data[] = {
+	[IPA_QSB_MASTER_DDR] = {
+		.max_writes		= 0,
+		.max_reads		= 0,	/* no limit (hardware max) */
+		.max_reads_beats	= 0,
+	},
+	[IPA_QSB_MASTER_PCIE] = {
+		.max_writes		= 0,
+		.max_reads		= 0,	/* no limit (hardware max) */
+		.max_reads_beats	= 0,
+	},
+};
+
+/* Endpoint configuration data for an SoC having IPA v5.0 */
+static const struct ipa_gsi_endpoint_data ipa_gsi_endpoint_data[] = {
+	[IPA_ENDPOINT_AP_COMMAND_TX] = {
+		.ee_id		= GSI_EE_AP,
+		.channel_id	= 12,
+		.endpoint_id	= 14,
+		.toward_ipa	= true,
+		.channel = {
+			.tre_count	= 256,
+			.event_count	= 256,
+			.tlv_count	= 20,
+		},
+		.endpoint = {
+			.config = {
+				.resource_group	= IPA_RSRC_GROUP_SRC_UL,
+				.dma_mode	= true,
+				.dma_endpoint	= IPA_ENDPOINT_AP_LAN_RX,
+				.tx = {
+					.seq_type = IPA_SEQ_DMA,
+				},
+			},
+		},
+	},
+	[IPA_ENDPOINT_AP_LAN_RX] = {
+		.ee_id		= GSI_EE_AP,
+		.channel_id	= 13,
+		.endpoint_id	= 16,
+		.toward_ipa	= false,
+		.channel = {
+			.tre_count	= 256,
+			.event_count	= 256,
+			.tlv_count	= 9,
+		},
+		.endpoint = {
+			.config = {
+				.resource_group	= IPA_RSRC_GROUP_DST_UL,
+				.aggregation	= true,
+				.status_enable	= true,
+				.rx = {
+					.buffer_size	= 8192,
+					.pad_align	= ilog2(sizeof(u32)),
+					.aggr_time_limit = 500,
+				},
+			},
+		},
+	},
+	[IPA_ENDPOINT_AP_MODEM_TX] = {
+		.ee_id		= GSI_EE_AP,
+		.channel_id	= 11,
+		.endpoint_id	= 2,
+		.toward_ipa	= true,
+		.channel = {
+			.tre_count	= 512,
+			.event_count	= 512,
+			.tlv_count	= 25,
+		},
+		.endpoint = {
+			.filter_support	= true,
+			.config = {
+				.resource_group	= IPA_RSRC_GROUP_SRC_UL,
+				.checksum       = true,
+				.qmap		= true,
+				.status_enable	= true,
+				.tx = {
+					.seq_type = IPA_SEQ_2_PASS_SKIP_LAST_UC,
+					.status_endpoint =
+						IPA_ENDPOINT_MODEM_AP_RX,
+				},
+			},
+		},
+	},
+	[IPA_ENDPOINT_AP_MODEM_RX] = {
+		.ee_id		= GSI_EE_AP,
+		.channel_id	= 1,
+		.endpoint_id	= 23,
+		.toward_ipa	= false,
+		.channel = {
+			.tre_count	= 256,
+			.event_count	= 256,
+			.tlv_count	= 9,
+		},
+		.endpoint = {
+			.config = {
+				.resource_group	= IPA_RSRC_GROUP_DST_DL,
+				.checksum       = true,
+				.qmap		= true,
+				.aggregation	= true,
+				.rx = {
+					.buffer_size	= 8192,
+					.aggr_time_limit = 500,
+					.aggr_close_eof	= true,
+				},
+			},
+		},
+	},
+	[IPA_ENDPOINT_MODEM_AP_TX] = {
+		.ee_id		= GSI_EE_MODEM,
+		.channel_id	= 0,
+		.endpoint_id	= 12,
+		.toward_ipa	= true,
+		.endpoint = {
+			.filter_support	= true,
+		},
+	},
+	[IPA_ENDPOINT_MODEM_AP_RX] = {
+		.ee_id		= GSI_EE_MODEM,
+		.channel_id	= 7,
+		.endpoint_id	= 21,
+		.toward_ipa	= false,
+	},
+	[IPA_ENDPOINT_MODEM_DL_NLO_TX] = {
+		.ee_id		= GSI_EE_MODEM,
+		.channel_id	= 2,
+		.endpoint_id	= 15,
+		.toward_ipa	= true,
+		.endpoint = {
+			.filter_support	= true,
+		},
+	},
+};
+
+/* Source resource configuration data for an SoC having IPA v5.0 */
+static const struct ipa_resource ipa_resource_src[] = {
+	[IPA_RESOURCE_TYPE_SRC_PKT_CONTEXTS] = {
+		.limits[IPA_RSRC_GROUP_SRC_UL] = {
+			.min = 3,	.max = 9,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_DL] = {
+			.min = 4,	.max = 10,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_URLLC] = {
+			.min = 1,	.max = 63,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_U_RX_QC] = {
+			.min = 0,	.max = 63,
+		},
+	},
+	[IPA_RESOURCE_TYPE_SRC_DESCRIPTOR_LISTS] = {
+		.limits[IPA_RSRC_GROUP_SRC_UL] = {
+			.min = 9,	.max = 9,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_DL] = {
+			.min = 12,	.max = 12,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_URLLC] = {
+			.min = 10,	.max = 10,
+		},
+	},
+	[IPA_RESOURCE_TYPE_SRC_DESCRIPTOR_BUFF] = {
+		.limits[IPA_RSRC_GROUP_SRC_UL] = {
+			.min = 9,	.max = 9,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_DL] = {
+			.min = 24,	.max = 24,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_URLLC] = {
+			.min = 20,	.max = 20,
+		},
+	},
+	[IPA_RESOURCE_TYPE_SRC_HPS_DMARS] = {
+		.limits[IPA_RSRC_GROUP_SRC_UL] = {
+			.min = 0,	.max = 63,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_DL] = {
+			.min = 0,	.max = 63,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_URLLC] = {
+			.min = 1,	.max = 63,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_U_RX_QC] = {
+			.min = 0,	.max = 63,
+		},
+	},
+	[IPA_RESOURCE_TYPE_SRC_ACK_ENTRIES] = {
+		.limits[IPA_RSRC_GROUP_SRC_UL] = {
+			.min = 22,	.max = 22,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_DL] = {
+			.min = 16,	.max = 16,
+		},
+		.limits[IPA_RSRC_GROUP_SRC_URLLC] = {
+			.min = 16,	.max = 16,
+		},
+	},
+};
+
+/* Destination resource configuration data for an SoC having IPA v5.0 */
+static const struct ipa_resource ipa_resource_dst[] = {
+	[IPA_RESOURCE_TYPE_DST_DATA_SECTORS] = {
+		.limits[IPA_RSRC_GROUP_DST_UL] = {
+			.min = 6,	.max = 6,
+		},
+		.limits[IPA_RSRC_GROUP_DST_DL] = {
+			.min = 5,	.max = 5,
+		},
+		.limits[IPA_RSRC_GROUP_DST_DRB_IP] = {
+			.min = 39,	.max = 39,
+		},
+	},
+	[IPA_RESOURCE_TYPE_DST_DPS_DMARS] = {
+		.limits[IPA_RSRC_GROUP_DST_UL] = {
+			.min = 0,	.max = 3,
+		},
+		.limits[IPA_RSRC_GROUP_DST_DL] = {
+			.min = 0,	.max = 3,
+		},
+	},
+	[IPA_RESOURCE_TYPE_DST_ULSO_SEGMENTS] = {
+		.limits[IPA_RSRC_GROUP_DST_UL] = {
+			.min = 0,	.max = 63,
+		},
+		.limits[IPA_RSRC_GROUP_DST_DL] = {
+			.min = 0,	.max = 63,
+		},
+	},
+};
+
+/* Resource configuration data for an SoC having IPA v5.0 */
+static const struct ipa_resource_data ipa_resource_data = {
+	.rsrc_group_dst_count	= IPA_RSRC_GROUP_DST_COUNT,
+	.rsrc_group_src_count	= IPA_RSRC_GROUP_SRC_COUNT,
+	.resource_src_count	= ARRAY_SIZE(ipa_resource_src),
+	.resource_src		= ipa_resource_src,
+	.resource_dst_count	= ARRAY_SIZE(ipa_resource_dst),
+	.resource_dst		= ipa_resource_dst,
+};
+
+/* IPA-resident memory region data for an SoC having IPA v5.0 */
+static const struct ipa_mem ipa_mem_local_data[] = {
+	{
+		.id		= IPA_MEM_UC_EVENT_RING,
+		.offset		= 0x0000,
+		.size		= 0x1000,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_UC_SHARED,
+		.offset		= 0x1000,
+		.size		= 0x0080,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_UC_INFO,
+		.offset		= 0x1080,
+		.size		= 0x0200,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_V4_FILTER_HASHED,
+		.offset		= 0x1288,
+		.size		= 0x0078,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V4_FILTER,
+		.offset		= 0x1308,
+		.size		= 0x0078,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V6_FILTER_HASHED,
+		.offset		= 0x1388,
+		.size		= 0x0078,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V6_FILTER,
+		.offset		= 0x1408,
+		.size		= 0x0078,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V4_ROUTE_HASHED,
+		.offset		= 0x1488,
+		.size		= 0x0098,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V4_ROUTE,
+		.offset		= 0x1528,
+		.size		= 0x0098,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V6_ROUTE_HASHED,
+		.offset		= 0x15c8,
+		.size		= 0x0098,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_V6_ROUTE,
+		.offset		= 0x1668,
+		.size		= 0x0098,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_MODEM_HEADER,
+		.offset		= 0x1708,
+		.size		= 0x0240,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_AP_HEADER,
+		.offset		= 0x1948,
+		.size		= 0x01e0,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_MODEM_PROC_CTX,
+		.offset		= 0x1b40,
+		.size		= 0x0b20,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_AP_PROC_CTX,
+		.offset		= 0x2660,
+		.size		= 0x0200,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_STATS_QUOTA_MODEM,
+		.offset		= 0x2868,
+		.size		= 0x0060,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_STATS_QUOTA_AP,
+		.offset		= 0x28c8,
+		.size		= 0x0048,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_AP_V4_FILTER,
+		.offset		= 0x2918,
+		.size		= 0x0118,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_AP_V6_FILTER,
+		.offset		= 0x2aa0,
+		.size		= 0x0228,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_STATS_FILTER_ROUTE,
+		.offset		= 0x2cd0,
+		.size		= 0x0ba0,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_STATS_DROP,
+		.offset		= 0x3870,
+		.size		= 0x0020,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_MODEM,
+		.offset		= 0x3898,
+		.size		= 0x0d48,
+		.canary_count	= 2,
+	},
+	{
+		.id		= IPA_MEM_NAT_TABLE,
+		.offset		= 0x45e0,
+		.size		= 0x0900,
+		.canary_count	= 0,
+	},
+	{
+		.id		= IPA_MEM_PDN_CONFIG,
+		.offset		= 0x4ee8,
+		.size		= 0x0100,
+		.canary_count	= 2,
+	},
+};
+
+/* Memory configuration data for an SoC having IPA v5.0 */
+static const struct ipa_mem_data ipa_mem_data = {
+	.local_count	= ARRAY_SIZE(ipa_mem_local_data),
+	.local		= ipa_mem_local_data,
+	.imem_addr	= 0x14688000,
+	.imem_size	= 0x00003000,
+	.smem_id	= 497,
+	.smem_size	= 0x00009000,
+};
+
+/* Interconnect rates are in 1000 byte/second units */
+static const struct ipa_interconnect_data ipa_interconnect_data[] = {
+	{
+		.name			= "memory",
+		.peak_bandwidth		= 1900000,	/* 1.9 GBps */
+		.average_bandwidth	= 600000,	/* 600 MBps */
+	},
+	/* Average rate is unused for the next interconnect */
+	{
+		.name			= "config",
+		.peak_bandwidth		= 76800,	/* 76.8 MBps */
+		.average_bandwidth	= 0,		/* unused */
+	},
+};
+
+/* Clock and interconnect configuration data for an SoC having IPA v5.0 */
+static const struct ipa_power_data ipa_power_data = {
+	.core_clock_rate	= 120 * 1000 * 1000,	/* Hz */
+	.interconnect_count	= ARRAY_SIZE(ipa_interconnect_data),
+	.interconnect_data	= ipa_interconnect_data,
+};
+
+/* Configuration data for an SoC having IPA v5.0. */
+const struct ipa_data ipa_data_v5_0 = {
+	.version		= IPA_VERSION_5_0,
+	.qsb_count		= ARRAY_SIZE(ipa_qsb_data),
+	.qsb_data		= ipa_qsb_data,
+	.modem_route_count	= 11,
+	.endpoint_count		= ARRAY_SIZE(ipa_gsi_endpoint_data),
+	.endpoint_data		= ipa_gsi_endpoint_data,
+	.resource_data		= &ipa_resource_data,
+	.mem_data		= &ipa_mem_data,
+	.power_data		= &ipa_power_data,
+};
diff --git a/drivers/net/ipa/ipa_data.h b/drivers/net/ipa/ipa_data.h
index 818e64114ed5..ce82b00fdc49 100644
--- a/drivers/net/ipa/ipa_data.h
+++ b/drivers/net/ipa/ipa_data.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 
 /* Copyright (c) 2012-2018, The Linux Foundation. All rights reserved.
- * Copyright (C) 2019-2022 Linaro Ltd.
+ * Copyright (C) 2019-2023 Linaro Ltd.
  */
 #ifndef _IPA_DATA_H_
 #define _IPA_DATA_H_
@@ -249,5 +249,6 @@ extern const struct ipa_data ipa_data_v4_5;
 extern const struct ipa_data ipa_data_v4_7;
 extern const struct ipa_data ipa_data_v4_9;
 extern const struct ipa_data ipa_data_v4_11;
+extern const struct ipa_data ipa_data_v5_0;
 
 #endif /* _IPA_DATA_H_ */
diff --git a/drivers/net/ipa/ipa_main.c b/drivers/net/ipa/ipa_main.c
index 4cc8d8d6bc9b..6a2f2fc2f501 100644
--- a/drivers/net/ipa/ipa_main.c
+++ b/drivers/net/ipa/ipa_main.c
@@ -285,7 +285,7 @@ static void ipa_hardware_config_comp(struct ipa *ipa)
 	} else if (ipa->version < IPA_VERSION_4_5) {
 		val |= reg_bit(reg, GSI_MULTI_AXI_MASTERS_DIS);
 	} else {
-		/* For IPA v4.5 FULL_FLUSH_WAIT_RS_CLOSURE_EN is 0 */
+		/* For IPA v4.5+ FULL_FLUSH_WAIT_RS_CLOSURE_EN is 0 */
 	}
 
 	val |= reg_bit(reg, GSI_MULTI_INORDER_RD_DIS);
@@ -684,6 +684,10 @@ static const struct of_device_id ipa_match[] = {
 		.compatible	= "qcom,sc7280-ipa",
 		.data		= &ipa_data_v4_11,
 	},
+	{
+		.compatible	= "qcom,sdx65-ipa",
+		.data		= &ipa_data_v5_0,
+	},
 	{ },
 };
 MODULE_DEVICE_TABLE(of, ipa_match);
-- 
cgit v1.2.3-70-g09d2


From b671c2067a04c0668df174ff5dfdb573d1f9b074 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:23:58 -0700
Subject: bpf: Retire the struct_ops map kvalue->refcnt.

We have replaced kvalue-refcnt with synchronize_rcu() to wait for an
RCU grace period.

Maintenance of kvalue->refcnt was a complicated task, as we had to
simultaneously keep track of two reference counts: one for the
reference count of bpf_map. When the kvalue->refcnt reaches zero, we
also have to reduce the reference count on bpf_map - yet these steps
are not performed in an atomic manner and require us to be vigilant
when managing them. By eliminating kvalue->refcnt, we can make our
maintenance more straightforward as the refcount of bpf_map is now
solely managed!

To prevent the trampoline image of a struct_ops from being released
while it is still in use, we wait for an RCU grace period. The
setsockopt(TCP_CONGESTION, "...") command allows you to change your
socket's congestion control algorithm and can result in releasing the
old struct_ops implementation. It is fine. However, this function is
exposed through bpf_setsockopt(), it may be accessed by BPF programs
as well. To ensure that the trampoline image belonging to struct_op
can be safely called while its method is in use, the trampoline
safeguarde the BPF program with rcu_read_lock(). Doing so prevents any
destruction of the associated images before returning from a
trampoline and requires us to wait for an RCU grace period.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Link: https://lore.kernel.org/r/20230323032405.3735486-2-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf.h         |  1 +
 kernel/bpf/bpf_struct_ops.c | 77 ++++++++++++++++++++++++++-------------------
 kernel/bpf/syscall.c        |  6 ++--
 3 files changed, 49 insertions(+), 35 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index ec0df059f562..f04098468d7a 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1945,6 +1945,7 @@ struct bpf_map *bpf_map_get_with_uref(u32 ufd);
 struct bpf_map *__bpf_map_get(struct fd f);
 void bpf_map_inc(struct bpf_map *map);
 void bpf_map_inc_with_uref(struct bpf_map *map);
+struct bpf_map *__bpf_map_inc_not_zero(struct bpf_map *map, bool uref);
 struct bpf_map * __must_check bpf_map_inc_not_zero(struct bpf_map *map);
 void bpf_map_put_with_uref(struct bpf_map *map);
 void bpf_map_put(struct bpf_map *map);
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index ba7a94276e3b..2f3c4a0e03ee 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -11,6 +11,7 @@
 #include <linux/refcount.h>
 #include <linux/mutex.h>
 #include <linux/btf_ids.h>
+#include <linux/rcupdate_wait.h>
 
 enum bpf_struct_ops_state {
 	BPF_STRUCT_OPS_STATE_INIT,
@@ -249,6 +250,7 @@ int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map, void *key,
 	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
 	struct bpf_struct_ops_value *uvalue, *kvalue;
 	enum bpf_struct_ops_state state;
+	s64 refcnt;
 
 	if (unlikely(*(u32 *)key != 0))
 		return -ENOENT;
@@ -267,7 +269,14 @@ int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map, void *key,
 	uvalue = value;
 	memcpy(uvalue, st_map->uvalue, map->value_size);
 	uvalue->state = state;
-	refcount_set(&uvalue->refcnt, refcount_read(&kvalue->refcnt));
+
+	/* This value offers the user space a general estimate of how
+	 * many sockets are still utilizing this struct_ops for TCP
+	 * congestion control. The number might not be exact, but it
+	 * should sufficiently meet our present goals.
+	 */
+	refcnt = atomic64_read(&map->refcnt) - atomic64_read(&map->usercnt);
+	refcount_set(&uvalue->refcnt, max_t(s64, refcnt, 0));
 
 	return 0;
 }
@@ -491,7 +500,6 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 		*(unsigned long *)(udata + moff) = prog->aux->id;
 	}
 
-	refcount_set(&kvalue->refcnt, 1);
 	bpf_map_inc(map);
 
 	set_memory_rox((long)st_map->image, 1);
@@ -536,8 +544,7 @@ static long bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key)
 	switch (prev_state) {
 	case BPF_STRUCT_OPS_STATE_INUSE:
 		st_map->st_ops->unreg(&st_map->kvalue.data);
-		if (refcount_dec_and_test(&st_map->kvalue.refcnt))
-			bpf_map_put(map);
+		bpf_map_put(map);
 		return 0;
 	case BPF_STRUCT_OPS_STATE_TOBEFREE:
 		return -EINPROGRESS;
@@ -570,7 +577,7 @@ static void bpf_struct_ops_map_seq_show_elem(struct bpf_map *map, void *key,
 	kfree(value);
 }
 
-static void bpf_struct_ops_map_free(struct bpf_map *map)
+static void __bpf_struct_ops_map_free(struct bpf_map *map)
 {
 	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
 
@@ -582,6 +589,28 @@ static void bpf_struct_ops_map_free(struct bpf_map *map)
 	bpf_map_area_free(st_map);
 }
 
+static void bpf_struct_ops_map_free(struct bpf_map *map)
+{
+	/* The struct_ops's function may switch to another struct_ops.
+	 *
+	 * For example, bpf_tcp_cc_x->init() may switch to
+	 * another tcp_cc_y by calling
+	 * setsockopt(TCP_CONGESTION, "tcp_cc_y").
+	 * During the switch,  bpf_struct_ops_put(tcp_cc_x) is called
+	 * and its refcount may reach 0 which then free its
+	 * trampoline image while tcp_cc_x is still running.
+	 *
+	 * A vanilla rcu gp is to wait for all bpf-tcp-cc prog
+	 * to finish. bpf-tcp-cc prog is non sleepable.
+	 * A rcu_tasks gp is to wait for the last few insn
+	 * in the tramopline image to finish before releasing
+	 * the trampoline image.
+	 */
+	synchronize_rcu_mult(call_rcu, call_rcu_tasks);
+
+	__bpf_struct_ops_map_free(map);
+}
+
 static int bpf_struct_ops_map_alloc_check(union bpf_attr *attr)
 {
 	if (attr->key_size != sizeof(unsigned int) || attr->max_entries != 1 ||
@@ -630,7 +659,7 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 				   NUMA_NO_NODE);
 	st_map->image = bpf_jit_alloc_exec(PAGE_SIZE);
 	if (!st_map->uvalue || !st_map->links || !st_map->image) {
-		bpf_struct_ops_map_free(map);
+		__bpf_struct_ops_map_free(map);
 		return ERR_PTR(-ENOMEM);
 	}
 
@@ -676,41 +705,23 @@ const struct bpf_map_ops bpf_struct_ops_map_ops = {
 bool bpf_struct_ops_get(const void *kdata)
 {
 	struct bpf_struct_ops_value *kvalue;
+	struct bpf_struct_ops_map *st_map;
+	struct bpf_map *map;
 
 	kvalue = container_of(kdata, struct bpf_struct_ops_value, data);
+	st_map = container_of(kvalue, struct bpf_struct_ops_map, kvalue);
 
-	return refcount_inc_not_zero(&kvalue->refcnt);
-}
-
-static void bpf_struct_ops_put_rcu(struct rcu_head *head)
-{
-	struct bpf_struct_ops_map *st_map;
-
-	st_map = container_of(head, struct bpf_struct_ops_map, rcu);
-	bpf_map_put(&st_map->map);
+	map = __bpf_map_inc_not_zero(&st_map->map, false);
+	return !IS_ERR(map);
 }
 
 void bpf_struct_ops_put(const void *kdata)
 {
 	struct bpf_struct_ops_value *kvalue;
+	struct bpf_struct_ops_map *st_map;
 
 	kvalue = container_of(kdata, struct bpf_struct_ops_value, data);
-	if (refcount_dec_and_test(&kvalue->refcnt)) {
-		struct bpf_struct_ops_map *st_map;
-
-		st_map = container_of(kvalue, struct bpf_struct_ops_map,
-				      kvalue);
-		/* The struct_ops's function may switch to another struct_ops.
-		 *
-		 * For example, bpf_tcp_cc_x->init() may switch to
-		 * another tcp_cc_y by calling
-		 * setsockopt(TCP_CONGESTION, "tcp_cc_y").
-		 * During the switch,  bpf_struct_ops_put(tcp_cc_x) is called
-		 * and its map->refcnt may reach 0 which then free its
-		 * trampoline image while tcp_cc_x is still running.
-		 *
-		 * Thus, a rcu grace period is needed here.
-		 */
-		call_rcu(&st_map->rcu, bpf_struct_ops_put_rcu);
-	}
+	st_map = container_of(kvalue, struct bpf_struct_ops_map, kvalue);
+
+	bpf_map_put(&st_map->map);
 }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 099e9068bcdd..cff0348a2871 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1303,8 +1303,10 @@ struct bpf_map *bpf_map_get_with_uref(u32 ufd)
 	return map;
 }
 
-/* map_idr_lock should have been held */
-static struct bpf_map *__bpf_map_inc_not_zero(struct bpf_map *map, bool uref)
+/* map_idr_lock should have been held or the map should have been
+ * protected by rcu read lock.
+ */
+struct bpf_map *__bpf_map_inc_not_zero(struct bpf_map *map, bool uref)
 {
 	int refold;
 
-- 
cgit v1.2.3-70-g09d2


From 8fb1a76a0f35c45a424c9eb84b0f97ffd51e6052 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:23:59 -0700
Subject: net: Update an existing TCP congestion control algorithm.

This feature lets you immediately transition to another congestion
control algorithm or implementation with the same name.  Once a name
is updated, new connections will apply this new algorithm.

The purpose is to update a customized algorithm implemented in BPF
struct_ops with a new version on the flight.  The following is an
example of using the userspace API implemented in later BPF patches.

   link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
   .......
   err = bpf_link__update_map(link, skel->maps.ca_update_2);

We first load and register an algorithm implemented in BPF struct_ops,
then swap it out with a new one using the same name. After that, newly
created connections will apply the updated algorithm, while older ones
retain the previous version already applied.

This patch also takes this chance to refactor the ca validation into
the new tcp_validate_congestion_control() function.

Cc: netdev@vger.kernel.org, Eric Dumazet <edumazet@google.com>
Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Link: https://lore.kernel.org/r/20230323032405.3735486-3-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/net/tcp.h   |  3 +++
 net/ipv4/tcp_cong.c | 66 +++++++++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 62 insertions(+), 7 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index db9f828e9d1e..2abb755e6a3a 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1117,6 +1117,9 @@ struct tcp_congestion_ops {
 
 int tcp_register_congestion_control(struct tcp_congestion_ops *type);
 void tcp_unregister_congestion_control(struct tcp_congestion_ops *type);
+int tcp_update_congestion_control(struct tcp_congestion_ops *type,
+				  struct tcp_congestion_ops *old_type);
+int tcp_validate_congestion_control(struct tcp_congestion_ops *ca);
 
 void tcp_assign_congestion_control(struct sock *sk);
 void tcp_init_congestion_control(struct sock *sk);
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index db8b4b488c31..1b34050a7538 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -75,14 +75,8 @@ struct tcp_congestion_ops *tcp_ca_find_key(u32 key)
 	return NULL;
 }
 
-/*
- * Attach new congestion control algorithm to the list
- * of available options.
- */
-int tcp_register_congestion_control(struct tcp_congestion_ops *ca)
+int tcp_validate_congestion_control(struct tcp_congestion_ops *ca)
 {
-	int ret = 0;
-
 	/* all algorithms must implement these */
 	if (!ca->ssthresh || !ca->undo_cwnd ||
 	    !(ca->cong_avoid || ca->cong_control)) {
@@ -90,6 +84,20 @@ int tcp_register_congestion_control(struct tcp_congestion_ops *ca)
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+/* Attach new congestion control algorithm to the list
+ * of available options.
+ */
+int tcp_register_congestion_control(struct tcp_congestion_ops *ca)
+{
+	int ret;
+
+	ret = tcp_validate_congestion_control(ca);
+	if (ret)
+		return ret;
+
 	ca->key = jhash(ca->name, sizeof(ca->name), strlen(ca->name));
 
 	spin_lock(&tcp_cong_list_lock);
@@ -130,6 +138,50 @@ void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca)
 }
 EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control);
 
+/* Replace a registered old ca with a new one.
+ *
+ * The new ca must have the same name as the old one, that has been
+ * registered.
+ */
+int tcp_update_congestion_control(struct tcp_congestion_ops *ca, struct tcp_congestion_ops *old_ca)
+{
+	struct tcp_congestion_ops *existing;
+	int ret;
+
+	ret = tcp_validate_congestion_control(ca);
+	if (ret)
+		return ret;
+
+	ca->key = jhash(ca->name, sizeof(ca->name), strlen(ca->name));
+
+	spin_lock(&tcp_cong_list_lock);
+	existing = tcp_ca_find_key(old_ca->key);
+	if (ca->key == TCP_CA_UNSPEC || !existing || strcmp(existing->name, ca->name)) {
+		pr_notice("%s not registered or non-unique key\n",
+			  ca->name);
+		ret = -EINVAL;
+	} else if (existing != old_ca) {
+		pr_notice("invalid old congestion control algorithm to replace\n");
+		ret = -EINVAL;
+	} else {
+		/* Add the new one before removing the old one to keep
+		 * one implementation available all the time.
+		 */
+		list_add_tail_rcu(&ca->list, &tcp_cong_list);
+		list_del_rcu(&existing->list);
+		pr_debug("%s updated\n", ca->name);
+	}
+	spin_unlock(&tcp_cong_list_lock);
+
+	/* Wait for outstanding readers to complete before the
+	 * module or struct_ops gets removed entirely.
+	 */
+	if (!ret)
+		synchronize_rcu();
+
+	return ret;
+}
+
 u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca)
 {
 	const struct tcp_congestion_ops *ca;
-- 
cgit v1.2.3-70-g09d2


From 68b04864ca425d1894c96b8141d4fba1181f11cb Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:24:00 -0700
Subject: bpf: Create links for BPF struct_ops maps.

Make bpf_link support struct_ops.  Previously, struct_ops were always
used alone without any associated links. Upon updating its value, a
struct_ops would be activated automatically. Yet other BPF program
types required to make a bpf_link with their instances before they
could become active. Now, however, you can create an inactive
struct_ops, and create a link to activate it later.

With bpf_links, struct_ops has a behavior similar to other BPF program
types. You can pin/unpin them from their links and the struct_ops will
be deactivated when its link is removed while previously need someone
to delete the value for it to be deactivated.

bpf_links are responsible for registering their associated
struct_ops. You can only use a struct_ops that has the BPF_F_LINK flag
set to create a bpf_link, while a structs without this flag behaves in
the same manner as before and is registered upon updating its value.

The BPF_LINK_TYPE_STRUCT_OPS serves a dual purpose. Not only is it
used to craft the links for BPF struct_ops programs, but also to
create links for BPF struct_ops them-self.  Since the links of BPF
struct_ops programs are only used to create trampolines internally,
they are never seen in other contexts. Thus, they can be reused for
struct_ops themself.

To maintain a reference to the map supporting this link, we add
bpf_struct_ops_link as an additional type. The pointer of the map is
RCU and won't be necessary until later in the patchset.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Link: https://lore.kernel.org/r/20230323032405.3735486-4-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf.h            |   7 ++
 include/uapi/linux/bpf.h       |  12 +++-
 kernel/bpf/bpf_struct_ops.c    | 143 ++++++++++++++++++++++++++++++++++++++++-
 kernel/bpf/syscall.c           |  23 ++++---
 net/ipv4/bpf_tcp_ca.c          |   8 ++-
 tools/include/uapi/linux/bpf.h |  12 +++-
 6 files changed, 190 insertions(+), 15 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index f04098468d7a..8552279efe46 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1518,6 +1518,7 @@ struct bpf_struct_ops {
 			   void *kdata, const void *udata);
 	int (*reg)(void *kdata);
 	void (*unreg)(void *kdata);
+	int (*validate)(void *kdata);
 	const struct btf_type *type;
 	const struct btf_type *value_type;
 	const char *name;
@@ -1552,6 +1553,7 @@ static inline void bpf_module_put(const void *data, struct module *owner)
 	else
 		module_put(owner);
 }
+int bpf_struct_ops_link_create(union bpf_attr *attr);
 
 #ifdef CONFIG_NET
 /* Define it here to avoid the use of forward declaration */
@@ -1592,6 +1594,11 @@ static inline int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map,
 {
 	return -EINVAL;
 }
+static inline int bpf_struct_ops_link_create(union bpf_attr *attr)
+{
+	return -EOPNOTSUPP;
+}
+
 #endif
 
 #if defined(CONFIG_CGROUP_BPF) && defined(CONFIG_BPF_LSM)
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 13129df937cd..42f40ee083bf 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1033,6 +1033,7 @@ enum bpf_attach_type {
 	BPF_PERF_EVENT,
 	BPF_TRACE_KPROBE_MULTI,
 	BPF_LSM_CGROUP,
+	BPF_STRUCT_OPS,
 	__MAX_BPF_ATTACH_TYPE
 };
 
@@ -1266,6 +1267,9 @@ enum {
 
 /* Create a map that is suitable to be an inner map with dynamic max entries */
 	BPF_F_INNER_MAP		= (1U << 12),
+
+/* Create a map that will be registered/unregesitered by the backed bpf_link */
+	BPF_F_LINK		= (1U << 13),
 };
 
 /* Flags for BPF_PROG_QUERY. */
@@ -1507,7 +1511,10 @@ union bpf_attr {
 	} task_fd_query;
 
 	struct { /* struct used by BPF_LINK_CREATE command */
-		__u32		prog_fd;	/* eBPF program to attach */
+		union {
+			__u32		prog_fd;	/* eBPF program to attach */
+			__u32		map_fd;		/* struct_ops to attach */
+		};
 		union {
 			__u32		target_fd;	/* object to attach to */
 			__u32		target_ifindex; /* target ifindex */
@@ -6379,6 +6386,9 @@ struct bpf_link_info {
 		struct {
 			__u32 ifindex;
 		} xdp;
+		struct {
+			__u32 map_id;
+		} struct_ops;
 	};
 } __attribute__((aligned(8)));
 
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 2f3c4a0e03ee..3d6b5240c25a 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -17,6 +17,7 @@ enum bpf_struct_ops_state {
 	BPF_STRUCT_OPS_STATE_INIT,
 	BPF_STRUCT_OPS_STATE_INUSE,
 	BPF_STRUCT_OPS_STATE_TOBEFREE,
+	BPF_STRUCT_OPS_STATE_READY,
 };
 
 #define BPF_STRUCT_OPS_COMMON_VALUE			\
@@ -59,6 +60,11 @@ struct bpf_struct_ops_map {
 	struct bpf_struct_ops_value kvalue;
 };
 
+struct bpf_struct_ops_link {
+	struct bpf_link link;
+	struct bpf_map __rcu *map;
+};
+
 #define VALUE_PREFIX "bpf_struct_ops_"
 #define VALUE_PREFIX_LEN (sizeof(VALUE_PREFIX) - 1)
 
@@ -500,11 +506,29 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 		*(unsigned long *)(udata + moff) = prog->aux->id;
 	}
 
-	bpf_map_inc(map);
+	if (st_map->map.map_flags & BPF_F_LINK) {
+		err = st_ops->validate(kdata);
+		if (err)
+			goto reset_unlock;
+		set_memory_rox((long)st_map->image, 1);
+		/* Let bpf_link handle registration & unregistration.
+		 *
+		 * Pair with smp_load_acquire() during lookup_elem().
+		 */
+		smp_store_release(&kvalue->state, BPF_STRUCT_OPS_STATE_READY);
+		goto unlock;
+	}
 
 	set_memory_rox((long)st_map->image, 1);
 	err = st_ops->reg(kdata);
 	if (likely(!err)) {
+		/* This refcnt increment on the map here after
+		 * 'st_ops->reg()' is secure since the state of the
+		 * map must be set to INIT at this moment, and thus
+		 * bpf_struct_ops_map_delete_elem() can't unregister
+		 * or transition it to TOBEFREE concurrently.
+		 */
+		bpf_map_inc(map);
 		/* Pair with smp_load_acquire() during lookup_elem().
 		 * It ensures the above udata updates (e.g. prog->aux->id)
 		 * can be seen once BPF_STRUCT_OPS_STATE_INUSE is set.
@@ -520,7 +544,6 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 	 */
 	set_memory_nx((long)st_map->image, 1);
 	set_memory_rw((long)st_map->image, 1);
-	bpf_map_put(map);
 
 reset_unlock:
 	bpf_struct_ops_map_put_progs(st_map);
@@ -538,6 +561,9 @@ static long bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key)
 	struct bpf_struct_ops_map *st_map;
 
 	st_map = (struct bpf_struct_ops_map *)map;
+	if (st_map->map.map_flags & BPF_F_LINK)
+		return -EOPNOTSUPP;
+
 	prev_state = cmpxchg(&st_map->kvalue.state,
 			     BPF_STRUCT_OPS_STATE_INUSE,
 			     BPF_STRUCT_OPS_STATE_TOBEFREE);
@@ -614,7 +640,7 @@ static void bpf_struct_ops_map_free(struct bpf_map *map)
 static int bpf_struct_ops_map_alloc_check(union bpf_attr *attr)
 {
 	if (attr->key_size != sizeof(unsigned int) || attr->max_entries != 1 ||
-	    attr->map_flags || !attr->btf_vmlinux_value_type_id)
+	    (attr->map_flags & ~BPF_F_LINK) || !attr->btf_vmlinux_value_type_id)
 		return -EINVAL;
 	return 0;
 }
@@ -638,6 +664,9 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 	if (attr->value_size != vt->size)
 		return ERR_PTR(-EINVAL);
 
+	if (attr->map_flags & BPF_F_LINK && !st_ops->validate)
+		return ERR_PTR(-EOPNOTSUPP);
+
 	t = st_ops->type;
 
 	st_map_size = sizeof(*st_map) +
@@ -725,3 +754,111 @@ void bpf_struct_ops_put(const void *kdata)
 
 	bpf_map_put(&st_map->map);
 }
+
+static bool bpf_struct_ops_valid_to_reg(struct bpf_map *map)
+{
+	struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map;
+
+	return map->map_type == BPF_MAP_TYPE_STRUCT_OPS &&
+		map->map_flags & BPF_F_LINK &&
+		/* Pair with smp_store_release() during map_update */
+		smp_load_acquire(&st_map->kvalue.state) == BPF_STRUCT_OPS_STATE_READY;
+}
+
+static void bpf_struct_ops_map_link_dealloc(struct bpf_link *link)
+{
+	struct bpf_struct_ops_link *st_link;
+	struct bpf_struct_ops_map *st_map;
+
+	st_link = container_of(link, struct bpf_struct_ops_link, link);
+	st_map = (struct bpf_struct_ops_map *)
+		rcu_dereference_protected(st_link->map, true);
+	if (st_map) {
+		/* st_link->map can be NULL if
+		 * bpf_struct_ops_link_create() fails to register.
+		 */
+		st_map->st_ops->unreg(&st_map->kvalue.data);
+		bpf_map_put(&st_map->map);
+	}
+	kfree(st_link);
+}
+
+static void bpf_struct_ops_map_link_show_fdinfo(const struct bpf_link *link,
+					    struct seq_file *seq)
+{
+	struct bpf_struct_ops_link *st_link;
+	struct bpf_map *map;
+
+	st_link = container_of(link, struct bpf_struct_ops_link, link);
+	rcu_read_lock();
+	map = rcu_dereference(st_link->map);
+	seq_printf(seq, "map_id:\t%d\n", map->id);
+	rcu_read_unlock();
+}
+
+static int bpf_struct_ops_map_link_fill_link_info(const struct bpf_link *link,
+					       struct bpf_link_info *info)
+{
+	struct bpf_struct_ops_link *st_link;
+	struct bpf_map *map;
+
+	st_link = container_of(link, struct bpf_struct_ops_link, link);
+	rcu_read_lock();
+	map = rcu_dereference(st_link->map);
+	info->struct_ops.map_id = map->id;
+	rcu_read_unlock();
+	return 0;
+}
+
+static const struct bpf_link_ops bpf_struct_ops_map_lops = {
+	.dealloc = bpf_struct_ops_map_link_dealloc,
+	.show_fdinfo = bpf_struct_ops_map_link_show_fdinfo,
+	.fill_link_info = bpf_struct_ops_map_link_fill_link_info,
+};
+
+int bpf_struct_ops_link_create(union bpf_attr *attr)
+{
+	struct bpf_struct_ops_link *link = NULL;
+	struct bpf_link_primer link_primer;
+	struct bpf_struct_ops_map *st_map;
+	struct bpf_map *map;
+	int err;
+
+	map = bpf_map_get(attr->link_create.map_fd);
+	if (!map)
+		return -EINVAL;
+
+	st_map = (struct bpf_struct_ops_map *)map;
+
+	if (!bpf_struct_ops_valid_to_reg(map)) {
+		err = -EINVAL;
+		goto err_out;
+	}
+
+	link = kzalloc(sizeof(*link), GFP_USER);
+	if (!link) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+	bpf_link_init(&link->link, BPF_LINK_TYPE_STRUCT_OPS, &bpf_struct_ops_map_lops, NULL);
+
+	err = bpf_link_prime(&link->link, &link_primer);
+	if (err)
+		goto err_out;
+
+	err = st_map->st_ops->reg(st_map->kvalue.data);
+	if (err) {
+		bpf_link_cleanup(&link_primer);
+		link = NULL;
+		goto err_out;
+	}
+	RCU_INIT_POINTER(link->map, map);
+
+	return bpf_link_settle(&link_primer);
+
+err_out:
+	bpf_map_put(map);
+	kfree(link);
+	return err;
+}
+
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index cff0348a2871..21f76698875c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2825,16 +2825,19 @@ static void bpf_link_show_fdinfo(struct seq_file *m, struct file *filp)
 	const struct bpf_prog *prog = link->prog;
 	char prog_tag[sizeof(prog->tag) * 2 + 1] = { };
 
-	bin2hex(prog_tag, prog->tag, sizeof(prog->tag));
 	seq_printf(m,
 		   "link_type:\t%s\n"
-		   "link_id:\t%u\n"
-		   "prog_tag:\t%s\n"
-		   "prog_id:\t%u\n",
+		   "link_id:\t%u\n",
 		   bpf_link_type_strs[link->type],
-		   link->id,
-		   prog_tag,
-		   prog->aux->id);
+		   link->id);
+	if (prog) {
+		bin2hex(prog_tag, prog->tag, sizeof(prog->tag));
+		seq_printf(m,
+			   "prog_tag:\t%s\n"
+			   "prog_id:\t%u\n",
+			   prog_tag,
+			   prog->aux->id);
+	}
 	if (link->ops->show_fdinfo)
 		link->ops->show_fdinfo(link, m);
 }
@@ -4314,7 +4317,8 @@ static int bpf_link_get_info_by_fd(struct file *file,
 
 	info.type = link->type;
 	info.id = link->id;
-	info.prog_id = link->prog->aux->id;
+	if (link->prog)
+		info.prog_id = link->prog->aux->id;
 
 	if (link->ops->fill_link_info) {
 		err = link->ops->fill_link_info(link, &info);
@@ -4577,6 +4581,9 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr)
 	if (CHECK_ATTR(BPF_LINK_CREATE))
 		return -EINVAL;
 
+	if (attr->link_create.attach_type == BPF_STRUCT_OPS)
+		return bpf_struct_ops_link_create(attr);
+
 	prog = bpf_prog_get(attr->link_create.prog_fd);
 	if (IS_ERR(prog))
 		return PTR_ERR(prog);
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index 13fc0c185cd9..bbbd5eb94db2 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -239,8 +239,6 @@ static int bpf_tcp_ca_init_member(const struct btf_type *t,
 		if (bpf_obj_name_cpy(tcp_ca->name, utcp_ca->name,
 				     sizeof(tcp_ca->name)) <= 0)
 			return -EINVAL;
-		if (tcp_ca_find(utcp_ca->name))
-			return -EEXIST;
 		return 1;
 	}
 
@@ -266,6 +264,11 @@ static void bpf_tcp_ca_unreg(void *kdata)
 	tcp_unregister_congestion_control(kdata);
 }
 
+static int bpf_tcp_ca_validate(void *kdata)
+{
+	return tcp_validate_congestion_control(kdata);
+}
+
 struct bpf_struct_ops bpf_tcp_congestion_ops = {
 	.verifier_ops = &bpf_tcp_ca_verifier_ops,
 	.reg = bpf_tcp_ca_reg,
@@ -273,6 +276,7 @@ struct bpf_struct_ops bpf_tcp_congestion_ops = {
 	.check_member = bpf_tcp_ca_check_member,
 	.init_member = bpf_tcp_ca_init_member,
 	.init = bpf_tcp_ca_init,
+	.validate = bpf_tcp_ca_validate,
 	.name = "tcp_congestion_ops",
 };
 
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 13129df937cd..9cf1deaf21f2 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1033,6 +1033,7 @@ enum bpf_attach_type {
 	BPF_PERF_EVENT,
 	BPF_TRACE_KPROBE_MULTI,
 	BPF_LSM_CGROUP,
+	BPF_STRUCT_OPS,
 	__MAX_BPF_ATTACH_TYPE
 };
 
@@ -1266,6 +1267,9 @@ enum {
 
 /* Create a map that is suitable to be an inner map with dynamic max entries */
 	BPF_F_INNER_MAP		= (1U << 12),
+
+/* Create a map that will be registered/unregesitered by the backed bpf_link */
+	BPF_F_LINK		= (1U << 13),
 };
 
 /* Flags for BPF_PROG_QUERY. */
@@ -1507,7 +1511,10 @@ union bpf_attr {
 	} task_fd_query;
 
 	struct { /* struct used by BPF_LINK_CREATE command */
-		__u32		prog_fd;	/* eBPF program to attach */
+		union {
+			__u32		prog_fd;	/* eBPF program to attach */
+			__u32		map_fd;		/* eBPF struct_ops to attach */
+		};
 		union {
 			__u32		target_fd;	/* object to attach to */
 			__u32		target_ifindex; /* target ifindex */
@@ -6379,6 +6386,9 @@ struct bpf_link_info {
 		struct {
 			__u32 ifindex;
 		} xdp;
+		struct {
+			__u32 map_id;
+		} struct_ops;
 	};
 } __attribute__((aligned(8)));
 
-- 
cgit v1.2.3-70-g09d2


From 8d1608d70927747da9c1a8770edf7b6ee68f8ebc Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:24:01 -0700
Subject: libbpf: Create a bpf_link in bpf_map__attach_struct_ops().

bpf_map__attach_struct_ops() was creating a dummy bpf_link as a
placeholder, but now it is constructing an authentic one by calling
bpf_link_create() if the map has the BPF_F_LINK flag.

You can flag a struct_ops map with BPF_F_LINK by calling
bpf_map__set_map_flags().

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230323032405.3735486-5-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/lib/bpf/libbpf.c | 95 +++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 71 insertions(+), 24 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 5d32aa8ea38a..4ff82baba0ea 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -116,6 +116,7 @@ static const char * const attach_type_name[] = {
 	[BPF_SK_REUSEPORT_SELECT_OR_MIGRATE]	= "sk_reuseport_select_or_migrate",
 	[BPF_PERF_EVENT]		= "perf_event",
 	[BPF_TRACE_KPROBE_MULTI]	= "trace_kprobe_multi",
+	[BPF_STRUCT_OPS]		= "struct_ops",
 };
 
 static const char * const link_type_name[] = {
@@ -7686,6 +7687,37 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 	return 0;
 }
 
+static void bpf_map_prepare_vdata(const struct bpf_map *map)
+{
+	struct bpf_struct_ops *st_ops;
+	__u32 i;
+
+	st_ops = map->st_ops;
+	for (i = 0; i < btf_vlen(st_ops->type); i++) {
+		struct bpf_program *prog = st_ops->progs[i];
+		void *kern_data;
+		int prog_fd;
+
+		if (!prog)
+			continue;
+
+		prog_fd = bpf_program__fd(prog);
+		kern_data = st_ops->kern_vdata + st_ops->kern_func_off[i];
+		*(unsigned long *)kern_data = prog_fd;
+	}
+}
+
+static int bpf_object_prepare_struct_ops(struct bpf_object *obj)
+{
+	int i;
+
+	for (i = 0; i < obj->nr_maps; i++)
+		if (bpf_map__is_struct_ops(&obj->maps[i]))
+			bpf_map_prepare_vdata(&obj->maps[i]);
+
+	return 0;
+}
+
 static int bpf_object_load(struct bpf_object *obj, int extra_log_level, const char *target_btf_path)
 {
 	int err, i;
@@ -7711,6 +7743,7 @@ static int bpf_object_load(struct bpf_object *obj, int extra_log_level, const ch
 	err = err ? : bpf_object__relocate(obj, obj->btf_custom_path ? : target_btf_path);
 	err = err ? : bpf_object__load_progs(obj, extra_log_level);
 	err = err ? : bpf_object_init_prog_arrays(obj);
+	err = err ? : bpf_object_prepare_struct_ops(obj);
 
 	if (obj->gen_loader) {
 		/* reset FDs */
@@ -11579,22 +11612,30 @@ struct bpf_link *bpf_program__attach(const struct bpf_program *prog)
 	return link;
 }
 
+struct bpf_link_struct_ops {
+	struct bpf_link link;
+	int map_fd;
+};
+
 static int bpf_link__detach_struct_ops(struct bpf_link *link)
 {
+	struct bpf_link_struct_ops *st_link;
 	__u32 zero = 0;
 
-	if (bpf_map_delete_elem(link->fd, &zero))
-		return -errno;
+	st_link = container_of(link, struct bpf_link_struct_ops, link);
 
-	return 0;
+	if (st_link->map_fd < 0)
+		/* w/o a real link */
+		return bpf_map_delete_elem(link->fd, &zero);
+
+	return close(link->fd);
 }
 
 struct bpf_link *bpf_map__attach_struct_ops(const struct bpf_map *map)
 {
-	struct bpf_struct_ops *st_ops;
-	struct bpf_link *link;
-	__u32 i, zero = 0;
-	int err;
+	struct bpf_link_struct_ops *link;
+	__u32 zero = 0;
+	int err, fd;
 
 	if (!bpf_map__is_struct_ops(map) || map->fd == -1)
 		return libbpf_err_ptr(-EINVAL);
@@ -11603,31 +11644,37 @@ struct bpf_link *bpf_map__attach_struct_ops(const struct bpf_map *map)
 	if (!link)
 		return libbpf_err_ptr(-EINVAL);
 
-	st_ops = map->st_ops;
-	for (i = 0; i < btf_vlen(st_ops->type); i++) {
-		struct bpf_program *prog = st_ops->progs[i];
-		void *kern_data;
-		int prog_fd;
+	/* kern_vdata should be prepared during the loading phase. */
+	err = bpf_map_update_elem(map->fd, &zero, map->st_ops->kern_vdata, 0);
+	/* It can be EBUSY if the map has been used to create or
+	 * update a link before.  We don't allow updating the value of
+	 * a struct_ops once it is set.  That ensures that the value
+	 * never changed.  So, it is safe to skip EBUSY.
+	 */
+	if (err && (!(map->def.map_flags & BPF_F_LINK) || err != -EBUSY)) {
+		free(link);
+		return libbpf_err_ptr(err);
+	}
 
-		if (!prog)
-			continue;
+	link->link.detach = bpf_link__detach_struct_ops;
 
-		prog_fd = bpf_program__fd(prog);
-		kern_data = st_ops->kern_vdata + st_ops->kern_func_off[i];
-		*(unsigned long *)kern_data = prog_fd;
+	if (!(map->def.map_flags & BPF_F_LINK)) {
+		/* w/o a real link */
+		link->link.fd = map->fd;
+		link->map_fd = -1;
+		return &link->link;
 	}
 
-	err = bpf_map_update_elem(map->fd, &zero, st_ops->kern_vdata, 0);
-	if (err) {
-		err = -errno;
+	fd = bpf_link_create(map->fd, 0, BPF_STRUCT_OPS, NULL);
+	if (fd < 0) {
 		free(link);
-		return libbpf_err_ptr(err);
+		return libbpf_err_ptr(fd);
 	}
 
-	link->detach = bpf_link__detach_struct_ops;
-	link->fd = map->fd;
+	link->link.fd = fd;
+	link->map_fd = map->fd;
 
-	return link;
+	return &link->link;
 }
 
 typedef enum bpf_perf_event_ret (*bpf_perf_event_print_t)(struct perf_event_header *hdr,
-- 
cgit v1.2.3-70-g09d2


From aef56f2e918bf8fc8de25f0b36e8c2aba44116ec Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:24:02 -0700
Subject: bpf: Update the struct_ops of a bpf_link.

By improving the BPF_LINK_UPDATE command of bpf(), it should allow you
to conveniently switch between different struct_ops on a single
bpf_link. This would enable smoother transitions from one struct_ops
to another.

The struct_ops maps passing along with BPF_LINK_UPDATE should have the
BPF_F_LINK flag.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230323032405.3735486-6-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf.h            |  3 +++
 include/uapi/linux/bpf.h       | 21 +++++++++++++-----
 kernel/bpf/bpf_struct_ops.c    | 48 +++++++++++++++++++++++++++++++++++++++++-
 kernel/bpf/syscall.c           | 34 ++++++++++++++++++++++++++++++
 net/ipv4/bpf_tcp_ca.c          |  6 ++++++
 tools/include/uapi/linux/bpf.h | 21 +++++++++++++-----
 6 files changed, 122 insertions(+), 11 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 8552279efe46..2d8f3f639e68 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1476,6 +1476,8 @@ struct bpf_link_ops {
 	void (*show_fdinfo)(const struct bpf_link *link, struct seq_file *seq);
 	int (*fill_link_info)(const struct bpf_link *link,
 			      struct bpf_link_info *info);
+	int (*update_map)(struct bpf_link *link, struct bpf_map *new_map,
+			  struct bpf_map *old_map);
 };
 
 struct bpf_tramp_link {
@@ -1518,6 +1520,7 @@ struct bpf_struct_ops {
 			   void *kdata, const void *udata);
 	int (*reg)(void *kdata);
 	void (*unreg)(void *kdata);
+	int (*update)(void *kdata, void *old_kdata);
 	int (*validate)(void *kdata);
 	const struct btf_type *type;
 	const struct btf_type *value_type;
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 42f40ee083bf..e3d3b5160d26 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1555,12 +1555,23 @@ union bpf_attr {
 
 	struct { /* struct used by BPF_LINK_UPDATE command */
 		__u32		link_fd;	/* link fd */
-		/* new program fd to update link with */
-		__u32		new_prog_fd;
+		union {
+			/* new program fd to update link with */
+			__u32		new_prog_fd;
+			/* new struct_ops map fd to update link with */
+			__u32           new_map_fd;
+		};
 		__u32		flags;		/* extra flags */
-		/* expected link's program fd; is specified only if
-		 * BPF_F_REPLACE flag is set in flags */
-		__u32		old_prog_fd;
+		union {
+			/* expected link's program fd; is specified only if
+			 * BPF_F_REPLACE flag is set in flags.
+			 */
+			__u32		old_prog_fd;
+			/* expected link's map fd; is specified only
+			 * if BPF_F_REPLACE flag is set.
+			 */
+			__u32           old_map_fd;
+		};
 	} link_update;
 
 	struct {
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 3d6b5240c25a..6401deca3b56 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -65,6 +65,8 @@ struct bpf_struct_ops_link {
 	struct bpf_map __rcu *map;
 };
 
+static DEFINE_MUTEX(update_mutex);
+
 #define VALUE_PREFIX "bpf_struct_ops_"
 #define VALUE_PREFIX_LEN (sizeof(VALUE_PREFIX) - 1)
 
@@ -664,7 +666,7 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 	if (attr->value_size != vt->size)
 		return ERR_PTR(-EINVAL);
 
-	if (attr->map_flags & BPF_F_LINK && !st_ops->validate)
+	if (attr->map_flags & BPF_F_LINK && (!st_ops->validate || !st_ops->update))
 		return ERR_PTR(-EOPNOTSUPP);
 
 	t = st_ops->type;
@@ -810,10 +812,54 @@ static int bpf_struct_ops_map_link_fill_link_info(const struct bpf_link *link,
 	return 0;
 }
 
+static int bpf_struct_ops_map_link_update(struct bpf_link *link, struct bpf_map *new_map,
+					  struct bpf_map *expected_old_map)
+{
+	struct bpf_struct_ops_map *st_map, *old_st_map;
+	struct bpf_map *old_map;
+	struct bpf_struct_ops_link *st_link;
+	int err = 0;
+
+	st_link = container_of(link, struct bpf_struct_ops_link, link);
+	st_map = container_of(new_map, struct bpf_struct_ops_map, map);
+
+	if (!bpf_struct_ops_valid_to_reg(new_map))
+		return -EINVAL;
+
+	mutex_lock(&update_mutex);
+
+	old_map = rcu_dereference_protected(st_link->map, lockdep_is_held(&update_mutex));
+	if (expected_old_map && old_map != expected_old_map) {
+		err = -EPERM;
+		goto err_out;
+	}
+
+	old_st_map = container_of(old_map, struct bpf_struct_ops_map, map);
+	/* The new and old struct_ops must be the same type. */
+	if (st_map->st_ops != old_st_map->st_ops) {
+		err = -EINVAL;
+		goto err_out;
+	}
+
+	err = st_map->st_ops->update(st_map->kvalue.data, old_st_map->kvalue.data);
+	if (err)
+		goto err_out;
+
+	bpf_map_inc(new_map);
+	rcu_assign_pointer(st_link->map, new_map);
+	bpf_map_put(old_map);
+
+err_out:
+	mutex_unlock(&update_mutex);
+
+	return err;
+}
+
 static const struct bpf_link_ops bpf_struct_ops_map_lops = {
 	.dealloc = bpf_struct_ops_map_link_dealloc,
 	.show_fdinfo = bpf_struct_ops_map_link_show_fdinfo,
 	.fill_link_info = bpf_struct_ops_map_link_fill_link_info,
+	.update_map = bpf_struct_ops_map_link_update,
 };
 
 int bpf_struct_ops_link_create(union bpf_attr *attr)
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 21f76698875c..b4d758fa5981 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -4682,6 +4682,35 @@ out:
 	return ret;
 }
 
+static int link_update_map(struct bpf_link *link, union bpf_attr *attr)
+{
+	struct bpf_map *new_map, *old_map = NULL;
+	int ret;
+
+	new_map = bpf_map_get(attr->link_update.new_map_fd);
+	if (IS_ERR(new_map))
+		return -EINVAL;
+
+	if (attr->link_update.flags & BPF_F_REPLACE) {
+		old_map = bpf_map_get(attr->link_update.old_map_fd);
+		if (IS_ERR(old_map)) {
+			ret = -EINVAL;
+			goto out_put;
+		}
+	} else if (attr->link_update.old_map_fd) {
+		ret = -EINVAL;
+		goto out_put;
+	}
+
+	ret = link->ops->update_map(link, new_map, old_map);
+
+	if (old_map)
+		bpf_map_put(old_map);
+out_put:
+	bpf_map_put(new_map);
+	return ret;
+}
+
 #define BPF_LINK_UPDATE_LAST_FIELD link_update.old_prog_fd
 
 static int link_update(union bpf_attr *attr)
@@ -4702,6 +4731,11 @@ static int link_update(union bpf_attr *attr)
 	if (IS_ERR(link))
 		return PTR_ERR(link);
 
+	if (link->ops->update_map) {
+		ret = link_update_map(link, attr);
+		goto out_put_link;
+	}
+
 	new_prog = bpf_prog_get(attr->link_update.new_prog_fd);
 	if (IS_ERR(new_prog)) {
 		ret = PTR_ERR(new_prog);
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index bbbd5eb94db2..e8b27826283e 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -264,6 +264,11 @@ static void bpf_tcp_ca_unreg(void *kdata)
 	tcp_unregister_congestion_control(kdata);
 }
 
+static int bpf_tcp_ca_update(void *kdata, void *old_kdata)
+{
+	return tcp_update_congestion_control(kdata, old_kdata);
+}
+
 static int bpf_tcp_ca_validate(void *kdata)
 {
 	return tcp_validate_congestion_control(kdata);
@@ -273,6 +278,7 @@ struct bpf_struct_ops bpf_tcp_congestion_ops = {
 	.verifier_ops = &bpf_tcp_ca_verifier_ops,
 	.reg = bpf_tcp_ca_reg,
 	.unreg = bpf_tcp_ca_unreg,
+	.update = bpf_tcp_ca_update,
 	.check_member = bpf_tcp_ca_check_member,
 	.init_member = bpf_tcp_ca_init_member,
 	.init = bpf_tcp_ca_init,
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 9cf1deaf21f2..d6c5a022ae28 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1555,12 +1555,23 @@ union bpf_attr {
 
 	struct { /* struct used by BPF_LINK_UPDATE command */
 		__u32		link_fd;	/* link fd */
-		/* new program fd to update link with */
-		__u32		new_prog_fd;
+		union {
+			/* new program fd to update link with */
+			__u32		new_prog_fd;
+			/* new struct_ops map fd to update link with */
+			__u32           new_map_fd;
+		};
 		__u32		flags;		/* extra flags */
-		/* expected link's program fd; is specified only if
-		 * BPF_F_REPLACE flag is set in flags */
-		__u32		old_prog_fd;
+		union {
+			/* expected link's program fd; is specified only if
+			 * BPF_F_REPLACE flag is set in flags.
+			 */
+			__u32		old_prog_fd;
+			/* expected link's map fd; is specified only
+			 * if BPF_F_REPLACE flag is set.
+			 */
+			__u32           old_map_fd;
+		};
 	} link_update;
 
 	struct {
-- 
cgit v1.2.3-70-g09d2


From 912dd4b0c2a50a839301520badb241f36664ae07 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:24:03 -0700
Subject: libbpf: Update a bpf_link with another struct_ops.

Introduce bpf_link__update_map(), which allows to atomically update
underlying struct_ops implementation for given struct_ops BPF link.

Also add old_map_fd to struct bpf_link_update_opts to handle
BPF_F_REPLACE feature.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Link: https://lore.kernel.org/r/20230323032405.3735486-7-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/lib/bpf/bpf.c      |  8 +++++++-
 tools/lib/bpf/bpf.h      |  3 ++-
 tools/lib/bpf/libbpf.c   | 35 +++++++++++++++++++++++++++++++++++
 tools/lib/bpf/libbpf.h   |  1 +
 tools/lib/bpf/libbpf.map |  1 +
 5 files changed, 46 insertions(+), 2 deletions(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index e750b6f5fcc3..767035900354 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -794,11 +794,17 @@ int bpf_link_update(int link_fd, int new_prog_fd,
 	if (!OPTS_VALID(opts, bpf_link_update_opts))
 		return libbpf_err(-EINVAL);
 
+	if (OPTS_GET(opts, old_prog_fd, 0) && OPTS_GET(opts, old_map_fd, 0))
+		return libbpf_err(-EINVAL);
+
 	memset(&attr, 0, attr_sz);
 	attr.link_update.link_fd = link_fd;
 	attr.link_update.new_prog_fd = new_prog_fd;
 	attr.link_update.flags = OPTS_GET(opts, flags, 0);
-	attr.link_update.old_prog_fd = OPTS_GET(opts, old_prog_fd, 0);
+	if (OPTS_GET(opts, old_prog_fd, 0))
+		attr.link_update.old_prog_fd = OPTS_GET(opts, old_prog_fd, 0);
+	else if (OPTS_GET(opts, old_map_fd, 0))
+		attr.link_update.old_map_fd = OPTS_GET(opts, old_map_fd, 0);
 
 	ret = sys_bpf(BPF_LINK_UPDATE, &attr, attr_sz);
 	return libbpf_err_errno(ret);
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index f0f786373238..b073e73439ef 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -336,8 +336,9 @@ struct bpf_link_update_opts {
 	size_t sz; /* size of this struct for forward/backward compatibility */
 	__u32 flags;	   /* extra flags */
 	__u32 old_prog_fd; /* expected old program FD */
+	__u32 old_map_fd;  /* expected old map FD */
 };
-#define bpf_link_update_opts__last_field old_prog_fd
+#define bpf_link_update_opts__last_field old_map_fd
 
 LIBBPF_API int bpf_link_update(int link_fd, int new_prog_fd,
 			       const struct bpf_link_update_opts *opts);
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 4ff82baba0ea..2b5ea2500b80 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -11677,6 +11677,41 @@ struct bpf_link *bpf_map__attach_struct_ops(const struct bpf_map *map)
 	return &link->link;
 }
 
+/*
+ * Swap the back struct_ops of a link with a new struct_ops map.
+ */
+int bpf_link__update_map(struct bpf_link *link, const struct bpf_map *map)
+{
+	struct bpf_link_struct_ops *st_ops_link;
+	__u32 zero = 0;
+	int err;
+
+	if (!bpf_map__is_struct_ops(map) || map->fd < 0)
+		return -EINVAL;
+
+	st_ops_link = container_of(link, struct bpf_link_struct_ops, link);
+	/* Ensure the type of a link is correct */
+	if (st_ops_link->map_fd < 0)
+		return -EINVAL;
+
+	err = bpf_map_update_elem(map->fd, &zero, map->st_ops->kern_vdata, 0);
+	/* It can be EBUSY if the map has been used to create or
+	 * update a link before.  We don't allow updating the value of
+	 * a struct_ops once it is set.  That ensures that the value
+	 * never changed.  So, it is safe to skip EBUSY.
+	 */
+	if (err && err != -EBUSY)
+		return err;
+
+	err = bpf_link_update(link->fd, map->fd, NULL);
+	if (err < 0)
+		return err;
+
+	st_ops_link->map_fd = map->fd;
+
+	return 0;
+}
+
 typedef enum bpf_perf_event_ret (*bpf_perf_event_print_t)(struct perf_event_header *hdr,
 							  void *private_data);
 
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index db4992a036f8..1615e55e2e79 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -719,6 +719,7 @@ bpf_program__attach_freplace(const struct bpf_program *prog,
 struct bpf_map;
 
 LIBBPF_API struct bpf_link *bpf_map__attach_struct_ops(const struct bpf_map *map);
+LIBBPF_API int bpf_link__update_map(struct bpf_link *link, const struct bpf_map *map);
 
 struct bpf_iter_attach_opts {
 	size_t sz; /* size of this struct for forward/backward compatibility */
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 50dde1f6521e..a5aa3a383d69 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -386,6 +386,7 @@ LIBBPF_1.1.0 {
 LIBBPF_1.2.0 {
 	global:
 		bpf_btf_get_info_by_fd;
+		bpf_link__update_map;
 		bpf_link_get_info_by_fd;
 		bpf_map_get_info_by_fd;
 		bpf_prog_get_info_by_fd;
-- 
cgit v1.2.3-70-g09d2


From 809a69d61899f6b79a8e9de474cd93249580fdaf Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:24:04 -0700
Subject: libbpf: Use .struct_ops.link section to indicate a struct_ops with a
 link.

Flags a struct_ops is to back a bpf_link by putting it to the
".struct_ops.link" section.  Once it is flagged, the created
struct_ops can be used to create a bpf_link or update a bpf_link that
has been backed by another struct_ops.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230323032405.3735486-8-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/lib/bpf/libbpf.c | 60 ++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 44 insertions(+), 16 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 2b5ea2500b80..f6a071db5c6e 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -468,6 +468,7 @@ struct bpf_struct_ops {
 #define KCONFIG_SEC ".kconfig"
 #define KSYMS_SEC ".ksyms"
 #define STRUCT_OPS_SEC ".struct_ops"
+#define STRUCT_OPS_LINK_SEC ".struct_ops.link"
 
 enum libbpf_map_type {
 	LIBBPF_MAP_UNSPEC,
@@ -597,6 +598,7 @@ struct elf_state {
 	Elf64_Ehdr *ehdr;
 	Elf_Data *symbols;
 	Elf_Data *st_ops_data;
+	Elf_Data *st_ops_link_data;
 	size_t shstrndx; /* section index for section name strings */
 	size_t strtabidx;
 	struct elf_sec_desc *secs;
@@ -606,6 +608,7 @@ struct elf_state {
 	int text_shndx;
 	int symbols_shndx;
 	int st_ops_shndx;
+	int st_ops_link_shndx;
 };
 
 struct usdt_manager;
@@ -1119,7 +1122,8 @@ static int bpf_object__init_kern_struct_ops_maps(struct bpf_object *obj)
 	return 0;
 }
 
-static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
+static int init_struct_ops_maps(struct bpf_object *obj, const char *sec_name,
+				int shndx, Elf_Data *data, __u32 map_flags)
 {
 	const struct btf_type *type, *datasec;
 	const struct btf_var_secinfo *vsi;
@@ -1130,15 +1134,15 @@ static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
 	struct bpf_map *map;
 	__u32 i;
 
-	if (obj->efile.st_ops_shndx == -1)
+	if (shndx == -1)
 		return 0;
 
 	btf = obj->btf;
-	datasec_id = btf__find_by_name_kind(btf, STRUCT_OPS_SEC,
+	datasec_id = btf__find_by_name_kind(btf, sec_name,
 					    BTF_KIND_DATASEC);
 	if (datasec_id < 0) {
 		pr_warn("struct_ops init: DATASEC %s not found\n",
-			STRUCT_OPS_SEC);
+			sec_name);
 		return -EINVAL;
 	}
 
@@ -1151,7 +1155,7 @@ static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
 		type_id = btf__resolve_type(obj->btf, vsi->type);
 		if (type_id < 0) {
 			pr_warn("struct_ops init: Cannot resolve var type_id %u in DATASEC %s\n",
-				vsi->type, STRUCT_OPS_SEC);
+				vsi->type, sec_name);
 			return -EINVAL;
 		}
 
@@ -1170,7 +1174,7 @@ static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
 		if (IS_ERR(map))
 			return PTR_ERR(map);
 
-		map->sec_idx = obj->efile.st_ops_shndx;
+		map->sec_idx = shndx;
 		map->sec_offset = vsi->offset;
 		map->name = strdup(var_name);
 		if (!map->name)
@@ -1180,6 +1184,7 @@ static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
 		map->def.key_size = sizeof(int);
 		map->def.value_size = type->size;
 		map->def.max_entries = 1;
+		map->def.map_flags = map_flags;
 
 		map->st_ops = calloc(1, sizeof(*map->st_ops));
 		if (!map->st_ops)
@@ -1192,14 +1197,14 @@ static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
 		if (!st_ops->data || !st_ops->progs || !st_ops->kern_func_off)
 			return -ENOMEM;
 
-		if (vsi->offset + type->size > obj->efile.st_ops_data->d_size) {
+		if (vsi->offset + type->size > data->d_size) {
 			pr_warn("struct_ops init: var %s is beyond the end of DATASEC %s\n",
-				var_name, STRUCT_OPS_SEC);
+				var_name, sec_name);
 			return -EINVAL;
 		}
 
 		memcpy(st_ops->data,
-		       obj->efile.st_ops_data->d_buf + vsi->offset,
+		       data->d_buf + vsi->offset,
 		       type->size);
 		st_ops->tname = tname;
 		st_ops->type = type;
@@ -1212,6 +1217,19 @@ static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
 	return 0;
 }
 
+static int bpf_object_init_struct_ops(struct bpf_object *obj)
+{
+	int err;
+
+	err = init_struct_ops_maps(obj, STRUCT_OPS_SEC, obj->efile.st_ops_shndx,
+				   obj->efile.st_ops_data, 0);
+	err = err ?: init_struct_ops_maps(obj, STRUCT_OPS_LINK_SEC,
+					  obj->efile.st_ops_link_shndx,
+					  obj->efile.st_ops_link_data,
+					  BPF_F_LINK);
+	return err;
+}
+
 static struct bpf_object *bpf_object__new(const char *path,
 					  const void *obj_buf,
 					  size_t obj_buf_sz,
@@ -1248,6 +1266,7 @@ static struct bpf_object *bpf_object__new(const char *path,
 	obj->efile.obj_buf_sz = obj_buf_sz;
 	obj->efile.btf_maps_shndx = -1;
 	obj->efile.st_ops_shndx = -1;
+	obj->efile.st_ops_link_shndx = -1;
 	obj->kconfig_map_idx = -1;
 
 	obj->kern_version = get_kernel_version();
@@ -1265,6 +1284,7 @@ static void bpf_object__elf_finish(struct bpf_object *obj)
 	obj->efile.elf = NULL;
 	obj->efile.symbols = NULL;
 	obj->efile.st_ops_data = NULL;
+	obj->efile.st_ops_link_data = NULL;
 
 	zfree(&obj->efile.secs);
 	obj->efile.sec_cnt = 0;
@@ -2619,7 +2639,7 @@ static int bpf_object__init_maps(struct bpf_object *obj,
 	err = bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
 	err = err ?: bpf_object__init_global_data_maps(obj);
 	err = err ?: bpf_object__init_kconfig_map(obj);
-	err = err ?: bpf_object__init_struct_ops_maps(obj);
+	err = err ?: bpf_object_init_struct_ops(obj);
 
 	return err;
 }
@@ -2753,12 +2773,13 @@ static bool libbpf_needs_btf(const struct bpf_object *obj)
 {
 	return obj->efile.btf_maps_shndx >= 0 ||
 	       obj->efile.st_ops_shndx >= 0 ||
+	       obj->efile.st_ops_link_shndx >= 0 ||
 	       obj->nr_extern > 0;
 }
 
 static bool kernel_needs_btf(const struct bpf_object *obj)
 {
-	return obj->efile.st_ops_shndx >= 0;
+	return obj->efile.st_ops_shndx >= 0 || obj->efile.st_ops_link_shndx >= 0;
 }
 
 static int bpf_object__init_btf(struct bpf_object *obj,
@@ -3451,6 +3472,9 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 			} else if (strcmp(name, STRUCT_OPS_SEC) == 0) {
 				obj->efile.st_ops_data = data;
 				obj->efile.st_ops_shndx = idx;
+			} else if (strcmp(name, STRUCT_OPS_LINK_SEC) == 0) {
+				obj->efile.st_ops_link_data = data;
+				obj->efile.st_ops_link_shndx = idx;
 			} else {
 				pr_info("elf: skipping unrecognized data section(%d) %s\n",
 					idx, name);
@@ -3465,6 +3489,7 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 			/* Only do relo for section with exec instructions */
 			if (!section_have_execinstr(obj, targ_sec_idx) &&
 			    strcmp(name, ".rel" STRUCT_OPS_SEC) &&
+			    strcmp(name, ".rel" STRUCT_OPS_LINK_SEC) &&
 			    strcmp(name, ".rel" MAPS_ELF_SEC)) {
 				pr_info("elf: skipping relo section(%d) %s for section(%d) %s\n",
 					idx, name, targ_sec_idx,
@@ -6611,7 +6636,7 @@ static int bpf_object__collect_relos(struct bpf_object *obj)
 			return -LIBBPF_ERRNO__INTERNAL;
 		}
 
-		if (idx == obj->efile.st_ops_shndx)
+		if (idx == obj->efile.st_ops_shndx || idx == obj->efile.st_ops_link_shndx)
 			err = bpf_object__collect_st_ops_relos(obj, shdr, data);
 		else if (idx == obj->efile.btf_maps_shndx)
 			err = bpf_object__collect_map_relos(obj, shdr, data);
@@ -8853,6 +8878,7 @@ const char *libbpf_bpf_prog_type_str(enum bpf_prog_type t)
 }
 
 static struct bpf_map *find_struct_ops_map_by_offset(struct bpf_object *obj,
+						     int sec_idx,
 						     size_t offset)
 {
 	struct bpf_map *map;
@@ -8862,7 +8888,8 @@ static struct bpf_map *find_struct_ops_map_by_offset(struct bpf_object *obj,
 		map = &obj->maps[i];
 		if (!bpf_map__is_struct_ops(map))
 			continue;
-		if (map->sec_offset <= offset &&
+		if (map->sec_idx == sec_idx &&
+		    map->sec_offset <= offset &&
 		    offset - map->sec_offset < map->def.value_size)
 			return map;
 	}
@@ -8904,7 +8931,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 		}
 
 		name = elf_sym_str(obj, sym->st_name) ?: "<?>";
-		map = find_struct_ops_map_by_offset(obj, rel->r_offset);
+		map = find_struct_ops_map_by_offset(obj, shdr->sh_info, rel->r_offset);
 		if (!map) {
 			pr_warn("struct_ops reloc: cannot find map at rel->r_offset %zu\n",
 				(size_t)rel->r_offset);
@@ -8971,8 +8998,9 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 		}
 
 		/* struct_ops BPF prog can be re-used between multiple
-		 * .struct_ops as long as it's the same struct_ops struct
-		 * definition and the same function pointer field
+		 * .struct_ops & .struct_ops.link as long as it's the
+		 * same struct_ops struct definition and the same
+		 * function pointer field
 		 */
 		if (prog->attach_btf_id != st_ops->type_id ||
 		    prog->expected_attach_type != member_idx) {
-- 
cgit v1.2.3-70-g09d2


From 06da9f3bd6418e06719f15340202996f7a4c258d Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <kuifeng@meta.com>
Date: Wed, 22 Mar 2023 20:24:05 -0700
Subject: selftests/bpf: Test switching TCP Congestion Control algorithms.

Create a pair of sockets that utilize the congestion control algorithm
under a particular name. Then switch up this congestion control
algorithm to another implementation and check whether newly created
connections using the same cc name now run the new implementation.

Also, try to update a link with a struct_ops that is without
BPF_F_LINK or with a wrong or different name.  These cases should fail
due to the violation of assumptions.  To update a bpf_link of a
struct_ops, it must be replaced with another struct_ops that is
identical in type and name and has the BPF_F_LINK flag.

The other test case is to create links from the same struct_ops more
than once.  It makes sure a struct_ops can be used repeatly.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Link: https://lore.kernel.org/r/20230323032405.3735486-9-kuifeng@meta.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 160 +++++++++++++++++++++
 tools/testing/selftests/bpf/progs/tcp_ca_update.c  |  80 +++++++++++
 2 files changed, 240 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/tcp_ca_update.c

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index e980188d4124..a53c254c6058 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -8,6 +8,7 @@
 #include "bpf_dctcp.skel.h"
 #include "bpf_cubic.skel.h"
 #include "bpf_tcp_nogpl.skel.h"
+#include "tcp_ca_update.skel.h"
 #include "bpf_dctcp_release.skel.h"
 #include "tcp_ca_write_sk_pacing.skel.h"
 #include "tcp_ca_incompl_cong_ops.skel.h"
@@ -381,6 +382,155 @@ static void test_unsupp_cong_op(void)
 	libbpf_set_print(old_print_fn);
 }
 
+static void test_update_ca(void)
+{
+	struct tcp_ca_update *skel;
+	struct bpf_link *link;
+	int saved_ca1_cnt;
+	int err;
+
+	skel = tcp_ca_update__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "open"))
+		return;
+
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
+	ASSERT_OK_PTR(link, "attach_struct_ops");
+
+	do_test("tcp_ca_update", NULL);
+	saved_ca1_cnt = skel->bss->ca1_cnt;
+	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
+
+	err = bpf_link__update_map(link, skel->maps.ca_update_2);
+	ASSERT_OK(err, "update_map");
+
+	do_test("tcp_ca_update", NULL);
+	ASSERT_EQ(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
+	ASSERT_GT(skel->bss->ca2_cnt, 0, "ca2_ca2_cnt");
+
+	bpf_link__destroy(link);
+	tcp_ca_update__destroy(skel);
+}
+
+static void test_update_wrong(void)
+{
+	struct tcp_ca_update *skel;
+	struct bpf_link *link;
+	int saved_ca1_cnt;
+	int err;
+
+	skel = tcp_ca_update__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "open"))
+		return;
+
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
+	ASSERT_OK_PTR(link, "attach_struct_ops");
+
+	do_test("tcp_ca_update", NULL);
+	saved_ca1_cnt = skel->bss->ca1_cnt;
+	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
+
+	err = bpf_link__update_map(link, skel->maps.ca_wrong);
+	ASSERT_ERR(err, "update_map");
+
+	do_test("tcp_ca_update", NULL);
+	ASSERT_GT(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
+
+	bpf_link__destroy(link);
+	tcp_ca_update__destroy(skel);
+}
+
+static void test_mixed_links(void)
+{
+	struct tcp_ca_update *skel;
+	struct bpf_link *link, *link_nl;
+	int err;
+
+	skel = tcp_ca_update__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "open"))
+		return;
+
+	link_nl = bpf_map__attach_struct_ops(skel->maps.ca_no_link);
+	ASSERT_OK_PTR(link_nl, "attach_struct_ops_nl");
+
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
+	ASSERT_OK_PTR(link, "attach_struct_ops");
+
+	do_test("tcp_ca_update", NULL);
+	ASSERT_GT(skel->bss->ca1_cnt, 0, "ca1_ca1_cnt");
+
+	err = bpf_link__update_map(link, skel->maps.ca_no_link);
+	ASSERT_ERR(err, "update_map");
+
+	bpf_link__destroy(link);
+	bpf_link__destroy(link_nl);
+	tcp_ca_update__destroy(skel);
+}
+
+static void test_multi_links(void)
+{
+	struct tcp_ca_update *skel;
+	struct bpf_link *link;
+
+	skel = tcp_ca_update__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "open"))
+		return;
+
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
+	ASSERT_OK_PTR(link, "attach_struct_ops_1st");
+	bpf_link__destroy(link);
+
+	/* A map should be able to be used to create links multiple
+	 * times.
+	 */
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
+	ASSERT_OK_PTR(link, "attach_struct_ops_2nd");
+	bpf_link__destroy(link);
+
+	tcp_ca_update__destroy(skel);
+}
+
+static void test_link_replace(void)
+{
+	DECLARE_LIBBPF_OPTS(bpf_link_update_opts, opts);
+	struct tcp_ca_update *skel;
+	struct bpf_link *link;
+	int err;
+
+	skel = tcp_ca_update__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "open"))
+		return;
+
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
+	ASSERT_OK_PTR(link, "attach_struct_ops_1st");
+	bpf_link__destroy(link);
+
+	link = bpf_map__attach_struct_ops(skel->maps.ca_update_2);
+	ASSERT_OK_PTR(link, "attach_struct_ops_2nd");
+
+	/* BPF_F_REPLACE with a wrong old map Fd. It should fail!
+	 *
+	 * With BPF_F_REPLACE, the link should be updated only if the
+	 * old map fd given here matches the map backing the link.
+	 */
+	opts.old_map_fd = bpf_map__fd(skel->maps.ca_update_1);
+	opts.flags = BPF_F_REPLACE;
+	err = bpf_link_update(bpf_link__fd(link),
+			      bpf_map__fd(skel->maps.ca_update_1),
+			      &opts);
+	ASSERT_ERR(err, "bpf_link_update_fail");
+
+	/* BPF_F_REPLACE with a correct old map Fd. It should success! */
+	opts.old_map_fd = bpf_map__fd(skel->maps.ca_update_2);
+	err = bpf_link_update(bpf_link__fd(link),
+			      bpf_map__fd(skel->maps.ca_update_1),
+			      &opts);
+	ASSERT_OK(err, "bpf_link_update_success");
+
+	bpf_link__destroy(link);
+
+	tcp_ca_update__destroy(skel);
+}
+
 void test_bpf_tcp_ca(void)
 {
 	if (test__start_subtest("dctcp"))
@@ -399,4 +549,14 @@ void test_bpf_tcp_ca(void)
 		test_incompl_cong_ops();
 	if (test__start_subtest("unsupp_cong_op"))
 		test_unsupp_cong_op();
+	if (test__start_subtest("update_ca"))
+		test_update_ca();
+	if (test__start_subtest("update_wrong"))
+		test_update_wrong();
+	if (test__start_subtest("mixed_links"))
+		test_mixed_links();
+	if (test__start_subtest("multi_links"))
+		test_multi_links();
+	if (test__start_subtest("link_replace"))
+		test_link_replace();
 }
diff --git a/tools/testing/selftests/bpf/progs/tcp_ca_update.c b/tools/testing/selftests/bpf/progs/tcp_ca_update.c
new file mode 100644
index 000000000000..b93a0ed33057
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/tcp_ca_update.c
@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include "vmlinux.h"
+
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+char _license[] SEC("license") = "GPL";
+
+int ca1_cnt = 0;
+int ca2_cnt = 0;
+
+static inline struct tcp_sock *tcp_sk(const struct sock *sk)
+{
+	return (struct tcp_sock *)sk;
+}
+
+SEC("struct_ops/ca_update_1_init")
+void BPF_PROG(ca_update_1_init, struct sock *sk)
+{
+	ca1_cnt++;
+}
+
+SEC("struct_ops/ca_update_2_init")
+void BPF_PROG(ca_update_2_init, struct sock *sk)
+{
+	ca2_cnt++;
+}
+
+SEC("struct_ops/ca_update_cong_control")
+void BPF_PROG(ca_update_cong_control, struct sock *sk,
+	      const struct rate_sample *rs)
+{
+}
+
+SEC("struct_ops/ca_update_ssthresh")
+__u32 BPF_PROG(ca_update_ssthresh, struct sock *sk)
+{
+	return tcp_sk(sk)->snd_ssthresh;
+}
+
+SEC("struct_ops/ca_update_undo_cwnd")
+__u32 BPF_PROG(ca_update_undo_cwnd, struct sock *sk)
+{
+	return tcp_sk(sk)->snd_cwnd;
+}
+
+SEC(".struct_ops.link")
+struct tcp_congestion_ops ca_update_1 = {
+	.init = (void *)ca_update_1_init,
+	.cong_control = (void *)ca_update_cong_control,
+	.ssthresh = (void *)ca_update_ssthresh,
+	.undo_cwnd = (void *)ca_update_undo_cwnd,
+	.name = "tcp_ca_update",
+};
+
+SEC(".struct_ops.link")
+struct tcp_congestion_ops ca_update_2 = {
+	.init = (void *)ca_update_2_init,
+	.cong_control = (void *)ca_update_cong_control,
+	.ssthresh = (void *)ca_update_ssthresh,
+	.undo_cwnd = (void *)ca_update_undo_cwnd,
+	.name = "tcp_ca_update",
+};
+
+SEC(".struct_ops.link")
+struct tcp_congestion_ops ca_wrong = {
+	.cong_control = (void *)ca_update_cong_control,
+	.ssthresh = (void *)ca_update_ssthresh,
+	.undo_cwnd = (void *)ca_update_undo_cwnd,
+	.name = "tcp_ca_wrong",
+};
+
+SEC(".struct_ops")
+struct tcp_congestion_ops ca_no_link = {
+	.cong_control = (void *)ca_update_cong_control,
+	.ssthresh = (void *)ca_update_ssthresh,
+	.undo_cwnd = (void *)ca_update_undo_cwnd,
+	.name = "tcp_ca_no_link",
+};
-- 
cgit v1.2.3-70-g09d2


From 5c4a9aa856c706def9239d1e43c4ea9fccb5c75a Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Tue, 21 Mar 2023 12:51:59 +0100
Subject: net: ipv4: Allow changing IPv4 address protocol

When IP address protocol field was added in commit 47f0bd503210 ("net: Add
new protocol attribute to IP addresses"), the semantics included the
ability to change the protocol for IPv6 addresses, but not for IPv4
addresses. It seems this was not deliberate, but rather by accident.

A userspace that wants to change the protocol of an address might drop and
recreate the address, but that disrupts routing and is just impractical.

So in this patch, when an IPv4 address is replaced (through RTM_NEWADDR
request with NLM_F_REPLACE flag), update the proto at the address to the
one given in the request, or zero if none is given. This matches the
behavior of IPv6. Previously, any new value given was simply ignored.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/devinet.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index b0acf6e19aed..5deac0517ef7 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -962,6 +962,7 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh,
 					 extack);
 	} else {
 		u32 new_metric = ifa->ifa_rt_priority;
+		u8 new_proto = ifa->ifa_proto;
 
 		inet_free_ifa(ifa);
 
@@ -975,6 +976,8 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh,
 			ifa->ifa_rt_priority = new_metric;
 		}
 
+		ifa->ifa_proto = new_proto;
+
 		set_ifa_lifetime(ifa, valid_lft, prefered_lft);
 		cancel_delayed_work(&check_lifetime_work);
 		queue_delayed_work(system_power_efficient_wq,
-- 
cgit v1.2.3-70-g09d2


From ecb3c1e675c719885ac05bb5473fa5c495d1ad24 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Tue, 21 Mar 2023 12:52:00 +0100
Subject: selftests: rtnetlink: Make the set of tests to run configurable

Extract the list of all tests into a variable, ALL_TESTS. Then assume the
environment variable TESTS holds the list of tests to actually run, falling
back to ALL_TESTS if TESTS is empty. This is the same interface that
forwarding selftests use to make the set of tests to run configurable.
In addition to this, allow setting the value explicitly through a command
line option "-t" along the lines of what fib_nexthops.sh does.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/rtnetlink.sh | 90 +++++++++++++++++---------------
 1 file changed, 48 insertions(+), 42 deletions(-)

diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh
index 275491be3da2..12caf9602353 100755
--- a/tools/testing/selftests/net/rtnetlink.sh
+++ b/tools/testing/selftests/net/rtnetlink.sh
@@ -4,6 +4,30 @@
 #
 # set -e
 
+ALL_TESTS="
+	kci_test_polrouting
+	kci_test_route_get
+	kci_test_addrlft
+	kci_test_promote_secondaries
+	kci_test_tc
+	kci_test_gre
+	kci_test_gretap
+	kci_test_ip6gretap
+	kci_test_erspan
+	kci_test_ip6erspan
+	kci_test_bridge
+	kci_test_addrlabel
+	kci_test_ifalias
+	kci_test_vrf
+	kci_test_encap
+	kci_test_macsec
+	kci_test_ipsec
+	kci_test_ipsec_offload
+	kci_test_fdb_get
+	kci_test_neigh_get
+	kci_test_bridge_parent_id
+"
+
 devdummy="test-dummy0"
 
 # Kselftest framework requirement - SKIP code is 4.
@@ -1227,60 +1251,34 @@ kci_test_bridge_parent_id()
 
 kci_test_rtnl()
 {
+	local current_test
 	local ret=0
+
 	kci_add_dummy
 	if [ $ret -ne 0 ];then
 		echo "FAIL: cannot add dummy interface"
 		return 1
 	fi
 
-	kci_test_polrouting
-	check_err $?
-	kci_test_route_get
-	check_err $?
-	kci_test_addrlft
-	check_err $?
-	kci_test_promote_secondaries
-	check_err $?
-	kci_test_tc
-	check_err $?
-	kci_test_gre
-	check_err $?
-	kci_test_gretap
-	check_err $?
-	kci_test_ip6gretap
-	check_err $?
-	kci_test_erspan
-	check_err $?
-	kci_test_ip6erspan
-	check_err $?
-	kci_test_bridge
-	check_err $?
-	kci_test_addrlabel
-	check_err $?
-	kci_test_ifalias
-	check_err $?
-	kci_test_vrf
-	check_err $?
-	kci_test_encap
-	check_err $?
-	kci_test_macsec
-	check_err $?
-	kci_test_ipsec
-	check_err $?
-	kci_test_ipsec_offload
-	check_err $?
-	kci_test_fdb_get
-	check_err $?
-	kci_test_neigh_get
-	check_err $?
-	kci_test_bridge_parent_id
-	check_err $?
+	for current_test in ${TESTS:-$ALL_TESTS}; do
+		$current_test
+		check_err $?
+	done
 
 	kci_del_dummy
 	return $ret
 }
 
+usage()
+{
+	cat <<EOF
+usage: ${0##*/} OPTS
+
+        -t <test>   Test(s) to run (default: all)
+                    (options: $(echo $ALL_TESTS))
+EOF
+}
+
 #check for needed privileges
 if [ "$(id -u)" -ne 0 ];then
 	echo "SKIP: Need root privileges"
@@ -1295,6 +1293,14 @@ for x in ip tc;do
 	fi
 done
 
+while getopts t:h o; do
+	case $o in
+		t) TESTS=$OPTARG;;
+		h) usage; exit 0;;
+		*) usage; exit 1;;
+	esac
+done
+
 kci_test_rtnl
 
 exit $?
-- 
cgit v1.2.3-70-g09d2


From 6a414fd77f613e374f2f1accb36beca90bab084d Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Tue, 21 Mar 2023 12:52:01 +0100
Subject: selftests: rtnetlink: Add an address proto test

Add coverage of "ip address {add,replace} ... proto" support.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/rtnetlink.sh | 91 ++++++++++++++++++++++++++++++++
 1 file changed, 91 insertions(+)

diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh
index 12caf9602353..3b15c686c03f 100755
--- a/tools/testing/selftests/net/rtnetlink.sh
+++ b/tools/testing/selftests/net/rtnetlink.sh
@@ -26,6 +26,7 @@ ALL_TESTS="
 	kci_test_fdb_get
 	kci_test_neigh_get
 	kci_test_bridge_parent_id
+	kci_test_address_proto
 "
 
 devdummy="test-dummy0"
@@ -1249,6 +1250,96 @@ kci_test_bridge_parent_id()
 	echo "PASS: bridge_parent_id"
 }
 
+address_get_proto()
+{
+	local addr=$1; shift
+
+	ip -N -j address show dev "$devdummy" |
+	    jq -e -r --arg addr "${addr%/*}" \
+	       '.[].addr_info[] | select(.local == $addr) | .protocol'
+}
+
+address_count()
+{
+	ip -N -j address show dev "$devdummy" "$@" |
+	    jq -e -r '[.[].addr_info[] | .local | select(. != null)] | length'
+}
+
+do_test_address_proto()
+{
+	local what=$1; shift
+	local addr=$1; shift
+	local addr2=${addr%/*}2/${addr#*/}
+	local addr3=${addr%/*}3/${addr#*/}
+	local proto
+	local count
+	local ret=0
+	local err
+
+	ip address add dev "$devdummy" "$addr3"
+	check_err $?
+	proto=$(address_get_proto "$addr3")
+	[[ "$proto" == null ]]
+	check_err $?
+
+	ip address add dev "$devdummy" "$addr2" proto 0x99
+	check_err $?
+	proto=$(address_get_proto "$addr2")
+	[[ "$proto" == 0x99 ]]
+	check_err $?
+
+	ip address add dev "$devdummy" "$addr" proto 0xab
+	check_err $?
+	proto=$(address_get_proto "$addr")
+	[[ "$proto" == 0xab ]]
+	check_err $?
+
+	ip address replace dev "$devdummy" "$addr" proto 0x11
+	proto=$(address_get_proto "$addr")
+	check_err $?
+	[[ "$proto" == 0x11 ]]
+	check_err $?
+
+	count=$(address_count)
+	check_err $?
+	(( count == 3 )) # $addr, $addr2 and $addr3
+
+	count=$(address_count proto 0)
+	check_err $?
+	(( count == 1 )) # just $addr2
+
+	count=$(address_count proto 0x11)
+	check_err $?
+	(( count == 2 )) # $addr and $addr2
+
+	count=$(address_count proto 0xab)
+	check_err $?
+	(( count == 1 )) # just $addr2
+
+	ip address del dev "$devdummy" "$addr"
+	ip address del dev "$devdummy" "$addr2"
+	ip address del dev "$devdummy" "$addr3"
+
+	if [ $ret -ne 0 ]; then
+		echo "FAIL: address proto $what"
+		return 1
+	fi
+	echo "PASS: address proto $what"
+}
+
+kci_test_address_proto()
+{
+	local ret=0
+
+	do_test_address_proto IPv4 192.0.2.1/28
+	check_err $?
+
+	do_test_address_proto IPv6 2001:db8:1::1/64
+	check_err $?
+
+	return $ret
+}
+
 kci_test_rtnl()
 {
 	local current_test
-- 
cgit v1.2.3-70-g09d2


From fcb3a4653bc5fb0525d957db0cc8b413252029f8 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Tue, 21 Mar 2023 19:33:45 -0300
Subject: net/sched: act_api: use the correct TCA_ACT attributes in dump

4 places in the act api code are using 'TCA_' definitions where they
should be using 'TCA_ACT_', which is confusing for the reader, although
functionally they are equivalent.

Cc: Hangbin Liu <haliu@redhat.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Acked-by: Hangbin Liu <haliu@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_api.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 296fc1afedd8..f7887f42d542 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -453,7 +453,7 @@ static size_t tcf_action_shared_attrs_size(const struct tc_action *act)
 		+ nla_total_size_64bit(sizeof(u64))
 		/* TCA_STATS_QUEUE */
 		+ nla_total_size_64bit(sizeof(struct gnet_stats_queue))
-		+ nla_total_size(0) /* TCA_OPTIONS nested */
+		+ nla_total_size(0) /* TCA_ACT_OPTIONS nested */
 		+ nla_total_size(sizeof(struct tcf_t)); /* TCA_GACT_TM */
 }
 
@@ -480,7 +480,7 @@ tcf_action_dump_terse(struct sk_buff *skb, struct tc_action *a, bool from_act)
 	unsigned char *b = skb_tail_pointer(skb);
 	struct tc_cookie *cookie;
 
-	if (nla_put_string(skb, TCA_KIND, a->ops->kind))
+	if (nla_put_string(skb, TCA_ACT_KIND, a->ops->kind))
 		goto nla_put_failure;
 	if (tcf_action_copy_stats(skb, a, 0))
 		goto nla_put_failure;
@@ -598,7 +598,7 @@ static int tcf_del_walker(struct tcf_idrinfo *idrinfo, struct sk_buff *skb,
 	nest = nla_nest_start_noflag(skb, 0);
 	if (nest == NULL)
 		goto nla_put_failure;
-	if (nla_put_string(skb, TCA_KIND, ops->kind))
+	if (nla_put_string(skb, TCA_ACT_KIND, ops->kind))
 		goto nla_put_failure;
 
 	ret = 0;
@@ -1189,7 +1189,7 @@ tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 	if (nla_put_u32(skb, TCA_ACT_IN_HW_COUNT, a->in_hw_count))
 		goto nla_put_failure;
 
-	nest = nla_nest_start_noflag(skb, TCA_OPTIONS);
+	nest = nla_nest_start_noflag(skb, TCA_ACT_OPTIONS);
 	if (nest == NULL)
 		goto nla_put_failure;
 	err = tcf_action_dump_old(skb, a, bind, ref);
-- 
cgit v1.2.3-70-g09d2


From 3eb8eea2a453463f5606ce3e46cf225f88671440 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 21 Mar 2023 22:38:48 -0700
Subject: docs: networking: document NAPI
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add basic documentation about NAPI. We can stop linking to the ancient
doc on the LF wiki.

Link: https://lore.kernel.org/all/20230315223044.471002-1-kuba@kernel.org/
Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> # for ctucanfd-driver.rst
Reviewed-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20230322053848.198452-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../device_drivers/can/ctu/ctucanfd-driver.rst     |   3 +-
 .../device_drivers/ethernet/intel/e100.rst         |   3 +-
 .../device_drivers/ethernet/intel/i40e.rst         |   4 +-
 .../device_drivers/ethernet/intel/ice.rst          |   4 +-
 Documentation/networking/index.rst                 |   1 +
 Documentation/networking/napi.rst                  | 254 +++++++++++++++++++++
 include/linux/netdevice.h                          |  13 +-
 7 files changed, 269 insertions(+), 13 deletions(-)
 create mode 100644 Documentation/networking/napi.rst

diff --git a/Documentation/networking/device_drivers/can/ctu/ctucanfd-driver.rst b/Documentation/networking/device_drivers/can/ctu/ctucanfd-driver.rst
index 1a4fc6607582..1661d13174d5 100644
--- a/Documentation/networking/device_drivers/can/ctu/ctucanfd-driver.rst
+++ b/Documentation/networking/device_drivers/can/ctu/ctucanfd-driver.rst
@@ -229,8 +229,7 @@ frames for a while. This has a potential to avoid the costly round of
 enabling interrupts, handling an incoming IRQ in ISR, re-enabling the
 softirq and switching context back to softirq.
 
-More detailed documentation of NAPI may be found on the pages of Linux
-Foundation `<https://wiki.linuxfoundation.org/networking/napi>`_.
+See :ref:`Documentation/networking/napi.rst <napi>` for more information.
 
 Integrating the core to Xilinx Zynq
 -----------------------------------
diff --git a/Documentation/networking/device_drivers/ethernet/intel/e100.rst b/Documentation/networking/device_drivers/ethernet/intel/e100.rst
index 3d4a9ba21946..371b7e5c3293 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e100.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e100.rst
@@ -151,8 +151,7 @@ NAPI
 
 NAPI (Rx polling mode) is supported in the e100 driver.
 
-See https://wiki.linuxfoundation.org/networking/napi for more
-information on NAPI.
+See :ref:`Documentation/networking/napi.rst <napi>` for more information.
 
 Multiple Interfaces on Same Ethernet Broadcast Network
 ------------------------------------------------------
diff --git a/Documentation/networking/device_drivers/ethernet/intel/i40e.rst b/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
index ac35bd472bdc..c495c4e16b3b 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
@@ -399,8 +399,8 @@ operate only in full duplex and only at their native speed.
 NAPI
 ----
 NAPI (Rx polling mode) is supported in the i40e driver.
-For more information on NAPI, see
-https://wiki.linuxfoundation.org/networking/napi
+
+See :ref:`Documentation/networking/napi.rst <napi>` for more information.
 
 Flow Control
 ------------
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ice.rst b/Documentation/networking/device_drivers/ethernet/intel/ice.rst
index 5efea4dd1251..2b6dc7880d7b 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ice.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ice.rst
@@ -817,10 +817,10 @@ NOTE:
 
 NAPI
 ----
+
 This driver supports NAPI (Rx polling mode).
-For more information on NAPI, see
-https://wiki.linuxfoundation.org/networking/napi
 
+See :ref:`Documentation/networking/napi.rst <napi>` for more information.
 
 MACVLAN
 -------
diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index 4ddcae33c336..24bb256d6d53 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -73,6 +73,7 @@ Contents:
    mpls-sysctl
    mptcp-sysctl
    multiqueue
+   napi
    netconsole
    netdev-features
    netdevices
diff --git a/Documentation/networking/napi.rst b/Documentation/networking/napi.rst
new file mode 100644
index 000000000000..a7a047742e93
--- /dev/null
+++ b/Documentation/networking/napi.rst
@@ -0,0 +1,254 @@
+.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+
+.. _napi:
+
+====
+NAPI
+====
+
+NAPI is the event handling mechanism used by the Linux networking stack.
+The name NAPI no longer stands for anything in particular [#]_.
+
+In basic operation the device notifies the host about new events
+via an interrupt.
+The host then schedules a NAPI instance to process the events.
+The device may also be polled for events via NAPI without receiving
+interrupts first (:ref:`busy polling<poll>`).
+
+NAPI processing usually happens in the software interrupt context,
+but there is an option to use :ref:`separate kernel threads<threaded>`
+for NAPI processing.
+
+All in all NAPI abstracts away from the drivers the context and configuration
+of event (packet Rx and Tx) processing.
+
+Driver API
+==========
+
+The two most important elements of NAPI are the struct napi_struct
+and the associated poll method. struct napi_struct holds the state
+of the NAPI instance while the method is the driver-specific event
+handler. The method will typically free Tx packets that have been
+transmitted and process newly received packets.
+
+.. _drv_ctrl:
+
+Control API
+-----------
+
+netif_napi_add() and netif_napi_del() add/remove a NAPI instance
+from the system. The instances are attached to the netdevice passed
+as argument (and will be deleted automatically when netdevice is
+unregistered). Instances are added in a disabled state.
+
+napi_enable() and napi_disable() manage the disabled state.
+A disabled NAPI can't be scheduled and its poll method is guaranteed
+to not be invoked. napi_disable() waits for ownership of the NAPI
+instance to be released.
+
+The control APIs are not idempotent. Control API calls are safe against
+concurrent use of datapath APIs but an incorrect sequence of control API
+calls may result in crashes, deadlocks, or race conditions. For example,
+calling napi_disable() multiple times in a row will deadlock.
+
+Datapath API
+------------
+
+napi_schedule() is the basic method of scheduling a NAPI poll.
+Drivers should call this function in their interrupt handler
+(see :ref:`drv_sched` for more info). A successful call to napi_schedule()
+will take ownership of the NAPI instance.
+
+Later, after NAPI is scheduled, the driver's poll method will be
+called to process the events/packets. The method takes a ``budget``
+argument - drivers can process completions for any number of Tx
+packets but should only process up to ``budget`` number of
+Rx packets. Rx processing is usually much more expensive.
+
+In other words, it is recommended to ignore the budget argument when
+performing TX buffer reclamation to ensure that the reclamation is not
+arbitrarily bounded; however, it is required to honor the budget argument
+for RX processing.
+
+.. warning::
+
+   The ``budget`` argument may be 0 if core tries to only process Tx completions
+   and no Rx packets.
+
+The poll method returns the amount of work done. If the driver still
+has outstanding work to do (e.g. ``budget`` was exhausted)
+the poll method should return exactly ``budget``. In that case,
+the NAPI instance will be serviced/polled again (without the
+need to be scheduled).
+
+If event processing has been completed (all outstanding packets
+processed) the poll method should call napi_complete_done()
+before returning. napi_complete_done() releases the ownership
+of the instance.
+
+.. warning::
+
+   The case of finishing all events and using exactly ``budget``
+   must be handled carefully. There is no way to report this
+   (rare) condition to the stack, so the driver must either
+   not call napi_complete_done() and wait to be called again,
+   or return ``budget - 1``.
+
+   If the ``budget`` is 0 napi_complete_done() should never be called.
+
+Call sequence
+-------------
+
+Drivers should not make assumptions about the exact sequencing
+of calls. The poll method may be called without the driver scheduling
+the instance (unless the instance is disabled). Similarly,
+it's not guaranteed that the poll method will be called, even
+if napi_schedule() succeeded (e.g. if the instance gets disabled).
+
+As mentioned in the :ref:`drv_ctrl` section - napi_disable() and subsequent
+calls to the poll method only wait for the ownership of the instance
+to be released, not for the poll method to exit. This means that
+drivers should avoid accessing any data structures after calling
+napi_complete_done().
+
+.. _drv_sched:
+
+Scheduling and IRQ masking
+--------------------------
+
+Drivers should keep the interrupts masked after scheduling
+the NAPI instance - until NAPI polling finishes any further
+interrupts are unnecessary.
+
+Drivers which have to mask the interrupts explicitly (as opposed
+to IRQ being auto-masked by the device) should use the napi_schedule_prep()
+and __napi_schedule() calls:
+
+.. code-block:: c
+
+  if (napi_schedule_prep(&v->napi)) {
+      mydrv_mask_rxtx_irq(v->idx);
+      /* schedule after masking to avoid races */
+      __napi_schedule(&v->napi);
+  }
+
+IRQ should only be unmasked after a successful call to napi_complete_done():
+
+.. code-block:: c
+
+  if (budget && napi_complete_done(&v->napi, work_done)) {
+    mydrv_unmask_rxtx_irq(v->idx);
+    return min(work_done, budget - 1);
+  }
+
+napi_schedule_irqoff() is a variant of napi_schedule() which takes advantage
+of guarantees given by being invoked in IRQ context (no need to
+mask interrupts). Note that PREEMPT_RT forces all interrupts
+to be threaded so the interrupt may need to be marked ``IRQF_NO_THREAD``
+to avoid issues on real-time kernel configurations.
+
+Instance to queue mapping
+-------------------------
+
+Modern devices have multiple NAPI instances (struct napi_struct) per
+interface. There is no strong requirement on how the instances are
+mapped to queues and interrupts. NAPI is primarily a polling/processing
+abstraction without specific user-facing semantics. That said, most networking
+devices end up using NAPI in fairly similar ways.
+
+NAPI instances most often correspond 1:1:1 to interrupts and queue pairs
+(queue pair is a set of a single Rx and single Tx queue).
+
+In less common cases a NAPI instance may be used for multiple queues
+or Rx and Tx queues can be serviced by separate NAPI instances on a single
+core. Regardless of the queue assignment, however, there is usually still
+a 1:1 mapping between NAPI instances and interrupts.
+
+It's worth noting that the ethtool API uses a "channel" terminology where
+each channel can be either ``rx``, ``tx`` or ``combined``. It's not clear
+what constitutes a channel; the recommended interpretation is to understand
+a channel as an IRQ/NAPI which services queues of a given type. For example,
+a configuration of 1 ``rx``, 1 ``tx`` and 1 ``combined`` channel is expected
+to utilize 3 interrupts, 2 Rx and 2 Tx queues.
+
+User API
+========
+
+User interactions with NAPI depend on NAPI instance ID. The instance IDs
+are only visible to the user thru the ``SO_INCOMING_NAPI_ID`` socket option.
+It's not currently possible to query IDs used by a given device.
+
+Software IRQ coalescing
+-----------------------
+
+NAPI does not perform any explicit event coalescing by default.
+In most scenarios batching happens due to IRQ coalescing which is done
+by the device. There are cases where software coalescing is helpful.
+
+NAPI can be configured to arm a repoll timer instead of unmasking
+the hardware interrupts as soon as all packets are processed.
+The ``gro_flush_timeout`` sysfs configuration of the netdevice
+is reused to control the delay of the timer, while
+``napi_defer_hard_irqs`` controls the number of consecutive empty polls
+before NAPI gives up and goes back to using hardware IRQs.
+
+.. _poll:
+
+Busy polling
+------------
+
+Busy polling allows a user process to check for incoming packets before
+the device interrupt fires. As is the case with any busy polling it trades
+off CPU cycles for lower latency (production uses of NAPI busy polling
+are not well known).
+
+Busy polling is enabled by either setting ``SO_BUSY_POLL`` on
+selected sockets or using the global ``net.core.busy_poll`` and
+``net.core.busy_read`` sysctls. An io_uring API for NAPI busy polling
+also exists.
+
+IRQ mitigation
+---------------
+
+While busy polling is supposed to be used by low latency applications,
+a similar mechanism can be used for IRQ mitigation.
+
+Very high request-per-second applications (especially routing/forwarding
+applications and especially applications using AF_XDP sockets) may not
+want to be interrupted until they finish processing a request or a batch
+of packets.
+
+Such applications can pledge to the kernel that they will perform a busy
+polling operation periodically, and the driver should keep the device IRQs
+permanently masked. This mode is enabled by using the ``SO_PREFER_BUSY_POLL``
+socket option. To avoid system misbehavior the pledge is revoked
+if ``gro_flush_timeout`` passes without any busy poll call.
+
+The NAPI budget for busy polling is lower than the default (which makes
+sense given the low latency intention of normal busy polling). This is
+not the case with IRQ mitigation, however, so the budget can be adjusted
+with the ``SO_BUSY_POLL_BUDGET`` socket option.
+
+.. _threaded:
+
+Threaded NAPI
+-------------
+
+Threaded NAPI is an operating mode that uses dedicated kernel
+threads rather than software IRQ context for NAPI processing.
+The configuration is per netdevice and will affect all
+NAPI instances of that device. Each NAPI instance will spawn a separate
+thread (called ``napi/${ifc-name}-${napi-id}``).
+
+It is recommended to pin each kernel thread to a single CPU, the same
+CPU as the CPU which services the interrupt. Note that the mapping
+between IRQs and NAPI instances may not be trivial (and is driver
+dependent). The NAPI instance IDs will be assigned in the opposite
+order than the process IDs of the kernel threads.
+
+Threaded NAPI is controlled by writing 0/1 to the ``threaded`` file in
+netdev's sysfs directory.
+
+.. rubric:: Footnotes
+
+.. [#] NAPI was originally referred to as New API in 2.4 Linux.
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 674ee5daa7b1..18a5be6ddd0f 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -509,15 +509,18 @@ static inline bool napi_reschedule(struct napi_struct *napi)
 	return false;
 }
 
-bool napi_complete_done(struct napi_struct *n, int work_done);
 /**
- *	napi_complete - NAPI processing complete
- *	@n: NAPI context
+ * napi_complete_done - NAPI processing complete
+ * @n: NAPI context
+ * @work_done: number of packets processed
  *
- * Mark NAPI processing as complete.
- * Consider using napi_complete_done() instead.
+ * Mark NAPI processing as complete. Should only be called if poll budget
+ * has not been completely consumed.
+ * Prefer over napi_complete().
  * Return false if device should avoid rearming interrupts.
  */
+bool napi_complete_done(struct napi_struct *n, int work_done);
+
 static inline bool napi_complete(struct napi_struct *n)
 {
 	return napi_complete_done(n, 0);
-- 
cgit v1.2.3-70-g09d2


From 8da3a5598f75cd0361e11b6d74697084380eb4b0 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@nvidia.com>
Date: Wed, 22 Mar 2023 16:42:42 +0100
Subject: ynl: allow to encode u8 attr

Playing with dpll netlink, I came across following issue:
$ sudo ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/dpll.yaml --do pin-set --json '{"id": 0, "pin-idx": 1, "pin-state": 1}'
Traceback (most recent call last):
  File "tools/net/ynl/cli.py", line 52, in <module>
    main()
  File "tools/net/ynl/cli.py", line 40, in main
    reply = ynl.do(args.do, attrs)
  File "tools/net/ynl/lib/ynl.py", line 520, in do
    return self._op(method, vals)
  File "tools/net/ynl/lib/ynl.py", line 476, in _op
    msg += self._add_attr(op.attr_set.name, name, value)
  File "tools/net/ynl/lib/ynl.py", line 344, in _add_attr
    raise Exception(f'Unknown type at {space} {name} {value} {attr["type"]}')
Exception: Unknown type at dpll pin-state 1 u8

I'm not that familiar with ynl code, but from a quick peek, I suspect
that couple other types are missing for both encoding and decoding.
Ignoring those here as I'm scratching my local itch only.

Fix the issue by adding u8 attr packing.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/20230322154242.1739136-1-jiri@resnulli.us
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/ynl.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 90764a83c646..bcb798c7734d 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -334,6 +334,8 @@ class YnlFamily(SpecFamily):
                 attr_payload += self._add_attr(attr['nested-attributes'], subname, subvalue)
         elif attr["type"] == 'flag':
             attr_payload = b''
+        elif attr["type"] == 'u8':
+            attr_payload = struct.pack("B", int(value))
         elif attr["type"] == 'u32':
             attr_payload = struct.pack("I", int(value))
         elif attr["type"] == 'string':
-- 
cgit v1.2.3-70-g09d2


From 0c04328ccf8550f29f1357ae90ce460e9568a5a9 Mon Sep 17 00:00:00 2001
From: Alex Elder <elder@linaro.org>
Date: Wed, 22 Mar 2023 09:47:42 -0500
Subject: net: ipa: add IPA v5.0 to ipa_version_string()

In the IPA device sysfs directory, the "version" file can be read to
find out what IPA version is implemented.  The content of this file
is supplied by ipa_version_string(), which needs to be updated to
properly handle IPA v5.0.

Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230322144742.2203947-1-elder@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ipa/ipa_sysfs.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ipa/ipa_sysfs.c b/drivers/net/ipa/ipa_sysfs.c
index 14bd2f903045..2ff09ce343b7 100644
--- a/drivers/net/ipa/ipa_sysfs.c
+++ b/drivers/net/ipa/ipa_sysfs.c
@@ -36,6 +36,8 @@ static const char *ipa_version_string(struct ipa *ipa)
 		return "4.9";
 	case IPA_VERSION_4_11:
 		return "4.11";
+	case IPA_VERSION_5_0:
+		return "5.0";
 	default:
 		return "0.0";	/* Won't happen (checked at probe time) */
 	}
-- 
cgit v1.2.3-70-g09d2


From 4691720f509ae037a3b2b33d008abfd8bb8ed505 Mon Sep 17 00:00:00 2001
From: Maxim Korotkov <korotkov.maxim.s@gmail.com>
Date: Wed, 22 Mar 2023 19:28:43 +0300
Subject: bnx2: remove deadcode in bnx2_init_cpus()

The load_cpu_fw function has no error return code
and always returns zero. Checking the value returned by
this function does not make sense.
Now checking the value of the return value is misleading when reading
the code. Path with error handling was deleted in 57579f7629a3
("bnx2: Use request_firmware()").
As a result, bnx2_init_cpus() will also return only zero
Therefore, it will be safe to change the type of functions
to void and remove checking to improving readability.

Found by Security Code and Linux Verification
Center (linuxtesting.org) with SVACE

Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Maxim Korotkov <korotkov.maxim.s@gmail.com>
Link: https://lore.kernel.org/r/20230322162843.3452-1-korotkov.maxim.s@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnx2.c | 31 ++++++++-----------------------
 1 file changed, 8 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index a66137b8d1a6..466e1d62bcf6 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -3828,7 +3828,7 @@ load_rv2p_fw(struct bnx2 *bp, u32 rv2p_proc,
 	return 0;
 }
 
-static int
+static void
 load_cpu_fw(struct bnx2 *bp, const struct cpu_reg *cpu_reg,
 	    const struct bnx2_mips_fw_file_entry *fw_entry)
 {
@@ -3896,48 +3896,34 @@ load_cpu_fw(struct bnx2 *bp, const struct cpu_reg *cpu_reg,
 	val &= ~cpu_reg->mode_value_halt;
 	bnx2_reg_wr_ind(bp, cpu_reg->state, cpu_reg->state_value_clear);
 	bnx2_reg_wr_ind(bp, cpu_reg->mode, val);
-
-	return 0;
 }
 
-static int
+static void
 bnx2_init_cpus(struct bnx2 *bp)
 {
 	const struct bnx2_mips_fw_file *mips_fw =
 		(const struct bnx2_mips_fw_file *) bp->mips_firmware->data;
 	const struct bnx2_rv2p_fw_file *rv2p_fw =
 		(const struct bnx2_rv2p_fw_file *) bp->rv2p_firmware->data;
-	int rc;
 
 	/* Initialize the RV2P processor. */
 	load_rv2p_fw(bp, RV2P_PROC1, &rv2p_fw->proc1);
 	load_rv2p_fw(bp, RV2P_PROC2, &rv2p_fw->proc2);
 
 	/* Initialize the RX Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_rxp, &mips_fw->rxp);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_rxp, &mips_fw->rxp);
 
 	/* Initialize the TX Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_txp, &mips_fw->txp);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_txp, &mips_fw->txp);
 
 	/* Initialize the TX Patch-up Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_tpat, &mips_fw->tpat);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_tpat, &mips_fw->tpat);
 
 	/* Initialize the Completion Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_com, &mips_fw->com);
-	if (rc)
-		goto init_cpu_err;
+	load_cpu_fw(bp, &cpu_reg_com, &mips_fw->com);
 
 	/* Initialize the Command Processor. */
-	rc = load_cpu_fw(bp, &cpu_reg_cp, &mips_fw->cp);
-
-init_cpu_err:
-	return rc;
+	load_cpu_fw(bp, &cpu_reg_cp, &mips_fw->cp);
 }
 
 static void
@@ -4950,8 +4936,7 @@ bnx2_init_chip(struct bnx2 *bp)
 	} else
 		bnx2_init_context(bp);
 
-	if ((rc = bnx2_init_cpus(bp)) != 0)
-		return rc;
+	bnx2_init_cpus(bp);
 
 	bnx2_init_nvram(bp);
 
-- 
cgit v1.2.3-70-g09d2


From e110ba65927151bf56d8012ddf39ce10f91cbdf9 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 22 Mar 2023 16:12:02 -0700
Subject: docs: netdev: add note about Changes Requested and revising commit
 messages

One of the most commonly asked questions is "I answered all questions
and don't need to make any code changes, why was the patch not applied".
Document our time honored tradition of asking people to repost with
improved commit messages, to record the answers to reviewer questions.

Take this opportunity to also recommend a change log format.

Link: https://lore.kernel.org/r/20230322231202.265835-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/process/maintainer-netdev.rst | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/Documentation/process/maintainer-netdev.rst b/Documentation/process/maintainer-netdev.rst
index 4a75686d35ab..e31d7a951073 100644
--- a/Documentation/process/maintainer-netdev.rst
+++ b/Documentation/process/maintainer-netdev.rst
@@ -109,6 +109,8 @@ Finally, the vX.Y gets released, and the whole cycle starts over.
 netdev patch review
 -------------------
 
+.. _patch_status:
+
 Patch status
 ~~~~~~~~~~~~
 
@@ -143,6 +145,33 @@ Asking the maintainer for status updates on your
 patch is a good way to ensure your patch is ignored or pushed to the
 bottom of the priority list.
 
+Changes requested
+~~~~~~~~~~~~~~~~~
+
+Patches :ref:`marked<patch_status>` as ``Changes Requested`` need
+to be revised. The new version should come with a change log,
+preferably including links to previous postings, for example::
+
+  [PATCH net-next v3] net: make cows go moo
+
+  Even users who don't drink milk appreciate hearing the cows go "moo".
+
+  The amount of mooing will depend on packet rate so should match
+  the diurnal cycle quite well.
+
+  Signed-of-by: Joe Defarmer <joe@barn.org>
+  ---
+  v3:
+    - add a note about time-of-day mooing fluctuation to the commit message
+  v2: https://lore.kernel.org/netdev/123themessageid@barn.org/
+    - fix missing argument in kernel doc for netif_is_bovine()
+    - fix memory leak in netdev_register_cow()
+  v1: https://lore.kernel.org/netdev/456getstheclicks@barn.org/
+
+The commit message should be revised to answer any questions reviewers
+had to ask in previous discussions. Occasionally the update of
+the commit message will be the only change in the new version.
+
 Partial resends
 ~~~~~~~~~~~~~~~
 
-- 
cgit v1.2.3-70-g09d2


From 2df9bf4d04d2eb458386a40bac01be22c8808450 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 22 Mar 2023 19:35:41 -0400
Subject: ipv6: prevent router_solicitations for team port

The issue fixed for bonding in commit c2edacf80e15 ("bonding / ipv6: no
addrconf for slaves separately from master") also exists in team driver.
However, we can't just disable ipv6 addrconf for team ports, as 'teamd'
will need it when nsns_ping watch is used in the user space.

Instead of preventing ipv6 addrconf, this patch only prevents RS packets
for team ports, as it did in commit b52e1cce31ca ("ipv6: Don't send rs
packets to the interface of ARPHRD_TUNNEL").

Note that we do not prevent DAD packets, to avoid the changes getting
intricate / hacky. Also, usually sysctl dad_transmits is set to 1 and
only 1 DAD packet will be sent, and by now no libteam user complains
about DAD packets on team ports, unlike RS packets.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/addrconf.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 31e0097878c5..3797917237d0 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -4223,7 +4223,8 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp, bool bump_id,
 		  ipv6_accept_ra(ifp->idev) &&
 		  ifp->idev->cnf.rtr_solicits != 0 &&
 		  (dev->flags & IFF_LOOPBACK) == 0 &&
-		  (dev->type != ARPHRD_TUNNEL);
+		  (dev->type != ARPHRD_TUNNEL) &&
+		  !netif_is_team_port(dev);
 	read_unlock_bh(&ifp->idev->lock);
 
 	/* While dad is in progress mld report's source address is in6_addrany.
-- 
cgit v1.2.3-70-g09d2


From c34ce2796228c3b00353c173a462debe62ab6265 Mon Sep 17 00:00:00 2001
From: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Date: Thu, 23 Mar 2023 06:21:01 +0100
Subject: ethernet: broadcom/sb1250-mac: clean up after SIBYTE_BCM1x55 removal

With commit b984d7b56dfc ("MIPS: sibyte: Remove unused config option
SIBYTE_BCM1x55"), some #if's in the Broadcom SiByte SOC built-in Ethernet
driver can be simplified.

Simplify prepreprocessor conditions after config SIBYTE_BCM1x55 removal.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/sb1250-mac.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c
index f02facb60fd1..3a6763c5e8b3 100644
--- a/drivers/net/ethernet/broadcom/sb1250-mac.c
+++ b/drivers/net/ethernet/broadcom/sb1250-mac.c
@@ -73,7 +73,7 @@ MODULE_PARM_DESC(int_timeout_rx, "RX timeout value");
 
 #include <asm/sibyte/board.h>
 #include <asm/sibyte/sb1250.h>
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+#if defined(CONFIG_SIBYTE_BCM1x80)
 #include <asm/sibyte/bcm1480_regs.h>
 #include <asm/sibyte/bcm1480_int.h>
 #define R_MAC_DMA_OODPKTLOST_RX	R_MAC_DMA_OODPKTLOST
@@ -87,7 +87,7 @@ MODULE_PARM_DESC(int_timeout_rx, "RX timeout value");
 #include <asm/sibyte/sb1250_mac.h>
 #include <asm/sibyte/sb1250_dma.h>
 
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+#if defined(CONFIG_SIBYTE_BCM1x80)
 #define UNIT_INT(n)		(K_BCM1480_INT_MAC_0 + ((n) * 2))
 #elif defined(CONFIG_SIBYTE_SB1250) || defined(CONFIG_SIBYTE_BCM112X)
 #define UNIT_INT(n)		(K_INT_MAC_0 + (n))
@@ -1527,7 +1527,7 @@ static void sbmac_channel_start(struct sbmac_softc *s)
 	 * Turn on the rest of the bits in the enable register
 	 */
 
-#if defined(CONFIG_SIBYTE_BCM1x55) || defined(CONFIG_SIBYTE_BCM1x80)
+#if defined(CONFIG_SIBYTE_BCM1x80)
 	__raw_writeq(M_MAC_RXDMA_EN0 |
 		       M_MAC_TXDMA_EN0, s->sbm_macenable);
 #elif defined(CONFIG_SIBYTE_SB1250) || defined(CONFIG_SIBYTE_BCM112X)
-- 
cgit v1.2.3-70-g09d2


From 4170f0ef582c1ff4557276b98fb5fad85c782c1f Mon Sep 17 00:00:00 2001
From: Taichi Nishimura <awkrail01@gmail.com>
Date: Thu, 23 Mar 2023 14:27:13 +0900
Subject: fix typos in net/sched/* files

This patch fixes typos in net/sched/* files.

Signed-off-by: Taichi Nishimura <awkrail01@gmail.com>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/cls_flower.c | 2 +-
 net/sched/em_meta.c    | 2 +-
 net/sched/sch_pie.c    | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 475fe222a855..cc49256d5318 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -1057,7 +1057,7 @@ static void fl_set_key_pppoe(struct nlattr **tb,
 	 * because ETH_P_PPP_SES was stored in basic.n_proto
 	 * which might get overwritten by ppp_proto
 	 * or might be set to 0, the role of key_val::type
-	 * is simmilar to vlan_key::tpid
+	 * is similar to vlan_key::tpid
 	 */
 	key_val->type = htons(ETH_P_PPP_SES);
 	key_mask->type = cpu_to_be16(~0);
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index 49bae3d5006b..af85a73c4c54 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -44,7 +44,7 @@
  * 	be provided for non-numeric types.
  *
  * 	Additionally, type dependent modifiers such as shift operators
- * 	or mask may be applied to extend the functionaliy. As of now,
+ * 	or mask may be applied to extend the functionality. As of now,
  * 	the variable length type supports shifting the byte string to
  * 	the right, eating up any number of octets and thus supporting
  * 	wildcard interface name comparisons such as "ppp%" matching
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index 265c238047a4..2152a56d73f8 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -319,7 +319,7 @@ void pie_calculate_probability(struct pie_params *params, struct pie_vars *vars,
 	}
 
 	/* If qdelay is zero and backlog is not, it means backlog is very small,
-	 * so we do not update probabilty in this round.
+	 * so we do not update probability in this round.
 	 */
 	if (qdelay == 0 && backlog != 0)
 		update_prob = false;
-- 
cgit v1.2.3-70-g09d2


From fc5aba60c24407f4a23f5e538585516068aa8746 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:00 +0800
Subject: net: liquidio: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/cavium/liquidio/lio_main.c    | 1 -
 drivers/net/ethernet/cavium/liquidio/lio_vf_main.c | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index fd7c80edb6e8..9bd1d2d7027d 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -1129,7 +1129,6 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 
 		fallthrough;
 	case OCT_DEV_PCI_ENABLE_DONE:
-		pci_clear_master(oct->pci_dev);
 		/* Disable the device, releasing the PCI INT */
 		pci_disable_device(oct->pci_dev);
 
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
index ac196883f07e..e2921aec3da0 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
@@ -577,7 +577,6 @@ static void octeon_destroy_resources(struct octeon_device *oct)
 
 		fallthrough;
 	case OCT_DEV_PCI_ENABLE_DONE:
-		pci_clear_master(oct->pci_dev);
 		/* Disable the device, releasing the PCI INT */
 		pci_disable_device(oct->pci_dev);
 
-- 
cgit v1.2.3-70-g09d2


From fc3e07e83e8ec31ebee5ed53940fb44f9b7c7cee Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:01 +0800
Subject: net: hisilicon: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c   | 7 ++-----
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 6 ++----
 2 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 07ad5f35219e..c3851a6e10c0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -11365,7 +11365,7 @@ static int hclge_pci_init(struct hclge_dev *hdev)
 	if (!hw->hw.io_base) {
 		dev_err(&pdev->dev, "Can't map configuration register space\n");
 		ret = -ENOMEM;
-		goto err_clr_master;
+		goto err_release_regions;
 	}
 
 	ret = hclge_dev_mem_map(hdev);
@@ -11378,8 +11378,7 @@ static int hclge_pci_init(struct hclge_dev *hdev)
 
 err_unmap_io_base:
 	pcim_iounmap(pdev, hdev->hw.hw.io_base);
-err_clr_master:
-	pci_clear_master(pdev);
+err_release_regions:
 	pci_release_regions(pdev);
 err_disable_device:
 	pci_disable_device(pdev);
@@ -11396,7 +11395,6 @@ static void hclge_pci_uninit(struct hclge_dev *hdev)
 
 	pcim_iounmap(pdev, hdev->hw.hw.io_base);
 	pci_free_irq_vectors(pdev);
-	pci_clear_master(pdev);
 	pci_release_mem_regions(pdev);
 	pci_disable_device(pdev);
 }
@@ -11743,7 +11741,6 @@ err_devlink_uninit:
 	hclge_devlink_uninit(hdev);
 err_pci_uninit:
 	pcim_iounmap(pdev, hdev->hw.hw.io_base);
-	pci_clear_master(pdev);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 out:
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index e84e5be8e59e..f24046250341 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -2598,7 +2598,7 @@ static int hclgevf_pci_init(struct hclgevf_dev *hdev)
 	if (!hw->hw.io_base) {
 		dev_err(&pdev->dev, "can't map configuration register space\n");
 		ret = -ENOMEM;
-		goto err_clr_master;
+		goto err_release_regions;
 	}
 
 	ret = hclgevf_dev_mem_map(hdev);
@@ -2609,8 +2609,7 @@ static int hclgevf_pci_init(struct hclgevf_dev *hdev)
 
 err_unmap_io_base:
 	pci_iounmap(pdev, hdev->hw.hw.io_base);
-err_clr_master:
-	pci_clear_master(pdev);
+err_release_regions:
 	pci_release_regions(pdev);
 err_disable_device:
 	pci_disable_device(pdev);
@@ -2626,7 +2625,6 @@ static void hclgevf_pci_uninit(struct hclgevf_dev *hdev)
 		devm_iounmap(&pdev->dev, hdev->hw.hw.mem_base);
 
 	pci_iounmap(pdev, hdev->hw.hw.io_base);
-	pci_clear_master(pdev);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 }
-- 
cgit v1.2.3-70-g09d2


From aae964bb7800d8e68a75c49a6508968579d253ae Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:02 +0800
Subject: net: cxgb4vf: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 63b2bd084130..9ba0864592e8 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -3258,7 +3258,6 @@ err_free_adapter:
 
 err_release_regions:
 	pci_release_regions(pdev);
-	pci_clear_master(pdev);
 
 err_disable_device:
 	pci_disable_device(pdev);
@@ -3338,7 +3337,6 @@ static void cxgb4vf_pci_remove(struct pci_dev *pdev)
 	 * Disable the device and release its PCI resources.
 	 */
 	pci_disable_device(pdev);
-	pci_clear_master(pdev);
 	pci_release_regions(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 8b91d5b62ce83b6678cfc876427dbd386798a6ef Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:03 +0800
Subject: net/fungible: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Acked-by: Dimitris Michailidis <dmichail@fungible.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/fungible/funcore/fun_dev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/fungible/funcore/fun_dev.c b/drivers/net/ethernet/fungible/funcore/fun_dev.c
index 3680f83feba2..a7fbd4cd560a 100644
--- a/drivers/net/ethernet/fungible/funcore/fun_dev.c
+++ b/drivers/net/ethernet/fungible/funcore/fun_dev.c
@@ -746,7 +746,6 @@ void fun_dev_disable(struct fun_dev *fdev)
 	bitmap_free(fdev->irq_map);
 	pci_free_irq_vectors(pdev);
 
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);
 
 	fun_unmap_bars(fdev);
@@ -821,7 +820,6 @@ int fun_dev_enable(struct fun_dev *fdev, struct pci_dev *pdev,
 disable_admin:
 	fun_disable_admin_queue(fdev);
 free_irq_mgr:
-	pci_clear_master(pdev);
 	bitmap_free(fdev->irq_map);
 free_irqs:
 	pci_free_irq_vectors(pdev);
-- 
cgit v1.2.3-70-g09d2


From 5b6f4bd24c8d45d2c67853e812453cb957b92f1a Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:04 +0800
Subject: net/mlx5: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index d39c3476b6d1..597174ceadc9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -918,7 +918,6 @@ static int mlx5_pci_init(struct mlx5_core_dev *dev, struct pci_dev *pdev,
 	return 0;
 
 err_clr_master:
-	pci_clear_master(dev->pdev);
 	release_bar(dev->pdev);
 err_disable:
 	mlx5_pci_disable_device(dev);
@@ -933,7 +932,6 @@ static void mlx5_pci_close(struct mlx5_core_dev *dev)
 	 */
 	mlx5_drain_health_wq(dev);
 	iounmap(dev->iseg);
-	pci_clear_master(dev->pdev);
 	release_bar(dev->pdev);
 	mlx5_pci_disable_device(dev);
 }
-- 
cgit v1.2.3-70-g09d2


From 2d59af8307526f2829fdec9c5c5898a857d55180 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:05 +0800
Subject: net: mana: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microsoft/mana/gdma_main.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c
index f9b8f372ec8a..8f3f78b68592 100644
--- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
+++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
@@ -1439,7 +1439,6 @@ free_gc:
 release_region:
 	pci_release_regions(pdev);
 disable_dev:
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);
 	dev_err(&pdev->dev, "gdma probe failed: err = %d\n", err);
 	return err;
@@ -1458,7 +1457,6 @@ static void mana_gd_remove(struct pci_dev *pdev)
 	vfree(gc);
 
 	pci_release_regions(pdev);
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From f686e9592734241eeddb406ef5f7aeab0a68c8b5 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:06 +0800
Subject: ionic: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Acked-by: Shannon Nelson <shannon.nelson@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
index e508f8eb43bf..b8678da1cce5 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
@@ -392,7 +392,6 @@ static void ionic_remove(struct pci_dev *pdev)
 	ionic_port_reset(ionic);
 	ionic_reset(ionic);
 	ionic_dev_teardown(ionic);
-	pci_clear_master(pdev);
 	ionic_unmap_bars(ionic);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-70-g09d2


From 3228150ba688b8cddc7459e5ae9a09bc2d2e47be Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 17:03:07 +0800
Subject: ethernet: ec_bhf: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ec_bhf.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/ec_bhf.c b/drivers/net/ethernet/ec_bhf.c
index 46e3a05e9582..c2c5c589a5e3 100644
--- a/drivers/net/ethernet/ec_bhf.c
+++ b/drivers/net/ethernet/ec_bhf.c
@@ -558,7 +558,6 @@ err_unmap:
 err_release_regions:
 	pci_release_regions(dev);
 err_disable_dev:
-	pci_clear_master(dev);
 	pci_disable_device(dev);
 
 	return err;
@@ -577,7 +576,6 @@ static void ec_bhf_remove(struct pci_dev *dev)
 	free_netdev(net_dev);
 
 	pci_release_regions(dev);
-	pci_clear_master(dev);
 	pci_disable_device(dev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From fac63186f11682fc5ffd0472d1a990256bd61348 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Thu, 23 Mar 2023 11:55:57 +0100
Subject: net: phy: micrel: Add support for PTP_PF_EXTTS for lan8841

Extend the PTP programmable gpios to implement also PTP_PF_EXTTS
function. The pins can be configured to capture both of rising
and falling edge. Once the event is seen, then an interrupt is
generated and the LTC is saved in the registers.

This was tested using:
ts2phc -m -l 7 -s generic -f ts2phc.cfg

Where the configuration was the following:
[global]
ts2phc.pin_index  6

[eth2]

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/micrel.c | 162 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 162 insertions(+)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index e26c6723caa4..cb87d38b4d9a 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -3437,6 +3437,7 @@ static void lan8841_ptp_process_rx_ts(struct kszphy_ptp_priv *ptp_priv)
 #define LAN8841_PTP_INT_STS_PTP_TX_TS_INT	BIT(12)
 #define LAN8841_PTP_INT_STS_PTP_RX_TS_OVRFL_INT	BIT(9)
 #define LAN8841_PTP_INT_STS_PTP_RX_TS_INT	BIT(8)
+#define LAN8841_PTP_INT_STS_PTP_GPIO_CAP_INT	BIT(2)
 
 static void lan8841_ptp_flush_fifo(struct kszphy_ptp_priv *ptp_priv, bool egress)
 {
@@ -3451,6 +3452,67 @@ static void lan8841_ptp_flush_fifo(struct kszphy_ptp_priv *ptp_priv, bool egress
 	phy_read_mmd(phydev, 2, LAN8841_PTP_INT_STS);
 }
 
+#define LAN8841_PTP_GPIO_CAP_STS			506
+#define LAN8841_PTP_GPIO_SEL				327
+#define LAN8841_PTP_GPIO_SEL_GPIO_SEL(gpio)		((gpio) << 8)
+#define LAN8841_PTP_GPIO_RE_LTC_SEC_HI_CAP		498
+#define LAN8841_PTP_GPIO_RE_LTC_SEC_LO_CAP		499
+#define LAN8841_PTP_GPIO_RE_LTC_NS_HI_CAP		500
+#define LAN8841_PTP_GPIO_RE_LTC_NS_LO_CAP		501
+#define LAN8841_PTP_GPIO_FE_LTC_SEC_HI_CAP		502
+#define LAN8841_PTP_GPIO_FE_LTC_SEC_LO_CAP		503
+#define LAN8841_PTP_GPIO_FE_LTC_NS_HI_CAP		504
+#define LAN8841_PTP_GPIO_FE_LTC_NS_LO_CAP		505
+
+static void lan8841_gpio_process_cap(struct kszphy_ptp_priv *ptp_priv)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	struct ptp_clock_event ptp_event = {0};
+	int pin, ret, tmp;
+	s32 sec, nsec;
+
+	pin = ptp_find_pin_unlocked(ptp_priv->ptp_clock, PTP_PF_EXTTS, 0);
+	if (pin == -1)
+		return;
+
+	tmp = phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_CAP_STS);
+	if (tmp < 0)
+		return;
+
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_GPIO_SEL,
+			    LAN8841_PTP_GPIO_SEL_GPIO_SEL(pin));
+	if (ret)
+		return;
+
+	mutex_lock(&ptp_priv->ptp_lock);
+	if (tmp & BIT(pin)) {
+		sec = phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_RE_LTC_SEC_HI_CAP);
+		sec <<= 16;
+		sec |= phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_RE_LTC_SEC_LO_CAP);
+
+		nsec = phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_RE_LTC_NS_HI_CAP) & 0x3fff;
+		nsec <<= 16;
+		nsec |= phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_RE_LTC_NS_LO_CAP);
+	} else {
+		sec = phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_FE_LTC_SEC_HI_CAP);
+		sec <<= 16;
+		sec |= phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_FE_LTC_SEC_LO_CAP);
+
+		nsec = phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_FE_LTC_NS_HI_CAP) & 0x3fff;
+		nsec <<= 16;
+		nsec |= phy_read_mmd(phydev, 2, LAN8841_PTP_GPIO_FE_LTC_NS_LO_CAP);
+	}
+	mutex_unlock(&ptp_priv->ptp_lock);
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_GPIO_SEL, 0);
+	if (ret)
+		return;
+
+	ptp_event.index = 0;
+	ptp_event.timestamp = ktime_set(sec, nsec);
+	ptp_event.type = PTP_CLOCK_EXTTS;
+	ptp_clock_event(ptp_priv->ptp_clock, &ptp_event);
+}
+
 static void lan8841_handle_ptp_interrupt(struct phy_device *phydev)
 {
 	struct kszphy_priv *priv = phydev->priv;
@@ -3459,12 +3521,16 @@ static void lan8841_handle_ptp_interrupt(struct phy_device *phydev)
 
 	do {
 		status = phy_read_mmd(phydev, 2, LAN8841_PTP_INT_STS);
+
 		if (status & LAN8841_PTP_INT_STS_PTP_TX_TS_INT)
 			lan8841_ptp_process_tx_ts(ptp_priv);
 
 		if (status & LAN8841_PTP_INT_STS_PTP_RX_TS_INT)
 			lan8841_ptp_process_rx_ts(ptp_priv);
 
+		if (status & LAN8841_PTP_INT_STS_PTP_GPIO_CAP_INT)
+			lan8841_gpio_process_cap(ptp_priv);
+
 		if (status & LAN8841_PTP_INT_STS_PTP_TX_TS_OVRFL_INT) {
 			lan8841_ptp_flush_fifo(ptp_priv, true);
 			skb_queue_purge(&ptp_priv->tx_queue);
@@ -3924,6 +3990,7 @@ static int lan8841_ptp_verify(struct ptp_clock_info *ptp, unsigned int pin,
 	switch (func) {
 	case PTP_PF_NONE:
 	case PTP_PF_PEROUT:
+	case PTP_PF_EXTTS:
 		break;
 	default:
 		return -1;
@@ -4191,10 +4258,104 @@ static int lan8841_ptp_perout(struct ptp_clock_info *ptp,
 	return ret;
 }
 
+#define LAN8841_PTP_GPIO_CAP_EN			496
+#define LAN8841_PTP_GPIO_CAP_EN_GPIO_RE_CAPTURE_ENABLE(gpio)	(BIT(gpio))
+#define LAN8841_PTP_GPIO_CAP_EN_GPIO_FE_CAPTURE_ENABLE(gpio)	(BIT(gpio) << 8)
+#define LAN8841_PTP_INT_EN_PTP_GPIO_CAP_EN	BIT(2)
+
+static int lan8841_ptp_extts_on(struct kszphy_ptp_priv *ptp_priv, int pin,
+				u32 flags)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	u16 tmp = 0;
+	int ret;
+
+	/* Set GPIO to be intput */
+	ret = phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_EN, BIT(pin));
+	if (ret)
+		return ret;
+
+	ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_BUF, BIT(pin));
+	if (ret)
+		return ret;
+
+	/* Enable capture on the edges of the pin */
+	if (flags & PTP_RISING_EDGE)
+		tmp |= LAN8841_PTP_GPIO_CAP_EN_GPIO_RE_CAPTURE_ENABLE(pin);
+	if (flags & PTP_FALLING_EDGE)
+		tmp |= LAN8841_PTP_GPIO_CAP_EN_GPIO_FE_CAPTURE_ENABLE(pin);
+	ret = phy_write_mmd(phydev, 2, LAN8841_PTP_GPIO_CAP_EN, tmp);
+	if (ret)
+		return ret;
+
+	/* Enable interrupt */
+	return phy_modify_mmd(phydev, 2, LAN8841_PTP_INT_EN,
+			      LAN8841_PTP_INT_EN_PTP_GPIO_CAP_EN,
+			      LAN8841_PTP_INT_EN_PTP_GPIO_CAP_EN);
+}
+
+static int lan8841_ptp_extts_off(struct kszphy_ptp_priv *ptp_priv, int pin)
+{
+	struct phy_device *phydev = ptp_priv->phydev;
+	int ret;
+
+	/* Set GPIO to be output */
+	ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_EN, BIT(pin));
+	if (ret)
+		return ret;
+
+	ret = phy_clear_bits_mmd(phydev, 2, LAN8841_GPIO_BUF, BIT(pin));
+	if (ret)
+		return ret;
+
+	/* Disable capture on both of the edges */
+	ret = phy_modify_mmd(phydev, 2, LAN8841_PTP_GPIO_CAP_EN,
+			     LAN8841_PTP_GPIO_CAP_EN_GPIO_RE_CAPTURE_ENABLE(pin) |
+			     LAN8841_PTP_GPIO_CAP_EN_GPIO_FE_CAPTURE_ENABLE(pin),
+			     0);
+	if (ret)
+		return ret;
+
+	/* Disable interrupt */
+	return phy_modify_mmd(phydev, 2, LAN8841_PTP_INT_EN,
+			      LAN8841_PTP_INT_EN_PTP_GPIO_CAP_EN,
+			      0);
+}
+
+static int lan8841_ptp_extts(struct ptp_clock_info *ptp,
+			     struct ptp_clock_request *rq, int on)
+{
+	struct kszphy_ptp_priv *ptp_priv = container_of(ptp, struct kszphy_ptp_priv,
+							ptp_clock_info);
+	int pin;
+	int ret;
+
+	/* Reject requests with unsupported flags */
+	if (rq->extts.flags & ~(PTP_ENABLE_FEATURE |
+				PTP_EXTTS_EDGES |
+				PTP_STRICT_FLAGS))
+		return -EOPNOTSUPP;
+
+	pin = ptp_find_pin(ptp_priv->ptp_clock, PTP_PF_EXTTS, rq->extts.index);
+	if (pin == -1 || pin >= LAN8841_PTP_GPIO_NUM)
+		return -EINVAL;
+
+	mutex_lock(&ptp_priv->ptp_lock);
+	if (on)
+		ret = lan8841_ptp_extts_on(ptp_priv, pin, rq->extts.flags);
+	else
+		ret = lan8841_ptp_extts_off(ptp_priv, pin);
+	mutex_unlock(&ptp_priv->ptp_lock);
+
+	return ret;
+}
+
 static int lan8841_ptp_enable(struct ptp_clock_info *ptp,
 			      struct ptp_clock_request *rq, int on)
 {
 	switch (rq->type) {
+	case PTP_CLK_REQ_EXTTS:
+		return lan8841_ptp_extts(ptp, rq, on);
 	case PTP_CLK_REQ_PEROUT:
 		return lan8841_ptp_perout(ptp, rq, on);
 	default:
@@ -4215,6 +4376,7 @@ static struct ptp_clock_info lan8841_ptp_clock_info = {
 	.verify         = lan8841_ptp_verify,
 	.enable         = lan8841_ptp_enable,
 	.n_per_out      = LAN8841_PTP_GPIO_NUM,
+	.n_ext_ts       = LAN8841_PTP_GPIO_NUM,
 	.n_pins         = LAN8841_PTP_GPIO_NUM,
 };
 
-- 
cgit v1.2.3-70-g09d2


From 950bfdcf17ec043072c87ed6e671605302215195 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:59:11 +0800
Subject: isdn: mISDN: netjet: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/isdn/hardware/mISDN/netjet.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/isdn/hardware/mISDN/netjet.c b/drivers/isdn/hardware/mISDN/netjet.c
index f8447135a902..566c790a9481 100644
--- a/drivers/isdn/hardware/mISDN/netjet.c
+++ b/drivers/isdn/hardware/mISDN/netjet.c
@@ -970,7 +970,6 @@ nj_release(struct tiger_hw *card)
 	write_lock_irqsave(&card_lock, flags);
 	list_del(&card->list);
 	write_unlock_irqrestore(&card_lock, flags);
-	pci_clear_master(card->pdev);
 	pci_disable_device(card->pdev);
 	pci_set_drvdata(card->pdev, NULL);
 	kfree(card);
-- 
cgit v1.2.3-70-g09d2


From c85bd3dacc80b99ae68c7082f8e2edd8b9f83c39 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 20:00:43 +0800
Subject: net/ism: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/ism_drv.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/s390/net/ism_drv.c b/drivers/s390/net/ism_drv.c
index 05749c877990..8acb9eba691b 100644
--- a/drivers/s390/net/ism_drv.c
+++ b/drivers/s390/net/ism_drv.c
@@ -675,7 +675,6 @@ static int ism_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	return 0;
 
 err_resource:
-	pci_clear_master(pdev);
 	pci_release_mem_regions(pdev);
 err_disable:
 	pci_disable_device(pdev);
@@ -738,7 +737,6 @@ static void ism_remove(struct pci_dev *pdev)
 	ism_dev_exit(ism);
 	mutex_unlock(&ism_dev_list.mutex);
 
-	pci_clear_master(pdev);
 	pci_release_mem_regions(pdev);
 	pci_disable_device(pdev);
 	device_del(&ism->dev);
-- 
cgit v1.2.3-70-g09d2


From 323fe43cf9aef79159ba8937218a3f076bf505af Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Thu, 23 Mar 2023 14:45:59 -0700
Subject: net: phy: Improved PHY error reporting in state machine

When the PHY library calls phy_error() something bad has happened, and
we halt the PHY state machine. Calling phy_error() from the main state
machine however is not precise enough to know whether the issue is
reading the link status or starting auto-negotiation.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy.c | 33 ++++++++++++++++++++++++---------
 1 file changed, 24 insertions(+), 9 deletions(-)

diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index b33e55a7364e..bcaf58636199 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -1169,6 +1169,22 @@ void phy_stop_machine(struct phy_device *phydev)
 	mutex_unlock(&phydev->lock);
 }
 
+static void phy_process_error(struct phy_device *phydev)
+{
+	mutex_lock(&phydev->lock);
+	phydev->state = PHY_HALTED;
+	mutex_unlock(&phydev->lock);
+
+	phy_trigger_machine(phydev);
+}
+
+static void phy_error_precise(struct phy_device *phydev,
+			      const void *func, int err)
+{
+	WARN(1, "%pS: returned: %d\n", func, err);
+	phy_process_error(phydev);
+}
+
 /**
  * phy_error - enter HALTED state for this PHY device
  * @phydev: target phy_device struct
@@ -1181,12 +1197,7 @@ void phy_stop_machine(struct phy_device *phydev)
 void phy_error(struct phy_device *phydev)
 {
 	WARN_ON(1);
-
-	mutex_lock(&phydev->lock);
-	phydev->state = PHY_HALTED;
-	mutex_unlock(&phydev->lock);
-
-	phy_trigger_machine(phydev);
+	phy_process_error(phydev);
 }
 EXPORT_SYMBOL(phy_error);
 
@@ -1378,6 +1389,7 @@ void phy_state_machine(struct work_struct *work)
 	struct net_device *dev = phydev->attached_dev;
 	bool needs_aneg = false, do_suspend = false;
 	enum phy_state old_state;
+	const void *func = NULL;
 	bool finished = false;
 	int err = 0;
 
@@ -1396,6 +1408,7 @@ void phy_state_machine(struct work_struct *work)
 	case PHY_NOLINK:
 	case PHY_RUNNING:
 		err = phy_check_link_status(phydev);
+		func = &phy_check_link_status;
 		break;
 	case PHY_CABLETEST:
 		err = phydev->drv->cable_test_get_status(phydev, &finished);
@@ -1425,16 +1438,18 @@ void phy_state_machine(struct work_struct *work)
 
 	mutex_unlock(&phydev->lock);
 
-	if (needs_aneg)
+	if (needs_aneg) {
 		err = phy_start_aneg(phydev);
-	else if (do_suspend)
+		func = &phy_start_aneg;
+	} else if (do_suspend) {
 		phy_suspend(phydev);
+	}
 
 	if (err == -ENODEV)
 		return;
 
 	if (err < 0)
-		phy_error(phydev);
+		phy_error_precise(phydev, func, err);
 
 	if (old_state != phydev->state) {
 		phydev_dbg(phydev, "PHY state change %s -> %s\n",
-- 
cgit v1.2.3-70-g09d2


From d24b2130146b31d2185296dbc08186fb67bb2c71 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Fri, 24 Mar 2023 00:35:43 +0200
Subject: wifi: iwlwifi: mvm: fix NULL deref in iwl_mvm_mld_disable_txq

Check sta pointer for NULL and don't crash if it is.

Fixes: 006c152ac9e5 ("wifi: iwlwifi: mvm: add support for the new STA related commands")
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/linux-wireless/20230314194113.132873ce015c.I7b12a77e5be066730762e6ceeeaa7190293c3df1@changeid/
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index d81fc9207ef0..51532b6379e2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -227,10 +227,15 @@ static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm,
 				   struct ieee80211_sta *sta,
 				   u16 *queueptr, u8 tid)
 {
-	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_sta *mvmsta;
 	int queue = *queueptr;
 	int ret = 0;
 
+	if (!sta)
+		return -EINVAL;
+
+	mvmsta = iwl_mvm_sta_from_mac80211(sta);
+
 	if (mvm->sta_remove_requires_queue_remove) {
 		u32 cmd_id = WIDE_ID(DATA_PATH_GROUP,
 				     SCD_QUEUE_CONFIG_CMD);
-- 
cgit v1.2.3-70-g09d2


From f102424befd3751386f3e2f2c70c5a1948248622 Mon Sep 17 00:00:00 2001
From: Bagas Sanjaya <bagasdotme@gmail.com>
Date: Thu, 23 Mar 2023 16:24:54 +0700
Subject: wifi: mac80211: use bullet list for amsdu_mesh_control formats list

Commit fe4a6d2db3ba ("wifi: mac80211: implement support for yet
another mesh A-MSDU format") expands amsdu_mesh_control list to
multi-line list. However, the expansion triggers Sphinx warning:

Documentation/driver-api/80211/mac80211-advanced:214: ./net/mac80211/sta_info.h:628: WARNING: Unexpected indentation.

Use bullet list instead to fix the warning.

Link: https://lore.kernel.org/linux-next/20230323141548.659479ef@canb.auug.org.au/
Fixes: fe4a6d2db3bad4 ("wifi: mac80211: implement support for yet another mesh A-MSDU format")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/sta_info.h | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index f354d470e174..195b563132d6 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -622,11 +622,13 @@ struct link_sta_info {
  *	taken from HT/VHT capabilities or VHT operating mode notification
  * @cparams: CoDel parameters for this station.
  * @reserved_tid: reserved TID (if any, otherwise IEEE80211_TID_UNRESERVED)
- * @amsdu_mesh_control: track the mesh A-MSDU format used by the peer
- *	(-1: not yet known,
- *	  0: non-mesh A-MSDU length field
- *	  1: big-endian mesh A-MSDU length field
- *	  2: little-endian mesh A-MSDU length field)
+ * @amsdu_mesh_control: track the mesh A-MSDU format used by the peer:
+ *
+ *	  * -1: not yet known
+ *	  * 0: non-mesh A-MSDU length field
+ *	  * 1: big-endian mesh A-MSDU length field
+ *	  * 2: little-endian mesh A-MSDU length field
+ *
  * @fast_tx: TX fastpath information
  * @fast_rx: RX fastpath information
  * @tdls_chandef: a TDLS peer can have a wider chandef that is compatible to
-- 
cgit v1.2.3-70-g09d2


From dbbb27e183b1568d5a907ace1cd144b0709ea52a Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Thu, 23 Mar 2023 04:38:00 -0700
Subject: cfg80211: support RNR for EMA AP

As per IEEE Std 802.11ax-2021, 11.1.3.8.3 Discovery of a nontransmitted
BSSID profile, an EMA AP that transmits a Beacon frame carrying a partial
list of nontransmitted BSSID profiles should include in the frame
a Reduced Neighbor Report element carrying information for at least the
nontransmitted BSSIDs that are not present in the Multiple BSSID element
carried in that frame.
Add new nested attribute NL80211_ATTR_EMA_RNR_ELEMS to support the above.
Number of RNR elements must be more than or equal to the number of
MBSSID elements. This attribute can be used only when EMA is enabled.
Userspace is responsible for splitting the RNR into multiple elements such
that each element excludes the non-transmitting profiles already included
in the MBSSID element (%NL80211_ATTR_MBSSID_ELEMS) at the same index.
Each EMA beacon will be generated by adding MBSSID and RNR elements
at the same index. If the userspace provides more RNR elements than the
number of MBSSID elements then these will be added in every EMA beacon.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Link: https://lore.kernel.org/r/20230323113801.6903-2-quic_alokad@quicinc.com
[Johannes: validate elements]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h       | 19 +++++++++++
 include/uapi/linux/nl80211.h | 13 ++++++++
 net/wireless/nl80211.c       | 79 ++++++++++++++++++++++++++++++++++++++++----
 3 files changed, 104 insertions(+), 7 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 86cb048dc924..3cf236520288 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -1178,6 +1178,23 @@ struct cfg80211_mbssid_elems {
 	} elem[];
 };
 
+/**
+ * struct cfg80211_rnr_elems - Reduced neighbor report (RNR) elements
+ *
+ * @cnt: Number of elements in array %elems.
+ *
+ * @elem: Array of RNR element(s) to be added into Beacon frames.
+ * @elem.data: Data for RNR elements.
+ * @elem.len: Length of data.
+ */
+struct cfg80211_rnr_elems {
+	u8 cnt;
+	struct {
+		const u8 *data;
+		size_t len;
+	} elem[];
+};
+
 /**
  * struct cfg80211_beacon_data - beacon data
  * @link_id: the link ID for the AP MLD link sending this beacon
@@ -1198,6 +1215,7 @@ struct cfg80211_mbssid_elems {
  * @probe_resp_len: length of probe response template (@probe_resp)
  * @probe_resp: probe response template (AP mode only)
  * @mbssid_ies: multiple BSSID elements
+ * @rnr_ies: reduced neighbor report elements
  * @ftm_responder: enable FTM responder functionality; -1 for no change
  *	(which also implies no change in LCI/civic location data)
  * @lci: Measurement Report element content, starting with Measurement Token
@@ -1221,6 +1239,7 @@ struct cfg80211_beacon_data {
 	const u8 *lci;
 	const u8 *civicloc;
 	struct cfg80211_mbssid_elems *mbssid_ies;
+	struct cfg80211_rnr_elems *rnr_ies;
 	s8 ftm_responder;
 
 	size_t head_len, tail_len;
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index 14e958a32b84..cf4fb981e131 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -2794,6 +2794,17 @@ enum nl80211_commands {
  * @NL80211_ATTR_HW_TIMESTAMP_ENABLED: Indicates whether HW timestamping should
  *	be enabled or not (flag attribute).
  *
+ * @NL80211_ATTR_EMA_RNR_ELEMS: Optional nested attribute for
+ *	reduced neighbor report (RNR) elements. This attribute can be used
+ *	only when NL80211_MBSSID_CONFIG_ATTR_EMA is enabled.
+ *	Userspace is responsible for splitting the RNR into multiple
+ *	elements such that each element excludes the non-transmitting
+ *	profiles already included in the MBSSID element
+ *	(%NL80211_ATTR_MBSSID_ELEMS) at the same index. Each EMA beacon
+ *	will be generated by adding MBSSID and RNR elements at the same
+ *	index. If the userspace includes more RNR elements than number of
+ *	MBSSID elements then these will be added in every EMA beacon.
+ *
  * @NUM_NL80211_ATTR: total number of nl80211_attrs available
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
@@ -3328,6 +3339,8 @@ enum nl80211_attrs {
 	NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS,
 	NL80211_ATTR_HW_TIMESTAMP_ENABLED,
 
+	NL80211_ATTR_EMA_RNR_ELEMS,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 0a31b1d2845d..80a20d69f285 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -809,6 +809,7 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = {
 
 	[NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS] = { .type = NLA_U16 },
 	[NL80211_ATTR_HW_TIMESTAMP_ENABLED] = { .type = NLA_FLAG },
+	[NL80211_ATTR_EMA_RNR_ELEMS] = { .type = NLA_NESTED },
 };
 
 /* policy for the key attributes */
@@ -5425,6 +5426,38 @@ nl80211_parse_mbssid_elems(struct wiphy *wiphy, struct nlattr *attrs)
 	return elems;
 }
 
+static struct cfg80211_rnr_elems *
+nl80211_parse_rnr_elems(struct wiphy *wiphy, struct nlattr *attrs,
+			struct netlink_ext_ack *extack)
+{
+	struct nlattr *nl_elems;
+	struct cfg80211_rnr_elems *elems;
+	int rem_elems;
+	u8 i = 0, num_elems = 0;
+
+	nla_for_each_nested(nl_elems, attrs, rem_elems) {
+		int ret;
+
+		ret = validate_ie_attr(nl_elems, extack);
+		if (ret)
+			return ERR_PTR(ret);
+
+		num_elems++;
+	}
+
+	elems = kzalloc(struct_size(elems, elem, num_elems), GFP_KERNEL);
+	if (!elems)
+		return ERR_PTR(-ENOMEM);
+
+	nla_for_each_nested(nl_elems, attrs, rem_elems) {
+		elems->elem[i].data = nla_data(nl_elems);
+		elems->elem[i].len = nla_len(nl_elems);
+		i++;
+	}
+	elems->cnt = num_elems;
+	return elems;
+}
+
 static int nl80211_parse_he_bss_color(struct nlattr *attrs,
 				      struct cfg80211_he_bss_color *he_bss_color)
 {
@@ -5451,7 +5484,8 @@ static int nl80211_parse_he_bss_color(struct nlattr *attrs,
 
 static int nl80211_parse_beacon(struct cfg80211_registered_device *rdev,
 				struct nlattr *attrs[],
-				struct cfg80211_beacon_data *bcn)
+				struct cfg80211_beacon_data *bcn,
+				struct netlink_ext_ack *extack)
 {
 	bool haveinfo = false;
 	int err;
@@ -5548,6 +5582,21 @@ static int nl80211_parse_beacon(struct cfg80211_registered_device *rdev,
 			return PTR_ERR(mbssid);
 
 		bcn->mbssid_ies = mbssid;
+
+		if (bcn->mbssid_ies && attrs[NL80211_ATTR_EMA_RNR_ELEMS]) {
+			struct cfg80211_rnr_elems *rnr =
+				nl80211_parse_rnr_elems(&rdev->wiphy,
+							attrs[NL80211_ATTR_EMA_RNR_ELEMS],
+							extack);
+
+			if (IS_ERR(rnr))
+				return PTR_ERR(rnr);
+
+			if (rnr && rnr->cnt < bcn->mbssid_ies->cnt)
+				return -EINVAL;
+
+			bcn->rnr_ies = rnr;
+		}
 	}
 
 	return 0;
@@ -5866,7 +5915,8 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
 	if (!params)
 		return -ENOMEM;
 
-	err = nl80211_parse_beacon(rdev, info->attrs, &params->beacon);
+	err = nl80211_parse_beacon(rdev, info->attrs, &params->beacon,
+				   info->extack);
 	if (err)
 		goto out;
 
@@ -6096,6 +6146,11 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
 			goto out_unlock;
 	}
 
+	if (!params->mbssid_config.ema && params->beacon.rnr_ies) {
+		err = -EINVAL;
+		goto out_unlock;
+	}
+
 	err = nl80211_calculate_ap_params(params);
 	if (err)
 		goto out_unlock;
@@ -6137,6 +6192,7 @@ out:
 	    params->mbssid_config.tx_wdev->netdev &&
 	    params->mbssid_config.tx_wdev->netdev != dev)
 		dev_put(params->mbssid_config.tx_wdev->netdev);
+	kfree(params->beacon.rnr_ies);
 	kfree(params);
 
 	return err;
@@ -6161,7 +6217,7 @@ static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info)
 	if (!wdev->links[link_id].ap.beacon_interval)
 		return -EINVAL;
 
-	err = nl80211_parse_beacon(rdev, info->attrs, &params);
+	err = nl80211_parse_beacon(rdev, info->attrs, &params, info->extack);
 	if (err)
 		goto out;
 
@@ -6171,6 +6227,7 @@ static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info)
 
 out:
 	kfree(params.mbssid_ies);
+	kfree(params.rnr_ies);
 	return err;
 }
 
@@ -10030,7 +10087,8 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info)
 	if (!need_new_beacon)
 		goto skip_beacons;
 
-	err = nl80211_parse_beacon(rdev, info->attrs, &params.beacon_after);
+	err = nl80211_parse_beacon(rdev, info->attrs, &params.beacon_after,
+				   info->extack);
 	if (err)
 		goto free;
 
@@ -10047,7 +10105,8 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto free;
 
-	err = nl80211_parse_beacon(rdev, csa_attrs, &params.beacon_csa);
+	err = nl80211_parse_beacon(rdev, csa_attrs, &params.beacon_csa,
+				   info->extack);
 	if (err)
 		goto free;
 
@@ -10167,6 +10226,8 @@ skip_beacons:
 free:
 	kfree(params.beacon_after.mbssid_ies);
 	kfree(params.beacon_csa.mbssid_ies);
+	kfree(params.beacon_after.rnr_ies);
+	kfree(params.beacon_csa.rnr_ies);
 	kfree(csa_attrs);
 	return err;
 }
@@ -15882,7 +15943,8 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info)
 	params.count = nla_get_u8(info->attrs[NL80211_ATTR_COLOR_CHANGE_COUNT]);
 	params.color = nla_get_u8(info->attrs[NL80211_ATTR_COLOR_CHANGE_COLOR]);
 
-	err = nl80211_parse_beacon(rdev, info->attrs, &params.beacon_next);
+	err = nl80211_parse_beacon(rdev, info->attrs, &params.beacon_next,
+				   info->extack);
 	if (err)
 		return err;
 
@@ -15896,7 +15958,8 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto out;
 
-	err = nl80211_parse_beacon(rdev, tb, &params.beacon_color_change);
+	err = nl80211_parse_beacon(rdev, tb, &params.beacon_color_change,
+				   info->extack);
 	if (err)
 		goto out;
 
@@ -15952,6 +16015,8 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info)
 out:
 	kfree(params.beacon_next.mbssid_ies);
 	kfree(params.beacon_color_change.mbssid_ies);
+	kfree(params.beacon_next.rnr_ies);
+	kfree(params.beacon_color_change.rnr_ies);
 	kfree(tb);
 	return err;
 }
-- 
cgit v1.2.3-70-g09d2


From 68b9bea267bfc1259e195dcac1bf69db0c0c28da Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Thu, 23 Mar 2023 04:38:01 -0700
Subject: mac80211: support RNR for EMA AP

Generate EMA beacons, each including MBSSID and RNR elements at a given
index. If number of stored RNR elements is more than the number of
MBSSID elements then add those in every EMA beacon.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Link: https://lore.kernel.org/r/20230323113801.6903-3-quic_alokad@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c         | 63 ++++++++++++++++++++++++++++++++++++++++++----
 net/mac80211/ieee80211_i.h | 21 +++++++++++++---
 net/mac80211/tx.c          | 10 ++++++++
 3 files changed, 86 insertions(+), 8 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index db5fa334b801..ebbd13d0c29a 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1084,6 +1084,23 @@ ieee80211_copy_mbssid_beacon(u8 *pos, struct cfg80211_mbssid_elems *dst,
 	return offset;
 }
 
+static int
+ieee80211_copy_rnr_beacon(u8 *pos, struct cfg80211_rnr_elems *dst,
+			  struct cfg80211_rnr_elems *src)
+{
+	int i, offset = 0;
+
+	for (i = 0; i < src->cnt; i++) {
+		memcpy(pos + offset, src->elem[i].data, src->elem[i].len);
+		dst->elem[i].len = src->elem[i].len;
+		dst->elem[i].data = pos + offset;
+		offset += dst->elem[i].len;
+	}
+	dst->cnt = src->cnt;
+
+	return offset;
+}
+
 static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata,
 				   struct ieee80211_link_data *link,
 				   struct cfg80211_beacon_data *params,
@@ -1091,6 +1108,7 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata,
 				   const struct ieee80211_color_change_settings *cca)
 {
 	struct cfg80211_mbssid_elems *mbssid = NULL;
+	struct cfg80211_rnr_elems *rnr = NULL;
 	struct beacon_data *new, *old;
 	int new_head_len, new_tail_len;
 	int size, err;
@@ -1122,11 +1140,21 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata,
 	if (params->mbssid_ies) {
 		mbssid = params->mbssid_ies;
 		size += struct_size(new->mbssid_ies, elem, mbssid->cnt);
-		size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt);
+		if (params->rnr_ies) {
+			rnr = params->rnr_ies;
+			size += struct_size(new->rnr_ies, elem, rnr->cnt);
+		}
+		size += ieee80211_get_mbssid_beacon_len(mbssid, rnr,
+							mbssid->cnt);
 	} else if (old && old->mbssid_ies) {
 		mbssid = old->mbssid_ies;
 		size += struct_size(new->mbssid_ies, elem, mbssid->cnt);
-		size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt);
+		if (old && old->rnr_ies) {
+			rnr = old->rnr_ies;
+			size += struct_size(new->rnr_ies, elem, rnr->cnt);
+		}
+		size += ieee80211_get_mbssid_beacon_len(mbssid, rnr,
+							mbssid->cnt);
 	}
 
 	new = kzalloc(size, GFP_KERNEL);
@@ -1137,7 +1165,7 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata,
 
 	/*
 	 * pointers go into the block we allocated,
-	 * memory is | beacon_data | head | tail | mbssid_ies
+	 * memory is | beacon_data | head | tail | mbssid_ies | rnr_ies
 	 */
 	new->head = ((u8 *) new) + sizeof(*new);
 	new->tail = new->head + new_head_len;
@@ -1149,7 +1177,13 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata,
 
 		new->mbssid_ies = (void *)pos;
 		pos += struct_size(new->mbssid_ies, elem, mbssid->cnt);
-		ieee80211_copy_mbssid_beacon(pos, new->mbssid_ies, mbssid);
+		pos += ieee80211_copy_mbssid_beacon(pos, new->mbssid_ies,
+						    mbssid);
+		if (rnr) {
+			new->rnr_ies = (void *)pos;
+			pos += struct_size(new->rnr_ies, elem, rnr->cnt);
+			ieee80211_copy_rnr_beacon(pos, new->rnr_ies, rnr);
+		}
 		/* update bssid_indicator */
 		link_conf->bssid_indicator =
 			ilog2(__roundup_pow_of_two(mbssid->cnt + 1));
@@ -1507,6 +1541,7 @@ static void ieee80211_free_next_beacon(struct ieee80211_link_data *link)
 		return;
 
 	kfree(link->u.ap.next_beacon->mbssid_ies);
+	kfree(link->u.ap.next_beacon->rnr_ies);
 	kfree(link->u.ap.next_beacon);
 	link->u.ap.next_beacon = NULL;
 }
@@ -3410,6 +3445,7 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon)
 
 	if (beacon->mbssid_ies)
 		len += ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies,
+						       beacon->rnr_ies,
 						       beacon->mbssid_ies->cnt);
 
 	new_beacon = kzalloc(sizeof(*new_beacon) + len, GFP_KERNEL);
@@ -3425,6 +3461,18 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon)
 			kfree(new_beacon);
 			return NULL;
 		}
+
+		if (beacon->rnr_ies && beacon->rnr_ies->cnt) {
+			new_beacon->rnr_ies =
+				kzalloc(struct_size(new_beacon->rnr_ies,
+						    elem, beacon->rnr_ies->cnt),
+					GFP_KERNEL);
+			if (!new_beacon->rnr_ies) {
+				kfree(new_beacon->mbssid_ies);
+				kfree(new_beacon);
+				return NULL;
+			}
+		}
 	}
 
 	pos = (u8 *)(new_beacon + 1);
@@ -3464,10 +3512,15 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon)
 		memcpy(pos, beacon->probe_resp, beacon->probe_resp_len);
 		pos += beacon->probe_resp_len;
 	}
-	if (beacon->mbssid_ies && beacon->mbssid_ies->cnt)
+	if (beacon->mbssid_ies && beacon->mbssid_ies->cnt) {
 		pos += ieee80211_copy_mbssid_beacon(pos,
 						    new_beacon->mbssid_ies,
 						    beacon->mbssid_ies);
+		if (beacon->rnr_ies && beacon->rnr_ies->cnt)
+			pos += ieee80211_copy_rnr_beacon(pos,
+							 new_beacon->rnr_ies,
+							 beacon->rnr_ies);
+	}
 
 	/* might copy -1, meaning no changes requested */
 	new_beacon->ftm_responder = beacon->ftm_responder;
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 84d10e993eca..85ecbf57d64e 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -270,6 +270,7 @@ struct beacon_data {
 	u16 cntdwn_counter_offsets[IEEE80211_MAX_CNTDWN_COUNTERS_NUM];
 	u8 cntdwn_current_counter;
 	struct cfg80211_mbssid_elems *mbssid_ies;
+	struct cfg80211_rnr_elems *rnr_ies;
 	struct rcu_head rcu_head;
 };
 
@@ -1186,20 +1187,34 @@ ieee80211_vif_get_shift(struct ieee80211_vif *vif)
 }
 
 static inline int
-ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems, u8 i)
+ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems,
+				struct cfg80211_rnr_elems *rnr_elems,
+				u8 i)
 {
 	int len = 0;
 
 	if (!elems || !elems->cnt || i > elems->cnt)
 		return 0;
 
-	if (i < elems->cnt)
-		return elems->elem[i].len;
+	if (i < elems->cnt) {
+		len = elems->elem[i].len;
+		if (rnr_elems) {
+			len += rnr_elems->elem[i].len;
+			for (i = elems->cnt; i < rnr_elems->cnt; i++)
+				len += rnr_elems->elem[i].len;
+		}
+		return len;
+	}
 
 	/* i == elems->cnt, calculate total length of all MBSSID elements */
 	for (i = 0; i < elems->cnt; i++)
 		len += elems->elem[i].len;
 
+	if (rnr_elems) {
+		for (i = 0; i < rnr_elems->cnt; i++)
+			len += rnr_elems->elem[i].len;
+	}
+
 	return len;
 }
 
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 139eec6c64da..dfe6b9c9b29e 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -5222,6 +5222,15 @@ ieee80211_beacon_add_mbssid(struct sk_buff *skb, struct beacon_data *beacon,
 	if (i < beacon->mbssid_ies->cnt) {
 		skb_put_data(skb, beacon->mbssid_ies->elem[i].data,
 			     beacon->mbssid_ies->elem[i].len);
+
+		if (beacon->rnr_ies && beacon->rnr_ies->cnt) {
+			skb_put_data(skb, beacon->rnr_ies->elem[i].data,
+				     beacon->rnr_ies->elem[i].len);
+
+			for (i = beacon->mbssid_ies->cnt; i < beacon->rnr_ies->cnt; i++)
+				skb_put_data(skb, beacon->rnr_ies->elem[i].data,
+					     beacon->rnr_ies->elem[i].len);
+		}
 		return;
 	}
 
@@ -5259,6 +5268,7 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw,
 	 * tail length, maximum TIM length and multiple BSSID length
 	 */
 	mbssid_len = ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies,
+						     beacon->rnr_ies,
 						     ema_index);
 
 	skb = dev_alloc_skb(local->tx_headroom + beacon->head_len +
-- 
cgit v1.2.3-70-g09d2


From 92d13386ec55816b3bdd6d578e59a71226e12806 Mon Sep 17 00:00:00 2001
From: Jaewan Kim <jaewan@google.com>
Date: Wed, 22 Mar 2023 13:16:33 +0000
Subject: mac80211_hwsim: add PMSR capability support

PMSR (a.k.a. peer measurement) is generalized measurement between two
Wi-Fi devices. And currently FTM (a.k.a. fine time measurement or flight
time measurement) is the one and only measurement. FTM is measured by
RTT (a.k.a. round trip time) of packets between two Wi-Fi devices.

Add necessary functionality to allow mac80211_hwsim to be configured with
PMSR capability. The capability is mandatory to accept incoming PMSR
request because nl80211_pmsr_start() ignores incoming the request without
the PMSR capability.

In detail, add new mac80211_hwsim attribute HWSIM_ATTR_PMSR_SUPPORT.
HWSIM_ATTR_PMSR_SUPPORT is used to set PMSR capability when creating a new
radio. To send extra capability details, HWSIM_ATTR_PMSR_SUPPORT can have
nested PMSR capability attributes defined in the nl80211.h. Data format is
the same as cfg80211_pmsr_capabilities.

If HWSIM_ATTR_PMSR_SUPPORT is specified, mac80211_hwsim builds
cfg80211_pmsr_capabilities and sets wiphy.pmsr_capa.

Signed-off-by: Jaewan Kim <jaewan@google.com>
Link: https://lore.kernel.org/r/20230322131637.2633968-2-jaewan@google.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 130 ++++++++++++++++++++++++++
 drivers/net/wireless/virtual/mac80211_hwsim.h |   3 +
 2 files changed, 133 insertions(+)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index 2aa7b740b1ce..d72cff77c157 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -719,6 +719,9 @@ struct mac80211_hwsim_data {
 	/* RSSI in rx status of the receiver */
 	int rx_rssi;
 
+	/* only used when pmsr capability is supplied */
+	struct cfg80211_pmsr_capabilities pmsr_capa;
+
 	struct mac80211_hwsim_link_data link_data[IEEE80211_MLD_MAX_NUM_LINKS];
 };
 
@@ -760,6 +763,34 @@ static const struct genl_multicast_group hwsim_mcgrps[] = {
 
 /* MAC80211_HWSIM netlink policy */
 
+static const struct nla_policy
+hwsim_ftm_capa_policy[NL80211_PMSR_FTM_CAPA_ATTR_MAX + 1] = {
+	[NL80211_PMSR_FTM_CAPA_ATTR_ASAP] = { .type = NLA_FLAG },
+	[NL80211_PMSR_FTM_CAPA_ATTR_NON_ASAP] = { .type = NLA_FLAG },
+	[NL80211_PMSR_FTM_CAPA_ATTR_REQ_LCI] = { .type = NLA_FLAG },
+	[NL80211_PMSR_FTM_CAPA_ATTR_REQ_CIVICLOC] = { .type = NLA_FLAG },
+	[NL80211_PMSR_FTM_CAPA_ATTR_PREAMBLES] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_CAPA_ATTR_BANDWIDTHS] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_CAPA_ATTR_MAX_BURSTS_EXPONENT] = NLA_POLICY_MAX(NLA_U8, 15),
+	[NL80211_PMSR_FTM_CAPA_ATTR_MAX_FTMS_PER_BURST] = NLA_POLICY_MAX(NLA_U8, 31),
+	[NL80211_PMSR_FTM_CAPA_ATTR_TRIGGER_BASED] = { .type = NLA_FLAG },
+	[NL80211_PMSR_FTM_CAPA_ATTR_NON_TRIGGER_BASED] = { .type = NLA_FLAG },
+};
+
+static const struct nla_policy
+hwsim_pmsr_capa_type_policy[NL80211_PMSR_TYPE_MAX + 1] = {
+	[NL80211_PMSR_TYPE_FTM] = NLA_POLICY_NESTED(hwsim_ftm_capa_policy),
+};
+
+static const struct nla_policy
+hwsim_pmsr_capa_policy[NL80211_PMSR_ATTR_MAX + 1] = {
+	[NL80211_PMSR_ATTR_MAX_PEERS] = { .type = NLA_U32 },
+	[NL80211_PMSR_ATTR_REPORT_AP_TSF] = { .type = NLA_FLAG },
+	[NL80211_PMSR_ATTR_RANDOMIZE_MAC_ADDR] = { .type = NLA_FLAG },
+	[NL80211_PMSR_ATTR_TYPE_CAPA] = NLA_POLICY_NESTED(hwsim_pmsr_capa_type_policy),
+	[NL80211_PMSR_ATTR_PEERS] = { .type = NLA_REJECT }, // only for request.
+};
+
 static const struct nla_policy hwsim_genl_policy[HWSIM_ATTR_MAX + 1] = {
 	[HWSIM_ATTR_ADDR_RECEIVER] = NLA_POLICY_ETH_ADDR_COMPAT,
 	[HWSIM_ATTR_ADDR_TRANSMITTER] = NLA_POLICY_ETH_ADDR_COMPAT,
@@ -788,6 +819,7 @@ static const struct nla_policy hwsim_genl_policy[HWSIM_ATTR_MAX + 1] = {
 	[HWSIM_ATTR_IFTYPE_SUPPORT] = { .type = NLA_U32 },
 	[HWSIM_ATTR_CIPHER_SUPPORT] = { .type = NLA_BINARY },
 	[HWSIM_ATTR_MLO_SUPPORT] = { .type = NLA_FLAG },
+	[HWSIM_ATTR_PMSR_SUPPORT] = NLA_POLICY_NESTED(hwsim_pmsr_capa_policy),
 };
 
 #if IS_REACHABLE(CONFIG_VIRTIO)
@@ -3217,6 +3249,7 @@ struct hwsim_new_radio_params {
 	u32 *ciphers;
 	u8 n_ciphers;
 	bool mlo;
+	const struct cfg80211_pmsr_capabilities *pmsr_capa;
 };
 
 static void hwsim_mcast_config_msg(struct sk_buff *mcast_skb,
@@ -4479,6 +4512,7 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 			      NL80211_EXT_FEATURE_MULTICAST_REGISTRATIONS);
 	wiphy_ext_feature_set(hw->wiphy,
 			      NL80211_EXT_FEATURE_BEACON_RATE_LEGACY);
+	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_ENABLE_FTM_RESPONDER);
 
 	wiphy_ext_feature_set(hw->wiphy,
 			      NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT);
@@ -4643,6 +4677,11 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 				    data->debugfs,
 				    data, &hwsim_simulate_radar);
 
+	if (param->pmsr_capa) {
+		data->pmsr_capa = *param->pmsr_capa;
+		hw->wiphy->pmsr_capa = &data->pmsr_capa;
+	}
+
 	spin_lock_bh(&hwsim_radio_lock);
 	err = rhashtable_insert_fast(&hwsim_radios_rht, &data->rht,
 				     hwsim_rht_params);
@@ -4752,6 +4791,7 @@ static int mac80211_hwsim_get_radio(struct sk_buff *skb,
 	param.regd = data->regd;
 	param.channels = data->channels;
 	param.hwname = wiphy_name(data->hw->wiphy);
+	param.pmsr_capa = &data->pmsr_capa;
 
 	res = append_radio_msg(skb, data->idx, &param);
 	if (res < 0)
@@ -5090,6 +5130,79 @@ static bool hwsim_known_ciphers(const u32 *ciphers, int n_ciphers)
 	return true;
 }
 
+static int parse_ftm_capa(const struct nlattr *ftm_capa, struct cfg80211_pmsr_capabilities *out,
+			  struct genl_info *info)
+{
+	struct nlattr *tb[NL80211_PMSR_FTM_CAPA_ATTR_MAX + 1];
+	int ret;
+
+	ret = nla_parse_nested(tb, NL80211_PMSR_FTM_CAPA_ATTR_MAX, ftm_capa, hwsim_ftm_capa_policy,
+			       NULL);
+	if (ret) {
+		NL_SET_ERR_MSG_ATTR(info->extack, ftm_capa, "malformed FTM capability");
+		return -EINVAL;
+	}
+
+	out->ftm.supported = 1;
+	if (tb[NL80211_PMSR_FTM_CAPA_ATTR_PREAMBLES])
+		out->ftm.preambles = nla_get_u32(tb[NL80211_PMSR_FTM_CAPA_ATTR_PREAMBLES]);
+	if (tb[NL80211_PMSR_FTM_CAPA_ATTR_BANDWIDTHS])
+		out->ftm.bandwidths = nla_get_u32(tb[NL80211_PMSR_FTM_CAPA_ATTR_BANDWIDTHS]);
+	if (tb[NL80211_PMSR_FTM_CAPA_ATTR_MAX_BURSTS_EXPONENT])
+		out->ftm.max_bursts_exponent =
+			nla_get_u8(tb[NL80211_PMSR_FTM_CAPA_ATTR_MAX_BURSTS_EXPONENT]);
+	if (tb[NL80211_PMSR_FTM_CAPA_ATTR_MAX_FTMS_PER_BURST])
+		out->ftm.max_ftms_per_burst =
+			nla_get_u8(tb[NL80211_PMSR_FTM_CAPA_ATTR_MAX_FTMS_PER_BURST]);
+	out->ftm.asap = !!tb[NL80211_PMSR_FTM_CAPA_ATTR_ASAP];
+	out->ftm.non_asap = !!tb[NL80211_PMSR_FTM_CAPA_ATTR_NON_ASAP];
+	out->ftm.request_lci = !!tb[NL80211_PMSR_FTM_CAPA_ATTR_REQ_LCI];
+	out->ftm.request_civicloc = !!tb[NL80211_PMSR_FTM_CAPA_ATTR_REQ_CIVICLOC];
+	out->ftm.trigger_based = !!tb[NL80211_PMSR_FTM_CAPA_ATTR_TRIGGER_BASED];
+	out->ftm.non_trigger_based = !!tb[NL80211_PMSR_FTM_CAPA_ATTR_NON_TRIGGER_BASED];
+
+	return 0;
+}
+
+static int parse_pmsr_capa(const struct nlattr *pmsr_capa, struct cfg80211_pmsr_capabilities *out,
+			   struct genl_info *info)
+{
+	struct nlattr *tb[NL80211_PMSR_ATTR_MAX + 1];
+	struct nlattr *nla;
+	int size;
+	int ret;
+
+	ret = nla_parse_nested(tb, NL80211_PMSR_ATTR_MAX, pmsr_capa, hwsim_pmsr_capa_policy, NULL);
+	if (ret) {
+		NL_SET_ERR_MSG_ATTR(info->extack, pmsr_capa, "malformed PMSR capability");
+		return -EINVAL;
+	}
+
+	if (tb[NL80211_PMSR_ATTR_MAX_PEERS])
+		out->max_peers = nla_get_u32(tb[NL80211_PMSR_ATTR_MAX_PEERS]);
+	out->report_ap_tsf = !!tb[NL80211_PMSR_ATTR_REPORT_AP_TSF];
+	out->randomize_mac_addr = !!tb[NL80211_PMSR_ATTR_RANDOMIZE_MAC_ADDR];
+
+	if (!tb[NL80211_PMSR_ATTR_TYPE_CAPA]) {
+		NL_SET_ERR_MSG_ATTR(info->extack, tb[NL80211_PMSR_ATTR_TYPE_CAPA],
+				    "malformed PMSR type");
+		return -EINVAL;
+	}
+
+	nla_for_each_nested(nla, tb[NL80211_PMSR_ATTR_TYPE_CAPA], size) {
+		switch (nla_type(nla)) {
+		case NL80211_PMSR_TYPE_FTM:
+			parse_ftm_capa(nla, out, info);
+			break;
+		default:
+			NL_SET_ERR_MSG_ATTR(info->extack, nla, "unsupported measurement type");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
 static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info)
 {
 	struct hwsim_new_radio_params param = { 0 };
@@ -5210,8 +5323,25 @@ static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info)
 		param.hwname = hwname;
 	}
 
+	if (info->attrs[HWSIM_ATTR_PMSR_SUPPORT]) {
+		struct cfg80211_pmsr_capabilities *pmsr_capa;
+
+		pmsr_capa = kmalloc(sizeof(*pmsr_capa), GFP_KERNEL);
+		if (!pmsr_capa) {
+			ret = -ENOMEM;
+			goto out_free;
+		}
+		ret = parse_pmsr_capa(info->attrs[HWSIM_ATTR_PMSR_SUPPORT], pmsr_capa, info);
+		if (ret)
+			goto out_free;
+		param.pmsr_capa = pmsr_capa;
+	}
+
 	ret = mac80211_hwsim_new_radio(info, &param);
+
+out_free:
 	kfree(hwname);
+	kfree(param.pmsr_capa);
 	return ret;
 }
 
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
index 527799b2de0f..d10fa7f4853b 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.h
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -142,6 +142,8 @@ enum {
  * @HWSIM_ATTR_CIPHER_SUPPORT: u32 array of supported cipher types
  * @HWSIM_ATTR_MLO_SUPPORT: claim MLO support (exact parameters TBD) for
  *	the new radio
+ * @HWSIM_ATTR_PMSR_SUPPORT: nested attribute used with %HWSIM_CMD_CREATE_RADIO
+ *	to provide peer measurement capabilities. (nl80211_peer_measurement_attrs)
  * @__HWSIM_ATTR_MAX: enum limit
  */
 
@@ -173,6 +175,7 @@ enum {
 	HWSIM_ATTR_IFTYPE_SUPPORT,
 	HWSIM_ATTR_CIPHER_SUPPORT,
 	HWSIM_ATTR_MLO_SUPPORT,
+	HWSIM_ATTR_PMSR_SUPPORT,
 	__HWSIM_ATTR_MAX,
 };
 #define HWSIM_ATTR_MAX (__HWSIM_ATTR_MAX - 1)
-- 
cgit v1.2.3-70-g09d2


From 5097f84437c9bd50b2c65b5f85395c34b2d545db Mon Sep 17 00:00:00 2001
From: Jaewan Kim <jaewan@google.com>
Date: Wed, 22 Mar 2023 13:16:34 +0000
Subject: wifi: nl80211: make nl80211_send_chandef non-static

Expose nl80211_send_chandef functionality for mac80211_hwsim or vendor
netlink can use it where needed.

Signed-off-by: Jaewan Kim <jaewan@google.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Link: https://lore.kernel.org/r/20230322131637.2633968-3-jaewan@google.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h | 9 +++++++++
 net/wireless/nl80211.c | 4 ++--
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 3cf236520288..9e04f69712b1 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -950,6 +950,15 @@ int cfg80211_chandef_dfs_required(struct wiphy *wiphy,
 				  const struct cfg80211_chan_def *chandef,
 				  enum nl80211_iftype iftype);
 
+/**
+ * nl80211_send_chandef - sends the channel definition.
+ * @msg: the msg to send channel definition
+ * @chandef: the channel definition to check
+ *
+ * Returns: 0 if sent the channel definition to msg, < 0 on error
+ **/
+int nl80211_send_chandef(struct sk_buff *msg, const struct cfg80211_chan_def *chandef);
+
 /**
  * ieee80211_chanwidth_rate_flags - return rate flags for channel width
  * @width: the channel width of the channel
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 80a20d69f285..f1cd3d9130dd 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -3765,8 +3765,7 @@ out:
 	return result;
 }
 
-static int nl80211_send_chandef(struct sk_buff *msg,
-				const struct cfg80211_chan_def *chandef)
+int nl80211_send_chandef(struct sk_buff *msg, const struct cfg80211_chan_def *chandef)
 {
 	if (WARN_ON(!cfg80211_chandef_valid(chandef)))
 		return -EINVAL;
@@ -3797,6 +3796,7 @@ static int nl80211_send_chandef(struct sk_buff *msg,
 		return -ENOBUFS;
 	return 0;
 }
+EXPORT_SYMBOL(nl80211_send_chandef);
 
 static int nl80211_send_iface(struct sk_buff *msg, u32 portid, u32 seq, int flags,
 			      struct cfg80211_registered_device *rdev,
-- 
cgit v1.2.3-70-g09d2


From 5530c04c87c5974d440ea95142e3e43ef7da4d5b Mon Sep 17 00:00:00 2001
From: Jaewan Kim <jaewan@google.com>
Date: Wed, 22 Mar 2023 13:16:35 +0000
Subject: mac80211_hwsim: add PMSR request support via virtio

PMSR (a.k.a. peer measurement) is generalized measurement between two
Wi-Fi devices. And currently FTM (a.k.a. fine time measurement or flight
time measurement) is the one and only measurement. FTM is measured by
RTT (a.k.a. round trip time) of packets between two Wi-Fi devices.

Add necessary functionalities for mac80211_hwsim to start PMSR request by
passthrough the request to wmediumd via virtio. mac80211_hwsim can't
measure RTT for real because mac80211_hwsim the software simulator and
packets are sent almost immediately for real. This change expect wmediumd
to have all the location information of devices, so passthrough requests
to wmediumd.

In detail, add new mac80211_hwsim command HWSIM_CMD_ABORT_PMSR. When
mac80211_hwsim receives the PMSR start request via
ieee80211_ops.start_pmsr, the received cfg80211_pmsr_request is resent to
the wmediumd with command HWSIM_CMD_START_PMSR and attribute
HWSIM_ATTR_PMSR_REQUEST. The attribute is formatted as the same way as
nl80211_pmsr_start() expects.

Signed-off-by: Jaewan Kim <jaewan@google.com>
Link: https://lore.kernel.org/r/20230322131637.2633968-4-jaewan@google.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 210 +++++++++++++++++++++++++-
 drivers/net/wireless/virtual/mac80211_hwsim.h |   6 +
 2 files changed, 215 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index d72cff77c157..c7cb4f66a168 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -721,6 +721,8 @@ struct mac80211_hwsim_data {
 
 	/* only used when pmsr capability is supplied */
 	struct cfg80211_pmsr_capabilities pmsr_capa;
+	struct cfg80211_pmsr_request *pmsr_request;
+	struct wireless_dev *pmsr_request_wdev;
 
 	struct mac80211_hwsim_link_data link_data[IEEE80211_MLD_MAX_NUM_LINKS];
 };
@@ -3170,6 +3172,211 @@ static int mac80211_hwsim_change_sta_links(struct ieee80211_hw *hw,
 	return 0;
 }
 
+static int mac80211_hwsim_send_pmsr_ftm_request_peer(struct sk_buff *msg,
+						     struct cfg80211_pmsr_ftm_request_peer *request)
+{
+	struct nlattr *ftm;
+
+	if (!request->requested)
+		return -EINVAL;
+
+	ftm = nla_nest_start(msg, NL80211_PMSR_TYPE_FTM);
+	if (!ftm)
+		return -ENOBUFS;
+
+	if (nla_put_u32(msg, NL80211_PMSR_FTM_REQ_ATTR_PREAMBLE, request->preamble))
+		return -ENOBUFS;
+
+	if (nla_put_u16(msg, NL80211_PMSR_FTM_REQ_ATTR_BURST_PERIOD, request->burst_period))
+		return -ENOBUFS;
+
+	if (request->asap && nla_put_flag(msg, NL80211_PMSR_FTM_REQ_ATTR_ASAP))
+		return -ENOBUFS;
+
+	if (request->request_lci && nla_put_flag(msg, NL80211_PMSR_FTM_REQ_ATTR_REQUEST_LCI))
+		return -ENOBUFS;
+
+	if (request->request_civicloc &&
+	    nla_put_flag(msg, NL80211_PMSR_FTM_REQ_ATTR_REQUEST_CIVICLOC))
+		return -ENOBUFS;
+
+	if (request->trigger_based && nla_put_flag(msg, NL80211_PMSR_FTM_REQ_ATTR_TRIGGER_BASED))
+		return -ENOBUFS;
+
+	if (request->non_trigger_based &&
+	    nla_put_flag(msg, NL80211_PMSR_FTM_REQ_ATTR_NON_TRIGGER_BASED))
+		return -ENOBUFS;
+
+	if (request->lmr_feedback && nla_put_flag(msg, NL80211_PMSR_FTM_REQ_ATTR_LMR_FEEDBACK))
+		return -ENOBUFS;
+
+	if (nla_put_u8(msg, NL80211_PMSR_FTM_REQ_ATTR_NUM_BURSTS_EXP, request->num_bursts_exp))
+		return -ENOBUFS;
+
+	if (nla_put_u8(msg, NL80211_PMSR_FTM_REQ_ATTR_BURST_DURATION, request->burst_duration))
+		return -ENOBUFS;
+
+	if (nla_put_u8(msg, NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST, request->ftms_per_burst))
+		return -ENOBUFS;
+
+	if (nla_put_u8(msg, NL80211_PMSR_FTM_REQ_ATTR_NUM_FTMR_RETRIES, request->ftmr_retries))
+		return -ENOBUFS;
+
+	if (nla_put_u8(msg, NL80211_PMSR_FTM_REQ_ATTR_BURST_DURATION, request->burst_duration))
+		return -ENOBUFS;
+
+	if (nla_put_u8(msg, NL80211_PMSR_FTM_REQ_ATTR_BSS_COLOR, request->bss_color))
+		return -ENOBUFS;
+
+	nla_nest_end(msg, ftm);
+
+	return 0;
+}
+
+static int mac80211_hwsim_send_pmsr_request_peer(struct sk_buff *msg,
+						 struct cfg80211_pmsr_request_peer *request)
+{
+	struct nlattr *peer, *chandef, *req, *data;
+	int err;
+
+	peer = nla_nest_start(msg, NL80211_PMSR_ATTR_PEERS);
+	if (!peer)
+		return -ENOBUFS;
+
+	if (nla_put(msg, NL80211_PMSR_PEER_ATTR_ADDR, ETH_ALEN,
+		    request->addr))
+		return -ENOBUFS;
+
+	chandef = nla_nest_start(msg, NL80211_PMSR_PEER_ATTR_CHAN);
+	if (!chandef)
+		return -ENOBUFS;
+
+	err = nl80211_send_chandef(msg, &request->chandef);
+	if (err)
+		return err;
+
+	nla_nest_end(msg, chandef);
+
+	req = nla_nest_start(msg, NL80211_PMSR_PEER_ATTR_REQ);
+	if (!req)
+		return -ENOBUFS;
+
+	if (request->report_ap_tsf && nla_put_flag(msg, NL80211_PMSR_REQ_ATTR_GET_AP_TSF))
+		return -ENOBUFS;
+
+	data = nla_nest_start(msg, NL80211_PMSR_REQ_ATTR_DATA);
+	if (!data)
+		return -ENOBUFS;
+
+	err = mac80211_hwsim_send_pmsr_ftm_request_peer(msg, &request->ftm);
+	if (err)
+		return err;
+
+	nla_nest_end(msg, data);
+	nla_nest_end(msg, req);
+	nla_nest_end(msg, peer);
+
+	return 0;
+}
+
+static int mac80211_hwsim_send_pmsr_request(struct sk_buff *msg,
+					    struct cfg80211_pmsr_request *request)
+{
+	struct nlattr *pmsr;
+	int err;
+
+	pmsr = nla_nest_start(msg, NL80211_ATTR_PEER_MEASUREMENTS);
+	if (!pmsr)
+		return -ENOBUFS;
+
+	if (nla_put_u32(msg, NL80211_ATTR_TIMEOUT, request->timeout))
+		return -ENOBUFS;
+
+	if (!is_zero_ether_addr(request->mac_addr)) {
+		if (nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, request->mac_addr))
+			return -ENOBUFS;
+		if (nla_put(msg, NL80211_ATTR_MAC_MASK, ETH_ALEN, request->mac_addr_mask))
+			return -ENOBUFS;
+	}
+
+	for (int i = 0; i < request->n_peers; i++) {
+		err = mac80211_hwsim_send_pmsr_request_peer(msg, &request->peers[i]);
+		if (err)
+			return err;
+	}
+
+	nla_nest_end(msg, pmsr);
+
+	return 0;
+}
+
+static int mac80211_hwsim_start_pmsr(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
+				     struct cfg80211_pmsr_request *request)
+{
+	struct mac80211_hwsim_data *data;
+	struct sk_buff *skb = NULL;
+	struct nlattr *pmsr;
+	void *msg_head;
+	u32 _portid;
+	int err = 0;
+
+	data = hw->priv;
+	_portid = READ_ONCE(data->wmediumd);
+	if (!_portid && !hwsim_virtio_enabled)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&data->mutex);
+
+	if (data->pmsr_request) {
+		err = -EBUSY;
+		goto out_free;
+	}
+
+	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+
+	if (!skb) {
+		err = -ENOMEM;
+		goto out_free;
+	}
+
+	msg_head = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0, HWSIM_CMD_START_PMSR);
+
+	if (nla_put(skb, HWSIM_ATTR_ADDR_TRANSMITTER,
+		    ETH_ALEN, data->addresses[1].addr)) {
+		err = -ENOMEM;
+		goto out_free;
+	}
+
+	pmsr = nla_nest_start(skb, HWSIM_ATTR_PMSR_REQUEST);
+	if (!pmsr) {
+		err = -ENOMEM;
+		goto out_free;
+	}
+
+	err = mac80211_hwsim_send_pmsr_request(skb, request);
+	if (err)
+		goto out_free;
+
+	nla_nest_end(skb, pmsr);
+
+	genlmsg_end(skb, msg_head);
+	if (hwsim_virtio_enabled)
+		hwsim_tx_virtio(data, skb);
+	else
+		hwsim_unicast_netgroup(data, skb, _portid);
+
+	data->pmsr_request = request;
+	data->pmsr_request_wdev = ieee80211_vif_to_wdev(vif);
+
+out_free:
+	if (err && skb)
+		nlmsg_free(skb);
+
+	mutex_unlock(&data->mutex);
+	return err;
+}
+
 #define HWSIM_COMMON_OPS					\
 	.tx = mac80211_hwsim_tx,				\
 	.wake_tx_queue = ieee80211_handle_wake_tx_queue,	\
@@ -3192,7 +3399,8 @@ static int mac80211_hwsim_change_sta_links(struct ieee80211_hw *hw,
 	.flush = mac80211_hwsim_flush,				\
 	.get_et_sset_count = mac80211_hwsim_get_et_sset_count,	\
 	.get_et_stats = mac80211_hwsim_get_et_stats,		\
-	.get_et_strings = mac80211_hwsim_get_et_strings,
+	.get_et_strings = mac80211_hwsim_get_et_strings,	\
+	.start_pmsr = mac80211_hwsim_start_pmsr,		\
 
 #define HWSIM_NON_MLO_OPS					\
 	.sta_add = mac80211_hwsim_sta_add,			\
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
index d10fa7f4853b..98e586a56582 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.h
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -81,6 +81,8 @@ enum hwsim_tx_control_flags {
  *	to this receiver address for a given station.
  * @HWSIM_CMD_DEL_MAC_ADDR: remove the MAC address again, the attributes
  *	are the same as to @HWSIM_CMD_ADD_MAC_ADDR.
+ * @HWSIM_CMD_START_PMSR: request to start peer measurement with the
+ *	%HWSIM_ATTR_PMSR_REQUEST.
  * @__HWSIM_CMD_MAX: enum limit
  */
 enum {
@@ -93,6 +95,7 @@ enum {
 	HWSIM_CMD_GET_RADIO,
 	HWSIM_CMD_ADD_MAC_ADDR,
 	HWSIM_CMD_DEL_MAC_ADDR,
+	HWSIM_CMD_START_PMSR,
 	__HWSIM_CMD_MAX,
 };
 #define HWSIM_CMD_MAX (_HWSIM_CMD_MAX - 1)
@@ -144,6 +147,8 @@ enum {
  *	the new radio
  * @HWSIM_ATTR_PMSR_SUPPORT: nested attribute used with %HWSIM_CMD_CREATE_RADIO
  *	to provide peer measurement capabilities. (nl80211_peer_measurement_attrs)
+ * @HWSIM_ATTR_PMSR_REQUEST: nested attribute used with %HWSIM_CMD_START_PMSR
+ *	to provide details about peer measurement request (nl80211_peer_measurement_attrs)
  * @__HWSIM_ATTR_MAX: enum limit
  */
 
@@ -176,6 +181,7 @@ enum {
 	HWSIM_ATTR_CIPHER_SUPPORT,
 	HWSIM_ATTR_MLO_SUPPORT,
 	HWSIM_ATTR_PMSR_SUPPORT,
+	HWSIM_ATTR_PMSR_REQUEST,
 	__HWSIM_ATTR_MAX,
 };
 #define HWSIM_ATTR_MAX (__HWSIM_ATTR_MAX - 1)
-- 
cgit v1.2.3-70-g09d2


From 8ba1da95053e02d9cb4a6849356eae6dc5687f62 Mon Sep 17 00:00:00 2001
From: Jaewan Kim <jaewan@google.com>
Date: Wed, 22 Mar 2023 13:16:36 +0000
Subject: mac80211_hwsim: add PMSR abort support via virtio

PMSR (a.k.a. peer measurement) is generalized measurement between two
devices with Wi-Fi support. And currently FTM (a.k.a. fine time
measurement or flight time measurement) is the one and only measurement.

Add necessary functionalities for mac80211_hwsim to abort previous PMSR
request. The abortion request is sent to the wmedium where the PMSR request
is actually handled.

In detail, add new mac80211_hwsim command HWSIM_CMD_ABORT_PMSR. When
mac80211_hwsim receives the PMSR abortion request via
ieee80211_ops.abort_pmsr, the received cfg80211_pmsr_request is resent to
the wmediumd with command HWSIM_CMD_ABORT_PMSR and attribute
HWSIM_ATTR_PMSR_REQUEST. The attribute is formatted as the same way as
nl80211_pmsr_start() expects.

Signed-off-by: Jaewan Kim <jaewan@google.com>
Link: https://lore.kernel.org/r/20230322131637.2633968-5-jaewan@google.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 62 +++++++++++++++++++++++++++
 drivers/net/wireless/virtual/mac80211_hwsim.h |  2 +
 2 files changed, 64 insertions(+)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index c7cb4f66a168..f04fb92dde2e 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -3377,6 +3377,67 @@ out_free:
 	return err;
 }
 
+static void mac80211_hwsim_abort_pmsr(struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif,
+				      struct cfg80211_pmsr_request *request)
+{
+	struct mac80211_hwsim_data *data;
+	struct sk_buff *skb = NULL;
+	struct nlattr *pmsr;
+	void *msg_head;
+	u32 _portid;
+	int err = 0;
+
+	data = hw->priv;
+	_portid = READ_ONCE(data->wmediumd);
+	if (!_portid && !hwsim_virtio_enabled)
+		return;
+
+	mutex_lock(&data->mutex);
+
+	if (data->pmsr_request != request) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!skb) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	msg_head = genlmsg_put(skb, 0, 0, &hwsim_genl_family, 0, HWSIM_CMD_ABORT_PMSR);
+
+	if (nla_put(skb, HWSIM_ATTR_ADDR_TRANSMITTER, ETH_ALEN, data->addresses[1].addr))
+		goto out;
+
+	pmsr = nla_nest_start(skb, HWSIM_ATTR_PMSR_REQUEST);
+	if (!pmsr) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = mac80211_hwsim_send_pmsr_request(skb, request);
+	if (err)
+		goto out;
+
+	err = nla_nest_end(skb, pmsr);
+	if (err)
+		goto out;
+
+	genlmsg_end(skb, msg_head);
+	if (hwsim_virtio_enabled)
+		hwsim_tx_virtio(data, skb);
+	else
+		hwsim_unicast_netgroup(data, skb, _portid);
+
+out:
+	if (err && skb)
+		nlmsg_free(skb);
+
+	mutex_unlock(&data->mutex);
+}
+
 #define HWSIM_COMMON_OPS					\
 	.tx = mac80211_hwsim_tx,				\
 	.wake_tx_queue = ieee80211_handle_wake_tx_queue,	\
@@ -3401,6 +3462,7 @@ out_free:
 	.get_et_stats = mac80211_hwsim_get_et_stats,		\
 	.get_et_strings = mac80211_hwsim_get_et_strings,	\
 	.start_pmsr = mac80211_hwsim_start_pmsr,		\
+	.abort_pmsr = mac80211_hwsim_abort_pmsr,
 
 #define HWSIM_NON_MLO_OPS					\
 	.sta_add = mac80211_hwsim_sta_add,			\
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
index 98e586a56582..383f3e39c911 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.h
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -83,6 +83,7 @@ enum hwsim_tx_control_flags {
  *	are the same as to @HWSIM_CMD_ADD_MAC_ADDR.
  * @HWSIM_CMD_START_PMSR: request to start peer measurement with the
  *	%HWSIM_ATTR_PMSR_REQUEST.
+ * @HWSIM_CMD_ABORT_PMSR: abort previously sent peer measurement
  * @__HWSIM_CMD_MAX: enum limit
  */
 enum {
@@ -96,6 +97,7 @@ enum {
 	HWSIM_CMD_ADD_MAC_ADDR,
 	HWSIM_CMD_DEL_MAC_ADDR,
 	HWSIM_CMD_START_PMSR,
+	HWSIM_CMD_ABORT_PMSR,
 	__HWSIM_CMD_MAX,
 };
 #define HWSIM_CMD_MAX (_HWSIM_CMD_MAX - 1)
-- 
cgit v1.2.3-70-g09d2


From 2af3b2a631b194a43551ce119cb71559d8f6b54b Mon Sep 17 00:00:00 2001
From: Jaewan Kim <jaewan@google.com>
Date: Wed, 22 Mar 2023 13:16:37 +0000
Subject: mac80211_hwsim: add PMSR report support via virtio

PMSR (a.k.a. peer measurement) is generalized measurement between two
devices with Wi-Fi support. And currently FTM (a.k.a. fine time measurement
or flight time measurement) is the one and only measurement.

Add the necessary functionality to allow mac80211_hwsim to report PMSR
result. The result would come from the wmediumd, where other Wi-Fi
devices' information are kept. mac80211_hwsim only need to deliver the
result to the userspace.

In detail, add new mac80211_hwsim attributes HWSIM_CMD_REPORT_PMSR, and
HWSIM_ATTR_PMSR_RESULT. When mac80211_hwsim receives the PMSR result with
command HWSIM_CMD_REPORT_PMSR and detail with attribute
HWSIM_ATTR_PMSR_RESULT, received data is parsed to cfg80211_pmsr_result and
resent to the userspace by cfg80211_pmsr_report().

To help receive the details of PMSR result, hwsim_rate_info_attributes is
added to receive rate_info without complex bitrate calculation. (i.e. send
rate_info without adding inverse of nl80211_put_sta_rate()).

Signed-off-by: Jaewan Kim <jaewan@google.com>
Link: https://lore.kernel.org/r/20230322131637.2633968-6-jaewan@google.com
[fix uninitialized return value when there are no reports]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 384 +++++++++++++++++++++++++-
 drivers/net/wireless/virtual/mac80211_hwsim.h |  51 +++-
 2 files changed, 425 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index f04fb92dde2e..f446d8f6e1f6 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -752,6 +752,11 @@ struct hwsim_radiotap_ack_hdr {
 	__le16 rt_chbitmask;
 } __packed;
 
+static struct mac80211_hwsim_data *get_hwsim_data_ref_from_addr(const u8 *addr)
+{
+	return rhashtable_lookup_fast(&hwsim_radios_rht, addr, hwsim_rht_params);
+}
+
 /* MAC80211_HWSIM netlink family */
 static struct genl_family hwsim_genl_family;
 
@@ -765,6 +770,76 @@ static const struct genl_multicast_group hwsim_mcgrps[] = {
 
 /* MAC80211_HWSIM netlink policy */
 
+static const struct nla_policy
+hwsim_rate_info_policy[HWSIM_RATE_INFO_ATTR_MAX + 1] = {
+	[HWSIM_RATE_INFO_ATTR_FLAGS] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_MCS] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_LEGACY] = { .type = NLA_U16 },
+	[HWSIM_RATE_INFO_ATTR_NSS] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_BW] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_HE_GI] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_HE_DCM] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_HE_RU_ALLOC] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_N_BOUNDED_CH] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_EHT_GI] = { .type = NLA_U8 },
+	[HWSIM_RATE_INFO_ATTR_EHT_RU_ALLOC] = { .type = NLA_U8 },
+};
+
+static const struct nla_policy
+hwsim_ftm_result_policy[NL80211_PMSR_FTM_RESP_ATTR_MAX + 1] = {
+	[NL80211_PMSR_FTM_RESP_ATTR_FAIL_REASON] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_RESP_ATTR_BURST_INDEX] = { .type = NLA_U16 },
+	[NL80211_PMSR_FTM_RESP_ATTR_NUM_FTMR_ATTEMPTS] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_RESP_ATTR_NUM_FTMR_SUCCESSES] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_RESP_ATTR_BUSY_RETRY_TIME] = { .type = NLA_U8 },
+	[NL80211_PMSR_FTM_RESP_ATTR_NUM_BURSTS_EXP] = { .type = NLA_U8 },
+	[NL80211_PMSR_FTM_RESP_ATTR_BURST_DURATION] = { .type = NLA_U8 },
+	[NL80211_PMSR_FTM_RESP_ATTR_FTMS_PER_BURST] = { .type = NLA_U8 },
+	[NL80211_PMSR_FTM_RESP_ATTR_RSSI_AVG] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_RESP_ATTR_RSSI_SPREAD] = { .type = NLA_U32 },
+	[NL80211_PMSR_FTM_RESP_ATTR_TX_RATE] = NLA_POLICY_NESTED(hwsim_rate_info_policy),
+	[NL80211_PMSR_FTM_RESP_ATTR_RX_RATE] = NLA_POLICY_NESTED(hwsim_rate_info_policy),
+	[NL80211_PMSR_FTM_RESP_ATTR_RTT_AVG] = { .type = NLA_U64 },
+	[NL80211_PMSR_FTM_RESP_ATTR_RTT_VARIANCE] = { .type = NLA_U64 },
+	[NL80211_PMSR_FTM_RESP_ATTR_RTT_SPREAD] = { .type = NLA_U64 },
+	[NL80211_PMSR_FTM_RESP_ATTR_DIST_AVG] = { .type = NLA_U64 },
+	[NL80211_PMSR_FTM_RESP_ATTR_DIST_VARIANCE] = { .type = NLA_U64 },
+	[NL80211_PMSR_FTM_RESP_ATTR_DIST_SPREAD] = { .type = NLA_U64 },
+	[NL80211_PMSR_FTM_RESP_ATTR_LCI] = { .type = NLA_STRING },
+	[NL80211_PMSR_FTM_RESP_ATTR_CIVICLOC] = { .type = NLA_STRING },
+};
+
+static const struct nla_policy
+hwsim_pmsr_resp_type_policy[NL80211_PMSR_TYPE_MAX + 1] = {
+	[NL80211_PMSR_TYPE_FTM] = NLA_POLICY_NESTED(hwsim_ftm_result_policy),
+};
+
+static const struct nla_policy
+hwsim_pmsr_resp_policy[NL80211_PMSR_RESP_ATTR_MAX + 1] = {
+	[NL80211_PMSR_RESP_ATTR_STATUS] = { .type = NLA_U32 },
+	[NL80211_PMSR_RESP_ATTR_HOST_TIME] = { .type = NLA_U64 },
+	[NL80211_PMSR_RESP_ATTR_AP_TSF] = { .type = NLA_U64 },
+	[NL80211_PMSR_RESP_ATTR_FINAL] = { .type = NLA_FLAG },
+	[NL80211_PMSR_RESP_ATTR_DATA] = NLA_POLICY_NESTED(hwsim_pmsr_resp_type_policy),
+};
+
+static const struct nla_policy
+hwsim_pmsr_peer_result_policy[NL80211_PMSR_PEER_ATTR_MAX + 1] = {
+	[NL80211_PMSR_PEER_ATTR_ADDR] = NLA_POLICY_ETH_ADDR_COMPAT,
+	[NL80211_PMSR_PEER_ATTR_CHAN] = { .type = NLA_REJECT },
+	[NL80211_PMSR_PEER_ATTR_REQ] = { .type = NLA_REJECT },
+	[NL80211_PMSR_PEER_ATTR_RESP] = NLA_POLICY_NESTED(hwsim_pmsr_resp_policy),
+};
+
+static const struct nla_policy
+hwsim_pmsr_peers_result_policy[NL80211_PMSR_ATTR_MAX + 1] = {
+	[NL80211_PMSR_ATTR_MAX_PEERS] = { .type = NLA_REJECT },
+	[NL80211_PMSR_ATTR_REPORT_AP_TSF] = { .type = NLA_REJECT },
+	[NL80211_PMSR_ATTR_RANDOMIZE_MAC_ADDR] = { .type = NLA_REJECT },
+	[NL80211_PMSR_ATTR_TYPE_CAPA] = { .type = NLA_REJECT },
+	[NL80211_PMSR_ATTR_PEERS] = NLA_POLICY_NESTED_ARRAY(hwsim_pmsr_peer_result_policy),
+};
+
 static const struct nla_policy
 hwsim_ftm_capa_policy[NL80211_PMSR_FTM_CAPA_ATTR_MAX + 1] = {
 	[NL80211_PMSR_FTM_CAPA_ATTR_ASAP] = { .type = NLA_FLAG },
@@ -822,6 +897,7 @@ static const struct nla_policy hwsim_genl_policy[HWSIM_ATTR_MAX + 1] = {
 	[HWSIM_ATTR_CIPHER_SUPPORT] = { .type = NLA_BINARY },
 	[HWSIM_ATTR_MLO_SUPPORT] = { .type = NLA_FLAG },
 	[HWSIM_ATTR_PMSR_SUPPORT] = NLA_POLICY_NESTED(hwsim_pmsr_capa_policy),
+	[HWSIM_ATTR_PMSR_RESULT] = NLA_POLICY_NESTED(hwsim_pmsr_peers_result_policy),
 };
 
 #if IS_REACHABLE(CONFIG_VIRTIO)
@@ -3438,6 +3514,297 @@ out:
 	mutex_unlock(&data->mutex);
 }
 
+static int mac80211_hwsim_parse_rate_info(struct nlattr *rateattr,
+					  struct rate_info *rate_info,
+					  struct genl_info *info)
+{
+	struct nlattr *tb[HWSIM_RATE_INFO_ATTR_MAX + 1];
+	int ret;
+
+	ret = nla_parse_nested(tb, HWSIM_RATE_INFO_ATTR_MAX,
+			       rateattr, hwsim_rate_info_policy, info->extack);
+	if (ret)
+		return ret;
+
+	if (tb[HWSIM_RATE_INFO_ATTR_FLAGS])
+		rate_info->flags = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_FLAGS]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_MCS])
+		rate_info->mcs = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_MCS]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_LEGACY])
+		rate_info->legacy = nla_get_u16(tb[HWSIM_RATE_INFO_ATTR_LEGACY]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_NSS])
+		rate_info->nss = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_NSS]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_BW])
+		rate_info->bw = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_BW]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_HE_GI])
+		rate_info->he_gi = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_HE_GI]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_HE_DCM])
+		rate_info->he_dcm = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_HE_DCM]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_HE_RU_ALLOC])
+		rate_info->he_ru_alloc =
+			nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_HE_RU_ALLOC]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_N_BOUNDED_CH])
+		rate_info->n_bonded_ch = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_N_BOUNDED_CH]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_EHT_GI])
+		rate_info->eht_gi = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_EHT_GI]);
+
+	if (tb[HWSIM_RATE_INFO_ATTR_EHT_RU_ALLOC])
+		rate_info->eht_ru_alloc = nla_get_u8(tb[HWSIM_RATE_INFO_ATTR_EHT_RU_ALLOC]);
+
+	return 0;
+}
+
+static int mac80211_hwsim_parse_ftm_result(struct nlattr *ftm,
+					   struct cfg80211_pmsr_ftm_result *result,
+					   struct genl_info *info)
+{
+	struct nlattr *tb[NL80211_PMSR_FTM_RESP_ATTR_MAX + 1];
+	int ret;
+
+	ret = nla_parse_nested(tb, NL80211_PMSR_FTM_RESP_ATTR_MAX,
+			       ftm, hwsim_ftm_result_policy, info->extack);
+	if (ret)
+		return ret;
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_FAIL_REASON])
+		result->failure_reason = nla_get_u32(tb[NL80211_PMSR_FTM_RESP_ATTR_FAIL_REASON]);
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_BURST_INDEX])
+		result->burst_index = nla_get_u16(tb[NL80211_PMSR_FTM_RESP_ATTR_BURST_INDEX]);
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_NUM_FTMR_ATTEMPTS]) {
+		result->num_ftmr_attempts_valid = 1;
+		result->num_ftmr_attempts =
+			nla_get_u32(tb[NL80211_PMSR_FTM_RESP_ATTR_NUM_FTMR_ATTEMPTS]);
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_NUM_FTMR_SUCCESSES]) {
+		result->num_ftmr_successes_valid = 1;
+		result->num_ftmr_successes =
+			nla_get_u32(tb[NL80211_PMSR_FTM_RESP_ATTR_NUM_FTMR_SUCCESSES]);
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_BUSY_RETRY_TIME])
+		result->busy_retry_time =
+			nla_get_u8(tb[NL80211_PMSR_FTM_RESP_ATTR_BUSY_RETRY_TIME]);
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_NUM_BURSTS_EXP])
+		result->num_bursts_exp = nla_get_u8(tb[NL80211_PMSR_FTM_RESP_ATTR_NUM_BURSTS_EXP]);
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_BURST_DURATION])
+		result->burst_duration = nla_get_u8(tb[NL80211_PMSR_FTM_RESP_ATTR_BURST_DURATION]);
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_FTMS_PER_BURST])
+		result->ftms_per_burst = nla_get_u8(tb[NL80211_PMSR_FTM_RESP_ATTR_FTMS_PER_BURST]);
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_RSSI_AVG]) {
+		result->rssi_avg_valid = 1;
+		result->rssi_avg = nla_get_s32(tb[NL80211_PMSR_FTM_RESP_ATTR_RSSI_AVG]);
+	}
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_RSSI_SPREAD]) {
+		result->rssi_spread_valid = 1;
+		result->rssi_spread =
+			nla_get_s32(tb[NL80211_PMSR_FTM_RESP_ATTR_RSSI_SPREAD]);
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_TX_RATE]) {
+		result->tx_rate_valid = 1;
+		ret = mac80211_hwsim_parse_rate_info(tb[NL80211_PMSR_FTM_RESP_ATTR_TX_RATE],
+						     &result->tx_rate, info);
+		if (ret)
+			return ret;
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_RX_RATE]) {
+		result->rx_rate_valid = 1;
+		ret = mac80211_hwsim_parse_rate_info(tb[NL80211_PMSR_FTM_RESP_ATTR_RX_RATE],
+						     &result->rx_rate, info);
+		if (ret)
+			return ret;
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_RTT_AVG]) {
+		result->rtt_avg_valid = 1;
+		result->rtt_avg =
+			nla_get_u64(tb[NL80211_PMSR_FTM_RESP_ATTR_RTT_AVG]);
+	}
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_RTT_VARIANCE]) {
+		result->rtt_variance_valid = 1;
+		result->rtt_variance =
+			nla_get_u64(tb[NL80211_PMSR_FTM_RESP_ATTR_RTT_VARIANCE]);
+	}
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_RTT_SPREAD]) {
+		result->rtt_spread_valid = 1;
+		result->rtt_spread =
+			nla_get_u64(tb[NL80211_PMSR_FTM_RESP_ATTR_RTT_SPREAD]);
+	}
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_DIST_AVG]) {
+		result->dist_avg_valid = 1;
+		result->dist_avg =
+			nla_get_u64(tb[NL80211_PMSR_FTM_RESP_ATTR_DIST_AVG]);
+	}
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_DIST_VARIANCE]) {
+		result->dist_variance_valid = 1;
+		result->dist_variance =
+			nla_get_u64(tb[NL80211_PMSR_FTM_RESP_ATTR_DIST_VARIANCE]);
+	}
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_DIST_SPREAD]) {
+		result->dist_spread_valid = 1;
+		result->dist_spread =
+			nla_get_u64(tb[NL80211_PMSR_FTM_RESP_ATTR_DIST_SPREAD]);
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_LCI]) {
+		result->lci = nla_data(tb[NL80211_PMSR_FTM_RESP_ATTR_LCI]);
+		result->lci_len = nla_len(tb[NL80211_PMSR_FTM_RESP_ATTR_LCI]);
+	}
+
+	if (tb[NL80211_PMSR_FTM_RESP_ATTR_CIVICLOC]) {
+		result->civicloc = nla_data(tb[NL80211_PMSR_FTM_RESP_ATTR_CIVICLOC]);
+		result->civicloc_len = nla_len(tb[NL80211_PMSR_FTM_RESP_ATTR_CIVICLOC]);
+	}
+
+	return 0;
+}
+
+static int mac80211_hwsim_parse_pmsr_resp(struct nlattr *resp,
+					  struct cfg80211_pmsr_result *result,
+					  struct genl_info *info)
+{
+	struct nlattr *tb[NL80211_PMSR_RESP_ATTR_MAX + 1];
+	struct nlattr *pmsr;
+	int rem;
+	int ret;
+
+	ret = nla_parse_nested(tb, NL80211_PMSR_RESP_ATTR_MAX, resp, hwsim_pmsr_resp_policy,
+			       info->extack);
+	if (ret)
+		return ret;
+
+	if (tb[NL80211_PMSR_RESP_ATTR_STATUS])
+		result->status = nla_get_u32(tb[NL80211_PMSR_RESP_ATTR_STATUS]);
+
+	if (tb[NL80211_PMSR_RESP_ATTR_HOST_TIME])
+		result->host_time = nla_get_u64(tb[NL80211_PMSR_RESP_ATTR_HOST_TIME]);
+
+	if (tb[NL80211_PMSR_RESP_ATTR_AP_TSF]) {
+		result->ap_tsf_valid = 1;
+		result->ap_tsf = nla_get_u64(tb[NL80211_PMSR_RESP_ATTR_AP_TSF]);
+	}
+
+	result->final = !!tb[NL80211_PMSR_RESP_ATTR_FINAL];
+
+	if (!tb[NL80211_PMSR_RESP_ATTR_DATA])
+		return 0;
+
+	nla_for_each_nested(pmsr, tb[NL80211_PMSR_RESP_ATTR_DATA], rem) {
+		switch (nla_type(pmsr)) {
+		case NL80211_PMSR_TYPE_FTM:
+			result->type = NL80211_PMSR_TYPE_FTM;
+			ret = mac80211_hwsim_parse_ftm_result(pmsr, &result->ftm, info);
+			if (ret)
+				return ret;
+			break;
+		default:
+			NL_SET_ERR_MSG_ATTR(info->extack, pmsr, "Unknown pmsr resp type");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int mac80211_hwsim_parse_pmsr_result(struct nlattr *peer,
+					    struct cfg80211_pmsr_result *result,
+					    struct genl_info *info)
+{
+	struct nlattr *tb[NL80211_PMSR_PEER_ATTR_MAX + 1];
+	int ret;
+
+	if (!peer)
+		return -EINVAL;
+
+	ret = nla_parse_nested(tb, NL80211_PMSR_PEER_ATTR_MAX, peer,
+			       hwsim_pmsr_peer_result_policy, info->extack);
+	if (ret)
+		return ret;
+
+	if (tb[NL80211_PMSR_PEER_ATTR_ADDR])
+		memcpy(result->addr, nla_data(tb[NL80211_PMSR_PEER_ATTR_ADDR]),
+		       ETH_ALEN);
+
+	if (tb[NL80211_PMSR_PEER_ATTR_RESP]) {
+		ret = mac80211_hwsim_parse_pmsr_resp(tb[NL80211_PMSR_PEER_ATTR_RESP], result, info);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+};
+
+static int hwsim_pmsr_report_nl(struct sk_buff *msg, struct genl_info *info)
+{
+	struct mac80211_hwsim_data *data;
+	struct nlattr *peers, *peer;
+	struct nlattr *reqattr;
+	const u8 *src;
+	int err;
+	int rem;
+
+	src = nla_data(info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER]);
+	data = get_hwsim_data_ref_from_addr(src);
+	if (!data)
+		return -EINVAL;
+
+	mutex_lock(&data->mutex);
+	if (!data->pmsr_request) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	reqattr = info->attrs[HWSIM_ATTR_PMSR_RESULT];
+	if (!reqattr) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	peers = nla_find_nested(reqattr, NL80211_PMSR_ATTR_PEERS);
+	if (!peers) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	nla_for_each_nested(peer, peers, rem) {
+		struct cfg80211_pmsr_result result;
+
+		err = mac80211_hwsim_parse_pmsr_result(peer, &result, info);
+		if (err)
+			goto out;
+
+		cfg80211_pmsr_report(data->pmsr_request_wdev,
+				     data->pmsr_request, &result, GFP_KERNEL);
+	}
+
+	cfg80211_pmsr_complete(data->pmsr_request_wdev, data->pmsr_request, GFP_KERNEL);
+
+	err = 0;
+out:
+	data->pmsr_request = NULL;
+	data->pmsr_request_wdev = NULL;
+
+	mutex_unlock(&data->mutex);
+	return err;
+}
+
 #define HWSIM_COMMON_OPS					\
 	.tx = mac80211_hwsim_tx,				\
 	.wake_tx_queue = ieee80211_handle_wake_tx_queue,	\
@@ -5113,13 +5480,6 @@ static void hwsim_mon_setup(struct net_device *dev)
 	eth_hw_addr_set(dev, addr);
 }
 
-static struct mac80211_hwsim_data *get_hwsim_data_ref_from_addr(const u8 *addr)
-{
-	return rhashtable_lookup_fast(&hwsim_radios_rht,
-				      addr,
-				      hwsim_rht_params);
-}
-
 static void hwsim_register_wmediumd(struct net *net, u32 portid)
 {
 	struct mac80211_hwsim_data *data;
@@ -5790,6 +6150,11 @@ static const struct genl_small_ops hwsim_ops[] = {
 		.doit = hwsim_get_radio_nl,
 		.dumpit = hwsim_dump_radio_nl,
 	},
+	{
+		.cmd = HWSIM_CMD_REPORT_PMSR,
+		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
+		.doit = hwsim_pmsr_report_nl,
+	},
 };
 
 static struct genl_family hwsim_genl_family __ro_after_init = {
@@ -5801,7 +6166,7 @@ static struct genl_family hwsim_genl_family __ro_after_init = {
 	.module = THIS_MODULE,
 	.small_ops = hwsim_ops,
 	.n_small_ops = ARRAY_SIZE(hwsim_ops),
-	.resv_start_op = HWSIM_CMD_DEL_MAC_ADDR + 1,
+	.resv_start_op = HWSIM_CMD_REPORT_PMSR + 1, // match with __HWSIM_CMD_MAX
 	.mcgrps = hwsim_mcgrps,
 	.n_mcgrps = ARRAY_SIZE(hwsim_mcgrps),
 };
@@ -5970,6 +6335,9 @@ static int hwsim_virtio_handle_cmd(struct sk_buff *skb)
 	case HWSIM_CMD_TX_INFO_FRAME:
 		hwsim_tx_info_frame_received_nl(skb, &info);
 		break;
+	case HWSIM_CMD_REPORT_PMSR:
+		hwsim_pmsr_report_nl(skb, &info);
+		break;
 	default:
 		pr_err_ratelimited("hwsim: invalid cmd: %d\n", gnlh->cmd);
 		return -EPROTO;
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
index 383f3e39c911..92126f02c58f 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.h
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -82,8 +82,8 @@ enum hwsim_tx_control_flags {
  * @HWSIM_CMD_DEL_MAC_ADDR: remove the MAC address again, the attributes
  *	are the same as to @HWSIM_CMD_ADD_MAC_ADDR.
  * @HWSIM_CMD_START_PMSR: request to start peer measurement with the
- *	%HWSIM_ATTR_PMSR_REQUEST.
- * @HWSIM_CMD_ABORT_PMSR: abort previously sent peer measurement
+ *	%HWSIM_ATTR_PMSR_REQUEST. Result will be sent back asynchronously
+ *	with %HWSIM_CMD_REPORT_PMSR.
  * @__HWSIM_CMD_MAX: enum limit
  */
 enum {
@@ -98,6 +98,7 @@ enum {
 	HWSIM_CMD_DEL_MAC_ADDR,
 	HWSIM_CMD_START_PMSR,
 	HWSIM_CMD_ABORT_PMSR,
+	HWSIM_CMD_REPORT_PMSR,
 	__HWSIM_CMD_MAX,
 };
 #define HWSIM_CMD_MAX (_HWSIM_CMD_MAX - 1)
@@ -151,6 +152,8 @@ enum {
  *	to provide peer measurement capabilities. (nl80211_peer_measurement_attrs)
  * @HWSIM_ATTR_PMSR_REQUEST: nested attribute used with %HWSIM_CMD_START_PMSR
  *	to provide details about peer measurement request (nl80211_peer_measurement_attrs)
+ * @HWSIM_ATTR_PMSR_RESULT: nested attributed used with %HWSIM_CMD_REPORT_PMSR
+ *	to provide peer measurement result (nl80211_peer_measurement_attrs)
  * @__HWSIM_ATTR_MAX: enum limit
  */
 
@@ -184,6 +187,7 @@ enum {
 	HWSIM_ATTR_MLO_SUPPORT,
 	HWSIM_ATTR_PMSR_SUPPORT,
 	HWSIM_ATTR_PMSR_REQUEST,
+	HWSIM_ATTR_PMSR_RESULT,
 	__HWSIM_ATTR_MAX,
 };
 #define HWSIM_ATTR_MAX (__HWSIM_ATTR_MAX - 1)
@@ -288,4 +292,47 @@ enum {
 	HWSIM_VQ_RX,
 	HWSIM_NUM_VQS,
 };
+
+/**
+ * enum hwsim_rate_info -- bitrate information.
+ *
+ * Information about a receiving or transmitting bitrate
+ * that can be mapped to struct rate_info
+ *
+ * @HWSIM_RATE_INFO_ATTR_FLAGS: bitflag of flags from &enum rate_info_flags
+ * @HWSIM_RATE_INFO_ATTR_MCS: mcs index if struct describes an HT/VHT/HE rate
+ * @HWSIM_RATE_INFO_ATTR_LEGACY: bitrate in 100kbit/s for 802.11abg
+ * @HWSIM_RATE_INFO_ATTR_NSS: number of streams (VHT & HE only)
+ * @HWSIM_RATE_INFO_ATTR_BW: bandwidth (from &enum rate_info_bw)
+ * @HWSIM_RATE_INFO_ATTR_HE_GI: HE guard interval (from &enum nl80211_he_gi)
+ * @HWSIM_RATE_INFO_ATTR_HE_DCM: HE DCM value
+ * @HWSIM_RATE_INFO_ATTR_HE_RU_ALLOC:  HE RU allocation (from &enum nl80211_he_ru_alloc,
+ *	only valid if bw is %RATE_INFO_BW_HE_RU)
+ * @HWSIM_RATE_INFO_ATTR_N_BOUNDED_CH: In case of EDMG the number of bonded channels (1-4)
+ * @HWSIM_RATE_INFO_ATTR_EHT_GI: EHT guard interval (from &enum nl80211_eht_gi)
+ * @HWSIM_RATE_INFO_ATTR_EHT_RU_ALLOC: EHT RU allocation (from &enum nl80211_eht_ru_alloc,
+ *	only valid if bw is %RATE_INFO_BW_EHT_RU)
+ * @NUM_HWSIM_RATE_INFO_ATTRS: internal
+ * @HWSIM_RATE_INFO_ATTR_MAX: highest attribute number
+ */
+enum hwsim_rate_info_attributes {
+	__HWSIM_RATE_INFO_ATTR_INVALID,
+
+	HWSIM_RATE_INFO_ATTR_FLAGS,
+	HWSIM_RATE_INFO_ATTR_MCS,
+	HWSIM_RATE_INFO_ATTR_LEGACY,
+	HWSIM_RATE_INFO_ATTR_NSS,
+	HWSIM_RATE_INFO_ATTR_BW,
+	HWSIM_RATE_INFO_ATTR_HE_GI,
+	HWSIM_RATE_INFO_ATTR_HE_DCM,
+	HWSIM_RATE_INFO_ATTR_HE_RU_ALLOC,
+	HWSIM_RATE_INFO_ATTR_N_BOUNDED_CH,
+	HWSIM_RATE_INFO_ATTR_EHT_GI,
+	HWSIM_RATE_INFO_ATTR_EHT_RU_ALLOC,
+
+	/* keep last */
+	NUM_HWSIM_RATE_INFO_ATTRS,
+	HWSIM_RATE_INFO_ATTR_MAX = NUM_HWSIM_RATE_INFO_ATTRS - 1
+};
+
 #endif /* __MAC80211_HWSIM_H */
-- 
cgit v1.2.3-70-g09d2


From 4b50d4205207ab706b3ecd0d122c014845b8e1f0 Mon Sep 17 00:00:00 2001
From: P Praneesh <quic_ppranees@quicinc.com>
Date: Fri, 17 Mar 2023 21:50:56 +0530
Subject: wifi: ath12k: fill peer meta data during reo_reinject

Firmware expects physical address in the reo entrance ring when MLO is
enabled and peer meta data in the queue_addr_lo for the non MLO cases.
Current implementation fills the physical address for the non MLO cases
and firmware decodes it differently which leads to SOC CSR region
access which is not related to firmware/MAC operation.

Fix this by sending only peer_meta data and destination indication bit
during non MLO case. This prevents firmware from invalid decoding.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Signed-off-by: Manish Dharanenthiran <quic_mdharane@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230317162056.20353-1-quic_mdharane@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 0adcbcfa0db5..b368494071ea 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -3047,10 +3047,14 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 	reo_ent_ring->rx_mpdu_info.peer_meta_data =
 		reo_dest_ring->rx_mpdu_info.peer_meta_data;
 
-	reo_ent_ring->queue_addr_lo = cpu_to_le32(lower_32_bits(rx_tid->paddr));
-	reo_ent_ring->info0 = le32_encode_bits(upper_32_bits(rx_tid->paddr),
-					       HAL_REO_ENTR_RING_INFO0_QUEUE_ADDR_HI) |
-		le32_encode_bits(dst_ind, HAL_REO_ENTR_RING_INFO0_DEST_IND);
+	/* Firmware expects physical address to be filled in queue_addr_lo in
+	 * the MLO scenario and in case of non MLO peer meta data needs to be
+	 * filled.
+	 * TODO: Need to handle for MLO scenario.
+	 */
+	reo_ent_ring->queue_addr_lo = reo_dest_ring->rx_mpdu_info.peer_meta_data;
+	reo_ent_ring->info0 = le32_encode_bits(dst_ind,
+					       HAL_REO_ENTR_RING_INFO0_DEST_IND);
 
 	reo_ent_ring->info1 = le32_encode_bits(rx_tid->cur_sn,
 					       HAL_REO_ENTR_RING_INFO1_MPDU_SEQ_NUM);
-- 
cgit v1.2.3-70-g09d2


From e93bbd65547ea8073b707c9034c3f051f8018614 Mon Sep 17 00:00:00 2001
From: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Date: Mon, 20 Mar 2023 10:54:27 +0530
Subject: wifi: ath12k: fix packets are sent in native wifi mode while we set
 raw mode

Currently, if we set raw mode as encap type, packets are sent
in native wifi mode which is default mode. This is caused by
no check being present for raw mode in ath12k_dp_tx_get_encap_type
function, so it's always going in native wifi mode.

Fix, this issue by adding check condition for raw mode.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230320052427.12271-1-quic_aarasahu@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_tx.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index fd8d850f9818..d3c7c76d6b75 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -13,6 +13,10 @@ static enum hal_tcl_encap_type
 ath12k_dp_tx_get_encap_type(struct ath12k_vif *arvif, struct sk_buff *skb)
 {
 	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
+	struct ath12k_base *ab = arvif->ar->ab;
+
+	if (test_bit(ATH12K_FLAG_RAW_MODE, &ab->dev_flags))
+		return HAL_TCL_ENCAP_TYPE_RAW;
 
 	if (tx_info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP)
 		return HAL_TCL_ENCAP_TYPE_ETHERNET;
-- 
cgit v1.2.3-70-g09d2


From 392b6b76fd3940ba9978ecd8fe30d9ed979d3bf7 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Mon, 20 Mar 2023 14:34:00 +0530
Subject: wifi: ath12k: fix incorrect handling of AMSDU frames

When a data traffic with size greater than 1500 is initiated,
say size 65500, it is sent as a set of AMSDUs (3 MSDUs)
in a MPDU.

These AMSDUs are incorrectly classified as multicast
and peer id gets changed, except for first msdu,
because of mcast_bcast bit derived from rx_desc
for every msdu received.

The mcast_bcast bit in rx_desc is only valid for
first msdu. Hence, add changes to derive mcast_bcast
only if it is a first msdu.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230320090400.30027-1-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index b368494071ea..e78478a5b978 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -196,7 +196,8 @@ static void ath12k_dp_rxdesc_set_msdu_len(struct ath12k_base *ab,
 static bool ath12k_dp_rx_h_is_mcbc(struct ath12k_base *ab,
 				   struct hal_rx_desc *desc)
 {
-	return ab->hw_params->hal_ops->rx_desc_is_mcbc(desc);
+	return (ath12k_dp_rx_h_first_msdu(ab, desc) &&
+		ab->hw_params->hal_ops->rx_desc_is_mcbc(desc));
 }
 
 static bool ath12k_dp_rxdesc_mac_addr2_valid(struct ath12k_base *ab,
-- 
cgit v1.2.3-70-g09d2


From 25d165c498025f401eed6503f03115c1fb896c47 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Mon, 20 Mar 2023 14:42:08 +0530
Subject: wifi: ath12k: incorrect channel survey dump

When underlying hardware supports more than one band, then wiphy
is aware of all the frequency bands present. Example, say
wiphy0 is aware of 2 GHz, 5 GHz and 6 GHz band information.
Hence, when an virtual AP interface is brought up in auto channel
selection on each band. Survey is initiated from the userspace application
for all the virtual AP interfaces simultaneously.

For survey dump collection, freq_to_idx() api is used to map the index of
particular frequency to populate its survey. When the dump was collected,
the index fetched was incorrect. Example, in case of multi-hardware,
the index of channel 36 was reported as 14, because wiphy0 had index 0-13
for 2 GHz channels, instead of 0 for that ar which resulted in invalid
survey sent to userspace application.

The incorrect index was because of the mapping did not consider whether
the freq band is present in its radio structure (ar) or not.

To handle this, the freq_to_idx() api should map index for the
particular band in the ar structure.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230320091208.25535-1-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wmi.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index af910296c41e..7ae0bb78b2b5 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -4937,6 +4937,9 @@ static int freq_to_idx(struct ath12k *ar, int freq)
 	int band, ch, idx = 0;
 
 	for (band = NL80211_BAND_2GHZ; band < NUM_NL80211_BANDS; band++) {
+		if (!ar->mac.sbands[band].channels)
+			continue;
+
 		sband = ar->hw->wiphy->bands[band];
 		if (!sband)
 			continue;
-- 
cgit v1.2.3-70-g09d2


From b6b88111c0dbcef04cd0e0bafd646e4c09728302 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan+linaro@kernel.org>
Date: Tue, 21 Mar 2023 10:40:10 +0100
Subject: dt-bindings: net: wireless: add ath11k pcie bindings

Add devicetree bindings for Qualcomm ath11k PCIe devices such as WCN6855
for which the calibration data variant may need to be described.

Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230321094011.9759-2-johan+linaro@kernel.org
---
 .../bindings/net/wireless/qcom,ath11k-pci.yaml     | 58 ++++++++++++++++++++++
 1 file changed, 58 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml
new file mode 100644
index 000000000000..817f02a8b481
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml
@@ -0,0 +1,58 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+# Copyright (c) 2023 Linaro Limited
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/wireless/qcom,ath11k-pci.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Technologies ath11k wireless devices (PCIe)
+
+maintainers:
+  - Kalle Valo <kvalo@kernel.org>
+
+description: |
+  Qualcomm Technologies IEEE 802.11ax PCIe devices
+
+properties:
+  compatible:
+    enum:
+      - pci17cb,1103  # WCN6855
+
+  reg:
+    maxItems: 1
+
+  qcom,ath11k-calibration-variant:
+    $ref: /schemas/types.yaml#/definitions/string
+    description: |
+      string to uniquely identify variant of the calibration data for designs
+      with colliding bus and device ids
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    pcie {
+        #address-cells = <3>;
+        #size-cells = <2>;
+
+        pcie@0 {
+            device_type = "pci";
+            reg = <0x0 0x0 0x0 0x0 0x0>;
+            #address-cells = <3>;
+            #size-cells = <2>;
+            ranges;
+
+            bus-range = <0x01 0xff>;
+
+            wifi@0 {
+                compatible = "pci17cb,1103";
+                reg = <0x10000 0x0 0x0 0x0 0x0>;
+
+                qcom,ath11k-calibration-variant = "LE_X13S";
+            };
+        };
+    };
-- 
cgit v1.2.3-70-g09d2


From 9fc093b756f64788ce33e3484f6f59a169704759 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Wed, 22 Mar 2023 08:28:55 -0400
Subject: wifi: ath10k: remove unused ath10k_get_ring_byte function

clang with W=1 reports:

drivers/net/wireless/ath/ath10k/ce.c:88:1: error:
  unused function 'ath10k_get_ring_byte' [-Werror,-Wunused-function]
ath10k_get_ring_byte(unsigned int offset,
^
This function is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230322122855.2570417-1-trix@redhat.com
---
 drivers/net/wireless/ath/ath10k/ce.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index b656cfc03648..c27b8204718a 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -84,13 +84,6 @@ ath10k_set_ring_byte(unsigned int offset,
 	return ((offset << addr_map->lsb) & addr_map->mask);
 }
 
-static inline unsigned int
-ath10k_get_ring_byte(unsigned int offset,
-		     struct ath10k_hw_ce_regs_addr_map *addr_map)
-{
-	return ((offset & addr_map->mask) >> (addr_map->lsb));
-}
-
 static inline u32 ath10k_ce_read32(struct ath10k *ar, u32 offset)
 {
 	struct ath10k_ce *ce = ath10k_ce_priv(ar);
-- 
cgit v1.2.3-70-g09d2


From 72383ed78c1c01d36612ca5c6fdd2f35fd859bbd Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Fri, 17 Mar 2023 11:30:47 -0600
Subject: wifi: carl9170: Fix multiple -Warray-bounds warnings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

GCC (and Clang)[1] does not like having a partially allocated object,
since it cannot reason about it for bounds checking. Instead, fully
allocate struct carl9170_cmd.

Fix the following warnings Seen under GCC 13:
drivers/net/wireless/ath/carl9170/cmd.c:125:30: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[4]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:126:30: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[4]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:125:30: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[20]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:126:30: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[20]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:161:12: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[20]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:162:12: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[20]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:163:12: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[20]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:164:12: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[20]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:125:30: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[8]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:126:30: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[8]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/cmd.c:220:12: warning: array subscript ‘struct carl9170_cmd[0]’ is partly outside array bounds of ‘unsigned char[8]’ [-Warray-bounds=]

Link: https://github.com/KSPP/linux/issues/268
Link: godbolt.org/z/KP97sxh3T [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Acked-by: Christian Lamparter <chunkeey@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/ZBSjx236+BTiRByf@work
---
 drivers/net/wireless/ath/carl9170/cmd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/carl9170/cmd.c b/drivers/net/wireless/ath/carl9170/cmd.c
index f2b4f537e4c1..b8ed193c0195 100644
--- a/drivers/net/wireless/ath/carl9170/cmd.c
+++ b/drivers/net/wireless/ath/carl9170/cmd.c
@@ -120,7 +120,7 @@ struct carl9170_cmd *carl9170_cmd_buf(struct ar9170 *ar,
 {
 	struct carl9170_cmd *tmp;
 
-	tmp = kzalloc(sizeof(struct carl9170_cmd_head) + len, GFP_ATOMIC);
+	tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC);
 	if (tmp) {
 		tmp->hdr.cmd = cmd;
 		tmp->hdr.len = len;
-- 
cgit v1.2.3-70-g09d2


From 1be3640cbb4a52ac9a2653ff20d68f983366092c Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Fri, 17 Mar 2023 11:39:36 -0600
Subject: wifi: carl9170: Replace fake flex-array with flexible-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Address the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
drivers/net/wireless/ath/carl9170/tx.c:702:61: warning: array subscript i is outside array bounds of ‘const struct _carl9170_tx_status[0]’ [-Warray-bounds=]
drivers/net/wireless/ath/carl9170/tx.c:701:65: warning: array subscript i is outside array bounds of ‘const struct _carl9170_tx_status[0]’ [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/267
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Acked-by: Christian Lamparter <chunkeey@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/ZBSl2M+aGIO1fnuG@work
---
 drivers/net/wireless/ath/carl9170/fwcmd.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/carl9170/fwcmd.h b/drivers/net/wireless/ath/carl9170/fwcmd.h
index ff4b3b50250c..e5bcc364f088 100644
--- a/drivers/net/wireless/ath/carl9170/fwcmd.h
+++ b/drivers/net/wireless/ath/carl9170/fwcmd.h
@@ -320,9 +320,9 @@ struct carl9170_rsp {
 		struct carl9170_u32_list	rreg_res;
 		struct carl9170_u32_list	echo;
 #ifdef __CARL9170FW__
-		struct carl9170_tx_status	tx_status[0];
+		DECLARE_FLEX_ARRAY(struct carl9170_tx_status, tx_status);
 #endif /* __CARL9170FW__ */
-		struct _carl9170_tx_status	_tx_status[0];
+		DECLARE_FLEX_ARRAY(struct _carl9170_tx_status, _tx_status);
 		struct carl9170_gpio		gpio;
 		struct carl9170_tsf_rsp		tsf;
 		struct carl9170_psm		psm;
-- 
cgit v1.2.3-70-g09d2


From a0340df7eca4f28e18afd3ecd6e464db5f2994c0 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 21 Mar 2023 17:15:00 +0100
Subject: can: rcar_canfd: Add transceiver support

Add support for CAN transceivers described as PHYs.

While simple CAN transceivers can do without, this is needed for CAN
transceivers like NXP TJR1443 that need a configuration step (like
pulling standby or enable lines), and/or impose a bitrate limit.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/1ce907572ac1d4e1733fa6ea7712250f2229cfcb.1679414936.git.geert+renesas@glider.be
[mkl: squash error message update from patch 2]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/rcar/rcar_canfd.c | 30 +++++++++++++++++++++++++-----
 1 file changed, 25 insertions(+), 5 deletions(-)

diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index ef4e1b9a9e1e..d8fbc3fca475 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -35,6 +35,7 @@
 #include <linux/netdevice.h>
 #include <linux/of.h>
 #include <linux/of_device.h>
+#include <linux/phy/phy.h>
 #include <linux/platform_device.h>
 #include <linux/reset.h>
 #include <linux/types.h>
@@ -530,6 +531,7 @@ struct rcar_canfd_channel {
 	struct net_device *ndev;
 	struct rcar_canfd_global *gpriv;	/* Controller reference */
 	void __iomem *base;			/* Register base address */
+	struct phy *transceiver;		/* Optional transceiver */
 	struct napi_struct napi;
 	u32 tx_head;				/* Incremented on xmit */
 	u32 tx_tail;				/* Incremented on xmit done */
@@ -1413,11 +1415,17 @@ static int rcar_canfd_open(struct net_device *ndev)
 	struct rcar_canfd_global *gpriv = priv->gpriv;
 	int err;
 
+	err = phy_power_on(priv->transceiver);
+	if (err) {
+		netdev_err(ndev, "failed to power on PHY: %pe\n", ERR_PTR(err));
+		return err;
+	}
+
 	/* Peripheral clock is already enabled in probe */
 	err = clk_prepare_enable(gpriv->can_clk);
 	if (err) {
 		netdev_err(ndev, "failed to enable CAN clock, error %d\n", err);
-		goto out_clock;
+		goto out_phy;
 	}
 
 	err = open_candev(ndev);
@@ -1437,7 +1445,8 @@ out_close:
 	close_candev(ndev);
 out_can_clock:
 	clk_disable_unprepare(gpriv->can_clk);
-out_clock:
+out_phy:
+	phy_power_off(priv->transceiver);
 	return err;
 }
 
@@ -1480,6 +1489,7 @@ static int rcar_canfd_close(struct net_device *ndev)
 	napi_disable(&priv->napi);
 	clk_disable_unprepare(gpriv->can_clk);
 	close_candev(ndev);
+	phy_power_off(priv->transceiver);
 	return 0;
 }
 
@@ -1711,7 +1721,7 @@ static const struct ethtool_ops rcar_canfd_ethtool_ops = {
 };
 
 static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
-				    u32 fcan_freq)
+				    u32 fcan_freq, struct phy *transceiver)
 {
 	const struct rcar_canfd_hw_info *info = gpriv->info;
 	struct platform_device *pdev = gpriv->pdev;
@@ -1732,8 +1742,11 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 	ndev->flags |= IFF_ECHO;
 	priv->ndev = ndev;
 	priv->base = gpriv->base;
+	priv->transceiver = transceiver;
 	priv->channel = ch;
 	priv->gpriv = gpriv;
+	if (transceiver)
+		priv->can.bitrate_max = transceiver->attrs.max_link_rate;
 	priv->can.clock.freq = fcan_freq;
 	dev_info(dev, "can_clk rate is %u\n", priv->can.clock.freq);
 
@@ -1836,6 +1849,7 @@ static void rcar_canfd_channel_remove(struct rcar_canfd_global *gpriv, u32 ch)
 
 static int rcar_canfd_probe(struct platform_device *pdev)
 {
+	struct phy *transceivers[RCANFD_NUM_CHANNELS] = { 0, };
 	const struct rcar_canfd_hw_info *info;
 	struct device *dev = &pdev->dev;
 	void __iomem *addr;
@@ -1857,9 +1871,14 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 	for (i = 0; i < info->max_channels; ++i) {
 		name[7] = '0' + i;
 		of_child = of_get_child_by_name(dev->of_node, name);
-		if (of_child && of_device_is_available(of_child))
+		if (of_child && of_device_is_available(of_child)) {
 			channels_mask |= BIT(i);
+			transceivers[i] = devm_of_phy_optional_get(dev,
+							of_child, NULL);
+		}
 		of_node_put(of_child);
+		if (IS_ERR(transceivers[i]))
+			return PTR_ERR(transceivers[i]);
 	}
 
 	if (info->shared_global_irqs) {
@@ -2035,7 +2054,8 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 	}
 
 	for_each_set_bit(ch, &gpriv->channels_mask, info->max_channels) {
-		err = rcar_canfd_channel_probe(gpriv, ch, fcan_freq);
+		err = rcar_canfd_channel_probe(gpriv, ch, fcan_freq,
+					       transceivers[ch]);
 		if (err)
 			goto fail_channel;
 	}
-- 
cgit v1.2.3-70-g09d2


From 33eced402b184b5f2127f45531099c265d34e269 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 21 Mar 2023 17:15:01 +0100
Subject: can: rcar_canfd: Improve error messages

Improve printed error messages:
  - Replace numerical error codes by mnemotechnic error codes, to
    improve the user experience in case of errors,
  - Drop parentheses around printed numbers, cfr.
    Documentation/process/coding-style.rst,
  - Drop printing of an error message in case of out-of-memory, as the
    core memory allocation code already takes care of this.

Suggested-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/4162cc46f72257ec191007675933985b6df394b9.1679414936.git.geert+renesas@glider.be
[mkl: use colon instead of comma]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/rcar/rcar_canfd.c | 41 +++++++++++++++++++--------------------
 1 file changed, 20 insertions(+), 21 deletions(-)

diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index d8fbc3fca475..701311dabab3 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -1424,13 +1424,13 @@ static int rcar_canfd_open(struct net_device *ndev)
 	/* Peripheral clock is already enabled in probe */
 	err = clk_prepare_enable(gpriv->can_clk);
 	if (err) {
-		netdev_err(ndev, "failed to enable CAN clock, error %d\n", err);
+		netdev_err(ndev, "failed to enable CAN clock: %pe\n", ERR_PTR(err));
 		goto out_phy;
 	}
 
 	err = open_candev(ndev);
 	if (err) {
-		netdev_err(ndev, "open_candev() failed, error %d\n", err);
+		netdev_err(ndev, "open_candev() failed: %pe\n", ERR_PTR(err));
 		goto out_can_clock;
 	}
 
@@ -1731,10 +1731,9 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 	int err = -ENODEV;
 
 	ndev = alloc_candev(sizeof(*priv), RCANFD_FIFO_DEPTH);
-	if (!ndev) {
-		dev_err(dev, "alloc_candev() failed\n");
+	if (!ndev)
 		return -ENOMEM;
-	}
+
 	priv = netdev_priv(ndev);
 
 	ndev->netdev_ops = &rcar_canfd_netdev_ops;
@@ -1777,8 +1776,8 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 				       rcar_canfd_channel_err_interrupt, 0,
 				       irq_name, priv);
 		if (err) {
-			dev_err(dev, "devm_request_irq CH Err(%d) failed, error %d\n",
-				err_irq, err);
+			dev_err(dev, "devm_request_irq CH Err %d failed: %pe\n",
+				err_irq, ERR_PTR(err));
 			goto fail;
 		}
 		irq_name = devm_kasprintf(dev, GFP_KERNEL, "canfd.ch%d_trx",
@@ -1791,8 +1790,8 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 				       rcar_canfd_channel_tx_interrupt, 0,
 				       irq_name, priv);
 		if (err) {
-			dev_err(dev, "devm_request_irq Tx (%d) failed, error %d\n",
-				tx_irq, err);
+			dev_err(dev, "devm_request_irq Tx %d failed: %pe\n",
+				tx_irq, ERR_PTR(err));
 			goto fail;
 		}
 	}
@@ -1823,7 +1822,7 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 	gpriv->ch[priv->channel] = priv;
 	err = register_candev(ndev);
 	if (err) {
-		dev_err(dev, "register_candev() failed, error %d\n", err);
+		dev_err(dev, "register_candev() failed: %pe\n", ERR_PTR(err));
 		goto fail_candev;
 	}
 	dev_info(dev, "device registered (channel %u)\n", priv->channel);
@@ -1967,16 +1966,16 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 				       rcar_canfd_channel_interrupt, 0,
 				       "canfd.ch_int", gpriv);
 		if (err) {
-			dev_err(dev, "devm_request_irq(%d) failed, error %d\n",
-				ch_irq, err);
+			dev_err(dev, "devm_request_irq %d failed: %pe\n",
+				ch_irq, ERR_PTR(err));
 			goto fail_dev;
 		}
 
 		err = devm_request_irq(dev, g_irq, rcar_canfd_global_interrupt,
 				       0, "canfd.g_int", gpriv);
 		if (err) {
-			dev_err(dev, "devm_request_irq(%d) failed, error %d\n",
-				g_irq, err);
+			dev_err(dev, "devm_request_irq %d failed: %pe\n",
+				g_irq, ERR_PTR(err));
 			goto fail_dev;
 		}
 	} else {
@@ -1985,8 +1984,8 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 				       "canfd.g_recc", gpriv);
 
 		if (err) {
-			dev_err(dev, "devm_request_irq(%d) failed, error %d\n",
-				g_recc_irq, err);
+			dev_err(dev, "devm_request_irq %d failed: %pe\n",
+				g_recc_irq, ERR_PTR(err));
 			goto fail_dev;
 		}
 
@@ -1994,8 +1993,8 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 				       rcar_canfd_global_err_interrupt, 0,
 				       "canfd.g_err", gpriv);
 		if (err) {
-			dev_err(dev, "devm_request_irq(%d) failed, error %d\n",
-				g_err_irq, err);
+			dev_err(dev, "devm_request_irq %d failed: %pe\n",
+				g_err_irq, ERR_PTR(err));
 			goto fail_dev;
 		}
 	}
@@ -2012,14 +2011,14 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 	/* Enable peripheral clock for register access */
 	err = clk_prepare_enable(gpriv->clkp);
 	if (err) {
-		dev_err(dev, "failed to enable peripheral clock, error %d\n",
-			err);
+		dev_err(dev, "failed to enable peripheral clock: %pe\n",
+			ERR_PTR(err));
 		goto fail_reset;
 	}
 
 	err = rcar_canfd_reset_controller(gpriv);
 	if (err) {
-		dev_err(dev, "reset controller failed\n");
+		dev_err(dev, "reset controller failed: %pe\n", ERR_PTR(err));
 		goto fail_clk;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 594503341de74f69dbccaedf4dbbb3aa02ec0534 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:33:15 +0800
Subject: can: c_can: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Link: https://lore.kernel.org/all/20230323113318.9473-1-cai.huoqing@linux.dev
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/c_can/c_can_pci.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/can/c_can/c_can_pci.c b/drivers/net/can/c_can/c_can_pci.c
index bf2f8c3da1c1..093bea597f4e 100644
--- a/drivers/net/can/c_can/c_can_pci.c
+++ b/drivers/net/can/c_can/c_can_pci.c
@@ -227,7 +227,6 @@ out_iounmap:
 	pci_iounmap(pdev, addr);
 out_release_regions:
 	pci_disable_msi(pdev);
-	pci_clear_master(pdev);
 	pci_release_regions(pdev);
 out_disable_device:
 	pci_disable_device(pdev);
@@ -247,7 +246,6 @@ static void c_can_pci_remove(struct pci_dev *pdev)
 
 	pci_iounmap(pdev, addr);
 	pci_disable_msi(pdev);
-	pci_clear_master(pdev);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 }
-- 
cgit v1.2.3-70-g09d2


From c9d23f9657cabfd2836a096bf6eddf8df2cf1434 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:33:16 +0800
Subject: can: ctucanfd: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Link: https://lore.kernel.org/all/20230323113318.9473-2-cai.huoqing@linux.dev
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/ctucanfd/ctucanfd_pci.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/net/can/ctucanfd/ctucanfd_pci.c b/drivers/net/can/ctucanfd/ctucanfd_pci.c
index 8f2956a8ae43..9da09e7dd63a 100644
--- a/drivers/net/can/ctucanfd/ctucanfd_pci.c
+++ b/drivers/net/can/ctucanfd/ctucanfd_pci.c
@@ -206,10 +206,8 @@ err_pci_iounmap_bar0:
 err_pci_iounmap_bar1:
 	pci_iounmap(pdev, addr);
 err_release_regions:
-	if (msi_ok) {
+	if (msi_ok)
 		pci_disable_msi(pdev);
-		pci_clear_master(pdev);
-	}
 	pci_release_regions(pdev);
 err_disable_device:
 	pci_disable_device(pdev);
@@ -257,10 +255,8 @@ static void ctucan_pci_remove(struct pci_dev *pdev)
 
 	pci_iounmap(pdev, bdata->bar1_base);
 
-	if (bdata->use_msi) {
+	if (bdata->use_msi)
 		pci_disable_msi(pdev);
-		pci_clear_master(pdev);
-	}
 
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-70-g09d2


From 8db931835fad6a2229166fc0719f55c5ba070516 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:33:17 +0800
Subject: can: kvaser_pciefd: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Link: https://lore.kernel.org/all/20230323113318.9473-3-cai.huoqing@linux.dev
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index bcad11709bc9..53e8a914c88b 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1907,7 +1907,6 @@ static void kvaser_pciefd_remove(struct pci_dev *pdev)
 
 	free_irq(pcie->pci->irq, pcie);
 
-	pci_clear_master(pdev);
 	pci_iounmap(pdev, pcie->reg_base);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-70-g09d2


From a57915aee31596d2380a875964c78f743162383c Mon Sep 17 00:00:00 2001
From: Frank Jungclaus <frank.jungclaus@esd.eu>
Date: Wed, 22 Feb 2023 17:37:54 +0100
Subject: can: esd_usb: Improve code readability by means of replacing struct
 esd_usb_msg with a union

As suggested by Vincent Mailhol, declare struct esd_usb_msg as a union
instead of a struct. Then replace all msg->msg.something constructs,
that make use of esd_usb_msg, with simpler and prettier looking
msg->something variants.

Link: https://lore.kernel.org/all/CAMZ6RqKRzJwmMShVT9QKwiQ5LJaQupYqkPkKjhRBsP=12QYpfA@mail.gmail.com/
Suggested-by: Vincent MAILHOL <mailhol.vincent@wanadoo.fr>
Signed-off-by: Frank Jungclaus <frank.jungclaus@esd.eu>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/20230222163754.3711766-1-frank.jungclaus@esd.eu
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/esd_usb.c | 166 +++++++++++++++++++++---------------------
 1 file changed, 82 insertions(+), 84 deletions(-)

diff --git a/drivers/net/can/usb/esd_usb.c b/drivers/net/can/usb/esd_usb.c
index 55b36973952d..e78bb468115a 100644
--- a/drivers/net/can/usb/esd_usb.c
+++ b/drivers/net/can/usb/esd_usb.c
@@ -174,17 +174,15 @@ struct set_baudrate_msg {
 };
 
 /* Main message type used between library and application */
-struct __packed esd_usb_msg {
-	union {
-		struct header_msg hdr;
-		struct version_msg version;
-		struct version_reply_msg version_reply;
-		struct rx_msg rx;
-		struct tx_msg tx;
-		struct tx_done_msg txdone;
-		struct set_baudrate_msg setbaud;
-		struct id_filter_msg filter;
-	} msg;
+union __packed esd_usb_msg {
+	struct header_msg hdr;
+	struct version_msg version;
+	struct version_reply_msg version_reply;
+	struct rx_msg rx;
+	struct tx_msg tx;
+	struct tx_done_msg txdone;
+	struct set_baudrate_msg setbaud;
+	struct id_filter_msg filter;
 };
 
 static struct usb_device_id esd_usb_table[] = {
@@ -229,22 +227,22 @@ struct esd_usb_net_priv {
 };
 
 static void esd_usb_rx_event(struct esd_usb_net_priv *priv,
-			     struct esd_usb_msg *msg)
+			     union esd_usb_msg *msg)
 {
 	struct net_device_stats *stats = &priv->netdev->stats;
 	struct can_frame *cf;
 	struct sk_buff *skb;
-	u32 id = le32_to_cpu(msg->msg.rx.id) & ESD_IDMASK;
+	u32 id = le32_to_cpu(msg->rx.id) & ESD_IDMASK;
 
 	if (id == ESD_EV_CAN_ERROR_EXT) {
-		u8 state = msg->msg.rx.ev_can_err_ext.status;
-		u8 ecc = msg->msg.rx.ev_can_err_ext.ecc;
-		u8 rxerr = msg->msg.rx.ev_can_err_ext.rec;
-		u8 txerr = msg->msg.rx.ev_can_err_ext.tec;
+		u8 state = msg->rx.ev_can_err_ext.status;
+		u8 ecc = msg->rx.ev_can_err_ext.ecc;
+		u8 rxerr = msg->rx.ev_can_err_ext.rec;
+		u8 txerr = msg->rx.ev_can_err_ext.tec;
 
 		netdev_dbg(priv->netdev,
 			   "CAN_ERR_EV_EXT: dlc=%#02x state=%02x ecc=%02x rec=%02x tec=%02x\n",
-			   msg->msg.rx.dlc, state, ecc, rxerr, txerr);
+			   msg->rx.dlc, state, ecc, rxerr, txerr);
 
 		skb = alloc_can_err_skb(priv->netdev, &cf);
 
@@ -322,7 +320,7 @@ static void esd_usb_rx_event(struct esd_usb_net_priv *priv,
 }
 
 static void esd_usb_rx_can_msg(struct esd_usb_net_priv *priv,
-			       struct esd_usb_msg *msg)
+			       union esd_usb_msg *msg)
 {
 	struct net_device_stats *stats = &priv->netdev->stats;
 	struct can_frame *cf;
@@ -333,7 +331,7 @@ static void esd_usb_rx_can_msg(struct esd_usb_net_priv *priv,
 	if (!netif_device_present(priv->netdev))
 		return;
 
-	id = le32_to_cpu(msg->msg.rx.id);
+	id = le32_to_cpu(msg->rx.id);
 
 	if (id & ESD_EVENT) {
 		esd_usb_rx_event(priv, msg);
@@ -345,17 +343,17 @@ static void esd_usb_rx_can_msg(struct esd_usb_net_priv *priv,
 		}
 
 		cf->can_id = id & ESD_IDMASK;
-		can_frame_set_cc_len(cf, msg->msg.rx.dlc & ~ESD_RTR,
+		can_frame_set_cc_len(cf, msg->rx.dlc & ~ESD_RTR,
 				     priv->can.ctrlmode);
 
 		if (id & ESD_EXTID)
 			cf->can_id |= CAN_EFF_FLAG;
 
-		if (msg->msg.rx.dlc & ESD_RTR) {
+		if (msg->rx.dlc & ESD_RTR) {
 			cf->can_id |= CAN_RTR_FLAG;
 		} else {
 			for (i = 0; i < cf->len; i++)
-				cf->data[i] = msg->msg.rx.data[i];
+				cf->data[i] = msg->rx.data[i];
 
 			stats->rx_bytes += cf->len;
 		}
@@ -366,7 +364,7 @@ static void esd_usb_rx_can_msg(struct esd_usb_net_priv *priv,
 }
 
 static void esd_usb_tx_done_msg(struct esd_usb_net_priv *priv,
-				struct esd_usb_msg *msg)
+				union esd_usb_msg *msg)
 {
 	struct net_device_stats *stats = &priv->netdev->stats;
 	struct net_device *netdev = priv->netdev;
@@ -375,9 +373,9 @@ static void esd_usb_tx_done_msg(struct esd_usb_net_priv *priv,
 	if (!netif_device_present(netdev))
 		return;
 
-	context = &priv->tx_contexts[msg->msg.txdone.hnd & (MAX_TX_URBS - 1)];
+	context = &priv->tx_contexts[msg->txdone.hnd & (MAX_TX_URBS - 1)];
 
-	if (!msg->msg.txdone.status) {
+	if (!msg->txdone.status) {
 		stats->tx_packets++;
 		stats->tx_bytes += can_get_echo_skb(netdev, context->echo_index,
 						    NULL);
@@ -417,32 +415,32 @@ static void esd_usb_read_bulk_callback(struct urb *urb)
 	}
 
 	while (pos < urb->actual_length) {
-		struct esd_usb_msg *msg;
+		union esd_usb_msg *msg;
 
-		msg = (struct esd_usb_msg *)(urb->transfer_buffer + pos);
+		msg = (union esd_usb_msg *)(urb->transfer_buffer + pos);
 
-		switch (msg->msg.hdr.cmd) {
+		switch (msg->hdr.cmd) {
 		case CMD_CAN_RX:
-			if (msg->msg.rx.net >= dev->net_count) {
+			if (msg->rx.net >= dev->net_count) {
 				dev_err(dev->udev->dev.parent, "format error\n");
 				break;
 			}
 
-			esd_usb_rx_can_msg(dev->nets[msg->msg.rx.net], msg);
+			esd_usb_rx_can_msg(dev->nets[msg->rx.net], msg);
 			break;
 
 		case CMD_CAN_TX:
-			if (msg->msg.txdone.net >= dev->net_count) {
+			if (msg->txdone.net >= dev->net_count) {
 				dev_err(dev->udev->dev.parent, "format error\n");
 				break;
 			}
 
-			esd_usb_tx_done_msg(dev->nets[msg->msg.txdone.net],
+			esd_usb_tx_done_msg(dev->nets[msg->txdone.net],
 					    msg);
 			break;
 		}
 
-		pos += msg->msg.hdr.len << 2;
+		pos += msg->hdr.len << 2;
 
 		if (pos > urb->actual_length) {
 			dev_err(dev->udev->dev.parent, "format error\n");
@@ -473,7 +471,7 @@ static void esd_usb_write_bulk_callback(struct urb *urb)
 	struct esd_tx_urb_context *context = urb->context;
 	struct esd_usb_net_priv *priv;
 	struct net_device *netdev;
-	size_t size = sizeof(struct esd_usb_msg);
+	size_t size = sizeof(union esd_usb_msg);
 
 	WARN_ON(!context);
 
@@ -529,20 +527,20 @@ static ssize_t nets_show(struct device *d,
 }
 static DEVICE_ATTR_RO(nets);
 
-static int esd_usb_send_msg(struct esd_usb *dev, struct esd_usb_msg *msg)
+static int esd_usb_send_msg(struct esd_usb *dev, union esd_usb_msg *msg)
 {
 	int actual_length;
 
 	return usb_bulk_msg(dev->udev,
 			    usb_sndbulkpipe(dev->udev, 2),
 			    msg,
-			    msg->msg.hdr.len << 2,
+			    msg->hdr.len << 2,
 			    &actual_length,
 			    1000);
 }
 
 static int esd_usb_wait_msg(struct esd_usb *dev,
-			    struct esd_usb_msg *msg)
+			    union esd_usb_msg *msg)
 {
 	int actual_length;
 
@@ -630,7 +628,7 @@ static int esd_usb_start(struct esd_usb_net_priv *priv)
 {
 	struct esd_usb *dev = priv->usb;
 	struct net_device *netdev = priv->netdev;
-	struct esd_usb_msg *msg;
+	union esd_usb_msg *msg;
 	int err, i;
 
 	msg = kmalloc(sizeof(*msg), GFP_KERNEL);
@@ -651,14 +649,14 @@ static int esd_usb_start(struct esd_usb_net_priv *priv)
 	 * the number of the starting bitmask (0..64) to the filter.option
 	 * field followed by only some bitmasks.
 	 */
-	msg->msg.hdr.cmd = CMD_IDADD;
-	msg->msg.hdr.len = 2 + ESD_MAX_ID_SEGMENT;
-	msg->msg.filter.net = priv->index;
-	msg->msg.filter.option = ESD_ID_ENABLE; /* start with segment 0 */
+	msg->hdr.cmd = CMD_IDADD;
+	msg->hdr.len = 2 + ESD_MAX_ID_SEGMENT;
+	msg->filter.net = priv->index;
+	msg->filter.option = ESD_ID_ENABLE; /* start with segment 0 */
 	for (i = 0; i < ESD_MAX_ID_SEGMENT; i++)
-		msg->msg.filter.mask[i] = cpu_to_le32(0xffffffff);
+		msg->filter.mask[i] = cpu_to_le32(0xffffffff);
 	/* enable 29bit extended IDs */
-	msg->msg.filter.mask[ESD_MAX_ID_SEGMENT] = cpu_to_le32(0x00000001);
+	msg->filter.mask[ESD_MAX_ID_SEGMENT] = cpu_to_le32(0x00000001);
 
 	err = esd_usb_send_msg(dev, msg);
 	if (err)
@@ -734,12 +732,12 @@ static netdev_tx_t esd_usb_start_xmit(struct sk_buff *skb,
 	struct esd_tx_urb_context *context = NULL;
 	struct net_device_stats *stats = &netdev->stats;
 	struct can_frame *cf = (struct can_frame *)skb->data;
-	struct esd_usb_msg *msg;
+	union esd_usb_msg *msg;
 	struct urb *urb;
 	u8 *buf;
 	int i, err;
 	int ret = NETDEV_TX_OK;
-	size_t size = sizeof(struct esd_usb_msg);
+	size_t size = sizeof(union esd_usb_msg);
 
 	if (can_dev_dropped_skb(netdev, skb))
 		return NETDEV_TX_OK;
@@ -761,24 +759,24 @@ static netdev_tx_t esd_usb_start_xmit(struct sk_buff *skb,
 		goto nobufmem;
 	}
 
-	msg = (struct esd_usb_msg *)buf;
+	msg = (union esd_usb_msg *)buf;
 
-	msg->msg.hdr.len = 3; /* minimal length */
-	msg->msg.hdr.cmd = CMD_CAN_TX;
-	msg->msg.tx.net = priv->index;
-	msg->msg.tx.dlc = can_get_cc_dlc(cf, priv->can.ctrlmode);
-	msg->msg.tx.id = cpu_to_le32(cf->can_id & CAN_ERR_MASK);
+	msg->hdr.len = 3; /* minimal length */
+	msg->hdr.cmd = CMD_CAN_TX;
+	msg->tx.net = priv->index;
+	msg->tx.dlc = can_get_cc_dlc(cf, priv->can.ctrlmode);
+	msg->tx.id = cpu_to_le32(cf->can_id & CAN_ERR_MASK);
 
 	if (cf->can_id & CAN_RTR_FLAG)
-		msg->msg.tx.dlc |= ESD_RTR;
+		msg->tx.dlc |= ESD_RTR;
 
 	if (cf->can_id & CAN_EFF_FLAG)
-		msg->msg.tx.id |= cpu_to_le32(ESD_EXTID);
+		msg->tx.id |= cpu_to_le32(ESD_EXTID);
 
 	for (i = 0; i < cf->len; i++)
-		msg->msg.tx.data[i] = cf->data[i];
+		msg->tx.data[i] = cf->data[i];
 
-	msg->msg.hdr.len += (cf->len + 3) >> 2;
+	msg->hdr.len += (cf->len + 3) >> 2;
 
 	for (i = 0; i < MAX_TX_URBS; i++) {
 		if (priv->tx_contexts[i].echo_index == MAX_TX_URBS) {
@@ -798,10 +796,10 @@ static netdev_tx_t esd_usb_start_xmit(struct sk_buff *skb,
 	context->echo_index = i;
 
 	/* hnd must not be 0 - MSB is stripped in txdone handling */
-	msg->msg.tx.hnd = 0x80000000 | i; /* returned in TX done message */
+	msg->tx.hnd = 0x80000000 | i; /* returned in TX done message */
 
 	usb_fill_bulk_urb(urb, dev->udev, usb_sndbulkpipe(dev->udev, 2), buf,
-			  msg->msg.hdr.len << 2,
+			  msg->hdr.len << 2,
 			  esd_usb_write_bulk_callback, context);
 
 	urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
@@ -855,7 +853,7 @@ nourbmem:
 static int esd_usb_close(struct net_device *netdev)
 {
 	struct esd_usb_net_priv *priv = netdev_priv(netdev);
-	struct esd_usb_msg *msg;
+	union esd_usb_msg *msg;
 	int i;
 
 	msg = kmalloc(sizeof(*msg), GFP_KERNEL);
@@ -863,21 +861,21 @@ static int esd_usb_close(struct net_device *netdev)
 		return -ENOMEM;
 
 	/* Disable all IDs (see esd_usb_start()) */
-	msg->msg.hdr.cmd = CMD_IDADD;
-	msg->msg.hdr.len = 2 + ESD_MAX_ID_SEGMENT;
-	msg->msg.filter.net = priv->index;
-	msg->msg.filter.option = ESD_ID_ENABLE; /* start with segment 0 */
+	msg->hdr.cmd = CMD_IDADD;
+	msg->hdr.len = 2 + ESD_MAX_ID_SEGMENT;
+	msg->filter.net = priv->index;
+	msg->filter.option = ESD_ID_ENABLE; /* start with segment 0 */
 	for (i = 0; i <= ESD_MAX_ID_SEGMENT; i++)
-		msg->msg.filter.mask[i] = 0;
+		msg->filter.mask[i] = 0;
 	if (esd_usb_send_msg(priv->usb, msg) < 0)
 		netdev_err(netdev, "sending idadd message failed\n");
 
 	/* set CAN controller to reset mode */
-	msg->msg.hdr.len = 2;
-	msg->msg.hdr.cmd = CMD_SETBAUD;
-	msg->msg.setbaud.net = priv->index;
-	msg->msg.setbaud.rsvd = 0;
-	msg->msg.setbaud.baud = cpu_to_le32(ESD_USB_NO_BAUDRATE);
+	msg->hdr.len = 2;
+	msg->hdr.cmd = CMD_SETBAUD;
+	msg->setbaud.net = priv->index;
+	msg->setbaud.rsvd = 0;
+	msg->setbaud.baud = cpu_to_le32(ESD_USB_NO_BAUDRATE);
 	if (esd_usb_send_msg(priv->usb, msg) < 0)
 		netdev_err(netdev, "sending setbaud message failed\n");
 
@@ -919,7 +917,7 @@ static int esd_usb2_set_bittiming(struct net_device *netdev)
 {
 	struct esd_usb_net_priv *priv = netdev_priv(netdev);
 	struct can_bittiming *bt = &priv->can.bittiming;
-	struct esd_usb_msg *msg;
+	union esd_usb_msg *msg;
 	int err;
 	u32 canbtr;
 	int sjw_shift;
@@ -950,11 +948,11 @@ static int esd_usb2_set_bittiming(struct net_device *netdev)
 	if (!msg)
 		return -ENOMEM;
 
-	msg->msg.hdr.len = 2;
-	msg->msg.hdr.cmd = CMD_SETBAUD;
-	msg->msg.setbaud.net = priv->index;
-	msg->msg.setbaud.rsvd = 0;
-	msg->msg.setbaud.baud = cpu_to_le32(canbtr);
+	msg->hdr.len = 2;
+	msg->hdr.cmd = CMD_SETBAUD;
+	msg->setbaud.net = priv->index;
+	msg->setbaud.rsvd = 0;
+	msg->setbaud.baud = cpu_to_le32(canbtr);
 
 	netdev_info(netdev, "setting BTR=%#x\n", canbtr);
 
@@ -1065,7 +1063,7 @@ static int esd_usb_probe(struct usb_interface *intf,
 			 const struct usb_device_id *id)
 {
 	struct esd_usb *dev;
-	struct esd_usb_msg *msg;
+	union esd_usb_msg *msg;
 	int i, err;
 
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
@@ -1087,11 +1085,11 @@ static int esd_usb_probe(struct usb_interface *intf,
 	}
 
 	/* query number of CAN interfaces (nets) */
-	msg->msg.hdr.cmd = CMD_VERSION;
-	msg->msg.hdr.len = 2;
-	msg->msg.version.rsvd = 0;
-	msg->msg.version.flags = 0;
-	msg->msg.version.drv_version = 0;
+	msg->hdr.cmd = CMD_VERSION;
+	msg->hdr.len = 2;
+	msg->version.rsvd = 0;
+	msg->version.flags = 0;
+	msg->version.drv_version = 0;
 
 	err = esd_usb_send_msg(dev, msg);
 	if (err < 0) {
@@ -1105,8 +1103,8 @@ static int esd_usb_probe(struct usb_interface *intf,
 		goto free_msg;
 	}
 
-	dev->net_count = (int)msg->msg.version_reply.nets;
-	dev->version = le32_to_cpu(msg->msg.version_reply.version);
+	dev->net_count = (int)msg->version_reply.nets;
+	dev->version = le32_to_cpu(msg->version_reply.version);
 
 	if (device_create_file(&intf->dev, &dev_attr_firmware))
 		dev_err(&intf->dev,
-- 
cgit v1.2.3-70-g09d2


From 73042934e4a30d9f45ff9cb0b3b029a01dbe7130 Mon Sep 17 00:00:00 2001
From: Markus Schneider-Pargmann <msp@baylibre.com>
Date: Wed, 15 Mar 2023 12:05:31 +0100
Subject: can: m_can: Remove repeated check for is_peripheral

Merge both if-blocks to fix this.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/all/20230315110546.2518305-2-msp@baylibre.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index 8e83d6963d85..563625a701fc 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -1592,10 +1592,8 @@ static int m_can_close(struct net_device *dev)
 		cdev->tx_skb = NULL;
 		destroy_workqueue(cdev->tx_wq);
 		cdev->tx_wq = NULL;
-	}
-
-	if (cdev->is_peripheral)
 		can_rx_offload_disable(&cdev->offload);
+	}
 
 	close_candev(dev);
 
-- 
cgit v1.2.3-70-g09d2


From 4ab639480900be41d7c7c0b89a6db7fa5fe951e4 Mon Sep 17 00:00:00 2001
From: Markus Schneider-Pargmann <msp@baylibre.com>
Date: Wed, 15 Mar 2023 12:05:32 +0100
Subject: can: m_can: Always acknowledge all interrupts

The code already exits the function on !ir before this condition. No
need to check again if anything is set as IR_ALL_INT is 0xffffffff.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/all/20230315110546.2518305-3-msp@baylibre.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index 563625a701fc..8eb327ae3bdf 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -1083,8 +1083,7 @@ static irqreturn_t m_can_isr(int irq, void *dev_id)
 		return IRQ_NONE;
 
 	/* ACK all irqs */
-	if (ir & IR_ALL_INT)
-		m_can_write(cdev, M_CAN_IR, ir);
+	m_can_write(cdev, M_CAN_IR, ir);
 
 	if (cdev->ops->clear_interrupts)
 		cdev->ops->clear_interrupts(cdev);
-- 
cgit v1.2.3-70-g09d2


From 71725bfdbbf263ec4897ce310087d3f6bd331c19 Mon Sep 17 00:00:00 2001
From: Markus Schneider-Pargmann <msp@baylibre.com>
Date: Wed, 15 Mar 2023 12:05:33 +0100
Subject: can: m_can: Remove double interrupt enable

Interrupts are enabled a few lines further down as well. Remove this
second call to enable all interrupts.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/all/20230315110546.2518305-4-msp@baylibre.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index 8eb327ae3bdf..5274d9642566 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -1364,7 +1364,6 @@ static int m_can_chip_config(struct net_device *dev)
 	m_can_write(cdev, M_CAN_TEST, test);
 
 	/* Enable interrupts */
-	m_can_write(cdev, M_CAN_IR, IR_ALL_INT);
 	if (!(cdev->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING))
 		if (cdev->version == 30)
 			m_can_write(cdev, M_CAN_IE, IR_ALL_INT &
-- 
cgit v1.2.3-70-g09d2


From 897e663218e2d05607a6f5b258134cb026a2fa85 Mon Sep 17 00:00:00 2001
From: Markus Schneider-Pargmann <msp@baylibre.com>
Date: Wed, 15 Mar 2023 12:05:34 +0100
Subject: can: m_can: Disable unused interrupts

There are a number of interrupts that are not used by the driver at the
moment. Disable all of these.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/all/20230315110546.2518305-5-msp@baylibre.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index 5274d9642566..e7aceeba3759 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -1261,6 +1261,7 @@ static int m_can_set_bittiming(struct net_device *dev)
 static int m_can_chip_config(struct net_device *dev)
 {
 	struct m_can_classdev *cdev = netdev_priv(dev);
+	u32 interrupts = IR_ALL_INT;
 	u32 cccr, test;
 	int err;
 
@@ -1270,6 +1271,11 @@ static int m_can_chip_config(struct net_device *dev)
 		return err;
 	}
 
+	/* Disable unused interrupts */
+	interrupts &= ~(IR_ARA | IR_ELO | IR_DRX | IR_TEFF | IR_TEFW | IR_TFE |
+			IR_TCF | IR_HPM | IR_RF1F | IR_RF1W | IR_RF1N |
+			IR_RF0F | IR_RF0W);
+
 	m_can_config_endisable(cdev, true);
 
 	/* RX Buffer/FIFO Element Size 64 bytes data field */
@@ -1364,15 +1370,13 @@ static int m_can_chip_config(struct net_device *dev)
 	m_can_write(cdev, M_CAN_TEST, test);
 
 	/* Enable interrupts */
-	if (!(cdev->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING))
+	if (!(cdev->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING)) {
 		if (cdev->version == 30)
-			m_can_write(cdev, M_CAN_IE, IR_ALL_INT &
-				    ~(IR_ERR_LEC_30X));
+			interrupts &= ~(IR_ERR_LEC_30X);
 		else
-			m_can_write(cdev, M_CAN_IE, IR_ALL_INT &
-				    ~(IR_ERR_LEC_31X));
-	else
-		m_can_write(cdev, M_CAN_IE, IR_ALL_INT);
+			interrupts &= ~(IR_ERR_LEC_31X);
+	}
+	m_can_write(cdev, M_CAN_IE, interrupts);
 
 	/* route all interrupts to INT0 */
 	m_can_write(cdev, M_CAN_ILS, ILS_ALL_INT0);
-- 
cgit v1.2.3-70-g09d2


From 9083e0b09df33376abef2bb28e6a4b0c88f4d00b Mon Sep 17 00:00:00 2001
From: Markus Schneider-Pargmann <msp@baylibre.com>
Date: Wed, 15 Mar 2023 12:05:35 +0100
Subject: can: m_can: Keep interrupts enabled during peripheral read

Interrupts currently get disabled if the interrupt status shows new
received data. Non-peripheral chips handle receiving in a worker thread,
but peripheral chips are handling the receive process in the threaded
interrupt routine itself without scheduling it for a different worker.
So there is no need to disable interrupts for peripheral chips.

Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/all/20230315110546.2518305-6-msp@baylibre.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index e7aceeba3759..a5003435802b 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -972,8 +972,8 @@ static int m_can_rx_peripheral(struct net_device *dev, u32 irqstatus)
 	/* Don't re-enable interrupts if the driver had a fatal error
 	 * (e.g., FIFO read failure).
 	 */
-	if (work_done >= 0)
-		m_can_enable_all_interrupts(cdev);
+	if (work_done < 0)
+		m_can_disable_all_interrupts(cdev);
 
 	return work_done;
 }
@@ -1095,11 +1095,12 @@ static irqreturn_t m_can_isr(int irq, void *dev_id)
 	 */
 	if ((ir & IR_RF0N) || (ir & IR_ERR_ALL_30X)) {
 		cdev->irqstatus = ir;
-		m_can_disable_all_interrupts(cdev);
-		if (!cdev->is_peripheral)
+		if (!cdev->is_peripheral) {
+			m_can_disable_all_interrupts(cdev);
 			napi_schedule(&cdev->napi);
-		else if (m_can_rx_peripheral(dev, ir) < 0)
+		} else if (m_can_rx_peripheral(dev, ir) < 0) {
 			goto out_fail;
+		}
 	}
 
 	if (cdev->version == 30) {
-- 
cgit v1.2.3-70-g09d2


From 55fbae05476df65e5eee8be54f61d0257af0240b Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Fri, 24 Mar 2023 11:42:41 -0700
Subject: bpf: Check IS_ERR for the bpf_map_get() return value

This patch fixes a mistake in checking NULL instead of
checking IS_ERR for the bpf_map_get() return value.

It also fixes the return value in link_update_map() from -EINVAL
to PTR_ERR(*_map).

Reported-by: syzbot+71ccc0fe37abb458406b@syzkaller.appspotmail.com
Fixes: 68b04864ca42 ("bpf: Create links for BPF struct_ops maps.")
Fixes: aef56f2e918b ("bpf: Update the struct_ops of a bpf_link.")
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Acked-by: Kui-Feng Lee <kuifeng@meta.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20230324184241.1387437-1-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_struct_ops.c | 4 ++--
 kernel/bpf/syscall.c        | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 6401deca3b56..d3f0a4825fa6 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -871,8 +871,8 @@ int bpf_struct_ops_link_create(union bpf_attr *attr)
 	int err;
 
 	map = bpf_map_get(attr->link_create.map_fd);
-	if (!map)
-		return -EINVAL;
+	if (IS_ERR(map))
+		return PTR_ERR(map);
 
 	st_map = (struct bpf_struct_ops_map *)map;
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index b4d758fa5981..a09597c95029 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -4689,12 +4689,12 @@ static int link_update_map(struct bpf_link *link, union bpf_attr *attr)
 
 	new_map = bpf_map_get(attr->link_update.new_map_fd);
 	if (IS_ERR(new_map))
-		return -EINVAL;
+		return PTR_ERR(new_map);
 
 	if (attr->link_update.flags & BPF_F_REPLACE) {
 		old_map = bpf_map_get(attr->link_update.old_map_fd);
 		if (IS_ERR(old_map)) {
-			ret = -EINVAL;
+			ret = PTR_ERR(old_map);
 			goto out_put;
 		}
 	} else if (attr->link_update.old_map_fd) {
-- 
cgit v1.2.3-70-g09d2


From 4e0473f1060aa49621d40a113afde24818101d37 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Wed, 8 Feb 2023 07:51:02 +0200
Subject: lib: cpu_rmap: Avoid use after free on rmap->obj array entries

When calling irq_set_affinity_notifier() with NULL at the notify
argument, it will cause freeing of the glue pointer in the
corresponding array entry but will leave the pointer in the array. A
subsequent call to free_irq_cpu_rmap() will try to free this entry again
leading to possible use after free.

Fix that by setting NULL to the array entry and checking that we have
non-zero at the array entry when iterating over the array in
free_irq_cpu_rmap().

The current code does not suffer from this since there are no cases
where irq_set_affinity_notifier(irq, NULL) (note the NULL passed for the
notify arg) is called, followed by a call to free_irq_cpu_rmap() so we
don't hit and issue. Subsequent patches in this series excersize this
flow, hence the required fix.

Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 lib/cpu_rmap.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/lib/cpu_rmap.c b/lib/cpu_rmap.c
index f08d9c56f712..e77f12bb3c77 100644
--- a/lib/cpu_rmap.c
+++ b/lib/cpu_rmap.c
@@ -232,7 +232,8 @@ void free_irq_cpu_rmap(struct cpu_rmap *rmap)
 
 	for (index = 0; index < rmap->used; index++) {
 		glue = rmap->obj[index];
-		irq_set_affinity_notifier(glue->notify.irq, NULL);
+		if (glue)
+			irq_set_affinity_notifier(glue->notify.irq, NULL);
 	}
 
 	cpu_rmap_put(rmap);
@@ -268,6 +269,7 @@ static void irq_cpu_rmap_release(struct kref *ref)
 		container_of(ref, struct irq_glue, notify.kref);
 
 	cpu_rmap_put(glue->rmap);
+	glue->rmap->obj[glue->index] = NULL;
 	kfree(glue);
 }
 
@@ -297,6 +299,7 @@ int irq_cpu_rmap_add(struct cpu_rmap *rmap, int irq)
 	rc = irq_set_affinity_notifier(irq, &glue->notify);
 	if (rc) {
 		cpu_rmap_put(glue->rmap);
+		rmap->obj[glue->index] = NULL;
 		kfree(glue);
 	}
 	return rc;
-- 
cgit v1.2.3-70-g09d2


From 9821d8d4628e630ab56f47a8e6b878a2576e069b Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Tue, 14 Feb 2023 09:29:46 +0200
Subject: lib: cpu_rmap: Use allocator for rmap entries

Use a proper allocator for rmap entries using a naive for loop. The
allocator relies on whether an entry is NULL to be considered free.
Remove the used field of rmap which is not needed.

Also, avoid crashing the kernel if an entry is not available.

Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 include/linux/cpu_rmap.h |  3 +--
 lib/cpu_rmap.c           | 43 ++++++++++++++++++++++++++++++++-----------
 2 files changed, 33 insertions(+), 13 deletions(-)

diff --git a/include/linux/cpu_rmap.h b/include/linux/cpu_rmap.h
index be8aea04d023..0ec745e6cd36 100644
--- a/include/linux/cpu_rmap.h
+++ b/include/linux/cpu_rmap.h
@@ -16,14 +16,13 @@
  * struct cpu_rmap - CPU affinity reverse-map
  * @refcount: kref for object
  * @size: Number of objects to be reverse-mapped
- * @used: Number of objects added
  * @obj: Pointer to array of object pointers
  * @near: For each CPU, the index and distance to the nearest object,
  *      based on affinity masks
  */
 struct cpu_rmap {
 	struct kref	refcount;
-	u16		size, used;
+	u16		size;
 	void		**obj;
 	struct {
 		u16	index;
diff --git a/lib/cpu_rmap.c b/lib/cpu_rmap.c
index e77f12bb3c77..5d4bf7a8b926 100644
--- a/lib/cpu_rmap.c
+++ b/lib/cpu_rmap.c
@@ -128,19 +128,31 @@ debug_print_rmap(const struct cpu_rmap *rmap, const char *prefix)
 }
 #endif
 
+static int get_free_index(struct cpu_rmap *rmap)
+{
+	int i;
+
+	for (i = 0; i < rmap->size; i++)
+		if (!rmap->obj[i])
+			return i;
+
+	return -ENOSPC;
+}
+
 /**
  * cpu_rmap_add - add object to a rmap
  * @rmap: CPU rmap allocated with alloc_cpu_rmap()
  * @obj: Object to add to rmap
  *
- * Return index of object.
+ * Return index of object or -ENOSPC if no free entry was found
  */
 int cpu_rmap_add(struct cpu_rmap *rmap, void *obj)
 {
-	u16 index;
+	int index = get_free_index(rmap);
+
+	if (index < 0)
+		return index;
 
-	BUG_ON(rmap->used >= rmap->size);
-	index = rmap->used++;
 	rmap->obj[index] = obj;
 	return index;
 }
@@ -230,7 +242,7 @@ void free_irq_cpu_rmap(struct cpu_rmap *rmap)
 	if (!rmap)
 		return;
 
-	for (index = 0; index < rmap->used; index++) {
+	for (index = 0; index < rmap->size; index++) {
 		glue = rmap->obj[index];
 		if (glue)
 			irq_set_affinity_notifier(glue->notify.irq, NULL);
@@ -295,13 +307,22 @@ int irq_cpu_rmap_add(struct cpu_rmap *rmap, int irq)
 	glue->notify.release = irq_cpu_rmap_release;
 	glue->rmap = rmap;
 	cpu_rmap_get(rmap);
-	glue->index = cpu_rmap_add(rmap, glue);
+	rc = cpu_rmap_add(rmap, glue);
+	if (rc < 0)
+		goto err_add;
+
+	glue->index = rc;
 	rc = irq_set_affinity_notifier(irq, &glue->notify);
-	if (rc) {
-		cpu_rmap_put(glue->rmap);
-		rmap->obj[glue->index] = NULL;
-		kfree(glue);
-	}
+	if (rc)
+		goto err_set;
+
+	return rc;
+
+err_set:
+	rmap->obj[glue->index] = NULL;
+err_add:
+	cpu_rmap_put(glue->rmap);
+	kfree(glue);
 	return rc;
 }
 EXPORT_SYMBOL(irq_cpu_rmap_add);
-- 
cgit v1.2.3-70-g09d2


From 71f0a2478605c100358a9f9e174849fa643bf8a7 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Tue, 14 Feb 2023 11:05:46 +0200
Subject: lib: cpu_rmap: Add irq_cpu_rmap_remove to complement irq_cpu_rmap_add

Add a function to complement irq_cpu_rmap_add(). It removes the irq from
the reverse mapping by setting the notifier to NULL. The function calls
irq_set_affinity_notifier() with NULL at the notify argument which then
cancel any pending notifier work and decrement reference on the
notifier. When ref count reaches zero, the glue pointer is kfree and the
rmap entry is set to NULL serving both to avoid second attempt to
release it and also making the rmap entry available for subsequent
mapping.

It should be noted the drivers usually creates the reverse mapping at
initialization time and remove it at unload time so we do not expect
failures in allocating rmap due to kref holding the glue entry.

Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 include/linux/cpu_rmap.h |  1 +
 lib/cpu_rmap.c           | 11 +++++++++++
 2 files changed, 12 insertions(+)

diff --git a/include/linux/cpu_rmap.h b/include/linux/cpu_rmap.h
index 0ec745e6cd36..cae324d10965 100644
--- a/include/linux/cpu_rmap.h
+++ b/include/linux/cpu_rmap.h
@@ -60,6 +60,7 @@ static inline struct cpu_rmap *alloc_irq_cpu_rmap(unsigned int size)
 }
 extern void free_irq_cpu_rmap(struct cpu_rmap *rmap);
 
+int irq_cpu_rmap_remove(struct cpu_rmap *rmap, int irq);
 extern int irq_cpu_rmap_add(struct cpu_rmap *rmap, int irq);
 
 #endif /* __LINUX_CPU_RMAP_H */
diff --git a/lib/cpu_rmap.c b/lib/cpu_rmap.c
index 5d4bf7a8b926..73c1636b927b 100644
--- a/lib/cpu_rmap.c
+++ b/lib/cpu_rmap.c
@@ -285,6 +285,17 @@ static void irq_cpu_rmap_release(struct kref *ref)
 	kfree(glue);
 }
 
+/**
+ * irq_cpu_rmap_remove - remove an IRQ from a CPU affinity reverse-map
+ * @rmap: The reverse-map
+ * @irq: The IRQ number
+ */
+int irq_cpu_rmap_remove(struct cpu_rmap *rmap, int irq)
+{
+	return irq_set_affinity_notifier(irq, NULL);
+}
+EXPORT_SYMBOL(irq_cpu_rmap_remove);
+
 /**
  * irq_cpu_rmap_add - add an IRQ to a CPU affinity reverse-map
  * @rmap: The reverse-map
-- 
cgit v1.2.3-70-g09d2


From b94616d9c6fd4e0d39616c68f46e33d7126add5a Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Wed, 28 Dec 2022 13:41:52 +0200
Subject: net/mlx5e: Coding style fix, add empty line

Add empty line between two function defnititions.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 38b32e98f3bd..b43121f64a80 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -637,6 +637,7 @@ static u16 async_eq_depth_devlink_param_get(struct mlx5_core_dev *dev)
 	mlx5_core_dbg(dev, "Failed to get param. using default. err = %d\n", err);
 	return MLX5_NUM_ASYNC_EQE;
 }
+
 static int create_async_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
-- 
cgit v1.2.3-70-g09d2


From 40a252c123c765a8d86d118b692c79aaa3b74ef6 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Wed, 28 Dec 2022 15:51:29 +0200
Subject: net/mlx5: Fix wrong comment

A control irq may be allocated from the parent device's pool in case
there is no SF dedicated pool. This could happen when there are not
enough vectors available for SFs.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 6bde18bcd42f..c72736f1571f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -394,7 +394,9 @@ struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev)
 		return ERR_PTR(-ENOMEM);
 	cpumask_copy(req_mask, cpu_online_mask);
 	if (!mlx5_irq_pool_is_sf_pool(pool)) {
-		/* In case we are allocating a control IRQ for PF/VF */
+		/* In case we are allocating a control IRQ from a pci device's pool.
+		 * This can happen also for a SF if the SFs pool is empty.
+		 */
 		if (!pool->xa_num_irqs.max) {
 			cpumask_clear(req_mask);
 			/* In case we only have a single IRQ for PF/VF */
-- 
cgit v1.2.3-70-g09d2


From 235a25fe28deb5558a500c9f9af4ff3b4797b86c Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Wed, 28 Dec 2022 16:30:52 +0200
Subject: net/mlx5: Modify struct mlx5_irq to use struct msi_map

Use the standard struct msi_map to store the vector number and irq
number pair in struct mlx5_irq.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 31 +++++++++++------------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index c72736f1571f..c79775d0ac24 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -29,8 +29,7 @@ struct mlx5_irq {
 	char name[MLX5_MAX_IRQ_NAME];
 	struct mlx5_irq_pool *pool;
 	int refcount;
-	u32 index;
-	int irqn;
+	struct msi_map map;
 };
 
 struct mlx5_irq_table {
@@ -128,14 +127,14 @@ static void irq_release(struct mlx5_irq *irq)
 {
 	struct mlx5_irq_pool *pool = irq->pool;
 
-	xa_erase(&pool->irqs, irq->index);
+	xa_erase(&pool->irqs, irq->map.index);
 	/* free_irq requires that affinity_hint and rmap will be cleared
 	 * before calling it. This is why there is asymmetry with set_rmap
 	 * which should be called after alloc_irq but before request_irq.
 	 */
-	irq_update_affinity_hint(irq->irqn, NULL);
+	irq_update_affinity_hint(irq->map.virq, NULL);
 	free_cpumask_var(irq->mask);
-	free_irq(irq->irqn, &irq->nh);
+	free_irq(irq->map.virq, &irq->nh);
 	kfree(irq);
 }
 
@@ -217,7 +216,7 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 	irq = kzalloc(sizeof(*irq), GFP_KERNEL);
 	if (!irq)
 		return ERR_PTR(-ENOMEM);
-	irq->irqn = pci_irq_vector(dev->pdev, i);
+	irq->map.virq = pci_irq_vector(dev->pdev, i);
 	if (!mlx5_irq_pool_is_sf_pool(pool))
 		irq_set_name(pool, name, i);
 	else
@@ -225,7 +224,7 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 	ATOMIC_INIT_NOTIFIER_HEAD(&irq->nh);
 	snprintf(irq->name, MLX5_MAX_IRQ_NAME,
 		 "%s@pci:%s", name, pci_name(dev->pdev));
-	err = request_irq(irq->irqn, irq_int_handler, 0, irq->name,
+	err = request_irq(irq->map.virq, irq_int_handler, 0, irq->name,
 			  &irq->nh);
 	if (err) {
 		mlx5_core_err(dev, "Failed to request irq. err = %d\n", err);
@@ -238,23 +237,23 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 	}
 	if (affinity) {
 		cpumask_copy(irq->mask, affinity);
-		irq_set_affinity_and_hint(irq->irqn, irq->mask);
+		irq_set_affinity_and_hint(irq->map.virq, irq->mask);
 	}
 	irq->pool = pool;
 	irq->refcount = 1;
-	irq->index = i;
-	err = xa_err(xa_store(&pool->irqs, irq->index, irq, GFP_KERNEL));
+	irq->map.index = i;
+	err = xa_err(xa_store(&pool->irqs, irq->map.index, irq, GFP_KERNEL));
 	if (err) {
 		mlx5_core_err(dev, "Failed to alloc xa entry for irq(%u). err = %d\n",
-			      irq->index, err);
+			      irq->map.index, err);
 		goto err_xa;
 	}
 	return irq;
 err_xa:
-	irq_update_affinity_hint(irq->irqn, NULL);
+	irq_update_affinity_hint(irq->map.virq, NULL);
 	free_cpumask_var(irq->mask);
 err_cpumask:
-	free_irq(irq->irqn, &irq->nh);
+	free_irq(irq->map.virq, &irq->nh);
 err_req_irq:
 	kfree(irq);
 	return ERR_PTR(err);
@@ -292,7 +291,7 @@ struct cpumask *mlx5_irq_get_affinity_mask(struct mlx5_irq *irq)
 
 int mlx5_irq_get_index(struct mlx5_irq *irq)
 {
-	return irq->index;
+	return irq->map.index;
 }
 
 /* irq_pool API */
@@ -364,7 +363,7 @@ static void mlx5_irqs_release(struct mlx5_irq **irqs, int nirqs)
 	int i;
 
 	for (i = 0; i < nirqs; i++) {
-		synchronize_irq(irqs[i]->irqn);
+		synchronize_irq(irqs[i]->map.virq);
 		mlx5_irq_put(irqs[i]);
 	}
 }
@@ -433,7 +432,7 @@ struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
 	if (IS_ERR(irq))
 		return irq;
 	mlx5_core_dbg(dev, "irq %u mapped to cpu %*pbl, %u EQs on this irq\n",
-		      irq->irqn, cpumask_pr_args(affinity),
+		      irq->map.virq, cpumask_pr_args(affinity),
 		      irq->refcount / MLX5_EQ_REFS_PER_IRQ);
 	return irq;
 }
-- 
cgit v1.2.3-70-g09d2


From bbac70c741830bd70382c8125bf34e1d9985c750 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Thu, 29 Dec 2022 11:02:19 +0200
Subject: net/mlx5: Use newer affinity descriptor

Use the more refined struct irq_affinity_desc to describe the required
IRQ affinity. For the async IRQs request unmanaged affinity and for
completion queues use managed affinity.

No functionality changes introduced. It will be used in a subsequent
patch when we use dynamic MSIX allocation.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 .../net/ethernet/mellanox/mlx5/core/irq_affinity.c | 39 ++++++++---------
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |  6 +--
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  | 49 ++++++++++------------
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h  |  2 +-
 4 files changed, 44 insertions(+), 52 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c b/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
index 380a208ab137..6535e8813178 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
@@ -45,30 +45,27 @@ static int cpu_get_least_loaded(struct mlx5_irq_pool *pool,
 
 /* Creating an IRQ from irq_pool */
 static struct mlx5_irq *
-irq_pool_request_irq(struct mlx5_irq_pool *pool, const struct cpumask *req_mask)
+irq_pool_request_irq(struct mlx5_irq_pool *pool, struct irq_affinity_desc *af_desc)
 {
-	cpumask_var_t auto_mask;
-	struct mlx5_irq *irq;
+	struct irq_affinity_desc auto_desc = {};
 	u32 irq_index;
 	int err;
 
-	if (!zalloc_cpumask_var(&auto_mask, GFP_KERNEL))
-		return ERR_PTR(-ENOMEM);
 	err = xa_alloc(&pool->irqs, &irq_index, NULL, pool->xa_num_irqs, GFP_KERNEL);
 	if (err)
 		return ERR_PTR(err);
 	if (pool->irqs_per_cpu) {
-		if (cpumask_weight(req_mask) > 1)
+		if (cpumask_weight(&af_desc->mask) > 1)
 			/* if req_mask contain more then one CPU, set the least loadad CPU
 			 * of req_mask
 			 */
-			cpumask_set_cpu(cpu_get_least_loaded(pool, req_mask), auto_mask);
+			cpumask_set_cpu(cpu_get_least_loaded(pool, &af_desc->mask),
+					&auto_desc.mask);
 		else
-			cpu_get(pool, cpumask_first(req_mask));
+			cpu_get(pool, cpumask_first(&af_desc->mask));
 	}
-	irq = mlx5_irq_alloc(pool, irq_index, cpumask_empty(auto_mask) ? req_mask : auto_mask);
-	free_cpumask_var(auto_mask);
-	return irq;
+	return mlx5_irq_alloc(pool, irq_index,
+			      cpumask_empty(&auto_desc.mask) ? af_desc : &auto_desc);
 }
 
 /* Looking for the IRQ with the smallest refcount that fits req_mask.
@@ -115,22 +112,22 @@ irq_pool_find_least_loaded(struct mlx5_irq_pool *pool, const struct cpumask *req
 /**
  * mlx5_irq_affinity_request - request an IRQ according to the given mask.
  * @pool: IRQ pool to request from.
- * @req_mask: cpumask requested for this IRQ.
+ * @af_desc: affinity descriptor for this IRQ.
  *
  * This function returns a pointer to IRQ, or ERR_PTR in case of error.
  */
 struct mlx5_irq *
-mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, const struct cpumask *req_mask)
+mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, struct irq_affinity_desc *af_desc)
 {
 	struct mlx5_irq *least_loaded_irq, *new_irq;
 
 	mutex_lock(&pool->lock);
-	least_loaded_irq = irq_pool_find_least_loaded(pool, req_mask);
+	least_loaded_irq = irq_pool_find_least_loaded(pool, &af_desc->mask);
 	if (least_loaded_irq &&
 	    mlx5_irq_read_locked(least_loaded_irq) < pool->min_threshold)
 		goto out;
 	/* We didn't find an IRQ with less than min_thres, try to allocate a new IRQ */
-	new_irq = irq_pool_request_irq(pool, req_mask);
+	new_irq = irq_pool_request_irq(pool, af_desc);
 	if (IS_ERR(new_irq)) {
 		if (!least_loaded_irq) {
 			/* We failed to create an IRQ and we didn't find an IRQ */
@@ -194,16 +191,15 @@ int mlx5_irq_affinity_irqs_request_auto(struct mlx5_core_dev *dev, int nirqs,
 					struct mlx5_irq **irqs)
 {
 	struct mlx5_irq_pool *pool = mlx5_irq_pool_get(dev);
-	cpumask_var_t req_mask;
+	struct irq_affinity_desc af_desc = {};
 	struct mlx5_irq *irq;
 	int i = 0;
 
-	if (!zalloc_cpumask_var(&req_mask, GFP_KERNEL))
-		return -ENOMEM;
-	cpumask_copy(req_mask, cpu_online_mask);
+	af_desc.is_managed = 1;
+	cpumask_copy(&af_desc.mask, cpu_online_mask);
 	for (i = 0; i < nirqs; i++) {
 		if (mlx5_irq_pool_is_sf_pool(pool))
-			irq = mlx5_irq_affinity_request(pool, req_mask);
+			irq = mlx5_irq_affinity_request(pool, &af_desc);
 		else
 			/* In case SF pool doesn't exists, fallback to the PF IRQs.
 			 * The PF IRQs are already allocated and binded to CPU
@@ -213,13 +209,12 @@ int mlx5_irq_affinity_irqs_request_auto(struct mlx5_core_dev *dev, int nirqs,
 		if (IS_ERR(irq))
 			break;
 		irqs[i] = irq;
-		cpumask_clear_cpu(cpumask_first(mlx5_irq_get_affinity_mask(irq)), req_mask);
+		cpumask_clear_cpu(cpumask_first(mlx5_irq_get_affinity_mask(irq)), &af_desc.mask);
 		mlx5_core_dbg(pool->dev, "IRQ %u mapped to cpu %*pbl, %u EQs on this irq\n",
 			      pci_irq_vector(dev->pdev, mlx5_irq_get_index(irq)),
 			      cpumask_pr_args(mlx5_irq_get_affinity_mask(irq)),
 			      mlx5_irq_read_locked(irq) / MLX5_EQ_REFS_PER_IRQ);
 	}
-	free_cpumask_var(req_mask);
 	if (!i)
 		return PTR_ERR(irq);
 	return i;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
index 23cb63fa4588..7e0dac74721e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
@@ -25,7 +25,7 @@ int mlx5_get_default_msix_vec_count(struct mlx5_core_dev *dev, int num_vfs);
 struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev);
 void mlx5_ctrl_irq_release(struct mlx5_irq *ctrl_irq);
 struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
-				  struct cpumask *affinity);
+				  struct irq_affinity_desc *af_desc);
 int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs,
 			      struct mlx5_irq **irqs);
 void mlx5_irqs_release_vectors(struct mlx5_irq **irqs, int nirqs);
@@ -39,7 +39,7 @@ struct mlx5_irq_pool;
 int mlx5_irq_affinity_irqs_request_auto(struct mlx5_core_dev *dev, int nirqs,
 					struct mlx5_irq **irqs);
 struct mlx5_irq *mlx5_irq_affinity_request(struct mlx5_irq_pool *pool,
-					   const struct cpumask *req_mask);
+					   struct irq_affinity_desc *af_desc);
 void mlx5_irq_affinity_irqs_release(struct mlx5_core_dev *dev, struct mlx5_irq **irqs,
 				    int num_irqs);
 #else
@@ -50,7 +50,7 @@ static inline int mlx5_irq_affinity_irqs_request_auto(struct mlx5_core_dev *dev,
 }
 
 static inline struct mlx5_irq *
-mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, const struct cpumask *req_mask)
+mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, struct irq_affinity_desc *af_desc)
 {
 	return ERR_PTR(-EOPNOTSUPP);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index c79775d0ac24..df6c2f8eb5df 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -206,7 +206,7 @@ static void irq_set_name(struct mlx5_irq_pool *pool, char *name, int vecidx)
 }
 
 struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
-				const struct cpumask *affinity)
+				struct irq_affinity_desc *af_desc)
 {
 	struct mlx5_core_dev *dev = pool->dev;
 	char name[MLX5_MAX_IRQ_NAME];
@@ -235,8 +235,8 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 		err = -ENOMEM;
 		goto err_cpumask;
 	}
-	if (affinity) {
-		cpumask_copy(irq->mask, affinity);
+	if (af_desc) {
+		cpumask_copy(irq->mask, &af_desc->mask);
 		irq_set_affinity_and_hint(irq->map.virq, irq->mask);
 	}
 	irq->pool = pool;
@@ -250,7 +250,8 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 	}
 	return irq;
 err_xa:
-	irq_update_affinity_hint(irq->map.virq, NULL);
+	if (af_desc)
+		irq_update_affinity_hint(irq->map.virq, NULL);
 	free_cpumask_var(irq->mask);
 err_cpumask:
 	free_irq(irq->map.virq, &irq->nh);
@@ -299,7 +300,7 @@ int mlx5_irq_get_index(struct mlx5_irq *irq)
 /* requesting an irq from a given pool according to given index */
 static struct mlx5_irq *
 irq_pool_request_vector(struct mlx5_irq_pool *pool, int vecidx,
-			struct cpumask *affinity)
+			struct irq_affinity_desc *af_desc)
 {
 	struct mlx5_irq *irq;
 
@@ -309,7 +310,7 @@ irq_pool_request_vector(struct mlx5_irq_pool *pool, int vecidx,
 		mlx5_irq_get_locked(irq);
 		goto unlock;
 	}
-	irq = mlx5_irq_alloc(pool, vecidx, affinity);
+	irq = mlx5_irq_alloc(pool, vecidx, af_desc);
 unlock:
 	mutex_unlock(&pool->lock);
 	return irq;
@@ -386,28 +387,26 @@ void mlx5_ctrl_irq_release(struct mlx5_irq *ctrl_irq)
 struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev)
 {
 	struct mlx5_irq_pool *pool = ctrl_irq_pool_get(dev);
-	cpumask_var_t req_mask;
+	struct irq_affinity_desc af_desc;
 	struct mlx5_irq *irq;
 
-	if (!zalloc_cpumask_var(&req_mask, GFP_KERNEL))
-		return ERR_PTR(-ENOMEM);
-	cpumask_copy(req_mask, cpu_online_mask);
+	cpumask_copy(&af_desc.mask, cpu_online_mask);
+	af_desc.is_managed = false;
 	if (!mlx5_irq_pool_is_sf_pool(pool)) {
 		/* In case we are allocating a control IRQ from a pci device's pool.
 		 * This can happen also for a SF if the SFs pool is empty.
 		 */
 		if (!pool->xa_num_irqs.max) {
-			cpumask_clear(req_mask);
+			cpumask_clear(&af_desc.mask);
 			/* In case we only have a single IRQ for PF/VF */
-			cpumask_set_cpu(cpumask_first(cpu_online_mask), req_mask);
+			cpumask_set_cpu(cpumask_first(cpu_online_mask), &af_desc.mask);
 		}
 		/* Allocate the IRQ in the last index of the pool */
-		irq = irq_pool_request_vector(pool, pool->xa_num_irqs.max, req_mask);
+		irq = irq_pool_request_vector(pool, pool->xa_num_irqs.max, &af_desc);
 	} else {
-		irq = mlx5_irq_affinity_request(pool, req_mask);
+		irq = mlx5_irq_affinity_request(pool, &af_desc);
 	}
 
-	free_cpumask_var(req_mask);
 	return irq;
 }
 
@@ -416,23 +415,23 @@ struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev)
  * @dev: mlx5 device that requesting the IRQ.
  * @vecidx: vector index of the IRQ. This argument is ignore if affinity is
  * provided.
- * @affinity: cpumask requested for this IRQ.
+ * @af_desc: affinity descriptor for this IRQ.
  *
  * This function returns a pointer to IRQ, or ERR_PTR in case of error.
  */
 struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
-				  struct cpumask *affinity)
+				  struct irq_affinity_desc *af_desc)
 {
 	struct mlx5_irq_table *irq_table = mlx5_irq_table_get(dev);
 	struct mlx5_irq_pool *pool;
 	struct mlx5_irq *irq;
 
 	pool = irq_table->pf_pool;
-	irq = irq_pool_request_vector(pool, vecidx, affinity);
+	irq = irq_pool_request_vector(pool, vecidx, af_desc);
 	if (IS_ERR(irq))
 		return irq;
 	mlx5_core_dbg(dev, "irq %u mapped to cpu %*pbl, %u EQs on this irq\n",
-		      irq->map.virq, cpumask_pr_args(affinity),
+		      irq->map.virq, cpumask_pr_args(&af_desc->mask),
 		      irq->refcount / MLX5_EQ_REFS_PER_IRQ);
 	return irq;
 }
@@ -463,22 +462,20 @@ void mlx5_irqs_release_vectors(struct mlx5_irq **irqs, int nirqs)
 int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs,
 			      struct mlx5_irq **irqs)
 {
-	cpumask_var_t req_mask;
+	struct irq_affinity_desc af_desc;
 	struct mlx5_irq *irq;
 	int i;
 
-	if (!zalloc_cpumask_var(&req_mask, GFP_KERNEL))
-		return -ENOMEM;
+	af_desc.is_managed = 1;
 	for (i = 0; i < nirqs; i++) {
-		cpumask_set_cpu(cpus[i], req_mask);
-		irq = mlx5_irq_request(dev, i, req_mask);
+		cpumask_set_cpu(cpus[i], &af_desc.mask);
+		irq = mlx5_irq_request(dev, i, &af_desc);
 		if (IS_ERR(irq))
 			break;
-		cpumask_clear(req_mask);
+		cpumask_clear(&af_desc.mask);
 		irqs[i] = irq;
 	}
 
-	free_cpumask_var(req_mask);
 	return i ? i : PTR_ERR(irq);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h
index 5c7e68bee43a..ea39e4027d8a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h
@@ -31,7 +31,7 @@ static inline bool mlx5_irq_pool_is_sf_pool(struct mlx5_irq_pool *pool)
 }
 
 struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
-				const struct cpumask *affinity);
+				struct irq_affinity_desc *af_desc);
 int mlx5_irq_get_locked(struct mlx5_irq *irq);
 int mlx5_irq_read_locked(struct mlx5_irq *irq);
 int mlx5_irq_put(struct mlx5_irq *irq);
-- 
cgit v1.2.3-70-g09d2


From 8bebfd767909eb884a7b529fb7a263e40c99627b Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Sun, 1 Jan 2023 09:22:34 +0200
Subject: net/mlx5: Improve naming of pci function vectors

The variable pf_vec is used to denote the number of vectors required for
the pci function's own use. To avoid confusion interpreting pf as
physical function, change the name to pcif_vec.

Same reasoning goes for pf_pool which is really pci function pool.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 46 +++++++++++------------
 1 file changed, 23 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index df6c2f8eb5df..c8e2b1ac7fe5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -33,7 +33,7 @@ struct mlx5_irq {
 };
 
 struct mlx5_irq_table {
-	struct mlx5_irq_pool *pf_pool;
+	struct mlx5_irq_pool *pcif_pool;
 	struct mlx5_irq_pool *sf_ctrl_pool;
 	struct mlx5_irq_pool *sf_comp_pool;
 };
@@ -337,7 +337,7 @@ struct mlx5_irq_pool *mlx5_irq_pool_get(struct mlx5_core_dev *dev)
 	/* In some configs, there won't be a pool of SFs IRQs. Hence, returning
 	 * the PF IRQs pool in case the SF pool doesn't exist.
 	 */
-	return pool ? pool : irq_table->pf_pool;
+	return pool ? pool : irq_table->pcif_pool;
 }
 
 static struct mlx5_irq_pool *ctrl_irq_pool_get(struct mlx5_core_dev *dev)
@@ -351,7 +351,7 @@ static struct mlx5_irq_pool *ctrl_irq_pool_get(struct mlx5_core_dev *dev)
 	/* In some configs, there won't be a pool of SFs IRQs. Hence, returning
 	 * the PF IRQs pool in case the SF pool doesn't exist.
 	 */
-	return pool ? pool : irq_table->pf_pool;
+	return pool ? pool : irq_table->pcif_pool;
 }
 
 /**
@@ -426,7 +426,7 @@ struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
 	struct mlx5_irq_pool *pool;
 	struct mlx5_irq *irq;
 
-	pool = irq_table->pf_pool;
+	pool = irq_table->pcif_pool;
 	irq = irq_pool_request_vector(pool, vecidx, af_desc);
 	if (IS_ERR(irq))
 		return irq;
@@ -519,7 +519,7 @@ static void irq_pool_free(struct mlx5_irq_pool *pool)
 	kvfree(pool);
 }
 
-static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pf_vec)
+static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec)
 {
 	struct mlx5_irq_table *table = dev->priv.irq_table;
 	int num_sf_ctrl_by_msix;
@@ -527,12 +527,12 @@ static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pf_vec)
 	int num_sf_ctrl;
 	int err;
 
-	/* init pf_pool */
-	table->pf_pool = irq_pool_alloc(dev, 0, pf_vec, NULL,
-					MLX5_EQ_SHARE_IRQ_MIN_COMP,
-					MLX5_EQ_SHARE_IRQ_MAX_COMP);
-	if (IS_ERR(table->pf_pool))
-		return PTR_ERR(table->pf_pool);
+	/* init pcif_pool */
+	table->pcif_pool = irq_pool_alloc(dev, 0, pcif_vec, NULL,
+					  MLX5_EQ_SHARE_IRQ_MIN_COMP,
+					  MLX5_EQ_SHARE_IRQ_MAX_COMP);
+	if (IS_ERR(table->pcif_pool))
+		return PTR_ERR(table->pcif_pool);
 	if (!mlx5_sf_max_functions(dev))
 		return 0;
 	if (sf_vec < MLX5_IRQ_VEC_COMP_BASE_SF) {
@@ -546,7 +546,7 @@ static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pf_vec)
 					  MLX5_SFS_PER_CTRL_IRQ);
 	num_sf_ctrl = min_t(int, num_sf_ctrl_by_msix, num_sf_ctrl_by_sfs);
 	num_sf_ctrl = min_t(int, MLX5_IRQ_CTRL_SF_MAX, num_sf_ctrl);
-	table->sf_ctrl_pool = irq_pool_alloc(dev, pf_vec, num_sf_ctrl,
+	table->sf_ctrl_pool = irq_pool_alloc(dev, pcif_vec, num_sf_ctrl,
 					     "mlx5_sf_ctrl",
 					     MLX5_EQ_SHARE_IRQ_MIN_CTRL,
 					     MLX5_EQ_SHARE_IRQ_MAX_CTRL);
@@ -555,7 +555,7 @@ static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pf_vec)
 		goto err_pf;
 	}
 	/* init sf_comp_pool */
-	table->sf_comp_pool = irq_pool_alloc(dev, pf_vec + num_sf_ctrl,
+	table->sf_comp_pool = irq_pool_alloc(dev, pcif_vec + num_sf_ctrl,
 					     sf_vec - num_sf_ctrl, "mlx5_sf_comp",
 					     MLX5_EQ_SHARE_IRQ_MIN_COMP,
 					     MLX5_EQ_SHARE_IRQ_MAX_COMP);
@@ -577,7 +577,7 @@ err_irqs_per_cpu:
 err_sf_ctrl:
 	irq_pool_free(table->sf_ctrl_pool);
 err_pf:
-	irq_pool_free(table->pf_pool);
+	irq_pool_free(table->pcif_pool);
 	return err;
 }
 
@@ -587,7 +587,7 @@ static void irq_pools_destroy(struct mlx5_irq_table *table)
 		irq_pool_free(table->sf_comp_pool);
 		irq_pool_free(table->sf_ctrl_pool);
 	}
-	irq_pool_free(table->pf_pool);
+	irq_pool_free(table->pcif_pool);
 }
 
 /* irq_table API */
@@ -618,9 +618,9 @@ void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev)
 
 int mlx5_irq_table_get_num_comp(struct mlx5_irq_table *table)
 {
-	if (!table->pf_pool->xa_num_irqs.max)
+	if (!table->pcif_pool->xa_num_irqs.max)
 		return 1;
-	return table->pf_pool->xa_num_irqs.max - table->pf_pool->xa_num_irqs.min;
+	return table->pcif_pool->xa_num_irqs.max - table->pcif_pool->xa_num_irqs.min;
 }
 
 int mlx5_irq_table_create(struct mlx5_core_dev *dev)
@@ -629,16 +629,16 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 		      MLX5_CAP_GEN(dev, max_num_eqs) :
 		      1 << MLX5_CAP_GEN(dev, log_max_eq);
 	int total_vec;
-	int pf_vec;
+	int pcif_vec;
 	int err;
 
 	if (mlx5_core_is_sf(dev))
 		return 0;
 
-	pf_vec = MLX5_CAP_GEN(dev, num_ports) * num_online_cpus() + 1;
-	pf_vec = min_t(int, pf_vec, num_eqs);
+	pcif_vec = MLX5_CAP_GEN(dev, num_ports) * num_online_cpus() + 1;
+	pcif_vec = min_t(int, pcif_vec, num_eqs);
 
-	total_vec = pf_vec;
+	total_vec = pcif_vec;
 	if (mlx5_sf_max_functions(dev))
 		total_vec += MLX5_IRQ_CTRL_SF_MAX +
 			MLX5_COMP_EQS_PER_SF * mlx5_sf_max_functions(dev);
@@ -646,9 +646,9 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 	total_vec = pci_alloc_irq_vectors(dev->pdev, 1, total_vec, PCI_IRQ_MSIX);
 	if (total_vec < 0)
 		return total_vec;
-	pf_vec = min(pf_vec, total_vec);
+	pcif_vec = min(pcif_vec, total_vec);
 
-	err = irq_pools_init(dev, total_vec - pf_vec, pf_vec);
+	err = irq_pools_init(dev, total_vec - pcif_vec, pcif_vec);
 	if (err)
 		pci_free_irq_vectors(dev->pdev);
 
-- 
cgit v1.2.3-70-g09d2


From b48a0f72bc3e5b10f55b3d40f051135fc9afd0e5 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Mon, 9 Jan 2023 12:08:53 +0200
Subject: net/mlx5: Refactor completion irq request/release code

Break the request and release functions into pci and sub-functions
devices handling for better readability, eventually making the code
symmetric in terms of request/release.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 70 ++++++++++++++++++----------
 1 file changed, 46 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index b43121f64a80..ed75b527280e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -804,44 +804,28 @@ void mlx5_eq_update_ci(struct mlx5_eq *eq, u32 cc, bool arm)
 }
 EXPORT_SYMBOL(mlx5_eq_update_ci);
 
-static void comp_irqs_release(struct mlx5_core_dev *dev)
+static void comp_irqs_release_pci(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
 
-	if (mlx5_core_is_sf(dev))
-		mlx5_irq_affinity_irqs_release(dev, table->comp_irqs, table->num_comp_eqs);
-	else
-		mlx5_irqs_release_vectors(table->comp_irqs, table->num_comp_eqs);
-	kfree(table->comp_irqs);
+	mlx5_irqs_release_vectors(table->comp_irqs, table->num_comp_eqs);
 }
 
-static int comp_irqs_request(struct mlx5_core_dev *dev)
+static int comp_irqs_request_pci(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
 	const struct cpumask *prev = cpu_none_mask;
 	const struct cpumask *mask;
-	int ncomp_eqs = table->num_comp_eqs;
+	int ncomp_eqs;
 	u16 *cpus;
 	int ret;
 	int cpu;
 	int i;
 
 	ncomp_eqs = table->num_comp_eqs;
-	table->comp_irqs = kcalloc(ncomp_eqs, sizeof(*table->comp_irqs), GFP_KERNEL);
-	if (!table->comp_irqs)
-		return -ENOMEM;
-	if (mlx5_core_is_sf(dev)) {
-		ret = mlx5_irq_affinity_irqs_request_auto(dev, ncomp_eqs, table->comp_irqs);
-		if (ret < 0)
-			goto free_irqs;
-		return ret;
-	}
-
 	cpus = kcalloc(ncomp_eqs, sizeof(*cpus), GFP_KERNEL);
-	if (!cpus) {
+	if (!cpus)
 		ret = -ENOMEM;
-		goto free_irqs;
-	}
 
 	i = 0;
 	rcu_read_lock();
@@ -857,12 +841,50 @@ spread_done:
 	rcu_read_unlock();
 	ret = mlx5_irqs_request_vectors(dev, cpus, ncomp_eqs, table->comp_irqs);
 	kfree(cpus);
-	if (ret < 0)
-		goto free_irqs;
 	return ret;
+}
+
+static void comp_irqs_release_sf(struct mlx5_core_dev *dev)
+{
+	struct mlx5_eq_table *table = dev->priv.eq_table;
+
+	mlx5_irq_affinity_irqs_release(dev, table->comp_irqs, table->num_comp_eqs);
+}
+
+static int comp_irqs_request_sf(struct mlx5_core_dev *dev)
+{
+	struct mlx5_eq_table *table = dev->priv.eq_table;
+	int ncomp_eqs = table->num_comp_eqs;
+
+	return mlx5_irq_affinity_irqs_request_auto(dev, ncomp_eqs, table->comp_irqs);
+}
+
+static void comp_irqs_release(struct mlx5_core_dev *dev)
+{
+	struct mlx5_eq_table *table = dev->priv.eq_table;
+
+	mlx5_core_is_sf(dev) ? comp_irqs_release_sf(dev) :
+			       comp_irqs_release_pci(dev);
 
-free_irqs:
 	kfree(table->comp_irqs);
+}
+
+static int comp_irqs_request(struct mlx5_core_dev *dev)
+{
+	struct mlx5_eq_table *table = dev->priv.eq_table;
+	int ncomp_eqs;
+	int ret;
+
+	ncomp_eqs = table->num_comp_eqs;
+	table->comp_irqs = kcalloc(ncomp_eqs, sizeof(*table->comp_irqs), GFP_KERNEL);
+	if (!table->comp_irqs)
+		return -ENOMEM;
+
+	ret = mlx5_core_is_sf(dev) ? comp_irqs_request_sf(dev) :
+				     comp_irqs_request_pci(dev);
+	if (ret < 0)
+		kfree(table->comp_irqs);
+
 	return ret;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 3354822cde5a9f72aa725b3c619188b149a71a33 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Sun, 1 Jan 2023 08:16:23 +0200
Subject: net/mlx5: Use dynamic msix vectors allocation

Current implementation calculates the number and the partitioaning of
available interrupts vectors and then allocates all the interrupt
vectors.

Here, whenever dynamic msix allocation is supported, we change this to
use msix vectors dynamically so a vectors is actually allocated only
when needed. The current pool logic is kept in place to take care of
partitioning the vectors between the consumers and take care of
reference counting. However, the vectors are allocated only when needed.

Subsequent patches will make use of this to allocate vectors for VDPA.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       | 114 +++++++++------------
 .../net/ethernet/mellanox/mlx5/core/irq_affinity.c |   5 +-
 drivers/net/ethernet/mellanox/mlx5/core/main.c     |   3 -
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |   6 +-
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  |  88 ++++++++++++----
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h  |   4 +-
 6 files changed, 129 insertions(+), 91 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index ed75b527280e..888ccfcbcce6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -18,6 +18,7 @@
 #include "lib/clock.h"
 #include "diag/fw_tracer.h"
 #include "mlx5_irq.h"
+#include "pci_irq.h"
 #include "devlink.h"
 #include "en_accel/ipsec.h"
 
@@ -61,9 +62,7 @@ struct mlx5_eq_table {
 	struct mlx5_irq_table	*irq_table;
 	struct mlx5_irq         **comp_irqs;
 	struct mlx5_irq         *ctrl_irq;
-#ifdef CONFIG_RFS_ACCEL
 	struct cpu_rmap		*rmap;
-#endif
 };
 
 #define MLX5_ASYNC_EVENT_MASK ((1ull << MLX5_EVENT_TYPE_PATH_MIG)	    | \
@@ -839,7 +838,7 @@ static int comp_irqs_request_pci(struct mlx5_core_dev *dev)
 	}
 spread_done:
 	rcu_read_unlock();
-	ret = mlx5_irqs_request_vectors(dev, cpus, ncomp_eqs, table->comp_irqs);
+	ret = mlx5_irqs_request_vectors(dev, cpus, ncomp_eqs, table->comp_irqs, &table->rmap);
 	kfree(cpus);
 	return ret;
 }
@@ -888,6 +887,40 @@ static int comp_irqs_request(struct mlx5_core_dev *dev)
 	return ret;
 }
 
+#ifdef CONFIG_RFS_ACCEL
+static int alloc_rmap(struct mlx5_core_dev *mdev)
+{
+	struct mlx5_eq_table *eq_table = mdev->priv.eq_table;
+
+	/* rmap is a mapping between irq number and queue number.
+	 * Each irq can be assigned only to a single rmap.
+	 * Since SFs share IRQs, rmap mapping cannot function correctly
+	 * for irqs that are shared between different core/netdev RX rings.
+	 * Hence we don't allow netdev rmap for SFs.
+	 */
+	if (mlx5_core_is_sf(mdev))
+		return 0;
+
+	eq_table->rmap = alloc_irq_cpu_rmap(eq_table->num_comp_eqs);
+	if (!eq_table->rmap)
+		return -ENOMEM;
+	return 0;
+}
+
+static void free_rmap(struct mlx5_core_dev *mdev)
+{
+	struct mlx5_eq_table *eq_table = mdev->priv.eq_table;
+
+	if (eq_table->rmap) {
+		free_irq_cpu_rmap(eq_table->rmap);
+		eq_table->rmap = NULL;
+	}
+}
+#else
+static int alloc_rmap(struct mlx5_core_dev *mdev) { return 0; }
+static void free_rmap(struct mlx5_core_dev *mdev) {}
+#endif
+
 static void destroy_comp_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
@@ -903,6 +936,7 @@ static void destroy_comp_eqs(struct mlx5_core_dev *dev)
 		kfree(eq);
 	}
 	comp_irqs_release(dev);
+	free_rmap(dev);
 }
 
 static u16 comp_eq_depth_devlink_param_get(struct mlx5_core_dev *dev)
@@ -929,9 +963,16 @@ static int create_comp_eqs(struct mlx5_core_dev *dev)
 	int err;
 	int i;
 
+	err = alloc_rmap(dev);
+	if (err)
+		return err;
+
 	ncomp_eqs = comp_irqs_request(dev);
-	if (ncomp_eqs < 0)
-		return ncomp_eqs;
+	if (ncomp_eqs < 0) {
+		err = ncomp_eqs;
+		goto err_irqs_req;
+	}
+
 	INIT_LIST_HEAD(&table->comp_eqs_list);
 	nent = comp_eq_depth_devlink_param_get(dev);
 
@@ -976,6 +1017,8 @@ clean_eq:
 	kfree(eq);
 clean:
 	destroy_comp_eqs(dev);
+err_irqs_req:
+	free_rmap(dev);
 	return err;
 }
 
@@ -1054,55 +1097,12 @@ struct mlx5_eq_comp *mlx5_eqn2comp_eq(struct mlx5_core_dev *dev, int eqn)
 	return ERR_PTR(-ENOENT);
 }
 
-static void clear_rmap(struct mlx5_core_dev *dev)
-{
-#ifdef CONFIG_RFS_ACCEL
-	struct mlx5_eq_table *eq_table = dev->priv.eq_table;
-
-	free_irq_cpu_rmap(eq_table->rmap);
-#endif
-}
-
-static int set_rmap(struct mlx5_core_dev *mdev)
-{
-	int err = 0;
-#ifdef CONFIG_RFS_ACCEL
-	struct mlx5_eq_table *eq_table = mdev->priv.eq_table;
-	int vecidx;
-
-	eq_table->rmap = alloc_irq_cpu_rmap(eq_table->num_comp_eqs);
-	if (!eq_table->rmap) {
-		err = -ENOMEM;
-		mlx5_core_err(mdev, "Failed to allocate cpu_rmap. err %d", err);
-		goto err_out;
-	}
-
-	for (vecidx = 0; vecidx < eq_table->num_comp_eqs; vecidx++) {
-		err = irq_cpu_rmap_add(eq_table->rmap,
-				       pci_irq_vector(mdev->pdev, vecidx));
-		if (err) {
-			mlx5_core_err(mdev, "irq_cpu_rmap_add failed. err %d",
-				      err);
-			goto err_irq_cpu_rmap_add;
-		}
-	}
-	return 0;
-
-err_irq_cpu_rmap_add:
-	clear_rmap(mdev);
-err_out:
-#endif
-	return err;
-}
-
 /* This function should only be called after mlx5_cmd_force_teardown_hca */
 void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
 
 	mutex_lock(&table->lock); /* sync with create/destroy_async_eq */
-	if (!mlx5_core_is_sf(dev))
-		clear_rmap(dev);
 	mlx5_irq_table_destroy(dev);
 	mutex_unlock(&table->lock);
 }
@@ -1139,18 +1139,6 @@ int mlx5_eq_table_create(struct mlx5_core_dev *dev)
 		goto err_async_eqs;
 	}
 
-	if (!mlx5_core_is_sf(dev)) {
-		/* rmap is a mapping between irq number and queue number.
-		 * each irq can be assign only to a single rmap.
-		 * since SFs share IRQs, rmap mapping cannot function correctly
-		 * for irqs that are shared for different core/netdev RX rings.
-		 * Hence we don't allow netdev rmap for SFs
-		 */
-		err = set_rmap(dev);
-		if (err)
-			goto err_rmap;
-	}
-
 	err = create_comp_eqs(dev);
 	if (err) {
 		mlx5_core_err(dev, "Failed to create completion EQs\n");
@@ -1158,10 +1146,8 @@ int mlx5_eq_table_create(struct mlx5_core_dev *dev)
 	}
 
 	return 0;
+
 err_comp_eqs:
-	if (!mlx5_core_is_sf(dev))
-		clear_rmap(dev);
-err_rmap:
 	destroy_async_eqs(dev);
 err_async_eqs:
 	return err;
@@ -1169,8 +1155,6 @@ err_async_eqs:
 
 void mlx5_eq_table_destroy(struct mlx5_core_dev *dev)
 {
-	if (!mlx5_core_is_sf(dev))
-		clear_rmap(dev);
 	destroy_comp_eqs(dev);
 	destroy_async_eqs(dev);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c b/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
index 6535e8813178..fa467335526e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
@@ -65,7 +65,8 @@ irq_pool_request_irq(struct mlx5_irq_pool *pool, struct irq_affinity_desc *af_de
 			cpu_get(pool, cpumask_first(&af_desc->mask));
 	}
 	return mlx5_irq_alloc(pool, irq_index,
-			      cpumask_empty(&auto_desc.mask) ? af_desc : &auto_desc);
+			      cpumask_empty(&auto_desc.mask) ? af_desc : &auto_desc,
+			      NULL);
 }
 
 /* Looking for the IRQ with the smallest refcount that fits req_mask.
@@ -205,7 +206,7 @@ int mlx5_irq_affinity_irqs_request_auto(struct mlx5_core_dev *dev, int nirqs,
 			 * The PF IRQs are already allocated and binded to CPU
 			 * at this point. Hence, only an index is needed.
 			 */
-			irq = mlx5_irq_request(dev, i, NULL);
+			irq = mlx5_irq_request(dev, i, NULL, NULL);
 		if (IS_ERR(irq))
 			break;
 		irqs[i] = irq;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index f1de152a6113..bbc9b4188212 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -46,9 +46,6 @@
 #include <linux/kmod.h>
 #include <linux/mlx5/mlx5_ifc.h>
 #include <linux/mlx5/vport.h>
-#ifdef CONFIG_RFS_ACCEL
-#include <linux/cpu_rmap.h>
-#endif
 #include <linux/version.h>
 #include <net/devlink.h>
 #include "mlx5_core.h"
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
index 7e0dac74721e..efd0c299c5c7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
@@ -9,6 +9,7 @@
 #define MLX5_COMP_EQS_PER_SF 8
 
 struct mlx5_irq;
+struct cpu_rmap;
 
 int mlx5_irq_table_init(struct mlx5_core_dev *dev);
 void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev);
@@ -25,9 +26,10 @@ int mlx5_get_default_msix_vec_count(struct mlx5_core_dev *dev, int num_vfs);
 struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev);
 void mlx5_ctrl_irq_release(struct mlx5_irq *ctrl_irq);
 struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
-				  struct irq_affinity_desc *af_desc);
+				  struct irq_affinity_desc *af_desc,
+				  struct cpu_rmap **rmap);
 int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs,
-			      struct mlx5_irq **irqs);
+			      struct mlx5_irq **irqs, struct cpu_rmap **rmap);
 void mlx5_irqs_release_vectors(struct mlx5_irq **irqs, int nirqs);
 int mlx5_irq_attach_nb(struct mlx5_irq *irq, struct notifier_block *nb);
 int mlx5_irq_detach_nb(struct mlx5_irq *irq, struct notifier_block *nb);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index c8e2b1ac7fe5..7fa63d31ae5b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -9,6 +9,7 @@
 #include "mlx5_irq.h"
 #include "pci_irq.h"
 #include "lib/sf.h"
+#include "lib/eq.h"
 #ifdef CONFIG_RFS_ACCEL
 #include <linux/cpu_rmap.h>
 #endif
@@ -126,15 +127,26 @@ out:
 static void irq_release(struct mlx5_irq *irq)
 {
 	struct mlx5_irq_pool *pool = irq->pool;
+#ifdef CONFIG_RFS_ACCEL
+	struct cpu_rmap *rmap;
+#endif
 
 	xa_erase(&pool->irqs, irq->map.index);
-	/* free_irq requires that affinity_hint and rmap will be cleared
-	 * before calling it. This is why there is asymmetry with set_rmap
-	 * which should be called after alloc_irq but before request_irq.
+	/* free_irq requires that affinity_hint and rmap will be cleared before
+	 * calling it. To satisfy this requirement, we call
+	 * irq_cpu_rmap_remove() to remove the notifier
 	 */
 	irq_update_affinity_hint(irq->map.virq, NULL);
+#ifdef CONFIG_RFS_ACCEL
+	rmap = mlx5_eq_table_get_rmap(pool->dev);
+	if (rmap && irq->map.index)
+		irq_cpu_rmap_remove(rmap, irq->map.virq);
+#endif
+
 	free_cpumask_var(irq->mask);
 	free_irq(irq->map.virq, &irq->nh);
+	if (irq->map.index && pci_msix_can_alloc_dyn(pool->dev->pdev))
+		pci_msix_free_irq(pool->dev->pdev, irq->map);
 	kfree(irq);
 }
 
@@ -197,7 +209,7 @@ static void irq_set_name(struct mlx5_irq_pool *pool, char *name, int vecidx)
 		return;
 	}
 
-	if (vecidx == pool->xa_num_irqs.max) {
+	if (!vecidx) {
 		snprintf(name, MLX5_MAX_IRQ_NAME, "mlx5_async%d", vecidx);
 		return;
 	}
@@ -206,7 +218,8 @@ static void irq_set_name(struct mlx5_irq_pool *pool, char *name, int vecidx)
 }
 
 struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
-				struct irq_affinity_desc *af_desc)
+				struct irq_affinity_desc *af_desc,
+				struct cpu_rmap **rmap)
 {
 	struct mlx5_core_dev *dev = pool->dev;
 	char name[MLX5_MAX_IRQ_NAME];
@@ -216,7 +229,28 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 	irq = kzalloc(sizeof(*irq), GFP_KERNEL);
 	if (!irq)
 		return ERR_PTR(-ENOMEM);
-	irq->map.virq = pci_irq_vector(dev->pdev, i);
+	if (!i || !pci_msix_can_alloc_dyn(dev->pdev)) {
+		/* The vector at index 0 was already allocated.
+		 * Just get the irq number. If dynamic irq is not supported
+		 * vectors have also been allocated.
+		 */
+		irq->map.virq = pci_irq_vector(dev->pdev, i);
+		irq->map.index = 0;
+	} else {
+		irq->map = pci_msix_alloc_irq_at(dev->pdev, MSI_ANY_INDEX, af_desc);
+		if (!irq->map.virq) {
+			err = irq->map.index;
+			goto err_alloc_irq;
+		}
+	}
+
+	if (i && rmap && *rmap) {
+#ifdef CONFIG_RFS_ACCEL
+		err = irq_cpu_rmap_add(*rmap, irq->map.virq);
+		if (err)
+			goto err_irq_rmap;
+#endif
+	}
 	if (!mlx5_irq_pool_is_sf_pool(pool))
 		irq_set_name(pool, name, i);
 	else
@@ -256,6 +290,16 @@ err_xa:
 err_cpumask:
 	free_irq(irq->map.virq, &irq->nh);
 err_req_irq:
+#ifdef CONFIG_RFS_ACCEL
+	if (i && rmap && *rmap) {
+		free_irq_cpu_rmap(*rmap);
+		*rmap = NULL;
+	}
+err_irq_rmap:
+#endif
+	if (i && pci_msix_can_alloc_dyn(dev->pdev))
+		pci_msix_free_irq(dev->pdev, irq->map);
+err_alloc_irq:
 	kfree(irq);
 	return ERR_PTR(err);
 }
@@ -300,7 +344,8 @@ int mlx5_irq_get_index(struct mlx5_irq *irq)
 /* requesting an irq from a given pool according to given index */
 static struct mlx5_irq *
 irq_pool_request_vector(struct mlx5_irq_pool *pool, int vecidx,
-			struct irq_affinity_desc *af_desc)
+			struct irq_affinity_desc *af_desc,
+			struct cpu_rmap **rmap)
 {
 	struct mlx5_irq *irq;
 
@@ -310,7 +355,7 @@ irq_pool_request_vector(struct mlx5_irq_pool *pool, int vecidx,
 		mlx5_irq_get_locked(irq);
 		goto unlock;
 	}
-	irq = mlx5_irq_alloc(pool, vecidx, af_desc);
+	irq = mlx5_irq_alloc(pool, vecidx, af_desc, rmap);
 unlock:
 	mutex_unlock(&pool->lock);
 	return irq;
@@ -401,8 +446,8 @@ struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev)
 			/* In case we only have a single IRQ for PF/VF */
 			cpumask_set_cpu(cpumask_first(cpu_online_mask), &af_desc.mask);
 		}
-		/* Allocate the IRQ in the last index of the pool */
-		irq = irq_pool_request_vector(pool, pool->xa_num_irqs.max, &af_desc);
+		/* Allocate the IRQ in index 0. The vector was already allocated */
+		irq = irq_pool_request_vector(pool, 0, &af_desc, NULL);
 	} else {
 		irq = mlx5_irq_affinity_request(pool, &af_desc);
 	}
@@ -416,18 +461,20 @@ struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev)
  * @vecidx: vector index of the IRQ. This argument is ignore if affinity is
  * provided.
  * @af_desc: affinity descriptor for this IRQ.
+ * @rmap: pointer to reverse map pointer for completion interrupts
  *
  * This function returns a pointer to IRQ, or ERR_PTR in case of error.
  */
 struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
-				  struct irq_affinity_desc *af_desc)
+				  struct irq_affinity_desc *af_desc,
+				  struct cpu_rmap **rmap)
 {
 	struct mlx5_irq_table *irq_table = mlx5_irq_table_get(dev);
 	struct mlx5_irq_pool *pool;
 	struct mlx5_irq *irq;
 
 	pool = irq_table->pcif_pool;
-	irq = irq_pool_request_vector(pool, vecidx, af_desc);
+	irq = irq_pool_request_vector(pool, vecidx, af_desc, rmap);
 	if (IS_ERR(irq))
 		return irq;
 	mlx5_core_dbg(dev, "irq %u mapped to cpu %*pbl, %u EQs on this irq\n",
@@ -452,6 +499,7 @@ void mlx5_irqs_release_vectors(struct mlx5_irq **irqs, int nirqs)
  * @cpus: CPUs array for binding the IRQs
  * @nirqs: number of IRQs to request.
  * @irqs: an output array of IRQs pointers.
+ * @rmap: pointer to reverse map pointer for completion interrupts
  *
  * Each IRQ is bound to at most 1 CPU.
  * This function is requests nirqs IRQs, starting from @vecidx.
@@ -460,7 +508,7 @@ void mlx5_irqs_release_vectors(struct mlx5_irq **irqs, int nirqs)
  * @nirqs), if successful, or a negative error code in case of an error.
  */
 int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs,
-			      struct mlx5_irq **irqs)
+			      struct mlx5_irq **irqs, struct cpu_rmap **rmap)
 {
 	struct irq_affinity_desc af_desc;
 	struct mlx5_irq *irq;
@@ -469,7 +517,7 @@ int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs,
 	af_desc.is_managed = 1;
 	for (i = 0; i < nirqs; i++) {
 		cpumask_set_cpu(cpus[i], &af_desc.mask);
-		irq = mlx5_irq_request(dev, i, &af_desc);
+		irq = mlx5_irq_request(dev, i + 1, &af_desc, rmap);
 		if (IS_ERR(irq))
 			break;
 		cpumask_clear(&af_desc.mask);
@@ -630,7 +678,9 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 		      1 << MLX5_CAP_GEN(dev, log_max_eq);
 	int total_vec;
 	int pcif_vec;
+	int req_vec;
 	int err;
+	int n;
 
 	if (mlx5_core_is_sf(dev))
 		return 0;
@@ -642,11 +692,13 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 	if (mlx5_sf_max_functions(dev))
 		total_vec += MLX5_IRQ_CTRL_SF_MAX +
 			MLX5_COMP_EQS_PER_SF * mlx5_sf_max_functions(dev);
+	total_vec = min_t(int, total_vec, pci_msix_vec_count(dev->pdev));
+	pcif_vec = min_t(int, pcif_vec, pci_msix_vec_count(dev->pdev));
 
-	total_vec = pci_alloc_irq_vectors(dev->pdev, 1, total_vec, PCI_IRQ_MSIX);
-	if (total_vec < 0)
-		return total_vec;
-	pcif_vec = min(pcif_vec, total_vec);
+	req_vec = pci_msix_can_alloc_dyn(dev->pdev) ? 1 : total_vec;
+	n = pci_alloc_irq_vectors(dev->pdev, 1, req_vec, PCI_IRQ_MSIX);
+	if (n < 0)
+		return n;
 
 	err = irq_pools_init(dev, total_vec - pcif_vec, pcif_vec);
 	if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h
index ea39e4027d8a..d3a77a0ab848 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.h
@@ -12,6 +12,7 @@
 #define MLX5_EQ_REFS_PER_IRQ (2)
 
 struct mlx5_irq;
+struct cpu_rmap;
 
 struct mlx5_irq_pool {
 	char name[MLX5_MAX_IRQ_NAME - MLX5_MAX_IRQ_IDX_CHARS];
@@ -31,7 +32,8 @@ static inline bool mlx5_irq_pool_is_sf_pool(struct mlx5_irq_pool *pool)
 }
 
 struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
-				struct irq_affinity_desc *af_desc);
+				struct irq_affinity_desc *af_desc,
+				struct cpu_rmap **rmap);
 int mlx5_irq_get_locked(struct mlx5_irq *irq);
 int mlx5_irq_read_locked(struct mlx5_irq *irq);
 int mlx5_irq_put(struct mlx5_irq *irq);
-- 
cgit v1.2.3-70-g09d2


From fe578cbb2f053f465e19d2671a523dcd01953888 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Tue, 3 Jan 2023 08:48:14 +0200
Subject: net/mlx5: Move devlink registration before mlx5_load

In order to allow reference to devlink parameters during driver load,
move the devlink registration before mlx5_load. Subsequent patch will
use it to control the number of completion vectors required based on
whether eth is enabled or not.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index bbc9b4188212..fb9ac7da6e4c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1399,16 +1399,16 @@ int mlx5_init_one(struct mlx5_core_dev *dev)
 		goto function_teardown;
 	}
 
+	err = mlx5_devlink_params_register(priv_to_devlink(dev));
+	if (err)
+		goto err_devlink_params_reg;
+
 	err = mlx5_load(dev);
 	if (err)
 		goto err_load;
 
 	set_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 
-	err = mlx5_devlink_params_register(priv_to_devlink(dev));
-	if (err)
-		goto err_devlink_params_reg;
-
 	err = mlx5_register_device(dev);
 	if (err)
 		goto err_register;
@@ -1418,11 +1418,11 @@ int mlx5_init_one(struct mlx5_core_dev *dev)
 	return 0;
 
 err_register:
-	mlx5_devlink_params_unregister(priv_to_devlink(dev));
-err_devlink_params_reg:
 	clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 	mlx5_unload(dev);
 err_load:
+	mlx5_devlink_params_unregister(priv_to_devlink(dev));
+err_devlink_params_reg:
 	mlx5_cleanup_once(dev);
 function_teardown:
 	mlx5_function_teardown(dev, true);
@@ -1441,7 +1441,6 @@ void mlx5_uninit_one(struct mlx5_core_dev *dev)
 	mutex_lock(&dev->intf_state_mutex);
 
 	mlx5_unregister_device(dev);
-	mlx5_devlink_params_unregister(priv_to_devlink(dev));
 
 	if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) {
 		mlx5_core_warn(dev, "%s: interface is down, NOP\n",
@@ -1452,6 +1451,7 @@ void mlx5_uninit_one(struct mlx5_core_dev *dev)
 
 	clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state);
 	mlx5_unload(dev);
+	mlx5_devlink_params_unregister(priv_to_devlink(dev));
 	mlx5_cleanup_once(dev);
 	mlx5_function_teardown(dev, true);
 out:
-- 
cgit v1.2.3-70-g09d2


From 1dc85133c207831dbaa91941a98f6c488411927a Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Tue, 3 Jan 2023 09:16:11 +0200
Subject: net/mlx5: Refactor calculation of required completion vectors

Move the calculation to a separate function. We will add more
functionality to it in a follow up patch.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 888ccfcbcce6..66696c935cc5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -1113,26 +1113,34 @@ void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev)
 #define MLX5_MAX_ASYNC_EQS 3
 #endif
 
-int mlx5_eq_table_create(struct mlx5_core_dev *dev)
+static int get_num_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *eq_table = dev->priv.eq_table;
-	int num_eqs = MLX5_CAP_GEN(dev, max_num_eqs) ?
+	int max_dev_eqs;
+	int max_eqs_sf;
+	int num_eqs;
+
+	max_dev_eqs = MLX5_CAP_GEN(dev, max_num_eqs) ?
 		      MLX5_CAP_GEN(dev, max_num_eqs) :
 		      1 << MLX5_CAP_GEN(dev, log_max_eq);
-	int max_eqs_sf;
-	int err;
 
-	eq_table->num_comp_eqs =
-		min_t(int,
-		      mlx5_irq_table_get_num_comp(eq_table->irq_table),
-		      num_eqs - MLX5_MAX_ASYNC_EQS);
+	num_eqs = min_t(int, mlx5_irq_table_get_num_comp(eq_table->irq_table),
+			max_dev_eqs - MLX5_MAX_ASYNC_EQS);
 	if (mlx5_core_is_sf(dev)) {
 		max_eqs_sf = min_t(int, MLX5_COMP_EQS_PER_SF,
 				   mlx5_irq_table_get_sfs_vec(eq_table->irq_table));
-		eq_table->num_comp_eqs = min_t(int, eq_table->num_comp_eqs,
-					       max_eqs_sf);
+		num_eqs = min_t(int, num_eqs, max_eqs_sf);
 	}
 
+	return num_eqs;
+}
+
+int mlx5_eq_table_create(struct mlx5_core_dev *dev)
+{
+	struct mlx5_eq_table *eq_table = dev->priv.eq_table;
+	int err;
+
+	eq_table->num_comp_eqs = get_num_eqs(dev);
 	err = create_async_eqs(dev);
 	if (err) {
 		mlx5_core_err(dev, "Failed to create async EQs\n");
-- 
cgit v1.2.3-70-g09d2


From b637ac5db0d0fd00cbe88469999ba2985355ac95 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Tue, 3 Jan 2023 09:37:23 +0200
Subject: net/mlx5: Use one completion vector if eth is disabled

If eth is disabled by devlink, use only a single completion vector to
have minimum performance of all users of completion vectors. This also
affects Infiniband performance.

The rest of the vectors can be used by other consumers on a first come
first served basis.

mlx5_vdpa will make use of this to allocate dedicated vectors for its
own use.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/dev.c     | 14 ++------------
 drivers/net/ethernet/mellanox/mlx5/core/devlink.h | 11 +++++++++++
 drivers/net/ethernet/mellanox/mlx5/core/eq.c      |  7 +++++++
 3 files changed, 20 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
index 445fe30c3d0b..e7739acc926e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
@@ -35,6 +35,7 @@
 #include <linux/mlx5/mlx5_ifc_vdpa.h>
 #include <linux/mlx5/vport.h>
 #include "mlx5_core.h"
+#include "devlink.h"
 
 /* intf dev list mutex */
 static DEFINE_MUTEX(mlx5_intf_mutex);
@@ -109,17 +110,6 @@ bool mlx5_eth_supported(struct mlx5_core_dev *dev)
 	return true;
 }
 
-static bool is_eth_enabled(struct mlx5_core_dev *dev)
-{
-	union devlink_param_value val;
-	int err;
-
-	err = devl_param_driverinit_value_get(priv_to_devlink(dev),
-					      DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH,
-					      &val);
-	return err ? false : val.vbool;
-}
-
 bool mlx5_vnet_supported(struct mlx5_core_dev *dev)
 {
 	if (!IS_ENABLED(CONFIG_MLX5_VDPA_NET))
@@ -251,7 +241,7 @@ static const struct mlx5_adev_device {
 					 .is_enabled = &is_ib_enabled },
 	[MLX5_INTERFACE_PROTOCOL_ETH] = { .suffix = "eth",
 					  .is_supported = &mlx5_eth_supported,
-					  .is_enabled = &is_eth_enabled },
+					  .is_enabled = &mlx5_core_is_eth_enabled },
 	[MLX5_INTERFACE_PROTOCOL_ETH_REP] = { .suffix = "eth-rep",
 					   .is_supported = &is_eth_rep_supported },
 	[MLX5_INTERFACE_PROTOCOL_IB_REP] = { .suffix = "rdma-rep",
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
index 212b12424146..f2601f9eef78 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h
@@ -44,4 +44,15 @@ void mlx5_devlink_free(struct devlink *devlink);
 int mlx5_devlink_params_register(struct devlink *devlink);
 void mlx5_devlink_params_unregister(struct devlink *devlink);
 
+static inline bool mlx5_core_is_eth_enabled(struct mlx5_core_dev *dev)
+{
+	union devlink_param_value val;
+	int err;
+
+	err = devl_param_driverinit_value_get(priv_to_devlink(dev),
+					      DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH,
+					      &val);
+	return err ? false : val.vbool;
+}
+
 #endif /* __MLX5_DEVLINK_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 66696c935cc5..eb41f0abf798 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -1120,6 +1120,13 @@ static int get_num_eqs(struct mlx5_core_dev *dev)
 	int max_eqs_sf;
 	int num_eqs;
 
+	/* If ethernet is disabled we use just a single completion vector to
+	 * have the other vectors available for other drivers using mlx5_core. For
+	 * example, mlx5_vdpa
+	 */
+	if (!mlx5_core_is_eth_enabled(dev) && mlx5_eth_supported(dev))
+		return 1;
+
 	max_dev_eqs = MLX5_CAP_GEN(dev, max_num_eqs) ?
 		      MLX5_CAP_GEN(dev, max_num_eqs) :
 		      1 << MLX5_CAP_GEN(dev, log_max_eq);
-- 
cgit v1.2.3-70-g09d2


From fb0a6a268dcd6fe144c99d60a1166e34c6991d5f Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Thu, 5 Jan 2023 11:31:46 +0200
Subject: net/mlx5: Provide external API for allocating vectors

Provide external API to be used by other drivers relying on mlx5_core,
for allocating MSIX vectors. An example for such a driver would be
mlx5_vdpa.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 52 +++++++++++++++++++++++
 include/linux/mlx5/driver.h                       |  6 +++
 2 files changed, 58 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 7fa63d31ae5b..e12e528c09f5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -483,6 +483,58 @@ struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
 	return irq;
 }
 
+/**
+ * mlx5_msix_alloc - allocate msix interrupt
+ * @dev: mlx5 device from which to request
+ * @handler: interrupt handler
+ * @affdesc: affinity descriptor
+ * @name: interrupt name
+ *
+ * Returns: struct msi_map with result encoded.
+ * Note: the caller must make sure to release the irq by calling
+ *       mlx5_msix_free() if shutdown was initiated.
+ */
+struct msi_map mlx5_msix_alloc(struct mlx5_core_dev *dev,
+			       irqreturn_t (*handler)(int, void *),
+			       const struct irq_affinity_desc *affdesc,
+			       const char *name)
+{
+	struct msi_map map;
+	int err;
+
+	if (!dev->pdev) {
+		map.virq = 0;
+		map.index = -EINVAL;
+		return map;
+	}
+
+	map = pci_msix_alloc_irq_at(dev->pdev, MSI_ANY_INDEX, affdesc);
+	if (!map.virq)
+		return map;
+
+	err = request_irq(map.virq, handler, 0, name, NULL);
+	if (err) {
+		mlx5_core_warn(dev, "err %d\n", err);
+		pci_msix_free_irq(dev->pdev, map);
+		map.virq = 0;
+		map.index = -ENOMEM;
+	}
+	return map;
+}
+EXPORT_SYMBOL(mlx5_msix_alloc);
+
+/**
+ * mlx5_msix_free - free a previously allocated msix interrupt
+ * @dev: mlx5 device associated with interrupt
+ * @map: map previously returned by mlx5_msix_alloc()
+ */
+void mlx5_msix_free(struct mlx5_core_dev *dev, struct msi_map map)
+{
+	free_irq(map.virq, NULL);
+	pci_msix_free_irq(dev->pdev, map);
+}
+EXPORT_SYMBOL(mlx5_msix_free);
+
 /**
  * mlx5_irqs_release_vectors - release one or more IRQs back to the system.
  * @irqs: IRQs to be released.
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index f33389b42209..df0f82110249 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1308,4 +1308,10 @@ enum {
 	MLX5_OCTWORD = 16,
 };
 
+struct msi_map mlx5_msix_alloc(struct mlx5_core_dev *dev,
+			       irqreturn_t (*handler)(int, void *),
+			       const struct irq_affinity_desc *affdesc,
+			       const char *name);
+void mlx5_msix_free(struct mlx5_core_dev *dev, struct msi_map map);
+
 #endif /* MLX5_DRIVER_H */
-- 
cgit v1.2.3-70-g09d2


From 2d08f3e128b1b380b0992ab40d60e0843db3c6b2 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Thu, 23 Mar 2023 12:13:43 -0400
Subject: mISDN: remove unused vpm_read_address and cpld_read_reg functions

clang with W=1 reports
drivers/isdn/hardware/mISDN/hfcmulti.c:667:1: error: unused function
  'vpm_read_address' [-Werror,-Wunused-function]
vpm_read_address(struct hfc_multi *c)
^

drivers/isdn/hardware/mISDN/hfcmulti.c:643:1: error: unused function
  'cpld_read_reg' [-Werror,-Wunused-function]
cpld_read_reg(struct hfc_multi *hc, unsigned char reg)
^

These functions are not used, so remove them.

Reported-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230323161343.2633836-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/isdn/hardware/mISDN/hfcmulti.c | 31 -------------------------------
 1 file changed, 31 deletions(-)

diff --git a/drivers/isdn/hardware/mISDN/hfcmulti.c b/drivers/isdn/hardware/mISDN/hfcmulti.c
index e840609c50eb..2e5cb9dde3ec 100644
--- a/drivers/isdn/hardware/mISDN/hfcmulti.c
+++ b/drivers/isdn/hardware/mISDN/hfcmulti.c
@@ -639,23 +639,6 @@ cpld_write_reg(struct hfc_multi *hc, unsigned char reg, unsigned char val)
 	return;
 }
 
-static inline unsigned char
-cpld_read_reg(struct hfc_multi *hc, unsigned char reg)
-{
-	unsigned char bytein;
-
-	cpld_set_reg(hc, reg);
-
-	/* Do data pin read low byte */
-	HFC_outb(hc, R_GPIO_OUT1, reg);
-
-	enablepcibridge(hc);
-	bytein = readpcibridge(hc, 1);
-	disablepcibridge(hc);
-
-	return bytein;
-}
-
 static inline void
 vpm_write_address(struct hfc_multi *hc, unsigned short addr)
 {
@@ -663,20 +646,6 @@ vpm_write_address(struct hfc_multi *hc, unsigned short addr)
 	cpld_write_reg(hc, 1, 0x01 & (addr >> 8));
 }
 
-static inline unsigned short
-vpm_read_address(struct hfc_multi *c)
-{
-	unsigned short addr;
-	unsigned short highbit;
-
-	addr = cpld_read_reg(c, 0);
-	highbit = cpld_read_reg(c, 1);
-
-	addr = addr | (highbit << 8);
-
-	return addr & 0x1ff;
-}
-
 static inline unsigned char
 vpm_in(struct hfc_multi *c, int which, unsigned short addr)
 {
-- 
cgit v1.2.3-70-g09d2


From bc77f7318da8812244cd2329d0bfc1312e75cfd9 Mon Sep 17 00:00:00 2001
From: Michal Michalik <michal.michalik@intel.com>
Date: Thu, 23 Mar 2023 20:08:02 +0100
Subject: tools: ynl: add the Python requirements.txt file

It is a good practice to state explicitly which are the required Python
packages needed in a particular project to run it. The most commonly
used way is to store them in the `requirements.txt` file*.

*URL: https://pip.pypa.io/en/stable/reference/requirements-file-format/

Currently user needs to figure out himself that Python needs `PyYAML`
and `jsonschema` (and theirs requirements) packages to use the tool.
Add the `requirements.txt` for user convenience.

How to use it:
1) (optional) Create and activate empty virtual environment:
  python3.X -m venv venv3X
  source ./venv3X/bin/activate
2) Install all the required packages:
  pip install -r requirements.txt
    or
  python -m pip install -r requirements.txt
3) Run the script!

The `requirements.txt` file was tested for:
* Python 3.6
* Python 3.8
* Python 3.10

Signed-off-by: Michal Michalik <michal.michalik@intel.com>
Link: https://lore.kernel.org/r/20230323190802.32206-1-michal.michalik@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/requirements.txt | 2 ++
 1 file changed, 2 insertions(+)
 create mode 100644 tools/net/ynl/requirements.txt

diff --git a/tools/net/ynl/requirements.txt b/tools/net/ynl/requirements.txt
new file mode 100644
index 000000000000..0db6ad0c1b39
--- /dev/null
+++ b/tools/net/ynl/requirements.txt
@@ -0,0 +1,2 @@
+jsonschema==4.*
+PyYAML==6.*
-- 
cgit v1.2.3-70-g09d2


From 1431d0b584a673ea690c88a5f7e1aedd9caf0e84 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Sat, 25 Mar 2023 16:31:42 -0500
Subject: bpf: Only invoke kptr dtor following non-NULL xchg

When a map value is being freed, we loop over all of the fields of the
corresponding BPF object and issue the appropriate cleanup calls
corresponding to the field's type. If the field is a referenced kptr, we
atomically xchg the value out of the map, and invoke the kptr's
destructor on whatever was there before (or bpf_obj_drop() it if it was
a local kptr).

Currently, we always invoke the destructor (either bpf_obj_drop() or the
kptr's registered destructor) on any KPTR_REF-type field in a map, even
if there wasn't a value in the map. This means that any function serving
as the kptr's KF_RELEASE destructor must always treat the argument as
possibly NULL, as the following can and regularly does happen:

void *xchgd_field;

/* No value was in the map, so xchgd_field is NULL */
xchgd_field = (void *)xchg(unsigned long *field_ptr, 0);
field->kptr.dtor(xchgd_field);

These are odd semantics to impose on KF_RELEASE kfuncs -- BPF programs
are prohibited by the verifier from passing NULL pointers to KF_RELEASE
kfuncs, so it doesn't make sense to require this of BPF programs, but
not the main kernel destructor path. It's also unnecessary to invoke any
cleanup logic for local kptrs. If there is no object there, there's
nothing to drop.

So as to allow KF_RELEASE kfuncs to fully assume that an argument is
non-NULL, this patch updates a KPTR_REF's destructor to only be invoked
when a non-NULL value is xchg'd out of the kptr map field.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230325213144.486885-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/syscall.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index a09597c95029..e18ac7fdc210 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -677,6 +677,9 @@ void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 			break;
 		case BPF_KPTR_REF:
 			xchgd_field = (void *)xchg((unsigned long *)field_ptr, 0);
+			if (!xchgd_field)
+				break;
+
 			if (!btf_is_kernel(field->kptr.btf)) {
 				pointee_struct_meta = btf_find_struct_meta(field->kptr.btf,
 									   field->kptr.btf_id);
-- 
cgit v1.2.3-70-g09d2


From fb2211a57c110b4ced3cb7f8570bd7246acf2d04 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Sat, 25 Mar 2023 16:31:45 -0500
Subject: bpf: Remove now-unnecessary NULL checks for KF_RELEASE kfuncs

Now that we're not invoking kfunc destructors when the kptr in a map was
NULL, we no longer require NULL checks in many of our KF_RELEASE kfuncs.
This patch removes those NULL checks.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230325213144.486885-3-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 drivers/hid/bpf/hid_bpf_dispatch.c | 3 ---
 kernel/bpf/cpumask.c               | 3 ---
 kernel/bpf/helpers.c               | 6 ------
 net/bpf/test_run.c                 | 3 ---
 net/netfilter/nf_conntrack_bpf.c   | 2 --
 5 files changed, 17 deletions(-)

diff --git a/drivers/hid/bpf/hid_bpf_dispatch.c b/drivers/hid/bpf/hid_bpf_dispatch.c
index 8a034a555d4c..d9ef45fcaeab 100644
--- a/drivers/hid/bpf/hid_bpf_dispatch.c
+++ b/drivers/hid/bpf/hid_bpf_dispatch.c
@@ -342,9 +342,6 @@ hid_bpf_release_context(struct hid_bpf_ctx *ctx)
 {
 	struct hid_bpf_ctx_kern *ctx_kern;
 
-	if (!ctx)
-		return;
-
 	ctx_kern = container_of(ctx, struct hid_bpf_ctx_kern, ctx);
 
 	kfree(ctx_kern);
diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c
index db9da2194c1a..e991af7dc13c 100644
--- a/kernel/bpf/cpumask.c
+++ b/kernel/bpf/cpumask.c
@@ -102,9 +102,6 @@ static void cpumask_free_cb(struct rcu_head *head)
  */
 __bpf_kfunc void bpf_cpumask_release(struct bpf_cpumask *cpumask)
 {
-	if (!cpumask)
-		return;
-
 	if (refcount_dec_and_test(&cpumask->usage))
 		call_rcu(&cpumask->rcu, cpumask_free_cb);
 }
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index f753676ef652..8980f6859443 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2089,9 +2089,6 @@ __bpf_kfunc struct task_struct *bpf_task_kptr_get(struct task_struct **pp)
  */
 __bpf_kfunc void bpf_task_release(struct task_struct *p)
 {
-	if (!p)
-		return;
-
 	put_task_struct(p);
 }
 
@@ -2148,9 +2145,6 @@ __bpf_kfunc struct cgroup *bpf_cgroup_kptr_get(struct cgroup **cgrpp)
  */
 __bpf_kfunc void bpf_cgroup_release(struct cgroup *cgrp)
 {
-	if (!cgrp)
-		return;
-
 	cgroup_put(cgrp);
 }
 
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 8d6b31209bd6..27587f1c5f36 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -615,9 +615,6 @@ bpf_kfunc_call_memb_acquire(void)
 
 __bpf_kfunc void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p)
 {
-	if (!p)
-		return;
-
 	refcount_dec(&p->cnt);
 }
 
diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c
index cd99e6dc1f35..002e9d24a1e9 100644
--- a/net/netfilter/nf_conntrack_bpf.c
+++ b/net/netfilter/nf_conntrack_bpf.c
@@ -401,8 +401,6 @@ __bpf_kfunc struct nf_conn *bpf_ct_insert_entry(struct nf_conn___init *nfct_i)
  */
 __bpf_kfunc void bpf_ct_release(struct nf_conn *nfct)
 {
-	if (!nfct)
-		return;
 	nf_ct_put(nfct);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 6c831c4684124a544f73f7c9b83bc7b2eb0b23d3 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Sat, 25 Mar 2023 16:31:46 -0500
Subject: bpf: Treat KF_RELEASE kfuncs as KF_TRUSTED_ARGS

KF_RELEASE kfuncs are not currently treated as having KF_TRUSTED_ARGS,
even though they have a superset of the requirements of KF_TRUSTED_ARGS.
Like KF_TRUSTED_ARGS, KF_RELEASE kfuncs require a 0-offset argument, and
don't allow NULL-able arguments. Unlike KF_TRUSTED_ARGS which require
_either_ an argument with ref_obj_id > 0, _or_ (ref->type &
BPF_REG_TRUSTED_MODIFIERS) (and no unsafe modifiers allowed), KF_RELEASE
only allows for ref_obj_id > 0.  Because KF_RELEASE today doesn't
automatically imply KF_TRUSTED_ARGS, some of these requirements are
enforced in different ways that can make the behavior of the verifier
feel unpredictable. For example, a KF_RELEASE kfunc with a NULL-able
argument will currently fail in the verifier with a message like, "arg#0
is ptr_or_null_ expected ptr_ or socket" rather than "Possibly NULL
pointer passed to trusted arg0". Our intention is the same, but the
semantics are different due to implemenetation details that kfunc authors
and BPF program writers should not need to care about.

Let's make the behavior of the verifier more consistent and intuitive by
having KF_RELEASE kfuncs imply the presence of KF_TRUSTED_ARGS. Our
eventual goal is to have all kfuncs assume KF_TRUSTED_ARGS by default
anyways, so this takes us a step in that direction.

Note that it does not make sense to assume KF_TRUSTED_ARGS for all
KF_ACQUIRE kfuncs. KF_ACQUIRE kfuncs can have looser semantics than
KF_RELEASE, with e.g. KF_RCU | KF_RET_NULL. We may want to have
KF_ACQUIRE imply KF_TRUSTED_ARGS _unless_ KF_RCU is specified, but that
can be left to another patch set, and there are no such subtleties to
address for KF_RELEASE.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230325213144.486885-4-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst                           |  7 ++++---
 kernel/bpf/cpumask.c                                   |  2 +-
 kernel/bpf/verifier.c                                  |  2 +-
 net/bpf/test_run.c                                     |  6 ++++++
 tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c |  4 ++--
 tools/testing/selftests/bpf/progs/task_kfunc_failure.c |  6 +++---
 tools/testing/selftests/bpf/verifier/calls.c           | 10 +++++++---
 tools/testing/selftests/bpf/verifier/ref_tracking.c    |  6 +++---
 8 files changed, 27 insertions(+), 16 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index 69eccf6f98ef..bf1b85941452 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -179,9 +179,10 @@ both are orthogonal to each other.
 ---------------------
 
 The KF_RELEASE flag is used to indicate that the kfunc releases the pointer
-passed in to it. There can be only one referenced pointer that can be passed in.
-All copies of the pointer being released are invalidated as a result of invoking
-kfunc with this flag.
+passed in to it. There can be only one referenced pointer that can be passed
+in. All copies of the pointer being released are invalidated as a result of
+invoking kfunc with this flag. KF_RELEASE kfuncs automatically receive the
+protection afforded by the KF_TRUSTED_ARGS flag described below.
 
 2.4.4 KF_KPTR_GET flag
 ----------------------
diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c
index e991af7dc13c..7efdf5d770ca 100644
--- a/kernel/bpf/cpumask.c
+++ b/kernel/bpf/cpumask.c
@@ -402,7 +402,7 @@ __diag_pop();
 
 BTF_SET8_START(cpumask_kfunc_btf_ids)
 BTF_ID_FLAGS(func, bpf_cpumask_create, KF_ACQUIRE | KF_RET_NULL)
-BTF_ID_FLAGS(func, bpf_cpumask_release, KF_RELEASE | KF_TRUSTED_ARGS)
+BTF_ID_FLAGS(func, bpf_cpumask_release, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_cpumask_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
 BTF_ID_FLAGS(func, bpf_cpumask_first, KF_RCU)
 BTF_ID_FLAGS(func, bpf_cpumask_first_zero, KF_RCU)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 64f06f6e16bf..20eb2015842f 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -9307,7 +9307,7 @@ static bool is_kfunc_release(struct bpf_kfunc_call_arg_meta *meta)
 
 static bool is_kfunc_trusted_args(struct bpf_kfunc_call_arg_meta *meta)
 {
-	return meta->kfunc_flags & KF_TRUSTED_ARGS;
+	return (meta->kfunc_flags & KF_TRUSTED_ARGS) || is_kfunc_release(meta);
 }
 
 static bool is_kfunc_sleepable(struct bpf_kfunc_call_arg_meta *meta)
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 27587f1c5f36..f1652f5fbd2e 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -606,6 +606,11 @@ bpf_kfunc_call_test_acquire(unsigned long *scalar_ptr)
 	return &prog_test_struct;
 }
 
+__bpf_kfunc void bpf_kfunc_call_test_offset(struct prog_test_ref_kfunc *p)
+{
+	WARN_ON_ONCE(1);
+}
+
 __bpf_kfunc struct prog_test_member *
 bpf_kfunc_call_memb_acquire(void)
 {
@@ -800,6 +805,7 @@ BTF_ID_FLAGS(func, bpf_kfunc_call_test_mem_len_fail2)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_ref, KF_TRUSTED_ARGS | KF_RCU)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_destructive, KF_DESTRUCTIVE)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_static_unused_arg)
+BTF_ID_FLAGS(func, bpf_kfunc_call_test_offset)
 BTF_SET8_END(test_sk_check_kfunc_ids)
 
 static void *bpf_test_init(const union bpf_attr *kattr, u32 user_size,
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
index 807fb0ac41e9..48b2034cadb3 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
@@ -206,7 +206,7 @@ int BPF_PROG(cgrp_kfunc_get_unreleased, struct cgroup *cgrp, const char *path)
 }
 
 SEC("tp_btf/cgroup_mkdir")
-__failure __msg("expects refcounted")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(cgrp_kfunc_release_untrusted, struct cgroup *cgrp, const char *path)
 {
 	struct __cgrps_kfunc_map_value *v;
@@ -234,7 +234,7 @@ int BPF_PROG(cgrp_kfunc_release_fp, struct cgroup *cgrp, const char *path)
 }
 
 SEC("tp_btf/cgroup_mkdir")
-__failure __msg("arg#0 is ptr_or_null_ expected ptr_ or socket")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(cgrp_kfunc_release_null, struct cgroup *cgrp, const char *path)
 {
 	struct __cgrps_kfunc_map_value local, *v;
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
index 27994d6b2914..2c374a7ffece 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
@@ -206,7 +206,7 @@ int BPF_PROG(task_kfunc_get_unreleased, struct task_struct *task, u64 clone_flag
 }
 
 SEC("tp_btf/task_newtask")
-__failure __msg("arg#0 is untrusted_ptr_or_null_ expected ptr_ or socket")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(task_kfunc_release_untrusted, struct task_struct *task, u64 clone_flags)
 {
 	struct __tasks_kfunc_map_value *v;
@@ -234,7 +234,7 @@ int BPF_PROG(task_kfunc_release_fp, struct task_struct *task, u64 clone_flags)
 }
 
 SEC("tp_btf/task_newtask")
-__failure __msg("arg#0 is ptr_or_null_ expected ptr_ or socket")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(task_kfunc_release_null, struct task_struct *task, u64 clone_flags)
 {
 	struct __tasks_kfunc_map_value local, *v;
@@ -277,7 +277,7 @@ int BPF_PROG(task_kfunc_release_unacquired, struct task_struct *task, u64 clone_
 }
 
 SEC("tp_btf/task_newtask")
-__failure __msg("arg#0 is ptr_or_null_ expected ptr_ or socket")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(task_kfunc_from_pid_no_null_check, struct task_struct *task, u64 clone_flags)
 {
 	struct task_struct *acquired;
diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c
index 5702fc9761ef..1bdf2b43e49e 100644
--- a/tools/testing/selftests/bpf/verifier/calls.c
+++ b/tools/testing/selftests/bpf/verifier/calls.c
@@ -109,7 +109,7 @@
 	},
 	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
 	.result = REJECT,
-	.errstr = "arg#0 is ptr_or_null_ expected ptr_ or socket",
+	.errstr = "Possibly NULL pointer passed to trusted arg0",
 	.fixup_kfunc_btf_id = {
 		{ "bpf_kfunc_call_test_acquire", 3 },
 		{ "bpf_kfunc_call_test_release", 5 },
@@ -165,19 +165,23 @@
 	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
 	BPF_ST_MEM(BPF_DW, BPF_REG_1, 0, 0),
 	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
 	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
 	BPF_EXIT_INSN(),
 	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, 16),
 	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -4),
 	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
 	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_MOV64_IMM(BPF_REG_0, 0),
 	BPF_EXIT_INSN(),
 	},
 	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
 	.fixup_kfunc_btf_id = {
 		{ "bpf_kfunc_call_test_acquire", 3 },
-		{ "bpf_kfunc_call_test_release", 9 },
+		{ "bpf_kfunc_call_test_offset", 9 },
+		{ "bpf_kfunc_call_test_release", 12 },
 	},
 	.result_unpriv = REJECT,
 	.result = REJECT,
diff --git a/tools/testing/selftests/bpf/verifier/ref_tracking.c b/tools/testing/selftests/bpf/verifier/ref_tracking.c
index 9540164712b7..5a2e154dd1e0 100644
--- a/tools/testing/selftests/bpf/verifier/ref_tracking.c
+++ b/tools/testing/selftests/bpf/verifier/ref_tracking.c
@@ -142,7 +142,7 @@
 	.kfunc = "bpf",
 	.expected_attach_type = BPF_LSM_MAC,
 	.flags = BPF_F_SLEEPABLE,
-	.errstr = "arg#0 is ptr_or_null_ expected ptr_ or socket",
+	.errstr = "Possibly NULL pointer passed to trusted arg0",
 	.fixup_kfunc_btf_id = {
 		{ "bpf_lookup_user_key", 2 },
 		{ "bpf_key_put", 4 },
@@ -163,7 +163,7 @@
 	.kfunc = "bpf",
 	.expected_attach_type = BPF_LSM_MAC,
 	.flags = BPF_F_SLEEPABLE,
-	.errstr = "arg#0 is ptr_or_null_ expected ptr_ or socket",
+	.errstr = "Possibly NULL pointer passed to trusted arg0",
 	.fixup_kfunc_btf_id = {
 		{ "bpf_lookup_system_key", 1 },
 		{ "bpf_key_put", 3 },
@@ -182,7 +182,7 @@
 	.kfunc = "bpf",
 	.expected_attach_type = BPF_LSM_MAC,
 	.flags = BPF_F_SLEEPABLE,
-	.errstr = "arg#0 pointer type STRUCT bpf_key must point to scalar, or struct with scalar",
+	.errstr = "Possibly NULL pointer passed to trusted arg0",
 	.fixup_kfunc_btf_id = {
 		{ "bpf_key_put", 1 },
 	},
-- 
cgit v1.2.3-70-g09d2


From 3e5329e193f463e6aaf98c33f7cb1308160880ab Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:42 +0200
Subject: selftests/bpf: Report program name on parse_test_spec error

Change test_loader.c:run_subtest() behavior to show BPF program name
when test spec for that program can't be parsed.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-2-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/test_loader.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index bf41390157bf..8ca5121b5329 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -232,8 +232,11 @@ void run_subtest(struct test_loader *tester,
 
 		/* if we can't derive test specification, go to the next test */
 		err = parse_test_spec(tester, obj, prog, &spec);
-		if (!ASSERT_OK(err, "parse_test_spec"))
+		if (err) {
+			PRINT_FAIL("Can't parse test spec for program '%s'\n",
+				   bpf_program__name(prog));
 			continue;
+		}
 
 		tobj = bpf_object__open_mem(obj_bytes, obj_byte_cnt, &open_opts);
 		if (!ASSERT_OK_PTR(tobj, "obj_open_mem")) /* shouldn't happen */
-- 
cgit v1.2.3-70-g09d2


From 207b1ba3019100d862931e97b49f76ff1e0a89f2 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:43 +0200
Subject: selftests/bpf: __imm_insn & __imm_const macro for bpf_misc.h

Add two convenience macro for BPF test cases,
allowing the following usage:

  #include <linux/filter.h>

  ...
  asm volatile (
  ...
  ".8byte %[raw_insn];"
  ...
  "r1 += %[st_foo_offset];"
  ...
  :
  : __imm_insn(raw_insn, BPF_RAW_INSN(...)),
    __imm_const(st_foo_offset, offsetof(struct st, foo))
  : __clobber_all);

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-3-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_misc.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 3c03ec8056ce..8b4681a96f89 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -35,8 +35,10 @@
 #define __clobber_all "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "memory"
 #define __clobber_common "r0", "r1", "r2", "r3", "r4", "r5", "memory"
 #define __imm(name) [name]"i"(name)
+#define __imm_const(name, expr) [name]"i"(expr)
 #define __imm_addr(name) [name]"i"(&name)
 #define __imm_ptr(name) [name]"p"(&name)
+#define __imm_insn(name, expr) [name]"i"(*(long *)&(expr))
 
 #if defined(__TARGET_ARCH_x86)
 #define SYSCALL_WRAPPER 1
-- 
cgit v1.2.3-70-g09d2


From 1d56ade032a49b2042f43b3f6bdf116928064267 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:44 +0200
Subject: selftests/bpf: Unprivileged tests for test_loader.c

Extends test_loader.c:test_loader__run_subtests() by allowing to
execute tests in unprivileged mode, similar to test_verifier.c.

Adds the following new attributes controlling test_loader behavior:

  __msg_unpriv
  __success_unpriv
  __failure_unpriv

* If any of these attributes is present the test would be loaded in
  unprivileged mode.
* If only "privileged" attributes are present the test would be loaded
  only in privileged mode.
* If both "privileged" and "unprivileged" attributes are present the
  test would be loaded in both modes.
* If test has to be executed in both modes, __msg(text) is specified
  and __msg_unpriv is not specified the behavior is the same as if
  __msg_unpriv(text) is specified.
* For test filtering purposes the name of the program loaded in
  unprivileged mode is derived from the usual program name by adding
  `@unpriv' suffix.

Also adds attribute '__description'. This attribute specifies text to
be used instead of a program name for display and filtering purposes.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-4-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/Makefile          |  10 +-
 tools/testing/selftests/bpf/autoconf_helper.h |   9 +
 tools/testing/selftests/bpf/progs/bpf_misc.h  |  25 ++
 tools/testing/selftests/bpf/test_loader.c     | 394 +++++++++++++++++++++-----
 tools/testing/selftests/bpf/test_verifier.c   |  25 +-
 tools/testing/selftests/bpf/unpriv_helpers.c  |  26 ++
 tools/testing/selftests/bpf/unpriv_helpers.h  |   7 +
 7 files changed, 395 insertions(+), 101 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/autoconf_helper.h
 create mode 100644 tools/testing/selftests/bpf/unpriv_helpers.c
 create mode 100644 tools/testing/selftests/bpf/unpriv_helpers.h

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index fc092582d16d..4a8ef118fd9d 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -231,8 +231,9 @@ TEST_GEN_PROGS_EXTENDED += $(TRUNNER_BPFTOOL)
 
 $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED): $(BPFOBJ)
 
-CGROUP_HELPERS	:= $(OUTPUT)/cgroup_helpers.o
 TESTING_HELPERS	:= $(OUTPUT)/testing_helpers.o
+CGROUP_HELPERS	:= $(OUTPUT)/cgroup_helpers.o
+UNPRIV_HELPERS  := $(OUTPUT)/unpriv_helpers.o
 TRACE_HELPERS	:= $(OUTPUT)/trace_helpers.o
 JSON_WRITER		:= $(OUTPUT)/json_writer.o
 CAP_HELPERS	:= $(OUTPUT)/cap_helpers.o
@@ -252,7 +253,7 @@ $(OUTPUT)/test_lirc_mode2_user: $(TESTING_HELPERS)
 $(OUTPUT)/xdping: $(TESTING_HELPERS)
 $(OUTPUT)/flow_dissector_load: $(TESTING_HELPERS)
 $(OUTPUT)/test_maps: $(TESTING_HELPERS)
-$(OUTPUT)/test_verifier: $(TESTING_HELPERS) $(CAP_HELPERS)
+$(OUTPUT)/test_verifier: $(TESTING_HELPERS) $(CAP_HELPERS) $(UNPRIV_HELPERS)
 $(OUTPUT)/xsk.o: $(BPFOBJ)
 
 BPFTOOL ?= $(DEFAULT_BPFTOOL)
@@ -560,8 +561,9 @@ TRUNNER_BPF_PROGS_DIR := progs
 TRUNNER_EXTRA_SOURCES := test_progs.c cgroup_helpers.c trace_helpers.c	\
 			 network_helpers.c testing_helpers.c		\
 			 btf_helpers.c flow_dissector_load.h		\
-			 cap_helpers.c test_loader.c xsk.c disasm.c \
-			 json_writer.c
+			 cap_helpers.c test_loader.c xsk.c disasm.c	\
+			 json_writer.c unpriv_helpers.c
+
 TRUNNER_EXTRA_FILES := $(OUTPUT)/urandom_read $(OUTPUT)/bpf_testmod.ko	\
 		       $(OUTPUT)/liburandom_read.so			\
 		       $(OUTPUT)/xdp_synproxy				\
diff --git a/tools/testing/selftests/bpf/autoconf_helper.h b/tools/testing/selftests/bpf/autoconf_helper.h
new file mode 100644
index 000000000000..5b243b9cdf8c
--- /dev/null
+++ b/tools/testing/selftests/bpf/autoconf_helper.h
@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#ifdef HAVE_GENHDR
+# include "autoconf.h"
+#else
+# if defined(__i386) || defined(__x86_64) || defined(__s390x__) || defined(__aarch64__)
+#  define CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS 1
+# endif
+#endif
diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 8b4681a96f89..9defc217a5bd 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -5,12 +5,33 @@
 /* This set of attributes controls behavior of the
  * test_loader.c:test_loader__run_subtests().
  *
+ * The test_loader sequentially loads each program in a skeleton.
+ * Programs could be loaded in privileged and unprivileged modes.
+ * - __success, __failure, __msg imply privileged mode;
+ * - __success_unpriv, __failure_unpriv, __msg_unpriv imply
+ *   unprivileged mode.
+ * If combination of privileged and unprivileged attributes is present
+ * both modes are used. If none are present privileged mode is implied.
+ *
+ * See test_loader.c:drop_capabilities() for exact set of capabilities
+ * that differ between privileged and unprivileged modes.
+ *
+ * For test filtering purposes the name of the program loaded in
+ * unprivileged mode is derived from the usual program name by adding
+ * `@unpriv' suffix.
+ *
  * __msg             Message expected to be found in the verifier log.
  *                   Multiple __msg attributes could be specified.
+ * __msg_unpriv      Same as __msg but for unprivileged mode.
  *
  * __success         Expect program load success in privileged mode.
+ * __success_unpriv  Expect program load success in unprivileged mode.
  *
  * __failure         Expect program load failure in privileged mode.
+ * __failure_unpriv  Expect program load failure in unprivileged mode.
+ *
+ * __description     Text to be used instead of a program name for display
+ *                   and filtering purposes.
  *
  * __log_level       Log level to use for the program, numeric value expected.
  *
@@ -27,6 +48,10 @@
 #define __msg(msg)		__attribute__((btf_decl_tag("comment:test_expect_msg=" msg)))
 #define __failure		__attribute__((btf_decl_tag("comment:test_expect_failure")))
 #define __success		__attribute__((btf_decl_tag("comment:test_expect_success")))
+#define __description(desc)	__attribute__((btf_decl_tag("comment:test_description=" desc)))
+#define __msg_unpriv(msg)	__attribute__((btf_decl_tag("comment:test_expect_msg_unpriv=" msg)))
+#define __failure_unpriv	__attribute__((btf_decl_tag("comment:test_expect_failure_unpriv")))
+#define __success_unpriv	__attribute__((btf_decl_tag("comment:test_expect_success_unpriv")))
 #define __log_level(lvl)	__attribute__((btf_decl_tag("comment:test_log_level="#lvl)))
 #define __flag(flag)		__attribute__((btf_decl_tag("comment:test_prog_flags="#flag)))
 
diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index 8ca5121b5329..41cddb303885 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -1,9 +1,14 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /* Copyright (c) 2022 Meta Platforms, Inc. and affiliates. */
+#include <linux/capability.h>
 #include <stdlib.h>
 #include <test_progs.h>
 #include <bpf/btf.h>
 
+#include "autoconf_helper.h"
+#include "unpriv_helpers.h"
+#include "cap_helpers.h"
+
 #define str_has_pfx(str, pfx) \
 	(strncmp(str, pfx, __builtin_constant_p(pfx) ? sizeof(pfx) - 1 : strlen(pfx)) == 0)
 
@@ -12,16 +17,40 @@
 #define TEST_TAG_EXPECT_FAILURE "comment:test_expect_failure"
 #define TEST_TAG_EXPECT_SUCCESS "comment:test_expect_success"
 #define TEST_TAG_EXPECT_MSG_PFX "comment:test_expect_msg="
+#define TEST_TAG_EXPECT_FAILURE_UNPRIV "comment:test_expect_failure_unpriv"
+#define TEST_TAG_EXPECT_SUCCESS_UNPRIV "comment:test_expect_success_unpriv"
+#define TEST_TAG_EXPECT_MSG_PFX_UNPRIV "comment:test_expect_msg_unpriv="
 #define TEST_TAG_LOG_LEVEL_PFX "comment:test_log_level="
 #define TEST_TAG_PROG_FLAGS_PFX "comment:test_prog_flags="
+#define TEST_TAG_DESCRIPTION_PFX "comment:test_description="
 
-struct test_spec {
-	const char *name;
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+#define EFFICIENT_UNALIGNED_ACCESS 1
+#else
+#define EFFICIENT_UNALIGNED_ACCESS 0
+#endif
+
+static int sysctl_unpriv_disabled = -1;
+
+enum mode {
+	PRIV = 1,
+	UNPRIV = 2
+};
+
+struct test_subspec {
+	char *name;
 	bool expect_failure;
 	const char **expect_msgs;
 	size_t expect_msg_cnt;
+};
+
+struct test_spec {
+	const char *prog_name;
+	struct test_subspec priv;
+	struct test_subspec unpriv;
 	int log_level;
 	int prog_flags;
+	int mode_mask;
 };
 
 static int tester_init(struct test_loader *tester)
@@ -44,17 +73,46 @@ void test_loader_fini(struct test_loader *tester)
 	free(tester->log_buf);
 }
 
+static void free_test_spec(struct test_spec *spec)
+{
+	free(spec->priv.name);
+	free(spec->unpriv.name);
+	free(spec->priv.expect_msgs);
+	free(spec->unpriv.expect_msgs);
+}
+
+static int push_msg(const char *msg, struct test_subspec *subspec)
+{
+	void *tmp;
+
+	tmp = realloc(subspec->expect_msgs, (1 + subspec->expect_msg_cnt) * sizeof(void *));
+	if (!tmp) {
+		ASSERT_FAIL("failed to realloc memory for messages\n");
+		return -ENOMEM;
+	}
+	subspec->expect_msgs = tmp;
+	subspec->expect_msgs[subspec->expect_msg_cnt++] = msg;
+
+	return 0;
+}
+
+/* Uses btf_decl_tag attributes to describe the expected test
+ * behavior, see bpf_misc.h for detailed description of each attribute
+ * and attribute combinations.
+ */
 static int parse_test_spec(struct test_loader *tester,
 			   struct bpf_object *obj,
 			   struct bpf_program *prog,
 			   struct test_spec *spec)
 {
+	const char *description = NULL;
+	bool has_unpriv_result = false;
+	int func_id, i, err = 0;
 	struct btf *btf;
-	int func_id, i;
 
 	memset(spec, 0, sizeof(*spec));
 
-	spec->name = bpf_program__name(prog);
+	spec->prog_name = bpf_program__name(prog);
 
 	btf = bpf_object__btf(obj);
 	if (!btf) {
@@ -62,15 +120,15 @@ static int parse_test_spec(struct test_loader *tester,
 		return -EINVAL;
 	}
 
-	func_id = btf__find_by_name_kind(btf, spec->name, BTF_KIND_FUNC);
+	func_id = btf__find_by_name_kind(btf, spec->prog_name, BTF_KIND_FUNC);
 	if (func_id < 0) {
-		ASSERT_FAIL("failed to find FUNC BTF type for '%s'", spec->name);
+		ASSERT_FAIL("failed to find FUNC BTF type for '%s'", spec->prog_name);
 		return -EINVAL;
 	}
 
 	for (i = 1; i < btf__type_cnt(btf); i++) {
+		const char *s, *val, *msg;
 		const struct btf_type *t;
-		const char *s, *val;
 		char *e;
 
 		t = btf__type_by_id(btf, i);
@@ -81,30 +139,42 @@ static int parse_test_spec(struct test_loader *tester,
 			continue;
 
 		s = btf__str_by_offset(btf, t->name_off);
-		if (strcmp(s, TEST_TAG_EXPECT_FAILURE) == 0) {
-			spec->expect_failure = true;
+		if (str_has_pfx(s, TEST_TAG_DESCRIPTION_PFX)) {
+			description = s + sizeof(TEST_TAG_DESCRIPTION_PFX) - 1;
+		} else if (strcmp(s, TEST_TAG_EXPECT_FAILURE) == 0) {
+			spec->priv.expect_failure = true;
+			spec->mode_mask |= PRIV;
 		} else if (strcmp(s, TEST_TAG_EXPECT_SUCCESS) == 0) {
-			spec->expect_failure = false;
+			spec->priv.expect_failure = false;
+			spec->mode_mask |= PRIV;
+		} else if (strcmp(s, TEST_TAG_EXPECT_FAILURE_UNPRIV) == 0) {
+			spec->unpriv.expect_failure = true;
+			spec->mode_mask |= UNPRIV;
+			has_unpriv_result = true;
+		} else if (strcmp(s, TEST_TAG_EXPECT_SUCCESS_UNPRIV) == 0) {
+			spec->unpriv.expect_failure = false;
+			spec->mode_mask |= UNPRIV;
+			has_unpriv_result = true;
 		} else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX)) {
-			void *tmp;
-			const char **msg;
-
-			tmp = realloc(spec->expect_msgs,
-				      (1 + spec->expect_msg_cnt) * sizeof(void *));
-			if (!tmp) {
-				ASSERT_FAIL("failed to realloc memory for messages\n");
-				return -ENOMEM;
-			}
-			spec->expect_msgs = tmp;
-			msg = &spec->expect_msgs[spec->expect_msg_cnt++];
-			*msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
+			msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
+			err = push_msg(msg, &spec->priv);
+			if (err)
+				goto cleanup;
+			spec->mode_mask |= PRIV;
+		} else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX_UNPRIV)) {
+			msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX_UNPRIV) - 1;
+			err = push_msg(msg, &spec->unpriv);
+			if (err)
+				goto cleanup;
+			spec->mode_mask |= UNPRIV;
 		} else if (str_has_pfx(s, TEST_TAG_LOG_LEVEL_PFX)) {
 			val = s + sizeof(TEST_TAG_LOG_LEVEL_PFX) - 1;
 			errno = 0;
 			spec->log_level = strtol(val, &e, 0);
 			if (errno || e[0] != '\0') {
-				ASSERT_FAIL("failed to parse test log level from '%s'", s);
-				return -EINVAL;
+				PRINT_FAIL("failed to parse test log level from '%s'\n", s);
+				err = -EINVAL;
+				goto cleanup;
 			}
 		} else if (str_has_pfx(s, TEST_TAG_PROG_FLAGS_PFX)) {
 			val = s + sizeof(TEST_TAG_PROG_FLAGS_PFX) - 1;
@@ -124,14 +194,70 @@ static int parse_test_spec(struct test_loader *tester,
 				errno = 0;
 				spec->prog_flags |= strtol(val, &e, 0);
 				if (errno || e[0] != '\0') {
-					ASSERT_FAIL("failed to parse test prog flags from '%s'", s);
-					return -EINVAL;
+					PRINT_FAIL("failed to parse test prog flags from '%s'\n",
+						   val);
+					err = -EINVAL;
+					goto cleanup;
 				}
 			}
 		}
 	}
 
+	if (spec->mode_mask == 0)
+		spec->mode_mask = PRIV;
+
+	if (!description)
+		description = spec->prog_name;
+
+	if (spec->mode_mask & PRIV) {
+		spec->priv.name = strdup(description);
+		if (!spec->priv.name) {
+			PRINT_FAIL("failed to allocate memory for priv.name\n");
+			err = -ENOMEM;
+			goto cleanup;
+		}
+	}
+
+	if (spec->mode_mask & UNPRIV) {
+		int descr_len = strlen(description);
+		const char *suffix = " @unpriv";
+		char *name;
+
+		name = malloc(descr_len + strlen(suffix) + 1);
+		if (!name) {
+			PRINT_FAIL("failed to allocate memory for unpriv.name\n");
+			err = -ENOMEM;
+			goto cleanup;
+		}
+
+		strcpy(name, description);
+		strcpy(&name[descr_len], suffix);
+		spec->unpriv.name = name;
+	}
+
+	if (spec->mode_mask & (PRIV | UNPRIV)) {
+		if (!has_unpriv_result)
+			spec->unpriv.expect_failure = spec->priv.expect_failure;
+
+		if (!spec->unpriv.expect_msgs) {
+			size_t sz = spec->priv.expect_msg_cnt * sizeof(void *);
+
+			spec->unpriv.expect_msgs = malloc(sz);
+			if (!spec->unpriv.expect_msgs) {
+				PRINT_FAIL("failed to allocate memory for unpriv.expect_msgs\n");
+				err = -ENOMEM;
+				goto cleanup;
+			}
+			memcpy(spec->unpriv.expect_msgs, spec->priv.expect_msgs, sz);
+			spec->unpriv.expect_msg_cnt = spec->priv.expect_msg_cnt;
+		}
+	}
+
 	return 0;
+
+cleanup:
+	free_test_spec(spec);
+	return err;
 }
 
 static void prepare_case(struct test_loader *tester,
@@ -148,7 +274,7 @@ static void prepare_case(struct test_loader *tester,
 
 	bpf_program__set_log_buf(prog, tester->log_buf, tester->log_buf_sz);
 
-	/* Make sure we set at least minimal log level, unless test requirest
+	/* Make sure we set at least minimal log level, unless test requires
 	 * even higher level already. Make sure to preserve independent log
 	 * level 4 (verifier stats), though.
 	 */
@@ -172,18 +298,18 @@ static void emit_verifier_log(const char *log_buf, bool force)
 }
 
 static void validate_case(struct test_loader *tester,
-			  struct test_spec *spec,
+			  struct test_subspec *subspec,
 			  struct bpf_object *obj,
 			  struct bpf_program *prog,
 			  int load_err)
 {
 	int i, j;
 
-	for (i = 0; i < spec->expect_msg_cnt; i++) {
+	for (i = 0; i < subspec->expect_msg_cnt; i++) {
 		char *match;
 		const char *expect_msg;
 
-		expect_msg = spec->expect_msgs[i];
+		expect_msg = subspec->expect_msgs[i];
 
 		match = strstr(tester->log_buf + tester->next_match_pos, expect_msg);
 		if (!ASSERT_OK_PTR(match, "expect_msg")) {
@@ -191,7 +317,8 @@ static void validate_case(struct test_loader *tester,
 			if (env.verbosity == VERBOSE_NONE)
 				emit_verifier_log(tester->log_buf, true /*force*/);
 			for (j = 0; j < i; j++)
-				fprintf(stderr, "MATCHED  MSG: '%s'\n", spec->expect_msgs[j]);
+				fprintf(stderr,
+					"MATCHED  MSG: '%s'\n", subspec->expect_msgs[j]);
 			fprintf(stderr, "EXPECTED MSG: '%s'\n", expect_msg);
 			return;
 		}
@@ -200,17 +327,169 @@ static void validate_case(struct test_loader *tester,
 	}
 }
 
+struct cap_state {
+	__u64 old_caps;
+	bool initialized;
+};
+
+static int drop_capabilities(struct cap_state *caps)
+{
+	const __u64 caps_to_drop = (1ULL << CAP_SYS_ADMIN | 1ULL << CAP_NET_ADMIN |
+				    1ULL << CAP_PERFMON   | 1ULL << CAP_BPF);
+	int err;
+
+	err = cap_disable_effective(caps_to_drop, &caps->old_caps);
+	if (err) {
+		PRINT_FAIL("failed to drop capabilities: %i, %s\n", err, strerror(err));
+		return err;
+	}
+
+	caps->initialized = true;
+	return 0;
+}
+
+static int restore_capabilities(struct cap_state *caps)
+{
+	int err;
+
+	if (!caps->initialized)
+		return 0;
+
+	err = cap_enable_effective(caps->old_caps, NULL);
+	if (err)
+		PRINT_FAIL("failed to restore capabilities: %i, %s\n", err, strerror(err));
+	caps->initialized = false;
+	return err;
+}
+
+static bool can_execute_unpriv(struct test_loader *tester, struct test_spec *spec)
+{
+	if (sysctl_unpriv_disabled < 0)
+		sysctl_unpriv_disabled = get_unpriv_disabled() ? 1 : 0;
+	if (sysctl_unpriv_disabled)
+		return false;
+	if ((spec->prog_flags & BPF_F_ANY_ALIGNMENT) && !EFFICIENT_UNALIGNED_ACCESS)
+		return false;
+	return true;
+}
+
+static bool is_unpriv_capable_map(struct bpf_map *map)
+{
+	enum bpf_map_type type;
+	__u32 flags;
+
+	type = bpf_map__type(map);
+
+	switch (type) {
+	case BPF_MAP_TYPE_HASH:
+	case BPF_MAP_TYPE_PERCPU_HASH:
+	case BPF_MAP_TYPE_HASH_OF_MAPS:
+		flags = bpf_map__map_flags(map);
+		return !(flags & BPF_F_ZERO_SEED);
+	case BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE:
+	case BPF_MAP_TYPE_ARRAY:
+	case BPF_MAP_TYPE_RINGBUF:
+	case BPF_MAP_TYPE_PROG_ARRAY:
+	case BPF_MAP_TYPE_CGROUP_ARRAY:
+	case BPF_MAP_TYPE_PERCPU_ARRAY:
+	case BPF_MAP_TYPE_USER_RINGBUF:
+	case BPF_MAP_TYPE_ARRAY_OF_MAPS:
+	case BPF_MAP_TYPE_CGROUP_STORAGE:
+	case BPF_MAP_TYPE_PERF_EVENT_ARRAY:
+		return true;
+	default:
+		return false;
+	}
+}
+
 /* this function is forced noinline and has short generic name to look better
  * in test_progs output (in case of a failure)
  */
 static noinline
 void run_subtest(struct test_loader *tester,
-		 const char *skel_name,
-		 skel_elf_bytes_fn elf_bytes_factory)
+		 struct bpf_object_open_opts *open_opts,
+		 const void *obj_bytes,
+		 size_t obj_byte_cnt,
+		 struct test_spec *spec,
+		 bool unpriv)
+{
+	struct test_subspec *subspec = unpriv ? &spec->unpriv : &spec->priv;
+	struct cap_state caps = {};
+	struct bpf_program *tprog;
+	struct bpf_object *tobj;
+	struct bpf_map *map;
+	int err;
+
+	if (!test__start_subtest(subspec->name))
+		return;
+
+	if (unpriv) {
+		if (!can_execute_unpriv(tester, spec)) {
+			test__skip();
+			test__end_subtest();
+			return;
+		}
+		if (drop_capabilities(&caps)) {
+			test__end_subtest();
+			return;
+		}
+	}
+
+	tobj = bpf_object__open_mem(obj_bytes, obj_byte_cnt, open_opts);
+	if (!ASSERT_OK_PTR(tobj, "obj_open_mem")) /* shouldn't happen */
+		goto subtest_cleanup;
+
+	bpf_object__for_each_program(tprog, tobj)
+		bpf_program__set_autoload(tprog, false);
+
+	bpf_object__for_each_program(tprog, tobj) {
+		/* only load specified program */
+		if (strcmp(bpf_program__name(tprog), spec->prog_name) == 0) {
+			bpf_program__set_autoload(tprog, true);
+			break;
+		}
+	}
+
+	prepare_case(tester, spec, tobj, tprog);
+
+	/* By default bpf_object__load() automatically creates all
+	 * maps declared in the skeleton. Some map types are only
+	 * allowed in priv mode. Disable autoload for such maps in
+	 * unpriv mode.
+	 */
+	bpf_object__for_each_map(map, tobj)
+		bpf_map__set_autocreate(map, !unpriv || is_unpriv_capable_map(map));
+
+	err = bpf_object__load(tobj);
+	if (subspec->expect_failure) {
+		if (!ASSERT_ERR(err, "unexpected_load_success")) {
+			emit_verifier_log(tester->log_buf, false /*force*/);
+			goto tobj_cleanup;
+		}
+	} else {
+		if (!ASSERT_OK(err, "unexpected_load_failure")) {
+			emit_verifier_log(tester->log_buf, true /*force*/);
+			goto tobj_cleanup;
+		}
+	}
+
+	emit_verifier_log(tester->log_buf, false /*force*/);
+	validate_case(tester, subspec, tobj, tprog, err);
+
+tobj_cleanup:
+	bpf_object__close(tobj);
+subtest_cleanup:
+	test__end_subtest();
+	restore_capabilities(&caps);
+}
+
+static void process_subtest(struct test_loader *tester,
+			    const char *skel_name,
+			    skel_elf_bytes_fn elf_bytes_factory)
 {
 	LIBBPF_OPTS(bpf_object_open_opts, open_opts, .object_name = skel_name);
-	struct bpf_object *obj = NULL, *tobj;
-	struct bpf_program *prog, *tprog;
+	struct bpf_object *obj = NULL;
+	struct bpf_program *prog;
 	const void *obj_bytes;
 	size_t obj_byte_cnt;
 	int err;
@@ -224,12 +503,8 @@ void run_subtest(struct test_loader *tester,
 		return;
 
 	bpf_object__for_each_program(prog, obj) {
-		const char *prog_name = bpf_program__name(prog);
 		struct test_spec spec;
 
-		if (!test__start_subtest(prog_name))
-			continue;
-
 		/* if we can't derive test specification, go to the next test */
 		err = parse_test_spec(tester, obj, prog, &spec);
 		if (err) {
@@ -238,41 +513,12 @@ void run_subtest(struct test_loader *tester,
 			continue;
 		}
 
-		tobj = bpf_object__open_mem(obj_bytes, obj_byte_cnt, &open_opts);
-		if (!ASSERT_OK_PTR(tobj, "obj_open_mem")) /* shouldn't happen */
-			continue;
+		if (spec.mode_mask & PRIV)
+			run_subtest(tester, &open_opts, obj_bytes, obj_byte_cnt, &spec, false);
+		if (spec.mode_mask & UNPRIV)
+			run_subtest(tester, &open_opts, obj_bytes, obj_byte_cnt, &spec, true);
 
-		bpf_object__for_each_program(tprog, tobj)
-			bpf_program__set_autoload(tprog, false);
-
-		bpf_object__for_each_program(tprog, tobj) {
-			/* only load specified program */
-			if (strcmp(bpf_program__name(tprog), prog_name) == 0) {
-				bpf_program__set_autoload(tprog, true);
-				break;
-			}
-		}
-
-		prepare_case(tester, &spec, tobj, tprog);
-
-		err = bpf_object__load(tobj);
-		if (spec.expect_failure) {
-			if (!ASSERT_ERR(err, "unexpected_load_success")) {
-				emit_verifier_log(tester->log_buf, false /*force*/);
-				goto tobj_cleanup;
-			}
-		} else {
-			if (!ASSERT_OK(err, "unexpected_load_failure")) {
-				emit_verifier_log(tester->log_buf, true /*force*/);
-				goto tobj_cleanup;
-			}
-		}
-
-		emit_verifier_log(tester->log_buf, false /*force*/);
-		validate_case(tester, &spec, tobj, tprog, err);
-
-tobj_cleanup:
-		bpf_object__close(tobj);
+		free_test_spec(&spec);
 	}
 
 	bpf_object__close(obj);
@@ -283,5 +529,5 @@ void test_loader__run_subtests(struct test_loader *tester,
 			       skel_elf_bytes_fn elf_bytes_factory)
 {
 	/* see comment in run_subtest() for why we do this function nesting */
-	run_subtest(tester, skel_name, elf_bytes_factory);
+	process_subtest(tester, skel_name, elf_bytes_factory);
 }
diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c
index 49a70d9beb0b..5b90eef09ade 100644
--- a/tools/testing/selftests/bpf/test_verifier.c
+++ b/tools/testing/selftests/bpf/test_verifier.c
@@ -33,13 +33,8 @@
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
-#ifdef HAVE_GENHDR
-# include "autoconf.h"
-#else
-# if defined(__i386) || defined(__x86_64) || defined(__s390x__) || defined(__aarch64__)
-#  define CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS 1
-# endif
-#endif
+#include "autoconf_helper.h"
+#include "unpriv_helpers.h"
 #include "cap_helpers.h"
 #include "bpf_rand.h"
 #include "bpf_util.h"
@@ -1665,22 +1660,6 @@ static bool is_admin(void)
 	return (caps & ADMIN_CAPS) == ADMIN_CAPS;
 }
 
-static void get_unpriv_disabled()
-{
-	char buf[2];
-	FILE *fd;
-
-	fd = fopen("/proc/sys/"UNPRIV_SYSCTL, "r");
-	if (!fd) {
-		perror("fopen /proc/sys/"UNPRIV_SYSCTL);
-		unpriv_disabled = true;
-		return;
-	}
-	if (fgets(buf, 2, fd) == buf && atoi(buf))
-		unpriv_disabled = true;
-	fclose(fd);
-}
-
 static bool test_as_unpriv(struct bpf_test *test)
 {
 #ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
diff --git a/tools/testing/selftests/bpf/unpriv_helpers.c b/tools/testing/selftests/bpf/unpriv_helpers.c
new file mode 100644
index 000000000000..2a6efbd0401e
--- /dev/null
+++ b/tools/testing/selftests/bpf/unpriv_helpers.c
@@ -0,0 +1,26 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <stdbool.h>
+#include <stdlib.h>
+#include <error.h>
+#include <stdio.h>
+
+#include "unpriv_helpers.h"
+
+bool get_unpriv_disabled(void)
+{
+	bool disabled;
+	char buf[2];
+	FILE *fd;
+
+	fd = fopen("/proc/sys/" UNPRIV_SYSCTL, "r");
+	if (fd) {
+		disabled = (fgets(buf, 2, fd) == buf && atoi(buf));
+		fclose(fd);
+	} else {
+		perror("fopen /proc/sys/" UNPRIV_SYSCTL);
+		disabled = true;
+	}
+
+	return disabled;
+}
diff --git a/tools/testing/selftests/bpf/unpriv_helpers.h b/tools/testing/selftests/bpf/unpriv_helpers.h
new file mode 100644
index 000000000000..151f67329665
--- /dev/null
+++ b/tools/testing/selftests/bpf/unpriv_helpers.h
@@ -0,0 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <stdbool.h>
+
+#define UNPRIV_SYSCTL "kernel/unprivileged_bpf_disabled"
+
+bool get_unpriv_disabled(void);
-- 
cgit v1.2.3-70-g09d2


From 19a8e06f5f9155caf1a5577a0f7969eee13d0cbb Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:45 +0200
Subject: selftests/bpf: Tests execution support for test_loader.c

Extends test_loader.c:test_loader__run_subtests() by allowing to
execute BPF_PROG_TEST_RUN bpf command for selected programs.
This is similar to functionality provided by test_verifier.

Adds the following new attributes controlling test_loader behavior:

  __retval(...)
  __retval_unpriv(...)

* If any of these attributes is present, the annotated program would
  be executed using libbpf's bpf_prog_test_run_opts() function.
* If __retval is present, the test run would be done for program
  loaded in privileged mode.
* If __retval_unpriv is present, the test run would be done for
  program loaded in unprivileged mode.
* To mimic test_verifier behavior, the actual run is initiated in
  privileged mode.
* The value returned by a test run is compared against retval
  parameter.

The retval attribute takes one of the following parameters:
- a decimal number
- a hexadecimal number (must start from '0x')
- any of a three special literals (provided for compatibility with
  test_verifier):
  - INT_MIN
  - POINTER_VALUE
  - TEST_DATA_LEN

An example of the attribute usage:

  SEC("socket")
  __description("return 42")
  __success __success_unpriv __retval(42)
  __naked void the_42_test(void)
  {
          asm volatile ("                                 \
          r0 = 42;                                        \
          exit;                                           \
  "       ::: __clobber_all);
  }

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-5-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_misc.h |  15 +++
 tools/testing/selftests/bpf/test_loader.c    | 149 ++++++++++++++++++++++++---
 2 files changed, 150 insertions(+), 14 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 9defc217a5bd..6e3b4903c541 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -30,6 +30,15 @@
  * __failure         Expect program load failure in privileged mode.
  * __failure_unpriv  Expect program load failure in unprivileged mode.
  *
+ * __retval          Execute the program using BPF_PROG_TEST_RUN command,
+ *                   expect return value to match passed parameter:
+ *                   - a decimal number
+ *                   - a hexadecimal number, when starts from 0x
+ *                   - literal INT_MIN
+ *                   - literal POINTER_VALUE (see definition below)
+ *                   - literal TEST_DATA_LEN (see definition below)
+ * __retval_unpriv   Same, but load program in unprivileged mode.
+ *
  * __description     Text to be used instead of a program name for display
  *                   and filtering purposes.
  *
@@ -54,6 +63,8 @@
 #define __success_unpriv	__attribute__((btf_decl_tag("comment:test_expect_success_unpriv")))
 #define __log_level(lvl)	__attribute__((btf_decl_tag("comment:test_log_level="#lvl)))
 #define __flag(flag)		__attribute__((btf_decl_tag("comment:test_prog_flags="#flag)))
+#define __retval(val)		__attribute__((btf_decl_tag("comment:test_retval="#val)))
+#define __retval_unpriv(val)	__attribute__((btf_decl_tag("comment:test_retval_unpriv="#val)))
 
 /* Convenience macro for use with 'asm volatile' blocks */
 #define __naked __attribute__((naked))
@@ -65,6 +76,10 @@
 #define __imm_ptr(name) [name]"p"(&name)
 #define __imm_insn(name, expr) [name]"i"(*(long *)&(expr))
 
+/* Magic constants used with __retval() */
+#define POINTER_VALUE	0xcafe4all
+#define TEST_DATA_LEN	64
+
 #if defined(__TARGET_ARCH_x86)
 #define SYSCALL_WRAPPER 1
 #define SYS_PREFIX "__x64_"
diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index 41cddb303885..47e9e076bc8f 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -23,6 +23,12 @@
 #define TEST_TAG_LOG_LEVEL_PFX "comment:test_log_level="
 #define TEST_TAG_PROG_FLAGS_PFX "comment:test_prog_flags="
 #define TEST_TAG_DESCRIPTION_PFX "comment:test_description="
+#define TEST_TAG_RETVAL_PFX "comment:test_retval="
+#define TEST_TAG_RETVAL_PFX_UNPRIV "comment:test_retval_unpriv="
+
+/* Warning: duplicated in bpf_misc.h */
+#define POINTER_VALUE	0xcafe4all
+#define TEST_DATA_LEN	64
 
 #ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 #define EFFICIENT_UNALIGNED_ACCESS 1
@@ -42,6 +48,8 @@ struct test_subspec {
 	bool expect_failure;
 	const char **expect_msgs;
 	size_t expect_msg_cnt;
+	int retval;
+	bool execute;
 };
 
 struct test_spec {
@@ -96,6 +104,46 @@ static int push_msg(const char *msg, struct test_subspec *subspec)
 	return 0;
 }
 
+static int parse_int(const char *str, int *val, const char *name)
+{
+	char *end;
+	long tmp;
+
+	errno = 0;
+	if (str_has_pfx(str, "0x"))
+		tmp = strtol(str + 2, &end, 16);
+	else
+		tmp = strtol(str, &end, 10);
+	if (errno || end[0] != '\0') {
+		PRINT_FAIL("failed to parse %s from '%s'\n", name, str);
+		return -EINVAL;
+	}
+	*val = tmp;
+	return 0;
+}
+
+static int parse_retval(const char *str, int *val, const char *name)
+{
+	struct {
+		char *name;
+		int val;
+	} named_values[] = {
+		{ "INT_MIN"      , INT_MIN },
+		{ "POINTER_VALUE", POINTER_VALUE },
+		{ "TEST_DATA_LEN", TEST_DATA_LEN },
+	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(named_values); ++i) {
+		if (strcmp(str, named_values[i].name) != 0)
+			continue;
+		*val = named_values[i].val;
+		return 0;
+	}
+
+	return parse_int(str, val, name);
+}
+
 /* Uses btf_decl_tag attributes to describe the expected test
  * behavior, see bpf_misc.h for detailed description of each attribute
  * and attribute combinations.
@@ -107,6 +155,7 @@ static int parse_test_spec(struct test_loader *tester,
 {
 	const char *description = NULL;
 	bool has_unpriv_result = false;
+	bool has_unpriv_retval = false;
 	int func_id, i, err = 0;
 	struct btf *btf;
 
@@ -129,7 +178,7 @@ static int parse_test_spec(struct test_loader *tester,
 	for (i = 1; i < btf__type_cnt(btf); i++) {
 		const char *s, *val, *msg;
 		const struct btf_type *t;
-		char *e;
+		int tmp;
 
 		t = btf__type_by_id(btf, i);
 		if (!btf_is_decl_tag(t))
@@ -167,15 +216,26 @@ static int parse_test_spec(struct test_loader *tester,
 			if (err)
 				goto cleanup;
 			spec->mode_mask |= UNPRIV;
+		} else if (str_has_pfx(s, TEST_TAG_RETVAL_PFX)) {
+			val = s + sizeof(TEST_TAG_RETVAL_PFX) - 1;
+			err = parse_retval(val, &spec->priv.retval, "__retval");
+			if (err)
+				goto cleanup;
+			spec->priv.execute = true;
+			spec->mode_mask |= PRIV;
+		} else if (str_has_pfx(s, TEST_TAG_RETVAL_PFX_UNPRIV)) {
+			val = s + sizeof(TEST_TAG_RETVAL_PFX_UNPRIV) - 1;
+			err = parse_retval(val, &spec->unpriv.retval, "__retval_unpriv");
+			if (err)
+				goto cleanup;
+			spec->mode_mask |= UNPRIV;
+			spec->unpriv.execute = true;
+			has_unpriv_retval = true;
 		} else if (str_has_pfx(s, TEST_TAG_LOG_LEVEL_PFX)) {
 			val = s + sizeof(TEST_TAG_LOG_LEVEL_PFX) - 1;
-			errno = 0;
-			spec->log_level = strtol(val, &e, 0);
-			if (errno || e[0] != '\0') {
-				PRINT_FAIL("failed to parse test log level from '%s'\n", s);
-				err = -EINVAL;
+			err = parse_int(val, &spec->log_level, "test log level");
+			if (err)
 				goto cleanup;
-			}
 		} else if (str_has_pfx(s, TEST_TAG_PROG_FLAGS_PFX)) {
 			val = s + sizeof(TEST_TAG_PROG_FLAGS_PFX) - 1;
 			if (strcmp(val, "BPF_F_STRICT_ALIGNMENT") == 0) {
@@ -191,14 +251,10 @@ static int parse_test_spec(struct test_loader *tester,
 			} else if (strcmp(val, "BPF_F_XDP_HAS_FRAGS") == 0) {
 				spec->prog_flags |= BPF_F_XDP_HAS_FRAGS;
 			} else /* assume numeric value */ {
-				errno = 0;
-				spec->prog_flags |= strtol(val, &e, 0);
-				if (errno || e[0] != '\0') {
-					PRINT_FAIL("failed to parse test prog flags from '%s'\n",
-						   val);
-					err = -EINVAL;
+				err = parse_int(val, &tmp, "test prog flags");
+				if (err)
 					goto cleanup;
-				}
+				spec->prog_flags |= tmp;
 			}
 		}
 	}
@@ -239,6 +295,11 @@ static int parse_test_spec(struct test_loader *tester,
 		if (!has_unpriv_result)
 			spec->unpriv.expect_failure = spec->priv.expect_failure;
 
+		if (!has_unpriv_retval) {
+			spec->unpriv.retval = spec->priv.retval;
+			spec->unpriv.execute = spec->priv.execute;
+		}
+
 		if (!spec->unpriv.expect_msgs) {
 			size_t sz = spec->priv.expect_msg_cnt * sizeof(void *);
 
@@ -402,6 +463,51 @@ static bool is_unpriv_capable_map(struct bpf_map *map)
 	}
 }
 
+static int do_prog_test_run(int fd_prog, int *retval)
+{
+	__u8 tmp_out[TEST_DATA_LEN << 2] = {};
+	__u8 tmp_in[TEST_DATA_LEN] = {};
+	int err, saved_errno;
+	LIBBPF_OPTS(bpf_test_run_opts, topts,
+		.data_in = tmp_in,
+		.data_size_in = sizeof(tmp_in),
+		.data_out = tmp_out,
+		.data_size_out = sizeof(tmp_out),
+		.repeat = 1,
+	);
+
+	err = bpf_prog_test_run_opts(fd_prog, &topts);
+	saved_errno = errno;
+
+	if (err) {
+		PRINT_FAIL("FAIL: Unexpected bpf_prog_test_run error: %d (%s) ",
+			   saved_errno, strerror(saved_errno));
+		return err;
+	}
+
+	ASSERT_OK(0, "bpf_prog_test_run");
+	*retval = topts.retval;
+
+	return 0;
+}
+
+static bool should_do_test_run(struct test_spec *spec, struct test_subspec *subspec)
+{
+	if (!subspec->execute)
+		return false;
+
+	if (subspec->expect_failure)
+		return false;
+
+	if ((spec->prog_flags & BPF_F_ANY_ALIGNMENT) && !EFFICIENT_UNALIGNED_ACCESS) {
+		if (env.verbosity != VERBOSE_NONE)
+			printf("alignment prevents execution\n");
+		return false;
+	}
+
+	return true;
+}
+
 /* this function is forced noinline and has short generic name to look better
  * in test_progs output (in case of a failure)
  */
@@ -418,6 +524,7 @@ void run_subtest(struct test_loader *tester,
 	struct bpf_program *tprog;
 	struct bpf_object *tobj;
 	struct bpf_map *map;
+	int retval;
 	int err;
 
 	if (!test__start_subtest(subspec->name))
@@ -476,6 +583,20 @@ void run_subtest(struct test_loader *tester,
 	emit_verifier_log(tester->log_buf, false /*force*/);
 	validate_case(tester, subspec, tobj, tprog, err);
 
+	if (should_do_test_run(spec, subspec)) {
+		/* For some reason test_verifier executes programs
+		 * with all capabilities restored. Do the same here.
+		 */
+		if (!restore_capabilities(&caps))
+			goto tobj_cleanup;
+
+		do_prog_test_run(bpf_program__fd(tprog), &retval);
+		if (retval != subspec->retval && subspec->retval != POINTER_VALUE) {
+			PRINT_FAIL("Unexpected retval: %d != %d\n", retval, subspec->retval);
+			goto tobj_cleanup;
+		}
+	}
+
 tobj_cleanup:
 	bpf_object__close(tobj);
 subtest_cleanup:
-- 
cgit v1.2.3-70-g09d2


From 55108621a35e42f773de5d4b20cf7a14d6d53503 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:46 +0200
Subject: selftests/bpf: prog_tests entry point for migrated test_verifier
 tests

prog_tests/verifier.c would be used as a host for verifier/*.c tests
migrated to use inline assembly and run from test_progs.

The run_test_aux() function mimics the test_verifier behavior
dropping CAP_SYS_ADMIN upon entry.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-6-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c | 29 +++++++++++++++++++++++
 1 file changed, 29 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/verifier.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
new file mode 100644
index 000000000000..aa63f5d84d97
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -0,0 +1,29 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <test_progs.h>
+
+#include "cap_helpers.h"
+
+__maybe_unused
+static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
+{
+	struct test_loader tester = {};
+	__u64 old_caps;
+	int err;
+
+	/* test_verifier tests are executed w/o CAP_SYS_ADMIN, do the same here */
+	err = cap_disable_effective(1ULL << CAP_SYS_ADMIN, &old_caps);
+	if (err) {
+		PRINT_FAIL("failed to drop CAP_SYS_ADMIN: %i, %s\n", err, strerror(err));
+		return;
+	}
+
+	test_loader__run_subtests(&tester, skel_name, elf_bytes_factory);
+	test_loader_fini(&tester);
+
+	err = cap_enable_effective(old_caps, NULL);
+	if (err)
+		PRINT_FAIL("failed to restore CAP_SYS_ADMIN: %i, %s\n", err, strerror(err));
+}
+
+#define RUN(skel) run_tests_aux(#skel, skel##__elf_bytes)
-- 
cgit v1.2.3-70-g09d2


From 9d0f1568ad5ba29feddc0897e2ccc7d6de6713c8 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:47 +0200
Subject: selftests/bpf: verifier/and.c converted to inline assembly

Test verifier/and.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-7-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   3 +
 tools/testing/selftests/bpf/progs/verifier_and.c  | 107 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/and.c        |  68 --------------
 3 files changed, 110 insertions(+), 68 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_and.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/and.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index aa63f5d84d97..34526f6d5ab1 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -3,6 +3,7 @@
 #include <test_progs.h>
 
 #include "cap_helpers.h"
+#include "verifier_and.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -27,3 +28,5 @@ static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_fac
 }
 
 #define RUN(skel) run_tests_aux(#skel, skel##__elf_bytes)
+
+void test_verifier_and(void)                  { RUN(verifier_and); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_and.c b/tools/testing/selftests/bpf/progs/verifier_and.c
new file mode 100644
index 000000000000..e97e518516b6
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_and.c
@@ -0,0 +1,107 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/and.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("socket")
+__description("invalid and of negative number")
+__failure __msg("R0 max value is outside of the allowed memory range")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void invalid_and_of_negative_number(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= -4;					\
+	r1 <<= 2;					\
+	r0 += r1;					\
+l0_%=:	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid range check")
+__failure __msg("R0 max value is outside of the allowed memory range")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void invalid_range_check(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	r9 = 1;						\
+	w1 %%= 2;					\
+	w1 += 1;					\
+	w9 &= w1;					\
+	w9 += 1;					\
+	w9 >>= 1;					\
+	w3 = 1;						\
+	w3 -= w9;					\
+	w3 *= 0x10000000;				\
+	r0 += r3;					\
+	*(u32*)(r0 + 0) = r3;				\
+l0_%=:	r0 = r0;					\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check known subreg with unknown reg")
+__success __failure_unpriv __msg_unpriv("R1 !read_ok")
+__retval(0)
+__naked void known_subreg_with_unknown_reg(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r0 <<= 32;					\
+	r0 += 1;					\
+	r0 &= 0xFFFF1234;				\
+	/* Upper bits are unknown but AND above masks out 1 zero'ing lower bits */\
+	if w0 < 1 goto l0_%=;				\
+	r1 = *(u32*)(r1 + 512);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/and.c b/tools/testing/selftests/bpf/verifier/and.c
deleted file mode 100644
index 7d7ebee5cc7a..000000000000
--- a/tools/testing/selftests/bpf/verifier/and.c
+++ /dev/null
@@ -1,68 +0,0 @@
-{
-	"invalid and of negative number",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, -4),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 max value is outside of the allowed memory range",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid range check",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 12),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_9, 1),
-	BPF_ALU32_IMM(BPF_MOD, BPF_REG_1, 2),
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_ALU32_REG(BPF_AND, BPF_REG_9, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_9, 1),
-	BPF_ALU32_IMM(BPF_RSH, BPF_REG_9, 1),
-	BPF_MOV32_IMM(BPF_REG_3, 1),
-	BPF_ALU32_REG(BPF_SUB, BPF_REG_3, BPF_REG_9),
-	BPF_ALU32_IMM(BPF_MUL, BPF_REG_3, 0x10000000),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_3),
-	BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_3, 0),
-	BPF_MOV64_REG(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 max value is outside of the allowed memory range",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"check known subreg with unknown reg",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_0, 32),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 0xFFFF1234),
-	/* Upper bits are unknown but AND above masks out 1 zero'ing lower bits */
-	BPF_JMP32_IMM(BPF_JLT, BPF_REG_0, 1, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_1, 512),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 !read_ok",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.retval = 0
-},
-- 
cgit v1.2.3-70-g09d2


From a3c830ae02093315a4526fa74fb7d1f66989d895 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:48 +0200
Subject: selftests/bpf: verifier/array_access.c converted to inline assembly

Test verifier/array_access.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-8-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_array_access.c    | 529 +++++++++++++++++++++
 .../testing/selftests/bpf/verifier/array_access.c  | 379 ---------------
 3 files changed, 531 insertions(+), 379 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_array_access.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/array_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 34526f6d5ab1..60eb0f38ed92 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -4,6 +4,7 @@
 
 #include "cap_helpers.h"
 #include "verifier_and.skel.h"
+#include "verifier_array_access.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -30,3 +31,4 @@ static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_fac
 #define RUN(skel) run_tests_aux(#skel, skel##__elf_bytes)
 
 void test_verifier_and(void)                  { RUN(verifier_and); }
+void test_verifier_array_access(void)         { RUN(verifier_array_access); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_array_access.c b/tools/testing/selftests/bpf/progs/verifier_array_access.c
new file mode 100644
index 000000000000..95d7ecc12963
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_array_access.c
@@ -0,0 +1,529 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/array_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+	__uint(map_flags, BPF_F_RDONLY_PROG);
+} map_array_ro SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+	__uint(map_flags, BPF_F_WRONLY_PROG);
+} map_array_wo SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("socket")
+__description("valid map access into an array with a constant")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0)
+__naked void an_array_with_a_constant_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("valid map access into an array with a register")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void an_array_with_a_register_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 4;						\
+	r1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("valid map access into an array with a variable")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void an_array_with_a_variable_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	if r1 >= %[max_entries] goto l0_%=;		\
+	r1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(max_entries, MAX_ENTRIES),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("valid map access into an array with a signed variable")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void array_with_a_signed_variable(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	if w1 s> 0xffffffff goto l1_%=;			\
+	w1 = 0;						\
+l1_%=:	w2 = %[max_entries];				\
+	if r2 s> r1 goto l2_%=;				\
+	w1 = 0;						\
+l2_%=:	w1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(max_entries, MAX_ENTRIES),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access into an array with a constant")
+__failure __msg("invalid access to map value, value_size=48 off=48 size=8")
+__failure_unpriv
+__naked void an_array_with_a_constant_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + %[__imm_0]) = r1;			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, (MAX_ENTRIES + 1) << 2),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access into an array with a register")
+__failure __msg("R0 min value is outside of the allowed memory range")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void an_array_with_a_register_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = %[__imm_0];				\
+	r1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, MAX_ENTRIES + 1),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access into an array with a variable")
+__failure
+__msg("R0 unbounded memory access, make sure to bounds check any such access")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void an_array_with_a_variable_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	r1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access into an array with no floor check")
+__failure __msg("R0 unbounded memory access")
+__failure_unpriv __msg_unpriv("R0 leaks addr")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void array_with_no_floor_check(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r0 + 0);				\
+	w2 = %[max_entries];				\
+	if r2 s> r1 goto l1_%=;				\
+	w1 = 0;						\
+l1_%=:	w1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(max_entries, MAX_ENTRIES),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access into an array with a invalid max check")
+__failure __msg("invalid access to map value, value_size=48 off=44 size=8")
+__failure_unpriv __msg_unpriv("R0 leaks addr")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void with_a_invalid_max_check_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	w2 = %[__imm_0];				\
+	if r2 > r1 goto l1_%=;				\
+	w1 = 0;						\
+l1_%=:	w1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, MAX_ENTRIES + 1),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access into an array with a invalid max check")
+__failure __msg("R0 pointer += pointer")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void with_a_invalid_max_check_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r8 = r0;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 += r8;					\
+	r0 = *(u32*)(r0 + %[test_val_foo]);		\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("valid read map access into a read-only array 1")
+__success __success_unpriv __retval(28)
+__naked void a_read_only_array_1_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_ro] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 = *(u32*)(r0 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_ro)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("valid read map access into a read-only array 2")
+__success __retval(65507)
+__naked void a_read_only_array_2_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_ro] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 &= 0xffff;					\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_ro)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid write map access into a read-only array 1")
+__failure __msg("write into map forbidden")
+__failure_unpriv
+__naked void a_read_only_array_1_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_ro] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 42;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_ro)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("invalid write map access into a read-only array 2")
+__failure __msg("write into map forbidden")
+__naked void a_read_only_array_2_2(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_ro] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = 0;						\
+	r3 = r0;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_skb_load_bytes),
+	  __imm_addr(map_array_ro)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("valid write map access into a write-only array 1")
+__success __success_unpriv __retval(1)
+__naked void a_write_only_array_1_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_wo] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 42;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_wo)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("valid write map access into a write-only array 2")
+__success __retval(0)
+__naked void a_write_only_array_2_1(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_wo] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = 0;						\
+	r3 = r0;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_skb_load_bytes),
+	  __imm_addr(map_array_wo)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid read map access into a write-only array 1")
+__failure __msg("read from map forbidden")
+__failure_unpriv
+__naked void a_write_only_array_1_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_wo] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 = *(u64*)(r0 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_wo)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("invalid read map access into a write-only array 2")
+__failure __msg("read from map forbidden")
+__naked void a_write_only_array_2_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_wo] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_wo)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/array_access.c b/tools/testing/selftests/bpf/verifier/array_access.c
deleted file mode 100644
index 1b138cd2b187..000000000000
--- a/tools/testing/selftests/bpf/verifier/array_access.c
+++ /dev/null
@@ -1,379 +0,0 @@
-{
-	"valid map access into an array with a constant",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"valid map access into an array with a register",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"valid map access into an array with a variable",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_1, MAX_ENTRIES, 3),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"valid map access into an array with a signed variable",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP32_IMM(BPF_JSGT, BPF_REG_1, 0xffffffff, 1),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, MAX_ENTRIES),
-	BPF_JMP_REG(BPF_JSGT, BPF_REG_2, BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_ALU32_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid map access into an array with a constant",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, (MAX_ENTRIES + 1) << 2,
-		   offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=48 size=8",
-	.result = REJECT,
-},
-{
-	"invalid map access into an array with a register",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_1, MAX_ENTRIES + 1),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 min value is outside of the allowed memory range",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid map access into an array with a variable",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 unbounded memory access, make sure to bounds check any such access",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid map access into an array with no floor check",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV32_IMM(BPF_REG_2, MAX_ENTRIES),
-	BPF_JMP_REG(BPF_JSGT, BPF_REG_2, BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_ALU32_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.errstr = "R0 unbounded memory access",
-	.result_unpriv = REJECT,
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid map access into an array with a invalid max check",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV32_IMM(BPF_REG_2, MAX_ENTRIES + 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_ALU32_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.errstr = "invalid access to map value, value_size=48 off=44 size=8",
-	.result_unpriv = REJECT,
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid map access into an array with a invalid max check",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 10),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_8),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0,
-		    offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3, 11 },
-	.errstr = "R0 pointer += pointer",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"valid read map access into a read-only array 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_ro = { 3 },
-	.result = ACCEPT,
-	.retval = 28,
-},
-{
-	"valid read map access into a read-only array 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-		     BPF_FUNC_csum_diff),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 0xffff),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_array_ro = { 3 },
-	.result = ACCEPT,
-	.retval = 65507,
-},
-{
-	"invalid write map access into a read-only array 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_ro = { 3 },
-	.result = REJECT,
-	.errstr = "write into map forbidden",
-},
-{
-	"invalid write map access into a read-only array 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-		     BPF_FUNC_skb_load_bytes),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_array_ro = { 4 },
-	.result = REJECT,
-	.errstr = "write into map forbidden",
-},
-{
-	"valid write map access into a write-only array 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 42),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_wo = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"valid write map access into a write-only array 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-		     BPF_FUNC_skb_load_bytes),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_array_wo = { 4 },
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"invalid read map access into a write-only array 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_wo = { 3 },
-	.result = REJECT,
-	.errstr = "read from map forbidden",
-},
-{
-	"invalid read map access into a write-only array 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-		     BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_array_wo = { 3 },
-	.result = REJECT,
-	.errstr = "read from map forbidden",
-},
-- 
cgit v1.2.3-70-g09d2


From 0ccbe4956d6c20fa0a09a72d2033c49f0976ed6c Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:49 +0200
Subject: selftests/bpf: verifier/basic_stack.c converted to inline assembly

Test verifier/basic_stack.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-9-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_basic_stack.c     | 100 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/basic_stack.c |  64 -------------
 3 files changed, 102 insertions(+), 64 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_basic_stack.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/basic_stack.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 60eb0f38ed92..95a3151db052 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -5,6 +5,7 @@
 #include "cap_helpers.h"
 #include "verifier_and.skel.h"
 #include "verifier_array_access.skel.h"
+#include "verifier_basic_stack.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -32,3 +33,4 @@ static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_fac
 
 void test_verifier_and(void)                  { RUN(verifier_and); }
 void test_verifier_array_access(void)         { RUN(verifier_array_access); }
+void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_basic_stack.c b/tools/testing/selftests/bpf/progs/verifier_basic_stack.c
new file mode 100644
index 000000000000..359df865a8f3
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_basic_stack.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/basic_stack.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("socket")
+__description("stack out of bounds")
+__failure __msg("invalid write to stack")
+__failure_unpriv
+__naked void stack_out_of_bounds(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 + 8) = r1;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("uninitialized stack1")
+__failure __msg("invalid indirect read from stack")
+__failure_unpriv
+__naked void uninitialized_stack1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("uninitialized stack2")
+__failure __msg("invalid read from stack")
+__failure_unpriv
+__naked void uninitialized_stack2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r0 = *(u64*)(r2 - 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid fp arithmetic")
+__failure __msg("R1 subtraction from stack pointer")
+__failure_unpriv
+__naked void invalid_fp_arithmetic(void)
+{
+	/* If this gets ever changed, make sure JITs can deal with it. */
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = r10;					\
+	r1 -= 8;					\
+	*(u64*)(r1 + 0) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("non-invalid fp arithmetic")
+__success __success_unpriv __retval(0)
+__naked void non_invalid_fp_arithmetic(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("misaligned read from stack")
+__failure __msg("misaligned stack access")
+__failure_unpriv
+__naked void misaligned_read_from_stack(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r0 = *(u64*)(r2 - 4);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/basic_stack.c b/tools/testing/selftests/bpf/verifier/basic_stack.c
deleted file mode 100644
index f995777dddb3..000000000000
--- a/tools/testing/selftests/bpf/verifier/basic_stack.c
+++ /dev/null
@@ -1,64 +0,0 @@
-{
-	"stack out of bounds",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, 8, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid write to stack",
-	.result = REJECT,
-},
-{
-	"uninitialized stack1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 2 },
-	.errstr = "invalid indirect read from stack",
-	.result = REJECT,
-},
-{
-	"uninitialized stack2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, -8),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid read from stack",
-	.result = REJECT,
-},
-{
-	"invalid fp arithmetic",
-	/* If this gets ever changed, make sure JITs can deal with it. */
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 subtraction from stack pointer",
-	.result = REJECT,
-},
-{
-	"non-invalid fp arithmetic",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-},
-{
-	"misaligned read from stack",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, -4),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "misaligned stack access",
-	.result = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From 7605f94b3492328f37815c9b5749ffba5c76da84 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:50 +0200
Subject: selftests/bpf: verifier/bounds_deduction.c converted to inline
 assembly

Test verifier/bounds_deduction.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-10-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_bounds_deduction.c          | 171 +++++++++++++++++++++
 .../selftests/bpf/verifier/bounds_deduction.c      | 136 ----------------
 3 files changed, 173 insertions(+), 136 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_bounds_deduction.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/bounds_deduction.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 95a3151db052..a8cfef92ed64 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -6,6 +6,7 @@
 #include "verifier_and.skel.h"
 #include "verifier_array_access.skel.h"
 #include "verifier_basic_stack.skel.h"
+#include "verifier_bounds_deduction.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -34,3 +35,4 @@ static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_fac
 void test_verifier_and(void)                  { RUN(verifier_and); }
 void test_verifier_array_access(void)         { RUN(verifier_array_access); }
 void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
+void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds_deduction.c b/tools/testing/selftests/bpf/progs/verifier_bounds_deduction.c
new file mode 100644
index 000000000000..c506afbdd936
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_bounds_deduction.c
@@ -0,0 +1,171 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/bounds_deduction.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("check deducing bounds from const, 1")
+__failure __msg("R0 tried to subtract pointer from scalar")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__naked void deducing_bounds_from_const_1(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	if r0 s>= 1 goto l0_%=;				\
+l0_%=:	r0 -= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 2")
+__success __failure_unpriv
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__retval(1)
+__naked void deducing_bounds_from_const_2(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	if r0 s>= 1 goto l0_%=;				\
+	exit;						\
+l0_%=:	if r0 s<= 1 goto l1_%=;				\
+	exit;						\
+l1_%=:	r1 -= r0;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 3")
+__failure __msg("R0 tried to subtract pointer from scalar")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__naked void deducing_bounds_from_const_3(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	if r0 s<= 0 goto l0_%=;				\
+l0_%=:	r0 -= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 4")
+__success __failure_unpriv
+__msg_unpriv("R6 has pointer with unsupported alu operation")
+__retval(0)
+__naked void deducing_bounds_from_const_4(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r0 = 0;						\
+	if r0 s<= 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	if r0 s>= 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r6 -= r0;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 5")
+__failure __msg("R0 tried to subtract pointer from scalar")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__naked void deducing_bounds_from_const_5(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	if r0 s>= 1 goto l0_%=;				\
+	r0 -= r1;					\
+l0_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 6")
+__failure __msg("R0 tried to subtract pointer from scalar")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__naked void deducing_bounds_from_const_6(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	if r0 s>= 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r0 -= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 7")
+__failure __msg("dereference of modified ctx ptr")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void deducing_bounds_from_const_7(void)
+{
+	asm volatile ("					\
+	r0 = %[__imm_0];				\
+	if r0 s>= 0 goto l0_%=;				\
+l0_%=:	r1 -= r0;					\
+	r0 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, ~0),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 8")
+__failure __msg("negative offset ctx ptr R1 off=-1 disallowed")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void deducing_bounds_from_const_8(void)
+{
+	asm volatile ("					\
+	r0 = %[__imm_0];				\
+	if r0 s>= 0 goto l0_%=;				\
+	r1 += r0;					\
+l0_%=:	r0 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, ~0),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 9")
+__failure __msg("R0 tried to subtract pointer from scalar")
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__naked void deducing_bounds_from_const_9(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	if r0 s>= 0 goto l0_%=;				\
+l0_%=:	r0 -= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from const, 10")
+__failure
+__msg("math between ctx pointer and register with unbounded min value is not allowed")
+__failure_unpriv
+__naked void deducing_bounds_from_const_10(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	if r0 s<= 0 goto l0_%=;				\
+l0_%=:	/* Marks reg as unknown. */			\
+	r0 = -r0;					\
+	r0 -= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/bounds_deduction.c b/tools/testing/selftests/bpf/verifier/bounds_deduction.c
deleted file mode 100644
index 3931c481e30c..000000000000
--- a/tools/testing/selftests/bpf/verifier/bounds_deduction.c
+++ /dev/null
@@ -1,136 +0,0 @@
-{
-	"check deducing bounds from const, 1",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 1),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 1, 0),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "R0 tried to subtract pointer from scalar",
-	.result = REJECT,
-},
-{
-	"check deducing bounds from const, 2",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 1),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 1, 1),
-		BPF_EXIT_INSN(),
-		BPF_JMP_IMM(BPF_JSLE, BPF_REG_0, 1, 1),
-		BPF_EXIT_INSN(),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_1, BPF_REG_0),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"check deducing bounds from const, 3",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JSLE, BPF_REG_0, 0, 0),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "R0 tried to subtract pointer from scalar",
-	.result = REJECT,
-},
-{
-	"check deducing bounds from const, 4",
-	.insns = {
-		BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JSLE, BPF_REG_0, 0, 1),
-		BPF_EXIT_INSN(),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 0, 1),
-		BPF_EXIT_INSN(),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_6, BPF_REG_0),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R6 has pointer with unsupported alu operation",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"check deducing bounds from const, 5",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 1, 1),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "R0 tried to subtract pointer from scalar",
-	.result = REJECT,
-},
-{
-	"check deducing bounds from const, 6",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 0, 1),
-		BPF_EXIT_INSN(),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "R0 tried to subtract pointer from scalar",
-	.result = REJECT,
-},
-{
-	"check deducing bounds from const, 7",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, ~0),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 0, 0),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_1, BPF_REG_0),
-		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-			    offsetof(struct __sk_buff, mark)),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "dereference of modified ctx ptr",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"check deducing bounds from const, 8",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, ~0),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 0, 1),
-		BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-			    offsetof(struct __sk_buff, mark)),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "negative offset ctx ptr R1 off=-1 disallowed",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"check deducing bounds from const, 9",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JSGE, BPF_REG_0, 0, 0),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-		BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.errstr = "R0 tried to subtract pointer from scalar",
-	.result = REJECT,
-},
-{
-	"check deducing bounds from const, 10",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JSLE, BPF_REG_0, 0, 0),
-		/* Marks reg as unknown. */
-		BPF_ALU64_IMM(BPF_NEG, BPF_REG_0, 0),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-		BPF_EXIT_INSN(),
-	},
-	.errstr = "math between ctx pointer and register with unbounded min value is not allowed",
-	.result = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From b14a702afd0d2da746294ed6070668b839a77793 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:51 +0200
Subject: selftests/bpf: verifier/bounds_mix_sign_unsign.c converted to inline
 assembly

Test verifier/bounds_mix_sign_unsign.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-11-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_bounds_mix_sign_unsign.c    | 554 +++++++++++++++++++++
 .../bpf/verifier/bounds_mix_sign_unsign.c          | 411 ---------------
 3 files changed, 556 insertions(+), 411 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/bounds_mix_sign_unsign.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index a8cfef92ed64..bbc39412fcd1 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -7,6 +7,7 @@
 #include "verifier_array_access.skel.h"
 #include "verifier_basic_stack.skel.h"
 #include "verifier_bounds_deduction.skel.h"
+#include "verifier_bounds_mix_sign_unsign.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -36,3 +37,4 @@ void test_verifier_and(void)                  { RUN(verifier_and); }
 void test_verifier_array_access(void)         { RUN(verifier_array_access); }
 void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
+void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c b/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c
new file mode 100644
index 000000000000..91a66357896a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c
@@ -0,0 +1,554 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/bounds_mix_sign_unsign.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, positive bounds")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_positive_bounds(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = 2;						\
+	if r2 >= r1 goto l0_%=;				\
+	if r1 s> 4 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void checks_mixing_signed_and_unsigned(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	if r1 > r2 goto l0_%=;				\
+	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 2")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_2(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	if r1 > r2 goto l0_%=;				\
+	r8 = 0;						\
+	r8 += r1;					\
+	if r8 s> 1 goto l0_%=;				\
+	r0 += r8;					\
+	r0 = 0;						\
+	*(u8*)(r8 + 0) = r0;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 3")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_3(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	if r1 > r2 goto l0_%=;				\
+	r8 = r1;					\
+	if r8 s> 1 goto l0_%=;				\
+	r0 += r8;					\
+	r0 = 0;						\
+	*(u8*)(r8 + 0) = r0;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 4")
+__success __success_unpriv __retval(0)
+__naked void signed_and_unsigned_variant_4(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = 1;						\
+	r1 &= r2;					\
+	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 5")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_5(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	if r1 > r2 goto l0_%=;				\
+	if r1 s> 1 goto l0_%=;				\
+	r0 += 4;					\
+	r0 -= r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 6")
+__failure __msg("R4 min value is negative, either use unsigned")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_6(void)
+{
+	asm volatile ("					\
+	r9 = r1;					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = r9;					\
+	r2 = 0;						\
+	r3 = r10;					\
+	r3 += -512;					\
+	r4 = *(u64*)(r10 - 16);				\
+	r6 = -1;					\
+	if r4 > r6 goto l0_%=;				\
+	if r4 s> 1 goto l0_%=;				\
+	r4 += 1;					\
+	r5 = 0;						\
+	r6 = 0;						\
+	*(u16*)(r10 - 512) = r6;			\
+	call %[bpf_skb_load_bytes];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 7")
+__success __success_unpriv __retval(0)
+__naked void signed_and_unsigned_variant_7(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = %[__imm_0];				\
+	if r1 > r2 goto l0_%=;				\
+	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__imm_0, 1024 * 1024 * 1024)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 8")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_8(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	if r2 > r1 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 9")
+__success __success_unpriv __retval(0)
+__naked void signed_and_unsigned_variant_9(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -9223372036854775808ULL ll;		\
+	if r2 > r1 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 10")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_10(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = 0;						\
+	if r2 > r1 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 11")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_11(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	if r2 >= r1 goto l1_%=;				\
+	/* Dead branch. */				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 12")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_12(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -6;					\
+	if r2 >= r1 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	if r1 s> 1 goto l0_%=;				\
+	r0 += r1;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 13")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_13(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = 2;						\
+	if r2 >= r1 goto l0_%=;				\
+	r7 = 1;						\
+	if r7 s> 0 goto l1_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+l1_%=:	r7 += r1;					\
+	if r7 s> 4 goto l2_%=;				\
+	r0 += r7;					\
+	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l2_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 14")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_14(void)
+{
+	asm volatile ("					\
+	r9 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -1;					\
+	r8 = 2;						\
+	if r9 == 42 goto l1_%=;				\
+	if r8 s> r1 goto l2_%=;				\
+l3_%=:	if r1 s> 1 goto l2_%=;				\
+	r0 += r1;					\
+l0_%=:	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+l2_%=:	r0 = 0;						\
+	exit;						\
+l1_%=:	if r1 > r2 goto l2_%=;				\
+	goto l3_%=;					\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks mixing signed and unsigned, variant 15")
+__failure __msg("unbounded min value")
+__failure_unpriv
+__naked void signed_and_unsigned_variant_15(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r10 - 16);				\
+	r2 = -6;					\
+	if r2 >= r1 goto l1_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 += r1;					\
+	if r0 > 1 goto l2_%=;				\
+	r0 = 0;						\
+	exit;						\
+l2_%=:	r1 = 0;						\
+	*(u8*)(r0 + 0) = r1;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/bounds_mix_sign_unsign.c b/tools/testing/selftests/bpf/verifier/bounds_mix_sign_unsign.c
deleted file mode 100644
index bf82b923c5fe..000000000000
--- a/tools/testing/selftests/bpf/verifier/bounds_mix_sign_unsign.c
+++ /dev/null
@@ -1,411 +0,0 @@
-{
-	"bounds checks mixing signed and unsigned, positive bounds",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, 2),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_2, BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 4, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, 3),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 2",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, 5),
-	BPF_MOV64_IMM(BPF_REG_8, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_8, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_8, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_8),
-	BPF_ST_MEM(BPF_B, BPF_REG_8, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 3",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, 4),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_8, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_8),
-	BPF_ST_MEM(BPF_B, BPF_REG_8, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 4",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, 1),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.result = ACCEPT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 5",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, 5),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 4),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 6",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_1),
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_9),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, -512),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_6, -1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_6, 5),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_4, 1, 4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 1),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_ST_MEM(BPF_H, BPF_REG_10, -512, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R4 min value is negative, either use unsigned",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 7",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, 1024 * 1024 * 1024),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, 3),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.result = ACCEPT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 8",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_1, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 9",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_LD_IMM64(BPF_REG_2, -9223372036854775808ULL),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_1, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.result = ACCEPT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 10",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_1, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 11",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_2, BPF_REG_1, 2),
-	/* Dead branch. */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 12",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -6),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_2, BPF_REG_1, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 13",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, 2),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_2, BPF_REG_1, 2),
-	BPF_MOV64_IMM(BPF_REG_7, 1),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_7, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_7, 4, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_7),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 14",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_9, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_MOV64_IMM(BPF_REG_8, 2),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_9, 42, 6),
-	BPF_JMP_REG(BPF_JSGT, BPF_REG_8, BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, -3),
-	BPF_JMP_IMM(BPF_JA, 0, 0, -7),
-	},
-	.fixup_map_hash_8b = { 6 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-{
-	"bounds checks mixing signed and unsigned, variant 15",
-	.insns = {
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_MOV64_IMM(BPF_REG_2, -6),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_2, BPF_REG_1, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_0, 1, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ST_MEM(BPF_B, BPF_REG_0, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "unbounded min value",
-	.result = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From 2f2047c22cda4fbbe6bb889cc6c5450cd90688f8 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:52 +0200
Subject: selftests/bpf: verifier/cfg.c converted to inline assembly

Test verifier/cfg.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-12-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   2 +
 tools/testing/selftests/bpf/progs/verifier_cfg.c  | 100 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/cfg.c        |  73 ----------------
 3 files changed, 102 insertions(+), 73 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_cfg.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/cfg.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index bbc39412fcd1..46182abecabb 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -8,6 +8,7 @@
 #include "verifier_basic_stack.skel.h"
 #include "verifier_bounds_deduction.skel.h"
 #include "verifier_bounds_mix_sign_unsign.skel.h"
+#include "verifier_cfg.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -38,3 +39,4 @@ void test_verifier_array_access(void)         { RUN(verifier_array_access); }
 void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
+void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_cfg.c b/tools/testing/selftests/bpf/progs/verifier_cfg.c
new file mode 100644
index 000000000000..df7697b94007
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_cfg.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/cfg.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("unreachable")
+__failure __msg("unreachable")
+__failure_unpriv
+__naked void unreachable(void)
+{
+	asm volatile ("					\
+	exit;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unreachable2")
+__failure __msg("unreachable")
+__failure_unpriv
+__naked void unreachable2(void)
+{
+	asm volatile ("					\
+	goto l0_%=;					\
+	goto l0_%=;					\
+l0_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("out of range jump")
+__failure __msg("jump out of range")
+__failure_unpriv
+__naked void out_of_range_jump(void)
+{
+	asm volatile ("					\
+	goto l0_%=;					\
+	exit;						\
+l0_%=:							\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("out of range jump2")
+__failure __msg("jump out of range")
+__failure_unpriv
+__naked void out_of_range_jump2(void)
+{
+	asm volatile ("					\
+	goto -2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("loop (back-edge)")
+__failure __msg("unreachable insn 1")
+__msg_unpriv("back-edge")
+__naked void loop_back_edge(void)
+{
+	asm volatile ("					\
+l0_%=:	goto l0_%=;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("loop2 (back-edge)")
+__failure __msg("unreachable insn 4")
+__msg_unpriv("back-edge")
+__naked void loop2_back_edge(void)
+{
+	asm volatile ("					\
+l0_%=:	r1 = r0;					\
+	r2 = r0;					\
+	r3 = r0;					\
+	goto l0_%=;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("conditional loop")
+__failure __msg("infinite loop detected")
+__msg_unpriv("back-edge")
+__naked void conditional_loop(void)
+{
+	asm volatile ("					\
+	r0 = r1;					\
+l0_%=:	r2 = r0;					\
+	r3 = r0;					\
+	if r1 == 0 goto l0_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/cfg.c b/tools/testing/selftests/bpf/verifier/cfg.c
deleted file mode 100644
index 4eb76ed739ce..000000000000
--- a/tools/testing/selftests/bpf/verifier/cfg.c
+++ /dev/null
@@ -1,73 +0,0 @@
-{
-	"unreachable",
-	.insns = {
-	BPF_EXIT_INSN(),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "unreachable",
-	.result = REJECT,
-},
-{
-	"unreachable2",
-	.insns = {
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "unreachable",
-	.result = REJECT,
-},
-{
-	"out of range jump",
-	.insns = {
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "jump out of range",
-	.result = REJECT,
-},
-{
-	"out of range jump2",
-	.insns = {
-	BPF_JMP_IMM(BPF_JA, 0, 0, -2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "jump out of range",
-	.result = REJECT,
-},
-{
-	"loop (back-edge)",
-	.insns = {
-	BPF_JMP_IMM(BPF_JA, 0, 0, -1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "unreachable insn 1",
-	.errstr_unpriv = "back-edge",
-	.result = REJECT,
-},
-{
-	"loop2 (back-edge)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, -4),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "unreachable insn 4",
-	.errstr_unpriv = "back-edge",
-	.result = REJECT,
-},
-{
-	"conditional loop",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, -3),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "infinite loop detected",
-	.errstr_unpriv = "back-edge",
-	.result = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From 047687a7f494d45198f112b51e72228aa054732c Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:53 +0200
Subject: selftests/bpf: verifier/cgroup_inv_retcode.c converted to inline
 assembly

Test verifier/cgroup_inv_retcode.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-13-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../bpf/progs/verifier_cgroup_inv_retcode.c        | 89 ++++++++++++++++++++++
 .../selftests/bpf/verifier/cgroup_inv_retcode.c    | 72 -----------------
 3 files changed, 91 insertions(+), 72 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_cgroup_inv_retcode.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/cgroup_inv_retcode.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 46182abecabb..b138c9894abb 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -9,6 +9,7 @@
 #include "verifier_bounds_deduction.skel.h"
 #include "verifier_bounds_mix_sign_unsign.skel.h"
 #include "verifier_cfg.skel.h"
+#include "verifier_cgroup_inv_retcode.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -40,3 +41,4 @@ void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
 void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
+void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_cgroup_inv_retcode.c b/tools/testing/selftests/bpf/progs/verifier_cgroup_inv_retcode.c
new file mode 100644
index 000000000000..d6c4a7f3f790
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_cgroup_inv_retcode.c
@@ -0,0 +1,89 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/cgroup_inv_retcode.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test1")
+__failure __msg("R0 has value (0x0; 0xffffffff)")
+__naked void with_invalid_return_code_test1(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test2")
+__success
+__naked void with_invalid_return_code_test2(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + 0);				\
+	r0 &= 1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test3")
+__failure __msg("R0 has value (0x0; 0x3)")
+__naked void with_invalid_return_code_test3(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + 0);				\
+	r0 &= 3;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test4")
+__success
+__naked void with_invalid_return_code_test4(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test5")
+__failure __msg("R0 has value (0x2; 0x0)")
+__naked void with_invalid_return_code_test5(void)
+{
+	asm volatile ("					\
+	r0 = 2;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test6")
+__failure __msg("R0 is not a known value (ctx)")
+__naked void with_invalid_return_code_test6(void)
+{
+	asm volatile ("					\
+	r0 = r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/sock")
+__description("bpf_exit with invalid return code. test7")
+__failure __msg("R0 has unknown scalar value")
+__naked void with_invalid_return_code_test7(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + 0);				\
+	r2 = *(u32*)(r1 + 4);				\
+	r0 *= r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/cgroup_inv_retcode.c b/tools/testing/selftests/bpf/verifier/cgroup_inv_retcode.c
deleted file mode 100644
index 6d65fe3e7321..000000000000
--- a/tools/testing/selftests/bpf/verifier/cgroup_inv_retcode.c
+++ /dev/null
@@ -1,72 +0,0 @@
-{
-	"bpf_exit with invalid return code. test1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 has value (0x0; 0xffffffff)",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-{
-	"bpf_exit with invalid return code. test2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-{
-	"bpf_exit with invalid return code. test3",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 3),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 has value (0x0; 0x3)",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-{
-	"bpf_exit with invalid return code. test4",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-{
-	"bpf_exit with invalid return code. test5",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 has value (0x2; 0x0)",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-{
-	"bpf_exit with invalid return code. test6",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 is not a known value (ctx)",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-{
-	"bpf_exit with invalid return code. test7",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_MUL, BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 has unknown scalar value",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-},
-- 
cgit v1.2.3-70-g09d2


From b1b6372535c0cc0cce4870b07a0938309f3a5d37 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:54 +0200
Subject: selftests/bpf: verifier/cgroup_skb.c converted to inline assembly

Test verifier/cgroup_skb.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-14-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_cgroup_skb.c      | 227 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/cgroup_skb.c  | 197 ------------------
 3 files changed, 229 insertions(+), 197 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_cgroup_skb.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/cgroup_skb.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index b138c9894abb..53e41af90821 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -10,6 +10,7 @@
 #include "verifier_bounds_mix_sign_unsign.skel.h"
 #include "verifier_cfg.skel.h"
 #include "verifier_cgroup_inv_retcode.skel.h"
+#include "verifier_cgroup_skb.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -42,3 +43,4 @@ void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction);
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
 void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
 void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
+void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_cgroup_skb.c b/tools/testing/selftests/bpf/progs/verifier_cgroup_skb.c
new file mode 100644
index 000000000000..5ee3d349d6d0
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_cgroup_skb.c
@@ -0,0 +1,227 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/cgroup_skb.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("cgroup/skb")
+__description("direct packet read test#1 for CGROUP_SKB")
+__success __failure_unpriv
+__msg_unpriv("invalid bpf_context access off=76 size=4")
+__retval(0)
+__naked void test_1_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r4 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r5 = *(u32*)(r1 + %[__sk_buff_pkt_type]);	\
+	r6 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	*(u32*)(r1 + %[__sk_buff_mark]) = r6;		\
+	r7 = *(u32*)(r1 + %[__sk_buff_queue_mapping]);	\
+	r8 = *(u32*)(r1 + %[__sk_buff_protocol]);	\
+	r9 = *(u32*)(r1 + %[__sk_buff_vlan_present]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(__sk_buff_pkt_type, offsetof(struct __sk_buff, pkt_type)),
+	  __imm_const(__sk_buff_protocol, offsetof(struct __sk_buff, protocol)),
+	  __imm_const(__sk_buff_queue_mapping, offsetof(struct __sk_buff, queue_mapping)),
+	  __imm_const(__sk_buff_vlan_present, offsetof(struct __sk_buff, vlan_present))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("direct packet read test#2 for CGROUP_SKB")
+__success __success_unpriv __retval(0)
+__naked void test_2_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r4 = *(u32*)(r1 + %[__sk_buff_vlan_tci]);	\
+	r5 = *(u32*)(r1 + %[__sk_buff_vlan_proto]);	\
+	r6 = *(u32*)(r1 + %[__sk_buff_priority]);	\
+	*(u32*)(r1 + %[__sk_buff_priority]) = r6;	\
+	r7 = *(u32*)(r1 + %[__sk_buff_ingress_ifindex]);\
+	r8 = *(u32*)(r1 + %[__sk_buff_tc_index]);	\
+	r9 = *(u32*)(r1 + %[__sk_buff_hash]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_hash, offsetof(struct __sk_buff, hash)),
+	  __imm_const(__sk_buff_ingress_ifindex, offsetof(struct __sk_buff, ingress_ifindex)),
+	  __imm_const(__sk_buff_priority, offsetof(struct __sk_buff, priority)),
+	  __imm_const(__sk_buff_tc_index, offsetof(struct __sk_buff, tc_index)),
+	  __imm_const(__sk_buff_vlan_proto, offsetof(struct __sk_buff, vlan_proto)),
+	  __imm_const(__sk_buff_vlan_tci, offsetof(struct __sk_buff, vlan_tci))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("direct packet read test#3 for CGROUP_SKB")
+__success __success_unpriv __retval(0)
+__naked void test_3_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r4 = *(u32*)(r1 + %[__sk_buff_cb_0]);		\
+	r5 = *(u32*)(r1 + %[__sk_buff_cb_1]);		\
+	r6 = *(u32*)(r1 + %[__sk_buff_cb_2]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_cb_3]);		\
+	r8 = *(u32*)(r1 + %[__sk_buff_cb_4]);		\
+	r9 = *(u32*)(r1 + %[__sk_buff_napi_id]);	\
+	*(u32*)(r1 + %[__sk_buff_cb_0]) = r4;		\
+	*(u32*)(r1 + %[__sk_buff_cb_1]) = r5;		\
+	*(u32*)(r1 + %[__sk_buff_cb_2]) = r6;		\
+	*(u32*)(r1 + %[__sk_buff_cb_3]) = r7;		\
+	*(u32*)(r1 + %[__sk_buff_cb_4]) = r8;		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0])),
+	  __imm_const(__sk_buff_cb_1, offsetof(struct __sk_buff, cb[1])),
+	  __imm_const(__sk_buff_cb_2, offsetof(struct __sk_buff, cb[2])),
+	  __imm_const(__sk_buff_cb_3, offsetof(struct __sk_buff, cb[3])),
+	  __imm_const(__sk_buff_cb_4, offsetof(struct __sk_buff, cb[4])),
+	  __imm_const(__sk_buff_napi_id, offsetof(struct __sk_buff, napi_id))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("direct packet read test#4 for CGROUP_SKB")
+__success __success_unpriv __retval(0)
+__naked void test_4_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_family]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_remote_ip4]);	\
+	r4 = *(u32*)(r1 + %[__sk_buff_local_ip4]);	\
+	r5 = *(u32*)(r1 + %[__sk_buff_remote_ip6_0]);	\
+	r5 = *(u32*)(r1 + %[__sk_buff_remote_ip6_1]);	\
+	r5 = *(u32*)(r1 + %[__sk_buff_remote_ip6_2]);	\
+	r5 = *(u32*)(r1 + %[__sk_buff_remote_ip6_3]);	\
+	r6 = *(u32*)(r1 + %[__sk_buff_local_ip6_0]);	\
+	r6 = *(u32*)(r1 + %[__sk_buff_local_ip6_1]);	\
+	r6 = *(u32*)(r1 + %[__sk_buff_local_ip6_2]);	\
+	r6 = *(u32*)(r1 + %[__sk_buff_local_ip6_3]);	\
+	r7 = *(u32*)(r1 + %[__sk_buff_remote_port]);	\
+	r8 = *(u32*)(r1 + %[__sk_buff_local_port]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_family, offsetof(struct __sk_buff, family)),
+	  __imm_const(__sk_buff_local_ip4, offsetof(struct __sk_buff, local_ip4)),
+	  __imm_const(__sk_buff_local_ip6_0, offsetof(struct __sk_buff, local_ip6[0])),
+	  __imm_const(__sk_buff_local_ip6_1, offsetof(struct __sk_buff, local_ip6[1])),
+	  __imm_const(__sk_buff_local_ip6_2, offsetof(struct __sk_buff, local_ip6[2])),
+	  __imm_const(__sk_buff_local_ip6_3, offsetof(struct __sk_buff, local_ip6[3])),
+	  __imm_const(__sk_buff_local_port, offsetof(struct __sk_buff, local_port)),
+	  __imm_const(__sk_buff_remote_ip4, offsetof(struct __sk_buff, remote_ip4)),
+	  __imm_const(__sk_buff_remote_ip6_0, offsetof(struct __sk_buff, remote_ip6[0])),
+	  __imm_const(__sk_buff_remote_ip6_1, offsetof(struct __sk_buff, remote_ip6[1])),
+	  __imm_const(__sk_buff_remote_ip6_2, offsetof(struct __sk_buff, remote_ip6[2])),
+	  __imm_const(__sk_buff_remote_ip6_3, offsetof(struct __sk_buff, remote_ip6[3])),
+	  __imm_const(__sk_buff_remote_port, offsetof(struct __sk_buff, remote_port))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid access of tc_classid for CGROUP_SKB")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void tc_classid_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_tc_classid]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_tc_classid, offsetof(struct __sk_buff, tc_classid))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid access of data_meta for CGROUP_SKB")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void data_meta_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_data_meta]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data_meta, offsetof(struct __sk_buff, data_meta))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid access of flow_keys for CGROUP_SKB")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void flow_keys_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_flow_keys]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_flow_keys, offsetof(struct __sk_buff, flow_keys))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid write access to napi_id for CGROUP_SKB")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void napi_id_for_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r9 = *(u32*)(r1 + %[__sk_buff_napi_id]);	\
+	*(u32*)(r1 + %[__sk_buff_napi_id]) = r9;	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_napi_id, offsetof(struct __sk_buff, napi_id))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("write tstamp from CGROUP_SKB")
+__success __failure_unpriv
+__msg_unpriv("invalid bpf_context access off=152 size=8")
+__retval(0)
+__naked void write_tstamp_from_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r1 + %[__sk_buff_tstamp]) = r0;		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_tstamp, offsetof(struct __sk_buff, tstamp))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("read tstamp from CGROUP_SKB")
+__success __success_unpriv __retval(0)
+__naked void read_tstamp_from_cgroup_skb(void)
+{
+	asm volatile ("					\
+	r0 = *(u64*)(r1 + %[__sk_buff_tstamp]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_tstamp, offsetof(struct __sk_buff, tstamp))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/cgroup_skb.c b/tools/testing/selftests/bpf/verifier/cgroup_skb.c
deleted file mode 100644
index 52e4c03b076b..000000000000
--- a/tools/testing/selftests/bpf/verifier/cgroup_skb.c
+++ /dev/null
@@ -1,197 +0,0 @@
-{
-	"direct packet read test#1 for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, pkt_type)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_6,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, queue_mapping)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, protocol)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_9, BPF_REG_1,
-		    offsetof(struct __sk_buff, vlan_present)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "invalid bpf_context access off=76 size=4",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"direct packet read test#2 for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct __sk_buff, vlan_tci)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, vlan_proto)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, priority)),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_6,
-		    offsetof(struct __sk_buff, priority)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, ingress_ifindex)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_index)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_9, BPF_REG_1,
-		    offsetof(struct __sk_buff, hash)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"direct packet read test#3 for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[1])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[2])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[3])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[4])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_9, BPF_REG_1,
-		    offsetof(struct __sk_buff, napi_id)),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_4,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_5,
-		    offsetof(struct __sk_buff, cb[1])),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_6,
-		    offsetof(struct __sk_buff, cb[2])),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_7,
-		    offsetof(struct __sk_buff, cb[3])),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_8,
-		    offsetof(struct __sk_buff, cb[4])),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"direct packet read test#4 for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, family)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, remote_ip4)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct __sk_buff, local_ip4)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, remote_ip6[0])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, remote_ip6[1])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, remote_ip6[2])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_1,
-		    offsetof(struct __sk_buff, remote_ip6[3])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, local_ip6[0])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, local_ip6[1])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, local_ip6[2])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, local_ip6[3])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, remote_port)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, local_port)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid access of tc_classid for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_classid)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid access of data_meta for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_meta)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid access of flow_keys for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, flow_keys)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid write access to napi_id for CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_9, BPF_REG_1,
-		    offsetof(struct __sk_buff, napi_id)),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_9,
-		    offsetof(struct __sk_buff, napi_id)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"write tstamp from CGROUP_SKB",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, tstamp)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "invalid bpf_context access off=152 size=8",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"read tstamp from CGROUP_SKB",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tstamp)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-- 
cgit v1.2.3-70-g09d2


From 8f16f3c07e460f81ff6f4d673c7edd413db19ffe Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:55 +0200
Subject: selftests/bpf: verifier/cgroup_storage.c converted to inline assembly

Test verifier/cgroup_storage.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-15-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_cgroup_storage.c  | 308 +++++++++++++++++++++
 .../selftests/bpf/verifier/cgroup_storage.c        | 220 ---------------
 3 files changed, 310 insertions(+), 220 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_cgroup_storage.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/cgroup_storage.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 53e41af90821..3b47620a1f42 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -11,6 +11,7 @@
 #include "verifier_cfg.skel.h"
 #include "verifier_cgroup_inv_retcode.skel.h"
 #include "verifier_cgroup_skb.skel.h"
+#include "verifier_cgroup_storage.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -44,3 +45,4 @@ void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_u
 void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
 void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
 void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
+void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_cgroup_storage.c b/tools/testing/selftests/bpf/progs/verifier_cgroup_storage.c
new file mode 100644
index 000000000000..9a13f5c11ac7
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_cgroup_storage.c
@@ -0,0 +1,308 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/cgroup_storage.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "../../../include/linux/filter.h"
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_CGROUP_STORAGE);
+	__uint(max_entries, 0);
+	__type(key, struct bpf_cgroup_storage_key);
+	__type(value, char[TEST_DATA_LEN]);
+} cgroup_storage SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE);
+	__uint(max_entries, 0);
+	__type(key, struct bpf_cgroup_storage_key);
+	__type(value, char[64]);
+} percpu_cgroup_storage SEC(".maps");
+
+SEC("cgroup/skb")
+__description("valid cgroup storage access")
+__success __success_unpriv __retval(0)
+__naked void valid_cgroup_storage_access(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[cgroup_storage] ll;			\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid cgroup storage access 1")
+__failure __msg("cannot pass map_type 1 into func bpf_get_local_storage")
+__failure_unpriv
+__naked void invalid_cgroup_storage_access_1(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid cgroup storage access 2")
+__failure __msg("fd 1 is not pointing to valid bpf_map")
+__failure_unpriv
+__naked void invalid_cgroup_storage_access_2(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	.8byte %[ld_map_fd];				\
+	.8byte 0;					\
+	call %[bpf_get_local_storage];			\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_insn(ld_map_fd, BPF_RAW_INSN(BPF_LD | BPF_DW | BPF_IMM, BPF_REG_1, BPF_PSEUDO_MAP_FD, 0, 1))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid cgroup storage access 3")
+__failure __msg("invalid access to map value, value_size=64 off=256 size=4")
+__failure_unpriv
+__naked void invalid_cgroup_storage_access_3(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[cgroup_storage] ll;			\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 256);				\
+	r1 += 1;					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid cgroup storage access 4")
+__failure __msg("invalid access to map value, value_size=64 off=-2 size=4")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void invalid_cgroup_storage_access_4(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[cgroup_storage] ll;			\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 - 2);				\
+	r0 = r1;					\
+	r1 += 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid cgroup storage access 5")
+__failure __msg("get_local_storage() doesn't support non-zero flags")
+__failure_unpriv
+__naked void invalid_cgroup_storage_access_5(void)
+{
+	asm volatile ("					\
+	r2 = 7;						\
+	r1 = %[cgroup_storage] ll;			\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid cgroup storage access 6")
+__failure __msg("get_local_storage() doesn't support non-zero flags")
+__msg_unpriv("R2 leaks addr into helper function")
+__naked void invalid_cgroup_storage_access_6(void)
+{
+	asm volatile ("					\
+	r2 = r1;					\
+	r1 = %[cgroup_storage] ll;			\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("valid per-cpu cgroup storage access")
+__success __success_unpriv __retval(0)
+__naked void per_cpu_cgroup_storage_access(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[percpu_cgroup_storage] ll;		\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(percpu_cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid per-cpu cgroup storage access 1")
+__failure __msg("cannot pass map_type 1 into func bpf_get_local_storage")
+__failure_unpriv
+__naked void cpu_cgroup_storage_access_1(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid per-cpu cgroup storage access 2")
+__failure __msg("fd 1 is not pointing to valid bpf_map")
+__failure_unpriv
+__naked void cpu_cgroup_storage_access_2(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	.8byte %[ld_map_fd];				\
+	.8byte 0;					\
+	call %[bpf_get_local_storage];			\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_insn(ld_map_fd, BPF_RAW_INSN(BPF_LD | BPF_DW | BPF_IMM, BPF_REG_1, BPF_PSEUDO_MAP_FD, 0, 1))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid per-cpu cgroup storage access 3")
+__failure __msg("invalid access to map value, value_size=64 off=256 size=4")
+__failure_unpriv
+__naked void cpu_cgroup_storage_access_3(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[percpu_cgroup_storage] ll;		\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 256);				\
+	r1 += 1;					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(percpu_cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid per-cpu cgroup storage access 4")
+__failure __msg("invalid access to map value, value_size=64 off=-2 size=4")
+__failure_unpriv
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void cpu_cgroup_storage_access_4(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	r1 = %[cgroup_storage] ll;			\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 - 2);				\
+	r0 = r1;					\
+	r1 += 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid per-cpu cgroup storage access 5")
+__failure __msg("get_local_storage() doesn't support non-zero flags")
+__failure_unpriv
+__naked void cpu_cgroup_storage_access_5(void)
+{
+	asm volatile ("					\
+	r2 = 7;						\
+	r1 = %[percpu_cgroup_storage] ll;		\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(percpu_cgroup_storage)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("invalid per-cpu cgroup storage access 6")
+__failure __msg("get_local_storage() doesn't support non-zero flags")
+__msg_unpriv("R2 leaks addr into helper function")
+__naked void cpu_cgroup_storage_access_6(void)
+{
+	asm volatile ("					\
+	r2 = r1;					\
+	r1 = %[percpu_cgroup_storage] ll;		\
+	call %[bpf_get_local_storage];			\
+	r1 = *(u32*)(r0 + 0);				\
+	r0 = r1;					\
+	r0 &= 1;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_local_storage),
+	  __imm_addr(percpu_cgroup_storage)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/cgroup_storage.c b/tools/testing/selftests/bpf/verifier/cgroup_storage.c
deleted file mode 100644
index 97057c0a1b8a..000000000000
--- a/tools/testing/selftests/bpf/verifier/cgroup_storage.c
+++ /dev/null
@@ -1,220 +0,0 @@
-{
-	"valid cgroup storage access",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_cgroup_storage = { 1 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid cgroup storage access 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	.result = REJECT,
-	.errstr = "cannot pass map_type 1 into func bpf_get_local_storage",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid cgroup storage access 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "fd 1 is not pointing to valid bpf_map",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid cgroup storage access 3",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 256),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=64 off=256 size=4",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid cgroup storage access 4",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, -2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=64 off=-2 size=4",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid cgroup storage access 5",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 7),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "get_local_storage() doesn't support non-zero flags",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid cgroup storage access 6",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_1),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "get_local_storage() doesn't support non-zero flags",
-	.errstr_unpriv = "R2 leaks addr into helper function",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"valid per-cpu cgroup storage access",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_percpu_cgroup_storage = { 1 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid per-cpu cgroup storage access 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	.result = REJECT,
-	.errstr = "cannot pass map_type 1 into func bpf_get_local_storage",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid per-cpu cgroup storage access 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "fd 1 is not pointing to valid bpf_map",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid per-cpu cgroup storage access 3",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 256),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_percpu_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=64 off=256 size=4",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid per-cpu cgroup storage access 4",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, -2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=64 off=-2 size=4",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid per-cpu cgroup storage access 5",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 7),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_percpu_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "get_local_storage() doesn't support non-zero flags",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"invalid per-cpu cgroup storage access 6",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_1),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_percpu_cgroup_storage = { 1 },
-	.result = REJECT,
-	.errstr = "get_local_storage() doesn't support non-zero flags",
-	.errstr_unpriv = "R2 leaks addr into helper function",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-- 
cgit v1.2.3-70-g09d2


From a2777eaad5d9b7b06917d5bd3e786a1733e9dc3c Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:56 +0200
Subject: selftests/bpf: verifier/const_or.c converted to inline assembly

Test verifier/const_or.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-16-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../selftests/bpf/progs/verifier_const_or.c        | 82 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/const_or.c    | 60 ----------------
 3 files changed, 84 insertions(+), 60 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_const_or.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/const_or.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 3b47620a1f42..36fdede7dcab 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -12,6 +12,7 @@
 #include "verifier_cgroup_inv_retcode.skel.h"
 #include "verifier_cgroup_skb.skel.h"
 #include "verifier_cgroup_storage.skel.h"
+#include "verifier_const_or.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -46,3 +47,4 @@ void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
 void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
 void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
 void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
+void test_verifier_const_or(void)             { RUN(verifier_const_or); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_const_or.c b/tools/testing/selftests/bpf/progs/verifier_const_or.c
new file mode 100644
index 000000000000..ba8922b2eebd
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_const_or.c
@@ -0,0 +1,82 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/const_or.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("tracepoint")
+__description("constant register |= constant should keep constant type")
+__success
+__naked void constant_should_keep_constant_type(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -48;					\
+	r2 = 34;					\
+	r2 |= 13;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("constant register |= constant should not bypass stack boundary checks")
+__failure __msg("invalid indirect access to stack R1 off=-48 size=58")
+__naked void not_bypass_stack_boundary_checks_1(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -48;					\
+	r2 = 34;					\
+	r2 |= 24;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("constant register |= constant register should keep constant type")
+__success
+__naked void register_should_keep_constant_type(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -48;					\
+	r2 = 34;					\
+	r4 = 13;					\
+	r2 |= r4;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("constant register |= constant register should not bypass stack boundary checks")
+__failure __msg("invalid indirect access to stack R1 off=-48 size=58")
+__naked void not_bypass_stack_boundary_checks_2(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -48;					\
+	r2 = 34;					\
+	r4 = 24;					\
+	r2 |= r4;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/const_or.c b/tools/testing/selftests/bpf/verifier/const_or.c
deleted file mode 100644
index 0719b0ddec04..000000000000
--- a/tools/testing/selftests/bpf/verifier/const_or.c
+++ /dev/null
@@ -1,60 +0,0 @@
-{
-	"constant register |= constant should keep constant type",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -48),
-	BPF_MOV64_IMM(BPF_REG_2, 34),
-	BPF_ALU64_IMM(BPF_OR, BPF_REG_2, 13),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"constant register |= constant should not bypass stack boundary checks",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -48),
-	BPF_MOV64_IMM(BPF_REG_2, 34),
-	BPF_ALU64_IMM(BPF_OR, BPF_REG_2, 24),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid indirect access to stack R1 off=-48 size=58",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"constant register |= constant register should keep constant type",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -48),
-	BPF_MOV64_IMM(BPF_REG_2, 34),
-	BPF_MOV64_IMM(BPF_REG_4, 13),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"constant register |= constant register should not bypass stack boundary checks",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -48),
-	BPF_MOV64_IMM(BPF_REG_2, 34),
-	BPF_MOV64_IMM(BPF_REG_4, 24),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid indirect access to stack R1 off=-48 size=58",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-- 
cgit v1.2.3-70-g09d2


From a58475a98903c756a7f731cffdf20242ed17b9b0 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:57 +0200
Subject: selftests/bpf: verifier/ctx_sk_msg.c converted to inline assembly

Test verifier/ctx_sk_msg.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-17-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_ctx_sk_msg.c      | 228 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/ctx_sk_msg.c  | 181 ----------------
 3 files changed, 230 insertions(+), 181 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_ctx_sk_msg.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/ctx_sk_msg.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 36fdede7dcab..29351c774ee2 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -13,6 +13,7 @@
 #include "verifier_cgroup_skb.skel.h"
 #include "verifier_cgroup_storage.skel.h"
 #include "verifier_const_or.skel.h"
+#include "verifier_ctx_sk_msg.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -48,3 +49,4 @@ void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode)
 void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
 void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
 void test_verifier_const_or(void)             { RUN(verifier_const_or); }
+void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_ctx_sk_msg.c b/tools/testing/selftests/bpf/progs/verifier_ctx_sk_msg.c
new file mode 100644
index 000000000000..65edc89799f9
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_ctx_sk_msg.c
@@ -0,0 +1,228 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/ctx_sk_msg.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("sk_msg")
+__description("valid access family in SK_MSG")
+__success
+__naked void access_family_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_family]);		\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_family, offsetof(struct sk_msg_md, family))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("valid access remote_ip4 in SK_MSG")
+__success
+__naked void remote_ip4_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_remote_ip4]);	\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_remote_ip4, offsetof(struct sk_msg_md, remote_ip4))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("valid access local_ip4 in SK_MSG")
+__success
+__naked void local_ip4_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_local_ip4]);	\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_local_ip4, offsetof(struct sk_msg_md, local_ip4))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("valid access remote_port in SK_MSG")
+__success
+__naked void remote_port_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_remote_port]);	\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_remote_port, offsetof(struct sk_msg_md, remote_port))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("valid access local_port in SK_MSG")
+__success
+__naked void local_port_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_local_port]);	\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_local_port, offsetof(struct sk_msg_md, local_port))
+	: __clobber_all);
+}
+
+SEC("sk_skb")
+__description("valid access remote_ip6 in SK_MSG")
+__success
+__naked void remote_ip6_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_remote_ip6_0]);	\
+	r0 = *(u32*)(r1 + %[sk_msg_md_remote_ip6_1]);	\
+	r0 = *(u32*)(r1 + %[sk_msg_md_remote_ip6_2]);	\
+	r0 = *(u32*)(r1 + %[sk_msg_md_remote_ip6_3]);	\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_remote_ip6_0, offsetof(struct sk_msg_md, remote_ip6[0])),
+	  __imm_const(sk_msg_md_remote_ip6_1, offsetof(struct sk_msg_md, remote_ip6[1])),
+	  __imm_const(sk_msg_md_remote_ip6_2, offsetof(struct sk_msg_md, remote_ip6[2])),
+	  __imm_const(sk_msg_md_remote_ip6_3, offsetof(struct sk_msg_md, remote_ip6[3]))
+	: __clobber_all);
+}
+
+SEC("sk_skb")
+__description("valid access local_ip6 in SK_MSG")
+__success
+__naked void local_ip6_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_local_ip6_0]);	\
+	r0 = *(u32*)(r1 + %[sk_msg_md_local_ip6_1]);	\
+	r0 = *(u32*)(r1 + %[sk_msg_md_local_ip6_2]);	\
+	r0 = *(u32*)(r1 + %[sk_msg_md_local_ip6_3]);	\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_local_ip6_0, offsetof(struct sk_msg_md, local_ip6[0])),
+	  __imm_const(sk_msg_md_local_ip6_1, offsetof(struct sk_msg_md, local_ip6[1])),
+	  __imm_const(sk_msg_md_local_ip6_2, offsetof(struct sk_msg_md, local_ip6[2])),
+	  __imm_const(sk_msg_md_local_ip6_3, offsetof(struct sk_msg_md, local_ip6[3]))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("valid access size in SK_MSG")
+__success
+__naked void access_size_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[sk_msg_md_size]);		\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_size, offsetof(struct sk_msg_md, size))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("invalid 64B read of size in SK_MSG")
+__failure __msg("invalid bpf_context access")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void of_size_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + %[sk_msg_md_size]);		\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_size, offsetof(struct sk_msg_md, size))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("invalid read past end of SK_MSG")
+__failure __msg("invalid bpf_context access")
+__naked void past_end_of_sk_msg(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__imm_0]);			\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, offsetof(struct sk_msg_md, size) + 4)
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("invalid read offset in SK_MSG")
+__failure __msg("invalid bpf_context access")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void read_offset_in_sk_msg(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__imm_0]);			\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, offsetof(struct sk_msg_md, family) + 1)
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("direct packet read for SK_MSG")
+__success
+__naked void packet_read_for_sk_msg(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + %[sk_msg_md_data]);		\
+	r3 = *(u64*)(r1 + %[sk_msg_md_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_data, offsetof(struct sk_msg_md, data)),
+	  __imm_const(sk_msg_md_data_end, offsetof(struct sk_msg_md, data_end))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("direct packet write for SK_MSG")
+__success
+__naked void packet_write_for_sk_msg(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + %[sk_msg_md_data]);		\
+	r3 = *(u64*)(r1 + %[sk_msg_md_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_data, offsetof(struct sk_msg_md, data)),
+	  __imm_const(sk_msg_md_data_end, offsetof(struct sk_msg_md, data_end))
+	: __clobber_all);
+}
+
+SEC("sk_msg")
+__description("overlapping checks for direct packet access SK_MSG")
+__success
+__naked void direct_packet_access_sk_msg(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + %[sk_msg_md_data]);		\
+	r3 = *(u64*)(r1 + %[sk_msg_md_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u16*)(r2 + 6);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(sk_msg_md_data, offsetof(struct sk_msg_md, data)),
+	  __imm_const(sk_msg_md_data_end, offsetof(struct sk_msg_md, data_end))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/ctx_sk_msg.c b/tools/testing/selftests/bpf/verifier/ctx_sk_msg.c
deleted file mode 100644
index c6c69220a569..000000000000
--- a/tools/testing/selftests/bpf/verifier/ctx_sk_msg.c
+++ /dev/null
@@ -1,181 +0,0 @@
-{
-	"valid access family in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, family)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"valid access remote_ip4 in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, remote_ip4)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"valid access local_ip4 in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, local_ip4)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"valid access remote_port in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, remote_port)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"valid access local_port in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, local_port)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"valid access remote_ip6 in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, remote_ip6[0])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, remote_ip6[1])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, remote_ip6[2])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, remote_ip6[3])),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_SKB,
-},
-{
-	"valid access local_ip6 in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, local_ip6[0])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, local_ip6[1])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, local_ip6[2])),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, local_ip6[3])),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_SKB,
-},
-{
-	"valid access size in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct sk_msg_md, size)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"invalid 64B read of size in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct sk_msg_md, size)),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid bpf_context access",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"invalid read past end of SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct sk_msg_md, size) + 4),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid bpf_context access",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"invalid read offset in SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct sk_msg_md, family) + 1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid bpf_context access",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet read for SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct sk_msg_md, data)),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct sk_msg_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"direct packet write for SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct sk_msg_md, data)),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct sk_msg_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-{
-	"overlapping checks for direct packet access SK_MSG",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct sk_msg_md, data)),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct sk_msg_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_2, 6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SK_MSG,
-},
-- 
cgit v1.2.3-70-g09d2


From 84988478fb2c9068c6adf107eb630c48c00ff690 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:58 +0200
Subject: selftests/bpf: verifier/direct_stack_access_wraparound.c converted to
 inline assembly

Test verifier/direct_stack_access_wraparound.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-18-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../verifier_direct_stack_access_wraparound.c      | 56 ++++++++++++++++++++++
 .../bpf/verifier/direct_stack_access_wraparound.c  | 40 ----------------
 3 files changed, 58 insertions(+), 40 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_direct_stack_access_wraparound.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/direct_stack_access_wraparound.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 29351c774ee2..8c33b8792a0a 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -14,6 +14,7 @@
 #include "verifier_cgroup_storage.skel.h"
 #include "verifier_const_or.skel.h"
 #include "verifier_ctx_sk_msg.skel.h"
+#include "verifier_direct_stack_access_wraparound.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -50,3 +51,4 @@ void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
 void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
 void test_verifier_const_or(void)             { RUN(verifier_const_or); }
 void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
+void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_direct_stack_access_wraparound.c b/tools/testing/selftests/bpf/progs/verifier_direct_stack_access_wraparound.c
new file mode 100644
index 000000000000..c538c6893552
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_direct_stack_access_wraparound.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/direct_stack_access_wraparound.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("direct stack access with 32-bit wraparound. test1")
+__failure __msg("fp pointer and 2147483647")
+__failure_unpriv
+__naked void with_32_bit_wraparound_test1(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += 0x7fffffff;				\
+	r1 += 0x7fffffff;				\
+	w0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("direct stack access with 32-bit wraparound. test2")
+__failure __msg("fp pointer and 1073741823")
+__failure_unpriv
+__naked void with_32_bit_wraparound_test2(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += 0x3fffffff;				\
+	r1 += 0x3fffffff;				\
+	w0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("direct stack access with 32-bit wraparound. test3")
+__failure __msg("fp pointer offset 1073741822")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void with_32_bit_wraparound_test3(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += 0x1fffffff;				\
+	r1 += 0x1fffffff;				\
+	w0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/direct_stack_access_wraparound.c b/tools/testing/selftests/bpf/verifier/direct_stack_access_wraparound.c
deleted file mode 100644
index 698e3779fdd2..000000000000
--- a/tools/testing/selftests/bpf/verifier/direct_stack_access_wraparound.c
+++ /dev/null
@@ -1,40 +0,0 @@
-{
-	"direct stack access with 32-bit wraparound. test1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x7fffffff),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x7fffffff),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "fp pointer and 2147483647",
-	.result = REJECT
-},
-{
-	"direct stack access with 32-bit wraparound. test2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x3fffffff),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x3fffffff),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "fp pointer and 1073741823",
-	.result = REJECT
-},
-{
-	"direct stack access with 32-bit wraparound. test3",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x1fffffff),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x1fffffff),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "fp pointer offset 1073741822",
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.result = REJECT
-},
-- 
cgit v1.2.3-70-g09d2


From 01a0925531a4ec962c88ceccb464c1c1178e9d81 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:54:59 +0200
Subject: selftests/bpf: verifier/div0.c converted to inline assembly

Test verifier/div0.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-19-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   2 +
 tools/testing/selftests/bpf/progs/verifier_div0.c | 213 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/div0.c       | 184 -------------------
 3 files changed, 215 insertions(+), 184 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_div0.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/div0.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 8c33b8792a0a..b172c41cdc61 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -15,6 +15,7 @@
 #include "verifier_const_or.skel.h"
 #include "verifier_ctx_sk_msg.skel.h"
 #include "verifier_direct_stack_access_wraparound.skel.h"
+#include "verifier_div0.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -52,3 +53,4 @@ void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
 void test_verifier_const_or(void)             { RUN(verifier_const_or); }
 void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
 void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
+void test_verifier_div0(void)                 { RUN(verifier_div0); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_div0.c b/tools/testing/selftests/bpf/progs/verifier_div0.c
new file mode 100644
index 000000000000..cca5ea18fc28
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_div0.c
@@ -0,0 +1,213 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/div0.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("DIV32 by 0, zero check 1")
+__success __success_unpriv __retval(42)
+__naked void by_0_zero_check_1_1(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	w1 = 0;						\
+	w2 = 1;						\
+	w2 /= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("DIV32 by 0, zero check 2")
+__success __success_unpriv __retval(42)
+__naked void by_0_zero_check_2_1(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	r1 = 0xffffffff00000000LL ll;			\
+	w2 = 1;						\
+	w2 /= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("DIV64 by 0, zero check")
+__success __success_unpriv __retval(42)
+__naked void div64_by_0_zero_check(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	w1 = 0;						\
+	w2 = 1;						\
+	r2 /= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("MOD32 by 0, zero check 1")
+__success __success_unpriv __retval(42)
+__naked void by_0_zero_check_1_2(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	w1 = 0;						\
+	w2 = 1;						\
+	w2 %%= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("MOD32 by 0, zero check 2")
+__success __success_unpriv __retval(42)
+__naked void by_0_zero_check_2_2(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	r1 = 0xffffffff00000000LL ll;			\
+	w2 = 1;						\
+	w2 %%= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("MOD64 by 0, zero check")
+__success __success_unpriv __retval(42)
+__naked void mod64_by_0_zero_check(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	w1 = 0;						\
+	w2 = 1;						\
+	r2 %%= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV32 by 0, zero check ok, cls")
+__success __retval(8)
+__naked void _0_zero_check_ok_cls_1(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	w1 = 2;						\
+	w2 = 16;					\
+	w2 /= w1;					\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV32 by 0, zero check 1, cls")
+__success __retval(0)
+__naked void _0_zero_check_1_cls_1(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w0 = 1;						\
+	w0 /= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV32 by 0, zero check 2, cls")
+__success __retval(0)
+__naked void _0_zero_check_2_cls_1(void)
+{
+	asm volatile ("					\
+	r1 = 0xffffffff00000000LL ll;			\
+	w0 = 1;						\
+	w0 /= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV64 by 0, zero check, cls")
+__success __retval(0)
+__naked void by_0_zero_check_cls(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w0 = 1;						\
+	r0 /= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD32 by 0, zero check ok, cls")
+__success __retval(2)
+__naked void _0_zero_check_ok_cls_2(void)
+{
+	asm volatile ("					\
+	w0 = 42;					\
+	w1 = 3;						\
+	w2 = 5;						\
+	w2 %%= w1;					\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD32 by 0, zero check 1, cls")
+__success __retval(1)
+__naked void _0_zero_check_1_cls_2(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w0 = 1;						\
+	w0 %%= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD32 by 0, zero check 2, cls")
+__success __retval(1)
+__naked void _0_zero_check_2_cls_2(void)
+{
+	asm volatile ("					\
+	r1 = 0xffffffff00000000LL ll;			\
+	w0 = 1;						\
+	w0 %%= w1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD64 by 0, zero check 1, cls")
+__success __retval(2)
+__naked void _0_zero_check_1_cls_3(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w0 = 2;						\
+	r0 %%= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD64 by 0, zero check 2, cls")
+__success __retval(-1)
+__naked void _0_zero_check_2_cls_3(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w0 = -1;					\
+	r0 %%= r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/div0.c b/tools/testing/selftests/bpf/verifier/div0.c
deleted file mode 100644
index 7685edfbcf71..000000000000
--- a/tools/testing/selftests/bpf/verifier/div0.c
+++ /dev/null
@@ -1,184 +0,0 @@
-{
-	"DIV32 by 0, zero check 1",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, 1),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"DIV32 by 0, zero check 2",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_LD_IMM64(BPF_REG_1, 0xffffffff00000000LL),
-	BPF_MOV32_IMM(BPF_REG_2, 1),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"DIV64 by 0, zero check",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, 1),
-	BPF_ALU64_REG(BPF_DIV, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"MOD32 by 0, zero check 1",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, 1),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"MOD32 by 0, zero check 2",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_LD_IMM64(BPF_REG_1, 0xffffffff00000000LL),
-	BPF_MOV32_IMM(BPF_REG_2, 1),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"MOD64 by 0, zero check",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, 1),
-	BPF_ALU64_REG(BPF_MOD, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"DIV32 by 0, zero check ok, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_MOV32_IMM(BPF_REG_1, 2),
-	BPF_MOV32_IMM(BPF_REG_2, 16),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_2, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 8,
-},
-{
-	"DIV32 by 0, zero check 1, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"DIV32 by 0, zero check 2, cls",
-	.insns = {
-	BPF_LD_IMM64(BPF_REG_1, 0xffffffff00000000LL),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"DIV64 by 0, zero check, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_ALU64_REG(BPF_DIV, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"MOD32 by 0, zero check ok, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, 42),
-	BPF_MOV32_IMM(BPF_REG_1, 3),
-	BPF_MOV32_IMM(BPF_REG_2, 5),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_2, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 2,
-},
-{
-	"MOD32 by 0, zero check 1, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"MOD32 by 0, zero check 2, cls",
-	.insns = {
-	BPF_LD_IMM64(BPF_REG_1, 0xffffffff00000000LL),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"MOD64 by 0, zero check 1, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_0, 2),
-	BPF_ALU64_REG(BPF_MOD, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 2,
-},
-{
-	"MOD64 by 0, zero check 2, cls",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_0, -1),
-	BPF_ALU64_REG(BPF_MOD, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = -1,
-},
-- 
cgit v1.2.3-70-g09d2


From 9553de70a8412a07b16703449b4b4c4e5d37c388 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:00 +0200
Subject: selftests/bpf: verifier/div_overflow.c converted to inline assembly

Test verifier/div_overflow.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-20-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_div_overflow.c    | 144 +++++++++++++++++++++
 .../testing/selftests/bpf/verifier/div_overflow.c  | 110 ----------------
 3 files changed, 146 insertions(+), 110 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_div_overflow.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/div_overflow.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index b172c41cdc61..d92211b4c3af 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -16,6 +16,7 @@
 #include "verifier_ctx_sk_msg.skel.h"
 #include "verifier_direct_stack_access_wraparound.skel.h"
 #include "verifier_div0.skel.h"
+#include "verifier_div_overflow.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -54,3 +55,4 @@ void test_verifier_const_or(void)             { RUN(verifier_const_or); }
 void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
 void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
 void test_verifier_div0(void)                 { RUN(verifier_div0); }
+void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_div_overflow.c b/tools/testing/selftests/bpf/progs/verifier_div_overflow.c
new file mode 100644
index 000000000000..458984da804c
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_div_overflow.c
@@ -0,0 +1,144 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/div_overflow.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <limits.h>
+#include "bpf_misc.h"
+
+/* Just make sure that JITs used udiv/umod as otherwise we get
+ * an exception from INT_MIN/-1 overflow similarly as with div
+ * by zero.
+ */
+
+SEC("tc")
+__description("DIV32 overflow, check 1")
+__success __retval(0)
+__naked void div32_overflow_check_1(void)
+{
+	asm volatile ("					\
+	w1 = -1;					\
+	w0 = %[int_min];				\
+	w0 /= w1;					\
+	exit;						\
+"	:
+	: __imm_const(int_min, INT_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV32 overflow, check 2")
+__success __retval(0)
+__naked void div32_overflow_check_2(void)
+{
+	asm volatile ("					\
+	w0 = %[int_min];				\
+	w0 /= -1;					\
+	exit;						\
+"	:
+	: __imm_const(int_min, INT_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV64 overflow, check 1")
+__success __retval(0)
+__naked void div64_overflow_check_1(void)
+{
+	asm volatile ("					\
+	r1 = -1;					\
+	r2 = %[llong_min] ll;				\
+	r2 /= r1;					\
+	w0 = 0;						\
+	if r0 == r2 goto l0_%=;				\
+	w0 = 1;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(llong_min, LLONG_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("DIV64 overflow, check 2")
+__success __retval(0)
+__naked void div64_overflow_check_2(void)
+{
+	asm volatile ("					\
+	r1 = %[llong_min] ll;				\
+	r1 /= -1;					\
+	w0 = 0;						\
+	if r0 == r1 goto l0_%=;				\
+	w0 = 1;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(llong_min, LLONG_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD32 overflow, check 1")
+__success __retval(INT_MIN)
+__naked void mod32_overflow_check_1(void)
+{
+	asm volatile ("					\
+	w1 = -1;					\
+	w0 = %[int_min];				\
+	w0 %%= w1;					\
+	exit;						\
+"	:
+	: __imm_const(int_min, INT_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD32 overflow, check 2")
+__success __retval(INT_MIN)
+__naked void mod32_overflow_check_2(void)
+{
+	asm volatile ("					\
+	w0 = %[int_min];				\
+	w0 %%= -1;					\
+	exit;						\
+"	:
+	: __imm_const(int_min, INT_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD64 overflow, check 1")
+__success __retval(1)
+__naked void mod64_overflow_check_1(void)
+{
+	asm volatile ("					\
+	r1 = -1;					\
+	r2 = %[llong_min] ll;				\
+	r3 = r2;					\
+	r2 %%= r1;					\
+	w0 = 0;						\
+	if r3 != r2 goto l0_%=;				\
+	w0 = 1;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(llong_min, LLONG_MIN)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("MOD64 overflow, check 2")
+__success __retval(1)
+__naked void mod64_overflow_check_2(void)
+{
+	asm volatile ("					\
+	r2 = %[llong_min] ll;				\
+	r3 = r2;					\
+	r2 %%= -1;					\
+	w0 = 0;						\
+	if r3 != r2 goto l0_%=;				\
+	w0 = 1;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(llong_min, LLONG_MIN)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/div_overflow.c b/tools/testing/selftests/bpf/verifier/div_overflow.c
deleted file mode 100644
index acab4f00819f..000000000000
--- a/tools/testing/selftests/bpf/verifier/div_overflow.c
+++ /dev/null
@@ -1,110 +0,0 @@
-/* Just make sure that JITs used udiv/umod as otherwise we get
- * an exception from INT_MIN/-1 overflow similarly as with div
- * by zero.
- */
-{
-	"DIV32 overflow, check 1",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_0, INT_MIN),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"DIV32 overflow, check 2",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, INT_MIN),
-	BPF_ALU32_IMM(BPF_DIV, BPF_REG_0, -1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"DIV64 overflow, check 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_LD_IMM64(BPF_REG_2, LLONG_MIN),
-	BPF_ALU64_REG(BPF_DIV, BPF_REG_2, BPF_REG_1),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_2, 1),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"DIV64 overflow, check 2",
-	.insns = {
-	BPF_LD_IMM64(BPF_REG_1, LLONG_MIN),
-	BPF_ALU64_IMM(BPF_DIV, BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"MOD32 overflow, check 1",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_0, INT_MIN),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = INT_MIN,
-},
-{
-	"MOD32 overflow, check 2",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_0, INT_MIN),
-	BPF_ALU32_IMM(BPF_MOD, BPF_REG_0, -1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = INT_MIN,
-},
-{
-	"MOD64 overflow, check 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_LD_IMM64(BPF_REG_2, LLONG_MIN),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_2),
-	BPF_ALU64_REG(BPF_MOD, BPF_REG_2, BPF_REG_1),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_JMP_REG(BPF_JNE, BPF_REG_3, BPF_REG_2, 1),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"MOD64 overflow, check 2",
-	.insns = {
-	BPF_LD_IMM64(BPF_REG_2, LLONG_MIN),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_MOD, BPF_REG_2, -1),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_JMP_REG(BPF_JNE, BPF_REG_3, BPF_REG_2, 1),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 1,
-},
-- 
cgit v1.2.3-70-g09d2


From b37d776b431ec056075feeeaddcdee4512c522f6 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:01 +0200
Subject: selftests/bpf: verifier/helper_access_var_len.c converted to inline
 assembly

Test verifier/helper_access_var_len.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-21-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_helper_access_var_len.c     | 825 +++++++++++++++++++++
 .../selftests/bpf/verifier/helper_access_var_len.c | 650 ----------------
 3 files changed, 827 insertions(+), 650 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_helper_access_var_len.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/helper_access_var_len.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index d92211b4c3af..22d7e152c05e 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -17,6 +17,7 @@
 #include "verifier_direct_stack_access_wraparound.skel.h"
 #include "verifier_div0.skel.h"
 #include "verifier_div_overflow.skel.h"
+#include "verifier_helper_access_var_len.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -56,3 +57,4 @@ void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
 void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
 void test_verifier_div0(void)                 { RUN(verifier_div0); }
 void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
+void test_verifier_helper_access_var_len(void) { RUN(verifier_helper_access_var_len); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_helper_access_var_len.c b/tools/testing/selftests/bpf/progs/verifier_helper_access_var_len.c
new file mode 100644
index 000000000000..50c6b22606f6
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_helper_access_var_len.c
@@ -0,0 +1,825 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/helper_access_var_len.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_RINGBUF);
+	__uint(max_entries, 4096);
+} map_ringbuf SEC(".maps");
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, bitwise AND + JMP, correct bounds")
+__success
+__naked void bitwise_and_jmp_correct_bounds(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -64;					\
+	r0 = 0;						\
+	*(u64*)(r10 - 64) = r0;				\
+	*(u64*)(r10 - 56) = r0;				\
+	*(u64*)(r10 - 48) = r0;				\
+	*(u64*)(r10 - 40) = r0;				\
+	*(u64*)(r10 - 32) = r0;				\
+	*(u64*)(r10 - 24) = r0;				\
+	*(u64*)(r10 - 16) = r0;				\
+	*(u64*)(r10 - 8) = r0;				\
+	r2 = 16;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	r2 &= 64;					\
+	r4 = 0;						\
+	if r4 >= r2 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("helper access to variable memory: stack, bitwise AND, zero included")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid indirect read from stack R2 off -64+0 size 64")
+__retval(0)
+__naked void stack_bitwise_and_zero_included(void)
+{
+	asm volatile ("					\
+	/* set max stack size */			\
+	r6 = 0;						\
+	*(u64*)(r10 - 128) = r6;			\
+	/* set r3 to a random value */			\
+	call %[bpf_get_prandom_u32];			\
+	r3 = r0;					\
+	/* use bitwise AND to limit r3 range to [0, 64] */\
+	r3 &= 64;					\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = r10;					\
+	r2 += -64;					\
+	r4 = 0;						\
+	/* Call bpf_ringbuf_output(), it is one of a few helper functions with\
+	 * ARG_CONST_SIZE_OR_ZERO parameter allowed in unpriv mode.\
+	 * For unpriv this should signal an error, because memory at &fp[-64] is\
+	 * not initialized.				\
+	 */						\
+	call %[bpf_ringbuf_output];			\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_ringbuf_output),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, bitwise AND + JMP, wrong max")
+__failure __msg("invalid indirect access to stack R1 off=-64 size=65")
+__naked void bitwise_and_jmp_wrong_max(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + 8);				\
+	r1 = r10;					\
+	r1 += -64;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	r2 &= 65;					\
+	r4 = 0;						\
+	if r4 >= r2 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, JMP, correct bounds")
+__success
+__naked void memory_stack_jmp_correct_bounds(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -64;					\
+	r0 = 0;						\
+	*(u64*)(r10 - 64) = r0;				\
+	*(u64*)(r10 - 56) = r0;				\
+	*(u64*)(r10 - 48) = r0;				\
+	*(u64*)(r10 - 40) = r0;				\
+	*(u64*)(r10 - 32) = r0;				\
+	*(u64*)(r10 - 24) = r0;				\
+	*(u64*)(r10 - 16) = r0;				\
+	*(u64*)(r10 - 8) = r0;				\
+	r2 = 16;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	if r2 > 64 goto l0_%=;				\
+	r4 = 0;						\
+	if r4 >= r2 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, JMP (signed), correct bounds")
+__success
+__naked void stack_jmp_signed_correct_bounds(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -64;					\
+	r0 = 0;						\
+	*(u64*)(r10 - 64) = r0;				\
+	*(u64*)(r10 - 56) = r0;				\
+	*(u64*)(r10 - 48) = r0;				\
+	*(u64*)(r10 - 40) = r0;				\
+	*(u64*)(r10 - 32) = r0;				\
+	*(u64*)(r10 - 24) = r0;				\
+	*(u64*)(r10 - 16) = r0;				\
+	*(u64*)(r10 - 8) = r0;				\
+	r2 = 16;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	if r2 s> 64 goto l0_%=;				\
+	r4 = 0;						\
+	if r4 s>= r2 goto l0_%=;			\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, JMP, bounds + offset")
+__failure __msg("invalid indirect access to stack R1 off=-64 size=65")
+__naked void memory_stack_jmp_bounds_offset(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + 8);				\
+	r1 = r10;					\
+	r1 += -64;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	if r2 > 64 goto l0_%=;				\
+	r4 = 0;						\
+	if r4 >= r2 goto l0_%=;				\
+	r2 += 1;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, JMP, wrong max")
+__failure __msg("invalid indirect access to stack R1 off=-64 size=65")
+__naked void memory_stack_jmp_wrong_max(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + 8);				\
+	r1 = r10;					\
+	r1 += -64;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	if r2 > 65 goto l0_%=;				\
+	r4 = 0;						\
+	if r4 >= r2 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, JMP, no max check")
+__failure
+/* because max wasn't checked, signed min is negative */
+__msg("R2 min value is negative, either use unsigned or 'var &= const'")
+__naked void stack_jmp_no_max_check(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + 8);				\
+	r1 = r10;					\
+	r1 += -64;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	r4 = 0;						\
+	if r4 >= r2 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("helper access to variable memory: stack, JMP, no min check")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid indirect read from stack R2 off -64+0 size 64")
+__retval(0)
+__naked void stack_jmp_no_min_check(void)
+{
+	asm volatile ("					\
+	/* set max stack size */			\
+	r6 = 0;						\
+	*(u64*)(r10 - 128) = r6;			\
+	/* set r3 to a random value */			\
+	call %[bpf_get_prandom_u32];			\
+	r3 = r0;					\
+	/* use JMP to limit r3 range to [0, 64] */	\
+	if r3 > 64 goto l0_%=;				\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = r10;					\
+	r2 += -64;					\
+	r4 = 0;						\
+	/* Call bpf_ringbuf_output(), it is one of a few helper functions with\
+	 * ARG_CONST_SIZE_OR_ZERO parameter allowed in unpriv mode.\
+	 * For unpriv this should signal an error, because memory at &fp[-64] is\
+	 * not initialized.				\
+	 */						\
+	call %[bpf_ringbuf_output];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_ringbuf_output),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: stack, JMP (signed), no min check")
+__failure __msg("R2 min value is negative")
+__naked void jmp_signed_no_min_check(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + 8);				\
+	r1 = r10;					\
+	r1 += -64;					\
+	*(u64*)(r1 - 128) = r2;				\
+	r2 = *(u64*)(r1 - 128);				\
+	if r2 s> 64 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: map, JMP, correct bounds")
+__success
+__naked void memory_map_jmp_correct_bounds(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[sizeof_test_val];			\
+	*(u64*)(r10 - 128) = r2;			\
+	r2 = *(u64*)(r10 - 128);			\
+	if r2 s> %[sizeof_test_val] goto l1_%=;		\
+	r4 = 0;						\
+	if r4 s>= r2 goto l1_%=;			\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(sizeof_test_val, sizeof(struct test_val))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: map, JMP, wrong max")
+__failure __msg("invalid access to map value, value_size=48 off=0 size=49")
+__naked void memory_map_jmp_wrong_max(void)
+{
+	asm volatile ("					\
+	r6 = *(u64*)(r1 + 8);				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = r6;					\
+	*(u64*)(r10 - 128) = r2;			\
+	r2 = *(u64*)(r10 - 128);			\
+	if r2 s> %[__imm_0] goto l1_%=;			\
+	r4 = 0;						\
+	if r4 s>= r2 goto l1_%=;			\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) + 1)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: map adjusted, JMP, correct bounds")
+__success
+__naked void map_adjusted_jmp_correct_bounds(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += 20;					\
+	r2 = %[sizeof_test_val];			\
+	*(u64*)(r10 - 128) = r2;			\
+	r2 = *(u64*)(r10 - 128);			\
+	if r2 s> %[__imm_0] goto l1_%=;			\
+	r4 = 0;						\
+	if r4 s>= r2 goto l1_%=;			\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - 20),
+	  __imm_const(sizeof_test_val, sizeof(struct test_val))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: map adjusted, JMP, wrong max")
+__failure __msg("R1 min value is outside of the allowed memory range")
+__naked void map_adjusted_jmp_wrong_max(void)
+{
+	asm volatile ("					\
+	r6 = *(u64*)(r1 + 8);				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += 20;					\
+	r2 = r6;					\
+	*(u64*)(r10 - 128) = r2;			\
+	r2 = *(u64*)(r10 - 128);			\
+	if r2 s> %[__imm_0] goto l1_%=;			\
+	r4 = 0;						\
+	if r4 s>= r2 goto l1_%=;			\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - 19)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size = 0 allowed on NULL (ARG_PTR_TO_MEM_OR_NULL)")
+__success __retval(0)
+__naked void ptr_to_mem_or_null_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	r2 = 0;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size > 0 not allowed on NULL (ARG_PTR_TO_MEM_OR_NULL)")
+__failure __msg("R1 type=scalar expected=fp")
+__naked void ptr_to_mem_or_null_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + 0);				\
+	r1 = 0;						\
+	*(u64*)(r10 - 128) = r2;			\
+	r2 = *(u64*)(r10 - 128);			\
+	r2 &= 64;					\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size = 0 allowed on != NULL stack pointer (ARG_PTR_TO_MEM_OR_NULL)")
+__success __retval(0)
+__naked void ptr_to_mem_or_null_3(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -8;					\
+	r2 = 0;						\
+	*(u64*)(r1 + 0) = r2;				\
+	r2 &= 8;					\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size = 0 allowed on != NULL map pointer (ARG_PTR_TO_MEM_OR_NULL)")
+__success __retval(0)
+__naked void ptr_to_mem_or_null_4(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = 0;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size possible = 0 allowed on != NULL stack pointer (ARG_PTR_TO_MEM_OR_NULL)")
+__success __retval(0)
+__naked void ptr_to_mem_or_null_5(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = *(u64*)(r0 + 0);				\
+	if r2 > 8 goto l0_%=;				\
+	r1 = r10;					\
+	r1 += -8;					\
+	*(u64*)(r1 + 0) = r2;				\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size possible = 0 allowed on != NULL map pointer (ARG_PTR_TO_MEM_OR_NULL)")
+__success __retval(0)
+__naked void ptr_to_mem_or_null_6(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = *(u64*)(r0 + 0);				\
+	if r2 > 8 goto l0_%=;				\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to variable memory: size possible = 0 allowed on != NULL packet pointer (ARG_PTR_TO_MEM_OR_NULL)")
+__success __retval(0)
+/* csum_diff of 64-byte packet */
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void ptr_to_mem_or_null_7(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r6;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = *(u64*)(r6 + 0);				\
+	if r2 > 8 goto l0_%=;				\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: size = 0 not allowed on NULL (!ARG_PTR_TO_MEM_OR_NULL)")
+__failure __msg("R1 type=scalar expected=fp")
+__naked void ptr_to_mem_or_null_8(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	r2 = 0;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: size > 0 not allowed on NULL (!ARG_PTR_TO_MEM_OR_NULL)")
+__failure __msg("R1 type=scalar expected=fp")
+__naked void ptr_to_mem_or_null_9(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	r2 = 1;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: size = 0 allowed on != NULL stack pointer (!ARG_PTR_TO_MEM_OR_NULL)")
+__success
+__naked void ptr_to_mem_or_null_10(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -8;					\
+	r2 = 0;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: size = 0 allowed on != NULL map pointer (!ARG_PTR_TO_MEM_OR_NULL)")
+__success
+__naked void ptr_to_mem_or_null_11(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = 0;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: size possible = 0 allowed on != NULL stack pointer (!ARG_PTR_TO_MEM_OR_NULL)")
+__success
+__naked void ptr_to_mem_or_null_12(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = *(u64*)(r0 + 0);				\
+	if r2 > 8 goto l0_%=;				\
+	r1 = r10;					\
+	r1 += -8;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: size possible = 0 allowed on != NULL map pointer (!ARG_PTR_TO_MEM_OR_NULL)")
+__success
+__naked void ptr_to_mem_or_null_13(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = *(u64*)(r0 + 0);				\
+	if r2 > 8 goto l0_%=;				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("helper access to variable memory: 8 bytes leak")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid indirect read from stack R2 off -64+32 size 64")
+__retval(0)
+__naked void variable_memory_8_bytes_leak(void)
+{
+	asm volatile ("					\
+	/* set max stack size */			\
+	r6 = 0;						\
+	*(u64*)(r10 - 128) = r6;			\
+	/* set r3 to a random value */			\
+	call %[bpf_get_prandom_u32];			\
+	r3 = r0;					\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = r10;					\
+	r2 += -64;					\
+	r0 = 0;						\
+	*(u64*)(r10 - 64) = r0;				\
+	*(u64*)(r10 - 56) = r0;				\
+	*(u64*)(r10 - 48) = r0;				\
+	*(u64*)(r10 - 40) = r0;				\
+	/* Note: fp[-32] left uninitialized */		\
+	*(u64*)(r10 - 24) = r0;				\
+	*(u64*)(r10 - 16) = r0;				\
+	*(u64*)(r10 - 8) = r0;				\
+	/* Limit r3 range to [1, 64] */			\
+	r3 &= 63;					\
+	r3 += 1;					\
+	r4 = 0;						\
+	/* Call bpf_ringbuf_output(), it is one of a few helper functions with\
+	 * ARG_CONST_SIZE_OR_ZERO parameter allowed in unpriv mode.\
+	 * For unpriv this should signal an error, because memory region [1, 64]\
+	 * at &fp[-64] is not fully initialized.	\
+	 */						\
+	call %[bpf_ringbuf_output];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_ringbuf_output),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to variable memory: 8 bytes no leak (init memory)")
+__success
+__naked void bytes_no_leak_init_memory(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r0 = 0;						\
+	r0 = 0;						\
+	*(u64*)(r10 - 64) = r0;				\
+	*(u64*)(r10 - 56) = r0;				\
+	*(u64*)(r10 - 48) = r0;				\
+	*(u64*)(r10 - 40) = r0;				\
+	*(u64*)(r10 - 32) = r0;				\
+	*(u64*)(r10 - 24) = r0;				\
+	*(u64*)(r10 - 16) = r0;				\
+	*(u64*)(r10 - 8) = r0;				\
+	r1 += -64;					\
+	r2 = 0;						\
+	r2 &= 32;					\
+	r2 += 32;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+	r1 = *(u64*)(r10 - 16);				\
+	exit;						\
+"	:
+	: __imm(bpf_probe_read_kernel)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/helper_access_var_len.c b/tools/testing/selftests/bpf/verifier/helper_access_var_len.c
deleted file mode 100644
index 9c4885885aba..000000000000
--- a/tools/testing/selftests/bpf/verifier/helper_access_var_len.c
+++ /dev/null
@@ -1,650 +0,0 @@
-{
-	"helper access to variable memory: stack, bitwise AND + JMP, correct bounds",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -56),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -48),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -40),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -32),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_MOV64_IMM(BPF_REG_2, 16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 64),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, bitwise AND, zero included",
-	.insns = {
-	/* set max stack size */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -128, 0),
-	/* set r3 to a random value */
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	/* use bitwise AND to limit r3 range to [0, 64] */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_3, 64),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -64),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	/* Call bpf_ringbuf_output(), it is one of a few helper functions with
-	 * ARG_CONST_SIZE_OR_ZERO parameter allowed in unpriv mode.
-	 * For unpriv this should signal an error, because memory at &fp[-64] is
-	 * not initialized.
-	 */
-	BPF_EMIT_CALL(BPF_FUNC_ringbuf_output),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 4 },
-	.errstr_unpriv = "invalid indirect read from stack R2 off -64+0 size 64",
-	.result_unpriv = REJECT,
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-{
-	"helper access to variable memory: stack, bitwise AND + JMP, wrong max",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 65),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid indirect access to stack R1 off=-64 size=65",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, JMP, correct bounds",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -56),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -48),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -40),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -32),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_MOV64_IMM(BPF_REG_2, 16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 64, 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, JMP (signed), correct bounds",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -56),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -48),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -40),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -32),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_MOV64_IMM(BPF_REG_2, 16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_2, 64, 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, JMP, bounds + offset",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 64, 5),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_4, BPF_REG_2, 3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid indirect access to stack R1 off=-64 size=65",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, JMP, wrong max",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 65, 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid indirect access to stack R1 off=-64 size=65",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, JMP, no max check",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	/* because max wasn't checked, signed min is negative */
-	.errstr = "R2 min value is negative, either use unsigned or 'var &= const'",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: stack, JMP, no min check",
-	.insns = {
-	/* set max stack size */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -128, 0),
-	/* set r3 to a random value */
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	/* use JMP to limit r3 range to [0, 64] */
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, 64, 6),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -64),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	/* Call bpf_ringbuf_output(), it is one of a few helper functions with
-	 * ARG_CONST_SIZE_OR_ZERO parameter allowed in unpriv mode.
-	 * For unpriv this should signal an error, because memory at &fp[-64] is
-	 * not initialized.
-	 */
-	BPF_EMIT_CALL(BPF_FUNC_ringbuf_output),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 4 },
-	.errstr_unpriv = "invalid indirect read from stack R2 off -64+0 size 64",
-	.result_unpriv = REJECT,
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-{
-	"helper access to variable memory: stack, JMP (signed), no min check",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_1, -128),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_2, 64, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R2 min value is negative",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: map, JMP, correct bounds",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 10),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, sizeof(struct test_val)),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -128),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_2, sizeof(struct test_val), 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: map, JMP, wrong max",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 10),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_6),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -128),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_2, sizeof(struct test_val) + 1, 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.errstr = "invalid access to map value, value_size=48 off=0 size=49",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: map adjusted, JMP, correct bounds",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 20),
-	BPF_MOV64_IMM(BPF_REG_2, sizeof(struct test_val)),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -128),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_2, sizeof(struct test_val) - 20, 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: map adjusted, JMP, wrong max",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 20),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_6),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -128),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_2, sizeof(struct test_val) - 19, 4),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_4, BPF_REG_2, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.errstr = "R1 min value is outside of the allowed memory range",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: size = 0 allowed on NULL (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to variable memory: size > 0 not allowed on NULL (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -128),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -128),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 64),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 type=scalar expected=fp",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to variable memory: size = 0 allowed on != NULL stack pointer (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to variable memory: size = 0 allowed on != NULL map pointer (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to variable memory: size possible = 0 allowed on != NULL stack pointer (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 8, 7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to variable memory: size possible = 0 allowed on != NULL map pointer (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 8, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to variable memory: size possible = 0 allowed on != NULL packet pointer (ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_6, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 8, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_csum_diff),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 0 /* csum_diff of 64-byte packet */,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"helper access to variable memory: size = 0 not allowed on NULL (!ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 type=scalar expected=fp",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: size > 0 not allowed on NULL (!ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 type=scalar expected=fp",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: size = 0 allowed on != NULL stack pointer (!ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: size = 0 allowed on != NULL map pointer (!ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: size possible = 0 allowed on != NULL stack pointer (!ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 8, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: size possible = 0 allowed on != NULL map pointer (!ARG_PTR_TO_MEM_OR_NULL)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_2, 8, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to variable memory: 8 bytes leak",
-	.insns = {
-	/* set max stack size */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -128, 0),
-	/* set r3 to a random value */
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -64),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -56),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -48),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -40),
-	/* Note: fp[-32] left uninitialized */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	/* Limit r3 range to [1, 64] */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_3, 63),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 1),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	/* Call bpf_ringbuf_output(), it is one of a few helper functions with
-	 * ARG_CONST_SIZE_OR_ZERO parameter allowed in unpriv mode.
-	 * For unpriv this should signal an error, because memory region [1, 64]
-	 * at &fp[-64] is not fully initialized.
-	 */
-	BPF_EMIT_CALL(BPF_FUNC_ringbuf_output),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 3 },
-	.errstr_unpriv = "invalid indirect read from stack R2 off -64+32 size 64",
-	.result_unpriv = REJECT,
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-{
-	"helper access to variable memory: 8 bytes no leak (init memory)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -64),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -56),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -48),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -40),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -32),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -64),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 32),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 32),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_10, -16),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-- 
cgit v1.2.3-70-g09d2


From fb179fe69e6a7bf66232d72a77f53ec10c026ee7 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:02 +0200
Subject: selftests/bpf: verifier/helper_packet_access.c converted to inline
 assembly

Test verifier/helper_packet_access.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-22-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_helper_packet_access.c      | 550 +++++++++++++++++++++
 .../selftests/bpf/verifier/helper_packet_access.c  | 460 -----------------
 3 files changed, 552 insertions(+), 460 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_helper_packet_access.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/helper_packet_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 22d7e152c05e..1cd162daf150 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -18,6 +18,7 @@
 #include "verifier_div0.skel.h"
 #include "verifier_div_overflow.skel.h"
 #include "verifier_helper_access_var_len.skel.h"
+#include "verifier_helper_packet_access.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -58,3 +59,4 @@ void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_st
 void test_verifier_div0(void)                 { RUN(verifier_div0); }
 void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
 void test_verifier_helper_access_var_len(void) { RUN(verifier_helper_access_var_len); }
+void test_verifier_helper_packet_access(void) { RUN(verifier_helper_packet_access); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_helper_packet_access.c b/tools/testing/selftests/bpf/progs/verifier_helper_packet_access.c
new file mode 100644
index 000000000000..74f5f9cd153d
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_helper_packet_access.c
@@ -0,0 +1,550 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/helper_packet_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("xdp")
+__description("helper access to packet: test1, valid packet_ptr range")
+__success __retval(0)
+__naked void test1_valid_packet_ptr_range(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = %[map_hash_8b] ll;				\
+	r3 = r2;					\
+	r4 = 0;						\
+	call %[bpf_map_update_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_update_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("helper access to packet: test2, unchecked packet_ptr")
+__failure __msg("invalid access to packet")
+__naked void packet_test2_unchecked_packet_ptr(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(xdp_md_data, offsetof(struct xdp_md, data))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("helper access to packet: test3, variable add")
+__success __retval(0)
+__naked void to_packet_test3_variable_add(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r4 = r2;					\
+	r4 += 8;					\
+	if r4 > r3 goto l0_%=;				\
+	r5 = *(u8*)(r2 + 0);				\
+	r4 = r2;					\
+	r4 += r5;					\
+	r5 = r4;					\
+	r5 += 8;					\
+	if r5 > r3 goto l0_%=;				\
+	r1 = %[map_hash_8b] ll;				\
+	r2 = r4;					\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("helper access to packet: test4, packet_ptr with bad range")
+__failure __msg("invalid access to packet")
+__naked void packet_ptr_with_bad_range_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r4 = r2;					\
+	r4 += 4;					\
+	if r4 > r3 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("helper access to packet: test5, packet_ptr with too short range")
+__failure __msg("invalid access to packet")
+__naked void ptr_with_too_short_range_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r2 += 1;					\
+	r4 = r2;					\
+	r4 += 7;					\
+	if r4 > r3 goto l0_%=;				\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test6, cls valid packet_ptr range")
+__success __retval(0)
+__naked void cls_valid_packet_ptr_range(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = %[map_hash_8b] ll;				\
+	r3 = r2;					\
+	r4 = 0;						\
+	call %[bpf_map_update_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_update_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test7, cls unchecked packet_ptr")
+__failure __msg("invalid access to packet")
+__naked void test7_cls_unchecked_packet_ptr(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test8, cls variable add")
+__success __retval(0)
+__naked void packet_test8_cls_variable_add(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r4 = r2;					\
+	r4 += 8;					\
+	if r4 > r3 goto l0_%=;				\
+	r5 = *(u8*)(r2 + 0);				\
+	r4 = r2;					\
+	r4 += r5;					\
+	r5 = r4;					\
+	r5 += 8;					\
+	if r5 > r3 goto l0_%=;				\
+	r1 = %[map_hash_8b] ll;				\
+	r2 = r4;					\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test9, cls packet_ptr with bad range")
+__failure __msg("invalid access to packet")
+__naked void packet_ptr_with_bad_range_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r4 = r2;					\
+	r4 += 4;					\
+	if r4 > r3 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test10, cls packet_ptr with too short range")
+__failure __msg("invalid access to packet")
+__naked void ptr_with_too_short_range_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r2 += 1;					\
+	r4 = r2;					\
+	r4 += 7;					\
+	if r4 > r3 goto l0_%=;				\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test11, cls unsuitable helper 1")
+__failure __msg("helper access to the packet")
+__naked void test11_cls_unsuitable_helper_1(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r3 = r6;					\
+	r3 += 7;					\
+	if r3 > r7 goto l0_%=;				\
+	r2 = 0;						\
+	r4 = 42;					\
+	r5 = 0;						\
+	call %[bpf_skb_store_bytes];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_skb_store_bytes),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test12, cls unsuitable helper 2")
+__failure __msg("helper access to the packet")
+__naked void test12_cls_unsuitable_helper_2(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r3 = r6;					\
+	r6 += 8;					\
+	if r6 > r7 goto l0_%=;				\
+	r2 = 0;						\
+	r4 = 4;						\
+	call %[bpf_skb_load_bytes];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test13, cls helper ok")
+__success __retval(0)
+__naked void packet_test13_cls_helper_ok(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test14, cls helper ok sub")
+__success __retval(0)
+__naked void test14_cls_helper_ok_sub(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 -= 4;					\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test15, cls helper fail sub")
+__failure __msg("invalid access to packet")
+__naked void test15_cls_helper_fail_sub(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 -= 12;					\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test16, cls helper fail range 1")
+__failure __msg("invalid access to packet")
+__naked void cls_helper_fail_range_1(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = 8;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test17, cls helper fail range 2")
+__failure __msg("R2 min value is negative")
+__naked void cls_helper_fail_range_2(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = -9;					\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test18, cls helper fail range 3")
+__failure __msg("R2 min value is negative")
+__naked void cls_helper_fail_range_3(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__imm_0, ~0),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test19, cls helper range zero")
+__success __retval(0)
+__naked void test19_cls_helper_range_zero(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = 0;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test20, pkt end as input")
+__failure __msg("R1 type=pkt_end expected=fp")
+__naked void test20_pkt_end_as_input(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r1 = r7;					\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("helper access to packet: test21, wrong reg")
+__failure __msg("invalid access to packet")
+__naked void to_packet_test21_wrong_reg(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r7 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r6 += 1;					\
+	r1 = r6;					\
+	r1 += 7;					\
+	if r1 > r7 goto l0_%=;				\
+	r2 = 4;						\
+	r3 = 0;						\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_csum_diff];				\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_csum_diff),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/helper_packet_access.c b/tools/testing/selftests/bpf/verifier/helper_packet_access.c
deleted file mode 100644
index ae54587e9829..000000000000
--- a/tools/testing/selftests/bpf/verifier/helper_packet_access.c
+++ /dev/null
@@ -1,460 +0,0 @@
-{
-	"helper access to packet: test1, valid packet_ptr range",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 5),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_2),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_update_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.result_unpriv = ACCEPT,
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"helper access to packet: test2, unchecked packet_ptr",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"helper access to packet: test3, variable add",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-			offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 10),
-	BPF_LDX_MEM(BPF_B, BPF_REG_5, BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_5),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_5, BPF_REG_3, 4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 11 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"helper access to packet: test4, packet_ptr with bad range",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 4),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 7 },
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"helper access to packet: test5, packet_ptr with too short range",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 1),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 6 },
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"helper access to packet: test6, cls valid packet_ptr range",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 5),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_2),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_update_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test7, cls unchecked packet_ptr",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test8, cls variable add",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-			offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-			offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 10),
-	BPF_LDX_MEM(BPF_B, BPF_REG_5, BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_5),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_5, BPF_REG_3, 4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 11 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test9, cls packet_ptr with bad range",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 4),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 7 },
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test10, cls packet_ptr with too short range",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 1),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 6 },
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test11, cls unsuitable helper 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_7, 4),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 42),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_store_bytes),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "helper access to the packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test12, cls unsuitable helper 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_7, 3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "helper access to the packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test13, cls helper ok",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test14, cls helper ok sub",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 4),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test15, cls helper fail sub",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 12),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test16, cls helper fail range 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test17, cls helper fail range 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, -9),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R2 min value is negative",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test18, cls helper fail range 3",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, ~0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R2 min value is negative",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test19, cls helper range zero",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test20, pkt end as input",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R1 type=pkt_end expected=fp",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"helper access to packet: test21, wrong reg",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_7, 6),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_csum_diff),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-- 
cgit v1.2.3-70-g09d2


From 77aa2563cb44a6241990cf4f082b55ee6f0a0623 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:03 +0200
Subject: selftests/bpf: verifier/helper_restricted.c converted to inline
 assembly

Test verifier/helper_restricted.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-23-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_helper_restricted.c         | 279 +++++++++++++++++++++
 .../selftests/bpf/verifier/helper_restricted.c     | 196 ---------------
 3 files changed, 281 insertions(+), 196 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_helper_restricted.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/helper_restricted.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 1cd162daf150..02983d1de218 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -19,6 +19,7 @@
 #include "verifier_div_overflow.skel.h"
 #include "verifier_helper_access_var_len.skel.h"
 #include "verifier_helper_packet_access.skel.h"
+#include "verifier_helper_restricted.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -60,3 +61,4 @@ void test_verifier_div0(void)                 { RUN(verifier_div0); }
 void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
 void test_verifier_helper_access_var_len(void) { RUN(verifier_helper_access_var_len); }
 void test_verifier_helper_packet_access(void) { RUN(verifier_helper_packet_access); }
+void test_verifier_helper_restricted(void)    { RUN(verifier_helper_restricted); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_helper_restricted.c b/tools/testing/selftests/bpf/progs/verifier_helper_restricted.c
new file mode 100644
index 000000000000..0ede0ccd090c
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_helper_restricted.c
@@ -0,0 +1,279 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/helper_restricted.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct val {
+	int cnt;
+	struct bpf_spin_lock l;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct val);
+} map_spin_lock SEC(".maps");
+
+struct timer {
+	struct bpf_timer t;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct timer);
+} map_timer SEC(".maps");
+
+SEC("kprobe")
+__description("bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_KPROBE")
+__failure __msg("unknown func bpf_ktime_get_coarse_ns")
+__naked void in_bpf_prog_type_kprobe_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_coarse_ns];		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_coarse_ns)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_TRACEPOINT")
+__failure __msg("unknown func bpf_ktime_get_coarse_ns")
+__naked void in_bpf_prog_type_tracepoint_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_coarse_ns];		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_coarse_ns)
+	: __clobber_all);
+}
+
+SEC("perf_event")
+__description("bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_PERF_EVENT")
+__failure __msg("unknown func bpf_ktime_get_coarse_ns")
+__naked void bpf_prog_type_perf_event_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_coarse_ns];		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_coarse_ns)
+	: __clobber_all);
+}
+
+SEC("raw_tracepoint")
+__description("bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_RAW_TRACEPOINT")
+__failure __msg("unknown func bpf_ktime_get_coarse_ns")
+__naked void bpf_prog_type_raw_tracepoint_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_coarse_ns];		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_coarse_ns)
+	: __clobber_all);
+}
+
+SEC("kprobe")
+__description("bpf_timer_init isn restricted in BPF_PROG_TYPE_KPROBE")
+__failure __msg("tracing progs cannot use bpf_timer yet")
+__naked void in_bpf_prog_type_kprobe_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_timer] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[map_timer] ll;				\
+	r3 = 1;						\
+l0_%=:	call %[bpf_timer_init];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_timer_init),
+	  __imm_addr(map_timer)
+	: __clobber_all);
+}
+
+SEC("perf_event")
+__description("bpf_timer_init is forbidden in BPF_PROG_TYPE_PERF_EVENT")
+__failure __msg("tracing progs cannot use bpf_timer yet")
+__naked void bpf_prog_type_perf_event_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_timer] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[map_timer] ll;				\
+	r3 = 1;						\
+l0_%=:	call %[bpf_timer_init];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_timer_init),
+	  __imm_addr(map_timer)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bpf_timer_init is forbidden in BPF_PROG_TYPE_TRACEPOINT")
+__failure __msg("tracing progs cannot use bpf_timer yet")
+__naked void in_bpf_prog_type_tracepoint_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_timer] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[map_timer] ll;				\
+	r3 = 1;						\
+l0_%=:	call %[bpf_timer_init];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_timer_init),
+	  __imm_addr(map_timer)
+	: __clobber_all);
+}
+
+SEC("raw_tracepoint")
+__description("bpf_timer_init is forbidden in BPF_PROG_TYPE_RAW_TRACEPOINT")
+__failure __msg("tracing progs cannot use bpf_timer yet")
+__naked void bpf_prog_type_raw_tracepoint_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_timer] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[map_timer] ll;				\
+	r3 = 1;						\
+l0_%=:	call %[bpf_timer_init];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_timer_init),
+	  __imm_addr(map_timer)
+	: __clobber_all);
+}
+
+SEC("kprobe")
+__description("bpf_spin_lock is forbidden in BPF_PROG_TYPE_KPROBE")
+__failure __msg("tracing progs cannot use bpf_spin_lock yet")
+__naked void in_bpf_prog_type_kprobe_3(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	call %[bpf_spin_lock];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bpf_spin_lock is forbidden in BPF_PROG_TYPE_TRACEPOINT")
+__failure __msg("tracing progs cannot use bpf_spin_lock yet")
+__naked void in_bpf_prog_type_tracepoint_3(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	call %[bpf_spin_lock];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("perf_event")
+__description("bpf_spin_lock is forbidden in BPF_PROG_TYPE_PERF_EVENT")
+__failure __msg("tracing progs cannot use bpf_spin_lock yet")
+__naked void bpf_prog_type_perf_event_3(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	call %[bpf_spin_lock];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("raw_tracepoint")
+__description("bpf_spin_lock is forbidden in BPF_PROG_TYPE_RAW_TRACEPOINT")
+__failure __msg("tracing progs cannot use bpf_spin_lock yet")
+__naked void bpf_prog_type_raw_tracepoint_3(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	call %[bpf_spin_lock];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/helper_restricted.c b/tools/testing/selftests/bpf/verifier/helper_restricted.c
deleted file mode 100644
index a067b7098b97..000000000000
--- a/tools/testing/selftests/bpf/verifier/helper_restricted.c
+++ /dev/null
@@ -1,196 +0,0 @@
-{
-	"bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_KPROBE",
-	.insns = {
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ktime_get_coarse_ns),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.errstr = "unknown func bpf_ktime_get_coarse_ns",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_KPROBE,
-},
-{
-	"bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_TRACEPOINT",
-	.insns = {
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ktime_get_coarse_ns),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.errstr = "unknown func bpf_ktime_get_coarse_ns",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_PERF_EVENT",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ktime_get_coarse_ns),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "unknown func bpf_ktime_get_coarse_ns",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_PERF_EVENT,
-},
-{
-	"bpf_ktime_get_coarse_ns is forbidden in BPF_PROG_TYPE_RAW_TRACEPOINT",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ktime_get_coarse_ns),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "unknown func bpf_ktime_get_coarse_ns",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT,
-},
-{
-	"bpf_timer_init isn restricted in BPF_PROG_TYPE_KPROBE",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_EMIT_CALL(BPF_FUNC_timer_init),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_timer = { 3, 8 },
-	.errstr = "tracing progs cannot use bpf_timer yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_KPROBE,
-},
-{
-	"bpf_timer_init is forbidden in BPF_PROG_TYPE_PERF_EVENT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_EMIT_CALL(BPF_FUNC_timer_init),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_timer = { 3, 8 },
-	.errstr = "tracing progs cannot use bpf_timer yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_PERF_EVENT,
-},
-{
-	"bpf_timer_init is forbidden in BPF_PROG_TYPE_TRACEPOINT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_EMIT_CALL(BPF_FUNC_timer_init),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_timer = { 3, 8 },
-	.errstr = "tracing progs cannot use bpf_timer yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bpf_timer_init is forbidden in BPF_PROG_TYPE_RAW_TRACEPOINT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_EMIT_CALL(BPF_FUNC_timer_init),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_timer = { 3, 8 },
-	.errstr = "tracing progs cannot use bpf_timer yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT,
-},
-{
-	"bpf_spin_lock is forbidden in BPF_PROG_TYPE_KPROBE",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_spin_lock),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.errstr = "tracing progs cannot use bpf_spin_lock yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_KPROBE,
-},
-{
-	"bpf_spin_lock is forbidden in BPF_PROG_TYPE_TRACEPOINT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_spin_lock),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.errstr = "tracing progs cannot use bpf_spin_lock yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bpf_spin_lock is forbidden in BPF_PROG_TYPE_PERF_EVENT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_spin_lock),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.errstr = "tracing progs cannot use bpf_spin_lock yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_PERF_EVENT,
-},
-{
-	"bpf_spin_lock is forbidden in BPF_PROG_TYPE_RAW_TRACEPOINT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_spin_lock),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.errstr = "tracing progs cannot use bpf_spin_lock yet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT,
-},
-- 
cgit v1.2.3-70-g09d2


From ecc424827b775860119f5a5e2c521d7485bcc74f Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:04 +0200
Subject: selftests/bpf: verifier/helper_value_access.c converted to inline
 assembly

Test verifier/helper_value_access.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-24-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |    2 +
 .../bpf/progs/verifier_helper_value_access.c       | 1245 ++++++++++++++++++++
 .../selftests/bpf/verifier/helper_value_access.c   |  953 ---------------
 3 files changed, 1247 insertions(+), 953 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_helper_value_access.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/helper_value_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 02983d1de218..2c3745a1fdcb 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -20,6 +20,7 @@
 #include "verifier_helper_access_var_len.skel.h"
 #include "verifier_helper_packet_access.skel.h"
 #include "verifier_helper_restricted.skel.h"
+#include "verifier_helper_value_access.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -62,3 +63,4 @@ void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
 void test_verifier_helper_access_var_len(void) { RUN(verifier_helper_access_var_len); }
 void test_verifier_helper_packet_access(void) { RUN(verifier_helper_packet_access); }
 void test_verifier_helper_restricted(void)    { RUN(verifier_helper_restricted); }
+void test_verifier_helper_value_access(void)  { RUN(verifier_helper_value_access); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_helper_value_access.c b/tools/testing/selftests/bpf/progs/verifier_helper_value_access.c
new file mode 100644
index 000000000000..692216c0ad3d
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_helper_value_access.c
@@ -0,0 +1,1245 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/helper_value_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct other_val {
+	long long foo;
+	long long bar;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct other_val);
+} map_hash_16b SEC(".maps");
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("tracepoint")
+__description("helper access to map: full range")
+__success
+__naked void access_to_map_full_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[sizeof_test_val];			\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(sizeof_test_val, sizeof(struct test_val))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: partial range")
+__success
+__naked void access_to_map_partial_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: empty range")
+__failure __msg("invalid access to map value, value_size=48 off=0 size=0")
+__naked void access_to_map_empty_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = 0;						\
+	call %[bpf_trace_printk];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_trace_printk),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: out-of-bound range")
+__failure __msg("invalid access to map value, value_size=48 off=0 size=56")
+__naked void map_out_of_bound_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) + 8)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: negative range")
+__failure __msg("R2 min value is negative")
+__naked void access_to_map_negative_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r2 = -8;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const imm): full range")
+__success
+__naked void via_const_imm_full_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += %[test_val_foo];				\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - offsetof(struct test_val, foo)),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const imm): partial range")
+__success
+__naked void via_const_imm_partial_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += %[test_val_foo];				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const imm): empty range")
+__failure __msg("invalid access to map value, value_size=48 off=4 size=0")
+__naked void via_const_imm_empty_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += %[test_val_foo];				\
+	r2 = 0;						\
+	call %[bpf_trace_printk];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_trace_printk),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const imm): out-of-bound range")
+__failure __msg("invalid access to map value, value_size=48 off=4 size=52")
+__naked void imm_out_of_bound_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += %[test_val_foo];				\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - offsetof(struct test_val, foo) + 8),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const imm): negative range (> adjustment)")
+__failure __msg("R2 min value is negative")
+__naked void const_imm_negative_range_adjustment_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += %[test_val_foo];				\
+	r2 = -8;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const imm): negative range (< adjustment)")
+__failure __msg("R2 min value is negative")
+__naked void const_imm_negative_range_adjustment_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r1 += %[test_val_foo];				\
+	r2 = -1;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const reg): full range")
+__success
+__naked void via_const_reg_full_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = %[test_val_foo];				\
+	r1 += r3;					\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - offsetof(struct test_val, foo)),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const reg): partial range")
+__success
+__naked void via_const_reg_partial_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = %[test_val_foo];				\
+	r1 += r3;					\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const reg): empty range")
+__failure __msg("R1 min value is outside of the allowed memory range")
+__naked void via_const_reg_empty_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = 0;						\
+	r1 += r3;					\
+	r2 = 0;						\
+	call %[bpf_trace_printk];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_trace_printk),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const reg): out-of-bound range")
+__failure __msg("invalid access to map value, value_size=48 off=4 size=52")
+__naked void reg_out_of_bound_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = %[test_val_foo];				\
+	r1 += r3;					\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - offsetof(struct test_val, foo) + 8),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const reg): negative range (> adjustment)")
+__failure __msg("R2 min value is negative")
+__naked void const_reg_negative_range_adjustment_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = %[test_val_foo];				\
+	r1 += r3;					\
+	r2 = -8;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via const reg): negative range (< adjustment)")
+__failure __msg("R2 min value is negative")
+__naked void const_reg_negative_range_adjustment_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = %[test_val_foo];				\
+	r1 += r3;					\
+	r2 = -1;					\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via variable): full range")
+__success
+__naked void map_via_variable_full_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 > %[test_val_foo] goto l0_%=;		\
+	r1 += r3;					\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - offsetof(struct test_val, foo)),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via variable): partial range")
+__success
+__naked void map_via_variable_partial_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 > %[test_val_foo] goto l0_%=;		\
+	r1 += r3;					\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via variable): empty range")
+__failure __msg("R1 min value is outside of the allowed memory range")
+__naked void map_via_variable_empty_range(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 > %[test_val_foo] goto l0_%=;		\
+	r1 += r3;					\
+	r2 = 0;						\
+	call %[bpf_trace_printk];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_trace_printk),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via variable): no max check")
+__failure __msg("R1 unbounded memory access")
+__naked void via_variable_no_max_check_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	r1 += r3;					\
+	r2 = 1;						\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to adjusted map (via variable): wrong max check")
+__failure __msg("invalid access to map value, value_size=48 off=4 size=45")
+__naked void via_variable_wrong_max_check_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 > %[test_val_foo] goto l0_%=;		\
+	r1 += r3;					\
+	r2 = %[__imm_0];				\
+	r3 = 0;						\
+	call %[bpf_probe_read_kernel];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_probe_read_kernel),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) - offsetof(struct test_val, foo) + 1),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using <, good access")
+__success
+__naked void bounds_check_using_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 < 32 goto l1_%=;				\
+	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using <, bad access")
+__failure __msg("R1 unbounded memory access")
+__naked void bounds_check_using_bad_access_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 < 32 goto l1_%=;				\
+	r1 += r3;					\
+l0_%=:	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using <=, good access")
+__success
+__naked void bounds_check_using_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 <= 32 goto l1_%=;				\
+	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using <=, bad access")
+__failure __msg("R1 unbounded memory access")
+__naked void bounds_check_using_bad_access_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 <= 32 goto l1_%=;				\
+	r1 += r3;					\
+l0_%=:	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using s<, good access")
+__success
+__naked void check_using_s_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 s< 32 goto l1_%=;				\
+l2_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	if r3 s< 0 goto l2_%=;				\
+	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using s<, good access 2")
+__success
+__naked void using_s_good_access_2_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 s< 32 goto l1_%=;				\
+l2_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	if r3 s< -3 goto l2_%=;				\
+	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using s<, bad access")
+__failure __msg("R1 min value is negative")
+__naked void check_using_s_bad_access_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u64*)(r0 + 0);				\
+	if r3 s< 32 goto l1_%=;				\
+l2_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	if r3 s< -3 goto l2_%=;				\
+	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using s<=, good access")
+__success
+__naked void check_using_s_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 s<= 32 goto l1_%=;			\
+l2_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	if r3 s<= 0 goto l2_%=;				\
+	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using s<=, good access 2")
+__success
+__naked void using_s_good_access_2_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 s<= 32 goto l1_%=;			\
+l2_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	if r3 s<= -3 goto l2_%=;			\
+	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("helper access to map: bounds check using s<=, bad access")
+__failure __msg("R1 min value is negative")
+__naked void check_using_s_bad_access_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	r3 = *(u64*)(r0 + 0);				\
+	if r3 s<= 32 goto l1_%=;			\
+l2_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+l1_%=:	if r3 s<= -3 goto l2_%=;			\
+	r1 += r3;					\
+	r0 = 0;						\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map lookup helper access to map")
+__success
+__naked void lookup_helper_access_to_map(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map update helper access to map")
+__success
+__naked void update_helper_access_to_map(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r4 = 0;						\
+	r3 = r0;					\
+	r2 = r0;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_update_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_map_update_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map update helper access to map: wrong size")
+__failure __msg("invalid access to map value, value_size=8 off=0 size=16")
+__naked void access_to_map_wrong_size(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r4 = 0;						\
+	r3 = r0;					\
+	r2 = r0;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_update_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_map_update_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via const imm)")
+__success
+__naked void adjusted_map_via_const_imm(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r2 += %[other_val_bar];				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(other_val_bar, offsetof(struct other_val, bar))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via const imm): out-of-bound 1")
+__failure __msg("invalid access to map value, value_size=16 off=12 size=8")
+__naked void imm_out_of_bound_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r2 += %[__imm_0];				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__imm_0, sizeof(struct other_val) - 4)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via const imm): out-of-bound 2")
+__failure __msg("invalid access to map value, value_size=16 off=-4 size=8")
+__naked void imm_out_of_bound_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r2 += -4;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via const reg)")
+__success
+__naked void adjusted_map_via_const_reg(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r3 = %[other_val_bar];				\
+	r2 += r3;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(other_val_bar, offsetof(struct other_val, bar))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via const reg): out-of-bound 1")
+__failure __msg("invalid access to map value, value_size=16 off=12 size=8")
+__naked void reg_out_of_bound_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r3 = %[__imm_0];				\
+	r2 += r3;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__imm_0, sizeof(struct other_val) - 4)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via const reg): out-of-bound 2")
+__failure __msg("invalid access to map value, value_size=16 off=-4 size=8")
+__naked void reg_out_of_bound_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r3 = -4;					\
+	r2 += r3;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via variable)")
+__success
+__naked void to_adjusted_map_via_variable(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 > %[other_val_bar] goto l0_%=;		\
+	r2 += r3;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(other_val_bar, offsetof(struct other_val, bar))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via variable): no max check")
+__failure
+__msg("R2 unbounded memory access, make sure to bounds check any such access")
+__naked void via_variable_no_max_check_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	r2 += r3;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("map helper access to adjusted map (via variable): wrong max check")
+__failure __msg("invalid access to map value, value_size=16 off=9 size=8")
+__naked void via_variable_wrong_max_check_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = r0;					\
+	r3 = *(u32*)(r0 + 0);				\
+	if r3 > %[__imm_0] goto l0_%=;			\
+	r2 += r3;					\
+	r1 = %[map_hash_16b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__imm_0, offsetof(struct other_val, bar) + 1)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/helper_value_access.c b/tools/testing/selftests/bpf/verifier/helper_value_access.c
deleted file mode 100644
index 1c7882ddfa63..000000000000
--- a/tools/testing/selftests/bpf/verifier/helper_value_access.c
+++ /dev/null
@@ -1,953 +0,0 @@
-{
-	"helper access to map: full range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, sizeof(struct test_val)),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: partial range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: empty range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_trace_printk),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=0 size=0",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: out-of-bound range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, sizeof(struct test_val) + 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=0 size=56",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: negative range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R2 min value is negative",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const imm): full range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_2,
-		      sizeof(struct test_val) -	offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const imm): partial range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const imm): empty range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_trace_printk),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=4 size=0",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const imm): out-of-bound range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_2,
-		      sizeof(struct test_val) - offsetof(struct test_val, foo) + 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=4 size=52",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const imm): negative range (> adjustment)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R2 min value is negative",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const imm): negative range (< adjustment)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R2 min value is negative",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const reg): full range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, offsetof(struct test_val, foo)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2,
-		      sizeof(struct test_val) - offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const reg): partial range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, offsetof(struct test_val, foo)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const reg): empty range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_trace_printk),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R1 min value is outside of the allowed memory range",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const reg): out-of-bound range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, offsetof(struct test_val, foo)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2,
-		      sizeof(struct test_val) -
-		      offsetof(struct test_val, foo) + 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=4 size=52",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const reg): negative range (> adjustment)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, offsetof(struct test_val, foo)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R2 min value is negative",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via const reg): negative range (< adjustment)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, offsetof(struct test_val, foo)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R2 min value is negative",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via variable): full range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, offsetof(struct test_val, foo), 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2,
-		      sizeof(struct test_val) - offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via variable): partial range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, offsetof(struct test_val, foo), 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via variable): empty range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, offsetof(struct test_val, foo), 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_trace_printk),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R1 min value is outside of the allowed memory range",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via variable): no max check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2, 1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R1 unbounded memory access",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to adjusted map (via variable): wrong max check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, offsetof(struct test_val, foo), 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_MOV64_IMM(BPF_REG_2,
-		      sizeof(struct test_val) -
-		      offsetof(struct test_val, foo) + 1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_probe_read_kernel),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "invalid access to map value, value_size=48 off=4 size=45",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using <, good access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using <, bad access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_3, 32, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 unbounded memory access",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using <=, good access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JLE, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using <=, bad access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JLE, BPF_REG_3, 32, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 unbounded memory access",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using s<, good access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_3, 0, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using s<, good access 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_3, -3, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using s<, bad access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_3, -3, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 min value is negative",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using s<=, good access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLE, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSLE, BPF_REG_3, 0, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using s<=, good access 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLE, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSLE, BPF_REG_3, -3, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"helper access to map: bounds check using s<=, bad access",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLE, BPF_REG_3, 32, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSLE, BPF_REG_3, -3, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 min value is negative",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map lookup helper access to map",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 8 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map update helper access to map",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_update_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 10 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map update helper access to map: wrong size",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_update_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.fixup_map_hash_16b = { 10 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=8 off=0 size=16",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via const imm)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, offsetof(struct other_val, bar)),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 9 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via const imm): out-of-bound 1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, sizeof(struct other_val) - 4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 9 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=16 off=12 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via const imm): out-of-bound 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 9 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=16 off=-4 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via const reg)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, offsetof(struct other_val, bar)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 10 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via const reg): out-of-bound 1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, sizeof(struct other_val) - 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 10 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=16 off=12 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via const reg): out-of-bound 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, -4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 10 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=16 off=-4 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via variable)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, offsetof(struct other_val, bar), 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 11 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via variable): no max check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 10 },
-	.result = REJECT,
-	.errstr = "R2 unbounded memory access, make sure to bounds check any such access",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"map helper access to adjusted map (via variable): wrong max check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, offsetof(struct other_val, bar) + 1, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 3, 11 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=16 off=9 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-- 
cgit v1.2.3-70-g09d2


From 01481e67dd4d1c2c62eb6a506a5f4803ee50f8a6 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:05 +0200
Subject: selftests/bpf: verifier/int_ptr.c converted to inline assembly

Test verifier/int_ptr.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-25-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_int_ptr.c | 157 ++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/int_ptr.c     | 161 ---------------------
 3 files changed, 159 insertions(+), 161 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_int_ptr.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/int_ptr.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 2c3745a1fdcb..d9180da30f1b 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -21,6 +21,7 @@
 #include "verifier_helper_packet_access.skel.h"
 #include "verifier_helper_restricted.skel.h"
 #include "verifier_helper_value_access.skel.h"
+#include "verifier_int_ptr.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -64,3 +65,4 @@ void test_verifier_helper_access_var_len(void) { RUN(verifier_helper_access_var_
 void test_verifier_helper_packet_access(void) { RUN(verifier_helper_packet_access); }
 void test_verifier_helper_restricted(void)    { RUN(verifier_helper_restricted); }
 void test_verifier_helper_value_access(void)  { RUN(verifier_helper_value_access); }
+void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_int_ptr.c b/tools/testing/selftests/bpf/progs/verifier_int_ptr.c
new file mode 100644
index 000000000000..b054f9c48143
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_int_ptr.c
@@ -0,0 +1,157 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/int_ptr.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("cgroup/sysctl")
+__description("ARG_PTR_TO_LONG uninitialized")
+__failure __msg("invalid indirect read from stack R4 off -16+0 size 8")
+__naked void arg_ptr_to_long_uninitialized(void)
+{
+	asm volatile ("					\
+	/* bpf_strtoul arg1 (buf) */			\
+	r7 = r10;					\
+	r7 += -8;					\
+	r0 = 0x00303036;				\
+	*(u64*)(r7 + 0) = r0;				\
+	r1 = r7;					\
+	/* bpf_strtoul arg2 (buf_len) */		\
+	r2 = 4;						\
+	/* bpf_strtoul arg3 (flags) */			\
+	r3 = 0;						\
+	/* bpf_strtoul arg4 (res) */			\
+	r7 += -8;					\
+	r4 = r7;					\
+	/* bpf_strtoul() */				\
+	call %[bpf_strtoul];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_strtoul)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ARG_PTR_TO_LONG half-uninitialized")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid indirect read from stack R4 off -16+4 size 8")
+__retval(0)
+__naked void ptr_to_long_half_uninitialized(void)
+{
+	asm volatile ("					\
+	/* bpf_strtoul arg1 (buf) */			\
+	r7 = r10;					\
+	r7 += -8;					\
+	r0 = 0x00303036;				\
+	*(u64*)(r7 + 0) = r0;				\
+	r1 = r7;					\
+	/* bpf_strtoul arg2 (buf_len) */		\
+	r2 = 4;						\
+	/* bpf_strtoul arg3 (flags) */			\
+	r3 = 0;						\
+	/* bpf_strtoul arg4 (res) */			\
+	r7 += -8;					\
+	*(u32*)(r7 + 0) = r0;				\
+	r4 = r7;					\
+	/* bpf_strtoul() */				\
+	call %[bpf_strtoul];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_strtoul)
+	: __clobber_all);
+}
+
+SEC("cgroup/sysctl")
+__description("ARG_PTR_TO_LONG misaligned")
+__failure __msg("misaligned stack access off (0x0; 0x0)+-20+0 size 8")
+__naked void arg_ptr_to_long_misaligned(void)
+{
+	asm volatile ("					\
+	/* bpf_strtoul arg1 (buf) */			\
+	r7 = r10;					\
+	r7 += -8;					\
+	r0 = 0x00303036;				\
+	*(u64*)(r7 + 0) = r0;				\
+	r1 = r7;					\
+	/* bpf_strtoul arg2 (buf_len) */		\
+	r2 = 4;						\
+	/* bpf_strtoul arg3 (flags) */			\
+	r3 = 0;						\
+	/* bpf_strtoul arg4 (res) */			\
+	r7 += -12;					\
+	r0 = 0;						\
+	*(u32*)(r7 + 0) = r0;				\
+	*(u64*)(r7 + 4) = r0;				\
+	r4 = r7;					\
+	/* bpf_strtoul() */				\
+	call %[bpf_strtoul];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_strtoul)
+	: __clobber_all);
+}
+
+SEC("cgroup/sysctl")
+__description("ARG_PTR_TO_LONG size < sizeof(long)")
+__failure __msg("invalid indirect access to stack R4 off=-4 size=8")
+__naked void to_long_size_sizeof_long(void)
+{
+	asm volatile ("					\
+	/* bpf_strtoul arg1 (buf) */			\
+	r7 = r10;					\
+	r7 += -16;					\
+	r0 = 0x00303036;				\
+	*(u64*)(r7 + 0) = r0;				\
+	r1 = r7;					\
+	/* bpf_strtoul arg2 (buf_len) */		\
+	r2 = 4;						\
+	/* bpf_strtoul arg3 (flags) */			\
+	r3 = 0;						\
+	/* bpf_strtoul arg4 (res) */			\
+	r7 += 12;					\
+	*(u32*)(r7 + 0) = r0;				\
+	r4 = r7;					\
+	/* bpf_strtoul() */				\
+	call %[bpf_strtoul];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_strtoul)
+	: __clobber_all);
+}
+
+SEC("cgroup/sysctl")
+__description("ARG_PTR_TO_LONG initialized")
+__success
+__naked void arg_ptr_to_long_initialized(void)
+{
+	asm volatile ("					\
+	/* bpf_strtoul arg1 (buf) */			\
+	r7 = r10;					\
+	r7 += -8;					\
+	r0 = 0x00303036;				\
+	*(u64*)(r7 + 0) = r0;				\
+	r1 = r7;					\
+	/* bpf_strtoul arg2 (buf_len) */		\
+	r2 = 4;						\
+	/* bpf_strtoul arg3 (flags) */			\
+	r3 = 0;						\
+	/* bpf_strtoul arg4 (res) */			\
+	r7 += -8;					\
+	*(u64*)(r7 + 0) = r0;				\
+	r4 = r7;					\
+	/* bpf_strtoul() */				\
+	call %[bpf_strtoul];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_strtoul)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/int_ptr.c b/tools/testing/selftests/bpf/verifier/int_ptr.c
deleted file mode 100644
index 02d9e004260b..000000000000
--- a/tools/testing/selftests/bpf/verifier/int_ptr.c
+++ /dev/null
@@ -1,161 +0,0 @@
-{
-	"ARG_PTR_TO_LONG uninitialized",
-	.insns = {
-		/* bpf_strtoul arg1 (buf) */
-		BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_MOV64_IMM(BPF_REG_0, 0x00303036),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-
-		BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-
-		/* bpf_strtoul arg2 (buf_len) */
-		BPF_MOV64_IMM(BPF_REG_2, 4),
-
-		/* bpf_strtoul arg3 (flags) */
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-
-		/* bpf_strtoul arg4 (res) */
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_MOV64_REG(BPF_REG_4, BPF_REG_7),
-
-		/* bpf_strtoul() */
-		BPF_EMIT_CALL(BPF_FUNC_strtoul),
-
-		BPF_MOV64_IMM(BPF_REG_0, 1),
-		BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SYSCTL,
-	.errstr = "invalid indirect read from stack R4 off -16+0 size 8",
-},
-{
-	"ARG_PTR_TO_LONG half-uninitialized",
-	.insns = {
-		/* bpf_strtoul arg1 (buf) */
-		BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_MOV64_IMM(BPF_REG_0, 0x00303036),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-
-		BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-
-		/* bpf_strtoul arg2 (buf_len) */
-		BPF_MOV64_IMM(BPF_REG_2, 4),
-
-		/* bpf_strtoul arg3 (flags) */
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-
-		/* bpf_strtoul arg4 (res) */
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_STX_MEM(BPF_W, BPF_REG_7, BPF_REG_0, 0),
-		BPF_MOV64_REG(BPF_REG_4, BPF_REG_7),
-
-		/* bpf_strtoul() */
-		BPF_EMIT_CALL(BPF_FUNC_strtoul),
-
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "invalid indirect read from stack R4 off -16+4 size 8",
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-{
-	"ARG_PTR_TO_LONG misaligned",
-	.insns = {
-		/* bpf_strtoul arg1 (buf) */
-		BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_MOV64_IMM(BPF_REG_0, 0x00303036),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-
-		BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-
-		/* bpf_strtoul arg2 (buf_len) */
-		BPF_MOV64_IMM(BPF_REG_2, 4),
-
-		/* bpf_strtoul arg3 (flags) */
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-
-		/* bpf_strtoul arg4 (res) */
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -12),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_STX_MEM(BPF_W, BPF_REG_7, BPF_REG_0, 0),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 4),
-		BPF_MOV64_REG(BPF_REG_4, BPF_REG_7),
-
-		/* bpf_strtoul() */
-		BPF_EMIT_CALL(BPF_FUNC_strtoul),
-
-		BPF_MOV64_IMM(BPF_REG_0, 1),
-		BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SYSCTL,
-	.errstr = "misaligned stack access off (0x0; 0x0)+-20+0 size 8",
-},
-{
-	"ARG_PTR_TO_LONG size < sizeof(long)",
-	.insns = {
-		/* bpf_strtoul arg1 (buf) */
-		BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -16),
-		BPF_MOV64_IMM(BPF_REG_0, 0x00303036),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-
-		BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-
-		/* bpf_strtoul arg2 (buf_len) */
-		BPF_MOV64_IMM(BPF_REG_2, 4),
-
-		/* bpf_strtoul arg3 (flags) */
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-
-		/* bpf_strtoul arg4 (res) */
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, 12),
-		BPF_STX_MEM(BPF_W, BPF_REG_7, BPF_REG_0, 0),
-		BPF_MOV64_REG(BPF_REG_4, BPF_REG_7),
-
-		/* bpf_strtoul() */
-		BPF_EMIT_CALL(BPF_FUNC_strtoul),
-
-		BPF_MOV64_IMM(BPF_REG_0, 1),
-		BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SYSCTL,
-	.errstr = "invalid indirect access to stack R4 off=-4 size=8",
-},
-{
-	"ARG_PTR_TO_LONG initialized",
-	.insns = {
-		/* bpf_strtoul arg1 (buf) */
-		BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_MOV64_IMM(BPF_REG_0, 0x00303036),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-
-		BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-
-		/* bpf_strtoul arg2 (buf_len) */
-		BPF_MOV64_IMM(BPF_REG_2, 4),
-
-		/* bpf_strtoul arg3 (flags) */
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-
-		/* bpf_strtoul arg4 (res) */
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -8),
-		BPF_STX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-		BPF_MOV64_REG(BPF_REG_4, BPF_REG_7),
-
-		/* bpf_strtoul() */
-		BPF_EMIT_CALL(BPF_FUNC_strtoul),
-
-		BPF_MOV64_IMM(BPF_REG_0, 1),
-		BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SYSCTL,
-},
-- 
cgit v1.2.3-70-g09d2


From e297875580662f2fdcde1cba4ea28bf70ea2ca49 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:06 +0200
Subject: selftests/bpf: verifier/ld_ind.c converted to inline assembly

Test verifier/ld_ind.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-26-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_ld_ind.c  | 110 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/ld_ind.c      |  72 --------------
 3 files changed, 112 insertions(+), 72 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_ld_ind.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/ld_ind.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index d9180da30f1b..d8d4464b6112 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -22,6 +22,7 @@
 #include "verifier_helper_restricted.skel.h"
 #include "verifier_helper_value_access.skel.h"
 #include "verifier_int_ptr.skel.h"
+#include "verifier_ld_ind.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -66,3 +67,4 @@ void test_verifier_helper_packet_access(void) { RUN(verifier_helper_packet_acces
 void test_verifier_helper_restricted(void)    { RUN(verifier_helper_restricted); }
 void test_verifier_helper_value_access(void)  { RUN(verifier_helper_value_access); }
 void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
+void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_ld_ind.c b/tools/testing/selftests/bpf/progs/verifier_ld_ind.c
new file mode 100644
index 000000000000..c925ba9a2e74
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_ld_ind.c
@@ -0,0 +1,110 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/ld_ind.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "../../../include/linux/filter.h"
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("ld_ind: check calling conv, r1")
+__failure __msg("R1 !read_ok")
+__failure_unpriv
+__naked void ind_check_calling_conv_r1(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_1, -0x200000))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ld_ind: check calling conv, r2")
+__failure __msg("R2 !read_ok")
+__failure_unpriv
+__naked void ind_check_calling_conv_r2(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r2 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r2;					\
+	exit;						\
+"	:
+	: __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_2, -0x200000))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ld_ind: check calling conv, r3")
+__failure __msg("R3 !read_ok")
+__failure_unpriv
+__naked void ind_check_calling_conv_r3(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r3 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r3;					\
+	exit;						\
+"	:
+	: __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_3, -0x200000))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ld_ind: check calling conv, r4")
+__failure __msg("R4 !read_ok")
+__failure_unpriv
+__naked void ind_check_calling_conv_r4(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r4 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r4;					\
+	exit;						\
+"	:
+	: __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_4, -0x200000))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ld_ind: check calling conv, r5")
+__failure __msg("R5 !read_ok")
+__failure_unpriv
+__naked void ind_check_calling_conv_r5(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r5 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r5;					\
+	exit;						\
+"	:
+	: __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_5, -0x200000))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ld_ind: check calling conv, r7")
+__success __success_unpriv __retval(1)
+__naked void ind_check_calling_conv_r7(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r7 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r7;					\
+	exit;						\
+"	:
+	: __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/ld_ind.c b/tools/testing/selftests/bpf/verifier/ld_ind.c
deleted file mode 100644
index 079734227538..000000000000
--- a/tools/testing/selftests/bpf/verifier/ld_ind.c
+++ /dev/null
@@ -1,72 +0,0 @@
-{
-	"ld_ind: check calling conv, r1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_1, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 !read_ok",
-	.result = REJECT,
-},
-{
-	"ld_ind: check calling conv, r2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_2, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_2, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R2 !read_ok",
-	.result = REJECT,
-},
-{
-	"ld_ind: check calling conv, r3",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_3, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R3 !read_ok",
-	.result = REJECT,
-},
-{
-	"ld_ind: check calling conv, r4",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_4, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_4, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_4),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R4 !read_ok",
-	.result = REJECT,
-},
-{
-	"ld_ind: check calling conv, r5",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_5, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_5, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_5),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R5 !read_ok",
-	.result = REJECT,
-},
-{
-	"ld_ind: check calling conv, r7",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_7, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_7),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 1,
-},
-- 
cgit v1.2.3-70-g09d2


From 583c7ce5be093277ef1691f35d9bec5ef9934e24 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:07 +0200
Subject: selftests/bpf: verifier/leak_ptr.c converted to inline assembly

Test verifier/leak_ptr.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-27-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../selftests/bpf/progs/verifier_leak_ptr.c        | 92 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/leak_ptr.c    | 67 ----------------
 3 files changed, 94 insertions(+), 67 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_leak_ptr.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/leak_ptr.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index d8d4464b6112..f8b3b6beba3f 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -23,6 +23,7 @@
 #include "verifier_helper_value_access.skel.h"
 #include "verifier_int_ptr.skel.h"
 #include "verifier_ld_ind.skel.h"
+#include "verifier_leak_ptr.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -68,3 +69,4 @@ void test_verifier_helper_restricted(void)    { RUN(verifier_helper_restricted);
 void test_verifier_helper_value_access(void)  { RUN(verifier_helper_value_access); }
 void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
 void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
+void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_leak_ptr.c b/tools/testing/selftests/bpf/progs/verifier_leak_ptr.c
new file mode 100644
index 000000000000..d153fbe50055
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_leak_ptr.c
@@ -0,0 +1,92 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/leak_ptr.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("socket")
+__description("leak pointer into ctx 1")
+__failure __msg("BPF_ATOMIC stores into R1 ctx is not allowed")
+__failure_unpriv __msg_unpriv("R2 leaks addr into mem")
+__naked void leak_pointer_into_ctx_1(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r2 = %[map_hash_8b] ll;				\
+	lock *(u64 *)(r1 + %[__sk_buff_cb_0]) += r2;	\
+	exit;						\
+"	:
+	: __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("leak pointer into ctx 2")
+__failure __msg("BPF_ATOMIC stores into R1 ctx is not allowed")
+__failure_unpriv __msg_unpriv("R10 leaks addr into mem")
+__naked void leak_pointer_into_ctx_2(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	lock *(u64 *)(r1 + %[__sk_buff_cb_0]) += r10;	\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("leak pointer into ctx 3")
+__success __failure_unpriv __msg_unpriv("R2 leaks addr into ctx")
+__retval(0)
+__naked void leak_pointer_into_ctx_3(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r2 = %[map_hash_8b] ll;				\
+	*(u64*)(r1 + %[__sk_buff_cb_0]) = r2;		\
+	exit;						\
+"	:
+	: __imm_addr(map_hash_8b),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("leak pointer into map val")
+__success __failure_unpriv __msg_unpriv("R6 leaks addr into mem")
+__retval(0)
+__naked void leak_pointer_into_map_val(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r3 = 0;						\
+	*(u64*)(r0 + 0) = r3;				\
+	lock *(u64 *)(r0 + 0) += r6;			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/leak_ptr.c b/tools/testing/selftests/bpf/verifier/leak_ptr.c
deleted file mode 100644
index 73f0dea95546..000000000000
--- a/tools/testing/selftests/bpf/verifier/leak_ptr.c
+++ /dev/null
@@ -1,67 +0,0 @@
-{
-	"leak pointer into ctx 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_1, BPF_REG_2,
-		      offsetof(struct __sk_buff, cb[0])),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 2 },
-	.errstr_unpriv = "R2 leaks addr into mem",
-	.result_unpriv = REJECT,
-	.result = REJECT,
-	.errstr = "BPF_ATOMIC stores into R1 ctx is not allowed",
-},
-{
-	"leak pointer into ctx 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_1, BPF_REG_10,
-		      offsetof(struct __sk_buff, cb[0])),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R10 leaks addr into mem",
-	.result_unpriv = REJECT,
-	.result = REJECT,
-	.errstr = "BPF_ATOMIC stores into R1 ctx is not allowed",
-},
-{
-	"leak pointer into ctx 3",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2,
-		      offsetof(struct __sk_buff, cb[0])),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	.errstr_unpriv = "R2 leaks addr into ctx",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"leak pointer into map val",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_0, BPF_REG_3, 0),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_0, BPF_REG_6, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.errstr_unpriv = "R6 leaks addr into mem",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-- 
cgit v1.2.3-70-g09d2


From caf345cf12073eb5905b03c3c5f905cc0964dc6c Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:08 +0200
Subject: selftests/bpf: verifier/map_ptr.c converted to inline assembly

Test verifier/map_ptr.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-28-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_map_ptr.c | 159 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/map_ptr.c     |  99 -------------
 3 files changed, 161 insertions(+), 99 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_map_ptr.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/map_ptr.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index f8b3b6beba3f..d2f3bff0e942 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -24,6 +24,7 @@
 #include "verifier_int_ptr.skel.h"
 #include "verifier_ld_ind.skel.h"
 #include "verifier_leak_ptr.skel.h"
+#include "verifier_map_ptr.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -70,3 +71,4 @@ void test_verifier_helper_value_access(void)  { RUN(verifier_helper_value_access
 void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
 void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
+void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_map_ptr.c b/tools/testing/selftests/bpf/progs/verifier_map_ptr.c
new file mode 100644
index 000000000000..11a079145966
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_map_ptr.c
@@ -0,0 +1,159 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/map_ptr.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+} map_array_48b SEC(".maps");
+
+struct other_val {
+	long long foo;
+	long long bar;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct other_val);
+} map_hash_16b SEC(".maps");
+
+SEC("socket")
+__description("bpf_map_ptr: read with negative offset rejected")
+__failure __msg("R1 is bpf_array invalid negative access: off=-8")
+__failure_unpriv
+__msg_unpriv("access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN")
+__naked void read_with_negative_offset_rejected(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 = %[map_array_48b] ll;			\
+	r6 = *(u64*)(r1 - 8);				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bpf_map_ptr: write rejected")
+__failure __msg("only read from bpf_array is supported")
+__failure_unpriv
+__msg_unpriv("access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN")
+__naked void bpf_map_ptr_write_rejected(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r0;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	*(u64*)(r1 + 0) = r2;				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bpf_map_ptr: read non-existent field rejected")
+__failure
+__msg("cannot access ptr member ops with moff 0 in struct bpf_map with off 1 size 4")
+__failure_unpriv
+__msg_unpriv("access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void read_non_existent_field_rejected(void)
+{
+	asm volatile ("					\
+	r6 = 0;						\
+	r1 = %[map_array_48b] ll;			\
+	r6 = *(u32*)(r1 + 1);				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bpf_map_ptr: read ops field accepted")
+__success __failure_unpriv
+__msg_unpriv("access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN")
+__retval(1)
+__naked void ptr_read_ops_field_accepted(void)
+{
+	asm volatile ("					\
+	r6 = 0;						\
+	r1 = %[map_array_48b] ll;			\
+	r6 = *(u64*)(r1 + 0);				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bpf_map_ptr: r = 0, map_ptr = map_ptr + r")
+__success __failure_unpriv
+__msg_unpriv("R1 has pointer with unsupported alu operation")
+__retval(0)
+__naked void map_ptr_map_ptr_r(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r0;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r0 = 0;						\
+	r1 = %[map_hash_16b] ll;			\
+	r1 += r0;					\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bpf_map_ptr: r = 0, r = r + map_ptr")
+__success __failure_unpriv
+__msg_unpriv("R0 has pointer with unsupported alu operation")
+__retval(0)
+__naked void _0_r_r_map_ptr(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r0;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	r0 = %[map_hash_16b] ll;			\
+	r1 += r0;					\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_16b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/map_ptr.c b/tools/testing/selftests/bpf/verifier/map_ptr.c
deleted file mode 100644
index 17ee84dc7766..000000000000
--- a/tools/testing/selftests/bpf/verifier/map_ptr.c
+++ /dev/null
@@ -1,99 +0,0 @@
-{
-	"bpf_map_ptr: read with negative offset rejected",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN",
-	.result = REJECT,
-	.errstr = "R1 is bpf_array invalid negative access: off=-8",
-},
-{
-	"bpf_map_ptr: write rejected",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN",
-	.result = REJECT,
-	.errstr = "only read from bpf_array is supported",
-},
-{
-	"bpf_map_ptr: read non-existent field rejected",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_6, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN",
-	.result = REJECT,
-	.errstr = "cannot access ptr member ops with moff 0 in struct bpf_map with off 1 size 4",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"bpf_map_ptr: read ops field accepted",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_6, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "access is allowed only to CAP_PERFMON and CAP_SYS_ADMIN",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"bpf_map_ptr: r = 0, map_ptr = map_ptr + r",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 4 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 has pointer with unsupported alu operation",
-	.result = ACCEPT,
-},
-{
-	"bpf_map_ptr: r = 0, r = r + map_ptr",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_LD_MAP_FD(BPF_REG_0, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 4 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 has pointer with unsupported alu operation",
-	.result = ACCEPT,
-},
-- 
cgit v1.2.3-70-g09d2


From 05e474ecbb56a5fadea6b36b7aa7d8d237aea064 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:09 +0200
Subject: selftests/bpf: verifier/map_ret_val.c converted to inline assembly

Test verifier/map_ret_val.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-29-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_map_ret_val.c     | 110 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/map_ret_val.c |  65 ------------
 3 files changed, 112 insertions(+), 65 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_map_ret_val.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/map_ret_val.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index d2f3bff0e942..5131a73fd225 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -25,6 +25,7 @@
 #include "verifier_ld_ind.skel.h"
 #include "verifier_leak_ptr.skel.h"
 #include "verifier_map_ptr.skel.h"
+#include "verifier_map_ret_val.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -72,3 +73,4 @@ void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
 void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
+void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_map_ret_val.c b/tools/testing/selftests/bpf/progs/verifier_map_ret_val.c
new file mode 100644
index 000000000000..1639628b832d
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_map_ret_val.c
@@ -0,0 +1,110 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/map_ret_val.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "../../../include/linux/filter.h"
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("socket")
+__description("invalid map_fd for function call")
+__failure __msg("fd 0 is not pointing to valid bpf_map")
+__failure_unpriv
+__naked void map_fd_for_function_call(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	.8byte %[ld_map_fd];				\
+	.8byte 0;					\
+	call %[bpf_map_delete_elem];			\
+	exit;						\
+"	:
+	: __imm(bpf_map_delete_elem),
+	  __imm_insn(ld_map_fd, BPF_RAW_INSN(BPF_LD | BPF_DW | BPF_IMM, BPF_REG_1, BPF_PSEUDO_MAP_FD, 0, 0))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("don't check return value before access")
+__failure __msg("R0 invalid mem access 'map_value_or_null'")
+__failure_unpriv
+__naked void check_return_value_before_access(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r1 = 0;						\
+	*(u64*)(r0 + 0) = r1;				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("access memory with incorrect alignment")
+__failure __msg("misaligned value access")
+__failure_unpriv
+__flag(BPF_F_STRICT_ALIGNMENT)
+__naked void access_memory_with_incorrect_alignment_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r0 + 4) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("sometimes access memory with incorrect alignment")
+__failure __msg("R0 invalid mem access")
+__msg_unpriv("R0 leaks addr")
+__flag(BPF_F_STRICT_ALIGNMENT)
+__naked void access_memory_with_incorrect_alignment_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r0 + 0) = r1;				\
+	exit;						\
+l0_%=:	r1 = 1;						\
+	*(u64*)(r0 + 0) = r1;				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/map_ret_val.c b/tools/testing/selftests/bpf/verifier/map_ret_val.c
deleted file mode 100644
index bdd0e8d18333..000000000000
--- a/tools/testing/selftests/bpf/verifier/map_ret_val.c
+++ /dev/null
@@ -1,65 +0,0 @@
-{
-	"invalid map_fd for function call",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_delete_elem),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "fd 0 is not pointing to valid bpf_map",
-	.result = REJECT,
-},
-{
-	"don't check return value before access",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 invalid mem access 'map_value_or_null'",
-	.result = REJECT,
-},
-{
-	"access memory with incorrect alignment",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 4, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "misaligned value access",
-	.result = REJECT,
-	.flags = F_LOAD_WITH_STRICT_ALIGNMENT,
-},
-{
-	"sometimes access memory with incorrect alignment",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 invalid mem access",
-	.errstr_unpriv = "R0 leaks addr",
-	.result = REJECT,
-	.flags = F_LOAD_WITH_STRICT_ALIGNMENT,
-},
-- 
cgit v1.2.3-70-g09d2


From ade3f08fc236ce8a5f00c1ffaf85ba42aa18ded4 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:10 +0200
Subject: selftests/bpf: verifier/masking.c converted to inline assembly

Test verifier/masking.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-30-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_masking.c | 410 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/masking.c     | 322 ----------------
 3 files changed, 412 insertions(+), 322 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_masking.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/masking.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 5131a73fd225..b23fcbe4f83b 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -26,6 +26,7 @@
 #include "verifier_leak_ptr.skel.h"
 #include "verifier_map_ptr.skel.h"
 #include "verifier_map_ret_val.skel.h"
+#include "verifier_masking.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -74,3 +75,4 @@ void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
+void test_verifier_masking(void)              { RUN(verifier_masking); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_masking.c b/tools/testing/selftests/bpf/progs/verifier_masking.c
new file mode 100644
index 000000000000..5732cc1b4c47
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_masking.c
@@ -0,0 +1,410 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/masking.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("masking, test out of bounds 1")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_1(void)
+{
+	asm volatile ("					\
+	w1 = 5;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 5 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 2")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_2(void)
+{
+	asm volatile ("					\
+	w1 = 1;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 3")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_3(void)
+{
+	asm volatile ("					\
+	w1 = 0xffffffff;				\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffffffff - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 4")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_4(void)
+{
+	asm volatile ("					\
+	w1 = 0xffffffff;				\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 5")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_5(void)
+{
+	asm volatile ("					\
+	w1 = -1;					\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 6")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_6(void)
+{
+	asm volatile ("					\
+	w1 = -1;					\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffffffff - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 7")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_7(void)
+{
+	asm volatile ("					\
+	r1 = 5;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 5 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 8")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_8(void)
+{
+	asm volatile ("					\
+	r1 = 1;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 9")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_9(void)
+{
+	asm volatile ("					\
+	r1 = 0xffffffff;				\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffffffff - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 10")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_10(void)
+{
+	asm volatile ("					\
+	r1 = 0xffffffff;				\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 11")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_11(void)
+{
+	asm volatile ("					\
+	r1 = -1;					\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test out of bounds 12")
+__success __success_unpriv __retval(0)
+__naked void test_out_of_bounds_12(void)
+{
+	asm volatile ("					\
+	r1 = -1;					\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffffffff - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 1")
+__success __success_unpriv __retval(4)
+__naked void masking_test_in_bounds_1(void)
+{
+	asm volatile ("					\
+	w1 = 4;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 5 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 2")
+__success __success_unpriv __retval(0)
+__naked void masking_test_in_bounds_2(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffffffff - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 3")
+__success __success_unpriv __retval(0xfffffffe)
+__naked void masking_test_in_bounds_3(void)
+{
+	asm volatile ("					\
+	w1 = 0xfffffffe;				\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffffffff - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 4")
+__success __success_unpriv __retval(0xabcde)
+__naked void masking_test_in_bounds_4(void)
+{
+	asm volatile ("					\
+	w1 = 0xabcde;					\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xabcdef - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 5")
+__success __success_unpriv __retval(0)
+__naked void masking_test_in_bounds_5(void)
+{
+	asm volatile ("					\
+	w1 = 0;						\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 1 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 6")
+__success __success_unpriv __retval(46)
+__naked void masking_test_in_bounds_6(void)
+{
+	asm volatile ("					\
+	w1 = 46;					\
+	w2 = %[__imm_0];				\
+	r2 -= r1;					\
+	r2 |= r1;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r1 &= r2;					\
+	r0 = r1;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 47 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 7")
+__success __success_unpriv __retval(46)
+__naked void masking_test_in_bounds_7(void)
+{
+	asm volatile ("					\
+	r3 = -46;					\
+	r3 *= -1;					\
+	w2 = %[__imm_0];				\
+	r2 -= r3;					\
+	r2 |= r3;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r3 &= r2;					\
+	r0 = r3;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 47 - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("masking, test in bounds 8")
+__success __success_unpriv __retval(0)
+__naked void masking_test_in_bounds_8(void)
+{
+	asm volatile ("					\
+	r3 = -47;					\
+	r3 *= -1;					\
+	w2 = %[__imm_0];				\
+	r2 -= r3;					\
+	r2 |= r3;					\
+	r2 = -r2;					\
+	r2 s>>= 63;					\
+	r3 &= r2;					\
+	r0 = r3;					\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 47 - 1)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/masking.c b/tools/testing/selftests/bpf/verifier/masking.c
deleted file mode 100644
index 6e1358c544fd..000000000000
--- a/tools/testing/selftests/bpf/verifier/masking.c
+++ /dev/null
@@ -1,322 +0,0 @@
-{
-	"masking, test out of bounds 1",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 5),
-	BPF_MOV32_IMM(BPF_REG_2, 5 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 2",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 3",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0xffffffff),
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 4",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0xffffffff),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 5",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 6",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 7",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 5),
-	BPF_MOV32_IMM(BPF_REG_2, 5 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 8",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 9",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0xffffffff),
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 10",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0xffffffff),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 11",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test out of bounds 12",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test in bounds 1",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 4),
-	BPF_MOV32_IMM(BPF_REG_2, 5 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 4,
-},
-{
-	"masking, test in bounds 2",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test in bounds 3",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0xfffffffe),
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0xfffffffe,
-},
-{
-	"masking, test in bounds 4",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0xabcde),
-	BPF_MOV32_IMM(BPF_REG_2, 0xabcdef - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0xabcde,
-},
-{
-	"masking, test in bounds 5",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_MOV32_IMM(BPF_REG_2, 1 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"masking, test in bounds 6",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_1, 46),
-	BPF_MOV32_IMM(BPF_REG_2, 47 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_1, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 46,
-},
-{
-	"masking, test in bounds 7",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, -46),
-	BPF_ALU64_IMM(BPF_MUL, BPF_REG_3, -1),
-	BPF_MOV32_IMM(BPF_REG_2, 47 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_3),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_3, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 46,
-},
-{
-	"masking, test in bounds 8",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, -47),
-	BPF_ALU64_IMM(BPF_MUL, BPF_REG_3, -1),
-	BPF_MOV32_IMM(BPF_REG_2, 47 - 1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_3),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_2, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_2, 0),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_2, 63),
-	BPF_ALU64_REG(BPF_AND, BPF_REG_3, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-- 
cgit v1.2.3-70-g09d2


From 65428312e38d896ce101ee3489403cff320d9b74 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:11 +0200
Subject: selftests/bpf: verifier/meta_access.c converted to inline assembly

Test verifier/meta_access.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-31-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_meta_access.c     | 284 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/meta_access.c | 235 -----------------
 3 files changed, 286 insertions(+), 235 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_meta_access.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/meta_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index b23fcbe4f83b..bd48a584a356 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -27,6 +27,7 @@
 #include "verifier_map_ptr.skel.h"
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
+#include "verifier_meta_access.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -76,3 +77,4 @@ void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
+void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_meta_access.c b/tools/testing/selftests/bpf/progs/verifier_meta_access.c
new file mode 100644
index 000000000000..d81722fb5f19
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_meta_access.c
@@ -0,0 +1,284 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/meta_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("xdp")
+__description("meta access, test1")
+__success __retval(0)
+__naked void meta_access_test1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test2")
+__failure __msg("invalid access to packet, off=-8")
+__naked void meta_access_test2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r0 = r2;					\
+	r0 -= 8;					\
+	r4 = r2;					\
+	r4 += 8;					\
+	if r4 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test3")
+__failure __msg("invalid access to packet")
+__naked void meta_access_test3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test4")
+__failure __msg("invalid access to packet")
+__naked void meta_access_test4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r4 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r0 = r4;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test5")
+__failure __msg("R3 !read_ok")
+__naked void meta_access_test5(void)
+{
+	asm volatile ("					\
+	r3 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r4 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r0 = r3;					\
+	r0 += 8;					\
+	if r0 > r4 goto l0_%=;				\
+	r2 = -8;					\
+	call %[bpf_xdp_adjust_meta];			\
+	r0 = *(u8*)(r3 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_xdp_adjust_meta),
+	  __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test6")
+__failure __msg("invalid access to packet")
+__naked void meta_access_test6(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r0 = r3;					\
+	r0 += 8;					\
+	r4 = r2;					\
+	r4 += 8;					\
+	if r4 > r0 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test7")
+__success __retval(0)
+__naked void meta_access_test7(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r0 = r3;					\
+	r0 += 8;					\
+	r4 = r2;					\
+	r4 += 8;					\
+	if r4 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test8")
+__success __retval(0)
+__naked void meta_access_test8(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r4 = r2;					\
+	r4 += 0xFFFF;					\
+	if r4 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test9")
+__failure __msg("invalid access to packet")
+__naked void meta_access_test9(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r4 = r2;					\
+	r4 += 0xFFFF;					\
+	r4 += 1;					\
+	if r4 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test10")
+__failure __msg("invalid access to packet")
+__naked void meta_access_test10(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r4 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r5 = 42;					\
+	r6 = 24;					\
+	*(u64*)(r10 - 8) = r5;				\
+	lock *(u64 *)(r10 - 8) += r6;			\
+	r5 = *(u64*)(r10 - 8);				\
+	if r5 > 100 goto l0_%=;				\
+	r3 += r5;					\
+	r5 = r3;					\
+	r6 = r2;					\
+	r6 += 8;					\
+	if r6 > r5 goto l0_%=;				\
+	r2 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test11")
+__success __retval(0)
+__naked void meta_access_test11(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r5 = 42;					\
+	r6 = 24;					\
+	*(u64*)(r10 - 8) = r5;				\
+	lock *(u64 *)(r10 - 8) += r6;			\
+	r5 = *(u64*)(r10 - 8);				\
+	if r5 > 100 goto l0_%=;				\
+	r2 += r5;					\
+	r5 = r2;					\
+	r6 = r2;					\
+	r6 += 8;					\
+	if r6 > r3 goto l0_%=;				\
+	r5 = *(u8*)(r5 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("meta access, test12")
+__success __retval(0)
+__naked void meta_access_test12(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r4 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r5 = r3;					\
+	r5 += 16;					\
+	if r5 > r4 goto l0_%=;				\
+	r0 = *(u8*)(r3 + 0);				\
+	r5 = r2;					\
+	r5 += 16;					\
+	if r5 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/meta_access.c b/tools/testing/selftests/bpf/verifier/meta_access.c
deleted file mode 100644
index b45e8af41420..000000000000
--- a/tools/testing/selftests/bpf/verifier/meta_access.c
+++ /dev/null
@@ -1,235 +0,0 @@
-{
-	"meta access, test1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_0, 8),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet, off=-8",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test3",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test4",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test5",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_4, 3),
-	BPF_MOV64_IMM(BPF_REG_2, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_xdp_adjust_meta),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_3, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R3 !read_ok",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test6",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_0, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test7",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test8",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 0xFFFF),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test9",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 0xFFFF),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test10",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_IMM(BPF_REG_5, 42),
-	BPF_MOV64_IMM(BPF_REG_6, 24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_5, -8),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_10, BPF_REG_6, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_5, BPF_REG_10, -8),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_5, 100, 6),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_5),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_3),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_5, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test11",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_IMM(BPF_REG_5, 42),
-	BPF_MOV64_IMM(BPF_REG_6, 24),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_5, -8),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_10, BPF_REG_6, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_5, BPF_REG_10, -8),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_5, 100, 6),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_5),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_5, BPF_REG_5, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"meta access, test12",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 16),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_5, BPF_REG_4, 5),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_3, 0),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 16),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_5, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-- 
cgit v1.2.3-70-g09d2


From 5a77a01f3320562db51dfdd24b2921768637daf2 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:12 +0200
Subject: selftests/bpf: verifier/raw_stack.c converted to inline assembly

Test verifier/raw_stack.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-32-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_raw_stack.c       | 371 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/raw_stack.c   | 305 -----------------
 3 files changed, 373 insertions(+), 305 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_raw_stack.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/raw_stack.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index bd48a584a356..4a73cac3f9ba 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -28,6 +28,7 @@
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
 #include "verifier_meta_access.skel.h"
+#include "verifier_raw_stack.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -78,3 +79,4 @@ void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
+void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_raw_stack.c b/tools/testing/selftests/bpf/progs/verifier_raw_stack.c
new file mode 100644
index 000000000000..efbfc3a4ad6a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_raw_stack.c
@@ -0,0 +1,371 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/raw_stack.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("tc")
+__description("raw_stack: no skb_load_bytes")
+__failure __msg("invalid read from stack R6 off=-8 size=8")
+__naked void stack_no_skb_load_bytes(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	r3 = r6;					\
+	r4 = 8;						\
+	/* Call to skb_load_bytes() omitted. */		\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, negative len")
+__failure __msg("R4 min value is negative")
+__naked void skb_load_bytes_negative_len(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	r3 = r6;					\
+	r4 = -8;					\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, negative len 2")
+__failure __msg("R4 min value is negative")
+__naked void load_bytes_negative_len_2(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	r3 = r6;					\
+	r4 = %[__imm_0];				\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes),
+	  __imm_const(__imm_0, ~0)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, zero len")
+__failure __msg("invalid zero-sized read")
+__naked void skb_load_bytes_zero_len(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	r3 = r6;					\
+	r4 = 0;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, no init")
+__success __retval(0)
+__naked void skb_load_bytes_no_init(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, init")
+__success __retval(0)
+__naked void stack_skb_load_bytes_init(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	r3 = 0xcafe;					\
+	*(u64*)(r6 + 0) = r3;				\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, spilled regs around bounds")
+__success __retval(0)
+__naked void bytes_spilled_regs_around_bounds(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -16;					\
+	*(u64*)(r6 - 8) = r1;				\
+	*(u64*)(r6 + 8) = r1;				\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 - 8);				\
+	r2 = *(u64*)(r6 + 8);				\
+	r0 = *(u32*)(r0 + %[__sk_buff_mark]);		\
+	r2 = *(u32*)(r2 + %[__sk_buff_priority]);	\
+	r0 += r2;					\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(__sk_buff_priority, offsetof(struct __sk_buff, priority))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, spilled regs corruption")
+__failure __msg("R0 invalid mem access 'scalar'")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void load_bytes_spilled_regs_corruption(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -8;					\
+	*(u64*)(r6 + 0) = r1;				\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	r0 = *(u32*)(r0 + %[__sk_buff_mark]);		\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, spilled regs corruption 2")
+__failure __msg("R3 invalid mem access 'scalar'")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void bytes_spilled_regs_corruption_2(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -16;					\
+	*(u64*)(r6 - 8) = r1;				\
+	*(u64*)(r6 + 0) = r1;				\
+	*(u64*)(r6 + 8) = r1;				\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 - 8);				\
+	r2 = *(u64*)(r6 + 8);				\
+	r3 = *(u64*)(r6 + 0);				\
+	r0 = *(u32*)(r0 + %[__sk_buff_mark]);		\
+	r2 = *(u32*)(r2 + %[__sk_buff_priority]);	\
+	r0 += r2;					\
+	r3 = *(u32*)(r3 + %[__sk_buff_pkt_type]);	\
+	r0 += r3;					\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(__sk_buff_pkt_type, offsetof(struct __sk_buff, pkt_type)),
+	  __imm_const(__sk_buff_priority, offsetof(struct __sk_buff, priority))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, spilled regs + data")
+__success __retval(0)
+__naked void load_bytes_spilled_regs_data(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -16;					\
+	*(u64*)(r6 - 8) = r1;				\
+	*(u64*)(r6 + 0) = r1;				\
+	*(u64*)(r6 + 8) = r1;				\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 - 8);				\
+	r2 = *(u64*)(r6 + 8);				\
+	r3 = *(u64*)(r6 + 0);				\
+	r0 = *(u32*)(r0 + %[__sk_buff_mark]);		\
+	r2 = *(u32*)(r2 + %[__sk_buff_priority]);	\
+	r0 += r2;					\
+	r0 += r3;					\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(__sk_buff_priority, offsetof(struct __sk_buff, priority))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, invalid access 1")
+__failure __msg("invalid indirect access to stack R3 off=-513 size=8")
+__naked void load_bytes_invalid_access_1(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -513;					\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, invalid access 2")
+__failure __msg("invalid indirect access to stack R3 off=-1 size=8")
+__naked void load_bytes_invalid_access_2(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -1;					\
+	r3 = r6;					\
+	r4 = 8;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, invalid access 3")
+__failure __msg("R4 min value is negative")
+__naked void load_bytes_invalid_access_3(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += 0xffffffff;				\
+	r3 = r6;					\
+	r4 = 0xffffffff;				\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, invalid access 4")
+__failure
+__msg("R4 unbounded memory access, use 'var &= const' or 'if (var < const)'")
+__naked void load_bytes_invalid_access_4(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -1;					\
+	r3 = r6;					\
+	r4 = 0x7fffffff;				\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, invalid access 5")
+__failure
+__msg("R4 unbounded memory access, use 'var &= const' or 'if (var < const)'")
+__naked void load_bytes_invalid_access_5(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -512;					\
+	r3 = r6;					\
+	r4 = 0x7fffffff;				\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, invalid access 6")
+__failure __msg("invalid zero-sized read")
+__naked void load_bytes_invalid_access_6(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -512;					\
+	r3 = r6;					\
+	r4 = 0;						\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("raw_stack: skb_load_bytes, large access")
+__success __retval(0)
+__naked void skb_load_bytes_large_access(void)
+{
+	asm volatile ("					\
+	r2 = 4;						\
+	r6 = r10;					\
+	r6 += -512;					\
+	r3 = r6;					\
+	r4 = 512;					\
+	call %[bpf_skb_load_bytes];			\
+	r0 = *(u64*)(r6 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skb_load_bytes)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/raw_stack.c b/tools/testing/selftests/bpf/verifier/raw_stack.c
deleted file mode 100644
index eb5ed936580b..000000000000
--- a/tools/testing/selftests/bpf/verifier/raw_stack.c
+++ /dev/null
@@ -1,305 +0,0 @@
-{
-	"raw_stack: no skb_load_bytes",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	/* Call to skb_load_bytes() omitted. */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid read from stack R6 off=-8 size=8",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, negative len",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R4 min value is negative",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, negative len 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, ~0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R4 min value is negative",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, zero len",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid zero-sized read",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, no init",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, init",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_6, 0, 0xcafe),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, spilled regs around bounds",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1,  8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_6,  8),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_2,
-		    offsetof(struct __sk_buff, priority)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, spilled regs corruption",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R0 invalid mem access 'scalar'",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"raw_stack: skb_load_bytes, spilled regs corruption 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1,  0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1,  8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_6,  8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_6,  0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_2,
-		    offsetof(struct __sk_buff, priority)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_3,
-		    offsetof(struct __sk_buff, pkt_type)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_3),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R3 invalid mem access 'scalar'",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"raw_stack: skb_load_bytes, spilled regs + data",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1,  0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1,  8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_6,  8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_6,  0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_2,
-		    offsetof(struct __sk_buff, priority)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_3),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, invalid access 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -513),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid indirect access to stack R3 off=-513 size=8",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, invalid access 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -1),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid indirect access to stack R3 off=-1 size=8",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, invalid access 3",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 0xffffffff),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 0xffffffff),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R4 min value is negative",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, invalid access 4",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -1),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 0x7fffffff),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R4 unbounded memory access, use 'var &= const' or 'if (var < const)'",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, invalid access 5",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -512),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 0x7fffffff),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R4 unbounded memory access, use 'var &= const' or 'if (var < const)'",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, invalid access 6",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -512),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid zero-sized read",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"raw_stack: skb_load_bytes, large access",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -512),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_4, 512),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_skb_load_bytes),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_6, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-- 
cgit v1.2.3-70-g09d2


From 18cdc2b531fbe13450c51132a5b511c670c77585 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:13 +0200
Subject: selftests/bpf: verifier/raw_tp_writable.c converted to inline
 assembly

Test verifier/raw_tp_writable.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-33-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../selftests/bpf/progs/verifier_raw_tp_writable.c | 50 ++++++++++++++++++++++
 .../selftests/bpf/verifier/raw_tp_writable.c       | 35 ---------------
 3 files changed, 52 insertions(+), 35 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_raw_tp_writable.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/raw_tp_writable.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 4a73cac3f9ba..f7488904f26e 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -29,6 +29,7 @@
 #include "verifier_masking.skel.h"
 #include "verifier_meta_access.skel.h"
 #include "verifier_raw_stack.skel.h"
+#include "verifier_raw_tp_writable.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -80,3 +81,4 @@ void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
+void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_raw_tp_writable.c b/tools/testing/selftests/bpf/progs/verifier_raw_tp_writable.c
new file mode 100644
index 000000000000..14a0172e2141
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_raw_tp_writable.c
@@ -0,0 +1,50 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/raw_tp_writable.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("raw_tracepoint.w")
+__description("raw_tracepoint_writable: reject variable offset")
+__failure
+__msg("R6 invalid variable buffer offset: off=0, var_off=(0x0; 0xffffffff)")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void tracepoint_writable_reject_variable_offset(void)
+{
+	asm volatile ("					\
+	/* r6 is our tp buffer */			\
+	r6 = *(u64*)(r1 + 0);				\
+	r1 = %[map_hash_8b] ll;				\
+	/* move the key (== 0) to r10-8 */		\
+	w0 = 0;						\
+	r2 = r10;					\
+	r2 += -8;					\
+	*(u64*)(r2 + 0) = r0;				\
+	/* lookup in the map */				\
+	call %[bpf_map_lookup_elem];			\
+	/* exit clean if null */			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	/* shift the buffer pointer to a variable location */\
+	r0 = *(u32*)(r0 + 0);				\
+	r6 += r0;					\
+	/* clobber whatever's there */			\
+	r7 = 4242;					\
+	*(u64*)(r6 + 0) = r7;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/raw_tp_writable.c b/tools/testing/selftests/bpf/verifier/raw_tp_writable.c
deleted file mode 100644
index 2978fb5a769d..000000000000
--- a/tools/testing/selftests/bpf/verifier/raw_tp_writable.c
+++ /dev/null
@@ -1,35 +0,0 @@
-{
-	"raw_tracepoint_writable: reject variable offset",
-	.insns = {
-		/* r6 is our tp buffer */
-		BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-
-		BPF_LD_MAP_FD(BPF_REG_1, 0),
-		/* move the key (== 0) to r10-8 */
-		BPF_MOV32_IMM(BPF_REG_0, 0),
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-		BPF_STX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-		/* lookup in the map */
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_map_lookup_elem),
-
-		/* exit clean if null */
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-		BPF_EXIT_INSN(),
-
-		/* shift the buffer pointer to a variable location */
-		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-		BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_0),
-		/* clobber whatever's there */
-		BPF_MOV64_IMM(BPF_REG_7, 4242),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_7, 0),
-
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1, },
-	.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
-	.errstr = "R6 invalid variable buffer offset: off=0, var_off=(0x0; 0xffffffff)",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-- 
cgit v1.2.3-70-g09d2


From b7e4203086eb4d85aa7bccd7f33c2835b34778d6 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:14 +0200
Subject: selftests/bpf: verifier/ringbuf.c converted to inline assembly

Test verifier/ringbuf.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-34-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_ringbuf.c | 131 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/ringbuf.c     |  95 ---------------
 3 files changed, 133 insertions(+), 95 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_ringbuf.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/ringbuf.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index f7488904f26e..df5fc6fe1647 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -30,6 +30,7 @@
 #include "verifier_meta_access.skel.h"
 #include "verifier_raw_stack.skel.h"
 #include "verifier_raw_tp_writable.skel.h"
+#include "verifier_ringbuf.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -82,3 +83,4 @@ void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
+void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_ringbuf.c b/tools/testing/selftests/bpf/progs/verifier_ringbuf.c
new file mode 100644
index 000000000000..ae1d521f326c
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_ringbuf.c
@@ -0,0 +1,131 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/ringbuf.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_RINGBUF);
+	__uint(max_entries, 4096);
+} map_ringbuf SEC(".maps");
+
+SEC("socket")
+__description("ringbuf: invalid reservation offset 1")
+__failure __msg("R1 must have zero offset when passed to release func")
+__failure_unpriv
+__naked void ringbuf_invalid_reservation_offset_1(void)
+{
+	asm volatile ("					\
+	/* reserve 8 byte ringbuf memory */		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_ringbuf_reserve];			\
+	/* store a pointer to the reserved memory in R6 */\
+	r6 = r0;					\
+	/* check whether the reservation was successful */\
+	if r0 == 0 goto l0_%=;				\
+	/* spill R6(mem) into the stack */		\
+	*(u64*)(r10 - 8) = r6;				\
+	/* fill it back in R7 */			\
+	r7 = *(u64*)(r10 - 8);				\
+	/* should be able to access *(R7) = 0 */	\
+	r1 = 0;						\
+	*(u64*)(r7 + 0) = r1;				\
+	/* submit the reserved ringbuf memory */	\
+	r1 = r7;					\
+	/* add invalid offset to reserved ringbuf memory */\
+	r1 += 0xcafe;					\
+	r2 = 0;						\
+	call %[bpf_ringbuf_submit];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ringbuf_reserve),
+	  __imm(bpf_ringbuf_submit),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ringbuf: invalid reservation offset 2")
+__failure __msg("R7 min value is outside of the allowed memory range")
+__failure_unpriv
+__naked void ringbuf_invalid_reservation_offset_2(void)
+{
+	asm volatile ("					\
+	/* reserve 8 byte ringbuf memory */		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_ringbuf_reserve];			\
+	/* store a pointer to the reserved memory in R6 */\
+	r6 = r0;					\
+	/* check whether the reservation was successful */\
+	if r0 == 0 goto l0_%=;				\
+	/* spill R6(mem) into the stack */		\
+	*(u64*)(r10 - 8) = r6;				\
+	/* fill it back in R7 */			\
+	r7 = *(u64*)(r10 - 8);				\
+	/* add invalid offset to reserved ringbuf memory */\
+	r7 += 0xcafe;					\
+	/* should be able to access *(R7) = 0 */	\
+	r1 = 0;						\
+	*(u64*)(r7 + 0) = r1;				\
+	/* submit the reserved ringbuf memory */	\
+	r1 = r7;					\
+	r2 = 0;						\
+	call %[bpf_ringbuf_submit];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ringbuf_reserve),
+	  __imm(bpf_ringbuf_submit),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("ringbuf: check passing rb mem to helpers")
+__success __retval(0)
+__naked void passing_rb_mem_to_helpers(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	/* reserve 8 byte ringbuf memory */		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_ringbuf_reserve];			\
+	r7 = r0;					\
+	/* check whether the reservation was successful */\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	/* pass allocated ring buffer memory to fib lookup */\
+	r1 = r6;					\
+	r2 = r0;					\
+	r3 = 8;						\
+	r4 = 0;						\
+	call %[bpf_fib_lookup];				\
+	/* submit the ringbuf memory */			\
+	r1 = r7;					\
+	r2 = 0;						\
+	call %[bpf_ringbuf_submit];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_fib_lookup),
+	  __imm(bpf_ringbuf_reserve),
+	  __imm(bpf_ringbuf_submit),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/ringbuf.c b/tools/testing/selftests/bpf/verifier/ringbuf.c
deleted file mode 100644
index 92e3f6a61a79..000000000000
--- a/tools/testing/selftests/bpf/verifier/ringbuf.c
+++ /dev/null
@@ -1,95 +0,0 @@
-{
-	"ringbuf: invalid reservation offset 1",
-	.insns = {
-	/* reserve 8 byte ringbuf memory */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_reserve),
-	/* store a pointer to the reserved memory in R6 */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* check whether the reservation was successful */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	/* spill R6(mem) into the stack */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_6, -8),
-	/* fill it back in R7 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_10, -8),
-	/* should be able to access *(R7) = 0 */
-	BPF_ST_MEM(BPF_DW, BPF_REG_7, 0, 0),
-	/* submit the reserved ringbuf memory */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	/* add invalid offset to reserved ringbuf memory */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0xcafe),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_submit),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 1 },
-	.result = REJECT,
-	.errstr = "R1 must have zero offset when passed to release func",
-},
-{
-	"ringbuf: invalid reservation offset 2",
-	.insns = {
-	/* reserve 8 byte ringbuf memory */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_reserve),
-	/* store a pointer to the reserved memory in R6 */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* check whether the reservation was successful */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	/* spill R6(mem) into the stack */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_6, -8),
-	/* fill it back in R7 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_10, -8),
-	/* add invalid offset to reserved ringbuf memory */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, 0xcafe),
-	/* should be able to access *(R7) = 0 */
-	BPF_ST_MEM(BPF_DW, BPF_REG_7, 0, 0),
-	/* submit the reserved ringbuf memory */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_submit),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 1 },
-	.result = REJECT,
-	.errstr = "R7 min value is outside of the allowed memory range",
-},
-{
-	"ringbuf: check passing rb mem to helpers",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	/* reserve 8 byte ringbuf memory */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_reserve),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* check whether the reservation was successful */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	/* pass allocated ring buffer memory to fib lookup */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 8),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_fib_lookup),
-	/* submit the ringbuf memory */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_submit),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 2 },
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.result = ACCEPT,
-},
-- 
cgit v1.2.3-70-g09d2


From f4fe3cfe6c3ac60a6ec086a291b1c0f59e7daaa2 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:15 +0200
Subject: selftests/bpf: verifier/spill_fill.c converted to inline assembly

Test verifier/spill_fill.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-35-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_spill_fill.c      | 374 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/spill_fill.c  | 345 -------------------
 3 files changed, 376 insertions(+), 345 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_spill_fill.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/spill_fill.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index df5fc6fe1647..e2b131d2ba94 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -31,6 +31,7 @@
 #include "verifier_raw_stack.skel.h"
 #include "verifier_raw_tp_writable.skel.h"
 #include "verifier_ringbuf.skel.h"
+#include "verifier_spill_fill.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -84,3 +85,4 @@ void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
+void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_spill_fill.c b/tools/testing/selftests/bpf/progs/verifier_spill_fill.c
new file mode 100644
index 000000000000..136e5530b72c
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_spill_fill.c
@@ -0,0 +1,374 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/spill_fill.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_RINGBUF);
+	__uint(max_entries, 4096);
+} map_ringbuf SEC(".maps");
+
+SEC("socket")
+__description("check valid spill/fill")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(POINTER_VALUE)
+__naked void check_valid_spill_fill(void)
+{
+	asm volatile ("					\
+	/* spill R1(ctx) into stack */			\
+	*(u64*)(r10 - 8) = r1;				\
+	/* fill it back into R2 */			\
+	r2 = *(u64*)(r10 - 8);				\
+	/* should be able to access R0 = *(R2 + 8) */	\
+	/* BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, 8), */\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check valid spill/fill, skb mark")
+__success __success_unpriv __retval(0)
+__naked void valid_spill_fill_skb_mark(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	*(u64*)(r10 - 8) = r6;				\
+	r0 = *(u64*)(r10 - 8);				\
+	r0 = *(u32*)(r0 + %[__sk_buff_mark]);		\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check valid spill/fill, ptr to mem")
+__success __success_unpriv __retval(0)
+__naked void spill_fill_ptr_to_mem(void)
+{
+	asm volatile ("					\
+	/* reserve 8 byte ringbuf memory */		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_ringbuf_reserve];			\
+	/* store a pointer to the reserved memory in R6 */\
+	r6 = r0;					\
+	/* check whether the reservation was successful */\
+	if r0 == 0 goto l0_%=;				\
+	/* spill R6(mem) into the stack */		\
+	*(u64*)(r10 - 8) = r6;				\
+	/* fill it back in R7 */			\
+	r7 = *(u64*)(r10 - 8);				\
+	/* should be able to access *(R7) = 0 */	\
+	r1 = 0;						\
+	*(u64*)(r7 + 0) = r1;				\
+	/* submit the reserved ringbuf memory */	\
+	r1 = r7;					\
+	r2 = 0;						\
+	call %[bpf_ringbuf_submit];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ringbuf_reserve),
+	  __imm(bpf_ringbuf_submit),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check with invalid reg offset 0")
+__failure __msg("R0 pointer arithmetic on ringbuf_mem_or_null prohibited")
+__failure_unpriv
+__naked void with_invalid_reg_offset_0(void)
+{
+	asm volatile ("					\
+	/* reserve 8 byte ringbuf memory */		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_ringbuf_reserve];			\
+	/* store a pointer to the reserved memory in R6 */\
+	r6 = r0;					\
+	/* add invalid offset to memory or NULL */	\
+	r0 += 1;					\
+	/* check whether the reservation was successful */\
+	if r0 == 0 goto l0_%=;				\
+	/* should not be able to access *(R7) = 0 */	\
+	r1 = 0;						\
+	*(u32*)(r6 + 0) = r1;				\
+	/* submit the reserved ringbuf memory */	\
+	r1 = r6;					\
+	r2 = 0;						\
+	call %[bpf_ringbuf_submit];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ringbuf_reserve),
+	  __imm(bpf_ringbuf_submit),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check corrupted spill/fill")
+__failure __msg("R0 invalid mem access 'scalar'")
+__msg_unpriv("attempt to corrupt spilled")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void check_corrupted_spill_fill(void)
+{
+	asm volatile ("					\
+	/* spill R1(ctx) into stack */			\
+	*(u64*)(r10 - 8) = r1;				\
+	/* mess up with R1 pointer on stack */		\
+	r0 = 0x23;					\
+	*(u8*)(r10 - 7) = r0;				\
+	/* fill back into R0 is fine for priv.		\
+	 * R0 now becomes SCALAR_VALUE.			\
+	 */						\
+	r0 = *(u64*)(r10 - 8);				\
+	/* Load from R0 should fail. */			\
+	r0 = *(u64*)(r0 + 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check corrupted spill/fill, LSB")
+__success __failure_unpriv __msg_unpriv("attempt to corrupt spilled")
+__retval(POINTER_VALUE)
+__naked void check_corrupted_spill_fill_lsb(void)
+{
+	asm volatile ("					\
+	*(u64*)(r10 - 8) = r1;				\
+	r0 = 0xcafe;					\
+	*(u16*)(r10 - 8) = r0;				\
+	r0 = *(u64*)(r10 - 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("check corrupted spill/fill, MSB")
+__success __failure_unpriv __msg_unpriv("attempt to corrupt spilled")
+__retval(POINTER_VALUE)
+__naked void check_corrupted_spill_fill_msb(void)
+{
+	asm volatile ("					\
+	*(u64*)(r10 - 8) = r1;				\
+	r0 = 0x12345678;				\
+	*(u32*)(r10 - 4) = r0;				\
+	r0 = *(u64*)(r10 - 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill and refill a u32 const scalar.  Offset to skb->data")
+__success __retval(0)
+__naked void scalar_offset_to_skb_data_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	w4 = 20;					\
+	*(u32*)(r10 - 8) = r4;				\
+	r4 = *(u32*)(r10 - 8);				\
+	r0 = r2;					\
+	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=20 */	\
+	r0 += r4;					\
+	/* if (r0 > r3) R0=pkt,off=20 R2=pkt R3=pkt_end R4=20 */\
+	if r0 > r3 goto l0_%=;				\
+	/* r0 = *(u32 *)r2 R0=pkt,off=20,r=20 R2=pkt,r=20 R3=pkt_end R4=20 */\
+	r0 = *(u32*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("Spill a u32 const, refill from another half of the uninit u32 from the stack")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid read from stack off -4+0 size 4")
+__retval(0)
+__naked void uninit_u32_from_the_stack(void)
+{
+	asm volatile ("					\
+	w4 = 20;					\
+	*(u32*)(r10 - 8) = r4;				\
+	/* r4 = *(u32 *)(r10 -4) fp-8=????rrrr*/	\
+	r4 = *(u32*)(r10 - 4);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill a u32 const scalar.  Refill as u16.  Offset to skb->data")
+__failure __msg("invalid access to packet")
+__naked void u16_offset_to_skb_data(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	w4 = 20;					\
+	*(u32*)(r10 - 8) = r4;				\
+	r4 = *(u16*)(r10 - 8);				\
+	r0 = r2;					\
+	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=65535 */\
+	r0 += r4;					\
+	/* if (r0 > r3) R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=umax=65535 */\
+	if r0 > r3 goto l0_%=;				\
+	/* r0 = *(u32 *)r2 R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=20 */\
+	r0 = *(u32*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill u32 const scalars.  Refill as u64.  Offset to skb->data")
+__failure __msg("invalid access to packet")
+__naked void u64_offset_to_skb_data(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	w6 = 0;						\
+	w7 = 20;					\
+	*(u32*)(r10 - 4) = r6;				\
+	*(u32*)(r10 - 8) = r7;				\
+	r4 = *(u16*)(r10 - 8);				\
+	r0 = r2;					\
+	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=65535 */\
+	r0 += r4;					\
+	/* if (r0 > r3) R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=umax=65535 */\
+	if r0 > r3 goto l0_%=;				\
+	/* r0 = *(u32 *)r2 R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=20 */\
+	r0 = *(u32*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill a u32 const scalar.  Refill as u16 from fp-6.  Offset to skb->data")
+__failure __msg("invalid access to packet")
+__naked void _6_offset_to_skb_data(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	w4 = 20;					\
+	*(u32*)(r10 - 8) = r4;				\
+	r4 = *(u16*)(r10 - 6);				\
+	r0 = r2;					\
+	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=65535 */\
+	r0 += r4;					\
+	/* if (r0 > r3) R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=umax=65535 */\
+	if r0 > r3 goto l0_%=;				\
+	/* r0 = *(u32 *)r2 R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=20 */\
+	r0 = *(u32*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill and refill a u32 const scalar at non 8byte aligned stack addr.  Offset to skb->data")
+__failure __msg("invalid access to packet")
+__naked void addr_offset_to_skb_data(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	w4 = 20;					\
+	*(u32*)(r10 - 8) = r4;				\
+	*(u32*)(r10 - 4) = r4;				\
+	r4 = *(u32*)(r10 - 4);				\
+	r0 = r2;					\
+	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=U32_MAX */\
+	r0 += r4;					\
+	/* if (r0 > r3) R0=pkt,umax=U32_MAX R2=pkt R3=pkt_end R4= */\
+	if r0 > r3 goto l0_%=;				\
+	/* r0 = *(u32 *)r2 R0=pkt,umax=U32_MAX R2=pkt R3=pkt_end R4= */\
+	r0 = *(u32*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill and refill a umax=40 bounded scalar.  Offset to skb->data")
+__success __retval(0)
+__naked void scalar_offset_to_skb_data_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r4 = *(u64*)(r1 + %[__sk_buff_tstamp]);		\
+	if r4 <= 40 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	/* *(u32 *)(r10 -8) = r4 R4=umax=40 */		\
+	*(u32*)(r10 - 8) = r4;				\
+	/* r4 = (*u32 *)(r10 - 8) */			\
+	r4 = *(u32*)(r10 - 8);				\
+	/* r2 += r4 R2=pkt R4=umax=40 */		\
+	r2 += r4;					\
+	/* r0 = r2 R2=pkt,umax=40 R4=umax=40 */		\
+	r0 = r2;					\
+	/* r2 += 20 R0=pkt,umax=40 R2=pkt,umax=40 */	\
+	r2 += 20;					\
+	/* if (r2 > r3) R0=pkt,umax=40 R2=pkt,off=20,umax=40 */\
+	if r2 > r3 goto l1_%=;				\
+	/* r0 = *(u32 *)r0 R0=pkt,r=20,umax=40 R2=pkt,off=20,r=20,umax=40 */\
+	r0 = *(u32*)(r0 + 0);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_tstamp, offsetof(struct __sk_buff, tstamp))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("Spill a u32 scalar at fp-4 and then at fp-8")
+__success __retval(0)
+__naked void and_then_at_fp_8(void)
+{
+	asm volatile ("					\
+	w4 = 4321;					\
+	*(u32*)(r10 - 4) = r4;				\
+	*(u32*)(r10 - 8) = r4;				\
+	r4 = *(u64*)(r10 - 8);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/spill_fill.c b/tools/testing/selftests/bpf/verifier/spill_fill.c
deleted file mode 100644
index d1463bf4949a..000000000000
--- a/tools/testing/selftests/bpf/verifier/spill_fill.c
+++ /dev/null
@@ -1,345 +0,0 @@
-{
-	"check valid spill/fill",
-	.insns = {
-	/* spill R1(ctx) into stack */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	/* fill it back into R2 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -8),
-	/* should be able to access R0 = *(R2 + 8) */
-	/* BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, 8), */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 leaks addr",
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.retval = POINTER_VALUE,
-},
-{
-	"check valid spill/fill, skb mark",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_6, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = ACCEPT,
-},
-{
-	"check valid spill/fill, ptr to mem",
-	.insns = {
-	/* reserve 8 byte ringbuf memory */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_reserve),
-	/* store a pointer to the reserved memory in R6 */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* check whether the reservation was successful */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	/* spill R6(mem) into the stack */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_6, -8),
-	/* fill it back in R7 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_10, -8),
-	/* should be able to access *(R7) = 0 */
-	BPF_ST_MEM(BPF_DW, BPF_REG_7, 0, 0),
-	/* submit the reserved ringbuf memory */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_submit),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 1 },
-	.result = ACCEPT,
-	.result_unpriv = ACCEPT,
-},
-{
-	"check with invalid reg offset 0",
-	.insns = {
-	/* reserve 8 byte ringbuf memory */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_reserve),
-	/* store a pointer to the reserved memory in R6 */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* add invalid offset to memory or NULL */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	/* check whether the reservation was successful */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	/* should not be able to access *(R7) = 0 */
-	BPF_ST_MEM(BPF_W, BPF_REG_6, 0, 0),
-	/* submit the reserved ringbuf memory */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_ringbuf_submit),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_ringbuf = { 1 },
-	.result = REJECT,
-	.errstr = "R0 pointer arithmetic on ringbuf_mem_or_null prohibited",
-},
-{
-	"check corrupted spill/fill",
-	.insns = {
-	/* spill R1(ctx) into stack */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	/* mess up with R1 pointer on stack */
-	BPF_ST_MEM(BPF_B, BPF_REG_10, -7, 0x23),
-	/* fill back into R0 is fine for priv.
-	 * R0 now becomes SCALAR_VALUE.
-	 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	/* Load from R0 should fail. */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "attempt to corrupt spilled",
-	.errstr = "R0 invalid mem access 'scalar'",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"check corrupted spill/fill, LSB",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_ST_MEM(BPF_H, BPF_REG_10, -8, 0xcafe),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "attempt to corrupt spilled",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.retval = POINTER_VALUE,
-},
-{
-	"check corrupted spill/fill, MSB",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0x12345678),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "attempt to corrupt spilled",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.retval = POINTER_VALUE,
-},
-{
-	"Spill and refill a u32 const scalar.  Offset to skb->data",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	/* r4 = 20 */
-	BPF_MOV32_IMM(BPF_REG_4, 20),
-	/* *(u32 *)(r10 -8) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* r4 = *(u32 *)(r10 -8) */
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_10, -8),
-	/* r0 = r2 */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=20 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	/* if (r0 > r3) R0=pkt,off=20 R2=pkt R3=pkt_end R4=20 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	/* r0 = *(u32 *)r2 R0=pkt,off=20,r=20 R2=pkt,r=20 R3=pkt_end R4=20 */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"Spill a u32 const, refill from another half of the uninit u32 from the stack",
-	.insns = {
-	/* r4 = 20 */
-	BPF_MOV32_IMM(BPF_REG_4, 20),
-	/* *(u32 *)(r10 -8) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* r4 = *(u32 *)(r10 -4) fp-8=????rrrr*/
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_10, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "invalid read from stack off -4+0 size 4",
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-{
-	"Spill a u32 const scalar.  Refill as u16.  Offset to skb->data",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	/* r4 = 20 */
-	BPF_MOV32_IMM(BPF_REG_4, 20),
-	/* *(u32 *)(r10 -8) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* r4 = *(u16 *)(r10 -8) */
-	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_10, -8),
-	/* r0 = r2 */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=65535 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	/* if (r0 > r3) R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=umax=65535 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	/* r0 = *(u32 *)r2 R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=20 */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"Spill u32 const scalars.  Refill as u64.  Offset to skb->data",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	/* r6 = 0 */
-	BPF_MOV32_IMM(BPF_REG_6, 0),
-	/* r7 = 20 */
-	BPF_MOV32_IMM(BPF_REG_7, 20),
-	/* *(u32 *)(r10 -4) = r6 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_6, -4),
-	/* *(u32 *)(r10 -8) = r7 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_7, -8),
-	/* r4 = *(u64 *)(r10 -8) */
-	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_10, -8),
-	/* r0 = r2 */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=65535 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	/* if (r0 > r3) R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=umax=65535 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	/* r0 = *(u32 *)r2 R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=20 */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"Spill a u32 const scalar.  Refill as u16 from fp-6.  Offset to skb->data",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	/* r4 = 20 */
-	BPF_MOV32_IMM(BPF_REG_4, 20),
-	/* *(u32 *)(r10 -8) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* r4 = *(u16 *)(r10 -6) */
-	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_10, -6),
-	/* r0 = r2 */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=65535 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	/* if (r0 > r3) R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=umax=65535 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	/* r0 = *(u32 *)r2 R0=pkt,umax=65535 R2=pkt R3=pkt_end R4=20 */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"Spill and refill a u32 const scalar at non 8byte aligned stack addr.  Offset to skb->data",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	/* r4 = 20 */
-	BPF_MOV32_IMM(BPF_REG_4, 20),
-	/* *(u32 *)(r10 -8) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* *(u32 *)(r10 -4) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -4),
-	/* r4 = *(u32 *)(r10 -4),  */
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_10, -4),
-	/* r0 = r2 */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	/* r0 += r4 R0=pkt R2=pkt R3=pkt_end R4=umax=U32_MAX */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	/* if (r0 > r3) R0=pkt,umax=U32_MAX R2=pkt R3=pkt_end R4= */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	/* r0 = *(u32 *)r2 R0=pkt,umax=U32_MAX R2=pkt R3=pkt_end R4= */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"Spill and refill a umax=40 bounded scalar.  Offset to skb->data",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct __sk_buff, tstamp)),
-	BPF_JMP_IMM(BPF_JLE, BPF_REG_4, 40, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	/* *(u32 *)(r10 -8) = r4 R4=umax=40 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* r4 = (*u32 *)(r10 - 8) */
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_10, -8),
-	/* r2 += r4 R2=pkt R4=umax=40 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_4),
-	/* r0 = r2 R2=pkt,umax=40 R4=umax=40 */
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	/* r2 += 20 R0=pkt,umax=40 R2=pkt,umax=40 */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 20),
-	/* if (r2 > r3) R0=pkt,umax=40 R2=pkt,off=20,umax=40 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_3, 1),
-	/* r0 = *(u32 *)r0 R0=pkt,r=20,umax=40 R2=pkt,off=20,r=20,umax=40 */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"Spill a u32 scalar at fp-4 and then at fp-8",
-	.insns = {
-	/* r4 = 4321 */
-	BPF_MOV32_IMM(BPF_REG_4, 4321),
-	/* *(u32 *)(r10 -4) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -4),
-	/* *(u32 *)(r10 -8) = r4 */
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_4, -8),
-	/* r4 = *(u64 *)(r10 -8) */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_10, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-- 
cgit v1.2.3-70-g09d2


From edff37b2f28f5c24f628981d0f26ca2fcd7e2ad5 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:16 +0200
Subject: selftests/bpf: verifier/stack_ptr.c converted to inline assembly

Test verifier/stack_ptr.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-36-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_stack_ptr.c       | 484 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/stack_ptr.c   | 359 ---------------
 3 files changed, 486 insertions(+), 359 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_stack_ptr.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/stack_ptr.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index e2b131d2ba94..ce1ca8c0c02e 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -32,6 +32,7 @@
 #include "verifier_raw_tp_writable.skel.h"
 #include "verifier_ringbuf.skel.h"
 #include "verifier_spill_fill.skel.h"
+#include "verifier_stack_ptr.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -86,3 +87,4 @@ void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
+void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_stack_ptr.c b/tools/testing/selftests/bpf/progs/verifier_stack_ptr.c
new file mode 100644
index 000000000000..e0f77e3e7869
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_stack_ptr.c
@@ -0,0 +1,484 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/stack_ptr.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <limits.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+} map_array_48b SEC(".maps");
+
+SEC("socket")
+__description("PTR_TO_STACK store/load")
+__success __success_unpriv __retval(0xfaceb00c)
+__naked void ptr_to_stack_store_load(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -10;					\
+	r0 = 0xfaceb00c;				\
+	*(u64*)(r1 + 2) = r0;				\
+	r0 = *(u64*)(r1 + 2);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK store/load - bad alignment on off")
+__failure __msg("misaligned stack access off (0x0; 0x0)+-8+2 size 8")
+__failure_unpriv
+__naked void load_bad_alignment_on_off(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -8;					\
+	r0 = 0xfaceb00c;				\
+	*(u64*)(r1 + 2) = r0;				\
+	r0 = *(u64*)(r1 + 2);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK store/load - bad alignment on reg")
+__failure __msg("misaligned stack access off (0x0; 0x0)+-10+8 size 8")
+__failure_unpriv
+__naked void load_bad_alignment_on_reg(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -10;					\
+	r0 = 0xfaceb00c;				\
+	*(u64*)(r1 + 8) = r0;				\
+	r0 = *(u64*)(r1 + 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK store/load - out of bounds low")
+__failure __msg("invalid write to stack R1 off=-79992 size=8")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void load_out_of_bounds_low(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -80000;					\
+	r0 = 0xfaceb00c;				\
+	*(u64*)(r1 + 8) = r0;				\
+	r0 = *(u64*)(r1 + 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK store/load - out of bounds high")
+__failure __msg("invalid write to stack R1 off=0 size=8")
+__failure_unpriv
+__naked void load_out_of_bounds_high(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -8;					\
+	r0 = 0xfaceb00c;				\
+	*(u64*)(r1 + 8) = r0;				\
+	r0 = *(u64*)(r1 + 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 1")
+__success __success_unpriv __retval(42)
+__naked void to_stack_check_high_1(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -1;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 2")
+__success __success_unpriv __retval(42)
+__naked void to_stack_check_high_2(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r0 = 42;					\
+	*(u8*)(r1 - 1) = r0;				\
+	r0 = *(u8*)(r1 - 1);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 3")
+__success __failure_unpriv
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__retval(42)
+__naked void to_stack_check_high_3(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += 0;					\
+	r0 = 42;					\
+	*(u8*)(r1 - 1) = r0;				\
+	r0 = *(u8*)(r1 - 1);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 4")
+__failure __msg("invalid write to stack R1 off=0 size=1")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_high_4(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += 0;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 5")
+__failure __msg("invalid write to stack R1")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_high_5(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[__imm_0];				\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, (1 << 29) - 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 6")
+__failure __msg("invalid write to stack")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_high_6(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[__imm_0];				\
+	r0 = 42;					\
+	*(u8*)(r1 + %[shrt_max]) = r0;			\
+	r0 = *(u8*)(r1 + %[shrt_max]);			\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, (1 << 29) - 1),
+	  __imm_const(shrt_max, SHRT_MAX)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check high 7")
+__failure __msg("fp pointer offset")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_high_7(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[__imm_0];				\
+	r1 += %[__imm_0];				\
+	r0 = 42;					\
+	*(u8*)(r1 + %[shrt_max]) = r0;			\
+	r0 = *(u8*)(r1 + %[shrt_max]);			\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, (1 << 29) - 1),
+	  __imm_const(shrt_max, SHRT_MAX)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 1")
+__success __success_unpriv __retval(42)
+__naked void to_stack_check_low_1(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -512;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 2")
+__success __failure_unpriv
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__retval(42)
+__naked void to_stack_check_low_2(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -513;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 1) = r0;				\
+	r0 = *(u8*)(r1 + 1);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 3")
+__failure __msg("invalid write to stack R1 off=-513 size=1")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_low_3(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -513;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 4")
+__failure __msg("math between fp pointer")
+__failure_unpriv
+__naked void to_stack_check_low_4(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[int_min];				\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	:
+	: __imm_const(int_min, INT_MIN)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 5")
+__failure __msg("invalid write to stack")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_low_5(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[__imm_0];				\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, -((1 << 29) - 1))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 6")
+__failure __msg("invalid write to stack")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_low_6(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[__imm_0];				\
+	r0 = 42;					\
+	*(u8*)(r1  %[shrt_min]) = r0;			\
+	r0 = *(u8*)(r1  %[shrt_min]);			\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, -((1 << 29) - 1)),
+	  __imm_const(shrt_min, SHRT_MIN)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK check low 7")
+__failure __msg("fp pointer offset")
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__naked void to_stack_check_low_7(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += %[__imm_0];				\
+	r1 += %[__imm_0];				\
+	r0 = 42;					\
+	*(u8*)(r1  %[shrt_min]) = r0;			\
+	r0 = *(u8*)(r1  %[shrt_min]);			\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, -((1 << 29) - 1)),
+	  __imm_const(shrt_min, SHRT_MIN)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK mixed reg/k, 1")
+__success __success_unpriv __retval(42)
+__naked void stack_mixed_reg_k_1(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -3;					\
+	r2 = -3;					\
+	r1 += r2;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK mixed reg/k, 2")
+__success __success_unpriv __retval(42)
+__naked void stack_mixed_reg_k_2(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r0;				\
+	r0 = 0;						\
+	*(u64*)(r10 - 16) = r0;				\
+	r1 = r10;					\
+	r1 += -3;					\
+	r2 = -3;					\
+	r1 += r2;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r5 = r10;					\
+	r0 = *(u8*)(r5 - 6);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK mixed reg/k, 3")
+__success __success_unpriv __retval(-3)
+__naked void stack_mixed_reg_k_3(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r1 += -3;					\
+	r2 = -3;					\
+	r1 += r2;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("PTR_TO_STACK reg")
+__success __success_unpriv __retval(42)
+__naked void ptr_to_stack_reg(void)
+{
+	asm volatile ("					\
+	r1 = r10;					\
+	r2 = -3;					\
+	r1 += r2;					\
+	r0 = 42;					\
+	*(u8*)(r1 + 0) = r0;				\
+	r0 = *(u8*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("stack pointer arithmetic")
+__success __success_unpriv __retval(0)
+__naked void stack_pointer_arithmetic(void)
+{
+	asm volatile ("					\
+	r1 = 4;						\
+	goto l0_%=;					\
+l0_%=:	r7 = r10;					\
+	r7 += -10;					\
+	r7 += -10;					\
+	r2 = r7;					\
+	r2 += r1;					\
+	r0 = 0;						\
+	*(u32*)(r2 + 4) = r0;				\
+	r2 = r7;					\
+	r2 += 8;					\
+	r0 = 0;						\
+	*(u32*)(r2 + 4) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("store PTR_TO_STACK in R10 to array map using BPF_B")
+__success __retval(42)
+__naked void array_map_using_bpf_b(void)
+{
+	asm volatile ("					\
+	/* Load pointer to map. */			\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	r0 = 2;						\
+	exit;						\
+l0_%=:	r1 = r0;					\
+	/* Copy R10 to R9. */				\
+	r9 = r10;					\
+	/* Pollute other registers with unaligned values. */\
+	r2 = -1;					\
+	r3 = -1;					\
+	r4 = -1;					\
+	r5 = -1;					\
+	r6 = -1;					\
+	r7 = -1;					\
+	r8 = -1;					\
+	/* Store both R9 and R10 with BPF_B and read back. */\
+	*(u8*)(r1 + 0) = r10;				\
+	r2 = *(u8*)(r1 + 0);				\
+	*(u8*)(r1 + 0) = r9;				\
+	r3 = *(u8*)(r1 + 0);				\
+	/* Should read back as same value. */		\
+	if r2 == r3 goto l1_%=;				\
+	r0 = 1;						\
+	exit;						\
+l1_%=:	r0 = 42;					\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/stack_ptr.c b/tools/testing/selftests/bpf/verifier/stack_ptr.c
deleted file mode 100644
index 8ab94d65f3d5..000000000000
--- a/tools/testing/selftests/bpf/verifier/stack_ptr.c
+++ /dev/null
@@ -1,359 +0,0 @@
-{
-	"PTR_TO_STACK store/load",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -10),
-	BPF_ST_MEM(BPF_DW, BPF_REG_1, 2, 0xfaceb00c),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, 2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0xfaceb00c,
-},
-{
-	"PTR_TO_STACK store/load - bad alignment on off",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_1, 2, 0xfaceb00c),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, 2),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "misaligned stack access off (0x0; 0x0)+-8+2 size 8",
-},
-{
-	"PTR_TO_STACK store/load - bad alignment on reg",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -10),
-	BPF_ST_MEM(BPF_DW, BPF_REG_1, 8, 0xfaceb00c),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, 8),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "misaligned stack access off (0x0; 0x0)+-10+8 size 8",
-},
-{
-	"PTR_TO_STACK store/load - out of bounds low",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -80000),
-	BPF_ST_MEM(BPF_DW, BPF_REG_1, 8, 0xfaceb00c),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, 8),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid write to stack R1 off=-79992 size=8",
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-},
-{
-	"PTR_TO_STACK store/load - out of bounds high",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_1, 8, 0xfaceb00c),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, 8),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid write to stack R1 off=0 size=8",
-},
-{
-	"PTR_TO_STACK check high 1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -1),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK check high 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, -1, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, -1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK check high 3",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, -1, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, -1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK check high 4",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "invalid write to stack R1 off=0 size=1",
-	.result = REJECT,
-},
-{
-	"PTR_TO_STACK check high 5",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, (1 << 29) - 1),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "invalid write to stack R1",
-},
-{
-	"PTR_TO_STACK check high 6",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, (1 << 29) - 1),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, SHRT_MAX, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, SHRT_MAX),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "invalid write to stack",
-},
-{
-	"PTR_TO_STACK check high 7",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, (1 << 29) - 1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, (1 << 29) - 1),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, SHRT_MAX, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, SHRT_MAX),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "fp pointer offset",
-},
-{
-	"PTR_TO_STACK check low 1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -512),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK check low 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -513),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 1, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK check low 3",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -513),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "invalid write to stack R1 off=-513 size=1",
-	.result = REJECT,
-},
-{
-	"PTR_TO_STACK check low 4",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, INT_MIN),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "math between fp pointer",
-},
-{
-	"PTR_TO_STACK check low 5",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -((1 << 29) - 1)),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "invalid write to stack",
-},
-{
-	"PTR_TO_STACK check low 6",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -((1 << 29) - 1)),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, SHRT_MIN, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, SHRT_MIN),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid write to stack",
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-},
-{
-	"PTR_TO_STACK check low 7",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -((1 << 29) - 1)),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -((1 << 29) - 1)),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, SHRT_MIN, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, SHRT_MIN),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.errstr = "fp pointer offset",
-},
-{
-	"PTR_TO_STACK mixed reg/k, 1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -3),
-	BPF_MOV64_IMM(BPF_REG_2, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_2),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK mixed reg/k, 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -16, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -3),
-	BPF_MOV64_IMM(BPF_REG_2, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_2),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_10),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_5, -6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"PTR_TO_STACK mixed reg/k, 3",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -3),
-	BPF_MOV64_IMM(BPF_REG_2, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_2),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = -3,
-},
-{
-	"PTR_TO_STACK reg",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_MOV64_IMM(BPF_REG_2, -3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_2),
-	BPF_ST_MEM(BPF_B, BPF_REG_1, 0, 42),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"stack pointer arithmetic",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 0),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, -10),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_1),
-	BPF_ST_MEM(0, BPF_REG_2, 4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 8),
-	BPF_ST_MEM(0, BPF_REG_2, 4, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-},
-{
-	"store PTR_TO_STACK in R10 to array map using BPF_B",
-	.insns = {
-	/* Load pointer to map. */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 2),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	/* Copy R10 to R9. */
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_10),
-	/* Pollute other registers with unaligned values. */
-	BPF_MOV64_IMM(BPF_REG_2, -1),
-	BPF_MOV64_IMM(BPF_REG_3, -1),
-	BPF_MOV64_IMM(BPF_REG_4, -1),
-	BPF_MOV64_IMM(BPF_REG_5, -1),
-	BPF_MOV64_IMM(BPF_REG_6, -1),
-	BPF_MOV64_IMM(BPF_REG_7, -1),
-	BPF_MOV64_IMM(BPF_REG_8, -1),
-	/* Store both R9 and R10 with BPF_B and read back. */
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_10, 0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_2, BPF_REG_1, 0),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_9, 0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_3, BPF_REG_1, 0),
-	/* Should read back as same value. */
-	BPF_JMP_REG(BPF_JEQ, BPF_REG_2, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 42,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-- 
cgit v1.2.3-70-g09d2


From ab839a58194633bd1c5ff42e6a9de87a746a67aa Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:17 +0200
Subject: selftests/bpf: verifier/uninit.c converted to inline assembly

Test verifier/uninit.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-37-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../testing/selftests/bpf/progs/verifier_uninit.c  | 61 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/uninit.c      | 39 --------------
 3 files changed, 63 insertions(+), 39 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_uninit.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/uninit.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index ce1ca8c0c02e..c6e69b3827dc 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -33,6 +33,7 @@
 #include "verifier_ringbuf.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
+#include "verifier_uninit.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -88,3 +89,4 @@ void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
+void test_verifier_uninit(void)               { RUN(verifier_uninit); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_uninit.c b/tools/testing/selftests/bpf/progs/verifier_uninit.c
new file mode 100644
index 000000000000..7718cd7d19ce
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_uninit.c
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/uninit.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "../../../include/linux/filter.h"
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("read uninitialized register")
+__failure __msg("R2 !read_ok")
+__failure_unpriv
+__naked void read_uninitialized_register(void)
+{
+	asm volatile ("					\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("read invalid register")
+__failure __msg("R15 is invalid")
+__failure_unpriv
+__naked void read_invalid_register(void)
+{
+	asm volatile ("					\
+	.8byte %[mov64_reg];				\
+	exit;						\
+"	:
+	: __imm_insn(mov64_reg, BPF_MOV64_REG(BPF_REG_0, -1))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("program doesn't init R0 before exit")
+__failure __msg("R0 !read_ok")
+__failure_unpriv
+__naked void t_init_r0_before_exit(void)
+{
+	asm volatile ("					\
+	r2 = r1;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("program doesn't init R0 before exit in all branches")
+__failure __msg("R0 !read_ok")
+__msg_unpriv("R1 pointer comparison")
+__naked void before_exit_in_all_branches(void)
+{
+	asm volatile ("					\
+	if r1 >= 0 goto l0_%=;				\
+	r0 = 1;						\
+	r0 += 2;					\
+l0_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/uninit.c b/tools/testing/selftests/bpf/verifier/uninit.c
deleted file mode 100644
index 987a5871ff1d..000000000000
--- a/tools/testing/selftests/bpf/verifier/uninit.c
+++ /dev/null
@@ -1,39 +0,0 @@
-{
-	"read uninitialized register",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R2 !read_ok",
-	.result = REJECT,
-},
-{
-	"read invalid register",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_0, -1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R15 is invalid",
-	.result = REJECT,
-},
-{
-	"program doesn't init R0 before exit",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_2, BPF_REG_1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 !read_ok",
-	.result = REJECT,
-},
-{
-	"program doesn't init R0 before exit in all branches",
-	.insns = {
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 2),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 !read_ok",
-	.errstr_unpriv = "R1 pointer comparison",
-	.result = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From 033914942da4696dcd9009ba88bc1bba06706549 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:18 +0200
Subject: selftests/bpf: verifier/value_adj_spill.c converted to inline
 assembly

Test verifier/value_adj_spill.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-38-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../selftests/bpf/progs/verifier_value_adj_spill.c | 78 ++++++++++++++++++++++
 .../selftests/bpf/verifier/value_adj_spill.c       | 43 ------------
 3 files changed, 80 insertions(+), 43 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_value_adj_spill.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/value_adj_spill.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index c6e69b3827dc..825c8583fecf 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -34,6 +34,7 @@
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_uninit.skel.h"
+#include "verifier_value_adj_spill.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -90,3 +91,4 @@ void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
+void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_value_adj_spill.c b/tools/testing/selftests/bpf/progs/verifier_value_adj_spill.c
new file mode 100644
index 000000000000..d7a5ba9bbe6a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_value_adj_spill.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/value_adj_spill.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("socket")
+__description("map element value is preserved across register spilling")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0)
+__naked void is_preserved_across_register_spilling(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 42;					\
+	*(u64*)(r0 + 0) = r1;				\
+	r1 = r10;					\
+	r1 += -184;					\
+	*(u64*)(r1 + 0) = r0;				\
+	r3 = *(u64*)(r1 + 0);				\
+	r1 = 42;					\
+	*(u64*)(r3 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value or null is marked on register spilling")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0)
+__naked void is_marked_on_register_spilling(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	r1 = r10;					\
+	r1 += -152;					\
+	*(u64*)(r1 + 0) = r0;				\
+	if r0 == 0 goto l0_%=;				\
+	r3 = *(u64*)(r1 + 0);				\
+	r1 = 42;					\
+	*(u64*)(r3 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/value_adj_spill.c b/tools/testing/selftests/bpf/verifier/value_adj_spill.c
deleted file mode 100644
index 7135e8021b81..000000000000
--- a/tools/testing/selftests/bpf/verifier/value_adj_spill.c
+++ /dev/null
@@ -1,43 +0,0 @@
-{
-	"map element value is preserved across register spilling",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 42),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -184),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_3, 0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-},
-{
-	"map element value or null is marked on register spilling",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -152),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_3, 0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From 8f59e87a3bc6a5618f0ed459f1d36838c16bcad7 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:19 +0200
Subject: selftests/bpf: verifier/value.c converted to inline assembly

Test verifier/value.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-39-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 tools/testing/selftests/bpf/progs/verifier_value.c | 158 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/value.c       | 104 --------------
 3 files changed, 160 insertions(+), 104 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_value.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/value.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 825c8583fecf..c77df746d650 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -35,6 +35,7 @@
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_uninit.skel.h"
 #include "verifier_value_adj_spill.skel.h"
+#include "verifier_value.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -92,3 +93,4 @@ void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
+void test_verifier_value(void)                { RUN(verifier_value); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_value.c b/tools/testing/selftests/bpf/progs/verifier_value.c
new file mode 100644
index 000000000000..b5af6b6f5acd
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_value.c
@@ -0,0 +1,158 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/value.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("socket")
+__description("map element value store of cleared call register")
+__failure __msg("R1 !read_ok")
+__failure_unpriv __msg_unpriv("R1 !read_ok")
+__naked void store_of_cleared_call_register(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value with unaligned store")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void element_value_with_unaligned_store(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 += 3;					\
+	r1 = 42;					\
+	*(u64*)(r0 + 0) = r1;				\
+	r1 = 43;					\
+	*(u64*)(r0 + 2) = r1;				\
+	r1 = 44;					\
+	*(u64*)(r0 - 2) = r1;				\
+	r8 = r0;					\
+	r1 = 32;					\
+	*(u64*)(r8 + 0) = r1;				\
+	r1 = 33;					\
+	*(u64*)(r8 + 2) = r1;				\
+	r1 = 34;					\
+	*(u64*)(r8 - 2) = r1;				\
+	r8 += 5;					\
+	r1 = 22;					\
+	*(u64*)(r8 + 0) = r1;				\
+	r1 = 23;					\
+	*(u64*)(r8 + 4) = r1;				\
+	r1 = 24;					\
+	*(u64*)(r8 - 7) = r1;				\
+	r7 = r8;					\
+	r7 += 3;					\
+	r1 = 22;					\
+	*(u64*)(r7 + 0) = r1;				\
+	r1 = 23;					\
+	*(u64*)(r7 + 4) = r1;				\
+	r1 = 24;					\
+	*(u64*)(r7 - 4) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value with unaligned load")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void element_value_with_unaligned_load(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	if r1 >= %[max_entries] goto l0_%=;		\
+	r0 += 3;					\
+	r7 = *(u64*)(r0 + 0);				\
+	r7 = *(u64*)(r0 + 2);				\
+	r8 = r0;					\
+	r7 = *(u64*)(r8 + 0);				\
+	r7 = *(u64*)(r8 + 2);				\
+	r0 += 5;					\
+	r7 = *(u64*)(r0 + 0);				\
+	r7 = *(u64*)(r0 + 4);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(max_entries, MAX_ENTRIES)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value is preserved across register spilling")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void is_preserved_across_register_spilling(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 += %[test_val_foo];				\
+	r1 = 42;					\
+	*(u64*)(r0 + 0) = r1;				\
+	r1 = r10;					\
+	r1 += -184;					\
+	*(u64*)(r1 + 0) = r0;				\
+	r3 = *(u64*)(r1 + 0);				\
+	r1 = 42;					\
+	*(u64*)(r3 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/value.c b/tools/testing/selftests/bpf/verifier/value.c
deleted file mode 100644
index 0e42592b1218..000000000000
--- a/tools/testing/selftests/bpf/verifier/value.c
+++ /dev/null
@@ -1,104 +0,0 @@
-{
-	"map element value store of cleared call register",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R1 !read_ok",
-	.errstr = "R1 !read_ok",
-	.result = REJECT,
-	.result_unpriv = REJECT,
-},
-{
-	"map element value with unaligned store",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 17),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 3),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 42),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 2, 43),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, -2, 44),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_8, 0, 32),
-	BPF_ST_MEM(BPF_DW, BPF_REG_8, 2, 33),
-	BPF_ST_MEM(BPF_DW, BPF_REG_8, -2, 34),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_8, 5),
-	BPF_ST_MEM(BPF_DW, BPF_REG_8, 0, 22),
-	BPF_ST_MEM(BPF_DW, BPF_REG_8, 4, 23),
-	BPF_ST_MEM(BPF_DW, BPF_REG_8, -7, 24),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_8),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_7, 3),
-	BPF_ST_MEM(BPF_DW, BPF_REG_7, 0, 22),
-	BPF_ST_MEM(BPF_DW, BPF_REG_7, 4, 23),
-	BPF_ST_MEM(BPF_DW, BPF_REG_7, -4, 24),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map element value with unaligned load",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_1, MAX_ENTRIES, 9),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 3),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 2),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_8, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_8, 2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 5),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_0, 4),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map element value is preserved across register spilling",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, offsetof(struct test_val, foo)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 42),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -184),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_1, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_3, 0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-- 
cgit v1.2.3-70-g09d2


From d330528617b78ae893c405bf06c6c604bd2e6357 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:20 +0200
Subject: selftests/bpf: verifier/value_or_null.c converted to inline assembly

Test verifier/value_or_null.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-40-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_value_or_null.c   | 288 +++++++++++++++++++++
 .../testing/selftests/bpf/verifier/value_or_null.c | 220 ----------------
 3 files changed, 290 insertions(+), 220 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_value_or_null.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/value_or_null.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index c77df746d650..54eb21ef9fad 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -36,6 +36,7 @@
 #include "verifier_uninit.skel.h"
 #include "verifier_value_adj_spill.skel.h"
 #include "verifier_value.skel.h"
+#include "verifier_value_or_null.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -94,3 +95,4 @@ void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
 void test_verifier_value(void)                { RUN(verifier_value); }
+void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_value_or_null.c b/tools/testing/selftests/bpf/progs/verifier_value_or_null.c
new file mode 100644
index 000000000000..8ff668a242eb
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_value_or_null.c
@@ -0,0 +1,288 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/value_or_null.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("tc")
+__description("multiple registers share map_lookup_elem result")
+__success __retval(0)
+__naked void share_map_lookup_elem_result(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r4 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("alu ops on ptr_to_map_value_or_null, 1")
+__failure __msg("R4 pointer arithmetic on map_value_or_null")
+__naked void map_value_or_null_1(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r4 = r0;					\
+	r4 += -2;					\
+	r4 += 2;					\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("alu ops on ptr_to_map_value_or_null, 2")
+__failure __msg("R4 pointer arithmetic on map_value_or_null")
+__naked void map_value_or_null_2(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r4 = r0;					\
+	r4 &= -1;					\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("alu ops on ptr_to_map_value_or_null, 3")
+__failure __msg("R4 pointer arithmetic on map_value_or_null")
+__naked void map_value_or_null_3(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r4 = r0;					\
+	r4 <<= 1;					\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("invalid memory access with multiple map_lookup_elem calls")
+__failure __msg("R4 !read_ok")
+__naked void multiple_map_lookup_elem_calls(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	r8 = r1;					\
+	r7 = r2;					\
+	call %[bpf_map_lookup_elem];			\
+	r4 = r0;					\
+	r1 = r8;					\
+	r2 = r7;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("valid indirect map_lookup_elem access with 2nd lookup in branch")
+__success __retval(0)
+__naked void with_2nd_lookup_in_branch(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	r8 = r1;					\
+	r7 = r2;					\
+	call %[bpf_map_lookup_elem];			\
+	r2 = 10;					\
+	if r2 != 0 goto l0_%=;				\
+	r1 = r8;					\
+	r2 = r7;					\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r4 = r0;					\
+	if r0 == 0 goto l1_%=;				\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid map access from else condition")
+__failure __msg("R0 unbounded memory access")
+__failure_unpriv __msg_unpriv("R0 leaks addr")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void map_access_from_else_condition(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	if r1 >= %[__imm_0] goto l1_%=;			\
+	r1 += 1;					\
+l1_%=:	r1 <<= 2;					\
+	r0 += r1;					\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, MAX_ENTRIES-1),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("map lookup and null branch prediction")
+__success __retval(0)
+__naked void lookup_and_null_branch_prediction(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r6 = r0;					\
+	if r6 == 0 goto l0_%=;				\
+	if r6 != 0 goto l0_%=;				\
+	r10 += 10;					\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("MAP_VALUE_OR_NULL check_ids() in regsafe()")
+__failure __msg("R8 invalid mem access 'map_value_or_null'")
+__failure_unpriv __msg_unpriv("")
+__flag(BPF_F_TEST_STATE_FREQ)
+__naked void null_check_ids_in_regsafe(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	/* r9 = map_lookup_elem(...) */			\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r9 = r0;					\
+	/* r8 = map_lookup_elem(...) */			\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r8 = r0;					\
+	/* r7 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r7 = r0;					\
+	/* r6 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r6 = r0;					\
+	/* if r6 > r7 goto +1    ; no new information about the state is derived from\
+	 *                       ; this check, thus produced verifier states differ\
+	 *                       ; only in 'insn_idx'	\
+	 * r9 = r8               ; optionally share ID between r9 and r8\
+	 */						\
+	if r6 > r7 goto l0_%=;				\
+	r9 = r8;					\
+l0_%=:	/* if r9 == 0 goto <exit> */			\
+	if r9 == 0 goto l1_%=;				\
+	/* read map value via r8, this is not always	\
+	 * safe because r8 might be not equal to r9.	\
+	 */						\
+	r0 = *(u64*)(r8 + 0);				\
+l1_%=:	/* exit 0 */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/value_or_null.c b/tools/testing/selftests/bpf/verifier/value_or_null.c
deleted file mode 100644
index 52a8bca14f03..000000000000
--- a/tools/testing/selftests/bpf/verifier/value_or_null.c
+++ /dev/null
@@ -1,220 +0,0 @@
-{
-	"multiple registers share map_lookup_elem result",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS
-},
-{
-	"alu ops on ptr_to_map_value_or_null, 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 2),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.errstr = "R4 pointer arithmetic on map_value_or_null",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS
-},
-{
-	"alu ops on ptr_to_map_value_or_null, 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_4, -1),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.errstr = "R4 pointer arithmetic on map_value_or_null",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS
-},
-{
-	"alu ops on ptr_to_map_value_or_null, 3",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_4, 1),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.errstr = "R4 pointer arithmetic on map_value_or_null",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS
-},
-{
-	"invalid memory access with multiple map_lookup_elem calls",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_2),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.result = REJECT,
-	.errstr = "R4 !read_ok",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS
-},
-{
-	"valid indirect map_lookup_elem access with 2nd lookup in branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_2),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_2, 10),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_2, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS
-},
-{
-	"invalid map access from else condition",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_1, MAX_ENTRIES-1, 1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 unbounded memory access",
-	.result = REJECT,
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map lookup and null branch prediction",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 2),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_6, 0, 1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_10, 10),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"MAP_VALUE_OR_NULL check_ids() in regsafe()",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* r9 = map_lookup_elem(...) */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_0),
-	/* r8 = map_lookup_elem(...) */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	/* r7 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r6 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* if r6 > r7 goto +1    ; no new information about the state is derived from
-	 *                       ; this check, thus produced verifier states differ
-	 *                       ; only in 'insn_idx'
-	 * r9 = r8               ; optionally share ID between r9 and r8
-	 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_7, 1),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_8),
-	/* if r9 == 0 goto <exit> */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_9, 0, 1),
-	/* read map value via r8, this is not always
-	 * safe because r8 might be not equal to r9.
-	 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_8, 0),
-	/* exit 0 */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.flags = BPF_F_TEST_STATE_FREQ,
-	.fixup_map_hash_8b = { 3, 9 },
-	.result = REJECT,
-	.errstr = "R8 invalid mem access 'map_value_or_null'",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-- 
cgit v1.2.3-70-g09d2


From d15f5b68b63ad7f47e05aba33c4794fb3bfaf1af Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:21 +0200
Subject: selftests/bpf: verifier/var_off.c converted to inline assembly

Test verifier/var_off.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-41-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_var_off.c | 349 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/var_off.c     | 291 -----------------
 3 files changed, 351 insertions(+), 291 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_var_off.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/var_off.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 54eb21ef9fad..44350e328da2 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -37,6 +37,7 @@
 #include "verifier_value_adj_spill.skel.h"
 #include "verifier_value.skel.h"
 #include "verifier_value_or_null.skel.h"
+#include "verifier_var_off.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -96,3 +97,4 @@ void test_verifier_uninit(void)               { RUN(verifier_uninit); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
 void test_verifier_value(void)                { RUN(verifier_value); }
 void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
+void test_verifier_var_off(void)              { RUN(verifier_var_off); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_var_off.c b/tools/testing/selftests/bpf/progs/verifier_var_off.c
new file mode 100644
index 000000000000..83a90afba785
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_var_off.c
@@ -0,0 +1,349 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/var_off.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("lwt_in")
+__description("variable-offset ctx access")
+__failure __msg("variable ctx access var_off=(0x0; 0x4)")
+__naked void variable_offset_ctx_access(void)
+{
+	asm volatile ("					\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned */		\
+	r2 &= 4;					\
+	/* add it to skb.  We now have either &skb->len or\
+	 * &skb->pkt_type, but we don't know which	\
+	 */						\
+	r1 += r2;					\
+	/* dereference it */				\
+	r0 = *(u32*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("variable-offset stack read, priv vs unpriv")
+__success __failure_unpriv
+__msg_unpriv("R2 variable stack access prohibited for !root")
+__retval(0)
+__naked void stack_read_priv_vs_unpriv(void)
+{
+	asm volatile ("					\
+	/* Fill the top 8 bytes of the stack */		\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r0;				\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned */		\
+	r2 &= 4;					\
+	r2 -= 8;					\
+	/* add it to fp.  We now have either fp-4 or fp-8, but\
+	 * we don't know which				\
+	 */						\
+	r2 += r10;					\
+	/* dereference it for a stack read */		\
+	r0 = *(u32*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("variable-offset stack read, uninitialized")
+__failure __msg("invalid variable-offset read from stack R2")
+__naked void variable_offset_stack_read_uninitialized(void)
+{
+	asm volatile ("					\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned */		\
+	r2 &= 4;					\
+	r2 -= 8;					\
+	/* add it to fp.  We now have either fp-4 or fp-8, but\
+	 * we don't know which				\
+	 */						\
+	r2 += r10;					\
+	/* dereference it for a stack read */		\
+	r0 = *(u32*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("variable-offset stack write, priv vs unpriv")
+__success __failure_unpriv
+/* Variable stack access is rejected for unprivileged.
+ */
+__msg_unpriv("R2 variable stack access prohibited for !root")
+__retval(0)
+__naked void stack_write_priv_vs_unpriv(void)
+{
+	asm volatile ("					\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 8-byte aligned */		\
+	r2 &= 8;					\
+	r2 -= 16;					\
+	/* Add it to fp.  We now have either fp-8 or fp-16, but\
+	 * we don't know which				\
+	 */						\
+	r2 += r10;					\
+	/* Dereference it for a stack write */		\
+	r0 = 0;						\
+	*(u64*)(r2 + 0) = r0;				\
+	/* Now read from the address we just wrote. This shows\
+	 * that, after a variable-offset write, a priviledged\
+	 * program can read the slots that were in the range of\
+	 * that write (even if the verifier doesn't actually know\
+	 * if the slot being read was really written to or not.\
+	 */						\
+	r3 = *(u64*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("variable-offset stack write clobbers spilled regs")
+__failure
+/* In the priviledged case, dereferencing a spilled-and-then-filled
+ * register is rejected because the previous variable offset stack
+ * write might have overwritten the spilled pointer (i.e. we lose track
+ * of the spilled register when we analyze the write).
+ */
+__msg("R2 invalid mem access 'scalar'")
+__failure_unpriv
+/* The unprivileged case is not too interesting; variable
+ * stack access is rejected.
+ */
+__msg_unpriv("R2 variable stack access prohibited for !root")
+__naked void stack_write_clobbers_spilled_regs(void)
+{
+	asm volatile ("					\
+	/* Dummy instruction; needed because we need to patch the next one\
+	 * and we can't patch the first instruction.	\
+	 */						\
+	r6 = 0;						\
+	/* Make R0 a map ptr */				\
+	r0 = %[map_hash_8b] ll;				\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 8-byte aligned */		\
+	r2 &= 8;					\
+	r2 -= 16;					\
+	/* Add it to fp. We now have either fp-8 or fp-16, but\
+	 * we don't know which.				\
+	 */						\
+	r2 += r10;					\
+	/* Spill R0(map ptr) into stack */		\
+	*(u64*)(r10 - 8) = r0;				\
+	/* Dereference the unknown value for a stack write */\
+	r0 = 0;						\
+	*(u64*)(r2 + 0) = r0;				\
+	/* Fill the register back into R2 */		\
+	r2 = *(u64*)(r10 - 8);				\
+	/* Try to dereference R2 for a memory load */	\
+	r0 = *(u64*)(r2 + 8);				\
+	exit;						\
+"	:
+	: __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("sockops")
+__description("indirect variable-offset stack access, unbounded")
+__failure __msg("invalid unbounded variable-offset indirect access to stack R4")
+__naked void variable_offset_stack_access_unbounded(void)
+{
+	asm volatile ("					\
+	r2 = 6;						\
+	r3 = 28;					\
+	/* Fill the top 16 bytes of the stack. */	\
+	r4 = 0;						\
+	*(u64*)(r10 - 16) = r4;				\
+	r4 = 0;						\
+	*(u64*)(r10 - 8) = r4;				\
+	/* Get an unknown value. */			\
+	r4 = *(u64*)(r1 + %[bpf_sock_ops_bytes_received]);\
+	/* Check the lower bound but don't check the upper one. */\
+	if r4 s< 0 goto l0_%=;				\
+	/* Point the lower bound to initialized stack. Offset is now in range\
+	 * from fp-16 to fp+0x7fffffffffffffef, i.e. max value is unbounded.\
+	 */						\
+	r4 -= 16;					\
+	r4 += r10;					\
+	r5 = 8;						\
+	/* Dereference it indirectly. */		\
+	call %[bpf_getsockopt];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_getsockopt),
+	  __imm_const(bpf_sock_ops_bytes_received, offsetof(struct bpf_sock_ops, bytes_received))
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("indirect variable-offset stack access, max out of bound")
+__failure __msg("invalid variable-offset indirect access to stack R2")
+__naked void access_max_out_of_bound(void)
+{
+	asm volatile ("					\
+	/* Fill the top 8 bytes of the stack */		\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned */		\
+	r2 &= 4;					\
+	r2 -= 8;					\
+	/* add it to fp.  We now have either fp-4 or fp-8, but\
+	 * we don't know which				\
+	 */						\
+	r2 += r10;					\
+	/* dereference it indirectly */			\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("indirect variable-offset stack access, min out of bound")
+__failure __msg("invalid variable-offset indirect access to stack R2")
+__naked void access_min_out_of_bound(void)
+{
+	asm volatile ("					\
+	/* Fill the top 8 bytes of the stack */		\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned */		\
+	r2 &= 4;					\
+	r2 -= 516;					\
+	/* add it to fp.  We now have either fp-516 or fp-512, but\
+	 * we don't know which				\
+	 */						\
+	r2 += r10;					\
+	/* dereference it indirectly */			\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("indirect variable-offset stack access, min_off < min_initialized")
+__failure __msg("invalid indirect read from stack R2 var_off")
+__naked void access_min_off_min_initialized(void)
+{
+	asm volatile ("					\
+	/* Fill only the top 8 bytes of the stack. */	\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned. */		\
+	r2 &= 4;					\
+	r2 -= 16;					\
+	/* Add it to fp.  We now have either fp-12 or fp-16, but we don't know\
+	 * which. fp-16 size 8 is partially uninitialized stack.\
+	 */						\
+	r2 += r10;					\
+	/* Dereference it indirectly. */		\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("indirect variable-offset stack access, priv vs unpriv")
+__success __failure_unpriv
+__msg_unpriv("R2 variable stack access prohibited for !root")
+__retval(0)
+__naked void stack_access_priv_vs_unpriv(void)
+{
+	asm volatile ("					\
+	/* Fill the top 16 bytes of the stack. */	\
+	r2 = 0;						\
+	*(u64*)(r10 - 16) = r2;				\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	/* Get an unknown value. */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned. */		\
+	r2 &= 4;					\
+	r2 -= 16;					\
+	/* Add it to fp.  We now have either fp-12 or fp-16, we don't know\
+	 * which, but either way it points to initialized stack.\
+	 */						\
+	r2 += r10;					\
+	/* Dereference it indirectly. */		\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("indirect variable-offset stack access, ok")
+__success __retval(0)
+__naked void variable_offset_stack_access_ok(void)
+{
+	asm volatile ("					\
+	/* Fill the top 16 bytes of the stack. */	\
+	r2 = 0;						\
+	*(u64*)(r10 - 16) = r2;				\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	/* Get an unknown value. */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* Make it small and 4-byte aligned. */		\
+	r2 &= 4;					\
+	r2 -= 16;					\
+	/* Add it to fp.  We now have either fp-12 or fp-16, we don't know\
+	 * which, but either way it points to initialized stack.\
+	 */						\
+	r2 += r10;					\
+	/* Dereference it indirectly. */		\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/var_off.c b/tools/testing/selftests/bpf/verifier/var_off.c
deleted file mode 100644
index b183e26c03f1..000000000000
--- a/tools/testing/selftests/bpf/verifier/var_off.c
+++ /dev/null
@@ -1,291 +0,0 @@
-{
-	"variable-offset ctx access",
-	.insns = {
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	/* add it to skb.  We now have either &skb->len or
-	 * &skb->pkt_type, but we don't know which
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_2),
-	/* dereference it */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "variable ctx access var_off=(0x0; 0x4)",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"variable-offset stack read, priv vs unpriv",
-	.insns = {
-	/* Fill the top 8 bytes of the stack */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 8),
-	/* add it to fp.  We now have either fp-4 or fp-8, but
-	 * we don't know which
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* dereference it for a stack read */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R2 variable stack access prohibited for !root",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"variable-offset stack read, uninitialized",
-	.insns = {
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 8),
-	/* add it to fp.  We now have either fp-4 or fp-8, but
-	 * we don't know which
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* dereference it for a stack read */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid variable-offset read from stack R2",
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"variable-offset stack write, priv vs unpriv",
-	.insns = {
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 8-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 8),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 16),
-	/* Add it to fp.  We now have either fp-8 or fp-16, but
-	 * we don't know which
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* Dereference it for a stack write */
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	/* Now read from the address we just wrote. This shows
-	 * that, after a variable-offset write, a priviledged
-	 * program can read the slots that were in the range of
-	 * that write (even if the verifier doesn't actually know
-	 * if the slot being read was really written to or not.
-	 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	/* Variable stack access is rejected for unprivileged.
-	 */
-	.errstr_unpriv = "R2 variable stack access prohibited for !root",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"variable-offset stack write clobbers spilled regs",
-	.insns = {
-	/* Dummy instruction; needed because we need to patch the next one
-	 * and we can't patch the first instruction.
-	 */
-	BPF_MOV64_IMM(BPF_REG_6, 0),
-	/* Make R0 a map ptr */
-	BPF_LD_MAP_FD(BPF_REG_0, 0),
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 8-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 8),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 16),
-	/* Add it to fp. We now have either fp-8 or fp-16, but
-	 * we don't know which.
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* Spill R0(map ptr) into stack */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	/* Dereference the unknown value for a stack write */
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	/* Fill the register back into R2 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -8),
-	/* Try to dereference R2 for a memory load */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, 8),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	/* The unprivileged case is not too interesting; variable
-	 * stack access is rejected.
-	 */
-	.errstr_unpriv = "R2 variable stack access prohibited for !root",
-	.result_unpriv = REJECT,
-	/* In the priviledged case, dereferencing a spilled-and-then-filled
-	 * register is rejected because the previous variable offset stack
-	 * write might have overwritten the spilled pointer (i.e. we lose track
-	 * of the spilled register when we analyze the write).
-	 */
-	.errstr = "R2 invalid mem access 'scalar'",
-	.result = REJECT,
-},
-{
-	"indirect variable-offset stack access, unbounded",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 6),
-	BPF_MOV64_IMM(BPF_REG_3, 28),
-	/* Fill the top 16 bytes of the stack. */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -16, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value. */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
-							   bytes_received)),
-	/* Check the lower bound but don't check the upper one. */
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_4, 0, 4),
-	/* Point the lower bound to initialized stack. Offset is now in range
-	 * from fp-16 to fp+0x7fffffffffffffef, i.e. max value is unbounded.
-	 */
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_4, 16),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_10),
-	BPF_MOV64_IMM(BPF_REG_5, 8),
-	/* Dereference it indirectly. */
-	BPF_EMIT_CALL(BPF_FUNC_getsockopt),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid unbounded variable-offset indirect access to stack R4",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SOCK_OPS,
-},
-{
-	"indirect variable-offset stack access, max out of bound",
-	.insns = {
-	/* Fill the top 8 bytes of the stack */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 8),
-	/* add it to fp.  We now have either fp-4 or fp-8, but
-	 * we don't know which
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* dereference it indirectly */
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "invalid variable-offset indirect access to stack R2",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"indirect variable-offset stack access, min out of bound",
-	.insns = {
-	/* Fill the top 8 bytes of the stack */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 516),
-	/* add it to fp.  We now have either fp-516 or fp-512, but
-	 * we don't know which
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* dereference it indirectly */
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "invalid variable-offset indirect access to stack R2",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"indirect variable-offset stack access, min_off < min_initialized",
-	.insns = {
-	/* Fill only the top 8 bytes of the stack. */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned. */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 16),
-	/* Add it to fp.  We now have either fp-12 or fp-16, but we don't know
-	 * which. fp-16 size 8 is partially uninitialized stack.
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* Dereference it indirectly. */
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 5 },
-	.errstr = "invalid indirect read from stack R2 var_off",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"indirect variable-offset stack access, priv vs unpriv",
-	.insns = {
-	/* Fill the top 16 bytes of the stack. */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -16, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value. */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned. */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 16),
-	/* Add it to fp.  We now have either fp-12 or fp-16, we don't know
-	 * which, but either way it points to initialized stack.
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* Dereference it indirectly. */
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 6 },
-	.errstr_unpriv = "R2 variable stack access prohibited for !root",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"indirect variable-offset stack access, ok",
-	.insns = {
-	/* Fill the top 16 bytes of the stack. */
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -16, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	/* Get an unknown value. */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* Make it small and 4-byte aligned. */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_2, 4),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_2, 16),
-	/* Add it to fp.  We now have either fp-12 or fp-16, we don't know
-	 * which, but either way it points to initialized stack.
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_10),
-	/* Dereference it indirectly. */
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 6 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-- 
cgit v1.2.3-70-g09d2


From a8036aea2d4f412c98a5fdbc0c987fa8a3c023ed Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:22 +0200
Subject: selftests/bpf: verifier/xadd.c converted to inline assembly

Test verifier/xadd.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-42-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   2 +
 tools/testing/selftests/bpf/progs/verifier_xadd.c | 124 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/xadd.c       |  97 -----------------
 3 files changed, 126 insertions(+), 97 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_xadd.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/xadd.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 44350e328da2..cd56fe520145 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -38,6 +38,7 @@
 #include "verifier_value.skel.h"
 #include "verifier_value_or_null.skel.h"
 #include "verifier_var_off.skel.h"
+#include "verifier_xadd.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -98,3 +99,4 @@ void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
 void test_verifier_value(void)                { RUN(verifier_value); }
 void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
 void test_verifier_var_off(void)              { RUN(verifier_var_off); }
+void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_xadd.c b/tools/testing/selftests/bpf/progs/verifier_xadd.c
new file mode 100644
index 000000000000..05a0a55adb45
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_xadd.c
@@ -0,0 +1,124 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/xadd.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("tc")
+__description("xadd/w check unaligned stack")
+__failure __msg("misaligned stack access off")
+__naked void xadd_w_check_unaligned_stack(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	*(u64*)(r10 - 8) = r0;				\
+	lock *(u32 *)(r10 - 7) += w0;			\
+	r0 = *(u64*)(r10 - 8);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("xadd/w check unaligned map")
+__failure __msg("misaligned value access off")
+__naked void xadd_w_check_unaligned_map(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = 1;						\
+	lock *(u32 *)(r0 + 3) += w1;			\
+	r0 = *(u32*)(r0 + 3);				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("xadd/w check unaligned pkt")
+__failure __msg("BPF_ATOMIC stores into R2 pkt is not allowed")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void xadd_w_check_unaligned_pkt(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	r0 = 99;					\
+	goto l1_%=;					\
+l0_%=:	r0 = 1;						\
+	r1 = 0;						\
+	*(u32*)(r2 + 0) = r1;				\
+	r1 = 0;						\
+	*(u32*)(r2 + 3) = r1;				\
+	lock *(u32 *)(r2 + 1) += w0;			\
+	lock *(u32 *)(r2 + 2) += w0;			\
+	r0 = *(u32*)(r2 + 1);				\
+l1_%=:	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("xadd/w check whether src/dst got mangled, 1")
+__success __retval(3)
+__naked void src_dst_got_mangled_1(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	r6 = r0;					\
+	r7 = r10;					\
+	*(u64*)(r10 - 8) = r0;				\
+	lock *(u64 *)(r10 - 8) += r0;			\
+	lock *(u64 *)(r10 - 8) += r0;			\
+	if r6 != r0 goto l0_%=;				\
+	if r7 != r10 goto l0_%=;			\
+	r0 = *(u64*)(r10 - 8);				\
+	exit;						\
+l0_%=:	r0 = 42;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("xadd/w check whether src/dst got mangled, 2")
+__success __retval(3)
+__naked void src_dst_got_mangled_2(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	r6 = r0;					\
+	r7 = r10;					\
+	*(u32*)(r10 - 8) = r0;				\
+	lock *(u32 *)(r10 - 8) += w0;			\
+	lock *(u32 *)(r10 - 8) += w0;			\
+	if r6 != r0 goto l0_%=;				\
+	if r7 != r10 goto l0_%=;			\
+	r0 = *(u32*)(r10 - 8);				\
+	exit;						\
+l0_%=:	r0 = 42;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/xadd.c b/tools/testing/selftests/bpf/verifier/xadd.c
deleted file mode 100644
index b96ef3526815..000000000000
--- a/tools/testing/selftests/bpf/verifier/xadd.c
+++ /dev/null
@@ -1,97 +0,0 @@
-{
-	"xadd/w check unaligned stack",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_10, BPF_REG_0, -7),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "misaligned stack access off",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"xadd/w check unaligned map",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_0, BPF_REG_1, 3),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 3),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = REJECT,
-	.errstr = "misaligned value access off",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"xadd/w check unaligned pkt",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 99),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 6),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_ST_MEM(BPF_W, BPF_REG_2, 0, 0),
-	BPF_ST_MEM(BPF_W, BPF_REG_2, 3, 0),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_2, BPF_REG_0, 1),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_2, BPF_REG_0, 2),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_2, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "BPF_ATOMIC stores into R2 pkt is not allowed",
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"xadd/w check whether src/dst got mangled, 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_10, BPF_REG_0, -8),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_10, BPF_REG_0, -8),
-	BPF_JMP_REG(BPF_JNE, BPF_REG_6, BPF_REG_0, 3),
-	BPF_JMP_REG(BPF_JNE, BPF_REG_7, BPF_REG_10, 2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 3,
-},
-{
-	"xadd/w check whether src/dst got mangled, 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_10),
-	BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -8),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_10, BPF_REG_0, -8),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_10, BPF_REG_0, -8),
-	BPF_JMP_REG(BPF_JNE, BPF_REG_6, BPF_REG_0, 3),
-	BPF_JMP_REG(BPF_JNE, BPF_REG_7, BPF_REG_10, 2),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_10, -8),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 42),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 3,
-},
-- 
cgit v1.2.3-70-g09d2


From ffb515c933a9e8000e50b03f76569ffb6ef4d39d Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Sat, 25 Mar 2023 04:55:23 +0200
Subject: selftests/bpf: verifier/xdp.c converted to inline assembly

Test verifier/xdp.c automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230325025524.144043-43-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |  2 ++
 tools/testing/selftests/bpf/progs/verifier_xdp.c  | 24 +++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/xdp.c        | 14 -------------
 3 files changed, 26 insertions(+), 14 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_xdp.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/xdp.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index cd56fe520145..a774d5b193f1 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -39,6 +39,7 @@
 #include "verifier_value_or_null.skel.h"
 #include "verifier_var_off.skel.h"
 #include "verifier_xadd.skel.h"
+#include "verifier_xdp.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -100,3 +101,4 @@ void test_verifier_value(void)                { RUN(verifier_value); }
 void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
 void test_verifier_var_off(void)              { RUN(verifier_var_off); }
 void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
+void test_verifier_xdp(void)                  { RUN(verifier_xdp); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_xdp.c b/tools/testing/selftests/bpf/progs/verifier_xdp.c
new file mode 100644
index 000000000000..50768ed179b3
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_xdp.c
@@ -0,0 +1,24 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/xdp.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("xdp")
+__description("XDP, using ifindex from netdev")
+__success __retval(1)
+__naked void xdp_using_ifindex_from_netdev(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r2 = *(u32*)(r1 + %[xdp_md_ingress_ifindex]);	\
+	if r2 < 1 goto l0_%=;				\
+	r0 = 1;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(xdp_md_ingress_ifindex, offsetof(struct xdp_md, ingress_ifindex))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/xdp.c b/tools/testing/selftests/bpf/verifier/xdp.c
deleted file mode 100644
index 5ac390508139..000000000000
--- a/tools/testing/selftests/bpf/verifier/xdp.c
+++ /dev/null
@@ -1,14 +0,0 @@
-{
-	"XDP, using ifindex from netdev",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, ingress_ifindex)),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_2, 1, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.retval = 1,
-},
-- 
cgit v1.2.3-70-g09d2


From e65a5c6edbc6ca4853e6076bd81db1a410592a09 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 22 Mar 2023 14:52:42 -0700
Subject: bpf: Add a few bpf mem allocator functions

This patch adds a few bpf mem allocator functions which will
be used in the bpf_local_storage in a later patch.

bpf_mem_cache_alloc_flags(..., gfp_t flags) is added. When the
flags == GFP_KERNEL, it will fallback to __alloc(..., GFP_KERNEL).
bpf_local_storage knows its running context is sleepable (GFP_KERNEL)
and provides a better guarantee on memory allocation.

bpf_local_storage has some uncommon cases that its selem
cannot be reused immediately. It handles its own
rcu_head and goes through a rcu_trace gp and then free it.
bpf_mem_cache_raw_free() is added for direct free purpose
without leaking the LLIST_NODE_SZ internal knowledge.
During free time, the 'struct bpf_mem_alloc *ma' is no longer
available. However, the caller should know if it is
percpu memory or not and it can call different raw_free functions.
bpf_local_storage does not support percpu value, so only
the non-percpu 'bpf_mem_cache_raw_free()' is added in
this patch.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230322215246.1675516-2-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_mem_alloc.h |  2 ++
 kernel/bpf/memalloc.c         | 59 ++++++++++++++++++++++++++++++++++++-------
 2 files changed, 52 insertions(+), 9 deletions(-)

diff --git a/include/linux/bpf_mem_alloc.h b/include/linux/bpf_mem_alloc.h
index a7104af61ab4..3929be5743f4 100644
--- a/include/linux/bpf_mem_alloc.h
+++ b/include/linux/bpf_mem_alloc.h
@@ -31,5 +31,7 @@ void bpf_mem_free(struct bpf_mem_alloc *ma, void *ptr);
 /* kmem_cache_alloc/free equivalent: */
 void *bpf_mem_cache_alloc(struct bpf_mem_alloc *ma);
 void bpf_mem_cache_free(struct bpf_mem_alloc *ma, void *ptr);
+void bpf_mem_cache_raw_free(void *ptr);
+void *bpf_mem_cache_alloc_flags(struct bpf_mem_alloc *ma, gfp_t flags);
 
 #endif /* _BPF_MEM_ALLOC_H */
diff --git a/kernel/bpf/memalloc.c b/kernel/bpf/memalloc.c
index 5fcdacbb8439..410637c225fb 100644
--- a/kernel/bpf/memalloc.c
+++ b/kernel/bpf/memalloc.c
@@ -121,15 +121,8 @@ static struct llist_node notrace *__llist_del_first(struct llist_head *head)
 	return entry;
 }
 
-static void *__alloc(struct bpf_mem_cache *c, int node)
+static void *__alloc(struct bpf_mem_cache *c, int node, gfp_t flags)
 {
-	/* Allocate, but don't deplete atomic reserves that typical
-	 * GFP_ATOMIC would do. irq_work runs on this cpu and kmalloc
-	 * will allocate from the current numa node which is what we
-	 * want here.
-	 */
-	gfp_t flags = GFP_NOWAIT | __GFP_NOWARN | __GFP_ACCOUNT;
-
 	if (c->percpu_size) {
 		void **obj = kmalloc_node(c->percpu_size, flags, node);
 		void *pptr = __alloc_percpu_gfp(c->unit_size, 8, flags);
@@ -185,7 +178,12 @@ static void alloc_bulk(struct bpf_mem_cache *c, int cnt, int node)
 		 */
 		obj = __llist_del_first(&c->free_by_rcu);
 		if (!obj) {
-			obj = __alloc(c, node);
+			/* Allocate, but don't deplete atomic reserves that typical
+			 * GFP_ATOMIC would do. irq_work runs on this cpu and kmalloc
+			 * will allocate from the current numa node which is what we
+			 * want here.
+			 */
+			obj = __alloc(c, node, GFP_NOWAIT | __GFP_NOWARN | __GFP_ACCOUNT);
 			if (!obj)
 				break;
 		}
@@ -676,3 +674,46 @@ void notrace bpf_mem_cache_free(struct bpf_mem_alloc *ma, void *ptr)
 
 	unit_free(this_cpu_ptr(ma->cache), ptr);
 }
+
+/* Directly does a kfree() without putting 'ptr' back to the free_llist
+ * for reuse and without waiting for a rcu_tasks_trace gp.
+ * The caller must first go through the rcu_tasks_trace gp for 'ptr'
+ * before calling bpf_mem_cache_raw_free().
+ * It could be used when the rcu_tasks_trace callback does not have
+ * a hold on the original bpf_mem_alloc object that allocated the
+ * 'ptr'. This should only be used in the uncommon code path.
+ * Otherwise, the bpf_mem_alloc's free_llist cannot be refilled
+ * and may affect performance.
+ */
+void bpf_mem_cache_raw_free(void *ptr)
+{
+	if (!ptr)
+		return;
+
+	kfree(ptr - LLIST_NODE_SZ);
+}
+
+/* When flags == GFP_KERNEL, it signals that the caller will not cause
+ * deadlock when using kmalloc. bpf_mem_cache_alloc_flags() will use
+ * kmalloc if the free_llist is empty.
+ */
+void notrace *bpf_mem_cache_alloc_flags(struct bpf_mem_alloc *ma, gfp_t flags)
+{
+	struct bpf_mem_cache *c;
+	void *ret;
+
+	c = this_cpu_ptr(ma->cache);
+
+	ret = unit_alloc(c);
+	if (!ret && flags == GFP_KERNEL) {
+		struct mem_cgroup *memcg, *old_memcg;
+
+		memcg = get_memcg(c);
+		old_memcg = set_active_memcg(memcg);
+		ret = __alloc(c, NUMA_NO_NODE, GFP_KERNEL | __GFP_NOWARN | __GFP_ACCOUNT);
+		set_active_memcg(old_memcg);
+		mem_cgroup_put(memcg);
+	}
+
+	return !ret ? NULL : ret + LLIST_NODE_SZ;
+}
-- 
cgit v1.2.3-70-g09d2


From 08a7ce384e33e53e0732c500a8af67a73f8fceca Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 22 Mar 2023 14:52:43 -0700
Subject: bpf: Use bpf_mem_cache_alloc/free in bpf_local_storage_elem

This patch uses bpf_mem_alloc for the task and cgroup local storage that
the bpf prog can easily get a hold of the storage owner's PTR_TO_BTF_ID.
eg. bpf_get_current_task_btf() can be used in some of the kmalloc code
path which will cause deadlock/recursion. bpf_mem_cache_alloc is
deadlock free and will solve a legit use case in [1].

For sk storage, its batch creation benchmark shows a few percent
regression when the sk create/destroy batch size is larger than 32.
The sk creation/destruction happens much more often and
depends on external traffic. Considering it is hypothetical
to be able to cause deadlock with sk storage, it can cross
the bridge to use bpf_mem_alloc till a legit (ie. useful)
use case comes up.

For inode storage, bpf_local_storage_destroy() is called before
waiting for a rcu gp and its memory cannot be reused immediately.
inode stays with kmalloc/kfree after the rcu [or tasks_trace] gp.

A 'bool bpf_ma' argument is added to bpf_local_storage_map_alloc().
Only task and cgroup storage have 'bpf_ma == true' which
means to use bpf_mem_cache_alloc/free(). This patch only changes
selem to use bpf_mem_alloc for task and cgroup. The next patch
will change the local_storage to use bpf_mem_alloc also for
task and cgroup.

Here is some more details on the changes:

* memory allocation:
After bpf_mem_cache_alloc(), the SDATA(selem)->data is zero-ed because
bpf_mem_cache_alloc() could return a reused selem. It is to keep
the existing bpf_map_kzalloc() behavior. Only SDATA(selem)->data
is zero-ed. SDATA(selem)->data is the visible part to the bpf prog.
No need to use zero_map_value() to do the zeroing because
bpf_selem_free(..., reuse_now = true) ensures no bpf prog is using
the selem before returning the selem through bpf_mem_cache_free().
For the internal fields of selem, they will be initialized when
linking to the new smap and the new local_storage.

When 'bpf_ma == false', nothing changes in this patch. It will
stay with the bpf_map_kzalloc().

* memory free:
The bpf_selem_free() and bpf_selem_free_rcu() are modified to handle
the bpf_ma == true case.

For the common selem free path where its owner is also being destroyed,
the mem is freed in bpf_local_storage_destroy(), the owner (task
and cgroup) has gone through a rcu gp. The memory can be reused
immediately, so bpf_local_storage_destroy() will call
bpf_selem_free(..., reuse_now = true) which will do
bpf_mem_cache_free() for immediate reuse consideration.

An exception is the delete elem code path. The delete elem code path
is called from the helper bpf_*_storage_delete() and the syscall
bpf_map_delete_elem(). This path is an unusual case for local
storage because the common use case is to have the local storage
staying with its owner life time so that the bpf prog and the user
space does not have to monitor the owner's destruction. For the delete
elem path, the selem cannot be reused immediately because there could
be bpf prog using it. It will call bpf_selem_free(..., reuse_now = false)
and it will wait for a rcu tasks trace gp before freeing the elem. The
rcu callback is changed to do bpf_mem_cache_raw_free() instead of kfree().

When 'bpf_ma == false', it should be the same as before.
__bpf_selem_free() is added to do the kfree_rcu and call_tasks_trace_rcu().
A few words on the 'reuse_now == true'. When 'reuse_now == true',
it is still racing with bpf_local_storage_map_free which is under rcu
protection, so it still needs to wait for a rcu gp instead of kfree().
Otherwise, the selem may be reused by slab for a totally different struct
while the bpf_local_storage_map_free() is still using it (as a
rcu reader). For the inode case, there may be other rcu readers also.
In short, when bpf_ma == false and reuse_now == true => vanilla rcu.

[1]: https://lore.kernel.org/bpf/20221118190109.1512674-1-namhyung@kernel.org/

Cc: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230322215246.1675516-3-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h |  6 ++-
 kernel/bpf/bpf_cgrp_storage.c     |  2 +-
 kernel/bpf/bpf_inode_storage.c    |  2 +-
 kernel/bpf/bpf_local_storage.c    | 95 +++++++++++++++++++++++++++++++++++----
 kernel/bpf/bpf_task_storage.c     |  2 +-
 net/core/bpf_sk_storage.c         |  2 +-
 6 files changed, 95 insertions(+), 14 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index a34f61467a2f..30efbcab2798 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -13,6 +13,7 @@
 #include <linux/list.h>
 #include <linux/hash.h>
 #include <linux/types.h>
+#include <linux/bpf_mem_alloc.h>
 #include <uapi/linux/btf.h>
 
 #define BPF_LOCAL_STORAGE_CACHE_SIZE	16
@@ -55,6 +56,8 @@ struct bpf_local_storage_map {
 	u32 bucket_log;
 	u16 elem_size;
 	u16 cache_idx;
+	struct bpf_mem_alloc selem_ma;
+	bool bpf_ma;
 };
 
 struct bpf_local_storage_data {
@@ -122,7 +125,8 @@ int bpf_local_storage_map_alloc_check(union bpf_attr *attr);
 
 struct bpf_map *
 bpf_local_storage_map_alloc(union bpf_attr *attr,
-			    struct bpf_local_storage_cache *cache);
+			    struct bpf_local_storage_cache *cache,
+			    bool bpf_ma);
 
 struct bpf_local_storage_data *
 bpf_local_storage_lookup(struct bpf_local_storage *local_storage,
diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c
index f5b016a5484d..d17d5b694668 100644
--- a/kernel/bpf/bpf_cgrp_storage.c
+++ b/kernel/bpf/bpf_cgrp_storage.c
@@ -149,7 +149,7 @@ static int notsupp_get_next_key(struct bpf_map *map, void *key, void *next_key)
 
 static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr)
 {
-	return bpf_local_storage_map_alloc(attr, &cgroup_cache);
+	return bpf_local_storage_map_alloc(attr, &cgroup_cache, true);
 }
 
 static void cgroup_storage_map_free(struct bpf_map *map)
diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c
index 9a5f05151898..e17ad581b9be 100644
--- a/kernel/bpf/bpf_inode_storage.c
+++ b/kernel/bpf/bpf_inode_storage.c
@@ -199,7 +199,7 @@ static int notsupp_get_next_key(struct bpf_map *map, void *key,
 
 static struct bpf_map *inode_storage_map_alloc(union bpf_attr *attr)
 {
-	return bpf_local_storage_map_alloc(attr, &inode_cache);
+	return bpf_local_storage_map_alloc(attr, &inode_cache, false);
 }
 
 static void inode_storage_map_free(struct bpf_map *map)
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 351d991694cb..309ea727a5cb 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -80,8 +80,24 @@ bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner,
 	if (charge_mem && mem_charge(smap, owner, smap->elem_size))
 		return NULL;
 
-	selem = bpf_map_kzalloc(&smap->map, smap->elem_size,
-				gfp_flags | __GFP_NOWARN);
+	if (smap->bpf_ma) {
+		migrate_disable();
+		selem = bpf_mem_cache_alloc_flags(&smap->selem_ma, gfp_flags);
+		migrate_enable();
+		if (selem)
+			/* Keep the original bpf_map_kzalloc behavior
+			 * before started using the bpf_mem_cache_alloc.
+			 *
+			 * No need to use zero_map_value. The bpf_selem_free()
+			 * only does bpf_mem_cache_free when there is
+			 * no other bpf prog is using the selem.
+			 */
+			memset(SDATA(selem)->data, 0, smap->map.value_size);
+	} else {
+		selem = bpf_map_kzalloc(&smap->map, smap->elem_size,
+					gfp_flags | __GFP_NOWARN);
+	}
+
 	if (selem) {
 		if (value)
 			copy_map_value(&smap->map, SDATA(selem)->data, value);
@@ -124,12 +140,34 @@ static void bpf_local_storage_free(struct bpf_local_storage *local_storage,
 		call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
 }
 
+/* rcu tasks trace callback for bpf_ma == false */
+static void __bpf_selem_free_trace_rcu(struct rcu_head *rcu)
+{
+	struct bpf_local_storage_elem *selem;
+
+	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
+	if (rcu_trace_implies_rcu_gp())
+		kfree(selem);
+	else
+		kfree_rcu(selem, rcu);
+}
+
+/* Handle bpf_ma == false */
+static void __bpf_selem_free(struct bpf_local_storage_elem *selem,
+			     bool vanilla_rcu)
+{
+	if (vanilla_rcu)
+		kfree_rcu(selem, rcu);
+	else
+		call_rcu_tasks_trace(&selem->rcu, __bpf_selem_free_trace_rcu);
+}
+
 static void bpf_selem_free_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage_elem *selem;
 
 	selem = container_of(rcu, struct bpf_local_storage_elem, rcu);
-	kfree(selem);
+	bpf_mem_cache_raw_free(selem);
 }
 
 static void bpf_selem_free_trace_rcu(struct rcu_head *rcu)
@@ -145,10 +183,23 @@ void bpf_selem_free(struct bpf_local_storage_elem *selem,
 		    bool reuse_now)
 {
 	bpf_obj_free_fields(smap->map.record, SDATA(selem)->data);
-	if (!reuse_now)
+
+	if (!smap->bpf_ma) {
+		__bpf_selem_free(selem, reuse_now);
+		return;
+	}
+
+	if (!reuse_now) {
 		call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu);
-	else
-		call_rcu(&selem->rcu, bpf_selem_free_rcu);
+	} else {
+		/* Instead of using the vanilla call_rcu(),
+		 * bpf_mem_cache_free will be able to reuse selem
+		 * immediately.
+		 */
+		migrate_disable();
+		bpf_mem_cache_free(&smap->selem_ma, selem);
+		migrate_enable();
+	}
 }
 
 /* local_storage->lock must be held and selem->local_storage == local_storage.
@@ -654,13 +705,25 @@ u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map)
 	return usage;
 }
 
+/* When bpf_ma == true, the bpf_mem_alloc is used to allocate and free memory.
+ * A deadlock free allocator is useful for storage that the bpf prog can easily
+ * get a hold of the owner PTR_TO_BTF_ID in any context. eg. bpf_get_current_task_btf.
+ * The task and cgroup storage fall into this case. The bpf_mem_alloc reuses
+ * memory immediately. To be reuse-immediate safe, the owner destruction
+ * code path needs to go through a rcu grace period before calling
+ * bpf_local_storage_destroy().
+ *
+ * When bpf_ma == false, the kmalloc and kfree are used.
+ */
 struct bpf_map *
 bpf_local_storage_map_alloc(union bpf_attr *attr,
-			    struct bpf_local_storage_cache *cache)
+			    struct bpf_local_storage_cache *cache,
+			    bool bpf_ma)
 {
 	struct bpf_local_storage_map *smap;
 	unsigned int i;
 	u32 nbuckets;
+	int err;
 
 	smap = bpf_map_area_alloc(sizeof(*smap), NUMA_NO_NODE);
 	if (!smap)
@@ -675,8 +738,8 @@ bpf_local_storage_map_alloc(union bpf_attr *attr,
 	smap->buckets = bpf_map_kvcalloc(&smap->map, sizeof(*smap->buckets),
 					 nbuckets, GFP_USER | __GFP_NOWARN);
 	if (!smap->buckets) {
-		bpf_map_area_free(smap);
-		return ERR_PTR(-ENOMEM);
+		err = -ENOMEM;
+		goto free_smap;
 	}
 
 	for (i = 0; i < nbuckets; i++) {
@@ -687,8 +750,20 @@ bpf_local_storage_map_alloc(union bpf_attr *attr,
 	smap->elem_size = offsetof(struct bpf_local_storage_elem,
 				   sdata.data[attr->value_size]);
 
+	smap->bpf_ma = bpf_ma;
+	if (bpf_ma) {
+		err = bpf_mem_alloc_init(&smap->selem_ma, smap->elem_size, false);
+		if (err)
+			goto free_smap;
+	}
+
 	smap->cache_idx = bpf_local_storage_cache_idx_get(cache);
 	return &smap->map;
+
+free_smap:
+	kvfree(smap->buckets);
+	bpf_map_area_free(smap);
+	return ERR_PTR(err);
 }
 
 void bpf_local_storage_map_free(struct bpf_map *map,
@@ -754,6 +829,8 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 	 */
 	synchronize_rcu();
 
+	if (smap->bpf_ma)
+		bpf_mem_alloc_destroy(&smap->selem_ma);
 	kvfree(smap->buckets);
 	bpf_map_area_free(smap);
 }
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index ab5bd1ef58c4..d1af0c8f9ce4 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -309,7 +309,7 @@ static int notsupp_get_next_key(struct bpf_map *map, void *key, void *next_key)
 
 static struct bpf_map *task_storage_map_alloc(union bpf_attr *attr)
 {
-	return bpf_local_storage_map_alloc(attr, &task_cache);
+	return bpf_local_storage_map_alloc(attr, &task_cache, true);
 }
 
 static void task_storage_map_free(struct bpf_map *map)
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index cb0f5a105b89..085025c7130a 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -68,7 +68,7 @@ static void bpf_sk_storage_map_free(struct bpf_map *map)
 
 static struct bpf_map *bpf_sk_storage_map_alloc(union bpf_attr *attr)
 {
-	return bpf_local_storage_map_alloc(attr, &sk_cache);
+	return bpf_local_storage_map_alloc(attr, &sk_cache, false);
 }
 
 static int notsupp_get_next_key(struct bpf_map *map, void *key,
-- 
cgit v1.2.3-70-g09d2


From 6ae9d5e99e1dd26babdd9502759fa25a3fd348ad Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 22 Mar 2023 14:52:44 -0700
Subject: bpf: Use bpf_mem_cache_alloc/free for bpf_local_storage

This patch uses bpf_mem_cache_alloc/free for allocating and freeing
bpf_local_storage for task and cgroup storage.

The changes are similar to the previous patch. A few things that
worth to mention for bpf_local_storage:

The local_storage is freed when the last selem is deleted.
Before deleting a selem from local_storage, it needs to retrieve the
local_storage->smap because the bpf_selem_unlink_storage_nolock()
may have set it to NULL. Note that local_storage->smap may have
already been NULL when the selem created this local_storage has
been removed. In this case, call_rcu will be used to free the
local_storage.
Also, the bpf_ma (true or false) value is needed before calling
bpf_local_storage_free(). The bpf_ma can either be obtained from
the local_storage->smap (if available) or any of its selem's smap.
A new helper check_storage_bpf_ma() is added to obtain
bpf_ma for a deleting bpf_local_storage.

When bpf_local_storage_alloc getting a reused memory, all
fields are either in the correct values or will be initialized.
'cache[]' must already be all NULLs. 'list' must be empty.
Others will be initialized.

Cc: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230322215246.1675516-4-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_local_storage.h |   1 +
 kernel/bpf/bpf_local_storage.c    | 130 +++++++++++++++++++++++++++++++++-----
 2 files changed, 116 insertions(+), 15 deletions(-)

diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h
index 30efbcab2798..173ec7f43ed1 100644
--- a/include/linux/bpf_local_storage.h
+++ b/include/linux/bpf_local_storage.h
@@ -57,6 +57,7 @@ struct bpf_local_storage_map {
 	u16 elem_size;
 	u16 cache_idx;
 	struct bpf_mem_alloc selem_ma;
+	struct bpf_mem_alloc storage_ma;
 	bool bpf_ma;
 };
 
diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 309ea727a5cb..dab2ff4c99d9 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -111,33 +111,74 @@ bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner,
 	return NULL;
 }
 
+/* rcu tasks trace callback for bpf_ma == false */
+static void __bpf_local_storage_free_trace_rcu(struct rcu_head *rcu)
+{
+	struct bpf_local_storage *local_storage;
+
+	/* If RCU Tasks Trace grace period implies RCU grace period, do
+	 * kfree(), else do kfree_rcu().
+	 */
+	local_storage = container_of(rcu, struct bpf_local_storage, rcu);
+	if (rcu_trace_implies_rcu_gp())
+		kfree(local_storage);
+	else
+		kfree_rcu(local_storage, rcu);
+}
+
 static void bpf_local_storage_free_rcu(struct rcu_head *rcu)
 {
 	struct bpf_local_storage *local_storage;
 
 	local_storage = container_of(rcu, struct bpf_local_storage, rcu);
-	kfree(local_storage);
+	bpf_mem_cache_raw_free(local_storage);
 }
 
 static void bpf_local_storage_free_trace_rcu(struct rcu_head *rcu)
 {
-	/* If RCU Tasks Trace grace period implies RCU grace period, do
-	 * kfree(), else do kfree_rcu().
-	 */
 	if (rcu_trace_implies_rcu_gp())
 		bpf_local_storage_free_rcu(rcu);
 	else
 		call_rcu(rcu, bpf_local_storage_free_rcu);
 }
 
+/* Handle bpf_ma == false */
+static void __bpf_local_storage_free(struct bpf_local_storage *local_storage,
+				     bool vanilla_rcu)
+{
+	if (vanilla_rcu)
+		kfree_rcu(local_storage, rcu);
+	else
+		call_rcu_tasks_trace(&local_storage->rcu,
+				     __bpf_local_storage_free_trace_rcu);
+}
+
 static void bpf_local_storage_free(struct bpf_local_storage *local_storage,
-				   bool reuse_now)
+				   struct bpf_local_storage_map *smap,
+				   bool bpf_ma, bool reuse_now)
 {
-	if (!reuse_now)
+	if (!bpf_ma) {
+		__bpf_local_storage_free(local_storage, reuse_now);
+		return;
+	}
+
+	if (!reuse_now) {
 		call_rcu_tasks_trace(&local_storage->rcu,
 				     bpf_local_storage_free_trace_rcu);
-	else
+		return;
+	}
+
+	if (smap) {
+		migrate_disable();
+		bpf_mem_cache_free(&smap->storage_ma, local_storage);
+		migrate_enable();
+	} else {
+		/* smap could be NULL if the selem that triggered
+		 * this 'local_storage' creation had been long gone.
+		 * In this case, directly do call_rcu().
+		 */
 		call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu);
+	}
 }
 
 /* rcu tasks trace callback for bpf_ma == false */
@@ -260,11 +301,47 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor
 	return free_local_storage;
 }
 
+static bool check_storage_bpf_ma(struct bpf_local_storage *local_storage,
+				 struct bpf_local_storage_map *storage_smap,
+				 struct bpf_local_storage_elem *selem)
+{
+
+	struct bpf_local_storage_map *selem_smap;
+
+	/* local_storage->smap may be NULL. If it is, get the bpf_ma
+	 * from any selem in the local_storage->list. The bpf_ma of all
+	 * local_storage and selem should have the same value
+	 * for the same map type.
+	 *
+	 * If the local_storage->list is already empty, the caller will not
+	 * care about the bpf_ma value also because the caller is not
+	 * responsibile to free the local_storage.
+	 */
+
+	if (storage_smap)
+		return storage_smap->bpf_ma;
+
+	if (!selem) {
+		struct hlist_node *n;
+
+		n = rcu_dereference_check(hlist_first_rcu(&local_storage->list),
+					  bpf_rcu_lock_held());
+		if (!n)
+			return false;
+
+		selem = hlist_entry(n, struct bpf_local_storage_elem, snode);
+	}
+	selem_smap = rcu_dereference_check(SDATA(selem)->smap, bpf_rcu_lock_held());
+
+	return selem_smap->bpf_ma;
+}
+
 static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 				     bool reuse_now)
 {
+	struct bpf_local_storage_map *storage_smap;
 	struct bpf_local_storage *local_storage;
-	bool free_local_storage = false;
+	bool bpf_ma, free_local_storage = false;
 	unsigned long flags;
 
 	if (unlikely(!selem_linked_to_storage_lockless(selem)))
@@ -273,6 +350,10 @@ static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 
 	local_storage = rcu_dereference_check(selem->local_storage,
 					      bpf_rcu_lock_held());
+	storage_smap = rcu_dereference_check(local_storage->smap,
+					     bpf_rcu_lock_held());
+	bpf_ma = check_storage_bpf_ma(local_storage, storage_smap, selem);
+
 	raw_spin_lock_irqsave(&local_storage->lock, flags);
 	if (likely(selem_linked_to_storage(selem)))
 		free_local_storage = bpf_selem_unlink_storage_nolock(
@@ -280,7 +361,7 @@ static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem,
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
 	if (free_local_storage)
-		bpf_local_storage_free(local_storage, reuse_now);
+		bpf_local_storage_free(local_storage, storage_smap, bpf_ma, reuse_now);
 }
 
 void bpf_selem_link_storage_nolock(struct bpf_local_storage *local_storage,
@@ -400,8 +481,15 @@ int bpf_local_storage_alloc(void *owner,
 	if (err)
 		return err;
 
-	storage = bpf_map_kzalloc(&smap->map, sizeof(*storage),
-				  gfp_flags | __GFP_NOWARN);
+	if (smap->bpf_ma) {
+		migrate_disable();
+		storage = bpf_mem_cache_alloc_flags(&smap->storage_ma, gfp_flags);
+		migrate_enable();
+	} else {
+		storage = bpf_map_kzalloc(&smap->map, sizeof(*storage),
+					  gfp_flags | __GFP_NOWARN);
+	}
+
 	if (!storage) {
 		err = -ENOMEM;
 		goto uncharge;
@@ -447,7 +535,7 @@ int bpf_local_storage_alloc(void *owner,
 	return 0;
 
 uncharge:
-	bpf_local_storage_free(storage, true);
+	bpf_local_storage_free(storage, smap, smap->bpf_ma, true);
 	mem_uncharge(smap, owner, sizeof(*storage));
 	return err;
 }
@@ -660,11 +748,15 @@ int bpf_local_storage_map_check_btf(const struct bpf_map *map,
 
 void bpf_local_storage_destroy(struct bpf_local_storage *local_storage)
 {
+	struct bpf_local_storage_map *storage_smap;
 	struct bpf_local_storage_elem *selem;
-	bool free_storage = false;
+	bool bpf_ma, free_storage = false;
 	struct hlist_node *n;
 	unsigned long flags;
 
+	storage_smap = rcu_dereference_check(local_storage->smap, bpf_rcu_lock_held());
+	bpf_ma = check_storage_bpf_ma(local_storage, storage_smap, NULL);
+
 	/* Neither the bpf_prog nor the bpf_map's syscall
 	 * could be modifying the local_storage->list now.
 	 * Thus, no elem can be added to or deleted from the
@@ -692,7 +784,7 @@ void bpf_local_storage_destroy(struct bpf_local_storage *local_storage)
 	raw_spin_unlock_irqrestore(&local_storage->lock, flags);
 
 	if (free_storage)
-		bpf_local_storage_free(local_storage, true);
+		bpf_local_storage_free(local_storage, storage_smap, bpf_ma, true);
 }
 
 u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map)
@@ -755,6 +847,12 @@ bpf_local_storage_map_alloc(union bpf_attr *attr,
 		err = bpf_mem_alloc_init(&smap->selem_ma, smap->elem_size, false);
 		if (err)
 			goto free_smap;
+
+		err = bpf_mem_alloc_init(&smap->storage_ma, sizeof(struct bpf_local_storage), false);
+		if (err) {
+			bpf_mem_alloc_destroy(&smap->selem_ma);
+			goto free_smap;
+		}
 	}
 
 	smap->cache_idx = bpf_local_storage_cache_idx_get(cache);
@@ -829,8 +927,10 @@ void bpf_local_storage_map_free(struct bpf_map *map,
 	 */
 	synchronize_rcu();
 
-	if (smap->bpf_ma)
+	if (smap->bpf_ma) {
 		bpf_mem_alloc_destroy(&smap->selem_ma);
+		bpf_mem_alloc_destroy(&smap->storage_ma);
+	}
 	kvfree(smap->buckets);
 	bpf_map_area_free(smap);
 }
-- 
cgit v1.2.3-70-g09d2


From d8db84d71c0e539f7ce902e2fe297e535ba4d46c Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 22 Mar 2023 14:52:45 -0700
Subject: selftests/bpf: Test task storage when local_storage->smap is NULL

The current sk storage test ensures the memory free works when
the local_storage->smap is NULL.

This patch adds a task storage test to ensure the memory free
code path works when local_storage->smap is NULL.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230322215246.1675516-5-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/test_local_storage.c  |  7 +--
 tools/testing/selftests/bpf/progs/local_storage.c  | 56 ++++++++++++++++------
 2 files changed, 46 insertions(+), 17 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_local_storage.c b/tools/testing/selftests/bpf/prog_tests/test_local_storage.c
index 563a9c746b7b..bcf2e1905ed7 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_local_storage.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_local_storage.c
@@ -23,7 +23,7 @@ struct storage {
 /* Fork and exec the provided rm binary and return the exit code of the
  * forked process and its pid.
  */
-static int run_self_unlink(int *monitored_pid, const char *rm_path)
+static int run_self_unlink(struct local_storage *skel, const char *rm_path)
 {
 	int child_pid, child_status, ret;
 	int null_fd;
@@ -35,7 +35,7 @@ static int run_self_unlink(int *monitored_pid, const char *rm_path)
 		dup2(null_fd, STDERR_FILENO);
 		close(null_fd);
 
-		*monitored_pid = getpid();
+		skel->bss->monitored_pid = getpid();
 		/* Use the copied /usr/bin/rm to delete itself
 		 * /tmp/copy_of_rm /tmp/copy_of_rm.
 		 */
@@ -44,6 +44,7 @@ static int run_self_unlink(int *monitored_pid, const char *rm_path)
 			exit(errno);
 	} else if (child_pid > 0) {
 		waitpid(child_pid, &child_status, 0);
+		ASSERT_EQ(skel->data->task_storage_result, 0, "task_storage_result");
 		return WEXITSTATUS(child_status);
 	}
 
@@ -133,7 +134,7 @@ void test_test_local_storage(void)
 	 * unlink its executable. This operation should be denied by the loaded
 	 * LSM program.
 	 */
-	err = run_self_unlink(&skel->bss->monitored_pid, tmp_exec_path);
+	err = run_self_unlink(skel, tmp_exec_path);
 	if (!ASSERT_EQ(err, EPERM, "run_self_unlink"))
 		goto close_prog_rmdir;
 
diff --git a/tools/testing/selftests/bpf/progs/local_storage.c b/tools/testing/selftests/bpf/progs/local_storage.c
index c8ba7207f5a5..bc8ea56671a1 100644
--- a/tools/testing/selftests/bpf/progs/local_storage.c
+++ b/tools/testing/selftests/bpf/progs/local_storage.c
@@ -16,6 +16,7 @@ char _license[] SEC("license") = "GPL";
 int monitored_pid = 0;
 int inode_storage_result = -1;
 int sk_storage_result = -1;
+int task_storage_result = -1;
 
 struct local_storage {
 	struct inode *exec_inode;
@@ -50,26 +51,57 @@ struct {
 	__type(value, struct local_storage);
 } task_storage_map SEC(".maps");
 
+struct {
+	__uint(type, BPF_MAP_TYPE_TASK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct local_storage);
+} task_storage_map2 SEC(".maps");
+
 SEC("lsm/inode_unlink")
 int BPF_PROG(unlink_hook, struct inode *dir, struct dentry *victim)
 {
 	__u32 pid = bpf_get_current_pid_tgid() >> 32;
+	struct bpf_local_storage *local_storage;
 	struct local_storage *storage;
+	struct task_struct *task;
 	bool is_self_unlink;
 
 	if (pid != monitored_pid)
 		return 0;
 
-	storage = bpf_task_storage_get(&task_storage_map,
-				       bpf_get_current_task_btf(), 0, 0);
-	if (storage) {
-		/* Don't let an executable delete itself */
-		is_self_unlink = storage->exec_inode == victim->d_inode;
-		if (is_self_unlink)
-			return -EPERM;
-	}
+	task = bpf_get_current_task_btf();
+	if (!task)
+		return 0;
 
-	return 0;
+	task_storage_result = -1;
+
+	storage = bpf_task_storage_get(&task_storage_map, task, 0, 0);
+	if (!storage)
+		return 0;
+
+	/* Don't let an executable delete itself */
+	is_self_unlink = storage->exec_inode == victim->d_inode;
+
+	storage = bpf_task_storage_get(&task_storage_map2, task, 0,
+				       BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (!storage || storage->value)
+		return 0;
+
+	if (bpf_task_storage_delete(&task_storage_map, task))
+		return 0;
+
+	/* Ensure that the task_storage_map is disconnected from the storage.
+	 * The storage memory should not be freed back to the
+	 * bpf_mem_alloc.
+	 */
+	local_storage = task->bpf_storage;
+	if (!local_storage || local_storage->smap)
+		return 0;
+
+	task_storage_result = 0;
+
+	return is_self_unlink ? -EPERM : 0;
 }
 
 SEC("lsm.s/inode_rename")
@@ -139,11 +171,7 @@ int BPF_PROG(socket_bind, struct socket *sock, struct sockaddr *address,
 	if (bpf_sk_storage_delete(&sk_storage_map, sock->sk))
 		return 0;
 
-	/* Ensure that the sk_storage_map is disconnected from the storage.
-	 * The storage memory should not be freed back to the
-	 * bpf_mem_alloc of the sk_bpf_storage_map because
-	 * sk_bpf_storage_map may have been gone.
-	 */
+	/* Ensure that the sk_storage_map is disconnected from the storage. */
 	if (!sock->sk->sk_bpf_storage || sock->sk->sk_bpf_storage->smap)
 		return 0;
 
-- 
cgit v1.2.3-70-g09d2


From cbe9d93d58b16b5912498ea42b5173022fff7c04 Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <martin.lau@kernel.org>
Date: Wed, 22 Mar 2023 14:52:46 -0700
Subject: selftests/bpf: Add bench for task storage creation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds a task storage benchmark to the existing
local-storage-create benchmark.

For task storage,
./bench --storage-type task --batch-size 32:
   bpf_ma: Summary: creates   30.456 ± 0.507k/s ( 30.456k/prod), 6.08 kmallocs/create
no bpf_ma: Summary: creates   31.962 ± 0.486k/s ( 31.962k/prod), 6.13 kmallocs/create

./bench --storage-type task --batch-size 64:
   bpf_ma: Summary: creates   30.197 ± 1.476k/s ( 30.197k/prod), 6.08 kmallocs/create
no bpf_ma: Summary: creates   31.103 ± 0.297k/s ( 31.103k/prod), 6.13 kmallocs/create

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20230322215246.1675516-6-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/bench.c                |   2 +
 .../bpf/benchs/bench_local_storage_create.c        | 151 +++++++++++++++++++--
 .../bpf/progs/bench_local_storage_create.c         |  25 ++++
 3 files changed, 164 insertions(+), 14 deletions(-)

diff --git a/tools/testing/selftests/bpf/bench.c b/tools/testing/selftests/bpf/bench.c
index dc3827c1f139..d9c080ac1796 100644
--- a/tools/testing/selftests/bpf/bench.c
+++ b/tools/testing/selftests/bpf/bench.c
@@ -278,6 +278,7 @@ extern struct argp bench_local_storage_argp;
 extern struct argp bench_local_storage_rcu_tasks_trace_argp;
 extern struct argp bench_strncmp_argp;
 extern struct argp bench_hashmap_lookup_argp;
+extern struct argp bench_local_storage_create_argp;
 
 static const struct argp_child bench_parsers[] = {
 	{ &bench_ringbufs_argp, 0, "Ring buffers benchmark", 0 },
@@ -288,6 +289,7 @@ static const struct argp_child bench_parsers[] = {
 	{ &bench_local_storage_rcu_tasks_trace_argp, 0,
 		"local_storage RCU Tasks Trace slowdown benchmark", 0 },
 	{ &bench_hashmap_lookup_argp, 0, "Hashmap lookup benchmark", 0 },
+	{ &bench_local_storage_create_argp, 0, "local-storage-create benchmark", 0 },
 	{},
 };
 
diff --git a/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c b/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
index f8b2a640ccbe..abb0321d4f34 100644
--- a/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
+++ b/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
@@ -3,19 +3,71 @@
 
 #include <sys/types.h>
 #include <sys/socket.h>
+#include <pthread.h>
+#include <argp.h>
 
 #include "bench.h"
 #include "bench_local_storage_create.skel.h"
 
-#define BATCH_SZ 32
-
 struct thread {
-	int fds[BATCH_SZ];
+	int *fds;
+	pthread_t *pthds;
+	int *pthd_results;
 };
 
 static struct bench_local_storage_create *skel;
 static struct thread *threads;
-static long socket_errs;
+static long create_owner_errs;
+static int storage_type = BPF_MAP_TYPE_SK_STORAGE;
+static int batch_sz = 32;
+
+enum {
+	ARG_BATCH_SZ = 9000,
+	ARG_STORAGE_TYPE = 9001,
+};
+
+static const struct argp_option opts[] = {
+	{ "batch-size", ARG_BATCH_SZ, "BATCH_SIZE", 0,
+	  "The number of storage creations in each batch" },
+	{ "storage-type", ARG_STORAGE_TYPE, "STORAGE_TYPE", 0,
+	  "The type of local storage to test (socket or task)" },
+	{},
+};
+
+static error_t parse_arg(int key, char *arg, struct argp_state *state)
+{
+	int ret;
+
+	switch (key) {
+	case ARG_BATCH_SZ:
+		ret = atoi(arg);
+		if (ret < 1) {
+			fprintf(stderr, "invalid batch-size\n");
+			argp_usage(state);
+		}
+		batch_sz = ret;
+		break;
+	case ARG_STORAGE_TYPE:
+		if (!strcmp(arg, "task")) {
+			storage_type = BPF_MAP_TYPE_TASK_STORAGE;
+		} else if (!strcmp(arg, "socket")) {
+			storage_type = BPF_MAP_TYPE_SK_STORAGE;
+		} else {
+			fprintf(stderr, "invalid storage-type (socket or task)\n");
+			argp_usage(state);
+		}
+		break;
+	default:
+		return ARGP_ERR_UNKNOWN;
+	}
+
+	return 0;
+}
+
+const struct argp bench_local_storage_create_argp = {
+	.options = opts,
+	.parser = parse_arg,
+};
 
 static void validate(void)
 {
@@ -28,6 +80,8 @@ static void validate(void)
 
 static void setup(void)
 {
+	int i;
+
 	skel = bench_local_storage_create__open_and_load();
 	if (!skel) {
 		fprintf(stderr, "error loading skel\n");
@@ -35,10 +89,16 @@ static void setup(void)
 	}
 
 	skel->bss->bench_pid = getpid();
-
-	if (!bpf_program__attach(skel->progs.socket_post_create)) {
-		fprintf(stderr, "Error attaching bpf program\n");
-		exit(1);
+	if (storage_type == BPF_MAP_TYPE_SK_STORAGE) {
+		if (!bpf_program__attach(skel->progs.socket_post_create)) {
+			fprintf(stderr, "Error attaching bpf program\n");
+			exit(1);
+		}
+	} else {
+		if (!bpf_program__attach(skel->progs.fork)) {
+			fprintf(stderr, "Error attaching bpf program\n");
+			exit(1);
+		}
 	}
 
 	if (!bpf_program__attach(skel->progs.kmalloc)) {
@@ -52,6 +112,29 @@ static void setup(void)
 		fprintf(stderr, "cannot alloc thread_res\n");
 		exit(1);
 	}
+
+	for (i = 0; i < env.producer_cnt; i++) {
+		struct thread *t = &threads[i];
+
+		if (storage_type == BPF_MAP_TYPE_SK_STORAGE) {
+			t->fds = malloc(batch_sz * sizeof(*t->fds));
+			if (!t->fds) {
+				fprintf(stderr, "cannot alloc t->fds\n");
+				exit(1);
+			}
+		} else {
+			t->pthds = malloc(batch_sz * sizeof(*t->pthds));
+			if (!t->pthds) {
+				fprintf(stderr, "cannot alloc t->pthds\n");
+				exit(1);
+			}
+			t->pthd_results = malloc(batch_sz * sizeof(*t->pthd_results));
+			if (!t->pthd_results) {
+				fprintf(stderr, "cannot alloc t->pthd_results\n");
+				exit(1);
+			}
+		}
+	}
 }
 
 static void measure(struct bench_res *res)
@@ -65,20 +148,20 @@ static void *consumer(void *input)
 	return NULL;
 }
 
-static void *producer(void *input)
+static void *sk_producer(void *input)
 {
 	struct thread *t = &threads[(long)(input)];
 	int *fds = t->fds;
 	int i;
 
 	while (true) {
-		for (i = 0; i < BATCH_SZ; i++) {
+		for (i = 0; i < batch_sz; i++) {
 			fds[i] = socket(AF_INET6, SOCK_DGRAM, 0);
 			if (fds[i] == -1)
-				atomic_inc(&socket_errs);
+				atomic_inc(&create_owner_errs);
 		}
 
-		for (i = 0; i < BATCH_SZ; i++) {
+		for (i = 0; i < batch_sz; i++) {
 			if (fds[i] != -1)
 				close(fds[i]);
 		}
@@ -87,6 +170,42 @@ static void *producer(void *input)
 	return NULL;
 }
 
+static void *thread_func(void *arg)
+{
+	return NULL;
+}
+
+static void *task_producer(void *input)
+{
+	struct thread *t = &threads[(long)(input)];
+	pthread_t *pthds = t->pthds;
+	int *pthd_results = t->pthd_results;
+	int i;
+
+	while (true) {
+		for (i = 0; i < batch_sz; i++) {
+			pthd_results[i] = pthread_create(&pthds[i], NULL, thread_func, NULL);
+			if (pthd_results[i])
+				atomic_inc(&create_owner_errs);
+		}
+
+		for (i = 0; i < batch_sz; i++) {
+			if (!pthd_results[i])
+				pthread_join(pthds[i], NULL);;
+		}
+	}
+
+	return NULL;
+}
+
+static void *producer(void *input)
+{
+	if (storage_type == BPF_MAP_TYPE_SK_STORAGE)
+		return sk_producer(input);
+	else
+		return task_producer(input);
+}
+
 static void report_progress(int iter, struct bench_res *res, long delta_ns)
 {
 	double creates_per_sec, kmallocs_per_create;
@@ -123,14 +242,18 @@ static void report_final(struct bench_res res[], int res_cnt)
 	printf("Summary: creates %8.3lf \u00B1 %5.3lfk/s (%7.3lfk/prod), ",
 	       creates_mean, creates_stddev, creates_mean / env.producer_cnt);
 	printf("%4.2lf kmallocs/create\n", (double)total_kmallocs / total_creates);
-	if (socket_errs || skel->bss->create_errs)
-		printf("socket() errors %ld create_errs %ld\n", socket_errs,
+	if (create_owner_errs || skel->bss->create_errs)
+		printf("%s() errors %ld create_errs %ld\n",
+		       storage_type == BPF_MAP_TYPE_SK_STORAGE ?
+		       "socket" : "pthread_create",
+		       create_owner_errs,
 		       skel->bss->create_errs);
 }
 
 /* Benchmark performance of creating bpf local storage  */
 const struct bench bench_local_storage_create = {
 	.name = "local-storage-create",
+	.argp = &bench_local_storage_create_argp,
 	.validate = validate,
 	.setup = setup,
 	.producer_thread = producer,
diff --git a/tools/testing/selftests/bpf/progs/bench_local_storage_create.c b/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
index 2814bab54d28..7c851c9d5e47 100644
--- a/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
+++ b/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
@@ -22,6 +22,13 @@ struct {
 	__type(value, struct storage);
 } sk_storage_map SEC(".maps");
 
+struct {
+	__uint(type, BPF_MAP_TYPE_TASK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct storage);
+} task_storage_map SEC(".maps");
+
 SEC("raw_tp/kmalloc")
 int BPF_PROG(kmalloc, unsigned long call_site, const void *ptr,
 	     size_t bytes_req, size_t bytes_alloc, gfp_t gfp_flags,
@@ -32,6 +39,24 @@ int BPF_PROG(kmalloc, unsigned long call_site, const void *ptr,
 	return 0;
 }
 
+SEC("tp_btf/sched_process_fork")
+int BPF_PROG(fork, struct task_struct *parent, struct task_struct *child)
+{
+	struct storage *stg;
+
+	if (parent->tgid != bench_pid)
+		return 0;
+
+	stg = bpf_task_storage_get(&task_storage_map, child, NULL,
+				   BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (stg)
+		__sync_fetch_and_add(&create_cnts, 1);
+	else
+		__sync_fetch_and_add(&create_errs, 1);
+
+	return 0;
+}
+
 SEC("lsm.s/socket_post_create")
 int BPF_PROG(socket_post_create, struct socket *sock, int family, int type,
 	     int protocol, int kern)
-- 
cgit v1.2.3-70-g09d2


From 8cfee110711ed60bfdd39af0107ddef01d6b72c3 Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler@microsoft.com>
Date: Sun, 26 Mar 2023 03:31:17 +0000
Subject: bpf, docs: Add extended call instructions

Add extended call instructions.  Uses the term "program-local" for
call by offset.  And there are instructions for calling helper functions
by "address" (the old way of using integer values), and for calling
helper functions by BTF ID (for kfuncs).

V1 -> V2: addressed comments from David Vernet

V2 -> V3: make descriptions in table consistent with updated names

V3 -> V4: addressed comments from Alexei

V4 -> V5: fixed alignment

Signed-off-by: Dave Thaler <dthaler@microsoft.com>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230326033117.1075-1-dthaler1968@googlemail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/instruction-set.rst | 59 ++++++++++++++++++++++-------------
 1 file changed, 37 insertions(+), 22 deletions(-)

diff --git a/Documentation/bpf/instruction-set.rst b/Documentation/bpf/instruction-set.rst
index b44640589055..b77280eb926f 100644
--- a/Documentation/bpf/instruction-set.rst
+++ b/Documentation/bpf/instruction-set.rst
@@ -243,27 +243,29 @@ Jump instructions
 otherwise identical operations.
 The 'code' field encodes the operation as below:
 
-========  =====  =========================  ============
-code      value  description                notes
-========  =====  =========================  ============
-BPF_JA    0x00   PC += off                  BPF_JMP only
-BPF_JEQ   0x10   PC += off if dst == src
-BPF_JGT   0x20   PC += off if dst > src     unsigned
-BPF_JGE   0x30   PC += off if dst >= src    unsigned
-BPF_JSET  0x40   PC += off if dst & src
-BPF_JNE   0x50   PC += off if dst != src
-BPF_JSGT  0x60   PC += off if dst > src     signed
-BPF_JSGE  0x70   PC += off if dst >= src    signed
-BPF_CALL  0x80   function call              see `Helper functions`_
-BPF_EXIT  0x90   function / program return  BPF_JMP only
-BPF_JLT   0xa0   PC += off if dst < src     unsigned
-BPF_JLE   0xb0   PC += off if dst <= src    unsigned
-BPF_JSLT  0xc0   PC += off if dst < src     signed
-BPF_JSLE  0xd0   PC += off if dst <= src    signed
-========  =====  =========================  ============
+========  =====  ===  ===========================================  =========================================
+code      value  src  description                                  notes
+========  =====  ===  ===========================================  =========================================
+BPF_JA    0x0    0x0  PC += offset                                 BPF_JMP only
+BPF_JEQ   0x1    any  PC += offset if dst == src
+BPF_JGT   0x2    any  PC += offset if dst > src                    unsigned
+BPF_JGE   0x3    any  PC += offset if dst >= src                   unsigned
+BPF_JSET  0x4    any  PC += offset if dst & src
+BPF_JNE   0x5    any  PC += offset if dst != src
+BPF_JSGT  0x6    any  PC += offset if dst > src                    signed
+BPF_JSGE  0x7    any  PC += offset if dst >= src                   signed
+BPF_CALL  0x8    0x0  call helper function by address              see `Helper functions`_
+BPF_CALL  0x8    0x1  call PC += offset                            see `Program-local functions`_
+BPF_CALL  0x8    0x2  call helper function by BTF ID               see `Helper functions`_
+BPF_EXIT  0x9    0x0  return                                       BPF_JMP only
+BPF_JLT   0xa    any  PC += offset if dst < src                    unsigned
+BPF_JLE   0xb    any  PC += offset if dst <= src                   unsigned
+BPF_JSLT  0xc    any  PC += offset if dst < src                    signed
+BPF_JSLE  0xd    any  PC += offset if dst <= src                   signed
+========  =====  ===  ===========================================  =========================================
 
 The eBPF program needs to store the return value into register R0 before doing a
-BPF_EXIT.
+``BPF_EXIT``.
 
 Example:
 
@@ -277,9 +279,22 @@ Helper functions
 ~~~~~~~~~~~~~~~~
 
 Helper functions are a concept whereby BPF programs can call into a
-set of function calls exposed by the runtime.  Each helper
-function is identified by an integer used in a ``BPF_CALL`` instruction.
-The available helper functions may differ for each program type.
+set of function calls exposed by the underlying platform.
+
+Historically, each helper function was identified by an address
+encoded in the imm field.  The available helper functions may differ
+for each program type, but address values are unique across all program types.
+
+Platforms that support the BPF Type Format (BTF) support identifying
+a helper function by a BTF ID encoded in the imm field, where the BTF ID
+identifies the helper name and type.
+
+Program-local functions
+~~~~~~~~~~~~~~~~~~~~~~~
+Program-local functions are functions exposed by the same BPF program as the
+caller, and are referenced by offset from the call instruction, similar to
+``BPF_JA``.  A ``BPF_EXIT`` within the program-local function will return to
+the caller.
 
 Load and store instructions
 ===========================
-- 
cgit v1.2.3-70-g09d2


From 5f5a7d8d8bd461f515543040ad7d107cc405d30c Mon Sep 17 00:00:00 2001
From: Nuno Gonçalves <nunog@fr24.com>
Date: Fri, 24 Mar 2023 10:02:22 +0000
Subject: xsk: allow remap of fill and/or completion rings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The remap of fill and completion rings was frowned upon as they
control the usage of UMEM which does not support concurrent use.
At the same time this would disallow the remap of these rings
into another process.

A possible use case is that the user wants to transfer the socket/
UMEM ownership to another process (via SYS_pidfd_getfd) and so
would need to also remap these rings.

This will have no impact on current usages and just relaxes the
remap limitation.

Signed-off-by: Nuno Gonçalves <nunog@fr24.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/r/20230324100222.13434-1-nunog@fr24.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/xdp/xsk.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 2ac58b282b5e..cc1e7f15fa73 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -1301,9 +1301,10 @@ static int xsk_mmap(struct file *file, struct socket *sock,
 	loff_t offset = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
 	unsigned long size = vma->vm_end - vma->vm_start;
 	struct xdp_sock *xs = xdp_sk(sock->sk);
+	int state = READ_ONCE(xs->state);
 	struct xsk_queue *q = NULL;
 
-	if (READ_ONCE(xs->state) != XSK_READY)
+	if (state != XSK_READY && state != XSK_BOUND)
 		return -EBUSY;
 
 	if (offset == XDP_PGOFF_RX_RING) {
@@ -1314,9 +1315,11 @@ static int xsk_mmap(struct file *file, struct socket *sock,
 		/* Matches the smp_wmb() in XDP_UMEM_REG */
 		smp_rmb();
 		if (offset == XDP_UMEM_PGOFF_FILL_RING)
-			q = READ_ONCE(xs->fq_tmp);
+			q = state == XSK_READY ? READ_ONCE(xs->fq_tmp) :
+						 READ_ONCE(xs->pool->fq);
 		else if (offset == XDP_UMEM_PGOFF_COMPLETION_RING)
-			q = READ_ONCE(xs->cq_tmp);
+			q = state == XSK_READY ? READ_ONCE(xs->cq_tmp) :
+						 READ_ONCE(xs->pool->cq);
 	}
 
 	if (!q)
-- 
cgit v1.2.3-70-g09d2


From a20869b3a7859135cd4e7800d8960966ad34a5b7 Mon Sep 17 00:00:00 2001
From: Álvaro Fernández Rojas <noltari@gmail.com>
Date: Fri, 24 Mar 2023 09:41:37 +0100
Subject: dt-bindings: net: dsa: b53: add BCM53134 support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BCM53134 are B53 switches connected by MDIO.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Acked-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml b/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml
index 57e0ef93b134..4c78c546343f 100644
--- a/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/brcm,b53.yaml
@@ -19,6 +19,7 @@ properties:
       - const: brcm,bcm53115
       - const: brcm,bcm53125
       - const: brcm,bcm53128
+      - const: brcm,bcm53134
       - const: brcm,bcm5365
       - const: brcm,bcm5395
       - const: brcm,bcm5389
-- 
cgit v1.2.3-70-g09d2


From f927e8ef1e93174780b7c26b239a046396914cd6 Mon Sep 17 00:00:00 2001
From: Paul Geurts <paul.geurts@prodrive-technologies.com>
Date: Fri, 24 Mar 2023 09:41:38 +0100
Subject: net: dsa: b53: mdio: add support for BCM53134
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add support for the BCM53134 Ethernet switch in the existing b53 dsa driver.
BCM53134 is very similar to the BCM58XX series.

Signed-off-by: Paul Geurts <paul.geurts@prodrive-technologies.com>
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/b53/b53_common.c | 15 +++++++++++++++
 drivers/net/dsa/b53/b53_mdio.c   |  5 ++++-
 drivers/net/dsa/b53/b53_priv.h   |  7 +++++--
 3 files changed, 24 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 1f9b251a5452..3464ce5e7470 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -2613,6 +2613,20 @@ static const struct b53_chip_data b53_switch_chips[] = {
 		.jumbo_pm_reg = B53_JUMBO_PORT_MASK,
 		.jumbo_size_reg = B53_JUMBO_MAX_SIZE,
 	},
+	{
+		.chip_id = BCM53134_DEVICE_ID,
+		.dev_name = "BCM53134",
+		.vlans = 4096,
+		.enabled_ports = 0x12f,
+		.imp_port = 8,
+		.cpu_port = B53_CPU_PORT,
+		.vta_regs = B53_VTA_REGS,
+		.arl_bins = 4,
+		.arl_buckets = 1024,
+		.duplex_reg = B53_DUPLEX_STAT_GE,
+		.jumbo_pm_reg = B53_JUMBO_PORT_MASK,
+		.jumbo_size_reg = B53_JUMBO_MAX_SIZE,
+	},
 };
 
 static int b53_switch_init(struct b53_device *dev)
@@ -2790,6 +2804,7 @@ int b53_switch_detect(struct b53_device *dev)
 		case BCM53012_DEVICE_ID:
 		case BCM53018_DEVICE_ID:
 		case BCM53019_DEVICE_ID:
+		case BCM53134_DEVICE_ID:
 			dev->chip_id = id32;
 			break;
 		default:
diff --git a/drivers/net/dsa/b53/b53_mdio.c b/drivers/net/dsa/b53/b53_mdio.c
index 6ddc03b58b28..8b422b298cd5 100644
--- a/drivers/net/dsa/b53/b53_mdio.c
+++ b/drivers/net/dsa/b53/b53_mdio.c
@@ -286,6 +286,7 @@ static const struct b53_io_ops b53_mdio_ops = {
 #define B53_BRCM_OUI_2	0x03625c00
 #define B53_BRCM_OUI_3	0x00406000
 #define B53_BRCM_OUI_4	0x01410c00
+#define B53_BRCM_OUI_5	0xae025000
 
 static int b53_mdio_probe(struct mdio_device *mdiodev)
 {
@@ -313,7 +314,8 @@ static int b53_mdio_probe(struct mdio_device *mdiodev)
 	if ((phy_id & 0xfffffc00) != B53_BRCM_OUI_1 &&
 	    (phy_id & 0xfffffc00) != B53_BRCM_OUI_2 &&
 	    (phy_id & 0xfffffc00) != B53_BRCM_OUI_3 &&
-	    (phy_id & 0xfffffc00) != B53_BRCM_OUI_4) {
+	    (phy_id & 0xfffffc00) != B53_BRCM_OUI_4 &&
+	    (phy_id & 0xfffffc00) != B53_BRCM_OUI_5) {
 		dev_err(&mdiodev->dev, "Unsupported device: 0x%08x\n", phy_id);
 		return -ENODEV;
 	}
@@ -375,6 +377,7 @@ static const struct of_device_id b53_of_match[] = {
 	{ .compatible = "brcm,bcm53115" },
 	{ .compatible = "brcm,bcm53125" },
 	{ .compatible = "brcm,bcm53128" },
+	{ .compatible = "brcm,bcm53134" },
 	{ .compatible = "brcm,bcm5365" },
 	{ .compatible = "brcm,bcm5389" },
 	{ .compatible = "brcm,bcm5395" },
diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h
index a689a6950189..fdcfd5081c28 100644
--- a/drivers/net/dsa/b53/b53_priv.h
+++ b/drivers/net/dsa/b53/b53_priv.h
@@ -80,6 +80,7 @@ enum {
 	BCM583XX_DEVICE_ID = 0x58300,
 	BCM7445_DEVICE_ID = 0x7445,
 	BCM7278_DEVICE_ID = 0x7278,
+	BCM53134_DEVICE_ID = 0x5075,
 };
 
 struct b53_pcs {
@@ -187,7 +188,8 @@ static inline int is531x5(struct b53_device *dev)
 {
 	return dev->chip_id == BCM53115_DEVICE_ID ||
 		dev->chip_id == BCM53125_DEVICE_ID ||
-		dev->chip_id == BCM53128_DEVICE_ID;
+		dev->chip_id == BCM53128_DEVICE_ID ||
+		dev->chip_id == BCM53134_DEVICE_ID;
 }
 
 static inline int is63xx(struct b53_device *dev)
@@ -215,7 +217,8 @@ static inline int is58xx(struct b53_device *dev)
 	return dev->chip_id == BCM58XX_DEVICE_ID ||
 		dev->chip_id == BCM583XX_DEVICE_ID ||
 		dev->chip_id == BCM7445_DEVICE_ID ||
-		dev->chip_id == BCM7278_DEVICE_ID;
+		dev->chip_id == BCM7278_DEVICE_ID ||
+		dev->chip_id == BCM53134_DEVICE_ID;
 }
 
 #define B53_63XX_RGMII0	4
-- 
cgit v1.2.3-70-g09d2


From 10c073e404693bf501afc44ef515316e255921e1 Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:46:56 -0700
Subject: octeon_ep: defer probe if firmware not ready

Defer probe if firmware is not ready for device usage.

Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: Satananda Burla <sburla@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/marvell/octeon_ep/octep_main.c    | 26 ++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index fdce78ceea87..0a50da52dc27 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -1016,6 +1016,26 @@ static void octep_device_cleanup(struct octep_device *oct)
 	oct->conf = NULL;
 }
 
+static bool get_fw_ready_status(struct pci_dev *pdev)
+{
+	u32 pos = 0;
+	u16 vsec_id;
+	u8 status;
+
+	while ((pos = pci_find_next_ext_capability(pdev, pos,
+						   PCI_EXT_CAP_ID_VNDR))) {
+		pci_read_config_word(pdev, pos + 4, &vsec_id);
+#define FW_STATUS_VSEC_ID  0xA3
+		if (vsec_id != FW_STATUS_VSEC_ID)
+			continue;
+
+		pci_read_config_byte(pdev, (pos + 8), &status);
+		dev_info(&pdev->dev, "Firmware ready status = %u\n", status);
+		return status;
+	}
+	return false;
+}
+
 /**
  * octep_probe() - Octeon PCI device probe handler.
  *
@@ -1051,6 +1071,12 @@ static int octep_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 
+	if (!get_fw_ready_status(pdev)) {
+		dev_notice(&pdev->dev, "Firmware not ready; defer probe.\n");
+		err = -EPROBE_DEFER;
+		goto err_alloc_netdev;
+	}
+
 	netdev = alloc_etherdev_mq(sizeof(struct octep_device),
 				   OCTEP_MAX_QUEUES);
 	if (!netdev) {
-- 
cgit v1.2.3-70-g09d2


From 24d4333233b378114106a1327d3d635a004f4387 Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:46:57 -0700
Subject: octeon_ep: poll for control messages

Poll for control messages until interrupts are enabled.
All the interrupts are enabled in ndo_open().
Add ability to listen for notifications from firmware before ndo_open().
Once interrupts are enabled, this polling is disabled and all the
messages are processed by bottom half of interrupt handler.

Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c | 47 ++++++++++------------
 .../net/ethernet/marvell/octeon_ep/octep_main.c    | 35 ++++++++++++++++
 .../net/ethernet/marvell/octeon_ep/octep_main.h    | 11 ++++-
 .../marvell/octeon_ep/octep_regs_cn9k_pf.h         |  4 ++
 4 files changed, 70 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c b/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
index 6ad88d0fe43f..adc2279bc66d 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
@@ -352,19 +352,28 @@ static void octep_setup_mbox_regs_cn93_pf(struct octep_device *oct, int q_no)
 	mbox->mbox_read_reg = oct->mmio[0].hw_addr + CN93_SDP_R_MBOX_VF_PF_DATA(q_no);
 }
 
-/* Mailbox Interrupt handler */
-static void cn93_handle_pf_mbox_intr(struct octep_device *oct)
+/* Process non-ioq interrupts required to keep pf interface running.
+ * OEI_RINT is needed for control mailbox
+ */
+static bool octep_poll_non_ioq_interrupts_cn93_pf(struct octep_device *oct)
 {
-	u64 mbox_int_val = 0ULL, val = 0ULL, qno = 0ULL;
+	bool handled = false;
+	u64 reg0;
 
-	mbox_int_val = readq(oct->mbox[0]->mbox_int_reg);
-	for (qno = 0; qno < OCTEP_MAX_VF; qno++) {
-		val = readq(oct->mbox[qno]->mbox_read_reg);
-		dev_dbg(&oct->pdev->dev,
-			"PF MBOX READ: val:%llx from VF:%llx\n", val, qno);
+	/* Check for OEI INTR */
+	reg0 = octep_read_csr64(oct, CN93_SDP_EPF_OEI_RINT);
+	if (reg0) {
+		dev_info(&oct->pdev->dev,
+			 "Received OEI_RINT intr: 0x%llx\n",
+			 reg0);
+		octep_write_csr64(oct, CN93_SDP_EPF_OEI_RINT, reg0);
+		if (reg0 & CN93_SDP_EPF_OEI_RINT_DATA_BIT_MBOX)
+			queue_work(octep_wq, &oct->ctrl_mbox_task);
+
+		handled = true;
 	}
 
-	writeq(mbox_int_val, oct->mbox[0]->mbox_int_reg);
+	return handled;
 }
 
 /* Interrupts handler for all non-queue generic interrupts. */
@@ -434,24 +443,9 @@ static irqreturn_t octep_non_ioq_intr_handler_cn93_pf(void *dev)
 		goto irq_handled;
 	}
 
-	/* Check for MBOX INTR */
-	reg_val = octep_read_csr64(oct, CN93_SDP_EPF_MBOX_RINT(0));
-	if (reg_val) {
-		dev_info(&pdev->dev,
-			 "Received MBOX_RINT intr: 0x%llx\n", reg_val);
-		cn93_handle_pf_mbox_intr(oct);
+	/* Check for MBOX INTR and OEI INTR */
+	if (octep_poll_non_ioq_interrupts_cn93_pf(oct))
 		goto irq_handled;
-	}
-
-	/* Check for OEI INTR */
-	reg_val = octep_read_csr64(oct, CN93_SDP_EPF_OEI_RINT);
-	if (reg_val) {
-		dev_info(&pdev->dev,
-			 "Received OEI_EINT intr: 0x%llx\n", reg_val);
-		octep_write_csr64(oct, CN93_SDP_EPF_OEI_RINT, reg_val);
-		queue_work(octep_wq, &oct->ctrl_mbox_task);
-		goto irq_handled;
-	}
 
 	/* Check for DMA INTR */
 	reg_val = octep_read_csr64(oct, CN93_SDP_EPF_DMA_RINT);
@@ -712,6 +706,7 @@ void octep_device_setup_cn93_pf(struct octep_device *oct)
 
 	oct->hw_ops.enable_interrupts = octep_enable_interrupts_cn93_pf;
 	oct->hw_ops.disable_interrupts = octep_disable_interrupts_cn93_pf;
+	oct->hw_ops.poll_non_ioq_interrupts = octep_poll_non_ioq_interrupts_cn93_pf;
 
 	oct->hw_ops.update_iq_read_idx = octep_update_iq_read_index_cn93_pf;
 
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index 0a50da52dc27..a3e4d9355681 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -17,6 +17,7 @@
 #include "octep_main.h"
 #include "octep_ctrl_net.h"
 
+#define OCTEP_INTR_POLL_TIME_MSECS    100
 struct workqueue_struct *octep_wq;
 
 /* Supported Devices */
@@ -511,6 +512,7 @@ static int octep_open(struct net_device *netdev)
 	ret = octep_get_link_status(oct);
 	if (!ret)
 		octep_set_link_status(oct, true);
+	oct->poll_non_ioq_intr = false;
 
 	/* Enable the input and output queues for this Octeon device */
 	oct->hw_ops.enable_io_queues(oct);
@@ -572,6 +574,11 @@ static int octep_stop(struct net_device *netdev)
 	oct->hw_ops.reset_io_queues(oct);
 	octep_free_oqs(oct);
 	octep_free_iqs(oct);
+
+	oct->poll_non_ioq_intr = true;
+	queue_delayed_work(octep_wq, &oct->intr_poll_task,
+			   msecs_to_jiffies(OCTEP_INTR_POLL_TIME_MSECS));
+
 	netdev_info(netdev, "Device stopped !!\n");
 	return 0;
 }
@@ -864,6 +871,28 @@ static const struct net_device_ops octep_netdev_ops = {
 	.ndo_change_mtu          = octep_change_mtu,
 };
 
+/**
+ * octep_intr_poll_task - work queue task to process non-ioq interrupts.
+ *
+ * @work: pointer to mbox work_struct
+ *
+ * Process non-ioq interrupts to handle control mailbox, pfvf mailbox.
+ **/
+static void octep_intr_poll_task(struct work_struct *work)
+{
+	struct octep_device *oct = container_of(work, struct octep_device,
+						intr_poll_task.work);
+
+	if (!oct->poll_non_ioq_intr) {
+		dev_info(&oct->pdev->dev, "Interrupt poll task stopped.\n");
+		return;
+	}
+
+	oct->hw_ops.poll_non_ioq_interrupts(oct);
+	queue_delayed_work(octep_wq, &oct->intr_poll_task,
+			   msecs_to_jiffies(OCTEP_INTR_POLL_TIME_MSECS));
+}
+
 /**
  * octep_ctrl_mbox_task - work queue task to handle ctrl mbox messages.
  *
@@ -1099,6 +1128,10 @@ static int octep_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 	INIT_WORK(&octep_dev->tx_timeout_task, octep_tx_timeout_task);
 	INIT_WORK(&octep_dev->ctrl_mbox_task, octep_ctrl_mbox_task);
+	INIT_DELAYED_WORK(&octep_dev->intr_poll_task, octep_intr_poll_task);
+	octep_dev->poll_non_ioq_intr = true;
+	queue_delayed_work(octep_wq, &octep_dev->intr_poll_task,
+			   msecs_to_jiffies(OCTEP_INTR_POLL_TIME_MSECS));
 
 	netdev->netdev_ops = &octep_netdev_ops;
 	octep_set_ethtool_ops(netdev);
@@ -1159,6 +1192,8 @@ static void octep_remove(struct pci_dev *pdev)
 	if (netdev->reg_state == NETREG_REGISTERED)
 		unregister_netdev(netdev);
 
+	oct->poll_non_ioq_intr = false;
+	cancel_delayed_work_sync(&oct->intr_poll_task);
 	octep_device_cleanup(oct);
 	pci_release_mem_regions(pdev);
 	free_netdev(netdev);
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.h b/drivers/net/ethernet/marvell/octeon_ep/octep_main.h
index 123ffc13754d..836d990ba3fa 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.h
@@ -73,6 +73,7 @@ struct octep_hw_ops {
 
 	void (*enable_interrupts)(struct octep_device *oct);
 	void (*disable_interrupts)(struct octep_device *oct);
+	bool (*poll_non_ioq_interrupts)(struct octep_device *oct);
 
 	void (*enable_io_queues)(struct octep_device *oct);
 	void (*disable_io_queues)(struct octep_device *oct);
@@ -270,7 +271,15 @@ struct octep_device {
 
 	/* Work entry to handle ctrl mbox interrupt */
 	struct work_struct ctrl_mbox_task;
-
+	/* Wait queue for host to firmware requests */
+	wait_queue_head_t ctrl_req_wait_q;
+	/* List of objects waiting for h2f response */
+	struct list_head ctrl_req_wait_list;
+
+	/* Enable non-ioq interrupt polling */
+	bool poll_non_ioq_intr;
+	/* Work entry to poll non-ioq interrupts */
+	struct delayed_work intr_poll_task;
 };
 
 static inline u16 OCTEP_MAJOR_REV(struct octep_device *oct)
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h b/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h
index 3d5d39a52fe6..0466fd9a002d 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h
@@ -364,4 +364,8 @@
 
 /* Number of non-queue interrupts in CN93xx */
 #define    CN93_NUM_NON_IOQ_INTR    16
+
+/* bit 0 for control mbox interrupt */
+#define CN93_SDP_EPF_OEI_RINT_DATA_BIT_MBOX	BIT_ULL(0)
+
 #endif /* _OCTEP_REGS_CN9K_PF_H_ */
-- 
cgit v1.2.3-70-g09d2


From 7c05d3d06c2382a2dee567b6ae5de4876a50de14 Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:46:58 -0700
Subject: octeon_ep: control mailbox for multiple PFs

Add control mailbox support for multiple PFs.
Update control mbox base address calculation based on PF function link.

Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c b/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
index adc2279bc66d..e2503c9bc8a1 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
@@ -13,6 +13,9 @@
 #include "octep_main.h"
 #include "octep_regs_cn9k_pf.h"
 
+#define CTRL_MBOX_MAX_PF	128
+#define CTRL_MBOX_SZ		((size_t)(0x400000 / CTRL_MBOX_MAX_PF))
+
 /* Names of Hardware non-queue generic interrupts */
 static char *cn93_non_ioq_msix_names[] = {
 	"epf_ire_rint",
@@ -198,7 +201,9 @@ static void octep_init_config_cn93_pf(struct octep_device *oct)
 {
 	struct octep_config *conf = oct->conf;
 	struct pci_dev *pdev = oct->pdev;
+	u8 link = 0;
 	u64 val;
+	int pos;
 
 	/* Read ring configuration:
 	 * PF ring count, number of VFs and rings per VF supported
@@ -234,7 +239,16 @@ static void octep_init_config_cn93_pf(struct octep_device *oct)
 	conf->msix_cfg.ioq_msix = conf->pf_ring_cfg.active_io_rings;
 	conf->msix_cfg.non_ioq_msix_names = cn93_non_ioq_msix_names;
 
-	conf->ctrl_mbox_cfg.barmem_addr = (void __iomem *)oct->mmio[2].hw_addr + (0x400000ull * 7);
+	pos = pci_find_ext_capability(oct->pdev, PCI_EXT_CAP_ID_SRIOV);
+	if (pos) {
+		pci_read_config_byte(oct->pdev,
+				     pos + PCI_SRIOV_FUNC_LINK,
+				     &link);
+		link = PCI_DEVFN(PCI_SLOT(oct->pdev->devfn), link);
+	}
+	conf->ctrl_mbox_cfg.barmem_addr = (void __iomem *)oct->mmio[2].hw_addr +
+					   (0x400000ull * 7) +
+					   (link * CTRL_MBOX_SZ);
 }
 
 /* Setup registers for a hardware Tx Queue  */
-- 
cgit v1.2.3-70-g09d2


From 577f0d1b1c5f3282fa2011177b0af692a7c21aee Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:46:59 -0700
Subject: octeon_ep: add separate mailbox command and response queues

Enhance control mailbox protocol to support following
 - separate command and response queues
    * command queue to send control commands to firmware.
    * response queue to receive responses and notifications from
      firmware.
 - variable size messages using scatter/gather

Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeon_ep/octep_ctrl_mbox.c   | 276 +++++++++--------
 .../ethernet/marvell/octeon_ep/octep_ctrl_mbox.h   |  88 +++---
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.c    | 330 ++++++++++++++-------
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.h    | 164 +++++-----
 .../net/ethernet/marvell/octeon_ep/octep_ethtool.c |   6 +-
 .../net/ethernet/marvell/octeon_ep/octep_main.c    |  76 ++---
 6 files changed, 529 insertions(+), 411 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c
index 39322e4dd100..a4ee6f3ae354 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c
@@ -24,41 +24,49 @@
 /* Time in msecs to wait for message response */
 #define OCTEP_CTRL_MBOX_MSG_WAIT_MS			10
 
-#define OCTEP_CTRL_MBOX_INFO_MAGIC_NUM_OFFSET(m)	(m)
-#define OCTEP_CTRL_MBOX_INFO_BARMEM_SZ_OFFSET(m)	((m) + 8)
-#define OCTEP_CTRL_MBOX_INFO_HOST_STATUS_OFFSET(m)	((m) + 24)
-#define OCTEP_CTRL_MBOX_INFO_FW_STATUS_OFFSET(m)	((m) + 144)
-
-#define OCTEP_CTRL_MBOX_H2FQ_INFO_OFFSET(m)		((m) + OCTEP_CTRL_MBOX_INFO_SZ)
-#define OCTEP_CTRL_MBOX_H2FQ_PROD_OFFSET(m)		(OCTEP_CTRL_MBOX_H2FQ_INFO_OFFSET(m))
-#define OCTEP_CTRL_MBOX_H2FQ_CONS_OFFSET(m)		((OCTEP_CTRL_MBOX_H2FQ_INFO_OFFSET(m)) + 4)
-#define OCTEP_CTRL_MBOX_H2FQ_ELEM_SZ_OFFSET(m)		((OCTEP_CTRL_MBOX_H2FQ_INFO_OFFSET(m)) + 8)
-#define OCTEP_CTRL_MBOX_H2FQ_ELEM_CNT_OFFSET(m)		((OCTEP_CTRL_MBOX_H2FQ_INFO_OFFSET(m)) + 12)
-
-#define OCTEP_CTRL_MBOX_F2HQ_INFO_OFFSET(m)		((m) + \
-							 OCTEP_CTRL_MBOX_INFO_SZ + \
-							 OCTEP_CTRL_MBOX_H2FQ_INFO_SZ)
-#define OCTEP_CTRL_MBOX_F2HQ_PROD_OFFSET(m)		(OCTEP_CTRL_MBOX_F2HQ_INFO_OFFSET(m))
-#define OCTEP_CTRL_MBOX_F2HQ_CONS_OFFSET(m)		((OCTEP_CTRL_MBOX_F2HQ_INFO_OFFSET(m)) + 4)
-#define OCTEP_CTRL_MBOX_F2HQ_ELEM_SZ_OFFSET(m)		((OCTEP_CTRL_MBOX_F2HQ_INFO_OFFSET(m)) + 8)
-#define OCTEP_CTRL_MBOX_F2HQ_ELEM_CNT_OFFSET(m)		((OCTEP_CTRL_MBOX_F2HQ_INFO_OFFSET(m)) + 12)
-
-#define OCTEP_CTRL_MBOX_Q_OFFSET(m, i)			((m) + \
-							 (sizeof(struct octep_ctrl_mbox_msg) * (i)))
-
-static u32 octep_ctrl_mbox_circq_inc(u32 index, u32 mask)
+/* Size of mbox info in bytes */
+#define OCTEP_CTRL_MBOX_INFO_SZ				256
+/* Size of mbox host to fw queue info in bytes */
+#define OCTEP_CTRL_MBOX_H2FQ_INFO_SZ			16
+/* Size of mbox fw to host queue info in bytes */
+#define OCTEP_CTRL_MBOX_F2HQ_INFO_SZ			16
+
+#define OCTEP_CTRL_MBOX_TOTAL_INFO_SZ	(OCTEP_CTRL_MBOX_INFO_SZ + \
+					 OCTEP_CTRL_MBOX_H2FQ_INFO_SZ + \
+					 OCTEP_CTRL_MBOX_F2HQ_INFO_SZ)
+
+#define OCTEP_CTRL_MBOX_INFO_MAGIC_NUM(m)	(m)
+#define OCTEP_CTRL_MBOX_INFO_BARMEM_SZ(m)	((m) + 8)
+#define OCTEP_CTRL_MBOX_INFO_HOST_STATUS(m)	((m) + 24)
+#define OCTEP_CTRL_MBOX_INFO_FW_STATUS(m)	((m) + 144)
+
+#define OCTEP_CTRL_MBOX_H2FQ_INFO(m)	((m) + OCTEP_CTRL_MBOX_INFO_SZ)
+#define OCTEP_CTRL_MBOX_H2FQ_PROD(m)	(OCTEP_CTRL_MBOX_H2FQ_INFO(m))
+#define OCTEP_CTRL_MBOX_H2FQ_CONS(m)	((OCTEP_CTRL_MBOX_H2FQ_INFO(m)) + 4)
+#define OCTEP_CTRL_MBOX_H2FQ_SZ(m)	((OCTEP_CTRL_MBOX_H2FQ_INFO(m)) + 8)
+
+#define OCTEP_CTRL_MBOX_F2HQ_INFO(m)	((m) + \
+					 OCTEP_CTRL_MBOX_INFO_SZ + \
+					 OCTEP_CTRL_MBOX_H2FQ_INFO_SZ)
+#define OCTEP_CTRL_MBOX_F2HQ_PROD(m)	(OCTEP_CTRL_MBOX_F2HQ_INFO(m))
+#define OCTEP_CTRL_MBOX_F2HQ_CONS(m)	((OCTEP_CTRL_MBOX_F2HQ_INFO(m)) + 4)
+#define OCTEP_CTRL_MBOX_F2HQ_SZ(m)	((OCTEP_CTRL_MBOX_F2HQ_INFO(m)) + 8)
+
+static const u32 mbox_hdr_sz = sizeof(union octep_ctrl_mbox_msg_hdr);
+
+static u32 octep_ctrl_mbox_circq_inc(u32 index, u32 inc, u32 sz)
 {
-	return (index + 1) & mask;
+	return (index + inc) % sz;
 }
 
-static u32 octep_ctrl_mbox_circq_space(u32 pi, u32 ci, u32 mask)
+static u32 octep_ctrl_mbox_circq_space(u32 pi, u32 ci, u32 sz)
 {
-	return mask - ((pi - ci) & mask);
+	return sz - (abs(pi - ci) % sz);
 }
 
-static u32 octep_ctrl_mbox_circq_depth(u32 pi, u32 ci, u32 mask)
+static u32 octep_ctrl_mbox_circq_depth(u32 pi, u32 ci, u32 sz)
 {
-	return ((pi - ci) & mask);
+	return (abs(pi - ci) % sz);
 }
 
 int octep_ctrl_mbox_init(struct octep_ctrl_mbox *mbox)
@@ -73,153 +81,170 @@ int octep_ctrl_mbox_init(struct octep_ctrl_mbox *mbox)
 		return -EINVAL;
 	}
 
-	magic_num = readq(OCTEP_CTRL_MBOX_INFO_MAGIC_NUM_OFFSET(mbox->barmem));
+	magic_num = readq(OCTEP_CTRL_MBOX_INFO_MAGIC_NUM(mbox->barmem));
 	if (magic_num != OCTEP_CTRL_MBOX_MAGIC_NUMBER) {
 		pr_info("octep_ctrl_mbox : Invalid magic number %llx\n", magic_num);
 		return -EINVAL;
 	}
 
-	status = readq(OCTEP_CTRL_MBOX_INFO_FW_STATUS_OFFSET(mbox->barmem));
+	status = readq(OCTEP_CTRL_MBOX_INFO_FW_STATUS(mbox->barmem));
 	if (status != OCTEP_CTRL_MBOX_STATUS_READY) {
 		pr_info("octep_ctrl_mbox : Firmware is not ready.\n");
 		return -EINVAL;
 	}
 
-	mbox->barmem_sz = readl(OCTEP_CTRL_MBOX_INFO_BARMEM_SZ_OFFSET(mbox->barmem));
+	mbox->barmem_sz = readl(OCTEP_CTRL_MBOX_INFO_BARMEM_SZ(mbox->barmem));
 
-	writeq(OCTEP_CTRL_MBOX_STATUS_INIT, OCTEP_CTRL_MBOX_INFO_HOST_STATUS_OFFSET(mbox->barmem));
+	writeq(OCTEP_CTRL_MBOX_STATUS_INIT,
+	       OCTEP_CTRL_MBOX_INFO_HOST_STATUS(mbox->barmem));
 
-	mbox->h2fq.elem_cnt = readl(OCTEP_CTRL_MBOX_H2FQ_ELEM_CNT_OFFSET(mbox->barmem));
-	mbox->h2fq.elem_sz = readl(OCTEP_CTRL_MBOX_H2FQ_ELEM_SZ_OFFSET(mbox->barmem));
-	mbox->h2fq.mask = (mbox->h2fq.elem_cnt - 1);
-	mutex_init(&mbox->h2fq_lock);
+	mbox->h2fq.sz = readl(OCTEP_CTRL_MBOX_H2FQ_SZ(mbox->barmem));
+	mbox->h2fq.hw_prod = OCTEP_CTRL_MBOX_H2FQ_PROD(mbox->barmem);
+	mbox->h2fq.hw_cons = OCTEP_CTRL_MBOX_H2FQ_CONS(mbox->barmem);
+	mbox->h2fq.hw_q = mbox->barmem + OCTEP_CTRL_MBOX_TOTAL_INFO_SZ;
 
-	mbox->f2hq.elem_cnt = readl(OCTEP_CTRL_MBOX_F2HQ_ELEM_CNT_OFFSET(mbox->barmem));
-	mbox->f2hq.elem_sz = readl(OCTEP_CTRL_MBOX_F2HQ_ELEM_SZ_OFFSET(mbox->barmem));
-	mbox->f2hq.mask = (mbox->f2hq.elem_cnt - 1);
-	mutex_init(&mbox->f2hq_lock);
-
-	mbox->h2fq.hw_prod = OCTEP_CTRL_MBOX_H2FQ_PROD_OFFSET(mbox->barmem);
-	mbox->h2fq.hw_cons = OCTEP_CTRL_MBOX_H2FQ_CONS_OFFSET(mbox->barmem);
-	mbox->h2fq.hw_q = mbox->barmem +
-			  OCTEP_CTRL_MBOX_INFO_SZ +
-			  OCTEP_CTRL_MBOX_H2FQ_INFO_SZ +
-			  OCTEP_CTRL_MBOX_F2HQ_INFO_SZ;
-
-	mbox->f2hq.hw_prod = OCTEP_CTRL_MBOX_F2HQ_PROD_OFFSET(mbox->barmem);
-	mbox->f2hq.hw_cons = OCTEP_CTRL_MBOX_F2HQ_CONS_OFFSET(mbox->barmem);
-	mbox->f2hq.hw_q = mbox->h2fq.hw_q +
-			  ((mbox->h2fq.elem_sz + sizeof(union octep_ctrl_mbox_msg_hdr)) *
-			   mbox->h2fq.elem_cnt);
+	mbox->f2hq.sz = readl(OCTEP_CTRL_MBOX_F2HQ_SZ(mbox->barmem));
+	mbox->f2hq.hw_prod = OCTEP_CTRL_MBOX_F2HQ_PROD(mbox->barmem);
+	mbox->f2hq.hw_cons = OCTEP_CTRL_MBOX_F2HQ_CONS(mbox->barmem);
+	mbox->f2hq.hw_q = mbox->barmem +
+			  OCTEP_CTRL_MBOX_TOTAL_INFO_SZ +
+			  mbox->h2fq.sz;
 
 	/* ensure ready state is seen after everything is initialized */
 	wmb();
-	writeq(OCTEP_CTRL_MBOX_STATUS_READY, OCTEP_CTRL_MBOX_INFO_HOST_STATUS_OFFSET(mbox->barmem));
+	writeq(OCTEP_CTRL_MBOX_STATUS_READY,
+	       OCTEP_CTRL_MBOX_INFO_HOST_STATUS(mbox->barmem));
 
 	pr_info("Octep ctrl mbox : Init successful.\n");
 
 	return 0;
 }
 
+static void
+octep_write_mbox_data(struct octep_ctrl_mbox_q *q, u32 *pi, u32 ci, void *buf, u32 w_sz)
+{
+	u8 __iomem *qbuf;
+	u32 cp_sz;
+
+	/* Assumption: Caller has ensured enough write space */
+	qbuf = (q->hw_q + *pi);
+	if (*pi < ci) {
+		/* copy entire w_sz */
+		memcpy_toio(qbuf, buf, w_sz);
+		*pi = octep_ctrl_mbox_circq_inc(*pi, w_sz, q->sz);
+	} else {
+		/* copy up to end of queue */
+		cp_sz = min((q->sz - *pi), w_sz);
+		memcpy_toio(qbuf, buf, cp_sz);
+		w_sz -= cp_sz;
+		*pi = octep_ctrl_mbox_circq_inc(*pi, cp_sz, q->sz);
+		if (w_sz) {
+			/* roll over and copy remaining w_sz */
+			buf += cp_sz;
+			qbuf = (q->hw_q + *pi);
+			memcpy_toio(qbuf, buf, w_sz);
+			*pi = octep_ctrl_mbox_circq_inc(*pi, w_sz, q->sz);
+		}
+	}
+}
+
 int octep_ctrl_mbox_send(struct octep_ctrl_mbox *mbox, struct octep_ctrl_mbox_msg *msg)
 {
-	unsigned long timeout = msecs_to_jiffies(OCTEP_CTRL_MBOX_MSG_TIMEOUT_MS);
-	unsigned long period = msecs_to_jiffies(OCTEP_CTRL_MBOX_MSG_WAIT_MS);
+	struct octep_ctrl_mbox_msg_buf *sg;
 	struct octep_ctrl_mbox_q *q;
-	unsigned long expire;
-	u64 *mbuf, *word0;
-	u8 __iomem *qidx;
-	u16 pi, ci;
-	int i;
+	u32 pi, ci, buf_sz, w_sz;
+	int s;
 
 	if (!mbox || !msg)
 		return -EINVAL;
 
+	if (readq(OCTEP_CTRL_MBOX_INFO_FW_STATUS(mbox->barmem)) != OCTEP_CTRL_MBOX_STATUS_READY)
+		return -EIO;
+
+	mutex_lock(&mbox->h2fq_lock);
 	q = &mbox->h2fq;
 	pi = readl(q->hw_prod);
 	ci = readl(q->hw_cons);
 
-	if (!octep_ctrl_mbox_circq_space(pi, ci, q->mask))
-		return -ENOMEM;
-
-	qidx = OCTEP_CTRL_MBOX_Q_OFFSET(q->hw_q, pi);
-	mbuf = (u64 *)msg->msg;
-	word0 = &msg->hdr.word0;
-
-	mutex_lock(&mbox->h2fq_lock);
-	for (i = 1; i <= msg->hdr.sizew; i++)
-		writeq(*mbuf++, (qidx + (i * 8)));
-
-	writeq(*word0, qidx);
+	if (octep_ctrl_mbox_circq_space(pi, ci, q->sz) < (msg->hdr.s.sz + mbox_hdr_sz)) {
+		mutex_unlock(&mbox->f2hq_lock);
+		return -EAGAIN;
+	}
 
-	pi = octep_ctrl_mbox_circq_inc(pi, q->mask);
+	octep_write_mbox_data(q, &pi, ci, (void *)&msg->hdr, mbox_hdr_sz);
+	buf_sz = msg->hdr.s.sz;
+	for (s = 0; ((s < msg->sg_num) && (buf_sz > 0)); s++) {
+		sg = &msg->sg_list[s];
+		w_sz = (sg->sz <= buf_sz) ? sg->sz : buf_sz;
+		octep_write_mbox_data(q, &pi, ci, sg->msg, w_sz);
+		buf_sz -= w_sz;
+	}
 	writel(pi, q->hw_prod);
 	mutex_unlock(&mbox->h2fq_lock);
 
-	/* don't check for notification response */
-	if (msg->hdr.flags & OCTEP_CTRL_MBOX_MSG_HDR_FLAG_NOTIFY)
-		return 0;
-
-	expire = jiffies + timeout;
-	while (true) {
-		*word0 = readq(qidx);
-		if (msg->hdr.flags == OCTEP_CTRL_MBOX_MSG_HDR_FLAG_RESP)
-			break;
-		schedule_timeout_interruptible(period);
-		if (signal_pending(current) || time_after(jiffies, expire)) {
-			pr_info("octep_ctrl_mbox: Timed out\n");
-			return -EBUSY;
+	return 0;
+}
+
+static void
+octep_read_mbox_data(struct octep_ctrl_mbox_q *q, u32 pi, u32 *ci, void *buf, u32 r_sz)
+{
+	u8 __iomem *qbuf;
+	u32 cp_sz;
+
+	/* Assumption: Caller has ensured enough read space */
+	qbuf = (q->hw_q + *ci);
+	if (*ci < pi) {
+		/* copy entire r_sz */
+		memcpy_fromio(buf, qbuf, r_sz);
+		*ci = octep_ctrl_mbox_circq_inc(*ci, r_sz, q->sz);
+	} else {
+		/* copy up to end of queue */
+		cp_sz = min((q->sz - *ci), r_sz);
+		memcpy_fromio(buf, qbuf, cp_sz);
+		r_sz -= cp_sz;
+		*ci = octep_ctrl_mbox_circq_inc(*ci, cp_sz, q->sz);
+		if (r_sz) {
+			/* roll over and copy remaining r_sz */
+			buf += cp_sz;
+			qbuf = (q->hw_q + *ci);
+			memcpy_fromio(buf, qbuf, r_sz);
+			*ci = octep_ctrl_mbox_circq_inc(*ci, r_sz, q->sz);
 		}
 	}
-	mbuf = (u64 *)msg->msg;
-	for (i = 1; i <= msg->hdr.sizew; i++)
-		*mbuf++ = readq(qidx + (i * 8));
-
-	return 0;
 }
 
 int octep_ctrl_mbox_recv(struct octep_ctrl_mbox *mbox, struct octep_ctrl_mbox_msg *msg)
 {
+	struct octep_ctrl_mbox_msg_buf *sg;
+	u32 pi, ci, r_sz, buf_sz, q_depth;
 	struct octep_ctrl_mbox_q *q;
-	u32 count, pi, ci;
-	u8 __iomem *qidx;
-	u64 *mbuf;
-	int i;
+	int s;
 
-	if (!mbox || !msg)
-		return -EINVAL;
+	if (readq(OCTEP_CTRL_MBOX_INFO_FW_STATUS(mbox->barmem)) != OCTEP_CTRL_MBOX_STATUS_READY)
+		return -EIO;
 
+	mutex_lock(&mbox->f2hq_lock);
 	q = &mbox->f2hq;
 	pi = readl(q->hw_prod);
 	ci = readl(q->hw_cons);
-	count = octep_ctrl_mbox_circq_depth(pi, ci, q->mask);
-	if (!count)
-		return -EAGAIN;
-
-	qidx = OCTEP_CTRL_MBOX_Q_OFFSET(q->hw_q, ci);
-	mbuf = (u64 *)msg->msg;
 
-	mutex_lock(&mbox->f2hq_lock);
-
-	msg->hdr.word0 = readq(qidx);
-	for (i = 1; i <= msg->hdr.sizew; i++)
-		*mbuf++ = readq(qidx + (i * 8));
+	q_depth = octep_ctrl_mbox_circq_depth(pi, ci, q->sz);
+	if (q_depth < mbox_hdr_sz) {
+		mutex_unlock(&mbox->f2hq_lock);
+		return -EAGAIN;
+	}
 
-	ci = octep_ctrl_mbox_circq_inc(ci, q->mask);
+	octep_read_mbox_data(q, pi, &ci, (void *)&msg->hdr, mbox_hdr_sz);
+	buf_sz = msg->hdr.s.sz;
+	for (s = 0; ((s < msg->sg_num) && (buf_sz > 0)); s++) {
+		sg = &msg->sg_list[s];
+		r_sz = (sg->sz <= buf_sz) ? sg->sz : buf_sz;
+		octep_read_mbox_data(q, pi, &ci, sg->msg, r_sz);
+		buf_sz -= r_sz;
+	}
 	writel(ci, q->hw_cons);
-
 	mutex_unlock(&mbox->f2hq_lock);
 
-	if (msg->hdr.flags != OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ || !mbox->process_req)
-		return 0;
-
-	mbox->process_req(mbox->user_ctx, msg);
-	mbuf = (u64 *)msg->msg;
-	for (i = 1; i <= msg->hdr.sizew; i++)
-		writeq(*mbuf++, (qidx + (i * 8)));
-
-	writeq(msg->hdr.word0, qidx);
-
 	return 0;
 }
 
@@ -227,18 +252,17 @@ int octep_ctrl_mbox_uninit(struct octep_ctrl_mbox *mbox)
 {
 	if (!mbox)
 		return -EINVAL;
+	if (!mbox->barmem)
+		return -EINVAL;
 
-	writeq(OCTEP_CTRL_MBOX_STATUS_UNINIT,
-	       OCTEP_CTRL_MBOX_INFO_HOST_STATUS_OFFSET(mbox->barmem));
+	writeq(OCTEP_CTRL_MBOX_STATUS_INVALID,
+	       OCTEP_CTRL_MBOX_INFO_HOST_STATUS(mbox->barmem));
 	/* ensure uninit state is written before uninitialization */
 	wmb();
 
 	mutex_destroy(&mbox->h2fq_lock);
 	mutex_destroy(&mbox->f2hq_lock);
 
-	writeq(OCTEP_CTRL_MBOX_STATUS_INVALID,
-	       OCTEP_CTRL_MBOX_INFO_HOST_STATUS_OFFSET(mbox->barmem));
-
 	pr_info("Octep ctrl mbox : Uninit successful.\n");
 
 	return 0;
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.h b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.h
index 2dc5753cfec6..9c4ff0fba6a0 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.h
@@ -27,50 +27,39 @@
  * |-------------------------------------------|
  * |producer index (4 bytes)                   |
  * |consumer index (4 bytes)                   |
- * |element size (4 bytes)                     |
- * |element count (4 bytes)                    |
+ * |max element size (4 bytes)                 |
+ * |reserved (4 bytes)                         |
  * |===========================================|
  * |Fw to Host Queue info (16 bytes)           |
  * |-------------------------------------------|
  * |producer index (4 bytes)                   |
  * |consumer index (4 bytes)                   |
- * |element size (4 bytes)                     |
- * |element count (4 bytes)                    |
+ * |max element size (4 bytes)                 |
+ * |reserved (4 bytes)                         |
  * |===========================================|
- * |Host to Fw Queue                           |
+ * |Host to Fw Queue ((total size-288/2) bytes)|
  * |-------------------------------------------|
- * |((elem_sz + hdr(8 bytes)) * elem_cnt) bytes|
+ * |                                           |
  * |===========================================|
  * |===========================================|
- * |Fw to Host Queue                           |
+ * |Fw to Host Queue ((total size-288/2) bytes)|
  * |-------------------------------------------|
- * |((elem_sz + hdr(8 bytes)) * elem_cnt) bytes|
+ * |                                           |
  * |===========================================|
  */
 
 #define OCTEP_CTRL_MBOX_MAGIC_NUMBER			0xdeaddeadbeefbeefull
 
-/* Size of mbox info in bytes */
-#define OCTEP_CTRL_MBOX_INFO_SZ				256
-/* Size of mbox host to target queue info in bytes */
-#define OCTEP_CTRL_MBOX_H2FQ_INFO_SZ			16
-/* Size of mbox target to host queue info in bytes */
-#define OCTEP_CTRL_MBOX_F2HQ_INFO_SZ			16
-/* Size of mbox queue in bytes */
-#define OCTEP_CTRL_MBOX_Q_SZ(sz, cnt)			(((sz) + 8) * (cnt))
-/* Size of mbox in bytes */
-#define OCTEP_CTRL_MBOX_SZ(hsz, hcnt, fsz, fcnt)	(OCTEP_CTRL_MBOX_INFO_SZ + \
-							 OCTEP_CTRL_MBOX_H2FQ_INFO_SZ + \
-							 OCTEP_CTRL_MBOX_F2HQ_INFO_SZ + \
-							 OCTEP_CTRL_MBOX_Q_SZ(hsz, hcnt) + \
-							 OCTEP_CTRL_MBOX_Q_SZ(fsz, fcnt))
-
 /* Valid request message */
 #define OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ		BIT(0)
 /* Valid response message */
 #define OCTEP_CTRL_MBOX_MSG_HDR_FLAG_RESP		BIT(1)
 /* Valid notification, no response required */
 #define OCTEP_CTRL_MBOX_MSG_HDR_FLAG_NOTIFY		BIT(2)
+/* Valid custom message */
+#define OCTEP_CTRL_MBOX_MSG_HDR_FLAG_CUSTOM		BIT(3)
+
+#define OCTEP_CTRL_MBOX_MSG_DESC_MAX			4
 
 enum octep_ctrl_mbox_status {
 	OCTEP_CTRL_MBOX_STATUS_INVALID = 0,
@@ -81,31 +70,48 @@ enum octep_ctrl_mbox_status {
 
 /* mbox message */
 union octep_ctrl_mbox_msg_hdr {
-	u64 word0;
+	u64 words[2];
 	struct {
+		/* must be 0 */
+		u16 reserved1:15;
+		/* vf_idx is valid if 1 */
+		u16 is_vf:1;
+		/* sender vf index 0-(n-1), 0 if (is_vf==0) */
+		u16 vf_idx;
+		/* total size of message excluding header */
+		u32 sz;
 		/* OCTEP_CTRL_MBOX_MSG_HDR_FLAG_* */
 		u32 flags;
-		/* size of message in words excluding header */
-		u32 sizew;
-	};
+		/* identifier to match responses */
+		u16 msg_id;
+		u16 reserved2;
+	} s;
+};
+
+/* mbox message buffer */
+struct octep_ctrl_mbox_msg_buf {
+	u32 reserved1;
+	u16 reserved2;
+	/* size of buffer */
+	u16 sz;
+	/* pointer to message buffer */
+	void *msg;
 };
 
 /* mbox message */
 struct octep_ctrl_mbox_msg {
 	/* mbox transaction header */
 	union octep_ctrl_mbox_msg_hdr hdr;
-	/* pointer to message buffer */
-	void *msg;
+	/* number of sg buffer's */
+	int sg_num;
+	/* message buffer's */
+	struct octep_ctrl_mbox_msg_buf sg_list[OCTEP_CTRL_MBOX_MSG_DESC_MAX];
 };
 
 /* Mbox queue */
 struct octep_ctrl_mbox_q {
-	/* q element size, should be aligned to unsigned long */
-	u16 elem_sz;
-	/* q element count, should be power of 2 */
-	u16 elem_cnt;
-	/* q mask */
-	u16 mask;
+	/* size of queue buffer */
+	u32 sz;
 	/* producer address in bar mem */
 	u8 __iomem *hw_prod;
 	/* consumer address in bar mem */
@@ -115,16 +121,10 @@ struct octep_ctrl_mbox_q {
 };
 
 struct octep_ctrl_mbox {
-	/* host driver version */
-	u64 version;
 	/* size of bar memory */
 	u32 barmem_sz;
 	/* pointer to BAR memory */
 	u8 __iomem *barmem;
-	/* user context for callback, can be null */
-	void *user_ctx;
-	/* callback handler for processing request, called from octep_ctrl_mbox_recv */
-	int (*process_req)(void *user_ctx, struct octep_ctrl_mbox_msg *msg);
 	/* host-to-fw queue */
 	struct octep_ctrl_mbox_q h2fq;
 	/* fw-to-host queue */
@@ -146,6 +146,8 @@ int octep_ctrl_mbox_init(struct octep_ctrl_mbox *mbox);
 /* Send mbox message.
  *
  * @param mbox: non-null pointer to struct octep_ctrl_mbox.
+ * @param msg:  non-null pointer to struct octep_ctrl_mbox_msg.
+ *              Caller should fill msg.sz and msg.desc.sz for each message.
  *
  * return value: 0 on success, -errno on failure.
  */
@@ -154,6 +156,8 @@ int octep_ctrl_mbox_send(struct octep_ctrl_mbox *mbox, struct octep_ctrl_mbox_ms
 /* Retrieve mbox message.
  *
  * @param mbox: non-null pointer to struct octep_ctrl_mbox.
+ * @param msg:  non-null pointer to struct octep_ctrl_mbox_msg.
+ *              Caller should fill msg.sz and msg.desc.sz for each message.
  *
  * return value: 0 on success, -errno on failure.
  */
@@ -161,7 +165,7 @@ int octep_ctrl_mbox_recv(struct octep_ctrl_mbox *mbox, struct octep_ctrl_mbox_ms
 
 /* Uninitialize control mbox.
  *
- * @param ep: non-null pointer to struct octep_ctrl_mbox.
+ * @param mbox: non-null pointer to struct octep_ctrl_mbox.
  *
  * return value: 0 on success, -errno on failure.
  */
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
index 7c00c896ab98..8d4d74e18a67 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
@@ -8,134 +8,194 @@
 #include <linux/types.h>
 #include <linux/etherdevice.h>
 #include <linux/pci.h>
+#include <linux/wait.h>
 
 #include "octep_config.h"
 #include "octep_main.h"
 #include "octep_ctrl_net.h"
 
-int octep_get_link_status(struct octep_device *oct)
+static const u32 req_hdr_sz = sizeof(union octep_ctrl_net_req_hdr);
+static const u32 mtu_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_mtu);
+static const u32 mac_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_mac);
+static const u32 state_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_state);
+static const u32 link_info_sz = sizeof(struct octep_ctrl_net_link_info);
+static const u32 get_stats_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_get_stats);
+static atomic_t ctrl_net_msg_id;
+
+static void init_send_req(struct octep_ctrl_mbox_msg *msg, void *buf, u16 sz)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_net_h2f_resp *resp;
-	struct octep_ctrl_mbox_msg msg = {};
-	int err;
+	msg->hdr.s.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
+	msg->hdr.s.msg_id = atomic_inc_return(&ctrl_net_msg_id) &
+			    GENMASK(sizeof(msg->hdr.s.msg_id) * BITS_PER_BYTE, 0);
+	msg->hdr.s.sz = req_hdr_sz + sz;
+	msg->sg_num = 1;
+	msg->sg_list[0].msg = buf;
+	msg->sg_list[0].sz = msg->hdr.s.sz;
+}
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_STATUS;
-	req.link.cmd = OCTEP_CTRL_NET_CMD_GET;
+static int octep_send_mbox_req(struct octep_device *oct,
+			       struct octep_ctrl_net_wait_data *d,
+			       bool wait_for_response)
+{
+	int err, ret;
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_STATE_REQ_SZW;
-	msg.msg = &req;
-	err = octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
-	if (err)
+	err = octep_ctrl_mbox_send(&oct->ctrl_mbox, &d->msg);
+	if (err < 0)
 		return err;
 
-	resp = (struct octep_ctrl_net_h2f_resp *)&req;
-	return resp->link.state;
+	if (!wait_for_response)
+		return 0;
+
+	d->done = 0;
+	INIT_LIST_HEAD(&d->list);
+	list_add_tail(&d->list, &oct->ctrl_req_wait_list);
+	ret = wait_event_interruptible_timeout(oct->ctrl_req_wait_q,
+					       (d->done != 0),
+					       jiffies + msecs_to_jiffies(500));
+	list_del(&d->list);
+	if (ret == 0 || ret == 1)
+		return -EAGAIN;
+
+	/**
+	 * (ret == 0)  cond = false && timeout, return 0
+	 * (ret < 0) interrupted by signal, return 0
+	 * (ret == 1) cond = true && timeout, return 1
+	 * (ret >= 1) cond = true && !timeout, return 1
+	 */
+
+	if (d->data.resp.hdr.s.reply != OCTEP_CTRL_NET_REPLY_OK)
+		return -EAGAIN;
+
+	return 0;
+}
+
+int octep_ctrl_net_init(struct octep_device *oct)
+{
+	struct octep_ctrl_mbox *ctrl_mbox;
+	struct pci_dev *pdev = oct->pdev;
+	int ret;
+
+	init_waitqueue_head(&oct->ctrl_req_wait_q);
+	INIT_LIST_HEAD(&oct->ctrl_req_wait_list);
+
+	/* Initialize control mbox */
+	ctrl_mbox = &oct->ctrl_mbox;
+	ctrl_mbox->barmem = CFG_GET_CTRL_MBOX_MEM_ADDR(oct->conf);
+	ret = octep_ctrl_mbox_init(ctrl_mbox);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to initialize control mbox\n");
+		return ret;
+	}
+	oct->ctrl_mbox_ifstats_offset = ctrl_mbox->barmem_sz;
+
+	return 0;
 }
 
-void octep_set_link_status(struct octep_device *oct, bool up)
+int octep_ctrl_net_get_link_status(struct octep_device *oct)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_mbox_msg msg = {};
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
+	int err;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_STATUS;
-	req.link.cmd = OCTEP_CTRL_NET_CMD_SET;
-	req.link.state = (up) ? OCTEP_CTRL_NET_STATE_UP : OCTEP_CTRL_NET_STATE_DOWN;
+	init_send_req(&d.msg, (void *)req, state_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_STATUS;
+	req->link.cmd = OCTEP_CTRL_NET_CMD_GET;
+	err = octep_send_mbox_req(oct, &d, true);
+	if (err < 0)
+		return err;
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_STATE_REQ_SZW;
-	msg.msg = &req;
-	octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
+	return d.data.resp.link.state;
 }
 
-void octep_set_rx_state(struct octep_device *oct, bool up)
+int octep_ctrl_net_set_link_status(struct octep_device *oct, bool up,
+				   bool wait_for_response)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_mbox_msg msg = {};
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_RX_STATE;
-	req.link.cmd = OCTEP_CTRL_NET_CMD_SET;
-	req.link.state = (up) ? OCTEP_CTRL_NET_STATE_UP : OCTEP_CTRL_NET_STATE_DOWN;
+	init_send_req(&d.msg, req, state_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_STATUS;
+	req->link.cmd = OCTEP_CTRL_NET_CMD_SET;
+	req->link.state = (up) ? OCTEP_CTRL_NET_STATE_UP :
+				OCTEP_CTRL_NET_STATE_DOWN;
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_STATE_REQ_SZW;
-	msg.msg = &req;
-	octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
+	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_get_mac_addr(struct octep_device *oct, u8 *addr)
+int octep_ctrl_net_set_rx_state(struct octep_device *oct, bool up,
+				bool wait_for_response)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_net_h2f_resp *resp;
-	struct octep_ctrl_mbox_msg msg = {};
-	int err;
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_MAC;
-	req.link.cmd = OCTEP_CTRL_NET_CMD_GET;
+	init_send_req(&d.msg, req, state_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_RX_STATE;
+	req->link.cmd = OCTEP_CTRL_NET_CMD_SET;
+	req->link.state = (up) ? OCTEP_CTRL_NET_STATE_UP :
+				OCTEP_CTRL_NET_STATE_DOWN;
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_MAC_REQ_SZW;
-	msg.msg = &req;
-	err = octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
-	if (err)
+	return octep_send_mbox_req(oct, &d, wait_for_response);
+}
+
+int octep_ctrl_net_get_mac_addr(struct octep_device *oct, u8 *addr)
+{
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
+	int err;
+
+	init_send_req(&d.msg, req, mac_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_MAC;
+	req->link.cmd = OCTEP_CTRL_NET_CMD_GET;
+	err = octep_send_mbox_req(oct, &d, true);
+	if (err < 0)
 		return err;
 
-	resp = (struct octep_ctrl_net_h2f_resp *)&req;
-	memcpy(addr, resp->mac.addr, ETH_ALEN);
+	memcpy(addr, d.data.resp.mac.addr, ETH_ALEN);
 
-	return err;
+	return 0;
 }
 
-int octep_set_mac_addr(struct octep_device *oct, u8 *addr)
+int octep_ctrl_net_set_mac_addr(struct octep_device *oct, u8 *addr,
+				bool wait_for_response)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_mbox_msg msg = {};
-
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_MAC;
-	req.mac.cmd = OCTEP_CTRL_NET_CMD_SET;
-	memcpy(&req.mac.addr, addr, ETH_ALEN);
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_MAC_REQ_SZW;
-	msg.msg = &req;
+	init_send_req(&d.msg, req, mac_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_MAC;
+	req->mac.cmd = OCTEP_CTRL_NET_CMD_SET;
+	memcpy(&req->mac.addr, addr, ETH_ALEN);
 
-	return octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
+	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_set_mtu(struct octep_device *oct, int mtu)
+int octep_ctrl_net_set_mtu(struct octep_device *oct, int mtu,
+			   bool wait_for_response)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_mbox_msg msg = {};
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_MTU;
-	req.mtu.cmd = OCTEP_CTRL_NET_CMD_SET;
-	req.mtu.val = mtu;
+	init_send_req(&d.msg, req, mtu_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_MTU;
+	req->mtu.cmd = OCTEP_CTRL_NET_CMD_SET;
+	req->mtu.val = mtu;
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_MTU_REQ_SZW;
-	msg.msg = &req;
-
-	return octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
+	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_get_if_stats(struct octep_device *oct)
+int octep_ctrl_net_get_if_stats(struct octep_device *oct)
 {
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 	void __iomem *iface_rx_stats;
 	void __iomem *iface_tx_stats;
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_mbox_msg msg = {};
 	int err;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_GET_IF_STATS;
-	req.mac.cmd = OCTEP_CTRL_NET_CMD_GET;
-	req.get_stats.offset = oct->ctrl_mbox_ifstats_offset;
-
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_GET_STATS_REQ_SZW;
-	msg.msg = &req;
-	err = octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
-	if (err)
+	init_send_req(&d.msg, req, get_stats_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_GET_IF_STATS;
+	req->get_stats.offset = oct->ctrl_mbox_ifstats_offset;
+	err = octep_send_mbox_req(oct, &d, true);
+	if (err < 0)
 		return err;
 
 	iface_rx_stats = oct->ctrl_mbox.barmem + oct->ctrl_mbox_ifstats_offset;
@@ -144,51 +204,101 @@ int octep_get_if_stats(struct octep_device *oct)
 	memcpy_fromio(&oct->iface_rx_stats, iface_rx_stats, sizeof(struct octep_iface_rx_stats));
 	memcpy_fromio(&oct->iface_tx_stats, iface_tx_stats, sizeof(struct octep_iface_tx_stats));
 
-	return err;
+	return 0;
 }
 
-int octep_get_link_info(struct octep_device *oct)
+int octep_ctrl_net_get_link_info(struct octep_device *oct)
 {
-	struct octep_ctrl_net_h2f_req req = {};
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 	struct octep_ctrl_net_h2f_resp *resp;
-	struct octep_ctrl_mbox_msg msg = {};
 	int err;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_INFO;
-	req.mac.cmd = OCTEP_CTRL_NET_CMD_GET;
-
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_LINK_INFO_REQ_SZW;
-	msg.msg = &req;
-	err = octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
-	if (err)
+	init_send_req(&d.msg, req, link_info_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_INFO;
+	req->link_info.cmd = OCTEP_CTRL_NET_CMD_GET;
+	err = octep_send_mbox_req(oct, &d, true);
+	if (err < 0)
 		return err;
 
-	resp = (struct octep_ctrl_net_h2f_resp *)&req;
+	resp = &d.data.resp;
 	oct->link_info.supported_modes = resp->link_info.supported_modes;
 	oct->link_info.advertised_modes = resp->link_info.advertised_modes;
 	oct->link_info.autoneg = resp->link_info.autoneg;
 	oct->link_info.pause = resp->link_info.pause;
 	oct->link_info.speed = resp->link_info.speed;
 
-	return err;
+	return 0;
+}
+
+int octep_ctrl_net_set_link_info(struct octep_device *oct,
+				 struct octep_iface_link_info *link_info,
+				 bool wait_for_response)
+{
+	struct octep_ctrl_net_wait_data d = {0};
+	struct octep_ctrl_net_h2f_req *req = &d.data.req;
+
+	init_send_req(&d.msg, req, link_info_sz);
+	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_INFO;
+	req->link_info.cmd = OCTEP_CTRL_NET_CMD_SET;
+	req->link_info.info.advertised_modes = link_info->advertised_modes;
+	req->link_info.info.autoneg = link_info->autoneg;
+	req->link_info.info.pause = link_info->pause;
+	req->link_info.info.speed = link_info->speed;
+
+	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_set_link_info(struct octep_device *oct, struct octep_iface_link_info *link_info)
+static void process_mbox_resp(struct octep_device *oct,
+			      struct octep_ctrl_mbox_msg *msg)
 {
-	struct octep_ctrl_net_h2f_req req = {};
-	struct octep_ctrl_mbox_msg msg = {};
+	struct octep_ctrl_net_wait_data *pos, *n;
+
+	list_for_each_entry_safe(pos, n, &oct->ctrl_req_wait_list, list) {
+		if (pos->msg.hdr.s.msg_id == msg->hdr.s.msg_id) {
+			memcpy(&pos->data.resp,
+			       msg->sg_list[0].msg,
+			       msg->hdr.s.sz);
+			pos->done = 1;
+			wake_up_interruptible_all(&oct->ctrl_req_wait_q);
+			break;
+		}
+	}
+}
+
+void octep_ctrl_net_recv_fw_messages(struct octep_device *oct)
+{
+	static u16 msg_sz = sizeof(union octep_ctrl_net_max_data);
+	union octep_ctrl_net_max_data data = {0};
+	struct octep_ctrl_mbox_msg msg = {0};
+	int ret;
+
+	msg.hdr.s.sz = msg_sz;
+	msg.sg_num = 1;
+	msg.sg_list[0].sz = msg_sz;
+	msg.sg_list[0].msg = &data;
+	while (true) {
+		/* mbox will overwrite msg.hdr.s.sz so initialize it */
+		msg.hdr.s.sz = msg_sz;
+		ret = octep_ctrl_mbox_recv(&oct->ctrl_mbox, (struct octep_ctrl_mbox_msg *)&msg);
+		if (ret < 0)
+			break;
+
+		if (msg.hdr.s.flags & OCTEP_CTRL_MBOX_MSG_HDR_FLAG_RESP)
+			process_mbox_resp(oct, &msg);
+	}
+}
+
+int octep_ctrl_net_uninit(struct octep_device *oct)
+{
+	struct octep_ctrl_net_wait_data *pos, *n;
+
+	list_for_each_entry_safe(pos, n, &oct->ctrl_req_wait_list, list)
+		pos->done = 1;
 
-	req.hdr.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_INFO;
-	req.link_info.cmd = OCTEP_CTRL_NET_CMD_SET;
-	req.link_info.info.advertised_modes = link_info->advertised_modes;
-	req.link_info.info.autoneg = link_info->autoneg;
-	req.link_info.info.pause = link_info->pause;
-	req.link_info.info.speed = link_info->speed;
+	wake_up_interruptible_all(&oct->ctrl_req_wait_q);
 
-	msg.hdr.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
-	msg.hdr.sizew = OCTEP_CTRL_NET_H2F_LINK_INFO_REQ_SZW;
-	msg.msg = &req;
+	octep_ctrl_mbox_uninit(&oct->ctrl_mbox);
 
-	return octep_ctrl_mbox_send(&oct->ctrl_mbox, &msg);
+	return 0;
 }
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
index f23b58381322..aba373ec28a4 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
@@ -45,15 +45,18 @@ enum octep_ctrl_net_f2h_cmd {
 	OCTEP_CTRL_NET_F2H_CMD_LINK_STATUS,
 };
 
-struct octep_ctrl_net_req_hdr {
-	/* sender id */
-	u16 sender;
-	/* receiver id */
-	u16 receiver;
-	/* octep_ctrl_net_h2t_cmd */
-	u16 cmd;
-	/* reserved */
-	u16 rsvd0;
+union octep_ctrl_net_req_hdr {
+	u64 words[1];
+	struct {
+		/* sender id */
+		u16 sender;
+		/* receiver id */
+		u16 receiver;
+		/* octep_ctrl_net_h2t_cmd */
+		u16 cmd;
+		/* reserved */
+		u16 rsvd0;
+	} s;
 };
 
 /* get/set mtu request */
@@ -110,7 +113,7 @@ struct octep_ctrl_net_h2f_req_cmd_link_info {
 
 /* Host to fw request data */
 struct octep_ctrl_net_h2f_req {
-	struct octep_ctrl_net_req_hdr hdr;
+	union octep_ctrl_net_req_hdr hdr;
 	union {
 		struct octep_ctrl_net_h2f_req_cmd_mtu mtu;
 		struct octep_ctrl_net_h2f_req_cmd_mac mac;
@@ -121,15 +124,18 @@ struct octep_ctrl_net_h2f_req {
 	};
 } __packed;
 
-struct octep_ctrl_net_resp_hdr {
-	/* sender id */
-	u16 sender;
-	/* receiver id */
-	u16 receiver;
-	/* octep_ctrl_net_h2t_cmd */
-	u16 cmd;
-	/* octep_ctrl_net_reply */
-	u16 reply;
+union octep_ctrl_net_resp_hdr {
+	u64 words[1];
+	struct {
+		/* sender id */
+		u16 sender;
+		/* receiver id */
+		u16 receiver;
+		/* octep_ctrl_net_h2t_cmd */
+		u16 cmd;
+		/* octep_ctrl_net_reply */
+		u16 reply;
+	} s;
 };
 
 /* get mtu response */
@@ -152,7 +158,7 @@ struct octep_ctrl_net_h2f_resp_cmd_state {
 
 /* Host to fw response data */
 struct octep_ctrl_net_h2f_resp {
-	struct octep_ctrl_net_resp_hdr hdr;
+	union octep_ctrl_net_resp_hdr hdr;
 	union {
 		struct octep_ctrl_net_h2f_resp_cmd_mtu mtu;
 		struct octep_ctrl_net_h2f_resp_cmd_mac mac;
@@ -170,7 +176,7 @@ struct octep_ctrl_net_f2h_req_cmd_state {
 
 /* Fw to host request data */
 struct octep_ctrl_net_f2h_req {
-	struct octep_ctrl_net_req_hdr hdr;
+	union octep_ctrl_net_req_hdr hdr;
 	union {
 		struct octep_ctrl_net_f2h_req_cmd_state link;
 	};
@@ -178,56 +184,34 @@ struct octep_ctrl_net_f2h_req {
 
 /* Fw to host response data */
 struct octep_ctrl_net_f2h_resp {
-	struct octep_ctrl_net_resp_hdr hdr;
+	union octep_ctrl_net_resp_hdr hdr;
 };
 
-/* Size of host to fw octep_ctrl_mbox queue element */
-union octep_ctrl_net_h2f_data_sz {
+/* Max data size to be transferred over mbox */
+union octep_ctrl_net_max_data {
 	struct octep_ctrl_net_h2f_req h2f_req;
 	struct octep_ctrl_net_h2f_resp h2f_resp;
-};
-
-/* Size of fw to host octep_ctrl_mbox queue element */
-union octep_ctrl_net_f2h_data_sz {
 	struct octep_ctrl_net_f2h_req f2h_req;
 	struct octep_ctrl_net_f2h_resp f2h_resp;
 };
 
-/* size of host to fw data in words */
-#define OCTEP_CTRL_NET_H2F_DATA_SZW		((sizeof(union octep_ctrl_net_h2f_data_sz)) / \
-						 (sizeof(unsigned long)))
-
-/* size of fw to host data in words */
-#define OCTEP_CTRL_NET_F2H_DATA_SZW		((sizeof(union octep_ctrl_net_f2h_data_sz)) / \
-						 (sizeof(unsigned long)))
-
-/* size in words of get/set mtu request */
-#define OCTEP_CTRL_NET_H2F_MTU_REQ_SZW			2
-/* size in words of get/set mac request */
-#define OCTEP_CTRL_NET_H2F_MAC_REQ_SZW			2
-/* size in words of get stats request */
-#define OCTEP_CTRL_NET_H2F_GET_STATS_REQ_SZW		2
-/* size in words of get/set state request */
-#define OCTEP_CTRL_NET_H2F_STATE_REQ_SZW		2
-/* size in words of get/set link info request */
-#define OCTEP_CTRL_NET_H2F_LINK_INFO_REQ_SZW		4
-
-/* size in words of get mtu response */
-#define OCTEP_CTRL_NET_H2F_GET_MTU_RESP_SZW		2
-/* size in words of set mtu response */
-#define OCTEP_CTRL_NET_H2F_SET_MTU_RESP_SZW		1
-/* size in words of get mac response */
-#define OCTEP_CTRL_NET_H2F_GET_MAC_RESP_SZW		2
-/* size in words of set mac response */
-#define OCTEP_CTRL_NET_H2F_SET_MAC_RESP_SZW		1
-/* size in words of get state request */
-#define OCTEP_CTRL_NET_H2F_GET_STATE_RESP_SZW		2
-/* size in words of set state request */
-#define OCTEP_CTRL_NET_H2F_SET_STATE_RESP_SZW		1
-/* size in words of get link info request */
-#define OCTEP_CTRL_NET_H2F_GET_LINK_INFO_RESP_SZW	4
-/* size in words of set link info request */
-#define OCTEP_CTRL_NET_H2F_SET_LINK_INFO_RESP_SZW	1
+struct octep_ctrl_net_wait_data {
+	struct list_head list;
+	int done;
+	struct octep_ctrl_mbox_msg msg;
+	union {
+		struct octep_ctrl_net_h2f_req req;
+		struct octep_ctrl_net_h2f_resp resp;
+	} data;
+};
+
+/** Initialize data for ctrl net.
+ *
+ * @param oct: non-null pointer to struct octep_device.
+ *
+ * return value: 0 on success, -errno on error.
+ */
+int octep_ctrl_net_init(struct octep_device *oct);
 
 /** Get link status from firmware.
  *
@@ -235,21 +219,29 @@ union octep_ctrl_net_f2h_data_sz {
  *
  * return value: link status 0=down, 1=up.
  */
-int octep_get_link_status(struct octep_device *oct);
+int octep_ctrl_net_get_link_status(struct octep_device *oct);
 
 /** Set link status in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
  * @param up: boolean status.
+ * @param wait_for_response: poll for response.
+ *
+ * return value: 0 on success, -errno on failure
  */
-void octep_set_link_status(struct octep_device *oct, bool up);
+int octep_ctrl_net_set_link_status(struct octep_device *oct, bool up,
+				   bool wait_for_response);
 
 /** Set rx state in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
  * @param up: boolean status.
+ * @param wait_for_response: poll for response.
+ *
+ * return value: 0 on success, -errno on failure.
  */
-void octep_set_rx_state(struct octep_device *oct, bool up);
+int octep_ctrl_net_set_rx_state(struct octep_device *oct, bool up,
+				bool wait_for_response);
 
 /** Get mac address from firmware.
  *
@@ -258,21 +250,29 @@ void octep_set_rx_state(struct octep_device *oct, bool up);
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_get_mac_addr(struct octep_device *oct, u8 *addr);
+int octep_ctrl_net_get_mac_addr(struct octep_device *oct, u8 *addr);
 
 /** Set mac address in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
  * @param addr: non-null pointer to mac address.
+ * @param wait_for_response: poll for response.
+ *
+ * return value: 0 on success, -errno on failure.
  */
-int octep_set_mac_addr(struct octep_device *oct, u8 *addr);
+int octep_ctrl_net_set_mac_addr(struct octep_device *oct, u8 *addr,
+				bool wait_for_response);
 
 /** Set mtu in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
  * @param mtu: mtu.
+ * @param wait_for_response: poll for response.
+ *
+ * return value: 0 on success, -errno on failure.
  */
-int octep_set_mtu(struct octep_device *oct, int mtu);
+int octep_ctrl_net_set_mtu(struct octep_device *oct, int mtu,
+			   bool wait_for_response);
 
 /** Get interface statistics from firmware.
  *
@@ -280,7 +280,7 @@ int octep_set_mtu(struct octep_device *oct, int mtu);
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_get_if_stats(struct octep_device *oct);
+int octep_ctrl_net_get_if_stats(struct octep_device *oct);
 
 /** Get link info from firmware.
  *
@@ -288,12 +288,32 @@ int octep_get_if_stats(struct octep_device *oct);
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_get_link_info(struct octep_device *oct);
+int octep_ctrl_net_get_link_info(struct octep_device *oct);
 
 /** Set link info in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param link_info: non-null pointer to struct octep_iface_link_info.
+ * @param wait_for_response: poll for response.
+ *
+ * return value: 0 on success, -errno on failure.
+ */
+int octep_ctrl_net_set_link_info(struct octep_device *oct,
+				 struct octep_iface_link_info *link_info,
+				 bool wait_for_response);
+
+/** Poll for firmware messages and process them.
+ *
+ * @param oct: non-null pointer to struct octep_device.
+ */
+void octep_ctrl_net_recv_fw_messages(struct octep_device *oct);
+
+/** Uninitialize data for ctrl net.
+ *
+ * @param oct: non-null pointer to struct octep_device.
+ *
+ * return value: 0 on success, -errno on error.
  */
-int octep_set_link_info(struct octep_device *oct, struct octep_iface_link_info *link_info);
+int octep_ctrl_net_uninit(struct octep_device *oct);
 
 #endif /* __OCTEP_CTRL_NET_H__ */
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
index 87ef129b269a..a1ddb64791f2 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
@@ -150,7 +150,7 @@ octep_get_ethtool_stats(struct net_device *netdev,
 	rx_packets = 0;
 	rx_bytes = 0;
 
-	octep_get_if_stats(oct);
+	octep_ctrl_net_get_if_stats(oct);
 	iface_tx_stats = &oct->iface_tx_stats;
 	iface_rx_stats = &oct->iface_rx_stats;
 
@@ -283,7 +283,7 @@ static int octep_get_link_ksettings(struct net_device *netdev,
 	ethtool_link_ksettings_zero_link_mode(cmd, supported);
 	ethtool_link_ksettings_zero_link_mode(cmd, advertising);
 
-	octep_get_link_info(oct);
+	octep_ctrl_net_get_link_info(oct);
 
 	advertised_modes = oct->link_info.advertised_modes;
 	supported_modes = oct->link_info.supported_modes;
@@ -439,7 +439,7 @@ static int octep_set_link_ksettings(struct net_device *netdev,
 	link_info_new.speed = cmd->base.speed;
 	link_info_new.autoneg = autoneg;
 
-	err = octep_set_link_info(oct, &link_info_new);
+	err = octep_ctrl_net_set_link_info(oct, &link_info_new, true);
 	if (err)
 		return err;
 
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index a3e4d9355681..43fff8979854 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -507,11 +507,8 @@ static int octep_open(struct net_device *netdev)
 	octep_napi_enable(oct);
 
 	oct->link_info.admin_up = 1;
-	octep_set_rx_state(oct, true);
-
-	ret = octep_get_link_status(oct);
-	if (!ret)
-		octep_set_link_status(oct, true);
+	octep_ctrl_net_set_rx_state(oct, true, false);
+	octep_ctrl_net_set_link_status(oct, true, false);
 	oct->poll_non_ioq_intr = false;
 
 	/* Enable the input and output queues for this Octeon device */
@@ -522,7 +519,7 @@ static int octep_open(struct net_device *netdev)
 
 	octep_oq_dbell_init(oct);
 
-	ret = octep_get_link_status(oct);
+	ret = octep_ctrl_net_get_link_status(oct);
 	if (ret > 0)
 		octep_link_up(netdev);
 
@@ -552,14 +549,14 @@ static int octep_stop(struct net_device *netdev)
 
 	netdev_info(netdev, "Stopping the device ...\n");
 
+	octep_ctrl_net_set_link_status(oct, false, false);
+	octep_ctrl_net_set_rx_state(oct, false, false);
+
 	/* Stop Tx from stack */
 	netif_tx_stop_all_queues(netdev);
 	netif_carrier_off(netdev);
 	netif_tx_disable(netdev);
 
-	octep_set_link_status(oct, false);
-	octep_set_rx_state(oct, false);
-
 	oct->link_info.admin_up = 0;
 	oct->link_info.oper_up = 0;
 
@@ -761,7 +758,9 @@ static void octep_get_stats64(struct net_device *netdev,
 	struct octep_device *oct = netdev_priv(netdev);
 	int q;
 
-	octep_get_if_stats(oct);
+	if (netif_running(netdev))
+		octep_ctrl_net_get_if_stats(oct);
+
 	tx_packets = 0;
 	tx_bytes = 0;
 	rx_packets = 0;
@@ -832,7 +831,7 @@ static int octep_set_mac(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	err = octep_set_mac_addr(oct, addr->sa_data);
+	err = octep_ctrl_net_set_mac_addr(oct, addr->sa_data, true);
 	if (err)
 		return err;
 
@@ -852,7 +851,7 @@ static int octep_change_mtu(struct net_device *netdev, int new_mtu)
 	if (link_info->mtu == new_mtu)
 		return 0;
 
-	err = octep_set_mtu(oct, new_mtu);
+	err = octep_ctrl_net_set_mtu(oct, new_mtu, true);
 	if (!err) {
 		oct->link_info.mtu = new_mtu;
 		netdev->mtu = new_mtu;
@@ -904,34 +903,8 @@ static void octep_ctrl_mbox_task(struct work_struct *work)
 {
 	struct octep_device *oct = container_of(work, struct octep_device,
 						ctrl_mbox_task);
-	struct net_device *netdev = oct->netdev;
-	struct octep_ctrl_net_f2h_req req = {};
-	struct octep_ctrl_mbox_msg msg;
-	int ret = 0;
-
-	msg.msg = &req;
-	while (true) {
-		ret = octep_ctrl_mbox_recv(&oct->ctrl_mbox, &msg);
-		if (ret)
-			break;
-
-		switch (req.hdr.cmd) {
-		case OCTEP_CTRL_NET_F2H_CMD_LINK_STATUS:
-			if (netif_running(netdev)) {
-				if (req.link.state) {
-					dev_info(&oct->pdev->dev, "netif_carrier_on\n");
-					netif_carrier_on(netdev);
-				} else {
-					dev_info(&oct->pdev->dev, "netif_carrier_off\n");
-					netif_carrier_off(netdev);
-				}
-			}
-			break;
-		default:
-			pr_info("Unknown mbox req : %u\n", req.hdr.cmd);
-			break;
-		}
-	}
+
+	octep_ctrl_net_recv_fw_messages(oct);
 }
 
 static const char *octep_devid_to_str(struct octep_device *oct)
@@ -955,9 +928,8 @@ static const char *octep_devid_to_str(struct octep_device *oct)
  */
 int octep_device_setup(struct octep_device *oct)
 {
-	struct octep_ctrl_mbox *ctrl_mbox;
 	struct pci_dev *pdev = oct->pdev;
-	int i, ret;
+	int i;
 
 	/* allocate memory for oct->conf */
 	oct->conf = kzalloc(sizeof(*oct->conf), GFP_KERNEL);
@@ -992,20 +964,7 @@ int octep_device_setup(struct octep_device *oct)
 
 	oct->pkind = CFG_GET_IQ_PKIND(oct->conf);
 
-	/* Initialize control mbox */
-	ctrl_mbox = &oct->ctrl_mbox;
-	ctrl_mbox->barmem = CFG_GET_CTRL_MBOX_MEM_ADDR(oct->conf);
-	ret = octep_ctrl_mbox_init(ctrl_mbox);
-	if (ret) {
-		dev_err(&pdev->dev, "Failed to initialize control mbox\n");
-		goto unsupported_dev;
-	}
-	oct->ctrl_mbox_ifstats_offset = OCTEP_CTRL_MBOX_SZ(ctrl_mbox->h2fq.elem_sz,
-							   ctrl_mbox->h2fq.elem_cnt,
-							   ctrl_mbox->f2hq.elem_sz,
-							   ctrl_mbox->f2hq.elem_cnt);
-
-	return 0;
+	return octep_ctrl_net_init(oct);
 
 unsupported_dev:
 	for (i = 0; i < OCTEP_MMIO_REGIONS; i++)
@@ -1033,7 +992,7 @@ static void octep_device_cleanup(struct octep_device *oct)
 		oct->mbox[i] = NULL;
 	}
 
-	octep_ctrl_mbox_uninit(&oct->ctrl_mbox);
+	octep_ctrl_net_uninit(oct);
 
 	oct->hw_ops.soft_reset(oct);
 	for (i = 0; i < OCTEP_MMIO_REGIONS; i++) {
@@ -1143,7 +1102,8 @@ static int octep_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->max_mtu = OCTEP_MAX_MTU;
 	netdev->mtu = OCTEP_DEFAULT_MTU;
 
-	err = octep_get_mac_addr(octep_dev, octep_dev->mac_addr);
+	err = octep_ctrl_net_get_mac_addr(octep_dev,
+					  octep_dev->mac_addr);
 	if (err) {
 		dev_err(&pdev->dev, "Failed to get mac address\n");
 		goto register_dev_err;
-- 
cgit v1.2.3-70-g09d2


From f25e596755b4642225c33243753c763982605416 Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:47:00 -0700
Subject: octeon_ep: include function id in mailbox commands

Extend control command structure to include vfid and
update APIs to accept VF ID.

Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.c    | 43 ++++++++++++----------
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.h    | 28 ++++++++++----
 .../net/ethernet/marvell/octeon_ep/octep_ethtool.c |  7 ++--
 .../net/ethernet/marvell/octeon_ep/octep_main.c    | 24 +++++++-----
 4 files changed, 63 insertions(+), 39 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
index 8d4d74e18a67..cef4bc3b1ec0 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
@@ -22,7 +22,8 @@ static const u32 link_info_sz = sizeof(struct octep_ctrl_net_link_info);
 static const u32 get_stats_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_get_stats);
 static atomic_t ctrl_net_msg_id;
 
-static void init_send_req(struct octep_ctrl_mbox_msg *msg, void *buf, u16 sz)
+static void init_send_req(struct octep_ctrl_mbox_msg *msg, void *buf,
+			  u16 sz, int vfid)
 {
 	msg->hdr.s.flags = OCTEP_CTRL_MBOX_MSG_HDR_FLAG_REQ;
 	msg->hdr.s.msg_id = atomic_inc_return(&ctrl_net_msg_id) &
@@ -31,6 +32,10 @@ static void init_send_req(struct octep_ctrl_mbox_msg *msg, void *buf, u16 sz)
 	msg->sg_num = 1;
 	msg->sg_list[0].msg = buf;
 	msg->sg_list[0].sz = msg->hdr.s.sz;
+	if (vfid != OCTEP_CTRL_NET_INVALID_VFID) {
+		msg->hdr.s.is_vf = 1;
+		msg->hdr.s.vf_idx = vfid;
+	}
 }
 
 static int octep_send_mbox_req(struct octep_device *oct,
@@ -91,13 +96,13 @@ int octep_ctrl_net_init(struct octep_device *oct)
 	return 0;
 }
 
-int octep_ctrl_net_get_link_status(struct octep_device *oct)
+int octep_ctrl_net_get_link_status(struct octep_device *oct, int vfid)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 	int err;
 
-	init_send_req(&d.msg, (void *)req, state_sz);
+	init_send_req(&d.msg, (void *)req, state_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_STATUS;
 	req->link.cmd = OCTEP_CTRL_NET_CMD_GET;
 	err = octep_send_mbox_req(oct, &d, true);
@@ -107,13 +112,13 @@ int octep_ctrl_net_get_link_status(struct octep_device *oct)
 	return d.data.resp.link.state;
 }
 
-int octep_ctrl_net_set_link_status(struct octep_device *oct, bool up,
+int octep_ctrl_net_set_link_status(struct octep_device *oct, int vfid, bool up,
 				   bool wait_for_response)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	init_send_req(&d.msg, req, state_sz);
+	init_send_req(&d.msg, req, state_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_STATUS;
 	req->link.cmd = OCTEP_CTRL_NET_CMD_SET;
 	req->link.state = (up) ? OCTEP_CTRL_NET_STATE_UP :
@@ -122,13 +127,13 @@ int octep_ctrl_net_set_link_status(struct octep_device *oct, bool up,
 	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_ctrl_net_set_rx_state(struct octep_device *oct, bool up,
+int octep_ctrl_net_set_rx_state(struct octep_device *oct, int vfid, bool up,
 				bool wait_for_response)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	init_send_req(&d.msg, req, state_sz);
+	init_send_req(&d.msg, req, state_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_RX_STATE;
 	req->link.cmd = OCTEP_CTRL_NET_CMD_SET;
 	req->link.state = (up) ? OCTEP_CTRL_NET_STATE_UP :
@@ -137,13 +142,13 @@ int octep_ctrl_net_set_rx_state(struct octep_device *oct, bool up,
 	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_ctrl_net_get_mac_addr(struct octep_device *oct, u8 *addr)
+int octep_ctrl_net_get_mac_addr(struct octep_device *oct, int vfid, u8 *addr)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 	int err;
 
-	init_send_req(&d.msg, req, mac_sz);
+	init_send_req(&d.msg, req, mac_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_MAC;
 	req->link.cmd = OCTEP_CTRL_NET_CMD_GET;
 	err = octep_send_mbox_req(oct, &d, true);
@@ -155,13 +160,13 @@ int octep_ctrl_net_get_mac_addr(struct octep_device *oct, u8 *addr)
 	return 0;
 }
 
-int octep_ctrl_net_set_mac_addr(struct octep_device *oct, u8 *addr,
+int octep_ctrl_net_set_mac_addr(struct octep_device *oct, int vfid, u8 *addr,
 				bool wait_for_response)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	init_send_req(&d.msg, req, mac_sz);
+	init_send_req(&d.msg, req, mac_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_MAC;
 	req->mac.cmd = OCTEP_CTRL_NET_CMD_SET;
 	memcpy(&req->mac.addr, addr, ETH_ALEN);
@@ -169,13 +174,13 @@ int octep_ctrl_net_set_mac_addr(struct octep_device *oct, u8 *addr,
 	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_ctrl_net_set_mtu(struct octep_device *oct, int mtu,
+int octep_ctrl_net_set_mtu(struct octep_device *oct, int vfid, int mtu,
 			   bool wait_for_response)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	init_send_req(&d.msg, req, mtu_sz);
+	init_send_req(&d.msg, req, mtu_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_MTU;
 	req->mtu.cmd = OCTEP_CTRL_NET_CMD_SET;
 	req->mtu.val = mtu;
@@ -183,7 +188,7 @@ int octep_ctrl_net_set_mtu(struct octep_device *oct, int mtu,
 	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_ctrl_net_get_if_stats(struct octep_device *oct)
+int octep_ctrl_net_get_if_stats(struct octep_device *oct, int vfid)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
@@ -191,7 +196,7 @@ int octep_ctrl_net_get_if_stats(struct octep_device *oct)
 	void __iomem *iface_tx_stats;
 	int err;
 
-	init_send_req(&d.msg, req, get_stats_sz);
+	init_send_req(&d.msg, req, get_stats_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_GET_IF_STATS;
 	req->get_stats.offset = oct->ctrl_mbox_ifstats_offset;
 	err = octep_send_mbox_req(oct, &d, true);
@@ -207,14 +212,14 @@ int octep_ctrl_net_get_if_stats(struct octep_device *oct)
 	return 0;
 }
 
-int octep_ctrl_net_get_link_info(struct octep_device *oct)
+int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 	struct octep_ctrl_net_h2f_resp *resp;
 	int err;
 
-	init_send_req(&d.msg, req, link_info_sz);
+	init_send_req(&d.msg, req, link_info_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_INFO;
 	req->link_info.cmd = OCTEP_CTRL_NET_CMD_GET;
 	err = octep_send_mbox_req(oct, &d, true);
@@ -231,14 +236,14 @@ int octep_ctrl_net_get_link_info(struct octep_device *oct)
 	return 0;
 }
 
-int octep_ctrl_net_set_link_info(struct octep_device *oct,
+int octep_ctrl_net_set_link_info(struct octep_device *oct, int vfid,
 				 struct octep_iface_link_info *link_info,
 				 bool wait_for_response)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
 
-	init_send_req(&d.msg, req, link_info_sz);
+	init_send_req(&d.msg, req, link_info_sz, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_LINK_INFO;
 	req->link_info.cmd = OCTEP_CTRL_NET_CMD_SET;
 	req->link_info.info.advertised_modes = link_info->advertised_modes;
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
index aba373ec28a4..de0759365f13 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
@@ -7,6 +7,8 @@
 #ifndef __OCTEP_CTRL_NET_H__
 #define __OCTEP_CTRL_NET_H__
 
+#define OCTEP_CTRL_NET_INVALID_VFID	(-1)
+
 /* Supported commands */
 enum octep_ctrl_net_cmd {
 	OCTEP_CTRL_NET_CMD_GET = 0,
@@ -216,89 +218,99 @@ int octep_ctrl_net_init(struct octep_device *oct);
 /** Get link status from firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  *
  * return value: link status 0=down, 1=up.
  */
-int octep_ctrl_net_get_link_status(struct octep_device *oct);
+int octep_ctrl_net_get_link_status(struct octep_device *oct, int vfid);
 
 /** Set link status in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  * @param up: boolean status.
  * @param wait_for_response: poll for response.
  *
  * return value: 0 on success, -errno on failure
  */
-int octep_ctrl_net_set_link_status(struct octep_device *oct, bool up,
+int octep_ctrl_net_set_link_status(struct octep_device *oct, int vfid, bool up,
 				   bool wait_for_response);
 
 /** Set rx state in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  * @param up: boolean status.
  * @param wait_for_response: poll for response.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_set_rx_state(struct octep_device *oct, bool up,
+int octep_ctrl_net_set_rx_state(struct octep_device *oct, int vfid, bool up,
 				bool wait_for_response);
 
 /** Get mac address from firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  * @param addr: non-null pointer to mac address.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_get_mac_addr(struct octep_device *oct, u8 *addr);
+int octep_ctrl_net_get_mac_addr(struct octep_device *oct, int vfid, u8 *addr);
 
 /** Set mac address in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  * @param addr: non-null pointer to mac address.
  * @param wait_for_response: poll for response.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_set_mac_addr(struct octep_device *oct, u8 *addr,
+int octep_ctrl_net_set_mac_addr(struct octep_device *oct, int vfid, u8 *addr,
 				bool wait_for_response);
 
 /** Set mtu in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  * @param mtu: mtu.
  * @param wait_for_response: poll for response.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_set_mtu(struct octep_device *oct, int mtu,
+int octep_ctrl_net_set_mtu(struct octep_device *oct, int vfid, int mtu,
 			   bool wait_for_response);
 
 /** Get interface statistics from firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_get_if_stats(struct octep_device *oct);
+int octep_ctrl_net_get_if_stats(struct octep_device *oct, int vfid);
 
 /** Get link info from firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_get_link_info(struct octep_device *oct);
+int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid);
 
 /** Set link info in firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
+ * @param vfid: Index of virtual function.
  * @param link_info: non-null pointer to struct octep_iface_link_info.
  * @param wait_for_response: poll for response.
  *
  * return value: 0 on success, -errno on failure.
  */
 int octep_ctrl_net_set_link_info(struct octep_device *oct,
+				 int vfid,
 				 struct octep_iface_link_info *link_info,
 				 bool wait_for_response);
 
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
index a1ddb64791f2..389042b57787 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
@@ -150,7 +150,7 @@ octep_get_ethtool_stats(struct net_device *netdev,
 	rx_packets = 0;
 	rx_bytes = 0;
 
-	octep_ctrl_net_get_if_stats(oct);
+	octep_ctrl_net_get_if_stats(oct, OCTEP_CTRL_NET_INVALID_VFID);
 	iface_tx_stats = &oct->iface_tx_stats;
 	iface_rx_stats = &oct->iface_rx_stats;
 
@@ -283,7 +283,7 @@ static int octep_get_link_ksettings(struct net_device *netdev,
 	ethtool_link_ksettings_zero_link_mode(cmd, supported);
 	ethtool_link_ksettings_zero_link_mode(cmd, advertising);
 
-	octep_ctrl_net_get_link_info(oct);
+	octep_ctrl_net_get_link_info(oct, OCTEP_CTRL_NET_INVALID_VFID);
 
 	advertised_modes = oct->link_info.advertised_modes;
 	supported_modes = oct->link_info.supported_modes;
@@ -439,7 +439,8 @@ static int octep_set_link_ksettings(struct net_device *netdev,
 	link_info_new.speed = cmd->base.speed;
 	link_info_new.autoneg = autoneg;
 
-	err = octep_ctrl_net_set_link_info(oct, &link_info_new, true);
+	err = octep_ctrl_net_set_link_info(oct, OCTEP_CTRL_NET_INVALID_VFID,
+					   &link_info_new, true);
 	if (err)
 		return err;
 
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index 43fff8979854..112d3825e544 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -507,8 +507,10 @@ static int octep_open(struct net_device *netdev)
 	octep_napi_enable(oct);
 
 	oct->link_info.admin_up = 1;
-	octep_ctrl_net_set_rx_state(oct, true, false);
-	octep_ctrl_net_set_link_status(oct, true, false);
+	octep_ctrl_net_set_rx_state(oct, OCTEP_CTRL_NET_INVALID_VFID, true,
+				    false);
+	octep_ctrl_net_set_link_status(oct, OCTEP_CTRL_NET_INVALID_VFID, true,
+				       false);
 	oct->poll_non_ioq_intr = false;
 
 	/* Enable the input and output queues for this Octeon device */
@@ -519,7 +521,7 @@ static int octep_open(struct net_device *netdev)
 
 	octep_oq_dbell_init(oct);
 
-	ret = octep_ctrl_net_get_link_status(oct);
+	ret = octep_ctrl_net_get_link_status(oct, OCTEP_CTRL_NET_INVALID_VFID);
 	if (ret > 0)
 		octep_link_up(netdev);
 
@@ -549,8 +551,10 @@ static int octep_stop(struct net_device *netdev)
 
 	netdev_info(netdev, "Stopping the device ...\n");
 
-	octep_ctrl_net_set_link_status(oct, false, false);
-	octep_ctrl_net_set_rx_state(oct, false, false);
+	octep_ctrl_net_set_link_status(oct, OCTEP_CTRL_NET_INVALID_VFID, false,
+				       false);
+	octep_ctrl_net_set_rx_state(oct, OCTEP_CTRL_NET_INVALID_VFID, false,
+				    false);
 
 	/* Stop Tx from stack */
 	netif_tx_stop_all_queues(netdev);
@@ -759,7 +763,7 @@ static void octep_get_stats64(struct net_device *netdev,
 	int q;
 
 	if (netif_running(netdev))
-		octep_ctrl_net_get_if_stats(oct);
+		octep_ctrl_net_get_if_stats(oct, OCTEP_CTRL_NET_INVALID_VFID);
 
 	tx_packets = 0;
 	tx_bytes = 0;
@@ -831,7 +835,8 @@ static int octep_set_mac(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	err = octep_ctrl_net_set_mac_addr(oct, addr->sa_data, true);
+	err = octep_ctrl_net_set_mac_addr(oct, OCTEP_CTRL_NET_INVALID_VFID,
+					  addr->sa_data, true);
 	if (err)
 		return err;
 
@@ -851,7 +856,8 @@ static int octep_change_mtu(struct net_device *netdev, int new_mtu)
 	if (link_info->mtu == new_mtu)
 		return 0;
 
-	err = octep_ctrl_net_set_mtu(oct, new_mtu, true);
+	err = octep_ctrl_net_set_mtu(oct, OCTEP_CTRL_NET_INVALID_VFID, new_mtu,
+				     true);
 	if (!err) {
 		oct->link_info.mtu = new_mtu;
 		netdev->mtu = new_mtu;
@@ -1102,7 +1108,7 @@ static int octep_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->max_mtu = OCTEP_MAX_MTU;
 	netdev->mtu = OCTEP_DEFAULT_MTU;
 
-	err = octep_ctrl_net_get_mac_addr(octep_dev,
+	err = octep_ctrl_net_get_mac_addr(octep_dev, OCTEP_CTRL_NET_INVALID_VFID,
 					  octep_dev->mac_addr);
 	if (err) {
 		dev_err(&pdev->dev, "Failed to get mac address\n");
-- 
cgit v1.2.3-70-g09d2


From baa987988777545c6bf93efdf62a4b2c74f8d739 Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:47:01 -0700
Subject: octeon_ep: support asynchronous notifications

Add asynchronous notification support to the control mailbox.

Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.c    | 29 ++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
index cef4bc3b1ec0..465eef2824e3 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
@@ -271,6 +271,33 @@ static void process_mbox_resp(struct octep_device *oct,
 	}
 }
 
+static int process_mbox_notify(struct octep_device *oct,
+			       struct octep_ctrl_mbox_msg *msg)
+{
+	struct net_device *netdev = oct->netdev;
+	struct octep_ctrl_net_f2h_req *req;
+
+	req = (struct octep_ctrl_net_f2h_req *)msg->sg_list[0].msg;
+	switch (req->hdr.s.cmd) {
+	case OCTEP_CTRL_NET_F2H_CMD_LINK_STATUS:
+		if (netif_running(netdev)) {
+			if (req->link.state) {
+				dev_info(&oct->pdev->dev, "netif_carrier_on\n");
+				netif_carrier_on(netdev);
+			} else {
+				dev_info(&oct->pdev->dev, "netif_carrier_off\n");
+				netif_carrier_off(netdev);
+			}
+		}
+		break;
+	default:
+		pr_info("Unknown mbox req : %u\n", req->hdr.s.cmd);
+		break;
+	}
+
+	return 0;
+}
+
 void octep_ctrl_net_recv_fw_messages(struct octep_device *oct)
 {
 	static u16 msg_sz = sizeof(union octep_ctrl_net_max_data);
@@ -291,6 +318,8 @@ void octep_ctrl_net_recv_fw_messages(struct octep_device *oct)
 
 		if (msg.hdr.s.flags & OCTEP_CTRL_MBOX_MSG_HDR_FLAG_RESP)
 			process_mbox_resp(oct, &msg);
+		else if (msg.hdr.s.flags & OCTEP_CTRL_MBOX_MSG_HDR_FLAG_NOTIFY)
+			process_mbox_notify(oct, &msg);
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From 0718693fb36cf891c88c64564553452406151d37 Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:47:02 -0700
Subject: octeon_ep: function id in link info and stats mailbox commands

Update control mailbox API to include function id in get stats and
link info.

Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.c    | 33 ++++++++++------------
 .../ethernet/marvell/octeon_ep/octep_ctrl_net.h    | 24 ++++++++++------
 .../net/ethernet/marvell/octeon_ep/octep_ethtool.c |  9 ++++--
 .../net/ethernet/marvell/octeon_ep/octep_main.c    |  5 +++-
 4 files changed, 40 insertions(+), 31 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
index 465eef2824e3..1cc6af2feb38 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.c
@@ -19,7 +19,6 @@ static const u32 mtu_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_mtu);
 static const u32 mac_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_mac);
 static const u32 state_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_state);
 static const u32 link_info_sz = sizeof(struct octep_ctrl_net_link_info);
-static const u32 get_stats_sz = sizeof(struct octep_ctrl_net_h2f_req_cmd_get_stats);
 static atomic_t ctrl_net_msg_id;
 
 static void init_send_req(struct octep_ctrl_mbox_msg *msg, void *buf,
@@ -188,31 +187,29 @@ int octep_ctrl_net_set_mtu(struct octep_device *oct, int vfid, int mtu,
 	return octep_send_mbox_req(oct, &d, wait_for_response);
 }
 
-int octep_ctrl_net_get_if_stats(struct octep_device *oct, int vfid)
+int octep_ctrl_net_get_if_stats(struct octep_device *oct, int vfid,
+				struct octep_iface_rx_stats *rx_stats,
+				struct octep_iface_tx_stats *tx_stats)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
-	void __iomem *iface_rx_stats;
-	void __iomem *iface_tx_stats;
+	struct octep_ctrl_net_h2f_resp *resp;
 	int err;
 
-	init_send_req(&d.msg, req, get_stats_sz, vfid);
+	init_send_req(&d.msg, req, 0, vfid);
 	req->hdr.s.cmd = OCTEP_CTRL_NET_H2F_CMD_GET_IF_STATS;
-	req->get_stats.offset = oct->ctrl_mbox_ifstats_offset;
 	err = octep_send_mbox_req(oct, &d, true);
 	if (err < 0)
 		return err;
 
-	iface_rx_stats = oct->ctrl_mbox.barmem + oct->ctrl_mbox_ifstats_offset;
-	iface_tx_stats = oct->ctrl_mbox.barmem + oct->ctrl_mbox_ifstats_offset +
-			 sizeof(struct octep_iface_rx_stats);
-	memcpy_fromio(&oct->iface_rx_stats, iface_rx_stats, sizeof(struct octep_iface_rx_stats));
-	memcpy_fromio(&oct->iface_tx_stats, iface_tx_stats, sizeof(struct octep_iface_tx_stats));
-
+	resp = &d.data.resp;
+	memcpy(rx_stats, &resp->if_stats.rx_stats, sizeof(struct octep_iface_rx_stats));
+	memcpy(tx_stats, &resp->if_stats.tx_stats, sizeof(struct octep_iface_tx_stats));
 	return 0;
 }
 
-int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid)
+int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid,
+				 struct octep_iface_link_info *link_info)
 {
 	struct octep_ctrl_net_wait_data d = {0};
 	struct octep_ctrl_net_h2f_req *req = &d.data.req;
@@ -227,11 +224,11 @@ int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid)
 		return err;
 
 	resp = &d.data.resp;
-	oct->link_info.supported_modes = resp->link_info.supported_modes;
-	oct->link_info.advertised_modes = resp->link_info.advertised_modes;
-	oct->link_info.autoneg = resp->link_info.autoneg;
-	oct->link_info.pause = resp->link_info.pause;
-	oct->link_info.speed = resp->link_info.speed;
+	link_info->supported_modes = resp->link_info.supported_modes;
+	link_info->advertised_modes = resp->link_info.advertised_modes;
+	link_info->autoneg = resp->link_info.autoneg;
+	link_info->pause = resp->link_info.pause;
+	link_info->speed = resp->link_info.speed;
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
index de0759365f13..37880dd79116 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_net.h
@@ -77,12 +77,6 @@ struct octep_ctrl_net_h2f_req_cmd_mac {
 	u8 addr[ETH_ALEN];
 };
 
-/* get if_stats, xstats, q_stats request */
-struct octep_ctrl_net_h2f_req_cmd_get_stats {
-	/* offset into barmem where fw should copy over stats */
-	u32 offset;
-};
-
 /* get/set link state, rx state */
 struct octep_ctrl_net_h2f_req_cmd_state {
 	/* enum octep_ctrl_net_cmd */
@@ -119,7 +113,6 @@ struct octep_ctrl_net_h2f_req {
 	union {
 		struct octep_ctrl_net_h2f_req_cmd_mtu mtu;
 		struct octep_ctrl_net_h2f_req_cmd_mac mac;
-		struct octep_ctrl_net_h2f_req_cmd_get_stats get_stats;
 		struct octep_ctrl_net_h2f_req_cmd_state link;
 		struct octep_ctrl_net_h2f_req_cmd_state rx;
 		struct octep_ctrl_net_h2f_req_cmd_link_info link_info;
@@ -152,6 +145,12 @@ struct octep_ctrl_net_h2f_resp_cmd_mac {
 	u8 addr[ETH_ALEN];
 };
 
+/* get if_stats, xstats, q_stats request */
+struct octep_ctrl_net_h2f_resp_cmd_get_stats {
+	struct octep_iface_rx_stats rx_stats;
+	struct octep_iface_tx_stats tx_stats;
+};
+
 /* get link state, rx state response */
 struct octep_ctrl_net_h2f_resp_cmd_state {
 	/* enum octep_ctrl_net_state */
@@ -164,6 +163,7 @@ struct octep_ctrl_net_h2f_resp {
 	union {
 		struct octep_ctrl_net_h2f_resp_cmd_mtu mtu;
 		struct octep_ctrl_net_h2f_resp_cmd_mac mac;
+		struct octep_ctrl_net_h2f_resp_cmd_get_stats if_stats;
 		struct octep_ctrl_net_h2f_resp_cmd_state link;
 		struct octep_ctrl_net_h2f_resp_cmd_state rx;
 		struct octep_ctrl_net_link_info link_info;
@@ -286,19 +286,25 @@ int octep_ctrl_net_set_mtu(struct octep_device *oct, int vfid, int mtu,
  *
  * @param oct: non-null pointer to struct octep_device.
  * @param vfid: Index of virtual function.
+ * @param rx_stats: non-null pointer struct octep_iface_rx_stats.
+ * @param tx_stats: non-null pointer struct octep_iface_tx_stats.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_get_if_stats(struct octep_device *oct, int vfid);
+int octep_ctrl_net_get_if_stats(struct octep_device *oct, int vfid,
+				struct octep_iface_rx_stats *rx_stats,
+				struct octep_iface_tx_stats *tx_stats);
 
 /** Get link info from firmware.
  *
  * @param oct: non-null pointer to struct octep_device.
  * @param vfid: Index of virtual function.
+ * @param link_info: non-null pointer to struct octep_iface_link_info.
  *
  * return value: 0 on success, -errno on failure.
  */
-int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid);
+int octep_ctrl_net_get_link_info(struct octep_device *oct, int vfid,
+				 struct octep_iface_link_info *link_info);
 
 /** Set link info in firmware.
  *
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
index 389042b57787..7d0124b283da 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ethtool.c
@@ -150,9 +150,12 @@ octep_get_ethtool_stats(struct net_device *netdev,
 	rx_packets = 0;
 	rx_bytes = 0;
 
-	octep_ctrl_net_get_if_stats(oct, OCTEP_CTRL_NET_INVALID_VFID);
 	iface_tx_stats = &oct->iface_tx_stats;
 	iface_rx_stats = &oct->iface_rx_stats;
+	octep_ctrl_net_get_if_stats(oct,
+				    OCTEP_CTRL_NET_INVALID_VFID,
+				    iface_rx_stats,
+				    iface_tx_stats);
 
 	for (q = 0; q < oct->num_oqs; q++) {
 		struct octep_iq *iq = oct->iq[q];
@@ -283,11 +286,11 @@ static int octep_get_link_ksettings(struct net_device *netdev,
 	ethtool_link_ksettings_zero_link_mode(cmd, supported);
 	ethtool_link_ksettings_zero_link_mode(cmd, advertising);
 
-	octep_ctrl_net_get_link_info(oct, OCTEP_CTRL_NET_INVALID_VFID);
+	link_info = &oct->link_info;
+	octep_ctrl_net_get_link_info(oct, OCTEP_CTRL_NET_INVALID_VFID, link_info);
 
 	advertised_modes = oct->link_info.advertised_modes;
 	supported_modes = oct->link_info.supported_modes;
-	link_info = &oct->link_info;
 
 	OCTEP_SET_ETHTOOL_LINK_MODES_BITMAP(supported_modes, cmd, supported);
 	OCTEP_SET_ETHTOOL_LINK_MODES_BITMAP(advertised_modes, cmd, advertising);
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index 112d3825e544..ba0d5fe3081d 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -763,7 +763,10 @@ static void octep_get_stats64(struct net_device *netdev,
 	int q;
 
 	if (netif_running(netdev))
-		octep_ctrl_net_get_if_stats(oct, OCTEP_CTRL_NET_INVALID_VFID);
+		octep_ctrl_net_get_if_stats(oct,
+					    OCTEP_CTRL_NET_INVALID_VFID,
+					    &oct->iface_rx_stats,
+					    &oct->iface_tx_stats);
 
 	tx_packets = 0;
 	tx_bytes = 0;
-- 
cgit v1.2.3-70-g09d2


From 5cb96c29aa0ea359e4f5d30585538dc6a847d69d Mon Sep 17 00:00:00 2001
From: Veerasenareddy Burru <vburru@marvell.com>
Date: Fri, 24 Mar 2023 10:47:03 -0700
Subject: octeon_ep: add heartbeat monitor

Monitor periodic heartbeat messages from device firmware.
Presence of heartbeat indicates the device is active and running.
If the heartbeat is missed for configured interval indicates
firmware has crashed and device is unusable; in this case, PF driver
stops and uninitialize the device.

Signed-off-by: Veerasenareddy Burru <vburru@marvell.com>
Signed-off-by: Abhijit Ayarekar <aayarekar@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c |  9 +++++
 .../net/ethernet/marvell/octeon_ep/octep_config.h  |  6 +++
 .../net/ethernet/marvell/octeon_ep/octep_main.c    | 44 +++++++++++++++++++++-
 .../net/ethernet/marvell/octeon_ep/octep_main.h    |  7 ++++
 .../marvell/octeon_ep/octep_regs_cn9k_pf.h         |  2 +
 5 files changed, 66 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c b/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
index e2503c9bc8a1..90c3a419932d 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_cn9k_pf.c
@@ -16,6 +16,9 @@
 #define CTRL_MBOX_MAX_PF	128
 #define CTRL_MBOX_SZ		((size_t)(0x400000 / CTRL_MBOX_MAX_PF))
 
+#define FW_HB_INTERVAL_IN_SECS		1
+#define FW_HB_MISS_COUNT		10
+
 /* Names of Hardware non-queue generic interrupts */
 static char *cn93_non_ioq_msix_names[] = {
 	"epf_ire_rint",
@@ -249,6 +252,10 @@ static void octep_init_config_cn93_pf(struct octep_device *oct)
 	conf->ctrl_mbox_cfg.barmem_addr = (void __iomem *)oct->mmio[2].hw_addr +
 					   (0x400000ull * 7) +
 					   (link * CTRL_MBOX_SZ);
+
+	conf->hb_interval = FW_HB_INTERVAL_IN_SECS;
+	conf->max_hb_miss_cnt = FW_HB_MISS_COUNT;
+
 }
 
 /* Setup registers for a hardware Tx Queue  */
@@ -383,6 +390,8 @@ static bool octep_poll_non_ioq_interrupts_cn93_pf(struct octep_device *oct)
 		octep_write_csr64(oct, CN93_SDP_EPF_OEI_RINT, reg0);
 		if (reg0 & CN93_SDP_EPF_OEI_RINT_DATA_BIT_MBOX)
 			queue_work(octep_wq, &oct->ctrl_mbox_task);
+		else if (reg0 & CN93_SDP_EPF_OEI_RINT_DATA_BIT_HBEAT)
+			atomic_set(&oct->hb_miss_cnt, 0);
 
 		handled = true;
 	}
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_config.h b/drivers/net/ethernet/marvell/octeon_ep/octep_config.h
index f208f3f9a447..df7cd39d9fce 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_config.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_config.h
@@ -200,5 +200,11 @@ struct octep_config {
 
 	/* ctrl mbox config */
 	struct octep_ctrl_mbox_config ctrl_mbox_cfg;
+
+	/* Configured maximum heartbeat miss count */
+	u32 max_hb_miss_cnt;
+
+	/* Configured firmware heartbeat interval in secs */
+	u32 hb_interval;
 };
 #endif /* _OCTEP_CONFIG_H_ */
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index ba0d5fe3081d..e1853da280f9 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -901,6 +901,37 @@ static void octep_intr_poll_task(struct work_struct *work)
 			   msecs_to_jiffies(OCTEP_INTR_POLL_TIME_MSECS));
 }
 
+/**
+ * octep_hb_timeout_task - work queue task to check firmware heartbeat.
+ *
+ * @work: pointer to hb work_struct
+ *
+ * Check for heartbeat miss count. Uninitialize oct device if miss count
+ * exceeds configured max heartbeat miss count.
+ *
+ **/
+static void octep_hb_timeout_task(struct work_struct *work)
+{
+	struct octep_device *oct = container_of(work, struct octep_device,
+						hb_task.work);
+
+	int miss_cnt;
+
+	miss_cnt = atomic_inc_return(&oct->hb_miss_cnt);
+	if (miss_cnt < oct->conf->max_hb_miss_cnt) {
+		queue_delayed_work(octep_wq, &oct->hb_task,
+				   msecs_to_jiffies(oct->conf->hb_interval * 1000));
+		return;
+	}
+
+	dev_err(&oct->pdev->dev, "Missed %u heartbeats. Uninitializing\n",
+		miss_cnt);
+	rtnl_lock();
+	if (netif_running(oct->netdev))
+		octep_stop(oct->netdev);
+	rtnl_unlock();
+}
+
 /**
  * octep_ctrl_mbox_task - work queue task to handle ctrl mbox messages.
  *
@@ -938,7 +969,7 @@ static const char *octep_devid_to_str(struct octep_device *oct)
 int octep_device_setup(struct octep_device *oct)
 {
 	struct pci_dev *pdev = oct->pdev;
-	int i;
+	int i, ret;
 
 	/* allocate memory for oct->conf */
 	oct->conf = kzalloc(sizeof(*oct->conf), GFP_KERNEL);
@@ -973,7 +1004,15 @@ int octep_device_setup(struct octep_device *oct)
 
 	oct->pkind = CFG_GET_IQ_PKIND(oct->conf);
 
-	return octep_ctrl_net_init(oct);
+	ret = octep_ctrl_net_init(oct);
+	if (ret)
+		return ret;
+
+	atomic_set(&oct->hb_miss_cnt, 0);
+	INIT_DELAYED_WORK(&oct->hb_task, octep_hb_timeout_task);
+	queue_delayed_work(octep_wq, &oct->hb_task,
+			   msecs_to_jiffies(oct->conf->hb_interval * 1000));
+	return 0;
 
 unsupported_dev:
 	for (i = 0; i < OCTEP_MMIO_REGIONS; i++)
@@ -1002,6 +1041,7 @@ static void octep_device_cleanup(struct octep_device *oct)
 	}
 
 	octep_ctrl_net_uninit(oct);
+	cancel_delayed_work_sync(&oct->hb_task);
 
 	oct->hw_ops.soft_reset(oct);
 	for (i = 0; i < OCTEP_MMIO_REGIONS; i++) {
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.h b/drivers/net/ethernet/marvell/octeon_ep/octep_main.h
index 836d990ba3fa..e0907a719133 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.h
@@ -280,6 +280,13 @@ struct octep_device {
 	bool poll_non_ioq_intr;
 	/* Work entry to poll non-ioq interrupts */
 	struct delayed_work intr_poll_task;
+
+	/* Firmware heartbeat timer */
+	struct timer_list hb_timer;
+	/* Firmware heartbeat miss count tracked by timer */
+	atomic_t hb_miss_cnt;
+	/* Task to reset device on heartbeat miss */
+	struct delayed_work hb_task;
 };
 
 static inline u16 OCTEP_MAJOR_REV(struct octep_device *oct)
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h b/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h
index 0466fd9a002d..b25c3093dc7b 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_regs_cn9k_pf.h
@@ -367,5 +367,7 @@
 
 /* bit 0 for control mbox interrupt */
 #define CN93_SDP_EPF_OEI_RINT_DATA_BIT_MBOX	BIT_ULL(0)
+/* bit 1 for firmware heartbeat interrupt */
+#define CN93_SDP_EPF_OEI_RINT_DATA_BIT_HBEAT	BIT_ULL(1)
 
 #endif /* _OCTEP_REGS_CN9K_PF_H_ */
-- 
cgit v1.2.3-70-g09d2


From d61157414d0a591d10d27d0ce5873916614e5e31 Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:27 -0400
Subject: net: sunhme: Fix uninitialized return code

Fix an uninitialized return code if we never found a qfe slot. It would be
a bug if we ever got into this situation, but it's good to return something
tracable.

Fixes: acb3f35f920b ("sunhme: forward the error code from pci_enable_device()")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index b0c7ab74a82e..7cf8210ebbec 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2834,7 +2834,7 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 	int i, qfe_slot = -1;
 	char prom_name[64];
 	u8 addr[ETH_ALEN];
-	int err;
+	int err = -ENODEV;
 
 	/* Now make sure pci_dev cookie is there. */
 #ifdef CONFIG_SPARC
-- 
cgit v1.2.3-70-g09d2


From 70b1b4b86227c024c29ad54278fa5b80ec535ca9 Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:28 -0400
Subject: net: sunhme: Just restart autonegotiation if we can't bring the link
 up

If we've tried regular autonegotiation and forcing the link mode, just
restart autonegotiation instead of reinitializing the whole NIC.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 245 ++++++++++++++++++--------------------
 1 file changed, 119 insertions(+), 126 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index 7cf8210ebbec..e52f2d9305bc 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -589,8 +589,6 @@ no_response:
 	return 1;
 }
 
-static int happy_meal_init(struct happy_meal *hp);
-
 static int is_lucent_phy(struct happy_meal *hp)
 {
 	void __iomem *tregs = hp->tcvregs;
@@ -606,6 +604,124 @@ static int is_lucent_phy(struct happy_meal *hp)
 	return ret;
 }
 
+/* hp->happy_lock must be held */
+static void
+happy_meal_begin_auto_negotiation(struct happy_meal *hp,
+				  void __iomem *tregs,
+				  const struct ethtool_link_ksettings *ep)
+{
+	int timeout;
+
+	/* Read all of the registers we are interested in now. */
+	hp->sw_bmsr      = happy_meal_tcvr_read(hp, tregs, MII_BMSR);
+	hp->sw_bmcr      = happy_meal_tcvr_read(hp, tregs, MII_BMCR);
+	hp->sw_physid1   = happy_meal_tcvr_read(hp, tregs, MII_PHYSID1);
+	hp->sw_physid2   = happy_meal_tcvr_read(hp, tregs, MII_PHYSID2);
+
+	/* XXX Check BMSR_ANEGCAPABLE, should not be necessary though. */
+
+	hp->sw_advertise = happy_meal_tcvr_read(hp, tregs, MII_ADVERTISE);
+	if (!ep || ep->base.autoneg == AUTONEG_ENABLE) {
+		/* Advertise everything we can support. */
+		if (hp->sw_bmsr & BMSR_10HALF)
+			hp->sw_advertise |= (ADVERTISE_10HALF);
+		else
+			hp->sw_advertise &= ~(ADVERTISE_10HALF);
+
+		if (hp->sw_bmsr & BMSR_10FULL)
+			hp->sw_advertise |= (ADVERTISE_10FULL);
+		else
+			hp->sw_advertise &= ~(ADVERTISE_10FULL);
+		if (hp->sw_bmsr & BMSR_100HALF)
+			hp->sw_advertise |= (ADVERTISE_100HALF);
+		else
+			hp->sw_advertise &= ~(ADVERTISE_100HALF);
+		if (hp->sw_bmsr & BMSR_100FULL)
+			hp->sw_advertise |= (ADVERTISE_100FULL);
+		else
+			hp->sw_advertise &= ~(ADVERTISE_100FULL);
+		happy_meal_tcvr_write(hp, tregs, MII_ADVERTISE, hp->sw_advertise);
+
+		/* XXX Currently no Happy Meal cards I know off support 100BaseT4,
+		 * XXX and this is because the DP83840 does not support it, changes
+		 * XXX would need to be made to the tx/rx logic in the driver as well
+		 * XXX so I completely skip checking for it in the BMSR for now.
+		 */
+
+		ASD("Advertising [ %s%s%s%s]\n",
+		    hp->sw_advertise & ADVERTISE_10HALF ? "10H " : "",
+		    hp->sw_advertise & ADVERTISE_10FULL ? "10F " : "",
+		    hp->sw_advertise & ADVERTISE_100HALF ? "100H " : "",
+		    hp->sw_advertise & ADVERTISE_100FULL ? "100F " : "");
+
+		/* Enable Auto-Negotiation, this is usually on already... */
+		hp->sw_bmcr |= BMCR_ANENABLE;
+		happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
+
+		/* Restart it to make sure it is going. */
+		hp->sw_bmcr |= BMCR_ANRESTART;
+		happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
+
+		/* BMCR_ANRESTART self clears when the process has begun. */
+
+		timeout = 64;  /* More than enough. */
+		while (--timeout) {
+			hp->sw_bmcr = happy_meal_tcvr_read(hp, tregs, MII_BMCR);
+			if (!(hp->sw_bmcr & BMCR_ANRESTART))
+				break; /* got it. */
+			udelay(10);
+		}
+		if (!timeout) {
+			netdev_err(hp->dev,
+				   "Happy Meal would not start auto negotiation BMCR=0x%04x\n",
+				   hp->sw_bmcr);
+			netdev_notice(hp->dev,
+				      "Performing force link detection.\n");
+			goto force_link;
+		} else {
+			hp->timer_state = arbwait;
+		}
+	} else {
+force_link:
+		/* Force the link up, trying first a particular mode.
+		 * Either we are here at the request of ethtool or
+		 * because the Happy Meal would not start to autoneg.
+		 */
+
+		/* Disable auto-negotiation in BMCR, enable the duplex and
+		 * speed setting, init the timer state machine, and fire it off.
+		 */
+		if (!ep || ep->base.autoneg == AUTONEG_ENABLE) {
+			hp->sw_bmcr = BMCR_SPEED100;
+		} else {
+			if (ep->base.speed == SPEED_100)
+				hp->sw_bmcr = BMCR_SPEED100;
+			else
+				hp->sw_bmcr = 0;
+			if (ep->base.duplex == DUPLEX_FULL)
+				hp->sw_bmcr |= BMCR_FULLDPLX;
+		}
+		happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
+
+		if (!is_lucent_phy(hp)) {
+			/* OK, seems we need do disable the transceiver for the first
+			 * tick to make sure we get an accurate link state at the
+			 * second tick.
+			 */
+			hp->sw_csconfig = happy_meal_tcvr_read(hp, tregs,
+							       DP83840_CSCONFIG);
+			hp->sw_csconfig &= ~(CSCONFIG_TCVDISAB);
+			happy_meal_tcvr_write(hp, tregs, DP83840_CSCONFIG,
+					      hp->sw_csconfig);
+		}
+		hp->timer_state = ltrywait;
+	}
+
+	hp->timer_ticks = 0;
+	hp->happy_timer.expires = jiffies + (12 * HZ)/10;  /* 1.2 sec. */
+	add_timer(&hp->happy_timer);
+}
+
 static void happy_meal_timer(struct timer_list *t)
 {
 	struct happy_meal *hp = from_timer(hp, t, happy_timer);
@@ -743,12 +859,7 @@ static void happy_meal_timer(struct timer_list *t)
 					netdev_notice(hp->dev,
 						      "Link down, cable problem?\n");
 
-					ret = happy_meal_init(hp);
-					if (ret) {
-						/* ho hum... */
-						netdev_err(hp->dev,
-							   "Error, cannot re-init the Happy Meal.\n");
-					}
+					happy_meal_begin_auto_negotiation(hp, tregs, NULL);
 					goto out;
 				}
 				if (!is_lucent_phy(hp)) {
@@ -1201,124 +1312,6 @@ static void happy_meal_init_rings(struct happy_meal *hp)
 	HMD("done\n");
 }
 
-/* hp->happy_lock must be held */
-static void
-happy_meal_begin_auto_negotiation(struct happy_meal *hp,
-				  void __iomem *tregs,
-				  const struct ethtool_link_ksettings *ep)
-{
-	int timeout;
-
-	/* Read all of the registers we are interested in now. */
-	hp->sw_bmsr      = happy_meal_tcvr_read(hp, tregs, MII_BMSR);
-	hp->sw_bmcr      = happy_meal_tcvr_read(hp, tregs, MII_BMCR);
-	hp->sw_physid1   = happy_meal_tcvr_read(hp, tregs, MII_PHYSID1);
-	hp->sw_physid2   = happy_meal_tcvr_read(hp, tregs, MII_PHYSID2);
-
-	/* XXX Check BMSR_ANEGCAPABLE, should not be necessary though. */
-
-	hp->sw_advertise = happy_meal_tcvr_read(hp, tregs, MII_ADVERTISE);
-	if (!ep || ep->base.autoneg == AUTONEG_ENABLE) {
-		/* Advertise everything we can support. */
-		if (hp->sw_bmsr & BMSR_10HALF)
-			hp->sw_advertise |= (ADVERTISE_10HALF);
-		else
-			hp->sw_advertise &= ~(ADVERTISE_10HALF);
-
-		if (hp->sw_bmsr & BMSR_10FULL)
-			hp->sw_advertise |= (ADVERTISE_10FULL);
-		else
-			hp->sw_advertise &= ~(ADVERTISE_10FULL);
-		if (hp->sw_bmsr & BMSR_100HALF)
-			hp->sw_advertise |= (ADVERTISE_100HALF);
-		else
-			hp->sw_advertise &= ~(ADVERTISE_100HALF);
-		if (hp->sw_bmsr & BMSR_100FULL)
-			hp->sw_advertise |= (ADVERTISE_100FULL);
-		else
-			hp->sw_advertise &= ~(ADVERTISE_100FULL);
-		happy_meal_tcvr_write(hp, tregs, MII_ADVERTISE, hp->sw_advertise);
-
-		/* XXX Currently no Happy Meal cards I know off support 100BaseT4,
-		 * XXX and this is because the DP83840 does not support it, changes
-		 * XXX would need to be made to the tx/rx logic in the driver as well
-		 * XXX so I completely skip checking for it in the BMSR for now.
-		 */
-
-		ASD("Advertising [ %s%s%s%s]\n",
-		    hp->sw_advertise & ADVERTISE_10HALF ? "10H " : "",
-		    hp->sw_advertise & ADVERTISE_10FULL ? "10F " : "",
-		    hp->sw_advertise & ADVERTISE_100HALF ? "100H " : "",
-		    hp->sw_advertise & ADVERTISE_100FULL ? "100F " : "");
-
-		/* Enable Auto-Negotiation, this is usually on already... */
-		hp->sw_bmcr |= BMCR_ANENABLE;
-		happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
-
-		/* Restart it to make sure it is going. */
-		hp->sw_bmcr |= BMCR_ANRESTART;
-		happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
-
-		/* BMCR_ANRESTART self clears when the process has begun. */
-
-		timeout = 64;  /* More than enough. */
-		while (--timeout) {
-			hp->sw_bmcr = happy_meal_tcvr_read(hp, tregs, MII_BMCR);
-			if (!(hp->sw_bmcr & BMCR_ANRESTART))
-				break; /* got it. */
-			udelay(10);
-		}
-		if (!timeout) {
-			netdev_err(hp->dev,
-				   "Happy Meal would not start auto negotiation BMCR=0x%04x\n",
-				   hp->sw_bmcr);
-			netdev_notice(hp->dev,
-				      "Performing force link detection.\n");
-			goto force_link;
-		} else {
-			hp->timer_state = arbwait;
-		}
-	} else {
-force_link:
-		/* Force the link up, trying first a particular mode.
-		 * Either we are here at the request of ethtool or
-		 * because the Happy Meal would not start to autoneg.
-		 */
-
-		/* Disable auto-negotiation in BMCR, enable the duplex and
-		 * speed setting, init the timer state machine, and fire it off.
-		 */
-		if (!ep || ep->base.autoneg == AUTONEG_ENABLE) {
-			hp->sw_bmcr = BMCR_SPEED100;
-		} else {
-			if (ep->base.speed == SPEED_100)
-				hp->sw_bmcr = BMCR_SPEED100;
-			else
-				hp->sw_bmcr = 0;
-			if (ep->base.duplex == DUPLEX_FULL)
-				hp->sw_bmcr |= BMCR_FULLDPLX;
-		}
-		happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
-
-		if (!is_lucent_phy(hp)) {
-			/* OK, seems we need do disable the transceiver for the first
-			 * tick to make sure we get an accurate link state at the
-			 * second tick.
-			 */
-			hp->sw_csconfig = happy_meal_tcvr_read(hp, tregs,
-							       DP83840_CSCONFIG);
-			hp->sw_csconfig &= ~(CSCONFIG_TCVDISAB);
-			happy_meal_tcvr_write(hp, tregs, DP83840_CSCONFIG,
-					      hp->sw_csconfig);
-		}
-		hp->timer_state = ltrywait;
-	}
-
-	hp->timer_ticks = 0;
-	hp->happy_timer.expires = jiffies + (12 * HZ)/10;  /* 1.2 sec. */
-	add_timer(&hp->happy_timer);
-}
-
 /* hp->happy_lock must be held */
 static int happy_meal_init(struct happy_meal *hp)
 {
-- 
cgit v1.2.3-70-g09d2


From 3427372d0bd8c75036768d017074c29c99ce6342 Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:29 -0400
Subject: net: sunhme: Remove residual polling code

The sunhme driver never used the hardware MII polling feature. Even the
if-def'd out happy_meal_poll_start was removed by 2002 [1]. Remove the
various places in the driver which needlessly guard against MII interrupts
which will never be enabled.

[1] https://lwn.net/2002/0411/a/2.5.8-pre3.php3

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 134 +++++---------------------------------
 drivers/net/ethernet/sun/sunhme.h |   6 +-
 2 files changed, 18 insertions(+), 122 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index e52f2d9305bc..b02cb0b493ae 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -985,32 +985,6 @@ static void happy_meal_get_counters(struct happy_meal *hp, void __iomem *bregs)
 	hme_write32(hp, bregs + BMAC_LTCTR, 0);
 }
 
-/* hp->happy_lock must be held */
-static void happy_meal_poll_stop(struct happy_meal *hp, void __iomem *tregs)
-{
-	/* If polling disabled or not polling already, nothing to do. */
-	if ((hp->happy_flags & (HFLAG_POLLENABLE | HFLAG_POLL)) !=
-	   (HFLAG_POLLENABLE | HFLAG_POLL)) {
-		ASD("not polling, return\n");
-		return;
-	}
-
-	/* Shut up the MIF. */
-	ASD("were polling, mif ints off, polling off\n");
-	hme_write32(hp, tregs + TCVR_IMASK, 0xffff);
-
-	/* Turn off polling. */
-	hme_write32(hp, tregs + TCVR_CFG,
-		    hme_read32(hp, tregs + TCVR_CFG) & ~(TCV_CFG_PENABLE));
-
-	/* We are no longer polling. */
-	hp->happy_flags &= ~(HFLAG_POLL);
-
-	/* Let the bits set. */
-	udelay(200);
-	ASD("done\n");
-}
-
 /* Only Sun can take such nice parts and fuck up the programming interface
  * like this.  Good job guys...
  */
@@ -1115,57 +1089,26 @@ static int happy_meal_tcvr_reset(struct happy_meal *hp, void __iomem *tregs)
 static void happy_meal_transceiver_check(struct happy_meal *hp, void __iomem *tregs)
 {
 	unsigned long tconfig = hme_read32(hp, tregs + TCVR_CFG);
+	u32 reread = hme_read32(hp, tregs + TCVR_CFG);
 
 	ASD("tcfg=%08lx\n", tconfig);
-	if (hp->happy_flags & HFLAG_POLL) {
-		/* If we are polling, we must stop to get the transceiver type. */
-		if (hp->tcvr_type == internal) {
-			if (tconfig & TCV_CFG_MDIO1) {
-				happy_meal_poll_stop(hp, tregs);
-				hp->paddr = TCV_PADDR_ETX;
-				hp->tcvr_type = external;
-				tconfig &= ~(TCV_CFG_PENABLE);
-				tconfig |= TCV_CFG_PSELECT;
-				hme_write32(hp, tregs + TCVR_CFG, tconfig);
-				ASD("poll stop, internal->external\n");
-			}
-		} else {
-			if (hp->tcvr_type == external) {
-				if (!(hme_read32(hp, tregs + TCVR_STATUS) >> 16)) {
-					happy_meal_poll_stop(hp, tregs);
-					hp->paddr = TCV_PADDR_ITX;
-					hp->tcvr_type = internal;
-					hme_write32(hp, tregs + TCVR_CFG,
-						    hme_read32(hp, tregs + TCVR_CFG) &
-						    ~(TCV_CFG_PSELECT));
-					ASD("poll stop, external->internal\n");
-				}
-			} else {
-				ASD("polling, none\n");
-			}
-		}
+	if (reread & TCV_CFG_MDIO1) {
+		hme_write32(hp, tregs + TCVR_CFG, tconfig | TCV_CFG_PSELECT);
+		hp->paddr = TCV_PADDR_ETX;
+		hp->tcvr_type = external;
+		ASD("not polling, external\n");
 	} else {
-		u32 reread = hme_read32(hp, tregs + TCVR_CFG);
-
-		/* Else we can just work off of the MDIO bits. */
-		if (reread & TCV_CFG_MDIO1) {
-			hme_write32(hp, tregs + TCVR_CFG, tconfig | TCV_CFG_PSELECT);
-			hp->paddr = TCV_PADDR_ETX;
-			hp->tcvr_type = external;
-			ASD("not polling, external\n");
+		if (reread & TCV_CFG_MDIO0) {
+			hme_write32(hp, tregs + TCVR_CFG,
+				    tconfig & ~(TCV_CFG_PSELECT));
+			hp->paddr = TCV_PADDR_ITX;
+			hp->tcvr_type = internal;
+			ASD("not polling, internal\n");
 		} else {
-			if (reread & TCV_CFG_MDIO0) {
-				hme_write32(hp, tregs + TCVR_CFG,
-					    tconfig & ~(TCV_CFG_PSELECT));
-				hp->paddr = TCV_PADDR_ITX;
-				hp->tcvr_type = internal;
-				ASD("not polling, internal\n");
-			} else {
-				netdev_err(hp->dev,
-					   "Transceiver and a coke please.");
-				hp->tcvr_type = none; /* Grrr... */
-				ASD("not polling, none\n");
-			}
+			netdev_err(hp->dev,
+				   "Transceiver and a coke please.");
+			hp->tcvr_type = none; /* Grrr... */
+			ASD("not polling, none\n");
 		}
 	}
 }
@@ -1334,10 +1277,6 @@ static int happy_meal_init(struct happy_meal *hp)
 		happy_meal_get_counters(hp, bregs);
 	}
 
-	/* Stop polling. */
-	HMD("to happy_meal_poll_stop\n");
-	happy_meal_poll_stop(hp, tregs);
-
 	/* Stop transmitter and receiver. */
 	HMD("to happy_meal_stop\n");
 	happy_meal_stop(hp, gregs);
@@ -1346,11 +1285,6 @@ static int happy_meal_init(struct happy_meal *hp)
 	HMD("to happy_meal_init_rings\n");
 	happy_meal_init_rings(hp);
 
-	/* Shut up the MIF. */
-	HMD("Disable all MIF irqs (old[%08x])\n",
-	    hme_read32(hp, tregs + TCVR_IMASK));
-	hme_write32(hp, tregs + TCVR_IMASK, 0xffff);
-
 	/* See if we can enable the MIF frame on this card to speak to the DP83840. */
 	if (hp->happy_flags & HFLAG_FENABLE) {
 		HMD("use frame old[%08x]\n",
@@ -1605,7 +1539,6 @@ static void happy_meal_set_initial_advertisement(struct happy_meal *hp)
 	void __iomem *gregs	= hp->gregs;
 
 	happy_meal_stop(hp, gregs);
-	hme_write32(hp, tregs + TCVR_IMASK, 0xffff);
 	if (hp->happy_flags & HFLAG_FENABLE)
 		hme_write32(hp, tregs + TCVR_CFG,
 			    hme_read32(hp, tregs + TCVR_CFG) & ~(TCV_CFG_BENABLE));
@@ -1762,34 +1695,6 @@ static int happy_meal_is_not_so_happy(struct happy_meal *hp, u32 status)
 	return 0;
 }
 
-/* hp->happy_lock must be held */
-static void happy_meal_mif_interrupt(struct happy_meal *hp)
-{
-	void __iomem *tregs = hp->tcvregs;
-
-	netdev_info(hp->dev, "Link status change.\n");
-	hp->sw_bmcr = happy_meal_tcvr_read(hp, tregs, MII_BMCR);
-	hp->sw_lpa = happy_meal_tcvr_read(hp, tregs, MII_LPA);
-
-	/* Use the fastest transmission protocol possible. */
-	if (hp->sw_lpa & LPA_100FULL) {
-		netdev_info(hp->dev, "Switching to 100Mbps at full duplex.\n");
-		hp->sw_bmcr |= (BMCR_FULLDPLX | BMCR_SPEED100);
-	} else if (hp->sw_lpa & LPA_100HALF) {
-		netdev_info(hp->dev, "Switching to 100MBps at half duplex.\n");
-		hp->sw_bmcr |= BMCR_SPEED100;
-	} else if (hp->sw_lpa & LPA_10FULL) {
-		netdev_info(hp->dev, "Switching to 10MBps at full duplex.\n");
-		hp->sw_bmcr |= BMCR_FULLDPLX;
-	} else {
-		netdev_info(hp->dev, "Using 10Mbps at half duplex.\n");
-	}
-	happy_meal_tcvr_write(hp, tregs, MII_BMCR, hp->sw_bmcr);
-
-	/* Finally stop polling and shut up the MIF. */
-	happy_meal_poll_stop(hp, tregs);
-}
-
 /* hp->happy_lock must be held */
 static void happy_meal_tx(struct happy_meal *hp)
 {
@@ -1973,9 +1878,6 @@ static irqreturn_t happy_meal_interrupt(int irq, void *dev_id)
 			goto out;
 	}
 
-	if (happy_status & GREG_STAT_MIFIRQ)
-		happy_meal_mif_interrupt(hp);
-
 	if (happy_status & GREG_STAT_TXALL)
 		happy_meal_tx(hp);
 
@@ -2003,7 +1905,6 @@ static irqreturn_t quattro_sbus_interrupt(int irq, void *cookie)
 		HMD("status=%08x\n", happy_status);
 
 		if (!(happy_status & (GREG_STAT_ERRORS |
-				      GREG_STAT_MIFIRQ |
 				      GREG_STAT_TXALL |
 				      GREG_STAT_RXTOHOST)))
 			continue;
@@ -2014,9 +1915,6 @@ static irqreturn_t quattro_sbus_interrupt(int irq, void *cookie)
 			if (happy_meal_is_not_so_happy(hp, happy_status))
 				goto next;
 
-		if (happy_status & GREG_STAT_MIFIRQ)
-			happy_meal_mif_interrupt(hp);
-
 		if (happy_status & GREG_STAT_TXALL)
 			happy_meal_tx(hp);
 
diff --git a/drivers/net/ethernet/sun/sunhme.h b/drivers/net/ethernet/sun/sunhme.h
index 9118c60c9426..258b4c7fe962 100644
--- a/drivers/net/ethernet/sun/sunhme.h
+++ b/drivers/net/ethernet/sun/sunhme.h
@@ -462,22 +462,20 @@ struct happy_meal {
 };
 
 /* Here are the happy flags. */
-#define HFLAG_POLL                0x00000001      /* We are doing MIF polling          */
 #define HFLAG_FENABLE             0x00000002      /* The MII frame is enabled          */
 #define HFLAG_LANCE               0x00000004      /* We are using lance-mode           */
 #define HFLAG_RXENABLE            0x00000008      /* Receiver is enabled               */
 #define HFLAG_AUTO                0x00000010      /* Using auto-negotiation, 0 = force */
 #define HFLAG_FULL                0x00000020      /* Full duplex enable                */
 #define HFLAG_MACFULL             0x00000040      /* Using full duplex in the MAC      */
-#define HFLAG_POLLENABLE          0x00000080      /* Actually try MIF polling          */
 #define HFLAG_RXCV                0x00000100      /* XXX RXCV ENABLE                   */
 #define HFLAG_INIT                0x00000200      /* Init called at least once         */
 #define HFLAG_LINKUP              0x00000400      /* 1 = Link is up                    */
 #define HFLAG_PCI                 0x00000800      /* PCI based Happy Meal              */
 #define HFLAG_QUATTRO		  0x00001000      /* On QFE/Quattro card	       */
 
-#define HFLAG_20_21  (HFLAG_POLLENABLE | HFLAG_FENABLE)
-#define HFLAG_NOT_A0 (HFLAG_POLLENABLE | HFLAG_FENABLE | HFLAG_LANCE | HFLAG_RXCV)
+#define HFLAG_20_21  HFLAG_FENABLE
+#define HFLAG_NOT_A0 (HFLAG_FENABLE | HFLAG_LANCE | HFLAG_RXCV)
 
 /* Support for QFE/Quattro cards. */
 struct quattro {
-- 
cgit v1.2.3-70-g09d2


From 27b9ea8f37a6f5d6bd4c6867fd1910a978371572 Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:30 -0400
Subject: net: sunhme: Unify IRQ requesting

Instead of registering one interrupt handler for all four SBUS Quattro
HMEs, let each HME register its own handler. To make this work, we don't
handle the IRQ if none of the status bits are set. This reduces the
complexity of the driver, and makes it easier to ensure things happen
before/after enabling IRQs.

I'm not really sure why we request IRQs in two different places (and leave
them running after removing the driver!). A lot of things in this driver
seem to just be crusty, and not necessarily intentional. I'm assuming
that's the case here as well.

This really needs to be tested by someone with an SBUS Quattro card.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 131 +++-----------------------------------
 1 file changed, 10 insertions(+), 121 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index b02cb0b493ae..4efcf60644aa 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -1870,6 +1870,8 @@ static irqreturn_t happy_meal_interrupt(int irq, void *dev_id)
 	u32 happy_status       = hme_read32(hp, hp->gregs + GREG_STAT);
 
 	HMD("status=%08x\n", happy_status);
+	if (!happy_status)
+		return IRQ_NONE;
 
 	spin_lock(&hp->happy_lock);
 
@@ -1891,62 +1893,16 @@ out:
 	return IRQ_HANDLED;
 }
 
-#ifdef CONFIG_SBUS
-static irqreturn_t quattro_sbus_interrupt(int irq, void *cookie)
-{
-	struct quattro *qp = (struct quattro *) cookie;
-	int i;
-
-	for (i = 0; i < 4; i++) {
-		struct net_device *dev = qp->happy_meals[i];
-		struct happy_meal *hp  = netdev_priv(dev);
-		u32 happy_status       = hme_read32(hp, hp->gregs + GREG_STAT);
-
-		HMD("status=%08x\n", happy_status);
-
-		if (!(happy_status & (GREG_STAT_ERRORS |
-				      GREG_STAT_TXALL |
-				      GREG_STAT_RXTOHOST)))
-			continue;
-
-		spin_lock(&hp->happy_lock);
-
-		if (happy_status & GREG_STAT_ERRORS)
-			if (happy_meal_is_not_so_happy(hp, happy_status))
-				goto next;
-
-		if (happy_status & GREG_STAT_TXALL)
-			happy_meal_tx(hp);
-
-		if (happy_status & GREG_STAT_RXTOHOST)
-			happy_meal_rx(hp, dev);
-
-	next:
-		spin_unlock(&hp->happy_lock);
-	}
-	HMD("done\n");
-
-	return IRQ_HANDLED;
-}
-#endif
-
 static int happy_meal_open(struct net_device *dev)
 {
 	struct happy_meal *hp = netdev_priv(dev);
 	int res;
 
-	/* On SBUS Quattro QFE cards, all hme interrupts are concentrated
-	 * into a single source which we register handling at probe time.
-	 */
-	if ((hp->happy_flags & (HFLAG_QUATTRO|HFLAG_PCI)) != HFLAG_QUATTRO) {
-		res = request_irq(hp->irq, happy_meal_interrupt, IRQF_SHARED,
-				  dev->name, dev);
-		if (res) {
-			HMD("EAGAIN\n");
-			netdev_err(dev, "Can't order irq %d to go.\n", hp->irq);
-
-			return -EAGAIN;
-		}
+	res = request_irq(hp->irq, happy_meal_interrupt, IRQF_SHARED,
+			  dev->name, dev);
+	if (res) {
+		netdev_err(dev, "Can't order irq %d to go.\n", hp->irq);
+		return res;
 	}
 
 	HMD("to happy_meal_init\n");
@@ -1955,7 +1911,7 @@ static int happy_meal_open(struct net_device *dev)
 	res = happy_meal_init(hp);
 	spin_unlock_irq(&hp->happy_lock);
 
-	if (res && ((hp->happy_flags & (HFLAG_QUATTRO|HFLAG_PCI)) != HFLAG_QUATTRO))
+	if (res)
 		free_irq(hp->irq, dev);
 	return res;
 }
@@ -1973,12 +1929,7 @@ static int happy_meal_close(struct net_device *dev)
 
 	spin_unlock_irq(&hp->happy_lock);
 
-	/* On Quattro QFE cards, all hme interrupts are concentrated
-	 * into a single source which we register handling at probe
-	 * time and never unregister.
-	 */
-	if ((hp->happy_flags & (HFLAG_QUATTRO|HFLAG_PCI)) != HFLAG_QUATTRO)
-		free_irq(hp->irq, dev);
+	free_irq(hp->irq, dev);
 
 	return 0;
 }
@@ -2311,59 +2262,6 @@ static struct quattro *quattro_sbus_find(struct platform_device *child)
 	platform_set_drvdata(op, qp);
 	return qp;
 }
-
-/* After all quattro cards have been probed, we call these functions
- * to register the IRQ handlers for the cards that have been
- * successfully probed and skip the cards that failed to initialize
- */
-static int __init quattro_sbus_register_irqs(void)
-{
-	struct quattro *qp;
-
-	for (qp = qfe_sbus_list; qp != NULL; qp = qp->next) {
-		struct platform_device *op = qp->quattro_dev;
-		int err, qfe_slot, skip = 0;
-
-		for (qfe_slot = 0; qfe_slot < 4; qfe_slot++) {
-			if (!qp->happy_meals[qfe_slot])
-				skip = 1;
-		}
-		if (skip)
-			continue;
-
-		err = request_irq(op->archdata.irqs[0],
-				  quattro_sbus_interrupt,
-				  IRQF_SHARED, "Quattro",
-				  qp);
-		if (err != 0) {
-			dev_err(&op->dev,
-				"Quattro HME: IRQ registration error %d.\n",
-				err);
-			return err;
-		}
-	}
-
-	return 0;
-}
-
-static void quattro_sbus_free_irqs(void)
-{
-	struct quattro *qp;
-
-	for (qp = qfe_sbus_list; qp != NULL; qp = qp->next) {
-		struct platform_device *op = qp->quattro_dev;
-		int qfe_slot, skip = 0;
-
-		for (qfe_slot = 0; qfe_slot < 4; qfe_slot++) {
-			if (!qp->happy_meals[qfe_slot])
-				skip = 1;
-		}
-		if (skip)
-			continue;
-
-		free_irq(op->archdata.irqs[0], qp);
-	}
-}
 #endif /* CONFIG_SBUS */
 
 #ifdef CONFIG_PCI
@@ -3005,8 +2903,6 @@ static int hme_sbus_remove(struct platform_device *op)
 
 	unregister_netdev(net_dev);
 
-	/* XXX qfe parent interrupt... */
-
 	of_iounmap(&op->resource[0], hp->gregs, GREG_REG_SIZE);
 	of_iounmap(&op->resource[1], hp->etxregs, ETX_REG_SIZE);
 	of_iounmap(&op->resource[2], hp->erxregs, ERX_REG_SIZE);
@@ -3050,19 +2946,12 @@ static struct platform_driver hme_sbus_driver = {
 
 static int __init happy_meal_sbus_init(void)
 {
-	int err;
-
-	err = platform_driver_register(&hme_sbus_driver);
-	if (!err)
-		err = quattro_sbus_register_irqs();
-
-	return err;
+	return platform_driver_register(&hme_sbus_driver);
 }
 
 static void happy_meal_sbus_exit(void)
 {
 	platform_driver_unregister(&hme_sbus_driver);
-	quattro_sbus_free_irqs();
 
 	while (qfe_sbus_list) {
 		struct quattro *qfe = qfe_sbus_list;
-- 
cgit v1.2.3-70-g09d2


From 1ff4f42aef609757c334947f60612b4691b95c8a Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:31 -0400
Subject: net: sunhme: Alphabetize includes

Alphabetize includes to make it clearer where to add new ones.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 55 ++++++++++++++++++---------------------
 1 file changed, 25 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index 4efcf60644aa..e9613db3bdee 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -14,48 +14,43 @@
  *     argument : macaddr=0x00,0x10,0x20,0x30,0x40,0x50
  */
 
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/types.h>
+#include <asm/byteorder.h>
+#include <asm/dma.h>
+#include <asm/irq.h>
+#include <linux/bitops.h>
+#include <linux/crc32.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/errno.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
 #include <linux/fcntl.h>
-#include <linux/interrupt.h>
-#include <linux/ioport.h>
 #include <linux/in.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/delay.h>
 #include <linux/init.h>
-#include <linux/ethtool.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/ioport.h>
+#include <linux/kernel.h>
 #include <linux/mii.h>
-#include <linux/crc32.h>
-#include <linux/random.h>
-#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/module.h>
 #include <linux/netdevice.h>
-#include <linux/etherdevice.h>
+#include <linux/of_device.h>
+#include <linux/of.h>
+#include <linux/pci.h>
+#include <linux/random.h>
 #include <linux/skbuff.h>
-#include <linux/mm.h>
-#include <linux/bitops.h>
-#include <linux/dma-mapping.h>
-
-#include <asm/io.h>
-#include <asm/dma.h>
-#include <asm/byteorder.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
 
 #ifdef CONFIG_SPARC
-#include <linux/of.h>
-#include <linux/of_device.h>
+#include <asm/auxio.h>
 #include <asm/idprom.h>
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/prom.h>
-#include <asm/auxio.h>
-#endif
-#include <linux/uaccess.h>
-
-#include <asm/irq.h>
-
-#ifdef CONFIG_PCI
-#include <linux/pci.h>
 #endif
 
 #include "sunhme.h"
-- 
cgit v1.2.3-70-g09d2


From cc216e4b44ceed03774f291140fee62255d947d8 Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:32 -0400
Subject: net: sunhme: Switch SBUS to devres

The PCI half of this driver was converted in commit 914d9b2711dd ("sunhme:
switch to devres"). Do the same for the SBUS half.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 118 +++++++++++---------------------------
 1 file changed, 35 insertions(+), 83 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index e9613db3bdee..9cd3448aca3e 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2313,29 +2313,28 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 	struct net_device *dev;
 	int i, qfe_slot = -1;
 	u8 addr[ETH_ALEN];
-	int err = -ENODEV;
+	int err;
 
 	sbus_dp = op->dev.parent->of_node;
 
 	/* We can match PCI devices too, do not accept those here. */
 	if (!of_node_name_eq(sbus_dp, "sbus") && !of_node_name_eq(sbus_dp, "sbi"))
-		return err;
+		return -ENODEV;
 
 	if (is_qfe) {
 		qp = quattro_sbus_find(op);
 		if (qp == NULL)
-			goto err_out;
+			return -ENODEV;
 		for (qfe_slot = 0; qfe_slot < 4; qfe_slot++)
 			if (qp->happy_meals[qfe_slot] == NULL)
 				break;
 		if (qfe_slot == 4)
-			goto err_out;
+			return -ENODEV;
 	}
 
-	err = -ENOMEM;
-	dev = alloc_etherdev(sizeof(struct happy_meal));
+	dev = devm_alloc_etherdev(&op->dev, sizeof(struct happy_meal));
 	if (!dev)
-		goto err_out;
+		return -ENOMEM;
 	SET_NETDEV_DEV(dev, &op->dev);
 
 	/* If user did not specify a MAC address specifically, use
@@ -2369,46 +2368,45 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 
 	spin_lock_init(&hp->happy_lock);
 
-	err = -ENODEV;
 	if (qp != NULL) {
 		hp->qfe_parent = qp;
 		hp->qfe_ent = qfe_slot;
 		qp->happy_meals[qfe_slot] = dev;
 	}
 
-	hp->gregs = of_ioremap(&op->resource[0], 0,
-			       GREG_REG_SIZE, "HME Global Regs");
-	if (!hp->gregs) {
+	hp->gregs = devm_platform_ioremap_resource(op, 0);
+	if (IS_ERR(hp->gregs)) {
 		dev_err(&op->dev, "Cannot map global registers.\n");
-		goto err_out_free_netdev;
+		err = PTR_ERR(hp->gregs);
+		goto err_out_clear_quattro;
 	}
 
-	hp->etxregs = of_ioremap(&op->resource[1], 0,
-				 ETX_REG_SIZE, "HME TX Regs");
-	if (!hp->etxregs) {
+	hp->etxregs = devm_platform_ioremap_resource(op, 1);
+	if (IS_ERR(hp->etxregs)) {
 		dev_err(&op->dev, "Cannot map MAC TX registers.\n");
-		goto err_out_iounmap;
+		err = PTR_ERR(hp->etxregs);
+		goto err_out_clear_quattro;
 	}
 
-	hp->erxregs = of_ioremap(&op->resource[2], 0,
-				 ERX_REG_SIZE, "HME RX Regs");
-	if (!hp->erxregs) {
+	hp->erxregs = devm_platform_ioremap_resource(op, 2);
+	if (IS_ERR(hp->erxregs)) {
 		dev_err(&op->dev, "Cannot map MAC RX registers.\n");
-		goto err_out_iounmap;
+		err = PTR_ERR(hp->erxregs);
+		goto err_out_clear_quattro;
 	}
 
-	hp->bigmacregs = of_ioremap(&op->resource[3], 0,
-				    BMAC_REG_SIZE, "HME BIGMAC Regs");
-	if (!hp->bigmacregs) {
+	hp->bigmacregs = devm_platform_ioremap_resource(op, 3);
+	if (IS_ERR(hp->bigmacregs)) {
 		dev_err(&op->dev, "Cannot map BIGMAC registers.\n");
-		goto err_out_iounmap;
+		err = PTR_ERR(hp->bigmacregs);
+		goto err_out_clear_quattro;
 	}
 
-	hp->tcvregs = of_ioremap(&op->resource[4], 0,
-				 TCVR_REG_SIZE, "HME Tranceiver Regs");
-	if (!hp->tcvregs) {
+	hp->tcvregs = devm_platform_ioremap_resource(op, 4);
+	if (IS_ERR(hp->tcvregs)) {
 		dev_err(&op->dev, "Cannot map TCVR registers.\n");
-		goto err_out_iounmap;
+		err = PTR_ERR(hp->tcvregs);
+		goto err_out_clear_quattro;
 	}
 
 	hp->hm_revision = of_getintprop_default(dp, "hm-rev", 0xff);
@@ -2428,13 +2426,12 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 	hp->happy_bursts = of_getintprop_default(sbus_dp,
 						 "burst-sizes", 0x00);
 
-	hp->happy_block = dma_alloc_coherent(hp->dma_dev,
-					     PAGE_SIZE,
-					     &hp->hblock_dvma,
-					     GFP_ATOMIC);
-	err = -ENOMEM;
-	if (!hp->happy_block)
-		goto err_out_iounmap;
+	hp->happy_block = dmam_alloc_coherent(&op->dev, PAGE_SIZE,
+					      &hp->hblock_dvma, GFP_KERNEL);
+	if (!hp->happy_block) {
+		err = -ENOMEM;
+		goto err_out_clear_quattro;
+	}
 
 	/* Force check of the link first time we are brought up. */
 	hp->linkcheck = 0;
@@ -2472,10 +2469,10 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 	happy_meal_set_initial_advertisement(hp);
 	spin_unlock_irq(&hp->happy_lock);
 
-	err = register_netdev(hp->dev);
+	err = devm_register_netdev(&op->dev, dev);
 	if (err) {
 		dev_err(&op->dev, "Cannot register net device, aborting.\n");
-		goto err_out_free_coherent;
+		goto err_out_clear_quattro;
 	}
 
 	platform_set_drvdata(op, hp);
@@ -2490,31 +2487,9 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 
 	return 0;
 
-err_out_free_coherent:
-	dma_free_coherent(hp->dma_dev,
-			  PAGE_SIZE,
-			  hp->happy_block,
-			  hp->hblock_dvma);
-
-err_out_iounmap:
-	if (hp->gregs)
-		of_iounmap(&op->resource[0], hp->gregs, GREG_REG_SIZE);
-	if (hp->etxregs)
-		of_iounmap(&op->resource[1], hp->etxregs, ETX_REG_SIZE);
-	if (hp->erxregs)
-		of_iounmap(&op->resource[2], hp->erxregs, ERX_REG_SIZE);
-	if (hp->bigmacregs)
-		of_iounmap(&op->resource[3], hp->bigmacregs, BMAC_REG_SIZE);
-	if (hp->tcvregs)
-		of_iounmap(&op->resource[4], hp->tcvregs, TCVR_REG_SIZE);
-
+err_out_clear_quattro:
 	if (qp)
 		qp->happy_meals[qfe_slot] = NULL;
-
-err_out_free_netdev:
-	free_netdev(dev);
-
-err_out:
 	return err;
 }
 #endif
@@ -2891,28 +2866,6 @@ static int hme_sbus_probe(struct platform_device *op)
 	return happy_meal_sbus_probe_one(op, is_qfe);
 }
 
-static int hme_sbus_remove(struct platform_device *op)
-{
-	struct happy_meal *hp = platform_get_drvdata(op);
-	struct net_device *net_dev = hp->dev;
-
-	unregister_netdev(net_dev);
-
-	of_iounmap(&op->resource[0], hp->gregs, GREG_REG_SIZE);
-	of_iounmap(&op->resource[1], hp->etxregs, ETX_REG_SIZE);
-	of_iounmap(&op->resource[2], hp->erxregs, ERX_REG_SIZE);
-	of_iounmap(&op->resource[3], hp->bigmacregs, BMAC_REG_SIZE);
-	of_iounmap(&op->resource[4], hp->tcvregs, TCVR_REG_SIZE);
-	dma_free_coherent(hp->dma_dev,
-			  PAGE_SIZE,
-			  hp->happy_block,
-			  hp->hblock_dvma);
-
-	free_netdev(net_dev);
-
-	return 0;
-}
-
 static const struct of_device_id hme_sbus_match[] = {
 	{
 		.name = "SUNW,hme",
@@ -2936,7 +2889,6 @@ static struct platform_driver hme_sbus_driver = {
 		.of_match_table = hme_sbus_match,
 	},
 	.probe		= hme_sbus_probe,
-	.remove		= hme_sbus_remove,
 };
 
 static int __init happy_meal_sbus_init(void)
-- 
cgit v1.2.3-70-g09d2


From 273fb669c62cf7ac237036da1335d85d3e99eeba Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:33 -0400
Subject: net: sunhme: Consolidate mac address initialization

The mac address initialization is braodly the same between PCI and SBUS,
and one was clearly copied from the other. Consolidate them. We still have
to have some ifdefs because pci_(un)map_rom is only implemented for PCI,
and idprom is only implemented for SPARC.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 282 ++++++++++++++++++--------------------
 1 file changed, 134 insertions(+), 148 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index 9cd3448aca3e..4fe67623b924 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2304,6 +2304,133 @@ static const struct net_device_ops hme_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
+#ifdef CONFIG_PCI
+static int is_quattro_p(struct pci_dev *pdev)
+{
+	struct pci_dev *busdev = pdev->bus->self;
+	struct pci_dev *this_pdev;
+	int n_hmes;
+
+	if (!busdev || busdev->vendor != PCI_VENDOR_ID_DEC ||
+	    busdev->device != PCI_DEVICE_ID_DEC_21153)
+		return 0;
+
+	n_hmes = 0;
+	list_for_each_entry(this_pdev, &pdev->bus->devices, bus_list) {
+		if (this_pdev->vendor == PCI_VENDOR_ID_SUN &&
+		    this_pdev->device == PCI_DEVICE_ID_SUN_HAPPYMEAL)
+			n_hmes++;
+	}
+
+	if (n_hmes != 4)
+		return 0;
+
+	return 1;
+}
+
+/* Fetch MAC address from vital product data of PCI ROM. */
+static int find_eth_addr_in_vpd(void __iomem *rom_base, int len, int index, unsigned char *dev_addr)
+{
+	int this_offset;
+
+	for (this_offset = 0x20; this_offset < len; this_offset++) {
+		void __iomem *p = rom_base + this_offset;
+
+		if (readb(p + 0) != 0x90 ||
+		    readb(p + 1) != 0x00 ||
+		    readb(p + 2) != 0x09 ||
+		    readb(p + 3) != 0x4e ||
+		    readb(p + 4) != 0x41 ||
+		    readb(p + 5) != 0x06)
+			continue;
+
+		this_offset += 6;
+		p += 6;
+
+		if (index == 0) {
+			int i;
+
+			for (i = 0; i < 6; i++)
+				dev_addr[i] = readb(p + i);
+			return 1;
+		}
+		index--;
+	}
+	return 0;
+}
+
+static void __maybe_unused get_hme_mac_nonsparc(struct pci_dev *pdev,
+						unsigned char *dev_addr)
+{
+	size_t size;
+	void __iomem *p = pci_map_rom(pdev, &size);
+
+	if (p) {
+		int index = 0;
+		int found;
+
+		if (is_quattro_p(pdev))
+			index = PCI_SLOT(pdev->devfn);
+
+		found = readb(p) == 0x55 &&
+			readb(p + 1) == 0xaa &&
+			find_eth_addr_in_vpd(p, (64 * 1024), index, dev_addr);
+		pci_unmap_rom(pdev, p);
+		if (found)
+			return;
+	}
+
+	/* Sun MAC prefix then 3 random bytes. */
+	dev_addr[0] = 0x08;
+	dev_addr[1] = 0x00;
+	dev_addr[2] = 0x20;
+	get_random_bytes(&dev_addr[3], 3);
+}
+#endif /* !(CONFIG_SPARC) */
+
+static void happy_meal_addr_init(struct happy_meal *hp,
+				 struct device_node *dp, int qfe_slot)
+{
+	int i;
+
+	for (i = 0; i < 6; i++) {
+		if (macaddr[i] != 0)
+			break;
+	}
+
+	if (i < 6) { /* a mac address was given */
+		u8 addr[ETH_ALEN];
+
+		for (i = 0; i < 6; i++)
+			addr[i] = macaddr[i];
+		eth_hw_addr_set(hp->dev, addr);
+		macaddr[5]++;
+	} else {
+#ifdef CONFIG_SPARC
+		const unsigned char *addr;
+		int len;
+
+		/* If user did not specify a MAC address specifically, use
+		 * the Quattro local-mac-address property...
+		 */
+		if (qfe_slot != -1) {
+			addr = of_get_property(dp, "local-mac-address", &len);
+			if (addr && len == 6) {
+				eth_hw_addr_set(hp->dev, addr);
+				return;
+			}
+		}
+
+		eth_hw_addr_set(hp->dev, idprom->id_ethaddr);
+#else
+		u8 addr[ETH_ALEN];
+
+		get_hme_mac_nonsparc(hp->happy_dev, addr);
+		eth_hw_addr_set(hp->dev, addr);
+#endif
+	}
+}
+
 #ifdef CONFIG_SBUS
 static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 {
@@ -2311,8 +2438,7 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 	struct quattro *qp = NULL;
 	struct happy_meal *hp;
 	struct net_device *dev;
-	int i, qfe_slot = -1;
-	u8 addr[ETH_ALEN];
+	int qfe_slot = -1;
 	int err;
 
 	sbus_dp = op->dev.parent->of_node;
@@ -2337,34 +2463,11 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 		return -ENOMEM;
 	SET_NETDEV_DEV(dev, &op->dev);
 
-	/* If user did not specify a MAC address specifically, use
-	 * the Quattro local-mac-address property...
-	 */
-	for (i = 0; i < 6; i++) {
-		if (macaddr[i] != 0)
-			break;
-	}
-	if (i < 6) { /* a mac address was given */
-		for (i = 0; i < 6; i++)
-			addr[i] = macaddr[i];
-		eth_hw_addr_set(dev, addr);
-		macaddr[5]++;
-	} else {
-		const unsigned char *addr;
-		int len;
-
-		addr = of_get_property(dp, "local-mac-address", &len);
-
-		if (qfe_slot != -1 && addr && len == ETH_ALEN)
-			eth_hw_addr_set(dev, addr);
-		else
-			eth_hw_addr_set(dev, idprom->id_ethaddr);
-	}
-
 	hp = netdev_priv(dev);
-
+	hp->dev = dev;
 	hp->happy_dev = op;
 	hp->dma_dev = &op->dev;
+	happy_meal_addr_init(hp, dp, qfe_slot);
 
 	spin_lock_init(&hp->happy_lock);
 
@@ -2442,7 +2545,6 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 
 	timer_setup(&hp->happy_timer, happy_meal_timer, 0);
 
-	hp->dev = dev;
 	dev->netdev_ops = &hme_netdev_ops;
 	dev->watchdog_timeo = 5*HZ;
 	dev->ethtool_ops = &hme_ethtool_ops;
@@ -2495,104 +2597,17 @@ err_out_clear_quattro:
 #endif
 
 #ifdef CONFIG_PCI
-#ifndef CONFIG_SPARC
-static int is_quattro_p(struct pci_dev *pdev)
-{
-	struct pci_dev *busdev = pdev->bus->self;
-	struct pci_dev *this_pdev;
-	int n_hmes;
-
-	if (busdev == NULL ||
-	    busdev->vendor != PCI_VENDOR_ID_DEC ||
-	    busdev->device != PCI_DEVICE_ID_DEC_21153)
-		return 0;
-
-	n_hmes = 0;
-	list_for_each_entry(this_pdev, &pdev->bus->devices, bus_list) {
-		if (this_pdev->vendor == PCI_VENDOR_ID_SUN &&
-		    this_pdev->device == PCI_DEVICE_ID_SUN_HAPPYMEAL)
-			n_hmes++;
-	}
-
-	if (n_hmes != 4)
-		return 0;
-
-	return 1;
-}
-
-/* Fetch MAC address from vital product data of PCI ROM. */
-static int find_eth_addr_in_vpd(void __iomem *rom_base, int len, int index, unsigned char *dev_addr)
-{
-	int this_offset;
-
-	for (this_offset = 0x20; this_offset < len; this_offset++) {
-		void __iomem *p = rom_base + this_offset;
-
-		if (readb(p + 0) != 0x90 ||
-		    readb(p + 1) != 0x00 ||
-		    readb(p + 2) != 0x09 ||
-		    readb(p + 3) != 0x4e ||
-		    readb(p + 4) != 0x41 ||
-		    readb(p + 5) != 0x06)
-			continue;
-
-		this_offset += 6;
-		p += 6;
-
-		if (index == 0) {
-			int i;
-
-			for (i = 0; i < 6; i++)
-				dev_addr[i] = readb(p + i);
-			return 1;
-		}
-		index--;
-	}
-	return 0;
-}
-
-static void get_hme_mac_nonsparc(struct pci_dev *pdev, unsigned char *dev_addr)
-{
-	size_t size;
-	void __iomem *p = pci_map_rom(pdev, &size);
-
-	if (p) {
-		int index = 0;
-		int found;
-
-		if (is_quattro_p(pdev))
-			index = PCI_SLOT(pdev->devfn);
-
-		found = readb(p) == 0x55 &&
-			readb(p + 1) == 0xaa &&
-			find_eth_addr_in_vpd(p, (64 * 1024), index, dev_addr);
-		pci_unmap_rom(pdev, p);
-		if (found)
-			return;
-	}
-
-	/* Sun MAC prefix then 3 random bytes. */
-	dev_addr[0] = 0x08;
-	dev_addr[1] = 0x00;
-	dev_addr[2] = 0x20;
-	get_random_bytes(&dev_addr[3], 3);
-}
-#endif /* !(CONFIG_SPARC) */
-
 static int happy_meal_pci_probe(struct pci_dev *pdev,
 				const struct pci_device_id *ent)
 {
+	struct device_node *dp = NULL;
 	struct quattro *qp = NULL;
-#ifdef CONFIG_SPARC
-	struct device_node *dp;
-#endif
 	struct happy_meal *hp;
 	struct net_device *dev;
 	void __iomem *hpreg_base;
 	struct resource *hpreg_res;
-	int i, qfe_slot = -1;
 	char prom_name[64];
-	u8 addr[ETH_ALEN];
+	int qfe_slot = -1;
 	int err = -ENODEV;
 
 	/* Now make sure pci_dev cookie is there. */
@@ -2634,7 +2649,7 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	hp = netdev_priv(dev);
-
+	hp->dev = dev;
 	hp->happy_dev = pdev;
 	hp->dma_dev = &pdev->dev;
 
@@ -2670,35 +2685,7 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 		goto err_out_clear_quattro;
 	}
 
-	for (i = 0; i < 6; i++) {
-		if (macaddr[i] != 0)
-			break;
-	}
-	if (i < 6) { /* a mac address was given */
-		for (i = 0; i < 6; i++)
-			addr[i] = macaddr[i];
-		eth_hw_addr_set(dev, addr);
-		macaddr[5]++;
-	} else {
-#ifdef CONFIG_SPARC
-		const unsigned char *addr;
-		int len;
-
-		if (qfe_slot != -1 &&
-		    (addr = of_get_property(dp, "local-mac-address", &len))
-			!= NULL &&
-		    len == 6) {
-			eth_hw_addr_set(dev, addr);
-		} else {
-			eth_hw_addr_set(dev, idprom->id_ethaddr);
-		}
-#else
-		u8 addr[ETH_ALEN];
-
-		get_hme_mac_nonsparc(pdev, addr);
-		eth_hw_addr_set(dev, addr);
-#endif
-	}
+	happy_meal_addr_init(hp, dp, qfe_slot);
 
 	/* Layout registers. */
 	hp->gregs      = (hpreg_base + 0x0000UL);
@@ -2747,7 +2734,6 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 	timer_setup(&hp->happy_timer, happy_meal_timer, 0);
 
 	hp->irq = pdev->irq;
-	hp->dev = dev;
 	dev->netdev_ops = &hme_netdev_ops;
 	dev->watchdog_timeo = 5*HZ;
 	dev->ethtool_ops = &hme_ethtool_ops;
-- 
cgit v1.2.3-70-g09d2


From d1f088196057a0e9b65156943fba71c9b6bb610d Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:34 -0400
Subject: net: sunhme: Clean up mac address init

Clean up some oddities suggested during review.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index 4fe67623b924..b51b8930bef1 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2348,9 +2348,7 @@ static int find_eth_addr_in_vpd(void __iomem *rom_base, int len, int index, unsi
 		p += 6;
 
 		if (index == 0) {
-			int i;
-
-			for (i = 0; i < 6; i++)
+			for (int i = 0; i < 6; i++)
 				dev_addr[i] = readb(p + i);
 			return 1;
 		}
@@ -2362,9 +2360,10 @@ static int find_eth_addr_in_vpd(void __iomem *rom_base, int len, int index, unsi
 static void __maybe_unused get_hme_mac_nonsparc(struct pci_dev *pdev,
 						unsigned char *dev_addr)
 {
+	void __iomem *p;
 	size_t size;
-	void __iomem *p = pci_map_rom(pdev, &size);
 
+	p = pci_map_rom(pdev, &size);
 	if (p) {
 		int index = 0;
 		int found;
@@ -2386,7 +2385,7 @@ static void __maybe_unused get_hme_mac_nonsparc(struct pci_dev *pdev,
 	dev_addr[2] = 0x20;
 	get_random_bytes(&dev_addr[3], 3);
 }
-#endif /* !(CONFIG_SPARC) */
+#endif
 
 static void happy_meal_addr_init(struct happy_meal *hp,
 				 struct device_node *dp, int qfe_slot)
-- 
cgit v1.2.3-70-g09d2


From 902fe6e903686ed36d030ae402aa86ada9f2710b Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:35 -0400
Subject: net: sunhme: Inline error returns

The err_out label used to have cleanup. Now that it just returns, inline it
everywhere.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index b51b8930bef1..bd1925f575c4 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2622,29 +2622,25 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 
 	err = pcim_enable_device(pdev);
 	if (err)
-		goto err_out;
+		return err;
 	pci_set_master(pdev);
 
 	if (!strcmp(prom_name, "SUNW,qfe") || !strcmp(prom_name, "qfe")) {
 		qp = quattro_pci_find(pdev);
-		if (IS_ERR(qp)) {
-			err = PTR_ERR(qp);
-			goto err_out;
-		}
+		if (IS_ERR(qp))
+			return PTR_ERR(qp);
 
 		for (qfe_slot = 0; qfe_slot < 4; qfe_slot++)
 			if (!qp->happy_meals[qfe_slot])
 				break;
 
 		if (qfe_slot == 4)
-			goto err_out;
+			return -ENODEV;
 	}
 
 	dev = devm_alloc_etherdev(&pdev->dev, sizeof(struct happy_meal));
-	if (!dev) {
-		err = -ENOMEM;
-		goto err_out;
-	}
+	if (!dev)
+		return -ENOMEM;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	hp = netdev_priv(dev);
@@ -2792,8 +2788,6 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 err_out_clear_quattro:
 	if (qp != NULL)
 		qp->happy_meals[qfe_slot] = NULL;
-
-err_out:
 	return err;
 }
 
-- 
cgit v1.2.3-70-g09d2


From ecdcd0428c594f6a6ee157fafb4c6265785be054 Mon Sep 17 00:00:00 2001
From: Sean Anderson <seanga2@gmail.com>
Date: Fri, 24 Mar 2023 13:51:36 -0400
Subject: net: sunhme: Consolidate common probe tasks

Most of the second half of the PCI/SBUS probe functions are the same.
Consolidate them into a common function.

Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sun/sunhme.c | 157 ++++++++++++++++----------------------
 1 file changed, 64 insertions(+), 93 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index bd1925f575c4..ec85aef35bf9 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2430,6 +2430,58 @@ static void happy_meal_addr_init(struct happy_meal *hp,
 	}
 }
 
+static int happy_meal_common_probe(struct happy_meal *hp,
+				   struct device_node *dp)
+{
+	struct net_device *dev = hp->dev;
+	int err;
+
+#ifdef CONFIG_SPARC
+	hp->hm_revision = of_getintprop_default(dp, "hm-rev", hp->hm_revision);
+#endif
+
+	/* Now enable the feature flags we can. */
+	if (hp->hm_revision == 0x20 || hp->hm_revision == 0x21)
+		hp->happy_flags |= HFLAG_20_21;
+	else if (hp->hm_revision != 0xa0)
+		hp->happy_flags |= HFLAG_NOT_A0;
+
+	hp->happy_block = dmam_alloc_coherent(hp->dma_dev, PAGE_SIZE,
+					      &hp->hblock_dvma, GFP_KERNEL);
+	if (!hp->happy_block)
+		return -ENOMEM;
+
+	/* Force check of the link first time we are brought up. */
+	hp->linkcheck = 0;
+
+	/* Force timer state to 'asleep' with count of zero. */
+	hp->timer_state = asleep;
+	hp->timer_ticks = 0;
+
+	timer_setup(&hp->happy_timer, happy_meal_timer, 0);
+
+	dev->netdev_ops = &hme_netdev_ops;
+	dev->watchdog_timeo = 5 * HZ;
+	dev->ethtool_ops = &hme_ethtool_ops;
+
+	/* Happy Meal can do it all... */
+	dev->hw_features = NETIF_F_SG | NETIF_F_HW_CSUM;
+	dev->features |= dev->hw_features | NETIF_F_RXCSUM;
+
+
+	/* Grrr, Happy Meal comes up by default not advertising
+	 * full duplex 100baseT capabilities, fix this.
+	 */
+	spin_lock_irq(&hp->happy_lock);
+	happy_meal_set_initial_advertisement(hp);
+	spin_unlock_irq(&hp->happy_lock);
+
+	err = devm_register_netdev(hp->dma_dev, dev);
+	if (err)
+		dev_err(hp->dma_dev, "Cannot register net device, aborting.\n");
+	return err;
+}
+
 #ifdef CONFIG_SBUS
 static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 {
@@ -2511,50 +2563,18 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 		goto err_out_clear_quattro;
 	}
 
-	hp->hm_revision = of_getintprop_default(dp, "hm-rev", 0xff);
-	if (hp->hm_revision == 0xff)
-		hp->hm_revision = 0xa0;
-
-	/* Now enable the feature flags we can. */
-	if (hp->hm_revision == 0x20 || hp->hm_revision == 0x21)
-		hp->happy_flags = HFLAG_20_21;
-	else if (hp->hm_revision != 0xa0)
-		hp->happy_flags = HFLAG_NOT_A0;
+	hp->hm_revision = 0xa0;
 
 	if (qp != NULL)
 		hp->happy_flags |= HFLAG_QUATTRO;
 
+	hp->irq = op->archdata.irqs[0];
+
 	/* Get the supported DVMA burst sizes from our Happy SBUS. */
 	hp->happy_bursts = of_getintprop_default(sbus_dp,
 						 "burst-sizes", 0x00);
 
-	hp->happy_block = dmam_alloc_coherent(&op->dev, PAGE_SIZE,
-					      &hp->hblock_dvma, GFP_KERNEL);
-	if (!hp->happy_block) {
-		err = -ENOMEM;
-		goto err_out_clear_quattro;
-	}
-
-	/* Force check of the link first time we are brought up. */
-	hp->linkcheck = 0;
-
-	/* Force timer state to 'asleep' with count of zero. */
-	hp->timer_state = asleep;
-	hp->timer_ticks = 0;
-
-	timer_setup(&hp->happy_timer, happy_meal_timer, 0);
-
-	dev->netdev_ops = &hme_netdev_ops;
-	dev->watchdog_timeo = 5*HZ;
-	dev->ethtool_ops = &hme_ethtool_ops;
-
-	/* Happy Meal can do it all... */
-	dev->hw_features = NETIF_F_SG | NETIF_F_HW_CSUM;
-	dev->features |= dev->hw_features | NETIF_F_RXCSUM;
-
-	hp->irq = op->archdata.irqs[0];
-
-#if defined(CONFIG_SBUS) && defined(CONFIG_PCI)
+#ifdef CONFIG_PCI
 	/* Hook up SBUS register/descriptor accessors. */
 	hp->read_desc32 = sbus_hme_read_desc32;
 	hp->write_txd = sbus_hme_write_txd;
@@ -2563,18 +2583,9 @@ static int happy_meal_sbus_probe_one(struct platform_device *op, int is_qfe)
 	hp->write32 = sbus_hme_write32;
 #endif
 
-	/* Grrr, Happy Meal comes up by default not advertising
-	 * full duplex 100baseT capabilities, fix this.
-	 */
-	spin_lock_irq(&hp->happy_lock);
-	happy_meal_set_initial_advertisement(hp);
-	spin_unlock_irq(&hp->happy_lock);
-
-	err = devm_register_netdev(&op->dev, dev);
-	if (err) {
-		dev_err(&op->dev, "Cannot register net device, aborting.\n");
+	err = happy_meal_common_probe(hp, dp);
+	if (err)
 		goto err_out_clear_quattro;
-	}
 
 	platform_set_drvdata(op, hp);
 
@@ -2689,20 +2700,10 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 	hp->bigmacregs = (hpreg_base + 0x6000UL);
 	hp->tcvregs    = (hpreg_base + 0x7000UL);
 
-#ifdef CONFIG_SPARC
-	hp->hm_revision = of_getintprop_default(dp, "hm-rev", 0xff);
-	if (hp->hm_revision == 0xff)
+	if (IS_ENABLED(CONFIG_SPARC))
 		hp->hm_revision = 0xc0 | (pdev->revision & 0x0f);
-#else
-	/* works with this on non-sparc hosts */
-	hp->hm_revision = 0x20;
-#endif
-
-	/* Now enable the feature flags we can. */
-	if (hp->hm_revision == 0x20 || hp->hm_revision == 0x21)
-		hp->happy_flags = HFLAG_20_21;
-	else if (hp->hm_revision != 0xa0 && hp->hm_revision != 0xc0)
-		hp->happy_flags = HFLAG_NOT_A0;
+	else
+		hp->hm_revision = 0x20;
 
 	if (qp != NULL)
 		hp->happy_flags |= HFLAG_QUATTRO;
@@ -2714,30 +2715,9 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 	/* Assume PCI happy meals can handle all burst sizes. */
 	hp->happy_bursts = DMA_BURSTBITS;
 #endif
-
-	hp->happy_block = dmam_alloc_coherent(&pdev->dev, PAGE_SIZE,
-					      &hp->hblock_dvma, GFP_KERNEL);
-	if (!hp->happy_block) {
-		err = -ENOMEM;
-		goto err_out_clear_quattro;
-	}
-
-	hp->linkcheck = 0;
-	hp->timer_state = asleep;
-	hp->timer_ticks = 0;
-
-	timer_setup(&hp->happy_timer, happy_meal_timer, 0);
-
 	hp->irq = pdev->irq;
-	dev->netdev_ops = &hme_netdev_ops;
-	dev->watchdog_timeo = 5*HZ;
-	dev->ethtool_ops = &hme_ethtool_ops;
 
-	/* Happy Meal can do it all... */
-	dev->hw_features = NETIF_F_SG | NETIF_F_HW_CSUM;
-	dev->features |= dev->hw_features | NETIF_F_RXCSUM;
-
-#if defined(CONFIG_SBUS) && defined(CONFIG_PCI)
+#ifdef CONFIG_SBUS
 	/* Hook up PCI register/descriptor accessors. */
 	hp->read_desc32 = pci_hme_read_desc32;
 	hp->write_txd = pci_hme_write_txd;
@@ -2746,18 +2726,9 @@ static int happy_meal_pci_probe(struct pci_dev *pdev,
 	hp->write32 = pci_hme_write32;
 #endif
 
-	/* Grrr, Happy Meal comes up by default not advertising
-	 * full duplex 100baseT capabilities, fix this.
-	 */
-	spin_lock_irq(&hp->happy_lock);
-	happy_meal_set_initial_advertisement(hp);
-	spin_unlock_irq(&hp->happy_lock);
-
-	err = devm_register_netdev(&pdev->dev, dev);
-	if (err) {
-		dev_err(&pdev->dev, "Cannot register net device, aborting.\n");
+	err = happy_meal_common_probe(hp, dp);
+	if (err)
 		goto err_out_clear_quattro;
-	}
 
 	pci_set_drvdata(pdev, hp);
 
-- 
cgit v1.2.3-70-g09d2


From dd3a7d58dcc2434632bed31683ff134b7d6d1da4 Mon Sep 17 00:00:00 2001
From: Michal Michalik <michal.michalik@intel.com>
Date: Fri, 24 Mar 2023 18:52:58 +0100
Subject: tools: ynl: Add missing types to encode/decode

While testing the tool I noticed we miss the u16 type on payload create.
On the code inspection it turned out we miss also u64 - add them.

We also miss the decoding of u16 despite the fact `NlAttr` class
supports it - add it.

Signed-off-by: Michal Michalik <michal.michalik@intel.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/net/ynl/lib/ynl.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 98ecfdb44a83..7eaf066b115e 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -336,8 +336,12 @@ class YnlFamily(SpecFamily):
             attr_payload = b''
         elif attr["type"] == 'u8':
             attr_payload = struct.pack("B", int(value))
+        elif attr["type"] == 'u16':
+            attr_payload = struct.pack("H", int(value))
         elif attr["type"] == 'u32':
             attr_payload = struct.pack("I", int(value))
+        elif attr["type"] == 'u64':
+            attr_payload = struct.pack("Q", int(value))
         elif attr["type"] == 'string':
             attr_payload = str(value).encode('ascii') + b'\x00'
         elif attr["type"] == 'binary':
@@ -373,6 +377,8 @@ class YnlFamily(SpecFamily):
                 decoded = subdict
             elif attr_spec['type'] == 'u8':
                 decoded = attr.as_u8()
+            elif attr_spec['type'] == 'u16':
+                decoded = attr.as_u16()
             elif attr_spec['type'] == 'u32':
                 decoded = attr.as_u32()
             elif attr_spec['type'] == 'u64':
-- 
cgit v1.2.3-70-g09d2


From a504d246d2129b6fe40059d372288ffb36e4588b Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Fri, 24 Mar 2023 11:17:57 -0700
Subject: selftests: tls: add a test for queuing data before setting the ULP

Other tests set up the connection fully on both ends before
communicating any data. Add a test which will queue up TLS
records to TCP before the TLS ULP is installed.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/tls.c | 45 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)

diff --git a/tools/testing/selftests/net/tls.c b/tools/testing/selftests/net/tls.c
index 2cbb12736596..e699548d4247 100644
--- a/tools/testing/selftests/net/tls.c
+++ b/tools/testing/selftests/net/tls.c
@@ -1820,4 +1820,49 @@ TEST(tls_v6ops) {
 	close(sfd);
 }
 
+TEST(prequeue) {
+	struct tls_crypto_info_keys tls12;
+	char buf[20000], buf2[20000];
+	struct sockaddr_in addr;
+	int sfd, cfd, ret, fd;
+	socklen_t len;
+
+	len = sizeof(addr);
+	memrnd(buf, sizeof(buf));
+
+	tls_crypto_info_init(TLS_1_2_VERSION, TLS_CIPHER_AES_GCM_256, &tls12);
+
+	addr.sin_family = AF_INET;
+	addr.sin_addr.s_addr = htonl(INADDR_ANY);
+	addr.sin_port = 0;
+
+	fd = socket(AF_INET, SOCK_STREAM, 0);
+	sfd = socket(AF_INET, SOCK_STREAM, 0);
+
+	ASSERT_EQ(bind(sfd, &addr, sizeof(addr)), 0);
+	ASSERT_EQ(listen(sfd, 10), 0);
+	ASSERT_EQ(getsockname(sfd, &addr, &len), 0);
+	ASSERT_EQ(connect(fd, &addr, sizeof(addr)), 0);
+	ASSERT_GE(cfd = accept(sfd, &addr, &len), 0);
+	close(sfd);
+
+	ret = setsockopt(fd, IPPROTO_TCP, TCP_ULP, "tls", sizeof("tls"));
+	if (ret) {
+		ASSERT_EQ(errno, ENOENT);
+		SKIP(return, "no TLS support");
+	}
+
+	ASSERT_EQ(setsockopt(fd, SOL_TLS, TLS_TX, &tls12, tls12.len), 0);
+	EXPECT_EQ(send(fd, buf, sizeof(buf), MSG_DONTWAIT), sizeof(buf));
+
+	ASSERT_EQ(setsockopt(cfd, IPPROTO_TCP, TCP_ULP, "tls", sizeof("tls")), 0);
+	ASSERT_EQ(setsockopt(cfd, SOL_TLS, TLS_RX, &tls12, tls12.len), 0);
+	EXPECT_EQ(recv(cfd, buf2, sizeof(buf2), MSG_WAITALL), sizeof(buf2));
+
+	EXPECT_EQ(memcmp(buf, buf2, sizeof(buf)), 0);
+
+	close(fd);
+	close(cfd);
+}
+
 TEST_HARNESS_MAIN
-- 
cgit v1.2.3-70-g09d2


From 4c6170d1ae2ccb20ed8d06d191868ed01e6bece0 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Fri, 24 Mar 2023 12:03:56 -0700
Subject: tools: ynl: default to treating enums as flags for mask generation

I was a bit too optimistic in commit bf51d27704c9 ("tools: ynl: fix
get_mask utility routine"), not every mask we use is necessarily
coming from an enum of type "flags". We also allow flipping an
enum into flags on per-attribute basis. That's done by
the 'enum-as-flags' property of an attribute.

Restore this functionality, it's not currently used by any in-tree
family.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/net/ynl/lib/nlspec.py | 8 ++++----
 tools/net/ynl/ynl-gen-c.py  | 3 ++-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index d04450c2a44a..dba70100124a 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -90,8 +90,8 @@ class SpecEnumEntry(SpecElement):
     def raw_value(self):
         return self.value
 
-    def user_value(self):
-        if self.enum_set['type'] == 'flags':
+    def user_value(self, as_flags=None):
+        if self.enum_set['type'] == 'flags' or as_flags:
             return 1 << self.value
         else:
             return self.value
@@ -136,10 +136,10 @@ class SpecEnumSet(SpecElement):
                 return True
         return False
 
-    def get_mask(self):
+    def get_mask(self, as_flags=None):
         mask = 0
         for e in self.entries.values():
-            mask += e.user_value()
+            mask += e.user_value(as_flags)
         return mask
 
 
diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index 972b87c7aaaf..cc2f8c945340 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -254,7 +254,8 @@ class TypeScalar(Type):
     def _attr_policy(self, policy):
         if 'flags-mask' in self.checks or self.is_bitfield:
             if self.is_bitfield:
-                mask = self.family.consts[self.attr['enum']].get_mask()
+                enum = self.family.consts[self.attr['enum']]
+                mask = enum.get_mask(as_flags=True)
             else:
                 flags = self.family.consts[self.checks['flags-mask']]
                 flag_cnt = len(flags['entries'])
-- 
cgit v1.2.3-70-g09d2


From 4228c3a23adc3a4369ecf7329af34ba753be3594 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Fri, 24 Mar 2023 22:23:58 +0100
Subject: net: phy: bcm7xxx: use devm_clk_get_optional_enabled to simplify the
 code

Use devm_clk_get_optional_enabled to simplify the code.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/bcm7xxx.c | 22 ++++------------------
 1 file changed, 4 insertions(+), 18 deletions(-)

diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c
index 75593e7d1118..06be71ecd2f8 100644
--- a/drivers/net/phy/bcm7xxx.c
+++ b/drivers/net/phy/bcm7xxx.c
@@ -45,7 +45,6 @@
 
 struct bcm7xxx_phy_priv {
 	u64	*stats;
-	struct clk *clk;
 };
 
 static int bcm7xxx_28nm_d0_afe_config_init(struct phy_device *phydev)
@@ -811,6 +810,7 @@ static void bcm7xxx_28nm_get_phy_stats(struct phy_device *phydev,
 static int bcm7xxx_28nm_probe(struct phy_device *phydev)
 {
 	struct bcm7xxx_phy_priv *priv;
+	struct clk *clk;
 	int ret = 0;
 
 	priv = devm_kzalloc(&phydev->mdio.dev, sizeof(*priv), GFP_KERNEL);
@@ -825,13 +825,9 @@ static int bcm7xxx_28nm_probe(struct phy_device *phydev)
 	if (!priv->stats)
 		return -ENOMEM;
 
-	priv->clk = devm_clk_get_optional(&phydev->mdio.dev, NULL);
-	if (IS_ERR(priv->clk))
-		return PTR_ERR(priv->clk);
-
-	ret = clk_prepare_enable(priv->clk);
-	if (ret)
-		return ret;
+	clk = devm_clk_get_optional_enabled(&phydev->mdio.dev, NULL);
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
 
 	/* Dummy read to a register to workaround an issue upon reset where the
 	 * internal inverter may not allow the first MDIO transaction to pass
@@ -844,13 +840,6 @@ static int bcm7xxx_28nm_probe(struct phy_device *phydev)
 	return ret;
 }
 
-static void bcm7xxx_28nm_remove(struct phy_device *phydev)
-{
-	struct bcm7xxx_phy_priv *priv = phydev->priv;
-
-	clk_disable_unprepare(priv->clk);
-}
-
 #define BCM7XXX_28NM_GPHY(_oui, _name)					\
 {									\
 	.phy_id		= (_oui),					\
@@ -866,7 +855,6 @@ static void bcm7xxx_28nm_remove(struct phy_device *phydev)
 	.get_strings	= bcm_phy_get_strings,				\
 	.get_stats	= bcm7xxx_28nm_get_phy_stats,			\
 	.probe		= bcm7xxx_28nm_probe,				\
-	.remove		= bcm7xxx_28nm_remove,				\
 }
 
 #define BCM7XXX_28NM_EPHY(_oui, _name)					\
@@ -882,7 +870,6 @@ static void bcm7xxx_28nm_remove(struct phy_device *phydev)
 	.get_strings	= bcm_phy_get_strings,				\
 	.get_stats	= bcm7xxx_28nm_get_phy_stats,			\
 	.probe		= bcm7xxx_28nm_probe,				\
-	.remove		= bcm7xxx_28nm_remove,				\
 	.read_mmd	= bcm7xxx_28nm_ephy_read_mmd,			\
 	.write_mmd	= bcm7xxx_28nm_ephy_write_mmd,			\
 }
@@ -908,7 +895,6 @@ static void bcm7xxx_28nm_remove(struct phy_device *phydev)
 	/* PHY_BASIC_FEATURES */					\
 	.flags		= PHY_IS_INTERNAL,				\
 	.probe		= bcm7xxx_28nm_probe,				\
-	.remove		= bcm7xxx_28nm_remove,				\
 	.config_init	= bcm7xxx_16nm_ephy_config_init,		\
 	.config_aneg	= genphy_config_aneg,				\
 	.read_status	= genphy_read_status,				\
-- 
cgit v1.2.3-70-g09d2


From e5b42483ccce50d5b957f474fd332afd4ef0c27b Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Fri, 24 Mar 2023 23:11:49 +0100
Subject: dev_ioctl: fix a W=1 warning

This fixes the following warning when compiled with GCC 12.2.0 and W=1.

net/core/dev_ioctl.c:475: warning: Function parameter or member 'data'
not described in 'dev_ioctl'

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev_ioctl.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 5cdbfbf9a7dc..846669426236 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -462,6 +462,7 @@ EXPORT_SYMBOL(dev_load);
  *	@net: the applicable net namespace
  *	@cmd: command to issue
  *	@ifr: pointer to a struct ifreq in user space
+ *	@data: data exchanged with userspace
  *	@need_copyout: whether or not copy_to_user() should be called
  *
  *	Issue ioctl functions to devices. This is normally called by the
-- 
cgit v1.2.3-70-g09d2


From f1cb927cdb6280e0ce283174cc1bf395dffc43d4 Mon Sep 17 00:00:00 2001
From: JP Kobryn <inwardvessel@gmail.com>
Date: Fri, 24 Mar 2023 18:08:45 -0700
Subject: libbpf: Ensure print callback usage is thread-safe

This patch prevents races on the print function pointer, allowing the
libbpf_set_print() function to become thread-safe.

Signed-off-by: JP Kobryn <inwardvessel@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230325010845.46000-1-inwardvessel@gmail.com
---
 tools/lib/bpf/libbpf.c | 9 ++++++---
 tools/lib/bpf/libbpf.h | 2 ++
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index f6a071db5c6e..15737d7b5a28 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -216,9 +216,10 @@ static libbpf_print_fn_t __libbpf_pr = __base_pr;
 
 libbpf_print_fn_t libbpf_set_print(libbpf_print_fn_t fn)
 {
-	libbpf_print_fn_t old_print_fn = __libbpf_pr;
+	libbpf_print_fn_t old_print_fn;
+
+	old_print_fn = __atomic_exchange_n(&__libbpf_pr, fn, __ATOMIC_RELAXED);
 
-	__libbpf_pr = fn;
 	return old_print_fn;
 }
 
@@ -227,8 +228,10 @@ void libbpf_print(enum libbpf_print_level level, const char *format, ...)
 {
 	va_list args;
 	int old_errno;
+	libbpf_print_fn_t print_fn;
 
-	if (!__libbpf_pr)
+	print_fn = __atomic_load_n(&__libbpf_pr, __ATOMIC_RELAXED);
+	if (!print_fn)
 		return;
 
 	old_errno = errno;
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 1615e55e2e79..0b7362397ea3 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -101,6 +101,8 @@ typedef int (*libbpf_print_fn_t)(enum libbpf_print_level level,
  * be used for libbpf warnings and informational messages.
  * @param fn The log print function. If NULL, libbpf won't print anything.
  * @return Pointer to old print function.
+ *
+ * This function is thread-safe.
  */
 LIBBPF_API libbpf_print_fn_t libbpf_set_print(libbpf_print_fn_t fn);
 
-- 
cgit v1.2.3-70-g09d2


From 7283137a7622292076dd8b7f3b8b2bb203ce5a14 Mon Sep 17 00:00:00 2001
From: Hengqi Chen <hengqi.chen@gmail.com>
Date: Sun, 26 Mar 2023 09:53:41 +0000
Subject: selftests/bpf: Don't assume page size is 4096

The verifier test creates BPF ringbuf maps using hard-coded
4096 as max_entries. Some tests will fail if the page size
of the running kernel is not 4096. Use getpagesize() instead.

Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230326095341.816023-1-hengqi.chen@gmail.com
---
 tools/testing/selftests/bpf/test_verifier.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c
index 5b90eef09ade..e4657c5bc3f1 100644
--- a/tools/testing/selftests/bpf/test_verifier.c
+++ b/tools/testing/selftests/bpf/test_verifier.c
@@ -1079,7 +1079,7 @@ static void do_test_fixup(struct bpf_test *test, enum bpf_prog_type prog_type,
 	}
 	if (*fixup_map_ringbuf) {
 		map_fds[20] = create_map(BPF_MAP_TYPE_RINGBUF, 0,
-					   0, 4096);
+					 0, getpagesize());
 		do {
 			prog[*fixup_map_ringbuf].imm = map_fds[20];
 			fixup_map_ringbuf++;
-- 
cgit v1.2.3-70-g09d2


From 3948b05950fdd64002a5f182c65ba5cf2d53cf71 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 23 Mar 2023 09:28:42 -0700
Subject: net: introduce a config option to tweak MAX_SKB_FRAGS

Currently, MAX_SKB_FRAGS value is 17.

For standard tcp sendmsg() traffic, no big deal because tcp_sendmsg()
attempts order-3 allocations, stuffing 32768 bytes per frag.

But with zero copy, we use order-0 pages.

For BIG TCP to show its full potential, we add a config option
to be able to fit up to 45 segments per skb.

This is also needed for BIG TCP rx zerocopy, as zerocopy currently
does not support skbs with frag list.

We have used MAX_SKB_FRAGS=45 value for years at Google before
we deployed 4K MTU, with no adverse effect, other than
a recent issue in mlx4, fixed in commit 26782aad00cc
("net/mlx4: MLX4_TX_BOUNCE_BUFFER_SIZE depends on MAX_SKB_FRAGS")

Back then, goal was to be able to receive full size (64KB) GRO
packets without the frag_list overhead.

Note that /proc/sys/net/core/max_skb_frags can also be used to limit
the number of fragments TCP can use in tx packets.

By default we keep the old/legacy value of 17 until we get
more coverage for the updated values.

Sizes of struct skb_shared_info on 64bit arches

MAX_SKB_FRAGS | sizeof(struct skb_shared_info):
==============================================
         17     320
         21     320+64  = 384
         25     320+128 = 448
         29     320+192 = 512
         33     320+256 = 576
         37     320+320 = 640
         41     320+384 = 704
         45     320+448 = 768

This inflation might cause problems for drivers assuming they could pack
both the incoming packet (for MTU=1500) and skb_shared_info in half a page,
using build_skb().

v3: fix build error when CONFIG_NET=n
v2: fix two build errors assuming MAX_SKB_FRAGS was "unsigned long"

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Link: https://lore.kernel.org/r/20230323162842.1935061-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/scsi/cxgbi/libcxgbi.c |  4 ++--
 include/linux/skbuff.h        | 16 +++++-----------
 net/Kconfig                   | 12 ++++++++++++
 net/packet/af_packet.c        |  4 ++--
 4 files changed, 21 insertions(+), 15 deletions(-)

diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c
index af281e271f88..3e1de4c842cc 100644
--- a/drivers/scsi/cxgbi/libcxgbi.c
+++ b/drivers/scsi/cxgbi/libcxgbi.c
@@ -2314,9 +2314,9 @@ static int cxgbi_sock_tx_queue_up(struct cxgbi_sock *csk, struct sk_buff *skb)
 		frags++;
 
 	if (frags >= SKB_WR_LIST_SIZE) {
-		pr_err("csk 0x%p, frags %u, %u,%u >%lu.\n",
+		pr_err("csk 0x%p, frags %u, %u,%u >%u.\n",
 		       csk, skb_shinfo(skb)->nr_frags, skb->len,
-		       skb->data_len, SKB_WR_LIST_SIZE);
+		       skb->data_len, (unsigned int)SKB_WR_LIST_SIZE);
 		return -EINVAL;
 	}
 
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index fe661011644b..82511b2f61ea 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -345,18 +345,12 @@ struct sk_buff_head {
 
 struct sk_buff;
 
-/* To allow 64K frame to be packed as single skb without frag_list we
- * require 64K/PAGE_SIZE pages plus 1 additional page to allow for
- * buffers which do not start on a page boundary.
- *
- * Since GRO uses frags we allocate at least 16 regardless of page
- * size.
- */
-#if (65536/PAGE_SIZE + 1) < 16
-#define MAX_SKB_FRAGS 16UL
-#else
-#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 1)
+#ifndef CONFIG_MAX_SKB_FRAGS
+# define CONFIG_MAX_SKB_FRAGS 17
 #endif
+
+#define MAX_SKB_FRAGS CONFIG_MAX_SKB_FRAGS
+
 extern int sysctl_max_skb_frags;
 
 /* Set skb_shinfo(skb)->gso_size to this in case you want skb_segment to
diff --git a/net/Kconfig b/net/Kconfig
index 48c33c222199..f806722bccf4 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -251,6 +251,18 @@ config PCPU_DEV_REFCNT
 	  network device refcount are using per cpu variables if this option is set.
 	  This can be forced to N to detect underflows (with a performance drop).
 
+config MAX_SKB_FRAGS
+	int "Maximum number of fragments per skb_shared_info"
+	range 17 45
+	default 17
+	help
+	  Having more fragments per skb_shared_info can help GRO efficiency.
+	  This helps BIG TCP workloads, but might expose bugs in some
+	  legacy drivers.
+	  This also increases memory overhead of small packets,
+	  and in drivers using build_skb().
+	  If unsure, say 17.
+
 config RPS
 	bool
 	depends on SMP && SYSFS
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 497193f73030..568f8d76e3c1 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2622,8 +2622,8 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 		nr_frags = skb_shinfo(skb)->nr_frags;
 
 		if (unlikely(nr_frags >= MAX_SKB_FRAGS)) {
-			pr_err("Packet exceed the number of skb frags(%lu)\n",
-			       MAX_SKB_FRAGS);
+			pr_err("Packet exceed the number of skb frags(%u)\n",
+			       (unsigned int)MAX_SKB_FRAGS);
 			return -EFAULT;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 2bcc74ffd21a748fb310135175a1f6b271397e0f Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Sat, 25 Mar 2023 20:17:33 -0400
Subject: qed: remove unused num_ooo_add_to_peninsula variable

clang with W=1 reports
drivers/net/ethernet/qlogic/qed/qed_ll2.c:649:6: error: variable
  'num_ooo_add_to_peninsula' set but not used [-Werror,-Wunused-but-set-variable]
        u32 num_ooo_add_to_peninsula = 0, cid;
            ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230326001733.1343274-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/qed/qed_ll2.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.c b/drivers/net/ethernet/qlogic/qed/qed_ll2.c
index e5116a86cfbc..717a0b3f89bd 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_ll2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.c
@@ -646,13 +646,13 @@ static int qed_ll2_lb_rxq_handler(struct qed_hwfn *p_hwfn,
 	struct qed_ll2_rx_queue *p_rx = &p_ll2_conn->rx_queue;
 	u16 packet_length = 0, parse_flags = 0, vlan = 0;
 	struct qed_ll2_rx_packet *p_pkt = NULL;
-	u32 num_ooo_add_to_peninsula = 0, cid;
 	union core_rx_cqe_union *cqe = NULL;
 	u16 cq_new_idx = 0, cq_old_idx = 0;
 	struct qed_ooo_buffer *p_buffer;
 	struct ooo_opaque *ooo_opq;
 	u8 placement_offset = 0;
 	u8 cqe_type;
+	u32 cid;
 
 	cq_new_idx = le16_to_cpu(*p_rx->p_fw_cons);
 	cq_old_idx = qed_chain_get_cons_idx(&p_rx->rcq_chain);
@@ -762,7 +762,6 @@ static int qed_ll2_lb_rxq_handler(struct qed_hwfn *p_hwfn,
 						   cid, ooo_opq->ooo_isle);
 				break;
 			case TCP_EVENT_ADD_PEN:
-				num_ooo_add_to_peninsula++;
 				qed_ooo_put_ready_buffer(p_hwfn,
 							 p_hwfn->p_ooo_info,
 							 p_buffer, true);
-- 
cgit v1.2.3-70-g09d2


From 3e4d5ba9a3f85f21f1ebdee5a5901bb43389abc5 Mon Sep 17 00:00:00 2001
From: Shay Agroskin <shayagr@amazon.com>
Date: Thu, 23 Mar 2023 18:36:04 +0200
Subject: netlink: Add a macro to set policy message with format string

Similar to NL_SET_ERR_MSG_FMT, add a macro which sets netlink policy
error message with a format string.

Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netlink.h | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 3e8743252167..19c0791ed9d5 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -161,9 +161,31 @@ struct netlink_ext_ack {
 	}							\
 } while (0)
 
+#define NL_SET_ERR_MSG_ATTR_POL_FMT(extack, attr, pol, fmt, args...) do {	\
+	struct netlink_ext_ack *__extack = (extack);				\
+										\
+	if (!__extack)								\
+		break;								\
+										\
+	if (snprintf(__extack->_msg_buf, NETLINK_MAX_FMTMSG_LEN,		\
+		     "%s" fmt "%s", "", ##args, "") >=				\
+	    NETLINK_MAX_FMTMSG_LEN)						\
+		net_warn_ratelimited("%s" fmt "%s", "truncated extack: ",       \
+				     ##args, "\n");				\
+										\
+	do_trace_netlink_extack(__extack->_msg_buf);				\
+										\
+	__extack->_msg = __extack->_msg_buf;					\
+	__extack->bad_attr = (attr);						\
+	__extack->policy = (pol);						\
+} while (0)
+
 #define NL_SET_ERR_MSG_ATTR(extack, attr, msg)		\
 	NL_SET_ERR_MSG_ATTR_POL(extack, attr, NULL, msg)
 
+#define NL_SET_ERR_MSG_ATTR_FMT(extack, attr, msg, args...) \
+	NL_SET_ERR_MSG_ATTR_POL_FMT(extack, attr, NULL, msg, ##args)
+
 #define NL_SET_ERR_ATTR_MISS(extack, nest, type)  do {	\
 	struct netlink_ext_ack *__extack = (extack);	\
 							\
-- 
cgit v1.2.3-70-g09d2


From 233eb4e786b57ea686b51c13a04cc2839fd682fc Mon Sep 17 00:00:00 2001
From: Shay Agroskin <shayagr@amazon.com>
Date: Thu, 23 Mar 2023 18:36:05 +0200
Subject: ethtool: Add support for configuring tx_push_buf_len

This attribute, which is part of ethtool's ring param configuration
allows the user to specify the maximum number of the packet's payload
that can be written directly to the device.

Example usage:
    # ethtool -G [interface] tx-push-buf-len [number of bytes]

Co-developed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ethtool.yaml     |  8 +++++
 Documentation/networking/ethtool-netlink.rst | 47 ++++++++++++++++++----------
 include/linux/ethtool.h                      | 14 ++++++---
 include/uapi/linux/ethtool_netlink.h         |  2 ++
 net/ethtool/netlink.h                        |  2 +-
 net/ethtool/rings.c                          | 34 ++++++++++++++++++--
 6 files changed, 84 insertions(+), 23 deletions(-)

diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 4727c067e2ba..6d8ae3d9a680 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -165,6 +165,12 @@ attribute-sets:
       -
         name: rx-push
         type: u8
+      -
+        name: tx-push-buf-len
+        type: u32
+      -
+        name: tx-push-buf-len-max
+        type: u32
 
   -
     name: mm-stat
@@ -311,6 +317,8 @@ operations:
             - cqe-size
             - tx-push
             - rx-push
+            - tx-push-buf-len
+            - tx-push-buf-len-max
       dump: *ring-get-op
     -
       name: rings-set
diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index e1bc6186d7ea..cd0973d4ba01 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -860,22 +860,24 @@ Request contents:
 
 Kernel response contents:
 
-  ====================================  ======  ===========================
-  ``ETHTOOL_A_RINGS_HEADER``            nested  reply header
-  ``ETHTOOL_A_RINGS_RX_MAX``            u32     max size of RX ring
-  ``ETHTOOL_A_RINGS_RX_MINI_MAX``       u32     max size of RX mini ring
-  ``ETHTOOL_A_RINGS_RX_JUMBO_MAX``      u32     max size of RX jumbo ring
-  ``ETHTOOL_A_RINGS_TX_MAX``            u32     max size of TX ring
-  ``ETHTOOL_A_RINGS_RX``                u32     size of RX ring
-  ``ETHTOOL_A_RINGS_RX_MINI``           u32     size of RX mini ring
-  ``ETHTOOL_A_RINGS_RX_JUMBO``          u32     size of RX jumbo ring
-  ``ETHTOOL_A_RINGS_TX``                u32     size of TX ring
-  ``ETHTOOL_A_RINGS_RX_BUF_LEN``        u32     size of buffers on the ring
-  ``ETHTOOL_A_RINGS_TCP_DATA_SPLIT``    u8      TCP header / data split
-  ``ETHTOOL_A_RINGS_CQE_SIZE``          u32     Size of TX/RX CQE
-  ``ETHTOOL_A_RINGS_TX_PUSH``           u8      flag of TX Push mode
-  ``ETHTOOL_A_RINGS_RX_PUSH``           u8      flag of RX Push mode
-  ====================================  ======  ===========================
+  =======================================   ======  ===========================
+  ``ETHTOOL_A_RINGS_HEADER``                nested  reply header
+  ``ETHTOOL_A_RINGS_RX_MAX``                u32     max size of RX ring
+  ``ETHTOOL_A_RINGS_RX_MINI_MAX``           u32     max size of RX mini ring
+  ``ETHTOOL_A_RINGS_RX_JUMBO_MAX``          u32     max size of RX jumbo ring
+  ``ETHTOOL_A_RINGS_TX_MAX``                u32     max size of TX ring
+  ``ETHTOOL_A_RINGS_RX``                    u32     size of RX ring
+  ``ETHTOOL_A_RINGS_RX_MINI``               u32     size of RX mini ring
+  ``ETHTOOL_A_RINGS_RX_JUMBO``              u32     size of RX jumbo ring
+  ``ETHTOOL_A_RINGS_TX``                    u32     size of TX ring
+  ``ETHTOOL_A_RINGS_RX_BUF_LEN``            u32     size of buffers on the ring
+  ``ETHTOOL_A_RINGS_TCP_DATA_SPLIT``        u8      TCP header / data split
+  ``ETHTOOL_A_RINGS_CQE_SIZE``              u32     Size of TX/RX CQE
+  ``ETHTOOL_A_RINGS_TX_PUSH``               u8      flag of TX Push mode
+  ``ETHTOOL_A_RINGS_RX_PUSH``               u8      flag of RX Push mode
+  ``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN``       u32     size of TX push buffer
+  ``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX``   u32     max size of TX push buffer
+  =======================================   ======  ===========================
 
 ``ETHTOOL_A_RINGS_TCP_DATA_SPLIT`` indicates whether the device is usable with
 page-flipping TCP zero-copy receive (``getsockopt(TCP_ZEROCOPY_RECEIVE)``).
@@ -891,6 +893,18 @@ through MMIO writes, thus reducing the latency. However, enabling this feature
 may increase the CPU cost. Drivers may enforce additional per-packet
 eligibility checks (e.g. on packet size).
 
+``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN`` specifies the maximum number of bytes of a
+transmitted packet a driver can push directly to the underlying device
+('push' mode). Pushing some of the payload bytes to the device has the
+advantages of reducing latency for small packets by avoiding DMA mapping (same
+as ``ETHTOOL_A_RINGS_TX_PUSH`` parameter) as well as allowing the underlying
+device to process packet headers ahead of fetching its payload.
+This can help the device to make fast actions based on the packet's headers.
+This is similar to the "tx-copybreak" parameter, which copies the packet to a
+preallocated DMA memory area instead of mapping new memory. However,
+tx-push-buff parameter copies the packet directly to the device to allow the
+device to take faster actions on the packet.
+
 RINGS_SET
 =========
 
@@ -908,6 +922,7 @@ Request contents:
   ``ETHTOOL_A_RINGS_CQE_SIZE``          u32     Size of TX/RX CQE
   ``ETHTOOL_A_RINGS_TX_PUSH``           u8      flag of TX Push mode
   ``ETHTOOL_A_RINGS_RX_PUSH``           u8      flag of RX Push mode
+  ``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN``   u32     size of TX push buffer
   ====================================  ======  ===========================
 
 Kernel checks that requested ring sizes do not exceed limits reported by
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 2792185dda22..798d35890118 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -75,6 +75,8 @@ enum {
  * @tx_push: The flag of tx push mode
  * @rx_push: The flag of rx push mode
  * @cqe_size: Size of TX/RX completion queue event
+ * @tx_push_buf_len: Size of TX push buffer
+ * @tx_push_buf_max_len: Maximum allowed size of TX push buffer
  */
 struct kernel_ethtool_ringparam {
 	u32	rx_buf_len;
@@ -82,6 +84,8 @@ struct kernel_ethtool_ringparam {
 	u8	tx_push;
 	u8	rx_push;
 	u32	cqe_size;
+	u32	tx_push_buf_len;
+	u32	tx_push_buf_max_len;
 };
 
 /**
@@ -90,12 +94,14 @@ struct kernel_ethtool_ringparam {
  * @ETHTOOL_RING_USE_CQE_SIZE: capture for setting cqe_size
  * @ETHTOOL_RING_USE_TX_PUSH: capture for setting tx_push
  * @ETHTOOL_RING_USE_RX_PUSH: capture for setting rx_push
+ * @ETHTOOL_RING_USE_TX_PUSH_BUF_LEN: capture for setting tx_push_buf_len
  */
 enum ethtool_supported_ring_param {
-	ETHTOOL_RING_USE_RX_BUF_LEN = BIT(0),
-	ETHTOOL_RING_USE_CQE_SIZE   = BIT(1),
-	ETHTOOL_RING_USE_TX_PUSH    = BIT(2),
-	ETHTOOL_RING_USE_RX_PUSH    = BIT(3),
+	ETHTOOL_RING_USE_RX_BUF_LEN		= BIT(0),
+	ETHTOOL_RING_USE_CQE_SIZE		= BIT(1),
+	ETHTOOL_RING_USE_TX_PUSH		= BIT(2),
+	ETHTOOL_RING_USE_RX_PUSH		= BIT(3),
+	ETHTOOL_RING_USE_TX_PUSH_BUF_LEN	= BIT(4),
 };
 
 #define __ETH_RSS_HASH_BIT(bit)	((u32)1 << (bit))
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index d39ce21381c5..1ebf8d455f07 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -357,6 +357,8 @@ enum {
 	ETHTOOL_A_RINGS_CQE_SIZE,			/* u32 */
 	ETHTOOL_A_RINGS_TX_PUSH,			/* u8 */
 	ETHTOOL_A_RINGS_RX_PUSH,			/* u8 */
+	ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN,		/* u32 */
+	ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX,		/* u32 */
 
 	/* add new constants above here */
 	__ETHTOOL_A_RINGS_CNT,
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index f7b189ed96b2..79424b34b553 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -413,7 +413,7 @@ extern const struct nla_policy ethnl_features_set_policy[ETHTOOL_A_FEATURES_WANT
 extern const struct nla_policy ethnl_privflags_get_policy[ETHTOOL_A_PRIVFLAGS_HEADER + 1];
 extern const struct nla_policy ethnl_privflags_set_policy[ETHTOOL_A_PRIVFLAGS_FLAGS + 1];
 extern const struct nla_policy ethnl_rings_get_policy[ETHTOOL_A_RINGS_HEADER + 1];
-extern const struct nla_policy ethnl_rings_set_policy[ETHTOOL_A_RINGS_RX_PUSH + 1];
+extern const struct nla_policy ethnl_rings_set_policy[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX + 1];
 extern const struct nla_policy ethnl_channels_get_policy[ETHTOOL_A_CHANNELS_HEADER + 1];
 extern const struct nla_policy ethnl_channels_set_policy[ETHTOOL_A_CHANNELS_COMBINED_COUNT + 1];
 extern const struct nla_policy ethnl_coalesce_get_policy[ETHTOOL_A_COALESCE_HEADER + 1];
diff --git a/net/ethtool/rings.c b/net/ethtool/rings.c
index f358cd57d094..1c4972526142 100644
--- a/net/ethtool/rings.c
+++ b/net/ethtool/rings.c
@@ -11,6 +11,7 @@ struct rings_reply_data {
 	struct ethnl_reply_data		base;
 	struct ethtool_ringparam	ringparam;
 	struct kernel_ethtool_ringparam	kernel_ringparam;
+	u32				supported_ring_params;
 };
 
 #define RINGS_REPDATA(__reply_base) \
@@ -32,6 +33,8 @@ static int rings_prepare_data(const struct ethnl_req_info *req_base,
 
 	if (!dev->ethtool_ops->get_ringparam)
 		return -EOPNOTSUPP;
+
+	data->supported_ring_params = dev->ethtool_ops->supported_ring_params;
 	ret = ethnl_ops_begin(dev);
 	if (ret < 0)
 		return ret;
@@ -57,7 +60,9 @@ static int rings_reply_size(const struct ethnl_req_info *req_base,
 	       nla_total_size(sizeof(u8))  +	/* _RINGS_TCP_DATA_SPLIT */
 	       nla_total_size(sizeof(u32)  +	/* _RINGS_CQE_SIZE */
 	       nla_total_size(sizeof(u8))  +	/* _RINGS_TX_PUSH */
-	       nla_total_size(sizeof(u8)));	/* _RINGS_RX_PUSH */
+	       nla_total_size(sizeof(u8))) +	/* _RINGS_RX_PUSH */
+	       nla_total_size(sizeof(u32)) +	/* _RINGS_TX_PUSH_BUF_LEN */
+	       nla_total_size(sizeof(u32));	/* _RINGS_TX_PUSH_BUF_LEN_MAX */
 }
 
 static int rings_fill_reply(struct sk_buff *skb,
@@ -67,6 +72,7 @@ static int rings_fill_reply(struct sk_buff *skb,
 	const struct rings_reply_data *data = RINGS_REPDATA(reply_base);
 	const struct kernel_ethtool_ringparam *kr = &data->kernel_ringparam;
 	const struct ethtool_ringparam *ringparam = &data->ringparam;
+	u32 supported_ring_params = data->supported_ring_params;
 
 	WARN_ON(kr->tcp_data_split > ETHTOOL_TCP_DATA_SPLIT_ENABLED);
 
@@ -98,7 +104,12 @@ static int rings_fill_reply(struct sk_buff *skb,
 	    (kr->cqe_size &&
 	     (nla_put_u32(skb, ETHTOOL_A_RINGS_CQE_SIZE, kr->cqe_size))) ||
 	    nla_put_u8(skb, ETHTOOL_A_RINGS_TX_PUSH, !!kr->tx_push) ||
-	    nla_put_u8(skb, ETHTOOL_A_RINGS_RX_PUSH, !!kr->rx_push))
+	    nla_put_u8(skb, ETHTOOL_A_RINGS_RX_PUSH, !!kr->rx_push) ||
+	    ((supported_ring_params & ETHTOOL_RING_USE_TX_PUSH_BUF_LEN) &&
+	     (nla_put_u32(skb, ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX,
+			  kr->tx_push_buf_max_len) ||
+	      nla_put_u32(skb, ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN,
+			  kr->tx_push_buf_len))))
 		return -EMSGSIZE;
 
 	return 0;
@@ -117,6 +128,7 @@ const struct nla_policy ethnl_rings_set_policy[] = {
 	[ETHTOOL_A_RINGS_CQE_SIZE]		= NLA_POLICY_MIN(NLA_U32, 1),
 	[ETHTOOL_A_RINGS_TX_PUSH]		= NLA_POLICY_MAX(NLA_U8, 1),
 	[ETHTOOL_A_RINGS_RX_PUSH]		= NLA_POLICY_MAX(NLA_U8, 1),
+	[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN]	= { .type = NLA_U32 },
 };
 
 static int
@@ -158,6 +170,14 @@ ethnl_set_rings_validate(struct ethnl_req_info *req_info,
 		return -EOPNOTSUPP;
 	}
 
+	if (tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN] &&
+	    !(ops->supported_ring_params & ETHTOOL_RING_USE_TX_PUSH_BUF_LEN)) {
+		NL_SET_ERR_MSG_ATTR(info->extack,
+				    tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN],
+				    "setting tx push buf len is not supported");
+		return -EOPNOTSUPP;
+	}
+
 	return ops->get_ringparam && ops->set_ringparam ? 1 : -EOPNOTSUPP;
 }
 
@@ -189,6 +209,8 @@ ethnl_set_rings(struct ethnl_req_info *req_info, struct genl_info *info)
 			tb[ETHTOOL_A_RINGS_TX_PUSH], &mod);
 	ethnl_update_u8(&kernel_ringparam.rx_push,
 			tb[ETHTOOL_A_RINGS_RX_PUSH], &mod);
+	ethnl_update_u32(&kernel_ringparam.tx_push_buf_len,
+			 tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN], &mod);
 	if (!mod)
 		return 0;
 
@@ -209,6 +231,14 @@ ethnl_set_rings(struct ethnl_req_info *req_info, struct genl_info *info)
 		return -EINVAL;
 	}
 
+	if (kernel_ringparam.tx_push_buf_len > kernel_ringparam.tx_push_buf_max_len) {
+		NL_SET_ERR_MSG_ATTR_FMT(info->extack, tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN],
+					"Requested TX push buffer exceeds the maximum of %u",
+					kernel_ringparam.tx_push_buf_max_len);
+
+		return -EINVAL;
+	}
+
 	ret = dev->ethtool_ops->set_ringparam(dev, &ringparam,
 					      &kernel_ringparam, info->extack);
 	return ret < 0 ? ret : 1;
-- 
cgit v1.2.3-70-g09d2


From 3a09108446b674a544a736318a7916c5bcea3f14 Mon Sep 17 00:00:00 2001
From: Shay Agroskin <shayagr@amazon.com>
Date: Thu, 23 Mar 2023 18:36:06 +0200
Subject: net: ena: Make few cosmetic preparations to support large LLQ

Move ena_calc_io_queue_size() implementation closer to the file's
beginning so that it can be later called from ena_device_init()
function without adding a function declaration.

Also add an empty line at some spots to separate logical blocks in
funcitons.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 133 ++++++++++++++-------------
 1 file changed, 67 insertions(+), 66 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index cbfe7f977270..794f8eec468a 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -3364,6 +3364,71 @@ static const struct net_device_ops ena_netdev_ops = {
 	.ndo_xdp_xmit		= ena_xdp_xmit,
 };
 
+static void ena_calc_io_queue_size(struct ena_adapter *adapter,
+				   struct ena_com_dev_get_features_ctx *get_feat_ctx)
+{
+	struct ena_admin_feature_llq_desc *llq = &get_feat_ctx->llq;
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+	u32 tx_queue_size = ENA_DEFAULT_RING_SIZE;
+	u32 rx_queue_size = ENA_DEFAULT_RING_SIZE;
+	u32 max_tx_queue_size;
+	u32 max_rx_queue_size;
+
+	if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) {
+		struct ena_admin_queue_ext_feature_fields *max_queue_ext =
+			&get_feat_ctx->max_queue_ext.max_queue_ext;
+		max_rx_queue_size = min_t(u32, max_queue_ext->max_rx_cq_depth,
+					  max_queue_ext->max_rx_sq_depth);
+		max_tx_queue_size = max_queue_ext->max_tx_cq_depth;
+
+		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  llq->max_llq_depth);
+		else
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  max_queue_ext->max_tx_sq_depth);
+
+		adapter->max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queue_ext->max_per_packet_tx_descs);
+		adapter->max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queue_ext->max_per_packet_rx_descs);
+	} else {
+		struct ena_admin_queue_feature_desc *max_queues =
+			&get_feat_ctx->max_queues;
+		max_rx_queue_size = min_t(u32, max_queues->max_cq_depth,
+					  max_queues->max_sq_depth);
+		max_tx_queue_size = max_queues->max_cq_depth;
+
+		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  llq->max_llq_depth);
+		else
+			max_tx_queue_size = min_t(u32, max_tx_queue_size,
+						  max_queues->max_sq_depth);
+
+		adapter->max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queues->max_packet_tx_descs);
+		adapter->max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
+						 max_queues->max_packet_rx_descs);
+	}
+
+	max_tx_queue_size = rounddown_pow_of_two(max_tx_queue_size);
+	max_rx_queue_size = rounddown_pow_of_two(max_rx_queue_size);
+
+	tx_queue_size = clamp_val(tx_queue_size, ENA_MIN_RING_SIZE,
+				  max_tx_queue_size);
+	rx_queue_size = clamp_val(rx_queue_size, ENA_MIN_RING_SIZE,
+				  max_rx_queue_size);
+
+	tx_queue_size = rounddown_pow_of_two(tx_queue_size);
+	rx_queue_size = rounddown_pow_of_two(rx_queue_size);
+
+	adapter->max_tx_ring_size  = max_tx_queue_size;
+	adapter->max_rx_ring_size = max_rx_queue_size;
+	adapter->requested_tx_ring_size = tx_queue_size;
+	adapter->requested_rx_ring_size = rx_queue_size;
+}
+
 static int ena_device_validate_params(struct ena_adapter *adapter,
 				      struct ena_com_dev_get_features_ctx *get_feat_ctx)
 {
@@ -4162,72 +4227,6 @@ static void ena_release_bars(struct ena_com_dev *ena_dev, struct pci_dev *pdev)
 	pci_release_selected_regions(pdev, release_bars);
 }
 
-
-static void ena_calc_io_queue_size(struct ena_adapter *adapter,
-				   struct ena_com_dev_get_features_ctx *get_feat_ctx)
-{
-	struct ena_admin_feature_llq_desc *llq = &get_feat_ctx->llq;
-	struct ena_com_dev *ena_dev = adapter->ena_dev;
-	u32 tx_queue_size = ENA_DEFAULT_RING_SIZE;
-	u32 rx_queue_size = ENA_DEFAULT_RING_SIZE;
-	u32 max_tx_queue_size;
-	u32 max_rx_queue_size;
-
-	if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) {
-		struct ena_admin_queue_ext_feature_fields *max_queue_ext =
-			&get_feat_ctx->max_queue_ext.max_queue_ext;
-		max_rx_queue_size = min_t(u32, max_queue_ext->max_rx_cq_depth,
-					  max_queue_ext->max_rx_sq_depth);
-		max_tx_queue_size = max_queue_ext->max_tx_cq_depth;
-
-		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
-			max_tx_queue_size = min_t(u32, max_tx_queue_size,
-						  llq->max_llq_depth);
-		else
-			max_tx_queue_size = min_t(u32, max_tx_queue_size,
-						  max_queue_ext->max_tx_sq_depth);
-
-		adapter->max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
-						 max_queue_ext->max_per_packet_tx_descs);
-		adapter->max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
-						 max_queue_ext->max_per_packet_rx_descs);
-	} else {
-		struct ena_admin_queue_feature_desc *max_queues =
-			&get_feat_ctx->max_queues;
-		max_rx_queue_size = min_t(u32, max_queues->max_cq_depth,
-					  max_queues->max_sq_depth);
-		max_tx_queue_size = max_queues->max_cq_depth;
-
-		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV)
-			max_tx_queue_size = min_t(u32, max_tx_queue_size,
-						  llq->max_llq_depth);
-		else
-			max_tx_queue_size = min_t(u32, max_tx_queue_size,
-						  max_queues->max_sq_depth);
-
-		adapter->max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
-						 max_queues->max_packet_tx_descs);
-		adapter->max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS,
-						 max_queues->max_packet_rx_descs);
-	}
-
-	max_tx_queue_size = rounddown_pow_of_two(max_tx_queue_size);
-	max_rx_queue_size = rounddown_pow_of_two(max_rx_queue_size);
-
-	tx_queue_size = clamp_val(tx_queue_size, ENA_MIN_RING_SIZE,
-				  max_tx_queue_size);
-	rx_queue_size = clamp_val(rx_queue_size, ENA_MIN_RING_SIZE,
-				  max_rx_queue_size);
-
-	tx_queue_size = rounddown_pow_of_two(tx_queue_size);
-	rx_queue_size = rounddown_pow_of_two(rx_queue_size);
-
-	adapter->max_tx_ring_size  = max_tx_queue_size;
-	adapter->max_rx_ring_size = max_rx_queue_size;
-	adapter->requested_tx_ring_size = tx_queue_size;
-	adapter->requested_rx_ring_size = rx_queue_size;
-}
-
 /* ena_probe - Device Initialization Routine
  * @pdev: PCI device information struct
  * @ent: entry in ena_pci_tbl
@@ -4364,6 +4363,7 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			"Failed to query interrupt moderation feature\n");
 		goto err_device_destroy;
 	}
+
 	ena_init_io_rings(adapter,
 			  0,
 			  adapter->xdp_num_queues +
@@ -4488,6 +4488,7 @@ static void __ena_shutoff(struct pci_dev *pdev, bool shutdown)
 	rtnl_lock(); /* lock released inside the below if-else block */
 	adapter->reset_reason = ENA_REGS_RESET_SHUTDOWN;
 	ena_destroy_device(adapter, true);
+
 	if (shutdown) {
 		netif_device_detach(netdev);
 		dev_close(netdev);
-- 
cgit v1.2.3-70-g09d2


From 1e36668886882d51641bd57ae659e678b3f1ca9c Mon Sep 17 00:00:00 2001
From: David Arinzon <darinzon@amazon.com>
Date: Thu, 23 Mar 2023 18:36:07 +0200
Subject: net: ena: Add an option to configure large LLQ headers

Allow configuring the device with large LLQ headers. The Low Latency
Queue (LLQ) allows the driver to write the first N bytes of the packet,
along with the rest of the TX descriptors directly into device (N can be
either 96 or 224 for large LLQ headers configuration).

Having L4 TCP/UDP headers contained in the first 96 bytes of the packet
is required to get maximum performance from the device.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Signed-off-by: David Arinzon <darinzon@amazon.com>
Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 90 +++++++++++++++++++++-------
 drivers/net/ethernet/amazon/ena/ena_netdev.h |  8 +++
 2 files changed, 76 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 794f8eec468a..97da4d841df3 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -3374,6 +3374,15 @@ static void ena_calc_io_queue_size(struct ena_adapter *adapter,
 	u32 max_tx_queue_size;
 	u32 max_rx_queue_size;
 
+	/* If this function is called after driver load, the ring sizes have already
+	 * been configured. Take it into account when recalculating ring size.
+	 */
+	if (adapter->tx_ring->ring_size)
+		tx_queue_size = adapter->tx_ring->ring_size;
+
+	if (adapter->rx_ring->ring_size)
+		rx_queue_size = adapter->rx_ring->ring_size;
+
 	if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) {
 		struct ena_admin_queue_ext_feature_fields *max_queue_ext =
 			&get_feat_ctx->max_queue_ext.max_queue_ext;
@@ -3415,6 +3424,24 @@ static void ena_calc_io_queue_size(struct ena_adapter *adapter,
 	max_tx_queue_size = rounddown_pow_of_two(max_tx_queue_size);
 	max_rx_queue_size = rounddown_pow_of_two(max_rx_queue_size);
 
+	/* When forcing large headers, we multiply the entry size by 2, and therefore divide
+	 * the queue size by 2, leaving the amount of memory used by the queues unchanged.
+	 */
+	if (adapter->large_llq_header_enabled) {
+		if ((llq->entry_size_ctrl_supported & ENA_ADMIN_LIST_ENTRY_SIZE_256B) &&
+		    ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
+			max_tx_queue_size /= 2;
+			dev_info(&adapter->pdev->dev,
+				 "Forcing large headers and decreasing maximum TX queue size to %d\n",
+				 max_tx_queue_size);
+		} else {
+			dev_err(&adapter->pdev->dev,
+				"Forcing large headers failed: LLQ is disabled or device does not support large headers\n");
+
+			adapter->large_llq_header_enabled = false;
+		}
+	}
+
 	tx_queue_size = clamp_val(tx_queue_size, ENA_MIN_RING_SIZE,
 				  max_tx_queue_size);
 	rx_queue_size = clamp_val(rx_queue_size, ENA_MIN_RING_SIZE,
@@ -3452,13 +3479,30 @@ static int ena_device_validate_params(struct ena_adapter *adapter,
 	return 0;
 }
 
-static void set_default_llq_configurations(struct ena_llq_configurations *llq_config)
+static void set_default_llq_configurations(struct ena_adapter *adapter,
+					   struct ena_llq_configurations *llq_config,
+					   struct ena_admin_feature_llq_desc *llq)
 {
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
+
 	llq_config->llq_header_location = ENA_ADMIN_INLINE_HEADER;
 	llq_config->llq_stride_ctrl = ENA_ADMIN_MULTIPLE_DESCS_PER_ENTRY;
 	llq_config->llq_num_decs_before_header = ENA_ADMIN_LLQ_NUM_DESCS_BEFORE_HEADER_2;
-	llq_config->llq_ring_entry_size = ENA_ADMIN_LIST_ENTRY_SIZE_128B;
-	llq_config->llq_ring_entry_size_value = 128;
+
+	adapter->large_llq_header_supported =
+		!!(ena_dev->supported_features & BIT(ENA_ADMIN_LLQ));
+	adapter->large_llq_header_supported &=
+		!!(llq->entry_size_ctrl_supported &
+			ENA_ADMIN_LIST_ENTRY_SIZE_256B);
+
+	if ((llq->entry_size_ctrl_supported & ENA_ADMIN_LIST_ENTRY_SIZE_256B) &&
+	    adapter->large_llq_header_enabled) {
+		llq_config->llq_ring_entry_size = ENA_ADMIN_LIST_ENTRY_SIZE_256B;
+		llq_config->llq_ring_entry_size_value = 256;
+	} else {
+		llq_config->llq_ring_entry_size = ENA_ADMIN_LIST_ENTRY_SIZE_128B;
+		llq_config->llq_ring_entry_size_value = 128;
+	}
 }
 
 static int ena_set_queues_placement_policy(struct pci_dev *pdev,
@@ -3477,6 +3521,13 @@ static int ena_set_queues_placement_policy(struct pci_dev *pdev,
 		return 0;
 	}
 
+	if (!ena_dev->mem_bar) {
+		netdev_err(ena_dev->net_device,
+			   "LLQ is advertised as supported but device doesn't expose mem bar\n");
+		ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
+		return 0;
+	}
+
 	rc = ena_com_config_dev_mode(ena_dev, llq, llq_default_configurations);
 	if (unlikely(rc)) {
 		dev_err(&pdev->dev,
@@ -3492,15 +3543,8 @@ static int ena_map_llq_mem_bar(struct pci_dev *pdev, struct ena_com_dev *ena_dev
 {
 	bool has_mem_bar = !!(bars & BIT(ENA_MEM_BAR));
 
-	if (!has_mem_bar) {
-		if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
-			dev_err(&pdev->dev,
-				"ENA device does not expose LLQ bar. Fallback to host mode policy.\n");
-			ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST;
-		}
-
+	if (!has_mem_bar)
 		return 0;
-	}
 
 	ena_dev->mem_bar = devm_ioremap_wc(&pdev->dev,
 					   pci_resource_start(pdev, ENA_MEM_BAR),
@@ -3512,10 +3556,11 @@ static int ena_map_llq_mem_bar(struct pci_dev *pdev, struct ena_com_dev *ena_dev
 	return 0;
 }
 
-static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
+static int ena_device_init(struct ena_adapter *adapter, struct pci_dev *pdev,
 			   struct ena_com_dev_get_features_ctx *get_feat_ctx,
 			   bool *wd_state)
 {
+	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	struct ena_llq_configurations llq_config;
 	struct device *dev = &pdev->dev;
 	bool readless_supported;
@@ -3600,7 +3645,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 
 	*wd_state = !!(aenq_groups & BIT(ENA_ADMIN_KEEP_ALIVE));
 
-	set_default_llq_configurations(&llq_config);
+	set_default_llq_configurations(adapter, &llq_config, &get_feat_ctx->llq);
 
 	rc = ena_set_queues_placement_policy(pdev, ena_dev, &get_feat_ctx->llq,
 					     &llq_config);
@@ -3609,6 +3654,8 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev,
 		goto err_admin_init;
 	}
 
+	ena_calc_io_queue_size(adapter, get_feat_ctx);
+
 	return 0;
 
 err_admin_init:
@@ -3707,7 +3754,7 @@ static int ena_restore_device(struct ena_adapter *adapter)
 	int rc;
 
 	set_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags);
-	rc = ena_device_init(ena_dev, adapter->pdev, &get_feat_ctx, &wd_state);
+	rc = ena_device_init(adapter, adapter->pdev, &get_feat_ctx, &wd_state);
 	if (rc) {
 		dev_err(&pdev->dev, "Can not initialize device\n");
 		goto err;
@@ -4309,18 +4356,18 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_drvdata(pdev, adapter);
 
-	rc = ena_device_init(ena_dev, pdev, &get_feat_ctx, &wd_state);
+	rc = ena_map_llq_mem_bar(pdev, ena_dev, bars);
 	if (rc) {
-		dev_err(&pdev->dev, "ENA device init failed\n");
-		if (rc == -ETIME)
-			rc = -EPROBE_DEFER;
+		dev_err(&pdev->dev, "ENA LLQ bar mapping failed\n");
 		goto err_netdev_destroy;
 	}
 
-	rc = ena_map_llq_mem_bar(pdev, ena_dev, bars);
+	rc = ena_device_init(adapter, pdev, &get_feat_ctx, &wd_state);
 	if (rc) {
-		dev_err(&pdev->dev, "ENA llq bar mapping failed\n");
-		goto err_device_destroy;
+		dev_err(&pdev->dev, "ENA device init failed\n");
+		if (rc == -ETIME)
+			rc = -EPROBE_DEFER;
+		goto err_netdev_destroy;
 	}
 
 	/* Initial TX and RX interrupt delay. Assumes 1 usec granularity.
@@ -4330,7 +4377,6 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	ena_dev->intr_moder_rx_interval = ENA_INTR_INITIAL_RX_INTERVAL_USECS;
 	ena_dev->intr_delay_resolution = ENA_DEFAULT_INTR_DELAY_RESOLUTION;
 	max_num_io_queues = ena_calc_max_io_queue_num(pdev, ena_dev, &get_feat_ctx);
-	ena_calc_io_queue_size(adapter, &get_feat_ctx);
 	if (unlikely(!max_num_io_queues)) {
 		rc = -EFAULT;
 		goto err_device_destroy;
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index 2cb141079474..3e8c4a66c7d8 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -334,6 +334,14 @@ struct ena_adapter {
 
 	u32 msg_enable;
 
+	/* large_llq_header_enabled is used for two purposes:
+	 * 1. Indicates that large LLQ has been requested.
+	 * 2. Indicates whether large LLQ is set or not after device
+	 *    initialization / configuration.
+	 */
+	bool large_llq_header_enabled;
+	bool large_llq_header_supported;
+
 	u16 max_tx_sgl_size;
 	u16 max_rx_sgl_size;
 
-- 
cgit v1.2.3-70-g09d2


From a416cb254d35d1e53ab3d4cbb81ea3835e5a10fd Mon Sep 17 00:00:00 2001
From: Shay Agroskin <shayagr@amazon.com>
Date: Thu, 23 Mar 2023 18:36:08 +0200
Subject: net: ena: Recalculate TX state variables every device reset

With the ability to modify LLQ entry size, the size of packet's
payload that can be written directly to the device changes.
This patch makes the driver recalculate this information every device
negotiation (also called device reset).

Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 97da4d841df3..ed5b019d27ea 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -3750,8 +3750,9 @@ static int ena_restore_device(struct ena_adapter *adapter)
 	struct ena_com_dev_get_features_ctx get_feat_ctx;
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	struct pci_dev *pdev = adapter->pdev;
+	struct ena_ring *txr;
+	int rc, count, i;
 	bool wd_state;
-	int rc;
 
 	set_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags);
 	rc = ena_device_init(adapter, adapter->pdev, &get_feat_ctx, &wd_state);
@@ -3761,6 +3762,13 @@ static int ena_restore_device(struct ena_adapter *adapter)
 	}
 	adapter->wd_state = wd_state;
 
+	count =  adapter->xdp_num_queues + adapter->num_io_queues;
+	for (i = 0 ; i < count; i++) {
+		txr = &adapter->tx_ring[i];
+		txr->tx_mem_queue_type = ena_dev->tx_mem_queue_type;
+		txr->tx_max_header_size = ena_dev->tx_max_header_size;
+	}
+
 	rc = ena_device_validate_params(adapter, &get_feat_ctx);
 	if (rc) {
 		dev_err(&pdev->dev, "Validation of device parameters failed\n");
-- 
cgit v1.2.3-70-g09d2


From b0c59e53968b55f6add088319bd31ca7dcdd8484 Mon Sep 17 00:00:00 2001
From: Shay Agroskin <shayagr@amazon.com>
Date: Thu, 23 Mar 2023 18:36:09 +0200
Subject: net: ena: Add support to changing tx_push_buf_len

The ENA driver allows for two distinct values for the number of bytes
of the packet's payload that can be written directly to the device.

For a value of 224 the driver turns on Large LLQ Header mode in which
the first 224 of the packet's payload are written to the LLQ.

Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_eth_com.h |  4 ++
 drivers/net/ethernet/amazon/ena/ena_ethtool.c | 57 +++++++++++++++++++++++++--
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 26 +++++++++---
 drivers/net/ethernet/amazon/ena/ena_netdev.h  |  7 ++--
 4 files changed, 82 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.h b/drivers/net/ethernet/amazon/ena/ena_eth_com.h
index 689313ee25a8..372b259279ec 100644
--- a/drivers/net/ethernet/amazon/ena/ena_eth_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.h
@@ -10,6 +10,10 @@
 
 /* head update threshold in units of (queue size / ENA_COMP_HEAD_THRESH) */
 #define ENA_COMP_HEAD_THRESH 4
+/* we allow 2 DMA descriptors per LLQ entry */
+#define ENA_LLQ_ENTRY_DESC_CHUNK_SIZE	(2 * sizeof(struct ena_eth_io_tx_desc))
+#define ENA_LLQ_HEADER		(128UL - ENA_LLQ_ENTRY_DESC_CHUNK_SIZE)
+#define ENA_LLQ_LARGE_HEADER	(256UL - ENA_LLQ_ENTRY_DESC_CHUNK_SIZE)
 
 struct ena_com_tx_ctx {
 	struct ena_com_tx_meta ena_meta;
diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index 1d4f2f4d10f2..de7ec02d8c09 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -476,6 +476,19 @@ static void ena_get_ringparam(struct net_device *netdev,
 
 	ring->tx_max_pending = adapter->max_tx_ring_size;
 	ring->rx_max_pending = adapter->max_rx_ring_size;
+	if (adapter->ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
+		bool large_llq_supported = adapter->large_llq_header_supported;
+
+		kernel_ring->tx_push_buf_len = adapter->ena_dev->tx_max_header_size;
+		if (large_llq_supported)
+			kernel_ring->tx_push_buf_max_len = ENA_LLQ_LARGE_HEADER;
+		else
+			kernel_ring->tx_push_buf_max_len = ENA_LLQ_HEADER;
+	} else {
+		kernel_ring->tx_push_buf_max_len = 0;
+		kernel_ring->tx_push_buf_len = 0;
+	}
+
 	ring->tx_pending = adapter->tx_ring[0].ring_size;
 	ring->rx_pending = adapter->rx_ring[0].ring_size;
 }
@@ -486,7 +499,8 @@ static int ena_set_ringparam(struct net_device *netdev,
 			     struct netlink_ext_ack *extack)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
-	u32 new_tx_size, new_rx_size;
+	u32 new_tx_size, new_rx_size, new_tx_push_buf_len;
+	bool changed = false;
 
 	new_tx_size = ring->tx_pending < ENA_MIN_RING_SIZE ?
 			ENA_MIN_RING_SIZE : ring->tx_pending;
@@ -496,11 +510,45 @@ static int ena_set_ringparam(struct net_device *netdev,
 			ENA_MIN_RING_SIZE : ring->rx_pending;
 	new_rx_size = rounddown_pow_of_two(new_rx_size);
 
-	if (new_tx_size == adapter->requested_tx_ring_size &&
-	    new_rx_size == adapter->requested_rx_ring_size)
+	changed |= new_tx_size != adapter->requested_tx_ring_size ||
+		   new_rx_size != adapter->requested_rx_ring_size;
+
+	/* This value is ignored if LLQ is not supported */
+	new_tx_push_buf_len = adapter->ena_dev->tx_max_header_size;
+
+	/* Validate that the push buffer is supported on the underlying device */
+	if (kernel_ring->tx_push_buf_len) {
+		enum ena_admin_placement_policy_type placement;
+
+		new_tx_push_buf_len = kernel_ring->tx_push_buf_len;
+
+		placement = adapter->ena_dev->tx_mem_queue_type;
+		if (placement == ENA_ADMIN_PLACEMENT_POLICY_HOST)
+			return -EOPNOTSUPP;
+
+		if (new_tx_push_buf_len != ENA_LLQ_HEADER &&
+		    new_tx_push_buf_len != ENA_LLQ_LARGE_HEADER) {
+			bool large_llq_sup = adapter->large_llq_header_supported;
+			char large_llq_size_str[40];
+
+			snprintf(large_llq_size_str, 40, ", %lu", ENA_LLQ_LARGE_HEADER);
+
+			NL_SET_ERR_MSG_FMT_MOD(extack,
+					       "Supported tx push buff values: [%lu%s]",
+					       ENA_LLQ_HEADER,
+					       large_llq_sup ? large_llq_size_str : "");
+
+			return -EINVAL;
+		}
+
+		changed |= new_tx_push_buf_len != adapter->ena_dev->tx_max_header_size;
+	}
+
+	if (!changed)
 		return 0;
 
-	return ena_update_queue_sizes(adapter, new_tx_size, new_rx_size);
+	return ena_update_queue_params(adapter, new_tx_size, new_rx_size,
+				       new_tx_push_buf_len);
 }
 
 static u32 ena_flow_hash_to_flow_type(u16 hash_fields)
@@ -909,6 +957,7 @@ static int ena_set_tunable(struct net_device *netdev,
 static const struct ethtool_ops ena_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
 				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
+	.supported_ring_params	= ETHTOOL_RING_USE_TX_PUSH_BUF_LEN,
 	.get_link_ksettings	= ena_get_link_ksettings,
 	.get_drvinfo		= ena_get_drvinfo,
 	.get_msglevel		= ena_get_msglevel,
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index ed5b019d27ea..e0588a82c8e5 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -2809,11 +2809,13 @@ static int ena_close(struct net_device *netdev)
 	return 0;
 }
 
-int ena_update_queue_sizes(struct ena_adapter *adapter,
-			   u32 new_tx_size,
-			   u32 new_rx_size)
+int ena_update_queue_params(struct ena_adapter *adapter,
+			    u32 new_tx_size,
+			    u32 new_rx_size,
+			    u32 new_llq_header_len)
 {
-	bool dev_was_up;
+	bool dev_was_up, large_llq_changed = false;
+	int rc = 0;
 
 	dev_was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
 	ena_close(adapter->netdev);
@@ -2823,7 +2825,21 @@ int ena_update_queue_sizes(struct ena_adapter *adapter,
 			  0,
 			  adapter->xdp_num_queues +
 			  adapter->num_io_queues);
-	return dev_was_up ? ena_up(adapter) : 0;
+
+	large_llq_changed = adapter->ena_dev->tx_mem_queue_type ==
+			    ENA_ADMIN_PLACEMENT_POLICY_DEV;
+	large_llq_changed &=
+		new_llq_header_len != adapter->ena_dev->tx_max_header_size;
+
+	/* a check that the configuration is valid is done by caller */
+	if (large_llq_changed) {
+		adapter->large_llq_header_enabled = !adapter->large_llq_header_enabled;
+
+		ena_destroy_device(adapter, false);
+		rc = ena_restore_device(adapter);
+	}
+
+	return dev_was_up && !rc ? ena_up(adapter) : rc;
 }
 
 int ena_set_rx_copybreak(struct ena_adapter *adapter, u32 rx_copybreak)
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index 3e8c4a66c7d8..5a0d4ee76172 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -396,9 +396,10 @@ void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf);
 
 int ena_update_hw_stats(struct ena_adapter *adapter);
 
-int ena_update_queue_sizes(struct ena_adapter *adapter,
-			   u32 new_tx_size,
-			   u32 new_rx_size);
+int ena_update_queue_params(struct ena_adapter *adapter,
+			    u32 new_tx_size,
+			    u32 new_rx_size,
+			    u32 new_llq_header_len);
 
 int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count);
 
-- 
cgit v1.2.3-70-g09d2


From 060cdac218eaf3650d7556e4ed8b6d2bd79d4a0c Mon Sep 17 00:00:00 2001
From: Shay Agroskin <shayagr@amazon.com>
Date: Thu, 23 Mar 2023 18:36:10 +0200
Subject: net: ena: Advertise TX push support

LLQ is auto enabled by the device and disabling it isn't supported on
new ENA generations while on old ones causes sub-optimal performance.

This patch adds advertisement of push-mode when LLQ mode is used, but
rejects an attempt to modify it.

Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Signed-off-by: Shay Agroskin <shayagr@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index de7ec02d8c09..d671df4b76bc 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -479,12 +479,14 @@ static void ena_get_ringparam(struct net_device *netdev,
 	if (adapter->ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) {
 		bool large_llq_supported = adapter->large_llq_header_supported;
 
+		kernel_ring->tx_push = true;
 		kernel_ring->tx_push_buf_len = adapter->ena_dev->tx_max_header_size;
 		if (large_llq_supported)
 			kernel_ring->tx_push_buf_max_len = ENA_LLQ_LARGE_HEADER;
 		else
 			kernel_ring->tx_push_buf_max_len = ENA_LLQ_HEADER;
 	} else {
+		kernel_ring->tx_push = false;
 		kernel_ring->tx_push_buf_max_len = 0;
 		kernel_ring->tx_push_buf_len = 0;
 	}
@@ -516,6 +518,12 @@ static int ena_set_ringparam(struct net_device *netdev,
 	/* This value is ignored if LLQ is not supported */
 	new_tx_push_buf_len = adapter->ena_dev->tx_max_header_size;
 
+	if ((adapter->ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) !=
+	    kernel_ring->tx_push) {
+		NL_SET_ERR_MSG_MOD(extack, "Push mode state cannot be modified");
+		return -EINVAL;
+	}
+
 	/* Validate that the push buffer is supported on the underlying device */
 	if (kernel_ring->tx_push_buf_len) {
 		enum ena_admin_placement_policy_type placement;
@@ -957,7 +965,8 @@ static int ena_set_tunable(struct net_device *netdev,
 static const struct ethtool_ops ena_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
 				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX,
-	.supported_ring_params	= ETHTOOL_RING_USE_TX_PUSH_BUF_LEN,
+	.supported_ring_params	= ETHTOOL_RING_USE_TX_PUSH_BUF_LEN |
+				  ETHTOOL_RING_USE_TX_PUSH,
 	.get_link_ksettings	= ena_get_link_ksettings,
 	.get_drvinfo		= ena_get_drvinfo,
 	.get_msglevel		= ena_get_msglevel,
-- 
cgit v1.2.3-70-g09d2


From d08ab82f59d55b0e5acfeb453081278dfc33f232 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Tue, 28 Mar 2023 03:47:38 +0300
Subject: libbpf: Fix double-free when linker processes empty sections

Double-free error in bpf_linker__free() was reported by James Hilliard.
The error is caused by miss-use of realloc() in extend_sec().
The error occurs when two files with empty sections of the same name
are linked:
- when first file is processed:
  - extend_sec() calls realloc(dst->raw_data, dst_align_sz)
    with dst->raw_data == NULL and dst_align_sz == 0;
  - dst->raw_data is set to a special pointer to a memory block of
    size zero;
- when second file is processed:
  - extend_sec() calls realloc(dst->raw_data, dst_align_sz)
    with dst->raw_data == <special pointer> and dst_align_sz == 0;
  - realloc() "frees" dst->raw_data special pointer and returns NULL;
  - extend_sec() exits with -ENOMEM, and the old dst->raw_data value
    is preserved (it is now invalid);
  - eventually, bpf_linker__free() attempts to free dst->raw_data again.

This patch fixes the bug by avoiding -ENOMEM exit for dst_align_sz == 0.
The fix was suggested by Andrii Nakryiko <andrii.nakryiko@gmail.com>.

Reported-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: James Hilliard <james.hilliard1@gmail.com>
Link: https://lore.kernel.org/bpf/CADvTj4o7ZWUikKwNTwFq0O_AaX+46t_+Ca9gvWMYdWdRtTGeHQ@mail.gmail.com/
Link: https://lore.kernel.org/bpf/20230328004738.381898-3-eddyz87@gmail.com
---
 tools/lib/bpf/linker.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/linker.c b/tools/lib/bpf/linker.c
index d7069780984a..5ced96d99f8c 100644
--- a/tools/lib/bpf/linker.c
+++ b/tools/lib/bpf/linker.c
@@ -1115,7 +1115,19 @@ static int extend_sec(struct bpf_linker *linker, struct dst_sec *dst, struct src
 
 	if (src->shdr->sh_type != SHT_NOBITS) {
 		tmp = realloc(dst->raw_data, dst_final_sz);
-		if (!tmp)
+		/* If dst_align_sz == 0, realloc() behaves in a special way:
+		 * 1. When dst->raw_data is NULL it returns:
+		 *    "either NULL or a pointer suitable to be passed to free()" [1].
+		 * 2. When dst->raw_data is not-NULL it frees dst->raw_data and returns NULL,
+		 *    thus invalidating any "pointer suitable to be passed to free()" obtained
+		 *    at step (1).
+		 *
+		 * The dst_align_sz > 0 check avoids error exit after (2), otherwise
+		 * dst->raw_data would be freed again in bpf_linker__free().
+		 *
+		 * [1] man 3 realloc
+		 */
+		if (!tmp && dst_align_sz > 0)
 			return -ENOMEM;
 		dst->raw_data = tmp;
 
-- 
cgit v1.2.3-70-g09d2


From 96d1c81e6a0478535342dff6c730adb076cd84e8 Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <socketcan@hartkopp.net>
Date: Sun, 26 Mar 2023 13:59:11 +0200
Subject: can: isotp: add module parameter for maximum pdu size

With ISO 15765-2:2016 the PDU size is not limited to 2^12 - 1 (4095)
bytes but can be represented as a 32 bit unsigned integer value which
allows 2^32 - 1 bytes (~4GB). The use-cases like automotive unified
diagnostic services (UDS) and flashing of ECUs still use the small
static buffers which are provided at socket creation time.

When a use-case requires to transfer PDUs up to 1025 kByte the maximum
PDU size can now be extended by setting the module parameter
max_pdu_size. The extended size buffers are only allocated on a
per-socket/connection base when needed at run-time.

changes since v2: https://lore.kernel.org/all/20230313172510.3851-1-socketcan@hartkopp.net
- use ARRAY_SIZE() to reference DEFAULT_MAX_PDU_SIZE only at one place

changes since v1: https://lore.kernel.org/all/20230311143446.3183-1-socketcan@hartkopp.net
- limit the minimum 'max_pdu_size' to 4095 to maintain the classic
  behavior before ISO 15765-2:2016

Link: https://github.com/raspberrypi/linux/issues/5371
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Link: https://lore.kernel.org/all/20230326115911.15094-1-socketcan@hartkopp.net
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 net/can/isotp.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 56 insertions(+), 9 deletions(-)

diff --git a/net/can/isotp.c b/net/can/isotp.c
index 9bc344851704..f1ea91772ae7 100644
--- a/net/can/isotp.c
+++ b/net/can/isotp.c
@@ -85,10 +85,21 @@ MODULE_ALIAS("can-proto-6");
 
 /* ISO 15765-2:2016 supports more than 4095 byte per ISO PDU as the FF_DL can
  * take full 32 bit values (4 Gbyte). We would need some good concept to handle
- * this between user space and kernel space. For now increase the static buffer
- * to something about 64 kbyte to be able to test this new functionality.
+ * this between user space and kernel space. For now set the static buffer to
+ * something about 8 kbyte to be able to test this new functionality.
  */
-#define MAX_MSG_LENGTH 66000
+#define DEFAULT_MAX_PDU_SIZE 8300
+
+/* maximum PDU size before ISO 15765-2:2016 extension was 4095 */
+#define MAX_12BIT_PDU_SIZE 4095
+
+/* limit the isotp pdu size from the optional module parameter to 1MByte */
+#define MAX_PDU_SIZE (1025 * 1024U)
+
+static unsigned int max_pdu_size __read_mostly = DEFAULT_MAX_PDU_SIZE;
+module_param(max_pdu_size, uint, 0444);
+MODULE_PARM_DESC(max_pdu_size, "maximum isotp pdu size (default "
+		 __stringify(DEFAULT_MAX_PDU_SIZE) ")");
 
 /* N_PCI type values in bits 7-4 of N_PCI bytes */
 #define N_PCI_SF 0x00	/* single frame */
@@ -123,13 +134,15 @@ enum {
 };
 
 struct tpcon {
-	unsigned int idx;
+	u8 *buf;
+	unsigned int buflen;
 	unsigned int len;
+	unsigned int idx;
 	u32 state;
 	u8 bs;
 	u8 sn;
 	u8 ll_dl;
-	u8 buf[MAX_MSG_LENGTH + 1];
+	u8 sbuf[DEFAULT_MAX_PDU_SIZE];
 };
 
 struct isotp_sock {
@@ -503,7 +516,17 @@ static int isotp_rcv_ff(struct sock *sk, struct canfd_frame *cf, int ae)
 	if (so->rx.len + ae + off + ff_pci_sz < so->rx.ll_dl)
 		return 1;
 
-	if (so->rx.len > MAX_MSG_LENGTH) {
+	/* PDU size > default => try max_pdu_size */
+	if (so->rx.len > so->rx.buflen && so->rx.buflen < max_pdu_size) {
+		u8 *newbuf = kmalloc(max_pdu_size, GFP_ATOMIC);
+
+		if (newbuf) {
+			so->rx.buf = newbuf;
+			so->rx.buflen = max_pdu_size;
+		}
+	}
+
+	if (so->rx.len > so->rx.buflen) {
 		/* send FC frame with overflow status */
 		isotp_send_fc(sk, ae, ISOTP_FC_OVFLW);
 		return 1;
@@ -807,7 +830,7 @@ static void isotp_create_fframe(struct canfd_frame *cf, struct isotp_sock *so,
 		cf->data[0] = so->opt.ext_address;
 
 	/* create N_PCI bytes with 12/32 bit FF_DL data length */
-	if (so->tx.len > 4095) {
+	if (so->tx.len > MAX_12BIT_PDU_SIZE) {
 		/* use 32 bit FF_DL notation */
 		cf->data[ae] = N_PCI_FF;
 		cf->data[ae + 1] = 0;
@@ -947,7 +970,17 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
 		so->tx.state = ISOTP_SENDING;
 	}
 
-	if (!size || size > MAX_MSG_LENGTH) {
+	/* PDU size > default => try max_pdu_size */
+	if (size > so->tx.buflen && so->tx.buflen < max_pdu_size) {
+		u8 *newbuf = kmalloc(max_pdu_size, GFP_KERNEL);
+
+		if (newbuf) {
+			so->tx.buf = newbuf;
+			so->tx.buflen = max_pdu_size;
+		}
+	}
+
+	if (!size || size > so->tx.buflen) {
 		err = -EINVAL;
 		goto err_out_drop;
 	}
@@ -1195,6 +1228,12 @@ static int isotp_release(struct socket *sock)
 	so->ifindex = 0;
 	so->bound = 0;
 
+	if (so->rx.buf != so->rx.sbuf)
+		kfree(so->rx.buf);
+
+	if (so->tx.buf != so->tx.sbuf)
+		kfree(so->tx.buf);
+
 	sock_orphan(sk);
 	sock->sk = NULL;
 
@@ -1591,6 +1630,11 @@ static int isotp_init(struct sock *sk)
 	so->rx.state = ISOTP_IDLE;
 	so->tx.state = ISOTP_IDLE;
 
+	so->rx.buf = so->rx.sbuf;
+	so->tx.buf = so->tx.sbuf;
+	so->rx.buflen = ARRAY_SIZE(so->rx.sbuf);
+	so->tx.buflen = ARRAY_SIZE(so->tx.sbuf);
+
 	hrtimer_init(&so->rxtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT);
 	so->rxtimer.function = isotp_rx_timer_handler;
 	hrtimer_init(&so->txtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT);
@@ -1658,7 +1702,10 @@ static __init int isotp_module_init(void)
 {
 	int err;
 
-	pr_info("can: isotp protocol\n");
+	max_pdu_size = max_t(unsigned int, max_pdu_size, MAX_12BIT_PDU_SIZE);
+	max_pdu_size = min_t(unsigned int, max_pdu_size, MAX_PDU_SIZE);
+
+	pr_info("can: isotp protocol (max_pdu_size %d)\n", max_pdu_size);
 
 	err = can_proto_register(&isotp_can_proto);
 	if (err < 0)
-- 
cgit v1.2.3-70-g09d2


From e5ab9eff46b04c5a04778e40d7092fed3fda52ca Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 23 Mar 2023 21:55:30 +0100
Subject: atomics: Provide atomic_add_negative() variants

atomic_add_negative() does not provide the relaxed/acquire/release
variants.

Provide them in preparation for a new scalable reference count algorithm.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20230323102800.101763813@linutronix.de
---
 include/linux/atomic/atomic-arch-fallback.h | 208 ++++++++++++++++++++++++++--
 include/linux/atomic/atomic-instrumented.h  |  68 ++++++++-
 include/linux/atomic/atomic-long.h          |  38 ++++-
 scripts/atomic/atomics.tbl                  |   2 +-
 scripts/atomic/fallbacks/add_negative       |  11 +-
 5 files changed, 309 insertions(+), 18 deletions(-)

diff --git a/include/linux/atomic/atomic-arch-fallback.h b/include/linux/atomic/atomic-arch-fallback.h
index 77bc5522e61c..4226379a232d 100644
--- a/include/linux/atomic/atomic-arch-fallback.h
+++ b/include/linux/atomic/atomic-arch-fallback.h
@@ -1208,15 +1208,21 @@ arch_atomic_inc_and_test(atomic_t *v)
 #define arch_atomic_inc_and_test arch_atomic_inc_and_test
 #endif
 
+#ifndef arch_atomic_add_negative_relaxed
+#ifdef arch_atomic_add_negative
+#define arch_atomic_add_negative_acquire arch_atomic_add_negative
+#define arch_atomic_add_negative_release arch_atomic_add_negative
+#define arch_atomic_add_negative_relaxed arch_atomic_add_negative
+#endif /* arch_atomic_add_negative */
+
 #ifndef arch_atomic_add_negative
 /**
- * arch_atomic_add_negative - add and test if negative
+ * arch_atomic_add_negative - Add and test if negative
  * @i: integer value to add
  * @v: pointer of type atomic_t
  *
- * Atomically adds @i to @v and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
  */
 static __always_inline bool
 arch_atomic_add_negative(int i, atomic_t *v)
@@ -1226,6 +1232,95 @@ arch_atomic_add_negative(int i, atomic_t *v)
 #define arch_atomic_add_negative arch_atomic_add_negative
 #endif
 
+#ifndef arch_atomic_add_negative_acquire
+/**
+ * arch_atomic_add_negative_acquire - Add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
+ */
+static __always_inline bool
+arch_atomic_add_negative_acquire(int i, atomic_t *v)
+{
+	return arch_atomic_add_return_acquire(i, v) < 0;
+}
+#define arch_atomic_add_negative_acquire arch_atomic_add_negative_acquire
+#endif
+
+#ifndef arch_atomic_add_negative_release
+/**
+ * arch_atomic_add_negative_release - Add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
+ */
+static __always_inline bool
+arch_atomic_add_negative_release(int i, atomic_t *v)
+{
+	return arch_atomic_add_return_release(i, v) < 0;
+}
+#define arch_atomic_add_negative_release arch_atomic_add_negative_release
+#endif
+
+#ifndef arch_atomic_add_negative_relaxed
+/**
+ * arch_atomic_add_negative_relaxed - Add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
+ */
+static __always_inline bool
+arch_atomic_add_negative_relaxed(int i, atomic_t *v)
+{
+	return arch_atomic_add_return_relaxed(i, v) < 0;
+}
+#define arch_atomic_add_negative_relaxed arch_atomic_add_negative_relaxed
+#endif
+
+#else /* arch_atomic_add_negative_relaxed */
+
+#ifndef arch_atomic_add_negative_acquire
+static __always_inline bool
+arch_atomic_add_negative_acquire(int i, atomic_t *v)
+{
+	bool ret = arch_atomic_add_negative_relaxed(i, v);
+	__atomic_acquire_fence();
+	return ret;
+}
+#define arch_atomic_add_negative_acquire arch_atomic_add_negative_acquire
+#endif
+
+#ifndef arch_atomic_add_negative_release
+static __always_inline bool
+arch_atomic_add_negative_release(int i, atomic_t *v)
+{
+	__atomic_release_fence();
+	return arch_atomic_add_negative_relaxed(i, v);
+}
+#define arch_atomic_add_negative_release arch_atomic_add_negative_release
+#endif
+
+#ifndef arch_atomic_add_negative
+static __always_inline bool
+arch_atomic_add_negative(int i, atomic_t *v)
+{
+	bool ret;
+	__atomic_pre_full_fence();
+	ret = arch_atomic_add_negative_relaxed(i, v);
+	__atomic_post_full_fence();
+	return ret;
+}
+#define arch_atomic_add_negative arch_atomic_add_negative
+#endif
+
+#endif /* arch_atomic_add_negative_relaxed */
+
 #ifndef arch_atomic_fetch_add_unless
 /**
  * arch_atomic_fetch_add_unless - add unless the number is already a given value
@@ -2329,15 +2424,21 @@ arch_atomic64_inc_and_test(atomic64_t *v)
 #define arch_atomic64_inc_and_test arch_atomic64_inc_and_test
 #endif
 
+#ifndef arch_atomic64_add_negative_relaxed
+#ifdef arch_atomic64_add_negative
+#define arch_atomic64_add_negative_acquire arch_atomic64_add_negative
+#define arch_atomic64_add_negative_release arch_atomic64_add_negative
+#define arch_atomic64_add_negative_relaxed arch_atomic64_add_negative
+#endif /* arch_atomic64_add_negative */
+
 #ifndef arch_atomic64_add_negative
 /**
- * arch_atomic64_add_negative - add and test if negative
+ * arch_atomic64_add_negative - Add and test if negative
  * @i: integer value to add
  * @v: pointer of type atomic64_t
  *
- * Atomically adds @i to @v and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
  */
 static __always_inline bool
 arch_atomic64_add_negative(s64 i, atomic64_t *v)
@@ -2347,6 +2448,95 @@ arch_atomic64_add_negative(s64 i, atomic64_t *v)
 #define arch_atomic64_add_negative arch_atomic64_add_negative
 #endif
 
+#ifndef arch_atomic64_add_negative_acquire
+/**
+ * arch_atomic64_add_negative_acquire - Add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic64_t
+ *
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
+ */
+static __always_inline bool
+arch_atomic64_add_negative_acquire(s64 i, atomic64_t *v)
+{
+	return arch_atomic64_add_return_acquire(i, v) < 0;
+}
+#define arch_atomic64_add_negative_acquire arch_atomic64_add_negative_acquire
+#endif
+
+#ifndef arch_atomic64_add_negative_release
+/**
+ * arch_atomic64_add_negative_release - Add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic64_t
+ *
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
+ */
+static __always_inline bool
+arch_atomic64_add_negative_release(s64 i, atomic64_t *v)
+{
+	return arch_atomic64_add_return_release(i, v) < 0;
+}
+#define arch_atomic64_add_negative_release arch_atomic64_add_negative_release
+#endif
+
+#ifndef arch_atomic64_add_negative_relaxed
+/**
+ * arch_atomic64_add_negative_relaxed - Add and test if negative
+ * @i: integer value to add
+ * @v: pointer of type atomic64_t
+ *
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
+ */
+static __always_inline bool
+arch_atomic64_add_negative_relaxed(s64 i, atomic64_t *v)
+{
+	return arch_atomic64_add_return_relaxed(i, v) < 0;
+}
+#define arch_atomic64_add_negative_relaxed arch_atomic64_add_negative_relaxed
+#endif
+
+#else /* arch_atomic64_add_negative_relaxed */
+
+#ifndef arch_atomic64_add_negative_acquire
+static __always_inline bool
+arch_atomic64_add_negative_acquire(s64 i, atomic64_t *v)
+{
+	bool ret = arch_atomic64_add_negative_relaxed(i, v);
+	__atomic_acquire_fence();
+	return ret;
+}
+#define arch_atomic64_add_negative_acquire arch_atomic64_add_negative_acquire
+#endif
+
+#ifndef arch_atomic64_add_negative_release
+static __always_inline bool
+arch_atomic64_add_negative_release(s64 i, atomic64_t *v)
+{
+	__atomic_release_fence();
+	return arch_atomic64_add_negative_relaxed(i, v);
+}
+#define arch_atomic64_add_negative_release arch_atomic64_add_negative_release
+#endif
+
+#ifndef arch_atomic64_add_negative
+static __always_inline bool
+arch_atomic64_add_negative(s64 i, atomic64_t *v)
+{
+	bool ret;
+	__atomic_pre_full_fence();
+	ret = arch_atomic64_add_negative_relaxed(i, v);
+	__atomic_post_full_fence();
+	return ret;
+}
+#define arch_atomic64_add_negative arch_atomic64_add_negative
+#endif
+
+#endif /* arch_atomic64_add_negative_relaxed */
+
 #ifndef arch_atomic64_fetch_add_unless
 /**
  * arch_atomic64_fetch_add_unless - add unless the number is already a given value
@@ -2456,4 +2646,4 @@ arch_atomic64_dec_if_positive(atomic64_t *v)
 #endif
 
 #endif /* _LINUX_ATOMIC_FALLBACK_H */
-// b5e87bdd5ede61470c29f7a7e4de781af3770f09
+// 00071fffa021cec66f6290d706d69c91df87bade
diff --git a/include/linux/atomic/atomic-instrumented.h b/include/linux/atomic/atomic-instrumented.h
index 7a139ec030b0..0496816738ca 100644
--- a/include/linux/atomic/atomic-instrumented.h
+++ b/include/linux/atomic/atomic-instrumented.h
@@ -592,6 +592,28 @@ atomic_add_negative(int i, atomic_t *v)
 	return arch_atomic_add_negative(i, v);
 }
 
+static __always_inline bool
+atomic_add_negative_acquire(int i, atomic_t *v)
+{
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic_add_negative_acquire(i, v);
+}
+
+static __always_inline bool
+atomic_add_negative_release(int i, atomic_t *v)
+{
+	kcsan_release();
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic_add_negative_release(i, v);
+}
+
+static __always_inline bool
+atomic_add_negative_relaxed(int i, atomic_t *v)
+{
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic_add_negative_relaxed(i, v);
+}
+
 static __always_inline int
 atomic_fetch_add_unless(atomic_t *v, int a, int u)
 {
@@ -1211,6 +1233,28 @@ atomic64_add_negative(s64 i, atomic64_t *v)
 	return arch_atomic64_add_negative(i, v);
 }
 
+static __always_inline bool
+atomic64_add_negative_acquire(s64 i, atomic64_t *v)
+{
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic64_add_negative_acquire(i, v);
+}
+
+static __always_inline bool
+atomic64_add_negative_release(s64 i, atomic64_t *v)
+{
+	kcsan_release();
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic64_add_negative_release(i, v);
+}
+
+static __always_inline bool
+atomic64_add_negative_relaxed(s64 i, atomic64_t *v)
+{
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic64_add_negative_relaxed(i, v);
+}
+
 static __always_inline s64
 atomic64_fetch_add_unless(atomic64_t *v, s64 a, s64 u)
 {
@@ -1830,6 +1874,28 @@ atomic_long_add_negative(long i, atomic_long_t *v)
 	return arch_atomic_long_add_negative(i, v);
 }
 
+static __always_inline bool
+atomic_long_add_negative_acquire(long i, atomic_long_t *v)
+{
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic_long_add_negative_acquire(i, v);
+}
+
+static __always_inline bool
+atomic_long_add_negative_release(long i, atomic_long_t *v)
+{
+	kcsan_release();
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic_long_add_negative_release(i, v);
+}
+
+static __always_inline bool
+atomic_long_add_negative_relaxed(long i, atomic_long_t *v)
+{
+	instrument_atomic_read_write(v, sizeof(*v));
+	return arch_atomic_long_add_negative_relaxed(i, v);
+}
+
 static __always_inline long
 atomic_long_fetch_add_unless(atomic_long_t *v, long a, long u)
 {
@@ -2083,4 +2149,4 @@ atomic_long_dec_if_positive(atomic_long_t *v)
 })
 
 #endif /* _LINUX_ATOMIC_INSTRUMENTED_H */
-// 764f741eb77a7ad565dc8d99ce2837d5542e8aee
+// 1b485de9cbaa4900de59e14ee2084357eaeb1c3a
diff --git a/include/linux/atomic/atomic-long.h b/include/linux/atomic/atomic-long.h
index 800b8c35992d..2fc51ba66beb 100644
--- a/include/linux/atomic/atomic-long.h
+++ b/include/linux/atomic/atomic-long.h
@@ -479,6 +479,24 @@ arch_atomic_long_add_negative(long i, atomic_long_t *v)
 	return arch_atomic64_add_negative(i, v);
 }
 
+static __always_inline bool
+arch_atomic_long_add_negative_acquire(long i, atomic_long_t *v)
+{
+	return arch_atomic64_add_negative_acquire(i, v);
+}
+
+static __always_inline bool
+arch_atomic_long_add_negative_release(long i, atomic_long_t *v)
+{
+	return arch_atomic64_add_negative_release(i, v);
+}
+
+static __always_inline bool
+arch_atomic_long_add_negative_relaxed(long i, atomic_long_t *v)
+{
+	return arch_atomic64_add_negative_relaxed(i, v);
+}
+
 static __always_inline long
 arch_atomic_long_fetch_add_unless(atomic_long_t *v, long a, long u)
 {
@@ -973,6 +991,24 @@ arch_atomic_long_add_negative(long i, atomic_long_t *v)
 	return arch_atomic_add_negative(i, v);
 }
 
+static __always_inline bool
+arch_atomic_long_add_negative_acquire(long i, atomic_long_t *v)
+{
+	return arch_atomic_add_negative_acquire(i, v);
+}
+
+static __always_inline bool
+arch_atomic_long_add_negative_release(long i, atomic_long_t *v)
+{
+	return arch_atomic_add_negative_release(i, v);
+}
+
+static __always_inline bool
+arch_atomic_long_add_negative_relaxed(long i, atomic_long_t *v)
+{
+	return arch_atomic_add_negative_relaxed(i, v);
+}
+
 static __always_inline long
 arch_atomic_long_fetch_add_unless(atomic_long_t *v, long a, long u)
 {
@@ -1011,4 +1047,4 @@ arch_atomic_long_dec_if_positive(atomic_long_t *v)
 
 #endif /* CONFIG_64BIT */
 #endif /* _LINUX_ATOMIC_LONG_H */
-// e8f0e08ff072b74d180eabe2ad001282b38c2c88
+// a194c07d7d2f4b0e178d3c118c919775d5d65f50
diff --git a/scripts/atomic/atomics.tbl b/scripts/atomic/atomics.tbl
index fbee2f6190d9..85ca8d9b5c27 100644
--- a/scripts/atomic/atomics.tbl
+++ b/scripts/atomic/atomics.tbl
@@ -33,7 +33,7 @@ try_cmpxchg		B	v	p:old	i:new
 sub_and_test		b	i	v
 dec_and_test		b	v
 inc_and_test		b	v
-add_negative		b	i	v
+add_negative		B	i	v
 add_unless		fb	v	i:a	i:u
 inc_not_zero		b	v
 inc_unless_negative	b	v
diff --git a/scripts/atomic/fallbacks/add_negative b/scripts/atomic/fallbacks/add_negative
index 15caa2eb2371..e5980abf5904 100755
--- a/scripts/atomic/fallbacks/add_negative
+++ b/scripts/atomic/fallbacks/add_negative
@@ -1,16 +1,15 @@
 cat <<EOF
 /**
- * arch_${atomic}_add_negative - add and test if negative
+ * arch_${atomic}_add_negative${order} - Add and test if negative
  * @i: integer value to add
  * @v: pointer of type ${atomic}_t
  *
- * Atomically adds @i to @v and returns true
- * if the result is negative, or false when
- * result is greater than or equal to zero.
+ * Atomically adds @i to @v and returns true if the result is negative,
+ * or false when the result is greater than or equal to zero.
  */
 static __always_inline bool
-arch_${atomic}_add_negative(${int} i, ${atomic}_t *v)
+arch_${atomic}_add_negative${order}(${int} i, ${atomic}_t *v)
 {
-	return arch_${atomic}_add_return(i, v) < 0;
+	return arch_${atomic}_add_return${order}(i, v) < 0;
 }
 EOF
-- 
cgit v1.2.3-70-g09d2


From ee1ee6db07795d9637bc5e8993a8ddcf886541ef Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 23 Mar 2023 21:55:31 +0100
Subject: atomics: Provide rcuref - scalable reference counting

atomic_t based reference counting, including refcount_t, uses
atomic_inc_not_zero() for acquiring a reference. atomic_inc_not_zero() is
implemented with a atomic_try_cmpxchg() loop. High contention of the
reference count leads to retry loops and scales badly. There is nothing to
improve on this implementation as the semantics have to be preserved.

Provide rcuref as a scalable alternative solution which is suitable for RCU
managed objects. Similar to refcount_t it comes with overflow and underflow
detection and mitigation.

rcuref treats the underlying atomic_t as an unsigned integer and partitions
this space into zones:

  0x00000000 - 0x7FFFFFFF	valid zone (1 .. (INT_MAX + 1) references)
  0x80000000 - 0xBFFFFFFF	saturation zone
  0xC0000000 - 0xFFFFFFFE	dead zone
  0xFFFFFFFF   			no reference

rcuref_get() unconditionally increments the reference count with
atomic_add_negative_relaxed(). rcuref_put() unconditionally decrements the
reference count with atomic_add_negative_release().

This unconditional increment avoids the inc_not_zero() problem, but
requires a more complex implementation on the put() side when the count
drops from 0 to -1.

When this transition is detected then it is attempted to mark the reference
count dead, by setting it to the midpoint of the dead zone with a single
atomic_cmpxchg_release() operation. This operation can fail due to a
concurrent rcuref_get() elevating the reference count from -1 to 0 again.

If the unconditional increment in rcuref_get() hits a reference count which
is marked dead (or saturated) it will detect it after the fact and bring
back the reference count to the midpoint of the respective zone. The zones
provide enough tolerance which makes it practically impossible to escape
from a zone.

The racy implementation of rcuref_put() requires to protect rcuref_put()
against a grace period ending in order to prevent a subtle use after
free. As RCU is the only mechanism which allows to protect against that, it
is not possible to fully replace the atomic_inc_not_zero() based
implementation of refcount_t with this scheme.

The final drop is slightly more expensive than the atomic_dec_return()
counterpart, but that's not the case which this is optimized for. The
optimization is on the high frequeunt get()/put() pairs and their
scalability.

The performance of an uncontended rcuref_get()/put() pair where the put()
is not dropping the last reference is still on par with the plain atomic
operations, while at the same time providing overflow and underflow
detection and mitigation.

The performance of rcuref compared to plain atomic_inc_not_zero() and
atomic_dec_return() based reference counting under contention:

 -  Micro benchmark: All CPUs running a increment/decrement loop on an
    elevated reference count, which means the 0 to -1 transition never
    happens.

    The performance gain depends on microarchitecture and the number of
    CPUs and has been observed in the range of 1.3X to 4.7X

 - Conversion of dst_entry::__refcnt to rcuref and testing with the
    localhost memtier/memcached benchmark. That benchmark shows the
    reference count contention prominently.

    The performance gain depends on microarchitecture and the number of
    CPUs and has been observed in the range of 1.1X to 2.6X over the
    previous fix for the false sharing issue vs. struct
    dst_entry::__refcnt.

    When memtier is run over a real 1Gb network connection, there is a
    small gain on top of the false sharing fix. The two changes combined
    result in a 2%-5% total gain for that networked test.

Reported-by: Wangyang Guo <wangyang.guo@intel.com>
Reported-by: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230323102800.158429195@linutronix.de
---
 include/linux/rcuref.h | 155 +++++++++++++++++++++++++++
 include/linux/types.h  |   6 ++
 lib/Makefile           |   2 +-
 lib/rcuref.c           | 281 +++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 443 insertions(+), 1 deletion(-)
 create mode 100644 include/linux/rcuref.h
 create mode 100644 lib/rcuref.c

diff --git a/include/linux/rcuref.h b/include/linux/rcuref.h
new file mode 100644
index 000000000000..2c8bfd0f1b6b
--- /dev/null
+++ b/include/linux/rcuref.h
@@ -0,0 +1,155 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef _LINUX_RCUREF_H
+#define _LINUX_RCUREF_H
+
+#include <linux/atomic.h>
+#include <linux/bug.h>
+#include <linux/limits.h>
+#include <linux/lockdep.h>
+#include <linux/preempt.h>
+#include <linux/rcupdate.h>
+
+#define RCUREF_ONEREF		0x00000000U
+#define RCUREF_MAXREF		0x7FFFFFFFU
+#define RCUREF_SATURATED	0xA0000000U
+#define RCUREF_RELEASED		0xC0000000U
+#define RCUREF_DEAD		0xE0000000U
+#define RCUREF_NOREF		0xFFFFFFFFU
+
+/**
+ * rcuref_init - Initialize a rcuref reference count with the given reference count
+ * @ref:	Pointer to the reference count
+ * @cnt:	The initial reference count typically '1'
+ */
+static inline void rcuref_init(rcuref_t *ref, unsigned int cnt)
+{
+	atomic_set(&ref->refcnt, cnt - 1);
+}
+
+/**
+ * rcuref_read - Read the number of held reference counts of a rcuref
+ * @ref:	Pointer to the reference count
+ *
+ * Return: The number of held references (0 ... N)
+ */
+static inline unsigned int rcuref_read(rcuref_t *ref)
+{
+	unsigned int c = atomic_read(&ref->refcnt);
+
+	/* Return 0 if within the DEAD zone. */
+	return c >= RCUREF_RELEASED ? 0 : c + 1;
+}
+
+extern __must_check bool rcuref_get_slowpath(rcuref_t *ref);
+
+/**
+ * rcuref_get - Acquire one reference on a rcuref reference count
+ * @ref:	Pointer to the reference count
+ *
+ * Similar to atomic_inc_not_zero() but saturates at RCUREF_MAXREF.
+ *
+ * Provides no memory ordering, it is assumed the caller has guaranteed the
+ * object memory to be stable (RCU, etc.). It does provide a control dependency
+ * and thereby orders future stores. See documentation in lib/rcuref.c
+ *
+ * Return:
+ *	False if the attempt to acquire a reference failed. This happens
+ *	when the last reference has been put already
+ *
+ *	True if a reference was successfully acquired
+ */
+static inline __must_check bool rcuref_get(rcuref_t *ref)
+{
+	/*
+	 * Unconditionally increase the reference count. The saturation and
+	 * dead zones provide enough tolerance for this.
+	 */
+	if (likely(!atomic_add_negative_relaxed(1, &ref->refcnt)))
+		return true;
+
+	/* Handle the cases inside the saturation and dead zones */
+	return rcuref_get_slowpath(ref);
+}
+
+extern __must_check bool rcuref_put_slowpath(rcuref_t *ref);
+
+/*
+ * Internal helper. Do not invoke directly.
+ */
+static __always_inline __must_check bool __rcuref_put(rcuref_t *ref)
+{
+	RCU_LOCKDEP_WARN(!rcu_read_lock_held() && preemptible(),
+			 "suspicious rcuref_put_rcusafe() usage");
+	/*
+	 * Unconditionally decrease the reference count. The saturation and
+	 * dead zones provide enough tolerance for this.
+	 */
+	if (likely(!atomic_add_negative_release(-1, &ref->refcnt)))
+		return false;
+
+	/*
+	 * Handle the last reference drop and cases inside the saturation
+	 * and dead zones.
+	 */
+	return rcuref_put_slowpath(ref);
+}
+
+/**
+ * rcuref_put_rcusafe -- Release one reference for a rcuref reference count RCU safe
+ * @ref:	Pointer to the reference count
+ *
+ * Provides release memory ordering, such that prior loads and stores are done
+ * before, and provides an acquire ordering on success such that free()
+ * must come after.
+ *
+ * Can be invoked from contexts, which guarantee that no grace period can
+ * happen which would free the object concurrently if the decrement drops
+ * the last reference and the slowpath races against a concurrent get() and
+ * put() pair. rcu_read_lock()'ed and atomic contexts qualify.
+ *
+ * Return:
+ *	True if this was the last reference with no future references
+ *	possible. This signals the caller that it can safely release the
+ *	object which is protected by the reference counter.
+ *
+ *	False if there are still active references or the put() raced
+ *	with a concurrent get()/put() pair. Caller is not allowed to
+ *	release the protected object.
+ */
+static inline __must_check bool rcuref_put_rcusafe(rcuref_t *ref)
+{
+	return __rcuref_put(ref);
+}
+
+/**
+ * rcuref_put -- Release one reference for a rcuref reference count
+ * @ref:	Pointer to the reference count
+ *
+ * Can be invoked from any context.
+ *
+ * Provides release memory ordering, such that prior loads and stores are done
+ * before, and provides an acquire ordering on success such that free()
+ * must come after.
+ *
+ * Return:
+ *
+ *	True if this was the last reference with no future references
+ *	possible. This signals the caller that it can safely schedule the
+ *	object, which is protected by the reference counter, for
+ *	deconstruction.
+ *
+ *	False if there are still active references or the put() raced
+ *	with a concurrent get()/put() pair. Caller is not allowed to
+ *	deconstruct the protected object.
+ */
+static inline __must_check bool rcuref_put(rcuref_t *ref)
+{
+	bool released;
+
+	preempt_disable();
+	released = __rcuref_put(ref);
+	preempt_enable();
+	return released;
+}
+
+#endif
diff --git a/include/linux/types.h b/include/linux/types.h
index ea8cf60a8a79..688fb943556a 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -175,6 +175,12 @@ typedef struct {
 } atomic64_t;
 #endif
 
+typedef struct {
+	atomic_t refcnt;
+} rcuref_t;
+
+#define RCUREF_INIT(i)	{ .refcnt = ATOMIC_INIT(i - 1) }
+
 struct list_head {
 	struct list_head *next, *prev;
 };
diff --git a/lib/Makefile b/lib/Makefile
index baf2821f7a00..31a3a257fd49 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -47,7 +47,7 @@ obj-y += bcd.o sort.o parser.o debug_locks.o random32.o \
 	 list_sort.o uuid.o iov_iter.o clz_ctz.o \
 	 bsearch.o find_bit.o llist.o memweight.o kfifo.o \
 	 percpu-refcount.o rhashtable.o base64.o \
-	 once.o refcount.o usercopy.o errseq.o bucket_locks.o \
+	 once.o refcount.o rcuref.o usercopy.o errseq.o bucket_locks.o \
 	 generic-radix-tree.o
 obj-$(CONFIG_STRING_SELFTEST) += test_string.o
 obj-y += string_helpers.o
diff --git a/lib/rcuref.c b/lib/rcuref.c
new file mode 100644
index 000000000000..5ec00a4a64d1
--- /dev/null
+++ b/lib/rcuref.c
@@ -0,0 +1,281 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+/*
+ * rcuref - A scalable reference count implementation for RCU managed objects
+ *
+ * rcuref is provided to replace open coded reference count implementations
+ * based on atomic_t. It protects explicitely RCU managed objects which can
+ * be visible even after the last reference has been dropped and the object
+ * is heading towards destruction.
+ *
+ * A common usage pattern is:
+ *
+ * get()
+ *	rcu_read_lock();
+ *	p = get_ptr();
+ *	if (p && !atomic_inc_not_zero(&p->refcnt))
+ *		p = NULL;
+ *	rcu_read_unlock();
+ *	return p;
+ *
+ * put()
+ *	if (!atomic_dec_return(&->refcnt)) {
+ *		remove_ptr(p);
+ *		kfree_rcu((p, rcu);
+ *	}
+ *
+ * atomic_inc_not_zero() is implemented with a try_cmpxchg() loop which has
+ * O(N^2) behaviour under contention with N concurrent operations.
+ *
+ * rcuref uses atomic_add_negative_relaxed() for the fast path, which scales
+ * better under contention.
+ *
+ * Why not refcount?
+ * =================
+ *
+ * In principle it should be possible to make refcount use the rcuref
+ * scheme, but the destruction race described below cannot be prevented
+ * unless the protected object is RCU managed.
+ *
+ * Theory of operation
+ * ===================
+ *
+ * rcuref uses an unsigned integer reference counter. As long as the
+ * counter value is greater than or equal to RCUREF_ONEREF and not larger
+ * than RCUREF_MAXREF the reference is alive:
+ *
+ * ONEREF   MAXREF               SATURATED             RELEASED      DEAD    NOREF
+ * 0        0x7FFFFFFF 0x8000000 0xA0000000 0xBFFFFFFF 0xC0000000 0xE0000000 0xFFFFFFFF
+ * <---valid --------> <-------saturation zone-------> <-----dead zone----->
+ *
+ * The get() and put() operations do unconditional increments and
+ * decrements. The result is checked after the operation. This optimizes
+ * for the fast path.
+ *
+ * If the reference count is saturated or dead, then the increments and
+ * decrements are not harmful as the reference count still stays in the
+ * respective zones and is always set back to STATURATED resp. DEAD. The
+ * zones have room for 2^28 racing operations in each direction, which
+ * makes it practically impossible to escape the zones.
+ *
+ * Once the last reference is dropped the reference count becomes
+ * RCUREF_NOREF which forces rcuref_put() into the slowpath operation. The
+ * slowpath then tries to set the reference count from RCUREF_NOREF to
+ * RCUREF_DEAD via a cmpxchg(). This opens a small window where a
+ * concurrent rcuref_get() can acquire the reference count and bring it
+ * back to RCUREF_ONEREF or even drop the reference again and mark it DEAD.
+ *
+ * If the cmpxchg() succeeds then a concurrent rcuref_get() will result in
+ * DEAD + 1, which is inside the dead zone. If that happens the reference
+ * count is put back to DEAD.
+ *
+ * The actual race is possible due to the unconditional increment and
+ * decrements in rcuref_get() and rcuref_put():
+ *
+ *	T1				T2
+ *	get()				put()
+ *					if (atomic_add_negative(-1, &ref->refcnt))
+ *		succeeds->			atomic_cmpxchg(&ref->refcnt, NOREF, DEAD);
+ *
+ *	atomic_add_negative(1, &ref->refcnt);	<- Elevates refcount to DEAD + 1
+ *
+ * As the result of T1's add is negative, the get() goes into the slow path
+ * and observes refcnt being in the dead zone which makes the operation fail.
+ *
+ * Possible critical states:
+ *
+ *	Context Counter	References	Operation
+ *	T1	0	1		init()
+ *	T2	1	2		get()
+ *	T1	0	1		put()
+ *	T2     -1	0		put() tries to mark dead
+ *	T1	0	1		get()
+ *	T2	0	1		put() mark dead fails
+ *	T1     -1	0		put() tries to mark dead
+ *	T1    DEAD	0		put() mark dead succeeds
+ *	T2    DEAD+1	0		get() fails and puts it back to DEAD
+ *
+ * Of course there are more complex scenarios, but the above illustrates
+ * the working principle. The rest is left to the imagination of the
+ * reader.
+ *
+ * Deconstruction race
+ * ===================
+ *
+ * The release operation must be protected by prohibiting a grace period in
+ * order to prevent a possible use after free:
+ *
+ *	T1				T2
+ *	put()				get()
+ *	// ref->refcnt = ONEREF
+ *	if (!atomic_add_negative(-1, &ref->refcnt))
+ *		return false;				<- Not taken
+ *
+ *	// ref->refcnt == NOREF
+ *	--> preemption
+ *					// Elevates ref->refcnt to ONEREF
+ *					if (!atomic_add_negative(1, &ref->refcnt))
+ *						return true;			<- taken
+ *
+ *					if (put(&p->ref)) { <-- Succeeds
+ *						remove_pointer(p);
+ *						kfree_rcu(p, rcu);
+ *					}
+ *
+ *		RCU grace period ends, object is freed
+ *
+ *	atomic_cmpxchg(&ref->refcnt, NOREF, DEAD);	<- UAF
+ *
+ * This is prevented by disabling preemption around the put() operation as
+ * that's in most kernel configurations cheaper than a rcu_read_lock() /
+ * rcu_read_unlock() pair and in many cases even a NOOP. In any case it
+ * prevents the grace period which keeps the object alive until all put()
+ * operations complete.
+ *
+ * Saturation protection
+ * =====================
+ *
+ * The reference count has a saturation limit RCUREF_MAXREF (INT_MAX).
+ * Once this is exceedded the reference count becomes stale by setting it
+ * to RCUREF_SATURATED, which will cause a memory leak, but it prevents
+ * wrap arounds which obviously cause worse problems than a memory
+ * leak. When saturation is reached a warning is emitted.
+ *
+ * Race conditions
+ * ===============
+ *
+ * All reference count increment/decrement operations are unconditional and
+ * only verified after the fact. This optimizes for the good case and takes
+ * the occasional race vs. a dead or already saturated refcount into
+ * account. The saturation and dead zones are large enough to accomodate
+ * for that.
+ *
+ * Memory ordering
+ * ===============
+ *
+ * Memory ordering rules are slightly relaxed wrt regular atomic_t functions
+ * and provide only what is strictly required for refcounts.
+ *
+ * The increments are fully relaxed; these will not provide ordering. The
+ * rationale is that whatever is used to obtain the object to increase the
+ * reference count on will provide the ordering. For locked data
+ * structures, its the lock acquire, for RCU/lockless data structures its
+ * the dependent load.
+ *
+ * rcuref_get() provides a control dependency ordering future stores which
+ * ensures that the object is not modified when acquiring a reference
+ * fails.
+ *
+ * rcuref_put() provides release order, i.e. all prior loads and stores
+ * will be issued before. It also provides a control dependency ordering
+ * against the subsequent destruction of the object.
+ *
+ * If rcuref_put() successfully dropped the last reference and marked the
+ * object DEAD it also provides acquire ordering.
+ */
+
+#include <linux/export.h>
+#include <linux/rcuref.h>
+
+/**
+ * rcuref_get_slowpath - Slowpath of rcuref_get()
+ * @ref:	Pointer to the reference count
+ *
+ * Invoked when the reference count is outside of the valid zone.
+ *
+ * Return:
+ *	False if the reference count was already marked dead
+ *
+ *	True if the reference count is saturated, which prevents the
+ *	object from being deconstructed ever.
+ */
+bool rcuref_get_slowpath(rcuref_t *ref)
+{
+	unsigned int cnt = atomic_read(&ref->refcnt);
+
+	/*
+	 * If the reference count was already marked dead, undo the
+	 * increment so it stays in the middle of the dead zone and return
+	 * fail.
+	 */
+	if (cnt >= RCUREF_RELEASED) {
+		atomic_set(&ref->refcnt, RCUREF_DEAD);
+		return false;
+	}
+
+	/*
+	 * If it was saturated, warn and mark it so. In case the increment
+	 * was already on a saturated value restore the saturation
+	 * marker. This keeps it in the middle of the saturation zone and
+	 * prevents the reference count from overflowing. This leaks the
+	 * object memory, but prevents the obvious reference count overflow
+	 * damage.
+	 */
+	if (WARN_ONCE(cnt > RCUREF_MAXREF, "rcuref saturated - leaking memory"))
+		atomic_set(&ref->refcnt, RCUREF_SATURATED);
+	return true;
+}
+EXPORT_SYMBOL_GPL(rcuref_get_slowpath);
+
+/**
+ * rcuref_put_slowpath - Slowpath of __rcuref_put()
+ * @ref:	Pointer to the reference count
+ *
+ * Invoked when the reference count is outside of the valid zone.
+ *
+ * Return:
+ *	True if this was the last reference with no future references
+ *	possible. This signals the caller that it can safely schedule the
+ *	object, which is protected by the reference counter, for
+ *	deconstruction.
+ *
+ *	False if there are still active references or the put() raced
+ *	with a concurrent get()/put() pair. Caller is not allowed to
+ *	deconstruct the protected object.
+ */
+bool rcuref_put_slowpath(rcuref_t *ref)
+{
+	unsigned int cnt = atomic_read(&ref->refcnt);
+
+	/* Did this drop the last reference? */
+	if (likely(cnt == RCUREF_NOREF)) {
+		/*
+		 * Carefully try to set the reference count to RCUREF_DEAD.
+		 *
+		 * This can fail if a concurrent get() operation has
+		 * elevated it again or the corresponding put() even marked
+		 * it dead already. Both are valid situations and do not
+		 * require a retry. If this fails the caller is not
+		 * allowed to deconstruct the object.
+		 */
+		if (atomic_cmpxchg_release(&ref->refcnt, RCUREF_NOREF, RCUREF_DEAD) != RCUREF_NOREF)
+			return false;
+
+		/*
+		 * The caller can safely schedule the object for
+		 * deconstruction. Provide acquire ordering.
+		 */
+		smp_acquire__after_ctrl_dep();
+		return true;
+	}
+
+	/*
+	 * If the reference count was already in the dead zone, then this
+	 * put() operation is imbalanced. Warn, put the reference count back to
+	 * DEAD and tell the caller to not deconstruct the object.
+	 */
+	if (WARN_ONCE(cnt >= RCUREF_RELEASED, "rcuref - imbalanced put()")) {
+		atomic_set(&ref->refcnt, RCUREF_DEAD);
+		return false;
+	}
+
+	/*
+	 * This is a put() operation on a saturated refcount. Restore the
+	 * mean saturation value and tell the caller to not deconstruct the
+	 * object.
+	 */
+	if (cnt > RCUREF_MAXREF)
+		atomic_set(&ref->refcnt, RCUREF_SATURATED);
+	return false;
+}
+EXPORT_SYMBOL_GPL(rcuref_put_slowpath);
-- 
cgit v1.2.3-70-g09d2


From b341be6de98caf746100b2740ed2db68cd083261 Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Tue, 28 Mar 2023 09:33:24 +0200
Subject: dt-bindings: arm: stm32: add compatible for syscon gcan node

Since commit ad440432d1f9 ("dt-bindings: mfd: Ensure 'syscon' has a
more specific compatible") it is required to provide at least 2 compatibles
string for syscon node.
This patch documents the new compatible for stm32f4 SoC to support
global/shared CAN registers access for bxCAN controllers.

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/all/20230328073328.3949796-2-dario.binacchi@amarulasolutions.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml b/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml
index b2b156cc160a..ad8e51aa01b0 100644
--- a/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml
+++ b/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml
@@ -20,6 +20,7 @@ properties:
               - st,stm32-syscfg
               - st,stm32-power-config
               - st,stm32-tamp
+              - st,stm32f4-gcan
           - const: syscon
       - items:
           - const: st,stm32-tamp
@@ -42,6 +43,7 @@ if:
       contains:
         enum:
           - st,stm32mp157-syscfg
+          - st,stm32f4-gcan
 then:
   required:
     - clocks
-- 
cgit v1.2.3-70-g09d2


From e43250c0ac8123e4560ac39777cd94ab6e75ee23 Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Tue, 28 Mar 2023 09:33:25 +0200
Subject: dt-bindings: net: can: add STM32 bxcan DT bindings

Add documentation of device tree bindings for the STM32 basic extended
CAN (bxcan) controller.

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/all/20230328073328.3949796-3-dario.binacchi@amarulasolutions.com
[mkl: drop unneeded quotes]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 .../bindings/net/can/st,stm32-bxcan.yaml           | 85 ++++++++++++++++++++++
 1 file changed, 85 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml

diff --git a/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml b/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml
new file mode 100644
index 000000000000..769fa5c27b76
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml
@@ -0,0 +1,85 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/can/st,stm32-bxcan.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: STMicroelectronics bxCAN controller
+
+description: STMicroelectronics BxCAN controller for CAN bus
+
+maintainers:
+  - Dario Binacchi <dario.binacchi@amarulasolutions.com>
+
+allOf:
+  - $ref: can-controller.yaml#
+
+properties:
+  compatible:
+    enum:
+      - st,stm32f4-bxcan
+
+  st,can-primary:
+    description:
+      Primary and secondary mode of the bxCAN peripheral is only relevant
+      if the chip has two CAN peripherals. In that case they share some
+      of the required logic.
+      To avoid misunderstandings, it should be noted that ST documentation
+      uses the terms master/slave instead of primary/secondary.
+    type: boolean
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    items:
+      - description: transmit interrupt
+      - description: FIFO 0 receive interrupt
+      - description: FIFO 1 receive interrupt
+      - description: status change error interrupt
+
+  interrupt-names:
+    items:
+      - const: tx
+      - const: rx0
+      - const: rx1
+      - const: sce
+
+  resets:
+    maxItems: 1
+
+  clocks:
+    maxItems: 1
+
+  st,gcan:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    description:
+      The phandle to the gcan node which allows to access the 512-bytes
+      SRAM memory shared by the two bxCAN cells (CAN1 primary and CAN2
+      secondary) in dual CAN peripheral configuration.
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - resets
+  - clocks
+  - st,gcan
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/stm32fx-clock.h>
+    #include <dt-bindings/mfd/stm32f4-rcc.h>
+
+    can1: can@40006400 {
+        compatible = "st,stm32f4-bxcan";
+        reg = <0x40006400 0x200>;
+        interrupts = <19>, <20>, <21>, <22>;
+        interrupt-names = "tx", "rx0", "rx1", "sce";
+        resets = <&rcc STM32F4_APB1_RESET(CAN1)>;
+        clocks = <&rcc 0 STM32F4_APB1_CLOCK(CAN1)>;
+        st,can-primary;
+        st,gcan = <&gcan>;
+    };
-- 
cgit v1.2.3-70-g09d2


From 7355ad1950f41e755e6dc451834be3b94f82acd4 Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Tue, 28 Mar 2023 09:33:26 +0200
Subject: ARM: dts: stm32: add CAN support on stm32f429

Add support for bxcan (Basic eXtended CAN controller) to STM32F429. The
chip contains two CAN peripherals, CAN1 the primary and CAN2 the secondary,
that share some of the required logic like clock and filters. This means
that the secondary CAN can't be used without the primary CAN.

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Link: https://lore.kernel.org/all/20230328073328.3949796-4-dario.binacchi@amarulasolutions.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 arch/arm/boot/dts/stm32f429.dtsi | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/arch/arm/boot/dts/stm32f429.dtsi b/arch/arm/boot/dts/stm32f429.dtsi
index c31ceb821231..c9e05e3540d6 100644
--- a/arch/arm/boot/dts/stm32f429.dtsi
+++ b/arch/arm/boot/dts/stm32f429.dtsi
@@ -362,6 +362,35 @@
 			status = "disabled";
 		};
 
+		can1: can@40006400 {
+			compatible = "st,stm32f4-bxcan";
+			reg = <0x40006400 0x200>;
+			interrupts = <19>, <20>, <21>, <22>;
+			interrupt-names = "tx", "rx0", "rx1", "sce";
+			resets = <&rcc STM32F4_APB1_RESET(CAN1)>;
+			clocks = <&rcc 0 STM32F4_APB1_CLOCK(CAN1)>;
+			st,can-primary;
+			st,gcan = <&gcan>;
+			status = "disabled";
+		};
+
+		gcan: gcan@40006600 {
+			compatible = "st,stm32f4-gcan", "syscon";
+			reg = <0x40006600 0x200>;
+			clocks = <&rcc 0 STM32F4_APB1_CLOCK(CAN1)>;
+		};
+
+		can2: can@40006800 {
+			compatible = "st,stm32f4-bxcan";
+			reg = <0x40006800 0x200>;
+			interrupts = <63>, <64>, <65>, <66>;
+			interrupt-names = "tx", "rx0", "rx1", "sce";
+			resets = <&rcc STM32F4_APB1_RESET(CAN2)>;
+			clocks = <&rcc 0 STM32F4_APB1_CLOCK(CAN2)>;
+			st,gcan = <&gcan>;
+			status = "disabled";
+		};
+
 		dac: dac@40007400 {
 			compatible = "st,stm32f4-dac-core";
 			reg = <0x40007400 0x400>;
-- 
cgit v1.2.3-70-g09d2


From 559a6e75b4bcf0fc9e41d34865e72cf742f67d8e Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Tue, 28 Mar 2023 09:33:27 +0200
Subject: ARM: dts: stm32: add pin map for CAN controller on stm32f4

Add pin configurations for using CAN controller on stm32f469-disco
board. They are located on the Arduino compatible connector CN5 (CAN1)
and on the extension connector CN12 (CAN2).

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Link: https://lore.kernel.org/all/20230328073328.3949796-5-dario.binacchi@amarulasolutions.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 arch/arm/boot/dts/stm32f4-pinctrl.dtsi | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/arch/arm/boot/dts/stm32f4-pinctrl.dtsi b/arch/arm/boot/dts/stm32f4-pinctrl.dtsi
index 4523c63475e4..3bb812d6399e 100644
--- a/arch/arm/boot/dts/stm32f4-pinctrl.dtsi
+++ b/arch/arm/boot/dts/stm32f4-pinctrl.dtsi
@@ -447,6 +447,36 @@
 					slew-rate = <2>;
 				};
 			};
+
+			can1_pins_a: can1-0 {
+				pins1 {
+					pinmux = <STM32_PINMUX('B', 9, AF9)>; /* CAN1_TX */
+				};
+				pins2 {
+					pinmux = <STM32_PINMUX('B', 8, AF9)>; /* CAN1_RX */
+					bias-pull-up;
+				};
+			};
+
+			can2_pins_a: can2-0 {
+				pins1 {
+					pinmux = <STM32_PINMUX('B', 13, AF9)>; /* CAN2_TX */
+				};
+				pins2 {
+					pinmux = <STM32_PINMUX('B', 5, AF9)>; /* CAN2_RX */
+					bias-pull-up;
+				};
+			};
+
+			can2_pins_b: can2-1 {
+				pins1 {
+					pinmux = <STM32_PINMUX('B', 13, AF9)>; /* CAN2_TX */
+				};
+				pins2 {
+					pinmux = <STM32_PINMUX('B', 12, AF9)>; /* CAN2_RX */
+					bias-pull-up;
+				};
+			};
 		};
 	};
 };
-- 
cgit v1.2.3-70-g09d2


From f00647d8127be4d3f37f7e07dace24c04689ec63 Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Tue, 28 Mar 2023 09:33:28 +0200
Subject: can: bxcan: add support for ST bxCAN controller

Add support for the basic extended CAN controller (bxCAN) found in many
low- to middle-end STM32 SoCs. It supports the Basic Extended CAN
protocol versions 2.0A and B with a maximum bit rate of 1 Mbit/s.

The controller supports two channels (CAN1 as primary and CAN2 as
secondary) and the driver can enable either or both of the channels. They
share some of the required logic (e. g. clocks and filters), and that means
you cannot use the secondary CAN without enabling some hardware resources
managed by the primary CAN.

Each channel has 3 transmit mailboxes, 2 receive FIFOs with 3 stages and
28 scalable filter banks.
It also manages 4 dedicated interrupt vectors:
- transmit interrupt
- FIFO 0 receive interrupt
- FIFO 1 receive interrupt
- status change error interrupt

Driver uses all 3 available mailboxes for transmission and FIFO 0 for
reception. Rx filter rules are configured to the minimum. They accept
all messages and assign filter 0 to CAN1 and filter 14 to CAN2 in
identifier mask mode with 32 bits width. It enables and uses transmit,
receive buffers for FIFO 0 and error and status change interrupts.

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/20230328073328.3949796-6-dario.binacchi@amarulasolutions.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 MAINTAINERS              |    7 +
 drivers/net/can/Kconfig  |   12 +
 drivers/net/can/Makefile |    1 +
 drivers/net/can/bxcan.c  | 1098 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 1118 insertions(+)
 create mode 100644 drivers/net/can/bxcan.c

diff --git a/MAINTAINERS b/MAINTAINERS
index fbbda4671e73..6a784baa57e8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4431,6 +4431,13 @@ S:	Maintained
 F:	drivers/scsi/BusLogic.*
 F:	drivers/scsi/FlashPoint.*
 
+BXCAN CAN NETWORK DRIVER
+M:	Dario Binacchi <dario.binacchi@amarulasolutions.com>
+L:	linux-can@vger.kernel.org
+S:	Maintained
+F:	Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml
+F:	drivers/net/can/bxcan.c
+
 C-MEDIA CMI8788 DRIVER
 M:	Clemens Ladisch <clemens@ladisch.de>
 L:	alsa-devel@alsa-project.org (moderated for non-subscribers)
diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index cd34e8dc9394..3ceccafd701b 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -93,6 +93,18 @@ config CAN_AT91
 	  This is a driver for the SoC CAN controller in Atmel's AT91SAM9263
 	  and AT91SAM9X5 processors.
 
+config CAN_BXCAN
+	tristate "STM32 Basic Extended CAN (bxCAN) devices"
+	depends on OF || ARCH_STM32 || COMPILE_TEST
+	depends on HAS_IOMEM
+	select CAN_RX_OFFLOAD
+	help
+	  Say yes here to build support for the STMicroelectronics STM32 basic
+	  extended CAN Controller (bxCAN).
+
+	  This driver can also be built as a module. If so, the module
+	  will be called bxcan.
+
 config CAN_CAN327
 	tristate "Serial / USB serial ELM327 based OBD-II Interfaces (can327)"
 	depends on TTY
diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile
index 52b0f6e10668..ff8f76295d13 100644
--- a/drivers/net/can/Makefile
+++ b/drivers/net/can/Makefile
@@ -14,6 +14,7 @@ obj-y				+= usb/
 obj-y				+= softing/
 
 obj-$(CONFIG_CAN_AT91)		+= at91_can.o
+obj-$(CONFIG_CAN_BXCAN)		+= bxcan.o
 obj-$(CONFIG_CAN_CAN327)	+= can327.o
 obj-$(CONFIG_CAN_CC770)		+= cc770/
 obj-$(CONFIG_CAN_C_CAN)		+= c_can/
diff --git a/drivers/net/can/bxcan.c b/drivers/net/can/bxcan.c
new file mode 100644
index 000000000000..e26ccd41e3cb
--- /dev/null
+++ b/drivers/net/can/bxcan.c
@@ -0,0 +1,1098 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// bxcan.c - STM32 Basic Extended CAN controller driver
+//
+// Copyright (c) 2022 Dario Binacchi <dario.binacchi@amarulasolutions.com>
+//
+// NOTE: The ST documentation uses the terms master/slave instead of
+// primary/secondary.
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/bitfield.h>
+#include <linux/can.h>
+#include <linux/can/dev.h>
+#include <linux/can/error.h>
+#include <linux/can/rx-offload.h>
+#include <linux/clk.h>
+#include <linux/ethtool.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+
+#define BXCAN_NAPI_WEIGHT 3
+#define BXCAN_TIMEOUT_US 10000
+
+#define BXCAN_RX_MB_NUM 2
+#define BXCAN_TX_MB_NUM 3
+
+/* Primary control register (MCR) bits */
+#define BXCAN_MCR_RESET BIT(15)
+#define BXCAN_MCR_TTCM BIT(7)
+#define BXCAN_MCR_ABOM BIT(6)
+#define BXCAN_MCR_AWUM BIT(5)
+#define BXCAN_MCR_NART BIT(4)
+#define BXCAN_MCR_RFLM BIT(3)
+#define BXCAN_MCR_TXFP BIT(2)
+#define BXCAN_MCR_SLEEP BIT(1)
+#define BXCAN_MCR_INRQ BIT(0)
+
+/* Primary status register (MSR) bits */
+#define BXCAN_MSR_ERRI BIT(2)
+#define BXCAN_MSR_SLAK BIT(1)
+#define BXCAN_MSR_INAK BIT(0)
+
+/* Transmit status register (TSR) bits */
+#define BXCAN_TSR_RQCP2 BIT(16)
+#define BXCAN_TSR_RQCP1 BIT(8)
+#define BXCAN_TSR_RQCP0 BIT(0)
+
+/* Receive FIFO 0 register (RF0R) bits */
+#define BXCAN_RF0R_RFOM0 BIT(5)
+#define BXCAN_RF0R_FMP0_MASK GENMASK(1, 0)
+
+/* Interrupt enable register (IER) bits */
+#define BXCAN_IER_SLKIE BIT(17)
+#define BXCAN_IER_WKUIE BIT(16)
+#define BXCAN_IER_ERRIE BIT(15)
+#define BXCAN_IER_LECIE BIT(11)
+#define BXCAN_IER_BOFIE BIT(10)
+#define BXCAN_IER_EPVIE BIT(9)
+#define BXCAN_IER_EWGIE BIT(8)
+#define BXCAN_IER_FOVIE1 BIT(6)
+#define BXCAN_IER_FFIE1 BIT(5)
+#define BXCAN_IER_FMPIE1 BIT(4)
+#define BXCAN_IER_FOVIE0 BIT(3)
+#define BXCAN_IER_FFIE0 BIT(2)
+#define BXCAN_IER_FMPIE0 BIT(1)
+#define BXCAN_IER_TMEIE BIT(0)
+
+/* Error status register (ESR) bits */
+#define BXCAN_ESR_REC_MASK GENMASK(31, 24)
+#define BXCAN_ESR_TEC_MASK GENMASK(23, 16)
+#define BXCAN_ESR_LEC_MASK GENMASK(6, 4)
+#define BXCAN_ESR_BOFF BIT(2)
+#define BXCAN_ESR_EPVF BIT(1)
+#define BXCAN_ESR_EWGF BIT(0)
+
+/* Bit timing register (BTR) bits */
+#define BXCAN_BTR_SILM BIT(31)
+#define BXCAN_BTR_LBKM BIT(30)
+#define BXCAN_BTR_SJW_MASK GENMASK(25, 24)
+#define BXCAN_BTR_TS2_MASK GENMASK(22, 20)
+#define BXCAN_BTR_TS1_MASK GENMASK(19, 16)
+#define BXCAN_BTR_BRP_MASK GENMASK(9, 0)
+
+/* TX mailbox identifier register (TIxR, x = 0..2) bits */
+#define BXCAN_TIxR_STID_MASK GENMASK(31, 21)
+#define BXCAN_TIxR_EXID_MASK GENMASK(31, 3)
+#define BXCAN_TIxR_IDE BIT(2)
+#define BXCAN_TIxR_RTR BIT(1)
+#define BXCAN_TIxR_TXRQ BIT(0)
+
+/* TX mailbox data length and time stamp register (TDTxR, x = 0..2 bits */
+#define BXCAN_TDTxR_DLC_MASK GENMASK(3, 0)
+
+/* RX FIFO mailbox identifier register (RIxR, x = 0..1 */
+#define BXCAN_RIxR_STID_MASK GENMASK(31, 21)
+#define BXCAN_RIxR_EXID_MASK GENMASK(31, 3)
+#define BXCAN_RIxR_IDE BIT(2)
+#define BXCAN_RIxR_RTR BIT(1)
+
+/* RX FIFO mailbox data length and timestamp register (RDTxR, x = 0..1) bits */
+#define BXCAN_RDTxR_TIME_MASK GENMASK(31, 16)
+#define BXCAN_RDTxR_DLC_MASK GENMASK(3, 0)
+
+#define BXCAN_FMR_REG 0x00
+#define BXCAN_FM1R_REG 0x04
+#define BXCAN_FS1R_REG 0x0c
+#define BXCAN_FFA1R_REG 0x14
+#define BXCAN_FA1R_REG 0x1c
+#define BXCAN_FiR1_REG(b) (0x40 + (b) * 8)
+#define BXCAN_FiR2_REG(b) (0x44 + (b) * 8)
+
+#define BXCAN_FILTER_ID(primary) (primary ? 0 : 14)
+
+/* Filter primary register (FMR) bits */
+#define BXCAN_FMR_CANSB_MASK GENMASK(13, 8)
+#define BXCAN_FMR_FINIT BIT(0)
+
+enum bxcan_lec_code {
+	BXCAN_LEC_NO_ERROR = 0,
+	BXCAN_LEC_STUFF_ERROR,
+	BXCAN_LEC_FORM_ERROR,
+	BXCAN_LEC_ACK_ERROR,
+	BXCAN_LEC_BIT1_ERROR,
+	BXCAN_LEC_BIT0_ERROR,
+	BXCAN_LEC_CRC_ERROR,
+	BXCAN_LEC_UNUSED
+};
+
+/* Structure of the message buffer */
+struct bxcan_mb {
+	u32 id;			/* can identifier */
+	u32 dlc;		/* data length control and timestamp */
+	u32 data[2];		/* data */
+};
+
+/* Structure of the hardware registers */
+struct bxcan_regs {
+	u32 mcr;			/* 0x00 - primary control */
+	u32 msr;			/* 0x04 - primary status */
+	u32 tsr;			/* 0x08 - transmit status */
+	u32 rf0r;			/* 0x0c - FIFO 0 */
+	u32 rf1r;			/* 0x10 - FIFO 1 */
+	u32 ier;			/* 0x14 - interrupt enable */
+	u32 esr;			/* 0x18 - error status */
+	u32 btr;			/* 0x1c - bit timing*/
+	u32 reserved0[88];		/* 0x20 */
+	struct bxcan_mb tx_mb[BXCAN_TX_MB_NUM];	/* 0x180 - tx mailbox */
+	struct bxcan_mb rx_mb[BXCAN_RX_MB_NUM];	/* 0x1b0 - rx mailbox */
+};
+
+struct bxcan_priv {
+	struct can_priv can;
+	struct can_rx_offload offload;
+	struct device *dev;
+	struct net_device *ndev;
+
+	struct bxcan_regs __iomem *regs;
+	struct regmap *gcan;
+	int tx_irq;
+	int sce_irq;
+	bool primary;
+	struct clk *clk;
+	spinlock_t rmw_lock;	/* lock for read-modify-write operations */
+	unsigned int tx_head;
+	unsigned int tx_tail;
+	u32 timestamp;
+};
+
+static const struct can_bittiming_const bxcan_bittiming_const = {
+	.name = KBUILD_MODNAME,
+	.tseg1_min = 1,
+	.tseg1_max = 16,
+	.tseg2_min = 1,
+	.tseg2_max = 8,
+	.sjw_max = 4,
+	.brp_min = 1,
+	.brp_max = 1024,
+	.brp_inc = 1,
+};
+
+static inline void bxcan_rmw(struct bxcan_priv *priv, void __iomem *addr,
+			     u32 clear, u32 set)
+{
+	unsigned long flags;
+	u32 old, val;
+
+	spin_lock_irqsave(&priv->rmw_lock, flags);
+	old = readl(addr);
+	val = (old & ~clear) | set;
+	if (val != old)
+		writel(val, addr);
+
+	spin_unlock_irqrestore(&priv->rmw_lock, flags);
+}
+
+static void bxcan_disable_filters(struct bxcan_priv *priv, bool primary)
+{
+	unsigned int fid = BXCAN_FILTER_ID(primary);
+	u32 fmask = BIT(fid);
+
+	regmap_update_bits(priv->gcan, BXCAN_FA1R_REG, fmask, 0);
+}
+
+static void bxcan_enable_filters(struct bxcan_priv *priv, bool primary)
+{
+	unsigned int fid = BXCAN_FILTER_ID(primary);
+	u32 fmask = BIT(fid);
+
+	/* Filter settings:
+	 *
+	 * Accept all messages.
+	 * Assign filter 0 to CAN1 and filter 14 to CAN2 in identifier
+	 * mask mode with 32 bits width.
+	 */
+
+	/* Enter filter initialization mode and assing filters to CAN
+	 * controllers.
+	 */
+	regmap_update_bits(priv->gcan, BXCAN_FMR_REG,
+			   BXCAN_FMR_CANSB_MASK | BXCAN_FMR_FINIT,
+			   FIELD_PREP(BXCAN_FMR_CANSB_MASK, 14) |
+			   BXCAN_FMR_FINIT);
+
+	/* Deactivate filter */
+	regmap_update_bits(priv->gcan, BXCAN_FA1R_REG, fmask, 0);
+
+	/* Two 32-bit registers in identifier mask mode */
+	regmap_update_bits(priv->gcan, BXCAN_FM1R_REG, fmask, 0);
+
+	/* Single 32-bit scale configuration */
+	regmap_update_bits(priv->gcan, BXCAN_FS1R_REG, fmask, fmask);
+
+	/* Assign filter to FIFO 0 */
+	regmap_update_bits(priv->gcan, BXCAN_FFA1R_REG, fmask, 0);
+
+	/* Accept all messages */
+	regmap_write(priv->gcan, BXCAN_FiR1_REG(fid), 0);
+	regmap_write(priv->gcan, BXCAN_FiR2_REG(fid), 0);
+
+	/* Activate filter */
+	regmap_update_bits(priv->gcan, BXCAN_FA1R_REG, fmask, fmask);
+
+	/* Exit filter initialization mode */
+	regmap_update_bits(priv->gcan, BXCAN_FMR_REG, BXCAN_FMR_FINIT, 0);
+}
+
+static inline u8 bxcan_get_tx_head(const struct bxcan_priv *priv)
+{
+	return priv->tx_head % BXCAN_TX_MB_NUM;
+}
+
+static inline u8 bxcan_get_tx_tail(const struct bxcan_priv *priv)
+{
+	return priv->tx_tail % BXCAN_TX_MB_NUM;
+}
+
+static inline u8 bxcan_get_tx_free(const struct bxcan_priv *priv)
+{
+	return BXCAN_TX_MB_NUM - (priv->tx_head - priv->tx_tail);
+}
+
+static bool bxcan_tx_busy(const struct bxcan_priv *priv)
+{
+	if (bxcan_get_tx_free(priv) > 0)
+		return false;
+
+	netif_stop_queue(priv->ndev);
+
+	/* Memory barrier before checking tx_free (head and tail) */
+	smp_mb();
+
+	if (bxcan_get_tx_free(priv) == 0) {
+		netdev_dbg(priv->ndev,
+			   "Stopping tx-queue (tx_head=0x%08x, tx_tail=0x%08x, len=%d).\n",
+			   priv->tx_head, priv->tx_tail,
+			   priv->tx_head - priv->tx_tail);
+
+		return true;
+	}
+
+	netif_start_queue(priv->ndev);
+
+	return false;
+}
+
+static int bxcan_chip_softreset(struct bxcan_priv *priv)
+{
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 value;
+
+	bxcan_rmw(priv, &regs->mcr, 0, BXCAN_MCR_RESET);
+	return readx_poll_timeout(readl, &regs->msr, value,
+				  value & BXCAN_MSR_SLAK, BXCAN_TIMEOUT_US,
+				  USEC_PER_SEC);
+}
+
+static int bxcan_enter_init_mode(struct bxcan_priv *priv)
+{
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 value;
+
+	bxcan_rmw(priv, &regs->mcr, 0, BXCAN_MCR_INRQ);
+	return readx_poll_timeout(readl, &regs->msr, value,
+				  value & BXCAN_MSR_INAK, BXCAN_TIMEOUT_US,
+				  USEC_PER_SEC);
+}
+
+static int bxcan_leave_init_mode(struct bxcan_priv *priv)
+{
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 value;
+
+	bxcan_rmw(priv, &regs->mcr, BXCAN_MCR_INRQ, 0);
+	return readx_poll_timeout(readl, &regs->msr, value,
+				  !(value & BXCAN_MSR_INAK), BXCAN_TIMEOUT_US,
+				  USEC_PER_SEC);
+}
+
+static int bxcan_enter_sleep_mode(struct bxcan_priv *priv)
+{
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 value;
+
+	bxcan_rmw(priv, &regs->mcr, 0, BXCAN_MCR_SLEEP);
+	return readx_poll_timeout(readl, &regs->msr, value,
+				  value & BXCAN_MSR_SLAK, BXCAN_TIMEOUT_US,
+				  USEC_PER_SEC);
+}
+
+static int bxcan_leave_sleep_mode(struct bxcan_priv *priv)
+{
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 value;
+
+	bxcan_rmw(priv, &regs->mcr, BXCAN_MCR_SLEEP, 0);
+	return readx_poll_timeout(readl, &regs->msr, value,
+				  !(value & BXCAN_MSR_SLAK), BXCAN_TIMEOUT_US,
+				  USEC_PER_SEC);
+}
+
+static inline
+struct bxcan_priv *rx_offload_to_priv(struct can_rx_offload *offload)
+{
+	return container_of(offload, struct bxcan_priv, offload);
+}
+
+static struct sk_buff *bxcan_mailbox_read(struct can_rx_offload *offload,
+					  unsigned int mbxno, u32 *timestamp,
+					  bool drop)
+{
+	struct bxcan_priv *priv = rx_offload_to_priv(offload);
+	struct bxcan_regs __iomem *regs = priv->regs;
+	struct bxcan_mb __iomem *mb_regs = &regs->rx_mb[0];
+	struct sk_buff *skb = NULL;
+	struct can_frame *cf;
+	u32 rf0r, id, dlc;
+
+	rf0r = readl(&regs->rf0r);
+	if (unlikely(drop)) {
+		skb = ERR_PTR(-ENOBUFS);
+		goto mark_as_read;
+	}
+
+	if (!(rf0r & BXCAN_RF0R_FMP0_MASK))
+		goto mark_as_read;
+
+	skb = alloc_can_skb(offload->dev, &cf);
+	if (unlikely(!skb)) {
+		skb = ERR_PTR(-ENOMEM);
+		goto mark_as_read;
+	}
+
+	id = readl(&mb_regs->id);
+	if (id & BXCAN_RIxR_IDE)
+		cf->can_id = FIELD_GET(BXCAN_RIxR_EXID_MASK, id) | CAN_EFF_FLAG;
+	else
+		cf->can_id = FIELD_GET(BXCAN_RIxR_STID_MASK, id) & CAN_SFF_MASK;
+
+	dlc = readl(&mb_regs->dlc);
+	priv->timestamp = FIELD_GET(BXCAN_RDTxR_TIME_MASK, dlc);
+	cf->len = can_cc_dlc2len(FIELD_GET(BXCAN_RDTxR_DLC_MASK, dlc));
+
+	if (id & BXCAN_RIxR_RTR) {
+		cf->can_id |= CAN_RTR_FLAG;
+	} else {
+		int i, j;
+
+		for (i = 0, j = 0; i < cf->len; i += 4, j++)
+			*(u32 *)(cf->data + i) = readl(&mb_regs->data[j]);
+	}
+
+ mark_as_read:
+	rf0r |= BXCAN_RF0R_RFOM0;
+	writel(rf0r, &regs->rf0r);
+	return skb;
+}
+
+static irqreturn_t bxcan_rx_isr(int irq, void *dev_id)
+{
+	struct net_device *ndev = dev_id;
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 rf0r;
+
+	rf0r = readl(&regs->rf0r);
+	if (!(rf0r & BXCAN_RF0R_FMP0_MASK))
+		return IRQ_NONE;
+
+	can_rx_offload_irq_offload_fifo(&priv->offload);
+	can_rx_offload_irq_finish(&priv->offload);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t bxcan_tx_isr(int irq, void *dev_id)
+{
+	struct net_device *ndev = dev_id;
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct bxcan_regs __iomem *regs = priv->regs;
+	struct net_device_stats *stats = &ndev->stats;
+	u32 tsr, rqcp_bit;
+	int idx;
+
+	tsr = readl(&regs->tsr);
+	if (!(tsr & (BXCAN_TSR_RQCP0 | BXCAN_TSR_RQCP1 | BXCAN_TSR_RQCP2)))
+		return IRQ_NONE;
+
+	while (priv->tx_head - priv->tx_tail > 0) {
+		idx = bxcan_get_tx_tail(priv);
+		rqcp_bit = BXCAN_TSR_RQCP0 << (idx << 3);
+		if (!(tsr & rqcp_bit))
+			break;
+
+		stats->tx_packets++;
+		stats->tx_bytes += can_get_echo_skb(ndev, idx, NULL);
+		priv->tx_tail++;
+	}
+
+	writel(tsr, &regs->tsr);
+
+	if (bxcan_get_tx_free(priv)) {
+		/* Make sure that anybody stopping the queue after
+		 * this sees the new tx_ring->tail.
+		 */
+		smp_mb();
+		netif_wake_queue(ndev);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static void bxcan_handle_state_change(struct net_device *ndev, u32 esr)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	enum can_state new_state = priv->can.state;
+	struct can_berr_counter bec;
+	enum can_state rx_state, tx_state;
+	struct sk_buff *skb;
+	struct can_frame *cf;
+
+	/* Early exit if no error flag is set */
+	if (!(esr & (BXCAN_ESR_EWGF | BXCAN_ESR_EPVF | BXCAN_ESR_BOFF)))
+		return;
+
+	bec.txerr = FIELD_GET(BXCAN_ESR_TEC_MASK, esr);
+	bec.rxerr = FIELD_GET(BXCAN_ESR_REC_MASK, esr);
+
+	if (esr & BXCAN_ESR_BOFF)
+		new_state = CAN_STATE_BUS_OFF;
+	else if (esr & BXCAN_ESR_EPVF)
+		new_state = CAN_STATE_ERROR_PASSIVE;
+	else if (esr & BXCAN_ESR_EWGF)
+		new_state = CAN_STATE_ERROR_WARNING;
+
+	/* state hasn't changed */
+	if (unlikely(new_state == priv->can.state))
+		return;
+
+	skb = alloc_can_err_skb(ndev, &cf);
+
+	tx_state = bec.txerr >= bec.rxerr ? new_state : 0;
+	rx_state = bec.txerr <= bec.rxerr ? new_state : 0;
+	can_change_state(ndev, cf, tx_state, rx_state);
+
+	if (new_state == CAN_STATE_BUS_OFF) {
+		can_bus_off(ndev);
+	} else if (skb) {
+		cf->can_id |= CAN_ERR_CNT;
+		cf->data[6] = bec.txerr;
+		cf->data[7] = bec.rxerr;
+	}
+
+	if (skb) {
+		int err;
+
+		err = can_rx_offload_queue_timestamp(&priv->offload, skb,
+						     priv->timestamp);
+		if (err)
+			ndev->stats.rx_fifo_errors++;
+	}
+}
+
+static void bxcan_handle_bus_err(struct net_device *ndev, u32 esr)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	enum bxcan_lec_code lec_code;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+
+	lec_code = FIELD_GET(BXCAN_ESR_LEC_MASK, esr);
+
+	/* Early exit if no lec update or no error.
+	 * No lec update means that no CAN bus event has been detected
+	 * since CPU wrote BXCAN_LEC_UNUSED value to status reg.
+	 */
+	if (lec_code == BXCAN_LEC_UNUSED || lec_code == BXCAN_LEC_NO_ERROR)
+		return;
+
+	/* Common for all type of bus errors */
+	priv->can.can_stats.bus_error++;
+
+	/* Propagate the error condition to the CAN stack */
+	skb = alloc_can_err_skb(ndev, &cf);
+	if (skb)
+		cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
+
+	switch (lec_code) {
+	case BXCAN_LEC_STUFF_ERROR:
+		netdev_dbg(ndev, "Stuff error\n");
+		ndev->stats.rx_errors++;
+		if (skb)
+			cf->data[2] |= CAN_ERR_PROT_STUFF;
+		break;
+
+	case BXCAN_LEC_FORM_ERROR:
+		netdev_dbg(ndev, "Form error\n");
+		ndev->stats.rx_errors++;
+		if (skb)
+			cf->data[2] |= CAN_ERR_PROT_FORM;
+		break;
+
+	case BXCAN_LEC_ACK_ERROR:
+		netdev_dbg(ndev, "Ack error\n");
+		ndev->stats.tx_errors++;
+		if (skb) {
+			cf->can_id |= CAN_ERR_ACK;
+			cf->data[3] = CAN_ERR_PROT_LOC_ACK;
+		}
+		break;
+
+	case BXCAN_LEC_BIT1_ERROR:
+		netdev_dbg(ndev, "Bit error (recessive)\n");
+		ndev->stats.tx_errors++;
+		if (skb)
+			cf->data[2] |= CAN_ERR_PROT_BIT1;
+		break;
+
+	case BXCAN_LEC_BIT0_ERROR:
+		netdev_dbg(ndev, "Bit error (dominant)\n");
+		ndev->stats.tx_errors++;
+		if (skb)
+			cf->data[2] |= CAN_ERR_PROT_BIT0;
+		break;
+
+	case BXCAN_LEC_CRC_ERROR:
+		netdev_dbg(ndev, "CRC error\n");
+		ndev->stats.rx_errors++;
+		if (skb) {
+			cf->data[2] |= CAN_ERR_PROT_BIT;
+			cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
+		}
+		break;
+
+	default:
+		break;
+	}
+
+	if (skb) {
+		int err;
+
+		err = can_rx_offload_queue_timestamp(&priv->offload, skb,
+						     priv->timestamp);
+		if (err)
+			ndev->stats.rx_fifo_errors++;
+	}
+}
+
+static irqreturn_t bxcan_state_change_isr(int irq, void *dev_id)
+{
+	struct net_device *ndev = dev_id;
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 msr, esr;
+
+	msr = readl(&regs->msr);
+	if (!(msr & BXCAN_MSR_ERRI))
+		return IRQ_NONE;
+
+	esr = readl(&regs->esr);
+	bxcan_handle_state_change(ndev, esr);
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING)
+		bxcan_handle_bus_err(ndev, esr);
+
+	msr |= BXCAN_MSR_ERRI;
+	writel(msr, &regs->msr);
+	can_rx_offload_irq_finish(&priv->offload);
+
+	return IRQ_HANDLED;
+}
+
+static int bxcan_chip_start(struct net_device *ndev)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct bxcan_regs __iomem *regs = priv->regs;
+	struct can_bittiming *bt = &priv->can.bittiming;
+	u32 clr, set;
+	int err;
+
+	err = bxcan_chip_softreset(priv);
+	if (err) {
+		netdev_err(ndev, "failed to reset chip, error %pe\n",
+			   ERR_PTR(err));
+		return err;
+	}
+
+	err = bxcan_leave_sleep_mode(priv);
+	if (err) {
+		netdev_err(ndev, "failed to leave sleep mode, error %pe\n",
+			   ERR_PTR(err));
+		goto failed_leave_sleep;
+	}
+
+	err = bxcan_enter_init_mode(priv);
+	if (err) {
+		netdev_err(ndev, "failed to enter init mode, error %pe\n",
+			   ERR_PTR(err));
+		goto failed_enter_init;
+	}
+
+	/* MCR
+	 *
+	 * select request order priority
+	 * enable time triggered mode
+	 * bus-off state left on sw request
+	 * sleep mode left on sw request
+	 * retransmit automatically on error
+	 * do not lock RX FIFO on overrun
+	 */
+	bxcan_rmw(priv, &regs->mcr,
+		  BXCAN_MCR_ABOM | BXCAN_MCR_AWUM | BXCAN_MCR_NART |
+		  BXCAN_MCR_RFLM, BXCAN_MCR_TTCM | BXCAN_MCR_TXFP);
+
+	/* Bit timing register settings */
+	set = FIELD_PREP(BXCAN_BTR_BRP_MASK, bt->brp - 1) |
+		FIELD_PREP(BXCAN_BTR_TS1_MASK, bt->phase_seg1 +
+			   bt->prop_seg - 1) |
+		FIELD_PREP(BXCAN_BTR_TS2_MASK, bt->phase_seg2 - 1) |
+		FIELD_PREP(BXCAN_BTR_SJW_MASK, bt->sjw - 1);
+
+	/* loopback + silent mode put the controller in test mode,
+	 * useful for hot self-test
+	 */
+	if (priv->can.ctrlmode & CAN_CTRLMODE_LOOPBACK)
+		set |= BXCAN_BTR_LBKM;
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY)
+		set |= BXCAN_BTR_SILM;
+
+	bxcan_rmw(priv, &regs->btr, BXCAN_BTR_SILM | BXCAN_BTR_LBKM |
+		  BXCAN_BTR_BRP_MASK | BXCAN_BTR_TS1_MASK | BXCAN_BTR_TS2_MASK |
+		  BXCAN_BTR_SJW_MASK, set);
+
+	bxcan_enable_filters(priv, priv->primary);
+
+	/* Clear all internal status */
+	priv->tx_head = 0;
+	priv->tx_tail = 0;
+
+	err = bxcan_leave_init_mode(priv);
+	if (err) {
+		netdev_err(ndev, "failed to leave init mode, error %pe\n",
+			   ERR_PTR(err));
+		goto failed_leave_init;
+	}
+
+	/* Set a `lec` value so that we can check for updates later */
+	bxcan_rmw(priv, &regs->esr, BXCAN_ESR_LEC_MASK,
+		  FIELD_PREP(BXCAN_ESR_LEC_MASK, BXCAN_LEC_UNUSED));
+
+	/* IER
+	 *
+	 * Enable interrupt for:
+	 * bus-off
+	 * passive error
+	 * warning error
+	 * last error code
+	 * RX FIFO pending message
+	 * TX mailbox empty
+	 */
+	clr = BXCAN_IER_WKUIE | BXCAN_IER_SLKIE |  BXCAN_IER_FOVIE1 |
+		BXCAN_IER_FFIE1 | BXCAN_IER_FMPIE1 | BXCAN_IER_FOVIE0 |
+		BXCAN_IER_FFIE0;
+	set = BXCAN_IER_ERRIE | BXCAN_IER_BOFIE | BXCAN_IER_EPVIE |
+		BXCAN_IER_EWGIE | BXCAN_IER_FMPIE0 | BXCAN_IER_TMEIE;
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING)
+		set |= BXCAN_IER_LECIE;
+	else
+		clr |= BXCAN_IER_LECIE;
+
+	bxcan_rmw(priv, &regs->ier, clr, set);
+
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+	return 0;
+
+failed_leave_init:
+failed_enter_init:
+failed_leave_sleep:
+	bxcan_chip_softreset(priv);
+	return err;
+}
+
+static int bxcan_open(struct net_device *ndev)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	int err;
+
+	err = clk_prepare_enable(priv->clk);
+	if (err) {
+		netdev_err(ndev, "failed to enable clock, error %pe\n",
+			   ERR_PTR(err));
+		return err;
+	}
+
+	err = open_candev(ndev);
+	if (err) {
+		netdev_err(ndev, "open_candev() failed, error %pe\n",
+			   ERR_PTR(err));
+		goto out_disable_clock;
+	}
+
+	can_rx_offload_enable(&priv->offload);
+	err = request_irq(ndev->irq, bxcan_rx_isr, IRQF_SHARED, ndev->name,
+			  ndev);
+	if (err) {
+		netdev_err(ndev, "failed to register rx irq(%d), error %pe\n",
+			   ndev->irq, ERR_PTR(err));
+		goto out_close_candev;
+	}
+
+	err = request_irq(priv->tx_irq, bxcan_tx_isr, IRQF_SHARED, ndev->name,
+			  ndev);
+	if (err) {
+		netdev_err(ndev, "failed to register tx irq(%d), error %pe\n",
+			   priv->tx_irq, ERR_PTR(err));
+		goto out_free_rx_irq;
+	}
+
+	err = request_irq(priv->sce_irq, bxcan_state_change_isr, IRQF_SHARED,
+			  ndev->name, ndev);
+	if (err) {
+		netdev_err(ndev, "failed to register sce irq(%d), error %pe\n",
+			   priv->sce_irq, ERR_PTR(err));
+		goto out_free_tx_irq;
+	}
+
+	err = bxcan_chip_start(ndev);
+	if (err)
+		goto out_free_sce_irq;
+
+	netif_start_queue(ndev);
+	return 0;
+
+out_free_sce_irq:
+	free_irq(priv->sce_irq, ndev);
+out_free_tx_irq:
+	free_irq(priv->tx_irq, ndev);
+out_free_rx_irq:
+	free_irq(ndev->irq, ndev);
+out_close_candev:
+	can_rx_offload_disable(&priv->offload);
+	close_candev(ndev);
+out_disable_clock:
+	clk_disable_unprepare(priv->clk);
+	return err;
+}
+
+static void bxcan_chip_stop(struct net_device *ndev)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct bxcan_regs __iomem *regs = priv->regs;
+
+	/* disable all interrupts */
+	bxcan_rmw(priv, &regs->ier, BXCAN_IER_SLKIE | BXCAN_IER_WKUIE |
+		  BXCAN_IER_ERRIE | BXCAN_IER_LECIE | BXCAN_IER_BOFIE |
+		  BXCAN_IER_EPVIE | BXCAN_IER_EWGIE | BXCAN_IER_FOVIE1 |
+		  BXCAN_IER_FFIE1 | BXCAN_IER_FMPIE1 | BXCAN_IER_FOVIE0 |
+		  BXCAN_IER_FFIE0 | BXCAN_IER_FMPIE0 | BXCAN_IER_TMEIE, 0);
+	bxcan_disable_filters(priv, priv->primary);
+	bxcan_enter_sleep_mode(priv);
+	priv->can.state = CAN_STATE_STOPPED;
+}
+
+static int bxcan_stop(struct net_device *ndev)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+
+	netif_stop_queue(ndev);
+	bxcan_chip_stop(ndev);
+	free_irq(ndev->irq, ndev);
+	free_irq(priv->tx_irq, ndev);
+	free_irq(priv->sce_irq, ndev);
+	can_rx_offload_disable(&priv->offload);
+	close_candev(ndev);
+	clk_disable_unprepare(priv->clk);
+	return 0;
+}
+
+static netdev_tx_t bxcan_start_xmit(struct sk_buff *skb,
+				    struct net_device *ndev)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct can_frame *cf = (struct can_frame *)skb->data;
+	struct bxcan_regs __iomem *regs = priv->regs;
+	struct bxcan_mb __iomem *mb_regs;
+	unsigned int idx;
+	u32 id;
+	int i, j;
+
+	if (can_dropped_invalid_skb(ndev, skb))
+		return NETDEV_TX_OK;
+
+	if (bxcan_tx_busy(priv))
+		return NETDEV_TX_BUSY;
+
+	idx = bxcan_get_tx_head(priv);
+	priv->tx_head++;
+	if (bxcan_get_tx_free(priv) == 0)
+		netif_stop_queue(ndev);
+
+	mb_regs = &regs->tx_mb[idx];
+	if (cf->can_id & CAN_EFF_FLAG)
+		id = FIELD_PREP(BXCAN_TIxR_EXID_MASK, cf->can_id) |
+			BXCAN_TIxR_IDE;
+	else
+		id = FIELD_PREP(BXCAN_TIxR_STID_MASK, cf->can_id);
+
+	if (cf->can_id & CAN_RTR_FLAG) { /* Remote transmission request */
+		id |= BXCAN_TIxR_RTR;
+	} else {
+		for (i = 0, j = 0; i < cf->len; i += 4, j++)
+			writel(*(u32 *)(cf->data + i), &mb_regs->data[j]);
+	}
+
+	writel(FIELD_PREP(BXCAN_TDTxR_DLC_MASK, cf->len), &mb_regs->dlc);
+
+	can_put_echo_skb(skb, ndev, idx, 0);
+
+	/* Start transmission */
+	writel(id | BXCAN_TIxR_TXRQ, &mb_regs->id);
+
+	return NETDEV_TX_OK;
+}
+
+static const struct net_device_ops bxcan_netdev_ops = {
+	.ndo_open = bxcan_open,
+	.ndo_stop = bxcan_stop,
+	.ndo_start_xmit = bxcan_start_xmit,
+	.ndo_change_mtu = can_change_mtu,
+};
+
+static const struct ethtool_ops bxcan_ethtool_ops = {
+	.get_ts_info = ethtool_op_get_ts_info,
+};
+
+static int bxcan_do_set_mode(struct net_device *ndev, enum can_mode mode)
+{
+	int err;
+
+	switch (mode) {
+	case CAN_MODE_START:
+		err = bxcan_chip_start(ndev);
+		if (err)
+			return err;
+
+		netif_wake_queue(ndev);
+		break;
+
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int bxcan_get_berr_counter(const struct net_device *ndev,
+				  struct can_berr_counter *bec)
+{
+	struct bxcan_priv *priv = netdev_priv(ndev);
+	struct bxcan_regs __iomem *regs = priv->regs;
+	u32 esr;
+	int err;
+
+	err = clk_prepare_enable(priv->clk);
+	if (err)
+		return err;
+
+	esr = readl(&regs->esr);
+	bec->txerr = FIELD_GET(BXCAN_ESR_TEC_MASK, esr);
+	bec->rxerr = FIELD_GET(BXCAN_ESR_REC_MASK, esr);
+	clk_disable_unprepare(priv->clk);
+	return 0;
+}
+
+static int bxcan_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct device *dev = &pdev->dev;
+	struct net_device *ndev;
+	struct bxcan_priv *priv;
+	struct clk *clk = NULL;
+	void __iomem *regs;
+	struct regmap *gcan;
+	bool primary;
+	int err, rx_irq, tx_irq, sce_irq;
+
+	regs = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(regs)) {
+		dev_err(dev, "failed to get base address\n");
+		return PTR_ERR(regs);
+	}
+
+	gcan = syscon_regmap_lookup_by_phandle(np, "st,gcan");
+	if (IS_ERR(gcan)) {
+		dev_err(dev, "failed to get shared memory base address\n");
+		return PTR_ERR(gcan);
+	}
+
+	primary = of_property_read_bool(np, "st,can-primary");
+	clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(clk)) {
+		dev_err(dev, "failed to get clock\n");
+		return PTR_ERR(clk);
+	}
+
+	rx_irq = platform_get_irq_byname(pdev, "rx0");
+	if (rx_irq < 0) {
+		dev_err(dev, "failed to get rx0 irq\n");
+		return rx_irq;
+	}
+
+	tx_irq = platform_get_irq_byname(pdev, "tx");
+	if (tx_irq < 0) {
+		dev_err(dev, "failed to get tx irq\n");
+		return tx_irq;
+	}
+
+	sce_irq = platform_get_irq_byname(pdev, "sce");
+	if (sce_irq < 0) {
+		dev_err(dev, "failed to get sce irq\n");
+		return sce_irq;
+	}
+
+	ndev = alloc_candev(sizeof(struct bxcan_priv), BXCAN_TX_MB_NUM);
+	if (!ndev) {
+		dev_err(dev, "alloc_candev() failed\n");
+		return -ENOMEM;
+	}
+
+	priv = netdev_priv(ndev);
+	platform_set_drvdata(pdev, ndev);
+	SET_NETDEV_DEV(ndev, dev);
+	ndev->netdev_ops = &bxcan_netdev_ops;
+	ndev->ethtool_ops = &bxcan_ethtool_ops;
+	ndev->irq = rx_irq;
+	ndev->flags |= IFF_ECHO;
+
+	priv->dev = dev;
+	priv->ndev = ndev;
+	priv->regs = regs;
+	priv->gcan = gcan;
+	priv->clk = clk;
+	priv->tx_irq = tx_irq;
+	priv->sce_irq = sce_irq;
+	priv->primary = primary;
+	priv->can.clock.freq = clk_get_rate(clk);
+	spin_lock_init(&priv->rmw_lock);
+	priv->tx_head = 0;
+	priv->tx_tail = 0;
+	priv->can.bittiming_const = &bxcan_bittiming_const;
+	priv->can.do_set_mode = bxcan_do_set_mode;
+	priv->can.do_get_berr_counter = bxcan_get_berr_counter;
+	priv->can.ctrlmode_supported = CAN_CTRLMODE_LOOPBACK |
+		CAN_CTRLMODE_LISTENONLY	| CAN_CTRLMODE_BERR_REPORTING;
+
+	priv->offload.mailbox_read = bxcan_mailbox_read;
+	err = can_rx_offload_add_fifo(ndev, &priv->offload, BXCAN_NAPI_WEIGHT);
+	if (err) {
+		dev_err(dev, "failed to add FIFO rx_offload\n");
+		goto out_free_candev;
+	}
+
+	err = register_candev(ndev);
+	if (err) {
+		dev_err(dev, "failed to register netdev\n");
+		goto out_can_rx_offload_del;
+	}
+
+	dev_info(dev, "clk: %d Hz, IRQs: %d, %d, %d\n", priv->can.clock.freq,
+		 tx_irq, rx_irq, sce_irq);
+	return 0;
+
+out_can_rx_offload_del:
+	can_rx_offload_del(&priv->offload);
+out_free_candev:
+	free_candev(ndev);
+	return err;
+}
+
+static int bxcan_remove(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct bxcan_priv *priv = netdev_priv(ndev);
+
+	unregister_candev(ndev);
+	clk_disable_unprepare(priv->clk);
+	can_rx_offload_del(&priv->offload);
+	free_candev(ndev);
+	return 0;
+}
+
+static int __maybe_unused bxcan_suspend(struct device *dev)
+{
+	struct net_device *ndev = dev_get_drvdata(dev);
+	struct bxcan_priv *priv = netdev_priv(ndev);
+
+	if (!netif_running(ndev))
+		return 0;
+
+	netif_stop_queue(ndev);
+	netif_device_detach(ndev);
+
+	bxcan_enter_sleep_mode(priv);
+	priv->can.state = CAN_STATE_SLEEPING;
+	clk_disable_unprepare(priv->clk);
+	return 0;
+}
+
+static int __maybe_unused bxcan_resume(struct device *dev)
+{
+	struct net_device *ndev = dev_get_drvdata(dev);
+	struct bxcan_priv *priv = netdev_priv(ndev);
+
+	if (!netif_running(ndev))
+		return 0;
+
+	clk_prepare_enable(priv->clk);
+	bxcan_leave_sleep_mode(priv);
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+
+	netif_device_attach(ndev);
+	netif_start_queue(ndev);
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(bxcan_pm_ops, bxcan_suspend, bxcan_resume);
+
+static const struct of_device_id bxcan_of_match[] = {
+	{.compatible = "st,stm32f4-bxcan"},
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, bxcan_of_match);
+
+static struct platform_driver bxcan_driver = {
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.pm = &bxcan_pm_ops,
+		.of_match_table = bxcan_of_match,
+	},
+	.probe = bxcan_probe,
+	.remove = bxcan_remove,
+};
+
+module_platform_driver(bxcan_driver);
+
+MODULE_AUTHOR("Dario Binacchi <dario.binacchi@amarulasolutions.com>");
+MODULE_DESCRIPTION("STMicroelectronics Basic Extended CAN controller driver");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-70-g09d2


From b68ffb1b3bee3762f7579bdcc9ca3d415ed0b9b4 Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Sun, 26 Mar 2023 01:03:52 +0300
Subject: virtio/vsock: allocate multiple skbuffs on tx

This adds small optimization for tx path: instead of allocating single
skbuff on every call to transport, allocate multiple skbuff's until
credit space allows, thus trying to send as much as possible data without
return to af_vsock.c.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/vmw_vsock/virtio_transport_common.c | 57 +++++++++++++++++++++++++--------
 1 file changed, 43 insertions(+), 14 deletions(-)

diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index 957cdc01c8e8..8c60fc7f94a5 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -196,7 +196,8 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
 	const struct virtio_transport *t_ops;
 	struct virtio_vsock_sock *vvs;
 	u32 pkt_len = info->pkt_len;
-	struct sk_buff *skb;
+	u32 rest_len;
+	int ret;
 
 	info->type = virtio_transport_get_type(sk_vsock(vsk));
 
@@ -216,10 +217,6 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
 
 	vvs = vsk->trans;
 
-	/* we can send less than pkt_len bytes */
-	if (pkt_len > VIRTIO_VSOCK_MAX_PKT_BUF_SIZE)
-		pkt_len = VIRTIO_VSOCK_MAX_PKT_BUF_SIZE;
-
 	/* virtio_transport_get_credit might return less than pkt_len credit */
 	pkt_len = virtio_transport_get_credit(vvs, pkt_len);
 
@@ -227,17 +224,49 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
 	if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
 		return pkt_len;
 
-	skb = virtio_transport_alloc_skb(info, pkt_len,
-					 src_cid, src_port,
-					 dst_cid, dst_port);
-	if (!skb) {
-		virtio_transport_put_credit(vvs, pkt_len);
-		return -ENOMEM;
-	}
+	rest_len = pkt_len;
+
+	do {
+		struct sk_buff *skb;
+		size_t skb_len;
+
+		skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE, rest_len);
+
+		skb = virtio_transport_alloc_skb(info, skb_len,
+						 src_cid, src_port,
+						 dst_cid, dst_port);
+		if (!skb) {
+			ret = -ENOMEM;
+			break;
+		}
 
-	virtio_transport_inc_tx_pkt(vvs, skb);
+		virtio_transport_inc_tx_pkt(vvs, skb);
 
-	return t_ops->send_pkt(skb);
+		ret = t_ops->send_pkt(skb);
+		if (ret < 0)
+			break;
+
+		/* Both virtio and vhost 'send_pkt()' returns 'skb_len',
+		 * but for reliability use 'ret' instead of 'skb_len'.
+		 * Also if partial send happens (e.g. 'ret' != 'skb_len')
+		 * somehow, we break this loop, but account such returned
+		 * value in 'virtio_transport_put_credit()'.
+		 */
+		rest_len -= ret;
+
+		if (WARN_ONCE(ret != skb_len,
+			      "'send_pkt()' returns %i, but %zu expected\n",
+			      ret, skb_len))
+			break;
+	} while (rest_len);
+
+	virtio_transport_put_credit(vvs, rest_len);
+
+	/* Return number of bytes, if any data has been sent. */
+	if (rest_len != pkt_len)
+		ret = pkt_len - rest_len;
+
+	return ret;
 }
 
 static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs,
-- 
cgit v1.2.3-70-g09d2


From e3ec366eb0d1d5f7cbe28e747c76180f7d45f00d Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Sun, 26 Mar 2023 01:04:50 +0300
Subject: virtio/vsock: check argument to avoid no effect call

Both of these functions have no effect when input argument is 0, so to
avoid useless spinlock access, check argument before it.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/vmw_vsock/virtio_transport_common.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index 8c60fc7f94a5..7fc178c3ee07 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -302,6 +302,9 @@ u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 credit)
 {
 	u32 ret;
 
+	if (!credit)
+		return 0;
+
 	spin_lock_bh(&vvs->tx_lock);
 	ret = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt);
 	if (ret > credit)
@@ -315,6 +318,9 @@ EXPORT_SYMBOL_GPL(virtio_transport_get_credit);
 
 void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit)
 {
+	if (!credit)
+		return;
+
 	spin_lock_bh(&vvs->tx_lock);
 	vvs->tx_cnt -= credit;
 	spin_unlock_bh(&vvs->tx_lock);
-- 
cgit v1.2.3-70-g09d2


From 5c8560c4a19fc3e400715a347ce82bd909e3ac74 Mon Sep 17 00:00:00 2001
From: Grygorii Strashko <grygorii.strashko@ti.com>
Date: Mon, 27 Mar 2023 14:27:58 +0530
Subject: net: ethernet: ti: am65-cpsw: add .ndo to set dma per-queue rate

Enable rate limiting TX DMA queues for CPSW interface by configuring the
rate in absolute Mb/s units per TX queue.

Example:
    ethtool -L eth0 tx 4

    echo 100 > /sys/class/net/eth0/queues/tx-0/tx_maxrate
    echo 200 > /sys/class/net/eth0/queues/tx-1/tx_maxrate
    echo 50 > /sys/class/net/eth0/queues/tx-2/tx_maxrate
    echo 30 > /sys/class/net/eth0/queues/tx-3/tx_maxrate

    # disable
    echo 0 > /sys/class/net/eth0/queues/tx-0/tx_maxrate

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Link: https://lore.kernel.org/r/20230327085758.3237155-1-s-vadapalli@ti.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c |  12 +++-
 drivers/net/ethernet/ti/am65-cpsw-nuss.h |   2 +
 drivers/net/ethernet/ti/am65-cpsw-qos.c  | 113 +++++++++++++++++++++++++++++++
 drivers/net/ethernet/ti/am65-cpsw-qos.h  |   4 ++
 4 files changed, 129 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 9ddb79776c88..44368ecd994a 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -428,6 +428,8 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common)
 	else
 		am65_cpsw_init_host_port_switch(common);
 
+	am65_cpsw_qos_tx_p0_rate_init(common);
+
 	for (i = 0; i < common->rx_chns.descs_num; i++) {
 		skb = __netdev_alloc_skb_ip_align(NULL,
 						  AM65_CPSW_MAX_PACKET_SIZE,
@@ -599,8 +601,12 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
 		goto runtime_put;
 	}
 
-	for (i = 0; i < common->tx_ch_num; i++)
-		netdev_tx_reset_queue(netdev_get_tx_queue(ndev, i));
+	for (i = 0; i < common->tx_ch_num; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(ndev, i);
+
+		netdev_tx_reset_queue(txq);
+		txq->tx_maxrate =  common->tx_chns[i].rate_mbps;
+	}
 
 	ret = am65_cpsw_nuss_common_open(common);
 	if (ret)
@@ -1425,6 +1431,7 @@ static const struct net_device_ops am65_cpsw_nuss_netdev_ops = {
 	.ndo_vlan_rx_kill_vid	= am65_cpsw_nuss_ndo_slave_kill_vid,
 	.ndo_eth_ioctl		= am65_cpsw_nuss_ndo_slave_ioctl,
 	.ndo_setup_tc           = am65_cpsw_qos_ndo_setup_tc,
+	.ndo_set_tx_maxrate	= am65_cpsw_qos_ndo_tx_p0_set_maxrate,
 };
 
 static void am65_cpsw_disable_phy(struct phy *phy)
@@ -1616,6 +1623,7 @@ void am65_cpsw_nuss_remove_tx_chns(struct am65_cpsw_common *common)
 
 	devm_remove_action(dev, am65_cpsw_nuss_free_tx_chns, common);
 
+	common->tx_ch_rate_msk = 0;
 	for (i = 0; i < common->tx_ch_num; i++) {
 		struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i];
 
diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.h b/drivers/net/ethernet/ti/am65-cpsw-nuss.h
index cad04662739c..bf40c88fbd9b 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.h
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.h
@@ -79,6 +79,7 @@ struct am65_cpsw_tx_chn {
 	u32 id;
 	u32 descs_num;
 	char tx_chn_name[128];
+	u32 rate_mbps;
 };
 
 struct am65_cpsw_rx_chn {
@@ -126,6 +127,7 @@ struct am65_cpsw_common {
 	int			usage_count; /* number of opened ports */
 	struct cpsw_ale		*ale;
 	int			tx_ch_num;
+	u32			tx_ch_rate_msk;
 	u32			rx_flow_id_base;
 
 	struct am65_cpsw_tx_chn	tx_chns[AM65_CPSW_MAX_TX_QUEUES];
diff --git a/drivers/net/ethernet/ti/am65-cpsw-qos.c b/drivers/net/ethernet/ti/am65-cpsw-qos.c
index 8dc2c3085dcf..3a908db6e5b2 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-qos.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-qos.c
@@ -19,6 +19,7 @@
 #define AM65_CPSW_PN_REG_CTL			0x004
 #define AM65_CPSW_PN_REG_FIFO_STATUS		0x050
 #define AM65_CPSW_PN_REG_EST_CTL		0x060
+#define AM65_CPSW_PN_REG_PRI_CIR(pri)		(0x140 + 4 * (pri))
 
 /* AM65_CPSW_REG_CTL register fields */
 #define AM65_CPSW_CTL_EST_EN			BIT(18)
@@ -819,3 +820,115 @@ void am65_cpsw_qos_link_down(struct net_device *ndev)
 
 	port->qos.link_speed = SPEED_UNKNOWN;
 }
+
+static u32
+am65_cpsw_qos_tx_rate_calc(u32 rate_mbps, unsigned long bus_freq)
+{
+	u32 ir;
+
+	bus_freq /= 1000000;
+	ir = DIV_ROUND_UP(((u64)rate_mbps * 32768),  bus_freq);
+	return ir;
+}
+
+static void
+am65_cpsw_qos_tx_p0_rate_apply(struct am65_cpsw_common *common,
+			       int tx_ch, u32 rate_mbps)
+{
+	struct am65_cpsw_host *host = am65_common_get_host(common);
+	u32 ch_cir;
+	int i;
+
+	ch_cir = am65_cpsw_qos_tx_rate_calc(rate_mbps, common->bus_freq);
+	writel(ch_cir, host->port_base + AM65_CPSW_PN_REG_PRI_CIR(tx_ch));
+
+	/* update rates for every port tx queues */
+	for (i = 0; i < common->port_num; i++) {
+		struct net_device *ndev = common->ports[i].ndev;
+
+		if (!ndev)
+			continue;
+		netdev_get_tx_queue(ndev, tx_ch)->tx_maxrate = rate_mbps;
+	}
+}
+
+int am65_cpsw_qos_ndo_tx_p0_set_maxrate(struct net_device *ndev,
+					int queue, u32 rate_mbps)
+{
+	struct am65_cpsw_port *port = am65_ndev_to_port(ndev);
+	struct am65_cpsw_common *common = port->common;
+	struct am65_cpsw_tx_chn *tx_chn;
+	u32 ch_rate, tx_ch_rate_msk_new;
+	u32 ch_msk = 0;
+	int ret;
+
+	dev_dbg(common->dev, "apply TX%d rate limiting %uMbps tx_rate_msk%x\n",
+		queue, rate_mbps, common->tx_ch_rate_msk);
+
+	if (common->pf_p0_rx_ptype_rrobin) {
+		dev_err(common->dev, "TX Rate Limiting failed - rrobin mode\n");
+		return -EINVAL;
+	}
+
+	ch_rate = netdev_get_tx_queue(ndev, queue)->tx_maxrate;
+	if (ch_rate == rate_mbps)
+		return 0;
+
+	ret = pm_runtime_get_sync(common->dev);
+	if (ret < 0) {
+		pm_runtime_put_noidle(common->dev);
+		return ret;
+	}
+	ret = 0;
+
+	tx_ch_rate_msk_new = common->tx_ch_rate_msk;
+	if (rate_mbps && !(tx_ch_rate_msk_new & BIT(queue))) {
+		tx_ch_rate_msk_new |= BIT(queue);
+		ch_msk = GENMASK(common->tx_ch_num - 1, queue);
+		ch_msk = tx_ch_rate_msk_new ^ ch_msk;
+	} else if (!rate_mbps) {
+		tx_ch_rate_msk_new &= ~BIT(queue);
+		ch_msk = queue ? GENMASK(queue - 1, 0) : 0;
+		ch_msk = tx_ch_rate_msk_new & ch_msk;
+	}
+
+	if (ch_msk) {
+		dev_err(common->dev, "TX rate limiting has to be enabled sequentially hi->lo tx_rate_msk:%x tx_rate_msk_new:%x\n",
+			common->tx_ch_rate_msk, tx_ch_rate_msk_new);
+		ret = -EINVAL;
+		goto exit_put;
+	}
+
+	tx_chn = &common->tx_chns[queue];
+	tx_chn->rate_mbps = rate_mbps;
+	common->tx_ch_rate_msk = tx_ch_rate_msk_new;
+
+	if (!common->usage_count)
+		/* will be applied on next netif up */
+		goto exit_put;
+
+	am65_cpsw_qos_tx_p0_rate_apply(common, queue, rate_mbps);
+
+exit_put:
+	pm_runtime_put(common->dev);
+	return ret;
+}
+
+void am65_cpsw_qos_tx_p0_rate_init(struct am65_cpsw_common *common)
+{
+	struct am65_cpsw_host *host = am65_common_get_host(common);
+	int tx_ch;
+
+	for (tx_ch = 0; tx_ch < common->tx_ch_num; tx_ch++) {
+		struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[tx_ch];
+		u32 ch_cir;
+
+		if (!tx_chn->rate_mbps)
+			continue;
+
+		ch_cir = am65_cpsw_qos_tx_rate_calc(tx_chn->rate_mbps,
+						    common->bus_freq);
+		writel(ch_cir,
+		       host->port_base + AM65_CPSW_PN_REG_PRI_CIR(tx_ch));
+	}
+}
diff --git a/drivers/net/ethernet/ti/am65-cpsw-qos.h b/drivers/net/ethernet/ti/am65-cpsw-qos.h
index fb223b43b196..0cc2a3b3d7f9 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-qos.h
+++ b/drivers/net/ethernet/ti/am65-cpsw-qos.h
@@ -8,6 +8,8 @@
 #include <linux/netdevice.h>
 #include <net/pkt_sched.h>
 
+struct am65_cpsw_common;
+
 struct am65_cpsw_est {
 	int buf;
 	/* has to be the last one */
@@ -33,5 +35,7 @@ int am65_cpsw_qos_ndo_setup_tc(struct net_device *ndev, enum tc_setup_type type,
 			       void *type_data);
 void am65_cpsw_qos_link_up(struct net_device *ndev, int link_speed);
 void am65_cpsw_qos_link_down(struct net_device *ndev);
+int am65_cpsw_qos_ndo_tx_p0_set_maxrate(struct net_device *ndev, int queue, u32 rate_mbps);
+void am65_cpsw_qos_tx_p0_rate_init(struct am65_cpsw_common *common);
 
 #endif /* AM65_CPSW_QOS_H_ */
-- 
cgit v1.2.3-70-g09d2


From 86e2eca4ddedc07d639c44c990e1c220cac3741e Mon Sep 17 00:00:00 2001
From: Grygorii Strashko <grygorii.strashko@ti.com>
Date: Mon, 27 Mar 2023 14:51:03 +0530
Subject: net: ethernet: ti: am65-cpsw: enable p0 host port rx_vlan_remap

By default, the tagged ingress packets to the switch from the host port
P0 get internal switch priority assigned equal to the DMA CPPI channel
number they came from, unless CPSW_P0_CONTROL_REG.RX_REMAP_VLAN is enabled.
This causes issues with applying QoS policies and mapping packets on
external port fifos, because the default configuration is vlan_aware and
DMA CPPI channels are shared between all external ports.

Hence enable CPSW_P0_CONTROL_REG.RX_REMAP_VLAN so packet will preserve
internal switch priority assigned following the VLAN(priority) tag no
matter through which DMA CPPI Channels packets enter the switch.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Link: https://lore.kernel.org/r/20230327092103.3256118-1-s-vadapalli@ti.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 44368ecd994a..d17757ecbf42 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -86,6 +86,7 @@
 
 /* AM65_CPSW_P0_REG_CTL */
 #define AM65_CPSW_P0_REG_CTL_RX_CHECKSUM_EN	BIT(0)
+#define AM65_CPSW_P0_REG_CTL_RX_REMAP_VLAN	BIT(16)
 
 /* AM65_CPSW_PORT_REG_PRI_CTL */
 #define AM65_CPSW_PORT_REG_PRI_CTL_RX_PTYPE_RROBIN	BIT(8)
@@ -385,8 +386,8 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common)
 	/* set base flow_id */
 	writel(common->rx_flow_id_base,
 	       host_p->port_base + AM65_CPSW_PORT0_REG_FLOW_ID_OFFSET);
-	/* en tx crc offload */
-	writel(AM65_CPSW_P0_REG_CTL_RX_CHECKSUM_EN, host_p->port_base + AM65_CPSW_P0_REG_CTL);
+	writel(AM65_CPSW_P0_REG_CTL_RX_CHECKSUM_EN | AM65_CPSW_P0_REG_CTL_RX_REMAP_VLAN,
+	       host_p->port_base + AM65_CPSW_P0_REG_CTL);
 
 	am65_cpsw_nuss_set_p0_ptype(common);
 
-- 
cgit v1.2.3-70-g09d2


From 09df037017799b723d03a97c3f0de069bae9f625 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Fri, 27 Jan 2023 17:06:12 +0200
Subject: net/mlx5e: RX, Remove mlx5e_alloc_unit argument in page allocation

Change internal page cache and page pool api to use a struct page **
instead of a mlx5e_alloc_unit *.

This is the first change in a series which is meant to remove the
mlx5e_alloc_unit altogether.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 35 +++++++++++++------------
 1 file changed, 18 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 3f7b63d6616b..36300118b6e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -292,7 +292,7 @@ static inline bool mlx5e_rx_cache_put(struct mlx5e_rq *rq, struct page *page)
 	return true;
 }
 
-static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, union mlx5e_alloc_unit *au)
+static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, struct page **pagep)
 {
 	struct mlx5e_page_cache *cache = &rq->page_cache;
 	struct mlx5e_rq_stats *stats = rq->stats;
@@ -308,35 +308,35 @@ static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, union mlx5e_alloc_uni
 		return false;
 	}
 
-	au->page = cache->page_cache[cache->head];
+	*pagep = cache->page_cache[cache->head];
 	cache->head = (cache->head + 1) & (MLX5E_CACHE_SIZE - 1);
 	stats->cache_reuse++;
 
-	addr = page_pool_get_dma_addr(au->page);
+	addr = page_pool_get_dma_addr(*pagep);
 	/* Non-XSK always uses PAGE_SIZE. */
 	dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, rq->buff.map_dir);
 	return true;
 }
 
-static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, union mlx5e_alloc_unit *au)
+static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, struct page **pagep)
 {
 	dma_addr_t addr;
 
-	if (mlx5e_rx_cache_get(rq, au))
+	if (mlx5e_rx_cache_get(rq, pagep))
 		return 0;
 
-	au->page = page_pool_dev_alloc_pages(rq->page_pool);
-	if (unlikely(!au->page))
+	*pagep = page_pool_dev_alloc_pages(rq->page_pool);
+	if (unlikely(!*pagep))
 		return -ENOMEM;
 
 	/* Non-XSK always uses PAGE_SIZE. */
-	addr = dma_map_page(rq->pdev, au->page, 0, PAGE_SIZE, rq->buff.map_dir);
+	addr = dma_map_page(rq->pdev, *pagep, 0, PAGE_SIZE, rq->buff.map_dir);
 	if (unlikely(dma_mapping_error(rq->pdev, addr))) {
-		page_pool_recycle_direct(rq->page_pool, au->page);
-		au->page = NULL;
+		page_pool_recycle_direct(rq->page_pool, *pagep);
+		*pagep = NULL;
 		return -ENOMEM;
 	}
-	page_pool_set_dma_addr(au->page, addr);
+	page_pool_set_dma_addr(*pagep, addr);
 
 	return 0;
 }
@@ -376,7 +376,7 @@ static inline int mlx5e_get_rx_frag(struct mlx5e_rq *rq,
 		 * offset) should just use the new one without replenishing again
 		 * by themselves.
 		 */
-		err = mlx5e_page_alloc_pool(rq, frag->au);
+		err = mlx5e_page_alloc_pool(rq, &frag->au->page);
 
 	return err;
 }
@@ -605,13 +605,14 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 		header_offset = (index & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) <<
 			MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE;
 		if (!(header_offset & (PAGE_SIZE - 1))) {
-			union mlx5e_alloc_unit au;
 
-			err = mlx5e_page_alloc_pool(rq, &au);
+			err = mlx5e_page_alloc_pool(rq, &page);
 			if (unlikely(err))
 				goto err_unmap;
-			page = dma_info->page = au.page;
-			addr = dma_info->addr = page_pool_get_dma_addr(au.page);
+
+			addr = page_pool_get_dma_addr(page);
+			dma_info->addr = addr;
+			dma_info->page = page;
 		} else {
 			dma_info->addr = addr + header_offset;
 			dma_info->page = page;
@@ -715,7 +716,7 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	for (i = 0; i < rq->mpwqe.pages_per_wqe; i++, au++) {
 		dma_addr_t addr;
 
-		err = mlx5e_page_alloc_pool(rq, au);
+		err = mlx5e_page_alloc_pool(rq, &au->page);
 		if (unlikely(err))
 			goto err_unmap;
 		addr = page_pool_get_dma_addr(au->page);
-- 
cgit v1.2.3-70-g09d2


From 8fb1814f58f691373c692df75b761668069e197a Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Fri, 27 Jan 2023 16:58:52 +0200
Subject: net/mlx5e: RX, Remove alloc unit layout constraint for legacy rq

The mlx5e_alloc_unit union is conveniently used to store arrays of
pointers to struct page or struct xdp_buff (for xsk). The union is
currently expected to have the size of a pointer for xsk batch
allocations to work. This is conveniet for the current state of the
code but makes it impossible to add a structure of a different size
to the alloc unit.

A further patch in the series will add the mlx5e_frag_page struct for
which the described size constraint will no longer hold.

This change removes the usage of mlx5e_alloc_unit union for legacy rq:

- A union of arrays is introduced (mlx5e_alloc_units) to replace the
  array of unions to allow structures of different sizes.

- Each fragment has a pointer to a unit in the mlx5e_alloc_units array.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       | 12 +++-
 .../net/ethernet/mellanox/mlx5/core/en/xsk/rx.c    | 19 +++--
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  | 82 ++++++++++++++--------
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 36 +++++-----
 4 files changed, 87 insertions(+), 62 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 02237e630d13..32036f23d962 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -606,11 +606,19 @@ struct mlx5e_icosq {
 } ____cacheline_aligned_in_smp;
 
 struct mlx5e_wqe_frag_info {
-	union mlx5e_alloc_unit *au;
+	union {
+		struct page **pagep;
+		struct xdp_buff **xskp;
+	};
 	u32 offset;
 	bool last_in_page;
 };
 
+union mlx5e_alloc_units {
+	DECLARE_FLEX_ARRAY(struct page *, pages);
+	DECLARE_FLEX_ARRAY(struct xdp_buff *, xsk_buffs);
+};
+
 struct mlx5e_mpw_info {
 	u16 consumed_strides;
 	DECLARE_BITMAP(xdp_xmit_bitmap, MLX5_MPWRQ_MAX_PAGES_PER_WQE);
@@ -702,7 +710,7 @@ struct mlx5e_rq {
 		struct {
 			struct mlx5_wq_cyc          wq;
 			struct mlx5e_wqe_frag_info *frags;
-			union mlx5e_alloc_unit     *alloc_units;
+			union mlx5e_alloc_units    *alloc_units;
 			struct mlx5e_rq_frags_info  info;
 			mlx5e_fp_skb_from_cqe       skb_from_cqe;
 		} wqe;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
index fab787600459..8a5ae80e6142 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
@@ -163,13 +163,10 @@ int mlx5e_xsk_alloc_rx_wqes_batched(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 	u32 contig, alloc;
 	int i;
 
-	/* mlx5e_init_frags_partition creates a 1:1 mapping between
-	 * rq->wqe.frags and rq->wqe.alloc_units, which allows us to
-	 * allocate XDP buffers straight into alloc_units.
+	/* Each rq->wqe.frags->xskp is 1:1 mapped to an element inside the
+	 * rq->wqe.alloc_units->xsk_buffs array allocated here.
 	 */
-	BUILD_BUG_ON(sizeof(rq->wqe.alloc_units[0]) !=
-		     sizeof(rq->wqe.alloc_units[0].xsk));
-	buffs = (struct xdp_buff **)rq->wqe.alloc_units;
+	buffs = rq->wqe.alloc_units->xsk_buffs;
 	contig = mlx5_wq_cyc_get_size(wq) - ix;
 	if (wqe_bulk <= contig) {
 		alloc = xsk_buff_alloc_batch(rq->xsk_pool, buffs + ix, wqe_bulk);
@@ -189,7 +186,7 @@ int mlx5e_xsk_alloc_rx_wqes_batched(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 		/* Assumes log_num_frags == 0. */
 		frag = &rq->wqe.frags[j];
 
-		addr = xsk_buff_xdp_get_frame_dma(frag->au->xsk);
+		addr = xsk_buff_xdp_get_frame_dma(*frag->xskp);
 		wqe->data[0].addr = cpu_to_be64(addr + rq->buff.headroom);
 	}
 
@@ -211,11 +208,11 @@ int mlx5e_xsk_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 		/* Assumes log_num_frags == 0. */
 		frag = &rq->wqe.frags[j];
 
-		frag->au->xsk = xsk_buff_alloc(rq->xsk_pool);
-		if (unlikely(!frag->au->xsk))
+		*frag->xskp = xsk_buff_alloc(rq->xsk_pool);
+		if (unlikely(!*frag->xskp))
 			return i;
 
-		addr = xsk_buff_xdp_get_frame_dma(frag->au->xsk);
+		addr = xsk_buff_xdp_get_frame_dma(*frag->xskp);
 		wqe->data[0].addr = cpu_to_be64(addr + rq->buff.headroom);
 	}
 
@@ -306,7 +303,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq,
 					      struct mlx5_cqe64 *cqe,
 					      u32 cqe_bcnt)
 {
-	struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->au->xsk);
+	struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(*wi->xskp);
 	struct bpf_prog *prog;
 
 	/* wi->offset is not used in this function, because xdp->data and the
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 90eac1aa5f6b..917b98d1da2d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -499,15 +499,9 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 	struct mlx5e_wqe_frag_info *prev = NULL;
 	int i;
 
-	if (rq->xsk_pool) {
-		/* Assumptions used by XSK batched allocator. */
-		WARN_ON(rq->wqe.info.num_frags != 1);
-		WARN_ON(rq->wqe.info.log_num_frags != 0);
-		WARN_ON(rq->wqe.info.arr[0].frag_stride != PAGE_SIZE);
-	}
-
-	next_frag.au = &rq->wqe.alloc_units[0];
+	WARN_ON(rq->xsk_pool);
 
+	next_frag.pagep = &rq->wqe.alloc_units->pages[0];
 	for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++) {
 		struct mlx5e_rq_frag_info *frag_info = &rq->wqe.info.arr[0];
 		struct mlx5e_wqe_frag_info *frag =
@@ -516,7 +510,8 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 
 		for (f = 0; f < rq->wqe.info.num_frags; f++, frag++) {
 			if (next_frag.offset + frag_info[f].frag_stride > PAGE_SIZE) {
-				next_frag.au++;
+				/* Pages are assigned at runtime. */
+				next_frag.pagep++;
 				next_frag.offset = 0;
 				if (prev)
 					prev->last_in_page = true;
@@ -533,22 +528,59 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 		prev->last_in_page = true;
 }
 
-static int mlx5e_init_au_list(struct mlx5e_rq *rq, int wq_sz, int node)
+static void mlx5e_init_xsk_buffs(struct mlx5e_rq *rq)
+{
+	int i;
+
+	/* Assumptions used by XSK batched allocator. */
+	WARN_ON(rq->wqe.info.num_frags != 1);
+	WARN_ON(rq->wqe.info.log_num_frags != 0);
+	WARN_ON(rq->wqe.info.arr[0].frag_stride != PAGE_SIZE);
+
+	/* Considering the above assumptions a fragment maps to a single
+	 * xsk_buff.
+	 */
+	for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++)
+		rq->wqe.frags[i].xskp = &rq->wqe.alloc_units->xsk_buffs[i];
+}
+
+static int mlx5e_init_wqe_alloc_info(struct mlx5e_rq *rq, int node)
 {
+	int wq_sz = mlx5_wq_cyc_get_size(&rq->wqe.wq);
 	int len = wq_sz << rq->wqe.info.log_num_frags;
+	struct mlx5e_wqe_frag_info *frags;
+	union mlx5e_alloc_units *aus;
+	int aus_sz;
+
+	if (rq->xsk_pool)
+		aus_sz = sizeof(*aus->xsk_buffs);
+	else
+		aus_sz = sizeof(*aus->pages);
+
+	aus = kvzalloc_node(array_size(len, aus_sz), GFP_KERNEL, node);
+	if (!aus)
+		return -ENOMEM;
 
-	rq->wqe.alloc_units = kvzalloc_node(array_size(len, sizeof(*rq->wqe.alloc_units)),
-					    GFP_KERNEL, node);
-	if (!rq->wqe.alloc_units)
+	frags = kvzalloc_node(array_size(len, sizeof(*frags)), GFP_KERNEL, node);
+	if (!frags) {
+		kvfree(aus);
 		return -ENOMEM;
+	}
+
+	rq->wqe.alloc_units = aus;
+	rq->wqe.frags = frags;
 
-	mlx5e_init_frags_partition(rq);
+	if (rq->xsk_pool)
+		mlx5e_init_xsk_buffs(rq);
+	else
+		mlx5e_init_frags_partition(rq);
 
 	return 0;
 }
 
-static void mlx5e_free_au_list(struct mlx5e_rq *rq)
+static void mlx5e_free_wqe_alloc_info(struct mlx5e_rq *rq)
 {
+	kvfree(rq->wqe.frags);
 	kvfree(rq->wqe.alloc_units);
 }
 
@@ -778,18 +810,9 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		rq->wqe.info = rqp->frags_info;
 		rq->buff.frame0_sz = rq->wqe.info.arr[0].frag_stride;
 
-		rq->wqe.frags =
-			kvzalloc_node(array_size(sizeof(*rq->wqe.frags),
-					(wq_sz << rq->wqe.info.log_num_frags)),
-				      GFP_KERNEL, node);
-		if (!rq->wqe.frags) {
-			err = -ENOMEM;
-			goto err_rq_wq_destroy;
-		}
-
-		err = mlx5e_init_au_list(rq, wq_sz, node);
+		err = mlx5e_init_wqe_alloc_info(rq, node);
 		if (err)
-			goto err_rq_frags;
+			goto err_rq_wq_destroy;
 	}
 
 	if (xsk) {
@@ -888,9 +911,7 @@ err_rq_drop_page:
 		mlx5e_free_mpwqe_rq_drop_page(rq);
 		break;
 	default: /* MLX5_WQ_TYPE_CYCLIC */
-		mlx5e_free_au_list(rq);
-err_rq_frags:
-		kvfree(rq->wqe.frags);
+		mlx5e_free_wqe_alloc_info(rq);
 	}
 err_rq_wq_destroy:
 	mlx5_wq_destroy(&rq->wq_ctrl);
@@ -921,8 +942,7 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq)
 		mlx5e_rq_free_shampo(rq);
 		break;
 	default: /* MLX5_WQ_TYPE_CYCLIC */
-		kvfree(rq->wqe.frags);
-		mlx5e_free_au_list(rq);
+		mlx5e_free_wqe_alloc_info(rq);
 	}
 
 	for (i = rq->page_cache.head; i != rq->page_cache.tail;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 36300118b6e4..74e7e00cf494 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -371,12 +371,12 @@ static inline int mlx5e_get_rx_frag(struct mlx5e_rq *rq,
 	int err = 0;
 
 	if (!frag->offset)
-		/* On first frag (offset == 0), replenish page (alloc_unit actually).
-		 * Other frags that point to the same alloc_unit (with a different
+		/* On first frag (offset == 0), replenish page.
+		 * Other frags that point to the same page (with a different
 		 * offset) should just use the new one without replenishing again
 		 * by themselves.
 		 */
-		err = mlx5e_page_alloc_pool(rq, &frag->au->page);
+		err = mlx5e_page_alloc_pool(rq, frag->pagep);
 
 	return err;
 }
@@ -386,7 +386,7 @@ static inline void mlx5e_put_rx_frag(struct mlx5e_rq *rq,
 				     bool recycle)
 {
 	if (frag->last_in_page)
-		mlx5e_page_release_dynamic(rq, frag->au->page, recycle);
+		mlx5e_page_release_dynamic(rq, *frag->pagep, recycle);
 }
 
 static inline struct mlx5e_wqe_frag_info *get_frag(struct mlx5e_rq *rq, u16 ix)
@@ -410,7 +410,7 @@ static int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe_cyc *wqe,
 			goto free_frags;
 
 		headroom = i == 0 ? rq->buff.headroom : 0;
-		addr = page_pool_get_dma_addr(frag->au->page);
+		addr = page_pool_get_dma_addr(*frag->pagep);
 		wqe->data[i].addr = cpu_to_be64(addr + frag->offset + headroom);
 	}
 
@@ -434,7 +434,7 @@ static inline void mlx5e_free_rx_wqe(struct mlx5e_rq *rq,
 		 * put into the Reuse Ring, because there is no way to return
 		 * the page to the userspace when the interface goes down.
 		 */
-		xsk_buff_free(wi->au->xsk);
+		xsk_buff_free(*wi->xskp);
 		return;
 	}
 
@@ -1587,8 +1587,8 @@ static struct sk_buff *
 mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 			  struct mlx5_cqe64 *cqe, u32 cqe_bcnt)
 {
-	union mlx5e_alloc_unit *au = wi->au;
 	u16 rx_headroom = rq->buff.headroom;
+	struct page *page = *wi->pagep;
 	struct bpf_prog *prog;
 	struct sk_buff *skb;
 	u32 metasize = 0;
@@ -1596,11 +1596,11 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 	dma_addr_t addr;
 	u32 frag_size;
 
-	va             = page_address(au->page) + wi->offset;
+	va             = page_address(page) + wi->offset;
 	data           = va + rx_headroom;
 	frag_size      = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt);
 
-	addr = page_pool_get_dma_addr(au->page);
+	addr = page_pool_get_dma_addr(page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset,
 				      frag_size, rq->buff.map_dir);
 	net_prefetch(data);
@@ -1624,7 +1624,7 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 		return NULL;
 
 	/* queue up for recycling/reuse */
-	page_ref_inc(au->page);
+	page_ref_inc(page);
 
 	return skb;
 }
@@ -1635,8 +1635,8 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 {
 	struct mlx5e_rq_frag_info *frag_info = &rq->wqe.info.arr[0];
 	struct mlx5e_wqe_frag_info *head_wi = wi;
-	union mlx5e_alloc_unit *au = wi->au;
 	u16 rx_headroom = rq->buff.headroom;
+	struct page *page = *wi->pagep;
 	struct skb_shared_info *sinfo;
 	struct mlx5e_xdp_buff mxbuf;
 	u32 frag_consumed_bytes;
@@ -1646,10 +1646,10 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	u32 truesize;
 	void *va;
 
-	va = page_address(au->page) + wi->offset;
+	va = page_address(page) + wi->offset;
 	frag_consumed_bytes = min_t(u32, frag_info->frag_size, cqe_bcnt);
 
-	addr = page_pool_get_dma_addr(au->page);
+	addr = page_pool_get_dma_addr(page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset,
 				      rq->buff.frame0_sz, rq->buff.map_dir);
 	net_prefetchw(va); /* xdp_frame data area */
@@ -1666,11 +1666,11 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	while (cqe_bcnt) {
 		skb_frag_t *frag;
 
-		au = wi->au;
+		page = *wi->pagep;
 
 		frag_consumed_bytes = min_t(u32, frag_info->frag_size, cqe_bcnt);
 
-		addr = page_pool_get_dma_addr(au->page);
+		addr = page_pool_get_dma_addr(page);
 		dma_sync_single_for_cpu(rq->pdev, addr + wi->offset,
 					frag_consumed_bytes, rq->buff.map_dir);
 
@@ -1684,11 +1684,11 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 		}
 
 		frag = &sinfo->frags[sinfo->nr_frags++];
-		__skb_frag_set_page(frag, au->page);
+		__skb_frag_set_page(frag, page);
 		skb_frag_off_set(frag, wi->offset);
 		skb_frag_size_set(frag, frag_consumed_bytes);
 
-		if (page_is_pfmemalloc(au->page))
+		if (page_is_pfmemalloc(page))
 			xdp_buff_set_frag_pfmemalloc(&mxbuf.xdp);
 
 		sinfo->xdp_frags_size += frag_consumed_bytes;
@@ -1717,7 +1717,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	if (unlikely(!skb))
 		return NULL;
 
-	page_ref_inc(head_wi->au->page);
+	page_ref_inc(*head_wi->pagep);
 
 	if (xdp_buff_has_frags(&mxbuf.xdp)) {
 		int i;
-- 
cgit v1.2.3-70-g09d2


From d39092caaedf89c67170c035cad87c79988d1527 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Mon, 30 Jan 2023 12:02:45 +0200
Subject: net/mlx5e: RX, Remove alloc unit layout constraint for striding rq

This change removes the usage of mlx5e_alloc_unit union for
striding rq. The change is more straightforward than legacy rq as
the alloc units union is already in place.

This patch only moves things around: instead of an array of unions make
it a union of arrays. This has the effect that each mlx5e_mpw_info will
allocate the largest possible size of the array member. For xsk this
means that the array of xdp_buff pointers for the wqe will still be
contiguous, but there will be some extra unused bytes at the end of the
array.

As further patch in the series will add the mlx5e_frag_page struct for
which the described size constraint will no longer hold.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  7 +--
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h  |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en/xsk/rx.c    | 29 +++++-----
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  3 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 61 ++++++++++++----------
 5 files changed, 51 insertions(+), 51 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 32036f23d962..ad4ad14853bf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -475,11 +475,6 @@ struct mlx5e_txqsq {
 	cqe_ts_to_ns               ptp_cyc2time;
 } ____cacheline_aligned_in_smp;
 
-union mlx5e_alloc_unit {
-	struct page *page;
-	struct xdp_buff *xsk;
-};
-
 /* XDP packets can be transmitted in different ways. On completion, we need to
  * distinguish between them to clean up things in a proper way.
  */
@@ -622,7 +617,7 @@ union mlx5e_alloc_units {
 struct mlx5e_mpw_info {
 	u16 consumed_strides;
 	DECLARE_BITMAP(xdp_xmit_bitmap, MLX5_MPWRQ_MAX_PAGES_PER_WQE);
-	union mlx5e_alloc_unit alloc_units[];
+	union mlx5e_alloc_units alloc_units;
 };
 
 #define MLX5E_MAX_RX_FRAGS 4
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 816ea83e6413..dab00a2c2eb7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -489,7 +489,7 @@ static inline bool mlx5e_icosq_can_post_wqe(struct mlx5e_icosq *sq, u16 wqe_size
 
 static inline struct mlx5e_mpw_info *mlx5e_get_mpw_info(struct mlx5e_rq *rq, int i)
 {
-	size_t isz = struct_size(rq->mpwqe.info, alloc_units, rq->mpwqe.pages_per_wqe);
+	size_t isz = struct_size(rq->mpwqe.info, alloc_units.pages, rq->mpwqe.pages_per_wqe);
 
 	return (struct mlx5e_mpw_info *)((char *)rq->mpwqe.info + array_size(i, isz));
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
index 8a5ae80e6142..b2c1af07c317 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
@@ -22,6 +22,7 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	struct mlx5e_icosq *icosq = rq->icosq;
 	struct mlx5_wq_cyc *wq = &icosq->wq;
 	struct mlx5e_umr_wqe *umr_wqe;
+	struct xdp_buff **xsk_buffs;
 	int batch, i;
 	u32 offset; /* 17-bit value with MTT. */
 	u16 pi;
@@ -29,9 +30,9 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	if (unlikely(!xsk_buff_can_alloc(rq->xsk_pool, rq->mpwqe.pages_per_wqe)))
 		goto err;
 
-	BUILD_BUG_ON(sizeof(wi->alloc_units[0]) != sizeof(wi->alloc_units[0].xsk));
 	XSK_CHECK_PRIV_TYPE(struct mlx5e_xdp_buff);
-	batch = xsk_buff_alloc_batch(rq->xsk_pool, (struct xdp_buff **)wi->alloc_units,
+	xsk_buffs = (struct xdp_buff **)wi->alloc_units.xsk_buffs;
+	batch = xsk_buff_alloc_batch(rq->xsk_pool, xsk_buffs,
 				     rq->mpwqe.pages_per_wqe);
 
 	/* If batch < pages_per_wqe, either:
@@ -41,8 +42,8 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	 * the first error, which will mean there are no more valid descriptors.
 	 */
 	for (; batch < rq->mpwqe.pages_per_wqe; batch++) {
-		wi->alloc_units[batch].xsk = xsk_buff_alloc(rq->xsk_pool);
-		if (unlikely(!wi->alloc_units[batch].xsk))
+		xsk_buffs[batch] = xsk_buff_alloc(rq->xsk_pool);
+		if (unlikely(!xsk_buffs[batch]))
 			goto err_reuse_batch;
 	}
 
@@ -52,8 +53,8 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 
 	if (likely(rq->mpwqe.umr_mode == MLX5E_MPWRQ_UMR_MODE_ALIGNED)) {
 		for (i = 0; i < batch; i++) {
-			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->alloc_units[i].xsk);
-			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(wi->alloc_units[i].xsk);
+			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(xsk_buffs[i]);
+			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(xsk_buffs[i]);
 
 			umr_wqe->inline_mtts[i] = (struct mlx5_mtt) {
 				.ptag = cpu_to_be64(addr | MLX5_EN_WR),
@@ -62,8 +63,8 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 		}
 	} else if (unlikely(rq->mpwqe.umr_mode == MLX5E_MPWRQ_UMR_MODE_UNALIGNED)) {
 		for (i = 0; i < batch; i++) {
-			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->alloc_units[i].xsk);
-			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(wi->alloc_units[i].xsk);
+			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(xsk_buffs[i]);
+			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(xsk_buffs[i]);
 
 			umr_wqe->inline_ksms[i] = (struct mlx5_ksm) {
 				.key = rq->mkey_be,
@@ -75,8 +76,8 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 		u32 mapping_size = 1 << (rq->mpwqe.page_shift - 2);
 
 		for (i = 0; i < batch; i++) {
-			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->alloc_units[i].xsk);
-			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(wi->alloc_units[i].xsk);
+			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(xsk_buffs[i]);
+			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(xsk_buffs[i]);
 
 			umr_wqe->inline_ksms[i << 2] = (struct mlx5_ksm) {
 				.key = rq->mkey_be,
@@ -102,8 +103,8 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 		__be32 frame_size = cpu_to_be32(rq->xsk_pool->chunk_size);
 
 		for (i = 0; i < batch; i++) {
-			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->alloc_units[i].xsk);
-			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(wi->alloc_units[i].xsk);
+			struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(xsk_buffs[i]);
+			dma_addr_t addr = xsk_buff_xdp_get_frame_dma(xsk_buffs[i]);
 
 			umr_wqe->inline_klms[i << 1] = (struct mlx5_klm) {
 				.key = rq->mkey_be,
@@ -149,7 +150,7 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 
 err_reuse_batch:
 	while (--batch >= 0)
-		xsk_buff_free(wi->alloc_units[batch].xsk);
+		xsk_buff_free(xsk_buffs[batch]);
 
 err:
 	rq->stats->buff_alloc_err++;
@@ -248,7 +249,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq,
 						    u32 head_offset,
 						    u32 page_idx)
 {
-	struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->alloc_units[page_idx].xsk);
+	struct mlx5e_xdp_buff *mxbuf = xsk_buff_to_mxbuf(wi->alloc_units.xsk_buffs[page_idx]);
 	struct bpf_prog *prog;
 
 	/* Check packet size. Note LRO doesn't use linear SKB */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 917b98d1da2d..0ed3f67f7dfc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -286,7 +286,8 @@ static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, int node)
 	int wq_sz = mlx5_wq_ll_get_size(&rq->mpwqe.wq);
 	size_t alloc_size;
 
-	alloc_size = array_size(wq_sz, struct_size(rq->mpwqe.info, alloc_units,
+	alloc_size = array_size(wq_sz, struct_size(rq->mpwqe.info,
+						   alloc_units.pages,
 						   rq->mpwqe.pages_per_wqe));
 
 	rq->mpwqe.info = kvzalloc_node(alloc_size, GFP_KERNEL, node);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 74e7e00cf494..d02f2f2af4ec 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -469,16 +469,16 @@ static int mlx5e_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 
 static inline void
 mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb,
-		   union mlx5e_alloc_unit *au, u32 frag_offset, u32 len,
+		   struct page *page, u32 frag_offset, u32 len,
 		   unsigned int truesize)
 {
-	dma_addr_t addr = page_pool_get_dma_addr(au->page);
+	dma_addr_t addr = page_pool_get_dma_addr(page);
 
 	dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len,
 				rq->buff.map_dir);
-	page_ref_inc(au->page);
+	page_ref_inc(page);
 	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
-			au->page, frag_offset, len, truesize);
+			page, frag_offset, len, truesize);
 }
 
 static inline void
@@ -498,7 +498,6 @@ mlx5e_copy_skb_header(struct mlx5e_rq *rq, struct sk_buff *skb,
 static void
 mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle)
 {
-	union mlx5e_alloc_unit *alloc_units = wi->alloc_units;
 	bool no_xdp_xmit;
 	int i;
 
@@ -509,17 +508,21 @@ mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle
 	no_xdp_xmit = bitmap_empty(wi->xdp_xmit_bitmap, rq->mpwqe.pages_per_wqe);
 
 	if (rq->xsk_pool) {
+		struct xdp_buff **xsk_buffs = wi->alloc_units.xsk_buffs;
+
 		/* The `recycle` parameter is ignored, and the page is always
 		 * put into the Reuse Ring, because there is no way to return
 		 * the page to the userspace when the interface goes down.
 		 */
 		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++)
 			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap))
-				xsk_buff_free(alloc_units[i].xsk);
+				xsk_buff_free(xsk_buffs[i]);
 	} else {
+		struct page **pages = wi->alloc_units.pages;
+
 		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++)
 			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap))
-				mlx5e_page_release_dynamic(rq, alloc_units[i].page, recycle);
+				mlx5e_page_release_dynamic(rq, pages[i], recycle);
 	}
 }
 
@@ -694,7 +697,7 @@ static int mlx5e_alloc_rx_hd_mpwqe(struct mlx5e_rq *rq)
 static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 {
 	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, ix);
-	union mlx5e_alloc_unit *au = &wi->alloc_units[0];
+	struct page **pagep = &wi->alloc_units.pages[0];
 	struct mlx5e_icosq *sq = rq->icosq;
 	struct mlx5_wq_cyc *wq = &sq->wq;
 	struct mlx5e_umr_wqe *umr_wqe;
@@ -713,13 +716,13 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	umr_wqe = mlx5_wq_cyc_get_wqe(wq, pi);
 	memcpy(umr_wqe, &rq->mpwqe.umr_wqe, sizeof(struct mlx5e_umr_wqe));
 
-	for (i = 0; i < rq->mpwqe.pages_per_wqe; i++, au++) {
+	for (i = 0; i < rq->mpwqe.pages_per_wqe; i++, pagep++) {
 		dma_addr_t addr;
 
-		err = mlx5e_page_alloc_pool(rq, &au->page);
+		err = mlx5e_page_alloc_pool(rq, pagep);
 		if (unlikely(err))
 			goto err_unmap;
-		addr = page_pool_get_dma_addr(au->page);
+		addr = page_pool_get_dma_addr(*pagep);
 		umr_wqe->inline_mtts[i] = (struct mlx5_mtt) {
 			.ptag = cpu_to_be64(addr | MLX5_EN_WR),
 		};
@@ -760,8 +763,8 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 
 err_unmap:
 	while (--i >= 0) {
-		au--;
-		mlx5e_page_release_dynamic(rq, au->page, true);
+		pagep--;
+		mlx5e_page_release_dynamic(rq, *pagep, true);
 	}
 
 err:
@@ -1914,7 +1917,7 @@ const struct mlx5e_rx_handlers mlx5e_rx_handlers_rep = {
 
 static void
 mlx5e_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
-		    union mlx5e_alloc_unit *au, u32 data_bcnt, u32 data_offset)
+		    struct page **pagep, u32 data_bcnt, u32 data_offset)
 {
 	net_prefetchw(skb->data);
 
@@ -1928,12 +1931,12 @@ mlx5e_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
 		else
 			truesize = ALIGN(pg_consumed_bytes, BIT(rq->mpwqe.log_stride_sz));
 
-		mlx5e_add_skb_frag(rq, skb, au, data_offset,
+		mlx5e_add_skb_frag(rq, skb, *pagep, data_offset,
 				   pg_consumed_bytes, truesize);
 
 		data_bcnt -= pg_consumed_bytes;
 		data_offset = 0;
-		au++;
+		pagep++;
 	}
 }
 
@@ -1942,11 +1945,11 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 				   struct mlx5_cqe64 *cqe, u16 cqe_bcnt, u32 head_offset,
 				   u32 page_idx)
 {
-	union mlx5e_alloc_unit *au = &wi->alloc_units[page_idx];
+	struct page **pagep = &wi->alloc_units.pages[page_idx];
 	u16 headlen = min_t(u16, MLX5E_RX_MAX_HEAD, cqe_bcnt);
 	u32 frag_offset    = head_offset + headlen;
 	u32 byte_cnt       = cqe_bcnt - headlen;
-	union mlx5e_alloc_unit *head_au = au;
+	struct page *head_page = *pagep;
 	struct sk_buff *skb;
 	dma_addr_t addr;
 
@@ -1961,14 +1964,14 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 
 	/* Non-linear mode, hence non-XSK, which always uses PAGE_SIZE. */
 	if (unlikely(frag_offset >= PAGE_SIZE)) {
-		au++;
+		pagep++;
 		frag_offset -= PAGE_SIZE;
 	}
 
-	mlx5e_fill_skb_data(skb, rq, au, byte_cnt, frag_offset);
+	mlx5e_fill_skb_data(skb, rq, pagep, byte_cnt, frag_offset);
 	/* copy header */
-	addr = page_pool_get_dma_addr(head_au->page);
-	mlx5e_copy_skb_header(rq, skb, head_au->page, addr,
+	addr = page_pool_get_dma_addr(head_page);
+	mlx5e_copy_skb_header(rq, skb, head_page, addr,
 			      head_offset, head_offset, headlen);
 	/* skb linear part was allocated with headlen and aligned to long */
 	skb->tail += headlen;
@@ -1982,7 +1985,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 				struct mlx5_cqe64 *cqe, u16 cqe_bcnt, u32 head_offset,
 				u32 page_idx)
 {
-	union mlx5e_alloc_unit *au = &wi->alloc_units[page_idx];
+	struct page *page = wi->alloc_units.pages[page_idx];
 	u16 rx_headroom = rq->buff.headroom;
 	struct bpf_prog *prog;
 	struct sk_buff *skb;
@@ -1997,11 +2000,11 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		return NULL;
 	}
 
-	va             = page_address(au->page) + head_offset;
+	va             = page_address(page) + head_offset;
 	data           = va + rx_headroom;
 	frag_size      = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt);
 
-	addr = page_pool_get_dma_addr(au->page);
+	addr = page_pool_get_dma_addr(page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, head_offset,
 				      frag_size, rq->buff.map_dir);
 	net_prefetch(data);
@@ -2028,7 +2031,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		return NULL;
 
 	/* queue up for recycling/reuse */
-	page_ref_inc(au->page);
+	page_ref_inc(page);
 
 	return skb;
 }
@@ -2146,7 +2149,6 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	bool match		= cqe->shampo.match;
 	struct mlx5e_rq_stats *stats = rq->stats;
 	struct mlx5e_rx_wqe_ll *wqe;
-	union mlx5e_alloc_unit *au;
 	struct mlx5e_mpw_info *wi;
 	struct mlx5_wq_ll *wq;
 
@@ -2196,8 +2198,9 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	}
 
 	if (likely(head_size)) {
-		au = &wi->alloc_units[page_idx];
-		mlx5e_fill_skb_data(*skb, rq, au, data_bcnt, data_offset);
+		struct page **pagep = &wi->alloc_units.pages[page_idx];
+
+		mlx5e_fill_skb_data(*skb, rq, pagep, data_bcnt, data_offset);
 	}
 
 	mlx5e_shampo_complete_rx_cqe(rq, cqe, cqe_bcnt, *skb);
-- 
cgit v1.2.3-70-g09d2


From ca6ef9f031946134030e1b583e172c2e47c9a992 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 14 Mar 2023 21:05:56 +0200
Subject: net/mlx5e: RX, Store SHAMPO header pages in array

Save allocated SHAMPO header pages to an array to which the
mlx5e_dma_info page will point to.

This change is a preparation for introducing mlx5e_frag_page structure
in a downstream patch. There's no new functionality introduced.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      |  8 +++--
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 21 ++++++++----
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c   | 42 ++++++++++++++---------
 3 files changed, 45 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index ad4ad14853bf..b38fbacbb4d1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -670,13 +670,17 @@ struct mlx5e_rq_frags_info {
 
 struct mlx5e_dma_info {
 	dma_addr_t addr;
-	struct page *page;
+	union {
+		struct page **pagep;
+		struct page *page;
+	};
 };
 
 struct mlx5e_shampo_hd {
 	u32 mkey;
 	struct mlx5e_dma_info *info;
-	struct page *last_page;
+	struct page **pages;
+	u16 curr_page_index;
 	u16 hd_per_wq;
 	u16 hd_per_wqe;
 	unsigned long *bitmap;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 0ed3f67f7dfc..77f81d74ff30 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -262,23 +262,30 @@ static int mlx5e_rq_shampo_hd_info_alloc(struct mlx5e_rq *rq, int node)
 
 	shampo->bitmap = bitmap_zalloc_node(shampo->hd_per_wq, GFP_KERNEL,
 					    node);
-	if (!shampo->bitmap)
-		return -ENOMEM;
-
 	shampo->info = kvzalloc_node(array_size(shampo->hd_per_wq,
 						sizeof(*shampo->info)),
 				     GFP_KERNEL, node);
-	if (!shampo->info) {
-		kvfree(shampo->bitmap);
-		return -ENOMEM;
-	}
+	shampo->pages = kvzalloc_node(array_size(shampo->hd_per_wq,
+						 sizeof(*shampo->pages)),
+				     GFP_KERNEL, node);
+	if (!shampo->bitmap || !shampo->info || !shampo->pages)
+		goto err_nomem;
+
 	return 0;
+
+err_nomem:
+	kvfree(shampo->info);
+	kvfree(shampo->bitmap);
+	kvfree(shampo->pages);
+
+	return -ENOMEM;
 }
 
 static void mlx5e_rq_shampo_hd_info_free(struct mlx5e_rq *rq)
 {
 	kvfree(rq->mpwqe.shampo->bitmap);
 	kvfree(rq->mpwqe.shampo->info);
+	kvfree(rq->mpwqe.shampo->pages);
 }
 
 static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, int node)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index d02f2f2af4ec..7057db954f6f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -586,10 +586,11 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
 	u16 entries, pi, header_offset, err, wqe_bbs, new_entries;
 	u32 lkey = rq->mdev->mlx5e_res.hw_objs.mkey;
-	struct page *page = shampo->last_page;
+	u16 page_index = shampo->curr_page_index;
 	u64 addr = shampo->last_addr;
 	struct mlx5e_dma_info *dma_info;
 	struct mlx5e_umr_wqe *umr_wqe;
+	struct page **pagep;
 	int headroom, i;
 
 	headroom = rq->buff.headroom;
@@ -600,6 +601,8 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 	umr_wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi);
 	build_klm_umr(sq, umr_wqe, shampo->key, index, entries, wqe_bbs);
 
+	pagep = &shampo->pages[page_index];
+
 	for (i = 0; i < entries; i++, index++) {
 		dma_info = &shampo->info[index];
 		if (i >= klm_entries || (index < shampo->pi && shampo->pi - index <
@@ -608,17 +611,20 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 		header_offset = (index & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) <<
 			MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE;
 		if (!(header_offset & (PAGE_SIZE - 1))) {
+			page_index = (page_index + 1) & (shampo->hd_per_wq - 1);
+			pagep = &shampo->pages[page_index];
 
-			err = mlx5e_page_alloc_pool(rq, &page);
+			err = mlx5e_page_alloc_pool(rq, pagep);
 			if (unlikely(err))
 				goto err_unmap;
 
-			addr = page_pool_get_dma_addr(page);
+			addr = page_pool_get_dma_addr(*pagep);
+
 			dma_info->addr = addr;
-			dma_info->page = page;
+			dma_info->pagep = pagep;
 		} else {
 			dma_info->addr = addr + header_offset;
-			dma_info->page = page;
+			dma_info->pagep = pagep;
 		}
 
 update_klm:
@@ -636,7 +642,7 @@ update_klm:
 	};
 
 	shampo->pi = (shampo->pi + new_entries) & (shampo->hd_per_wq - 1);
-	shampo->last_page = page;
+	shampo->curr_page_index = page_index;
 	shampo->last_addr = addr;
 	sq->pc += wqe_bbs;
 	sq->doorbell_cseg = &umr_wqe->ctrl;
@@ -648,7 +654,7 @@ err_unmap:
 		dma_info = &shampo->info[--index];
 		if (!(i & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1))) {
 			dma_info->addr = ALIGN_DOWN(dma_info->addr, PAGE_SIZE);
-			mlx5e_page_release_dynamic(rq, dma_info->page, true);
+			mlx5e_page_release_dynamic(rq, *dma_info->pagep, true);
 		}
 	}
 	rq->stats->buff_alloc_err++;
@@ -783,7 +789,7 @@ void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close
 {
 	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
 	int hd_per_wq = shampo->hd_per_wq;
-	struct page *deleted_page = NULL;
+	struct page **deleted_page = NULL;
 	struct mlx5e_dma_info *hd_info;
 	int i, index = start;
 
@@ -796,9 +802,9 @@ void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close
 
 		hd_info = &shampo->info[index];
 		hd_info->addr = ALIGN_DOWN(hd_info->addr, PAGE_SIZE);
-		if (hd_info->page != deleted_page) {
-			deleted_page = hd_info->page;
-			mlx5e_page_release_dynamic(rq, hd_info->page, false);
+		if (hd_info->pagep != deleted_page) {
+			deleted_page = hd_info->pagep;
+			mlx5e_page_release_dynamic(rq, *hd_info->pagep, false);
 		}
 	}
 
@@ -1137,7 +1143,7 @@ static void *mlx5e_shampo_get_packet_hd(struct mlx5e_rq *rq, u16 header_index)
 	struct mlx5e_dma_info *last_head = &rq->mpwqe.shampo->info[header_index];
 	u16 head_offset = (last_head->addr & (PAGE_SIZE - 1)) + rq->buff.headroom;
 
-	return page_address(last_head->page) + head_offset;
+	return page_address(*last_head->pagep) + head_offset;
 }
 
 static void mlx5e_shampo_update_ipv4_udp_hdr(struct mlx5e_rq *rq, struct iphdr *ipv4)
@@ -2048,7 +2054,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 	void *hdr, *data;
 	u32 frag_size;
 
-	hdr		= page_address(head->page) + head_offset;
+	hdr		= page_address(*head->pagep) + head_offset;
 	data		= hdr + rx_headroom;
 	frag_size	= MLX5_SKB_FRAG_SZ(rx_headroom + head_size);
 
@@ -2063,7 +2069,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 			return NULL;
 
 		/* queue up for recycling/reuse */
-		page_ref_inc(head->page);
+		page_ref_inc(*head->pagep);
 
 	} else {
 		/* allocate SKB and copy header for large header */
@@ -2076,7 +2082,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		}
 
 		prefetchw(skb->data);
-		mlx5e_copy_skb_header(rq, skb, head->page, head->addr,
+		mlx5e_copy_skb_header(rq, skb, *head->pagep, head->addr,
 				      head_offset + rx_headroom,
 				      rx_headroom, head_size);
 		/* skb linear part was allocated with headlen and aligned to long */
@@ -2127,8 +2133,10 @@ mlx5e_free_rx_shampo_hd_entry(struct mlx5e_rq *rq, u16 header_index)
 	u64 addr = shampo->info[header_index].addr;
 
 	if (((header_index + 1) & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) == 0) {
-		shampo->info[header_index].addr = ALIGN_DOWN(addr, PAGE_SIZE);
-		mlx5e_page_release_dynamic(rq, shampo->info[header_index].page, true);
+		struct mlx5e_dma_info *dma_info = &shampo->info[header_index];
+
+		dma_info->addr = ALIGN_DOWN(addr, PAGE_SIZE);
+		mlx5e_page_release_dynamic(rq, *dma_info->pagep, true);
 	}
 	bitmap_clear(shampo->bitmap, header_index, 1);
 }
-- 
cgit v1.2.3-70-g09d2


From 08c9b61b071ca780ac2740b9de755d2ebac2a2e5 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 13 Dec 2022 14:37:07 +0200
Subject: net/mlx5e: RX, Remove internal page_cache

This patch removes the internal rx page_cache and uses the generic
page_pool api only. It used to be that the page_pool couldn't handle all
the mlx5 driver usecases, but with the introduction of skb recycling and
page fragmentaton in the page_pool full switch can now be made. Some
benfits of this transition:
* Better page recycling in the cases when the page_cache was suffering
  from head of queue blocking. The page_pool doesn't have this issue.
* DMA mapping/unmapping can be managed by the page_pool.
* mlx5e_rq size reduced by more than 50% due to the page_cache array
  being deleted.

This patch only removes the page_cache. Downstream patches will enable
the required page_pool features and will add further fine-tuning.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      |  6 ---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 13 ------
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c   | 53 -----------------------
 3 files changed, 72 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index b38fbacbb4d1..2684e7af5a7a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -628,11 +628,6 @@ struct mlx5e_mpw_info {
 #define MLX5E_CACHE_UNIT (MLX5_MPWRQ_MAX_PAGES_PER_WQE > NAPI_POLL_WEIGHT ? \
 			  MLX5_MPWRQ_MAX_PAGES_PER_WQE : NAPI_POLL_WEIGHT)
 #define MLX5E_CACHE_SIZE	(4 * roundup_pow_of_two(MLX5E_CACHE_UNIT))
-struct mlx5e_page_cache {
-	u32 head;
-	u32 tail;
-	struct page *page_cache[MLX5E_CACHE_SIZE];
-};
 
 struct mlx5e_rq;
 typedef void (*mlx5e_fp_handle_rx_cqe)(struct mlx5e_rq*, struct mlx5_cqe64*);
@@ -745,7 +740,6 @@ struct mlx5e_rq {
 	struct mlx5e_rq_stats *stats;
 	struct mlx5e_cq        cq;
 	struct mlx5e_cq_decomp cqd;
-	struct mlx5e_page_cache page_cache;
 	struct hwtstamp_config *tstamp;
 	struct mlx5_clock      *clock;
 	struct mlx5e_icosq    *icosq;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 77f81d74ff30..b0322a20b71b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -900,9 +900,6 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		rq->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
 	}
 
-	rq->page_cache.head = 0;
-	rq->page_cache.tail = 0;
-
 	return 0;
 
 err_destroy_page_pool:
@@ -933,7 +930,6 @@ err_rq_xdp_prog:
 static void mlx5e_free_rq(struct mlx5e_rq *rq)
 {
 	struct bpf_prog *old_prog;
-	int i;
 
 	if (xdp_rxq_info_is_reg(&rq->xdp_rxq)) {
 		old_prog = rcu_dereference_protected(rq->xdp_prog,
@@ -953,15 +949,6 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq)
 		mlx5e_free_wqe_alloc_info(rq);
 	}
 
-	for (i = rq->page_cache.head; i != rq->page_cache.tail;
-	     i = (i + 1) & (MLX5E_CACHE_SIZE - 1)) {
-		/* With AF_XDP, page_cache is not used, so this loop is not
-		 * entered, and it's safe to call mlx5e_page_release_dynamic
-		 * directly.
-		 */
-		mlx5e_page_release_dynamic(rq, rq->page_cache.page_cache[i], false);
-	}
-
 	xdp_rxq_info_unreg(&rq->xdp_rxq);
 	page_pool_destroy(rq->page_pool);
 	mlx5_wq_destroy(&rq->wq_ctrl);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 7057db954f6f..192f12a7d9a9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -271,60 +271,10 @@ static inline u32 mlx5e_decompress_cqes_start(struct mlx5e_rq *rq,
 	return mlx5e_decompress_cqes_cont(rq, wq, 1, budget_rem);
 }
 
-static inline bool mlx5e_rx_cache_put(struct mlx5e_rq *rq, struct page *page)
-{
-	struct mlx5e_page_cache *cache = &rq->page_cache;
-	u32 tail_next = (cache->tail + 1) & (MLX5E_CACHE_SIZE - 1);
-	struct mlx5e_rq_stats *stats = rq->stats;
-
-	if (tail_next == cache->head) {
-		stats->cache_full++;
-		return false;
-	}
-
-	if (!dev_page_is_reusable(page)) {
-		stats->cache_waive++;
-		return false;
-	}
-
-	cache->page_cache[cache->tail] = page;
-	cache->tail = tail_next;
-	return true;
-}
-
-static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, struct page **pagep)
-{
-	struct mlx5e_page_cache *cache = &rq->page_cache;
-	struct mlx5e_rq_stats *stats = rq->stats;
-	dma_addr_t addr;
-
-	if (unlikely(cache->head == cache->tail)) {
-		stats->cache_empty++;
-		return false;
-	}
-
-	if (page_ref_count(cache->page_cache[cache->head]) != 1) {
-		stats->cache_busy++;
-		return false;
-	}
-
-	*pagep = cache->page_cache[cache->head];
-	cache->head = (cache->head + 1) & (MLX5E_CACHE_SIZE - 1);
-	stats->cache_reuse++;
-
-	addr = page_pool_get_dma_addr(*pagep);
-	/* Non-XSK always uses PAGE_SIZE. */
-	dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, rq->buff.map_dir);
-	return true;
-}
-
 static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, struct page **pagep)
 {
 	dma_addr_t addr;
 
-	if (mlx5e_rx_cache_get(rq, pagep))
-		return 0;
-
 	*pagep = page_pool_dev_alloc_pages(rq->page_pool);
 	if (unlikely(!*pagep))
 		return -ENOMEM;
@@ -353,9 +303,6 @@ void mlx5e_page_dma_unmap(struct mlx5e_rq *rq, struct page *page)
 void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, struct page *page, bool recycle)
 {
 	if (likely(recycle)) {
-		if (mlx5e_rx_cache_put(rq, page))
-			return;
-
 		mlx5e_page_dma_unmap(rq, page);
 		page_pool_recycle_direct(rq->page_pool, page);
 	} else {
-- 
cgit v1.2.3-70-g09d2


From 4a5c5e25008f374e525178f5ba2581cfa3303a0c Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Wed, 14 Dec 2022 15:44:33 +0200
Subject: net/mlx5e: RX, Enable dma map and sync from page_pool allocator

Remove driver dma mapping and unmapping of pages. Let the
page_pool api do it.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h |  1 -
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c  |  2 --
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  6 ++++--
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c   | 22 ----------------------
 4 files changed, 4 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index dab00a2c2eb7..04419f56ac85 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -65,7 +65,6 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget);
 int mlx5e_poll_ico_cq(struct mlx5e_cq *cq);
 
 /* RX */
-void mlx5e_page_dma_unmap(struct mlx5e_rq *rq, struct page *page);
 void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, struct page *page, bool recycle);
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq));
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index c5dae48b7932..5e6ef602c748 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -209,8 +209,6 @@ bool mlx5e_xdp_handle(struct mlx5e_rq *rq,
 			goto xdp_abort;
 		__set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags);
 		__set_bit(MLX5E_RQ_FLAG_XDP_REDIRECT, rq->flags);
-		if (xdp->rxq->mem.type != MEM_TYPE_XSK_BUFF_POOL)
-			mlx5e_page_dma_unmap(rq, virt_to_page(xdp->data));
 		rq->stats->xdp_redirect++;
 		return true;
 	default:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index b0322a20b71b..2a73680021c2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -733,7 +733,6 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 			  struct mlx5e_rq_param *rqp,
 			  int node, struct mlx5e_rq *rq)
 {
-	struct page_pool_params pp_params = { 0 };
 	struct mlx5_core_dev *mdev = rq->mdev;
 	void *rqc = rqp->rqc;
 	void *rqc_wq = MLX5_ADDR_OF(rqc, rqc, wq);
@@ -829,12 +828,15 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		xsk_pool_set_rxq_info(rq->xsk_pool, &rq->xdp_rxq);
 	} else {
 		/* Create a page_pool and register it with rxq */
+		struct page_pool_params pp_params = { 0 };
+
 		pp_params.order     = 0;
-		pp_params.flags     = 0; /* No-internal DMA mapping in page_pool */
+		pp_params.flags     = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
 		pp_params.pool_size = pool_size;
 		pp_params.nid       = node;
 		pp_params.dev       = rq->pdev;
 		pp_params.dma_dir   = rq->buff.map_dir;
+		pp_params.max_len   = PAGE_SIZE;
 
 		/* page_pool can be used even when there is no rq->xdp_prog,
 		 * given page_pool does not handle DMA mapping there is no
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 192f12a7d9a9..01c789b89cb9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -273,40 +273,18 @@ static inline u32 mlx5e_decompress_cqes_start(struct mlx5e_rq *rq,
 
 static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, struct page **pagep)
 {
-	dma_addr_t addr;
-
 	*pagep = page_pool_dev_alloc_pages(rq->page_pool);
 	if (unlikely(!*pagep))
 		return -ENOMEM;
 
-	/* Non-XSK always uses PAGE_SIZE. */
-	addr = dma_map_page(rq->pdev, *pagep, 0, PAGE_SIZE, rq->buff.map_dir);
-	if (unlikely(dma_mapping_error(rq->pdev, addr))) {
-		page_pool_recycle_direct(rq->page_pool, *pagep);
-		*pagep = NULL;
-		return -ENOMEM;
-	}
-	page_pool_set_dma_addr(*pagep, addr);
-
 	return 0;
 }
 
-void mlx5e_page_dma_unmap(struct mlx5e_rq *rq, struct page *page)
-{
-	dma_addr_t dma_addr = page_pool_get_dma_addr(page);
-
-	dma_unmap_page_attrs(rq->pdev, dma_addr, PAGE_SIZE, rq->buff.map_dir,
-			     DMA_ATTR_SKIP_CPU_SYNC);
-	page_pool_set_dma_addr(page, 0);
-}
-
 void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, struct page *page, bool recycle)
 {
 	if (likely(recycle)) {
-		mlx5e_page_dma_unmap(rq, page);
 		page_pool_recycle_direct(rq->page_pool, page);
 	} else {
-		mlx5e_page_dma_unmap(rq, page);
 		page_pool_release_page(rq->page_pool, page);
 		put_page(page);
 	}
-- 
cgit v1.2.3-70-g09d2


From 6f5742846053c7656992e70726aad26a5129cf19 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Wed, 18 Jan 2023 17:08:51 +0200
Subject: net/mlx5e: RX, Enable skb page recycling through the page_pool

Start using the page_pool skb recycling api to recycle all pages back to
the page pool and stop using atomic page reference counting.

The mlx5e driver used to manage in-flight pages using page refcounting:
for each fragment there were 2 atomic write operations happening (one
for building the skb and one on skb release).

The page_pool api introduced a method to track page fragments more
optimally:
* The page's pp_fragment_count is set to a large bias on page alloc
  (1 x atomic write operation).
* The driver tracks the actual page fragments in a non atomic variable.
* When the skb is recycled, pp_fragment_count is decremented
  (atomic write operation).
* When page is released in the driver, the unused number of fragments
  (relative to the bias) is deducted from pp_fragment_count (atomic
  write operation).
* Last page defragmentation will only be an atomic read.

So in total there are `number of fragments + 1` atomic write ops. As
opposed to previously: `2 * frags` atomic writes ops.

Pages are wrapped in a mlx5e_frag_page structure which also contains the
number of fragments. This makes it easy to count the fragments in the
driver.

This change brings performance improvements for the case when the old rx
page_cache had low recycling rates due to head of queue blocking. For a
iperf3 TCP test with a single stream, on a single core (iperf and receive
queue running on same core), the following improvements can be noticed:

* Striding rq:
  - before (net-next baseline): bitrate = 30.1 Gbits/sec
  - after                     : bitrate = 31.4 Gbits/sec (diff: 4.14 %)

* Legacy rq:
  - before (net-next baseline): bitrate = 30.2 Gbits/sec
  - after                     : bitrate = 33.0 Gbits/sec (diff: 8.48 %)

There are 2 temporary performance degradations introduced:

1) TCP streams that had a good recycling rate with the old page_cache
   have a degradation for both striding and linear rq. This is due to
   very low page pool cache recycling: the pages are released during skb
   recycle which will release pages to the page pool ring for safety.
   The following patches in this series will tackle this problem by
   deferring the page release in the driver to increase the
   chance of having pages recycled to the cache.

2) XDP performance is now lower (4-5 %) due to the higher number of
   atomic operations used for fragment management. But this opens the
   door for supporting multiple packets per page in XDP, which will
   bring a big gain.

Otherwise, performance is similar to baseline.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      |  12 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h |   3 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c  |   3 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  11 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c   | 180 +++++++++++++---------
 5 files changed, 121 insertions(+), 88 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 2684e7af5a7a..0862556105d5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -600,9 +600,14 @@ struct mlx5e_icosq {
 	struct work_struct         recover_work;
 } ____cacheline_aligned_in_smp;
 
+struct mlx5e_frag_page {
+	struct page *page;
+	u16 frags;
+};
+
 struct mlx5e_wqe_frag_info {
 	union {
-		struct page **pagep;
+		struct mlx5e_frag_page *frag_page;
 		struct xdp_buff **xskp;
 	};
 	u32 offset;
@@ -610,6 +615,7 @@ struct mlx5e_wqe_frag_info {
 };
 
 union mlx5e_alloc_units {
+	DECLARE_FLEX_ARRAY(struct mlx5e_frag_page, frag_pages);
 	DECLARE_FLEX_ARRAY(struct page *, pages);
 	DECLARE_FLEX_ARRAY(struct xdp_buff *, xsk_buffs);
 };
@@ -666,7 +672,7 @@ struct mlx5e_rq_frags_info {
 struct mlx5e_dma_info {
 	dma_addr_t addr;
 	union {
-		struct page **pagep;
+		struct mlx5e_frag_page *frag_page;
 		struct page *page;
 	};
 };
@@ -674,7 +680,7 @@ struct mlx5e_dma_info {
 struct mlx5e_shampo_hd {
 	u32 mkey;
 	struct mlx5e_dma_info *info;
-	struct page **pages;
+	struct mlx5e_frag_page *pages;
 	u16 curr_page_index;
 	u16 hd_per_wq;
 	u16 hd_per_wqe;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 04419f56ac85..cd7779a9d046 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -65,7 +65,6 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget);
 int mlx5e_poll_ico_cq(struct mlx5e_cq *cq);
 
 /* RX */
-void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, struct page *page, bool recycle);
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq));
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq));
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
@@ -488,7 +487,7 @@ static inline bool mlx5e_icosq_can_post_wqe(struct mlx5e_icosq *sq, u16 wqe_size
 
 static inline struct mlx5e_mpw_info *mlx5e_get_mpw_info(struct mlx5e_rq *rq, int i)
 {
-	size_t isz = struct_size(rq->mpwqe.info, alloc_units.pages, rq->mpwqe.pages_per_wqe);
+	size_t isz = struct_size(rq->mpwqe.info, alloc_units.frag_pages, rq->mpwqe.pages_per_wqe);
 
 	return (struct mlx5e_mpw_info *)((char *)rq->mpwqe.info + array_size(i, isz));
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 5e6ef602c748..ca6ac9772d22 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -523,7 +523,8 @@ static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq,
 			break;
 		case MLX5E_XDP_XMIT_MODE_PAGE:
 			/* XDP_TX from the regular RQ */
-			mlx5e_page_release_dynamic(xdpi.page.rq, xdpi.page.page, recycle);
+			page_pool_put_defragged_page(xdpi.page.rq->page_pool,
+						     xdpi.page.page, -1, recycle);
 			break;
 		case MLX5E_XDP_XMIT_MODE_XSK:
 			/* AF_XDP send */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 2a73680021c2..eca9a11454e5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -294,7 +294,7 @@ static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, int node)
 	size_t alloc_size;
 
 	alloc_size = array_size(wq_sz, struct_size(rq->mpwqe.info,
-						   alloc_units.pages,
+						   alloc_units.frag_pages,
 						   rq->mpwqe.pages_per_wqe));
 
 	rq->mpwqe.info = kvzalloc_node(alloc_size, GFP_KERNEL, node);
@@ -509,7 +509,8 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 
 	WARN_ON(rq->xsk_pool);
 
-	next_frag.pagep = &rq->wqe.alloc_units->pages[0];
+	next_frag.frag_page = &rq->wqe.alloc_units->frag_pages[0];
+
 	for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++) {
 		struct mlx5e_rq_frag_info *frag_info = &rq->wqe.info.arr[0];
 		struct mlx5e_wqe_frag_info *frag =
@@ -519,7 +520,7 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 		for (f = 0; f < rq->wqe.info.num_frags; f++, frag++) {
 			if (next_frag.offset + frag_info[f].frag_stride > PAGE_SIZE) {
 				/* Pages are assigned at runtime. */
-				next_frag.pagep++;
+				next_frag.frag_page++;
 				next_frag.offset = 0;
 				if (prev)
 					prev->last_in_page = true;
@@ -563,7 +564,7 @@ static int mlx5e_init_wqe_alloc_info(struct mlx5e_rq *rq, int node)
 	if (rq->xsk_pool)
 		aus_sz = sizeof(*aus->xsk_buffs);
 	else
-		aus_sz = sizeof(*aus->pages);
+		aus_sz = sizeof(*aus->frag_pages);
 
 	aus = kvzalloc_node(array_size(len, aus_sz), GFP_KERNEL, node);
 	if (!aus)
@@ -831,7 +832,7 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		struct page_pool_params pp_params = { 0 };
 
 		pp_params.order     = 0;
-		pp_params.flags     = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+		pp_params.flags     = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV | PP_FLAG_PAGE_FRAG;
 		pp_params.pool_size = pool_size;
 		pp_params.nid       = node;
 		pp_params.dev       = rq->pdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 01c789b89cb9..7724e30ec133 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -271,23 +271,36 @@ static inline u32 mlx5e_decompress_cqes_start(struct mlx5e_rq *rq,
 	return mlx5e_decompress_cqes_cont(rq, wq, 1, budget_rem);
 }
 
-static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, struct page **pagep)
+#define MLX5E_PAGECNT_BIAS_MAX (PAGE_SIZE / 64)
+
+static int mlx5e_page_alloc_fragmented(struct mlx5e_rq *rq,
+				       struct mlx5e_frag_page *frag_page)
 {
-	*pagep = page_pool_dev_alloc_pages(rq->page_pool);
-	if (unlikely(!*pagep))
+	struct page *page;
+
+	page = page_pool_dev_alloc_pages(rq->page_pool);
+	if (unlikely(!page))
 		return -ENOMEM;
 
+	page_pool_fragment_page(page, MLX5E_PAGECNT_BIAS_MAX);
+
+	*frag_page = (struct mlx5e_frag_page) {
+		.page	= page,
+		.frags	= 0,
+	};
+
 	return 0;
 }
 
-void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, struct page *page, bool recycle)
+static void mlx5e_page_release_fragmented(struct mlx5e_rq *rq,
+					  struct mlx5e_frag_page *frag_page,
+					  bool recycle)
 {
-	if (likely(recycle)) {
-		page_pool_recycle_direct(rq->page_pool, page);
-	} else {
-		page_pool_release_page(rq->page_pool, page);
-		put_page(page);
-	}
+	u16 drain_count = MLX5E_PAGECNT_BIAS_MAX - frag_page->frags;
+	struct page *page = frag_page->page;
+
+	if (page_pool_defrag_page(page, drain_count) == 0)
+		page_pool_put_defragged_page(rq->page_pool, page, -1, recycle);
 }
 
 static inline int mlx5e_get_rx_frag(struct mlx5e_rq *rq,
@@ -301,7 +314,7 @@ static inline int mlx5e_get_rx_frag(struct mlx5e_rq *rq,
 		 * offset) should just use the new one without replenishing again
 		 * by themselves.
 		 */
-		err = mlx5e_page_alloc_pool(rq, frag->pagep);
+		err = mlx5e_page_alloc_fragmented(rq, frag->frag_page);
 
 	return err;
 }
@@ -311,7 +324,7 @@ static inline void mlx5e_put_rx_frag(struct mlx5e_rq *rq,
 				     bool recycle)
 {
 	if (frag->last_in_page)
-		mlx5e_page_release_dynamic(rq, *frag->pagep, recycle);
+		mlx5e_page_release_fragmented(rq, frag->frag_page, recycle);
 }
 
 static inline struct mlx5e_wqe_frag_info *get_frag(struct mlx5e_rq *rq, u16 ix)
@@ -335,7 +348,7 @@ static int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe_cyc *wqe,
 			goto free_frags;
 
 		headroom = i == 0 ? rq->buff.headroom : 0;
-		addr = page_pool_get_dma_addr(*frag->pagep);
+		addr = page_pool_get_dma_addr(frag->frag_page->page);
 		wqe->data[i].addr = cpu_to_be64(addr + frag->offset + headroom);
 	}
 
@@ -401,7 +414,6 @@ mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb,
 
 	dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len,
 				rq->buff.map_dir);
-	page_ref_inc(page);
 	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
 			page, frag_offset, len, truesize);
 }
@@ -443,11 +455,14 @@ mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle
 			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap))
 				xsk_buff_free(xsk_buffs[i]);
 	} else {
-		struct page **pages = wi->alloc_units.pages;
+		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++) {
+			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap)) {
+				struct mlx5e_frag_page *frag_page;
 
-		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++)
-			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap))
-				mlx5e_page_release_dynamic(rq, pages[i], recycle);
+				frag_page = &wi->alloc_units.frag_pages[i];
+				mlx5e_page_release_fragmented(rq, frag_page, recycle);
+			}
+		}
 	}
 }
 
@@ -512,10 +527,10 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 	u16 entries, pi, header_offset, err, wqe_bbs, new_entries;
 	u32 lkey = rq->mdev->mlx5e_res.hw_objs.mkey;
 	u16 page_index = shampo->curr_page_index;
+	struct mlx5e_frag_page *frag_page;
 	u64 addr = shampo->last_addr;
 	struct mlx5e_dma_info *dma_info;
 	struct mlx5e_umr_wqe *umr_wqe;
-	struct page **pagep;
 	int headroom, i;
 
 	headroom = rq->buff.headroom;
@@ -526,7 +541,7 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 	umr_wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi);
 	build_klm_umr(sq, umr_wqe, shampo->key, index, entries, wqe_bbs);
 
-	pagep = &shampo->pages[page_index];
+	frag_page = &shampo->pages[page_index];
 
 	for (i = 0; i < entries; i++, index++) {
 		dma_info = &shampo->info[index];
@@ -537,19 +552,19 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 			MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE;
 		if (!(header_offset & (PAGE_SIZE - 1))) {
 			page_index = (page_index + 1) & (shampo->hd_per_wq - 1);
-			pagep = &shampo->pages[page_index];
+			frag_page = &shampo->pages[page_index];
 
-			err = mlx5e_page_alloc_pool(rq, pagep);
+			err = mlx5e_page_alloc_fragmented(rq, frag_page);
 			if (unlikely(err))
 				goto err_unmap;
 
-			addr = page_pool_get_dma_addr(*pagep);
+			addr = page_pool_get_dma_addr(frag_page->page);
 
 			dma_info->addr = addr;
-			dma_info->pagep = pagep;
+			dma_info->frag_page = frag_page;
 		} else {
 			dma_info->addr = addr + header_offset;
-			dma_info->pagep = pagep;
+			dma_info->frag_page = frag_page;
 		}
 
 update_klm:
@@ -579,7 +594,7 @@ err_unmap:
 		dma_info = &shampo->info[--index];
 		if (!(i & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1))) {
 			dma_info->addr = ALIGN_DOWN(dma_info->addr, PAGE_SIZE);
-			mlx5e_page_release_dynamic(rq, *dma_info->pagep, true);
+			mlx5e_page_release_fragmented(rq, dma_info->frag_page, true);
 		}
 	}
 	rq->stats->buff_alloc_err++;
@@ -628,8 +643,8 @@ static int mlx5e_alloc_rx_hd_mpwqe(struct mlx5e_rq *rq)
 static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 {
 	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, ix);
-	struct page **pagep = &wi->alloc_units.pages[0];
 	struct mlx5e_icosq *sq = rq->icosq;
+	struct mlx5e_frag_page *frag_page;
 	struct mlx5_wq_cyc *wq = &sq->wq;
 	struct mlx5e_umr_wqe *umr_wqe;
 	u32 offset; /* 17-bit value with MTT. */
@@ -647,13 +662,15 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	umr_wqe = mlx5_wq_cyc_get_wqe(wq, pi);
 	memcpy(umr_wqe, &rq->mpwqe.umr_wqe, sizeof(struct mlx5e_umr_wqe));
 
-	for (i = 0; i < rq->mpwqe.pages_per_wqe; i++, pagep++) {
+	frag_page = &wi->alloc_units.frag_pages[0];
+
+	for (i = 0; i < rq->mpwqe.pages_per_wqe; i++, frag_page++) {
 		dma_addr_t addr;
 
-		err = mlx5e_page_alloc_pool(rq, pagep);
+		err = mlx5e_page_alloc_fragmented(rq, frag_page);
 		if (unlikely(err))
 			goto err_unmap;
-		addr = page_pool_get_dma_addr(*pagep);
+		addr = page_pool_get_dma_addr(frag_page->page);
 		umr_wqe->inline_mtts[i] = (struct mlx5_mtt) {
 			.ptag = cpu_to_be64(addr | MLX5_EN_WR),
 		};
@@ -694,8 +711,8 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 
 err_unmap:
 	while (--i >= 0) {
-		pagep--;
-		mlx5e_page_release_dynamic(rq, *pagep, true);
+		frag_page--;
+		mlx5e_page_release_fragmented(rq, frag_page, true);
 	}
 
 err:
@@ -713,8 +730,8 @@ err:
 void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close)
 {
 	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
+	struct mlx5e_frag_page *deleted_page = NULL;
 	int hd_per_wq = shampo->hd_per_wq;
-	struct page **deleted_page = NULL;
 	struct mlx5e_dma_info *hd_info;
 	int i, index = start;
 
@@ -727,10 +744,12 @@ void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close
 
 		hd_info = &shampo->info[index];
 		hd_info->addr = ALIGN_DOWN(hd_info->addr, PAGE_SIZE);
-		if (hd_info->pagep != deleted_page) {
-			deleted_page = hd_info->pagep;
-			mlx5e_page_release_dynamic(rq, *hd_info->pagep, false);
+		if (hd_info->frag_page && hd_info->frag_page != deleted_page) {
+			deleted_page = hd_info->frag_page;
+			mlx5e_page_release_fragmented(rq, hd_info->frag_page, false);
 		}
+
+		hd_info->frag_page = NULL;
 	}
 
 	if (start + len > hd_per_wq) {
@@ -1068,7 +1087,7 @@ static void *mlx5e_shampo_get_packet_hd(struct mlx5e_rq *rq, u16 header_index)
 	struct mlx5e_dma_info *last_head = &rq->mpwqe.shampo->info[header_index];
 	u16 head_offset = (last_head->addr & (PAGE_SIZE - 1)) + rq->buff.headroom;
 
-	return page_address(*last_head->pagep) + head_offset;
+	return page_address(last_head->frag_page->page) + head_offset;
 }
 
 static void mlx5e_shampo_update_ipv4_udp_hdr(struct mlx5e_rq *rq, struct iphdr *ipv4)
@@ -1521,8 +1540,8 @@ static struct sk_buff *
 mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 			  struct mlx5_cqe64 *cqe, u32 cqe_bcnt)
 {
+	struct mlx5e_frag_page *frag_page = wi->frag_page;
 	u16 rx_headroom = rq->buff.headroom;
-	struct page *page = *wi->pagep;
 	struct bpf_prog *prog;
 	struct sk_buff *skb;
 	u32 metasize = 0;
@@ -1530,11 +1549,11 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 	dma_addr_t addr;
 	u32 frag_size;
 
-	va             = page_address(page) + wi->offset;
+	va             = page_address(frag_page->page) + wi->offset;
 	data           = va + rx_headroom;
 	frag_size      = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt);
 
-	addr = page_pool_get_dma_addr(page);
+	addr = page_pool_get_dma_addr(frag_page->page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset,
 				      frag_size, rq->buff.map_dir);
 	net_prefetch(data);
@@ -1558,7 +1577,8 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 		return NULL;
 
 	/* queue up for recycling/reuse */
-	page_ref_inc(page);
+	skb_mark_for_recycle(skb);
+	frag_page->frags++;
 
 	return skb;
 }
@@ -1570,7 +1590,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	struct mlx5e_rq_frag_info *frag_info = &rq->wqe.info.arr[0];
 	struct mlx5e_wqe_frag_info *head_wi = wi;
 	u16 rx_headroom = rq->buff.headroom;
-	struct page *page = *wi->pagep;
+	struct mlx5e_frag_page *frag_page;
 	struct skb_shared_info *sinfo;
 	struct mlx5e_xdp_buff mxbuf;
 	u32 frag_consumed_bytes;
@@ -1580,10 +1600,12 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	u32 truesize;
 	void *va;
 
-	va = page_address(page) + wi->offset;
+	frag_page = wi->frag_page;
+
+	va = page_address(frag_page->page) + wi->offset;
 	frag_consumed_bytes = min_t(u32, frag_info->frag_size, cqe_bcnt);
 
-	addr = page_pool_get_dma_addr(page);
+	addr = page_pool_get_dma_addr(frag_page->page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset,
 				      rq->buff.frame0_sz, rq->buff.map_dir);
 	net_prefetchw(va); /* xdp_frame data area */
@@ -1600,11 +1622,11 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	while (cqe_bcnt) {
 		skb_frag_t *frag;
 
-		page = *wi->pagep;
+		frag_page = wi->frag_page;
 
 		frag_consumed_bytes = min_t(u32, frag_info->frag_size, cqe_bcnt);
 
-		addr = page_pool_get_dma_addr(page);
+		addr = page_pool_get_dma_addr(frag_page->page);
 		dma_sync_single_for_cpu(rq->pdev, addr + wi->offset,
 					frag_consumed_bytes, rq->buff.map_dir);
 
@@ -1618,11 +1640,12 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 		}
 
 		frag = &sinfo->frags[sinfo->nr_frags++];
-		__skb_frag_set_page(frag, page);
+
+		__skb_frag_set_page(frag, frag_page->page);
 		skb_frag_off_set(frag, wi->offset);
 		skb_frag_size_set(frag, frag_consumed_bytes);
 
-		if (page_is_pfmemalloc(page))
+		if (page_is_pfmemalloc(frag_page->page))
 			xdp_buff_set_frag_pfmemalloc(&mxbuf.xdp);
 
 		sinfo->xdp_frags_size += frag_consumed_bytes;
@@ -1651,21 +1674,17 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	if (unlikely(!skb))
 		return NULL;
 
-	page_ref_inc(*head_wi->pagep);
+	skb_mark_for_recycle(skb);
+	head_wi->frag_page->frags++;
 
 	if (xdp_buff_has_frags(&mxbuf.xdp)) {
-		int i;
-
 		/* sinfo->nr_frags is reset by build_skb, calculate again. */
 		xdp_update_skb_shared_info(skb, wi - head_wi - 1,
 					   sinfo->xdp_frags_size, truesize,
 					   xdp_buff_is_frag_pfmemalloc(&mxbuf.xdp));
 
-		for (i = 0; i < sinfo->nr_frags; i++) {
-			skb_frag_t *frag = &sinfo->frags[i];
-
-			page_ref_inc(skb_frag_page(frag));
-		}
+		for (struct mlx5e_wqe_frag_info *pwi = head_wi + 1; pwi < wi; pwi++)
+			pwi->frag_page->frags++;
 	}
 
 	return skb;
@@ -1848,7 +1867,8 @@ const struct mlx5e_rx_handlers mlx5e_rx_handlers_rep = {
 
 static void
 mlx5e_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
-		    struct page **pagep, u32 data_bcnt, u32 data_offset)
+		    struct mlx5e_frag_page *frag_page,
+		    u32 data_bcnt, u32 data_offset)
 {
 	net_prefetchw(skb->data);
 
@@ -1862,12 +1882,13 @@ mlx5e_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
 		else
 			truesize = ALIGN(pg_consumed_bytes, BIT(rq->mpwqe.log_stride_sz));
 
-		mlx5e_add_skb_frag(rq, skb, *pagep, data_offset,
+		frag_page->frags++;
+		mlx5e_add_skb_frag(rq, skb, frag_page->page, data_offset,
 				   pg_consumed_bytes, truesize);
 
 		data_bcnt -= pg_consumed_bytes;
 		data_offset = 0;
-		pagep++;
+		frag_page++;
 	}
 }
 
@@ -1876,11 +1897,11 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 				   struct mlx5_cqe64 *cqe, u16 cqe_bcnt, u32 head_offset,
 				   u32 page_idx)
 {
-	struct page **pagep = &wi->alloc_units.pages[page_idx];
+	struct mlx5e_frag_page *frag_page = &wi->alloc_units.frag_pages[page_idx];
 	u16 headlen = min_t(u16, MLX5E_RX_MAX_HEAD, cqe_bcnt);
+	struct mlx5e_frag_page *head_page = frag_page;
 	u32 frag_offset    = head_offset + headlen;
 	u32 byte_cnt       = cqe_bcnt - headlen;
-	struct page *head_page = *pagep;
 	struct sk_buff *skb;
 	dma_addr_t addr;
 
@@ -1895,14 +1916,15 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 
 	/* Non-linear mode, hence non-XSK, which always uses PAGE_SIZE. */
 	if (unlikely(frag_offset >= PAGE_SIZE)) {
-		pagep++;
+		frag_page++;
 		frag_offset -= PAGE_SIZE;
 	}
 
-	mlx5e_fill_skb_data(skb, rq, pagep, byte_cnt, frag_offset);
+	skb_mark_for_recycle(skb);
+	mlx5e_fill_skb_data(skb, rq, frag_page, byte_cnt, frag_offset);
 	/* copy header */
-	addr = page_pool_get_dma_addr(head_page);
-	mlx5e_copy_skb_header(rq, skb, head_page, addr,
+	addr = page_pool_get_dma_addr(head_page->page);
+	mlx5e_copy_skb_header(rq, skb, head_page->page, addr,
 			      head_offset, head_offset, headlen);
 	/* skb linear part was allocated with headlen and aligned to long */
 	skb->tail += headlen;
@@ -1916,7 +1938,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 				struct mlx5_cqe64 *cqe, u16 cqe_bcnt, u32 head_offset,
 				u32 page_idx)
 {
-	struct page *page = wi->alloc_units.pages[page_idx];
+	struct mlx5e_frag_page *frag_page = &wi->alloc_units.frag_pages[page_idx];
 	u16 rx_headroom = rq->buff.headroom;
 	struct bpf_prog *prog;
 	struct sk_buff *skb;
@@ -1931,11 +1953,11 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		return NULL;
 	}
 
-	va             = page_address(page) + head_offset;
+	va             = page_address(frag_page->page) + head_offset;
 	data           = va + rx_headroom;
 	frag_size      = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt);
 
-	addr = page_pool_get_dma_addr(page);
+	addr = page_pool_get_dma_addr(frag_page->page);
 	dma_sync_single_range_for_cpu(rq->pdev, addr, head_offset,
 				      frag_size, rq->buff.map_dir);
 	net_prefetch(data);
@@ -1962,7 +1984,8 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		return NULL;
 
 	/* queue up for recycling/reuse */
-	page_ref_inc(page);
+	skb_mark_for_recycle(skb);
+	frag_page->frags++;
 
 	return skb;
 }
@@ -1979,7 +2002,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 	void *hdr, *data;
 	u32 frag_size;
 
-	hdr		= page_address(*head->pagep) + head_offset;
+	hdr		= page_address(head->frag_page->page) + head_offset;
 	data		= hdr + rx_headroom;
 	frag_size	= MLX5_SKB_FRAG_SZ(rx_headroom + head_size);
 
@@ -1993,9 +2016,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		if (unlikely(!skb))
 			return NULL;
 
-		/* queue up for recycling/reuse */
-		page_ref_inc(*head->pagep);
-
+		head->frag_page->frags++;
 	} else {
 		/* allocate SKB and copy header for large header */
 		rq->stats->gro_large_hds++;
@@ -2007,13 +2028,17 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		}
 
 		prefetchw(skb->data);
-		mlx5e_copy_skb_header(rq, skb, *head->pagep, head->addr,
+		mlx5e_copy_skb_header(rq, skb, head->frag_page->page, head->addr,
 				      head_offset + rx_headroom,
 				      rx_headroom, head_size);
 		/* skb linear part was allocated with headlen and aligned to long */
 		skb->tail += head_size;
 		skb->len  += head_size;
 	}
+
+	/* queue up for recycling/reuse */
+	skb_mark_for_recycle(skb);
+
 	return skb;
 }
 
@@ -2061,7 +2086,7 @@ mlx5e_free_rx_shampo_hd_entry(struct mlx5e_rq *rq, u16 header_index)
 		struct mlx5e_dma_info *dma_info = &shampo->info[header_index];
 
 		dma_info->addr = ALIGN_DOWN(addr, PAGE_SIZE);
-		mlx5e_page_release_dynamic(rq, *dma_info->pagep, true);
+		mlx5e_page_release_fragmented(rq, dma_info->frag_page, true);
 	}
 	bitmap_clear(shampo->bitmap, header_index, 1);
 }
@@ -2131,9 +2156,10 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	}
 
 	if (likely(head_size)) {
-		struct page **pagep = &wi->alloc_units.pages[page_idx];
+		struct mlx5e_frag_page *frag_page;
 
-		mlx5e_fill_skb_data(*skb, rq, pagep, data_bcnt, data_offset);
+		frag_page = &wi->alloc_units.frag_pages[page_idx];
+		mlx5e_fill_skb_data(*skb, rq, frag_page, data_bcnt, data_offset);
 	}
 
 	mlx5e_shampo_complete_rx_cqe(rq, cqe, cqe_bcnt, *skb);
-- 
cgit v1.2.3-70-g09d2


From 38a36efccd902939cc415f3757fb0d9a0f0618f9 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Fri, 17 Feb 2023 12:56:30 +0200
Subject: net/mlx5e: RX, Rename xdp_xmit_bitmap to a more generic name

The xdp_xmit_bitmap currently serves only one purpose: to avoid
releasing pages that are still in use due to XDP TX.

A following patch will use this bitmap in a slightly different context
but for the same purpose. So rename the bitmap to a more generic name
that reflects the purpose not the context.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h        |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c |  4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c     | 12 ++++++------
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 0862556105d5..566ddf7a7aa9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -622,7 +622,7 @@ union mlx5e_alloc_units {
 
 struct mlx5e_mpw_info {
 	u16 consumed_strides;
-	DECLARE_BITMAP(xdp_xmit_bitmap, MLX5_MPWRQ_MAX_PAGES_PER_WQE);
+	DECLARE_BITMAP(skip_release_bitmap, MLX5_MPWRQ_MAX_PAGES_PER_WQE);
 	union mlx5e_alloc_units alloc_units;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
index b2c1af07c317..3cde264cbe4e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
@@ -120,7 +120,7 @@ int mlx5e_xsk_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 		}
 	}
 
-	bitmap_zero(wi->xdp_xmit_bitmap, rq->mpwqe.pages_per_wqe);
+	bitmap_zero(wi->skip_release_bitmap, rq->mpwqe.pages_per_wqe);
 	wi->consumed_strides = 0;
 
 	umr_wqe->ctrl.opmod_idx_opcode =
@@ -289,7 +289,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq,
 	prog = rcu_dereference(rq->xdp_prog);
 	if (likely(prog && mlx5e_xdp_handle(rq, prog, mxbuf))) {
 		if (likely(__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)))
-			__set_bit(page_idx, wi->xdp_xmit_bitmap); /* non-atomic */
+			__set_bit(page_idx, wi->skip_release_bitmap); /* non-atomic */
 		return NULL; /* page/packet was consumed by XDP */
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 7724e30ec133..eab8cba33ce4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -439,10 +439,10 @@ mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle
 	int i;
 
 	/* A common case for AF_XDP. */
-	if (bitmap_full(wi->xdp_xmit_bitmap, rq->mpwqe.pages_per_wqe))
+	if (bitmap_full(wi->skip_release_bitmap, rq->mpwqe.pages_per_wqe))
 		return;
 
-	no_xdp_xmit = bitmap_empty(wi->xdp_xmit_bitmap, rq->mpwqe.pages_per_wqe);
+	no_xdp_xmit = bitmap_empty(wi->skip_release_bitmap, rq->mpwqe.pages_per_wqe);
 
 	if (rq->xsk_pool) {
 		struct xdp_buff **xsk_buffs = wi->alloc_units.xsk_buffs;
@@ -452,11 +452,11 @@ mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle
 		 * the page to the userspace when the interface goes down.
 		 */
 		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++)
-			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap))
+			if (no_xdp_xmit || !test_bit(i, wi->skip_release_bitmap))
 				xsk_buff_free(xsk_buffs[i]);
 	} else {
 		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++) {
-			if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap)) {
+			if (no_xdp_xmit || !test_bit(i, wi->skip_release_bitmap)) {
 				struct mlx5e_frag_page *frag_page;
 
 				frag_page = &wi->alloc_units.frag_pages[i];
@@ -687,7 +687,7 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 		       sizeof(*umr_wqe->inline_mtts) * pad);
 	}
 
-	bitmap_zero(wi->xdp_xmit_bitmap, rq->mpwqe.pages_per_wqe);
+	bitmap_zero(wi->skip_release_bitmap, rq->mpwqe.pages_per_wqe);
 	wi->consumed_strides = 0;
 
 	umr_wqe->ctrl.opmod_idx_opcode =
@@ -1970,7 +1970,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, cqe_bcnt, &mxbuf);
 		if (mlx5e_xdp_handle(rq, prog, &mxbuf)) {
 			if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags))
-				__set_bit(page_idx, wi->xdp_xmit_bitmap); /* non-atomic */
+				__set_bit(page_idx, wi->skip_release_bitmap); /* non-atomic */
 			return NULL; /* page/packet was consumed by XDP */
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 4c2a1323680709078736a2886c754004e9fe42e6 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 14 Feb 2023 12:01:40 +0200
Subject: net/mlx5e: RX, Defer page release in striding rq for better recycling

Currently, for striding RQ, fragmented pages from the page pool can
get released in two ways:

1) In the mlx5e driver when trimming off the unused fragments AND the
   associated skb fragments have been released. This path allows
   recycling of pages to the page pool cache (allow_direct == true).

2) On the skb release path (last fragment release), which
   will always release pages to the page pool ring
   (allow_direct == false).

Whichever is releasing the last fragment will be decisive on
where the page gets released: the cache or the ring. So we
obviously want to maximize for doing the release from 1.

This patch does that by deferring the release of page fragments
right before requesting new ones from the page pool. Extra care
needs to be taken for the corner cases:

* On first call, make sure that release is not called. The
  skip_release_bitmap is used for this purpose.

* On rq shutdown, make sure that all wqes that were not
  in the linked list are released.

For a single ring, single core, default MTU (1500) TCP stream
test the number of pages allocated from the cache directly
(rx_pp_recycle_cached) increases from 31 % to 98 %:

+----------------------------------------------+
| Page Pool stats (/sec)  |  Before |   After  |
+-------------------------+---------+----------+
|rx_pp_alloc_fast         | 2137754 |  2261033 |
|rx_pp_alloc_slow         |      47 |        9 |
|rx_pp_alloc_empty        |      47 |        9 |
|rx_pp_alloc_refill       |   23230 |      819 |
|rx_pp_alloc_waive        |       0 |        0 |
|rx_pp_recycle_cached     |  672182 |  2209015 |
|rx_pp_recycle_cache_full |    1789 |        0 |
|rx_pp_recycle_ring       | 1485848 |    52259 |
|rx_pp_recycle_ring_full  |    3003 |      584 |
+----------------------------------------------+

With this patch, the performance in striding rq for the above test is
back to baseline.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/reporter_rx.c    |  4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h   |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c   | 21 +++++++++++++++++----
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c     |  8 +++++---
 4 files changed, 25 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
index b621f735cdc3..a047a2a4ddac 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
@@ -121,9 +121,9 @@ static int mlx5e_rx_reporter_err_icosq_cqe_recover(void *ctx)
 
 	mlx5e_reset_icosq_cc_pc(icosq);
 
-	mlx5e_free_rx_in_progress_descs(rq);
+	mlx5e_free_rx_missing_descs(rq);
 	if (xskrq)
-		mlx5e_free_rx_in_progress_descs(xskrq);
+		mlx5e_free_rx_missing_descs(xskrq);
 
 	clear_bit(MLX5E_SQ_STATE_RECOVERING, &icosq->state);
 	mlx5e_activate_icosq(icosq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index cd7779a9d046..651be7aaf7d5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -69,7 +69,7 @@ INDIRECT_CALLABLE_DECLARE(bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq));
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq));
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
 void mlx5e_free_rx_descs(struct mlx5e_rq *rq);
-void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq);
+void mlx5e_free_rx_missing_descs(struct mlx5e_rq *rq);
 
 static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index eca9a11454e5..53eef689f225 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -301,6 +301,15 @@ static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, int node)
 	if (!rq->mpwqe.info)
 		return -ENOMEM;
 
+	/* For deferred page release (release right before alloc), make sure
+	 * that on first round release is not called.
+	 */
+	for (int i = 0; i < wq_sz; i++) {
+		struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, i);
+
+		bitmap_fill(wi->skip_release_bitmap, rq->mpwqe.pages_per_wqe);
+	}
+
 	mlx5e_build_umr_wqe(rq, rq->icosq, &rq->mpwqe.umr_wqe);
 
 	return 0;
@@ -1112,7 +1121,7 @@ int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq, int wait_time)
 	return -ETIMEDOUT;
 }
 
-void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq)
+void mlx5e_free_rx_missing_descs(struct mlx5e_rq *rq)
 {
 	struct mlx5_wq_ll *wq;
 	u16 head;
@@ -1124,8 +1133,12 @@ void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq)
 	wq = &rq->mpwqe.wq;
 	head = wq->head;
 
-	/* Outstanding UMR WQEs (in progress) start at wq->head */
-	for (i = 0; i < rq->mpwqe.umr_in_progress; i++) {
+	/* Release WQEs that are in missing state: they have been
+	 * popped from the list after completion but were not freed
+	 * due to deferred release.
+	 * Also free the linked-list reserved entry, hence the "+ 1".
+	 */
+	for (i = 0; i < mlx5_wq_ll_missing(wq) + 1; i++) {
 		rq->dealloc_wqe(rq, head);
 		head = mlx5_wq_ll_get_wqe_next_ix(wq, head);
 	}
@@ -1152,7 +1165,7 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq)
 	if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) {
 		struct mlx5_wq_ll *wq = &rq->mpwqe.wq;
 
-		mlx5e_free_rx_in_progress_descs(rq);
+		mlx5e_free_rx_missing_descs(rq);
 
 		while (!mlx5_wq_ll_is_empty(wq)) {
 			struct mlx5e_rx_wqe_ll *wqe;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index eab8cba33ce4..73bc373bf27d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -983,6 +983,11 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq)
 	head = rq->mpwqe.actual_wq_head;
 	i = missing;
 	do {
+		struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, head);
+
+		/* Deferred free for better page pool cache usage. */
+		mlx5e_free_rx_mpwqe(rq, wi, true);
+
 		alloc_err = rq->xsk_pool ? mlx5e_xsk_alloc_rx_mpwqe(rq, head) :
 					   mlx5e_alloc_rx_mpwqe(rq, head);
 
@@ -1855,7 +1860,6 @@ mpwrq_cqe_out:
 
 	wq  = &rq->mpwqe.wq;
 	wqe = mlx5_wq_ll_get_wqe(wq, wqe_id);
-	mlx5e_free_rx_mpwqe(rq, wi, true);
 	mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index);
 }
 
@@ -2173,7 +2177,6 @@ mpwrq_cqe_out:
 
 	wq  = &rq->mpwqe.wq;
 	wqe = mlx5_wq_ll_get_wqe(wq, wqe_id);
-	mlx5e_free_rx_mpwqe(rq, wi, true);
 	mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index);
 }
 
@@ -2233,7 +2236,6 @@ mpwrq_cqe_out:
 
 	wq  = &rq->mpwqe.wq;
 	wqe = mlx5_wq_ll_get_wqe(wq, wqe_id);
-	mlx5e_free_rx_mpwqe(rq, wi, true);
 	mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 625dff29df3957f8a864c05d21a327231ebeff94 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 21 Feb 2023 17:24:09 +0200
Subject: net/mlx5e: RX, Change wqe last_in_page field from bool to bit flags

Change the bool flag to a bitfield as we'll use it in a downstream patch
in the series to add signaling about skipping a fragment release.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      | 6 +++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c   | 2 +-
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 566ddf7a7aa9..9ef4b7163e5a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -605,13 +605,17 @@ struct mlx5e_frag_page {
 	u16 frags;
 };
 
+enum mlx5e_wqe_frag_flag {
+	MLX5E_WQE_FRAG_LAST_IN_PAGE,
+};
+
 struct mlx5e_wqe_frag_info {
 	union {
 		struct mlx5e_frag_page *frag_page;
 		struct xdp_buff **xskp;
 	};
 	u32 offset;
-	bool last_in_page;
+	u8 flags;
 };
 
 union mlx5e_alloc_units {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 53eef689f225..bb1cbf008876 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -532,7 +532,7 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 				next_frag.frag_page++;
 				next_frag.offset = 0;
 				if (prev)
-					prev->last_in_page = true;
+					prev->flags |= BIT(MLX5E_WQE_FRAG_LAST_IN_PAGE);
 			}
 			*frag = next_frag;
 
@@ -543,7 +543,7 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 	}
 
 	if (prev)
-		prev->last_in_page = true;
+		prev->flags |= BIT(MLX5E_WQE_FRAG_LAST_IN_PAGE);
 }
 
 static void mlx5e_init_xsk_buffs(struct mlx5e_rq *rq)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 73bc373bf27d..f98212596c1e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -323,7 +323,7 @@ static inline void mlx5e_put_rx_frag(struct mlx5e_rq *rq,
 				     struct mlx5e_wqe_frag_info *frag,
 				     bool recycle)
 {
-	if (frag->last_in_page)
+	if (frag->flags & BIT(MLX5E_WQE_FRAG_LAST_IN_PAGE))
 		mlx5e_page_release_fragmented(rq, frag->frag_page, recycle);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 3f93f82988bc6b4104314bf9aa485d47fbb7e241 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 21 Feb 2023 20:25:15 +0200
Subject: net/mlx5e: RX, Defer page release in legacy rq for better recycling

Currently, fragmented pages from the page pool can be released
in two ways:

1) In the mlx5e driver when trimming off the unused fragments AND the
associated skb fragments have been released. This path allows
recycling of pages to the page pool cache (allow_direct == true).

2) On the skb release path (last fragment release), which
will always release pages to the page pool ring
(allow_direct == false).

Whichever is releasing the last fragment will be decisive on
where the page gets released: the cache or the ring. So we
obviously want to maximize for doing the release from 1.

This patch does that by deferring the release of page fragments
right before requesting new ones from the page pool. A flag is
added to make sure that there's no release before first alloc
and that XDP_TX fragments are not released prematurely.

This is a preparation patch that doesn't unlock the performance
improvements yet. A followup patch will do that.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  1 +
 .../net/ethernet/mellanox/mlx5/core/en/xsk/rx.c    |  2 +
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  | 20 ++++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 66 ++++++++++++++--------
 4 files changed, 66 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 9ef4b7163e5a..613a7daf9595 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -607,6 +607,7 @@ struct mlx5e_frag_page {
 
 enum mlx5e_wqe_frag_flag {
 	MLX5E_WQE_FRAG_LAST_IN_PAGE,
+	MLX5E_WQE_FRAG_SKIP_RELEASE,
 };
 
 struct mlx5e_wqe_frag_info {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
index 3cde264cbe4e..d97e6df66f45 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
@@ -189,6 +189,7 @@ int mlx5e_xsk_alloc_rx_wqes_batched(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 
 		addr = xsk_buff_xdp_get_frame_dma(*frag->xskp);
 		wqe->data[0].addr = cpu_to_be64(addr + rq->buff.headroom);
+		frag->flags &= ~BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
 	}
 
 	return alloc;
@@ -215,6 +216,7 @@ int mlx5e_xsk_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 
 		addr = xsk_buff_xdp_get_frame_dma(*frag->xskp);
 		wqe->data[0].addr = cpu_to_be64(addr + rq->buff.headroom);
+		frag->flags &= ~BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
 	}
 
 	return wqe_bulk;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index bb1cbf008876..3cf7d2b59037 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -520,6 +520,9 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq)
 
 	next_frag.frag_page = &rq->wqe.alloc_units->frag_pages[0];
 
+	/* Skip first release due to deferred release. */
+	next_frag.flags = BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
+
 	for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++) {
 		struct mlx5e_rq_frag_info *frag_info = &rq->wqe.info.arr[0];
 		struct mlx5e_wqe_frag_info *frag =
@@ -558,8 +561,14 @@ static void mlx5e_init_xsk_buffs(struct mlx5e_rq *rq)
 	/* Considering the above assumptions a fragment maps to a single
 	 * xsk_buff.
 	 */
-	for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++)
+	for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++) {
 		rq->wqe.frags[i].xskp = &rq->wqe.alloc_units->xsk_buffs[i];
+
+		/* Skip first release due to deferred release as WQES are
+		 * not allocated yet.
+		 */
+		rq->wqe.frags[i].flags |= BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
+	}
 }
 
 static int mlx5e_init_wqe_alloc_info(struct mlx5e_rq *rq, int node)
@@ -1183,12 +1192,21 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq)
 						0, true);
 	} else {
 		struct mlx5_wq_cyc *wq = &rq->wqe.wq;
+		u16 missing = mlx5_wq_cyc_missing(wq);
+		u16 head = mlx5_wq_cyc_get_head(wq);
 
 		while (!mlx5_wq_cyc_is_empty(wq)) {
 			wqe_ix = mlx5_wq_cyc_get_tail(wq);
 			rq->dealloc_wqe(rq, wqe_ix);
 			mlx5_wq_cyc_pop(wq);
 		}
+		/* Missing slots might also contain unreleased pages due to
+		 * deferred release.
+		 */
+		while (missing--) {
+			wqe_ix = mlx5_wq_cyc_ctr2ix(wq, head++);
+			rq->dealloc_wqe(rq, wqe_ix);
+		}
 	}
 
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index f98212596c1e..0675ffa5f8df 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -319,11 +319,21 @@ static inline int mlx5e_get_rx_frag(struct mlx5e_rq *rq,
 	return err;
 }
 
+static bool mlx5e_frag_can_release(struct mlx5e_wqe_frag_info *frag)
+{
+#define CAN_RELEASE_MASK \
+	(BIT(MLX5E_WQE_FRAG_LAST_IN_PAGE) | BIT(MLX5E_WQE_FRAG_SKIP_RELEASE))
+
+#define CAN_RELEASE_VALUE BIT(MLX5E_WQE_FRAG_LAST_IN_PAGE)
+
+	return (frag->flags & CAN_RELEASE_MASK) == CAN_RELEASE_VALUE;
+}
+
 static inline void mlx5e_put_rx_frag(struct mlx5e_rq *rq,
 				     struct mlx5e_wqe_frag_info *frag,
 				     bool recycle)
 {
-	if (frag->flags & BIT(MLX5E_WQE_FRAG_LAST_IN_PAGE))
+	if (mlx5e_frag_can_release(frag))
 		mlx5e_page_release_fragmented(rq, frag->frag_page, recycle);
 }
 
@@ -347,6 +357,8 @@ static int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe_cyc *wqe,
 		if (unlikely(err))
 			goto free_frags;
 
+		frag->flags &= ~BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
+
 		headroom = i == 0 ? rq->buff.headroom : 0;
 		addr = page_pool_get_dma_addr(frag->frag_page->page);
 		wqe->data[i].addr = cpu_to_be64(addr + frag->offset + headroom);
@@ -367,7 +379,7 @@ static inline void mlx5e_free_rx_wqe(struct mlx5e_rq *rq,
 {
 	int i;
 
-	if (rq->xsk_pool) {
+	if (rq->xsk_pool && !(wi->flags & BIT(MLX5E_WQE_FRAG_SKIP_RELEASE))) {
 		/* The `recycle` parameter is ignored, and the page is always
 		 * put into the Reuse Ring, because there is no way to return
 		 * the page to the userspace when the interface goes down.
@@ -387,6 +399,20 @@ static void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix)
 	mlx5e_free_rx_wqe(rq, wi, false);
 }
 
+static void mlx5e_free_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
+{
+	struct mlx5_wq_cyc *wq = &rq->wqe.wq;
+	int i;
+
+	for (i = 0; i < wqe_bulk; i++) {
+		int j = mlx5_wq_cyc_ctr2ix(wq, ix + i);
+		struct mlx5e_wqe_frag_info *wi;
+
+		wi = get_frag(rq, j);
+		mlx5e_free_rx_wqe(rq, wi, true);
+	}
+}
+
 static int mlx5e_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 {
 	struct mlx5_wq_cyc *wq = &rq->wqe.wq;
@@ -792,6 +818,8 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq)
 	 */
 	wqe_bulk -= (head + wqe_bulk) & rq->wqe.info.wqe_index_mask;
 
+	mlx5e_free_rx_wqes(rq, head, wqe_bulk);
+
 	if (!rq->xsk_pool)
 		count = mlx5e_alloc_rx_wqes(rq, head, wqe_bulk);
 	else if (likely(!rq->xsk_pool->dma_need_sync))
@@ -1727,7 +1755,7 @@ static void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 
 	if (unlikely(MLX5E_RX_ERR_CQE(cqe))) {
 		mlx5e_handle_rx_err_cqe(rq, cqe);
-		goto free_wqe;
+		goto wq_cyc_pop;
 	}
 
 	skb = INDIRECT_CALL_3(rq->wqe.skb_from_cqe,
@@ -1741,9 +1769,9 @@ static void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 			/* do not return page to cache,
 			 * it will be returned on XDP_TX completion.
 			 */
-			goto wq_cyc_pop;
+			wi->flags |= BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
 		}
-		goto free_wqe;
+		goto wq_cyc_pop;
 	}
 
 	mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb);
@@ -1751,13 +1779,11 @@ static void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 	if (mlx5e_cqe_regb_chain(cqe))
 		if (!mlx5e_tc_update_skb_nic(cqe, skb)) {
 			dev_kfree_skb_any(skb);
-			goto free_wqe;
+			goto wq_cyc_pop;
 		}
 
 	napi_gro_receive(rq->cq.napi, skb);
 
-free_wqe:
-	mlx5e_free_rx_wqe(rq, wi, true);
 wq_cyc_pop:
 	mlx5_wq_cyc_pop(wq);
 }
@@ -1781,7 +1807,7 @@ static void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 
 	if (unlikely(MLX5E_RX_ERR_CQE(cqe))) {
 		mlx5e_handle_rx_err_cqe(rq, cqe);
-		goto free_wqe;
+		goto wq_cyc_pop;
 	}
 
 	skb = INDIRECT_CALL_2(rq->wqe.skb_from_cqe,
@@ -1794,9 +1820,9 @@ static void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 			/* do not return page to cache,
 			 * it will be returned on XDP_TX completion.
 			 */
-			goto wq_cyc_pop;
+			wi->flags |= BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
 		}
-		goto free_wqe;
+		goto wq_cyc_pop;
 	}
 
 	mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb);
@@ -1806,8 +1832,6 @@ static void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 
 	mlx5e_rep_tc_receive(cqe, rq, skb);
 
-free_wqe:
-	mlx5e_free_rx_wqe(rq, wi, true);
 wq_cyc_pop:
 	mlx5_wq_cyc_pop(wq);
 }
@@ -2454,7 +2478,7 @@ static void mlx5i_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 
 	if (unlikely(MLX5E_RX_ERR_CQE(cqe))) {
 		rq->stats->wqe_err++;
-		goto wq_free_wqe;
+		goto wq_cyc_pop;
 	}
 
 	skb = INDIRECT_CALL_2(rq->wqe.skb_from_cqe,
@@ -2462,17 +2486,16 @@ static void mlx5i_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 			      mlx5e_skb_from_cqe_nonlinear,
 			      rq, wi, cqe, cqe_bcnt);
 	if (!skb)
-		goto wq_free_wqe;
+		goto wq_cyc_pop;
 
 	mlx5i_complete_rx_cqe(rq, cqe, cqe_bcnt, skb);
 	if (unlikely(!skb->dev)) {
 		dev_kfree_skb_any(skb);
-		goto wq_free_wqe;
+		goto wq_cyc_pop;
 	}
 	napi_gro_receive(rq->cq.napi, skb);
 
-wq_free_wqe:
-	mlx5e_free_rx_wqe(rq, wi, true);
+wq_cyc_pop:
 	mlx5_wq_cyc_pop(wq);
 }
 
@@ -2547,12 +2570,12 @@ static void mlx5e_trap_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe
 
 	if (unlikely(MLX5E_RX_ERR_CQE(cqe))) {
 		rq->stats->wqe_err++;
-		goto free_wqe;
+		goto wq_cyc_pop;
 	}
 
 	skb = mlx5e_skb_from_cqe_nonlinear(rq, wi, cqe, cqe_bcnt);
 	if (!skb)
-		goto free_wqe;
+		goto wq_cyc_pop;
 
 	mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb);
 	skb_push(skb, ETH_HLEN);
@@ -2561,8 +2584,7 @@ static void mlx5e_trap_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe
 				 rq->netdev->devlink_port);
 	dev_kfree_skb_any(skb);
 
-free_wqe:
-	mlx5e_free_rx_wqe(rq, wi, false);
+wq_cyc_pop:
 	mlx5_wq_cyc_pop(wq);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 76238d0fbd21948dd9423e53f57e7354898967d3 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 21 Feb 2023 20:31:39 +0200
Subject: net/mlx5e: RX, Split off release path for xsk buffers for legacy rq

Don't mix xsk buffer releases with page releases anymore. This is
needed for handling of deferred page release.

Add a new bulk free function for xsk buffers from wqe frags.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 50 +++++++++++++++++--------
 1 file changed, 35 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 0675ffa5f8df..9c5270eb9dc6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -379,24 +379,42 @@ static inline void mlx5e_free_rx_wqe(struct mlx5e_rq *rq,
 {
 	int i;
 
-	if (rq->xsk_pool && !(wi->flags & BIT(MLX5E_WQE_FRAG_SKIP_RELEASE))) {
-		/* The `recycle` parameter is ignored, and the page is always
-		 * put into the Reuse Ring, because there is no way to return
-		 * the page to the userspace when the interface goes down.
-		 */
-		xsk_buff_free(*wi->xskp);
-		return;
-	}
-
 	for (i = 0; i < rq->wqe.info.num_frags; i++, wi++)
 		mlx5e_put_rx_frag(rq, wi, recycle);
 }
 
+static void mlx5e_xsk_free_rx_wqe(struct mlx5e_wqe_frag_info *wi)
+{
+	if (!(wi->flags & BIT(MLX5E_WQE_FRAG_SKIP_RELEASE)))
+		xsk_buff_free(*wi->xskp);
+}
+
 static void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix)
 {
 	struct mlx5e_wqe_frag_info *wi = get_frag(rq, ix);
 
-	mlx5e_free_rx_wqe(rq, wi, false);
+	if (rq->xsk_pool)
+		mlx5e_xsk_free_rx_wqe(wi);
+	else
+		mlx5e_free_rx_wqe(rq, wi, false);
+}
+
+static void mlx5e_xsk_free_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
+{
+	struct mlx5_wq_cyc *wq = &rq->wqe.wq;
+	int i;
+
+	for (i = 0; i < wqe_bulk; i++) {
+		int j = mlx5_wq_cyc_ctr2ix(wq, ix + i);
+		struct mlx5e_wqe_frag_info *wi;
+
+		wi = get_frag(rq, j);
+		/* The page is always put into the Reuse Ring, because there
+		 * is no way to return the page to the userspace when the
+		 * interface goes down.
+		 */
+		mlx5e_xsk_free_rx_wqe(wi);
+	}
 }
 
 static void mlx5e_free_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
@@ -818,19 +836,21 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq)
 	 */
 	wqe_bulk -= (head + wqe_bulk) & rq->wqe.info.wqe_index_mask;
 
-	mlx5e_free_rx_wqes(rq, head, wqe_bulk);
-
-	if (!rq->xsk_pool)
+	if (!rq->xsk_pool) {
+		mlx5e_free_rx_wqes(rq, head, wqe_bulk);
 		count = mlx5e_alloc_rx_wqes(rq, head, wqe_bulk);
-	else if (likely(!rq->xsk_pool->dma_need_sync))
+	} else if (likely(!rq->xsk_pool->dma_need_sync)) {
+		mlx5e_xsk_free_rx_wqes(rq, head, wqe_bulk);
 		count = mlx5e_xsk_alloc_rx_wqes_batched(rq, head, wqe_bulk);
-	else
+	} else {
+		mlx5e_xsk_free_rx_wqes(rq, head, wqe_bulk);
 		/* If dma_need_sync is true, it's more efficient to call
 		 * xsk_buff_alloc in a loop, rather than xsk_buff_alloc_batch,
 		 * because the latter does the same check and returns only one
 		 * frame.
 		 */
 		count = mlx5e_xsk_alloc_rx_wqes(rq, head, wqe_bulk);
+	}
 
 	mlx5_wq_cyc_push_n(wq, count);
 	if (unlikely(count != wqe_bulk)) {
-- 
cgit v1.2.3-70-g09d2


From 4ba2b4988c98ce9b56b77a1610c3a7b70ee30b57 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 21 Feb 2023 20:42:48 +0200
Subject: net/mlx5e: RX, Increase WQE bulk size for legacy rq

Deferred page release was added to legacy rq but its desired effect
(driver releases last fragment to page pool cache) is not yet visible
due to the WQE bulks being too small.

This patch increases the WQE bulk size to span 512 KB and clip it to
one quarter of the rx queue size.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en/params.c    | 47 ++++++++++++++++++++--
 2 files changed, 44 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 613a7daf9595..a087c433366b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -670,7 +670,7 @@ struct mlx5e_rq_frags_info {
 	struct mlx5e_rq_frag_info arr[MLX5E_MAX_RX_FRAGS];
 	u8 num_frags;
 	u8 log_num_frags;
-	u8 wqe_bulk;
+	u16 wqe_bulk;
 	u8 wqe_index_mask;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 561da78d3b5c..40218d77ef34 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -667,6 +667,43 @@ static int mlx5e_max_nonlinear_mtu(int first_frag_size, int frag_size, bool xdp)
 	return first_frag_size + (MLX5E_MAX_RX_FRAGS - 2) * frag_size + PAGE_SIZE;
 }
 
+static void mlx5e_rx_compute_wqe_bulk_params(struct mlx5e_params *params,
+					     struct mlx5e_rq_frags_info *info)
+{
+	u16 bulk_bound_rq_size = (1 << params->log_rq_mtu_frames) / 4;
+	u32 bulk_bound_rq_size_in_bytes;
+	u32 sum_frag_strides = 0;
+	u32 wqe_bulk_in_bytes;
+	u32 wqe_bulk;
+	int i;
+
+	for (i = 0; i < info->num_frags; i++)
+		sum_frag_strides += info->arr[i].frag_stride;
+
+	/* For MTUs larger than PAGE_SIZE, align to PAGE_SIZE to reflect
+	 * amount of consumed pages per wqe in bytes.
+	 */
+	if (sum_frag_strides > PAGE_SIZE)
+		sum_frag_strides = ALIGN(sum_frag_strides, PAGE_SIZE);
+
+	bulk_bound_rq_size_in_bytes = bulk_bound_rq_size * sum_frag_strides;
+
+#define MAX_WQE_BULK_BYTES(xdp) ((xdp ? 256 : 512) * 1024)
+
+	/* A WQE bulk should not exceed min(512KB, 1/4 of rq size). For XDP
+	 * keep bulk size smaller to avoid filling the page_pool cache on
+	 * every bulk refill.
+	 */
+	wqe_bulk_in_bytes = min_t(u32, MAX_WQE_BULK_BYTES(params->xdp_prog),
+				  bulk_bound_rq_size_in_bytes);
+	wqe_bulk = DIV_ROUND_UP(wqe_bulk_in_bytes, sum_frag_strides);
+
+	/* Make sure that allocations don't start when the page is still used
+	 * by older WQEs.
+	 */
+	info->wqe_bulk = max_t(u16, info->wqe_index_mask + 1, wqe_bulk);
+}
+
 #define DEFAULT_FRAG_SIZE (2048)
 
 static int mlx5e_build_rq_frags_info(struct mlx5_core_dev *mdev,
@@ -774,11 +811,13 @@ static int mlx5e_build_rq_frags_info(struct mlx5_core_dev *mdev,
 	}
 
 out:
-	/* Bulking optimization to skip allocation until at least 8 WQEs can be
-	 * allocated in a row. At the same time, never start allocation when
-	 * the page is still used by older WQEs.
+	/* Bulking optimization to skip allocation until a large enough number
+	 * of WQEs can be allocated in a row. Bulking also influences how well
+	 * deferred page release works.
 	 */
-	info->wqe_bulk = max_t(u8, info->wqe_index_mask + 1, 8);
+	mlx5e_rx_compute_wqe_bulk_params(params, info);
+
+	mlx5_core_dbg(mdev, "%s: wqe_bulk = %u\n", __func__, info->wqe_bulk);
 
 	info->log_num_frags = order_base_2(info->num_frags);
 
-- 
cgit v1.2.3-70-g09d2


From cd640b050368d5be6bccf1edb51b1e4c553555e6 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 21 Feb 2023 21:05:07 +0200
Subject: net/mlx5e: RX, Break the wqe bulk refill in smaller chunks

To avoid overflowing the page pool's cache, don't release the
whole bulk which is usually larger than the cache refill size.
Group release+alloc instead into cache refill units that
allow releasing to the cache and then allocating from the cache.

A refill_unit variable is added as a iteration unit over the
wqe_bulk when doing release+alloc.

For a single ring, single core, default MTU (1500) TCP stream
test the number of pages allocated from the cache directly
(rx_pp_recycle_cached) increases from 0% to 52%:

+---------------------------------------------+
| Page Pool stats (/sec)  |  Before |   After |
+-------------------------+---------+---------+
|rx_pp_alloc_fast         | 2145422 | 2193802 |
|rx_pp_alloc_slow         |       2 |       0 |
|rx_pp_alloc_empty        |       2 |       0 |
|rx_pp_alloc_refill       |   34059 |   16634 |
|rx_pp_alloc_waive        |       0 |       0 |
|rx_pp_recycle_cached     |       0 | 1145818 |
|rx_pp_recycle_cache_full |       0 |       0 |
|rx_pp_recycle_ring       | 2179361 | 1064616 |
|rx_pp_recycle_ring_full  |     121 |       0 |
+---------------------------------------------+

With this patch, the performance for legacy rq for the above test is
back to baseline.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  1 +
 .../net/ethernet/mellanox/mlx5/core/en/params.c    |  8 ++++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 28 ++++++++++++++++++++--
 3 files changed, 34 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index a087c433366b..ba615b74bb8e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -671,6 +671,7 @@ struct mlx5e_rq_frags_info {
 	u8 num_frags;
 	u8 log_num_frags;
 	u16 wqe_bulk;
+	u16 refill_unit;
 	u8 wqe_index_mask;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 40218d77ef34..31f3c6e51d9e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -674,6 +674,7 @@ static void mlx5e_rx_compute_wqe_bulk_params(struct mlx5e_params *params,
 	u32 bulk_bound_rq_size_in_bytes;
 	u32 sum_frag_strides = 0;
 	u32 wqe_bulk_in_bytes;
+	u16 split_factor;
 	u32 wqe_bulk;
 	int i;
 
@@ -702,6 +703,10 @@ static void mlx5e_rx_compute_wqe_bulk_params(struct mlx5e_params *params,
 	 * by older WQEs.
 	 */
 	info->wqe_bulk = max_t(u16, info->wqe_index_mask + 1, wqe_bulk);
+
+	split_factor = DIV_ROUND_UP(MAX_WQE_BULK_BYTES(params->xdp_prog),
+				    PP_ALLOC_CACHE_REFILL * PAGE_SIZE);
+	info->refill_unit = DIV_ROUND_UP(info->wqe_bulk, split_factor);
 }
 
 #define DEFAULT_FRAG_SIZE (2048)
@@ -817,7 +822,8 @@ out:
 	 */
 	mlx5e_rx_compute_wqe_bulk_params(params, info);
 
-	mlx5_core_dbg(mdev, "%s: wqe_bulk = %u\n", __func__, info->wqe_bulk);
+	mlx5_core_dbg(mdev, "%s: wqe_bulk = %u, wqe_bulk_refill_unit = %u\n",
+		      __func__, info->wqe_bulk, info->refill_unit);
 
 	info->log_num_frags = order_base_2(info->num_frags);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 9c5270eb9dc6..df5dbef9e5ec 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -449,6 +449,31 @@ static int mlx5e_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 	return i;
 }
 
+static int mlx5e_refill_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
+{
+	int remaining = wqe_bulk;
+	int i = 0;
+
+	/* The WQE bulk is split into smaller bulks that are sized
+	 * according to the page pool cache refill size to avoid overflowing
+	 * the page pool cache due to too many page releases at once.
+	 */
+	do {
+		int refill = min_t(u16, rq->wqe.info.refill_unit, remaining);
+		int alloc_count;
+
+		mlx5e_free_rx_wqes(rq, ix + i, refill);
+		alloc_count = mlx5e_alloc_rx_wqes(rq, ix + i, refill);
+		i += alloc_count;
+		if (unlikely(alloc_count != refill))
+			break;
+
+		remaining -= refill;
+	} while (remaining);
+
+	return i;
+}
+
 static inline void
 mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb,
 		   struct page *page, u32 frag_offset, u32 len,
@@ -837,8 +862,7 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq)
 	wqe_bulk -= (head + wqe_bulk) & rq->wqe.info.wqe_index_mask;
 
 	if (!rq->xsk_pool) {
-		mlx5e_free_rx_wqes(rq, head, wqe_bulk);
-		count = mlx5e_alloc_rx_wqes(rq, head, wqe_bulk);
+		count = mlx5e_refill_rx_wqes(rq, head, wqe_bulk);
 	} else if (likely(!rq->xsk_pool->dma_need_sync)) {
 		mlx5e_xsk_free_rx_wqes(rq, head, wqe_bulk);
 		count = mlx5e_xsk_alloc_rx_wqes_batched(rq, head, wqe_bulk);
-- 
cgit v1.2.3-70-g09d2


From 3905f8d64ccc2c640d8c1179f4452f2bf8f1df56 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Wed, 22 Feb 2023 11:27:38 +0200
Subject: net/mlx5e: RX, Remove unnecessary recycle parameter and page_cache
 stats

The recycle parameter used during page release is no longer
necessary: the page pool can detect when the page cannot be
recycled to the cache or ring without any outside hint.

The page pool will also take care of cleaning up after itself
once all the inflight pages have been released. So no need to
explicitly release pages to the system.

Remove the internal page_cache stats as the mlx5e_page_cache
struct no longer exists.

Delete the documentation entries along with the stats.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/counters.rst            | 26 ------------
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c   |  7 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 47 ++++++++++------------
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 20 ---------
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 10 -----
 5 files changed, 25 insertions(+), 85 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
index 4cd8e869762b..6b2d1fe74ecf 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
@@ -346,32 +346,6 @@ the software port.
      - The number of receive packets with CQE compression on ring i [#accel]_.
      - Acceleration
 
-   * - `rx[i]_cache_reuse`
-     - The number of events of successful reuse of a page from a driver's
-       internal page cache.
-     - Acceleration
-
-   * - `rx[i]_cache_full`
-     - The number of events of full internal page cache where driver can't put a
-       page back to the cache for recycling (page will be freed).
-     - Acceleration
-
-   * - `rx[i]_cache_empty`
-     - The number of events where cache was empty - no page to give. Driver
-       shall allocate new page.
-     - Acceleration
-
-   * - `rx[i]_cache_busy`
-     - The number of events where cache head was busy and cannot be recycled.
-       Driver allocated new page.
-     - Acceleration
-
-   * - `rx[i]_cache_waive`
-     - The number of cache evacuation. This can occur due to page move to
-       another NUMA node or page was pfmemalloc-ed and should be freed as soon
-       as possible.
-     - Acceleration
-
    * - `rx[i]_arfs_err`
      - Number of flow rules that failed to be added to the flow table.
      - Error
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index ca6ac9772d22..15d15d0e5ef9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -505,7 +505,6 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq,
 				  struct mlx5e_xdp_wqe_info *wi,
 				  u32 *xsk_frames,
-				  bool recycle,
 				  struct xdp_frame_bulk *bq)
 {
 	struct mlx5e_xdp_info_fifo *xdpi_fifo = &sq->db.xdpi_fifo;
@@ -524,7 +523,7 @@ static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq,
 		case MLX5E_XDP_XMIT_MODE_PAGE:
 			/* XDP_TX from the regular RQ */
 			page_pool_put_defragged_page(xdpi.page.rq->page_pool,
-						     xdpi.page.page, -1, recycle);
+						     xdpi.page.page, -1, true);
 			break;
 		case MLX5E_XDP_XMIT_MODE_XSK:
 			/* AF_XDP send */
@@ -578,7 +577,7 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq)
 
 			sqcc += wi->num_wqebbs;
 
-			mlx5e_free_xdpsq_desc(sq, wi, &xsk_frames, true, &bq);
+			mlx5e_free_xdpsq_desc(sq, wi, &xsk_frames, &bq);
 		} while (!last_wqe);
 
 		if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ)) {
@@ -625,7 +624,7 @@ void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq)
 
 		sq->cc += wi->num_wqebbs;
 
-		mlx5e_free_xdpsq_desc(sq, wi, &xsk_frames, false, &bq);
+		mlx5e_free_xdpsq_desc(sq, wi, &xsk_frames, &bq);
 	}
 
 	xdp_flush_frame_bulk(&bq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index df5dbef9e5ec..1049805571c6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -293,14 +293,13 @@ static int mlx5e_page_alloc_fragmented(struct mlx5e_rq *rq,
 }
 
 static void mlx5e_page_release_fragmented(struct mlx5e_rq *rq,
-					  struct mlx5e_frag_page *frag_page,
-					  bool recycle)
+					  struct mlx5e_frag_page *frag_page)
 {
 	u16 drain_count = MLX5E_PAGECNT_BIAS_MAX - frag_page->frags;
 	struct page *page = frag_page->page;
 
 	if (page_pool_defrag_page(page, drain_count) == 0)
-		page_pool_put_defragged_page(rq->page_pool, page, -1, recycle);
+		page_pool_put_defragged_page(rq->page_pool, page, -1, true);
 }
 
 static inline int mlx5e_get_rx_frag(struct mlx5e_rq *rq,
@@ -330,11 +329,10 @@ static bool mlx5e_frag_can_release(struct mlx5e_wqe_frag_info *frag)
 }
 
 static inline void mlx5e_put_rx_frag(struct mlx5e_rq *rq,
-				     struct mlx5e_wqe_frag_info *frag,
-				     bool recycle)
+				     struct mlx5e_wqe_frag_info *frag)
 {
 	if (mlx5e_frag_can_release(frag))
-		mlx5e_page_release_fragmented(rq, frag->frag_page, recycle);
+		mlx5e_page_release_fragmented(rq, frag->frag_page);
 }
 
 static inline struct mlx5e_wqe_frag_info *get_frag(struct mlx5e_rq *rq, u16 ix)
@@ -368,19 +366,18 @@ static int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq, struct mlx5e_rx_wqe_cyc *wqe,
 
 free_frags:
 	while (--i >= 0)
-		mlx5e_put_rx_frag(rq, --frag, true);
+		mlx5e_put_rx_frag(rq, --frag);
 
 	return err;
 }
 
 static inline void mlx5e_free_rx_wqe(struct mlx5e_rq *rq,
-				     struct mlx5e_wqe_frag_info *wi,
-				     bool recycle)
+				     struct mlx5e_wqe_frag_info *wi)
 {
 	int i;
 
 	for (i = 0; i < rq->wqe.info.num_frags; i++, wi++)
-		mlx5e_put_rx_frag(rq, wi, recycle);
+		mlx5e_put_rx_frag(rq, wi);
 }
 
 static void mlx5e_xsk_free_rx_wqe(struct mlx5e_wqe_frag_info *wi)
@@ -396,7 +393,7 @@ static void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix)
 	if (rq->xsk_pool)
 		mlx5e_xsk_free_rx_wqe(wi);
 	else
-		mlx5e_free_rx_wqe(rq, wi, false);
+		mlx5e_free_rx_wqe(rq, wi);
 }
 
 static void mlx5e_xsk_free_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
@@ -427,7 +424,7 @@ static void mlx5e_free_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 		struct mlx5e_wqe_frag_info *wi;
 
 		wi = get_frag(rq, j);
-		mlx5e_free_rx_wqe(rq, wi, true);
+		mlx5e_free_rx_wqe(rq, wi);
 	}
 }
 
@@ -502,7 +499,7 @@ mlx5e_copy_skb_header(struct mlx5e_rq *rq, struct sk_buff *skb,
 }
 
 static void
-mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle)
+mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi)
 {
 	bool no_xdp_xmit;
 	int i;
@@ -516,9 +513,9 @@ mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle
 	if (rq->xsk_pool) {
 		struct xdp_buff **xsk_buffs = wi->alloc_units.xsk_buffs;
 
-		/* The `recycle` parameter is ignored, and the page is always
-		 * put into the Reuse Ring, because there is no way to return
-		 * the page to the userspace when the interface goes down.
+		/* The page is always put into the Reuse Ring, because there
+		 * is no way to return the page to userspace when the interface
+		 * goes down.
 		 */
 		for (i = 0; i < rq->mpwqe.pages_per_wqe; i++)
 			if (no_xdp_xmit || !test_bit(i, wi->skip_release_bitmap))
@@ -529,7 +526,7 @@ mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle
 				struct mlx5e_frag_page *frag_page;
 
 				frag_page = &wi->alloc_units.frag_pages[i];
-				mlx5e_page_release_fragmented(rq, frag_page, recycle);
+				mlx5e_page_release_fragmented(rq, frag_page);
 			}
 		}
 	}
@@ -663,7 +660,7 @@ err_unmap:
 		dma_info = &shampo->info[--index];
 		if (!(i & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1))) {
 			dma_info->addr = ALIGN_DOWN(dma_info->addr, PAGE_SIZE);
-			mlx5e_page_release_fragmented(rq, dma_info->frag_page, true);
+			mlx5e_page_release_fragmented(rq, dma_info->frag_page);
 		}
 	}
 	rq->stats->buff_alloc_err++;
@@ -781,7 +778,7 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 err_unmap:
 	while (--i >= 0) {
 		frag_page--;
-		mlx5e_page_release_fragmented(rq, frag_page, true);
+		mlx5e_page_release_fragmented(rq, frag_page);
 	}
 
 err:
@@ -815,7 +812,7 @@ void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close
 		hd_info->addr = ALIGN_DOWN(hd_info->addr, PAGE_SIZE);
 		if (hd_info->frag_page && hd_info->frag_page != deleted_page) {
 			deleted_page = hd_info->frag_page;
-			mlx5e_page_release_fragmented(rq, hd_info->frag_page, false);
+			mlx5e_page_release_fragmented(rq, hd_info->frag_page);
 		}
 
 		hd_info->frag_page = NULL;
@@ -833,8 +830,8 @@ void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close
 static void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 {
 	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, ix);
-	/* Don't recycle, this function is called on rq/netdev close */
-	mlx5e_free_rx_mpwqe(rq, wi, false);
+	/* This function is called on rq/netdev close. */
+	mlx5e_free_rx_mpwqe(rq, wi);
 }
 
 INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq)
@@ -1058,7 +1055,7 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq)
 		struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, head);
 
 		/* Deferred free for better page pool cache usage. */
-		mlx5e_free_rx_mpwqe(rq, wi, true);
+		mlx5e_free_rx_mpwqe(rq, wi);
 
 		alloc_err = rq->xsk_pool ? mlx5e_xsk_alloc_rx_mpwqe(rq, head) :
 					   mlx5e_alloc_rx_mpwqe(rq, head);
@@ -1739,7 +1736,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 			int i;
 
 			for (i = wi - head_wi; i < rq->wqe.info.num_frags; i++)
-				mlx5e_put_rx_frag(rq, &head_wi[i], true);
+				mlx5e_put_rx_frag(rq, &head_wi[i]);
 		}
 		return NULL; /* page/packet was consumed by XDP */
 	}
@@ -2158,7 +2155,7 @@ mlx5e_free_rx_shampo_hd_entry(struct mlx5e_rq *rq, u16 header_index)
 		struct mlx5e_dma_info *dma_info = &shampo->info[header_index];
 
 		dma_info->addr = ALIGN_DOWN(addr, PAGE_SIZE);
-		mlx5e_page_release_fragmented(rq, dma_info->frag_page, true);
+		mlx5e_page_release_fragmented(rq, dma_info->frag_page);
 	}
 	bitmap_clear(shampo->bitmap, header_index, 1);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 4478223c1720..f1d9596905c6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -179,11 +179,6 @@ static const struct counter_desc sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_buff_alloc_err) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cqe_compress_blks) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cqe_compress_pkts) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cache_reuse) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cache_full) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cache_empty) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cache_busy) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cache_waive) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_congst_umr) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_arfs_err) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_recover) },
@@ -358,11 +353,6 @@ static void mlx5e_stats_grp_sw_update_stats_rq_stats(struct mlx5e_sw_stats *s,
 	s->rx_buff_alloc_err          += rq_stats->buff_alloc_err;
 	s->rx_cqe_compress_blks       += rq_stats->cqe_compress_blks;
 	s->rx_cqe_compress_pkts       += rq_stats->cqe_compress_pkts;
-	s->rx_cache_reuse             += rq_stats->cache_reuse;
-	s->rx_cache_full              += rq_stats->cache_full;
-	s->rx_cache_empty             += rq_stats->cache_empty;
-	s->rx_cache_busy              += rq_stats->cache_busy;
-	s->rx_cache_waive             += rq_stats->cache_waive;
 	s->rx_congst_umr              += rq_stats->congst_umr;
 	s->rx_arfs_err                += rq_stats->arfs_err;
 	s->rx_recover                 += rq_stats->recover;
@@ -1978,11 +1968,6 @@ static const struct counter_desc rq_stats_desc[] = {
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, buff_alloc_err) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cqe_compress_blks) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cqe_compress_pkts) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cache_reuse) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cache_full) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cache_empty) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cache_busy) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cache_waive) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, congst_umr) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, arfs_err) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, recover) },
@@ -2163,11 +2148,6 @@ static const struct counter_desc ptp_rq_stats_desc[] = {
 	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, buff_alloc_err) },
 	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cqe_compress_blks) },
 	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cqe_compress_pkts) },
-	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cache_reuse) },
-	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cache_full) },
-	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cache_empty) },
-	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cache_busy) },
-	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, cache_waive) },
 	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, congst_umr) },
 	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, arfs_err) },
 	{ MLX5E_DECLARE_PTP_RQ_STAT(struct mlx5e_rq_stats, recover) },
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index b77100b60b50..1ff8a06027dc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -193,11 +193,6 @@ struct mlx5e_sw_stats {
 	u64 rx_buff_alloc_err;
 	u64 rx_cqe_compress_blks;
 	u64 rx_cqe_compress_pkts;
-	u64 rx_cache_reuse;
-	u64 rx_cache_full;
-	u64 rx_cache_empty;
-	u64 rx_cache_busy;
-	u64 rx_cache_waive;
 	u64 rx_congst_umr;
 	u64 rx_arfs_err;
 	u64 rx_recover;
@@ -362,11 +357,6 @@ struct mlx5e_rq_stats {
 	u64 buff_alloc_err;
 	u64 cqe_compress_blks;
 	u64 cqe_compress_pkts;
-	u64 cache_reuse;
-	u64 cache_full;
-	u64 cache_empty;
-	u64 cache_busy;
-	u64 cache_waive;
 	u64 congst_umr;
 	u64 arfs_err;
 	u64 recover;
-- 
cgit v1.2.3-70-g09d2


From 6e9e141a7a28520a1cd13c96ad9127860e32ffbb Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Tue, 28 Mar 2023 05:08:12 +0300
Subject: selftests/bpf: Verifier/xdp_direct_packet_access.c converted to
 inline assembly

Test verifier/xdp_direct_packet_access.c automatically converted to use inline assembly.
Original test would be removed in the next patch.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230328020813.392560-2-eddyz87@gmail.com
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |    2 +
 .../bpf/progs/verifier_xdp_direct_packet_access.c  | 1722 ++++++++++++++++++++
 2 files changed, 1724 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_xdp_direct_packet_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index a774d5b193f1..efc8cf2e18d0 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -40,6 +40,7 @@
 #include "verifier_var_off.skel.h"
 #include "verifier_xadd.skel.h"
 #include "verifier_xdp.skel.h"
+#include "verifier_xdp_direct_packet_access.skel.h"
 
 __maybe_unused
 static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
@@ -102,3 +103,4 @@ void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
 void test_verifier_var_off(void)              { RUN(verifier_var_off); }
 void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
 void test_verifier_xdp(void)                  { RUN(verifier_xdp); }
+void test_verifier_xdp_direct_packet_access(void) { RUN(verifier_xdp_direct_packet_access); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_xdp_direct_packet_access.c b/tools/testing/selftests/bpf/progs/verifier_xdp_direct_packet_access.c
new file mode 100644
index 000000000000..df2dfd1b15d1
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_xdp_direct_packet_access.c
@@ -0,0 +1,1722 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/xdp_direct_packet_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end mangling, bad access 1")
+__failure __msg("R3 pointer arithmetic on pkt_end")
+__naked void end_mangling_bad_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	r3 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end mangling, bad access 2")
+__failure __msg("R3 pointer arithmetic on pkt_end")
+__naked void end_mangling_bad_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	r3 -= 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' > pkt_end, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void end_corner_case_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' > pkt_end, bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_end_bad_access_1_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 4);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' > pkt_end, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_end_bad_access_2_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' > pkt_end, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 9);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' > pkt_end, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end > pkt_data', good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void end_pkt_data_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end > pkt_data', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 6);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end > pkt_data', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 > r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end > pkt_data', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end > pkt_data', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' < pkt_end, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_pkt_end_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' < pkt_end, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 6);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' < pkt_end, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_end_bad_access_2_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' < pkt_end, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void end_corner_case_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' < pkt_end, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end < pkt_data', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end < pkt_data', bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_1_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 4);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end < pkt_data', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 < r1 goto l0_%=;				\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end < pkt_data', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 9);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end < pkt_data', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' >= pkt_end, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_pkt_end_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u32*)(r1 - 5);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' >= pkt_end, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_5(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 6);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' >= pkt_end, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_end_bad_access_2_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 >= r3 goto l0_%=;				\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' >= pkt_end, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void end_corner_case_good_access_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' >= pkt_end, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_5(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end >= pkt_data', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end >= pkt_data', bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_1_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 4);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end >= pkt_data', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 >= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end >= pkt_data', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_6(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 9);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end >= pkt_data', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_6(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' <= pkt_end, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void end_corner_case_good_access_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' <= pkt_end, bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_end_bad_access_1_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 4);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' <= pkt_end, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_end_bad_access_2_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 <= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' <= pkt_end, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_7(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 9);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data' <= pkt_end, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_7(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end <= pkt_data', good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void end_pkt_data_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u32*)(r1 - 5);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end <= pkt_data', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_8(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 6);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end <= pkt_data', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 <= r1 goto l0_%=;				\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end <= pkt_data', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_end <= pkt_data', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_8(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' > pkt_data, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_5(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' > pkt_data, bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_1_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 4);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' > pkt_data, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_5(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 > r3 goto l0_%=;				\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' > pkt_data, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_9(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 9);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' > pkt_data, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_9(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data > pkt_meta', good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_pkt_meta_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data > pkt_meta', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_10(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 6);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data > pkt_meta', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_meta_bad_access_2_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 > r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data > pkt_meta', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void meta_corner_case_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data > pkt_meta', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_10(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 > r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' < pkt_data, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void meta_pkt_data_good_access_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' < pkt_data, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_11(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 6);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' < pkt_data, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_6(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' < pkt_data, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_6(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' < pkt_data, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_11(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 < r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data < pkt_meta', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void meta_corner_case_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data < pkt_meta', bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_meta_bad_access_1_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 4);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data < pkt_meta', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_meta_bad_access_2_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 < r1 goto l0_%=;				\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data < pkt_meta', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_12(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 9);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data < pkt_meta', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_12(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 < r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' >= pkt_data, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void meta_pkt_data_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u32*)(r1 - 5);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' >= pkt_data, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_13(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 6);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' >= pkt_data, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_7(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 >= r3 goto l0_%=;				\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' >= pkt_data, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_7(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' >= pkt_data, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_13(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 >= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data >= pkt_meta', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void meta_corner_case_good_access_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data >= pkt_meta', bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_meta_bad_access_1_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 4);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data >= pkt_meta', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_meta_bad_access_2_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 >= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data >= pkt_meta', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_14(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 9);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data >= pkt_meta', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_14(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 >= r1 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' <= pkt_data, corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_corner_case_good_access_8(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' <= pkt_data, bad access 1")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_1_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 4);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' <= pkt_data, bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_data_bad_access_2_8(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r1 <= r3 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' <= pkt_data, corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_15(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 9;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 9);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_meta' <= pkt_data, corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_15(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r1 <= r3 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r1 - 7);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data <= pkt_meta', good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void data_pkt_meta_good_access_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u32*)(r1 - 5);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data <= pkt_meta', corner case -1, bad access")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_bad_access_16(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 6);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data <= pkt_meta', bad access 2")
+__failure __msg("R1 offset is outside of the packet")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void pkt_meta_bad_access_2_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 <= r1 goto l0_%=;				\
+l0_%=:	r0 = *(u32*)(r1 - 5);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data <= pkt_meta', corner case, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void meta_corner_case_good_access_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 7;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 7);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("XDP pkt read, pkt_data <= pkt_meta', corner case +1, good access")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void corner_case_1_good_access_16(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data_meta]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r1 = r2;					\
+	r1 += 8;					\
+	if r3 <= r1 goto l0_%=;				\
+	r0 = *(u64*)(r1 - 8);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_meta, offsetof(struct xdp_md, data_meta))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From c63a7d8bbb54a904f3ab8ff0aae39cd571b2c39c Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Tue, 28 Mar 2023 05:08:13 +0300
Subject: selftests/bpf: Remove verifier/xdp_direct_packet_access.c, converted
 to progs/verifier_xdp_direct_packet_access.c

Removing verifier/xdp_direct_packet_access.c.c as it was automatically converted to use
inline assembly in the previous commit. It is available in
progs/verifier_xdp_direct_packet_access.c.c.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230328020813.392560-3-eddyz87@gmail.com
---
 .../bpf/verifier/xdp_direct_packet_access.c        | 1468 --------------------
 1 file changed, 1468 deletions(-)
 delete mode 100644 tools/testing/selftests/bpf/verifier/xdp_direct_packet_access.c

diff --git a/tools/testing/selftests/bpf/verifier/xdp_direct_packet_access.c b/tools/testing/selftests/bpf/verifier/xdp_direct_packet_access.c
deleted file mode 100644
index b4ec228eb95d..000000000000
--- a/tools/testing/selftests/bpf/verifier/xdp_direct_packet_access.c
+++ /dev/null
@@ -1,1468 +0,0 @@
-{
-	"XDP pkt read, pkt_end mangling, bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R3 pointer arithmetic on pkt_end",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"XDP pkt read, pkt_end mangling, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_3, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R3 pointer arithmetic on pkt_end",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"XDP pkt read, pkt_data' > pkt_end, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' > pkt_end, bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' > pkt_end, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' > pkt_end, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' > pkt_end, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end > pkt_data', good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end > pkt_data', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end > pkt_data', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end > pkt_data', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end > pkt_data', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' < pkt_end, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' < pkt_end, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' < pkt_end, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' < pkt_end, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' < pkt_end, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end < pkt_data', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end < pkt_data', bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end < pkt_data', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end < pkt_data', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end < pkt_data', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' >= pkt_end, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' >= pkt_end, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' >= pkt_end, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' >= pkt_end, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' >= pkt_end, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end >= pkt_data', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end >= pkt_data', bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end >= pkt_data', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end >= pkt_data', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end >= pkt_data', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' <= pkt_end, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' <= pkt_end, bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' <= pkt_end, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' <= pkt_end, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data' <= pkt_end, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end <= pkt_data', good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end <= pkt_data', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end <= pkt_data', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end <= pkt_data', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_end <= pkt_data', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' > pkt_data, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' > pkt_data, bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' > pkt_data, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' > pkt_data, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' > pkt_data, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data > pkt_meta', good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data > pkt_meta', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data > pkt_meta', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data > pkt_meta', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data > pkt_meta', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' < pkt_data, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' < pkt_data, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' < pkt_data, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' < pkt_data, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' < pkt_data, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data < pkt_meta', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data < pkt_meta', bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data < pkt_meta', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data < pkt_meta', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data < pkt_meta', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' >= pkt_data, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' >= pkt_data, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' >= pkt_data, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' >= pkt_data, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' >= pkt_data, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data >= pkt_meta', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data >= pkt_meta', bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data >= pkt_meta', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data >= pkt_meta', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data >= pkt_meta', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' <= pkt_data, corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' <= pkt_data, bad access 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' <= pkt_data, bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' <= pkt_data, corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 9),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -9),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_meta' <= pkt_data, corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_1, BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data <= pkt_meta', good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data <= pkt_meta', corner case -1, bad access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data <= pkt_meta', bad access 2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, -5),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R1 offset is outside of the packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data <= pkt_meta', corner case, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 7),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"XDP pkt read, pkt_data <= pkt_meta', corner case +1, good access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct xdp_md, data_meta)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-- 
cgit v1.2.3-70-g09d2


From 163c2c7059178898ad7512294ca3a768df2c1515 Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@nvidia.com>
Date: Tue, 28 Mar 2023 13:07:23 -0700
Subject: net/mlx5e: Fix build break on 32bit

The cited commit caused the following build break in mlx5 due to a change
in size of MAX_SKB_FRAGS.

error: format '%lu' expects argument of type 'long unsigned int',
       but argument 7 has type 'unsigned int' [-Werror=format=]

Fix this by explicit casting.

Fixes: 3948b05950fd ("net: introduce a config option to tweak MAX_SKB_FRAGS")
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20230328200723.125122-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 90eac1aa5f6b..6db1aff8778d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -5725,8 +5725,8 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv)
 
 	/* Validate the max_wqe_size_sq capability. */
 	if (WARN_ON_ONCE(mlx5e_get_max_sq_wqebbs(priv->mdev) < MLX5E_MAX_TX_WQEBBS)) {
-		mlx5_core_warn(priv->mdev, "MLX5E: Max SQ WQEBBs firmware capability: %u, needed %lu\n",
-			       mlx5e_get_max_sq_wqebbs(priv->mdev), MLX5E_MAX_TX_WQEBBS);
+		mlx5_core_warn(priv->mdev, "MLX5E: Max SQ WQEBBs firmware capability: %u, needed %u\n",
+			       mlx5e_get_max_sq_wqebbs(priv->mdev), (unsigned int)MLX5E_MAX_TX_WQEBBS);
 		return -EIO;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From d288a162dd1c73507da582966f17dd226e34a0c0 Mon Sep 17 00:00:00 2001
From: Wangyang Guo <wangyang.guo@intel.com>
Date: Thu, 23 Mar 2023 21:55:29 +0100
Subject: net: dst: Prevent false sharing vs. dst_entry:: __refcnt

dst_entry::__refcnt is highly contended in scenarios where many connections
happen from and to the same IP. The reference count is an atomic_t, so the
reference count operations have to take the cache-line exclusive.

Aside of the unavoidable reference count contention there is another
significant problem which is caused by that: False sharing.

perf top identified two affected read accesses. dst_entry::lwtstate and
rtable::rt_genid.

dst_entry:__refcnt is located at offset 64 of dst_entry, which puts it into
a seperate cacheline vs. the read mostly members located at the beginning
of the struct.

That prevents false sharing vs. the struct members in the first 64
bytes of the structure, but there is also

  dst_entry::lwtstate

which is located after the reference count and in the same cache line. This
member is read after a reference count has been acquired.

struct rtable embeds a struct dst_entry at offset 0. struct dst_entry has a
size of 112 bytes, which means that the struct members of rtable which
follow the dst member share the same cache line as dst_entry::__refcnt.
Especially

  rtable::rt_genid

is also read by the contexts which have a reference count acquired
already.

When dst_entry:__refcnt is incremented or decremented via an atomic
operation these read accesses stall. This was found when analysing the
memtier benchmark in 1:100 mode, which amplifies the problem extremly.

Move the rt[6i]_uncached[_list] members out of struct rtable and struct
rt6_info into struct dst_entry to provide padding and move the lwtstate
member after that so it ends up in the same cache line.

The resulting improvement depends on the micro-architecture and the number
of CPUs. It ranges from +20% to +120% with a localhost memtier/memcached
benchmark.

[ tglx: Rearrange struct ]

Signed-off-by: Wangyang Guo <wangyang.guo@intel.com>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230323102800.042297517@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/dst.h       | 15 ++++++++++++++-
 include/net/ip6_fib.h   |  3 ---
 include/net/ip6_route.h |  2 +-
 include/net/route.h     |  3 ---
 net/ipv4/route.c        | 20 ++++++++++----------
 net/ipv4/xfrm4_policy.c |  4 ++--
 net/ipv6/route.c        | 26 +++++++++++++-------------
 net/ipv6/xfrm6_policy.c |  4 ++--
 8 files changed, 42 insertions(+), 35 deletions(-)

diff --git a/include/net/dst.h b/include/net/dst.h
index d67fda89cd0f..81f2279ea911 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -69,15 +69,28 @@ struct dst_entry {
 #endif
 	int			__use;
 	unsigned long		lastuse;
-	struct lwtunnel_state   *lwtstate;
 	struct rcu_head		rcu_head;
 	short			error;
 	short			__pad;
 	__u32			tclassid;
 #ifndef CONFIG_64BIT
+	struct lwtunnel_state   *lwtstate;
 	atomic_t		__refcnt;	/* 32-bit offset 64 */
 #endif
 	netdevice_tracker	dev_tracker;
+
+	/*
+	 * Used by rtable and rt6_info. Moves lwtstate into the next cache
+	 * line on 64bit so that lwtstate does not cause false sharing with
+	 * __refcnt under contention of __refcnt. This also puts the
+	 * frequently accessed members of rtable and rt6_info out of the
+	 * __refcnt cache line.
+	 */
+	struct list_head	rt_uncached;
+	struct uncached_list	*rt_uncached_list;
+#ifdef CONFIG_64BIT
+	struct lwtunnel_state   *lwtstate;
+#endif
 };
 
 struct dst_metrics {
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 6268963d9599..79570cb4ea9c 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -217,9 +217,6 @@ struct rt6_info {
 	struct inet6_dev		*rt6i_idev;
 	u32				rt6i_flags;
 
-	struct list_head		rt6i_uncached;
-	struct uncached_list		*rt6i_uncached_list;
-
 	/* more non-fragment space at head required */
 	unsigned short			rt6i_nfheader_len;
 };
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 81ee387a1fc4..3556595ce59a 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -100,7 +100,7 @@ static inline struct dst_entry *ip6_route_output(struct net *net,
 static inline void ip6_rt_put_flags(struct rt6_info *rt, int flags)
 {
 	if (!(flags & RT6_LOOKUP_F_DST_NOREF) ||
-	    !list_empty(&rt->rt6i_uncached))
+	    !list_empty(&rt->dst.rt_uncached))
 		ip6_rt_put(rt);
 }
 
diff --git a/include/net/route.h b/include/net/route.h
index fe00b0a2e475..bcc367cf3aa2 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -78,9 +78,6 @@ struct rtable {
 	/* Miscellaneous cached information */
 	u32			rt_mtu_locked:1,
 				rt_pmtu:31;
-
-	struct list_head	rt_uncached;
-	struct uncached_list	*rt_uncached_list;
 };
 
 static inline bool rt_is_input_route(const struct rtable *rt)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 6a0a0bb452e9..2a3d14d95ada 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1508,20 +1508,20 @@ void rt_add_uncached_list(struct rtable *rt)
 {
 	struct uncached_list *ul = raw_cpu_ptr(&rt_uncached_list);
 
-	rt->rt_uncached_list = ul;
+	rt->dst.rt_uncached_list = ul;
 
 	spin_lock_bh(&ul->lock);
-	list_add_tail(&rt->rt_uncached, &ul->head);
+	list_add_tail(&rt->dst.rt_uncached, &ul->head);
 	spin_unlock_bh(&ul->lock);
 }
 
 void rt_del_uncached_list(struct rtable *rt)
 {
-	if (!list_empty(&rt->rt_uncached)) {
-		struct uncached_list *ul = rt->rt_uncached_list;
+	if (!list_empty(&rt->dst.rt_uncached)) {
+		struct uncached_list *ul = rt->dst.rt_uncached_list;
 
 		spin_lock_bh(&ul->lock);
-		list_del_init(&rt->rt_uncached);
+		list_del_init(&rt->dst.rt_uncached);
 		spin_unlock_bh(&ul->lock);
 	}
 }
@@ -1546,13 +1546,13 @@ void rt_flush_dev(struct net_device *dev)
 			continue;
 
 		spin_lock_bh(&ul->lock);
-		list_for_each_entry_safe(rt, safe, &ul->head, rt_uncached) {
+		list_for_each_entry_safe(rt, safe, &ul->head, dst.rt_uncached) {
 			if (rt->dst.dev != dev)
 				continue;
 			rt->dst.dev = blackhole_netdev;
 			netdev_ref_replace(dev, blackhole_netdev,
 					   &rt->dst.dev_tracker, GFP_ATOMIC);
-			list_move(&rt->rt_uncached, &ul->quarantine);
+			list_move(&rt->dst.rt_uncached, &ul->quarantine);
 		}
 		spin_unlock_bh(&ul->lock);
 	}
@@ -1644,7 +1644,7 @@ struct rtable *rt_dst_alloc(struct net_device *dev,
 		rt->rt_uses_gateway = 0;
 		rt->rt_gw_family = 0;
 		rt->rt_gw4 = 0;
-		INIT_LIST_HEAD(&rt->rt_uncached);
+		INIT_LIST_HEAD(&rt->dst.rt_uncached);
 
 		rt->dst.output = ip_output;
 		if (flags & RTCF_LOCAL)
@@ -1675,7 +1675,7 @@ struct rtable *rt_dst_clone(struct net_device *dev, struct rtable *rt)
 			new_rt->rt_gw4 = rt->rt_gw4;
 		else if (rt->rt_gw_family == AF_INET6)
 			new_rt->rt_gw6 = rt->rt_gw6;
-		INIT_LIST_HEAD(&new_rt->rt_uncached);
+		INIT_LIST_HEAD(&new_rt->dst.rt_uncached);
 
 		new_rt->dst.input = rt->dst.input;
 		new_rt->dst.output = rt->dst.output;
@@ -2859,7 +2859,7 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or
 		else if (rt->rt_gw_family == AF_INET6)
 			rt->rt_gw6 = ort->rt_gw6;
 
-		INIT_LIST_HEAD(&rt->rt_uncached);
+		INIT_LIST_HEAD(&rt->dst.rt_uncached);
 	}
 
 	dst_release(dst_orig);
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 3d0dfa6cf9f9..47861c8b7340 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -91,7 +91,7 @@ static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
 		xdst->u.rt.rt_gw6 = rt->rt_gw6;
 	xdst->u.rt.rt_pmtu = rt->rt_pmtu;
 	xdst->u.rt.rt_mtu_locked = rt->rt_mtu_locked;
-	INIT_LIST_HEAD(&xdst->u.rt.rt_uncached);
+	INIT_LIST_HEAD(&xdst->u.rt.dst.rt_uncached);
 	rt_add_uncached_list(&xdst->u.rt);
 
 	return 0;
@@ -121,7 +121,7 @@ static void xfrm4_dst_destroy(struct dst_entry *dst)
 	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
 
 	dst_destroy_metrics_generic(dst);
-	if (xdst->u.rt.rt_uncached_list)
+	if (xdst->u.rt.dst.rt_uncached_list)
 		rt_del_uncached_list(&xdst->u.rt);
 	xfrm_dst_destroy(xdst);
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 244df77fac87..12e163dec34e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -139,20 +139,20 @@ void rt6_uncached_list_add(struct rt6_info *rt)
 {
 	struct uncached_list *ul = raw_cpu_ptr(&rt6_uncached_list);
 
-	rt->rt6i_uncached_list = ul;
+	rt->dst.rt_uncached_list = ul;
 
 	spin_lock_bh(&ul->lock);
-	list_add_tail(&rt->rt6i_uncached, &ul->head);
+	list_add_tail(&rt->dst.rt_uncached, &ul->head);
 	spin_unlock_bh(&ul->lock);
 }
 
 void rt6_uncached_list_del(struct rt6_info *rt)
 {
-	if (!list_empty(&rt->rt6i_uncached)) {
-		struct uncached_list *ul = rt->rt6i_uncached_list;
+	if (!list_empty(&rt->dst.rt_uncached)) {
+		struct uncached_list *ul = rt->dst.rt_uncached_list;
 
 		spin_lock_bh(&ul->lock);
-		list_del_init(&rt->rt6i_uncached);
+		list_del_init(&rt->dst.rt_uncached);
 		spin_unlock_bh(&ul->lock);
 	}
 }
@@ -169,7 +169,7 @@ static void rt6_uncached_list_flush_dev(struct net_device *dev)
 			continue;
 
 		spin_lock_bh(&ul->lock);
-		list_for_each_entry_safe(rt, safe, &ul->head, rt6i_uncached) {
+		list_for_each_entry_safe(rt, safe, &ul->head, dst.rt_uncached) {
 			struct inet6_dev *rt_idev = rt->rt6i_idev;
 			struct net_device *rt_dev = rt->dst.dev;
 			bool handled = false;
@@ -188,7 +188,7 @@ static void rt6_uncached_list_flush_dev(struct net_device *dev)
 				handled = true;
 			}
 			if (handled)
-				list_move(&rt->rt6i_uncached,
+				list_move(&rt->dst.rt_uncached,
 					  &ul->quarantine);
 		}
 		spin_unlock_bh(&ul->lock);
@@ -334,7 +334,7 @@ static const struct rt6_info ip6_blk_hole_entry_template = {
 static void rt6_info_init(struct rt6_info *rt)
 {
 	memset_after(rt, 0, dst);
-	INIT_LIST_HEAD(&rt->rt6i_uncached);
+	INIT_LIST_HEAD(&rt->dst.rt_uncached);
 }
 
 /* allocate dst with ip6_dst_ops */
@@ -2638,7 +2638,7 @@ struct dst_entry *ip6_route_output_flags(struct net *net,
 	dst = ip6_route_output_flags_noref(net, sk, fl6, flags);
 	rt6 = (struct rt6_info *)dst;
 	/* For dst cached in uncached_list, refcnt is already taken. */
-	if (list_empty(&rt6->rt6i_uncached) && !dst_hold_safe(dst)) {
+	if (list_empty(&rt6->dst.rt_uncached) && !dst_hold_safe(dst)) {
 		dst = &net->ipv6.ip6_null_entry->dst;
 		dst_hold(dst);
 	}
@@ -2748,7 +2748,7 @@ INDIRECT_CALLABLE_SCOPE struct dst_entry *ip6_dst_check(struct dst_entry *dst,
 	from = rcu_dereference(rt->from);
 
 	if (from && (rt->rt6i_flags & RTF_PCPU ||
-	    unlikely(!list_empty(&rt->rt6i_uncached))))
+	    unlikely(!list_empty(&rt->dst.rt_uncached))))
 		dst_ret = rt6_dst_from_check(rt, from, cookie);
 	else
 		dst_ret = rt6_check(rt, from, cookie);
@@ -6477,7 +6477,7 @@ static int __net_init ip6_route_net_init(struct net *net)
 	net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops;
 	dst_init_metrics(&net->ipv6.ip6_null_entry->dst,
 			 ip6_template_metrics, true);
-	INIT_LIST_HEAD(&net->ipv6.ip6_null_entry->rt6i_uncached);
+	INIT_LIST_HEAD(&net->ipv6.ip6_null_entry->dst.rt_uncached);
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	net->ipv6.fib6_has_custom_rules = false;
@@ -6489,7 +6489,7 @@ static int __net_init ip6_route_net_init(struct net *net)
 	net->ipv6.ip6_prohibit_entry->dst.ops = &net->ipv6.ip6_dst_ops;
 	dst_init_metrics(&net->ipv6.ip6_prohibit_entry->dst,
 			 ip6_template_metrics, true);
-	INIT_LIST_HEAD(&net->ipv6.ip6_prohibit_entry->rt6i_uncached);
+	INIT_LIST_HEAD(&net->ipv6.ip6_prohibit_entry->dst.rt_uncached);
 
 	net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
 					       sizeof(*net->ipv6.ip6_blk_hole_entry),
@@ -6499,7 +6499,7 @@ static int __net_init ip6_route_net_init(struct net *net)
 	net->ipv6.ip6_blk_hole_entry->dst.ops = &net->ipv6.ip6_dst_ops;
 	dst_init_metrics(&net->ipv6.ip6_blk_hole_entry->dst,
 			 ip6_template_metrics, true);
-	INIT_LIST_HEAD(&net->ipv6.ip6_blk_hole_entry->rt6i_uncached);
+	INIT_LIST_HEAD(&net->ipv6.ip6_blk_hole_entry->dst.rt_uncached);
 #ifdef CONFIG_IPV6_SUBTREES
 	net->ipv6.fib6_routes_require_src = 0;
 #endif
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index ea435eba3053..2b493f8d0091 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -89,7 +89,7 @@ static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
 	xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway;
 	xdst->u.rt6.rt6i_dst = rt->rt6i_dst;
 	xdst->u.rt6.rt6i_src = rt->rt6i_src;
-	INIT_LIST_HEAD(&xdst->u.rt6.rt6i_uncached);
+	INIT_LIST_HEAD(&xdst->u.rt6.dst.rt_uncached);
 	rt6_uncached_list_add(&xdst->u.rt6);
 
 	return 0;
@@ -121,7 +121,7 @@ static void xfrm6_dst_destroy(struct dst_entry *dst)
 	if (likely(xdst->u.rt6.rt6i_idev))
 		in6_dev_put(xdst->u.rt6.rt6i_idev);
 	dst_destroy_metrics_generic(dst);
-	if (xdst->u.rt6.rt6i_uncached_list)
+	if (xdst->u.rt6.dst.rt_uncached_list)
 		rt6_uncached_list_del(&xdst->u.rt6);
 	xfrm_dst_destroy(xdst);
 }
-- 
cgit v1.2.3-70-g09d2


From bc9d3a9f2afca189a6ae40225b6985e3c775375e Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 23 Mar 2023 21:55:32 +0100
Subject: net: dst: Switch to rcuref_t reference counting

Under high contention dst_entry::__refcnt becomes a significant bottleneck.

atomic_inc_not_zero() is implemented with a cmpxchg() loop, which goes into
high retry rates on contention.

Switch the reference count to rcuref_t which results in a significant
performance gain. Rename the reference count member to __rcuref to reflect
the change.

The gain depends on the micro-architecture and the number of concurrent
operations and has been measured in the range of +25% to +130% with a
localhost memtier/memcached benchmark which amplifies the problem
massively.

Running the memtier/memcached benchmark over a real (1Gb) network
connection the conversion on top of the false sharing fix for struct
dst_entry::__refcnt results in a total gain in the 2%-5% range over the
upstream baseline.

Reported-by: Wangyang Guo <wangyang.guo@intel.com>
Reported-by: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20230307125538.989175656@linutronix.de
Link: https://lore.kernel.org/r/20230323102800.215027837@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/dst.h               | 19 ++++++++++---------
 include/net/sock.h              |  2 +-
 net/bridge/br_nf_core.c         |  2 +-
 net/core/dst.c                  | 26 +++++---------------------
 net/core/rtnetlink.c            |  2 +-
 net/ipv6/route.c                |  6 +++---
 net/netfilter/ipvs/ip_vs_xmit.c |  4 ++--
 7 files changed, 23 insertions(+), 38 deletions(-)

diff --git a/include/net/dst.h b/include/net/dst.h
index 81f2279ea911..78884429deed 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -16,6 +16,7 @@
 #include <linux/bug.h>
 #include <linux/jiffies.h>
 #include <linux/refcount.h>
+#include <linux/rcuref.h>
 #include <net/neighbour.h>
 #include <asm/processor.h>
 #include <linux/indirect_call_wrapper.h>
@@ -61,11 +62,11 @@ struct dst_entry {
 	unsigned short		trailer_len;	/* space to reserve at tail */
 
 	/*
-	 * __refcnt wants to be on a different cache line from
+	 * __rcuref wants to be on a different cache line from
 	 * input/output/ops or performance tanks badly
 	 */
 #ifdef CONFIG_64BIT
-	atomic_t		__refcnt;	/* 64-bit offset 64 */
+	rcuref_t		__rcuref;	/* 64-bit offset 64 */
 #endif
 	int			__use;
 	unsigned long		lastuse;
@@ -75,16 +76,16 @@ struct dst_entry {
 	__u32			tclassid;
 #ifndef CONFIG_64BIT
 	struct lwtunnel_state   *lwtstate;
-	atomic_t		__refcnt;	/* 32-bit offset 64 */
+	rcuref_t		__rcuref;	/* 32-bit offset 64 */
 #endif
 	netdevice_tracker	dev_tracker;
 
 	/*
 	 * Used by rtable and rt6_info. Moves lwtstate into the next cache
 	 * line on 64bit so that lwtstate does not cause false sharing with
-	 * __refcnt under contention of __refcnt. This also puts the
+	 * __rcuref under contention of __rcuref. This also puts the
 	 * frequently accessed members of rtable and rt6_info out of the
-	 * __refcnt cache line.
+	 * __rcuref cache line.
 	 */
 	struct list_head	rt_uncached;
 	struct uncached_list	*rt_uncached_list;
@@ -238,10 +239,10 @@ static inline void dst_hold(struct dst_entry *dst)
 {
 	/*
 	 * If your kernel compilation stops here, please check
-	 * the placement of __refcnt in struct dst_entry
+	 * the placement of __rcuref in struct dst_entry
 	 */
-	BUILD_BUG_ON(offsetof(struct dst_entry, __refcnt) & 63);
-	WARN_ON(atomic_inc_not_zero(&dst->__refcnt) == 0);
+	BUILD_BUG_ON(offsetof(struct dst_entry, __rcuref) & 63);
+	WARN_ON(!rcuref_get(&dst->__rcuref));
 }
 
 static inline void dst_use_noref(struct dst_entry *dst, unsigned long time)
@@ -305,7 +306,7 @@ static inline void skb_dst_copy(struct sk_buff *nskb, const struct sk_buff *oskb
  */
 static inline bool dst_hold_safe(struct dst_entry *dst)
 {
-	return atomic_inc_not_zero(&dst->__refcnt);
+	return rcuref_get(&dst->__rcuref);
 }
 
 /**
diff --git a/include/net/sock.h b/include/net/sock.h
index 573f2bf7e0de..5edf0038867c 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2131,7 +2131,7 @@ sk_dst_get(struct sock *sk)
 
 	rcu_read_lock();
 	dst = rcu_dereference(sk->sk_dst_cache);
-	if (dst && !atomic_inc_not_zero(&dst->__refcnt))
+	if (dst && !rcuref_get(&dst->__rcuref))
 		dst = NULL;
 	rcu_read_unlock();
 	return dst;
diff --git a/net/bridge/br_nf_core.c b/net/bridge/br_nf_core.c
index 8c69f0c95a8e..98aea5485aae 100644
--- a/net/bridge/br_nf_core.c
+++ b/net/bridge/br_nf_core.c
@@ -73,7 +73,7 @@ void br_netfilter_rtable_init(struct net_bridge *br)
 {
 	struct rtable *rt = &br->fake_rtable;
 
-	atomic_set(&rt->dst.__refcnt, 1);
+	rcuref_init(&rt->dst.__rcuref, 1);
 	rt->dst.dev = br->dev;
 	dst_init_metrics(&rt->dst, br_dst_default_metrics, true);
 	rt->dst.flags	= DST_NOXFRM | DST_FAKE_RTABLE;
diff --git a/net/core/dst.c b/net/core/dst.c
index 31c08a3386d3..3247e84045ca 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -66,7 +66,7 @@ void dst_init(struct dst_entry *dst, struct dst_ops *ops,
 	dst->tclassid = 0;
 #endif
 	dst->lwtstate = NULL;
-	atomic_set(&dst->__refcnt, initial_ref);
+	rcuref_init(&dst->__rcuref, initial_ref);
 	dst->__use = 0;
 	dst->lastuse = jiffies;
 	dst->flags = flags;
@@ -162,31 +162,15 @@ EXPORT_SYMBOL(dst_dev_put);
 
 void dst_release(struct dst_entry *dst)
 {
-	if (dst) {
-		int newrefcnt;
-
-		newrefcnt = atomic_dec_return(&dst->__refcnt);
-		if (WARN_ONCE(newrefcnt < 0, "dst_release underflow"))
-			net_warn_ratelimited("%s: dst:%p refcnt:%d\n",
-					     __func__, dst, newrefcnt);
-		if (!newrefcnt)
-			call_rcu_hurry(&dst->rcu_head, dst_destroy_rcu);
-	}
+	if (dst && rcuref_put(&dst->__rcuref))
+		call_rcu_hurry(&dst->rcu_head, dst_destroy_rcu);
 }
 EXPORT_SYMBOL(dst_release);
 
 void dst_release_immediate(struct dst_entry *dst)
 {
-	if (dst) {
-		int newrefcnt;
-
-		newrefcnt = atomic_dec_return(&dst->__refcnt);
-		if (WARN_ONCE(newrefcnt < 0, "dst_release_immediate underflow"))
-			net_warn_ratelimited("%s: dst:%p refcnt:%d\n",
-					     __func__, dst, newrefcnt);
-		if (!newrefcnt)
-			dst_destroy(dst);
-	}
+	if (dst && rcuref_put(&dst->__rcuref))
+		dst_destroy(dst);
 }
 EXPORT_SYMBOL(dst_release_immediate);
 
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index b7b1661d0d56..906aebdc566b 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -843,7 +843,7 @@ int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
 	if (dst) {
 		ci.rta_lastuse = jiffies_delta_to_clock_t(jiffies - dst->lastuse);
 		ci.rta_used = dst->__use;
-		ci.rta_clntref = atomic_read(&dst->__refcnt);
+		ci.rta_clntref = rcuref_read(&dst->__rcuref);
 	}
 	if (expires) {
 		unsigned long clock;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 12e163dec34e..35085fc0cf15 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -293,7 +293,7 @@ static const struct fib6_info fib6_null_entry_template = {
 
 static const struct rt6_info ip6_null_entry_template = {
 	.dst = {
-		.__refcnt	= ATOMIC_INIT(1),
+		.__rcuref	= RCUREF_INIT(1),
 		.__use		= 1,
 		.obsolete	= DST_OBSOLETE_FORCE_CHK,
 		.error		= -ENETUNREACH,
@@ -307,7 +307,7 @@ static const struct rt6_info ip6_null_entry_template = {
 
 static const struct rt6_info ip6_prohibit_entry_template = {
 	.dst = {
-		.__refcnt	= ATOMIC_INIT(1),
+		.__rcuref	= RCUREF_INIT(1),
 		.__use		= 1,
 		.obsolete	= DST_OBSOLETE_FORCE_CHK,
 		.error		= -EACCES,
@@ -319,7 +319,7 @@ static const struct rt6_info ip6_prohibit_entry_template = {
 
 static const struct rt6_info ip6_blk_hole_entry_template = {
 	.dst = {
-		.__refcnt	= ATOMIC_INIT(1),
+		.__rcuref	= RCUREF_INIT(1),
 		.__use		= 1,
 		.obsolete	= DST_OBSOLETE_FORCE_CHK,
 		.error		= -EINVAL,
diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
index 80448885c3d7..99c349c0d968 100644
--- a/net/netfilter/ipvs/ip_vs_xmit.c
+++ b/net/netfilter/ipvs/ip_vs_xmit.c
@@ -339,7 +339,7 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
 			spin_unlock_bh(&dest->dst_lock);
 			IP_VS_DBG(10, "new dst %pI4, src %pI4, refcnt=%d\n",
 				  &dest->addr.ip, &dest_dst->dst_saddr.ip,
-				  atomic_read(&rt->dst.__refcnt));
+				  rcuref_read(&rt->dst.__rcuref));
 		}
 		if (ret_saddr)
 			*ret_saddr = dest_dst->dst_saddr.ip;
@@ -507,7 +507,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
 			spin_unlock_bh(&dest->dst_lock);
 			IP_VS_DBG(10, "new dst %pI6, src %pI6, refcnt=%d\n",
 				  &dest->addr.in6, &dest_dst->dst_saddr.in6,
-				  atomic_read(&rt->dst.__refcnt));
+				  rcuref_read(&rt->dst.__rcuref));
 		}
 		if (ret_saddr)
 			*ret_saddr = dest_dst->dst_saddr.in6;
-- 
cgit v1.2.3-70-g09d2


From cdeccd13a03fc6e97c090c12a27566d02dace6d9 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Mon, 27 Mar 2023 17:21:12 +0200
Subject: Revert "sh_eth: remove open coded netif_running()"

This reverts commit ce1fdb065695f49ef6f126d35c1abbfe645d62d5. It turned
out this actually introduces a race condition. netif_running() is not a
suitable check for get_stats.

Reported-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230327152112.15635-1-wsa+renesas@sang-engineering.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/renesas/sh_eth.c | 6 +++++-
 drivers/net/ethernet/renesas/sh_eth.h | 1 +
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 2d9787231099..d8ec729825be 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -2441,6 +2441,8 @@ static int sh_eth_open(struct net_device *ndev)
 
 	netif_start_queue(ndev);
 
+	mdp->is_opened = 1;
+
 	return ret;
 
 out_free_irq:
@@ -2563,7 +2565,7 @@ static struct net_device_stats *sh_eth_get_stats(struct net_device *ndev)
 	if (mdp->cd->no_tx_cntrs)
 		return &ndev->stats;
 
-	if (!netif_running(ndev))
+	if (!mdp->is_opened)
 		return &ndev->stats;
 
 	sh_eth_update_stat(ndev, &ndev->stats.tx_dropped, TROCR);
@@ -2612,6 +2614,8 @@ static int sh_eth_close(struct net_device *ndev)
 	/* Free all the skbuffs in the Rx queue and the DMA buffer. */
 	sh_eth_ring_free(ndev);
 
+	mdp->is_opened = 0;
+
 	pm_runtime_put(&mdp->pdev->dev);
 
 	return 0;
diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h
index f56dbc8a064a..a5c07c6ff44a 100644
--- a/drivers/net/ethernet/renesas/sh_eth.h
+++ b/drivers/net/ethernet/renesas/sh_eth.h
@@ -560,6 +560,7 @@ struct sh_eth_private {
 
 	unsigned no_ether_link:1;
 	unsigned ether_link_active_low:1;
+	unsigned is_opened:1;
 	unsigned wol_enabled:1;
 };
 
-- 
cgit v1.2.3-70-g09d2


From e48cefb9c8d1308cd64c40ebabfd51d79bf6e93b Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Mon, 27 Mar 2023 19:54:23 -0400
Subject: net: ethernet: 8390: axnet_cs: remove unused xfer_count variable

clang with W=1 reports
drivers/net/ethernet/8390/axnet_cs.c:653:9: error: variable
  'xfer_count' set but not used [-Werror,-Wunused-but-set-variable]
    int xfer_count = count;
        ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230327235423.1777590-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/8390/axnet_cs.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/8390/axnet_cs.c b/drivers/net/ethernet/8390/axnet_cs.c
index 3aef959fc25b..78f985885547 100644
--- a/drivers/net/ethernet/8390/axnet_cs.c
+++ b/drivers/net/ethernet/8390/axnet_cs.c
@@ -650,7 +650,6 @@ static void block_input(struct net_device *dev, int count,
 {
     unsigned int nic_base = dev->base_addr;
     struct ei_device *ei_local = netdev_priv(dev);
-    int xfer_count = count;
     char *buf = skb->data;
 
     if ((netif_msg_rx_status(ei_local)) && (count != 4))
@@ -662,9 +661,7 @@ static void block_input(struct net_device *dev, int count,
     insw(nic_base + AXNET_DATAPORT,buf,count>>1);
     if (count & 0x01) {
 	buf[count-1] = inb(nic_base + AXNET_DATAPORT);
-	xfer_count++;
     }
-
 }
 
 /*====================================================================*/
-- 
cgit v1.2.3-70-g09d2


From e70f94c6c75c2e3274df474b4c0bf6153e0533c3 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 27 Mar 2023 10:26:46 -0700
Subject: docs: netdev: clarify the need to sending reverts as patches

We don't state explicitly that reverts need to be submitted
as a patch. It occasionally comes up.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230327172646.2622943-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/process/maintainer-netdev.rst | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/Documentation/process/maintainer-netdev.rst b/Documentation/process/maintainer-netdev.rst
index e31d7a951073..f73ac9e175a8 100644
--- a/Documentation/process/maintainer-netdev.rst
+++ b/Documentation/process/maintainer-netdev.rst
@@ -184,11 +184,18 @@ Handling misapplied patches
 
 Occasionally a patch series gets applied before receiving critical feedback,
 or the wrong version of a series gets applied.
-There is no revert possible, once it is pushed out, it stays like that.
+
+Making the patch disappear once it is pushed out is not possible, the commit
+history in netdev trees is immutable.
 Please send incremental versions on top of what has been merged in order to fix
 the patches the way they would look like if your latest patch series was to be
 merged.
 
+In cases where full revert is needed the revert has to be submitted
+as a patch to the list with a commit message explaining the technical
+problems with the reverted commit. Reverts should be used as a last resort,
+when original change is completely wrong; incremental fixes are preferred.
+
 Stable tree
 ~~~~~~~~~~~
 
-- 
cgit v1.2.3-70-g09d2


From bec0b7a2db358718a371e6fe0103220685552fae Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:32 +0100
Subject: tools: ynl: Add struct parsing to nlspec

Add python classes for struct definitions to nlspec

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/nlspec.py | 43 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index dba70100124a..83de2a1a3cc6 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -214,6 +214,44 @@ class SpecAttrSet(SpecElement):
         return self.attrs.items()
 
 
+class SpecStructMember(SpecElement):
+    """Struct member attribute
+
+    Represents a single struct member attribute.
+
+    Attributes:
+        type    string, type of the member attribute
+    """
+    def __init__(self, family, yaml):
+        super().__init__(family, yaml)
+        self.type = yaml['type']
+
+
+class SpecStruct(SpecElement):
+    """Netlink struct type
+
+    Represents a C struct definition.
+
+    Attributes:
+        members   ordered list of struct members
+    """
+    def __init__(self, family, yaml):
+        super().__init__(family, yaml)
+
+        self.members = []
+        for member in yaml.get('members', []):
+            self.members.append(self.new_member(family, member))
+
+    def new_member(self, family, elem):
+        return SpecStructMember(family, elem)
+
+    def __iter__(self):
+        yield from self.members
+
+    def items(self):
+        return self.members.items()
+
+
 class SpecOperation(SpecElement):
     """Netlink Operation
 
@@ -344,6 +382,9 @@ class SpecFamily(SpecElement):
     def new_attr_set(self, elem):
         return SpecAttrSet(self, elem)
 
+    def new_struct(self, elem):
+        return SpecStruct(self, elem)
+
     def new_operation(self, elem, req_val, rsp_val):
         return SpecOperation(self, elem, req_val, rsp_val)
 
@@ -399,6 +440,8 @@ class SpecFamily(SpecElement):
         for elem in definitions:
             if elem['type'] == 'enum' or elem['type'] == 'flags':
                 self.consts[elem['name']] = self.new_enum(elem)
+            elif elem['type'] == 'struct':
+                self.consts[elem['name']] = self.new_struct(elem)
             else:
                 self.consts[elem['name']] = elem
 
-- 
cgit v1.2.3-70-g09d2


From b423c3c86325192259380ac870aafd370a683e73 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:33 +0100
Subject: tools: ynl: Add C array attribute decoding to ynl

Add support for decoding C arrays from binay blobs in genetlink-legacy
messages.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/nlspec.py |  7 +++++--
 tools/net/ynl/lib/ynl.py    | 18 +++++++++++++++++-
 2 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index 83de2a1a3cc6..6cc9b7646ae8 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -149,8 +149,10 @@ class SpecAttr(SpecElement):
     Represents a single attribute type within an attr space.
 
     Attributes:
-        value      numerical ID when serialized
-        attr_set   Attribute Set containing this attr
+        value         numerical ID when serialized
+        attr_set      Attribute Set containing this attr
+        is_multi      bool, attr may repeat multiple times
+        sub_type      string, name of sub type
     """
     def __init__(self, family, attr_set, yaml, value):
         super().__init__(family, yaml)
@@ -158,6 +160,7 @@ class SpecAttr(SpecElement):
         self.value = value
         self.attr_set = attr_set
         self.is_multi = yaml.get('multi-attr', False)
+        self.sub_type = yaml.get('sub-type')
 
 
 class SpecAttrSet(SpecElement):
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 7eaf066b115e..eada229402fa 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -68,6 +68,11 @@ class Netlink:
 
 
 class NlAttr:
+    type_formats = { 'u8' : ('B', 1), 's8' : ('b', 1),
+                     'u16': ('H', 2), 's16': ('h', 2),
+                     'u32': ('I', 4), 's32': ('i', 4),
+                     'u64': ('Q', 8), 's64': ('q', 8) }
+
     def __init__(self, raw, offset):
         self._len, self._type = struct.unpack("HH", raw[offset:offset + 4])
         self.type = self._type & ~Netlink.NLA_TYPE_MASK
@@ -93,6 +98,10 @@ class NlAttr:
     def as_bin(self):
         return self.raw
 
+    def as_c_array(self, type):
+        format, _ = self.type_formats[type]
+        return list({ x[0] for x in struct.iter_unpack(format, self.raw) })
+
     def __repr__(self):
         return f"[type:{self.type} len:{self._len}] {self.raw}"
 
@@ -367,6 +376,13 @@ class YnlFamily(SpecFamily):
             value = enum.entries_by_val[raw - i].name
         rsp[attr_spec['name']] = value
 
+    def _decode_binary(self, attr, attr_spec):
+        if attr_spec.sub_type:
+            decoded = attr.as_c_array(attr_spec.sub_type)
+        else:
+            decoded = attr.as_bin()
+        return decoded
+
     def _decode(self, attrs, space):
         attr_space = self.attr_sets[space]
         rsp = dict()
@@ -386,7 +402,7 @@ class YnlFamily(SpecFamily):
             elif attr_spec["type"] == 'string':
                 decoded = attr.as_strz()
             elif attr_spec["type"] == 'binary':
-                decoded = attr.as_bin()
+                decoded = self._decode_binary(attr, attr_spec)
             elif attr_spec["type"] == 'flag':
                 decoded = True
             else:
-- 
cgit v1.2.3-70-g09d2


From 2607191395bd4db544db05452625cd7e98bc0848 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:34 +0100
Subject: tools: ynl: Add struct attr decoding to ynl

Add support for decoding attributes that contain C structs.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/genetlink-legacy.yaml |  5 +++++
 tools/net/ynl/lib/nlspec.py                 |  2 ++
 tools/net/ynl/lib/ynl.py                    | 15 ++++++++++++++-
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/Documentation/netlink/genetlink-legacy.yaml b/Documentation/netlink/genetlink-legacy.yaml
index 5dc6f1c07a97..d50c78b9f42d 100644
--- a/Documentation/netlink/genetlink-legacy.yaml
+++ b/Documentation/netlink/genetlink-legacy.yaml
@@ -218,6 +218,11 @@ properties:
                     description: Max length for a string or a binary attribute.
                     $ref: '#/$defs/len-or-define'
               sub-type: *attr-type
+              # Start genetlink-legacy
+              struct:
+                description: Name of the struct type used for the attribute.
+                type: string
+              # End genetlink-legacy
 
       # Make sure name-prefix does not appear in subsets (subsets inherit naming)
       dependencies:
diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index 6cc9b7646ae8..d1e5f60af580 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -152,6 +152,7 @@ class SpecAttr(SpecElement):
         value         numerical ID when serialized
         attr_set      Attribute Set containing this attr
         is_multi      bool, attr may repeat multiple times
+        struct_name   string, name of struct definition
         sub_type      string, name of sub type
     """
     def __init__(self, family, attr_set, yaml, value):
@@ -160,6 +161,7 @@ class SpecAttr(SpecElement):
         self.value = value
         self.attr_set = attr_set
         self.is_multi = yaml.get('multi-attr', False)
+        self.struct_name = yaml.get('struct')
         self.sub_type = yaml.get('sub-type')
 
 
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index eada229402fa..63af3bd9787d 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -102,6 +102,17 @@ class NlAttr:
         format, _ = self.type_formats[type]
         return list({ x[0] for x in struct.iter_unpack(format, self.raw) })
 
+    def as_struct(self, members):
+        value = dict()
+        offset = 0
+        for m in members:
+            # TODO: handle non-scalar members
+            format, size = self.type_formats[m.type]
+            decoded = struct.unpack_from(format, self.raw, offset)
+            offset += size
+            value[m.name] = decoded[0]
+        return value
+
     def __repr__(self):
         return f"[type:{self.type} len:{self._len}] {self.raw}"
 
@@ -377,7 +388,9 @@ class YnlFamily(SpecFamily):
         rsp[attr_spec['name']] = value
 
     def _decode_binary(self, attr, attr_spec):
-        if attr_spec.sub_type:
+        if attr_spec.struct_name:
+            decoded = attr.as_struct(self.consts[attr_spec.struct_name])
+        elif attr_spec.sub_type:
             decoded = attr.as_c_array(attr_spec.sub_type)
         else:
             decoded = attr.as_bin()
-- 
cgit v1.2.3-70-g09d2


From f036d936ca57e8bc1f39b92cadfbac27095dc4e7 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:35 +0100
Subject: tools: ynl: Add fixed-header support to ynl

Add support for netlink families that add an optional fixed header structure
after the genetlink header and before any attributes. The fixed-header can be
specified on a per op basis, or once for all operations, which serves as a
default value that can be overridden.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/genetlink-legacy.yaml | 11 +++++++++++
 tools/net/ynl/lib/nlspec.py                 | 21 +++++++++++++--------
 tools/net/ynl/lib/ynl.py                    | 24 ++++++++++++++++++++----
 3 files changed, 44 insertions(+), 12 deletions(-)

diff --git a/Documentation/netlink/genetlink-legacy.yaml b/Documentation/netlink/genetlink-legacy.yaml
index d50c78b9f42d..b33541a51d6b 100644
--- a/Documentation/netlink/genetlink-legacy.yaml
+++ b/Documentation/netlink/genetlink-legacy.yaml
@@ -261,6 +261,14 @@ properties:
       async-enum:
         description: Name for the enum type with notifications/events.
         type: string
+      # Start genetlink-legacy
+      fixed-header: &fixed-header
+        description: |
+          Name of the structure defining the optional fixed-length protocol
+          header. This header is placed in a message after the netlink and
+          genetlink headers and before any attributes.
+        type: string
+      # End genetlink-legacy
       list:
         description: List of commands
         type: array
@@ -293,6 +301,9 @@ properties:
               type: array
               items:
                 enum: [ strict, dump ]
+            # Start genetlink-legacy
+            fixed-header: *fixed-header
+            # End genetlink-legacy
             do: &subop-type
               description: Main command handler.
               type: object
diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index d1e5f60af580..06a906d74f0e 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -263,16 +263,17 @@ class SpecOperation(SpecElement):
     Information about a single Netlink operation.
 
     Attributes:
-        value       numerical ID when serialized, None if req/rsp values differ
+        value           numerical ID when serialized, None if req/rsp values differ
 
-        req_value   numerical ID when serialized, user -> kernel
-        rsp_value   numerical ID when serialized, user <- kernel
-        is_call     bool, whether the operation is a call
-        is_async    bool, whether the operation is a notification
-        is_resv     bool, whether the operation does not exist (it's just a reserved ID)
-        attr_set    attribute set name
+        req_value       numerical ID when serialized, user -> kernel
+        rsp_value       numerical ID when serialized, user <- kernel
+        is_call         bool, whether the operation is a call
+        is_async        bool, whether the operation is a notification
+        is_resv         bool, whether the operation does not exist (it's just a reserved ID)
+        attr_set        attribute set name
+        fixed_header    string, optional name of fixed header struct
 
-        yaml        raw spec as loaded from the spec file
+        yaml            raw spec as loaded from the spec file
     """
     def __init__(self, family, yaml, req_value, rsp_value):
         super().__init__(family, yaml)
@@ -284,6 +285,7 @@ class SpecOperation(SpecElement):
         self.is_call = 'do' in yaml or 'dump' in yaml
         self.is_async = 'notify' in yaml or 'event' in yaml
         self.is_resv = not self.is_async and not self.is_call
+        self.fixed_header = self.yaml.get('fixed-header', family.fixed_header)
 
         # Added by resolve:
         self.attr_set = None
@@ -324,6 +326,7 @@ class SpecFamily(SpecElement):
         msgs_by_value  dict of all messages (indexed by name)
         ops        dict of all valid requests / responses
         consts     dict of all constants/enums
+        fixed_header  string, optional name of family default fixed header struct
     """
     def __init__(self, spec_path, schema_path=None):
         with open(spec_path, "r") as stream:
@@ -397,6 +400,7 @@ class SpecFamily(SpecElement):
         self._resolution_list.append(elem)
 
     def _dictify_ops_unified(self):
+        self.fixed_header = self.yaml['operations'].get('fixed-header')
         val = 1
         for elem in self.yaml['operations']['list']:
             if 'value' in elem:
@@ -408,6 +412,7 @@ class SpecFamily(SpecElement):
             self.msgs[op.name] = op
 
     def _dictify_ops_directional(self):
+        self.fixed_header = self.yaml['operations'].get('fixed-header')
         req_val = rsp_val = 1
         for elem in self.yaml['operations']['list']:
             if 'notify' in elem:
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 63af3bd9787d..ec40918152e1 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -278,14 +278,22 @@ def _genl_load_families():
 
 
 class GenlMsg:
-    def __init__(self, nl_msg):
+    def __init__(self, nl_msg, fixed_header_members=[]):
         self.nl = nl_msg
 
         self.hdr = nl_msg.raw[0:4]
-        self.raw = nl_msg.raw[4:]
+        offset = 4
 
         self.genl_cmd, self.genl_version, _ = struct.unpack("BBH", self.hdr)
 
+        self.fixed_header_attrs = dict()
+        for m in fixed_header_members:
+            format, size = NlAttr.type_formats[m.type]
+            decoded = struct.unpack_from(format, nl_msg.raw, offset)
+            offset += size
+            self.fixed_header_attrs[m.name] = decoded[0]
+
+        self.raw = nl_msg.raw[offset:]
         self.raw_attrs = NlAttrs(self.raw)
 
     def __repr__(self):
@@ -509,6 +517,13 @@ class YnlFamily(SpecFamily):
 
         req_seq = random.randint(1024, 65535)
         msg = _genl_msg(self.family.family_id, nl_flags, op.req_value, 1, req_seq)
+        fixed_header_members = []
+        if op.fixed_header:
+            fixed_header_members = self.consts[op.fixed_header].members
+            for m in fixed_header_members:
+                value = vals.pop(m.name)
+                format, _ = NlAttr.type_formats[m.type]
+                msg += struct.pack(format, value)
         for name, value in vals.items():
             msg += self._add_attr(op.attr_set.name, name, value)
         msg = _genl_msg_finalize(msg)
@@ -535,7 +550,7 @@ class YnlFamily(SpecFamily):
                     done = True
                     break
 
-                gm = GenlMsg(nl_msg)
+                gm = GenlMsg(nl_msg, fixed_header_members)
                 # Check if this is a reply to our request
                 if nl_msg.nl_seq != req_seq or gm.genl_cmd != op.rsp_value:
                     if gm.genl_cmd in self.async_msg_ids:
@@ -545,7 +560,8 @@ class YnlFamily(SpecFamily):
                         print('Unexpected message: ' + repr(gm))
                         continue
 
-                rsp.append(self._decode(gm.raw_attrs, op.attr_set.name))
+                rsp.append(self._decode(gm.raw_attrs, op.attr_set.name)
+                           | gm.fixed_header_attrs)
 
         if not rsp:
             return None
-- 
cgit v1.2.3-70-g09d2


From 643ef4a676e3a1ff1312e9fea6ae777c9a92fcd3 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:36 +0100
Subject: netlink: specs: add partial specification for openvswitch

The openvswitch family has a fixed header, uses struct attrs and has array
values. This partial spec demonstrates these features in the YNL CLI. These
specs are sufficient to create, delete and dump datapaths and to dump vports:

$ ./tools/net/ynl/cli.py \
    --spec Documentation/netlink/specs/ovs_datapath.yaml \
    --do dp-new --json '{ "dp-ifindex": 0, "name": "demo", "upcall-pid": 0}'
None

$ ./tools/net/ynl/cli.py \
    --spec Documentation/netlink/specs/ovs_datapath.yaml \
    --dump dp-get --json '{ "dp-ifindex": 0 }'
[{'dp-ifindex': 3,
  'masks-cache-size': 256,
  'megaflow-stats': {'cache-hits': 0,
                     'mask-hit': 0,
                     'masks': 0,
                     'pad1': 0,
                     'padding': 0},
  'name': 'test',
  'stats': {'flows': 0, 'hit': 0, 'lost': 0, 'missed': 0},
  'user-features': {'dispatch-upcall-per-cpu',
                    'tc-recirc-sharing',
                    'unaligned'}},
 {'dp-ifindex': 48,
  'masks-cache-size': 256,
  'megaflow-stats': {'cache-hits': 0,
                     'mask-hit': 0,
                     'masks': 0,
                     'pad1': 0,
                     'padding': 0},
  'name': 'demo',
  'stats': {'flows': 0, 'hit': 0, 'lost': 0, 'missed': 0},
  'user-features': set()}]

$ ./tools/net/ynl/cli.py \
    --spec Documentation/netlink/specs/ovs_datapath.yaml \
    --do dp-del --json '{ "dp-ifindex": 0, "name": "demo"}'
None

$ ./tools/net/ynl/cli.py \
    --spec Documentation/netlink/specs/ovs_vport.yaml \
    --dump vport-get --json '{ "dp-ifindex": 3 }'
[{'dp-ifindex': 3,
  'ifindex': 3,
  'name': 'test',
  'port-no': 0,
  'stats': {'rx-bytes': 0,
            'rx-dropped': 0,
            'rx-errors': 0,
            'rx-packets': 0,
            'tx-bytes': 0,
            'tx-dropped': 0,
            'tx-errors': 0,
            'tx-packets': 0},
  'type': 'internal',
  'upcall-pid': [0],
  'upcall-stats': {'fail': 0, 'success': 0}}]

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ovs_datapath.yaml | 153 ++++++++++++++++++++++++++
 Documentation/netlink/specs/ovs_vport.yaml    | 139 +++++++++++++++++++++++
 2 files changed, 292 insertions(+)
 create mode 100644 Documentation/netlink/specs/ovs_datapath.yaml
 create mode 100644 Documentation/netlink/specs/ovs_vport.yaml

diff --git a/Documentation/netlink/specs/ovs_datapath.yaml b/Documentation/netlink/specs/ovs_datapath.yaml
new file mode 100644
index 000000000000..6d71db8c4416
--- /dev/null
+++ b/Documentation/netlink/specs/ovs_datapath.yaml
@@ -0,0 +1,153 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: ovs_datapath
+version: 2
+protocol: genetlink-legacy
+
+doc:
+  OVS datapath configuration over generic netlink.
+
+definitions:
+  -
+    name: ovs-header
+    type: struct
+    members:
+      -
+        name: dp-ifindex
+        type: u32
+  -
+    name: user-features
+    type: flags
+    entries:
+      -
+        name: unaligned
+        doc: Allow last Netlink attribute to be unaligned
+      -
+        name: vport-pids
+        doc: Allow datapath to associate multiple Netlink PIDs to each vport
+      -
+        name: tc-recirc-sharing
+        doc: Allow tc offload recirc sharing
+      -
+        name: dispatch-upcall-per-cpu
+        doc: Allow per-cpu dispatch of upcalls
+  -
+    name: datapath-stats
+    type: struct
+    members:
+      -
+        name: hit
+        type: u64
+      -
+        name: missed
+        type: u64
+      -
+        name: lost
+        type: u64
+      -
+        name: flows
+        type: u64
+  -
+    name: megaflow-stats
+    type: struct
+    members:
+      -
+        name: mask-hit
+        type: u64
+      -
+        name: masks
+        type: u32
+      -
+        name: padding
+        type: u32
+      -
+        name: cache-hits
+        type: u64
+      -
+        name: pad1
+        type: u64
+
+attribute-sets:
+  -
+    name: datapath
+    attributes:
+      -
+        name: name
+        type: string
+      -
+        name: upcall-pid
+        doc: upcall pid
+        type: u32
+      -
+        name: stats
+        type: binary
+        struct: datapath-stats
+      -
+        name: megaflow-stats
+        type: binary
+        struct: megaflow-stats
+      -
+        name: user-features
+        type: u32
+        enum: user-features
+        enum-as-flags: true
+      -
+        name: pad
+        type: unused
+      -
+        name: masks-cache-size
+        type: u32
+      -
+        name: per-cpu-pids
+        type: binary
+        sub-type: u32
+
+operations:
+  fixed-header: ovs-header
+  list:
+    -
+      name: dp-get
+      doc: Get / dump OVS data path configuration and state
+      value: 3
+      attribute-set: datapath
+      do: &dp-get-op
+        request:
+          attributes:
+            - name
+        reply:
+          attributes:
+            - name
+            - upcall-pid
+            - stats
+            - megaflow-stats
+            - user-features
+            - masks-cache-size
+            - per-cpu-pids
+      dump: *dp-get-op
+    -
+      name: dp-new
+      doc: Create new OVS data path
+      value: 1
+      attribute-set: datapath
+      do:
+        request:
+          attributes:
+            - dp-ifindex
+            - name
+            - upcall-pid
+            - user-features
+    -
+      name: dp-del
+      doc: Delete existing OVS data path
+      value: 2
+      attribute-set: datapath
+      do:
+        request:
+          attributes:
+            - dp-ifindex
+            - name
+
+mcast-groups:
+  list:
+    -
+      name: ovs_datapath
diff --git a/Documentation/netlink/specs/ovs_vport.yaml b/Documentation/netlink/specs/ovs_vport.yaml
new file mode 100644
index 000000000000..8e55622ddf11
--- /dev/null
+++ b/Documentation/netlink/specs/ovs_vport.yaml
@@ -0,0 +1,139 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: ovs_vport
+version: 2
+protocol: genetlink-legacy
+
+doc:
+  OVS vport configuration over generic netlink.
+
+definitions:
+  -
+    name: ovs-header
+    type: struct
+    members:
+      -
+        name: dp-ifindex
+        type: u32
+  -
+    name: vport-type
+    type: enum
+    entries: [ unspec, netdev, internal, gre, vxlan, geneve ]
+  -
+    name: vport-stats
+    type: struct
+    members:
+      -
+        name: rx-packets
+        type: u64
+      -
+        name: tx-packets
+        type: u64
+      -
+        name: rx-bytes
+        type: u64
+      -
+        name: tx-bytes
+        type: u64
+      -
+        name: rx-errors
+        type: u64
+      -
+        name: tx-errors
+        type: u64
+      -
+        name: rx-dropped
+        type: u64
+      -
+        name: tx-dropped
+        type: u64
+
+attribute-sets:
+  -
+    name: vport-options
+    attributes:
+      -
+        name: dst-port
+        type: u32
+      -
+        name: extension
+        type: u32
+  -
+    name: upcall-stats
+    attributes:
+      -
+        name: success
+        type: u64
+        value: 0
+      -
+        name: fail
+        type: u64
+  -
+    name: vport
+    attributes:
+      -
+        name: port-no
+        type: u32
+      -
+        name: type
+        type: u32
+        enum: vport-type
+      -
+        name: name
+        type: string
+      -
+        name: options
+        type: nest
+        nested-attributes: vport-options
+      -
+        name: upcall-pid
+        type: binary
+        sub-type: u32
+      -
+        name: stats
+        type: binary
+        struct: vport-stats
+      -
+        name: pad
+        type: unused
+      -
+        name: ifindex
+        type: u32
+      -
+        name: netnsid
+        type: u32
+      -
+        name: upcall-stats
+        type: nest
+        nested-attributes: upcall-stats
+
+operations:
+  list:
+    -
+      name: vport-get
+      doc: Get / dump OVS vport configuration and state
+      value: 3
+      attribute-set: vport
+      fixed-header: ovs-header
+      do: &vport-get-op
+        request:
+          attributes:
+            - dp-ifindex
+            - name
+        reply: &dev-all
+          attributes:
+            - dp-ifindex
+            - port-no
+            - type
+            - name
+            - upcall-pid
+            - stats
+            - ifindex
+            - netnsid
+            - upcall-stats
+      dump: *vport-get-op
+
+mcast-groups:
+  list:
+    -
+      name: ovs_vport
-- 
cgit v1.2.3-70-g09d2


From 88e288968412ec1ca3d3b2d96956baa543fdfe82 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:37 +0100
Subject: docs: netlink: document struct support for genetlink-legacy

Describe the genetlink-legacy support for using struct definitions
for fixed headers and for binary attributes.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../userspace-api/netlink/genetlink-legacy.rst     | 74 +++++++++++++++++++++-
 1 file changed, 71 insertions(+), 3 deletions(-)

diff --git a/Documentation/userspace-api/netlink/genetlink-legacy.rst b/Documentation/userspace-api/netlink/genetlink-legacy.rst
index 3bf0bcdf21d8..b8fdcf7f6615 100644
--- a/Documentation/userspace-api/netlink/genetlink-legacy.rst
+++ b/Documentation/userspace-api/netlink/genetlink-legacy.rst
@@ -162,9 +162,77 @@ Other quirks (todo)
 Structures
 ----------
 
-Legacy families can define C structures both to be used as the contents
-of an attribute and as a fixed message header. The plan is to define
-the structs in ``definitions`` and link the appropriate attrs.
+Legacy families can define C structures both to be used as the contents of
+an attribute and as a fixed message header. Structures are defined in
+``definitions``  and referenced in operations or attributes. Note that
+structures defined in YAML are implicitly packed according to C
+conventions. For example, the following struct is 4 bytes, not 6 bytes:
+
+.. code-block:: c
+
+  struct {
+          u8 a;
+          u16 b;
+          u8 c;
+  }
+
+Any padding must be explicitly added and C-like languages should infer the
+need for explicit padding from whether the members are naturally aligned.
+
+Here is the struct definition from above, declared in YAML:
+
+.. code-block:: yaml
+
+  definitions:
+    -
+      name: message-header
+      type: struct
+      members:
+        -
+          name: a
+          type: u8
+        -
+          name: b
+          type: u16
+        -
+          name: c
+          type: u8
+
+Fixed Headers
+~~~~~~~~~~~~~
+
+Fixed message headers can be added to operations using ``fixed-header``.
+The default ``fixed-header`` can be set in ``operations`` and it can be set
+or overridden for each operation.
+
+.. code-block:: yaml
+
+  operations:
+    fixed-header: message-header
+    list:
+      -
+        name: get
+        fixed-header: custom-header
+        attribute-set: message-attrs
+
+Attributes
+~~~~~~~~~~
+
+A ``binary`` attribute can be interpreted as a C structure using a
+``struct`` property with the name of the structure definition. The
+``struct`` property implies ``sub-type: struct`` so it is not necessary to
+specify a sub-type.
+
+.. code-block:: yaml
+
+  attribute-sets:
+    -
+      name: stats-attrs
+      attributes:
+        -
+          name: stats
+          type: binary
+          struct: vport-stats
 
 Multi-message DO
 ----------------
-- 
cgit v1.2.3-70-g09d2


From 04eac39361d3506c82ccf932cb288e84d5746969 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Mon, 27 Mar 2023 09:31:38 +0100
Subject: docs: netlink: document the sub-type attribute property

Add a definition for sub-type to the protocol spec doc and a description of
its usage for C arrays in genetlink-legacy.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/userspace-api/netlink/genetlink-legacy.rst | 14 ++++++++++++++
 Documentation/userspace-api/netlink/specs.rst            | 10 ++++++++++
 2 files changed, 24 insertions(+)

diff --git a/Documentation/userspace-api/netlink/genetlink-legacy.rst b/Documentation/userspace-api/netlink/genetlink-legacy.rst
index b8fdcf7f6615..802875a37a27 100644
--- a/Documentation/userspace-api/netlink/genetlink-legacy.rst
+++ b/Documentation/userspace-api/netlink/genetlink-legacy.rst
@@ -234,6 +234,20 @@ specify a sub-type.
           type: binary
           struct: vport-stats
 
+C Arrays
+--------
+
+Legacy families also use ``binary`` attributes to encapsulate C arrays. The
+``sub-type`` is used to identify the type of scalar to extract.
+
+.. code-block:: yaml
+
+  attributes:
+    -
+      name: ports
+      type: binary
+      sub-type: u32
+
 Multi-message DO
 ----------------
 
diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst
index a22442ba1d30..2e4acde890b7 100644
--- a/Documentation/userspace-api/netlink/specs.rst
+++ b/Documentation/userspace-api/netlink/specs.rst
@@ -254,6 +254,16 @@ rather than depend on what is specified in the spec file.
 The validation policy in the kernel is formed by combining the type
 definition (``type`` and ``nested-attributes``) and the ``checks``.
 
+sub-type
+~~~~~~~~
+
+Legacy families have special ways of expressing arrays. ``sub-type`` can be
+used to define the type of array members in case array members are not
+fully defined as attributes (in a bona fide attribute space). For instance
+a C array of u32 values can be specified with ``type: binary`` and
+``sub-type: u32``. Binary types and legacy array formats are described in
+more detail in :doc:`genetlink-legacy`.
+
 operations
 ----------
 
-- 
cgit v1.2.3-70-g09d2


From 24265c2c91ad6aae9446e18472566cd83e92b602 Mon Sep 17 00:00:00 2001
From: Bobby Eshleman <bobby.eshleman@bytedance.com>
Date: Mon, 27 Mar 2023 22:16:06 +0000
Subject: testing/vsock: add vsock_perf to gitignore

This adds the vsock_perf binary to the gitignore file.

Fixes: 8abbffd27ced ("test/vsock: vsock_perf utility")
Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com>
Reviewed-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/r/20230327-vsock-add-vsock-perf-to-ignore-v1-1-f28a84f3606b@bytedance.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/vsock/.gitignore | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/vsock/.gitignore b/tools/testing/vsock/.gitignore
index 87ca2731cff9..a8adcfdc292b 100644
--- a/tools/testing/vsock/.gitignore
+++ b/tools/testing/vsock/.gitignore
@@ -2,3 +2,4 @@
 *.d
 vsock_test
 vsock_diag_test
+vsock_perf
-- 
cgit v1.2.3-70-g09d2


From 634f1a7110b439c65fd8a809171c1d2d28bcea6f Mon Sep 17 00:00:00 2001
From: Bobby Eshleman <bobby.eshleman@bytedance.com>
Date: Mon, 27 Mar 2023 19:11:51 +0000
Subject: vsock: support sockmap

This patch adds sockmap support for vsock sockets. It is intended to be
usable by all transports, but only the virtio and loopback transports
are implemented.

SOCK_STREAM, SOCK_DGRAM, and SOCK_SEQPACKET are all supported.

Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/vhost/vsock.c                   |   1 +
 include/linux/virtio_vsock.h            |   1 +
 include/net/af_vsock.h                  |  17 ++++
 net/vmw_vsock/Makefile                  |   1 +
 net/vmw_vsock/af_vsock.c                |  64 ++++++++++--
 net/vmw_vsock/virtio_transport.c        |   2 +
 net/vmw_vsock/virtio_transport_common.c |  25 +++++
 net/vmw_vsock/vsock_bpf.c               | 174 ++++++++++++++++++++++++++++++++
 net/vmw_vsock/vsock_loopback.c          |   2 +
 9 files changed, 281 insertions(+), 6 deletions(-)
 create mode 100644 net/vmw_vsock/vsock_bpf.c

diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index c8e6087769a1..6578db78f0ae 100644
--- a/drivers/vhost/vsock.c
+++ b/drivers/vhost/vsock.c
@@ -439,6 +439,7 @@ static struct virtio_transport vhost_transport = {
 		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
 		.notify_buffer_size       = virtio_transport_notify_buffer_size,
 
+		.read_skb = virtio_transport_read_skb,
 	},
 
 	.send_pkt = vhost_transport_send_pkt,
diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
index 3f9c16611306..c58453699ee9 100644
--- a/include/linux/virtio_vsock.h
+++ b/include/linux/virtio_vsock.h
@@ -245,4 +245,5 @@ u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 wanted);
 void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit);
 void virtio_transport_deliver_tap_pkt(struct sk_buff *skb);
 int virtio_transport_purge_skbs(void *vsk, struct sk_buff_head *list);
+int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t read_actor);
 #endif /* _LINUX_VIRTIO_VSOCK_H */
diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index 568a87c5e0d0..0e7504a42925 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -75,6 +75,7 @@ struct vsock_sock {
 	void *trans;
 };
 
+s64 vsock_connectible_has_data(struct vsock_sock *vsk);
 s64 vsock_stream_has_data(struct vsock_sock *vsk);
 s64 vsock_stream_has_space(struct vsock_sock *vsk);
 struct sock *vsock_create_connected(struct sock *parent);
@@ -173,6 +174,9 @@ struct vsock_transport {
 
 	/* Addressing. */
 	u32 (*get_local_cid)(void);
+
+	/* Read a single skb */
+	int (*read_skb)(struct vsock_sock *, skb_read_actor_t);
 };
 
 /**** CORE ****/
@@ -225,5 +229,18 @@ int vsock_init_tap(void);
 int vsock_add_tap(struct vsock_tap *vt);
 int vsock_remove_tap(struct vsock_tap *vt);
 void vsock_deliver_tap(struct sk_buff *build_skb(void *opaque), void *opaque);
+int vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
+			      int flags);
+int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
+			size_t len, int flags);
+
+#ifdef CONFIG_BPF_SYSCALL
+extern struct proto vsock_proto;
+int vsock_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore);
+void __init vsock_bpf_build_proto(void);
+#else
+static inline void __init vsock_bpf_build_proto(void)
+{}
+#endif
 
 #endif /* __AF_VSOCK_H__ */
diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile
index 6a943ec95c4a..5da74c4a9f1d 100644
--- a/net/vmw_vsock/Makefile
+++ b/net/vmw_vsock/Makefile
@@ -8,6 +8,7 @@ obj-$(CONFIG_HYPERV_VSOCKETS) += hv_sock.o
 obj-$(CONFIG_VSOCKETS_LOOPBACK) += vsock_loopback.o
 
 vsock-y += af_vsock.o af_vsock_tap.o vsock_addr.o
+vsock-$(CONFIG_BPF_SYSCALL) += vsock_bpf.o
 
 vsock_diag-y += diag.o
 
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 19aea7cba26e..5f2dda35c980 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -116,10 +116,13 @@ static void vsock_sk_destruct(struct sock *sk);
 static int vsock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
 
 /* Protocol family. */
-static struct proto vsock_proto = {
+struct proto vsock_proto = {
 	.name = "AF_VSOCK",
 	.owner = THIS_MODULE,
 	.obj_size = sizeof(struct vsock_sock),
+#ifdef CONFIG_BPF_SYSCALL
+	.psock_update_sk_prot = vsock_bpf_update_proto,
+#endif
 };
 
 /* The default peer timeout indicates how long we will wait for a peer response
@@ -865,7 +868,7 @@ s64 vsock_stream_has_data(struct vsock_sock *vsk)
 }
 EXPORT_SYMBOL_GPL(vsock_stream_has_data);
 
-static s64 vsock_connectible_has_data(struct vsock_sock *vsk)
+s64 vsock_connectible_has_data(struct vsock_sock *vsk)
 {
 	struct sock *sk = sk_vsock(vsk);
 
@@ -874,6 +877,7 @@ static s64 vsock_connectible_has_data(struct vsock_sock *vsk)
 	else
 		return vsock_stream_has_data(vsk);
 }
+EXPORT_SYMBOL_GPL(vsock_connectible_has_data);
 
 s64 vsock_stream_has_space(struct vsock_sock *vsk)
 {
@@ -1131,6 +1135,13 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock,
 	return mask;
 }
 
+static int vsock_read_skb(struct sock *sk, skb_read_actor_t read_actor)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+
+	return vsk->transport->read_skb(vsk, read_actor);
+}
+
 static int vsock_dgram_sendmsg(struct socket *sock, struct msghdr *msg,
 			       size_t len)
 {
@@ -1242,18 +1253,42 @@ static int vsock_dgram_connect(struct socket *sock,
 	memcpy(&vsk->remote_addr, remote_addr, sizeof(vsk->remote_addr));
 	sock->state = SS_CONNECTED;
 
+	/* sock map disallows redirection of non-TCP sockets with sk_state !=
+	 * TCP_ESTABLISHED (see sock_map_redirect_allowed()), so we set
+	 * TCP_ESTABLISHED here to allow redirection of connected vsock dgrams.
+	 *
+	 * This doesn't seem to be abnormal state for datagram sockets, as the
+	 * same approach can be see in other datagram socket types as well
+	 * (such as unix sockets).
+	 */
+	sk->sk_state = TCP_ESTABLISHED;
+
 out:
 	release_sock(sk);
 	return err;
 }
 
-static int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
-			       size_t len, int flags)
+int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
+			size_t len, int flags)
 {
-	struct vsock_sock *vsk = vsock_sk(sock->sk);
+#ifdef CONFIG_BPF_SYSCALL
+	const struct proto *prot;
+#endif
+	struct vsock_sock *vsk;
+	struct sock *sk;
+
+	sk = sock->sk;
+	vsk = vsock_sk(sk);
+
+#ifdef CONFIG_BPF_SYSCALL
+	prot = READ_ONCE(sk->sk_prot);
+	if (prot != &vsock_proto)
+		return prot->recvmsg(sk, msg, len, flags, NULL);
+#endif
 
 	return vsk->transport->dgram_dequeue(vsk, msg, len, flags);
 }
+EXPORT_SYMBOL_GPL(vsock_dgram_recvmsg);
 
 static const struct proto_ops vsock_dgram_ops = {
 	.family = PF_VSOCK,
@@ -1272,6 +1307,7 @@ static const struct proto_ops vsock_dgram_ops = {
 	.recvmsg = vsock_dgram_recvmsg,
 	.mmap = sock_no_mmap,
 	.sendpage = sock_no_sendpage,
+	.read_skb = vsock_read_skb,
 };
 
 static int vsock_transport_cancel_pkt(struct vsock_sock *vsk)
@@ -2086,13 +2122,16 @@ out:
 	return err;
 }
 
-static int
+int
 vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 			  int flags)
 {
 	struct sock *sk;
 	struct vsock_sock *vsk;
 	const struct vsock_transport *transport;
+#ifdef CONFIG_BPF_SYSCALL
+	const struct proto *prot;
+#endif
 	int err;
 
 	sk = sock->sk;
@@ -2139,6 +2178,14 @@ vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 		goto out;
 	}
 
+#ifdef CONFIG_BPF_SYSCALL
+	prot = READ_ONCE(sk->sk_prot);
+	if (prot != &vsock_proto) {
+		release_sock(sk);
+		return prot->recvmsg(sk, msg, len, flags, NULL);
+	}
+#endif
+
 	if (sk->sk_type == SOCK_STREAM)
 		err = __vsock_stream_recvmsg(sk, msg, len, flags);
 	else
@@ -2148,6 +2195,7 @@ out:
 	release_sock(sk);
 	return err;
 }
+EXPORT_SYMBOL_GPL(vsock_connectible_recvmsg);
 
 static int vsock_set_rcvlowat(struct sock *sk, int val)
 {
@@ -2188,6 +2236,7 @@ static const struct proto_ops vsock_stream_ops = {
 	.mmap = sock_no_mmap,
 	.sendpage = sock_no_sendpage,
 	.set_rcvlowat = vsock_set_rcvlowat,
+	.read_skb = vsock_read_skb,
 };
 
 static const struct proto_ops vsock_seqpacket_ops = {
@@ -2209,6 +2258,7 @@ static const struct proto_ops vsock_seqpacket_ops = {
 	.recvmsg = vsock_connectible_recvmsg,
 	.mmap = sock_no_mmap,
 	.sendpage = sock_no_sendpage,
+	.read_skb = vsock_read_skb,
 };
 
 static int vsock_create(struct net *net, struct socket *sock,
@@ -2348,6 +2398,8 @@ static int __init vsock_init(void)
 		goto err_unregister_proto;
 	}
 
+	vsock_bpf_build_proto();
+
 	return 0;
 
 err_unregister_proto:
diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index 28b5a8e8e094..e95df847176b 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -457,6 +457,8 @@ static struct virtio_transport virtio_transport = {
 		.notify_send_pre_enqueue  = virtio_transport_notify_send_pre_enqueue,
 		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
 		.notify_buffer_size       = virtio_transport_notify_buffer_size,
+
+		.read_skb = virtio_transport_read_skb,
 	},
 
 	.send_pkt = virtio_transport_send_pkt,
diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index 7fc178c3ee07..f39639dd6eb5 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -1418,6 +1418,31 @@ int virtio_transport_purge_skbs(void *vsk, struct sk_buff_head *queue)
 }
 EXPORT_SYMBOL_GPL(virtio_transport_purge_skbs);
 
+int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t recv_actor)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	struct sock *sk = sk_vsock(vsk);
+	struct sk_buff *skb;
+	int off = 0;
+	int copied;
+	int err;
+
+	spin_lock_bh(&vvs->rx_lock);
+	/* Use __skb_recv_datagram() for race-free handling of the receive. It
+	 * works for types other than dgrams.
+	 */
+	skb = __skb_recv_datagram(sk, &vvs->rx_queue, MSG_DONTWAIT, &off, &err);
+	spin_unlock_bh(&vvs->rx_lock);
+
+	if (!skb)
+		return err;
+
+	copied = recv_actor(sk, skb);
+	kfree_skb(skb);
+	return copied;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_read_skb);
+
 MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Asias He");
 MODULE_DESCRIPTION("common code for virtio vsock");
diff --git a/net/vmw_vsock/vsock_bpf.c b/net/vmw_vsock/vsock_bpf.c
new file mode 100644
index 000000000000..a3c97546ab84
--- /dev/null
+++ b/net/vmw_vsock/vsock_bpf.c
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2022 Bobby Eshleman <bobby.eshleman@bytedance.com>
+ *
+ * Based off of net/unix/unix_bpf.c
+ */
+
+#include <linux/bpf.h>
+#include <linux/module.h>
+#include <linux/skmsg.h>
+#include <linux/socket.h>
+#include <linux/wait.h>
+#include <net/af_vsock.h>
+#include <net/sock.h>
+
+#define vsock_sk_has_data(__sk, __psock)				\
+		({	!skb_queue_empty(&(__sk)->sk_receive_queue) ||	\
+			!skb_queue_empty(&(__psock)->ingress_skb) ||	\
+			!list_empty(&(__psock)->ingress_msg);		\
+		})
+
+static struct proto *vsock_prot_saved __read_mostly;
+static DEFINE_SPINLOCK(vsock_prot_lock);
+static struct proto vsock_bpf_prot;
+
+static bool vsock_has_data(struct sock *sk, struct sk_psock *psock)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+	s64 ret;
+
+	ret = vsock_connectible_has_data(vsk);
+	if (ret > 0)
+		return true;
+
+	return vsock_sk_has_data(sk, psock);
+}
+
+static bool vsock_msg_wait_data(struct sock *sk, struct sk_psock *psock, long timeo)
+{
+	bool ret;
+
+	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+
+	if (sk->sk_shutdown & RCV_SHUTDOWN)
+		return true;
+
+	if (!timeo)
+		return false;
+
+	add_wait_queue(sk_sleep(sk), &wait);
+	sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk);
+	ret = vsock_has_data(sk, psock);
+	if (!ret) {
+		wait_woken(&wait, TASK_INTERRUPTIBLE, timeo);
+		ret = vsock_has_data(sk, psock);
+	}
+	sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk);
+	remove_wait_queue(sk_sleep(sk), &wait);
+	return ret;
+}
+
+static int __vsock_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
+{
+	struct socket *sock = sk->sk_socket;
+	int err;
+
+	if (sk->sk_type == SOCK_STREAM || sk->sk_type == SOCK_SEQPACKET)
+		err = vsock_connectible_recvmsg(sock, msg, len, flags);
+	else if (sk->sk_type == SOCK_DGRAM)
+		err = vsock_dgram_recvmsg(sock, msg, len, flags);
+	else
+		err = -EPROTOTYPE;
+
+	return err;
+}
+
+static int vsock_bpf_recvmsg(struct sock *sk, struct msghdr *msg,
+			     size_t len, int flags, int *addr_len)
+{
+	struct sk_psock *psock;
+	int copied;
+
+	psock = sk_psock_get(sk);
+	if (unlikely(!psock))
+		return __vsock_recvmsg(sk, msg, len, flags);
+
+	lock_sock(sk);
+	if (vsock_has_data(sk, psock) && sk_psock_queue_empty(psock)) {
+		release_sock(sk);
+		sk_psock_put(sk, psock);
+		return __vsock_recvmsg(sk, msg, len, flags);
+	}
+
+	copied = sk_msg_recvmsg(sk, psock, msg, len, flags);
+	while (copied == 0) {
+		long timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+
+		if (!vsock_msg_wait_data(sk, psock, timeo)) {
+			copied = -EAGAIN;
+			break;
+		}
+
+		if (sk_psock_queue_empty(psock)) {
+			release_sock(sk);
+			sk_psock_put(sk, psock);
+			return __vsock_recvmsg(sk, msg, len, flags);
+		}
+
+		copied = sk_msg_recvmsg(sk, psock, msg, len, flags);
+	}
+
+	release_sock(sk);
+	sk_psock_put(sk, psock);
+
+	return copied;
+}
+
+/* Copy of original proto with updated sock_map methods */
+static struct proto vsock_bpf_prot = {
+	.close = sock_map_close,
+	.recvmsg = vsock_bpf_recvmsg,
+	.sock_is_readable = sk_msg_is_readable,
+	.unhash = sock_map_unhash,
+};
+
+static void vsock_bpf_rebuild_protos(struct proto *prot, const struct proto *base)
+{
+	*prot        = *base;
+	prot->close  = sock_map_close;
+	prot->recvmsg = vsock_bpf_recvmsg;
+	prot->sock_is_readable = sk_msg_is_readable;
+}
+
+static void vsock_bpf_check_needs_rebuild(struct proto *ops)
+{
+	/* Paired with the smp_store_release() below. */
+	if (unlikely(ops != smp_load_acquire(&vsock_prot_saved))) {
+		spin_lock_bh(&vsock_prot_lock);
+		if (likely(ops != vsock_prot_saved)) {
+			vsock_bpf_rebuild_protos(&vsock_bpf_prot, ops);
+			/* Make sure proto function pointers are updated before publishing the
+			 * pointer to the struct.
+			 */
+			smp_store_release(&vsock_prot_saved, ops);
+		}
+		spin_unlock_bh(&vsock_prot_lock);
+	}
+}
+
+int vsock_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore)
+{
+	struct vsock_sock *vsk;
+
+	if (restore) {
+		sk->sk_write_space = psock->saved_write_space;
+		sock_replace_proto(sk, psock->sk_proto);
+		return 0;
+	}
+
+	vsk = vsock_sk(sk);
+	if (!vsk->transport)
+		return -ENODEV;
+
+	if (!vsk->transport->read_skb)
+		return -EOPNOTSUPP;
+
+	vsock_bpf_check_needs_rebuild(psock->sk_proto);
+	sock_replace_proto(sk, &vsock_bpf_prot);
+	return 0;
+}
+
+void __init vsock_bpf_build_proto(void)
+{
+	vsock_bpf_rebuild_protos(&vsock_bpf_prot, &vsock_proto);
+}
diff --git a/net/vmw_vsock/vsock_loopback.c b/net/vmw_vsock/vsock_loopback.c
index 671e03240fc5..40753b661c13 100644
--- a/net/vmw_vsock/vsock_loopback.c
+++ b/net/vmw_vsock/vsock_loopback.c
@@ -94,6 +94,8 @@ static struct virtio_transport loopback_transport = {
 		.notify_send_pre_enqueue  = virtio_transport_notify_send_pre_enqueue,
 		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
 		.notify_buffer_size       = virtio_transport_notify_buffer_size,
+
+		.read_skb = virtio_transport_read_skb,
 	},
 
 	.send_pkt = vsock_loopback_send_pkt,
-- 
cgit v1.2.3-70-g09d2


From c7c605c982d63b2140280d005a9df321d74f81dc Mon Sep 17 00:00:00 2001
From: Bobby Eshleman <bobby.eshleman@bytedance.com>
Date: Mon, 27 Mar 2023 19:11:52 +0000
Subject: selftests/bpf: add vsock to vmtest.sh

Add vsock loopback to the test kernel.

This allows sockmap for vsock to be tested.

Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/bpf/config.aarch64 | 2 ++
 tools/testing/selftests/bpf/config.s390x   | 3 +++
 tools/testing/selftests/bpf/config.x86_64  | 3 +++
 3 files changed, 8 insertions(+)

diff --git a/tools/testing/selftests/bpf/config.aarch64 b/tools/testing/selftests/bpf/config.aarch64
index 1f0437644186..253821494884 100644
--- a/tools/testing/selftests/bpf/config.aarch64
+++ b/tools/testing/selftests/bpf/config.aarch64
@@ -176,6 +176,8 @@ CONFIG_VIRTIO_MMIO_CMDLINE_DEVICES=y
 CONFIG_VIRTIO_MMIO=y
 CONFIG_VIRTIO_NET=y
 CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VSOCKETS_COMMON=y
 CONFIG_VLAN_8021Q=y
 CONFIG_VSOCKETS=y
+CONFIG_VSOCKETS_LOOPBACK=y
 CONFIG_XFRM_USER=y
diff --git a/tools/testing/selftests/bpf/config.s390x b/tools/testing/selftests/bpf/config.s390x
index d49f6170e7bd..2ba92167be35 100644
--- a/tools/testing/selftests/bpf/config.s390x
+++ b/tools/testing/selftests/bpf/config.s390x
@@ -140,5 +140,8 @@ CONFIG_VIRTIO_BALLOON=y
 CONFIG_VIRTIO_BLK=y
 CONFIG_VIRTIO_NET=y
 CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VSOCKETS_COMMON=y
 CONFIG_VLAN_8021Q=y
+CONFIG_VSOCKETS=y
+CONFIG_VSOCKETS_LOOPBACK=y
 CONFIG_XFRM_USER=y
diff --git a/tools/testing/selftests/bpf/config.x86_64 b/tools/testing/selftests/bpf/config.x86_64
index dd97d61d325c..b650b2e617b8 100644
--- a/tools/testing/selftests/bpf/config.x86_64
+++ b/tools/testing/selftests/bpf/config.x86_64
@@ -234,7 +234,10 @@ CONFIG_VIRTIO_BLK=y
 CONFIG_VIRTIO_CONSOLE=y
 CONFIG_VIRTIO_NET=y
 CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VSOCKETS_COMMON=y
 CONFIG_VLAN_8021Q=y
+CONFIG_VSOCKETS=y
+CONFIG_VSOCKETS_LOOPBACK=y
 CONFIG_X86_ACPI_CPUFREQ=y
 CONFIG_X86_CPUID=y
 CONFIG_X86_MSR=y
-- 
cgit v1.2.3-70-g09d2


From d61bd8c1fd02cfc8aed00a58c20bd58c72549e3a Mon Sep 17 00:00:00 2001
From: Bobby Eshleman <bobby.eshleman@bytedance.com>
Date: Mon, 27 Mar 2023 19:11:53 +0000
Subject: selftests/bpf: add a test case for vsock sockmap

Add a test case testing the redirection from connectible AF_VSOCK
sockets to connectible AF_UNIX sockets.

Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/bpf/prog_tests/sockmap_listen.c      | 163 +++++++++++++++++++++
 1 file changed, 163 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c b/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
index 567e07c19ecc..8f09e1ea3ba7 100644
--- a/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
+++ b/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
@@ -18,6 +18,7 @@
 #include <string.h>
 #include <sys/select.h>
 #include <unistd.h>
+#include <linux/vm_sockets.h>
 
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
@@ -251,6 +252,16 @@ static void init_addr_loopback6(struct sockaddr_storage *ss, socklen_t *len)
 	*len = sizeof(*addr6);
 }
 
+static void init_addr_loopback_vsock(struct sockaddr_storage *ss, socklen_t *len)
+{
+	struct sockaddr_vm *addr = memset(ss, 0, sizeof(*ss));
+
+	addr->svm_family = AF_VSOCK;
+	addr->svm_port = VMADDR_PORT_ANY;
+	addr->svm_cid = VMADDR_CID_LOCAL;
+	*len = sizeof(*addr);
+}
+
 static void init_addr_loopback(int family, struct sockaddr_storage *ss,
 			       socklen_t *len)
 {
@@ -261,6 +272,9 @@ static void init_addr_loopback(int family, struct sockaddr_storage *ss,
 	case AF_INET6:
 		init_addr_loopback6(ss, len);
 		return;
+	case AF_VSOCK:
+		init_addr_loopback_vsock(ss, len);
+		return;
 	default:
 		FAIL("unsupported address family %d", family);
 	}
@@ -1478,6 +1492,8 @@ static const char *family_str(sa_family_t family)
 		return "IPv6";
 	case AF_UNIX:
 		return "Unix";
+	case AF_VSOCK:
+		return "VSOCK";
 	default:
 		return "unknown";
 	}
@@ -1689,6 +1705,151 @@ static void test_unix_redir(struct test_sockmap_listen *skel, struct bpf_map *ma
 	unix_skb_redir_to_connected(skel, map, sotype);
 }
 
+/* Returns two connected loopback vsock sockets */
+static int vsock_socketpair_connectible(int sotype, int *v0, int *v1)
+{
+	struct sockaddr_storage addr;
+	socklen_t len = sizeof(addr);
+	int s, p, c;
+
+	s = socket_loopback(AF_VSOCK, sotype);
+	if (s < 0)
+		return -1;
+
+	c = xsocket(AF_VSOCK, sotype | SOCK_NONBLOCK, 0);
+	if (c == -1)
+		goto close_srv;
+
+	if (getsockname(s, sockaddr(&addr), &len) < 0)
+		goto close_cli;
+
+	if (connect(c, sockaddr(&addr), len) < 0 && errno != EINPROGRESS) {
+		FAIL_ERRNO("connect");
+		goto close_cli;
+	}
+
+	len = sizeof(addr);
+	p = accept_timeout(s, sockaddr(&addr), &len, IO_TIMEOUT_SEC);
+	if (p < 0)
+		goto close_cli;
+
+	*v0 = p;
+	*v1 = c;
+
+	return 0;
+
+close_cli:
+	close(c);
+close_srv:
+	close(s);
+
+	return -1;
+}
+
+static void vsock_unix_redir_connectible(int sock_mapfd, int verd_mapfd,
+					 enum redir_mode mode, int sotype)
+{
+	const char *log_prefix = redir_mode_str(mode);
+	char a = 'a', b = 'b';
+	int u0, u1, v0, v1;
+	int sfd[2];
+	unsigned int pass;
+	int err, n;
+	u32 key;
+
+	zero_verdict_count(verd_mapfd);
+
+	if (socketpair(AF_UNIX, SOCK_STREAM | SOCK_NONBLOCK, 0, sfd))
+		return;
+
+	u0 = sfd[0];
+	u1 = sfd[1];
+
+	err = vsock_socketpair_connectible(sotype, &v0, &v1);
+	if (err) {
+		FAIL("vsock_socketpair_connectible() failed");
+		goto close_uds;
+	}
+
+	err = add_to_sockmap(sock_mapfd, u0, v0);
+	if (err) {
+		FAIL("add_to_sockmap failed");
+		goto close_vsock;
+	}
+
+	n = write(v1, &a, sizeof(a));
+	if (n < 0)
+		FAIL_ERRNO("%s: write", log_prefix);
+	if (n == 0)
+		FAIL("%s: incomplete write", log_prefix);
+	if (n < 1)
+		goto out;
+
+	n = recv(mode == REDIR_INGRESS ? u0 : u1, &b, sizeof(b), MSG_DONTWAIT);
+	if (n < 0)
+		FAIL("%s: recv() err, errno=%d", log_prefix, errno);
+	if (n == 0)
+		FAIL("%s: incomplete recv", log_prefix);
+	if (b != a)
+		FAIL("%s: vsock socket map failed, %c != %c", log_prefix, a, b);
+
+	key = SK_PASS;
+	err = xbpf_map_lookup_elem(verd_mapfd, &key, &pass);
+	if (err)
+		goto out;
+	if (pass != 1)
+		FAIL("%s: want pass count 1, have %d", log_prefix, pass);
+out:
+	key = 0;
+	bpf_map_delete_elem(sock_mapfd, &key);
+	key = 1;
+	bpf_map_delete_elem(sock_mapfd, &key);
+
+close_vsock:
+	close(v0);
+	close(v1);
+
+close_uds:
+	close(u0);
+	close(u1);
+}
+
+static void vsock_unix_skb_redir_connectible(struct test_sockmap_listen *skel,
+					     struct bpf_map *inner_map,
+					     int sotype)
+{
+	int verdict = bpf_program__fd(skel->progs.prog_skb_verdict);
+	int verdict_map = bpf_map__fd(skel->maps.verdict_map);
+	int sock_map = bpf_map__fd(inner_map);
+	int err;
+
+	err = xbpf_prog_attach(verdict, sock_map, BPF_SK_SKB_VERDICT, 0);
+	if (err)
+		return;
+
+	skel->bss->test_ingress = false;
+	vsock_unix_redir_connectible(sock_map, verdict_map, REDIR_EGRESS, sotype);
+	skel->bss->test_ingress = true;
+	vsock_unix_redir_connectible(sock_map, verdict_map, REDIR_INGRESS, sotype);
+
+	xbpf_prog_detach2(verdict, sock_map, BPF_SK_SKB_VERDICT);
+}
+
+static void test_vsock_redir(struct test_sockmap_listen *skel, struct bpf_map *map)
+{
+	const char *family_name, *map_name;
+	char s[MAX_TEST_NAME];
+
+	family_name = family_str(AF_VSOCK);
+	map_name = map_type_str(map);
+	snprintf(s, sizeof(s), "%s %s %s", map_name, family_name, __func__);
+	if (!test__start_subtest(s))
+		return;
+
+	vsock_unix_skb_redir_connectible(skel, map, SOCK_STREAM);
+	vsock_unix_skb_redir_connectible(skel, map, SOCK_SEQPACKET);
+}
+
 static void test_reuseport(struct test_sockmap_listen *skel,
 			   struct bpf_map *map, int family, int sotype)
 {
@@ -2060,12 +2221,14 @@ void serial_test_sockmap_listen(void)
 	run_tests(skel, skel->maps.sock_map, AF_INET6);
 	test_unix_redir(skel, skel->maps.sock_map, SOCK_DGRAM);
 	test_unix_redir(skel, skel->maps.sock_map, SOCK_STREAM);
+	test_vsock_redir(skel, skel->maps.sock_map);
 
 	skel->bss->test_sockmap = false;
 	run_tests(skel, skel->maps.sock_hash, AF_INET);
 	run_tests(skel, skel->maps.sock_hash, AF_INET6);
 	test_unix_redir(skel, skel->maps.sock_hash, SOCK_DGRAM);
 	test_unix_redir(skel, skel->maps.sock_hash, SOCK_STREAM);
+	test_vsock_redir(skel, skel->maps.sock_hash);
 
 	test_sockmap_listen__destroy(skel);
 }
-- 
cgit v1.2.3-70-g09d2


From 8cdc3223e78c43e1b60ea1c536a103e32fdca3c5 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Mon, 27 Mar 2023 16:54:54 -0700
Subject: ipv6: Remove in6addr_any alternatives.

Some code defines the IPv6 wildcard address as a local variable and
use it with memcmp() or ipv6_addr_equal().

Let's use in6addr_any and ipv6_addr_any() instead.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c |  5 ++---
 include/net/ip6_fib.h                                     |  9 +++------
 include/trace/events/fib.h                                |  5 ++---
 include/trace/events/fib6.h                               |  5 +----
 net/ethtool/ioctl.c                                       | 10 +++++-----
 net/ipv4/inet_hashtables.c                                | 11 ++++-------
 6 files changed, 17 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
index a108e73c9f66..20c2d2ecaf93 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
@@ -98,7 +98,6 @@ int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow,
 #if IS_ENABLED(CONFIG_INET) && IS_ENABLED(CONFIG_IPV6)
 	else if (ip_version == 6) {
 		int ipv6_size = MLX5_FLD_SZ_BYTES(ipv6_layout, ipv6);
-		struct in6_addr zerov6 = {};
 
 		daddr = MLX5_ADDR_OF(fte_match_param, spec->match_value,
 				     outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6);
@@ -106,8 +105,8 @@ int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow,
 				     outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6);
 		memcpy(&tun_attr->dst_ip.v6, daddr, ipv6_size);
 		memcpy(&tun_attr->src_ip.v6, saddr, ipv6_size);
-		if (!memcmp(&tun_attr->dst_ip.v6, &zerov6, sizeof(zerov6)) ||
-		    !memcmp(&tun_attr->src_ip.v6, &zerov6, sizeof(zerov6)))
+		if (ipv6_addr_any(&tun_attr->dst_ip.v6) ||
+		    ipv6_addr_any(&tun_attr->src_ip.v6))
 			return 0;
 	}
 #endif
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 79570cb4ea9c..05e6f756feaf 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -469,13 +469,10 @@ void rt6_get_prefsrc(const struct rt6_info *rt, struct in6_addr *addr)
 	rcu_read_lock();
 
 	from = rcu_dereference(rt->from);
-	if (from) {
+	if (from)
 		*addr = from->fib6_prefsrc.addr;
-	} else {
-		struct in6_addr in6_zero = {};
-
-		*addr = in6_zero;
-	}
+	else
+		*addr = in6addr_any;
 
 	rcu_read_unlock();
 }
diff --git a/include/trace/events/fib.h b/include/trace/events/fib.h
index c2300c407f58..76297ecd4935 100644
--- a/include/trace/events/fib.h
+++ b/include/trace/events/fib.h
@@ -36,7 +36,6 @@ TRACE_EVENT(fib_table_lookup,
 	),
 
 	TP_fast_assign(
-		struct in6_addr in6_zero = {};
 		struct net_device *dev;
 		struct in6_addr *in6;
 		__be32 *p32;
@@ -74,7 +73,7 @@ TRACE_EVENT(fib_table_lookup,
 				*p32 = nhc->nhc_gw.ipv4;
 
 				in6 = (struct in6_addr *)__entry->gw6;
-				*in6 = in6_zero;
+				*in6 = in6addr_any;
 			} else if (nhc->nhc_gw_family == AF_INET6) {
 				p32 = (__be32 *) __entry->gw4;
 				*p32 = 0;
@@ -87,7 +86,7 @@ TRACE_EVENT(fib_table_lookup,
 			*p32 = 0;
 
 			in6 = (struct in6_addr *)__entry->gw6;
-			*in6 = in6_zero;
+			*in6 = in6addr_any;
 		}
 	),
 
diff --git a/include/trace/events/fib6.h b/include/trace/events/fib6.h
index 6e821eb79450..4d3e607b3cde 100644
--- a/include/trace/events/fib6.h
+++ b/include/trace/events/fib6.h
@@ -68,11 +68,8 @@ TRACE_EVENT(fib6_table_lookup,
 			strcpy(__entry->name, "-");
 		}
 		if (res->f6i == net->ipv6.fib6_null_entry) {
-			struct in6_addr in6_zero = {};
-
 			in6 = (struct in6_addr *)__entry->gw;
-			*in6 = in6_zero;
-
+			*in6 = in6addr_any;
 		} else if (res->nh) {
 			in6 = (struct in6_addr *)__entry->gw;
 			*in6 = res->nh->fib_nh_gw6;
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 646b3e490c71..59adc4e6e9ee 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -27,6 +27,7 @@
 #include <linux/net.h>
 #include <linux/pm_runtime.h>
 #include <net/devlink.h>
+#include <net/ipv6.h>
 #include <net/xdp_sock_drv.h>
 #include <net/flow_offload.h>
 #include <linux/ethtool_netlink.h>
@@ -3127,7 +3128,6 @@ struct ethtool_rx_flow_rule *
 ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input)
 {
 	const struct ethtool_rx_flow_spec *fs = input->fs;
-	static struct in6_addr zero_addr = {};
 	struct ethtool_rx_flow_match *match;
 	struct ethtool_rx_flow_rule *flow;
 	struct flow_action_entry *act;
@@ -3233,20 +3233,20 @@ ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input)
 
 		v6_spec = &fs->h_u.tcp_ip6_spec;
 		v6_m_spec = &fs->m_u.tcp_ip6_spec;
-		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) {
+		if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6src)) {
 			memcpy(&match->key.ipv6.src, v6_spec->ip6src,
 			       sizeof(match->key.ipv6.src));
 			memcpy(&match->mask.ipv6.src, v6_m_spec->ip6src,
 			       sizeof(match->mask.ipv6.src));
 		}
-		if (memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) {
+		if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6dst)) {
 			memcpy(&match->key.ipv6.dst, v6_spec->ip6dst,
 			       sizeof(match->key.ipv6.dst));
 			memcpy(&match->mask.ipv6.dst, v6_m_spec->ip6dst,
 			       sizeof(match->mask.ipv6.dst));
 		}
-		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr)) ||
-		    memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) {
+		if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6src) ||
+		    !ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6dst)) {
 			match->dissector.used_keys |=
 				BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS);
 			match->dissector.offset[FLOW_DISSECTOR_KEY_IPV6_ADDRS] =
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 6edae3886885..e7391bf310a7 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -826,13 +826,11 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const
 				      unsigned short port, int l3mdev, const struct sock *sk)
 {
 #if IS_ENABLED(CONFIG_IPV6)
-	struct in6_addr addr_any = {};
-
 	if (sk->sk_family != tb->family) {
 		if (sk->sk_family == AF_INET)
 			return net_eq(ib2_net(tb), net) && tb->port == port &&
 				tb->l3mdev == l3mdev &&
-				ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any);
+				ipv6_addr_any(&tb->v6_rcv_saddr);
 
 		return false;
 	}
@@ -840,7 +838,7 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const
 	if (sk->sk_family == AF_INET6)
 		return net_eq(ib2_net(tb), net) && tb->port == port &&
 			tb->l3mdev == l3mdev &&
-			ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any);
+			ipv6_addr_any(&tb->v6_rcv_saddr);
 	else
 #endif
 		return net_eq(ib2_net(tb), net) && tb->port == port &&
@@ -866,11 +864,10 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
 {
 	struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
 	u32 hash;
-#if IS_ENABLED(CONFIG_IPV6)
-	struct in6_addr addr_any = {};
 
+#if IS_ENABLED(CONFIG_IPV6)
 	if (sk->sk_family == AF_INET6)
-		hash = ipv6_portaddr_hash(net, &addr_any, port);
+		hash = ipv6_portaddr_hash(net, &in6addr_any, port);
 	else
 #endif
 		hash = ipv4_portaddr_hash(net, 0, port);
-- 
cgit v1.2.3-70-g09d2


From be689c719eb6153b0fbd83536ae656dca8981f12 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Mon, 27 Mar 2023 16:54:55 -0700
Subject: 6lowpan: Remove redundant initialisation.

We'll call memset(&tmp, 0, sizeof(tmp)) later.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/6lowpan/iphc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c
index 52fad5dad9f7..e116d308a8df 100644
--- a/net/6lowpan/iphc.c
+++ b/net/6lowpan/iphc.c
@@ -848,7 +848,7 @@ static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct net_device *dev,
 				   const struct lowpan_iphc_ctx *ctx,
 				   const unsigned char *lladdr, bool sam)
 {
-	struct in6_addr tmp = {};
+	struct in6_addr tmp;
 	u8 dam;
 
 	switch (lowpan_dev(dev)->lltype) {
-- 
cgit v1.2.3-70-g09d2


From 2bb9a37f0e194ed95c70603b0efc7898a5a0d9b4 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Mon, 27 Mar 2023 12:22:21 +0200
Subject: mptcp: avoid unneeded address copy

In the syn_recv fallback path, the msk is unused. We can skip
setting the socket address.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/subflow.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index dadaf85db720..a11f4c525e01 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -821,8 +821,6 @@ create_child:
 				goto dispose_child;
 			}
 
-			if (new_msk)
-				mptcp_copy_inaddrs(new_msk, child);
 			mptcp_subflow_drop_ctx(child);
 			goto out;
 		}
-- 
cgit v1.2.3-70-g09d2


From a88d0092b24b8cddce57fe0e88e60a9e29e0b515 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Mon, 27 Mar 2023 12:22:22 +0200
Subject: mptcp: simplify subflow_syn_recv_sock()

Postpone the msk cloning to the child process creation
so that we can avoid a bunch of conditionals.

Link: https://github.com/multipath-tcp/mptcp_net-next/issues/61
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/subflow.c | 41 +++++++++++++----------------------------
 1 file changed, 13 insertions(+), 28 deletions(-)

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index a11f4c525e01..33dd27765116 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -696,14 +696,6 @@ static bool subflow_hmac_valid(const struct request_sock *req,
 	return !crypto_memneq(hmac, mp_opt->hmac, MPTCPOPT_HMAC_LEN);
 }
 
-static void mptcp_force_close(struct sock *sk)
-{
-	/* the msk is not yet exposed to user-space, and refcount is 2 */
-	inet_sk_state_store(sk, TCP_CLOSE);
-	sk_common_release(sk);
-	sock_put(sk);
-}
-
 static void subflow_ulp_fallback(struct sock *sk,
 				 struct mptcp_subflow_context *old_ctx)
 {
@@ -755,7 +747,6 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 	struct mptcp_subflow_request_sock *subflow_req;
 	struct mptcp_options_received mp_opt;
 	bool fallback, fallback_is_fatal;
-	struct sock *new_msk = NULL;
 	struct mptcp_sock *owner;
 	struct sock *child;
 
@@ -784,14 +775,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 		 * options.
 		 */
 		mptcp_get_options(skb, &mp_opt);
-		if (!(mp_opt.suboptions & OPTIONS_MPTCP_MPC)) {
+		if (!(mp_opt.suboptions & OPTIONS_MPTCP_MPC))
 			fallback = true;
-			goto create_child;
-		}
 
-		new_msk = mptcp_sk_clone(listener->conn, &mp_opt, req);
-		if (!new_msk)
-			fallback = true;
 	} else if (subflow_req->mp_join) {
 		mptcp_get_options(skb, &mp_opt);
 		if (!(mp_opt.suboptions & OPTIONS_MPTCP_MPJ) ||
@@ -820,21 +806,23 @@ create_child:
 				subflow_add_reset_reason(skb, MPTCP_RST_EMPTCP);
 				goto dispose_child;
 			}
-
-			mptcp_subflow_drop_ctx(child);
-			goto out;
+			goto fallback;
 		}
 
 		/* ssk inherits options of listener sk */
 		ctx->setsockopt_seq = listener->setsockopt_seq;
 
 		if (ctx->mp_capable) {
-			owner = mptcp_sk(new_msk);
+			ctx->conn = mptcp_sk_clone(listener->conn, &mp_opt, req);
+			if (!ctx->conn)
+				goto fallback;
+
+			owner = mptcp_sk(ctx->conn);
 
 			/* this can't race with mptcp_close(), as the msk is
 			 * not yet exposted to user-space
 			 */
-			inet_sk_state_store((void *)new_msk, TCP_ESTABLISHED);
+			inet_sk_state_store(ctx->conn, TCP_ESTABLISHED);
 
 			/* record the newly created socket as the first msk
 			 * subflow, but don't link it yet into conn_list
@@ -844,11 +832,9 @@ create_child:
 			/* new mpc subflow takes ownership of the newly
 			 * created mptcp socket
 			 */
-			mptcp_sk(new_msk)->setsockopt_seq = ctx->setsockopt_seq;
+			owner->setsockopt_seq = ctx->setsockopt_seq;
 			mptcp_pm_new_connection(owner, child, 1);
 			mptcp_token_accept(subflow_req, owner);
-			ctx->conn = new_msk;
-			new_msk = NULL;
 
 			/* set msk addresses early to ensure mptcp_pm_get_local_id()
 			 * uses the correct data
@@ -898,11 +884,6 @@ create_child:
 		}
 	}
 
-out:
-	/* dispose of the left over mptcp master, if any */
-	if (unlikely(new_msk))
-		mptcp_force_close(new_msk);
-
 	/* check for expected invariant - should never trigger, just help
 	 * catching eariler subtle bugs
 	 */
@@ -920,6 +901,10 @@ dispose_child:
 
 	/* The last child reference will be released by the caller */
 	return child;
+
+fallback:
+	mptcp_subflow_drop_ctx(child);
+	return child;
 }
 
 static struct inet_connection_sock_af_ops subflow_specific __ro_after_init;
-- 
cgit v1.2.3-70-g09d2


From e925a0322ada48a3bfdfaf37be46945af3ad6e2e Mon Sep 17 00:00:00 2001
From: Matthieu Baerts <matthieu.baerts@tessares.net>
Date: Mon, 27 Mar 2023 12:22:23 +0200
Subject: mptcp: do not fill info not used by the PM in used

Only the in-kernel PM uses the number of address and subflow limits
allowed per connection.

It then makes more sense not to display such info when other PMs are
used not to confuse the userspace by showing limits not being used.

While at it, we can get rid of the "val" variable and add indentations
instead.

It would have been good to have done this modification directly in
commit 4d25247d3ae4 ("mptcp: bypass in-kernel PM restrictions for non-kernel PMs")
but as we change a bit the behaviour, it is fine not to backport it to
stable.

Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/sockopt.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c
index 5cef4d3d21ac..b655cebda0f3 100644
--- a/net/mptcp/sockopt.c
+++ b/net/mptcp/sockopt.c
@@ -885,7 +885,6 @@ out:
 void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info)
 {
 	u32 flags = 0;
-	u8 val;
 
 	memset(info, 0, sizeof(*info));
 
@@ -893,12 +892,19 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info)
 	info->mptcpi_add_addr_signal = READ_ONCE(msk->pm.add_addr_signaled);
 	info->mptcpi_add_addr_accepted = READ_ONCE(msk->pm.add_addr_accepted);
 	info->mptcpi_local_addr_used = READ_ONCE(msk->pm.local_addr_used);
-	info->mptcpi_subflows_max = mptcp_pm_get_subflows_max(msk);
-	val = mptcp_pm_get_add_addr_signal_max(msk);
-	info->mptcpi_add_addr_signal_max = val;
-	val = mptcp_pm_get_add_addr_accept_max(msk);
-	info->mptcpi_add_addr_accepted_max = val;
-	info->mptcpi_local_addr_max = mptcp_pm_get_local_addr_max(msk);
+
+	/* The following limits only make sense for the in-kernel PM */
+	if (mptcp_pm_is_kernel(msk)) {
+		info->mptcpi_subflows_max =
+			mptcp_pm_get_subflows_max(msk);
+		info->mptcpi_add_addr_signal_max =
+			mptcp_pm_get_add_addr_signal_max(msk);
+		info->mptcpi_add_addr_accepted_max =
+			mptcp_pm_get_add_addr_accept_max(msk);
+		info->mptcpi_local_addr_max =
+			mptcp_pm_get_local_addr_max(msk);
+	}
+
 	if (test_bit(MPTCP_FALLBACK_DONE, &msk->flags))
 		flags |= MPTCP_INFO_FLAG_FALLBACK;
 	if (READ_ONCE(msk->can_ack))
-- 
cgit v1.2.3-70-g09d2


From 9095ce97bf8a4f5123dc9dbfd1cabcf2a78f5883 Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliang.tang@suse.com>
Date: Mon, 27 Mar 2023 12:22:24 +0200
Subject: selftests: mptcp: add mptcp_info tests

This patch adds the mptcp_info fields tests in endpoint_tests(). Add a
new function chk_mptcp_info() to check the given number of the given
mptcp_info field.

Link: https://github.com/multipath-tcp/mptcp_net-next/issues/330
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 47 ++++++++++++++++++++++++-
 1 file changed, 46 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 42e3bd1a05f5..fafd19ec7e1f 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -1719,6 +1719,46 @@ chk_subflow_nr()
 	fi
 }
 
+chk_mptcp_info()
+{
+	local nr_info=$1
+	local info
+	local cnt1
+	local cnt2
+	local dump_stats
+
+	if [[ $nr_info = "subflows_"* ]]; then
+		info="subflows"
+		nr_info=${nr_info:9}
+	else
+		echo "[fail] unsupported argument: $nr_info"
+		fail_test
+		return 1
+	fi
+
+	printf "%-${nr_blank}s %-30s" " " "mptcp_info $info=$nr_info"
+
+	cnt1=$(ss -N $ns1 -inmHM | grep "$info:" |
+		sed -n 's/.*\('"$info"':\)\([[:digit:]]*\).*$/\2/p;q')
+	[ -z "$cnt1" ] && cnt1=0
+	cnt2=$(ss -N $ns2 -inmHM | grep "$info:" |
+		sed -n 's/.*\('"$info"':\)\([[:digit:]]*\).*$/\2/p;q')
+	[ -z "$cnt2" ] && cnt2=0
+	if [ "$cnt1" != "$nr_info" ] || [ "$cnt2" != "$nr_info" ]; then
+		echo "[fail] got $cnt1:$cnt2 $info expected $nr_info"
+		fail_test
+		dump_stats=1
+	else
+		echo "[ ok ]"
+	fi
+
+	if [ "$dump_stats" = 1 ]; then
+		ss -N $ns1 -inmHM
+		ss -N $ns2 -inmHM
+		dump_stats
+	fi
+}
+
 chk_link_usage()
 {
 	local ns=$1
@@ -3118,13 +3158,18 @@ endpoint_tests()
 		run_tests $ns1 $ns2 10.0.1.1 4 0 0 speed_20 2>/dev/null &
 
 		wait_mpj $ns2
+		chk_subflow_nr needtitle "before delete" 2
+		chk_mptcp_info subflows_1
+
 		pm_nl_del_endpoint $ns2 2 10.0.2.2
 		sleep 0.5
-		chk_subflow_nr needtitle "after delete" 1
+		chk_subflow_nr "" "after delete" 1
+		chk_mptcp_info subflows_0
 
 		pm_nl_add_endpoint $ns2 10.0.2.2 dev ns2eth2 flags subflow
 		wait_mpj $ns2
 		chk_subflow_nr "" "after re-add" 2
+		chk_mptcp_info subflows_1
 		kill_tests_wait
 	fi
 }
-- 
cgit v1.2.3-70-g09d2


From d45276e75e90f1b67eb689fb3b4c556963796351 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 28 Mar 2023 10:57:56 +0800
Subject: macvlan: Skip broadcast queue if multicast with single receiver

As it stands all broadcast and multicast packets are queued and
processed in a work queue.  This is so that we don't overwhelm
the receive softirq path by generating thousands of packets or
more (see commit 412ca1550cbe "macvlan: Move broadcasts into a
work queue").

As such all multicast packets will be delayed, even if they will
be received by a single macvlan device.  As using a workqueue
is not free in terms of latency, we should avoid this where possible.

This patch adds a new filter to determine which addresses should
be delayed and which ones won't.  This is done using a crude
counter of how many times an address has been added to the macvlan
port (ha->synced).  For now if an address has been added more than
once, then it will be considered to be broadcast.  This could be
tuned further by making this threshold configurable.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macvlan.c | 74 ++++++++++++++++++++++++++++++++-------------------
 1 file changed, 46 insertions(+), 28 deletions(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 99a971929c8e..62b4748d3836 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -50,6 +50,7 @@ struct macvlan_port {
 	u32			flags;
 	int			count;
 	struct hlist_head	vlan_source_hash[MACVLAN_HASH_SIZE];
+	DECLARE_BITMAP(bc_filter, MACVLAN_MC_FILTER_SZ);
 	DECLARE_BITMAP(mc_filter, MACVLAN_MC_FILTER_SZ);
 	unsigned char           perm_addr[ETH_ALEN];
 };
@@ -291,6 +292,31 @@ static void macvlan_broadcast(struct sk_buff *skb,
 	}
 }
 
+static void macvlan_multicast_rx(const struct macvlan_port *port,
+				 const struct macvlan_dev *src,
+				 struct sk_buff *skb)
+{
+	if (!src)
+		/* frame comes from an external address */
+		macvlan_broadcast(skb, port, NULL,
+				  MACVLAN_MODE_PRIVATE |
+				  MACVLAN_MODE_VEPA    |
+				  MACVLAN_MODE_PASSTHRU|
+				  MACVLAN_MODE_BRIDGE);
+	else if (src->mode == MACVLAN_MODE_VEPA)
+		/* flood to everyone except source */
+		macvlan_broadcast(skb, port, src->dev,
+				  MACVLAN_MODE_VEPA |
+				  MACVLAN_MODE_BRIDGE);
+	else
+		/*
+		 * flood only to VEPA ports, bridge ports
+		 * already saw the frame on the way out.
+		 */
+		macvlan_broadcast(skb, port, src->dev,
+				  MACVLAN_MODE_VEPA);
+}
+
 static void macvlan_process_broadcast(struct work_struct *w)
 {
 	struct macvlan_port *port = container_of(w, struct macvlan_port,
@@ -308,27 +334,7 @@ static void macvlan_process_broadcast(struct work_struct *w)
 		const struct macvlan_dev *src = MACVLAN_SKB_CB(skb)->src;
 
 		rcu_read_lock();
-
-		if (!src)
-			/* frame comes from an external address */
-			macvlan_broadcast(skb, port, NULL,
-					  MACVLAN_MODE_PRIVATE |
-					  MACVLAN_MODE_VEPA    |
-					  MACVLAN_MODE_PASSTHRU|
-					  MACVLAN_MODE_BRIDGE);
-		else if (src->mode == MACVLAN_MODE_VEPA)
-			/* flood to everyone except source */
-			macvlan_broadcast(skb, port, src->dev,
-					  MACVLAN_MODE_VEPA |
-					  MACVLAN_MODE_BRIDGE);
-		else
-			/*
-			 * flood only to VEPA ports, bridge ports
-			 * already saw the frame on the way out.
-			 */
-			macvlan_broadcast(skb, port, src->dev,
-					  MACVLAN_MODE_VEPA);
-
+		macvlan_multicast_rx(port, src, skb);
 		rcu_read_unlock();
 
 		if (src)
@@ -476,8 +482,10 @@ static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb)
 		}
 
 		hash = mc_hash(NULL, eth->h_dest);
-		if (test_bit(hash, port->mc_filter))
+		if (test_bit(hash, port->bc_filter))
 			macvlan_broadcast_enqueue(port, src, skb);
+		else if (test_bit(hash, port->mc_filter))
+			macvlan_multicast_rx(port, src, skb);
 
 		return RX_HANDLER_PASS;
 	}
@@ -780,20 +788,27 @@ static void macvlan_change_rx_flags(struct net_device *dev, int change)
 
 static void macvlan_compute_filter(unsigned long *mc_filter,
 				   struct net_device *dev,
-				   struct macvlan_dev *vlan)
+				   struct macvlan_dev *vlan, int cutoff)
 {
 	if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) {
-		bitmap_fill(mc_filter, MACVLAN_MC_FILTER_SZ);
+		if (cutoff >= 0)
+			bitmap_fill(mc_filter, MACVLAN_MC_FILTER_SZ);
+		else
+			bitmap_zero(mc_filter, MACVLAN_MC_FILTER_SZ);
 	} else {
-		struct netdev_hw_addr *ha;
 		DECLARE_BITMAP(filter, MACVLAN_MC_FILTER_SZ);
+		struct netdev_hw_addr *ha;
 
 		bitmap_zero(filter, MACVLAN_MC_FILTER_SZ);
 		netdev_for_each_mc_addr(ha, dev) {
+			if (cutoff >= 0 && ha->synced <= cutoff)
+				continue;
+
 			__set_bit(mc_hash(vlan, ha->addr), filter);
 		}
 
-		__set_bit(mc_hash(vlan, dev->broadcast), filter);
+		if (cutoff >= 0)
+			__set_bit(mc_hash(vlan, dev->broadcast), filter);
 
 		bitmap_copy(mc_filter, filter, MACVLAN_MC_FILTER_SZ);
 	}
@@ -803,7 +818,7 @@ static void macvlan_set_mac_lists(struct net_device *dev)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
 
-	macvlan_compute_filter(vlan->mc_filter, dev, vlan);
+	macvlan_compute_filter(vlan->mc_filter, dev, vlan, 0);
 
 	dev_uc_sync(vlan->lowerdev, dev);
 	dev_mc_sync(vlan->lowerdev, dev);
@@ -821,7 +836,10 @@ static void macvlan_set_mac_lists(struct net_device *dev)
 	 * The solution is to maintain a list of broadcast addresses like
 	 * we do for uc/mc, if you care.
 	 */
-	macvlan_compute_filter(vlan->port->mc_filter, vlan->lowerdev, NULL);
+	macvlan_compute_filter(vlan->port->mc_filter, vlan->lowerdev, NULL,
+			       0);
+	macvlan_compute_filter(vlan->port->bc_filter, vlan->lowerdev, NULL,
+			       1);
 }
 
 static int macvlan_change_mtu(struct net_device *dev, int new_mtu)
-- 
cgit v1.2.3-70-g09d2


From 954d1fa1ac93aa8a66f7d9a9ba545cf7f020d348 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 28 Mar 2023 10:57:59 +0800
Subject: macvlan: Add netlink attribute for broadcast cutoff

Make the broadcast cutoff configurable through netlink.  Note
that macvlan is weird because there is no central device for
us to configure (the lowerdev could be anything).  So all the
options are duplicated over what could be thousands of child
devices.

IFLA_MACVLAN_BC_QUEUE_LEN took the approach of taking the maximum
of all child device settings.  This is unnecessary as we could
simply store the option in the port device and take the last
child device that gets updated as the value to use.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macvlan.c              | 31 +++++++++++++++++++++++++++++--
 include/uapi/linux/if_link.h       |  1 +
 tools/include/uapi/linux/if_link.h |  1 +
 3 files changed, 31 insertions(+), 2 deletions(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 62b4748d3836..4215106adc40 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -47,6 +47,7 @@ struct macvlan_port {
 	struct sk_buff_head	bc_queue;
 	struct work_struct	bc_work;
 	u32			bc_queue_len_used;
+	int			bc_cutoff;
 	u32			flags;
 	int			count;
 	struct hlist_head	vlan_source_hash[MACVLAN_HASH_SIZE];
@@ -814,6 +815,12 @@ static void macvlan_compute_filter(unsigned long *mc_filter,
 	}
 }
 
+static void macvlan_recompute_bc_filter(struct macvlan_dev *vlan)
+{
+	macvlan_compute_filter(vlan->port->bc_filter, vlan->lowerdev, NULL,
+			       vlan->port->bc_cutoff);
+}
+
 static void macvlan_set_mac_lists(struct net_device *dev)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
@@ -838,8 +845,16 @@ static void macvlan_set_mac_lists(struct net_device *dev)
 	 */
 	macvlan_compute_filter(vlan->port->mc_filter, vlan->lowerdev, NULL,
 			       0);
-	macvlan_compute_filter(vlan->port->bc_filter, vlan->lowerdev, NULL,
-			       1);
+	macvlan_recompute_bc_filter(vlan);
+}
+
+static void update_port_bc_cutoff(struct macvlan_dev *vlan, int cutoff)
+{
+	if (vlan->port->bc_cutoff == cutoff)
+		return;
+
+	vlan->port->bc_cutoff = cutoff;
+	macvlan_recompute_bc_filter(vlan);
 }
 
 static int macvlan_change_mtu(struct net_device *dev, int new_mtu)
@@ -1254,6 +1269,7 @@ static int macvlan_port_create(struct net_device *dev)
 		INIT_HLIST_HEAD(&port->vlan_source_hash[i]);
 
 	port->bc_queue_len_used = 0;
+	port->bc_cutoff = 1;
 	skb_queue_head_init(&port->bc_queue);
 	INIT_WORK(&port->bc_work, macvlan_process_broadcast);
 
@@ -1527,6 +1543,10 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev,
 	if (data && data[IFLA_MACVLAN_BC_QUEUE_LEN])
 		vlan->bc_queue_len_req = nla_get_u32(data[IFLA_MACVLAN_BC_QUEUE_LEN]);
 
+	if (data && data[IFLA_MACVLAN_BC_CUTOFF])
+		update_port_bc_cutoff(
+			vlan, nla_get_s32(data[IFLA_MACVLAN_BC_CUTOFF]));
+
 	err = register_netdevice(dev);
 	if (err < 0)
 		goto destroy_macvlan_port;
@@ -1623,6 +1643,10 @@ static int macvlan_changelink(struct net_device *dev,
 		update_port_bc_queue_len(vlan->port);
 	}
 
+	if (data && data[IFLA_MACVLAN_BC_CUTOFF])
+		update_port_bc_cutoff(
+			vlan, nla_get_s32(data[IFLA_MACVLAN_BC_CUTOFF]));
+
 	if (set_mode)
 		vlan->mode = mode;
 	if (data && data[IFLA_MACVLAN_MACADDR_MODE]) {
@@ -1703,6 +1727,9 @@ static int macvlan_fill_info(struct sk_buff *skb,
 		goto nla_put_failure;
 	if (nla_put_u32(skb, IFLA_MACVLAN_BC_QUEUE_LEN_USED, port->bc_queue_len_used))
 		goto nla_put_failure;
+	if (port->bc_cutoff != 1 &&
+	    nla_put_s32(skb, IFLA_MACVLAN_BC_CUTOFF, port->bc_cutoff))
+		goto nla_put_failure;
 	return 0;
 
 nla_put_failure:
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 57ceb788250f..8d679688efe0 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -635,6 +635,7 @@ enum {
 	IFLA_MACVLAN_MACADDR_COUNT,
 	IFLA_MACVLAN_BC_QUEUE_LEN,
 	IFLA_MACVLAN_BC_QUEUE_LEN_USED,
+	IFLA_MACVLAN_BC_CUTOFF,
 	__IFLA_MACVLAN_MAX,
 };
 
diff --git a/tools/include/uapi/linux/if_link.h b/tools/include/uapi/linux/if_link.h
index 901d98b865a1..39e659c83cfd 100644
--- a/tools/include/uapi/linux/if_link.h
+++ b/tools/include/uapi/linux/if_link.h
@@ -605,6 +605,7 @@ enum {
 	IFLA_MACVLAN_MACADDR_COUNT,
 	IFLA_MACVLAN_BC_QUEUE_LEN,
 	IFLA_MACVLAN_BC_QUEUE_LEN_USED,
+	IFLA_MACVLAN_BC_CUTOFF,
 	__IFLA_MACVLAN_MAX,
 };
 
-- 
cgit v1.2.3-70-g09d2


From edd025ca088719770c77a27266bf03f2f39ce83d Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Mon, 27 Mar 2023 11:36:03 +0100
Subject: sfc: document TC-to-EF100-MAE action translation concepts

Includes an explanation of the lifetime of the 'cursor' action-set `act`.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/tc.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c
index 2b07bb2fd735..34c1ff87ba5e 100644
--- a/drivers/net/ethernet/sfc/tc.c
+++ b/drivers/net/ethernet/sfc/tc.c
@@ -66,7 +66,7 @@ static const struct rhashtable_params efx_tc_match_action_ht_params = {
 static void efx_tc_free_action_set(struct efx_nic *efx,
 				   struct efx_tc_action_set *act, bool in_hw)
 {
-	/* Failure paths calling this on the 'running action' set in_hw=false,
+	/* Failure paths calling this on the 'cursor' action set in_hw=false,
 	 * because if the alloc had succeeded we'd've put it in acts.list and
 	 * not still have it in act.
 	 */
@@ -407,6 +407,30 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 		goto release;
 	}
 
+	/**
+	 * DOC: TC action translation
+	 *
+	 * Actions in TC are sequential and cumulative, with delivery actions
+	 * potentially anywhere in the order.  The EF100 MAE, however, takes
+	 * an 'action set list' consisting of 'action sets', each of which is
+	 * applied to the _original_ packet, and consists of a set of optional
+	 * actions in a fixed order with delivery at the end.
+	 * To translate between these two models, we maintain a 'cursor', @act,
+	 * which describes the cumulative effect of all the packet-mutating
+	 * actions encountered so far; on handling a delivery (mirred or drop)
+	 * action, once the action-set has been inserted into hardware, we
+	 * append @act to the action-set list (@rule->acts); if this is a pipe
+	 * action (mirred mirror) we then allocate a new @act with a copy of
+	 * the cursor state _before_ the delivery action, otherwise we set @act
+	 * to %NULL.
+	 * This ensures that every allocated action-set is either attached to
+	 * @rule->acts or pointed to by @act (and never both), and that only
+	 * those action-sets in @rule->acts exist in hardware.  Consequently,
+	 * in the failure path, @act only needs to be freed in memory, whereas
+	 * for @rule->acts we remove each action-set from hardware before
+	 * freeing it (efx_tc_free_action_set_list()), even if the action-set
+	 * list itself is not in hardware.
+	 */
 	flow_action_for_each(i, fa, &fr->action) {
 		struct efx_tc_action_set save;
 		u16 tci;
-- 
cgit v1.2.3-70-g09d2


From b9d5c9b7d8a47ca4af5e32f576265c388a467af5 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Mon, 27 Mar 2023 11:36:04 +0100
Subject: sfc: add notion of match on enc keys to MAE machinery

Extend the MAE caps check to validate that the hardware supports these
 outer-header matches where used by the driver.
Extend efx_mae_populate_match_criteria() to fill in the outer rule ID
 and VNI match fields.
Nothing yet populates these match fields, nor creates outer rules.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mae.c | 97 +++++++++++++++++++++++++++++++++++++++++-
 drivers/net/ethernet/sfc/mae.h |  3 ++
 drivers/net/ethernet/sfc/tc.h  | 24 +++++++++++
 3 files changed, 122 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/sfc/mae.c b/drivers/net/ethernet/sfc/mae.c
index c53d354c1fb2..2290a63908c5 100644
--- a/drivers/net/ethernet/sfc/mae.c
+++ b/drivers/net/ethernet/sfc/mae.c
@@ -254,13 +254,23 @@ static int efx_mae_get_rule_fields(struct efx_nic *efx, u32 cmd,
 	size_t outlen;
 	int rc, i;
 
+	/* AR and OR caps MCDIs have identical layout, so we are using the
+	 * same code for both.
+	 */
+	BUILD_BUG_ON(MC_CMD_MAE_GET_AR_CAPS_OUT_LEN(MAE_NUM_FIELDS) <
+		     MC_CMD_MAE_GET_OR_CAPS_OUT_LEN(MAE_NUM_FIELDS));
 	BUILD_BUG_ON(MC_CMD_MAE_GET_AR_CAPS_IN_LEN);
+	BUILD_BUG_ON(MC_CMD_MAE_GET_OR_CAPS_IN_LEN);
 
 	rc = efx_mcdi_rpc(efx, cmd, NULL, 0, outbuf, sizeof(outbuf), &outlen);
 	if (rc)
 		return rc;
+	BUILD_BUG_ON(MC_CMD_MAE_GET_AR_CAPS_OUT_COUNT_OFST !=
+		     MC_CMD_MAE_GET_OR_CAPS_OUT_COUNT_OFST);
 	count = MCDI_DWORD(outbuf, MAE_GET_AR_CAPS_OUT_COUNT);
 	memset(field_support, MAE_FIELD_UNSUPPORTED, MAE_NUM_FIELDS);
+	BUILD_BUG_ON(MC_CMD_MAE_GET_AR_CAPS_OUT_FIELD_FLAGS_OFST !=
+		     MC_CMD_MAE_GET_OR_CAPS_OUT_FIELD_FLAGS_OFST);
 	caps = _MCDI_DWORD(outbuf, MAE_GET_AR_CAPS_OUT_FIELD_FLAGS);
 	/* We're only interested in the support status enum, not any other
 	 * flags, so just extract that from each entry.
@@ -278,8 +288,12 @@ int efx_mae_get_caps(struct efx_nic *efx, struct mae_caps *caps)
 	rc = efx_mae_get_basic_caps(efx, caps);
 	if (rc)
 		return rc;
-	return efx_mae_get_rule_fields(efx, MC_CMD_MAE_GET_AR_CAPS,
-				       caps->action_rule_fields);
+	rc = efx_mae_get_rule_fields(efx, MC_CMD_MAE_GET_AR_CAPS,
+				     caps->action_rule_fields);
+	if (rc)
+		return rc;
+	return efx_mae_get_rule_fields(efx, MC_CMD_MAE_GET_OR_CAPS,
+				       caps->outer_rule_fields);
 }
 
 /* Bit twiddling:
@@ -432,11 +446,67 @@ int efx_mae_match_check_caps(struct efx_nic *efx,
 	    CHECK_BIT(IP_FIRST_FRAG, ip_firstfrag) ||
 	    CHECK(RECIRC_ID, recirc_id))
 		return rc;
+	/* Matches on outer fields are done in a separate hardware table,
+	 * the Outer Rule table.  Thus the Action Rule merely does an
+	 * exact match on Outer Rule ID if any outer field matches are
+	 * present.  The exception is the VNI/VSID (enc_keyid), which is
+	 * available to the Action Rule match iff the Outer Rule matched
+	 * (and thus identified the encap protocol to use to extract it).
+	 */
+	if (efx_tc_match_is_encap(mask)) {
+		rc = efx_mae_match_check_cap_typ(
+				supported_fields[MAE_FIELD_OUTER_RULE_ID],
+				MASK_ONES);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "No support for encap rule ID matches");
+			return rc;
+		}
+		if (CHECK(ENC_VNET_ID, enc_keyid))
+			return rc;
+	} else if (mask->enc_keyid) {
+		NL_SET_ERR_MSG_MOD(extack, "Match on enc_keyid requires other encap fields");
+		return -EINVAL;
+	}
 	return 0;
 }
 #undef CHECK_BIT
 #undef CHECK
 
+#define CHECK(_mcdi)	({						       \
+	rc = efx_mae_match_check_cap_typ(supported_fields[MAE_FIELD_ ## _mcdi],\
+					 MASK_ONES);			       \
+	if (rc)								       \
+		NL_SET_ERR_MSG_FMT_MOD(extack,				       \
+				       "No support for field %s", #_mcdi);     \
+	rc;								       \
+})
+/* Checks that the fields needed for encap-rule matches are supported by the
+ * MAE.  All the fields are exact-match.
+ */
+int efx_mae_check_encap_match_caps(struct efx_nic *efx, bool ipv6,
+				   struct netlink_ext_ack *extack)
+{
+	u8 *supported_fields = efx->tc->caps->outer_rule_fields;
+	int rc;
+
+	if (CHECK(ENC_ETHER_TYPE))
+		return rc;
+	if (ipv6) {
+		if (CHECK(ENC_SRC_IP6) ||
+		    CHECK(ENC_DST_IP6))
+			return rc;
+	} else {
+		if (CHECK(ENC_SRC_IP4) ||
+		    CHECK(ENC_DST_IP4))
+			return rc;
+	}
+	if (CHECK(ENC_L4_DPORT) ||
+	    CHECK(ENC_IP_PROTO))
+		return rc;
+	return 0;
+}
+#undef CHECK
+
 int efx_mae_allocate_counter(struct efx_nic *efx, struct efx_tc_counter *cnt)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_COUNTER_ALLOC_OUT_LEN(1));
@@ -941,6 +1011,29 @@ static int efx_mae_populate_match_criteria(MCDI_DECLARE_STRUCT_PTR(match_crit),
 				match->value.tcp_flags);
 	MCDI_STRUCT_SET_WORD_BE(match_crit, MAE_FIELD_MASK_VALUE_PAIRS_V2_TCP_FLAGS_BE_MASK,
 				match->mask.tcp_flags);
+	/* enc-keys are handled indirectly, through encap_match ID */
+	if (match->encap) {
+		MCDI_STRUCT_SET_DWORD(match_crit, MAE_FIELD_MASK_VALUE_PAIRS_V2_OUTER_RULE_ID,
+				      match->encap->fw_id);
+		MCDI_STRUCT_SET_DWORD(match_crit, MAE_FIELD_MASK_VALUE_PAIRS_V2_OUTER_RULE_ID_MASK,
+				      U32_MAX);
+		/* enc_keyid (VNI/VSID) is not part of the encap_match */
+		MCDI_STRUCT_SET_DWORD_BE(match_crit, MAE_FIELD_MASK_VALUE_PAIRS_V2_ENC_VNET_ID_BE,
+					 match->value.enc_keyid);
+		MCDI_STRUCT_SET_DWORD_BE(match_crit, MAE_FIELD_MASK_VALUE_PAIRS_V2_ENC_VNET_ID_BE_MASK,
+					 match->mask.enc_keyid);
+	} else if (WARN_ON_ONCE(match->mask.enc_src_ip) ||
+		   WARN_ON_ONCE(match->mask.enc_dst_ip) ||
+		   WARN_ON_ONCE(!ipv6_addr_any(&match->mask.enc_src_ip6)) ||
+		   WARN_ON_ONCE(!ipv6_addr_any(&match->mask.enc_dst_ip6)) ||
+		   WARN_ON_ONCE(match->mask.enc_ip_tos) ||
+		   WARN_ON_ONCE(match->mask.enc_ip_ttl) ||
+		   WARN_ON_ONCE(match->mask.enc_sport) ||
+		   WARN_ON_ONCE(match->mask.enc_dport) ||
+		   WARN_ON_ONCE(match->mask.enc_keyid)) {
+		/* No enc-keys should appear in a rule without an encap_match */
+		return -EOPNOTSUPP;
+	}
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/sfc/mae.h b/drivers/net/ethernet/sfc/mae.h
index bec293a06733..2ccbc62d79b9 100644
--- a/drivers/net/ethernet/sfc/mae.h
+++ b/drivers/net/ethernet/sfc/mae.h
@@ -72,6 +72,7 @@ struct mae_caps {
 	u32 match_field_count;
 	u32 action_prios;
 	u8 action_rule_fields[MAE_NUM_FIELDS];
+	u8 outer_rule_fields[MAE_NUM_FIELDS];
 };
 
 int efx_mae_get_caps(struct efx_nic *efx, struct mae_caps *caps);
@@ -79,6 +80,8 @@ int efx_mae_get_caps(struct efx_nic *efx, struct mae_caps *caps);
 int efx_mae_match_check_caps(struct efx_nic *efx,
 			     const struct efx_tc_match_fields *mask,
 			     struct netlink_ext_ack *extack);
+int efx_mae_check_encap_match_caps(struct efx_nic *efx, bool ipv6,
+				   struct netlink_ext_ack *extack);
 
 int efx_mae_allocate_counter(struct efx_nic *efx, struct efx_tc_counter *cnt);
 int efx_mae_free_counter(struct efx_nic *efx, struct efx_tc_counter *cnt);
diff --git a/drivers/net/ethernet/sfc/tc.h b/drivers/net/ethernet/sfc/tc.h
index 542853f60c2a..c1485679507c 100644
--- a/drivers/net/ethernet/sfc/tc.h
+++ b/drivers/net/ethernet/sfc/tc.h
@@ -48,11 +48,35 @@ struct efx_tc_match_fields {
 	/* L4 */
 	__be16 l4_sport, l4_dport; /* Ports (UDP, TCP) */
 	__be16 tcp_flags;
+	/* Encap.  The following are *outer* fields.  Note that there are no
+	 * outer eth (L2) fields; this is because TC doesn't have them.
+	 */
+	__be32 enc_src_ip, enc_dst_ip;
+	struct in6_addr enc_src_ip6, enc_dst_ip6;
+	u8 enc_ip_tos, enc_ip_ttl;
+	__be16 enc_sport, enc_dport;
+	__be32 enc_keyid; /* e.g. VNI, VSID */
+};
+
+static inline bool efx_tc_match_is_encap(const struct efx_tc_match_fields *mask)
+{
+	return mask->enc_src_ip || mask->enc_dst_ip ||
+	       !ipv6_addr_any(&mask->enc_src_ip6) ||
+	       !ipv6_addr_any(&mask->enc_dst_ip6) || mask->enc_ip_tos ||
+	       mask->enc_ip_ttl || mask->enc_sport || mask->enc_dport;
+}
+
+struct efx_tc_encap_match {
+	__be32 src_ip, dst_ip;
+	struct in6_addr src_ip6, dst_ip6;
+	__be16 udp_dport;
+	u32 fw_id; /* index of this entry in firmware encap match table */
 };
 
 struct efx_tc_match {
 	struct efx_tc_match_fields value;
 	struct efx_tc_match_fields mask;
+	struct efx_tc_encap_match *encap;
 };
 
 struct efx_tc_action_set_list {
-- 
cgit v1.2.3-70-g09d2


From b7f5e17b3bb96165979f55f2821190a77f984d0e Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Mon, 27 Mar 2023 11:36:05 +0100
Subject: sfc: handle enc keys in efx_tc_flower_parse_match()

Translate the fields from flow dissector into struct efx_tc_match.
In efx_tc_flower_replace(), reject filters that match on them, because
 only 'foreign' filters (i.e. those for which the ingress dev is not
 the sfc netdev or any of its representors, e.g. a tunnel netdev) can
 use them.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/tc.c | 61 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 61 insertions(+)

diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c
index 34c1ff87ba5e..21eb79b20978 100644
--- a/drivers/net/ethernet/sfc/tc.c
+++ b/drivers/net/ethernet/sfc/tc.c
@@ -193,6 +193,11 @@ static int efx_tc_flower_parse_match(struct efx_nic *efx,
 	      BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
 	      BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
 	      BIT(FLOW_DISSECTOR_KEY_PORTS) |
+	      BIT(FLOW_DISSECTOR_KEY_ENC_KEYID) |
+	      BIT(FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) |
+	      BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) |
+	      BIT(FLOW_DISSECTOR_KEY_ENC_PORTS) |
+	      BIT(FLOW_DISSECTOR_KEY_ENC_CONTROL) |
 	      BIT(FLOW_DISSECTOR_KEY_TCP) |
 	      BIT(FLOW_DISSECTOR_KEY_IP))) {
 		NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported flower keys %#x",
@@ -280,6 +285,57 @@ static int efx_tc_flower_parse_match(struct efx_nic *efx,
 	MAP_KEY_AND_MASK(PORTS, ports, src, l4_sport);
 	MAP_KEY_AND_MASK(PORTS, ports, dst, l4_dport);
 	MAP_KEY_AND_MASK(TCP, tcp, flags, tcp_flags);
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) {
+		struct flow_match_control fm;
+
+		flow_rule_match_enc_control(rule, &fm);
+		if (fm.mask->flags) {
+			NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported match on enc_control.flags %#x",
+					       fm.mask->flags);
+			return -EOPNOTSUPP;
+		}
+		if (!IS_ALL_ONES(fm.mask->addr_type)) {
+			NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported enc addr_type mask %u (key %u)",
+					       fm.mask->addr_type,
+					       fm.key->addr_type);
+			return -EOPNOTSUPP;
+		}
+		switch (fm.key->addr_type) {
+		case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
+			MAP_ENC_KEY_AND_MASK(IPV4_ADDRS, ipv4_addrs, enc_ipv4_addrs,
+					     src, enc_src_ip);
+			MAP_ENC_KEY_AND_MASK(IPV4_ADDRS, ipv4_addrs, enc_ipv4_addrs,
+					     dst, enc_dst_ip);
+			break;
+#ifdef CONFIG_IPV6
+		case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
+			MAP_ENC_KEY_AND_MASK(IPV6_ADDRS, ipv6_addrs, enc_ipv6_addrs,
+					     src, enc_src_ip6);
+			MAP_ENC_KEY_AND_MASK(IPV6_ADDRS, ipv6_addrs, enc_ipv6_addrs,
+					     dst, enc_dst_ip6);
+			break;
+#endif
+		default:
+			NL_SET_ERR_MSG_FMT_MOD(extack,
+					       "Unsupported enc addr_type %u (supported are IPv4, IPv6)",
+					       fm.key->addr_type);
+			return -EOPNOTSUPP;
+		}
+		MAP_ENC_KEY_AND_MASK(IP, ip, enc_ip, tos, enc_ip_tos);
+		MAP_ENC_KEY_AND_MASK(IP, ip, enc_ip, ttl, enc_ip_ttl);
+		MAP_ENC_KEY_AND_MASK(PORTS, ports, enc_ports, src, enc_sport);
+		MAP_ENC_KEY_AND_MASK(PORTS, ports, enc_ports, dst, enc_dport);
+		MAP_ENC_KEY_AND_MASK(KEYID, enc_keyid, enc_keyid, keyid, enc_keyid);
+	} else if (dissector->used_keys &
+		   (BIT(FLOW_DISSECTOR_KEY_ENC_KEYID) |
+		    BIT(FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) |
+		    BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) |
+		    BIT(FLOW_DISSECTOR_KEY_ENC_IP) |
+		    BIT(FLOW_DISSECTOR_KEY_ENC_PORTS))) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Flower enc keys require enc_control (keys: %#x)",
+				       dissector->used_keys);
+		return -EOPNOTSUPP;
+	}
 
 	return 0;
 }
@@ -373,6 +429,11 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 	rc = efx_tc_flower_parse_match(efx, fr, &match, extack);
 	if (rc)
 		return rc;
+	if (efx_tc_match_is_encap(&match.mask)) {
+		NL_SET_ERR_MSG_MOD(extack, "Ingress enc_key matches not supported");
+		rc = -EOPNOTSUPP;
+		goto release;
+	}
 
 	if (tc->common.chain_index) {
 		NL_SET_ERR_MSG_MOD(extack, "No support for nonzero chain_index");
-- 
cgit v1.2.3-70-g09d2


From 2245eb0086d81e54dfc6492f649059465ebbed79 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Mon, 27 Mar 2023 11:36:06 +0100
Subject: sfc: add functions to insert encap matches into the MAE

An encap match corresponds to an entry in the exact-match Outer Rule
 table; the lookup response includes the encap type (protocol) allowing
 the hardware to continue parsing into the inner headers.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mae.c | 105 +++++++++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/mae.h |   5 ++
 drivers/net/ethernet/sfc/tc.h  |   1 +
 3 files changed, 111 insertions(+)

diff --git a/drivers/net/ethernet/sfc/mae.c b/drivers/net/ethernet/sfc/mae.c
index 2290a63908c5..92f1383ee4b9 100644
--- a/drivers/net/ethernet/sfc/mae.c
+++ b/drivers/net/ethernet/sfc/mae.c
@@ -558,6 +558,20 @@ int efx_mae_free_counter(struct efx_nic *efx, struct efx_tc_counter *cnt)
 	return 0;
 }
 
+static int efx_mae_encap_type_to_mae_type(enum efx_encap_type type)
+{
+	switch (type & EFX_ENCAP_TYPES_MASK) {
+	case EFX_ENCAP_TYPE_NONE:
+		return MAE_MCDI_ENCAP_TYPE_NONE;
+	case EFX_ENCAP_TYPE_VXLAN:
+		return MAE_MCDI_ENCAP_TYPE_VXLAN;
+	case EFX_ENCAP_TYPE_GENEVE:
+		return MAE_MCDI_ENCAP_TYPE_GENEVE;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
 int efx_mae_lookup_mport(struct efx_nic *efx, u32 vf_idx, u32 *id)
 {
 	struct ef100_nic_data *nic_data = efx->nic_data;
@@ -915,6 +929,97 @@ int efx_mae_free_action_set_list(struct efx_nic *efx,
 	return 0;
 }
 
+int efx_mae_register_encap_match(struct efx_nic *efx,
+				 struct efx_tc_encap_match *encap)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_MAE_OUTER_RULE_INSERT_IN_LEN(MAE_ENC_FIELD_PAIRS_LEN));
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_OUTER_RULE_INSERT_OUT_LEN);
+	MCDI_DECLARE_STRUCT_PTR(match_crit);
+	size_t outlen;
+	int rc;
+
+	rc = efx_mae_encap_type_to_mae_type(encap->tun_type);
+	if (rc < 0)
+		return rc;
+	match_crit = _MCDI_DWORD(inbuf, MAE_OUTER_RULE_INSERT_IN_FIELD_MATCH_CRITERIA);
+	/* The struct contains IP src and dst, and udp dport.
+	 * So we actually need to filter on IP src and dst, L4 dport, and
+	 * ipproto == udp.
+	 */
+	MCDI_SET_DWORD(inbuf, MAE_OUTER_RULE_INSERT_IN_ENCAP_TYPE, rc);
+#ifdef CONFIG_IPV6
+	if (encap->src_ip | encap->dst_ip) {
+#endif
+		MCDI_STRUCT_SET_DWORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_SRC_IP4_BE,
+					 encap->src_ip);
+		MCDI_STRUCT_SET_DWORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_SRC_IP4_BE_MASK,
+					 ~(__be32)0);
+		MCDI_STRUCT_SET_DWORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_DST_IP4_BE,
+					 encap->dst_ip);
+		MCDI_STRUCT_SET_DWORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_DST_IP4_BE_MASK,
+					 ~(__be32)0);
+		MCDI_STRUCT_SET_WORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_ETHER_TYPE_BE,
+					htons(ETH_P_IP));
+#ifdef CONFIG_IPV6
+	} else {
+		memcpy(MCDI_STRUCT_PTR(match_crit, MAE_ENC_FIELD_PAIRS_ENC_SRC_IP6_BE),
+		       &encap->src_ip6, sizeof(encap->src_ip6));
+		memset(MCDI_STRUCT_PTR(match_crit, MAE_ENC_FIELD_PAIRS_ENC_SRC_IP6_BE_MASK),
+		       0xff, sizeof(encap->src_ip6));
+		memcpy(MCDI_STRUCT_PTR(match_crit, MAE_ENC_FIELD_PAIRS_ENC_DST_IP6_BE),
+		       &encap->dst_ip6, sizeof(encap->dst_ip6));
+		memset(MCDI_STRUCT_PTR(match_crit, MAE_ENC_FIELD_PAIRS_ENC_DST_IP6_BE_MASK),
+		       0xff, sizeof(encap->dst_ip6));
+		MCDI_STRUCT_SET_WORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_ETHER_TYPE_BE,
+					htons(ETH_P_IPV6));
+	}
+#endif
+	MCDI_STRUCT_SET_WORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_ETHER_TYPE_BE_MASK,
+				~(__be16)0);
+	MCDI_STRUCT_SET_WORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_L4_DPORT_BE,
+				encap->udp_dport);
+	MCDI_STRUCT_SET_WORD_BE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_L4_DPORT_BE_MASK,
+				~(__be16)0);
+	MCDI_STRUCT_SET_BYTE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_IP_PROTO, IPPROTO_UDP);
+	MCDI_STRUCT_SET_BYTE(match_crit, MAE_ENC_FIELD_PAIRS_ENC_IP_PROTO_MASK, ~0);
+	rc = efx_mcdi_rpc(efx, MC_CMD_MAE_OUTER_RULE_INSERT, inbuf,
+			  sizeof(inbuf), outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+	if (outlen < sizeof(outbuf))
+		return -EIO;
+	encap->fw_id = MCDI_DWORD(outbuf, MAE_OUTER_RULE_INSERT_OUT_OR_ID);
+	return 0;
+}
+
+int efx_mae_unregister_encap_match(struct efx_nic *efx,
+				   struct efx_tc_encap_match *encap)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_OUTER_RULE_REMOVE_OUT_LEN(1));
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_MAE_OUTER_RULE_REMOVE_IN_LEN(1));
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, MAE_OUTER_RULE_REMOVE_IN_OR_ID, encap->fw_id);
+	rc = efx_mcdi_rpc(efx, MC_CMD_MAE_OUTER_RULE_REMOVE, inbuf,
+			  sizeof(inbuf), outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+	if (outlen < sizeof(outbuf))
+		return -EIO;
+	/* FW freed a different ID than we asked for, should also never happen.
+	 * Warn because it means we've now got a different idea to the FW of
+	 * what encap_mds exist, which could cause mayhem later.
+	 */
+	if (WARN_ON(MCDI_DWORD(outbuf, MAE_OUTER_RULE_REMOVE_OUT_REMOVED_OR_ID) != encap->fw_id))
+		return -EIO;
+	/* We're probably about to free @encap, but let's just make sure its
+	 * fw_id is blatted so that it won't look valid if it leaks out.
+	 */
+	encap->fw_id = MC_CMD_MAE_OUTER_RULE_INSERT_OUT_OUTER_RULE_ID_NULL;
+	return 0;
+}
+
 static int efx_mae_populate_match_criteria(MCDI_DECLARE_STRUCT_PTR(match_crit),
 					   const struct efx_tc_match *match)
 {
diff --git a/drivers/net/ethernet/sfc/mae.h b/drivers/net/ethernet/sfc/mae.h
index 2ccbc62d79b9..79d315a129ab 100644
--- a/drivers/net/ethernet/sfc/mae.h
+++ b/drivers/net/ethernet/sfc/mae.h
@@ -94,6 +94,11 @@ int efx_mae_alloc_action_set_list(struct efx_nic *efx,
 int efx_mae_free_action_set_list(struct efx_nic *efx,
 				 struct efx_tc_action_set_list *acts);
 
+int efx_mae_register_encap_match(struct efx_nic *efx,
+				 struct efx_tc_encap_match *encap);
+int efx_mae_unregister_encap_match(struct efx_nic *efx,
+				   struct efx_tc_encap_match *encap);
+
 int efx_mae_insert_rule(struct efx_nic *efx, const struct efx_tc_match *match,
 			u32 prio, u32 acts_id, u32 *id);
 int efx_mae_delete_rule(struct efx_nic *efx, u32 id);
diff --git a/drivers/net/ethernet/sfc/tc.h b/drivers/net/ethernet/sfc/tc.h
index c1485679507c..fff73960698a 100644
--- a/drivers/net/ethernet/sfc/tc.h
+++ b/drivers/net/ethernet/sfc/tc.h
@@ -70,6 +70,7 @@ struct efx_tc_encap_match {
 	__be32 src_ip, dst_ip;
 	struct in6_addr src_ip6, dst_ip6;
 	__be16 udp_dport;
+	enum efx_encap_type tun_type;
 	u32 fw_id; /* index of this entry in firmware encap match table */
 };
 
-- 
cgit v1.2.3-70-g09d2


From 746224cdef013e2a7a4b83fa52881c5482f0c47d Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Mon, 27 Mar 2023 11:36:07 +0100
Subject: sfc: add code to register and unregister encap matches

Add a hashtable to detect duplicate and conflicting matches.  If match
 is not a duplicate, call MAE functions to add/remove it from OR table.
Calling code not added yet, so mark the new functions as unused.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/tc.c | 163 ++++++++++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/sfc/tc.h |  11 +++
 2 files changed, 174 insertions(+)

diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c
index 21eb79b20978..69d3ee5d23d1 100644
--- a/drivers/net/ethernet/sfc/tc.c
+++ b/drivers/net/ethernet/sfc/tc.c
@@ -57,6 +57,12 @@ static s64 efx_tc_flower_external_mport(struct efx_nic *efx, struct efx_rep *efv
 	return mport;
 }
 
+static const struct rhashtable_params efx_tc_encap_match_ht_params = {
+	.key_len	= offsetof(struct efx_tc_encap_match, linkage),
+	.key_offset	= 0,
+	.head_offset	= offsetof(struct efx_tc_encap_match, linkage),
+};
+
 static const struct rhashtable_params efx_tc_match_action_ht_params = {
 	.key_len	= sizeof(unsigned long),
 	.key_offset	= offsetof(struct efx_tc_flow_rule, cookie),
@@ -340,6 +346,144 @@ static int efx_tc_flower_parse_match(struct efx_nic *efx,
 	return 0;
 }
 
+__always_unused
+static int efx_tc_flower_record_encap_match(struct efx_nic *efx,
+					    struct efx_tc_match *match,
+					    enum efx_encap_type type,
+					    struct netlink_ext_ack *extack)
+{
+	struct efx_tc_encap_match *encap, *old;
+	bool ipv6 = false;
+	int rc;
+
+	/* We require that the socket-defining fields (IP addrs and UDP dest
+	 * port) are present and exact-match.  Other fields are currently not
+	 * allowed.  This meets what OVS will ask for, and means that we don't
+	 * need to handle difficult checks for overlapping matches as could
+	 * come up if we allowed masks or varying sets of match fields.
+	 */
+	if (match->mask.enc_dst_ip | match->mask.enc_src_ip) {
+		if (!IS_ALL_ONES(match->mask.enc_dst_ip)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Egress encap match is not exact on dst IP address");
+			return -EOPNOTSUPP;
+		}
+		if (!IS_ALL_ONES(match->mask.enc_src_ip)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Egress encap match is not exact on src IP address");
+			return -EOPNOTSUPP;
+		}
+#ifdef CONFIG_IPV6
+		if (!ipv6_addr_any(&match->mask.enc_dst_ip6) ||
+		    !ipv6_addr_any(&match->mask.enc_src_ip6)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Egress encap match on both IPv4 and IPv6, don't understand");
+			return -EOPNOTSUPP;
+		}
+	} else {
+		ipv6 = true;
+		if (!efx_ipv6_addr_all_ones(&match->mask.enc_dst_ip6)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Egress encap match is not exact on dst IP address");
+			return -EOPNOTSUPP;
+		}
+		if (!efx_ipv6_addr_all_ones(&match->mask.enc_src_ip6)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Egress encap match is not exact on src IP address");
+			return -EOPNOTSUPP;
+		}
+#endif
+	}
+	if (!IS_ALL_ONES(match->mask.enc_dport)) {
+		NL_SET_ERR_MSG_MOD(extack, "Egress encap match is not exact on dst UDP port");
+		return -EOPNOTSUPP;
+	}
+	if (match->mask.enc_sport) {
+		NL_SET_ERR_MSG_MOD(extack, "Egress encap match on src UDP port not supported");
+		return -EOPNOTSUPP;
+	}
+	if (match->mask.enc_ip_tos) {
+		NL_SET_ERR_MSG_MOD(extack, "Egress encap match on IP ToS not supported");
+		return -EOPNOTSUPP;
+	}
+	if (match->mask.enc_ip_ttl) {
+		NL_SET_ERR_MSG_MOD(extack, "Egress encap match on IP TTL not supported");
+		return -EOPNOTSUPP;
+	}
+
+	rc = efx_mae_check_encap_match_caps(efx, ipv6, extack);
+	if (rc) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "MAE hw reports no support for IPv%d encap matches",
+				       ipv6 ? 6 : 4);
+		return -EOPNOTSUPP;
+	}
+
+	encap = kzalloc(sizeof(*encap), GFP_USER);
+	if (!encap)
+		return -ENOMEM;
+	encap->src_ip = match->value.enc_src_ip;
+	encap->dst_ip = match->value.enc_dst_ip;
+#ifdef CONFIG_IPV6
+	encap->src_ip6 = match->value.enc_src_ip6;
+	encap->dst_ip6 = match->value.enc_dst_ip6;
+#endif
+	encap->udp_dport = match->value.enc_dport;
+	encap->tun_type = type;
+	old = rhashtable_lookup_get_insert_fast(&efx->tc->encap_match_ht,
+						&encap->linkage,
+						efx_tc_encap_match_ht_params);
+	if (old) {
+		/* don't need our new entry */
+		kfree(encap);
+		if (old->tun_type != type) {
+			NL_SET_ERR_MSG_FMT_MOD(extack,
+					       "Egress encap match with conflicting tun_type %u != %u",
+					       old->tun_type, type);
+			return -EEXIST;
+		}
+		if (!refcount_inc_not_zero(&old->ref))
+			return -EAGAIN;
+		/* existing entry found */
+		encap = old;
+	} else {
+		rc = efx_mae_register_encap_match(efx, encap);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to record egress encap match in HW");
+			goto fail;
+		}
+		refcount_set(&encap->ref, 1);
+	}
+	match->encap = encap;
+	return 0;
+fail:
+	rhashtable_remove_fast(&efx->tc->encap_match_ht, &encap->linkage,
+			       efx_tc_encap_match_ht_params);
+	kfree(encap);
+	return rc;
+}
+
+__always_unused
+static void efx_tc_flower_release_encap_match(struct efx_nic *efx,
+					      struct efx_tc_encap_match *encap)
+{
+	int rc;
+
+	if (!refcount_dec_and_test(&encap->ref))
+		return; /* still in use */
+
+	rc = efx_mae_unregister_encap_match(efx, encap);
+	if (rc)
+		/* Display message but carry on and remove entry from our
+		 * SW tables, because there's not much we can do about it.
+		 */
+		netif_err(efx, drv, efx->net_dev,
+			  "Failed to release encap match %#x, rc %d\n",
+			  encap->fw_id, rc);
+	rhashtable_remove_fast(&efx->tc->encap_match_ht, &encap->linkage,
+			       efx_tc_encap_match_ht_params);
+	kfree(encap);
+}
+
 /* For details of action order constraints refer to SF-123102-TC-1§12.6.1 */
 enum efx_tc_action_order {
 	EFX_TC_AO_VLAN_POP,
@@ -974,6 +1118,18 @@ void efx_fini_tc(struct efx_nic *efx)
 	efx->tc->up = false;
 }
 
+/* At teardown time, all TC filter rules (and thus all resources they created)
+ * should already have been removed.  If we find any in our hashtables, make a
+ * cursory attempt to clean up the software side.
+ */
+static void efx_tc_encap_match_free(void *ptr, void *__unused)
+{
+	struct efx_tc_encap_match *encap = ptr;
+
+	WARN_ON(refcount_read(&encap->ref));
+	kfree(encap);
+}
+
 int efx_init_struct_tc(struct efx_nic *efx)
 {
 	int rc;
@@ -996,6 +1152,9 @@ int efx_init_struct_tc(struct efx_nic *efx)
 	rc = efx_tc_init_counters(efx);
 	if (rc < 0)
 		goto fail_counters;
+	rc = rhashtable_init(&efx->tc->encap_match_ht, &efx_tc_encap_match_ht_params);
+	if (rc < 0)
+		goto fail_encap_match_ht;
 	rc = rhashtable_init(&efx->tc->match_action_ht, &efx_tc_match_action_ht_params);
 	if (rc < 0)
 		goto fail_match_action_ht;
@@ -1008,6 +1167,8 @@ int efx_init_struct_tc(struct efx_nic *efx)
 	efx->extra_channel_type[EFX_EXTRA_CHANNEL_TC] = &efx_tc_channel_type;
 	return 0;
 fail_match_action_ht:
+	rhashtable_destroy(&efx->tc->encap_match_ht);
+fail_encap_match_ht:
 	efx_tc_destroy_counters(efx);
 fail_counters:
 	mutex_destroy(&efx->tc->mutex);
@@ -1030,6 +1191,8 @@ void efx_fini_struct_tc(struct efx_nic *efx)
 			     MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL);
 	rhashtable_free_and_destroy(&efx->tc->match_action_ht, efx_tc_flow_free,
 				    efx);
+	rhashtable_free_and_destroy(&efx->tc->encap_match_ht,
+				    efx_tc_encap_match_free, NULL);
 	efx_tc_fini_counters(efx);
 	mutex_unlock(&efx->tc->mutex);
 	mutex_destroy(&efx->tc->mutex);
diff --git a/drivers/net/ethernet/sfc/tc.h b/drivers/net/ethernet/sfc/tc.h
index fff73960698a..7d06c5031b56 100644
--- a/drivers/net/ethernet/sfc/tc.h
+++ b/drivers/net/ethernet/sfc/tc.h
@@ -18,6 +18,13 @@
 
 #define IS_ALL_ONES(v)	(!(typeof (v))~(v))
 
+#ifdef CONFIG_IPV6
+static inline bool efx_ipv6_addr_all_ones(struct in6_addr *addr)
+{
+	return !memchr_inv(addr, 0xff, sizeof(*addr));
+}
+#endif
+
 struct efx_tc_action_set {
 	u16 vlan_push:2;
 	u16 vlan_pop:2;
@@ -70,7 +77,9 @@ struct efx_tc_encap_match {
 	__be32 src_ip, dst_ip;
 	struct in6_addr src_ip6, dst_ip6;
 	__be16 udp_dport;
+	struct rhash_head linkage;
 	enum efx_encap_type tun_type;
+	refcount_t ref;
 	u32 fw_id; /* index of this entry in firmware encap match table */
 };
 
@@ -107,6 +116,7 @@ enum efx_tc_rule_prios {
  * @mutex: Used to serialise operations on TC hashtables
  * @counter_ht: Hashtable of TC counters (FW IDs and counter values)
  * @counter_id_ht: Hashtable mapping TC counter cookies to counters
+ * @encap_match_ht: Hashtable of TC encap matches
  * @match_action_ht: Hashtable of TC match-action rules
  * @reps_mport_id: MAE port allocated for representor RX
  * @reps_filter_uc: VNIC filter for representor unicast RX (promisc)
@@ -130,6 +140,7 @@ struct efx_tc_state {
 	struct mutex mutex;
 	struct rhashtable counter_ht;
 	struct rhashtable counter_id_ht;
+	struct rhashtable encap_match_ht;
 	struct rhashtable match_action_ht;
 	u32 reps_mport_id, reps_mport_vport_id;
 	s32 reps_filter_uc, reps_filter_mc;
-- 
cgit v1.2.3-70-g09d2


From 17654d84b47c69ca788fea7afbdb275504e4c6dc Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Mon, 27 Mar 2023 11:36:08 +0100
Subject: sfc: add offloading of 'foreign' TC (decap) rules

A 'foreign' rule is one for which the net_dev is not the sfc netdevice
 or any of its representors.  The driver registers indirect flow blocks
 for tunnel netdevs so that it can offload decap rules.  For example:

    tc filter add dev vxlan0 parent ffff: protocol ipv4 flower \
        enc_src_ip 10.1.0.2 enc_dst_ip 10.1.0.1 \
        enc_key_id 1000 enc_dst_port 4789 \
        action tunnel_key unset \
        action mirred egress redirect dev $REPRESENTOR

When notified of a rule like this, register an encap match on the IP
 and dport tuple (creating an Outer Rule table entry) and insert an MAE
 action rule to perform the decapsulation and deliver to the representee.

Moved efx_tc_delete_rule() below efx_tc_flower_release_encap_match() to
 avoid the need for a forward declaration.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/sfc/mae.c |  25 ++-
 drivers/net/ethernet/sfc/mae.h |   3 +
 drivers/net/ethernet/sfc/tc.c  | 354 +++++++++++++++++++++++++++++++++++++++--
 drivers/net/ethernet/sfc/tc.h  |   1 +
 4 files changed, 366 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/sfc/mae.c b/drivers/net/ethernet/sfc/mae.c
index 92f1383ee4b9..49706a7b94bf 100644
--- a/drivers/net/ethernet/sfc/mae.c
+++ b/drivers/net/ethernet/sfc/mae.c
@@ -241,6 +241,7 @@ static int efx_mae_get_basic_caps(struct efx_nic *efx, struct mae_caps *caps)
 	if (outlen < sizeof(outbuf))
 		return -EIO;
 	caps->match_field_count = MCDI_DWORD(outbuf, MAE_GET_CAPS_OUT_MATCH_FIELD_COUNT);
+	caps->encap_types = MCDI_DWORD(outbuf, MAE_GET_CAPS_OUT_ENCAP_TYPES_SUPPORTED);
 	caps->action_prios = MCDI_DWORD(outbuf, MAE_GET_CAPS_OUT_ACTION_PRIOS);
 	return 0;
 }
@@ -507,6 +508,25 @@ int efx_mae_check_encap_match_caps(struct efx_nic *efx, bool ipv6,
 }
 #undef CHECK
 
+int efx_mae_check_encap_type_supported(struct efx_nic *efx, enum efx_encap_type typ)
+{
+	unsigned int bit;
+
+	switch (typ & EFX_ENCAP_TYPES_MASK) {
+	case EFX_ENCAP_TYPE_VXLAN:
+		bit = MC_CMD_MAE_GET_CAPS_OUT_ENCAP_TYPE_VXLAN_LBN;
+		break;
+	case EFX_ENCAP_TYPE_GENEVE:
+		bit = MC_CMD_MAE_GET_CAPS_OUT_ENCAP_TYPE_GENEVE_LBN;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+	if (efx->tc->caps->encap_types & BIT(bit))
+		return 0;
+	return -EOPNOTSUPP;
+}
+
 int efx_mae_allocate_counter(struct efx_nic *efx, struct efx_tc_counter *cnt)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_COUNTER_ALLOC_OUT_LEN(1));
@@ -766,9 +786,10 @@ int efx_mae_alloc_action_set(struct efx_nic *efx, struct efx_tc_action_set *act)
 	size_t outlen;
 	int rc;
 
-	MCDI_POPULATE_DWORD_2(inbuf, MAE_ACTION_SET_ALLOC_IN_FLAGS,
+	MCDI_POPULATE_DWORD_3(inbuf, MAE_ACTION_SET_ALLOC_IN_FLAGS,
 			      MAE_ACTION_SET_ALLOC_IN_VLAN_PUSH, act->vlan_push,
-			      MAE_ACTION_SET_ALLOC_IN_VLAN_POP, act->vlan_pop);
+			      MAE_ACTION_SET_ALLOC_IN_VLAN_POP, act->vlan_pop,
+			      MAE_ACTION_SET_ALLOC_IN_DECAP, act->decap);
 
 	MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_SRC_MAC_ID,
 		       MC_CMD_MAE_MAC_ADDR_ALLOC_OUT_MAC_ID_NULL);
diff --git a/drivers/net/ethernet/sfc/mae.h b/drivers/net/ethernet/sfc/mae.h
index 79d315a129ab..9226219491a0 100644
--- a/drivers/net/ethernet/sfc/mae.h
+++ b/drivers/net/ethernet/sfc/mae.h
@@ -70,6 +70,7 @@ void efx_mae_counters_grant_credits(struct work_struct *work);
 
 struct mae_caps {
 	u32 match_field_count;
+	u32 encap_types;
 	u32 action_prios;
 	u8 action_rule_fields[MAE_NUM_FIELDS];
 	u8 outer_rule_fields[MAE_NUM_FIELDS];
@@ -82,6 +83,8 @@ int efx_mae_match_check_caps(struct efx_nic *efx,
 			     struct netlink_ext_ack *extack);
 int efx_mae_check_encap_match_caps(struct efx_nic *efx, bool ipv6,
 				   struct netlink_ext_ack *extack);
+int efx_mae_check_encap_type_supported(struct efx_nic *efx,
+				       enum efx_encap_type typ);
 
 int efx_mae_allocate_counter(struct efx_nic *efx, struct efx_tc_counter *cnt);
 int efx_mae_free_counter(struct efx_nic *efx, struct efx_tc_counter *cnt);
diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c
index 69d3ee5d23d1..0327639a628a 100644
--- a/drivers/net/ethernet/sfc/tc.c
+++ b/drivers/net/ethernet/sfc/tc.c
@@ -10,12 +10,24 @@
  */
 
 #include <net/pkt_cls.h>
+#include <net/vxlan.h>
+#include <net/geneve.h>
 #include "tc.h"
 #include "tc_bindings.h"
 #include "mae.h"
 #include "ef100_rep.h"
 #include "efx.h"
 
+static enum efx_encap_type efx_tc_indr_netdev_type(struct net_device *net_dev)
+{
+	if (netif_is_vxlan(net_dev))
+		return EFX_ENCAP_TYPE_VXLAN;
+	if (netif_is_geneve(net_dev))
+		return EFX_ENCAP_TYPE_GENEVE;
+
+	return EFX_ENCAP_TYPE_NONE;
+}
+
 #define EFX_EFV_PF	NULL
 /* Look up the representor information (efv) for a device.
  * May return NULL for the PF (us), or an error pointer for a device that
@@ -43,6 +55,20 @@ static struct efx_rep *efx_tc_flower_lookup_efv(struct efx_nic *efx,
 	return efv;
 }
 
+/* Convert a driver-internal vport ID into an internal device (PF or VF) */
+static s64 efx_tc_flower_internal_mport(struct efx_nic *efx, struct efx_rep *efv)
+{
+	u32 mport;
+
+	if (IS_ERR(efv))
+		return PTR_ERR(efv);
+	if (!efv) /* device is PF (us) */
+		efx_mae_mport_uplink(efx, &mport);
+	else /* device is repr */
+		efx_mae_mport_mport(efx, efv->mport, &mport);
+	return mport;
+}
+
 /* Convert a driver-internal vport ID into an external device (wire or VF) */
 static s64 efx_tc_flower_external_mport(struct efx_nic *efx, struct efx_rep *efv)
 {
@@ -106,15 +132,6 @@ static void efx_tc_free_action_set_list(struct efx_nic *efx,
 	/* Don't kfree, as acts is embedded inside a struct efx_tc_flow_rule */
 }
 
-static void efx_tc_delete_rule(struct efx_nic *efx, struct efx_tc_flow_rule *rule)
-{
-	efx_mae_delete_rule(efx, rule->fw_id);
-
-	/* Release entries in subsidiary tables */
-	efx_tc_free_action_set_list(efx, &rule->acts, true);
-	rule->fw_id = MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL;
-}
-
 static void efx_tc_flow_free(void *ptr, void *arg)
 {
 	struct efx_tc_flow_rule *rule = ptr;
@@ -346,7 +363,6 @@ static int efx_tc_flower_parse_match(struct efx_nic *efx,
 	return 0;
 }
 
-__always_unused
 static int efx_tc_flower_record_encap_match(struct efx_nic *efx,
 					    struct efx_tc_match *match,
 					    enum efx_encap_type type,
@@ -462,7 +478,6 @@ fail:
 	return rc;
 }
 
-__always_unused
 static void efx_tc_flower_release_encap_match(struct efx_nic *efx,
 					      struct efx_tc_encap_match *encap)
 {
@@ -484,8 +499,35 @@ static void efx_tc_flower_release_encap_match(struct efx_nic *efx,
 	kfree(encap);
 }
 
+static void efx_tc_delete_rule(struct efx_nic *efx, struct efx_tc_flow_rule *rule)
+{
+	efx_mae_delete_rule(efx, rule->fw_id);
+
+	/* Release entries in subsidiary tables */
+	efx_tc_free_action_set_list(efx, &rule->acts, true);
+	if (rule->match.encap)
+		efx_tc_flower_release_encap_match(efx, rule->match.encap);
+	rule->fw_id = MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL;
+}
+
+static const char *efx_tc_encap_type_name(enum efx_encap_type typ)
+{
+	switch (typ) {
+	case EFX_ENCAP_TYPE_NONE:
+		return "none";
+	case EFX_ENCAP_TYPE_VXLAN:
+		return "vxlan";
+	case EFX_ENCAP_TYPE_GENEVE:
+		return "geneve";
+	default:
+		pr_warn_once("Unknown efx_encap_type %d encountered\n", typ);
+		return "unknown";
+	}
+}
+
 /* For details of action order constraints refer to SF-123102-TC-1§12.6.1 */
 enum efx_tc_action_order {
+	EFX_TC_AO_DECAP,
 	EFX_TC_AO_VLAN_POP,
 	EFX_TC_AO_VLAN_PUSH,
 	EFX_TC_AO_COUNT,
@@ -496,6 +538,10 @@ static bool efx_tc_flower_action_order_ok(const struct efx_tc_action_set *act,
 					  enum efx_tc_action_order new)
 {
 	switch (new) {
+	case EFX_TC_AO_DECAP:
+		if (act->decap)
+			return false;
+		fallthrough;
 	case EFX_TC_AO_VLAN_POP:
 		if (act->vlan_pop >= 2)
 			return false;
@@ -523,6 +569,286 @@ static bool efx_tc_flower_action_order_ok(const struct efx_tc_action_set *act,
 	}
 }
 
+static int efx_tc_flower_replace_foreign(struct efx_nic *efx,
+					 struct net_device *net_dev,
+					 struct flow_cls_offload *tc)
+{
+	struct flow_rule *fr = flow_cls_offload_flow_rule(tc);
+	struct netlink_ext_ack *extack = tc->common.extack;
+	struct efx_tc_flow_rule *rule = NULL, *old = NULL;
+	struct efx_tc_action_set *act = NULL;
+	bool found = false, uplinked = false;
+	const struct flow_action_entry *fa;
+	struct efx_tc_match match;
+	struct efx_rep *to_efv;
+	s64 rc;
+	int i;
+
+	/* Parse match */
+	memset(&match, 0, sizeof(match));
+	rc = efx_tc_flower_parse_match(efx, fr, &match, NULL);
+	if (rc)
+		return rc;
+	/* The rule as given to us doesn't specify a source netdevice.
+	 * But, determining whether packets from a VF should match it is
+	 * complicated, so leave those to the software slowpath: qualify
+	 * the filter with source m-port == wire.
+	 */
+	rc = efx_tc_flower_external_mport(efx, EFX_EFV_PF);
+	if (rc < 0) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to identify ingress m-port for foreign filter");
+		return rc;
+	}
+	match.value.ingress_port = rc;
+	match.mask.ingress_port = ~0;
+
+	if (tc->common.chain_index) {
+		NL_SET_ERR_MSG_MOD(extack, "No support for nonzero chain_index");
+		return -EOPNOTSUPP;
+	}
+	match.mask.recirc_id = 0xff;
+
+	flow_action_for_each(i, fa, &fr->action) {
+		switch (fa->id) {
+		case FLOW_ACTION_REDIRECT:
+		case FLOW_ACTION_MIRRED: /* mirred means mirror here */
+			to_efv = efx_tc_flower_lookup_efv(efx, fa->dev);
+			if (IS_ERR(to_efv))
+				continue;
+			found = true;
+			break;
+		default:
+			break;
+		}
+	}
+	if (!found) { /* We don't care. */
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Ignoring foreign filter that doesn't egdev us\n");
+		rc = -EOPNOTSUPP;
+		goto release;
+	}
+
+	rc = efx_mae_match_check_caps(efx, &match.mask, NULL);
+	if (rc)
+		goto release;
+
+	if (efx_tc_match_is_encap(&match.mask)) {
+		enum efx_encap_type type;
+
+		type = efx_tc_indr_netdev_type(net_dev);
+		if (type == EFX_ENCAP_TYPE_NONE) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Egress encap match on unsupported tunnel device");
+			rc = -EOPNOTSUPP;
+			goto release;
+		}
+
+		rc = efx_mae_check_encap_type_supported(efx, type);
+		if (rc) {
+			NL_SET_ERR_MSG_FMT_MOD(extack,
+					       "Firmware reports no support for %s encap match",
+					       efx_tc_encap_type_name(type));
+			goto release;
+		}
+
+		rc = efx_tc_flower_record_encap_match(efx, &match, type,
+						      extack);
+		if (rc)
+			goto release;
+	} else {
+		/* This is not a tunnel decap rule, ignore it */
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Ignoring foreign filter without encap match\n");
+		rc = -EOPNOTSUPP;
+		goto release;
+	}
+
+	rule = kzalloc(sizeof(*rule), GFP_USER);
+	if (!rule) {
+		rc = -ENOMEM;
+		goto release;
+	}
+	INIT_LIST_HEAD(&rule->acts.list);
+	rule->cookie = tc->cookie;
+	old = rhashtable_lookup_get_insert_fast(&efx->tc->match_action_ht,
+						&rule->linkage,
+						efx_tc_match_action_ht_params);
+	if (old) {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Ignoring already-offloaded rule (cookie %lx)\n",
+			  tc->cookie);
+		rc = -EEXIST;
+		goto release;
+	}
+
+	act = kzalloc(sizeof(*act), GFP_USER);
+	if (!act) {
+		rc = -ENOMEM;
+		goto release;
+	}
+
+	/* Parse actions.  For foreign rules we only support decap & redirect.
+	 * See corresponding code in efx_tc_flower_replace() for theory of
+	 * operation & how 'act' cursor is used.
+	 */
+	flow_action_for_each(i, fa, &fr->action) {
+		struct efx_tc_action_set save;
+
+		switch (fa->id) {
+		case FLOW_ACTION_REDIRECT:
+		case FLOW_ACTION_MIRRED:
+			/* See corresponding code in efx_tc_flower_replace() for
+			 * long explanations of what's going on here.
+			 */
+			save = *act;
+			if (fa->hw_stats) {
+				struct efx_tc_counter_index *ctr;
+
+				if (!(fa->hw_stats & FLOW_ACTION_HW_STATS_DELAYED)) {
+					NL_SET_ERR_MSG_FMT_MOD(extack,
+							       "hw_stats_type %u not supported (only 'delayed')",
+							       fa->hw_stats);
+					rc = -EOPNOTSUPP;
+					goto release;
+				}
+				if (!efx_tc_flower_action_order_ok(act, EFX_TC_AO_COUNT)) {
+					rc = -EOPNOTSUPP;
+					goto release;
+				}
+
+				ctr = efx_tc_flower_get_counter_index(efx,
+								      tc->cookie,
+								      EFX_TC_COUNTER_TYPE_AR);
+				if (IS_ERR(ctr)) {
+					rc = PTR_ERR(ctr);
+					NL_SET_ERR_MSG_MOD(extack, "Failed to obtain a counter");
+					goto release;
+				}
+				act->count = ctr;
+			}
+
+			if (!efx_tc_flower_action_order_ok(act, EFX_TC_AO_DELIVER)) {
+				/* can't happen */
+				rc = -EOPNOTSUPP;
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Deliver action violates action order (can't happen)");
+				goto release;
+			}
+			to_efv = efx_tc_flower_lookup_efv(efx, fa->dev);
+			/* PF implies egdev is us, in which case we really
+			 * want to deliver to the uplink (because this is an
+			 * ingress filter).  If we don't recognise the egdev
+			 * at all, then we'd better trap so SW can handle it.
+			 */
+			if (IS_ERR(to_efv))
+				to_efv = EFX_EFV_PF;
+			if (to_efv == EFX_EFV_PF) {
+				if (uplinked)
+					break;
+				uplinked = true;
+			}
+			rc = efx_tc_flower_internal_mport(efx, to_efv);
+			if (rc < 0) {
+				NL_SET_ERR_MSG_MOD(extack, "Failed to identify egress m-port");
+				goto release;
+			}
+			act->dest_mport = rc;
+			act->deliver = 1;
+			rc = efx_mae_alloc_action_set(efx, act);
+			if (rc) {
+				NL_SET_ERR_MSG_MOD(extack,
+						   "Failed to write action set to hw (mirred)");
+				goto release;
+			}
+			list_add_tail(&act->list, &rule->acts.list);
+			act = NULL;
+			if (fa->id == FLOW_ACTION_REDIRECT)
+				break; /* end of the line */
+			/* Mirror, so continue on with saved act */
+			act = kzalloc(sizeof(*act), GFP_USER);
+			if (!act) {
+				rc = -ENOMEM;
+				goto release;
+			}
+			*act = save;
+			break;
+		case FLOW_ACTION_TUNNEL_DECAP:
+			if (!efx_tc_flower_action_order_ok(act, EFX_TC_AO_DECAP)) {
+				rc = -EINVAL;
+				NL_SET_ERR_MSG_MOD(extack, "Decap action violates action order");
+				goto release;
+			}
+			act->decap = 1;
+			/* If we previously delivered/trapped to uplink, now
+			 * that we've decapped we'll want another copy if we
+			 * try to deliver/trap to uplink again.
+			 */
+			uplinked = false;
+			break;
+		default:
+			NL_SET_ERR_MSG_FMT_MOD(extack, "Unhandled action %u",
+					       fa->id);
+			rc = -EOPNOTSUPP;
+			goto release;
+		}
+	}
+
+	if (act) {
+		if (!uplinked) {
+			/* Not shot/redirected, so deliver to default dest (which is
+			 * the uplink, as this is an ingress filter)
+			 */
+			efx_mae_mport_uplink(efx, &act->dest_mport);
+			act->deliver = 1;
+		}
+		rc = efx_mae_alloc_action_set(efx, act);
+		if (rc) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to write action set to hw (deliver)");
+			goto release;
+		}
+		list_add_tail(&act->list, &rule->acts.list);
+		act = NULL; /* Prevent double-free in error path */
+	}
+
+	rule->match = match;
+
+	netif_dbg(efx, drv, efx->net_dev,
+		  "Successfully parsed foreign filter (cookie %lx)\n",
+		  tc->cookie);
+
+	rc = efx_mae_alloc_action_set_list(efx, &rule->acts);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to write action set list to hw");
+		goto release;
+	}
+	rc = efx_mae_insert_rule(efx, &rule->match, EFX_TC_PRIO_TC,
+				 rule->acts.fw_id, &rule->fw_id);
+	if (rc) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to insert rule in hw");
+		goto release_acts;
+	}
+	return 0;
+
+release_acts:
+	efx_mae_free_action_set_list(efx, &rule->acts);
+release:
+	/* We failed to insert the rule, so free up any entries we created in
+	 * subsidiary tables.
+	 */
+	if (act)
+		efx_tc_free_action_set(efx, act, false);
+	if (rule) {
+		rhashtable_remove_fast(&efx->tc->match_action_ht,
+				       &rule->linkage,
+				       efx_tc_match_action_ht_params);
+		efx_tc_free_action_set_list(efx, &rule->acts, false);
+	}
+	kfree(rule);
+	if (match.encap)
+		efx_tc_flower_release_encap_match(efx, match.encap);
+	return rc;
+}
+
 static int efx_tc_flower_replace(struct efx_nic *efx,
 				 struct net_device *net_dev,
 				 struct flow_cls_offload *tc,
@@ -547,10 +873,8 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 
 	from_efv = efx_tc_flower_lookup_efv(efx, net_dev);
 	if (IS_ERR(from_efv)) {
-		/* Might be a tunnel decap rule from an indirect block.
-		 * Support for those not implemented yet.
-		 */
-		return -EOPNOTSUPP;
+		/* Not from our PF or representors, so probably a tunnel dev */
+		return efx_tc_flower_replace_foreign(efx, net_dev, tc);
 	}
 
 	if (efv != from_efv) {
diff --git a/drivers/net/ethernet/sfc/tc.h b/drivers/net/ethernet/sfc/tc.h
index 7d06c5031b56..04cced6a2d39 100644
--- a/drivers/net/ethernet/sfc/tc.h
+++ b/drivers/net/ethernet/sfc/tc.h
@@ -28,6 +28,7 @@ static inline bool efx_ipv6_addr_all_ones(struct in6_addr *addr)
 struct efx_tc_action_set {
 	u16 vlan_push:2;
 	u16 vlan_pop:2;
+	u16 decap:1;
 	u16 deliver:1;
 	__be16 vlan_tci[2]; /* TCIs for vlan_push */
 	__be16 vlan_proto[2]; /* Ethertypes for vlan_push */
-- 
cgit v1.2.3-70-g09d2


From 3b064f541be822dc095991c6dda20a75eb51db5e Mon Sep 17 00:00:00 2001
From: Hao Lan <lanhao@huawei.com>
Date: Mon, 27 Mar 2023 21:55:04 +0800
Subject: net: hns3: support wake on lan configuration and query

The HNS3 driver supports Wake-on-LAN, which can wake up
the server from power off state to power on state by magic
packet or magic security packet.

ChangeLog:
v1->v2:
Deleted the debugfs function that overlaps with the ethtool function
from suggestion of Andrew Lunn.

v2->v3:
Return the wol configuration stored in driver,
suggested by Alexander H Duyck.

v3->v4:
Add a helper to go from netdev to the local struct,
suggested by Simon Horman and Jakub Kicinski.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Hao Lan <lanhao@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/hisilicon/hns3/hnae3.h        |  12 ++
 .../hisilicon/hns3/hns3_common/hclge_comm_cmd.c    |   1 +
 .../hisilicon/hns3/hns3_common/hclge_comm_cmd.h    |   3 +
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c |   3 +
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.h    |   6 +
 drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c |  27 +++++
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h |  12 ++
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 130 +++++++++++++++++++++
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.h    |   8 ++
 9 files changed, 202 insertions(+)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index 40f4306449eb..9c9c72dc57e0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -100,6 +100,7 @@ enum HNAE3_DEV_CAP_BITS {
 	HNAE3_DEV_SUPPORT_CQ_B,
 	HNAE3_DEV_SUPPORT_FEC_STATS_B,
 	HNAE3_DEV_SUPPORT_LANE_NUM_B,
+	HNAE3_DEV_SUPPORT_WOL_B,
 };
 
 #define hnae3_ae_dev_fd_supported(ae_dev) \
@@ -168,6 +169,9 @@ enum HNAE3_DEV_CAP_BITS {
 #define hnae3_ae_dev_lane_num_supported(ae_dev) \
 	test_bit(HNAE3_DEV_SUPPORT_LANE_NUM_B, (ae_dev)->caps)
 
+#define hnae3_ae_dev_wol_supported(ae_dev) \
+	test_bit(HNAE3_DEV_SUPPORT_WOL_B, (ae_dev)->caps)
+
 enum HNAE3_PF_CAP_BITS {
 	HNAE3_PF_SUPPORT_VLAN_FLTR_MDF_B = 0,
 };
@@ -561,6 +565,10 @@ struct hnae3_ae_dev {
  *   Get phc info
  * clean_vf_config
  *   Clean residual vf info after disable sriov
+ * get_wol
+ *   Get wake on lan info
+ * set_wol
+ *   Config wake on lan
  */
 struct hnae3_ae_ops {
 	int (*init_ae_dev)(struct hnae3_ae_dev *ae_dev);
@@ -760,6 +768,10 @@ struct hnae3_ae_ops {
 	void (*clean_vf_config)(struct hnae3_ae_dev *ae_dev, int num_vfs);
 	int (*get_dscp_prio)(struct hnae3_handle *handle, u8 dscp,
 			     u8 *tc_map_mode, u8 *priority);
+	void (*get_wol)(struct hnae3_handle *handle,
+			struct ethtool_wolinfo *wol);
+	int (*set_wol)(struct hnae3_handle *handle,
+		       struct ethtool_wolinfo *wol);
 };
 
 struct hnae3_dcb_ops {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
index f671a63cecde..cbbab5b2b402 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
@@ -155,6 +155,7 @@ static const struct hclge_comm_caps_bit_map hclge_pf_cmd_caps[] = {
 	{HCLGE_COMM_CAP_FD_B, HNAE3_DEV_SUPPORT_FD_B},
 	{HCLGE_COMM_CAP_FEC_STATS_B, HNAE3_DEV_SUPPORT_FEC_STATS_B},
 	{HCLGE_COMM_CAP_LANE_NUM_B, HNAE3_DEV_SUPPORT_LANE_NUM_B},
+	{HCLGE_COMM_CAP_WOL_B, HNAE3_DEV_SUPPORT_WOL_B},
 };
 
 static const struct hclge_comm_caps_bit_map hclge_vf_cmd_caps[] = {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h
index b1f9383b418f..de72ecbfd5ad 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h
@@ -294,6 +294,8 @@ enum hclge_opcode_type {
 	HCLGE_PPP_CMD0_INT_CMD		= 0x2100,
 	HCLGE_PPP_CMD1_INT_CMD		= 0x2101,
 	HCLGE_MAC_ETHERTYPE_IDX_RD      = 0x2105,
+	HCLGE_OPC_WOL_GET_SUPPORTED_MODE	= 0x2201,
+	HCLGE_OPC_WOL_CFG		= 0x2202,
 	HCLGE_NCSI_INT_EN		= 0x2401,
 
 	/* ROH MAC commands */
@@ -345,6 +347,7 @@ enum HCLGE_COMM_CAP_BITS {
 	HCLGE_COMM_CAP_FD_B = 21,
 	HCLGE_COMM_CAP_FEC_STATS_B = 25,
 	HCLGE_COMM_CAP_LANE_NUM_B = 27,
+	HCLGE_COMM_CAP_WOL_B = 28,
 };
 
 enum HCLGE_COMM_API_CAP_BITS {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
index 66feb23f7b7b..4c3e90a1c4d0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
@@ -408,6 +408,9 @@ static struct hns3_dbg_cap_info hns3_dbg_cap[] = {
 	}, {
 		.name = "support lane num",
 		.cap_bit = HNAE3_DEV_SUPPORT_LANE_NUM_B,
+	}, {
+		.name = "support wake on lan",
+		.cap_bit = HNAE3_DEV_SUPPORT_WOL_B,
 	}
 };
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
index 294a14b4fdef..88af34bbee34 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h
@@ -695,6 +695,12 @@ static inline unsigned int hns3_page_order(struct hns3_enet_ring *ring)
 #define hns3_get_handle(ndev) \
 	(((struct hns3_nic_priv *)netdev_priv(ndev))->ae_handle)
 
+#define hns3_get_ae_dev(handle) \
+	(pci_get_drvdata((handle)->pdev))
+
+#define hns3_get_ops(handle) \
+	((handle)->ae_algo->ops)
+
 #define hns3_gl_usec_to_reg(int_gl) ((int_gl) >> 1)
 #define hns3_gl_round_down(int_gl) round_down(int_gl, 2)
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
index 55306fe8a540..51d1278b18f6 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
@@ -2063,6 +2063,31 @@ static int hns3_get_link_ext_state(struct net_device *netdev,
 	return -ENODATA;
 }
 
+static void hns3_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+
+	if (!hnae3_ae_dev_wol_supported(ae_dev))
+		return;
+
+	ops->get_wol(handle, wol);
+}
+
+static int hns3_set_wol(struct net_device *netdev,
+			struct ethtool_wolinfo *wol)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+	const struct hnae3_ae_ops *ops = hns3_get_ops(handle);
+	struct hnae3_ae_dev *ae_dev = hns3_get_ae_dev(handle);
+
+	if (!hnae3_ae_dev_wol_supported(ae_dev))
+		return -EOPNOTSUPP;
+
+	return ops->set_wol(handle, wol);
+}
+
 static const struct ethtool_ops hns3vf_ethtool_ops = {
 	.supported_coalesce_params = HNS3_ETHTOOL_COALESCE,
 	.supported_ring_params = HNS3_ETHTOOL_RING,
@@ -2139,6 +2164,8 @@ static const struct ethtool_ops hns3_ethtool_ops = {
 	.set_tunable = hns3_set_tunable,
 	.reset = hns3_set_reset,
 	.get_link_ext_state = hns3_get_link_ext_state,
+	.get_wol = hns3_get_wol,
+	.set_wol = hns3_set_wol,
 };
 
 void hns3_ethtool_set_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
index 43cada51d8cb..91c173f40701 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h
@@ -872,6 +872,18 @@ struct hclge_phy_reg_cmd {
 	u8 rsv1[18];
 };
 
+struct hclge_wol_cfg_cmd {
+	__le32 wake_on_lan_mode;
+	u8 sopass[SOPASS_MAX];
+	u8 sopass_size;
+	u8 rsv[13];
+};
+
+struct hclge_query_wol_supported_cmd {
+	__le32 supported_wake_mode;
+	u8 rsv[20];
+};
+
 struct hclge_hw;
 int hclge_cmd_send(struct hclge_hw *hw, struct hclge_desc *desc, int num);
 enum hclge_comm_cmd_status hclge_cmd_mdio_write(struct hclge_hw *hw,
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index c3851a6e10c0..4fb5406c1951 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -11522,6 +11522,124 @@ static void hclge_uninit_rxd_adv_layout(struct hclge_dev *hdev)
 		hclge_write_dev(&hdev->hw, HCLGE_RXD_ADV_LAYOUT_EN_REG, 0);
 }
 
+static struct hclge_wol_info *hclge_get_wol_info(struct hnae3_handle *handle)
+{
+	struct hclge_vport *vport = hclge_get_vport(handle);
+
+	return &vport->back->hw.mac.wol;
+}
+
+static int hclge_get_wol_supported_mode(struct hclge_dev *hdev,
+					u32 *wol_supported)
+{
+	struct hclge_query_wol_supported_cmd *wol_supported_cmd;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_WOL_GET_SUPPORTED_MODE,
+				   true);
+	wol_supported_cmd = (struct hclge_query_wol_supported_cmd *)desc.data;
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret) {
+		dev_err(&hdev->pdev->dev,
+			"failed to query wol supported, ret = %d\n", ret);
+		return ret;
+	}
+
+	*wol_supported = le32_to_cpu(wol_supported_cmd->supported_wake_mode);
+
+	return 0;
+}
+
+static int hclge_set_wol_cfg(struct hclge_dev *hdev,
+			     struct hclge_wol_info *wol_info)
+{
+	struct hclge_wol_cfg_cmd *wol_cfg_cmd;
+	struct hclge_desc desc;
+	int ret;
+
+	hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_WOL_CFG, false);
+	wol_cfg_cmd = (struct hclge_wol_cfg_cmd *)desc.data;
+	wol_cfg_cmd->wake_on_lan_mode = cpu_to_le32(wol_info->wol_current_mode);
+	wol_cfg_cmd->sopass_size = wol_info->wol_sopass_size;
+	memcpy(wol_cfg_cmd->sopass, wol_info->wol_sopass, SOPASS_MAX);
+
+	ret = hclge_cmd_send(&hdev->hw, &desc, 1);
+	if (ret)
+		dev_err(&hdev->pdev->dev,
+			"failed to set wol config, ret = %d\n", ret);
+
+	return ret;
+}
+
+static int hclge_update_wol(struct hclge_dev *hdev)
+{
+	struct hclge_wol_info *wol_info = &hdev->hw.mac.wol;
+
+	if (!hnae3_ae_dev_wol_supported(hdev->ae_dev))
+		return 0;
+
+	return hclge_set_wol_cfg(hdev, wol_info);
+}
+
+static int hclge_init_wol(struct hclge_dev *hdev)
+{
+	struct hclge_wol_info *wol_info = &hdev->hw.mac.wol;
+	int ret;
+
+	if (!hnae3_ae_dev_wol_supported(hdev->ae_dev))
+		return 0;
+
+	memset(wol_info, 0, sizeof(struct hclge_wol_info));
+	ret = hclge_get_wol_supported_mode(hdev,
+					   &wol_info->wol_support_mode);
+	if (ret) {
+		wol_info->wol_support_mode = 0;
+		return ret;
+	}
+
+	return hclge_update_wol(hdev);
+}
+
+static void hclge_get_wol(struct hnae3_handle *handle,
+			  struct ethtool_wolinfo *wol)
+{
+	struct hclge_wol_info *wol_info = hclge_get_wol_info(handle);
+
+	wol->supported = wol_info->wol_support_mode;
+	wol->wolopts = wol_info->wol_current_mode;
+	if (wol_info->wol_current_mode & WAKE_MAGICSECURE)
+		memcpy(wol->sopass, wol_info->wol_sopass, SOPASS_MAX);
+}
+
+static int hclge_set_wol(struct hnae3_handle *handle,
+			 struct ethtool_wolinfo *wol)
+{
+	struct hclge_wol_info *wol_info = hclge_get_wol_info(handle);
+	struct hclge_vport *vport = hclge_get_vport(handle);
+	u32 wol_mode;
+	int ret;
+
+	wol_mode = wol->wolopts;
+	if (wol_mode & ~wol_info->wol_support_mode)
+		return -EINVAL;
+
+	wol_info->wol_current_mode = wol_mode;
+	if (wol_mode & WAKE_MAGICSECURE) {
+		memcpy(wol_info->wol_sopass, wol->sopass, SOPASS_MAX);
+		wol_info->wol_sopass_size = SOPASS_MAX;
+	} else {
+		wol_info->wol_sopass_size = 0;
+	}
+
+	ret = hclge_set_wol_cfg(vport->back, wol_info);
+	if (ret)
+		wol_info->wol_current_mode = 0;
+
+	return ret;
+}
+
 static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 {
 	struct pci_dev *pdev = ae_dev->pdev;
@@ -11718,6 +11836,11 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev)
 	/* Enable MISC vector(vector0) */
 	hclge_enable_vector(&hdev->misc_vector, true);
 
+	ret = hclge_init_wol(hdev);
+	if (ret)
+		dev_warn(&pdev->dev,
+			 "failed to wake on lan init, ret = %d\n", ret);
+
 	hclge_state_init(hdev);
 	hdev->last_reset_time = jiffies;
 
@@ -12096,6 +12219,11 @@ static int hclge_reset_ae_dev(struct hnae3_ae_dev *ae_dev)
 
 	hclge_init_rxd_adv_layout(hdev);
 
+	ret = hclge_update_wol(hdev);
+	if (ret)
+		dev_warn(&pdev->dev,
+			 "failed to update wol config, ret = %d\n", ret);
+
 	dev_info(&pdev->dev, "Reset done, %s driver initialization finished.\n",
 		 HCLGE_DRIVER_NAME);
 
@@ -13142,6 +13270,8 @@ static const struct hnae3_ae_ops hclge_ops = {
 	.get_link_diagnosis_info = hclge_get_link_diagnosis_info,
 	.clean_vf_config = hclge_clean_vport_config,
 	.get_dscp_prio = hclge_get_dscp_prio,
+	.get_wol = hclge_get_wol,
+	.set_wol = hclge_set_wol,
 };
 
 static struct hnae3_ae_algo ae_algo = {
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
index 13f23d606e77..81aa6b0facf5 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h
@@ -249,6 +249,13 @@ enum HCLGE_MAC_DUPLEX {
 #define QUERY_SFP_SPEED		0
 #define QUERY_ACTIVE_SPEED	1
 
+struct hclge_wol_info {
+	u32 wol_support_mode; /* store the wake on lan info */
+	u32 wol_current_mode;
+	u8 wol_sopass[SOPASS_MAX];
+	u8 wol_sopass_size;
+};
+
 struct hclge_mac {
 	u8 mac_id;
 	u8 phy_addr;
@@ -268,6 +275,7 @@ struct hclge_mac {
 	u32 user_fec_mode;
 	u32 fec_ability;
 	int link;	/* store the link status of mac & phy (if phy exists) */
+	struct hclge_wol_info wol;
 	struct phy_device *phydev;
 	struct mii_bus *mdio_bus;
 	phy_interface_t phy_if;
-- 
cgit v1.2.3-70-g09d2


From d8d8b008629ffd69f1e204010cb3299bb633877e Mon Sep 17 00:00:00 2001
From: Manu Bretelle <chantr4@gmail.com>
Date: Wed, 29 Mar 2023 00:30:02 -0700
Subject: tools: bpftool: json: Fix backslash escape typo in jsonw_puts

This is essentially a backport of iproute2's
commit ed54f76484b5 ("json: fix backslash escape typo in jsonw_puts")

Also added the stdio.h include in json_writer.h to be able to compile
and run the json_writer test as used below).

Before this fix:

$ gcc -D notused -D TEST -I../../include -o json_writer  json_writer.c
json_writer.h
$ ./json_writer
{
    "Vyatta": {
        "url": "http://vyatta.com",
        "downloads": 2000000,
        "stock": 8.16,
        "ARGV": [],
        "empty": [],
        "NIL": {},
        "my_null": null,
        "special chars": [
            "slash": "/",
            "newline": "\n",
            "tab": "\t",
            "ff": "\f",
            "quote": "\"",
            "tick": "'",
            "backslash": "\n"
        ]
    }
}

After:

$ gcc -D notused -D TEST -I../../include -o json_writer  json_writer.c
json_writer.h
$ ./json_writer
{
    "Vyatta": {
        "url": "http://vyatta.com",
        "downloads": 2000000,
        "stock": 8.16,
        "ARGV": [],
        "empty": [],
        "NIL": {},
        "my_null": null,
        "special chars": [
            "slash": "/",
            "newline": "\n",
            "tab": "\t",
            "ff": "\f",
            "quote": "\"",
            "tick": "'",
            "backslash": "\\"
        ]
    }
}

Signed-off-by: Manu Bretelle <chantr4@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/bpf/20230329073002.2026563-1-chantr4@gmail.com
---
 tools/bpf/bpftool/json_writer.c | 2 +-
 tools/bpf/bpftool/json_writer.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/json_writer.c b/tools/bpf/bpftool/json_writer.c
index bca5dd0a59e3..be379613d118 100644
--- a/tools/bpf/bpftool/json_writer.c
+++ b/tools/bpf/bpftool/json_writer.c
@@ -75,7 +75,7 @@ static void jsonw_puts(json_writer_t *self, const char *str)
 			fputs("\\b", self->out);
 			break;
 		case '\\':
-			fputs("\\n", self->out);
+			fputs("\\\\", self->out);
 			break;
 		case '"':
 			fputs("\\\"", self->out);
diff --git a/tools/bpf/bpftool/json_writer.h b/tools/bpf/bpftool/json_writer.h
index 8ace65cdb92f..5aaffd3b837b 100644
--- a/tools/bpf/bpftool/json_writer.h
+++ b/tools/bpf/bpftool/json_writer.h
@@ -14,6 +14,7 @@
 #include <stdbool.h>
 #include <stdint.h>
 #include <stdarg.h>
+#include <stdio.h>
 #include <linux/compiler.h>
 
 /* Opaque class structure */
-- 
cgit v1.2.3-70-g09d2


From 562dc56a88983421a6c5a46e0feb891873d118a1 Mon Sep 17 00:00:00 2001
From: Yixin Shen <bobankhshen@gmail.com>
Date: Wed, 29 Mar 2023 07:35:57 +0000
Subject: bpf: allow a TCP CC to write app_limited

A CC that implements tcp_congestion_ops.cong_control() should be able to
write app_limited. A built-in CC or one from a kernel module is already
able to write to this member of struct tcp_sock.
For a BPF program, write access has not been allowed, yet.

Signed-off-by: Yixin Shen <bobankhshen@gmail.com>
Link: https://lore.kernel.org/r/20230329073558.8136-2-bobankhshen@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 net/ipv4/bpf_tcp_ca.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index e8b27826283e..ea21c96c03aa 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -113,6 +113,9 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log,
 	case offsetof(struct tcp_sock, ecn_flags):
 		end = offsetofend(struct tcp_sock, ecn_flags);
 		break;
+	case offsetof(struct tcp_sock, app_limited):
+		end = offsetofend(struct tcp_sock, app_limited);
+		break;
 	default:
 		bpf_log(log, "no write support to tcp_sock at off %d\n", off);
 		return -EACCES;
-- 
cgit v1.2.3-70-g09d2


From 4239561b69feb94e52e43d93685cc46fb9dbcae5 Mon Sep 17 00:00:00 2001
From: Yixin Shen <bobankhshen@gmail.com>
Date: Wed, 29 Mar 2023 07:35:58 +0000
Subject: selftests/bpf: test a BPF CC writing app_limited

Test whether a TCP CC implemented in BPF is allowed to write
app_limited in struct tcp_sock. This is already allowed for
the built-in TCP CC.

Signed-off-by: Yixin Shen <bobankhshen@gmail.com>
Link: https://lore.kernel.org/r/20230329073558.8136-3-bobankhshen@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/progs/tcp_ca_write_sk_pacing.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/tcp_ca_write_sk_pacing.c b/tools/testing/selftests/bpf/progs/tcp_ca_write_sk_pacing.c
index 43447704cf0e..0724a79cec78 100644
--- a/tools/testing/selftests/bpf/progs/tcp_ca_write_sk_pacing.c
+++ b/tools/testing/selftests/bpf/progs/tcp_ca_write_sk_pacing.c
@@ -16,6 +16,16 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk)
 	return (struct tcp_sock *)sk;
 }
 
+static inline unsigned int tcp_left_out(const struct tcp_sock *tp)
+{
+	return tp->sacked_out + tp->lost_out;
+}
+
+static inline unsigned int tcp_packets_in_flight(const struct tcp_sock *tp)
+{
+	return tp->packets_out - tcp_left_out(tp) + tp->retrans_out;
+}
+
 SEC("struct_ops/write_sk_pacing_init")
 void BPF_PROG(write_sk_pacing_init, struct sock *sk)
 {
@@ -31,11 +41,12 @@ SEC("struct_ops/write_sk_pacing_cong_control")
 void BPF_PROG(write_sk_pacing_cong_control, struct sock *sk,
 	      const struct rate_sample *rs)
 {
-	const struct tcp_sock *tp = tcp_sk(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
 	unsigned long rate =
 		((tp->snd_cwnd * tp->mss_cache * USEC_PER_SEC) << 3) /
 		(tp->srtt_us ?: 1U << 3);
 	sk->sk_pacing_rate = min(rate, sk->sk_max_pacing_rate);
+	tp->app_limited = (tp->delivered + tcp_packets_in_flight(tp)) ?: 1;
 }
 
 SEC("struct_ops/write_sk_pacing_ssthresh")
-- 
cgit v1.2.3-70-g09d2


From d6e6286a12e7b8a4ddc66237c4ccf6f531ef1c82 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 27 Mar 2023 11:52:00 -0700
Subject: libbpf: disassociate section handler on explicit
 bpf_program__set_type() call

If user explicitly overrides programs's type with
bpf_program__set_type() API call, we need to disassociate whatever
SEC_DEF handler libbpf determined initially based on program's SEC()
definition, as it's not goind to be valid anymore and could lead to
crashes and/or confusing failures.

Also, fix up bpf_prog_test_load() helper in selftests/bpf, which is
force-setting program type (even if that's completely unnecessary; this
is quite a legacy piece of code), and thus should expect auto-attach to
not work, yet one of the tests explicitly relies on auto-attach for
testing.

Instead, force-set program type only if it differs from the desired one.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230327185202.1929145-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf.c                        | 1 +
 tools/testing/selftests/bpf/testing_helpers.c | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 15737d7b5a28..49cd304ae3bc 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -8468,6 +8468,7 @@ int bpf_program__set_type(struct bpf_program *prog, enum bpf_prog_type type)
 		return libbpf_err(-EBUSY);
 
 	prog->type = type;
+	prog->sec_def = NULL;
 	return 0;
 }
 
diff --git a/tools/testing/selftests/bpf/testing_helpers.c b/tools/testing/selftests/bpf/testing_helpers.c
index 6c44153755e6..ecfea13f938b 100644
--- a/tools/testing/selftests/bpf/testing_helpers.c
+++ b/tools/testing/selftests/bpf/testing_helpers.c
@@ -195,7 +195,7 @@ int bpf_prog_test_load(const char *file, enum bpf_prog_type type,
 		goto err_out;
 	}
 
-	if (type != BPF_PROG_TYPE_UNSPEC)
+	if (type != BPF_PROG_TYPE_UNSPEC && bpf_program__type(prog) != type)
 		bpf_program__set_type(prog, type);
 
 	flags = bpf_program__flags(prog) | BPF_F_TEST_RND_HI32;
-- 
cgit v1.2.3-70-g09d2


From b3c63d7ad81ad6f43921d59af18fc25c64327a74 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 27 Mar 2023 11:52:01 -0700
Subject: veristat: add -d debug mode option to see debug libbpf log

Add -d option to allow requesting libbpf debug logs from veristat.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230327185202.1929145-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index 83231456d3c5..263df32fbda8 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -135,6 +135,7 @@ static struct env {
 	char **filenames;
 	int filename_cnt;
 	bool verbose;
+	bool debug;
 	bool quiet;
 	int log_level;
 	enum resfmt out_fmt;
@@ -169,7 +170,7 @@ static int libbpf_print_fn(enum libbpf_print_level level, const char *format, va
 {
 	if (!env.verbose)
 		return 0;
-	if (level == LIBBPF_DEBUG /* && !env.verbose */)
+	if (level == LIBBPF_DEBUG  && !env.debug)
 		return 0;
 	return vfprintf(stderr, format, args);
 }
@@ -186,6 +187,7 @@ static const struct argp_option opts[] = {
 	{ NULL, 'h', NULL, OPTION_HIDDEN, "Show the full help" },
 	{ "verbose", 'v', NULL, 0, "Verbose mode" },
 	{ "log-level", 'l', "LEVEL", 0, "Verifier log level (default 0 for normal mode, 1 for verbose mode)" },
+	{ "debug", 'd', NULL, 0, "Debug mode (turns on libbpf debug logging)" },
 	{ "quiet", 'q', NULL, 0, "Quiet mode" },
 	{ "emit", 'e', "SPEC", 0, "Specify stats to be emitted" },
 	{ "sort", 's', "SPEC", 0, "Specify sort order" },
@@ -212,6 +214,10 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 	case 'v':
 		env.verbose = true;
 		break;
+	case 'd':
+		env.debug = true;
+		env.verbose = true;
+		break;
 	case 'q':
 		env.quiet = true;
 		break;
-- 
cgit v1.2.3-70-g09d2


From fa7cc90620870e4444bb5184c08148495b1627c6 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 27 Mar 2023 11:52:02 -0700
Subject: veristat: guess and substitue underlying program type for freplace
 (EXT) progs

SEC("freplace") (i.e., BPF_PROG_TYPE_EXT) programs are not loadable as
is through veristat, as kernel expects actual program's FD during
BPF_PROG_LOAD time, which veristat has no way of knowing.

Unfortunately, freplace programs are a pretty important class of
programs, especially when dealing with XDP chaining solutions, which
rely on EXT programs.

So let's do our best and teach veristat to try to guess the original
program type, based on program's context argument type. And if guessing
process succeeds, we manually override freplace/EXT with guessed program
type using bpf_program__set_type() setter to increase chances of proper
BPF verification.

We rely on BTF and maintain a simple lookup table. This process is
obviously not 100% bulletproof, as valid program might not use context
and thus wouldn't have to specify correct type. Also, __sk_buff is very
ambiguous and is the context type across many different program types.
We pick BPF_PROG_TYPE_CGROUP_SKB for now, which seems to work fine in
practice so far. Similarly, some program types require specifying attach
type, and so we pick one out of possible few variants.

Best effort at its best. But this makes veristat even more widely
applicable.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230327185202.1929145-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 121 +++++++++++++++++++++++++++++++--
 1 file changed, 117 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index 263df32fbda8..055df1abd7ca 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -15,6 +15,7 @@
 #include <sys/sysinfo.h>
 #include <sys/stat.h>
 #include <bpf/libbpf.h>
+#include <bpf/btf.h>
 #include <libelf.h>
 #include <gelf.h>
 #include <float.h>
@@ -778,7 +779,62 @@ static int parse_verif_log(char * const buf, size_t buf_sz, struct verif_stats *
 	return 0;
 }
 
-static void fixup_obj(struct bpf_object *obj)
+static int guess_prog_type_by_ctx_name(const char *ctx_name,
+				       enum bpf_prog_type *prog_type,
+				       enum bpf_attach_type *attach_type)
+{
+	/* We need to guess program type based on its declared context type.
+	 * This guess can't be perfect as many different program types might
+	 * share the same context type.  So we can only hope to reasonably
+	 * well guess this and get lucky.
+	 *
+	 * Just in case, we support both UAPI-side type names and
+	 * kernel-internal names.
+	 */
+	static struct {
+		const char *uapi_name;
+		const char *kern_name;
+		enum bpf_prog_type prog_type;
+		enum bpf_attach_type attach_type;
+	} ctx_map[] = {
+		/* __sk_buff is most ambiguous, for now we assume cgroup_skb */
+		{ "__sk_buff", "sk_buff", BPF_PROG_TYPE_CGROUP_SKB, BPF_CGROUP_INET_INGRESS },
+		{ "bpf_sock", "sock", BPF_PROG_TYPE_CGROUP_SOCK, BPF_CGROUP_INET4_POST_BIND },
+		{ "bpf_sock_addr", "bpf_sock_addr_kern",  BPF_PROG_TYPE_CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_BIND },
+		{ "bpf_sock_ops", "bpf_sock_ops_kern", BPF_PROG_TYPE_SOCK_OPS, BPF_CGROUP_SOCK_OPS },
+		{ "sk_msg_md", "sk_msg", BPF_PROG_TYPE_SK_MSG, BPF_SK_MSG_VERDICT },
+		{ "bpf_cgroup_dev_ctx", "bpf_cgroup_dev_ctx", BPF_PROG_TYPE_CGROUP_DEVICE, BPF_CGROUP_DEVICE },
+		{ "bpf_sysctl", "bpf_sysctl_kern", BPF_PROG_TYPE_CGROUP_SYSCTL, BPF_CGROUP_SYSCTL },
+		{ "bpf_sockopt", "bpf_sockopt_kern", BPF_PROG_TYPE_CGROUP_SOCKOPT, BPF_CGROUP_SETSOCKOPT },
+		{ "sk_reuseport_md", "sk_reuseport_kern", BPF_PROG_TYPE_SK_REUSEPORT, BPF_SK_REUSEPORT_SELECT_OR_MIGRATE },
+		{ "bpf_sk_lookup", "bpf_sk_lookup_kern", BPF_PROG_TYPE_SK_LOOKUP, BPF_SK_LOOKUP },
+		{ "xdp_md", "xdp_buff", BPF_PROG_TYPE_XDP, BPF_XDP },
+		/* tracing types with no expected attach type */
+		{ "bpf_user_pt_regs_t", "pt_regs", BPF_PROG_TYPE_KPROBE },
+		{ "bpf_perf_event_data", "bpf_perf_event_data_kern", BPF_PROG_TYPE_PERF_EVENT },
+		/* raw_tp programs use u64[] from kernel side, we don't want
+		 * to match on that, probably; so NULL for kern-side type
+		 */
+		{ "bpf_raw_tracepoint_args", NULL, BPF_PROG_TYPE_RAW_TRACEPOINT },
+	};
+	int i;
+
+	if (!ctx_name)
+		return -EINVAL;
+
+	for (i = 0; i < ARRAY_SIZE(ctx_map); i++) {
+		if (strcmp(ctx_map[i].uapi_name, ctx_name) == 0 ||
+		    (ctx_map[i].kern_name && strcmp(ctx_map[i].kern_name, ctx_name) == 0)) {
+			*prog_type = ctx_map[i].prog_type;
+			*attach_type = ctx_map[i].attach_type;
+			return 0;
+		}
+	}
+
+	return -ESRCH;
+}
+
+static void fixup_obj(struct bpf_object *obj, struct bpf_program *prog, const char *filename)
 {
 	struct bpf_map *map;
 
@@ -798,18 +854,75 @@ static void fixup_obj(struct bpf_object *obj)
 				bpf_map__set_max_entries(map, 1);
 		}
 	}
+
+	/* SEC(freplace) programs can't be loaded with veristat as is,
+	 * but we can try guessing their target program's expected type by
+	 * looking at the type of program's first argument and substituting
+	 * corresponding program type
+	 */
+	if (bpf_program__type(prog) == BPF_PROG_TYPE_EXT) {
+		const struct btf *btf = bpf_object__btf(obj);
+		const char *prog_name = bpf_program__name(prog);
+		enum bpf_prog_type prog_type;
+		enum bpf_attach_type attach_type;
+		const struct btf_type *t;
+		const char *ctx_name;
+		int id;
+
+		if (!btf)
+			goto skip_freplace_fixup;
+
+		id = btf__find_by_name_kind(btf, prog_name, BTF_KIND_FUNC);
+		t = btf__type_by_id(btf, id);
+		t = btf__type_by_id(btf, t->type);
+		if (!btf_is_func_proto(t) || btf_vlen(t) != 1)
+			goto skip_freplace_fixup;
+
+		/* context argument is a pointer to a struct/typedef */
+		t = btf__type_by_id(btf, btf_params(t)[0].type);
+		while (t && btf_is_mod(t))
+			t = btf__type_by_id(btf, t->type);
+		if (!t || !btf_is_ptr(t))
+			goto skip_freplace_fixup;
+		t = btf__type_by_id(btf, t->type);
+		while (t && btf_is_mod(t))
+			t = btf__type_by_id(btf, t->type);
+		if (!t)
+			goto skip_freplace_fixup;
+
+		ctx_name = btf__name_by_offset(btf, t->name_off);
+
+		if (guess_prog_type_by_ctx_name(ctx_name, &prog_type, &attach_type) == 0) {
+			bpf_program__set_type(prog, prog_type);
+			bpf_program__set_expected_attach_type(prog, attach_type);
+
+			if (!env.quiet) {
+				printf("Using guessed program type '%s' for %s/%s...\n",
+					libbpf_bpf_prog_type_str(prog_type),
+					filename, prog_name);
+			}
+		} else {
+			if (!env.quiet) {
+				printf("Failed to guess program type for freplace program with context type name '%s' for %s/%s. Consider using canonical type names to help veristat...\n",
+					ctx_name, filename, prog_name);
+			}
+		}
+	}
+skip_freplace_fixup:
+	return;
 }
 
 static int process_prog(const char *filename, struct bpf_object *obj, struct bpf_program *prog)
 {
 	const char *prog_name = bpf_program__name(prog);
+	const char *base_filename = basename(filename);
 	size_t buf_sz = sizeof(verif_log_buf);
 	char *buf = verif_log_buf;
 	struct verif_stats *stats;
 	int err = 0;
 	void *tmp;
 
-	if (!should_process_file_prog(basename(filename), bpf_program__name(prog))) {
+	if (!should_process_file_prog(base_filename, bpf_program__name(prog))) {
 		env.progs_skipped++;
 		return 0;
 	}
@@ -835,12 +948,12 @@ static int process_prog(const char *filename, struct bpf_object *obj, struct bpf
 	verif_log_buf[0] = '\0';
 
 	/* increase chances of successful BPF object loading */
-	fixup_obj(obj);
+	fixup_obj(obj, prog, base_filename);
 
 	err = bpf_object__load(obj);
 	env.progs_processed++;
 
-	stats->file_name = strdup(basename(filename));
+	stats->file_name = strdup(base_filename);
 	stats->prog_name = strdup(bpf_program__name(prog));
 	stats->stats[VERDICT] = err == 0; /* 1 - success, 0 - failure */
 	parse_verif_log(buf, buf_sz, stats);
-- 
cgit v1.2.3-70-g09d2


From 4ca13d1002f37c10038ff4ed3cfdc70dbe049d60 Mon Sep 17 00:00:00 2001
From: Xu Kuohai <xukuohai@huawei.com>
Date: Tue, 28 Mar 2023 21:10:48 -0400
Subject: selftests/bpf: Rewrite two infinite loops in bound check cases

The two infinite loops in bound check cases added by commit
1a3148fc171f ("selftests/bpf: Check when bounds are not in the 32-bit range")
increased the execution time of test_verifier from about 6 seconds to
about 9 seconds. Rewrite these two infinite loops to finite loops to get
rid of this extra time cost.

Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Link: https://lore.kernel.org/r/20230329011048.1721937-1-xukuohai@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/verifier/bounds.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/bpf/verifier/bounds.c b/tools/testing/selftests/bpf/verifier/bounds.c
index 74b1917d4208..43942ce8cf15 100644
--- a/tools/testing/selftests/bpf/verifier/bounds.c
+++ b/tools/testing/selftests/bpf/verifier/bounds.c
@@ -784,22 +784,26 @@
 	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
 	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
 	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 8),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 13),
 
 	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
 	BPF_LD_IMM64(BPF_REG_0, 0x7fffffffffffff10),
 	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
 
+	BPF_LD_IMM64(BPF_REG_2, 0x8000000000000fff),
 	BPF_LD_IMM64(BPF_REG_0, 0x8000000000000000),
 	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
+	BPF_JMP_REG(BPF_JSGT, BPF_REG_0, BPF_REG_2, 3),
 	/* r1 signed range is [S64_MIN, S64_MAX] */
-	BPF_JMP_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -2),
+	BPF_JMP_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -3),
+
+	BPF_MOV64_IMM(BPF_REG_0, 1),
+	BPF_EXIT_INSN(),
 
 	BPF_MOV64_IMM(BPF_REG_0, 0),
 	BPF_EXIT_INSN(),
 	},
-	.errstr = "BPF program is too large",
-	.result = REJECT,
+	.result = ACCEPT,
 	.prog_type = BPF_PROG_TYPE_XDP,
 },
 {
@@ -856,21 +860,25 @@
 	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
 	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
 	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 6),
+	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 10),
 
 	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
 	BPF_MOV32_IMM(BPF_REG_0, 0x7fffff10),
 	BPF_ALU32_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
 
+	BPF_MOV32_IMM(BPF_REG_2, 0x80000fff),
 	BPF_MOV32_IMM(BPF_REG_0, 0x80000000),
 	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 1),
+	BPF_JMP32_REG(BPF_JSGT, BPF_REG_0, BPF_REG_2, 3),
 	/* r1 signed range is [S32_MIN, S32_MAX] */
-	BPF_JMP32_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -2),
+	BPF_JMP32_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -3),
+
+	BPF_MOV64_IMM(BPF_REG_0, 1),
+	BPF_EXIT_INSN(),
 
 	BPF_MOV64_IMM(BPF_REG_0, 0),
 	BPF_EXIT_INSN(),
 	},
-	.errstr = "BPF program is too large",
-	.result = REJECT,
+	.result = ACCEPT,
 	.prog_type = BPF_PROG_TYPE_XDP,
 },
-- 
cgit v1.2.3-70-g09d2


From 615927f1a487bd9066425df48d06cec3bb3d8dbb Mon Sep 17 00:00:00 2001
From: Tianfei Zhang <tianfei.zhang@intel.com>
Date: Tue, 28 Mar 2023 10:24:55 -0400
Subject: ptp: add ToD device driver for Intel FPGA cards
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Adding a DFL (Device Feature List) device driver of ToD device for
Intel FPGA cards.

The Intel FPGA Time of Day(ToD) IP within the FPGA DFL bus is exposed
as PTP Hardware clock(PHC) device to the Linux PTP stack to synchronize
the system clock to its ToD information using phc2sys utility of the
Linux PTP stack. The DFL is a hardware List within FPGA, which defines
a linked list of feature headers within the device MMIO space to provide
an extensible way of adding subdevice features.

Signed-off-by: Raghavendra Khadatare <raghavendrax.anand.khadatare@intel.com>
Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20230328142455.481146-1-tianfei.zhang@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS               |   7 +
 drivers/ptp/Kconfig       |  14 ++
 drivers/ptp/Makefile      |   1 +
 drivers/ptp/ptp_dfl_tod.c | 332 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 354 insertions(+)
 create mode 100644 drivers/ptp/ptp_dfl_tod.c

diff --git a/MAINTAINERS b/MAINTAINERS
index fbbda4671e73..91201c2b8190 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15630,6 +15630,13 @@ L:	netdev@vger.kernel.org
 S:	Maintained
 F:	drivers/ptp/ptp_ocp.c
 
+INTEL PTP DFL ToD DRIVER
+M:	Tianfei Zhang <tianfei.zhang@intel.com>
+L:	linux-fpga@vger.kernel.org
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/ptp/ptp_dfl_tod.c
+
 OPENCORES I2C BUS DRIVER
 M:	Peter Korsgaard <peter@korsgaard.com>
 M:	Andrew Lunn <andrew@lunn.ch>
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index fe4971b65c64..b00201d81313 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -186,4 +186,18 @@ config PTP_1588_CLOCK_OCP
 
 	  More information is available at http://www.timingcard.com/
 
+config PTP_DFL_TOD
+	tristate "FPGA DFL ToD Driver"
+	depends on FPGA_DFL
+	depends on PTP_1588_CLOCK
+	help
+	  The DFL (Device Feature List) device driver for the Intel ToD
+	  (Time-of-Day) device in FPGA card. The ToD IP within the FPGA
+	  is exposed as PTP Hardware Clock (PHC) device to the Linux PTP
+	  stack to synchronize the system clock to its ToD information
+	  using phc2sys utility of the Linux PTP stack.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called ptp_dfl_tod.
+
 endmenu
diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile
index 28a6fe342d3e..553f18bf3c83 100644
--- a/drivers/ptp/Makefile
+++ b/drivers/ptp/Makefile
@@ -18,3 +18,4 @@ obj-$(CONFIG_PTP_1588_CLOCK_IDTCM)	+= ptp_clockmatrix.o
 obj-$(CONFIG_PTP_1588_CLOCK_IDT82P33)	+= ptp_idt82p33.o
 obj-$(CONFIG_PTP_1588_CLOCK_VMW)	+= ptp_vmw.o
 obj-$(CONFIG_PTP_1588_CLOCK_OCP)	+= ptp_ocp.o
+obj-$(CONFIG_PTP_DFL_TOD)		+= ptp_dfl_tod.o
diff --git a/drivers/ptp/ptp_dfl_tod.c b/drivers/ptp/ptp_dfl_tod.c
new file mode 100644
index 000000000000..f699d541b360
--- /dev/null
+++ b/drivers/ptp/ptp_dfl_tod.c
@@ -0,0 +1,332 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * DFL device driver for Time-of-Day (ToD) private feature
+ *
+ * Copyright (C) 2023 Intel Corporation
+ */
+
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/dfl.h>
+#include <linux/gcd.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/spinlock.h>
+#include <linux/units.h>
+
+#define FME_FEATURE_ID_TOD		0x22
+
+/* ToD clock register space. */
+#define TOD_CLK_FREQ			0x038
+
+/*
+ * The read sequence of ToD timestamp registers: TOD_NANOSEC, TOD_SECONDSL and
+ * TOD_SECONDSH, because there is a hardware snapshot whenever the TOD_NANOSEC
+ * register is read.
+ *
+ * The ToD IP requires writing registers in the reverse order to the read sequence.
+ * The timestamp is corrected when the TOD_NANOSEC register is written, so the
+ * sequence of write TOD registers: TOD_SECONDSH, TOD_SECONDSL and TOD_NANOSEC.
+ */
+#define TOD_SECONDSH			0x100
+#define TOD_SECONDSL			0x104
+#define TOD_NANOSEC			0x108
+#define TOD_PERIOD			0x110
+#define TOD_ADJUST_PERIOD		0x114
+#define TOD_ADJUST_COUNT		0x118
+#define TOD_DRIFT_ADJUST		0x11c
+#define TOD_DRIFT_ADJUST_RATE		0x120
+#define PERIOD_FRAC_OFFSET		16
+#define SECONDS_MSB			GENMASK_ULL(47, 32)
+#define SECONDS_LSB			GENMASK_ULL(31, 0)
+#define TOD_SECONDSH_SEC_MSB		GENMASK_ULL(15, 0)
+
+#define CAL_SECONDS(m, l)		((FIELD_GET(TOD_SECONDSH_SEC_MSB, (m)) << 32) | (l))
+
+#define TOD_PERIOD_MASK		GENMASK_ULL(19, 0)
+#define TOD_PERIOD_MAX			FIELD_MAX(TOD_PERIOD_MASK)
+#define TOD_PERIOD_MIN			0
+#define TOD_DRIFT_ADJUST_MASK		GENMASK_ULL(15, 0)
+#define TOD_DRIFT_ADJUST_FNS_MAX	FIELD_MAX(TOD_DRIFT_ADJUST_MASK)
+#define TOD_DRIFT_ADJUST_RATE_MAX	TOD_DRIFT_ADJUST_FNS_MAX
+#define TOD_ADJUST_COUNT_MASK		GENMASK_ULL(19, 0)
+#define TOD_ADJUST_COUNT_MAX		FIELD_MAX(TOD_ADJUST_COUNT_MASK)
+#define TOD_ADJUST_INTERVAL_US		10
+#define TOD_ADJUST_MS			\
+		(((TOD_PERIOD_MAX >> 16) + 1) * (TOD_ADJUST_COUNT_MAX + 1))
+#define TOD_ADJUST_MS_MAX		(TOD_ADJUST_MS / MICRO)
+#define TOD_ADJUST_MAX_US		(TOD_ADJUST_MS_MAX * USEC_PER_MSEC)
+#define TOD_MAX_ADJ			(500 * MEGA)
+
+struct dfl_tod {
+	struct ptp_clock_info ptp_clock_ops;
+	struct device *dev;
+	struct ptp_clock *ptp_clock;
+
+	/* ToD Clock address space */
+	void __iomem *tod_ctrl;
+
+	/* ToD clock registers protection */
+	spinlock_t tod_lock;
+};
+
+/*
+ * A fine ToD HW clock offset adjustment. To perform the fine offset adjustment, the
+ * adjust_period and adjust_count argument are used to update the TOD_ADJUST_PERIOD
+ * and TOD_ADJUST_COUNT register for in hardware. The dt->tod_lock spinlock must be
+ * held when calling this function.
+ */
+static int fine_adjust_tod_clock(struct dfl_tod *dt, u32 adjust_period,
+				 u32 adjust_count)
+{
+	void __iomem *base = dt->tod_ctrl;
+	u32 val;
+
+	writel(adjust_period, base + TOD_ADJUST_PERIOD);
+	writel(adjust_count, base + TOD_ADJUST_COUNT);
+
+	/* Wait for present offset adjustment update to complete */
+	return readl_poll_timeout_atomic(base + TOD_ADJUST_COUNT, val, !val, TOD_ADJUST_INTERVAL_US,
+				  TOD_ADJUST_MAX_US);
+}
+
+/*
+ * A coarse ToD HW clock offset adjustment. The coarse time adjustment performs by
+ * adding or subtracting the delta value from the current ToD HW clock time.
+ */
+static int coarse_adjust_tod_clock(struct dfl_tod *dt, s64 delta)
+{
+	u32 seconds_msb, seconds_lsb, nanosec;
+	void __iomem *base = dt->tod_ctrl;
+	u64 seconds, now;
+
+	if (delta == 0)
+		return 0;
+
+	nanosec = readl(base + TOD_NANOSEC);
+	seconds_lsb = readl(base + TOD_SECONDSL);
+	seconds_msb = readl(base + TOD_SECONDSH);
+
+	/* Calculate new time */
+	seconds = CAL_SECONDS(seconds_msb, seconds_lsb);
+	now = seconds * NSEC_PER_SEC + nanosec + delta;
+
+	seconds = div_u64_rem(now, NSEC_PER_SEC, &nanosec);
+	seconds_msb = FIELD_GET(SECONDS_MSB, seconds);
+	seconds_lsb = FIELD_GET(SECONDS_LSB, seconds);
+
+	writel(seconds_msb, base + TOD_SECONDSH);
+	writel(seconds_lsb, base + TOD_SECONDSL);
+	writel(nanosec, base + TOD_NANOSEC);
+
+	return 0;
+}
+
+static int dfl_tod_adjust_fine(struct ptp_clock_info *ptp, long scaled_ppm)
+{
+	struct dfl_tod *dt = container_of(ptp, struct dfl_tod, ptp_clock_ops);
+	u32 tod_period, tod_rem, tod_drift_adjust_fns, tod_drift_adjust_rate;
+	void __iomem *base = dt->tod_ctrl;
+	unsigned long flags, rate;
+	u64 ppb;
+
+	/* Get the clock rate from clock frequency register offset */
+	rate = readl(base + TOD_CLK_FREQ);
+
+	/* add GIGA as nominal ppb */
+	ppb = scaled_ppm_to_ppb(scaled_ppm) + GIGA;
+
+	tod_period = div_u64_rem(ppb << PERIOD_FRAC_OFFSET, rate, &tod_rem);
+	if (tod_period > TOD_PERIOD_MAX)
+		return -ERANGE;
+
+	/*
+	 * The drift of ToD adjusted periodically by adding a drift_adjust_fns
+	 * correction value every drift_adjust_rate count of clock cycles.
+	 */
+	tod_drift_adjust_fns = tod_rem / gcd(tod_rem, rate);
+	tod_drift_adjust_rate = rate / gcd(tod_rem, rate);
+
+	while ((tod_drift_adjust_fns > TOD_DRIFT_ADJUST_FNS_MAX) ||
+	       (tod_drift_adjust_rate > TOD_DRIFT_ADJUST_RATE_MAX)) {
+		tod_drift_adjust_fns >>= 1;
+		tod_drift_adjust_rate >>= 1;
+	}
+
+	if (tod_drift_adjust_fns == 0)
+		tod_drift_adjust_rate = 0;
+
+	spin_lock_irqsave(&dt->tod_lock, flags);
+	writel(tod_period, base + TOD_PERIOD);
+	writel(0, base + TOD_ADJUST_PERIOD);
+	writel(0, base + TOD_ADJUST_COUNT);
+	writel(tod_drift_adjust_fns, base + TOD_DRIFT_ADJUST);
+	writel(tod_drift_adjust_rate, base + TOD_DRIFT_ADJUST_RATE);
+	spin_unlock_irqrestore(&dt->tod_lock, flags);
+
+	return 0;
+}
+
+static int dfl_tod_adjust_time(struct ptp_clock_info *ptp, s64 delta)
+{
+	struct dfl_tod *dt = container_of(ptp, struct dfl_tod, ptp_clock_ops);
+	u32 period, diff, rem, rem_period, adj_period;
+	void __iomem *base = dt->tod_ctrl;
+	unsigned long flags;
+	bool neg_adj;
+	u64 count;
+	int ret;
+
+	neg_adj = delta < 0;
+	if (neg_adj)
+		delta = -delta;
+
+	spin_lock_irqsave(&dt->tod_lock, flags);
+
+	/*
+	 * Get the maximum possible value of the Period register offset
+	 * adjustment in nanoseconds scale. This depends on the current
+	 * Period register setting and the maximum and minimum possible
+	 * values of the Period register.
+	 */
+	period = readl(base + TOD_PERIOD);
+
+	if (neg_adj) {
+		diff = (period - TOD_PERIOD_MIN) >> PERIOD_FRAC_OFFSET;
+		adj_period = period - (diff << PERIOD_FRAC_OFFSET);
+		count = div_u64_rem(delta, diff, &rem);
+		rem_period = period - (rem << PERIOD_FRAC_OFFSET);
+	} else {
+		diff = (TOD_PERIOD_MAX - period) >> PERIOD_FRAC_OFFSET;
+		adj_period = period + (diff << PERIOD_FRAC_OFFSET);
+		count = div_u64_rem(delta, diff, &rem);
+		rem_period = period + (rem << PERIOD_FRAC_OFFSET);
+	}
+
+	ret = 0;
+
+	if (count > TOD_ADJUST_COUNT_MAX) {
+		ret = coarse_adjust_tod_clock(dt, delta);
+	} else {
+		/* Adjust the period by count cycles to adjust the time */
+		if (count)
+			ret = fine_adjust_tod_clock(dt, adj_period, count);
+
+		/* If there is a remainder, adjust the period for an additional cycle */
+		if (rem)
+			ret = fine_adjust_tod_clock(dt, rem_period, 1);
+	}
+
+	spin_unlock_irqrestore(&dt->tod_lock, flags);
+
+	return ret;
+}
+
+static int dfl_tod_get_timex(struct ptp_clock_info *ptp, struct timespec64 *ts,
+			     struct ptp_system_timestamp *sts)
+{
+	struct dfl_tod *dt = container_of(ptp, struct dfl_tod, ptp_clock_ops);
+	u32 seconds_msb, seconds_lsb, nanosec;
+	void __iomem *base = dt->tod_ctrl;
+	unsigned long flags;
+	u64 seconds;
+
+	spin_lock_irqsave(&dt->tod_lock, flags);
+	ptp_read_system_prets(sts);
+	nanosec = readl(base + TOD_NANOSEC);
+	seconds_lsb = readl(base + TOD_SECONDSL);
+	seconds_msb = readl(base + TOD_SECONDSH);
+	ptp_read_system_postts(sts);
+	spin_unlock_irqrestore(&dt->tod_lock, flags);
+
+	seconds = CAL_SECONDS(seconds_msb, seconds_lsb);
+
+	ts->tv_nsec = nanosec;
+	ts->tv_sec = seconds;
+
+	return 0;
+}
+
+static int dfl_tod_set_time(struct ptp_clock_info *ptp,
+			    const struct timespec64 *ts)
+{
+	struct dfl_tod *dt = container_of(ptp, struct dfl_tod, ptp_clock_ops);
+	u32 seconds_msb = FIELD_GET(SECONDS_MSB, ts->tv_sec);
+	u32 seconds_lsb = FIELD_GET(SECONDS_LSB, ts->tv_sec);
+	u32 nanosec = FIELD_GET(SECONDS_LSB, ts->tv_nsec);
+	void __iomem *base = dt->tod_ctrl;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dt->tod_lock, flags);
+	writel(seconds_msb, base + TOD_SECONDSH);
+	writel(seconds_lsb, base + TOD_SECONDSL);
+	writel(nanosec, base + TOD_NANOSEC);
+	spin_unlock_irqrestore(&dt->tod_lock, flags);
+
+	return 0;
+}
+
+static struct ptp_clock_info dfl_tod_clock_ops = {
+	.owner = THIS_MODULE,
+	.name = "dfl_tod",
+	.max_adj = TOD_MAX_ADJ,
+	.adjfine = dfl_tod_adjust_fine,
+	.adjtime = dfl_tod_adjust_time,
+	.gettimex64 = dfl_tod_get_timex,
+	.settime64 = dfl_tod_set_time,
+};
+
+static int dfl_tod_probe(struct dfl_device *ddev)
+{
+	struct device *dev = &ddev->dev;
+	struct dfl_tod *dt;
+
+	dt = devm_kzalloc(dev, sizeof(*dt), GFP_KERNEL);
+	if (!dt)
+		return -ENOMEM;
+
+	dt->tod_ctrl = devm_ioremap_resource(dev, &ddev->mmio_res);
+	if (IS_ERR(dt->tod_ctrl))
+		return PTR_ERR(dt->tod_ctrl);
+
+	dt->dev = dev;
+	spin_lock_init(&dt->tod_lock);
+	dev_set_drvdata(dev, dt);
+
+	dt->ptp_clock_ops = dfl_tod_clock_ops;
+
+	dt->ptp_clock = ptp_clock_register(&dt->ptp_clock_ops, dev);
+	if (IS_ERR(dt->ptp_clock))
+		return dev_err_probe(dt->dev, PTR_ERR(dt->ptp_clock),
+				     "Unable to register PTP clock\n");
+
+	return 0;
+}
+
+static void dfl_tod_remove(struct dfl_device *ddev)
+{
+	struct dfl_tod *dt = dev_get_drvdata(&ddev->dev);
+
+	ptp_clock_unregister(dt->ptp_clock);
+}
+
+static const struct dfl_device_id dfl_tod_ids[] = {
+	{ FME_ID, FME_FEATURE_ID_TOD },
+	{ }
+};
+MODULE_DEVICE_TABLE(dfl, dfl_tod_ids);
+
+static struct dfl_driver dfl_tod_driver = {
+	.drv = {
+		.name = "dfl-tod",
+	},
+	.id_table = dfl_tod_ids,
+	.probe = dfl_tod_probe,
+	.remove = dfl_tod_remove,
+};
+module_dfl_driver(dfl_tod_driver);
+
+MODULE_DESCRIPTION("FPGA DFL ToD driver");
+MODULE_AUTHOR("Intel Corporation");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-70-g09d2


From 765f3604641e1524389e1d6431974d2bbcab14a7 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Wed, 29 Mar 2023 09:51:37 +0300
Subject: octeon_ep: unlock the correct lock on error path

The h and the f letters are swapped so it unlocks the wrong lock.

Fixes: 577f0d1b1c5f ("octeon_ep: add separate mailbox command and response queues")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Link: https://lore.kernel.org/r/251aa2a2-913e-4868-aac9-0a90fc3eeeda@kili.mountain
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c
index a4ee6f3ae354..035ead7935c7 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_ctrl_mbox.c
@@ -167,7 +167,7 @@ int octep_ctrl_mbox_send(struct octep_ctrl_mbox *mbox, struct octep_ctrl_mbox_ms
 	ci = readl(q->hw_cons);
 
 	if (octep_ctrl_mbox_circq_space(pi, ci, q->sz) < (msg->hdr.s.sz + mbox_hdr_sz)) {
-		mutex_unlock(&mbox->f2hq_lock);
+		mutex_unlock(&mbox->h2fq_lock);
 		return -EAGAIN;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From c5370374bb1bf692167c7276be8b56c02565d535 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Tue, 28 Mar 2023 17:19:58 +0200
Subject: net: ena: removed unused tx_bytes variable

clang 16.0.0 with W=1 reports:

drivers/net/ethernet/amazon/ena/ena_netdev.c:1901:6: error: variable 'tx_bytes' set but not used [-Werror,-Wunused-but-set-variable]
        u32 tx_bytes = 0;

The variable is not used so remove it.

Signed-off-by: Simon Horman <horms@kernel.org>
Acked-by: Shay Agroskin <shayagr@amazon.com>
Link: https://lore.kernel.org/r/20230328151958.410687-1-horms@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index e0588a82c8e5..e6a6efaeb87c 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -1898,7 +1898,6 @@ static int ena_clean_xdp_irq(struct ena_ring *xdp_ring, u32 budget)
 {
 	u32 total_done = 0;
 	u16 next_to_clean;
-	u32 tx_bytes = 0;
 	int tx_pkts = 0;
 	u16 req_id;
 	int rc;
@@ -1936,7 +1935,6 @@ static int ena_clean_xdp_irq(struct ena_ring *xdp_ring, u32 budget)
 			  "tx_poll: q %d skb %p completed\n", xdp_ring->qid,
 			  xdpf);
 
-		tx_bytes += xdpf->len;
 		tx_pkts++;
 		total_done += tx_info->tx_descs;
 
-- 
cgit v1.2.3-70-g09d2


From 6cf882d9aa9e8f1f2d63182e179ac4b2e59c00db Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Wed, 29 Mar 2023 10:32:46 -0700
Subject: wifi: iwlwifi: mvm: Use 64-bit division helper in
 iwl_mvm_get_crosstimestamp_fw()

There is a 64-bit division in iwl_mvm_get_crosstimestamp_fw(), which
results in a link failure when building 32-bit architectures with clang:

  ld.lld: error: undefined symbol: __udivdi3
  >>> referenced by ptp.c
  >>>               drivers/net/wireless/intel/iwlwifi/mvm/ptp.o:(iwl_mvm_phc_get_crosstimestamp) in archive vmlinux.a

GCC has optimizations for division by a constant that clang does not
implement, so this issue is not visible when building with GCC.

Use the 64-bit division helper div_u64(), which takes a u64 dividend and
u32 divisor, which matches this situation and prevents the emission of a
libcall for the division.

Fixes: 21fb8da6ebe4 ("wifi: iwlwifi: mvm: read synced time from firmware if supported")
Reported-by: Arnd Bergmann <arnd@arndb.de>
Link: https://github.com/ClangBuiltLinux/linux/issues/1826
Reported-by: "kernelci.org bot" <bot@kernelci.org>
Link: https://lore.kernel.org/6423173a.620a0220.3d5cc.6358@mx.google.com/
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
index 5c2bfc8ed88d..e89259de6f4c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ptp.c
@@ -116,7 +116,7 @@ iwl_mvm_get_crosstimestamp_fw(struct iwl_mvm *mvm, u32 *gp2, u64 *sys_time)
 
 	gp2_10ns = (u64)le32_to_cpu(resp->gp2_timestamp_hi) << 32 |
 		le32_to_cpu(resp->gp2_timestamp_lo);
-	*gp2 = gp2_10ns / 100;
+	*gp2 = div_u64(gp2_10ns, 100);
 
 	*sys_time = (u64)le32_to_cpu(resp->platform_timestamp_hi) << 32 |
 		le32_to_cpu(resp->platform_timestamp_lo);
-- 
cgit v1.2.3-70-g09d2


From 968a768d28530cadd13aab51531f8fec70bf9eae Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Sat, 25 Mar 2023 09:26:10 -0400
Subject: mac80211: minstrel_ht: remove unused n_supported variable

clang with W=1 reports
net/mac80211/rc80211_minstrel_ht.c:1711:6: error: variable
  'n_supported' set but not used [-Werror,-Wunused-but-set-variable]
        int n_supported = 0;
            ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230325132610.1334820-1-trix@redhat.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/rc80211_minstrel_ht.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c
index 762346598338..b34c80522047 100644
--- a/net/mac80211/rc80211_minstrel_ht.c
+++ b/net/mac80211/rc80211_minstrel_ht.c
@@ -1708,7 +1708,6 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband,
 	struct sta_info *sta_info;
 	bool ldpc, erp;
 	int use_vht;
-	int n_supported = 0;
 	int ack_dur;
 	int stbc;
 	int i;
@@ -1791,8 +1790,6 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband,
 				continue;
 
 			mi->supported[i] = mcs->rx_mask[nss - 1];
-			if (mi->supported[i])
-				n_supported++;
 			continue;
 		}
 
@@ -1819,9 +1816,6 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband,
 
 		mi->supported[i] = minstrel_get_valid_vht_rates(bw, nss,
 				vht_cap->vht_mcs.tx_mcs_map);
-
-		if (mi->supported[i])
-			n_supported++;
 	}
 
 	sta_info = container_of(sta, struct sta_info, sta);
-- 
cgit v1.2.3-70-g09d2


From 0333a81bc83431d7f90391d38aa09e856c5e5b25 Mon Sep 17 00:00:00 2001
From: Kieran Frewen <kieran.frewen@morsemicro.com>
Date: Thu, 23 Feb 2023 16:25:12 +1300
Subject: wifi: mac80211: S1G capabilities information element in probe request

Add the missing S1G capabilities information element to probe requests.

Signed-off-by: Kieran Frewen <kieran.frewen@morsemicro.com>
Co-developed-by: Gilad Itzkovitch <gilad.itzkovitch@morsemicro.com>
Signed-off-by: Gilad Itzkovitch <gilad.itzkovitch@morsemicro.com>
Link: https://lore.kernel.org/r/20230223032512.3848105-1-gilad.itzkovitch@virscient.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  1 +
 net/mac80211/util.c        | 23 +++++++++++++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 85ecbf57d64e..cdc80285efd4 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2491,6 +2491,7 @@ void ieee80211_add_s1g_capab_ie(struct ieee80211_sub_if_data *sdata,
 				struct sk_buff *skb);
 void ieee80211_add_aid_request_ie(struct ieee80211_sub_if_data *sdata,
 				  struct sk_buff *skb);
+u8 *ieee80211_ie_build_s1g_cap(u8 *pos, struct ieee80211_sta_s1g_cap *s1g_cap);
 
 /* channel management */
 bool ieee80211_chandef_ht_oper(const struct ieee80211_ht_operation *ht_oper,
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 1a28fe5cb614..10ac9819ed91 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -1959,6 +1959,14 @@ static int ieee80211_build_preq_ies_band(struct ieee80211_sub_if_data *sdata,
 	rate_flags = ieee80211_chandef_rate_flags(chandef);
 	shift = ieee80211_chandef_get_shift(chandef);
 
+	/* For direct scan add S1G IE and consider its override bits */
+	if (band == NL80211_BAND_S1GHZ) {
+		if (end - pos < 2 + sizeof(struct ieee80211_s1g_cap))
+			goto out_err;
+		pos = ieee80211_ie_build_s1g_cap(pos, &sband->s1g_cap);
+		goto done;
+	}
+
 	num_rates = 0;
 	for (i = 0; i < sband->n_bitrates; i++) {
 		if ((BIT(i) & rate_mask) == 0)
@@ -3020,6 +3028,21 @@ size_t ieee80211_ie_split_vendor(const u8 *ies, size_t ielen, size_t offset)
 	return pos;
 }
 
+u8 *ieee80211_ie_build_s1g_cap(u8 *pos, struct ieee80211_sta_s1g_cap *s1g_cap)
+{
+	*pos++ = WLAN_EID_S1G_CAPABILITIES;
+	*pos++ = sizeof(struct ieee80211_s1g_cap);
+	memset(pos, 0, sizeof(struct ieee80211_s1g_cap));
+
+	memcpy(pos, &s1g_cap->cap, sizeof(s1g_cap->cap));
+	pos += sizeof(s1g_cap->cap);
+
+	memcpy(pos, &s1g_cap->nss_mcs, sizeof(s1g_cap->nss_mcs));
+	pos += sizeof(s1g_cap->nss_mcs);
+
+	return pos;
+}
+
 u8 *ieee80211_ie_build_ht_cap(u8 *pos, struct ieee80211_sta_ht_cap *ht_cap,
 			      u16 cap)
 {
-- 
cgit v1.2.3-70-g09d2


From 9a8aac92eba90b3b7c71d0531db535f5588388f5 Mon Sep 17 00:00:00 2001
From: Kieran Frewen <kieran.frewen@morsemicro.com>
Date: Fri, 24 Feb 2023 10:29:17 +1300
Subject: wifi: nl80211: support advertising S1G capabilities

Include S1G capabilities in netlink band info messages.

Signed-off-by: Kieran Frewen <kieran.frewen@morsemicro.com>
Co-developed-by: Gilad Itzkovitch <gilad.itzkovitch@morsemicro.com>
Signed-off-by: Gilad Itzkovitch <gilad.itzkovitch@morsemicro.com>
Link: https://lore.kernel.org/r/20230223212917.4010246-1-gilad.itzkovitch@virscient.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/uapi/linux/nl80211.h |  7 +++++++
 net/wireless/nl80211.c       | 10 ++++++++++
 2 files changed, 17 insertions(+)

diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index cf4fb981e131..c59fec406da5 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -4061,6 +4061,10 @@ enum nl80211_band_iftype_attr {
  * @NL80211_BAND_ATTR_EDMG_BW_CONFIG: Channel BW Configuration subfield encodes
  *	the allowed channel bandwidth configurations.
  *	Defined by IEEE P802.11ay/D4.0 section 9.4.2.251, Table 13.
+ * @NL80211_BAND_ATTR_S1G_MCS_NSS_SET: S1G capabilities, supported S1G-MCS and NSS
+ *	set subfield, as in the S1G information IE, 5 bytes
+ * @NL80211_BAND_ATTR_S1G_CAPA: S1G capabilities information subfield as in the
+ *	S1G information IE, 10 bytes
  * @NL80211_BAND_ATTR_MAX: highest band attribute currently defined
  * @__NL80211_BAND_ATTR_AFTER_LAST: internal use
  */
@@ -4081,6 +4085,9 @@ enum nl80211_band_attr {
 	NL80211_BAND_ATTR_EDMG_CHANNELS,
 	NL80211_BAND_ATTR_EDMG_BW_CONFIG,
 
+	NL80211_BAND_ATTR_S1G_MCS_NSS_SET,
+	NL80211_BAND_ATTR_S1G_CAPA,
+
 	/* keep last */
 	__NL80211_BAND_ATTR_AFTER_LAST,
 	NL80211_BAND_ATTR_MAX = __NL80211_BAND_ATTR_AFTER_LAST - 1
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index f1cd3d9130dd..2c9edb015652 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1961,6 +1961,16 @@ static int nl80211_send_band_rateinfo(struct sk_buff *msg,
 
 	nla_nest_end(msg, nl_rates);
 
+	/* S1G capabilities */
+	if (sband->band == NL80211_BAND_S1GHZ && sband->s1g_cap.s1g &&
+	    (nla_put(msg, NL80211_BAND_ATTR_S1G_CAPA,
+		     sizeof(sband->s1g_cap.cap),
+		     sband->s1g_cap.cap) ||
+	     nla_put(msg, NL80211_BAND_ATTR_S1G_MCS_NSS_SET,
+		     sizeof(sband->s1g_cap.nss_mcs),
+		     sband->s1g_cap.nss_mcs)))
+		return -ENOBUFS;
+
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 650cadb730105f1894b6c8afacc57a368dd18b91 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:58:41 +0300
Subject: wifi: iwlwifi: mvm: vif preparation for MLO

In MLO, some fields of iwl_mvm_vif should be defined in the
context of a link. Define a separate structure for these fields and
add a deflink object to hold it as part of iwl_mvm_vif. Non-MLO legacy
code will use only deflink object while MLO related code will use the
corresponding link from the link array.

It follows the strategy applied in mac80211 for introducing MLO
changes.

The below spatch takes care of updating all driver code to access
fields separated into MLD specific data structure via deflink (need
to convert all references to the fields listed in var to deflink.var
and also to take care of calls like iwl_mvm_vif_from_mac80211(vif)->field).

  @iwl_mld_vif@
  struct iwl_mvm_vif *v;
  struct ieee80211_vif *vv;
  identifier fn;
  identifier var = {bssid, ap_sta_id, bcast_sta, mcast_sta,
                    beacon_stats, smps_requests, probe_resp_data,
		    he_ru_2mhz_block, cab_queue, phy_ctxt,
                    queue_params};
  @@

  (
    v->
  -    var
  +    deflink.var
  |
    fn(vv)->
  -    var
  +    deflink.var
  )

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.4896576f0a9f.Ifaf0187c96b9fe52b24bd629331165831a877691@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/binding.c   |  13 +-
 drivers/net/wireless/intel/iwlwifi/mvm/coex.c      |  29 +++--
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c        |  39 +++---
 .../net/wireless/intel/iwlwifi/mvm/debugfs-vif.c   |  14 +--
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c |   6 +-
 .../net/wireless/intel/iwlwifi/mvm/ftm-responder.c |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      |  14 +--
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  |  28 ++---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 138 +++++++++++----------
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c   |  12 +-
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  19 +--
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   |  21 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 121 +++++++++---------
 drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c  |   4 +-
 drivers/net/wireless/intel/iwlwifi/mvm/power.c     |  22 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/quota.c     |  11 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c        |  26 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c      |  15 +--
 drivers/net/wireless/intel/iwlwifi/mvm/sf.c        |   6 +-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       |  78 ++++++------
 .../net/wireless/intel/iwlwifi/mvm/time-event.c    |   6 +-
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c        |   8 +-
 drivers/net/wireless/intel/iwlwifi/mvm/utils.c     |  31 ++---
 23 files changed, 355 insertions(+), 308 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/binding.c b/drivers/net/wireless/intel/iwlwifi/mvm/binding.c
index 0aac306304cb..ef50ccabcc73 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/binding.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/binding.c
@@ -2,6 +2,7 @@
 /*
  * Copyright (C) 2012-2014, 2020 Intel Corporation
  * Copyright (C) 2016 Intel Deutschland GmbH
+ * Copyright (C) 2022 Intel Corporation
  */
 #include <net/mac80211.h>
 #include "fw-api.h"
@@ -75,7 +76,7 @@ static void iwl_mvm_iface_iterator(void *_data, u8 *mac,
 	if (vif == data->ignore_vif)
 		return;
 
-	if (mvmvif->phy_ctxt != data->phyctxt)
+	if (mvmvif->deflink.phy_ctxt != data->phyctxt)
 		return;
 
 	if (WARN_ON_ONCE(data->idx >= MAX_MACS_IN_BINDING))
@@ -132,7 +133,7 @@ int iwl_mvm_binding_add_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
 		return -EINVAL;
 
 	/*
@@ -142,7 +143,8 @@ int iwl_mvm_binding_add_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	if (iwl_mvm_sf_update(mvm, vif, false))
 		return -EINVAL;
 
-	return iwl_mvm_binding_update(mvm, vif, mvmvif->phy_ctxt, true);
+	return iwl_mvm_binding_update(mvm, vif, mvmvif->deflink.phy_ctxt,
+				      true);
 }
 
 int iwl_mvm_binding_remove_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
@@ -150,10 +152,11 @@ int iwl_mvm_binding_remove_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
 
-	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
 		return -EINVAL;
 
-	ret = iwl_mvm_binding_update(mvm, vif, mvmvif->phy_ctxt, false);
+	ret = iwl_mvm_binding_update(mvm, vif, mvmvif->deflink.phy_ctxt,
+				     false);
 
 	if (!ret)
 		if (iwl_mvm_sf_update(mvm, vif, true))
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
index ee3c8a786199..73ef1da3ccac 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
@@ -292,7 +292,8 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 			/* ... relax constraints and disable rssi events */
 			iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
 					    smps_mode);
-			iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id,
+			iwl_mvm_bt_coex_reduced_txp(mvm,
+						    mvmvif->deflink.ap_sta_id,
 						    false);
 			iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
 		}
@@ -314,8 +315,8 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	if (!vif->cfg.assoc)
 		smps_mode = IEEE80211_SMPS_AUTOMATIC;
 
-	if (mvmvif->phy_ctxt &&
-	    (mvm->last_bt_notif.rrc_status & BIT(mvmvif->phy_ctxt->id)))
+	if (mvmvif->deflink.phy_ctxt &&
+	    (mvm->last_bt_notif.rrc_status & BIT(mvmvif->deflink.phy_ctxt->id)))
 		smps_mode = IEEE80211_SMPS_AUTOMATIC;
 
 	IWL_DEBUG_COEX(data->mvm,
@@ -384,7 +385,8 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	if (iwl_get_coex_type(mvm, vif) == BT_COEX_LOOSE_LUT ||
 	    mvm->cfg->bt_shared_single_ant || !vif->cfg.assoc ||
 	    le32_to_cpu(mvm->last_bt_notif.bt_activity_grading) == BT_OFF) {
-		iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id, false);
+		iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->deflink.ap_sta_id,
+					    false);
 		iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
 		return;
 	}
@@ -396,10 +398,12 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	if (!ave_rssi)
 		ave_rssi = -100;
 	if (ave_rssi > -IWL_MVM_BT_COEX_EN_RED_TXP_THRESH) {
-		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id, true))
+		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->deflink.ap_sta_id,
+						true))
 			IWL_ERR(mvm, "Couldn't send BT_CONFIG cmd\n");
 	} else if (ave_rssi < -IWL_MVM_BT_COEX_DIS_RED_TXP_THRESH) {
-		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id, false))
+		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->deflink.ap_sta_id,
+						false))
 			IWL_ERR(mvm, "Couldn't send BT_CONFIG cmd\n");
 	}
 
@@ -521,7 +525,7 @@ void iwl_mvm_bt_rssi_event(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	 * Rssi update while not associated - can happen since the statistics
 	 * are handled asynchronously
 	 */
-	if (mvmvif->ap_sta_id == IWL_MVM_INVALID_STA)
+	if (mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA)
 		return;
 
 	/* No BT - reports should be disabled */
@@ -537,10 +541,13 @@ void iwl_mvm_bt_rssi_event(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	 */
 	if (rssi_event == RSSI_EVENT_LOW || mvm->cfg->bt_shared_single_ant ||
 	    iwl_get_coex_type(mvm, vif) == BT_COEX_LOOSE_LUT)
-		ret = iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id,
+		ret = iwl_mvm_bt_coex_reduced_txp(mvm,
+						  mvmvif->deflink.ap_sta_id,
 						  false);
 	else
-		ret = iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->ap_sta_id, true);
+		ret = iwl_mvm_bt_coex_reduced_txp(mvm,
+						  mvmvif->deflink.ap_sta_id,
+						  true);
 
 	if (ret)
 		IWL_ERR(mvm, "couldn't send BT_CONFIG HCMD upon RSSI event\n");
@@ -554,7 +561,7 @@ u16 iwl_mvm_coex_agg_time_limit(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(mvmsta->vif);
-	struct iwl_mvm_phy_ctxt *phy_ctxt = mvmvif->phy_ctxt;
+	struct iwl_mvm_phy_ctxt *phy_ctxt = mvmvif->deflink.phy_ctxt;
 	enum iwl_bt_coex_lut_type lut_type;
 
 	if (mvm->last_bt_notif.ttc_status & BIT(phy_ctxt->id))
@@ -578,7 +585,7 @@ bool iwl_mvm_bt_coex_is_mimo_allowed(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(mvmsta->vif);
-	struct iwl_mvm_phy_ctxt *phy_ctxt = mvmvif->phy_ctxt;
+	struct iwl_mvm_phy_ctxt *phy_ctxt = mvmvif->deflink.phy_ctxt;
 	enum iwl_bt_coex_lut_type lut_type;
 
 	if (mvm->last_bt_notif.ttc_status & BIT(phy_ctxt->id))
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index c5ad34b063df..40adf789c8fc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -470,7 +470,7 @@ static int iwl_mvm_wowlan_config_rsc_tsc(struct iwl_mvm *mvm,
 		for (i = 0; i < ARRAY_SIZE(data.rsc->mcast_key_id_map); i++)
 			data.rsc->mcast_key_id_map[i] =
 				IWL_MCAST_KEY_MAP_INVALID;
-		data.rsc->sta_id = cpu_to_le32(mvmvif->ap_sta_id);
+		data.rsc->sta_id = cpu_to_le32(mvmvif->deflink.ap_sta_id);
 
 		ieee80211_iter_keys(mvm->hw, vif,
 				    iwl_mvm_wowlan_get_rsc_v5_data,
@@ -493,7 +493,8 @@ static int iwl_mvm_wowlan_config_rsc_tsc(struct iwl_mvm *mvm,
 
 		if (ver == 4) {
 			size = sizeof(*data.rsc_tsc);
-			data.rsc_tsc->sta_id = cpu_to_le32(mvmvif->ap_sta_id);
+			data.rsc_tsc->sta_id =
+				cpu_to_le32(mvmvif->deflink.ap_sta_id);
 		} else {
 			/* ver == 2 || ver == IWL_FW_CMD_VER_UNKNOWN */
 			size = sizeof(data.rsc_tsc->params);
@@ -691,7 +692,7 @@ static int iwl_mvm_send_patterns(struct iwl_mvm *mvm,
 
 	pattern_cmd->n_patterns = wowlan->n_patterns;
 	if (ver >= 3)
-		pattern_cmd->sta_id = mvmvif->ap_sta_id;
+		pattern_cmd->sta_id = mvmvif->deflink.ap_sta_id;
 
 	for (i = 0; i < wowlan->n_patterns; i++) {
 		int mask_len = DIV_ROUND_UP(wowlan->patterns[i].pattern_len, 8);
@@ -732,7 +733,7 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		return -EINVAL;
 
 	/* add back the PHY */
-	if (WARN_ON(!mvmvif->phy_ctxt))
+	if (WARN_ON(!mvmvif->deflink.phy_ctxt))
 		return -EINVAL;
 
 	rcu_read_lock();
@@ -746,7 +747,7 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	chains_dynamic = ctx->rx_chains_dynamic;
 	rcu_read_unlock();
 
-	ret = iwl_mvm_phy_ctxt_add(mvm, mvmvif->phy_ctxt, &chandef,
+	ret = iwl_mvm_phy_ctxt_add(mvm, mvmvif->deflink.phy_ctxt, &chandef,
 				   chains_static, chains_dynamic);
 	if (ret)
 		return ret;
@@ -763,12 +764,12 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	/* add back binding - XXX refactor? */
 	binding_cmd.id_and_color =
-		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->phy_ctxt->id,
-						mvmvif->phy_ctxt->color));
+		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->deflink.phy_ctxt->id,
+						mvmvif->deflink.phy_ctxt->color));
 	binding_cmd.action = cpu_to_le32(FW_CTXT_ACTION_ADD);
 	binding_cmd.phy =
-		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->phy_ctxt->id,
-						mvmvif->phy_ctxt->color));
+		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->deflink.phy_ctxt->id,
+						mvmvif->deflink.phy_ctxt->color));
 	binding_cmd.macs[0] = cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->id,
 							      mvmvif->color));
 	for (i = 1; i < MAX_MACS_IN_BINDING; i++)
@@ -791,7 +792,8 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	ret = iwl_mvm_sta_send_to_fw(mvm, ap_sta, false, 0);
 	if (ret)
 		return ret;
-	rcu_assign_pointer(mvm->fw_id_to_mac_id[mvmvif->ap_sta_id], ap_sta);
+	rcu_assign_pointer(mvm->fw_id_to_mac_id[mvmvif->deflink.ap_sta_id],
+			   ap_sta);
 
 	ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
 	if (ret)
@@ -800,8 +802,8 @@ static int iwl_mvm_d3_reprogram(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	/* and some quota */
 	quota = iwl_mvm_quota_cmd_get_quota(mvm, &quota_cmd, 0);
 	quota->id_and_color =
-		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->phy_ctxt->id,
-						mvmvif->phy_ctxt->color));
+		cpu_to_le32(FW_CMD_ID_AND_COLOR(mvmvif->deflink.phy_ctxt->id,
+						mvmvif->deflink.phy_ctxt->color));
 	quota->quota = cpu_to_le32(IWL_MVM_MAX_QUOTA);
 	quota->max_duration = cpu_to_le32(IWL_MVM_MAX_QUOTA);
 
@@ -1027,7 +1029,7 @@ static int iwl_mvm_wowlan_config_key_params(struct iwl_mvm *mvm,
 		if (ver == 2) {
 			size = sizeof(tkip_data.tkip);
 			tkip_data.tkip.sta_id =
-				cpu_to_le32(mvmvif->ap_sta_id);
+				cpu_to_le32(mvmvif->deflink.ap_sta_id);
 		} else if (ver == 1 || ver == IWL_FW_CMD_VER_UNKNOWN) {
 			size = sizeof(struct iwl_wowlan_tkip_params_cmd_ver_1);
 		} else {
@@ -1076,7 +1078,7 @@ static int iwl_mvm_wowlan_config_key_params(struct iwl_mvm *mvm,
 		kek_kck_cmd.kek_len = cpu_to_le16(mvmvif->rekey_data.kek_len);
 		kek_kck_cmd.replay_ctr = mvmvif->rekey_data.replay_ctr;
 		kek_kck_cmd.akm = cpu_to_le32(mvmvif->rekey_data.akm);
-		kek_kck_cmd.sta_id = cpu_to_le32(mvmvif->ap_sta_id);
+		kek_kck_cmd.sta_id = cpu_to_le32(mvmvif->deflink.ap_sta_id);
 
 		if (cmd_ver == 4) {
 			cmd_size = sizeof(struct iwl_wowlan_kek_kck_material_cmd_v4);
@@ -1269,7 +1271,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 
 	mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	if (mvmvif->ap_sta_id == IWL_MVM_INVALID_STA) {
+	if (mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA) {
 		/* if we're not associated, this must be netdetect */
 		if (!wowlan->nd_config) {
 			ret = 1;
@@ -1285,10 +1287,10 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 	} else {
 		struct iwl_wowlan_config_cmd wowlan_config_cmd = {};
 
-		wowlan_config_cmd.sta_id = mvmvif->ap_sta_id;
+		wowlan_config_cmd.sta_id = mvmvif->deflink.ap_sta_id;
 
 		ap_sta = rcu_dereference_protected(
-			mvm->fw_id_to_mac_id[mvmvif->ap_sta_id],
+			mvm->fw_id_to_mac_id[mvmvif->deflink.ap_sta_id],
 			lockdep_is_held(&mvm->mutex));
 		if (IS_ERR_OR_NULL(ap_sta)) {
 			ret = -EINVAL;
@@ -2575,7 +2577,8 @@ iwl_mvm_choose_query_wakeup_reasons(struct iwl_mvm *mvm,
 	/* if FW uses status notification, status shouldn't be NULL here */
 	if (!d3_data->status) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-		u8 sta_id = mvm->net_detect ? IWL_MVM_INVALID_STA : mvmvif->ap_sta_id;
+		u8 sta_id = mvm->net_detect ? IWL_MVM_INVALID_STA :
+					      mvmvif->deflink.ap_sta_id;
 
 		d3_data->status = iwl_mvm_send_wowlan_get_status(mvm, sta_id);
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
index 3779ac040ba0..3613b1fdc5d9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
@@ -179,7 +179,7 @@ static ssize_t iwl_dbgfs_mac_params_read(struct file *file,
 
 	mutex_lock(&mvm->mutex);
 
-	ap_sta_id = mvmvif->ap_sta_id;
+	ap_sta_id = mvmvif->deflink.ap_sta_id;
 
 	switch (ieee80211_vif_type_p2p(vif)) {
 	case NL80211_IFTYPE_ADHOC:
@@ -211,14 +211,14 @@ static ssize_t iwl_dbgfs_mac_params_read(struct file *file,
 	pos += scnprintf(buf+pos, bufsz-pos, "Load: %d\n",
 			 mvm->tcm.result.load[mvmvif->id]);
 	pos += scnprintf(buf+pos, bufsz-pos, "QoS:\n");
-	for (i = 0; i < ARRAY_SIZE(mvmvif->queue_params); i++)
+	for (i = 0; i < ARRAY_SIZE(mvmvif->deflink.queue_params); i++)
 		pos += scnprintf(buf+pos, bufsz-pos,
 				 "\t%d: txop:%d - cw_min:%d - cw_max = %d - aifs = %d upasd = %d\n",
-				 i, mvmvif->queue_params[i].txop,
-				 mvmvif->queue_params[i].cw_min,
-				 mvmvif->queue_params[i].cw_max,
-				 mvmvif->queue_params[i].aifs,
-				 mvmvif->queue_params[i].uapsd);
+				 i, mvmvif->deflink.queue_params[i].txop,
+				 mvmvif->deflink.queue_params[i].cw_min,
+				 mvmvif->deflink.queue_params[i].cw_max,
+				 mvmvif->deflink.queue_params[i].aifs,
+				 mvmvif->deflink.queue_params[i].uapsd);
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
 	    ap_sta_id != IWL_MVM_INVALID_STA) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 05f3136b1c43..379da4bec5dd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -73,7 +73,7 @@ int iwl_mvm_ftm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		struct ieee80211_sta *sta;
 
 		rcu_read_lock();
-		sta = rcu_dereference(mvm->fw_id_to_mac_id[mvmvif->ap_sta_id]);
+		sta = rcu_dereference(mvm->fw_id_to_mac_id[mvmvif->deflink.ap_sta_id]);
 		if (!IS_ERR_OR_NULL(sta) && sta->mfp)
 			expected_tk_len = 0;
 		rcu_read_unlock();
@@ -510,13 +510,13 @@ iwl_mvm_ftm_put_target(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 		rcu_read_lock();
 
-		sta = rcu_dereference(mvm->fw_id_to_mac_id[mvmvif->ap_sta_id]);
+		sta = rcu_dereference(mvm->fw_id_to_mac_id[mvmvif->deflink.ap_sta_id]);
 		if (sta->mfp && (peer->ftm.trigger_based || peer->ftm.non_trigger_based))
 			FTM_PUT_FLAG(PMF);
 
 		rcu_read_unlock();
 
-		target->sta_id = mvmvif->ap_sta_id;
+		target->sta_id = mvmvif->deflink.ap_sta_id;
 	} else {
 		target->sta_id = IWL_MVM_INVALID_STA;
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
index e862d1b43f21..c37d793d6a4a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
@@ -119,7 +119,7 @@ iwl_mvm_ftm_responder_cmd(struct iwl_mvm *mvm,
 			cpu_to_le32(IWL_TOF_RESPONDER_CMD_VALID_CHAN_INFO |
 				    IWL_TOF_RESPONDER_CMD_VALID_BSSID |
 				    IWL_TOF_RESPONDER_CMD_VALID_STA_ID),
-		.sta_id = mvmvif->bcast_sta.sta_id,
+		.sta_id = mvmvif->deflink.bcast_sta.sta_id,
 	};
 	u8 cmd_ver = iwl_fw_lookup_cmd_ver(mvm->fw, cmd_id, 6);
 	int err;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 2688cb49c951..21e45f89c5f9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -25,7 +25,7 @@ int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_link_config_cmd cmd = {};
 
-	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
 		return -EINVAL;
 
 	/* Update SF - Disable if needed. if this fails, SF might still be on
@@ -34,9 +34,9 @@ int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	if (iwl_mvm_sf_update(mvm, vif, false))
 		return -EINVAL;
 
-	cmd.link_id = cpu_to_le32(mvmvif->phy_ctxt->id);
+	cmd.link_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
-	cmd.phy_id = cpu_to_le32(mvmvif->phy_ctxt->id);
+	cmd.phy_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
 
 	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
 
@@ -47,7 +47,7 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			 u32 changes, bool active)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_phy_ctxt *phyctxt = mvmvif->phy_ctxt;
+	struct iwl_mvm_phy_ctxt *phyctxt = mvmvif->deflink.phy_ctxt;
 	struct iwl_link_config_cmd cmd = {};
 	u32 ht_flag, flags = 0, flags_mask = 0;
 
@@ -136,7 +136,7 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	cmd.frame_time_rts_th = cpu_to_le16(vif->bss_conf.frame_time_rts_th);
 
 	/* Block 26-tone RU OFDMA transmissions */
-	if (mvmvif->he_ru_2mhz_block) {
+	if (mvmvif->deflink.he_ru_2mhz_block) {
 		flags |= LINK_FLG_RU_2MHZ_BLOCK;
 		flags_mask |= LINK_FLG_RU_2MHZ_BLOCK;
 	}
@@ -161,10 +161,10 @@ int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	struct iwl_link_config_cmd cmd = {};
 	int ret;
 
-	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
 		return -EINVAL;
 
-	cmd.link_id = cpu_to_le32(mvmvif->phy_ctxt->id);
+	cmd.link_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
 	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_REMOVE);
 
 	if (!ret)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 4d7ad91c53d4..10dc1fdbdeb3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -293,15 +293,15 @@ int iwl_mvm_mac_ctxt_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		 * For TVQM this will be overwritten later with the FW assigned
 		 * queue value (when queue is enabled).
 		 */
-		mvmvif->cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
+		mvmvif->deflink.cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
 	}
 
-	mvmvif->bcast_sta.sta_id = IWL_MVM_INVALID_STA;
-	mvmvif->mcast_sta.sta_id = IWL_MVM_INVALID_STA;
-	mvmvif->ap_sta_id = IWL_MVM_INVALID_STA;
+	mvmvif->deflink.bcast_sta.sta_id = IWL_MVM_INVALID_STA;
+	mvmvif->deflink.mcast_sta.sta_id = IWL_MVM_INVALID_STA;
+	mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
 
 	for (i = 0; i < NUM_IWL_MVM_SMPS_REQ; i++)
-		mvmvif->smps_requests[i] = IEEE80211_SMPS_AUTOMATIC;
+		mvmvif->deflink.smps_requests[i] = IEEE80211_SMPS_AUTOMATIC;
 
 	return 0;
 
@@ -469,12 +469,12 @@ void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		u8 ucode_ac = iwl_mvm_mac80211_ac_to_ucode_ac(i);
 
 		ac[ucode_ac].cw_min =
-			cpu_to_le16(mvmvif->queue_params[i].cw_min);
+			cpu_to_le16(mvmvif->deflink.queue_params[i].cw_min);
 		ac[ucode_ac].cw_max =
-			cpu_to_le16(mvmvif->queue_params[i].cw_max);
+			cpu_to_le16(mvmvif->deflink.queue_params[i].cw_max);
 		ac[ucode_ac].edca_txop =
-			cpu_to_le16(mvmvif->queue_params[i].txop * 32);
-		ac[ucode_ac].aifsn = mvmvif->queue_params[i].aifs;
+			cpu_to_le16(mvmvif->deflink.queue_params[i].txop * 32);
+		ac[ucode_ac].aifsn = mvmvif->deflink.queue_params[i].aifs;
 		ac[ucode_ac].fifos_mask = BIT(txf);
 	}
 
@@ -944,7 +944,7 @@ static void iwl_mvm_mac_ctxt_set_tx(struct iwl_mvm *mvm,
 
 	/* Set up TX command fields */
 	tx->len = cpu_to_le16((u16)beacon->len);
-	tx->sta_id = mvmvif->bcast_sta.sta_id;
+	tx->sta_id = mvmvif->deflink.bcast_sta.sta_id;
 	tx->life_time = cpu_to_le32(TX_CMD_LIFE_TIME_INFINITE);
 	tx_flags = TX_CMD_FLG_SEQ_CTL | TX_CMD_FLG_TSF;
 	tx_flags |=
@@ -1214,7 +1214,7 @@ static void iwl_mvm_mac_ctxt_cmd_fill_ap(struct iwl_mvm *mvm,
 
 	if (!fw_has_api(&mvm->fw->ucode_capa,
 			IWL_UCODE_TLV_API_STA_TYPE))
-		ctxt_ap->mcast_qid = cpu_to_le32(mvmvif->cab_queue);
+		ctxt_ap->mcast_qid = cpu_to_le32(mvmvif->deflink.cab_queue);
 
 	/*
 	 * Only set the beacon time when the MAC is being added, when we
@@ -1663,9 +1663,9 @@ void iwl_mvm_probe_resp_data_notif(struct iwl_mvm *mvm,
 	    sizeof(struct ieee80211_p2p_noa_desc) + 2)
 		new_data->noa_len -= sizeof(struct ieee80211_p2p_noa_desc);
 
-	old_data = rcu_dereference_protected(mvmvif->probe_resp_data,
-					lockdep_is_held(&mvmvif->mvm->mutex));
-	rcu_assign_pointer(mvmvif->probe_resp_data, new_data);
+	old_data = rcu_dereference_protected(mvmvif->deflink.probe_resp_data,
+					     lockdep_is_held(&mvmvif->mvm->mutex));
+	rcu_assign_pointer(mvmvif->deflink.probe_resp_data, new_data);
 
 	if (old_data)
 		kfree_rcu(old_data, rcu_head);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 33582f4eea91..163515db6828 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -736,7 +736,7 @@ static void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
 	    !offchannel) {
 		struct iwl_mvm_vif *mvmvif =
 			iwl_mvm_vif_from_mac80211(info->control.vif);
-		u8 ap_sta_id = READ_ONCE(mvmvif->ap_sta_id);
+		u8 ap_sta_id = READ_ONCE(mvmvif->deflink.ap_sta_id);
 
 		if (ap_sta_id < mvm->fw->ucode_capa.num_stations) {
 			/* mac80211 holds rcu read lock */
@@ -921,7 +921,7 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 
 	switch (action) {
 	case IEEE80211_AMPDU_RX_START:
-		if (iwl_mvm_vif_from_mac80211(vif)->ap_sta_id ==
+		if (iwl_mvm_vif_from_mac80211(vif)->deflink.ap_sta_id ==
 				iwl_mvm_sta_from_mac80211(sta)->sta_id) {
 			struct iwl_mvm_vif *mvmvif;
 			u16 macid = iwl_mvm_vif_from_mac80211(vif)->id;
@@ -987,15 +987,16 @@ static void iwl_mvm_cleanup_iterator(void *data, u8 *mac,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
 	mvmvif->uploaded = false;
-	mvmvif->ap_sta_id = IWL_MVM_INVALID_STA;
+	mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
 
 	spin_lock_bh(&mvm->time_event_lock);
 	iwl_mvm_te_clear_data(mvm, &mvmvif->time_event_data);
 	spin_unlock_bh(&mvm->time_event_lock);
 
-	mvmvif->phy_ctxt = NULL;
+	mvmvif->deflink.phy_ctxt = NULL;
 	memset(&mvmvif->bf_data, 0, sizeof(mvmvif->bf_data));
-	memset(&mvmvif->probe_resp_data, 0, sizeof(mvmvif->probe_resp_data));
+	memset(&mvmvif->deflink.probe_resp_data, 0,
+	       sizeof(mvmvif->deflink.probe_resp_data));
 }
 
 static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm)
@@ -1330,7 +1331,7 @@ static int iwl_mvm_post_channel_switch(struct ieee80211_hw *hw,
 
 		mvmvif->csa_bcn_pending = false;
 		mvmsta = iwl_mvm_sta_from_staid_protected(mvm,
-							  mvmvif->ap_sta_id);
+							  mvmvif->deflink.ap_sta_id);
 
 		if (WARN_ON(!mvmsta)) {
 			ret = -EIO;
@@ -1447,7 +1448,10 @@ bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
 	lockdep_assert_held(&mvm->mutex);
 
 	mvmvif->mvm = mvm;
-	RCU_INIT_POINTER(mvmvif->probe_resp_data, NULL);
+	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
+
+	/* the first link always points to the default one */
+	mvmvif->link[0] = &mvmvif->deflink;
 
 	/*
 	 * Not much to do here. The stack will not allow interface
@@ -1457,8 +1461,8 @@ bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
 
 	/* make sure that beacon statistics don't go backwards with FW reset */
 	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
-		mvmvif->beacon_stats.accu_num_beacons +=
-			mvmvif->beacon_stats.num_beacons;
+		mvmvif->deflink.beacon_stats.accu_num_beacons +=
+			mvmvif->deflink.beacon_stats.num_beacons;
 
 	/* Allocate resources for the MAC context, and add it to the fw  */
 	*ret = iwl_mvm_mac_ctxt_init(mvm, vif);
@@ -1485,6 +1489,7 @@ bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
 	}
 
 	mvmvif->features |= hw->netdev_features;
+
 	return false;
 }
 
@@ -1506,7 +1511,8 @@ static int iwl_mvm_alloc_bcast_mcast_sta(struct iwl_mvm *mvm,
 	 * Only queue for this station is the mcast queue,
 	 * which shouldn't be in TFD mask anyway
 	 */
-	return iwl_mvm_allocate_int_sta(mvm, &mvmvif->mcast_sta, 0, vif->type,
+	return iwl_mvm_allocate_int_sta(mvm, &mvmvif->deflink.mcast_sta, 0,
+					vif->type,
 					IWL_STA_MULTICAST);
 }
 
@@ -1550,13 +1556,13 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 	 */
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 
-		mvmvif->phy_ctxt = iwl_mvm_get_free_phy_ctxt(mvm);
-		if (!mvmvif->phy_ctxt) {
+		mvmvif->deflink.phy_ctxt = iwl_mvm_get_free_phy_ctxt(mvm);
+		if (!mvmvif->deflink.phy_ctxt) {
 			ret = -ENOSPC;
 			goto out_free_bf;
 		}
 
-		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->phy_ctxt);
+		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
 		ret = iwl_mvm_binding_add_vif(mvm, vif);
 		if (ret)
 			goto out_unref_phy;
@@ -1600,7 +1606,7 @@ out:
  out_unbind:
 	iwl_mvm_binding_remove_vif(mvm, vif);
  out_unref_phy:
-	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
+	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
  out_free_bf:
 	if (mvm->bf_allowed_vif == mvmvif) {
 		mvm->bf_allowed_vif = NULL;
@@ -1608,7 +1614,7 @@ out:
 				       IEEE80211_VIF_SUPPORTS_CQM_RSSI);
 	}
  out_remove_mac:
-	mvmvif->phy_ctxt = NULL;
+	mvmvif->deflink.phy_ctxt = NULL;
 	iwl_mvm_mac_ctxt_remove(mvm, vif);
  out_unlock:
 	mutex_unlock(&mvm->mutex);
@@ -1653,9 +1659,9 @@ bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
 		mvm->csme_vif = NULL;
 	}
 
-	probe_data = rcu_dereference_protected(mvmvif->probe_resp_data,
+	probe_data = rcu_dereference_protected(mvmvif->deflink.probe_resp_data,
 					       lockdep_is_held(&mvm->mutex));
-	RCU_INIT_POINTER(mvmvif->probe_resp_data, NULL);
+	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
 	if (probe_data)
 		kfree_rcu(probe_data, rcu_head);
 
@@ -1702,8 +1708,8 @@ static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
 		mvm->p2p_device_vif = NULL;
 		iwl_mvm_rm_p2p_bcast_sta(mvm, vif);
 		iwl_mvm_binding_remove_vif(mvm, vif);
-		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
-		mvmvif->phy_ctxt = NULL;
+		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
+		mvmvif->deflink.phy_ctxt = NULL;
 	}
 
 	iwl_mvm_mac_ctxt_remove(mvm, vif);
@@ -1716,7 +1722,7 @@ static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
 out:
 	if (vif->type == NL80211_IFTYPE_AP ||
 	    vif->type == NL80211_IFTYPE_ADHOC) {
-		iwl_mvm_dealloc_int_sta(mvm, &mvmvif->mcast_sta);
+		iwl_mvm_dealloc_int_sta(mvm, &mvmvif->deflink.mcast_sta);
 		iwl_mvm_dealloc_bcast_sta(mvm, vif);
 	}
 
@@ -2207,10 +2213,10 @@ bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
 
 	for (i = 0; i < IEEE80211_NUM_ACS; i++) {
 		struct ieee80211_he_mu_edca_param_ac_rec *mu_edca =
-			&mvmvif->queue_params[i].mu_edca_param_rec;
+			&mvmvif->deflink.queue_params[i].mu_edca_param_rec;
 		u8 ac = iwl_mvm_mac80211_ac_to_ucode_ac(i);
 
-		if (!mvmvif->queue_params[i].mu_edca) {
+		if (!mvmvif->deflink.queue_params[i].mu_edca) {
 			mu_edca_enabled = false;
 			break;
 		}
@@ -2335,7 +2341,7 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 	flags = 0;
 
 	/* Block 26-tone RU OFDMA transmissions */
-	if (mvmvif->he_ru_2mhz_block)
+	if (mvmvif->deflink.he_ru_2mhz_block)
 		flags |= STA_CTXT_HE_RU_2MHZ_BLOCK;
 
 	/* HTC flags */
@@ -2474,7 +2480,7 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 		     !iwlwifi_mod_params.disable_11ax) ||
 		    (vif->bss_conf.eht_support &&
 		     !iwlwifi_mod_params.disable_11be))
-			iwl_mvm_cfg_he_sta(mvm, vif, mvmvif->ap_sta_id);
+			iwl_mvm_cfg_he_sta(mvm, vif, mvmvif->deflink.ap_sta_id);
 
 		iwl_mvm_mac_ctxt_recalc_tsf_id(mvm, vif);
 	}
@@ -2486,7 +2492,7 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 	      !iwlwifi_mod_params.disable_11ax) ||
 	     (vif->bss_conf.eht_support &&
 	      !iwlwifi_mod_params.disable_11be)))
-		iwl_mvm_cfg_he_sta(mvm, vif, mvmvif->ap_sta_id);
+		iwl_mvm_cfg_he_sta(mvm, vif, mvmvif->deflink.ap_sta_id);
 
 	/*
 	 * If we're not associated yet, take the (new) BSSID before associating
@@ -2495,22 +2501,22 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 	 * branch for disassociation below.
 	 */
 	if (changes & BSS_CHANGED_BSSID && !mvmvif->associated)
-		memcpy(mvmvif->bssid, bss_conf->bssid, ETH_ALEN);
+		memcpy(mvmvif->deflink.bssid, bss_conf->bssid, ETH_ALEN);
 
-	ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false, mvmvif->bssid);
+	ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false, mvmvif->deflink.bssid);
 	if (ret)
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
 
 	/* after sending it once, adopt mac80211 data */
-	memcpy(mvmvif->bssid, bss_conf->bssid, ETH_ALEN);
+	memcpy(mvmvif->deflink.bssid, bss_conf->bssid, ETH_ALEN);
 	mvmvif->associated = vif->cfg.assoc;
 
 	if (changes & BSS_CHANGED_ASSOC) {
 		if (vif->cfg.assoc) {
 			/* clear statistics to get clean beacon counter */
 			iwl_mvm_request_statistics(mvm, true);
-			memset(&mvmvif->beacon_stats, 0,
-			       sizeof(mvmvif->beacon_stats));
+			memset(&mvmvif->deflink.beacon_stats, 0,
+			       sizeof(mvmvif->deflink.beacon_stats));
 
 			/* add quota for this interface */
 			ret = iwl_mvm_update_quotas(mvm, true, NULL);
@@ -2567,7 +2573,7 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 						    IWL_MVM_SMPS_REQ_PROT,
 						    IEEE80211_SMPS_DYNAMIC);
 			}
-		} else if (mvmvif->ap_sta_id != IWL_MVM_INVALID_STA) {
+		} else if (mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
 			iwl_mvm_mei_host_disassociated(mvm);
 			/*
 			 * If update fails - SF might be running in associated
@@ -2597,12 +2603,12 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 				 * the MAC is unassoc
 				 */
 				ret = iwl_mvm_rm_sta_id(mvm, vif,
-							mvmvif->ap_sta_id);
+							mvmvif->deflink.ap_sta_id);
 				if (ret)
 					IWL_ERR(mvm,
 						"failed to remove AP station\n");
 
-				mvmvif->ap_sta_id = IWL_MVM_INVALID_STA;
+				mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
 			}
 
 			/* remove quota for this interface */
@@ -3305,7 +3311,7 @@ static void iwl_mvm_check_he_obss_narrow_bw_ru(struct ieee80211_hw *hw,
 	};
 
 	if (!(vif->bss_conf.chandef.chan->flags & IEEE80211_CHAN_RADAR)) {
-		mvmvif->he_ru_2mhz_block = false;
+		mvmvif->deflink.he_ru_2mhz_block = false;
 		return;
 	}
 
@@ -3317,7 +3323,7 @@ static void iwl_mvm_check_he_obss_narrow_bw_ru(struct ieee80211_hw *hw,
 	 * If there is at least one AP on radar channel that cannot
 	 * tolerate 26-tone RU UL OFDMA transmissions using HE TB PPDU.
 	 */
-	mvmvif->he_ru_2mhz_block = !iter_data.tolerated;
+	mvmvif->deflink.he_ru_2mhz_block = !iter_data.tolerated;
 }
 
 static void iwl_mvm_reset_cca_40mhz_workaround(struct iwl_mvm *mvm,
@@ -3434,7 +3440,7 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 			   sta->addr, old_state, new_state);
 
 	/* this would be a mac80211 bug ... but don't crash */
-	if (WARN_ON_ONCE(!mvmvif->phy_ctxt))
+	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
 		return test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED, &mvm->status) ? 0 : -EINVAL;
 
 	/*
@@ -3532,14 +3538,15 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 		} else if (vif->type == NL80211_IFTYPE_STATION) {
 			vif->bss_conf.he_support = sta->deflink.he_cap.has_he;
 
-			mvmvif->he_ru_2mhz_block = false;
+			mvmvif->deflink.he_ru_2mhz_block = false;
 			if (sta->deflink.he_cap.has_he)
 				iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif);
 
 			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
 		}
 
-		iwl_mvm_rs_rate_init(mvm, sta, mvmvif->phy_ctxt->channel->band,
+		iwl_mvm_rs_rate_init(mvm, sta,
+				     mvmvif->deflink.phy_ctxt->channel->band,
 				     false);
 		ret = iwl_mvm_update_sta(mvm, vif, sta);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
@@ -3563,14 +3570,16 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 			iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
 		}
 
-		iwl_mvm_rs_rate_init(mvm, sta, mvmvif->phy_ctxt->channel->band,
+		iwl_mvm_rs_rate_init(mvm, sta,
+				     mvmvif->deflink.phy_ctxt->channel->band,
 				     true);
 	} else if (old_state == IEEE80211_STA_AUTHORIZED &&
 		   new_state == IEEE80211_STA_ASSOC) {
 		/* once we move into assoc state, need to update rate scale to
 		 * disable using wide bandwidth
 		 */
-		iwl_mvm_rs_rate_init(mvm, sta, mvmvif->phy_ctxt->channel->band,
+		iwl_mvm_rs_rate_init(mvm, sta,
+				     mvmvif->deflink.phy_ctxt->channel->band,
 				     false);
 		if (!sta->tdls) {
 			/*
@@ -3650,7 +3659,8 @@ static void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw,
 	if (changed & (IEEE80211_RC_BW_CHANGED |
 		       IEEE80211_RC_SUPP_RATES_CHANGED |
 		       IEEE80211_RC_NSS_CHANGED))
-		iwl_mvm_rs_rate_init(mvm, sta, mvmvif->phy_ctxt->channel->band,
+		iwl_mvm_rs_rate_init(mvm, sta,
+				     mvmvif->deflink.phy_ctxt->channel->band,
 				     true);
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
@@ -3666,7 +3676,7 @@ static int iwl_mvm_mac_conf_tx(struct ieee80211_hw *hw,
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	mvmvif->queue_params[ac] = *params;
+	mvmvif->deflink.queue_params[ac] = *params;
 
 	/*
 	 * No need to update right away, we'll get BSS_CHANGED_QOS
@@ -4237,7 +4247,7 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 
 	for (i = 0; i < NUM_PHY_CTX; i++) {
 		phy_ctxt = &mvm->phy_ctxts[i];
-		if (phy_ctxt->ref == 0 || mvmvif->phy_ctxt == phy_ctxt)
+		if (phy_ctxt->ref == 0 || mvmvif->deflink.phy_ctxt == phy_ctxt)
 			continue;
 
 		if (phy_ctxt->ref && channel == phy_ctxt->channel) {
@@ -4249,22 +4259,22 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 			if (WARN(ret, "Failed unbinding P2P_DEVICE\n"))
 				goto out_unlock;
 
-			iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
+			iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
 
 			/* Bind the P2P_DEVICE to the current PHY Context */
-			mvmvif->phy_ctxt = phy_ctxt;
+			mvmvif->deflink.phy_ctxt = phy_ctxt;
 
 			ret = iwl_mvm_binding_add_vif(mvm, vif);
 			if (WARN(ret, "Failed binding P2P_DEVICE\n"))
 				goto out_unlock;
 
-			iwl_mvm_phy_ctxt_ref(mvm, mvmvif->phy_ctxt);
+			iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
 			goto schedule_time_event;
 		}
 	}
 
 	/* Need to update the PHY context only if the ROC channel changed */
-	if (channel == mvmvif->phy_ctxt->channel)
+	if (channel == mvmvif->deflink.phy_ctxt->channel)
 		goto schedule_time_event;
 
 	cfg80211_chandef_create(&chandef, channel, NL80211_CHAN_NO_HT);
@@ -4278,14 +4288,14 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 	band_change_removal =
 		fw_has_capa(&mvm->fw->ucode_capa,
 			    IWL_UCODE_TLV_CAPA_BINDING_CDB_SUPPORT) &&
-		mvmvif->phy_ctxt->channel->band != chandef.chan->band;
+		mvmvif->deflink.phy_ctxt->channel->band != chandef.chan->band;
 
-	if (mvmvif->phy_ctxt->ref == 1 && !band_change_removal) {
+	if (mvmvif->deflink.phy_ctxt->ref == 1 && !band_change_removal) {
 		/*
 		 * Change the PHY context configuration as it is currently
 		 * referenced only by the P2P Device MAC (and we can modify it)
 		 */
-		ret = iwl_mvm_phy_ctxt_changed(mvm, mvmvif->phy_ctxt,
+		ret = iwl_mvm_phy_ctxt_changed(mvm, mvmvif->deflink.phy_ctxt,
 					       &chandef, 1, 1);
 		if (ret)
 			goto out_unlock;
@@ -4313,16 +4323,16 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 		if (WARN(ret, "Failed unbinding P2P_DEVICE\n"))
 			goto out_unlock;
 
-		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
+		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
 
 		/* Bind the P2P_DEVICE to the new allocated PHY context */
-		mvmvif->phy_ctxt = phy_ctxt;
+		mvmvif->deflink.phy_ctxt = phy_ctxt;
 
 		ret = iwl_mvm_binding_add_vif(mvm, vif);
 		if (WARN(ret, "Failed binding P2P_DEVICE\n"))
 			goto out_unlock;
 
-		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->phy_ctxt);
+		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
 	}
 
 schedule_time_event:
@@ -4505,7 +4515,7 @@ bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
 
 	lockdep_assert_held(&mvm->mutex);
 
-	mvmvif->phy_ctxt = phy_ctxt;
+	mvmvif->deflink.phy_ctxt = phy_ctxt;
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_AP:
@@ -4609,7 +4619,7 @@ out_remove_binding:
 	iwl_mvm_power_update_mac(mvm);
 out:
 	if (ret)
-		mvmvif->phy_ctxt = NULL;
+		mvmvif->deflink.phy_ctxt = NULL;
 	return ret;
 }
 
@@ -4641,7 +4651,8 @@ bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
 	lockdep_assert_held(&mvm->mutex);
-	iwl_mvm_remove_time_event(mvm, mvmvif, &mvmvif->time_event_data);
+	iwl_mvm_remove_time_event(mvm, mvmvif,
+				  &mvmvif->time_event_data);
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_ADHOC:
@@ -4700,7 +4711,7 @@ out:
 	if (fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_CHANNEL_SWITCH_CMD) &&
 	    switching_chanctx)
 		return;
-	mvmvif->phy_ctxt = NULL;
+	mvmvif->deflink.phy_ctxt = NULL;
 	iwl_mvm_power_update_mac(mvm);
 }
 
@@ -5243,7 +5254,7 @@ static void iwl_mvm_mac_flush(struct ieee80211_hw *hw,
 			continue;
 
 		/* make sure only TDLS peers or the AP are flushed */
-		WARN_ON(i != mvmvif->ap_sta_id && !sta->tdls);
+		WARN_ON(i != mvmvif->deflink.ap_sta_id && !sta->tdls);
 
 		if (drop) {
 			if (iwl_mvm_flush_sta(mvm, mvmsta, false))
@@ -5480,18 +5491,19 @@ static void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
-	if (mvmvif->ap_sta_id != mvmsta->sta_id)
+	if (mvmvif->deflink.ap_sta_id != mvmsta->sta_id)
 		goto unlock;
 
 	if (iwl_mvm_request_statistics(mvm, false))
 		goto unlock;
 
-	sinfo->rx_beacon = mvmvif->beacon_stats.num_beacons +
-			   mvmvif->beacon_stats.accu_num_beacons;
+	sinfo->rx_beacon = mvmvif->deflink.beacon_stats.num_beacons +
+			   mvmvif->deflink.beacon_stats.accu_num_beacons;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_BEACON_RX);
-	if (mvmvif->beacon_stats.avg_signal) {
+	if (mvmvif->deflink.beacon_stats.avg_signal) {
 		/* firmware only reports a value after RXing a few beacons */
-		sinfo->rx_beacon_signal_avg = mvmvif->beacon_stats.avg_signal;
+		sinfo->rx_beacon_signal_avg =
+			mvmvif->deflink.beacon_stats.avg_signal;
 		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_BEACON_SIGNAL_AVG);
 	}
  unlock:
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index e27c893502f7..4eaba89ba0c5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -17,7 +17,7 @@ static u32 iwl_mvm_get_sec_sta_mask(struct iwl_mvm *mvm,
 
 	if (vif->type == NL80211_IFTYPE_AP &&
 	    !(keyconf->flags & IEEE80211_KEY_FLAG_PAIRWISE))
-		return BIT(mvmvif->mcast_sta.sta_id);
+		return BIT(mvmvif->deflink.mcast_sta.sta_id);
 
 	if (sta) {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
@@ -26,8 +26,8 @@ static u32 iwl_mvm_get_sec_sta_mask(struct iwl_mvm *mvm,
 	}
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->ap_sta_id != IWL_MVM_INVALID_STA)
-		return BIT(mvmvif->ap_sta_id);
+	    mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA)
+		return BIT(mvmvif->deflink.ap_sta_id);
 
 	/* invalid */
 	return 0;
@@ -70,8 +70,8 @@ static u32 iwl_mvm_get_sec_flags(struct iwl_mvm *mvm,
 
 	rcu_read_lock();
 	if (!sta && vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->ap_sta_id != IWL_MVM_INVALID_STA) {
-		u8 sta_id = mvmvif->ap_sta_id;
+	    mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
+		u8 sta_id = mvmvif->deflink.ap_sta_id;
 
 		sta = rcu_dereference_check(mvm->fw_id_to_mac_id[sta_id],
 					    lockdep_is_held(&mvm->mutex));
@@ -214,7 +214,7 @@ void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
 	u8 sec_key_ver = iwl_fw_lookup_cmd_ver(mvm->fw, sec_key_id, 0);
 
 	if (WARN_ON(vif->type != NL80211_IFTYPE_STATION ||
-		    mvmvif->ap_sta_id == IWL_MVM_INVALID_STA))
+		    mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA))
 		return;
 
 	if (!sec_key_ver)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index f800bf0b6d90..fe55efce8dd5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -43,13 +43,13 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	 * MAC context is bound to it at this stage.
 	 */
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
-		mvmvif->phy_ctxt = iwl_mvm_get_free_phy_ctxt(mvm);
-		if (!mvmvif->phy_ctxt) {
+		mvmvif->deflink.phy_ctxt = iwl_mvm_get_free_phy_ctxt(mvm);
+		if (!mvmvif->deflink.phy_ctxt) {
 			ret = -ENOSPC;
 			goto out_free_bf;
 		}
 
-		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->phy_ctxt);
+		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
 		ret = iwl_mvm_add_link(mvm, vif);
 		if (ret)
 			goto out_unref_phy;
@@ -97,7 +97,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
 	iwl_mvm_remove_link(mvm, vif);
  out_unref_phy:
-	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
+	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
  out_free_bf:
 	if (mvm->bf_allowed_vif == mvmvif) {
 		mvm->bf_allowed_vif = NULL;
@@ -105,7 +105,8 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 				       IEEE80211_VIF_SUPPORTS_CQM_RSSI);
 	}
  out_remove_mac:
-	mvmvif->phy_ctxt = NULL;
+	mvmvif->deflink.phy_ctxt = NULL;
+	mvmvif->link[0] = NULL;
 	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
  out_unlock:
 	mutex_unlock(&mvm->mutex);
@@ -129,8 +130,8 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 		iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
 				     false);
 		iwl_mvm_remove_link(mvm, vif);
-		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->phy_ctxt);
-		mvmvif->phy_ctxt = NULL;
+		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
+		mvmvif->deflink.phy_ctxt = NULL;
 	}
 
 	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
@@ -187,7 +188,7 @@ out_remove_link:
 	iwl_mvm_power_update_mac(mvm);
 out:
 	if (ret)
-		mvmvif->phy_ctxt = NULL;
+		mvmvif->deflink.phy_ctxt = NULL;
 	return ret;
 }
 
@@ -230,7 +231,7 @@ static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
 out:
 	if (switching_chanctx)
 		return;
-	mvmvif->phy_ctxt = NULL;
+	mvmvif->deflink.phy_ctxt = NULL;
 	iwl_mvm_power_update_mac(mvm);
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 51532b6379e2..b2a7fa09f900 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -147,7 +147,7 @@ static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
 int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *bsta = &mvmvif->bcast_sta;
+	struct iwl_mvm_int_sta *bsta = &mvmvif->deflink.bcast_sta;
 	static const u8 _baddr[] = {0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF};
 	const u8 *baddr = _baddr;
 	unsigned int wdg_timeout =
@@ -172,7 +172,7 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	return iwl_mvm_mld_add_int_sta(mvm, bsta, queue,
 				       ieee80211_vif_type_p2p(vif),
 				       STATION_TYPE_BCAST_MGMT,
-				       mvmvif->phy_ctxt->id, baddr,
+				       mvmvif->deflink.phy_ctxt->id, baddr,
 				       IWL_MAX_TID_COUNT, &wdg_timeout);
 }
 
@@ -183,7 +183,7 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *msta = &mvmvif->mcast_sta;
+	struct iwl_mvm_int_sta *msta = &mvmvif->deflink.mcast_sta;
 	static const u8 _maddr[] = {0x03, 0x00, 0x00, 0x00, 0x00, 0x00};
 	const u8 *maddr = _maddr;
 	unsigned int timeout = iwl_mvm_get_wd_timeout(mvm, vif, false, false);
@@ -200,11 +200,11 @@ int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 * changes in mac80211 layer.
 	 */
 	if (vif->type == NL80211_IFTYPE_ADHOC)
-		mvmvif->cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
+		mvmvif->deflink.cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
 
-	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvmvif->cab_queue,
+	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvmvif->deflink.cab_queue,
 				       vif->type, STATION_TYPE_MCAST,
-				       mvmvif->phy_ctxt->id, maddr, 0,
+				       mvmvif->deflink.phy_ctxt->id, maddr, 0,
 				       &timeout);
 }
 
@@ -219,7 +219,7 @@ int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	return iwl_mvm_mld_add_int_sta(mvm, &mvm->snif_sta, &mvm->snif_queue,
 				       vif->type, STATION_TYPE_BCAST_MGMT,
-				       mvmvif->phy_ctxt->id, NULL,
+				       mvmvif->deflink.phy_ctxt->id, NULL,
 				       IWL_MAX_TID_COUNT, NULL);
 }
 
@@ -305,7 +305,7 @@ int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		return -EINVAL;
 	}
 
-	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->bcast_sta, true,
+	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->deflink.bcast_sta, true,
 				      IWL_MAX_TID_COUNT, queueptr);
 }
 
@@ -318,8 +318,9 @@ int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	lockdep_assert_held(&mvm->mutex);
 
-	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->mcast_sta, true, 0,
-				      &mvmvif->cab_queue);
+	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->deflink.mcast_sta, true,
+				      0,
+				      &mvmvif->deflink.cab_queue);
 }
 
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index cea94250d197..72056cccc77e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -280,12 +280,55 @@ struct iwl_probe_resp_data {
 	int noa_len;
 };
 
+/**
+ * struct iwl_mvm_vif_link_info - per link data in Virtual Interface
+ * @ap_sta_id: the sta_id of the AP - valid only if VIF type is STA
+ * @bssid: BSSID for this (client) interface
+ * @bcast_sta: station used for broadcast packets. Used by the following
+ *	vifs: P2P_DEVICE, GO and AP.
+ * @beacon_stats: beacon statistics, containing the # of received beacons,
+ *	# of received beacons accumulated over FW restart, and the current
+ *	average signal of beacons retrieved from the firmware
+ * @smps_requests: the SMPS requests of different parts of the driver,
+ *	combined on update to yield the overall request to mac80211.
+ * @probe_resp_data: data from FW notification to store NOA and CSA related
+ *	data to be inserted into probe response.
+ * @he_ru_2mhz_block: 26-tone RU OFDMA transmissions should be blocked
+ * @queue_params: QoS params for this MAC
+ */
+struct iwl_mvm_vif_link_info {
+	u8 bssid[ETH_ALEN];
+	u8 ap_sta_id;
+
+	struct iwl_mvm_int_sta bcast_sta;
+	struct iwl_mvm_int_sta mcast_sta;
+
+	struct {
+		u32 num_beacons, accu_num_beacons;
+		u8 avg_signal;
+	} beacon_stats;
+
+	enum ieee80211_smps_mode smps_requests[NUM_IWL_MVM_SMPS_REQ];
+	struct iwl_probe_resp_data __rcu *probe_resp_data;
+
+	bool he_ru_2mhz_block;
+
+	u16 cab_queue;
+	/* Assigned while mac80211 has the link in a channel context,
+	 * or, for P2P Device, while it exists.
+	 */
+	struct iwl_mvm_phy_ctxt *phy_ctxt;
+	/* QoS data from mac80211, need to store this here
+	 * as mac80211 has a separate callback but we need
+	 * to have the data for the MAC context
+	 */
+	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
+};
+
 /**
  * struct iwl_mvm_vif - data per Virtual Interface, it is a MAC context
  * @id: between 0 and 3
  * @color: to solve races upon MAC addition and removal
- * @ap_sta_id: the sta_id of the AP - valid only if VIF type is STA
- * @bssid: BSSID for this (client) interface
  * @associated: indicates that we're currently associated, used only for
  *	managing the firmware state in iwl_mvm_bss_info_changed_station()
  * @ap_assoc_sta_count: count of stations associated to us - valid only
@@ -293,7 +336,7 @@ struct iwl_probe_resp_data {
  * @uploaded: indicates the MAC context has been added to the device
  * @ap_ibss_active: indicates that AP/IBSS is configured and that the interface
  *	should get quota etc.
- * @pm_enabled - Indicate if MAC power management is allowed
+ * @pm_enabled - indicate if MAC power management is allowed
  * @monitor_active: indicates that monitor context is configured, and that the
  *	interface should get quota etc.
  * @low_latency: bit flags for low latency
@@ -302,68 +345,31 @@ struct iwl_probe_resp_data {
  *	as a result from low_latency bit flags and takes force into account.
  * @authorized: indicates the AP station was set to authorized
  * @ps_disabled: indicates that this interface requires PS to be disabled
- * @queue_params: QoS params for this MAC
- * @bcast_sta: station used for broadcast packets. Used by the following
- *  vifs: P2P_DEVICE, GO and AP.
- * @beacon_skb: the skb used to hold the AP/GO beacon template
- * @smps_requests: the SMPS requests of different parts of the driver,
- *	combined on update to yield the overall request to mac80211.
- * @beacon_stats: beacon statistics, containing the # of received beacons,
- *	# of received beacons accumulated over FW restart, and the current
- *	average signal of beacons retrieved from the firmware
+ * @csa_countdown: indicates that CSA countdown may be started
  * @csa_failed: CSA failed to schedule time event, report an error later
+ * @csa_bcn_pending: indicates that we are waiting for a beacon on a new channel
  * @features: hw features active for this vif
- * @probe_resp_data: data from FW notification to store NOA and CSA related
- *	data to be inserted into probe response.
  */
 struct iwl_mvm_vif {
 	struct iwl_mvm *mvm;
 	u16 id;
 	u16 color;
-	u8 ap_sta_id;
 
-	u8 bssid[ETH_ALEN];
 	bool associated;
 	u8 ap_assoc_sta_count;
-
-	u16 cab_queue;
-
 	bool uploaded;
 	bool ap_ibss_active;
 	bool pm_enabled;
 	bool monitor_active;
+
 	u8 low_latency: 6;
 	u8 low_latency_actual: 1;
+
 	u8 authorized:1;
 	bool ps_disabled;
-	struct iwl_mvm_vif_bf_data bf_data;
-
-	struct {
-		u32 num_beacons, accu_num_beacons;
-		u8 avg_signal;
-	} beacon_stats;
 
 	u32 ap_beacon_time;
-
-	enum iwl_tsf_id tsf_id;
-
-	/*
-	 * QoS data from mac80211, need to store this here
-	 * as mac80211 has a separate callback but we need
-	 * to have the data for the MAC context
-	 */
-	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
-	struct iwl_mvm_time_event_data time_event_data;
-	struct iwl_mvm_time_event_data hs_time_event_data;
-
-	struct iwl_mvm_int_sta bcast_sta;
-	struct iwl_mvm_int_sta mcast_sta;
-
-	/*
-	 * Assigned while mac80211 has the interface in a channel context,
-	 * or, for P2P Device, while it exists.
-	 */
-	struct iwl_mvm_phy_ctxt *phy_ctxt;
+	struct iwl_mvm_vif_bf_data bf_data;
 
 #ifdef CONFIG_PM
 	/* WoWLAN GTK rekey data */
@@ -399,40 +405,43 @@ struct iwl_mvm_vif {
 	int dbgfs_quota_min;
 #endif
 
-	enum ieee80211_smps_mode smps_requests[NUM_IWL_MVM_SMPS_REQ];
-
 	/* FW identified misbehaving AP */
 	u8 uapsd_misbehaving_bssid[ETH_ALEN];
-
 	struct delayed_work uapsd_nonagg_detected_wk;
 
-	/* Indicates that CSA countdown may be started */
 	bool csa_countdown;
 	bool csa_failed;
+	bool csa_bcn_pending;
 	u16 csa_target_freq;
 	u16 csa_count;
 	u16 csa_misbehave;
 	struct delayed_work csa_work;
 
-	/* Indicates that we are waiting for a beacon on a new channel */
-	bool csa_bcn_pending;
+	enum iwl_tsf_id tsf_id;
+
+	struct iwl_mvm_time_event_data time_event_data;
+	struct iwl_mvm_time_event_data hs_time_event_data;
 
 	/* TCP Checksum Offload */
 	netdev_features_t features;
 
-	struct iwl_probe_resp_data __rcu *probe_resp_data;
-
 	/* we can only have 2 GTK + 2 IGTK active at a time */
 	struct ieee80211_key_conf *ap_early_keys[4];
 
-	/* 26-tone RU OFDMA transmissions should be blocked */
-	bool he_ru_2mhz_block;
-
 	struct {
 		struct ieee80211_key_conf __rcu *keys[2];
 	} bcn_prot;
+
+	struct iwl_mvm_vif_link_info deflink;
+	struct iwl_mvm_vif_link_info *link[IEEE80211_MLD_MAX_NUM_LINKS];
 };
 
+#define for_each_mvm_vif_valid_link(mvm_vif, link_id)			\
+	for (link_id = 0;						\
+	     link_id < ARRAY_SIZE((mvm_vif)->link);			\
+	     link_id++)							\
+		if ((mvm_vif)->link[link_id])
+
 static inline struct iwl_mvm_vif *
 iwl_mvm_vif_from_mac80211(struct ieee80211_vif *vif)
 {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c
index 06f4203fb989..3ab6fb83a175 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c
@@ -382,12 +382,12 @@ static void iwl_mvm_binding_iterator(void *_data, u8 *mac,
 	unsigned long *data = _data;
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	if (!mvmvif->phy_ctxt)
+	if (!mvmvif->deflink.phy_ctxt)
 		return;
 
 	if (vif->type == NL80211_IFTYPE_STATION ||
 	    vif->type == NL80211_IFTYPE_AP)
-		__set_bit(mvmvif->phy_ctxt->id, data);
+		__set_bit(mvmvif->deflink.phy_ctxt->id, data);
 }
 
 int iwl_mvm_phy_ctx_count(struct iwl_mvm *mvm)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index f5744162d0d8..6e2e3127843f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -150,7 +150,7 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 #endif
 
 	for (ac = IEEE80211_AC_VO; ac <= IEEE80211_AC_BK; ac++) {
-		if (!mvmvif->queue_params[ac].uapsd)
+		if (!mvmvif->deflink.queue_params[ac].uapsd)
 			continue;
 
 		if (!test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status))
@@ -160,7 +160,7 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 		cmd->uapsd_ac_flags |= BIT(ac);
 
 		/* QNDP TID - the highest TID with no admission control */
-		if (!tid_found && !mvmvif->queue_params[ac].acm) {
+		if (!tid_found && !mvmvif->deflink.queue_params[ac].acm) {
 			tid_found = true;
 			switch (ac) {
 			case IEEE80211_AC_VO:
@@ -509,8 +509,9 @@ static void iwl_mvm_power_uapsd_misbehav_ap_iterator(void *_data, u8 *mac,
 	/* The ap_sta_id is not expected to change during current association
 	 * so no explicit protection is needed
 	 */
-	if (mvmvif->ap_sta_id == *ap_sta_id)
-		memcpy(mvmvif->uapsd_misbehaving_bssid, vif->bss_conf.bssid,
+	if (mvmvif->deflink.ap_sta_id == *ap_sta_id)
+		memcpy(mvmvif->uapsd_misbehaving_bssid,
+		       vif->bss_conf.bssid,
 		       ETH_ALEN);
 }
 
@@ -552,7 +553,8 @@ static void iwl_mvm_power_ps_disabled_iterator(void *_data, u8* mac,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	bool *disable_ps = _data;
 
-	if (mvmvif->phy_ctxt && mvmvif->phy_ctxt->id < NUM_PHY_CTX)
+	if (mvmvif->deflink.phy_ctxt &&
+	    mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX)
 		*disable_ps |= mvmvif->ps_disabled;
 }
 
@@ -561,7 +563,7 @@ static void iwl_mvm_power_get_vifs_iterator(void *_data, u8 *mac,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_power_vifs *power_iterator = _data;
-	bool active = mvmvif->phy_ctxt && mvmvif->phy_ctxt->id < NUM_PHY_CTX;
+	bool active = mvmvif->deflink.phy_ctxt && mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX;
 
 	if (!mvmvif->uploaded)
 		return;
@@ -649,11 +651,11 @@ static void iwl_mvm_power_set_pm(struct iwl_mvm *mvm,
 	}
 
 	if (vifs->bss_active && vifs->p2p_active)
-		client_same_channel = (bss_mvmvif->phy_ctxt->id ==
-				       p2p_mvmvif->phy_ctxt->id);
+		client_same_channel = (bss_mvmvif->deflink.phy_ctxt->id ==
+				       p2p_mvmvif->deflink.phy_ctxt->id);
 	if (vifs->bss_active && vifs->ap_active)
-		ap_same_channel = (bss_mvmvif->phy_ctxt->id ==
-				   ap_mvmvif->phy_ctxt->id);
+		ap_same_channel = (bss_mvmvif->deflink.phy_ctxt->id ==
+				   ap_mvmvif->deflink.phy_ctxt->id);
 
 	/* clients are not stand alone: enable PM if DCM */
 	if (!(client_same_channel || ap_same_channel)) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/quota.c b/drivers/net/wireless/intel/iwlwifi/mvm/quota.c
index cea1a34f9130..aad2614af9ad 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/quota.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/quota.c
@@ -33,11 +33,11 @@ static void iwl_mvm_quota_iterator(void *_data, u8 *mac,
 	if (vif == data->disabled_vif)
 		return;
 
-	if (!mvmvif->phy_ctxt)
+	if (!mvmvif->deflink.phy_ctxt)
 		return;
 
 	/* currently, PHY ID == binding ID */
-	id = mvmvif->phy_ctxt->id;
+	id = mvmvif->deflink.phy_ctxt->id;
 
 	/* need at least one binding per PHY */
 	BUILD_BUG_ON(NUM_PHY_CTX > MAX_BINDINGS);
@@ -67,9 +67,10 @@ static void iwl_mvm_quota_iterator(void *_data, u8 *mac,
 	}
 
 	if (data->colors[id] < 0)
-		data->colors[id] = mvmvif->phy_ctxt->color;
+		data->colors[id] = mvmvif->deflink.phy_ctxt->color;
 	else
-		WARN_ON_ONCE(data->colors[id] != mvmvif->phy_ctxt->color);
+		WARN_ON_ONCE(data->colors[id] !=
+			     mvmvif->deflink.phy_ctxt->color);
 
 	data->n_interfaces[id]++;
 
@@ -99,7 +100,7 @@ static void iwl_mvm_adjust_quota_for_noa(struct iwl_mvm *mvm,
 	if (!mvmvif->ap_ibss_active)
 		return;
 
-	phy_id = mvmvif->phy_ctxt->id;
+	phy_id = mvmvif->deflink.phy_ctxt->id;
 	beacon_int = mvm->noa_vif->bss_conf.beacon_int;
 
 	for (i = 0; i < MAX_BINDINGS; i++) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index d2ce414879aa..4105b79db467 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -237,11 +237,11 @@ static void iwl_mvm_rx_handle_tcm(struct iwl_mvm *mvm,
 
 	if (mdata->opened_rx_ba_sessions ||
 	    mdata->uapsd_nonagg_detect.detected ||
-	    (!mvmvif->queue_params[IEEE80211_AC_VO].uapsd &&
-	     !mvmvif->queue_params[IEEE80211_AC_VI].uapsd &&
-	     !mvmvif->queue_params[IEEE80211_AC_BE].uapsd &&
-	     !mvmvif->queue_params[IEEE80211_AC_BK].uapsd) ||
-	    mvmsta->sta_id != mvmvif->ap_sta_id)
+	    (!mvmvif->deflink.queue_params[IEEE80211_AC_VO].uapsd &&
+	     !mvmvif->deflink.queue_params[IEEE80211_AC_VI].uapsd &&
+	     !mvmvif->deflink.queue_params[IEEE80211_AC_BE].uapsd &&
+	     !mvmvif->deflink.queue_params[IEEE80211_AC_BK].uapsd) ||
+	    mvmsta->sta_id != mvmvif->deflink.ap_sta_id)
 		return;
 
 	if (rate_n_flags & RATE_MCS_HT_MSK_V1) {
@@ -628,9 +628,9 @@ static void iwl_mvm_stat_iterator(void *_data, u8 *mac,
 	 * data copied into the "data" struct, but rather the data from
 	 * the notification directly.
 	 */
-	mvmvif->beacon_stats.num_beacons =
+	mvmvif->deflink.beacon_stats.num_beacons =
 		le32_to_cpu(data->beacon_counter[vif_id]);
-	mvmvif->beacon_stats.avg_signal =
+	mvmvif->deflink.beacon_stats.avg_signal =
 		-data->beacon_average_energy[vif_id];
 
 	if (mvmvif->id != id)
@@ -643,8 +643,8 @@ static void iwl_mvm_stat_iterator(void *_data, u8 *mac,
 	 * request to clear statistics
 	 */
 	if (le32_to_cpu(data->flags) & IWL_STATISTICS_REPLY_FLG_CLEAR)
-		mvmvif->beacon_stats.accu_num_beacons +=
-			mvmvif->beacon_stats.num_beacons;
+		mvmvif->deflink.beacon_stats.accu_num_beacons +=
+			mvmvif->deflink.beacon_stats.num_beacons;
 
 	iwl_mvm_update_vif_sig(vif, sig);
 }
@@ -666,17 +666,17 @@ static void iwl_mvm_stat_iterator_all_macs(void *_data, u8 *mac,
 
 	mac_stats = &data->per_mac_stats[vif_id];
 
-	mvmvif->beacon_stats.num_beacons =
+	mvmvif->deflink.beacon_stats.num_beacons =
 		le32_to_cpu(mac_stats->beacon_counter);
-	mvmvif->beacon_stats.avg_signal =
+	mvmvif->deflink.beacon_stats.avg_signal =
 		-le32_to_cpu(mac_stats->beacon_average_energy);
 
 	/* make sure that beacon statistics don't go backwards with TCM
 	 * request to clear statistics
 	 */
 	if (le32_to_cpu(data->flags) & IWL_STATISTICS_REPLY_FLG_CLEAR)
-		mvmvif->beacon_stats.accu_num_beacons +=
-			mvmvif->beacon_stats.num_beacons;
+		mvmvif->deflink.beacon_stats.accu_num_beacons +=
+			mvmvif->deflink.beacon_stats.num_beacons;
 
 	sig = -le32_to_cpu(mac_stats->beacon_filter_average_energy);
 	iwl_mvm_update_vif_sig(vif, sig);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index acd8803dbcdd..975edc02e637 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -193,8 +193,9 @@ static void iwl_mvm_scan_iterator(void *_data, u8 *mac,
 	struct iwl_mvm_scan_iter_data *data = _data;
 	struct iwl_mvm_vif *curr_mvmvif;
 
-	if (vif->type != NL80211_IFTYPE_P2P_DEVICE && mvmvif->phy_ctxt &&
-	    mvmvif->phy_ctxt->id < NUM_PHY_CTX)
+	if (vif->type != NL80211_IFTYPE_P2P_DEVICE &&
+	    mvmvif->deflink.phy_ctxt &&
+	    mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX)
 		data->global_cnt += 1;
 
 	if (!data->current_vif || vif == data->current_vif)
@@ -203,8 +204,8 @@ static void iwl_mvm_scan_iterator(void *_data, u8 *mac,
 	curr_mvmvif = iwl_mvm_vif_from_mac80211(data->current_vif);
 
 	if (vif->type == NL80211_IFTYPE_AP && vif->p2p &&
-	    mvmvif->phy_ctxt && curr_mvmvif->phy_ctxt &&
-	    mvmvif->phy_ctxt->id != curr_mvmvif->phy_ctxt->id)
+	    mvmvif->deflink.phy_ctxt && curr_mvmvif->deflink.phy_ctxt &&
+	    mvmvif->deflink.phy_ctxt->id != curr_mvmvif->deflink.phy_ctxt->id)
 		data->is_dcm_with_p2p_go = true;
 }
 
@@ -2677,9 +2678,9 @@ static void iwl_mvm_scan_respect_p2p_go_iter(void *_data, u8 *mac,
 		return;
 
 	if (vif->type == NL80211_IFTYPE_AP && vif->p2p &&
-	    mvmvif->phy_ctxt->id < NUM_PHY_CTX &&
+	    mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX &&
 	    (data->band == NUM_NL80211_BANDS ||
-	     mvmvif->phy_ctxt->channel->band == data->band))
+	     mvmvif->deflink.phy_ctxt->channel->band == data->band))
 		data->p2p_go = true;
 }
 
@@ -2980,7 +2981,7 @@ void iwl_mvm_rx_umac_scan_complete_notif(struct iwl_mvm *mvm,
 			.scan_start_tsf = mvm->scan_start,
 		};
 
-		memcpy(info.tsf_bssid, mvm->scan_vif->bssid, ETH_ALEN);
+		memcpy(info.tsf_bssid, mvm->scan_vif->deflink.bssid, ETH_ALEN);
 		ieee80211_scan_completed(mvm->hw, &info);
 		mvm->scan_vif = NULL;
 		cancel_delayed_work(&mvm->scan_timeout_dwork);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
index 1f4ac1e93cee..84b750d02848 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
@@ -23,14 +23,14 @@ static void iwl_mvm_bound_iface_iterator(void *_data, u8 *mac,
 	struct iwl_mvm_active_iface_iterator_data *data = _data;
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	if (vif == data->ignore_vif || !mvmvif->phy_ctxt ||
+	if (vif == data->ignore_vif || !mvmvif->deflink.phy_ctxt ||
 	    vif->type == NL80211_IFTYPE_P2P_DEVICE)
 		return;
 
 	data->num_active_macs++;
 
 	if (vif->type == NL80211_IFTYPE_STATION) {
-		data->sta_vif_ap_sta_id = mvmvif->ap_sta_id;
+		data->sta_vif_ap_sta_id = mvmvif->deflink.ap_sta_id;
 		if (vif->cfg.assoc)
 			data->sta_vif_state = SF_FULL_ON;
 		else
@@ -264,7 +264,7 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 			} else if (changed_vif->cfg.assoc &&
 				   changed_vif->bss_conf.dtim_period) {
 				mvmvif = iwl_mvm_vif_from_mac80211(changed_vif);
-				sta_id = mvmvif->ap_sta_id;
+				sta_id = mvmvif->deflink.ap_sta_id;
 				new_state = SF_FULL_ON;
 			} else {
 				new_state = SF_INIT_OFF;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index ec843f1b0165..c3b81021898b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1785,10 +1785,10 @@ update_fw:
 
 	if (vif->type == NL80211_IFTYPE_STATION) {
 		if (!sta->tdls) {
-			WARN_ON(mvmvif->ap_sta_id != IWL_MVM_INVALID_STA);
-			mvmvif->ap_sta_id = sta_id;
+			WARN_ON(mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA);
+			mvmvif->deflink.ap_sta_id = sta_id;
 		} else {
-			WARN_ON(mvmvif->ap_sta_id == IWL_MVM_INVALID_STA);
+			WARN_ON(mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA);
 		}
 	}
 
@@ -1953,7 +1953,7 @@ static bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	}
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->ap_sta_id == sta_id) {
+	    mvmvif->deflink.ap_sta_id == sta_id) {
 		/* if associated - we can't remove the AP STA now */
 		if (vif->cfg.assoc)
 			return true;
@@ -1962,7 +1962,7 @@ static bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		iwl_mvm_sec_key_remove_ap(mvm, vif);
 
 		/* unassoc - go ahead - remove the AP STA now */
-		mvmvif->ap_sta_id = IWL_MVM_INVALID_STA;
+		mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
 	}
 
 	/*
@@ -2227,7 +2227,7 @@ void iwl_mvm_dealloc_snif_sta(struct iwl_mvm *mvm)
 int iwl_mvm_send_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *bsta = &mvmvif->bcast_sta;
+	struct iwl_mvm_int_sta *bsta = &mvmvif->deflink.bcast_sta;
 	static const u8 _baddr[] = {0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF};
 	const u8 *baddr = _baddr;
 	int queue;
@@ -2236,7 +2236,7 @@ int iwl_mvm_send_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		iwl_mvm_get_wd_timeout(mvm, vif, false, false);
 	struct iwl_trans_txq_scd_cfg cfg = {
 		.fifo = IWL_MVM_TX_FIFO_VO,
-		.sta_id = mvmvif->bcast_sta.sta_id,
+		.sta_id = mvmvif->deflink.bcast_sta.sta_id,
 		.tid = IWL_MAX_TID_COUNT,
 		.aggregate = false,
 		.frame_limit = IWL_FRAME_LIMIT,
@@ -2302,7 +2302,7 @@ void iwl_mvm_free_bcast_sta_queues(struct iwl_mvm *mvm,
 
 	lockdep_assert_held(&mvm->mutex);
 
-	iwl_mvm_flush_sta(mvm, &mvmvif->bcast_sta, true);
+	iwl_mvm_flush_sta(mvm, &mvmvif->deflink.bcast_sta, true);
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_AP:
@@ -2319,13 +2319,13 @@ void iwl_mvm_free_bcast_sta_queues(struct iwl_mvm *mvm,
 	}
 
 	queue = *queueptr;
-	iwl_mvm_disable_txq(mvm, NULL, mvmvif->bcast_sta.sta_id,
+	iwl_mvm_disable_txq(mvm, NULL, mvmvif->deflink.bcast_sta.sta_id,
 			    queueptr, IWL_MAX_TID_COUNT);
 	if (iwl_mvm_has_new_tx_api(mvm))
 		return;
 
-	WARN_ON(!(mvmvif->bcast_sta.tfd_queue_msk & BIT(queue)));
-	mvmvif->bcast_sta.tfd_queue_msk &= ~BIT(queue);
+	WARN_ON(!(mvmvif->deflink.bcast_sta.tfd_queue_msk & BIT(queue)));
+	mvmvif->deflink.bcast_sta.tfd_queue_msk &= ~BIT(queue);
 }
 
 /* Send the FW a request to remove the station from it's internal data
@@ -2339,7 +2339,7 @@ int iwl_mvm_send_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	iwl_mvm_free_bcast_sta_queues(mvm, vif);
 
-	ret = iwl_mvm_rm_sta_common(mvm, mvmvif->bcast_sta.sta_id);
+	ret = iwl_mvm_rm_sta_common(mvm, mvmvif->deflink.bcast_sta.sta_id);
 	if (ret)
 		IWL_WARN(mvm, "Failed sending remove station\n");
 	return ret;
@@ -2351,7 +2351,7 @@ int iwl_mvm_alloc_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	lockdep_assert_held(&mvm->mutex);
 
-	return iwl_mvm_allocate_int_sta(mvm, &mvmvif->bcast_sta, 0,
+	return iwl_mvm_allocate_int_sta(mvm, &mvmvif->deflink.bcast_sta, 0,
 					ieee80211_vif_type_p2p(vif),
 					IWL_STA_GENERAL_PURPOSE);
 }
@@ -2366,7 +2366,7 @@ int iwl_mvm_alloc_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 int iwl_mvm_add_p2p_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *bsta = &mvmvif->bcast_sta;
+	struct iwl_mvm_int_sta *bsta = &mvmvif->deflink.bcast_sta;
 	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -2387,7 +2387,7 @@ void iwl_mvm_dealloc_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	iwl_mvm_dealloc_int_sta(mvm, &mvmvif->bcast_sta);
+	iwl_mvm_dealloc_int_sta(mvm, &mvmvif->deflink.bcast_sta);
 }
 
 /*
@@ -2418,7 +2418,7 @@ int iwl_mvm_rm_p2p_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 int iwl_mvm_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *msta = &mvmvif->mcast_sta;
+	struct iwl_mvm_int_sta *msta = &mvmvif->deflink.mcast_sta;
 	static const u8 _maddr[] = {0x03, 0x00, 0x00, 0x00, 0x00, 0x00};
 	const u8 *maddr = _maddr;
 	struct iwl_trans_txq_scd_cfg cfg = {
@@ -2445,7 +2445,7 @@ int iwl_mvm_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 * changes in mac80211 layer.
 	 */
 	if (vif->type == NL80211_IFTYPE_ADHOC)
-		mvmvif->cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
+		mvmvif->deflink.cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
 
 	/*
 	 * While in previous FWs we had to exclude cab queue from TFD queue
@@ -2453,9 +2453,10 @@ int iwl_mvm_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 */
 	if (!iwl_mvm_has_new_tx_api(mvm) &&
 	    fw_has_api(&mvm->fw->ucode_capa, IWL_UCODE_TLV_API_STA_TYPE)) {
-		iwl_mvm_enable_txq(mvm, NULL, mvmvif->cab_queue, 0, &cfg,
+		iwl_mvm_enable_txq(mvm, NULL, mvmvif->deflink.cab_queue, 0,
+				   &cfg,
 				   timeout);
-		msta->tfd_queue_msk |= BIT(mvmvif->cab_queue);
+		msta->tfd_queue_msk |= BIT(mvmvif->deflink.cab_queue);
 	}
 	ret = iwl_mvm_add_int_sta_common(mvm, msta, maddr,
 					 mvmvif->id, mvmvif->color);
@@ -2476,10 +2477,11 @@ int iwl_mvm_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 			ret = queue;
 			goto err;
 		}
-		mvmvif->cab_queue = queue;
+		mvmvif->deflink.cab_queue = queue;
 	} else if (!fw_has_api(&mvm->fw->ucode_capa,
 			       IWL_UCODE_TLV_API_STA_TYPE))
-		iwl_mvm_enable_txq(mvm, NULL, mvmvif->cab_queue, 0, &cfg,
+		iwl_mvm_enable_txq(mvm, NULL, mvmvif->deflink.cab_queue, 0,
+				   &cfg,
 				   timeout);
 
 	return 0;
@@ -2552,12 +2554,12 @@ int iwl_mvm_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	lockdep_assert_held(&mvm->mutex);
 
-	iwl_mvm_flush_sta(mvm, &mvmvif->mcast_sta, true);
+	iwl_mvm_flush_sta(mvm, &mvmvif->deflink.mcast_sta, true);
 
-	iwl_mvm_disable_txq(mvm, NULL, mvmvif->mcast_sta.sta_id,
-			    &mvmvif->cab_queue, 0);
+	iwl_mvm_disable_txq(mvm, NULL, mvmvif->deflink.mcast_sta.sta_id,
+			    &mvmvif->deflink.cab_queue, 0);
 
-	ret = iwl_mvm_rm_sta_common(mvm, mvmvif->mcast_sta.sta_id);
+	ret = iwl_mvm_rm_sta_common(mvm, mvmvif->deflink.mcast_sta.sta_id);
 	if (ret)
 		IWL_WARN(mvm, "Failed sending remove station\n");
 
@@ -3383,8 +3385,8 @@ static struct iwl_mvm_sta *iwl_mvm_get_key_sta(struct iwl_mvm *mvm,
 	 * station ID, then use AP's station ID.
 	 */
 	if (vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->ap_sta_id != IWL_MVM_INVALID_STA) {
-		u8 sta_id = mvmvif->ap_sta_id;
+	    mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
+		u8 sta_id = mvmvif->deflink.ap_sta_id;
 
 		sta = rcu_dereference_check(mvm->fw_id_to_mac_id[sta_id],
 					    lockdep_is_held(&mvm->mutex));
@@ -3661,8 +3663,8 @@ static inline u8 *iwl_mvm_get_mac_addr(struct iwl_mvm *mvm,
 		return sta->addr;
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->ap_sta_id != IWL_MVM_INVALID_STA) {
-		u8 sta_id = mvmvif->ap_sta_id;
+	    mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
+		u8 sta_id = mvmvif->deflink.ap_sta_id;
 		sta = rcu_dereference_protected(mvm->fw_id_to_mac_id[sta_id],
 						lockdep_is_held(&mvm->mutex));
 		return sta->addr;
@@ -3694,7 +3696,7 @@ static int __iwl_mvm_set_sta_key(struct iwl_mvm *mvm,
 		   !(keyconf->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-		sta_id = mvmvif->mcast_sta.sta_id;
+		sta_id = mvmvif->deflink.mcast_sta.sta_id;
 	} else {
 		IWL_ERR(mvm, "Failed to find station id\n");
 		return -EINVAL;
@@ -3759,7 +3761,7 @@ int iwl_mvm_set_sta_key(struct iwl_mvm *mvm,
 	} else {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-		sta_id = mvmvif->mcast_sta.sta_id;
+		sta_id = mvmvif->deflink.mcast_sta.sta_id;
 	}
 
 	if (keyconf->cipher == WLAN_CIPHER_SUITE_AES_CMAC ||
@@ -3834,7 +3836,7 @@ int iwl_mvm_remove_sta_key(struct iwl_mvm *mvm,
 	if (mvm_sta)
 		sta_id = mvm_sta->sta_id;
 	else if (!sta && vif->type == NL80211_IFTYPE_AP && mcast)
-		sta_id = iwl_mvm_vif_from_mac80211(vif)->mcast_sta.sta_id;
+		sta_id = iwl_mvm_vif_from_mac80211(vif)->deflink.mcast_sta.sta_id;
 
 
 	IWL_DEBUG_WEP(mvm, "mvm remove dynamic key: idx=%d sta=%d\n",
@@ -4113,17 +4115,19 @@ void iwl_mvm_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 		return;
 
 	/* Need to block/unblock also multicast station */
-	if (mvmvif->mcast_sta.sta_id != IWL_MVM_INVALID_STA)
+	if (mvmvif->deflink.mcast_sta.sta_id != IWL_MVM_INVALID_STA)
 		iwl_mvm_int_sta_modify_disable_tx(mvm, mvmvif,
-						  &mvmvif->mcast_sta, disable);
+						  &mvmvif->deflink.mcast_sta,
+						  disable);
 
 	/*
 	 * Only unblock the broadcast station (FW blocks it for immediate
 	 * quiet, not the driver)
 	 */
-	if (!disable && mvmvif->bcast_sta.sta_id != IWL_MVM_INVALID_STA)
+	if (!disable && mvmvif->deflink.bcast_sta.sta_id != IWL_MVM_INVALID_STA)
 		iwl_mvm_int_sta_modify_disable_tx(mvm, mvmvif,
-						  &mvmvif->bcast_sta, disable);
+						  &mvmvif->deflink.bcast_sta,
+						  disable);
 }
 
 void iwl_mvm_csa_client_absent(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
@@ -4133,7 +4137,7 @@ void iwl_mvm_csa_client_absent(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	rcu_read_lock();
 
-	mvmsta = iwl_mvm_sta_from_staid_rcu(mvm, mvmvif->ap_sta_id);
+	mvmsta = iwl_mvm_sta_from_staid_rcu(mvm, mvmvif->deflink.ap_sta_id);
 
 	if (mvmsta)
 		iwl_mvm_sta_modify_disable_tx(mvm, mvmsta, true);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index e403a240a82f..3b17dd998d19 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -79,7 +79,8 @@ void iwl_mvm_roc_done_wk(struct work_struct *wk)
 
 		if (!WARN_ON(!mvm->p2p_device_vif)) {
 			mvmvif = iwl_mvm_vif_from_mac80211(mvm->p2p_device_vif);
-			iwl_mvm_flush_sta(mvm, &mvmvif->bcast_sta, true);
+			iwl_mvm_flush_sta(mvm, &mvmvif->deflink.bcast_sta,
+					  true);
 		}
 	}
 
@@ -170,7 +171,8 @@ static bool iwl_mvm_te_check_disconnect(struct iwl_mvm *mvm,
 		struct iwl_mvm_sta *mvmsta;
 
 		rcu_read_lock();
-		mvmsta = iwl_mvm_sta_from_staid_rcu(mvm, mvmvif->ap_sta_id);
+		mvmsta = iwl_mvm_sta_from_staid_rcu(mvm,
+						    mvmvif->deflink.ap_sta_id);
 		if (!WARN_ON(!mvmsta))
 			iwl_mvm_sta_modify_disable_tx(mvm, mvmsta, false);
 		rcu_read_unlock();
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 6e2e0791d848..890fe93896ba 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -631,7 +631,7 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 
 		if (!ieee80211_has_order(fc) && !ieee80211_is_probe_req(fc) &&
 		    is_multicast_ether_addr(hdr->addr1))
-			return mvmvif->cab_queue;
+			return mvmvif->deflink.cab_queue;
 
 		WARN_ONCE(info->control.vif->type != NL80211_IFTYPE_ADHOC,
 			  "fc=0x%02x", le16_to_cpu(fc));
@@ -668,7 +668,7 @@ static void iwl_mvm_probe_resp_set_noa(struct iwl_mvm *mvm,
 
 	rcu_read_lock();
 
-	resp_data = rcu_dereference(mvmvif->probe_resp_data);
+	resp_data = rcu_dereference(mvmvif->deflink.probe_resp_data);
 	if (!resp_data)
 		goto out;
 
@@ -740,9 +740,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 		    info.control.vif->type == NL80211_IFTYPE_AP ||
 		    info.control.vif->type == NL80211_IFTYPE_ADHOC) {
 			if (!ieee80211_is_data(hdr->frame_control))
-				sta_id = mvmvif->bcast_sta.sta_id;
+				sta_id = mvmvif->deflink.bcast_sta.sta_id;
 			else
-				sta_id = mvmvif->mcast_sta.sta_id;
+				sta_id = mvmvif->deflink.mcast_sta.sta_id;
 
 			queue = iwl_mvm_get_ctrl_vif_queue(mvm, &info, hdr);
 		} else if (info.control.vif->type == NL80211_IFTYPE_MONITOR) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 14b2de65bd84..c1f6dd8f7d10 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -294,13 +294,13 @@ void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		return;
 
 	mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	mvmvif->smps_requests[req_type] = smps_request;
+	mvmvif->deflink.smps_requests[req_type] = smps_request;
 	for (i = 0; i < NUM_IWL_MVM_SMPS_REQ; i++) {
-		if (mvmvif->smps_requests[i] == IEEE80211_SMPS_STATIC) {
+		if (mvmvif->deflink.smps_requests[i] == IEEE80211_SMPS_STATIC) {
 			smps_mode = IEEE80211_SMPS_STATIC;
 			break;
 		}
-		if (mvmvif->smps_requests[i] == IEEE80211_SMPS_DYNAMIC)
+		if (mvmvif->deflink.smps_requests[i] == IEEE80211_SMPS_DYNAMIC)
 			smps_mode = IEEE80211_SMPS_DYNAMIC;
 	}
 
@@ -392,12 +392,12 @@ static void iwl_mvm_diversity_iter(void *_data, u8 *mac,
 	struct iwl_mvm_diversity_iter_data *data = _data;
 	int i;
 
-	if (mvmvif->phy_ctxt != data->ctxt)
+	if (mvmvif->deflink.phy_ctxt != data->ctxt)
 		return;
 
 	for (i = 0; i < NUM_IWL_MVM_SMPS_REQ; i++) {
-		if (mvmvif->smps_requests[i] == IEEE80211_SMPS_STATIC ||
-		    mvmvif->smps_requests[i] == IEEE80211_SMPS_DYNAMIC) {
+		if (mvmvif->deflink.smps_requests[i] == IEEE80211_SMPS_STATIC ||
+		    mvmvif->deflink.smps_requests[i] == IEEE80211_SMPS_DYNAMIC) {
 			data->result = false;
 			break;
 		}
@@ -495,10 +495,10 @@ static void iwl_mvm_ll_iter(void *_data, u8 *mac, struct ieee80211_vif *vif)
 	if (iwl_mvm_vif_low_latency(mvmvif)) {
 		result->result = true;
 
-		if (!mvmvif->phy_ctxt)
+		if (!mvmvif->deflink.phy_ctxt)
 			return;
 
-		band = mvmvif->phy_ctxt->channel->band;
+		band = mvmvif->deflink.phy_ctxt->channel->band;
 		result->result_per_band[band] = true;
 	}
 }
@@ -819,10 +819,10 @@ static void iwl_mvm_uapsd_agg_disconnect(struct iwl_mvm *mvm,
 	if (!vif->cfg.assoc)
 		return;
 
-	if (!mvmvif->queue_params[IEEE80211_AC_VO].uapsd &&
-	    !mvmvif->queue_params[IEEE80211_AC_VI].uapsd &&
-	    !mvmvif->queue_params[IEEE80211_AC_BE].uapsd &&
-	    !mvmvif->queue_params[IEEE80211_AC_BK].uapsd)
+	if (!mvmvif->deflink.queue_params[IEEE80211_AC_VO].uapsd &&
+	    !mvmvif->deflink.queue_params[IEEE80211_AC_VI].uapsd &&
+	    !mvmvif->deflink.queue_params[IEEE80211_AC_BE].uapsd &&
+	    !mvmvif->deflink.queue_params[IEEE80211_AC_BK].uapsd)
 		return;
 
 	if (mvm->tcm.data[mvmvif->id].uapsd_nonagg_detect.detected)
@@ -831,7 +831,8 @@ static void iwl_mvm_uapsd_agg_disconnect(struct iwl_mvm *mvm,
 	mvm->tcm.data[mvmvif->id].uapsd_nonagg_detect.detected = true;
 	IWL_INFO(mvm,
 		 "detected AP should do aggregation but isn't, likely due to U-APSD\n");
-	schedule_delayed_work(&mvmvif->uapsd_nonagg_detected_wk, 15 * HZ);
+	schedule_delayed_work(&mvmvif->uapsd_nonagg_detected_wk,
+			      15 * HZ);
 }
 
 static void iwl_mvm_check_uapsd_agg_expected_tpt(struct iwl_mvm *mvm,
@@ -883,10 +884,10 @@ static void iwl_mvm_tcm_iterator(void *_data, u8 *mac,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	u32 *band = _data;
 
-	if (!mvmvif->phy_ctxt)
+	if (!mvmvif->deflink.phy_ctxt)
 		return;
 
-	band[mvmvif->id] = mvmvif->phy_ctxt->channel->band;
+	band[mvmvif->id] = mvmvif->deflink.phy_ctxt->channel->band;
 }
 
 static unsigned long iwl_mvm_calc_tcm_stats(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From c8ee33e184e4163586cc6c786d5c14dcdf6f4e07 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:58:42 +0300
Subject: wifi: iwlwifi: mvm: sta preparation for MLO

Split iwl_mvm_sta into general and link specific parts. As a first
step, all link dependent parameters reside in deflink.

The change was done mostly using the spatch below with some manual
adjustments.

@iwl_mvm_sta@
struct iwl_mvm_sta *s;
identifier var = {sta_id, lq_sta, avg_energy};
@@

(
  s->
-    var
+    deflink.var
)

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.34eace06d583.I1f8c5e919a71b21030460fbdd220d42401b688b1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/coex.c     |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c  |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  19 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c  |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c  |   4 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c    |  10 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c       |  42 ++++----
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c       |   6 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c     |   5 +-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c      | 113 ++++++++++++----------
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h      |  37 +++++--
 drivers/net/wireless/intel/iwlwifi/mvm/tdls.c     |   8 +-
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c       |  34 ++++---
 13 files changed, 161 insertions(+), 123 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
index 73ef1da3ccac..21ba3c7dd9b4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
@@ -194,7 +194,7 @@ static int iwl_mvm_bt_coex_reduced_txp(struct iwl_mvm *mvm, u8 sta_id,
 	if (mvmsta->bt_reduced_txpower == enable)
 		return 0;
 
-	value = mvmsta->sta_id;
+	value = mvmsta->deflink.sta_id;
 
 	if (enable)
 		value |= BT_REDUCED_TX_POWER_BIT;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 2a6f42f07c9e..b5eb0ee298ad 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -376,7 +376,7 @@ static ssize_t iwl_dbgfs_rs_data_read(struct file *file, char __user *user_buf,
 {
 	struct ieee80211_sta *sta = file->private_data;
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->lq_sta.rs_fw;
+	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
 	struct iwl_mvm *mvm = lq_sta->pers.drv;
 	static const size_t bufsz = 2048;
 	char *buff;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 163515db6828..282c15e74f08 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -922,7 +922,7 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 	switch (action) {
 	case IEEE80211_AMPDU_RX_START:
 		if (iwl_mvm_vif_from_mac80211(vif)->deflink.ap_sta_id ==
-				iwl_mvm_sta_from_mac80211(sta)->sta_id) {
+		    iwl_mvm_sta_from_mac80211(sta)->deflink.sta_id) {
 			struct iwl_mvm_vif *mvmvif;
 			u16 macid = iwl_mvm_vif_from_mac80211(vif)->id;
 			struct iwl_mvm_tcm_mac *mdata = &mvm->tcm.data[macid];
@@ -3116,7 +3116,7 @@ static void __iwl_mvm_mac_sta_notify(struct ieee80211_hw *hw,
 		 */
 		break;
 	case STA_NOTIFY_AWAKE:
-		if (WARN_ON(mvmsta->sta_id == IWL_MVM_INVALID_STA))
+		if (WARN_ON(mvmsta->deflink.sta_id == IWL_MVM_INVALID_STA))
 			break;
 
 		if (txqs)
@@ -3206,8 +3206,8 @@ static void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 	 * callback deleted the station.
 	 */
 	mutex_lock(&mvm->mutex);
-	if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[mvm_sta->sta_id]))
-		rcu_assign_pointer(mvm->fw_id_to_mac_id[mvm_sta->sta_id],
+	if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id]))
+		rcu_assign_pointer(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id],
 				   ERR_PTR(-ENOENT));
 
 	mutex_unlock(&mvm->mutex);
@@ -3534,7 +3534,8 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 			     !iwlwifi_mod_params.disable_11ax) ||
 			    (vif->bss_conf.eht_support &&
 			     !iwlwifi_mod_params.disable_11be))
-				iwl_mvm_cfg_he_sta(mvm, vif, mvm_sta->sta_id);
+				iwl_mvm_cfg_he_sta(mvm, vif,
+						   mvm_sta->deflink.sta_id);
 		} else if (vif->type == NL80211_IFTYPE_STATION) {
 			vif->bss_conf.he_support = sta->deflink.he_cap.has_he;
 
@@ -5470,13 +5471,13 @@ static void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 
-	if (mvmsta->avg_energy) {
-		sinfo->signal_avg = -(s8)mvmsta->avg_energy;
+	if (mvmsta->deflink.avg_energy) {
+		sinfo->signal_avg = -(s8)mvmsta->deflink.avg_energy;
 		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_SIGNAL_AVG);
 	}
 
 	if (iwl_mvm_has_tlc_offload(mvm)) {
-		struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->lq_sta.rs_fw;
+		struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
 
 		iwl_mvm_set_sta_rate(lq_sta->last_rate_n_flags, &sinfo->txrate);
 		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
@@ -5491,7 +5492,7 @@ static void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
-	if (mvmvif->deflink.ap_sta_id != mvmsta->sta_id)
+	if (mvmvif->deflink.ap_sta_id != mvmsta->deflink.sta_id)
 		goto unlock;
 
 	if (iwl_mvm_request_statistics(mvm, false))
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index 4eaba89ba0c5..154f145396ce 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -22,7 +22,7 @@ static u32 iwl_mvm_get_sec_sta_mask(struct iwl_mvm *mvm,
 	if (sta) {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 
-		return BIT(mvmsta->sta_id);
+		return BIT(mvmsta->deflink.sta_id);
 	}
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index b2a7fa09f900..c063383247f9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -243,7 +243,7 @@ static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm,
 			.operation = cpu_to_le32(IWL_SCD_QUEUE_REMOVE),
 			.u.remove.tid = cpu_to_le32(tid),
 			.u.remove.sta_mask =
-				cpu_to_le32(BIT(mvmsta->sta_id)),
+				cpu_to_le32(BIT(mvmsta->deflink.sta_id)),
 		};
 
 		ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0,
@@ -348,7 +348,7 @@ static void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
 
 	mvm_sta->disable_tx = disable;
 
-	cmd.sta_id = cpu_to_le32(mvm_sta->sta_id);
+	cmd.sta_id = cpu_to_le32(mvm_sta->deflink.sta_id);
 	cmd.disable = cpu_to_le32(disable);
 
 	ret = iwl_mvm_send_cmd_pdu(mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index e3fb1b2cea6d..a33673c6ae7d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -438,7 +438,7 @@ void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 
 	flags = le32_to_cpu(notif->flags);
 
-	lq_sta = &mvmsta->lq_sta.rs_fw;
+	lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
 
 	if (flags & IWL_TLC_NOTIF_FLAG_RATE) {
 		char pretty_rate[100];
@@ -547,12 +547,12 @@ void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 {
 	struct ieee80211_hw *hw = mvm->hw;
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->lq_sta.rs_fw;
+	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
 	u32 cmd_id = WIDE_ID(DATA_PATH_GROUP, TLC_MNG_CONFIG_CMD);
 	struct ieee80211_supported_band *sband = hw->wiphy->bands[band];
 	u16 max_amsdu_len = rs_fw_get_max_amsdu_len(sta);
 	struct iwl_tlc_config_cmd_v4 cfg_cmd = {
-		.sta_id = mvmsta->sta_id,
+		.sta_id = mvmsta->deflink.sta_id,
 		.max_ch_width = update ?
 			rs_fw_bw_from_sta_bw(sta) : IWL_TLC_MNG_CH_WIDTH_20MHZ,
 		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, sta, sband)),
@@ -641,12 +641,12 @@ int rs_fw_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 
 void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta)
 {
-	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->lq_sta.rs_fw;
+	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
 
 	IWL_DEBUG_RATE(mvm, "create station rate scale window\n");
 
 	lq_sta->pers.drv = mvm;
-	lq_sta->pers.sta_id = mvmsta->sta_id;
+	lq_sta->pers.sta_id = mvmsta->deflink.sta_id;
 	lq_sta->pers.chains = 0;
 	memset(lq_sta->pers.chain_signal, 0, sizeof(lq_sta->pers.chain_signal));
 	lq_sta->pers.last_rssi = S8_MIN;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index bd135b7168cb..1b0e99377156 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -754,7 +754,7 @@ static int rs_collect_tlc_data(struct iwl_mvm *mvm,
 		return -EINVAL;
 
 	if (tbl->column != RS_COLUMN_INVALID) {
-		struct lq_sta_pers *pers = &mvmsta->lq_sta.rs_drv.pers;
+		struct lq_sta_pers *pers = &mvmsta->deflink.lq_sta.rs_drv.pers;
 
 		pers->tx_stats[tbl->column][scale_index].total += attempts;
 		pers->tx_stats[tbl->column][scale_index].success += successes;
@@ -2599,7 +2599,7 @@ void rs_update_last_rssi(struct iwl_mvm *mvm,
 			 struct iwl_mvm_sta *mvmsta,
 			 struct ieee80211_rx_status *rx_status)
 {
-	struct iwl_lq_sta *lq_sta = &mvmsta->lq_sta.rs_drv;
+	struct iwl_lq_sta *lq_sta = &mvmsta->deflink.lq_sta.rs_drv;
 	int i;
 
 	lq_sta->pers.chains = rx_status->chains;
@@ -2711,7 +2711,7 @@ static void *rs_drv_alloc_sta(void *mvm_rate, struct ieee80211_sta *sta,
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_op_mode *op_mode = (struct iwl_op_mode *)mvm_rate;
 	struct iwl_mvm *mvm  = IWL_OP_MODE_GET_MVM(op_mode);
-	struct iwl_lq_sta *lq_sta = &mvmsta->lq_sta.rs_drv;
+	struct iwl_lq_sta *lq_sta = &mvmsta->deflink.lq_sta.rs_drv;
 
 	IWL_DEBUG_RATE(mvm, "create station rate scale window\n");
 
@@ -2917,18 +2917,18 @@ static void rs_drv_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
 	struct ieee80211_sta_vht_cap *vht_cap = &sta->deflink.vht_cap;
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_lq_sta *lq_sta = &mvmsta->lq_sta.rs_drv;
+	struct iwl_lq_sta *lq_sta = &mvmsta->deflink.lq_sta.rs_drv;
 	struct ieee80211_supported_band *sband;
 	unsigned long supp; /* must be unsigned long for for_each_set_bit */
 
-	lockdep_assert_held(&mvmsta->lq_sta.rs_drv.pers.lock);
+	lockdep_assert_held(&mvmsta->deflink.lq_sta.rs_drv.pers.lock);
 
 	/* clear all non-persistent lq data */
 	memset(lq_sta, 0, offsetof(typeof(*lq_sta), pers));
 
 	sband = hw->wiphy->bands[band];
 
-	lq_sta->lq.sta_id = mvmsta->sta_id;
+	lq_sta->lq.sta_id = mvmsta->deflink.sta_id;
 	mvmsta->amsdu_enabled = 0;
 	mvmsta->max_amsdu_len = sta->cur->max_amsdu_len;
 
@@ -2940,7 +2940,7 @@ static void rs_drv_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 
 	IWL_DEBUG_RATE(mvm,
 		       "LQ: *** rate scale station global init for station %d ***\n",
-		       mvmsta->sta_id);
+		       mvmsta->deflink.sta_id);
 	/* TODO: what is a good starting rate for STA? About middle? Maybe not
 	 * the lowest or the highest rate.. Could consider using RSSI from
 	 * previous packets? Need to have IEEE 802.1X auth succeed immediately
@@ -3032,7 +3032,7 @@ static void __iwl_mvm_rs_tx_status(struct iwl_mvm *mvm,
 	u8 lq_color = RS_DRV_DATA_LQ_COLOR_GET(tlc_info);
 	u32 tx_resp_hwrate = (uintptr_t)info->status.status_driver_data[1];
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_lq_sta *lq_sta = &mvmsta->lq_sta.rs_drv;
+	struct iwl_lq_sta *lq_sta = &mvmsta->deflink.lq_sta.rs_drv;
 
 	if (!lq_sta->pers.drv) {
 		IWL_DEBUG_RATE(mvm, "Rate scaling not initialized yet.\n");
@@ -3256,11 +3256,11 @@ void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	/* If it's locked we are in middle of init flow
 	 * just wait for next tx status to update the lq_sta data
 	 */
-	if (!spin_trylock(&mvmsta->lq_sta.rs_drv.pers.lock))
+	if (!spin_trylock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock))
 		return;
 
 	__iwl_mvm_rs_tx_status(mvm, sta, tid, info, ndp);
-	spin_unlock(&mvmsta->lq_sta.rs_drv.pers.lock);
+	spin_unlock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock);
 }
 
 #ifdef CONFIG_MAC80211_DEBUGFS
@@ -3436,7 +3436,7 @@ static void rs_bfer_active_iter(void *_data,
 {
 	struct rs_bfer_active_iter_data *data = _data;
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_lq_cmd *lq_cmd = &mvmsta->lq_sta.rs_drv.lq;
+	struct iwl_lq_cmd *lq_cmd = &mvmsta->deflink.lq_sta.rs_drv.lq;
 	u32 ss_params = le32_to_cpu(lq_cmd->ss_params);
 
 	if (sta == data->exclude_sta)
@@ -3467,7 +3467,8 @@ static int rs_bfer_priority(struct iwl_mvm_sta *sta)
 		prio = 1;
 		break;
 	default:
-		WARN_ONCE(true, "viftype %d sta_id %d", viftype, sta->sta_id);
+		WARN_ONCE(true, "viftype %d sta_id %d", viftype,
+			  sta->deflink.sta_id);
 		prio = -1;
 	}
 
@@ -3544,12 +3545,12 @@ static void rs_set_lq_ss_params(struct iwl_mvm *mvm,
 	}
 
 	IWL_DEBUG_RATE(mvm, "Found existing sta %d with BFER activated\n",
-		       bfer_mvmsta->sta_id);
+		       bfer_mvmsta->deflink.sta_id);
 
 	/* Disallow BFER on another STA if active and we're a higher priority */
 	if (rs_bfer_priority_cmp(mvmsta, bfer_mvmsta) > 0) {
 		struct iwl_lq_cmd *bfersta_lq_cmd =
-			&bfer_mvmsta->lq_sta.rs_drv.lq;
+			&bfer_mvmsta->deflink.lq_sta.rs_drv.lq;
 		u32 bfersta_ss_params = le32_to_cpu(bfersta_lq_cmd->ss_params);
 
 		bfersta_ss_params &= ~LQ_SS_BFER_ALLOWED;
@@ -3559,7 +3560,7 @@ static void rs_set_lq_ss_params(struct iwl_mvm *mvm,
 		ss_params |= LQ_SS_BFER_ALLOWED;
 		IWL_DEBUG_RATE(mvm,
 			       "Lower priority BFER sta found (%d). Switch BFER\n",
-			       bfer_mvmsta->sta_id);
+			       bfer_mvmsta->deflink.sta_id);
 	}
 out:
 	lq_cmd->ss_params = cpu_to_le32(ss_params);
@@ -3744,7 +3745,7 @@ static ssize_t rs_sta_dbgfs_scale_table_read(struct file *file,
 
 	struct iwl_lq_sta *lq_sta = file->private_data;
 	struct iwl_mvm_sta *mvmsta =
-		container_of(lq_sta, struct iwl_mvm_sta, lq_sta.rs_drv);
+		container_of(lq_sta, struct iwl_mvm_sta, deflink.lq_sta.rs_drv);
 	struct iwl_mvm *mvm;
 	struct iwl_scale_tbl_info *tbl = &(lq_sta->lq_info[lq_sta->active_tbl]);
 	struct rs_rate *rate = &tbl->rate;
@@ -4045,7 +4046,8 @@ static void rs_drv_add_sta_debugfs(void *mvm, void *priv_sta,
 	struct iwl_lq_sta *lq_sta = priv_sta;
 	struct iwl_mvm_sta *mvmsta;
 
-	mvmsta = container_of(lq_sta, struct iwl_mvm_sta, lq_sta.rs_drv);
+	mvmsta = container_of(lq_sta, struct iwl_mvm_sta,
+			      deflink.lq_sta.rs_drv);
 
 	if (!mvmsta->vif)
 		return;
@@ -4102,9 +4104,9 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	} else {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 
-		spin_lock(&mvmsta->lq_sta.rs_drv.pers.lock);
+		spin_lock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock);
 		rs_drv_rate_init(mvm, sta, band);
-		spin_unlock(&mvmsta->lq_sta.rs_drv.pers.lock);
+		spin_unlock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock);
 	}
 }
 
@@ -4121,7 +4123,7 @@ void iwl_mvm_rate_control_unregister(void)
 static int rs_drv_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 				bool enable)
 {
-	struct iwl_lq_cmd *lq = &mvmsta->lq_sta.rs_drv.lq;
+	struct iwl_lq_cmd *lq = &mvmsta->deflink.lq_sta.rs_drv.lq;
 
 	lockdep_assert_held(&mvm->mutex);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index 4105b79db467..e08dca8d2baa 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -241,7 +241,7 @@ static void iwl_mvm_rx_handle_tcm(struct iwl_mvm *mvm,
 	     !mvmvif->deflink.queue_params[IEEE80211_AC_VI].uapsd &&
 	     !mvmvif->deflink.queue_params[IEEE80211_AC_BE].uapsd &&
 	     !mvmvif->deflink.queue_params[IEEE80211_AC_BK].uapsd) ||
-	    mvmsta->sta_id != mvmvif->deflink.ap_sta_id)
+	    mvmsta->deflink.sta_id != mvmvif->deflink.ap_sta_id)
 		return;
 
 	if (rate_n_flags & RATE_MCS_HT_MSK_V1) {
@@ -712,14 +712,14 @@ static void iwl_mvm_stats_energy_iter(void *_data,
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	u8 *energy = _data;
-	u32 sta_id = mvmsta->sta_id;
+	u32 sta_id = mvmsta->deflink.sta_id;
 
 	if (WARN_ONCE(sta_id >= IWL_MVM_STATION_COUNT_MAX, "sta_id %d >= %d",
 		      sta_id, IWL_MVM_STATION_COUNT_MAX))
 		return;
 
 	if (energy[sta_id])
-		mvmsta->avg_energy = energy[sta_id];
+		mvmsta->deflink.avg_energy = energy[sta_id];
 
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 6286132f165a..b160ae137c4a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -979,9 +979,10 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
 		return false;
 	}
 
-	if (WARN(tid != baid_data->tid || mvm_sta->sta_id != baid_data->sta_id,
+	if (WARN(tid != baid_data->tid ||
+		 mvm_sta->deflink.sta_id != baid_data->sta_id,
 		 "baid 0x%x is mapped to sta:%d tid:%d, but was received for sta:%d tid:%d\n",
-		 baid, baid_data->sta_id, baid_data->tid, mvm_sta->sta_id,
+		 baid, baid_data->sta_id, baid_data->tid, mvm_sta->deflink.sta_id,
 		 tid))
 		return false;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index c3b81021898b..d01c67e1ce2a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -119,7 +119,7 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_add_sta_cmd add_sta_cmd = {
-		.sta_id = mvm_sta->sta_id,
+		.sta_id = mvm_sta->deflink.sta_id,
 		.mac_id_n_color = cpu_to_le32(mvm_sta->mac_id_n_color),
 		.add_modify = update ? 1 : 0,
 		.station_flags_msk = cpu_to_le32(STA_FLG_FAT_EN_MSK |
@@ -314,7 +314,7 @@ static int iwl_mvm_invalidate_sta_queue(struct iwl_mvm *mvm, int queue,
 	mvmsta->tid_disable_agg |= disable_agg_tids;
 
 	cmd.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color);
-	cmd.sta_id = mvmsta->sta_id;
+	cmd.sta_id = mvmsta->deflink.sta_id;
 	cmd.add_modify = STA_MODE_MODIFY;
 	cmd.modify_mask = STA_MODIFY_QUEUES;
 	if (disable_agg_tids)
@@ -856,15 +856,15 @@ static int iwl_mvm_sta_alloc_queue_tvqm(struct iwl_mvm *mvm,
 
 	IWL_DEBUG_TX_QUEUES(mvm,
 			    "Allocating queue for sta %d on tid %d\n",
-			    mvmsta->sta_id, tid);
-	queue = iwl_mvm_tvqm_enable_txq(mvm, sta, mvmsta->sta_id,
+			    mvmsta->deflink.sta_id, tid);
+	queue = iwl_mvm_tvqm_enable_txq(mvm, sta, mvmsta->deflink.sta_id,
 					tid, wdg_timeout);
 	if (queue < 0)
 		return queue;
 
 	mvmtxq->txq_id = queue;
 	mvm->tvqm_info[queue].txq_tid = tid;
-	mvm->tvqm_info[queue].sta_id = mvmsta->sta_id;
+	mvm->tvqm_info[queue].sta_id = mvmsta->deflink.sta_id;
 
 	IWL_DEBUG_TX_QUEUES(mvm, "Allocated queue is %d\n", queue);
 
@@ -1049,7 +1049,7 @@ static void iwl_mvm_unshare_queue(struct iwl_mvm *mvm, int queue)
 		mvmsta->tid_disable_agg &= ~BIT(tid);
 
 		cmd.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color);
-		cmd.sta_id = mvmsta->sta_id;
+		cmd.sta_id = mvmsta->deflink.sta_id;
 		cmd.add_modify = STA_MODE_MODIFY;
 		cmd.modify_mask = STA_MODIFY_TID_DISABLE_TX;
 		cmd.tfd_queue_msk = cpu_to_le32(mvmsta->tfd_queue_msk);
@@ -1274,7 +1274,7 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm,
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_trans_txq_scd_cfg cfg = {
 		.fifo = iwl_mvm_mac_ac_to_tx_fifo(mvm, ac),
-		.sta_id = mvmsta->sta_id,
+		.sta_id = mvmsta->deflink.sta_id,
 		.tid = tid,
 		.frame_limit = IWL_FRAME_LIMIT,
 	};
@@ -1300,7 +1300,7 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm,
 	spin_unlock_bh(&mvmsta->lock);
 
 	if (tid == IWL_MAX_TID_COUNT) {
-		queue = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id,
+		queue = iwl_mvm_find_free_queue(mvm, mvmsta->deflink.sta_id,
 						IWL_MVM_DQA_MIN_MGMT_QUEUE,
 						IWL_MVM_DQA_MAX_MGMT_QUEUE);
 		if (queue >= IWL_MVM_DQA_MIN_MGMT_QUEUE)
@@ -1319,12 +1319,12 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm,
 	}
 
 	if (queue < 0)
-		queue = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id,
+		queue = iwl_mvm_find_free_queue(mvm, mvmsta->deflink.sta_id,
 						IWL_MVM_DQA_MIN_DATA_QUEUE,
 						IWL_MVM_DQA_MAX_DATA_QUEUE);
 	if (queue < 0) {
 		/* try harder - perhaps kill an inactive queue */
-		queue = iwl_mvm_inactivity_check(mvm, mvmsta->sta_id);
+		queue = iwl_mvm_inactivity_check(mvm, mvmsta->deflink.sta_id);
 	}
 
 	/* No free queue - we'll have to share */
@@ -1364,7 +1364,7 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm,
 	IWL_DEBUG_TX_QUEUES(mvm,
 			    "Allocating %squeue #%d to sta %d on tid %d\n",
 			    shared_queue ? "shared " : "", queue,
-			    mvmsta->sta_id, tid);
+			    mvmsta->deflink.sta_id, tid);
 
 	if (shared_queue) {
 		/* Disable any open aggs on this queue */
@@ -1431,7 +1431,7 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm,
 
 out_err:
 	queue_tmp = queue;
-	iwl_mvm_disable_txq(mvm, sta, mvmsta->sta_id, &queue_tmp, tid);
+	iwl_mvm_disable_txq(mvm, sta, mvmsta->deflink.sta_id, &queue_tmp, tid);
 
 	return ret;
 }
@@ -1500,12 +1500,12 @@ static int iwl_mvm_reserve_sta_stream(struct iwl_mvm *mvm,
 	     IWL_MVM_QUEUE_FREE))
 		queue = IWL_MVM_DQA_BSS_CLIENT_QUEUE;
 	else
-		queue = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id,
+		queue = iwl_mvm_find_free_queue(mvm, mvmsta->deflink.sta_id,
 						IWL_MVM_DQA_MIN_DATA_QUEUE,
 						IWL_MVM_DQA_MAX_DATA_QUEUE);
 	if (queue < 0) {
 		/* try again - this time kick out a queue if needed */
-		queue = iwl_mvm_inactivity_check(mvm, mvmsta->sta_id);
+		queue = iwl_mvm_inactivity_check(mvm, mvmsta->deflink.sta_id);
 		if (queue < 0) {
 			IWL_ERR(mvm, "No available queues for new station\n");
 			return -ENOSPC;
@@ -1516,7 +1516,7 @@ static int iwl_mvm_reserve_sta_stream(struct iwl_mvm *mvm,
 	mvmsta->reserved_queue = queue;
 
 	IWL_DEBUG_TX_QUEUES(mvm, "Reserving data queue #%d for sta_id %d\n",
-			    queue, mvmsta->sta_id);
+			    queue, mvmsta->deflink.sta_id);
 
 	return 0;
 }
@@ -1536,7 +1536,7 @@ static void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
 		iwl_mvm_get_wd_timeout(mvm, mvm_sta->vif, false, false);
 	int i;
 	struct iwl_trans_txq_scd_cfg cfg = {
-		.sta_id = mvm_sta->sta_id,
+		.sta_id = mvm_sta->deflink.sta_id,
 		.frame_limit = IWL_FRAME_LIMIT,
 	};
 
@@ -1558,9 +1558,9 @@ static void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
 		if (iwl_mvm_has_new_tx_api(mvm)) {
 			IWL_DEBUG_TX_QUEUES(mvm,
 					    "Re-mapping sta %d tid %d\n",
-					    mvm_sta->sta_id, i);
+					    mvm_sta->deflink.sta_id, i);
 			txq_id = iwl_mvm_tvqm_enable_txq(mvm, sta,
-							 mvm_sta->sta_id,
+							 mvm_sta->deflink.sta_id,
 							 i, wdg);
 			/*
 			 * on failures, just set it to IWL_MVM_INVALID_QUEUE
@@ -1589,7 +1589,8 @@ static void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
 
 			IWL_DEBUG_TX_QUEUES(mvm,
 					    "Re-mapping sta %d tid %d to queue %d\n",
-					    mvm_sta->sta_id, i, txq_id);
+					    mvm_sta->deflink.sta_id, i,
+					    txq_id);
 
 			iwl_mvm_enable_txq(mvm, sta, txq_id, seq, &cfg, wdg);
 			mvm->queue_info[txq_id].status = IWL_MVM_QUEUE_READY;
@@ -1658,7 +1659,9 @@ static int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	lockdep_assert_held(&mvm->mutex);
 
-	mvm_sta->sta_id = sta_id;
+	mvm_sta->deflink.sta_id = sta_id;
+	rcu_assign_pointer(mvm_sta->link[0], &mvm_sta->deflink);
+
 	mvm_sta->mac_id_n_color = FW_CMD_ID_AND_COLOR(mvmvif->id,
 						      mvmvif->color);
 	mvm_sta->vif = vif;
@@ -1724,7 +1727,7 @@ static int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (iwl_mvm_has_tlc_offload(mvm))
 		iwl_mvm_rs_add_sta(mvm, mvm_sta);
 	else
-		spin_lock_init(&mvm_sta->lq_sta.rs_drv.pers.lock);
+		spin_lock_init(&mvm_sta->deflink.lq_sta.rs_drv.pers.lock);
 
 	iwl_mvm_toggle_tx_ant(mvm, &mvm_sta->tx_ant);
 
@@ -1747,7 +1750,7 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 		sta_id = iwl_mvm_find_free_sta_id(mvm,
 						  ieee80211_vif_type_p2p(vif));
 	else
-		sta_id = mvm_sta->sta_id;
+		sta_id = mvm_sta->deflink.sta_id;
 
 	if (sta_id == IWL_MVM_INVALID_STA)
 		return -ENOSPC;
@@ -1810,7 +1813,7 @@ int iwl_mvm_drain_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 	lockdep_assert_held(&mvm->mutex);
 
 	cmd.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color);
-	cmd.sta_id = mvmsta->sta_id;
+	cmd.sta_id = mvmsta->deflink.sta_id;
 	cmd.add_modify = STA_MODE_MODIFY;
 	cmd.station_flags = drain ? cpu_to_le32(STA_FLG_DRAIN_FLOW) : 0;
 	cmd.station_flags_msk = cpu_to_le32(STA_FLG_DRAIN_FLOW);
@@ -1825,12 +1828,12 @@ int iwl_mvm_drain_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 	switch (status & IWL_ADD_STA_STATUS_MASK) {
 	case ADD_STA_SUCCESS:
 		IWL_DEBUG_INFO(mvm, "Frames for staid %d will drained in fw\n",
-			       mvmsta->sta_id);
+			       mvmsta->deflink.sta_id);
 		break;
 	default:
 		ret = -EIO;
 		IWL_ERR(mvm, "Couldn't drain frames for staid %d\n",
-			mvmsta->sta_id);
+			mvmsta->deflink.sta_id);
 		break;
 	}
 
@@ -1882,7 +1885,7 @@ static void iwl_mvm_disable_sta_queues(struct iwl_mvm *mvm,
 		if (mvm_sta->tid_data[i].txq_id == IWL_MVM_INVALID_QUEUE)
 			continue;
 
-		iwl_mvm_disable_txq(mvm, sta, mvm_sta->sta_id,
+		iwl_mvm_disable_txq(mvm, sta, mvm_sta->deflink.sta_id,
 				    &mvm_sta->tid_data[i].txq_id, i);
 		mvm_sta->tid_data[i].txq_id = IWL_MVM_INVALID_QUEUE;
 	}
@@ -1926,7 +1929,7 @@ static bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-	u8 sta_id = mvm_sta->sta_id;
+	u8 sta_id = mvm_sta->deflink.sta_id;
 
 	lockdep_assert_held(&mvm->mutex);
 
@@ -2022,8 +2025,8 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 	if (iwl_mvm_sta_del(mvm, vif, sta, &ret))
 		return ret;
 
-	ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->sta_id);
-	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->sta_id], NULL);
+	ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->deflink.sta_id);
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id], NULL);
 
 	return ret;
 }
@@ -2654,7 +2657,7 @@ static int iwl_mvm_fw_baid_op_sta(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_add_sta_cmd cmd = {
 		.mac_id_n_color = cpu_to_le32(mvm_sta->mac_id_n_color),
-		.sta_id = mvm_sta->sta_id,
+		.sta_id = mvm_sta->deflink.sta_id,
 		.add_modify = STA_MODE_MODIFY,
 	};
 	u32 status;
@@ -2710,7 +2713,8 @@ static int iwl_mvm_fw_baid_op_cmd(struct iwl_mvm *mvm,
 	BUILD_BUG_ON(sizeof(struct iwl_rx_baid_cfg_resp) != sizeof(baid));
 
 	if (start) {
-		cmd.alloc.sta_id_mask = cpu_to_le32(BIT(mvm_sta->sta_id));
+		cmd.alloc.sta_id_mask =
+			cpu_to_le32(BIT(mvm_sta->deflink.sta_id));
 		cmd.alloc.tid = tid;
 		cmd.alloc.ssn = cpu_to_le16(ssn);
 		cmd.alloc.win_size = cpu_to_le16(buf_size);
@@ -2719,7 +2723,8 @@ static int iwl_mvm_fw_baid_op_cmd(struct iwl_mvm *mvm,
 		cmd.remove_v1.baid = cpu_to_le32(baid);
 		BUILD_BUG_ON(sizeof(cmd.remove_v1) > sizeof(cmd.remove));
 	} else {
-		cmd.remove.sta_id_mask = cpu_to_le32(BIT(mvm_sta->sta_id));
+		cmd.remove.sta_id_mask =
+			cpu_to_le32(BIT(mvm_sta->deflink.sta_id));
 		cmd.remove.tid = cpu_to_le32(tid);
 	}
 
@@ -2843,7 +2848,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			    iwl_mvm_rx_agg_session_expired, 0);
 		baid_data->mvm = mvm;
 		baid_data->tid = tid;
-		baid_data->sta_id = mvm_sta->sta_id;
+		baid_data->sta_id = mvm_sta->deflink.sta_id;
 
 		mvm_sta->tid_to_baid[tid] = baid;
 		if (timeout)
@@ -2858,7 +2863,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		 * RX is being processed in parallel
 		 */
 		IWL_DEBUG_HT(mvm, "Sta %d(%d) is assigned to BAID %d\n",
-			     mvm_sta->sta_id, tid, baid);
+			     mvm_sta->deflink.sta_id, tid, baid);
 		WARN_ON(rcu_access_pointer(mvm->baid_map[baid]));
 		rcu_assign_pointer(mvm->baid_map[baid], baid_data);
 	} else  {
@@ -2920,7 +2925,7 @@ int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	}
 
 	cmd.mac_id_n_color = cpu_to_le32(mvm_sta->mac_id_n_color);
-	cmd.sta_id = mvm_sta->sta_id;
+	cmd.sta_id = mvm_sta->deflink.sta_id;
 	cmd.add_modify = STA_MODE_MODIFY;
 	if (!iwl_mvm_has_new_tx_api(mvm))
 		cmd.modify_mask = STA_MODIFY_QUEUES;
@@ -3012,7 +3017,7 @@ int iwl_mvm_sta_tx_agg_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	 */
 	txq_id = mvmsta->tid_data[tid].txq_id;
 	if (txq_id == IWL_MVM_INVALID_QUEUE) {
-		ret = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id,
+		ret = iwl_mvm_find_free_queue(mvm, mvmsta->deflink.sta_id,
 					      IWL_MVM_DQA_MIN_DATA_QUEUE,
 					      IWL_MVM_DQA_MAX_DATA_QUEUE);
 		if (ret < 0) {
@@ -3050,7 +3055,8 @@ int iwl_mvm_sta_tx_agg_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	IWL_DEBUG_TX_QUEUES(mvm,
 			    "Start AGG: sta %d tid %d queue %d - ssn = %d, next_recl = %d\n",
-			    mvmsta->sta_id, tid, txq_id, tid_data->ssn,
+			    mvmsta->deflink.sta_id, tid, txq_id,
+			    tid_data->ssn,
 			    tid_data->next_reclaimed);
 
 	/*
@@ -3089,7 +3095,7 @@ int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	u16 ssn;
 
 	struct iwl_trans_txq_scd_cfg cfg = {
-		.sta_id = mvmsta->sta_id,
+		.sta_id = mvmsta->deflink.sta_id,
 		.tid = tid,
 		.frame_limit = buf_size,
 		.aggregate = true,
@@ -3161,7 +3167,7 @@ int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		}
 
 		ret = iwl_mvm_reconfig_scd(mvm, queue, cfg.fifo,
-					   mvmsta->sta_id, tid,
+					   mvmsta->deflink.sta_id, tid,
 					   buf_size, ssn);
 		if (ret) {
 			IWL_ERR(mvm,
@@ -3194,12 +3200,13 @@ out:
 	 */
 	mvmsta->max_agg_bufsize =
 		min(mvmsta->max_agg_bufsize, buf_size);
-	mvmsta->lq_sta.rs_drv.lq.agg_frame_cnt_limit = mvmsta->max_agg_bufsize;
+	mvmsta->deflink.lq_sta.rs_drv.lq.agg_frame_cnt_limit =
+		mvmsta->max_agg_bufsize;
 
 	IWL_DEBUG_HT(mvm, "Tx aggregation enabled on ra = %pM tid = %d\n",
 		     sta->addr, tid);
 
-	return iwl_mvm_send_lq_cmd(mvm, &mvmsta->lq_sta.rs_drv.lq);
+	return iwl_mvm_send_lq_cmd(mvm, &mvmsta->deflink.lq_sta.rs_drv.lq);
 }
 
 static void iwl_mvm_unreserve_agg_queue(struct iwl_mvm *mvm,
@@ -3248,7 +3255,8 @@ int iwl_mvm_sta_tx_agg_stop(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	txq_id = tid_data->txq_id;
 
 	IWL_DEBUG_TX_QUEUES(mvm, "Stop AGG: sta %d tid %d q %d state %d\n",
-			    mvmsta->sta_id, tid, txq_id, tid_data->state);
+			    mvmsta->deflink.sta_id, tid, txq_id,
+			    tid_data->state);
 
 	mvmsta->agg_tids &= ~BIT(tid);
 
@@ -3287,7 +3295,7 @@ int iwl_mvm_sta_tx_agg_stop(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	default:
 		IWL_ERR(mvm,
 			"Stopping AGG while state not ON or starting for %d on %d (%d)\n",
-			mvmsta->sta_id, tid, tid_data->state);
+			mvmsta->deflink.sta_id, tid, tid_data->state);
 		IWL_ERR(mvm,
 			"\ttid_data->txq_id = %d\n", tid_data->txq_id);
 		err = -EINVAL;
@@ -3313,7 +3321,8 @@ int iwl_mvm_sta_tx_agg_flush(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	spin_lock_bh(&mvmsta->lock);
 	txq_id = tid_data->txq_id;
 	IWL_DEBUG_TX_QUEUES(mvm, "Flush AGG: sta %d tid %d q %d state %d\n",
-			    mvmsta->sta_id, tid, txq_id, tid_data->state);
+			    mvmsta->deflink.sta_id, tid, txq_id,
+			    tid_data->state);
 	old_state = tid_data->state;
 	tid_data->state = IWL_AGG_OFF;
 	mvmsta->agg_tids &= ~BIT(tid);
@@ -3325,7 +3334,7 @@ int iwl_mvm_sta_tx_agg_flush(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		iwl_mvm_drain_sta(mvm, mvmsta, true);
 
 		if (iwl_mvm_has_new_tx_api(mvm)) {
-			if (iwl_mvm_flush_sta_tids(mvm, mvmsta->sta_id,
+			if (iwl_mvm_flush_sta_tids(mvm, mvmsta->deflink.sta_id,
 						   BIT(tid)))
 				IWL_ERR(mvm, "Couldn't flush the AGG queue\n");
 			iwl_trans_wait_txq_empty(mvm->trans, txq_id);
@@ -3690,7 +3699,7 @@ static int __iwl_mvm_set_sta_key(struct iwl_mvm *mvm,
 	if (sta) {
 		struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 
-		sta_id = mvm_sta->sta_id;
+		sta_id = mvm_sta->deflink.sta_id;
 		mfp = sta->mfp;
 	} else if (vif->type == NL80211_IFTYPE_AP &&
 		   !(keyconf->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {
@@ -3739,7 +3748,7 @@ int iwl_mvm_set_sta_key(struct iwl_mvm *mvm,
 			IWL_ERR(mvm, "Failed to find station\n");
 			return -EINVAL;
 		}
-		sta_id = mvm_sta->sta_id;
+		sta_id = mvm_sta->deflink.sta_id;
 
 		/*
 		 * It is possible that the 'sta' parameter is NULL, and thus
@@ -3834,7 +3843,7 @@ int iwl_mvm_remove_sta_key(struct iwl_mvm *mvm,
 	/* Get the station from the mvm local station table */
 	mvm_sta = iwl_mvm_get_key_sta(mvm, vif, sta);
 	if (mvm_sta)
-		sta_id = mvm_sta->sta_id;
+		sta_id = mvm_sta->deflink.sta_id;
 	else if (!sta && vif->type == NL80211_IFTYPE_AP && mcast)
 		sta_id = iwl_mvm_vif_from_mac80211(vif)->deflink.mcast_sta.sta_id;
 
@@ -3892,7 +3901,7 @@ void iwl_mvm_update_tkip_key(struct iwl_mvm *mvm,
 	mvm_sta = iwl_mvm_get_key_sta(mvm, vif, sta);
 	if (WARN_ON_ONCE(!mvm_sta))
 		goto unlock;
-	iwl_mvm_send_sta_key(mvm, mvm_sta->sta_id, keyconf, mcast,
+	iwl_mvm_send_sta_key(mvm, mvm_sta->deflink.sta_id, keyconf, mcast,
 			     iv32, phase1key, CMD_ASYNC, keyconf->hw_key_idx,
 			     mfp);
 
@@ -3906,7 +3915,7 @@ void iwl_mvm_sta_modify_ps_wake(struct iwl_mvm *mvm,
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_add_sta_cmd cmd = {
 		.add_modify = STA_MODE_MODIFY,
-		.sta_id = mvmsta->sta_id,
+		.sta_id = mvmsta->deflink.sta_id,
 		.station_flags_msk = cpu_to_le32(STA_FLG_PS),
 		.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color),
 	};
@@ -3927,7 +3936,7 @@ void iwl_mvm_sta_modify_sleep_tx_count(struct iwl_mvm *mvm,
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_add_sta_cmd cmd = {
 		.add_modify = STA_MODE_MODIFY,
-		.sta_id = mvmsta->sta_id,
+		.sta_id = mvmsta->deflink.sta_id,
 		.modify_mask = STA_MODIFY_SLEEPING_STA_TX_COUNT,
 		.sleep_tx_count = cpu_to_le16(cnt),
 		.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color),
@@ -4024,7 +4033,7 @@ void iwl_mvm_sta_modify_disable_tx(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_add_sta_cmd cmd = {
 		.add_modify = STA_MODE_MODIFY,
-		.sta_id = mvmsta->sta_id,
+		.sta_id = mvmsta->deflink.sta_id,
 		.station_flags = disable ? cpu_to_le32(STA_FLG_DISABLE_TX) : 0,
 		.station_flags_msk = cpu_to_le32(STA_FLG_DISABLE_TX),
 		.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index a0c13c409ccd..bfd44c489286 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -330,9 +330,27 @@ struct iwl_mvm_rxq_dup_data {
 	u8 last_sub_frame[IWL_MAX_TID_COUNT + 1];
 } ____cacheline_aligned_in_smp;
 
+/**
+ * struct iwl_mvm_link_sta - link specific parameters of a station
+ * @rcu_head: used for freeing the data
+ * @sta_id: the index of the station in the fw
+ * @lq_sta: holds rate scaling data, either for the case when RS is done in
+ *	the driver - %rs_drv or in the FW - %rs_fw.
+ * @avg_energy: energy as reported by FW statistics notification
+ */
+struct iwl_mvm_link_sta {
+	struct rcu_head rcu_head;
+	u32 sta_id;
+	union {
+		struct iwl_lq_sta_rs_fw rs_fw;
+		struct iwl_lq_sta rs_drv;
+	} lq_sta;
+
+	u8 avg_energy;
+};
+
 /**
  * struct iwl_mvm_sta - representation of a station in the driver
- * @sta_id: the index of the station in the fw (will be replaced by id_n_color)
  * @tfd_queue_msk: the tfd queues used by the station
  * @mac_id_n_color: the MAC context this station is linked to
  * @tid_disable_agg: bitmap: if bit(tid) is set, the fw won't send ampdus for
@@ -347,8 +365,6 @@ struct iwl_mvm_rxq_dup_data {
  * and from Tx response flow, it needs a spinlock.
  * @tid_data: per tid data + mgmt. Look at %iwl_mvm_tid_data.
  * @tid_to_baid: a simple map of TID to baid
- * @lq_sta: holds rate scaling data, either for the case when RS is done in
- *	the driver - %rs_drv or in the FW - %rs_fw.
  * @reserved_queue: the queue reserved for this STA for DQA purposes
  *	Every STA has is given one reserved queue to allow it to operate. If no
  *	such queue can be guaranteed, the STA addition will fail.
@@ -374,6 +390,12 @@ struct iwl_mvm_rxq_dup_data {
  *	used during connection establishment (e.g. for the 4 way handshake
  *	exchange).
  * @pairwise_cipher: used to feed iwlmei upon authorization
+ * @deflink: the default link station, for non-MLO STA, all link specific data
+ *	is accessed via deflink (or link[0]). For MLO, it will hold data of the
+ *	first added link STA.
+ * @link: per link sta entries. For non-MLO only link[0] holds data. For MLO,
+ *	link[0] points to deflink and link[link_id] is allocated when new link
+ *	sta is added.
  *
  * When mac80211 creates a station it reserves some space (hw->sta_data_size)
  * in the structure for use by driver. This structure is placed in that
@@ -381,7 +403,6 @@ struct iwl_mvm_rxq_dup_data {
  *
  */
 struct iwl_mvm_sta {
-	u32 sta_id;
 	u32 tfd_queue_msk;
 	u32 mac_id_n_color;
 	u16 tid_disable_agg;
@@ -393,10 +414,6 @@ struct iwl_mvm_sta {
 	spinlock_t lock;
 	struct iwl_mvm_tid_data tid_data[IWL_MAX_TID_COUNT + 1];
 	u8 tid_to_baid[IWL_MAX_TID_COUNT];
-	union {
-		struct iwl_lq_sta_rs_fw rs_fw;
-		struct iwl_lq_sta rs_drv;
-	} lq_sta;
 	struct ieee80211_vif *vif;
 	struct iwl_mvm_key_pn __rcu *ptk_pn[4];
 	struct iwl_mvm_rxq_dup_data *dup_data;
@@ -414,9 +431,11 @@ struct iwl_mvm_sta {
 	bool sleeping;
 	u8 agg_tids;
 	u8 sleep_tx_count;
-	u8 avg_energy;
 	u8 tx_ant;
 	u32 pairwise_cipher;
+
+	struct iwl_mvm_link_sta deflink;
+	struct iwl_mvm_link_sta __rcu *link[IEEE80211_MLD_MAX_NUM_LINKS];
 };
 
 u16 iwl_mvm_tid_queued(struct iwl_mvm *mvm, struct iwl_mvm_tid_data *tid_data);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c b/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c
index 674dd137fb9f..dae6f2a1aad9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c
@@ -369,7 +369,7 @@ iwl_mvm_tdls_config_channel_switch(struct iwl_mvm *mvm,
 		goto out;
 	}
 	mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	cmd.peer_sta_id = cpu_to_le32(mvmsta->sta_id);
+	cmd.peer_sta_id = cpu_to_le32(mvmsta->deflink.sta_id);
 
 	if (!chandef) {
 		if (mvm->tdls_cs.state == IWL_MVM_TDLS_SW_REQ_SENT &&
@@ -414,7 +414,7 @@ iwl_mvm_tdls_config_channel_switch(struct iwl_mvm *mvm,
 	}
 
 	iwl_mvm_set_tx_cmd(mvm, skb, &tail->frame.tx_cmd, info,
-			   mvmsta->sta_id);
+			   mvmsta->deflink.sta_id);
 
 	iwl_mvm_set_tx_cmd_rate(mvm, &tail->frame.tx_cmd, info, sta,
 				hdr->frame_control);
@@ -431,7 +431,7 @@ iwl_mvm_tdls_config_channel_switch(struct iwl_mvm *mvm,
 
 	/* channel switch has started, update state */
 	if (type != TDLS_MOVE_CH) {
-		mvm->tdls_cs.cur_sta_id = mvmsta->sta_id;
+		mvm->tdls_cs.cur_sta_id = mvmsta->deflink.sta_id;
 		iwl_mvm_tdls_update_cs_state(mvm,
 					     type == TDLS_SEND_CHAN_SW_REQ ?
 					     IWL_MVM_TDLS_SW_REQ_SENT :
@@ -541,7 +541,7 @@ iwl_mvm_tdls_channel_switch(struct ieee80211_hw *hw,
 	}
 
 	mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	mvm->tdls_cs.peer.sta_id = mvmsta->sta_id;
+	mvm->tdls_cs.peer.sta_id = mvmsta->deflink.sta_id;
 	mvm->tdls_cs.peer.chandef = *chandef;
 	mvm->tdls_cs.peer.initiator = sta->tdls_initiator;
 	mvm->tdls_cs.peer.op_class = oper_class;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 890fe93896ba..1737f2ad21ad 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -1084,7 +1084,7 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (WARN_ON_ONCE(!mvmsta))
 		return -1;
 
-	if (WARN_ON_ONCE(mvmsta->sta_id == IWL_MVM_INVALID_STA))
+	if (WARN_ON_ONCE(mvmsta->deflink.sta_id == IWL_MVM_INVALID_STA))
 		return -1;
 
 	if (unlikely(ieee80211_is_any_nullfunc(fc)) && sta->deflink.he_cap.has_he)
@@ -1094,7 +1094,7 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 		iwl_mvm_probe_resp_set_noa(mvm, skb);
 
 	dev_cmd = iwl_mvm_set_tx_params(mvm, skb, info, hdrlen,
-					sta, mvmsta->sta_id);
+					sta, mvmsta->deflink.sta_id);
 	if (!dev_cmd)
 		goto drop;
 
@@ -1170,7 +1170,7 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX(mvm, "TX to [%d|%d] Q:%d - seq: 0x%x len %d\n",
-		     mvmsta->sta_id, tid, txq_id,
+		     mvmsta->deflink.sta_id, tid, txq_id,
 		     IEEE80211_SEQ_TO_SN(seq_number), skb->len);
 
 	/* From now on, we cannot access info->control */
@@ -1205,7 +1205,8 @@ drop_unlock_sta:
 	iwl_trans_free_tx_cmd(mvm->trans, dev_cmd);
 	spin_unlock(&mvmsta->lock);
 drop:
-	IWL_DEBUG_TX(mvm, "TX to [%d|%d] dropped\n", mvmsta->sta_id, tid);
+	IWL_DEBUG_TX(mvm, "TX to [%d|%d] dropped\n", mvmsta->deflink.sta_id,
+		     tid);
 	return -1;
 }
 
@@ -1222,7 +1223,7 @@ int iwl_mvm_tx_skb_sta(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (WARN_ON_ONCE(!mvmsta))
 		return -1;
 
-	if (WARN_ON_ONCE(mvmsta->sta_id == IWL_MVM_INVALID_STA))
+	if (WARN_ON_ONCE(mvmsta->deflink.sta_id == IWL_MVM_INVALID_STA))
 		return -1;
 
 	memcpy(&info, skb->cb, sizeof(info));
@@ -2230,17 +2231,22 @@ free_rsp:
 
 int iwl_mvm_flush_sta(struct iwl_mvm *mvm, void *sta, bool internal)
 {
-	struct iwl_mvm_int_sta *int_sta = sta;
-	struct iwl_mvm_sta *mvm_sta = sta;
+	u32 sta_id, tfd_queue_msk;
 
-	BUILD_BUG_ON(offsetof(struct iwl_mvm_int_sta, sta_id) !=
-		     offsetof(struct iwl_mvm_sta, sta_id));
+	if (internal) {
+		struct iwl_mvm_int_sta *int_sta = sta;
 
-	if (iwl_mvm_has_new_tx_api(mvm))
-		return iwl_mvm_flush_sta_tids(mvm, mvm_sta->sta_id, 0xffff);
+		sta_id = int_sta->sta_id;
+		tfd_queue_msk = int_sta->tfd_queue_msk;
+	} else {
+		struct iwl_mvm_sta *mvm_sta = sta;
 
-	if (internal)
-		return iwl_mvm_flush_tx_path(mvm, int_sta->tfd_queue_msk);
+		sta_id = mvm_sta->deflink.sta_id;
+		tfd_queue_msk = mvm_sta->tfd_queue_msk;
+	}
+
+	if (iwl_mvm_has_new_tx_api(mvm))
+		return iwl_mvm_flush_sta_tids(mvm, sta_id, 0xffff);
 
-	return iwl_mvm_flush_tx_path(mvm, mvm_sta->tfd_queue_msk);
+	return iwl_mvm_flush_tx_path(mvm, tfd_queue_msk);
 }
-- 
cgit v1.2.3-70-g09d2


From 9e2de88d28b235a65db6209e331496f8f2fd203a Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:43 +0300
Subject: wifi: iwlwifi: mvm: add an indication that the new MLD API is used

WE can't mix between the new MLD API and the old API.
I.e. - we can't send one of the new cmds and then one of the old ones.
This will cause a FW assert. So we need an indication what API should be
used. We use the new API if:
1. FW supports it
2. We are registered to mac80211 with the new MLD ops
Add an indication which will only be true if both conditions are true.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.5756b0907403.I0adce36d1783cce23d0e080e3c4a8953db33b515@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 1 +
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 72056cccc77e..443f0b80b63e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1155,6 +1155,7 @@ struct iwl_mvm {
 	unsigned long last_reset_or_resume_time_jiffies;
 
 	bool sta_remove_requires_queue_remove;
+	bool mld_api_is_used;
 
 	bool pldr_sync;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 70926bb1df6c..5c8784ddfa66 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1295,6 +1295,9 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	mvm->sta_remove_requires_queue_remove =
 		trans_cfg.queue_alloc_cmd_ver > 0;
 
+	/* Until we register with the MLD ops - we don't use the MLD API */
+	mvm->mld_api_is_used = false;
+
 	/* Configure transport layer */
 	iwl_trans_configure(mvm->trans, &trans_cfg);
 
-- 
cgit v1.2.3-70-g09d2


From 87f7e2435cb7ddafbc644b759aa1a9dffaab1515 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:44 +0300
Subject: wifi: iwlwifi: mvm: add sta handling flows for MLD mode

In MLD mode we have a new STA cmd. As a result, it is also changes
the flows of adding/updating/removing and handling state of
a station. Add these flows.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.b5548cfd8fe3.I70f9c8f3c95e18d5c9af0a5681e0830893509531@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  58 ++++--
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  18 ++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 205 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   7 +
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       |  25 +--
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  45 +++++
 6 files changed, 333 insertions(+), 25 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 282c15e74f08..b730b605da74 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2103,9 +2103,8 @@ static void iwl_mvm_get_optimal_ppe_info(struct iwl_he_pkt_ext_v2 *pkt_ext,
 }
 
 /* Set the pkt_ext field according to PPE Thresholds element */
-static int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm,
-				   struct ieee80211_sta *sta,
-				   struct iwl_he_pkt_ext_v2 *pkt_ext)
+int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			    struct iwl_he_pkt_ext_v2 *pkt_ext)
 {
 	u8 nominal_padding;
 	int i, ret = 0;
@@ -2250,7 +2249,7 @@ bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 			       IEEE80211_HE_MAC_CAP2_ACK_EN));
 }
 
-static __le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta)
+__le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta)
 {
 	u8 *mac_cap_info = &sta->deflink.he_cap.he_cap_elem.mac_cap_info[0];
 	__le32 htc_flags = 0;
@@ -3425,11 +3424,37 @@ static void iwl_mvm_mei_host_associated(struct iwl_mvm *mvm,
 #endif
 }
 
+static int iwl_mvm_mac_ctxt_changed_wrapper(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
+					    bool force_assoc_off)
+{
+	return iwl_mvm_mac_ctxt_changed(mvm, vif, force_assoc_off, NULL);
+}
+
 static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 				 struct ieee80211_vif *vif,
 				 struct ieee80211_sta *sta,
 				 enum ieee80211_sta_state old_state,
 				 enum ieee80211_sta_state new_state)
+{
+	struct iwl_mvm_sta_state_ops callbacks = {
+		.add_sta = iwl_mvm_add_sta,
+		.update_sta = iwl_mvm_update_sta,
+		.rm_sta = iwl_mvm_rm_sta,
+		.mac_ctxt_changed = iwl_mvm_mac_ctxt_changed_wrapper,
+	};
+
+	return iwl_mvm_mac_sta_state_common(hw, vif, sta, old_state, new_state,
+					    &callbacks);
+}
+
+/* Common part for MLD and non-MLD modes */
+int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_sta *sta,
+				 enum ieee80211_sta_state old_state,
+				 enum ieee80211_sta_state new_state,
+				 struct iwl_mvm_sta_state_ops *callbacks)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -3507,7 +3532,7 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 			goto out_unlock;
 		}
 
-		ret = iwl_mvm_add_sta(mvm, vif, sta);
+		ret = callbacks->add_sta(mvm, vif, sta);
 		if (sta->tdls && ret == 0) {
 			iwl_mvm_recalc_tdls_state(mvm, vif, true);
 			iwl_mvm_tdls_check_trigger(mvm, vif, sta->addr,
@@ -3529,11 +3554,12 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 		if (vif->type == NL80211_IFTYPE_AP) {
 			vif->bss_conf.he_support = sta->deflink.he_cap.has_he;
 			mvmvif->ap_assoc_sta_count++;
-			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
-			if ((vif->bss_conf.he_support &&
+			callbacks->mac_ctxt_changed(mvm, vif, false);
+			if (!mvm->mld_api_is_used &&
+			    ((vif->bss_conf.he_support &&
 			     !iwlwifi_mod_params.disable_11ax) ||
 			    (vif->bss_conf.eht_support &&
-			     !iwlwifi_mod_params.disable_11be))
+			     !iwlwifi_mod_params.disable_11be)))
 				iwl_mvm_cfg_he_sta(mvm, vif,
 						   mvm_sta->deflink.sta_id);
 		} else if (vif->type == NL80211_IFTYPE_STATION) {
@@ -3543,13 +3569,19 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 			if (sta->deflink.he_cap.has_he)
 				iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif);
 
-			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
+			callbacks->mac_ctxt_changed(mvm, vif, false);
+
+			if (mvm->mld_api_is_used)
+				iwl_mvm_link_changed(mvm, vif,
+						     LINK_CONTEXT_MODIFY_ALL &
+						     ~LINK_CONTEXT_MODIFY_ACTIVE,
+						     true);
 		}
 
 		iwl_mvm_rs_rate_init(mvm, sta,
 				     mvmvif->deflink.phy_ctxt->channel->band,
 				     false);
-		ret = iwl_mvm_update_sta(mvm, vif, sta);
+		ret = callbacks->update_sta(mvm, vif, sta);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTHORIZED) {
 		ret = 0;
@@ -3567,7 +3599,7 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 
 			mvmvif->authorized = 1;
 
-			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
+			callbacks->mac_ctxt_changed(mvm, vif, false);
 			iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
 		}
 
@@ -3601,7 +3633,7 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 		   new_state == IEEE80211_STA_AUTH) {
 		if (vif->type == NL80211_IFTYPE_AP) {
 			mvmvif->ap_assoc_sta_count--;
-			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
+			callbacks->mac_ctxt_changed(mvm, vif, false);
 		} else if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
 			iwl_mvm_stop_session_protection(mvm, vif);
 		ret = 0;
@@ -3612,7 +3644,7 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 		   new_state == IEEE80211_STA_NOTEXIST) {
 		if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
 			iwl_mvm_stop_session_protection(mvm, vif);
-		ret = iwl_mvm_rm_sta(mvm, vif, sta);
+		ret = callbacks->rm_sta(mvm, vif, sta);
 		if (sta->tdls) {
 			iwl_mvm_recalc_tdls_state(mvm, vif, false);
 			iwl_mvm_tdls_check_trigger(mvm, vif, sta->addr,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index fe55efce8dd5..36c87d755d5c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -352,6 +352,23 @@ static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
+static int iwl_mvm_mld_mac_sta_state(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_sta *sta,
+				     enum ieee80211_sta_state old_state,
+				     enum ieee80211_sta_state new_state)
+{
+	struct iwl_mvm_sta_state_ops callbacks = {
+		.add_sta = iwl_mvm_mld_add_sta,
+		.update_sta = iwl_mvm_mld_update_sta,
+		.rm_sta = iwl_mvm_mld_rm_sta,
+		.mac_ctxt_changed = iwl_mvm_mld_mac_ctxt_changed,
+	};
+
+	return iwl_mvm_mac_sta_state_common(hw, vif, sta, old_state, new_state,
+					    &callbacks);
+}
+
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
@@ -359,4 +376,5 @@ const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
 	.join_ibss = iwl_mvm_mld_start_ap_ibss,
 	.leave_ibss = iwl_mvm_mld_stop_ap_ibss,
+	.sta_state = iwl_mvm_mld_mac_sta_state,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index c063383247f9..0e9c87bc224c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -3,6 +3,7 @@
  * Copyright (C) 2022 Intel Corporation
  */
 #include "mvm.h"
+#include "time-sync.h"
 
 static int iwl_mvm_mld_send_sta_cmd(struct iwl_mvm *mvm,
 				    struct iwl_mvm_sta_cfg_cmd *cmd)
@@ -331,6 +332,210 @@ int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 				      IWL_MAX_TID_COUNT, &mvm->snif_queue);
 }
 
+/* send a cfg sta command to add/update a sta in firmware */
+static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			       struct ieee80211_vif *vif, u16 phy_id)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_sta_cfg_cmd cmd = {
+		.sta_id = cpu_to_le32(mvm_sta->deflink.sta_id),
+		.link_id = cpu_to_le32(phy_id),
+		.station_type = cpu_to_le32(mvm_sta->sta_type),
+		.mfp = cpu_to_le32(sta->mfp),
+	};
+	u32 agg_size = 0, mpdu_dens = 0;
+
+	/* For now the link addr is the same as the mld addr */
+	if (vif->type == NL80211_IFTYPE_AP) {
+		memcpy(&cmd.peer_mld_address, sta->addr, ETH_ALEN);
+		memcpy(&cmd.peer_link_address, sta->addr, ETH_ALEN);
+	} else if (vif->bss_conf.bssid) {
+		memcpy(&cmd.peer_mld_address, vif->bss_conf.bssid, ETH_ALEN);
+		memcpy(&cmd.peer_link_address, vif->bss_conf.bssid, ETH_ALEN);
+	}
+
+	if (mvm_sta->sta_state >= IEEE80211_STA_ASSOC)
+		cmd.assoc_id = cpu_to_le32(sta->aid);
+
+	switch (sta->deflink.rx_nss) {
+	case 1:
+		cmd.mimo = cpu_to_le32(0);
+		break;
+	case 2 ... 8:
+		cmd.mimo = cpu_to_le32(1);
+		break;
+	}
+
+	switch (sta->deflink.smps_mode) {
+	case IEEE80211_SMPS_AUTOMATIC:
+	case IEEE80211_SMPS_NUM_MODES:
+		WARN_ON(1);
+		break;
+	case IEEE80211_SMPS_STATIC:
+		/* override NSS */
+		cmd.mimo = cpu_to_le32(0);
+		break;
+	case IEEE80211_SMPS_DYNAMIC:
+		cmd.mimo_protection = cpu_to_le32(1);
+		break;
+	case IEEE80211_SMPS_OFF:
+		/* nothing */
+		break;
+	}
+
+	mpdu_dens = iwl_mvm_get_sta_ampdu_dens(sta, &agg_size);
+	cmd.tx_ampdu_spacing = cpu_to_le32(mpdu_dens);
+	cmd.tx_ampdu_max_size = cpu_to_le32(agg_size);
+
+	if (sta->wme) {
+		cmd.sp_length =
+			cpu_to_le32(sta->max_sp ? sta->max_sp * 2 : 128);
+		cmd.uapsd_acs = cpu_to_le32(iwl_mvm_get_sta_uapsd_acs(sta));
+	}
+
+	if (sta->deflink.he_cap.has_he) {
+		cmd.trig_rnd_alloc =
+			cpu_to_le32(vif->bss_conf.uora_exists ? 1 : 0);
+
+		/* PPE Thresholds */
+		iwl_mvm_set_sta_pkt_ext(mvm, sta, &cmd.pkt_ext);
+
+		/* HTC flags */
+		cmd.htc_flags = iwl_mvm_get_sta_htc_flags(sta);
+
+		if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] &
+		    IEEE80211_HE_MAC_CAP2_ACK_EN)
+			cmd.ack_enabled = cpu_to_le32(1);
+	}
+
+	return iwl_mvm_mld_send_sta_cmd(mvm, &cmd);
+}
+
+int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	int sta_id, ret = 0;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+		sta_id = iwl_mvm_find_free_sta_id(mvm,
+						  ieee80211_vif_type_p2p(vif));
+	else
+		sta_id = mvm_sta->deflink.sta_id;
+
+	if (sta_id == IWL_MVM_INVALID_STA)
+		return -ENOSPC;
+
+	spin_lock_init(&mvm_sta->lock);
+
+	/* if this is a HW restart re-alloc existing queues */
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
+		struct iwl_mvm_int_sta tmp_sta = {
+			.sta_id = sta_id,
+			.type = mvm_sta->sta_type,
+		};
+
+		/* First add an empty station since allocating
+		 * a queue requires a valid station
+		 */
+		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, &tmp_sta,
+						    vif->bss_conf.bssid,
+						    mvmvif->deflink.phy_ctxt->id);
+		if (ret)
+			return ret;
+
+		iwl_mvm_realloc_queues_after_restart(mvm, sta);
+	} else {
+		ret = iwl_mvm_sta_init(mvm, vif, sta, sta_id,
+				       STATION_TYPE_PEER);
+	}
+
+	ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif, mvmvif->deflink.phy_ctxt->id);
+	if (ret)
+		return ret;
+
+	if (vif->type == NL80211_IFTYPE_STATION) {
+		if (!sta->tdls) {
+			WARN_ON(mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA);
+			mvmvif->deflink.ap_sta_id = sta_id;
+		} else {
+			WARN_ON(mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA);
+		}
+	}
+
+	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
+
+	return 0;
+}
+
+int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			   struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	lockdep_assert_held(&mvm->mutex);
+
+	return iwl_mvm_mld_cfg_sta(mvm, sta, vif, mvmvif->deflink.phy_ctxt->id);
+}
+
+static void iwl_mvm_mld_disable_sta_queues(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	int i;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	for (i = 0; i < ARRAY_SIZE(mvm_sta->tid_data); i++) {
+		if (mvm_sta->tid_data[i].txq_id == IWL_MVM_INVALID_QUEUE)
+			continue;
+
+		iwl_mvm_mld_disable_txq(mvm, sta, &mvm_sta->tid_data[i].txq_id,
+					i);
+		mvm_sta->tid_data[i].txq_id = IWL_MVM_INVALID_QUEUE;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(sta->txq); i++) {
+		struct iwl_mvm_txq *mvmtxq =
+			iwl_mvm_txq_from_mac80211(sta->txq[i]);
+
+		mvmtxq->txq_id = IWL_MVM_INVALID_QUEUE;
+	}
+}
+
+int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		       struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	kfree(mvm_sta->dup_data);
+
+	/* flush its queues here since we are freeing mvm_sta */
+	ret = iwl_mvm_flush_sta(mvm, mvm_sta, false);
+	if (ret)
+		return ret;
+	ret = iwl_mvm_wait_sta_queues_empty(mvm, mvm_sta);
+	if (ret)
+		return ret;
+
+	iwl_mvm_mld_disable_sta_queues(mvm, vif, sta);
+
+	if (iwl_mvm_sta_del(mvm, vif, sta, &ret))
+		return ret;
+
+	ret = iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_sta->deflink.sta_id);
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id], NULL);
+
+	return ret;
+}
+
 static void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
 					      struct ieee80211_sta *sta,
 					      bool disable)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 443f0b80b63e..24884e102673 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1573,6 +1573,13 @@ int iwl_mvm_flush_tx_path(struct iwl_mvm *mvm, u32 tfd_msk);
 int iwl_mvm_flush_sta(struct iwl_mvm *mvm, void *sta, bool internal);
 int iwl_mvm_flush_sta_tids(struct iwl_mvm *mvm, u32 sta_id, u16 tids);
 
+/* Utils to extract sta related data */
+__le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta);
+u8 iwl_mvm_get_sta_uapsd_acs(struct ieee80211_sta *sta);
+u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta, u32 *_agg_size);
+int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			    struct iwl_he_pkt_ext_v2 *pkt_ext);
+
 void iwl_mvm_async_handlers_purge(struct iwl_mvm *mvm);
 
 static inline void iwl_mvm_set_tx_cmd_ccmp(struct ieee80211_tx_info *info,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index d01c67e1ce2a..0ff66ffc5bee 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -24,8 +24,7 @@ static inline int iwl_mvm_add_sta_cmd_size(struct iwl_mvm *mvm)
 		return sizeof(struct iwl_mvm_add_sta_cmd_v7);
 }
 
-static int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm,
-				    enum nl80211_iftype iftype)
+int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm, enum nl80211_iftype iftype)
 {
 	int sta_id;
 	u32 reserved_ids = 0;
@@ -52,8 +51,7 @@ static int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm,
 }
 
 /* Calculate the ampdu density and max size */
-static u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta,
-				      u32 *_agg_size)
+u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta, u32 *_agg_size)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	u32 agg_size = 0, mpdu_dens = 0;
@@ -97,7 +95,7 @@ static u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta,
 	return mpdu_dens;
 }
 
-static u8 iwl_mvm_get_sta_uapsd_acs(struct ieee80211_sta *sta)
+u8 iwl_mvm_get_sta_uapsd_acs(struct ieee80211_sta *sta)
 {
 	u8 uapsd_acs = 0;
 
@@ -1528,8 +1526,8 @@ static int iwl_mvm_reserve_sta_stream(struct iwl_mvm *mvm,
  *
  * Note that re-enabling aggregations isn't done in this function.
  */
-static void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
-						 struct ieee80211_sta *sta)
+void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
+					  struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	unsigned int wdg =
@@ -1649,8 +1647,8 @@ static int iwl_mvm_add_int_sta_common(struct iwl_mvm *mvm,
 }
 
 /* Initialize driver data of a new sta */
-static int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			    struct ieee80211_sta *sta, int sta_id, u8 sta_type)
+int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		     struct ieee80211_sta *sta, int sta_id, u8 sta_type)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
@@ -1923,9 +1921,12 @@ int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
 	return 0;
 }
 
-/* Execute the common part for both MLD and non-MLD modes */
-static bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			    struct ieee80211_sta *sta, int *ret)
+/* Execute the common part for both MLD and non-MLD modes.
+ * Returns if we're done with removing the station, either
+ * with error or success
+ */
+bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		     struct ieee80211_sta *sta, int *ret)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index bfd44c489286..fa644928746a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -471,6 +471,9 @@ struct iwl_mvm_int_sta {
  */
 int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			   bool update, unsigned int flags);
+int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm, enum nl80211_iftype iftype);
+int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		     struct ieee80211_sta *sta, int sta_id, u8 sta_type);
 int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 		    struct ieee80211_vif *vif,
 		    struct ieee80211_sta *sta);
@@ -482,8 +485,12 @@ static inline int iwl_mvm_update_sta(struct iwl_mvm *mvm,
 	return iwl_mvm_sta_send_to_fw(mvm, sta, true, 0);
 }
 
+void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
+					  struct ieee80211_sta *sta);
 int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
 				  struct iwl_mvm_sta *mvm_sta);
+bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		     struct ieee80211_sta *sta, int *ret);
 int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 		   struct ieee80211_vif *vif,
 		   struct ieee80211_sta *sta);
@@ -578,6 +585,38 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 			    struct ieee80211_sta *sta,
 			    u8 sta_id, u8 tid, unsigned int timeout);
 
+/* Sta state */
+/**
+ * struct iwl_mvm_sta_state_ops - callbacks for the sta_state() ops
+ *
+ * Since the only difference between both MLD and
+ * non-MLD versions of sta_state() is these function calls,
+ * each version will send its specific function calls to
+ * %iwl_mvm_mac_sta_state_common().
+ *
+ * @add_sta: pointer to the function that adds a new sta
+ * @update_sta: pointer to the function that updates a sta
+ * @rm_sta: pointer to the functions that removes a sta
+ * @mac_ctxt_changed: pointer to the function that handles a change in mac ctxt
+ */
+struct iwl_mvm_sta_state_ops {
+	int (*add_sta)(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		       struct ieee80211_sta *sta);
+	int (*update_sta)(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			  struct ieee80211_sta *sta);
+	int (*rm_sta)(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		      struct ieee80211_sta *sta);
+	int (*mac_ctxt_changed)(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				bool force_assoc_off);
+};
+
+int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_sta *sta,
+				 enum ieee80211_sta_state old_state,
+				 enum ieee80211_sta_state new_state,
+				 struct iwl_mvm_sta_state_ops *callbacks);
+
 /* New MLD STA related APIs */
 /* STA */
 int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
@@ -586,6 +625,12 @@ int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			struct ieee80211_sta *sta);
+int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			   struct ieee80211_sta *sta);
+int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		       struct ieee80211_sta *sta);
 
 /* Queues */
 void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From 660eba5af0ddf80f04efd970a63958d23c4d1052 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:45 +0300
Subject: wifi: iwlwifi: mvm: add some new MLD ops

Add MLD version of bss_info_changed/switch_vif_chanctx/
config_iface_filter and conf_tx() callbacks.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.9c83c253d610.Ibf2006be9ece87896c17cb43dfe3654ac73d81ff@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 213 +++++++++++--------
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 227 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   |  11 +
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  73 +++++++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |   2 +
 5 files changed, 446 insertions(+), 80 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index b730b605da74..dae5e1b7d1cc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2433,9 +2433,8 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 		IWL_ERR(mvm, "Failed to config FW to work HE!\n");
 }
 
-static void iwl_mvm_protect_assoc(struct iwl_mvm *mvm,
-				  struct ieee80211_vif *vif,
-				  u32 duration_override)
+void iwl_mvm_protect_assoc(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			   u32 duration_override)
 {
 	u32 duration = IWL_MVM_TE_SESSION_PROTECTION_MAX_TIME_MS;
 	u32 min_duration = IWL_MVM_TE_SESSION_PROTECTION_MIN_TIME_MS;
@@ -2461,6 +2460,82 @@ static void iwl_mvm_protect_assoc(struct iwl_mvm *mvm,
 					min_duration, 500, false);
 }
 
+/* Handle association common part to MLD and non-MLD modes */
+void iwl_mvm_bss_info_changed_station_assoc(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
+					    u64 changes)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	/* The firmware tracks the MU-MIMO group on its own.
+	 * However, on HW restart we should restore this data.
+	 */
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
+	    (changes & BSS_CHANGED_MU_GROUPS) && vif->bss_conf.mu_mimo_owner) {
+		ret = iwl_mvm_update_mu_groups(mvm, vif);
+		if (ret)
+			IWL_ERR(mvm,
+				"failed to update VHT MU_MIMO groups\n");
+	}
+
+	iwl_mvm_recalc_multicast(mvm);
+
+	/* reset rssi values */
+	mvmvif->bf_data.ave_beacon_signal = 0;
+
+	iwl_mvm_bt_coex_vif_change(mvm);
+	iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_TT,
+			    IEEE80211_SMPS_AUTOMATIC);
+	if (fw_has_capa(&mvm->fw->ucode_capa,
+			IWL_UCODE_TLV_CAPA_UMAC_SCAN))
+		iwl_mvm_config_scan(mvm);
+}
+
+/* Execute the common part for MLD and non-MLD modes */
+void iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
+					     struct ieee80211_vif *vif,
+					     u64 changes)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	if (changes & BSS_CHANGED_BEACON_INFO) {
+		/* We received a beacon from the associated AP so
+		 * remove the session protection.
+		 */
+		iwl_mvm_stop_session_protection(mvm, vif);
+
+		iwl_mvm_sf_update(mvm, vif, false);
+		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
+	}
+
+	if (changes & (BSS_CHANGED_PS | BSS_CHANGED_P2P_PS | BSS_CHANGED_QOS |
+		       /* Send power command on every beacon change,
+			* because we may have not enabled beacon abort yet.
+			*/
+		       BSS_CHANGED_BEACON_INFO)) {
+		ret = iwl_mvm_power_update_mac(mvm);
+		if (ret)
+			IWL_ERR(mvm, "failed to update power mode\n");
+	}
+
+	if (changes & BSS_CHANGED_CQM) {
+		IWL_DEBUG_MAC80211(mvm, "cqm info_changed\n");
+		/* reset cqm events tracking */
+		mvmvif->bf_data.last_cqm_event = 0;
+		if (mvmvif->bf_data.bf_enabled) {
+			ret = iwl_mvm_enable_beacon_filter(mvm, vif, 0);
+			if (ret)
+				IWL_ERR(mvm,
+					"failed to update CQM thresholds\n");
+		}
+	}
+
+	if (changes & BSS_CHANGED_BANDWIDTH)
+		iwl_mvm_apply_fw_smps_request(vif);
+}
+
 static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 					     struct ieee80211_vif *vif,
 					     struct ieee80211_bss_conf *bss_conf,
@@ -2623,67 +2698,10 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 					vif->addr);
 		}
 
-		/*
-		 * The firmware tracks the MU-MIMO group on its own.
-		 * However, on HW restart we should restore this data.
-		 */
-		if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
-		    (changes & BSS_CHANGED_MU_GROUPS) && vif->bss_conf.mu_mimo_owner) {
-			ret = iwl_mvm_update_mu_groups(mvm, vif);
-			if (ret)
-				IWL_ERR(mvm,
-					"failed to update VHT MU_MIMO groups\n");
-		}
-
-		iwl_mvm_recalc_multicast(mvm);
-
-		/* reset rssi values */
-		mvmvif->bf_data.ave_beacon_signal = 0;
-
-		iwl_mvm_bt_coex_vif_change(mvm);
-		iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_TT,
-				    IEEE80211_SMPS_AUTOMATIC);
-		if (fw_has_capa(&mvm->fw->ucode_capa,
-				IWL_UCODE_TLV_CAPA_UMAC_SCAN))
-			iwl_mvm_config_scan(mvm);
-	}
-
-	if (changes & BSS_CHANGED_BEACON_INFO) {
-		/*
-		 * We received a beacon from the associated AP so
-		 * remove the session protection.
-		 */
-		iwl_mvm_stop_session_protection(mvm, vif);
-
-		iwl_mvm_sf_update(mvm, vif, false);
-		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
+		iwl_mvm_bss_info_changed_station_assoc(mvm, vif, changes);
 	}
 
-	if (changes & (BSS_CHANGED_PS | BSS_CHANGED_P2P_PS | BSS_CHANGED_QOS |
-		       /*
-			* Send power command on every beacon change,
-			* because we may have not enabled beacon abort yet.
-			*/
-		       BSS_CHANGED_BEACON_INFO)) {
-		ret = iwl_mvm_power_update_mac(mvm);
-		if (ret)
-			IWL_ERR(mvm, "failed to update power mode\n");
-	}
-
-	if (changes & BSS_CHANGED_CQM) {
-		IWL_DEBUG_MAC80211(mvm, "cqm info_changed\n");
-		/* reset cqm events tracking */
-		mvmvif->bf_data.last_cqm_event = 0;
-		if (mvmvif->bf_data.bf_enabled) {
-			ret = iwl_mvm_enable_beacon_filter(mvm, vif, 0);
-			if (ret)
-				IWL_ERR(mvm,
-					"failed to update CQM thresholds\n");
-		}
-	}
-
-	if (changes & BSS_CHANGED_BANDWIDTH)
-		iwl_mvm_apply_fw_smps_request(vif);
+	iwl_mvm_bss_info_changed_station_common(mvm, vif, changes);
 }
 
 bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
@@ -2967,6 +2985,22 @@ static void iwl_mvm_bss_info_changed(struct ieee80211_hw *hw,
 				     struct ieee80211_vif *vif,
 				     struct ieee80211_bss_conf *bss_conf,
 				     u64 changes)
+{
+	struct iwl_mvm_bss_info_changed_ops callbacks = {
+		.bss_info_changed_sta = iwl_mvm_bss_info_changed_station,
+		.bss_info_changed_ap_ibss = iwl_mvm_bss_info_changed_ap_ibss,
+	};
+
+	iwl_mvm_bss_info_changed_common(hw, vif, bss_conf, &callbacks,
+					changes);
+}
+
+void
+iwl_mvm_bss_info_changed_common(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *bss_conf,
+				struct iwl_mvm_bss_info_changed_ops *callbacks,
+				u64 changes)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -2977,11 +3011,12 @@ static void iwl_mvm_bss_info_changed(struct ieee80211_hw *hw,
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
-		iwl_mvm_bss_info_changed_station(mvm, vif, bss_conf, changes);
+		callbacks->bss_info_changed_sta(mvm, vif, bss_conf, changes);
 		break;
 	case NL80211_IFTYPE_AP:
 	case NL80211_IFTYPE_ADHOC:
-		iwl_mvm_bss_info_changed_ap_ibss(mvm, vif, bss_conf, changes);
+		callbacks->bss_info_changed_ap_ibss(mvm, vif, bss_conf,
+						    changes);
 		break;
 	case NL80211_IFTYPE_MONITOR:
 		if (changes & BSS_CHANGED_MU_GROUPS)
@@ -4762,12 +4797,13 @@ static void iwl_mvm_unassign_vif_chanctx(struct ieee80211_hw *hw,
 
 static int
 iwl_mvm_switch_vif_chanctx_swap(struct iwl_mvm *mvm,
-				struct ieee80211_vif_chanctx_switch *vifs)
+				struct ieee80211_vif_chanctx_switch *vifs,
+				struct iwl_mvm_switch_vif_chanctx_ops *ops)
 {
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	__iwl_mvm_unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx, true);
+	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx, true);
 	__iwl_mvm_remove_chanctx(mvm, vifs[0].old_ctx);
 
 	ret = __iwl_mvm_add_chanctx(mvm, vifs[0].new_ctx);
@@ -4776,7 +4812,7 @@ iwl_mvm_switch_vif_chanctx_swap(struct iwl_mvm *mvm,
 		goto out_reassign;
 	}
 
-	ret = __iwl_mvm_assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].new_ctx,
+	ret = ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].new_ctx,
 					   true);
 	if (ret) {
 		IWL_ERR(mvm,
@@ -4799,8 +4835,8 @@ out_reassign:
 		goto out_restart;
 	}
 
-	if (__iwl_mvm_assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx,
-					 true)) {
+	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx,
+				      true)) {
 		IWL_ERR(mvm, "failed to reassign old_ctx after failure.\n");
 		goto out_restart;
 	}
@@ -4819,14 +4855,15 @@ out:
 
 static int
 iwl_mvm_switch_vif_chanctx_reassign(struct iwl_mvm *mvm,
-				    struct ieee80211_vif_chanctx_switch *vifs)
+				    struct ieee80211_vif_chanctx_switch *vifs,
+				    struct iwl_mvm_switch_vif_chanctx_ops *ops)
 {
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	__iwl_mvm_unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx, true);
+	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx, true);
 
-	ret = __iwl_mvm_assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].new_ctx,
+	ret = ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].new_ctx,
 					   true);
 	if (ret) {
 		IWL_ERR(mvm,
@@ -4837,8 +4874,8 @@ iwl_mvm_switch_vif_chanctx_reassign(struct iwl_mvm *mvm,
 	goto out;
 
 out_reassign:
-	if (__iwl_mvm_assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx,
-					 true)) {
+	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx,
+				      true)) {
 		IWL_ERR(mvm, "failed to reassign old_ctx after failure.\n");
 		goto out_restart;
 	}
@@ -4855,10 +4892,13 @@ out:
 	return ret;
 }
 
-static int iwl_mvm_switch_vif_chanctx(struct ieee80211_hw *hw,
-				      struct ieee80211_vif_chanctx_switch *vifs,
-				      int n_vifs,
-				      enum ieee80211_chanctx_switch_mode mode)
+/* Execute the common part for both MLD and non-MLD modes */
+int
+iwl_mvm_switch_vif_chanctx_common(struct ieee80211_hw *hw,
+				  struct ieee80211_vif_chanctx_switch *vifs,
+				  int n_vifs,
+				  enum ieee80211_chanctx_switch_mode mode,
+				  struct iwl_mvm_switch_vif_chanctx_ops *ops)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -4869,10 +4909,10 @@ static int iwl_mvm_switch_vif_chanctx(struct ieee80211_hw *hw,
 
 	switch (mode) {
 	case CHANCTX_SWMODE_SWAP_CONTEXTS:
-		ret = iwl_mvm_switch_vif_chanctx_swap(mvm, vifs);
+		ret = iwl_mvm_switch_vif_chanctx_swap(mvm, vifs, ops);
 		break;
 	case CHANCTX_SWMODE_REASSIGN_VIF:
-		ret = iwl_mvm_switch_vif_chanctx_reassign(mvm, vifs);
+		ret = iwl_mvm_switch_vif_chanctx_reassign(mvm, vifs, ops);
 		break;
 	default:
 		ret = -EOPNOTSUPP;
@@ -4882,6 +4922,19 @@ static int iwl_mvm_switch_vif_chanctx(struct ieee80211_hw *hw,
 	return ret;
 }
 
+static int iwl_mvm_switch_vif_chanctx(struct ieee80211_hw *hw,
+				      struct ieee80211_vif_chanctx_switch *vifs,
+				      int n_vifs,
+				      enum ieee80211_chanctx_switch_mode mode)
+{
+	struct iwl_mvm_switch_vif_chanctx_ops ops = {
+		.__assign_vif_chanctx = __iwl_mvm_assign_vif_chanctx,
+		.__unassign_vif_chanctx = __iwl_mvm_unassign_vif_chanctx,
+	};
+
+	return iwl_mvm_switch_vif_chanctx_common(hw, vifs, n_vifs, mode, &ops);
+}
+
 static int iwl_mvm_tx_last_beacon(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 36c87d755d5c..9d97098ab57c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -369,12 +369,239 @@ static int iwl_mvm_mld_mac_sta_state(struct ieee80211_hw *hw,
 					    &callbacks);
 }
 
+static void
+iwl_mvm_mld_bss_info_changed_station(struct iwl_mvm *mvm,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *bss_conf,
+				     u64 changes)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+	u32 link_changes = 0;
+	bool has_he = vif->bss_conf.he_support &&
+			  !iwlwifi_mod_params.disable_11ax;
+	bool has_eht = vif->bss_conf.eht_support &&
+			  !iwlwifi_mod_params.disable_11be;
+
+	if (changes & BSS_CHANGED_ASSOC && vif->cfg.assoc &&
+	    (has_he || has_eht)) {
+		IWL_DEBUG_MAC80211(mvm, "Associated in HE mode\n");
+		link_changes |= LINK_CONTEXT_MODIFY_HE_PARAMS;
+	}
+
+	/* Update MU EDCA params */
+	if (changes & BSS_CHANGED_QOS && vif->cfg.assoc &&
+	    (has_he || has_eht))
+		link_changes |= LINK_CONTEXT_MODIFY_QOS_PARAMS;
+
+	/* Update EHT Puncturing info */
+	if (changes & BSS_CHANGED_EHT_PUNCTURING && vif->cfg.assoc && has_eht)
+		link_changes |= LINK_CONTEXT_MODIFY_EHT_PARAMS;
+
+	if (link_changes) {
+		ret = iwl_mvm_link_changed(mvm, vif, link_changes, true);
+		if (ret)
+			IWL_ERR(mvm, "failed to update link\n");
+	}
+
+	ret = iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
+	if (ret)
+		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
+
+	memcpy(mvmvif->deflink.bssid, bss_conf->bssid, ETH_ALEN);
+	mvmvif->associated = vif->cfg.assoc;
+
+	if (changes & BSS_CHANGED_ASSOC) {
+		if (vif->cfg.assoc) {
+			/* clear statistics to get clean beacon counter */
+			iwl_mvm_request_statistics(mvm, true);
+			memset(&mvmvif->deflink.beacon_stats, 0,
+			       sizeof(mvmvif->deflink.beacon_stats));
+
+			if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART,
+				      &mvm->status) &&
+			    !vif->bss_conf.dtim_period) {
+				/* If we're not restarting and still haven't
+				 * heard a beacon (dtim period unknown) then
+				 * make sure we still have enough minimum time
+				 * remaining in the time event, since the auth
+				 * might actually have taken quite a while
+				 * (especially for SAE) and so the remaining
+				 * time could be small without us having heard
+				 * a beacon yet.
+				 */
+				iwl_mvm_protect_assoc(mvm, vif, 0);
+			}
+
+			iwl_mvm_sf_update(mvm, vif, false);
+			iwl_mvm_power_vif_assoc(mvm, vif);
+			if (vif->p2p) {
+				iwl_mvm_update_smps(mvm, vif,
+						    IWL_MVM_SMPS_REQ_PROT,
+						    IEEE80211_SMPS_DYNAMIC);
+			}
+		} else if (mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
+			iwl_mvm_mei_host_disassociated(mvm);
+			/* If update fails - SF might be running in associated
+			 * mode while disassociated - which is forbidden.
+			 */
+			ret = iwl_mvm_sf_update(mvm, vif, false);
+			WARN_ONCE(ret &&
+				  !test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED,
+					    &mvm->status),
+				  "Failed to update SF upon disassociation\n");
+
+			/* If we get an assert during the connection (after the
+			 * station has been added, but before the vif is set
+			 * to associated), mac80211 will re-add the station and
+			 * then configure the vif. Since the vif is not
+			 * associated, we would remove the station here and
+			 * this would fail the recovery.
+			 */
+			if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART,
+				      &mvm->status)) {
+				/* first remove remaining keys */
+				iwl_mvm_sec_key_remove_ap(mvm, vif);
+
+				/* Remove AP station now that
+				 * the MAC is unassoc
+				 */
+				ret = iwl_mvm_mld_rm_sta_id(mvm, vif,
+							    mvmvif->deflink.ap_sta_id);
+				if (ret)
+					IWL_ERR(mvm,
+						"failed to remove AP station\n");
+
+				mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
+			}
+		}
+
+		iwl_mvm_bss_info_changed_station_assoc(mvm, vif, changes);
+	}
+
+	iwl_mvm_bss_info_changed_station_common(mvm, vif, changes);
+}
+
+static void
+iwl_mvm_mld_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *bss_conf,
+				     u64 changes)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	u32 link_changes = LINK_CONTEXT_MODIFY_PROTECT_FLAGS |
+			   LINK_CONTEXT_MODIFY_QOS_PARAMS;
+
+	/* Changes will be applied when the AP/IBSS is started */
+	if (!mvmvif->ap_ibss_active)
+		return;
+
+	if (changes & (BSS_CHANGED_ERP_CTS_PROT | BSS_CHANGED_HT |
+		       BSS_CHANGED_BANDWIDTH | BSS_CHANGED_QOS) &&
+		       iwl_mvm_link_changed(mvm, vif, link_changes, true))
+		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
+
+	/* Need to send a new beacon template to the FW */
+	if (changes & BSS_CHANGED_BEACON &&
+	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif))
+		IWL_WARN(mvm, "Failed updating beacon data\n");
+
+	if (changes & BSS_CHANGED_FTM_RESPONDER) {
+		int ret = iwl_mvm_ftm_start_responder(mvm, vif);
+
+		if (ret)
+			IWL_WARN(mvm, "Failed to enable FTM responder (%d)\n",
+				 ret);
+	}
+}
+
+static void iwl_mvm_mld_bss_info_changed(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif,
+					 struct ieee80211_bss_conf *bss_conf,
+					 u64 changes)
+{
+	struct iwl_mvm_bss_info_changed_ops callbacks = {
+		.bss_info_changed_sta = iwl_mvm_mld_bss_info_changed_station,
+		.bss_info_changed_ap_ibss =
+			iwl_mvm_mld_bss_info_changed_ap_ibss,
+	};
+
+	iwl_mvm_bss_info_changed_common(hw, vif, bss_conf, &callbacks,
+					changes);
+}
+
+static int
+iwl_mvm_mld_switch_vif_chanctx(struct ieee80211_hw *hw,
+			       struct ieee80211_vif_chanctx_switch *vifs,
+			       int n_vifs,
+			       enum ieee80211_chanctx_switch_mode mode)
+{
+	struct iwl_mvm_switch_vif_chanctx_ops ops = {
+		.__assign_vif_chanctx = __iwl_mvm_mld_assign_vif_chanctx,
+		.__unassign_vif_chanctx = __iwl_mvm_mld_unassign_vif_chanctx,
+	};
+
+	return iwl_mvm_switch_vif_chanctx_common(hw, vifs, n_vifs, mode, &ops);
+}
+
+static void iwl_mvm_mld_config_iface_filter(struct ieee80211_hw *hw,
+					    struct ieee80211_vif *vif,
+					    unsigned int filter_flags,
+					    unsigned int changed_flags)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	/* We support only filter for probe requests */
+	if (!(changed_flags & FIF_PROBE_REQ))
+		return;
+
+	/* Supported only for p2p client interfaces */
+	if (vif->type != NL80211_IFTYPE_STATION || !vif->cfg.assoc ||
+	    !vif->p2p)
+		return;
+
+	mutex_lock(&mvm->mutex);
+	iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
+	mutex_unlock(&mvm->mutex);
+}
+
+static int
+iwl_mvm_mld_mac_conf_tx(struct ieee80211_hw *hw,
+			struct ieee80211_vif *vif,
+			unsigned int link_id, u16 ac,
+			const struct ieee80211_tx_queue_params *params)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	mvmvif->deflink.queue_params[ac] = *params;
+
+	/* No need to update right away, we'll get BSS_CHANGED_QOS
+	 * The exception is P2P_DEVICE interface which needs immediate update.
+	 */
+	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
+		int ret;
+
+		mutex_lock(&mvm->mutex);
+		ret = iwl_mvm_link_changed(mvm, vif,
+					   LINK_CONTEXT_MODIFY_QOS_PARAMS,
+					   true);
+		mutex_unlock(&mvm->mutex);
+		return ret;
+	}
+	return 0;
+}
+
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
+	.config_iface_filter = iwl_mvm_mld_config_iface_filter,
 	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
 	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
+	.switch_vif_chanctx = iwl_mvm_mld_switch_vif_chanctx,
 	.join_ibss = iwl_mvm_mld_start_ap_ibss,
 	.leave_ibss = iwl_mvm_mld_stop_ap_ibss,
 	.sta_state = iwl_mvm_mld_mac_sta_state,
+	.conf_tx = iwl_mvm_mld_mac_conf_tx,
+	.bss_info_changed = iwl_mvm_mld_bss_info_changed,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 0e9c87bc224c..16723d581a3d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -536,6 +536,17 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	return ret;
 }
 
+int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			  u8 sta_id)
+{
+	int ret = iwl_mvm_mld_rm_sta_from_fw(mvm, sta_id);
+
+	lockdep_assert_held(&mvm->mutex);
+
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta_id], NULL);
+	return ret;
+}
+
 static void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
 					      struct ieee80211_sta *sta,
 					      bool disable)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 24884e102673..2ea224acdea1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1789,6 +1789,48 @@ bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
 void iwl_mvm_stop_ap_ibss_common(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif);
 
+/* BSS Info */
+/**
+ * struct iwl_mvm_bss_info_changed_ops - callbacks for the bss_info_changed()
+ *
+ * Since the only difference between both MLD and
+ * non-MLD versions of bss_info_changed() is these function calls,
+ * each version will send its specific function calls to
+ * %iwl_mvm_bss_info_changed_common().
+ *
+ * @bss_info_changed_sta: pointer to the function that handles changes
+ *	in bss_info in sta mode
+ * @bss_info_changed_ap_ibss: pointer to the function that handles changes
+ *	in bss_info in ap and ibss modes
+ */
+struct iwl_mvm_bss_info_changed_ops {
+	void (*bss_info_changed_sta)(struct iwl_mvm *mvm,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *bss_conf,
+				     u64 changes);
+	void (*bss_info_changed_ap_ibss)(struct iwl_mvm *mvm,
+					 struct ieee80211_vif *vif,
+					 struct ieee80211_bss_conf *bss_conf,
+					 u64 changes);
+};
+
+void
+iwl_mvm_bss_info_changed_common(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *bss_conf,
+				struct iwl_mvm_bss_info_changed_ops *callbacks,
+				u64 changes);
+void iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
+					     struct ieee80211_vif *vif,
+					     u64 changes);
+void iwl_mvm_bss_info_changed_station_assoc(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
+					    u64 changes);
+
+/*Session Protection */
+void iwl_mvm_protect_assoc(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			   u32 duration_override);
+
 /* Quota management */
 static inline size_t iwl_mvm_quota_cmd_size(struct iwl_mvm *mvm)
 {
@@ -2232,6 +2274,37 @@ static inline u8 iwl_mvm_phy_band_from_nl80211(enum nl80211_band band)
 void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk);
 
 /* Channel Context */
+/**
+ * struct iwl_mvm_switch_vif_chanctx_ops - callbacks for switch_vif_chanctx()
+ *
+ * Since the only difference between both MLD and
+ * non-MLD versions of switch_vif_chanctx() is these function calls,
+ * each version will send its specific function calls to
+ * %iwl_mvm_switch_vif_chanctx_common().
+ *
+ * @__assign_vif_chanctx: pointer to the function that assigns a chanctx to
+ *	a given vif
+ * @__unassign_vif_chanctx: pointer to the function that unassigns a chanctx to
+ *	a given vif
+ */
+struct iwl_mvm_switch_vif_chanctx_ops {
+	int (*__assign_vif_chanctx)(struct iwl_mvm *mvm,
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_chanctx_conf *ctx,
+				    bool switching_chanctx);
+	void (*__unassign_vif_chanctx)(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_chanctx_conf *ctx,
+				       bool switching_chanctx);
+};
+
+int
+iwl_mvm_switch_vif_chanctx_common(struct ieee80211_hw *hw,
+				  struct ieee80211_vif_chanctx_switch *vifs,
+				  int n_vifs,
+				  enum ieee80211_chanctx_switch_mode mode,
+				  struct iwl_mvm_switch_vif_chanctx_ops *ops);
+
 bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
 					 struct ieee80211_vif *vif,
 					 struct ieee80211_chanctx_conf *ctx,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index fa644928746a..770516b20303 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -631,6 +631,8 @@ int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta);
 int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta);
+int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			  u8 sta_id);
 
 /* Queues */
 void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From feebebae1fc3a9c02395ce48074e763cbcdecc8e Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:46 +0300
Subject: wifi: iwlwifi: mvm: refactor iwl_mvm_roc()

This flow is almost the same for both MLD and non-MLD modes,
except for some function calls. Therefore there is no reason to
add an MLD version of this flow. Instead - put the parts that are unique
for each mode in helper functions, and in the next patch each version of
this flow will call the common part with pointers to its specific
helper functions.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.61bc077a7f3c.Ia3aa81d3293792bf8f80528dbc67a711ce334b32@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 100 ++++++++++++----------
 1 file changed, 56 insertions(+), 44 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index dae5e1b7d1cc..bb7e4e26d99f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4258,6 +4258,52 @@ static int iwl_mvm_send_aux_roc_cmd(struct iwl_mvm *mvm,
 	return res;
 }
 
+static int iwl_mvm_add_aux_sta_for_hs20(struct iwl_mvm *mvm, u32 lmac_id)
+{
+	int ret = 0;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (!fw_has_capa(&mvm->fw->ucode_capa,
+			 IWL_UCODE_TLV_CAPA_HOTSPOT_SUPPORT)) {
+		IWL_ERR(mvm, "hotspot not supported\n");
+		return -EINVAL;
+	}
+
+	if (iwl_fw_lookup_cmd_ver(mvm->fw, ADD_STA, 0) >= 12) {
+		ret = iwl_mvm_add_aux_sta(mvm, lmac_id);
+		WARN(ret, "Failed to allocate aux station");
+	}
+
+	return ret;
+}
+
+static int iwl_mvm_roc_switch_binding(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct iwl_mvm_phy_ctxt *new_phy_ctxt)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret = 0;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	/* Unbind the P2P_DEVICE from the current PHY context,
+	 * and if the PHY context is not used remove it.
+	 */
+	ret = iwl_mvm_binding_remove_vif(mvm, vif);
+	if (WARN(ret, "Failed unbinding P2P_DEVICE\n"))
+		return ret;
+
+	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
+
+	/* Bind the P2P_DEVICE to the current PHY Context */
+	mvmvif->deflink.phy_ctxt = new_phy_ctxt;
+
+	ret = iwl_mvm_binding_add_vif(mvm, vif);
+	WARN(ret, "Failed binding P2P_DEVICE\n");
+	return ret;
+}
+
 static int iwl_mvm_roc(struct ieee80211_hw *hw,
 		       struct ieee80211_vif *vif,
 		       struct ieee80211_channel *channel,
@@ -4270,6 +4316,7 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 	struct iwl_mvm_phy_ctxt *phy_ctxt;
 	bool band_change_removal;
 	int ret, i;
+	u32 lmac_id;
 
 	IWL_DEBUG_MAC80211(mvm, "enter (%d, %d, %d)\n", channel->hw_value,
 			   duration, type);
@@ -4284,25 +4331,13 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
-		if (fw_has_capa(&mvm->fw->ucode_capa,
-				IWL_UCODE_TLV_CAPA_HOTSPOT_SUPPORT)) {
-			/* Use aux roc framework (HS20) */
-			if (iwl_fw_lookup_cmd_ver(mvm->fw, ADD_STA, 0) >= 12) {
-				u32 lmac_id;
-
-				lmac_id = iwl_mvm_get_lmac_id(mvm->fw,
-							      channel->band);
-				ret = iwl_mvm_add_aux_sta(mvm, lmac_id);
-				if (WARN(ret,
-					 "Failed to allocate aux station"))
-					goto out_unlock;
-			}
+		lmac_id = iwl_mvm_get_lmac_id(mvm->fw, channel->band);
+
+		/* Use aux roc framework (HS20) */
+		ret = iwl_mvm_add_aux_sta_for_hs20(mvm, lmac_id);
+		if (!ret)
 			ret = iwl_mvm_send_aux_roc_cmd(mvm, channel,
 						       vif, duration);
-			goto out_unlock;
-		}
-		IWL_ERR(mvm, "hotspot not supported\n");
-		ret = -EINVAL;
 		goto out_unlock;
 	case NL80211_IFTYPE_P2P_DEVICE:
 		/* handle below */
@@ -4319,21 +4354,8 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 			continue;
 
 		if (phy_ctxt->ref && channel == phy_ctxt->channel) {
-			/*
-			 * Unbind the P2P_DEVICE from the current PHY context,
-			 * and if the PHY context is not used remove it.
-			 */
-			ret = iwl_mvm_binding_remove_vif(mvm, vif);
-			if (WARN(ret, "Failed unbinding P2P_DEVICE\n"))
-				goto out_unlock;
-
-			iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
-
-			/* Bind the P2P_DEVICE to the current PHY Context */
-			mvmvif->deflink.phy_ctxt = phy_ctxt;
-
-			ret = iwl_mvm_binding_add_vif(mvm, vif);
-			if (WARN(ret, "Failed binding P2P_DEVICE\n"))
+			ret = iwl_mvm_roc_switch_binding(mvm, vif, phy_ctxt);
+			if (ret)
 				goto out_unlock;
 
 			iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
@@ -4386,18 +4408,8 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 			goto out_unlock;
 		}
 
-		/* Unbind the P2P_DEVICE from the current PHY context */
-		ret = iwl_mvm_binding_remove_vif(mvm, vif);
-		if (WARN(ret, "Failed unbinding P2P_DEVICE\n"))
-			goto out_unlock;
-
-		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
-
-		/* Bind the P2P_DEVICE to the new allocated PHY context */
-		mvmvif->deflink.phy_ctxt = phy_ctxt;
-
-		ret = iwl_mvm_binding_add_vif(mvm, vif);
-		if (WARN(ret, "Failed binding P2P_DEVICE\n"))
+		ret = iwl_mvm_roc_switch_binding(mvm, vif, phy_ctxt);
+		if (ret)
 			goto out_unlock;
 
 		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
-- 
cgit v1.2.3-70-g09d2


From fe8b2ad361fb92f4b2b7b003c4609a2a7ed1fbda Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:47 +0300
Subject: wifi: iwlwifi: mvm: add cancel/remain_on_channel for MLD mode

Add an MLD version of the remain_on_channel and
cancel_remain_on_channel callbacks.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.b51813dbebd4.Ia25bbd63d3138e4759237ce2be0cd0436fe01c0a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/mac-cfg.h    |  3 ++
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 24 +++++++++---
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 43 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 44 +++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 26 +++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  2 +
 .../net/wireless/intel/iwlwifi/mvm/time-event.c    |  6 +++
 7 files changed, 142 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index a111e7366d49..3c540aa9c0c8 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -491,11 +491,14 @@ struct iwl_link_config_cmd {
  * @STATION_TYPE_MCAST: the station used for BCAST / MCAST in GO. Will be
  *	suspended / resumed at the right timing depending on the clients'
  *	power save state and the DTIM timing
+ * @STATION_TYPE_AUX: aux sta. In the FW there is no need for a special type
+ *	for the aux sta, so this type is only for driver - internal use.
  */
 enum iwl_fw_sta_type {
 	STATION_TYPE_PEER,
 	STATION_TYPE_BCAST_MGMT,
 	STATION_TYPE_MCAST,
+	STATION_TYPE_AUX,
 }; /* STATION_TYPE_E_VER_1 */
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index bb7e4e26d99f..d36319bd9247 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4309,6 +4309,20 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 		       struct ieee80211_channel *channel,
 		       int duration,
 		       enum ieee80211_roc_type type)
+{
+	struct iwl_mvm_roc_ops ops = {
+		.add_aux_sta_for_hs20 = iwl_mvm_add_aux_sta_for_hs20,
+		.switch_phy_ctxt = iwl_mvm_roc_switch_binding,
+	};
+
+	return iwl_mvm_roc_common(hw, vif, channel, duration, type, &ops);
+}
+
+/* Execute the common part for MLD and non-MLD modes */
+int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       struct ieee80211_channel *channel, int duration,
+		       enum ieee80211_roc_type type,
+		       struct iwl_mvm_roc_ops *ops)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -4334,7 +4348,7 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 		lmac_id = iwl_mvm_get_lmac_id(mvm->fw, channel->band);
 
 		/* Use aux roc framework (HS20) */
-		ret = iwl_mvm_add_aux_sta_for_hs20(mvm, lmac_id);
+		ret = ops->add_aux_sta_for_hs20(mvm, lmac_id);
 		if (!ret)
 			ret = iwl_mvm_send_aux_roc_cmd(mvm, channel,
 						       vif, duration);
@@ -4354,7 +4368,7 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 			continue;
 
 		if (phy_ctxt->ref && channel == phy_ctxt->channel) {
-			ret = iwl_mvm_roc_switch_binding(mvm, vif, phy_ctxt);
+			ret = ops->switch_phy_ctxt(mvm, vif, phy_ctxt);
 			if (ret)
 				goto out_unlock;
 
@@ -4408,7 +4422,7 @@ static int iwl_mvm_roc(struct ieee80211_hw *hw,
 			goto out_unlock;
 		}
 
-		ret = iwl_mvm_roc_switch_binding(mvm, vif, phy_ctxt);
+		ret = ops->switch_phy_ctxt(mvm, vif, phy_ctxt);
 		if (ret)
 			goto out_unlock;
 
@@ -4425,8 +4439,8 @@ out_unlock:
 	return ret;
 }
 
-static int iwl_mvm_cancel_roc(struct ieee80211_hw *hw,
-			      struct ieee80211_vif *vif)
+int iwl_mvm_cancel_roc(struct ieee80211_hw *hw,
+		       struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 9d97098ab57c..871da586ebb4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -592,10 +592,53 @@ iwl_mvm_mld_mac_conf_tx(struct ieee80211_hw *hw,
 	return 0;
 }
 
+static int iwl_mvm_link_switch_phy_ctx(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif,
+				       struct iwl_mvm_phy_ctxt *new_phy_ctxt)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret = 0;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	/* Inorder to change the phy_ctx of a link, the link needs to be
+	 * inactive. Therefore, first deactivate the link, then change its
+	 * phy_ctx, and then activate it again.
+	 */
+	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	if (WARN(ret, "Failed to deactivate link\n"))
+		return ret;
+
+	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
+
+	mvmvif->deflink.phy_ctxt = new_phy_ctxt;
+
+	ret = iwl_mvm_link_changed(mvm, vif, 0, false);
+	if (WARN(ret, "Failed to deactivate link\n"))
+		return ret;
+
+	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, true);
+	WARN(ret, "Failed binding P2P_DEVICE\n");
+	return ret;
+}
+
+static int iwl_mvm_mld_roc(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			   struct ieee80211_channel *channel, int duration,
+			   enum ieee80211_roc_type type)
+{
+	struct iwl_mvm_roc_ops ops = {
+		.add_aux_sta_for_hs20 = iwl_mvm_mld_add_aux_sta,
+		.switch_phy_ctxt = iwl_mvm_link_switch_phy_ctx,
+	};
+
+	return iwl_mvm_roc_common(hw, vif, channel, duration, type, &ops);
+}
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
 	.config_iface_filter = iwl_mvm_mld_config_iface_filter,
+	.remain_on_channel = iwl_mvm_mld_roc,
+	.cancel_remain_on_channel = iwl_mvm_cancel_roc,
 	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
 	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
 	.switch_vif_chanctx = iwl_mvm_mld_switch_vif_chanctx,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 16723d581a3d..643eb58317f0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -75,6 +75,24 @@ static int iwl_mvm_mld_rm_sta_from_fw(struct iwl_mvm *mvm, u32 sta_id)
 	return 0;
 }
 
+static int iwl_mvm_add_aux_sta_to_fw(struct iwl_mvm *mvm,
+				     struct iwl_mvm_int_sta *sta,
+				     u32 lmac_id)
+{
+	int ret;
+
+	struct iwl_mvm_aux_sta_cmd cmd = {
+		.sta_id = cpu_to_le32(sta->sta_id),
+		.lmac_id = cpu_to_le32(lmac_id),
+	};
+
+	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(MAC_CONF_GROUP, AUX_STA_CMD),
+				   0, sizeof(cmd), &cmd);
+	if (ret)
+		IWL_ERR(mvm, "Failed to send AUX_STA_CMD\n");
+	return ret;
+}
+
 /*
  * Adds an internal sta to the FW table with its queues
  */
@@ -91,7 +109,10 @@ static int iwl_mvm_mld_add_int_sta_with_queue(struct iwl_mvm *mvm,
 	if (WARN_ON_ONCE(sta->sta_id == IWL_MVM_INVALID_STA))
 		return -ENOSPC;
 
-	ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, phy_id);
+	if (sta->type == STATION_TYPE_AUX)
+		ret = iwl_mvm_add_aux_sta_to_fw(mvm, sta, phy_id);
+	else
+		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, phy_id);
 	if (ret)
 		return ret;
 
@@ -224,6 +245,19 @@ int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 				       IWL_MAX_TID_COUNT, NULL);
 }
 
+int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
+{
+	lockdep_assert_held(&mvm->mutex);
+
+	/* In CDB NICs we need to specify which lmac to use for aux activity
+	 * using the phy_id argument place to send lmac_id to the function
+	 */
+	return iwl_mvm_mld_add_int_sta(mvm, &mvm->aux_sta, &mvm->aux_queue,
+				       NL80211_IFTYPE_UNSPECIFIED,
+				       STATION_TYPE_AUX, lmac_id, NULL,
+				       IWL_MAX_TID_COUNT, NULL);
+}
+
 static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm,
 				   struct ieee80211_sta *sta,
 				   u16 *queueptr, u8 tid)
@@ -332,6 +366,14 @@ int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 				      IWL_MAX_TID_COUNT, &mvm->snif_queue);
 }
 
+int iwl_mvm_mld_rm_aux_sta(struct iwl_mvm *mvm)
+{
+	lockdep_assert_held(&mvm->mutex);
+
+	return iwl_mvm_mld_rm_int_sta(mvm, &mvm->aux_sta, false,
+				      IWL_MAX_TID_COUNT, &mvm->aux_queue);
+}
+
 /* send a cfg sta command to add/update a sta in firmware */
 static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			       struct ieee80211_vif *vif, u16 phy_id)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 2ea224acdea1..ba59a6c30529 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1827,6 +1827,32 @@ void iwl_mvm_bss_info_changed_station_assoc(struct iwl_mvm *mvm,
 					    struct ieee80211_vif *vif,
 					    u64 changes);
 
+/* ROC */
+/**
+ * struct iwl_mvm_roc_ops - callbacks for the remain_on_channel()
+ *
+ * Since the only difference between both MLD and
+ * non-MLD versions of remain_on_channel() is these function calls,
+ * each version will send its specific function calls to
+ * %iwl_mvm_roc_common().
+ *
+ * @add_aux_sta_for_hs20: pointer to the function that adds an aux sta
+ *	for Hot Spot 2.0
+ * @switch_phy_ctxt: pointer to the function that switches a vif from one
+ *	phy_ctx to another
+ */
+struct iwl_mvm_roc_ops {
+	int (*add_aux_sta_for_hs20)(struct iwl_mvm *mvm, u32 lmac_id);
+	int (*switch_phy_ctxt)(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			       struct iwl_mvm_phy_ctxt *new_phy_ctxt);
+};
+
+int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       struct ieee80211_channel *channel, int duration,
+		       enum ieee80211_roc_type type,
+		       struct iwl_mvm_roc_ops *ops);
+int iwl_mvm_cancel_roc(struct ieee80211_hw *hw,
+		       struct ieee80211_vif *vif);
 /*Session Protection */
 void iwl_mvm_protect_assoc(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			   u32 duration_override);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 770516b20303..5436e52ca639 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -622,9 +622,11 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id);
 int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_rm_aux_sta(struct iwl_mvm *mvm);
 int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta);
 int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index 3b17dd998d19..6b7b6250f1bb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -95,6 +95,11 @@ void iwl_mvm_roc_done_wk(struct work_struct *wk)
 		/* do the same in case of hot spot 2.0 */
 		iwl_mvm_flush_sta(mvm, &mvm->aux_sta, true);
 
+		if (mvm->mld_api_is_used) {
+			iwl_mvm_mld_rm_aux_sta(mvm);
+			goto out_unlock;
+		}
+
 		/* In newer version of this command an aux station is added only
 		 * in cases of dedicated tx queue and need to be removed in end
 		 * of use */
@@ -102,6 +107,7 @@ void iwl_mvm_roc_done_wk(struct work_struct *wk)
 			iwl_mvm_rm_aux_sta(mvm);
 	}
 
+out_unlock:
 	mutex_unlock(&mvm->mutex);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 56f4f12ba8f6d8cb52f776dfcd261f2e1040d38c Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:48 +0300
Subject: wifi: iwlwifi: mvm: unite sta_modify_disable_tx flows

These flows are the same in both MLD API and the current API,
except for the commands that are being sent during this flows.
Instead of checking each time before calling these floews
what API we use and then call the correct function, call always the old
one, which in turn will call the new one in case we're using the MLD
API.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.5692d8dea9be.Ib1882b2c2f0b0603abc4b7d4a0ecc45cd1fbf9a7@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  6 ++--
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  4 ---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 36 +++++++++++++---------
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c       | 18 ++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  6 ++++
 5 files changed, 47 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d36319bd9247..ce2f502bfff8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4762,6 +4762,9 @@ bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
 
 		mvmvif->csa_countdown = false;
 
+		/* Set CS bit on all the stations */
+		iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, true);
+
 		/* Save blocked iface, the timeout is set on the next beacon */
 		rcu_assign_pointer(mvm->csa_tx_blocked_vif, vif);
 
@@ -4787,9 +4790,6 @@ static void __iwl_mvm_unassign_vif_chanctx(struct iwl_mvm *mvm,
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		iwl_mvm_rm_snif_sta(mvm, vif);
 
-	if (vif->type == NL80211_IFTYPE_AP)
-		/* Set CS bit on all the stations */
-		iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, true);
 
 	if (vif->type == NL80211_IFTYPE_STATION && switching_chanctx) {
 		disabled_vif = vif;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 871da586ebb4..92f174231f60 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -220,10 +220,6 @@ static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		iwl_mvm_mld_rm_snif_sta(mvm, vif);
 
-	if (vif->type == NL80211_IFTYPE_AP)
-		/* Set CS bit on all the stations */
-		iwl_mvm_mld_modify_all_sta_disable_tx(mvm, mvmvif, true);
-
 	/* Link needs to be deactivated before removal */
 	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
 	iwl_mvm_remove_link(mvm, vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 643eb58317f0..a6eb18830cd9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -589,24 +589,14 @@ int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	return ret;
 }
 
-static void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
-					      struct ieee80211_sta *sta,
-					      bool disable)
+void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
+				       struct iwl_mvm_sta *mvmsta,
+				       bool disable)
 {
-	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_sta_disable_tx_cmd cmd;
 	int ret;
 
-	spin_lock_bh(&mvm_sta->lock);
-
-	if (mvm_sta->disable_tx == disable) {
-		spin_unlock_bh(&mvm_sta->lock);
-		return;
-	}
-
-	mvm_sta->disable_tx = disable;
-
-	cmd.sta_id = cpu_to_le32(mvm_sta->deflink.sta_id);
+	cmd.sta_id = cpu_to_le32(mvmsta->deflink.sta_id);
 	cmd.disable = cpu_to_le32(disable);
 
 	ret = iwl_mvm_send_cmd_pdu(mvm,
@@ -616,6 +606,22 @@ static void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
 		IWL_ERR(mvm,
 			"Failed to send STA_DISABLE_TX_CMD command (%d)\n",
 			ret);
+}
+
+void iwl_mvm_mld_sta_modify_disable_tx_ap(struct iwl_mvm *mvm,
+					  struct ieee80211_sta *sta,
+					  bool disable)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+
+	spin_lock_bh(&mvm_sta->lock);
+
+	if (mvm_sta->disable_tx == disable) {
+		spin_unlock_bh(&mvm_sta->lock);
+		return;
+	}
+
+	iwl_mvm_mld_sta_modify_disable_tx(mvm, mvm_sta, disable);
 
 	spin_unlock_bh(&mvm_sta->lock);
 }
@@ -641,7 +647,7 @@ void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 		    FW_CMD_ID_AND_COLOR(mvmvif->id, mvmvif->color))
 			continue;
 
-		iwl_mvm_mld_sta_modify_disable_tx(mvm, sta, disable);
+		iwl_mvm_mld_sta_modify_disable_tx(mvm, mvm_sta, disable);
 	}
 
 	rcu_read_unlock();
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 0ff66ffc5bee..54ef749be488 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -4030,7 +4030,8 @@ void iwl_mvm_rx_eosp_notif(struct iwl_mvm *mvm,
 }
 
 void iwl_mvm_sta_modify_disable_tx(struct iwl_mvm *mvm,
-				   struct iwl_mvm_sta *mvmsta, bool disable)
+				   struct iwl_mvm_sta *mvmsta,
+				   bool disable)
 {
 	struct iwl_mvm_add_sta_cmd cmd = {
 		.add_modify = STA_MODE_MODIFY,
@@ -4041,6 +4042,11 @@ void iwl_mvm_sta_modify_disable_tx(struct iwl_mvm *mvm,
 	};
 	int ret;
 
+	if (mvm->mld_api_is_used) {
+		iwl_mvm_mld_sta_modify_disable_tx(mvm, mvmsta, disable);
+		return;
+	}
+
 	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC,
 				   iwl_mvm_add_sta_cmd_size(mvm), &cmd);
 	if (ret)
@@ -4053,6 +4059,11 @@ void iwl_mvm_sta_modify_disable_tx_ap(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 
+	if (mvm->mld_api_is_used) {
+		iwl_mvm_mld_sta_modify_disable_tx_ap(mvm, sta, disable);
+		return;
+	}
+
 	spin_lock_bh(&mvm_sta->lock);
 
 	if (mvm_sta->disable_tx == disable) {
@@ -4103,6 +4114,11 @@ void iwl_mvm_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 	struct iwl_mvm_sta *mvm_sta;
 	int i;
 
+	if (mvm->mld_api_is_used) {
+		iwl_mvm_mld_modify_all_sta_disable_tx(mvm, mvmvif, disable);
+		return;
+	}
+
 	rcu_read_lock();
 
 	/* Block/unblock all the stations of the given mvmvif */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 5436e52ca639..3a9002eeb5fb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -640,4 +640,10 @@ int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 					   struct iwl_mvm_vif *mvmvif,
 					   bool disable);
+void iwl_mvm_mld_sta_modify_disable_tx(struct iwl_mvm *mvm,
+				       struct iwl_mvm_sta *mvm_sta,
+				       bool disable);
+void iwl_mvm_mld_sta_modify_disable_tx_ap(struct iwl_mvm *mvm,
+					  struct ieee80211_sta *sta,
+					  bool disable);
 #endif /* __sta_h__ */
-- 
cgit v1.2.3-70-g09d2


From 03117f30b998430d83842057a1ecf0382618acfb Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:49 +0300
Subject: wifi: iwlwifi: mvm: add support for post_channel_switch in MLD mode

Adjust the existing iwl_mvm_post_channel_switch()
to the new MLD API and use it in the new MLD ieee80211_ops

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.fa3992f7dfd2.Ie298a9b1522e956d7b699f0432795548bc6e47f9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c     | 10 ++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c |  1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h          |  2 ++
 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index ce2f502bfff8..631d1becc6ab 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1317,8 +1317,8 @@ static int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	return iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0, len, &cmd);
 }
 
-static int iwl_mvm_post_channel_switch(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif)
+int iwl_mvm_post_channel_switch(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
@@ -1339,8 +1339,10 @@ static int iwl_mvm_post_channel_switch(struct ieee80211_hw *hw,
 		}
 
 		iwl_mvm_sta_modify_disable_tx(mvm, mvmsta, false);
-
-		iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
+		if (mvm->mld_api_is_used)
+			iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
+		else
+			iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
 
 		if (!fw_has_capa(&mvm->fw->ucode_capa,
 				 IWL_UCODE_TLV_CAPA_CHANNEL_SWITCH_CMD)) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 92f174231f60..818aed050b73 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -640,6 +640,7 @@ const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.switch_vif_chanctx = iwl_mvm_mld_switch_vif_chanctx,
 	.join_ibss = iwl_mvm_mld_start_ap_ibss,
 	.leave_ibss = iwl_mvm_mld_stop_ap_ibss,
+	.post_channel_switch = iwl_mvm_post_channel_switch,
 	.sta_state = iwl_mvm_mld_mac_sta_state,
 	.conf_tx = iwl_mvm_mld_mac_conf_tx,
 	.bss_info_changed = iwl_mvm_mld_bss_info_changed,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ba59a6c30529..bf1de6405cd6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2298,6 +2298,8 @@ static inline u8 iwl_mvm_phy_band_from_nl80211(enum nl80211_band band)
 
 /* Channel Switch */
 void iwl_mvm_channel_switch_disconnect_wk(struct work_struct *wk);
+int iwl_mvm_post_channel_switch(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif);
 
 /* Channel Context */
 /**
-- 
cgit v1.2.3-70-g09d2


From cbce62a315f67e3badfd3f4a7b33e21c90ddc415 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:50 +0300
Subject: wifi: iwlwifi: mvm: add all missing ops to iwl_mvm_mld_ops

Add all the callbacks that are not changing with the
new MLD API and register to mac80211 with the new ops.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.a2f724342522.I5d1d6a8f5f14e6275da56ea704c3c0063fee5226@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 200 ++++++++++-----------
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 114 +++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 135 ++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |   4 +-
 4 files changed, 334 insertions(+), 119 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 631d1becc6ab..e922cc45c390 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -258,8 +258,7 @@ static const struct wiphy_iftype_ext_capab add_iftypes_ext_capa[] = {
 	},
 };
 
-static int
-iwl_mvm_op_get_antenna(struct ieee80211_hw *hw, u32 *tx_ant, u32 *rx_ant)
+int iwl_mvm_op_get_antenna(struct ieee80211_hw *hw, u32 *tx_ant, u32 *rx_ant)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	*tx_ant = iwl_mvm_get_valid_tx_ant(mvm);
@@ -703,9 +702,8 @@ static void iwl_mvm_tx_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 	ieee80211_free_txskb(mvm->hw, skb);
 }
 
-static void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
-			   struct ieee80211_tx_control *control,
-			   struct sk_buff *skb)
+void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
+		    struct ieee80211_tx_control *control, struct sk_buff *skb)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct ieee80211_sta *sta = control->sta;
@@ -801,8 +799,8 @@ void iwl_mvm_mac_itxq_xmit(struct ieee80211_hw *hw, struct ieee80211_txq *txq)
 	rcu_read_unlock();
 }
 
-static void iwl_mvm_mac_wake_tx_queue(struct ieee80211_hw *hw,
-				      struct ieee80211_txq *txq)
+void iwl_mvm_mac_wake_tx_queue(struct ieee80211_hw *hw,
+			       struct ieee80211_txq *txq)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_txq *mvmtxq = iwl_mvm_txq_from_mac80211(txq);
@@ -897,9 +895,9 @@ iwl_mvm_ampdu_check_trigger(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	}
 }
 
-static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
-				    struct ieee80211_vif *vif,
-				    struct ieee80211_ampdu_params *params)
+int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_ampdu_params *params)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -1095,7 +1093,7 @@ int __iwl_mvm_mac_start(struct iwl_mvm *mvm)
 	return ret;
 }
 
-static int iwl_mvm_mac_start(struct ieee80211_hw *hw)
+int iwl_mvm_mac_start(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -1164,9 +1162,8 @@ static void iwl_mvm_restart_complete(struct iwl_mvm *mvm)
 	mutex_unlock(&mvm->mutex);
 }
 
-static void
-iwl_mvm_mac_reconfig_complete(struct ieee80211_hw *hw,
-			      enum ieee80211_reconfig_type reconfig_type)
+void iwl_mvm_mac_reconfig_complete(struct ieee80211_hw *hw,
+				   enum ieee80211_reconfig_type reconfig_type)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -1228,7 +1225,7 @@ void __iwl_mvm_mac_stop(struct iwl_mvm *mvm)
 	}
 }
 
-static void iwl_mvm_mac_stop(struct ieee80211_hw *hw)
+void iwl_mvm_mac_stop(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -1366,8 +1363,8 @@ out_unlock:
 	return ret;
 }
 
-static void iwl_mvm_abort_channel_switch(struct ieee80211_hw *hw,
-					 struct ieee80211_vif *vif)
+void iwl_mvm_abort_channel_switch(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -1731,11 +1728,6 @@ out:
 	mutex_unlock(&mvm->mutex);
 }
 
-static int iwl_mvm_mac_config(struct ieee80211_hw *hw, u32 changed)
-{
-	return 0;
-}
-
 struct iwl_mvm_mc_iter_data {
 	struct iwl_mvm *mvm;
 	int port_id;
@@ -1807,8 +1799,8 @@ static void iwl_mvm_recalc_multicast(struct iwl_mvm *mvm)
 		IWL_ERR(mvm, "Failed to synchronize multicast groups update\n");
 }
 
-static u64 iwl_mvm_prepare_multicast(struct ieee80211_hw *hw,
-				     struct netdev_hw_addr_list *mc_list)
+u64 iwl_mvm_prepare_multicast(struct ieee80211_hw *hw,
+			      struct netdev_hw_addr_list *mc_list)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mcast_filter_cmd *cmd;
@@ -1844,10 +1836,9 @@ static u64 iwl_mvm_prepare_multicast(struct ieee80211_hw *hw,
 	return (u64)(unsigned long)cmd;
 }
 
-static void iwl_mvm_configure_filter(struct ieee80211_hw *hw,
-				     unsigned int changed_flags,
-				     unsigned int *total_flags,
-				     u64 multicast)
+void iwl_mvm_configure_filter(struct ieee80211_hw *hw,
+			      unsigned int changed_flags,
+			      unsigned int *total_flags, u64 multicast)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mcast_filter_cmd *cmd = (void *)(unsigned long)multicast;
@@ -3038,9 +3029,8 @@ iwl_mvm_bss_info_changed_common(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static int iwl_mvm_mac_hw_scan(struct ieee80211_hw *hw,
-			       struct ieee80211_vif *vif,
-			       struct ieee80211_scan_request *hw_req)
+int iwl_mvm_mac_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			struct ieee80211_scan_request *hw_req)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -3056,8 +3046,8 @@ static int iwl_mvm_mac_hw_scan(struct ieee80211_hw *hw,
 	return ret;
 }
 
-static void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif)
+void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -3076,7 +3066,7 @@ static void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static void
+void
 iwl_mvm_mac_allow_buffered_frames(struct ieee80211_hw *hw,
 				  struct ieee80211_sta *sta, u16 tids,
 				  int num_frames,
@@ -3091,7 +3081,7 @@ iwl_mvm_mac_allow_buffered_frames(struct ieee80211_hw *hw,
 					  tids, more_data, false);
 }
 
-static void
+void
 iwl_mvm_mac_release_buffered_frames(struct ieee80211_hw *hw,
 				    struct ieee80211_sta *sta, u16 tids,
 				    int num_frames,
@@ -3165,10 +3155,8 @@ static void __iwl_mvm_mac_sta_notify(struct ieee80211_hw *hw,
 	spin_unlock_bh(&mvmsta->lock);
 }
 
-static void iwl_mvm_mac_sta_notify(struct ieee80211_hw *hw,
-				   struct ieee80211_vif *vif,
-				   enum sta_notify_cmd cmd,
-				   struct ieee80211_sta *sta)
+void iwl_mvm_mac_sta_notify(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    enum sta_notify_cmd cmd, struct ieee80211_sta *sta)
 {
 	__iwl_mvm_mac_sta_notify(hw, cmd, sta);
 }
@@ -3226,9 +3214,9 @@ void iwl_mvm_sta_pm_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
 	rcu_read_unlock();
 }
 
-static void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       struct ieee80211_sta *sta)
+void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
@@ -3710,7 +3698,7 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 	return ret;
 }
 
-static int iwl_mvm_mac_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
+int iwl_mvm_mac_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -3719,9 +3707,8 @@ static int iwl_mvm_mac_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 	return 0;
 }
 
-static void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw,
-				  struct ieee80211_vif *vif,
-				  struct ieee80211_sta *sta, u32 changed)
+void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			   struct ieee80211_sta *sta, u32 changed)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -3763,9 +3750,9 @@ static int iwl_mvm_mac_conf_tx(struct ieee80211_hw *hw,
 	return 0;
 }
 
-static void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       struct ieee80211_prep_tx_info *info)
+void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_prep_tx_info *info)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -3774,9 +3761,9 @@ static void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static void iwl_mvm_mac_mgd_complete_tx(struct ieee80211_hw *hw,
-					struct ieee80211_vif *vif,
-					struct ieee80211_prep_tx_info *info)
+void iwl_mvm_mac_mgd_complete_tx(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_prep_tx_info *info)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -3789,10 +3776,10 @@ static void iwl_mvm_mac_mgd_complete_tx(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static int iwl_mvm_mac_sched_scan_start(struct ieee80211_hw *hw,
-					struct ieee80211_vif *vif,
-					struct cfg80211_sched_scan_request *req,
-					struct ieee80211_scan_ies *ies)
+int iwl_mvm_mac_sched_scan_start(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct cfg80211_sched_scan_request *req,
+				 struct ieee80211_scan_ies *ies)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -3812,8 +3799,8 @@ out:
 	return ret;
 }
 
-static int iwl_mvm_mac_sched_scan_stop(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif)
+int iwl_mvm_mac_sched_scan_stop(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -4069,11 +4056,9 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 	return ret;
 }
 
-static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
-			       enum set_key_cmd cmd,
-			       struct ieee80211_vif *vif,
-			       struct ieee80211_sta *sta,
-			       struct ieee80211_key_conf *key)
+int iwl_mvm_mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
+			struct ieee80211_vif *vif, struct ieee80211_sta *sta,
+			struct ieee80211_key_conf *key)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -4085,11 +4070,11 @@ static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 	return ret;
 }
 
-static void iwl_mvm_mac_update_tkip_key(struct ieee80211_hw *hw,
-					struct ieee80211_vif *vif,
-					struct ieee80211_key_conf *keyconf,
-					struct ieee80211_sta *sta,
-					u32 iv32, u16 *phase1key)
+void iwl_mvm_mac_update_tkip_key(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_key_conf *keyconf,
+				 struct ieee80211_sta *sta,
+				 u32 iv32, u16 *phase1key)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -4519,8 +4504,8 @@ out:
 	return ret;
 }
 
-static int iwl_mvm_add_chanctx(struct ieee80211_hw *hw,
-			       struct ieee80211_chanctx_conf *ctx)
+int iwl_mvm_add_chanctx(struct ieee80211_hw *hw,
+			struct ieee80211_chanctx_conf *ctx)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -4543,8 +4528,8 @@ static void __iwl_mvm_remove_chanctx(struct iwl_mvm *mvm,
 	iwl_mvm_phy_ctxt_unref(mvm, phy_ctxt);
 }
 
-static void iwl_mvm_remove_chanctx(struct ieee80211_hw *hw,
-				   struct ieee80211_chanctx_conf *ctx)
+void iwl_mvm_remove_chanctx(struct ieee80211_hw *hw,
+			    struct ieee80211_chanctx_conf *ctx)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -4553,9 +4538,8 @@ static void iwl_mvm_remove_chanctx(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static void iwl_mvm_change_chanctx(struct ieee80211_hw *hw,
-				   struct ieee80211_chanctx_conf *ctx,
-				   u32 changed)
+void iwl_mvm_change_chanctx(struct ieee80211_hw *hw,
+			    struct ieee80211_chanctx_conf *ctx, u32 changed)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	u16 *phy_ctxt_id = (u16 *)ctx->drv_priv;
@@ -4963,16 +4947,15 @@ static int iwl_mvm_switch_vif_chanctx(struct ieee80211_hw *hw,
 	return iwl_mvm_switch_vif_chanctx_common(hw, vifs, n_vifs, mode, &ops);
 }
 
-static int iwl_mvm_tx_last_beacon(struct ieee80211_hw *hw)
+int iwl_mvm_tx_last_beacon(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
 	return mvm->ibss_manager;
 }
 
-static int iwl_mvm_set_tim(struct ieee80211_hw *hw,
-			   struct ieee80211_sta *sta,
-			   bool set)
+int iwl_mvm_set_tim(struct ieee80211_hw *hw, struct ieee80211_sta *sta,
+		    bool set)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
@@ -5038,9 +5021,9 @@ static int __iwl_mvm_mac_testmode_cmd(struct iwl_mvm *mvm,
 	return -EOPNOTSUPP;
 }
 
-static int iwl_mvm_mac_testmode_cmd(struct ieee80211_hw *hw,
-				    struct ieee80211_vif *vif,
-				    void *data, int len)
+int iwl_mvm_mac_testmode_cmd(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     void *data, int len)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int err;
@@ -5053,9 +5036,8 @@ static int iwl_mvm_mac_testmode_cmd(struct ieee80211_hw *hw,
 }
 #endif
 
-static void iwl_mvm_channel_switch(struct ieee80211_hw *hw,
-				   struct ieee80211_vif *vif,
-				   struct ieee80211_channel_switch *chsw)
+void iwl_mvm_channel_switch(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    struct ieee80211_channel_switch *chsw)
 {
 	/* By implementing this operation, we prevent mac80211 from
 	 * starting its own channel switch timer, so that we can call
@@ -5130,9 +5112,9 @@ static int iwl_mvm_old_pre_chan_sw_sta(struct iwl_mvm *mvm,
 }
 
 #define IWL_MAX_CSA_BLOCK_TX 1500
-static int iwl_mvm_pre_channel_switch(struct ieee80211_hw *hw,
-				      struct ieee80211_vif *vif,
-				      struct ieee80211_channel_switch *chsw)
+int iwl_mvm_pre_channel_switch(struct ieee80211_hw *hw,
+			       struct ieee80211_vif *vif,
+			       struct ieee80211_channel_switch *chsw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct ieee80211_vif *csa_vif;
@@ -5245,9 +5227,9 @@ out_unlock:
 	return ret;
 }
 
-static void iwl_mvm_channel_switch_rx_beacon(struct ieee80211_hw *hw,
-					     struct ieee80211_vif *vif,
-					     struct ieee80211_channel_switch *chsw)
+void iwl_mvm_channel_switch_rx_beacon(struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_channel_switch *chsw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -5332,8 +5314,8 @@ static void iwl_mvm_flush_no_vif(struct iwl_mvm *mvm, u32 queues, bool drop)
 	mutex_unlock(&mvm->mutex);
 }
 
-static void iwl_mvm_mac_flush(struct ieee80211_hw *hw,
-			      struct ieee80211_vif *vif, u32 queues, bool drop)
+void iwl_mvm_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       u32 queues, bool drop)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif;
@@ -5390,8 +5372,8 @@ static void iwl_mvm_mac_flush(struct ieee80211_hw *hw,
 		iwl_trans_wait_tx_queues_empty(mvm->trans, msk);
 }
 
-static int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
-				  struct survey_info *survey)
+int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
+			   struct survey_info *survey)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -5575,10 +5557,10 @@ static void iwl_mvm_set_sta_rate(u32 rate_n_flags, struct rate_info *rinfo)
 	}
 }
 
-static void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       struct ieee80211_sta *sta,
-				       struct station_info *sinfo)
+void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta,
+				struct station_info *sinfo)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -5719,9 +5701,9 @@ static void iwl_mvm_event_bar_rx_callback(struct iwl_mvm *mvm,
 				event->u.ba.ssn);
 }
 
-static void iwl_mvm_mac_event_callback(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       const struct ieee80211_event *event)
+void iwl_mvm_mac_event_callback(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				const struct ieee80211_event *event)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -5803,7 +5785,7 @@ out:
 	}
 }
 
-static void iwl_mvm_sync_rx_queues(struct ieee80211_hw *hw)
+void iwl_mvm_sync_rx_queues(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -5812,7 +5794,7 @@ static void iwl_mvm_sync_rx_queues(struct ieee80211_hw *hw)
 	mutex_unlock(&mvm->mutex);
 }
 
-static int
+int
 iwl_mvm_mac_get_ftm_responder_stats(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
 				    struct cfg80211_ftm_responder_stats *stats)
@@ -5841,9 +5823,8 @@ iwl_mvm_mac_get_ftm_responder_stats(struct ieee80211_hw *hw,
 	return 0;
 }
 
-static int iwl_mvm_start_pmsr(struct ieee80211_hw *hw,
-			      struct ieee80211_vif *vif,
-			      struct cfg80211_pmsr_request *request)
+int iwl_mvm_start_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       struct cfg80211_pmsr_request *request)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
@@ -5855,9 +5836,8 @@ static int iwl_mvm_start_pmsr(struct ieee80211_hw *hw,
 	return ret;
 }
 
-static void iwl_mvm_abort_pmsr(struct ieee80211_hw *hw,
-			       struct ieee80211_vif *vif,
-			       struct cfg80211_pmsr_request *request)
+void iwl_mvm_abort_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			struct cfg80211_pmsr_request *request)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 818aed050b73..5f54e57de8e6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -244,7 +244,8 @@ static void iwl_mvm_mld_unassign_vif_chanctx(struct ieee80211_hw *hw,
 }
 
 static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
-				     struct ieee80211_vif *vif)
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -319,8 +320,22 @@ out_unlock:
 	return ret;
 }
 
+static int iwl_mvm_mld_start_ap(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *link_conf)
+{
+	return iwl_mvm_mld_start_ap_ibss(hw, vif, link_conf);
+}
+
+static int iwl_mvm_mld_start_ibss(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif)
+{
+	return iwl_mvm_mld_start_ap_ibss(hw, vif, &vif->bss_conf);
+}
+
 static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
-				     struct ieee80211_vif *vif)
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
@@ -348,6 +363,19 @@ static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
+static void iwl_mvm_mld_stop_ap(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *link_conf)
+{
+	iwl_mvm_mld_stop_ap_ibss(hw, vif, link_conf);
+}
+
+static void iwl_mvm_mld_stop_ibss(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif)
+{
+	iwl_mvm_mld_stop_ap_ibss(hw, vif, &vif->bss_conf);
+}
+
 static int iwl_mvm_mld_mac_sta_state(struct ieee80211_hw *hw,
 				     struct ieee80211_vif *vif,
 				     struct ieee80211_sta *sta,
@@ -630,18 +658,90 @@ static int iwl_mvm_mld_roc(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	return iwl_mvm_roc_common(hw, vif, channel, duration, type, &ops);
 }
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
+	.tx = iwl_mvm_mac_tx,
+	.wake_tx_queue = iwl_mvm_mac_wake_tx_queue,
+	.ampdu_action = iwl_mvm_mac_ampdu_action,
+	.get_antenna = iwl_mvm_op_get_antenna,
+	.start = iwl_mvm_mac_start,
+	.reconfig_complete = iwl_mvm_mac_reconfig_complete,
+	.stop = iwl_mvm_mac_stop,
 	.add_interface = iwl_mvm_mld_mac_add_interface,
 	.remove_interface = iwl_mvm_mld_mac_remove_interface,
+	.config = iwl_mvm_mac_config,
+	.prepare_multicast = iwl_mvm_prepare_multicast,
+	.configure_filter = iwl_mvm_configure_filter,
 	.config_iface_filter = iwl_mvm_mld_config_iface_filter,
+	.bss_info_changed = iwl_mvm_mld_bss_info_changed,
+	.hw_scan = iwl_mvm_mac_hw_scan,
+	.cancel_hw_scan = iwl_mvm_mac_cancel_hw_scan,
+	.sta_pre_rcu_remove = iwl_mvm_sta_pre_rcu_remove,
+	.sta_state = iwl_mvm_mld_mac_sta_state,
+	.sta_notify = iwl_mvm_mac_sta_notify,
+	.allow_buffered_frames = iwl_mvm_mac_allow_buffered_frames,
+	.release_buffered_frames = iwl_mvm_mac_release_buffered_frames,
+	.set_rts_threshold = iwl_mvm_mac_set_rts_threshold,
+	.sta_rc_update = iwl_mvm_sta_rc_update,
+	.conf_tx = iwl_mvm_mld_mac_conf_tx,
+	.mgd_prepare_tx = iwl_mvm_mac_mgd_prepare_tx,
+	.mgd_complete_tx = iwl_mvm_mac_mgd_complete_tx,
+	.mgd_protect_tdls_discover = iwl_mvm_mac_mgd_protect_tdls_discover,
+	.flush = iwl_mvm_mac_flush,
+	.sched_scan_start = iwl_mvm_mac_sched_scan_start,
+	.sched_scan_stop = iwl_mvm_mac_sched_scan_stop,
+	.set_key = iwl_mvm_mac_set_key,
+	.update_tkip_key = iwl_mvm_mac_update_tkip_key,
 	.remain_on_channel = iwl_mvm_mld_roc,
 	.cancel_remain_on_channel = iwl_mvm_cancel_roc,
+	.add_chanctx = iwl_mvm_add_chanctx,
+	.remove_chanctx = iwl_mvm_remove_chanctx,
+	.change_chanctx = iwl_mvm_change_chanctx,
 	.assign_vif_chanctx = iwl_mvm_mld_assign_vif_chanctx,
 	.unassign_vif_chanctx = iwl_mvm_mld_unassign_vif_chanctx,
 	.switch_vif_chanctx = iwl_mvm_mld_switch_vif_chanctx,
-	.join_ibss = iwl_mvm_mld_start_ap_ibss,
-	.leave_ibss = iwl_mvm_mld_stop_ap_ibss,
+
+	.start_ap = iwl_mvm_mld_start_ap,
+	.stop_ap = iwl_mvm_mld_stop_ap,
+	.join_ibss = iwl_mvm_mld_start_ibss,
+	.leave_ibss = iwl_mvm_mld_stop_ibss,
+
+	.tx_last_beacon = iwl_mvm_tx_last_beacon,
+
+	.set_tim = iwl_mvm_set_tim,
+
+	.channel_switch = iwl_mvm_channel_switch,
+	.pre_channel_switch = iwl_mvm_pre_channel_switch,
 	.post_channel_switch = iwl_mvm_post_channel_switch,
-	.sta_state = iwl_mvm_mld_mac_sta_state,
-	.conf_tx = iwl_mvm_mld_mac_conf_tx,
-	.bss_info_changed = iwl_mvm_mld_bss_info_changed,
+	.abort_channel_switch = iwl_mvm_abort_channel_switch,
+	.channel_switch_rx_beacon = iwl_mvm_channel_switch_rx_beacon,
+
+	.tdls_channel_switch = iwl_mvm_tdls_channel_switch,
+	.tdls_cancel_channel_switch = iwl_mvm_tdls_cancel_channel_switch,
+	.tdls_recv_channel_switch = iwl_mvm_tdls_recv_channel_switch,
+
+	.event_callback = iwl_mvm_mac_event_callback,
+
+	.sync_rx_queues = iwl_mvm_sync_rx_queues,
+
+	CFG80211_TESTMODE_CMD(iwl_mvm_mac_testmode_cmd)
+
+#ifdef CONFIG_PM_SLEEP
+	/* look at d3.c */
+	.suspend = iwl_mvm_suspend,
+	.resume = iwl_mvm_resume,
+	.set_wakeup = iwl_mvm_set_wakeup,
+	.set_rekey_data = iwl_mvm_set_rekey_data,
+#if IS_ENABLED(CONFIG_IPV6)
+	.ipv6_addr_change = iwl_mvm_ipv6_addr_change,
+#endif
+	.set_default_unicast_key = iwl_mvm_set_default_unicast_key,
+#endif
+	.get_survey = iwl_mvm_mac_get_survey,
+	.sta_statistics = iwl_mvm_mac_sta_statistics,
+	.get_ftm_responder_stats = iwl_mvm_mac_get_ftm_responder_stats,
+	.start_pmsr = iwl_mvm_start_pmsr,
+	.abort_pmsr = iwl_mvm_abort_pmsr,
+
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+	.sta_add_debugfs = iwl_mvm_sta_add_debugfs,
+#endif
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index bf1de6405cd6..6f4ed5d1efb9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1384,6 +1384,28 @@ static inline bool iwl_mvm_has_new_rx_api(struct iwl_mvm *mvm)
 			   IWL_UCODE_TLV_CAPA_MULTI_QUEUE_RX_SUPPORT);
 }
 
+static inline bool iwl_mvm_has_mld_api(const struct iwl_fw *fw)
+{
+	return (iwl_fw_lookup_cmd_ver(fw, LINK_CONFIG_CMD,
+				      IWL_FW_CMD_VER_UNKNOWN) !=
+				IWL_FW_CMD_VER_UNKNOWN) &&
+		(iwl_fw_lookup_cmd_ver(fw, MAC_CONFIG_CMD,
+				       IWL_FW_CMD_VER_UNKNOWN) !=
+				IWL_FW_CMD_VER_UNKNOWN) &&
+		(iwl_fw_lookup_cmd_ver(fw, STA_CONFIG_CMD,
+				       IWL_FW_CMD_VER_UNKNOWN) !=
+				IWL_FW_CMD_VER_UNKNOWN) &&
+		(iwl_fw_lookup_cmd_ver(fw, AUX_STA_CMD,
+				       IWL_FW_CMD_VER_UNKNOWN) !=
+				IWL_FW_CMD_VER_UNKNOWN) &&
+		(iwl_fw_lookup_cmd_ver(fw, STA_REMOVE_CMD,
+				       IWL_FW_CMD_VER_UNKNOWN) !=
+				IWL_FW_CMD_VER_UNKNOWN) &&
+		(iwl_fw_lookup_cmd_ver(fw, STA_DISABLE_TX_CMD,
+				       IWL_FW_CMD_VER_UNKNOWN) !=
+				IWL_FW_CMD_VER_UNKNOWN);
+}
+
 static inline bool iwl_mvm_has_new_tx_api(struct iwl_mvm *mvm)
 {
 	/* TODO - replace with TLV once defined */
@@ -2462,4 +2484,117 @@ void iwl_mvm_send_roaming_forbidden_event(struct iwl_mvm *mvm,
 					  struct ieee80211_vif *vif,
 					  bool forbidden);
 bool iwl_mvm_is_vendor_in_approved_list(void);
+
+/* Callbacks for ieee80211_ops */
+void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
+		    struct ieee80211_tx_control *control, struct sk_buff *skb);
+void iwl_mvm_mac_wake_tx_queue(struct ieee80211_hw *hw,
+			       struct ieee80211_txq *txq);
+
+int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_ampdu_params *params);
+int iwl_mvm_op_get_antenna(struct ieee80211_hw *hw, u32 *tx_ant, u32 *rx_ant);
+int iwl_mvm_mac_start(struct ieee80211_hw *hw);
+void iwl_mvm_mac_reconfig_complete(struct ieee80211_hw *hw,
+				   enum ieee80211_reconfig_type reconfig_type);
+void iwl_mvm_mac_stop(struct ieee80211_hw *hw);
+static inline int iwl_mvm_mac_config(struct ieee80211_hw *hw, u32 changed)
+{
+	return 0;
+}
+
+u64 iwl_mvm_prepare_multicast(struct ieee80211_hw *hw,
+			      struct netdev_hw_addr_list *mc_list);
+
+void iwl_mvm_configure_filter(struct ieee80211_hw *hw,
+			      unsigned int changed_flags,
+			      unsigned int *total_flags, u64 multicast);
+int iwl_mvm_mac_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			struct ieee80211_scan_request *hw_req);
+void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif);
+void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta);
+void iwl_mvm_mac_sta_notify(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    enum sta_notify_cmd cmd,
+			    struct ieee80211_sta *sta);
+void
+iwl_mvm_mac_allow_buffered_frames(struct ieee80211_hw *hw,
+				  struct ieee80211_sta *sta, u16 tids,
+				  int num_frames,
+				  enum ieee80211_frame_release_type reason,
+				  bool more_data);
+void
+iwl_mvm_mac_release_buffered_frames(struct ieee80211_hw *hw,
+				    struct ieee80211_sta *sta, u16 tids,
+				    int num_frames,
+				    enum ieee80211_frame_release_type reason,
+				    bool more_data);
+int iwl_mvm_mac_set_rts_threshold(struct ieee80211_hw *hw, u32 value);
+void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			   struct ieee80211_sta *sta, u32 changed);
+void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_prep_tx_info *info);
+void iwl_mvm_mac_mgd_complete_tx(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_prep_tx_info *info);
+void iwl_mvm_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       u32 queues, bool drop);
+int iwl_mvm_mac_sched_scan_start(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct cfg80211_sched_scan_request *req,
+				 struct ieee80211_scan_ies *ies);
+int iwl_mvm_mac_sched_scan_stop(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif);
+int iwl_mvm_mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
+			struct ieee80211_vif *vif, struct ieee80211_sta *sta,
+			struct ieee80211_key_conf *key);
+void iwl_mvm_mac_update_tkip_key(struct ieee80211_hw *hw,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_key_conf *keyconf,
+				 struct ieee80211_sta *sta,
+				 u32 iv32, u16 *phase1key);
+int iwl_mvm_add_chanctx(struct ieee80211_hw *hw,
+			struct ieee80211_chanctx_conf *ctx);
+void iwl_mvm_remove_chanctx(struct ieee80211_hw *hw,
+			    struct ieee80211_chanctx_conf *ctx);
+void iwl_mvm_change_chanctx(struct ieee80211_hw *hw,
+			    struct ieee80211_chanctx_conf *ctx, u32 changed);
+int iwl_mvm_tx_last_beacon(struct ieee80211_hw *hw);
+int iwl_mvm_set_tim(struct ieee80211_hw *hw, struct ieee80211_sta *sta,
+		    bool set);
+void iwl_mvm_channel_switch(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			    struct ieee80211_channel_switch *chsw);
+int iwl_mvm_pre_channel_switch(struct ieee80211_hw *hw,
+			       struct ieee80211_vif *vif,
+			       struct ieee80211_channel_switch *chsw);
+void iwl_mvm_abort_channel_switch(struct ieee80211_hw *hw,
+				  struct ieee80211_vif *vif);
+void iwl_mvm_channel_switch_rx_beacon(struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_channel_switch *chsw);
+void iwl_mvm_mac_event_callback(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				const struct ieee80211_event *event);
+void iwl_mvm_sync_rx_queues(struct ieee80211_hw *hw);
+int iwl_mvm_mac_testmode_cmd(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     void *data, int len);
+int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
+			   struct survey_info *survey);
+void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta,
+				struct station_info *sinfo);
+int
+iwl_mvm_mac_get_ftm_responder_stats(struct ieee80211_hw *hw,
+				    struct ieee80211_vif *vif,
+				    struct cfg80211_ftm_responder_stats *stats);
+int iwl_mvm_start_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       struct cfg80211_pmsr_request *request);
+void iwl_mvm_abort_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			struct cfg80211_pmsr_request *request);
 #endif /* __IWL_MVM_H__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 5c8784ddfa66..b355b9ebad0a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1116,6 +1116,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	 ********************************/
 	hw = ieee80211_alloc_hw(sizeof(struct iwl_op_mode) +
 				sizeof(struct iwl_mvm),
+				iwl_mvm_has_mld_api(fw) ? &iwl_mvm_mld_hw_ops :
 				&iwl_mvm_hw_ops);
 	if (!hw)
 		return NULL;
@@ -1295,8 +1296,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	mvm->sta_remove_requires_queue_remove =
 		trans_cfg.queue_alloc_cmd_ver > 0;
 
-	/* Until we register with the MLD ops - we don't use the MLD API */
-	mvm->mld_api_is_used = false;
+	mvm->mld_api_is_used = iwl_mvm_has_mld_api(mvm->fw);
 
 	/* Configure transport layer */
 	iwl_trans_configure(mvm->trans, &trans_cfg);
-- 
cgit v1.2.3-70-g09d2


From cea05a8651759135ee3c92d35bd13553669635d0 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:51 +0300
Subject: wifi: iwlwifi: mvm: fix "modify_mask" value in the link cmd.

This bitmap indicates what fields of the cmd got changed.
A field will be ignored by the FW if the corresponding flag wasn't set.
There are a few cases in which we currently set the wrong bits when
sending this cmd, which caused FW asserts. Fix this by setting the
correct bits in each case.

Fixes: 1ab26632332e ("wifi: iwlwifi: mvm: Add an add_interface() callback for mld mode")
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.19ddbee0c98d.I595abb79d0419c9a21e5234303c2c3fd5290a52a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 5f54e57de8e6..33368e5945f4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -54,9 +54,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 		if (ret)
 			goto out_unref_phy;
 
-		ret = iwl_mvm_link_changed(mvm, vif,
-					   LINK_CONTEXT_MODIFY_ACTIVE |
-					   LINK_CONTEXT_MODIFY_RATES_INFO,
+		ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
 					   true);
 		if (ret)
 			goto out_remove_link;
@@ -162,7 +160,8 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	ret = iwl_mvm_add_link(mvm, vif);
 	if (ret)
 		goto out;
-	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
+	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE |
+				   LINK_CONTEXT_MODIFY_RATES_INFO,
 				   true);
 	if (ret)
 		goto out_remove_link;
@@ -270,7 +269,7 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	if (ret)
 		goto out_remove_mac;
 
-	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
+	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ALL,
 				   true);
 	if (ret)
 		goto out_remove_link;
-- 
cgit v1.2.3-70-g09d2


From 75700ee1644f23ccb2742dbe623a1094a9aab753 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:52 +0300
Subject: wifi: iwlwifi: mvm: fix crash on queue removal for MLD API too

The patch linked below fixes the crash on queue removal bug only
for the non-MLD API. Do the same for the MLD API.

Fixes: c5a976cf6a75 ("wifi: iwlwifi: modify new queue allocation command")
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.527dace26147.Ia215df5833634f95688a979f39fae70c1ac4e027@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index a6eb18830cd9..fc380e59c740 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -258,27 +258,19 @@ int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
 				       IWL_MAX_TID_COUNT, NULL);
 }
 
-static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm,
-				   struct ieee80211_sta *sta,
+static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm, int sta_id,
 				   u16 *queueptr, u8 tid)
 {
-	struct iwl_mvm_sta *mvmsta;
 	int queue = *queueptr;
 	int ret = 0;
 
-	if (!sta)
-		return -EINVAL;
-
-	mvmsta = iwl_mvm_sta_from_mac80211(sta);
-
 	if (mvm->sta_remove_requires_queue_remove) {
 		u32 cmd_id = WIDE_ID(DATA_PATH_GROUP,
 				     SCD_QUEUE_CONFIG_CMD);
 		struct iwl_scd_queue_cfg_cmd remove_cmd = {
 			.operation = cpu_to_le32(IWL_SCD_QUEUE_REMOVE),
 			.u.remove.tid = cpu_to_le32(tid),
-			.u.remove.sta_mask =
-				cpu_to_le32(BIT(mvmsta->deflink.sta_id)),
+			.u.remove.sta_mask = cpu_to_le32(BIT(sta_id)),
 		};
 
 		ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0,
@@ -308,7 +300,7 @@ static int iwl_mvm_mld_rm_int_sta(struct iwl_mvm *mvm,
 	if (flush)
 		iwl_mvm_flush_sta(mvm, int_sta, true);
 
-	iwl_mvm_mld_disable_txq(mvm, NULL, queuptr, tid);
+	iwl_mvm_mld_disable_txq(mvm, int_sta->sta_id, queuptr, tid);
 
 	ret = iwl_mvm_mld_rm_sta_from_fw(mvm, int_sta->sta_id);
 	if (ret)
@@ -536,8 +528,8 @@ static void iwl_mvm_mld_disable_sta_queues(struct iwl_mvm *mvm,
 		if (mvm_sta->tid_data[i].txq_id == IWL_MVM_INVALID_QUEUE)
 			continue;
 
-		iwl_mvm_mld_disable_txq(mvm, sta, &mvm_sta->tid_data[i].txq_id,
-					i);
+		iwl_mvm_mld_disable_txq(mvm, mvm_sta->deflink.sta_id,
+					&mvm_sta->tid_data[i].txq_id, i);
 		mvm_sta->tid_data[i].txq_id = IWL_MVM_INVALID_QUEUE;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 786810de9cf735ebfdbd0a09a0126364ae217ca7 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:53 +0300
Subject: wifi: iwlwifi: mvm: modify link instead of removing it during csa

During CSA the PHY used by the link is changing, So the driver
needs to modify the links phy to the FW. Currently the driver is doing
it by removing the link and adding a new one with the new phy_id, but the
FW expects the link only to be modified. Fix this by modifying the links
phy_id instead of removing it and adding a new one

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.c07ca7aace29.I4ed5c77f4afe1b5da19322734e2f84d51aa541ad@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 33368e5945f4..5be8058ca669 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -157,9 +157,16 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 						switching_chanctx, &ret))
 		goto out;
 
-	ret = iwl_mvm_add_link(mvm, vif);
-	if (ret)
-		goto out;
+	if (!switching_chanctx) {
+		ret = iwl_mvm_add_link(mvm, vif);
+		if (ret)
+			goto out;
+	} else {
+		ret = iwl_mvm_link_changed(mvm, vif, 0, false);
+		if (ret)
+			goto out_remove_link;
+	}
+
 	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE |
 				   LINK_CONTEXT_MODIFY_RATES_INFO,
 				   true);
@@ -219,10 +226,9 @@ static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		iwl_mvm_mld_rm_snif_sta(mvm, vif);
 
-	/* Link needs to be deactivated before removal */
 	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
-	iwl_mvm_remove_link(mvm, vif);
-
+	if (!switching_chanctx)
+		iwl_mvm_remove_link(mvm, vif);
 out:
 	if (switching_chanctx)
 		return;
-- 
cgit v1.2.3-70-g09d2


From 5ed461be3ca93197cef6e8fc7fe2b523a892b317 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:54 +0300
Subject: wifi: iwlwifi: mvm: always use the sta->addr as the peers addr

Currently we're setting the sta->addr as the peers address only if
the iftype is NL80211_IFTYPE_AP, otherwise we are setting the bssid to
be the addr. This causes bugs in TDLS. Fix this by always using the
sta->addr.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.4c104c3074c4.I78912bb85251033e60db99a65165890779203612@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index fc380e59c740..41f6be7fe75e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -380,13 +380,8 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	u32 agg_size = 0, mpdu_dens = 0;
 
 	/* For now the link addr is the same as the mld addr */
-	if (vif->type == NL80211_IFTYPE_AP) {
-		memcpy(&cmd.peer_mld_address, sta->addr, ETH_ALEN);
-		memcpy(&cmd.peer_link_address, sta->addr, ETH_ALEN);
-	} else if (vif->bss_conf.bssid) {
-		memcpy(&cmd.peer_mld_address, vif->bss_conf.bssid, ETH_ALEN);
-		memcpy(&cmd.peer_link_address, vif->bss_conf.bssid, ETH_ALEN);
-	}
+	memcpy(&cmd.peer_mld_address, sta->addr, ETH_ALEN);
+	memcpy(&cmd.peer_link_address, sta->addr, ETH_ALEN);
 
 	if (mvm_sta->sta_state >= IEEE80211_STA_ASSOC)
 		cmd.assoc_id = cpu_to_le32(sta->aid);
-- 
cgit v1.2.3-70-g09d2


From 6b5a87df8ec8e3e24ffde62b1356aad9b4e9182d Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 28 Mar 2023 10:58:55 +0300
Subject: wifi: iwlwifi: mvm: align to the LINK cmd update in the FW

The LINK cmd host api has been updated. Align the driver to the
new changes. Also, temporary use mac_id for link_id.
Using the phy_id as the link_id is wrong since we might have 2 macs
operating on the same phy - in this case we will have 2 different
links (one for each mac) with the same link_id. On the other hand,
since we don't have MLO implemented yet, we won't have 2 different
links of the same mac. Therefore, we can use the mac_id as the
link_id.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104948.78ae716884fe.Icfeb2794d9652baaccf9b0cdddbd751d0db4f952@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/mac-cfg.h    | 17 ++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      | 22 +++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 40 ++++++++++------------
 3 files changed, 37 insertions(+), 42 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index 3c540aa9c0c8..b644274e4a31 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -383,11 +383,14 @@ enum iwl_link_ctx_protection_flags {
  * @LINK_FLG_RU_2MHZ_BLOCK: indicates that 26-tone RU OFDMA transmission are
  *      not allowed (as there are OBSS that might classify such transmissions as
  *      radar pulses).
+ * @LINK_FLG_NDP_FEEDBACK_ENABLED: mark support for NDP feedback and change
+ *	of threshold
  */
 enum iwl_link_ctx_flags {
 	LINK_FLG_BSS_COLOR_DIS		= BIT(0),
 	LINK_FLG_MU_EDCA_CW		= BIT(1),
 	LINK_FLG_RU_2MHZ_BLOCK		= BIT(2),
+	LINK_FLG_NDP_FEEDBACK_ENABLED	= BIT(3),
 }; /* LINK_CONTEXT_FLAG_E_VER_1 */
 
 /**
@@ -423,13 +426,9 @@ enum iwl_link_ctx_flags {
  * @rand_alloc_ecwmax: random CWmax = 2**ECWmax-1
  * @ndp_fdbk_buff_th_exp: set exponent for the NDP feedback buffered threshold
  * @trig_based_txf: MU EDCA Parameter set for the trigger based traffic queues
- * @dtim_time: DTIM arrival time in system time
- * @dtim_tsf: DTIM arrival time in TSF
- * @assoc_beacon_arrive_time: TSF of first beacon after association
  * @bi: beacon interval in TU, applicable only when associated
  * @dtim_interval: DTIM interval in TU.
  *	Relevant only for GO, otherwise this is offloaded.
- * @beacon_template: beacon template ID. For GO only
  * @puncture_mask: puncture mask for EHT
  * @frame_time_rts_th: HE duration RTS threshold, in units of 32us
  * @flags: a combination from &enum iwl_link_ctx_flags
@@ -440,6 +439,9 @@ enum iwl_link_ctx_flags {
  * @bssid_index: index of the associated VAP
  * @bss_color: 11ax AP ID that is used in the HE SIG-A to mark inter BSS frame
  * @reserved: alignment
+ * @ibss_bssid_addr: bssid for ibss
+ * @reserved_for_ibss_bssid_addr: reserved
+ * @reserved1: reserved for future use
  */
 struct iwl_link_config_cmd {
 	__le32 action;
@@ -464,12 +466,8 @@ struct iwl_link_config_cmd {
 	u8 rand_alloc_ecwmax;
 	u8 ndp_fdbk_buff_th_exp;
 	struct iwl_he_backoff_conf trig_based_txf[AC_NUM];
-	__le32 dtim_time;
-	__le64 dtim_tsf;
-	__le32 assoc_beacon_arrive_time;
 	__le32 bi;
 	__le32 dtim_interval;
-	__le32 beacon_template;
 	__le16 puncture_mask;
 	__le16 frame_time_rts_th;
 	__le32 flags;
@@ -480,6 +478,9 @@ struct iwl_link_config_cmd {
 	u8 bssid_index;
 	u8 bss_color;
 	u8 reserved[2];
+	u8 ibss_bssid_addr[6];
+	__le16 reserved_for_ibss_bssid_addr;
+	__le32 reserved1[8];
 } __packed; /* LINK_CONTEXT_CONFIG_CMD_API_S_VER_1 */
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 21e45f89c5f9..4905dac3a048 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -34,12 +34,15 @@ int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	if (iwl_mvm_sf_update(mvm, vif, false))
 		return -EINVAL;
 
-	cmd.link_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
+	cmd.link_id = cpu_to_le32(mvmvif->id);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
 	cmd.phy_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
 
 	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
 
+	if (vif->type == NL80211_IFTYPE_ADHOC && vif->bss_conf.bssid)
+		memcpy(cmd.ibss_bssid_addr, vif->bss_conf.bssid, ETH_ALEN);
+
 	return iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_ADD);
 }
 
@@ -54,7 +57,7 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (!phyctxt)
 		return -EINVAL;
 
-	cmd.link_id = cpu_to_le32(phyctxt->id);
+	cmd.link_id = cpu_to_le32(mvmvif->id);
 
 	/* The phy_id, link address and listen_lmac can be modified only until
 	 * the link becomes active, otherwise they will be ignored.
@@ -66,6 +69,9 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	cmd.active = cpu_to_le32(active);
 
+	if (vif->type == NL80211_IFTYPE_ADHOC && vif->bss_conf.bssid)
+		memcpy(cmd.ibss_bssid_addr, vif->bss_conf.bssid, ETH_ALEN);
+
 	/* TODO: set a value to cmd.listen_lmac when system requiremens
 	 * will define it
 	 */
@@ -82,21 +88,11 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	iwl_mvm_set_fw_qos_params(mvm, vif, &cmd.ac[0], &cmd.qos_flags);
 
-	/* We need the dtim_period to set the MAC as associated */
-	if (vif->cfg.assoc && vif->bss_conf.dtim_period)
-		iwl_mvm_set_fw_dtim_tbtt(mvm, vif, &cmd.dtim_tsf,
-					 &cmd.dtim_time,
-					 &cmd.assoc_beacon_arrive_time);
-	else
-		changes &= ~LINK_CONTEXT_MODIFY_BEACON_TIMING;
 
 	cmd.bi = cpu_to_le32(vif->bss_conf.beacon_int);
 	cmd.dtim_interval = cpu_to_le32(vif->bss_conf.beacon_int *
 					vif->bss_conf.dtim_period);
 
-	/* TODO: Assumes that the beacon id == mac context id */
-	cmd.beacon_template = cpu_to_le32(mvmvif->id);
-
 	if (!vif->bss_conf.he_support || iwlwifi_mod_params.disable_11ax ||
 	    !vif->cfg.assoc) {
 		changes &= ~LINK_CONTEXT_MODIFY_HE_PARAMS;
@@ -164,7 +160,7 @@ int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
 		return -EINVAL;
 
-	cmd.link_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
+	cmd.link_id = cpu_to_le32(mvmvif->id);
 	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_REMOVE);
 
 	if (!ret)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 41f6be7fe75e..1608ac371bb0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -22,7 +22,7 @@ static int iwl_mvm_mld_send_sta_cmd(struct iwl_mvm *mvm,
 static int iwl_mvm_mld_add_int_sta_to_fw(struct iwl_mvm *mvm,
 					 struct iwl_mvm_int_sta *sta,
 					 const u8 *addr,
-					 u16 phy_id)
+					 u16 mac_id)
 {
 	struct iwl_mvm_sta_cfg_cmd cmd;
 
@@ -31,7 +31,7 @@ static int iwl_mvm_mld_add_int_sta_to_fw(struct iwl_mvm *mvm,
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.sta_id = cpu_to_le32((u8)sta->sta_id);
 
-	cmd.link_id = cpu_to_le32(phy_id);
+	cmd.link_id = cpu_to_le32(mac_id);
 
 	cmd.station_type = cpu_to_le32(sta->type);
 
@@ -98,7 +98,7 @@ static int iwl_mvm_add_aux_sta_to_fw(struct iwl_mvm *mvm,
  */
 static int iwl_mvm_mld_add_int_sta_with_queue(struct iwl_mvm *mvm,
 					      struct iwl_mvm_int_sta *sta,
-					      const u8 *addr, int phy_id,
+					      const u8 *addr, int mac_id,
 					      u16 *queue, u8 tid,
 					      unsigned int *_wdg_timeout)
 {
@@ -110,9 +110,9 @@ static int iwl_mvm_mld_add_int_sta_with_queue(struct iwl_mvm *mvm,
 		return -ENOSPC;
 
 	if (sta->type == STATION_TYPE_AUX)
-		ret = iwl_mvm_add_aux_sta_to_fw(mvm, sta, phy_id);
+		ret = iwl_mvm_add_aux_sta_to_fw(mvm, sta, mac_id);
 	else
-		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, phy_id);
+		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, mac_id);
 	if (ret)
 		return ret;
 
@@ -139,7 +139,7 @@ static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
 				   struct iwl_mvm_int_sta *int_sta, u16 *queue,
 				   enum nl80211_iftype iftype,
 				   enum iwl_fw_sta_type sta_type,
-				   int phy_id, const u8 *addr, u8 tid,
+				   int mac_id, const u8 *addr, u8 tid,
 				   unsigned int *wdg_timeout)
 {
 	int ret;
@@ -152,7 +152,7 @@ static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
 	if (ret)
 		return ret;
 
-	ret = iwl_mvm_mld_add_int_sta_with_queue(mvm, int_sta, addr, phy_id,
+	ret = iwl_mvm_mld_add_int_sta_with_queue(mvm, int_sta, addr, mac_id,
 						 queue, tid, wdg_timeout);
 	if (ret) {
 		iwl_mvm_dealloc_int_sta(mvm, int_sta);
@@ -194,8 +194,8 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	return iwl_mvm_mld_add_int_sta(mvm, bsta, queue,
 				       ieee80211_vif_type_p2p(vif),
 				       STATION_TYPE_BCAST_MGMT,
-				       mvmvif->deflink.phy_ctxt->id, baddr,
-				       IWL_MAX_TID_COUNT, &wdg_timeout);
+				       mvmvif->id, baddr, IWL_MAX_TID_COUNT,
+				       &wdg_timeout);
 }
 
 /* Allocate a new station entry for the broadcast station to the given vif,
@@ -226,8 +226,7 @@ int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvmvif->deflink.cab_queue,
 				       vif->type, STATION_TYPE_MCAST,
-				       mvmvif->deflink.phy_ctxt->id, maddr, 0,
-				       &timeout);
+				       mvmvif->id, maddr, 0, &timeout);
 }
 
 /* Allocate a new station entry for the sniffer station to the given vif,
@@ -241,8 +240,8 @@ int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	return iwl_mvm_mld_add_int_sta(mvm, &mvm->snif_sta, &mvm->snif_queue,
 				       vif->type, STATION_TYPE_BCAST_MGMT,
-				       mvmvif->deflink.phy_ctxt->id, NULL,
-				       IWL_MAX_TID_COUNT, NULL);
+				       mvmvif->id, NULL, IWL_MAX_TID_COUNT,
+				       NULL);
 }
 
 int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
@@ -250,7 +249,7 @@ int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
 	lockdep_assert_held(&mvm->mutex);
 
 	/* In CDB NICs we need to specify which lmac to use for aux activity
-	 * using the phy_id argument place to send lmac_id to the function
+	 * using the mac_id argument place to send lmac_id to the function
 	 */
 	return iwl_mvm_mld_add_int_sta(mvm, &mvm->aux_sta, &mvm->aux_queue,
 				       NL80211_IFTYPE_UNSPECIFIED,
@@ -368,12 +367,13 @@ int iwl_mvm_mld_rm_aux_sta(struct iwl_mvm *mvm)
 
 /* send a cfg sta command to add/update a sta in firmware */
 static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
-			       struct ieee80211_vif *vif, u16 phy_id)
+			       struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta_cfg_cmd cmd = {
 		.sta_id = cpu_to_le32(mvm_sta->deflink.sta_id),
-		.link_id = cpu_to_le32(phy_id),
+		.link_id = cpu_to_le32(mvmvif->id),
 		.station_type = cpu_to_le32(mvm_sta->sta_type),
 		.mfp = cpu_to_le32(sta->mfp),
 	};
@@ -472,7 +472,7 @@ int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		 */
 		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, &tmp_sta,
 						    vif->bss_conf.bssid,
-						    mvmvif->deflink.phy_ctxt->id);
+						    mvmvif->id);
 		if (ret)
 			return ret;
 
@@ -482,7 +482,7 @@ int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				       STATION_TYPE_PEER);
 	}
 
-	ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif, mvmvif->deflink.phy_ctxt->id);
+	ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif);
 	if (ret)
 		return ret;
 
@@ -503,11 +503,9 @@ int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta)
 {
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-
 	lockdep_assert_held(&mvm->mutex);
 
-	return iwl_mvm_mld_cfg_sta(mvm, sta, vif, mvmvif->deflink.phy_ctxt->id);
+	return iwl_mvm_mld_cfg_sta(mvm, sta, vif);
 }
 
 static void iwl_mvm_mld_disable_sta_queues(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From 203b22acb1a93ca154c104fe29cfbc312571f2a0 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:58:56 +0300
Subject: wifi: iwlwifi: mvm: adjust smart fifo configuration to MLO

Consider properties of all available links for smart FIFO
configuration.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.870ff175b82c.Iada56a474a0770ae5469b31d635cc05405dbfb87@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sf.c | 32 ++++++++++++++++++++++++-----
 1 file changed, 27 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
index 84b750d02848..7c5f41e40e7a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
@@ -98,6 +98,10 @@ static void iwl_mvm_fill_sf_command(struct iwl_mvm *mvm,
 				    struct ieee80211_sta *sta)
 {
 	int i, j, watermark;
+	u8 max_rx_nss = 0;
+	bool is_legacy = true;
+	struct ieee80211_link_sta *link_sta;
+	unsigned int link_id;
 
 	sf_cmd->watermark[SF_LONG_DELAY_ON] = cpu_to_le32(SF_W_MARK_SCAN);
 
@@ -106,10 +110,25 @@ static void iwl_mvm_fill_sf_command(struct iwl_mvm *mvm,
 	 * capabilities of the AP station, and choose the watermark accordingly.
 	 */
 	if (sta) {
-		if (sta->deflink.ht_cap.ht_supported ||
-		    sta->deflink.vht_cap.vht_supported ||
-		    sta->deflink.he_cap.has_he) {
-			switch (sta->deflink.rx_nss) {
+		/* find the maximal NSS number among all links (if relevant) */
+		rcu_read_lock();
+		for (link_id = 0; link_id < ARRAY_SIZE(sta->link); link_id++) {
+			link_sta = rcu_dereference(sta->link[link_id]);
+			if (!link_sta)
+				continue;
+
+			if (link_sta->ht_cap.ht_supported ||
+			    link_sta->vht_cap.vht_supported ||
+			    link_sta->eht_cap.has_eht ||
+			    link_sta->he_cap.has_he) {
+				is_legacy = false;
+				max_rx_nss = max(max_rx_nss, link_sta->rx_nss);
+			}
+		}
+		rcu_read_unlock();
+
+		if (!is_legacy) {
+			switch (max_rx_nss) {
 			case 1:
 				watermark = SF_W_MARK_SISO;
 				break;
@@ -151,7 +170,6 @@ static void iwl_mvm_fill_sf_command(struct iwl_mvm *mvm,
 		memcpy(sf_cmd->full_on_timeouts, sf_full_timeout_def,
 		       sizeof(sf_full_timeout_def));
 	}
-
 }
 
 static int iwl_mvm_sf_config(struct iwl_mvm *mvm, u8 sta_id,
@@ -275,5 +293,9 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 		/* If there are multiple active macs - change to SF_UNINIT */
 		new_state = SF_UNINIT;
 	}
+
+	/* For MLO it's ok to use deflink->sta_id as it's needed only to get
+	 * a pointer to mac80211 sta
+	 */
 	return iwl_mvm_sf_config(mvm, sta_id, new_state);
 }
-- 
cgit v1.2.3-70-g09d2


From 36cf537798cb6c738b94a67ec1604c571e15b2b8 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:58:57 +0300
Subject: wifi: iwlwifi: mvm: adjust mld_mac_ctxt_/beacon_changed() for MLO

HE/EHT support is reported to FW if there's at least one link
that supports it. Configure beacon separately for each link.
Don't send the beacon template before adding the MAC.

Co-developed-by: Johannes Berg <johannes.berg@intel.com>
Co-developed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.5ef4efeda2dd.I6ebda2b71c964b9aa63240c9fa1ee0d28099fe6e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/tx.h     | 10 +++--
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c   | 22 ++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  | 24 +++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 39 ++++++++++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c   | 49 +++++++++++++++++++---
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 12 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  6 ++-
 7 files changed, 115 insertions(+), 47 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
index ecc6706f66ed..97edf5477ba7 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
@@ -800,7 +800,7 @@ enum iwl_mac_beacon_flags {
  *	is &enum iwl_mac_beacon_flags.
  * @short_ssid: Short SSID
  * @reserved: reserved
- * @template_id: currently equal to the mac context id of the coresponding mac.
+ * @link_id: the firmware id of the link that will use this beacon
  * @tim_idx: the offset of the tim IE in the beacon
  * @tim_size: the length of the tim IE
  * @ecsa_offset: offset to the ECSA IE if present
@@ -812,15 +812,17 @@ struct iwl_mac_beacon_cmd {
 	__le16 flags;
 	__le32 short_ssid;
 	__le32 reserved;
-	__le32 template_id;
+	__le32 link_id;
 	__le32 tim_idx;
 	__le32 tim_size;
 	__le32 ecsa_offset;
 	__le32 csa_offset;
 	struct ieee80211_hdr frame[];
 } __packed; /* BEACON_TEMPLATE_CMD_API_S_VER_10,
-	       BEACON_TEMPLATE_CMD_API_S_VER_11,
-	       BEACON_TEMPLATE_CMD_API_S_VER_12 */
+	     * BEACON_TEMPLATE_CMD_API_S_VER_11,
+	     * BEACON_TEMPLATE_CMD_API_S_VER_12,
+	     * BEACON_TEMPLATE_CMD_API_S_VER_13
+	     */
 
 struct iwl_beacon_notif {
 	struct iwl_mvm_tx_resp beacon_notify_hdr;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index b5eb0ee298ad..2c72f5cb19f1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -1388,6 +1388,7 @@ static int _iwl_dbgfs_inject_beacon_ie(struct iwl_mvm *mvm, char *bin, int len)
 	struct sk_buff *beacon;
 	struct ieee80211_tx_info *info;
 	struct iwl_mac_beacon_cmd beacon_cmd = {};
+	unsigned int link_id;
 	u8 rate;
 	int i;
 
@@ -1436,17 +1437,20 @@ static int _iwl_dbgfs_inject_beacon_ie(struct iwl_mvm *mvm, char *bin, int len)
 	info = IEEE80211_SKB_CB(beacon);
 	rate = iwl_mvm_mac_ctxt_get_beacon_rate(mvm, info, vif);
 
-	beacon_cmd.flags =
-		cpu_to_le16(iwl_mvm_mac_ctxt_get_beacon_flags(mvm->fw, rate));
-	beacon_cmd.byte_cnt = cpu_to_le16((u16)beacon->len);
-	beacon_cmd.template_id = cpu_to_le32((u32)mvmvif->id);
+	for_each_mvm_vif_valid_link(mvmvif, link_id) {
+		beacon_cmd.flags =
+			cpu_to_le16(iwl_mvm_mac_ctxt_get_beacon_flags(mvm->fw,
+								      rate));
+		beacon_cmd.byte_cnt = cpu_to_le16((u16)beacon->len);
+		beacon_cmd.link_id = cpu_to_le32((u32)mvmvif->id);
 
-	iwl_mvm_mac_ctxt_set_tim(mvm, &beacon_cmd.tim_idx,
-				 &beacon_cmd.tim_size,
-				 beacon->data, beacon->len);
+		iwl_mvm_mac_ctxt_set_tim(mvm, &beacon_cmd.tim_idx,
+					 &beacon_cmd.tim_size,
+					 beacon->data, beacon->len);
 
-	iwl_mvm_mac_ctxt_send_beacon_cmd(mvm, beacon, &beacon_cmd,
-					 sizeof(beacon_cmd));
+		iwl_mvm_mac_ctxt_send_beacon_cmd(mvm, beacon, &beacon_cmd,
+						 sizeof(beacon_cmd));
+	}
 	mutex_unlock(&mvm->mutex);
 
 	dev_kfree_skb(beacon);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 10dc1fdbdeb3..a06513989469 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -1039,7 +1039,8 @@ static int iwl_mvm_mac_ctxt_send_beacon_v7(struct iwl_mvm *mvm,
 
 static int iwl_mvm_mac_ctxt_send_beacon_v9(struct iwl_mvm *mvm,
 					   struct ieee80211_vif *vif,
-					   struct sk_buff *beacon)
+					   struct sk_buff *beacon,
+					   struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(beacon);
@@ -1052,7 +1053,7 @@ static int iwl_mvm_mac_ctxt_send_beacon_v9(struct iwl_mvm *mvm,
 
 	/* Enable FILS on PSC channels only */
 	rcu_read_lock();
-	ctx = rcu_dereference(vif->bss_conf.chanctx_conf);
+	ctx = rcu_dereference(link_conf->chanctx_conf);
 	channel = ieee80211_frequency_to_channel(ctx->def.chan->center_freq);
 	WARN_ON(channel == 0);
 	if (cfg80211_channel_is_psc(ctx->def.chan) &&
@@ -1069,7 +1070,7 @@ static int iwl_mvm_mac_ctxt_send_beacon_v9(struct iwl_mvm *mvm,
 
 	beacon_cmd.flags = cpu_to_le16(flags);
 	beacon_cmd.byte_cnt = cpu_to_le16((u16)beacon->len);
-	beacon_cmd.template_id = cpu_to_le32((u32)mvmvif->id);
+	beacon_cmd.link_id = cpu_to_le32((u32)mvmvif->id);
 
 	if (vif->type == NL80211_IFTYPE_AP)
 		iwl_mvm_mac_ctxt_set_tim(mvm, &beacon_cmd.tim_idx,
@@ -1091,7 +1092,8 @@ static int iwl_mvm_mac_ctxt_send_beacon_v9(struct iwl_mvm *mvm,
 
 int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif,
-				 struct sk_buff *beacon)
+				 struct sk_buff *beacon,
+				 struct ieee80211_bss_conf *link_conf)
 {
 	if (WARN_ON(!beacon))
 		return -EINVAL;
@@ -1105,14 +1107,16 @@ int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm,
 
 	if (fw_has_api(&mvm->fw->ucode_capa,
 		       IWL_UCODE_TLV_API_NEW_BEACON_TEMPLATE))
-		return iwl_mvm_mac_ctxt_send_beacon_v9(mvm, vif, beacon);
+		return iwl_mvm_mac_ctxt_send_beacon_v9(mvm, vif, beacon,
+						       link_conf);
 
 	return iwl_mvm_mac_ctxt_send_beacon_v7(mvm, vif, beacon);
 }
 
 /* The beacon template for the AP/GO/IBSS has changed and needs update */
 int iwl_mvm_mac_ctxt_beacon_changed(struct iwl_mvm *mvm,
-				    struct ieee80211_vif *vif)
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_bss_conf *link_conf)
 {
 	struct sk_buff *beacon;
 	int ret;
@@ -1120,7 +1124,8 @@ int iwl_mvm_mac_ctxt_beacon_changed(struct iwl_mvm *mvm,
 	WARN_ON(vif->type != NL80211_IFTYPE_AP &&
 		vif->type != NL80211_IFTYPE_ADHOC);
 
-	beacon = ieee80211_beacon_get_template(mvm->hw, vif, NULL, 0);
+	beacon = ieee80211_beacon_get_template(mvm->hw, vif, NULL,
+					       link_conf->link_id);
 	if (!beacon)
 		return -ENOMEM;
 
@@ -1131,7 +1136,7 @@ int iwl_mvm_mac_ctxt_beacon_changed(struct iwl_mvm *mvm,
 	}
 #endif
 
-	ret = iwl_mvm_mac_ctxt_send_beacon(mvm, vif, beacon);
+	ret = iwl_mvm_mac_ctxt_send_beacon(mvm, vif, beacon, link_conf);
 	dev_kfree_skb(beacon);
 	return ret;
 }
@@ -1398,7 +1403,8 @@ static void iwl_mvm_csa_count_down(struct iwl_mvm *mvm,
 	if (!ieee80211_beacon_cntdwn_is_complete(csa_vif)) {
 		int c = ieee80211_beacon_update_cntdwn(csa_vif);
 
-		iwl_mvm_mac_ctxt_beacon_changed(mvm, csa_vif);
+		iwl_mvm_mac_ctxt_beacon_changed(mvm, csa_vif,
+						&csa_vif->bss_conf);
 		if (csa_vif->p2p &&
 		    !iwl_mvm_te_scheduled(&mvmvif->time_event_data) && gp2 &&
 		    tx_success) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index e922cc45c390..d9caf326535f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2748,11 +2748,6 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
-	/* Send the beacon template */
-	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif);
-	if (ret)
-		goto out_unlock;
-
 	/*
 	 * Re-calculate the tsf id, as the leader-follower relations depend on
 	 * the beacon interval, which was not known when the AP interface
@@ -2761,10 +2756,31 @@ static int iwl_mvm_start_ap_ibss(struct ieee80211_hw *hw,
 	if (vif->type == NL80211_IFTYPE_AP)
 		iwl_mvm_mac_ctxt_recalc_tsf_id(mvm, vif);
 
-	/* Add the mac context */
-	ret = iwl_mvm_mac_ctxt_add(mvm, vif);
-	if (ret)
-		goto out_unlock;
+	/* For older devices need to send beacon template before adding mac
+	 * context. For the newer, the beacon is a resource that belongs to a
+	 * MAC, so need to send beacon template after adding the mac.
+	 */
+	if (mvm->trans->trans_cfg->device_family > IWL_DEVICE_FAMILY_22000) {
+		/* Add the mac context */
+		ret = iwl_mvm_mac_ctxt_add(mvm, vif);
+		if (ret)
+			goto out_unlock;
+
+		/* Send the beacon template */
+		ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf);
+		if (ret)
+			goto out_unlock;
+	} else {
+		/* Send the beacon template */
+		ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf);
+		if (ret)
+			goto out_unlock;
+
+		/* Add the mac context */
+		ret = iwl_mvm_mac_ctxt_add(mvm, vif);
+		if (ret)
+			goto out_unlock;
+	}
 
 	/* Perform the binding */
 	ret = iwl_mvm_binding_add_vif(mvm, vif);
@@ -2961,7 +2977,7 @@ iwl_mvm_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 
 	/* Need to send a new beacon template to the FW */
 	if (changes & BSS_CHANGED_BEACON &&
-	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif))
+	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, &vif->bss_conf))
 		IWL_WARN(mvm, "Failed updating beacon data\n");
 
 	if (changes & BSS_CHANGED_FTM_RESPONDER) {
@@ -4965,7 +4981,8 @@ int iwl_mvm_set_tim(struct ieee80211_hw *hw, struct ieee80211_sta *sta,
 		return -EINVAL;
 	}
 
-	return iwl_mvm_mac_ctxt_beacon_changed(mvm, mvm_sta->vif);
+	return iwl_mvm_mac_ctxt_beacon_changed(mvm, mvm_sta->vif,
+					       &mvm_sta->vif->bss_conf);
 }
 
 #ifdef CONFIG_NL80211_TESTMODE
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
index 240f51d2fd53..c9a0f68c5d48 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
@@ -10,6 +10,8 @@ static void iwl_mvm_mld_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 					    u32 action)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct ieee80211_bss_conf *link_conf;
+	unsigned int link_id;
 
 	cmd->id_and_color = cpu_to_le32(mvmvif->id);
 	cmd->action = cpu_to_le32(action);
@@ -18,19 +20,53 @@ static void iwl_mvm_mld_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 
 	memcpy(cmd->local_mld_addr, vif->addr, ETH_ALEN);
 
-	cmd->filter_flags = cpu_to_le32(0);
-	cmd->he_support = cpu_to_le32(0);
-	cmd->eht_support = cpu_to_le32(0);
+	cmd->he_support = 0;
+	cmd->eht_support = 0;
+
+	/* should be set by specific context type handler */
+	cmd->filter_flags = 0;
 
 	cmd->nic_not_ack_enabled =
 		cpu_to_le32(!iwl_mvm_is_nic_ack_enabled(mvm, vif));
 
 	if (iwlwifi_mod_params.disable_11ax)
 		return;
-	cmd->he_support = cpu_to_le32(vif->bss_conf.he_support);
 
-	if (!iwlwifi_mod_params.disable_11be && cmd->he_support)
-		cmd->eht_support = cpu_to_le32(vif->bss_conf.eht_support);
+	/* If we have MLO enabled, then the firmware needs to enable
+	 * address translation for the station(s) we add. That depends
+	 * on having EHT enabled in firmware, which in turn depends on
+	 * mac80211 in the code below.
+	 * However, mac80211 doesn't enable HE/EHT until it has parsed
+	 * the association response successfully, so just skip all that
+	 * and enable both when we have MLO.
+	 */
+	if (vif->valid_links) {
+		cmd->he_support = cpu_to_le32(1);
+		cmd->eht_support = cpu_to_le32(1);
+		return;
+	}
+
+	rcu_read_lock();
+	for (link_id = 0; link_id < ARRAY_SIZE((vif)->link_conf); link_id++) {
+		link_conf = rcu_dereference(vif->link_conf[link_id]);
+		if (!link_conf)
+			continue;
+
+		if (link_conf->he_support)
+			cmd->he_support = cpu_to_le32(1);
+
+		/* it's not reasonable to have EHT without HE and FW API doesn't
+		 * support it. Ignore EHT in this case.
+		 */
+		if (!link_conf->he_support && link_conf->eht_support)
+			continue;
+
+		if (link_conf->eht_support) {
+			cmd->eht_support = cpu_to_le32(1);
+			break;
+		}
+	}
+	rcu_read_unlock();
 }
 
 static int iwl_mvm_mld_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
@@ -66,6 +102,7 @@ static int iwl_mvm_mld_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 		cmd.client.ctwin =
 			iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(mvm, vif);
 
+	/* FIXME: assume for now that DTIM period is the same for all links */
 	if (vif->cfg.assoc && vif->bss_conf.dtim_period &&
 	    !force_assoc_off) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 5be8058ca669..ec0789484d8b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -258,11 +258,6 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
-	/* Send the beacon template */
-	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif);
-	if (ret)
-		goto out_unlock;
-
 	/* No need to re-calculate the tsf_is, as it was offloaded */
 
 	/* Add the mac context */
@@ -270,6 +265,11 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	if (ret)
 		goto out_unlock;
 
+	/* Send the beacon template */
+	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf);
+	if (ret)
+		goto out_unlock;
+
 	/* Add link and activate it */
 	ret = iwl_mvm_add_link(mvm, vif);
 	if (ret)
@@ -532,7 +532,7 @@ iwl_mvm_mld_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 
 	/* Need to send a new beacon template to the FW */
 	if (changes & BSS_CHANGED_BEACON &&
-	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif))
+	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, &vif->bss_conf))
 		IWL_WARN(mvm, "Failed updating beacon data\n");
 
 	if (changes & BSS_CHANGED_FTM_RESPONDER) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 6f4ed5d1efb9..9c5e772d619e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1759,10 +1759,12 @@ int iwl_mvm_mac_ctxt_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			     bool force_assoc_off, const u8 *bssid_override);
 int iwl_mvm_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mac_ctxt_beacon_changed(struct iwl_mvm *mvm,
-				    struct ieee80211_vif *vif);
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif,
-				 struct sk_buff *beacon);
+				 struct sk_buff *beacon,
+				 struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_mac_ctxt_send_beacon_cmd(struct iwl_mvm *mvm,
 				     struct sk_buff *beacon,
 				     void *data, int len);
-- 
cgit v1.2.3-70-g09d2


From f551d013bd3111413a1c57214796ab3fe3b2f694 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:58:58 +0300
Subject: wifi: iwlwifi: mvm: adjust some PS and PM methods to MLD

When using mvm vif PS/PM related properties, consider all links.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.8d9e18f7d74b.I048fc17405ecdddc1f6fe72859013cc4878b2a3b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  4 +--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  6 +++++
 drivers/net/wireless/intel/iwlwifi/mvm/power.c    | 16 ++++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/utils.c    | 33 +++++++++++++++++++++++
 4 files changed, 50 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d9caf326535f..2989285d451a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1278,8 +1278,8 @@ struct iwl_mvm_phy_ctxt *iwl_mvm_get_free_phy_ctxt(struct iwl_mvm *mvm)
 	return NULL;
 }
 
-static int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-				s16 tx_power)
+int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 s16 tx_power)
 {
 	u32 cmd_id = REDUCE_TX_POWER_CMD;
 	int len;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 9c5e772d619e..ae9c165197e8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2599,4 +2599,10 @@ int iwl_mvm_start_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       struct cfg80211_pmsr_request *request);
 void iwl_mvm_abort_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			struct cfg80211_pmsr_request *request);
+
+bool iwl_mvm_have_links_same_channel(struct iwl_mvm_vif *vif1,
+				     struct iwl_mvm_vif *vif2);
+bool iwl_mvm_vif_is_active(struct iwl_mvm_vif *mvmvif);
+int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 s16 tx_power);
 #endif /* __IWL_MVM_H__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 6e2e3127843f..45a652cce491 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -553,8 +553,7 @@ static void iwl_mvm_power_ps_disabled_iterator(void *_data, u8* mac,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	bool *disable_ps = _data;
 
-	if (mvmvif->deflink.phy_ctxt &&
-	    mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX)
+	if (iwl_mvm_vif_is_active(mvmvif))
 		*disable_ps |= mvmvif->ps_disabled;
 }
 
@@ -563,11 +562,13 @@ static void iwl_mvm_power_get_vifs_iterator(void *_data, u8 *mac,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_power_vifs *power_iterator = _data;
-	bool active = mvmvif->deflink.phy_ctxt && mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX;
+	bool active;
 
 	if (!mvmvif->uploaded)
 		return;
 
+	active = iwl_mvm_vif_is_active(mvmvif);
+
 	switch (ieee80211_vif_type_p2p(vif)) {
 	case NL80211_IFTYPE_P2P_DEVICE:
 		break;
@@ -651,11 +652,12 @@ static void iwl_mvm_power_set_pm(struct iwl_mvm *mvm,
 	}
 
 	if (vifs->bss_active && vifs->p2p_active)
-		client_same_channel = (bss_mvmvif->deflink.phy_ctxt->id ==
-				       p2p_mvmvif->deflink.phy_ctxt->id);
+		client_same_channel =
+			iwl_mvm_have_links_same_channel(bss_mvmvif, p2p_mvmvif);
+
 	if (vifs->bss_active && vifs->ap_active)
-		ap_same_channel = (bss_mvmvif->deflink.phy_ctxt->id ==
-				   ap_mvmvif->deflink.phy_ctxt->id);
+		ap_same_channel =
+			iwl_mvm_have_links_same_channel(bss_mvmvif, ap_mvmvif);
 
 	/* clients are not stand alone: enable PM if DCM */
 	if (!(client_same_channel || ap_same_channel)) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index c1f6dd8f7d10..026856c8f5f2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -1138,3 +1138,36 @@ void iwl_mvm_get_sync_time(struct iwl_mvm *mvm, int clock_type,
 		iwl_mvm_power_update_device(mvm);
 	}
 }
+
+/* Find if at least two links from different vifs use same channel
+ * FIXME: consider having a refcount array in struct iwl_mvm_vif for
+ * used phy_ctxt ids.
+ */
+bool iwl_mvm_have_links_same_channel(struct iwl_mvm_vif *vif1,
+				     struct iwl_mvm_vif *vif2)
+{
+	unsigned int i, j;
+
+	for_each_mvm_vif_valid_link(vif1, i) {
+		for_each_mvm_vif_valid_link(vif2, j) {
+			if (vif1->link[i]->phy_ctxt == vif2->link[j]->phy_ctxt)
+				return true;
+		}
+	}
+
+	return false;
+}
+
+bool iwl_mvm_vif_is_active(struct iwl_mvm_vif *mvmvif)
+{
+	unsigned int i;
+
+	/* FIXME: can it fail when phy_ctxt is assigned? */
+	for_each_mvm_vif_valid_link(mvmvif, i) {
+		if (mvmvif->link[i]->phy_ctxt &&
+		    mvmvif->link[i]->phy_ctxt->id < NUM_PHY_CTX)
+			return true;
+	}
+
+	return false;
+}
-- 
cgit v1.2.3-70-g09d2


From cb145863e7abbbf4248ab11d2783a1b619349d1b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 28 Mar 2023 10:58:59 +0300
Subject: wifi: iwlwifi: mvm: add link to firmware earlier

The firmware now allows adding a link that's not yet bound
to a PHY context. Make use of that to align the driver with
mac80211's API expectations. For now, just add the link at
the same time as the MAC since we don't yet have real MLD
support, but that'll obviously change later. This fixes an
issue with apStaId tracking in the firmware.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.097e5008b637.I4e75c6c11e21c08d28ff6a066be36629d3975db6@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      |  21 ++-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  17 +-
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 173 +++++++++++++--------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   7 +-
 4 files changed, 133 insertions(+), 85 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 4905dac3a048..d8902e0ed2ae 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -23,11 +23,9 @@ static int iwl_mvm_link_cmd_send(struct iwl_mvm *mvm,
 int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_phy_ctxt *phyctxt = mvmvif->deflink.phy_ctxt;
 	struct iwl_link_config_cmd cmd = {};
 
-	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
-		return -EINVAL;
-
 	/* Update SF - Disable if needed. if this fails, SF might still be on
 	 * while many macs are bound, which is forbidden - so fail the binding.
 	 */
@@ -36,7 +34,11 @@ int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	cmd.link_id = cpu_to_le32(mvmvif->id);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
-	cmd.phy_id = cpu_to_le32(mvmvif->deflink.phy_ctxt->id);
+	/* P2P-Device already has a valid PHY context during add */
+	if (phyctxt)
+		cmd.phy_id = cpu_to_le32(phyctxt->id);
+	else
+		cmd.phy_id = cpu_to_le32(FW_CTXT_INVALID);
 
 	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
 
@@ -54,15 +56,15 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	struct iwl_link_config_cmd cmd = {};
 	u32 ht_flag, flags = 0, flags_mask = 0;
 
-	if (!phyctxt)
-		return -EINVAL;
-
 	cmd.link_id = cpu_to_le32(mvmvif->id);
 
 	/* The phy_id, link address and listen_lmac can be modified only until
 	 * the link becomes active, otherwise they will be ignored.
 	 */
-	cmd.phy_id = cpu_to_le32(phyctxt->id);
+	if (phyctxt)
+		cmd.phy_id = cpu_to_le32(phyctxt->id);
+	else
+		cmd.phy_id = cpu_to_le32(FW_CTXT_INVALID);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
 
 	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
@@ -157,9 +159,6 @@ int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	struct iwl_link_config_cmd cmd = {};
 	int ret;
 
-	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
-		return -EINVAL;
-
 	cmd.link_id = cpu_to_le32(mvmvif->id);
 	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_REMOVE);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 2989285d451a..18e7ba98daa0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1437,10 +1437,13 @@ iwl_mvm_chandef_get_primary_80(struct cfg80211_chan_def *chandef)
 /*
  * Returns true if addding the interface is done
  * (either with success or failure)
+ *
+ * FIXME: remove this again and merge it in
  */
-bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
-				      struct ieee80211_hw *hw,
-				      struct ieee80211_vif *vif, int *ret)
+static bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
+					     struct ieee80211_hw *hw,
+					     struct ieee80211_vif *vif,
+					     int *ret)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
@@ -1621,8 +1624,8 @@ out:
 	return ret;
 }
 
-static void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
-					struct ieee80211_vif *vif)
+void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif)
 {
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 		/*
@@ -1638,8 +1641,8 @@ static void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
  * both - MLD and non-MLD modes. Returns true if removing the interface
  * is done
  */
-bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
-					 struct ieee80211_vif *vif)
+static bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
+						struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index ec0789484d8b..1d2da26408e6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -13,17 +13,34 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 
 	mutex_lock(&mvm->mutex);
 
-	/* Common for MLD and non-MLD API */
-	if (iwl_mvm_mac_add_interface_common(mvm, hw, vif, &ret))
-		goto out_unlock;
+	mvmvif->mvm = mvm;
+	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
 
-	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
+	/* Not much to do here. The stack will not allow interface
+	 * types or combinations that we didn't advertise, so we
+	 * don't really have to check the types.
+	 */
+
+	/* make sure that beacon statistics don't go backwards with FW reset */
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+		mvmvif->deflink.beacon_stats.accu_num_beacons +=
+			mvmvif->deflink.beacon_stats.num_beacons;
+
+	/* Allocate resources for the MAC context, and add it to the fw  */
+	ret = iwl_mvm_mac_ctxt_init(mvm, vif);
 	if (ret)
 		goto out_unlock;
 
-	ret = iwl_mvm_power_update_mac(mvm);
+	rcu_assign_pointer(mvm->vif_id_to_mac[mvmvif->id], vif);
+
+	mvmvif->features |= hw->netdev_features;
+
+	/* the first link always points to the default one */
+	mvmvif->link[0] = &mvmvif->deflink;
+
+	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
 	if (ret)
-		goto out_remove_mac;
+		goto out_unlock;
 
 	/* beacon filtering */
 	ret = iwl_mvm_disable_beacon_filter(mvm, vif, 0);
@@ -67,8 +84,16 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 		 * update the p2p device MAC when a GO is started/stopped
 		 */
 		mvm->p2p_device_vif = vif;
+	} else {
+		ret = iwl_mvm_add_link(mvm, vif);
+		if (ret)
+			goto out_free_bf;
 	}
 
+	ret = iwl_mvm_power_update_mac(mvm);
+	if (ret)
+		goto out_free_bf;
+
 	iwl_mvm_tcm_add_vif(mvm, vif);
 	INIT_DELAYED_WORK(&mvmvif->csa_work,
 			  iwl_mvm_channel_switch_disconnect_wk);
@@ -117,9 +142,52 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_probe_resp_data *probe_data;
 
-	if (iwl_mvm_mac_remove_interface_common(hw, vif))
-		goto out;
+	iwl_mvm_prepare_mac_removal(mvm, vif);
+
+	if (!(vif->type == NL80211_IFTYPE_AP ||
+	      vif->type == NL80211_IFTYPE_ADHOC))
+		iwl_mvm_tcm_rm_vif(mvm, vif);
+
+	mutex_lock(&mvm->mutex);
+
+	if (vif == mvm->csme_vif) {
+		iwl_mei_set_netdev(NULL);
+		mvm->csme_vif = NULL;
+	}
+
+	probe_data = rcu_dereference_protected(mvmvif->deflink.probe_resp_data,
+					       lockdep_is_held(&mvm->mutex));
+	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
+	if (probe_data)
+		kfree_rcu(probe_data, rcu_head);
+
+	if (mvm->bf_allowed_vif == mvmvif) {
+		mvm->bf_allowed_vif = NULL;
+		vif->driver_flags &= ~(IEEE80211_VIF_BEACON_FILTER |
+				       IEEE80211_VIF_SUPPORTS_CQM_RSSI);
+	}
+
+	if (vif->bss_conf.ftm_responder)
+		memset(&mvm->ftm_resp_stats, 0, sizeof(mvm->ftm_resp_stats));
+
+	iwl_mvm_vif_dbgfs_clean(mvm, vif);
+
+	/* For AP/GO interface, the tear down of the resources allocated to the
+	 * interface is be handled as part of the stop_ap flow.
+	 */
+	if (vif->type == NL80211_IFTYPE_AP ||
+	    vif->type == NL80211_IFTYPE_ADHOC) {
+#ifdef CONFIG_NL80211_TESTMODE
+		if (vif == mvm->noa_vif) {
+			mvm->noa_vif = NULL;
+			mvm->noa_duration = 0;
+		}
+#endif
+	}
+
+	iwl_mvm_power_update_mac(mvm);
 
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 		mvm->p2p_device_vif = NULL;
@@ -130,6 +198,8 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 		iwl_mvm_remove_link(mvm, vif);
 		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
 		mvmvif->deflink.phy_ctxt = NULL;
+	} else {
+		iwl_mvm_remove_link(mvm, vif);
 	}
 
 	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
@@ -141,7 +211,6 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 		__clear_bit(IEEE80211_HW_RX_INCLUDES_FCS, mvm->hw->flags);
 	}
 
-out:
 	mutex_unlock(&mvm->mutex);
 }
 
@@ -150,28 +219,30 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 					    struct ieee80211_chanctx_conf *ctx,
 					    bool switching_chanctx)
 {
+	u16 *phy_ctxt_id = (u16 *)ctx->drv_priv;
+	struct iwl_mvm_phy_ctxt *phy_ctxt = &mvm->phy_ctxts[*phy_ctxt_id];
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
 
-	if (__iwl_mvm_assign_vif_chanctx_common(mvm, vif, ctx,
-						switching_chanctx, &ret))
-		goto out;
+	mvmvif->deflink.phy_ctxt = phy_ctxt;
 
-	if (!switching_chanctx) {
-		ret = iwl_mvm_add_link(mvm, vif);
-		if (ret)
-			goto out;
-	} else {
-		ret = iwl_mvm_link_changed(mvm, vif, 0, false);
-		if (ret)
-			goto out_remove_link;
+	if (switching_chanctx) {
+		/* reactivate if we turned this off during channel switch */
+		if (vif->type == NL80211_IFTYPE_AP)
+			mvmvif->ap_ibss_active = true;
 	}
 
+	/* send it first with phy context ID */
+	ret = iwl_mvm_link_changed(mvm, vif, 0, false);
+	if (ret)
+		goto out;
+
+	/* then activate */
 	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE |
 				   LINK_CONTEXT_MODIFY_RATES_INFO,
 				   true);
 	if (ret)
-		goto out_remove_link;
+		goto out;
 
 	/*
 	 * Power state must be updated before quotas,
@@ -182,19 +253,16 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	if (vif->type == NL80211_IFTYPE_MONITOR) {
 		ret = iwl_mvm_mld_add_snif_sta(mvm, vif);
 		if (ret)
-			goto out_remove_link;
+			goto deactivate;
 	}
 
-	goto out;
+	return 0;
 
-out_remove_link:
-	/* Link needs to be deactivated before removal */
+deactivate:
 	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
-	iwl_mvm_remove_link(mvm, vif);
-	iwl_mvm_power_update_mac(mvm);
 out:
-	if (ret)
-		mvmvif->deflink.phy_ctxt = NULL;
+	mvmvif->deflink.phy_ctxt = NULL;
+	iwl_mvm_power_update_mac(mvm);
 	return ret;
 }
 
@@ -220,16 +288,23 @@ static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-	if (__iwl_mvm_unassign_vif_chanctx_common(mvm, vif, switching_chanctx))
-		goto out;
+	if (vif->type == NL80211_IFTYPE_AP && switching_chanctx) {
+		mvmvif->csa_countdown = false;
+
+		/* Set CS bit on all the stations */
+		iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, true);
+
+		/* Save blocked iface, the timeout is set on the next beacon */
+		rcu_assign_pointer(mvm->csa_tx_blocked_vif, vif);
+
+		mvmvif->ap_ibss_active = false;
+	}
 
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		iwl_mvm_mld_rm_snif_sta(mvm, vif);
 
 	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
-	if (!switching_chanctx)
-		iwl_mvm_remove_link(mvm, vif);
-out:
+
 	if (switching_chanctx)
 		return;
 	mvmvif->deflink.phy_ctxt = NULL;
@@ -257,32 +332,19 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-
-	/* No need to re-calculate the tsf_is, as it was offloaded */
-
-	/* Add the mac context */
-	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
-	if (ret)
-		goto out_unlock;
-
 	/* Send the beacon template */
 	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf);
 	if (ret)
 		goto out_unlock;
 
-	/* Add link and activate it */
-	ret = iwl_mvm_add_link(mvm, vif);
-	if (ret)
-		goto out_remove_mac;
-
 	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ALL,
 				   true);
 	if (ret)
-		goto out_remove_link;
+		goto out_unlock;
 
 	ret = iwl_mvm_mld_add_mcast_sta(mvm, vif);
 	if (ret)
-		goto out_remove_link;
+		goto out_unlock;
 
 	/* Send the bcast station. At this stage the TBTT and DTIM time
 	 * events are added and applied to the scheduler
@@ -314,12 +376,6 @@ out_failed:
 	iwl_mvm_mld_rm_bcast_sta(mvm, vif);
 out_rm_mcast:
 	iwl_mvm_mld_rm_mcast_sta(mvm, vif);
-out_remove_link:
-	/* Link needs to be deactivated before removal */
-	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
-	iwl_mvm_remove_link(mvm, vif);
-out_remove_mac:
-	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
 out_unlock:
 	mutex_unlock(&mvm->mutex);
 	return ret;
@@ -357,14 +413,7 @@ static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
 	iwl_mvm_mld_rm_bcast_sta(mvm, vif);
 	iwl_mvm_mld_rm_mcast_sta(mvm, vif);
 
-	/* Link needs to be deactivated before removal */
-	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
-	iwl_mvm_remove_link(mvm, vif);
-
 	iwl_mvm_power_update_mac(mvm);
-
-	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
-
 	mutex_unlock(&mvm->mutex);
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ae9c165197e8..af2059155794 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1718,11 +1718,8 @@ u8 iwl_mvm_get_ctrl_pos(struct cfg80211_chan_def *chandef);
 
 /* MAC (virtual interface) programming */
 
-bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
-				      struct ieee80211_hw *hw,
-				      struct ieee80211_vif *vif, int *ret);
-bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
-					 struct ieee80211_vif *vif);
+void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif);
 void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				__le32 *cck_rates, __le32 *ofdm_rates);
 void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From be8897e24415f2c23b05711c68d32392ba8541f7 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Tue, 28 Mar 2023 10:59:00 +0300
Subject: wifi: iwlwifi: mvm: add set_hw_timestamp to mld ops

In mld ops, set_hw_timestamp was not set. Fix it.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.7b24a582b763.Ieb3139d41f1a01b2b566adecd32450da1bcdd5fa@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c     | 6 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h          | 3 +++
 3 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 18e7ba98daa0..4dbc9a57746b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -5896,9 +5896,9 @@ static bool iwl_mvm_mac_can_aggregate(struct ieee80211_hw *hw,
 	return iwl_mvm_can_hw_csum(skb) == iwl_mvm_can_hw_csum(head);
 }
 
-static int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
-				    struct ieee80211_vif *vif,
-				    struct cfg80211_set_hw_timestamp *hwts)
+int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct cfg80211_set_hw_timestamp *hwts)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	u32 protocols = 0;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 1d2da26408e6..15beed41b305 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -798,4 +798,5 @@ const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	.sta_add_debugfs = iwl_mvm_sta_add_debugfs,
 #endif
+	.set_hw_timestamp = iwl_mvm_set_hw_timestamp,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index af2059155794..9b04ae9c41ff 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2602,4 +2602,7 @@ bool iwl_mvm_have_links_same_channel(struct iwl_mvm_vif *vif1,
 bool iwl_mvm_vif_is_active(struct iwl_mvm_vif *mvmvif);
 int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			 s16 tx_power);
+int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct cfg80211_set_hw_timestamp *hwts);
 #endif /* __IWL_MVM_H__ */
-- 
cgit v1.2.3-70-g09d2


From 1a3e70391e55d12010073935e383629ab856528a Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:01 +0300
Subject: wifi: iwlwifi: mvm: adjust SMPS for MLO

Configure SMPS per-link. Add link_id parameter to
iwl_mvm_update_smps() and refactor iwl_mvm_intf_dual_chain_req()
since it has to handle all active links.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.da6a19db562c.Ic88b02338c8973f2934439ac3ee77c7451bc0054@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/coex.c      |  4 +--
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 19 ++++++++------
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  4 +--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 19 ++++++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       | 23 +++++++++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/tt.c        |  4 +--
 drivers/net/wireless/intel/iwlwifi/mvm/utils.c     | 29 ++++++++++++++++++----
 7 files changed, 73 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
index 21ba3c7dd9b4..daaa173aaed6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
@@ -291,7 +291,7 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 		if (vif->type == NL80211_IFTYPE_STATION) {
 			/* ... relax constraints and disable rssi events */
 			iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
-					    smps_mode);
+					    smps_mode, 0);
 			iwl_mvm_bt_coex_reduced_txp(mvm,
 						    mvmvif->deflink.ap_sta_id,
 						    false);
@@ -325,7 +325,7 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 
 	if (vif->type == NL80211_IFTYPE_STATION)
 		iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
-				    smps_mode);
+				    smps_mode, 0);
 
 	/* low latency is always primary */
 	if (iwl_mvm_vif_low_latency(mvmvif)) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 4dbc9a57746b..ef3eb57b806a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2481,17 +2481,19 @@ void iwl_mvm_bss_info_changed_station_assoc(struct iwl_mvm *mvm,
 	mvmvif->bf_data.ave_beacon_signal = 0;
 
 	iwl_mvm_bt_coex_vif_change(mvm);
-	iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_TT,
-			    IEEE80211_SMPS_AUTOMATIC);
+	iwl_mvm_update_smps_on_active_links(mvm, vif, IWL_MVM_SMPS_REQ_TT,
+					    IEEE80211_SMPS_AUTOMATIC);
 	if (fw_has_capa(&mvm->fw->ucode_capa,
 			IWL_UCODE_TLV_CAPA_UMAC_SCAN))
 		iwl_mvm_config_scan(mvm);
 }
 
 /* Execute the common part for MLD and non-MLD modes */
-void iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
-					     struct ieee80211_vif *vif,
-					     u64 changes)
+void
+iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					struct ieee80211_bss_conf *link_conf,
+					u64 changes)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
@@ -2529,7 +2531,7 @@ void iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
 	}
 
 	if (changes & BSS_CHANGED_BANDWIDTH)
-		iwl_mvm_apply_fw_smps_request(vif);
+		iwl_mvm_update_link_smps(vif, link_conf);
 }
 
 static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
@@ -2641,7 +2643,7 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 			if (vif->p2p) {
 				iwl_mvm_update_smps(mvm, vif,
 						    IWL_MVM_SMPS_REQ_PROT,
-						    IEEE80211_SMPS_DYNAMIC);
+						    IEEE80211_SMPS_DYNAMIC, 0);
 			}
 		} else if (mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
 			iwl_mvm_mei_host_disassociated(mvm);
@@ -2697,7 +2699,8 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 		iwl_mvm_bss_info_changed_station_assoc(mvm, vif, changes);
 	}
 
-	iwl_mvm_bss_info_changed_station_common(mvm, vif, changes);
+	iwl_mvm_bss_info_changed_station_common(mvm, vif, &vif->bss_conf,
+						changes);
 }
 
 bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 15beed41b305..80f1a1ccf16d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -516,7 +516,7 @@ iwl_mvm_mld_bss_info_changed_station(struct iwl_mvm *mvm,
 			if (vif->p2p) {
 				iwl_mvm_update_smps(mvm, vif,
 						    IWL_MVM_SMPS_REQ_PROT,
-						    IEEE80211_SMPS_DYNAMIC);
+						    IEEE80211_SMPS_DYNAMIC, 0);
 			}
 		} else if (mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
 			iwl_mvm_mei_host_disassociated(mvm);
@@ -557,7 +557,7 @@ iwl_mvm_mld_bss_info_changed_station(struct iwl_mvm *mvm,
 		iwl_mvm_bss_info_changed_station_assoc(mvm, vif, changes);
 	}
 
-	iwl_mvm_bss_info_changed_station_common(mvm, vif, changes);
+	iwl_mvm_bss_info_changed_station_common(mvm, vif, &vif->bss_conf, changes);
 }
 
 static void
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 9b04ae9c41ff..d62341269744 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1841,9 +1841,11 @@ iwl_mvm_bss_info_changed_common(struct ieee80211_hw *hw,
 				struct ieee80211_bss_conf *bss_conf,
 				struct iwl_mvm_bss_info_changed_ops *callbacks,
 				u64 changes);
-void iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
-					     struct ieee80211_vif *vif,
-					     u64 changes);
+void
+iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					struct ieee80211_bss_conf *link_conf,
+					u64 changes);
 void iwl_mvm_bss_info_changed_station_assoc(struct iwl_mvm *mvm,
 					    struct ieee80211_vif *vif,
 					    u64 changes);
@@ -2057,10 +2059,17 @@ int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 /* SMPS */
 void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				enum iwl_mvm_smps_type_request req_type,
-				enum ieee80211_smps_mode smps_request);
+				enum ieee80211_smps_mode smps_request,
+				unsigned int link_id);
+void
+iwl_mvm_update_smps_on_active_links(struct iwl_mvm *mvm,
+				    struct ieee80211_vif *vif,
+				    enum iwl_mvm_smps_type_request req_type,
+				    enum ieee80211_smps_mode smps_request);
 bool iwl_mvm_rx_diversity_allowed(struct iwl_mvm *mvm,
 				  struct iwl_mvm_phy_ctxt *ctxt);
-void iwl_mvm_apply_fw_smps_request(struct ieee80211_vif *vif);
+void iwl_mvm_update_link_smps(struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf);
 
 /* Low latency */
 int iwl_mvm_update_low_latency(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index b355b9ebad0a..4a679ac132c5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -209,24 +209,37 @@ static void iwl_mvm_rx_monitor_notif(struct iwl_mvm *mvm,
 	ieee80211_disconnect(vif, true);
 }
 
-void iwl_mvm_apply_fw_smps_request(struct ieee80211_vif *vif)
+void iwl_mvm_update_link_smps(struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm *mvm = mvmvif->mvm;
 	enum ieee80211_smps_mode mode = IEEE80211_SMPS_AUTOMATIC;
 
+	if (!link_conf)
+		return;
+
 	if (mvm->fw_static_smps_request &&
-	    vif->bss_conf.chandef.width == NL80211_CHAN_WIDTH_160 &&
-	    vif->bss_conf.he_support)
+	    link_conf->chandef.width == NL80211_CHAN_WIDTH_160 &&
+	    link_conf->he_support)
 		mode = IEEE80211_SMPS_STATIC;
 
-	iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_FW, mode);
+	iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_FW, mode,
+			    link_conf->link_id);
 }
 
 static void iwl_mvm_intf_dual_chain_req(void *data, u8 *mac,
 					struct ieee80211_vif *vif)
 {
-	iwl_mvm_apply_fw_smps_request(vif);
+	struct ieee80211_bss_conf *link_conf;
+	unsigned int link_id;
+
+	rcu_read_lock();
+
+	for_each_vif_active_link(vif, link_conf, link_id)
+		iwl_mvm_update_link_smps(vif, link_conf);
+
+	rcu_read_unlock();
 }
 
 static void iwl_mvm_rx_thermal_dual_chain_req(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 232c200af38f..4d0e161a92e0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2012-2014, 2019-2021 Intel Corporation
+ * Copyright (C) 2012-2014, 2019-2022 Intel Corporation
  * Copyright (C) 2013-2014 Intel Mobile Communications GmbH
  * Copyright (C) 2015-2016 Intel Deutschland GmbH
  */
@@ -334,7 +334,7 @@ static void iwl_mvm_tt_smps_iterator(void *_data, u8 *mac,
 	if (vif->type != NL80211_IFTYPE_STATION)
 		return;
 
-	iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_TT, smps_mode);
+	iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_TT, smps_mode, 0);
 }
 
 static void iwl_mvm_tt_tx_protection(struct iwl_mvm *mvm, bool enable)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 026856c8f5f2..cf72b9377411 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -272,13 +272,15 @@ int iwl_mvm_send_lq_cmd(struct iwl_mvm *mvm, struct iwl_lq_cmd *lq)
  * @vif: Pointer to the ieee80211_vif structure
  * @req_type: The part of the driver who call for a change.
  * @smps_request: The request to change the SMPS mode.
+ * @link_id: for MLO link_id, otherwise 0 (deflink)
  *
  * Get a requst to change the SMPS mode,
  * and change it according to all other requests in the driver.
  */
 void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			 enum iwl_mvm_smps_type_request req_type,
-			 enum ieee80211_smps_mode smps_request)
+			 enum ieee80211_smps_mode smps_request,
+			 unsigned int link_id)
 {
 	struct iwl_mvm_vif *mvmvif;
 	enum ieee80211_smps_mode smps_mode = IEEE80211_SMPS_AUTOMATIC;
@@ -294,17 +296,34 @@ void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		return;
 
 	mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	mvmvif->deflink.smps_requests[req_type] = smps_request;
+	mvmvif->link[link_id]->smps_requests[req_type] = smps_request;
 	for (i = 0; i < NUM_IWL_MVM_SMPS_REQ; i++) {
-		if (mvmvif->deflink.smps_requests[i] == IEEE80211_SMPS_STATIC) {
+		if (mvmvif->link[link_id]->smps_requests[i] ==
+		    IEEE80211_SMPS_STATIC) {
 			smps_mode = IEEE80211_SMPS_STATIC;
 			break;
 		}
-		if (mvmvif->deflink.smps_requests[i] == IEEE80211_SMPS_DYNAMIC)
+		if (mvmvif->link[link_id]->smps_requests[i] ==
+		    IEEE80211_SMPS_DYNAMIC)
 			smps_mode = IEEE80211_SMPS_DYNAMIC;
 	}
 
-	ieee80211_request_smps(vif, 0, smps_mode);
+	ieee80211_request_smps(vif, link_id, smps_mode);
+}
+
+void iwl_mvm_update_smps_on_active_links(struct iwl_mvm *mvm,
+					 struct ieee80211_vif *vif,
+					 enum iwl_mvm_smps_type_request req_type,
+					 enum ieee80211_smps_mode smps_request)
+{
+	struct ieee80211_bss_conf *link_conf;
+	unsigned int link_id;
+
+	rcu_read_lock();
+	for_each_vif_active_link(vif, link_conf, link_id)
+		iwl_mvm_update_smps(mvm, vif, req_type, smps_request,
+				    link_id);
+	rcu_read_unlock();
 }
 
 static bool iwl_wait_stats_complete(struct iwl_notif_wait_data *notif_wait,
-- 
cgit v1.2.3-70-g09d2


From ac251da91afa209fc4eba504fc42a504f9d31d5d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 28 Mar 2023 10:59:02 +0300
Subject: wifi: iwlwifi: mvm: don't check dtim_period in new API

In newer firmware APIs the firmware is responsible for tracking
the DTIM period and other beacon timing, so we don't need to
wait with setting associated. In real MLO operation, mac80211
isn't tracking this anyway, and connections wouldn't work.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.02354241fede.Id957bed3851fdf1fe902d79a1b0338c6d80bc0e1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
index c9a0f68c5d48..ab0ba85936b4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
@@ -102,9 +102,7 @@ static int iwl_mvm_mld_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 		cmd.client.ctwin =
 			iwl_mvm_mac_ctxt_cmd_p2p_sta_get_oppps_ctwin(mvm, vif);
 
-	/* FIXME: assume for now that DTIM period is the same for all links */
-	if (vif->cfg.assoc && vif->bss_conf.dtim_period &&
-	    !force_assoc_off) {
+	if (vif->cfg.assoc && !force_assoc_off) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
 		cmd.client.is_assoc = cpu_to_le32(1);
-- 
cgit v1.2.3-70-g09d2


From cacc1d42a489f6f61cd5bdc42d216ec29b144317 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:03 +0300
Subject: wifi: iwlwifi: mvm: add link_conf parameter for add/remove/change
 link

Add link_conf parameter and change all the relevant calls accordingly.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.374015eed2e3.Icbf15a18e2599b53f4fa1c92fe3db64b551b84b1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      | 81 +++++++++++++---------
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  | 46 +++++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  2 +-
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 50 +++++++------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 13 +++-
 5 files changed, 118 insertions(+), 74 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index d8902e0ed2ae..bbaac2ba9079 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -20,11 +20,16 @@ static int iwl_mvm_link_cmd_send(struct iwl_mvm *mvm,
 	return ret;
 }
 
-int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_phy_ctxt *phyctxt = mvmvif->deflink.phy_ctxt;
+	unsigned int link_id = link_conf->link_id;
 	struct iwl_link_config_cmd cmd = {};
+	struct iwl_mvm_phy_ctxt *phyctxt;
+
+	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+		return -EINVAL;
 
 	/* Update SF - Disable if needed. if this fails, SF might still be on
 	 * while many macs are bound, which is forbidden - so fail the binding.
@@ -32,82 +37,94 @@ int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	if (iwl_mvm_sf_update(mvm, vif, false))
 		return -EINVAL;
 
+	/* FIXME: add proper link id allocation */
 	cmd.link_id = cpu_to_le32(mvmvif->id);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
 	/* P2P-Device already has a valid PHY context during add */
+	phyctxt = mvmvif->link[link_id]->phy_ctxt;
 	if (phyctxt)
 		cmd.phy_id = cpu_to_le32(phyctxt->id);
 	else
 		cmd.phy_id = cpu_to_le32(FW_CTXT_INVALID);
 
-	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
+	memcpy(cmd.local_link_addr, link_conf->addr, ETH_ALEN);
 
-	if (vif->type == NL80211_IFTYPE_ADHOC && vif->bss_conf.bssid)
-		memcpy(cmd.ibss_bssid_addr, vif->bss_conf.bssid, ETH_ALEN);
+	if (vif->type == NL80211_IFTYPE_ADHOC && link_conf->bssid)
+		memcpy(cmd.ibss_bssid_addr, link_conf->bssid, ETH_ALEN);
 
 	return iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_ADD);
 }
 
 int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 struct ieee80211_bss_conf *link_conf,
 			 u32 changes, bool active)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_phy_ctxt *phyctxt = mvmvif->deflink.phy_ctxt;
+	unsigned int link_id = link_conf->link_id;
+	struct iwl_mvm_phy_ctxt *phyctxt;
 	struct iwl_link_config_cmd cmd = {};
 	u32 ht_flag, flags = 0, flags_mask = 0;
 
+	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+		return -EINVAL;
+
+	/* FIXME: add proper link id allocation */
 	cmd.link_id = cpu_to_le32(mvmvif->id);
 
 	/* The phy_id, link address and listen_lmac can be modified only until
 	 * the link becomes active, otherwise they will be ignored.
 	 */
+	phyctxt = mvmvif->link[link_id]->phy_ctxt;
 	if (phyctxt)
 		cmd.phy_id = cpu_to_le32(phyctxt->id);
 	else
 		cmd.phy_id = cpu_to_le32(FW_CTXT_INVALID);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
 
-	memcpy(cmd.local_link_addr, vif->addr, ETH_ALEN);
+	memcpy(cmd.local_link_addr, link_conf->addr, ETH_ALEN);
 
 	cmd.active = cpu_to_le32(active);
 
-	if (vif->type == NL80211_IFTYPE_ADHOC && vif->bss_conf.bssid)
-		memcpy(cmd.ibss_bssid_addr, vif->bss_conf.bssid, ETH_ALEN);
+	if (vif->type == NL80211_IFTYPE_ADHOC && link_conf->bssid)
+		memcpy(cmd.ibss_bssid_addr, link_conf->bssid, ETH_ALEN);
 
 	/* TODO: set a value to cmd.listen_lmac when system requiremens
 	 * will define it
 	 */
 
-	iwl_mvm_set_fw_basic_rates(mvm, vif, &cmd.cck_rates, &cmd.ofdm_rates);
+	iwl_mvm_set_fw_basic_rates(mvm, vif, link_conf,
+				   &cmd.cck_rates, &cmd.ofdm_rates);
 
-	cmd.cck_short_preamble = cpu_to_le32(vif->bss_conf.use_short_preamble);
-	cmd.short_slot = cpu_to_le32(vif->bss_conf.use_short_slot);
+	cmd.cck_short_preamble = cpu_to_le32(link_conf->use_short_preamble);
+	cmd.short_slot = cpu_to_le32(link_conf->use_short_slot);
 
 	/* The fw does not distinguish between ht and fat */
 	ht_flag = LINK_PROT_FLG_HT_PROT | LINK_PROT_FLG_FAT_PROT;
-	iwl_mvm_set_fw_protection_flags(mvm, vif, &cmd.protection_flags,
+	iwl_mvm_set_fw_protection_flags(mvm, vif, link_conf,
+					&cmd.protection_flags,
 					ht_flag, LINK_PROT_FLG_TGG_PROTECT);
 
-	iwl_mvm_set_fw_qos_params(mvm, vif, &cmd.ac[0], &cmd.qos_flags);
+	iwl_mvm_set_fw_qos_params(mvm, vif, link_conf, &cmd.ac[0],
+				  &cmd.qos_flags);
 
 
-	cmd.bi = cpu_to_le32(vif->bss_conf.beacon_int);
-	cmd.dtim_interval = cpu_to_le32(vif->bss_conf.beacon_int *
-					vif->bss_conf.dtim_period);
+	cmd.bi = cpu_to_le32(link_conf->beacon_int);
+	cmd.dtim_interval = cpu_to_le32(link_conf->beacon_int *
+					link_conf->dtim_period);
 
-	if (!vif->bss_conf.he_support || iwlwifi_mod_params.disable_11ax ||
+	if (!link_conf->he_support || iwlwifi_mod_params.disable_11ax ||
 	    !vif->cfg.assoc) {
 		changes &= ~LINK_CONTEXT_MODIFY_HE_PARAMS;
 		goto send_cmd;
 	}
 
-	cmd.htc_trig_based_pkt_ext = vif->bss_conf.htc_trig_based_pkt_ext;
+	cmd.htc_trig_based_pkt_ext = link_conf->htc_trig_based_pkt_ext;
 
-	if (vif->bss_conf.uora_exists) {
+	if (link_conf->uora_exists) {
 		cmd.rand_alloc_ecwmin =
-			vif->bss_conf.uora_ocw_range & 0x7;
+			link_conf->uora_ocw_range & 0x7;
 		cmd.rand_alloc_ecwmax =
-			(vif->bss_conf.uora_ocw_range >> 3) & 0x7;
+			(link_conf->uora_ocw_range >> 3) & 0x7;
 	}
 
 	/* TODO  how to set ndp_fdbk_buff_th_exp? */
@@ -118,20 +135,20 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		flags_mask |= LINK_FLG_MU_EDCA_CW;
 	}
 
-	if (vif->bss_conf.eht_puncturing && !iwlwifi_mod_params.disable_11be)
-		cmd.puncture_mask = cpu_to_le16(vif->bss_conf.eht_puncturing);
+	if (link_conf->eht_puncturing && !iwlwifi_mod_params.disable_11be)
+		cmd.puncture_mask = cpu_to_le16(link_conf->eht_puncturing);
 	else
 		/* This flag can be set only if the MAC has eht support */
 		changes &= ~LINK_CONTEXT_MODIFY_EHT_PARAMS;
 
-	cmd.bss_color = vif->bss_conf.he_bss_color.color;
+	cmd.bss_color = link_conf->he_bss_color.color;
 
-	if (!vif->bss_conf.he_bss_color.enabled) {
+	if (!link_conf->he_bss_color.enabled) {
 		flags |= LINK_FLG_BSS_COLOR_DIS;
 		flags_mask |= LINK_FLG_BSS_COLOR_DIS;
 	}
 
-	cmd.frame_time_rts_th = cpu_to_le16(vif->bss_conf.frame_time_rts_th);
+	cmd.frame_time_rts_th = cpu_to_le16(link_conf->frame_time_rts_th);
 
 	/* Block 26-tone RU OFDMA transmissions */
 	if (mvmvif->deflink.he_ru_2mhz_block) {
@@ -139,10 +156,10 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		flags_mask |= LINK_FLG_RU_2MHZ_BLOCK;
 	}
 
-	if (vif->bss_conf.nontransmitted) {
+	if (link_conf->nontransmitted) {
 		ether_addr_copy(cmd.ref_bssid_addr,
-				vif->bss_conf.transmitter_bssid);
-		cmd.bssid_index = vif->bss_conf.bssid_index;
+				link_conf->transmitter_bssid);
+		cmd.bssid_index = link_conf->bssid_index;
 	}
 
 send_cmd:
@@ -153,12 +170,14 @@ send_cmd:
 	return iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_MODIFY);
 }
 
-int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_link_config_cmd cmd = {};
 	int ret;
 
+	/* FIXME: add proper link id allocation */
 	cmd.link_id = cpu_to_le32(mvmvif->id);
 	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_REMOVE);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index a06513989469..432b9cbcd8a4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -397,16 +397,18 @@ static void iwl_mvm_ack_rates(struct iwl_mvm *mvm,
 }
 
 void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *link_conf,
 				__le32 *cck_rates, __le32 *ofdm_rates)
 {
 	struct ieee80211_chanctx_conf *chanctx;
-	u8 cck_ack_rates, ofdm_ack_rates;
+	u8 cck_ack_rates = 0, ofdm_ack_rates = 0;
 
 	rcu_read_lock();
-	chanctx = rcu_dereference(vif->bss_conf.chanctx_conf);
+	chanctx = rcu_dereference(link_conf->chanctx_conf);
 	iwl_mvm_ack_rates(mvm, vif, chanctx ? chanctx->def.chan->band
 					    : NL80211_BAND_2GHZ,
 			  &cck_ack_rates, &ofdm_ack_rates);
+
 	rcu_read_unlock();
 
 	*cck_rates = cpu_to_le32((u32)cck_ack_rates);
@@ -415,21 +417,22 @@ void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
 				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *link_conf,
 				     __le32 *protection_flags, u32 ht_flag,
 				     u32 tgg_flag)
 {
 	/* for both sta and ap, ht_operation_mode hold the protection_mode */
-	u8 protection_mode = vif->bss_conf.ht_operation_mode &
+	u8 protection_mode = link_conf->ht_operation_mode &
 				 IEEE80211_HT_OP_MODE_PROTECTION;
-	bool ht_enabled = !!(vif->bss_conf.ht_operation_mode &
+	bool ht_enabled = !!(link_conf->ht_operation_mode &
 			     IEEE80211_HT_OP_MODE_PROTECTION);
 
-	if (vif->bss_conf.use_cts_prot)
+	if (link_conf->use_cts_prot)
 		*protection_flags |= cpu_to_le32(tgg_flag);
 
 	IWL_DEBUG_RATE(mvm, "use_cts_prot %d, ht_operation_mode %d\n",
-		       vif->bss_conf.use_cts_prot,
-		       vif->bss_conf.ht_operation_mode);
+		       link_conf->use_cts_prot,
+		       link_conf->ht_operation_mode);
 
 	if (!ht_enabled)
 		return;
@@ -448,7 +451,7 @@ void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
 		break;
 	case IEEE80211_HT_OP_MODE_PROTECTION_20MHZ:
 		/* Protect when channel wider than 20MHz */
-		if (vif->bss_conf.chandef.width > NL80211_CHAN_WIDTH_20)
+		if (link_conf->chandef.width > NL80211_CHAN_WIDTH_20)
 			*protection_flags |= cpu_to_le32(ht_flag);
 		break;
 	default:
@@ -459,6 +462,7 @@ void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
 }
 
 void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			       struct ieee80211_bss_conf *link_conf,
 			       struct iwl_ac_qos *ac, __le32 *qos_flags)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -478,10 +482,10 @@ void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		ac[ucode_ac].fifos_mask = BIT(txf);
 	}
 
-	if (vif->bss_conf.qos)
+	if (link_conf->qos)
 		*qos_flags |= cpu_to_le32(MAC_QOS_FLG_UPDATE_EDCA);
 
-	if (vif->bss_conf.chandef.width != NL80211_CHAN_WIDTH_20_NOHT)
+	if (link_conf->chandef.width != NL80211_CHAN_WIDTH_20_NOHT)
 		*qos_flags |= cpu_to_le32(MAC_QOS_FLG_TGN);
 }
 
@@ -538,7 +542,7 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 	else
 		eth_broadcast_addr(cmd->bssid_addr);
 
-	iwl_mvm_set_fw_basic_rates(mvm, vif, &cmd->cck_rates,
+	iwl_mvm_set_fw_basic_rates(mvm, vif, &vif->bss_conf, &cmd->cck_rates,
 				   &cmd->ofdm_rates);
 
 	cmd->cck_short_preamble =
@@ -550,11 +554,13 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 
 	cmd->filter_flags = 0;
 
-	iwl_mvm_set_fw_qos_params(mvm, vif, &cmd->ac[0], &cmd->qos_flags);
+	iwl_mvm_set_fw_qos_params(mvm, vif, &vif->bss_conf, &cmd->ac[0],
+				  &cmd->qos_flags);
 
 	/* The fw does not distinguish between ht and fat */
 	ht_flag = MAC_PROT_FLG_HT_PROT | MAC_PROT_FLG_FAT_PROT;
-	iwl_mvm_set_fw_protection_flags(mvm, vif, &cmd->protection_flags,
+	iwl_mvm_set_fw_protection_flags(mvm, vif, &vif->bss_conf,
+					&cmd->protection_flags,
 					ht_flag, MAC_PROT_FLG_TGG_PROTECT);
 }
 
@@ -570,6 +576,7 @@ static int iwl_mvm_mac_ctxt_send_cmd(struct iwl_mvm *mvm,
 }
 
 void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf,
 			      __le64 *dtim_tsf, __le32 *dtim_time,
 			      __le32 *assoc_beacon_arrive_time)
 {
@@ -590,17 +597,17 @@ void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	 * 384us in the longest case), this is currently not relevant
 	 * as the firmware wakes up around 2ms before the TBTT.
 	 */
-	dtim_offs = vif->bss_conf.sync_dtim_count *
-			vif->bss_conf.beacon_int;
+	dtim_offs = link_conf->sync_dtim_count *
+			link_conf->beacon_int;
 	/* convert TU to usecs */
 	dtim_offs *= 1024;
 
 	*dtim_tsf =
-		cpu_to_le64(vif->bss_conf.sync_tsf + dtim_offs);
+		cpu_to_le64(link_conf->sync_tsf + dtim_offs);
 	*dtim_time =
-		cpu_to_le32(vif->bss_conf.sync_device_ts + dtim_offs);
+		cpu_to_le32(link_conf->sync_device_ts + dtim_offs);
 	*assoc_beacon_arrive_time =
-		cpu_to_le32(vif->bss_conf.sync_device_ts);
+		cpu_to_le32(link_conf->sync_device_ts);
 
 	IWL_DEBUG_INFO(mvm, "DTIM TBTT is 0x%llx/0x%x, offset %d\n",
 		       le64_to_cpu(*dtim_tsf),
@@ -666,7 +673,8 @@ static int iwl_mvm_mac_ctxt_cmd_sta(struct iwl_mvm *mvm,
 	    !force_assoc_off) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
-		iwl_mvm_set_fw_dtim_tbtt(mvm, vif, &ctxt_sta->dtim_tsf,
+		iwl_mvm_set_fw_dtim_tbtt(mvm, vif, &vif->bss_conf,
+					 &ctxt_sta->dtim_tsf,
 					 &ctxt_sta->dtim_time,
 					 &ctxt_sta->assoc_beacon_arrive_time);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index ef3eb57b806a..cff24bf7e799 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3619,7 +3619,7 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 			callbacks->mac_ctxt_changed(mvm, vif, false);
 
 			if (mvm->mld_api_is_used)
-				iwl_mvm_link_changed(mvm, vif,
+				iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
 						     LINK_CONTEXT_MODIFY_ALL &
 						     ~LINK_CONTEXT_MODIFY_ACTIVE,
 						     true);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 80f1a1ccf16d..d02b3165f1d5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -67,11 +67,13 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 		}
 
 		iwl_mvm_phy_ctxt_ref(mvm, mvmvif->deflink.phy_ctxt);
-		ret = iwl_mvm_add_link(mvm, vif);
+		ret = iwl_mvm_add_link(mvm, vif, &vif->bss_conf);
 		if (ret)
 			goto out_unref_phy;
 
-		ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
+		ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+					   LINK_CONTEXT_MODIFY_ACTIVE |
+					   LINK_CONTEXT_MODIFY_RATES_INFO,
 					   true);
 		if (ret)
 			goto out_remove_link;
@@ -85,7 +87,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 		 */
 		mvm->p2p_device_vif = vif;
 	} else {
-		ret = iwl_mvm_add_link(mvm, vif);
+		ret = iwl_mvm_add_link(mvm, vif, &vif->bss_conf);
 		if (ret)
 			goto out_free_bf;
 	}
@@ -117,8 +119,9 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 
  out_remove_link:
 	/* Link needs to be deactivated before removal */
-	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
-	iwl_mvm_remove_link(mvm, vif);
+	iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+			     LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_remove_link(mvm, vif, &vif->bss_conf);
  out_unref_phy:
 	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
  out_free_bf:
@@ -193,13 +196,13 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 		mvm->p2p_device_vif = NULL;
 		iwl_mvm_mld_rm_bcast_sta(mvm, vif);
 		/* Link needs to be deactivated before removal */
-		iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE,
-				     false);
-		iwl_mvm_remove_link(mvm, vif);
+		iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+				     LINK_CONTEXT_MODIFY_ACTIVE, false);
+		iwl_mvm_remove_link(mvm, vif, &vif->bss_conf);
 		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
 		mvmvif->deflink.phy_ctxt = NULL;
 	} else {
-		iwl_mvm_remove_link(mvm, vif);
+		iwl_mvm_remove_link(mvm, vif, &vif->bss_conf);
 	}
 
 	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
@@ -233,12 +236,13 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	}
 
 	/* send it first with phy context ID */
-	ret = iwl_mvm_link_changed(mvm, vif, 0, false);
+	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf, 0, false);
 	if (ret)
 		goto out;
 
 	/* then activate */
-	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE |
+	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+				   LINK_CONTEXT_MODIFY_ACTIVE |
 				   LINK_CONTEXT_MODIFY_RATES_INFO,
 				   true);
 	if (ret)
@@ -259,7 +263,8 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	return 0;
 
 deactivate:
-	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+			     LINK_CONTEXT_MODIFY_ACTIVE, false);
 out:
 	mvmvif->deflink.phy_ctxt = NULL;
 	iwl_mvm_power_update_mac(mvm);
@@ -303,7 +308,8 @@ static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		iwl_mvm_mld_rm_snif_sta(mvm, vif);
 
-	iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+			     LINK_CONTEXT_MODIFY_ACTIVE, false);
 
 	if (switching_chanctx)
 		return;
@@ -337,7 +343,7 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	if (ret)
 		goto out_unlock;
 
-	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ALL,
+	ret = iwl_mvm_link_changed(mvm, vif, link_conf, LINK_CONTEXT_MODIFY_ALL,
 				   true);
 	if (ret)
 		goto out_unlock;
@@ -477,7 +483,8 @@ iwl_mvm_mld_bss_info_changed_station(struct iwl_mvm *mvm,
 		link_changes |= LINK_CONTEXT_MODIFY_EHT_PARAMS;
 
 	if (link_changes) {
-		ret = iwl_mvm_link_changed(mvm, vif, link_changes, true);
+		ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+					   link_changes, true);
 		if (ret)
 			IWL_ERR(mvm, "failed to update link\n");
 	}
@@ -576,7 +583,8 @@ iwl_mvm_mld_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 
 	if (changes & (BSS_CHANGED_ERP_CTS_PROT | BSS_CHANGED_HT |
 		       BSS_CHANGED_BANDWIDTH | BSS_CHANGED_QOS) &&
-		       iwl_mvm_link_changed(mvm, vif, link_changes, true))
+		       iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+					    link_changes, true))
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
 
 	/* Need to send a new beacon template to the FW */
@@ -661,7 +669,7 @@ iwl_mvm_mld_mac_conf_tx(struct ieee80211_hw *hw,
 		int ret;
 
 		mutex_lock(&mvm->mutex);
-		ret = iwl_mvm_link_changed(mvm, vif,
+		ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
 					   LINK_CONTEXT_MODIFY_QOS_PARAMS,
 					   true);
 		mutex_unlock(&mvm->mutex);
@@ -683,7 +691,8 @@ static int iwl_mvm_link_switch_phy_ctx(struct iwl_mvm *mvm,
 	 * inactive. Therefore, first deactivate the link, then change its
 	 * phy_ctx, and then activate it again.
 	 */
-	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, false);
+	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+				   LINK_CONTEXT_MODIFY_ACTIVE, false);
 	if (WARN(ret, "Failed to deactivate link\n"))
 		return ret;
 
@@ -691,11 +700,12 @@ static int iwl_mvm_link_switch_phy_ctx(struct iwl_mvm *mvm,
 
 	mvmvif->deflink.phy_ctxt = new_phy_ctxt;
 
-	ret = iwl_mvm_link_changed(mvm, vif, 0, false);
+	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf, 0, false);
 	if (WARN(ret, "Failed to deactivate link\n"))
 		return ret;
 
-	ret = iwl_mvm_link_changed(mvm, vif, LINK_CONTEXT_MODIFY_ACTIVE, true);
+	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+				   LINK_CONTEXT_MODIFY_ACTIVE, true);
 	WARN(ret, "Failed binding P2P_DEVICE\n");
 	return ret;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index d62341269744..03710c0cb57d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1721,17 +1721,21 @@ u8 iwl_mvm_get_ctrl_pos(struct cfg80211_chan_def *chandef);
 void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif);
 void iwl_mvm_set_fw_basic_rates(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *link_conf,
 				__le32 *cck_rates, __le32 *ofdm_rates);
 void iwl_mvm_set_fw_protection_flags(struct iwl_mvm *mvm,
 				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *link_conf,
 				     __le32 *protection_flags, u32 ht_flag,
-				  u32 tgg_flag);
+				     u32 tgg_flag);
 void iwl_mvm_set_fw_qos_params(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			       struct ieee80211_bss_conf *link_conf,
 			       struct iwl_ac_qos *ac, __le32 *qos_flags);
 bool iwl_mvm_set_fw_mu_edca_params(struct iwl_mvm *mvm,
 				   struct iwl_mvm_vif *mvmvif,
 				   struct iwl_he_backoff_conf *trig_based_txf);
 void iwl_mvm_set_fw_dtim_tbtt(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf,
 			      __le64 *dtim_tsf, __le32 *dtim_time,
 			      __le32 *assoc_beacon_arrive_time);
 __le32 iwl_mac_ctxt_p2p_dev_has_extended_disc(struct iwl_mvm *mvm,
@@ -1799,10 +1803,13 @@ int iwl_mvm_binding_add_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_binding_remove_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 
 /* Links */
-int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		     struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 struct ieee80211_bss_conf *link_conf,
 			 u32 changes, bool active);
-int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			struct ieee80211_bss_conf *link_conf);
 
 /* AP and IBSS */
 bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
-- 
cgit v1.2.3-70-g09d2


From 22c588343529d34e971b7cafae99d23f80dff7e9 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:04 +0300
Subject: wifi: iwlwifi: mvm: replace bss_info_changed() with
 vif_cfg/link_info_changed()

These are two new handlers for MLO. As the configurations done in
bss_info_changed() are now split into two separate flows, use MLO
specific implementation instead of common functions with the non-MLO
code.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.7b238cae0895.Ieb87f204787fb1c7cb7562e1cbf54ef518d87123@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |   6 +-
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 297 ++++++++++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   1 +
 4 files changed, 206 insertions(+), 100 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index bbaac2ba9079..320a3bc4e9c0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -113,7 +113,7 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 					link_conf->dtim_period);
 
 	if (!link_conf->he_support || iwlwifi_mod_params.disable_11ax ||
-	    !vif->cfg.assoc) {
+	    (vif->type == NL80211_IFTYPE_STATION && !vif->cfg.assoc)) {
 		changes &= ~LINK_CONTEXT_MODIFY_HE_PARAMS;
 		goto send_cmd;
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index cff24bf7e799..b90b0a973cec 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1888,8 +1888,7 @@ static void iwl_mvm_config_iface_filter(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static int iwl_mvm_update_mu_groups(struct iwl_mvm *mvm,
-				    struct ieee80211_vif *vif)
+int iwl_mvm_update_mu_groups(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 	struct iwl_mu_group_mgmt_cmd cmd = {};
 
@@ -2523,6 +2522,9 @@ iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
 		/* reset cqm events tracking */
 		mvmvif->bf_data.last_cqm_event = 0;
 		if (mvmvif->bf_data.bf_enabled) {
+			/* FIXME: need to update per link when FW API will
+			 * support it
+			 */
 			ret = iwl_mvm_enable_beacon_filter(mvm, vif, 0);
 			if (ret)
 				IWL_ERR(mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index d02b3165f1d5..950f252c1a5a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -454,37 +454,41 @@ static int iwl_mvm_mld_mac_sta_state(struct ieee80211_hw *hw,
 }
 
 static void
-iwl_mvm_mld_bss_info_changed_station(struct iwl_mvm *mvm,
-				     struct ieee80211_vif *vif,
-				     struct ieee80211_bss_conf *bss_conf,
-				     u64 changes)
+iwl_mvm_mld_link_info_changed_station(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_bss_conf *link_conf,
+				      u64 changes)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	int ret;
+	bool has_he, has_eht;
 	u32 link_changes = 0;
-	bool has_he = vif->bss_conf.he_support &&
-			  !iwlwifi_mod_params.disable_11ax;
-	bool has_eht = vif->bss_conf.eht_support &&
-			  !iwlwifi_mod_params.disable_11be;
+	int ret;
+
+	if (WARN_ON_ONCE(!mvmvif->link[link_conf->link_id]))
+		return;
+
+	has_he = link_conf->he_support && !iwlwifi_mod_params.disable_11ax;
+	has_eht = link_conf->eht_support && !iwlwifi_mod_params.disable_11be;
 
-	if (changes & BSS_CHANGED_ASSOC && vif->cfg.assoc &&
-	    (has_he || has_eht)) {
+	/* Update EDCA params */
+	if (changes & BSS_CHANGED_QOS && vif->cfg.assoc && link_conf->qos)
+		link_changes |= LINK_CONTEXT_MODIFY_QOS_PARAMS;
+
+	if (changes & BSS_CHANGED_ERP_SLOT)
+		link_changes |= LINK_CONTEXT_MODIFY_RATES_INFO;
+
+	if (vif->cfg.assoc && (has_he || has_eht)) {
 		IWL_DEBUG_MAC80211(mvm, "Associated in HE mode\n");
 		link_changes |= LINK_CONTEXT_MODIFY_HE_PARAMS;
 	}
 
-	/* Update MU EDCA params */
-	if (changes & BSS_CHANGED_QOS && vif->cfg.assoc &&
-	    (has_he || has_eht))
-		link_changes |= LINK_CONTEXT_MODIFY_QOS_PARAMS;
-
 	/* Update EHT Puncturing info */
 	if (changes & BSS_CHANGED_EHT_PUNCTURING && vif->cfg.assoc && has_eht)
 		link_changes |= LINK_CONTEXT_MODIFY_EHT_PARAMS;
 
 	if (link_changes) {
-		ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
-					   link_changes, true);
+		ret = iwl_mvm_link_changed(mvm, vif, link_conf, link_changes,
+					   true);
 		if (ret)
 			IWL_ERR(mvm, "failed to update link\n");
 	}
@@ -493,85 +497,139 @@ iwl_mvm_mld_bss_info_changed_station(struct iwl_mvm *mvm,
 	if (ret)
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
 
-	memcpy(mvmvif->deflink.bssid, bss_conf->bssid, ETH_ALEN);
+	memcpy(mvmvif->link[link_conf->link_id]->bssid, link_conf->bssid,
+	       ETH_ALEN);
+
+	iwl_mvm_bss_info_changed_station_common(mvm, vif, link_conf, changes);
+}
+
+static bool iwl_mvm_mld_vif_have_valid_ap_sta(struct iwl_mvm_vif *mvmvif)
+{
+	int i;
+
+	for_each_mvm_vif_valid_link(mvmvif, i) {
+		if (mvmvif->link[i]->ap_sta_id != IWL_MVM_INVALID_STA)
+			return true;
+	}
+
+	return false;
+}
+
+static void iwl_mvm_mld_vif_delete_all_stas(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int i, ret;
+
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+		return;
+
+	iwl_mvm_sec_key_remove_ap(mvm, vif);
+
+	for_each_mvm_vif_valid_link(mvmvif, i) {
+		struct iwl_mvm_vif_link_info *link = mvmvif->link[i];
+
+		if (!link)
+			continue;
+
+		ret = iwl_mvm_mld_rm_sta_id(mvm, vif, link->ap_sta_id);
+		if (ret)
+			IWL_ERR(mvm, "failed to remove AP station\n");
+
+		link->ap_sta_id = IWL_MVM_INVALID_STA;
+	}
+}
+
+static void iwl_mvm_mld_vif_cfg_changed_station(struct iwl_mvm *mvm,
+						struct ieee80211_vif *vif,
+						u64 changes)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct ieee80211_bss_conf *link_conf;
+	bool protect = false;
+	unsigned int i;
+	int ret;
+
+	ret = iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
+	if (ret)
+		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
+
 	mvmvif->associated = vif->cfg.assoc;
 
-	if (changes & BSS_CHANGED_ASSOC) {
-		if (vif->cfg.assoc) {
-			/* clear statistics to get clean beacon counter */
-			iwl_mvm_request_statistics(mvm, true);
-			memset(&mvmvif->deflink.beacon_stats, 0,
-			       sizeof(mvmvif->deflink.beacon_stats));
-
-			if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART,
-				      &mvm->status) &&
-			    !vif->bss_conf.dtim_period) {
-				/* If we're not restarting and still haven't
-				 * heard a beacon (dtim period unknown) then
-				 * make sure we still have enough minimum time
-				 * remaining in the time event, since the auth
-				 * might actually have taken quite a while
-				 * (especially for SAE) and so the remaining
-				 * time could be small without us having heard
-				 * a beacon yet.
-				 */
-				iwl_mvm_protect_assoc(mvm, vif, 0);
-			}
+	if (!(changes & BSS_CHANGED_ASSOC))
+		return;
+
+	if (vif->cfg.assoc) {
+		/* clear statistics to get clean beacon counter */
+		iwl_mvm_request_statistics(mvm, true);
+		iwl_mvm_sf_update(mvm, vif, false);
+		iwl_mvm_power_vif_assoc(mvm, vif);
+
+		for_each_mvm_vif_valid_link(mvmvif, i) {
+			memset(&mvmvif->link[i]->beacon_stats, 0,
+			       sizeof(mvmvif->link[i]->beacon_stats));
 
-			iwl_mvm_sf_update(mvm, vif, false);
-			iwl_mvm_power_vif_assoc(mvm, vif);
 			if (vif->p2p) {
 				iwl_mvm_update_smps(mvm, vif,
 						    IWL_MVM_SMPS_REQ_PROT,
-						    IEEE80211_SMPS_DYNAMIC, 0);
+						    IEEE80211_SMPS_DYNAMIC, i);
 			}
-		} else if (mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
-			iwl_mvm_mei_host_disassociated(mvm);
-			/* If update fails - SF might be running in associated
-			 * mode while disassociated - which is forbidden.
-			 */
-			ret = iwl_mvm_sf_update(mvm, vif, false);
-			WARN_ONCE(ret &&
-				  !test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED,
-					    &mvm->status),
-				  "Failed to update SF upon disassociation\n");
-
-			/* If we get an assert during the connection (after the
-			 * station has been added, but before the vif is set
-			 * to associated), mac80211 will re-add the station and
-			 * then configure the vif. Since the vif is not
-			 * associated, we would remove the station here and
-			 * this would fail the recovery.
+
+			rcu_read_lock();
+			link_conf = rcu_dereference(vif->link_conf[i]);
+			if (link_conf && !link_conf->dtim_period)
+				protect = true;
+			rcu_read_unlock();
+		}
+
+		if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
+		    protect) {
+			/* If we're not restarting and still haven't
+			 * heard a beacon (dtim period unknown) then
+			 * make sure we still have enough minimum time
+			 * remaining in the time event, since the auth
+			 * might actually have taken quite a while
+			 * (especially for SAE) and so the remaining
+			 * time could be small without us having heard
+			 * a beacon yet.
 			 */
-			if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART,
-				      &mvm->status)) {
-				/* first remove remaining keys */
-				iwl_mvm_sec_key_remove_ap(mvm, vif);
-
-				/* Remove AP station now that
-				 * the MAC is unassoc
-				 */
-				ret = iwl_mvm_mld_rm_sta_id(mvm, vif,
-							    mvmvif->deflink.ap_sta_id);
-				if (ret)
-					IWL_ERR(mvm,
-						"failed to remove AP station\n");
-
-				mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
-			}
+			iwl_mvm_protect_assoc(mvm, vif, 0);
 		}
 
-		iwl_mvm_bss_info_changed_station_assoc(mvm, vif, changes);
+		iwl_mvm_sf_update(mvm, vif, false);
+
+		/* FIXME: need to decide about misbehaving AP handling */
+		iwl_mvm_power_vif_assoc(mvm, vif);
+	} else if (iwl_mvm_mld_vif_have_valid_ap_sta(mvmvif)) {
+		iwl_mvm_mei_host_disassociated(mvm);
+
+		/* If update fails - SF might be running in associated
+		 * mode while disassociated - which is forbidden.
+		 */
+		ret = iwl_mvm_sf_update(mvm, vif, false);
+		WARN_ONCE(ret &&
+			  !test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED,
+				    &mvm->status),
+			  "Failed to update SF upon disassociation\n");
+
+		/* If we get an assert during the connection (after the
+		 * station has been added, but before the vif is set
+		 * to associated), mac80211 will re-add the station and
+		 * then configure the vif. Since the vif is not
+		 * associated, we would remove the station here and
+		 * this would fail the recovery.
+		 */
+		iwl_mvm_mld_vif_delete_all_stas(mvm, vif);
 	}
 
-	iwl_mvm_bss_info_changed_station_common(mvm, vif, &vif->bss_conf, changes);
+	iwl_mvm_bss_info_changed_station_assoc(mvm, vif, changes);
 }
 
 static void
-iwl_mvm_mld_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
-				     struct ieee80211_vif *vif,
-				     struct ieee80211_bss_conf *bss_conf,
-				     u64 changes)
+iwl_mvm_mld_link_info_changed_ap_ibss(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_bss_conf *link_conf,
+				      u64 changes)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	u32 link_changes = LINK_CONTEXT_MODIFY_PROTECT_FLAGS |
@@ -581,17 +639,22 @@ iwl_mvm_mld_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 	if (!mvmvif->ap_ibss_active)
 		return;
 
+	if (link_conf->he_support)
+		link_changes |= LINK_CONTEXT_MODIFY_HE_PARAMS;
+
 	if (changes & (BSS_CHANGED_ERP_CTS_PROT | BSS_CHANGED_HT |
-		       BSS_CHANGED_BANDWIDTH | BSS_CHANGED_QOS) &&
-		       iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+		       BSS_CHANGED_BANDWIDTH | BSS_CHANGED_QOS |
+		       BSS_CHANGED_HE_BSS_COLOR) &&
+		       iwl_mvm_link_changed(mvm, vif, link_conf,
 					    link_changes, true))
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
 
 	/* Need to send a new beacon template to the FW */
 	if (changes & BSS_CHANGED_BEACON &&
-	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, &vif->bss_conf))
+	    iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf))
 		IWL_WARN(mvm, "Failed updating beacon data\n");
 
+	/* FIXME: need to decide if we need FTM responder per link */
 	if (changes & BSS_CHANGED_FTM_RESPONDER) {
 		int ret = iwl_mvm_ftm_start_responder(mvm, vif);
 
@@ -601,19 +664,58 @@ iwl_mvm_mld_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 	}
 }
 
-static void iwl_mvm_mld_bss_info_changed(struct ieee80211_hw *hw,
-					 struct ieee80211_vif *vif,
-					 struct ieee80211_bss_conf *bss_conf,
-					 u64 changes)
+static void iwl_mvm_mld_link_info_changed(struct ieee80211_hw *hw,
+					  struct ieee80211_vif *vif,
+					  struct ieee80211_bss_conf *link_conf,
+					  u64 changes)
+{
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+
+	mutex_lock(&mvm->mutex);
+
+	switch (vif->type) {
+	case NL80211_IFTYPE_STATION:
+		iwl_mvm_mld_link_info_changed_station(mvm, vif, link_conf,
+						      changes);
+		break;
+	case NL80211_IFTYPE_AP:
+	case NL80211_IFTYPE_ADHOC:
+		iwl_mvm_mld_link_info_changed_ap_ibss(mvm, vif, link_conf,
+						      changes);
+		break;
+	case NL80211_IFTYPE_MONITOR:
+		if (changes & BSS_CHANGED_MU_GROUPS)
+			iwl_mvm_update_mu_groups(mvm, vif);
+		break;
+	default:
+		/* shouldn't happen */
+		WARN_ON_ONCE(1);
+	}
+
+	if (changes & BSS_CHANGED_TXPOWER) {
+		IWL_DEBUG_CALIB(mvm, "Changing TX Power to %d dBm\n",
+				link_conf->txpower);
+		iwl_mvm_set_tx_power(mvm, vif, link_conf->txpower);
+	}
+
+	mutex_unlock(&mvm->mutex);
+}
+
+static void iwl_mvm_mld_vif_cfg_changed(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					u64 changes)
 {
-	struct iwl_mvm_bss_info_changed_ops callbacks = {
-		.bss_info_changed_sta = iwl_mvm_mld_bss_info_changed_station,
-		.bss_info_changed_ap_ibss =
-			iwl_mvm_mld_bss_info_changed_ap_ibss,
-	};
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	iwl_mvm_bss_info_changed_common(hw, vif, bss_conf, &callbacks,
-					changes);
+	mutex_lock(&mvm->mutex);
+
+	if (changes & BSS_CHANGED_IDLE && !vif->cfg.idle)
+		iwl_mvm_scan_stop(mvm, IWL_MVM_SCAN_SCHED, true);
+
+	if (vif->type == NL80211_IFTYPE_STATION)
+		iwl_mvm_mld_vif_cfg_changed_station(mvm, vif, changes);
+
+	mutex_unlock(&mvm->mutex);
 }
 
 static int
@@ -735,7 +837,8 @@ const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.prepare_multicast = iwl_mvm_prepare_multicast,
 	.configure_filter = iwl_mvm_configure_filter,
 	.config_iface_filter = iwl_mvm_mld_config_iface_filter,
-	.bss_info_changed = iwl_mvm_mld_bss_info_changed,
+	.link_info_changed = iwl_mvm_mld_link_info_changed,
+	.vif_cfg_changed = iwl_mvm_mld_vif_cfg_changed,
 	.hw_scan = iwl_mvm_mac_hw_scan,
 	.cancel_hw_scan = iwl_mvm_mac_cancel_hw_scan,
 	.sta_pre_rcu_remove = iwl_mvm_sta_pre_rcu_remove,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 03710c0cb57d..03e035b12783 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2621,4 +2621,5 @@ int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
 			     struct ieee80211_vif *vif,
 			     struct cfg80211_set_hw_timestamp *hwts);
+int iwl_mvm_update_mu_groups(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 #endif /* __IWL_MVM_H__ */
-- 
cgit v1.2.3-70-g09d2


From 62e0ccb2170e7a20d62d38f781f90a8ef031c450 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:05 +0300
Subject: wifi: iwlwifi: mvm: adjust internal stations to MLO

Internal stations are mcast/bcast and sniffer. They need some link
specific data like bss id/phy id. Add link_conf as a parameter.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.25b013fd30de.I7f9f5906123acae05040cceec470c9acb07b9d42@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 21 +++++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 40 ++++++++++++++--------
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       | 15 +++++---
 3 files changed, 47 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 950f252c1a5a..c81b92d2e8f7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -78,7 +78,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 		if (ret)
 			goto out_remove_link;
 
-		ret = iwl_mvm_mld_add_bcast_sta(mvm, vif);
+		ret = iwl_mvm_mld_add_bcast_sta(mvm, vif, &vif->bss_conf);
 		if (ret)
 			goto out_remove_link;
 
@@ -194,7 +194,9 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 		mvm->p2p_device_vif = NULL;
-		iwl_mvm_mld_rm_bcast_sta(mvm, vif);
+
+		/* P2P device uses only one link */
+		iwl_mvm_mld_rm_bcast_sta(mvm, vif, &vif->bss_conf);
 		/* Link needs to be deactivated before removal */
 		iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
 				     LINK_CONTEXT_MODIFY_ACTIVE, false);
@@ -255,7 +257,8 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	iwl_mvm_power_update_mac(mvm);
 
 	if (vif->type == NL80211_IFTYPE_MONITOR) {
-		ret = iwl_mvm_mld_add_snif_sta(mvm, vif);
+		ret = iwl_mvm_mld_add_snif_sta(mvm, vif,
+					       &vif->bss_conf);
 		if (ret)
 			goto deactivate;
 	}
@@ -348,14 +351,14 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	if (ret)
 		goto out_unlock;
 
-	ret = iwl_mvm_mld_add_mcast_sta(mvm, vif);
+	ret = iwl_mvm_mld_add_mcast_sta(mvm, vif, link_conf);
 	if (ret)
 		goto out_unlock;
 
 	/* Send the bcast station. At this stage the TBTT and DTIM time
 	 * events are added and applied to the scheduler
 	 */
-	ret = iwl_mvm_mld_add_bcast_sta(mvm, vif);
+	ret = iwl_mvm_mld_add_bcast_sta(mvm, vif, link_conf);
 	if (ret)
 		goto out_rm_mcast;
 
@@ -379,9 +382,9 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 out_failed:
 	iwl_mvm_power_update_mac(mvm);
 	mvmvif->ap_ibss_active = false;
-	iwl_mvm_mld_rm_bcast_sta(mvm, vif);
+	iwl_mvm_mld_rm_bcast_sta(mvm, vif, link_conf);
 out_rm_mcast:
-	iwl_mvm_mld_rm_mcast_sta(mvm, vif);
+	iwl_mvm_mld_rm_mcast_sta(mvm, vif, link_conf);
 out_unlock:
 	mutex_unlock(&mvm->mutex);
 	return ret;
@@ -416,8 +419,8 @@ static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
 
 	iwl_mvm_ftm_responder_clear(mvm, vif);
 
-	iwl_mvm_mld_rm_bcast_sta(mvm, vif);
-	iwl_mvm_mld_rm_mcast_sta(mvm, vif);
+	iwl_mvm_mld_rm_bcast_sta(mvm, vif, link_conf);
+	iwl_mvm_mld_rm_mcast_sta(mvm, vif, link_conf);
 
 	iwl_mvm_power_update_mac(mvm);
 	mutex_unlock(&mvm->mutex);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 1608ac371bb0..75d4585c0b21 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -166,10 +166,13 @@ static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
  * and send it to the FW.
  * Note that each P2P mac should have its own broadcast station.
  */
-int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *bsta = &mvmvif->deflink.bcast_sta;
+	struct iwl_mvm_vif_link_info *mvm_link =
+		mvmvif->link[link_conf->link_id];
+	struct iwl_mvm_int_sta *bsta = &mvm_link->bcast_sta;
 	static const u8 _baddr[] = {0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF};
 	const u8 *baddr = _baddr;
 	unsigned int wdg_timeout =
@@ -179,7 +182,7 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	lockdep_assert_held(&mvm->mutex);
 
 	if (vif->type == NL80211_IFTYPE_ADHOC)
-		baddr = vif->bss_conf.bssid;
+		baddr = link_conf->bssid;
 
 	if (vif->type == NL80211_IFTYPE_AP ||
 	    vif->type == NL80211_IFTYPE_ADHOC) {
@@ -202,10 +205,13 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
  * and send it to the FW.
  * Note that each AP/GO mac should have its own multicast station.
  */
-int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_mvm_int_sta *msta = &mvmvif->deflink.mcast_sta;
+	struct iwl_mvm_vif_link_info *mvm_link =
+		mvmvif->link[link_conf->link_id];
+	struct iwl_mvm_int_sta *msta = &mvm_link->mcast_sta;
 	static const u8 _maddr[] = {0x03, 0x00, 0x00, 0x00, 0x00, 0x00};
 	const u8 *maddr = _maddr;
 	unsigned int timeout = iwl_mvm_get_wd_timeout(mvm, vif, false, false);
@@ -222,9 +228,9 @@ int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 * changes in mac80211 layer.
 	 */
 	if (vif->type == NL80211_IFTYPE_ADHOC)
-		mvmvif->deflink.cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
+		mvm_link->cab_queue = IWL_MVM_DQA_GCAST_QUEUE;
 
-	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvmvif->deflink.cab_queue,
+	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvm_link->cab_queue,
 				       vif->type, STATION_TYPE_MCAST,
 				       mvmvif->id, maddr, 0, &timeout);
 }
@@ -232,7 +238,8 @@ int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 /* Allocate a new station entry for the sniffer station to the given vif,
  * and send it to the FW.
  */
-int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
@@ -310,9 +317,11 @@ static int iwl_mvm_mld_rm_int_sta(struct iwl_mvm *mvm,
 	return ret;
 }
 
-int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif_link_info *link = mvmvif->link[link_conf->link_id];
 	u16 *queueptr;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -331,22 +340,23 @@ int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		return -EINVAL;
 	}
 
-	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->deflink.bcast_sta, true,
-				      IWL_MAX_TID_COUNT, queueptr);
+	return iwl_mvm_mld_rm_int_sta(mvm, &link->bcast_sta,
+				      true, IWL_MAX_TID_COUNT, queueptr);
 }
 
 /* Send the FW a request to remove the station from it's internal data
  * structures, and in addition remove it from the local data structure.
  */
-int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
+int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif_link_info *link = mvmvif->link[link_conf->link_id];
 
 	lockdep_assert_held(&mvm->mutex);
 
-	return iwl_mvm_mld_rm_int_sta(mvm, &mvmvif->deflink.mcast_sta, true,
-				      0,
-				      &mvmvif->deflink.cab_queue);
+	return iwl_mvm_mld_rm_int_sta(mvm, &link->mcast_sta, true, 0,
+				      &link->cab_queue);
 }
 
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 3a9002eeb5fb..7148263dddf8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -619,13 +619,18 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 
 /* New MLD STA related APIs */
 /* STA */
-int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
-int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
-int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf);
+int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *link_conf);
+int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			      struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id);
-int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_mld_rm_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
-int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
+int iwl_mvm_mld_rm_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			     struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_mld_rm_aux_sta(struct iwl_mvm *mvm);
 int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta);
-- 
cgit v1.2.3-70-g09d2


From d6f6b0d804e08e1cab1dcebe4f62dc949bb646e9 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:06 +0300
Subject: wifi: iwlwifi: mvm: add fw link id allocation

Driver uses link_id as an index in the array. FW currently can
support only 2 concurrently active links per vif with the ids in the
range 0-3. Add a mapping of dirver link ids to fw link id and track the
number of active link ids.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.a53e5df49c33.I02b25648d2d5ca370c0697bf19d0d34724eae8a1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/mac-cfg.h    |  6 ++
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c   |  6 +-
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c        |  2 +
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      | 87 ++++++++++++++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  |  6 +-
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 25 +++++--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 10 +++
 8 files changed, 121 insertions(+), 22 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index b644274e4a31..e236d1b0aae2 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -483,6 +483,12 @@ struct iwl_link_config_cmd {
 	__le32 reserved1[8];
 } __packed; /* LINK_CONTEXT_CONFIG_CMD_API_S_VER_1 */
 
+/* Currently FW supports link ids in the range 0-3 and can have
+ * at most two active links for each vif.
+ */
+#define IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM 2
+#define IWL_MVM_FW_MAX_LINK_ID 3
+
 /**
  * enum iwl_fw_sta_type - FW station types
  * @STATION_TYPE_PEER: represents a peer - AP in BSS, a TDLS sta, a client in
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 2c72f5cb19f1..527daaf46f96 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -1442,7 +1442,11 @@ static int _iwl_dbgfs_inject_beacon_ie(struct iwl_mvm *mvm, char *bin, int len)
 			cpu_to_le16(iwl_mvm_mac_ctxt_get_beacon_flags(mvm->fw,
 								      rate));
 		beacon_cmd.byte_cnt = cpu_to_le16((u16)beacon->len);
-		beacon_cmd.link_id = cpu_to_le32((u32)mvmvif->id);
+		if (iwl_fw_lookup_cmd_ver(mvm->fw, BEACON_TEMPLATE_CMD, 0) > 12)
+			beacon_cmd.link_id =
+				cpu_to_le32(mvmvif->link[link_id]->fw_link_id);
+		else
+			beacon_cmd.link_id = cpu_to_le32((u32)mvmvif->id);
 
 		iwl_mvm_mac_ctxt_set_tim(mvm, &beacon_cmd.tim_idx,
 					 &beacon_cmd.tim_size,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 6d6fa35c2b21..d302d98b7631 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1573,6 +1573,8 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++)
 		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[i], NULL);
 
+	memset(&mvm->fw_link_ids_map, 0, sizeof(mvm->fw_link_ids_map));
+
 	mvm->tdls_cs.peer.sta_id = IWL_MVM_INVALID_STA;
 
 	/* reset quota debouncing buffer - 0xff will yield invalid data */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 320a3bc4e9c0..7a70b6ee65ac 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -1,9 +1,34 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2022 Intel Corporation
+ * Copyright (C) 2022 - 2023 Intel Corporation
  */
 #include "mvm.h"
 
+static u32 iwl_mvm_get_free_fw_link_id(struct iwl_mvm *mvm,
+				       struct iwl_mvm_vif *mvm_vif)
+{
+	u32 link_id;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	link_id = ffz(mvm->fw_link_ids_map);
+
+	/* this case can happen if there're deactivated but not removed links */
+	if (link_id > IWL_MVM_FW_MAX_LINK_ID)
+		return IWL_MVM_FW_LINK_ID_INVALID;
+
+	mvm->fw_link_ids_map |= BIT(link_id);
+	return link_id;
+}
+
+static void iwl_mvm_release_fw_link_id(struct iwl_mvm *mvm, u32 link_id)
+{
+	lockdep_assert_held(&mvm->mutex);
+
+	if (!WARN_ON(link_id > IWL_MVM_FW_MAX_LINK_ID))
+		mvm->fw_link_ids_map &= ~BIT(link_id);
+}
+
 static int iwl_mvm_link_cmd_send(struct iwl_mvm *mvm,
 				 struct iwl_link_config_cmd *cmd,
 				 enum iwl_ctxt_action action)
@@ -25,23 +50,30 @@ int iwl_mvm_add_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	unsigned int link_id = link_conf->link_id;
+	struct iwl_mvm_vif_link_info *link_info = mvmvif->link[link_id];
 	struct iwl_link_config_cmd cmd = {};
 	struct iwl_mvm_phy_ctxt *phyctxt;
 
-	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+	if (WARN_ON_ONCE(!link_info))
 		return -EINVAL;
 
+	if (link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID) {
+		link_info->fw_link_id = iwl_mvm_get_free_fw_link_id(mvm,
+								    mvmvif);
+		if (link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID)
+			return -EINVAL;
+	}
+
 	/* Update SF - Disable if needed. if this fails, SF might still be on
 	 * while many macs are bound, which is forbidden - so fail the binding.
 	 */
 	if (iwl_mvm_sf_update(mvm, vif, false))
 		return -EINVAL;
 
-	/* FIXME: add proper link id allocation */
-	cmd.link_id = cpu_to_le32(mvmvif->id);
+	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
 	cmd.mac_id = cpu_to_le32(mvmvif->id);
 	/* P2P-Device already has a valid PHY context during add */
-	phyctxt = mvmvif->link[link_id]->phy_ctxt;
+	phyctxt = link_info->phy_ctxt;
 	if (phyctxt)
 		cmd.phy_id = cpu_to_le32(phyctxt->id);
 	else
@@ -61,20 +93,27 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	unsigned int link_id = link_conf->link_id;
+	struct iwl_mvm_vif_link_info *link_info = mvmvif->link[link_id];
 	struct iwl_mvm_phy_ctxt *phyctxt;
 	struct iwl_link_config_cmd cmd = {};
 	u32 ht_flag, flags = 0, flags_mask = 0;
+	int ret;
+
+	if (WARN_ON_ONCE(!link_info ||
+			 link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID))
+		return -EINVAL;
 
-	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+	/* cannot activate third link */
+	if (!link_info->active && active &&
+	    mvmvif->fw_active_links_num >= IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM)
 		return -EINVAL;
 
-	/* FIXME: add proper link id allocation */
-	cmd.link_id = cpu_to_le32(mvmvif->id);
+	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
 
 	/* The phy_id, link address and listen_lmac can be modified only until
 	 * the link becomes active, otherwise they will be ignored.
 	 */
-	phyctxt = mvmvif->link[link_id]->phy_ctxt;
+	phyctxt = link_info->phy_ctxt;
 	if (phyctxt)
 		cmd.phy_id = cpu_to_le32(phyctxt->id);
 	else
@@ -151,7 +190,7 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	cmd.frame_time_rts_th = cpu_to_le16(link_conf->frame_time_rts_th);
 
 	/* Block 26-tone RU OFDMA transmissions */
-	if (mvmvif->deflink.he_ru_2mhz_block) {
+	if (link_info->he_ru_2mhz_block) {
 		flags |= LINK_FLG_RU_2MHZ_BLOCK;
 		flags_mask |= LINK_FLG_RU_2MHZ_BLOCK;
 	}
@@ -167,18 +206,40 @@ send_cmd:
 	cmd.flags = cpu_to_le32(flags);
 	cmd.flags_mask = cpu_to_le32(flags_mask);
 
-	return iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_MODIFY);
+	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_MODIFY);
+	if (!ret) {
+		/* the FW is updated, so now it's possible to update the
+		 * activation status. If activating a link, it was already
+		 * checked above that we didn't reach the FW limit.
+		 */
+		if (link_info->active && !active)
+			mvmvif->fw_active_links_num--;
+		else if (!link_info->active && active)
+			mvmvif->fw_active_links_num++;
+
+		link_info->active = active;
+	}
+
+	return ret;
 }
 
 int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned int link_id = link_conf->link_id;
+	struct iwl_mvm_vif_link_info *link_info = mvmvif->link[link_id];
 	struct iwl_link_config_cmd cmd = {};
 	int ret;
 
-	/* FIXME: add proper link id allocation */
-	cmd.link_id = cpu_to_le32(mvmvif->id);
+	if (WARN_ON(!link_info ||
+		    link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID))
+		return -EINVAL;
+
+	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
+	iwl_mvm_release_fw_link_id(mvm, link_info->fw_link_id);
+	link_info->fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
+
 	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_REMOVE);
 
 	if (!ret)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 432b9cbcd8a4..82fad042a281 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -1078,7 +1078,11 @@ static int iwl_mvm_mac_ctxt_send_beacon_v9(struct iwl_mvm *mvm,
 
 	beacon_cmd.flags = cpu_to_le16(flags);
 	beacon_cmd.byte_cnt = cpu_to_le16((u16)beacon->len);
-	beacon_cmd.link_id = cpu_to_le32((u32)mvmvif->id);
+	if (iwl_fw_lookup_cmd_ver(mvm->fw, BEACON_TEMPLATE_CMD, 0) > 12)
+		beacon_cmd.link_id =
+			cpu_to_le32(mvmvif->link[link_conf->link_id]->fw_link_id);
+	else
+		beacon_cmd.link_id = cpu_to_le32((u32)mvmvif->id);
 
 	if (vif->type == NL80211_IFTYPE_AP)
 		iwl_mvm_mac_ctxt_set_tim(mvm, &beacon_cmd.tim_idx,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index c81b92d2e8f7..b861b5478e08 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -36,6 +36,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	mvmvif->features |= hw->netdev_features;
 
 	/* the first link always points to the default one */
+	mvmvif->deflink.fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
 	mvmvif->link[0] = &mvmvif->deflink;
 
 	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 75d4585c0b21..3791f37afa78 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -197,8 +197,8 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	return iwl_mvm_mld_add_int_sta(mvm, bsta, queue,
 				       ieee80211_vif_type_p2p(vif),
 				       STATION_TYPE_BCAST_MGMT,
-				       mvmvif->id, baddr, IWL_MAX_TID_COUNT,
-				       &wdg_timeout);
+				       mvm_link->fw_link_id, baddr,
+				       IWL_MAX_TID_COUNT, &wdg_timeout);
 }
 
 /* Allocate a new station entry for the broadcast station to the given vif,
@@ -232,7 +232,8 @@ int iwl_mvm_mld_add_mcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	return iwl_mvm_mld_add_int_sta(mvm, msta, &mvm_link->cab_queue,
 				       vif->type, STATION_TYPE_MCAST,
-				       mvmvif->id, maddr, 0, &timeout);
+				       mvm_link->fw_link_id, maddr, 0,
+				       &timeout);
 }
 
 /* Allocate a new station entry for the sniffer station to the given vif,
@@ -242,13 +243,15 @@ int iwl_mvm_mld_add_snif_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			     struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif_link_info *mvm_link =
+		mvmvif->link[link_conf->link_id];
 
 	lockdep_assert_held(&mvm->mutex);
 
 	return iwl_mvm_mld_add_int_sta(mvm, &mvm->snif_sta, &mvm->snif_queue,
 				       vif->type, STATION_TYPE_BCAST_MGMT,
-				       mvmvif->id, NULL, IWL_MAX_TID_COUNT,
-				       NULL);
+				       mvm_link->fw_link_id, NULL,
+				       IWL_MAX_TID_COUNT, NULL);
 }
 
 int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
@@ -380,15 +383,23 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			       struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif *mvm_vif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta_cfg_cmd cmd = {
 		.sta_id = cpu_to_le32(mvm_sta->deflink.sta_id),
-		.link_id = cpu_to_le32(mvmvif->id),
 		.station_type = cpu_to_le32(mvm_sta->sta_type),
 		.mfp = cpu_to_le32(sta->mfp),
 	};
+	/* FIXME: use proper link_id */
+	unsigned int link_id = 0;
+	struct iwl_mvm_vif_link_info *link_info = mvm_vif->link[link_id];
 	u32 agg_size = 0, mpdu_dens = 0;
 
+	/* when adding sta, link should exist in FW */
+	if (WARN_ON(link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID))
+		return -EINVAL;
+
+	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
+
 	/* For now the link addr is the same as the mld addr */
 	memcpy(&cmd.peer_mld_address, sta->addr, ETH_ALEN);
 	memcpy(&cmd.peer_link_address, sta->addr, ETH_ALEN);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 03e035b12783..ad67a9b71807 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -73,6 +73,9 @@
 /* offchannel queue towards mac80211 */
 #define IWL_MVM_OFFCHANNEL_QUEUE 0
 
+/* invalid value for FW link id */
+#define IWL_MVM_FW_LINK_ID_INVALID 0xff
+
 extern const struct ieee80211_ops iwl_mvm_hw_ops;
 extern const struct ieee80211_ops iwl_mvm_mld_hw_ops;
 
@@ -283,6 +286,7 @@ struct iwl_probe_resp_data {
 /**
  * struct iwl_mvm_vif_link_info - per link data in Virtual Interface
  * @ap_sta_id: the sta_id of the AP - valid only if VIF type is STA
+ * @fw_link_id: the id of the link according to the FW API
  * @bssid: BSSID for this (client) interface
  * @bcast_sta: station used for broadcast packets. Used by the following
  *	vifs: P2P_DEVICE, GO and AP.
@@ -299,6 +303,7 @@ struct iwl_probe_resp_data {
 struct iwl_mvm_vif_link_info {
 	u8 bssid[ETH_ALEN];
 	u8 ap_sta_id;
+	u8 fw_link_id;
 
 	struct iwl_mvm_int_sta bcast_sta;
 	struct iwl_mvm_int_sta mcast_sta;
@@ -312,6 +317,7 @@ struct iwl_mvm_vif_link_info {
 	struct iwl_probe_resp_data __rcu *probe_resp_data;
 
 	bool he_ru_2mhz_block;
+	bool active;
 
 	u16 cab_queue;
 	/* Assigned while mac80211 has the link in a channel context,
@@ -349,6 +355,7 @@ struct iwl_mvm_vif_link_info {
  * @csa_failed: CSA failed to schedule time event, report an error later
  * @csa_bcn_pending: indicates that we are waiting for a beacon on a new channel
  * @features: hw features active for this vif
+ * @fw_active_links_num: the number of activated links
  */
 struct iwl_mvm_vif {
 	struct iwl_mvm *mvm;
@@ -432,6 +439,8 @@ struct iwl_mvm_vif {
 		struct ieee80211_key_conf __rcu *keys[2];
 	} bcn_prot;
 
+	u32 fw_active_links_num;
+
 	struct iwl_mvm_vif_link_info deflink;
 	struct iwl_mvm_vif_link_info *link[IEEE80211_MLD_MAX_NUM_LINKS];
 };
@@ -896,6 +905,7 @@ struct iwl_mvm {
 	/* data related to data path */
 	struct iwl_rx_phy_info last_phy_info;
 	struct ieee80211_sta __rcu *fw_id_to_mac_id[IWL_MVM_STATION_COUNT_MAX];
+	unsigned long fw_link_ids_map;
 	u8 rx_ba_sessions;
 
 	/* configured by mac80211 */
-- 
cgit v1.2.3-70-g09d2


From 4263ac7fe5b2d6d5b3acc4c4192cebe7054bc623 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:07 +0300
Subject: wifi: iwlwifi: mvm: adjust to MLO
 assign/unassign/switch_vif_chanctx()

Pass link_conf when it's necessary and change the code accordingly.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.dc1074239c5b.I30db722af9009f1b274fa929f276fd2b78520dea@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 46 +++++++++++--------
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 52 +++++++++++++---------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       | 10 +----
 3 files changed, 61 insertions(+), 47 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index b90b0a973cec..82e7a66c5828 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4608,10 +4608,11 @@ out_unlock:
  * Returns true if we're done assigning the chanctx
  * (either on failure or success)
  */
-bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
-					 struct ieee80211_vif *vif,
-					 struct ieee80211_chanctx_conf *ctx,
-					 bool switching_chanctx, int *ret)
+static bool
+__iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_chanctx_conf *ctx,
+				    bool switching_chanctx, int *ret)
 {
 	u16 *phy_ctxt_id = (u16 *)ctx->drv_priv;
 	struct iwl_mvm_phy_ctxt *phy_ctxt = &mvm->phy_ctxts[*phy_ctxt_id];
@@ -4651,12 +4652,16 @@ bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
 
 static int __iwl_mvm_assign_vif_chanctx(struct iwl_mvm *mvm,
 					struct ieee80211_vif *vif,
+					struct ieee80211_bss_conf *link_conf,
 					struct ieee80211_chanctx_conf *ctx,
 					bool switching_chanctx)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
 
+	if (WARN_ON(!link_conf))
+		return -EINVAL;
+
 	if (__iwl_mvm_assign_vif_chanctx_common(mvm, vif, ctx,
 						switching_chanctx, &ret))
 		goto out;
@@ -4736,7 +4741,7 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	ret = __iwl_mvm_assign_vif_chanctx(mvm, vif, ctx, false);
+	ret = __iwl_mvm_assign_vif_chanctx(mvm, vif, link_conf, ctx, false);
 	mutex_unlock(&mvm->mutex);
 
 	return ret;
@@ -4748,9 +4753,9 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
  * Returns if chanctx unassign chanctx is done
  * (either on failure or success)
  */
-bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
-					   struct ieee80211_vif *vif,
-					   bool switching_chanctx)
+static bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
+						  struct ieee80211_vif *vif,
+						  bool switching_chanctx)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
@@ -4788,6 +4793,7 @@ bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
 
 static void __iwl_mvm_unassign_vif_chanctx(struct iwl_mvm *mvm,
 					   struct ieee80211_vif *vif,
+					   struct ieee80211_bss_conf *link_conf,
 					   struct ieee80211_chanctx_conf *ctx,
 					   bool switching_chanctx)
 {
@@ -4827,7 +4833,7 @@ static void iwl_mvm_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
 	mutex_lock(&mvm->mutex);
-	__iwl_mvm_unassign_vif_chanctx(mvm, vif, ctx, false);
+	__iwl_mvm_unassign_vif_chanctx(mvm, vif, link_conf, ctx, false);
 	mutex_unlock(&mvm->mutex);
 }
 
@@ -4839,7 +4845,8 @@ iwl_mvm_switch_vif_chanctx_swap(struct iwl_mvm *mvm,
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx, true);
+	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
+				    vifs[0].old_ctx, true);
 	__iwl_mvm_remove_chanctx(mvm, vifs[0].old_ctx);
 
 	ret = __iwl_mvm_add_chanctx(mvm, vifs[0].new_ctx);
@@ -4848,8 +4855,8 @@ iwl_mvm_switch_vif_chanctx_swap(struct iwl_mvm *mvm,
 		goto out_reassign;
 	}
 
-	ret = ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].new_ctx,
-					   true);
+	ret = ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
+					vifs[0].new_ctx, true);
 	if (ret) {
 		IWL_ERR(mvm,
 			"failed to assign new_ctx during channel switch\n");
@@ -4871,8 +4878,8 @@ out_reassign:
 		goto out_restart;
 	}
 
-	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx,
-				      true)) {
+	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
+				      vifs[0].old_ctx, true)) {
 		IWL_ERR(mvm, "failed to reassign old_ctx after failure.\n");
 		goto out_restart;
 	}
@@ -4897,10 +4904,11 @@ iwl_mvm_switch_vif_chanctx_reassign(struct iwl_mvm *mvm,
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx, true);
+	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
+				    vifs[0].old_ctx, true);
 
-	ret = ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].new_ctx,
-					   true);
+	ret = ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
+					vifs[0].new_ctx, true);
 	if (ret) {
 		IWL_ERR(mvm,
 			"failed to assign new_ctx during channel switch\n");
@@ -4910,8 +4918,8 @@ iwl_mvm_switch_vif_chanctx_reassign(struct iwl_mvm *mvm,
 	goto out;
 
 out_reassign:
-	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].old_ctx,
-				      true)) {
+	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
+				      vifs[0].old_ctx, true)) {
 		IWL_ERR(mvm, "failed to reassign old_ctx after failure.\n");
 		goto out_restart;
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index b861b5478e08..0c233d151231 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -220,17 +220,23 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 	mutex_unlock(&mvm->mutex);
 }
 
-static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
-					    struct ieee80211_vif *vif,
-					    struct ieee80211_chanctx_conf *ctx,
-					    bool switching_chanctx)
+static int
+__iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_bss_conf *link_conf,
+				 struct ieee80211_chanctx_conf *ctx,
+				 bool switching_chanctx)
 {
 	u16 *phy_ctxt_id = (u16 *)ctx->drv_priv;
 	struct iwl_mvm_phy_ctxt *phy_ctxt = &mvm->phy_ctxts[*phy_ctxt_id];
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned int link_id = link_conf->link_id;
 	int ret;
 
-	mvmvif->deflink.phy_ctxt = phy_ctxt;
+	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+		return -EINVAL;
+
+	mvmvif->link[link_id]->phy_ctxt = phy_ctxt;
 
 	if (switching_chanctx) {
 		/* reactivate if we turned this off during channel switch */
@@ -239,12 +245,12 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	}
 
 	/* send it first with phy context ID */
-	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf, 0, false);
+	ret = iwl_mvm_link_changed(mvm, vif, link_conf, 0, false);
 	if (ret)
 		goto out;
 
 	/* then activate */
-	ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+	ret = iwl_mvm_link_changed(mvm, vif, link_conf,
 				   LINK_CONTEXT_MODIFY_ACTIVE |
 				   LINK_CONTEXT_MODIFY_RATES_INFO,
 				   true);
@@ -258,8 +264,7 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	iwl_mvm_power_update_mac(mvm);
 
 	if (vif->type == NL80211_IFTYPE_MONITOR) {
-		ret = iwl_mvm_mld_add_snif_sta(mvm, vif,
-					       &vif->bss_conf);
+		ret = iwl_mvm_mld_add_snif_sta(mvm, vif, link_conf);
 		if (ret)
 			goto deactivate;
 	}
@@ -267,10 +272,10 @@ static int __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	return 0;
 
 deactivate:
-	iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
-			     LINK_CONTEXT_MODIFY_ACTIVE, false);
+	iwl_mvm_link_changed(mvm, vif, link_conf, LINK_CONTEXT_MODIFY_ACTIVE,
+			     false);
 out:
-	mvmvif->deflink.phy_ctxt = NULL;
+	mvmvif->link[link_id]->phy_ctxt = NULL;
 	iwl_mvm_power_update_mac(mvm);
 	return ret;
 }
@@ -284,18 +289,25 @@ static int iwl_mvm_mld_assign_vif_chanctx(struct ieee80211_hw *hw,
 	int ret;
 
 	mutex_lock(&mvm->mutex);
-	ret = __iwl_mvm_mld_assign_vif_chanctx(mvm, vif, ctx, false);
+	ret = __iwl_mvm_mld_assign_vif_chanctx(mvm, vif, link_conf, ctx, false);
 	mutex_unlock(&mvm->mutex);
 
 	return ret;
 }
 
-static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
-					       struct ieee80211_vif *vif,
-					       struct ieee80211_chanctx_conf *ctx,
-					       bool switching_chanctx)
+static void
+__iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_bss_conf *link_conf,
+				   struct ieee80211_chanctx_conf *ctx,
+				   bool switching_chanctx)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned int link_id = link_conf->link_id;
+
+	/* shouldn't happen, but verify link_id is valid before accessing */
+	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+		return;
 
 	if (vif->type == NL80211_IFTYPE_AP && switching_chanctx) {
 		mvmvif->csa_countdown = false;
@@ -312,12 +324,12 @@ static void __iwl_mvm_mld_unassign_vif_chanctx(struct iwl_mvm *mvm,
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		iwl_mvm_mld_rm_snif_sta(mvm, vif);
 
-	iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+	iwl_mvm_link_changed(mvm, vif, link_conf,
 			     LINK_CONTEXT_MODIFY_ACTIVE, false);
 
 	if (switching_chanctx)
 		return;
-	mvmvif->deflink.phy_ctxt = NULL;
+	mvmvif->link[link_id]->phy_ctxt = NULL;
 	iwl_mvm_power_update_mac(mvm);
 }
 
@@ -329,7 +341,7 @@ static void iwl_mvm_mld_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
 	mutex_lock(&mvm->mutex);
-	__iwl_mvm_mld_unassign_vif_chanctx(mvm, vif, ctx, false);
+	__iwl_mvm_mld_unassign_vif_chanctx(mvm, vif, link_conf, ctx, false);
 	mutex_unlock(&mvm->mutex);
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ad67a9b71807..811647c8aa71 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2365,10 +2365,12 @@ int iwl_mvm_post_channel_switch(struct ieee80211_hw *hw,
 struct iwl_mvm_switch_vif_chanctx_ops {
 	int (*__assign_vif_chanctx)(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif,
+				    struct ieee80211_bss_conf *link_conf,
 				    struct ieee80211_chanctx_conf *ctx,
 				    bool switching_chanctx);
 	void (*__unassign_vif_chanctx)(struct iwl_mvm *mvm,
 				       struct ieee80211_vif *vif,
+				       struct ieee80211_bss_conf *link_conf,
 				       struct ieee80211_chanctx_conf *ctx,
 				       bool switching_chanctx);
 };
@@ -2380,14 +2382,6 @@ iwl_mvm_switch_vif_chanctx_common(struct ieee80211_hw *hw,
 				  enum ieee80211_chanctx_switch_mode mode,
 				  struct iwl_mvm_switch_vif_chanctx_ops *ops);
 
-bool __iwl_mvm_assign_vif_chanctx_common(struct iwl_mvm *mvm,
-					 struct ieee80211_vif *vif,
-					 struct ieee80211_chanctx_conf *ctx,
-					 bool switching_chanctx, int *ret);
-bool __iwl_mvm_unassign_vif_chanctx_common(struct iwl_mvm *mvm,
-					   struct ieee80211_vif *vif,
-					   bool switching_chanctx);
-
 /* Channel info utils */
 static inline bool iwl_mvm_has_ultra_hb_channel(struct iwl_mvm *mvm)
 {
-- 
cgit v1.2.3-70-g09d2


From 2c9b92209fce94cbd586e46ef8ffb0faaf180525 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:08 +0300
Subject: wifi: iwlwifi: mvm: update iwl_mvm_tx_reclaim() for MLO

vif->bss_conf is used in this function only when TLC is not offloaded,
so not in MLO flow. Simplify the related "if" condition and call
iwl_mvm_hwrate_to_tx_status() only for driver rate scale.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.c6826d5b5477.Ib56ec6025c0da3a381aaf88e71085ce9b96a9e65@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 1737f2ad21ad..61a95ee3ef2d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -1976,9 +1976,11 @@ static void iwl_mvm_tx_reclaim(struct iwl_mvm *mvm, int sta_id, int tid,
 	 * possible (i.e. first MPDU in the aggregation wasn't acked)
 	 * Still it's important to update RS about sent vs. acked.
 	 */
-	if (!is_flush && skb_queue_empty(&reclaimed_skbs)) {
+	if (!is_flush && skb_queue_empty(&reclaimed_skbs) &&
+	    !iwl_mvm_has_tlc_offload(mvm)) {
 		struct ieee80211_chanctx_conf *chanctx_conf = NULL;
 
+		/* no TLC offload, so non-MLD mode */
 		if (mvmsta->vif)
 			chanctx_conf =
 				rcu_dereference(mvmsta->vif->bss_conf.chanctx_conf);
@@ -1989,11 +1991,8 @@ static void iwl_mvm_tx_reclaim(struct iwl_mvm *mvm, int sta_id, int tid,
 		tx_info->band = chanctx_conf->def.chan->band;
 		iwl_mvm_hwrate_to_tx_status(mvm->fw, rate, tx_info);
 
-		if (!iwl_mvm_has_tlc_offload(mvm)) {
-			IWL_DEBUG_TX_REPLY(mvm,
-					   "No reclaim. Update rs directly\n");
-			iwl_mvm_rs_tx_status(mvm, sta, tid, tx_info, false);
-		}
+		IWL_DEBUG_TX_REPLY(mvm, "No reclaim. Update rs directly\n");
+		iwl_mvm_rs_tx_status(mvm, sta, tid, tx_info, false);
 	}
 
 out:
-- 
cgit v1.2.3-70-g09d2


From 57974a55d995468a9a476e24693eb741c649b25f Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:09 +0300
Subject: wifi: iwlwifi: mvm: refactor iwl_mvm_mac_sta_state_common()

Move code handling specific state transitions into separate handlers
and adjust them for MLO. Adjust relevant callbacks as well.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.5a3f8a849723.I0670d20436858a1cd3c055e03c7528db81292811@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 430 +++++++++++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c  | 309 +++++++++++++---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  10 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c    |  28 +-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c      | 137 +++++--
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h      |   3 +-
 6 files changed, 671 insertions(+), 246 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 82e7a66c5828..f778ac69c671 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2013,12 +2013,13 @@ set_thresholds:
 }
 
 static void iwl_mvm_set_pkt_ext_from_he_ppe(struct iwl_mvm *mvm,
-					    struct ieee80211_sta *sta,
+					    struct ieee80211_link_sta *link_sta,
 					    struct iwl_he_pkt_ext_v2 *pkt_ext,
 					    bool inheritance)
 {
-	u8 nss = (sta->deflink.he_cap.ppe_thres[0] & IEEE80211_PPE_THRES_NSS_MASK) + 1;
-	u8 *ppe = &sta->deflink.he_cap.ppe_thres[0];
+	u8 nss = (link_sta->he_cap.ppe_thres[0] &
+		  IEEE80211_PPE_THRES_NSS_MASK) + 1;
+	u8 *ppe = &link_sta->he_cap.ppe_thres[0];
 	u8 ru_index_bitmap =
 		u8_get_bits(*ppe,
 			    IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK);
@@ -2098,31 +2099,35 @@ static void iwl_mvm_get_optimal_ppe_info(struct iwl_he_pkt_ext_v2 *pkt_ext,
 }
 
 /* Set the pkt_ext field according to PPE Thresholds element */
-int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm,
+			    struct ieee80211_link_sta *link_sta,
 			    struct iwl_he_pkt_ext_v2 *pkt_ext)
 {
 	u8 nominal_padding;
 	int i, ret = 0;
 
+	if (WARN_ON(!link_sta))
+		return -EINVAL;
+
 	/* Initialize the PPE thresholds to "None" (7), as described in Table
 	 * 9-262ac of 80211.ax/D3.0.
 	 */
 	memset(pkt_ext, IWL_HE_PKT_EXT_NONE,
 	       sizeof(struct iwl_he_pkt_ext_v2));
 
-	if (sta->deflink.eht_cap.has_eht) {
+	if (link_sta->eht_cap.has_eht) {
 		nominal_padding =
-			u8_get_bits(sta->deflink.eht_cap.eht_cap_elem.phy_cap_info[5],
+			u8_get_bits(link_sta->eht_cap.eht_cap_elem.phy_cap_info[5],
 				    IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK);
 
 		/* If PPE Thresholds exists, parse them into a FW-familiar
 		 * format.
 		 */
-		if (sta->deflink.eht_cap.eht_cap_elem.phy_cap_info[5] &
+		if (link_sta->eht_cap.eht_cap_elem.phy_cap_info[5] &
 		    IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT) {
-			u8 nss = (sta->deflink.eht_cap.eht_ppe_thres[0] &
+			u8 nss = (link_sta->eht_cap.eht_ppe_thres[0] &
 				IEEE80211_EHT_PPE_THRES_NSS_MASK) + 1;
-			u8 *ppe = &sta->deflink.eht_cap.eht_ppe_thres[0];
+			u8 *ppe = &link_sta->eht_cap.eht_ppe_thres[0];
 			u8 ru_index_bitmap =
 				u16_get_bits(*ppe,
 					     IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK);
@@ -2131,17 +2136,17 @@ int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 
 			iwl_mvm_parse_ppe(mvm, pkt_ext, nss, ru_index_bitmap,
 					  ppe, ppe_pos_bit, true);
-		/* EHT PPE Thresholds doesn't exist - set the API according
-		 * to HE PPE Tresholds
+		/* EHT PPE Thresholds doesn't exist - set the API according to
+		 * HE PPE Tresholds
 		 */
-		} else if (sta->deflink.he_cap.he_cap_elem.phy_cap_info[6] &
+		} else if (link_sta->he_cap.he_cap_elem.phy_cap_info[6] &
 			   IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
 			/* Even though HE Capabilities IE doesn't contain PPE
 			 * Thresholds for BW 320Mhz, thresholds for this BW will
 			 * be filled in with the same values as 160Mhz, due to
 			 * the inheritance, as required.
 			 */
-			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, sta, pkt_ext,
+			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, link_sta, pkt_ext,
 							true);
 
 			/* According to the requirements, for MCSs 12-13 the
@@ -2157,18 +2162,18 @@ int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			ret = iwl_mvm_set_pkt_ext_from_nominal_padding(pkt_ext,
 								       nominal_padding);
 		}
-	} else if (sta->deflink.he_cap.has_he) {
+	} else if (link_sta->he_cap.has_he) {
 		/* If PPE Thresholds exist, parse them into a FW-familiar format. */
-		if (sta->deflink.he_cap.he_cap_elem.phy_cap_info[6] &
+		if (link_sta->he_cap.he_cap_elem.phy_cap_info[6] &
 			IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
-			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, sta, pkt_ext,
+			iwl_mvm_set_pkt_ext_from_he_ppe(mvm, link_sta, pkt_ext,
 							false);
 		/* PPE Thresholds doesn't exist - set the API PPE values
 		 * according to Common Nominal Packet Padding field.
 		 */
 		} else {
 			nominal_padding =
-				u8_get_bits(sta->deflink.he_cap.he_cap_elem.phy_cap_info[9],
+				u8_get_bits(link_sta->he_cap.he_cap_elem.phy_cap_info[9],
 					    IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK);
 			if (nominal_padding != IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED)
 				ret = iwl_mvm_set_pkt_ext_from_nominal_padding(pkt_ext,
@@ -2244,9 +2249,11 @@ bool iwl_mvm_is_nic_ack_enabled(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 			       IEEE80211_HE_MAC_CAP2_ACK_EN));
 }
 
-__le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta)
+__le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta,
+				 struct ieee80211_link_sta *link_sta)
 {
-	u8 *mac_cap_info = &sta->deflink.he_cap.he_cap_elem.mac_cap_info[0];
+	u8 *mac_cap_info =
+		&link_sta->he_cap.he_cap_elem.mac_cap_info[0];
 	__le32 htc_flags = 0;
 
 	if (mac_cap_info[0] & IEEE80211_HE_MAC_CAP0_HTC_HE)
@@ -2339,10 +2346,10 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
 		flags |= STA_CTXT_HE_RU_2MHZ_BLOCK;
 
 	/* HTC flags */
-	sta_ctxt_cmd.htc_flags = iwl_mvm_get_sta_htc_flags(sta);
+	sta_ctxt_cmd.htc_flags = iwl_mvm_get_sta_htc_flags(sta, &sta->deflink);
 
 	/* PPE Thresholds */
-	if (!iwl_mvm_set_sta_pkt_ext(mvm, sta, &sta_ctxt_cmd.pkt_ext))
+	if (!iwl_mvm_set_sta_pkt_ext(mvm, &sta->deflink, &sta_ctxt_cmd.pkt_ext))
 		flags |= STA_CTXT_HE_PACKET_EXT;
 
 	if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] &
@@ -3497,6 +3504,253 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
 					    &callbacks);
 }
 
+/* FIXME: temporary making two assumptions in all sta handling functions:
+ *	(1) when setting sta state, the link exists and protected
+ *	(2) if a link is valid in sta then it's valid in vif (can
+ *	use same index in the link array)
+ */
+
+#define iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, update) do {		\
+	typeof(mvmvif) _mvmvif = mvmvif;					\
+	unsigned int _i;							\
+	for_each_mvm_vif_valid_link(_mvmvif, _i)				\
+		iwl_mvm_rs_rate_init((mvm), (sta),				\
+				     _mvmvif->link[_i]->phy_ctxt->channel->band,\
+				     (update));					\
+} while (0)
+
+#define IWL_MVM_MIN_BEACON_INTERVAL_TU 16
+
+static bool iwl_mvm_vif_conf_from_sta(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta)
+{
+	unsigned int i;
+
+	/* Beacon interval check - firmware will crash if the beacon
+	 * interval is less than 16. We can't avoid connecting at all,
+	 * so refuse the station state change, this will cause mac80211
+	 * to abandon attempts to connect to this AP, and eventually
+	 * wpa_s will blocklist the AP...
+	 */
+
+	for_each_set_bit(i, (unsigned long *)&sta->valid_links,
+			 IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_link_sta *link_sta =
+			rcu_dereference_protected(sta->link[i], 1);
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference_protected(vif->link_conf[i], 1);
+
+		if (!link_conf || !link_sta)
+			continue;
+
+		if (link_conf->beacon_int < IWL_MVM_MIN_BEACON_INTERVAL_TU) {
+			IWL_ERR(mvm,
+				"Beacon interval %d for AP %pM is too small\n",
+				link_conf->beacon_int, link_sta->addr);
+			return false;
+		}
+
+		link_conf->he_support = link_sta->he_cap.has_he;
+	}
+
+	return true;
+}
+
+static void iwl_mvm_vif_set_he_support(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_sta *sta,
+				       bool is_sta)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned int i;
+
+	for_each_set_bit(i, (unsigned long *)&sta->valid_links,
+			 IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_link_sta *link_sta =
+			rcu_dereference_protected(sta->link[i], 1);
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference_protected(vif->link_conf[i], 1);
+
+		if (!link_conf || !link_sta)
+			continue;
+
+		link_conf->he_support = link_sta->he_cap.has_he;
+
+		if (is_sta) {
+			mvmvif->link[i]->he_ru_2mhz_block = false;
+			if (link_sta->he_cap.has_he)
+				iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif);
+		}
+	}
+}
+
+static int
+iwl_mvm_sta_state_notexist_to_none(struct iwl_mvm *mvm,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_sta *sta,
+				   struct iwl_mvm_sta_state_ops *callbacks)
+{
+	unsigned int i;
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (vif->type == NL80211_IFTYPE_STATION &&
+	    !iwl_mvm_vif_conf_from_sta(mvm, vif, sta))
+		return -EINVAL;
+
+	if (sta->tdls &&
+	    (vif->p2p ||
+	     iwl_mvm_tdls_sta_count(mvm, NULL) == IWL_MVM_TDLS_STA_COUNT ||
+	     iwl_mvm_phy_ctx_count(mvm) > 1)) {
+		IWL_DEBUG_MAC80211(mvm, "refusing TDLS sta\n");
+		return -EBUSY;
+	}
+
+	ret = callbacks->add_sta(mvm, vif, sta);
+	if (sta->tdls && ret == 0) {
+		iwl_mvm_recalc_tdls_state(mvm, vif, true);
+		iwl_mvm_tdls_check_trigger(mvm, vif, sta->addr,
+					   NL80211_TDLS_SETUP);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(sta->link); i++) {
+		struct ieee80211_link_sta *link_sta;
+
+		link_sta = link_sta_dereference_protected(sta, i);
+		if (!link_sta)
+			continue;
+
+		link_sta->agg.max_rc_amsdu_len = 1;
+	}
+	ieee80211_sta_recalc_aggregates(sta);
+
+	return 0;
+}
+
+static int
+iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw,
+				struct iwl_mvm *mvm,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta,
+				struct iwl_mvm_sta_state_ops *callbacks)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	unsigned int i;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (vif->type == NL80211_IFTYPE_AP) {
+		iwl_mvm_vif_set_he_support(hw, vif, sta, false);
+		mvmvif->ap_assoc_sta_count++;
+		callbacks->mac_ctxt_changed(mvm, vif, false);
+
+		/* since the below is not for MLD API, it's ok to use
+		 * the default bss_conf
+		 */
+		if (!mvm->mld_api_is_used &&
+		    ((vif->bss_conf.he_support &&
+		      !iwlwifi_mod_params.disable_11ax) ||
+		    (vif->bss_conf.eht_support &&
+		     !iwlwifi_mod_params.disable_11be)))
+			iwl_mvm_cfg_he_sta(mvm, vif, mvm_sta->deflink.sta_id);
+	} else if (vif->type == NL80211_IFTYPE_STATION) {
+		iwl_mvm_vif_set_he_support(hw, vif, sta, true);
+
+		callbacks->mac_ctxt_changed(mvm, vif, false);
+
+		if (!mvm->mld_api_is_used)
+			goto out;
+
+		for_each_set_bit(i, (unsigned long *)&sta->valid_links,
+				 IEEE80211_MLD_MAX_NUM_LINKS) {
+			struct ieee80211_bss_conf *link_conf =
+				rcu_dereference_protected(vif->link_conf[i], 1);
+
+			if (WARN_ON(!link_conf))
+				return -EINVAL;
+
+			iwl_mvm_link_changed(mvm, vif, link_conf,
+					     LINK_CONTEXT_MODIFY_ALL &
+					     ~LINK_CONTEXT_MODIFY_ACTIVE,
+					     true);
+		}
+	}
+
+out:
+	iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, false);
+
+	return callbacks->update_sta(mvm, vif, sta);
+}
+
+static int
+iwl_mvm_sta_state_assoc_to_authorized(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta,
+				      struct iwl_mvm_sta_state_ops *callbacks)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+
+	lockdep_assert_held(&mvm->mutex);
+
+	/* we don't support TDLS during DCM */
+	if (iwl_mvm_phy_ctx_count(mvm) > 1)
+		iwl_mvm_teardown_tdls_peers(mvm);
+
+	if (sta->tdls) {
+		iwl_mvm_tdls_check_trigger(mvm, vif, sta->addr,
+					   NL80211_TDLS_ENABLE_LINK);
+	} else {
+		/* enable beacon filtering */
+		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
+
+		mvmvif->authorized = 1;
+
+		callbacks->mac_ctxt_changed(mvm, vif, false);
+		iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
+	}
+
+	iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, true);
+
+	return 0;
+}
+
+static int
+iwl_mvm_sta_state_authorized_to_assoc(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta,
+				      struct iwl_mvm_sta_state_ops *callbacks)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	/* once we move into assoc state, need to update rate scale to
+	 * disable using wide bandwidth
+	 */
+	iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, false);
+
+	if (!sta->tdls) {
+		/* Set this but don't call iwl_mvm_mac_ctxt_changed()
+		 * yet to avoid sending high prio again for a little
+		 * time.
+		 */
+		mvmvif->authorized = 0;
+
+		/* disable beacon filtering */
+		ret = iwl_mvm_disable_beacon_filter(mvm, vif, 0);
+		WARN_ON(ret &&
+			!test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED,
+				  &mvm->status));
+	}
+
+	return 0;
+}
+
 /* Common part for MLD and non-MLD modes */
 int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 				 struct ieee80211_vif *vif,
@@ -3508,15 +3762,12 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	unsigned int link_id;
 	int ret;
 
 	IWL_DEBUG_MAC80211(mvm, "station %pM state change %d->%d\n",
 			   sta->addr, old_state, new_state);
 
-	/* this would be a mac80211 bug ... but don't crash */
-	if (WARN_ON_ONCE(!mvmvif->deflink.phy_ctxt))
-		return test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED, &mvm->status) ? 0 : -EINVAL;
-
 	/*
 	 * If we are in a STA removal flow and in DQA mode:
 	 *
@@ -3547,48 +3798,25 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 	}
 
 	mutex_lock(&mvm->mutex);
+
+	/* this would be a mac80211 bug ... but don't crash */
+	for_each_mvm_vif_valid_link(mvmvif, link_id) {
+		if (WARN_ON_ONCE(!mvmvif->link[link_id]->phy_ctxt)) {
+			mutex_unlock(&mvm->mutex);
+			return test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED,
+					&mvm->status) ? 0 : -EINVAL;
+		}
+	}
+
 	/* track whether or not the station is associated */
 	mvm_sta->sta_state = new_state;
 
 	if (old_state == IEEE80211_STA_NOTEXIST &&
 	    new_state == IEEE80211_STA_NONE) {
-		/*
-		 * Firmware bug - it'll crash if the beacon interval is less
-		 * than 16. We can't avoid connecting at all, so refuse the
-		 * station state change, this will cause mac80211 to abandon
-		 * attempts to connect to this AP, and eventually wpa_s will
-		 * blocklist the AP...
-		 */
-		if (vif->type == NL80211_IFTYPE_STATION &&
-		    vif->bss_conf.beacon_int < 16) {
-			IWL_ERR(mvm,
-				"AP %pM beacon interval is %d, refusing due to firmware bug!\n",
-				sta->addr, vif->bss_conf.beacon_int);
-			ret = -EINVAL;
+		ret = iwl_mvm_sta_state_notexist_to_none(mvm, vif, sta,
+							 callbacks);
+		if (ret < 0)
 			goto out_unlock;
-		}
-
-		if (vif->type == NL80211_IFTYPE_STATION)
-			vif->bss_conf.he_support = sta->deflink.he_cap.has_he;
-
-		if (sta->tdls &&
-		    (vif->p2p ||
-		     iwl_mvm_tdls_sta_count(mvm, NULL) ==
-						IWL_MVM_TDLS_STA_COUNT ||
-		     iwl_mvm_phy_ctx_count(mvm) > 1)) {
-			IWL_DEBUG_MAC80211(mvm, "refusing TDLS sta\n");
-			ret = -EBUSY;
-			goto out_unlock;
-		}
-
-		ret = callbacks->add_sta(mvm, vif, sta);
-		if (sta->tdls && ret == 0) {
-			iwl_mvm_recalc_tdls_state(mvm, vif, true);
-			iwl_mvm_tdls_check_trigger(mvm, vif, sta->addr,
-						   NL80211_TDLS_SETUP);
-		}
-
-		sta->deflink.agg.max_rc_amsdu_len = 1;
 	} else if (old_state == IEEE80211_STA_NONE &&
 		   new_state == IEEE80211_STA_AUTH) {
 		/*
@@ -3600,84 +3828,16 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 		ret = 0;
 	} else if (old_state == IEEE80211_STA_AUTH &&
 		   new_state == IEEE80211_STA_ASSOC) {
-		if (vif->type == NL80211_IFTYPE_AP) {
-			vif->bss_conf.he_support = sta->deflink.he_cap.has_he;
-			mvmvif->ap_assoc_sta_count++;
-			callbacks->mac_ctxt_changed(mvm, vif, false);
-			if (!mvm->mld_api_is_used &&
-			    ((vif->bss_conf.he_support &&
-			     !iwlwifi_mod_params.disable_11ax) ||
-			    (vif->bss_conf.eht_support &&
-			     !iwlwifi_mod_params.disable_11be)))
-				iwl_mvm_cfg_he_sta(mvm, vif,
-						   mvm_sta->deflink.sta_id);
-		} else if (vif->type == NL80211_IFTYPE_STATION) {
-			vif->bss_conf.he_support = sta->deflink.he_cap.has_he;
-
-			mvmvif->deflink.he_ru_2mhz_block = false;
-			if (sta->deflink.he_cap.has_he)
-				iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif);
-
-			callbacks->mac_ctxt_changed(mvm, vif, false);
-
-			if (mvm->mld_api_is_used)
-				iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
-						     LINK_CONTEXT_MODIFY_ALL &
-						     ~LINK_CONTEXT_MODIFY_ACTIVE,
-						     true);
-		}
-
-		iwl_mvm_rs_rate_init(mvm, sta,
-				     mvmvif->deflink.phy_ctxt->channel->band,
-				     false);
-		ret = callbacks->update_sta(mvm, vif, sta);
+		ret = iwl_mvm_sta_state_auth_to_assoc(hw, mvm, vif, sta,
+						      callbacks);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTHORIZED) {
-		ret = 0;
-
-		/* we don't support TDLS during DCM */
-		if (iwl_mvm_phy_ctx_count(mvm) > 1)
-			iwl_mvm_teardown_tdls_peers(mvm);
-
-		if (sta->tdls) {
-			iwl_mvm_tdls_check_trigger(mvm, vif, sta->addr,
-						   NL80211_TDLS_ENABLE_LINK);
-		} else {
-			/* enable beacon filtering */
-			WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
-
-			mvmvif->authorized = 1;
-
-			callbacks->mac_ctxt_changed(mvm, vif, false);
-			iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
-		}
-
-		iwl_mvm_rs_rate_init(mvm, sta,
-				     mvmvif->deflink.phy_ctxt->channel->band,
-				     true);
+		ret = iwl_mvm_sta_state_assoc_to_authorized(mvm, vif, sta,
+							    callbacks);
 	} else if (old_state == IEEE80211_STA_AUTHORIZED &&
 		   new_state == IEEE80211_STA_ASSOC) {
-		/* once we move into assoc state, need to update rate scale to
-		 * disable using wide bandwidth
-		 */
-		iwl_mvm_rs_rate_init(mvm, sta,
-				     mvmvif->deflink.phy_ctxt->channel->band,
-				     false);
-		if (!sta->tdls) {
-			/*
-			 * Set this but don't call iwl_mvm_mac_ctxt_changed()
-			 * yet to avoid sending high prio again for a little
-			 * time.
-			 */
-			mvmvif->authorized = 0;
-
-			/* disable beacon filtering */
-			ret = iwl_mvm_disable_beacon_filter(mvm, vif, 0);
-			WARN_ON(ret &&
-				!test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED,
-					  &mvm->status));
-		}
-		ret = 0;
+		ret = iwl_mvm_sta_state_authorized_to_assoc(mvm, vif, sta,
+							    callbacks);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTH) {
 		if (vif->type == NL80211_IFTYPE_AP) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 3791f37afa78..7a01295949fa 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -50,18 +50,14 @@ static int iwl_mvm_mld_add_int_sta_to_fw(struct iwl_mvm *mvm,
  */
 static int iwl_mvm_mld_rm_sta_from_fw(struct iwl_mvm *mvm, u32 sta_id)
 {
-	struct ieee80211_sta *sta;
 	struct iwl_mvm_remove_sta_cmd rm_sta_cmd = {
 		.sta_id = cpu_to_le32(sta_id),
 	};
 	int ret;
 
-	sta = rcu_dereference_protected(mvm->fw_id_to_mac_id[sta_id],
-					lockdep_is_held(&mvm->mutex));
-
 	/* Note: internal stations are marked as error values */
-	if (!sta) {
-		IWL_ERR(mvm, "Invalid station id\n");
+	if (!rcu_access_pointer(mvm->fw_id_to_mac_id[sta_id])) {
+		IWL_ERR(mvm, "Invalid station id %d\n", sta_id);
 		return -EINVAL;
 	}
 
@@ -380,18 +376,20 @@ int iwl_mvm_mld_rm_aux_sta(struct iwl_mvm *mvm)
 
 /* send a cfg sta command to add/update a sta in firmware */
 static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
-			       struct ieee80211_vif *vif)
+			       struct ieee80211_vif *vif,
+			       struct ieee80211_link_sta *link_sta,
+			       struct ieee80211_bss_conf *link_conf,
+			       struct iwl_mvm_link_sta *mvm_link_sta)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_vif *mvm_vif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif_link_info *link_info =
+					mvm_vif->link[link_conf->link_id];
 	struct iwl_mvm_sta_cfg_cmd cmd = {
-		.sta_id = cpu_to_le32(mvm_sta->deflink.sta_id),
+		.sta_id = cpu_to_le32(mvm_link_sta->sta_id),
 		.station_type = cpu_to_le32(mvm_sta->sta_type),
 		.mfp = cpu_to_le32(sta->mfp),
 	};
-	/* FIXME: use proper link_id */
-	unsigned int link_id = 0;
-	struct iwl_mvm_vif_link_info *link_info = mvm_vif->link[link_id];
 	u32 agg_size = 0, mpdu_dens = 0;
 
 	/* when adding sta, link should exist in FW */
@@ -407,7 +405,7 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	if (mvm_sta->sta_state >= IEEE80211_STA_ASSOC)
 		cmd.assoc_id = cpu_to_le32(sta->aid);
 
-	switch (sta->deflink.rx_nss) {
+	switch (link_sta->rx_nss) {
 	case 1:
 		cmd.mimo = cpu_to_le32(0);
 		break;
@@ -433,7 +431,7 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		break;
 	}
 
-	mpdu_dens = iwl_mvm_get_sta_ampdu_dens(sta, &agg_size);
+	mpdu_dens = iwl_mvm_get_sta_ampdu_dens(link_sta, link_conf, &agg_size);
 	cmd.tx_ampdu_spacing = cpu_to_le32(mpdu_dens);
 	cmd.tx_ampdu_max_size = cpu_to_le32(agg_size);
 
@@ -443,17 +441,17 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		cmd.uapsd_acs = cpu_to_le32(iwl_mvm_get_sta_uapsd_acs(sta));
 	}
 
-	if (sta->deflink.he_cap.has_he) {
+	if (link_sta->he_cap.has_he) {
 		cmd.trig_rnd_alloc =
-			cpu_to_le32(vif->bss_conf.uora_exists ? 1 : 0);
+			cpu_to_le32(link_conf->uora_exists ? 1 : 0);
 
 		/* PPE Thresholds */
-		iwl_mvm_set_sta_pkt_ext(mvm, sta, &cmd.pkt_ext);
+		iwl_mvm_set_sta_pkt_ext(mvm, link_sta, &cmd.pkt_ext);
 
 		/* HTC flags */
-		cmd.htc_flags = iwl_mvm_get_sta_htc_flags(sta);
+		cmd.htc_flags = iwl_mvm_get_sta_htc_flags(sta, link_sta);
 
-		if (sta->deflink.he_cap.he_cap_elem.mac_cap_info[2] &
+		if (link_sta->he_cap.he_cap_elem.mac_cap_info[2] &
 		    IEEE80211_HE_MAC_CAP2_ACK_EN)
 			cmd.ack_enabled = cpu_to_le32(1);
 	}
@@ -461,72 +459,243 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	return iwl_mvm_mld_send_sta_cmd(mvm, &cmd);
 }
 
-int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			struct ieee80211_sta *sta)
+static void iwl_mvm_mld_sta_rm_all_sta_links(struct iwl_mvm *mvm,
+					     struct iwl_mvm_sta *mvm_sta)
+{
+	unsigned int link_id;
+
+	for (link_id = 0; link_id < ARRAY_SIZE(mvm_sta->link); link_id++) {
+		struct iwl_mvm_link_sta *link =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (!link)
+			continue;
+
+		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[link->sta_id], NULL);
+		RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
+
+		if (link != &mvm_sta->deflink)
+			kfree_rcu(link, rcu_head);
+	}
+}
+
+/* allocate all the links of a sta, called when the station is first added */
+static int iwl_mvm_mld_alloc_sta_links(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_sta *sta)
 {
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-	int sta_id, ret = 0;
+	struct iwl_mvm_link_sta *link;
+	unsigned int link_id;
+	u32 sta_id;
+	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+	for (link_id = 0; link_id < ARRAY_SIZE(sta->link); link_id++) {
+		if (!rcu_access_pointer(sta->link[link_id]) ||
+		    mvm_sta->link[link_id])
+			continue;
+
 		sta_id = iwl_mvm_find_free_sta_id(mvm,
 						  ieee80211_vif_type_p2p(vif));
-	else
-		sta_id = mvm_sta->deflink.sta_id;
 
-	if (sta_id == IWL_MVM_INVALID_STA)
-		return -ENOSPC;
+		if (sta_id == IWL_MVM_INVALID_STA) {
+			ret = -ENOSPC;
+			goto err;
+		}
 
-	spin_lock_init(&mvm_sta->lock);
+		if (rcu_access_pointer(sta->link[link_id]) == &sta->deflink) {
+			link = &mvm_sta->deflink;
+		} else {
+			link = kzalloc(sizeof(*link), GFP_KERNEL);
+			if (!link) {
+				ret = -ENOMEM;
+				goto err;
+			}
+		}
 
-	/* if this is a HW restart re-alloc existing queues */
-	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
-		struct iwl_mvm_int_sta tmp_sta = {
-			.sta_id = sta_id,
-			.type = mvm_sta->sta_type,
-		};
+		link->sta_id = sta_id;
+		rcu_assign_pointer(mvm_sta->link[link_id], link);
+		rcu_assign_pointer(mvm->fw_id_to_mac_id[link->sta_id], sta);
+	}
 
-		/* First add an empty station since allocating
-		 * a queue requires a valid station
-		 */
+	return 0;
+
+err:
+	iwl_mvm_mld_sta_rm_all_sta_links(mvm, mvm_sta);
+	return ret;
+}
+
+static void iwl_mvm_mld_set_ap_sta_id(struct ieee80211_sta *sta,
+				      struct iwl_mvm_vif_link_info *vif_link,
+				      struct iwl_mvm_link_sta *sta_link)
+{
+	if (!sta->tdls) {
+		WARN_ON(vif_link->ap_sta_id != IWL_MVM_INVALID_STA);
+		vif_link->ap_sta_id = sta_link->sta_id;
+	} else {
+		WARN_ON(vif_link->ap_sta_id == IWL_MVM_INVALID_STA);
+	}
+}
+
+/* FIXME: consider waiting for mac80211 to add the STA instead of allocating
+ * queues here
+ */
+static int iwl_mvm_alloc_sta_after_restart(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct ieee80211_link_sta *link_sta;
+	unsigned int link_id;
+	struct iwl_mvm_int_sta tmp_sta = {
+		.type = mvm_sta->sta_type,
+	};
+	int sta_id, ret;
+
+	/* First add an empty station since allocating a queue requires
+	 * a valid station. Since we need a link_id to allocate a station,
+	 * pick up the first valid one.
+	 */
+	for_each_sta_active_link(vif, sta, link_sta, link_id) {
+		struct iwl_mvm_vif_link_info *mvm_link;
+		struct ieee80211_bss_conf *link_conf =
+			link_conf_dereference_protected(vif, link_id);
+		struct iwl_mvm_link_sta *mvm_link_sta =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (!link_conf)
+			continue;
+
+		mvm_link = mvmvif->link[link_conf->link_id];
+
+		if (!mvm_link || !mvm_link_sta)
+			continue;
+
+		sta_id = mvm_link_sta->sta_id;
+		tmp_sta.sta_id = sta_id;
 		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, &tmp_sta,
 						    vif->bss_conf.bssid,
-						    mvmvif->id);
+						    mvm_link->fw_link_id);
 		if (ret)
 			return ret;
 
+		rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
 		iwl_mvm_realloc_queues_after_restart(mvm, sta);
-	} else {
-		ret = iwl_mvm_sta_init(mvm, vif, sta, sta_id,
-				       STATION_TYPE_PEER);
+
+		/* since we need only one station, no need to continue */
+		return 0;
 	}
 
-	ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif);
-	if (ret)
-		return ret;
+	/* no active link found */
+	return -EINVAL;
+}
 
-	if (vif->type == NL80211_IFTYPE_STATION) {
-		if (!sta->tdls) {
-			WARN_ON(mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA);
-			mvmvif->deflink.ap_sta_id = sta_id;
-		} else {
-			WARN_ON(mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA);
-		}
+int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			struct ieee80211_sta *sta)
+{
+	struct iwl_mvm_vif *mvm_vif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	unsigned long link_sta_added_to_fw = 0;
+	struct ieee80211_link_sta *link_sta;
+	int ret = 0;
+	unsigned int link_id;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
+		ret = iwl_mvm_mld_alloc_sta_links(mvm, vif, sta);
+		if (ret)
+			return ret;
 	}
 
-	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
+	spin_lock_init(&mvm_sta->lock);
+
+	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+		ret = iwl_mvm_alloc_sta_after_restart(mvm, vif, sta);
+	else
+		ret = iwl_mvm_sta_init(mvm, vif, sta, IWL_MVM_INVALID_STA,
+				       STATION_TYPE_PEER);
+	if (ret)
+		goto err;
+
+	/* at this stage sta link pointers are already allocated */
+	ret = iwl_mvm_mld_update_sta(mvm, vif, sta);
+
+	for_each_sta_active_link(vif, sta, link_sta, link_id) {
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference_protected(vif->link_conf[link_id], 1);
+		struct iwl_mvm_link_sta *mvm_link_sta =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (WARN_ON(!link_conf || !mvm_link_sta))
+			goto err;
+
+		ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif, link_sta, link_conf,
+					  mvm_link_sta);
+		if (ret)
+			goto err;
+
+		link_sta_added_to_fw |= BIT(link_id);
+
+		if (vif->type == NL80211_IFTYPE_STATION)
+			iwl_mvm_mld_set_ap_sta_id(sta, mvm_vif->link[link_id],
+						  mvm_link_sta);
+	}
 
 	return 0;
+
+err:
+	/* remove all already allocated stations in FW */
+	for_each_set_bit(link_id, &link_sta_added_to_fw,
+			 IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct iwl_mvm_link_sta *mvm_link_sta =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_link_sta->sta_id);
+	}
+
+	/* free all sta resources in the driver */
+	iwl_mvm_mld_sta_rm_all_sta_links(mvm, mvm_sta);
+	return ret;
 }
 
 int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta)
 {
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct ieee80211_link_sta *link_sta;
+	unsigned int link_id;
+	int ret = 0;
+
 	lockdep_assert_held(&mvm->mutex);
 
-	return iwl_mvm_mld_cfg_sta(mvm, sta, vif);
+	for_each_sta_active_link(vif, sta, link_sta, link_id) {
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference_protected(vif->link_conf[link_id], 1);
+		struct iwl_mvm_link_sta *mvm_link_sta =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (WARN_ON(!link_conf || !mvm_link_sta))
+			return -EINVAL;
+
+		ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif, link_sta, link_conf,
+					  mvm_link_sta);
+
+		if (ret) {
+			IWL_ERR(mvm, "Failed to update sta link %d\n", link_id);
+			break;
+		}
+	}
+
+	return ret;
 }
 
 static void iwl_mvm_mld_disable_sta_queues(struct iwl_mvm *mvm,
@@ -559,6 +728,8 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct ieee80211_link_sta *link_sta;
+	unsigned int link_id;
 	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -566,20 +737,38 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	kfree(mvm_sta->dup_data);
 
 	/* flush its queues here since we are freeing mvm_sta */
-	ret = iwl_mvm_flush_sta(mvm, mvm_sta, false);
-	if (ret)
-		return ret;
+	for_each_sta_active_link(vif, sta, link_sta, link_id) {
+		struct iwl_mvm_link_sta *mvm_link_sta =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (WARN_ON(!mvm_link_sta))
+			return -EINVAL;
+
+		ret = iwl_mvm_flush_sta_tids(mvm, mvm_link_sta->sta_id,
+					     0xffff);
+		if (ret)
+			return ret;
+	}
+
 	ret = iwl_mvm_wait_sta_queues_empty(mvm, mvm_sta);
 	if (ret)
 		return ret;
 
 	iwl_mvm_mld_disable_sta_queues(mvm, vif, sta);
 
-	if (iwl_mvm_sta_del(mvm, vif, sta, &ret))
-		return ret;
+	for_each_sta_active_link(vif, sta, link_sta, link_id) {
+		struct iwl_mvm_link_sta *mvm_link_sta =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (iwl_mvm_sta_del(mvm, vif, sta, mvm_link_sta, &ret))
+			return ret;
+
+		ret = iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_link_sta->sta_id);
+	}
 
-	ret = iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_sta->deflink.sta_id);
-	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id], NULL);
+	iwl_mvm_mld_sta_rm_all_sta_links(mvm, mvm_sta);
 
 	return ret;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 811647c8aa71..3a9ece67aff7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1606,10 +1606,14 @@ int iwl_mvm_flush_sta(struct iwl_mvm *mvm, void *sta, bool internal);
 int iwl_mvm_flush_sta_tids(struct iwl_mvm *mvm, u32 sta_id, u16 tids);
 
 /* Utils to extract sta related data */
-__le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta);
+__le32 iwl_mvm_get_sta_htc_flags(struct ieee80211_sta *sta,
+				 struct ieee80211_link_sta *link_sta);
 u8 iwl_mvm_get_sta_uapsd_acs(struct ieee80211_sta *sta);
-u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta, u32 *_agg_size);
-int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_link_sta *link_sta,
+			       struct ieee80211_bss_conf *link_conf,
+			       u32 *_agg_size);
+int iwl_mvm_set_sta_pkt_ext(struct iwl_mvm *mvm,
+			    struct ieee80211_link_sta *link_sta,
 			    struct iwl_he_pkt_ext_v2 *pkt_ext);
 
 void iwl_mvm_async_handlers_purge(struct iwl_mvm *mvm);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index a33673c6ae7d..778c9237fe87 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -641,18 +641,30 @@ int rs_fw_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 
 void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta)
 {
-	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
+	unsigned int link_id;
 
 	IWL_DEBUG_RATE(mvm, "create station rate scale window\n");
 
-	lq_sta->pers.drv = mvm;
-	lq_sta->pers.sta_id = mvmsta->deflink.sta_id;
-	lq_sta->pers.chains = 0;
-	memset(lq_sta->pers.chain_signal, 0, sizeof(lq_sta->pers.chain_signal));
-	lq_sta->pers.last_rssi = S8_MIN;
-	lq_sta->last_rate_n_flags = 0;
+	for (link_id = 0; link_id < ARRAY_SIZE(mvmsta->link); link_id++) {
+		struct iwl_lq_sta_rs_fw *lq_sta;
+		struct iwl_mvm_link_sta *link =
+			rcu_dereference_protected(mvmsta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+		if (!link)
+			continue;
+
+		lq_sta = &link->lq_sta.rs_fw;
+
+		lq_sta->pers.drv = mvm;
+		lq_sta->pers.sta_id = link->sta_id;
+		lq_sta->pers.chains = 0;
+		memset(lq_sta->pers.chain_signal, 0,
+		       sizeof(lq_sta->pers.chain_signal));
+		lq_sta->pers.last_rssi = S8_MIN;
+		lq_sta->last_rate_n_flags = 0;
 
 #ifdef CONFIG_MAC80211_DEBUGFS
-	lq_sta->pers.dbg_fixed_rate = 0;
+		lq_sta->pers.dbg_fixed_rate = 0;
 #endif
+	}
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 54ef749be488..eb57acaef96a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -51,26 +51,31 @@ int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm, enum nl80211_iftype iftype)
 }
 
 /* Calculate the ampdu density and max size */
-u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta, u32 *_agg_size)
+u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_link_sta *link_sta,
+			       struct ieee80211_bss_conf *link_conf,
+			       u32 *_agg_size)
 {
-	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	u32 agg_size = 0, mpdu_dens = 0;
 
-	if (sta->deflink.ht_cap.ht_supported)
-		mpdu_dens = sta->deflink.ht_cap.ampdu_density;
+	if (WARN_ON(!link_sta))
+		return 0;
+
+	if (link_sta->ht_cap.ht_supported)
+		mpdu_dens = link_sta->ht_cap.ampdu_density;
 
-	if (mvm_sta->vif->bss_conf.chandef.chan->band == NL80211_BAND_6GHZ) {
-		mpdu_dens = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+	if (link_conf->chandef.chan->band ==
+	    NL80211_BAND_6GHZ) {
+		mpdu_dens = le16_get_bits(link_sta->he_6ghz_capa.capa,
 					  IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START);
-		agg_size = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+		agg_size = le16_get_bits(link_sta->he_6ghz_capa.capa,
 					 IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP);
-	} else if (sta->deflink.vht_cap.vht_supported) {
-		agg_size = sta->deflink.vht_cap.cap &
+	} else if (link_sta->vht_cap.vht_supported) {
+		agg_size = link_sta->vht_cap.cap &
 			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK;
 		agg_size >>=
 			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT;
-	} else if (sta->deflink.ht_cap.ht_supported) {
-		agg_size = sta->deflink.ht_cap.ampdu_factor;
+	} else if (link_sta->ht_cap.ht_supported) {
+		agg_size = link_sta->ht_cap.ampdu_factor;
 	}
 
 	/* D6.0 10.12.2 A-MPDU length limit rules
@@ -81,10 +86,10 @@ u32 iwl_mvm_get_sta_ampdu_dens(struct ieee80211_sta *sta, u32 *_agg_size)
 	 * Maximum AMPDU Length Exponent Extension field in its HE
 	 * Capabilities element
 	 */
-	if (sta->deflink.he_cap.has_he)
+	if (link_sta->he_cap.has_he)
 		agg_size +=
-		    u8_get_bits(sta->deflink.he_cap.he_cap_elem.mac_cap_info[3],
-				IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK);
+			u8_get_bits(link_sta->he_cap.he_cap_elem.mac_cap_info[3],
+				    IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK);
 
 	/* Limit to max A-MPDU supported by FW */
 	if (agg_size > (STA_FLG_MAX_AGG_SIZE_4M >> STA_FLG_MAX_AGG_SIZE_SHIFT))
@@ -200,7 +205,9 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			cpu_to_le32(STA_FLG_MAX_AGG_SIZE_MSK |
 				    STA_FLG_AGG_MPDU_DENS_MSK);
 
-	mpdu_dens = iwl_mvm_get_sta_ampdu_dens(sta, &agg_size);
+	mpdu_dens = iwl_mvm_get_sta_ampdu_dens(&sta->deflink,
+					       &mvm_sta->vif->bss_conf,
+					       &agg_size);
 	add_sta_cmd.station_flags |=
 		cpu_to_le32(agg_size << STA_FLG_MAX_AGG_SIZE_SHIFT);
 	add_sta_cmd.station_flags |=
@@ -784,19 +791,35 @@ static int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 sta_id,
 
 static int iwl_mvm_get_queue_size(struct ieee80211_sta *sta)
 {
+	int max_size = IWL_DEFAULT_QUEUE_SIZE;
+	unsigned int link_id;
+
 	/* this queue isn't used for traffic (cab_queue) */
 	if (!sta)
 		return IWL_MGMT_QUEUE_SIZE;
 
-	/* support for 1k ba size */
-	if (sta->deflink.eht_cap.has_eht)
-		return IWL_DEFAULT_QUEUE_SIZE_EHT;
+	rcu_read_lock();
 
-	/* support for 256 ba size */
-	if (sta->deflink.he_cap.has_he)
-		return IWL_DEFAULT_QUEUE_SIZE_HE;
+	for (link_id = 0; link_id < ARRAY_SIZE(sta->link); link_id++) {
+		struct ieee80211_link_sta *link =
+			rcu_dereference(sta->link[link_id]);
 
-	return IWL_DEFAULT_QUEUE_SIZE;
+		if (!link)
+			continue;
+
+		/* support for 1k ba size */
+		if (link->eht_cap.has_eht &&
+		    max_size < IWL_DEFAULT_QUEUE_SIZE_EHT)
+			max_size = IWL_DEFAULT_QUEUE_SIZE_EHT;
+
+		/* support for 256 ba size */
+		if (link->he_cap.has_he &&
+		    max_size < IWL_DEFAULT_QUEUE_SIZE_HE)
+			max_size = IWL_DEFAULT_QUEUE_SIZE_HE;
+	}
+
+	rcu_read_unlock();
+	return max_size;
 }
 
 int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
@@ -804,6 +827,7 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 			    u8 sta_id, u8 tid, unsigned int timeout)
 {
 	int queue, size;
+	u32 sta_mask = 0;
 
 	if (tid == IWL_MAX_TID_COUNT) {
 		tid = IWL_MGMT_TID;
@@ -819,22 +843,45 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 	/* size needs to be power of 2 values for calculating read/write pointers */
 	size = rounddown_pow_of_two(size);
 
+	if (sta) {
+		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
+		unsigned int link_id;
+
+		for (link_id = 0;
+		     link_id < ARRAY_SIZE(mvmsta->link);
+		     link_id++) {
+			struct iwl_mvm_link_sta *link =
+				rcu_dereference_protected(mvmsta->link[link_id],
+							  lockdep_is_held(&mvm->mutex));
+
+			if (!link)
+				continue;
+
+			sta_mask |= BIT(link->sta_id);
+		}
+	} else {
+		sta_mask |= BIT(sta_id);
+	}
+
+	if (!sta_mask)
+		return -EINVAL;
+
 	do {
-		queue = iwl_trans_txq_alloc(mvm->trans, 0, BIT(sta_id),
+		queue = iwl_trans_txq_alloc(mvm->trans, 0, sta_mask,
 					    tid, size, timeout);
 
 		if (queue < 0)
 			IWL_DEBUG_TX_QUEUES(mvm,
-					    "Failed allocating TXQ of size %d for sta %d tid %d, ret: %d\n",
-					    size, sta_id, tid, queue);
+					    "Failed allocating TXQ of size %d for sta mask %x tid %d, ret: %d\n",
+					    size, sta_mask, tid, queue);
 		size /= 2;
 	} while (queue < 0 && size >= 16);
 
 	if (queue < 0)
 		return queue;
 
-	IWL_DEBUG_TX_QUEUES(mvm, "Enabling TXQ #%d for sta %d tid %d\n",
-			    queue, sta_id, tid);
+	IWL_DEBUG_TX_QUEUES(mvm, "Enabling TXQ #%d for sta mask 0x%x tid %d\n",
+			    queue, sta_mask, tid);
 
 	return queue;
 }
@@ -1657,16 +1704,28 @@ int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	lockdep_assert_held(&mvm->mutex);
 
-	mvm_sta->deflink.sta_id = sta_id;
-	rcu_assign_pointer(mvm_sta->link[0], &mvm_sta->deflink);
-
 	mvm_sta->mac_id_n_color = FW_CMD_ID_AND_COLOR(mvmvif->id,
 						      mvmvif->color);
 	mvm_sta->vif = vif;
-	if (!mvm->trans->trans_cfg->gen2)
-		mvm_sta->max_agg_bufsize = LINK_QUAL_AGG_FRAME_LIMIT_DEF;
-	else
-		mvm_sta->max_agg_bufsize = LINK_QUAL_AGG_FRAME_LIMIT_GEN2_DEF;
+
+	/* for MLD sta_id(s) should be allocated for each link before calling
+	 * this function
+	 */
+	if (!mvm->mld_api_is_used) {
+		if (WARN_ON(sta_id == IWL_MVM_INVALID_STA))
+			return -EINVAL;
+
+		mvm_sta->deflink.sta_id = sta_id;
+		rcu_assign_pointer(mvm_sta->link[0], &mvm_sta->deflink);
+
+		if (!mvm->trans->trans_cfg->gen2)
+			mvm_sta->max_agg_bufsize =
+				LINK_QUAL_AGG_FRAME_LIMIT_DEF;
+		else
+			mvm_sta->max_agg_bufsize =
+				LINK_QUAL_AGG_FRAME_LIMIT_GEN2_DEF;
+	}
+
 	mvm_sta->tt_tx_protection = false;
 	mvm_sta->sta_type = sta_type;
 
@@ -1926,11 +1985,12 @@ int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
  * with error or success
  */
 bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-		     struct ieee80211_sta *sta, int *ret)
+		     struct ieee80211_sta *sta,
+		     struct iwl_mvm_link_sta *mvm_link_sta, int *ret)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-	u8 sta_id = mvm_sta->deflink.sta_id;
+	u8 sta_id = mvm_link_sta->sta_id;
 
 	lockdep_assert_held(&mvm->mutex);
 
@@ -1956,8 +2016,7 @@ bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		*status = IWL_MVM_QUEUE_FREE;
 	}
 
-	if (vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->deflink.ap_sta_id == sta_id) {
+	if (vif->type == NL80211_IFTYPE_STATION) {
 		/* if associated - we can't remove the AP STA now */
 		if (vif->cfg.assoc)
 			return true;
@@ -2023,7 +2082,7 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 
 	iwl_mvm_disable_sta_queues(mvm, vif, sta);
 
-	if (iwl_mvm_sta_del(mvm, vif, sta, &ret))
+	if (iwl_mvm_sta_del(mvm, vif, sta, &mvm_sta->deflink, &ret))
 		return ret;
 
 	ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->deflink.sta_id);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 7148263dddf8..6d4db666368b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -490,7 +490,8 @@ void iwl_mvm_realloc_queues_after_restart(struct iwl_mvm *mvm,
 int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
 				  struct iwl_mvm_sta *mvm_sta);
 bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-		     struct ieee80211_sta *sta, int *ret);
+		     struct ieee80211_sta *sta,
+		     struct iwl_mvm_link_sta *mvm_link_sta, int *ret);
 int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 		   struct ieee80211_vif *vif,
 		   struct ieee80211_sta *sta);
-- 
cgit v1.2.3-70-g09d2


From 79faae3a408e2a03e550de2d60f9619902251248 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 28 Mar 2023 10:59:10 +0300
Subject: wifi: iwlwifi: mvm: adjust some cleanup functions to MLO

Clean up data for all allocated links of a vif.
Invalidate all link sta pointers.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.948a7bee8b09.Ic6b202959c158258d391ccce77f66e1281879690@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 32 ++++++++++++++++++-----
 1 file changed, 25 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index f778ac69c671..9e825e3cf630 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -983,18 +983,23 @@ static void iwl_mvm_cleanup_iterator(void *data, u8 *mac,
 {
 	struct iwl_mvm *mvm = data;
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned int link_id;
 
 	mvmvif->uploaded = false;
-	mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
 
 	spin_lock_bh(&mvm->time_event_lock);
 	iwl_mvm_te_clear_data(mvm, &mvmvif->time_event_data);
 	spin_unlock_bh(&mvm->time_event_lock);
 
-	mvmvif->deflink.phy_ctxt = NULL;
 	memset(&mvmvif->bf_data, 0, sizeof(mvmvif->bf_data));
-	memset(&mvmvif->deflink.probe_resp_data, 0,
-	       sizeof(mvmvif->deflink.probe_resp_data));
+
+	for_each_mvm_vif_valid_link(mvmvif, link_id) {
+		mvmvif->link[link_id]->ap_sta_id = IWL_MVM_INVALID_STA;
+		mvmvif->link[link_id]->fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
+		mvmvif->link[link_id]->phy_ctxt = NULL;
+		memset(&mvmvif->link[link_id]->probe_resp_data, 0,
+		       sizeof(mvmvif->link[link_id]->probe_resp_data));
+	}
 }
 
 static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm)
@@ -3251,6 +3256,7 @@ void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	unsigned int link_id;
 
 	/*
 	 * This is called before mac80211 does RCU synchronisation,
@@ -3259,12 +3265,24 @@ void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 	 * be able to find the station this way, and we don't rely
 	 * on further RCU synchronisation after the sta_state()
 	 * callback deleted the station.
+	 * Since there's mvm->mutex here, no need to have RCU lock for
+	 * mvm_sta->link access.
 	 */
 	mutex_lock(&mvm->mutex);
-	if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id]))
-		rcu_assign_pointer(mvm->fw_id_to_mac_id[mvm_sta->deflink.sta_id],
-				   ERR_PTR(-ENOENT));
+	for (link_id = 0; link_id < ARRAY_SIZE(mvm_sta->link); link_id++) {
+		struct iwl_mvm_link_sta *link_sta;
+		u32 sta_id;
+
+		if (!mvm_sta->link[link_id])
+			continue;
 
+		link_sta = rcu_dereference_protected(mvm_sta->link[link_id],
+						     lockdep_is_held(&mvm->mutex));
+		sta_id = link_sta->sta_id;
+		if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[sta_id]))
+			rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id],
+					   ERR_PTR(-ENOENT));
+	}
 	mutex_unlock(&mvm->mutex);
 }
 
-- 
cgit v1.2.3-70-g09d2


From bf976c814c864ec45fa32303e7080fc2e6efe4f2 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 28 Mar 2023 10:59:11 +0300
Subject: wifi: iwlwifi: mvm: implement link change ops

Implement the link change ops for links and stations.
Note that the stations one is empty for now as we only
have support for a single link so far, and then the
stations are created with the first link as deflink by
mac80211, so right now we don't really need anything.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230328104949.6186c5a37e99.Ifd00d3ee93356ddef273aa18f1e081cd8f2c84ae@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c      |  43 ++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  12 +-
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 145 +++++++++++++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   2 +
 4 files changed, 187 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 7a70b6ee65ac..0cd40672fade 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -3,6 +3,7 @@
  * Copyright (C) 2022 - 2023 Intel Corporation
  */
 #include "mvm.h"
+#include "time-event.h"
 
 static u32 iwl_mvm_get_free_fw_link_id(struct iwl_mvm *mvm,
 				       struct iwl_mvm_vif *mvm_vif)
@@ -108,6 +109,28 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	    mvmvif->fw_active_links_num >= IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM)
 		return -EINVAL;
 
+	if (changes & LINK_CONTEXT_MODIFY_ACTIVE) {
+		/* When activating a link, phy context should be valid;
+		 * when deactivating a link, it also should be valid since
+		 * the link was active before. So, do nothing in this case.
+		 * Since a link is added first with FW_CTXT_INVALID, then we
+		 * can get here in case it's removed before it was activated.
+		 */
+		if (!link_info->phy_ctxt)
+			return 0;
+
+		/* Catch early if driver tries to activate or deactivate a link
+		 * twice.
+		 */
+		WARN_ON_ONCE(active == link_info->active);
+
+		/* When deactivating a link session protection should
+		 * be stopped
+		 */
+		if (!active && vif->type == NL80211_IFTYPE_STATION)
+			iwl_mvm_stop_session_protection(mvm, vif);
+	}
+
 	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
 
 	/* The phy_id, link address and listen_lmac can be modified only until
@@ -248,3 +271,23 @@ int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	return ret;
 }
+
+/* link should be deactivated before removal, so in most cases we need to
+ * perform these two operations together
+ */
+int iwl_mvm_disable_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 struct ieee80211_bss_conf *link_conf)
+{
+	int ret;
+
+	ret = iwl_mvm_link_changed(mvm, vif, link_conf,
+				   LINK_CONTEXT_MODIFY_ACTIVE, false);
+	if (ret)
+		return ret;
+
+	ret = iwl_mvm_remove_link(mvm, vif, link_conf);
+	if (ret)
+		return ret;
+
+	return ret;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 9e825e3cf630..c60aff0df801 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -983,6 +983,7 @@ static void iwl_mvm_cleanup_iterator(void *data, u8 *mac,
 {
 	struct iwl_mvm *mvm = data;
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_probe_resp_data *probe_data;
 	unsigned int link_id;
 
 	mvmvif->uploaded = false;
@@ -993,13 +994,19 @@ static void iwl_mvm_cleanup_iterator(void *data, u8 *mac,
 
 	memset(&mvmvif->bf_data, 0, sizeof(mvmvif->bf_data));
 
+	mvmvif->fw_active_links_num = 0;
 	for_each_mvm_vif_valid_link(mvmvif, link_id) {
 		mvmvif->link[link_id]->ap_sta_id = IWL_MVM_INVALID_STA;
 		mvmvif->link[link_id]->fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
 		mvmvif->link[link_id]->phy_ctxt = NULL;
-		memset(&mvmvif->link[link_id]->probe_resp_data, 0,
-		       sizeof(mvmvif->link[link_id]->probe_resp_data));
+		mvmvif->link[link_id]->active = 0;
 	}
+
+	probe_data = rcu_dereference_protected(mvmvif->deflink.probe_resp_data,
+					       lockdep_is_held(&mvm->mutex));
+	if (probe_data)
+		kfree_rcu(probe_data, rcu_head);
+	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
 }
 
 static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm)
@@ -1455,7 +1462,6 @@ static bool iwl_mvm_mac_add_interface_common(struct iwl_mvm *mvm,
 	lockdep_assert_held(&mvm->mutex);
 
 	mvmvif->mvm = mvm;
-	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
 
 	/* the first link always points to the default one */
 	mvmvif->link[0] = &mvmvif->deflink;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 0c233d151231..7463cb8a271a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -14,7 +14,6 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	mutex_lock(&mvm->mutex);
 
 	mvmvif->mvm = mvm;
-	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
 
 	/* Not much to do here. The stack will not allow interface
 	 * types or combinations that we didn't advertise, so we
@@ -35,8 +34,10 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 
 	mvmvif->features |= hw->netdev_features;
 
-	/* the first link always points to the default one */
+	/* reset deflink MLO parameters */
 	mvmvif->deflink.fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
+	mvmvif->deflink.active = 0;
+	/* the first link always points to the default one */
 	mvmvif->link[0] = &mvmvif->deflink;
 
 	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
@@ -119,10 +120,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	goto out_unlock;
 
  out_remove_link:
-	/* Link needs to be deactivated before removal */
-	iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
-			     LINK_CONTEXT_MODIFY_ACTIVE, false);
-	iwl_mvm_remove_link(mvm, vif, &vif->bss_conf);
+	iwl_mvm_disable_link(mvm, vif, &vif->bss_conf);
  out_unref_phy:
 	iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
  out_free_bf:
@@ -198,14 +196,11 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 
 		/* P2P device uses only one link */
 		iwl_mvm_mld_rm_bcast_sta(mvm, vif, &vif->bss_conf);
-		/* Link needs to be deactivated before removal */
-		iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
-				     LINK_CONTEXT_MODIFY_ACTIVE, false);
-		iwl_mvm_remove_link(mvm, vif, &vif->bss_conf);
+		iwl_mvm_disable_link(mvm, vif, &vif->bss_conf);
 		iwl_mvm_phy_ctxt_unref(mvm, mvmvif->deflink.phy_ctxt);
 		mvmvif->deflink.phy_ctxt = NULL;
 	} else {
-		iwl_mvm_remove_link(mvm, vif, &vif->bss_conf);
+		iwl_mvm_disable_link(mvm, vif, &vif->bss_conf);
 	}
 
 	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
@@ -359,7 +354,10 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	if (ret)
 		goto out_unlock;
 
-	ret = iwl_mvm_link_changed(mvm, vif, link_conf, LINK_CONTEXT_MODIFY_ALL,
+	/* the link should be already activated when assigning chan context */
+	ret = iwl_mvm_link_changed(mvm, vif, link_conf,
+				   LINK_CONTEXT_MODIFY_ALL &
+				   ~LINK_CONTEXT_MODIFY_ACTIVE,
 				   true);
 	if (ret)
 		goto out_unlock;
@@ -839,6 +837,126 @@ static int iwl_mvm_mld_roc(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	return iwl_mvm_roc_common(hw, vif, channel, duration, type, &ops);
 }
+
+static int
+iwl_mvm_mld_change_vif_links(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     u16 old_links, u16 new_links,
+			     struct ieee80211_bss_conf *old[IEEE80211_MLD_MAX_NUM_LINKS])
+{
+	struct iwl_mvm_vif_link_info *new_link[IEEE80211_MLD_MAX_NUM_LINKS] = {};
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	u16 removed = old_links & ~new_links;
+	u16 added = new_links & ~old_links;
+	int err, i;
+
+	if (hweight16(new_links) > 2) {
+		return -EOPNOTSUPP;
+	} else if (hweight16(new_links) > 1) {
+		unsigned int n_active = 0;
+
+		for (i = 0; i < IEEE80211_MLD_MAX_NUM_LINKS; i++) {
+			struct ieee80211_bss_conf *link_conf;
+
+			link_conf = link_conf_dereference_protected(vif, i);
+			if (link_conf &&
+			    rcu_access_pointer(link_conf->chanctx_conf))
+				n_active++;
+		}
+
+		if (n_active > 1)
+			return -EOPNOTSUPP;
+	}
+
+	for (i = 0; i < IEEE80211_MLD_MAX_NUM_LINKS; i++) {
+		int r;
+
+		if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
+			break;
+
+		if (!(added & BIT(i)))
+			continue;
+		new_link[i] = kzalloc(sizeof(*new_link[i]), GFP_KERNEL);
+		if (!new_link[i]) {
+			err = -ENOMEM;
+			goto free;
+		}
+
+		new_link[i]->bcast_sta.sta_id = IWL_MVM_INVALID_STA;
+		new_link[i]->mcast_sta.sta_id = IWL_MVM_INVALID_STA;
+		new_link[i]->ap_sta_id = IWL_MVM_INVALID_STA;
+		new_link[i]->fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
+
+		for (r = 0; r < NUM_IWL_MVM_SMPS_REQ; r++)
+			new_link[i]->smps_requests[r] =
+				IEEE80211_SMPS_AUTOMATIC;
+	}
+
+	mutex_lock(&mvm->mutex);
+
+	if (old_links == 0) {
+		err = iwl_mvm_disable_link(mvm, vif, &vif->bss_conf);
+		if (err)
+			goto out_err;
+		mvmvif->link[0] = NULL;
+	}
+
+	for (i = 0; i < IEEE80211_MLD_MAX_NUM_LINKS; i++) {
+		if (removed & BIT(i)) {
+			struct ieee80211_bss_conf *link_conf = old[i];
+
+			err = iwl_mvm_disable_link(mvm, vif, link_conf);
+			if (err)
+				goto out_err;
+			kfree(mvmvif->link[i]);
+			mvmvif->link[i] = NULL;
+		}
+
+		if (added & BIT(i)) {
+			struct ieee80211_bss_conf *link_conf;
+
+			/* FIXME: allow use of sdata_dereference()? */
+			link_conf = rcu_dereference_protected(vif->link_conf[i],
+							      1);
+			if (WARN_ON(!link_conf))
+				continue;
+
+			if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART,
+				      &mvm->status))
+				mvmvif->link[i] = new_link[i];
+			new_link[i] = NULL;
+			err = iwl_mvm_add_link(mvm, vif, link_conf);
+			if (err)
+				goto out_err;
+		}
+	}
+
+	err = 0;
+	if (new_links == 0) {
+		mvmvif->link[0] = &mvmvif->deflink;
+		err = iwl_mvm_add_link(mvm, vif, &vif->bss_conf);
+	}
+
+out_err:
+	/* we really don't have a good way to roll back here ... */
+	mutex_unlock(&mvm->mutex);
+
+free:
+	for (i = 0; i < IEEE80211_MLD_MAX_NUM_LINKS; i++)
+		kfree(new_link[i]);
+	return err;
+}
+
+static int
+iwl_mvm_mld_change_sta_links(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_sta *sta,
+			     u16 old_links, u16 new_links)
+{
+	return 0;
+}
+
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.tx = iwl_mvm_mac_tx,
 	.wake_tx_queue = iwl_mvm_mac_wake_tx_queue,
@@ -928,4 +1046,7 @@ const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
 	.sta_add_debugfs = iwl_mvm_sta_add_debugfs,
 #endif
 	.set_hw_timestamp = iwl_mvm_set_hw_timestamp,
+
+	.change_vif_links = iwl_mvm_mld_change_vif_links,
+	.change_sta_links = iwl_mvm_mld_change_sta_links,
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 3a9ece67aff7..dfe1aff10548 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1824,6 +1824,8 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			 u32 changes, bool active);
 int iwl_mvm_remove_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			struct ieee80211_bss_conf *link_conf);
+int iwl_mvm_disable_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			 struct ieee80211_bss_conf *link_conf);
 
 /* AP and IBSS */
 bool iwl_mvm_start_ap_ibss_common(struct ieee80211_hw *hw,
-- 
cgit v1.2.3-70-g09d2


From ef2b47b89514ae2a7090d4552bae9229c5c54ca8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:07 +0300
Subject: wifi: iwlwifi: mvm: make some HW flags conditional

With newer MLD firmware API we no longer need beacon timing
information from the host, so disable TIMING_BEACON_ONLY.

If MLO is enabled (currently only for testing), then we must
not have DEAUTH_NEED_MGD_TX_PREP because mac80211 doesn't
support it yet, we'll have to fix that later.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.28ec631487bf.I6eb27c3d4c0289a5ec3682f573aae3424f45619f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index c60aff0df801..c76c6b1f0218 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -291,17 +291,28 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	ieee80211_hw_set(hw, SUPPORTS_PS);
 	ieee80211_hw_set(hw, SUPPORTS_DYNAMIC_PS);
 	ieee80211_hw_set(hw, AMPDU_AGGREGATION);
-	ieee80211_hw_set(hw, TIMING_BEACON_ONLY);
 	ieee80211_hw_set(hw, CONNECTION_MONITOR);
 	ieee80211_hw_set(hw, CHANCTX_STA_CSA);
 	ieee80211_hw_set(hw, SUPPORT_FAST_XMIT);
 	ieee80211_hw_set(hw, SUPPORTS_CLONED_SKBS);
 	ieee80211_hw_set(hw, SUPPORTS_AMSDU_IN_AMPDU);
 	ieee80211_hw_set(hw, NEEDS_UNIQUE_STA_ADDR);
-	ieee80211_hw_set(hw, DEAUTH_NEED_MGD_TX_PREP);
 	ieee80211_hw_set(hw, SUPPORTS_VHT_EXT_NSS_BW);
 	ieee80211_hw_set(hw, BUFF_MMPDU_TXQ);
 	ieee80211_hw_set(hw, STA_MMPDU_TXQ);
+
+	/* With MLD FW API, it tracks timing by itself,
+	 * no need for any timing from the host
+	 */
+	if (!mvm->mld_api_is_used)
+		ieee80211_hw_set(hw, TIMING_BEACON_ONLY);
+
+	/* We should probably have this, but mac80211
+	 * currently doesn't support it for MLO.
+	 */
+	if (!(hw->wiphy->flags & WIPHY_FLAG_SUPPORTS_MLO))
+		ieee80211_hw_set(hw, DEAUTH_NEED_MGD_TX_PREP);
+
 	/*
 	 * On older devices, enabling TX A-MSDU occasionally leads to
 	 * something getting messed up, the command read from the FIFO
-- 
cgit v1.2.3-70-g09d2


From 6e1b59565d7bacd7e4b91c81ec4b7780c81bfb07 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:08 +0300
Subject: wifi: iwlwifi: mvm: fix narrow RU check for MLO

This needs to apply to the link, not deflink.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.0347fa77d522.Ibdf7a40cbe86e64ce086f243330fc27f43aeb3aa@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index c76c6b1f0218..9eb95ed48d02 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3392,20 +3392,27 @@ static void iwl_mvm_check_he_obss_narrow_bw_ru_iter(struct wiphy *wiphy,
 	rcu_read_unlock();
 }
 
-static void iwl_mvm_check_he_obss_narrow_bw_ru(struct ieee80211_hw *hw,
-					       struct ieee80211_vif *vif)
+static void
+iwl_mvm_check_he_obss_narrow_bw_ru(struct ieee80211_hw *hw,
+				   struct ieee80211_vif *vif,
+				   unsigned int link_id,
+				   struct ieee80211_bss_conf *link_conf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_he_obss_narrow_bw_ru_data iter_data = {
 		.tolerated = true,
 	};
 
-	if (!(vif->bss_conf.chandef.chan->flags & IEEE80211_CHAN_RADAR)) {
-		mvmvif->deflink.he_ru_2mhz_block = false;
+	if (WARN_ON_ONCE(!link_conf->chandef.chan ||
+			 !mvmvif->link[link_id]))
+		return;
+
+	if (!(link_conf->chandef.chan->flags & IEEE80211_CHAN_RADAR)) {
+		mvmvif->link[link_id]->he_ru_2mhz_block = false;
 		return;
 	}
 
-	cfg80211_bss_iter(hw->wiphy, &vif->bss_conf.chandef,
+	cfg80211_bss_iter(hw->wiphy, &link_conf->chandef,
 			  iwl_mvm_check_he_obss_narrow_bw_ru_iter,
 			  &iter_data);
 
@@ -3413,7 +3420,7 @@ static void iwl_mvm_check_he_obss_narrow_bw_ru(struct ieee80211_hw *hw,
 	 * If there is at least one AP on radar channel that cannot
 	 * tolerate 26-tone RU UL OFDMA transmissions using HE TB PPDU.
 	 */
-	mvmvif->deflink.he_ru_2mhz_block = !iter_data.tolerated;
+	mvmvif->link[link_id]->he_ru_2mhz_block = !iter_data.tolerated;
 }
 
 static void iwl_mvm_reset_cca_40mhz_workaround(struct iwl_mvm *mvm,
@@ -3615,7 +3622,8 @@ static void iwl_mvm_vif_set_he_support(struct ieee80211_hw *hw,
 		if (is_sta) {
 			mvmvif->link[i]->he_ru_2mhz_block = false;
 			if (link_sta->he_cap.has_he)
-				iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif);
+				iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif, i,
+								   link_conf);
 		}
 	}
 }
-- 
cgit v1.2.3-70-g09d2


From 6c07b73e70c89269c534620b20d5946ec2d6e217 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:09 +0300
Subject: wifi: iwlwifi: mvm: skip MEI update for MLO

For now, if using MLO, skip the MEI update. It probably
needs new API for multiple channels.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.aeaa746b14e7.I42b084f3cdc20bb52cf867700a3c2e1f5e7b4ecb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 9eb95ed48d02..294e1c1de6b6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3464,7 +3464,6 @@ static void iwl_mvm_mei_host_associated(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mei_conn_info conn_info = {
 		.ssid_len = vif->cfg.ssid_len,
-		.channel = vif->bss_conf.chandef.chan->hw_value,
 	};
 
 	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
@@ -3473,6 +3472,12 @@ static void iwl_mvm_mei_host_associated(struct iwl_mvm *mvm,
 	if (!mvm->mei_registered)
 		return;
 
+	/* FIXME: MEI needs to be updated for MLO */
+	if (!vif->bss_conf.chandef.chan)
+		return;
+
+	conn_info.channel = vif->bss_conf.chandef.chan->hw_value;
+
 	switch (mvm_sta->pairwise_cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
 		conn_info.pairwise_cipher = IWL_MEI_CIPHER_TKIP;
-- 
cgit v1.2.3-70-g09d2


From b7e39d565fe1d32c6d8253340d577cc53fad03d8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:10 +0300
Subject: wifi: iwlwifi: mvm: use STA link address

For now we only support a single link, so just use the
deflink's address for the link address, instead of the
STA's (MLD) address, but use the link address anyway
in order to facilitate MLO connections.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.c853c8ced3ba.I2e1915d1090c526e6a4c718440b45a7192bbbb03@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 7a01295949fa..da9d2cefa509 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -398,9 +398,9 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 
 	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
 
-	/* For now the link addr is the same as the mld addr */
 	memcpy(&cmd.peer_mld_address, sta->addr, ETH_ALEN);
-	memcpy(&cmd.peer_link_address, sta->addr, ETH_ALEN);
+	/* FIXME: use the correct link */
+	memcpy(&cmd.peer_link_address, sta->deflink.addr, ETH_ALEN);
 
 	if (mvm_sta->sta_state >= IEEE80211_STA_ASSOC)
 		cmd.assoc_id = cpu_to_le32(sta->aid);
-- 
cgit v1.2.3-70-g09d2


From 42564a74507c450a0edc93f4f06b5537967283fe Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:11 +0300
Subject: wifi: iwlwifi: mvm: rs-fw: don't crash on missing channel

This is more of a workaround, with MLO we can get here with
the deflink not assigned. It's not critical right now that
we have this right, so WARN_ON_ONCE() and don't crash until
we can fix this area of the code.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.2b0d31bdb60b.I39d23c76eec16ac49f6ae3a6d5f7652041bde855@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 778c9237fe87..ed4200d9bd04 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -506,6 +506,9 @@ u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta)
 	const struct ieee80211_sta_vht_cap *vht_cap = &sta->deflink.vht_cap;
 	const struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
 
+	if (WARN_ON_ONCE(!mvmsta->vif->bss_conf.chandef.chan))
+		return IEEE80211_MAX_MPDU_LEN_VHT_3895;
+
 	if (mvmsta->vif->bss_conf.chandef.chan->band == NL80211_BAND_6GHZ) {
 		switch (le16_get_bits(sta->deflink.he_6ghz_capa.capa,
 				      IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN)) {
-- 
cgit v1.2.3-70-g09d2


From 3ca2207932fbe227c29fbfdb62f72b824a16ba8e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:12 +0300
Subject: wifi: iwlwifi: mvm: coex: start handling multiple links

Iterate all the links here and check which ones are active,
then adjust them accordingly.

This is still wrong as far as the RSSI event is concerned
(calling iwl_mvm_bt_coex_enable_rssi_event) since we call
that now multiple times, which could overwrite the data;
we need to either make that per link or call it only once,
but need to sort out first what the firmware will be doing
for beacon filtering/beacon abort in MLO.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.fe813fd8de5d.I216236717876510e51e4c21c7fac7691925443f1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/coex.c  | 89 +++++++++++++++++---------
 drivers/net/wireless/intel/iwlwifi/mvm/utils.c |  4 ++
 2 files changed, 64 insertions(+), 29 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
index daaa173aaed6..5a5b1128e75c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c
@@ -257,33 +257,35 @@ static void iwl_mvm_bt_coex_tcm_based_ci(struct iwl_mvm *mvm,
 	swap(data->primary, data->secondary);
 }
 
-/* must be called under rcu_read_lock */
-static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
-				      struct ieee80211_vif *vif)
+static void iwl_mvm_bt_notif_per_link(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct iwl_bt_iterator_data *data,
+				      unsigned int link_id)
 {
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_bt_iterator_data *data = _data;
-	struct iwl_mvm *mvm = data->mvm;
-	struct ieee80211_chanctx_conf *chanctx_conf;
 	/* default smps_mode is AUTOMATIC - only used for client modes */
 	enum ieee80211_smps_mode smps_mode = IEEE80211_SMPS_AUTOMATIC;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	u32 bt_activity_grading, min_ag_for_static_smps;
+	struct ieee80211_chanctx_conf *chanctx_conf;
+	struct iwl_mvm_vif_link_info *link_info;
+	struct ieee80211_bss_conf *link_conf;
 	int ave_rssi;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	switch (vif->type) {
-	case NL80211_IFTYPE_STATION:
-		break;
-	case NL80211_IFTYPE_AP:
-		if (!mvmvif->ap_ibss_active)
-			return;
-		break;
-	default:
+	link_info = mvmvif->link[link_id];
+	if (!link_info)
 		return;
-	}
 
-	chanctx_conf = rcu_dereference(vif->bss_conf.chanctx_conf);
+	link_conf = rcu_dereference(vif->link_conf[link_id]);
+	/* This can happen due to races: if we receive the notification
+	 * and have the mutex held, while mac80211 is stuck on our mutex
+	 * in the middle of removing the link.
+	 */
+	if (!link_conf)
+		return;
+
+	chanctx_conf = rcu_dereference(link_conf->chanctx_conf);
 
 	/* If channel context is invalid or not on 2.4GHz .. */
 	if ((!chanctx_conf ||
@@ -291,10 +293,10 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 		if (vif->type == NL80211_IFTYPE_STATION) {
 			/* ... relax constraints and disable rssi events */
 			iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
-					    smps_mode, 0);
-			iwl_mvm_bt_coex_reduced_txp(mvm,
-						    mvmvif->deflink.ap_sta_id,
+					    smps_mode, link_id);
+			iwl_mvm_bt_coex_reduced_txp(mvm, link_info->ap_sta_id,
 						    false);
+			/* FIXME: should this be per link? */
 			iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
 		}
 		return;
@@ -315,17 +317,18 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	if (!vif->cfg.assoc)
 		smps_mode = IEEE80211_SMPS_AUTOMATIC;
 
-	if (mvmvif->deflink.phy_ctxt &&
-	    (mvm->last_bt_notif.rrc_status & BIT(mvmvif->deflink.phy_ctxt->id)))
+	if (link_info->phy_ctxt &&
+	    (mvm->last_bt_notif.rrc_status & BIT(link_info->phy_ctxt->id)))
 		smps_mode = IEEE80211_SMPS_AUTOMATIC;
 
 	IWL_DEBUG_COEX(data->mvm,
-		       "mac %d: bt_activity_grading %d smps_req %d\n",
-		       mvmvif->id, bt_activity_grading, smps_mode);
+		       "mac %d link %d: bt_activity_grading %d smps_req %d\n",
+		       mvmvif->id, link_info->fw_link_id,
+		       bt_activity_grading, smps_mode);
 
 	if (vif->type == NL80211_IFTYPE_STATION)
 		iwl_mvm_update_smps(mvm, vif, IWL_MVM_SMPS_REQ_BT_COEX,
-				    smps_mode, 0);
+				    smps_mode, link_id);
 
 	/* low latency is always primary */
 	if (iwl_mvm_vif_low_latency(mvmvif)) {
@@ -354,6 +357,7 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 			data->secondary = chanctx_conf;
 		}
 
+		/* FIXME: TCM load per interface? or need something per link? */
 		if (data->primary == chanctx_conf)
 			data->primary_load = mvm->tcm.result.load[mvmvif->id];
 		else if (data->secondary == chanctx_conf)
@@ -371,6 +375,7 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 		/* if secondary is not NULL, it might be a GO */
 		data->secondary = chanctx_conf;
 
+	/* FIXME: TCM load per interface? or need something per link? */
 	if (data->primary == chanctx_conf)
 		data->primary_load = mvm->tcm.result.load[mvmvif->id];
 	else if (data->secondary == chanctx_conf)
@@ -385,8 +390,8 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	if (iwl_get_coex_type(mvm, vif) == BT_COEX_LOOSE_LUT ||
 	    mvm->cfg->bt_shared_single_ant || !vif->cfg.assoc ||
 	    le32_to_cpu(mvm->last_bt_notif.bt_activity_grading) == BT_OFF) {
-		iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->deflink.ap_sta_id,
-					    false);
+		iwl_mvm_bt_coex_reduced_txp(mvm, link_info->ap_sta_id, false);
+		/* FIXME: should this be per link? */
 		iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, false, 0);
 		return;
 	}
@@ -398,11 +403,11 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	if (!ave_rssi)
 		ave_rssi = -100;
 	if (ave_rssi > -IWL_MVM_BT_COEX_EN_RED_TXP_THRESH) {
-		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->deflink.ap_sta_id,
+		if (iwl_mvm_bt_coex_reduced_txp(mvm, link_info->ap_sta_id,
 						true))
 			IWL_ERR(mvm, "Couldn't send BT_CONFIG cmd\n");
 	} else if (ave_rssi < -IWL_MVM_BT_COEX_DIS_RED_TXP_THRESH) {
-		if (iwl_mvm_bt_coex_reduced_txp(mvm, mvmvif->deflink.ap_sta_id,
+		if (iwl_mvm_bt_coex_reduced_txp(mvm, link_info->ap_sta_id,
 						false))
 			IWL_ERR(mvm, "Couldn't send BT_CONFIG cmd\n");
 	}
@@ -411,6 +416,32 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
 	iwl_mvm_bt_coex_enable_rssi_event(mvm, vif, true, ave_rssi);
 }
 
+/* must be called under rcu_read_lock */
+static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac,
+				      struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_bt_iterator_data *data = _data;
+	struct iwl_mvm *mvm = data->mvm;
+	unsigned int link_id;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	switch (vif->type) {
+	case NL80211_IFTYPE_STATION:
+		break;
+	case NL80211_IFTYPE_AP:
+		if (!mvmvif->ap_ibss_active)
+			return;
+		break;
+	default:
+		return;
+	}
+
+	for (link_id = 0; link_id < IEEE80211_MLD_MAX_NUM_LINKS; link_id++)
+		iwl_mvm_bt_notif_per_link(mvm, vif, data, link_id);
+}
+
 static void iwl_mvm_bt_coex_notif_handle(struct iwl_mvm *mvm)
 {
 	struct iwl_bt_iterator_data data = {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index cf72b9377411..af31b09c3966 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -296,6 +296,10 @@ void iwl_mvm_update_smps(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		return;
 
 	mvmvif = iwl_mvm_vif_from_mac80211(vif);
+
+	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
+		return;
+
 	mvmvif->link[link_id]->smps_requests[req_type] = smps_request;
 	for (i = 0; i < NUM_IWL_MVM_SMPS_REQ; i++) {
 		if (mvmvif->link[link_id]->smps_requests[i] ==
-- 
cgit v1.2.3-70-g09d2


From 6a2a71e563d5cb53037d01785d3817650371213a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:13 +0300
Subject: wifi: iwlwifi: mvm: make a few warnings only trigger once

We're hitting these while starting to enable MLO in the
driver, but getting them each and every time isn't very
useful one way or the other. Make these warnings trigger
only once.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.f333741d3dcf.If063d4cfe8a583f0f980a1b0ae4e63e17ba4ddc9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c  | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 294e1c1de6b6..2b2221d7582b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -5591,7 +5591,7 @@ void iwl_mvm_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			continue;
 
 		/* make sure only TDLS peers or the AP are flushed */
-		WARN_ON(i != mvmvif->deflink.ap_sta_id && !sta->tdls);
+		WARN_ON_ONCE(i != mvmvif->deflink.ap_sta_id && !sta->tdls);
 
 		if (drop) {
 			if (iwl_mvm_flush_sta(mvm, mvmsta, false))
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index 154f145396ce..aa8b4fb9c201 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -213,8 +213,8 @@ void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
 	u32 sec_key_id = WIDE_ID(DATA_PATH_GROUP, SEC_KEY_CMD);
 	u8 sec_key_ver = iwl_fw_lookup_cmd_ver(mvm->fw, sec_key_id, 0);
 
-	if (WARN_ON(vif->type != NL80211_IFTYPE_STATION ||
-		    mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA))
+	if (WARN_ON_ONCE(vif->type != NL80211_IFTYPE_STATION ||
+			 mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA))
 		return;
 
 	if (!sec_key_ver)
-- 
cgit v1.2.3-70-g09d2


From ba9eef6ba636e72c00b565610d29d10984711d68 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:14 +0300
Subject: wifi: iwlwifi: mvm: adjust iwl_mvm_sec_key_remove_ap to MLO

It has to be done per link. We still don't support keys
configuration for several links, but the single vif link can
still point to a link different from deflink. For now handle
the removal of keys for the default link.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.19d729cc4654.I4ebe8e3eb5fc00a994761f7c0ad40107382705ca@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c     | 3 ++-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c      | 6 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 3 +--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h          | 3 ++-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c          | 2 +-
 5 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 2b2221d7582b..44542b00fa14 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2699,7 +2699,8 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 			if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART,
 				      &mvm->status)) {
 				/* first remove remaining keys */
-				iwl_mvm_sec_key_remove_ap(mvm, vif);
+				iwl_mvm_sec_key_remove_ap(mvm, vif,
+							  &mvmvif->deflink);
 
 				/*
 				 * Remove AP station now that
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index aa8b4fb9c201..44206e3e1a87 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -207,14 +207,14 @@ static void iwl_mvm_sec_key_remove_ap_iter(struct ieee80211_hw *hw,
 }
 
 void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
-			       struct ieee80211_vif *vif)
+			       struct ieee80211_vif *vif,
+			       struct iwl_mvm_vif_link_info *link)
 {
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	u32 sec_key_id = WIDE_ID(DATA_PATH_GROUP, SEC_KEY_CMD);
 	u8 sec_key_ver = iwl_fw_lookup_cmd_ver(mvm->fw, sec_key_id, 0);
 
 	if (WARN_ON_ONCE(vif->type != NL80211_IFTYPE_STATION ||
-			 mvmvif->deflink.ap_sta_id == IWL_MVM_INVALID_STA))
+			 link->ap_sta_id == IWL_MVM_INVALID_STA))
 		return;
 
 	if (!sec_key_ver)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 7463cb8a271a..be9b2efa5ef5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -538,14 +538,13 @@ static void iwl_mvm_mld_vif_delete_all_stas(struct iwl_mvm *mvm,
 	if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
 		return;
 
-	iwl_mvm_sec_key_remove_ap(mvm, vif);
-
 	for_each_mvm_vif_valid_link(mvmvif, i) {
 		struct iwl_mvm_vif_link_info *link = mvmvif->link[i];
 
 		if (!link)
 			continue;
 
+		iwl_mvm_sec_key_remove_ap(mvm, vif, link);
 		ret = iwl_mvm_mld_rm_sta_id(mvm, vif, link->ap_sta_id);
 		if (ret)
 			IWL_ERR(mvm, "failed to remove AP station\n");
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index dfe1aff10548..6421edd53404 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2326,7 +2326,8 @@ int iwl_mvm_sec_key_del(struct iwl_mvm *mvm,
 			struct ieee80211_sta *sta,
 			struct ieee80211_key_conf *keyconf);
 void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
-			       struct ieee80211_vif *vif);
+			       struct ieee80211_vif *vif,
+			       struct iwl_mvm_vif_link_info *link);
 
 int iwl_rfi_send_config_cmd(struct iwl_mvm *mvm,
 			    struct iwl_rfi_lut_entry *rfi_table);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index eb57acaef96a..f9f512a85ef1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2022,7 +2022,7 @@ bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			return true;
 
 		/* first remove remaining keys */
-		iwl_mvm_sec_key_remove_ap(mvm, vif);
+		iwl_mvm_sec_key_remove_ap(mvm, vif, &mvmvif->deflink);
 
 		/* unassoc - go ahead - remove the AP STA now */
 		mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
-- 
cgit v1.2.3-70-g09d2


From 5ae9daf1e31cc9a4c0898ea660ee96016f3c1729 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:15 +0300
Subject: wifi: iwlwifi: mvm: adjust radar detection to MLO

Make the decision about radar detection status depending
on each active MLO link.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.a61b1d3f681a.I27bb8726d8806ad725c22bc7256856b1b6636745@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/power.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 45a652cce491..2033d89dc9e5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -279,18 +279,24 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 static bool iwl_mvm_power_is_radar(struct ieee80211_vif *vif)
 {
 	struct ieee80211_chanctx_conf *chanctx_conf;
-	struct ieee80211_channel *chan;
+	struct ieee80211_bss_conf *link_conf;
 	bool radar_detect = false;
+	unsigned int link_id;
 
 	rcu_read_lock();
-	chanctx_conf = rcu_dereference(vif->bss_conf.chanctx_conf);
-	WARN_ON(!chanctx_conf);
-	if (chanctx_conf) {
-		chan = chanctx_conf->def.chan;
-		radar_detect = chan->flags & IEEE80211_CHAN_RADAR;
+	for_each_vif_active_link(vif, link_conf, link_id) {
+		chanctx_conf = rcu_dereference(link_conf->chanctx_conf);
+		if (WARN_ON(!chanctx_conf))
+			continue;
+
+		radar_detect = !!(chanctx_conf->def.chan->flags &
+				  IEEE80211_CHAN_RADAR);
+		if (radar_detect)
+			goto out;
 	}
-	rcu_read_unlock();
 
+out:
+	rcu_read_unlock();
 	return radar_detect;
 }
 
-- 
cgit v1.2.3-70-g09d2


From f53be9c4bef6858df46d7bd9246a72789e63d5fc Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:16 +0300
Subject: wifi: iwlwifi: mvm: adjust rs init to MLO

Rate scale initialization needs some parameters stored
separately for each link. Pass link_conf and link_sta
pointers and adjust the relevant code accordingly.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.e2f458198844.I98bf3ea807dd8ae8d703915ce9c01e7b7d5ccb42@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 36 ++++++++++++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c    | 36 ++++++++++++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c       | 17 +++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h       | 10 +++++--
 4 files changed, 71 insertions(+), 28 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 44542b00fa14..92567ab34b88 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3557,15 +3557,28 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
  *	(2) if a link is valid in sta then it's valid in vif (can
  *	use same index in the link array)
  */
+static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
+					   struct ieee80211_vif *vif,
+					   struct ieee80211_sta *sta,
+					   bool update)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	unsigned int link_id;
 
-#define iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, update) do {		\
-	typeof(mvmvif) _mvmvif = mvmvif;					\
-	unsigned int _i;							\
-	for_each_mvm_vif_valid_link(_mvmvif, _i)				\
-		iwl_mvm_rs_rate_init((mvm), (sta),				\
-				     _mvmvif->link[_i]->phy_ctxt->channel->band,\
-				     (update));					\
-} while (0)
+	for_each_mvm_vif_valid_link(mvmvif, link_id) {
+		struct ieee80211_bss_conf *conf =
+			rcu_dereference_protected(vif->link_conf[link_id], 1);
+		struct ieee80211_link_sta *link_sta =
+			rcu_dereference_protected(sta->link[link_id], 1);
+
+		if (!conf || !link_sta || !mvmvif->link[link_id]->phy_ctxt)
+			continue;
+
+		iwl_mvm_rs_rate_init(mvm, sta, conf, link_sta,
+				     mvmvif->link[link_id]->phy_ctxt->channel->band,
+				     update);
+	}
+}
 
 #define IWL_MVM_MIN_BEACON_INTERVAL_TU 16
 
@@ -3729,7 +3742,7 @@ iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw,
 	}
 
 out:
-	iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, false);
+	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, false);
 
 	return callbacks->update_sta(mvm, vif, sta);
 }
@@ -3762,7 +3775,7 @@ iwl_mvm_sta_state_assoc_to_authorized(struct iwl_mvm *mvm,
 		iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
 	}
 
-	iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, true);
+	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, true);
 
 	return 0;
 }
@@ -3781,7 +3794,7 @@ iwl_mvm_sta_state_authorized_to_assoc(struct iwl_mvm *mvm,
 	/* once we move into assoc state, need to update rate scale to
 	 * disable using wide bandwidth
 	 */
-	iwl_mvm_rs_rate_init_all_links(mvm, mvmvif, sta, false);
+	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, false);
 
 	if (!sta->tdls) {
 		/* Set this but don't call iwl_mvm_mac_ctxt_changed()
@@ -3950,6 +3963,7 @@ void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       IEEE80211_RC_SUPP_RATES_CHANGED |
 		       IEEE80211_RC_NSS_CHANGED))
 		iwl_mvm_rs_rate_init(mvm, sta,
+				     &vif->bss_conf, &sta->deflink,
 				     mvmvif->deflink.phy_ctxt->channel->band,
 				     true);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index ed4200d9bd04..d1a308a451cc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -500,16 +500,17 @@ out:
 	rcu_read_unlock();
 }
 
-u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta)
+u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta,
+			    struct ieee80211_bss_conf *link_conf,
+			    struct ieee80211_link_sta *link_sta)
 {
-	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	const struct ieee80211_sta_vht_cap *vht_cap = &sta->deflink.vht_cap;
-	const struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
+	const struct ieee80211_sta_vht_cap *vht_cap = &link_sta->vht_cap;
+	const struct ieee80211_sta_ht_cap *ht_cap = &link_sta->ht_cap;
 
-	if (WARN_ON_ONCE(!mvmsta->vif->bss_conf.chandef.chan))
+	if (WARN_ON_ONCE(!link_conf->chandef.chan))
 		return IEEE80211_MAX_MPDU_LEN_VHT_3895;
 
-	if (mvmsta->vif->bss_conf.chandef.chan->band == NL80211_BAND_6GHZ) {
+	if (link_conf->chandef.chan->band == NL80211_BAND_6GHZ) {
 		switch (le16_get_bits(sta->deflink.he_6ghz_capa.capa,
 				      IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN)) {
 		case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454:
@@ -546,16 +547,18 @@ u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta)
 }
 
 void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+		     struct ieee80211_bss_conf *link_conf,
+		     struct ieee80211_link_sta *link_sta,
 		     enum nl80211_band band, bool update)
 {
 	struct ieee80211_hw *hw = mvm->hw;
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_lq_sta_rs_fw *lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
 	u32 cmd_id = WIDE_ID(DATA_PATH_GROUP, TLC_MNG_CONFIG_CMD);
 	struct ieee80211_supported_band *sband = hw->wiphy->bands[band];
-	u16 max_amsdu_len = rs_fw_get_max_amsdu_len(sta);
+	u16 max_amsdu_len = rs_fw_get_max_amsdu_len(sta, link_conf, link_sta);
+	struct iwl_mvm_link_sta *mvm_link_sta;
+	struct iwl_lq_sta_rs_fw *lq_sta;
 	struct iwl_tlc_config_cmd_v4 cfg_cmd = {
-		.sta_id = mvmsta->deflink.sta_id,
 		.max_ch_width = update ?
 			rs_fw_bw_from_sta_bw(sta) : IWL_TLC_MNG_CH_WIDTH_20MHZ,
 		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, sta, sband)),
@@ -564,11 +567,24 @@ void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		.max_mpdu_len = iwl_mvm_is_csum_supported(mvm) ?
 				cpu_to_le16(max_amsdu_len) : 0,
 	};
-	int ret;
+	unsigned int link_id = link_conf->link_id;
 	int cmd_ver;
+	int ret;
 
+	rcu_read_lock();
+	mvm_link_sta = rcu_dereference(mvmsta->link[link_id]);
+	if (WARN_ON_ONCE(!mvm_link_sta)) {
+		rcu_read_unlock();
+		return;
+	}
+
+	cfg_cmd.sta_id = mvm_link_sta->sta_id;
+
+	lq_sta = &mvm_link_sta->lq_sta.rs_fw;
 	memset(lq_sta, 0, offsetof(typeof(*lq_sta), pers));
 
+	rcu_read_unlock();
+
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	iwl_mvm_reset_frame_stats(mvm);
 #endif
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 1b0e99377156..161a32929848 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -1487,9 +1487,11 @@ static void rs_set_amsdu_len(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			     enum rs_action scale_action)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
+	struct ieee80211_bss_conf *bss_conf = &mvmsta->vif->bss_conf;
 	int i;
 
-	sta->deflink.agg.max_amsdu_len = rs_fw_get_max_amsdu_len(sta);
+	sta->deflink.agg.max_amsdu_len =
+		rs_fw_get_max_amsdu_len(sta, bss_conf, &sta->deflink);
 
 	/*
 	 * In case TLC offload is not active amsdu_enabled is either 0xFFFF
@@ -1502,7 +1504,7 @@ static void rs_set_amsdu_len(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	else
 		mvmsta->amsdu_enabled = 0xFFFF;
 
-	if (mvmsta->vif->bss_conf.he_support &&
+	if (bss_conf->he_support &&
 	    !iwlwifi_mod_params.disable_11ax)
 		mvmsta->max_amsdu_len = sta->deflink.agg.max_amsdu_len;
 	else
@@ -3002,17 +3004,20 @@ static void rs_drv_rate_update(void *mvm_r,
 			       void *priv_sta, u32 changed)
 {
 	struct iwl_op_mode *op_mode = mvm_r;
+	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm *mvm __maybe_unused = IWL_OP_MODE_GET_MVM(op_mode);
 	u8 tid;
 
-	if (!iwl_mvm_sta_from_mac80211(sta)->vif)
+	if (!mvmsta->vif)
 		return;
 
 	/* Stop any ongoing aggregations as rs starts off assuming no agg */
 	for (tid = 0; tid < IWL_MAX_TID_COUNT; tid++)
 		ieee80211_stop_tx_ba_session(sta, tid);
 
-	iwl_mvm_rs_rate_init(mvm, sta, sband->band, true);
+	iwl_mvm_rs_rate_init(mvm, sta,
+			     &mvmsta->vif->bss_conf, &sta->deflink,
+			     sband->band, true);
 }
 
 static void __iwl_mvm_rs_tx_status(struct iwl_mvm *mvm,
@@ -4097,10 +4102,12 @@ static const struct rate_control_ops rs_mvm_ops_drv = {
 };
 
 void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			  struct ieee80211_bss_conf *link_conf,
+			  struct ieee80211_link_sta *link_sta,
 			  enum nl80211_band band, bool update)
 {
 	if (iwl_mvm_has_tlc_offload(mvm)) {
-		rs_fw_rate_init(mvm, sta, band, update);
+		rs_fw_rate_init(mvm, sta, link_conf, link_sta, band, update);
 	} else {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index b7bc8c1b2dda..c3145cb8f354 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -393,7 +393,9 @@ struct iwl_lq_sta {
 
 /* Initialize station's rate scaling information after adding station */
 void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
-			  enum nl80211_band band, bool init);
+			  struct ieee80211_bss_conf *link_conf,
+			  struct ieee80211_link_sta *link_sta,
+			  enum nl80211_band band, bool update);
 
 /* Notify RS about Tx status */
 void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
@@ -430,11 +432,15 @@ void iwl_mvm_reset_frame_stats(struct iwl_mvm *mvm);
 
 void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta);
 void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+		     struct ieee80211_bss_conf *link_conf,
+		     struct ieee80211_link_sta *link_sta,
 		     enum nl80211_band band, bool update);
 int rs_fw_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 			bool enable);
 void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 			      struct iwl_rx_cmd_buffer *rxb);
 
-u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta);
+u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta,
+			    struct ieee80211_bss_conf *link_conf,
+			    struct ieee80211_link_sta *link_sta);
 #endif /* __rs__ */
-- 
cgit v1.2.3-70-g09d2


From 29df2a64811914f8399fc2d1fb5b13df03336f50 Mon Sep 17 00:00:00 2001
From: Shaul Triebitz <shaul.triebitz@intel.com>
Date: Wed, 29 Mar 2023 10:05:17 +0300
Subject: wifi: iwlwifi: mvm: use the link sta address

Replace the deflink.addr with the proper link address
for setting the peer_link_address in the station command.
For a non-MLD station, it will be the deflink.

Signed-off-by: Shaul Triebitz <shaul.triebitz@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.03ab287da0ae.I88fb5ab4e3ea9c886a3fac7ce09c4791469c3c8e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index da9d2cefa509..c877c94dd1d6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -399,8 +399,7 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	cmd.link_id = cpu_to_le32(link_info->fw_link_id);
 
 	memcpy(&cmd.peer_mld_address, sta->addr, ETH_ALEN);
-	/* FIXME: use the correct link */
-	memcpy(&cmd.peer_link_address, sta->deflink.addr, ETH_ALEN);
+	memcpy(&cmd.peer_link_address, link_sta->addr, ETH_ALEN);
 
 	if (mvm_sta->sta_state >= IEEE80211_STA_ASSOC)
 		cmd.assoc_id = cpu_to_le32(sta->aid);
-- 
cgit v1.2.3-70-g09d2


From 6e4198d35c2e7e107328866c183e14d2b17c740b Mon Sep 17 00:00:00 2001
From: Shaul Triebitz <shaul.triebitz@intel.com>
Date: Wed, 29 Mar 2023 10:05:18 +0300
Subject: wifi: iwlwifi: mvm: implement mac80211 callback change_sta_links

Add/removed from iwl driver and firmware station links.
Update the station queues accordingly (which station links
are connected to the station queues).

Signed-off-by: Shaul Triebitz <shaul.triebitz@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.156d1aae5de1.I32973141be1190222169879f8caf7038c1a8f769@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |   9 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 246 +++++++++++++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |   4 +
 3 files changed, 236 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index be9b2efa5ef5..e475d0daf512 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -953,7 +953,14 @@ iwl_mvm_mld_change_sta_links(struct ieee80211_hw *hw,
 			     struct ieee80211_sta *sta,
 			     u16 old_links, u16 new_links)
 {
-	return 0;
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	int ret;
+
+	mutex_lock(&mvm->mutex);
+	ret = iwl_mvm_mld_update_sta_links(mvm, vif, sta, old_links, new_links);
+	mutex_unlock(&mvm->mutex);
+
+	return ret;
 }
 
 const struct ieee80211_ops iwl_mvm_mld_hw_ops = {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index c877c94dd1d6..9be6d858411d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -479,15 +479,53 @@ static void iwl_mvm_mld_sta_rm_all_sta_links(struct iwl_mvm *mvm,
 	}
 }
 
+static void iwl_mvm_mld_free_sta_link(struct iwl_mvm *mvm,
+				      struct iwl_mvm_sta *mvm_sta,
+				      struct iwl_mvm_link_sta *mvm_sta_link,
+				      unsigned int link_id)
+{
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta_link->sta_id], NULL);
+	RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
+
+	if (mvm_sta_link != &mvm_sta->deflink)
+		kfree_rcu(mvm_sta_link, rcu_head);
+}
+
+static int iwl_mvm_mld_alloc_sta_link(struct iwl_mvm *mvm,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_sta *sta,
+				      unsigned int link_id)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_link_sta *link;
+	u32 sta_id = iwl_mvm_find_free_sta_id(mvm,
+					  ieee80211_vif_type_p2p(vif));
+
+	if (sta_id == IWL_MVM_INVALID_STA)
+		return -ENOSPC;
+
+	if (rcu_access_pointer(sta->link[link_id]) == &sta->deflink) {
+		link = &mvm_sta->deflink;
+	} else {
+		link = kzalloc(sizeof(*link), GFP_KERNEL);
+		if (!link)
+			return -ENOMEM;
+	}
+
+	link->sta_id = sta_id;
+	rcu_assign_pointer(mvm_sta->link[link_id], link);
+	rcu_assign_pointer(mvm->fw_id_to_mac_id[link->sta_id], sta);
+
+	return 0;
+}
+
 /* allocate all the links of a sta, called when the station is first added */
 static int iwl_mvm_mld_alloc_sta_links(struct iwl_mvm *mvm,
 				       struct ieee80211_vif *vif,
 				       struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-	struct iwl_mvm_link_sta *link;
 	unsigned int link_id;
-	u32 sta_id;
 	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -497,27 +535,9 @@ static int iwl_mvm_mld_alloc_sta_links(struct iwl_mvm *mvm,
 		    mvm_sta->link[link_id])
 			continue;
 
-		sta_id = iwl_mvm_find_free_sta_id(mvm,
-						  ieee80211_vif_type_p2p(vif));
-
-		if (sta_id == IWL_MVM_INVALID_STA) {
-			ret = -ENOSPC;
+		ret = iwl_mvm_mld_alloc_sta_link(mvm, vif, sta, link_id);
+		if (ret)
 			goto err;
-		}
-
-		if (rcu_access_pointer(sta->link[link_id]) == &sta->deflink) {
-			link = &mvm_sta->deflink;
-		} else {
-			link = kzalloc(sizeof(*link), GFP_KERNEL);
-			if (!link) {
-				ret = -ENOMEM;
-				goto err;
-			}
-		}
-
-		link->sta_id = sta_id;
-		rcu_assign_pointer(mvm_sta->link[link_id], link);
-		rcu_assign_pointer(mvm->fw_id_to_mac_id[link->sta_id], sta);
 	}
 
 	return 0;
@@ -846,3 +866,185 @@ void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 
 	rcu_read_unlock();
 }
+
+static int iwl_mvm_mld_update_sta_queue(struct iwl_mvm *mvm,
+					struct iwl_mvm_sta *mvm_sta,
+					u32 old_sta_mask,
+					u32 new_sta_mask)
+{
+	struct iwl_scd_queue_cfg_cmd cmd = {
+		.operation = cpu_to_le32(IWL_SCD_QUEUE_MODIFY),
+		.u.modify.old_sta_mask = cpu_to_le32(old_sta_mask),
+		.u.modify.new_sta_mask = cpu_to_le32(new_sta_mask),
+	};
+	struct iwl_host_cmd hcmd = {
+		.id = WIDE_ID(DATA_PATH_GROUP, SCD_QUEUE_CONFIG_CMD),
+		.len[0] = sizeof(cmd),
+		.data[0] = &cmd
+	};
+	int tid;
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	for (tid = 0; tid <= IWL_MAX_TID_COUNT; tid++) {
+		struct iwl_mvm_tid_data *tid_data = &mvm_sta->tid_data[tid];
+		int txq_id = tid_data->txq_id;
+
+		if (txq_id == IWL_MVM_INVALID_QUEUE)
+			continue;
+
+		if (tid == IWL_MAX_TID_COUNT)
+			cmd.u.modify.tid = cpu_to_le32(IWL_MGMT_TID);
+		else
+			cmd.u.modify.tid = cpu_to_le32(tid);
+
+		ret = iwl_mvm_send_cmd(mvm, &hcmd);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_sta *sta,
+				 u16 old_links, u16 new_links)
+{
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_vif *mvm_vif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_link_sta *mvm_sta_link;
+	struct iwl_mvm_vif_link_info *mvm_vif_link;
+	unsigned long links_to_add = ~old_links & new_links;
+	unsigned long links_to_rem = old_links & ~new_links;
+	unsigned long old_links_long = old_links;
+	u32 current_sta_mask = 0, sta_mask_added = 0, sta_mask_to_rem = 0;
+	unsigned long link_sta_added_to_fw = 0, link_sta_allocated = 0;
+	unsigned int link_id;
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	for_each_set_bit(link_id, &old_links_long,
+			 IEEE80211_MLD_MAX_NUM_LINKS) {
+		mvm_sta_link =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (WARN_ON(!mvm_sta_link)) {
+			ret = -EINVAL;
+			goto err;
+		}
+
+		current_sta_mask |= BIT(mvm_sta_link->sta_id);
+		if (links_to_rem & BIT(link_id))
+			sta_mask_to_rem |= BIT(mvm_sta_link->sta_id);
+	}
+
+	if (sta_mask_to_rem) {
+		ret = iwl_mvm_mld_update_sta_queue(mvm, mvm_sta,
+						   current_sta_mask,
+						   current_sta_mask & ~sta_mask_to_rem);
+		if (WARN_ON(ret))
+			goto err;
+
+		current_sta_mask &= ~sta_mask_to_rem;
+	}
+
+	for_each_set_bit(link_id, &links_to_rem, IEEE80211_MLD_MAX_NUM_LINKS) {
+		mvm_sta_link =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+		mvm_vif_link = mvm_vif->link[link_id];
+
+		if (WARN_ON(!mvm_sta_link || !mvm_vif_link)) {
+			ret = -EINVAL;
+			goto err;
+		}
+
+		ret = iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_sta_link->sta_id);
+		if (WARN_ON(ret))
+			goto err;
+
+		if (vif->type == NL80211_IFTYPE_STATION)
+			mvm_vif_link->ap_sta_id = IWL_MVM_INVALID_STA;
+
+		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_sta_link, link_id);
+	}
+
+	for_each_set_bit(link_id, &links_to_add, IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference_protected(vif->link_conf[link_id], 1);
+		struct ieee80211_link_sta *link_sta =
+			rcu_dereference_protected(sta->link[link_id], 1);
+		mvm_vif_link = mvm_vif->link[link_id];
+
+		if (WARN_ON(!mvm_vif_link || !link_conf || !link_sta ||
+			    mvm_sta->link[link_id])) {
+			ret = -EINVAL;
+			goto err;
+		}
+
+		ret = iwl_mvm_mld_alloc_sta_link(mvm, vif, sta, link_id);
+		if (WARN_ON(ret))
+			goto err;
+
+		mvm_sta_link =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		if (WARN_ON(!mvm_sta_link)) {
+			ret = -EINVAL;
+			goto err;
+		}
+
+		if (vif->type == NL80211_IFTYPE_STATION)
+			iwl_mvm_mld_set_ap_sta_id(sta, mvm_vif_link,
+						  mvm_sta_link);
+
+		link_sta_allocated |= BIT(link_id);
+
+		sta_mask_added |= BIT(mvm_sta_link->sta_id);
+
+		ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif, link_sta, link_conf,
+					  mvm_sta_link);
+		if (WARN_ON(ret))
+			goto err;
+
+		link_sta_added_to_fw |= BIT(link_id);
+	}
+
+	if (sta_mask_added) {
+		ret = iwl_mvm_mld_update_sta_queue(mvm, mvm_sta,
+						   current_sta_mask,
+						   current_sta_mask | sta_mask_added);
+		if (WARN_ON(ret))
+			goto err;
+	}
+
+	return 0;
+
+err:
+	/* remove all already allocated stations in FW */
+	for_each_set_bit(link_id, &link_sta_added_to_fw,
+			 IEEE80211_MLD_MAX_NUM_LINKS) {
+		mvm_sta_link =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_sta_link->sta_id);
+	}
+
+	/* remove all already allocated station links in driver */
+	for_each_set_bit(link_id, &link_sta_allocated,
+			 IEEE80211_MLD_MAX_NUM_LINKS) {
+		mvm_sta_link =
+			rcu_dereference_protected(mvm_sta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+
+		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_sta_link, link_id);
+	}
+
+	return ret;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 6d4db666368b..2402b41bc0bb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -641,6 +641,10 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta);
 int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			  u8 sta_id);
+int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif,
+				 struct ieee80211_sta *sta,
+				 u16 old_links, u16 new_links);
 
 /* Queues */
 void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From 90723da664f53928c8a72c87b2b6d875ef657043 Mon Sep 17 00:00:00 2001
From: Shaul Triebitz <shaul.triebitz@intel.com>
Date: Wed, 29 Mar 2023 10:05:19 +0300
Subject: wifi: iwlwifi: mvm: translate management frame address

For management frames sent by an AP interface, translate
the MLD addresses to LINK addresses (for an MLD AP).
AP (non-bufferable) management frames are sent via
the broadcast station so the translation cannot be
done by the firmware.

Signed-off-by: Shaul Triebitz <shaul.triebitz@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.3cb4292f51e8.Ia662c00ff271c70eda927c12ed49b045b1eb8489@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 92567ab34b88..4c8ab1db1f19 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -722,6 +722,9 @@ void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	bool offchannel = IEEE80211_SKB_CB(skb)->flags &
 		IEEE80211_TX_CTL_TX_OFFCHAN;
+	u32 link_id = u32_get_bits(info->control.flags,
+				   IEEE80211_TX_CTRL_MLO_LINK);
+	struct ieee80211_sta *tmp_sta = sta;
 
 	if (iwl_mvm_is_radio_killed(mvm)) {
 		IWL_DEBUG_DROP(mvm, "Dropping - RF/CT KILL\n");
@@ -755,6 +758,25 @@ void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
 		}
 	}
 
+	if (tmp_sta && !sta && link_id != IEEE80211_LINK_UNSPECIFIED &&
+	    !ieee80211_is_probe_resp(hdr->frame_control)) {
+		/* translate MLD addresses to LINK addresses */
+		struct ieee80211_link_sta *link_sta =
+			rcu_dereference(tmp_sta->link[link_id]);
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference(info->control.vif->link_conf[link_id]);
+		struct ieee80211_mgmt *mgmt;
+
+		if (WARN_ON(!link_sta || !link_conf))
+			goto drop;
+
+		/* if sta is NULL, the frame is a management frame */
+		mgmt = (void *)hdr;
+		memcpy(mgmt->da, link_sta->addr, ETH_ALEN);
+		memcpy(mgmt->sa, link_conf->addr, ETH_ALEN);
+		memcpy(mgmt->bssid, link_conf->bssid, ETH_ALEN);
+	}
+
 	iwl_mvm_tx_skb(mvm, skb, sta);
 	return;
  drop:
-- 
cgit v1.2.3-70-g09d2


From 6e3ac4260bc8918919f5cedc7735fb8dcda22143 Mon Sep 17 00:00:00 2001
From: Shaul Triebitz <shaul.triebitz@intel.com>
Date: Wed, 29 Mar 2023 10:05:20 +0300
Subject: wifi: iwlwifi: mvm: use bcast/mcast link station id

For an MLD AP, use the correct link mcast or bcast station id.

Signed-off-by: Shaul Triebitz <shaul.triebitz@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.0cffa6c45242.I342e17e7bca87b7f05939eb2ebd36fc2aff0b49f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 61a95ee3ef2d..ca1284f11f8b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -739,10 +739,23 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 		if (info.control.vif->type == NL80211_IFTYPE_P2P_DEVICE ||
 		    info.control.vif->type == NL80211_IFTYPE_AP ||
 		    info.control.vif->type == NL80211_IFTYPE_ADHOC) {
+			u32 link_id = u32_get_bits(info.control.flags,
+						   IEEE80211_TX_CTRL_MLO_LINK);
+			struct iwl_mvm_vif_link_info *link;
+
+			if (link_id == IEEE80211_LINK_UNSPECIFIED) {
+				if (info.control.vif->active_links)
+					link_id = ffs(info.control.vif->active_links) - 1;
+				else
+					link_id = 0;
+			}
+
+			link = mvmvif->link[link_id];
+
 			if (!ieee80211_is_data(hdr->frame_control))
-				sta_id = mvmvif->deflink.bcast_sta.sta_id;
+				sta_id = link->bcast_sta.sta_id;
 			else
-				sta_id = mvmvif->deflink.mcast_sta.sta_id;
+				sta_id = link->mcast_sta.sta_id;
 
 			queue = iwl_mvm_get_ctrl_vif_queue(mvm, &info, hdr);
 		} else if (info.control.vif->type == NL80211_IFTYPE_MONITOR) {
-- 
cgit v1.2.3-70-g09d2


From b8a85a1d42d76916bd476a5c91bd7bc98751378f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:21 +0300
Subject: wifi: iwlwifi: mvm: rxmq: report link ID to mac80211

Add a fw_id_to_link_sta array in mvm to track the link
STA for each firmware station ID, and then use that to
report the link a frame was received on (since we know
the station ID from firmware).

Notably, this fixes beacon tracking for the correct link
since mac80211 now queues and processes those on the one
link identified by the link ID only.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.c7dd3ec18077.I12ef9eb4a5b8b5c2b9d6bcaa1fda73b59eba39d8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c       |  8 +++++--
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  4 +++-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c  |  8 +++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  1 +
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c     | 28 +++++++++++++++++------
 5 files changed, 39 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index d302d98b7631..7fe733dcc748 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1570,8 +1570,10 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	}
 
 	/* init the fw <-> mac80211 STA mapping */
-	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++)
+	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++) {
 		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[i], NULL);
+		RCU_INIT_POINTER(mvm->fw_id_to_link_sta[i], NULL);
+	}
 
 	memset(&mvm->fw_link_ids_map, 0, sizeof(mvm->fw_link_ids_map));
 
@@ -1757,8 +1759,10 @@ int iwl_mvm_load_d3_fw(struct iwl_mvm *mvm)
 		goto error;
 
 	/* init the fw <-> mac80211 STA mapping */
-	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++)
+	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++) {
 		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[i], NULL);
+		RCU_INIT_POINTER(mvm->fw_id_to_link_sta[i], NULL);
+	}
 
 	if (iwl_fw_lookup_cmd_ver(mvm->fw, ADD_STA, 0) < 12) {
 		/*
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 4c8ab1db1f19..568f0eaeb0fb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3319,9 +3319,11 @@ void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 		link_sta = rcu_dereference_protected(mvm_sta->link[link_id],
 						     lockdep_is_held(&mvm->mutex));
 		sta_id = link_sta->sta_id;
-		if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[sta_id]))
+		if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[sta_id])) {
 			rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id],
 					   ERR_PTR(-ENOENT));
+			RCU_INIT_POINTER(mvm->fw_id_to_link_sta[sta_id], NULL);
+		}
 	}
 	mutex_unlock(&mvm->mutex);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 9be6d858411d..a713b8a10781 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -472,6 +472,7 @@ static void iwl_mvm_mld_sta_rm_all_sta_links(struct iwl_mvm *mvm,
 			continue;
 
 		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[link->sta_id], NULL);
+		RCU_INIT_POINTER(mvm->fw_id_to_link_sta[link->sta_id], NULL);
 		RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
 
 		if (link != &mvm_sta->deflink)
@@ -485,6 +486,7 @@ static void iwl_mvm_mld_free_sta_link(struct iwl_mvm *mvm,
 				      unsigned int link_id)
 {
 	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta_link->sta_id], NULL);
+	RCU_INIT_POINTER(mvm->fw_id_to_link_sta[mvm_sta_link->sta_id], NULL);
 	RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
 
 	if (mvm_sta_link != &mvm_sta->deflink)
@@ -496,6 +498,8 @@ static int iwl_mvm_mld_alloc_sta_link(struct iwl_mvm *mvm,
 				      struct ieee80211_sta *sta,
 				      unsigned int link_id)
 {
+	struct ieee80211_link_sta *link_sta =
+		rcu_dereference_protected(sta->link[link_id], 1);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_link_sta *link;
 	u32 sta_id = iwl_mvm_find_free_sta_id(mvm,
@@ -515,6 +519,8 @@ static int iwl_mvm_mld_alloc_sta_link(struct iwl_mvm *mvm,
 	link->sta_id = sta_id;
 	rcu_assign_pointer(mvm_sta->link[link_id], link);
 	rcu_assign_pointer(mvm->fw_id_to_mac_id[link->sta_id], sta);
+	rcu_assign_pointer(mvm->fw_id_to_link_sta[link->sta_id],
+			   link_sta);
 
 	return 0;
 }
@@ -604,6 +610,7 @@ static int iwl_mvm_alloc_sta_after_restart(struct iwl_mvm *mvm,
 			return ret;
 
 		rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
+		rcu_assign_pointer(mvm->fw_id_to_link_sta[sta_id], link_sta);
 		iwl_mvm_realloc_queues_after_restart(mvm, sta);
 
 		/* since we need only one station, no need to continue */
@@ -800,6 +807,7 @@ int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	lockdep_assert_held(&mvm->mutex);
 
 	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta_id], NULL);
+	RCU_INIT_POINTER(mvm->fw_id_to_link_sta[sta_id], NULL);
 	return ret;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 6421edd53404..a7a7018edd2b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -905,6 +905,7 @@ struct iwl_mvm {
 	/* data related to data path */
 	struct iwl_rx_phy_info last_phy_info;
 	struct ieee80211_sta __rcu *fw_id_to_mac_id[IWL_MVM_STATION_COUNT_MAX];
+	struct ieee80211_link_sta __rcu *fw_id_to_link_sta[IWL_MVM_STATION_COUNT_MAX];
 	unsigned long fw_link_ids_map;
 	u8 rx_ba_sessions;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index b160ae137c4a..5d803e537b00 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -253,12 +253,22 @@ static void iwl_mvm_add_rtap_sniffer_config(struct iwl_mvm *mvm,
 static void iwl_mvm_pass_packet_to_mac80211(struct iwl_mvm *mvm,
 					    struct napi_struct *napi,
 					    struct sk_buff *skb, int queue,
-					    struct ieee80211_sta *sta)
+					    struct ieee80211_sta *sta,
+					    struct ieee80211_link_sta *link_sta)
 {
-	if (iwl_mvm_check_pn(mvm, skb, queue, sta))
+	if (unlikely(iwl_mvm_check_pn(mvm, skb, queue, sta))) {
 		kfree_skb(skb);
-	else
-		ieee80211_rx_napi(mvm->hw, sta, skb, napi);
+		return;
+	}
+
+	if (sta && sta->valid_links && link_sta) {
+		struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
+
+		rx_status->link_valid = 1;
+		rx_status->link_id = link_sta->link_id;
+	}
+
+	ieee80211_rx_napi(mvm->hw, sta, skb, napi);
 }
 
 static void iwl_mvm_get_signal_strength(struct iwl_mvm *mvm,
@@ -631,7 +641,7 @@ static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
 		while ((skb = __skb_dequeue(skb_list))) {
 			iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb,
 							reorder_buf->queue,
-							sta);
+							sta, NULL /* FIXME */);
 			reorder_buf->num_stored--;
 		}
 	}
@@ -2298,6 +2308,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 	u32 len;
 	u32 pkt_len = iwl_rx_packet_payload_len(pkt);
 	struct ieee80211_sta *sta = NULL;
+	struct ieee80211_link_sta *link_sta = NULL;
 	struct sk_buff *skb;
 	u8 crypt_len = 0;
 	size_t desc_size;
@@ -2454,6 +2465,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			sta = rcu_dereference(mvm->fw_id_to_mac_id[id]);
 			if (IS_ERR(sta))
 				sta = NULL;
+			link_sta = rcu_dereference(mvm->fw_id_to_link_sta[id]);
 		}
 	} else if (!is_multicast_ether_addr(hdr->addr2)) {
 		/*
@@ -2588,8 +2600,10 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 	}
 
 	if (!iwl_mvm_reorder(mvm, napi, queue, sta, skb, desc) &&
-	    (likely(!iwl_mvm_time_sync_frame(mvm, skb, hdr->addr2))))
-		iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, queue, sta);
+	    likely(!iwl_mvm_time_sync_frame(mvm, skb, hdr->addr2)))
+		iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, queue, sta,
+						link_sta);
+
 out:
 	rcu_read_unlock();
 }
-- 
cgit v1.2.3-70-g09d2


From 2f17227d65804ae83c20ecfd3175a56ee8671e63 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 29 Mar 2023 10:05:22 +0300
Subject: wifi: iwlwifi: mvm: adjust iwl_mvm_scan_respect_p2p_go_iter() for MLO

When looking for a GO on for setting the scan parameters, iterate
over all the available links.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.4eb25d5655d0.Ie69f7313e4337f78c262a835aea3f707273a4209@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 975edc02e637..07045092c717 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -2677,11 +2677,23 @@ static void iwl_mvm_scan_respect_p2p_go_iter(void *_data, u8 *mac,
 	if (vif == data->current_vif)
 		return;
 
-	if (vif->type == NL80211_IFTYPE_AP && vif->p2p &&
-	    mvmvif->deflink.phy_ctxt->id < NUM_PHY_CTX &&
-	    (data->band == NUM_NL80211_BANDS ||
-	     mvmvif->deflink.phy_ctxt->channel->band == data->band))
-		data->p2p_go = true;
+	if (vif->type == NL80211_IFTYPE_AP && vif->p2p) {
+		u32 link_id;
+
+		for (link_id = 0;
+		     link_id < ARRAY_SIZE(mvmvif->link);
+		     link_id++) {
+			struct iwl_mvm_vif_link_info *link =
+				mvmvif->link[link_id];
+
+			if (link && link->phy_ctxt->id < NUM_PHY_CTX &&
+			    (data->band == NUM_NL80211_BANDS ||
+			     link->phy_ctxt->channel->band == data->band)) {
+				data->p2p_go = true;
+				break;
+			}
+		}
+	}
 }
 
 static bool _iwl_mvm_get_respect_p2p_go(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From f14ad95a5d9086efba664eeb9f1f02fd8bc08c65 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:23 +0300
Subject: wifi: iwlwifi: mvm: skip inactive links

When iterating station links, skip the links that are not
yet active by checking for mvmvif->link[] existence.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.bd9b4e64c478.Ie21422c3bf2589d22942c3c57d26e6330d2e3afc@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 568f0eaeb0fb..1fd7e5bf594d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3657,7 +3657,7 @@ static void iwl_mvm_vif_set_he_support(struct ieee80211_hw *hw,
 		struct ieee80211_bss_conf *link_conf =
 			rcu_dereference_protected(vif->link_conf[i], 1);
 
-		if (!link_conf || !link_sta)
+		if (!link_conf || !link_sta || !mvmvif->link[i])
 			continue;
 
 		link_conf->he_support = link_sta->he_cap.has_he;
@@ -3757,6 +3757,8 @@ iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw,
 
 			if (WARN_ON(!link_conf))
 				return -EINVAL;
+			if (!mvmvif->link[i])
+				continue;
 
 			iwl_mvm_link_changed(mvm, vif, link_conf,
 					     LINK_CONTEXT_MODIFY_ALL &
-- 
cgit v1.2.3-70-g09d2


From 072573f6971dcedeb0e221259a41718e42154c8f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:24 +0300
Subject: wifi: iwlwifi: mvm: remove only link-specific AP keys

When we remove the AP station, we iterate over the links
and remove all the keys, however, the key iteration will
return all keys for all links, so skip the ones that we
don't need based on the link ID.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.e724878f502e.I66870d4629244b4b309be79e11cbbd384bdf93be@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c     | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c      | 9 +++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h          | 3 ++-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c          | 2 +-
 5 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 1fd7e5bf594d..98c940271648 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2722,7 +2722,7 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 				      &mvm->status)) {
 				/* first remove remaining keys */
 				iwl_mvm_sec_key_remove_ap(mvm, vif,
-							  &mvmvif->deflink);
+							  &mvmvif->deflink, 0);
 
 				/*
 				 * Remove AP station now that
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index 44206e3e1a87..f4785c0a0b84 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -195,6 +195,7 @@ static void iwl_mvm_sec_key_remove_ap_iter(struct ieee80211_hw *hw,
 					   void *data)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	unsigned int link_id = (uintptr_t)data;
 
 	if (key->hw_key_idx == STA_KEY_IDX_INVALID)
 		return;
@@ -202,13 +203,17 @@ static void iwl_mvm_sec_key_remove_ap_iter(struct ieee80211_hw *hw,
 	if (sta)
 		return;
 
+	if (key->link_id >= 0 && key->link_id != link_id)
+		return;
+
 	_iwl_mvm_sec_key_del(mvm, vif, NULL, key, CMD_ASYNC);
 	key->hw_key_idx = STA_KEY_IDX_INVALID;
 }
 
 void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
 			       struct ieee80211_vif *vif,
-			       struct iwl_mvm_vif_link_info *link)
+			       struct iwl_mvm_vif_link_info *link,
+			       unsigned int link_id)
 {
 	u32 sec_key_id = WIDE_ID(DATA_PATH_GROUP, SEC_KEY_CMD);
 	u8 sec_key_ver = iwl_fw_lookup_cmd_ver(mvm->fw, sec_key_id, 0);
@@ -222,5 +227,5 @@ void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
 
 	ieee80211_iter_keys_rcu(mvm->hw, vif,
 				iwl_mvm_sec_key_remove_ap_iter,
-				NULL);
+				(void *)(uintptr_t)link_id);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index e475d0daf512..1a97bf925817 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -544,7 +544,7 @@ static void iwl_mvm_mld_vif_delete_all_stas(struct iwl_mvm *mvm,
 		if (!link)
 			continue;
 
-		iwl_mvm_sec_key_remove_ap(mvm, vif, link);
+		iwl_mvm_sec_key_remove_ap(mvm, vif, link, i);
 		ret = iwl_mvm_mld_rm_sta_id(mvm, vif, link->ap_sta_id);
 		if (ret)
 			IWL_ERR(mvm, "failed to remove AP station\n");
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index a7a7018edd2b..23d797247915 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2328,7 +2328,8 @@ int iwl_mvm_sec_key_del(struct iwl_mvm *mvm,
 			struct ieee80211_key_conf *keyconf);
 void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
 			       struct ieee80211_vif *vif,
-			       struct iwl_mvm_vif_link_info *link);
+			       struct iwl_mvm_vif_link_info *link,
+			       unsigned int link_id);
 
 int iwl_rfi_send_config_cmd(struct iwl_mvm *mvm,
 			    struct iwl_rfi_lut_entry *rfi_table);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index f9f512a85ef1..9a3a7eabded8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2022,7 +2022,7 @@ bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			return true;
 
 		/* first remove remaining keys */
-		iwl_mvm_sec_key_remove_ap(mvm, vif, &mvmvif->deflink);
+		iwl_mvm_sec_key_remove_ap(mvm, vif, &mvmvif->deflink, 0);
 
 		/* unassoc - go ahead - remove the AP STA now */
 		mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
-- 
cgit v1.2.3-70-g09d2


From 8930ed56734d2975f954e0447e65472c142aa88b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:25 +0300
Subject: wifi: iwlwifi: mvm: avoid sending MAC context for idle

If we change only idle, avoid sending the MAC context
command since it doesn't depend on idle state. This
also fixes an issue with the firmware because without
this we send the command during link switching, as we
just deactivated the first link, and we cannot send
this command when there's no active link.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.f5218f8453ec.I1325ff14ec07a27dd7ea2c1c210a1721d969839f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 1a97bf925817..8ed76fc83702 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -563,6 +563,12 @@ static void iwl_mvm_mld_vif_cfg_changed_station(struct iwl_mvm *mvm,
 	unsigned int i;
 	int ret;
 
+	/* This might get called without active links during the
+	 * chanctx switch, but we don't care about it anyway.
+	 */
+	if (changes == BSS_CHANGED_IDLE)
+		return;
+
 	ret = iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
 	if (ret)
 		IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
-- 
cgit v1.2.3-70-g09d2


From 3aff89d0b392c9bdf8a3297f8ac0cb22e398f5b3 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:26 +0300
Subject: wifi: iwlwifi: mvm: remove chanctx WARN_ON

During link switching there might be a link that's marked
active but has no chanctx assigned (so it's not active from
our driver's POV), skip such links in power recalculation
without any warning.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.c629090bd5d2.If7a680d5e349d454f2122f936c21522b9528a55f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/power.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 2033d89dc9e5..ac1dae52556f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -286,7 +286,8 @@ static bool iwl_mvm_power_is_radar(struct ieee80211_vif *vif)
 	rcu_read_lock();
 	for_each_vif_active_link(vif, link_conf, link_id) {
 		chanctx_conf = rcu_dereference(link_conf->chanctx_conf);
-		if (WARN_ON(!chanctx_conf))
+		/* this happens on link switching, just ignore inactive ones */
+		if (!chanctx_conf)
 			continue;
 
 		radar_detect = !!(chanctx_conf->def.chan->flags &
-- 
cgit v1.2.3-70-g09d2


From 0d504ca1f19b1a150fa17e59c38e1fb42a585b4f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:27 +0300
Subject: wifi: iwlwifi: mvm: use the new lockdep-checking macros

Use the new macros from mac80211 that do lockdep checking
on the RCU dereferences, instead of hard-coding 1 as the
argument to rcu_dereference_protected().

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.112df5c8dec2.I1a1008f5566e509953d988f254d15c9e58630418@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c     | 14 +++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c |  4 +---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c      | 10 +++++-----
 3 files changed, 13 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 98c940271648..7fb00220e6cc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3591,9 +3591,9 @@ static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
 
 	for_each_mvm_vif_valid_link(mvmvif, link_id) {
 		struct ieee80211_bss_conf *conf =
-			rcu_dereference_protected(vif->link_conf[link_id], 1);
+			link_conf_dereference_protected(vif, link_id);
 		struct ieee80211_link_sta *link_sta =
-			rcu_dereference_protected(sta->link[link_id], 1);
+			link_sta_dereference_protected(sta, link_id);
 
 		if (!conf || !link_sta || !mvmvif->link[link_id]->phy_ctxt)
 			continue;
@@ -3622,9 +3622,9 @@ static bool iwl_mvm_vif_conf_from_sta(struct iwl_mvm *mvm,
 	for_each_set_bit(i, (unsigned long *)&sta->valid_links,
 			 IEEE80211_MLD_MAX_NUM_LINKS) {
 		struct ieee80211_link_sta *link_sta =
-			rcu_dereference_protected(sta->link[i], 1);
+			link_sta_dereference_protected(sta, i);
 		struct ieee80211_bss_conf *link_conf =
-			rcu_dereference_protected(vif->link_conf[i], 1);
+			link_conf_dereference_protected(vif, i);
 
 		if (!link_conf || !link_sta)
 			continue;
@@ -3653,9 +3653,9 @@ static void iwl_mvm_vif_set_he_support(struct ieee80211_hw *hw,
 	for_each_set_bit(i, (unsigned long *)&sta->valid_links,
 			 IEEE80211_MLD_MAX_NUM_LINKS) {
 		struct ieee80211_link_sta *link_sta =
-			rcu_dereference_protected(sta->link[i], 1);
+			link_sta_dereference_protected(sta, i);
 		struct ieee80211_bss_conf *link_conf =
-			rcu_dereference_protected(vif->link_conf[i], 1);
+			link_conf_dereference_protected(vif, i);
 
 		if (!link_conf || !link_sta || !mvmvif->link[i])
 			continue;
@@ -3753,7 +3753,7 @@ iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw,
 		for_each_set_bit(i, (unsigned long *)&sta->valid_links,
 				 IEEE80211_MLD_MAX_NUM_LINKS) {
 			struct ieee80211_bss_conf *link_conf =
-				rcu_dereference_protected(vif->link_conf[i], 1);
+				link_conf_dereference_protected(vif, i);
 
 			if (WARN_ON(!link_conf))
 				return -EINVAL;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 8ed76fc83702..ddc09012549b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -921,9 +921,7 @@ iwl_mvm_mld_change_vif_links(struct ieee80211_hw *hw,
 		if (added & BIT(i)) {
 			struct ieee80211_bss_conf *link_conf;
 
-			/* FIXME: allow use of sdata_dereference()? */
-			link_conf = rcu_dereference_protected(vif->link_conf[i],
-							      1);
+			link_conf = link_conf_dereference_protected(vif, i);
 			if (WARN_ON(!link_conf))
 				continue;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index a713b8a10781..34b85a9a27af 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -499,7 +499,7 @@ static int iwl_mvm_mld_alloc_sta_link(struct iwl_mvm *mvm,
 				      unsigned int link_id)
 {
 	struct ieee80211_link_sta *link_sta =
-		rcu_dereference_protected(sta->link[link_id], 1);
+		link_sta_dereference_protected(sta, link_id);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_link_sta *link;
 	u32 sta_id = iwl_mvm_find_free_sta_id(mvm,
@@ -654,7 +654,7 @@ int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	for_each_sta_active_link(vif, sta, link_sta, link_id) {
 		struct ieee80211_bss_conf *link_conf =
-			rcu_dereference_protected(vif->link_conf[link_id], 1);
+			link_conf_dereference_protected(vif, link_id);
 		struct iwl_mvm_link_sta *mvm_link_sta =
 			rcu_dereference_protected(mvm_sta->link[link_id],
 						  lockdep_is_held(&mvm->mutex));
@@ -704,7 +704,7 @@ int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	for_each_sta_active_link(vif, sta, link_sta, link_id) {
 		struct ieee80211_bss_conf *link_conf =
-			rcu_dereference_protected(vif->link_conf[link_id], 1);
+			link_conf_dereference_protected(vif, link_id);
 		struct iwl_mvm_link_sta *mvm_link_sta =
 			rcu_dereference_protected(mvm_sta->link[link_id],
 						  lockdep_is_held(&mvm->mutex));
@@ -983,9 +983,9 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 
 	for_each_set_bit(link_id, &links_to_add, IEEE80211_MLD_MAX_NUM_LINKS) {
 		struct ieee80211_bss_conf *link_conf =
-			rcu_dereference_protected(vif->link_conf[link_id], 1);
+			link_conf_dereference_protected(vif, link_id);
 		struct ieee80211_link_sta *link_sta =
-			rcu_dereference_protected(sta->link[link_id], 1);
+			link_sta_dereference_protected(sta, link_id);
 		mvm_vif_link = mvm_vif->link[link_id];
 
 		if (WARN_ON(!mvm_vif_link || !link_conf || !link_sta ||
-- 
cgit v1.2.3-70-g09d2


From 9213f809bb6319f87296103db39401834bdbe0cd Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:28 +0300
Subject: wifi: iwlwifi: mvm: use appropriate link for rate selection

We were still using the deflink in most cases, update the code to use
the appropriate link.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.fa1025502fb4.Iaba0c64740fdcf04a521e2f213bd3f3e27862472@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c | 109 ++++++++++++++-----------
 1 file changed, 61 insertions(+), 48 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index d1a308a451cc..c8ba2fe3e4a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -9,9 +9,9 @@
 #include "iwl-op-mode.h"
 #include "mvm.h"
 
-static u8 rs_fw_bw_from_sta_bw(const struct ieee80211_sta *sta)
+static u8 rs_fw_bw_from_sta_bw(const struct ieee80211_link_sta *link_sta)
 {
-	switch (sta->deflink.bandwidth) {
+	switch (link_sta->bandwidth) {
 	case IEEE80211_STA_RX_BW_320:
 		return IWL_TLC_MNG_CH_WIDTH_320MHZ;
 	case IEEE80211_STA_RX_BW_160:
@@ -38,11 +38,11 @@ static u8 rs_fw_set_active_chains(u8 chains)
 	return fw_chains;
 }
 
-static u8 rs_fw_sgi_cw_support(struct ieee80211_sta *sta)
+static u8 rs_fw_sgi_cw_support(struct ieee80211_link_sta *link_sta)
 {
-	struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
-	struct ieee80211_sta_vht_cap *vht_cap = &sta->deflink.vht_cap;
-	struct ieee80211_sta_he_cap *he_cap = &sta->deflink.he_cap;
+	struct ieee80211_sta_ht_cap *ht_cap = &link_sta->ht_cap;
+	struct ieee80211_sta_vht_cap *vht_cap = &link_sta->vht_cap;
+	struct ieee80211_sta_he_cap *he_cap = &link_sta->he_cap;
 	u8 supp = 0;
 
 	if (he_cap->has_he)
@@ -61,12 +61,12 @@ static u8 rs_fw_sgi_cw_support(struct ieee80211_sta *sta)
 }
 
 static u16 rs_fw_get_config_flags(struct iwl_mvm *mvm,
-				  struct ieee80211_sta *sta,
+				  struct ieee80211_link_sta *link_sta,
 				  struct ieee80211_supported_band *sband)
 {
-	struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
-	struct ieee80211_sta_vht_cap *vht_cap = &sta->deflink.vht_cap;
-	struct ieee80211_sta_he_cap *he_cap = &sta->deflink.he_cap;
+	struct ieee80211_sta_ht_cap *ht_cap = &link_sta->ht_cap;
+	struct ieee80211_sta_vht_cap *vht_cap = &link_sta->vht_cap;
+	struct ieee80211_sta_he_cap *he_cap = &link_sta->he_cap;
 	bool vht_ena = vht_cap->vht_supported;
 	u16 flags = 0;
 
@@ -132,20 +132,20 @@ int rs_fw_vht_highest_rx_mcs_index(const struct ieee80211_sta_vht_cap *vht_cap,
 }
 
 static void
-rs_fw_vht_set_enabled_rates(const struct ieee80211_sta *sta,
+rs_fw_vht_set_enabled_rates(const struct ieee80211_link_sta *link_sta,
 			    const struct ieee80211_sta_vht_cap *vht_cap,
 			    struct iwl_tlc_config_cmd_v4 *cmd)
 {
 	u16 supp;
 	int i, highest_mcs;
-	u8 max_nss = sta->deflink.rx_nss;
+	u8 max_nss = link_sta->rx_nss;
 	struct ieee80211_vht_cap ieee_vht_cap = {
 		.vht_cap_info = cpu_to_le32(vht_cap->cap),
 		.supp_mcs = vht_cap->vht_mcs,
 	};
 
 	/* the station support only a single receive chain */
-	if (sta->deflink.smps_mode == IEEE80211_SMPS_STATIC)
+	if (link_sta->smps_mode == IEEE80211_SMPS_STATIC)
 		max_nss = 1;
 
 	for (i = 0; i < max_nss && i < IWL_TLC_NSS_MAX; i++) {
@@ -156,7 +156,7 @@ rs_fw_vht_set_enabled_rates(const struct ieee80211_sta *sta,
 			continue;
 
 		supp = BIT(highest_mcs + 1) - 1;
-		if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_20)
+		if (link_sta->bandwidth == IEEE80211_STA_RX_BW_20)
 			supp &= ~BIT(IWL_TLC_MNG_HT_RATE_MCS9);
 
 		cmd->ht_rates[i][IWL_TLC_MCS_PER_BW_80] = cpu_to_le16(supp);
@@ -165,7 +165,7 @@ rs_fw_vht_set_enabled_rates(const struct ieee80211_sta *sta,
 		 * configuration is supported - only for MCS 0 since we already
 		 * decoded the MCS bits anyway ourselves.
 		 */
-		if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_160 &&
+		if (link_sta->bandwidth == IEEE80211_STA_RX_BW_160 &&
 		    ieee80211_get_vht_max_nss(&ieee_vht_cap,
 					      IEEE80211_VHT_CHANWIDTH_160MHZ,
 					      0, true, nss) >= nss)
@@ -192,11 +192,11 @@ static u16 rs_fw_he_ieee80211_mcs_to_rs_mcs(u16 mcs)
 }
 
 static void
-rs_fw_he_set_enabled_rates(const struct ieee80211_sta *sta,
+rs_fw_he_set_enabled_rates(const struct ieee80211_link_sta *link_sta,
 			   struct ieee80211_supported_band *sband,
 			   struct iwl_tlc_config_cmd_v4 *cmd)
 {
-	const struct ieee80211_sta_he_cap *he_cap = &sta->deflink.he_cap;
+	const struct ieee80211_sta_he_cap *he_cap = &link_sta->he_cap;
 	u16 mcs_160 = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_160);
 	u16 mcs_80 = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_80);
 	u16 tx_mcs_80 =
@@ -204,10 +204,10 @@ rs_fw_he_set_enabled_rates(const struct ieee80211_sta *sta,
 	u16 tx_mcs_160 =
 		le16_to_cpu(sband->iftype_data->he_cap.he_mcs_nss_supp.tx_mcs_160);
 	int i;
-	u8 nss = sta->deflink.rx_nss;
+	u8 nss = link_sta->rx_nss;
 
 	/* the station support only a single receive chain */
-	if (sta->deflink.smps_mode == IEEE80211_SMPS_STATIC)
+	if (link_sta->smps_mode == IEEE80211_SMPS_STATIC)
 		nss = 1;
 
 	for (i = 0; i < nss && i < IWL_TLC_NSS_MAX; i++) {
@@ -282,13 +282,14 @@ rs_fw_rs_mcs2eht_mcs(enum IWL_TLC_MCS_PER_BW bw,
 	}
 }
 
-static void rs_fw_eht_set_enabled_rates(const struct ieee80211_sta *sta,
-					struct ieee80211_supported_band *sband,
-					struct iwl_tlc_config_cmd_v4 *cmd)
+static void
+rs_fw_eht_set_enabled_rates(const struct ieee80211_link_sta *link_sta,
+			    struct ieee80211_supported_band *sband,
+			    struct iwl_tlc_config_cmd_v4 *cmd)
 {
 	/* peer RX mcs capa */
 	const struct ieee80211_eht_mcs_nss_supp *eht_rx_mcs =
-		&sta->deflink.eht_cap.eht_mcs_nss_supp;
+		&link_sta->eht_cap.eht_mcs_nss_supp;
 	/* our TX mcs capa */
 	const struct ieee80211_eht_mcs_nss_supp *eht_tx_mcs =
 		&sband->iftype_data->eht_cap.eht_mcs_nss_supp;
@@ -298,7 +299,7 @@ static void rs_fw_eht_set_enabled_rates(const struct ieee80211_sta *sta,
 	struct ieee80211_eht_mcs_nss_supp_20mhz_only mcs_tx_20;
 
 	/* peer is 20Mhz only */
-	if (!(sta->deflink.he_cap.he_cap_elem.phy_cap_info[0] &
+	if (!(link_sta->he_cap.he_cap_elem.phy_cap_info[0] &
 	      IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK_ALL)) {
 		mcs_rx_20 = eht_rx_mcs->only_20mhz;
 	} else {
@@ -354,25 +355,25 @@ static void rs_fw_eht_set_enabled_rates(const struct ieee80211_sta *sta,
 	}
 
 	/* the station support only a single receive chain */
-	if (sta->deflink.smps_mode == IEEE80211_SMPS_STATIC ||
-	    sta->deflink.rx_nss < 2)
+	if (link_sta->smps_mode == IEEE80211_SMPS_STATIC ||
+	    link_sta->rx_nss < 2)
 		memset(cmd->ht_rates[IWL_TLC_NSS_2], 0,
 		       sizeof(cmd->ht_rates[IWL_TLC_NSS_2]));
 }
 
-static void rs_fw_set_supp_rates(struct ieee80211_sta *sta,
+static void rs_fw_set_supp_rates(struct ieee80211_link_sta *link_sta,
 				 struct ieee80211_supported_band *sband,
 				 struct iwl_tlc_config_cmd_v4 *cmd)
 {
 	int i;
 	u16 supp = 0;
 	unsigned long tmp; /* must be unsigned long for for_each_set_bit */
-	const struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
-	const struct ieee80211_sta_vht_cap *vht_cap = &sta->deflink.vht_cap;
-	const struct ieee80211_sta_he_cap *he_cap = &sta->deflink.he_cap;
+	const struct ieee80211_sta_ht_cap *ht_cap = &link_sta->ht_cap;
+	const struct ieee80211_sta_vht_cap *vht_cap = &link_sta->vht_cap;
+	const struct ieee80211_sta_he_cap *he_cap = &link_sta->he_cap;
 
 	/* non HT rates */
-	tmp = sta->deflink.supp_rates[sband->band];
+	tmp = link_sta->supp_rates[sband->band];
 	for_each_set_bit(i, &tmp, BITS_PER_LONG)
 		supp |= BIT(sband->bitrates[i].hw_value);
 
@@ -380,22 +381,22 @@ static void rs_fw_set_supp_rates(struct ieee80211_sta *sta,
 	cmd->mode = IWL_TLC_MNG_MODE_NON_HT;
 
 	/* HT/VHT rates */
-	if (sta->deflink.eht_cap.has_eht) {
+	if (link_sta->eht_cap.has_eht) {
 		cmd->mode = IWL_TLC_MNG_MODE_EHT;
-		rs_fw_eht_set_enabled_rates(sta, sband, cmd);
+		rs_fw_eht_set_enabled_rates(link_sta, sband, cmd);
 	} else if (he_cap->has_he) {
 		cmd->mode = IWL_TLC_MNG_MODE_HE;
-		rs_fw_he_set_enabled_rates(sta, sband, cmd);
+		rs_fw_he_set_enabled_rates(link_sta, sband, cmd);
 	} else if (vht_cap->vht_supported) {
 		cmd->mode = IWL_TLC_MNG_MODE_VHT;
-		rs_fw_vht_set_enabled_rates(sta, vht_cap, cmd);
+		rs_fw_vht_set_enabled_rates(link_sta, vht_cap, cmd);
 	} else if (ht_cap->ht_supported) {
 		cmd->mode = IWL_TLC_MNG_MODE_HT;
 		cmd->ht_rates[IWL_TLC_NSS_1][IWL_TLC_MCS_PER_BW_80] =
 			cpu_to_le16(ht_cap->mcs.rx_mask[0]);
 
 		/* the station support only a single receive chain */
-		if (sta->deflink.smps_mode == IEEE80211_SMPS_STATIC)
+		if (link_sta->smps_mode == IEEE80211_SMPS_STATIC)
 			cmd->ht_rates[IWL_TLC_NSS_2][IWL_TLC_MCS_PER_BW_80] =
 				0;
 		else
@@ -410,15 +411,18 @@ void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 	struct iwl_rx_packet *pkt = rxb_addr(rxb);
 	struct iwl_tlc_update_notif *notif;
 	struct ieee80211_sta *sta;
+	struct ieee80211_link_sta *link_sta;
 	struct iwl_mvm_sta *mvmsta;
+	struct iwl_mvm_link_sta *mvm_link_sta;
 	struct iwl_lq_sta_rs_fw *lq_sta;
 	u32 flags;
 
 	rcu_read_lock();
 
 	notif = (void *)pkt->data;
+	link_sta = rcu_dereference(mvm->fw_id_to_link_sta[notif->sta_id]);
 	sta = rcu_dereference(mvm->fw_id_to_mac_id[notif->sta_id]);
-	if (IS_ERR_OR_NULL(sta)) {
+	if (IS_ERR_OR_NULL(sta) || !link_sta) {
 		/* can happen in remove station flow where mvm removed internally
 		 * the station before removing from FW
 		 */
@@ -438,7 +442,14 @@ void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 
 	flags = le32_to_cpu(notif->flags);
 
-	lq_sta = &mvmsta->deflink.lq_sta.rs_fw;
+	mvm_link_sta = rcu_dereference(mvmsta->link[link_sta->link_id]);
+	if (!mvm_link_sta) {
+		IWL_DEBUG_RATE(mvm,
+			       "Invalid mvmsta RCU pointer for link (%d) of  sta id (%d) in TLC notification\n",
+			       link_sta->link_id, notif->sta_id);
+		goto out;
+	}
+	lq_sta = &mvm_link_sta->lq_sta.rs_fw;
 
 	if (flags & IWL_TLC_NOTIF_FLAG_RATE) {
 		char pretty_rate[100];
@@ -465,9 +476,9 @@ void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 		u16 size = le32_to_cpu(notif->amsdu_size);
 		int i;
 
-		if (sta->deflink.agg.max_amsdu_len < size) {
+		if (link_sta->agg.max_amsdu_len < size) {
 			/*
-			 * In debug sta->deflink.agg.max_amsdu_len < size
+			 * In debug link_sta->agg.max_amsdu_len < size
 			 * so also check with orig_amsdu_len which holds the
 			 * original data before debugfs changed the value
 			 */
@@ -477,18 +488,18 @@ void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 
 		mvmsta->amsdu_enabled = le32_to_cpu(notif->amsdu_enabled);
 		mvmsta->max_amsdu_len = size;
-		sta->deflink.agg.max_rc_amsdu_len = mvmsta->max_amsdu_len;
+		link_sta->agg.max_rc_amsdu_len = mvmsta->max_amsdu_len;
 
 		for (i = 0; i < IWL_MAX_TID_COUNT; i++) {
 			if (mvmsta->amsdu_enabled & BIT(i))
-				sta->deflink.agg.max_tid_amsdu_len[i] =
+				link_sta->agg.max_tid_amsdu_len[i] =
 					iwl_mvm_max_amsdu_size(mvm, sta, i);
 			else
 				/*
 				 * Not so elegant, but this will effectively
 				 * prevent AMSDU on this TID
 				 */
-				sta->deflink.agg.max_tid_amsdu_len[i] = 1;
+				link_sta->agg.max_tid_amsdu_len[i] = 1;
 		}
 
 		IWL_DEBUG_RATE(mvm,
@@ -511,7 +522,7 @@ u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta,
 		return IEEE80211_MAX_MPDU_LEN_VHT_3895;
 
 	if (link_conf->chandef.chan->band == NL80211_BAND_6GHZ) {
-		switch (le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+		switch (le16_get_bits(link_sta->he_6ghz_capa.capa,
 				      IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN)) {
 		case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454:
 			return IEEE80211_MAX_MPDU_LEN_VHT_11454;
@@ -560,10 +571,12 @@ void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	struct iwl_lq_sta_rs_fw *lq_sta;
 	struct iwl_tlc_config_cmd_v4 cfg_cmd = {
 		.max_ch_width = update ?
-			rs_fw_bw_from_sta_bw(sta) : IWL_TLC_MNG_CH_WIDTH_20MHZ,
-		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, sta, sband)),
+			rs_fw_bw_from_sta_bw(link_sta) :
+			IWL_TLC_MNG_CH_WIDTH_20MHZ,
+		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, link_sta,
+							    sband)),
 		.chains = rs_fw_set_active_chains(iwl_mvm_get_valid_tx_ant(mvm)),
-		.sgi_ch_width_supp = rs_fw_sgi_cw_support(sta),
+		.sgi_ch_width_supp = rs_fw_sgi_cw_support(link_sta),
 		.max_mpdu_len = iwl_mvm_is_csum_supported(mvm) ?
 				cpu_to_le16(max_amsdu_len) : 0,
 	};
@@ -588,7 +601,7 @@ void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	iwl_mvm_reset_frame_stats(mvm);
 #endif
-	rs_fw_set_supp_rates(sta, sband, &cfg_cmd);
+	rs_fw_set_supp_rates(link_sta, sband, &cfg_cmd);
 
 	/*
 	 * since TLC offload works with one mode we can assume
-- 
cgit v1.2.3-70-g09d2


From 24dc33cc54e61a7a97049d6f663c5ac92e097d68 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:29 +0300
Subject: wifi: iwlwifi: mvm: initialize max_rc_amsdu_len per-link

Initialize max_rc_amsdu_len per-link both on state change and when a new
link is added.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.5bf521fe58b8.I73fe585f0ff75d41b5afd32077e3d6e48c90db2a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 34b85a9a27af..5e382981bfa2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -998,6 +998,9 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 		if (WARN_ON(ret))
 			goto err;
 
+		link_sta->agg.max_rc_amsdu_len = 1;
+		ieee80211_sta_recalc_aggregates(sta);
+
 		mvm_sta_link =
 			rcu_dereference_protected(mvm_sta->link[link_id],
 						  lockdep_is_held(&mvm->mutex));
-- 
cgit v1.2.3-70-g09d2


From 881d054896f700ca80b3bb2e5ad0b93631bf0113 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:30 +0300
Subject: wifi: iwlwifi: mvm: fix station link data leak

When we remove the station, we call iwl_mvm_sta_del() which
returns true if we cannot remove it from the firmware yet,
which happens if this is the station ID for the AP station
that's still used because the MAC is still associated.

However, we still must free the link data as the station is
only kept alive in the firmware, in mac80211 and driver the
data structures are destroyed.

To fix that, we need to make iwl_mvm_mld_free_sta_link()
track whether or not the station is still alive in FW, as
otherwise we might reuse the station ID in the meantime and
iwl_mvm_mld_rm_sta_from_fw() would reject the later delete
from the firmware. Add an argument to it for that. Then we
can use the return value of iwl_mvm_sta_del() for that to
fix the issue, and call iwl_mvm_mld_rm_sta_from_fw() only
if we need to not keep the station in FW.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100039.1d81d4c71f35.I8fc60ac28ffc1147e9b1250e5e6237b3cb5516ac@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c   | 59 ++++++++++++----------
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h       |  3 +-
 3 files changed, 33 insertions(+), 31 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index ddc09012549b..beef706d0598 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -545,7 +545,7 @@ static void iwl_mvm_mld_vif_delete_all_stas(struct iwl_mvm *mvm,
 			continue;
 
 		iwl_mvm_sec_key_remove_ap(mvm, vif, link, i);
-		ret = iwl_mvm_mld_rm_sta_id(mvm, vif, link->ap_sta_id);
+		ret = iwl_mvm_mld_rm_sta_id(mvm, link->ap_sta_id);
 		if (ret)
 			IWL_ERR(mvm, "failed to remove AP station\n");
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 5e382981bfa2..b3377b4d7924 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -458,6 +458,21 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	return iwl_mvm_mld_send_sta_cmd(mvm, &cmd);
 }
 
+static void iwl_mvm_mld_free_sta_link(struct iwl_mvm *mvm,
+				      struct iwl_mvm_sta *mvm_sta,
+				      struct iwl_mvm_link_sta *mvm_sta_link,
+				      unsigned int link_id,
+				      bool is_in_fw)
+{
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta_link->sta_id],
+			 is_in_fw ? ERR_PTR(-EINVAL) : NULL);
+	RCU_INIT_POINTER(mvm->fw_id_to_link_sta[mvm_sta_link->sta_id], NULL);
+	RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
+
+	if (mvm_sta_link != &mvm_sta->deflink)
+		kfree_rcu(mvm_sta_link, rcu_head);
+}
+
 static void iwl_mvm_mld_sta_rm_all_sta_links(struct iwl_mvm *mvm,
 					     struct iwl_mvm_sta *mvm_sta)
 {
@@ -471,28 +486,10 @@ static void iwl_mvm_mld_sta_rm_all_sta_links(struct iwl_mvm *mvm,
 		if (!link)
 			continue;
 
-		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[link->sta_id], NULL);
-		RCU_INIT_POINTER(mvm->fw_id_to_link_sta[link->sta_id], NULL);
-		RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
-
-		if (link != &mvm_sta->deflink)
-			kfree_rcu(link, rcu_head);
+		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, link, link_id, false);
 	}
 }
 
-static void iwl_mvm_mld_free_sta_link(struct iwl_mvm *mvm,
-				      struct iwl_mvm_sta *mvm_sta,
-				      struct iwl_mvm_link_sta *mvm_sta_link,
-				      unsigned int link_id)
-{
-	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[mvm_sta_link->sta_id], NULL);
-	RCU_INIT_POINTER(mvm->fw_id_to_link_sta[mvm_sta_link->sta_id], NULL);
-	RCU_INIT_POINTER(mvm_sta->link[link_id], NULL);
-
-	if (mvm_sta_link != &mvm_sta->deflink)
-		kfree_rcu(mvm_sta_link, rcu_head);
-}
-
 static int iwl_mvm_mld_alloc_sta_link(struct iwl_mvm *mvm,
 				      struct ieee80211_vif *vif,
 				      struct ieee80211_sta *sta,
@@ -787,20 +784,24 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		struct iwl_mvm_link_sta *mvm_link_sta =
 			rcu_dereference_protected(mvm_sta->link[link_id],
 						  lockdep_is_held(&mvm->mutex));
+		bool stay_in_fw;
 
-		if (iwl_mvm_sta_del(mvm, vif, sta, mvm_link_sta, &ret))
-			return ret;
+		stay_in_fw = iwl_mvm_sta_del(mvm, vif, sta, mvm_link_sta, &ret);
+		if (ret)
+			break;
 
-		ret = iwl_mvm_mld_rm_sta_from_fw(mvm, mvm_link_sta->sta_id);
-	}
+		if (!stay_in_fw)
+			ret = iwl_mvm_mld_rm_sta_from_fw(mvm,
+							 mvm_link_sta->sta_id);
 
-	iwl_mvm_mld_sta_rm_all_sta_links(mvm, mvm_sta);
+		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_link_sta,
+					  link_id, stay_in_fw);
+	}
 
 	return ret;
 }
 
-int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			  u8 sta_id)
+int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, u8 sta_id)
 {
 	int ret = iwl_mvm_mld_rm_sta_from_fw(mvm, sta_id);
 
@@ -978,7 +979,8 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 		if (vif->type == NL80211_IFTYPE_STATION)
 			mvm_vif_link->ap_sta_id = IWL_MVM_INVALID_STA;
 
-		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_sta_link, link_id);
+		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_sta_link, link_id,
+					  false);
 	}
 
 	for_each_set_bit(link_id, &links_to_add, IEEE80211_MLD_MAX_NUM_LINKS) {
@@ -1054,7 +1056,8 @@ err:
 			rcu_dereference_protected(mvm_sta->link[link_id],
 						  lockdep_is_held(&mvm->mutex));
 
-		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_sta_link, link_id);
+		iwl_mvm_mld_free_sta_link(mvm, mvm_sta, mvm_sta_link, link_id,
+					  false);
 	}
 
 	return ret;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 2402b41bc0bb..824d12047019 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -639,8 +639,7 @@ int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta);
 int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta);
-int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			  u8 sta_id);
+int iwl_mvm_mld_rm_sta_id(struct iwl_mvm *mvm, u8 sta_id);
 int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif,
 				 struct ieee80211_sta *sta,
-- 
cgit v1.2.3-70-g09d2


From 2152662d4e14610aaec84b8aba1266114ee78dd5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:31 +0300
Subject: wifi: iwlwifi: mvm: clean up mac_id vs. link_id in MLD sta

Here we always have a link ID, not MAC ID, so clean up the
naming.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.3def62de34b5.I10c9cf5dbfd1fc1e9c9c7d6d4cefcf0c08f1f2da@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index b3377b4d7924..4d713c78b508 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -21,8 +21,7 @@ static int iwl_mvm_mld_send_sta_cmd(struct iwl_mvm *mvm,
  */
 static int iwl_mvm_mld_add_int_sta_to_fw(struct iwl_mvm *mvm,
 					 struct iwl_mvm_int_sta *sta,
-					 const u8 *addr,
-					 u16 mac_id)
+					 const u8 *addr, int link_id)
 {
 	struct iwl_mvm_sta_cfg_cmd cmd;
 
@@ -31,7 +30,7 @@ static int iwl_mvm_mld_add_int_sta_to_fw(struct iwl_mvm *mvm,
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.sta_id = cpu_to_le32((u8)sta->sta_id);
 
-	cmd.link_id = cpu_to_le32(mac_id);
+	cmd.link_id = cpu_to_le32(link_id);
 
 	cmd.station_type = cpu_to_le32(sta->type);
 
@@ -94,7 +93,7 @@ static int iwl_mvm_add_aux_sta_to_fw(struct iwl_mvm *mvm,
  */
 static int iwl_mvm_mld_add_int_sta_with_queue(struct iwl_mvm *mvm,
 					      struct iwl_mvm_int_sta *sta,
-					      const u8 *addr, int mac_id,
+					      const u8 *addr, int link_id,
 					      u16 *queue, u8 tid,
 					      unsigned int *_wdg_timeout)
 {
@@ -106,9 +105,9 @@ static int iwl_mvm_mld_add_int_sta_with_queue(struct iwl_mvm *mvm,
 		return -ENOSPC;
 
 	if (sta->type == STATION_TYPE_AUX)
-		ret = iwl_mvm_add_aux_sta_to_fw(mvm, sta, mac_id);
+		ret = iwl_mvm_add_aux_sta_to_fw(mvm, sta, link_id);
 	else
-		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, mac_id);
+		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, sta, addr, link_id);
 	if (ret)
 		return ret;
 
@@ -135,7 +134,7 @@ static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
 				   struct iwl_mvm_int_sta *int_sta, u16 *queue,
 				   enum nl80211_iftype iftype,
 				   enum iwl_fw_sta_type sta_type,
-				   int mac_id, const u8 *addr, u8 tid,
+				   int link_id, const u8 *addr, u8 tid,
 				   unsigned int *wdg_timeout)
 {
 	int ret;
@@ -148,7 +147,7 @@ static int iwl_mvm_mld_add_int_sta(struct iwl_mvm *mvm,
 	if (ret)
 		return ret;
 
-	ret = iwl_mvm_mld_add_int_sta_with_queue(mvm, int_sta, addr, mac_id,
+	ret = iwl_mvm_mld_add_int_sta_with_queue(mvm, int_sta, addr, link_id,
 						 queue, tid, wdg_timeout);
 	if (ret) {
 		iwl_mvm_dealloc_int_sta(mvm, int_sta);
@@ -254,8 +253,8 @@ int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
 {
 	lockdep_assert_held(&mvm->mutex);
 
-	/* In CDB NICs we need to specify which lmac to use for aux activity
-	 * using the mac_id argument place to send lmac_id to the function
+	/* In CDB NICs we need to specify which lmac to use for aux activity;
+	 * use the link_id argument place to send lmac_id to the function.
 	 */
 	return iwl_mvm_mld_add_int_sta(mvm, &mvm->aux_sta, &mvm->aux_queue,
 				       NL80211_IFTYPE_UNSPECIFIED,
-- 
cgit v1.2.3-70-g09d2


From f699f039cdb6e028d1f40cb54e73f85cd5d459bd Mon Sep 17 00:00:00 2001
From: Shaul Triebitz <shaul.triebitz@intel.com>
Date: Wed, 29 Mar 2023 10:05:32 +0300
Subject: wifi: iwlwifi: mvm: use the correct link queue

For bcase/mcast tx frames use the link queue.

Signed-off-by: Shaul Triebitz <shaul.triebitz@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.ccd7218e4be2.I40f608a0441190cc26137b039f7cb7b065fd4e0c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index ca1284f11f8b..f5b92d99f39d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -604,11 +604,10 @@ static void iwl_mvm_skb_prepare_status(struct sk_buff *skb,
 }
 
 static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
+				      struct iwl_mvm_vif_link_info *link,
 				      struct ieee80211_tx_info *info,
 				      struct ieee80211_hdr *hdr)
 {
-	struct iwl_mvm_vif *mvmvif =
-		iwl_mvm_vif_from_mac80211(info->control.vif);
 	__le16 fc = hdr->frame_control;
 
 	switch (info->control.vif->type) {
@@ -631,7 +630,7 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 
 		if (!ieee80211_has_order(fc) && !ieee80211_is_probe_req(fc) &&
 		    is_multicast_ether_addr(hdr->addr1))
-			return mvmvif->deflink.cab_queue;
+			return link->cab_queue;
 
 		WARN_ONCE(info->control.vif->type != NL80211_IFTYPE_ADHOC,
 			  "fc=0x%02x", le16_to_cpu(fc));
@@ -757,7 +756,8 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 			else
 				sta_id = link->mcast_sta.sta_id;
 
-			queue = iwl_mvm_get_ctrl_vif_queue(mvm, &info, hdr);
+			queue = iwl_mvm_get_ctrl_vif_queue(mvm, link, &info,
+							   hdr);
 		} else if (info.control.vif->type == NL80211_IFTYPE_MONITOR) {
 			queue = mvm->snif_queue;
 			sta_id = mvm->snif_sta.sta_id;
-- 
cgit v1.2.3-70-g09d2


From ef36f1df198b57e2c9c1fe249f537cd64a762f9f Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:33 +0300
Subject: wifi: iwlwifi: mvm: update mac config when assigning chanctx

Some mac parameters, such as HE support, can change at this stage.
Update mac config before updating links.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.d7882a0d6e04.Ie38cd854a237c46cf85fd7143dc757326f30da6e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index beef706d0598..4d56b2fc5f33 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -231,6 +231,18 @@ __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	if (WARN_ON_ONCE(!mvmvif->link[link_id]))
 		return -EINVAL;
 
+	/* mac parameters such as HE support can change at this stage
+	 * For sta, need first to configure correct state from drv_sta_state
+	 * and only after that update mac config.
+	 */
+	if (vif->type == NL80211_IFTYPE_AP) {
+		ret = iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
+		if (ret) {
+			IWL_ERR(mvm, "failed to update MAC %pM\n", vif->addr);
+			return -EINVAL;
+		}
+	}
+
 	mvmvif->link[link_id]->phy_ctxt = phy_ctxt;
 
 	if (switching_chanctx) {
-- 
cgit v1.2.3-70-g09d2


From 9deccfcd48b7c358245ed05c11b142cc945daf5e Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:34 +0300
Subject: wifi: iwlwifi: mvm: rework active links counting

Remove fw_active_links_num counter since we now have a bitmap of
active links in vif. Also, update link activation status only when
LINK_CONTEXT_MODIFY_ACTIVE bit set in changes parameter.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.1ecfb27b6b84.I3a5e0bc32b3728e4caae8a231bc3f04ea1d89cad@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c     | 33 ++++++++++++-----------
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  1 -
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  3 ---
 3 files changed, 17 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 0cd40672fade..eb828de40a3c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -104,11 +104,6 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			 link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID))
 		return -EINVAL;
 
-	/* cannot activate third link */
-	if (!link_info->active && active &&
-	    mvmvif->fw_active_links_num >= IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM)
-		return -EINVAL;
-
 	if (changes & LINK_CONTEXT_MODIFY_ACTIVE) {
 		/* When activating a link, phy context should be valid;
 		 * when deactivating a link, it also should be valid since
@@ -119,6 +114,22 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		if (!link_info->phy_ctxt)
 			return 0;
 
+		/* check there aren't too many active links */
+		if (!link_info->active && active) {
+			int i, count = 0;
+
+			/* link with phy_ctxt is active in FW */
+			for_each_mvm_vif_valid_link(mvmvif, i)
+				if (mvmvif->link[i]->phy_ctxt)
+					count++;
+
+			/* FIXME: IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM should be
+			 * defined per HW
+			 */
+			if (count >= IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM)
+				return -EINVAL;
+		}
+
 		/* Catch early if driver tries to activate or deactivate a link
 		 * twice.
 		 */
@@ -230,18 +241,8 @@ send_cmd:
 	cmd.flags_mask = cpu_to_le32(flags_mask);
 
 	ret = iwl_mvm_link_cmd_send(mvm, &cmd, FW_CTXT_ACTION_MODIFY);
-	if (!ret) {
-		/* the FW is updated, so now it's possible to update the
-		 * activation status. If activating a link, it was already
-		 * checked above that we didn't reach the FW limit.
-		 */
-		if (link_info->active && !active)
-			mvmvif->fw_active_links_num--;
-		else if (!link_info->active && active)
-			mvmvif->fw_active_links_num++;
-
+	if (!ret && (changes & LINK_CONTEXT_MODIFY_ACTIVE))
 		link_info->active = active;
-	}
 
 	return ret;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 7fb00220e6cc..ff03c8d52277 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1027,7 +1027,6 @@ static void iwl_mvm_cleanup_iterator(void *data, u8 *mac,
 
 	memset(&mvmvif->bf_data, 0, sizeof(mvmvif->bf_data));
 
-	mvmvif->fw_active_links_num = 0;
 	for_each_mvm_vif_valid_link(mvmvif, link_id) {
 		mvmvif->link[link_id]->ap_sta_id = IWL_MVM_INVALID_STA;
 		mvmvif->link[link_id]->fw_link_id = IWL_MVM_FW_LINK_ID_INVALID;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 23d797247915..022f7bd64a5d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -355,7 +355,6 @@ struct iwl_mvm_vif_link_info {
  * @csa_failed: CSA failed to schedule time event, report an error later
  * @csa_bcn_pending: indicates that we are waiting for a beacon on a new channel
  * @features: hw features active for this vif
- * @fw_active_links_num: the number of activated links
  */
 struct iwl_mvm_vif {
 	struct iwl_mvm *mvm;
@@ -439,8 +438,6 @@ struct iwl_mvm_vif {
 		struct ieee80211_key_conf __rcu *keys[2];
 	} bcn_prot;
 
-	u32 fw_active_links_num;
-
 	struct iwl_mvm_vif_link_info deflink;
 	struct iwl_mvm_vif_link_info *link[IEEE80211_MLD_MAX_NUM_LINKS];
 };
-- 
cgit v1.2.3-70-g09d2


From 164a52d45e4aac669be811d7a345856cccd8cc8e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:35 +0300
Subject: wifi: iwlwifi: mvm: send full STA during HW restart

By using the internal station add the station is installed in
firmware with zeroed MAC addresses, which is wrong. Use the
full installation function instead, to fill all data.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.62d5371bb3c7.Ie25b62125a3a022f76a36bae5fed9796c18698aa@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 23 +++++++++--------------
 1 file changed, 9 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 4d713c78b508..26686cc7ff4c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -572,10 +572,9 @@ static int iwl_mvm_alloc_sta_after_restart(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct ieee80211_link_sta *link_sta;
 	unsigned int link_id;
-	struct iwl_mvm_int_sta tmp_sta = {
-		.type = mvm_sta->sta_type,
-	};
-	int sta_id, ret;
+	/* no active link found */
+	int ret = -EINVAL;
+	int sta_id;
 
 	/* First add an empty station since allocating a queue requires
 	 * a valid station. Since we need a link_id to allocate a station,
@@ -598,23 +597,19 @@ static int iwl_mvm_alloc_sta_after_restart(struct iwl_mvm *mvm,
 			continue;
 
 		sta_id = mvm_link_sta->sta_id;
-		tmp_sta.sta_id = sta_id;
-		ret = iwl_mvm_mld_add_int_sta_to_fw(mvm, &tmp_sta,
-						    vif->bss_conf.bssid,
-						    mvm_link->fw_link_id);
+		ret = iwl_mvm_mld_cfg_sta(mvm, sta, vif, link_sta,
+					  link_conf, mvm_link_sta);
 		if (ret)
 			return ret;
 
 		rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id], sta);
 		rcu_assign_pointer(mvm->fw_id_to_link_sta[sta_id], link_sta);
-		iwl_mvm_realloc_queues_after_restart(mvm, sta);
-
-		/* since we need only one station, no need to continue */
-		return 0;
+		ret = 0;
 	}
 
-	/* no active link found */
-	return -EINVAL;
+	iwl_mvm_realloc_queues_after_restart(mvm, sta);
+
+	return ret;
 }
 
 int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-- 
cgit v1.2.3-70-g09d2


From 59c2d94bba838b61733186e490d4febaec8e93a6 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:36 +0300
Subject: wifi: iwlwifi: mvm: move max_agg_bufsize into host TLC lq_sta

This field is used only for host TLC, so it can reside inside
the corresponding lq_sta struct. Also, TLC lq_sta is cleared
in iwl_mvm_rs_rate_init() upon association, but max_agg_bufsize
is set earlier in iwl_mvm_sta_init(). Thus, place this field
in the persistent part of lq_sta to retain its value.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.d55361064e39.Ib79d30f27d94607d097f0192af2aacd455a17958@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c  |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h  |  6 ++++--
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 11 ++++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h |  2 --
 4 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 161a32929848..ab82965bc0f4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -3607,7 +3607,7 @@ static void rs_fill_lq_cmd(struct iwl_mvm *mvm,
 	    num_of_ant(initial_rate->ant) == 1)
 		lq_cmd->single_stream_ant_msk = initial_rate->ant;
 
-	lq_cmd->agg_frame_cnt_limit = mvmsta->max_agg_bufsize;
+	lq_cmd->agg_frame_cnt_limit = lq_sta->pers.max_agg_bufsize;
 
 	/*
 	 * In case of low latency, tell the firmware to leave a frame in the
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index c3145cb8f354..f99603b0f693 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -1,10 +1,9 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /******************************************************************************
  *
- * Copyright(c) 2003 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2015 Intel Mobile Communications GmbH
  * Copyright(c) 2017 Intel Deutschland GmbH
- * Copyright(c) 2018 - 2019 Intel Corporation
+ * Copyright (C) 2003 - 2014, 2018 - 2022 Intel Corporation
  *****************************************************************************/
 
 #ifndef __rs_h__
@@ -204,6 +203,7 @@ struct rs_rate {
 /**
  * struct iwl_lq_sta_rs_fw - rate and related statistics for RS in FW
  * @last_rate_n_flags: last rate reported by FW
+ * @max_agg_bufsize: the maximal size of the AGG buffer for this station
  * @sta_id: the id of the station
 #ifdef CONFIG_MAC80211_DEBUGFS
  * @dbg_fixed_rate: for debug, use fixed rate if not 0
@@ -353,6 +353,7 @@ struct iwl_lq_sta {
 
 	/* last tx rate_n_flags */
 	u32 last_rate_n_flags;
+
 	/* packets destined for this STA are aggregated */
 	u8 is_agg;
 
@@ -371,6 +372,7 @@ struct iwl_lq_sta {
 		u8 chains;
 		s8 chain_signal[IEEE80211_MAX_CHAINS];
 		s8 last_rssi;
+		u16 max_agg_bufsize;
 		struct rs_rate_stats tx_stats[RS_COLUMN_COUNT][IWL_RATE_COUNT];
 		struct iwl_mvm *drv;
 		spinlock_t lock; /* for races in reinit/update table */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 9a3a7eabded8..db6a3419bd31 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1719,10 +1719,10 @@ int iwl_mvm_sta_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		rcu_assign_pointer(mvm_sta->link[0], &mvm_sta->deflink);
 
 		if (!mvm->trans->trans_cfg->gen2)
-			mvm_sta->max_agg_bufsize =
+			mvm_sta->deflink.lq_sta.rs_drv.pers.max_agg_bufsize =
 				LINK_QUAL_AGG_FRAME_LIMIT_DEF;
 		else
-			mvm_sta->max_agg_bufsize =
+			mvm_sta->deflink.lq_sta.rs_drv.pers.max_agg_bufsize =
 				LINK_QUAL_AGG_FRAME_LIMIT_GEN2_DEF;
 	}
 
@@ -3258,10 +3258,11 @@ out:
 	 * for each station. Therefore, use the minimum of all the
 	 * aggregation sessions and our default value.
 	 */
-	mvmsta->max_agg_bufsize =
-		min(mvmsta->max_agg_bufsize, buf_size);
+	mvmsta->deflink.lq_sta.rs_drv.pers.max_agg_bufsize =
+		min(mvmsta->deflink.lq_sta.rs_drv.pers.max_agg_bufsize,
+		    buf_size);
 	mvmsta->deflink.lq_sta.rs_drv.lq.agg_frame_cnt_limit =
-		mvmsta->max_agg_bufsize;
+		mvmsta->deflink.lq_sta.rs_drv.pers.max_agg_bufsize;
 
 	IWL_DEBUG_HT(mvm, "Tx aggregation enabled on ra = %pM tid = %d\n",
 		     sta->addr, tid);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 824d12047019..625e1dfd47a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -355,7 +355,6 @@ struct iwl_mvm_link_sta {
  * @mac_id_n_color: the MAC context this station is linked to
  * @tid_disable_agg: bitmap: if bit(tid) is set, the fw won't send ampdus for
  *	tid.
- * @max_agg_bufsize: the maximal size of the AGG buffer for this station
  * @sta_type: station type
  * @sta_state: station state according to enum %ieee80211_sta_state
  * @bt_reduced_txpower: is reduced tx power enabled for this station
@@ -406,7 +405,6 @@ struct iwl_mvm_sta {
 	u32 tfd_queue_msk;
 	u32 mac_id_n_color;
 	u16 tid_disable_agg;
-	u16 max_agg_bufsize;
 	u8 sta_type;
 	enum ieee80211_sta_state sta_state;
 	bool bt_reduced_txpower;
-- 
cgit v1.2.3-70-g09d2


From e0c7ee3a20cee04d39ac466e7c4b74bfc1328f14 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 29 Mar 2023 10:05:37 +0300
Subject: wifi: iwlwifi: bump FW API to 75 for AX devices

Start supporting API version 75 for AX devices.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.f08a27944fc6.Iafe3a2db2b91072a559038b85eca7b6b322be3ff@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index e649daaf842c..0b10b34a77a8 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_22000_UCODE_API_MAX	74
+#define IWL_22000_UCODE_API_MAX	75
 
 /* Lowest firmware API version supported */
 #define IWL_22000_UCODE_API_MIN	39
-- 
cgit v1.2.3-70-g09d2


From 8ca86d61798f04c98fc2c303d52e552247dc433b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:38 +0300
Subject: wifi: iwlwifi: mvm: free probe_resp_data later

In the MLD code, we free probe_resp_data before we remove
the MAC from the firmware, so we might receive another one
from the device after freeing, and thus might leak it. Fix
that by moving the free later.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.152b1715fc13.Ibd37fed1b24cd25012923ad9170d1fe33ab35c5c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 4d56b2fc5f33..203f2513e7ea 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -159,12 +159,6 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 		mvm->csme_vif = NULL;
 	}
 
-	probe_data = rcu_dereference_protected(mvmvif->deflink.probe_resp_data,
-					       lockdep_is_held(&mvm->mutex));
-	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
-	if (probe_data)
-		kfree_rcu(probe_data, rcu_head);
-
 	if (mvm->bf_allowed_vif == mvmvif) {
 		mvm->bf_allowed_vif = NULL;
 		vif->driver_flags &= ~(IEEE80211_VIF_BEACON_FILTER |
@@ -207,6 +201,12 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 
 	RCU_INIT_POINTER(mvm->vif_id_to_mac[mvmvif->id], NULL);
 
+	probe_data = rcu_dereference_protected(mvmvif->deflink.probe_resp_data,
+					       lockdep_is_held(&mvm->mutex));
+	RCU_INIT_POINTER(mvmvif->deflink.probe_resp_data, NULL);
+	if (probe_data)
+		kfree_rcu(probe_data, rcu_head);
+
 	if (vif->type == NL80211_IFTYPE_MONITOR) {
 		mvm->monitor_on = false;
 		__clear_bit(IEEE80211_HW_RX_INCLUDES_FCS, mvm->hw->flags);
-- 
cgit v1.2.3-70-g09d2


From de50140b701911a34c4652d3a4212cce8f4fccd7 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:39 +0300
Subject: wifi: iwlwifi: separate AP link management queues

The link management queues associated with the broadcast stations
were forgotten and so the same queue was used with both broadcast
stations. This leads to lost frames and warnings on cleanup and
HW restart.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.0671fa976832.Id5aa9856fd5984e447f247e6d0c3979d9794a21a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c |  4 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h     |  3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c     | 16 ++++++++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c      |  4 ++--
 4 files changed, 21 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 26686cc7ff4c..ca788a427fa6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -181,7 +181,7 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	if (vif->type == NL80211_IFTYPE_AP ||
 	    vif->type == NL80211_IFTYPE_ADHOC) {
-		queue = &mvm->probe_queue;
+		queue = &mvm_link->mgmt_queue;
 	} else if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 		queue = &mvm->p2p_dev_queue;
 	} else {
@@ -327,7 +327,7 @@ int iwl_mvm_mld_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	switch (vif->type) {
 	case NL80211_IFTYPE_AP:
 	case NL80211_IFTYPE_ADHOC:
-		queueptr = &mvm->probe_queue;
+		queueptr = &link->mgmt_queue;
 		break;
 	case NL80211_IFTYPE_P2P_DEVICE:
 		queueptr = &mvm->p2p_dev_queue;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 022f7bd64a5d..707c930a278f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -299,6 +299,7 @@ struct iwl_probe_resp_data {
  *	data to be inserted into probe response.
  * @he_ru_2mhz_block: 26-tone RU OFDMA transmissions should be blocked
  * @queue_params: QoS params for this MAC
+ * @mgmt_queue: queue number for unbufferable management frames
  */
 struct iwl_mvm_vif_link_info {
 	u8 bssid[ETH_ALEN];
@@ -329,6 +330,8 @@ struct iwl_mvm_vif_link_info {
 	 * to have the data for the MAC context
 	 */
 	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
+
+	u16 mgmt_queue;
 };
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index db6a3419bd31..a57de37f6e02 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2348,10 +2348,18 @@ int iwl_mvm_send_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 		}
 
 		if (vif->type == NL80211_IFTYPE_AP ||
-		    vif->type == NL80211_IFTYPE_ADHOC)
+		    vif->type == NL80211_IFTYPE_ADHOC) {
+			/* for queue management */
 			mvm->probe_queue = queue;
-		else if (vif->type == NL80211_IFTYPE_P2P_DEVICE)
+			/* for use in TX */
+			mvmvif->deflink.mgmt_queue = queue;
+		} else if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
 			mvm->p2p_dev_queue = queue;
+		}
+	} else if (vif->type == NL80211_IFTYPE_AP ||
+		   vif->type == NL80211_IFTYPE_ADHOC) {
+		/* set it for use in TX */
+		mvmvif->deflink.mgmt_queue = mvm->probe_queue;
 	}
 
 	return 0;
@@ -2384,6 +2392,10 @@ void iwl_mvm_free_bcast_sta_queues(struct iwl_mvm *mvm,
 	queue = *queueptr;
 	iwl_mvm_disable_txq(mvm, NULL, mvmvif->deflink.bcast_sta.sta_id,
 			    queueptr, IWL_MAX_TID_COUNT);
+
+	if (vif->type == NL80211_IFTYPE_AP || vif->type == NL80211_IFTYPE_ADHOC)
+		mvmvif->deflink.mgmt_queue = mvm->probe_queue;
+
 	if (iwl_mvm_has_new_tx_api(mvm))
 		return;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index f5b92d99f39d..2c842938656d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -626,7 +626,7 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 		if (ieee80211_is_mgmt(fc) &&
 		    (!ieee80211_is_bufferable_mmpdu(fc) ||
 		     ieee80211_is_deauth(fc) || ieee80211_is_disassoc(fc)))
-			return mvm->probe_queue;
+			return link->mgmt_queue;
 
 		if (!ieee80211_has_order(fc) && !ieee80211_is_probe_req(fc) &&
 		    is_multicast_ether_addr(hdr->addr1))
@@ -634,7 +634,7 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 
 		WARN_ONCE(info->control.vif->type != NL80211_IFTYPE_ADHOC,
 			  "fc=0x%02x", le16_to_cpu(fc));
-		return mvm->probe_queue;
+		return link->mgmt_queue;
 	case NL80211_IFTYPE_P2P_DEVICE:
 		if (ieee80211_is_mgmt(fc))
 			return mvm->p2p_dev_queue;
-- 
cgit v1.2.3-70-g09d2


From a6ef8a88fc38cc1bf5c440be73fe520d04301424 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 29 Mar 2023 10:05:40 +0300
Subject: wifi: iwlwifi: mvm: correctly use link in iwl_mvm_sta_del()

This function can be invoked for both MLO and non-MLO, so
it must deal with multi-link correctly. Notable, on auth
timeout, we'd otherwise get a warning due to the erroneous
deflink usage in MLO cases.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230329100040.b85f6052d51a.Iedfef4b4c4f3ca557aebc0093fdc3f5cfb49b507@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c     | 18 +++++++++++++-----
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h     |  2 +-
 3 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index ca788a427fa6..78d4f186cd99 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -780,7 +780,7 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 						  lockdep_is_held(&mvm->mutex));
 		bool stay_in_fw;
 
-		stay_in_fw = iwl_mvm_sta_del(mvm, vif, sta, mvm_link_sta, &ret);
+		stay_in_fw = iwl_mvm_sta_del(mvm, vif, sta, link_sta, &ret);
 		if (ret)
 			break;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index a57de37f6e02..8489554e58b1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1986,14 +1986,22 @@ int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
  */
 bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		     struct ieee80211_sta *sta,
-		     struct iwl_mvm_link_sta *mvm_link_sta, int *ret)
+		     struct ieee80211_link_sta *link_sta, int *ret)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif_link_info *mvm_link =
+		mvmvif->link[link_sta->link_id];
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-	u8 sta_id = mvm_link_sta->sta_id;
+	struct iwl_mvm_link_sta *mvm_link_sta;
+	u8 sta_id;
 
 	lockdep_assert_held(&mvm->mutex);
 
+	mvm_link_sta =
+		rcu_dereference_protected(mvm_sta->link[link_sta->link_id],
+					  lockdep_is_held(&mvm->mutex));
+	sta_id = mvm_link_sta->sta_id;
+
 	/* If there is a TXQ still marked as reserved - free it */
 	if (mvm_sta->reserved_queue != IEEE80211_INVAL_HW_QUEUE) {
 		u8 reserved_txq = mvm_sta->reserved_queue;
@@ -2022,10 +2030,10 @@ bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			return true;
 
 		/* first remove remaining keys */
-		iwl_mvm_sec_key_remove_ap(mvm, vif, &mvmvif->deflink, 0);
+		iwl_mvm_sec_key_remove_ap(mvm, vif, mvm_link, 0);
 
 		/* unassoc - go ahead - remove the AP STA now */
-		mvmvif->deflink.ap_sta_id = IWL_MVM_INVALID_STA;
+		mvm_link->ap_sta_id = IWL_MVM_INVALID_STA;
 	}
 
 	/*
@@ -2082,7 +2090,7 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 
 	iwl_mvm_disable_sta_queues(mvm, vif, sta);
 
-	if (iwl_mvm_sta_del(mvm, vif, sta, &mvm_sta->deflink, &ret))
+	if (iwl_mvm_sta_del(mvm, vif, sta, &sta->deflink, &ret))
 		return ret;
 
 	ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->deflink.sta_id);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 625e1dfd47a2..7b9e91935aa0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -489,7 +489,7 @@ int iwl_mvm_wait_sta_queues_empty(struct iwl_mvm *mvm,
 				  struct iwl_mvm_sta *mvm_sta);
 bool iwl_mvm_sta_del(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		     struct ieee80211_sta *sta,
-		     struct iwl_mvm_link_sta *mvm_link_sta, int *ret);
+		     struct ieee80211_link_sta *link_sta, int *ret);
 int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 		   struct ieee80211_vif *vif,
 		   struct ieee80211_sta *sta);
-- 
cgit v1.2.3-70-g09d2


From 8fcb76b934daff12cde76adeab3d502eeb0734b1 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 Mar 2023 23:50:18 +0000
Subject: net: napi_schedule_rps() cleanup

napi_schedule_rps() return value is ignored, remove it.

Change the comment to clarify the intent.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Tested-by: Jason Xing <kerneljasonxing@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/dev.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 7172334a418f..f7050b95d125 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4582,11 +4582,18 @@ static void trigger_rx_softirq(void *data)
 }
 
 /*
- * Check if this softnet_data structure is another cpu one
- * If yes, queue it to our IPI list and return 1
- * If no, return 0
+ * After we queued a packet into sd->input_pkt_queue,
+ * we need to make sure this queue is serviced soon.
+ *
+ * - If this is another cpu queue, link it to our rps_ipi_list,
+ *   and make sure we will process rps_ipi_list from net_rx_action().
+ *   As we do not know yet if we are called from net_rx_action(),
+ *   we have to raise NET_RX_SOFTIRQ. This might change in the future.
+ *
+ * - If this is our own queue, NAPI schedule our backlog.
+ *   Note that this also raises NET_RX_SOFTIRQ.
  */
-static int napi_schedule_rps(struct softnet_data *sd)
+static void napi_schedule_rps(struct softnet_data *sd)
 {
 	struct softnet_data *mysd = this_cpu_ptr(&softnet_data);
 
@@ -4596,11 +4603,10 @@ static int napi_schedule_rps(struct softnet_data *sd)
 		mysd->rps_ipi_list = sd;
 
 		__raise_softirq_irqoff(NET_RX_SOFTIRQ);
-		return 1;
+		return;
 	}
 #endif /* CONFIG_RPS */
 	__napi_schedule_irqoff(&mysd->backlog);
-	return 0;
 }
 
 #ifdef CONFIG_NET_FLOW_LIMIT
-- 
cgit v1.2.3-70-g09d2


From c59647c0dc679008886756a888368da1c6d4ccd3 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 Mar 2023 23:50:19 +0000
Subject: net: add softnet_data.in_net_rx_action

We want to make two optimizations in napi_schedule_rps() and
____napi_schedule() which require to know if these helpers are
called from net_rx_action(), instead of being called from
other contexts.

sd.in_net_rx_action is only read/written by the owning cpu.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Tested-by: Jason Xing <kerneljasonxing@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/linux/netdevice.h | 1 +
 net/core/dev.c            | 4 ++++
 2 files changed, 5 insertions(+)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 18a5be6ddd0f..c8c634091a65 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3188,6 +3188,7 @@ struct softnet_data {
 #ifdef CONFIG_RPS
 	struct softnet_data	*rps_ipi_list;
 #endif
+	bool			in_net_rx_action;
 #ifdef CONFIG_NET_FLOW_LIMIT
 	struct sd_flow_limit __rcu *flow_limit;
 #endif
diff --git a/net/core/dev.c b/net/core/dev.c
index f7050b95d125..15331edbacf4 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6646,6 +6646,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
 	LIST_HEAD(list);
 	LIST_HEAD(repoll);
 
+	sd->in_net_rx_action = true;
 	local_irq_disable();
 	list_splice_init(&sd->poll_list, &list);
 	local_irq_enable();
@@ -6656,6 +6657,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
 		skb_defer_free_flush(sd);
 
 		if (list_empty(&list)) {
+			sd->in_net_rx_action = false;
 			if (!sd_has_rps_ipi_waiting(sd) && list_empty(&repoll))
 				goto end;
 			break;
@@ -6682,6 +6684,8 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
 	list_splice(&list, &sd->poll_list);
 	if (!list_empty(&sd->poll_list))
 		__raise_softirq_irqoff(NET_RX_SOFTIRQ);
+	else
+		sd->in_net_rx_action = false;
 
 	net_rps_action_and_irq_enable(sd);
 end:;
-- 
cgit v1.2.3-70-g09d2


From 821eba962d95806beb0440742c4062a9da8a386b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 Mar 2023 23:50:20 +0000
Subject: net: optimize napi_schedule_rps()

Based on initial patch from Jason Xing.

Idea is to not raise NET_RX_SOFTIRQ from napi_schedule_rps()
when we queued a packet into another cpu backlog.

We can do this only in the context of us being called indirectly
from net_rx_action(), to have the guarantee our rps_ipi_list
will be processed before we exit from net_rx_action().

Link: https://lore.kernel.org/lkml/20230325152417.5403-1-kerneljasonxing@gmail.com/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Tested-by: Jason Xing <kerneljasonxing@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/dev.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 15331edbacf4..f34ce93f2f02 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4587,8 +4587,6 @@ static void trigger_rx_softirq(void *data)
  *
  * - If this is another cpu queue, link it to our rps_ipi_list,
  *   and make sure we will process rps_ipi_list from net_rx_action().
- *   As we do not know yet if we are called from net_rx_action(),
- *   we have to raise NET_RX_SOFTIRQ. This might change in the future.
  *
  * - If this is our own queue, NAPI schedule our backlog.
  *   Note that this also raises NET_RX_SOFTIRQ.
@@ -4602,7 +4600,11 @@ static void napi_schedule_rps(struct softnet_data *sd)
 		sd->rps_ipi_next = mysd->rps_ipi_list;
 		mysd->rps_ipi_list = sd;
 
-		__raise_softirq_irqoff(NET_RX_SOFTIRQ);
+		/* If not called from net_rx_action()
+		 * we have to raise NET_RX_SOFTIRQ.
+		 */
+		if (!mysd->in_net_rx_action)
+			__raise_softirq_irqoff(NET_RX_SOFTIRQ);
 		return;
 	}
 #endif /* CONFIG_RPS */
-- 
cgit v1.2.3-70-g09d2


From 8b43fd3d1d7d88293eb15e92090826e6b7cc13e4 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 Mar 2023 23:50:21 +0000
Subject: net: optimize ____napi_schedule() to avoid extra NET_RX_SOFTIRQ

____napi_schedule() adds a napi into current cpu softnet_data poll_list,
then raises NET_RX_SOFTIRQ to make sure net_rx_action() will process it.

Idea of this patch is to not raise NET_RX_SOFTIRQ when being called indirectly
from net_rx_action(), because we can process poll_list from this point,
without going to full softirq loop.

This needs a change in net_rx_action() to make sure we restart
its main loop if sd->poll_list was updated without NET_RX_SOFTIRQ
being raised.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Tested-by: Jason Xing <kerneljasonxing@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/dev.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index f34ce93f2f02..0c4b21291348 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4360,7 +4360,11 @@ static inline void ____napi_schedule(struct softnet_data *sd,
 	}
 
 	list_add_tail(&napi->poll_list, &sd->poll_list);
-	__raise_softirq_irqoff(NET_RX_SOFTIRQ);
+	/* If not called from net_rx_action()
+	 * we have to raise NET_RX_SOFTIRQ.
+	 */
+	if (!sd->in_net_rx_action)
+		__raise_softirq_irqoff(NET_RX_SOFTIRQ);
 }
 
 #ifdef CONFIG_RPS
@@ -6648,6 +6652,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
 	LIST_HEAD(list);
 	LIST_HEAD(repoll);
 
+start:
 	sd->in_net_rx_action = true;
 	local_irq_disable();
 	list_splice_init(&sd->poll_list, &list);
@@ -6659,9 +6664,18 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
 		skb_defer_free_flush(sd);
 
 		if (list_empty(&list)) {
-			sd->in_net_rx_action = false;
-			if (!sd_has_rps_ipi_waiting(sd) && list_empty(&repoll))
-				goto end;
+			if (list_empty(&repoll)) {
+				sd->in_net_rx_action = false;
+				barrier();
+				/* We need to check if ____napi_schedule()
+				 * had refilled poll_list while
+				 * sd->in_net_rx_action was true.
+				 */
+				if (!list_empty(&sd->poll_list))
+					goto start;
+				if (!sd_has_rps_ipi_waiting(sd))
+					goto end;
+			}
 			break;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From da617cd8d90608582eb8d0b58026f31f1a9bfb1d Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Wed, 29 Mar 2023 08:44:14 +0200
Subject: smsc911x: remove superfluous variable init

phydev is assigned a value right away, no need to initialize it.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20230329064414.25028-1-wsa+renesas@sang-engineering.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/smsc/smsc911x.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c
index 39446d4e94b6..c653ea792dd1 100644
--- a/drivers/net/ethernet/smsc/smsc911x.c
+++ b/drivers/net/ethernet/smsc/smsc911x.c
@@ -1016,7 +1016,7 @@ static void smsc911x_phy_adjust_link(struct net_device *dev)
 static int smsc911x_mii_probe(struct net_device *dev)
 {
 	struct smsc911x_data *pdata = netdev_priv(dev);
-	struct phy_device *phydev = NULL;
+	struct phy_device *phydev;
 	int ret;
 
 	/* find the first phy */
-- 
cgit v1.2.3-70-g09d2


From 8ba732befd6ff44cd8ecc809bd6df850e246f31a Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Wed, 29 Mar 2023 10:22:55 -0700
Subject: Documentation/eth/intel: Update address for driver support

Update the email address for support to use Intel Wired LAN, the mailing
list used for kernel development.

Suggested-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
---
 Documentation/networking/device_drivers/ethernet/intel/e100.rst    | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/e1000.rst   | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/e1000e.rst  | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/fm10k.rst   | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/i40e.rst    | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/iavf.rst    | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/ice.rst     | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/igb.rst     | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/igbvf.rst   | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst   | 2 +-
 Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst | 2 +-
 11 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/intel/e100.rst b/Documentation/networking/device_drivers/ethernet/intel/e100.rst
index 371b7e5c3293..4f613949782c 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e100.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e100.rst
@@ -184,4 +184,4 @@ or the Intel Wired Networking project hosted by Sourceforge at:
 http://sourceforge.net/projects/e1000
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/e1000.rst b/Documentation/networking/device_drivers/ethernet/intel/e1000.rst
index 4aaae0f7d6ba..7b15b8c72be0 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e1000.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e1000.rst
@@ -460,4 +460,4 @@ or the Intel Wired Networking project hosted by Sourceforge at:
 
 If an issue is identified with the released source code on the supported
 kernel with a supported adapter, email the specific information related
-to the issue to e1000-devel@lists.sf.net
+to the issue to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst b/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst
index f49cd370e7bf..7a9cbfa9e0f3 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst
@@ -380,4 +380,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst b/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst
index 9258ef6f515c..9d9c2ec2152e 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst
@@ -139,4 +139,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/i40e.rst b/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
index c495c4e16b3b..5b13fe0fec82 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
@@ -768,4 +768,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/iavf.rst b/Documentation/networking/device_drivers/ethernet/intel/iavf.rst
index 151af0a8da9c..079847666125 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/iavf.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/iavf.rst
@@ -328,4 +328,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on the supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ice.rst b/Documentation/networking/device_drivers/ethernet/intel/ice.rst
index 2b6dc7880d7b..246bf6455f64 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ice.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ice.rst
@@ -1031,7 +1031,7 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
 
 
 Trademarks
diff --git a/Documentation/networking/device_drivers/ethernet/intel/igb.rst b/Documentation/networking/device_drivers/ethernet/intel/igb.rst
index d46289e182cf..ee149bdb42b9 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/igb.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/igb.rst
@@ -210,4 +210,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst b/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst
index 40fa210c5e14..78ceb3cdbfdb 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst
@@ -62,4 +62,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst b/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst
index 0a233b17c664..8d4f7ede2ff8 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst
@@ -554,4 +554,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst b/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst
index 76bbde736f21..9201c74e4c36 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst
@@ -64,4 +64,4 @@ https://sourceforge.net/projects/e1000
 
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
-to e1000-devel@lists.sf.net.
+to intel-wired-lan@lists.osuosl.org.
-- 
cgit v1.2.3-70-g09d2


From 79d872c62b16e6c640e125fd4613c6beadac5210 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Wed, 29 Mar 2023 10:22:56 -0700
Subject: Documentation/eth/intel: Remove references to SourceForge

The out-of-tree driver is hosted on SourceForge, as this does not apply
to the kernel driver remove references to it. Also do some minor
formatting changes around this section.

Suggested-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
---
 Documentation/networking/device_drivers/ethernet/intel/e100.rst    | 2 --
 Documentation/networking/device_drivers/ethernet/intel/e1000.rst   | 7 +------
 Documentation/networking/device_drivers/ethernet/intel/e1000e.rst  | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/fm10k.rst   | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/i40e.rst    | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/iavf.rst    | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/ice.rst     | 3 ---
 Documentation/networking/device_drivers/ethernet/intel/igb.rst     | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/igbvf.rst   | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst   | 5 -----
 Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst | 5 -----
 11 files changed, 1 insertion(+), 51 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/intel/e100.rst b/Documentation/networking/device_drivers/ethernet/intel/e100.rst
index 4f613949782c..5dee1b53e977 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e100.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e100.rst
@@ -180,8 +180,6 @@ Support
 For general information, go to the Intel support website at:
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-http://sourceforge.net/projects/e1000
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/e1000.rst b/Documentation/networking/device_drivers/ethernet/intel/e1000.rst
index 7b15b8c72be0..52a7fb9ce8d9 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e1000.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e1000.rst
@@ -451,12 +451,7 @@ Support
 =======
 
 For general information, go to the Intel support website at:
-
-    http://support.intel.com
-
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-    http://sourceforge.net/projects/e1000
+http://support.intel.com
 
 If an issue is identified with the released source code on the supported
 kernel with a supported adapter, email the specific information related
diff --git a/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst b/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst
index 7a9cbfa9e0f3..d8f810afdd49 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/e1000e.rst
@@ -371,13 +371,8 @@ NOTE: Wake on LAN is only supported on port A for the following devices:
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst b/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst
index 9d9c2ec2152e..396a2c8c3db1 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/fm10k.rst
@@ -130,13 +130,8 @@ the Intel Ethernet Controller XL710.
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/i40e.rst b/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
index 5b13fe0fec82..4fbaa1a2d674 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/i40e.rst
@@ -759,13 +759,8 @@ enabled when setting up DCB on your switch.
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/iavf.rst b/Documentation/networking/device_drivers/ethernet/intel/iavf.rst
index 079847666125..eb926c3bd4cd 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/iavf.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/iavf.rst
@@ -319,13 +319,8 @@ This is caused by the way the Linux kernel reports this stressed condition.
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://support.intel.com
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on the supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ice.rst b/Documentation/networking/device_drivers/ethernet/intel/ice.rst
index 246bf6455f64..69695e5511f4 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ice.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ice.rst
@@ -1026,9 +1026,6 @@ Support
 For general information, go to the Intel support website at:
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/igb.rst b/Documentation/networking/device_drivers/ethernet/intel/igb.rst
index ee149bdb42b9..fbd590b6a0d6 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/igb.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/igb.rst
@@ -201,13 +201,8 @@ NOTE: This feature is exclusive to i210 models.
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst b/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst
index 78ceb3cdbfdb..11a9017f3069 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/igbvf.rst
@@ -53,13 +53,8 @@ https://www.kernel.org/pub/software/network/ethtool/
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst b/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst
index 8d4f7ede2ff8..1e5f16993f69 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ixgbe.rst
@@ -545,13 +545,8 @@ on the Intel Ethernet Controller XL710.
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
diff --git a/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst b/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst
index 9201c74e4c36..08dc0d368a48 100644
--- a/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst
+++ b/Documentation/networking/device_drivers/ethernet/intel/ixgbevf.rst
@@ -55,13 +55,8 @@ VLANs: There is a limit of a total of 64 shared VLANs to 1 or more VFs.
 Support
 =======
 For general information, go to the Intel support website at:
-
 https://www.intel.com/support/
 
-or the Intel Wired Networking project hosted by Sourceforge at:
-
-https://sourceforge.net/projects/e1000
-
 If an issue is identified with the released source code on a supported kernel
 with a supported adapter, email the specific information related to the issue
 to intel-wired-lan@lists.osuosl.org.
-- 
cgit v1.2.3-70-g09d2


From 503d473c983bf04b978d97f522e6039723caf224 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Mon, 27 Feb 2023 14:15:00 -0800
Subject: ice: remove comment about not supporting driver reinit

Since commit 31c8db2c4fa7 ("ice: implement devlink reinit action"), the ice
driver does support driver re-initialization via devlink reload. Remove the
stale comment indicating that the driver lacks this support from the
ice_devlink_ops structure.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_devlink.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.c b/drivers/net/ethernet/intel/ice/ice_devlink.c
index 05f216af8c81..bc44cc220818 100644
--- a/drivers/net/ethernet/intel/ice/ice_devlink.c
+++ b/drivers/net/ethernet/intel/ice/ice_devlink.c
@@ -1254,7 +1254,6 @@ static const struct devlink_ops ice_devlink_ops = {
 	.supported_flash_update_params = DEVLINK_SUPPORT_FLASH_UPDATE_OVERWRITE_MASK,
 	.reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
 			  BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE),
-	/* The ice driver currently does not support driver reinit */
 	.reload_down = ice_devlink_reload_down,
 	.reload_up = ice_devlink_reload_up,
 	.port_split = ice_devlink_port_split,
-- 
cgit v1.2.3-70-g09d2


From 356e2adb3f3b9fe4427224eec2f8819a3078665f Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 28 Mar 2023 18:44:15 +0200
Subject: netfilter: nfnetlink_log: remove rcu_bh usage

structure is free'd via call_rcu, so its safe to use rcu_read_lock only.

While at it, skip rcu_read_lock for lookup from packet path, its always
called with rcu held.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nfnetlink_log.c | 36 +++++++++++++++++++++++-------------
 1 file changed, 23 insertions(+), 13 deletions(-)

diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index d97eb280cb2e..e57eb168ee13 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -103,9 +103,9 @@ static inline u_int8_t instance_hashfn(u_int16_t group_num)
 }
 
 static struct nfulnl_instance *
-__instance_lookup(struct nfnl_log_net *log, u_int16_t group_num)
+__instance_lookup(const struct nfnl_log_net *log, u16 group_num)
 {
-	struct hlist_head *head;
+	const struct hlist_head *head;
 	struct nfulnl_instance *inst;
 
 	head = &log->instance_table[instance_hashfn(group_num)];
@@ -123,15 +123,25 @@ instance_get(struct nfulnl_instance *inst)
 }
 
 static struct nfulnl_instance *
-instance_lookup_get(struct nfnl_log_net *log, u_int16_t group_num)
+instance_lookup_get_rcu(const struct nfnl_log_net *log, u16 group_num)
 {
 	struct nfulnl_instance *inst;
 
-	rcu_read_lock_bh();
 	inst = __instance_lookup(log, group_num);
 	if (inst && !refcount_inc_not_zero(&inst->use))
 		inst = NULL;
-	rcu_read_unlock_bh();
+
+	return inst;
+}
+
+static struct nfulnl_instance *
+instance_lookup_get(const struct nfnl_log_net *log, u16 group_num)
+{
+	struct nfulnl_instance *inst;
+
+	rcu_read_lock();
+	inst = instance_lookup_get_rcu(log, group_num);
+	rcu_read_unlock();
 
 	return inst;
 }
@@ -698,7 +708,7 @@ nfulnl_log_packet(struct net *net,
 	else
 		li = &default_loginfo;
 
-	inst = instance_lookup_get(log, li->u.ulog.group);
+	inst = instance_lookup_get_rcu(log, li->u.ulog.group);
 	if (!inst)
 		return;
 
@@ -1030,7 +1040,7 @@ static struct hlist_node *get_first(struct net *net, struct iter_state *st)
 		struct hlist_head *head = &log->instance_table[st->bucket];
 
 		if (!hlist_empty(head))
-			return rcu_dereference_bh(hlist_first_rcu(head));
+			return rcu_dereference(hlist_first_rcu(head));
 	}
 	return NULL;
 }
@@ -1038,7 +1048,7 @@ static struct hlist_node *get_first(struct net *net, struct iter_state *st)
 static struct hlist_node *get_next(struct net *net, struct iter_state *st,
 				   struct hlist_node *h)
 {
-	h = rcu_dereference_bh(hlist_next_rcu(h));
+	h = rcu_dereference(hlist_next_rcu(h));
 	while (!h) {
 		struct nfnl_log_net *log;
 		struct hlist_head *head;
@@ -1048,7 +1058,7 @@ static struct hlist_node *get_next(struct net *net, struct iter_state *st,
 
 		log = nfnl_log_pernet(net);
 		head = &log->instance_table[st->bucket];
-		h = rcu_dereference_bh(hlist_first_rcu(head));
+		h = rcu_dereference(hlist_first_rcu(head));
 	}
 	return h;
 }
@@ -1066,9 +1076,9 @@ static struct hlist_node *get_idx(struct net *net, struct iter_state *st,
 }
 
 static void *seq_start(struct seq_file *s, loff_t *pos)
-	__acquires(rcu_bh)
+	__acquires(rcu)
 {
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	return get_idx(seq_file_net(s), s->private, *pos);
 }
 
@@ -1079,9 +1089,9 @@ static void *seq_next(struct seq_file *s, void *v, loff_t *pos)
 }
 
 static void seq_stop(struct seq_file *s, void *v)
-	__releases(rcu_bh)
+	__releases(rcu)
 {
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 }
 
 static int seq_show(struct seq_file *s, void *v)
-- 
cgit v1.2.3-70-g09d2


From 28c1b6df436819a7ed8a781835766e45139771a3 Mon Sep 17 00:00:00 2001
From: Eric Sage <eric_sage@apple.com>
Date: Mon, 27 Mar 2023 13:44:49 -0400
Subject: netfilter: nfnetlink_queue: enable classid socket info retrieval

This enables associating a socket with a v1 net_cls cgroup. Useful for
applying a per-cgroup policy when processing packets in userspace.

Signed-off-by: Eric Sage <eric_sage@apple.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/uapi/linux/netfilter/nfnetlink_queue.h |  1 +
 net/netfilter/nfnetlink_queue.c                | 20 ++++++++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/include/uapi/linux/netfilter/nfnetlink_queue.h b/include/uapi/linux/netfilter/nfnetlink_queue.h
index ef7c97f21a15..efcb7c044a74 100644
--- a/include/uapi/linux/netfilter/nfnetlink_queue.h
+++ b/include/uapi/linux/netfilter/nfnetlink_queue.h
@@ -62,6 +62,7 @@ enum nfqnl_attr_type {
 	NFQA_VLAN,			/* nested attribute: packet vlan info */
 	NFQA_L2HDR,			/* full L2 header */
 	NFQA_PRIORITY,			/* skb->priority */
+	NFQA_CGROUP_CLASSID,		/* __u32 cgroup classid */
 
 	__NFQA_MAX
 };
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 87a9009d5234..e311462f6d98 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -29,6 +29,7 @@
 #include <linux/netfilter/nfnetlink_queue.h>
 #include <linux/netfilter/nf_conntrack_common.h>
 #include <linux/list.h>
+#include <linux/cgroup-defs.h>
 #include <net/sock.h>
 #include <net/tcp_states.h>
 #include <net/netfilter/nf_queue.h>
@@ -301,6 +302,19 @@ nla_put_failure:
 	return -1;
 }
 
+static int nfqnl_put_sk_classid(struct sk_buff *skb, struct sock *sk)
+{
+#if IS_ENABLED(CONFIG_CGROUP_NET_CLASSID)
+	if (sk && sk_fullsock(sk)) {
+		u32 classid = sock_cgroup_classid(&sk->sk_cgrp_data);
+
+		if (classid && nla_put_be32(skb, NFQA_CGROUP_CLASSID, htonl(classid)))
+			return -1;
+	}
+#endif
+	return 0;
+}
+
 static u32 nfqnl_get_sk_secctx(struct sk_buff *skb, char **secdata)
 {
 	u32 seclen = 0;
@@ -406,6 +420,9 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 		+ nla_total_size(sizeof(u_int32_t))	/* priority */
 		+ nla_total_size(sizeof(struct nfqnl_msg_packet_hw))
 		+ nla_total_size(sizeof(u_int32_t))	/* skbinfo */
+#if IS_ENABLED(CONFIG_CGROUP_NET_CLASSID)
+		+ nla_total_size(sizeof(u_int32_t))	/* classid */
+#endif
 		+ nla_total_size(sizeof(u_int32_t));	/* cap_len */
 
 	tstamp = skb_tstamp_cond(entskb, false);
@@ -599,6 +616,9 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 	    nfqnl_put_sk_uidgid(skb, entskb->sk) < 0)
 		goto nla_put_failure;
 
+	if (nfqnl_put_sk_classid(skb, entskb->sk) < 0)
+		goto nla_put_failure;
+
 	if (seclen && nla_put(skb, NFQA_SECCTX, seclen, secdata))
 		goto nla_put_failure;
 
-- 
cgit v1.2.3-70-g09d2


From a25b8b7136ad43760bd876af62b6e59abd30496c Mon Sep 17 00:00:00 2001
From: Matthieu De Beule <matthieu.debeule@proton.ch>
Date: Wed, 29 Mar 2023 12:52:18 +0000
Subject: netfilter: Correct documentation errors in nf_tables.h

NFTA_RANGE_OP incorrectly says nft_cmp_ops instead of nft_range_ops.
NFTA_LOG_GROUP and NFTA_LOG_QTHRESHOLD claim NLA_U32 instead of NLA_U16
NFTA_EXTHDR_SREG isn't documented as a register

Signed-off-by: Matthieu De Beule <matthieu.debeule@proton.ch>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/uapi/linux/netfilter/nf_tables.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 9c6f02c26054..c4d4d8e42dc8 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -685,7 +685,7 @@ enum nft_range_ops {
  * enum nft_range_attributes - nf_tables range expression netlink attributes
  *
  * @NFTA_RANGE_SREG: source register of data to compare (NLA_U32: nft_registers)
- * @NFTA_RANGE_OP: cmp operation (NLA_U32: nft_cmp_ops)
+ * @NFTA_RANGE_OP: cmp operation (NLA_U32: nft_range_ops)
  * @NFTA_RANGE_FROM_DATA: data range from (NLA_NESTED: nft_data_attributes)
  * @NFTA_RANGE_TO_DATA: data range to (NLA_NESTED: nft_data_attributes)
  */
@@ -878,7 +878,7 @@ enum nft_exthdr_op {
  * @NFTA_EXTHDR_LEN: extension header length (NLA_U32)
  * @NFTA_EXTHDR_FLAGS: extension header flags (NLA_U32)
  * @NFTA_EXTHDR_OP: option match type (NLA_U32)
- * @NFTA_EXTHDR_SREG: option match type (NLA_U32)
+ * @NFTA_EXTHDR_SREG: source register (NLA_U32: nft_registers)
  */
 enum nft_exthdr_attributes {
 	NFTA_EXTHDR_UNSPEC,
@@ -1262,10 +1262,10 @@ enum nft_last_attributes {
 /**
  * enum nft_log_attributes - nf_tables log expression netlink attributes
  *
- * @NFTA_LOG_GROUP: netlink group to send messages to (NLA_U32)
+ * @NFTA_LOG_GROUP: netlink group to send messages to (NLA_U16)
  * @NFTA_LOG_PREFIX: prefix to prepend to log messages (NLA_STRING)
  * @NFTA_LOG_SNAPLEN: length of payload to include in netlink message (NLA_U32)
- * @NFTA_LOG_QTHRESHOLD: queue threshold (NLA_U32)
+ * @NFTA_LOG_QTHRESHOLD: queue threshold (NLA_U16)
  * @NFTA_LOG_LEVEL: log level (NLA_U32)
  * @NFTA_LOG_FLAGS: logging flags (NLA_U32)
  */
-- 
cgit v1.2.3-70-g09d2


From 9b7c68b3911aef84afa4cbfc31bce20f10570d51 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Wed, 22 Mar 2023 09:35:32 +0200
Subject: netfilter: ctnetlink: Support offloaded conntrack entry deletion

Currently, offloaded conntrack entries (flows) can only be deleted
after they are removed from offload, which is either by timeout,
tcp state change or tc ct rule deletion. This can cause issues for
users wishing to manually delete or flush existing entries.

Support deletion of offloaded conntrack entries.

Example usage:
 # Delete all offloaded (and non offloaded) conntrack entries
 # whose source address is 1.2.3.4
 $ conntrack -D -s 1.2.3.4
 # Delete all entries
 $ conntrack -F

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nf_conntrack_netlink.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index bfc3aaa2c872..fbc47e4b7bc3 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1554,9 +1554,6 @@ static const struct nla_policy ct_nla_policy[CTA_MAX+1] = {
 
 static int ctnetlink_flush_iterate(struct nf_conn *ct, void *data)
 {
-	if (test_bit(IPS_OFFLOAD_BIT, &ct->status))
-		return 0;
-
 	return ctnetlink_filter_match(ct, data);
 }
 
@@ -1626,11 +1623,6 @@ static int ctnetlink_del_conntrack(struct sk_buff *skb,
 
 	ct = nf_ct_tuplehash_to_ctrack(h);
 
-	if (test_bit(IPS_OFFLOAD_BIT, &ct->status)) {
-		nf_ct_put(ct);
-		return -EBUSY;
-	}
-
 	if (cda[CTA_ID]) {
 		__be32 id = nla_get_be32(cda[CTA_ID]);
 
-- 
cgit v1.2.3-70-g09d2


From aa2aa818cd1198cfa2498116d57cd9f13fea80e4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 30 Mar 2023 22:48:45 +0200
Subject: wifi: clean up erroneously introduced file

Evidently Gregory sent this file but I (apparently every else) missed
it entirely, remove that.

Fixes: cf85123a210f ("wifi: iwlwifi: mvm: support enabling and disabling HW timestamping")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 intc-scripts/chromeOS/adjustments/024-time-sync.spatch | 6 ------
 1 file changed, 6 deletions(-)
 delete mode 100644 intc-scripts/chromeOS/adjustments/024-time-sync.spatch

diff --git a/intc-scripts/chromeOS/adjustments/024-time-sync.spatch b/intc-scripts/chromeOS/adjustments/024-time-sync.spatch
deleted file mode 100644
index 425e4700b343..000000000000
--- a/intc-scripts/chromeOS/adjustments/024-time-sync.spatch
+++ /dev/null
@@ -1,6 +0,0 @@
-@@
-struct ieee80211_hw *hw;
-expression val;
-@@
--hw->wiphy->hw_timestamp_max_peers = val
-+hw_timestamp_max_peers(hw)
-- 
cgit v1.2.3-70-g09d2


From d816129530e77b905b492631651eb09a18789692 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 30 Mar 2023 12:01:15 -0700
Subject: veristat: change guess for __sk_buff from CGROUP_SKB to SCHED_CLS

SCHED_CLS seems to be a better option as a default guess for freplace
programs that have __sk_buff as a context type.

Reported-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230330190115.3942962-1-andrii@kernel.org
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index 055df1abd7ca..7888c03ba631 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -798,7 +798,7 @@ static int guess_prog_type_by_ctx_name(const char *ctx_name,
 		enum bpf_attach_type attach_type;
 	} ctx_map[] = {
 		/* __sk_buff is most ambiguous, for now we assume cgroup_skb */
-		{ "__sk_buff", "sk_buff", BPF_PROG_TYPE_CGROUP_SKB, BPF_CGROUP_INET_INGRESS },
+		{ "__sk_buff", "sk_buff", BPF_PROG_TYPE_SCHED_CLS },
 		{ "bpf_sock", "sock", BPF_PROG_TYPE_CGROUP_SOCK, BPF_CGROUP_INET4_POST_BIND },
 		{ "bpf_sock_addr", "bpf_sock_addr_kern",  BPF_PROG_TYPE_CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_BIND },
 		{ "bpf_sock_ops", "bpf_sock_ops_kern", BPF_PROG_TYPE_SOCK_OPS, BPF_CGROUP_SOCK_OPS },
-- 
cgit v1.2.3-70-g09d2


From e4c2acab95a5947fe7948140a83e4f4918c6c048 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 30 Mar 2023 09:52:02 -0500
Subject: bpf: Handle PTR_MAYBE_NULL case in PTR_TO_BTF_ID helper call arg

When validating a helper function argument, we use check_reg_type() to
ensure that the register containing the argument is of the correct type.
When the register's base type is PTR_TO_BTF_ID, there is some
supplemental logic where we do extra checks for various combinations of
PTR_TO_BTF_ID type modifiers. For example, for PTR_TO_BTF_ID,
PTR_TO_BTF_ID | PTR_TRUSTED, and PTR_TO_BTF_ID | MEM_RCU, we call
map_kptr_match_type() for bpf_kptr_xchg() calls, and
btf_struct_ids_match() for other helper calls.

When an unhandled PTR_TO_BTF_ID type modifier combination is passed to
check_reg_type(), the verifier fails with an internal verifier error
message. This can currently be triggered by passing a PTR_MAYBE_NULL
pointer to helper functions (currently just bpf_kptr_xchg()) with an
ARG_PTR_TO_BTF_ID_OR_NULL arg type. For example, by callin
bpf_kptr_xchg(&v->kptr, bpf_cpumask_create()).

Whether or not passing a PTR_MAYBE_NULL arg to an
ARG_PTR_TO_BTF_ID_OR_NULL argument is valid is an interesting question.
In a vacuum, it seems fine. A helper function with an
ARG_PTR_TO_BTF_ID_OR_NULL arg would seem to be implying that it can
handle either a NULL or non-NULL arg, and has logic in place to detect
and gracefully handle each. This is the case for bpf_kptr_xchg(), which
of course simply does an xchg(). On the other hand, bpf_kptr_xchg() also
specifies OBJ_RELEASE, and refcounting semantics for a PTR_MAYBE_NULL
pointer is different than handling it for a NULL _OR_ non-NULL pointer.
For example, with a non-NULL arg, we should always fail if there was not
a nonzero refcount for the value in the register being passed to the
helper. For PTR_MAYBE_NULL on the other hand, it's unclear. If the
pointer is NULL it would be fine, but if it's not NULL, it would be
incorrect to load the program.

The current solution to this is to just fail if PTR_MAYBE_NULL is
passed, and to instead require programs to have a NULL check to
explicitly handle the NULL and non-NULL cases. This seems reasonable.
Not only would it possibly be quite complicated to correctly handle
PTR_MAYBE_NULL refcounting in the verifier, but it's also an arguably
odd programming pattern in general to not explicitly handle the NULL
case anyways. For example, it seems odd to not care about whether a
pointer you're passing to bpf_kptr_xchg() was successfully allocated in
a program such as the following:

private(MASK) static struct bpf_cpumask __kptr * global_mask;

SEC("tp_btf/task_newtask")
int BPF_PROG(example, struct task_struct *task, u64 clone_flags)
{
        struct bpf_cpumask *prev;

	/* bpf_cpumask_create() returns PTR_MAYBE_NULL */
	prev = bpf_kptr_xchg(&global_mask, bpf_cpumask_create());
	if (prev)
		bpf_cpumask_release(prev);

	return 0;
}

This patch therefore updates the verifier to explicitly check for
PTR_MAYBE_NULL in check_reg_type(), and fail gracefully if it's
observed. This isn't really "fixing" anything unsafe or incorrect. We're
just updating the verifier to fail gracefully, and explicitly handle
this pattern rather than unintentionally falling back to an internal
verifier error path. A subsequent patch will update selftests.

Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230330145203.80506-1-void@manifault.com
---
 kernel/bpf/verifier.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 20eb2015842f..52738f9dcb15 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -7204,6 +7204,10 @@ found:
 		}
 		break;
 	}
+	case PTR_TO_BTF_ID | PTR_MAYBE_NULL:
+	case PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_RCU:
+		verbose(env, "Possibly NULL pointer passed to helper arg%d\n", regno);
+		return -EACCES;
 	case PTR_TO_BTF_ID | MEM_ALLOC:
 		if (meta->func_id != BPF_FUNC_spin_lock && meta->func_id != BPF_FUNC_spin_unlock &&
 		    meta->func_id != BPF_FUNC_kptr_xchg) {
-- 
cgit v1.2.3-70-g09d2


From 67efbd57bc6e57de276b964f023f8f947bc52460 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Thu, 30 Mar 2023 09:52:03 -0500
Subject: selftests/bpf: Add testcases for ptr_*_or_null_ in bpf_kptr_xchg

The second argument of the bpf_kptr_xchg() helper function is
ARG_PTR_TO_BTF_ID_OR_NULL. A recent patch fixed a bug whereby the
verifier would fail with an internal error message if a program invoked
the helper with a PTR_TO_BTF_ID | PTR_MAYBE_NULL register. This testcase
adds some testcases to ensure that it fails gracefully moving forward.

Before the fix, these testcases would have failed an error resembling
the following:

; p = bpf_kfunc_call_test_acquire(&(unsigned long){0});
99: (7b) *(u64 *)(r10 -16) = r7       ; frame1: ...
100: (bf) r1 = r10                    ; frame1: ...
101: (07) r1 += -16                   ; frame1: ...
; p = bpf_kfunc_call_test_acquire(&(unsigned long){0});
102: (85) call bpf_kfunc_call_test_acquire#13908
; frame1: R0_w=ptr_or_null_prog_test_ref_kfunc...
; p = bpf_kptr_xchg(&v->ref_ptr, p);
103: (bf) r1 = r6                     ; frame1: ...
104: (bf) r2 = r0
; frame1: R0_w=ptr_or_null_prog_test_ref_kfunc...
105: (85) call bpf_kptr_xchg#194
verifier internal error: invalid PTR_TO_BTF_ID register for type match

Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230330145203.80506-2-void@manifault.com
---
 .../testing/selftests/bpf/progs/cpumask_failure.c  | 25 ++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/map_kptr_fail.c  | 23 ++++++++++++++++++++
 2 files changed, 48 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/cpumask_failure.c b/tools/testing/selftests/bpf/progs/cpumask_failure.c
index db4f94e72b61..a9bf6ea336cf 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_failure.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_failure.c
@@ -165,3 +165,28 @@ int BPF_PROG(test_global_mask_no_null_check, struct task_struct *task, u64 clone
 
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+__failure __msg("Possibly NULL pointer passed to helper arg2")
+int BPF_PROG(test_global_mask_rcu_no_null_check, struct task_struct *task, u64 clone_flags)
+{
+	struct bpf_cpumask *prev, *curr;
+
+	curr = bpf_cpumask_create();
+	if (!curr)
+		return 0;
+
+	prev = bpf_kptr_xchg(&global_mask, curr);
+	if (prev)
+		bpf_cpumask_release(prev);
+
+	bpf_rcu_read_lock();
+	curr = global_mask;
+	/* PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_RCU passed to bpf_kptr_xchg() */
+	prev = bpf_kptr_xchg(&global_mask, curr);
+	bpf_rcu_read_unlock();
+	if (prev)
+		bpf_cpumask_release(prev);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/map_kptr_fail.c b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
index 08f9ec18c345..15bf3127dba3 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
@@ -20,6 +20,7 @@ struct array_map {
 } array_map SEC(".maps");
 
 extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
+extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
 extern struct prog_test_ref_kfunc *
 bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **p, int a, int b) __ksym;
 
@@ -442,4 +443,26 @@ int kptr_get_ref_state(struct __sk_buff *ctx)
 	return 0;
 }
 
+SEC("?tc")
+__failure __msg("Possibly NULL pointer passed to helper arg2")
+int kptr_xchg_possibly_null(struct __sk_buff *ctx)
+{
+	struct prog_test_ref_kfunc *p;
+	struct map_value *v;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&array_map, &key);
+	if (!v)
+		return 0;
+
+	p = bpf_kfunc_call_test_acquire(&(unsigned long){0});
+
+	/* PTR_TO_BTF_ID | PTR_MAYBE_NULL passed to bpf_kptr_xchg() */
+	p = bpf_kptr_xchg(&v->ref_ptr, p);
+	if (p)
+		bpf_kfunc_call_test_release(p);
+
+	return 0;
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 9a865a98a3360d06f92549839abda1bc96104aea Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Wed, 29 Mar 2023 08:59:29 -0400
Subject: net: ksz884x: remove unused change variable

clang with W=1 reports
drivers/net/ethernet/micrel/ksz884x.c:3216:6: error: variable
  'change' set but not used [-Werror,-Wunused-but-set-variable]
        int change = 0;
            ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230329125929.1808420-1-trix@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/micrel/ksz884x.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c
index e6acd1e7b263..f78e8ead8c36 100644
--- a/drivers/net/ethernet/micrel/ksz884x.c
+++ b/drivers/net/ethernet/micrel/ksz884x.c
@@ -3213,7 +3213,6 @@ static void port_get_link_speed(struct ksz_port *port)
 	u8 remote;
 	int i;
 	int p;
-	int change = 0;
 
 	interrupt = hw_block_intr(hw);
 
@@ -3260,17 +3259,14 @@ static void port_get_link_speed(struct ksz_port *port)
 					port_cfg_back_pressure(hw, p,
 						(1 == info->duplex));
 				}
-				change |= 1 << i;
 				port_cfg_change(hw, port, info, status);
 			}
 			info->state = media_connected;
 		} else {
-			if (media_disconnected != info->state) {
-				change |= 1 << i;
-
-				/* Indicate the link just goes down. */
+			/* Indicate the link just goes down. */
+			if (media_disconnected != info->state)
 				hw->port_mib[p].link_down = 1;
-			}
+
 			info->state = media_disconnected;
 		}
 		hw->port_mib[p].state = (u8) info->state;
-- 
cgit v1.2.3-70-g09d2


From eb1ab7650d358b553cc946035fd7c7bdda1856e3 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 29 Mar 2023 16:38:19 +0300
Subject: net: dsa: fix db type confusion in host fdb/mdb add/del

We have the following code paths:

Host FDB (unicast RX filtering):

dsa_port_standalone_host_fdb_add()   dsa_port_bridge_host_fdb_add()
               |                                     |
               +--------------+         +------------+
                              |         |
                              v         v
                         dsa_port_host_fdb_add()

dsa_port_standalone_host_fdb_del()   dsa_port_bridge_host_fdb_del()
               |                                     |
               +--------------+         +------------+
                              |         |
                              v         v
                         dsa_port_host_fdb_del()

Host MDB (multicast RX filtering):

dsa_port_standalone_host_mdb_add()   dsa_port_bridge_host_mdb_add()
               |                                     |
               +--------------+         +------------+
                              |         |
                              v         v
                         dsa_port_host_mdb_add()

dsa_port_standalone_host_mdb_del()   dsa_port_bridge_host_mdb_del()
               |                                     |
               +--------------+         +------------+
                              |         |
                              v         v
                         dsa_port_host_mdb_del()

The logic added by commit 5e8a1e03aa4d ("net: dsa: install secondary
unicast and multicast addresses as host FDB/MDB") zeroes out
db.bridge.num if the switch doesn't support ds->fdb_isolation
(the majority doesn't). This is done for a reason explained in commit
c26933639b54 ("net: dsa: request drivers to perform FDB isolation").

Taking a single code path as example - dsa_port_host_fdb_add() - the
others are similar - the problem is that this function handles:
- DSA_DB_PORT databases, when called from
  dsa_port_standalone_host_fdb_add()
- DSA_DB_BRIDGE databases, when called from
  dsa_port_bridge_host_fdb_add()

So, if dsa_port_host_fdb_add() were to make any change on the
"bridge.num" attribute of the database, this would only be correct for a
DSA_DB_BRIDGE, and a type confusion for a DSA_DB_PORT bridge.

However, this bug is without consequences, for 2 reasons:

- dsa_port_standalone_host_fdb_add() is only called from code which is
  (in)directly guarded by dsa_switch_supports_uc_filtering(ds), and that
  function only returns true if ds->fdb_isolation is set. So, the code
  only executed for DSA_DB_BRIDGE databases.

- Even if the code was not dead for DSA_DB_PORT, we have the following
  memory layout:

struct dsa_bridge {
	struct net_device *dev;
	unsigned int num;
	bool tx_fwd_offload;
	refcount_t refcount;
};

struct dsa_db {
	enum dsa_db_type type;

	union {
		const struct dsa_port *dp; // DSA_DB_PORT
		struct dsa_lag lag;
		struct dsa_bridge bridge; // DSA_DB_BRIDGE
	};
};

So, the zeroization of dsa_db :: bridge :: num on a dsa_db structure of
type DSA_DB_PORT would access memory which is unused, because we only
use dsa_db :: dp for DSA_DB_PORT, and this is mapped at the same address
with dsa_db :: dev for DSA_DB_BRIDGE, thanks to the union definition.

It is correct to fix up dsa_db :: bridge :: num only from code paths
that come from the bridge / switchdev, so move these there.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230329133819.697642-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/dsa/port.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/net/dsa/port.c b/net/dsa/port.c
index 67ad1adec2a2..15cee17769e9 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -1028,9 +1028,6 @@ static int dsa_port_host_fdb_add(struct dsa_port *dp,
 		.db = db,
 	};
 
-	if (!dp->ds->fdb_isolation)
-		info.db.bridge.num = 0;
-
 	return dsa_port_notify(dp, DSA_NOTIFIER_HOST_FDB_ADD, &info);
 }
 
@@ -1055,6 +1052,9 @@ int dsa_port_bridge_host_fdb_add(struct dsa_port *dp,
 	};
 	int err;
 
+	if (!dp->ds->fdb_isolation)
+		db.bridge.num = 0;
+
 	/* Avoid a call to __dev_set_promiscuity() on the master, which
 	 * requires rtnl_lock(), since we can't guarantee that is held here,
 	 * and we can't take it either.
@@ -1079,9 +1079,6 @@ static int dsa_port_host_fdb_del(struct dsa_port *dp,
 		.db = db,
 	};
 
-	if (!dp->ds->fdb_isolation)
-		info.db.bridge.num = 0;
-
 	return dsa_port_notify(dp, DSA_NOTIFIER_HOST_FDB_DEL, &info);
 }
 
@@ -1106,6 +1103,9 @@ int dsa_port_bridge_host_fdb_del(struct dsa_port *dp,
 	};
 	int err;
 
+	if (!dp->ds->fdb_isolation)
+		db.bridge.num = 0;
+
 	if (master->priv_flags & IFF_UNICAST_FLT) {
 		err = dev_uc_del(master, addr);
 		if (err)
@@ -1210,9 +1210,6 @@ static int dsa_port_host_mdb_add(const struct dsa_port *dp,
 		.db = db,
 	};
 
-	if (!dp->ds->fdb_isolation)
-		info.db.bridge.num = 0;
-
 	return dsa_port_notify(dp, DSA_NOTIFIER_HOST_MDB_ADD, &info);
 }
 
@@ -1237,6 +1234,9 @@ int dsa_port_bridge_host_mdb_add(const struct dsa_port *dp,
 	};
 	int err;
 
+	if (!dp->ds->fdb_isolation)
+		db.bridge.num = 0;
+
 	err = dev_mc_add(master, mdb->addr);
 	if (err)
 		return err;
@@ -1254,9 +1254,6 @@ static int dsa_port_host_mdb_del(const struct dsa_port *dp,
 		.db = db,
 	};
 
-	if (!dp->ds->fdb_isolation)
-		info.db.bridge.num = 0;
-
 	return dsa_port_notify(dp, DSA_NOTIFIER_HOST_MDB_DEL, &info);
 }
 
@@ -1281,6 +1278,9 @@ int dsa_port_bridge_host_mdb_del(const struct dsa_port *dp,
 	};
 	int err;
 
+	if (!dp->ds->fdb_isolation)
+		db.bridge.num = 0;
+
 	err = dev_mc_del(master, mdb->addr);
 	if (err)
 		return err;
-- 
cgit v1.2.3-70-g09d2


From 3292004c90c8aba74600a5cd8d10f8186fd48269 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Wed, 29 Mar 2023 08:08:01 -0700
Subject: net: ethernet: ti: Fix format specifier in netcp_create_interface()

After commit 3948b05950fd ("net: introduce a config option to tweak
MAX_SKB_FRAGS"), clang warns:

  drivers/net/ethernet/ti/netcp_core.c:2085:4: warning: format specifies type 'long' but the argument has type 'int' [-Wformat]
                          MAX_SKB_FRAGS);
                          ^~~~~~~~~~~~~
  include/linux/dev_printk.h:144:65: note: expanded from macro 'dev_err'
          dev_printk_index_wrap(_dev_err, KERN_ERR, dev, dev_fmt(fmt), ##__VA_ARGS__)
                                                                 ~~~     ^~~~~~~~~~~
  include/linux/dev_printk.h:110:23: note: expanded from macro 'dev_printk_index_wrap'
                  _p_func(dev, fmt, ##__VA_ARGS__);                       \
                               ~~~    ^~~~~~~~~~~
  include/linux/skbuff.h:352:23: note: expanded from macro 'MAX_SKB_FRAGS'
  #define MAX_SKB_FRAGS CONFIG_MAX_SKB_FRAGS
                        ^~~~~~~~~~~~~~~~~~~~
  ./include/generated/autoconf.h:11789:30: note: expanded from macro 'CONFIG_MAX_SKB_FRAGS'
  #define CONFIG_MAX_SKB_FRAGS 17
                               ^~
  1 warning generated.

Follow the pattern of the rest of the tree by changing the specifier to
'%u' and casting MAX_SKB_FRAGS explicitly to 'unsigned int', which
eliminates the warning.

Fixes: 3948b05950fd ("net: introduce a config option to tweak MAX_SKB_FRAGS")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Link: https://lore.kernel.org/r/20230329-net-ethernet-ti-wformat-v1-1-83d0f799b553@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/netcp_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index 1bb596a9d8a2..d829113c16ee 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -2081,8 +2081,8 @@ static int netcp_create_interface(struct netcp_device *netcp_device,
 	netcp->tx_pool_region_id = temp[1];
 
 	if (netcp->tx_pool_size < MAX_SKB_FRAGS) {
-		dev_err(dev, "tx-pool size too small, must be at least %ld\n",
-			MAX_SKB_FRAGS);
+		dev_err(dev, "tx-pool size too small, must be at least %u\n",
+			(unsigned int)MAX_SKB_FRAGS);
 		ret = -ENODEV;
 		goto quit;
 	}
-- 
cgit v1.2.3-70-g09d2


From 46e9acb7ae2a36643b8c43f5107c072cbbc48572 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Wed, 29 Mar 2023 17:24:53 +0200
Subject: selftests: rtnetlink: Fix do_test_address_proto()

This selftest was introduced recently in the commit cited below. It misses
several check_err() invocations to actually verify that the previous
command succeeded. When these are added, the first one fails, because
besides the addresses added by hand, there can be a link-local address
added by the kernel. Adjust the check to expect at least three addresses
instead of exactly three, and add the missing check_err's.

Furthermore, the explanatory comments assume that the address with no
protocol is $addr2, when in fact it is $addr3. Update the comments.

Fixes: 6a414fd77f61 ("selftests: rtnetlink: Add an address proto test")
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://lore.kernel.org/r/53a579bc883e1bf2fe490d58427cf22c2d1aa21f.1680102695.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/rtnetlink.sh | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh
index 3b15c686c03f..383ac6fc037d 100755
--- a/tools/testing/selftests/net/rtnetlink.sh
+++ b/tools/testing/selftests/net/rtnetlink.sh
@@ -1302,19 +1302,23 @@ do_test_address_proto()
 
 	count=$(address_count)
 	check_err $?
-	(( count == 3 )) # $addr, $addr2 and $addr3
+	(( count >= 3 )) # $addr, $addr2 and $addr3 plus any kernel addresses
+	check_err $?
 
 	count=$(address_count proto 0)
 	check_err $?
-	(( count == 1 )) # just $addr2
+	(( count == 1 )) # just $addr3
+	check_err $?
 
 	count=$(address_count proto 0x11)
 	check_err $?
-	(( count == 2 )) # $addr and $addr2
+	(( count == 2 )) # $addr and $addr3
+	check_err $?
 
 	count=$(address_count proto 0xab)
 	check_err $?
-	(( count == 1 )) # just $addr2
+	(( count == 1 )) # just $addr3
+	check_err $?
 
 	ip address del dev "$devdummy" "$addr"
 	ip address del dev "$devdummy" "$addr2"
-- 
cgit v1.2.3-70-g09d2


From 2384127e98db52a6ac2577924ad9cae25f3e7472 Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Wed, 29 Mar 2023 11:54:52 +0200
Subject: net/sched: act_tunnel_key: add support for "don't fragment"

extend "act_tunnel_key" to allow specifying TUNNEL_DONT_FRAGMENT.

Suggested-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Pedro Tammela <pctammela@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/uapi/linux/tc_act/tc_tunnel_key.h | 1 +
 net/sched/act_tunnel_key.c                | 5 +++++
 2 files changed, 6 insertions(+)

diff --git a/include/uapi/linux/tc_act/tc_tunnel_key.h b/include/uapi/linux/tc_act/tc_tunnel_key.h
index 49ad4033951b..37c6f612f161 100644
--- a/include/uapi/linux/tc_act/tc_tunnel_key.h
+++ b/include/uapi/linux/tc_act/tc_tunnel_key.h
@@ -34,6 +34,7 @@ enum {
 					 */
 	TCA_TUNNEL_KEY_ENC_TOS,		/* u8 */
 	TCA_TUNNEL_KEY_ENC_TTL,		/* u8 */
+	TCA_TUNNEL_KEY_NO_FRAG,		/* flag */
 	__TCA_TUNNEL_KEY_MAX,
 };
 
diff --git a/net/sched/act_tunnel_key.c b/net/sched/act_tunnel_key.c
index 2d12d2626415..0c8aa7e686ea 100644
--- a/net/sched/act_tunnel_key.c
+++ b/net/sched/act_tunnel_key.c
@@ -420,6 +420,9 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla,
 		    nla_get_u8(tb[TCA_TUNNEL_KEY_NO_CSUM]))
 			flags &= ~TUNNEL_CSUM;
 
+		if (nla_get_flag(tb[TCA_TUNNEL_KEY_NO_FRAG]))
+			flags |= TUNNEL_DONT_FRAGMENT;
+
 		if (tb[TCA_TUNNEL_KEY_ENC_DST_PORT])
 			dst_port = nla_get_be16(tb[TCA_TUNNEL_KEY_ENC_DST_PORT]);
 
@@ -747,6 +750,8 @@ static int tunnel_key_dump(struct sk_buff *skb, struct tc_action *a,
 				   key->tp_dst)) ||
 		    nla_put_u8(skb, TCA_TUNNEL_KEY_NO_CSUM,
 			       !(key->tun_flags & TUNNEL_CSUM)) ||
+		    ((key->tun_flags & TUNNEL_DONT_FRAGMENT) &&
+		     nla_put_flag(skb, TCA_TUNNEL_KEY_NO_FRAG)) ||
 		    tunnel_key_opts_dump(skb, info))
 			goto nla_put_failure;
 
-- 
cgit v1.2.3-70-g09d2


From 7f3f86402609315d6b98f6c76713e07b5a9562aa Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Wed, 29 Mar 2023 11:54:53 +0200
Subject: selftests: tc-testing: add "depends_on" property to skip tests

currently, users can skip individual test cases by means of writing

  "skip": "yes"

in the scenario file. Extend this functionality, introducing 'dependsOn':
it's optional property like "skip", but the value contains a command (for
example, a probe on iproute2 to check if it supports a specific feature).
If such property is present, tdc executes that command and skips the test
when the return value is non-zero.

Reviewed-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../tc-testing/creating-testcases/AddingTestCases.txt       |  2 ++
 tools/testing/selftests/tc-testing/tdc.py                   | 13 +++++++++++++
 2 files changed, 15 insertions(+)

diff --git a/tools/testing/selftests/tc-testing/creating-testcases/AddingTestCases.txt b/tools/testing/selftests/tc-testing/creating-testcases/AddingTestCases.txt
index a28571aff0e1..ff956d8c99c5 100644
--- a/tools/testing/selftests/tc-testing/creating-testcases/AddingTestCases.txt
+++ b/tools/testing/selftests/tc-testing/creating-testcases/AddingTestCases.txt
@@ -38,6 +38,8 @@ skip:         A completely optional key, if the corresponding value is "yes"
               this test case will still appear in the results output but
               marked as skipped. This key can be placed anywhere inside the
               test case at the top level.
+dependsOn:    Same as 'skip', but the value is executed as a command. The test
+              is skipped when the command returns non-zero.
 category:     A list of single-word descriptions covering what the command
               under test is testing. Example: filter, actions, u32, gact, etc.
 setup:        The list of commands required to ensure the command under test
diff --git a/tools/testing/selftests/tc-testing/tdc.py b/tools/testing/selftests/tc-testing/tdc.py
index 7bd94f8e490a..b98256f38447 100755
--- a/tools/testing/selftests/tc-testing/tdc.py
+++ b/tools/testing/selftests/tc-testing/tdc.py
@@ -369,6 +369,19 @@ def run_one_test(pm, args, index, tidx):
             pm.call_post_execute()
             return res
 
+    if 'dependsOn' in tidx:
+        if (args.verbose > 0):
+            print('probe command for test skip')
+        (p, procout) = exec_cmd(args, pm, 'execute', tidx['dependsOn'])
+        if p:
+            if (p.returncode != 0):
+                res = TestResult(tidx['id'], tidx['name'])
+                res.set_result(ResultState.skip)
+                res.set_errormsg('probe command: test skipped.')
+                pm.call_pre_case(tidx, test_skip=True)
+                pm.call_post_execute()
+                return res
+
     # populate NAMES with TESTID for this test
     NAMES['TESTID'] = tidx['id']
 
-- 
cgit v1.2.3-70-g09d2


From b8617f8eed84e12c26f2220733280187466a0615 Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Wed, 29 Mar 2023 11:54:54 +0200
Subject: selftests: tc-testing: add tunnel_key "nofrag" test case

# ./tdc.py -e 6bda -l
 6bda: (actions, tunnel_key) Add tunnel_key action with nofrag option

Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../tc-testing/tc-tests/actions/tunnel_key.json    | 25 ++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/tools/testing/selftests/tc-testing/tc-tests/actions/tunnel_key.json b/tools/testing/selftests/tc-testing/tc-tests/actions/tunnel_key.json
index b40ee602918a..b5b47fbf6c00 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/actions/tunnel_key.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/actions/tunnel_key.json
@@ -983,5 +983,30 @@
         "teardown": [
             "$TC actions flush action tunnel_key"
         ]
+    },
+    {
+        "id": "6bda",
+        "name": "Add tunnel_key action with nofrag option",
+        "category": [
+            "actions",
+            "tunnel_key"
+        ],
+        "dependsOn": "$TC actions add action tunnel_key help 2>&1 | grep -q nofrag",
+        "setup": [
+            [
+                "$TC action flush action tunnel_key",
+                0,
+                1,
+                255
+            ]
+        ],
+        "cmdUnderTest": "$TC actions add action tunnel_key set src_ip 10.10.10.1 dst_ip 10.10.10.2 id 1111 nofrag index 222",
+        "expExitCode": "0",
+        "verifyCmd": "$TC actions get action tunnel_key index 222",
+        "matchPattern": "action order [0-9]+: tunnel_key.*src_ip 10.10.10.1.*dst_ip 10.10.10.2.*key_id 1111.*csum.*nofrag pipe.*index 222",
+        "matchCount": "1",
+        "teardown": [
+            "$TC actions flush action tunnel_key"
+        ]
     }
 ]
-- 
cgit v1.2.3-70-g09d2


From 533a89b1940f527c6819f046e2aefb60df8a13d3 Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Wed, 29 Mar 2023 11:54:55 +0200
Subject: selftests: forwarding: add tunnel_key "nofrag" test case

Add a selftest that configures metadata tunnel encapsulation using the TC
"tunnel_key" action: it includes a test case for setting "nofrag" flag.

Example output:

 # selftests: net/forwarding: tc_tunnel_key.sh
 # TEST: tunnel_key nofrag (skip_hw)                                   [ OK ]
 # INFO: Could not test offloaded functionality
 ok 1 selftests: net/forwarding: tc_tunnel_key.sh

Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/Makefile    |   1 +
 .../selftests/net/forwarding/tc_tunnel_key.sh      | 161 +++++++++++++++++++++
 2 files changed, 162 insertions(+)
 create mode 100755 tools/testing/selftests/net/forwarding/tc_tunnel_key.sh

diff --git a/tools/testing/selftests/net/forwarding/Makefile b/tools/testing/selftests/net/forwarding/Makefile
index 91201ab3c4fc..236f6b796a52 100644
--- a/tools/testing/selftests/net/forwarding/Makefile
+++ b/tools/testing/selftests/net/forwarding/Makefile
@@ -85,6 +85,7 @@ TEST_PROGS = bridge_igmp.sh \
 	tc_mpls_l2vpn.sh \
 	tc_police.sh \
 	tc_shblocks.sh \
+	tc_tunnel_key.sh \
 	tc_vlan_modify.sh \
 	vxlan_asymmetric_ipv6.sh \
 	vxlan_asymmetric.sh \
diff --git a/tools/testing/selftests/net/forwarding/tc_tunnel_key.sh b/tools/testing/selftests/net/forwarding/tc_tunnel_key.sh
new file mode 100755
index 000000000000..5ac184d51809
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/tc_tunnel_key.sh
@@ -0,0 +1,161 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+ALL_TESTS="tunnel_key_nofrag_test"
+
+NUM_NETIFS=4
+source tc_common.sh
+source lib.sh
+
+tcflags="skip_hw"
+
+h1_create()
+{
+	simple_if_init $h1 192.0.2.1/24
+	forwarding_enable
+	mtu_set $h1 1500
+	tunnel_create h1-et vxlan 192.0.2.1 192.0.2.2 dev $h1 dstport 0 external
+	tc qdisc add dev h1-et clsact
+	mtu_set h1-et 1230
+	mtu_restore $h1
+	mtu_set $h1 1000
+}
+
+h1_destroy()
+{
+	tc qdisc del dev h1-et clsact
+	tunnel_destroy h1-et
+	forwarding_restore
+	mtu_restore $h1
+	simple_if_fini $h1 192.0.2.1/24
+}
+
+h2_create()
+{
+	simple_if_init $h2 192.0.2.2/24
+}
+
+h2_destroy()
+{
+	simple_if_fini $h2 192.0.2.2/24
+}
+
+switch_create()
+{
+	simple_if_init $swp1 192.0.2.2/24
+	tc qdisc add dev $swp1 clsact
+	simple_if_init $swp2 192.0.2.1/24
+}
+
+switch_destroy()
+{
+	simple_if_fini $swp2 192.0.2.1/24
+	tc qdisc del dev $swp1 clsact
+	simple_if_fini $swp1 192.0.2.2/24
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	h1mac=$(mac_get $h1)
+	h2mac=$(mac_get $h2)
+
+	swp1origmac=$(mac_get $swp1)
+	swp2origmac=$(mac_get $swp2)
+	ip link set $swp1 address $h2mac
+	ip link set $swp2 address $h1mac
+
+	vrf_prepare
+
+	h1_create
+	h2_create
+	switch_create
+
+	if ! tc action add action tunnel_key help 2>&1 | grep -q nofrag; then
+		log_test "SKIP: iproute doesn't support nofrag"
+		exit $ksft_skip
+	fi
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	switch_destroy
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+
+	ip link set $swp2 address $swp2origmac
+	ip link set $swp1 address $swp1origmac
+}
+
+tunnel_key_nofrag_test()
+{
+	RET=0
+	local i
+
+	tc filter add dev $swp1 ingress protocol ip pref 100 handle 100 \
+		flower ip_flags nofrag action drop
+	tc filter add dev $swp1 ingress protocol ip pref 101 handle 101 \
+		flower ip_flags firstfrag action drop
+	tc filter add dev $swp1 ingress protocol ip pref 102 handle 102 \
+		flower ip_flags nofirstfrag action drop
+
+	# test 'nofrag' set
+	tc filter add dev h1-et egress protocol all pref 1 handle 1 matchall $tcflags \
+		action tunnel_key set src_ip 192.0.2.1 dst_ip 192.0.2.2 id 42 nofrag index 10
+	$MZ h1-et -c 1 -p 930 -a 00:aa:bb:cc:dd:ee -b 00:ee:dd:cc:bb:aa -t ip -q
+	tc_check_packets "dev $swp1 ingress" 100 1
+	check_err $? "packet smaller than MTU was not tunneled"
+
+	$MZ h1-et -c 1 -p 931 -a 00:aa:bb:cc:dd:ee -b 00:ee:dd:cc:bb:aa -t ip -q
+	tc_check_packets "dev $swp1 ingress" 100 1
+	check_err $? "packet bigger than MTU matched nofrag (nofrag was set)"
+	tc_check_packets "dev $swp1 ingress" 101 0
+	check_err $? "packet bigger than MTU matched firstfrag (nofrag was set)"
+	tc_check_packets "dev $swp1 ingress" 102 0
+	check_err $? "packet bigger than MTU matched nofirstfrag (nofrag was set)"
+
+	# test 'nofrag' cleared
+	tc actions change action tunnel_key set src_ip 192.0.2.1 dst_ip 192.0.2.2 id 42 index 10
+	$MZ h1-et -c 1 -p 931 -a 00:aa:bb:cc:dd:ee -b 00:ee:dd:cc:bb:aa -t ip -q
+	tc_check_packets "dev $swp1  ingress" 100 1
+	check_err $? "packet bigger than MTU matched nofrag (nofrag was unset)"
+	tc_check_packets "dev $swp1  ingress" 101 1
+	check_err $? "packet bigger than MTU didn't match firstfrag (nofrag was unset) "
+	tc_check_packets "dev $swp1 ingress" 102 1
+	check_err $? "packet bigger than MTU didn't match nofirstfrag (nofrag was unset) "
+
+	for i in 100 101 102; do
+		tc filter del dev $swp1 ingress protocol ip pref $i handle $i flower
+	done
+	tc filter del dev h1-et egress pref 1 handle 1 matchall
+
+	log_test "tunnel_key nofrag ($tcflags)"
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+tc_offload_check
+if [[ $? -ne 0 ]]; then
+	log_info "Could not test offloaded functionality"
+else
+	tcflags="skip_sw"
+	tests_run
+fi
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-70-g09d2


From 709d0b880ceae9503ce4282fcfaf4cb59c6dd267 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Wed, 29 Mar 2023 13:23:56 +0200
Subject: octeontx2-af: update type of prof fields in nix_aw_enq_req

Update type of prof and prof_mask fields in nix_as_enq_req
from u64 to struct nix_bandprof_s, which is 128 bits wide.

This is to address warnings with compiling with gcc-12 W=1
regarding string fortification.

Although the union of which these fields are a member is 128bits
wide, and thus writing a 128bit entity is safe, the compiler flags
a problem as the field being written is only 64 bits wide.

  CC [M]  drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.o
scripts/Makefile.build:252: ./drivers/net/ethernet/marvell/octeontx2/nic/Makefile: otx2_dcbnl.o is added to multiple modules: rvu_nicpf rvu_nicvf
  CC [M]  drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.o
  CC [M]  drivers/net/ethernet/marvell/octeontx2/nic/qos_sq.o
  CC [M]  drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.o
  CC [M]  drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.o
In file included from ./include/linux/string.h:254,
                 from ./include/linux/bitmap.h:11,
                 from ./include/linux/cpumask.h:12,
                 from ./arch/x86/include/asm/paravirt.h:17,
                 from ./arch/x86/include/asm/cpuid.h:62,
                 from ./arch/x86/include/asm/processor.h:19,
                 from ./arch/x86/include/asm/timex.h:5,
                 from ./include/linux/timex.h:67,
                 from ./include/linux/time32.h:13,
                 from ./include/linux/time.h:60,
                 from ./include/linux/stat.h:19,
                 from ./include/linux/module.h:13,
                 from drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c:8:
In function 'fortify_memcpy_chk',
    inlined from 'rvu_nix_blk_aq_enq_inst' at drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c:969:4:
./include/linux/fortify-string.h:529:25: error: call to '__read_overflow2_field' declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror=attribute-warning]
  529 |                         __read_overflow2_field(q_size_field, size);
      |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In function 'fortify_memcpy_chk',
    inlined from 'rvu_nix_blk_aq_enq_inst' at drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c:984:4:
./include/linux/fortify-string.h:529:25: error: call to '__read_overflow2_field' declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror=attribute-warning]
  529 |                         __read_overflow2_field(q_size_field, size);
      |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors

Compile tested only!

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Link: https://lore.kernel.org/r/20230329112356.458072-1-horms@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index 5727d67e0259..8fb5cae7285b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -936,7 +936,7 @@ struct nix_aq_enq_req {
 		struct nix_cq_ctx_s cq;
 		struct nix_rsse_s   rss;
 		struct nix_rx_mce_s mce;
-		u64 prof;
+		struct nix_bandprof_s prof;
 	};
 	union {
 		struct nix_rq_ctx_s rq_mask;
@@ -944,7 +944,7 @@ struct nix_aq_enq_req {
 		struct nix_cq_ctx_s cq_mask;
 		struct nix_rsse_s   rss_mask;
 		struct nix_rx_mce_s mce_mask;
-		u64 prof_mask;
+		struct nix_bandprof_s prof_mask;
 	};
 };
 
-- 
cgit v1.2.3-70-g09d2


From 9f7cc57fe5508c495d3a75efd7f942aeec0013e0 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Wed, 29 Mar 2023 15:16:52 -0700
Subject: tools: ynl: support byte-order in cli

Used by ethtool spec.

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/nlspec.py |  1 +
 tools/net/ynl/lib/ynl.py    | 35 +++++++++++++++++++++++------------
 2 files changed, 24 insertions(+), 12 deletions(-)

diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index 06a906d74f0e..0c43022ff7e6 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -163,6 +163,7 @@ class SpecAttr(SpecElement):
         self.is_multi = yaml.get('multi-attr', False)
         self.struct_name = yaml.get('struct')
         self.sub_type = yaml.get('sub-type')
+        self.byte_order = yaml.get('byte-order')
 
 
 class SpecAttrSet(SpecElement):
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index ec40918152e1..8778994d40c0 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -80,17 +80,25 @@ class NlAttr:
         self.full_len = (self.payload_len + 3) & ~3
         self.raw = raw[offset + 4:offset + self.payload_len]
 
+    def format_byte_order(byte_order):
+        if byte_order:
+            return ">" if byte_order == "big-endian" else "<"
+        return ""
+
     def as_u8(self):
         return struct.unpack("B", self.raw)[0]
 
-    def as_u16(self):
-        return struct.unpack("H", self.raw)[0]
+    def as_u16(self, byte_order=None):
+        endian = NlAttr.format_byte_order(byte_order)
+        return struct.unpack(f"{endian}H", self.raw)[0]
 
-    def as_u32(self):
-        return struct.unpack("I", self.raw)[0]
+    def as_u32(self, byte_order=None):
+        endian = NlAttr.format_byte_order(byte_order)
+        return struct.unpack(f"{endian}I", self.raw)[0]
 
-    def as_u64(self):
-        return struct.unpack("Q", self.raw)[0]
+    def as_u64(self, byte_order=None):
+        endian = NlAttr.format_byte_order(byte_order)
+        return struct.unpack(f"{endian}Q", self.raw)[0]
 
     def as_strz(self):
         return self.raw.decode('ascii')[:-1]
@@ -365,11 +373,14 @@ class YnlFamily(SpecFamily):
         elif attr["type"] == 'u8':
             attr_payload = struct.pack("B", int(value))
         elif attr["type"] == 'u16':
-            attr_payload = struct.pack("H", int(value))
+            endian = NlAttr.format_byte_order(attr.byte_order)
+            attr_payload = struct.pack(f"{endian}H", int(value))
         elif attr["type"] == 'u32':
-            attr_payload = struct.pack("I", int(value))
+            endian = NlAttr.format_byte_order(attr.byte_order)
+            attr_payload = struct.pack(f"{endian}I", int(value))
         elif attr["type"] == 'u64':
-            attr_payload = struct.pack("Q", int(value))
+            endian = NlAttr.format_byte_order(attr.byte_order)
+            attr_payload = struct.pack(f"{endian}Q", int(value))
         elif attr["type"] == 'string':
             attr_payload = str(value).encode('ascii') + b'\x00'
         elif attr["type"] == 'binary':
@@ -415,11 +426,11 @@ class YnlFamily(SpecFamily):
             elif attr_spec['type'] == 'u8':
                 decoded = attr.as_u8()
             elif attr_spec['type'] == 'u16':
-                decoded = attr.as_u16()
+                decoded = attr.as_u16(attr_spec.byte_order)
             elif attr_spec['type'] == 'u32':
-                decoded = attr.as_u32()
+                decoded = attr.as_u32(attr_spec.byte_order)
             elif attr_spec['type'] == 'u64':
-                decoded = attr.as_u64()
+                decoded = attr.as_u64(attr_spec.byte_order)
             elif attr_spec["type"] == 'string':
                 decoded = attr.as_strz()
             elif attr_spec["type"] == 'binary':
-- 
cgit v1.2.3-70-g09d2


From a353318ebf24100bcc7254d293cd9f041f4075dd Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Wed, 29 Mar 2023 15:16:53 -0700
Subject: tools: ynl: populate most of the ethtool spec

Things that are not implemented:
- cable tests
- bitmaks in the requests don't work (needs multi-attr support in ynl.py)
- stats-get seems to return nonsense (not passing a bitmask properly?)
- notifications are not tested

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ethtool.yaml | 1480 +++++++++++++++++++++++++++---
 1 file changed, 1367 insertions(+), 113 deletions(-)

diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 6d8ae3d9a680..129f413ea349 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -6,6 +6,12 @@ protocol: genetlink-legacy
 
 doc: Partial family for Ethtool Netlink.
 
+definitions:
+  -
+    name: udp-tunnel-type
+    type: enum
+    entries: [ vxlan, geneve, vxlan-gpe ]
+
 attribute-sets:
   -
     name: header
@@ -38,6 +44,7 @@ attribute-sets:
       -
         name: bit
         type: nest
+        multi-attr: true
         nested-attributes: bitset-bit
   -
     name: bitset
@@ -53,6 +60,22 @@ attribute-sets:
         type: nest
         nested-attributes: bitset-bits
 
+  -
+    name: u64-array
+    attributes:
+      -
+        name: u64
+        type: nest
+        multi-attr: true
+        nested-attributes: u64
+  -
+    name: s32-array
+    attributes:
+      -
+        name: s32
+        type: nest
+        multi-attr: true
+        nested-attributes: s32
   -
     name: string
     attributes:
@@ -234,118 +257,1351 @@ attribute-sets:
         name: stats
         type: nest
         nested-attributes: mm-stat
+  -
+    name: linkinfo
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: port
+        type: u8
+      -
+        name: phyaddr
+        type: u8
+      -
+        name: tp-mdix
+        type: u8
+      -
+        name: tp-mdix-ctrl
+        type: u8
+      -
+        name: transceiver
+        type: u8
+  -
+    name: linkmodes
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: autoneg
+        type: u8
+      -
+        name: ours
+        type: nest
+        nested-attributes: bitset
+      -
+        name: peer
+        type: nest
+        nested-attributes: bitset
+      -
+        name: speed
+        type: u32
+      -
+        name: duplex
+        type: u8
+      -
+        name: master-slave-cfg
+        type: u8
+      -
+        name: master-slave-state
+        type: u8
+      -
+        name: master-slave-lanes
+        type: u32
+      -
+        name: rate-matching
+        type: u8
+  -
+    name: linkstate
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: link
+        type: u8
+      -
+        name: sqi
+        type: u32
+      -
+        name: sqi-max
+        type: u32
+      -
+        name: ext-state
+        type: u8
+      -
+        name: ext-substate
+        type: u8
+      -
+        name: down-cnt
+        type: u32
+  -
+    name: debug
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: msgmask
+        type: nest
+        nested-attributes: bitset
+  -
+    name: wol
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: modes
+        type: nest
+        nested-attributes: bitset
+      -
+        name: sopass
+        type: binary
+  -
+    name: features
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: hw
+        type: nest
+        nested-attributes: bitset
+      -
+        name: wanted
+        type: nest
+        nested-attributes: bitset
+      -
+        name: active
+        type: nest
+        nested-attributes: bitset
+      -
+        name: nochange
+        type: nest
+        nested-attributes: bitset
+  -
+    name: channels
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: rx-max
+        type: u32
+      -
+        name: tx-max
+        type: u32
+      -
+        name: other-max
+        type: u32
+      -
+        name: combined-max
+        type: u32
+      -
+        name: rx-count
+        type: u32
+      -
+        name: tx-count
+        type: u32
+      -
+        name: other-count
+        type: u32
+      -
+        name: combined-count
+        type: u32
 
-operations:
-  enum-model: directional
-  list:
-    -
-      name: strset-get
-      doc: Get string set from the kernel.
-
-      attribute-set: strset
-
-      do: &strset-get-op
-        request:
-          attributes:
-            - header
-            - stringsets
-            - counts-only
-        reply:
-          attributes:
-            - header
-            - stringsets
-      dump: *strset-get-op
-
-    # TODO: fill in the requests in between
-
-    -
-      name: privflags-get
-      doc: Get device private flags.
-
-      attribute-set: privflags
-
-      do: &privflag-get-op
-        request:
-          value: 13
-          attributes:
-            - header
-        reply:
-          value: 14
-          attributes:
-            - header
-            - flags
-      dump: *privflag-get-op
-    -
-      name: privflags-set
-      doc: Set device private flags.
-
-      attribute-set: privflags
-
-      do:
-        request:
-          attributes:
-            - header
-            - flags
-    -
-      name: privflags-ntf
-      doc: Notification for change in device private flags.
-      notify: privflags-get
-
-    -
-      name: rings-get
-      doc: Get ring params.
-
-      attribute-set: rings
-
-      do: &ring-get-op
-        request:
-          attributes:
-            - header
-        reply:
-          attributes:
-            - header
-            - rx-max
-            - rx-mini-max
-            - rx-jumbo-max
-            - tx-max
-            - rx
-            - rx-mini
-            - rx-jumbo
-            - tx
-            - rx-buf-len
-            - tcp-data-split
-            - cqe-size
-            - tx-push
-            - rx-push
-            - tx-push-buf-len
-            - tx-push-buf-len-max
-      dump: *ring-get-op
-    -
-      name: rings-set
-      doc: Set ring params.
-
-      attribute-set: rings
-
-      do:
-        request:
-          attributes:
-            - header
-            - rx
-            - rx-mini
-            - rx-jumbo
-            - tx
-            - rx-buf-len
-            - tcp-data-split
-            - cqe-size
-            - tx-push
-            - rx-push
-    -
-      name: rings-ntf
-      doc: Notification for change in ring params.
-      notify: rings-get
-
-    # TODO: fill in the requests in between
-
+  -
+    name: coalesce
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: rx-usecs
+        type: u32
+      -
+        name: rx-max-frames
+        type: u32
+      -
+        name: rx-usecs-irq
+        type: u32
+      -
+        name: rx-max-frames-irq
+        type: u32
+      -
+        name: tx-usecs
+        type: u32
+      -
+        name: tx-max-frames
+        type: u32
+      -
+        name: tx-usecs-irq
+        type: u32
+      -
+        name: tx-max-frames-irq
+        type: u32
+      -
+        name: stats-block-usecs
+        type: u32
+      -
+        name: use-adaptive-rx
+        type: u8
+      -
+        name: use-adaptive-tx
+        type: u8
+      -
+        name: pkt-rate-low
+        type: u32
+      -
+        name: rx-usecs-low
+        type: u32
+      -
+        name: rx-max-frames-low
+        type: u32
+      -
+        name: tx-usecs-low
+        type: u32
+      -
+        name: tx-max-frames-low
+        type: u32
+      -
+        name: pkt-rate-high
+        type: u32
+      -
+        name: rx-usecs-high
+        type: u32
+      -
+        name: rx-max-frames-high
+        type: u32
+      -
+        name: tx-usecs-high
+        type: u32
+      -
+        name: tx-max-frames-high
+        type: u32
+      -
+        name: rate-sample-interval
+        type: u32
+      -
+        name: use-cqe-mode-tx
+        type: u8
+      -
+        name: use-cqe-mode-rx
+        type: u8
+      -
+        name: tx-aggr-max-bytes
+        type: u32
+      -
+        name: tx-aggr-max-frames
+        type: u32
+      -
+        name: tx-aggr-time-usecs
+        type: u32
+  -
+    name: pause-stat
+    attributes:
+      -
+        name: pad
+        type: u32
+      -
+        name: tx-frames
+        type: u64
+      -
+        name: rx-frames
+        type: u64
+  -
+    name: pause
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: autoneg
+        type: u8
+      -
+        name: rx
+        type: u8
+      -
+        name: tx
+        type: u8
+      -
+        name: stats
+        type: nest
+        nested-attributes: pause-stat
+      -
+        name: stats-src
+        type: u32
+  -
+    name: eee
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: modes-ours
+        type: nest
+        nested-attributes: bitset
+      -
+        name: modes-peer
+        type: nest
+        nested-attributes: bitset
+      -
+        name: active
+        type: u8
+      -
+        name: enabled
+        type: u8
+      -
+        name: tx-lpi-enabled
+        type: u8
+      -
+        name: tx-lpi-timer
+        type: u32
+  -
+    name: tsinfo
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: timestamping
+        type: nest
+        nested-attributes: bitset
+      -
+        name: tx-types
+        type: nest
+        nested-attributes: bitset
+      -
+        name: rx-filters
+        type: nest
+        nested-attributes: bitset
+      -
+        name: phc-index
+        type: u32
+  -
+    name: cable-test-nft-nest-result
+    attributes:
+      -
+        name: pair
+        type: u8
+      -
+        name: code
+        type: u8
+  -
+    name: cable-test-nft-nest-fault-length
+    attributes:
+      -
+        name: pair
+        type: u8
+      -
+        name: cm
+        type: u32
+  -
+    name: cable-test-nft-nest
+    attributes:
+      -
+        name: result
+        type: nest
+        nested-attributes: cable-test-nft-nest-result
+      -
+        name: fault-length
+        type: nest
+        nested-attributes: cable-test-nft-nest-fault-length
+  -
+    name: cable-test
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: status
+        type: u8
+      -
+        name: nest
+        type: nest
+        nested-attributes: cable-test-nft-nest
+  -
+    name: cable-test-tdr-cfg
+    attributes:
+      -
+        name: first
+        type: u32
+      -
+        name: last
+        type: u32
+      -
+        name: step
+        type: u32
+      -
+        name: pari
+        type: u8
+  -
+    name: cable-test-tdr
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: cfg
+        type: nest
+        nested-attributes: cable-test-tdr-cfg
+  -
+    name: tunnel-info-udp-entry
+    attributes:
+      -
+        name: port
+        type: u16
+        byte-order: big-endian
+      -
+        name: type
+        type: u32
+        enum: udp-tunnel-type
+  -
+    name: tunnel-info-udp-table
+    attributes:
+      -
+        name: size
+        type: u32
+      -
+        name: types
+        type: nest
+        nested-attributes: bitset
+      -
+        name: udp-ports
+        type: nest
+        nested-attributes: tunnel-info-udp-entry
+  -
+    name: tunnel-info
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: udp-ports
+        type: nest
+        nested-attributes: tunnel-info-udp-table
+  -
+    name: fec-stat
+    attributes:
+      -
+        name: pad
+        type: u8
+      -
+        name: corrected
+        type: nest
+        nested-attributes: u64-array
+      -
+        name: uncorr
+        type: nest
+        nested-attributes: u64-array
+      -
+        name: corr-bits
+        type: nest
+        nested-attributes: u64-array
+  -
+    name: fec
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: modes
+        type: nest
+        nested-attributes: bitset
+      -
+        name: auto
+        type: u8
+      -
+        name: active
+        type: u32
+      -
+        name: stats
+        type: nest
+        nested-attributes: fec-stat
+  -
+    name: module-eeprom
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: offset
+        type: u32
+      -
+        name: length
+        type: u32
+      -
+        name: page
+        type: u8
+      -
+        name: bank
+        type: u8
+      -
+        name: i2c-address
+        type: u8
+      -
+        name: data
+        type: binary
+  -
+    name: stats-grp
+    attributes:
+      -
+        name: pad
+        type: u32
+      -
+        name: id
+        type: u32
+      -
+        name: ss-id
+        type: u32
+      -
+        name: stat
+        type: nest
+        nested-attributes: u64
+      -
+        name: hist-rx
+        type: nest
+        nested-attributes: u64
+      -
+        name: hist-tx
+        type: nest
+        nested-attributes: u64
+      -
+        name: hist-bkt-low
+        type: u32
+      -
+        name: hist-bkt-hi
+        type: u32
+      -
+        name: hist-bkt-val
+        type: u64
+  -
+    name: stats
+    attributes:
+      -
+        name: pad
+        type: u32
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: groups
+        type: nest
+        nested-attributes: bitset
+      -
+        name: grp
+        type: nest
+        nested-attributes: stats-grp
+      -
+        name: src
+        type: u32
+  -
+    name: phc-vclocks
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: num
+        type: u32
+      -
+        name: index
+        type: nest
+        nested-attributes: s32-array
+  -
+    name: module
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: power-mode-policy
+        type: u8
+      -
+        name: power-mode
+        type: u8
+  -
+    name: pse
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: admin-state
+        type: u32
+      -
+        name: admin-control
+        type: u32
+      -
+        name: pw-d-status
+        type: u32
+  -
+    name: rss
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: context
+        type: u32
+      -
+        name: hfunc
+        type: u32
+      -
+        name: indir
+        type: binary
+      -
+        name: hkey
+        type: binary
+  -
+    name: plca
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: version
+        type: u16
+      -
+        name: enabled
+        type: u8
+      -
+        name: status
+        type: u8
+      -
+        name: node-cnt
+        type: u32
+      -
+        name: node-id
+        type: u32
+      -
+        name: to-tmr
+        type: u32
+      -
+        name: burst-cnt
+        type: u32
+      -
+        name: burst-tmr
+        type: u32
+
+operations:
+  enum-model: directional
+  list:
+    -
+      name: strset-get
+      doc: Get string set from the kernel.
+
+      attribute-set: strset
+
+      do: &strset-get-op
+        request:
+          attributes:
+            - header
+            - stringsets
+            - counts-only
+        reply:
+          attributes:
+            - header
+            - stringsets
+      dump: *strset-get-op
+    -
+      name: linkinfo-get
+      doc: Get link info.
+
+      attribute-set: linkinfo
+
+      do: &linkinfo-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &linkinfo
+            - header
+            - port
+            - phyaddr
+            - tp-mdix
+            - tp-mdix-ctrl
+            - transceiver
+      dump: *linkinfo-get-op
+    -
+      name: linkinfo-set
+      doc: Set link info.
+
+      attribute-set: linkinfo
+
+      do:
+        request:
+          attributes: *linkinfo
+    -
+      name: linkinfo-ntf
+      doc: Notification for change in link info.
+      notify: linkinfo-get
+    -
+      name: linkmodes-get
+      doc: Get link modes.
+
+      attribute-set: linkmodes
+
+      do: &linkmodes-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &linkmodes
+            - header
+            - autoneg
+            - ours
+            - peer
+            - speed
+            - duplex
+            - master-slave-cfg
+            - master-slave-state
+            - master-slave-lanes
+            - rate-matching
+      dump: *linkmodes-get-op
+    -
+      name: linkmodes-set
+      doc: Set link modes.
+
+      attribute-set: linkmodes
+
+      do:
+        request:
+          attributes: *linkmodes
+    -
+      name: linkmodes-ntf
+      doc: Notification for change in link modes.
+      notify: linkmodes-get
+    -
+      name: linkstate-get
+      doc: Get link state.
+
+      attribute-set: linkstate
+
+      do: &linkstate-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - link
+            - sqi
+            - sqi-max
+            - ext-state
+            - ext-substate
+            - down-cnt
+      dump: *linkstate-get-op
+    -
+      name: debug-get
+      doc: Get debug message mask.
+
+      attribute-set: debug
+
+      do: &debug-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &debug
+            - header
+            - msgmask
+      dump: *debug-get-op
+    -
+      name: debug-set
+      doc: Set debug message mask.
+
+      attribute-set: debug
+
+      do:
+        request:
+          attributes: *debug
+    -
+      name: debug-ntf
+      doc: Notification for change in debug message mask.
+      notify: debug-get
+    -
+      name: wol-get
+      doc: Get WOL params.
+
+      attribute-set: wol
+
+      do: &wol-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &wol
+            - header
+            - modes
+            - sopass
+      dump: *wol-get-op
+    -
+      name: wol-set
+      doc: Set WOL params.
+
+      attribute-set: wol
+
+      do:
+        request:
+          attributes: *wol
+    -
+      name: wol-ntf
+      doc: Notification for change in WOL params.
+      notify: wol-get
+    -
+      name: features-get
+      doc: Get features.
+
+      attribute-set: features
+
+      do: &feature-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &feature
+            - header
+            # User-changeable features.
+            - hw
+            # User-requested features.
+            - wanted
+            # Currently active features.
+            - active
+            # Unchangeable features.
+            - nochange
+      dump: *feature-get-op
+    -
+      name: features-set
+      doc: Set features.
+
+      attribute-set: features
+
+      do: &feature-set-op
+        request:
+          attributes: *feature
+        reply:
+          attributes: *feature
+    -
+      name: features-ntf
+      doc: Notification for change in features.
+      notify: features-get
+    -
+      name: privflags-get
+      doc: Get device private flags.
+
+      attribute-set: privflags
+
+      do: &privflag-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &privflag
+            - header
+            - flags
+      dump: *privflag-get-op
+    -
+      name: privflags-set
+      doc: Set device private flags.
+
+      attribute-set: privflags
+
+      do:
+        request:
+          attributes: *privflag
+    -
+      name: privflags-ntf
+      doc: Notification for change in device private flags.
+      notify: privflags-get
+
+    -
+      name: rings-get
+      doc: Get ring params.
+
+      attribute-set: rings
+
+      do: &ring-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &ring
+            - header
+            - rx-max
+            - rx-mini-max
+            - rx-jumbo-max
+            - tx-max
+            - rx
+            - rx-mini
+            - rx-jumbo
+            - tx
+            - rx-buf-len
+            - tcp-data-split
+            - cqe-size
+            - tx-push
+            - rx-push
+            - tx-push-buf-len
+            - tx-push-buf-len-max
+      dump: *ring-get-op
+    -
+      name: rings-set
+      doc: Set ring params.
+
+      attribute-set: rings
+
+      do:
+        request:
+          attributes: *ring
+    -
+      name: rings-ntf
+      doc: Notification for change in ring params.
+      notify: rings-get
+    -
+      name: channels-get
+      doc: Get channel params.
+
+      attribute-set: channels
+
+      do: &channel-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &channel
+            - header
+            - rx-max
+            - tx-max
+            - other-max
+            - combined-max
+            - rx-count
+            - tx-count
+            - other-count
+            - combined-count
+      dump: *channel-get-op
+    -
+      name: channels-set
+      doc: Set channel params.
+
+      attribute-set: channels
+
+      do:
+        request:
+          attributes: *channel
+    -
+      name: channels-ntf
+      doc: Notification for change in channel params.
+      notify: channels-get
+    -
+      name: coalesce-get
+      doc: Get coalesce params.
+
+      attribute-set: coalesce
+
+      do: &coalesce-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &coalesce
+            - header
+            - rx-usecs
+            - rx-max-frames
+            - rx-usecs-irq
+            - rx-max-frames-irq
+            - tx-usecs
+            - tx-max-frames
+            - tx-usecs-irq
+            - tx-max-frames-irq
+            - stats-block-usecs
+            - use-adaptive-rx
+            - use-adaptive-tx
+            - pkt-rate-low
+            - rx-usecs-low
+            - rx-max-frames-low
+            - tx-usecs-low
+            - tx-max-frames-low
+            - pkt-rate-high
+            - rx-usecs-high
+            - rx-max-frames-high
+            - tx-usecs-high
+            - tx-max-frames-high
+            - rate-sample-interval
+            - use-cqe-mode-tx
+            - use-cqe-mode-rx
+            - tx-aggr-max-bytes
+            - tx-aggr-max-frames
+            - tx-aggr-time-usecs
+      dump: *coalesce-get-op
+    -
+      name: coalesce-set
+      doc: Set coalesce params.
+
+      attribute-set: coalesce
+
+      do:
+        request:
+          attributes: *coalesce
+    -
+      name: coalesce-ntf
+      doc: Notification for change in coalesce params.
+      notify: coalesce-get
+    -
+      name: pause-get
+      doc: Get pause params.
+
+      attribute-set: pause
+
+      do: &pause-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &pause
+            - header
+            - autoneg
+            - rx
+            - tx
+            - stats
+            - stats-src
+      dump: *pause-get-op
+    -
+      name: pause-set
+      doc: Set pause params.
+
+      attribute-set: pause
+
+      do:
+        request:
+          attributes: *pause
+    -
+      name: pause-ntf
+      doc: Notification for change in pause params.
+      notify: pause-get
+    -
+      name: eee-get
+      doc: Get eee params.
+
+      attribute-set: eee
+
+      do: &eee-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &eee
+            - header
+            - modes-ours
+            - modes-peer
+            - active
+            - enabled
+            - tx-lpi-enabled
+            - tx-lpi-timer
+      dump: *eee-get-op
+    -
+      name: eee-set
+      doc: Set eee params.
+
+      attribute-set: eee
+
+      do:
+        request:
+          attributes: *eee
+    -
+      name: eee-ntf
+      doc: Notification for change in eee params.
+      notify: eee-get
+    -
+      name: tsinfo-get
+      doc: Get tsinfo params.
+
+      attribute-set: tsinfo
+
+      do: &tsinfo-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - timestamping
+            - tx-types
+            - rx-filters
+            - phc-index
+      dump: *tsinfo-get-op
+    -
+      name: cable-test-act
+      doc: Cable test.
+
+      attribute-set: cable-test
+
+      do:
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - cable-test-nft-nest
+    -
+      name: cable-test-tdr-act
+      doc: Cable test TDR.
+
+      attribute-set: cable-test-tdr
+
+      do:
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - cable-test-tdr-cfg
+    -
+      name: tunnel-info-get
+      doc: Get tsinfo params.
+
+      attribute-set: tunnel-info
+
+      do: &tunnel-info-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - udp-ports
+      dump: *tunnel-info-get-op
+    -
+      name: fec-get
+      doc: Get FEC params.
+
+      attribute-set: fec
+
+      do: &fec-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &fec
+            - header
+            - modes
+            - auto
+            - active
+            - stats
+      dump: *fec-get-op
+    -
+      name: fec-set
+      doc: Set FEC params.
+
+      attribute-set: fec
+
+      do:
+        request:
+          attributes: *fec
+    -
+      name: fec-ntf
+      doc: Notification for change in FEC params.
+      notify: fec-get
+    -
+      name: module-eeprom-get
+      doc: Get module EEPROM params.
+
+      attribute-set: module-eeprom
+
+      do: &module-eeprom-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - offset
+            - length
+            - page
+            - bank
+            - i2c-address
+            - data
+      dump: *module-eeprom-get-op
+    -
+      name: stats-get
+      doc: Get statistics.
+
+      attribute-set: stats
+
+      do: &stats-get-op
+        request:
+          attributes:
+            - header
+            - groups
+        reply:
+          attributes:
+            - header
+            - groups
+            - grp
+            - src
+      dump: *stats-get-op
+    -
+      name: phc-vclocks-get
+      doc: Get PHC VCLOCKs.
+
+      attribute-set: phc-vclocks
+
+      do: &phc-vclocks-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - num
+      dump: *phc-vclocks-get-op
+    -
+      name: module-get
+      doc: Get module params.
+
+      attribute-set: module
+
+      do: &module-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &module
+            - header
+            - power-mode-policy
+            - power-mode
+      dump: *module-get-op
+    -
+      name: module-set
+      doc: Set module params.
+
+      attribute-set: module
+
+      do:
+        request:
+          attributes: *module
+    -
+      name: module-ntf
+      doc: Notification for change in module params.
+      notify: module-get
+    -
+      name: pse-get
+      doc: Get Power Sourcing Equipment params.
+
+      attribute-set: pse
+
+      do: &pse-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &pse
+            - header
+            - admin-state
+            - admin-control
+            - pw-d-status
+      dump: *pse-get-op
+    -
+      name: pse-set
+      doc: Set Power Sourcing Equipment params.
+
+      attribute-set: pse
+
+      do:
+        request:
+          attributes: *pse
+    -
+      name: rss-get
+      doc: Get RSS params.
+
+      attribute-set: rss
+
+      do: &rss-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes:
+            - header
+            - context
+            - hfunc
+            - indir
+            - hkey
+      dump: *rss-get-op
+    -
+      name: plca-get
+      doc: Get PLCA params.
+
+      attribute-set: plca
+
+      do: &plca-get-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: &plca
+            - header
+            - version
+            - enabled
+            - status
+            - node-cnt
+            - node-id
+            - to-tmr
+            - burst-cnt
+            - burst-tmr
+      dump: *plca-get-op
+    -
+      name: plca-set
+      doc: Set PLCA params.
+
+      attribute-set: plca
+
+      do:
+        request:
+          attributes: *plca
+    -
+      name: plca-get-status
+      doc: Get PLCA status params.
+
+      attribute-set: plca
+
+      do: &plca-get-status-op
+        request:
+          attributes:
+            - header
+        reply:
+          attributes: *plca
+      dump: *plca-get-status-op
+    -
+      name: plca-ntf
+      doc: Notification for change in PLCA params.
+      notify: plca-get
     -
       name: mm-get
       doc: Get MAC Merge configuration and state
@@ -354,11 +1610,9 @@ operations:
 
       do: &mm-get-op
         request:
-          value: 42
           attributes:
             - header
         reply:
-          value: 42
           attributes:
             - header
             - pmac-enabled
-- 
cgit v1.2.3-70-g09d2


From 48993e22d23ae1bda1db3616f5d9baa4e7b18d35 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Wed, 29 Mar 2023 15:16:54 -0700
Subject: tools: ynl: replace print with NlError

Instead of dumping the error on the stdout, make the callee and
opportunity to decide what to do with it. This is mostly for the
ethtool testing.

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/ynl.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 8778994d40c0..373c0edb5f83 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -67,6 +67,14 @@ class Netlink:
     NLMSGERR_ATTR_MISS_NEST = 6
 
 
+class NlError(Exception):
+  def __init__(self, nl_msg):
+    self.nl_msg = nl_msg
+
+  def __str__(self):
+    return f"Netlink error: {os.strerror(-self.nl_msg.error)}\n{self.nl_msg}"
+
+
 class NlAttr:
     type_formats = { 'u8' : ('B', 1), 's8' : ('b', 1),
                      'u16': ('H', 2), 's16': ('h', 2),
@@ -551,9 +559,7 @@ class YnlFamily(SpecFamily):
                     self._decode_extack(msg, op.attr_set, nl_msg.extack)
 
                 if nl_msg.error:
-                    print("Netlink error:", os.strerror(-nl_msg.error))
-                    print(nl_msg)
-                    return
+                    raise NlError(nl_msg)
                 if nl_msg.done:
                     if nl_msg.extack:
                         print("Netlink warning:")
-- 
cgit v1.2.3-70-g09d2


From f3d07b02b2b8eba5b0e168405614e15cd6617a43 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Wed, 29 Mar 2023 15:16:55 -0700
Subject: tools: ynl: ethtool testing tool

This is what I've been using to see whether the spec makes sense.
A small subset of getters (mostly the unprivileged ones) is implemented.
Some setters (channels) also work.
Setters for messages with bitmasks are not implemented.

Initially I was trying to make this tool look 1:1 like real ethtool,
but eventually gave up :-)

Sample output:

$ ./tools/net/ynl/ethtool enp0s31f6
Settings for enp0s31f6:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half
100baseT/Full 1000baseT/Full
Supported pause frame use: no
Supports auto-negotiation: yes
Supported FEC modes: Not reported
Speed: Unknown!
Duplex: Unknown! (255)
Auto-negotiation: on
Port: Twisted Pair
PHYAD: 2
Transceiver: Internal
MDI-X: Unknown (auto)
Current message level: drv probe link
Link detected: no

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ethtool       | 424 ++++++++++++++++++++++++++++++++++++++++++++
 tools/net/ynl/lib/nlspec.py |   9 +
 tools/net/ynl/lib/ynl.py    |  11 ++
 3 files changed, 444 insertions(+)
 create mode 100755 tools/net/ynl/ethtool

diff --git a/tools/net/ynl/ethtool b/tools/net/ynl/ethtool
new file mode 100755
index 000000000000..5fb1d670693a
--- /dev/null
+++ b/tools/net/ynl/ethtool
@@ -0,0 +1,424 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+import argparse
+import json
+import pprint
+import sys
+import re
+
+from lib import YnlFamily
+
+def args_to_req(ynl, op_name, args, req):
+    """
+    Verify and convert command-line arguments to the ynl-compatible request.
+    """
+    valid_attrs = ynl.operation_do_attributes(op_name)
+    valid_attrs.remove('header') # not user-provided
+
+    if len(args) == 0:
+        print(f'no attributes, expected: {valid_attrs}')
+        sys.exit(1)
+
+    i = 0
+    while i < len(args):
+        attr = args[i]
+        if i + 1 >= len(args):
+            print(f'expected value for \'{attr}\'')
+            sys.exit(1)
+
+        if attr not in valid_attrs:
+            print(f'invalid attribute \'{attr}\', expected: {valid_attrs}')
+            sys.exit(1)
+
+        val = args[i+1]
+        i += 2
+
+        req[attr] = val
+
+def print_field(reply, *desc):
+    """
+    Pretty-print a set of fields from the reply. desc specifies the
+    fields and the optional type (bool/yn).
+    """
+    if len(desc) == 0:
+        return print_field(reply, *zip(reply.keys(), reply.keys()))
+
+    for spec in desc:
+        try:
+            field, name, tp = spec
+        except:
+            field, name = spec
+            tp = 'int'
+
+        value = reply.get(field, None)
+        if tp == 'yn':
+            value = 'yes' if value else 'no'
+        elif tp == 'bool' or isinstance(value, bool):
+            value = 'on' if value else 'off'
+        else:
+            value = 'n/a' if value is None else value
+
+        print(f'{name}: {value}')
+
+def print_speed(name, value):
+    """
+    Print out the speed-like strings from the value dict.
+    """
+    speed_re = re.compile(r'[0-9]+base[^/]+/.+')
+    speed = [ k for k, v in value.items() if v and speed_re.match(k) ]
+    print(f'{name}: {" ".join(speed)}')
+
+def doit(ynl, args, op_name):
+    """
+    Prepare request header, parse arguments and doit.
+    """
+    req = {
+        'header': {
+          'dev-name': args.device,
+        },
+    }
+
+    args_to_req(ynl, op_name, args.args, req)
+    ynl.do(op_name, req)
+
+def dumpit(ynl, args, op_name, extra = {}):
+    """
+    Prepare request header, parse arguments and dumpit (filtering out the
+    devices we're not interested in).
+    """
+    reply = ynl.dump(op_name, { 'header': {} } | extra)
+    if not reply:
+        return {}
+
+    for msg in reply:
+        if msg['header']['dev-name'] == args.device:
+            if args.json:
+                pprint.PrettyPrinter().pprint(msg)
+                sys.exit(0)
+            msg.pop('header', None)
+            return msg
+
+    print(f"Not supported for device {args.device}")
+    sys.exit(1)
+
+def bits_to_dict(attr):
+    """
+    Convert ynl-formatted bitmask to a dict of bit=value.
+    """
+    ret = {}
+    if 'bits' not in attr:
+        return dict()
+    if 'bit' not in attr['bits']:
+        return dict()
+    for bit in attr['bits']['bit']:
+        if bit['name'] == '':
+            continue
+        name = bit['name']
+        value = bit.get('value', False)
+        ret[name] = value
+    return ret
+
+def main():
+    parser = argparse.ArgumentParser(description='ethtool wannabe')
+    parser.add_argument('--json', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--show-priv-flags', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--set-priv-flags', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--show-eee', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--set-eee', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-a', '--show-pause', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-A', '--set-pause', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-c', '--show-coalesce', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-C', '--set-coalesce', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-g', '--show-ring', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-G', '--set-ring', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-k', '--show-features', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-K', '--set-features', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-l', '--show-channels', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-L', '--set-channels', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-T', '--show-time-stamping', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-S', '--statistics', action=argparse.BooleanOptionalAction)
+    # TODO: --show-tunnels        tunnel-info-get
+    # TODO: --show-module         module-get
+    # TODO: --get-plca-cfg        plca-get
+    # TODO: --get-plca-status     plca-get-status
+    # TODO: --show-mm             mm-get
+    # TODO: --show-fec            fec-get
+    # TODO: --dump-module-eerpom  module-eeprom-get
+    # TODO:                       pse-get
+    # TODO:                       rss-get
+    parser.add_argument('device', metavar='device', type=str)
+    parser.add_argument('args', metavar='args', type=str, nargs='*')
+    global args
+    args = parser.parse_args()
+
+    spec = '/usr/local/google/home/sdf/src/linux/Documentation/netlink/specs/ethtool.yaml'
+    schema = '/usr/local/google/home/sdf/src/linux/Documentation/netlink/genetlink-legacy.yaml'
+
+    ynl = YnlFamily(spec, schema)
+
+    if args.set_priv_flags:
+        # TODO: parse the bitmask
+        print("not implemented")
+        return
+
+    if args.set_eee:
+        return doit(ynl, args, 'eee-set')
+
+    if args.set_pause:
+        return doit(ynl, args, 'pause-set')
+
+    if args.set_coalesce:
+        return doit(ynl, args, 'coalesce-set')
+
+    if args.set_features:
+        # TODO: parse the bitmask
+        print("not implemented")
+        return
+
+    if args.set_channels:
+        return doit(ynl, args, 'channels-set')
+
+    if args.set_ring:
+        return doit(ynl, args, 'rings-set')
+
+    if args.show_priv_flags:
+        flags = bits_to_dict(dumpit(ynl, args, 'privflags-get')['flags'])
+        print_field(flags)
+        return
+
+    if args.show_eee:
+        eee = dumpit(ynl, args, 'eee-get')
+        ours = bits_to_dict(eee['modes-ours'])
+        peer = bits_to_dict(eee['modes-peer'])
+
+        if 'enabled' in eee:
+            status = 'enabled' if eee['enabled'] else 'disabled'
+            if 'active' in eee and eee['active']:
+                status = status + ' - active'
+            else:
+                status = status + ' - inactive'
+        else:
+            status = 'not supported'
+
+        print(f'EEE status: {status}')
+        print_field(eee, ('tx-lpi-timer', 'Tx LPI'))
+        print_speed('Advertised EEE link modes', ours)
+        print_speed('Link partner advertised EEE link modes', peer)
+
+        return
+
+    if args.show_pause:
+        print_field(dumpit(ynl, args, 'pause-get'),
+                ('autoneg', 'Autonegotiate', 'bool'),
+                ('rx', 'RX', 'bool'),
+                ('tx', 'TX', 'bool'))
+        return
+
+    if args.show_coalesce:
+        print_field(dumpit(ynl, args, 'coalesce-get'))
+        return
+
+    if args.show_features:
+        reply = dumpit(ynl, args, 'features-get')
+        available = bits_to_dict(reply['hw'])
+        requested = bits_to_dict(reply['wanted']).keys()
+        active = bits_to_dict(reply['active']).keys()
+        never_changed = bits_to_dict(reply['nochange']).keys()
+
+        for f in sorted(available):
+            value = "off"
+            if f in active:
+                value = "on"
+
+            fixed = ""
+            if f not in available or f in never_changed:
+                fixed = " [fixed]"
+
+            req = ""
+            if f in requested:
+                if f in active:
+                    req = " [requested on]"
+                else:
+                    req = " [requested off]"
+
+            print(f'{f}: {value}{fixed}{req}')
+
+        return
+
+    if args.show_channels:
+        reply = dumpit(ynl, args, 'channels-get')
+        print(f'Channel parameters for {args.device}:')
+
+        print(f'Pre-set maximums:')
+        print_field(reply,
+            ('rx-max', 'RX'),
+            ('tx-max', 'TX'),
+            ('other-max', 'Other'),
+            ('combined-max', 'Combined'))
+
+        print(f'Current hardware settings:')
+        print_field(reply,
+            ('rx-count', 'RX'),
+            ('tx-count', 'TX'),
+            ('other-count', 'Other'),
+            ('combined-count', 'Combined'))
+
+        return
+
+    if args.show_ring:
+        reply = dumpit(ynl, args, 'channels-get')
+
+        print(f'Ring parameters for {args.device}:')
+
+        print(f'Pre-set maximums:')
+        print_field(reply,
+            ('rx-max', 'RX'),
+            ('rx-mini-max', 'RX Mini'),
+            ('rx-jumbo-max', 'RX Jumbo'),
+            ('tx-max', 'TX'))
+
+        print(f'Current hardware settings:')
+        print_field(reply,
+            ('rx', 'RX'),
+            ('rx-mini', 'RX Mini'),
+            ('rx-jumbo', 'RX Jumbo'),
+            ('tx', 'TX'))
+
+        print_field(reply,
+            ('rx-buf-len', 'RX Buf Len'),
+            ('cqe-size', 'CQE Size'),
+            ('tx-push', 'TX Push', 'bool'))
+
+        return
+
+    if args.statistics:
+        print(f'NIC statistics:')
+
+        # TODO: pass id?
+        strset = dumpit(ynl, args, 'strset-get')
+        pprint.PrettyPrinter().pprint(strset)
+
+        req = {
+          'groups': {
+            'size': 1,
+            'bits': {
+              'bit':
+                # TODO: support passing the bitmask
+                #[
+                  #{ 'name': 'eth-phy', 'value': True },
+                  { 'name': 'eth-mac', 'value': True },
+                  #{ 'name': 'eth-ctrl', 'value': True },
+                  #{ 'name': 'rmon', 'value': True },
+                #],
+            },
+          },
+        }
+
+        rsp = dumpit(ynl, args, 'stats-get', req)
+        pprint.PrettyPrinter().pprint(rsp)
+        return
+
+    if args.show_time_stamping:
+        tsinfo = dumpit(ynl, args, 'tsinfo-get')
+
+        print(f'Time stamping parameters for {args.device}:')
+
+        print('Capabilities:')
+        [print(f'\t{v}') for v in bits_to_dict(tsinfo['timestamping'])]
+
+        print(f'PTP Hardware Clock: {tsinfo["phc-index"]}')
+
+        print('Hardware Transmit Timestamp Modes:')
+        [print(f'\t{v}') for v in bits_to_dict(tsinfo['tx-types'])]
+
+        print('Hardware Receive Filter Modes:')
+        [print(f'\t{v}') for v in bits_to_dict(tsinfo['rx-filters'])]
+        return
+
+    print(f'Settings for {args.device}:')
+    linkmodes = dumpit(ynl, args, 'linkmodes-get')
+    ours = bits_to_dict(linkmodes['ours'])
+
+    supported_ports = ('TP',  'AUI', 'BNC', 'MII', 'FIBRE', 'Backplane')
+    ports = [ p for p in supported_ports if ours.get(p, False)]
+    print(f'Supported ports: [ {" ".join(ports)} ]')
+
+    print_speed('Supported link modes', ours)
+
+    print_field(ours, ('Pause', 'Supported pause frame use', 'yn'))
+    print_field(ours, ('Autoneg', 'Supports auto-negotiation', 'yn'))
+
+    supported_fec = ('None',  'PS', 'BASER', 'LLRS')
+    fec = [ p for p in supported_fec if ours.get(p, False)]
+    fec_str = " ".join(fec)
+    if len(fec) == 0:
+        fec_str = "Not reported"
+
+    print(f'Supported FEC modes: {fec_str}')
+
+    speed = 'Unknown!'
+    if linkmodes['speed'] > 0 and linkmodes['speed'] < 0xffffffff:
+        speed = f'{linkmodes["speed"]}Mb/s'
+    print(f'Speed: {speed}')
+
+    duplex_modes = {
+            0: 'Half',
+            1: 'Full',
+    }
+    duplex = duplex_modes.get(linkmodes["duplex"], None)
+    if not duplex:
+        duplex = f'Unknown! ({linkmodes["duplex"]})'
+    print(f'Duplex: {duplex}')
+
+    autoneg = "off"
+    if linkmodes.get("autoneg", 0) != 0:
+        autoneg = "on"
+    print(f'Auto-negotiation: {autoneg}')
+
+    ports = {
+            0: 'Twisted Pair',
+            1: 'AUI',
+            2: 'MII',
+            3: 'FIBRE',
+            4: 'BNC',
+            5: 'Directly Attached Copper',
+            0xef: 'None',
+    }
+    linkinfo = dumpit(ynl, args, 'linkinfo-get')
+    print(f'Port: {ports.get(linkinfo["port"], "Other")}')
+
+    print_field(linkinfo, ('phyaddr', 'PHYAD'))
+
+    transceiver = {
+            0: 'Internal',
+            1: 'External',
+    }
+    print(f'Transceiver: {transceiver.get(linkinfo["transceiver"], "Unknown")}')
+
+    mdix_ctrl = {
+            1: 'off',
+            2: 'on',
+    }
+    mdix = mdix_ctrl.get(linkinfo['tp-mdix-ctrl'], None)
+    if mdix:
+        mdix = mdix + ' (forced)'
+    else:
+        mdix = mdix_ctrl.get(linkinfo['tp-mdix'], 'Unknown (auto)')
+    print(f'MDI-X: {mdix}')
+
+    debug = dumpit(ynl, args, 'debug-get')
+    msgmask = bits_to_dict(debug.get("msgmask", [])).keys()
+    print(f'Current message level: {" ".join(msgmask)}')
+
+    linkstate = dumpit(ynl, args, 'linkstate-get')
+    detected_states = {
+            0: 'no',
+            1: 'yes',
+    }
+    # TODO: wol-get
+    detected = detected_states.get(linkstate['link'], 'unknown')
+    print(f'Link detected: {detected}')
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index 0c43022ff7e6..a0241add3839 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -444,6 +444,15 @@ class SpecFamily(SpecElement):
 
             self.msgs[op.name] = op
 
+    def find_operation(self, name):
+      """
+      For a given operation name, find and return operation spec.
+      """
+      for op in self.yaml['operations']['list']:
+        if name == op['name']:
+          return op
+      return None
+
     def resolve(self):
         self.resolve_up(super())
 
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 373c0edb5f83..7690e0b0cb3f 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -527,6 +527,17 @@ class YnlFamily(SpecFamily):
 
                 self.handle_ntf(nl_msg, gm)
 
+    def operation_do_attributes(self, name):
+      """
+      For a given operation name, find and return a supported
+      set of attributes (as a dict).
+      """
+      op = self.find_operation(name)
+      if not op:
+        return None
+
+      return op['do']['request']['attributes'].copy()
+
     def _op(self, method, vals, dump=False):
         op = self.ops[method]
 
-- 
cgit v1.2.3-70-g09d2


From ae63ad9b2cc766bc4b3b6328e68dd27995abb0fe Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 29 Mar 2023 16:45:33 +0800
Subject: macvlan: Fix mc_filter calculation

On Wed, Mar 29, 2023 at 08:10:26AM +0000, patchwork-bot+netdevbpf@kernel.org wrote:
>
> Here is the summary with links:
>   - [1/2] macvlan: Skip broadcast queue if multicast with single receiver
>     https://git.kernel.org/netdev/net-next/c/d45276e75e90
>   - [2/2] macvlan: Add netlink attribute for broadcast cutoff
>     https://git.kernel.org/netdev/net-next/c/954d1fa1ac93

Sorry, I made an error and posted my patches from an earlier
revision so a follow-up fix was missing:

---8<---
The bc_cutoff patch broke the calculation of mc_filter causing
some multicast packets to not make it through to the targeted
device.

Fix this by checking whether vlan is set instead of cutoff >= 0.

Also move the cutoff < 0 logic into macvlan_recompute_bc_filter
so that it doesn't change the mc_filter at all.

Fixes: d45276e75e90 ("macvlan: Skip broadcast queue if multicast with single receiver")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macvlan.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 4215106adc40..4a53debf9d7c 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -792,24 +792,20 @@ static void macvlan_compute_filter(unsigned long *mc_filter,
 				   struct macvlan_dev *vlan, int cutoff)
 {
 	if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) {
-		if (cutoff >= 0)
-			bitmap_fill(mc_filter, MACVLAN_MC_FILTER_SZ);
-		else
-			bitmap_zero(mc_filter, MACVLAN_MC_FILTER_SZ);
+		bitmap_fill(mc_filter, MACVLAN_MC_FILTER_SZ);
 	} else {
 		DECLARE_BITMAP(filter, MACVLAN_MC_FILTER_SZ);
 		struct netdev_hw_addr *ha;
 
 		bitmap_zero(filter, MACVLAN_MC_FILTER_SZ);
 		netdev_for_each_mc_addr(ha, dev) {
-			if (cutoff >= 0 && ha->synced <= cutoff)
+			if (!vlan && ha->synced <= cutoff)
 				continue;
 
 			__set_bit(mc_hash(vlan, ha->addr), filter);
 		}
 
-		if (cutoff >= 0)
-			__set_bit(mc_hash(vlan, dev->broadcast), filter);
+		__set_bit(mc_hash(vlan, dev->broadcast), filter);
 
 		bitmap_copy(mc_filter, filter, MACVLAN_MC_FILTER_SZ);
 	}
@@ -817,6 +813,11 @@ static void macvlan_compute_filter(unsigned long *mc_filter,
 
 static void macvlan_recompute_bc_filter(struct macvlan_dev *vlan)
 {
+	if (vlan->port->bc_cutoff < 0) {
+		bitmap_zero(vlan->port->bc_filter, MACVLAN_MC_FILTER_SZ);
+		return;
+	}
+
 	macvlan_compute_filter(vlan->port->bc_filter, vlan->lowerdev, NULL,
 			       vlan->port->bc_cutoff);
 }
-- 
cgit v1.2.3-70-g09d2


From ee05d90d0ac7f1f6e8e3403f02a39d83e0ce6729 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Wed, 29 Mar 2023 13:13:48 -0700
Subject: tcp: Refine SYN handling for PAWS.

Our Network Load Balancer (NLB) [0] has multiple nodes with different
IP addresses, and each node forwards TCP flows from clients to backend
targets.  NLB has an option to preserve the client's source IP address
and port when routing packets to backend targets. [1]

When a client connects to two different NLB nodes, they may select the
same backend target.  Then, if the client has used the same source IP
and port, the two flows at the backend side will have the same 4-tuple.

While testing around such cases, I saw these sequences on the backend
target.

IP 10.0.0.215.60000 > 10.0.3.249.10000: Flags [S], seq 2819965599, win 62727, options [mss 8365,sackOK,TS val 1029816180 ecr 0,nop,wscale 7], length 0
IP 10.0.3.249.10000 > 10.0.0.215.60000: Flags [S.], seq 3040695044, ack 2819965600, win 62643, options [mss 8961,sackOK,TS val 1224784076 ecr 1029816180,nop,wscale 7], length 0
IP 10.0.0.215.60000 > 10.0.3.249.10000: Flags [.], ack 1, win 491, options [nop,nop,TS val 1029816181 ecr 1224784076], length 0
IP 10.0.0.215.60000 > 10.0.3.249.10000: Flags [S], seq 2681819307, win 62727, options [mss 8365,sackOK,TS val 572088282 ecr 0,nop,wscale 7], length 0
IP 10.0.3.249.10000 > 10.0.0.215.60000: Flags [.], ack 1, win 490, options [nop,nop,TS val 1224794914 ecr 1029816181,nop,nop,sack 1 {4156821004:4156821005}], length 0

It seems to be working correctly, but the last ACK was generated by
tcp_send_dupack() and PAWSEstab was increased.  This is because the
second connection has a smaller timestamp than the first one.

In this case, we should send a dup ACK in tcp_send_challenge_ack()
to increase the correct counter and rate-limit it properly.

Let's check the SYN flag after the PAWS tests to avoid adding unnecessary
overhead for most packets.

Link: https://docs.aws.amazon.com/elasticloadbalancing/latest/network/introduction.html [0]
Link: https://docs.aws.amazon.com/elasticloadbalancing/latest/network/load-balancer-target-groups.html#client-ip-preservation [1]
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_input.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2b75cd9e2e92..a057330d6f59 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5714,6 +5714,8 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
 	    tp->rx_opt.saw_tstamp &&
 	    tcp_paws_discard(sk, skb)) {
 		if (!th->rst) {
+			if (unlikely(th->syn))
+				goto syn_challenge;
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
 			if (!tcp_oow_rate_limited(sock_net(sk), skb,
 						  LINUX_MIB_TCPACKSKIPPEDPAWS,
-- 
cgit v1.2.3-70-g09d2


From df1875c4f345092716f1529d6e8c43e7fc7f250a Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Tue, 28 Mar 2023 01:07:42 +0800
Subject: wifi: mac80211: enable EHT mesh support

Similar to AP beacon, this enables the basic mesh EHT mode, including
EHT operation IE and the fixed field of EHT operation information IE.
As for the optional part (i.e. preamble puncturing bitmap) will be
added in future patch.

Tested-by: Lian Chen <lian.chen@mediatek.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Link: https://lore.kernel.org/r/1e0ddb9001312451c3e99c4eed2072caf8075f61.1679935259.git.ryder.lee@mediatek.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  2 ++
 net/mac80211/mesh.c        | 73 ++++++++++++++++++++++++++++++++++++++++++++--
 net/mac80211/mesh.h        |  4 +++
 net/mac80211/mesh_plink.c  | 16 ++++++++--
 net/mac80211/util.c        | 71 ++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 162 insertions(+), 4 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 8c058351e629..9b7e184430b8 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2479,6 +2479,8 @@ void ieee80211_ie_build_he_6ghz_cap(struct ieee80211_sub_if_data *sdata,
 				    enum ieee80211_smps_mode smps_mode,
 				    struct sk_buff *skb);
 u8 *ieee80211_ie_build_he_oper(u8 *pos, struct cfg80211_chan_def *chandef);
+u8 *ieee80211_ie_build_eht_oper(u8 *pos, struct cfg80211_chan_def *chandef,
+				const struct ieee80211_sta_eht_cap *eht_cap);
 int ieee80211_parse_bitrates(enum nl80211_chan_width width,
 			     const struct ieee80211_supported_band *sband,
 			     const u8 *srates, int srates_len, u32 *rates);
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index 6b94cf2a4046..f72333201903 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -105,7 +105,7 @@ bool mesh_matches_local(struct ieee80211_sub_if_data *sdata,
 	ieee80211_chandef_vht_oper(&sdata->local->hw, vht_cap_info,
 				   ie->vht_operation, ie->ht_operation,
 				   &sta_chan_def);
-	ieee80211_chandef_he_6ghz_oper(sdata, ie->he_operation, NULL,
+	ieee80211_chandef_he_6ghz_oper(sdata, ie->he_operation, ie->eht_operation,
 				       &sta_chan_def);
 
 	if (!cfg80211_chandef_compatible(&sdata->vif.bss_conf.chandef,
@@ -639,6 +639,65 @@ int mesh_add_he_6ghz_cap_ie(struct ieee80211_sub_if_data *sdata,
 	return 0;
 }
 
+int mesh_add_eht_cap_ie(struct ieee80211_sub_if_data *sdata,
+			struct sk_buff *skb, u8 ie_len)
+{
+	const struct ieee80211_sta_he_cap *he_cap;
+	const struct ieee80211_sta_eht_cap *eht_cap;
+	struct ieee80211_supported_band *sband;
+	u8 *pos;
+
+	sband = ieee80211_get_sband(sdata);
+	if (!sband)
+		return -EINVAL;
+
+	he_cap = ieee80211_get_he_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT);
+	eht_cap = ieee80211_get_eht_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT);
+	if (!he_cap || !eht_cap ||
+	    sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_20_NOHT ||
+	    sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_5 ||
+	    sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_10)
+		return 0;
+
+	if (skb_tailroom(skb) < ie_len)
+		return -ENOMEM;
+
+	pos = skb_put(skb, ie_len);
+	ieee80211_ie_build_eht_cap(pos, he_cap, eht_cap, pos + ie_len, false);
+
+	return 0;
+}
+
+int mesh_add_eht_oper_ie(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb)
+{
+	const struct ieee80211_sta_eht_cap *eht_cap;
+	struct ieee80211_supported_band *sband;
+	u32 len;
+	u8 *pos;
+
+	sband = ieee80211_get_sband(sdata);
+	if (!sband)
+		return -EINVAL;
+
+	eht_cap = ieee80211_get_eht_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT);
+	if (!eht_cap ||
+	    sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_20_NOHT ||
+	    sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_5 ||
+	    sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_10)
+		return 0;
+
+	len = 2 + 1 + offsetof(struct ieee80211_eht_operation, optional) +
+		      offsetof(struct ieee80211_eht_operation_info, optional);
+
+	if (skb_tailroom(skb) < len)
+		return -ENOMEM;
+
+	pos = skb_put(skb, len);
+	ieee80211_ie_build_eht_oper(pos, &sdata->vif.bss_conf.chandef, eht_cap);
+
+	return 0;
+}
+
 static void ieee80211_mesh_path_timer(struct timer_list *t)
 {
 	struct ieee80211_sub_if_data *sdata =
@@ -697,6 +756,9 @@ ieee80211_mesh_update_bss_params(struct ieee80211_sub_if_data *sdata,
 	if (he_oper)
 		sdata->vif.bss_conf.he_oper.params =
 			__le32_to_cpu(he_oper->he_oper_params);
+
+	sdata->vif.bss_conf.eht_support =
+		!!ieee80211_get_eht_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT);
 }
 
 bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
@@ -903,7 +965,7 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh)
 	struct ieee80211_chanctx_conf *chanctx_conf;
 	struct mesh_csa_settings *csa;
 	enum nl80211_band band;
-	u8 ie_len_he_cap;
+	u8 ie_len_he_cap, ie_len_eht_cap;
 	u8 *pos;
 	struct ieee80211_sub_if_data *sdata;
 	int hdr_len = offsetofend(struct ieee80211_mgmt, u.beacon);
@@ -916,6 +978,8 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh)
 
 	ie_len_he_cap = ieee80211_ie_len_he_cap(sdata,
 						NL80211_IFTYPE_MESH_POINT);
+	ie_len_eht_cap = ieee80211_ie_len_eht_cap(sdata,
+						  NL80211_IFTYPE_MESH_POINT);
 	head_len = hdr_len +
 		   2 + /* NULL SSID */
 		   /* Channel Switch Announcement */
@@ -939,6 +1003,9 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh)
 		   2 + 1 + sizeof(struct ieee80211_he_operation) +
 			   sizeof(struct ieee80211_he_6ghz_oper) +
 		   2 + 1 + sizeof(struct ieee80211_he_6ghz_capa) +
+		   ie_len_eht_cap +
+		   2 + 1 + offsetof(struct ieee80211_eht_operation, optional) +
+			   offsetof(struct ieee80211_eht_operation_info, optional) +
 		   ifmsh->ie_len;
 
 	bcn = kzalloc(sizeof(*bcn) + head_len + tail_len, GFP_KERNEL);
@@ -1059,6 +1126,8 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh)
 	    mesh_add_he_cap_ie(sdata, skb, ie_len_he_cap) ||
 	    mesh_add_he_oper_ie(sdata, skb) ||
 	    mesh_add_he_6ghz_cap_ie(sdata, skb) ||
+	    mesh_add_eht_cap_ie(sdata, skb, ie_len_eht_cap) ||
+	    mesh_add_eht_oper_ie(sdata, skb) ||
 	    mesh_add_vendor_ies(sdata, skb))
 		goto out_free;
 
diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h
index 13f394e677ae..022f41292a05 100644
--- a/net/mac80211/mesh.h
+++ b/net/mac80211/mesh.h
@@ -234,6 +234,10 @@ int mesh_add_he_oper_ie(struct ieee80211_sub_if_data *sdata,
 			struct sk_buff *skb);
 int mesh_add_he_6ghz_cap_ie(struct ieee80211_sub_if_data *sdata,
 			    struct sk_buff *skb);
+int mesh_add_eht_cap_ie(struct ieee80211_sub_if_data *sdata,
+			struct sk_buff *skb, u8 ie_len);
+int mesh_add_eht_oper_ie(struct ieee80211_sub_if_data *sdata,
+			 struct sk_buff *skb);
 void mesh_rmc_free(struct ieee80211_sub_if_data *sdata);
 int mesh_rmc_init(struct ieee80211_sub_if_data *sdata);
 void ieee80211s_init(void);
diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c
index ddfe5102b9a4..8f168bc4e4b8 100644
--- a/net/mac80211/mesh_plink.c
+++ b/net/mac80211/mesh_plink.c
@@ -219,12 +219,14 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata,
 	bool include_plid = false;
 	u16 peering_proto = 0;
 	u8 *pos, ie_len = 4;
-	u8 ie_len_he_cap;
+	u8 ie_len_he_cap, ie_len_eht_cap;
 	int hdr_len = offsetofend(struct ieee80211_mgmt, u.action.u.self_prot);
 	int err = -ENOMEM;
 
 	ie_len_he_cap = ieee80211_ie_len_he_cap(sdata,
 						NL80211_IFTYPE_MESH_POINT);
+	ie_len_eht_cap = ieee80211_ie_len_eht_cap(sdata,
+						  NL80211_IFTYPE_MESH_POINT);
 	skb = dev_alloc_skb(local->tx_headroom +
 			    hdr_len +
 			    2 + /* capability info */
@@ -241,6 +243,9 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata,
 			    2 + 1 + sizeof(struct ieee80211_he_operation) +
 				    sizeof(struct ieee80211_he_6ghz_oper) +
 			    2 + 1 + sizeof(struct ieee80211_he_6ghz_capa) +
+			    ie_len_eht_cap +
+			    2 + 1 + offsetof(struct ieee80211_eht_operation, optional) +
+				    offsetof(struct ieee80211_eht_operation_info, optional) +
 			    2 + 8 + /* peering IE */
 			    sdata->u.mesh.ie_len);
 	if (!skb)
@@ -332,7 +337,9 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata,
 		    mesh_add_vht_oper_ie(sdata, skb) ||
 		    mesh_add_he_cap_ie(sdata, skb, ie_len_he_cap) ||
 		    mesh_add_he_oper_ie(sdata, skb) ||
-		    mesh_add_he_6ghz_cap_ie(sdata, skb))
+		    mesh_add_he_6ghz_cap_ie(sdata, skb) ||
+		    mesh_add_eht_cap_ie(sdata, skb, ie_len_eht_cap) ||
+		    mesh_add_eht_oper_ie(sdata, skb))
 			goto free;
 	}
 
@@ -451,6 +458,11 @@ static void mesh_sta_info_init(struct ieee80211_sub_if_data *sdata,
 					  elems->he_6ghz_capa,
 					  &sta->deflink);
 
+	ieee80211_eht_cap_ie_to_sta_eht_cap(sdata, sband, elems->he_cap,
+					    elems->he_cap_len,
+					    elems->eht_cap, elems->eht_cap_len,
+					    &sta->deflink);
+
 	if (bw != sta->sta.deflink.bandwidth)
 		changed |= IEEE80211_RC_BW_CHANGED;
 
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 9f5cc16538f3..1527d6aafc14 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3485,6 +3485,77 @@ out:
 	return pos;
 }
 
+u8 *ieee80211_ie_build_eht_oper(u8 *pos, struct cfg80211_chan_def *chandef,
+				const struct ieee80211_sta_eht_cap *eht_cap)
+
+{
+	const struct ieee80211_eht_mcs_nss_supp_20mhz_only *eht_mcs_nss =
+					&eht_cap->eht_mcs_nss_supp.only_20mhz;
+	struct ieee80211_eht_operation *eht_oper;
+	struct ieee80211_eht_operation_info *eht_oper_info;
+	u8 eht_oper_len = offsetof(struct ieee80211_eht_operation, optional);
+	u8 eht_oper_info_len =
+		offsetof(struct ieee80211_eht_operation_info, optional);
+	u8 chan_width = 0;
+
+	*pos++ = WLAN_EID_EXTENSION;
+	*pos++ = 1 + eht_oper_len + eht_oper_info_len;
+	*pos++ = WLAN_EID_EXT_EHT_OPERATION;
+
+	eht_oper = (struct ieee80211_eht_operation *)pos;
+
+	memcpy(&eht_oper->basic_mcs_nss, eht_mcs_nss, sizeof(*eht_mcs_nss));
+	eht_oper->params |= IEEE80211_EHT_OPER_INFO_PRESENT;
+	pos += eht_oper_len;
+
+	eht_oper_info =
+		(struct ieee80211_eht_operation_info *)eht_oper->optional;
+
+	eht_oper_info->ccfs0 =
+		ieee80211_frequency_to_channel(chandef->center_freq1);
+	if (chandef->center_freq2)
+		eht_oper_info->ccfs1 =
+			ieee80211_frequency_to_channel(chandef->center_freq2);
+	else
+		eht_oper_info->ccfs1 = 0;
+
+	switch (chandef->width) {
+	case NL80211_CHAN_WIDTH_320:
+		chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_320MHZ;
+		eht_oper_info->ccfs1 = eht_oper_info->ccfs0;
+		if (chandef->chan->center_freq < chandef->center_freq1)
+			eht_oper_info->ccfs0 -= 16;
+		else
+			eht_oper_info->ccfs0 += 16;
+		break;
+	case NL80211_CHAN_WIDTH_160:
+		eht_oper_info->ccfs1 = eht_oper_info->ccfs0;
+		if (chandef->chan->center_freq < chandef->center_freq1)
+			eht_oper_info->ccfs0 -= 8;
+		else
+			eht_oper_info->ccfs0 += 8;
+		fallthrough;
+	case NL80211_CHAN_WIDTH_80P80:
+		chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_160MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_80MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_40:
+		chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_40MHZ;
+		break;
+	default:
+		chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_20MHZ;
+		break;
+	}
+	eht_oper_info->control = chan_width;
+	pos += eht_oper_info_len;
+
+	/* TODO: eht_oper_info->optional */
+
+	return pos;
+}
+
 bool ieee80211_chandef_ht_oper(const struct ieee80211_ht_operation *ht_oper,
 			       struct cfg80211_chan_def *chandef)
 {
-- 
cgit v1.2.3-70-g09d2


From c7e39d70613c113b5ac667beeb3cacb242c6a4cb Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Sun, 19 Mar 2023 09:54:18 -0400
Subject: wifi: ipw2x00: remove unused _ipw_read16 function

clang with W=1 reports
drivers/net/wireless/intel/ipw2x00/ipw2200.c:381:19: error:
  unused function '_ipw_read16' [-Werror,-Wunused-function]
static inline u16 _ipw_read16(struct ipw_priv *ipw, unsigned long ofs)
                  ^
This function and its wrapping marco are not used, so remove them.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230319135418.1703380-1-trix@redhat.com
---
 drivers/net/wireless/intel/ipw2x00/ipw2200.c | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2200.c b/drivers/net/wireless/intel/ipw2x00/ipw2200.c
index b91b1a2d0be7..dfe0f74369e6 100644
--- a/drivers/net/wireless/intel/ipw2x00/ipw2200.c
+++ b/drivers/net/wireless/intel/ipw2x00/ipw2200.c
@@ -377,19 +377,6 @@ static inline u8 _ipw_read8(struct ipw_priv *ipw, unsigned long ofs)
 	_ipw_read8(ipw, ofs); \
 })
 
-/* 16-bit direct read (low 4K) */
-static inline u16 _ipw_read16(struct ipw_priv *ipw, unsigned long ofs)
-{
-	return readw(ipw->hw_base + ofs);
-}
-
-/* alias to 16-bit direct read (low 4K of SRAM/regs), with debug wrapper */
-#define ipw_read16(ipw, ofs) ({ \
-	IPW_DEBUG_IO("%s %d: read_direct16(0x%08X)\n", __FILE__, __LINE__, \
-			(u32)(ofs)); \
-	_ipw_read16(ipw, ofs); \
-})
-
 /* 32-bit direct read (low 4K) */
 static inline u32 _ipw_read32(struct ipw_priv *ipw, unsigned long ofs)
 {
-- 
cgit v1.2.3-70-g09d2


From 8de7838acfa3309fcc4c6160cb230ee79f582b18 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Mon, 20 Mar 2023 17:13:42 +0100
Subject: wifi: rsi: Slightly simplify rsi_set_channel()

There is no point in allocating 'skb' and then freeing it if !channel.

Make the sanity check first to slightly simplify the code.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/29bf0296bd939e3f6952272bfdcc73b22edbc374.1679328588.git.christophe.jaillet@wanadoo.fr
---
 drivers/net/wireless/rsi/rsi_91x_mgmt.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/rsi/rsi_91x_mgmt.c b/drivers/net/wireless/rsi/rsi_91x_mgmt.c
index 1b309e47a1f1..7f2c1608f2ce 100644
--- a/drivers/net/wireless/rsi/rsi_91x_mgmt.c
+++ b/drivers/net/wireless/rsi/rsi_91x_mgmt.c
@@ -1127,6 +1127,9 @@ int rsi_set_channel(struct rsi_common *common,
 	rsi_dbg(MGMT_TX_ZONE,
 		"%s: Sending scan req frame\n", __func__);
 
+	if (!channel)
+		return 0;
+
 	skb = dev_alloc_skb(frame_len);
 	if (!skb) {
 		rsi_dbg(ERR_ZONE, "%s: Failed in allocation of skb\n",
@@ -1134,10 +1137,6 @@ int rsi_set_channel(struct rsi_common *common,
 		return -ENOMEM;
 	}
 
-	if (!channel) {
-		dev_kfree_skb(skb);
-		return 0;
-	}
 	memset(skb->data, 0, frame_len);
 	chan_cfg = (struct rsi_chan_config *)skb->data;
 
-- 
cgit v1.2.3-70-g09d2


From c9b6111a6f94bd186d1f0f7a04a9b37030a27738 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Mon, 20 Mar 2023 19:34:48 -0400
Subject: wifi: rtw88: remove unused rtw_pci_get_tx_desc function

clang with W=1 reports
drivers/net/wireless/realtek/rtw88/pci.c:92:21: error:
  unused function 'rtw_pci_get_tx_desc' [-Werror,-Wunused-function]
static inline void *rtw_pci_get_tx_desc(struct rtw_pci_tx_ring *tx_ring, u8 idx)
                    ^
This function is not used, so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320233448.1729899-1-trix@redhat.com
---
 drivers/net/wireless/realtek/rtw88/pci.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index b4bd831c9845..6a8e6ee82069 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -89,13 +89,6 @@ static void rtw_pci_write32(struct rtw_dev *rtwdev, u32 addr, u32 val)
 	writel(val, rtwpci->mmap + addr);
 }
 
-static inline void *rtw_pci_get_tx_desc(struct rtw_pci_tx_ring *tx_ring, u8 idx)
-{
-	int offset = tx_ring->r.desc_size * idx;
-
-	return tx_ring->r.head + offset;
-}
-
 static void rtw_pci_free_tx_ring_skbs(struct rtw_dev *rtwdev,
 				      struct rtw_pci_tx_ring *tx_ring)
 {
-- 
cgit v1.2.3-70-g09d2


From 703f15b6dce1e3d38b2455b10c660fafd22e5528 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Wed, 22 Mar 2023 12:07:31 +0300
Subject: wifi: rndis_wlan: clean up a type issue

The rndis_bss_info_update() returns type bool.  Change the NULL
returns to false.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/d1f250a3-653a-497d-9c92-12dd628b7927@kili.mountain
---
 drivers/net/wireless/legacy/rndis_wlan.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/legacy/rndis_wlan.c b/drivers/net/wireless/legacy/rndis_wlan.c
index bf72e5fd39cf..cc361d35a90f 100644
--- a/drivers/net/wireless/legacy/rndis_wlan.c
+++ b/drivers/net/wireless/legacy/rndis_wlan.c
@@ -1972,7 +1972,7 @@ static bool rndis_bss_info_update(struct usbnet *usbdev,
 
 	if (bssid_len < sizeof(struct ndis_80211_bssid_ex) +
 			sizeof(struct ndis_80211_fixed_ies))
-		return NULL;
+		return false;
 
 	fixed = (struct ndis_80211_fixed_ies *)bssid->ies;
 
@@ -1981,13 +1981,13 @@ static bool rndis_bss_info_update(struct usbnet *usbdev,
 					(int)le32_to_cpu(bssid->ie_length));
 	ie_len -= sizeof(struct ndis_80211_fixed_ies);
 	if (ie_len < 0)
-		return NULL;
+		return false;
 
 	/* extract data for cfg80211_inform_bss */
 	channel = ieee80211_get_channel(priv->wdev.wiphy,
 			KHZ_TO_MHZ(le32_to_cpu(bssid->config.ds_config)));
 	if (!channel)
-		return NULL;
+		return false;
 
 	signal = level_to_qual(le32_to_cpu(bssid->rssi));
 	timestamp = le64_to_cpu(*(__le64 *)fixed->timestamp);
-- 
cgit v1.2.3-70-g09d2


From 06dabcccc08b6323c56062c7fdbb9dba88168565 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Wed, 22 Mar 2023 12:26:53 -0600
Subject: wifi: rndis_wlan: Replace fake flex-array with flexible-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Address the following warning found with GCC-13 and
-fstrict-flex-array=3 enabled:
drivers/net/wireless/rndis_wlan.c:2902:23: warning: array subscript 0 is outside array bounds of ‘struct ndis_80211_auth_request[0]’ [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/274
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/ZBtIbU77L9eXqa4j@work
---
 drivers/net/wireless/legacy/rndis_wlan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/legacy/rndis_wlan.c b/drivers/net/wireless/legacy/rndis_wlan.c
index cc361d35a90f..712038d46bdb 100644
--- a/drivers/net/wireless/legacy/rndis_wlan.c
+++ b/drivers/net/wireless/legacy/rndis_wlan.c
@@ -209,7 +209,7 @@ struct ndis_80211_status_indication {
 	union {
 		__le32					media_stream_mode;
 		__le32					radio_status;
-		struct ndis_80211_auth_request		auth_request[0];
+		DECLARE_FLEX_ARRAY(struct ndis_80211_auth_request, auth_request);
 		struct ndis_80211_pmkid_cand_list	cand_list;
 	} u;
 } __packed;
-- 
cgit v1.2.3-70-g09d2


From e665c6d67e541ac7520fa8c5175ad2c0213f247e Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:26:12 +0800
Subject: wifi: rtw88: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230323112613.7550-4-cai.huoqing@linux.dev
---
 drivers/net/wireless/realtek/rtw88/pci.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index 6a8e6ee82069..672ddde80816 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -1545,7 +1545,6 @@ static int rtw_pci_claim(struct rtw_dev *rtwdev, struct pci_dev *pdev)
 
 static void rtw_pci_declaim(struct rtw_dev *rtwdev, struct pci_dev *pdev)
 {
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 82d6077901c673546490daa0c1715d6662b06bcf Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Thu, 23 Mar 2023 19:11:35 -0600
Subject: wifi: rtlwifi: Replace fake flex-array with flex-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Address the following warning found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
In function ‘fortify_memset_chk’,
    inlined from ‘rtl_usb_probe’ at drivers/net/wireless/realtek/rtlwifi/usb.c:1044:2:
./include/linux/fortify-string.h:430:25: warning: call to ‘__write_overflow_field’ declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning]
  430 |                         __write_overflow_field(p_size_field, size);
      |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/277
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/ZBz4x+MWoI/f65o1@work
---
 drivers/net/wireless/realtek/rtlwifi/wifi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index 31f9e9e5c680..082af216760f 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -2831,7 +2831,7 @@ struct rtl_priv {
 	 * beyond  this structure like:
 	 * rtl_pci_priv or rtl_usb_priv
 	 */
-	u8 priv[0] __aligned(sizeof(void *));
+	u8 priv[] __aligned(sizeof(void *));
 };
 
 #define rtl_priv(hw)		(((struct rtl_priv *)(hw)->priv))
-- 
cgit v1.2.3-70-g09d2


From 4c7f8c237d322a98026a57f6ba846fc997f165df Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Fri, 24 Mar 2023 09:50:22 -0400
Subject: wifi: b43legacy: remove unused freq_r3A_value function

clang with W=1 reports
drivers/net/wireless/broadcom/b43legacy/radio.c:1713:5: error:
  unused function 'freq_r3A_value' [-Werror,-Wunused-function]
u16 freq_r3A_value(u16 frequency)
    ^
This function is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230324135022.2649735-1-trix@redhat.com
---
 drivers/net/wireless/broadcom/b43legacy/radio.c | 17 -----------------
 1 file changed, 17 deletions(-)

diff --git a/drivers/net/wireless/broadcom/b43legacy/radio.c b/drivers/net/wireless/broadcom/b43legacy/radio.c
index fdf78c10a05c..8d7eb89c1628 100644
--- a/drivers/net/wireless/broadcom/b43legacy/radio.c
+++ b/drivers/net/wireless/broadcom/b43legacy/radio.c
@@ -1709,23 +1709,6 @@ u16 b43legacy_radio_init2050(struct b43legacy_wldev *dev)
 	return ret;
 }
 
-static inline
-u16 freq_r3A_value(u16 frequency)
-{
-	u16 value;
-
-	if (frequency < 5091)
-		value = 0x0040;
-	else if (frequency < 5321)
-		value = 0x0000;
-	else if (frequency < 5806)
-		value = 0x0080;
-	else
-		value = 0x0040;
-
-	return value;
-}
-
 int b43legacy_radio_selectchannel(struct b43legacy_wldev *dev,
 				  u8 channel,
 				  int synthetic_pu_workaround)
-- 
cgit v1.2.3-70-g09d2


From 5aeb763a27c2b444928639d35cbe94520c405969 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Sat, 25 Mar 2023 09:03:43 -0400
Subject: wifi: brcmsmac: remove unused has_5g variable

clang with W=1 reports
drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c:1051:6: error:
  variable 'has_5g' set but not used [-Werror,-Wunused-but-set-variable]
        int has_5g = 0;
            ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230325130343.1334209-1-trix@redhat.com
---
 drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
index a8333e6adbda..0bd4e679a359 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
@@ -1048,7 +1048,6 @@ static int ieee_hw_rate_init(struct ieee80211_hw *hw)
 	struct brcms_info *wl = hw->priv;
 	struct brcms_c_info *wlc = wl->wlc;
 	struct ieee80211_supported_band *band;
-	int has_5g = 0;
 	u16 phy_type;
 
 	hw->wiphy->bands[NL80211_BAND_2GHZ] = NULL;
@@ -1070,7 +1069,6 @@ static int ieee_hw_rate_init(struct ieee80211_hw *hw)
 
 	/* Assume all bands use the same phy.  True for 11n devices. */
 	if (wl->pub->_nbands > 1) {
-		has_5g++;
 		if (phy_type == PHY_TYPE_N || phy_type == PHY_TYPE_LCN) {
 			band = &wlc->bandstate[BAND_5G_INDEX]->band;
 			*band = brcms_band_5GHz_nphy_template;
-- 
cgit v1.2.3-70-g09d2


From 905a9241e4e8c15d2c084fee916280514848fe35 Mon Sep 17 00:00:00 2001
From: Wei Chen <harperchen1110@gmail.com>
Date: Sun, 26 Mar 2023 05:31:38 +0000
Subject: wifi: rtlwifi: fix incorrect error codes in
 rtl_debugfs_set_write_rfreg()

If there is a failure during copy_from_user or user-provided data buffer
is invalid, rtl_debugfs_set_write_rfreg should return negative error code
instead of a positive value count.

Fix this bug by returning correct error code. Moreover, the check of buffer
against null is removed since it will be handled by copy_from_user.

Fixes: 610247f46feb ("rtlwifi: Improve debugging by using debugfs")
Signed-off-by: Wei Chen <harperchen1110@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230326053138.91338-1-harperchen1110@gmail.com
---
 drivers/net/wireless/realtek/rtlwifi/debug.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/debug.c b/drivers/net/wireless/realtek/rtlwifi/debug.c
index 0b1bc04cb6ad..602717928887 100644
--- a/drivers/net/wireless/realtek/rtlwifi/debug.c
+++ b/drivers/net/wireless/realtek/rtlwifi/debug.c
@@ -375,8 +375,8 @@ static ssize_t rtl_debugfs_set_write_rfreg(struct file *filp,
 
 	tmp_len = (count > sizeof(tmp) - 1 ? sizeof(tmp) - 1 : count);
 
-	if (!buffer || copy_from_user(tmp, buffer, tmp_len))
-		return count;
+	if (copy_from_user(tmp, buffer, tmp_len))
+		return -EFAULT;
 
 	tmp[tmp_len] = '\0';
 
@@ -386,7 +386,7 @@ static ssize_t rtl_debugfs_set_write_rfreg(struct file *filp,
 	if (num != 4) {
 		rtl_dbg(rtlpriv, COMP_ERR, DBG_DMESG,
 			"Format is <path> <addr> <mask> <data>\n");
-		return count;
+		return -EINVAL;
 	}
 
 	rtl_set_rfreg(hw, path, addr, bitmask, data);
-- 
cgit v1.2.3-70-g09d2


From 5dbe1f8eb8c5ac69394400a5b86fd81775e96c43 Mon Sep 17 00:00:00 2001
From: Wei Chen <harperchen1110@gmail.com>
Date: Sun, 26 Mar 2023 05:42:17 +0000
Subject: wifi: rtlwifi: fix incorrect error codes in
 rtl_debugfs_set_write_reg()

If there is a failure during copy_from_user or user-provided data buffer is
invalid, rtl_debugfs_set_write_reg should return negative error code instead
of a positive value count.

Fix this bug by returning correct error code. Moreover, the check of buffer
against null is removed since it will be handled by copy_from_user.

Fixes: 610247f46feb ("rtlwifi: Improve debugging by using debugfs")
Signed-off-by: Wei Chen <harperchen1110@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230326054217.93492-1-harperchen1110@gmail.com
---
 drivers/net/wireless/realtek/rtlwifi/debug.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/debug.c b/drivers/net/wireless/realtek/rtlwifi/debug.c
index 602717928887..9eb26dfe4ca9 100644
--- a/drivers/net/wireless/realtek/rtlwifi/debug.c
+++ b/drivers/net/wireless/realtek/rtlwifi/debug.c
@@ -278,8 +278,8 @@ static ssize_t rtl_debugfs_set_write_reg(struct file *filp,
 
 	tmp_len = (count > sizeof(tmp) - 1 ? sizeof(tmp) - 1 : count);
 
-	if (!buffer || copy_from_user(tmp, buffer, tmp_len))
-		return count;
+	if (copy_from_user(tmp, buffer, tmp_len))
+		return -EFAULT;
 
 	tmp[tmp_len] = '\0';
 
@@ -287,7 +287,7 @@ static ssize_t rtl_debugfs_set_write_reg(struct file *filp,
 	num = sscanf(tmp, "%x %x %x", &addr, &val, &len);
 
 	if (num !=  3)
-		return count;
+		return -EINVAL;
 
 	switch (len) {
 	case 1:
-- 
cgit v1.2.3-70-g09d2


From 2f73f04b7f93f031c94a5c473dd9b7f1cdf91bc0 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Mon, 27 Mar 2023 11:11:51 -0400
Subject: wifi: brcmsmac: ampdu: remove unused suc_mpdu variable

clang with W=1 reports
drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c:848:5: error: variable
  'suc_mpdu' set but not used [-Werror,-Wunused-but-set-variable]
        u8 suc_mpdu = 0, tot_mpdu = 0;
           ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230327151151.1771350-1-trix@redhat.com
---
 drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c
index 2631eb7569eb..e24228e60027 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c
@@ -845,7 +845,7 @@ brcms_c_ampdu_dotxstatus_complete(struct ampdu_info *ampdu, struct scb *scb,
 	u16 seq, start_seq = 0, bindex, index, mcl;
 	u8 mcs = 0;
 	bool ba_recd = false, ack_recd = false;
-	u8 suc_mpdu = 0, tot_mpdu = 0;
+	u8 tot_mpdu = 0;
 	uint supr_status;
 	bool retry = true;
 	u16 mimoantsel = 0;
@@ -975,7 +975,6 @@ brcms_c_ampdu_dotxstatus_complete(struct ampdu_info *ampdu, struct scb *scb,
 				ieee80211_tx_status_irqsafe(wlc->pub->ieee_hw,
 							    p);
 				ack_recd = true;
-				suc_mpdu++;
 			}
 		}
 		/* either retransmit or send bar if ack not recd */
-- 
cgit v1.2.3-70-g09d2


From 074d0a1ae1fefce3dfe73bdda855e62eb8eb2a78 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Wed, 29 Mar 2023 09:14:44 -0400
Subject: wifi: mwifiex: remove unused evt_buf variable

clang with W=1 reports
drivers/net/wireless/marvell/mwifiex/11h.c:198:6: error: variable
  'evt_buf' set but not used [-Werror,-Wunused-but-set-variable]
        u8 *evt_buf;
            ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230329131444.1809018-1-trix@redhat.com
---
 drivers/net/wireless/marvell/mwifiex/11h.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/wireless/marvell/mwifiex/11h.c b/drivers/net/wireless/marvell/mwifiex/11h.c
index b0c40a776a2e..2ea03725f188 100644
--- a/drivers/net/wireless/marvell/mwifiex/11h.c
+++ b/drivers/net/wireless/marvell/mwifiex/11h.c
@@ -195,7 +195,6 @@ int mwifiex_11h_handle_chanrpt_ready(struct mwifiex_private *priv,
 {
 	struct host_cmd_ds_chan_rpt_event *rpt_event;
 	struct mwifiex_ie_types_chan_rpt_data *rpt;
-	u8 *evt_buf;
 	u16 event_len, tlv_len;
 
 	rpt_event = (void *)(skb->data + sizeof(u32));
@@ -208,8 +207,6 @@ int mwifiex_11h_handle_chanrpt_ready(struct mwifiex_private *priv,
 		return -1;
 	}
 
-	evt_buf = (void *)&rpt_event->tlvbuf;
-
 	while (event_len >= sizeof(struct mwifiex_ie_types_header)) {
 		rpt = (void *)&rpt_event->tlvbuf;
 		tlv_len = le16_to_cpu(rpt->header.len);
@@ -231,7 +228,6 @@ int mwifiex_11h_handle_chanrpt_ready(struct mwifiex_private *priv,
 			break;
 		}
 
-		evt_buf += (tlv_len + sizeof(rpt->header));
 		event_len -= (tlv_len + sizeof(rpt->header));
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 767d011dc39dc504b17a373ecba4966aef2b0205 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Sun, 19 Mar 2023 11:31:59 -0500
Subject: bcma: Use of_address_to_resource()

Replace of_get_address() and of_translate_address() calls with single
call to of_address_to_resource().

Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230319163159.225773-1-robh@kernel.org
---
 drivers/bcma/main.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/bcma/main.c b/drivers/bcma/main.c
index 7b39f010bbb3..5e438f74ee4c 100644
--- a/drivers/bcma/main.c
+++ b/drivers/bcma/main.c
@@ -140,17 +140,17 @@ static struct device_node *bcma_of_find_child_device(struct device *parent,
 						     struct bcma_device *core)
 {
 	struct device_node *node;
-	u64 size;
-	const __be32 *reg;
+	int ret;
 
 	if (!parent->of_node)
 		return NULL;
 
 	for_each_child_of_node(parent->of_node, node) {
-		reg = of_get_address(node, 0, &size, NULL);
-		if (!reg)
+		struct resource res;
+		ret = of_address_to_resource(node, 0, &res);
+		if (ret)
 			continue;
-		if (of_translate_address(node, reg) == core->addr)
+		if (res.start == core->addr)
 			return node;
 	}
 	return NULL;
-- 
cgit v1.2.3-70-g09d2


From 21898a40b2f84deb9438c2ef63f7a5072d6ab2af Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Fri, 24 Mar 2023 15:06:47 -0400
Subject: bcma: remove unused mips_read32 function

clang with W=1 reports
drivers/bcma/driver_mips.c:49:19: error: unused function
  'mips_read32' [-Werror,-Wunused-function]
static inline u32 mips_read32(struct bcma_drv_mips *mcore,
                  ^
This function is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230324190647.2654293-1-trix@redhat.com
---
 drivers/bcma/driver_mips.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c
index 4f01e6b17bb9..9be0806eb033 100644
--- a/drivers/bcma/driver_mips.c
+++ b/drivers/bcma/driver_mips.c
@@ -46,12 +46,6 @@ static inline bool bcma_core_mips_bcm5357b0_quirk(struct bcma_device *dev)
 	       dev->id.id == BCMA_CORE_USB20_HOST;
 }
 
-static inline u32 mips_read32(struct bcma_drv_mips *mcore,
-			      u16 offset)
-{
-	return bcma_read32(mcore->core, offset);
-}
-
 static u32 bcma_core_mips_irqflag(struct bcma_device *dev)
 {
 	u32 flag;
-- 
cgit v1.2.3-70-g09d2


From 328bafc9a373da5f268d82533dd8f2e66526d168 Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Fri, 31 Mar 2023 11:31:55 +0200
Subject: selftests/bpf: Add err.h header

Moving error macros from profiler.inc.h to new err.h header.
It will be used in following changes.

Also adding PTR_ERR macro that will be used in following changes.

Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20230331093157.1749137-2-jolsa@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/err.h          | 18 ++++++++++++++++++
 tools/testing/selftests/bpf/progs/profiler.inc.h |  3 +--
 2 files changed, 19 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/err.h

diff --git a/tools/testing/selftests/bpf/progs/err.h b/tools/testing/selftests/bpf/progs/err.h
new file mode 100644
index 000000000000..d66d283d9e59
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/err.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ERR_H__
+#define __ERR_H__
+
+#define MAX_ERRNO 4095
+#define IS_ERR_VALUE(x) (unsigned long)(void *)(x) >= (unsigned long)-MAX_ERRNO
+
+static inline int IS_ERR_OR_NULL(const void *ptr)
+{
+	return !ptr || IS_ERR_VALUE((unsigned long)ptr);
+}
+
+static inline long PTR_ERR(const void *ptr)
+{
+	return (long) ptr;
+}
+
+#endif /* __ERR_H__ */
diff --git a/tools/testing/selftests/bpf/progs/profiler.inc.h b/tools/testing/selftests/bpf/progs/profiler.inc.h
index 875513866032..f799d87e8700 100644
--- a/tools/testing/selftests/bpf/progs/profiler.inc.h
+++ b/tools/testing/selftests/bpf/progs/profiler.inc.h
@@ -6,6 +6,7 @@
 #include <bpf/bpf_tracing.h>
 
 #include "profiler.h"
+#include "err.h"
 
 #ifndef NULL
 #define NULL 0
@@ -16,7 +17,6 @@
 #define O_DIRECTORY 00200000
 #define __O_TMPFILE 020000000
 #define O_TMPFILE (__O_TMPFILE | O_DIRECTORY)
-#define MAX_ERRNO 4095
 #define S_IFMT 00170000
 #define S_IFSOCK 0140000
 #define S_IFLNK 0120000
@@ -34,7 +34,6 @@
 #define S_ISBLK(m) (((m)&S_IFMT) == S_IFBLK)
 #define S_ISFIFO(m) (((m)&S_IFMT) == S_IFIFO)
 #define S_ISSOCK(m) (((m)&S_IFMT) == S_IFSOCK)
-#define IS_ERR_VALUE(x) (unsigned long)(void*)(x) >= (unsigned long)-MAX_ERRNO
 
 #define KILL_DATA_ARRAY_SIZE 8
 
-- 
cgit v1.2.3-70-g09d2


From 88dc8b3605b38a440fba45edcc53a6c7a98eee3b Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Fri, 31 Mar 2023 11:31:56 +0200
Subject: selftests/bpf: Add read_build_id function

Adding read_build_id function that parses out build id from
specified binary.

It will replace extract_build_id and also be used in following
changes.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20230331093157.1749137-3-jolsa@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/trace_helpers.c | 82 +++++++++++++++++++++++++++++
 tools/testing/selftests/bpf/trace_helpers.h |  5 ++
 2 files changed, 87 insertions(+)

diff --git a/tools/testing/selftests/bpf/trace_helpers.c b/tools/testing/selftests/bpf/trace_helpers.c
index 934bf28fc888..9b070cdf44ac 100644
--- a/tools/testing/selftests/bpf/trace_helpers.c
+++ b/tools/testing/selftests/bpf/trace_helpers.c
@@ -11,6 +11,9 @@
 #include <linux/perf_event.h>
 #include <sys/mman.h>
 #include "trace_helpers.h"
+#include <linux/limits.h>
+#include <libelf.h>
+#include <gelf.h>
 
 #define TRACEFS_PIPE	"/sys/kernel/tracing/trace_pipe"
 #define DEBUGFS_PIPE	"/sys/kernel/debug/tracing/trace_pipe"
@@ -234,3 +237,82 @@ ssize_t get_rel_offset(uintptr_t addr)
 	fclose(f);
 	return -EINVAL;
 }
+
+static int
+parse_build_id_buf(const void *note_start, Elf32_Word note_size, char *build_id)
+{
+	Elf32_Word note_offs = 0;
+
+	while (note_offs + sizeof(Elf32_Nhdr) < note_size) {
+		Elf32_Nhdr *nhdr = (Elf32_Nhdr *)(note_start + note_offs);
+
+		if (nhdr->n_type == 3 && nhdr->n_namesz == sizeof("GNU") &&
+		    !strcmp((char *)(nhdr + 1), "GNU") && nhdr->n_descsz > 0 &&
+		    nhdr->n_descsz <= BPF_BUILD_ID_SIZE) {
+			memcpy(build_id, note_start + note_offs +
+			       ALIGN(sizeof("GNU"), 4) + sizeof(Elf32_Nhdr), nhdr->n_descsz);
+			memset(build_id + nhdr->n_descsz, 0, BPF_BUILD_ID_SIZE - nhdr->n_descsz);
+			return (int) nhdr->n_descsz;
+		}
+
+		note_offs = note_offs + sizeof(Elf32_Nhdr) +
+			   ALIGN(nhdr->n_namesz, 4) + ALIGN(nhdr->n_descsz, 4);
+	}
+
+	return -ENOENT;
+}
+
+/* Reads binary from *path* file and returns it in the *build_id* buffer
+ * with *size* which is expected to be at least BPF_BUILD_ID_SIZE bytes.
+ * Returns size of build id on success. On error the error value is
+ * returned.
+ */
+int read_build_id(const char *path, char *build_id, size_t size)
+{
+	int fd, err = -EINVAL;
+	Elf *elf = NULL;
+	GElf_Ehdr ehdr;
+	size_t max, i;
+
+	if (size < BPF_BUILD_ID_SIZE)
+		return -EINVAL;
+
+	fd = open(path, O_RDONLY | O_CLOEXEC);
+	if (fd < 0)
+		return -errno;
+
+	(void)elf_version(EV_CURRENT);
+
+	elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
+	if (!elf)
+		goto out;
+	if (elf_kind(elf) != ELF_K_ELF)
+		goto out;
+	if (!gelf_getehdr(elf, &ehdr))
+		goto out;
+
+	for (i = 0; i < ehdr.e_phnum; i++) {
+		GElf_Phdr mem, *phdr;
+		char *data;
+
+		phdr = gelf_getphdr(elf, i, &mem);
+		if (!phdr)
+			goto out;
+		if (phdr->p_type != PT_NOTE)
+			continue;
+		data = elf_rawfile(elf, &max);
+		if (!data)
+			goto out;
+		if (phdr->p_offset + phdr->p_memsz > max)
+			goto out;
+		err = parse_build_id_buf(data + phdr->p_offset, phdr->p_memsz, build_id);
+		if (err > 0)
+			break;
+	}
+
+out:
+	if (elf)
+		elf_end(elf);
+	close(fd);
+	return err;
+}
diff --git a/tools/testing/selftests/bpf/trace_helpers.h b/tools/testing/selftests/bpf/trace_helpers.h
index 53efde0e2998..876f3e711df6 100644
--- a/tools/testing/selftests/bpf/trace_helpers.h
+++ b/tools/testing/selftests/bpf/trace_helpers.h
@@ -4,6 +4,9 @@
 
 #include <bpf/libbpf.h>
 
+#define __ALIGN_MASK(x, mask)	(((x)+(mask))&~(mask))
+#define ALIGN(x, a)		__ALIGN_MASK(x, (typeof(x))(a)-1)
+
 struct ksym {
 	long addr;
 	char *name;
@@ -23,4 +26,6 @@ void read_trace_pipe(void);
 ssize_t get_uprobe_offset(const void *addr);
 ssize_t get_rel_offset(uintptr_t addr);
 
+int read_build_id(const char *path, char *build_id, size_t size);
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From dcc46f51d770bde625e4845cac42e808b3302b62 Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Fri, 31 Mar 2023 11:31:57 +0200
Subject: selftests/bpf: Replace extract_build_id with read_build_id

Replacing extract_build_id with read_build_id that parses out
build id directly from elf without using readelf tool.

Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20230331093157.1749137-4-jolsa@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/stacktrace_build_id.c | 19 ++++++----------
 .../bpf/prog_tests/stacktrace_build_id_nmi.c       | 17 ++++++---------
 tools/testing/selftests/bpf/test_progs.c           | 25 ----------------------
 tools/testing/selftests/bpf/test_progs.h           |  1 -
 4 files changed, 13 insertions(+), 49 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
index 9ad09a6c538a..b7ba5cd47d96 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
@@ -7,13 +7,12 @@ void test_stacktrace_build_id(void)
 
 	int control_map_fd, stackid_hmap_fd, stackmap_fd, stack_amap_fd;
 	struct test_stacktrace_build_id *skel;
-	int err, stack_trace_len;
+	int err, stack_trace_len, build_id_size;
 	__u32 key, prev_key, val, duration = 0;
-	char buf[256];
-	int i, j;
+	char buf[BPF_BUILD_ID_SIZE];
 	struct bpf_stack_build_id id_offs[PERF_MAX_STACK_DEPTH];
 	int build_id_matches = 0;
-	int retry = 1;
+	int i, retry = 1;
 
 retry:
 	skel = test_stacktrace_build_id__open_and_load();
@@ -52,9 +51,10 @@ retry:
 		  "err %d errno %d\n", err, errno))
 		goto cleanup;
 
-	err = extract_build_id(buf, 256);
+	build_id_size = read_build_id("urandom_read", buf, sizeof(buf));
+	err = build_id_size < 0 ? build_id_size : 0;
 
-	if (CHECK(err, "get build_id with readelf",
+	if (CHECK(err, "read_build_id",
 		  "err %d errno %d\n", err, errno))
 		goto cleanup;
 
@@ -64,8 +64,6 @@ retry:
 		goto cleanup;
 
 	do {
-		char build_id[64];
-
 		err = bpf_map_lookup_elem(stackmap_fd, &key, id_offs);
 		if (CHECK(err, "lookup_elem from stackmap",
 			  "err %d, errno %d\n", err, errno))
@@ -73,10 +71,7 @@ retry:
 		for (i = 0; i < PERF_MAX_STACK_DEPTH; ++i)
 			if (id_offs[i].status == BPF_STACK_BUILD_ID_VALID &&
 			    id_offs[i].offset != 0) {
-				for (j = 0; j < 20; ++j)
-					sprintf(build_id + 2 * j, "%02x",
-						id_offs[i].build_id[j] & 0xff);
-				if (strstr(buf, build_id) != NULL)
+				if (memcmp(buf, id_offs[i].build_id, build_id_size) == 0)
 					build_id_matches = 1;
 			}
 		prev_key = key;
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
index f4ea1a215ce4..47558b0d7f66 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
@@ -28,11 +28,10 @@ void test_stacktrace_build_id_nmi(void)
 		.config = PERF_COUNT_HW_CPU_CYCLES,
 	};
 	__u32 key, prev_key, val, duration = 0;
-	char buf[256];
-	int i, j;
+	char buf[BPF_BUILD_ID_SIZE];
 	struct bpf_stack_build_id id_offs[PERF_MAX_STACK_DEPTH];
-	int build_id_matches = 0;
-	int retry = 1;
+	int build_id_matches = 0, build_id_size;
+	int i, retry = 1;
 
 	attr.sample_freq = read_perf_max_sample_freq();
 
@@ -94,7 +93,8 @@ retry:
 		  "err %d errno %d\n", err, errno))
 		goto cleanup;
 
-	err = extract_build_id(buf, 256);
+	build_id_size = read_build_id("urandom_read", buf, sizeof(buf));
+	err = build_id_size < 0 ? build_id_size : 0;
 
 	if (CHECK(err, "get build_id with readelf",
 		  "err %d errno %d\n", err, errno))
@@ -106,8 +106,6 @@ retry:
 		goto cleanup;
 
 	do {
-		char build_id[64];
-
 		err = bpf_map__lookup_elem(skel->maps.stackmap, &key, sizeof(key),
 					   id_offs, sizeof(id_offs), 0);
 		if (CHECK(err, "lookup_elem from stackmap",
@@ -116,10 +114,7 @@ retry:
 		for (i = 0; i < PERF_MAX_STACK_DEPTH; ++i)
 			if (id_offs[i].status == BPF_STACK_BUILD_ID_VALID &&
 			    id_offs[i].offset != 0) {
-				for (j = 0; j < 20; ++j)
-					sprintf(build_id + 2 * j, "%02x",
-						id_offs[i].build_id[j] & 0xff);
-				if (strstr(buf, build_id) != NULL)
+				if (memcmp(buf, id_offs[i].build_id, build_id_size) == 0)
 					build_id_matches = 1;
 			}
 		prev_key = key;
diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index d903e6a72a96..ea82921110da 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -629,31 +629,6 @@ out:
 	return err;
 }
 
-int extract_build_id(char *build_id, size_t size)
-{
-	FILE *fp;
-	char *line = NULL;
-	size_t len = 0;
-
-	fp = popen("readelf -n ./urandom_read | grep 'Build ID'", "r");
-	if (fp == NULL)
-		return -1;
-
-	if (getline(&line, &len, fp) == -1)
-		goto err;
-	pclose(fp);
-
-	if (len > size)
-		len = size;
-	memcpy(build_id, line, len);
-	build_id[len] = '\0';
-	free(line);
-	return 0;
-err:
-	pclose(fp);
-	return -1;
-}
-
 static int finit_module(int fd, const char *param_values, int flags)
 {
 	return syscall(__NR_finit_module, fd, param_values, flags);
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 4b06b8347cd4..10ba43250668 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -405,7 +405,6 @@ static inline void *u64_to_ptr(__u64 ptr)
 int bpf_find_map(const char *test, struct bpf_object *obj, const char *name);
 int compare_map_keys(int map1_fd, int map2_fd);
 int compare_stack_ips(int smap_fd, int amap_fd, int stack_trace_len);
-int extract_build_id(char *build_id, size_t size);
 int kern_sync_rcu(void);
 int trigger_module_test_read(int read_sz);
 int trigger_module_test_write(int write_sz);
-- 
cgit v1.2.3-70-g09d2


From 99b3a769cd8ace9e2d38b3c41208a5a261510e70 Mon Sep 17 00:00:00 2001
From: Peng Fan <peng.fan@nxp.com>
Date: Tue, 28 Mar 2023 14:15:18 +0800
Subject: dt-bindings: net: fec: add power-domains property

Add optional power domains property

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20230328061518.1985981-1-peng.fan@oss.nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/fsl,fec.yaml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/fsl,fec.yaml b/Documentation/devicetree/bindings/net/fsl,fec.yaml
index e6f2045f05de..b494e009326e 100644
--- a/Documentation/devicetree/bindings/net/fsl,fec.yaml
+++ b/Documentation/devicetree/bindings/net/fsl,fec.yaml
@@ -144,6 +144,9 @@ properties:
     description:
       Regulator that powers the Ethernet PHY.
 
+  power-domains:
+    maxItems: 1
+
   fsl,num-tx-queues:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-- 
cgit v1.2.3-70-g09d2


From 9af0f555ae4add25f0950753fb410c509aa71f50 Mon Sep 17 00:00:00 2001
From: James Hilliard <james.hilliard1@gmail.com>
Date: Fri, 31 Mar 2023 01:58:42 -0600
Subject: selftests/bpf: Fix conflicts with built-in functions in
 bench_local_storage_create

The fork function in gcc is considered a built in function due to
being used by libgcov when building with gnu extensions.

Rename fork to sched_process_fork to prevent this conflict.

See details:
https://github.com/gcc-mirror/gcc/commit/d1c38823924506d389ca58d02926ace21bdf82fa
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82457

Fixes the following error:

In file included from progs/bench_local_storage_create.c:6:
progs/bench_local_storage_create.c:43:14: error: conflicting types for
built-in function 'fork'; expected 'int(void)'
[-Werror=builtin-declaration-mismatch]
   43 | int BPF_PROG(fork, struct task_struct *parent, struct
task_struct *child)
      |              ^~~~

Fixes: cbe9d93d58b1 ("selftests/bpf: Add bench for task storage creation")
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230331075848.1642814-1-james.hilliard1@gmail.com
---
 tools/testing/selftests/bpf/benchs/bench_local_storage_create.c | 2 +-
 tools/testing/selftests/bpf/progs/bench_local_storage_create.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c b/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
index abb0321d4f34..cff703f90e95 100644
--- a/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
+++ b/tools/testing/selftests/bpf/benchs/bench_local_storage_create.c
@@ -95,7 +95,7 @@ static void setup(void)
 			exit(1);
 		}
 	} else {
-		if (!bpf_program__attach(skel->progs.fork)) {
+		if (!bpf_program__attach(skel->progs.sched_process_fork)) {
 			fprintf(stderr, "Error attaching bpf program\n");
 			exit(1);
 		}
diff --git a/tools/testing/selftests/bpf/progs/bench_local_storage_create.c b/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
index 7c851c9d5e47..e4bfbba6c193 100644
--- a/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
+++ b/tools/testing/selftests/bpf/progs/bench_local_storage_create.c
@@ -40,7 +40,7 @@ int BPF_PROG(kmalloc, unsigned long call_site, const void *ptr,
 }
 
 SEC("tp_btf/sched_process_fork")
-int BPF_PROG(fork, struct task_struct *parent, struct task_struct *child)
+int BPF_PROG(sched_process_fork, struct task_struct *parent, struct task_struct *child)
 {
 	struct storage *stg;
 
-- 
cgit v1.2.3-70-g09d2


From 3ed85ae80283885ef8491d07cdcd7124328bed35 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Fri, 31 Mar 2023 15:24:02 -0700
Subject: veristat: relicense veristat.c as dual GPL-2.0-only or BSD-2-Clause
 licensed

Dual-license veristat.c to dual GPL-2.0-only or BSD-2-Clause license.
This is needed to mirror it to Github to make it convenient for distro
packagers to package veristat as a separate package.

Veristat grew into a useful tool by itself, and there are already
a bunch of users relying on veristat as generic BPF loading and
verification helper tool. So making it easy to packagers by providing
Github mirror just like we do for bpftool and libbpf is the next step to
get veristat into the hands of users.

Apart from few typo fixes, I'm the sole contributor to veristat.c so
far, so no extra Acks should be needed for relicensing.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230331222405.3468634-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index 7888c03ba631..612ca52c6fba 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -1,4 +1,4 @@
-// SPDX-License-Identifier: GPL-2.0
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
 /* Copyright (c) 2022 Meta Platforms, Inc. and affiliates. */
 #define _GNU_SOURCE
 #include <argp.h>
-- 
cgit v1.2.3-70-g09d2


From 71c8c39f517787af19d9f35fe60463f7eec914e1 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Fri, 31 Mar 2023 15:24:03 -0700
Subject: veristat: improve version reporting

For packaging version of the tool is important, so add a simple way to
specify veristat version for upstream mirror at Github.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230331222405.3468634-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index 612ca52c6fba..daac72b76508 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -140,6 +140,7 @@ static struct env {
 	bool quiet;
 	int log_level;
 	enum resfmt out_fmt;
+	bool show_version;
 	bool comparison_mode;
 	bool replay_mode;
 
@@ -176,16 +177,22 @@ static int libbpf_print_fn(enum libbpf_print_level level, const char *format, va
 	return vfprintf(stderr, format, args);
 }
 
-const char *argp_program_version = "veristat";
+#ifndef VERISTAT_VERSION
+#define VERISTAT_VERSION "<kernel>"
+#endif
+
+const char *argp_program_version = "veristat v" VERISTAT_VERSION;
 const char *argp_program_bug_address = "<bpf@vger.kernel.org>";
 const char argp_program_doc[] =
 "veristat    BPF verifier stats collection and comparison tool.\n"
 "\n"
 "USAGE: veristat <obj-file> [<obj-file>...]\n"
-"   OR: veristat -C <baseline.csv> <comparison.csv>\n";
+"   OR: veristat -C <baseline.csv> <comparison.csv>\n"
+"   OR: veristat -R <results.csv>\n";
 
 static const struct argp_option opts[] = {
 	{ NULL, 'h', NULL, OPTION_HIDDEN, "Show the full help" },
+	{ "version", 'V', NULL, 0, "Print version" },
 	{ "verbose", 'v', NULL, 0, "Verbose mode" },
 	{ "log-level", 'l', "LEVEL", 0, "Verifier log level (default 0 for normal mode, 1 for verbose mode)" },
 	{ "debug", 'd', NULL, 0, "Debug mode (turns on libbpf debug logging)" },
@@ -212,6 +219,9 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 	case 'h':
 		argp_state_help(state, stderr, ARGP_HELP_STD_HELP);
 		break;
+	case 'V':
+		env.show_version = true;
+		break;
 	case 'v':
 		env.verbose = true;
 		break;
@@ -1991,6 +2001,11 @@ int main(int argc, char **argv)
 	if (argp_parse(&argp, argc, argv, 0, NULL, NULL))
 		return 1;
 
+	if (env.show_version) {
+		printf("%s\n", argp_program_version);
+		return 0;
+	}
+
 	if (env.verbose && env.quiet) {
 		fprintf(stderr, "Verbose and quiet modes are incompatible, please specify just one or neither!\n\n");
 		argp_help(&argp, stderr, ARGP_HELP_USAGE, "veristat");
-- 
cgit v1.2.3-70-g09d2


From e3b65c0c1a5b8ed06818b7eeb0c44165ea817d52 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Fri, 31 Mar 2023 15:24:04 -0700
Subject: veristat: avoid using kernel-internal headers

Drop linux/compiler.h include, which seems to be needed for ARRAY_SIZE
macro only. Redefine own version of ARRAY_SIZE instead.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230331222405.3468634-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index daac72b76508..e592d05bccb2 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -4,7 +4,6 @@
 #include <argp.h>
 #include <string.h>
 #include <stdlib.h>
-#include <linux/compiler.h>
 #include <sched.h>
 #include <pthread.h>
 #include <dirent.h>
@@ -20,6 +19,10 @@
 #include <gelf.h>
 #include <float.h>
 
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]))
+#endif
+
 enum stat_id {
 	VERDICT,
 	DURATION,
-- 
cgit v1.2.3-70-g09d2


From ebf390c9d0136e01f327439c012ab5741971e72d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Fri, 31 Mar 2023 15:24:05 -0700
Subject: veristat: small fixed found in -O2 mode

Fix few potentially unitialized variables uses, found while building
veristat.c in release (-O2) mode.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230331222405.3468634-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/veristat.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index e592d05bccb2..53d7ec168268 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -810,7 +810,7 @@ static int guess_prog_type_by_ctx_name(const char *ctx_name,
 		enum bpf_prog_type prog_type;
 		enum bpf_attach_type attach_type;
 	} ctx_map[] = {
-		/* __sk_buff is most ambiguous, for now we assume cgroup_skb */
+		/* __sk_buff is most ambiguous, we assume TC program */
 		{ "__sk_buff", "sk_buff", BPF_PROG_TYPE_SCHED_CLS },
 		{ "bpf_sock", "sock", BPF_PROG_TYPE_CGROUP_SOCK, BPF_CGROUP_INET4_POST_BIND },
 		{ "bpf_sock_addr", "bpf_sock_addr_kern",  BPF_PROG_TYPE_CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_BIND },
@@ -1045,6 +1045,7 @@ static int process_obj(const char *filename)
 			goto cleanup;
 		}
 
+		lprog = NULL;
 		bpf_object__for_each_program(tprog, tobj) {
 			const char *tprog_name = bpf_program__name(tprog);
 
@@ -1855,6 +1856,7 @@ static int handle_comparison_mode(void)
 one_more_time:
 	output_comp_headers(cur_fmt);
 
+	last_idx = -1;
 	for (i = 0; i < env.join_stat_cnt; i++) {
 		const struct verif_stats_join *join = &env.join_stats[i];
 
-- 
cgit v1.2.3-70-g09d2


From d02c48fa113953aba0b330ec6c35f50c7d1d7986 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Fri, 31 Mar 2023 14:57:31 -0500
Subject: bpf: Make struct task_struct an RCU-safe type

struct task_struct objects are a bit interesting in terms of how their
lifetime is protected by refcounts. task structs have two refcount
fields:

1. refcount_t usage: Protects the memory backing the task struct. When
   this refcount drops to 0, the task is immediately freed, without
   waiting for an RCU grace period to elapse. This is the field that
   most callers in the kernel currently use to ensure that a task
   remains valid while it's being referenced, and is what's currently
   tracked with bpf_task_acquire() and bpf_task_release().

2. refcount_t rcu_users: A refcount field which, when it drops to 0,
   schedules an RCU callback that drops a reference held on the 'usage'
   field above (which is acquired when the task is first created). This
   field therefore provides a form of RCU protection on the task by
   ensuring that at least one 'usage' refcount will be held until an RCU
   grace period has elapsed. The qualifier "a form of" is important
   here, as a task can remain valid after task->rcu_users has dropped to
   0 and the subsequent RCU gp has elapsed.

In terms of BPF, we want to use task->rcu_users to protect tasks that
function as referenced kptrs, and to allow tasks stored as referenced
kptrs in maps to be accessed with RCU protection.

Let's first determine whether we can safely use task->rcu_users to
protect tasks stored in maps. All of the bpf_task* kfuncs can only be
called from tracepoint, struct_ops, or BPF_PROG_TYPE_SCHED_CLS, program
types. For tracepoint and struct_ops programs, the struct task_struct
passed to a program handler will always be trusted, so it will always be
safe to call bpf_task_acquire() with any task passed to a program.
Note, however, that we must update bpf_task_acquire() to be KF_RET_NULL,
as it is possible that the task has exited by the time the program is
invoked, even if the pointer is still currently valid because the main
kernel holds a task->usage refcount. For BPF_PROG_TYPE_SCHED_CLS, tasks
should never be passed as an argument to the any program handlers, so it
should not be relevant.

The second question is whether it's safe to use RCU to access a task
that was acquired with bpf_task_acquire(), and stored in a map. Because
bpf_task_acquire() now uses task->rcu_users, it follows that if the task
is present in the map, that it must have had at least one
task->rcu_users refcount by the time the current RCU cs was started.
Therefore, it's safe to access that task until the end of the current
RCU cs.

With all that said, this patch makes struct task_struct is an
RCU-protected object. In doing so, we also change bpf_task_acquire() to
be KF_ACQUIRE | KF_RCU | KF_RET_NULL, and adjust any selftests as
necessary. A subsequent patch will remove bpf_task_kptr_get(), and
bpf_task_acquire_not_zero() respectively.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230331195733.699708-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c                               | 11 +--
 kernel/bpf/verifier.c                              |  1 +
 .../testing/selftests/bpf/prog_tests/task_kfunc.c  |  1 +
 .../selftests/bpf/progs/task_kfunc_common.h        |  5 ++
 .../selftests/bpf/progs/task_kfunc_failure.c       | 80 +++++++++++++++++++---
 .../selftests/bpf/progs/task_kfunc_success.c       | 26 ++++++-
 6 files changed, 108 insertions(+), 16 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 8980f6859443..e71a4a54ce99 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -18,6 +18,7 @@
 #include <linux/pid_namespace.h>
 #include <linux/poison.h>
 #include <linux/proc_ns.h>
+#include <linux/sched/task.h>
 #include <linux/security.h>
 #include <linux/btf_ids.h>
 #include <linux/bpf_mem_alloc.h>
@@ -2013,7 +2014,9 @@ __bpf_kfunc struct bpf_rb_node *bpf_rbtree_first(struct bpf_rb_root *root)
  */
 __bpf_kfunc struct task_struct *bpf_task_acquire(struct task_struct *p)
 {
-	return get_task_struct(p);
+	if (refcount_inc_not_zero(&p->rcu_users))
+		return p;
+	return NULL;
 }
 
 /**
@@ -2089,7 +2092,7 @@ __bpf_kfunc struct task_struct *bpf_task_kptr_get(struct task_struct **pp)
  */
 __bpf_kfunc void bpf_task_release(struct task_struct *p)
 {
-	put_task_struct(p);
+	put_task_struct_rcu_user(p);
 }
 
 #ifdef CONFIG_CGROUPS
@@ -2199,7 +2202,7 @@ __bpf_kfunc struct task_struct *bpf_task_from_pid(s32 pid)
 	rcu_read_lock();
 	p = find_task_by_pid_ns(pid, &init_pid_ns);
 	if (p)
-		bpf_task_acquire(p);
+		p = bpf_task_acquire(p);
 	rcu_read_unlock();
 
 	return p;
@@ -2371,7 +2374,7 @@ BTF_ID_FLAGS(func, bpf_list_push_front)
 BTF_ID_FLAGS(func, bpf_list_push_back)
 BTF_ID_FLAGS(func, bpf_list_pop_front, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_list_pop_back, KF_ACQUIRE | KF_RET_NULL)
-BTF_ID_FLAGS(func, bpf_task_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
+BTF_ID_FLAGS(func, bpf_task_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_acquire_not_zero, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_release, KF_RELEASE)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 52738f9dcb15..92ae4e8ab87b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4600,6 +4600,7 @@ BTF_SET_START(rcu_protected_types)
 BTF_ID(struct, prog_test_ref_kfunc)
 BTF_ID(struct, cgroup)
 BTF_ID(struct, bpf_cpumask)
+BTF_ID(struct, task_struct)
 BTF_SET_END(rcu_protected_types)
 
 static bool rcu_protected_object(const struct btf *btf, u32 btf_id)
diff --git a/tools/testing/selftests/bpf/prog_tests/task_kfunc.c b/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
index f79fa5bc9a8d..330133ece3f6 100644
--- a/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
+++ b/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
@@ -78,6 +78,7 @@ static const char * const success_tests[] = {
 	"test_task_from_pid_arg",
 	"test_task_from_pid_current",
 	"test_task_from_pid_invalid",
+	"task_kfunc_acquire_trusted_walked",
 };
 
 void test_task_kfunc(void)
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_common.h b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
index 4c2a4b0e3a25..bf0d1da9aff8 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
@@ -24,6 +24,8 @@ struct task_struct *bpf_task_acquire(struct task_struct *p) __ksym;
 struct task_struct *bpf_task_kptr_get(struct task_struct **pp) __ksym;
 void bpf_task_release(struct task_struct *p) __ksym;
 struct task_struct *bpf_task_from_pid(s32 pid) __ksym;
+void bpf_rcu_read_lock(void) __ksym;
+void bpf_rcu_read_unlock(void) __ksym;
 
 static inline struct __tasks_kfunc_map_value *tasks_kfunc_map_value_lookup(struct task_struct *p)
 {
@@ -60,6 +62,9 @@ static inline int tasks_kfunc_map_insert(struct task_struct *p)
 	}
 
 	acquired = bpf_task_acquire(p);
+	if (!acquired)
+		return -ENOENT;
+
 	old = bpf_kptr_xchg(&v->task, acquired);
 	if (old) {
 		bpf_task_release(old);
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
index 2c374a7ffece..63aef547da87 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
@@ -40,6 +40,9 @@ int BPF_PROG(task_kfunc_acquire_untrusted, struct task_struct *task, u64 clone_f
 
 	/* Can't invoke bpf_task_acquire() on an untrusted pointer. */
 	acquired = bpf_task_acquire(v->task);
+	if (!acquired)
+		return 0;
+
 	bpf_task_release(acquired);
 
 	return 0;
@@ -53,38 +56,49 @@ int BPF_PROG(task_kfunc_acquire_fp, struct task_struct *task, u64 clone_flags)
 
 	/* Can't invoke bpf_task_acquire() on a random frame pointer. */
 	acquired = bpf_task_acquire((struct task_struct *)&stack_task);
+	if (!acquired)
+		return 0;
+
 	bpf_task_release(acquired);
 
 	return 0;
 }
 
 SEC("kretprobe/free_task")
-__failure __msg("reg type unsupported for arg#0 function")
+__failure __msg("calling kernel function bpf_task_acquire is not allowed")
 int BPF_PROG(task_kfunc_acquire_unsafe_kretprobe, struct task_struct *task, u64 clone_flags)
 {
 	struct task_struct *acquired;
 
+	/* Can't call bpf_task_acquire() or bpf_task_release() in an untrusted prog. */
 	acquired = bpf_task_acquire(task);
-	/* Can't release a bpf_task_acquire()'d task without a NULL check. */
+	if (!acquired)
+		return 0;
 	bpf_task_release(acquired);
 
 	return 0;
 }
 
-SEC("tp_btf/task_newtask")
-__failure __msg("R1 must be referenced or trusted")
-int BPF_PROG(task_kfunc_acquire_trusted_walked, struct task_struct *task, u64 clone_flags)
+SEC("kretprobe/free_task")
+__failure __msg("calling kernel function bpf_task_acquire is not allowed")
+int BPF_PROG(task_kfunc_acquire_unsafe_kretprobe_rcu, struct task_struct *task, u64 clone_flags)
 {
 	struct task_struct *acquired;
 
-	/* Can't invoke bpf_task_acquire() on a trusted pointer obtained from walking a struct. */
-	acquired = bpf_task_acquire(task->group_leader);
-	bpf_task_release(acquired);
+	bpf_rcu_read_lock();
+	if (!task) {
+		bpf_rcu_read_unlock();
+		return 0;
+	}
+	/* Can't call bpf_task_acquire() or bpf_task_release() in an untrusted prog. */
+	acquired = bpf_task_acquire(task);
+	if (acquired)
+		bpf_task_release(acquired);
+	bpf_rcu_read_unlock();
 
 	return 0;
 }
 
-
 SEC("tp_btf/task_newtask")
 __failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(task_kfunc_acquire_null, struct task_struct *task, u64 clone_flags)
@@ -137,6 +151,8 @@ int BPF_PROG(task_kfunc_get_non_kptr_acquired, struct task_struct *task, u64 clo
 	struct task_struct *kptr, *acquired;
 
 	acquired = bpf_task_acquire(task);
+	if (!acquired)
+		return 0;
 
 	/* Cannot use bpf_task_kptr_get() on a non-kptr, even if it was acquired. */
 	kptr = bpf_task_kptr_get(&acquired);
@@ -185,6 +201,19 @@ int BPF_PROG(task_kfunc_xchg_unreleased, struct task_struct *task, u64 clone_fla
 	return 0;
 }
 
+SEC("tp_btf/task_newtask")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
+int BPF_PROG(task_kfunc_acquire_release_no_null_check, struct task_struct *task, u64 clone_flags)
+{
+	struct task_struct *acquired;
+
+	acquired = bpf_task_acquire(task);
+	/* Can't invoke bpf_task_release() on an acquired task without a NULL check. */
+	bpf_task_release(acquired);
+
+	return 0;
+}
+
 SEC("tp_btf/task_newtask")
 __failure __msg("Unreleased reference")
 int BPF_PROG(task_kfunc_get_unreleased, struct task_struct *task, u64 clone_flags)
@@ -256,12 +285,13 @@ int BPF_PROG(task_kfunc_release_null, struct task_struct *task, u64 clone_flags)
 		return -ENOENT;
 
 	acquired = bpf_task_acquire(task);
+	if (!acquired)
+		return -EEXIST;
 
 	old = bpf_kptr_xchg(&v->task, acquired);
 
 	/* old cannot be passed to bpf_task_release() without a NULL check. */
 	bpf_task_release(old);
-	bpf_task_release(old);
 
 	return 0;
 }
@@ -298,6 +328,9 @@ int BPF_PROG(task_kfunc_from_lsm_task_free, struct task_struct *task)
 
 	/* the argument of lsm task_free hook is untrusted. */
 	acquired = bpf_task_acquire(task);
+	if (!acquired)
+		return 0;
+
 	bpf_task_release(acquired);
 	return 0;
 }
@@ -337,3 +370,30 @@ int BPF_PROG(task_access_comm4, struct task_struct *task, const char *buf, bool
 	bpf_strncmp(task->comm, 16, "foo");
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+__failure __msg("R1 must be referenced or trusted")
+int BPF_PROG(task_kfunc_release_in_map, struct task_struct *task, u64 clone_flags)
+{
+	struct task_struct *local;
+	struct __tasks_kfunc_map_value *v;
+
+	if (tasks_kfunc_map_insert(task))
+		return 0;
+
+	v = tasks_kfunc_map_value_lookup(task);
+	if (!v)
+		return 0;
+
+	bpf_rcu_read_lock();
+	local = v->task;
+	if (!local) {
+		bpf_rcu_read_unlock();
+		return 0;
+	}
+	/* Can't release a kptr that's still stored in a map. */
+	bpf_task_release(local);
+	bpf_rcu_read_unlock();
+
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index cfa7f12b84e8..a75304a5e860 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -47,7 +47,10 @@ static int test_acquire_release(struct task_struct *task)
 	}
 
 	acquired = bpf_task_acquire(task);
-	bpf_task_release(acquired);
+	if (acquired)
+		bpf_task_release(acquired);
+	else
+		err = 6;
 
 	return 0;
 }
@@ -166,7 +169,10 @@ int BPF_PROG(test_task_current_acquire_release, struct task_struct *task, u64 cl
 
 	current = bpf_get_current_task_btf();
 	acquired = bpf_task_acquire(current);
-	bpf_task_release(acquired);
+	if (acquired)
+		bpf_task_release(acquired);
+	else
+		err = 1;
 
 	return 0;
 }
@@ -241,3 +247,19 @@ int BPF_PROG(test_task_from_pid_invalid, struct task_struct *task, u64 clone_fla
 
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+int BPF_PROG(task_kfunc_acquire_trusted_walked, struct task_struct *task, u64 clone_flags)
+{
+	struct task_struct *acquired;
+
+	/* task->group_leader is listed as a trusted, non-NULL field of task struct. */
+	acquired = bpf_task_acquire(task->group_leader);
+	if (acquired)
+		bpf_task_release(acquired);
+	else
+		err = 1;
+
+
+	return 0;
+}
-- 
cgit v1.2.3-70-g09d2


From f85671c6ef46d490a90dac719e0c0e0adbacfd9b Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Fri, 31 Mar 2023 14:57:32 -0500
Subject: bpf: Remove now-defunct task kfuncs

In commit 22df776a9a86 ("tasks: Extract rcu_users out of union"), the
'refcount_t rcu_users' field was extracted out of a union with the
'struct rcu_head rcu' field. This allows us to safely perform a
refcount_inc_not_zero() on task->rcu_users when acquiring a reference on
a task struct. A prior patch leveraged this by making struct task_struct
an RCU-protected object in the verifier, and by bpf_task_acquire() to
use the task->rcu_users field for synchronization.

Now that we can use RCU to protect tasks, we no longer need
bpf_task_kptr_get(), or bpf_task_acquire_not_zero(). bpf_task_kptr_get()
is truly completely unnecessary, as we can just use RCU to get the
object. bpf_task_acquire_not_zero() is now equivalent to
bpf_task_acquire().

In addition to these changes, this patch also updates the associated
selftests to no longer use these kfuncs.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230331195733.699708-3-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c                               | 69 --------------------
 .../testing/selftests/bpf/prog_tests/task_kfunc.c  |  2 +-
 tools/testing/selftests/bpf/progs/rcu_read_lock.c  |  9 +--
 .../selftests/bpf/progs/task_kfunc_common.h        |  1 -
 .../selftests/bpf/progs/task_kfunc_failure.c       | 73 ----------------------
 .../selftests/bpf/progs/task_kfunc_success.c       | 22 +++----
 6 files changed, 14 insertions(+), 162 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index e71a4a54ce99..6be16db9f188 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2019,73 +2019,6 @@ __bpf_kfunc struct task_struct *bpf_task_acquire(struct task_struct *p)
 	return NULL;
 }
 
-/**
- * bpf_task_acquire_not_zero - Acquire a reference to a rcu task object. A task
- * acquired by this kfunc which is not stored in a map as a kptr, must be
- * released by calling bpf_task_release().
- * @p: The task on which a reference is being acquired.
- */
-__bpf_kfunc struct task_struct *bpf_task_acquire_not_zero(struct task_struct *p)
-{
-	/* For the time being this function returns NULL, as it's not currently
-	 * possible to safely acquire a reference to a task with RCU protection
-	 * using get_task_struct() and put_task_struct(). This is due to the
-	 * slightly odd mechanics of p->rcu_users, and how task RCU protection
-	 * works.
-	 *
-	 * A struct task_struct is refcounted by two different refcount_t
-	 * fields:
-	 *
-	 * 1. p->usage:     The "true" refcount field which tracks a task's
-	 *		    lifetime. The task is freed as soon as this
-	 *		    refcount drops to 0.
-	 *
-	 * 2. p->rcu_users: An "RCU users" refcount field which is statically
-	 *		    initialized to 2, and is co-located in a union with
-	 *		    a struct rcu_head field (p->rcu). p->rcu_users
-	 *		    essentially encapsulates a single p->usage
-	 *		    refcount, and when p->rcu_users goes to 0, an RCU
-	 *		    callback is scheduled on the struct rcu_head which
-	 *		    decrements the p->usage refcount.
-	 *
-	 * There are two important implications to this task refcounting logic
-	 * described above. The first is that
-	 * refcount_inc_not_zero(&p->rcu_users) cannot be used anywhere, as
-	 * after the refcount goes to 0, the RCU callback being scheduled will
-	 * cause the memory backing the refcount to again be nonzero due to the
-	 * fields sharing a union. The other is that we can't rely on RCU to
-	 * guarantee that a task is valid in a BPF program. This is because a
-	 * task could have already transitioned to being in the TASK_DEAD
-	 * state, had its rcu_users refcount go to 0, and its rcu callback
-	 * invoked in which it drops its single p->usage reference. At this
-	 * point the task will be freed as soon as the last p->usage reference
-	 * goes to 0, without waiting for another RCU gp to elapse. The only
-	 * way that a BPF program can guarantee that a task is valid is in this
-	 * scenario is to hold a p->usage refcount itself.
-	 *
-	 * Until we're able to resolve this issue, either by pulling
-	 * p->rcu_users and p->rcu out of the union, or by getting rid of
-	 * p->usage and just using p->rcu_users for refcounting, we'll just
-	 * return NULL here.
-	 */
-	return NULL;
-}
-
-/**
- * bpf_task_kptr_get - Acquire a reference on a struct task_struct kptr. A task
- * kptr acquired by this kfunc which is not subsequently stored in a map, must
- * be released by calling bpf_task_release().
- * @pp: A pointer to a task kptr on which a reference is being acquired.
- */
-__bpf_kfunc struct task_struct *bpf_task_kptr_get(struct task_struct **pp)
-{
-	/* We must return NULL here until we have clarity on how to properly
-	 * leverage RCU for ensuring a task's lifetime. See the comment above
-	 * in bpf_task_acquire_not_zero() for more details.
-	 */
-	return NULL;
-}
-
 /**
  * bpf_task_release - Release the reference acquired on a task.
  * @p: The task on which a reference is being released.
@@ -2375,8 +2308,6 @@ BTF_ID_FLAGS(func, bpf_list_push_back)
 BTF_ID_FLAGS(func, bpf_list_pop_front, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_list_pop_back, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
-BTF_ID_FLAGS(func, bpf_task_acquire_not_zero, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
-BTF_ID_FLAGS(func, bpf_task_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_release, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_rbtree_remove, KF_ACQUIRE)
 BTF_ID_FLAGS(func, bpf_rbtree_add)
diff --git a/tools/testing/selftests/bpf/prog_tests/task_kfunc.c b/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
index 330133ece3f6..740d5f644b40 100644
--- a/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
+++ b/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
@@ -73,7 +73,7 @@ static const char * const success_tests[] = {
 	"test_task_acquire_release_current",
 	"test_task_acquire_leave_in_map",
 	"test_task_xchg_release",
-	"test_task_get_release",
+	"test_task_map_acquire_release",
 	"test_task_current_acquire_release",
 	"test_task_from_pid_arg",
 	"test_task_from_pid_current",
diff --git a/tools/testing/selftests/bpf/progs/rcu_read_lock.c b/tools/testing/selftests/bpf/progs/rcu_read_lock.c
index 6a8c88e58df2..14fb01437fb8 100644
--- a/tools/testing/selftests/bpf/progs/rcu_read_lock.c
+++ b/tools/testing/selftests/bpf/progs/rcu_read_lock.c
@@ -23,7 +23,7 @@ struct bpf_key *bpf_lookup_user_key(__u32 serial, __u64 flags) __ksym;
 void bpf_key_put(struct bpf_key *key) __ksym;
 void bpf_rcu_read_lock(void) __ksym;
 void bpf_rcu_read_unlock(void) __ksym;
-struct task_struct *bpf_task_acquire_not_zero(struct task_struct *p) __ksym;
+struct task_struct *bpf_task_acquire(struct task_struct *p) __ksym;
 void bpf_task_release(struct task_struct *p) __ksym;
 
 SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
@@ -159,13 +159,8 @@ int task_acquire(void *ctx)
 		goto out;
 
 	/* acquire a reference which can be used outside rcu read lock region */
-	gparent = bpf_task_acquire_not_zero(gparent);
+	gparent = bpf_task_acquire(gparent);
 	if (!gparent)
-		/* Until we resolve the issues with using task->rcu_users, we
-		 * expect bpf_task_acquire_not_zero() to return a NULL task.
-		 * See the comment at the definition of
-		 * bpf_task_acquire_not_zero() for more details.
-		 */
 		goto out;
 
 	(void)bpf_task_storage_get(&map_a, gparent, 0, 0);
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_common.h b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
index bf0d1da9aff8..41f2d44f49cb 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
@@ -21,7 +21,6 @@ struct hash_map {
 } __tasks_kfunc_map SEC(".maps");
 
 struct task_struct *bpf_task_acquire(struct task_struct *p) __ksym;
-struct task_struct *bpf_task_kptr_get(struct task_struct **pp) __ksym;
 void bpf_task_release(struct task_struct *p) __ksym;
 struct task_struct *bpf_task_from_pid(s32 pid) __ksym;
 void bpf_rcu_read_lock(void) __ksym;
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
index 63aef547da87..dcdea3127086 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
@@ -128,59 +128,6 @@ int BPF_PROG(task_kfunc_acquire_unreleased, struct task_struct *task, u64 clone_
 	return 0;
 }
 
-SEC("tp_btf/task_newtask")
-__failure __msg("arg#0 expected pointer to map value")
-int BPF_PROG(task_kfunc_get_non_kptr_param, struct task_struct *task, u64 clone_flags)
-{
-	struct task_struct *kptr;
-
-	/* Cannot use bpf_task_kptr_get() on a non-kptr, even on a valid task. */
-	kptr = bpf_task_kptr_get(&task);
-	if (!kptr)
-		return 0;
-
-	bpf_task_release(kptr);
-
-	return 0;
-}
-
-SEC("tp_btf/task_newtask")
-__failure __msg("arg#0 expected pointer to map value")
-int BPF_PROG(task_kfunc_get_non_kptr_acquired, struct task_struct *task, u64 clone_flags)
-{
-	struct task_struct *kptr, *acquired;
-
-	acquired = bpf_task_acquire(task);
-	if (!acquired)
-		return 0;
-
-	/* Cannot use bpf_task_kptr_get() on a non-kptr, even if it was acquired. */
-	kptr = bpf_task_kptr_get(&acquired);
-	bpf_task_release(acquired);
-	if (!kptr)
-		return 0;
-
-	bpf_task_release(kptr);
-
-	return 0;
-}
-
-SEC("tp_btf/task_newtask")
-__failure __msg("arg#0 expected pointer to map value")
-int BPF_PROG(task_kfunc_get_null, struct task_struct *task, u64 clone_flags)
-{
-	struct task_struct *kptr;
-
-	/* Cannot use bpf_task_kptr_get() on a NULL pointer. */
-	kptr = bpf_task_kptr_get(NULL);
-	if (!kptr)
-		return 0;
-
-	bpf_task_release(kptr);
-
-	return 0;
-}
-
 SEC("tp_btf/task_newtask")
 __failure __msg("Unreleased reference")
 int BPF_PROG(task_kfunc_xchg_unreleased, struct task_struct *task, u64 clone_flags)
@@ -214,26 +161,6 @@ int BPF_PROG(task_kfunc_acquire_release_no_null_check, struct task_struct *task,
 	return 0;
 }
 
-SEC("tp_btf/task_newtask")
-__failure __msg("Unreleased reference")
-int BPF_PROG(task_kfunc_get_unreleased, struct task_struct *task, u64 clone_flags)
-{
-	struct task_struct *kptr;
-	struct __tasks_kfunc_map_value *v;
-
-	v = insert_lookup_task(task);
-	if (!v)
-		return 0;
-
-	kptr = bpf_task_kptr_get(&v->task);
-	if (!kptr)
-		return 0;
-
-	/* Kptr acquired above is never released. */
-
-	return 0;
-}
-
 SEC("tp_btf/task_newtask")
 __failure __msg("Possibly NULL pointer passed to trusted arg0")
 int BPF_PROG(task_kfunc_release_untrusted, struct task_struct *task, u64 clone_flags)
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index a75304a5e860..b09371bba204 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -122,7 +122,7 @@ int BPF_PROG(test_task_xchg_release, struct task_struct *task, u64 clone_flags)
 }
 
 SEC("tp_btf/task_newtask")
-int BPF_PROG(test_task_get_release, struct task_struct *task, u64 clone_flags)
+int BPF_PROG(test_task_map_acquire_release, struct task_struct *task, u64 clone_flags)
 {
 	struct task_struct *kptr;
 	struct __tasks_kfunc_map_value *v;
@@ -143,18 +143,18 @@ int BPF_PROG(test_task_get_release, struct task_struct *task, u64 clone_flags)
 		return 0;
 	}
 
-	kptr = bpf_task_kptr_get(&v->task);
-	if (kptr) {
-		/* Until we resolve the issues with using task->rcu_users, we
-		 * expect bpf_task_kptr_get() to return a NULL task. See the
-		 * comment at the definition of bpf_task_acquire_not_zero() for
-		 * more details.
-		 */
-		bpf_task_release(kptr);
+	bpf_rcu_read_lock();
+	kptr = v->task;
+	if (!kptr) {
 		err = 3;
-		return 0;
+	} else {
+		kptr = bpf_task_acquire(kptr);
+		if (!kptr)
+			err = 4;
+		else
+			bpf_task_release(kptr);
 	}
-
+	bpf_rcu_read_unlock();
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From db9d479ab59b21d719486e6bf673f83f129dae32 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Fri, 31 Mar 2023 14:57:33 -0500
Subject: bpf,docs: Update documentation to reflect new task kfuncs

Now that struct task_struct objects are RCU safe, and bpf_task_acquire()
can return NULL, we should update the BPF task kfunc documentation to
reflect the current state of the API.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230331195733.699708-4-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst | 49 ++++++++++++++++++++++++++++++++++++++------
 1 file changed, 43 insertions(+), 6 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index bf1b85941452..d8a16c4bef7f 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -471,13 +471,50 @@ struct_ops callback arg. For example:
 		struct task_struct *acquired;
 
 		acquired = bpf_task_acquire(task);
+		if (acquired)
+			/*
+			 * In a typical program you'd do something like store
+			 * the task in a map, and the map will automatically
+			 * release it later. Here, we release it manually.
+			 */
+			bpf_task_release(acquired);
+		return 0;
+	}
+
+
+References acquired on ``struct task_struct *`` objects are RCU protected.
+Therefore, when in an RCU read region, you can obtain a pointer to a task
+embedded in a map value without having to acquire a reference:
+
+.. code-block:: c
+
+	#define private(name) SEC(".data." #name) __hidden __attribute__((aligned(8)))
+	private(TASK) static struct task_struct *global;
+
+	/**
+	 * A trivial example showing how to access a task stored
+	 * in a map using RCU.
+	 */
+	SEC("tp_btf/task_newtask")
+	int BPF_PROG(task_rcu_read_example, struct task_struct *task, u64 clone_flags)
+	{
+		struct task_struct *local_copy;
+
+		bpf_rcu_read_lock();
+		local_copy = global;
+		if (local_copy)
+			/*
+			 * We could also pass local_copy to kfuncs or helper functions here,
+			 * as we're guaranteed that local_copy will be valid until we exit
+			 * the RCU read region below.
+			 */
+			bpf_printk("Global task %s is valid", local_copy->comm);
+		else
+			bpf_printk("No global task found");
+		bpf_rcu_read_unlock();
+
+		/* At this point we can no longer reference local_copy. */
 
-		/*
-		 * In a typical program you'd do something like store
-		 * the task in a map, and the map will automatically
-		 * release it later. Here, we release it manually.
-		 */
-		bpf_task_release(acquired);
 		return 0;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 5b85575ad4280171c382e888b006cb8d12c35bde Mon Sep 17 00:00:00 2001
From: Anton Protopopov <aspsk@isovalent.com>
Date: Sat, 1 Apr 2023 20:06:02 +0000
Subject: bpf: optimize hashmap lookups when key_size is divisible by 4

The BPF hashmap uses the jhash() hash function. There is an optimized version
of this hash function which may be used if hash size is a multiple of 4. Apply
this optimization to the hashmap in a similar way as it is done in the bloom
filter map.

On practice the optimization is only noticeable for smaller key sizes, which,
however, is sufficient for many applications. An example is listed in the
following table of measurements (a hashmap of 65536 elements was used):

    --------------------------------------------------------------------
    | key_size | fullness | lookups /sec | lookups (opt) /sec |   gain |
    --------------------------------------------------------------------
    |        4 |      25% |      42.990M |            46.000M |   7.0% |
    |        4 |      50% |      37.910M |            39.094M |   3.1% |
    |        4 |      75% |      34.486M |            36.124M |   4.7% |
    |        4 |     100% |      31.760M |            32.719M |   3.0% |
    --------------------------------------------------------------------
    |        8 |      25% |      43.855M |            49.626M |  13.2% |
    |        8 |      50% |      38.328M |            42.152M |  10.0% |
    |        8 |      75% |      34.483M |            38.088M |  10.5% |
    |        8 |     100% |      31.306M |            34.686M |  10.8% |
    --------------------------------------------------------------------
    |       12 |      25% |      38.398M |            43.770M |  14.0% |
    |       12 |      50% |      33.336M |            37.712M |  13.1% |
    |       12 |      75% |      29.917M |            34.440M |  15.1% |
    |       12 |     100% |      27.322M |            30.480M |  11.6% |
    --------------------------------------------------------------------
    |       16 |      25% |      41.491M |            41.921M |   1.0% |
    |       16 |      50% |      36.206M |            36.474M |   0.7% |
    |       16 |      75% |      32.529M |            33.027M |   1.5% |
    |       16 |     100% |      29.581M |            30.325M |   2.5% |
    --------------------------------------------------------------------
    |       20 |      25% |      34.240M |            36.787M |   7.4% |
    |       20 |      50% |      30.328M |            32.663M |   7.7% |
    |       20 |      75% |      27.536M |            29.354M |   6.6% |
    |       20 |     100% |      24.847M |            26.505M |   6.7% |
    --------------------------------------------------------------------
    |       24 |      25% |      36.329M |            40.608M |  11.8% |
    |       24 |      50% |      31.444M |            35.059M |  11.5% |
    |       24 |      75% |      28.426M |            31.452M |  10.6% |
    |       24 |     100% |      26.278M |            28.741M |   9.4% |
    --------------------------------------------------------------------
    |       28 |      25% |      31.540M |            31.944M |   1.3% |
    |       28 |      50% |      27.739M |            28.063M |   1.2% |
    |       28 |      75% |      24.993M |            25.814M |   3.3% |
    |       28 |     100% |      23.513M |            23.500M |  -0.1% |
    --------------------------------------------------------------------
    |       32 |      25% |      32.116M |            33.953M |   5.7% |
    |       32 |      50% |      28.879M |            29.859M |   3.4% |
    |       32 |      75% |      26.227M |            26.948M |   2.7% |
    |       32 |     100% |      23.829M |            24.613M |   3.3% |
    --------------------------------------------------------------------
    |       64 |      25% |      22.535M |            22.554M |   0.1% |
    |       64 |      50% |      20.471M |            20.675M |   1.0% |
    |       64 |      75% |      19.077M |            19.146M |   0.4% |
    |       64 |     100% |      17.710M |            18.131M |   2.4% |
    --------------------------------------------------------------------

The following script was used to gather the results (SMT & frequency off):

    cd tools/testing/selftests/bpf
    for key_size in 4 8 12 16 20 24 28 32 64; do
            for nr_entries in `seq 16384 16384 65536`; do
                    fullness=$(printf '%3s' $((nr_entries*100/65536)))
                    echo -n "key_size=$key_size: $fullness% full: "
                    sudo ./bench -d2 -a bpf-hashmap-lookup --key_size=$key_size --nr_entries=$nr_entries --max_entries=65536 --nr_loops=2000000 --map_flags=0x40 | grep cpu
            done
            echo
    done

Signed-off-by: Anton Protopopov <aspsk@isovalent.com>
Link: https://lore.kernel.org/r/20230401200602.3275-1-aspsk@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/hashtab.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 96b645bba3a4..00c253b84bf5 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -607,6 +607,8 @@ free_htab:
 
 static inline u32 htab_map_hash(const void *key, u32 key_len, u32 hashrnd)
 {
+	if (likely(key_len % 4 == 0))
+		return jhash2(key, key_len / 4, hashrnd);
 	return jhash(key, key_len, hashrnd);
 }
 
-- 
cgit v1.2.3-70-g09d2


From ceb29474bbbc377e11be3a70589a0005305e4fc3 Mon Sep 17 00:00:00 2001
From: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Date: Thu, 30 Mar 2023 10:00:22 -0700
Subject: i40e: Add support for VF to specify its primary MAC address

Currently in the i40e driver there is no implementation of different
MAC address handling depending on whether it is a legacy or primary.
Introduce new checks for VF to be able to specify its primary MAC
address based on the VIRTCHNL_ETHER_ADDR_PRIMARY type.

Primary MAC address are treated differently compared to legacy
ones in a scenario where:
1. If a unicast MAC is being added and it's specified as
VIRTCHNL_ETHER_ADDR_PRIMARY, then replace the current
default_lan_addr.addr.
2. If a unicast MAC is being deleted and it's type
is specified as VIRTCHNL_ETHER_ADDR_PRIMARY, then zero the
hw_lan_addr.addr.

Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Mateusz Palczewski <mateusz.palczewski@intel.com>
Tested-by: Rafal Romanowski <rafal.romanowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 74 ++++++++++++++++++++--
 1 file changed, 70 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 8a4587585acd..be59ba3774e1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -2914,6 +2914,72 @@ static inline int i40e_check_vf_permission(struct i40e_vf *vf,
 	return 0;
 }
 
+/**
+ * i40e_vc_ether_addr_type - get type of virtchnl_ether_addr
+ * @vc_ether_addr: used to extract the type
+ **/
+static u8
+i40e_vc_ether_addr_type(struct virtchnl_ether_addr *vc_ether_addr)
+{
+	return vc_ether_addr->type & VIRTCHNL_ETHER_ADDR_TYPE_MASK;
+}
+
+/**
+ * i40e_is_vc_addr_legacy
+ * @vc_ether_addr: VIRTCHNL structure that contains MAC and type
+ *
+ * check if the MAC address is from an older VF
+ **/
+static bool
+i40e_is_vc_addr_legacy(struct virtchnl_ether_addr *vc_ether_addr)
+{
+	return i40e_vc_ether_addr_type(vc_ether_addr) ==
+		VIRTCHNL_ETHER_ADDR_LEGACY;
+}
+
+/**
+ * i40e_is_vc_addr_primary
+ * @vc_ether_addr: VIRTCHNL structure that contains MAC and type
+ *
+ * check if the MAC address is the VF's primary MAC
+ * This function should only be called when the MAC address in
+ * virtchnl_ether_addr is a valid unicast MAC
+ **/
+static bool
+i40e_is_vc_addr_primary(struct virtchnl_ether_addr *vc_ether_addr)
+{
+	return i40e_vc_ether_addr_type(vc_ether_addr) ==
+		VIRTCHNL_ETHER_ADDR_PRIMARY;
+}
+
+/**
+ * i40e_update_vf_mac_addr
+ * @vf: VF to update
+ * @vc_ether_addr: structure from VIRTCHNL with MAC to add
+ *
+ * update the VF's cached hardware MAC if allowed
+ **/
+static void
+i40e_update_vf_mac_addr(struct i40e_vf *vf,
+			struct virtchnl_ether_addr *vc_ether_addr)
+{
+	u8 *mac_addr = vc_ether_addr->addr;
+
+	if (!is_valid_ether_addr(mac_addr))
+		return;
+
+	/* If request to add MAC filter is a primary request update its default
+	 * MAC address with the requested one. If it is a legacy request then
+	 * check if current default is empty if so update the default MAC
+	 */
+	if (i40e_is_vc_addr_primary(vc_ether_addr)) {
+		ether_addr_copy(vf->default_lan_addr.addr, mac_addr);
+	} else if (i40e_is_vc_addr_legacy(vc_ether_addr)) {
+		if (is_zero_ether_addr(vf->default_lan_addr.addr))
+			ether_addr_copy(vf->default_lan_addr.addr, mac_addr);
+	}
+}
+
 /**
  * i40e_vc_add_mac_addr_msg
  * @vf: pointer to the VF info
@@ -2965,11 +3031,8 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg)
 				spin_unlock_bh(&vsi->mac_filter_hash_lock);
 				goto error_param;
 			}
-			if (is_valid_ether_addr(al->list[i].addr) &&
-			    is_zero_ether_addr(vf->default_lan_addr.addr))
-				ether_addr_copy(vf->default_lan_addr.addr,
-						al->list[i].addr);
 		}
+		i40e_update_vf_mac_addr(vf, &al->list[i]);
 	}
 	spin_unlock_bh(&vsi->mac_filter_hash_lock);
 
@@ -3032,6 +3095,9 @@ static int i40e_vc_del_mac_addr_msg(struct i40e_vf *vf, u8 *msg)
 
 	spin_unlock_bh(&vsi->mac_filter_hash_lock);
 
+	if (was_unimac_deleted)
+		eth_zero_addr(vf->default_lan_addr.addr);
+
 	/* program the updated filter list */
 	ret = i40e_sync_vsi_filters(vsi);
 	if (ret)
-- 
cgit v1.2.3-70-g09d2


From dd2d6604407da5b1b260faee409cd601fe914ce9 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 30 Mar 2023 21:47:31 -0700
Subject: net: minor reshuffle of napi_struct

napi_id is read by GRO and drivers to mark skbs, and it currently
sits at the end of the structure, in a mostly unused cache line.
Move it up into a hole, and separate the clearly control path
fields from the important ones.

Before:

struct napi_struct {
	struct list_head           poll_list;            /*     0    16 */
	long unsigned int          state;                /*    16     8 */
	int                        weight;               /*    24     4 */
	int                        defer_hard_irqs_count; /*    28     4 */
	long unsigned int          gro_bitmask;          /*    32     8 */
	int                        (*poll)(struct napi_struct *, int); /*    40     8 */
	int                        poll_owner;           /*    48     4 */

	/* XXX 4 bytes hole, try to pack */

	struct net_device *        dev;                  /*    56     8 */
	/* --- cacheline 1 boundary (64 bytes) --- */
	struct gro_list            gro_hash[8];          /*    64   192 */
	/* --- cacheline 4 boundary (256 bytes) --- */
	struct sk_buff *           skb;                  /*   256     8 */
	struct list_head           rx_list;              /*   264    16 */
	int                        rx_count;             /*   280     4 */

	/* XXX 4 bytes hole, try to pack */

	struct hrtimer             timer;                /*   288    64 */

	/* XXX last struct has 4 bytes of padding */

	/* --- cacheline 5 boundary (320 bytes) was 32 bytes ago --- */
	struct list_head           dev_list;             /*   352    16 */
	struct hlist_node          napi_hash_node;       /*   368    16 */
	/* --- cacheline 6 boundary (384 bytes) --- */
	unsigned int               napi_id;              /*   384     4 */

	/* XXX 4 bytes hole, try to pack */

	struct task_struct *       thread;               /*   392     8 */

	/* size: 400, cachelines: 7, members: 17 */
	/* sum members: 388, holes: 3, sum holes: 12 */
	/* paddings: 1, sum paddings: 4 */
	/* last cacheline: 16 bytes */
};

After:

struct napi_struct {
	struct list_head           poll_list;            /*     0    16 */
	long unsigned int          state;                /*    16     8 */
	int                        weight;               /*    24     4 */
	int                        defer_hard_irqs_count; /*    28     4 */
	long unsigned int          gro_bitmask;          /*    32     8 */
	int                        (*poll)(struct napi_struct *, int); /*    40     8 */
	int                        poll_owner;           /*    48     4 */

	/* XXX 4 bytes hole, try to pack */

	struct net_device *        dev;                  /*    56     8 */
	/* --- cacheline 1 boundary (64 bytes) --- */
	struct gro_list            gro_hash[8];          /*    64   192 */
	/* --- cacheline 4 boundary (256 bytes) --- */
	struct sk_buff *           skb;                  /*   256     8 */
	struct list_head           rx_list;              /*   264    16 */
	int                        rx_count;             /*   280     4 */
	unsigned int               napi_id;              /*   284     4 */
	struct hrtimer             timer;                /*   288    64 */

	/* XXX last struct has 4 bytes of padding */

	/* --- cacheline 5 boundary (320 bytes) was 32 bytes ago --- */
	struct task_struct *       thread;               /*   352     8 */
	struct list_head           dev_list;             /*   360    16 */
	struct hlist_node          napi_hash_node;       /*   376    16 */

	/* size: 392, cachelines: 7, members: 17 */
	/* sum members: 388, holes: 1, sum holes: 4 */
	/* paddings: 1, sum paddings: 4 */
	/* forced alignments: 1 */
	/* last cacheline: 8 bytes */
} __attribute__((__aligned__(8)));

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c8c634091a65..62e093a6d6d1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -367,11 +367,12 @@ struct napi_struct {
 	struct sk_buff		*skb;
 	struct list_head	rx_list; /* Pending GRO_NORMAL skbs */
 	int			rx_count; /* length of rx_list */
+	unsigned int		napi_id;
 	struct hrtimer		timer;
+	struct task_struct	*thread;
+	/* control-path-only fields follow */
 	struct list_head	dev_list;
 	struct hlist_node	napi_hash_node;
-	unsigned int		napi_id;
-	struct task_struct	*thread;
 };
 
 enum {
-- 
cgit v1.2.3-70-g09d2


From 5601ef91fba8efea483a7ca3bd6c42a581e4daa1 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Fri, 31 Mar 2023 16:17:30 +0200
Subject: mlxsw: core_thermal: Use static trip points for transceiver modules

The driver registers a thermal zone for each transceiver module and
tries to set the trip point temperatures according to the thresholds
read from the transceiver. If a threshold cannot be read or if a
transceiver is unplugged, the trip point temperature is set to zero,
which means that it is disabled as far as the thermal subsystem is
concerned.

A recent change in the thermal core made it so that such trip points are
no longer marked as disabled, which lead the thermal subsystem to
incorrectly set the associated cooling devices to the their maximum
state [1]. A fix to restore this behavior was merged in commit
f1b80a3878b2 ("thermal: core: Restore behavior regarding invalid trip
points"). However, the thermal maintainer suggested to not rely on this
behavior and instead always register a valid array of trip points [2].

Therefore, create a static array of trip points with sane defaults
(suggested by Vadim) and register it with the thermal zone of each
transceiver module. User space can choose to override these defaults
using the thermal zone sysfs interface since these files are writeable.

Before:

 $ cat /sys/class/thermal/thermal_zone11/type
 mlxsw-module11
 $ cat /sys/class/thermal/thermal_zone11/trip_point_*_temp
 65000
 75000
 80000

After:

 $ cat /sys/class/thermal/thermal_zone11/type
 mlxsw-module11
 $ cat /sys/class/thermal/thermal_zone11/trip_point_*_temp
 55000
 65000
 80000

Also tested by reverting commit f1b80a3878b2 ("thermal: core: Restore
behavior regarding invalid trip points") and making sure that the
associated cooling devices are not set to their maximum state.

[1] https://lore.kernel.org/linux-pm/ZA3CFNhU4AbtsP4G@shredder/
[2] https://lore.kernel.org/linux-pm/f78e6b70-a963-c0ca-a4b2-0d4c6aeef1fb@linaro.org/

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/core_thermal.c | 110 +++++----------------
 1 file changed, 25 insertions(+), 85 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index 09ed6e5fa6c3..ece5075b7dbf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -19,6 +19,9 @@
 #define MLXSW_THERMAL_ASIC_TEMP_NORM	75000	/* 75C */
 #define MLXSW_THERMAL_ASIC_TEMP_HIGH	85000	/* 85C */
 #define MLXSW_THERMAL_ASIC_TEMP_HOT	105000	/* 105C */
+#define MLXSW_THERMAL_MODULE_TEMP_NORM	55000	/* 55C */
+#define MLXSW_THERMAL_MODULE_TEMP_HIGH	65000	/* 65C */
+#define MLXSW_THERMAL_MODULE_TEMP_HOT	80000	/* 80C */
 #define MLXSW_THERMAL_HYSTERESIS_TEMP	5000	/* 5C */
 #define MLXSW_THERMAL_MODULE_TEMP_SHIFT	(MLXSW_THERMAL_HYSTERESIS_TEMP * 2)
 #define MLXSW_THERMAL_MAX_STATE	10
@@ -30,12 +33,6 @@ static char * const mlxsw_thermal_external_allowed_cdev[] = {
 	"mlxreg_fan",
 };
 
-enum mlxsw_thermal_trips {
-	MLXSW_THERMAL_TEMP_TRIP_NORM,
-	MLXSW_THERMAL_TEMP_TRIP_HIGH,
-	MLXSW_THERMAL_TEMP_TRIP_HOT,
-};
-
 struct mlxsw_cooling_states {
 	int	min_state;
 	int	max_state;
@@ -59,6 +56,24 @@ static const struct thermal_trip default_thermal_trips[] = {
 	},
 };
 
+static const struct thermal_trip default_thermal_module_trips[] = {
+	{	/* In range - 0-40% PWM */
+		.type		= THERMAL_TRIP_ACTIVE,
+		.temperature	= MLXSW_THERMAL_MODULE_TEMP_NORM,
+		.hysteresis	= MLXSW_THERMAL_HYSTERESIS_TEMP,
+	},
+	{
+		/* In range - 40-100% PWM */
+		.type		= THERMAL_TRIP_ACTIVE,
+		.temperature	= MLXSW_THERMAL_MODULE_TEMP_HIGH,
+		.hysteresis	= MLXSW_THERMAL_HYSTERESIS_TEMP,
+	},
+	{	/* Warning */
+		.type		= THERMAL_TRIP_HOT,
+		.temperature	= MLXSW_THERMAL_MODULE_TEMP_HOT,
+	},
+};
+
 static const struct mlxsw_cooling_states default_cooling_states[] = {
 	{
 		.min_state	= 0,
@@ -140,63 +155,6 @@ static int mlxsw_get_cooling_device_idx(struct mlxsw_thermal *thermal,
 	return -ENODEV;
 }
 
-static void
-mlxsw_thermal_module_trips_reset(struct mlxsw_thermal_module *tz)
-{
-	tz->trips[MLXSW_THERMAL_TEMP_TRIP_NORM].temperature = 0;
-	tz->trips[MLXSW_THERMAL_TEMP_TRIP_HIGH].temperature = 0;
-	tz->trips[MLXSW_THERMAL_TEMP_TRIP_HOT].temperature = 0;
-}
-
-static int
-mlxsw_thermal_module_trips_update(struct device *dev, struct mlxsw_core *core,
-				  struct mlxsw_thermal_module *tz,
-				  int crit_temp, int emerg_temp)
-{
-	int err;
-
-	/* Do not try to query temperature thresholds directly from the module's
-	 * EEPROM if we got valid thresholds from MTMP.
-	 */
-	if (!emerg_temp || !crit_temp) {
-		err = mlxsw_env_module_temp_thresholds_get(core, tz->slot_index,
-							   tz->module,
-							   SFP_TEMP_HIGH_WARN,
-							   &crit_temp);
-		if (err)
-			return err;
-
-		err = mlxsw_env_module_temp_thresholds_get(core, tz->slot_index,
-							   tz->module,
-							   SFP_TEMP_HIGH_ALARM,
-							   &emerg_temp);
-		if (err)
-			return err;
-	}
-
-	if (crit_temp > emerg_temp) {
-		dev_warn(dev, "%s : Critical threshold %d is above emergency threshold %d\n",
-			 tz->tzdev->type, crit_temp, emerg_temp);
-		return 0;
-	}
-
-	/* According to the system thermal requirements, the thermal zones are
-	 * defined with three trip points. The critical and emergency
-	 * temperature thresholds, provided by QSFP module are set as "active"
-	 * and "hot" trip points, "normal" trip point is derived from "active"
-	 * by subtracting double hysteresis value.
-	 */
-	if (crit_temp >= MLXSW_THERMAL_MODULE_TEMP_SHIFT)
-		tz->trips[MLXSW_THERMAL_TEMP_TRIP_NORM].temperature = crit_temp -
-					MLXSW_THERMAL_MODULE_TEMP_SHIFT;
-	else
-		tz->trips[MLXSW_THERMAL_TEMP_TRIP_NORM].temperature = crit_temp;
-	tz->trips[MLXSW_THERMAL_TEMP_TRIP_HIGH].temperature = crit_temp;
-	tz->trips[MLXSW_THERMAL_TEMP_TRIP_HOT].temperature = emerg_temp;
-
-	return 0;
-}
-
 static int mlxsw_thermal_bind(struct thermal_zone_device *tzdev,
 			      struct thermal_cooling_device *cdev)
 {
@@ -358,10 +316,8 @@ static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 	struct mlxsw_thermal_module *tz = tzdev->devdata;
 	struct mlxsw_thermal *thermal = tz->parent;
 	int temp, crit_temp, emerg_temp;
-	struct device *dev;
 	u16 sensor_index;
 
-	dev = thermal->bus_info->dev;
 	sensor_index = MLXSW_REG_MTMP_MODULE_INDEX_MIN + tz->module;
 
 	/* Read module temperature and thresholds. */
@@ -371,13 +327,6 @@ static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 						     &crit_temp, &emerg_temp);
 	*p_temp = temp;
 
-	if (!temp)
-		return 0;
-
-	/* Update trip points. */
-	mlxsw_thermal_module_trips_update(dev, thermal->core, tz,
-					  crit_temp, emerg_temp);
-
 	return 0;
 }
 
@@ -527,10 +476,7 @@ mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 			  struct mlxsw_thermal_area *area, u8 module)
 {
 	struct mlxsw_thermal_module *module_tz;
-	int dummy_temp, crit_temp, emerg_temp;
-	u16 sensor_index;
 
-	sensor_index = MLXSW_REG_MTMP_MODULE_INDEX_MIN + module;
 	module_tz = &area->tz_module_arr[module];
 	/* Skip if parent is already set (case of port split). */
 	if (module_tz->parent)
@@ -538,19 +484,13 @@ mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 	module_tz->module = module;
 	module_tz->slot_index = area->slot_index;
 	module_tz->parent = thermal;
-	memcpy(module_tz->trips, default_thermal_trips,
+	BUILD_BUG_ON(ARRAY_SIZE(default_thermal_module_trips) !=
+		     MLXSW_THERMAL_NUM_TRIPS);
+	memcpy(module_tz->trips, default_thermal_module_trips,
 	       sizeof(thermal->trips));
 	memcpy(module_tz->cooling_states, default_cooling_states,
 	       sizeof(thermal->cooling_states));
-	/* Initialize all trip point. */
-	mlxsw_thermal_module_trips_reset(module_tz);
-	/* Read module temperature and thresholds. */
-	mlxsw_thermal_module_temp_and_thresholds_get(core, area->slot_index,
-						     sensor_index, &dummy_temp,
-						     &crit_temp, &emerg_temp);
-	/* Update trip point according to the module data. */
-	return mlxsw_thermal_module_trips_update(dev, core, module_tz,
-						 crit_temp, emerg_temp);
+	return 0;
 }
 
 static void mlxsw_thermal_module_fini(struct mlxsw_thermal_module *module_tz)
-- 
cgit v1.2.3-70-g09d2


From c1536d856e18c6d37a6b7d1fc68d3d27aa7f8e9d Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Fri, 31 Mar 2023 16:17:31 +0200
Subject: mlxsw: core_thermal: Make mlxsw_thermal_module_init() void

The function can no longer fail so make it void and remove the
associated error path.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/core_thermal.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index ece5075b7dbf..f0c5a2c59075 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -470,7 +470,7 @@ static void mlxsw_thermal_module_tz_fini(struct thermal_zone_device *tzdev)
 	thermal_zone_device_unregister(tzdev);
 }
 
-static int
+static void
 mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 			  struct mlxsw_thermal *thermal,
 			  struct mlxsw_thermal_area *area, u8 module)
@@ -480,7 +480,7 @@ mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 	module_tz = &area->tz_module_arr[module];
 	/* Skip if parent is already set (case of port split). */
 	if (module_tz->parent)
-		return 0;
+		return;
 	module_tz->module = module;
 	module_tz->slot_index = area->slot_index;
 	module_tz->parent = thermal;
@@ -490,7 +490,6 @@ mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 	       sizeof(thermal->trips));
 	memcpy(module_tz->cooling_states, default_cooling_states,
 	       sizeof(thermal->cooling_states));
-	return 0;
 }
 
 static void mlxsw_thermal_module_fini(struct mlxsw_thermal_module *module_tz)
@@ -529,11 +528,8 @@ mlxsw_thermal_modules_init(struct device *dev, struct mlxsw_core *core,
 	if (!area->tz_module_arr)
 		return -ENOMEM;
 
-	for (i = 0; i < area->tz_module_num; i++) {
-		err = mlxsw_thermal_module_init(dev, core, thermal, area, i);
-		if (err)
-			goto err_thermal_module_init;
-	}
+	for (i = 0; i < area->tz_module_num; i++)
+		mlxsw_thermal_module_init(dev, core, thermal, area, i);
 
 	for (i = 0; i < area->tz_module_num; i++) {
 		module_tz = &area->tz_module_arr[i];
@@ -547,7 +543,6 @@ mlxsw_thermal_modules_init(struct device *dev, struct mlxsw_core *core,
 	return 0;
 
 err_thermal_module_tz_init:
-err_thermal_module_init:
 	for (i = area->tz_module_num - 1; i >= 0; i--)
 		mlxsw_thermal_module_fini(&area->tz_module_arr[i]);
 	kfree(area->tz_module_arr);
-- 
cgit v1.2.3-70-g09d2


From cc19439f703b8c979133bf91690d6c0dffd8dc46 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Fri, 31 Mar 2023 16:17:32 +0200
Subject: mlxsw: core_thermal: Simplify transceiver module get_temp() callback

The get_temp() callback of a thermal zone associated with a transceiver
module no longer needs to read the temperature thresholds of the module.
Therefore, simplify the callback by only reading the temperature.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/core_thermal.c | 44 ++++------------------
 1 file changed, 8 insertions(+), 36 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index f0c5a2c59075..deac4bced98c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -283,50 +283,22 @@ static int mlxsw_thermal_module_unbind(struct thermal_zone_device *tzdev,
 	return err;
 }
 
-static void
-mlxsw_thermal_module_temp_and_thresholds_get(struct mlxsw_core *core,
-					     u8 slot_index, u16 sensor_index,
-					     int *p_temp, int *p_crit_temp,
-					     int *p_emerg_temp)
-{
-	char mtmp_pl[MLXSW_REG_MTMP_LEN];
-	int err;
-
-	/* Read module temperature and thresholds. */
-	mlxsw_reg_mtmp_pack(mtmp_pl, slot_index, sensor_index,
-			    false, false);
-	err = mlxsw_reg_query(core, MLXSW_REG(mtmp), mtmp_pl);
-	if (err) {
-		/* Set temperature and thresholds to zero to avoid passing
-		 * uninitialized data back to the caller.
-		 */
-		*p_temp = 0;
-		*p_crit_temp = 0;
-		*p_emerg_temp = 0;
-
-		return;
-	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, p_temp, NULL, p_crit_temp, p_emerg_temp,
-			      NULL);
-}
-
 static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 					 int *p_temp)
 {
 	struct mlxsw_thermal_module *tz = tzdev->devdata;
 	struct mlxsw_thermal *thermal = tz->parent;
-	int temp, crit_temp, emerg_temp;
+	char mtmp_pl[MLXSW_REG_MTMP_LEN];
 	u16 sensor_index;
+	int err;
 
 	sensor_index = MLXSW_REG_MTMP_MODULE_INDEX_MIN + tz->module;
-
-	/* Read module temperature and thresholds. */
-	mlxsw_thermal_module_temp_and_thresholds_get(thermal->core,
-						     tz->slot_index,
-						     sensor_index, &temp,
-						     &crit_temp, &emerg_temp);
-	*p_temp = temp;
-
+	mlxsw_reg_mtmp_pack(mtmp_pl, tz->slot_index, sensor_index,
+			    false, false);
+	err = mlxsw_reg_query(thermal->core, MLXSW_REG(mtmp), mtmp_pl);
+	if (err)
+		return err;
+	mlxsw_reg_mtmp_unpack(mtmp_pl, p_temp, NULL, NULL, NULL, NULL);
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 51aaa68222d6c34f0373cf95223ce2f230329e8f Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Fri, 31 Mar 2023 16:55:45 -0400
Subject: net: alteon: remove unused len variable

clang with W=1 reports
drivers/net/ethernet/alteon/acenic.c:2438:10: error: variable
  'len' set but not used [-Werror,-Wunused-but-set-variable]
                int i, len = 0;
                       ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/alteon/acenic.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c
index d7762da8b2c0..eafef84fe3be 100644
--- a/drivers/net/ethernet/alteon/acenic.c
+++ b/drivers/net/ethernet/alteon/acenic.c
@@ -2435,7 +2435,7 @@ restart:
 	} else {
 		dma_addr_t mapping;
 		u32 vlan_tag = 0;
-		int i, len = 0;
+		int i;
 
 		mapping = ace_map_tx_skb(ap, skb, NULL, idx);
 		flagsize = (skb_headlen(skb) << 16);
@@ -2454,7 +2454,6 @@ restart:
 			const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 			struct tx_ring_info *info;
 
-			len += skb_frag_size(frag);
 			info = ap->skb->tx_skbuff + idx;
 			desc = ap->tx_ring + idx;
 
-- 
cgit v1.2.3-70-g09d2


From 92b2e810f0d3a2c05d8cf12a800592b238d458df Mon Sep 17 00:00:00 2001
From: Anton Protopopov <aspsk@isovalent.com>
Date: Sun, 2 Apr 2023 11:43:40 +0000
Subject: bpf: compute hashes in bloom filter similar to hashmap

If the value size in a bloom filter is a multiple of 4, then the jhash2()
function is used to compute hashes. The length parameter of this function
equals to the number of 32-bit words in input. Compute it in the hot path
instead of pre-computing it, as this is translated to one extra shift to
divide the length by four vs. one extra memory load of a pre-computed length.

Signed-off-by: Anton Protopopov <aspsk@isovalent.com>
Link: https://lore.kernel.org/r/20230402114340.3441-1-aspsk@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bloom_filter.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

diff --git a/kernel/bpf/bloom_filter.c b/kernel/bpf/bloom_filter.c
index db19784601a7..540331b610a9 100644
--- a/kernel/bpf/bloom_filter.c
+++ b/kernel/bpf/bloom_filter.c
@@ -16,13 +16,6 @@ struct bpf_bloom_filter {
 	struct bpf_map map;
 	u32 bitset_mask;
 	u32 hash_seed;
-	/* If the size of the values in the bloom filter is u32 aligned,
-	 * then it is more performant to use jhash2 as the underlying hash
-	 * function, else we use jhash. This tracks the number of u32s
-	 * in an u32-aligned value size. If the value size is not u32 aligned,
-	 * this will be 0.
-	 */
-	u32 aligned_u32_count;
 	u32 nr_hash_funcs;
 	unsigned long bitset[];
 };
@@ -32,9 +25,8 @@ static u32 hash(struct bpf_bloom_filter *bloom, void *value,
 {
 	u32 h;
 
-	if (bloom->aligned_u32_count)
-		h = jhash2(value, bloom->aligned_u32_count,
-			   bloom->hash_seed + index);
+	if (likely(value_size % 4 == 0))
+		h = jhash2(value, value_size / 4, bloom->hash_seed + index);
 	else
 		h = jhash(value, value_size, bloom->hash_seed + index);
 
@@ -152,11 +144,6 @@ static struct bpf_map *bloom_map_alloc(union bpf_attr *attr)
 	bloom->nr_hash_funcs = nr_hash_funcs;
 	bloom->bitset_mask = bitset_mask;
 
-	/* Check whether the value size is u32-aligned */
-	if ((attr->value_size & (sizeof(u32) - 1)) == 0)
-		bloom->aligned_u32_count =
-			attr->value_size / sizeof(u32);
-
 	if (!(attr->map_flags & BPF_F_ZERO_SEED))
 		bloom->hash_seed = get_random_u32();
 
-- 
cgit v1.2.3-70-g09d2


From 16b7c970cc8192e929dbd5192ccc1867e19d7bda Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler@microsoft.com>
Date: Sun, 26 Mar 2023 05:49:46 +0000
Subject: bpf, docs: Add docs on extended 64-bit immediate instructions

Add docs on extended 64-bit immediate instructions, including six instructions
previously undocumented.  Include a brief description of maps and variables,
as used by those instructions.

V1 -> V2: rebased on top of latest master

V2 -> V3: addressed comments from Alexei

V3 -> V4: addressed comments from David Vernet

Signed-off-by: Dave Thaler <dthaler@microsoft.com>
Link: https://lore.kernel.org/r/20230326054946.2331-1-dthaler1968@googlemail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/instruction-set.rst | 58 ++++++++++++++++++++++++++++++-----
 Documentation/bpf/linux-notes.rst     | 22 +++++++++++++
 2 files changed, 72 insertions(+), 8 deletions(-)

diff --git a/Documentation/bpf/instruction-set.rst b/Documentation/bpf/instruction-set.rst
index b77280eb926f..492980ece1ab 100644
--- a/Documentation/bpf/instruction-set.rst
+++ b/Documentation/bpf/instruction-set.rst
@@ -416,14 +416,56 @@ and loaded back to ``R0``.
 -----------------------------
 
 Instructions with the ``BPF_IMM`` 'mode' modifier use the wide instruction
-encoding for an extra imm64 value.
-
-There is currently only one such instruction.
-
-``BPF_LD | BPF_DW | BPF_IMM`` means::
-
-  dst = imm64
-
+encoding defined in `Instruction encoding`_, and use the 'src' field of the
+basic instruction to hold an opcode subtype.
+
+The following table defines a set of ``BPF_IMM | BPF_DW | BPF_LD`` instructions
+with opcode subtypes in the 'src' field, using new terms such as "map"
+defined further below:
+
+=========================  ======  ===  =========================================  ===========  ==============
+opcode construction        opcode  src  pseudocode                                 imm type     dst type
+=========================  ======  ===  =========================================  ===========  ==============
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x0  dst = imm64                                integer      integer
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x1  dst = map_by_fd(imm)                       map fd       map
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x2  dst = map_val(map_by_fd(imm)) + next_imm   map fd       data pointer
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x3  dst = var_addr(imm)                        variable id  data pointer
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x4  dst = code_addr(imm)                       integer      code pointer
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x5  dst = map_by_idx(imm)                      map index    map
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x6  dst = map_val(map_by_idx(imm)) + next_imm  map index    data pointer
+=========================  ======  ===  =========================================  ===========  ==============
+
+where
+
+* map_by_fd(imm) means to convert a 32-bit file descriptor into an address of a map (see `Maps`_)
+* map_by_idx(imm) means to convert a 32-bit index into an address of a map
+* map_val(map) gets the address of the first value in a given map
+* var_addr(imm) gets the address of a platform variable (see `Platform Variables`_) with a given id
+* code_addr(imm) gets the address of the instruction at a specified relative offset in number of (64-bit) instructions
+* the 'imm type' can be used by disassemblers for display
+* the 'dst type' can be used for verification and JIT compilation purposes
+
+Maps
+~~~~
+
+Maps are shared memory regions accessible by eBPF programs on some platforms.
+A map can have various semantics as defined in a separate document, and may or
+may not have a single contiguous memory region, but the 'map_val(map)' is
+currently only defined for maps that do have a single contiguous memory region.
+
+Each map can have a file descriptor (fd) if supported by the platform, where
+'map_by_fd(imm)' means to get the map with the specified file descriptor. Each
+BPF program can also be defined to use a set of maps associated with the
+program at load time, and 'map_by_idx(imm)' means to get the map with the given
+index in the set associated with the BPF program containing the instruction.
+
+Platform Variables
+~~~~~~~~~~~~~~~~~~
+
+Platform variables are memory regions, identified by integer ids, exposed by
+the runtime and accessible by BPF programs on some platforms.  The
+'var_addr(imm)' operation means to get the address of the memory region
+identified by the given id.
 
 Legacy BPF Packet access instructions
 -------------------------------------
diff --git a/Documentation/bpf/linux-notes.rst b/Documentation/bpf/linux-notes.rst
index f43b9c797bcb..508d009d3bed 100644
--- a/Documentation/bpf/linux-notes.rst
+++ b/Documentation/bpf/linux-notes.rst
@@ -20,6 +20,28 @@ integer would be read from a specified register, is not currently supported
 by the verifier.  Any programs with this instruction will fail to load
 until such support is added.
 
+Maps
+====
+
+Linux only supports the 'map_val(map)' operation on array maps with a single element.
+
+Linux uses an fd_array to store maps associated with a BPF program. Thus,
+map_by_idx(imm) uses the fd at that index in the array.
+
+Variables
+=========
+
+The following 64-bit immediate instruction specifies that a variable address,
+which corresponds to some integer stored in the 'imm' field, should be loaded:
+
+=========================  ======  ===  =========================================  ===========  ==============
+opcode construction        opcode  src  pseudocode                                 imm type     dst type
+=========================  ======  ===  =========================================  ===========  ==============
+BPF_IMM | BPF_DW | BPF_LD  0x18    0x3  dst = var_addr(imm)                        variable id  data pointer
+=========================  ======  ===  =========================================  ===========  ==============
+
+On Linux, this integer is a BTF ID.
+
 Legacy BPF Packet access instructions
 =====================================
 
-- 
cgit v1.2.3-70-g09d2


From 509541b440e64e155201dd304411e2daa152fe8f Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:22 +0300
Subject: net/mlx5e: Factor out IPsec ASO update function

The ASO update is common operation which is going to be used in next
patch, so as a preparation, let's refactor the code for future reuse.

As part of this refactoring, not used function argument was removed too.

Link: https://lore.kernel.org/r/d04770b959822fed51c22c13e798f04d760a682e.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    | 27 +++++++++++-----------
 1 file changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 5342b0b07681..43cfa4df1311 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -275,26 +275,21 @@ void mlx5_accel_esp_modify_xfrm(struct mlx5e_ipsec_sa_entry *sa_entry,
 	memcpy(&sa_entry->attrs, attrs, sizeof(sa_entry->attrs));
 }
 
-static void
-mlx5e_ipsec_aso_update_esn(struct mlx5e_ipsec_sa_entry *sa_entry,
-			   const struct mlx5_accel_esp_xfrm_attrs *attrs)
+static void mlx5e_ipsec_aso_update(struct mlx5e_ipsec_sa_entry *sa_entry,
+				   struct mlx5_wqe_aso_ctrl_seg *data)
 {
-	struct mlx5_wqe_aso_ctrl_seg data = {};
+	data->data_mask_mode = MLX5_ASO_DATA_MASK_MODE_BITWISE_64BIT << 6;
+	data->condition_1_0_operand = MLX5_ASO_ALWAYS_TRUE |
+				      MLX5_ASO_ALWAYS_TRUE << 4;
 
-	data.data_mask_mode = MLX5_ASO_DATA_MASK_MODE_BITWISE_64BIT << 6;
-	data.condition_1_0_operand = MLX5_ASO_ALWAYS_TRUE | MLX5_ASO_ALWAYS_TRUE
-								    << 4;
-	data.data_offset_condition_operand = MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET;
-	data.bitwise_data = cpu_to_be64(BIT_ULL(54));
-	data.data_mask = data.bitwise_data;
-
-	mlx5e_ipsec_aso_query(sa_entry, &data);
+	mlx5e_ipsec_aso_query(sa_entry, data);
 }
 
 static void mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry,
 					 u32 mode_param)
 {
 	struct mlx5_accel_esp_xfrm_attrs attrs = {};
+	struct mlx5_wqe_aso_ctrl_seg data = {};
 
 	if (mode_param < MLX5E_IPSEC_ESN_SCOPE_MID) {
 		sa_entry->esn_state.esn++;
@@ -305,7 +300,13 @@ static void mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry,
 
 	mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &attrs);
 	mlx5_accel_esp_modify_xfrm(sa_entry, &attrs);
-	mlx5e_ipsec_aso_update_esn(sa_entry, &attrs);
+
+	data.data_offset_condition_operand =
+		MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET;
+	data.bitwise_data = cpu_to_be64(BIT_ULL(54));
+	data.data_mask = data.bitwise_data;
+
+	mlx5e_ipsec_aso_update(sa_entry, &data);
 }
 
 static void mlx5e_ipsec_handle_event(struct work_struct *_work)
-- 
cgit v1.2.3-70-g09d2


From 2da961d21663a72610f36a4f69b1090f2c61f23f Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:23 +0300
Subject: net/mlx5e: Prevent zero IPsec soft/hard limits

Hardware triggers limit events when the packets arrive and are processed
through the device. In case zero was configured as a limit, the HW won't
be able to arm event as it happens at the end of execution pipeline.

Let's prevent such configuration.

Link: https://lore.kernel.org/r/80d0ba33e21fb28b1b91d306d1da39df3d990b68.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 91fa0a366316..c2e4f30d1f76 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -283,6 +283,11 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 			NL_SET_ERR_MSG_MOD(extack, "Hard packet limit must be greater than soft one");
 			return -EINVAL;
 		}
+
+		if (!x->lft.soft_packet_limit || !x->lft.hard_packet_limit) {
+			NL_SET_ERR_MSG_MOD(extack, "Soft/hard packet limits can't be 0");
+			return -EINVAL;
+		}
 		break;
 	default:
 		NL_SET_ERR_MSG_MOD(extack, "Unsupported xfrm offload type");
-- 
cgit v1.2.3-70-g09d2


From d05971a413d3c3dd207c604a6dbfc702070da63e Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:24 +0300
Subject: net/mlx5e: Add SW implementation to support IPsec 64 bit soft and
 hard limits

The CX7 cards which support IPsec packet offload use 32 bits to
configure soft and hard packet limits. This is not enough as the
software part using 64 bits.

The needed functionality of supporting 64 bits is implemented through
mlx5 abstraction layer, which will ensure that HW is reconfigured
on-demand every 2^31 packets.

To simulate the 64 bit IPsec soft/hard limits, we divide the soft/hard
limits to multiple interrupts (rounds). Each round counts 2^31 packets.
Once the counter is less than or equal to 2^31, the soft event is raised
and software sets the bit 31 of the counter and decrement the round
counter.

Link: https://lore.kernel.org/r/5a86c890b6dccb6865acf9042a8b03f899d1f3f9.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 114 +++++++++++++++++---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  17 ++-
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    | 119 +++++++++++++++++++--
 3 files changed, 227 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index c2e4f30d1f76..3612cdd37b5a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -87,25 +87,113 @@ static void mlx5e_ipsec_init_limits(struct mlx5e_ipsec_sa_entry *sa_entry,
 				    struct mlx5_accel_esp_xfrm_attrs *attrs)
 {
 	struct xfrm_state *x = sa_entry->x;
+	s64 start_value, n;
 
-	attrs->hard_packet_limit = x->lft.hard_packet_limit;
+	attrs->lft.hard_packet_limit = x->lft.hard_packet_limit;
+	attrs->lft.soft_packet_limit = x->lft.soft_packet_limit;
 	if (x->lft.soft_packet_limit == XFRM_INF)
 		return;
 
-	/* Hardware decrements hard_packet_limit counter through
-	 * the operation. While fires an event when soft_packet_limit
-	 * is reached. It emans that we need substitute the numbers
-	 * in order to properly count soft limit.
+	/* Compute hard limit initial value and number of rounds.
 	 *
-	 * As an example:
-	 * XFRM user sets soft limit is 2 and hard limit is 9 and
-	 * expects to see soft event after 2 packets and hard event
-	 * after 9 packets. In our case, the hard limit will be set
-	 * to 9 and soft limit is comparator to 7 so user gets the
-	 * soft event after 2 packeta
+	 * The counting pattern of hardware counter goes:
+	 *                value  -> 2^31-1
+	 *      2^31  | (2^31-1) -> 2^31-1
+	 *      2^31  | (2^31-1) -> 2^31-1
+	 *      [..]
+	 *      2^31  | (2^31-1) -> 0
+	 *
+	 * The pattern is created by using an ASO operation to atomically set
+	 * bit 31 after the down counter clears bit 31. This is effectively an
+	 * atomic addition of 2**31 to the counter.
+	 *
+	 * We wish to configure the counter, within the above pattern, so that
+	 * when it reaches 0, it has hit the hard limit. This is defined by this
+	 * system of equations:
+	 *
+	 *      hard_limit == start_value + n * 2^31
+	 *      n >= 0
+	 *      start_value < 2^32, start_value >= 0
+	 *
+	 * These equations are not single-solution, there are often two choices:
+	 *      hard_limit == start_value + n * 2^31
+	 *      hard_limit == (start_value+2^31) + (n-1) * 2^31
+	 *
+	 * The algorithm selects the solution that keeps the counter value
+	 * above 2^31 until the final iteration.
+	 */
+
+	/* Start by estimating n and compute start_value */
+	n = attrs->lft.hard_packet_limit / BIT_ULL(31);
+	start_value = attrs->lft.hard_packet_limit - n * BIT_ULL(31);
+
+	/* Choose the best of the two solutions: */
+	if (n >= 1)
+		n -= 1;
+
+	/* Computed values solve the system of equations: */
+	start_value = attrs->lft.hard_packet_limit - n * BIT_ULL(31);
+
+	/* The best solution means: when there are multiple iterations we must
+	 * start above 2^31 and count down to 2**31 to get the interrupt.
+	 */
+	attrs->lft.hard_packet_limit = lower_32_bits(start_value);
+	attrs->lft.numb_rounds_hard = (u64)n;
+
+	/* Compute soft limit initial value and number of rounds.
+	 *
+	 * The soft_limit is achieved by adjusting the counter's
+	 * interrupt_value. This is embedded in the counting pattern created by
+	 * hard packet calculations above.
+	 *
+	 * We wish to compute the interrupt_value for the soft_limit. This is
+	 * defined by this system of equations:
+	 *
+	 *      soft_limit == start_value - soft_value + n * 2^31
+	 *      n >= 0
+	 *      soft_value < 2^32, soft_value >= 0
+	 *      for n == 0 start_value > soft_value
+	 *
+	 * As with compute_hard_n_value() the equations are not single-solution.
+	 * The algorithm selects the solution that has:
+	 *      2^30 <= soft_limit < 2^31 + 2^30
+	 * for the interior iterations, which guarantees a large guard band
+	 * around the counter hard limit and next interrupt.
+	 */
+
+	/* Start by estimating n and compute soft_value */
+	n = (x->lft.soft_packet_limit - attrs->lft.hard_packet_limit) / BIT_ULL(31);
+	start_value = attrs->lft.hard_packet_limit + n * BIT_ULL(31) -
+		      x->lft.soft_packet_limit;
+
+	/* Compare against constraints and adjust n */
+	if (n < 0)
+		n = 0;
+	else if (start_value >= BIT_ULL(32))
+		n -= 1;
+	else if (start_value < 0)
+		n += 1;
+
+	/* Choose the best of the two solutions: */
+	start_value = attrs->lft.hard_packet_limit + n * BIT_ULL(31) - start_value;
+	if (n != attrs->lft.numb_rounds_hard && start_value < BIT_ULL(30))
+		n += 1;
+
+	/* Note that the upper limit of soft_value happens naturally because we
+	 * always select the lowest soft_value.
+	 */
+
+	/* Computed values solve the system of equations: */
+	start_value = attrs->lft.hard_packet_limit + n * BIT_ULL(31) - start_value;
+
+	/* The best solution means: when there are multiple iterations we must
+	 * not fall below 2^30 as that would get too close to the false
+	 * hard_limit and when we reach an interior iteration for soft_limit it
+	 * has to be far away from 2**32-1 which is the counter reset point
+	 * after the +2^31 to accommodate latency.
 	 */
-	attrs->soft_packet_limit =
-		x->lft.hard_packet_limit - x->lft.soft_packet_limit;
+	attrs->lft.soft_packet_limit = lower_32_bits(start_value);
+	attrs->lft.numb_rounds_soft = (u64)n;
 }
 
 void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 68ae5230eb75..0c58c3583b0f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -60,6 +60,13 @@ struct upspec {
 	u8 proto;
 };
 
+struct mlx5_ipsec_lft {
+	u64 hard_packet_limit;
+	u64 soft_packet_limit;
+	u64 numb_rounds_hard;
+	u64 numb_rounds_soft;
+};
+
 struct mlx5_accel_esp_xfrm_attrs {
 	u32   esn;
 	u32   spi;
@@ -85,8 +92,7 @@ struct mlx5_accel_esp_xfrm_attrs {
 	u32 replay_window;
 	u32 authsize;
 	u32 reqid;
-	u64 hard_packet_limit;
-	u64 soft_packet_limit;
+	struct mlx5_ipsec_lft lft;
 };
 
 enum mlx5_ipsec_cap {
@@ -170,6 +176,12 @@ struct mlx5e_ipsec_modify_state_work {
 	struct mlx5_accel_esp_xfrm_attrs attrs;
 };
 
+struct mlx5e_ipsec_limits {
+	u64 round;
+	u8 soft_limit_hit : 1;
+	u8 fix_limit : 1;
+};
+
 struct mlx5e_ipsec_sa_entry {
 	struct mlx5e_ipsec_esn_state esn_state;
 	struct xfrm_state *x;
@@ -181,6 +193,7 @@ struct mlx5e_ipsec_sa_entry {
 	u32 enc_key_id;
 	struct mlx5e_ipsec_rule ipsec_rule;
 	struct mlx5e_ipsec_modify_state_work modify_work;
+	struct mlx5e_ipsec_limits limits;
 };
 
 struct mlx5_accel_pol_xfrm_attrs {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 43cfa4df1311..684de9739e69 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -8,6 +8,7 @@
 
 enum {
 	MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET,
+	MLX5_IPSEC_ASO_REMOVE_FLOW_SOFT_LFT_OFFSET,
 };
 
 u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev)
@@ -100,15 +101,15 @@ static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn,
 	if (attrs->dir == XFRM_DEV_OFFLOAD_OUT)
 		MLX5_SET(ipsec_aso, aso_ctx, mode, MLX5_IPSEC_ASO_INC_SN);
 
-	if (attrs->hard_packet_limit != XFRM_INF) {
+	if (attrs->lft.hard_packet_limit != XFRM_INF) {
 		MLX5_SET(ipsec_aso, aso_ctx, remove_flow_pkt_cnt,
-			 lower_32_bits(attrs->hard_packet_limit));
+			 attrs->lft.hard_packet_limit);
 		MLX5_SET(ipsec_aso, aso_ctx, hard_lft_arm, 1);
 	}
 
-	if (attrs->soft_packet_limit != XFRM_INF) {
+	if (attrs->lft.soft_packet_limit != XFRM_INF) {
 		MLX5_SET(ipsec_aso, aso_ctx, remove_flow_soft_lft,
-			 lower_32_bits(attrs->soft_packet_limit));
+			 attrs->lft.soft_packet_limit);
 
 		MLX5_SET(ipsec_aso, aso_ctx, soft_lft_arm, 1);
 	}
@@ -309,6 +310,110 @@ static void mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry,
 	mlx5e_ipsec_aso_update(sa_entry, &data);
 }
 
+static void mlx5e_ipsec_aso_update_hard(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct mlx5_wqe_aso_ctrl_seg data = {};
+
+	data.data_offset_condition_operand =
+		MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET;
+	data.bitwise_data = cpu_to_be64(BIT_ULL(57) + BIT_ULL(31));
+	data.data_mask = data.bitwise_data;
+	mlx5e_ipsec_aso_update(sa_entry, &data);
+}
+
+static void mlx5e_ipsec_aso_update_soft(struct mlx5e_ipsec_sa_entry *sa_entry,
+					u32 val)
+{
+	struct mlx5_wqe_aso_ctrl_seg data = {};
+
+	data.data_offset_condition_operand =
+		MLX5_IPSEC_ASO_REMOVE_FLOW_SOFT_LFT_OFFSET;
+	data.bitwise_data = cpu_to_be64(val);
+	data.data_mask = cpu_to_be64(U32_MAX);
+	mlx5e_ipsec_aso_update(sa_entry, &data);
+}
+
+static void mlx5e_ipsec_handle_limits(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs;
+	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
+	struct mlx5e_ipsec_aso *aso = ipsec->aso;
+	bool soft_arm, hard_arm;
+	u64 hard_cnt;
+
+	lockdep_assert_held(&sa_entry->x->lock);
+
+	soft_arm = !MLX5_GET(ipsec_aso, aso->ctx, soft_lft_arm);
+	hard_arm = !MLX5_GET(ipsec_aso, aso->ctx, hard_lft_arm);
+	if (!soft_arm && !hard_arm)
+		/* It is not lifetime event */
+		return;
+
+	hard_cnt = MLX5_GET(ipsec_aso, aso->ctx, remove_flow_pkt_cnt);
+	if (!hard_cnt || hard_arm) {
+		/* It is possible to see packet counter equal to zero without
+		 * hard limit event armed. Such situation can be if packet
+		 * decreased, while we handled soft limit event.
+		 *
+		 * However it will be HW/FW bug if hard limit event is raised
+		 * and packet counter is not zero.
+		 */
+		WARN_ON_ONCE(hard_arm && hard_cnt);
+
+		/* Notify about hard limit */
+		xfrm_state_check_expire(sa_entry->x);
+		return;
+	}
+
+	/* We are in soft limit event. */
+	if (!sa_entry->limits.soft_limit_hit &&
+	    sa_entry->limits.round == attrs->lft.numb_rounds_soft) {
+		sa_entry->limits.soft_limit_hit = true;
+		/* Notify about soft limit */
+		xfrm_state_check_expire(sa_entry->x);
+
+		if (sa_entry->limits.round == attrs->lft.numb_rounds_hard)
+			goto hard;
+
+		if (attrs->lft.soft_packet_limit > BIT_ULL(31)) {
+			/* We cannot avoid a soft_value that might have the high
+			 * bit set. For instance soft_value=2^31+1 cannot be
+			 * adjusted to the low bit clear version of soft_value=1
+			 * because it is too close to 0.
+			 *
+			 * Thus we have this corner case where we can hit the
+			 * soft_limit with the high bit set, but cannot adjust
+			 * the counter. Thus we set a temporary interrupt_value
+			 * at least 2^30 away from here and do the adjustment
+			 * then.
+			 */
+			mlx5e_ipsec_aso_update_soft(sa_entry,
+						    BIT_ULL(31) - BIT_ULL(30));
+			sa_entry->limits.fix_limit = true;
+			return;
+		}
+
+		sa_entry->limits.fix_limit = true;
+	}
+
+hard:
+	if (sa_entry->limits.round == attrs->lft.numb_rounds_hard) {
+		mlx5e_ipsec_aso_update_soft(sa_entry, 0);
+		attrs->lft.soft_packet_limit = XFRM_INF;
+		return;
+	}
+
+	mlx5e_ipsec_aso_update_hard(sa_entry);
+	sa_entry->limits.round++;
+	if (sa_entry->limits.round == attrs->lft.numb_rounds_soft)
+		mlx5e_ipsec_aso_update_soft(sa_entry,
+					    attrs->lft.soft_packet_limit);
+	if (sa_entry->limits.fix_limit) {
+		sa_entry->limits.fix_limit = false;
+		mlx5e_ipsec_aso_update_soft(sa_entry, BIT_ULL(31) - 1);
+	}
+}
+
 static void mlx5e_ipsec_handle_event(struct work_struct *_work)
 {
 	struct mlx5e_ipsec_work *work =
@@ -339,10 +444,8 @@ static void mlx5e_ipsec_handle_event(struct work_struct *_work)
 		mlx5e_ipsec_update_esn_state(sa_entry, mode_param);
 	}
 
-	if (attrs->soft_packet_limit != XFRM_INF)
-		if (!MLX5_GET(ipsec_aso, aso->ctx, soft_lft_arm) ||
-		    !MLX5_GET(ipsec_aso, aso->ctx, hard_lft_arm))
-			xfrm_state_check_expire(sa_entry->x);
+	if (attrs->lft.soft_packet_limit != XFRM_INF)
+		mlx5e_ipsec_handle_limits(sa_entry);
 
 unlock:
 	spin_unlock(&sa_entry->x->lock);
-- 
cgit v1.2.3-70-g09d2


From 76e463f6508b5909887df3c360ca55f1caa3b94a Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:25 +0300
Subject: net/mlx5e: Overcome slow response for first IPsec ASO WQE

First ASO WQE causes to cache miss in hardware, which can't return
result immediately. It causes to the situation where such WQE is polled
earlier than it is needed. Add logic to retry ASO CQ polling operation.

Link: https://lore.kernel.org/r/eb92a758c533ff3f058e0dcb4f8d2324355304ad.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 684de9739e69..6971e5e36820 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -568,6 +568,7 @@ int mlx5e_ipsec_aso_query(struct mlx5e_ipsec_sa_entry *sa_entry,
 	struct mlx5_wqe_aso_ctrl_seg *ctrl;
 	struct mlx5e_hw_objs *res;
 	struct mlx5_aso_wqe *wqe;
+	unsigned long expires;
 	u8 ds_cnt;
 	int ret;
 
@@ -589,7 +590,12 @@ int mlx5e_ipsec_aso_query(struct mlx5e_ipsec_sa_entry *sa_entry,
 	mlx5e_ipsec_aso_copy(ctrl, data);
 
 	mlx5_aso_post_wqe(aso->aso, false, &wqe->ctrl);
-	ret = mlx5_aso_poll_cq(aso->aso, false);
+	expires = jiffies + msecs_to_jiffies(10);
+	do {
+		ret = mlx5_aso_poll_cq(aso->aso, false);
+		if (ret)
+			usleep_range(2, 10);
+	} while (ret && time_is_after_jiffies(expires));
 	spin_unlock_bh(&aso->lock);
 	return ret;
 }
-- 
cgit v1.2.3-70-g09d2


From 00d521b39307c3232bdc685c2e9fd82ed973ac24 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:49 +0300
Subject: net: don't abuse "default" case for unknown ioctl in dev_ifsioc()

The "switch (cmd)" block from dev_ifsioc() gained a bit too much
unnecessary manual handling of "cmd" in the "default" case, starting
with the private ioctls.

Clean that up by using the "ellipsis" gcc extension, adding separate
cases for the rest of the ioctls, and letting the default case only
return -EINVAL.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev_ioctl.c | 42 +++++++++++++++++++-----------------------
 1 file changed, 19 insertions(+), 23 deletions(-)

diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 846669426236..1c0256ea522f 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -391,36 +391,32 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data,
 		rtnl_lock();
 		return err;
 
+	case SIOCDEVPRIVATE ... SIOCDEVPRIVATE + 15:
+		return dev_siocdevprivate(dev, ifr, data, cmd);
+
 	case SIOCSHWTSTAMP:
 		err = net_hwtstamp_validate(ifr);
 		if (err)
 			return err;
 		fallthrough;
 
-	/*
-	 *	Unknown or private ioctl
-	 */
-	default:
-		if (cmd >= SIOCDEVPRIVATE &&
-		    cmd <= SIOCDEVPRIVATE + 15)
-			return dev_siocdevprivate(dev, ifr, data, cmd);
-
-		if (cmd == SIOCGMIIPHY ||
-		    cmd == SIOCGMIIREG ||
-		    cmd == SIOCSMIIREG ||
-		    cmd == SIOCSHWTSTAMP ||
-		    cmd == SIOCGHWTSTAMP) {
-			err = dev_eth_ioctl(dev, ifr, cmd);
-		} else if (cmd == SIOCBONDENSLAVE ||
-		    cmd == SIOCBONDRELEASE ||
-		    cmd == SIOCBONDSETHWADDR ||
-		    cmd == SIOCBONDSLAVEINFOQUERY ||
-		    cmd == SIOCBONDINFOQUERY ||
-		    cmd == SIOCBONDCHANGEACTIVE) {
-			err = dev_siocbond(dev, ifr, cmd);
-		} else
-			err = -EINVAL;
+	case SIOCGHWTSTAMP:
+	case SIOCGMIIPHY:
+	case SIOCGMIIREG:
+	case SIOCSMIIREG:
+		return dev_eth_ioctl(dev, ifr, cmd);
+
+	case SIOCBONDENSLAVE:
+	case SIOCBONDRELEASE:
+	case SIOCBONDSETHWADDR:
+	case SIOCBONDSLAVEINFOQUERY:
+	case SIOCBONDINFOQUERY:
+	case SIOCBONDCHANGEACTIVE:
+		return dev_siocbond(dev, ifr, cmd);
 
+	/* Unknown ioctl */
+	default:
+		err = -EINVAL;
 	}
 	return err;
 }
-- 
cgit v1.2.3-70-g09d2


From 1193db2a55b6b04f296f03affdfa80c16ecc3814 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:50 +0300
Subject: net: simplify handling of dsa_ndo_eth_ioctl() return code

In the expression "x == 0 || x != -95", the term "x == 0" does not
change the expression's logical value, because 0 != -95, and so,
if x is 0, the expression would still be true by virtue of the second
term. If x is non-zero, the expression depends on the truth value of
the second term anyway. As such, the first term is redundant and can
be deleted.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev_ioctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 1c0256ea522f..b299fb23fcfa 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -249,7 +249,7 @@ static int dev_eth_ioctl(struct net_device *dev,
 	int err;
 
 	err = dsa_ndo_eth_ioctl(dev, ifr, cmd);
-	if (err == 0 || err != -EOPNOTSUPP)
+	if (err != -EOPNOTSUPP)
 		return err;
 
 	if (ops->ndo_eth_ioctl) {
-- 
cgit v1.2.3-70-g09d2


From 4ee58e1e56800b589afe31c34547e2bc0c59f586 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:51 +0300
Subject: net: promote SIOCSHWTSTAMP and SIOCGHWTSTAMP ioctls to dedicated
 handlers

DSA does not want to intercept all ioctls handled by dev_eth_ioctl(),
only SIOCSHWTSTAMP. This can be seen from commit f685e609a301 ("net:
dsa: Deny PTP on master if switch supports it"). However, the way in
which the dsa_ndo_eth_ioctl() is called would suggest otherwise.

Split the handling of SIOCSHWTSTAMP and SIOCGHWTSTAMP ioctls into
separate case statements of dev_ifsioc(), and make each one call its own
sub-function. This also removes the dsa_ndo_eth_ioctl() call from
dev_eth_ioctl(), which from now on exclusively handles PHY ioctls.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev_ioctl.c | 39 ++++++++++++++++++++++++++-------------
 1 file changed, 26 insertions(+), 13 deletions(-)

diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index b299fb23fcfa..3b1402f6897c 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -246,20 +246,34 @@ static int dev_eth_ioctl(struct net_device *dev,
 			 struct ifreq *ifr, unsigned int cmd)
 {
 	const struct net_device_ops *ops = dev->netdev_ops;
+
+	if (!ops->ndo_eth_ioctl)
+		return -EOPNOTSUPP;
+
+	if (!netif_device_present(dev))
+		return -ENODEV;
+
+	return ops->ndo_eth_ioctl(dev, ifr, cmd);
+}
+
+static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr)
+{
+	return dev_eth_ioctl(dev, ifr, SIOCGHWTSTAMP);
+}
+
+static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
+{
 	int err;
 
-	err = dsa_ndo_eth_ioctl(dev, ifr, cmd);
-	if (err != -EOPNOTSUPP)
+	err = net_hwtstamp_validate(ifr);
+	if (err)
 		return err;
 
-	if (ops->ndo_eth_ioctl) {
-		if (netif_device_present(dev))
-			err = ops->ndo_eth_ioctl(dev, ifr, cmd);
-		else
-			err = -ENODEV;
-	}
+	err = dsa_ndo_eth_ioctl(dev, ifr, SIOCSHWTSTAMP);
+	if (err != -EOPNOTSUPP)
+		return err;
 
-	return err;
+	return dev_eth_ioctl(dev, ifr, SIOCSHWTSTAMP);
 }
 
 static int dev_siocbond(struct net_device *dev,
@@ -395,12 +409,11 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data,
 		return dev_siocdevprivate(dev, ifr, data, cmd);
 
 	case SIOCSHWTSTAMP:
-		err = net_hwtstamp_validate(ifr);
-		if (err)
-			return err;
-		fallthrough;
+		return dev_set_hwtstamp(dev, ifr);
 
 	case SIOCGHWTSTAMP:
+		return dev_get_hwtstamp(dev, ifr);
+
 	case SIOCGMIIPHY:
 	case SIOCGMIIREG:
 	case SIOCSMIIREG:
-- 
cgit v1.2.3-70-g09d2


From d5d5fd8f2552f6b9c50d3937c77b6783f99fbe83 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:52 +0300
Subject: net: move copy_from_user() out of net_hwtstamp_validate()

The kernel will want to start using the more meaningful struct
hwtstamp_config pointer in more places, so move the copy_from_user() at
the beginning of dev_set_hwtstamp() in order to get to that, and pass
this argument to net_hwtstamp_validate().

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev_ioctl.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 3b1402f6897c..34a0da5fbcfc 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -183,22 +183,18 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm
 	return err;
 }
 
-static int net_hwtstamp_validate(struct ifreq *ifr)
+static int net_hwtstamp_validate(const struct hwtstamp_config *cfg)
 {
-	struct hwtstamp_config cfg;
 	enum hwtstamp_tx_types tx_type;
 	enum hwtstamp_rx_filters rx_filter;
 	int tx_type_valid = 0;
 	int rx_filter_valid = 0;
 
-	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
-		return -EFAULT;
-
-	if (cfg.flags & ~HWTSTAMP_FLAG_MASK)
+	if (cfg->flags & ~HWTSTAMP_FLAG_MASK)
 		return -EINVAL;
 
-	tx_type = cfg.tx_type;
-	rx_filter = cfg.rx_filter;
+	tx_type = cfg->tx_type;
+	rx_filter = cfg->rx_filter;
 
 	switch (tx_type) {
 	case HWTSTAMP_TX_OFF:
@@ -263,9 +259,13 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 
 static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 {
+	struct hwtstamp_config cfg;
 	int err;
 
-	err = net_hwtstamp_validate(ifr);
+	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
+		return -EFAULT;
+
+	err = net_hwtstamp_validate(&cfg);
 	if (err)
 		return err;
 
-- 
cgit v1.2.3-70-g09d2


From c4bffeaa8d50b7279c5a76597efa4b06e709df63 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:53 +0300
Subject: net: add struct kernel_hwtstamp_config and make
 net_hwtstamp_validate() use it

Jakub Kicinski suggested that we may want to add new UAPI for
controlling hardware timestamping through netlink in the future, and in
that case, we will be limited to the struct hwtstamp_config that is
currently passed in fixed binary format through the SIOCGHWTSTAMP and
SIOCSHWTSTAMP ioctls. It would be good if new kernel code already
started operating on an extensible kernel variant of that structure,
similar in concept to struct kernel_ethtool_coalesce vs struct
ethtool_coalesce.

Since struct hwtstamp_config is in include/uapi/linux/net_tstamp.h, here
we introduce include/linux/net_tstamp.h which shadows that other header,
but also includes it, so that existing includers of this header work as
before. In addition to that, we add the definition for the kernel-only
structure, and a helper which translates all fields by manual copying.
I am doing a manual copy in order to not force the alignment (or type)
of the fields of struct kernel_hwtstamp_config to be the same as of
struct hwtstamp_config, even though now, they are the same.

Link: https://lore.kernel.org/netdev/20230330223519.36ce7d23@kernel.org/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net_tstamp.h | 33 +++++++++++++++++++++++++++++++++
 net/core/dev_ioctl.c       |  7 +++++--
 2 files changed, 38 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/net_tstamp.h

diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
new file mode 100644
index 000000000000..fd67f3cc0c4b
--- /dev/null
+++ b/include/linux/net_tstamp.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _LINUX_NET_TIMESTAMPING_H_
+#define _LINUX_NET_TIMESTAMPING_H_
+
+#include <uapi/linux/net_tstamp.h>
+
+/**
+ * struct kernel_hwtstamp_config - Kernel copy of struct hwtstamp_config
+ *
+ * @flags: see struct hwtstamp_config
+ * @tx_type: see struct hwtstamp_config
+ * @rx_filter: see struct hwtstamp_config
+ *
+ * Prefer using this structure for in-kernel processing of hardware
+ * timestamping configuration, over the inextensible struct hwtstamp_config
+ * exposed to the %SIOCGHWTSTAMP and %SIOCSHWTSTAMP ioctl UAPI.
+ */
+struct kernel_hwtstamp_config {
+	int flags;
+	int tx_type;
+	int rx_filter;
+};
+
+static inline void hwtstamp_config_to_kernel(struct kernel_hwtstamp_config *kernel_cfg,
+					     const struct hwtstamp_config *cfg)
+{
+	kernel_cfg->flags = cfg->flags;
+	kernel_cfg->tx_type = cfg->tx_type;
+	kernel_cfg->rx_filter = cfg->rx_filter;
+}
+
+#endif /* _LINUX_NET_TIMESTAMPING_H_ */
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 34a0da5fbcfc..c532ef4d5dff 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -183,7 +183,7 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm
 	return err;
 }
 
-static int net_hwtstamp_validate(const struct hwtstamp_config *cfg)
+static int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg)
 {
 	enum hwtstamp_tx_types tx_type;
 	enum hwtstamp_rx_filters rx_filter;
@@ -259,13 +259,16 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 
 static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 {
+	struct kernel_hwtstamp_config kernel_cfg;
 	struct hwtstamp_config cfg;
 	int err;
 
 	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
 		return -EFAULT;
 
-	err = net_hwtstamp_validate(&cfg);
+	hwtstamp_config_to_kernel(&kernel_cfg, &cfg);
+
+	err = net_hwtstamp_validate(&kernel_cfg);
 	if (err)
 		return err;
 
-- 
cgit v1.2.3-70-g09d2


From ff6ac4d013e680a5e7a38ee83ca59ffe1846915d Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:54 +0300
Subject: net: dsa: make dsa_port_supports_hwtstamp() construct a fake ifreq

dsa_master_ioctl() is in the process of getting converted to a different
API, where we won't have access to a struct ifreq * anymore, but rather,
to a struct kernel_hwtstamp_config.

Since ds->ops->port_hwtstamp_get() still uses struct ifreq *, this
creates a difficult situation where we have to make up such a dummy
pointer.

The conversion is a bit messy, because it forces a "good" implementation
of ds->ops->port_hwtstamp_get() to return -EFAULT in copy_to_user()
because of the NULL ifr->ifr_data pointer. However, it works, and it is
only a transient step until ds->ops->port_hwtstamp_get() gets converted
to the new API which passes struct kernel_hwtstamp_config and does not
call copy_to_user().

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/master.c |  2 +-
 net/dsa/port.c   | 10 ++++++----
 net/dsa/port.h   |  2 +-
 3 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/net/dsa/master.c b/net/dsa/master.c
index 22d3f16b0e6d..e397641382ca 100644
--- a/net/dsa/master.c
+++ b/net/dsa/master.c
@@ -212,7 +212,7 @@ static int dsa_master_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		 * switch in the tree that is PTP capable.
 		 */
 		list_for_each_entry(dp, &dst->ports, list)
-			if (dsa_port_supports_hwtstamp(dp, ifr))
+			if (dsa_port_supports_hwtstamp(dp))
 				return -EBUSY;
 		break;
 	}
diff --git a/net/dsa/port.c b/net/dsa/port.c
index 15cee17769e9..71ba30538411 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -114,19 +114,21 @@ static bool dsa_port_can_configure_learning(struct dsa_port *dp)
 	return !err;
 }
 
-bool dsa_port_supports_hwtstamp(struct dsa_port *dp, struct ifreq *ifr)
+bool dsa_port_supports_hwtstamp(struct dsa_port *dp)
 {
 	struct dsa_switch *ds = dp->ds;
+	struct ifreq ifr = {};
 	int err;
 
 	if (!ds->ops->port_hwtstamp_get || !ds->ops->port_hwtstamp_set)
 		return false;
 
 	/* "See through" shim implementations of the "get" method.
-	 * This will clobber the ifreq structure, but we will either return an
-	 * error, or the master will overwrite it with proper values.
+	 * Since we can't cook up a complete ioctl request structure, this will
+	 * fail in copy_to_user() with -EFAULT, which hopefully is enough to
+	 * detect a valid implementation.
 	 */
-	err = ds->ops->port_hwtstamp_get(ds, dp->index, ifr);
+	err = ds->ops->port_hwtstamp_get(ds, dp->index, &ifr);
 	return err != -EOPNOTSUPP;
 }
 
diff --git a/net/dsa/port.h b/net/dsa/port.h
index 9c218660d223..dc812512fd0e 100644
--- a/net/dsa/port.h
+++ b/net/dsa/port.h
@@ -15,7 +15,7 @@ struct switchdev_obj_port_mdb;
 struct switchdev_vlan_msti;
 struct phy_device;
 
-bool dsa_port_supports_hwtstamp(struct dsa_port *dp, struct ifreq *ifr);
+bool dsa_port_supports_hwtstamp(struct dsa_port *dp);
 void dsa_port_set_tag_protocol(struct dsa_port *cpu_dp,
 			       const struct dsa_device_ops *tag_ops);
 int dsa_port_set_state(struct dsa_port *dp, u8 state, bool do_fast_age);
-- 
cgit v1.2.3-70-g09d2


From 88c0a6b503b7f4fffb68a8d49c3987870c5b1d6b Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 2 Apr 2023 15:37:55 +0300
Subject: net: create a netdev notifier for DSA to reject PTP on DSA master

The fact that PTP 2-step TX timestamping is broken on DSA switches if
the master also timestamps the same packets is documented by commit
f685e609a301 ("net: dsa: Deny PTP on master if switch supports it").
We attempt to help the users avoid shooting themselves in the foot by
making DSA reject the timestamping ioctls on an interface that is a DSA
master, and the switch tree beneath it contains switches which are aware
of PTP.

The only problem is that there isn't an established way of intercepting
ndo_eth_ioctl calls, so DSA creates avoidable burden upon the network
stack by creating a struct dsa_netdevice_ops with overlaid function
pointers that are manually checked from the relevant call sites. There
used to be 2 such dsa_netdevice_ops, but now, ndo_eth_ioctl is the only
one left.

There is an ongoing effort to migrate driver-visible hardware timestamping
control from the ndo_eth_ioctl() based API to a new ndo_hwtstamp_set()
model, but DSA actively prevents that migration, since dsa_master_ioctl()
is currently coded to manually call the master's legacy ndo_eth_ioctl(),
and so, whenever a network device driver would be converted to the new
API, DSA's restrictions would be circumvented, because any device could
be used as a DSA master.

The established way for unrelated modules to react on a net device event
is via netdevice notifiers. So we create a new notifier which gets
called whenever there is an attempt to change hardware timestamping
settings on a device.

Finally, there is another reason why a netdev notifier will be a good
idea, besides strictly DSA, and this has to do with PHY timestamping.

With ndo_eth_ioctl(), all MAC drivers must manually call
phy_has_hwtstamp() before deciding whether to act upon SIOCSHWTSTAMP,
otherwise they must pass this ioctl to the PHY driver via
phy_mii_ioctl().

With the new ndo_hwtstamp_set() API, it will be desirable to simply not
make any calls into the MAC device driver when timestamping should be
performed at the PHY level.

But there exist drivers, such as the lan966x switch, which need to
install packet traps for PTP regardless of whether they are the layer
that provides the hardware timestamps, or the PHY is. That would be
impossible to support with the new API.

The proposal there, too, is to introduce a netdev notifier which acts as
a better cue for switching drivers to add or remove PTP packet traps,
than ndo_hwtstamp_set(). The one introduced here "almost" works there as
well, except for the fact that packet traps should only be installed if
the PHY driver succeeded to enable hardware timestamping, whereas here,
we need to deny hardware timestamping on the DSA master before it
actually gets enabled. This is why this notifier is called "PRE_", and
the notifier that would get used for PHY timestamping and packet traps
would be called NETDEV_CHANGE_HWTSTAMP. This isn't a new concept, for
example NETDEV_CHANGEUPPER and NETDEV_PRECHANGEUPPER do the same thing.

In expectation of future netlink UAPI, we also pass a non-NULL extack
pointer to the netdev notifier, and we make DSA populate it with an
informative reason for the rejection. To avoid making it go to waste, we
make the ioctl-based dev_set_hwtstamp() create a fake extack and print
the message to the kernel log.

Link: https://lore.kernel.org/netdev/20230401191215.tvveoi3lkawgg6g4@skbuf/
Link: https://lore.kernel.org/netdev/20230310164451.ls7bbs6pdzs4m6pw@skbuf/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  9 ++++++++-
 include/net/dsa.h         | 51 -----------------------------------------------
 net/core/dev.c            |  8 +++-----
 net/core/dev_ioctl.c      | 16 +++++++++++++--
 net/dsa/master.c          | 50 ++++++++++++++--------------------------------
 net/dsa/master.h          |  3 +++
 net/dsa/slave.c           | 11 ++++++++++
 7 files changed, 54 insertions(+), 94 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 62e093a6d6d1..a740be3bb911 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2878,6 +2878,7 @@ enum netdev_cmd {
 	NETDEV_OFFLOAD_XSTATS_REPORT_USED,
 	NETDEV_OFFLOAD_XSTATS_REPORT_DELTA,
 	NETDEV_XDP_FEAT_CHANGE,
+	NETDEV_PRE_CHANGE_HWTSTAMP,
 };
 const char *netdev_cmd_to_name(enum netdev_cmd cmd);
 
@@ -2928,6 +2929,11 @@ struct netdev_notifier_pre_changeaddr_info {
 	const unsigned char *dev_addr;
 };
 
+struct netdev_notifier_hwtstamp_info {
+	struct netdev_notifier_info info; /* must be first */
+	struct kernel_hwtstamp_config *config;
+};
+
 enum netdev_offload_xstats_type {
 	NETDEV_OFFLOAD_XSTATS_TYPE_L3 = 1,
 };
@@ -2984,7 +2990,8 @@ netdev_notifier_info_to_extack(const struct netdev_notifier_info *info)
 }
 
 int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
-
+int call_netdevice_notifiers_info(unsigned long val,
+				  struct netdev_notifier_info *info);
 
 extern rwlock_t				dev_base_lock;		/* Device list lock */
 
diff --git a/include/net/dsa.h b/include/net/dsa.h
index a15f17a38eca..8903053fa5aa 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -109,16 +109,6 @@ struct dsa_device_ops {
 	bool promisc_on_master;
 };
 
-/* This structure defines the control interfaces that are overlayed by the
- * DSA layer on top of the DSA CPU/management net_device instance. This is
- * used by the core net_device layer while calling various net_device_ops
- * function pointers.
- */
-struct dsa_netdevice_ops {
-	int (*ndo_eth_ioctl)(struct net_device *dev, struct ifreq *ifr,
-			     int cmd);
-};
-
 struct dsa_lag {
 	struct net_device *dev;
 	unsigned int id;
@@ -317,11 +307,6 @@ struct dsa_port {
 	 */
 	const struct ethtool_ops *orig_ethtool_ops;
 
-	/*
-	 * Original copy of the master netdev net_device_ops
-	 */
-	const struct dsa_netdevice_ops *netdev_ops;
-
 	/* List of MAC addresses that must be forwarded on this port.
 	 * These are only valid on CPU ports and DSA links.
 	 */
@@ -1339,42 +1324,6 @@ static inline void dsa_tag_generic_flow_dissect(const struct sk_buff *skb,
 #endif
 }
 
-#if IS_ENABLED(CONFIG_NET_DSA)
-static inline int __dsa_netdevice_ops_check(struct net_device *dev)
-{
-	int err = -EOPNOTSUPP;
-
-	if (!dev->dsa_ptr)
-		return err;
-
-	if (!dev->dsa_ptr->netdev_ops)
-		return err;
-
-	return 0;
-}
-
-static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr,
-				    int cmd)
-{
-	const struct dsa_netdevice_ops *ops;
-	int err;
-
-	err = __dsa_netdevice_ops_check(dev);
-	if (err)
-		return err;
-
-	ops = dev->dsa_ptr->netdev_ops;
-
-	return ops->ndo_eth_ioctl(dev, ifr, cmd);
-}
-#else
-static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr,
-				    int cmd)
-{
-	return -EOPNOTSUPP;
-}
-#endif
-
 void dsa_unregister_switch(struct dsa_switch *ds);
 int dsa_register_switch(struct dsa_switch *ds);
 void dsa_switch_shutdown(struct dsa_switch *ds);
diff --git a/net/core/dev.c b/net/core/dev.c
index 0c4b21291348..7ce5985be84b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -160,8 +160,6 @@ struct list_head ptype_base[PTYPE_HASH_SIZE] __read_mostly;
 struct list_head ptype_all __read_mostly;	/* Taps */
 
 static int netif_rx_internal(struct sk_buff *skb);
-static int call_netdevice_notifiers_info(unsigned long val,
-					 struct netdev_notifier_info *info);
 static int call_netdevice_notifiers_extack(unsigned long val,
 					   struct net_device *dev,
 					   struct netlink_ext_ack *extack);
@@ -1614,7 +1612,7 @@ const char *netdev_cmd_to_name(enum netdev_cmd cmd)
 	N(SVLAN_FILTER_PUSH_INFO) N(SVLAN_FILTER_DROP_INFO)
 	N(PRE_CHANGEADDR) N(OFFLOAD_XSTATS_ENABLE) N(OFFLOAD_XSTATS_DISABLE)
 	N(OFFLOAD_XSTATS_REPORT_USED) N(OFFLOAD_XSTATS_REPORT_DELTA)
-	N(XDP_FEAT_CHANGE)
+	N(XDP_FEAT_CHANGE) N(PRE_CHANGE_HWTSTAMP)
 	}
 #undef N
 	return "UNKNOWN_NETDEV_EVENT";
@@ -1919,8 +1917,8 @@ static void move_netdevice_notifiers_dev_net(struct net_device *dev,
  *	are as for raw_notifier_call_chain().
  */
 
-static int call_netdevice_notifiers_info(unsigned long val,
-					 struct netdev_notifier_info *info)
+int call_netdevice_notifiers_info(unsigned long val,
+				  struct netdev_notifier_info *info)
 {
 	struct net *net = dev_net(info->dev);
 	int ret;
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index c532ef4d5dff..6d772837eb3f 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -259,7 +259,11 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 
 static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 {
+	struct netdev_notifier_hwtstamp_info info = {
+		.info.dev = dev,
+	};
 	struct kernel_hwtstamp_config kernel_cfg;
+	struct netlink_ext_ack extack = {};
 	struct hwtstamp_config cfg;
 	int err;
 
@@ -272,9 +276,17 @@ static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 	if (err)
 		return err;
 
-	err = dsa_ndo_eth_ioctl(dev, ifr, SIOCSHWTSTAMP);
-	if (err != -EOPNOTSUPP)
+	info.info.extack = &extack;
+	info.config = &kernel_cfg;
+
+	err = call_netdevice_notifiers_info(NETDEV_PRE_CHANGE_HWTSTAMP,
+					    &info.info);
+	err = notifier_to_errno(err);
+	if (err) {
+		if (extack._msg)
+			netdev_err(dev, "%s\n", extack._msg);
 		return err;
+	}
 
 	return dev_eth_ioctl(dev, ifr, SIOCSHWTSTAMP);
 }
diff --git a/net/dsa/master.c b/net/dsa/master.c
index e397641382ca..c2cabe6248b1 100644
--- a/net/dsa/master.c
+++ b/net/dsa/master.c
@@ -195,38 +195,31 @@ static void dsa_master_get_strings(struct net_device *dev, uint32_t stringset,
 	}
 }
 
-static int dsa_master_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+/* Deny PTP operations on master if there is at least one switch in the tree
+ * that is PTP capable.
+ */
+int dsa_master_pre_change_hwtstamp(struct net_device *dev,
+				   const struct kernel_hwtstamp_config *config,
+				   struct netlink_ext_ack *extack)
 {
 	struct dsa_port *cpu_dp = dev->dsa_ptr;
 	struct dsa_switch *ds = cpu_dp->ds;
 	struct dsa_switch_tree *dst;
-	int err = -EOPNOTSUPP;
 	struct dsa_port *dp;
 
 	dst = ds->dst;
 
-	switch (cmd) {
-	case SIOCGHWTSTAMP:
-	case SIOCSHWTSTAMP:
-		/* Deny PTP operations on master if there is at least one
-		 * switch in the tree that is PTP capable.
-		 */
-		list_for_each_entry(dp, &dst->ports, list)
-			if (dsa_port_supports_hwtstamp(dp))
-				return -EBUSY;
-		break;
+	list_for_each_entry(dp, &dst->ports, list) {
+		if (dsa_port_supports_hwtstamp(dp)) {
+			NL_SET_ERR_MSG(extack,
+				       "HW timestamping not allowed on DSA master when switch supports the operation");
+			return -EBUSY;
+		}
 	}
 
-	if (dev->netdev_ops->ndo_eth_ioctl)
-		err = dev->netdev_ops->ndo_eth_ioctl(dev, ifr, cmd);
-
-	return err;
+	return 0;
 }
 
-static const struct dsa_netdevice_ops dsa_netdev_ops = {
-	.ndo_eth_ioctl = dsa_master_ioctl,
-};
-
 static int dsa_master_ethtool_setup(struct net_device *dev)
 {
 	struct dsa_port *cpu_dp = dev->dsa_ptr;
@@ -267,15 +260,6 @@ static void dsa_master_ethtool_teardown(struct net_device *dev)
 	cpu_dp->orig_ethtool_ops = NULL;
 }
 
-static void dsa_netdev_ops_set(struct net_device *dev,
-			       const struct dsa_netdevice_ops *ops)
-{
-	if (netif_is_lag_master(dev))
-		return;
-
-	dev->dsa_ptr->netdev_ops = ops;
-}
-
 /* Keep the master always promiscuous if the tagging protocol requires that
  * (garbles MAC DA) or if it doesn't support unicast filtering, case in which
  * it would revert to promiscuous mode as soon as we call dev_uc_add() on it
@@ -414,16 +398,13 @@ int dsa_master_setup(struct net_device *dev, struct dsa_port *cpu_dp)
 	if (ret)
 		goto out_err_reset_promisc;
 
-	dsa_netdev_ops_set(dev, &dsa_netdev_ops);
-
 	ret = sysfs_create_group(&dev->dev.kobj, &dsa_group);
 	if (ret)
-		goto out_err_ndo_teardown;
+		goto out_err_ethtool_teardown;
 
 	return ret;
 
-out_err_ndo_teardown:
-	dsa_netdev_ops_set(dev, NULL);
+out_err_ethtool_teardown:
 	dsa_master_ethtool_teardown(dev);
 out_err_reset_promisc:
 	dsa_master_set_promiscuity(dev, -1);
@@ -433,7 +414,6 @@ out_err_reset_promisc:
 void dsa_master_teardown(struct net_device *dev)
 {
 	sysfs_remove_group(&dev->dev.kobj, &dsa_group);
-	dsa_netdev_ops_set(dev, NULL);
 	dsa_master_ethtool_teardown(dev);
 	dsa_master_reset_mtu(dev);
 	dsa_master_set_promiscuity(dev, -1);
diff --git a/net/dsa/master.h b/net/dsa/master.h
index 3fc0e610b5b5..80842f4e27f7 100644
--- a/net/dsa/master.h
+++ b/net/dsa/master.h
@@ -15,5 +15,8 @@ int dsa_master_lag_setup(struct net_device *lag_dev, struct dsa_port *cpu_dp,
 			 struct netlink_ext_ack *extack);
 void dsa_master_lag_teardown(struct net_device *lag_dev,
 			     struct dsa_port *cpu_dp);
+int dsa_master_pre_change_hwtstamp(struct net_device *dev,
+				   const struct kernel_hwtstamp_config *config,
+				   struct netlink_ext_ack *extack);
 
 #endif
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 165bb2cb8431..8abc1658ac47 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -3289,6 +3289,7 @@ static int dsa_master_changeupper(struct net_device *dev,
 static int dsa_slave_netdevice_event(struct notifier_block *nb,
 				     unsigned long event, void *ptr)
 {
+	struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr);
 	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
 
 	switch (event) {
@@ -3418,6 +3419,16 @@ static int dsa_slave_netdevice_event(struct notifier_block *nb,
 
 		return NOTIFY_OK;
 	}
+	case NETDEV_PRE_CHANGE_HWTSTAMP: {
+		struct netdev_notifier_hwtstamp_info *info = ptr;
+		int err;
+
+		if (!netdev_uses_dsa(dev))
+			return NOTIFY_DONE;
+
+		err = dsa_master_pre_change_hwtstamp(dev, info->config, extack);
+		return notifier_from_errno(err);
+	}
 	default:
 		break;
 	}
-- 
cgit v1.2.3-70-g09d2


From fc281d78b6863ba86baec220651fef815341d3a0 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:11:40 +0200
Subject: net: phy: smsc: rename flag energy_enable

Rename the flag to edpd_enable, as we're not enabling energy but
edpd (energy detect power down) mode. In addition change the
type to a bit field member in preparation of adding further flags.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 730964b856ab..928cf6d8b2e6 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -44,7 +44,7 @@ static struct smsc_hw_stat smsc_hw_stats[] = {
 };
 
 struct smsc_phy_priv {
-	bool energy_enable;
+	unsigned int edpd_enable:1;
 };
 
 static int smsc_phy_ack_interrupt(struct phy_device *phydev)
@@ -102,7 +102,7 @@ int smsc_phy_config_init(struct phy_device *phydev)
 {
 	struct smsc_phy_priv *priv = phydev->priv;
 
-	if (!priv || !priv->energy_enable || phydev->irq != PHY_POLL)
+	if (!priv || !priv->edpd_enable || phydev->irq != PHY_POLL)
 		return 0;
 
 	/* Enable energy detect power down mode */
@@ -198,7 +198,7 @@ int lan87xx_read_status(struct phy_device *phydev)
 	if (err)
 		return err;
 
-	if (!phydev->link && priv && priv->energy_enable &&
+	if (!phydev->link && priv && priv->edpd_enable &&
 	    phydev->irq == PHY_POLL) {
 		/* Disable EDPD to wake up PHY */
 		int rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
@@ -284,10 +284,10 @@ int smsc_phy_probe(struct phy_device *phydev)
 	if (!priv)
 		return -ENOMEM;
 
-	priv->energy_enable = true;
+	priv->edpd_enable = true;
 
 	if (device_property_present(dev, "smsc,disable-energy-detect"))
-		priv->energy_enable = false;
+		priv->edpd_enable = false;
 
 	phydev->priv = priv;
 
-- 
cgit v1.2.3-70-g09d2


From 89946e31ff4f7f43b3cf4837eb5b9f141eb6f1d6 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:12:20 +0200
Subject: net: phy: smsc: add helper smsc_phy_config_edpd

Add helper smsc_phy_config_edpd() and explicitly clear bit
MII_LAN83C185_EDPWRDOWN is edpd_enable isn't set.
Boot loader may have left whatever value.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 928cf6d8b2e6..1b588366e075 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -77,6 +77,18 @@ int smsc_phy_config_intr(struct phy_device *phydev)
 }
 EXPORT_SYMBOL_GPL(smsc_phy_config_intr);
 
+static int smsc_phy_config_edpd(struct phy_device *phydev)
+{
+	struct smsc_phy_priv *priv = phydev->priv;
+
+	if (priv->edpd_enable)
+		return phy_set_bits(phydev, MII_LAN83C185_CTRL_STATUS,
+				    MII_LAN83C185_EDPWRDOWN);
+	else
+		return phy_clear_bits(phydev, MII_LAN83C185_CTRL_STATUS,
+				      MII_LAN83C185_EDPWRDOWN);
+}
+
 irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev)
 {
 	int irq_status;
@@ -105,9 +117,7 @@ int smsc_phy_config_init(struct phy_device *phydev)
 	if (!priv || !priv->edpd_enable || phydev->irq != PHY_POLL)
 		return 0;
 
-	/* Enable energy detect power down mode */
-	return phy_set_bits(phydev, MII_LAN83C185_CTRL_STATUS,
-			    MII_LAN83C185_EDPWRDOWN);
+	return smsc_phy_config_edpd(phydev);
 }
 EXPORT_SYMBOL_GPL(smsc_phy_config_init);
 
-- 
cgit v1.2.3-70-g09d2


From d56417ad1133fc41752bb9fe37da7ae3187395a4 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:13:05 +0200
Subject: net: phy: smsc: clear edpd_enable if interrupt mode is used

Clear edpd_enable if interupt mode is used, this avoids
having to check for PHY_POLL multiple times.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 1b588366e075..f5ecd8bea0e3 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -114,9 +114,12 @@ int smsc_phy_config_init(struct phy_device *phydev)
 {
 	struct smsc_phy_priv *priv = phydev->priv;
 
-	if (!priv || !priv->edpd_enable || phydev->irq != PHY_POLL)
+	if (!priv)
 		return 0;
 
+	if (phydev->irq != PHY_POLL)
+		priv->edpd_enable = false;
+
 	return smsc_phy_config_edpd(phydev);
 }
 EXPORT_SYMBOL_GPL(smsc_phy_config_init);
@@ -208,8 +211,7 @@ int lan87xx_read_status(struct phy_device *phydev)
 	if (err)
 		return err;
 
-	if (!phydev->link && priv && priv->edpd_enable &&
-	    phydev->irq == PHY_POLL) {
+	if (!phydev->link && priv && priv->edpd_enable) {
 		/* Disable EDPD to wake up PHY */
 		int rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
 		if (rc < 0)
-- 
cgit v1.2.3-70-g09d2


From a620511080960a72a40c58c55945e41cc3052f6f Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:13:35 +0200
Subject: net: phy: smsc: add flag edpd_mode_set_by_user

Add flag edpd_mode_set_by_user in preparation of adding edpd phy tunable
support. This flag will allow users to override the default behavior
of edpd being disabled if interrupt mode is used.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index f5ecd8bea0e3..25b9cd474e83 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -45,6 +45,7 @@ static struct smsc_hw_stat smsc_hw_stats[] = {
 
 struct smsc_phy_priv {
 	unsigned int edpd_enable:1;
+	unsigned int edpd_mode_set_by_user:1;
 };
 
 static int smsc_phy_ack_interrupt(struct phy_device *phydev)
@@ -117,7 +118,8 @@ int smsc_phy_config_init(struct phy_device *phydev)
 	if (!priv)
 		return 0;
 
-	if (phydev->irq != PHY_POLL)
+	/* don't use EDPD in irq mode except overridden by user */
+	if (!priv->edpd_mode_set_by_user && phydev->irq != PHY_POLL)
 		priv->edpd_enable = false;
 
 	return smsc_phy_config_edpd(phydev);
-- 
cgit v1.2.3-70-g09d2


From 1ce658693b089ee753260eae78a3ea572bc4b3fb Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:14:35 +0200
Subject: net: phy: smsc: prepare for making edpd wait period configurable

Add a member edpd_max_wait_ms to the private data structure in preparation
of making the wait period configurable by supporting the edpd phy tunable.

v2:
- rename constant to EDPD_MAX_WAIT_DFLT_MS

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 25b9cd474e83..659a3ab10d82 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -33,6 +33,8 @@
 #define SPECIAL_CTRL_STS_AMDIX_ENABLE_	0x4000
 #define SPECIAL_CTRL_STS_AMDIX_STATE_	0x2000
 
+#define EDPD_MAX_WAIT_DFLT_MS		640
+
 struct smsc_hw_stat {
 	const char *string;
 	u8 reg;
@@ -46,6 +48,7 @@ static struct smsc_hw_stat smsc_hw_stats[] = {
 struct smsc_phy_priv {
 	unsigned int edpd_enable:1;
 	unsigned int edpd_mode_set_by_user:1;
+	unsigned int edpd_max_wait_ms;
 };
 
 static int smsc_phy_ack_interrupt(struct phy_device *phydev)
@@ -213,9 +216,13 @@ int lan87xx_read_status(struct phy_device *phydev)
 	if (err)
 		return err;
 
-	if (!phydev->link && priv && priv->edpd_enable) {
+	if (!phydev->link && priv && priv->edpd_enable &&
+	    priv->edpd_max_wait_ms) {
+		unsigned int max_wait = priv->edpd_max_wait_ms * 1000;
+		int rc;
+
 		/* Disable EDPD to wake up PHY */
-		int rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
+		rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
 		if (rc < 0)
 			return rc;
 
@@ -229,7 +236,7 @@ int lan87xx_read_status(struct phy_device *phydev)
 		 */
 		read_poll_timeout(phy_read, rc,
 				  rc & MII_LAN83C185_ENERGYON || rc < 0,
-				  10000, 640000, true, phydev,
+				  10000, max_wait, true, phydev,
 				  MII_LAN83C185_CTRL_STATUS);
 		if (rc < 0)
 			return rc;
@@ -299,6 +306,7 @@ int smsc_phy_probe(struct phy_device *phydev)
 		return -ENOMEM;
 
 	priv->edpd_enable = true;
+	priv->edpd_max_wait_ms = EDPD_MAX_WAIT_DFLT_MS;
 
 	if (device_property_present(dev, "smsc,disable-energy-detect"))
 		priv->edpd_enable = false;
-- 
cgit v1.2.3-70-g09d2


From 657de1cf258dbe2489906c81bd91e4af536de255 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:16:56 +0200
Subject: net: phy: smsc: add support for edpd tunable

This adds support for the EDPD PHY tunable.
Per default EDPD is disabled in interrupt mode, the tunable can be used
to override this, e.g. if the link partner doesn't use EDPD.
The interval to check for energy can be chosen between 1000ms and
2000ms. Note that this value consists of the 1000ms phylib interval
for state machine runs plus the time to wait for energy being detected.

v2:
- consider that phylib core holds phydev->lock when calling the
  phy tunable hooks

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c  | 75 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/smscphy.h |  4 +++
 2 files changed, 79 insertions(+)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 659a3ab10d82..0eba69ad5745 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -34,6 +34,8 @@
 #define SPECIAL_CTRL_STS_AMDIX_STATE_	0x2000
 
 #define EDPD_MAX_WAIT_DFLT_MS		640
+/* interval between phylib state machine runs in ms */
+#define PHY_STATE_MACH_MS		1000
 
 struct smsc_hw_stat {
 	const char *string;
@@ -295,6 +297,79 @@ static void smsc_get_stats(struct phy_device *phydev,
 		data[i] = smsc_get_stat(phydev, i);
 }
 
+static int smsc_phy_get_edpd(struct phy_device *phydev, u16 *edpd)
+{
+	struct smsc_phy_priv *priv = phydev->priv;
+
+	if (!priv)
+		return -EOPNOTSUPP;
+
+	if (!priv->edpd_enable)
+		*edpd = ETHTOOL_PHY_EDPD_DISABLE;
+	else if (!priv->edpd_max_wait_ms)
+		*edpd = ETHTOOL_PHY_EDPD_NO_TX;
+	else
+		*edpd = PHY_STATE_MACH_MS + priv->edpd_max_wait_ms;
+
+	return 0;
+}
+
+static int smsc_phy_set_edpd(struct phy_device *phydev, u16 edpd)
+{
+	struct smsc_phy_priv *priv = phydev->priv;
+
+	if (!priv)
+		return -EOPNOTSUPP;
+
+	switch (edpd) {
+	case ETHTOOL_PHY_EDPD_DISABLE:
+		priv->edpd_enable = false;
+		break;
+	case ETHTOOL_PHY_EDPD_NO_TX:
+		priv->edpd_enable = true;
+		priv->edpd_max_wait_ms = 0;
+		break;
+	case ETHTOOL_PHY_EDPD_DFLT_TX_MSECS:
+		edpd = PHY_STATE_MACH_MS + EDPD_MAX_WAIT_DFLT_MS;
+		fallthrough;
+	default:
+		if (phydev->irq != PHY_POLL)
+			return -EOPNOTSUPP;
+		if (edpd < PHY_STATE_MACH_MS || edpd > PHY_STATE_MACH_MS + 1000)
+			return -EINVAL;
+		priv->edpd_enable = true;
+		priv->edpd_max_wait_ms = edpd - PHY_STATE_MACH_MS;
+	}
+
+	priv->edpd_mode_set_by_user = true;
+
+	return smsc_phy_config_edpd(phydev);
+}
+
+int smsc_phy_get_tunable(struct phy_device *phydev,
+			 struct ethtool_tunable *tuna, void *data)
+{
+	switch (tuna->id) {
+	case ETHTOOL_PHY_EDPD:
+		return smsc_phy_get_edpd(phydev, data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+EXPORT_SYMBOL_GPL(smsc_phy_get_tunable);
+
+int smsc_phy_set_tunable(struct phy_device *phydev,
+			 struct ethtool_tunable *tuna, const void *data)
+{
+	switch (tuna->id) {
+	case ETHTOOL_PHY_EDPD:
+		return smsc_phy_set_edpd(phydev, *(u16 *)data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+EXPORT_SYMBOL_GPL(smsc_phy_set_tunable);
+
 int smsc_phy_probe(struct phy_device *phydev)
 {
 	struct device *dev = &phydev->mdio.dev;
diff --git a/include/linux/smscphy.h b/include/linux/smscphy.h
index 80f37c1dba58..e1c88627755a 100644
--- a/include/linux/smscphy.h
+++ b/include/linux/smscphy.h
@@ -32,6 +32,10 @@ int smsc_phy_config_intr(struct phy_device *phydev);
 irqreturn_t smsc_phy_handle_interrupt(struct phy_device *phydev);
 int smsc_phy_config_init(struct phy_device *phydev);
 int lan87xx_read_status(struct phy_device *phydev);
+int smsc_phy_get_tunable(struct phy_device *phydev,
+			 struct ethtool_tunable *tuna, void *data);
+int smsc_phy_set_tunable(struct phy_device *phydev,
+			 struct ethtool_tunable *tuna, const void *data);
 int smsc_phy_probe(struct phy_device *phydev);
 
 #endif /* __LINUX_SMSCPHY_H__ */
-- 
cgit v1.2.3-70-g09d2


From 3c4c3b3e6d414c50f6858f027a2ca2206e1ce583 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Sun, 2 Apr 2023 17:17:35 +0200
Subject: net: phy: smsc: enable edpd tunable support

Enable EDPD PHY tunable support for all drivers using
lan87xx_read_status.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/smsc.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 0eba69ad5745..692930750215 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -469,6 +469,9 @@ static struct phy_driver smsc_phy_driver[] = {
 	.get_strings	= smsc_get_strings,
 	.get_stats	= smsc_get_stats,
 
+	.get_tunable	= smsc_phy_get_tunable,
+	.set_tunable	= smsc_phy_set_tunable,
+
 	.suspend	= genphy_suspend,
 	.resume		= genphy_resume,
 }, {
@@ -513,6 +516,9 @@ static struct phy_driver smsc_phy_driver[] = {
 	.get_strings	= smsc_get_strings,
 	.get_stats	= smsc_get_stats,
 
+	.get_tunable	= smsc_phy_get_tunable,
+	.set_tunable	= smsc_phy_set_tunable,
+
 	.suspend	= genphy_suspend,
 	.resume		= genphy_resume,
 }, {
@@ -539,6 +545,9 @@ static struct phy_driver smsc_phy_driver[] = {
 	.get_strings	= smsc_get_strings,
 	.get_stats	= smsc_get_stats,
 
+	.get_tunable	= smsc_phy_get_tunable,
+	.set_tunable	= smsc_phy_set_tunable,
+
 	.suspend	= genphy_suspend,
 	.resume		= genphy_resume,
 }, {
@@ -569,6 +578,9 @@ static struct phy_driver smsc_phy_driver[] = {
 	.get_strings	= smsc_get_strings,
 	.get_stats	= smsc_get_stats,
 
+	.get_tunable	= smsc_phy_get_tunable,
+	.set_tunable	= smsc_phy_set_tunable,
+
 	.suspend	= genphy_suspend,
 	.resume		= genphy_resume,
 } };
-- 
cgit v1.2.3-70-g09d2


From b6f56cddb5f57a0b8da0ce582232a2f1933558c6 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:17:19 +0100
Subject: net: dsa: mt7530: make some noise if register read fails

Simply returning the negative error value instead of the read value
doesn't seem like a good idea. Return 0 instead and add WARN_ON_ONCE(1)
so this kind of error will not go unnoticed.

Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index a0d99af897ac..18d4aa6bb996 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -224,9 +224,10 @@ mt7530_mii_read(struct mt7530_priv *priv, u32 reg)
 	/* MT7530 uses 31 as the pseudo port */
 	ret = bus->write(bus, 0x1f, 0x1f, page);
 	if (ret < 0) {
+		WARN_ON_ONCE(1);
 		dev_err(&bus->dev,
 			"failed to read mt7530 register\n");
-		return ret;
+		return 0;
 	}
 
 	lo = bus->read(bus, 0x1f, r);
-- 
cgit v1.2.3-70-g09d2


From 9ecc00164dc2300dfcd40afe549a8ee951dfea9f Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:17:30 +0100
Subject: net: dsa: mt7530: refactor SGMII PCS creation

Instead of macro templates use a dedidated function and allocated
regmap_config when creating the regmaps for the pcs-mtk-lynxi
instances.
This is in preparation to switching to use unlocked regmap accessors
and have regmap's locking API handle locking for us.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 74 ++++++++++++++++++++++++++++++++----------------
 1 file changed, 50 insertions(+), 24 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 18d4aa6bb996..5685c71bc917 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -2927,26 +2927,56 @@ static const struct regmap_bus mt7531_regmap_bus = {
 	.reg_update_bits = mt7530_regmap_update_bits,
 };
 
-#define MT7531_PCS_REGMAP_CONFIG(_name, _reg_base) \
-	{				\
-		.name = _name,		\
-		.reg_bits = 16,		\
-		.val_bits = 32,		\
-		.reg_stride = 4,	\
-		.reg_base = _reg_base,	\
-		.max_register = 0x17c,	\
-	}
-
-static const struct regmap_config mt7531_pcs_config[] = {
-	MT7531_PCS_REGMAP_CONFIG("port5", MT7531_SGMII_REG_BASE(5)),
-	MT7531_PCS_REGMAP_CONFIG("port6", MT7531_SGMII_REG_BASE(6)),
-};
+static int
+mt7531_create_sgmii(struct mt7530_priv *priv)
+{
+	struct regmap_config *mt7531_pcs_config[2];
+	struct phylink_pcs *pcs;
+	struct regmap *regmap;
+	int i, ret = 0;
+
+	for (i = 0; i < 2; i++) {
+		mt7531_pcs_config[i] = devm_kzalloc(priv->dev,
+						    sizeof(struct regmap_config),
+						    GFP_KERNEL);
+		if (!mt7531_pcs_config[i]) {
+			ret = -ENOMEM;
+			break;
+		}
+
+		mt7531_pcs_config[i]->name = i ? "port6" : "port5";
+		mt7531_pcs_config[i]->reg_bits = 16;
+		mt7531_pcs_config[i]->val_bits = 32;
+		mt7531_pcs_config[i]->reg_stride = 4;
+		mt7531_pcs_config[i]->reg_base = MT7531_SGMII_REG_BASE(5 + i);
+		mt7531_pcs_config[i]->max_register = 0x17c;
+
+		regmap = devm_regmap_init(priv->dev,
+					  &mt7531_regmap_bus, priv,
+					  mt7531_pcs_config[i]);
+		if (IS_ERR(regmap)) {
+			ret = PTR_ERR(regmap);
+			break;
+		}
+		pcs = mtk_pcs_lynxi_create(priv->dev, regmap,
+					   MT7531_PHYA_CTRL_SIGNAL3, 0);
+		if (!pcs) {
+			ret = -ENXIO;
+			break;
+		}
+		priv->ports[5 + i].sgmii_pcs = pcs;
+	}
+
+	if (ret && i)
+		mtk_pcs_lynxi_destroy(priv->ports[5].sgmii_pcs);
+
+	return ret;
+}
 
 static int
 mt753x_setup(struct dsa_switch *ds)
 {
 	struct mt7530_priv *priv = ds->priv;
-	struct regmap *regmap;
 	int i, ret;
 
 	/* Initialise the PCS devices */
@@ -2968,15 +2998,11 @@ mt753x_setup(struct dsa_switch *ds)
 	if (ret && priv->irq)
 		mt7530_free_irq_common(priv);
 
-	if (priv->id == ID_MT7531)
-		for (i = 0; i < 2; i++) {
-			regmap = devm_regmap_init(ds->dev,
-						  &mt7531_regmap_bus, priv,
-						  &mt7531_pcs_config[i]);
-			priv->ports[5 + i].sgmii_pcs =
-				mtk_pcs_lynxi_create(ds->dev, regmap,
-						     MT7531_PHYA_CTRL_SIGNAL3, 0);
-		}
+	if (priv->id == ID_MT7531) {
+		ret = mt7531_create_sgmii(priv);
+		if (ret && priv->irq)
+			mt7530_free_irq_common(priv);
+	}
 
 	return ret;
 }
-- 
cgit v1.2.3-70-g09d2


From 1bd099c49f65ed923b9f19b8c4b3cd1ff0024091 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:17:40 +0100
Subject: net: dsa: mt7530: use unlocked regmap accessors

Instead of wrapping the locked register accessor functions, use the
unlocked variants and add locking wrapper functions to let regmap
handle the locking.

This is a preparation towards being able to always use regmap to
access switch registers instead of open-coded accessor functions.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 5685c71bc917..d8b041d79f2b 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -2900,7 +2900,7 @@ static int mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val
 {
 	struct mt7530_priv *priv = context;
 
-	*val = mt7530_read(priv, reg);
+	*val = mt7530_mii_read(priv, reg);
 	return 0;
 };
 
@@ -2908,23 +2908,25 @@ static int mt7530_regmap_write(void *context, unsigned int reg, unsigned int val
 {
 	struct mt7530_priv *priv = context;
 
-	mt7530_write(priv, reg, val);
+	mt7530_mii_write(priv, reg, val);
 	return 0;
 };
 
-static int mt7530_regmap_update_bits(void *context, unsigned int reg,
-				     unsigned int mask, unsigned int val)
+static void
+mt7530_mdio_regmap_lock(void *mdio_lock)
 {
-	struct mt7530_priv *priv = context;
+	mutex_lock_nested(mdio_lock, MDIO_MUTEX_NESTED);
+}
 
-	mt7530_rmw(priv, reg, mask, val);
-	return 0;
-};
+static void
+mt7530_mdio_regmap_unlock(void *mdio_lock)
+{
+	mutex_unlock(mdio_lock);
+}
 
 static const struct regmap_bus mt7531_regmap_bus = {
 	.reg_write = mt7530_regmap_write,
 	.reg_read = mt7530_regmap_read,
-	.reg_update_bits = mt7530_regmap_update_bits,
 };
 
 static int
@@ -2950,6 +2952,9 @@ mt7531_create_sgmii(struct mt7530_priv *priv)
 		mt7531_pcs_config[i]->reg_stride = 4;
 		mt7531_pcs_config[i]->reg_base = MT7531_SGMII_REG_BASE(5 + i);
 		mt7531_pcs_config[i]->max_register = 0x17c;
+		mt7531_pcs_config[i]->lock = mt7530_mdio_regmap_lock;
+		mt7531_pcs_config[i]->unlock = mt7530_mdio_regmap_unlock;
+		mt7531_pcs_config[i]->lock_arg = &priv->bus->mdio_lock;
 
 		regmap = devm_regmap_init(priv->dev,
 					  &mt7531_regmap_bus, priv,
-- 
cgit v1.2.3-70-g09d2


From a08c045580e060a6886bbb656c50ae20b0c780b5 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:17:52 +0100
Subject: net: dsa: mt7530: use regmap to access switch register space

Use regmap API to access the switch register space.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 99 +++++++++++++++++++++++++++++-------------------
 drivers/net/dsa/mt7530.h |  2 +
 2 files changed, 62 insertions(+), 39 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index d8b041d79f2b..a0e1e2e015f0 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -183,9 +183,9 @@ core_clear(struct mt7530_priv *priv, u32 reg, u32 val)
 }
 
 static int
-mt7530_mii_write(struct mt7530_priv *priv, u32 reg, u32 val)
+mt7530_regmap_write(void *context, unsigned int reg, unsigned int val)
 {
-	struct mii_bus *bus = priv->bus;
+	struct mii_bus *bus = context;
 	u16 page, r, lo, hi;
 	int ret;
 
@@ -197,24 +197,34 @@ mt7530_mii_write(struct mt7530_priv *priv, u32 reg, u32 val)
 	/* MT7530 uses 31 as the pseudo port */
 	ret = bus->write(bus, 0x1f, 0x1f, page);
 	if (ret < 0)
-		goto err;
+		return ret;
 
 	ret = bus->write(bus, 0x1f, r,  lo);
 	if (ret < 0)
-		goto err;
+		return ret;
 
 	ret = bus->write(bus, 0x1f, 0x10, hi);
-err:
+	return ret;
+}
+
+static int
+mt7530_mii_write(struct mt7530_priv *priv, u32 reg, u32 val)
+{
+	int ret;
+
+	ret = regmap_write(priv->regmap, reg, val);
+
 	if (ret < 0)
-		dev_err(&bus->dev,
+		dev_err(priv->dev,
 			"failed to write mt7530 register\n");
+
 	return ret;
 }
 
-static u32
-mt7530_mii_read(struct mt7530_priv *priv, u32 reg)
+static int
+mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val)
 {
-	struct mii_bus *bus = priv->bus;
+	struct mii_bus *bus = context;
 	u16 page, r, lo, hi;
 	int ret;
 
@@ -223,17 +233,32 @@ mt7530_mii_read(struct mt7530_priv *priv, u32 reg)
 
 	/* MT7530 uses 31 as the pseudo port */
 	ret = bus->write(bus, 0x1f, 0x1f, page);
-	if (ret < 0) {
+	if (ret < 0)
+		return ret;
+
+	lo = bus->read(bus, 0x1f, r);
+	hi = bus->read(bus, 0x1f, 0x10);
+
+	*val = (hi << 16) | (lo & 0xffff);
+
+	return 0;
+}
+
+static u32
+mt7530_mii_read(struct mt7530_priv *priv, u32 reg)
+{
+	int ret;
+	u32 val;
+
+	ret = regmap_read(priv->regmap, reg, &val);
+	if (ret) {
 		WARN_ON_ONCE(1);
-		dev_err(&bus->dev,
+		dev_err(priv->dev,
 			"failed to read mt7530 register\n");
 		return 0;
 	}
 
-	lo = bus->read(bus, 0x1f, r);
-	hi = bus->read(bus, 0x1f, 0x10);
-
-	return (hi << 16) | (lo & 0xffff);
+	return val;
 }
 
 static void
@@ -283,14 +308,10 @@ mt7530_rmw(struct mt7530_priv *priv, u32 reg,
 	   u32 mask, u32 set)
 {
 	struct mii_bus *bus = priv->bus;
-	u32 val;
 
 	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
 
-	val = mt7530_mii_read(priv, reg);
-	val &= ~mask;
-	val |= set;
-	mt7530_mii_write(priv, reg, val);
+	regmap_update_bits(priv->regmap, reg, mask, set);
 
 	mutex_unlock(&bus->mdio_lock);
 }
@@ -298,7 +319,7 @@ mt7530_rmw(struct mt7530_priv *priv, u32 reg,
 static void
 mt7530_set(struct mt7530_priv *priv, u32 reg, u32 val)
 {
-	mt7530_rmw(priv, reg, 0, val);
+	mt7530_rmw(priv, reg, val, val);
 }
 
 static void
@@ -2896,22 +2917,6 @@ static const struct phylink_pcs_ops mt7530_pcs_ops = {
 	.pcs_an_restart = mt7530_pcs_an_restart,
 };
 
-static int mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val)
-{
-	struct mt7530_priv *priv = context;
-
-	*val = mt7530_mii_read(priv, reg);
-	return 0;
-};
-
-static int mt7530_regmap_write(void *context, unsigned int reg, unsigned int val)
-{
-	struct mt7530_priv *priv = context;
-
-	mt7530_mii_write(priv, reg, val);
-	return 0;
-};
-
 static void
 mt7530_mdio_regmap_lock(void *mdio_lock)
 {
@@ -2924,7 +2929,7 @@ mt7530_mdio_regmap_unlock(void *mdio_lock)
 	mutex_unlock(mdio_lock);
 }
 
-static const struct regmap_bus mt7531_regmap_bus = {
+static const struct regmap_bus mt7530_regmap_bus = {
 	.reg_write = mt7530_regmap_write,
 	.reg_read = mt7530_regmap_read,
 };
@@ -2957,7 +2962,7 @@ mt7531_create_sgmii(struct mt7530_priv *priv)
 		mt7531_pcs_config[i]->lock_arg = &priv->bus->mdio_lock;
 
 		regmap = devm_regmap_init(priv->dev,
-					  &mt7531_regmap_bus, priv,
+					  &mt7530_regmap_bus, priv->bus,
 					  mt7531_pcs_config[i]);
 		if (IS_ERR(regmap)) {
 			ret = PTR_ERR(regmap);
@@ -3128,6 +3133,7 @@ MODULE_DEVICE_TABLE(of, mt7530_of_match);
 static int
 mt7530_probe(struct mdio_device *mdiodev)
 {
+	static struct regmap_config *regmap_config;
 	struct mt7530_priv *priv;
 	struct device_node *dn;
 
@@ -3207,6 +3213,21 @@ mt7530_probe(struct mdio_device *mdiodev)
 	mutex_init(&priv->reg_mutex);
 	dev_set_drvdata(&mdiodev->dev, priv);
 
+	regmap_config = devm_kzalloc(&mdiodev->dev, sizeof(*regmap_config),
+				     GFP_KERNEL);
+	if (!regmap_config)
+		return -ENOMEM;
+
+	regmap_config->reg_bits = 16;
+	regmap_config->val_bits = 32;
+	regmap_config->reg_stride = 4;
+	regmap_config->max_register = MT7530_CREV;
+	regmap_config->disable_locking = true;
+	priv->regmap = devm_regmap_init(priv->dev, &mt7530_regmap_bus,
+					priv->bus, regmap_config);
+	if (IS_ERR(priv->regmap))
+		return PTR_ERR(priv->regmap);
+
 	return dsa_register_switch(priv->ds);
 }
 
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index c5d29f3fc1d8..39aaca50961b 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -754,6 +754,7 @@ struct mt753x_info {
  * @dev:		The device pointer
  * @ds:			The pointer to the dsa core structure
  * @bus:		The bus used for the device and built-in PHY
+ * @regmap:		The regmap instance representing all switch registers
  * @rstc:		The pointer to reset control used by MCM
  * @core_pwr:		The power supplied into the core
  * @io_pwr:		The power supplied into the I/O
@@ -774,6 +775,7 @@ struct mt7530_priv {
 	struct device		*dev;
 	struct dsa_switch	*ds;
 	struct mii_bus		*bus;
+	struct regmap		*regmap;
 	struct reset_control	*rstc;
 	struct regulator	*core_pwr;
 	struct regulator	*io_pwr;
-- 
cgit v1.2.3-70-g09d2


From 6de2852297737171ba96b91e89bf302ca1fda869 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:18:04 +0100
Subject: net: dsa: mt7530: move SGMII PCS creation to mt7530_probe function

Move creating the SGMII PCS from mt753x_setup() to the more appropriate
mt7530_probe() function.
This is done also in preparation of moving all functions related to
MDIO-connected MT753x switches to a separate module.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index a0e1e2e015f0..d285a60aaf68 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -3008,12 +3008,6 @@ mt753x_setup(struct dsa_switch *ds)
 	if (ret && priv->irq)
 		mt7530_free_irq_common(priv);
 
-	if (priv->id == ID_MT7531) {
-		ret = mt7531_create_sgmii(priv);
-		if (ret && priv->irq)
-			mt7530_free_irq_common(priv);
-	}
-
 	return ret;
 }
 
@@ -3136,6 +3130,7 @@ mt7530_probe(struct mdio_device *mdiodev)
 	static struct regmap_config *regmap_config;
 	struct mt7530_priv *priv;
 	struct device_node *dn;
+	int ret;
 
 	dn = mdiodev->dev.of_node;
 
@@ -3228,6 +3223,12 @@ mt7530_probe(struct mdio_device *mdiodev)
 	if (IS_ERR(priv->regmap))
 		return PTR_ERR(priv->regmap);
 
+	if (priv->id == ID_MT7531) {
+		ret = mt7531_create_sgmii(priv);
+		if (ret)
+			return ret;
+	}
+
 	return dsa_register_switch(priv->ds);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 1557c679f71c82a994eae0baadbaeb62b71e15bf Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:18:16 +0100
Subject: net: dsa: mt7530: introduce mutex helpers

As the MDIO bus lock only needs to be involved if actually operating
on an MDIO-connected switch we will need to skip locking for built-in
switches which are accessed via MMIO.
Create helper functions which simplify that upcoming change.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 73 ++++++++++++++++++++++++------------------------
 1 file changed, 36 insertions(+), 37 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index d285a60aaf68..9d49cbc8ddcb 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -143,31 +143,40 @@ err:
 }
 
 static void
-core_write(struct mt7530_priv *priv, u32 reg, u32 val)
+mt7530_mutex_lock(struct mt7530_priv *priv)
 {
-	struct mii_bus *bus = priv->bus;
+	mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
+}
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+static void
+mt7530_mutex_unlock(struct mt7530_priv *priv)
+{
+	mutex_unlock(&priv->bus->mdio_lock);
+}
+
+static void
+core_write(struct mt7530_priv *priv, u32 reg, u32 val)
+{
+	mt7530_mutex_lock(priv);
 
 	core_write_mmd_indirect(priv, reg, MDIO_MMD_VEND2, val);
 
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 }
 
 static void
 core_rmw(struct mt7530_priv *priv, u32 reg, u32 mask, u32 set)
 {
-	struct mii_bus *bus = priv->bus;
 	u32 val;
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	val = core_read_mmd_indirect(priv, reg, MDIO_MMD_VEND2);
 	val &= ~mask;
 	val |= set;
 	core_write_mmd_indirect(priv, reg, MDIO_MMD_VEND2, val);
 
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 }
 
 static void
@@ -264,13 +273,11 @@ mt7530_mii_read(struct mt7530_priv *priv, u32 reg)
 static void
 mt7530_write(struct mt7530_priv *priv, u32 reg, u32 val)
 {
-	struct mii_bus *bus = priv->bus;
-
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	mt7530_mii_write(priv, reg, val);
 
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 }
 
 static u32
@@ -282,14 +289,13 @@ _mt7530_unlocked_read(struct mt7530_dummy_poll *p)
 static u32
 _mt7530_read(struct mt7530_dummy_poll *p)
 {
-	struct mii_bus		*bus = p->priv->bus;
 	u32 val;
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(p->priv);
 
 	val = mt7530_mii_read(p->priv, p->reg);
 
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(p->priv);
 
 	return val;
 }
@@ -307,13 +313,11 @@ static void
 mt7530_rmw(struct mt7530_priv *priv, u32 reg,
 	   u32 mask, u32 set)
 {
-	struct mii_bus *bus = priv->bus;
-
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	regmap_update_bits(priv->regmap, reg, mask, set);
 
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 }
 
 static void
@@ -657,14 +661,13 @@ static int
 mt7531_ind_c45_phy_read(struct mt7530_priv *priv, int port, int devad,
 			int regnum)
 {
-	struct mii_bus *bus = priv->bus;
 	struct mt7530_dummy_poll p;
 	u32 reg, val;
 	int ret;
 
 	INIT_MT7530_DUMMY_POLL(&p, priv, MT7531_PHY_IAC);
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	ret = readx_poll_timeout(_mt7530_unlocked_read, &p, val,
 				 !(val & MT7531_PHY_ACS_ST), 20, 100000);
@@ -697,7 +700,7 @@ mt7531_ind_c45_phy_read(struct mt7530_priv *priv, int port, int devad,
 
 	ret = val & MT7531_MDIO_RW_DATA_MASK;
 out:
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 
 	return ret;
 }
@@ -706,14 +709,13 @@ static int
 mt7531_ind_c45_phy_write(struct mt7530_priv *priv, int port, int devad,
 			 int regnum, u16 data)
 {
-	struct mii_bus *bus = priv->bus;
 	struct mt7530_dummy_poll p;
 	u32 val, reg;
 	int ret;
 
 	INIT_MT7530_DUMMY_POLL(&p, priv, MT7531_PHY_IAC);
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	ret = readx_poll_timeout(_mt7530_unlocked_read, &p, val,
 				 !(val & MT7531_PHY_ACS_ST), 20, 100000);
@@ -745,7 +747,7 @@ mt7531_ind_c45_phy_write(struct mt7530_priv *priv, int port, int devad,
 	}
 
 out:
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 
 	return ret;
 }
@@ -753,14 +755,13 @@ out:
 static int
 mt7531_ind_c22_phy_read(struct mt7530_priv *priv, int port, int regnum)
 {
-	struct mii_bus *bus = priv->bus;
 	struct mt7530_dummy_poll p;
 	int ret;
 	u32 val;
 
 	INIT_MT7530_DUMMY_POLL(&p, priv, MT7531_PHY_IAC);
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	ret = readx_poll_timeout(_mt7530_unlocked_read, &p, val,
 				 !(val & MT7531_PHY_ACS_ST), 20, 100000);
@@ -783,7 +784,7 @@ mt7531_ind_c22_phy_read(struct mt7530_priv *priv, int port, int regnum)
 
 	ret = val & MT7531_MDIO_RW_DATA_MASK;
 out:
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 
 	return ret;
 }
@@ -792,14 +793,13 @@ static int
 mt7531_ind_c22_phy_write(struct mt7530_priv *priv, int port, int regnum,
 			 u16 data)
 {
-	struct mii_bus *bus = priv->bus;
 	struct mt7530_dummy_poll p;
 	int ret;
 	u32 reg;
 
 	INIT_MT7530_DUMMY_POLL(&p, priv, MT7531_PHY_IAC);
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	ret = readx_poll_timeout(_mt7530_unlocked_read, &p, reg,
 				 !(reg & MT7531_PHY_ACS_ST), 20, 100000);
@@ -821,7 +821,7 @@ mt7531_ind_c22_phy_write(struct mt7530_priv *priv, int port, int regnum,
 	}
 
 out:
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 
 	return ret;
 }
@@ -1102,7 +1102,6 @@ static int
 mt7530_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
 {
 	struct mt7530_priv *priv = ds->priv;
-	struct mii_bus *bus = priv->bus;
 	int length;
 	u32 val;
 
@@ -1113,7 +1112,7 @@ mt7530_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
 	if (!dsa_is_cpu_port(ds, port))
 		return 0;
 
-	mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 
 	val = mt7530_mii_read(priv, MT7530_GMACCR);
 	val &= ~MAX_RX_PKT_LEN_MASK;
@@ -1134,7 +1133,7 @@ mt7530_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
 
 	mt7530_mii_write(priv, MT7530_GMACCR, val);
 
-	mutex_unlock(&bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 
 	return 0;
 }
@@ -1935,10 +1934,10 @@ mt7530_irq_thread_fn(int irq, void *dev_id)
 	u32 val;
 	int p;
 
-	mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 	val = mt7530_mii_read(priv, MT7530_SYS_INT_STS);
 	mt7530_mii_write(priv, MT7530_SYS_INT_STS, val);
-	mutex_unlock(&priv->bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 
 	for (p = 0; p < MT7530_NUM_PHYS; p++) {
 		if (BIT(p) & val) {
@@ -1974,7 +1973,7 @@ mt7530_irq_bus_lock(struct irq_data *d)
 {
 	struct mt7530_priv *priv = irq_data_get_irq_chip_data(d);
 
-	mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
+	mt7530_mutex_lock(priv);
 }
 
 static void
@@ -1983,7 +1982,7 @@ mt7530_irq_bus_sync_unlock(struct irq_data *d)
 	struct mt7530_priv *priv = irq_data_get_irq_chip_data(d);
 
 	mt7530_mii_write(priv, MT7530_SYS_INT_EN, priv->irq_enable);
-	mutex_unlock(&priv->bus->mdio_lock);
+	mt7530_mutex_unlock(priv);
 }
 
 static struct irq_chip mt7530_irq_chip = {
-- 
cgit v1.2.3-70-g09d2


From 25d15dee34a1a40d5fd71636a205e3211f09fd1d Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:18:28 +0100
Subject: net: dsa: mt7530: move p5_intf_modes() function to mt7530.c

In preparation of splitting mt7530.c into a driver for MDIO-connected
as well as MDIO-accessed built-in switches on one hand and MMIO-accessed
built-in switches move the p5_inft_modes() function from mt7530.h to
mt7530.c. The function is only needed there and will trigger a compiler
warning about a defined but unused function otherwise when including
mt7530.h in the to-be-introduced bus-specific drivers.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 18 ++++++++++++++++++
 drivers/net/dsa/mt7530.h | 18 ------------------
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 9d49cbc8ddcb..1215d5e4dd38 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -943,6 +943,24 @@ mt7530_set_ageing_time(struct dsa_switch *ds, unsigned int msecs)
 	return 0;
 }
 
+static const char *p5_intf_modes(unsigned int p5_interface)
+{
+	switch (p5_interface) {
+	case P5_DISABLED:
+		return "DISABLED";
+	case P5_INTF_SEL_PHY_P0:
+		return "PHY P0";
+	case P5_INTF_SEL_PHY_P4:
+		return "PHY P4";
+	case P5_INTF_SEL_GMAC5:
+		return "GMAC5";
+	case P5_INTF_SEL_GMAC5_SGMII:
+		return "GMAC5_SGMII";
+	default:
+		return "unknown";
+	}
+}
+
 static void mt7530_setup_port5(struct dsa_switch *ds, phy_interface_t interface)
 {
 	struct mt7530_priv *priv = ds->priv;
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index 39aaca50961b..2a611173a7d0 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -682,24 +682,6 @@ enum p5_interface_select {
 	P5_INTF_SEL_GMAC5_SGMII,
 };
 
-static const char *p5_intf_modes(unsigned int p5_interface)
-{
-	switch (p5_interface) {
-	case P5_DISABLED:
-		return "DISABLED";
-	case P5_INTF_SEL_PHY_P0:
-		return "PHY P0";
-	case P5_INTF_SEL_PHY_P4:
-		return "PHY P4";
-	case P5_INTF_SEL_GMAC5:
-		return "GMAC5";
-	case P5_INTF_SEL_GMAC5_SGMII:
-		return "GMAC5_SGMII";
-	default:
-		return "unknown";
-	}
-}
-
 struct mt7530_priv;
 
 struct mt753x_pcs {
-- 
cgit v1.2.3-70-g09d2


From 37c9c0d8d0b2e24f8c9af72ecd4edd31537284d3 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:18:39 +0100
Subject: net: dsa: mt7530: introduce mt7530_probe_common helper function

Move commonly used parts from mt7530_probe into new mt7530_probe_common
helper function which will be used by both, mt7530_probe and the
to-be-introduced mt7988_probe.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 98 ++++++++++++++++++++++++++----------------------
 1 file changed, 54 insertions(+), 44 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 1215d5e4dd38..a60ed8706ce3 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -3142,44 +3142,21 @@ static const struct of_device_id mt7530_of_match[] = {
 MODULE_DEVICE_TABLE(of, mt7530_of_match);
 
 static int
-mt7530_probe(struct mdio_device *mdiodev)
+mt7530_probe_common(struct mt7530_priv *priv)
 {
-	static struct regmap_config *regmap_config;
-	struct mt7530_priv *priv;
-	struct device_node *dn;
-	int ret;
-
-	dn = mdiodev->dev.of_node;
-
-	priv = devm_kzalloc(&mdiodev->dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
+	struct device *dev = priv->dev;
 
-	priv->ds = devm_kzalloc(&mdiodev->dev, sizeof(*priv->ds), GFP_KERNEL);
+	priv->ds = devm_kzalloc(dev, sizeof(*priv->ds), GFP_KERNEL);
 	if (!priv->ds)
 		return -ENOMEM;
 
-	priv->ds->dev = &mdiodev->dev;
+	priv->ds->dev = dev;
 	priv->ds->num_ports = MT7530_NUM_PORTS;
 
-	/* Use medatek,mcm property to distinguish hardware type that would
-	 * casues a little bit differences on power-on sequence.
-	 */
-	priv->mcm = of_property_read_bool(dn, "mediatek,mcm");
-	if (priv->mcm) {
-		dev_info(&mdiodev->dev, "MT7530 adapts as multi-chip module\n");
-
-		priv->rstc = devm_reset_control_get(&mdiodev->dev, "mcm");
-		if (IS_ERR(priv->rstc)) {
-			dev_err(&mdiodev->dev, "Couldn't get our reset line\n");
-			return PTR_ERR(priv->rstc);
-		}
-	}
-
 	/* Get the hardware identifier from the devicetree node.
 	 * We will need it for some of the clock and regulator setup.
 	 */
-	priv->info = of_device_get_match_data(&mdiodev->dev);
+	priv->info = of_device_get_match_data(dev);
 	if (!priv->info)
 		return -EINVAL;
 
@@ -3193,23 +3170,53 @@ mt7530_probe(struct mdio_device *mdiodev)
 		return -EINVAL;
 
 	priv->id = priv->info->id;
+	priv->dev = dev;
+	priv->ds->priv = priv;
+	priv->ds->ops = &mt7530_switch_ops;
+	mutex_init(&priv->reg_mutex);
+	dev_set_drvdata(dev, priv);
 
-	if (priv->id == ID_MT7530) {
-		priv->core_pwr = devm_regulator_get(&mdiodev->dev, "core");
-		if (IS_ERR(priv->core_pwr))
-			return PTR_ERR(priv->core_pwr);
+	return 0;
+}
 
-		priv->io_pwr = devm_regulator_get(&mdiodev->dev, "io");
-		if (IS_ERR(priv->io_pwr))
-			return PTR_ERR(priv->io_pwr);
-	}
+static int
+mt7530_probe(struct mdio_device *mdiodev)
+{
+	static struct regmap_config *regmap_config;
+	struct mt7530_priv *priv;
+	struct device_node *dn;
+	int ret;
+
+	dn = mdiodev->dev.of_node;
 
-	/* Not MCM that indicates switch works as the remote standalone
+	priv = devm_kzalloc(&mdiodev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->bus = mdiodev->bus;
+	priv->dev = &mdiodev->dev;
+
+	ret = mt7530_probe_common(priv);
+	if (ret)
+		return ret;
+
+	/* Use medatek,mcm property to distinguish hardware type that would
+	 * cause a little bit differences on power-on sequence.
+	 * Not MCM that indicates switch works as the remote standalone
 	 * integrated circuit so the GPIO pin would be used to complete
 	 * the reset, otherwise memory-mapped register accessing used
 	 * through syscon provides in the case of MCM.
 	 */
-	if (!priv->mcm) {
+	priv->mcm = of_property_read_bool(dn, "mediatek,mcm");
+	if (priv->mcm) {
+		dev_info(&mdiodev->dev, "MT7530 adapts as multi-chip module\n");
+
+		priv->rstc = devm_reset_control_get(&mdiodev->dev, "mcm");
+		if (IS_ERR(priv->rstc)) {
+			dev_err(&mdiodev->dev, "Couldn't get our reset line\n");
+			return PTR_ERR(priv->rstc);
+		}
+	} else {
 		priv->reset = devm_gpiod_get_optional(&mdiodev->dev, "reset",
 						      GPIOD_OUT_LOW);
 		if (IS_ERR(priv->reset)) {
@@ -3218,12 +3225,15 @@ mt7530_probe(struct mdio_device *mdiodev)
 		}
 	}
 
-	priv->bus = mdiodev->bus;
-	priv->dev = &mdiodev->dev;
-	priv->ds->priv = priv;
-	priv->ds->ops = &mt7530_switch_ops;
-	mutex_init(&priv->reg_mutex);
-	dev_set_drvdata(&mdiodev->dev, priv);
+	if (priv->id == ID_MT7530) {
+		priv->core_pwr = devm_regulator_get(&mdiodev->dev, "core");
+		if (IS_ERR(priv->core_pwr))
+			return PTR_ERR(priv->core_pwr);
+
+		priv->io_pwr = devm_regulator_get(&mdiodev->dev, "io");
+		if (IS_ERR(priv->io_pwr))
+			return PTR_ERR(priv->io_pwr);
+	}
 
 	regmap_config = devm_kzalloc(&mdiodev->dev, sizeof(*regmap_config),
 				     GFP_KERNEL);
-- 
cgit v1.2.3-70-g09d2


From 720d736351761574af02ed093658ab60de60576c Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:18:50 +0100
Subject: net: dsa: mt7530: introduce mt7530_remove_common helper function

Move commonly used parts from mt7530_remove into new
mt7530_remove_common helper function which will be used by both,
mt7530_remove and the to-be-introduced mt7988_remove.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index a60ed8706ce3..9404675c4722 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -3259,6 +3259,17 @@ mt7530_probe(struct mdio_device *mdiodev)
 	return dsa_register_switch(priv->ds);
 }
 
+static void
+mt7530_remove_common(struct mt7530_priv *priv)
+{
+	if (priv->irq)
+		mt7530_free_irq(priv);
+
+	dsa_unregister_switch(priv->ds);
+
+	mutex_destroy(&priv->reg_mutex);
+}
+
 static void
 mt7530_remove(struct mdio_device *mdiodev)
 {
@@ -3278,15 +3289,10 @@ mt7530_remove(struct mdio_device *mdiodev)
 		dev_err(priv->dev, "Failed to disable io pwr: %d\n",
 			ret);
 
-	if (priv->irq)
-		mt7530_free_irq(priv);
-
-	dsa_unregister_switch(priv->ds);
+	mt7530_remove_common(priv);
 
 	for (i = 0; i < 2; ++i)
 		mtk_pcs_lynxi_destroy(priv->ports[5 + i].sgmii_pcs);
-
-	mutex_destroy(&priv->reg_mutex);
 }
 
 static void mt7530_shutdown(struct mdio_device *mdiodev)
-- 
cgit v1.2.3-70-g09d2


From 7f54cc9772ced2d76ac11832f0ada43798443ac9 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:19:02 +0100
Subject: net: dsa: mt7530: split-off common parts from mt7531_setup

MT7988 shares a significant part of the setup function with MT7531.
Split-off those parts into a shared function which is going to be used
also by mt7988_setup.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 99 +++++++++++++++++++++++++++---------------------
 1 file changed, 55 insertions(+), 44 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 9404675c4722..47e3c413ccf1 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -2347,12 +2347,65 @@ mt7530_setup(struct dsa_switch *ds)
 	return 0;
 }
 
+static int
+mt7531_setup_common(struct dsa_switch *ds)
+{
+	struct mt7530_priv *priv = ds->priv;
+	struct dsa_port *cpu_dp;
+	int ret, i;
+
+	/* BPDU to CPU port */
+	dsa_switch_for_each_cpu_port(cpu_dp, ds) {
+		mt7530_rmw(priv, MT7531_CFC, MT7531_CPU_PMAP_MASK,
+			   BIT(cpu_dp->index));
+		break;
+	}
+	mt7530_rmw(priv, MT753X_BPC, MT753X_BPDU_PORT_FW_MASK,
+		   MT753X_BPDU_CPU_ONLY);
+
+	/* Enable and reset MIB counters */
+	mt7530_mib_reset(ds);
+
+	for (i = 0; i < MT7530_NUM_PORTS; i++) {
+		/* Disable forwarding by default on all ports */
+		mt7530_rmw(priv, MT7530_PCR_P(i), PCR_MATRIX_MASK,
+			   PCR_MATRIX_CLR);
+
+		/* Disable learning by default on all ports */
+		mt7530_set(priv, MT7530_PSC_P(i), SA_DIS);
+
+		mt7530_set(priv, MT7531_DBG_CNT(i), MT7531_DIS_CLR);
+
+		if (dsa_is_cpu_port(ds, i)) {
+			ret = mt753x_cpu_port_enable(ds, i);
+			if (ret)
+				return ret;
+		} else {
+			mt7530_port_disable(ds, i);
+
+			/* Set default PVID to 0 on all user ports */
+			mt7530_rmw(priv, MT7530_PPBV1_P(i), G0_PORT_VID_MASK,
+				   G0_PORT_VID_DEF);
+		}
+
+		/* Enable consistent egress tag */
+		mt7530_rmw(priv, MT7530_PVC_P(i), PVC_EG_TAG_MASK,
+			   PVC_EG_TAG(MT7530_VLAN_EG_CONSISTENT));
+	}
+
+	/* Flush the FDB table */
+	ret = mt7530_fdb_cmd(priv, MT7530_FDB_FLUSH, NULL);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
 static int
 mt7531_setup(struct dsa_switch *ds)
 {
 	struct mt7530_priv *priv = ds->priv;
 	struct mt7530_dummy_poll p;
-	struct dsa_port *cpu_dp;
 	u32 val, id;
 	int ret, i;
 
@@ -2430,44 +2483,7 @@ mt7531_setup(struct dsa_switch *ds)
 	mt7531_ind_c45_phy_write(priv, MT753X_CTRL_PHY_ADDR, MDIO_MMD_VEND2,
 				 CORE_PLL_GROUP4, val);
 
-	/* BPDU to CPU port */
-	dsa_switch_for_each_cpu_port(cpu_dp, ds) {
-		mt7530_rmw(priv, MT7531_CFC, MT7531_CPU_PMAP_MASK,
-			   BIT(cpu_dp->index));
-		break;
-	}
-	mt7530_rmw(priv, MT753X_BPC, MT753X_BPDU_PORT_FW_MASK,
-		   MT753X_BPDU_CPU_ONLY);
-
-	/* Enable and reset MIB counters */
-	mt7530_mib_reset(ds);
-
-	for (i = 0; i < MT7530_NUM_PORTS; i++) {
-		/* Disable forwarding by default on all ports */
-		mt7530_rmw(priv, MT7530_PCR_P(i), PCR_MATRIX_MASK,
-			   PCR_MATRIX_CLR);
-
-		/* Disable learning by default on all ports */
-		mt7530_set(priv, MT7530_PSC_P(i), SA_DIS);
-
-		mt7530_set(priv, MT7531_DBG_CNT(i), MT7531_DIS_CLR);
-
-		if (dsa_is_cpu_port(ds, i)) {
-			ret = mt753x_cpu_port_enable(ds, i);
-			if (ret)
-				return ret;
-		} else {
-			mt7530_port_disable(ds, i);
-
-			/* Set default PVID to 0 on all user ports */
-			mt7530_rmw(priv, MT7530_PPBV1_P(i), G0_PORT_VID_MASK,
-				   G0_PORT_VID_DEF);
-		}
-
-		/* Enable consistent egress tag */
-		mt7530_rmw(priv, MT7530_PVC_P(i), PVC_EG_TAG_MASK,
-			   PVC_EG_TAG(MT7530_VLAN_EG_CONSISTENT));
-	}
+	mt7531_setup_common(ds);
 
 	/* Setup VLAN ID 0 for VLAN-unaware bridges */
 	ret = mt7530_setup_vlan0(priv);
@@ -2477,11 +2493,6 @@ mt7531_setup(struct dsa_switch *ds)
 	ds->assisted_learning_on_cpu_port = true;
 	ds->mtu_enforcement_ingress = true;
 
-	/* Flush the FDB table */
-	ret = mt7530_fdb_cmd(priv, MT7530_FDB_FLUSH, NULL);
-	if (ret < 0)
-		return ret;
-
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From cb675afcddbbeb2bfa6596e3bc236bc026cd425f Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:19:13 +0100
Subject: net: dsa: mt7530: introduce separate MDIO driver

Split MT7530 switch driver into a common part and a part specific
for MDIO connected switches and multi-chip modules.
Move MDIO-specific functions to newly introduced mt7530-mdio.c while
keeping the common parts in mt7530.c.
Introduce new Kconfig symbol CONFIG_NET_DSA_MT7530_MDIO which is
implied by CONFIG_NET_DSA_MT7530.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                   |   1 +
 drivers/net/dsa/Kconfig       |  15 ++-
 drivers/net/dsa/Makefile      |   1 +
 drivers/net/dsa/mt7530-mdio.c | 271 ++++++++++++++++++++++++++++++++++++++++++
 drivers/net/dsa/mt7530.c      | 264 ++--------------------------------------
 drivers/net/dsa/mt7530.h      |   6 +
 6 files changed, 300 insertions(+), 258 deletions(-)
 create mode 100644 drivers/net/dsa/mt7530-mdio.c

diff --git a/MAINTAINERS b/MAINTAINERS
index dffcc253563b..4970a6c6a0a8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -13177,6 +13177,7 @@ M:	Landen Chao <Landen.Chao@mediatek.com>
 M:	DENG Qingfang <dqfext@gmail.com>
 L:	netdev@vger.kernel.org
 S:	Maintained
+F:	drivers/net/dsa/mt7530-mdio.c
 F:	drivers/net/dsa/mt7530.*
 F:	net/dsa/tag_mtk.c
 
diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index 6b45fa8b6907..941d020fe4b6 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -38,11 +38,22 @@ config NET_DSA_MT7530
 	tristate "MediaTek MT7530 and MT7531 Ethernet switch support"
 	select NET_DSA_TAG_MTK
 	select MEDIATEK_GE_PHY
-	select PCS_MTK_LYNXI
+	imply NET_DSA_MT7530_MDIO
 	help
 	  This enables support for the MediaTek MT7530 and MT7531 Ethernet
 	  switch chips. Multi-chip module MT7530 in MT7621AT, MT7621DAT,
-	  MT7621ST and MT7623AI SoCs is supported.
+	  MT7621ST and MT7623AI SoCs, and built-in switch in MT7988 SoC are
+	  supported as well.
+
+config NET_DSA_MT7530_MDIO
+	tristate "MediaTek MT7530 MDIO interface driver"
+	depends on NET_DSA_MT7530
+	select PCS_MTK_LYNXI
+	help
+	  This enables support for the MediaTek MT7530 and MT7531 switch
+	  chips which are connected via MDIO, as well as multi-chip
+	  module MT7530 which can be found in the MT7621AT, MT7621DAT,
+	  MT7621ST and MT7623AI SoCs.
 
 config NET_DSA_MV88E6060
 	tristate "Marvell 88E6060 ethernet switch chip support"
diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile
index 16eb879e0cb4..c5da75cb0a6b 100644
--- a/drivers/net/dsa/Makefile
+++ b/drivers/net/dsa/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_FIXED_PHY)		+= dsa_loop_bdinfo.o
 endif
 obj-$(CONFIG_NET_DSA_LANTIQ_GSWIP) += lantiq_gswip.o
 obj-$(CONFIG_NET_DSA_MT7530)	+= mt7530.o
+obj-$(CONFIG_NET_DSA_MT7530_MDIO) += mt7530-mdio.o
 obj-$(CONFIG_NET_DSA_MV88E6060) += mv88e6060.o
 obj-$(CONFIG_NET_DSA_RZN1_A5PSW) += rzn1_a5psw.o
 obj-$(CONFIG_NET_DSA_SMSC_LAN9303) += lan9303-core.o
diff --git a/drivers/net/dsa/mt7530-mdio.c b/drivers/net/dsa/mt7530-mdio.c
new file mode 100644
index 000000000000..34a547b88e49
--- /dev/null
+++ b/drivers/net/dsa/mt7530-mdio.c
@@ -0,0 +1,271 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/gpio/consumer.h>
+#include <linux/mdio.h>
+#include <linux/module.h>
+#include <linux/pcs/pcs-mtk-lynxi.h>
+#include <linux/of_irq.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/of_platform.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+#include <linux/regulator/consumer.h>
+#include <net/dsa.h>
+
+#include "mt7530.h"
+
+static int
+mt7530_regmap_write(void *context, unsigned int reg, unsigned int val)
+{
+	struct mii_bus *bus = context;
+	u16 page, r, lo, hi;
+	int ret;
+
+	page = (reg >> 6) & 0x3ff;
+	r  = (reg >> 2) & 0xf;
+	lo = val & 0xffff;
+	hi = val >> 16;
+
+	/* MT7530 uses 31 as the pseudo port */
+	ret = bus->write(bus, 0x1f, 0x1f, page);
+	if (ret < 0)
+		return ret;
+
+	ret = bus->write(bus, 0x1f, r,  lo);
+	if (ret < 0)
+		return ret;
+
+	ret = bus->write(bus, 0x1f, 0x10, hi);
+	return ret;
+}
+
+static int
+mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val)
+{
+	struct mii_bus *bus = context;
+	u16 page, r, lo, hi;
+	int ret;
+
+	page = (reg >> 6) & 0x3ff;
+	r = (reg >> 2) & 0xf;
+
+	/* MT7530 uses 31 as the pseudo port */
+	ret = bus->write(bus, 0x1f, 0x1f, page);
+	if (ret < 0)
+		return ret;
+
+	lo = bus->read(bus, 0x1f, r);
+	hi = bus->read(bus, 0x1f, 0x10);
+
+	*val = (hi << 16) | (lo & 0xffff);
+
+	return 0;
+}
+
+static void
+mt7530_mdio_regmap_lock(void *mdio_lock)
+{
+	mutex_lock_nested(mdio_lock, MDIO_MUTEX_NESTED);
+}
+
+static void
+mt7530_mdio_regmap_unlock(void *mdio_lock)
+{
+	mutex_unlock(mdio_lock);
+}
+
+static const struct regmap_bus mt7530_regmap_bus = {
+	.reg_write = mt7530_regmap_write,
+	.reg_read = mt7530_regmap_read,
+};
+
+static int
+mt7531_create_sgmii(struct mt7530_priv *priv)
+{
+	struct regmap_config *mt7531_pcs_config[2];
+	struct phylink_pcs *pcs;
+	struct regmap *regmap;
+	int i, ret = 0;
+
+	for (i = 0; i < 2; i++) {
+		mt7531_pcs_config[i] = devm_kzalloc(priv->dev,
+						    sizeof(struct regmap_config),
+						    GFP_KERNEL);
+		if (!mt7531_pcs_config[i]) {
+			ret = -ENOMEM;
+			break;
+		}
+
+		mt7531_pcs_config[i]->name = i ? "port6" : "port5";
+		mt7531_pcs_config[i]->reg_bits = 16;
+		mt7531_pcs_config[i]->val_bits = 32;
+		mt7531_pcs_config[i]->reg_stride = 4;
+		mt7531_pcs_config[i]->reg_base = MT7531_SGMII_REG_BASE(5 + i);
+		mt7531_pcs_config[i]->max_register = 0x17c;
+		mt7531_pcs_config[i]->lock = mt7530_mdio_regmap_lock;
+		mt7531_pcs_config[i]->unlock = mt7530_mdio_regmap_unlock;
+		mt7531_pcs_config[i]->lock_arg = &priv->bus->mdio_lock;
+
+		regmap = devm_regmap_init(priv->dev,
+					  &mt7530_regmap_bus, priv->bus,
+					  mt7531_pcs_config[i]);
+		if (IS_ERR(regmap)) {
+			ret = PTR_ERR(regmap);
+			break;
+		}
+		pcs = mtk_pcs_lynxi_create(priv->dev, regmap,
+					   MT7531_PHYA_CTRL_SIGNAL3, 0);
+		if (!pcs) {
+			ret = -ENXIO;
+			break;
+		}
+		priv->ports[5 + i].sgmii_pcs = pcs;
+	}
+
+	if (ret && i)
+		mtk_pcs_lynxi_destroy(priv->ports[5].sgmii_pcs);
+
+	return ret;
+}
+
+static const struct of_device_id mt7530_of_match[] = {
+	{ .compatible = "mediatek,mt7621", .data = &mt753x_table[ID_MT7621], },
+	{ .compatible = "mediatek,mt7530", .data = &mt753x_table[ID_MT7530], },
+	{ .compatible = "mediatek,mt7531", .data = &mt753x_table[ID_MT7531], },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, mt7530_of_match);
+
+static int
+mt7530_probe(struct mdio_device *mdiodev)
+{
+	static struct regmap_config *regmap_config;
+	struct mt7530_priv *priv;
+	struct device_node *dn;
+	int ret;
+
+	dn = mdiodev->dev.of_node;
+
+	priv = devm_kzalloc(&mdiodev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->bus = mdiodev->bus;
+	priv->dev = &mdiodev->dev;
+
+	ret = mt7530_probe_common(priv);
+	if (ret)
+		return ret;
+
+	/* Use medatek,mcm property to distinguish hardware type that would
+	 * cause a little bit differences on power-on sequence.
+	 * Not MCM that indicates switch works as the remote standalone
+	 * integrated circuit so the GPIO pin would be used to complete
+	 * the reset, otherwise memory-mapped register accessing used
+	 * through syscon provides in the case of MCM.
+	 */
+	priv->mcm = of_property_read_bool(dn, "mediatek,mcm");
+	if (priv->mcm) {
+		dev_info(&mdiodev->dev, "MT7530 adapts as multi-chip module\n");
+
+		priv->rstc = devm_reset_control_get(&mdiodev->dev, "mcm");
+		if (IS_ERR(priv->rstc)) {
+			dev_err(&mdiodev->dev, "Couldn't get our reset line\n");
+			return PTR_ERR(priv->rstc);
+		}
+	} else {
+		priv->reset = devm_gpiod_get_optional(&mdiodev->dev, "reset",
+						      GPIOD_OUT_LOW);
+		if (IS_ERR(priv->reset)) {
+			dev_err(&mdiodev->dev, "Couldn't get our reset line\n");
+			return PTR_ERR(priv->reset);
+		}
+	}
+
+	if (priv->id == ID_MT7530) {
+		priv->core_pwr = devm_regulator_get(&mdiodev->dev, "core");
+		if (IS_ERR(priv->core_pwr))
+			return PTR_ERR(priv->core_pwr);
+
+		priv->io_pwr = devm_regulator_get(&mdiodev->dev, "io");
+		if (IS_ERR(priv->io_pwr))
+			return PTR_ERR(priv->io_pwr);
+	}
+
+	regmap_config = devm_kzalloc(&mdiodev->dev, sizeof(*regmap_config),
+				     GFP_KERNEL);
+	if (!regmap_config)
+		return -ENOMEM;
+
+	regmap_config->reg_bits = 16;
+	regmap_config->val_bits = 32;
+	regmap_config->reg_stride = 4;
+	regmap_config->max_register = MT7530_CREV;
+	regmap_config->disable_locking = true;
+	priv->regmap = devm_regmap_init(priv->dev, &mt7530_regmap_bus,
+					priv->bus, regmap_config);
+	if (IS_ERR(priv->regmap))
+		return PTR_ERR(priv->regmap);
+
+	if (priv->id == ID_MT7531) {
+		ret = mt7531_create_sgmii(priv);
+		if (ret)
+			return ret;
+	}
+
+	return dsa_register_switch(priv->ds);
+}
+
+static void
+mt7530_remove(struct mdio_device *mdiodev)
+{
+	struct mt7530_priv *priv = dev_get_drvdata(&mdiodev->dev);
+	int ret = 0, i;
+
+	if (!priv)
+		return;
+
+	ret = regulator_disable(priv->core_pwr);
+	if (ret < 0)
+		dev_err(priv->dev,
+			"Failed to disable core power: %d\n", ret);
+
+	ret = regulator_disable(priv->io_pwr);
+	if (ret < 0)
+		dev_err(priv->dev, "Failed to disable io pwr: %d\n",
+			ret);
+
+	mt7530_remove_common(priv);
+
+	for (i = 0; i < 2; ++i)
+		mtk_pcs_lynxi_destroy(priv->ports[5 + i].sgmii_pcs);
+}
+
+static void mt7530_shutdown(struct mdio_device *mdiodev)
+{
+	struct mt7530_priv *priv = dev_get_drvdata(&mdiodev->dev);
+
+	if (!priv)
+		return;
+
+	dsa_switch_shutdown(priv->ds);
+
+	dev_set_drvdata(&mdiodev->dev, NULL);
+}
+
+static struct mdio_driver mt7530_mdio_driver = {
+	.probe  = mt7530_probe,
+	.remove = mt7530_remove,
+	.shutdown = mt7530_shutdown,
+	.mdiodrv.driver = {
+		.name = "mt7530-mdio",
+		.of_match_table = mt7530_of_match,
+	},
+};
+
+mdio_module_driver(mt7530_mdio_driver);
+
+MODULE_AUTHOR("Sean Wang <sean.wang@mediatek.com>");
+MODULE_DESCRIPTION("Driver for Mediatek MT7530 Switch (MDIO)");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 47e3c413ccf1..ac485ee83ed0 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -14,7 +14,6 @@
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
 #include <linux/of_platform.h>
-#include <linux/pcs/pcs-mtk-lynxi.h>
 #include <linux/phylink.h>
 #include <linux/regmap.h>
 #include <linux/regulator/consumer.h>
@@ -191,31 +190,6 @@ core_clear(struct mt7530_priv *priv, u32 reg, u32 val)
 	core_rmw(priv, reg, val, 0);
 }
 
-static int
-mt7530_regmap_write(void *context, unsigned int reg, unsigned int val)
-{
-	struct mii_bus *bus = context;
-	u16 page, r, lo, hi;
-	int ret;
-
-	page = (reg >> 6) & 0x3ff;
-	r  = (reg >> 2) & 0xf;
-	lo = val & 0xffff;
-	hi = val >> 16;
-
-	/* MT7530 uses 31 as the pseudo port */
-	ret = bus->write(bus, 0x1f, 0x1f, page);
-	if (ret < 0)
-		return ret;
-
-	ret = bus->write(bus, 0x1f, r,  lo);
-	if (ret < 0)
-		return ret;
-
-	ret = bus->write(bus, 0x1f, 0x10, hi);
-	return ret;
-}
-
 static int
 mt7530_mii_write(struct mt7530_priv *priv, u32 reg, u32 val)
 {
@@ -230,29 +204,6 @@ mt7530_mii_write(struct mt7530_priv *priv, u32 reg, u32 val)
 	return ret;
 }
 
-static int
-mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val)
-{
-	struct mii_bus *bus = context;
-	u16 page, r, lo, hi;
-	int ret;
-
-	page = (reg >> 6) & 0x3ff;
-	r = (reg >> 2) & 0xf;
-
-	/* MT7530 uses 31 as the pseudo port */
-	ret = bus->write(bus, 0x1f, 0x1f, page);
-	if (ret < 0)
-		return ret;
-
-	lo = bus->read(bus, 0x1f, r);
-	hi = bus->read(bus, 0x1f, 0x10);
-
-	*val = (hi << 16) | (lo & 0xffff);
-
-	return 0;
-}
-
 static u32
 mt7530_mii_read(struct mt7530_priv *priv, u32 reg)
 {
@@ -2945,72 +2896,6 @@ static const struct phylink_pcs_ops mt7530_pcs_ops = {
 	.pcs_an_restart = mt7530_pcs_an_restart,
 };
 
-static void
-mt7530_mdio_regmap_lock(void *mdio_lock)
-{
-	mutex_lock_nested(mdio_lock, MDIO_MUTEX_NESTED);
-}
-
-static void
-mt7530_mdio_regmap_unlock(void *mdio_lock)
-{
-	mutex_unlock(mdio_lock);
-}
-
-static const struct regmap_bus mt7530_regmap_bus = {
-	.reg_write = mt7530_regmap_write,
-	.reg_read = mt7530_regmap_read,
-};
-
-static int
-mt7531_create_sgmii(struct mt7530_priv *priv)
-{
-	struct regmap_config *mt7531_pcs_config[2];
-	struct phylink_pcs *pcs;
-	struct regmap *regmap;
-	int i, ret = 0;
-
-	for (i = 0; i < 2; i++) {
-		mt7531_pcs_config[i] = devm_kzalloc(priv->dev,
-						    sizeof(struct regmap_config),
-						    GFP_KERNEL);
-		if (!mt7531_pcs_config[i]) {
-			ret = -ENOMEM;
-			break;
-		}
-
-		mt7531_pcs_config[i]->name = i ? "port6" : "port5";
-		mt7531_pcs_config[i]->reg_bits = 16;
-		mt7531_pcs_config[i]->val_bits = 32;
-		mt7531_pcs_config[i]->reg_stride = 4;
-		mt7531_pcs_config[i]->reg_base = MT7531_SGMII_REG_BASE(5 + i);
-		mt7531_pcs_config[i]->max_register = 0x17c;
-		mt7531_pcs_config[i]->lock = mt7530_mdio_regmap_lock;
-		mt7531_pcs_config[i]->unlock = mt7530_mdio_regmap_unlock;
-		mt7531_pcs_config[i]->lock_arg = &priv->bus->mdio_lock;
-
-		regmap = devm_regmap_init(priv->dev,
-					  &mt7530_regmap_bus, priv->bus,
-					  mt7531_pcs_config[i]);
-		if (IS_ERR(regmap)) {
-			ret = PTR_ERR(regmap);
-			break;
-		}
-		pcs = mtk_pcs_lynxi_create(priv->dev, regmap,
-					   MT7531_PHYA_CTRL_SIGNAL3, 0);
-		if (!pcs) {
-			ret = -ENXIO;
-			break;
-		}
-		priv->ports[5 + i].sgmii_pcs = pcs;
-	}
-
-	if (ret && i)
-		mtk_pcs_lynxi_destroy(priv->ports[5].sgmii_pcs);
-
-	return ret;
-}
-
 static int
 mt753x_setup(struct dsa_switch *ds)
 {
@@ -3069,7 +2954,7 @@ static int mt753x_set_mac_eee(struct dsa_switch *ds, int port,
 	return 0;
 }
 
-static const struct dsa_switch_ops mt7530_switch_ops = {
+const struct dsa_switch_ops mt7530_switch_ops = {
 	.get_tag_protocol	= mtk_get_tag_protocol,
 	.setup			= mt753x_setup,
 	.get_strings		= mt7530_get_strings,
@@ -3103,8 +2988,9 @@ static const struct dsa_switch_ops mt7530_switch_ops = {
 	.get_mac_eee		= mt753x_get_mac_eee,
 	.set_mac_eee		= mt753x_set_mac_eee,
 };
+EXPORT_SYMBOL_GPL(mt7530_switch_ops);
 
-static const struct mt753x_info mt753x_table[] = {
+const struct mt753x_info mt753x_table[] = {
 	[ID_MT7621] = {
 		.id = ID_MT7621,
 		.pcs_ops = &mt7530_pcs_ops,
@@ -3143,16 +3029,9 @@ static const struct mt753x_info mt753x_table[] = {
 		.mac_port_config = mt7531_mac_config,
 	},
 };
+EXPORT_SYMBOL_GPL(mt753x_table);
 
-static const struct of_device_id mt7530_of_match[] = {
-	{ .compatible = "mediatek,mt7621", .data = &mt753x_table[ID_MT7621], },
-	{ .compatible = "mediatek,mt7530", .data = &mt753x_table[ID_MT7530], },
-	{ .compatible = "mediatek,mt7531", .data = &mt753x_table[ID_MT7531], },
-	{ /* sentinel */ },
-};
-MODULE_DEVICE_TABLE(of, mt7530_of_match);
-
-static int
+int
 mt7530_probe_common(struct mt7530_priv *priv)
 {
 	struct device *dev = priv->dev;
@@ -3189,88 +3068,9 @@ mt7530_probe_common(struct mt7530_priv *priv)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(mt7530_probe_common);
 
-static int
-mt7530_probe(struct mdio_device *mdiodev)
-{
-	static struct regmap_config *regmap_config;
-	struct mt7530_priv *priv;
-	struct device_node *dn;
-	int ret;
-
-	dn = mdiodev->dev.of_node;
-
-	priv = devm_kzalloc(&mdiodev->dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
-
-	priv->bus = mdiodev->bus;
-	priv->dev = &mdiodev->dev;
-
-	ret = mt7530_probe_common(priv);
-	if (ret)
-		return ret;
-
-	/* Use medatek,mcm property to distinguish hardware type that would
-	 * cause a little bit differences on power-on sequence.
-	 * Not MCM that indicates switch works as the remote standalone
-	 * integrated circuit so the GPIO pin would be used to complete
-	 * the reset, otherwise memory-mapped register accessing used
-	 * through syscon provides in the case of MCM.
-	 */
-	priv->mcm = of_property_read_bool(dn, "mediatek,mcm");
-	if (priv->mcm) {
-		dev_info(&mdiodev->dev, "MT7530 adapts as multi-chip module\n");
-
-		priv->rstc = devm_reset_control_get(&mdiodev->dev, "mcm");
-		if (IS_ERR(priv->rstc)) {
-			dev_err(&mdiodev->dev, "Couldn't get our reset line\n");
-			return PTR_ERR(priv->rstc);
-		}
-	} else {
-		priv->reset = devm_gpiod_get_optional(&mdiodev->dev, "reset",
-						      GPIOD_OUT_LOW);
-		if (IS_ERR(priv->reset)) {
-			dev_err(&mdiodev->dev, "Couldn't get our reset line\n");
-			return PTR_ERR(priv->reset);
-		}
-	}
-
-	if (priv->id == ID_MT7530) {
-		priv->core_pwr = devm_regulator_get(&mdiodev->dev, "core");
-		if (IS_ERR(priv->core_pwr))
-			return PTR_ERR(priv->core_pwr);
-
-		priv->io_pwr = devm_regulator_get(&mdiodev->dev, "io");
-		if (IS_ERR(priv->io_pwr))
-			return PTR_ERR(priv->io_pwr);
-	}
-
-	regmap_config = devm_kzalloc(&mdiodev->dev, sizeof(*regmap_config),
-				     GFP_KERNEL);
-	if (!regmap_config)
-		return -ENOMEM;
-
-	regmap_config->reg_bits = 16;
-	regmap_config->val_bits = 32;
-	regmap_config->reg_stride = 4;
-	regmap_config->max_register = MT7530_CREV;
-	regmap_config->disable_locking = true;
-	priv->regmap = devm_regmap_init(priv->dev, &mt7530_regmap_bus,
-					priv->bus, regmap_config);
-	if (IS_ERR(priv->regmap))
-		return PTR_ERR(priv->regmap);
-
-	if (priv->id == ID_MT7531) {
-		ret = mt7531_create_sgmii(priv);
-		if (ret)
-			return ret;
-	}
-
-	return dsa_register_switch(priv->ds);
-}
-
-static void
+void
 mt7530_remove_common(struct mt7530_priv *priv)
 {
 	if (priv->irq)
@@ -3280,55 +3080,7 @@ mt7530_remove_common(struct mt7530_priv *priv)
 
 	mutex_destroy(&priv->reg_mutex);
 }
-
-static void
-mt7530_remove(struct mdio_device *mdiodev)
-{
-	struct mt7530_priv *priv = dev_get_drvdata(&mdiodev->dev);
-	int ret = 0, i;
-
-	if (!priv)
-		return;
-
-	ret = regulator_disable(priv->core_pwr);
-	if (ret < 0)
-		dev_err(priv->dev,
-			"Failed to disable core power: %d\n", ret);
-
-	ret = regulator_disable(priv->io_pwr);
-	if (ret < 0)
-		dev_err(priv->dev, "Failed to disable io pwr: %d\n",
-			ret);
-
-	mt7530_remove_common(priv);
-
-	for (i = 0; i < 2; ++i)
-		mtk_pcs_lynxi_destroy(priv->ports[5 + i].sgmii_pcs);
-}
-
-static void mt7530_shutdown(struct mdio_device *mdiodev)
-{
-	struct mt7530_priv *priv = dev_get_drvdata(&mdiodev->dev);
-
-	if (!priv)
-		return;
-
-	dsa_switch_shutdown(priv->ds);
-
-	dev_set_drvdata(&mdiodev->dev, NULL);
-}
-
-static struct mdio_driver mt7530_mdio_driver = {
-	.probe  = mt7530_probe,
-	.remove = mt7530_remove,
-	.shutdown = mt7530_shutdown,
-	.mdiodrv.driver = {
-		.name = "mt7530",
-		.of_match_table = mt7530_of_match,
-	},
-};
-
-mdio_module_driver(mt7530_mdio_driver);
+EXPORT_SYMBOL_GPL(mt7530_remove_common);
 
 MODULE_AUTHOR("Sean Wang <sean.wang@mediatek.com>");
 MODULE_DESCRIPTION("Driver for Mediatek MT7530 Switch");
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index 2a611173a7d0..ce02aa592a7a 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -814,4 +814,10 @@ static inline void INIT_MT7530_DUMMY_POLL(struct mt7530_dummy_poll *p,
 	p->reg = reg;
 }
 
+int mt7530_probe_common(struct mt7530_priv *priv);
+void mt7530_remove_common(struct mt7530_priv *priv);
+
+extern const struct dsa_switch_ops mt7530_switch_ops;
+extern const struct mt753x_info mt753x_table[];
+
 #endif /* __MT7530_H */
-- 
cgit v1.2.3-70-g09d2


From 54d4147a121cec5004a673a58572da346e4458f8 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:19:28 +0100
Subject: net: dsa: mt7530: skip locking if MDIO bus isn't present

As MT7530 and MT7531 internally use 32-bit wide registers, each access
to any register of the switch requires several operations on the MDIO
bus. Hence if there is congruent access, e.g. due to PCS or PHY
polling, this can mess up and interfere with another ongoing register
access sequence.

However, the MDIO bus mutex is only relevant for MDIO-connected
switches. Prepare switches which have there registers directly mapped
into the SoCs register space via MMIO which do not require such
locking. There we can simply use regmap's default locking mechanism.

Hence guard mutex operations to only be performed in case of MDIO
connected switches.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index ac485ee83ed0..394fada7ddf1 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -144,13 +144,15 @@ err:
 static void
 mt7530_mutex_lock(struct mt7530_priv *priv)
 {
-	mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
+	if (priv->bus)
+		mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
 }
 
 static void
 mt7530_mutex_unlock(struct mt7530_priv *priv)
 {
-	mutex_unlock(&priv->bus->mdio_lock);
+	if (priv->bus)
+		mutex_unlock(&priv->bus->mdio_lock);
 }
 
 static void
-- 
cgit v1.2.3-70-g09d2


From 110c18bfed41421edd677935dd33be5e6507ba92 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:19:40 +0100
Subject: net: dsa: mt7530: introduce driver for MT7988 built-in switch

Add driver for the built-in Gigabit Ethernet switch which can be found
in the MediaTek MT7988 SoC.

The switch shares most of its design with MT7530 and MT7531, but has
it's registers mapped into the SoCs register space rather than being
connected externally or internally via MDIO.

Introduce a new platform driver to support that.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                   |   2 +
 drivers/net/dsa/Kconfig       |  12 ++++
 drivers/net/dsa/Makefile      |   1 +
 drivers/net/dsa/mt7530-mmio.c | 101 +++++++++++++++++++++++++++++++
 drivers/net/dsa/mt7530.c      | 137 ++++++++++++++++++++++++++++++++++++++++--
 drivers/net/dsa/mt7530.h      |  12 ++--
 6 files changed, 255 insertions(+), 10 deletions(-)
 create mode 100644 drivers/net/dsa/mt7530-mmio.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 4970a6c6a0a8..7812f0e251ad 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -13175,9 +13175,11 @@ MEDIATEK SWITCH DRIVER
 M:	Sean Wang <sean.wang@mediatek.com>
 M:	Landen Chao <Landen.Chao@mediatek.com>
 M:	DENG Qingfang <dqfext@gmail.com>
+M:	Daniel Golle <daniel@makrotopia.org>
 L:	netdev@vger.kernel.org
 S:	Maintained
 F:	drivers/net/dsa/mt7530-mdio.c
+F:	drivers/net/dsa/mt7530-mmio.c
 F:	drivers/net/dsa/mt7530.*
 F:	net/dsa/tag_mtk.c
 
diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index 941d020fe4b6..3ed5391bb18d 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -39,6 +39,7 @@ config NET_DSA_MT7530
 	select NET_DSA_TAG_MTK
 	select MEDIATEK_GE_PHY
 	imply NET_DSA_MT7530_MDIO
+	imply NET_DSA_MT7530_MMIO
 	help
 	  This enables support for the MediaTek MT7530 and MT7531 Ethernet
 	  switch chips. Multi-chip module MT7530 in MT7621AT, MT7621DAT,
@@ -55,6 +56,17 @@ config NET_DSA_MT7530_MDIO
 	  module MT7530 which can be found in the MT7621AT, MT7621DAT,
 	  MT7621ST and MT7623AI SoCs.
 
+config NET_DSA_MT7530_MMIO
+	tristate "MediaTek MT7530 MMIO interface driver"
+	depends on NET_DSA_MT7530
+	depends on HAS_IOMEM
+	help
+	  This enables support for the built-in Ethernet switch found
+	  in the MediaTek MT7988 SoC.
+	  The switch is a similar design as MT7531, but the switch registers
+	  are directly mapped into the SoCs register space rather than being
+	  accessible via MDIO.
+
 config NET_DSA_MV88E6060
 	tristate "Marvell 88E6060 ethernet switch chip support"
 	select NET_DSA_TAG_TRAILER
diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile
index c5da75cb0a6b..cb9a97340e58 100644
--- a/drivers/net/dsa/Makefile
+++ b/drivers/net/dsa/Makefile
@@ -8,6 +8,7 @@ endif
 obj-$(CONFIG_NET_DSA_LANTIQ_GSWIP) += lantiq_gswip.o
 obj-$(CONFIG_NET_DSA_MT7530)	+= mt7530.o
 obj-$(CONFIG_NET_DSA_MT7530_MDIO) += mt7530-mdio.o
+obj-$(CONFIG_NET_DSA_MT7530_MMIO) += mt7530-mmio.o
 obj-$(CONFIG_NET_DSA_MV88E6060) += mv88e6060.o
 obj-$(CONFIG_NET_DSA_RZN1_A5PSW) += rzn1_a5psw.o
 obj-$(CONFIG_NET_DSA_SMSC_LAN9303) += lan9303-core.o
diff --git a/drivers/net/dsa/mt7530-mmio.c b/drivers/net/dsa/mt7530-mmio.c
new file mode 100644
index 000000000000..1a3d4b692f34
--- /dev/null
+++ b/drivers/net/dsa/mt7530-mmio.c
@@ -0,0 +1,101 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/module.h>
+#include <linux/of_platform.h>
+#include <linux/regmap.h>
+#include <linux/regulator/consumer.h>
+#include <linux/reset.h>
+#include <net/dsa.h>
+
+#include "mt7530.h"
+
+static const struct of_device_id mt7988_of_match[] = {
+	{ .compatible = "mediatek,mt7988-switch", .data = &mt753x_table[ID_MT7988], },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, mt7988_of_match);
+
+static int
+mt7988_probe(struct platform_device *pdev)
+{
+	static struct regmap_config *sw_regmap_config;
+	struct mt7530_priv *priv;
+	void __iomem *base_addr;
+	int ret;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->bus = NULL;
+	priv->dev = &pdev->dev;
+
+	ret = mt7530_probe_common(priv);
+	if (ret)
+		return ret;
+
+	priv->rstc = devm_reset_control_get(&pdev->dev, NULL);
+	if (IS_ERR(priv->rstc)) {
+		dev_err(&pdev->dev, "Couldn't get our reset line\n");
+		return PTR_ERR(priv->rstc);
+	}
+
+	base_addr = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(base_addr)) {
+		dev_err(&pdev->dev, "cannot request I/O memory space\n");
+		return -ENXIO;
+	}
+
+	sw_regmap_config = devm_kzalloc(&pdev->dev, sizeof(*sw_regmap_config), GFP_KERNEL);
+	if (!sw_regmap_config)
+		return -ENOMEM;
+
+	sw_regmap_config->name = "switch";
+	sw_regmap_config->reg_bits = 16;
+	sw_regmap_config->val_bits = 32;
+	sw_regmap_config->reg_stride = 4;
+	sw_regmap_config->max_register = MT7530_CREV;
+	priv->regmap = devm_regmap_init_mmio(&pdev->dev, base_addr, sw_regmap_config);
+	if (IS_ERR(priv->regmap))
+		return PTR_ERR(priv->regmap);
+
+	return dsa_register_switch(priv->ds);
+}
+
+static int
+mt7988_remove(struct platform_device *pdev)
+{
+	struct mt7530_priv *priv = platform_get_drvdata(pdev);
+
+	if (priv)
+		mt7530_remove_common(priv);
+
+	return 0;
+}
+
+static void mt7988_shutdown(struct platform_device *pdev)
+{
+	struct mt7530_priv *priv = platform_get_drvdata(pdev);
+
+	if (!priv)
+		return;
+
+	dsa_switch_shutdown(priv->ds);
+
+	dev_set_drvdata(&pdev->dev, NULL);
+}
+
+static struct platform_driver mt7988_platform_driver = {
+	.probe  = mt7988_probe,
+	.remove = mt7988_remove,
+	.shutdown = mt7988_shutdown,
+	.driver = {
+		.name = "mt7530-mmio",
+		.of_match_table = mt7988_of_match,
+	},
+};
+module_platform_driver(mt7988_platform_driver);
+
+MODULE_AUTHOR("Daniel Golle <daniel@makrotopia.org>");
+MODULE_DESCRIPTION("Driver for Mediatek MT7530 Switch (MMIO)");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 394fada7ddf1..e4bb5037d352 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -1981,6 +1981,47 @@ static const struct irq_domain_ops mt7530_irq_domain_ops = {
 	.xlate = irq_domain_xlate_onecell,
 };
 
+static void
+mt7988_irq_mask(struct irq_data *d)
+{
+	struct mt7530_priv *priv = irq_data_get_irq_chip_data(d);
+
+	priv->irq_enable &= ~BIT(d->hwirq);
+	mt7530_mii_write(priv, MT7530_SYS_INT_EN, priv->irq_enable);
+}
+
+static void
+mt7988_irq_unmask(struct irq_data *d)
+{
+	struct mt7530_priv *priv = irq_data_get_irq_chip_data(d);
+
+	priv->irq_enable |= BIT(d->hwirq);
+	mt7530_mii_write(priv, MT7530_SYS_INT_EN, priv->irq_enable);
+}
+
+static struct irq_chip mt7988_irq_chip = {
+	.name = KBUILD_MODNAME,
+	.irq_mask = mt7988_irq_mask,
+	.irq_unmask = mt7988_irq_unmask,
+};
+
+static int
+mt7988_irq_map(struct irq_domain *domain, unsigned int irq,
+	       irq_hw_number_t hwirq)
+{
+	irq_set_chip_data(irq, domain->host_data);
+	irq_set_chip_and_handler(irq, &mt7988_irq_chip, handle_simple_irq);
+	irq_set_nested_thread(irq, true);
+	irq_set_noprobe(irq);
+
+	return 0;
+}
+
+static const struct irq_domain_ops mt7988_irq_domain_ops = {
+	.map = mt7988_irq_map,
+	.xlate = irq_domain_xlate_onecell,
+};
+
 static void
 mt7530_setup_mdio_irq(struct mt7530_priv *priv)
 {
@@ -2015,8 +2056,15 @@ mt7530_setup_irq(struct mt7530_priv *priv)
 		return priv->irq ? : -EINVAL;
 	}
 
-	priv->irq_domain = irq_domain_add_linear(np, MT7530_NUM_PHYS,
-						 &mt7530_irq_domain_ops, priv);
+	if (priv->id == ID_MT7988)
+		priv->irq_domain = irq_domain_add_linear(np, MT7530_NUM_PHYS,
+							 &mt7988_irq_domain_ops,
+							 priv);
+	else
+		priv->irq_domain = irq_domain_add_linear(np, MT7530_NUM_PHYS,
+							 &mt7530_irq_domain_ops,
+							 priv);
+
 	if (!priv->irq_domain) {
 		dev_err(dev, "failed to create IRQ domain\n");
 		return -ENOMEM;
@@ -2511,6 +2559,25 @@ static void mt7531_mac_port_get_caps(struct dsa_switch *ds, int port,
 	}
 }
 
+static void mt7988_mac_port_get_caps(struct dsa_switch *ds, int port,
+				     struct phylink_config *config)
+{
+	phy_interface_zero(config->supported_interfaces);
+
+	switch (port) {
+	case 0 ... 4: /* Internal phy */
+		__set_bit(PHY_INTERFACE_MODE_INTERNAL,
+			  config->supported_interfaces);
+		break;
+
+	case 6:
+		__set_bit(PHY_INTERFACE_MODE_INTERNAL,
+			  config->supported_interfaces);
+		config->mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
+					   MAC_10000FD;
+	}
+}
+
 static int
 mt753x_pad_setup(struct dsa_switch *ds, const struct phylink_link_state *state)
 {
@@ -2586,6 +2653,17 @@ static bool mt753x_is_mac_port(u32 port)
 	return (port == 5 || port == 6);
 }
 
+static int
+mt7988_mac_config(struct dsa_switch *ds, int port, unsigned int mode,
+		  phy_interface_t interface)
+{
+	if (dsa_is_cpu_port(ds, port) &&
+	    interface == PHY_INTERFACE_MODE_INTERNAL)
+		return 0;
+
+	return -EINVAL;
+}
+
 static int
 mt7531_mac_config(struct dsa_switch *ds, int port, unsigned int mode,
 		  phy_interface_t interface)
@@ -2656,7 +2734,8 @@ mt753x_phylink_mac_config(struct dsa_switch *ds, int port, unsigned int mode,
 
 	switch (port) {
 	case 0 ... 4: /* Internal phy */
-		if (state->interface != PHY_INTERFACE_MODE_GMII)
+		if (state->interface != PHY_INTERFACE_MODE_GMII &&
+		    state->interface != PHY_INTERFACE_MODE_INTERNAL)
 			goto unsupported;
 		break;
 	case 5: /* 2nd cpu port with phy of port 0 or 4 / external phy */
@@ -2734,7 +2813,8 @@ static void mt753x_phylink_mac_link_up(struct dsa_switch *ds, int port,
 	/* MT753x MAC works in 1G full duplex mode for all up-clocked
 	 * variants.
 	 */
-	if (interface == PHY_INTERFACE_MODE_TRGMII ||
+	if (interface == PHY_INTERFACE_MODE_INTERNAL ||
+	    interface == PHY_INTERFACE_MODE_TRGMII ||
 	    (phy_interface_mode_is_8023z(interface))) {
 		speed = SPEED_1000;
 		duplex = DUPLEX_FULL;
@@ -2814,6 +2894,21 @@ mt7531_cpu_port_config(struct dsa_switch *ds, int port)
 	return 0;
 }
 
+static int
+mt7988_cpu_port_config(struct dsa_switch *ds, int port)
+{
+	struct mt7530_priv *priv = ds->priv;
+
+	mt7530_write(priv, MT7530_PMCR_P(port),
+		     PMCR_CPU_PORT_SETTING(priv->id));
+
+	mt753x_phylink_mac_link_up(ds, port, MLO_AN_FIXED,
+				   PHY_INTERFACE_MODE_INTERNAL, NULL,
+				   SPEED_10000, DUPLEX_FULL, true, true);
+
+	return 0;
+}
+
 static void mt753x_phylink_get_caps(struct dsa_switch *ds, int port,
 				    struct phylink_config *config)
 {
@@ -2956,6 +3051,27 @@ static int mt753x_set_mac_eee(struct dsa_switch *ds, int port,
 	return 0;
 }
 
+static int mt7988_pad_setup(struct dsa_switch *ds, phy_interface_t interface)
+{
+	return 0;
+}
+
+static int mt7988_setup(struct dsa_switch *ds)
+{
+	struct mt7530_priv *priv = ds->priv;
+
+	/* Reset the switch */
+	reset_control_assert(priv->rstc);
+	usleep_range(20, 50);
+	reset_control_deassert(priv->rstc);
+	usleep_range(20, 50);
+
+	/* Reset the switch PHYs */
+	mt7530_write(priv, MT7530_SYS_CTRL, SYS_CTRL_PHY_RST);
+
+	return mt7531_setup_common(ds);
+}
+
 const struct dsa_switch_ops mt7530_switch_ops = {
 	.get_tag_protocol	= mtk_get_tag_protocol,
 	.setup			= mt753x_setup,
@@ -3030,6 +3146,19 @@ const struct mt753x_info mt753x_table[] = {
 		.mac_port_get_caps = mt7531_mac_port_get_caps,
 		.mac_port_config = mt7531_mac_config,
 	},
+	[ID_MT7988] = {
+		.id = ID_MT7988,
+		.pcs_ops = &mt7530_pcs_ops,
+		.sw_setup = mt7988_setup,
+		.phy_read_c22 = mt7531_ind_c22_phy_read,
+		.phy_write_c22 = mt7531_ind_c22_phy_write,
+		.phy_read_c45 = mt7531_ind_c45_phy_read,
+		.phy_write_c45 = mt7531_ind_c45_phy_write,
+		.pad_setup = mt7988_pad_setup,
+		.cpu_port_config = mt7988_cpu_port_config,
+		.mac_port_get_caps = mt7988_mac_port_get_caps,
+		.mac_port_config = mt7988_mac_config,
+	},
 };
 EXPORT_SYMBOL_GPL(mt753x_table);
 
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index ce02aa592a7a..01db5c9724fa 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -18,6 +18,7 @@ enum mt753x_id {
 	ID_MT7530 = 0,
 	ID_MT7621 = 1,
 	ID_MT7531 = 2,
+	ID_MT7988 = 3,
 };
 
 #define	NUM_TRGMII_CTRL			5
@@ -54,11 +55,11 @@ enum mt753x_id {
 #define  MT7531_MIRROR_PORT_SET(x)	(((x) & MIRROR_MASK) << 16)
 #define  MT7531_CPU_PMAP_MASK		GENMASK(7, 0)
 
-#define MT753X_MIRROR_REG(id)		(((id) == ID_MT7531) ? \
+#define MT753X_MIRROR_REG(id)		((((id) == ID_MT7531) || ((id) == ID_MT7988)) ?	\
 					 MT7531_CFC : MT7530_MFC)
-#define MT753X_MIRROR_EN(id)		(((id) == ID_MT7531) ? \
+#define MT753X_MIRROR_EN(id)		((((id) == ID_MT7531) || ((id) == ID_MT7988)) ?	\
 					 MT7531_MIRROR_EN : MIRROR_EN)
-#define MT753X_MIRROR_MASK(id)		(((id) == ID_MT7531) ? \
+#define MT753X_MIRROR_MASK(id)		((((id) == ID_MT7531) || ((id) == ID_MT7988)) ?	\
 					 MT7531_MIRROR_MASK : MIRROR_MASK)
 
 /* Registers for BPDU and PAE frame control*/
@@ -295,9 +296,8 @@ enum mt7530_vlan_port_acc_frm {
 					 MT7531_FORCE_DPX | \
 					 MT7531_FORCE_RX_FC | \
 					 MT7531_FORCE_TX_FC)
-#define  PMCR_FORCE_MODE_ID(id)		(((id) == ID_MT7531) ? \
-					 MT7531_FORCE_MODE : \
-					 PMCR_FORCE_MODE)
+#define  PMCR_FORCE_MODE_ID(id)		((((id) == ID_MT7531) || ((id) == ID_MT7988)) ?	\
+					 MT7531_FORCE_MODE : PMCR_FORCE_MODE)
 #define  PMCR_LINK_SETTINGS_MASK	(PMCR_TX_EN | PMCR_FORCE_SPEED_1000 | \
 					 PMCR_RX_EN | PMCR_FORCE_SPEED_100 | \
 					 PMCR_TX_FC_EN | PMCR_RX_FC_EN | \
-- 
cgit v1.2.3-70-g09d2


From 386f5fc9061b0c8429c7c0cc57f12c05f8662e23 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 3 Apr 2023 02:19:51 +0100
Subject: dt-bindings: net: dsa: mediatek,mt7530: add mediatek,mt7988-switch

Add documentation for the built-in switch which can be found in the
MediaTek MT7988 SoC.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../bindings/net/dsa/mediatek,mt7530.yaml          | 26 ++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml b/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
index 5ae9cd8f99a2..e532c6b795f4 100644
--- a/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
@@ -11,16 +11,23 @@ maintainers:
   - Landen Chao <Landen.Chao@mediatek.com>
   - DENG Qingfang <dqfext@gmail.com>
   - Sean Wang <sean.wang@mediatek.com>
+  - Daniel Golle <daniel@makrotopia.org>
 
 description: |
-  There are two versions of MT7530, standalone and in a multi-chip module.
+  There are three versions of MT7530, standalone, in a multi-chip module and
+  built-into a SoC.
 
   MT7530 is a part of the multi-chip module in MT7620AN, MT7620DA, MT7620DAN,
   MT7620NN, MT7621AT, MT7621DAT, MT7621ST and MT7623AI SoCs.
 
+  The MT7988 SoC comes with a built-in switch similar to MT7531 as well as four
+  Gigabit Ethernet PHYs. The switch registers are directly mapped into the SoC's
+  memory map rather than using MDIO. The switch got an internally connected 10G
+  CPU port and 4 user ports connected to the built-in Gigabit Ethernet PHYs.
+
   MT7530 in MT7620AN, MT7620DA, MT7620DAN and MT7620NN SoCs has got 10/100 PHYs
   and the switch registers are directly mapped into SoC's memory map rather than
-  using MDIO. The DSA driver currently doesn't support this.
+  using MDIO. The DSA driver currently doesn't support MT7620 variants.
 
   There is only the standalone version of MT7531.
 
@@ -81,6 +88,10 @@ properties:
           Multi-chip module MT7530 in MT7621AT, MT7621DAT and MT7621ST SoCs
         const: mediatek,mt7621
 
+      - description:
+          Built-in switch of the MT7988 SoC
+        const: mediatek,mt7988-switch
+
   reg:
     maxItems: 1
 
@@ -268,6 +279,17 @@ allOf:
       required:
         - mediatek,mcm
 
+  - if:
+      properties:
+        compatible:
+          const: mediatek,mt7988-switch
+    then:
+      $ref: "#/$defs/mt7530-dsa-port"
+      properties:
+        gpio-controller: false
+        mediatek,mcm: false
+        reset-names: false
+
 unevaluatedProperties: false
 
 examples:
-- 
cgit v1.2.3-70-g09d2


From 70a13e5f00e9ee5497b0c25c2787347c4431b5e4 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Tue, 14 Mar 2023 10:06:13 +0800
Subject: wifi: rtw89: coex: Add LPS protocol radio state for RTL8852B

This LPS state will not turn off RF, and it can still do some basic
traffic, only RTL8852B has the state. Coexistence need let Bluetooth
know WiFi is still alive to prevent some Bluetooth performance issue.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230314020617.28193-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 10 +++++++++-
 drivers/net/wireless/realtek/rtw89/coex.h |  1 +
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index 3dbd4ee14c70..bd06388ea211 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -5728,6 +5728,11 @@ void rtw89_btc_ntfy_radio_state(struct rtw89_dev *rtwdev, enum btc_rfctrl rf_sta
 		wl->status.map.lps = BTC_LPS_RF_OFF;
 		wl->status.map.busy = 0;
 		break;
+	case BTC_RFCTRL_LPS_WL_ON: /* LPS-Protocol (RFon) */
+		wl->status.map.rf_off = 0;
+		wl->status.map.lps = BTC_LPS_RF_ON;
+		wl->status.map.busy = 0;
+		break;
 	case BTC_RFCTRL_WL_ON:
 	default:
 		wl->status.map.rf_off = 0;
@@ -5745,6 +5750,9 @@ void rtw89_btc_ntfy_radio_state(struct rtw89_dev *rtwdev, enum btc_rfctrl rf_sta
 		rtw89_btc_fw_en_rpt(rtwdev, RPT_EN_ALL, false);
 		if (rf_state == BTC_RFCTRL_WL_OFF)
 			_write_scbd(rtwdev, BTC_WSCB_ALL, false);
+		else if (rf_state == BTC_RFCTRL_LPS_WL_ON &&
+			 wl->status.map.lps_pre != BTC_LPS_OFF)
+			_update_bt_scbd(rtwdev, true);
 	}
 
 	btc->dm.cnt_dm[BTC_DCNT_BTCNT_HANG] = 0;
@@ -5755,7 +5763,7 @@ void rtw89_btc_ntfy_radio_state(struct rtw89_dev *rtwdev, enum btc_rfctrl rf_sta
 		btc->dm.tdma_instant_excute = 0;
 
 	_run_coex(rtwdev, BTC_RSN_NTFY_RADIO_STATE);
-
+	btc->dm.tdma_instant_excute = 0;
 	wl->status.map.rf_off_pre = wl->status.map.rf_off;
 	wl->status.map.lps_pre = wl->status.map.lps;
 }
diff --git a/drivers/net/wireless/realtek/rtw89/coex.h b/drivers/net/wireless/realtek/rtw89/coex.h
index 38cc53a505c3..f16421cb30ef 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.h
+++ b/drivers/net/wireless/realtek/rtw89/coex.h
@@ -131,6 +131,7 @@ enum btc_role_state {
 enum btc_rfctrl {
 	BTC_RFCTRL_WL_OFF,
 	BTC_RFCTRL_WL_ON,
+	BTC_RFCTRL_LPS_WL_ON,
 	BTC_RFCTRL_FW_CTRL,
 	BTC_RFCTRL_MAX
 };
-- 
cgit v1.2.3-70-g09d2


From 829b3a8b212a71b55598b9bd5861a41906a611ae Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Tue, 14 Mar 2023 10:06:14 +0800
Subject: wifi: rtw89: coex: Not to enable firmware report when WiFi is power
 saving

If driver enable firmware report during WiFi power saving, the
firmware timer will lead to some power saving issue like,
fail to enter LPS, can not leave LPS or some unexpected issue.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230314020617.28193-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index bd06388ea211..d4a2ed254213 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -1741,10 +1741,14 @@ static void rtw89_btc_fw_en_rpt(struct rtw89_dev *rtwdev,
 				u32 rpt_map, bool rpt_state)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_wl_smap *wl_smap = &btc->cx.wl.status.map;
 	struct rtw89_btc_btf_fwinfo *fwinfo = &btc->fwinfo;
 	struct rtw89_btc_btf_set_report r = {0};
 	u32 val, bit_map;
 
+	if ((wl_smap->rf_off || wl_smap->lps != BTC_LPS_OFF) && rpt_state != 0)
+		return;
+
 	bit_map = rtw89_btc_fw_rpt_ver(rtwdev, rpt_map);
 
 	rtw89_debug(rtwdev, RTW89_DBG_BTC,
-- 
cgit v1.2.3-70-g09d2


From 20595db3c0681cc034e50fbcba85a1e6cf3325b5 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Tue, 14 Mar 2023 10:06:15 +0800
Subject: wifi: rtw89: coex: Update RTL8852B LNA2 hardware parameter

The LNA gain didn't set before, it will lead some WiFi RX issue.
And the setting can increase both of WiFi & BT performance while
they are both RX.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230314020617.28193-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h     |  3 +-
 drivers/net/wireless/realtek/rtw89/rtw8852a.c | 11 ++++--
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 51 ++++++++++++++++++++++++++-
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 11 ++++--
 4 files changed, 68 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index e447bfec508b..3b13f2a6b42d 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -2186,12 +2186,13 @@ struct rtw89_btc_dm {
 	u32 wl_stb_chg: 1;
 	u32 pta_owner: 1;
 	u32 tdma_instant_excute: 1;
-	u32 rsvd: 1;
 
 	u16 slot_dur[CXST_MAX];
 
 	u8 run_reason;
 	u8 run_action;
+
+	u8 wl_lna2: 1;
 };
 
 struct rtw89_btc_ctrl {
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index 9c42b6abd223..152f715d55d5 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -1947,20 +1947,25 @@ static void rtw8852a_set_wl_lna2(struct rtw89_dev *rtwdev, u8 level)
 
 static void rtw8852a_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
 {
+	struct rtw89_btc *btc = &rtwdev->btc;
+
 	switch (level) {
 	case 0: /* original */
+	default:
 		rtw8852a_bb_ctrl_btc_preagc(rtwdev, false);
-		rtw8852a_set_wl_lna2(rtwdev, 0);
+		btc->dm.wl_lna2 = 0;
 		break;
 	case 1: /* for FDD free-run */
 		rtw8852a_bb_ctrl_btc_preagc(rtwdev, true);
-		rtw8852a_set_wl_lna2(rtwdev, 0);
+		btc->dm.wl_lna2 = 0;
 		break;
 	case 2: /* for BTG Co-Rx*/
 		rtw8852a_bb_ctrl_btc_preagc(rtwdev, false);
-		rtw8852a_set_wl_lna2(rtwdev, 1);
+		btc->dm.wl_lna2 = 1;
 		break;
 	}
+
+	rtw8852a_set_wl_lna2(rtwdev, btc->dm.wl_lna2);
 }
 
 static void rtw8852a_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 499ae0389c71..7bab6d2ab776 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2284,15 +2284,64 @@ static void rtw8852b_btc_wl_s1_standby(struct rtw89_dev *rtwdev, bool state)
 
 	/* set WL standby = Rx for GNT_BT_Tx = 1->0 settle issue */
 	if (state)
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x579);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x179);
 	else
 		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x20);
 
 	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
 }
 
+static void rtw8852b_btc_set_wl_lna2(struct rtw89_dev *rtwdev, u8 level)
+{
+	switch (level) {
+	case 0: /* default */
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x1000);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x0);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x17);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x2);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x3);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x17);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
+		break;
+	case 1: /* Fix LNA2=5  */
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x1000);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x0);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x5);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x2);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x3);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x5);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
+		break;
+	}
+}
+
 static void rtw8852b_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
 {
+	struct rtw89_btc *btc = &rtwdev->btc;
+
+	switch (level) {
+	case 0: /* original */
+	default:
+		rtw8852b_bb_ctrl_btc_preagc(rtwdev, false);
+		btc->dm.wl_lna2 = 0;
+		break;
+	case 1: /* for FDD free-run */
+		rtw8852b_bb_ctrl_btc_preagc(rtwdev, true);
+		btc->dm.wl_lna2 = 0;
+		break;
+	case 2: /* for BTG Co-Rx*/
+		rtw8852b_bb_ctrl_btc_preagc(rtwdev, false);
+		btc->dm.wl_lna2 = 1;
+		break;
+	}
+
+	rtw8852b_btc_set_wl_lna2(rtwdev, btc->dm.wl_lna2);
 }
 
 static void rtw8852b_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 8af813132f71..758021287552 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2633,20 +2633,25 @@ static void rtw8852c_set_wl_lna2(struct rtw89_dev *rtwdev, u8 level)
 
 static void rtw8852c_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
 {
+	struct rtw89_btc *btc = &rtwdev->btc;
+
 	switch (level) {
 	case 0: /* original */
+	default:
 		rtw8852c_bb_ctrl_btc_preagc(rtwdev, false);
-		rtw8852c_set_wl_lna2(rtwdev, 0);
+		btc->dm.wl_lna2 = 0;
 		break;
 	case 1: /* for FDD free-run */
 		rtw8852c_bb_ctrl_btc_preagc(rtwdev, true);
-		rtw8852c_set_wl_lna2(rtwdev, 0);
+		btc->dm.wl_lna2 = 0;
 		break;
 	case 2: /* for BTG Co-Rx*/
 		rtw8852c_bb_ctrl_btc_preagc(rtwdev, false);
-		rtw8852c_set_wl_lna2(rtwdev, 1);
+		btc->dm.wl_lna2 = 1;
 		break;
 	}
+
+	rtw8852c_set_wl_lna2(rtwdev, btc->dm.wl_lna2);
 }
 
 static void rtw8852c_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
-- 
cgit v1.2.3-70-g09d2


From d7904ca8a04062e1c926498966a9fab4abfab161 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Tue, 14 Mar 2023 10:06:16 +0800
Subject: wifi: rtw89: coex: Add report control v5 variation

In order to reduce firmware code size cost, remove some counter value from
the structure. But firmware didn't update version code. To parse the
correct report, add another variation version v105 to parse it.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230314020617.28193-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 148 ++++++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/core.h |  23 +++++
 2 files changed, 171 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index d4a2ed254213..c0ca37473f03 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -1022,6 +1022,11 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 		} else if (ver->fcxbtcrpt == 5) {
 			pfinfo = &pfwinfo->rpt_ctrl.finfo.v5;
 			pcinfo->req_len = sizeof(pfwinfo->rpt_ctrl.finfo.v5);
+		} else if (ver->fcxbtcrpt == 105) {
+			pfinfo = &pfwinfo->rpt_ctrl.finfo.v105;
+			pcinfo->req_len = sizeof(pfwinfo->rpt_ctrl.finfo.v105);
+			pcinfo->req_fver = 5;
+			break;
 		} else {
 			goto err;
 		}
@@ -1263,6 +1268,33 @@ static u32 _chk_btc_report(struct rtw89_dev *rtwdev,
 			_chk_btc_err(rtwdev, BTC_DCNT_RPT_HANG,
 				     pfwinfo->event[BTF_EVNT_RPT]);
 
+			dm->error.map.bt_rfk_timeout = bt->rfk_info.map.timeout;
+		} else if (ver->fcxbtcrpt == 105) {
+			prpt->v105 = pfwinfo->rpt_ctrl.finfo.v105;
+			pfwinfo->rpt_en_map = le32_to_cpu(prpt->v105.rpt_info.en);
+			wl->ver_info.fw_coex = le32_to_cpu(prpt->v105.rpt_info.cx_ver);
+			wl->ver_info.fw = le32_to_cpu(prpt->v105.rpt_info.fw_ver);
+			dm->wl_fw_cx_offload = 0;
+
+			for (i = RTW89_PHY_0; i < RTW89_PHY_MAX; i++)
+				memcpy(&dm->gnt.band[i], &prpt->v105.gnt_val[i][0],
+				       sizeof(dm->gnt.band[i]));
+
+			btc->cx.cnt_bt[BTC_BCNT_HIPRI_TX] =
+				le16_to_cpu(prpt->v105.bt_cnt[BTC_BCNT_HI_TX_V105]);
+			btc->cx.cnt_bt[BTC_BCNT_HIPRI_RX] =
+				le16_to_cpu(prpt->v105.bt_cnt[BTC_BCNT_HI_RX_V105]);
+			btc->cx.cnt_bt[BTC_BCNT_LOPRI_TX] =
+				le16_to_cpu(prpt->v105.bt_cnt[BTC_BCNT_LO_TX_V105]);
+			btc->cx.cnt_bt[BTC_BCNT_LOPRI_RX] =
+				le16_to_cpu(prpt->v105.bt_cnt[BTC_BCNT_LO_RX_V105]);
+			btc->cx.cnt_bt[BTC_BCNT_POLUT] =
+				le16_to_cpu(prpt->v105.bt_cnt[BTC_BCNT_POLLUTED_V105]);
+
+			_chk_btc_err(rtwdev, BTC_DCNT_BTCNT_HANG, 0);
+			_chk_btc_err(rtwdev, BTC_DCNT_RPT_HANG,
+				     pfwinfo->event[BTF_EVNT_RPT]);
+
 			dm->error.map.bt_rfk_timeout = bt->rfk_info.map.timeout;
 		} else {
 			goto err;
@@ -8165,7 +8197,121 @@ static void _show_summary_v5(struct rtw89_dev *rtwdev, struct seq_file *m)
 			   le16_to_cpu(prptctrl->rpt_info.cnt_aoac_rf_on),
 			   le16_to_cpu(prptctrl->rpt_info.cnt_aoac_rf_off));
 	} else {
+		seq_printf(m,
+			   " %-15s : h2c_cnt=%d(fail:%d), c2h_cnt=%d",
+			   "[summary]", pfwinfo->cnt_h2c,
+			   pfwinfo->cnt_h2c_fail, pfwinfo->cnt_c2h);
+	}
+
+	if (!pcinfo->valid || pfwinfo->len_mismch || pfwinfo->fver_mismch ||
+	    pfwinfo->err[BTFRE_EXCEPTION]) {
+		seq_puts(m, "\n");
+		seq_printf(m,
+			   " %-15s : WL FW rpt error!![rpt_ctrl_valid:%d/len:"
+			   "0x%x/ver:0x%x/ex:%d/lps=%d/rf_off=%d]",
+			   "[ERROR]", pcinfo->valid, pfwinfo->len_mismch,
+			   pfwinfo->fver_mismch, pfwinfo->err[BTFRE_EXCEPTION],
+			   wl->status.map.lps, wl->status.map.rf_off);
+	}
+
+	for (i = 0; i < BTC_NCNT_NUM; i++)
+		cnt_sum += dm->cnt_notify[i];
+
+	seq_puts(m, "\n");
+	seq_printf(m,
+		   " %-15s : total=%d, show_coex_info=%d, power_on=%d, init_coex=%d, ",
+		   "[notify_cnt]",
+		   cnt_sum, cnt[BTC_NCNT_SHOW_COEX_INFO],
+		   cnt[BTC_NCNT_POWER_ON], cnt[BTC_NCNT_INIT_COEX]);
+
+	seq_printf(m,
+		   "power_off=%d, radio_state=%d, role_info=%d, wl_rfk=%d, wl_sta=%d",
+		   cnt[BTC_NCNT_POWER_OFF], cnt[BTC_NCNT_RADIO_STATE],
+		   cnt[BTC_NCNT_ROLE_INFO], cnt[BTC_NCNT_WL_RFK],
+		   cnt[BTC_NCNT_WL_STA]);
+
+	seq_puts(m, "\n");
+	seq_printf(m,
+		   " %-15s : scan_start=%d, scan_finish=%d, switch_band=%d, special_pkt=%d, ",
+		   "[notify_cnt]",
+		   cnt[BTC_NCNT_SCAN_START], cnt[BTC_NCNT_SCAN_FINISH],
+		   cnt[BTC_NCNT_SWITCH_BAND], cnt[BTC_NCNT_SPECIAL_PACKET]);
+
+	seq_printf(m,
+		   "timer=%d, control=%d, customerize=%d",
+		   cnt[BTC_NCNT_TIMER], cnt[BTC_NCNT_CONTROL],
+		   cnt[BTC_NCNT_CUSTOMERIZE]);
+}
+
+static void _show_summary_v105(struct rtw89_dev *rtwdev, struct seq_file *m)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	struct rtw89_btc_btf_fwinfo *pfwinfo = &btc->fwinfo;
+	struct rtw89_btc_fbtc_rpt_ctrl_v105 *prptctrl;
+	struct rtw89_btc_rpt_cmn_info *pcinfo;
+	struct rtw89_btc_cx *cx = &btc->cx;
+	struct rtw89_btc_dm *dm = &btc->dm;
+	struct rtw89_btc_wl_info *wl = &cx->wl;
+	u32 cnt_sum = 0, *cnt = btc->dm.cnt_notify;
+	u8 i;
+
+	if (!(dm->coex_info_map & BTC_COEX_INFO_SUMMARY))
+		return;
+
+	seq_puts(m, "========== [Statistics] ==========\n");
+
+	pcinfo = &pfwinfo->rpt_ctrl.cinfo;
+	if (pcinfo->valid && !wl->status.map.lps && !wl->status.map.rf_off) {
+		prptctrl = &pfwinfo->rpt_ctrl.finfo.v105;
+
+		seq_printf(m,
+			   " %-15s : h2c_cnt=%d(fail:%d, fw_recv:%d), c2h_cnt=%d(fw_send:%d, len:%d), ",
+			   "[summary]", pfwinfo->cnt_h2c, pfwinfo->cnt_h2c_fail,
+			   le16_to_cpu(prptctrl->rpt_info.cnt_h2c),
+			   pfwinfo->cnt_c2h,
+			   le16_to_cpu(prptctrl->rpt_info.cnt_c2h),
+			   le16_to_cpu(prptctrl->rpt_info.len_c2h));
+
+		seq_printf(m,
+			   "rpt_cnt=%d(fw_send:%d), rpt_map=0x%x",
+			   pfwinfo->event[BTF_EVNT_RPT],
+			   le16_to_cpu(prptctrl->rpt_info.cnt),
+			   le32_to_cpu(prptctrl->rpt_info.en));
+
+		if (dm->error.map.wl_fw_hang)
+			seq_puts(m, " (WL FW Hang!!)");
 		seq_puts(m, "\n");
+		seq_printf(m,
+			   " %-15s : send_ok:%d, send_fail:%d, recv:%d, ",
+			   "[mailbox]",
+			   le32_to_cpu(prptctrl->bt_mbx_info.cnt_send_ok),
+			   le32_to_cpu(prptctrl->bt_mbx_info.cnt_send_fail),
+			   le32_to_cpu(prptctrl->bt_mbx_info.cnt_recv));
+
+		seq_printf(m,
+			   "A2DP_empty:%d(stop:%d, tx:%d, ack:%d, nack:%d)\n",
+			   le32_to_cpu(prptctrl->bt_mbx_info.a2dp.cnt_empty),
+			   le32_to_cpu(prptctrl->bt_mbx_info.a2dp.cnt_flowctrl),
+			   le32_to_cpu(prptctrl->bt_mbx_info.a2dp.cnt_tx),
+			   le32_to_cpu(prptctrl->bt_mbx_info.a2dp.cnt_ack),
+			   le32_to_cpu(prptctrl->bt_mbx_info.a2dp.cnt_nack));
+
+		seq_printf(m,
+			   " %-15s : wl_rfk[req:%d/go:%d/reject:%d/tout:%d]",
+			   "[RFK/LPS]", cx->cnt_wl[BTC_WCNT_RFK_REQ],
+			   cx->cnt_wl[BTC_WCNT_RFK_GO],
+			   cx->cnt_wl[BTC_WCNT_RFK_REJECT],
+			   cx->cnt_wl[BTC_WCNT_RFK_TIMEOUT]);
+
+		seq_printf(m,
+			   ", bt_rfk[req:%d]",
+			   le16_to_cpu(prptctrl->bt_cnt[BTC_BCNT_RFK_REQ]));
+
+		seq_printf(m,
+			   ", AOAC[RF_on:%d/RF_off:%d]",
+			   le16_to_cpu(prptctrl->rpt_info.cnt_aoac_rf_on),
+			   le16_to_cpu(prptctrl->rpt_info.cnt_aoac_rf_off));
+	} else {
 		seq_printf(m,
 			   " %-15s : h2c_cnt=%d(fail:%d), c2h_cnt=%d",
 			   "[summary]", pfwinfo->cnt_h2c,
@@ -8256,6 +8402,8 @@ void rtw89_btc_dump_info(struct rtw89_dev *rtwdev, struct seq_file *m)
 		_show_summary_v4(rtwdev, m);
 	else if (ver->fcxbtcrpt == 5)
 		_show_summary_v5(rtwdev, m);
+	else if (ver->fcxbtcrpt == 105)
+		_show_summary_v105(rtwdev, m);
 }
 
 void rtw89_coex_recognize_ver(struct rtw89_dev *rtwdev)
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 3b13f2a6b42d..ea2eb3ac1df0 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1576,6 +1576,16 @@ enum rtw89_btc_bt_sta_counter {
 	BTC_BCNT_STA_MAX
 };
 
+enum rtw89_btc_bt_sta_counter_v105 {
+	BTC_BCNT_RFK_REQ_V105 = 0,
+	BTC_BCNT_HI_TX_V105 = 1,
+	BTC_BCNT_HI_RX_V105 = 2,
+	BTC_BCNT_LO_TX_V105 = 3,
+	BTC_BCNT_LO_RX_V105 = 4,
+	BTC_BCNT_POLLUTED_V105 = 5,
+	BTC_BCNT_STA_MAX_V105
+};
+
 struct rtw89_btc_fbtc_rpt_ctrl_v1 {
 	u16 fver; /* btc_ver::fcxbtcrpt */
 	u16 rpt_cnt; /* tmr counters */
@@ -1666,10 +1676,23 @@ struct rtw89_btc_fbtc_rpt_ctrl_v5 {
 	struct rtw89_btc_fbtc_rpt_ctrl_bt_mailbox bt_mbx_info;
 } __packed;
 
+struct rtw89_btc_fbtc_rpt_ctrl_v105 {
+	u8 fver;
+	u8 rsvd;
+	__le16 rsvd1;
+
+	u8 gnt_val[RTW89_PHY_MAX][4];
+	__le16 bt_cnt[BTC_BCNT_STA_MAX_V105];
+
+	struct rtw89_btc_fbtc_rpt_ctrl_info_v5 rpt_info;
+	struct rtw89_btc_fbtc_rpt_ctrl_bt_mailbox bt_mbx_info;
+} __packed;
+
 union rtw89_btc_fbtc_rpt_ctrl_ver_info {
 	struct rtw89_btc_fbtc_rpt_ctrl_v1 v1;
 	struct rtw89_btc_fbtc_rpt_ctrl_v4 v4;
 	struct rtw89_btc_fbtc_rpt_ctrl_v5 v5;
+	struct rtw89_btc_fbtc_rpt_ctrl_v105 v105;
 };
 
 enum rtw89_fbtc_ext_ctrl_type {
-- 
cgit v1.2.3-70-g09d2


From 7527251f77664dd19716f36f53b723845d537eec Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Tue, 14 Mar 2023 10:06:17 +0800
Subject: wifi: rtw89: coex: Update Wi-Fi Bluetooth coexistence version to
 7.0.1

Update driver with the supported firmware version of the below item.
Bluetooth firmware BT_Coex_Ver: 0x07
Wi-Fi firmware version:
  RTL8852C->v0.27.56.10
  RTL8852A->v0.24.36
  RTL8852B->v0.29.29

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230314020617.28193-6-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index c0ca37473f03..acb3fac0c96d 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -9,7 +9,7 @@
 #include "ps.h"
 #include "reg.h"
 
-#define RTW89_COEX_VERSION 0x07000013
+#define RTW89_COEX_VERSION 0x07000113
 #define FCXDEF_STEP 50 /* MUST <= FCXMAX_STEP and match with wl fw*/
 
 enum btc_fbtc_tdma_template {
@@ -148,6 +148,13 @@ static const struct rtw89_btc_ver rtw89_btc_ver_defs[] = {
 	 .fwlrole = 1,   .frptmap = 2,    .fcxctrl = 1,
 	 .info_buf = 1280, .max_role_num = 5,
 	},
+	{RTL8852B, RTW89_FW_VER_CODE(0, 29, 29, 0),
+	 .fcxbtcrpt = 105, .fcxtdma = 3,  .fcxslots = 1, .fcxcysta = 5,
+	 .fcxstep = 3,   .fcxnullsta = 2, .fcxmreg = 2,  .fcxgpiodbg = 1,
+	 .fcxbtver = 1,  .fcxbtscan = 2,  .fcxbtafh = 2, .fcxbtdevinfo = 1,
+	 .fwlrole = 1,   .frptmap = 3,    .fcxctrl = 1,
+	 .info_buf = 1800, .max_role_num = 6,
+	},
 	{RTL8852B, RTW89_FW_VER_CODE(0, 29, 14, 0),
 	 .fcxbtcrpt = 5, .fcxtdma = 3,    .fcxslots = 1, .fcxcysta = 4,
 	 .fcxstep = 3,   .fcxnullsta = 2, .fcxmreg = 1,  .fcxgpiodbg = 1,
-- 
cgit v1.2.3-70-g09d2


From e749ef968f14a92155b2c3138d7720cf39e65e00 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 16 Mar 2023 14:39:56 +0800
Subject: wifi: rtw89: add counters of register-based H2C/C2H

The register-based H2C/C2H are used to exchange information between driver
and firmware, but only apply to narrow area because its data size is
smaller than regular packet-based H2C/C2H.

This kind of H2C/C2H must be paired. To identify if any H2C/C2H is missing,
update counters to help diagnose this kind of problems.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230316063956.71687-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h     |  4 ++++
 drivers/net/wireless/realtek/rtw89/fw.c       | 11 +++++++++++
 drivers/net/wireless/realtek/rtw89/mac.c      |  2 ++
 drivers/net/wireless/realtek/rtw89/reg.h      |  5 +++++
 drivers/net/wireless/realtek/rtw89/rtw8852a.c |  2 ++
 drivers/net/wireless/realtek/rtw89/rtw8852b.c |  2 ++
 drivers/net/wireless/realtek/rtw89/rtw8852c.c |  2 ++
 7 files changed, 28 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index ea2eb3ac1df0..62770d445713 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3140,8 +3140,10 @@ struct rtw89_chip_info {
 	u32 txwd_body_size;
 	u32 h2c_ctrl_reg;
 	const u32 *h2c_regs;
+	struct rtw89_reg_def h2c_counter_reg;
 	u32 c2h_ctrl_reg;
 	const u32 *c2h_regs;
+	struct rtw89_reg_def c2h_counter_reg;
 	const struct rtw89_page_regs *page_regs;
 	bool cfo_src_fd;
 	const struct rtw89_reg_def *dcfo_comp;
@@ -3268,6 +3270,8 @@ struct rtw89_fw_info {
 	struct completion completion;
 	u8 h2c_seq;
 	u8 rec_seq;
+	u8 h2c_counter;
+	u8 c2h_counter;
 	struct rtw89_fw_suit normal;
 	struct rtw89_fw_suit wowlan;
 	bool fw_log_enable;
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 713aefd0cf8d..5fa6863d36b3 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -615,6 +615,8 @@ int rtw89_fw_download(struct rtw89_dev *rtwdev, enum rtw89_fw_type type)
 
 	fw_info->h2c_seq = 0;
 	fw_info->rec_seq = 0;
+	fw_info->h2c_counter = 0;
+	fw_info->c2h_counter = 0;
 	rtwdev->mac.rpwm_seq_num = RPWM_SEQ_NUM_MAX;
 	rtwdev->mac.cpwm_seq_num = CPWM_SEQ_NUM_MAX;
 
@@ -2724,6 +2726,7 @@ static int rtw89_fw_write_h2c_reg(struct rtw89_dev *rtwdev,
 				  struct rtw89_mac_h2c_info *info)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
+	struct rtw89_fw_info *fw_info = &rtwdev->fw;
 	const u32 *h2c_reg = chip->h2c_regs;
 	u8 i, val, len;
 	int ret;
@@ -2743,6 +2746,9 @@ static int rtw89_fw_write_h2c_reg(struct rtw89_dev *rtwdev,
 	for (i = 0; i < RTW89_H2CREG_MAX; i++)
 		rtw89_write32(rtwdev, h2c_reg[i], info->h2creg[i]);
 
+	fw_info->h2c_counter++;
+	rtw89_write8_mask(rtwdev, chip->h2c_counter_reg.addr,
+			  chip->h2c_counter_reg.mask, fw_info->h2c_counter);
 	rtw89_write8(rtwdev, chip->h2c_ctrl_reg, B_AX_H2CREG_TRIGGER);
 
 	return 0;
@@ -2752,6 +2758,7 @@ static int rtw89_fw_read_c2h_reg(struct rtw89_dev *rtwdev,
 				 struct rtw89_mac_c2h_info *info)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
+	struct rtw89_fw_info *fw_info = &rtwdev->fw;
 	const u32 *c2h_reg = chip->c2h_regs;
 	u32 ret;
 	u8 i, val;
@@ -2775,6 +2782,10 @@ static int rtw89_fw_read_c2h_reg(struct rtw89_dev *rtwdev,
 	info->content_len = (RTW89_GET_C2H_HDR_LEN(*info->c2hreg) << 2) -
 				RTW89_C2HREG_HDR_LEN;
 
+	fw_info->c2h_counter++;
+	rtw89_write8_mask(rtwdev, chip->c2h_counter_reg.addr,
+			  chip->c2h_counter_reg.mask, fw_info->c2h_counter);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 7866fe925d24..d0e138f8b880 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -3398,6 +3398,8 @@ int rtw89_mac_enable_cpu(struct rtw89_dev *rtwdev, u8 boot_reason, bool dlfw)
 	if (rtw89_read32(rtwdev, R_AX_PLATFORM_ENABLE) & B_AX_WCPU_EN)
 		return -EFAULT;
 
+	rtw89_write32(rtwdev, R_AX_UDM1, 0);
+	rtw89_write32(rtwdev, R_AX_UDM2, 0);
 	rtw89_write32(rtwdev, R_AX_HALT_H2C_CTRL, 0);
 	rtw89_write32(rtwdev, R_AX_HALT_C2H_CTRL, 0);
 	rtw89_write32(rtwdev, R_AX_HALT_H2C, 0);
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index e5c0ab43ab7a..c75aee61f383 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -207,6 +207,11 @@
 
 #define R_AX_UDM0 0x01F0
 #define R_AX_UDM1 0x01F4
+#define B_AX_UDM1_MASK GENMASK(31, 16)
+#define B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK GENMASK(15, 12)
+#define B_AX_UDM1_HALMAC_H2C_DEQ_CNT_MASK GENMASK(11, 8)
+#define B_AX_UDM1_WCPU_C2H_ENQ_CNT_MASK GENMASK(7, 4)
+#define B_AX_UDM1_WCPU_H2C_DEQ_CNT_MASK GENMASK(3, 0)
 #define R_AX_UDM2 0x01F8
 #define R_AX_UDM3 0x01FC
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index 152f715d55d5..e6f8df929a95 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -2136,9 +2136,11 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 	.h2c_desc_size		= sizeof(struct rtw89_txwd_body),
 	.txwd_body_size		= sizeof(struct rtw89_txwd_body),
 	.h2c_ctrl_reg		= R_AX_H2CREG_CTRL,
+	.h2c_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_H2C_DEQ_CNT_MASK >> 8},
 	.h2c_regs		= rtw8852a_h2c_regs,
 	.c2h_ctrl_reg		= R_AX_C2HREG_CTRL,
 	.c2h_regs		= rtw8852a_c2h_regs,
+	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.page_regs		= &rtw8852a_page_regs,
 	.cfo_src_fd		= false,
 	.dcfo_comp		= &rtw8852a_dcfo_comp,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 7bab6d2ab776..27733a3da6df 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2557,8 +2557,10 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 	.h2c_desc_size		= sizeof(struct rtw89_txwd_body),
 	.txwd_body_size		= sizeof(struct rtw89_txwd_body),
 	.h2c_ctrl_reg		= R_AX_H2CREG_CTRL,
+	.h2c_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_H2C_DEQ_CNT_MASK >> 8},
 	.h2c_regs		= rtw8852b_h2c_regs,
 	.c2h_ctrl_reg		= R_AX_C2HREG_CTRL,
+	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.c2h_regs		= rtw8852b_c2h_regs,
 	.page_regs		= &rtw8852b_page_regs,
 	.cfo_src_fd		= true,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 758021287552..a82bee5c8f0a 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2872,8 +2872,10 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 	.h2c_desc_size		= sizeof(struct rtw89_rxdesc_short),
 	.txwd_body_size		= sizeof(struct rtw89_txwd_body_v1),
 	.h2c_ctrl_reg		= R_AX_H2CREG_CTRL_V1,
+	.h2c_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_H2C_DEQ_CNT_MASK >> 8},
 	.h2c_regs		= rtw8852c_h2c_regs,
 	.c2h_ctrl_reg		= R_AX_C2HREG_CTRL_V1,
+	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.c2h_regs		= rtw8852c_c2h_regs,
 	.page_regs		= &rtw8852c_page_regs,
 	.cfo_src_fd		= false,
-- 
cgit v1.2.3-70-g09d2


From e5307c9cd7ee50accd4e5f28d6a6e09ec11849e3 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Mon, 20 Mar 2023 17:31:10 +0800
Subject: wifi: rtw89: set data lowest rate according to AP supported rate

By default the driver uses the 1M and 6M rate (0x0) for data frames in
2 GHz and 5/6 GHz bands respectively. But the rate that might not AP
supported. Therefore, We modify the data lowest rate according to the
lowest of AP supported rate.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320093112.30466-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 38 +++++++++++++++++++++++--------
 1 file changed, 28 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 489fa7a86160..208fd589ffa2 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -686,6 +686,33 @@ desc_bk:
 	desc_info->bk = true;
 }
 
+static u16 rtw89_core_get_data_rate(struct rtw89_dev *rtwdev,
+				    struct rtw89_core_tx_request *tx_req)
+{
+	struct ieee80211_vif *vif = tx_req->vif;
+	struct ieee80211_sta *sta = tx_req->sta;
+	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
+	struct rtw89_phy_rate_pattern *rate_pattern = &rtwvif->rate_pattern;
+	enum rtw89_sub_entity_idx idx = rtwvif->sub_entity_idx;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, idx);
+	u16 lowest_rate;
+
+	if (rate_pattern->enable)
+		return rate_pattern->rate;
+
+	if (vif->p2p)
+		lowest_rate = RTW89_HW_RATE_OFDM6;
+	else if (chan->band_type == RTW89_BAND_2G)
+		lowest_rate = RTW89_HW_RATE_CCK1;
+	else
+		lowest_rate = RTW89_HW_RATE_OFDM6;
+
+	if (!sta->deflink.supp_rates[chan->band_type])
+		return lowest_rate;
+
+	return __ffs(sta->deflink.supp_rates[chan->band_type]) + lowest_rate;
+}
+
 static void
 rtw89_core_tx_update_data_info(struct rtw89_dev *rtwdev,
 			       struct rtw89_core_tx_request *tx_req)
@@ -694,8 +721,6 @@ rtw89_core_tx_update_data_info(struct rtw89_dev *rtwdev,
 	struct ieee80211_sta *sta = tx_req->sta;
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
 	struct rtw89_sta *rtwsta = sta_to_rtwsta_safe(sta);
-	struct rtw89_phy_rate_pattern *rate_pattern = &rtwvif->rate_pattern;
-	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
 	struct rtw89_tx_desc_info *desc_info = &tx_req->desc_info;
 	struct sk_buff *skb = tx_req->skb;
 	u8 tid, tid_indicate;
@@ -719,14 +744,7 @@ rtw89_core_tx_update_data_info(struct rtw89_dev *rtwdev,
 	if (IEEE80211_SKB_CB(skb)->control.hw_key)
 		rtw89_core_tx_update_sec_key(rtwdev, tx_req);
 
-	if (vif->p2p)
-		desc_info->data_retry_lowest_rate = RTW89_HW_RATE_OFDM6;
-	else if (rate_pattern->enable)
-		desc_info->data_retry_lowest_rate = rate_pattern->rate;
-	else if (chan->band_type == RTW89_BAND_2G)
-		desc_info->data_retry_lowest_rate = RTW89_HW_RATE_CCK1;
-	else
-		desc_info->data_retry_lowest_rate = RTW89_HW_RATE_OFDM6;
+	desc_info->data_retry_lowest_rate = rtw89_core_get_data_rate(rtwdev, tx_req);
 }
 
 static enum btc_pkt_type
-- 
cgit v1.2.3-70-g09d2


From 41209362d700f868ae7f0f30dc7634d0e0026880 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Mon, 20 Mar 2023 17:31:11 +0800
Subject: wifi: rtw89: remove superfluous H2C of join_info

We find that when starting WoWLAN, the second join_info H2C is
unnecessary and leads WoWLAN not enter power save mode if using new
firmware, so remove it.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320093112.30466-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/wow.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c
index c78ee2ab732c..7cff9c1d8d37 100644
--- a/drivers/net/wireless/realtek/rtw89/wow.c
+++ b/drivers/net/wireless/realtek/rtw89/wow.c
@@ -420,14 +420,11 @@ static int rtw89_wow_cfg_wake(struct rtw89_dev *rtwdev, bool wow)
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)wow_vif->drv_priv;
 	struct ieee80211_sta *wow_sta;
 	struct rtw89_sta *rtwsta = NULL;
-	bool is_conn = true;
 	int ret;
 
 	wow_sta = ieee80211_find_sta(wow_vif, rtwvif->bssid);
 	if (wow_sta)
 		rtwsta = (struct rtw89_sta *)wow_sta->drv_priv;
-	else
-		is_conn = false;
 
 	if (wow) {
 		if (rtw_wow->pattern_cnt)
@@ -454,12 +451,6 @@ static int rtw89_wow_cfg_wake(struct rtw89_dev *rtwdev, bool wow)
 		}
 	}
 
-	ret = rtw89_fw_h2c_join_info(rtwdev, rtwvif, rtwsta, !is_conn);
-	if (ret) {
-		rtw89_warn(rtwdev, "failed to send h2c join info\n");
-		return ret;
-	}
-
 	ret = rtw89_fw_h2c_cam(rtwdev, rtwvif, rtwsta, NULL);
 	if (ret) {
 		rtw89_warn(rtwdev, "failed to send h2c cam\n");
-- 
cgit v1.2.3-70-g09d2


From 40711486c7ba91a35ba495564e45be0f2a1b3729 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Mon, 20 Mar 2023 17:31:12 +0800
Subject: wifi: rtw89: fix incorrect channel info during scan due to ppdu_sts
 filtering

We use ppdu_sts to obtain channel information from hardware, to ensure
that the scan results have correct channel information. However, some of
ppdu_sts that is filtered by the to_self check is also needed for the scan
results. Therefore, we modify the filter part in front of CFO count.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320093112.30466-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 208fd589ffa2..448597835761 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -1220,6 +1220,10 @@ static void rtw89_core_parse_phy_status_ie01(struct rtw89_dev *rtwdev, u8 *addr,
 	phy_ppdu->chan_idx = RTW89_GET_PHY_STS_IE01_CH_IDX(addr);
 	if (phy_ppdu->rate < RTW89_HW_RATE_OFDM6)
 		return;
+
+	if (!phy_ppdu->to_self)
+		return;
+
 	/* sign conversion for S(12,2) */
 	if (rtwdev->chip->cfo_src_fd)
 		cfo = sign_extend32(RTW89_GET_PHY_STS_IE01_FD_CFO(addr), 11);
@@ -1284,9 +1288,6 @@ static int rtw89_core_rx_parse_phy_sts(struct rtw89_dev *rtwdev,
 	if (phy_ppdu->ie < RTW89_CCK_PKT)
 		return -EINVAL;
 
-	if (!phy_ppdu->to_self)
-		return 0;
-
 	pos = (u8 *)phy_ppdu->buf + PHY_STS_HDR_LEN;
 	end = (u8 *)phy_ppdu->buf + phy_ppdu->len;
 	while (pos < end) {
-- 
cgit v1.2.3-70-g09d2


From 280c444745aaf0ff7f2e54b000bd9fff9f1bbaf2 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Wed, 22 Mar 2023 14:02:38 +0800
Subject: wifi: rtw89: config EDCCA threshold during scan to prevent TX failed

Need to configure EDCCA threshold to default value before scan, and recall
original value after scan to prevent probe request can't be sent out.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230322060238.43922-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c     |  2 ++
 drivers/net/wireless/realtek/rtw89/core.h     |  3 +++
 drivers/net/wireless/realtek/rtw89/phy.c      | 19 +++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/phy.h      |  1 +
 drivers/net/wireless/realtek/rtw89/reg.h      |  5 +++++
 drivers/net/wireless/realtek/rtw89/rtw8852a.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852b.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852c.c |  1 +
 8 files changed, 33 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 448597835761..0ec52a8b83b0 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3264,6 +3264,7 @@ void rtw89_core_scan_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
 	rtw89_btc_ntfy_scan_start(rtwdev, RTW89_PHY_0, chan->band_type);
 	rtw89_chip_rfk_scan(rtwdev, true);
 	rtw89_hci_recalc_int_mit(rtwdev);
+	rtw89_phy_config_edcca(rtwdev, true);
 
 	rtw89_fw_h2c_cam(rtwdev, rtwvif, NULL, mac_addr);
 }
@@ -3281,6 +3282,7 @@ void rtw89_core_scan_complete(struct rtw89_dev *rtwdev,
 
 	rtw89_chip_rfk_scan(rtwdev, false);
 	rtw89_btc_ntfy_scan_finish(rtwdev, RTW89_PHY_0);
+	rtw89_phy_config_edcca(rtwdev, false);
 
 	rtwdev->scanning = false;
 	rtwdev->dig.bypass_dig = true;
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 62770d445713..40fb18b613d9 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3152,6 +3152,7 @@ struct rtw89_chip_info {
 	const struct rtw89_rrsr_cfgs *rrsr_cfgs;
 	u32 bss_clr_map_reg;
 	u32 dma_ch_mask;
+	u32 edcca_lvl_reg;
 	const struct wiphy_wowlan_support *wowlan_stub;
 };
 
@@ -3363,6 +3364,8 @@ struct rtw89_hal {
 
 	bool entity_active;
 	enum rtw89_entity_mode entity_mode;
+
+	u32 edcca_bak;
 };
 
 #define RTW89_MAX_MAC_ID_NUM 128
diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index d8b035972dd4..cb0f6cc51d6b 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -4366,3 +4366,22 @@ void rtw89_decode_chan_idx(struct rtw89_dev *rtwdev, u8 chan_idx,
 	*ch = rtw89_ch_base_table[idx] + (offset << 1);
 }
 EXPORT_SYMBOL(rtw89_decode_chan_idx);
+
+#define EDCCA_DEFAULT 249
+void rtw89_phy_config_edcca(struct rtw89_dev *rtwdev, bool scan)
+{
+	u32 reg = rtwdev->chip->edcca_lvl_reg;
+	struct rtw89_hal *hal = &rtwdev->hal;
+	u32 val;
+
+	if (scan) {
+		hal->edcca_bak = rtw89_phy_read32(rtwdev, reg);
+		val = hal->edcca_bak;
+		u32p_replace_bits(&val, EDCCA_DEFAULT, B_SEG0R_EDCCA_LVL_A_MSK);
+		u32p_replace_bits(&val, EDCCA_DEFAULT, B_SEG0R_EDCCA_LVL_P_MSK);
+		u32p_replace_bits(&val, EDCCA_DEFAULT, B_SEG0R_PPDU_LVL_MSK);
+		rtw89_phy_write32(rtwdev, reg, val);
+	} else {
+		rtw89_phy_write32(rtwdev, reg, hal->edcca_bak);
+	}
+}
diff --git a/drivers/net/wireless/realtek/rtw89/phy.h b/drivers/net/wireless/realtek/rtw89/phy.h
index de0a9abf646e..7535867d0f48 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.h
+++ b/drivers/net/wireless/realtek/rtw89/phy.h
@@ -558,5 +558,6 @@ void rtw89_phy_ul_tb_ctrl_track(struct rtw89_dev *rtwdev);
 u8 rtw89_encode_chan_idx(struct rtw89_dev *rtwdev, u8 central_ch, u8 band);
 void rtw89_decode_chan_idx(struct rtw89_dev *rtwdev, u8 chan_idx,
 			   u8 *ch, enum nl80211_band *band);
+void rtw89_phy_config_edcca(struct rtw89_dev *rtwdev, bool scan);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index c75aee61f383..37a1777a25a3 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -4280,6 +4280,11 @@
 #define B_PKT_POP_EN BIT(8)
 #define R_SEG0R_PD 0x481C
 #define R_SEG0R_PD_V1 0x4860
+#define R_SEG0R_EDCCA_LVL 0x4840
+#define R_SEG0R_EDCCA_LVL_V1 0x4884
+#define B_SEG0R_PPDU_LVL_MSK GENMASK(31, 24)
+#define B_SEG0R_EDCCA_LVL_P_MSK GENMASK(15, 8)
+#define B_SEG0R_EDCCA_LVL_A_MSK GENMASK(7, 0)
 #define B_SEG0R_PD_SPATIAL_REUSE_EN_MSK_V1 BIT(30)
 #define B_SEG0R_PD_SPATIAL_REUSE_EN_MSK BIT(29)
 #define B_SEG0R_PD_LOWER_BOUND_MSK GENMASK(10, 6)
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index e6f8df929a95..46907748778b 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -2149,6 +2149,7 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 	.rrsr_cfgs		= &rtw8852a_rrsr_cfgs,
 	.bss_clr_map_reg	= R_BSS_CLR_MAP,
 	.dma_ch_mask		= 0,
+	.edcca_lvl_reg		= R_SEG0R_EDCCA_LVL,
 #ifdef CONFIG_PM
 	.wowlan_stub		= &rtw_wowlan_stub_8852a,
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 27733a3da6df..bae80984cfd5 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2572,6 +2572,7 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 	.dma_ch_mask		= BIT(RTW89_DMA_ACH4) | BIT(RTW89_DMA_ACH5) |
 				  BIT(RTW89_DMA_ACH6) | BIT(RTW89_DMA_ACH7) |
 				  BIT(RTW89_DMA_B1MG) | BIT(RTW89_DMA_B1HI),
+	.edcca_lvl_reg		= R_SEG0R_EDCCA_LVL_V1,
 };
 EXPORT_SYMBOL(rtw8852b_chip_info);
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index a82bee5c8f0a..ba728fca9188 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2885,6 +2885,7 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 	.rrsr_cfgs		= &rtw8852c_rrsr_cfgs,
 	.bss_clr_map_reg	= R_BSS_CLR_MAP,
 	.dma_ch_mask		= 0,
+	.edcca_lvl_reg		= R_SEG0R_EDCCA_LVL,
 #ifdef CONFIG_PM
 	.wowlan_stub		= &rtw_wowlan_stub_8852c,
 #endif
-- 
cgit v1.2.3-70-g09d2


From 47515664ecfbde11425dff121f298ae4499425c9 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 23 Mar 2023 16:28:39 +0800
Subject: wifi: rtw89: fix potential race condition between napi_init and
 napi_enable

A race condition can happen if netdev is registered, but NAPI isn't
initialized yet, and meanwhile user space starts the netdev that will
enable NAPI. Then, it hits BUG_ON():

 kernel BUG at net/core/dev.c:6423!
 invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
 CPU: 0 PID: 417 Comm: iwd Not tainted 6.2.7-slab-dirty #3 eb0f5a8a9d91
 Hardware name: LENOVO 21DL/LNVNB161216, BIOS JPCN20WW(V1.06) 09/20/2022
 RIP: 0010:napi_enable+0x3f/0x50
 Code: 48 89 c2 48 83 e2 f6 f6 81 89 08 00 00 02 74 0d 48 83 ...
 RSP: 0018:ffffada1414f3548 EFLAGS: 00010246
 RAX: 0000000000000000 RBX: ffffa01425802080 RCX: 0000000000000000
 RDX: 00000000000002ff RSI: ffffada14e50c614 RDI: ffffa01425808dc0
 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
 R10: 0000000000000001 R11: 0000000000000100 R12: ffffa01425808f58
 R13: 0000000000000000 R14: ffffa01423498940 R15: 0000000000000001
 FS:  00007f5577c0a740(0000) GS:ffffa0169fc00000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 00007f5577a19972 CR3: 0000000125a7a000 CR4: 0000000000750ef0
 PKRU: 55555554
 Call Trace:
  <TASK>
  rtw89_pci_ops_start+0x1c/0x70 [rtw89_pci 6cbc75429515c181cbc386478d5cfb32ffc5a0f8]
  rtw89_core_start+0xbe/0x160 [rtw89_core fe07ecb874820b6d778370d4acb6ef8a37847f22]
  rtw89_ops_start+0x26/0x40 [rtw89_core fe07ecb874820b6d778370d4acb6ef8a37847f22]
  drv_start+0x42/0x100 [mac80211 c07fa22af8c3cf3f7d7ab3884ca990784d72e2d2]
  ieee80211_do_open+0x311/0x7d0 [mac80211 c07fa22af8c3cf3f7d7ab3884ca990784d72e2d2]
  ieee80211_open+0x6a/0x90 [mac80211 c07fa22af8c3cf3f7d7ab3884ca990784d72e2d2]
  __dev_open+0xe0/0x180
  __dev_change_flags+0x1da/0x250
  dev_change_flags+0x26/0x70
  do_setlink+0x37c/0x12c0
  ? ep_poll_callback+0x246/0x290
  ? __nla_validate_parse+0x61/0xd00
  ? __wake_up_common_lock+0x8f/0xd0

To fix this, follow Jonas' suggestion to switch the order of these
functions and move register netdev to be the last step of PCI probe.
Also, correct the error handling of rtw89_core_register_hw().

Fixes: e3ec7017f6a2 ("rtw89: add Realtek 802.11ax driver")
Cc: stable@vger.kernel.org
Reported-by: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Link: https://lore.kernel.org/linux-wireless/CAOiHx=n7EwK2B9CnBR07FVA=sEzFagb8TkS4XC_qBNq8OwcYUg@mail.gmail.com/T/#t
Suggested-by: Jonas Gorski <jonas.gorski@gmail.com>
Tested-by: Larry Finger<Larry.Finger@lwfinger.net>
Reviewed-by: Larry Finger<Larry.Finger@lwfinger.net>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230323082839.20474-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 10 +++++++---
 drivers/net/wireless/realtek/rtw89/pci.c  | 19 ++++++++++---------
 2 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 0ec52a8b83b0..56a13be2e283 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3455,18 +3455,22 @@ static int rtw89_core_register_hw(struct rtw89_dev *rtwdev)
 	ret = ieee80211_register_hw(hw);
 	if (ret) {
 		rtw89_err(rtwdev, "failed to register hw\n");
-		goto err;
+		goto err_free_supported_band;
 	}
 
 	ret = rtw89_regd_init(rtwdev, rtw89_regd_notifier);
 	if (ret) {
 		rtw89_err(rtwdev, "failed to init regd\n");
-		goto err;
+		goto err_unregister_hw;
 	}
 
 	return 0;
 
-err:
+err_unregister_hw:
+	ieee80211_unregister_hw(hw);
+err_free_supported_band:
+	rtw89_core_clr_supported_band(rtwdev);
+
 	return ret;
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index ec8bb5f10482..4025aabe0604 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -3874,25 +3874,26 @@ int rtw89_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	rtw89_pci_link_cfg(rtwdev);
 	rtw89_pci_l1ss_cfg(rtwdev);
 
-	ret = rtw89_core_register(rtwdev);
-	if (ret) {
-		rtw89_err(rtwdev, "failed to register core\n");
-		goto err_clear_resource;
-	}
-
 	rtw89_core_napi_init(rtwdev);
 
 	ret = rtw89_pci_request_irq(rtwdev, pdev);
 	if (ret) {
 		rtw89_err(rtwdev, "failed to request pci irq\n");
-		goto err_unregister;
+		goto err_deinit_napi;
+	}
+
+	ret = rtw89_core_register(rtwdev);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to register core\n");
+		goto err_free_irq;
 	}
 
 	return 0;
 
-err_unregister:
+err_free_irq:
+	rtw89_pci_free_irq(rtwdev, pdev);
+err_deinit_napi:
 	rtw89_core_napi_deinit(rtwdev);
-	rtw89_core_unregister(rtwdev);
 err_clear_resource:
 	rtw89_pci_clear_resource(rtwdev, pdev);
 err_declaim_pci:
-- 
cgit v1.2.3-70-g09d2


From 5995f746310108b11c7e377b0ebac4c98e53f9dc Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:26:13 +0800
Subject: wifi: rtw89: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230323112613.7550-5-cai.huoqing@linux.dev
---
 drivers/net/wireless/realtek/rtw89/pci.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index 4025aabe0604..68f0fed6d31e 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -2694,7 +2694,6 @@ static int rtw89_pci_claim_device(struct rtw89_dev *rtwdev,
 static void rtw89_pci_declaim_device(struct rtw89_dev *rtwdev,
 				     struct pci_dev *pdev)
 {
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From e83ce86aa7d914484038b181725718136387c7e1 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Thu, 30 Mar 2023 10:18:41 +0800
Subject: wifi: b43legacy: Remove the unused function prev_slot()

The function prev_slot is defined in the dma.c file, but not called
elsewhere, so remove this unused function.

drivers/net/wireless/broadcom/b43legacy/dma.c:130:19: warning: unused function 'prev_slot'.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4642
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330021841.67724-1-jiapeng.chong@linux.alibaba.com
---
 drivers/net/wireless/broadcom/b43legacy/dma.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/net/wireless/broadcom/b43legacy/dma.c b/drivers/net/wireless/broadcom/b43legacy/dma.c
index 6869f2bf1bae..60e41de72f29 100644
--- a/drivers/net/wireless/broadcom/b43legacy/dma.c
+++ b/drivers/net/wireless/broadcom/b43legacy/dma.c
@@ -127,14 +127,6 @@ static inline int next_slot(struct b43legacy_dmaring *ring, int slot)
 	return slot + 1;
 }
 
-static inline int prev_slot(struct b43legacy_dmaring *ring, int slot)
-{
-	B43legacy_WARN_ON(!(slot >= 0 && slot <= ring->nr_slots - 1));
-	if (slot == 0)
-		return ring->nr_slots - 1;
-	return slot - 1;
-}
-
 #ifdef CONFIG_B43LEGACY_DEBUG
 static void update_max_used_slots(struct b43legacy_dmaring *ring,
 				  int current_used_slots)
-- 
cgit v1.2.3-70-g09d2


From cbef9a83c51dfcb07f77cfa6ac26f53a1ea86f49 Mon Sep 17 00:00:00 2001
From: Armin Wolf <W_Armin@gmx.de>
Date: Thu, 30 Mar 2023 23:56:37 +0200
Subject: wifi: rt2x00: Fix memory leak when handling surveys

When removing a rt2x00 device, its associated channel surveys
are not freed, causing a memory leak observable with kmemleak:

unreferenced object 0xffff9620f0881a00 (size 512):
  comm "systemd-udevd", pid 2290, jiffies 4294906974 (age 33.768s)
  hex dump (first 32 bytes):
    70 44 12 00 00 00 00 00 92 8a 00 00 00 00 00 00  pD..............
    00 00 00 00 00 00 00 00 ab 87 01 00 00 00 00 00  ................
  backtrace:
    [<ffffffffb0ed858b>] __kmalloc+0x4b/0x130
    [<ffffffffc1b0f29b>] rt2800_probe_hw+0xc2b/0x1380 [rt2800lib]
    [<ffffffffc1a9496e>] rt2800usb_probe_hw+0xe/0x60 [rt2800usb]
    [<ffffffffc1ae491a>] rt2x00lib_probe_dev+0x21a/0x7d0 [rt2x00lib]
    [<ffffffffc1b3b83e>] rt2x00usb_probe+0x1be/0x980 [rt2x00usb]
    [<ffffffffc05981e2>] usb_probe_interface+0xe2/0x310 [usbcore]
    [<ffffffffb13be2d5>] really_probe+0x1a5/0x410
    [<ffffffffb13be5c8>] __driver_probe_device+0x78/0x180
    [<ffffffffb13be6fe>] driver_probe_device+0x1e/0x90
    [<ffffffffb13be972>] __driver_attach+0xd2/0x1c0
    [<ffffffffb13bbc57>] bus_for_each_dev+0x77/0xd0
    [<ffffffffb13bd2a2>] bus_add_driver+0x112/0x210
    [<ffffffffb13bfc6c>] driver_register+0x5c/0x120
    [<ffffffffc0596ae8>] usb_register_driver+0x88/0x150 [usbcore]
    [<ffffffffb0c011c4>] do_one_initcall+0x44/0x220
    [<ffffffffb0d6134c>] do_init_module+0x4c/0x220

Fix this by freeing the channel surveys on device removal.

Tested with a RT3070 based USB wireless adapter.

Fixes: 5447626910f5 ("rt2x00: save survey for every channel visited")
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Stanislaw Gruszka <stf_xl@wp.pl>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330215637.4332-1-W_Armin@gmx.de
---
 drivers/net/wireless/ralink/rt2x00/rt2x00dev.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c b/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c
index 3a035afcf7f9..9a9cfd0ce402 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c
@@ -1091,6 +1091,7 @@ static void rt2x00lib_remove_hw(struct rt2x00_dev *rt2x00dev)
 	}
 
 	kfree(rt2x00dev->spec.channels_info);
+	kfree(rt2x00dev->chan_survey);
 }
 
 static const struct ieee80211_tpt_blink rt2x00_tpt_blink[] = {
-- 
cgit v1.2.3-70-g09d2


From f6a6a5a976288e4d0d94eb1c6c9e983e8e5cdb31 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Mon, 3 Apr 2023 13:00:27 -0700
Subject: bpf: Fix struct_meta lookup for bpf_obj_free_fields kfunc call

bpf_obj_drop_impl has a void return type. In check_kfunc_call, the "else
if" which sets insn_aux->kptr_struct_meta for bpf_obj_drop_impl is
surrounded by a larger if statement which checks btf_type_is_ptr. As a
result:

  * The bpf_obj_drop_impl-specific code will never execute
  * The btf_struct_meta input to bpf_obj_drop is always NULL
  * __bpf_obj_drop_impl will always see a NULL btf_record when called
    from BPF program, and won't call bpf_obj_free_fields
  * program-allocated kptrs which have fields that should be cleaned up
    by bpf_obj_free_fields may instead leak resources

This patch adds a btf_type_is_void branch to the larger if and moves
special handling for bpf_obj_drop_impl there, fixing the issue.

Fixes: ac9f06050a35 ("bpf: Introduce bpf_obj_drop")
Cc: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230403200027.2271029-1-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 92ae4e8ab87b..eaf9c5291cf0 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -10743,10 +10743,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 				insn_aux->obj_new_size = ret_t->size;
 				insn_aux->kptr_struct_meta =
 					btf_find_struct_meta(ret_btf, ret_btf_id);
-			} else if (meta.func_id == special_kfunc_list[KF_bpf_obj_drop_impl]) {
-				insn_aux->kptr_struct_meta =
-					btf_find_struct_meta(meta.arg_obj_drop.btf,
-							     meta.arg_obj_drop.btf_id);
 			} else if (meta.func_id == special_kfunc_list[KF_bpf_list_pop_front] ||
 				   meta.func_id == special_kfunc_list[KF_bpf_list_pop_back]) {
 				struct btf_field *field = meta.arg_list_head.field;
@@ -10875,7 +10871,15 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 
 		if (reg_may_point_to_spin_lock(&regs[BPF_REG_0]) && !regs[BPF_REG_0].id)
 			regs[BPF_REG_0].id = ++env->id_gen;
-	} /* else { add_kfunc_call() ensures it is btf_type_is_void(t) } */
+	} else if (btf_type_is_void(t)) {
+		if (meta.btf == btf_vmlinux && btf_id_set_contains(&special_kfunc_set, meta.func_id)) {
+			if (meta.func_id == special_kfunc_list[KF_bpf_obj_drop_impl]) {
+				insn_aux->kptr_struct_meta =
+					btf_find_struct_meta(meta.arg_obj_drop.btf,
+							     meta.arg_obj_drop.btf_id);
+			}
+		}
+	}
 
 	nargs = btf_type_vlen(meta.func_proto);
 	args = (const struct btf_param *)(meta.func_proto + 1);
-- 
cgit v1.2.3-70-g09d2


From e790fc15bfbf3d9e6edd4e42b6288a4dd678e2f4 Mon Sep 17 00:00:00 2001
From: Íñigo Huguet <ihuguet@redhat.com>
Date: Fri, 31 Mar 2023 13:14:01 +0200
Subject: sfc: store PTP filters in a list
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of using a fixed sized array for the PTP filters, use a list.

This is not actually necessary at this point because the filters for
multicast PTP are a fixed number, but this is a preparation for the
following patches adding support for unicast PTP.

To avoid confusion with the new struct type efx_ptp_rxfilter, change the
name of some local variables from rxfilter to spec, given they're of the
type efx_filter_spec.

Reported-by: Yalin Li <yalli@redhat.com>
Signed-off-by: Íñigo Huguet <ihuguet@redhat.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ptp.c | 81 +++++++++++++++++++++++++++---------------
 1 file changed, 52 insertions(+), 29 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index 9f07e1ba7780..8418a25ed724 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -33,6 +33,7 @@
 #include <linux/ip.h>
 #include <linux/udp.h>
 #include <linux/time.h>
+#include <linux/errno.h>
 #include <linux/ktime.h>
 #include <linux/module.h>
 #include <linux/pps_kernel.h>
@@ -118,8 +119,6 @@
 
 #define	PTP_MIN_LENGTH		63
 
-#define PTP_RXFILTERS_LEN	5
-
 #define PTP_ADDR_IPV4		0xe0000181	/* 224.0.1.129 */
 #define PTP_ADDR_IPV6		{0xff, 0x0e, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
 				0, 0x01, 0x81}	/* ff0e::181 */
@@ -213,6 +212,16 @@ struct efx_ptp_timeset {
 	u32 window;	/* Derived: end - start, allowing for wrap */
 };
 
+/**
+ * struct efx_ptp_rxfilter - Filter for PTP packets
+ * @list: Node of the list where the filter is added
+ * @handle: Handle ID for the MCDI filters table
+ */
+struct efx_ptp_rxfilter {
+	struct list_head list;
+	int handle;
+};
+
 /**
  * struct efx_ptp_data - Precision Time Protocol (PTP) state
  * @efx: The NIC context
@@ -229,8 +238,7 @@ struct efx_ptp_timeset {
  * @work: Work task
  * @reset_required: A serious error has occurred and the PTP task needs to be
  *                  reset (disable, enable).
- * @rxfilters: Receive filters when operating
- * @rxfilters_count: Num of installed rxfilters, should be == PTP_RXFILTERS_LEN
+ * @rxfilters_mcast: Receive filters for multicast PTP packets
  * @config: Current timestamp configuration
  * @enabled: PTP operation enabled
  * @mode: Mode in which PTP operating (PTP version)
@@ -299,8 +307,7 @@ struct efx_ptp_data {
 	struct workqueue_struct *workwq;
 	struct work_struct work;
 	bool reset_required;
-	u32 rxfilters[PTP_RXFILTERS_LEN];
-	size_t rxfilters_count;
+	struct list_head rxfilters_mcast;
 	struct hwtstamp_config config;
 	bool enabled;
 	unsigned int mode;
@@ -1292,11 +1299,13 @@ static inline void efx_ptp_process_rx(struct efx_nic *efx, struct sk_buff *skb)
 static void efx_ptp_remove_multicast_filters(struct efx_nic *efx)
 {
 	struct efx_ptp_data *ptp = efx->ptp_data;
+	struct efx_ptp_rxfilter *rxfilter, *tmp;
 
-	while (ptp->rxfilters_count) {
-		ptp->rxfilters_count--;
+	list_for_each_entry_safe(rxfilter, tmp, &ptp->rxfilters_mcast, list) {
 		efx_filter_remove_id_safe(efx, EFX_FILTER_PRI_REQUIRED,
-					  ptp->rxfilters[ptp->rxfilters_count]);
+					  rxfilter->handle);
+		list_del(&rxfilter->list);
+		kfree(rxfilter);
 	}
 }
 
@@ -1311,48 +1320,60 @@ static void efx_ptp_init_filter(struct efx_nic *efx,
 }
 
 static int efx_ptp_insert_filter(struct efx_nic *efx,
-				 struct efx_filter_spec *rxfilter)
+				 struct efx_filter_spec *spec)
 {
 	struct efx_ptp_data *ptp = efx->ptp_data;
+	struct efx_ptp_rxfilter *rxfilter;
+	int rc;
+
+	rxfilter = kzalloc(sizeof(*rxfilter), GFP_KERNEL);
+	if (!rxfilter)
+		return -ENOMEM;
 
-	int rc = efx_filter_insert_filter(efx, rxfilter, true);
+	rc = efx_filter_insert_filter(efx, spec, true);
 	if (rc < 0)
-		return rc;
-	ptp->rxfilters[ptp->rxfilters_count] = rc;
-	ptp->rxfilters_count++;
+		goto fail;
+
+	rxfilter->handle = rc;
+	list_add(&rxfilter->list, &ptp->rxfilters_mcast);
+
 	return 0;
+
+fail:
+	kfree(rxfilter);
+	return rc;
 }
 
 static int efx_ptp_insert_ipv4_filter(struct efx_nic *efx, u16 port)
 {
-	struct efx_filter_spec rxfilter;
+	struct efx_filter_spec spec;
 
-	efx_ptp_init_filter(efx, &rxfilter);
-	efx_filter_set_ipv4_local(&rxfilter, IPPROTO_UDP, htonl(PTP_ADDR_IPV4),
+	efx_ptp_init_filter(efx, &spec);
+	efx_filter_set_ipv4_local(&spec, IPPROTO_UDP, htonl(PTP_ADDR_IPV4),
 				  htons(port));
-	return efx_ptp_insert_filter(efx, &rxfilter);
+	return efx_ptp_insert_filter(efx, &spec);
 }
 
 static int efx_ptp_insert_ipv6_filter(struct efx_nic *efx, u16 port)
 {
 	const struct in6_addr addr = {{PTP_ADDR_IPV6}};
-	struct efx_filter_spec rxfilter;
+	struct efx_filter_spec spec;
 
-	efx_ptp_init_filter(efx, &rxfilter);
-	efx_filter_set_ipv6_local(&rxfilter, IPPROTO_UDP, &addr, htons(port));
-	return efx_ptp_insert_filter(efx, &rxfilter);
+	efx_ptp_init_filter(efx, &spec);
+	efx_filter_set_ipv6_local(&spec, IPPROTO_UDP, &addr, htons(port));
+	return efx_ptp_insert_filter(efx, &spec);
 }
 
 static int efx_ptp_insert_eth_filter(struct efx_nic *efx)
 {
 	const u8 addr[ETH_ALEN] = PTP_ADDR_ETHER;
-	struct efx_filter_spec rxfilter;
+	struct efx_filter_spec spec;
 
-	efx_ptp_init_filter(efx, &rxfilter);
-	efx_filter_set_eth_local(&rxfilter, EFX_FILTER_VID_UNSPEC, addr);
-	rxfilter.match_flags |= EFX_FILTER_MATCH_ETHER_TYPE;
-	rxfilter.ether_type = htons(ETH_P_1588);
-	return efx_ptp_insert_filter(efx, &rxfilter);
+	efx_ptp_init_filter(efx, &spec);
+	efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, addr);
+	spec.match_flags |= EFX_FILTER_MATCH_ETHER_TYPE;
+	spec.ether_type = htons(ETH_P_1588);
+	return efx_ptp_insert_filter(efx, &spec);
 }
 
 static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
@@ -1360,7 +1381,7 @@ static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
 	struct efx_ptp_data *ptp = efx->ptp_data;
 	int rc;
 
-	if (!ptp->channel || ptp->rxfilters_count)
+	if (!ptp->channel || !list_empty(&ptp->rxfilters_mcast))
 		return 0;
 
 	/* Must filter on both event and general ports to ensure
@@ -1566,6 +1587,8 @@ int efx_ptp_probe(struct efx_nic *efx, struct efx_channel *channel)
 	for (pos = 0; pos < MAX_RECEIVE_EVENTS; pos++)
 		list_add(&ptp->rx_evts[pos].link, &ptp->evt_free_list);
 
+	INIT_LIST_HEAD(&ptp->rxfilters_mcast);
+
 	/* Get the NIC PTP attributes and set up time conversions */
 	rc = efx_ptp_get_attributes(efx);
 	if (rc < 0)
-- 
cgit v1.2.3-70-g09d2


From 75687cd06620e5229a74c127e6a46048aa6307f3 Mon Sep 17 00:00:00 2001
From: Íñigo Huguet <ihuguet@redhat.com>
Date: Fri, 31 Mar 2023 13:14:02 +0200
Subject: sfc: allow insertion of filters for unicast PTP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a second list for unicast filters and generalize the
efx_ptp_insert/remove_filters functions to allow acting in any of the 2
lists.

No filters for unicast are inserted yet. That will be done in the next
patch.

The reason to use 2 different lists instead of a single one is that, in
next patches, we will want to check if unicast filters are already added
and if they're expired. We don't need that for multicast filters.

Reported-by: Yalin Li <yalli@redhat.com>
Signed-off-by: Íñigo Huguet <ihuguet@redhat.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ptp.c | 59 ++++++++++++++++++++++++++----------------
 1 file changed, 36 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index 8418a25ed724..d5631e02e626 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -239,6 +239,7 @@ struct efx_ptp_rxfilter {
  * @reset_required: A serious error has occurred and the PTP task needs to be
  *                  reset (disable, enable).
  * @rxfilters_mcast: Receive filters for multicast PTP packets
+ * @rxfilters_ucast: Receive filters for unicast PTP packets
  * @config: Current timestamp configuration
  * @enabled: PTP operation enabled
  * @mode: Mode in which PTP operating (PTP version)
@@ -308,6 +309,7 @@ struct efx_ptp_data {
 	struct work_struct work;
 	bool reset_required;
 	struct list_head rxfilters_mcast;
+	struct list_head rxfilters_ucast;
 	struct hwtstamp_config config;
 	bool enabled;
 	unsigned int mode;
@@ -1296,12 +1298,12 @@ static inline void efx_ptp_process_rx(struct efx_nic *efx, struct sk_buff *skb)
 	local_bh_enable();
 }
 
-static void efx_ptp_remove_multicast_filters(struct efx_nic *efx)
+static void efx_ptp_remove_filters(struct efx_nic *efx,
+				   struct list_head *filter_list)
 {
-	struct efx_ptp_data *ptp = efx->ptp_data;
 	struct efx_ptp_rxfilter *rxfilter, *tmp;
 
-	list_for_each_entry_safe(rxfilter, tmp, &ptp->rxfilters_mcast, list) {
+	list_for_each_entry_safe(rxfilter, tmp, filter_list, list) {
 		efx_filter_remove_id_safe(efx, EFX_FILTER_PRI_REQUIRED,
 					  rxfilter->handle);
 		list_del(&rxfilter->list);
@@ -1320,9 +1322,9 @@ static void efx_ptp_init_filter(struct efx_nic *efx,
 }
 
 static int efx_ptp_insert_filter(struct efx_nic *efx,
+				 struct list_head *filter_list,
 				 struct efx_filter_spec *spec)
 {
-	struct efx_ptp_data *ptp = efx->ptp_data;
 	struct efx_ptp_rxfilter *rxfilter;
 	int rc;
 
@@ -1335,7 +1337,7 @@ static int efx_ptp_insert_filter(struct efx_nic *efx,
 		goto fail;
 
 	rxfilter->handle = rc;
-	list_add(&rxfilter->list, &ptp->rxfilters_mcast);
+	list_add(&rxfilter->list, filter_list);
 
 	return 0;
 
@@ -1344,28 +1346,31 @@ fail:
 	return rc;
 }
 
-static int efx_ptp_insert_ipv4_filter(struct efx_nic *efx, u16 port)
+static int efx_ptp_insert_ipv4_filter(struct efx_nic *efx,
+				      struct list_head *filter_list,
+				      __be32 addr, u16 port)
 {
 	struct efx_filter_spec spec;
 
 	efx_ptp_init_filter(efx, &spec);
-	efx_filter_set_ipv4_local(&spec, IPPROTO_UDP, htonl(PTP_ADDR_IPV4),
-				  htons(port));
-	return efx_ptp_insert_filter(efx, &spec);
+	efx_filter_set_ipv4_local(&spec, IPPROTO_UDP, addr, htons(port));
+	return efx_ptp_insert_filter(efx, filter_list, &spec);
 }
 
-static int efx_ptp_insert_ipv6_filter(struct efx_nic *efx, u16 port)
+static int efx_ptp_insert_ipv6_filter(struct efx_nic *efx,
+				      struct list_head *filter_list,
+				      struct in6_addr *addr, u16 port)
 {
-	const struct in6_addr addr = {{PTP_ADDR_IPV6}};
 	struct efx_filter_spec spec;
 
 	efx_ptp_init_filter(efx, &spec);
-	efx_filter_set_ipv6_local(&spec, IPPROTO_UDP, &addr, htons(port));
-	return efx_ptp_insert_filter(efx, &spec);
+	efx_filter_set_ipv6_local(&spec, IPPROTO_UDP, addr, htons(port));
+	return efx_ptp_insert_filter(efx, filter_list, &spec);
 }
 
-static int efx_ptp_insert_eth_filter(struct efx_nic *efx)
+static int efx_ptp_insert_eth_multicast_filter(struct efx_nic *efx)
 {
+	struct efx_ptp_data *ptp = efx->ptp_data;
 	const u8 addr[ETH_ALEN] = PTP_ADDR_ETHER;
 	struct efx_filter_spec spec;
 
@@ -1373,7 +1378,7 @@ static int efx_ptp_insert_eth_filter(struct efx_nic *efx)
 	efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, addr);
 	spec.match_flags |= EFX_FILTER_MATCH_ETHER_TYPE;
 	spec.ether_type = htons(ETH_P_1588);
-	return efx_ptp_insert_filter(efx, &spec);
+	return efx_ptp_insert_filter(efx, &ptp->rxfilters_mcast, &spec);
 }
 
 static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
@@ -1387,11 +1392,13 @@ static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
 	/* Must filter on both event and general ports to ensure
 	 * that there is no packet re-ordering.
 	 */
-	rc = efx_ptp_insert_ipv4_filter(efx, PTP_EVENT_PORT);
+	rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_mcast,
+					htonl(PTP_ADDR_IPV4), PTP_EVENT_PORT);
 	if (rc < 0)
 		goto fail;
 
-	rc = efx_ptp_insert_ipv4_filter(efx, PTP_GENERAL_PORT);
+	rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_mcast,
+					htonl(PTP_ADDR_IPV4), PTP_GENERAL_PORT);
 	if (rc < 0)
 		goto fail;
 
@@ -1399,15 +1406,19 @@ static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
 	 * PTP over IPv6 and Ethernet
 	 */
 	if (efx_ptp_use_mac_tx_timestamps(efx)) {
-		rc = efx_ptp_insert_ipv6_filter(efx, PTP_EVENT_PORT);
+		struct in6_addr ipv6_addr = {{PTP_ADDR_IPV6}};
+
+		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_mcast,
+						&ipv6_addr, PTP_EVENT_PORT);
 		if (rc < 0)
 			goto fail;
 
-		rc = efx_ptp_insert_ipv6_filter(efx, PTP_GENERAL_PORT);
+		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_mcast,
+						&ipv6_addr, PTP_GENERAL_PORT);
 		if (rc < 0)
 			goto fail;
 
-		rc = efx_ptp_insert_eth_filter(efx);
+		rc = efx_ptp_insert_eth_multicast_filter(efx);
 		if (rc < 0)
 			goto fail;
 	}
@@ -1415,7 +1426,7 @@ static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
 	return 0;
 
 fail:
-	efx_ptp_remove_multicast_filters(efx);
+	efx_ptp_remove_filters(efx, &ptp->rxfilters_mcast);
 	return rc;
 }
 
@@ -1440,7 +1451,7 @@ static int efx_ptp_start(struct efx_nic *efx)
 	return 0;
 
 fail:
-	efx_ptp_remove_multicast_filters(efx);
+	efx_ptp_remove_filters(efx, &ptp->rxfilters_mcast);
 	return rc;
 }
 
@@ -1456,7 +1467,8 @@ static int efx_ptp_stop(struct efx_nic *efx)
 
 	rc = efx_ptp_disable(efx);
 
-	efx_ptp_remove_multicast_filters(efx);
+	efx_ptp_remove_filters(efx, &ptp->rxfilters_mcast);
+	efx_ptp_remove_filters(efx, &ptp->rxfilters_ucast);
 
 	/* Make sure RX packets are really delivered */
 	efx_ptp_deliver_rx_queue(&efx->ptp_data->rxq);
@@ -1588,6 +1600,7 @@ int efx_ptp_probe(struct efx_nic *efx, struct efx_channel *channel)
 		list_add(&ptp->rx_evts[pos].link, &ptp->evt_free_list);
 
 	INIT_LIST_HEAD(&ptp->rxfilters_mcast);
+	INIT_LIST_HEAD(&ptp->rxfilters_ucast);
 
 	/* Get the NIC PTP attributes and set up time conversions */
 	rc = efx_ptp_get_attributes(efx);
-- 
cgit v1.2.3-70-g09d2


From 49ed35a0b6aafacd5e4d7c24ed6c02fbfbadaedc Mon Sep 17 00:00:00 2001
From: Íñigo Huguet <ihuguet@redhat.com>
Date: Fri, 31 Mar 2023 13:14:03 +0200
Subject: sfc: support unicast PTP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When sending a PTP event packet, add the correct filters that will make
that future incoming unicast PTP event packets will be timestamped.
The unicast address for the filter is gotten from the outgoing skb
before sending it.

Until now they were not timestamped because only filters that match with
the PTP multicast addressed were being configured into the NIC for the
PTP special channel. Packets received through different channels are not
timestamped, getting "received SYNC without timestamp" error in ptp4l.

Note that the inserted filters are never removed unless the NIC is stopped
or reconfigured, so efx_ptp_stop is called. Removal of old filters will
be handled by the next patch.

Additionally, cleanup a bit efx_ptp_xmit_skb_mc to use the reverse xmas
tree convention and remove an unnecessary assignment to rc variable in
void function.

Reported-by: Yalin Li <yalli@redhat.com>
Signed-off-by: Íñigo Huguet <ihuguet@redhat.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ptp.c | 101 +++++++++++++++++++++++++++++++++++++++--
 1 file changed, 98 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index d5631e02e626..b495f3d0616b 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -215,10 +215,16 @@ struct efx_ptp_timeset {
 /**
  * struct efx_ptp_rxfilter - Filter for PTP packets
  * @list: Node of the list where the filter is added
+ * @ether_type: Network protocol of the filter (ETHER_P_IP / ETHER_P_IPV6)
+ * @loc_port: UDP port of the filter (PTP_EVENT_PORT / PTP_GENERAL_PORT)
+ * @loc_host: IPv4/v6 address of the filter
  * @handle: Handle ID for the MCDI filters table
  */
 struct efx_ptp_rxfilter {
 	struct list_head list;
+	__be16 ether_type;
+	__be16 loc_port;
+	__be32 loc_host[4];
 	int handle;
 };
 
@@ -367,6 +373,8 @@ static int efx_phc_settime(struct ptp_clock_info *ptp,
 			   const struct timespec64 *e_ts);
 static int efx_phc_enable(struct ptp_clock_info *ptp,
 			  struct ptp_clock_request *request, int on);
+static int efx_ptp_insert_unicast_filter(struct efx_nic *efx,
+					 struct sk_buff *skb);
 
 bool efx_ptp_use_mac_tx_timestamps(struct efx_nic *efx)
 {
@@ -1112,6 +1120,8 @@ static void efx_ptp_xmit_skb_queue(struct efx_nic *efx, struct sk_buff *skb)
 
 	tx_queue = efx_channel_get_tx_queue(ptp_data->channel, type);
 	if (tx_queue && tx_queue->timestamping) {
+		skb_get(skb);
+
 		/* This code invokes normal driver TX code which is always
 		 * protected from softirqs when called from generic TX code,
 		 * which in turn disables preemption. Look at __dev_queue_xmit
@@ -1135,6 +1145,13 @@ static void efx_ptp_xmit_skb_queue(struct efx_nic *efx, struct sk_buff *skb)
 		local_bh_disable();
 		efx_enqueue_skb(tx_queue, skb);
 		local_bh_enable();
+
+		/* We need to add the filters after enqueuing the packet.
+		 * Otherwise, there's high latency in sending back the
+		 * timestamp, causing ptp4l timeouts
+		 */
+		efx_ptp_insert_unicast_filter(efx, skb);
+		dev_consume_skb_any(skb);
 	} else {
 		WARN_ONCE(1, "PTP channel has no timestamped tx queue\n");
 		dev_kfree_skb_any(skb);
@@ -1144,11 +1161,11 @@ static void efx_ptp_xmit_skb_queue(struct efx_nic *efx, struct sk_buff *skb)
 /* Transmit a PTP packet, via the MCDI interface, to the wire. */
 static void efx_ptp_xmit_skb_mc(struct efx_nic *efx, struct sk_buff *skb)
 {
+	MCDI_DECLARE_BUF(txtime, MC_CMD_PTP_OUT_TRANSMIT_LEN);
 	struct efx_ptp_data *ptp_data = efx->ptp_data;
 	struct skb_shared_hwtstamps timestamps;
-	int rc = -EIO;
-	MCDI_DECLARE_BUF(txtime, MC_CMD_PTP_OUT_TRANSMIT_LEN);
 	size_t len;
+	int rc;
 
 	MCDI_SET_DWORD(ptp_data->txbuf, PTP_IN_OP, MC_CMD_PTP_OP_TRANSMIT);
 	MCDI_SET_DWORD(ptp_data->txbuf, PTP_IN_PERIPH_ID, 0);
@@ -1182,7 +1199,10 @@ static void efx_ptp_xmit_skb_mc(struct efx_nic *efx, struct sk_buff *skb)
 
 	skb_tstamp_tx(skb, &timestamps);
 
-	rc = 0;
+	/* Add the filters after sending back the timestamp to avoid delaying it
+	 * or ptp4l may timeout.
+	 */
+	efx_ptp_insert_unicast_filter(efx, skb);
 
 fail:
 	dev_kfree_skb_any(skb);
@@ -1298,6 +1318,21 @@ static inline void efx_ptp_process_rx(struct efx_nic *efx, struct sk_buff *skb)
 	local_bh_enable();
 }
 
+static bool efx_ptp_filter_exists(struct list_head *filter_list,
+				  struct efx_filter_spec *spec)
+{
+	struct efx_ptp_rxfilter *rxfilter;
+
+	list_for_each_entry(rxfilter, filter_list, list) {
+		if (rxfilter->ether_type == spec->ether_type &&
+		    rxfilter->loc_port == spec->loc_port &&
+		    !memcmp(rxfilter->loc_host, spec->loc_host, sizeof(spec->loc_host)))
+			return true;
+	}
+
+	return false;
+}
+
 static void efx_ptp_remove_filters(struct efx_nic *efx,
 				   struct list_head *filter_list)
 {
@@ -1328,6 +1363,9 @@ static int efx_ptp_insert_filter(struct efx_nic *efx,
 	struct efx_ptp_rxfilter *rxfilter;
 	int rc;
 
+	if (efx_ptp_filter_exists(filter_list, spec))
+		return 0;
+
 	rxfilter = kzalloc(sizeof(*rxfilter), GFP_KERNEL);
 	if (!rxfilter)
 		return -ENOMEM;
@@ -1337,6 +1375,9 @@ static int efx_ptp_insert_filter(struct efx_nic *efx,
 		goto fail;
 
 	rxfilter->handle = rc;
+	rxfilter->ether_type = spec->ether_type;
+	rxfilter->loc_port = spec->loc_port;
+	memcpy(rxfilter->loc_host, spec->loc_host, sizeof(spec->loc_host));
 	list_add(&rxfilter->list, filter_list);
 
 	return 0;
@@ -1430,6 +1471,60 @@ fail:
 	return rc;
 }
 
+static bool efx_ptp_valid_unicast_event_pkt(struct sk_buff *skb)
+{
+	if (skb->protocol == htons(ETH_P_IP)) {
+		return ip_hdr(skb)->daddr != htonl(PTP_ADDR_IPV4) &&
+			ip_hdr(skb)->protocol == IPPROTO_UDP &&
+			udp_hdr(skb)->source == htons(PTP_EVENT_PORT);
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		struct in6_addr mcast_addr = {{PTP_ADDR_IPV6}};
+
+		return !ipv6_addr_equal(&ipv6_hdr(skb)->daddr, &mcast_addr) &&
+			ipv6_hdr(skb)->nexthdr == IPPROTO_UDP &&
+			udp_hdr(skb)->source == htons(PTP_EVENT_PORT);
+	}
+	return false;
+}
+
+static int efx_ptp_insert_unicast_filter(struct efx_nic *efx,
+					 struct sk_buff *skb)
+{
+	struct efx_ptp_data *ptp = efx->ptp_data;
+	int rc;
+
+	if (!efx_ptp_valid_unicast_event_pkt(skb))
+		return -EINVAL;
+
+	if (skb->protocol == htons(ETH_P_IP)) {
+		__be32 addr = ip_hdr(skb)->saddr;
+
+		rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_ucast,
+						addr, PTP_EVENT_PORT);
+		if (rc < 0)
+			goto out;
+
+		rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_ucast,
+						addr, PTP_GENERAL_PORT);
+	} else if (efx_ptp_use_mac_tx_timestamps(efx)) {
+		/* IPv6 PTP only supported by devices with MAC hw timestamp */
+		struct in6_addr *addr = &ipv6_hdr(skb)->saddr;
+
+		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_ucast,
+						addr, PTP_EVENT_PORT);
+		if (rc < 0)
+			goto out;
+
+		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_ucast,
+						addr, PTP_GENERAL_PORT);
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+out:
+	return rc;
+}
+
 static int efx_ptp_start(struct efx_nic *efx)
 {
 	struct efx_ptp_data *ptp = efx->ptp_data;
-- 
cgit v1.2.3-70-g09d2


From ad47655eadc8918a60c4165237a4c7da383c1d08 Mon Sep 17 00:00:00 2001
From: Íñigo Huguet <ihuguet@redhat.com>
Date: Fri, 31 Mar 2023 13:14:04 +0200
Subject: sfc: remove expired unicast PTP filters
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Filters inserted to support unicast PTP mode might become unused after
some time, so we need to remove them to avoid accumulating many of them.

Refresh the expiration time of a filter each time it's used. Then check
periodically if any filter hasn't been used for a long time (30s) and
remove it.

Reported-by: Yalin Li <yalli@redhat.com>
Signed-off-by: Íñigo Huguet <ihuguet@redhat.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ptp.c | 97 +++++++++++++++++++++++++++++++-----------
 1 file changed, 72 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index b495f3d0616b..0c40571133cb 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -75,6 +75,9 @@
 /* How long an unmatched event or packet can be held */
 #define PKT_EVENT_LIFETIME_MS		10
 
+/* How long unused unicast filters can be held */
+#define UCAST_FILTER_EXPIRY_JIFFIES	msecs_to_jiffies(30000)
+
 /* Offsets into PTP packet for identification.  These offsets are from the
  * start of the IP header, not the MAC header.  Note that neither PTP V1 nor
  * PTP V2 permit the use of IPV4 options.
@@ -218,6 +221,7 @@ struct efx_ptp_timeset {
  * @ether_type: Network protocol of the filter (ETHER_P_IP / ETHER_P_IPV6)
  * @loc_port: UDP port of the filter (PTP_EVENT_PORT / PTP_GENERAL_PORT)
  * @loc_host: IPv4/v6 address of the filter
+ * @expiry: time when the filter expires, in jiffies
  * @handle: Handle ID for the MCDI filters table
  */
 struct efx_ptp_rxfilter {
@@ -225,6 +229,7 @@ struct efx_ptp_rxfilter {
 	__be16 ether_type;
 	__be16 loc_port;
 	__be32 loc_host[4];
+	unsigned long expiry;
 	int handle;
 };
 
@@ -242,6 +247,7 @@ struct efx_ptp_rxfilter {
  * @rx_evts: Instantiated events (on evt_list and evt_free_list)
  * @workwq: Work queue for processing pending PTP operations
  * @work: Work task
+ * @cleanup_work: Work task for periodic cleanup
  * @reset_required: A serious error has occurred and the PTP task needs to be
  *                  reset (disable, enable).
  * @rxfilters_mcast: Receive filters for multicast PTP packets
@@ -313,6 +319,7 @@ struct efx_ptp_data {
 	struct efx_ptp_event_rx rx_evts[MAX_RECEIVE_EVENTS];
 	struct workqueue_struct *workwq;
 	struct work_struct work;
+	struct delayed_work cleanup_work;
 	bool reset_required;
 	struct list_head rxfilters_mcast;
 	struct list_head rxfilters_ucast;
@@ -1318,8 +1325,8 @@ static inline void efx_ptp_process_rx(struct efx_nic *efx, struct sk_buff *skb)
 	local_bh_enable();
 }
 
-static bool efx_ptp_filter_exists(struct list_head *filter_list,
-				  struct efx_filter_spec *spec)
+static struct efx_ptp_rxfilter *
+efx_ptp_find_filter(struct list_head *filter_list, struct efx_filter_spec *spec)
 {
 	struct efx_ptp_rxfilter *rxfilter;
 
@@ -1327,10 +1334,19 @@ static bool efx_ptp_filter_exists(struct list_head *filter_list,
 		if (rxfilter->ether_type == spec->ether_type &&
 		    rxfilter->loc_port == spec->loc_port &&
 		    !memcmp(rxfilter->loc_host, spec->loc_host, sizeof(spec->loc_host)))
-			return true;
+			return rxfilter;
 	}
 
-	return false;
+	return NULL;
+}
+
+static void efx_ptp_remove_one_filter(struct efx_nic *efx,
+				      struct efx_ptp_rxfilter *rxfilter)
+{
+	efx_filter_remove_id_safe(efx, EFX_FILTER_PRI_REQUIRED,
+				  rxfilter->handle);
+	list_del(&rxfilter->list);
+	kfree(rxfilter);
 }
 
 static void efx_ptp_remove_filters(struct efx_nic *efx,
@@ -1338,12 +1354,8 @@ static void efx_ptp_remove_filters(struct efx_nic *efx,
 {
 	struct efx_ptp_rxfilter *rxfilter, *tmp;
 
-	list_for_each_entry_safe(rxfilter, tmp, filter_list, list) {
-		efx_filter_remove_id_safe(efx, EFX_FILTER_PRI_REQUIRED,
-					  rxfilter->handle);
-		list_del(&rxfilter->list);
-		kfree(rxfilter);
-	}
+	list_for_each_entry_safe(rxfilter, tmp, filter_list, list)
+		efx_ptp_remove_one_filter(efx, rxfilter);
 }
 
 static void efx_ptp_init_filter(struct efx_nic *efx,
@@ -1358,13 +1370,18 @@ static void efx_ptp_init_filter(struct efx_nic *efx,
 
 static int efx_ptp_insert_filter(struct efx_nic *efx,
 				 struct list_head *filter_list,
-				 struct efx_filter_spec *spec)
+				 struct efx_filter_spec *spec,
+				 unsigned long expiry)
 {
+	struct efx_ptp_data *ptp = efx->ptp_data;
 	struct efx_ptp_rxfilter *rxfilter;
 	int rc;
 
-	if (efx_ptp_filter_exists(filter_list, spec))
+	rxfilter = efx_ptp_find_filter(filter_list, spec);
+	if (rxfilter) {
+		rxfilter->expiry = expiry;
 		return 0;
+	}
 
 	rxfilter = kzalloc(sizeof(*rxfilter), GFP_KERNEL);
 	if (!rxfilter)
@@ -1378,8 +1395,12 @@ static int efx_ptp_insert_filter(struct efx_nic *efx,
 	rxfilter->ether_type = spec->ether_type;
 	rxfilter->loc_port = spec->loc_port;
 	memcpy(rxfilter->loc_host, spec->loc_host, sizeof(spec->loc_host));
+	rxfilter->expiry = expiry;
 	list_add(&rxfilter->list, filter_list);
 
+	queue_delayed_work(ptp->workwq, &ptp->cleanup_work,
+			   UCAST_FILTER_EXPIRY_JIFFIES + 1);
+
 	return 0;
 
 fail:
@@ -1389,24 +1410,26 @@ fail:
 
 static int efx_ptp_insert_ipv4_filter(struct efx_nic *efx,
 				      struct list_head *filter_list,
-				      __be32 addr, u16 port)
+				      __be32 addr, u16 port,
+				      unsigned long expiry)
 {
 	struct efx_filter_spec spec;
 
 	efx_ptp_init_filter(efx, &spec);
 	efx_filter_set_ipv4_local(&spec, IPPROTO_UDP, addr, htons(port));
-	return efx_ptp_insert_filter(efx, filter_list, &spec);
+	return efx_ptp_insert_filter(efx, filter_list, &spec, expiry);
 }
 
 static int efx_ptp_insert_ipv6_filter(struct efx_nic *efx,
 				      struct list_head *filter_list,
-				      struct in6_addr *addr, u16 port)
+				      struct in6_addr *addr, u16 port,
+				      unsigned long expiry)
 {
 	struct efx_filter_spec spec;
 
 	efx_ptp_init_filter(efx, &spec);
 	efx_filter_set_ipv6_local(&spec, IPPROTO_UDP, addr, htons(port));
-	return efx_ptp_insert_filter(efx, filter_list, &spec);
+	return efx_ptp_insert_filter(efx, filter_list, &spec, expiry);
 }
 
 static int efx_ptp_insert_eth_multicast_filter(struct efx_nic *efx)
@@ -1419,7 +1442,7 @@ static int efx_ptp_insert_eth_multicast_filter(struct efx_nic *efx)
 	efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, addr);
 	spec.match_flags |= EFX_FILTER_MATCH_ETHER_TYPE;
 	spec.ether_type = htons(ETH_P_1588);
-	return efx_ptp_insert_filter(efx, &ptp->rxfilters_mcast, &spec);
+	return efx_ptp_insert_filter(efx, &ptp->rxfilters_mcast, &spec, 0);
 }
 
 static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
@@ -1434,12 +1457,14 @@ static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
 	 * that there is no packet re-ordering.
 	 */
 	rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_mcast,
-					htonl(PTP_ADDR_IPV4), PTP_EVENT_PORT);
+					htonl(PTP_ADDR_IPV4), PTP_EVENT_PORT,
+					0);
 	if (rc < 0)
 		goto fail;
 
 	rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_mcast,
-					htonl(PTP_ADDR_IPV4), PTP_GENERAL_PORT);
+					htonl(PTP_ADDR_IPV4), PTP_GENERAL_PORT,
+					0);
 	if (rc < 0)
 		goto fail;
 
@@ -1450,12 +1475,12 @@ static int efx_ptp_insert_multicast_filters(struct efx_nic *efx)
 		struct in6_addr ipv6_addr = {{PTP_ADDR_IPV6}};
 
 		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_mcast,
-						&ipv6_addr, PTP_EVENT_PORT);
+						&ipv6_addr, PTP_EVENT_PORT, 0);
 		if (rc < 0)
 			goto fail;
 
 		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_mcast,
-						&ipv6_addr, PTP_GENERAL_PORT);
+						&ipv6_addr, PTP_GENERAL_PORT, 0);
 		if (rc < 0)
 			goto fail;
 
@@ -1491,32 +1516,35 @@ static int efx_ptp_insert_unicast_filter(struct efx_nic *efx,
 					 struct sk_buff *skb)
 {
 	struct efx_ptp_data *ptp = efx->ptp_data;
+	unsigned long expiry;
 	int rc;
 
 	if (!efx_ptp_valid_unicast_event_pkt(skb))
 		return -EINVAL;
 
+	expiry = jiffies + UCAST_FILTER_EXPIRY_JIFFIES;
+
 	if (skb->protocol == htons(ETH_P_IP)) {
 		__be32 addr = ip_hdr(skb)->saddr;
 
 		rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_ucast,
-						addr, PTP_EVENT_PORT);
+						addr, PTP_EVENT_PORT, expiry);
 		if (rc < 0)
 			goto out;
 
 		rc = efx_ptp_insert_ipv4_filter(efx, &ptp->rxfilters_ucast,
-						addr, PTP_GENERAL_PORT);
+						addr, PTP_GENERAL_PORT, expiry);
 	} else if (efx_ptp_use_mac_tx_timestamps(efx)) {
 		/* IPv6 PTP only supported by devices with MAC hw timestamp */
 		struct in6_addr *addr = &ipv6_hdr(skb)->saddr;
 
 		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_ucast,
-						addr, PTP_EVENT_PORT);
+						addr, PTP_EVENT_PORT, expiry);
 		if (rc < 0)
 			goto out;
 
 		rc = efx_ptp_insert_ipv6_filter(efx, &ptp->rxfilters_ucast,
-						addr, PTP_GENERAL_PORT);
+						addr, PTP_GENERAL_PORT, expiry);
 	} else {
 		return -EOPNOTSUPP;
 	}
@@ -1627,6 +1655,23 @@ static void efx_ptp_worker(struct work_struct *work)
 		efx_ptp_process_rx(efx, skb);
 }
 
+static void efx_ptp_cleanup_worker(struct work_struct *work)
+{
+	struct efx_ptp_data *ptp =
+		container_of(work, struct efx_ptp_data, cleanup_work.work);
+	struct efx_ptp_rxfilter *rxfilter, *tmp;
+
+	list_for_each_entry_safe(rxfilter, tmp, &ptp->rxfilters_ucast, list) {
+		if (time_is_before_jiffies(rxfilter->expiry))
+			efx_ptp_remove_one_filter(ptp->efx, rxfilter);
+	}
+
+	if (!list_empty(&ptp->rxfilters_ucast)) {
+		queue_delayed_work(ptp->workwq, &ptp->cleanup_work,
+				   UCAST_FILTER_EXPIRY_JIFFIES + 1);
+	}
+}
+
 static const struct ptp_clock_info efx_phc_clock_info = {
 	.owner		= THIS_MODULE,
 	.name		= "sfc",
@@ -1685,6 +1730,7 @@ int efx_ptp_probe(struct efx_nic *efx, struct efx_channel *channel)
 	}
 
 	INIT_WORK(&ptp->work, efx_ptp_worker);
+	INIT_DELAYED_WORK(&ptp->cleanup_work, efx_ptp_cleanup_worker);
 	ptp->config.flags = 0;
 	ptp->config.tx_type = HWTSTAMP_TX_OFF;
 	ptp->config.rx_filter = HWTSTAMP_FILTER_NONE;
@@ -1776,6 +1822,7 @@ void efx_ptp_remove(struct efx_nic *efx)
 	(void)efx_ptp_disable(efx);
 
 	cancel_work_sync(&efx->ptp_data->work);
+	cancel_delayed_work_sync(&efx->ptp_data->cleanup_work);
 	if (efx->ptp_data->pps_workwq)
 		cancel_work_sync(&efx->ptp_data->pps_work);
 
-- 
cgit v1.2.3-70-g09d2


From f59f3006ca7b2fc9bf38fbf44e3dc4b1fbd686c1 Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Mon, 3 Apr 2023 14:24:17 +0300
Subject: vsock/vmci: convert VMCI error code to -ENOMEM on receive

This adds conversion of VMCI specific error code to general -ENOMEM. It
is preparation for the next patch, which changes af_vsock.c behaviour
on receive to pass value returned from transport to the user.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Vishnu Dasa <vdasa@vmware.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/vmw_vsock/vmci_transport.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 36eb16a40745..a5375c97f5b0 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -1831,10 +1831,17 @@ static ssize_t vmci_transport_stream_dequeue(
 	size_t len,
 	int flags)
 {
+	ssize_t err;
+
 	if (flags & MSG_PEEK)
-		return vmci_qpair_peekv(vmci_trans(vsk)->qpair, msg, len, 0);
+		err = vmci_qpair_peekv(vmci_trans(vsk)->qpair, msg, len, 0);
 	else
-		return vmci_qpair_dequev(vmci_trans(vsk)->qpair, msg, len, 0);
+		err = vmci_qpair_dequev(vmci_trans(vsk)->qpair, msg, len, 0);
+
+	if (err < 0)
+		err = -ENOMEM;
+
+	return err;
 }
 
 static ssize_t vmci_transport_stream_enqueue(
-- 
cgit v1.2.3-70-g09d2


From 02ab696febab622c4c837d509f6825884c952c22 Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Mon, 3 Apr 2023 14:25:12 +0300
Subject: vsock: return errors other than -ENOMEM to socket

This removes behaviour, where error code returned from any transport
was always switched to ENOMEM. This works in the same way as:
commit
c43170b7e157 ("vsock: return errors other than -ENOMEM to socket"),
but for receive calls.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/vmw_vsock/af_vsock.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 5f2dda35c980..413407bb646c 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -2043,7 +2043,7 @@ static int __vsock_stream_recvmsg(struct sock *sk, struct msghdr *msg,
 
 		read = transport->stream_dequeue(vsk, msg, len - copied, flags);
 		if (read < 0) {
-			err = -ENOMEM;
+			err = read;
 			break;
 		}
 
@@ -2094,7 +2094,7 @@ static int __vsock_seqpacket_recvmsg(struct sock *sk, struct msghdr *msg,
 	msg_len = transport->seqpacket_dequeue(vsk, msg, flags);
 
 	if (msg_len < 0) {
-		err = -ENOMEM;
+		err = msg_len;
 		goto out;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From b5d54eb5899a7c2d478909041446499f379c716c Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Mon, 3 Apr 2023 14:26:18 +0300
Subject: vsock/test: update expected return values

This updates expected return values for invalid buffer test. Now such
values are returned from transport, not from af_vsock.c.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 tools/testing/vsock/vsock_test.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
index 12b97c92fbb2..ac1bd3ac1533 100644
--- a/tools/testing/vsock/vsock_test.c
+++ b/tools/testing/vsock/vsock_test.c
@@ -723,7 +723,7 @@ static void test_seqpacket_invalid_rec_buffer_server(const struct test_opts *opt
 		exit(EXIT_FAILURE);
 	}
 
-	if (errno != ENOMEM) {
+	if (errno != EFAULT) {
 		perror("unexpected errno of 'broken_buf'");
 		exit(EXIT_FAILURE);
 	}
@@ -887,7 +887,7 @@ static void test_inv_buf_client(const struct test_opts *opts, bool stream)
 		exit(EXIT_FAILURE);
 	}
 
-	if (errno != ENOMEM) {
+	if (errno != EFAULT) {
 		fprintf(stderr, "unexpected recv(2) errno %d\n", errno);
 		exit(EXIT_FAILURE);
 	}
-- 
cgit v1.2.3-70-g09d2


From 8e85d550c127d8c559d94c8a2bb7b48e1d75432d Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 29 Mar 2023 08:40:55 +0200
Subject: can: rcar_canfd: rcar_canfd_probe(): fix plain integer in
 transceivers[] init

Fix the following compile warning with C=1:

| drivers/net/can/rcar/rcar_canfd.c:1852:59: warning: Using plain integer as NULL pointer

Fixes: a0340df7eca4f28e ("can: rcar_canfd: Add transceiver support")
Reported-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230328145658.7fdbc394@kernel.org
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com>
Link: https://lore.kernel.org/all/7f7b0dde0caa2d2977b4fb5b65b63036e75f5022.1680071972.git.geert+renesas@glider.be
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/rcar/rcar_canfd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index 701311dabab3..963c42f43755 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -1848,7 +1848,7 @@ static void rcar_canfd_channel_remove(struct rcar_canfd_global *gpriv, u32 ch)
 
 static int rcar_canfd_probe(struct platform_device *pdev)
 {
-	struct phy *transceivers[RCANFD_NUM_CHANNELS] = { 0, };
+	struct phy *transceivers[RCANFD_NUM_CHANNELS] = { NULL, };
 	const struct rcar_canfd_hw_info *info;
 	struct device *dev = &pdev->dev;
 	void __iomem *addr;
-- 
cgit v1.2.3-70-g09d2


From 066b41a599d687fb0409bd0b0f3c41615f070d77 Mon Sep 17 00:00:00 2001
From: Peng Fan <peng.fan@nxp.com>
Date: Tue, 28 Mar 2023 13:46:02 +0800
Subject: dt-bindings: can: fsl,flexcan: add optional power-domains property

Add optional power-domains property for i.MX8 usage.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/all/20230328054602.1974255-1-peng.fan@oss.nxp.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 Documentation/devicetree/bindings/net/can/fsl,flexcan.yaml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/can/fsl,flexcan.yaml b/Documentation/devicetree/bindings/net/can/fsl,flexcan.yaml
index 6e59bd2a6094..4162469c3c08 100644
--- a/Documentation/devicetree/bindings/net/can/fsl,flexcan.yaml
+++ b/Documentation/devicetree/bindings/net/can/fsl,flexcan.yaml
@@ -63,6 +63,9 @@ properties:
       boot loader. This property should only be used the used operating system
       doesn't support the clocks and clock-names property.
 
+  power-domains:
+    maxItems: 1
+
   xceiver-supply:
     description: Regulator that powers the CAN transceiver.
 
-- 
cgit v1.2.3-70-g09d2


From c42fc36949234bc8f8b5afe734acf6b49377f37a Mon Sep 17 00:00:00 2001
From: Frank Jungclaus <frank.jungclaus@esd.eu>
Date: Thu, 30 Mar 2023 20:44:46 +0200
Subject: can: esd_usb: Add support for CAN_CTRLMODE_BERR_REPORTING

Announce that the driver supports CAN_CTRLMODE_BERR_REPORTING by means
of priv->can.ctrlmode_supported. Until now berr reporting always has
been active without taking care of the berr-reporting parameter given
to an "ip link set ..." command.

Additionally apply some changes to function esd_usb_rx_event():
- If berr reporting is off and it is also no state change, then
immediately return.
- Unconditionally (even in case of the above "immediate return") store
tx- and rx-error counters, so directly use priv->bec.txerr and
priv->bec.rxerr instead of intermediate variables.
- Not directly related, but to better point out the linkage between a
failed alloc_can_err_skb() and stats->rx_dropped++:
Move the increment of the rx_dropped statistic counter (back) to
directly behind the err_skb allocation.

Signed-off-by: Frank Jungclaus <frank.jungclaus@esd.eu>
Link: https://lore.kernel.org/all/20230330184446.2802135-1-frank.jungclaus@esd.eu
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/esd_usb.c | 35 ++++++++++++++++++++---------------
 1 file changed, 20 insertions(+), 15 deletions(-)

diff --git a/drivers/net/can/usb/esd_usb.c b/drivers/net/can/usb/esd_usb.c
index e78bb468115a..d33bac3a6c10 100644
--- a/drivers/net/can/usb/esd_usb.c
+++ b/drivers/net/can/usb/esd_usb.c
@@ -237,14 +237,23 @@ static void esd_usb_rx_event(struct esd_usb_net_priv *priv,
 	if (id == ESD_EV_CAN_ERROR_EXT) {
 		u8 state = msg->rx.ev_can_err_ext.status;
 		u8 ecc = msg->rx.ev_can_err_ext.ecc;
-		u8 rxerr = msg->rx.ev_can_err_ext.rec;
-		u8 txerr = msg->rx.ev_can_err_ext.tec;
+
+		priv->bec.rxerr = msg->rx.ev_can_err_ext.rec;
+		priv->bec.txerr = msg->rx.ev_can_err_ext.tec;
 
 		netdev_dbg(priv->netdev,
 			   "CAN_ERR_EV_EXT: dlc=%#02x state=%02x ecc=%02x rec=%02x tec=%02x\n",
-			   msg->rx.dlc, state, ecc, rxerr, txerr);
+			   msg->rx.dlc, state, ecc,
+			   priv->bec.rxerr, priv->bec.txerr);
+
+		/* if berr-reporting is off, only pass through on state change ... */
+		if (!(priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING) &&
+		    state == priv->old_state)
+			return;
 
 		skb = alloc_can_err_skb(priv->netdev, &cf);
+		if (!skb)
+			stats->rx_dropped++;
 
 		if (state != priv->old_state) {
 			enum can_state tx_state, rx_state;
@@ -265,14 +274,14 @@ static void esd_usb_rx_event(struct esd_usb_net_priv *priv,
 				break;
 			default:
 				new_state = CAN_STATE_ERROR_ACTIVE;
-				txerr = 0;
-				rxerr = 0;
+				priv->bec.txerr = 0;
+				priv->bec.rxerr = 0;
 				break;
 			}
 
 			if (new_state != priv->can.state) {
-				tx_state = (txerr >= rxerr) ? new_state : 0;
-				rx_state = (txerr <= rxerr) ? new_state : 0;
+				tx_state = (priv->bec.txerr >= priv->bec.rxerr) ? new_state : 0;
+				rx_state = (priv->bec.txerr <= priv->bec.rxerr) ? new_state : 0;
 				can_change_state(priv->netdev, cf,
 						 tx_state, rx_state);
 			}
@@ -304,17 +313,12 @@ static void esd_usb_rx_event(struct esd_usb_net_priv *priv,
 			cf->data[3] = ecc & SJA1000_ECC_SEG;
 		}
 
-		priv->bec.txerr = txerr;
-		priv->bec.rxerr = rxerr;
-
 		if (skb) {
 			cf->can_id |= CAN_ERR_CNT;
-			cf->data[6] = txerr;
-			cf->data[7] = rxerr;
+			cf->data[6] = priv->bec.txerr;
+			cf->data[7] = priv->bec.rxerr;
 
 			netif_rx(skb);
-		} else {
-			stats->rx_dropped++;
 		}
 	}
 }
@@ -1016,7 +1020,8 @@ static int esd_usb_probe_one_net(struct usb_interface *intf, int index)
 
 	priv->can.state = CAN_STATE_STOPPED;
 	priv->can.ctrlmode_supported = CAN_CTRLMODE_LISTENONLY |
-		CAN_CTRLMODE_CC_LEN8_DLC;
+		CAN_CTRLMODE_CC_LEN8_DLC |
+		CAN_CTRLMODE_BERR_REPORTING;
 
 	if (le16_to_cpu(dev->udev->descriptor.idProduct) ==
 	    USB_CANUSBM_PRODUCT_ID)
-- 
cgit v1.2.3-70-g09d2


From 1afae605e0b2da606d5fff4a8a5849da207128b3 Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <socketcan@hartkopp.net>
Date: Wed, 29 Mar 2023 11:09:15 +0200
Subject: kvaser_usb: convert USB IDs to hexadecimal values

USB IDs are usually represented in 16 bit hexadecimal values. To match
the common representation in lsusb and for searching USB IDs in the
internet convert the decimal values to lowercase hexadecimal.

changes since v1: https://lore.kernel.org/all/20230327175344.4668-1-socketcan@hartkopp.net
- drop the aligned block indentation (suggested by Jimmy)
- use lowercase hex values (suggested by Alex)

Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Acked-by: Jimmy Assarsson <extja@kvaser.com>
Reviewed-by: Alexander Dahl <ada@thorsis.com>
Link: https://lore.kernel.org/all/20230329090915.3127-1-socketcan@hartkopp.net
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c | 102 +++++++++++------------
 1 file changed, 51 insertions(+), 51 deletions(-)

diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
index d4c5356d5884..7135ec851341 100644
--- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
+++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
@@ -31,63 +31,63 @@
 #include "kvaser_usb.h"
 
 /* Kvaser USB vendor id. */
-#define KVASER_VENDOR_ID			0x0bfd
+#define KVASER_VENDOR_ID 0x0bfd
 
 /* Kvaser Leaf USB devices product ids */
-#define USB_LEAF_DEVEL_PRODUCT_ID		10
-#define USB_LEAF_LITE_PRODUCT_ID		11
-#define USB_LEAF_PRO_PRODUCT_ID			12
-#define USB_LEAF_SPRO_PRODUCT_ID		14
-#define USB_LEAF_PRO_LS_PRODUCT_ID		15
-#define USB_LEAF_PRO_SWC_PRODUCT_ID		16
-#define USB_LEAF_PRO_LIN_PRODUCT_ID		17
-#define USB_LEAF_SPRO_LS_PRODUCT_ID		18
-#define USB_LEAF_SPRO_SWC_PRODUCT_ID		19
-#define USB_MEMO2_DEVEL_PRODUCT_ID		22
-#define USB_MEMO2_HSHS_PRODUCT_ID		23
-#define USB_UPRO_HSHS_PRODUCT_ID		24
-#define USB_LEAF_LITE_GI_PRODUCT_ID		25
-#define USB_LEAF_PRO_OBDII_PRODUCT_ID		26
-#define USB_MEMO2_HSLS_PRODUCT_ID		27
-#define USB_LEAF_LITE_CH_PRODUCT_ID		28
-#define USB_BLACKBIRD_SPRO_PRODUCT_ID		29
-#define USB_OEM_MERCURY_PRODUCT_ID		34
-#define USB_OEM_LEAF_PRODUCT_ID			35
-#define USB_CAN_R_PRODUCT_ID			39
-#define USB_LEAF_LITE_V2_PRODUCT_ID		288
-#define USB_MINI_PCIE_HS_PRODUCT_ID		289
-#define USB_LEAF_LIGHT_HS_V2_OEM_PRODUCT_ID	290
-#define USB_USBCAN_LIGHT_2HS_PRODUCT_ID		291
-#define USB_MINI_PCIE_2HS_PRODUCT_ID		292
-#define USB_USBCAN_R_V2_PRODUCT_ID		294
-#define USB_LEAF_LIGHT_R_V2_PRODUCT_ID		295
-#define USB_LEAF_LIGHT_HS_V2_OEM2_PRODUCT_ID	296
+#define USB_LEAF_DEVEL_PRODUCT_ID 0x000a
+#define USB_LEAF_LITE_PRODUCT_ID 0x000b
+#define USB_LEAF_PRO_PRODUCT_ID 0x000c
+#define USB_LEAF_SPRO_PRODUCT_ID 0x000e
+#define USB_LEAF_PRO_LS_PRODUCT_ID 0x000f
+#define USB_LEAF_PRO_SWC_PRODUCT_ID 0x0010
+#define USB_LEAF_PRO_LIN_PRODUCT_ID 0x0011
+#define USB_LEAF_SPRO_LS_PRODUCT_ID 0x0012
+#define USB_LEAF_SPRO_SWC_PRODUCT_ID 0x0013
+#define USB_MEMO2_DEVEL_PRODUCT_ID 0x0016
+#define USB_MEMO2_HSHS_PRODUCT_ID 0x0017
+#define USB_UPRO_HSHS_PRODUCT_ID 0x0018
+#define USB_LEAF_LITE_GI_PRODUCT_ID 0x0019
+#define USB_LEAF_PRO_OBDII_PRODUCT_ID 0x001a
+#define USB_MEMO2_HSLS_PRODUCT_ID 0x001b
+#define USB_LEAF_LITE_CH_PRODUCT_ID 0x001c
+#define USB_BLACKBIRD_SPRO_PRODUCT_ID 0x001d
+#define USB_OEM_MERCURY_PRODUCT_ID 0x0022
+#define USB_OEM_LEAF_PRODUCT_ID 0x0023
+#define USB_CAN_R_PRODUCT_ID 0x0027
+#define USB_LEAF_LITE_V2_PRODUCT_ID 0x0120
+#define USB_MINI_PCIE_HS_PRODUCT_ID 0x0121
+#define USB_LEAF_LIGHT_HS_V2_OEM_PRODUCT_ID 0x0122
+#define USB_USBCAN_LIGHT_2HS_PRODUCT_ID 0x0123
+#define USB_MINI_PCIE_2HS_PRODUCT_ID 0x0124
+#define USB_USBCAN_R_V2_PRODUCT_ID 0x0126
+#define USB_LEAF_LIGHT_R_V2_PRODUCT_ID 0x0127
+#define USB_LEAF_LIGHT_HS_V2_OEM2_PRODUCT_ID 0x0128
 
 /* Kvaser USBCan-II devices product ids */
-#define USB_USBCAN_REVB_PRODUCT_ID		2
-#define USB_VCI2_PRODUCT_ID			3
-#define USB_USBCAN2_PRODUCT_ID			4
-#define USB_MEMORATOR_PRODUCT_ID		5
+#define USB_USBCAN_REVB_PRODUCT_ID 0x0002
+#define USB_VCI2_PRODUCT_ID 0x0003
+#define USB_USBCAN2_PRODUCT_ID 0x0004
+#define USB_MEMORATOR_PRODUCT_ID 0x0005
 
 /* Kvaser Minihydra USB devices product ids */
-#define USB_BLACKBIRD_V2_PRODUCT_ID		258
-#define USB_MEMO_PRO_5HS_PRODUCT_ID		260
-#define USB_USBCAN_PRO_5HS_PRODUCT_ID		261
-#define USB_USBCAN_LIGHT_4HS_PRODUCT_ID		262
-#define USB_LEAF_PRO_HS_V2_PRODUCT_ID		263
-#define USB_USBCAN_PRO_2HS_V2_PRODUCT_ID	264
-#define USB_MEMO_2HS_PRODUCT_ID			265
-#define USB_MEMO_PRO_2HS_V2_PRODUCT_ID		266
-#define USB_HYBRID_2CANLIN_PRODUCT_ID		267
-#define USB_ATI_USBCAN_PRO_2HS_V2_PRODUCT_ID	268
-#define USB_ATI_MEMO_PRO_2HS_V2_PRODUCT_ID	269
-#define USB_HYBRID_PRO_2CANLIN_PRODUCT_ID	270
-#define USB_U100_PRODUCT_ID			273
-#define USB_U100P_PRODUCT_ID			274
-#define USB_U100S_PRODUCT_ID			275
-#define USB_USBCAN_PRO_4HS_PRODUCT_ID		276
-#define USB_HYBRID_CANLIN_PRODUCT_ID		277
-#define USB_HYBRID_PRO_CANLIN_PRODUCT_ID	278
+#define USB_BLACKBIRD_V2_PRODUCT_ID 0x0102
+#define USB_MEMO_PRO_5HS_PRODUCT_ID 0x0104
+#define USB_USBCAN_PRO_5HS_PRODUCT_ID 0x0105
+#define USB_USBCAN_LIGHT_4HS_PRODUCT_ID 0x0106
+#define USB_LEAF_PRO_HS_V2_PRODUCT_ID 0x0107
+#define USB_USBCAN_PRO_2HS_V2_PRODUCT_ID 0x0108
+#define USB_MEMO_2HS_PRODUCT_ID 0x0109
+#define USB_MEMO_PRO_2HS_V2_PRODUCT_ID 0x010a
+#define USB_HYBRID_2CANLIN_PRODUCT_ID 0x010b
+#define USB_ATI_USBCAN_PRO_2HS_V2_PRODUCT_ID 0x010c
+#define USB_ATI_MEMO_PRO_2HS_V2_PRODUCT_ID 0x010d
+#define USB_HYBRID_PRO_2CANLIN_PRODUCT_ID 0x010e
+#define USB_U100_PRODUCT_ID 0x0111
+#define USB_U100P_PRODUCT_ID 0x0112
+#define USB_U100S_PRODUCT_ID 0x0113
+#define USB_USBCAN_PRO_4HS_PRODUCT_ID 0x0114
+#define USB_HYBRID_CANLIN_PRODUCT_ID 0x0115
+#define USB_HYBRID_PRO_CANLIN_PRODUCT_ID 0x0116
 
 static const struct kvaser_usb_driver_info kvaser_usb_driver_info_hydra = {
 	.quirks = KVASER_USB_QUIRK_HAS_HARDWARE_TIMESTAMP,
-- 
cgit v1.2.3-70-g09d2


From 8fc59c26d212c23d6fd5ad47a10651cf72d83b4a Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Mon, 3 Apr 2023 19:29:35 +0200
Subject: selftests/bpf: Add RESOLVE_BTFIDS dependency to bpf_testmod.ko

bpf_testmod.ko sometimes fails to build from a clean checkout:

    BTF [M] linux/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.ko
    /bin/sh: 1: linux-build//tools/build/resolve_btfids/resolve_btfids: not found

The reason is that RESOLVE_BTFIDS may not yet be built. Fix by adding a
dependency.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20230403172935.1553022-1-iii@linux.ibm.com
---
 tools/testing/selftests/bpf/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 4a8ef118fd9d..febd1dae6c88 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -201,7 +201,7 @@ $(OUTPUT)/sign-file: ../../../../scripts/sign-file.c
 		  $< -o $@ \
 		  $(shell $(HOSTPKG_CONFIG) --libs libcrypto 2> /dev/null || echo -lcrypto)
 
-$(OUTPUT)/bpf_testmod.ko: $(VMLINUX_BTF) $(wildcard bpf_testmod/Makefile bpf_testmod/*.[ch])
+$(OUTPUT)/bpf_testmod.ko: $(VMLINUX_BTF) $(RESOLVE_BTFIDS) $(wildcard bpf_testmod/Makefile bpf_testmod/*.[ch])
 	$(call msg,MOD,,$@)
 	$(Q)$(RM) bpf_testmod/bpf_testmod.ko # force re-compilation
 	$(Q)$(MAKE) $(submake_extras) RESOLVE_BTFIDS=$(RESOLVE_BTFIDS) -C bpf_testmod
-- 
cgit v1.2.3-70-g09d2


From 7d64c513284408fee5178a0953a686e9410f2399 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:22 -0700
Subject: bpf: Invoke btf_struct_access() callback only for writes.

Remove duplicated if (atype == BPF_READ) btf_struct_access() from
btf_struct_access() callback and invoke it only for writes. This is
possible to do because currently btf_struct_access() custom callback
always delegates to generic btf_struct_access() helper for BPF_READ
accesses.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-2-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c          | 2 +-
 net/bpf/bpf_dummy_struct_ops.c | 2 +-
 net/core/filter.c              | 6 ------
 net/ipv4/bpf_tcp_ca.c          | 3 ---
 4 files changed, 2 insertions(+), 11 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index eaf9c5291cf0..83984568ccb4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5504,7 +5504,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 		return -EACCES;
 	}
 
-	if (env->ops->btf_struct_access && !type_is_alloc(reg->type)) {
+	if (env->ops->btf_struct_access && !type_is_alloc(reg->type) && atype == BPF_WRITE) {
 		if (!btf_is_kernel(reg->btf)) {
 			verbose(env, "verifier internal error: reg->btf must be kernel btf\n");
 			return -EFAULT;
diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c
index ff4f89a2b02a..9535c8506cda 100644
--- a/net/bpf/bpf_dummy_struct_ops.c
+++ b/net/bpf/bpf_dummy_struct_ops.c
@@ -198,7 +198,7 @@ static int bpf_dummy_ops_btf_struct_access(struct bpf_verifier_log *log,
 	if (err < 0)
 		return err;
 
-	return atype == BPF_READ ? err : NOT_INIT;
+	return NOT_INIT;
 }
 
 static const struct bpf_verifier_ops bpf_dummy_verifier_ops = {
diff --git a/net/core/filter.c b/net/core/filter.c
index 3370efad1dda..8b9f409a2ec3 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -8753,9 +8753,6 @@ static int tc_cls_act_btf_struct_access(struct bpf_verifier_log *log,
 {
 	int ret = -EACCES;
 
-	if (atype == BPF_READ)
-		return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
-
 	mutex_lock(&nf_conn_btf_access_lock);
 	if (nfct_btf_struct_access)
 		ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
@@ -8830,9 +8827,6 @@ static int xdp_btf_struct_access(struct bpf_verifier_log *log,
 {
 	int ret = -EACCES;
 
-	if (atype == BPF_READ)
-		return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
-
 	mutex_lock(&nf_conn_btf_access_lock);
 	if (nfct_btf_struct_access)
 		ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index ea21c96c03aa..d6465876bbf6 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -78,9 +78,6 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log,
 	const struct btf_type *t;
 	size_t end;
 
-	if (atype == BPF_READ)
-		return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
-
 	t = btf_type_by_id(reg->btf, reg->btf_id);
 	if (t != tcp_sock_type) {
 		bpf_log(log, "only read is supported\n");
-- 
cgit v1.2.3-70-g09d2


From b7e852a9ec96635168c04204fb7cf1f7390b9a8c Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:23 -0700
Subject: bpf: Remove unused arguments from btf_struct_access().

Remove unused arguments from btf_struct_access() callback.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-3-alexei.starovoitov@gmail.com
---
 include/linux/bpf.h              |  3 +--
 include/linux/filter.h           |  3 +--
 kernel/bpf/verifier.c            |  4 ++--
 net/bpf/bpf_dummy_struct_ops.c   | 12 +++++-------
 net/core/filter.c                | 13 +++++--------
 net/ipv4/bpf_tcp_ca.c            |  3 +--
 net/netfilter/nf_conntrack_bpf.c |  3 +--
 7 files changed, 16 insertions(+), 25 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 2d8f3f639e68..4f689dda748f 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -893,8 +893,7 @@ struct bpf_verifier_ops {
 				  struct bpf_prog *prog, u32 *target_size);
 	int (*btf_struct_access)(struct bpf_verifier_log *log,
 				 const struct bpf_reg_state *reg,
-				 int off, int size, enum bpf_access_type atype,
-				 u32 *next_btf_id, enum bpf_type_flag *flag);
+				 int off, int size);
 };
 
 struct bpf_prog_offload_ops {
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 23c08c31bea9..5364b0c52c1d 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -571,8 +571,7 @@ DECLARE_STATIC_KEY_FALSE(bpf_stats_enabled_key);
 extern struct mutex nf_conn_btf_access_lock;
 extern int (*nfct_btf_struct_access)(struct bpf_verifier_log *log,
 				     const struct bpf_reg_state *reg,
-				     int off, int size, enum bpf_access_type atype,
-				     u32 *next_btf_id, enum bpf_type_flag *flag);
+				     int off, int size);
 
 typedef unsigned int (*bpf_dispatcher_fn)(const void *ctx,
 					  const struct bpf_insn *insnsi,
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 83984568ccb4..5ca520e5eddf 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5459,7 +5459,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 	const struct btf_type *t = btf_type_by_id(reg->btf, reg->btf_id);
 	const char *tname = btf_name_by_offset(reg->btf, t->name_off);
 	enum bpf_type_flag flag = 0;
-	u32 btf_id;
+	u32 btf_id = 0;
 	int ret;
 
 	if (!env->allow_ptr_leaks) {
@@ -5509,7 +5509,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 			verbose(env, "verifier internal error: reg->btf must be kernel btf\n");
 			return -EFAULT;
 		}
-		ret = env->ops->btf_struct_access(&env->log, reg, off, size, atype, &btf_id, &flag);
+		ret = env->ops->btf_struct_access(&env->log, reg, off, size);
 	} else {
 		/* Writes are permitted with default btf_struct_access for
 		 * program allocated objects (which always have ref_obj_id > 0),
diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c
index 9535c8506cda..5918d1b32e19 100644
--- a/net/bpf/bpf_dummy_struct_ops.c
+++ b/net/bpf/bpf_dummy_struct_ops.c
@@ -173,14 +173,11 @@ static int bpf_dummy_ops_check_member(const struct btf_type *t,
 
 static int bpf_dummy_ops_btf_struct_access(struct bpf_verifier_log *log,
 					   const struct bpf_reg_state *reg,
-					   int off, int size, enum bpf_access_type atype,
-					   u32 *next_btf_id,
-					   enum bpf_type_flag *flag)
+					   int off, int size)
 {
 	const struct btf_type *state;
 	const struct btf_type *t;
 	s32 type_id;
-	int err;
 
 	type_id = btf_find_by_name_kind(reg->btf, "bpf_dummy_ops_state",
 					BTF_KIND_STRUCT);
@@ -194,9 +191,10 @@ static int bpf_dummy_ops_btf_struct_access(struct bpf_verifier_log *log,
 		return -EACCES;
 	}
 
-	err = btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
-	if (err < 0)
-		return err;
+	if (off + size > sizeof(struct bpf_dummy_ops_state)) {
+		bpf_log(log, "write access at off %d with size %d\n", off, size);
+		return -EACCES;
+	}
 
 	return NOT_INIT;
 }
diff --git a/net/core/filter.c b/net/core/filter.c
index 8b9f409a2ec3..1f2abf0f60e6 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -8742,20 +8742,18 @@ EXPORT_SYMBOL_GPL(nf_conn_btf_access_lock);
 
 int (*nfct_btf_struct_access)(struct bpf_verifier_log *log,
 			      const struct bpf_reg_state *reg,
-			      int off, int size, enum bpf_access_type atype,
-			      u32 *next_btf_id, enum bpf_type_flag *flag);
+			      int off, int size);
 EXPORT_SYMBOL_GPL(nfct_btf_struct_access);
 
 static int tc_cls_act_btf_struct_access(struct bpf_verifier_log *log,
 					const struct bpf_reg_state *reg,
-					int off, int size, enum bpf_access_type atype,
-					u32 *next_btf_id, enum bpf_type_flag *flag)
+					int off, int size)
 {
 	int ret = -EACCES;
 
 	mutex_lock(&nf_conn_btf_access_lock);
 	if (nfct_btf_struct_access)
-		ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
+		ret = nfct_btf_struct_access(log, reg, off, size);
 	mutex_unlock(&nf_conn_btf_access_lock);
 
 	return ret;
@@ -8822,14 +8820,13 @@ EXPORT_SYMBOL_GPL(bpf_warn_invalid_xdp_action);
 
 static int xdp_btf_struct_access(struct bpf_verifier_log *log,
 				 const struct bpf_reg_state *reg,
-				 int off, int size, enum bpf_access_type atype,
-				 u32 *next_btf_id, enum bpf_type_flag *flag)
+				 int off, int size)
 {
 	int ret = -EACCES;
 
 	mutex_lock(&nf_conn_btf_access_lock);
 	if (nfct_btf_struct_access)
-		ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag);
+		ret = nfct_btf_struct_access(log, reg, off, size);
 	mutex_unlock(&nf_conn_btf_access_lock);
 
 	return ret;
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index d6465876bbf6..4406d796cc2f 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -72,8 +72,7 @@ static bool bpf_tcp_ca_is_valid_access(int off, int size,
 
 static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log,
 					const struct bpf_reg_state *reg,
-					int off, int size, enum bpf_access_type atype,
-					u32 *next_btf_id, enum bpf_type_flag *flag)
+					int off, int size)
 {
 	const struct btf_type *t;
 	size_t end;
diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c
index 002e9d24a1e9..3f821b7ba646 100644
--- a/net/netfilter/nf_conntrack_bpf.c
+++ b/net/netfilter/nf_conntrack_bpf.c
@@ -192,8 +192,7 @@ BTF_ID(struct, nf_conn___init)
 /* Check writes into `struct nf_conn` */
 static int _nf_conntrack_btf_struct_access(struct bpf_verifier_log *log,
 					   const struct bpf_reg_state *reg,
-					   int off, int size, enum bpf_access_type atype,
-					   u32 *next_btf_id, enum bpf_type_flag *flag)
+					   int off, int size)
 {
 	const struct btf_type *ncit, *nct, *t;
 	size_t end;
-- 
cgit v1.2.3-70-g09d2


From 63260df1396578226ac3134cf7f764690002e70e Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:24 -0700
Subject: bpf: Refactor btf_nested_type_is_trusted().

btf_nested_type_is_trusted() tries to find a struct member at corresponding offset.
It works for flat structures and falls apart in more complex structs with nested structs.
The offset->member search is already performed by btf_struct_walk() including nested structs.
Reuse this work and pass {field name, field btf id} into btf_nested_type_is_trusted()
instead of offset to make BTF_TYPE_SAFE*() logic more robust.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-4-alexei.starovoitov@gmail.com
---
 include/linux/bpf.h   |  7 ++++---
 kernel/bpf/btf.c      | 44 +++++++++++++++++---------------------------
 kernel/bpf/verifier.c | 23 ++++++++++++-----------
 3 files changed, 33 insertions(+), 41 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 4f689dda748f..002a811b6b90 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2263,7 +2263,7 @@ static inline bool bpf_tracing_btf_ctx_access(int off, int size,
 int btf_struct_access(struct bpf_verifier_log *log,
 		      const struct bpf_reg_state *reg,
 		      int off, int size, enum bpf_access_type atype,
-		      u32 *next_btf_id, enum bpf_type_flag *flag);
+		      u32 *next_btf_id, enum bpf_type_flag *flag, const char **field_name);
 bool btf_struct_ids_match(struct bpf_verifier_log *log,
 			  const struct btf *btf, u32 id, int off,
 			  const struct btf *need_btf, u32 need_type_id,
@@ -2302,7 +2302,7 @@ struct bpf_core_ctx {
 
 bool btf_nested_type_is_trusted(struct bpf_verifier_log *log,
 				const struct bpf_reg_state *reg,
-				int off, const char *suffix);
+				const char *field_name, u32 btf_id, const char *suffix);
 
 bool btf_type_ids_nocast_alias(struct bpf_verifier_log *log,
 			       const struct btf *reg_btf, u32 reg_id,
@@ -2517,7 +2517,8 @@ static inline struct bpf_prog *bpf_prog_by_id(u32 id)
 static inline int btf_struct_access(struct bpf_verifier_log *log,
 				    const struct bpf_reg_state *reg,
 				    int off, int size, enum bpf_access_type atype,
-				    u32 *next_btf_id, enum bpf_type_flag *flag)
+				    u32 *next_btf_id, enum bpf_type_flag *flag,
+				    const char **field_name)
 {
 	return -EACCES;
 }
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index b7e5a5510b91..593c45a294d0 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -6166,7 +6166,8 @@ enum bpf_struct_walk_result {
 
 static int btf_struct_walk(struct bpf_verifier_log *log, const struct btf *btf,
 			   const struct btf_type *t, int off, int size,
-			   u32 *next_btf_id, enum bpf_type_flag *flag)
+			   u32 *next_btf_id, enum bpf_type_flag *flag,
+			   const char **field_name)
 {
 	u32 i, moff, mtrue_end, msize = 0, total_nelems = 0;
 	const struct btf_type *mtype, *elem_type = NULL;
@@ -6395,6 +6396,8 @@ error:
 			if (btf_type_is_struct(stype)) {
 				*next_btf_id = id;
 				*flag |= tmp_flag;
+				if (field_name)
+					*field_name = mname;
 				return WALK_PTR;
 			}
 		}
@@ -6421,7 +6424,8 @@ error:
 int btf_struct_access(struct bpf_verifier_log *log,
 		      const struct bpf_reg_state *reg,
 		      int off, int size, enum bpf_access_type atype __maybe_unused,
-		      u32 *next_btf_id, enum bpf_type_flag *flag)
+		      u32 *next_btf_id, enum bpf_type_flag *flag,
+		      const char **field_name)
 {
 	const struct btf *btf = reg->btf;
 	enum bpf_type_flag tmp_flag = 0;
@@ -6453,7 +6457,7 @@ int btf_struct_access(struct bpf_verifier_log *log,
 
 	t = btf_type_by_id(btf, id);
 	do {
-		err = btf_struct_walk(log, btf, t, off, size, &id, &tmp_flag);
+		err = btf_struct_walk(log, btf, t, off, size, &id, &tmp_flag, field_name);
 
 		switch (err) {
 		case WALK_PTR:
@@ -6528,7 +6532,7 @@ again:
 	type = btf_type_by_id(btf, id);
 	if (!type)
 		return false;
-	err = btf_struct_walk(log, btf, type, off, 1, &id, &flag);
+	err = btf_struct_walk(log, btf, type, off, 1, &id, &flag, NULL);
 	if (err != WALK_STRUCT)
 		return false;
 
@@ -8488,16 +8492,15 @@ out:
 
 bool btf_nested_type_is_trusted(struct bpf_verifier_log *log,
 				const struct bpf_reg_state *reg,
-				int off, const char *suffix)
+				const char *field_name, u32 btf_id, const char *suffix)
 {
 	struct btf *btf = reg->btf;
 	const struct btf_type *walk_type, *safe_type;
 	const char *tname;
 	char safe_tname[64];
 	long ret, safe_id;
-	const struct btf_member *member, *m_walk = NULL;
+	const struct btf_member *member;
 	u32 i;
-	const char *walk_name;
 
 	walk_type = btf_type_by_id(btf, reg->btf_id);
 	if (!walk_type)
@@ -8517,30 +8520,17 @@ bool btf_nested_type_is_trusted(struct bpf_verifier_log *log,
 	if (!safe_type)
 		return false;
 
-	for_each_member(i, walk_type, member) {
-		u32 moff;
-
-		/* We're looking for the PTR_TO_BTF_ID member in the struct
-		 * type we're walking which matches the specified offset.
-		 * Below, we'll iterate over the fields in the safe variant of
-		 * the struct and see if any of them has a matching type /
-		 * name.
-		 */
-		moff = __btf_member_bit_offset(walk_type, member) / 8;
-		if (off == moff) {
-			m_walk = member;
-			break;
-		}
-	}
-	if (m_walk == NULL)
-		return false;
-
-	walk_name = __btf_name_by_offset(btf, m_walk->name_off);
 	for_each_member(i, safe_type, member) {
 		const char *m_name = __btf_name_by_offset(btf, member->name_off);
+		const struct btf_type *mtype = btf_type_by_id(btf, member->type);
+		u32 id;
+
+		if (!btf_type_is_ptr(mtype))
+			continue;
 
+		btf_type_skip_modifiers(btf, mtype->type, &id);
 		/* If we match on both type and name, the field is considered trusted. */
-		if (m_walk->type == member->type && !strcmp(walk_name, m_name))
+		if (btf_id == id && !strcmp(field_name, m_name))
 			return true;
 	}
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5ca520e5eddf..2cd2e0b725cd 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5400,12 +5400,12 @@ BTF_TYPE_SAFE_RCU(struct css_set) {
 
 /* full trusted: these fields are trusted even outside of RCU CS and never NULL */
 BTF_TYPE_SAFE_TRUSTED(struct bpf_iter_meta) {
-	__bpf_md_ptr(struct seq_file *, seq);
+	struct seq_file *seq;
 };
 
 BTF_TYPE_SAFE_TRUSTED(struct bpf_iter__task) {
-	__bpf_md_ptr(struct bpf_iter_meta *, meta);
-	__bpf_md_ptr(struct task_struct *, task);
+	struct bpf_iter_meta *meta;
+	struct task_struct *task;
 };
 
 BTF_TYPE_SAFE_TRUSTED(struct linux_binprm) {
@@ -5427,17 +5427,17 @@ BTF_TYPE_SAFE_TRUSTED(struct socket) {
 
 static bool type_is_rcu(struct bpf_verifier_env *env,
 			struct bpf_reg_state *reg,
-			int off)
+			const char *field_name, u32 btf_id)
 {
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct task_struct));
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct css_set));
 
-	return btf_nested_type_is_trusted(&env->log, reg, off, "__safe_rcu");
+	return btf_nested_type_is_trusted(&env->log, reg, field_name, btf_id, "__safe_rcu");
 }
 
 static bool type_is_trusted(struct bpf_verifier_env *env,
 			    struct bpf_reg_state *reg,
-			    int off)
+			    const char *field_name, u32 btf_id)
 {
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct bpf_iter_meta));
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct bpf_iter__task));
@@ -5446,7 +5446,7 @@ static bool type_is_trusted(struct bpf_verifier_env *env,
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct dentry));
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_TRUSTED(struct socket));
 
-	return btf_nested_type_is_trusted(&env->log, reg, off, "__safe_trusted");
+	return btf_nested_type_is_trusted(&env->log, reg, field_name, btf_id, "__safe_trusted");
 }
 
 static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
@@ -5458,6 +5458,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 	struct bpf_reg_state *reg = regs + regno;
 	const struct btf_type *t = btf_type_by_id(reg->btf, reg->btf_id);
 	const char *tname = btf_name_by_offset(reg->btf, t->name_off);
+	const char *field_name = NULL;
 	enum bpf_type_flag flag = 0;
 	u32 btf_id = 0;
 	int ret;
@@ -5526,7 +5527,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 			return -EFAULT;
 		}
 
-		ret = btf_struct_access(&env->log, reg, off, size, atype, &btf_id, &flag);
+		ret = btf_struct_access(&env->log, reg, off, size, atype, &btf_id, &flag, &field_name);
 	}
 
 	if (ret < 0)
@@ -5554,10 +5555,10 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 		 * A regular RCU-protected pointer with __rcu tag can also be deemed
 		 * trusted if we are in an RCU CS. Such pointer can be NULL.
 		 */
-		if (type_is_trusted(env, reg, off)) {
+		if (type_is_trusted(env, reg, field_name, btf_id)) {
 			flag |= PTR_TRUSTED;
 		} else if (in_rcu_cs(env) && !type_may_be_null(reg->type)) {
-			if (type_is_rcu(env, reg, off)) {
+			if (type_is_rcu(env, reg, field_name, btf_id)) {
 				/* ignore __rcu tag and mark it MEM_RCU */
 				flag |= MEM_RCU;
 			} else if (flag & MEM_RCU) {
@@ -5640,7 +5641,7 @@ static int check_ptr_to_map_access(struct bpf_verifier_env *env,
 	/* Simulate access to a PTR_TO_BTF_ID */
 	memset(&map_reg, 0, sizeof(map_reg));
 	mark_btf_ld_reg(env, &map_reg, 0, PTR_TO_BTF_ID, btf_vmlinux, *map->ops->map_btf_id, 0);
-	ret = btf_struct_access(&env->log, &map_reg, off, size, atype, &btf_id, &flag);
+	ret = btf_struct_access(&env->log, &map_reg, off, size, atype, &btf_id, &flag, NULL);
 	if (ret < 0)
 		return ret;
 
-- 
cgit v1.2.3-70-g09d2


From 91571a515d1bcdc280bb46423bb697ea7eb42ff3 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:25 -0700
Subject: bpf: Teach verifier that certain helpers accept NULL pointer.

bpf_[sk|inode|task|cgrp]_storage_[get|delete]() and bpf_get_socket_cookie() helpers
perform run-time check that sk|inode|task|cgrp pointer != NULL.
Teach verifier about this fact and allow bpf programs to pass
PTR_TO_BTF_ID | PTR_MAYBE_NULL into such helpers.
It will be used in the subsequent patch that will do
bpf_sk_storage_get(.., skb->sk, ...);
Even when 'skb' pointer is trusted the 'sk' pointer may be NULL.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-5-alexei.starovoitov@gmail.com
---
 kernel/bpf/bpf_cgrp_storage.c  | 4 ++--
 kernel/bpf/bpf_inode_storage.c | 4 ++--
 kernel/bpf/bpf_task_storage.c  | 8 ++++----
 net/core/bpf_sk_storage.c      | 4 ++--
 net/core/filter.c              | 2 +-
 5 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c
index d17d5b694668..d44fe8dd9732 100644
--- a/kernel/bpf/bpf_cgrp_storage.c
+++ b/kernel/bpf/bpf_cgrp_storage.c
@@ -224,7 +224,7 @@ const struct bpf_func_proto bpf_cgrp_storage_get_proto = {
 	.gpl_only	= false,
 	.ret_type	= RET_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg1_type	= ARG_CONST_MAP_PTR,
-	.arg2_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id	= &bpf_cgroup_btf_id[0],
 	.arg3_type	= ARG_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg4_type	= ARG_ANYTHING,
@@ -235,6 +235,6 @@ const struct bpf_func_proto bpf_cgrp_storage_delete_proto = {
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
 	.arg1_type	= ARG_CONST_MAP_PTR,
-	.arg2_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id	= &bpf_cgroup_btf_id[0],
 };
diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c
index e17ad581b9be..a4d93df78c75 100644
--- a/kernel/bpf/bpf_inode_storage.c
+++ b/kernel/bpf/bpf_inode_storage.c
@@ -229,7 +229,7 @@ const struct bpf_func_proto bpf_inode_storage_get_proto = {
 	.gpl_only	= false,
 	.ret_type	= RET_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg1_type	= ARG_CONST_MAP_PTR,
-	.arg2_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id	= &bpf_inode_storage_btf_ids[0],
 	.arg3_type	= ARG_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg4_type	= ARG_ANYTHING,
@@ -240,6 +240,6 @@ const struct bpf_func_proto bpf_inode_storage_delete_proto = {
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
 	.arg1_type	= ARG_CONST_MAP_PTR,
-	.arg2_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id	= &bpf_inode_storage_btf_ids[0],
 };
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index d1af0c8f9ce4..adf6dfe0ba68 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -338,7 +338,7 @@ const struct bpf_func_proto bpf_task_storage_get_recur_proto = {
 	.gpl_only = false,
 	.ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg1_type = ARG_CONST_MAP_PTR,
-	.arg2_type = ARG_PTR_TO_BTF_ID,
+	.arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK],
 	.arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg4_type = ARG_ANYTHING,
@@ -349,7 +349,7 @@ const struct bpf_func_proto bpf_task_storage_get_proto = {
 	.gpl_only = false,
 	.ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg1_type = ARG_CONST_MAP_PTR,
-	.arg2_type = ARG_PTR_TO_BTF_ID,
+	.arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK],
 	.arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg4_type = ARG_ANYTHING,
@@ -360,7 +360,7 @@ const struct bpf_func_proto bpf_task_storage_delete_recur_proto = {
 	.gpl_only = false,
 	.ret_type = RET_INTEGER,
 	.arg1_type = ARG_CONST_MAP_PTR,
-	.arg2_type = ARG_PTR_TO_BTF_ID,
+	.arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK],
 };
 
@@ -369,6 +369,6 @@ const struct bpf_func_proto bpf_task_storage_delete_proto = {
 	.gpl_only = false,
 	.ret_type = RET_INTEGER,
 	.arg1_type = ARG_CONST_MAP_PTR,
-	.arg2_type = ARG_PTR_TO_BTF_ID,
+	.arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK],
 };
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index 085025c7130a..d4172534dfa8 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -412,7 +412,7 @@ const struct bpf_func_proto bpf_sk_storage_get_tracing_proto = {
 	.gpl_only	= false,
 	.ret_type	= RET_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg1_type	= ARG_CONST_MAP_PTR,
-	.arg2_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id	= &btf_sock_ids[BTF_SOCK_TYPE_SOCK_COMMON],
 	.arg3_type	= ARG_PTR_TO_MAP_VALUE_OR_NULL,
 	.arg4_type	= ARG_ANYTHING,
@@ -424,7 +424,7 @@ const struct bpf_func_proto bpf_sk_storage_delete_tracing_proto = {
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
 	.arg1_type	= ARG_CONST_MAP_PTR,
-	.arg2_type	= ARG_PTR_TO_BTF_ID,
+	.arg2_type	= ARG_PTR_TO_BTF_ID_OR_NULL,
 	.arg2_btf_id	= &btf_sock_ids[BTF_SOCK_TYPE_SOCK_COMMON],
 	.allowed	= bpf_sk_storage_tracing_allowed,
 };
diff --git a/net/core/filter.c b/net/core/filter.c
index 1f2abf0f60e6..727c5269867d 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -4998,7 +4998,7 @@ const struct bpf_func_proto bpf_get_socket_ptr_cookie_proto = {
 	.func		= bpf_get_socket_ptr_cookie,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
-	.arg1_type	= ARG_PTR_TO_BTF_ID_SOCK_COMMON,
+	.arg1_type	= ARG_PTR_TO_BTF_ID_SOCK_COMMON | PTR_MAYBE_NULL,
 };
 
 BPF_CALL_1(bpf_get_socket_cookie_sock_ops, struct bpf_sock_ops_kern *, ctx)
-- 
cgit v1.2.3-70-g09d2


From add68b843f33d4e5dcbdc7ba6dffe7750a964159 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:26 -0700
Subject: bpf: Refactor NULL-ness check in check_reg_type().

check_reg_type() unconditionally disallows PTR_TO_BTF_ID | PTR_MAYBE_NULL.
It's problematic for helpers that allow ARG_PTR_TO_BTF_ID_OR_NULL like
bpf_sk_storage_get(). Allow passing PTR_TO_BTF_ID | PTR_MAYBE_NULL into such
helpers. That technically includes bpf_kptr_xchg() helper, but in practice:
  bpf_kptr_xchg(..., bpf_cpumask_create());
is still disallowed because bpf_cpumask_create() returns ref counted pointer
with ref_obj_id > 0.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-6-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 2cd2e0b725cd..4e7d671497f4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -7168,6 +7168,8 @@ found:
 	case PTR_TO_BTF_ID:
 	case PTR_TO_BTF_ID | PTR_TRUSTED:
 	case PTR_TO_BTF_ID | MEM_RCU:
+	case PTR_TO_BTF_ID | PTR_MAYBE_NULL:
+	case PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_RCU:
 	{
 		/* For bpf_sk_release, it needs to match against first member
 		 * 'struct sock_common', hence make an exception for it. This
@@ -7176,6 +7178,12 @@ found:
 		bool strict_type_match = arg_type_is_release(arg_type) &&
 					 meta->func_id != BPF_FUNC_sk_release;
 
+		if (type_may_be_null(reg->type) &&
+		    (!type_may_be_null(arg_type) || arg_type_is_release(arg_type))) {
+			verbose(env, "Possibly NULL pointer passed to helper arg%d\n", regno);
+			return -EACCES;
+		}
+
 		if (!arg_btf_id) {
 			if (!compatible->btf_id) {
 				verbose(env, "verifier internal error: missing arg compatible BTF ID\n");
@@ -7206,10 +7214,6 @@ found:
 		}
 		break;
 	}
-	case PTR_TO_BTF_ID | PTR_MAYBE_NULL:
-	case PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_RCU:
-		verbose(env, "Possibly NULL pointer passed to helper arg%d\n", regno);
-		return -EACCES;
 	case PTR_TO_BTF_ID | MEM_ALLOC:
 		if (meta->func_id != BPF_FUNC_spin_lock && meta->func_id != BPF_FUNC_spin_unlock &&
 		    meta->func_id != BPF_FUNC_kptr_xchg) {
-- 
cgit v1.2.3-70-g09d2


From 30ee9821f9430c34a6254134a5f0e8db227510be Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:27 -0700
Subject: bpf: Allowlist few fields similar to __rcu tag.

Allow bpf program access cgrp->kn, mm->exe_file, skb->sk, req->sk.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-7-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c | 39 +++++++++++++++++++++++++++++++++++++--
 1 file changed, 37 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 4e7d671497f4..fd90ba498ccc 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5378,6 +5378,7 @@ static int bpf_map_direct_read(struct bpf_map *map, int off, int size, u64 *val)
 }
 
 #define BTF_TYPE_SAFE_RCU(__type)  __PASTE(__type, __safe_rcu)
+#define BTF_TYPE_SAFE_RCU_OR_NULL(__type)  __PASTE(__type, __safe_rcu_or_null)
 #define BTF_TYPE_SAFE_TRUSTED(__type)  __PASTE(__type, __safe_trusted)
 
 /*
@@ -5394,10 +5395,31 @@ BTF_TYPE_SAFE_RCU(struct task_struct) {
 	struct task_struct *group_leader;
 };
 
+BTF_TYPE_SAFE_RCU(struct cgroup) {
+	/* cgrp->kn is always accessible as documented in kernel/cgroup/cgroup.c */
+	struct kernfs_node *kn;
+};
+
 BTF_TYPE_SAFE_RCU(struct css_set) {
 	struct cgroup *dfl_cgrp;
 };
 
+/* RCU trusted: these fields are trusted in RCU CS and can be NULL */
+BTF_TYPE_SAFE_RCU_OR_NULL(struct mm_struct) {
+	struct file __rcu *exe_file;
+};
+
+/* skb->sk, req->sk are not RCU protected, but we mark them as such
+ * because bpf prog accessible sockets are SOCK_RCU_FREE.
+ */
+BTF_TYPE_SAFE_RCU_OR_NULL(struct sk_buff) {
+	struct sock *sk;
+};
+
+BTF_TYPE_SAFE_RCU_OR_NULL(struct request_sock) {
+	struct sock *sk;
+};
+
 /* full trusted: these fields are trusted even outside of RCU CS and never NULL */
 BTF_TYPE_SAFE_TRUSTED(struct bpf_iter_meta) {
 	struct seq_file *seq;
@@ -5430,11 +5452,23 @@ static bool type_is_rcu(struct bpf_verifier_env *env,
 			const char *field_name, u32 btf_id)
 {
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct task_struct));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct cgroup));
 	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU(struct css_set));
 
 	return btf_nested_type_is_trusted(&env->log, reg, field_name, btf_id, "__safe_rcu");
 }
 
+static bool type_is_rcu_or_null(struct bpf_verifier_env *env,
+				struct bpf_reg_state *reg,
+				const char *field_name, u32 btf_id)
+{
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU_OR_NULL(struct mm_struct));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU_OR_NULL(struct sk_buff));
+	BTF_TYPE_EMIT(BTF_TYPE_SAFE_RCU_OR_NULL(struct request_sock));
+
+	return btf_nested_type_is_trusted(&env->log, reg, field_name, btf_id, "__safe_rcu_or_null");
+}
+
 static bool type_is_trusted(struct bpf_verifier_env *env,
 			    struct bpf_reg_state *reg,
 			    const char *field_name, u32 btf_id)
@@ -5561,9 +5595,10 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 			if (type_is_rcu(env, reg, field_name, btf_id)) {
 				/* ignore __rcu tag and mark it MEM_RCU */
 				flag |= MEM_RCU;
-			} else if (flag & MEM_RCU) {
+			} else if (flag & MEM_RCU ||
+				   type_is_rcu_or_null(env, reg, field_name, btf_id)) {
 				/* __rcu tagged pointers can be NULL */
-				flag |= PTR_MAYBE_NULL;
+				flag |= MEM_RCU | PTR_MAYBE_NULL;
 			} else if (flag & (MEM_PERCPU | MEM_USER)) {
 				/* keep as-is */
 			} else {
-- 
cgit v1.2.3-70-g09d2


From afeebf9f57a4965e0be90e4dc87f8f3a1417376d Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:28 -0700
Subject: bpf: Undo strict enforcement for walking untagged fields.

The commit 6fcd486b3a0a ("bpf: Refactor RCU enforcement in the verifier.")
broke several tracing bpf programs. Even in clang compiled kernels there are
many fields that are not marked with __rcu that are safe to read and pass into
helpers, but the verifier doesn't know that they're safe. Aggressively marking
them as PTR_UNTRUSTED was premature.

Fixes: 6fcd486b3a0a ("bpf: Refactor RCU enforcement in the verifier.")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-8-alexei.starovoitov@gmail.com
---
 kernel/bpf/verifier.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index fd90ba498ccc..56f569811f70 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4974,6 +4974,11 @@ static bool is_rcu_reg(const struct bpf_reg_state *reg)
 	return reg->type & MEM_RCU;
 }
 
+static void clear_trusted_flags(enum bpf_type_flag *flag)
+{
+	*flag &= ~(BPF_REG_TRUSTED_MODIFIERS | MEM_RCU);
+}
+
 static int check_pkt_ptr_alignment(struct bpf_verifier_env *env,
 				   const struct bpf_reg_state *reg,
 				   int off, int size, bool strict)
@@ -5602,8 +5607,8 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 			} else if (flag & (MEM_PERCPU | MEM_USER)) {
 				/* keep as-is */
 			} else {
-				/* walking unknown pointers yields untrusted pointer */
-				flag = PTR_UNTRUSTED;
+				/* walking unknown pointers yields old deprecated PTR_TO_BTF_ID */
+				clear_trusted_flags(&flag);
 			}
 		} else {
 			/*
@@ -5617,7 +5622,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
 		}
 	} else {
 		/* Old compat. Deprecated */
-		flag &= ~PTR_TRUSTED;
+		clear_trusted_flags(&flag);
 	}
 
 	if (atype == BPF_READ && value_regno >= 0)
-- 
cgit v1.2.3-70-g09d2


From 69f41a787761633b752d71166786eb642bad4913 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 3 Apr 2023 21:50:29 -0700
Subject: selftests/bpf: Add tracing tests for walking skb and req.

Add tracing tests for walking skb->sk and req->sk.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230404045029.82870-9-alexei.starovoitov@gmail.com
---
 .../selftests/bpf/progs/test_sk_storage_tracing.c        | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/test_sk_storage_tracing.c b/tools/testing/selftests/bpf/progs/test_sk_storage_tracing.c
index 6dc1f28fc4b6..02e718f06e0f 100644
--- a/tools/testing/selftests/bpf/progs/test_sk_storage_tracing.c
+++ b/tools/testing/selftests/bpf/progs/test_sk_storage_tracing.c
@@ -92,4 +92,20 @@ int BPF_PROG(inet_csk_accept, struct sock *sk, int flags, int *err, bool kern,
 	return 0;
 }
 
+SEC("tp_btf/tcp_retransmit_synack")
+int BPF_PROG(tcp_retransmit_synack, struct sock* sk, struct request_sock* req)
+{
+	/* load only test */
+	bpf_sk_storage_get(&sk_stg_map, sk, 0, 0);
+	bpf_sk_storage_get(&sk_stg_map, req->sk, 0, 0);
+	return 0;
+}
+
+SEC("tp_btf/tcp_bad_csum")
+int BPF_PROG(tcp_bad_csum, struct sk_buff* skb)
+{
+	bpf_sk_storage_get(&sk_stg_map, skb->sk, 0, 0);
+	return 0;
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From d099f594ad5650e8d8232b5f31f5f90104e65def Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Tue, 4 Apr 2023 00:02:54 +0200
Subject: kallsyms: Disable preemption for find_kallsyms_symbol_value

Artem reported suspicious RCU usage [1]. The reason is that verifier
calls find_kallsyms_symbol_value with preemption enabled which will
trigger suspicious RCU usage warning in rcu_dereference_sched call.

Disabling preemption in find_kallsyms_symbol_value and adding
__find_kallsyms_symbol_value function.

Fixes: 31bf1dbccfb0 ("bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules")
Reported-by: Artem Savkov <asavkov@redhat.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Artem Savkov <asavkov@redhat.com>
Reviewed-by: Zhen Lei <thunder.leizhen@huawei.com>
Link: https://lore.kernel.org/bpf/20230403220254.2191240-1-jolsa@kernel.org

[1] https://lore.kernel.org/bpf/ZBrPMkv8YVRiWwCR@samus.usersys.redhat.com/
---
 kernel/module/kallsyms.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/kernel/module/kallsyms.c b/kernel/module/kallsyms.c
index ab2376a1be88..bdc911dbcde5 100644
--- a/kernel/module/kallsyms.c
+++ b/kernel/module/kallsyms.c
@@ -442,7 +442,7 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
 }
 
 /* Given a module and name of symbol, find and return the symbol's value */
-unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name)
+static unsigned long __find_kallsyms_symbol_value(struct module *mod, const char *name)
 {
 	unsigned int i;
 	struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms);
@@ -466,7 +466,7 @@ static unsigned long __module_kallsyms_lookup_name(const char *name)
 	if (colon) {
 		mod = find_module_all(name, colon - name, false);
 		if (mod)
-			return find_kallsyms_symbol_value(mod, colon + 1);
+			return __find_kallsyms_symbol_value(mod, colon + 1);
 		return 0;
 	}
 
@@ -475,7 +475,7 @@ static unsigned long __module_kallsyms_lookup_name(const char *name)
 
 		if (mod->state == MODULE_STATE_UNFORMED)
 			continue;
-		ret = find_kallsyms_symbol_value(mod, name);
+		ret = __find_kallsyms_symbol_value(mod, name);
 		if (ret)
 			return ret;
 	}
@@ -494,6 +494,16 @@ unsigned long module_kallsyms_lookup_name(const char *name)
 	return ret;
 }
 
+unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name)
+{
+	unsigned long ret;
+
+	preempt_disable();
+	ret = __find_kallsyms_symbol_value(mod, name);
+	preempt_enable();
+	return ret;
+}
+
 int module_kallsyms_on_each_symbol(const char *modname,
 				   int (*fn)(void *, const char *,
 					     struct module *, unsigned long),
-- 
cgit v1.2.3-70-g09d2


From 992e76908e9288d0c725addbbd72b5930e99b2de Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 3 Apr 2023 21:35:46 +0200
Subject: net: phy: meson-gxl: enable edpd tunable support for G12A internal
 PHY

Enable EDPD PHY tunable support for the G12A internal PHY, reusing the
recently added tunable support in the smsc driver.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/8d309575-067c-7321-33cf-6ffac11f7c8d@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/meson-gxl.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/phy/meson-gxl.c b/drivers/net/phy/meson-gxl.c
index 3dea7c752057..bb9b33b6bce2 100644
--- a/drivers/net/phy/meson-gxl.c
+++ b/drivers/net/phy/meson-gxl.c
@@ -210,6 +210,10 @@ static struct phy_driver meson_gxl_phy[] = {
 		.read_status	= lan87xx_read_status,
 		.config_intr	= smsc_phy_config_intr,
 		.handle_interrupt = smsc_phy_handle_interrupt,
+
+		.get_tunable	= smsc_phy_get_tunable,
+		.set_tunable	= smsc_phy_set_tunable,
+
 		.suspend        = genphy_suspend,
 		.resume         = genphy_resume,
 		.read_mmd	= genphy_read_mmd_unsupported,
-- 
cgit v1.2.3-70-g09d2


From 054fbf7ff8143d35ca7d3bb5414bb44ee1574194 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Mon, 3 Apr 2023 17:43:16 +0200
Subject: net: qrtr: correct types of trace event parameters

The arguments passed to the trace events are of type unsigned int,
however the signature of the events used __le32 parameters.

I may be missing the point here, but sparse flagged this and it
does seem incorrect to me.

  net/qrtr/ns.c: note: in included file (through include/trace/trace_events.h, include/trace/define_trace.h, include/trace/events/qrtr.h):
  ./include/trace/events/qrtr.h:11:1: warning: cast to restricted __le32
  ./include/trace/events/qrtr.h:11:1: warning: restricted __le32 degrades to integer
  ./include/trace/events/qrtr.h:11:1: warning: restricted __le32 degrades to integer
  ... (a lot more similar warnings)
  net/qrtr/ns.c:115:47:    expected restricted __le32 [usertype] service
  net/qrtr/ns.c:115:47:    got unsigned int service
  net/qrtr/ns.c:115:61: warning: incorrect type in argument 2 (different base types)
  ... (a lot more similar warnings)

Fixes: dfddb54043f0 ("net: qrtr: Add tracepoint support")
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230402-qrtr-trace-types-v1-1-92ad55008dd3@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/trace/events/qrtr.h | 33 ++++++++++++++++++---------------
 1 file changed, 18 insertions(+), 15 deletions(-)

diff --git a/include/trace/events/qrtr.h b/include/trace/events/qrtr.h
index b1de14c3bb93..441132c67133 100644
--- a/include/trace/events/qrtr.h
+++ b/include/trace/events/qrtr.h
@@ -10,15 +10,16 @@
 
 TRACE_EVENT(qrtr_ns_service_announce_new,
 
-	TP_PROTO(__le32 service, __le32 instance, __le32 node, __le32 port),
+	TP_PROTO(unsigned int service, unsigned int instance,
+		 unsigned int node, unsigned int port),
 
 	TP_ARGS(service, instance, node, port),
 
 	TP_STRUCT__entry(
-		__field(__le32, service)
-		__field(__le32, instance)
-		__field(__le32, node)
-		__field(__le32, port)
+		__field(unsigned int, service)
+		__field(unsigned int, instance)
+		__field(unsigned int, node)
+		__field(unsigned int, port)
 	),
 
 	TP_fast_assign(
@@ -36,15 +37,16 @@ TRACE_EVENT(qrtr_ns_service_announce_new,
 
 TRACE_EVENT(qrtr_ns_service_announce_del,
 
-	TP_PROTO(__le32 service, __le32 instance, __le32 node, __le32 port),
+	TP_PROTO(unsigned int service, unsigned int instance,
+		 unsigned int node, unsigned int port),
 
 	TP_ARGS(service, instance, node, port),
 
 	TP_STRUCT__entry(
-		__field(__le32, service)
-		__field(__le32, instance)
-		__field(__le32, node)
-		__field(__le32, port)
+		__field(unsigned int, service)
+		__field(unsigned int, instance)
+		__field(unsigned int, node)
+		__field(unsigned int, port)
 	),
 
 	TP_fast_assign(
@@ -62,15 +64,16 @@ TRACE_EVENT(qrtr_ns_service_announce_del,
 
 TRACE_EVENT(qrtr_ns_server_add,
 
-	TP_PROTO(__le32 service, __le32 instance, __le32 node, __le32 port),
+	TP_PROTO(unsigned int service, unsigned int instance,
+		 unsigned int node, unsigned int port),
 
 	TP_ARGS(service, instance, node, port),
 
 	TP_STRUCT__entry(
-		__field(__le32, service)
-		__field(__le32, instance)
-		__field(__le32, node)
-		__field(__le32, port)
+		__field(unsigned int, service)
+		__field(unsigned int, instance)
+		__field(unsigned int, node)
+		__field(unsigned int, port)
 	),
 
 	TP_fast_assign(
-- 
cgit v1.2.3-70-g09d2


From 9af87166944b3ff33d1399f7a1924ef0175e96b2 Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Mon, 3 Apr 2023 15:01:51 +0200
Subject: selftests: xsk: Add xskxceiver.h dependency to Makefile

xskxceiver depends on xskxceiver.h so tell make about it.

Signed-off-by: Kal Conley <kal.conley@dectris.com>
Link: https://lore.kernel.org/r/20230403130151.31195-1-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index febd1dae6c88..b5ffdd89b86f 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -612,7 +612,7 @@ $(OUTPUT)/test_verifier: test_verifier.c verifier/tests.h $(BPFOBJ) | $(OUTPUT)
 	$(call msg,BINARY,,$@)
 	$(Q)$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@
 
-$(OUTPUT)/xskxceiver: xskxceiver.c $(OUTPUT)/xsk.o $(OUTPUT)/xsk_xdp_progs.skel.h $(BPFOBJ) | $(OUTPUT)
+$(OUTPUT)/xskxceiver: xskxceiver.c xskxceiver.h $(OUTPUT)/xsk.o $(OUTPUT)/xsk_xdp_progs.skel.h $(BPFOBJ) | $(OUTPUT)
 	$(call msg,BINARY,,$@)
 	$(Q)$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@
 
-- 
cgit v1.2.3-70-g09d2


From 7a2050df244e2c9a4e90882052b7907450ad10ed Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Mon, 3 Apr 2023 16:50:46 +0200
Subject: selftests: xsk: Use correct UMEM size in testapp_invalid_desc

Avoid UMEM_SIZE macro in testapp_invalid_desc which is incorrect when
the frame size is not XSK_UMEM__DEFAULT_FRAME_SIZE. Also remove the
macro since it's no longer being used.

Fixes: 909f0e28207c ("selftests: xsk: Add tests for 2K frame size")
Signed-off-by: Kal Conley <kal.conley@dectris.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/r/20230403145047.33065-2-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/xskxceiver.c | 9 +++++----
 tools/testing/selftests/bpf/xskxceiver.h | 1 -
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index b65e0645b0cd..3956f5db84f3 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -1652,6 +1652,7 @@ static void testapp_single_pkt(struct test_spec *test)
 
 static void testapp_invalid_desc(struct test_spec *test)
 {
+	u64 umem_size = test->ifobj_tx->umem->num_frames * test->ifobj_tx->umem->frame_size;
 	struct pkt pkts[] = {
 		/* Zero packet address allowed */
 		{0, PKT_SIZE, 0, true},
@@ -1662,9 +1663,9 @@ static void testapp_invalid_desc(struct test_spec *test)
 		/* Packet too large */
 		{0x2000, XSK_UMEM__INVALID_FRAME_SIZE, 0, false},
 		/* After umem ends */
-		{UMEM_SIZE, PKT_SIZE, 0, false},
+		{umem_size, PKT_SIZE, 0, false},
 		/* Straddle the end of umem */
-		{UMEM_SIZE - PKT_SIZE / 2, PKT_SIZE, 0, false},
+		{umem_size - PKT_SIZE / 2, PKT_SIZE, 0, false},
 		/* Straddle a page boundrary */
 		{0x3000 - PKT_SIZE / 2, PKT_SIZE, 0, false},
 		/* Straddle a 2K boundrary */
@@ -1682,8 +1683,8 @@ static void testapp_invalid_desc(struct test_spec *test)
 	}
 
 	if (test->ifobj_tx->shared_umem) {
-		pkts[4].addr += UMEM_SIZE;
-		pkts[5].addr += UMEM_SIZE;
+		pkts[4].addr += umem_size;
+		pkts[5].addr += umem_size;
 	}
 
 	pkt_stream_generate_custom(test, pkts, ARRAY_SIZE(pkts));
diff --git a/tools/testing/selftests/bpf/xskxceiver.h b/tools/testing/selftests/bpf/xskxceiver.h
index bdb4efedf3a9..cc24ab72f3ff 100644
--- a/tools/testing/selftests/bpf/xskxceiver.h
+++ b/tools/testing/selftests/bpf/xskxceiver.h
@@ -53,7 +53,6 @@
 #define THREAD_TMOUT 3
 #define DEFAULT_PKT_CNT (4 * 1024)
 #define DEFAULT_UMEM_BUFFERS (DEFAULT_PKT_CNT / 4)
-#define UMEM_SIZE (DEFAULT_UMEM_BUFFERS * XSK_UMEM__DEFAULT_FRAME_SIZE)
 #define RX_FULL_RXQSIZE 32
 #define UMEM_HEADROOM_TEST_SIZE 128
 #define XSK_UMEM__INVALID_FRAME_SIZE (XSK_UMEM__DEFAULT_FRAME_SIZE + 1)
-- 
cgit v1.2.3-70-g09d2


From ccd1b2933f8cbc09a8667992425996f19bf62c15 Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Mon, 3 Apr 2023 16:50:47 +0200
Subject: selftests: xsk: Add test case for packets at end of UMEM

Add test case to testapp_invalid_desc for valid packets at the end of
the UMEM.

Signed-off-by: Kal Conley <kal.conley@dectris.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/r/20230403145047.33065-3-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/xskxceiver.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index 3956f5db84f3..34a1f32fe752 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -1662,6 +1662,8 @@ static void testapp_invalid_desc(struct test_spec *test)
 		{-2, PKT_SIZE, 0, false},
 		/* Packet too large */
 		{0x2000, XSK_UMEM__INVALID_FRAME_SIZE, 0, false},
+		/* Up to end of umem allowed */
+		{umem_size - PKT_SIZE, PKT_SIZE, 0, true},
 		/* After umem ends */
 		{umem_size, PKT_SIZE, 0, false},
 		/* Straddle the end of umem */
@@ -1675,16 +1677,17 @@ static void testapp_invalid_desc(struct test_spec *test)
 
 	if (test->ifobj_tx->umem->unaligned_mode) {
 		/* Crossing a page boundrary allowed */
-		pkts[6].valid = true;
+		pkts[7].valid = true;
 	}
 	if (test->ifobj_tx->umem->frame_size == XSK_UMEM__DEFAULT_FRAME_SIZE / 2) {
 		/* Crossing a 2K frame size boundrary not allowed */
-		pkts[7].valid = false;
+		pkts[8].valid = false;
 	}
 
 	if (test->ifobj_tx->shared_umem) {
 		pkts[4].addr += umem_size;
 		pkts[5].addr += umem_size;
+		pkts[6].addr += umem_size;
 	}
 
 	pkt_stream_generate_custom(test, pkts, ARRAY_SIZE(pkts));
-- 
cgit v1.2.3-70-g09d2


From f2b50f17268390567bc0e95642170d88f336c8f4 Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Wed, 5 Apr 2023 10:29:04 +0200
Subject: selftests: xsk: Disable IPv6 on VETH1

This change fixes flakiness in the BIDIRECTIONAL test:

    # [is_pkt_valid] expected length [60], got length [90]
    not ok 1 FAIL: SKB BUSY-POLL BIDIRECTIONAL

When IPv6 is enabled, the interface will periodically send MLDv1 and
MLDv2 packets. These packets can cause the BIDIRECTIONAL test to fail
since it uses VETH0 for RX.

For other tests, this was not a problem since they only receive on VETH1
and IPv6 was already disabled on VETH0.

Fixes: a89052572ebb ("selftests/bpf: Xsk selftests framework")
Signed-off-by: Kal Conley <kal.conley@dectris.com>
Link: https://lore.kernel.org/r/20230405082905.6303-1-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/test_xsk.sh | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/selftests/bpf/test_xsk.sh b/tools/testing/selftests/bpf/test_xsk.sh
index b077cf58f825..377fb157a57c 100755
--- a/tools/testing/selftests/bpf/test_xsk.sh
+++ b/tools/testing/selftests/bpf/test_xsk.sh
@@ -116,6 +116,7 @@ setup_vethPairs() {
 	ip link add ${VETH0} numtxqueues 4 numrxqueues 4 type veth peer name ${VETH1} numtxqueues 4 numrxqueues 4
 	if [ -f /proc/net/if_inet6 ]; then
 		echo 1 > /proc/sys/net/ipv6/conf/${VETH0}/disable_ipv6
+		echo 1 > /proc/sys/net/ipv6/conf/${VETH1}/disable_ipv6
 	fi
 	if [[ $verbose -eq 1 ]]; then
 	        echo "setting up ${VETH1}"
-- 
cgit v1.2.3-70-g09d2


From 68e7322142f5e731af222892d384d311835db0f1 Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Mon, 3 Apr 2023 14:03:59 +0200
Subject: selftests: xsk: Deflakify STATS_RX_DROPPED test

Fix flaky STATS_RX_DROPPED test. The receiver calls getsockopt after
receiving the last (valid) packet which is not the final packet sent in
the test (valid and invalid packets are sent in alternating fashion with
the final packet being invalid). Since the last packet may or may not
have been dropped already, both outcomes must be allowed.

This issue could also be fixed by making sure the last packet sent is
valid. This alternative is left as an exercise to the reader (or the
benevolent maintainers of this file).

This problem was quite visible on certain setups. On one machine this
failure was observed 50% of the time.

Also, remove a redundant assignment of pkt_stream->nb_pkts. This field
is already initialized by __pkt_stream_alloc.

Fixes: 27e934bec35b ("selftests: xsk: make stat tests not spin on getsockopt")
Signed-off-by: Kal Conley <kal.conley@dectris.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/r/20230403120400.31018-1-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/xskxceiver.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index 34a1f32fe752..1a4bdd5aa78c 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -633,7 +633,6 @@ static struct pkt_stream *pkt_stream_generate(struct xsk_umem_info *umem, u32 nb
 	if (!pkt_stream)
 		exit_with_error(ENOMEM);
 
-	pkt_stream->nb_pkts = nb_pkts;
 	for (i = 0; i < nb_pkts; i++) {
 		pkt_set(umem, &pkt_stream->pkts[i], (i % umem->num_frames) * umem->frame_size,
 			pkt_len);
@@ -1141,7 +1140,14 @@ static int validate_rx_dropped(struct ifobject *ifobject)
 	if (err)
 		return TEST_FAILURE;
 
-	if (stats.rx_dropped == ifobject->pkt_stream->nb_pkts / 2)
+	/* The receiver calls getsockopt after receiving the last (valid)
+	 * packet which is not the final packet sent in this test (valid and
+	 * invalid packets are sent in alternating fashion with the final
+	 * packet being invalid). Since the last packet may or may not have
+	 * been dropped already, both outcomes must be allowed.
+	 */
+	if (stats.rx_dropped == ifobject->pkt_stream->nb_pkts / 2 ||
+	    stats.rx_dropped == ifobject->pkt_stream->nb_pkts / 2 - 1)
 		return TEST_PASS;
 
 	return TEST_FAILURE;
-- 
cgit v1.2.3-70-g09d2


From 5af607a861d43ffff830fc1890033e579ec44799 Mon Sep 17 00:00:00 2001
From: YiFei Zhu <zhuyifei@google.com>
Date: Wed, 5 Apr 2023 19:33:54 +0000
Subject: selftests/bpf: Wait for receive in cg_storage_multi test

In some cases the loopback latency might be large enough, causing
the assertion on invocations to be run before ingress prog getting
executed. The assertion would fail and the test would flake.

This can be reliably reproduced by arbitrarily increasing the
loopback latency (thanks to [1]):
  tc qdisc add dev lo root handle 1: htb default 12
  tc class add dev lo parent 1:1 classid 1:12 htb rate 20kbps ceil 20kbps
  tc qdisc add dev lo parent 1:12 netem delay 100ms

Fix this by waiting on the receive end, instead of instantly
returning to the assert. The call to read() will wait for the
default SO_RCVTIMEO timeout of 3 seconds provided by
start_server().

[1] https://gist.github.com/kstevens715/4598301

Reported-by: Martin KaFai Lau <martin.lau@linux.dev>
Link: https://lore.kernel.org/bpf/9c5c8b7e-1d89-a3af-5400-14fde81f4429@linux.dev/
Fixes: 3573f384014f ("selftests/bpf: Test CGROUP_STORAGE behavior on shared egress + ingress")
Acked-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: YiFei Zhu <zhuyifei@google.com>
Link: https://lore.kernel.org/r/20230405193354.1956209-1-zhuyifei@google.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c
index 621c57222191..63ee892bc757 100644
--- a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c
+++ b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c
@@ -56,8 +56,9 @@ static bool assert_storage_noexist(struct bpf_map *map, const void *key)
 
 static bool connect_send(const char *cgroup_path)
 {
-	bool res = true;
 	int server_fd = -1, client_fd = -1;
+	char message[] = "message";
+	bool res = true;
 
 	if (join_cgroup(cgroup_path))
 		goto out_clean;
@@ -70,7 +71,10 @@ static bool connect_send(const char *cgroup_path)
 	if (client_fd < 0)
 		goto out_clean;
 
-	if (send(client_fd, "message", strlen("message"), 0) < 0)
+	if (send(client_fd, &message, sizeof(message), 0) < 0)
+		goto out_clean;
+
+	if (read(server_fd, &message, sizeof(message)) < 0)
 		goto out_clean;
 
 	res = false;
-- 
cgit v1.2.3-70-g09d2


From c8f1f2e946757e5130882d23c6beded32d87ed0c Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Tue, 4 Apr 2023 15:42:13 -0500
Subject: dt-bindings: net: ethernet-switch: Make "#address-cells/#size-cells"
 required

The schema doesn't allow for a single (unaddressed) ethernet port node
nor does a single port switch make much sense. So if there's always
multiple child nodes, "#address-cells" and "#size-cells" should be
required.

Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230404204213.635773-1-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/ethernet-switch.yaml   | 4 ++++
 Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml | 4 +++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/ethernet-switch.yaml b/Documentation/devicetree/bindings/net/ethernet-switch.yaml
index a04f8ef744aa..2ceccce6cbd7 100644
--- a/Documentation/devicetree/bindings/net/ethernet-switch.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-switch.yaml
@@ -40,6 +40,10 @@ patternProperties:
         type: object
         description: Ethernet switch ports
 
+    required:
+      - "#address-cells"
+      - "#size-cells"
+
 oneOf:
   - required:
       - ports
diff --git a/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml b/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml
index 144001ff840c..164704338ef0 100644
--- a/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml
+++ b/Documentation/devicetree/bindings/net/qcom,ipq8064-mdio.yaml
@@ -53,7 +53,9 @@ examples:
             reg = <0x10>;
 
             ports {
-              /* ... */
+                #address-cells = <1>;
+                #size-cells = <0>;
+                /* ... */
             };
         };
     };
-- 
cgit v1.2.3-70-g09d2


From f037897669051d53551a2b198b2519356d59d491 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Tue, 4 Apr 2023 15:41:52 -0500
Subject: dt-bindings: net: dsa: brcm,sf2: Drop unneeded
 "#address-cells/#size-cells"

There's no need for "#address-cells/#size-cells" in the brcm,sf2 node as
no immediate child nodes have an address. What was probably intended was
to put them in the 'ports' node, but that's not necessary as that is
covered by ethernet-switch.yaml via dsa.yaml.

Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230404204152.635400-1-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml b/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml
index 37bf33bd4670..c745407f2f68 100644
--- a/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/brcm,sf2.yaml
@@ -76,12 +76,6 @@ properties:
       supports reporting the number of packets in-flight in a switch queue
     type: boolean
 
-  "#address-cells":
-    const: 1
-
-  "#size-cells":
-    const: 0
-
   ports:
     type: object
 
@@ -99,8 +93,6 @@ properties:
 required:
   - reg
   - interrupts
-  - "#address-cells"
-  - "#size-cells"
 
 allOf:
   - $ref: dsa.yaml#
@@ -145,8 +137,6 @@ examples:
   - |
     switch@f0b00000 {
             compatible = "brcm,bcm7445-switch-v4.0";
-            #address-cells = <1>;
-            #size-cells = <0>;
             reg = <0xf0b00000 0x40000>,
                   <0xf0b40000 0x110>,
                   <0xf0b40340 0x30>,
-- 
cgit v1.2.3-70-g09d2


From 8b0f256530d97f2a2310c4f8336ea2c477c8e6c4 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Tue, 4 Apr 2023 17:34:49 -0300
Subject: net/sched: sch_mqprio: use netlink payload helpers

For the sake of readability, use the netlink payload helpers from
the 'nla_get_*()' family to parse the attributes.

tdc results:
1..5
ok 1 9903 - Add mqprio Qdisc to multi-queue device (8 queues)
ok 2 453a - Delete nonexistent mqprio Qdisc
ok 3 5292 - Delete mqprio Qdisc twice
ok 4 45a9 - Add mqprio Qdisc to single-queue device
ok 5 2ba9 - Show mqprio class

Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230404203449.1627033-1-pctammela@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/sch_mqprio.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 48ed87b91086..fdd6a6575a54 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -178,12 +178,12 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 
 	if (tb[TCA_MQPRIO_MODE]) {
 		priv->flags |= TC_MQPRIO_F_MODE;
-		priv->mode = *(u16 *)nla_data(tb[TCA_MQPRIO_MODE]);
+		priv->mode = nla_get_u16(tb[TCA_MQPRIO_MODE]);
 	}
 
 	if (tb[TCA_MQPRIO_SHAPER]) {
 		priv->flags |= TC_MQPRIO_F_SHAPER;
-		priv->shaper = *(u16 *)nla_data(tb[TCA_MQPRIO_SHAPER]);
+		priv->shaper = nla_get_u16(tb[TCA_MQPRIO_SHAPER]);
 	}
 
 	if (tb[TCA_MQPRIO_MIN_RATE64]) {
@@ -196,7 +196,7 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 				return -EINVAL;
 			if (i >= qopt->num_tc)
 				break;
-			priv->min_rate[i] = *(u64 *)nla_data(attr);
+			priv->min_rate[i] = nla_get_u64(attr);
 			i++;
 		}
 		priv->flags |= TC_MQPRIO_F_MIN_RATE;
@@ -212,7 +212,7 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 				return -EINVAL;
 			if (i >= qopt->num_tc)
 				break;
-			priv->max_rate[i] = *(u64 *)nla_data(attr);
+			priv->max_rate[i] = nla_get_u64(attr);
 			i++;
 		}
 		priv->flags |= TC_MQPRIO_F_MAX_RATE;
-- 
cgit v1.2.3-70-g09d2


From 0840c9f7d80b1baf959d0ddcedc97c9f383ded1c Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Mon, 13 Mar 2023 15:38:52 +0200
Subject: net/mlx5e: Set default can_offload action

Many parsers of tc actions just return true on their can_offload()
implementation, without checking the input flow/action.
Set the default can_offload action to true (allow), and avoid
having many can_offload implementations that do just that.

This patch doesn't change any functionality.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c     | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c       | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/pedit.c      | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ptype.c      | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c       | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/tun.c        | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c       | 10 ----------
 .../net/ethernet/mellanox/mlx5/core/en/tc/act/vlan_mangle.c    | 10 ----------
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c                |  2 +-
 9 files changed, 1 insertion(+), 81 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c
index a278f52d52b0..9db1b5307a8d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c
@@ -4,15 +4,6 @@
 #include "act.h"
 #include "en/tc_priv.h"
 
-static bool
-tc_act_can_offload_accept(struct mlx5e_tc_act_parse_state *parse_state,
-			  const struct flow_action_entry *act,
-			  int act_index,
-			  struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_accept(struct mlx5e_tc_act_parse_state *parse_state,
 		    const struct flow_action_entry *act,
@@ -26,7 +17,6 @@ tc_act_parse_accept(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_accept = {
-	.can_offload = tc_act_can_offload_accept,
 	.parse_action = tc_act_parse_accept,
 	.is_terminating_action = true,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c
index 7d16aeabb119..5dc81715d625 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c
@@ -4,15 +4,6 @@
 #include "act.h"
 #include "en/tc_priv.h"
 
-static bool
-tc_act_can_offload_drop(struct mlx5e_tc_act_parse_state *parse_state,
-			const struct flow_action_entry *act,
-			int act_index,
-			struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_drop(struct mlx5e_tc_act_parse_state *parse_state,
 		  const struct flow_action_entry *act,
@@ -25,7 +16,6 @@ tc_act_parse_drop(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_drop = {
-	.can_offload = tc_act_can_offload_drop,
 	.parse_action = tc_act_parse_drop,
 	.is_terminating_action = true,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/pedit.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/pedit.c
index 47597c524e59..3b272bbf4c53 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/pedit.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/pedit.c
@@ -78,15 +78,6 @@ out_err:
 	return err;
 }
 
-static bool
-tc_act_can_offload_pedit(struct mlx5e_tc_act_parse_state *parse_state,
-			 const struct flow_action_entry *act,
-			 int act_index,
-			 struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_pedit(struct mlx5e_tc_act_parse_state *parse_state,
 		   const struct flow_action_entry *act,
@@ -114,6 +105,5 @@ tc_act_parse_pedit(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_pedit = {
-	.can_offload = tc_act_can_offload_pedit,
 	.parse_action = tc_act_parse_pedit,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ptype.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ptype.c
index 6454b031ff7a..80b4bc64380a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ptype.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ptype.c
@@ -4,15 +4,6 @@
 #include "act.h"
 #include "en/tc_priv.h"
 
-static bool
-tc_act_can_offload_ptype(struct mlx5e_tc_act_parse_state *parse_state,
-			 const struct flow_action_entry *act,
-			 int act_index,
-			 struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_ptype(struct mlx5e_tc_act_parse_state *parse_state,
 		   const struct flow_action_entry *act,
@@ -31,6 +22,5 @@ tc_act_parse_ptype(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_ptype = {
-	.can_offload = tc_act_can_offload_ptype,
 	.parse_action = tc_act_parse_ptype,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c
index 915ce201aeb2..1b78bd9c106a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c
@@ -5,15 +5,6 @@
 #include "en/tc_priv.h"
 #include "eswitch.h"
 
-static bool
-tc_act_can_offload_trap(struct mlx5e_tc_act_parse_state *parse_state,
-			const struct flow_action_entry *act,
-			int act_index,
-			struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_trap(struct mlx5e_tc_act_parse_state *parse_state,
 		  const struct flow_action_entry *act,
@@ -27,6 +18,5 @@ tc_act_parse_trap(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_trap = {
-	.can_offload = tc_act_can_offload_trap,
 	.parse_action = tc_act_parse_trap,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/tun.c
index b4fa2de9711d..f1cae21c2c37 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/tun.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/tun.c
@@ -32,15 +32,6 @@ tc_act_parse_tun_encap(struct mlx5e_tc_act_parse_state *parse_state,
 	return 0;
 }
 
-static bool
-tc_act_can_offload_tun_decap(struct mlx5e_tc_act_parse_state *parse_state,
-			     const struct flow_action_entry *act,
-			     int act_index,
-			     struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_tun_decap(struct mlx5e_tc_act_parse_state *parse_state,
 		       const struct flow_action_entry *act,
@@ -58,6 +49,5 @@ struct mlx5e_tc_act mlx5e_tc_act_tun_encap = {
 };
 
 struct mlx5e_tc_act mlx5e_tc_act_tun_decap = {
-	.can_offload = tc_act_can_offload_tun_decap,
 	.parse_action = tc_act_parse_tun_decap,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c
index 2e0d88b513aa..c8a3eaf189f6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan.c
@@ -141,15 +141,6 @@ mlx5e_tc_act_vlan_add_pop_action(struct mlx5e_priv *priv,
 	return err;
 }
 
-static bool
-tc_act_can_offload_vlan(struct mlx5e_tc_act_parse_state *parse_state,
-			const struct flow_action_entry *act,
-			int act_index,
-			struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_vlan(struct mlx5e_tc_act_parse_state *parse_state,
 		  const struct flow_action_entry *act,
@@ -205,7 +196,6 @@ tc_act_post_parse_vlan(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_vlan = {
-	.can_offload = tc_act_can_offload_vlan,
 	.parse_action = tc_act_parse_vlan,
 	.post_parse = tc_act_post_parse_vlan,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan_mangle.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan_mangle.c
index 9a8a1a6bd99e..310b99230760 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan_mangle.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/vlan_mangle.c
@@ -50,15 +50,6 @@ mlx5e_tc_act_vlan_add_rewrite_action(struct mlx5e_priv *priv, int namespace,
 	return err;
 }
 
-static bool
-tc_act_can_offload_vlan_mangle(struct mlx5e_tc_act_parse_state *parse_state,
-			       const struct flow_action_entry *act,
-			       int act_index,
-			       struct mlx5_flow_attr *attr)
-{
-	return true;
-}
-
 static int
 tc_act_parse_vlan_mangle(struct mlx5e_tc_act_parse_state *parse_state,
 			 const struct flow_action_entry *act,
@@ -81,6 +72,5 @@ tc_act_parse_vlan_mangle(struct mlx5e_tc_act_parse_state *parse_state,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_vlan_mangle = {
-	.can_offload = tc_act_can_offload_vlan_mangle,
 	.parse_action = tc_act_parse_vlan_mangle,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 083ce31f95b6..4140155e3487 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -4096,7 +4096,7 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 			goto out_free;
 		}
 
-		if (!tc_act->can_offload(parse_state, act, i, attr)) {
+		if (tc_act->can_offload && !tc_act->can_offload(parse_state, act, i, attr)) {
 			err = -EOPNOTSUPP;
 			goto out_free;
 		}
-- 
cgit v1.2.3-70-g09d2


From 7195d9a0c8df0ab78c9d7a587809d16b00432426 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Thu, 1 Dec 2022 12:04:58 +0200
Subject: net/mlx5e: TC, Remove unused vf_tun variable

vf_tun is being assigned but never being used so remove it.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 4140155e3487..a4cf1818e8fe 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1785,8 +1785,7 @@ out:
 static void
 clean_encap_dests(struct mlx5e_priv *priv,
 		  struct mlx5e_tc_flow *flow,
-		  struct mlx5_flow_attr *attr,
-		  bool *vf_tun)
+		  struct mlx5_flow_attr *attr)
 {
 	struct mlx5_esw_flow_attr *esw_attr;
 	int out_index;
@@ -1795,17 +1794,11 @@ clean_encap_dests(struct mlx5e_priv *priv,
 		return;
 
 	esw_attr = attr->esw_attr;
-	*vf_tun = false;
 
 	for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) {
 		if (!(esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP))
 			continue;
 
-		if (esw_attr->dests[out_index].flags &
-		    MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE &&
-		    !esw_attr->dest_int_port)
-			*vf_tun = true;
-
 		mlx5e_detach_encap(priv, flow, attr, out_index);
 		kfree(attr->parse_attr->tun_info[out_index]);
 	}
@@ -2039,7 +2032,6 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv,
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5_flow_attr *attr = flow->attr;
 	struct mlx5_esw_flow_attr *esw_attr;
-	bool vf_tun;
 
 	esw_attr = attr->esw_attr;
 	mlx5e_put_flow_tunnel_id(flow);
@@ -2061,7 +2053,7 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv,
 	if (flow->decap_route)
 		mlx5e_detach_decap_route(priv, flow);
 
-	clean_encap_dests(priv, flow, attr, &vf_tun);
+	clean_encap_dests(priv, flow, attr);
 
 	mlx5_tc_ct_match_del(get_ct_priv(priv), &flow->attr->ct_attr);
 
@@ -4445,7 +4437,6 @@ static void
 mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr)
 {
 	struct mlx5_core_dev *counter_dev = get_flow_counter_dev(flow);
-	bool vf_tun;
 
 	if (!attr)
 		return;
@@ -4453,7 +4444,7 @@ mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr)
 	if (attr->post_act_handle)
 		mlx5e_tc_post_act_del(get_post_action(flow->priv), attr->post_act_handle);
 
-	clean_encap_dests(flow->priv, flow, attr, &vf_tun);
+	clean_encap_dests(flow->priv, flow, attr);
 
 	if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT)
 		mlx5_fc_destroy(counter_dev, attr->counter);
-- 
cgit v1.2.3-70-g09d2


From a830ec485e8368a29e328d08e2eb28750bbc483f Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Wed, 25 Jan 2023 11:54:22 +0200
Subject: net/mlx5e: TC, Move main flow attribute cleanup to helper func

Actions that can be setup per flow attribute (so per split rule)
are cleaned up from mlx5_free_flow_attr(), mlx5e_tc_del_fdb_flow(),
and free_flow_post_acts().

Remove the duplication by re-using the helper function for
the main flow attribute and split rules attributes.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 27 +++++++++----------------
 1 file changed, 9 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index a4cf1818e8fe..cafab7634fb8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -177,7 +177,8 @@ static struct lock_class_key tc_ht_wq_key;
 
 static void mlx5e_put_flow_tunnel_id(struct mlx5e_tc_flow *flow);
 static void free_flow_post_acts(struct mlx5e_tc_flow *flow);
-static void mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr);
+static void mlx5_free_flow_attr_actions(struct mlx5e_tc_flow *flow,
+					struct mlx5_flow_attr *attr);
 
 void
 mlx5e_tc_match_to_reg_match(struct mlx5_flow_spec *spec,
@@ -2021,7 +2022,7 @@ static void free_branch_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *
 	if (!attr)
 		return;
 
-	mlx5_free_flow_attr(flow, attr);
+	mlx5_free_flow_attr_actions(flow, attr);
 	kvfree(attr->parse_attr);
 	kfree(attr);
 }
@@ -2053,18 +2054,8 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv,
 	if (flow->decap_route)
 		mlx5e_detach_decap_route(priv, flow);
 
-	clean_encap_dests(priv, flow, attr);
-
 	mlx5_tc_ct_match_del(get_ct_priv(priv), &flow->attr->ct_attr);
 
-	if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) {
-		mlx5e_mod_hdr_dealloc(&attr->parse_attr->mod_hdr_acts);
-		mlx5e_tc_detach_mod_hdr(priv, flow, attr);
-	}
-
-	if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT)
-		mlx5_fc_destroy(esw_attr->counter_dev, attr->counter);
-
 	if (esw_attr->int_port)
 		mlx5e_tc_int_port_put(mlx5e_get_int_port_priv(priv), esw_attr->int_port);
 
@@ -2077,8 +2068,7 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv,
 	mlx5e_tc_act_stats_del_flow(get_act_stats_handle(priv), flow);
 
 	free_flow_post_acts(flow);
-	free_branch_attr(flow, attr->branch_true);
-	free_branch_attr(flow, attr->branch_false);
+	mlx5_free_flow_attr_actions(flow, attr);
 
 	kvfree(attr->esw_attr->rx_tun_attr);
 	kvfree(attr->parse_attr);
@@ -3797,9 +3787,7 @@ free_flow_post_acts(struct mlx5e_tc_flow *flow)
 		if (list_is_last(&attr->list, &flow->attrs))
 			break;
 
-		mlx5_free_flow_attr(flow, attr);
-		free_branch_attr(flow, attr->branch_true);
-		free_branch_attr(flow, attr->branch_false);
+		mlx5_free_flow_attr_actions(flow, attr);
 
 		list_del(&attr->list);
 		kvfree(attr->parse_attr);
@@ -4434,7 +4422,7 @@ mlx5_alloc_flow_attr(enum mlx5_flow_namespace_type type)
 }
 
 static void
-mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr)
+mlx5_free_flow_attr_actions(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr)
 {
 	struct mlx5_core_dev *counter_dev = get_flow_counter_dev(flow);
 
@@ -4453,6 +4441,9 @@ mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr)
 		mlx5e_mod_hdr_dealloc(&attr->parse_attr->mod_hdr_acts);
 		mlx5e_tc_detach_mod_hdr(flow->priv, flow, attr);
 	}
+
+	free_branch_attr(flow, attr->branch_true);
+	free_branch_attr(flow, attr->branch_false);
 }
 
 static int
-- 
cgit v1.2.3-70-g09d2


From 13aca17b450e87a8de4e4a3b3ad454efbc576740 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Sun, 12 Feb 2023 19:11:07 +0200
Subject: net/mlx5e: CT: Use per action stats

CT action can miss in a middle of an action list, use
per action stats to correctly report stats for missed
packets.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en/tc/act/act.h   |  2 ++
 .../net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c    |  9 ++++++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c       | 19 +++++++++++++++++--
 3 files changed, 27 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
index 8346557eeaf6..cdcddf6e1b08 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
@@ -56,6 +56,8 @@ struct mlx5e_tc_act {
 				   const struct flow_action_entry *act,
 				   struct mlx5_flow_attr *attr);
 
+	bool (*is_missable)(const struct flow_action_entry *act);
+
 	int (*offload_action)(struct mlx5e_priv *priv,
 			      struct flow_offload_action *fl_act,
 			      struct flow_action_entry *act);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
index a829c94289c1..fce1c0fd2453 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
@@ -95,10 +95,17 @@ tc_act_is_multi_table_act_ct(struct mlx5e_priv *priv,
 	return true;
 }
 
+static bool
+tc_act_is_missable_ct(const struct flow_action_entry *act)
+{
+	return !(act->ct.action & TCA_CT_ACT_CLEAR);
+}
+
 struct mlx5e_tc_act mlx5e_tc_act_ct = {
 	.can_offload = tc_act_can_offload_ct,
 	.parse_action = tc_act_parse_ct,
-	.is_multi_table_act = tc_act_is_multi_table_act_ct,
 	.post_parse = tc_act_post_parse_ct,
+	.is_multi_table_act = tc_act_is_multi_table_act_ct,
+	.is_missable = tc_act_is_missable_ct,
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index cafab7634fb8..32b50d685312 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -4052,7 +4052,9 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 	enum mlx5_flow_namespace_type ns_type;
 	struct mlx5e_priv *priv = flow->priv;
 	struct flow_action_entry *act, **_act;
+	struct mlx5_flow_attr *prev_attr;
 	struct mlx5e_tc_act *tc_act;
+	bool is_missable;
 	int err, i;
 
 	flow_action_reorder.num_entries = flow_action->num_entries;
@@ -4069,6 +4071,9 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 	flow_action_for_each(i, _act, &flow_action_reorder) {
 		jump_state.jump_target = false;
 		act = *_act;
+		is_missable = false;
+		prev_attr = attr;
+
 		tc_act = mlx5e_tc_act_get(act->id, ns_type);
 		if (!tc_act) {
 			NL_SET_ERR_MSG_MOD(extack, "Not implemented offload action");
@@ -4092,14 +4097,14 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 			goto out_free;
 
 		parse_state->actions |= attr->action;
-		if (!tc_act->stats_action)
-			attr->tc_act_cookies[attr->tc_act_cookies_count++] = act->cookie;
 
 		/* Split attr for multi table act if not the last act. */
 		if (jump_state.jump_target ||
 		    (tc_act->is_multi_table_act &&
 		    tc_act->is_multi_table_act(priv, act, attr) &&
 		    i < flow_action_reorder.num_entries - 1)) {
+			is_missable = tc_act->is_missable ? tc_act->is_missable(act) : false;
+
 			err = mlx5e_tc_act_post_parse(parse_state, flow_action, attr, ns_type);
 			if (err)
 				goto out_free;
@@ -4112,6 +4117,16 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 
 			list_add(&attr->list, &flow->attrs);
 		}
+
+		if (is_missable) {
+			/* Add counter to prev, and assign act to new (next) attr */
+			prev_attr->action |= MLX5_FLOW_CONTEXT_ACTION_COUNT;
+			flow_flag_set(flow, USE_ACT_STATS);
+
+			attr->tc_act_cookies[attr->tc_act_cookies_count++] = act->cookie;
+		} else if (!tc_act->stats_action) {
+			prev_attr->tc_act_cookies[prev_attr->tc_act_cookies_count++] = act->cookie;
+		}
 	}
 
 	kfree(flow_action_reorder.entries);
-- 
cgit v1.2.3-70-g09d2


From 67efaf45930df662111acf7c706d545c83f83999 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Sun, 4 Dec 2022 16:53:56 +0200
Subject: net/mlx5e: TC, Remove CT action reordering

CT action reordering was done as a workaround when CT misses
used to restore the relevant filter's tc chain and continuing sw processing
from that chain. As such, there was a need to reorder CT action to be before
any packet modifying actions (e.g mac rewrite).

Currently (after patch "net/mlx5e: TC, Set CT miss to the specific ct
action instance"), CT misses continues from the relevant ct action in
software, and so reordering isn't needed anymore.

Remove the reordering.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/tc/act/act.c    | 20 --------------
 .../ethernet/mellanox/mlx5/core/en/tc/act/act.h    |  4 ---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 32 ++++++----------------
 3 files changed, 9 insertions(+), 47 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c
index eba0c8698926..fc923a99b6a4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c
@@ -82,26 +82,6 @@ mlx5e_tc_act_init_parse_state(struct mlx5e_tc_act_parse_state *parse_state,
 	parse_state->flow_action = flow_action;
 }
 
-void
-mlx5e_tc_act_reorder_flow_actions(struct flow_action *flow_action,
-				  struct mlx5e_tc_flow_action *flow_action_reorder)
-{
-	struct flow_action_entry *act;
-	int i, j = 0;
-
-	flow_action_for_each(i, act, flow_action) {
-		/* Add CT action to be first. */
-		if (act->id == FLOW_ACTION_CT)
-			flow_action_reorder->entries[j++] = act;
-	}
-
-	flow_action_for_each(i, act, flow_action) {
-		if (act->id == FLOW_ACTION_CT)
-			continue;
-		flow_action_reorder->entries[j++] = act;
-	}
-}
-
 int
 mlx5e_tc_act_post_parse(struct mlx5e_tc_act_parse_state *parse_state,
 			struct flow_action *flow_action,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
index cdcddf6e1b08..d7615e329e6d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
@@ -112,10 +112,6 @@ mlx5e_tc_act_init_parse_state(struct mlx5e_tc_act_parse_state *parse_state,
 			      struct flow_action *flow_action,
 			      struct netlink_ext_ack *extack);
 
-void
-mlx5e_tc_act_reorder_flow_actions(struct flow_action *flow_action,
-				  struct mlx5e_tc_flow_action *flow_action_reorder);
-
 int
 mlx5e_tc_act_post_parse(struct mlx5e_tc_act_parse_state *parse_state,
 			struct flow_action *flow_action,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 32b50d685312..15deb1df4df3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -4045,32 +4045,22 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 		 struct flow_action *flow_action)
 {
 	struct netlink_ext_ack *extack = parse_state->extack;
-	struct mlx5e_tc_flow_action flow_action_reorder;
 	struct mlx5e_tc_flow *flow = parse_state->flow;
 	struct mlx5e_tc_jump_state jump_state = {};
 	struct mlx5_flow_attr *attr = flow->attr;
 	enum mlx5_flow_namespace_type ns_type;
 	struct mlx5e_priv *priv = flow->priv;
-	struct flow_action_entry *act, **_act;
 	struct mlx5_flow_attr *prev_attr;
+	struct flow_action_entry *act;
 	struct mlx5e_tc_act *tc_act;
 	bool is_missable;
 	int err, i;
 
-	flow_action_reorder.num_entries = flow_action->num_entries;
-	flow_action_reorder.entries = kcalloc(flow_action->num_entries,
-					      sizeof(flow_action), GFP_KERNEL);
-	if (!flow_action_reorder.entries)
-		return -ENOMEM;
-
-	mlx5e_tc_act_reorder_flow_actions(flow_action, &flow_action_reorder);
-
 	ns_type = mlx5e_get_flow_namespace(flow);
 	list_add(&attr->list, &flow->attrs);
 
-	flow_action_for_each(i, _act, &flow_action_reorder) {
+	flow_action_for_each(i, act, flow_action) {
 		jump_state.jump_target = false;
-		act = *_act;
 		is_missable = false;
 		prev_attr = attr;
 
@@ -4078,23 +4068,23 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 		if (!tc_act) {
 			NL_SET_ERR_MSG_MOD(extack, "Not implemented offload action");
 			err = -EOPNOTSUPP;
-			goto out_free;
+			goto out_free_post_acts;
 		}
 
 		if (tc_act->can_offload && !tc_act->can_offload(parse_state, act, i, attr)) {
 			err = -EOPNOTSUPP;
-			goto out_free;
+			goto out_free_post_acts;
 		}
 
 		err = tc_act->parse_action(parse_state, act, priv, attr);
 		if (err)
-			goto out_free;
+			goto out_free_post_acts;
 
 		dec_jump_count(act, tc_act, attr, priv, &jump_state);
 
 		err = parse_branch_ctrl(act, tc_act, flow, attr, &jump_state, extack);
 		if (err)
-			goto out_free;
+			goto out_free_post_acts;
 
 		parse_state->actions |= attr->action;
 
@@ -4102,17 +4092,17 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 		if (jump_state.jump_target ||
 		    (tc_act->is_multi_table_act &&
 		    tc_act->is_multi_table_act(priv, act, attr) &&
-		    i < flow_action_reorder.num_entries - 1)) {
+		    i < flow_action->num_entries - 1)) {
 			is_missable = tc_act->is_missable ? tc_act->is_missable(act) : false;
 
 			err = mlx5e_tc_act_post_parse(parse_state, flow_action, attr, ns_type);
 			if (err)
-				goto out_free;
+				goto out_free_post_acts;
 
 			attr = mlx5e_clone_flow_attr_for_post_act(flow->attr, ns_type);
 			if (!attr) {
 				err = -ENOMEM;
-				goto out_free;
+				goto out_free_post_acts;
 			}
 
 			list_add(&attr->list, &flow->attrs);
@@ -4129,8 +4119,6 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 		}
 	}
 
-	kfree(flow_action_reorder.entries);
-
 	err = mlx5e_tc_act_post_parse(parse_state, flow_action, attr, ns_type);
 	if (err)
 		goto out_free_post_acts;
@@ -4141,8 +4129,6 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state,
 
 	return 0;
 
-out_free:
-	kfree(flow_action_reorder.entries);
 out_free_post_acts:
 	free_flow_post_acts(flow);
 
-- 
cgit v1.2.3-70-g09d2


From 08fe94ec5f77136177197c297a794ea00c3677d7 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Wed, 25 Jan 2023 16:41:05 +0200
Subject: net/mlx5e: TC, Remove special handling of CT action

CT action has special treating as a per-flow action since
it was assumed to be singular and reordered to be first on
the action list.

This isn't the case anymore, and can be converted to just a
FWD to pre_ct + MODIFY_HEAD, and handled per post_act rule.

Remove special handling of CT action, and offload it while
post parsing each ct attribute.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/tc/act/act.h    |   1 -
 .../net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c |  39 +----
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 169 +++++----------------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h |  31 +---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |  31 +---
 5 files changed, 58 insertions(+), 213 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
index d7615e329e6d..033afd44fc4f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
@@ -18,7 +18,6 @@ struct mlx5e_tc_act_parse_state {
 	struct netlink_ext_ack *extack;
 	u32 actions;
 	bool ct;
-	bool ct_clear;
 	bool encap;
 	bool decap;
 	bool mpls_push;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
index fce1c0fd2453..36bfed07d400 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
@@ -28,30 +28,16 @@ tc_act_parse_ct(struct mlx5e_tc_act_parse_state *parse_state,
 		struct mlx5e_priv *priv,
 		struct mlx5_flow_attr *attr)
 {
-	bool clear_action = act->ct.action & TCA_CT_ACT_CLEAR;
 	int err;
 
-	/* It's redundant to do ct clear more than once. */
-	if (clear_action && parse_state->ct_clear)
-		return 0;
-
-	err = mlx5_tc_ct_parse_action(parse_state->ct_priv, attr,
-				      &attr->parse_attr->mod_hdr_acts,
-				      act, parse_state->extack);
+	err = mlx5_tc_ct_parse_action(parse_state->ct_priv, attr, act, parse_state->extack);
 	if (err)
 		return err;
 
-
 	if (mlx5e_is_eswitch_flow(parse_state->flow))
 		attr->esw_attr->split_count = attr->esw_attr->out_count;
 
-	if (clear_action) {
-		parse_state->ct_clear = true;
-	} else {
-		attr->flags |= MLX5_ATTR_FLAG_CT;
-		flow_flag_set(parse_state->flow, CT);
-		parse_state->ct = true;
-	}
+	attr->flags |= MLX5_ATTR_FLAG_CT;
 
 	return 0;
 }
@@ -61,27 +47,10 @@ tc_act_post_parse_ct(struct mlx5e_tc_act_parse_state *parse_state,
 		     struct mlx5e_priv *priv,
 		     struct mlx5_flow_attr *attr)
 {
-	struct mlx5e_tc_mod_hdr_acts *mod_acts = &attr->parse_attr->mod_hdr_acts;
-	int err;
-
-	/* If ct action exist, we can ignore previous ct_clear actions */
-	if (parse_state->ct)
+	if (!(attr->flags & MLX5_ATTR_FLAG_CT))
 		return 0;
 
-	if (parse_state->ct_clear) {
-		err = mlx5_tc_ct_set_ct_clear_regs(parse_state->ct_priv, mod_acts);
-		if (err) {
-			NL_SET_ERR_MSG_MOD(parse_state->extack,
-					   "Failed to set registers for ct clear");
-			return err;
-		}
-		attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
-
-		/* Prevent handling of additional, redundant clear actions */
-		parse_state->ct_clear = false;
-	}
-
-	return 0;
+	return mlx5_tc_ct_flow_offload(parse_state->ct_priv, attr);
 }
 
 static bool
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index 314983bc6f08..24badbaad935 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -83,12 +83,6 @@ struct mlx5_tc_ct_priv {
 	struct mlx5_tc_ct_debugfs debugfs;
 };
 
-struct mlx5_ct_flow {
-	struct mlx5_flow_attr *pre_ct_attr;
-	struct mlx5_flow_handle *pre_ct_rule;
-	struct mlx5_ct_ft *ft;
-};
-
 struct mlx5_ct_zone_rule {
 	struct mlx5_ct_fs_rule *rule;
 	struct mlx5e_mod_hdr_handle *mh;
@@ -598,12 +592,6 @@ mlx5_tc_ct_entry_set_registers(struct mlx5_tc_ct_priv *ct_priv,
 	return 0;
 }
 
-int mlx5_tc_ct_set_ct_clear_regs(struct mlx5_tc_ct_priv *priv,
-				 struct mlx5e_tc_mod_hdr_acts *mod_acts)
-{
-		return mlx5_tc_ct_entry_set_registers(priv, mod_acts, 0, 0, 0, 0);
-}
-
 static int
 mlx5_tc_ct_parse_mangle_to_mod_act(struct flow_action_entry *act,
 				   char *modact)
@@ -1545,7 +1533,6 @@ mlx5_tc_ct_match_add(struct mlx5_tc_ct_priv *priv,
 int
 mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv,
 			struct mlx5_flow_attr *attr,
-			struct mlx5e_tc_mod_hdr_acts *mod_acts,
 			const struct flow_action_entry *act,
 			struct netlink_ext_ack *extack)
 {
@@ -1555,8 +1542,8 @@ mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
+	attr->ct_attr.ct_action |= act->ct.action; /* So we can have clear + ct */
 	attr->ct_attr.zone = act->ct.zone;
-	attr->ct_attr.ct_action = act->ct.action;
 	attr->ct_attr.nf_ft = act->ct.flow_table;
 	attr->ct_attr.act_miss_cookie = act->miss_cookie;
 
@@ -1892,14 +1879,14 @@ mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft)
 
 /* We translate the tc filter with CT action to the following HW model:
  *
- *	+---------------------+
- *	+ ft prio (tc chain)  +
- *	+ original match      +
- *	+---------------------+
+ *	+-----------------------+
+ *	+ rule (either original +
+ *	+ or post_act rule)     +
+ *	+-----------------------+
  *		 | set act_miss_cookie mapping
  *		 | set fte_id
  *		 | set tunnel_id
- *		 | do decap
+ *		 | rest of actions before the CT action (for this orig/post_act rule)
  *		 |
  * +-------------+
  * | Chain 0	 |
@@ -1924,32 +1911,21 @@ mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft)
  *		 | do nat (if needed)
  *		 v
  *	+--------------+
- *	+ post_act     + original filter actions
+ *	+ post_act     + rest of parsed filter's actions
  *	+ fte_id match +------------------------>
  *	+--------------+
  *
  */
-static struct mlx5_flow_handle *
+static int
 __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv,
-			  struct mlx5_flow_spec *orig_spec,
 			  struct mlx5_flow_attr *attr)
 {
 	bool nat = attr->ct_attr.ct_action & TCA_CT_ACT_NAT;
 	struct mlx5e_priv *priv = netdev_priv(ct_priv->netdev);
-	struct mlx5e_tc_mod_hdr_acts *pre_mod_acts;
-	u32 attr_sz = ns_to_attr_sz(ct_priv->ns_type);
-	struct mlx5_flow_attr *pre_ct_attr;
-	struct mlx5_modify_hdr *mod_hdr;
-	struct mlx5_ct_flow *ct_flow;
 	int act_miss_mapping = 0, err;
 	struct mlx5_ct_ft *ft;
 	u16 zone;
 
-	ct_flow = kzalloc(sizeof(*ct_flow), GFP_KERNEL);
-	if (!ct_flow) {
-		return ERR_PTR(-ENOMEM);
-	}
-
 	/* Register for CT established events */
 	ft = mlx5_tc_ct_add_ft_cb(ct_priv, attr->ct_attr.zone,
 				  attr->ct_attr.nf_ft);
@@ -1958,23 +1934,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv,
 		ct_dbg("Failed to register to ft callback");
 		goto err_ft;
 	}
-	ct_flow->ft = ft;
-
-	/* Base flow attributes of both rules on original rule attribute */
-	ct_flow->pre_ct_attr = mlx5_alloc_flow_attr(ct_priv->ns_type);
-	if (!ct_flow->pre_ct_attr) {
-		err = -ENOMEM;
-		goto err_alloc_pre;
-	}
-
-	pre_ct_attr = ct_flow->pre_ct_attr;
-	memcpy(pre_ct_attr, attr, attr_sz);
-	pre_mod_acts = &pre_ct_attr->parse_attr->mod_hdr_acts;
-
-	/* Modify the original rule's action to fwd and modify, leave decap */
-	pre_ct_attr->action = attr->action & MLX5_FLOW_CONTEXT_ACTION_DECAP;
-	pre_ct_attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-			       MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+	attr->ct_attr.ft = ft;
 
 	err = mlx5e_tc_action_miss_mapping_get(ct_priv->priv, attr, attr->ct_attr.act_miss_cookie,
 					       &act_miss_mapping);
@@ -1982,136 +1942,89 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv,
 		ct_dbg("Failed to get register mapping for act miss");
 		goto err_get_act_miss;
 	}
-	attr->ct_attr.act_miss_mapping = act_miss_mapping;
 
-	err = mlx5e_tc_match_to_reg_set(priv->mdev, pre_mod_acts, ct_priv->ns_type,
-					MAPPED_OBJ_TO_REG, act_miss_mapping);
+	err = mlx5e_tc_match_to_reg_set(priv->mdev, &attr->parse_attr->mod_hdr_acts,
+					ct_priv->ns_type, MAPPED_OBJ_TO_REG, act_miss_mapping);
 	if (err) {
 		ct_dbg("Failed to set act miss register mapping");
 		goto err_mapping;
 	}
 
-	/* If original flow is decap, we do it before going into ct table
-	 * so add a rewrite for the tunnel match_id.
-	 */
-	if ((pre_ct_attr->action & MLX5_FLOW_CONTEXT_ACTION_DECAP) &&
-	    attr->chain == 0) {
-		err = mlx5e_tc_match_to_reg_set(priv->mdev, pre_mod_acts,
-						ct_priv->ns_type,
-						TUNNEL_TO_REG,
-						attr->tunnel_id);
-		if (err) {
-			ct_dbg("Failed to set tunnel register mapping");
-			goto err_mapping;
-		}
-	}
-
-	/* Change original rule point to ct table
-	 * Chain 0 sets the zone and jumps to ct table
+	/* Chain 0 sets the zone and jumps to ct table
 	 * Other chains jump to pre_ct table to align with act_ct cached logic
 	 */
-	pre_ct_attr->dest_chain = 0;
 	if (!attr->chain) {
 		zone = ft->zone & MLX5_CT_ZONE_MASK;
-		err = mlx5e_tc_match_to_reg_set(priv->mdev, pre_mod_acts, ct_priv->ns_type,
-						ZONE_TO_REG, zone);
+		err = mlx5e_tc_match_to_reg_set(priv->mdev, &attr->parse_attr->mod_hdr_acts,
+						ct_priv->ns_type, ZONE_TO_REG, zone);
 		if (err) {
 			ct_dbg("Failed to set zone register mapping");
 			goto err_mapping;
 		}
 
-		pre_ct_attr->dest_ft = nat ? ct_priv->ct_nat : ct_priv->ct;
+		attr->dest_ft = nat ? ct_priv->ct_nat : ct_priv->ct;
 	} else {
-		pre_ct_attr->dest_ft = nat ? ft->pre_ct_nat.ft : ft->pre_ct.ft;
-	}
-
-	mod_hdr = mlx5_modify_header_alloc(priv->mdev, ct_priv->ns_type,
-					   pre_mod_acts->num_actions,
-					   pre_mod_acts->actions);
-	if (IS_ERR(mod_hdr)) {
-		err = PTR_ERR(mod_hdr);
-		ct_dbg("Failed to create pre ct mod hdr");
-		goto err_mapping;
-	}
-	pre_ct_attr->modify_hdr = mod_hdr;
-	ct_flow->pre_ct_rule = mlx5_tc_rule_insert(priv, orig_spec,
-						   pre_ct_attr);
-	if (IS_ERR(ct_flow->pre_ct_rule)) {
-		err = PTR_ERR(ct_flow->pre_ct_rule);
-		ct_dbg("Failed to add pre ct rule");
-		goto err_insert_orig;
+		attr->dest_ft = nat ? ft->pre_ct_nat.ft : ft->pre_ct.ft;
 	}
 
-	attr->ct_attr.ct_flow = ct_flow;
-	mlx5e_mod_hdr_dealloc(pre_mod_acts);
+	attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+	attr->ct_attr.act_miss_mapping = act_miss_mapping;
 
-	return ct_flow->pre_ct_rule;
+	return 0;
 
-err_insert_orig:
-	mlx5_modify_header_dealloc(priv->mdev, pre_ct_attr->modify_hdr);
 err_mapping:
-	mlx5e_mod_hdr_dealloc(pre_mod_acts);
 	mlx5e_tc_action_miss_mapping_put(ct_priv->priv, attr, act_miss_mapping);
 err_get_act_miss:
-	kfree(ct_flow->pre_ct_attr);
-err_alloc_pre:
 	mlx5_tc_ct_del_ft_cb(ct_priv, ft);
 err_ft:
-	kfree(ct_flow);
 	netdev_warn(priv->netdev, "Failed to offload ct flow, err %d\n", err);
-	return ERR_PTR(err);
+	return err;
 }
 
-struct mlx5_flow_handle *
-mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv,
-			struct mlx5_flow_spec *spec,
-			struct mlx5_flow_attr *attr,
-			struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts)
+int
+mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv, struct mlx5_flow_attr *attr)
 {
-	struct mlx5_flow_handle *rule;
+	int err;
 
 	if (!priv)
-		return ERR_PTR(-EOPNOTSUPP);
+		return -EOPNOTSUPP;
+
+	if (attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR) {
+		err = mlx5_tc_ct_entry_set_registers(priv, &attr->parse_attr->mod_hdr_acts,
+						     0, 0, 0, 0);
+		if (err)
+			return err;
+
+		attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+	}
+
+	if (!attr->ct_attr.nf_ft) /* means only ct clear action, and not ct_clear,ct() */
+		return 0;
 
 	mutex_lock(&priv->control_lock);
-	rule = __mlx5_tc_ct_flow_offload(priv, spec, attr);
+	err = __mlx5_tc_ct_flow_offload(priv, attr);
 	mutex_unlock(&priv->control_lock);
 
-	return rule;
+	return err;
 }
 
 static void
 __mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *ct_priv,
-			 struct mlx5_ct_flow *ct_flow,
 			 struct mlx5_flow_attr *attr)
 {
-	struct mlx5_flow_attr *pre_ct_attr = ct_flow->pre_ct_attr;
-	struct mlx5e_priv *priv = netdev_priv(ct_priv->netdev);
-
-	mlx5_tc_rule_delete(priv, ct_flow->pre_ct_rule, pre_ct_attr);
-	mlx5_modify_header_dealloc(priv->mdev, pre_ct_attr->modify_hdr);
-
 	mlx5e_tc_action_miss_mapping_put(ct_priv->priv, attr, attr->ct_attr.act_miss_mapping);
-	mlx5_tc_ct_del_ft_cb(ct_priv, ct_flow->ft);
-
-	kfree(ct_flow->pre_ct_attr);
-	kfree(ct_flow);
+	mlx5_tc_ct_del_ft_cb(ct_priv, attr->ct_attr.ft);
 }
 
 void
 mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *priv,
 		       struct mlx5_flow_attr *attr)
 {
-	struct mlx5_ct_flow *ct_flow = attr->ct_attr.ct_flow;
-
-	/* We are called on error to clean up stuff from parsing
-	 * but we don't have anything for now
-	 */
-	if (!ct_flow)
+	if (!attr->ct_attr.nf_ft) /* means only ct clear action, and not ct_clear,ct() */
 		return;
 
 	mutex_lock(&priv->control_lock);
-	__mlx5_tc_ct_delete_flow(priv, ct_flow, attr);
+	__mlx5_tc_ct_delete_flow(priv, attr);
 	mutex_unlock(&priv->control_lock);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h
index 5c5ddaa83055..8e9316fa46d4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h
@@ -25,11 +25,11 @@ struct nf_flowtable;
 struct mlx5_ct_attr {
 	u16 zone;
 	u16 ct_action;
-	struct mlx5_ct_flow *ct_flow;
 	struct nf_flowtable *nf_ft;
 	u32 ct_labels_id;
 	u32 act_miss_mapping;
 	u64 act_miss_cookie;
+	struct mlx5_ct_ft *ft;
 };
 
 #define zone_to_reg_ct {\
@@ -113,15 +113,12 @@ int mlx5_tc_ct_add_no_trk_match(struct mlx5_flow_spec *spec);
 int
 mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv,
 			struct mlx5_flow_attr *attr,
-			struct mlx5e_tc_mod_hdr_acts *mod_acts,
 			const struct flow_action_entry *act,
 			struct netlink_ext_ack *extack);
 
-struct mlx5_flow_handle *
-mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv,
-			struct mlx5_flow_spec *spec,
-			struct mlx5_flow_attr *attr,
-			struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts);
+int
+mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv, struct mlx5_flow_attr *attr);
+
 void
 mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *priv,
 		       struct mlx5_flow_attr *attr);
@@ -130,10 +127,6 @@ bool
 mlx5e_tc_ct_restore_flow(struct mlx5_tc_ct_priv *ct_priv,
 			 struct sk_buff *skb, u8 zone_restore_id);
 
-int
-mlx5_tc_ct_set_ct_clear_regs(struct mlx5_tc_ct_priv *priv,
-			     struct mlx5e_tc_mod_hdr_acts *mod_acts);
-
 #else /* CONFIG_MLX5_TC_CT */
 
 static inline struct mlx5_tc_ct_priv *
@@ -175,17 +168,9 @@ mlx5_tc_ct_add_no_trk_match(struct mlx5_flow_spec *spec)
 	return 0;
 }
 
-static inline int
-mlx5_tc_ct_set_ct_clear_regs(struct mlx5_tc_ct_priv *priv,
-			     struct mlx5e_tc_mod_hdr_acts *mod_acts)
-{
-	return -EOPNOTSUPP;
-}
-
 static inline int
 mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv,
 			struct mlx5_flow_attr *attr,
-			struct mlx5e_tc_mod_hdr_acts *mod_acts,
 			const struct flow_action_entry *act,
 			struct netlink_ext_ack *extack)
 {
@@ -193,13 +178,11 @@ mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv,
 	return -EOPNOTSUPP;
 }
 
-static inline struct mlx5_flow_handle *
+static inline int
 mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv,
-			struct mlx5_flow_spec *spec,
-			struct mlx5_flow_attr *attr,
-			struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts)
+			struct mlx5_flow_attr *attr)
 {
-	return ERR_PTR(-EOPNOTSUPP);
+	return -EOPNOTSUPP;
 }
 
 static inline void
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 15deb1df4df3..d8fa8f0c0fbf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -488,15 +488,6 @@ mlx5e_tc_rule_offload(struct mlx5e_priv *priv,
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	int err;
 
-	if (attr->flags & MLX5_ATTR_FLAG_CT) {
-		struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts =
-			&attr->parse_attr->mod_hdr_acts;
-
-		return mlx5_tc_ct_flow_offload(get_ct_priv(priv),
-					       spec, attr,
-					       mod_hdr_acts);
-	}
-
 	if (!is_mdev_switchdev_mode(priv->mdev))
 		return mlx5e_add_offloaded_nic_rule(priv, spec, attr);
 
@@ -519,11 +510,6 @@ mlx5e_tc_rule_unoffload(struct mlx5e_priv *priv,
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 
-	if (attr->flags & MLX5_ATTR_FLAG_CT) {
-		mlx5_tc_ct_delete_flow(get_ct_priv(priv), attr);
-		return;
-	}
-
 	if (!is_mdev_switchdev_mode(priv->mdev)) {
 		mlx5e_del_offloaded_nic_rule(priv, rule, attr);
 		return;
@@ -1396,13 +1382,7 @@ mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv,
 			return err;
 	}
 
-	if (attr->flags & MLX5_ATTR_FLAG_CT)
-		flow->rule[0] = mlx5_tc_ct_flow_offload(get_ct_priv(priv), &parse_attr->spec,
-							attr, &parse_attr->mod_hdr_acts);
-	else
-		flow->rule[0] = mlx5e_add_offloaded_nic_rule(priv, &parse_attr->spec,
-							     attr);
-
+	flow->rule[0] = mlx5e_add_offloaded_nic_rule(priv, &parse_attr->spec, attr);
 	return PTR_ERR_OR_ZERO(flow->rule[0]);
 }
 
@@ -1433,9 +1413,7 @@ static void mlx5e_tc_del_nic_flow(struct mlx5e_priv *priv,
 
 	flow_flag_clear(flow, OFFLOADED);
 
-	if (attr->flags & MLX5_ATTR_FLAG_CT)
-		mlx5_tc_ct_delete_flow(get_ct_priv(flow->priv), attr);
-	else if (!IS_ERR_OR_NULL(flow->rule[0]))
+	if (!IS_ERR_OR_NULL(flow->rule[0]))
 		mlx5e_del_offloaded_nic_rule(priv, flow->rule[0], attr);
 
 	/* Remove root table if no rules are left to avoid
@@ -3734,6 +3712,7 @@ mlx5e_clone_flow_attr_for_post_act(struct mlx5_flow_attr *attr,
 	attr2->dest_chain = 0;
 	attr2->dest_ft = NULL;
 	attr2->act_id_restore_rule = NULL;
+	memset(&attr2->ct_attr, 0, sizeof(attr2->ct_attr));
 
 	if (ns_type == MLX5_FLOW_NAMESPACE_FDB) {
 		attr2->esw_attr->out_count = 0;
@@ -4443,6 +4422,8 @@ mlx5_free_flow_attr_actions(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *a
 		mlx5e_tc_detach_mod_hdr(flow->priv, flow, attr);
 	}
 
+	mlx5_tc_ct_delete_flow(get_ct_priv(flow->priv), attr);
+
 	free_branch_attr(flow, attr->branch_true);
 	free_branch_attr(flow, attr->branch_false);
 }
@@ -4906,7 +4887,7 @@ int mlx5e_stats_flower(struct net_device *dev, struct mlx5e_priv *priv,
 		goto errout;
 	}
 
-	if (mlx5e_is_offloaded_flow(flow) || flow_flag_test(flow, CT)) {
+	if (mlx5e_is_offloaded_flow(flow)) {
 		if (flow_flag_test(flow, USE_ACT_STATS)) {
 			f->use_act_stats = true;
 		} else {
-- 
cgit v1.2.3-70-g09d2


From d0cc0853640d81397572d8152d068cd673a3e739 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Mon, 23 Jan 2023 10:46:10 +0200
Subject: net/mlx5e: TC, Remove multiple ct actions limitation

Multiple ct actions was not supported when only chain was
restored on misses, as CT is a modifying action which could
modify the packet and cause the sw ct rule to not match again
and continue processing.

Now that the misses continue from the relevant tc ct action
in software, this case is supported.

Remove this limitation.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en/tc/act/act.h    |  1 -
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c | 18 ------------------
 2 files changed, 19 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
index 033afd44fc4f..0e6e1872ac62 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h
@@ -17,7 +17,6 @@ struct mlx5e_tc_act_parse_state {
 	struct mlx5e_tc_flow *flow;
 	struct netlink_ext_ack *extack;
 	u32 actions;
-	bool ct;
 	bool encap;
 	bool decap;
 	bool mpls_push;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
index 36bfed07d400..92d3952dfa8b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/ct.c
@@ -5,23 +5,6 @@
 #include "en/tc_priv.h"
 #include "en/tc_ct.h"
 
-static bool
-tc_act_can_offload_ct(struct mlx5e_tc_act_parse_state *parse_state,
-		      const struct flow_action_entry *act,
-		      int act_index,
-		      struct mlx5_flow_attr *attr)
-{
-	bool clear_action = act->ct.action & TCA_CT_ACT_CLEAR;
-	struct netlink_ext_ack *extack = parse_state->extack;
-
-	if (parse_state->ct && !clear_action) {
-		NL_SET_ERR_MSG_MOD(extack, "Multiple CT actions are not supported");
-		return false;
-	}
-
-	return true;
-}
-
 static int
 tc_act_parse_ct(struct mlx5e_tc_act_parse_state *parse_state,
 		const struct flow_action_entry *act,
@@ -71,7 +54,6 @@ tc_act_is_missable_ct(const struct flow_action_entry *act)
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_ct = {
-	.can_offload = tc_act_can_offload_ct,
 	.parse_action = tc_act_parse_ct,
 	.post_parse = tc_act_post_parse_ct,
 	.is_multi_table_act = tc_act_is_multi_table_act_ct,
-- 
cgit v1.2.3-70-g09d2


From 5d7cb06eb91a3a45de6b7248dd30a93d7c820022 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Wed, 25 Jan 2023 11:56:28 +0200
Subject: net/mlx5e: TC, Remove tuple rewrite and ct limitation

Tuple rewrite and ct action was not supported when only chain was
restored on misses. To work around that limitation, ct action was
reordered to be first (so if hw misses on ct action, packet wasn't
modified). This reordering wasn't possible for tuple rewrite
actions before ct action since the ct action result was
dependent on the tuple info.

Now that the misses continue from the relevant tc ct action
in software and ct action is no longer reordered, this case
is supported.

Remove this limitation.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 142 ++++++------------------
 1 file changed, 33 insertions(+), 109 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index d8fa8f0c0fbf..b6469abc7012 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3423,114 +3423,59 @@ struct ipv6_hoplimit_word {
 };
 
 static bool
-is_action_keys_supported(const struct flow_action_entry *act, bool ct_flow,
-			 bool *modify_ip_header, bool *modify_tuple,
-			 struct netlink_ext_ack *extack)
+is_flow_action_modify_ip_header(struct flow_action *flow_action)
 {
+	const struct flow_action_entry *act;
 	u32 mask, offset;
 	u8 htype;
+	int i;
 
-	htype = act->mangle.htype;
-	offset = act->mangle.offset;
-	mask = ~act->mangle.mask;
 	/* For IPv4 & IPv6 header check 4 byte word,
 	 * to determine that modified fields
 	 * are NOT ttl & hop_limit only.
 	 */
-	if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP4) {
-		struct ip_ttl_word *ttl_word =
-			(struct ip_ttl_word *)&mask;
-
-		if (offset != offsetof(struct iphdr, ttl) ||
-		    ttl_word->protocol ||
-		    ttl_word->check) {
-			*modify_ip_header = true;
-		}
-
-		if (offset >= offsetof(struct iphdr, saddr))
-			*modify_tuple = true;
+	flow_action_for_each(i, act, flow_action) {
+		if (act->id != FLOW_ACTION_MANGLE &&
+		    act->id != FLOW_ACTION_ADD)
+			continue;
 
-		if (ct_flow && *modify_tuple) {
-			NL_SET_ERR_MSG_MOD(extack,
-					   "can't offload re-write of ipv4 address with action ct");
-			return false;
-		}
-	} else if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP6) {
-		struct ipv6_hoplimit_word *hoplimit_word =
-			(struct ipv6_hoplimit_word *)&mask;
-
-		if (offset != offsetof(struct ipv6hdr, payload_len) ||
-		    hoplimit_word->payload_len ||
-		    hoplimit_word->nexthdr) {
-			*modify_ip_header = true;
-		}
-
-		if (ct_flow && offset >= offsetof(struct ipv6hdr, saddr))
-			*modify_tuple = true;
-
-		if (ct_flow && *modify_tuple) {
-			NL_SET_ERR_MSG_MOD(extack,
-					   "can't offload re-write of ipv6 address with action ct");
-			return false;
-		}
-	} else if (htype == FLOW_ACT_MANGLE_HDR_TYPE_TCP ||
-		   htype == FLOW_ACT_MANGLE_HDR_TYPE_UDP) {
-		*modify_tuple = true;
-		if (ct_flow) {
-			NL_SET_ERR_MSG_MOD(extack,
-					   "can't offload re-write of transport header ports with action ct");
-			return false;
+		htype = act->mangle.htype;
+		offset = act->mangle.offset;
+		mask = ~act->mangle.mask;
+
+		if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP4) {
+			struct ip_ttl_word *ttl_word =
+				(struct ip_ttl_word *)&mask;
+
+			if (offset != offsetof(struct iphdr, ttl) ||
+			    ttl_word->protocol ||
+			    ttl_word->check)
+				return true;
+		} else if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP6) {
+			struct ipv6_hoplimit_word *hoplimit_word =
+				(struct ipv6_hoplimit_word *)&mask;
+
+			if (offset != offsetof(struct ipv6hdr, payload_len) ||
+			    hoplimit_word->payload_len ||
+			    hoplimit_word->nexthdr)
+				return true;
 		}
 	}
 
-	return true;
-}
-
-static bool modify_tuple_supported(bool modify_tuple, bool ct_clear,
-				   bool ct_flow, struct netlink_ext_ack *extack,
-				   struct mlx5e_priv *priv,
-				   struct mlx5_flow_spec *spec)
-{
-	if (!modify_tuple || ct_clear)
-		return true;
-
-	if (ct_flow) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "can't offload tuple modification with non-clear ct()");
-		netdev_info(priv->netdev,
-			    "can't offload tuple modification with non-clear ct()");
-		return false;
-	}
-
-	/* Add ct_state=-trk match so it will be offloaded for non ct flows
-	 * (or after clear action), as otherwise, since the tuple is changed,
-	 * we can't restore ct state
-	 */
-	if (mlx5_tc_ct_add_no_trk_match(spec)) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "can't offload tuple modification with ct matches and no ct(clear) action");
-		netdev_info(priv->netdev,
-			    "can't offload tuple modification with ct matches and no ct(clear) action");
-		return false;
-	}
-
-	return true;
+	return false;
 }
 
 static bool modify_header_match_supported(struct mlx5e_priv *priv,
 					  struct mlx5_flow_spec *spec,
 					  struct flow_action *flow_action,
-					  u32 actions, bool ct_flow,
-					  bool ct_clear,
+					  u32 actions,
 					  struct netlink_ext_ack *extack)
 {
-	const struct flow_action_entry *act;
-	bool modify_ip_header, modify_tuple;
+	bool modify_ip_header;
 	void *headers_c;
 	void *headers_v;
 	u16 ethertype;
 	u8 ip_proto;
-	int i;
 
 	headers_c = mlx5e_get_match_headers_criteria(actions, spec);
 	headers_v = mlx5e_get_match_headers_value(actions, spec);
@@ -3541,23 +3486,7 @@ static bool modify_header_match_supported(struct mlx5e_priv *priv,
 	    ethertype != ETH_P_IP && ethertype != ETH_P_IPV6)
 		goto out_ok;
 
-	modify_ip_header = false;
-	modify_tuple = false;
-	flow_action_for_each(i, act, flow_action) {
-		if (act->id != FLOW_ACTION_MANGLE &&
-		    act->id != FLOW_ACTION_ADD)
-			continue;
-
-		if (!is_action_keys_supported(act, ct_flow,
-					      &modify_ip_header,
-					      &modify_tuple, extack))
-			return false;
-	}
-
-	if (!modify_tuple_supported(modify_tuple, ct_clear, ct_flow, extack,
-				    priv, spec))
-		return false;
-
+	modify_ip_header = is_flow_action_modify_ip_header(flow_action);
 	ip_proto = MLX5_GET(fte_match_set_lyr_2_4, headers_v, ip_protocol);
 	if (modify_ip_header && ip_proto != IPPROTO_TCP &&
 	    ip_proto != IPPROTO_UDP && ip_proto != IPPROTO_ICMP) {
@@ -3611,14 +3540,9 @@ actions_match_supported(struct mlx5e_priv *priv,
 			struct mlx5e_tc_flow *flow,
 			struct netlink_ext_ack *extack)
 {
-	bool ct_flow, ct_clear;
-
-	ct_clear = flow->attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR;
-	ct_flow = flow_flag_test(flow, CT) && !ct_clear;
-
 	if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR &&
-	    !modify_header_match_supported(priv, &parse_attr->spec, flow_action,
-					   actions, ct_flow, ct_clear, extack))
+	    !modify_header_match_supported(priv, &parse_attr->spec, flow_action, actions,
+					   extack))
 		return false;
 
 	if (mlx5e_is_eswitch_flow(flow) &&
-- 
cgit v1.2.3-70-g09d2


From dc614025e22810eab057fe683cd0042abde3930c Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Wed, 25 Jan 2023 12:01:47 +0200
Subject: net/mlx5e: TC, Remove mirror and ct limitation

Mirror action before a ct nat action was not supported when only
chain was restored on misses. As to work around that limitation,
ct action was reordered to be first (so if hw misses on ct
action, packet wasn't modified). This reordering wasn't possible
if there was mirror action before the ct nat action, as we had to
mirror the packet before the nat operation.

Now that the misses continue from the relevant tc ct action
in software and ct action is no longer reordered, this case
is supported.

Remove this limitation.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index b6469abc7012..728b82ce4031 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -3507,19 +3507,6 @@ actions_match_supported_fdb(struct mlx5e_priv *priv,
 			    struct netlink_ext_ack *extack)
 {
 	struct mlx5_esw_flow_attr *esw_attr = flow->attr->esw_attr;
-	bool ct_flow, ct_clear;
-
-	ct_clear = flow->attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR;
-	ct_flow = flow_flag_test(flow, CT) && !ct_clear;
-
-	if (esw_attr->split_count && ct_flow &&
-	    !MLX5_CAP_GEN(esw_attr->in_mdev, reg_c_preserve)) {
-		/* All registers used by ct are cleared when using
-		 * split rules.
-		 */
-		NL_SET_ERR_MSG_MOD(extack, "Can't offload mirroring with action ct");
-		return false;
-	}
 
 	if (esw_attr->split_count > 0 && !mlx5_esw_has_fwd_fdb(priv->mdev)) {
 		NL_SET_ERR_MSG_MOD(extack,
-- 
cgit v1.2.3-70-g09d2


From 35c8de16d84632dd73ac17f619f3cb824803c89e Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb@nvidia.com>
Date: Wed, 25 Jan 2023 12:03:58 +0200
Subject: net/mlx5e: TC, Remove sample and ct limitation

Sample action before a ct nat action was not supported when only
chain was restored on misses. As to work around that limitation,
ct action was reordered to be first (so if hw misses on ct
action, packet wasn't modified). This reordering wasn't possible
if there was a sample action before the ct nat action, as we had to
sample the packet before the nat operation.

Now that the misses continue from the relevant tc ct action
in software and ct action is no longer reordered, this case
is supported.

Remove this limitation.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/tc/act/sample.c   | 20 --------------------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h | 11 +++++------
 2 files changed, 5 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/sample.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/sample.c
index 2c0196431302..2df02f99cecf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/sample.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/sample.c
@@ -6,25 +6,6 @@
 #include "en/tc_priv.h"
 #include "en/tc/act/sample.h"
 
-static bool
-tc_act_can_offload_sample(struct mlx5e_tc_act_parse_state *parse_state,
-			  const struct flow_action_entry *act,
-			  int act_index,
-			  struct mlx5_flow_attr *attr)
-{
-	struct netlink_ext_ack *extack = parse_state->extack;
-	bool ct_nat;
-
-	ct_nat = attr->ct_attr.ct_action & TCA_CT_ACT_NAT;
-
-	if (flow_flag_test(parse_state->flow, CT) && ct_nat) {
-		NL_SET_ERR_MSG_MOD(extack, "Sample action with CT NAT is not supported");
-		return false;
-	}
-
-	return true;
-}
-
 static int
 tc_act_parse_sample(struct mlx5e_tc_act_parse_state *parse_state,
 		    const struct flow_action_entry *act,
@@ -65,7 +46,6 @@ tc_act_is_multi_table_act_sample(struct mlx5e_priv *priv,
 }
 
 struct mlx5e_tc_act mlx5e_tc_act_sample = {
-	.can_offload = tc_act_can_offload_sample,
 	.parse_action = tc_act_parse_sample,
 	.is_multi_table_act = tc_act_is_multi_table_act_sample,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h
index 451fd4342a5a..ba2b1f24ff14 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h
@@ -25,12 +25,11 @@ enum {
 	MLX5E_TC_FLOW_FLAG_DUP                   = MLX5E_TC_FLOW_BASE + 4,
 	MLX5E_TC_FLOW_FLAG_NOT_READY             = MLX5E_TC_FLOW_BASE + 5,
 	MLX5E_TC_FLOW_FLAG_DELETED               = MLX5E_TC_FLOW_BASE + 6,
-	MLX5E_TC_FLOW_FLAG_CT                    = MLX5E_TC_FLOW_BASE + 7,
-	MLX5E_TC_FLOW_FLAG_L3_TO_L2_DECAP        = MLX5E_TC_FLOW_BASE + 8,
-	MLX5E_TC_FLOW_FLAG_TUN_RX                = MLX5E_TC_FLOW_BASE + 9,
-	MLX5E_TC_FLOW_FLAG_FAILED                = MLX5E_TC_FLOW_BASE + 10,
-	MLX5E_TC_FLOW_FLAG_SAMPLE                = MLX5E_TC_FLOW_BASE + 11,
-	MLX5E_TC_FLOW_FLAG_USE_ACT_STATS	 = MLX5E_TC_FLOW_BASE + 12,
+	MLX5E_TC_FLOW_FLAG_L3_TO_L2_DECAP        = MLX5E_TC_FLOW_BASE + 7,
+	MLX5E_TC_FLOW_FLAG_TUN_RX                = MLX5E_TC_FLOW_BASE + 8,
+	MLX5E_TC_FLOW_FLAG_FAILED                = MLX5E_TC_FLOW_BASE + 9,
+	MLX5E_TC_FLOW_FLAG_SAMPLE                = MLX5E_TC_FLOW_BASE + 10,
+	MLX5E_TC_FLOW_FLAG_USE_ACT_STATS         = MLX5E_TC_FLOW_BASE + 11,
 };
 
 struct mlx5e_tc_flow_parse_attr {
-- 
cgit v1.2.3-70-g09d2


From 1a62ffcaaabfbefe2094d98f94a241b0b6fa34be Mon Sep 17 00:00:00 2001
From: Emeel Hakim <ehakim@nvidia.com>
Date: Thu, 29 Dec 2022 17:17:47 +0200
Subject: net/mlx5e: Remove redundant macsec code

Currently macsec_fs_tx_create uses memset to set
two parameters to zeros when they are already
initialized to zeros.

Don't pass macsec_ctx to mlx5e_macsec_fs_add_rule
since it's not used.

Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
index 5b658a5588c6..9173b67becef 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
@@ -292,8 +292,6 @@ static int macsec_fs_tx_create(struct mlx5e_macsec_fs *macsec_fs)
 	}
 
 	/* Tx crypto table MKE rule - MKE packets shouldn't be offloaded */
-	memset(&flow_act, 0, sizeof(flow_act));
-	memset(spec, 0, sizeof(*spec));
 	spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
 
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype);
@@ -1109,7 +1107,6 @@ static void macsec_fs_rx_setup_fte(struct mlx5_flow_spec *spec,
 
 static union mlx5e_macsec_rule *
 macsec_fs_rx_add_rule(struct mlx5e_macsec_fs *macsec_fs,
-		      const struct macsec_context *macsec_ctx,
 		      struct mlx5_macsec_rule_attrs *attrs,
 		      u32 fs_id)
 {
@@ -1334,7 +1331,7 @@ mlx5e_macsec_fs_add_rule(struct mlx5e_macsec_fs *macsec_fs,
 {
 	return (attrs->action == MLX5_ACCEL_MACSEC_ACTION_ENCRYPT) ?
 		macsec_fs_tx_add_rule(macsec_fs, macsec_ctx, attrs, sa_fs_id) :
-		macsec_fs_rx_add_rule(macsec_fs, macsec_ctx, attrs, *sa_fs_id);
+		macsec_fs_rx_add_rule(macsec_fs, attrs, *sa_fs_id);
 }
 
 void mlx5e_macsec_fs_del_rule(struct mlx5e_macsec_fs *macsec_fs,
-- 
cgit v1.2.3-70-g09d2


From 6a40109275626267ebf413ceda81c64719b5c431 Mon Sep 17 00:00:00 2001
From: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Date: Wed, 8 Mar 2023 08:18:22 -0800
Subject: net/mlx5: Update cyclecounter shift value to improve ptp free running
 mode precision

Multiplier values are equivalent to 2^(shift constant) since all mlx5
devices advertise a 1Ghz frequency for the internal timer. The previous
shift constant of 23 led to internal timer adjustments only taking place
when the provided adjustment values were greater than or equal to ~120 ppb
or ~7864 scaled ppm. Using a shift constant of 31 enables adjustments when
an adjustment parameter is greater than or equal to ~0.47 ppb or ~30.8
scaled ppm.

Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Bar Shapira <bshapira@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
index 4c9a40211059..932fbc843c69 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
@@ -39,7 +39,7 @@
 #include "clock.h"
 
 enum {
-	MLX5_CYCLES_SHIFT	= 23
+	MLX5_CYCLES_SHIFT	= 31
 };
 
 enum {
-- 
cgit v1.2.3-70-g09d2


From cf1cccae79836aceaecc1d6a2d1c955d1e2e8b0c Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Mon, 30 Jan 2023 15:52:08 +0200
Subject: net/mlx5e: Rename misleading skb_pc/cc references in ptp code

The 'skb_pc/cc' naming is misleading as the values hold the
producer/consumer indices (masked values), not the counters. Rename to
'skb_pi/ci'.

Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Adham Faris <afaris@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c
index eb5aeba3addf..eb5abd0e55d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c
@@ -81,23 +81,23 @@ void mlx5e_skb_cb_hwtstamp_handler(struct sk_buff *skb, int hwtstamp_type,
 
 #define PTP_WQE_CTR2IDX(val) ((val) & ptpsq->ts_cqe_ctr_mask)
 
-static bool mlx5e_ptp_ts_cqe_drop(struct mlx5e_ptpsq *ptpsq, u16 skb_cc, u16 skb_id)
+static bool mlx5e_ptp_ts_cqe_drop(struct mlx5e_ptpsq *ptpsq, u16 skb_ci, u16 skb_id)
 {
-	return (ptpsq->ts_cqe_ctr_mask && (skb_cc != skb_id));
+	return (ptpsq->ts_cqe_ctr_mask && (skb_ci != skb_id));
 }
 
 static bool mlx5e_ptp_ts_cqe_ooo(struct mlx5e_ptpsq *ptpsq, u16 skb_id)
 {
-	u16 skb_cc = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_cc);
-	u16 skb_pc = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_pc);
+	u16 skb_ci = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_cc);
+	u16 skb_pi = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_pc);
 
-	if (PTP_WQE_CTR2IDX(skb_id - skb_cc) >= PTP_WQE_CTR2IDX(skb_pc - skb_cc))
+	if (PTP_WQE_CTR2IDX(skb_id - skb_ci) >= PTP_WQE_CTR2IDX(skb_pi - skb_ci))
 		return true;
 
 	return false;
 }
 
-static void mlx5e_ptp_skb_fifo_ts_cqe_resync(struct mlx5e_ptpsq *ptpsq, u16 skb_cc,
+static void mlx5e_ptp_skb_fifo_ts_cqe_resync(struct mlx5e_ptpsq *ptpsq, u16 skb_ci,
 					     u16 skb_id, int budget)
 {
 	struct skb_shared_hwtstamps hwts = {};
@@ -105,13 +105,13 @@ static void mlx5e_ptp_skb_fifo_ts_cqe_resync(struct mlx5e_ptpsq *ptpsq, u16 skb_
 
 	ptpsq->cq_stats->resync_event++;
 
-	while (skb_cc != skb_id) {
+	while (skb_ci != skb_id) {
 		skb = mlx5e_skb_fifo_pop(&ptpsq->skb_fifo);
 		hwts.hwtstamp = mlx5e_skb_cb_get_hwts(skb)->cqe_hwtstamp;
 		skb_tstamp_tx(skb, &hwts);
 		ptpsq->cq_stats->resync_cqe++;
 		napi_consume_skb(skb, budget);
-		skb_cc = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_cc);
+		skb_ci = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_cc);
 	}
 }
 
@@ -120,7 +120,7 @@ static void mlx5e_ptp_handle_ts_cqe(struct mlx5e_ptpsq *ptpsq,
 				    int budget)
 {
 	u16 skb_id = PTP_WQE_CTR2IDX(be16_to_cpu(cqe->wqe_counter));
-	u16 skb_cc = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_cc);
+	u16 skb_ci = PTP_WQE_CTR2IDX(ptpsq->skb_fifo_cc);
 	struct mlx5e_txqsq *sq = &ptpsq->txqsq;
 	struct sk_buff *skb;
 	ktime_t hwtstamp;
@@ -131,13 +131,13 @@ static void mlx5e_ptp_handle_ts_cqe(struct mlx5e_ptpsq *ptpsq,
 		goto out;
 	}
 
-	if (mlx5e_ptp_ts_cqe_drop(ptpsq, skb_cc, skb_id)) {
+	if (mlx5e_ptp_ts_cqe_drop(ptpsq, skb_ci, skb_id)) {
 		if (mlx5e_ptp_ts_cqe_ooo(ptpsq, skb_id)) {
 			/* already handled by a previous resync */
 			ptpsq->cq_stats->ooo_cqe_drop++;
 			return;
 		}
-		mlx5e_ptp_skb_fifo_ts_cqe_resync(ptpsq, skb_cc, skb_id, budget);
+		mlx5e_ptp_skb_fifo_ts_cqe_resync(ptpsq, skb_ci, skb_id, budget);
 	}
 
 	skb = mlx5e_skb_fifo_pop(&ptpsq->skb_fifo);
-- 
cgit v1.2.3-70-g09d2


From 6bd0f349ae70ce1fd325edee23fbe1d27dfbecb6 Mon Sep 17 00:00:00 2001
From: Adham Faris <afaris@nvidia.com>
Date: Wed, 15 Mar 2023 15:12:45 +0200
Subject: net/mlx5e: Fix RQ SW state layout in RQ devlink health diagnostics

Remove nesting level before RQ's SW state title and before RQ's SW
state capabilities line.

Preceding the RQ's SW state with a nameless nesting, wraps the inner SW
state map/dictionary with a nameless dictionary which is prohibited in
JSON file format.

Removing preceding SW state nest by removing function call
devlink_fmsg_obj_nest_start() and devlink_fmsg_obj_nest_end().

Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
index a047a2a4ddac..e8eea9ffd5eb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c
@@ -259,10 +259,6 @@ static int mlx5e_health_rq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_
 
 	BUILD_BUG_ON_MSG(ARRAY_SIZE(rq_sw_state_type_name) != MLX5E_NUM_RQ_STATES,
 			 "rq_sw_state_type_name string array must be consistent with MLX5E_RQ_STATE_* enum in en.h");
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
 	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
 	if (err)
 		return err;
@@ -274,11 +270,7 @@ static int mlx5e_health_rq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_
 			return err;
 	}
 
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return devlink_fmsg_obj_nest_end(fmsg);
+	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
 static int
-- 
cgit v1.2.3-70-g09d2


From b0d87ed27be7853af8f897c3cfcb9ddc5179a2a3 Mon Sep 17 00:00:00 2001
From: Adham Faris <afaris@nvidia.com>
Date: Wed, 15 Mar 2023 16:36:22 +0200
Subject: net/mlx5e: Fix SQ SW state layout in SQ devlink health diagnostics

Remove nesting level before SQ's SW state title and before SQ's SW
state capabilities line.

Preceding the SQ's SW state with a nameless nesting, wraps the inner SW
state map/dictionary with a nameless dictionary which is prohibited in
JSON file format.

Removing preceding SW state nest by removing function call
devlink_fmsg_obj_nest_start() and devlink_fmsg_obj_nest_end().

Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
index 44c1926843a1..b35ff289af49 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
@@ -57,10 +57,6 @@ static int mlx5e_health_sq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_
 
 	BUILD_BUG_ON_MSG(ARRAY_SIZE(sq_sw_state_type_name) != MLX5E_NUM_SQ_STATES,
 			 "sq_sw_state_type_name string array must be consistent with MLX5E_SQ_STATE_* enum in en.h");
-	err = devlink_fmsg_obj_nest_start(fmsg);
-	if (err)
-		return err;
-
 	err = mlx5e_health_fmsg_named_obj_nest_start(fmsg, "SW State");
 	if (err)
 		return err;
@@ -72,11 +68,7 @@ static int mlx5e_health_sq_put_sw_state(struct devlink_fmsg *fmsg, struct mlx5e_
 			return err;
 	}
 
-	err = mlx5e_health_fmsg_named_obj_nest_end(fmsg);
-	if (err)
-		return err;
-
-	return devlink_fmsg_obj_nest_end(fmsg);
+	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);
 }
 
 static int mlx5e_tx_reporter_err_cqe_recover(void *ctx)
-- 
cgit v1.2.3-70-g09d2


From 10739ea3132861b1344264b110ea48d951a0e3b0 Mon Sep 17 00:00:00 2001
From: Shenwei Wang <shenwei.wang@nxp.com>
Date: Mon, 3 Apr 2023 17:23:01 -0500
Subject: net: stmmac: add support for platform specific reset

This patch adds support for platform-specific reset logic in the
stmmac driver. Some SoCs require a different reset mechanism than
the standard dwmac IP reset. To support these platforms, a new function
pointer 'fix_soc_reset' is added to the plat_stmmacenet_data structure.
The stmmac_reset in hwif.h is modified to call the 'fix_soc_reset'
function if it exists. This enables the driver to use the platform-specific
reset logic when necessary.

Signed-off-by: Shenwei Wang <shenwei.wang@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230403222302.328262-1-shenwei.wang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/hwif.c | 13 +++++++++++++
 drivers/net/ethernet/stmicro/stmmac/hwif.h |  3 +--
 include/linux/stmmac.h                     |  1 +
 3 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c
index bb7114f970f8..b8ba8f2d8041 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.c
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c
@@ -87,6 +87,19 @@ static int stmmac_dwxlgmac_quirks(struct stmmac_priv *priv)
 	return 0;
 }
 
+int stmmac_reset(struct stmmac_priv *priv, void __iomem *ioaddr)
+{
+	struct plat_stmmacenet_data *plat = priv ? priv->plat : NULL;
+
+	if (!priv)
+		return -EINVAL;
+
+	if (plat && plat->fix_soc_reset)
+		return plat->fix_soc_reset(plat, ioaddr);
+
+	return stmmac_do_callback(priv, dma, reset, ioaddr);
+}
+
 static const struct stmmac_hwif_entry {
 	bool gmac;
 	bool gmac4;
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 16a7421715cb..1cc286b000b6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -214,8 +214,6 @@ struct stmmac_dma_ops {
 	int (*enable_tbs)(void __iomem *ioaddr, bool en, u32 chan);
 };
 
-#define stmmac_reset(__priv, __args...) \
-	stmmac_do_callback(__priv, dma, reset, __args)
 #define stmmac_dma_init(__priv, __args...) \
 	stmmac_do_void_callback(__priv, dma, init, __args)
 #define stmmac_init_chan(__priv, __args...) \
@@ -640,6 +638,7 @@ extern const struct stmmac_mmc_ops dwxgmac_mmc_ops;
 #define GMAC_VERSION		0x00000020	/* GMAC CORE Version */
 #define GMAC4_VERSION		0x00000110	/* GMAC4+ CORE Version */
 
+int stmmac_reset(struct stmmac_priv *priv, void __iomem *ioaddr);
 int stmmac_hwif_init(struct stmmac_priv *priv);
 
 #endif /* __STMMAC_HWIF_H__ */
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index a2414c187483..dafa001e9e7a 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -223,6 +223,7 @@ struct plat_stmmacenet_data {
 	struct stmmac_rxq_cfg rx_queues_cfg[MTL_MAX_RX_QUEUES];
 	struct stmmac_txq_cfg tx_queues_cfg[MTL_MAX_TX_QUEUES];
 	void (*fix_mac_speed)(void *priv, unsigned int speed);
+	int (*fix_soc_reset)(void *priv, void __iomem *ioaddr);
 	int (*serdes_powerup)(struct net_device *ndev, void *priv);
 	void (*serdes_powerdown)(struct net_device *ndev, void *priv);
 	void (*speed_mode_2500)(struct net_device *ndev, void *priv);
-- 
cgit v1.2.3-70-g09d2


From b536f32b5b034f592df0f0ba129ad59fa0f3e532 Mon Sep 17 00:00:00 2001
From: Shenwei Wang <shenwei.wang@nxp.com>
Date: Mon, 3 Apr 2023 17:23:02 -0500
Subject: net: stmmac: dwmac-imx: use platform specific reset for imx93 SoCs

The patch addresses an issue with the reset logic on the i.MX93 SoC, which
requires configuration of the correct interface speed under RMII mode to
complete the reset. The patch implements a fix_soc_reset function and uses
it specifically for the i.MX93 SoCs.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Shenwei Wang <shenwei.wang@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230403222302.328262-2-shenwei.wang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c | 27 +++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c
index 2a2be65d65a0..7c228bd0d099 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c
@@ -37,10 +37,15 @@
 #define MX93_GPR_ENET_QOS_INTF_SEL_RGMII	(0x1 << 1)
 #define MX93_GPR_ENET_QOS_CLK_GEN_EN		(0x1 << 0)
 
+#define DMA_BUS_MODE			0x00001000
+#define DMA_BUS_MODE_SFT_RESET		(0x1 << 0)
+#define RMII_RESET_SPEED		(0x3 << 14)
+
 struct imx_dwmac_ops {
 	u32 addr_width;
 	bool mac_rgmii_txclk_auto_adj;
 
+	int (*fix_soc_reset)(void *priv, void __iomem *ioaddr);
 	int (*set_intf_mode)(struct plat_stmmacenet_data *plat_dat);
 };
 
@@ -207,6 +212,25 @@ static void imx_dwmac_fix_speed(void *priv, unsigned int speed)
 		dev_err(dwmac->dev, "failed to set tx rate %lu\n", rate);
 }
 
+static int imx_dwmac_mx93_reset(void *priv, void __iomem *ioaddr)
+{
+	struct plat_stmmacenet_data *plat_dat = priv;
+	u32 value = readl(ioaddr + DMA_BUS_MODE);
+
+	/* DMA SW reset */
+	value |= DMA_BUS_MODE_SFT_RESET;
+	writel(value, ioaddr + DMA_BUS_MODE);
+
+	if (plat_dat->interface == PHY_INTERFACE_MODE_RMII) {
+		usleep_range(100, 200);
+		writel(RMII_RESET_SPEED, ioaddr + MAC_CTRL_REG);
+	}
+
+	return readl_poll_timeout(ioaddr + DMA_BUS_MODE, value,
+				 !(value & DMA_BUS_MODE_SFT_RESET),
+				 10000, 1000000);
+}
+
 static int
 imx_dwmac_parse_dt(struct imx_priv_data *dwmac, struct device *dev)
 {
@@ -304,6 +328,8 @@ static int imx_dwmac_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_dwmac_init;
 
+	dwmac->plat_dat->fix_soc_reset = dwmac->ops->fix_soc_reset;
+
 	ret = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res);
 	if (ret)
 		goto err_drv_probe;
@@ -337,6 +363,7 @@ static struct imx_dwmac_ops imx93_dwmac_data = {
 	.addr_width = 32,
 	.mac_rgmii_txclk_auto_adj = true,
 	.set_intf_mode = imx93_set_intf_mode,
+	.fix_soc_reset = imx_dwmac_mx93_reset,
 };
 
 static const struct of_device_id imx_dwmac_match[] = {
-- 
cgit v1.2.3-70-g09d2


From e27f0f1620b693ecf312c21ff8a52affb8a5be08 Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:14 +0100
Subject: bpftool: Fix documentation about line info display for prog dumps

The documentation states that when line_info is available when dumping a
program, the source line will be displayed "by default". There is no
notion of "default" here: the line is always displayed if available,
there is no way currently to turn it off.

In the next sentence, the documentation states that if "linum" is used
on the command line, the relevant filename, line, and column will be
displayed "on top of the source line". This is incorrect, as they are
currently displayed on the right side of the source line (or on top of
the eBPF instruction, not the source).

This commit fixes the documentation to address these points.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-2-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/Documentation/bpftool-prog.rst | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
index 14de72544995..06d1e4314406 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
@@ -106,9 +106,8 @@ DESCRIPTION
 		  CFG in DOT format, on standard output.
 
 		  If the programs have line_info available, the source line will
-		  be displayed by default.  If **linum** is specified,
-		  the filename, line number and line column will also be
-		  displayed on top of the source line.
+		  be displayed.  If **linum** is specified, the filename, line
+		  number and line column will also be displayed.
 
 	**bpftool prog dump jited**  *PROG* [{ **file** *FILE* | **opcodes** | **linum** }]
 		  Dump jited image (host machine code) of the program.
@@ -120,9 +119,8 @@ DESCRIPTION
 		  **opcodes** controls if raw opcodes will be printed.
 
 		  If the prog has line_info available, the source line will
-		  be displayed by default.  If **linum** is specified,
-		  the filename, line number and line column will also be
-		  displayed on top of the source line.
+		  be displayed.  If **linum** is specified, the filename, line
+		  number and line column will also be displayed.
 
 	**bpftool prog pin** *PROG* *FILE*
 		  Pin program *PROG* as *FILE*.
-- 
cgit v1.2.3-70-g09d2


From 67cf52cdb6c8fa6365d29106555dacf95c9fd374 Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:15 +0100
Subject: bpftool: Fix bug for long instructions in program CFG dumps

When dumping the control flow graphs for programs using the 16-byte long
load instruction, we need to skip the second part of this instruction
when looking for the next instruction to process. Otherwise, we end up
printing "BUG_ld_00" from the kernel disassembler in the CFG.

Fixes: efcef17a6d65 ("tools: bpftool: generate .dot graph from CFG information")
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-3-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/xlated_dumper.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/tools/bpf/bpftool/xlated_dumper.c b/tools/bpf/bpftool/xlated_dumper.c
index 6fe3134ae45d..3daa05d9bbb7 100644
--- a/tools/bpf/bpftool/xlated_dumper.c
+++ b/tools/bpf/bpftool/xlated_dumper.c
@@ -372,8 +372,15 @@ void dump_xlated_for_graph(struct dump_data *dd, void *buf_start, void *buf_end,
 	struct bpf_insn *insn_start = buf_start;
 	struct bpf_insn *insn_end = buf_end;
 	struct bpf_insn *cur = insn_start;
+	bool double_insn = false;
 
 	for (; cur <= insn_end; cur++) {
+		if (double_insn) {
+			double_insn = false;
+			continue;
+		}
+		double_insn = cur->code == (BPF_LD | BPF_IMM | BPF_DW);
+
 		printf("% 4d: ", (int)(cur - insn_start + start_idx));
 		print_bpf_insn(&cbs, cur, true);
 		if (cur != insn_end)
-- 
cgit v1.2.3-70-g09d2


From 9fd496848b1c4cd04fee5f152bb7bcec11e1b901 Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:16 +0100
Subject: bpftool: Support inline annotations when dumping the CFG of a program

We support dumping the control flow graph of loaded programs to the DOT
format with bpftool, but so far this feature wouldn't display the source
code lines available through BTF along with the eBPF bytecode. Let's add
support for these annotations, to make it easier to read the graph.

In prog.c, we move the call to dump_xlated_cfg() in order to pass and
use the full struct dump_data, instead of creating a minimal one in
draw_bb_node().

We pass the pointer to this struct down to dump_xlated_for_graph() in
xlated_dumper.c, where most of the logics is added. We deal with BTF
mostly like we do for plain or JSON output, except that we cannot use a
"nr_skip" value to skip a given number of linfo records (we don't
process the BPF instructions linearly, and apart from the root of the
graph we don't know how many records we should skip, so we just store
the last linfo and make sure the new one we find is different before
printing it).

When printing the source instructions to the label of a DOT graph node,
there are a few subtleties to address. We want some special newline
markers, and there are some characters that we must escape. To deal with
them, we introduce a new dedicated function btf_dump_linfo_dotlabel() in
btf_dumper.c. We'll reuse this function in a later commit to format the
filepath, line, and column references as well.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-4-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/btf_dumper.c    | 34 ++++++++++++++++++++++++++++++++++
 tools/bpf/bpftool/cfg.c           | 23 ++++++++++-------------
 tools/bpf/bpftool/cfg.h           |  4 +++-
 tools/bpf/bpftool/main.h          |  2 ++
 tools/bpf/bpftool/prog.c          | 17 +++++++----------
 tools/bpf/bpftool/xlated_dumper.c | 34 ++++++++++++++++++++++++++++++++--
 6 files changed, 88 insertions(+), 26 deletions(-)

diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c
index e7f6ec3a8f35..583aa843df92 100644
--- a/tools/bpf/bpftool/btf_dumper.c
+++ b/tools/bpf/bpftool/btf_dumper.c
@@ -821,3 +821,37 @@ void btf_dump_linfo_json(const struct btf *btf,
 					BPF_LINE_INFO_LINE_COL(linfo->line_col));
 	}
 }
+
+static void dotlabel_puts(const char *s)
+{
+	for (; *s; ++s) {
+		switch (*s) {
+		case '\\':
+		case '"':
+		case '{':
+		case '}':
+		case '<':
+		case '>':
+		case '|':
+		case ' ':
+			putchar('\\');
+			__fallthrough;
+		default:
+			putchar(*s);
+		}
+	}
+}
+
+void btf_dump_linfo_dotlabel(const struct btf *btf,
+			     const struct bpf_line_info *linfo)
+{
+	const char *line = btf__name_by_offset(btf, linfo->line_off);
+
+	if (!line || !strlen(line))
+		return;
+	line = ltrim(line);
+
+	printf("; ");
+	dotlabel_puts(line);
+	printf("\\l\\\n");
+}
diff --git a/tools/bpf/bpftool/cfg.c b/tools/bpf/bpftool/cfg.c
index 1951219a9af7..9fdc1f0cdd6e 100644
--- a/tools/bpf/bpftool/cfg.c
+++ b/tools/bpf/bpftool/cfg.c
@@ -380,7 +380,8 @@ static void cfg_destroy(struct cfg *cfg)
 	}
 }
 
-static void draw_bb_node(struct func_node *func, struct bb_node *bb)
+static void
+draw_bb_node(struct func_node *func, struct bb_node *bb, struct dump_data *dd)
 {
 	const char *shape;
 
@@ -398,13 +399,9 @@ static void draw_bb_node(struct func_node *func, struct bb_node *bb)
 		printf("EXIT");
 	} else {
 		unsigned int start_idx;
-		struct dump_data dd = {};
-
-		printf("{");
-		kernel_syms_load(&dd);
+		printf("{\\\n");
 		start_idx = bb->head - func->start;
-		dump_xlated_for_graph(&dd, bb->head, bb->tail, start_idx);
-		kernel_syms_destroy(&dd);
+		dump_xlated_for_graph(dd, bb->head, bb->tail, start_idx);
 		printf("}");
 	}
 
@@ -430,12 +427,12 @@ static void draw_bb_succ_edges(struct func_node *func, struct bb_node *bb)
 	}
 }
 
-static void func_output_bb_def(struct func_node *func)
+static void func_output_bb_def(struct func_node *func, struct dump_data *dd)
 {
 	struct bb_node *bb;
 
 	list_for_each_entry(bb, &func->bbs, l) {
-		draw_bb_node(func, bb);
+		draw_bb_node(func, bb, dd);
 	}
 }
 
@@ -455,7 +452,7 @@ static void func_output_edges(struct func_node *func)
 	       func_idx, ENTRY_BLOCK_INDEX, func_idx, EXIT_BLOCK_INDEX);
 }
 
-static void cfg_dump(struct cfg *cfg)
+static void cfg_dump(struct cfg *cfg, struct dump_data *dd)
 {
 	struct func_node *func;
 
@@ -463,14 +460,14 @@ static void cfg_dump(struct cfg *cfg)
 	list_for_each_entry(func, &cfg->funcs, l) {
 		printf("subgraph \"cluster_%d\" {\n\tstyle=\"dashed\";\n\tcolor=\"black\";\n\tlabel=\"func_%d ()\";\n",
 		       func->idx, func->idx);
-		func_output_bb_def(func);
+		func_output_bb_def(func, dd);
 		func_output_edges(func);
 		printf("}\n");
 	}
 	printf("}\n");
 }
 
-void dump_xlated_cfg(void *buf, unsigned int len)
+void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len)
 {
 	struct bpf_insn *insn = buf;
 	struct cfg cfg;
@@ -479,7 +476,7 @@ void dump_xlated_cfg(void *buf, unsigned int len)
 	if (cfg_build(&cfg, insn, len))
 		return;
 
-	cfg_dump(&cfg);
+	cfg_dump(&cfg, dd);
 
 	cfg_destroy(&cfg);
 }
diff --git a/tools/bpf/bpftool/cfg.h b/tools/bpf/bpftool/cfg.h
index e144257ea6d2..909d17e6d4c2 100644
--- a/tools/bpf/bpftool/cfg.h
+++ b/tools/bpf/bpftool/cfg.h
@@ -4,6 +4,8 @@
 #ifndef __BPF_TOOL_CFG_H
 #define __BPF_TOOL_CFG_H
 
-void dump_xlated_cfg(void *buf, unsigned int len);
+#include "xlated_dumper.h"
+
+void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len);
 
 #endif /* __BPF_TOOL_CFG_H */
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 0ef373cef4c7..e9ee514b22d4 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -229,6 +229,8 @@ void btf_dump_linfo_plain(const struct btf *btf,
 			  const char *prefix, bool linum);
 void btf_dump_linfo_json(const struct btf *btf,
 			 const struct bpf_line_info *linfo, bool linum);
+void btf_dump_linfo_dotlabel(const struct btf *btf,
+			     const struct bpf_line_info *linfo);
 
 struct nlattr;
 struct ifinfomsg;
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index afbe3ec342c8..d855118f0d96 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -840,11 +840,6 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode,
 					      false))
 				goto exit_free;
 		}
-	} else if (visual) {
-		if (json_output)
-			jsonw_null(json_wtr);
-		else
-			dump_xlated_cfg(buf, member_len);
 	} else {
 		kernel_syms_load(&dd);
 		dd.nr_jited_ksyms = info->nr_jited_ksyms;
@@ -854,12 +849,14 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode,
 		dd.finfo_rec_size = info->func_info_rec_size;
 		dd.prog_linfo = prog_linfo;
 
-		if (json_output)
-			dump_xlated_json(&dd, buf, member_len, opcodes,
-					 linum);
+		if (json_output && visual)
+			jsonw_null(json_wtr);
+		else if (json_output)
+			dump_xlated_json(&dd, buf, member_len, opcodes, linum);
+		else if (visual)
+			dump_xlated_cfg(&dd, buf, member_len);
 		else
-			dump_xlated_plain(&dd, buf, member_len, opcodes,
-					  linum);
+			dump_xlated_plain(&dd, buf, member_len, opcodes, linum);
 		kernel_syms_destroy(&dd);
 	}
 
diff --git a/tools/bpf/bpftool/xlated_dumper.c b/tools/bpf/bpftool/xlated_dumper.c
index 3daa05d9bbb7..2ee83561b945 100644
--- a/tools/bpf/bpftool/xlated_dumper.c
+++ b/tools/bpf/bpftool/xlated_dumper.c
@@ -369,21 +369,51 @@ void dump_xlated_for_graph(struct dump_data *dd, void *buf_start, void *buf_end,
 		.cb_imm		= print_imm,
 		.private_data	= dd,
 	};
+	const struct bpf_prog_linfo *prog_linfo = dd->prog_linfo;
+	const struct bpf_line_info *last_linfo = NULL;
+	struct bpf_func_info *record = dd->func_info;
 	struct bpf_insn *insn_start = buf_start;
 	struct bpf_insn *insn_end = buf_end;
 	struct bpf_insn *cur = insn_start;
+	struct btf *btf = dd->btf;
 	bool double_insn = false;
+	char func_sig[1024];
 
 	for (; cur <= insn_end; cur++) {
+		unsigned int insn_off;
+
 		if (double_insn) {
 			double_insn = false;
 			continue;
 		}
 		double_insn = cur->code == (BPF_LD | BPF_IMM | BPF_DW);
 
-		printf("% 4d: ", (int)(cur - insn_start + start_idx));
+		insn_off = (unsigned int)(cur - insn_start + start_idx);
+		if (btf && record) {
+			if (record->insn_off == insn_off) {
+				btf_dumper_type_only(btf, record->type_id,
+						     func_sig,
+						     sizeof(func_sig));
+				if (func_sig[0] != '\0')
+					printf("; %s:\\l\\\n", func_sig);
+				record = (void *)record + dd->finfo_rec_size;
+			}
+		}
+
+		if (prog_linfo) {
+			const struct bpf_line_info *linfo;
+
+			linfo = bpf_prog_linfo__lfind(prog_linfo, insn_off, 0);
+			if (linfo && linfo != last_linfo) {
+				btf_dump_linfo_dotlabel(btf, linfo);
+				last_linfo = linfo;
+			}
+		}
+
+		printf("%d: ", insn_off);
 		print_bpf_insn(&cbs, cur, true);
+
 		if (cur != insn_end)
-			printf(" | ");
+			printf("| ");
 	}
 }
-- 
cgit v1.2.3-70-g09d2


From 05a06be722896e51f65dbbb6a3610f85a8353d6b Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:17 +0100
Subject: bpftool: Return an error on prog dumps if both CFG and JSON are
 required

We do not support JSON output for control flow graphs of programs with
bpftool. So far, requiring both the CFG and JSON output would result in
producing a null JSON object. It makes more sense to raise an error
directly when parsing command line arguments and options, so that users
know they won't get any output they might expect.

If JSON is required for the graph, we leave it to Graphviz instead:

    # bpftool prog dump xlated <REF> visual | dot -Tjson

Suggested-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-5-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/bash-completion/bpftool | 9 ++++++---
 tools/bpf/bpftool/prog.c                  | 8 +++++---
 2 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 35f26f7c1124..a3cb07172789 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -255,16 +255,19 @@ _bpftool_map_update_get_name()
 
 _bpftool()
 {
-    local cur prev words objword
+    local cur prev words objword json=0
     _init_completion || return
 
     # Deal with options
     if [[ ${words[cword]} == -* ]]; then
         local c='--version --json --pretty --bpffs --mapcompat --debug \
-	       --use-loader --base-btf'
+            --use-loader --base-btf'
         COMPREPLY=( $( compgen -W "$c" -- "$cur" ) )
         return 0
     fi
+    if _bpftool_search_list -j --json -p --pretty; then
+        json=1
+    fi
 
     # Deal with simplest keywords
     case $prev in
@@ -367,7 +370,7 @@ _bpftool()
                             ;;
                         *)
                             _bpftool_once_attr 'file'
-                            if _bpftool_search_list 'xlated'; then
+                            if _bpftool_search_list 'xlated' && [[ "$json" == 0 ]]; then
                                 COMPREPLY+=( $( compgen -W 'opcodes visual linum' -- \
                                     "$cur" ) )
                             else
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index d855118f0d96..736defc6e5d0 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -849,9 +849,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode,
 		dd.finfo_rec_size = info->func_info_rec_size;
 		dd.prog_linfo = prog_linfo;
 
-		if (json_output && visual)
-			jsonw_null(json_wtr);
-		else if (json_output)
+		if (json_output)
 			dump_xlated_json(&dd, buf, member_len, opcodes, linum);
 		else if (visual)
 			dump_xlated_cfg(&dd, buf, member_len);
@@ -940,6 +938,10 @@ static int do_dump(int argc, char **argv)
 		usage();
 		goto exit_close;
 	}
+	if (json_output && visual) {
+		p_err("'visual' is not compatible with JSON output");
+		goto exit_close;
+	}
 
 	if (json_output && nb_fds > 1)
 		jsonw_start_array(json_wtr);	/* root array */
-- 
cgit v1.2.3-70-g09d2


From 9b79f02722bbf24f060b2ab79513ad6e22c8e2f0 Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:18 +0100
Subject: bpftool: Support "opcodes", "linum", "visual" simultaneously

When dumping a program, the keywords "opcodes" (for printing the raw
opcodes), "linum" (for displaying the filename, line number, column
number along with the source code), and "visual" (for generating the
control flow graph for translated programs) are mutually exclusive. But
there's no reason why they should be. Let's make it possible to pass
several of them at once. The "file FILE" option, which makes bpftool
output a binary image to a file, remains incompatible with the others.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-6-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/Documentation/bpftool-prog.rst |  8 ++--
 tools/bpf/bpftool/bash-completion/bpftool        | 17 ++++---
 tools/bpf/bpftool/prog.c                         | 61 +++++++++++++-----------
 3 files changed, 47 insertions(+), 39 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
index 06d1e4314406..9443c524bb76 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
@@ -28,8 +28,8 @@ PROG COMMANDS
 =============
 
 |	**bpftool** **prog** { **show** | **list** } [*PROG*]
-|	**bpftool** **prog dump xlated** *PROG* [{**file** *FILE* | **opcodes** | **visual** | **linum**}]
-|	**bpftool** **prog dump jited**  *PROG* [{**file** *FILE* | **opcodes** | **linum**}]
+|	**bpftool** **prog dump xlated** *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] [**visual**] }]
+|	**bpftool** **prog dump jited**  *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] }]
 |	**bpftool** **prog pin** *PROG* *FILE*
 |	**bpftool** **prog** { **load** | **loadall** } *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] [**autoattach**]
 |	**bpftool** **prog attach** *PROG* *ATTACH_TYPE* [*MAP*]
@@ -88,7 +88,7 @@ DESCRIPTION
 		  programs. On such kernels bpftool will automatically emit this
 		  information as well.
 
-	**bpftool prog dump xlated** *PROG* [{ **file** *FILE* | **opcodes** | **visual** | **linum** }]
+	**bpftool prog dump xlated** *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] [**visual**] }]
 		  Dump eBPF instructions of the programs from the kernel. By
 		  default, eBPF will be disassembled and printed to standard
 		  output in human-readable format. In this case, **opcodes**
@@ -109,7 +109,7 @@ DESCRIPTION
 		  be displayed.  If **linum** is specified, the filename, line
 		  number and line column will also be displayed.
 
-	**bpftool prog dump jited**  *PROG* [{ **file** *FILE* | **opcodes** | **linum** }]
+	**bpftool prog dump jited**  *PROG* [{ **file** *FILE* | [**opcodes**] [**linum**] }]
 		  Dump jited image (host machine code) of the program.
 
 		  If *FILE* is specified image will be written to a file,
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index a3cb07172789..69c64dc18b1d 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -271,7 +271,7 @@ _bpftool()
 
     # Deal with simplest keywords
     case $prev in
-        help|hex|opcodes|visual|linum)
+        help|hex)
             return 0
             ;;
         tag)
@@ -369,13 +369,16 @@ _bpftool()
                             return 0
                             ;;
                         *)
-                            _bpftool_once_attr 'file'
+                            # "file" is not compatible with other keywords here
+                            if _bpftool_search_list 'file'; then
+                                return 0
+                            fi
+                            if ! _bpftool_search_list 'linum opcodes visual'; then
+                                _bpftool_once_attr 'file'
+                            fi
+                            _bpftool_once_attr 'linum opcodes'
                             if _bpftool_search_list 'xlated' && [[ "$json" == 0 ]]; then
-                                COMPREPLY+=( $( compgen -W 'opcodes visual linum' -- \
-                                    "$cur" ) )
-                            else
-                                COMPREPLY+=( $( compgen -W 'opcodes linum' -- \
-                                    "$cur" ) )
+                                _bpftool_once_attr 'visual'
                             fi
                             return 0
                             ;;
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 736defc6e5d0..092525a6933a 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -905,37 +905,42 @@ static int do_dump(int argc, char **argv)
 	if (nb_fds < 1)
 		goto exit_free;
 
-	if (is_prefix(*argv, "file")) {
-		NEXT_ARG();
-		if (!argc) {
-			p_err("expected file path");
-			goto exit_close;
-		}
-		if (nb_fds > 1) {
-			p_err("several programs matched");
-			goto exit_close;
-		}
+	while (argc) {
+		if (is_prefix(*argv, "file")) {
+			NEXT_ARG();
+			if (!argc) {
+				p_err("expected file path");
+				goto exit_close;
+			}
+			if (nb_fds > 1) {
+				p_err("several programs matched");
+				goto exit_close;
+			}
 
-		filepath = *argv;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "opcodes")) {
-		opcodes = true;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "visual")) {
-		if (nb_fds > 1) {
-			p_err("several programs matched");
+			filepath = *argv;
+			NEXT_ARG();
+		} else if (is_prefix(*argv, "opcodes")) {
+			opcodes = true;
+			NEXT_ARG();
+		} else if (is_prefix(*argv, "visual")) {
+			if (nb_fds > 1) {
+				p_err("several programs matched");
+				goto exit_close;
+			}
+
+			visual = true;
+			NEXT_ARG();
+		} else if (is_prefix(*argv, "linum")) {
+			linum = true;
+			NEXT_ARG();
+		} else {
+			usage();
 			goto exit_close;
 		}
-
-		visual = true;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "linum")) {
-		linum = true;
-		NEXT_ARG();
 	}
 
-	if (argc) {
-		usage();
+	if (filepath && (opcodes || visual || linum)) {
+		p_err("'file' is not compatible with 'opcodes', 'visual', or 'linum'");
 		goto exit_close;
 	}
 	if (json_output && visual) {
@@ -2419,8 +2424,8 @@ static int do_help(int argc, char **argv)
 
 	fprintf(stderr,
 		"Usage: %1$s %2$s { show | list } [PROG]\n"
-		"       %1$s %2$s dump xlated PROG [{ file FILE | opcodes | visual | linum }]\n"
-		"       %1$s %2$s dump jited  PROG [{ file FILE | opcodes | linum }]\n"
+		"       %1$s %2$s dump xlated PROG [{ file FILE | [opcodes] [linum] [visual] }]\n"
+		"       %1$s %2$s dump jited  PROG [{ file FILE | [opcodes] [linum] }]\n"
 		"       %1$s %2$s pin   PROG FILE\n"
 		"       %1$s %2$s { load | loadall } OBJ  PATH \\\n"
 		"                         [type TYPE] [dev NAME] \\\n"
-- 
cgit v1.2.3-70-g09d2


From 7483a7a70a12d7c00c9f80574d533b01689d39a7 Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:19 +0100
Subject: bpftool: Support printing opcodes and source file references in CFG

Add support for displaying opcodes or/and file references (filepath,
line and column numbers) when dumping the control flow graphs of loaded
BPF programs with bpftool.

The filepaths in the records are absolute. To avoid blocks on the graph
to get too wide, we truncate them when they get too long (but we always
keep the entire file name). In the unlikely case where the resulting
file name is ambiguous, it remains possible to get the full path with a
regular dump (no CFG).

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-7-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/btf_dumper.c    | 51 ++++++++++++++++++++++++++++++++++++++-
 tools/bpf/bpftool/cfg.c           | 22 +++++++++++------
 tools/bpf/bpftool/cfg.h           |  3 ++-
 tools/bpf/bpftool/main.h          |  2 +-
 tools/bpf/bpftool/prog.c          |  2 +-
 tools/bpf/bpftool/xlated_dumper.c | 15 ++++++++++--
 tools/bpf/bpftool/xlated_dumper.h |  3 ++-
 7 files changed, 83 insertions(+), 15 deletions(-)

diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c
index 583aa843df92..6c5e0e82da22 100644
--- a/tools/bpf/bpftool/btf_dumper.c
+++ b/tools/bpf/bpftool/btf_dumper.c
@@ -842,8 +842,37 @@ static void dotlabel_puts(const char *s)
 	}
 }
 
+static const char *shorten_path(const char *path)
+{
+	const unsigned int MAX_PATH_LEN = 32;
+	size_t len = strlen(path);
+	const char *shortpath;
+
+	if (len <= MAX_PATH_LEN)
+		return path;
+
+	/* Search for last '/' under the MAX_PATH_LEN limit */
+	shortpath = strchr(path + len - MAX_PATH_LEN, '/');
+	if (shortpath) {
+		if (shortpath < path + strlen("..."))
+			/* We removed a very short prefix, e.g. "/w", and we'll
+			 * make the path longer by prefixing with the ellipsis.
+			 * Not worth it, keep initial path.
+			 */
+			return path;
+		return shortpath;
+	}
+
+	/* File base name length is > MAX_PATH_LEN, search for last '/' */
+			shortpath = strrchr(path, '/');
+	if (shortpath)
+		return shortpath;
+
+	return path;
+}
+
 void btf_dump_linfo_dotlabel(const struct btf *btf,
-			     const struct bpf_line_info *linfo)
+			     const struct bpf_line_info *linfo, bool linum)
 {
 	const char *line = btf__name_by_offset(btf, linfo->line_off);
 
@@ -851,6 +880,26 @@ void btf_dump_linfo_dotlabel(const struct btf *btf,
 		return;
 	line = ltrim(line);
 
+	if (linum) {
+		const char *file = btf__name_by_offset(btf, linfo->file_name_off);
+		const char *shortfile;
+
+		/* More forgiving on file because linum option is
+		 * expected to provide more info than the already
+		 * available src line.
+		 */
+		if (!file)
+			shortfile = "";
+		else
+			shortfile = shorten_path(file);
+
+		printf("; [%s", shortfile > file ? "..." : "");
+		dotlabel_puts(shortfile);
+		printf(" line:%u col:%u]\\l\\\n",
+		       BPF_LINE_INFO_LINE_NUM(linfo->line_col),
+		       BPF_LINE_INFO_LINE_COL(linfo->line_col));
+	}
+
 	printf("; ");
 	dotlabel_puts(line);
 	printf("\\l\\\n");
diff --git a/tools/bpf/bpftool/cfg.c b/tools/bpf/bpftool/cfg.c
index 9fdc1f0cdd6e..eec437cca2ea 100644
--- a/tools/bpf/bpftool/cfg.c
+++ b/tools/bpf/bpftool/cfg.c
@@ -381,7 +381,8 @@ static void cfg_destroy(struct cfg *cfg)
 }
 
 static void
-draw_bb_node(struct func_node *func, struct bb_node *bb, struct dump_data *dd)
+draw_bb_node(struct func_node *func, struct bb_node *bb, struct dump_data *dd,
+	     bool opcodes, bool linum)
 {
 	const char *shape;
 
@@ -401,7 +402,8 @@ draw_bb_node(struct func_node *func, struct bb_node *bb, struct dump_data *dd)
 		unsigned int start_idx;
 		printf("{\\\n");
 		start_idx = bb->head - func->start;
-		dump_xlated_for_graph(dd, bb->head, bb->tail, start_idx);
+		dump_xlated_for_graph(dd, bb->head, bb->tail, start_idx,
+				      opcodes, linum);
 		printf("}");
 	}
 
@@ -427,12 +429,14 @@ static void draw_bb_succ_edges(struct func_node *func, struct bb_node *bb)
 	}
 }
 
-static void func_output_bb_def(struct func_node *func, struct dump_data *dd)
+static void
+func_output_bb_def(struct func_node *func, struct dump_data *dd,
+		   bool opcodes, bool linum)
 {
 	struct bb_node *bb;
 
 	list_for_each_entry(bb, &func->bbs, l) {
-		draw_bb_node(func, bb, dd);
+		draw_bb_node(func, bb, dd, opcodes, linum);
 	}
 }
 
@@ -452,7 +456,8 @@ static void func_output_edges(struct func_node *func)
 	       func_idx, ENTRY_BLOCK_INDEX, func_idx, EXIT_BLOCK_INDEX);
 }
 
-static void cfg_dump(struct cfg *cfg, struct dump_data *dd)
+static void
+cfg_dump(struct cfg *cfg, struct dump_data *dd, bool opcodes, bool linum)
 {
 	struct func_node *func;
 
@@ -460,14 +465,15 @@ static void cfg_dump(struct cfg *cfg, struct dump_data *dd)
 	list_for_each_entry(func, &cfg->funcs, l) {
 		printf("subgraph \"cluster_%d\" {\n\tstyle=\"dashed\";\n\tcolor=\"black\";\n\tlabel=\"func_%d ()\";\n",
 		       func->idx, func->idx);
-		func_output_bb_def(func, dd);
+		func_output_bb_def(func, dd, opcodes, linum);
 		func_output_edges(func);
 		printf("}\n");
 	}
 	printf("}\n");
 }
 
-void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len)
+void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len,
+		     bool opcodes, bool linum)
 {
 	struct bpf_insn *insn = buf;
 	struct cfg cfg;
@@ -476,7 +482,7 @@ void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len)
 	if (cfg_build(&cfg, insn, len))
 		return;
 
-	cfg_dump(&cfg, dd);
+	cfg_dump(&cfg, dd, opcodes, linum);
 
 	cfg_destroy(&cfg);
 }
diff --git a/tools/bpf/bpftool/cfg.h b/tools/bpf/bpftool/cfg.h
index 909d17e6d4c2..b3793f4e1783 100644
--- a/tools/bpf/bpftool/cfg.h
+++ b/tools/bpf/bpftool/cfg.h
@@ -6,6 +6,7 @@
 
 #include "xlated_dumper.h"
 
-void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len);
+void dump_xlated_cfg(struct dump_data *dd, void *buf, unsigned int len,
+		     bool opcodes, bool linum);
 
 #endif /* __BPF_TOOL_CFG_H */
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index e9ee514b22d4..00d11ca6d3f2 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -230,7 +230,7 @@ void btf_dump_linfo_plain(const struct btf *btf,
 void btf_dump_linfo_json(const struct btf *btf,
 			 const struct bpf_line_info *linfo, bool linum);
 void btf_dump_linfo_dotlabel(const struct btf *btf,
-			     const struct bpf_line_info *linfo);
+			     const struct bpf_line_info *linfo, bool linum);
 
 struct nlattr;
 struct ifinfomsg;
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 092525a6933a..430f72306409 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -852,7 +852,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode,
 		if (json_output)
 			dump_xlated_json(&dd, buf, member_len, opcodes, linum);
 		else if (visual)
-			dump_xlated_cfg(&dd, buf, member_len);
+			dump_xlated_cfg(&dd, buf, member_len, opcodes, linum);
 		else
 			dump_xlated_plain(&dd, buf, member_len, opcodes, linum);
 		kernel_syms_destroy(&dd);
diff --git a/tools/bpf/bpftool/xlated_dumper.c b/tools/bpf/bpftool/xlated_dumper.c
index 2ee83561b945..da608e10c843 100644
--- a/tools/bpf/bpftool/xlated_dumper.c
+++ b/tools/bpf/bpftool/xlated_dumper.c
@@ -361,7 +361,8 @@ void dump_xlated_plain(struct dump_data *dd, void *buf, unsigned int len,
 }
 
 void dump_xlated_for_graph(struct dump_data *dd, void *buf_start, void *buf_end,
-			   unsigned int start_idx)
+			   unsigned int start_idx,
+			   bool opcodes, bool linum)
 {
 	const struct bpf_insn_cbs cbs = {
 		.cb_print	= print_insn_for_graph,
@@ -405,7 +406,7 @@ void dump_xlated_for_graph(struct dump_data *dd, void *buf_start, void *buf_end,
 
 			linfo = bpf_prog_linfo__lfind(prog_linfo, insn_off, 0);
 			if (linfo && linfo != last_linfo) {
-				btf_dump_linfo_dotlabel(btf, linfo);
+				btf_dump_linfo_dotlabel(btf, linfo, linum);
 				last_linfo = linfo;
 			}
 		}
@@ -413,6 +414,16 @@ void dump_xlated_for_graph(struct dump_data *dd, void *buf_start, void *buf_end,
 		printf("%d: ", insn_off);
 		print_bpf_insn(&cbs, cur, true);
 
+		if (opcodes) {
+			printf("\\ \\ \\ \\ ");
+			fprint_hex(stdout, cur, 8, " ");
+			if (double_insn && cur <= insn_end - 1) {
+				printf(" ");
+				fprint_hex(stdout, cur + 1, 8, " ");
+			}
+			printf("\\l\\\n");
+		}
+
 		if (cur != insn_end)
 			printf("| ");
 	}
diff --git a/tools/bpf/bpftool/xlated_dumper.h b/tools/bpf/bpftool/xlated_dumper.h
index 54847e174273..9a946377b0e6 100644
--- a/tools/bpf/bpftool/xlated_dumper.h
+++ b/tools/bpf/bpftool/xlated_dumper.h
@@ -34,6 +34,7 @@ void dump_xlated_json(struct dump_data *dd, void *buf, unsigned int len,
 void dump_xlated_plain(struct dump_data *dd, void *buf, unsigned int len,
 		       bool opcodes, bool linum);
 void dump_xlated_for_graph(struct dump_data *dd, void *buf, void *buf_end,
-			   unsigned int start_index);
+			   unsigned int start_index,
+			   bool opcodes, bool linum);
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 7319296855f1bac0e7fb003388f44ecbd4515102 Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Wed, 5 Apr 2023 14:21:20 +0100
Subject: bpftool: Clean up _bpftool_once_attr() calls in bash completion

In bpftool's bash completion file, function _bpftool_once_attr() is able
to process multiple arguments. There are a few locations where this
function is called multiple times in a row, each time for a single
argument; let's pass all arguments instead to minimize the number of
function calls required for the completion.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230405132120.59886-8-quentin@isovalent.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/bash-completion/bpftool | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 69c64dc18b1d..e7234d1a5306 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -508,10 +508,7 @@ _bpftool()
                             ;;
                         *)
                             COMPREPLY=( $( compgen -W "map" -- "$cur" ) )
-                            _bpftool_once_attr 'type'
-                            _bpftool_once_attr 'dev'
-                            _bpftool_once_attr 'pinmaps'
-                            _bpftool_once_attr 'autoattach'
+                            _bpftool_once_attr 'type dev pinmaps autoattach'
                             return 0
                             ;;
                     esac
@@ -736,16 +733,10 @@ _bpftool()
                             esac
                             ;;
                         *)
-                            _bpftool_once_attr 'type'
-                            _bpftool_once_attr 'key'
-                            _bpftool_once_attr 'value'
-                            _bpftool_once_attr 'entries'
-                            _bpftool_once_attr 'name'
-                            _bpftool_once_attr 'flags'
+                            _bpftool_once_attr 'type key value entries name flags dev'
                             if _bpftool_search_list 'array_of_maps' 'hash_of_maps'; then
                                 _bpftool_once_attr 'inner_map'
                             fi
-                            _bpftool_once_attr 'dev'
                             return 0
                             ;;
                     esac
@@ -886,8 +877,7 @@ _bpftool()
                             return 0
                             ;;
                         *)
-                            _bpftool_once_attr 'cpu'
-                            _bpftool_once_attr 'index'
+                            _bpftool_once_attr 'cpu index'
                             return 0
                             ;;
                     esac
-- 
cgit v1.2.3-70-g09d2


From 34bf93472f8fb60b4189aa2872471017e739cf0a Mon Sep 17 00:00:00 2001
From: Viktor Malik <vmalik@redhat.com>
Date: Thu, 30 Mar 2023 12:20:01 +0200
Subject: kallsyms: move module-related functions under correct configs

Functions for searching module kallsyms should have non-empty
definitions only if CONFIG_MODULES=y and CONFIG_KALLSYMS=y. Until now,
only CONFIG_MODULES check was used for many of these, which may have
caused complilation errors on some configs.

This patch moves all relevant functions under the correct configs.

Fixes: bd5314f8dd2d ("kallsyms, bpf: Move find_kallsyms_symbol_value out of internal header")
Signed-off-by: Viktor Malik <vmalik@redhat.com>
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202303181535.RFDCnz3E-lkp@intel.com/
Link: https://lore.kernel.org/r/20230330102001.2183693-1-vmalik@redhat.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/module.h | 135 +++++++++++++++++++++++++++----------------------
 1 file changed, 74 insertions(+), 61 deletions(-)

diff --git a/include/linux/module.h b/include/linux/module.h
index 41cfd3be57e5..886d24877c7c 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -608,16 +608,6 @@ static inline bool within_module(unsigned long addr, const struct module *mod)
 /* Search for module by name: must be in a RCU-sched critical section. */
 struct module *find_module(const char *name);
 
-/* Returns 0 and fills in value, defined and namebuf, or -ERANGE if
-   symnum out of range. */
-int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
-			char *name, char *module_name, int *exported);
-
-/* Look for this name: can be of form module:name. */
-unsigned long module_kallsyms_lookup_name(const char *name);
-
-unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name);
-
 extern void __noreturn __module_put_and_kthread_exit(struct module *mod,
 			long code);
 #define module_put_and_kthread_exit(code) __module_put_and_kthread_exit(THIS_MODULE, code)
@@ -664,17 +654,6 @@ static inline void __module_get(struct module *module)
 /* Dereference module function descriptor */
 void *dereference_module_function_descriptor(struct module *mod, void *ptr);
 
-/* For kallsyms to ask for address resolution.  namebuf should be at
- * least KSYM_NAME_LEN long: a pointer to namebuf is returned if
- * found, otherwise NULL. */
-const char *module_address_lookup(unsigned long addr,
-			    unsigned long *symbolsize,
-			    unsigned long *offset,
-			    char **modname, const unsigned char **modbuildid,
-			    char *namebuf);
-int lookup_module_symbol_name(unsigned long addr, char *symname);
-int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name);
-
 int register_module_notifier(struct notifier_block *nb);
 int unregister_module_notifier(struct notifier_block *nb);
 
@@ -765,45 +744,6 @@ static inline void module_put(struct module *module)
 
 #define module_name(mod) "kernel"
 
-/* For kallsyms to ask for address resolution.  NULL means not found. */
-static inline const char *module_address_lookup(unsigned long addr,
-					  unsigned long *symbolsize,
-					  unsigned long *offset,
-					  char **modname,
-					  const unsigned char **modbuildid,
-					  char *namebuf)
-{
-	return NULL;
-}
-
-static inline int lookup_module_symbol_name(unsigned long addr, char *symname)
-{
-	return -ERANGE;
-}
-
-static inline int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name)
-{
-	return -ERANGE;
-}
-
-static inline int module_get_kallsym(unsigned int symnum, unsigned long *value,
-					char *type, char *name,
-					char *module_name, int *exported)
-{
-	return -ERANGE;
-}
-
-static inline unsigned long module_kallsyms_lookup_name(const char *name)
-{
-	return 0;
-}
-
-static inline unsigned long find_kallsyms_symbol_value(struct module *mod,
-						       const char *name)
-{
-	return 0;
-}
-
 static inline int register_module_notifier(struct notifier_block *nb)
 {
 	/* no events will happen anyway, so this can always succeed */
@@ -899,7 +839,36 @@ int module_kallsyms_on_each_symbol(const char *modname,
 				   int (*fn)(void *, const char *,
 					     struct module *, unsigned long),
 				   void *data);
-#else
+
+/* For kallsyms to ask for address resolution.  namebuf should be at
+ * least KSYM_NAME_LEN long: a pointer to namebuf is returned if
+ * found, otherwise NULL.
+ */
+const char *module_address_lookup(unsigned long addr,
+				  unsigned long *symbolsize,
+				  unsigned long *offset,
+				  char **modname, const unsigned char **modbuildid,
+				  char *namebuf);
+int lookup_module_symbol_name(unsigned long addr, char *symname);
+int lookup_module_symbol_attrs(unsigned long addr,
+			       unsigned long *size,
+			       unsigned long *offset,
+			       char *modname,
+			       char *name);
+
+/* Returns 0 and fills in value, defined and namebuf, or -ERANGE if
+ * symnum out of range.
+ */
+int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
+		       char *name, char *module_name, int *exported);
+
+/* Look for this name: can be of form module:name. */
+unsigned long module_kallsyms_lookup_name(const char *name);
+
+unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name);
+
+#else	/* CONFIG_MODULES && CONFIG_KALLSYMS */
+
 static inline int module_kallsyms_on_each_symbol(const char *modname,
 						 int (*fn)(void *, const char *,
 						 struct module *, unsigned long),
@@ -907,6 +876,50 @@ static inline int module_kallsyms_on_each_symbol(const char *modname,
 {
 	return -EOPNOTSUPP;
 }
+
+/* For kallsyms to ask for address resolution.  NULL means not found. */
+static inline const char *module_address_lookup(unsigned long addr,
+						unsigned long *symbolsize,
+						unsigned long *offset,
+						char **modname,
+						const unsigned char **modbuildid,
+						char *namebuf)
+{
+	return NULL;
+}
+
+static inline int lookup_module_symbol_name(unsigned long addr, char *symname)
+{
+	return -ERANGE;
+}
+
+static inline int lookup_module_symbol_attrs(unsigned long addr,
+					     unsigned long *size,
+					     unsigned long *offset,
+					     char *modname,
+					     char *name)
+{
+	return -ERANGE;
+}
+
+static inline int module_get_kallsym(unsigned int symnum, unsigned long *value,
+				     char *type, char *name,
+				     char *module_name, int *exported)
+{
+	return -ERANGE;
+}
+
+static inline unsigned long module_kallsyms_lookup_name(const char *name)
+{
+	return 0;
+}
+
+static inline unsigned long find_kallsyms_symbol_value(struct module *mod,
+						       const char *name)
+{
+	return 0;
+}
+
 #endif  /* CONFIG_MODULES && CONFIG_KALLSYMS */
 
 #endif /* _LINUX_MODULE_H */
-- 
cgit v1.2.3-70-g09d2


From 3e1c957f9a3b248f47f8b39b607002d948fd17d4 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:26 +0300
Subject: xfrm: don't require advance ESN callback for packet offload

In packet offload mode, the hardware is responsible to manage
replay window and advance ESN. In that mode, there won't any
call to .xdo_dev_state_advance_esn callback.

So relax current check for existence of that callback.

Link: https://lore.kernel.org/r/9f3dfc3fef2cfcd191f0c5eee7cf0aa74e7f7786.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Acked-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 net/xfrm/xfrm_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c
index 95f1436bf6a2..bef28c6187eb 100644
--- a/net/xfrm/xfrm_device.c
+++ b/net/xfrm/xfrm_device.c
@@ -287,7 +287,7 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
 		return (is_packet_offload) ? -EINVAL : 0;
 	}
 
-	if (x->props.flags & XFRM_STATE_ESN &&
+	if (!is_packet_offload && x->props.flags & XFRM_STATE_ESN &&
 	    !dev->xfrmdev_ops->xdo_dev_state_advance_esn) {
 		NL_SET_ERR_MSG(extack, "Device doesn't support offload with ESN");
 		xso->dev = NULL;
-- 
cgit v1.2.3-70-g09d2


From f4979e2667c56f031a3c2e84364dbcde01c2cdf3 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:27 +0300
Subject: net/mlx5e: Remove ESN callbacks if it is not supported

There is no need in implementation of .xdo_dev_state_advance_esn() and
setting work as it will never be called in packet offload mode.

Link: https://lore.kernel.org/r/2fc9fade32e31f03b100d6086a82ad36269349dc.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 51 ++++++++++++++++------
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  1 -
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    |  3 --
 3 files changed, 38 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 3612cdd37b5a..067704307851 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -56,11 +56,6 @@ static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry)
 	u32 seq_bottom = 0;
 	u8 overlap;
 
-	if (!(sa_entry->x->props.flags & XFRM_STATE_ESN)) {
-		sa_entry->esn_state.trigger = 0;
-		return false;
-	}
-
 	replay_esn = sa_entry->x->replay_esn;
 	if (replay_esn->seq >= replay_esn->replay_window)
 		seq_bottom = replay_esn->seq - replay_esn->replay_window + 1;
@@ -70,7 +65,6 @@ static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry)
 	sa_entry->esn_state.esn = xfrm_replay_seqhi(sa_entry->x,
 						    htonl(seq_bottom));
 
-	sa_entry->esn_state.trigger = 1;
 	if (unlikely(overlap && seq_bottom < MLX5E_IPSEC_ESN_SCOPE_MID)) {
 		sa_entry->esn_state.overlap = 0;
 		return true;
@@ -229,7 +223,7 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	aes_gcm->icv_len = x->aead->alg_icv_len;
 
 	/* esn */
-	if (sa_entry->esn_state.trigger) {
+	if (x->props.flags & XFRM_STATE_ESN) {
 		attrs->esn_trigger = true;
 		attrs->esn = sa_entry->esn_state.esn;
 		attrs->esn_overlap = sa_entry->esn_state.overlap;
@@ -394,6 +388,22 @@ static void _update_xfrm_state(struct work_struct *work)
 	mlx5_accel_esp_modify_xfrm(sa_entry, &modify_work->attrs);
 }
 
+static void mlx5e_ipsec_set_esn_ops(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct xfrm_state *x = sa_entry->x;
+
+	if (x->xso.type != XFRM_DEV_OFFLOAD_CRYPTO ||
+	    x->xso.dir != XFRM_DEV_OFFLOAD_OUT)
+		return;
+
+	if (x->props.flags & XFRM_STATE_ESN) {
+		sa_entry->set_iv_op = mlx5e_ipsec_set_iv_esn;
+		return;
+	}
+
+	sa_entry->set_iv_op = mlx5e_ipsec_set_iv;
+}
+
 static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 				struct netlink_ext_ack *extack)
 {
@@ -425,7 +435,8 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 		goto err_xfrm;
 
 	/* check esn */
-	mlx5e_ipsec_update_esn_state(sa_entry);
+	if (x->props.flags & XFRM_STATE_ESN)
+		mlx5e_ipsec_update_esn_state(sa_entry);
 
 	mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &sa_entry->attrs);
 	/* create hw context */
@@ -446,11 +457,17 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 	if (err)
 		goto err_add_rule;
 
-	if (x->xso.dir == XFRM_DEV_OFFLOAD_OUT)
-		sa_entry->set_iv_op = (x->props.flags & XFRM_STATE_ESN) ?
-				mlx5e_ipsec_set_iv_esn : mlx5e_ipsec_set_iv;
+	mlx5e_ipsec_set_esn_ops(sa_entry);
 
-	INIT_WORK(&sa_entry->modify_work.work, _update_xfrm_state);
+	switch (x->xso.type) {
+	case XFRM_DEV_OFFLOAD_CRYPTO:
+		if (x->props.flags & XFRM_STATE_ESN)
+			INIT_WORK(&sa_entry->modify_work.work,
+				  _update_xfrm_state);
+		break;
+	default:
+		break;
+	}
 out:
 	x->xso.offload_handle = (unsigned long)sa_entry;
 	return 0;
@@ -485,7 +502,15 @@ static void mlx5e_xfrm_free_state(struct xfrm_state *x)
 	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
 		goto sa_entry_free;
 
-	cancel_work_sync(&sa_entry->modify_work.work);
+	switch (x->xso.type) {
+	case XFRM_DEV_OFFLOAD_CRYPTO:
+		if (x->props.flags & XFRM_STATE_ESN)
+			cancel_work_sync(&sa_entry->modify_work.work);
+		break;
+	default:
+		break;
+	}
+
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 	mlx5_ipsec_free_sa_ctx(sa_entry);
 sa_entry_free:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 0c58c3583b0f..e4a606364a45 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -160,7 +160,6 @@ struct mlx5e_ipsec {
 
 struct mlx5e_ipsec_esn_state {
 	u32 esn;
-	u8 trigger: 1;
 	u8 overlap: 1;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 6971e5e36820..a2e9af5e51e1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -225,9 +225,6 @@ static int mlx5_modify_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry,
 	void *obj;
 	int err;
 
-	if (!attrs->esn_trigger)
-		return 0;
-
 	general_obj_types = MLX5_CAP_GEN_64(mdev, general_obj_types);
 	if (!(general_obj_types & MLX5_HCA_CAP_GENERAL_OBJECT_TYPES_IPSEC))
 		return -EINVAL;
-- 
cgit v1.2.3-70-g09d2


From 7db21ef4566ef7a24e46808e8668a603531163c3 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:28 +0300
Subject: net/mlx5e: Set IPsec replay sequence numbers

"ip xfrm state ..." command allows users to configure replay sequence
numbers with replay-seq* arguments for RX and replay-oseq* for TX.

Add the needed driver logic to support setting them.

Link: https://lore.kernel.org/r/a9b17827eff2b29a4951225efa684a6cd38f74fe.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 48 +++++++++++++++++-----
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   | 14 +++++--
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    | 22 +++++-----
 3 files changed, 60 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 067704307851..b8058f89365e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -52,18 +52,46 @@ static struct mlx5e_ipsec_pol_entry *to_ipsec_pol_entry(struct xfrm_policy *x)
 
 static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry)
 {
-	struct xfrm_replay_state_esn *replay_esn;
+	struct xfrm_state *x = sa_entry->x;
 	u32 seq_bottom = 0;
+	u32 esn, esn_msb;
 	u8 overlap;
 
-	replay_esn = sa_entry->x->replay_esn;
-	if (replay_esn->seq >= replay_esn->replay_window)
-		seq_bottom = replay_esn->seq - replay_esn->replay_window + 1;
+	switch (x->xso.type) {
+	case XFRM_DEV_OFFLOAD_PACKET:
+		switch (x->xso.dir) {
+		case XFRM_DEV_OFFLOAD_IN:
+			esn = x->replay_esn->seq;
+			esn_msb = x->replay_esn->seq_hi;
+			break;
+		case XFRM_DEV_OFFLOAD_OUT:
+			esn = x->replay_esn->oseq;
+			esn_msb = x->replay_esn->oseq_hi;
+			break;
+		default:
+			WARN_ON(true);
+			return false;
+		}
+		break;
+	case XFRM_DEV_OFFLOAD_CRYPTO:
+		/* Already parsed by XFRM core */
+		esn = x->replay_esn->seq;
+		break;
+	default:
+		WARN_ON(true);
+		return false;
+	}
 
 	overlap = sa_entry->esn_state.overlap;
 
-	sa_entry->esn_state.esn = xfrm_replay_seqhi(sa_entry->x,
-						    htonl(seq_bottom));
+	if (esn >= x->replay_esn->replay_window)
+		seq_bottom = esn - x->replay_esn->replay_window + 1;
+
+	if (x->xso.type == XFRM_DEV_OFFLOAD_CRYPTO)
+		esn_msb = xfrm_replay_seqhi(x, htonl(seq_bottom));
+
+	sa_entry->esn_state.esn = esn;
+	sa_entry->esn_state.esn_msb = esn_msb;
 
 	if (unlikely(overlap && seq_bottom < MLX5E_IPSEC_ESN_SCOPE_MID)) {
 		sa_entry->esn_state.overlap = 0;
@@ -224,10 +252,10 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 
 	/* esn */
 	if (x->props.flags & XFRM_STATE_ESN) {
-		attrs->esn_trigger = true;
-		attrs->esn = sa_entry->esn_state.esn;
-		attrs->esn_overlap = sa_entry->esn_state.overlap;
-		attrs->replay_window = x->replay_esn->replay_window;
+		attrs->replay_esn.trigger = true;
+		attrs->replay_esn.esn = sa_entry->esn_state.esn;
+		attrs->replay_esn.esn_msb = sa_entry->esn_state.esn_msb;
+		attrs->replay_esn.overlap = sa_entry->esn_state.overlap;
 	}
 
 	attrs->dir = x->xso.dir;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index e4a606364a45..8d5ce65def9f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -67,8 +67,15 @@ struct mlx5_ipsec_lft {
 	u64 numb_rounds_soft;
 };
 
+struct mlx5_replay_esn {
+	u32 replay_window;
+	u32 esn;
+	u32 esn_msb;
+	u8 overlap : 1;
+	u8 trigger : 1;
+};
+
 struct mlx5_accel_esp_xfrm_attrs {
-	u32   esn;
 	u32   spi;
 	u32   flags;
 	struct aes_gcm_keymat aes_gcm;
@@ -85,11 +92,9 @@ struct mlx5_accel_esp_xfrm_attrs {
 
 	struct upspec upspec;
 	u8 dir : 2;
-	u8 esn_overlap : 1;
-	u8 esn_trigger : 1;
 	u8 type : 2;
 	u8 family;
-	u32 replay_window;
+	struct mlx5_replay_esn replay_esn;
 	u32 authsize;
 	u32 reqid;
 	struct mlx5_ipsec_lft lft;
@@ -160,6 +165,7 @@ struct mlx5e_ipsec {
 
 struct mlx5e_ipsec_esn_state {
 	u32 esn;
+	u32 esn_msb;
 	u8 overlap: 1;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index a2e9af5e51e1..c974c6153d89 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -76,15 +76,17 @@ static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn,
 	void *aso_ctx;
 
 	aso_ctx = MLX5_ADDR_OF(ipsec_obj, obj, ipsec_aso);
-	if (attrs->esn_trigger) {
+	if (attrs->replay_esn.trigger) {
 		MLX5_SET(ipsec_aso, aso_ctx, esn_event_arm, 1);
 
 		if (attrs->dir == XFRM_DEV_OFFLOAD_IN) {
 			MLX5_SET(ipsec_aso, aso_ctx, window_sz,
-				 attrs->replay_window / 64);
+				 attrs->replay_esn.replay_window / 64);
 			MLX5_SET(ipsec_aso, aso_ctx, mode,
 				 MLX5_IPSEC_ASO_REPLAY_PROTECTION);
-			}
+		}
+		MLX5_SET(ipsec_aso, aso_ctx, mode_parameter,
+			 attrs->replay_esn.esn);
 	}
 
 	/* ASO context */
@@ -136,10 +138,10 @@ static int mlx5_create_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry)
 	salt_iv_p = MLX5_ADDR_OF(ipsec_obj, obj, implicit_iv);
 	memcpy(salt_iv_p, &aes_gcm->seq_iv, sizeof(aes_gcm->seq_iv));
 	/* esn */
-	if (attrs->esn_trigger) {
+	if (attrs->replay_esn.trigger) {
 		MLX5_SET(ipsec_obj, obj, esn_en, 1);
-		MLX5_SET(ipsec_obj, obj, esn_msb, attrs->esn);
-		MLX5_SET(ipsec_obj, obj, esn_overlap, attrs->esn_overlap);
+		MLX5_SET(ipsec_obj, obj, esn_msb, attrs->replay_esn.esn_msb);
+		MLX5_SET(ipsec_obj, obj, esn_overlap, attrs->replay_esn.overlap);
 	}
 
 	MLX5_SET(ipsec_obj, obj, dekn, sa_entry->enc_key_id);
@@ -252,8 +254,8 @@ static int mlx5_modify_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry,
 	MLX5_SET64(ipsec_obj, obj, modify_field_select,
 		   MLX5_MODIFY_IPSEC_BITMASK_ESN_OVERLAP |
 			   MLX5_MODIFY_IPSEC_BITMASK_ESN_MSB);
-	MLX5_SET(ipsec_obj, obj, esn_msb, attrs->esn);
-	MLX5_SET(ipsec_obj, obj, esn_overlap, attrs->esn_overlap);
+	MLX5_SET(ipsec_obj, obj, esn_msb, attrs->replay_esn.esn_msb);
+	MLX5_SET(ipsec_obj, obj, esn_overlap, attrs->replay_esn.overlap);
 
 	/* general object fields set */
 	MLX5_SET(general_obj_in_cmd_hdr, in, opcode, MLX5_CMD_OP_MODIFY_GENERAL_OBJECT);
@@ -290,7 +292,7 @@ static void mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry,
 	struct mlx5_wqe_aso_ctrl_seg data = {};
 
 	if (mode_param < MLX5E_IPSEC_ESN_SCOPE_MID) {
-		sa_entry->esn_state.esn++;
+		sa_entry->esn_state.esn_msb++;
 		sa_entry->esn_state.overlap = 0;
 	} else {
 		sa_entry->esn_state.overlap = 1;
@@ -434,7 +436,7 @@ static void mlx5e_ipsec_handle_event(struct work_struct *_work)
 	if (ret)
 		goto unlock;
 
-	if (attrs->esn_trigger &&
+	if (attrs->replay_esn.trigger &&
 	    !MLX5_GET(ipsec_aso, aso->ctx, esn_event_arm)) {
 		u32 mode_param = MLX5_GET(ipsec_aso, aso->ctx, mode_parameter);
 
-- 
cgit v1.2.3-70-g09d2


From 20fbdab21e2e21bc1e9f0a0dd736c087a9fe0382 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:29 +0300
Subject: net/mlx5e: Reduce contention in IPsec workqueue

IPsec workqueue shouldn't be declared as ordered queue with one work
per-CPU, and can be safely changed to be unordered with default number
of works per-CPU.

Link: https://lore.kernel.org/r/5dc224a4decd09c14f645d38173e1a1710802cd8.1680162300.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index b8058f89365e..fa66f4f3cba7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -561,8 +561,8 @@ void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 
 	xa_init_flags(&ipsec->sadb, XA_FLAGS_ALLOC);
 	ipsec->mdev = priv->mdev;
-	ipsec->wq = alloc_ordered_workqueue("mlx5e_ipsec: %s", 0,
-					    priv->netdev->name);
+	ipsec->wq = alloc_workqueue("mlx5e_ipsec: %s", WQ_UNBOUND, 0,
+				    priv->netdev->name);
 	if (!ipsec->wq)
 		goto err_wq;
 
-- 
cgit v1.2.3-70-g09d2


From 4562116f8a565fdeadb0af5503cb7edb8e9d5761 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:30 +0300
Subject: net/mlx5e: Generalize IPsec work structs

IPsec logic has two work structs which are submitted to same workqueue.
As a preparation to addition of new work which needs to be submitted
too, let's generalize struct mlx5e_ipsec_work.

Link: https://lore.kernel.org/r/285a1550242363de181bab3a07a69296f66ad9a8.1680162300.git.leonro@nvidia.com
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 86 ++++++++++++++--------
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   | 11 +--
 .../mellanox/mlx5/core/en_accel/ipsec_offload.c    | 19 ++---
 3 files changed, 68 insertions(+), 48 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index fa66f4f3cba7..ef7f942b5ac7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -406,14 +406,16 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 	return 0;
 }
 
-static void _update_xfrm_state(struct work_struct *work)
+static void mlx5e_ipsec_modify_state(struct work_struct *_work)
 {
-	struct mlx5e_ipsec_modify_state_work *modify_work =
-		container_of(work, struct mlx5e_ipsec_modify_state_work, work);
-	struct mlx5e_ipsec_sa_entry *sa_entry = container_of(
-		modify_work, struct mlx5e_ipsec_sa_entry, modify_work);
+	struct mlx5e_ipsec_work *work =
+		container_of(_work, struct mlx5e_ipsec_work, work);
+	struct mlx5e_ipsec_sa_entry *sa_entry = work->sa_entry;
+	struct mlx5_accel_esp_xfrm_attrs *attrs;
 
-	mlx5_accel_esp_modify_xfrm(sa_entry, &modify_work->attrs);
+	attrs = &((struct mlx5e_ipsec_sa_entry *)work->data)->attrs;
+
+	mlx5_accel_esp_modify_xfrm(sa_entry, attrs);
 }
 
 static void mlx5e_ipsec_set_esn_ops(struct mlx5e_ipsec_sa_entry *sa_entry)
@@ -432,6 +434,36 @@ static void mlx5e_ipsec_set_esn_ops(struct mlx5e_ipsec_sa_entry *sa_entry)
 	sa_entry->set_iv_op = mlx5e_ipsec_set_iv;
 }
 
+static int mlx5_ipsec_create_work(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct xfrm_state *x = sa_entry->x;
+	struct mlx5e_ipsec_work *work;
+
+	switch (x->xso.type) {
+	case XFRM_DEV_OFFLOAD_CRYPTO:
+		if (!(x->props.flags & XFRM_STATE_ESN))
+			return 0;
+		break;
+	default:
+		return 0;
+	}
+
+	work = kzalloc(sizeof(*work), GFP_KERNEL);
+	if (!work)
+		return -ENOMEM;
+
+	work->data = kzalloc(sizeof(*sa_entry), GFP_KERNEL);
+	if (!work->data) {
+		kfree(work);
+		return -ENOMEM;
+	}
+
+	INIT_WORK(&work->work, mlx5e_ipsec_modify_state);
+	work->sa_entry = sa_entry;
+	sa_entry->work = work;
+	return 0;
+}
+
 static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 				struct netlink_ext_ack *extack)
 {
@@ -467,10 +499,15 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 		mlx5e_ipsec_update_esn_state(sa_entry);
 
 	mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &sa_entry->attrs);
+
+	err = mlx5_ipsec_create_work(sa_entry);
+	if (err)
+		goto err_xfrm;
+
 	/* create hw context */
 	err = mlx5_ipsec_create_sa_ctx(sa_entry);
 	if (err)
-		goto err_xfrm;
+		goto release_work;
 
 	err = mlx5e_accel_ipsec_fs_add_rule(sa_entry);
 	if (err)
@@ -486,16 +523,6 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 		goto err_add_rule;
 
 	mlx5e_ipsec_set_esn_ops(sa_entry);
-
-	switch (x->xso.type) {
-	case XFRM_DEV_OFFLOAD_CRYPTO:
-		if (x->props.flags & XFRM_STATE_ESN)
-			INIT_WORK(&sa_entry->modify_work.work,
-				  _update_xfrm_state);
-		break;
-	default:
-		break;
-	}
 out:
 	x->xso.offload_handle = (unsigned long)sa_entry;
 	return 0;
@@ -504,6 +531,9 @@ err_add_rule:
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 err_hw_ctx:
 	mlx5_ipsec_free_sa_ctx(sa_entry);
+release_work:
+	kfree(sa_entry->work->data);
+	kfree(sa_entry->work);
 err_xfrm:
 	kfree(sa_entry);
 	NL_SET_ERR_MSG_MOD(extack, "Device failed to offload this policy");
@@ -530,17 +560,13 @@ static void mlx5e_xfrm_free_state(struct xfrm_state *x)
 	if (x->xso.flags & XFRM_DEV_OFFLOAD_FLAG_ACQ)
 		goto sa_entry_free;
 
-	switch (x->xso.type) {
-	case XFRM_DEV_OFFLOAD_CRYPTO:
-		if (x->props.flags & XFRM_STATE_ESN)
-			cancel_work_sync(&sa_entry->modify_work.work);
-		break;
-	default:
-		break;
-	}
+	if (sa_entry->work)
+		cancel_work_sync(&sa_entry->work->work);
 
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 	mlx5_ipsec_free_sa_ctx(sa_entry);
+	kfree(sa_entry->work->data);
+	kfree(sa_entry->work);
 sa_entry_free:
 	kfree(sa_entry);
 }
@@ -626,16 +652,18 @@ static bool mlx5e_ipsec_offload_ok(struct sk_buff *skb, struct xfrm_state *x)
 static void mlx5e_xfrm_advance_esn_state(struct xfrm_state *x)
 {
 	struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x);
-	struct mlx5e_ipsec_modify_state_work *modify_work =
-		&sa_entry->modify_work;
+	struct mlx5e_ipsec_work *work = sa_entry->work;
+	struct mlx5e_ipsec_sa_entry *sa_entry_shadow;
 	bool need_update;
 
 	need_update = mlx5e_ipsec_update_esn_state(sa_entry);
 	if (!need_update)
 		return;
 
-	mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &modify_work->attrs);
-	queue_work(sa_entry->ipsec->wq, &modify_work->work);
+	sa_entry_shadow = work->data;
+	memset(sa_entry_shadow, 0x00, sizeof(*sa_entry_shadow));
+	mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &sa_entry_shadow->attrs);
+	queue_work(sa_entry->ipsec->wq, &work->work);
 }
 
 static void mlx5e_xfrm_update_curlft(struct xfrm_state *x)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 8d5ce65def9f..ab48fb9b4698 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -136,8 +136,8 @@ struct mlx5e_ipsec_tx;
 
 struct mlx5e_ipsec_work {
 	struct work_struct work;
-	struct mlx5e_ipsec *ipsec;
-	u32 id;
+	struct mlx5e_ipsec_sa_entry *sa_entry;
+	void *data;
 };
 
 struct mlx5e_ipsec_aso {
@@ -176,11 +176,6 @@ struct mlx5e_ipsec_rule {
 	struct mlx5_fc *fc;
 };
 
-struct mlx5e_ipsec_modify_state_work {
-	struct work_struct		work;
-	struct mlx5_accel_esp_xfrm_attrs attrs;
-};
-
 struct mlx5e_ipsec_limits {
 	u64 round;
 	u8 soft_limit_hit : 1;
@@ -197,7 +192,7 @@ struct mlx5e_ipsec_sa_entry {
 	u32 ipsec_obj_id;
 	u32 enc_key_id;
 	struct mlx5e_ipsec_rule ipsec_rule;
-	struct mlx5e_ipsec_modify_state_work modify_work;
+	struct mlx5e_ipsec_work *work;
 	struct mlx5e_ipsec_limits limits;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index c974c6153d89..5fddb86bb35e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -417,18 +417,12 @@ static void mlx5e_ipsec_handle_event(struct work_struct *_work)
 {
 	struct mlx5e_ipsec_work *work =
 		container_of(_work, struct mlx5e_ipsec_work, work);
+	struct mlx5e_ipsec_sa_entry *sa_entry = work->data;
 	struct mlx5_accel_esp_xfrm_attrs *attrs;
-	struct mlx5e_ipsec_sa_entry *sa_entry;
 	struct mlx5e_ipsec_aso *aso;
-	struct mlx5e_ipsec *ipsec;
 	int ret;
 
-	sa_entry = xa_load(&work->ipsec->sadb, work->id);
-	if (!sa_entry)
-		goto out;
-
-	ipsec = sa_entry->ipsec;
-	aso = ipsec->aso;
+	aso = sa_entry->ipsec->aso;
 	attrs = &sa_entry->attrs;
 
 	spin_lock(&sa_entry->x->lock);
@@ -448,7 +442,6 @@ static void mlx5e_ipsec_handle_event(struct work_struct *_work)
 
 unlock:
 	spin_unlock(&sa_entry->x->lock);
-out:
 	kfree(work);
 }
 
@@ -456,6 +449,7 @@ static int mlx5e_ipsec_event(struct notifier_block *nb, unsigned long event,
 			     void *data)
 {
 	struct mlx5e_ipsec *ipsec = container_of(nb, struct mlx5e_ipsec, nb);
+	struct mlx5e_ipsec_sa_entry *sa_entry;
 	struct mlx5_eqe_obj_change *object;
 	struct mlx5e_ipsec_work *work;
 	struct mlx5_eqe *eqe = data;
@@ -470,13 +464,16 @@ static int mlx5e_ipsec_event(struct notifier_block *nb, unsigned long event,
 	if (type != MLX5_GENERAL_OBJECT_TYPES_IPSEC)
 		return NOTIFY_DONE;
 
+	sa_entry = xa_load(&ipsec->sadb, be32_to_cpu(object->obj_id));
+	if (!sa_entry)
+		return NOTIFY_DONE;
+
 	work = kmalloc(sizeof(*work), GFP_ATOMIC);
 	if (!work)
 		return NOTIFY_DONE;
 
 	INIT_WORK(&work->work, mlx5e_ipsec_handle_event);
-	work->ipsec = ipsec;
-	work->id = be32_to_cpu(object->obj_id);
+	work->data = sa_entry;
 
 	queue_work(ipsec->wq, &work->work);
 	return NOTIFY_OK;
-- 
cgit v1.2.3-70-g09d2


From b2f7b01d36a9b94fbd7489bd1228025ea7e7a2f4 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 30 Mar 2023 11:02:31 +0300
Subject: net/mlx5e: Simulate missing IPsec TX limits hardware functionality

ConnectX-7 devices don't have ability to send TX hard/soft limits
events. As a possible workaround, let's rely on existing infrastructure
and use periodic check of cached flow counter. In these periodic checks,
we call to xfrm_state_check_expire() to check and mark state accordingly.

Once the state is marked as XFRM_STATE_EXPIRED, the SA flow rule is
changed to drop all the traffic.

Link: https://lore.kernel.org/r/94a5d82c0c399747117d8a558f9beebfbcf26154.1680162300.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 65 +++++++++++++++++++++-
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  8 +++
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 31 +++++++++--
 3 files changed, 99 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index ef7f942b5ac7..def01bfde610 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -40,6 +40,8 @@
 #include "ipsec.h"
 #include "ipsec_rxtx.h"
 
+#define MLX5_IPSEC_RESCHED msecs_to_jiffies(1000)
+
 static struct mlx5e_ipsec_sa_entry *to_ipsec_sa_entry(struct xfrm_state *x)
 {
 	return (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle;
@@ -50,6 +52,28 @@ static struct mlx5e_ipsec_pol_entry *to_ipsec_pol_entry(struct xfrm_policy *x)
 	return (struct mlx5e_ipsec_pol_entry *)x->xdo.offload_handle;
 }
 
+static void mlx5e_ipsec_handle_tx_limit(struct work_struct *_work)
+{
+	struct mlx5e_ipsec_dwork *dwork =
+		container_of(_work, struct mlx5e_ipsec_dwork, dwork.work);
+	struct mlx5e_ipsec_sa_entry *sa_entry = dwork->sa_entry;
+	struct xfrm_state *x = sa_entry->x;
+
+	spin_lock(&x->lock);
+	xfrm_state_check_expire(x);
+	if (x->km.state == XFRM_STATE_EXPIRED) {
+		sa_entry->attrs.drop = true;
+		mlx5e_accel_ipsec_fs_modify(sa_entry);
+	}
+	spin_unlock(&x->lock);
+
+	if (sa_entry->attrs.drop)
+		return;
+
+	queue_delayed_work(sa_entry->ipsec->wq, &dwork->dwork,
+			   MLX5_IPSEC_RESCHED);
+}
+
 static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry)
 {
 	struct xfrm_state *x = sa_entry->x;
@@ -464,6 +488,31 @@ static int mlx5_ipsec_create_work(struct mlx5e_ipsec_sa_entry *sa_entry)
 	return 0;
 }
 
+static int mlx5e_ipsec_create_dwork(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct xfrm_state *x = sa_entry->x;
+	struct mlx5e_ipsec_dwork *dwork;
+
+	if (x->xso.type != XFRM_DEV_OFFLOAD_PACKET)
+		return 0;
+
+	if (x->xso.dir != XFRM_DEV_OFFLOAD_OUT)
+		return 0;
+
+	if (x->lft.soft_packet_limit == XFRM_INF &&
+	    x->lft.hard_packet_limit == XFRM_INF)
+		return 0;
+
+	dwork = kzalloc(sizeof(*dwork), GFP_KERNEL);
+	if (!dwork)
+		return -ENOMEM;
+
+	dwork->sa_entry = sa_entry;
+	INIT_DELAYED_WORK(&dwork->dwork, mlx5e_ipsec_handle_tx_limit);
+	sa_entry->dwork = dwork;
+	return 0;
+}
+
 static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 				struct netlink_ext_ack *extack)
 {
@@ -504,10 +553,14 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 	if (err)
 		goto err_xfrm;
 
+	err = mlx5e_ipsec_create_dwork(sa_entry);
+	if (err)
+		goto release_work;
+
 	/* create hw context */
 	err = mlx5_ipsec_create_sa_ctx(sa_entry);
 	if (err)
-		goto release_work;
+		goto release_dwork;
 
 	err = mlx5e_accel_ipsec_fs_add_rule(sa_entry);
 	if (err)
@@ -523,6 +576,10 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 		goto err_add_rule;
 
 	mlx5e_ipsec_set_esn_ops(sa_entry);
+
+	if (sa_entry->dwork)
+		queue_delayed_work(ipsec->wq, &sa_entry->dwork->dwork,
+				   MLX5_IPSEC_RESCHED);
 out:
 	x->xso.offload_handle = (unsigned long)sa_entry;
 	return 0;
@@ -531,6 +588,8 @@ err_add_rule:
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 err_hw_ctx:
 	mlx5_ipsec_free_sa_ctx(sa_entry);
+release_dwork:
+	kfree(sa_entry->dwork);
 release_work:
 	kfree(sa_entry->work->data);
 	kfree(sa_entry->work);
@@ -563,8 +622,12 @@ static void mlx5e_xfrm_free_state(struct xfrm_state *x)
 	if (sa_entry->work)
 		cancel_work_sync(&sa_entry->work->work);
 
+	if (sa_entry->dwork)
+		cancel_delayed_work_sync(&sa_entry->dwork->dwork);
+
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 	mlx5_ipsec_free_sa_ctx(sa_entry);
+	kfree(sa_entry->dwork);
 	kfree(sa_entry->work->data);
 	kfree(sa_entry->work);
 sa_entry_free:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index ab48fb9b4698..52890d7dce6b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -93,6 +93,7 @@ struct mlx5_accel_esp_xfrm_attrs {
 	struct upspec upspec;
 	u8 dir : 2;
 	u8 type : 2;
+	u8 drop : 1;
 	u8 family;
 	struct mlx5_replay_esn replay_esn;
 	u32 authsize;
@@ -140,6 +141,11 @@ struct mlx5e_ipsec_work {
 	void *data;
 };
 
+struct mlx5e_ipsec_dwork {
+	struct delayed_work dwork;
+	struct mlx5e_ipsec_sa_entry *sa_entry;
+};
+
 struct mlx5e_ipsec_aso {
 	u8 __aligned(64) ctx[MLX5_ST_SZ_BYTES(ipsec_aso)];
 	dma_addr_t dma_addr;
@@ -193,6 +199,7 @@ struct mlx5e_ipsec_sa_entry {
 	u32 enc_key_id;
 	struct mlx5e_ipsec_rule ipsec_rule;
 	struct mlx5e_ipsec_work *work;
+	struct mlx5e_ipsec_dwork *dwork;
 	struct mlx5e_ipsec_limits limits;
 };
 
@@ -235,6 +242,7 @@ int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry);
 void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry);
 int mlx5e_accel_ipsec_fs_add_pol(struct mlx5e_ipsec_pol_entry *pol_entry);
 void mlx5e_accel_ipsec_fs_del_pol(struct mlx5e_ipsec_pol_entry *pol_entry);
+void mlx5e_accel_ipsec_fs_modify(struct mlx5e_ipsec_sa_entry *sa_entry);
 
 int mlx5_ipsec_create_sa_ctx(struct mlx5e_ipsec_sa_entry *sa_entry);
 void mlx5_ipsec_free_sa_ctx(struct mlx5e_ipsec_sa_entry *sa_entry);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 0539640a4d88..b47794d4146e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -926,9 +926,12 @@ static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC;
 	flow_act.crypto.obj_id = sa_entry->ipsec_obj_id;
 	flow_act.flags |= FLOW_ACT_NO_APPEND;
-	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-			   MLX5_FLOW_CONTEXT_ACTION_CRYPTO_DECRYPT |
+	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_CRYPTO_DECRYPT |
 			   MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	if (attrs->drop)
+		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_DROP;
+	else
+		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dest[0].ft = rx->ft.status;
 	dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
@@ -1018,9 +1021,13 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry)
 	flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC;
 	flow_act.crypto.obj_id = sa_entry->ipsec_obj_id;
 	flow_act.flags |= FLOW_ACT_NO_APPEND;
-	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
-			   MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT |
+	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT |
 			   MLX5_FLOW_CONTEXT_ACTION_COUNT;
+	if (attrs->drop)
+		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_DROP;
+	else
+		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+
 	dest[0].ft = tx->ft.status;
 	dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
@@ -1430,3 +1437,19 @@ err_rx_ipv4:
 	kfree(ipsec->tx);
 	return err;
 }
+
+void mlx5e_accel_ipsec_fs_modify(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct mlx5e_ipsec_sa_entry sa_entry_shadow = {};
+	int err;
+
+	memcpy(&sa_entry_shadow, sa_entry, sizeof(*sa_entry));
+	memset(&sa_entry_shadow.ipsec_rule, 0x00, sizeof(sa_entry->ipsec_rule));
+
+	err = mlx5e_accel_ipsec_fs_add_rule(&sa_entry_shadow);
+	if (err)
+		return;
+
+	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
+	memcpy(sa_entry, &sa_entry_shadow, sizeof(*sa_entry));
+}
-- 
cgit v1.2.3-70-g09d2


From f6636ff69ec4f2c94a5ee1d032b21cfe1e0a5678 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:36 +0200
Subject: net: dsa: microchip: ksz8: Separate static MAC table operations for
 code reuse

Move static MAC table operations to separate functions in order to reuse
the code for add/del_fdb. This is needed to address kernel warnings
caused by the lack of fdb add function support in the current driver.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8795.c | 34 +++++++++++++++++++++++-----------
 1 file changed, 23 insertions(+), 11 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index 3fffd5da8d3b..e47d08cf3f26 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -977,8 +977,8 @@ int ksz8_fdb_dump(struct ksz_device *dev, int port,
 	return ret;
 }
 
-int ksz8_mdb_add(struct ksz_device *dev, int port,
-		 const struct switchdev_obj_port_mdb *mdb, struct dsa_db db)
+static int ksz8_add_sta_mac(struct ksz_device *dev, int port,
+			    const unsigned char *addr, u16 vid)
 {
 	struct alu_struct alu;
 	int index;
@@ -988,8 +988,8 @@ int ksz8_mdb_add(struct ksz_device *dev, int port,
 	for (index = 0; index < dev->info->num_statics; index++) {
 		if (!ksz8_r_sta_mac_table(dev, index, &alu)) {
 			/* Found one already in static MAC table. */
-			if (!memcmp(alu.mac, mdb->addr, ETH_ALEN) &&
-			    alu.fid == mdb->vid)
+			if (!memcmp(alu.mac, addr, ETH_ALEN) &&
+			    alu.fid == vid)
 				break;
 		/* Remember the first empty entry. */
 		} else if (!empty) {
@@ -1005,23 +1005,23 @@ int ksz8_mdb_add(struct ksz_device *dev, int port,
 	if (index == dev->info->num_statics) {
 		index = empty - 1;
 		memset(&alu, 0, sizeof(alu));
-		memcpy(alu.mac, mdb->addr, ETH_ALEN);
+		memcpy(alu.mac, addr, ETH_ALEN);
 		alu.is_static = true;
 	}
 	alu.port_forward |= BIT(port);
-	if (mdb->vid) {
+	if (vid) {
 		alu.is_use_fid = true;
 
 		/* Need a way to map VID to FID. */
-		alu.fid = mdb->vid;
+		alu.fid = vid;
 	}
 	ksz8_w_sta_mac_table(dev, index, &alu);
 
 	return 0;
 }
 
-int ksz8_mdb_del(struct ksz_device *dev, int port,
-		 const struct switchdev_obj_port_mdb *mdb, struct dsa_db db)
+static int ksz8_del_sta_mac(struct ksz_device *dev, int port,
+			    const unsigned char *addr, u16 vid)
 {
 	struct alu_struct alu;
 	int index;
@@ -1029,8 +1029,8 @@ int ksz8_mdb_del(struct ksz_device *dev, int port,
 	for (index = 0; index < dev->info->num_statics; index++) {
 		if (!ksz8_r_sta_mac_table(dev, index, &alu)) {
 			/* Found one already in static MAC table. */
-			if (!memcmp(alu.mac, mdb->addr, ETH_ALEN) &&
-			    alu.fid == mdb->vid)
+			if (!memcmp(alu.mac, addr, ETH_ALEN) &&
+			    alu.fid == vid)
 				break;
 		}
 	}
@@ -1049,6 +1049,18 @@ exit:
 	return 0;
 }
 
+int ksz8_mdb_add(struct ksz_device *dev, int port,
+		 const struct switchdev_obj_port_mdb *mdb, struct dsa_db db)
+{
+	return ksz8_add_sta_mac(dev, port, mdb->addr, mdb->vid);
+}
+
+int ksz8_mdb_del(struct ksz_device *dev, int port,
+		 const struct switchdev_obj_port_mdb *mdb, struct dsa_db db)
+{
+	return ksz8_del_sta_mac(dev, port, mdb->addr, mdb->vid);
+}
+
 int ksz8_port_vlan_filtering(struct ksz_device *dev, int port, bool flag,
 			     struct netlink_ext_ack *extack)
 {
-- 
cgit v1.2.3-70-g09d2


From 57795412a447812553a8b61bbd968d06a54a41b4 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:37 +0200
Subject: net: dsa: microchip: ksz8: Implement add/del_fdb and use static MAC
 table operations

Add support for add/del_fdb operations and utilize the refactored static
MAC table code. This resolves kernel warnings caused by the lack of fdb
add function support in the current driver.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8.h       |  4 ++++
 drivers/net/dsa/microchip/ksz8795.c    | 12 ++++++++++++
 drivers/net/dsa/microchip/ksz_common.c |  2 ++
 3 files changed, 18 insertions(+)

diff --git a/drivers/net/dsa/microchip/ksz8.h b/drivers/net/dsa/microchip/ksz8.h
index ea05abfbd51d..4d4552ac2792 100644
--- a/drivers/net/dsa/microchip/ksz8.h
+++ b/drivers/net/dsa/microchip/ksz8.h
@@ -32,6 +32,10 @@ void ksz8_freeze_mib(struct ksz_device *dev, int port, bool freeze);
 void ksz8_port_init_cnt(struct ksz_device *dev, int port);
 int ksz8_fdb_dump(struct ksz_device *dev, int port,
 		  dsa_fdb_dump_cb_t *cb, void *data);
+int ksz8_fdb_add(struct ksz_device *dev, int port, const unsigned char *addr,
+		 u16 vid, struct dsa_db db);
+int ksz8_fdb_del(struct ksz_device *dev, int port, const unsigned char *addr,
+		 u16 vid, struct dsa_db db);
 int ksz8_mdb_add(struct ksz_device *dev, int port,
 		 const struct switchdev_obj_port_mdb *mdb, struct dsa_db db);
 int ksz8_mdb_del(struct ksz_device *dev, int port,
diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index e47d08cf3f26..7143e90bf35f 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -1061,6 +1061,18 @@ int ksz8_mdb_del(struct ksz_device *dev, int port,
 	return ksz8_del_sta_mac(dev, port, mdb->addr, mdb->vid);
 }
 
+int ksz8_fdb_add(struct ksz_device *dev, int port, const unsigned char *addr,
+		 u16 vid, struct dsa_db db)
+{
+	return ksz8_add_sta_mac(dev, port, addr, vid);
+}
+
+int ksz8_fdb_del(struct ksz_device *dev, int port, const unsigned char *addr,
+		 u16 vid, struct dsa_db db)
+{
+	return ksz8_del_sta_mac(dev, port, addr, vid);
+}
+
 int ksz8_port_vlan_filtering(struct ksz_device *dev, int port, bool flag,
 			     struct netlink_ext_ack *extack)
 {
diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index 5568d7b22135..a4428be5f483 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -200,6 +200,8 @@ static const struct ksz_dev_ops ksz8_dev_ops = {
 	.freeze_mib = ksz8_freeze_mib,
 	.port_init_cnt = ksz8_port_init_cnt,
 	.fdb_dump = ksz8_fdb_dump,
+	.fdb_add = ksz8_fdb_add,
+	.fdb_del = ksz8_fdb_del,
 	.mdb_add = ksz8_mdb_add,
 	.mdb_del = ksz8_mdb_del,
 	.vlan_filtering = ksz8_port_vlan_filtering,
-- 
cgit v1.2.3-70-g09d2


From b5751cdd7dbe618a03951bdd4c982a71ba448b1b Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:38 +0200
Subject: net: dsa: microchip: ksz8: Make ksz8_r_sta_mac_table() static

As ksz8_r_sta_mac_table() is only used within ksz8795.c, there is no need
to export it. Make the function static for better encapsulation.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8.h    | 2 --
 drivers/net/dsa/microchip/ksz8795.c | 4 ++--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz8.h b/drivers/net/dsa/microchip/ksz8.h
index 4d4552ac2792..ce0f7b71268d 100644
--- a/drivers/net/dsa/microchip/ksz8.h
+++ b/drivers/net/dsa/microchip/ksz8.h
@@ -21,8 +21,6 @@ int ksz8_r_phy(struct ksz_device *dev, u16 phy, u16 reg, u16 *val);
 int ksz8_w_phy(struct ksz_device *dev, u16 phy, u16 reg, u16 val);
 int ksz8_r_dyn_mac_table(struct ksz_device *dev, u16 addr, u8 *mac_addr,
 			 u8 *fid, u8 *src_port, u8 *timestamp, u16 *entries);
-int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
-			 struct alu_struct *alu);
 void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
 			  struct alu_struct *alu);
 void ksz8_r_mib_cnt(struct ksz_device *dev, int port, u16 addr, u64 *cnt);
diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index 7143e90bf35f..a66372614226 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -457,8 +457,8 @@ int ksz8_r_dyn_mac_table(struct ksz_device *dev, u16 addr, u8 *mac_addr,
 	return rc;
 }
 
-int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
-			 struct alu_struct *alu)
+static int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
+				struct alu_struct *alu)
 {
 	u32 data_hi, data_lo;
 	const u8 *shifts;
-- 
cgit v1.2.3-70-g09d2


From 559901b46810e82ba5321a5e789f994b65d3bc3d Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:39 +0200
Subject: net: dsa: microchip: ksz8_r_sta_mac_table(): Avoid using error code
 for empty entries

Prepare for the next patch by ensuring that ksz8_r_sta_mac_table() does
not use error codes for empty entries. This change will enable better
handling of read/write errors in the upcoming patch.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8795.c | 87 +++++++++++++++++++++----------------
 1 file changed, 50 insertions(+), 37 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index a66372614226..b2fd31be4455 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -458,7 +458,7 @@ int ksz8_r_dyn_mac_table(struct ksz_device *dev, u16 addr, u8 *mac_addr,
 }
 
 static int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
-				struct alu_struct *alu)
+				struct alu_struct *alu, bool *valid)
 {
 	u32 data_hi, data_lo;
 	const u8 *shifts;
@@ -471,28 +471,32 @@ static int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
 	ksz8_r_table(dev, TABLE_STATIC_MAC, addr, &data);
 	data_hi = data >> 32;
 	data_lo = (u32)data;
-	if (data_hi & (masks[STATIC_MAC_TABLE_VALID] |
-			masks[STATIC_MAC_TABLE_OVERRIDE])) {
-		alu->mac[5] = (u8)data_lo;
-		alu->mac[4] = (u8)(data_lo >> 8);
-		alu->mac[3] = (u8)(data_lo >> 16);
-		alu->mac[2] = (u8)(data_lo >> 24);
-		alu->mac[1] = (u8)data_hi;
-		alu->mac[0] = (u8)(data_hi >> 8);
-		alu->port_forward =
-			(data_hi & masks[STATIC_MAC_TABLE_FWD_PORTS]) >>
-				shifts[STATIC_MAC_FWD_PORTS];
-		alu->is_override =
-			(data_hi & masks[STATIC_MAC_TABLE_OVERRIDE]) ? 1 : 0;
-		data_hi >>= 1;
-		alu->is_static = true;
-		alu->is_use_fid =
-			(data_hi & masks[STATIC_MAC_TABLE_USE_FID]) ? 1 : 0;
-		alu->fid = (data_hi & masks[STATIC_MAC_TABLE_FID]) >>
-				shifts[STATIC_MAC_FID];
+
+	if (!(data_hi & (masks[STATIC_MAC_TABLE_VALID] |
+			 masks[STATIC_MAC_TABLE_OVERRIDE]))) {
+		*valid = false;
 		return 0;
 	}
-	return -ENXIO;
+
+	alu->mac[5] = (u8)data_lo;
+	alu->mac[4] = (u8)(data_lo >> 8);
+	alu->mac[3] = (u8)(data_lo >> 16);
+	alu->mac[2] = (u8)(data_lo >> 24);
+	alu->mac[1] = (u8)data_hi;
+	alu->mac[0] = (u8)(data_hi >> 8);
+	alu->port_forward =
+		(data_hi & masks[STATIC_MAC_TABLE_FWD_PORTS]) >>
+			shifts[STATIC_MAC_FWD_PORTS];
+	alu->is_override = (data_hi & masks[STATIC_MAC_TABLE_OVERRIDE]) ? 1 : 0;
+	data_hi >>= 1;
+	alu->is_static = true;
+	alu->is_use_fid = (data_hi & masks[STATIC_MAC_TABLE_USE_FID]) ? 1 : 0;
+	alu->fid = (data_hi & masks[STATIC_MAC_TABLE_FID]) >>
+		shifts[STATIC_MAC_FID];
+
+	*valid = true;
+
+	return 0;
 }
 
 void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
@@ -981,20 +985,25 @@ static int ksz8_add_sta_mac(struct ksz_device *dev, int port,
 			    const unsigned char *addr, u16 vid)
 {
 	struct alu_struct alu;
-	int index;
+	int index, ret;
 	int empty = 0;
 
 	alu.port_forward = 0;
 	for (index = 0; index < dev->info->num_statics; index++) {
-		if (!ksz8_r_sta_mac_table(dev, index, &alu)) {
-			/* Found one already in static MAC table. */
-			if (!memcmp(alu.mac, addr, ETH_ALEN) &&
-			    alu.fid == vid)
-				break;
-		/* Remember the first empty entry. */
-		} else if (!empty) {
-			empty = index + 1;
+		bool valid;
+
+		ret = ksz8_r_sta_mac_table(dev, index, &alu, &valid);
+		if (ret)
+			return ret;
+		if (!valid) {
+			/* Remember the first empty entry. */
+			if (!empty)
+				empty = index + 1;
+			continue;
 		}
+
+		if (!memcmp(alu.mac, addr, ETH_ALEN) && alu.fid == vid)
+			break;
 	}
 
 	/* no available entry */
@@ -1024,15 +1033,19 @@ static int ksz8_del_sta_mac(struct ksz_device *dev, int port,
 			    const unsigned char *addr, u16 vid)
 {
 	struct alu_struct alu;
-	int index;
+	int index, ret;
 
 	for (index = 0; index < dev->info->num_statics; index++) {
-		if (!ksz8_r_sta_mac_table(dev, index, &alu)) {
-			/* Found one already in static MAC table. */
-			if (!memcmp(alu.mac, addr, ETH_ALEN) &&
-			    alu.fid == vid)
-				break;
-		}
+		bool valid;
+
+		ret = ksz8_r_sta_mac_table(dev, index, &alu, &valid);
+		if (ret)
+			return ret;
+		if (!valid)
+			continue;
+
+		if (!memcmp(alu.mac, addr, ETH_ALEN) && alu.fid == vid)
+			break;
 	}
 
 	/* no available entry */
-- 
cgit v1.2.3-70-g09d2


From ec2312f337356cd48f538c49f7cf1c2047ec1866 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:40 +0200
Subject: net: dsa: microchip: ksz8_r_sta_mac_table(): Utilize error values
 from read/write functions

Take advantage of the return values provided by read/write functions in
ksz8_r_sta_mac_table() to handle cases where read/write operations may
fail.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8795.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index b2fd31be4455..35a326e6fa0d 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -336,19 +336,26 @@ void ksz8_port_init_cnt(struct ksz_device *dev, int port)
 	}
 }
 
-static void ksz8_r_table(struct ksz_device *dev, int table, u16 addr, u64 *data)
+static int ksz8_r_table(struct ksz_device *dev, int table, u16 addr, u64 *data)
 {
 	const u16 *regs;
 	u16 ctrl_addr;
+	int ret;
 
 	regs = dev->info->regs;
 
 	ctrl_addr = IND_ACC_TABLE(table | TABLE_READ) | addr;
 
 	mutex_lock(&dev->alu_mutex);
-	ksz_write16(dev, regs[REG_IND_CTRL_0], ctrl_addr);
-	ksz_read64(dev, regs[REG_IND_DATA_HI], data);
+	ret = ksz_write16(dev, regs[REG_IND_CTRL_0], ctrl_addr);
+	if (ret)
+		goto unlock_alu;
+
+	ret = ksz_read64(dev, regs[REG_IND_DATA_HI], data);
+unlock_alu:
 	mutex_unlock(&dev->alu_mutex);
+
+	return ret;
 }
 
 static void ksz8_w_table(struct ksz_device *dev, int table, u16 addr, u64 data)
@@ -464,11 +471,15 @@ static int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
 	const u8 *shifts;
 	const u32 *masks;
 	u64 data;
+	int ret;
 
 	shifts = dev->info->shifts;
 	masks = dev->info->masks;
 
-	ksz8_r_table(dev, TABLE_STATIC_MAC, addr, &data);
+	ret = ksz8_r_table(dev, TABLE_STATIC_MAC, addr, &data);
+	if (ret)
+		return ret;
+
 	data_hi = data >> 32;
 	data_lo = (u32)data;
 
-- 
cgit v1.2.3-70-g09d2


From c8e04374f9e1194a99c1c67980a21aa3f56a79d0 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:41 +0200
Subject: net: dsa: microchip: Make ksz8_w_sta_mac_table() static

Since ksz8_w_sta_mac_table() is only used within ksz8795.c, make it static
to limit its scope.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8.h    | 2 --
 drivers/net/dsa/microchip/ksz8795.c | 4 ++--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz8.h b/drivers/net/dsa/microchip/ksz8.h
index ce0f7b71268d..e68465fdf6b9 100644
--- a/drivers/net/dsa/microchip/ksz8.h
+++ b/drivers/net/dsa/microchip/ksz8.h
@@ -21,8 +21,6 @@ int ksz8_r_phy(struct ksz_device *dev, u16 phy, u16 reg, u16 *val);
 int ksz8_w_phy(struct ksz_device *dev, u16 phy, u16 reg, u16 val);
 int ksz8_r_dyn_mac_table(struct ksz_device *dev, u16 addr, u8 *mac_addr,
 			 u8 *fid, u8 *src_port, u8 *timestamp, u16 *entries);
-void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
-			  struct alu_struct *alu);
 void ksz8_r_mib_cnt(struct ksz_device *dev, int port, u16 addr, u64 *cnt);
 void ksz8_r_mib_pkt(struct ksz_device *dev, int port, u16 addr,
 		    u64 *dropped, u64 *cnt);
diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index 35a326e6fa0d..f394fb7d9ea1 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -510,8 +510,8 @@ static int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
 	return 0;
 }
 
-void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
-			  struct alu_struct *alu)
+static void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
+				 struct alu_struct *alu)
 {
 	u32 data_hi, data_lo;
 	const u8 *shifts;
-- 
cgit v1.2.3-70-g09d2


From 3c2e6b54e4e98f85bc9e17e20d6c702dd907afd3 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Tue, 4 Apr 2023 12:18:42 +0200
Subject: net: dsa: microchip: Utilize error values in ksz8_w_sta_mac_table()

To handle potential read/write operation failures, update
ksz8_w_sta_mac_table() to make use of the return values provided by
read/write functions.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/ksz8795.c | 33 ++++++++++++++++++---------------
 1 file changed, 18 insertions(+), 15 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c
index f394fb7d9ea1..23614a937cc3 100644
--- a/drivers/net/dsa/microchip/ksz8795.c
+++ b/drivers/net/dsa/microchip/ksz8795.c
@@ -358,19 +358,26 @@ unlock_alu:
 	return ret;
 }
 
-static void ksz8_w_table(struct ksz_device *dev, int table, u16 addr, u64 data)
+static int ksz8_w_table(struct ksz_device *dev, int table, u16 addr, u64 data)
 {
 	const u16 *regs;
 	u16 ctrl_addr;
+	int ret;
 
 	regs = dev->info->regs;
 
 	ctrl_addr = IND_ACC_TABLE(table) | addr;
 
 	mutex_lock(&dev->alu_mutex);
-	ksz_write64(dev, regs[REG_IND_DATA_HI], data);
-	ksz_write16(dev, regs[REG_IND_CTRL_0], ctrl_addr);
+	ret = ksz_write64(dev, regs[REG_IND_DATA_HI], data);
+	if (ret)
+		goto unlock_alu;
+
+	ret = ksz_write16(dev, regs[REG_IND_CTRL_0], ctrl_addr);
+unlock_alu:
 	mutex_unlock(&dev->alu_mutex);
+
+	return ret;
 }
 
 static int ksz8_valid_dyn_entry(struct ksz_device *dev, u8 *data)
@@ -510,8 +517,8 @@ static int ksz8_r_sta_mac_table(struct ksz_device *dev, u16 addr,
 	return 0;
 }
 
-static void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
-				 struct alu_struct *alu)
+static int ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
+				struct alu_struct *alu)
 {
 	u32 data_hi, data_lo;
 	const u8 *shifts;
@@ -539,7 +546,8 @@ static void ksz8_w_sta_mac_table(struct ksz_device *dev, u16 addr,
 		data_hi &= ~masks[STATIC_MAC_TABLE_OVERRIDE];
 
 	data = (u64)data_hi << 32 | data_lo;
-	ksz8_w_table(dev, TABLE_STATIC_MAC, addr, data);
+
+	return ksz8_w_table(dev, TABLE_STATIC_MAC, addr, data);
 }
 
 static void ksz8_from_vlan(struct ksz_device *dev, u32 vlan, u8 *fid,
@@ -1035,9 +1043,8 @@ static int ksz8_add_sta_mac(struct ksz_device *dev, int port,
 		/* Need a way to map VID to FID. */
 		alu.fid = vid;
 	}
-	ksz8_w_sta_mac_table(dev, index, &alu);
 
-	return 0;
+	return ksz8_w_sta_mac_table(dev, index, &alu);
 }
 
 static int ksz8_del_sta_mac(struct ksz_device *dev, int port,
@@ -1061,16 +1068,14 @@ static int ksz8_del_sta_mac(struct ksz_device *dev, int port,
 
 	/* no available entry */
 	if (index == dev->info->num_statics)
-		goto exit;
+		return 0;
 
 	/* clear port */
 	alu.port_forward &= ~BIT(port);
 	if (!alu.port_forward)
 		alu.is_static = false;
-	ksz8_w_sta_mac_table(dev, index, &alu);
 
-exit:
-	return 0;
+	return ksz8_w_sta_mac_table(dev, index, &alu);
 }
 
 int ksz8_mdb_add(struct ksz_device *dev, int port,
@@ -1394,9 +1399,7 @@ int ksz8_enable_stp_addr(struct ksz_device *dev)
 	alu.is_override = true;
 	alu.port_forward = dev->info->cpu_ports;
 
-	ksz8_w_sta_mac_table(dev, 0, &alu);
-
-	return 0;
+	return ksz8_w_sta_mac_table(dev, 0, &alu);
 }
 
 int ksz8_setup(struct dsa_switch *ds)
-- 
cgit v1.2.3-70-g09d2


From 905a9eb5f636f3312964b162362e4d6ca4e37378 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 5 Apr 2023 07:15:56 +0000
Subject: selftests/net: fix typo in tcp_mmap

kernel test robot reported the following warning:

All warnings (new ones prefixed by >>):

   tcp_mmap.c: In function 'child_thread':
>> tcp_mmap.c:211:61: warning: 'lu' may be used uninitialized in this function [-Wmaybe-uninitialized]
     211 |                         zc.length = min(chunk_size, FILE_SZ - lu);

We want to read FILE_SZ bytes, so the correct expression
should be (FILE_SZ - total)

Fixes: 5c5945dc695c ("selftests/net: Add SHA256 computation over data sent in tcp_mmap")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304042104.UFIuevBp-lkp@intel.com/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Xiaoyan Li <lixiaoyan@google.com>
Cc: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230405071556.1019623-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 tools/testing/selftests/net/tcp_mmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/tcp_mmap.c b/tools/testing/selftests/net/tcp_mmap.c
index 607cc9ad8d1b..6e59b1461dcc 100644
--- a/tools/testing/selftests/net/tcp_mmap.c
+++ b/tools/testing/selftests/net/tcp_mmap.c
@@ -208,7 +208,7 @@ void *child_thread(void *arg)
 
 			memset(&zc, 0, sizeof(zc));
 			zc.address = (__u64)((unsigned long)addr);
-			zc.length = min(chunk_size, FILE_SZ - lu);
+			zc.length = min(chunk_size, FILE_SZ - total);
 
 			res = getsockopt(fd, IPPROTO_TCP, TCP_ZEROCOPY_RECEIVE,
 					 &zc, &zc_len);
-- 
cgit v1.2.3-70-g09d2


From 0ebd4fd6b9064764a3af3d671463b1350abffb6c Mon Sep 17 00:00:00 2001
From: Yinjun Zhang <yinjun.zhang@corigine.com>
Date: Wed, 5 Apr 2023 14:08:29 +0200
Subject: nfp: initialize netdev's dev_port with correct id

`dev_port` is used to differentiate devices that instantiate from
the same function, which is the case in most of NFP NICs.

In some customized scenario, `dev_port` is used to rename netdev
instead of `phys_port_name`. Example rules using `dev_port`:

  SUBSYSTEM=="net", ACTION=="add", KERNELS=="0000:e1:00.0", ATTR{dev_port}=="0", NAME:="ens8np0"
  SUBSYSTEM=="net", ACTION=="add", KERNELS=="0000:e1:00.0", ATTR{dev_port}=="1", NAME:="ens8np1"

To take port split case into account, here we initialize `dev_port`
according to the port sequence in eth_table from management firmware
instead of using port label id directly. And management firmware
makes sure that port sequence matches its label id.

Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Link: https://lore.kernel.org/r/20230405120829.28817-1-louis.peens@corigine.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/netronome/nfp/nfp_port.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/netronome/nfp/nfp_port.c b/drivers/net/ethernet/netronome/nfp/nfp_port.c
index 4f2308570dcf..54640bcb70fb 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_port.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_port.c
@@ -189,6 +189,7 @@ int nfp_port_init_phy_port(struct nfp_pf *pf, struct nfp_app *app,
 
 	port->eth_port = &pf->eth_tbl->ports[id];
 	port->eth_id = pf->eth_tbl->ports[id].index;
+	port->netdev->dev_port = id;
 	if (pf->mac_stats_mem)
 		port->eth_stats =
 			pf->mac_stats_mem + port->eth_id * NFP_MAC_STATS_SIZE;
-- 
cgit v1.2.3-70-g09d2


From d769ccaf957fe7391f357c0a923de71f594b8a2b Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Thu, 6 Apr 2023 01:59:18 +0200
Subject: xsk: Fix unaligned descriptor validation

Make sure unaligned descriptors that straddle the end of the UMEM are
considered invalid. Currently, descriptor validation is broken for
zero-copy mode which only checks descriptors at page granularity.
For example, descriptors in zero-copy mode that overrun the end of the
UMEM but not a page boundary are (incorrectly) considered valid. The
UMEM boundary check needs to happen before the page boundary and
contiguity checks in xp_desc_crosses_non_contig_pg(). Do this check in
xp_unaligned_validate_desc() instead like xp_check_unaligned() already
does.

Fixes: 2b43470add8c ("xsk: Introduce AF_XDP buffer allocation API")
Signed-off-by: Kal Conley <kal.conley@dectris.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/r/20230405235920.7305-2-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/net/xsk_buff_pool.h | 9 ++-------
 net/xdp/xsk_queue.h         | 1 +
 2 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h
index 3e952e569418..d318c769b445 100644
--- a/include/net/xsk_buff_pool.h
+++ b/include/net/xsk_buff_pool.h
@@ -180,13 +180,8 @@ static inline bool xp_desc_crosses_non_contig_pg(struct xsk_buff_pool *pool,
 	if (likely(!cross_pg))
 		return false;
 
-	if (pool->dma_pages_cnt) {
-		return !(pool->dma_pages[addr >> PAGE_SHIFT] &
-			 XSK_NEXT_PG_CONTIG_MASK);
-	}
-
-	/* skb path */
-	return addr + len > pool->addrs_cnt;
+	return pool->dma_pages_cnt &&
+	       !(pool->dma_pages[addr >> PAGE_SHIFT] & XSK_NEXT_PG_CONTIG_MASK);
 }
 
 static inline u64 xp_aligned_extract_addr(struct xsk_buff_pool *pool, u64 addr)
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index bfb2a7e50c26..66c6f57c9c44 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -162,6 +162,7 @@ static inline bool xp_unaligned_validate_desc(struct xsk_buff_pool *pool,
 		return false;
 
 	if (base_addr >= pool->addrs_cnt || addr >= pool->addrs_cnt ||
+	    addr + desc->len > pool->addrs_cnt ||
 	    xp_desc_crosses_non_contig_pg(pool, addr, desc->len))
 		return false;
 
-- 
cgit v1.2.3-70-g09d2


From c0801598e5430d9da2d406ed32fcedbef23977fc Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Thu, 6 Apr 2023 01:59:19 +0200
Subject: selftests: xsk: Add test UNALIGNED_INV_DESC_4K1_FRAME_SIZE

Add unaligned descriptor test for frame size of 4001. Using an odd frame
size ensures that the end of the UMEM is not near a page boundary. This
allows testing descriptors that staddle the end of the UMEM but not a
page.

This test used to fail without the previous commit ("xsk: Fix unaligned
descriptor validation").

Signed-off-by: Kal Conley <kal.conley@dectris.com>
Link: https://lore.kernel.org/r/20230405235920.7305-3-kal.conley@dectris.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/xskxceiver.c | 24 ++++++++++++++++++++++++
 tools/testing/selftests/bpf/xskxceiver.h |  1 +
 2 files changed, 25 insertions(+)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index 1a4bdd5aa78c..5a9691e942de 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -69,6 +69,7 @@
  */
 
 #define _GNU_SOURCE
+#include <assert.h>
 #include <fcntl.h>
 #include <errno.h>
 #include <getopt.h>
@@ -1876,6 +1877,29 @@ static void run_pkt_test(struct test_spec *test, enum test_mode mode, enum test_
 		test->ifobj_rx->umem->unaligned_mode = true;
 		testapp_invalid_desc(test);
 		break;
+	case TEST_TYPE_UNALIGNED_INV_DESC_4K1_FRAME: {
+		u64 page_size, umem_size;
+
+		if (!hugepages_present(test->ifobj_tx)) {
+			ksft_test_result_skip("No 2M huge pages present.\n");
+			return;
+		}
+		test_spec_set_name(test, "UNALIGNED_INV_DESC_4K1_FRAME_SIZE");
+		/* Odd frame size so the UMEM doesn't end near a page boundary. */
+		test->ifobj_tx->umem->frame_size = 4001;
+		test->ifobj_rx->umem->frame_size = 4001;
+		test->ifobj_tx->umem->unaligned_mode = true;
+		test->ifobj_rx->umem->unaligned_mode = true;
+		/* This test exists to test descriptors that staddle the end of
+		 * the UMEM but not a page.
+		 */
+		page_size = sysconf(_SC_PAGESIZE);
+		umem_size = test->ifobj_tx->umem->num_frames * test->ifobj_tx->umem->frame_size;
+		assert(umem_size % page_size > PKT_SIZE);
+		assert(umem_size % page_size < page_size - PKT_SIZE);
+		testapp_invalid_desc(test);
+		break;
+	}
 	case TEST_TYPE_UNALIGNED:
 		if (!testapp_unaligned(test))
 			return;
diff --git a/tools/testing/selftests/bpf/xskxceiver.h b/tools/testing/selftests/bpf/xskxceiver.h
index cc24ab72f3ff..919327807a4e 100644
--- a/tools/testing/selftests/bpf/xskxceiver.h
+++ b/tools/testing/selftests/bpf/xskxceiver.h
@@ -78,6 +78,7 @@ enum test_type {
 	TEST_TYPE_ALIGNED_INV_DESC,
 	TEST_TYPE_ALIGNED_INV_DESC_2K_FRAME,
 	TEST_TYPE_UNALIGNED_INV_DESC,
+	TEST_TYPE_UNALIGNED_INV_DESC_4K1_FRAME,
 	TEST_TYPE_HEADROOM,
 	TEST_TYPE_TEARDOWN,
 	TEST_TYPE_BIDI,
-- 
cgit v1.2.3-70-g09d2


From 13fbcee55706db45ce047a7cea14811d68f94ee3 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Thu, 6 Apr 2023 09:44:55 -0700
Subject: bpf: Improve verifier JEQ/JNE insn branch taken checking

Currently, for BPF_JEQ/BPF_JNE insn, verifier determines
whether the branch is taken or not only if both operands
are constants. Therefore, for the following code snippet,
  0: (85) call bpf_ktime_get_ns#5       ; R0_w=scalar()
  1: (a5) if r0 < 0x3 goto pc+2         ; R0_w=scalar(umin=3)
  2: (b7) r2 = 2                        ; R2_w=2
  3: (1d) if r0 == r2 goto pc+2 6

At insn 3, since r0 is not a constant, verifier assumes both branch
can be taken which may lead inproper verification failure.

Add comparing umin/umax value and the constant. If the umin value
is greater than the constant, or umax value is smaller than the constant,
for JEQ the branch must be not-taken, and for JNE the branch must be taken.
The jmp32 mode JEQ/JNE branch taken checking is also handled similarly.

The following lists the veristat result w.r.t. changed number
of processes insns during verification:

File                                                   Program                                               Insns (A)  Insns (B)  Insns    (DIFF)
-----------------------------------------------------  ----------------------------------------------------  ---------  ---------  ---------------
test_cls_redirect.bpf.linked3.o                        cls_redirect                                              64980      73472  +8492 (+13.07%)
test_seg6_loop.bpf.linked3.o                           __add_egr_x                                               12425      12423      -2 (-0.02%)
test_tcp_hdr_options.bpf.linked3.o                     estab                                                      2634       2558     -76 (-2.89%)
test_parse_tcp_hdr_opt.bpf.linked3.o                   xdp_ingress_v6                                             1421       1420      -1 (-0.07%)
test_parse_tcp_hdr_opt_dynptr.bpf.linked3.o            xdp_ingress_v6                                             1238       1237      -1 (-0.08%)
test_tc_dtime.bpf.linked3.o                            egress_fwdns_prio100                                        414        411      -3 (-0.72%)

Mostly a small improvement but test_cls_redirect.bpf.linked3.o has a 13% regression.
I checked with verifier log and found it this is due to pruning.
For some JEQ/JNE branches impacted by this patch,
one branch is explored and the other has state equivalence and
pruned.

Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Dave Marchevsky <davemarchevsky@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230406164455.1045294-1-yhs@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 56f569811f70..5c6b90e384a5 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -12651,10 +12651,14 @@ static int is_branch32_taken(struct bpf_reg_state *reg, u32 val, u8 opcode)
 	case BPF_JEQ:
 		if (tnum_is_const(subreg))
 			return !!tnum_equals_const(subreg, val);
+		else if (val < reg->u32_min_value || val > reg->u32_max_value)
+			return 0;
 		break;
 	case BPF_JNE:
 		if (tnum_is_const(subreg))
 			return !tnum_equals_const(subreg, val);
+		else if (val < reg->u32_min_value || val > reg->u32_max_value)
+			return 1;
 		break;
 	case BPF_JSET:
 		if ((~subreg.mask & subreg.value) & val)
@@ -12724,10 +12728,14 @@ static int is_branch64_taken(struct bpf_reg_state *reg, u64 val, u8 opcode)
 	case BPF_JEQ:
 		if (tnum_is_const(reg->var_off))
 			return !!tnum_equals_const(reg->var_off, val);
+		else if (val < reg->umin_value || val > reg->umax_value)
+			return 0;
 		break;
 	case BPF_JNE:
 		if (tnum_is_const(reg->var_off))
 			return !tnum_equals_const(reg->var_off, val);
+		else if (val < reg->umin_value || val > reg->umax_value)
+			return 1;
 		break;
 	case BPF_JSET:
 		if ((~reg->var_off.mask & reg->var_off.value) & val)
-- 
cgit v1.2.3-70-g09d2


From aec08d677b4d0adeb7412fa98547cf07bfce6fea Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Thu, 6 Apr 2023 09:45:00 -0700
Subject: selftests/bpf: Add tests for non-constant cond_op NE/EQ bound
 deduction

Add various tests for code pattern '<non-const> NE/EQ <const>' implemented
in the previous verifier patch. Without the verifier patch, these new
tests will fail.

Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230406164500.1045715-1-yhs@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../progs/verifier_bounds_deduction_non_const.c    | 179 +++++++++++++++++++++
 2 files changed, 181 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index efc8cf2e18d0..73dff693d411 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -7,6 +7,7 @@
 #include "verifier_array_access.skel.h"
 #include "verifier_basic_stack.skel.h"
 #include "verifier_bounds_deduction.skel.h"
+#include "verifier_bounds_deduction_non_const.skel.h"
 #include "verifier_bounds_mix_sign_unsign.skel.h"
 #include "verifier_cfg.skel.h"
 #include "verifier_cgroup_inv_retcode.skel.h"
@@ -70,6 +71,7 @@ void test_verifier_and(void)                  { RUN(verifier_and); }
 void test_verifier_array_access(void)         { RUN(verifier_array_access); }
 void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
+void test_verifier_bounds_deduction_non_const(void)     { RUN(verifier_bounds_deduction_non_const); }
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
 void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
 void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c b/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c
new file mode 100644
index 000000000000..fe570d866139
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c
@@ -0,0 +1,179 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <non_const> == <const>, 1")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 < 3 goto l0_%=;				\
+	r2 = 2;						\
+	if r0 == r2 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <non_const> == <const>, 2")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_2(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 > 3 goto l0_%=;				\
+	r2 = 4;						\
+	if r0 == r2 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <non_const> != <const>, 1")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_3(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 < 3 goto l0_%=;				\
+	r2 = 2;						\
+	if r0 != r2 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <non_const> != <const>, 2")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_4(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 > 3 goto l0_%=;				\
+	r2 = 4;						\
+	if r0 != r2 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <non_const> == <const>, 1")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_5(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 < 4 goto l0_%=;				\
+	w2 = 3;						\
+	if w0 == w2 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <non_const> == <const>, 2")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_6(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 > 4 goto l0_%=;				\
+	w2 = 5;						\
+	if w0 == w2 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <non_const> != <const>, 1")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_7(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 < 3 goto l0_%=;				\
+	w2 = 2;						\
+	if w0 != w2 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <non_const> != <const>, 2")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_8(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 > 3 goto l0_%=;				\
+	w2 = 4;						\
+	if w0 != w2 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 953d9f5beaf75e88c69a13d70ce424cd606a29f5 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Thu, 6 Apr 2023 09:45:05 -0700
Subject: bpf: Improve handling of pattern '<const> <cond_op> <non_const>' in
 verifier

Currently, the verifier does not handle '<const> <cond_op> <non_const>' well.
For example,
  ...
  10: (79) r1 = *(u64 *)(r10 -16)       ; R1_w=scalar() R10=fp0
  11: (b7) r2 = 0                       ; R2_w=0
  12: (2d) if r2 > r1 goto pc+2
  13: (b7) r0 = 0
  14: (95) exit
  15: (65) if r1 s> 0x1 goto pc+3
  16: (0f) r0 += r1
  ...
At insn 12, verifier decides both true and false branch are possible, but
actually only false branch is possible.

Currently, the verifier already supports patterns '<non_const> <cond_op> <const>.
Add support for patterns '<const> <cond_op> <non_const>' in a similar way.

Also fix selftest 'verifier_bounds_mix_sign_unsign/bounds checks mixing signed and unsigned, variant 10'
due to this change.

Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Dave Marchevsky <davemarchevsky@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230406164505.1046801-1-yhs@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c                                        | 12 ++++++++++++
 .../selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c    |  2 +-
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5c6b90e384a5..3660b573048a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -13356,6 +13356,18 @@ static int check_cond_jmp_op(struct bpf_verifier_env *env,
 				       src_reg->var_off.value,
 				       opcode,
 				       is_jmp32);
+	} else if (dst_reg->type == SCALAR_VALUE &&
+		   is_jmp32 && tnum_is_const(tnum_subreg(dst_reg->var_off))) {
+		pred = is_branch_taken(src_reg,
+				       tnum_subreg(dst_reg->var_off).value,
+				       flip_opcode(opcode),
+				       is_jmp32);
+	} else if (dst_reg->type == SCALAR_VALUE &&
+		   !is_jmp32 && tnum_is_const(dst_reg->var_off)) {
+		pred = is_branch_taken(src_reg,
+				       dst_reg->var_off.value,
+				       flip_opcode(opcode),
+				       is_jmp32);
 	} else if (reg_is_pkt_pointer_any(dst_reg) &&
 		   reg_is_pkt_pointer_any(src_reg) &&
 		   !is_jmp32) {
diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c b/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c
index 91a66357896a..4f40144748a5 100644
--- a/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c
+++ b/tools/testing/selftests/bpf/progs/verifier_bounds_mix_sign_unsign.c
@@ -354,7 +354,7 @@ __naked void signed_and_unsigned_variant_10(void)
 	call %[bpf_map_lookup_elem];			\
 	if r0 == 0 goto l0_%=;				\
 	r1 = *(u64*)(r10 - 16);				\
-	r2 = 0;						\
+	r2 = -1;						\
 	if r2 > r1 goto l1_%=;				\
 	r0 = 0;						\
 	exit;						\
-- 
cgit v1.2.3-70-g09d2


From 23a88fae9f20d47bb3aed99b1e08d0d6cf65cf0c Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Thu, 6 Apr 2023 09:45:10 -0700
Subject: selftests/bpf: Add verifier tests for code pattern '<const> <cond_op>
 <non_const>'

Add various tests for code pattern '<const> <cond_op> <non_const>' to
exercise the previous verifier patch.

The following are veristat changed number of processed insns stat
comparing the previous patch vs. this patch:

File                                                   Program                                               Insns (A)  Insns (B)  Insns  (DIFF)
-----------------------------------------------------  ----------------------------------------------------  ---------  ---------  -------------
test_seg6_loop.bpf.linked3.o                           __add_egr_x                                               12423      12314  -109 (-0.88%)

Only one program is affected with minor change.

Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230406164510.1047757-1-yhs@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../progs/verifier_bounds_deduction_non_const.c    | 460 +++++++++++++++++++++
 1 file changed, 460 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c b/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c
index fe570d866139..823f727cf210 100644
--- a/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c
+++ b/tools/testing/selftests/bpf/progs/verifier_bounds_deduction_non_const.c
@@ -176,4 +176,464 @@ l1_%=:							\
 	: __clobber_all);
 }
 
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> > <non_const>, 1")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_9(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	r2 = 0;						\
+	if r2 > r0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> > <non_const>, 2")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_10(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 < 4 goto l0_%=;				\
+	r2 = 4;						\
+	if r2 > r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> >= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_11(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 < 4 goto l0_%=;				\
+	r2 = 3;						\
+	if r2 >= r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> < <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_12(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 > 4 goto l0_%=;				\
+	r2 = 4;						\
+	if r2 < r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> <= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_13(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 >= 4 goto l0_%=;				\
+	r2 = 4;						\
+	if r2 <= r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> == <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_14(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 < 3 goto l0_%=;				\
+	r2 = 2;						\
+	if r2 == r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> s> <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_15(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 s< 4 goto l0_%=;				\
+	r2 = 4;						\
+	if r2 s> r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> s>= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_16(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 s< 4 goto l0_%=;				\
+	r2 = 3;						\
+	if r2 s>= r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> s< <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_17(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 s> 4 goto l0_%=;				\
+	r2 = 4;						\
+	if r2 s< r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> s<= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_18(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 s> 4 goto l0_%=;				\
+	r2 = 5;						\
+	if r2 s<= r0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp64, <const> != <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_19(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if r0 < 3 goto l0_%=;				\
+	r2 = 2;						\
+	if r2 != r0 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> > <non_const>, 1")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_20(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	w2 = 0;						\
+	if w2 > w0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> > <non_const>, 2")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_21(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 < 4 goto l0_%=;				\
+	w2 = 4;						\
+	if w2 > w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> >= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_22(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 < 4 goto l0_%=;				\
+	w2 = 3;						\
+	if w2 >= w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> < <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_23(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 > 4 goto l0_%=;				\
+	w2 = 4;						\
+	if w2 < w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> <= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_24(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 >= 4 goto l0_%=;				\
+	w2 = 4;						\
+	if w2 <= w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> == <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_25(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 < 4 goto l0_%=;				\
+	w2 = 3;						\
+	if w2 == w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> s> <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_26(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 s< 4 goto l0_%=;				\
+	w2 = 4;						\
+	if w2 s> w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> s>= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_27(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 s< 4 goto l0_%=;				\
+	w2 = 3;						\
+	if w2 s>= w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> s< <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_28(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 s> 4 goto l0_%=;				\
+	w2 = 5;						\
+	if w2 s< w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> s<= <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_29(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 s>= 4 goto l0_%=;				\
+	w2 = 4;						\
+	if w2 s<= w0 goto l1_%=;				\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check deducing bounds from non-const, jmp32, <const> != <non_const>")
+__success __retval(0)
+__naked void deducing_bounds_from_non_const_30(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	if w0 < 3 goto l0_%=;				\
+	w2 = 2;						\
+	if w2 != w0 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:							\
+	r0 = 0;						\
+	exit;						\
+l1_%=:							\
+	r0 -= r1;					\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From f3f21349779776135349a8e6f114a1485b2476b7 Mon Sep 17 00:00:00 2001
From: Barret Rhoden <brho@google.com>
Date: Thu, 6 Apr 2023 20:18:08 -0400
Subject: bpf: ensure all memory is initialized in bpf_get_current_comm

BPF helpers that take an ARG_PTR_TO_UNINIT_MEM must ensure that all of
the memory is set, including beyond the end of the string.

Signed-off-by: Barret Rhoden <brho@google.com>
Link: https://lore.kernel.org/r/20230407001808.1622968-1-brho@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 6be16db9f188..b6a5cda5bb59 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -258,7 +258,7 @@ BPF_CALL_2(bpf_get_current_comm, char *, buf, u32, size)
 		goto err_clear;
 
 	/* Verifier guarantees that size > 0 */
-	strscpy(buf, task->comm, size);
+	strscpy_pad(buf, task->comm, size);
 	return 0;
 err_clear:
 	memset(buf, 0, size);
-- 
cgit v1.2.3-70-g09d2


From f8b648bf6628eb0a6c25be68e512bed6df2e5f24 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Wed, 5 Apr 2023 19:29:48 +0200
Subject: net: sunhme: move asm includes to below linux includes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A recent rearrangement of includes has lead to a problem on m68k
as flagged by the kernel test robot.

Resolve this by moving the block asm includes to below linux includes.
A side effect i that non-Sparc asm includes are now immediately
before Sparc asm includes, which seems nice.

Using sparse v0.6.4 I was able to reproduce this problem as follows
using the config provided by the kernel test robot:

$ wget https://download.01.org/0day-ci/archive/20230404/202304041748.0sQc4K4l-lkp@intel.com/config
$ cp config .config
$ make ARCH=m68k oldconfig
$ make ARCH=m68k C=2 M=drivers/net/ethernet/sun
   CC [M]  drivers/net/ethernet/sun/sunhme.o
 In file included from drivers/net/ethernet/sun/sunhme.c:19:
 ./arch/m68k/include/asm/irq.h:78:11: error: expected ‘;’ before ‘void’
    78 | asmlinkage void do_IRQ(int irq, struct pt_regs *regs);
       |           ^~~~~
       |           ;
 ./arch/m68k/include/asm/irq.h:78:40: warning: ‘struct pt_regs’ declared inside parameter list will not be visible outside of this definition or declaration
    78 | asmlinkage void do_IRQ(int irq, struct pt_regs *regs);
       |                                        ^~~~~~~

Compile tested only.

Fixes: 1ff4f42aef60 ("net: sunhme: Alphabetize includes")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304041748.0sQc4K4l-lkp@intel.com/
Signed-off-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230405-sunhme-includes-fix-v1-1-bf17cc5de20d@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sun/sunhme.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index ec85aef35bf9..b93613cd1994 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -14,9 +14,6 @@
  *     argument : macaddr=0x00,0x10,0x20,0x30,0x40,0x50
  */
 
-#include <asm/byteorder.h>
-#include <asm/dma.h>
-#include <asm/irq.h>
 #include <linux/bitops.h>
 #include <linux/crc32.h>
 #include <linux/delay.h>
@@ -45,6 +42,10 @@
 #include <linux/types.h>
 #include <linux/uaccess.h>
 
+#include <asm/byteorder.h>
+#include <asm/dma.h>
+#include <asm/irq.h>
+
 #ifdef CONFIG_SPARC
 #include <asm/auxio.h>
 #include <asm/idprom.h>
-- 
cgit v1.2.3-70-g09d2


From a9fda7a0b0331250c4af006f1862752dbefcab9c Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Wed, 5 Apr 2023 16:25:12 +0200
Subject: selftests: forwarding: hw_stats_l3: Detect failure to install
 counters

Running this test makes little sense if the enabled l3_stats are not
actually reported as "used". This can signify a failure of a driver to
install the necessary counters, or simply lack of support for enabling
in-HW counters on a given netdevice. It is generally impossible to tell
from the outside which it is. But more likely than not, if somebody is
running this on veth pairs, they do not intend to actually test that a
certain piece of HW can install in-HW counters for the veth. It is more
likely they are e.g. running the test by mistake.

Therefore detect that the counter has not been actually installed. In that
case, if the netdevice is one end of a veth pair, SKIP. Otherwise FAIL.

Suggested-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Tested-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/a86817961903cca5cb0aebf2b2a06294b8aa7dea.1680704172.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/hw_stats_l3.sh | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/forwarding/hw_stats_l3.sh b/tools/testing/selftests/net/forwarding/hw_stats_l3.sh
index 9c1f76e108af..432fe8469851 100755
--- a/tools/testing/selftests/net/forwarding/hw_stats_l3.sh
+++ b/tools/testing/selftests/net/forwarding/hw_stats_l3.sh
@@ -319,6 +319,19 @@ trap cleanup EXIT
 setup_prepare
 setup_wait
 
-tests_run
+used=$(ip -j stats show dev $rp1.200 group offload subgroup hw_stats_info |
+	   jq '.[].info.l3_stats.used')
+kind=$(ip -j -d link show dev $rp1 |
+	   jq -r '.[].linkinfo.info_kind')
+if [[ $used != true ]]; then
+	if [[ $kind == veth ]]; then
+		log_test_skip "l3_stats not offloaded on veth interface"
+		EXIT_STATUS=$ksft_skip
+	else
+		RET=1 log_test "l3_stats not offloaded"
+	fi
+else
+	tests_run
+fi
 
 exit $EXIT_STATUS
-- 
cgit v1.2.3-70-g09d2


From 05f3ab7780b3c0cfe26a8134606bdf641c4f4bb2 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Wed, 5 Apr 2023 17:10:25 +0200
Subject: net: ethernet: mtk_eth_soc: add code for offloading flows from wlan
 devices

WED version 2 (on MT7986 and later) can offload flows originating from
wireless devices.
In order to make that work, ndo_setup_tc needs to be implemented on the
netdevs. This adds the required code to offload flows coming in from WED,
while keeping track of the incoming wed index used for selecting the
correct PPE device.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.h     |   3 +
 drivers/net/ethernet/mediatek/mtk_ppe_offload.c |  40 ++++++----
 drivers/net/ethernet/mediatek/mtk_wed.c         | 101 ++++++++++++++++++++++++
 include/linux/soc/mediatek/mtk_wed.h            |   6 ++
 4 files changed, 136 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index 23c7abeb5c14..cdcf8534283e 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -1276,6 +1276,9 @@ int mtk_gmac_rgmii_path_setup(struct mtk_eth *eth, int mac_id);
 int mtk_eth_offload_init(struct mtk_eth *eth);
 int mtk_eth_setup_tc(struct net_device *dev, enum tc_setup_type type,
 		     void *type_data);
+int mtk_flow_offload_cmd(struct mtk_eth *eth, struct flow_cls_offload *cls,
+			 int ppe_index);
+void mtk_flow_offload_cleanup(struct mtk_eth *eth, struct list_head *list);
 void mtk_eth_set_dma_device(struct mtk_eth *eth, struct device *dma_dev);
 
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index 46634dc29d2f..02eebff02d45 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -235,7 +235,8 @@ out:
 }
 
 static int
-mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
+mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f,
+			 int ppe_index)
 {
 	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
 	struct flow_action_entry *act;
@@ -452,6 +453,7 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	entry->cookie = f->cookie;
 	memcpy(&entry->data, &foe, sizeof(entry->data));
 	entry->wed_index = wed_index;
+	entry->ppe_index = ppe_index;
 
 	err = mtk_foe_entry_commit(eth->ppe[entry->ppe_index], entry);
 	if (err < 0)
@@ -520,25 +522,15 @@ mtk_flow_offload_stats(struct mtk_eth *eth, struct flow_cls_offload *f)
 
 static DEFINE_MUTEX(mtk_flow_offload_mutex);
 
-static int
-mtk_eth_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv)
+int mtk_flow_offload_cmd(struct mtk_eth *eth, struct flow_cls_offload *cls,
+			 int ppe_index)
 {
-	struct flow_cls_offload *cls = type_data;
-	struct net_device *dev = cb_priv;
-	struct mtk_mac *mac = netdev_priv(dev);
-	struct mtk_eth *eth = mac->hw;
 	int err;
 
-	if (!tc_can_offload(dev))
-		return -EOPNOTSUPP;
-
-	if (type != TC_SETUP_CLSFLOWER)
-		return -EOPNOTSUPP;
-
 	mutex_lock(&mtk_flow_offload_mutex);
 	switch (cls->command) {
 	case FLOW_CLS_REPLACE:
-		err = mtk_flow_offload_replace(eth, cls);
+		err = mtk_flow_offload_replace(eth, cls, ppe_index);
 		break;
 	case FLOW_CLS_DESTROY:
 		err = mtk_flow_offload_destroy(eth, cls);
@@ -555,6 +547,26 @@ mtk_eth_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_pri
 	return err;
 }
 
+static int
+mtk_eth_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv)
+{
+	struct flow_cls_offload *cls = type_data;
+	struct net_device *dev = cb_priv;
+	struct mtk_mac *mac;
+	struct mtk_eth *eth;
+
+	mac = netdev_priv(dev);
+	eth = mac->hw;
+
+	if (!tc_can_offload(dev))
+		return -EOPNOTSUPP;
+
+	if (type != TC_SETUP_CLSFLOWER)
+		return -EOPNOTSUPP;
+
+	return mtk_flow_offload_cmd(eth, cls, 0);
+}
+
 static int
 mtk_eth_setup_tc_block(struct net_device *dev, struct flow_block_offload *f)
 {
diff --git a/drivers/net/ethernet/mediatek/mtk_wed.c b/drivers/net/ethernet/mediatek/mtk_wed.c
index 95d890870984..4c205afbd230 100644
--- a/drivers/net/ethernet/mediatek/mtk_wed.c
+++ b/drivers/net/ethernet/mediatek/mtk_wed.c
@@ -13,6 +13,8 @@
 #include <linux/mfd/syscon.h>
 #include <linux/debugfs.h>
 #include <linux/soc/mediatek/mtk_wed.h>
+#include <net/flow_offload.h>
+#include <net/pkt_cls.h>
 #include "mtk_eth_soc.h"
 #include "mtk_wed_regs.h"
 #include "mtk_wed.h"
@@ -41,6 +43,11 @@
 static struct mtk_wed_hw *hw_list[2];
 static DEFINE_MUTEX(hw_lock);
 
+struct mtk_wed_flow_block_priv {
+	struct mtk_wed_hw *hw;
+	struct net_device *dev;
+};
+
 static void
 wed_m32(struct mtk_wed_device *dev, u32 reg, u32 mask, u32 val)
 {
@@ -1745,6 +1752,99 @@ out:
 	mutex_unlock(&hw_lock);
 }
 
+static int
+mtk_wed_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv)
+{
+	struct mtk_wed_flow_block_priv *priv = cb_priv;
+	struct flow_cls_offload *cls = type_data;
+	struct mtk_wed_hw *hw = priv->hw;
+
+	if (!tc_can_offload(priv->dev))
+		return -EOPNOTSUPP;
+
+	if (type != TC_SETUP_CLSFLOWER)
+		return -EOPNOTSUPP;
+
+	return mtk_flow_offload_cmd(hw->eth, cls, hw->index);
+}
+
+static int
+mtk_wed_setup_tc_block(struct mtk_wed_hw *hw, struct net_device *dev,
+		       struct flow_block_offload *f)
+{
+	struct mtk_wed_flow_block_priv *priv;
+	static LIST_HEAD(block_cb_list);
+	struct flow_block_cb *block_cb;
+	struct mtk_eth *eth = hw->eth;
+	flow_setup_cb_t *cb;
+
+	if (!eth->soc->offload_version)
+		return -EOPNOTSUPP;
+
+	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
+		return -EOPNOTSUPP;
+
+	cb = mtk_wed_setup_tc_block_cb;
+	f->driver_block_list = &block_cb_list;
+
+	switch (f->command) {
+	case FLOW_BLOCK_BIND:
+		block_cb = flow_block_cb_lookup(f->block, cb, dev);
+		if (block_cb) {
+			flow_block_cb_incref(block_cb);
+			return 0;
+		}
+
+		priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+		if (!priv)
+			return -ENOMEM;
+
+		priv->hw = hw;
+		priv->dev = dev;
+		block_cb = flow_block_cb_alloc(cb, dev, priv, NULL);
+		if (IS_ERR(block_cb)) {
+			kfree(priv);
+			return PTR_ERR(block_cb);
+		}
+
+		flow_block_cb_incref(block_cb);
+		flow_block_cb_add(block_cb, f);
+		list_add_tail(&block_cb->driver_list, &block_cb_list);
+		return 0;
+	case FLOW_BLOCK_UNBIND:
+		block_cb = flow_block_cb_lookup(f->block, cb, dev);
+		if (!block_cb)
+			return -ENOENT;
+
+		if (!flow_block_cb_decref(block_cb)) {
+			flow_block_cb_remove(block_cb, f);
+			list_del(&block_cb->driver_list);
+			kfree(block_cb->cb_priv);
+		}
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int
+mtk_wed_setup_tc(struct mtk_wed_device *wed, struct net_device *dev,
+		 enum tc_setup_type type, void *type_data)
+{
+	struct mtk_wed_hw *hw = wed->hw;
+
+	if (hw->version < 2)
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case TC_SETUP_BLOCK:
+	case TC_SETUP_FT:
+		return mtk_wed_setup_tc_block(hw, dev, type_data);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
 void mtk_wed_add_hw(struct device_node *np, struct mtk_eth *eth,
 		    void __iomem *wdma, phys_addr_t wdma_phy,
 		    int index)
@@ -1764,6 +1864,7 @@ void mtk_wed_add_hw(struct device_node *np, struct mtk_eth *eth,
 		.irq_set_mask = mtk_wed_irq_set_mask,
 		.detach = mtk_wed_detach,
 		.ppe_check = mtk_wed_ppe_check,
+		.setup_tc = mtk_wed_setup_tc,
 	};
 	struct device_node *eth_np = eth->dev->of_node;
 	struct platform_device *pdev;
diff --git a/include/linux/soc/mediatek/mtk_wed.h b/include/linux/soc/mediatek/mtk_wed.h
index fd0b0605cf90..b2b28180dff7 100644
--- a/include/linux/soc/mediatek/mtk_wed.h
+++ b/include/linux/soc/mediatek/mtk_wed.h
@@ -6,6 +6,7 @@
 #include <linux/regmap.h>
 #include <linux/pci.h>
 #include <linux/skbuff.h>
+#include <linux/netdevice.h>
 
 #define MTK_WED_TX_QUEUES		2
 #define MTK_WED_RX_QUEUES		2
@@ -179,6 +180,8 @@ struct mtk_wed_ops {
 
 	u32 (*irq_get)(struct mtk_wed_device *dev, u32 mask);
 	void (*irq_set_mask)(struct mtk_wed_device *dev, u32 mask);
+	int (*setup_tc)(struct mtk_wed_device *wed, struct net_device *dev,
+			enum tc_setup_type type, void *type_data);
 };
 
 extern const struct mtk_wed_ops __rcu *mtk_soc_wed_ops;
@@ -237,6 +240,8 @@ mtk_wed_get_rx_capa(struct mtk_wed_device *dev)
 	(_dev)->ops->msg_update(_dev, _id, _msg, _len)
 #define mtk_wed_device_stop(_dev) (_dev)->ops->stop(_dev)
 #define mtk_wed_device_dma_reset(_dev) (_dev)->ops->reset_dma(_dev)
+#define mtk_wed_device_setup_tc(_dev, _netdev, _type, _type_data) \
+	(_dev)->ops->setup_tc(_dev, _netdev, _type, _type_data)
 #else
 static inline bool mtk_wed_device_active(struct mtk_wed_device *dev)
 {
@@ -255,6 +260,7 @@ static inline bool mtk_wed_device_active(struct mtk_wed_device *dev)
 #define mtk_wed_device_update_msg(_dev, _id, _msg, _len) -ENODEV
 #define mtk_wed_device_stop(_dev) do {} while (0)
 #define mtk_wed_device_dma_reset(_dev) do {} while (0)
+#define mtk_wed_device_setup_tc(_dev, _netdev, _type, _type_data) -EOPNOTSUPP
 #endif
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From e28531143b257a4cbfdc1c3358eff18cc3024783 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Wed, 5 Apr 2023 17:10:26 +0200
Subject: net: ethernet: mtk_eth_soc: mtk_ppe: prefer newly added l2 flows

When a device is roaming between interfaces and a new flow entry is
created, we should assume that its output device is more up to date than
whatever entry existed already.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mediatek/mtk_ppe.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index f9c9f2ea4206..dd9581334b05 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -635,10 +635,20 @@ void mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 static int
 mtk_foe_entry_commit_l2(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 {
+	struct mtk_flow_entry *prev;
+
 	entry->type = MTK_FLOW_TYPE_L2;
 
-	return rhashtable_insert_fast(&ppe->l2_flows, &entry->l2_node,
-				      mtk_flow_l2_ht_params);
+	prev = rhashtable_lookup_get_insert_fast(&ppe->l2_flows, &entry->l2_node,
+						 mtk_flow_l2_ht_params);
+	if (likely(!prev))
+		return 0;
+
+	if (IS_ERR(prev))
+		return PTR_ERR(prev);
+
+	return rhashtable_replace_fast(&ppe->l2_flows, &prev->l2_node,
+				       &entry->l2_node, mtk_flow_l2_ht_params);
 }
 
 int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
-- 
cgit v1.2.3-70-g09d2


From 3ebf5212bf042954666b19fe4ff5a98911b08128 Mon Sep 17 00:00:00 2001
From: Song Liu <song@kernel.org>
Date: Fri, 7 Apr 2023 12:01:30 -0700
Subject: selftests/bpf: Use PERF_COUNT_HW_CPU_CYCLES event for
 get_branch_snapshot

perf_event with type=PERF_TYPE_RAW and config=0x1b00 turned out to be not
reliable in ensuring LBR is active. Thus, test_progs:get_branch_snapshot is
not reliable in some systems. Replace it with PERF_COUNT_HW_CPU_CYCLES
event, which gives more consistent results.

Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20230407190130.2093736-1-song@kernel.org
---
 tools/testing/selftests/bpf/prog_tests/get_branch_snapshot.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/get_branch_snapshot.c b/tools/testing/selftests/bpf/prog_tests/get_branch_snapshot.c
index 3948da12a528..0394a1156d99 100644
--- a/tools/testing/selftests/bpf/prog_tests/get_branch_snapshot.c
+++ b/tools/testing/selftests/bpf/prog_tests/get_branch_snapshot.c
@@ -37,8 +37,8 @@ static int create_perf_events(void)
 
 	/* create perf event */
 	attr.size = sizeof(attr);
-	attr.type = PERF_TYPE_RAW;
-	attr.config = 0x1b00;
+	attr.type = PERF_TYPE_HARDWARE;
+	attr.config = PERF_COUNT_HW_CPU_CYCLES;
 	attr.sample_type = PERF_SAMPLE_BRANCH_STACK;
 	attr.branch_sample_type = PERF_SAMPLE_BRANCH_KERNEL |
 		PERF_SAMPLE_BRANCH_USER | PERF_SAMPLE_BRANCH_ANY;
-- 
cgit v1.2.3-70-g09d2


From b24f0b049e706c6fc6e822dfc519ee33c3865092 Mon Sep 17 00:00:00 2001
From: Wei Yongjun <weiyongjun1@huawei.com>
Date: Fri, 7 Apr 2023 08:14:26 +0000
Subject: bpftool: Set program type only if it differs from the desired one

After commit d6e6286a12e7 ("libbpf: disassociate section handler on explicit
bpf_program__set_type() call"), bpf_program__set_type() will force cleanup
the program's SEC() definition, this commit fixed the test helper but missed
the bpftool, which leads to bpftool prog autoattach broken as follows:

  $ bpftool prog load spi-xfer-r1v1.o /sys/fs/bpf/test autoattach
  Program spi_xfer_r1v1 does not support autoattach, falling back to pinning

This patch fix bpftool to set program type only if it differs.

Fixes: d6e6286a12e7 ("libbpf: disassociate section handler on explicit bpf_program__set_type() call")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230407081427.2621590-1-weiyongjun@huaweicloud.com
---
 tools/bpf/bpftool/prog.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 430f72306409..e5b613a7974c 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -1685,7 +1685,8 @@ static int load_with_options(int argc, char **argv, bool first_prog_only)
 		}
 
 		bpf_program__set_ifindex(pos, ifindex);
-		bpf_program__set_type(pos, prog_type);
+		if (bpf_program__type(pos) != prog_type)
+			bpf_program__set_type(pos, prog_type);
 		bpf_program__set_expected_attach_type(pos, expected_attach_type);
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 5855b0999de4213bf51d856a345c4b53f2304e33 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 7 Apr 2023 18:41:25 +0300
Subject: selftests/bpf: Prevent infinite loop in veristat when base file is
 too short

The following example forces veristat to loop indefinitely:

$ cat two-ok
file_name,prog_name,verdict,total_states
file-a,a,success,12
file-b,b,success,67

$ cat add-failure
file_name,prog_name,verdict,total_states
file-a,a,success,12
file-b,b,success,67
file-b,c,failure,32

$ veristat -C two-ok add-failure
  <does not return>

The loop is caused by handle_comparison_mode() not checking if `base`
variable points to `fallback_stats` prior advancing joined results
using `base`.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230407154125.896927-1-eddyz87@gmail.com
---
 tools/testing/selftests/bpf/veristat.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index 53d7ec168268..e05954e20bba 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -1824,18 +1824,22 @@ static int handle_comparison_mode(void)
 			join->stats_b = comp;
 			i++;
 			j++;
-		} else if (comp == &fallback_stats || r < 0) {
+		} else if (base != &fallback_stats && (comp == &fallback_stats || r < 0)) {
 			join->file_name = base->file_name;
 			join->prog_name = base->prog_name;
 			join->stats_a = base;
 			join->stats_b = NULL;
 			i++;
-		} else {
+		} else if (comp != &fallback_stats && (base == &fallback_stats || r > 0)) {
 			join->file_name = comp->file_name;
 			join->prog_name = comp->prog_name;
 			join->stats_a = NULL;
 			join->stats_b = comp;
 			j++;
+		} else {
+			fprintf(stderr, "%s:%d: should never reach here i=%i, j=%i",
+				__FILE__, __LINE__, i, j);
+			return -EINVAL;
 		}
 		env.join_stat_cnt += 1;
 	}
-- 
cgit v1.2.3-70-g09d2


From ce639b767139b27500fa5bc3f8d7126b18d0d310 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 4 Apr 2023 11:44:57 +0530
Subject: net: ethernet: ti: am65-cpsw: Move mode specific config to
 mac_config()

Move the interface mode specific configuration to the mac_config()
callback am65_cpsw_nuss_mac_config().

Also, do not reset the MAC Control register on mac_link_down(). Only
clear those bits that can possibly be set in mac_link_up().

Let the MAC remain in IDLE state after mac_link_down(). Bring it out of
the IDLE state on mac_link_up().

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 035dc0355b49..f08f0e7d00d5 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -1506,9 +1506,13 @@ static void am65_cpsw_nuss_mac_config(struct phylink_config *config, unsigned in
 	struct am65_cpsw_common *common = port->common;
 
 	if (common->pdata.extra_modes & BIT(state->interface)) {
-		if (state->interface == PHY_INTERFACE_MODE_SGMII)
+		if (state->interface == PHY_INTERFACE_MODE_SGMII) {
 			writel(ADVERTISE_SGMII,
 			       port->sgmii_base + AM65_CPSW_SGMII_MR_ADV_ABILITY_REG);
+			cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_EXT_EN);
+		} else {
+			cpsw_sl_ctl_clr(port->slave.mac_sl, CPSW_SL_CTL_EXT_EN);
+		}
 
 		writel(AM65_CPSW_SGMII_CONTROL_MR_AN_ENABLE,
 		       port->sgmii_base + AM65_CPSW_SGMII_CONTROL_REG);
@@ -1523,6 +1527,7 @@ static void am65_cpsw_nuss_mac_link_down(struct phylink_config *config, unsigned
 	struct am65_cpsw_port *port = container_of(slave, struct am65_cpsw_port, slave);
 	struct am65_cpsw_common *common = port->common;
 	struct net_device *ndev = port->ndev;
+	u32 mac_control;
 	int tmo;
 
 	/* disable forwarding */
@@ -1534,7 +1539,14 @@ static void am65_cpsw_nuss_mac_link_down(struct phylink_config *config, unsigned
 	dev_dbg(common->dev, "down msc_sl %08x tmo %d\n",
 		cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_MACSTATUS), tmo);
 
-	cpsw_sl_ctl_reset(port->slave.mac_sl);
+	/* All the bits that am65_cpsw_nuss_mac_link_up() can possibly set */
+	mac_control = CPSW_SL_CTL_GMII_EN | CPSW_SL_CTL_GIG | CPSW_SL_CTL_IFCTL_A |
+		      CPSW_SL_CTL_FULLDUPLEX | CPSW_SL_CTL_RX_FLOW_EN | CPSW_SL_CTL_TX_FLOW_EN;
+	/* If interface mode is RGMII, CPSW_SL_CTL_EXT_EN might have been set for 10 Mbps */
+	if (phy_interface_mode_is_rgmii(interface))
+		mac_control |= CPSW_SL_CTL_EXT_EN;
+	/* Only clear those bits that can be set by am65_cpsw_nuss_mac_link_up() */
+	cpsw_sl_ctl_clr(port->slave.mac_sl, mac_control);
 
 	am65_cpsw_qos_link_down(ndev);
 	netif_tx_stop_all_queues(ndev);
@@ -1551,10 +1563,12 @@ static void am65_cpsw_nuss_mac_link_up(struct phylink_config *config, struct phy
 	u32 mac_control = CPSW_SL_CTL_GMII_EN;
 	struct net_device *ndev = port->ndev;
 
+	/* Bring the port out of idle state */
+	cpsw_sl_ctl_clr(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE);
+
 	if (speed == SPEED_1000)
 		mac_control |= CPSW_SL_CTL_GIG;
-	if (interface == PHY_INTERFACE_MODE_SGMII)
-		mac_control |= CPSW_SL_CTL_EXT_EN;
+	/* TODO: Verify whether in-band is necessary for 10 Mbps RGMII */
 	if (speed == SPEED_10 && phy_interface_mode_is_rgmii(interface))
 		/* Can be used with in band mode only */
 		mac_control |= CPSW_SL_CTL_EXT_EN;
-- 
cgit v1.2.3-70-g09d2


From 4e003d61e79504532babfb1d4e6b0767b32739d1 Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 4 Apr 2023 11:44:58 +0530
Subject: net: ethernet: ti: am65-cpsw: Enable QSGMII for J784S4 CPSW9G

TI's J784S4 SoC supports QSGMII mode with the CPSW9G instance of the
CPSW Ethernet Switch. Add a new compatible for J784S4 SoC and enable
QSGMII support for it by adding QSGMII mode to the extra_modes member of
the "j784s4_cpswxg_pdata" SoC data.

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index f08f0e7d00d5..fb458120de43 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2810,12 +2810,20 @@ static const struct am65_cpsw_pdata j721e_cpswxg_pdata = {
 	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII) | BIT(PHY_INTERFACE_MODE_SGMII),
 };
 
+static const struct am65_cpsw_pdata j784s4_cpswxg_pdata = {
+	.quirks = 0,
+	.ale_dev_id = "am64-cpswxg",
+	.fdqring_mode = K3_RINGACC_RING_MODE_MESSAGE,
+	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII),
+};
+
 static const struct of_device_id am65_cpsw_nuss_of_mtable[] = {
 	{ .compatible = "ti,am654-cpsw-nuss", .data = &am65x_sr1_0},
 	{ .compatible = "ti,j721e-cpsw-nuss", .data = &j721e_pdata},
 	{ .compatible = "ti,am642-cpsw-nuss", .data = &am64x_cpswxg_pdata},
 	{ .compatible = "ti,j7200-cpswxg-nuss", .data = &j7200_cpswxg_pdata},
 	{ .compatible = "ti,j721e-cpswxg-nuss", .data = &j721e_cpswxg_pdata},
+	{ .compatible = "ti,j784s4-cpswxg-nuss", .data = &j784s4_cpswxg_pdata},
 	{ /* sentinel */ },
 };
 MODULE_DEVICE_TABLE(of, am65_cpsw_nuss_of_mtable);
-- 
cgit v1.2.3-70-g09d2


From 8e672b560e0b52ab9497fa48f361d5c174937a3e Mon Sep 17 00:00:00 2001
From: Siddharth Vadapalli <s-vadapalli@ti.com>
Date: Tue, 4 Apr 2023 11:44:59 +0530
Subject: net: ethernet: ti: am65-cpsw: Enable USXGMII mode for J784S4 CPSW9G

TI's J784S4 SoC supports USXGMII mode. Add USXGMII mode to the
extra_modes member of the J784S4 SoC data.

Configure MAC control register for supporting USXGMII mode and add
MAC_5000FD in the "mac_capabilities" member of struct "phylink_config".

Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index fb458120de43..11cbcd9e2c72 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -1514,6 +1514,14 @@ static void am65_cpsw_nuss_mac_config(struct phylink_config *config, unsigned in
 			cpsw_sl_ctl_clr(port->slave.mac_sl, CPSW_SL_CTL_EXT_EN);
 		}
 
+		if (state->interface == PHY_INTERFACE_MODE_USXGMII) {
+			cpsw_sl_ctl_set(port->slave.mac_sl,
+					CPSW_SL_CTL_XGIG | CPSW_SL_CTL_XGMII_EN);
+		} else {
+			cpsw_sl_ctl_clr(port->slave.mac_sl,
+					CPSW_SL_CTL_XGIG | CPSW_SL_CTL_XGMII_EN);
+		}
+
 		writel(AM65_CPSW_SGMII_CONTROL_MR_AN_ENABLE,
 		       port->sgmii_base + AM65_CPSW_SGMII_CONTROL_REG);
 	}
@@ -2171,7 +2179,8 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 	/* Configuring Phylink */
 	port->slave.phylink_config.dev = &port->ndev->dev;
 	port->slave.phylink_config.type = PHYLINK_NETDEV;
-	port->slave.phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100 | MAC_1000FD;
+	port->slave.phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100 |
+						      MAC_1000FD | MAC_5000FD;
 	port->slave.phylink_config.mac_managed_pm = true; /* MAC does PM */
 
 	switch (port->slave.phy_if) {
@@ -2189,6 +2198,7 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 
 	case PHY_INTERFACE_MODE_QSGMII:
 	case PHY_INTERFACE_MODE_SGMII:
+	case PHY_INTERFACE_MODE_USXGMII:
 		if (common->pdata.extra_modes & BIT(port->slave.phy_if)) {
 			__set_bit(port->slave.phy_if,
 				  port->slave.phylink_config.supported_interfaces);
@@ -2814,7 +2824,7 @@ static const struct am65_cpsw_pdata j784s4_cpswxg_pdata = {
 	.quirks = 0,
 	.ale_dev_id = "am64-cpswxg",
 	.fdqring_mode = K3_RINGACC_RING_MODE_MESSAGE,
-	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII),
+	.extra_modes = BIT(PHY_INTERFACE_MODE_QSGMII) | BIT(PHY_INTERFACE_MODE_USXGMII),
 };
 
 static const struct of_device_id am65_cpsw_nuss_of_mtable[] = {
-- 
cgit v1.2.3-70-g09d2


From 07e75db6b1b2e96f2cb727c2d2f64c8ed36de5dc Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 6 Apr 2023 15:54:12 +0300
Subject: net: stmmac: remove set but unused mask in
 stmmac_ethtool_set_link_ksettings()

This was never used since the commit that added it - e58bb43f5e43
("stmmac: initial support to manage pcs modes").

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230406125412.48790-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 35c8dd92d369..2ae73ab842d4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -393,19 +393,10 @@ stmmac_ethtool_set_link_ksettings(struct net_device *dev,
 
 	if (priv->hw->pcs & STMMAC_PCS_RGMII ||
 	    priv->hw->pcs & STMMAC_PCS_SGMII) {
-		u32 mask = ADVERTISED_Autoneg | ADVERTISED_Pause;
-
 		/* Only support ANE */
 		if (cmd->base.autoneg != AUTONEG_ENABLE)
 			return -EINVAL;
 
-		mask &= (ADVERTISED_1000baseT_Half |
-			ADVERTISED_1000baseT_Full |
-			ADVERTISED_100baseT_Half |
-			ADVERTISED_100baseT_Full |
-			ADVERTISED_10baseT_Half |
-			ADVERTISED_10baseT_Full);
-
 		mutex_lock(&priv->lock);
 		stmmac_pcs_ctrl_ane(priv, priv->ioaddr, 1, priv->hw->ps, 0);
 		mutex_unlock(&priv->lock);
-- 
cgit v1.2.3-70-g09d2


From 9f12541d684b9250ed14bfca2cc1b106db1c0e74 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Thu, 6 Apr 2023 19:30:09 +0200
Subject: net: stmmac: dwmac-anarion: Use annotation __iomem for register base

Use __iomem annotation the register base: the ctl_block field of struct
anarion_gmac. I believe this is the normal practice for such variables.

By doing so some casting is avoided.
And sparse no longer reports:

 .../dwmac-anarion.c:29:23: warning: incorrect type in argument 1 (different address spaces)
 .../dwmac-anarion.c:29:23:    expected void const volatile [noderef] __iomem *addr
 .../dwmac-anarion.c:29:23:    got void *
 .../dwmac-anarion.c:34:22: warning: incorrect type in argument 2 (different address spaces)
 .../dwmac-anarion.c:34:22:    expected void volatile [noderef] __iomem *addr
 .../dwmac-anarion.c:34:22:    got void *

No functional change intended.
Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c
index dfbaea06d108..2357e77434fb 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c
@@ -20,18 +20,18 @@
 #define  GMAC_CONFIG_INTF_RGMII		(0x1 << 0)
 
 struct anarion_gmac {
-	uintptr_t ctl_block;
+	void __iomem *ctl_block;
 	uint32_t phy_intf_sel;
 };
 
 static uint32_t gmac_read_reg(struct anarion_gmac *gmac, uint8_t reg)
 {
-	return readl((void *)(gmac->ctl_block + reg));
+	return readl(gmac->ctl_block + reg);
 };
 
 static void gmac_write_reg(struct anarion_gmac *gmac, uint8_t reg, uint32_t val)
 {
-	writel(val, (void *)(gmac->ctl_block + reg));
+	writel(val, gmac->ctl_block + reg);
 }
 
 static int anarion_gmac_init(struct platform_device *pdev, void *priv)
@@ -77,7 +77,7 @@ static struct anarion_gmac *anarion_config_dt(struct platform_device *pdev)
 	if (!gmac)
 		return ERR_PTR(-ENOMEM);
 
-	gmac->ctl_block = (uintptr_t)ctl_block;
+	gmac->ctl_block = ctl_block;
 
 	err = of_get_phy_mode(pdev->dev.of_node, &phy_mode);
 	if (err)
-- 
cgit v1.2.3-70-g09d2


From 51fe084b17e795f0315d787077510364172cc350 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Thu, 6 Apr 2023 19:30:10 +0200
Subject: net: stmmac: dwmac-anarion: Always return struct anarion_gmac * from
 anarion_config_dt()

Always return struct anarion_gmac * from anarion_config_dt().
In the case where ctl_block was an error pointer it was being
returned directly. Which sparse flags as follows:

 .../dwmac-anarion.c:73:24: warning: incorrect type in return expression (different address spaces)
 .../dwmac-anarion.c:73:24:    expected struct anarion_gmac *
 .../dwmac-anarion.c:73:24:    got void [noderef] __iomem *[assigned] ctl_block

Avoid this by converting the error pointer to an error.
And then reversing the conversion.

As a side effect, the error can be used for logging purposes,
subjectively, leading to a minor cleanup.

No functional change intended.
Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c
index 2357e77434fb..9354bf419112 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c
@@ -68,9 +68,9 @@ static struct anarion_gmac *anarion_config_dt(struct platform_device *pdev)
 
 	ctl_block = devm_platform_ioremap_resource(pdev, 1);
 	if (IS_ERR(ctl_block)) {
-		dev_err(&pdev->dev, "Cannot get reset region (%ld)!\n",
-			PTR_ERR(ctl_block));
-		return ctl_block;
+		err = PTR_ERR(ctl_block);
+		dev_err(&pdev->dev, "Cannot get reset region (%d)!\n", err);
+		return ERR_PTR(err);
 	}
 
 	gmac = devm_kzalloc(&pdev->dev, sizeof(*gmac), GFP_KERNEL);
-- 
cgit v1.2.3-70-g09d2


From 48b7ea1d22ddf657e2692fb2b399138a02d31c17 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 6 Apr 2023 19:46:34 +0000
Subject: net: make SO_BUSY_POLL available to all users

After commit 217f69743681 ("net: busy-poll: allow preemption
in sk_busy_loop()"), a thread willing to use busy polling
is not hurting other threads anymore in a non preempt kernel.

I think it is safe to remove CAP_NET_ADMIN check.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230406194634.1804691-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/sock.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index c25888795390..5440e67bcfe3 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1396,15 +1396,10 @@ set_sndbuf:
 
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	case SO_BUSY_POLL:
-		/* allow unprivileged users to decrease the value */
-		if ((val > sk->sk_ll_usec) && !sockopt_capable(CAP_NET_ADMIN))
-			ret = -EPERM;
-		else {
-			if (val < 0)
-				ret = -EINVAL;
-			else
-				WRITE_ONCE(sk->sk_ll_usec, val);
-		}
+		if (val < 0)
+			ret = -EINVAL;
+		else
+			WRITE_ONCE(sk->sk_ll_usec, val);
 		break;
 	case SO_PREFER_BUSY_POLL:
 		if (valbool && !sockopt_capable(CAP_NET_ADMIN))
-- 
cgit v1.2.3-70-g09d2


From 5a17818682cf43ad0fdd6035945f3b7a8c9dc5e9 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 6 Apr 2023 14:42:46 +0300
Subject: net: dsa: replace NETDEV_PRE_CHANGE_HWTSTAMP notifier with a stub

There was a sort of rush surrounding commit 88c0a6b503b7 ("net: create a
netdev notifier for DSA to reject PTP on DSA master"), due to a desire
to convert DSA's attempt to deny TX timestamping on a DSA master to
something that doesn't block the kernel-wide API conversion from
ndo_eth_ioctl() to ndo_hwtstamp_set().

What was required was a mechanism that did not depend on ndo_eth_ioctl(),
and what was provided was a mechanism that did not depend on
ndo_eth_ioctl(), while at the same time introducing something that
wasn't absolutely necessary - a new netdev notifier.

There have been objections from Jakub Kicinski that using notifiers in
general when they are not absolutely necessary creates complications to
the control flow and difficulties to maintainers who look at the code.
So there is a desire to not use notifiers.

In addition to that, the notifier chain gets called even if there is no
DSA in the system and no one is interested in applying any restriction.

Take the model of udp_tunnel_nic_ops and introduce a stub mechanism,
through which net/core/dev_ioctl.c can call into DSA even when
CONFIG_NET_DSA=m.

Compared to the code that existed prior to the notifier conversion, aka
what was added in commits:
- 4cfab3566710 ("net: dsa: Add wrappers for overloaded ndo_ops")
- 3369afba1e46 ("net: Call into DSA netdevice_ops wrappers")

this is different because we are not overloading any struct
net_device_ops of the DSA master anymore, but rather, we are exposing a
rather specific functionality which is orthogonal to which API is used
to enable it - ndo_eth_ioctl() or ndo_hwtstamp_set().

Also, what is similar is that both approaches use function pointers to
get from built-in code to DSA.

There is no point in replicating the function pointers towards
__dsa_master_hwtstamp_validate() once for every CPU port (dev->dsa_ptr).
Instead, it is sufficient to introduce a singleton struct dsa_stubs,
built into the kernel, which contains a single function pointer to
__dsa_master_hwtstamp_validate().

I find this approach preferable to what we had originally, because
dev->dsa_ptr->netdev_ops->ndo_do_ioctl() used to require going through
struct dsa_port (dev->dsa_ptr), and so, this was incompatible with any
attempts to add any data encapsulation and hide DSA data structures from
the outside world.

Link: https://lore.kernel.org/netdev/20230403083019.120b72fd@kernel.org/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  6 ------
 include/net/dsa_stubs.h   | 48 +++++++++++++++++++++++++++++++++++++++++++++++
 net/Makefile              |  2 +-
 net/core/dev.c            |  2 +-
 net/core/dev_ioctl.c      | 12 ++----------
 net/dsa/Makefile          |  6 ++++++
 net/dsa/dsa.c             | 19 +++++++++++++++++++
 net/dsa/master.c          |  2 +-
 net/dsa/master.h          |  2 +-
 net/dsa/slave.c           | 11 -----------
 net/dsa/stubs.c           | 10 ++++++++++
 11 files changed, 89 insertions(+), 31 deletions(-)
 create mode 100644 include/net/dsa_stubs.h
 create mode 100644 net/dsa/stubs.c

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a740be3bb911..1c25b39681b3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2878,7 +2878,6 @@ enum netdev_cmd {
 	NETDEV_OFFLOAD_XSTATS_REPORT_USED,
 	NETDEV_OFFLOAD_XSTATS_REPORT_DELTA,
 	NETDEV_XDP_FEAT_CHANGE,
-	NETDEV_PRE_CHANGE_HWTSTAMP,
 };
 const char *netdev_cmd_to_name(enum netdev_cmd cmd);
 
@@ -2929,11 +2928,6 @@ struct netdev_notifier_pre_changeaddr_info {
 	const unsigned char *dev_addr;
 };
 
-struct netdev_notifier_hwtstamp_info {
-	struct netdev_notifier_info info; /* must be first */
-	struct kernel_hwtstamp_config *config;
-};
-
 enum netdev_offload_xstats_type {
 	NETDEV_OFFLOAD_XSTATS_TYPE_L3 = 1,
 };
diff --git a/include/net/dsa_stubs.h b/include/net/dsa_stubs.h
new file mode 100644
index 000000000000..361811750a54
--- /dev/null
+++ b/include/net/dsa_stubs.h
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * include/net/dsa_stubs.h - Stubs for the Distributed Switch Architecture framework
+ */
+
+#include <linux/mutex.h>
+#include <linux/netdevice.h>
+#include <linux/net_tstamp.h>
+#include <net/dsa.h>
+
+#if IS_ENABLED(CONFIG_NET_DSA)
+
+extern const struct dsa_stubs *dsa_stubs;
+
+struct dsa_stubs {
+	int (*master_hwtstamp_validate)(struct net_device *dev,
+					const struct kernel_hwtstamp_config *config,
+					struct netlink_ext_ack *extack);
+};
+
+static inline int dsa_master_hwtstamp_validate(struct net_device *dev,
+					       const struct kernel_hwtstamp_config *config,
+					       struct netlink_ext_ack *extack)
+{
+	if (!netdev_uses_dsa(dev))
+		return 0;
+
+	/* rtnl_lock() is a sufficient guarantee, because as long as
+	 * netdev_uses_dsa() returns true, the dsa_core module is still
+	 * registered, and so, dsa_unregister_stubs() couldn't have run.
+	 * For netdev_uses_dsa() to start returning false, it would imply that
+	 * dsa_master_teardown() has executed, which requires rtnl_lock().
+	 */
+	ASSERT_RTNL();
+
+	return dsa_stubs->master_hwtstamp_validate(dev, config, extack);
+}
+
+#else
+
+static inline int dsa_master_hwtstamp_validate(struct net_device *dev,
+					       const struct kernel_hwtstamp_config *config,
+					       struct netlink_ext_ack *extack)
+{
+	return 0;
+}
+
+#endif
diff --git a/net/Makefile b/net/Makefile
index 0914bea9c335..87592009366f 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -24,7 +24,7 @@ obj-$(CONFIG_PACKET)		+= packet/
 obj-$(CONFIG_NET_KEY)		+= key/
 obj-$(CONFIG_BRIDGE)		+= bridge/
 obj-$(CONFIG_NET_DEVLINK)	+= devlink/
-obj-$(CONFIG_NET_DSA)		+= dsa/
+obj-y				+= dsa/
 obj-$(CONFIG_ATALK)		+= appletalk/
 obj-$(CONFIG_X25)		+= x25/
 obj-$(CONFIG_LAPB)		+= lapb/
diff --git a/net/core/dev.c b/net/core/dev.c
index 7ce5985be84b..480600a075ce 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1612,7 +1612,7 @@ const char *netdev_cmd_to_name(enum netdev_cmd cmd)
 	N(SVLAN_FILTER_PUSH_INFO) N(SVLAN_FILTER_DROP_INFO)
 	N(PRE_CHANGEADDR) N(OFFLOAD_XSTATS_ENABLE) N(OFFLOAD_XSTATS_DISABLE)
 	N(OFFLOAD_XSTATS_REPORT_USED) N(OFFLOAD_XSTATS_REPORT_DELTA)
-	N(XDP_FEAT_CHANGE) N(PRE_CHANGE_HWTSTAMP)
+	N(XDP_FEAT_CHANGE)
 	}
 #undef N
 	return "UNKNOWN_NETDEV_EVENT";
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 6d772837eb3f..3730945ee294 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -7,7 +7,7 @@
 #include <linux/net_tstamp.h>
 #include <linux/wireless.h>
 #include <linux/if_bridge.h>
-#include <net/dsa.h>
+#include <net/dsa_stubs.h>
 #include <net/wext.h>
 
 #include "dev.h"
@@ -259,9 +259,6 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 
 static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 {
-	struct netdev_notifier_hwtstamp_info info = {
-		.info.dev = dev,
-	};
 	struct kernel_hwtstamp_config kernel_cfg;
 	struct netlink_ext_ack extack = {};
 	struct hwtstamp_config cfg;
@@ -276,12 +273,7 @@ static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 	if (err)
 		return err;
 
-	info.info.extack = &extack;
-	info.config = &kernel_cfg;
-
-	err = call_netdevice_notifiers_info(NETDEV_PRE_CHANGE_HWTSTAMP,
-					    &info.info);
-	err = notifier_to_errno(err);
+	err = dsa_master_hwtstamp_validate(dev, &kernel_cfg, &extack);
 	if (err) {
 		if (extack._msg)
 			netdev_err(dev, "%s\n", extack._msg);
diff --git a/net/dsa/Makefile b/net/dsa/Makefile
index cc7e93a562fe..3835de286116 100644
--- a/net/dsa/Makefile
+++ b/net/dsa/Makefile
@@ -1,4 +1,10 @@
 # SPDX-License-Identifier: GPL-2.0
+
+# the stubs are built-in whenever DSA is built-in or module
+ifdef CONFIG_NET_DSA
+obj-y := stubs.o
+endif
+
 # the core
 obj-$(CONFIG_NET_DSA) += dsa_core.o
 dsa_core-y += \
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index e5f156940c67..ab1afe67fd18 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -17,6 +17,7 @@
 #include <linux/of.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
+#include <net/dsa_stubs.h>
 #include <net/sch_generic.h>
 
 #include "devlink.h"
@@ -1702,6 +1703,20 @@ bool dsa_mdb_present_in_other_db(struct dsa_switch *ds, int port,
 }
 EXPORT_SYMBOL_GPL(dsa_mdb_present_in_other_db);
 
+static const struct dsa_stubs __dsa_stubs = {
+	.master_hwtstamp_validate = __dsa_master_hwtstamp_validate,
+};
+
+static void dsa_register_stubs(void)
+{
+	dsa_stubs = &__dsa_stubs;
+}
+
+static void dsa_unregister_stubs(void)
+{
+	dsa_stubs = NULL;
+}
+
 static int __init dsa_init_module(void)
 {
 	int rc;
@@ -1721,6 +1736,8 @@ static int __init dsa_init_module(void)
 	if (rc)
 		goto netlink_register_fail;
 
+	dsa_register_stubs();
+
 	return 0;
 
 netlink_register_fail:
@@ -1735,6 +1752,8 @@ module_init(dsa_init_module);
 
 static void __exit dsa_cleanup_module(void)
 {
+	dsa_unregister_stubs();
+
 	rtnl_link_unregister(&dsa_link_ops);
 
 	dsa_slave_unregister_notifier();
diff --git a/net/dsa/master.c b/net/dsa/master.c
index c2cabe6248b1..6be89ab0cc01 100644
--- a/net/dsa/master.c
+++ b/net/dsa/master.c
@@ -198,7 +198,7 @@ static void dsa_master_get_strings(struct net_device *dev, uint32_t stringset,
 /* Deny PTP operations on master if there is at least one switch in the tree
  * that is PTP capable.
  */
-int dsa_master_pre_change_hwtstamp(struct net_device *dev,
+int __dsa_master_hwtstamp_validate(struct net_device *dev,
 				   const struct kernel_hwtstamp_config *config,
 				   struct netlink_ext_ack *extack)
 {
diff --git a/net/dsa/master.h b/net/dsa/master.h
index 80842f4e27f7..76e39d3ec909 100644
--- a/net/dsa/master.h
+++ b/net/dsa/master.h
@@ -15,7 +15,7 @@ int dsa_master_lag_setup(struct net_device *lag_dev, struct dsa_port *cpu_dp,
 			 struct netlink_ext_ack *extack);
 void dsa_master_lag_teardown(struct net_device *lag_dev,
 			     struct dsa_port *cpu_dp);
-int dsa_master_pre_change_hwtstamp(struct net_device *dev,
+int __dsa_master_hwtstamp_validate(struct net_device *dev,
 				   const struct kernel_hwtstamp_config *config,
 				   struct netlink_ext_ack *extack);
 
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 8abc1658ac47..165bb2cb8431 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -3289,7 +3289,6 @@ static int dsa_master_changeupper(struct net_device *dev,
 static int dsa_slave_netdevice_event(struct notifier_block *nb,
 				     unsigned long event, void *ptr)
 {
-	struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr);
 	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
 
 	switch (event) {
@@ -3419,16 +3418,6 @@ static int dsa_slave_netdevice_event(struct notifier_block *nb,
 
 		return NOTIFY_OK;
 	}
-	case NETDEV_PRE_CHANGE_HWTSTAMP: {
-		struct netdev_notifier_hwtstamp_info *info = ptr;
-		int err;
-
-		if (!netdev_uses_dsa(dev))
-			return NOTIFY_DONE;
-
-		err = dsa_master_pre_change_hwtstamp(dev, info->config, extack);
-		return notifier_from_errno(err);
-	}
 	default:
 		break;
 	}
diff --git a/net/dsa/stubs.c b/net/dsa/stubs.c
new file mode 100644
index 000000000000..2ed8a6c85fbf
--- /dev/null
+++ b/net/dsa/stubs.c
@@ -0,0 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Stubs for DSA functionality called by the core network stack.
+ * These are necessary because CONFIG_NET_DSA can be a module, and built-in
+ * code cannot directly call symbols exported by modules.
+ */
+#include <net/dsa_stubs.h>
+
+const struct dsa_stubs *dsa_stubs;
+EXPORT_SYMBOL_GPL(dsa_stubs);
-- 
cgit v1.2.3-70-g09d2


From 87f1c15e8759a006e39ddb7556f6829067f8dc3e Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:04 +0200
Subject: net: nfp: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230407145911.79642-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/netronome/nfp/nfp_hwmon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/netronome/nfp/nfp_hwmon.c b/drivers/net/ethernet/netronome/nfp/nfp_hwmon.c
index 5cabb1aa9c0c..0d6c59d6d4ae 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_hwmon.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_hwmon.c
@@ -115,7 +115,7 @@ static const struct hwmon_channel_info nfp_power = {
 	.config = nfp_power_config,
 };
 
-static const struct hwmon_channel_info *nfp_hwmon_info[] = {
+static const struct hwmon_channel_info * const nfp_hwmon_info[] = {
 	&nfp_chip,
 	&nfp_temp,
 	&nfp_power,
-- 
cgit v1.2.3-70-g09d2


From bc1585f611b2fe0562ef7815ed02f1fb0708ce72 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:05 +0200
Subject: net: aquantia: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20230407145911.79642-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c b/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c
index d3526cd38f3d..414b2e448d59 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_drvinfo.c
@@ -124,7 +124,7 @@ static const struct hwmon_channel_info aq_hwmon_temp = {
 	.config = aq_hwmon_temp_config,
 };
 
-static const struct hwmon_channel_info *aq_hwmon_info[] = {
+static const struct hwmon_channel_info * const aq_hwmon_info[] = {
 	&aq_hwmon_temp,
 	NULL,
 };
-- 
cgit v1.2.3-70-g09d2


From 234d79a5f82698fe100ec30ab8b7b8a91c17dd34 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:06 +0200
Subject: net: phy: aquantia: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230407145911.79642-3-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/aquantia_hwmon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/aquantia_hwmon.c b/drivers/net/phy/aquantia_hwmon.c
index 19c4c280a6cd..0da451e46f69 100644
--- a/drivers/net/phy/aquantia_hwmon.c
+++ b/drivers/net/phy/aquantia_hwmon.c
@@ -210,7 +210,7 @@ static const struct hwmon_channel_info aqr_hwmon_temp = {
 	.config = aqr_hwmon_temp_config,
 };
 
-static const struct hwmon_channel_info *aqr_hwmon_info[] = {
+static const struct hwmon_channel_info * const aqr_hwmon_info[] = {
 	&aqr_hwmon_chip,
 	&aqr_hwmon_temp,
 	NULL,
-- 
cgit v1.2.3-70-g09d2


From 211f70be25773a2802470f8d1ea418b9baf14e61 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:07 +0200
Subject: net: phy: bcm54140: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230407145911.79642-4-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/bcm54140.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/bcm54140.c b/drivers/net/phy/bcm54140.c
index d8f3024860dc..d43076592f81 100644
--- a/drivers/net/phy/bcm54140.c
+++ b/drivers/net/phy/bcm54140.c
@@ -364,7 +364,7 @@ static int bcm54140_hwmon_write(struct device *dev,
 	}
 }
 
-static const struct hwmon_channel_info *bcm54140_hwmon_info[] = {
+static const struct hwmon_channel_info * const bcm54140_hwmon_info[] = {
 	HWMON_CHANNEL_INFO(temp,
 			   HWMON_T_INPUT | HWMON_T_MIN | HWMON_T_MAX |
 			   HWMON_T_ALARM),
-- 
cgit v1.2.3-70-g09d2


From ff0805e2bde04bbb3a11669c40ecaf418e392ed1 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:08 +0200
Subject: net: phy: marvell: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230407145911.79642-5-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/marvell.c    | 2 +-
 drivers/net/phy/marvell10g.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 63a3644d86c9..24853e9a889e 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -2735,7 +2735,7 @@ static const struct hwmon_channel_info marvell_hwmon_temp = {
 	.config = marvell_hwmon_temp_config,
 };
 
-static const struct hwmon_channel_info *marvell_hwmon_info[] = {
+static const struct hwmon_channel_info * const marvell_hwmon_info[] = {
 	&marvell_hwmon_chip,
 	&marvell_hwmon_temp,
 	NULL
diff --git a/drivers/net/phy/marvell10g.c b/drivers/net/phy/marvell10g.c
index 383a9c9f36e5..55d9d7acc32e 100644
--- a/drivers/net/phy/marvell10g.c
+++ b/drivers/net/phy/marvell10g.c
@@ -243,7 +243,7 @@ static const struct hwmon_channel_info mv3310_hwmon_temp = {
 	.config = mv3310_hwmon_temp_config,
 };
 
-static const struct hwmon_channel_info *mv3310_hwmon_info[] = {
+static const struct hwmon_channel_info * const mv3310_hwmon_info[] = {
 	&mv3310_hwmon_chip,
 	&mv3310_hwmon_temp,
 	NULL,
-- 
cgit v1.2.3-70-g09d2


From 2ed84c0c6f75434091552aba91c6d53e48b95ecb Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:09 +0200
Subject: net: phy: mxl: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230407145911.79642-6-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/mxl-gpy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/mxl-gpy.c b/drivers/net/phy/mxl-gpy.c
index 8e6bb97b5f85..6301a9abfb95 100644
--- a/drivers/net/phy/mxl-gpy.c
+++ b/drivers/net/phy/mxl-gpy.c
@@ -182,7 +182,7 @@ static umode_t gpy_hwmon_is_visible(const void *data,
 	return 0444;
 }
 
-static const struct hwmon_channel_info *gpy_hwmon_info[] = {
+static const struct hwmon_channel_info * const gpy_hwmon_info[] = {
 	HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT),
 	NULL
 };
-- 
cgit v1.2.3-70-g09d2


From 0e76f1dcf487be6de4482e31076cb71e19c40c3d Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:10 +0200
Subject: net: phy: nxp-tja11xx: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230407145911.79642-7-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/nxp-tja11xx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/nxp-tja11xx.c b/drivers/net/phy/nxp-tja11xx.c
index ec91e671f8aa..b13e15310feb 100644
--- a/drivers/net/phy/nxp-tja11xx.c
+++ b/drivers/net/phy/nxp-tja11xx.c
@@ -477,7 +477,7 @@ static umode_t tja11xx_hwmon_is_visible(const void *data,
 	return 0;
 }
 
-static const struct hwmon_channel_info *tja11xx_hwmon_info[] = {
+static const struct hwmon_channel_info * const tja11xx_hwmon_info[] = {
 	HWMON_CHANNEL_INFO(in, HWMON_I_LCRIT_ALARM),
 	HWMON_CHANNEL_INFO(temp, HWMON_T_CRIT_ALARM),
 	NULL
-- 
cgit v1.2.3-70-g09d2


From 490fde262f17c5e4db27ee871cef970846e84599 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Fri, 7 Apr 2023 16:59:11 +0200
Subject: net: phy: sfp: constify pointers to hwmon_channel_info

Statically allocated array of pointed to hwmon_channel_info can be made
const for safety.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230407145911.79642-8-krzysztof.kozlowski@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/sfp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 5e515165ceab..cc9d29b71098 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -1382,7 +1382,7 @@ static const struct hwmon_ops sfp_hwmon_ops = {
 	.read_string = sfp_hwmon_read_string,
 };
 
-static const struct hwmon_channel_info *sfp_hwmon_info[] = {
+static const struct hwmon_channel_info * const sfp_hwmon_info[] = {
 	HWMON_CHANNEL_INFO(chip,
 			   HWMON_C_REGISTER_TZ),
 	HWMON_CHANNEL_INFO(in,
-- 
cgit v1.2.3-70-g09d2


From c4d3b488a90be95f4f9413dc7eae5fc113d15fe9 Mon Sep 17 00:00:00 2001
From: Manu Bretelle <chantr4@gmail.com>
Date: Fri, 7 Apr 2023 19:29:19 -0700
Subject: selftests/bpf: Reset err when symbol name already exist in
 kprobe_multi_test

When trying to add a name to the hashmap, an error code of EEXIST is
returned and we continue as names are possibly duplicated in the sys
file.

If the last name in the file is a duplicate, we will continue to the
next iteration of the while loop, and exit the loop with a value of err
set to EEXIST and enter the error label with err set, which causes the
test to fail when it should not.

This change reset err to 0 before continue-ing into the next iteration,
this way, if there is no more data to read from the file we iterate
through, err will be set to 0.

Behaviour prior to this change:
```
test_kprobe_multi_bench_attach:FAIL:get_syms unexpected error: -17
(errno 2)

All error logs:
test_kprobe_multi_bench_attach:FAIL:get_syms unexpected error: -17
(errno 2)
Summary: 0/1 PASSED, 0 SKIPPED, 1 FAILED
```

After this change:
```
Summary: 1/2 PASSED, 0 SKIPPED, 0 FAILED
```

Signed-off-by: Manu Bretelle <chantr4@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230408022919.54601-1-chantr4@gmail.com
---
 tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c b/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c
index 22be0a9a5a0a..2173c4bb555e 100644
--- a/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/kprobe_multi_test.c
@@ -381,8 +381,10 @@ static int get_syms(char ***symsp, size_t *cntp, bool kernel)
 			continue;
 
 		err = hashmap__add(map, name, 0);
-		if (err == -EEXIST)
+		if (err == -EEXIST) {
+			err = 0;
 			continue;
+		}
 		if (err)
 			goto error;
 
-- 
cgit v1.2.3-70-g09d2


From d2f5c68e3f7157e874a759e382a5eaffa775b869 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:30 -0700
Subject: docs: net: reformat driver.rst from a list to sections

driver.rst had a historical form of list of common problems.
In the age os Sphinx and rendered documentation it's better
to use the more usual title + text format.

This will allow us to render kdoc into the output more naturally.

No changes to the actual text.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/driver.rst | 91 +++++++++++++++++++++++--------------
 1 file changed, 56 insertions(+), 35 deletions(-)

diff --git a/Documentation/networking/driver.rst b/Documentation/networking/driver.rst
index 64f7236ff10b..3040a74d421c 100644
--- a/Documentation/networking/driver.rst
+++ b/Documentation/networking/driver.rst
@@ -4,15 +4,19 @@
 Softnet Driver Issues
 =====================
 
-Transmit path guidelines:
+Transmit path guidelines
+========================
 
-1) The ndo_start_xmit method must not return NETDEV_TX_BUSY under
-   any normal circumstances.  It is considered a hard error unless
-   there is no way your device can tell ahead of time when its
-   transmit function will become busy.
+Stop queues in advance
+----------------------
 
-   Instead it must maintain the queue properly.  For example,
-   for a driver implementing scatter-gather this means::
+The ndo_start_xmit method must not return NETDEV_TX_BUSY under
+any normal circumstances.  It is considered a hard error unless
+there is no way your device can tell ahead of time when its
+transmit function will become busy.
+
+Instead it must maintain the queue properly.  For example,
+for a driver implementing scatter-gather this means::
 
 	static netdev_tx_t drv_hard_start_xmit(struct sk_buff *skb,
 					       struct net_device *dev)
@@ -42,56 +46,73 @@ Transmit path guidelines:
 		return NETDEV_TX_OK;
 	}
 
-   And then at the end of your TX reclamation event handling::
+And then at the end of your TX reclamation event handling::
 
 	if (netif_queue_stopped(dp->dev) &&
 	    TX_BUFFS_AVAIL(dp) > (MAX_SKB_FRAGS + 1))
 		netif_wake_queue(dp->dev);
 
-   For a non-scatter-gather supporting card, the three tests simply become::
+For a non-scatter-gather supporting card, the three tests simply become::
 
 		/* This is a hard error log it. */
 		if (TX_BUFFS_AVAIL(dp) <= 0)
 
-   and::
+and::
 
 		if (TX_BUFFS_AVAIL(dp) == 0)
 
-   and::
+and::
 
 	if (netif_queue_stopped(dp->dev) &&
 	    TX_BUFFS_AVAIL(dp) > 0)
 		netif_wake_queue(dp->dev);
 
-2) An ndo_start_xmit method must not modify the shared parts of a
-   cloned SKB.
+No exclusive ownership
+----------------------
+
+An ndo_start_xmit method must not modify the shared parts of a
+cloned SKB.
+
+Timely completions
+------------------
+
+Do not forget that once you return NETDEV_TX_OK from your
+ndo_start_xmit method, it is your driver's responsibility to free
+up the SKB and in some finite amount of time.
 
-3) Do not forget that once you return NETDEV_TX_OK from your
-   ndo_start_xmit method, it is your driver's responsibility to free
-   up the SKB and in some finite amount of time.
+For example, this means that it is not allowed for your TX
+mitigation scheme to let TX packets "hang out" in the TX
+ring unreclaimed forever if no new TX packets are sent.
+This error can deadlock sockets waiting for send buffer room
+to be freed up.
 
-   For example, this means that it is not allowed for your TX
-   mitigation scheme to let TX packets "hang out" in the TX
-   ring unreclaimed forever if no new TX packets are sent.
-   This error can deadlock sockets waiting for send buffer room
-   to be freed up.
+If you return NETDEV_TX_BUSY from the ndo_start_xmit method, you
+must not keep any reference to that SKB and you must not attempt
+to free it up.
 
-   If you return NETDEV_TX_BUSY from the ndo_start_xmit method, you
-   must not keep any reference to that SKB and you must not attempt
-   to free it up.
+Probing guidelines
+==================
 
-Probing guidelines:
+Address validation
+------------------
+
+Any hardware layer address you obtain for your device should
+be verified.  For example, for ethernet check it with
+linux/etherdevice.h:is_valid_ether_addr()
+
+Close/stop guidelines
+=====================
 
-1) Any hardware layer address you obtain for your device should
-   be verified.  For example, for ethernet check it with
-   linux/etherdevice.h:is_valid_ether_addr()
+Quiescence
+----------
 
-Close/stop guidelines:
+After the ndo_stop routine has been called, the hardware must
+not receive or transmit any data.  All in flight packets must
+be aborted. If necessary, poll or wait for completion of
+any reset commands.
 
-1) After the ndo_stop routine has been called, the hardware must
-   not receive or transmit any data.  All in flight packets must
-   be aborted. If necessary, poll or wait for completion of
-   any reset commands.
+Auto-close
+----------
 
-2) The ndo_stop routine will be called by unregister_netdevice
-   if device is still UP.
+The ndo_stop routine will be called by unregister_netdevice
+if device is still UP.
-- 
cgit v1.2.3-70-g09d2


From da4f0f82ee9d5a128b48c959e8db7f41b59848c8 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:31 -0700
Subject: docs: net: move the probe and open/close sections of driver.rst up

Somehow it feels more right to start from the probe then open,
then tx... Much like the lifetime of the driver itself.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/driver.rst | 54 ++++++++++++++++++-------------------
 1 file changed, 27 insertions(+), 27 deletions(-)

diff --git a/Documentation/networking/driver.rst b/Documentation/networking/driver.rst
index 3040a74d421c..bfbd66871bb3 100644
--- a/Documentation/networking/driver.rst
+++ b/Documentation/networking/driver.rst
@@ -4,6 +4,33 @@
 Softnet Driver Issues
 =====================
 
+Probing guidelines
+==================
+
+Address validation
+------------------
+
+Any hardware layer address you obtain for your device should
+be verified.  For example, for ethernet check it with
+linux/etherdevice.h:is_valid_ether_addr()
+
+Close/stop guidelines
+=====================
+
+Quiescence
+----------
+
+After the ndo_stop routine has been called, the hardware must
+not receive or transmit any data.  All in flight packets must
+be aborted. If necessary, poll or wait for completion of
+any reset commands.
+
+Auto-close
+----------
+
+The ndo_stop routine will be called by unregister_netdevice
+if device is still UP.
+
 Transmit path guidelines
 ========================
 
@@ -89,30 +116,3 @@ to be freed up.
 If you return NETDEV_TX_BUSY from the ndo_start_xmit method, you
 must not keep any reference to that SKB and you must not attempt
 to free it up.
-
-Probing guidelines
-==================
-
-Address validation
-------------------
-
-Any hardware layer address you obtain for your device should
-be verified.  For example, for ethernet check it with
-linux/etherdevice.h:is_valid_ether_addr()
-
-Close/stop guidelines
-=====================
-
-Quiescence
-----------
-
-After the ndo_stop routine has been called, the hardware must
-not receive or transmit any data.  All in flight packets must
-be aborted. If necessary, poll or wait for completion of
-any reset commands.
-
-Auto-close
-----------
-
-The ndo_stop routine will be called by unregister_netdevice
-if device is still UP.
-- 
cgit v1.2.3-70-g09d2


From 8336462539ae751cbf7d59822eb331602a69df99 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:32 -0700
Subject: docs: net: use C syntax highlight in driver.rst

Use syntax highlight, comment out the "..." since they are
not valid C.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/driver.rst | 30 ++++++++++++++++++++----------
 1 file changed, 20 insertions(+), 10 deletions(-)

diff --git a/Documentation/networking/driver.rst b/Documentation/networking/driver.rst
index bfbd66871bb3..19c363291d04 100644
--- a/Documentation/networking/driver.rst
+++ b/Documentation/networking/driver.rst
@@ -43,7 +43,9 @@ there is no way your device can tell ahead of time when its
 transmit function will become busy.
 
 Instead it must maintain the queue properly.  For example,
-for a driver implementing scatter-gather this means::
+for a driver implementing scatter-gather this means:
+
+.. code-block:: c
 
 	static netdev_tx_t drv_hard_start_xmit(struct sk_buff *skb,
 					       struct net_device *dev)
@@ -51,7 +53,7 @@ for a driver implementing scatter-gather this means::
 		struct drv *dp = netdev_priv(dev);
 
 		lock_tx(dp);
-		...
+		//...
 		/* This is a hard error log it. */
 		if (TX_BUFFS_AVAIL(dp) <= (skb_shinfo(skb)->nr_frags + 1)) {
 			netif_stop_queue(dev);
@@ -61,34 +63,42 @@ for a driver implementing scatter-gather this means::
 			return NETDEV_TX_BUSY;
 		}
 
-		... queue packet to card ...
-		... update tx consumer index ...
+		//... queue packet to card ...
+		//... update tx consumer index ...
 
 		if (TX_BUFFS_AVAIL(dp) <= (MAX_SKB_FRAGS + 1))
 			netif_stop_queue(dev);
 
-		...
+		//...
 		unlock_tx(dp);
-		...
+		//...
 		return NETDEV_TX_OK;
 	}
 
-And then at the end of your TX reclamation event handling::
+And then at the end of your TX reclamation event handling:
+
+.. code-block:: c
 
 	if (netif_queue_stopped(dp->dev) &&
 	    TX_BUFFS_AVAIL(dp) > (MAX_SKB_FRAGS + 1))
 		netif_wake_queue(dp->dev);
 
-For a non-scatter-gather supporting card, the three tests simply become::
+For a non-scatter-gather supporting card, the three tests simply become:
+
+.. code-block:: c
 
 		/* This is a hard error log it. */
 		if (TX_BUFFS_AVAIL(dp) <= 0)
 
-and::
+and:
+
+.. code-block:: c
 
 		if (TX_BUFFS_AVAIL(dp) == 0)
 
-and::
+and:
+
+.. code-block:: c
 
 	if (netif_queue_stopped(dp->dev) &&
 	    TX_BUFFS_AVAIL(dp) > 0)
-- 
cgit v1.2.3-70-g09d2


From c91c46de6bbc1147ae5dfe046b87f5f3d6593215 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:33 -0700
Subject: net: provide macros for commonly copied lockless queue stop/wake code

A lot of drivers follow the same scheme to stop / start queues
without introducing locks between xmit and NAPI tx completions.
I'm guessing they all copy'n'paste each other's code.
The original code dates back all the way to e1000 and Linux 2.6.19.

Smaller drivers shy away from the scheme and introduce a lock
which may cause deadlocks in netpoll.

Provide macros which encapsulate the necessary logic.

The macros do not prevent false wake ups, the extra barrier
required to close that race is not worth it. See discussion in:
https://lore.kernel.org/all/c39312a2-4537-14b4-270c-9fe1fbb91e89@gmail.com/

Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/driver.rst |   6 ++
 include/linux/netdevice.h           |   1 +
 include/net/netdev_queues.h         | 144 ++++++++++++++++++++++++++++++++++++
 3 files changed, 151 insertions(+)
 create mode 100644 include/net/netdev_queues.h

diff --git a/Documentation/networking/driver.rst b/Documentation/networking/driver.rst
index 19c363291d04..4071f2c00f8b 100644
--- a/Documentation/networking/driver.rst
+++ b/Documentation/networking/driver.rst
@@ -104,6 +104,12 @@ and:
 	    TX_BUFFS_AVAIL(dp) > 0)
 		netif_wake_queue(dp->dev);
 
+Lockless queue stop / wake helper macros
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. kernel-doc:: include/net/netdev_queues.h
+   :doc: Lockless queue stopping / waking helpers.
+
 No exclusive ownership
 ----------------------
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 1c25b39681b3..7bec9a2be8ef 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3335,6 +3335,7 @@ static inline void netif_tx_wake_all_queues(struct net_device *dev)
 
 static __always_inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
 {
+	/* Must be an atomic op see netif_txq_try_stop() */
 	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
 }
 
diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
new file mode 100644
index 000000000000..5236d78bbdeb
--- /dev/null
+++ b/include/net/netdev_queues.h
@@ -0,0 +1,144 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_NET_QUEUES_H
+#define _LINUX_NET_QUEUES_H
+
+#include <linux/netdevice.h>
+
+/**
+ * DOC: Lockless queue stopping / waking helpers.
+ *
+ * The netif_txq_maybe_stop() and __netif_txq_completed_wake()
+ * macros are designed to safely implement stopping
+ * and waking netdev queues without full lock protection.
+ *
+ * We assume that there can be no concurrent stop attempts and no concurrent
+ * wake attempts. The try-stop should happen from the xmit handler,
+ * while wake up should be triggered from NAPI poll context.
+ * The two may run concurrently (single producer, single consumer).
+ *
+ * The try-stop side is expected to run from the xmit handler and therefore
+ * it does not reschedule Tx (netif_tx_start_queue() instead of
+ * netif_tx_wake_queue()). Uses of the ``stop`` macros outside of the xmit
+ * handler may lead to xmit queue being enabled but not run.
+ * The waking side does not have similar context restrictions.
+ *
+ * The macros guarantee that rings will not remain stopped if there's
+ * space available, but they do *not* prevent false wake ups when
+ * the ring is full! Drivers should check for ring full at the start
+ * for the xmit handler.
+ *
+ * All descriptor ring indexes (and other relevant shared state) must
+ * be updated before invoking the macros.
+ */
+
+#define netif_txq_try_stop(txq, get_desc, start_thrs)			\
+	({								\
+		int _res;						\
+									\
+		netif_tx_stop_queue(txq);				\
+		/* Producer index and stop bit must be visible		\
+		 * to consumer before we recheck.			\
+		 * Pairs with a barrier in __netif_txq_maybe_wake().	\
+		 */							\
+		smp_mb__after_atomic();					\
+									\
+		/* We need to check again in a case another		\
+		 * CPU has just made room available.			\
+		 */							\
+		_res = 0;						\
+		if (unlikely(get_desc >= start_thrs)) {			\
+			netif_tx_start_queue(txq);			\
+			_res = -1;					\
+		}							\
+		_res;							\
+	})								\
+
+/**
+ * netif_txq_maybe_stop() - locklessly stop a Tx queue, if needed
+ * @txq:	struct netdev_queue to stop/start
+ * @get_desc:	get current number of free descriptors (see requirements below!)
+ * @stop_thrs:	minimal number of available descriptors for queue to be left
+ *		enabled
+ * @start_thrs:	minimal number of descriptors to re-enable the queue, can be
+ *		equal to @stop_thrs or higher to avoid frequent waking
+ *
+ * All arguments may be evaluated multiple times, beware of side effects.
+ * @get_desc must be a formula or a function call, it must always
+ * return up-to-date information when evaluated!
+ * Expected to be used from ndo_start_xmit, see the comment on top of the file.
+ *
+ * Returns:
+ *	 0 if the queue was stopped
+ *	 1 if the queue was left enabled
+ *	-1 if the queue was re-enabled (raced with waking)
+ */
+#define netif_txq_maybe_stop(txq, get_desc, stop_thrs, start_thrs)	\
+	({								\
+		int _res;						\
+									\
+		_res = 1;						\
+		if (unlikely(get_desc < stop_thrs))			\
+			_res = netif_txq_try_stop(txq, get_desc, start_thrs); \
+		_res;							\
+	})								\
+
+
+/**
+ * __netif_txq_maybe_wake() - locklessly wake a Tx queue, if needed
+ * @txq:	struct netdev_queue to stop/start
+ * @get_desc:	get current number of free descriptors (see requirements below!)
+ * @start_thrs:	minimal number of descriptors to re-enable the queue
+ * @down_cond:	down condition, predicate indicating that the queue should
+ *		not be woken up even if descriptors are available
+ *
+ * All arguments may be evaluated multiple times.
+ * @get_desc must be a formula or a function call, it must always
+ * return up-to-date information when evaluated!
+ *
+ * Returns:
+ *	 0 if the queue was woken up
+ *	 1 if the queue was already enabled (or disabled but @down_cond is true)
+ *	-1 if the queue was left unchanged (@start_thrs not reached)
+ */
+#define __netif_txq_maybe_wake(txq, get_desc, start_thrs, down_cond)	\
+	({								\
+		int _res;						\
+									\
+		_res = -1;						\
+		if (likely(get_desc > start_thrs)) {			\
+			/* Make sure that anybody stopping the queue after \
+			 * this sees the new next_to_clean.		\
+			 */						\
+			smp_mb();					\
+			_res = 1;					\
+			if (unlikely(netif_tx_queue_stopped(txq)) &&	\
+			    !(down_cond)) {				\
+				netif_tx_wake_queue(txq);		\
+				_res = 0;				\
+			}						\
+		}							\
+		_res;							\
+	})
+
+#define netif_txq_maybe_wake(txq, get_desc, start_thrs)		\
+	__netif_txq_maybe_wake(txq, get_desc, start_thrs, false)
+
+/* subqueue variants follow */
+
+#define netif_subqueue_try_stop(dev, idx, get_desc, start_thrs)		\
+	({								\
+		struct netdev_queue *txq;				\
+									\
+		txq = netdev_get_tx_queue(dev, idx);			\
+		netif_txq_try_stop(txq, get_desc, start_thrs);		\
+	})
+
+#define netif_subqueue_maybe_stop(dev, idx, get_desc, stop_thrs, start_thrs) \
+	({								\
+		struct netdev_queue *txq;				\
+									\
+		txq = netdev_get_tx_queue(dev, idx);			\
+		netif_txq_maybe_stop(txq, get_desc, stop_thrs, start_thrs); \
+	})
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From 9ded5bc77fe5f396a58da95627695a89a427c962 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:34 -0700
Subject: ixgbe: use new queue try_stop/try_wake macros

Convert ixgbe to use the new macros, I think a lot of people
copy the ixgbe code. The only functional change is that the
unlikely() in ixgbe_clean_tx_irq() turns into a likely()
inside the new macro and no longer includes

  total_packets && netif_carrier_ok(tx_ring->netdev)

which is probably for the best, anyway.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 38 +++++++--------------------
 1 file changed, 10 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 773c35fecace..cbbddee55db1 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -36,6 +36,7 @@
 #include <net/tc_act/tc_mirred.h>
 #include <net/vxlan.h>
 #include <net/mpls.h>
+#include <net/netdev_queues.h>
 #include <net/xdp_sock_drv.h>
 #include <net/xfrm.h>
 
@@ -1119,6 +1120,7 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 	unsigned int total_bytes = 0, total_packets = 0, total_ipsec = 0;
 	unsigned int budget = q_vector->tx.work_limit;
 	unsigned int i = tx_ring->next_to_clean;
+	struct netdev_queue *txq;
 
 	if (test_bit(__IXGBE_DOWN, &adapter->state))
 		return true;
@@ -1253,20 +1255,12 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 				  total_packets, total_bytes);
 
 #define TX_WAKE_THRESHOLD (DESC_NEEDED * 2)
-	if (unlikely(total_packets && netif_carrier_ok(tx_ring->netdev) &&
-		     (ixgbe_desc_unused(tx_ring) >= TX_WAKE_THRESHOLD))) {
-		/* Make sure that anybody stopping the queue after this
-		 * sees the new next_to_clean.
-		 */
-		smp_mb();
-		if (__netif_subqueue_stopped(tx_ring->netdev,
-					     tx_ring->queue_index)
-		    && !test_bit(__IXGBE_DOWN, &adapter->state)) {
-			netif_wake_subqueue(tx_ring->netdev,
-					    tx_ring->queue_index);
-			++tx_ring->tx_stats.restart_queue;
-		}
-	}
+	txq = netdev_get_tx_queue(tx_ring->netdev, tx_ring->queue_index);
+	if (total_packets && netif_carrier_ok(tx_ring->netdev) &&
+	    !__netif_txq_maybe_wake(txq, ixgbe_desc_unused(tx_ring),
+				    TX_WAKE_THRESHOLD,
+				    test_bit(__IXGBE_DOWN, &adapter->state)))
+		++tx_ring->tx_stats.restart_queue;
 
 	return !!budget;
 }
@@ -8270,22 +8264,10 @@ static void ixgbe_tx_olinfo_status(union ixgbe_adv_tx_desc *tx_desc,
 
 static int __ixgbe_maybe_stop_tx(struct ixgbe_ring *tx_ring, u16 size)
 {
-	netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index);
-
-	/* Herbert's original patch had:
-	 *  smp_mb__after_netif_stop_queue();
-	 * but since that doesn't exist yet, just open code it.
-	 */
-	smp_mb();
-
-	/* We need to check again in a case another CPU has just
-	 * made room available.
-	 */
-	if (likely(ixgbe_desc_unused(tx_ring) < size))
+	if (!netif_subqueue_try_stop(tx_ring->netdev, tx_ring->queue_index,
+				     ixgbe_desc_unused(tx_ring), size))
 		return -EBUSY;
 
-	/* A reprieve! - use start_queue because it doesn't call schedule */
-	netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index);
 	++tx_ring->tx_stats.restart_queue;
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 08a096780d9239e69909c48f4b1fcd99c860b2ef Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:35 -0700
Subject: bnxt: use new queue try_stop/try_wake macros

Convert bnxt to use new macros rather than open code the logic.
Two differences:
(1) bnxt_tx_int() will now only issue a memory barrier if it sees
    enough space on the ring to wake the queue. This should be fine,
    the mb() is between the writes to the ring pointers and checking
    queue state.
(2) we'll start the queue instead of waking on race, this should
    be safe inside the xmit handler.

Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 40 ++++++-------------------------
 1 file changed, 7 insertions(+), 33 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 8ff5a4f98d6f..de97bee25249 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -56,6 +56,7 @@
 #include <linux/hwmon-sysfs.h>
 #include <net/page_pool.h>
 #include <linux/align.h>
+#include <net/netdev_queues.h>
 
 #include "bnxt_hsi.h"
 #include "bnxt.h"
@@ -331,26 +332,6 @@ static void bnxt_txr_db_kick(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 	txr->kick_pending = 0;
 }
 
-static bool bnxt_txr_netif_try_stop_queue(struct bnxt *bp,
-					  struct bnxt_tx_ring_info *txr,
-					  struct netdev_queue *txq)
-{
-	netif_tx_stop_queue(txq);
-
-	/* netif_tx_stop_queue() must be done before checking
-	 * tx index in bnxt_tx_avail() below, because in
-	 * bnxt_tx_int(), we update tx index before checking for
-	 * netif_tx_queue_stopped().
-	 */
-	smp_mb();
-	if (bnxt_tx_avail(bp, txr) >= bp->tx_wake_thresh) {
-		netif_tx_wake_queue(txq);
-		return false;
-	}
-
-	return true;
-}
-
 static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct bnxt *bp = netdev_priv(dev);
@@ -384,7 +365,8 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		if (net_ratelimit() && txr->kick_pending)
 			netif_warn(bp, tx_err, dev,
 				   "bnxt: ring busy w/ flush pending!\n");
-		if (bnxt_txr_netif_try_stop_queue(bp, txr, txq))
+		if (!netif_txq_try_stop(txq, bnxt_tx_avail(bp, txr),
+					bp->tx_wake_thresh))
 			return NETDEV_TX_BUSY;
 	}
 
@@ -614,7 +596,8 @@ tx_done:
 		if (netdev_xmit_more() && !tx_buf->is_push)
 			bnxt_txr_db_kick(bp, txr, prod);
 
-		bnxt_txr_netif_try_stop_queue(bp, txr, txq);
+		netif_txq_try_stop(txq, bnxt_tx_avail(bp, txr),
+				   bp->tx_wake_thresh);
 	}
 	return NETDEV_TX_OK;
 
@@ -708,17 +691,8 @@ next_tx_int:
 	netdev_tx_completed_queue(txq, nr_pkts, tx_bytes);
 	txr->tx_cons = cons;
 
-	/* Need to make the tx_cons update visible to bnxt_start_xmit()
-	 * before checking for netif_tx_queue_stopped().  Without the
-	 * memory barrier, there is a small possibility that bnxt_start_xmit()
-	 * will miss it and cause the queue to be stopped forever.
-	 */
-	smp_mb();
-
-	if (unlikely(netif_tx_queue_stopped(txq)) &&
-	    bnxt_tx_avail(bp, txr) >= bp->tx_wake_thresh &&
-	    READ_ONCE(txr->dev_state) != BNXT_DEV_STATE_CLOSING)
-		netif_tx_wake_queue(txq);
+	__netif_txq_maybe_wake(txq, bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
+			       READ_ONCE(txr->dev_state) != BNXT_DEV_STATE_CLOSING);
 }
 
 static struct page *__bnxt_alloc_rx_page(struct bnxt *bp, dma_addr_t *mapping,
-- 
cgit v1.2.3-70-g09d2


From 301f227fc860624d37ba5dae9da57dcf371268db Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Apr 2023 18:25:36 -0700
Subject: net: piggy back on the memory barrier in bql when waking queues

Drivers call netdev_tx_completed_queue() right before
netif_txq_maybe_wake(). If BQL is enabled netdev_tx_completed_queue()
should issue a memory barrier, so we can depend on that separating
the stop check from the consumer index update, instead of adding
another barrier in netif_txq_maybe_wake().

This matters more than the barriers on the xmit path, because
the wake condition is almost always true. So we issue the
consumer side barrier often.

Wrap netdev_tx_completed_queue() in a local helper to issue
the barrier even if BQL is disabled. Keep the same semantics
as netdev_tx_completed_queue() (barrier only if bytes != 0)
to make it clear that the barrier is conditional.

Plus since macro gets pkt/byte counts as arguments now -
we can skip waking if there were no packets completed.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     |  6 ++---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 12 ++++-----
 include/linux/netdevice.h                     |  2 +-
 include/net/netdev_queues.h                   | 39 ++++++++++++++++++++-------
 4 files changed, 38 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index de97bee25249..f7602d8d79e3 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -688,11 +688,11 @@ next_tx_int:
 		dev_kfree_skb_any(skb);
 	}
 
-	netdev_tx_completed_queue(txq, nr_pkts, tx_bytes);
 	txr->tx_cons = cons;
 
-	__netif_txq_maybe_wake(txq, bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
-			       READ_ONCE(txr->dev_state) != BNXT_DEV_STATE_CLOSING);
+	__netif_txq_completed_wake(txq, nr_pkts, tx_bytes,
+				   bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
+				   READ_ONCE(txr->dev_state) != BNXT_DEV_STATE_CLOSING);
 }
 
 static struct page *__bnxt_alloc_rx_page(struct bnxt *bp, dma_addr_t *mapping,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index cbbddee55db1..f2604fc05991 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1251,15 +1251,13 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 	if (ring_is_xdp(tx_ring))
 		return !!budget;
 
-	netdev_tx_completed_queue(txring_txq(tx_ring),
-				  total_packets, total_bytes);
-
 #define TX_WAKE_THRESHOLD (DESC_NEEDED * 2)
 	txq = netdev_get_tx_queue(tx_ring->netdev, tx_ring->queue_index);
-	if (total_packets && netif_carrier_ok(tx_ring->netdev) &&
-	    !__netif_txq_maybe_wake(txq, ixgbe_desc_unused(tx_ring),
-				    TX_WAKE_THRESHOLD,
-				    test_bit(__IXGBE_DOWN, &adapter->state)))
+	if (!__netif_txq_completed_wake(txq, total_packets, total_bytes,
+					ixgbe_desc_unused(tx_ring),
+					TX_WAKE_THRESHOLD,
+					netif_carrier_ok(tx_ring->netdev) &&
+					test_bit(__IXGBE_DOWN, &adapter->state)))
 		++tx_ring->tx_stats.restart_queue;
 
 	return !!budget;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 7bec9a2be8ef..fe355592dfde 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3532,7 +3532,7 @@ static inline void netdev_tx_completed_queue(struct netdev_queue *dev_queue,
 	 * netdev_tx_sent_queue will miss the update and cause the queue to
 	 * be stopped forever
 	 */
-	smp_mb();
+	smp_mb(); /* NOTE: netdev_txq_completed_mb() assumes this exists */
 
 	if (unlikely(dql_avail(&dev_queue->dql) < 0))
 		return;
diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
index 5236d78bbdeb..b26fdb441e39 100644
--- a/include/net/netdev_queues.h
+++ b/include/net/netdev_queues.h
@@ -38,7 +38,7 @@
 		netif_tx_stop_queue(txq);				\
 		/* Producer index and stop bit must be visible		\
 		 * to consumer before we recheck.			\
-		 * Pairs with a barrier in __netif_txq_maybe_wake().	\
+		 * Pairs with a barrier in __netif_txq_completed_wake(). \
 		 */							\
 		smp_mb__after_atomic();					\
 									\
@@ -82,10 +82,24 @@
 		_res;							\
 	})								\
 
+/* Variant of netdev_tx_completed_queue() which guarantees smp_mb() if
+ * @bytes != 0, regardless of kernel config.
+ */
+static inline void
+netdev_txq_completed_mb(struct netdev_queue *dev_queue,
+			unsigned int pkts, unsigned int bytes)
+{
+	if (IS_ENABLED(CONFIG_BQL))
+		netdev_tx_completed_queue(dev_queue, pkts, bytes);
+	else if (bytes)
+		smp_mb();
+}
 
 /**
- * __netif_txq_maybe_wake() - locklessly wake a Tx queue, if needed
+ * __netif_txq_completed_wake() - locklessly wake a Tx queue, if needed
  * @txq:	struct netdev_queue to stop/start
+ * @pkts:	number of packets completed
+ * @bytes:	number of bytes completed
  * @get_desc:	get current number of free descriptors (see requirements below!)
  * @start_thrs:	minimal number of descriptors to re-enable the queue
  * @down_cond:	down condition, predicate indicating that the queue should
@@ -94,22 +108,27 @@
  * All arguments may be evaluated multiple times.
  * @get_desc must be a formula or a function call, it must always
  * return up-to-date information when evaluated!
+ * Reports completed pkts/bytes to BQL.
  *
  * Returns:
  *	 0 if the queue was woken up
  *	 1 if the queue was already enabled (or disabled but @down_cond is true)
  *	-1 if the queue was left unchanged (@start_thrs not reached)
  */
-#define __netif_txq_maybe_wake(txq, get_desc, start_thrs, down_cond)	\
+#define __netif_txq_completed_wake(txq, pkts, bytes,			\
+				   get_desc, start_thrs, down_cond)	\
 	({								\
 		int _res;						\
 									\
+		/* Report to BQL and piggy back on its barrier.		\
+		 * Barrier makes sure that anybody stopping the queue	\
+		 * after this point sees the new consumer index.	\
+		 * Pairs with barrier in netif_txq_try_stop().		\
+		 */							\
+		netdev_txq_completed_mb(txq, pkts, bytes);		\
+									\
 		_res = -1;						\
-		if (likely(get_desc > start_thrs)) {			\
-			/* Make sure that anybody stopping the queue after \
-			 * this sees the new next_to_clean.		\
-			 */						\
-			smp_mb();					\
+		if (pkts && likely(get_desc > start_thrs)) {		\
 			_res = 1;					\
 			if (unlikely(netif_tx_queue_stopped(txq)) &&	\
 			    !(down_cond)) {				\
@@ -120,8 +139,8 @@
 		_res;							\
 	})
 
-#define netif_txq_maybe_wake(txq, get_desc, start_thrs)		\
-	__netif_txq_maybe_wake(txq, get_desc, start_thrs, false)
+#define netif_txq_completed_wake(txq, pkts, bytes, get_desc, start_thrs) \
+	__netif_txq_completed_wake(txq, pkts, bytes, get_desc, start_thrs, false)
 
 /* subqueue variants follow */
 
-- 
cgit v1.2.3-70-g09d2


From 9bc11460bea751b5c805ac793de35a55629adb9b Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Fri, 7 Apr 2023 15:26:51 +0200
Subject: net: ethernet: mtk_eth_soc: use be32 type to store be32 values

n_addr is used to store be32 values,
so a sparse-friendly array of be32 to store these values.

Flagged by sparse:
  .../mtk_ppe_debugfs.c:59:27: warning: incorrect type in assignment (different base types)
  .../mtk_ppe_debugfs.c:59:27:    expected unsigned int
  .../mtk_ppe_debugfs.c:59:27:    got restricted __be32 [usertype]
  .../mtk_ppe_debugfs.c:161:46: warning: cast to restricted __be16

No functional changes intended.
Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230401-mtk_eth_soc-sparse-v2-1-963becba3cb7@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c b/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
index 53cf87e9acbb..316fe2e70fea 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
@@ -47,7 +47,7 @@ static const char *mtk_foe_pkt_type_str(int type)
 static void
 mtk_print_addr(struct seq_file *m, u32 *addr, bool ipv6)
 {
-	u32 n_addr[4];
+	__be32 n_addr[4];
 	int i;
 
 	if (!ipv6) {
-- 
cgit v1.2.3-70-g09d2


From 89863a3b5f0240cc89b7cbf6756e5c6c220c056e Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Fri, 7 Apr 2023 11:41:57 +0800
Subject: net: fddi: skfp: rmt: Clean up some inconsistent indenting

No functional modification involved.

drivers/net/fddi/skfp/rmt.c:236 rmt_fsm() warn: if statement not indented.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4736
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Link: https://lore.kernel.org/r/20230407034157.61276-1-jiapeng.chong@linux.alibaba.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/fddi/skfp/rmt.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/fddi/skfp/rmt.c b/drivers/net/fddi/skfp/rmt.c
index 37a89675dbeb..314623650a84 100644
--- a/drivers/net/fddi/skfp/rmt.c
+++ b/drivers/net/fddi/skfp/rmt.c
@@ -234,9 +234,9 @@ static void rmt_fsm(struct s_smc *smc, int cmd)
 		if (smc->r.rm_join) {
 			smc->r.sm_ma_avail = TRUE ;
 			if (smc->mib.m[MAC0].fddiMACMA_UnitdataEnable)
-			smc->mib.m[MAC0].fddiMACMA_UnitdataAvailable = TRUE ;
-				else
-			smc->mib.m[MAC0].fddiMACMA_UnitdataAvailable = FALSE ;
+				smc->mib.m[MAC0].fddiMACMA_UnitdataAvailable = TRUE;
+			else
+				smc->mib.m[MAC0].fddiMACMA_UnitdataAvailable = FALSE;
 		}
 		DB_RMTN(1, "RMT : RING UP");
 		RS_CLEAR(smc,RS_NORINGOP) ;
-- 
cgit v1.2.3-70-g09d2


From d8f975594da8dc59b463db9bcc0bec3fd961630b Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Mon, 10 Apr 2023 17:07:20 +0300
Subject: wifi: iwlwifi: mvm: fix the order of TIMING_MEASUREMENT notifications

Host commands and notifications are expected to be ordered in
iwl_mvm_groups array. Fix the order of two new TIMING_MEASUREMENT
notifications.

Fixes: c7eca79def44 ("wifi: iwlwifi: mvm: report hardware timestamps in RX/TX status")
Reported-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/linux-wireless/20230331175121.GA3127046@dev-arch.thelio-3990X/
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20230410140721.897683-1-gregory.greenman@intel.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 56a4e9d6ae33..6f082059025a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -472,8 +472,8 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
 	HCMD_NAME(SCAN_OFFLOAD_PROFILES_QUERY_CMD),
 	HCMD_NAME(BT_COEX_UPDATE_REDUCED_TXP),
 	HCMD_NAME(BT_COEX_CI),
-	HCMD_NAME(WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION),
 	HCMD_NAME(WNM_80211V_TIMING_MEASUREMENT_NOTIFICATION),
+	HCMD_NAME(WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION),
 	HCMD_NAME(PHY_CONFIGURATION_CMD),
 	HCMD_NAME(CALIB_RES_NOTIF_PHY_DB),
 	HCMD_NAME(PHY_DB_CMD),
-- 
cgit v1.2.3-70-g09d2


From fc3c2f0ed86b65dff4b6844c59c597b977cae533 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Mon, 10 Apr 2023 17:07:21 +0300
Subject: wifi: iwlwifi: fix duplicate entry in iwl_dev_info_table

There're two identical entries for ax1650 device in
iwl_dev_info_table. Remove one of the duplicate entries.

Fixes: 953e66a7238b ("iwlwifi: add new ax1650 killer device")
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230410140721.897683-2-gregory.greenman@intel.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 6c935d73943a..86b094683631 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -565,7 +565,6 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 	IWL_DEV_INFO(0x43F0, 0x1652, killer1650i_2ax_cfg_qu_b0_hr_b0, iwl_ax201_killer_1650i_name),
 	IWL_DEV_INFO(0x43F0, 0x2074, iwl_ax201_cfg_qu_hr, NULL),
 	IWL_DEV_INFO(0x43F0, 0x4070, iwl_ax201_cfg_qu_hr, NULL),
-	IWL_DEV_INFO(0x43F0, 0x1651, killer1650s_2ax_cfg_qu_b0_hr_b0, iwl_ax201_killer_1650s_name),
 	IWL_DEV_INFO(0xA0F0, 0x0070, iwl_ax201_cfg_qu_hr, NULL),
 	IWL_DEV_INFO(0xA0F0, 0x0074, iwl_ax201_cfg_qu_hr, NULL),
 	IWL_DEV_INFO(0xA0F0, 0x0078, iwl_ax201_cfg_qu_hr, NULL),
-- 
cgit v1.2.3-70-g09d2


From ebe3bdc4359e07b4e347af8d5324fb436302bbe1 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Fri, 7 Apr 2023 07:56:09 -0700
Subject: tools: ynl: throw a more meaningful exception if family not supported

cli.py currently throws a pure KeyError if kernel doesn't support
a netlink family. Users who did not write ynl (hah) may waste
their time investigating what's wrong with the Python code.
Improve the error message:

Traceback (most recent call last):
  File "/home/kicinski/devel/linux/tools/net/ynl/lib/ynl.py", line 362, in __init__
    self.family = GenlFamily(self.yaml['name'])
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/kicinski/devel/linux/tools/net/ynl/lib/ynl.py", line 331, in __init__
    self.genl_family = genl_family_name_to_id[family_name]
                       ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
KeyError: 'netdev'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/kicinski/devel/linux/./tools/net/ynl/cli.py", line 52, in <module>
    main()
  File "/home/kicinski/devel/linux/./tools/net/ynl/cli.py", line 31, in main
    ynl = YnlFamily(args.spec, args.schema)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/kicinski/devel/linux/tools/net/ynl/lib/ynl.py", line 364, in __init__
    raise Exception(f"Family '{self.yaml['name']}' not supported by the kernel")
Exception: Family 'netdev' not supported by the kernel

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230407145609.297525-1-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 tools/net/ynl/lib/ynl.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 7690e0b0cb3f..aa77bcae4807 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -358,7 +358,10 @@ class YnlFamily(SpecFamily):
             bound_f = functools.partial(self._op, op_name)
             setattr(self, op.ident_name, bound_f)
 
-        self.family = GenlFamily(self.yaml['name'])
+        try:
+            self.family = GenlFamily(self.yaml['name'])
+        except KeyError:
+            raise Exception(f"Family '{self.yaml['name']}' not supported by the kernel")
 
     def ntf_subscribe(self, mcast_name):
         if mcast_name not in self.family.genl_family['mcast']:
-- 
cgit v1.2.3-70-g09d2


From 2bef4d1fb8b3401d70f28e5d62380a25cefc9d3e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 4 Apr 2023 17:16:58 +0000
Subject: wifi: mac80211_hwsim: fix potential NULL deref in
 hwsim_pmsr_report_nl()

syzbot reported a NULL dereference caused by a missing check
in hwsim_pmsr_report_nl(), and bisected the issue to cited commit.

v2: test the nlattr before using nla_data() on it (Simon Horman)

general protection fault, probably for non-canonical address 0xdffffc0000000001: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f]
CPU: 0 PID: 5084 Comm: syz-executor104 Not tainted 6.3.0-rc4-next-20230331-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/02/2023
RIP: 0010:jhash+0x339/0x610 include/linux/jhash.h:95
Code: 83 fd 01 0f 84 5f ff ff ff eb de 83 fd 05 74 3a e8 ac f5 71 fd 48 8d 7b 05 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <0f> b6 04 02 48 89 fa 83 e2 07 38 d0 7f 08 84 c0 0f 85 96 02 00 00
RSP: 0018:ffffc90003abf298 EFLAGS: 00010202
RAX: dffffc0000000000 RBX: 0000000000000004 RCX: 0000000000000000
RDX: 0000000000000001 RSI: ffffffff84111ba4 RDI: 0000000000000009
RBP: 0000000000000006 R08: 0000000000000005 R09: 000000000000000c
R10: 0000000000000006 R11: 0000000000000000 R12: 000000004d2c27cd
R13: 000000002bd9e6c2 R14: 000000002bd9e6c2 R15: 000000002bd9e6c2
FS: 0000555556847300(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000045ad50 CR3: 0000000078aa6000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
rht_key_hashfn include/linux/rhashtable.h:159 [inline]
__rhashtable_lookup include/linux/rhashtable.h:604 [inline]
rhashtable_lookup include/linux/rhashtable.h:646 [inline]
rhashtable_lookup_fast include/linux/rhashtable.h:672 [inline]
get_hwsim_data_ref_from_addr+0xb9/0x600 drivers/net/wireless/virtual/mac80211_hwsim.c:757
hwsim_pmsr_report_nl+0xe7/0xd50 drivers/net/wireless/virtual/mac80211_hwsim.c:3764
genl_family_rcv_msg_doit.isra.0+0x1e6/0x2d0 net/netlink/genetlink.c:968
genl_family_rcv_msg net/netlink/genetlink.c:1048 [inline]
genl_rcv_msg+0x4ff/0x7e0 net/netlink/genetlink.c:1065
netlink_rcv_skb+0x165/0x440 net/netlink/af_netlink.c:2572
genl_rcv+0x28/0x40 net/netlink/genetlink.c:1076
netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline]
netlink_unicast+0x547/0x7f0 net/netlink/af_netlink.c:1365
netlink_sendmsg+0x925/0xe30 net/netlink/af_netlink.c:1942
sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg+0xde/0x190 net/socket.c:747
____sys_sendmsg+0x71c/0x900 net/socket.c:2501
___sys_sendmsg+0x110/0x1b0 net/socket.c:2555
__sys_sendmsg+0xf7/0x1c0 net/socket.c:2584
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

Fixes: 2af3b2a631b1 ("mac80211_hwsim: add PMSR report support via virtio")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jaewan Kim <jaewan@google.com>
Cc: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Jaewan Kim <jaewan@google.com>
Link: https://lore.kernel.org/r/20230404171658.917361-1-edumazet@google.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index f446d8f6e1f6..2211fa58fe41 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -3760,6 +3760,9 @@ static int hwsim_pmsr_report_nl(struct sk_buff *msg, struct genl_info *info)
 	int err;
 	int rem;
 
+	if (!info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER])
+		return -EINVAL;
+
 	src = nla_data(info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER]);
 	data = get_hwsim_data_ref_from_addr(src);
 	if (!data)
-- 
cgit v1.2.3-70-g09d2


From e3e0ca32cf478e78c579b02cd9c1657d93c97add Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Mon, 10 Apr 2023 13:03:32 -0700
Subject: wifi: mac80211: set EHT support flag in AP mode

Set 'eht_support' flag if EHT capabilities are present.

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Link: https://lore.kernel.org/r/20230410200332.32265-1-quic_alokad@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 473915606715..7317e4a5d1ff 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1342,6 +1342,10 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev,
 	}
 
 	if (params->eht_cap) {
+		if (!link_conf->he_support)
+			return -EOPNOTSUPP;
+
+		link_conf->eht_support = true;
 		link_conf->eht_puncturing = params->punct_bitmap;
 		changed |= BSS_CHANGED_EHT_PUNCTURING;
 
-- 
cgit v1.2.3-70-g09d2


From 4de00f0acc722f43046dca06fe1336597d1250ab Mon Sep 17 00:00:00 2001
From: Shailend Chand <shailend@google.com>
Date: Fri, 7 Apr 2023 11:48:30 -0700
Subject: gve: Unify duplicate GQ min pkt desc size constants

The two constants accomplish the same thing.

Signed-off-by: Shailend Chand <shailend@google.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230407184830.309398-1-shailend@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/google/gve/gve.h    | 2 --
 drivers/net/ethernet/google/gve/gve_tx.c | 4 ++--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index 4d01c66d2d65..98eb78d98e9f 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -49,8 +49,6 @@
 
 #define GVE_XDP_ACTIONS 5
 
-#define GVE_TX_MAX_HEADER_SIZE 182
-
 #define GVE_GQ_TX_MIN_PKT_DESC_BYTES 182
 
 /* Each slot in the desc ring has a 1:1 mapping to a slot in the data ring */
diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c
index fe6b933e1df1..813da572abca 100644
--- a/drivers/net/ethernet/google/gve/gve_tx.c
+++ b/drivers/net/ethernet/google/gve/gve_tx.c
@@ -730,7 +730,7 @@ static int gve_tx_fill_xdp(struct gve_priv *priv, struct gve_tx_ring *tx,
 	u32 reqi = tx->req;
 
 	pad = gve_tx_fifo_pad_alloc_one_frag(&tx->tx_fifo, len);
-	if (pad >= GVE_TX_MAX_HEADER_SIZE)
+	if (pad >= GVE_GQ_TX_MIN_PKT_DESC_BYTES)
 		pad = 0;
 	info = &tx->info[reqi & tx->mask];
 	info->xdp_frame = frame_p;
@@ -810,7 +810,7 @@ int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx,
 {
 	int nsegs;
 
-	if (!gve_can_tx(tx, len + GVE_TX_MAX_HEADER_SIZE - 1))
+	if (!gve_can_tx(tx, len + GVE_GQ_TX_MIN_PKT_DESC_BYTES - 1))
 		return -EBUSY;
 
 	nsegs = gve_tx_fill_xdp(priv, tx, data, len, frame_p, false);
-- 
cgit v1.2.3-70-g09d2


From eafa92152e2ec6318e32b6ddda9c1d95d161000a Mon Sep 17 00:00:00 2001
From: Alejandro Colomar <alx@kernel.org>
Date: Tue, 11 Apr 2023 15:47:47 +0100
Subject: bpf: Remove extra whitespace in SPDX tag for syscall/helpers man
 pages

There is an extra whitespace in the SPDX tag, before the license name,
in the script for generating man pages for the bpf() syscall and the
helpers. It has caused problems in Debian packaging, in the tool that
autodetects licenses. Let's clean it up.

Fixes: 5cb62b7598f2 ("bpf, docs: Use SPDX license identifier in bpf_doc.py")
Signed-off-by: Alejandro Colomar <alx@kernel.org>
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230411144747.66734-1-quentin@isovalent.com
---
 scripts/bpf_doc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/bpf_doc.py b/scripts/bpf_doc.py
index 38d51e05c7a2..eaae2ce78381 100755
--- a/scripts/bpf_doc.py
+++ b/scripts/bpf_doc.py
@@ -383,7 +383,7 @@ class PrinterRST(Printer):
 .. Copyright (C) All BPF authors and contributors from 2014 to present.
 .. See git log include/uapi/linux/bpf.h in kernel tree for details.
 .. 
-.. SPDX-License-Identifier:  Linux-man-pages-copyleft
+.. SPDX-License-Identifier: Linux-man-pages-copyleft
 .. 
 .. Please do not edit this file. It was generated from the documentation
 .. located in file include/uapi/linux/bpf.h of the Linux kernel sources
-- 
cgit v1.2.3-70-g09d2


From 4294a0a7ab6282c3d92f03de84e762dda993c93d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:47 -0700
Subject: bpf: Split off basic BPF verifier log into separate file

kernel/bpf/verifier.c file is large and growing larger all the time. So
it's good to start splitting off more or less self-contained parts into
separate files to keep source code size (somewhat) somewhat under
control.

This patch is a one step in this direction, moving some of BPF verifier log
routines into a separate kernel/bpf/log.c. Right now it's most low-level
and isolated routines to append data to log, reset log to previous
position, etc. Eventually we could probably move verifier state
printing logic here as well, but this patch doesn't attempt to do that
yet.

Subsequent patches will add more logic to verifier log management, so
having basics in a separate file will make sure verifier.c doesn't grow
more with new changes.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-2-andrii@kernel.org
---
 include/linux/bpf_verifier.h | 19 ++++------
 kernel/bpf/Makefile          |  3 +-
 kernel/bpf/log.c             | 85 ++++++++++++++++++++++++++++++++++++++++++++
 kernel/bpf/verifier.c        | 69 -----------------------------------
 4 files changed, 94 insertions(+), 82 deletions(-)
 create mode 100644 kernel/bpf/log.c

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 81d525d057c7..83dff25545ee 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -498,11 +498,6 @@ struct bpf_verifier_log {
 	u32 len_total;
 };
 
-static inline bool bpf_verifier_log_full(const struct bpf_verifier_log *log)
-{
-	return log->len_used >= log->len_total - 1;
-}
-
 #define BPF_LOG_LEVEL1	1
 #define BPF_LOG_LEVEL2	2
 #define BPF_LOG_STATS	4
@@ -512,6 +507,11 @@ static inline bool bpf_verifier_log_full(const struct bpf_verifier_log *log)
 #define BPF_LOG_MIN_ALIGNMENT 8U
 #define BPF_LOG_ALIGNMENT 40U
 
+static inline bool bpf_verifier_log_full(const struct bpf_verifier_log *log)
+{
+	return log->len_used >= log->len_total - 1;
+}
+
 static inline bool bpf_verifier_log_needed(const struct bpf_verifier_log *log)
 {
 	return log &&
@@ -519,13 +519,6 @@ static inline bool bpf_verifier_log_needed(const struct bpf_verifier_log *log)
 		 log->level == BPF_LOG_KERNEL);
 }
 
-static inline bool
-bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
-{
-	return log->len_total >= 128 && log->len_total <= UINT_MAX >> 2 &&
-	       log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
-}
-
 #define BPF_MAX_SUBPROGS 256
 
 struct bpf_subprog_info {
@@ -608,12 +601,14 @@ struct bpf_verifier_env {
 	char type_str_buf[TYPE_STR_BUF_LEN];
 };
 
+bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log);
 __printf(2, 0) void bpf_verifier_vlog(struct bpf_verifier_log *log,
 				      const char *fmt, va_list args);
 __printf(2, 3) void bpf_verifier_log_write(struct bpf_verifier_env *env,
 					   const char *fmt, ...);
 __printf(2, 3) void bpf_log(struct bpf_verifier_log *log,
 			    const char *fmt, ...);
+void bpf_vlog_reset(struct bpf_verifier_log *log, u32 new_pos);
 
 static inline struct bpf_func_state *cur_func(struct bpf_verifier_env *env)
 {
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index 02242614dcc7..1d3892168d32 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -6,7 +6,8 @@ cflags-nogcse-$(CONFIG_X86)$(CONFIG_CC_IS_GCC) := -fno-gcse
 endif
 CFLAGS_core.o += $(call cc-disable-warning, override-init) $(cflags-nogcse-yy)
 
-obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o tnum.o bpf_iter.o map_iter.o task_iter.o prog_iter.o link_iter.o
+obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o tnum.o log.o
+obj-$(CONFIG_BPF_SYSCALL) += bpf_iter.o map_iter.o task_iter.o prog_iter.o link_iter.o
 obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o percpu_freelist.o bpf_lru_list.o lpm_trie.o map_in_map.o bloom_filter.o
 obj-$(CONFIG_BPF_SYSCALL) += local_storage.o queue_stack_maps.o ringbuf.o
 obj-$(CONFIG_BPF_SYSCALL) += bpf_local_storage.o bpf_task_storage.o
diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
new file mode 100644
index 000000000000..920061e38d2e
--- /dev/null
+++ b/kernel/bpf/log.c
@@ -0,0 +1,85 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
+ * Copyright (c) 2016 Facebook
+ * Copyright (c) 2018 Covalent IO, Inc. http://covalent.io
+ */
+#include <uapi/linux/btf.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/bpf.h>
+#include <linux/bpf_verifier.h>
+
+bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
+{
+	return log->len_total >= 128 && log->len_total <= UINT_MAX >> 2 &&
+	       log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
+}
+
+void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
+		       va_list args)
+{
+	unsigned int n;
+
+	n = vscnprintf(log->kbuf, BPF_VERIFIER_TMP_LOG_SIZE, fmt, args);
+
+	WARN_ONCE(n >= BPF_VERIFIER_TMP_LOG_SIZE - 1,
+		  "verifier log line truncated - local buffer too short\n");
+
+	if (log->level == BPF_LOG_KERNEL) {
+		bool newline = n > 0 && log->kbuf[n - 1] == '\n';
+
+		pr_err("BPF: %s%s", log->kbuf, newline ? "" : "\n");
+		return;
+	}
+
+	n = min(log->len_total - log->len_used - 1, n);
+	log->kbuf[n] = '\0';
+	if (!copy_to_user(log->ubuf + log->len_used, log->kbuf, n + 1))
+		log->len_used += n;
+	else
+		log->ubuf = NULL;
+}
+
+void bpf_vlog_reset(struct bpf_verifier_log *log, u32 new_pos)
+{
+	char zero = 0;
+
+	if (!bpf_verifier_log_needed(log))
+		return;
+
+	log->len_used = new_pos;
+	if (put_user(zero, log->ubuf + new_pos))
+		log->ubuf = NULL;
+}
+
+/* log_level controls verbosity level of eBPF verifier.
+ * bpf_verifier_log_write() is used to dump the verification trace to the log,
+ * so the user can figure out what's wrong with the program
+ */
+__printf(2, 3) void bpf_verifier_log_write(struct bpf_verifier_env *env,
+					   const char *fmt, ...)
+{
+	va_list args;
+
+	if (!bpf_verifier_log_needed(&env->log))
+		return;
+
+	va_start(args, fmt);
+	bpf_verifier_vlog(&env->log, fmt, args);
+	va_end(args);
+}
+EXPORT_SYMBOL_GPL(bpf_verifier_log_write);
+
+__printf(2, 3) void bpf_log(struct bpf_verifier_log *log,
+			    const char *fmt, ...)
+{
+	va_list args;
+
+	if (!bpf_verifier_log_needed(log))
+		return;
+
+	va_start(args, fmt);
+	bpf_verifier_vlog(log, fmt, args);
+	va_end(args);
+}
+EXPORT_SYMBOL_GPL(bpf_log);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 3660b573048a..745ae0cd01d4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -335,61 +335,6 @@ find_linfo(const struct bpf_verifier_env *env, u32 insn_off)
 	return &linfo[i - 1];
 }
 
-void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
-		       va_list args)
-{
-	unsigned int n;
-
-	n = vscnprintf(log->kbuf, BPF_VERIFIER_TMP_LOG_SIZE, fmt, args);
-
-	WARN_ONCE(n >= BPF_VERIFIER_TMP_LOG_SIZE - 1,
-		  "verifier log line truncated - local buffer too short\n");
-
-	if (log->level == BPF_LOG_KERNEL) {
-		bool newline = n > 0 && log->kbuf[n - 1] == '\n';
-
-		pr_err("BPF: %s%s", log->kbuf, newline ? "" : "\n");
-		return;
-	}
-
-	n = min(log->len_total - log->len_used - 1, n);
-	log->kbuf[n] = '\0';
-	if (!copy_to_user(log->ubuf + log->len_used, log->kbuf, n + 1))
-		log->len_used += n;
-	else
-		log->ubuf = NULL;
-}
-
-static void bpf_vlog_reset(struct bpf_verifier_log *log, u32 new_pos)
-{
-	char zero = 0;
-
-	if (!bpf_verifier_log_needed(log))
-		return;
-
-	log->len_used = new_pos;
-	if (put_user(zero, log->ubuf + new_pos))
-		log->ubuf = NULL;
-}
-
-/* log_level controls verbosity level of eBPF verifier.
- * bpf_verifier_log_write() is used to dump the verification trace to the log,
- * so the user can figure out what's wrong with the program
- */
-__printf(2, 3) void bpf_verifier_log_write(struct bpf_verifier_env *env,
-					   const char *fmt, ...)
-{
-	va_list args;
-
-	if (!bpf_verifier_log_needed(&env->log))
-		return;
-
-	va_start(args, fmt);
-	bpf_verifier_vlog(&env->log, fmt, args);
-	va_end(args);
-}
-EXPORT_SYMBOL_GPL(bpf_verifier_log_write);
-
 __printf(2, 3) static void verbose(void *private_data, const char *fmt, ...)
 {
 	struct bpf_verifier_env *env = private_data;
@@ -403,20 +348,6 @@ __printf(2, 3) static void verbose(void *private_data, const char *fmt, ...)
 	va_end(args);
 }
 
-__printf(2, 3) void bpf_log(struct bpf_verifier_log *log,
-			    const char *fmt, ...)
-{
-	va_list args;
-
-	if (!bpf_verifier_log_needed(log))
-		return;
-
-	va_start(args, fmt);
-	bpf_verifier_vlog(log, fmt, args);
-	va_end(args);
-}
-EXPORT_SYMBOL_GPL(bpf_log);
-
 static const char *ltrim(const char *s)
 {
 	while (isspace(*s))
-- 
cgit v1.2.3-70-g09d2


From 03cc3aa6a53394481f01c16231f99298332066f9 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:48 -0700
Subject: bpf: Remove minimum size restrictions on verifier log buffer

It's not clear why we have 128 as minimum size, but it makes testing
harder and seems unnecessary, as we carefully handle truncation
scenarios and use proper snprintf variants. So remove this limitation
and just enforce positive length for log buffer.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-3-andrii@kernel.org
---
 kernel/bpf/log.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 920061e38d2e..1974891fc324 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -11,7 +11,7 @@
 
 bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
 {
-	return log->len_total >= 128 && log->len_total <= UINT_MAX >> 2 &&
+	return log->len_total > 0 && log->len_total <= UINT_MAX >> 2 &&
 	       log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 1216640938035e63bdbd32438e91c9bcc1fd8ee1 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:49 -0700
Subject: bpf: Switch BPF verifier log to be a rotating log by default

Currently, if user-supplied log buffer to collect BPF verifier log turns
out to be too small to contain full log, bpf() syscall returns -ENOSPC,
fails BPF program verification/load, and preserves first N-1 bytes of
the verifier log (where N is the size of user-supplied buffer).

This is problematic in a bunch of common scenarios, especially when
working with real-world BPF programs that tend to be pretty complex as
far as verification goes and require big log buffers. Typically, it's
when debugging tricky cases at log level 2 (verbose). Also, when BPF program
is successfully validated, log level 2 is the only way to actually see
verifier state progression and all the important details.

Even with log level 1, it's possible to get -ENOSPC even if the final
verifier log fits in log buffer, if there is a code path that's deep
enough to fill up entire log, even if normally it would be reset later
on (there is a logic to chop off successfully validated portions of BPF
verifier log).

In short, it's not always possible to pre-size log buffer. Also, what's
worse, in practice, the end of the log most often is way more important
than the beginning, but verifier stops emitting log as soon as initial
log buffer is filled up.

This patch switches BPF verifier log behavior to effectively behave as
rotating log. That is, if user-supplied log buffer turns out to be too
short, verifier will keep overwriting previously written log,
effectively treating user's log buffer as a ring buffer. -ENOSPC is
still going to be returned at the end, to notify user that log contents
was truncated, but the important last N bytes of the log would be
returned, which might be all that user really needs. This consistent
-ENOSPC behavior, regardless of rotating or fixed log behavior, allows
to prevent backwards compatibility breakage. The only user-visible
change is which portion of verifier log user ends up seeing *if buffer
is too small*. Given contents of verifier log itself is not an ABI,
there is no breakage due to this behavior change. Specialized tools that
rely on specific contents of verifier log in -ENOSPC scenario are
expected to be easily adapted to accommodate old and new behaviors.

Importantly, though, to preserve good user experience and not require
every user-space application to adopt to this new behavior, before
exiting to user-space verifier will rotate log (in place) to make it
start at the very beginning of user buffer as a continuous
zero-terminated string. The contents will be a chopped off N-1 last
bytes of full verifier log, of course.

Given beginning of log is sometimes important as well, we add
BPF_LOG_FIXED (which equals 8) flag to force old behavior, which allows
tools like veristat to request first part of verifier log, if necessary.
BPF_LOG_FIXED flag is also a simple and straightforward way to check if
BPF verifier supports rotating behavior.

On the implementation side, conceptually, it's all simple. We maintain
64-bit logical start and end positions. If we need to truncate the log,
start position will be adjusted accordingly to lag end position by
N bytes. We then use those logical positions to calculate their matching
actual positions in user buffer and handle wrap around the end of the
buffer properly. Finally, right before returning from bpf_check(), we
rotate user log buffer contents in-place as necessary, to make log
contents contiguous. See comments in relevant functions for details.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-4-andrii@kernel.org
---
 include/linux/bpf_verifier.h                       |  33 +++-
 kernel/bpf/btf.c                                   |   3 +-
 kernel/bpf/log.c                                   | 198 ++++++++++++++++++++-
 kernel/bpf/verifier.c                              |  19 +-
 tools/testing/selftests/bpf/prog_tests/log_fixup.c |   1 +
 5 files changed, 228 insertions(+), 26 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 83dff25545ee..4c926227f612 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -491,25 +491,42 @@ struct bpf_insn_aux_data {
 #define BPF_VERIFIER_TMP_LOG_SIZE	1024
 
 struct bpf_verifier_log {
-	u32 level;
-	char kbuf[BPF_VERIFIER_TMP_LOG_SIZE];
+	/* Logical start and end positions of a "log window" of the verifier log.
+	 * start_pos == 0 means we haven't truncated anything.
+	 * Once truncation starts to happen, start_pos + len_total == end_pos,
+	 * except during log reset situations, in which (end_pos - start_pos)
+	 * might get smaller than len_total (see bpf_vlog_reset()).
+	 * Generally, (end_pos - start_pos) gives number of useful data in
+	 * user log buffer.
+	 */
+	u64 start_pos;
+	u64 end_pos;
 	char __user *ubuf;
-	u32 len_used;
+	u32 level;
 	u32 len_total;
+	char kbuf[BPF_VERIFIER_TMP_LOG_SIZE];
 };
 
 #define BPF_LOG_LEVEL1	1
 #define BPF_LOG_LEVEL2	2
 #define BPF_LOG_STATS	4
+#define BPF_LOG_FIXED	8
 #define BPF_LOG_LEVEL	(BPF_LOG_LEVEL1 | BPF_LOG_LEVEL2)
-#define BPF_LOG_MASK	(BPF_LOG_LEVEL | BPF_LOG_STATS)
+#define BPF_LOG_MASK	(BPF_LOG_LEVEL | BPF_LOG_STATS | BPF_LOG_FIXED)
 #define BPF_LOG_KERNEL	(BPF_LOG_MASK + 1) /* kernel internal flag */
 #define BPF_LOG_MIN_ALIGNMENT 8U
 #define BPF_LOG_ALIGNMENT 40U
 
+static inline u32 bpf_log_used(const struct bpf_verifier_log *log)
+{
+	return log->end_pos - log->start_pos;
+}
+
 static inline bool bpf_verifier_log_full(const struct bpf_verifier_log *log)
 {
-	return log->len_used >= log->len_total - 1;
+	if (log->level & BPF_LOG_FIXED)
+		return bpf_log_used(log) >= log->len_total - 1;
+	return false;
 }
 
 static inline bool bpf_verifier_log_needed(const struct bpf_verifier_log *log)
@@ -596,7 +613,7 @@ struct bpf_verifier_env {
 	u32 scratched_regs;
 	/* Same as scratched_regs but for stack slots */
 	u64 scratched_stack_slots;
-	u32 prev_log_len, prev_insn_print_len;
+	u64 prev_log_pos, prev_insn_print_pos;
 	/* buffer used in reg_type_str() to generate reg_type string */
 	char type_str_buf[TYPE_STR_BUF_LEN];
 };
@@ -608,7 +625,9 @@ __printf(2, 3) void bpf_verifier_log_write(struct bpf_verifier_env *env,
 					   const char *fmt, ...);
 __printf(2, 3) void bpf_log(struct bpf_verifier_log *log,
 			    const char *fmt, ...);
-void bpf_vlog_reset(struct bpf_verifier_log *log, u32 new_pos);
+void bpf_vlog_reset(struct bpf_verifier_log *log, u64 new_pos);
+void bpf_vlog_finalize(struct bpf_verifier_log *log);
+bool bpf_vlog_truncated(const struct bpf_verifier_log *log);
 
 static inline struct bpf_func_state *cur_func(struct bpf_verifier_env *env)
 {
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 593c45a294d0..20a05b8932db 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5593,7 +5593,8 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 		}
 	}
 
-	if (log->level && bpf_verifier_log_full(log)) {
+	bpf_vlog_finalize(log);
+	if (log->level && bpf_vlog_truncated(log)) {
 		err = -ENOSPC;
 		goto errout_meta;
 	}
diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 1974891fc324..92b1c8ad6601 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -8,6 +8,7 @@
 #include <linux/types.h>
 #include <linux/bpf.h>
 #include <linux/bpf_verifier.h>
+#include <linux/math64.h>
 
 bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
 {
@@ -32,23 +33,202 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 		return;
 	}
 
-	n = min(log->len_total - log->len_used - 1, n);
-	log->kbuf[n] = '\0';
-	if (!copy_to_user(log->ubuf + log->len_used, log->kbuf, n + 1))
-		log->len_used += n;
-	else
-		log->ubuf = NULL;
+	if (log->level & BPF_LOG_FIXED) {
+		n = min(log->len_total - bpf_log_used(log) - 1, n);
+		log->kbuf[n] = '\0';
+		n += 1;
+
+		if (copy_to_user(log->ubuf + log->end_pos, log->kbuf, n))
+			goto fail;
+
+		log->end_pos += n - 1; /* don't count terminating '\0' */
+	} else {
+		u64 new_end, new_start, cur_pos;
+		u32 buf_start, buf_end, new_n;
+
+		n += 1;
+
+		new_end = log->end_pos + n;
+		if (new_end - log->start_pos >= log->len_total)
+			new_start = new_end - log->len_total;
+		else
+			new_start = log->start_pos;
+		new_n = min(n, log->len_total);
+		cur_pos = new_end - new_n;
+
+		div_u64_rem(cur_pos, log->len_total, &buf_start);
+		div_u64_rem(new_end, log->len_total, &buf_end);
+		/* new_end and buf_end are exclusive indices, so if buf_end is
+		 * exactly zero, then it actually points right to the end of
+		 * ubuf and there is no wrap around
+		 */
+		if (buf_end == 0)
+			buf_end = log->len_total;
+
+		/* if buf_start > buf_end, we wrapped around;
+		 * if buf_start == buf_end, then we fill ubuf completely; we
+		 * can't have buf_start == buf_end to mean that there is
+		 * nothing to write, because we always write at least
+		 * something, even if terminal '\0'
+		 */
+		if (buf_start < buf_end) {
+			/* message fits within contiguous chunk of ubuf */
+			if (copy_to_user(log->ubuf + buf_start,
+					 log->kbuf + n - new_n,
+					 buf_end - buf_start))
+				goto fail;
+		} else {
+			/* message wraps around the end of ubuf, copy in two chunks */
+			if (copy_to_user(log->ubuf + buf_start,
+					 log->kbuf + n - new_n,
+					 log->len_total - buf_start))
+				goto fail;
+			if (copy_to_user(log->ubuf,
+					 log->kbuf + n - buf_end,
+					 buf_end))
+				goto fail;
+		}
+
+		log->start_pos = new_start;
+		log->end_pos = new_end - 1; /* don't count terminating '\0' */
+	}
+
+	return;
+fail:
+	log->ubuf = NULL;
 }
 
-void bpf_vlog_reset(struct bpf_verifier_log *log, u32 new_pos)
+void bpf_vlog_reset(struct bpf_verifier_log *log, u64 new_pos)
 {
 	char zero = 0;
+	u32 pos;
+
+	if (WARN_ON_ONCE(new_pos > log->end_pos))
+		return;
 
 	if (!bpf_verifier_log_needed(log))
 		return;
 
-	log->len_used = new_pos;
-	if (put_user(zero, log->ubuf + new_pos))
+	/* if position to which we reset is beyond current log window,
+	 * then we didn't preserve any useful content and should adjust
+	 * start_pos to end up with an empty log (start_pos == end_pos)
+	 */
+	log->end_pos = new_pos;
+	if (log->end_pos < log->start_pos)
+		log->start_pos = log->end_pos;
+	div_u64_rem(new_pos, log->len_total, &pos);
+	if (put_user(zero, log->ubuf + pos))
+		log->ubuf = NULL;
+}
+
+static void bpf_vlog_reverse_kbuf(char *buf, int len)
+{
+	int i, j;
+
+	for (i = 0, j = len - 1; i < j; i++, j--)
+		swap(buf[i], buf[j]);
+}
+
+static int bpf_vlog_reverse_ubuf(struct bpf_verifier_log *log, int start, int end)
+{
+	/* we split log->kbuf into two equal parts for both ends of array */
+	int n = sizeof(log->kbuf) / 2, nn;
+	char *lbuf = log->kbuf, *rbuf = log->kbuf + n;
+
+	/* Read ubuf's section [start, end) two chunks at a time, from left
+	 * and right side; within each chunk, swap all the bytes; after that
+	 * reverse the order of lbuf and rbuf and write result back to ubuf.
+	 * This way we'll end up with swapped contents of specified
+	 * [start, end) ubuf segment.
+	 */
+	while (end - start > 1) {
+		nn = min(n, (end - start ) / 2);
+
+		if (copy_from_user(lbuf, log->ubuf + start, nn))
+			return -EFAULT;
+		if (copy_from_user(rbuf, log->ubuf + end - nn, nn))
+			return -EFAULT;
+
+		bpf_vlog_reverse_kbuf(lbuf, nn);
+		bpf_vlog_reverse_kbuf(rbuf, nn);
+
+		/* we write lbuf to the right end of ubuf, while rbuf to the
+		 * left one to end up with properly reversed overall ubuf
+		 */
+		if (copy_to_user(log->ubuf + start, rbuf, nn))
+			return -EFAULT;
+		if (copy_to_user(log->ubuf + end - nn, lbuf, nn))
+			return -EFAULT;
+
+		start += nn;
+		end -= nn;
+	}
+
+	return 0;
+}
+
+bool bpf_vlog_truncated(const struct bpf_verifier_log *log)
+{
+	if (log->level & BPF_LOG_FIXED)
+		return bpf_log_used(log) >= log->len_total - 1;
+	else
+		return log->start_pos > 0;
+}
+
+void bpf_vlog_finalize(struct bpf_verifier_log *log)
+{
+	u32 sublen;
+	int err;
+
+	if (!log || !log->level || !log->ubuf)
+		return;
+	if ((log->level & BPF_LOG_FIXED) || log->level == BPF_LOG_KERNEL)
+		return;
+
+	/* If we never truncated log, there is nothing to move around. */
+	if (log->start_pos == 0)
+		return;
+
+	/* Otherwise we need to rotate log contents to make it start from the
+	 * buffer beginning and be a continuous zero-terminated string. Note
+	 * that if log->start_pos != 0 then we definitely filled up entire log
+	 * buffer with no gaps, and we just need to shift buffer contents to
+	 * the left by (log->start_pos % log->len_total) bytes.
+	 *
+	 * Unfortunately, user buffer could be huge and we don't want to
+	 * allocate temporary kernel memory of the same size just to shift
+	 * contents in a straightforward fashion. Instead, we'll be clever and
+	 * do in-place array rotation. This is a leetcode-style problem, which
+	 * could be solved by three rotations.
+	 *
+	 * Let's say we have log buffer that has to be shifted left by 7 bytes
+	 * (spaces and vertical bar is just for demonstrative purposes):
+	 *   E F G H I J K | A B C D
+	 *
+	 * First, we reverse entire array:
+	 *   D C B A | K J I H G F E
+	 *
+	 * Then we rotate first 4 bytes (DCBA) and separately last 7 bytes
+	 * (KJIHGFE), resulting in a properly rotated array:
+	 *   A B C D | E F G H I J K
+	 *
+	 * We'll utilize log->kbuf to read user memory chunk by chunk, swap
+	 * bytes, and write them back. Doing it byte-by-byte would be
+	 * unnecessarily inefficient. Altogether we are going to read and
+	 * write each byte twice, for total 4 memory copies between kernel and
+	 * user space.
+	 */
+
+	/* length of the chopped off part that will be the beginning;
+	 * len(ABCD) in the example above
+	 */
+	div_u64_rem(log->start_pos, log->len_total, &sublen);
+	sublen = log->len_total - sublen;
+
+	err = bpf_vlog_reverse_ubuf(log, 0, log->len_total);
+	err = err ?: bpf_vlog_reverse_ubuf(log, 0, sublen);
+	err = err ?: bpf_vlog_reverse_ubuf(log, sublen, log->len_total);
+	if (err)
 		log->ubuf = NULL;
 }
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 745ae0cd01d4..a476bb319685 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1439,10 +1439,10 @@ static inline u32 vlog_alignment(u32 pos)
 static void print_insn_state(struct bpf_verifier_env *env,
 			     const struct bpf_func_state *state)
 {
-	if (env->prev_log_len && env->prev_log_len == env->log.len_used) {
+	if (env->prev_log_pos && env->prev_log_pos == env->log.end_pos) {
 		/* remove new line character */
-		bpf_vlog_reset(&env->log, env->prev_log_len - 1);
-		verbose(env, "%*c;", vlog_alignment(env->prev_insn_print_len), ' ');
+		bpf_vlog_reset(&env->log, env->prev_log_pos - 1);
+		verbose(env, "%*c;", vlog_alignment(env->prev_insn_print_pos), ' ');
 	} else {
 		verbose(env, "%d:", env->insn_idx);
 	}
@@ -1750,7 +1750,7 @@ static struct bpf_verifier_state *push_stack(struct bpf_verifier_env *env,
 	elem->insn_idx = insn_idx;
 	elem->prev_insn_idx = prev_insn_idx;
 	elem->next = env->head;
-	elem->log_pos = env->log.len_used;
+	elem->log_pos = env->log.end_pos;
 	env->head = elem;
 	env->stack_size++;
 	err = copy_verifier_state(&elem->st, cur);
@@ -2286,7 +2286,7 @@ static struct bpf_verifier_state *push_async_cb(struct bpf_verifier_env *env,
 	elem->insn_idx = insn_idx;
 	elem->prev_insn_idx = prev_insn_idx;
 	elem->next = env->head;
-	elem->log_pos = env->log.len_used;
+	elem->log_pos = env->log.end_pos;
 	env->head = elem;
 	env->stack_size++;
 	if (env->stack_size > BPF_COMPLEXITY_LIMIT_JMP_SEQ) {
@@ -15638,11 +15638,11 @@ static int do_check(struct bpf_verifier_env *env)
 				print_insn_state(env, state->frame[state->curframe]);
 
 			verbose_linfo(env, env->insn_idx, "; ");
-			env->prev_log_len = env->log.len_used;
+			env->prev_log_pos = env->log.end_pos;
 			verbose(env, "%d: ", env->insn_idx);
 			print_bpf_insn(&cbs, insn, env->allow_ptr_leaks);
-			env->prev_insn_print_len = env->log.len_used - env->prev_log_len;
-			env->prev_log_len = env->log.len_used;
+			env->prev_insn_print_pos = env->log.end_pos - env->prev_log_pos;
+			env->prev_log_pos = env->log.end_pos;
 		}
 
 		if (bpf_prog_is_offloaded(env->prog->aux)) {
@@ -18860,7 +18860,8 @@ skip_full_check:
 	print_verification_stats(env);
 	env->prog->aux->verified_insns = env->insn_processed;
 
-	if (log->level && bpf_verifier_log_full(log))
+	bpf_vlog_finalize(log);
+	if (log->level && bpf_vlog_truncated(log))
 		ret = -ENOSPC;
 	if (log->level && !log->ubuf) {
 		ret = -EFAULT;
diff --git a/tools/testing/selftests/bpf/prog_tests/log_fixup.c b/tools/testing/selftests/bpf/prog_tests/log_fixup.c
index 239e1c5753b0..bc27170bdeb0 100644
--- a/tools/testing/selftests/bpf/prog_tests/log_fixup.c
+++ b/tools/testing/selftests/bpf/prog_tests/log_fixup.c
@@ -24,6 +24,7 @@ static void bad_core_relo(size_t log_buf_size, enum trunc_type trunc_type)
 	bpf_program__set_autoload(skel->progs.bad_relo, true);
 	memset(log_buf, 0, sizeof(log_buf));
 	bpf_program__set_log_buf(skel->progs.bad_relo, log_buf, log_buf_size ?: sizeof(log_buf));
+	bpf_program__set_log_level(skel->progs.bad_relo, 1 | 8); /* BPF_LOG_FIXED to force truncation */
 
 	err = test_log_fixup__load(skel);
 	if (!ASSERT_ERR(err, "load_fail"))
-- 
cgit v1.2.3-70-g09d2


From e0aee1facccf9f12136600031be4ce21eb810a78 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:50 -0700
Subject: libbpf: Don't enforce unnecessary verifier log restrictions on libbpf
 side

This basically prevents any forward compatibility. And we either way
just return -EINVAL, which would otherwise be returned from bpf()
syscall anyways.

Similarly, drop enforcement of non-NULL log_buf when log_level > 0. This
won't be true anymore soon.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-5-andrii@kernel.org
---
 tools/lib/bpf/bpf.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 767035900354..f1d04ee14d83 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -290,10 +290,6 @@ int bpf_prog_load(enum bpf_prog_type prog_type,
 
 	if (!!log_buf != !!log_size)
 		return libbpf_err(-EINVAL);
-	if (log_level > (4 | 2 | 1))
-		return libbpf_err(-EINVAL);
-	if (log_level && !log_buf)
-		return libbpf_err(-EINVAL);
 
 	func_info_rec_size = OPTS_GET(opts, func_info_rec_size, 0);
 	func_info = OPTS_GET(opts, func_info, NULL);
-- 
cgit v1.2.3-70-g09d2


From d0d75c67c45abd3930967dcafc82fd4505400665 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:51 -0700
Subject: veristat: Add more veristat control over verifier log options

Add --log-size to be able to customize log buffer sent to bpf() syscall
for BPF program verification logging.

Add --log-fixed to enforce BPF_LOG_FIXED behavior for BPF verifier log.
This is useful in unlikely event that beginning of truncated verifier
log is more important than the end of it (which with rotating verifier
log behavior is the default now).

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230406234205.323208-6-andrii@kernel.org
---
 tools/testing/selftests/bpf/veristat.c | 44 +++++++++++++++++++++++++++-------
 1 file changed, 35 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
index e05954e20bba..1db7185181da 100644
--- a/tools/testing/selftests/bpf/veristat.c
+++ b/tools/testing/selftests/bpf/veristat.c
@@ -141,12 +141,15 @@ static struct env {
 	bool verbose;
 	bool debug;
 	bool quiet;
-	int log_level;
 	enum resfmt out_fmt;
 	bool show_version;
 	bool comparison_mode;
 	bool replay_mode;
 
+	int log_level;
+	int log_size;
+	bool log_fixed;
+
 	struct verif_stats *prog_stats;
 	int prog_stat_cnt;
 
@@ -193,12 +196,19 @@ const char argp_program_doc[] =
 "   OR: veristat -C <baseline.csv> <comparison.csv>\n"
 "   OR: veristat -R <results.csv>\n";
 
+enum {
+	OPT_LOG_FIXED = 1000,
+	OPT_LOG_SIZE = 1001,
+};
+
 static const struct argp_option opts[] = {
 	{ NULL, 'h', NULL, OPTION_HIDDEN, "Show the full help" },
 	{ "version", 'V', NULL, 0, "Print version" },
 	{ "verbose", 'v', NULL, 0, "Verbose mode" },
-	{ "log-level", 'l', "LEVEL", 0, "Verifier log level (default 0 for normal mode, 1 for verbose mode)" },
 	{ "debug", 'd', NULL, 0, "Debug mode (turns on libbpf debug logging)" },
+	{ "log-level", 'l', "LEVEL", 0, "Verifier log level (default 0 for normal mode, 1 for verbose mode)" },
+	{ "log-fixed", OPT_LOG_FIXED, NULL, 0, "Disable verifier log rotation" },
+	{ "log-size", OPT_LOG_SIZE, "BYTES", 0, "Customize verifier log size (default to 16MB)" },
 	{ "quiet", 'q', NULL, 0, "Quiet mode" },
 	{ "emit", 'e', "SPEC", 0, "Specify stats to be emitted" },
 	{ "sort", 's', "SPEC", 0, "Specify sort order" },
@@ -263,6 +273,17 @@ static error_t parse_arg(int key, char *arg, struct argp_state *state)
 			argp_usage(state);
 		}
 		break;
+	case OPT_LOG_FIXED:
+		env.log_fixed = true;
+		break;
+	case OPT_LOG_SIZE:
+		errno = 0;
+		env.log_size = strtol(arg, NULL, 10);
+		if (errno) {
+			fprintf(stderr, "invalid log size: %s\n", arg);
+			argp_usage(state);
+		}
+		break;
 	case 'C':
 		env.comparison_mode = true;
 		break;
@@ -929,8 +950,8 @@ static int process_prog(const char *filename, struct bpf_object *obj, struct bpf
 {
 	const char *prog_name = bpf_program__name(prog);
 	const char *base_filename = basename(filename);
-	size_t buf_sz = sizeof(verif_log_buf);
-	char *buf = verif_log_buf;
+	char *buf;
+	int buf_sz, log_level;
 	struct verif_stats *stats;
 	int err = 0;
 	void *tmp;
@@ -948,18 +969,23 @@ static int process_prog(const char *filename, struct bpf_object *obj, struct bpf
 	memset(stats, 0, sizeof(*stats));
 
 	if (env.verbose) {
-		buf_sz = 16 * 1024 * 1024;
+		buf_sz = env.log_size ? env.log_size : 16 * 1024 * 1024;
 		buf = malloc(buf_sz);
 		if (!buf)
 			return -ENOMEM;
-		bpf_program__set_log_buf(prog, buf, buf_sz);
-		bpf_program__set_log_level(prog, env.log_level | 4); /* stats + log */
+		/* ensure we always request stats */
+		log_level = env.log_level | 4 | (env.log_fixed ? 8 : 0);
 	} else {
-		bpf_program__set_log_buf(prog, buf, buf_sz);
-		bpf_program__set_log_level(prog, 4); /* only verifier stats */
+		buf = verif_log_buf;
+		buf_sz = sizeof(verif_log_buf);
+		/* request only verifier stats */
+		log_level = 4 | (env.log_fixed ? 8 : 0);
 	}
 	verif_log_buf[0] = '\0';
 
+	bpf_program__set_log_buf(prog, buf, buf_sz);
+	bpf_program__set_log_level(prog, log_level);
+
 	/* increase chances of successful BPF object loading */
 	fixup_obj(obj, prog, base_filename);
 
-- 
cgit v1.2.3-70-g09d2


From b1a7a480a1120d4f70305f5e8859f527e0efe4a5 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:52 -0700
Subject: selftests/bpf: Add fixed vs rotating verifier log tests

Add selftests validating BPF_LOG_FIXED behavior, which used to be the
only behavior, and now default rotating BPF verifier log, which returns
just up to last N bytes of full verifier log, instead of returning
-ENOSPC.

To stress test correctness of in-kernel verifier log logic, we force it
to truncate program's verifier log to all lengths from 1 all the way to
its full size (about 450 bytes today). This was a useful stress test
while developing the feature.

For both fixed and rotating log modes we expect -ENOSPC if log contents
doesn't fit in user-supplied log buffer.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-7-andrii@kernel.org
---
 .../selftests/bpf/prog_tests/verifier_log.c        | 179 +++++++++++++++++++++
 1 file changed, 179 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/verifier_log.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier_log.c b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
new file mode 100644
index 000000000000..3284108a6ce8
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
@@ -0,0 +1,179 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <test_progs.h>
+#include <bpf/btf.h>
+
+#include "test_log_buf.skel.h"
+
+
+static bool check_prog_load(int prog_fd, bool expect_err, const char *tag)
+{
+	if (expect_err) {
+		if (!ASSERT_LT(prog_fd, 0, tag)) {
+			close(prog_fd);
+			return false;
+		}
+	} else /* !expect_err */ {
+		if (!ASSERT_GT(prog_fd, 0, tag))
+			return false;
+	}
+	return true;
+}
+
+static void verif_log_subtest(const char *name, bool expect_load_error, int log_level)
+{
+	LIBBPF_OPTS(bpf_prog_load_opts, opts);
+	struct {
+		/* strategically placed before others to avoid accidental modification by kernel */
+		char filler[1024];
+		char buf[1024];
+		/* strategically placed after buf[] to catch more accidental corruptions */
+		char reference[1024];
+	} logs;
+	char *exp_log, prog_name[16], op_name[32];
+	struct test_log_buf *skel;
+	struct bpf_program *prog;
+	const struct bpf_insn *insns;
+	size_t insn_cnt, fixed_log_sz;
+	int i, mode, err, prog_fd;
+
+	skel = test_log_buf__open();
+	if (!ASSERT_OK_PTR(skel, "skel_open"))
+		return;
+
+	bpf_object__for_each_program(prog, skel->obj) {
+		if (strcmp(bpf_program__name(prog), name) == 0)
+			bpf_program__set_autoload(prog, true);
+		else
+			bpf_program__set_autoload(prog, false);
+	}
+
+	err = test_log_buf__load(skel);
+	if (!expect_load_error && !ASSERT_OK(err, "unexpected_load_failure"))
+		goto cleanup;
+	if (expect_load_error && !ASSERT_ERR(err, "unexpected_load_success"))
+		goto cleanup;
+
+	insns = bpf_program__insns(skel->progs.good_prog);
+	insn_cnt = bpf_program__insn_cnt(skel->progs.good_prog);
+
+	opts.log_buf = logs.reference;
+	opts.log_size = sizeof(logs.reference);
+	opts.log_level = log_level | 8 /* BPF_LOG_FIXED */;
+	prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, "log_fixed",
+				"GPL", insns, insn_cnt, &opts);
+	if (!check_prog_load(prog_fd, expect_load_error, "fixed_buf_prog_load"))
+		goto cleanup;
+	close(prog_fd);
+
+	fixed_log_sz = strlen(logs.reference) + 1;
+	if (!ASSERT_GT(fixed_log_sz, 50, "fixed_log_sz"))
+		goto cleanup;
+	memset(logs.reference + fixed_log_sz, 0, sizeof(logs.reference) - fixed_log_sz);
+
+	/* validate BPF_LOG_FIXED works as verifier log used to work, that is:
+	 * we get -ENOSPC and beginning of the full verifier log. This only
+	 * works for log_level 2 and log_level 1 + failed program. For log
+	 * level 2 we don't reset log at all. For log_level 1 + failed program
+	 * we don't get to verification stats output. With log level 1
+	 * for successful program  final result will be just verifier stats.
+	 * But if provided too short log buf, kernel will NULL-out log->ubuf
+	 * and will stop emitting further log. This means we'll never see
+	 * predictable verifier stats.
+	 * Long story short, we do the following -ENOSPC test only for
+	 * predictable combinations.
+	 */
+	if (log_level >= 2 || expect_load_error) {
+		opts.log_buf = logs.buf;
+		opts.log_level = log_level | 8; /* fixed-length log */
+		opts.log_size = 25;
+
+		prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, "log_fixed50",
+					"GPL", insns, insn_cnt, &opts);
+		if (!ASSERT_EQ(prog_fd, -ENOSPC, "unexpected_log_fixed_prog_load_result")) {
+			if (prog_fd >= 0)
+				close(prog_fd);
+			goto cleanup;
+		}
+		if (!ASSERT_EQ(strlen(logs.buf), 24, "log_fixed_25"))
+			goto cleanup;
+		if (!ASSERT_STRNEQ(logs.buf, logs.reference, 24, op_name))
+			goto cleanup;
+	}
+
+	/* validate rolling verifier log logic: try all variations of log buf
+	 * length to force various truncation scenarios
+	 */
+	opts.log_buf = logs.buf;
+
+	/* rotating mode, then fixed mode */
+	for (mode = 1; mode >= 0; mode--) {
+		/* prefill logs.buf with 'A's to detect any write beyond allowed length */
+		memset(logs.filler, 'A', sizeof(logs.filler));
+		logs.filler[sizeof(logs.filler) - 1] = '\0';
+		memset(logs.buf, 'A', sizeof(logs.buf));
+		logs.buf[sizeof(logs.buf) - 1] = '\0';
+
+		for (i = 1; i < fixed_log_sz; i++) {
+			opts.log_size = i;
+			opts.log_level = log_level | (mode ? 0 : 8 /* BPF_LOG_FIXED */);
+
+			snprintf(prog_name, sizeof(prog_name),
+				 "log_%s_%d", mode ? "roll" : "fixed", i);
+			prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, prog_name,
+						"GPL", insns, insn_cnt, &opts);
+
+			snprintf(op_name, sizeof(op_name),
+				 "log_%s_prog_load_%d", mode ? "roll" : "fixed", i);
+			if (!ASSERT_EQ(prog_fd, -ENOSPC, op_name)) {
+				if (prog_fd >= 0)
+					close(prog_fd);
+				goto cleanup;
+			}
+
+			snprintf(op_name, sizeof(op_name),
+				 "log_%s_strlen_%d", mode ? "roll" : "fixed", i);
+			ASSERT_EQ(strlen(logs.buf), i - 1, op_name);
+
+			if (mode)
+				exp_log = logs.reference + fixed_log_sz - i;
+			else
+				exp_log = logs.reference;
+
+			snprintf(op_name, sizeof(op_name),
+				 "log_%s_contents_%d", mode ? "roll" : "fixed", i);
+			if (!ASSERT_STRNEQ(logs.buf, exp_log, i - 1, op_name)) {
+				printf("CMP:%d\nS1:'%s'\nS2:'%s'\n",
+					strncmp(logs.buf, exp_log, i - 1),
+					logs.buf, exp_log);
+				goto cleanup;
+			}
+
+			/* check that unused portions of logs.buf is not overwritten */
+			snprintf(op_name, sizeof(op_name),
+				 "log_%s_unused_%d", mode ? "roll" : "fixed", i);
+			if (!ASSERT_STREQ(logs.buf + i, logs.filler + i, op_name)) {
+				printf("CMP:%d\nS1:'%s'\nS2:'%s'\n",
+					strcmp(logs.buf + i, logs.filler + i),
+					logs.buf + i, logs.filler + i);
+				goto cleanup;
+			}
+		}
+	}
+
+cleanup:
+	test_log_buf__destroy(skel);
+}
+
+void test_verifier_log(void)
+{
+	if (test__start_subtest("good_prog-level1"))
+		verif_log_subtest("good_prog", false, 1);
+	if (test__start_subtest("good_prog-level2"))
+		verif_log_subtest("good_prog", false, 2);
+	if (test__start_subtest("bad_prog-level1"))
+		verif_log_subtest("bad_prog", true, 1);
+	if (test__start_subtest("bad_prog-level2"))
+		verif_log_subtest("bad_prog", true, 2);
+}
-- 
cgit v1.2.3-70-g09d2


From 24bc80887adb4d6fc0057d4f14fabeaa4502b2a0 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:53 -0700
Subject: bpf: Ignore verifier log reset in BPF_LOG_KERNEL mode

Verifier log position reset is meaningless in BPF_LOG_KERNEL mode, so
just exit early in bpf_vlog_reset() if log->level is BPF_LOG_KERNEL.

This avoid meaningless put_user() into NULL log->ubuf.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-8-andrii@kernel.org
---
 kernel/bpf/log.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 92b1c8ad6601..d99a50f07187 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -106,7 +106,7 @@ void bpf_vlog_reset(struct bpf_verifier_log *log, u64 new_pos)
 	if (WARN_ON_ONCE(new_pos > log->end_pos))
 		return;
 
-	if (!bpf_verifier_log_needed(log))
+	if (!bpf_verifier_log_needed(log) || log->level == BPF_LOG_KERNEL)
 		return;
 
 	/* if position to which we reset is beyond current log window,
-- 
cgit v1.2.3-70-g09d2


From 971fb5057d787d0a7e7c8cb910207c82e2db920e Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:54 -0700
Subject: bpf: Fix missing -EFAULT return on user log buf error in btf_parse()

btf_parse() is missing -EFAULT error return if log->ubuf was NULL-ed out
due to error while copying data into user-provided buffer. Add it, but
handle a special case of BPF_LOG_KERNEL in which log->ubuf is always NULL.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-9-andrii@kernel.org
---
 kernel/bpf/btf.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 20a05b8932db..6372c144a294 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5598,6 +5598,10 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 		err = -ENOSPC;
 		goto errout_meta;
 	}
+	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf) {
+		err = -EFAULT;
+		goto errout_meta;
+	}
 
 	btf_verifier_env_free(env);
 	refcount_set(&btf->refcnt, 1);
-- 
cgit v1.2.3-70-g09d2


From cbedb42a0da3bb48819b2200af4b4cb5d922c518 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:55 -0700
Subject: bpf: Avoid incorrect -EFAULT error in BPF_LOG_KERNEL mode

If verifier log is in BPF_LOG_KERNEL mode, no log->ubuf is expected and
it stays NULL throughout entire verification process. Don't erroneously
return -EFAULT in such case.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-10-andrii@kernel.org
---
 kernel/bpf/verifier.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index a476bb319685..0323149803f5 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18863,7 +18863,7 @@ skip_full_check:
 	bpf_vlog_finalize(log);
 	if (log->level && bpf_vlog_truncated(log))
 		ret = -ENOSPC;
-	if (log->level && !log->ubuf) {
+	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf) {
 		ret = -EFAULT;
 		goto err_release_maps;
 	}
-- 
cgit v1.2.3-70-g09d2


From 8a6ca6bc553e3c878fa53c506bc6ec281efdc039 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:56 -0700
Subject: bpf: Simplify logging-related error conditions handling

Move log->level == 0 check into bpf_vlog_truncated() instead of doing it
explicitly. Also remove unnecessary goto in kernel/bpf/verifier.c.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-11-andrii@kernel.org
---
 kernel/bpf/btf.c      | 2 +-
 kernel/bpf/log.c      | 4 +++-
 kernel/bpf/verifier.c | 6 ++----
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 6372c144a294..5aa540ee611f 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5594,7 +5594,7 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 	}
 
 	bpf_vlog_finalize(log);
-	if (log->level && bpf_vlog_truncated(log)) {
+	if (bpf_vlog_truncated(log)) {
 		err = -ENOSPC;
 		goto errout_meta;
 	}
diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index d99a50f07187..c778f3b290cb 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -169,7 +169,9 @@ static int bpf_vlog_reverse_ubuf(struct bpf_verifier_log *log, int start, int en
 
 bool bpf_vlog_truncated(const struct bpf_verifier_log *log)
 {
-	if (log->level & BPF_LOG_FIXED)
+	if (!log->level)
+		return false;
+	else if (log->level & BPF_LOG_FIXED)
 		return bpf_log_used(log) >= log->len_total - 1;
 	else
 		return log->start_pos > 0;
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 0323149803f5..a98cbc046d1e 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18861,12 +18861,10 @@ skip_full_check:
 	env->prog->aux->verified_insns = env->insn_processed;
 
 	bpf_vlog_finalize(log);
-	if (log->level && bpf_vlog_truncated(log))
+	if (bpf_vlog_truncated(log))
 		ret = -ENOSPC;
-	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf) {
+	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf)
 		ret = -EFAULT;
-		goto err_release_maps;
-	}
 
 	if (ret)
 		goto err_release_maps;
-- 
cgit v1.2.3-70-g09d2


From fa1c7d5cc404ac3b6e6b4ab6d00b07c76bd819be Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:57 -0700
Subject: bpf: Keep track of total log content size in both fixed and rolling
 modes

Change how we do accounting in BPF_LOG_FIXED mode and adopt log->end_pos
as *logical* log position. This means that we can go beyond physical log
buffer size now and be able to tell what log buffer size should be to
fit entire log contents without -ENOSPC.

To do this for BPF_LOG_FIXED mode, we need to remove a short-circuiting
logic of not vsnprintf()'ing further log content once we filled up
user-provided buffer, which is done by bpf_verifier_log_needed() checks.
We modify these checks to always keep going if log->level is non-zero
(i.e., log is requested), even if log->ubuf was NULL'ed out due to
copying data to user-space, or if entire log buffer is physically full.
We adopt bpf_verifier_vlog() routine to work correctly with
log->ubuf == NULL condition, performing log formatting into temporary
kernel buffer, doing all the necessary accounting, but just avoiding
copying data out if buffer is full or NULL'ed out.

With these changes, it's now possible to do this sort of determination of
log contents size in both BPF_LOG_FIXED and default rolling log mode.
We need to keep in mind bpf_vlog_reset(), though, which shrinks log
contents after successful verification of a particular code path. This
log reset means that log->end_pos isn't always increasing, so to return
back to users what should be the log buffer size to fit all log content
without causing -ENOSPC even in the presence of log resetting, we need
to keep maximum over "lifetime" of logging. We do this accounting in
bpf_vlog_update_len_max() helper.

A related and subtle aspect is that with this logical log->end_pos even in
BPF_LOG_FIXED mode we could temporary "overflow" buffer, but then reset
it back with bpf_vlog_reset() to a position inside user-supplied
log_buf. In such situation we still want to properly maintain
terminating zero. We will eventually return -ENOSPC even if final log
buffer is small (we detect this through log->len_max check). This
behavior is simpler to reason about and is consistent with current
behavior of verifier log. Handling of this required a small addition to
bpf_vlog_reset() logic to avoid doing put_user() beyond physical log
buffer dimensions.

Another issue to keep in mind is that we limit log buffer size to 32-bit
value and keep such log length as u32, but theoretically verifier could
produce huge log stretching beyond 4GB. Instead of keeping (and later
returning) 64-bit log length, we cap it at UINT_MAX. Current UAPI makes
it impossible to specify log buffer size bigger than 4GB anyways, so we
don't really loose anything here and keep everything consistently 32-bit
in UAPI. This property will be utilized in next patch.

Doing the same determination of maximum log buffer for rolling mode is
trivial, as log->end_pos and log->start_pos are already logical
positions, so there is nothing new there.

These changes do incidentally fix one small issue with previous logging
logic. Previously, if use provided log buffer of size N, and actual log
output was exactly N-1 bytes + terminating \0, kernel logic coun't
distinguish this condition from log truncation scenario which would end
up with truncated log contents of N-1 bytes + terminating \0 as well.

But now with log->end_pos being logical position that could go beyond
actual log buffer size, we can distinguish these two conditions, which
we do in this patch. This plays nicely with returning log_size_actual
(implemented in UAPI in the next patch), as we can now guarantee that if
user takes such log_size_actual and provides log buffer of that exact
size, they will not get -ENOSPC in return.

All in all, all these changes do conceptually unify fixed and rolling
log modes much better, and allow a nice feature requested by users:
knowing what should be the size of the buffer to avoid -ENOSPC.

We'll plumb this through the UAPI and the code in the next patch.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-12-andrii@kernel.org
---
 include/linux/bpf_verifier.h | 12 ++------
 kernel/bpf/log.c             | 67 ++++++++++++++++++++++++++++++--------------
 2 files changed, 49 insertions(+), 30 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 4c926227f612..98d2eb382dbb 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -504,6 +504,7 @@ struct bpf_verifier_log {
 	char __user *ubuf;
 	u32 level;
 	u32 len_total;
+	u32 len_max;
 	char kbuf[BPF_VERIFIER_TMP_LOG_SIZE];
 };
 
@@ -517,23 +518,16 @@ struct bpf_verifier_log {
 #define BPF_LOG_MIN_ALIGNMENT 8U
 #define BPF_LOG_ALIGNMENT 40U
 
-static inline u32 bpf_log_used(const struct bpf_verifier_log *log)
-{
-	return log->end_pos - log->start_pos;
-}
-
 static inline bool bpf_verifier_log_full(const struct bpf_verifier_log *log)
 {
 	if (log->level & BPF_LOG_FIXED)
-		return bpf_log_used(log) >= log->len_total - 1;
+		return log->end_pos >= log->len_total;
 	return false;
 }
 
 static inline bool bpf_verifier_log_needed(const struct bpf_verifier_log *log)
 {
-	return log &&
-		((log->level && log->ubuf && !bpf_verifier_log_full(log)) ||
-		 log->level == BPF_LOG_KERNEL);
+	return log && log->level;
 }
 
 #define BPF_MAX_SUBPROGS 256
diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index c778f3b290cb..47bea2fad6fe 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -16,10 +16,26 @@ bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
 	       log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
 }
 
+static void bpf_vlog_update_len_max(struct bpf_verifier_log *log, u32 add_len)
+{
+	/* add_len includes terminal \0, so no need for +1. */
+	u64 len = log->end_pos + add_len;
+
+	/* log->len_max could be larger than our current len due to
+	 * bpf_vlog_reset() calls, so we maintain the max of any length at any
+	 * previous point
+	 */
+	if (len > UINT_MAX)
+		log->len_max = UINT_MAX;
+	else if (len > log->len_max)
+		log->len_max = len;
+}
+
 void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 		       va_list args)
 {
-	unsigned int n;
+	u64 cur_pos;
+	u32 new_n, n;
 
 	n = vscnprintf(log->kbuf, BPF_VERIFIER_TMP_LOG_SIZE, fmt, args);
 
@@ -33,21 +49,27 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 		return;
 	}
 
-	if (log->level & BPF_LOG_FIXED) {
-		n = min(log->len_total - bpf_log_used(log) - 1, n);
-		log->kbuf[n] = '\0';
-		n += 1;
+	n += 1; /* include terminating zero */
+	bpf_vlog_update_len_max(log, n);
 
-		if (copy_to_user(log->ubuf + log->end_pos, log->kbuf, n))
-			goto fail;
+	if (log->level & BPF_LOG_FIXED) {
+		/* check if we have at least something to put into user buf */
+		new_n = 0;
+		if (log->end_pos < log->len_total) {
+			new_n = min_t(u32, log->len_total - log->end_pos, n);
+			log->kbuf[new_n - 1] = '\0';
+		}
 
+		cur_pos = log->end_pos;
 		log->end_pos += n - 1; /* don't count terminating '\0' */
+
+		if (log->ubuf && new_n &&
+		    copy_to_user(log->ubuf + cur_pos, log->kbuf, new_n))
+			goto fail;
 	} else {
-		u64 new_end, new_start, cur_pos;
+		u64 new_end, new_start;
 		u32 buf_start, buf_end, new_n;
 
-		n += 1;
-
 		new_end = log->end_pos + n;
 		if (new_end - log->start_pos >= log->len_total)
 			new_start = new_end - log->len_total;
@@ -65,6 +87,12 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 		if (buf_end == 0)
 			buf_end = log->len_total;
 
+		log->start_pos = new_start;
+		log->end_pos = new_end - 1; /* don't count terminating '\0' */
+
+		if (!log->ubuf)
+			return;
+
 		/* if buf_start > buf_end, we wrapped around;
 		 * if buf_start == buf_end, then we fill ubuf completely; we
 		 * can't have buf_start == buf_end to mean that there is
@@ -88,9 +116,6 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 					 buf_end))
 				goto fail;
 		}
-
-		log->start_pos = new_start;
-		log->end_pos = new_end - 1; /* don't count terminating '\0' */
 	}
 
 	return;
@@ -116,8 +141,13 @@ void bpf_vlog_reset(struct bpf_verifier_log *log, u64 new_pos)
 	log->end_pos = new_pos;
 	if (log->end_pos < log->start_pos)
 		log->start_pos = log->end_pos;
-	div_u64_rem(new_pos, log->len_total, &pos);
-	if (put_user(zero, log->ubuf + pos))
+
+	if (log->level & BPF_LOG_FIXED)
+		pos = log->end_pos + 1;
+	else
+		div_u64_rem(new_pos, log->len_total, &pos);
+
+	if (log->ubuf && pos < log->len_total && put_user(zero, log->ubuf + pos))
 		log->ubuf = NULL;
 }
 
@@ -169,12 +199,7 @@ static int bpf_vlog_reverse_ubuf(struct bpf_verifier_log *log, int start, int en
 
 bool bpf_vlog_truncated(const struct bpf_verifier_log *log)
 {
-	if (!log->level)
-		return false;
-	else if (log->level & BPF_LOG_FIXED)
-		return bpf_log_used(log) >= log->len_total - 1;
-	else
-		return log->start_pos > 0;
+	return log->len_max > log->len_total;
 }
 
 void bpf_vlog_finalize(struct bpf_verifier_log *log)
-- 
cgit v1.2.3-70-g09d2


From 47a71c1f9af0a334c9dfa97633c41de4feda4287 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:58 -0700
Subject: bpf: Add log_true_size output field to return necessary log buffer
 size

Add output-only log_true_size and btf_log_true_size field to
BPF_PROG_LOAD and BPF_BTF_LOAD commands, respectively. It will return
the size of log buffer necessary to fit in all the log contents at
specified log_level. This is very useful for BPF loader libraries like
libbpf to be able to size log buffer correctly, but could be used by
users directly, if necessary, as well.

This patch plumbs all this through the code, taking into account actual
bpf_attr size provided by user to determine if these new fields are
expected by users. And if they are, set them from kernel on return.

We refactory btf_parse() function to accommodate this, moving attr and
uattr handling inside it. The rest is very straightforward code, which
is split from the logging accounting changes in the previous patch to
make it simpler to review logic vs UAPI changes.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-13-andrii@kernel.org
---
 include/linux/bpf.h            |  2 +-
 include/linux/btf.h            |  2 +-
 include/uapi/linux/bpf.h       | 10 ++++++++++
 kernel/bpf/btf.c               | 32 ++++++++++++++++++--------------
 kernel/bpf/syscall.c           | 16 ++++++++--------
 kernel/bpf/verifier.c          |  8 +++++++-
 tools/include/uapi/linux/bpf.h | 12 +++++++++++-
 7 files changed, 56 insertions(+), 26 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 002a811b6b90..2c6095bd7d69 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2175,7 +2175,7 @@ int bpf_check_uarg_tail_zero(bpfptr_t uaddr, size_t expected_size,
 			     size_t actual_size);
 
 /* verify correctness of eBPF program */
-int bpf_check(struct bpf_prog **fp, union bpf_attr *attr, bpfptr_t uattr);
+int bpf_check(struct bpf_prog **fp, union bpf_attr *attr, bpfptr_t uattr, u32 uattr_size);
 
 #ifndef CONFIG_BPF_JIT_ALWAYS_ON
 void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
diff --git a/include/linux/btf.h b/include/linux/btf.h
index d53b10cc55f2..495250162422 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -125,7 +125,7 @@ extern const struct file_operations btf_fops;
 
 void btf_get(struct btf *btf);
 void btf_put(struct btf *btf);
-int btf_new_fd(const union bpf_attr *attr, bpfptr_t uattr);
+int btf_new_fd(const union bpf_attr *attr, bpfptr_t uattr, u32 uattr_sz);
 struct btf *btf_get_by_fd(int fd);
 int btf_get_info_by_fd(const struct btf *btf,
 		       const union bpf_attr *attr,
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index e3d3b5160d26..3823100b7934 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1407,6 +1407,11 @@ union bpf_attr {
 		__aligned_u64	fd_array;	/* array of FDs */
 		__aligned_u64	core_relos;
 		__u32		core_relo_rec_size; /* sizeof(struct bpf_core_relo) */
+		/* output: actual total log contents size (including termintaing zero).
+		 * It could be both larger than original log_size (if log was
+		 * truncated), or smaller (if log buffer wasn't filled completely).
+		 */
+		__u32		log_true_size;
 	};
 
 	struct { /* anonymous struct used by BPF_OBJ_* commands */
@@ -1492,6 +1497,11 @@ union bpf_attr {
 		__u32		btf_size;
 		__u32		btf_log_size;
 		__u32		btf_log_level;
+		/* output: actual total log contents size (including termintaing zero).
+		 * It could be both larger than original log_size (if log was
+		 * truncated), or smaller (if log buffer wasn't filled completely).
+		 */
+		__u32		btf_log_true_size;
 	};
 
 	struct {
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 5aa540ee611f..0748cf4b8ab6 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5504,9 +5504,10 @@ static int btf_check_type_tags(struct btf_verifier_env *env,
 	return 0;
 }
 
-static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
-			     u32 log_level, char __user *log_ubuf, u32 log_size)
+static struct btf *btf_parse(const union bpf_attr *attr, bpfptr_t uattr, u32 uattr_size)
 {
+	bpfptr_t btf_data = make_bpfptr(attr->btf, uattr.is_kernel);
+	char __user *log_ubuf = u64_to_user_ptr(attr->btf_log_buf);
 	struct btf_struct_metas *struct_meta_tab;
 	struct btf_verifier_env *env = NULL;
 	struct bpf_verifier_log *log;
@@ -5514,7 +5515,7 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 	u8 *data;
 	int err;
 
-	if (btf_data_size > BTF_MAX_SIZE)
+	if (attr->btf_size > BTF_MAX_SIZE)
 		return ERR_PTR(-E2BIG);
 
 	env = kzalloc(sizeof(*env), GFP_KERNEL | __GFP_NOWARN);
@@ -5522,13 +5523,13 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 		return ERR_PTR(-ENOMEM);
 
 	log = &env->log;
-	if (log_level || log_ubuf || log_size) {
+	if (attr->btf_log_level || log_ubuf || attr->btf_log_size) {
 		/* user requested verbose verifier output
 		 * and supplied buffer to store the verification trace
 		 */
-		log->level = log_level;
+		log->level = attr->btf_log_level;
 		log->ubuf = log_ubuf;
-		log->len_total = log_size;
+		log->len_total = attr->btf_log_size;
 
 		/* log attributes have to be sane */
 		if (!bpf_verifier_log_attr_valid(log)) {
@@ -5544,16 +5545,16 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 	}
 	env->btf = btf;
 
-	data = kvmalloc(btf_data_size, GFP_KERNEL | __GFP_NOWARN);
+	data = kvmalloc(attr->btf_size, GFP_KERNEL | __GFP_NOWARN);
 	if (!data) {
 		err = -ENOMEM;
 		goto errout;
 	}
 
 	btf->data = data;
-	btf->data_size = btf_data_size;
+	btf->data_size = attr->btf_size;
 
-	if (copy_from_bpfptr(data, btf_data, btf_data_size)) {
+	if (copy_from_bpfptr(data, btf_data, attr->btf_size)) {
 		err = -EFAULT;
 		goto errout;
 	}
@@ -5594,6 +5595,12 @@ static struct btf *btf_parse(bpfptr_t btf_data, u32 btf_data_size,
 	}
 
 	bpf_vlog_finalize(log);
+	if (uattr_size >= offsetofend(union bpf_attr, btf_log_true_size) &&
+	    copy_to_bpfptr_offset(uattr, offsetof(union bpf_attr, btf_log_true_size),
+				  &log->len_max, sizeof(log->len_max))) {
+		err = -EFAULT;
+		goto errout_meta;
+	}
 	if (bpf_vlog_truncated(log)) {
 		err = -ENOSPC;
 		goto errout_meta;
@@ -7218,15 +7225,12 @@ static int __btf_new_fd(struct btf *btf)
 	return anon_inode_getfd("btf", &btf_fops, btf, O_RDONLY | O_CLOEXEC);
 }
 
-int btf_new_fd(const union bpf_attr *attr, bpfptr_t uattr)
+int btf_new_fd(const union bpf_attr *attr, bpfptr_t uattr, u32 uattr_size)
 {
 	struct btf *btf;
 	int ret;
 
-	btf = btf_parse(make_bpfptr(attr->btf, uattr.is_kernel),
-			attr->btf_size, attr->btf_log_level,
-			u64_to_user_ptr(attr->btf_log_buf),
-			attr->btf_log_size);
+	btf = btf_parse(attr, uattr, uattr_size);
 	if (IS_ERR(btf))
 		return PTR_ERR(btf);
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index e18ac7fdc210..6d575505f89c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2501,9 +2501,9 @@ static bool is_perfmon_prog_type(enum bpf_prog_type prog_type)
 }
 
 /* last field in 'union bpf_attr' used by this command */
-#define	BPF_PROG_LOAD_LAST_FIELD core_relo_rec_size
+#define	BPF_PROG_LOAD_LAST_FIELD log_true_size
 
-static int bpf_prog_load(union bpf_attr *attr, bpfptr_t uattr)
+static int bpf_prog_load(union bpf_attr *attr, bpfptr_t uattr, u32 uattr_size)
 {
 	enum bpf_prog_type type = attr->prog_type;
 	struct bpf_prog *prog, *dst_prog = NULL;
@@ -2653,7 +2653,7 @@ static int bpf_prog_load(union bpf_attr *attr, bpfptr_t uattr)
 		goto free_prog_sec;
 
 	/* run eBPF verifier */
-	err = bpf_check(&prog, attr, uattr);
+	err = bpf_check(&prog, attr, uattr, uattr_size);
 	if (err < 0)
 		goto free_used_maps;
 
@@ -4371,9 +4371,9 @@ static int bpf_obj_get_info_by_fd(const union bpf_attr *attr,
 	return err;
 }
 
-#define BPF_BTF_LOAD_LAST_FIELD btf_log_level
+#define BPF_BTF_LOAD_LAST_FIELD btf_log_true_size
 
-static int bpf_btf_load(const union bpf_attr *attr, bpfptr_t uattr)
+static int bpf_btf_load(const union bpf_attr *attr, bpfptr_t uattr, __u32 uattr_size)
 {
 	if (CHECK_ATTR(BPF_BTF_LOAD))
 		return -EINVAL;
@@ -4381,7 +4381,7 @@ static int bpf_btf_load(const union bpf_attr *attr, bpfptr_t uattr)
 	if (!bpf_capable())
 		return -EPERM;
 
-	return btf_new_fd(attr, uattr);
+	return btf_new_fd(attr, uattr, uattr_size);
 }
 
 #define BPF_BTF_GET_FD_BY_ID_LAST_FIELD btf_id
@@ -5059,7 +5059,7 @@ static int __sys_bpf(int cmd, bpfptr_t uattr, unsigned int size)
 		err = map_freeze(&attr);
 		break;
 	case BPF_PROG_LOAD:
-		err = bpf_prog_load(&attr, uattr);
+		err = bpf_prog_load(&attr, uattr, size);
 		break;
 	case BPF_OBJ_PIN:
 		err = bpf_obj_pin(&attr);
@@ -5104,7 +5104,7 @@ static int __sys_bpf(int cmd, bpfptr_t uattr, unsigned int size)
 		err = bpf_raw_tracepoint_open(&attr);
 		break;
 	case BPF_BTF_LOAD:
-		err = bpf_btf_load(&attr, uattr);
+		err = bpf_btf_load(&attr, uattr, size);
 		break;
 	case BPF_BTF_GET_FD_BY_ID:
 		err = bpf_btf_get_fd_by_id(&attr);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index a98cbc046d1e..308e7abeb979 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18694,7 +18694,7 @@ struct btf *bpf_get_btf_vmlinux(void)
 	return btf_vmlinux;
 }
 
-int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, bpfptr_t uattr)
+int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, bpfptr_t uattr, __u32 uattr_size)
 {
 	u64 start_time = ktime_get_ns();
 	struct bpf_verifier_env *env;
@@ -18861,6 +18861,12 @@ skip_full_check:
 	env->prog->aux->verified_insns = env->insn_processed;
 
 	bpf_vlog_finalize(log);
+	if (uattr_size >= offsetofend(union bpf_attr, log_true_size) &&
+	    copy_to_bpfptr_offset(uattr, offsetof(union bpf_attr, log_true_size),
+				  &log->len_max, sizeof(log->len_max))) {
+		ret = -EFAULT;
+		goto err_release_maps;
+	}
 	if (bpf_vlog_truncated(log))
 		ret = -ENOSPC;
 	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf)
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index d6c5a022ae28..3823100b7934 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1407,6 +1407,11 @@ union bpf_attr {
 		__aligned_u64	fd_array;	/* array of FDs */
 		__aligned_u64	core_relos;
 		__u32		core_relo_rec_size; /* sizeof(struct bpf_core_relo) */
+		/* output: actual total log contents size (including termintaing zero).
+		 * It could be both larger than original log_size (if log was
+		 * truncated), or smaller (if log buffer wasn't filled completely).
+		 */
+		__u32		log_true_size;
 	};
 
 	struct { /* anonymous struct used by BPF_OBJ_* commands */
@@ -1492,6 +1497,11 @@ union bpf_attr {
 		__u32		btf_size;
 		__u32		btf_log_size;
 		__u32		btf_log_level;
+		/* output: actual total log contents size (including termintaing zero).
+		 * It could be both larger than original log_size (if log was
+		 * truncated), or smaller (if log buffer wasn't filled completely).
+		 */
+		__u32		btf_log_true_size;
 	};
 
 	struct {
@@ -1513,7 +1523,7 @@ union bpf_attr {
 	struct { /* struct used by BPF_LINK_CREATE command */
 		union {
 			__u32		prog_fd;	/* eBPF program to attach */
-			__u32		map_fd;		/* eBPF struct_ops to attach */
+			__u32		map_fd;		/* struct_ops to attach */
 		};
 		union {
 			__u32		target_fd;	/* object to attach to */
-- 
cgit v1.2.3-70-g09d2


From bdcab4144f5da97cc0fa7e1dd63b8475e10c8f0a Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:41:59 -0700
Subject: bpf: Simplify internal verifier log interface

Simplify internal verifier log API down to bpf_vlog_init() and
bpf_vlog_finalize(). The former handles input arguments validation in
one place and makes it easier to change it. The latter subsumes -ENOSPC
(truncation) and -EFAULT handling and simplifies both caller's code
(bpf_check() and btf_parse()).

For btf_parse(), this patch also makes sure that verifier log
finalization happens even if there is some error condition during BTF
verification process prior to normal finalization step.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-14-andrii@kernel.org
---
 include/linux/bpf_verifier.h | 13 ++-------
 kernel/bpf/btf.c             | 65 ++++++++++++++++++++++----------------------
 kernel/bpf/log.c             | 48 ++++++++++++++++++++++++++------
 kernel/bpf/verifier.c        | 39 +++++++++++---------------
 4 files changed, 90 insertions(+), 75 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 98d2eb382dbb..f03852b89d28 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -518,13 +518,6 @@ struct bpf_verifier_log {
 #define BPF_LOG_MIN_ALIGNMENT 8U
 #define BPF_LOG_ALIGNMENT 40U
 
-static inline bool bpf_verifier_log_full(const struct bpf_verifier_log *log)
-{
-	if (log->level & BPF_LOG_FIXED)
-		return log->end_pos >= log->len_total;
-	return false;
-}
-
 static inline bool bpf_verifier_log_needed(const struct bpf_verifier_log *log)
 {
 	return log && log->level;
@@ -612,16 +605,16 @@ struct bpf_verifier_env {
 	char type_str_buf[TYPE_STR_BUF_LEN];
 };
 
-bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log);
 __printf(2, 0) void bpf_verifier_vlog(struct bpf_verifier_log *log,
 				      const char *fmt, va_list args);
 __printf(2, 3) void bpf_verifier_log_write(struct bpf_verifier_env *env,
 					   const char *fmt, ...);
 __printf(2, 3) void bpf_log(struct bpf_verifier_log *log,
 			    const char *fmt, ...);
+int bpf_vlog_init(struct bpf_verifier_log *log, u32 log_level,
+		  char __user *log_buf, u32 log_size);
 void bpf_vlog_reset(struct bpf_verifier_log *log, u64 new_pos);
-void bpf_vlog_finalize(struct bpf_verifier_log *log);
-bool bpf_vlog_truncated(const struct bpf_verifier_log *log);
+int bpf_vlog_finalize(struct bpf_verifier_log *log, u32 *log_size_actual);
 
 static inline struct bpf_func_state *cur_func(struct bpf_verifier_env *env)
 {
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 0748cf4b8ab6..ffc31a1c84af 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5504,16 +5504,30 @@ static int btf_check_type_tags(struct btf_verifier_env *env,
 	return 0;
 }
 
+static int finalize_log(struct bpf_verifier_log *log, bpfptr_t uattr, u32 uattr_size)
+{
+	u32 log_true_size;
+	int err;
+
+	err = bpf_vlog_finalize(log, &log_true_size);
+
+	if (uattr_size >= offsetofend(union bpf_attr, btf_log_true_size) &&
+	    copy_to_bpfptr_offset(uattr, offsetof(union bpf_attr, btf_log_true_size),
+				  &log_true_size, sizeof(log_true_size)))
+		err = -EFAULT;
+
+	return err;
+}
+
 static struct btf *btf_parse(const union bpf_attr *attr, bpfptr_t uattr, u32 uattr_size)
 {
 	bpfptr_t btf_data = make_bpfptr(attr->btf, uattr.is_kernel);
 	char __user *log_ubuf = u64_to_user_ptr(attr->btf_log_buf);
 	struct btf_struct_metas *struct_meta_tab;
 	struct btf_verifier_env *env = NULL;
-	struct bpf_verifier_log *log;
 	struct btf *btf = NULL;
 	u8 *data;
-	int err;
+	int err, ret;
 
 	if (attr->btf_size > BTF_MAX_SIZE)
 		return ERR_PTR(-E2BIG);
@@ -5522,21 +5536,13 @@ static struct btf *btf_parse(const union bpf_attr *attr, bpfptr_t uattr, u32 uat
 	if (!env)
 		return ERR_PTR(-ENOMEM);
 
-	log = &env->log;
-	if (attr->btf_log_level || log_ubuf || attr->btf_log_size) {
-		/* user requested verbose verifier output
-		 * and supplied buffer to store the verification trace
-		 */
-		log->level = attr->btf_log_level;
-		log->ubuf = log_ubuf;
-		log->len_total = attr->btf_log_size;
-
-		/* log attributes have to be sane */
-		if (!bpf_verifier_log_attr_valid(log)) {
-			err = -EINVAL;
-			goto errout;
-		}
-	}
+	/* user could have requested verbose verifier output
+	 * and supplied buffer to store the verification trace
+	 */
+	err = bpf_vlog_init(&env->log, attr->btf_log_level,
+			    log_ubuf, attr->btf_log_size);
+	if (err)
+		goto errout_free;
 
 	btf = kzalloc(sizeof(*btf), GFP_KERNEL | __GFP_NOWARN);
 	if (!btf) {
@@ -5577,7 +5583,7 @@ static struct btf *btf_parse(const union bpf_attr *attr, bpfptr_t uattr, u32 uat
 	if (err)
 		goto errout;
 
-	struct_meta_tab = btf_parse_struct_metas(log, btf);
+	struct_meta_tab = btf_parse_struct_metas(&env->log, btf);
 	if (IS_ERR(struct_meta_tab)) {
 		err = PTR_ERR(struct_meta_tab);
 		goto errout;
@@ -5594,21 +5600,9 @@ static struct btf *btf_parse(const union bpf_attr *attr, bpfptr_t uattr, u32 uat
 		}
 	}
 
-	bpf_vlog_finalize(log);
-	if (uattr_size >= offsetofend(union bpf_attr, btf_log_true_size) &&
-	    copy_to_bpfptr_offset(uattr, offsetof(union bpf_attr, btf_log_true_size),
-				  &log->len_max, sizeof(log->len_max))) {
-		err = -EFAULT;
-		goto errout_meta;
-	}
-	if (bpf_vlog_truncated(log)) {
-		err = -ENOSPC;
-		goto errout_meta;
-	}
-	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf) {
-		err = -EFAULT;
-		goto errout_meta;
-	}
+	err = finalize_log(&env->log, uattr, uattr_size);
+	if (err)
+		goto errout_free;
 
 	btf_verifier_env_free(env);
 	refcount_set(&btf->refcnt, 1);
@@ -5617,6 +5611,11 @@ static struct btf *btf_parse(const union bpf_attr *attr, bpfptr_t uattr, u32 uat
 errout_meta:
 	btf_free_struct_meta_tab(btf);
 errout:
+	/* overwrite err with -ENOSPC or -EFAULT */
+	ret = finalize_log(&env->log, uattr, uattr_size);
+	if (ret)
+		err = ret;
+errout_free:
 	btf_verifier_env_free(env);
 	if (btf)
 		btf_free(btf);
diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 47bea2fad6fe..1fae2c5d7ae4 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -10,12 +10,26 @@
 #include <linux/bpf_verifier.h>
 #include <linux/math64.h>
 
-bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
+static bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
 {
 	return log->len_total > 0 && log->len_total <= UINT_MAX >> 2 &&
 	       log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
 }
 
+int bpf_vlog_init(struct bpf_verifier_log *log, u32 log_level,
+		  char __user *log_buf, u32 log_size)
+{
+	log->level = log_level;
+	log->ubuf = log_buf;
+	log->len_total = log_size;
+
+	/* log attributes have to be sane */
+	if (!bpf_verifier_log_attr_valid(log))
+		return -EINVAL;
+
+	return 0;
+}
+
 static void bpf_vlog_update_len_max(struct bpf_verifier_log *log, u32 add_len)
 {
 	/* add_len includes terminal \0, so no need for +1. */
@@ -197,24 +211,25 @@ static int bpf_vlog_reverse_ubuf(struct bpf_verifier_log *log, int start, int en
 	return 0;
 }
 
-bool bpf_vlog_truncated(const struct bpf_verifier_log *log)
+static bool bpf_vlog_truncated(const struct bpf_verifier_log *log)
 {
 	return log->len_max > log->len_total;
 }
 
-void bpf_vlog_finalize(struct bpf_verifier_log *log)
+int bpf_vlog_finalize(struct bpf_verifier_log *log, u32 *log_size_actual)
 {
 	u32 sublen;
 	int err;
 
-	if (!log || !log->level || !log->ubuf)
-		return;
-	if ((log->level & BPF_LOG_FIXED) || log->level == BPF_LOG_KERNEL)
-		return;
+	*log_size_actual = 0;
+	if (!log || log->level == 0 || log->level == BPF_LOG_KERNEL)
+		return 0;
 
+	if (!log->ubuf)
+		goto skip_log_rotate;
 	/* If we never truncated log, there is nothing to move around. */
-	if (log->start_pos == 0)
-		return;
+	if ((log->level & BPF_LOG_FIXED) || log->start_pos == 0)
+		goto skip_log_rotate;
 
 	/* Otherwise we need to rotate log contents to make it start from the
 	 * buffer beginning and be a continuous zero-terminated string. Note
@@ -257,6 +272,21 @@ void bpf_vlog_finalize(struct bpf_verifier_log *log)
 	err = err ?: bpf_vlog_reverse_ubuf(log, sublen, log->len_total);
 	if (err)
 		log->ubuf = NULL;
+
+skip_log_rotate:
+	*log_size_actual = log->len_max;
+
+	/* properly initialized log has either both ubuf!=NULL and len_total>0
+	 * or ubuf==NULL and len_total==0, so if this condition doesn't hold,
+	 * we got a fault somewhere along the way, so report it back
+	 */
+	if (!!log->ubuf != !!log->len_total)
+		return -EFAULT;
+
+	if (bpf_vlog_truncated(log))
+		return -ENOSPC;
+
+	return 0;
 }
 
 /* log_level controls verbosity level of eBPF verifier.
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 308e7abeb979..d6db6de3e9ea 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18698,8 +18698,8 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, bpfptr_t uattr, __u3
 {
 	u64 start_time = ktime_get_ns();
 	struct bpf_verifier_env *env;
-	struct bpf_verifier_log *log;
-	int i, len, ret = -EINVAL;
+	int i, len, ret = -EINVAL, err;
+	u32 log_true_size;
 	bool is_priv;
 
 	/* no program is valid */
@@ -18712,7 +18712,6 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, bpfptr_t uattr, __u3
 	env = kzalloc(sizeof(struct bpf_verifier_env), GFP_KERNEL);
 	if (!env)
 		return -ENOMEM;
-	log = &env->log;
 
 	len = (*prog)->len;
 	env->insn_aux_data =
@@ -18733,20 +18732,14 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, bpfptr_t uattr, __u3
 	if (!is_priv)
 		mutex_lock(&bpf_verifier_lock);
 
-	if (attr->log_level || attr->log_buf || attr->log_size) {
-		/* user requested verbose verifier output
-		 * and supplied buffer to store the verification trace
-		 */
-		log->level = attr->log_level;
-		log->ubuf = (char __user *) (unsigned long) attr->log_buf;
-		log->len_total = attr->log_size;
-
-		/* log attributes have to be sane */
-		if (!bpf_verifier_log_attr_valid(log)) {
-			ret = -EINVAL;
-			goto err_unlock;
-		}
-	}
+	/* user could have requested verbose verifier output
+	 * and supplied buffer to store the verification trace
+	 */
+	ret = bpf_vlog_init(&env->log, attr->log_level,
+			    (char __user *) (unsigned long) attr->log_buf,
+			    attr->log_size);
+	if (ret)
+		goto err_unlock;
 
 	mark_verifier_state_clean(env);
 
@@ -18860,17 +18853,17 @@ skip_full_check:
 	print_verification_stats(env);
 	env->prog->aux->verified_insns = env->insn_processed;
 
-	bpf_vlog_finalize(log);
+	/* preserve original error even if log finalization is successful */
+	err = bpf_vlog_finalize(&env->log, &log_true_size);
+	if (err)
+		ret = err;
+
 	if (uattr_size >= offsetofend(union bpf_attr, log_true_size) &&
 	    copy_to_bpfptr_offset(uattr, offsetof(union bpf_attr, log_true_size),
-				  &log->len_max, sizeof(log->len_max))) {
+				  &log_true_size, sizeof(log_true_size))) {
 		ret = -EFAULT;
 		goto err_release_maps;
 	}
-	if (bpf_vlog_truncated(log))
-		ret = -ENOSPC;
-	if (log->level && log->level != BPF_LOG_KERNEL && !log->ubuf)
-		ret = -EFAULT;
 
 	if (ret)
 		goto err_release_maps;
-- 
cgit v1.2.3-70-g09d2


From fac08d45e2531f91d8fb3d11fc6576f588049476 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:42:00 -0700
Subject: bpf: Relax log_buf NULL conditions when log_level>0 is requested

Drop the log_size>0 and log_buf!=NULL condition when log_level>0. This
allows users to request log_true_size of a full log without providing
actual (even if small) log buffer. Verifier log handling code was mostly
ready to handle NULL log->ubuf, so only few small changes were necessary
to prevent NULL log->ubuf from causing problems.

Note, that if user provided NULL log_buf with log_level>0 we don't
consider this a log truncation, and thus won't return -ENOSPC.

We also enforce that either (log_buf==NULL && log_size==0) or
(log_buf!=NULL && log_size>0).

Suggested-by: Lorenz Bauer <lmb@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-15-andrii@kernel.org
---
 kernel/bpf/log.c | 42 +++++++++++++++++++++++++-----------------
 1 file changed, 25 insertions(+), 17 deletions(-)

diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 1fae2c5d7ae4..046ddff37a76 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -12,8 +12,17 @@
 
 static bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
 {
-	return log->len_total > 0 && log->len_total <= UINT_MAX >> 2 &&
-	       log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
+	/* ubuf and len_total should both be specified (or not) together */
+	if (!!log->ubuf != !!log->len_total)
+		return false;
+	/* log buf without log_level is meaningless */
+	if (log->ubuf && log->level == 0)
+		return false;
+	if (log->level & ~BPF_LOG_MASK)
+		return false;
+	if (log->len_total > UINT_MAX >> 2)
+		return false;
+	return true;
 }
 
 int bpf_vlog_init(struct bpf_verifier_log *log, u32 log_level,
@@ -89,9 +98,15 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 			new_start = new_end - log->len_total;
 		else
 			new_start = log->start_pos;
+
+		log->start_pos = new_start;
+		log->end_pos = new_end - 1; /* don't count terminating '\0' */
+
+		if (!log->ubuf)
+			return;
+
 		new_n = min(n, log->len_total);
 		cur_pos = new_end - new_n;
-
 		div_u64_rem(cur_pos, log->len_total, &buf_start);
 		div_u64_rem(new_end, log->len_total, &buf_end);
 		/* new_end and buf_end are exclusive indices, so if buf_end is
@@ -101,12 +116,6 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 		if (buf_end == 0)
 			buf_end = log->len_total;
 
-		log->start_pos = new_start;
-		log->end_pos = new_end - 1; /* don't count terminating '\0' */
-
-		if (!log->ubuf)
-			return;
-
 		/* if buf_start > buf_end, we wrapped around;
 		 * if buf_start == buf_end, then we fill ubuf completely; we
 		 * can't have buf_start == buf_end to mean that there is
@@ -156,12 +165,15 @@ void bpf_vlog_reset(struct bpf_verifier_log *log, u64 new_pos)
 	if (log->end_pos < log->start_pos)
 		log->start_pos = log->end_pos;
 
+	if (!log->ubuf)
+		return;
+
 	if (log->level & BPF_LOG_FIXED)
 		pos = log->end_pos + 1;
 	else
 		div_u64_rem(new_pos, log->len_total, &pos);
 
-	if (log->ubuf && pos < log->len_total && put_user(zero, log->ubuf + pos))
+	if (pos < log->len_total && put_user(zero, log->ubuf + pos))
 		log->ubuf = NULL;
 }
 
@@ -211,11 +223,6 @@ static int bpf_vlog_reverse_ubuf(struct bpf_verifier_log *log, int start, int en
 	return 0;
 }
 
-static bool bpf_vlog_truncated(const struct bpf_verifier_log *log)
-{
-	return log->len_max > log->len_total;
-}
-
 int bpf_vlog_finalize(struct bpf_verifier_log *log, u32 *log_size_actual)
 {
 	u32 sublen;
@@ -228,7 +235,7 @@ int bpf_vlog_finalize(struct bpf_verifier_log *log, u32 *log_size_actual)
 	if (!log->ubuf)
 		goto skip_log_rotate;
 	/* If we never truncated log, there is nothing to move around. */
-	if ((log->level & BPF_LOG_FIXED) || log->start_pos == 0)
+	if (log->start_pos == 0)
 		goto skip_log_rotate;
 
 	/* Otherwise we need to rotate log contents to make it start from the
@@ -283,7 +290,8 @@ skip_log_rotate:
 	if (!!log->ubuf != !!log->len_total)
 		return -EFAULT;
 
-	if (bpf_vlog_truncated(log))
+	/* did truncation actually happen? */
+	if (log->ubuf && log->len_max > log->len_total)
 		return -ENOSPC;
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 94e55c0fdaf4268bdda2d3b375bc61daba056e85 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:42:01 -0700
Subject: libbpf: Wire through log_true_size returned from kernel for
 BPF_PROG_LOAD

Add output-only log_true_size field to bpf_prog_load_opts to return
bpf_attr->log_true_size value back from bpf() syscall.

Note, that we have to drop const modifier from opts in bpf_prog_load().
This could potentially cause compilation error for some users. But
the usual practice is to define bpf_prog_load_ops
as a local variable next to bpf_prog_load() call and pass pointer to it,
so const vs non-const makes no difference and won't even come up in most
(if not all) cases.

There are no runtime and ABI backwards/forward compatibility issues at all.
If user provides old struct bpf_prog_load_opts, libbpf won't set new
fields. If old libbpf is provided new bpf_prog_load_opts, nothing will
happen either as old libbpf doesn't yet know about this new field.

Adding a new variant of bpf_prog_load() just for this seems like a big
and unnecessary overkill. As a corroborating evidence is the fact that
entire selftests/bpf code base required not adjustment whatsoever.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230406234205.323208-16-andrii@kernel.org
---
 tools/lib/bpf/bpf.c |  7 +++++--
 tools/lib/bpf/bpf.h | 11 +++++++++--
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index f1d04ee14d83..a031de0a707a 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -230,9 +230,9 @@ alloc_zero_tailing_info(const void *orecord, __u32 cnt,
 int bpf_prog_load(enum bpf_prog_type prog_type,
 		  const char *prog_name, const char *license,
 		  const struct bpf_insn *insns, size_t insn_cnt,
-		  const struct bpf_prog_load_opts *opts)
+		  struct bpf_prog_load_opts *opts)
 {
-	const size_t attr_sz = offsetofend(union bpf_attr, fd_array);
+	const size_t attr_sz = offsetofend(union bpf_attr, log_true_size);
 	void *finfo = NULL, *linfo = NULL;
 	const char *func_info, *line_info;
 	__u32 log_size, log_level, attach_prog_fd, attach_btf_obj_fd;
@@ -312,6 +312,7 @@ int bpf_prog_load(enum bpf_prog_type prog_type,
 	}
 
 	fd = sys_bpf_prog_load(&attr, attr_sz, attempts);
+	OPTS_SET(opts, log_true_size, attr.log_true_size);
 	if (fd >= 0)
 		return fd;
 
@@ -352,6 +353,7 @@ int bpf_prog_load(enum bpf_prog_type prog_type,
 		}
 
 		fd = sys_bpf_prog_load(&attr, attr_sz, attempts);
+		OPTS_SET(opts, log_true_size, attr.log_true_size);
 		if (fd >= 0)
 			goto done;
 	}
@@ -366,6 +368,7 @@ int bpf_prog_load(enum bpf_prog_type prog_type,
 		attr.log_level = 1;
 
 		fd = sys_bpf_prog_load(&attr, attr_sz, attempts);
+		OPTS_SET(opts, log_true_size, attr.log_true_size);
 	}
 done:
 	/* free() doesn't affect errno, so we don't need to restore it */
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index b073e73439ef..cfa0cdfa50f8 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -96,13 +96,20 @@ struct bpf_prog_load_opts {
 	__u32 log_level;
 	__u32 log_size;
 	char *log_buf;
+	/* output: actual total log contents size (including termintaing zero).
+	 * It could be both larger than original log_size (if log was
+	 * truncated), or smaller (if log buffer wasn't filled completely).
+	 * If kernel doesn't support this feature, log_size is left unchanged.
+	 */
+	__u32 log_true_size;
+	size_t :0;
 };
-#define bpf_prog_load_opts__last_field log_buf
+#define bpf_prog_load_opts__last_field log_true_size
 
 LIBBPF_API int bpf_prog_load(enum bpf_prog_type prog_type,
 			     const char *prog_name, const char *license,
 			     const struct bpf_insn *insns, size_t insn_cnt,
-			     const struct bpf_prog_load_opts *opts);
+			     struct bpf_prog_load_opts *opts);
 
 /* Flags to direct loading requirements */
 #define MAPS_RELAX_COMPAT	0x01
-- 
cgit v1.2.3-70-g09d2


From 097d8002b754a865beef880e5c1cdc3ef7c2163d Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:42:02 -0700
Subject: libbpf: Wire through log_true_size for bpf_btf_load() API

Similar to what we did for bpf_prog_load() in previous patch, wire
returning of log_true_size value from kernel back to the user through
OPTS out field.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230406234205.323208-17-andrii@kernel.org
---
 tools/lib/bpf/bpf.c |  6 ++++--
 tools/lib/bpf/bpf.h | 11 +++++++++--
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index a031de0a707a..128ac723c4ea 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -1083,9 +1083,9 @@ int bpf_raw_tracepoint_open(const char *name, int prog_fd)
 	return libbpf_err_errno(fd);
 }
 
-int bpf_btf_load(const void *btf_data, size_t btf_size, const struct bpf_btf_load_opts *opts)
+int bpf_btf_load(const void *btf_data, size_t btf_size, struct bpf_btf_load_opts *opts)
 {
-	const size_t attr_sz = offsetofend(union bpf_attr, btf_log_level);
+	const size_t attr_sz = offsetofend(union bpf_attr, btf_log_true_size);
 	union bpf_attr attr;
 	char *log_buf;
 	size_t log_size;
@@ -1128,6 +1128,8 @@ int bpf_btf_load(const void *btf_data, size_t btf_size, const struct bpf_btf_loa
 		attr.btf_log_level = 1;
 		fd = sys_bpf_fd(BPF_BTF_LOAD, &attr, attr_sz);
 	}
+
+	OPTS_SET(opts, log_true_size, attr.btf_log_true_size);
 	return libbpf_err_errno(fd);
 }
 
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index cfa0cdfa50f8..a2c091389b18 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -124,11 +124,18 @@ struct bpf_btf_load_opts {
 	char *log_buf;
 	__u32 log_level;
 	__u32 log_size;
+	/* output: actual total log contents size (including termintaing zero).
+	 * It could be both larger than original log_size (if log was
+	 * truncated), or smaller (if log buffer wasn't filled completely).
+	 * If kernel doesn't support this feature, log_size is left unchanged.
+	 */
+	__u32 log_true_size;
+	size_t :0;
 };
-#define bpf_btf_load_opts__last_field log_size
+#define bpf_btf_load_opts__last_field log_true_size
 
 LIBBPF_API int bpf_btf_load(const void *btf_data, size_t btf_size,
-			    const struct bpf_btf_load_opts *opts);
+			    struct bpf_btf_load_opts *opts);
 
 LIBBPF_API int bpf_map_update_elem(int fd, const void *key, const void *value,
 				   __u64 flags);
-- 
cgit v1.2.3-70-g09d2


From 5787540827a9e2cdecf38166e648b2924a57443f Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:42:03 -0700
Subject: selftests/bpf: Add tests to validate log_true_size feature

Add additional test cases validating that log_true_size is consistent
between fixed and rotating log modes, and that log_true_size can be
used *exactly* without causing -ENOSPC, while using just 1 byte shorter
log buffer would cause -ENOSPC.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-18-andrii@kernel.org
---
 .../selftests/bpf/prog_tests/verifier_log.c        | 92 ++++++++++++++++++----
 1 file changed, 76 insertions(+), 16 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier_log.c b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
index 3284108a6ce8..2ec82fc60c03 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier_log.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
@@ -18,25 +18,41 @@ static bool check_prog_load(int prog_fd, bool expect_err, const char *tag)
 		if (!ASSERT_GT(prog_fd, 0, tag))
 			return false;
 	}
+	if (prog_fd >= 0)
+		close(prog_fd);
 	return true;
 }
 
+static struct {
+	/* strategically placed before others to avoid accidental modification by kernel */
+	char filler[1024];
+	char buf[1024];
+	/* strategically placed after buf[] to catch more accidental corruptions */
+	char reference[1024];
+} logs;
+static const struct bpf_insn *insns;
+static size_t insn_cnt;
+
+static int load_prog(struct bpf_prog_load_opts *opts, bool expect_load_error)
+{
+	int prog_fd;
+
+	prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, "log_prog",
+				"GPL", insns, insn_cnt, opts);
+	check_prog_load(prog_fd, expect_load_error, "prog_load");
+
+	return prog_fd;
+}
+
 static void verif_log_subtest(const char *name, bool expect_load_error, int log_level)
 {
 	LIBBPF_OPTS(bpf_prog_load_opts, opts);
-	struct {
-		/* strategically placed before others to avoid accidental modification by kernel */
-		char filler[1024];
-		char buf[1024];
-		/* strategically placed after buf[] to catch more accidental corruptions */
-		char reference[1024];
-	} logs;
 	char *exp_log, prog_name[16], op_name[32];
 	struct test_log_buf *skel;
 	struct bpf_program *prog;
-	const struct bpf_insn *insns;
-	size_t insn_cnt, fixed_log_sz;
-	int i, mode, err, prog_fd;
+	size_t fixed_log_sz;
+	__u32 log_true_sz_fixed, log_true_sz_rolling;
+	int i, mode, err, prog_fd, res;
 
 	skel = test_log_buf__open();
 	if (!ASSERT_OK_PTR(skel, "skel_open"))
@@ -61,11 +77,7 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 	opts.log_buf = logs.reference;
 	opts.log_size = sizeof(logs.reference);
 	opts.log_level = log_level | 8 /* BPF_LOG_FIXED */;
-	prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, "log_fixed",
-				"GPL", insns, insn_cnt, &opts);
-	if (!check_prog_load(prog_fd, expect_load_error, "fixed_buf_prog_load"))
-		goto cleanup;
-	close(prog_fd);
+	load_prog(&opts, expect_load_error);
 
 	fixed_log_sz = strlen(logs.reference) + 1;
 	if (!ASSERT_GT(fixed_log_sz, 50, "fixed_log_sz"))
@@ -89,7 +101,7 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 		opts.log_level = log_level | 8; /* fixed-length log */
 		opts.log_size = 25;
 
-		prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, "log_fixed50",
+		prog_fd = bpf_prog_load(BPF_PROG_TYPE_RAW_TRACEPOINT, "log_fixed25",
 					"GPL", insns, insn_cnt, &opts);
 		if (!ASSERT_EQ(prog_fd, -ENOSPC, "unexpected_log_fixed_prog_load_result")) {
 			if (prog_fd >= 0)
@@ -162,6 +174,54 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 		}
 	}
 
+	/* (FIXED) get actual log size */
+	opts.log_buf = logs.buf;
+	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
+	opts.log_size = sizeof(logs.buf);
+	res = load_prog(&opts, expect_load_error);
+	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_fixed");
+
+	log_true_sz_fixed = opts.log_true_size;
+	ASSERT_GT(log_true_sz_fixed, 0, "log_true_sz_fixed");
+
+	/* (ROLLING) get actual log size */
+	opts.log_buf = logs.buf;
+	opts.log_level = log_level;
+	opts.log_size = sizeof(logs.buf);
+	res = load_prog(&opts, expect_load_error);
+	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_rolling");
+
+	log_true_sz_rolling = opts.log_true_size;
+	ASSERT_EQ(log_true_sz_rolling, log_true_sz_fixed, "log_true_sz_eq");
+
+	/* (FIXED) expect -ENOSPC for one byte short log */
+	opts.log_buf = logs.buf;
+	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
+	opts.log_size = log_true_sz_fixed - 1;
+	res = load_prog(&opts, true /* should fail */);
+	ASSERT_EQ(res, -ENOSPC, "prog_load_res_too_short_fixed");
+
+	/* (FIXED) expect *not* -ENOSPC with exact log_true_size buffer */
+	opts.log_buf = logs.buf;
+	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
+	opts.log_size = log_true_sz_fixed;
+	res = load_prog(&opts, expect_load_error);
+	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_just_right_fixed");
+
+	/* (ROLLING) expect -ENOSPC for one byte short log */
+	opts.log_buf = logs.buf;
+	opts.log_level = log_level;
+	opts.log_size = log_true_sz_rolling - 1;
+	res = load_prog(&opts, true /* should fail */);
+	ASSERT_EQ(res, -ENOSPC, "prog_load_res_too_short_rolling");
+
+	/* (ROLLING) expect *not* -ENOSPC with exact log_true_size buffer */
+	opts.log_buf = logs.buf;
+	opts.log_level = log_level;
+	opts.log_size = log_true_sz_rolling;
+	res = load_prog(&opts, expect_load_error);
+	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_just_right_rolling");
+
 cleanup:
 	test_log_buf__destroy(skel);
 }
-- 
cgit v1.2.3-70-g09d2


From be983f44274f575e42025130e3c62b8718b0a29a Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:42:04 -0700
Subject: selftests/bpf: Add testing of log_buf==NULL condition for
 BPF_PROG_LOAD

Add few extra test conditions to validate that it's ok to pass
log_buf==NULL and log_size==0 to BPF_PROG_LOAD command with the intent
to get log_true_size without providing a buffer.

Test that log_buf==NULL condition *does not* return -ENOSPC.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-19-andrii@kernel.org
---
 .../selftests/bpf/prog_tests/verifier_log.c        | 23 ++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier_log.c b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
index 2ec82fc60c03..9ae0ac6e3b25 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier_log.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
@@ -178,26 +178,47 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 	opts.log_buf = logs.buf;
 	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
 	opts.log_size = sizeof(logs.buf);
+	opts.log_true_size = 0;
 	res = load_prog(&opts, expect_load_error);
 	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_fixed");
 
 	log_true_sz_fixed = opts.log_true_size;
 	ASSERT_GT(log_true_sz_fixed, 0, "log_true_sz_fixed");
 
+	/* (FIXED, NULL) get actual log size */
+	opts.log_buf = NULL;
+	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
+	opts.log_size = 0;
+	opts.log_true_size = 0;
+	res = load_prog(&opts, expect_load_error);
+	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_fixed_null");
+	ASSERT_EQ(opts.log_true_size, log_true_sz_fixed, "log_sz_fixed_null_eq");
+
 	/* (ROLLING) get actual log size */
 	opts.log_buf = logs.buf;
 	opts.log_level = log_level;
 	opts.log_size = sizeof(logs.buf);
+	opts.log_true_size = 0;
 	res = load_prog(&opts, expect_load_error);
 	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_rolling");
 
 	log_true_sz_rolling = opts.log_true_size;
 	ASSERT_EQ(log_true_sz_rolling, log_true_sz_fixed, "log_true_sz_eq");
 
+	/* (ROLLING, NULL) get actual log size */
+	opts.log_buf = NULL;
+	opts.log_level = log_level;
+	opts.log_size = 0;
+	opts.log_true_size = 0;
+	res = load_prog(&opts, expect_load_error);
+	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_rolling_null");
+	ASSERT_EQ(opts.log_true_size, log_true_sz_rolling, "log_true_sz_null_eq");
+
 	/* (FIXED) expect -ENOSPC for one byte short log */
 	opts.log_buf = logs.buf;
 	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
 	opts.log_size = log_true_sz_fixed - 1;
+	opts.log_true_size = 0;
 	res = load_prog(&opts, true /* should fail */);
 	ASSERT_EQ(res, -ENOSPC, "prog_load_res_too_short_fixed");
 
@@ -205,6 +226,7 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 	opts.log_buf = logs.buf;
 	opts.log_level = log_level | 8; /* BPF_LOG_FIXED */
 	opts.log_size = log_true_sz_fixed;
+	opts.log_true_size = 0;
 	res = load_prog(&opts, expect_load_error);
 	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_just_right_fixed");
 
@@ -219,6 +241,7 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 	opts.log_buf = logs.buf;
 	opts.log_level = log_level;
 	opts.log_size = log_true_sz_rolling;
+	opts.log_true_size = 0;
 	res = load_prog(&opts, expect_load_error);
 	ASSERT_NEQ(res, -ENOSPC, "prog_load_res_just_right_rolling");
 
-- 
cgit v1.2.3-70-g09d2


From 054b6c7866c7a2537fffd4aa12d88aac47db60f9 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Thu, 6 Apr 2023 16:42:05 -0700
Subject: selftests/bpf: Add verifier log tests for BPF_BTF_LOAD command

Add verifier log tests for BPF_BTF_LOAD command, which are very similar,
conceptually, to BPF_PROG_LOAD tests. These are two separate commands
dealing with verbose verifier log, so should be both tested separately.

Test that log_buf==NULL condition *does not* return -ENOSPC.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/bpf/20230406234205.323208-20-andrii@kernel.org
---
 .../selftests/bpf/prog_tests/verifier_log.c        | 188 +++++++++++++++++++++
 1 file changed, 188 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier_log.c b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
index 9ae0ac6e3b25..475092a78deb 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier_log.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
@@ -249,6 +249,190 @@ cleanup:
 	test_log_buf__destroy(skel);
 }
 
+static const void *btf_data;
+static u32 btf_data_sz;
+
+static int load_btf(struct bpf_btf_load_opts *opts, bool expect_err)
+{
+	int fd;
+
+	fd = bpf_btf_load(btf_data, btf_data_sz, opts);
+	if (fd >= 0)
+		close(fd);
+	if (expect_err)
+		ASSERT_LT(fd, 0, "btf_load_failure");
+	else /* !expect_err */
+		ASSERT_GT(fd, 0, "btf_load_success");
+	return fd;
+}
+
+static void verif_btf_log_subtest(bool bad_btf)
+{
+	LIBBPF_OPTS(bpf_btf_load_opts, opts);
+	struct btf *btf;
+	struct btf_type *t;
+	char *exp_log, op_name[32];
+	size_t fixed_log_sz;
+	__u32 log_true_sz_fixed, log_true_sz_rolling;
+	int i, res;
+
+	/* prepare simple BTF contents */
+	btf = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf, "btf_new_empty"))
+		return;
+	res = btf__add_int(btf, "whatever", 4, 0);
+	if (!ASSERT_GT(res, 0, "btf_add_int_id"))
+		goto cleanup;
+	if (bad_btf) {
+		/* btf__add_int() doesn't allow bad value of size, so we'll just
+		 * force-cast btf_type pointer and manually override size to invalid
+		 * 3 if we need to simulate failure
+		 */
+		t = (void *)btf__type_by_id(btf, res);
+		if (!ASSERT_OK_PTR(t, "int_btf_type"))
+			goto cleanup;
+		t->size = 3;
+	}
+
+	btf_data = btf__raw_data(btf, &btf_data_sz);
+	if (!ASSERT_OK_PTR(btf_data, "btf_data"))
+		goto cleanup;
+
+	load_btf(&opts, bad_btf);
+
+	opts.log_buf = logs.reference;
+	opts.log_size = sizeof(logs.reference);
+	opts.log_level = 1 | 8 /* BPF_LOG_FIXED */;
+	load_btf(&opts, bad_btf);
+
+	fixed_log_sz = strlen(logs.reference) + 1;
+	if (!ASSERT_GT(fixed_log_sz, 50, "fixed_log_sz"))
+		goto cleanup;
+	memset(logs.reference + fixed_log_sz, 0, sizeof(logs.reference) - fixed_log_sz);
+
+	/* validate BPF_LOG_FIXED truncation works as verifier log used to work */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1 | 8; /* fixed-length log */
+	opts.log_size = 25;
+	res = load_btf(&opts, true);
+	ASSERT_EQ(res, -ENOSPC, "half_log_fd");
+	ASSERT_EQ(strlen(logs.buf), 24, "log_fixed_25");
+	ASSERT_STRNEQ(logs.buf, logs.reference, 24, op_name);
+
+	/* validate rolling verifier log logic: try all variations of log buf
+	 * length to force various truncation scenarios
+	 */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1; /* rolling log */
+
+	/* prefill logs.buf with 'A's to detect any write beyond allowed length */
+	memset(logs.filler, 'A', sizeof(logs.filler));
+	logs.filler[sizeof(logs.filler) - 1] = '\0';
+	memset(logs.buf, 'A', sizeof(logs.buf));
+	logs.buf[sizeof(logs.buf) - 1] = '\0';
+
+	for (i = 1; i < fixed_log_sz; i++) {
+		opts.log_size = i;
+
+		snprintf(op_name, sizeof(op_name), "log_roll_btf_load_%d", i);
+		res = load_btf(&opts, true);
+		if (!ASSERT_EQ(res, -ENOSPC, op_name))
+			goto cleanup;
+
+		exp_log = logs.reference + fixed_log_sz - i;
+		snprintf(op_name, sizeof(op_name), "log_roll_contents_%d", i);
+		if (!ASSERT_STREQ(logs.buf, exp_log, op_name)) {
+			printf("CMP:%d\nS1:'%s'\nS2:'%s'\n",
+				strcmp(logs.buf, exp_log),
+				logs.buf, exp_log);
+			goto cleanup;
+		}
+
+		/* check that unused portions of logs.buf are not overwritten */
+		snprintf(op_name, sizeof(op_name), "log_roll_unused_tail_%d", i);
+		if (!ASSERT_STREQ(logs.buf + i, logs.filler + i, op_name)) {
+			printf("CMP:%d\nS1:'%s'\nS2:'%s'\n",
+				strcmp(logs.buf + i, logs.filler + i),
+				logs.buf + i, logs.filler + i);
+			goto cleanup;
+		}
+	}
+
+	/* (FIXED) get actual log size */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1 | 8; /* BPF_LOG_FIXED */
+	opts.log_size = sizeof(logs.buf);
+	opts.log_true_size = 0;
+	res = load_btf(&opts, bad_btf);
+	ASSERT_NEQ(res, -ENOSPC, "btf_load_res_fixed");
+
+	log_true_sz_fixed = opts.log_true_size;
+	ASSERT_GT(log_true_sz_fixed, 0, "log_true_sz_fixed");
+
+	/* (FIXED, NULL) get actual log size */
+	opts.log_buf = NULL;
+	opts.log_level = 1 | 8; /* BPF_LOG_FIXED */
+	opts.log_size = 0;
+	opts.log_true_size = 0;
+	res = load_btf(&opts, bad_btf);
+	ASSERT_NEQ(res, -ENOSPC, "btf_load_res_fixed_null");
+	ASSERT_EQ(opts.log_true_size, log_true_sz_fixed, "log_sz_fixed_null_eq");
+
+	/* (ROLLING) get actual log size */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1;
+	opts.log_size = sizeof(logs.buf);
+	opts.log_true_size = 0;
+	res = load_btf(&opts, bad_btf);
+	ASSERT_NEQ(res, -ENOSPC, "btf_load_res_rolling");
+
+	log_true_sz_rolling = opts.log_true_size;
+	ASSERT_EQ(log_true_sz_rolling, log_true_sz_fixed, "log_true_sz_eq");
+
+	/* (ROLLING, NULL) get actual log size */
+	opts.log_buf = NULL;
+	opts.log_level = 1;
+	opts.log_size = 0;
+	opts.log_true_size = 0;
+	res = load_btf(&opts, bad_btf);
+	ASSERT_NEQ(res, -ENOSPC, "btf_load_res_rolling_null");
+	ASSERT_EQ(opts.log_true_size, log_true_sz_rolling, "log_true_sz_null_eq");
+
+	/* (FIXED) expect -ENOSPC for one byte short log */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1 | 8; /* BPF_LOG_FIXED */
+	opts.log_size = log_true_sz_fixed - 1;
+	opts.log_true_size = 0;
+	res = load_btf(&opts, true);
+	ASSERT_EQ(res, -ENOSPC, "btf_load_res_too_short_fixed");
+
+	/* (FIXED) expect *not* -ENOSPC with exact log_true_size buffer */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1 | 8; /* BPF_LOG_FIXED */
+	opts.log_size = log_true_sz_fixed;
+	opts.log_true_size = 0;
+	res = load_btf(&opts, bad_btf);
+	ASSERT_NEQ(res, -ENOSPC, "btf_load_res_just_right_fixed");
+
+	/* (ROLLING) expect -ENOSPC for one byte short log */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1;
+	opts.log_size = log_true_sz_rolling - 1;
+	res = load_btf(&opts, true);
+	ASSERT_EQ(res, -ENOSPC, "btf_load_res_too_short_rolling");
+
+	/* (ROLLING) expect *not* -ENOSPC with exact log_true_size buffer */
+	opts.log_buf = logs.buf;
+	opts.log_level = 1;
+	opts.log_size = log_true_sz_rolling;
+	opts.log_true_size = 0;
+	res = load_btf(&opts, bad_btf);
+	ASSERT_NEQ(res, -ENOSPC, "btf_load_res_just_right_rolling");
+
+cleanup:
+	btf__free(btf);
+}
+
 void test_verifier_log(void)
 {
 	if (test__start_subtest("good_prog-level1"))
@@ -259,4 +443,8 @@ void test_verifier_log(void)
 		verif_log_subtest("bad_prog", true, 1);
 	if (test__start_subtest("bad_prog-level2"))
 		verif_log_subtest("bad_prog", true, 2);
+	if (test__start_subtest("bad_btf"))
+		verif_btf_log_subtest(true /* bad btf */);
+	if (test__start_subtest("good_btf"))
+		verif_btf_log_subtest(false /* !bad btf */);
 }
-- 
cgit v1.2.3-70-g09d2


From 91f2dc6838c19342f7f2993627c622835cc24890 Mon Sep 17 00:00:00 2001
From: Feng Zhou <zhoufeng.zf@bytedance.com>
Date: Mon, 10 Apr 2023 16:59:07 +0800
Subject: bpf/btf: Fix is_int_ptr()

When tracing a kernel function with arg type is u32*, btf_ctx_access()
would report error: arg2 type INT is not a struct.

The commit bb6728d75611 ("bpf: Allow access to int pointer arguments
in tracing programs") added support for int pointer, but did not skip
modifiers before checking it's type. This patch fixes it.

Fixes: bb6728d75611 ("bpf: Allow access to int pointer arguments in tracing programs")
Co-developed-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Feng Zhou <zhoufeng.zf@bytedance.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20230410085908.98493-2-zhoufeng.zf@bytedance.com
---
 kernel/bpf/btf.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index ffc31a1c84af..913b9d717a4a 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5921,12 +5921,8 @@ struct btf *bpf_prog_get_target_btf(const struct bpf_prog *prog)
 
 static bool is_int_ptr(struct btf *btf, const struct btf_type *t)
 {
-	/* t comes in already as a pointer */
-	t = btf_type_by_id(btf, t->type);
-
-	/* allow const */
-	if (BTF_INFO_KIND(t->info) == BTF_KIND_CONST)
-		t = btf_type_by_id(btf, t->type);
+	/* skip modifiers */
+	t = btf_type_skip_modifiers(btf, t->type, NULL);
 
 	return btf_type_is_int(t);
 }
-- 
cgit v1.2.3-70-g09d2


From 75dcef8d3609d0b1d3497d6ed4809096513e0b83 Mon Sep 17 00:00:00 2001
From: Feng Zhou <zhoufeng.zf@bytedance.com>
Date: Mon, 10 Apr 2023 16:59:08 +0800
Subject: selftests/bpf: Add test to access u32 ptr argument in tracing program

Adding verifier test for accessing u32 pointer argument in
tracing programs.

The test program loads 1nd argument of bpf_fentry_test9 function
which is u32 pointer and checks that verifier allows that.

Co-developed-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Feng Zhou <zhoufeng.zf@bytedance.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20230410085908.98493-3-zhoufeng.zf@bytedance.com
---
 net/bpf/test_run.c                                    |  8 +++++++-
 tools/testing/selftests/bpf/verifier/btf_ctx_access.c | 13 +++++++++++++
 2 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index f1652f5fbd2e..68bdfc041a7b 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -541,6 +541,11 @@ int noinline bpf_fentry_test8(struct bpf_fentry_test_t *arg)
 	return (long)arg->a;
 }
 
+__bpf_kfunc u32 bpf_fentry_test9(u32 *a)
+{
+	return *a;
+}
+
 __bpf_kfunc int bpf_modify_return_test(int a, int *b)
 {
 	*b += 1;
@@ -855,7 +860,8 @@ int bpf_prog_test_run_tracing(struct bpf_prog *prog,
 		    bpf_fentry_test5(11, (void *)12, 13, 14, 15) != 65 ||
 		    bpf_fentry_test6(16, (void *)17, 18, 19, (void *)20, 21) != 111 ||
 		    bpf_fentry_test7((struct bpf_fentry_test_t *)0) != 0 ||
-		    bpf_fentry_test8(&arg) != 0)
+		    bpf_fentry_test8(&arg) != 0 ||
+		    bpf_fentry_test9(&retval) != 0)
 			goto out;
 		break;
 	case BPF_MODIFY_RETURN:
diff --git a/tools/testing/selftests/bpf/verifier/btf_ctx_access.c b/tools/testing/selftests/bpf/verifier/btf_ctx_access.c
index 6340db6b46dc..0484d3de040d 100644
--- a/tools/testing/selftests/bpf/verifier/btf_ctx_access.c
+++ b/tools/testing/selftests/bpf/verifier/btf_ctx_access.c
@@ -10,3 +10,16 @@
 	.expected_attach_type = BPF_TRACE_FENTRY,
 	.kfunc = "bpf_modify_return_test",
 },
+
+{
+	"btf_ctx_access u32 pointer accept",
+	.insns = {
+	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),	/* load 1nd argument value (u32 pointer) */
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = ACCEPT,
+	.prog_type = BPF_PROG_TYPE_TRACING,
+	.expected_attach_type = BPF_TRACE_FENTRY,
+	.kfunc = "bpf_fentry_test9",
+},
-- 
cgit v1.2.3-70-g09d2


From e5688f6fb9e3d0254a8fb1c714c38e407f0baa64 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Wed, 1 Mar 2023 10:33:55 +0100
Subject: net/mlx5: Add mlx5_ifc definitions for bridge multicast support

Add the required hardware definitions to mlx5_ifc: fdb_uplink_hairpin,
fdb_multi_path_any_table_limit_regc, fdb_multi_path_any_table.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 include/linux/mlx5/mlx5_ifc.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index e47d6c58da35..02c628f4fe26 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -880,7 +880,12 @@ enum {
 
 struct mlx5_ifc_flow_table_eswitch_cap_bits {
 	u8      fdb_to_vport_reg_c_id[0x8];
-	u8      reserved_at_8[0xd];
+	u8      reserved_at_8[0x5];
+	u8      fdb_uplink_hairpin[0x1];
+	u8      fdb_multi_path_any_table_limit_regc[0x1];
+	u8      reserved_at_f[0x3];
+	u8      fdb_multi_path_any_table[0x1];
+	u8      reserved_at_13[0x2];
 	u8      fdb_modify_header_fwd_to_table[0x1];
 	u8      fdb_ipv4_ttl_modify[0x1];
 	u8      flow_source[0x1];
-- 
cgit v1.2.3-70-g09d2


From 9071b423c302b70eff80da715d724c4e75c1c46a Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Thu, 5 Jan 2023 15:28:29 +0100
Subject: net/mlx5: Bridge, increase bridge tables sizes

Bridge ingress and egress tables got more flow groups recently for QinQ
support and will get more in following patches of this series. Increase the
sizes of the tables to allow offloading more flows in each mode.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index 3cdcb0e0b20f..e45f9bb80535 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -13,8 +13,8 @@
 #define CREATE_TRACE_POINTS
 #include "diag/bridge_tracepoint.h"
 
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE 12000
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE 16000
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE 131072
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE 524288
 #define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_FROM 0
 #define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO		\
 	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
@@ -40,10 +40,10 @@
 	 MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE - 1)
 #define MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE			\
 	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_TO + 1)
-static_assert(MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE == 64000);
+static_assert(MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE == 1048576);
 
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE 16000
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_SIZE (32000 - 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE 131072
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_SIZE (262144 - 1)
 #define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_FROM 0
 #define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_TO		\
 	(MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE - 1)
@@ -63,7 +63,7 @@ static_assert(MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE == 64000);
 	MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_FROM
 #define MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE			\
 	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_TO + 1)
-static_assert(MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE == 64000);
+static_assert(MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE == 524288);
 
 #define MLX5_ESW_BRIDGE_SKIP_TABLE_SIZE 0
 
-- 
cgit v1.2.3-70-g09d2


From 6767c97d7adc3c4283c0167a9716de3953f251b1 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Sun, 19 Feb 2023 18:27:04 +0100
Subject: net/mlx5: Bridge, move additional data structures to priv header

Following patches in series will require accessing flow tables and groups
sizes, table levels and struct mlx5_esw_bridge from new the new source file
dedicated to multicast code. Expose these data in bridge_priv.h to reduce
clutter in following patches that will implement the actual functionality.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   | 85 ----------------------
 .../ethernet/mellanox/mlx5/core/esw/bridge_priv.h  | 85 ++++++++++++++++++++++
 2 files changed, 85 insertions(+), 85 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index e45f9bb80535..ec052fff7712 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -13,66 +13,6 @@
 #define CREATE_TRACE_POINTS
 #include "diag/bridge_tracepoint.h"
 
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE 131072
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE 524288
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_FROM 0
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO		\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_FROM	\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_TO		\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_FROM +	\
-	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_FROM			\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_TO			\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_FROM +		\
-	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_FROM	\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_TO		\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_FROM +	\
-	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_FROM			\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_TO			\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_FROM +		\
-	 MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE			\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_TO + 1)
-static_assert(MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE == 1048576);
-
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE 131072
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_SIZE (262144 - 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_FROM 0
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_TO		\
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_FROM		\
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_TO			\
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_FROM +		\
-	 MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_FROM \
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_TO			\
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_FROM +		\
-	 MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_FROM \
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_TO + 1)
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_TO	\
-	MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_FROM
-#define MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE			\
-	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_TO + 1)
-static_assert(MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE == 524288);
-
-#define MLX5_ESW_BRIDGE_SKIP_TABLE_SIZE 0
-
-enum {
-	MLX5_ESW_BRIDGE_LEVEL_INGRESS_TABLE,
-	MLX5_ESW_BRIDGE_LEVEL_EGRESS_TABLE,
-	MLX5_ESW_BRIDGE_LEVEL_SKIP_TABLE,
-};
-
 static const struct rhashtable_params fdb_ht_params = {
 	.key_offset = offsetof(struct mlx5_esw_bridge_fdb_entry, key),
 	.key_len = sizeof(struct mlx5_esw_bridge_fdb_key),
@@ -80,31 +20,6 @@ static const struct rhashtable_params fdb_ht_params = {
 	.automatic_shrinking = true,
 };
 
-enum {
-	MLX5_ESW_BRIDGE_VLAN_FILTERING_FLAG = BIT(0),
-};
-
-struct mlx5_esw_bridge {
-	int ifindex;
-	int refcnt;
-	struct list_head list;
-	struct mlx5_esw_bridge_offloads *br_offloads;
-
-	struct list_head fdb_list;
-	struct rhashtable fdb_ht;
-
-	struct mlx5_flow_table *egress_ft;
-	struct mlx5_flow_group *egress_vlan_fg;
-	struct mlx5_flow_group *egress_qinq_fg;
-	struct mlx5_flow_group *egress_mac_fg;
-	struct mlx5_flow_group *egress_miss_fg;
-	struct mlx5_pkt_reformat *egress_miss_pkt_reformat;
-	struct mlx5_flow_handle *egress_miss_handle;
-	unsigned long ageing_time;
-	u32 flags;
-	u16 vlan_proto;
-};
-
 static void
 mlx5_esw_bridge_fdb_offload_notify(struct net_device *dev, const unsigned char *addr, u16 vid,
 				   unsigned long val)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
index 878311fe950a..b99761e73c1b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
@@ -12,6 +12,70 @@
 #include <linux/xarray.h>
 #include "fs_core.h"
 
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE 131072
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE 524288
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_FROM 0
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_FROM	\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_FROM +	\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_FROM			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_FROM	\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_FROM +	\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_FROM			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_QINQ_FILTER_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MAC_GRP_IDX_TO + 1)
+static_assert(MLX5_ESW_BRIDGE_INGRESS_TABLE_SIZE == 1048576);
+
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE 131072
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_SIZE (262144 - 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_FROM 0
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_FROM		\
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_EGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_FROM \
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_QINQ_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_FROM \
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MAC_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_TO	\
+	MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_FROM
+#define MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE			\
+	(MLX5_ESW_BRIDGE_EGRESS_TABLE_MISS_GRP_IDX_TO + 1)
+static_assert(MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE == 524288);
+
+#define MLX5_ESW_BRIDGE_SKIP_TABLE_SIZE 0
+
+enum {
+	MLX5_ESW_BRIDGE_LEVEL_INGRESS_TABLE,
+	MLX5_ESW_BRIDGE_LEVEL_EGRESS_TABLE,
+	MLX5_ESW_BRIDGE_LEVEL_SKIP_TABLE,
+};
+
+enum {
+	MLX5_ESW_BRIDGE_VLAN_FILTERING_FLAG = BIT(0),
+};
+
 struct mlx5_esw_bridge_fdb_key {
 	unsigned char addr[ETH_ALEN];
 	u16 vid;
@@ -60,4 +124,25 @@ struct mlx5_esw_bridge_port {
 	struct xarray vlans;
 };
 
+struct mlx5_esw_bridge {
+	int ifindex;
+	int refcnt;
+	struct list_head list;
+	struct mlx5_esw_bridge_offloads *br_offloads;
+
+	struct list_head fdb_list;
+	struct rhashtable fdb_ht;
+
+	struct mlx5_flow_table *egress_ft;
+	struct mlx5_flow_group *egress_vlan_fg;
+	struct mlx5_flow_group *egress_qinq_fg;
+	struct mlx5_flow_group *egress_mac_fg;
+	struct mlx5_flow_group *egress_miss_fg;
+	struct mlx5_pkt_reformat *egress_miss_pkt_reformat;
+	struct mlx5_flow_handle *egress_miss_handle;
+	unsigned long ageing_time;
+	u32 flags;
+	u16 vlan_proto;
+};
+
 #endif /* _MLX5_ESW_BRIDGE_PRIVATE_ */
-- 
cgit v1.2.3-70-g09d2


From b99c4ef29e278ab0b91406e52df4fa9b634b117e Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Mon, 13 Mar 2023 16:22:49 +0100
Subject: net/mlx5: Bridge, extract code to lookup parent bridge of port

The pattern when function looks up a port by vport_num+vhca_id tuple in
order to just obtain its parent bridge is repeated multiple times in
bridge.c file. Further commits in this series use the pattern even more.
Extract the pattern to standalone mlx5_esw_bridge_from_port_lookup()
function to improve code readability.

This commits doesn't change functionality.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   | 47 ++++++++++++----------
 1 file changed, 26 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index ec052fff7712..bbbf982bbbc0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -933,6 +933,19 @@ static void mlx5_esw_bridge_port_erase(struct mlx5_esw_bridge_port *port,
 	xa_erase(&br_offloads->ports, mlx5_esw_bridge_port_key(port));
 }
 
+static struct mlx5_esw_bridge *
+mlx5_esw_bridge_from_port_lookup(u16 vport_num, u16 esw_owner_vhca_id,
+				 struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	struct mlx5_esw_bridge_port *port;
+
+	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!port)
+		return NULL;
+
+	return port->bridge;
+}
+
 static void mlx5_esw_bridge_fdb_entry_refresh(struct mlx5_esw_bridge_fdb_entry *entry)
 {
 	trace_mlx5_esw_bridge_fdb_entry_refresh(entry);
@@ -1388,28 +1401,26 @@ err_ingress_fc_create:
 int mlx5_esw_bridge_ageing_time_set(u16 vport_num, u16 esw_owner_vhca_id, unsigned long ageing_time,
 				    struct mlx5_esw_bridge_offloads *br_offloads)
 {
-	struct mlx5_esw_bridge_port *port;
+	struct mlx5_esw_bridge *bridge;
 
-	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
-	if (!port)
+	bridge = mlx5_esw_bridge_from_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!bridge)
 		return -EINVAL;
 
-	port->bridge->ageing_time = clock_t_to_jiffies(ageing_time);
+	bridge->ageing_time = clock_t_to_jiffies(ageing_time);
 	return 0;
 }
 
 int mlx5_esw_bridge_vlan_filtering_set(u16 vport_num, u16 esw_owner_vhca_id, bool enable,
 				       struct mlx5_esw_bridge_offloads *br_offloads)
 {
-	struct mlx5_esw_bridge_port *port;
 	struct mlx5_esw_bridge *bridge;
 	bool filtering;
 
-	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
-	if (!port)
+	bridge = mlx5_esw_bridge_from_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!bridge)
 		return -EINVAL;
 
-	bridge = port->bridge;
 	filtering = bridge->flags & MLX5_ESW_BRIDGE_VLAN_FILTERING_FLAG;
 	if (filtering == enable)
 		return 0;
@@ -1426,15 +1437,13 @@ int mlx5_esw_bridge_vlan_filtering_set(u16 vport_num, u16 esw_owner_vhca_id, boo
 int mlx5_esw_bridge_vlan_proto_set(u16 vport_num, u16 esw_owner_vhca_id, u16 proto,
 				   struct mlx5_esw_bridge_offloads *br_offloads)
 {
-	struct mlx5_esw_bridge_port *port;
 	struct mlx5_esw_bridge *bridge;
 
-	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id,
-					   br_offloads);
-	if (!port)
+	bridge = mlx5_esw_bridge_from_port_lookup(vport_num, esw_owner_vhca_id,
+						  br_offloads);
+	if (!bridge)
 		return -EINVAL;
 
-	bridge = port->bridge;
 	if (bridge->vlan_proto == proto)
 		return 0;
 	if (proto != ETH_P_8021Q && proto != ETH_P_8021AD) {
@@ -1626,14 +1635,12 @@ void mlx5_esw_bridge_fdb_update_used(struct net_device *dev, u16 vport_num, u16
 				     struct switchdev_notifier_fdb_info *fdb_info)
 {
 	struct mlx5_esw_bridge_fdb_entry *entry;
-	struct mlx5_esw_bridge_port *port;
 	struct mlx5_esw_bridge *bridge;
 
-	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
-	if (!port)
+	bridge = mlx5_esw_bridge_from_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!bridge)
 		return;
 
-	bridge = port->bridge;
 	entry = mlx5_esw_bridge_fdb_lookup(bridge, fdb_info->addr, fdb_info->vid);
 	if (!entry) {
 		esw_debug(br_offloads->esw->dev,
@@ -1680,14 +1687,12 @@ void mlx5_esw_bridge_fdb_remove(struct net_device *dev, u16 vport_num, u16 esw_o
 {
 	struct mlx5_eswitch *esw = br_offloads->esw;
 	struct mlx5_esw_bridge_fdb_entry *entry;
-	struct mlx5_esw_bridge_port *port;
 	struct mlx5_esw_bridge *bridge;
 
-	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
-	if (!port)
+	bridge = mlx5_esw_bridge_from_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!bridge)
 		return;
 
-	bridge = port->bridge;
 	entry = mlx5_esw_bridge_fdb_lookup(bridge, fdb_info->addr, fdb_info->vid);
 	if (!entry) {
 		esw_debug(esw->dev,
-- 
cgit v1.2.3-70-g09d2


From 18c2916cee12137f32669d42ac1fbb87cba343bb Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Tue, 21 Feb 2023 21:31:48 +0100
Subject: net/mlx5: Bridge, snoop igmp/mld packets

Handle SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED attribute notification to
dynamically toggle bridge multicast offload. Set new
MLX5_ESW_BRIDGE_MCAST_FLAG bridge flag when multicast offload is enabled.
Put multicast-specific code into new bridge_mcast.c file.

When initializing bridge multicast pipeline create a static rule for
snooping on IGMP traffic and three rules for snooping on MLD traffic (for
query, report and done message types). Note that matching MLD traffic
requires having flexparser MLX5_FLEX_PROTO_ICMPV6 capability enabled.

By default Linux bridge is created with multicast enabled which can be
modified by 'mcast_snooping' argument:

$ ip link set name my_bridge type bridge mcast_snooping 0

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
 .../ethernet/mellanox/mlx5/core/en/rep/bridge.c    |   4 +
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   |  31 ++
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.h   |   9 +
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c | 316 +++++++++++++++++++++
 .../ethernet/mellanox/mlx5/core/esw/bridge_priv.h  |  27 +-
 6 files changed, 384 insertions(+), 5 deletions(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 6c2f1d4a58ab..4a009b720bee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -75,7 +75,7 @@ mlx5_core-$(CONFIG_MLX5_ESWITCH)   += esw/acl/helper.o \
 				      esw/acl/egress_lgcy.o esw/acl/egress_ofld.o \
 				      esw/acl/ingress_lgcy.o esw/acl/ingress_ofld.o
 
-mlx5_core-$(CONFIG_MLX5_BRIDGE)    += esw/bridge.o en/rep/bridge.o
+mlx5_core-$(CONFIG_MLX5_BRIDGE)    += esw/bridge.o esw/bridge_mcast.o en/rep/bridge.o
 
 mlx5_core-$(CONFIG_THERMAL)        += thermal.o
 mlx5_core-$(CONFIG_MLX5_MPFS)      += lib/mpfs.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
index ce85b48d327d..9701eb3c5f1b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
@@ -306,6 +306,10 @@ mlx5_esw_bridge_port_obj_attr_set(struct net_device *dev,
 						     attr->u.vlan_protocol,
 						     br_offloads);
 		break;
+	case SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED:
+		err = mlx5_esw_bridge_mcast_set(vport_num, esw_owner_vhca_id,
+						!attr->u.mc_disabled, br_offloads);
+		break;
 	default:
 		err = -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index bbbf982bbbc0..35436aa9548d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -868,6 +868,7 @@ static void mlx5_esw_bridge_put(struct mlx5_esw_bridge_offloads *br_offloads,
 		return;
 
 	mlx5_esw_bridge_egress_table_cleanup(bridge);
+	mlx5_esw_bridge_mcast_disable(bridge);
 	list_del(&bridge->list);
 	rhashtable_destroy(&bridge->fdb_ht);
 	kvfree(bridge);
@@ -1458,6 +1459,36 @@ int mlx5_esw_bridge_vlan_proto_set(u16 vport_num, u16 esw_owner_vhca_id, u16 pro
 	return 0;
 }
 
+int mlx5_esw_bridge_mcast_set(u16 vport_num, u16 esw_owner_vhca_id, bool enable,
+			      struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	struct mlx5_eswitch *esw = br_offloads->esw;
+	struct mlx5_esw_bridge *bridge;
+	int err = 0;
+	bool mcast;
+
+	if (!(MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_multi_path_any_table) ||
+	      MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_multi_path_any_table_limit_regc)) ||
+	    !MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_uplink_hairpin) ||
+	    !MLX5_CAP_ESW_FLOWTABLE_FDB((esw)->dev, ignore_flow_level))
+		return -EOPNOTSUPP;
+
+	bridge = mlx5_esw_bridge_from_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!bridge)
+		return -EINVAL;
+
+	mcast = bridge->flags & MLX5_ESW_BRIDGE_MCAST_FLAG;
+	if (mcast == enable)
+		return 0;
+
+	if (enable)
+		err = mlx5_esw_bridge_mcast_enable(bridge);
+	else
+		mlx5_esw_bridge_mcast_disable(bridge);
+
+	return err;
+}
+
 static int mlx5_esw_bridge_vport_init(u16 vport_num, u16 esw_owner_vhca_id, u16 flags,
 				      struct mlx5_esw_bridge_offloads *br_offloads,
 				      struct mlx5_esw_bridge *bridge)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
index 10851a515bca..b18f137173d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
@@ -25,12 +25,19 @@ struct mlx5_esw_bridge_offloads {
 	struct delayed_work update_work;
 
 	struct mlx5_flow_table *ingress_ft;
+	struct mlx5_flow_group *ingress_igmp_fg;
+	struct mlx5_flow_group *ingress_mld_fg;
 	struct mlx5_flow_group *ingress_vlan_fg;
 	struct mlx5_flow_group *ingress_vlan_filter_fg;
 	struct mlx5_flow_group *ingress_qinq_fg;
 	struct mlx5_flow_group *ingress_qinq_filter_fg;
 	struct mlx5_flow_group *ingress_mac_fg;
 
+	struct mlx5_flow_handle *igmp_handle;
+	struct mlx5_flow_handle *mld_query_handle;
+	struct mlx5_flow_handle *mld_report_handle;
+	struct mlx5_flow_handle *mld_done_handle;
+
 	struct mlx5_flow_table *skip_ft;
 };
 
@@ -64,6 +71,8 @@ int mlx5_esw_bridge_vlan_filtering_set(u16 vport_num, u16 esw_owner_vhca_id, boo
 				       struct mlx5_esw_bridge_offloads *br_offloads);
 int mlx5_esw_bridge_vlan_proto_set(u16 vport_num, u16 esw_owner_vhca_id, u16 proto,
 				   struct mlx5_esw_bridge_offloads *br_offloads);
+int mlx5_esw_bridge_mcast_set(u16 vport_num, u16 esw_owner_vhca_id, bool enable,
+			      struct mlx5_esw_bridge_offloads *br_offloads);
 int mlx5_esw_bridge_port_vlan_add(u16 vport_num, u16 esw_owner_vhca_id, u16 vid, u16 flags,
 				  struct mlx5_esw_bridge_offloads *br_offloads,
 				  struct netlink_ext_ack *extack);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
new file mode 100644
index 000000000000..d5a89a86c9e8
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
@@ -0,0 +1,316 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
+
+#include "bridge.h"
+#include "eswitch.h"
+#include "bridge_priv.h"
+
+static struct mlx5_flow_group *
+mlx5_esw_bridge_ingress_igmp_fg_create(struct mlx5_eswitch *esw,
+				       struct mlx5_flow_table *ingress_ft)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_group *fg;
+	u32 *in, *match;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return ERR_PTR(-ENOMEM);
+
+	MLX5_SET(create_flow_group_in, in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	match = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+
+	MLX5_SET_TO_ONES(fte_match_param, match, outer_headers.ip_version);
+	MLX5_SET_TO_ONES(fte_match_param, match, outer_headers.ip_protocol);
+
+	MLX5_SET(create_flow_group_in, in, start_flow_index,
+		 MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_IDX_FROM);
+	MLX5_SET(create_flow_group_in, in, end_flow_index,
+		 MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_IDX_TO);
+
+	fg = mlx5_create_flow_group(ingress_ft, in);
+	kvfree(in);
+	if (IS_ERR(fg))
+		esw_warn(esw->dev,
+			 "Failed to create IGMP flow group for bridge ingress table (err=%pe)\n",
+			 fg);
+
+	return fg;
+}
+
+static struct mlx5_flow_group *
+mlx5_esw_bridge_ingress_mld_fg_create(struct mlx5_eswitch *esw,
+				      struct mlx5_flow_table *ingress_ft)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_group *fg;
+	u32 *in, *match;
+
+	if (!(MLX5_CAP_GEN(esw->dev, flex_parser_protocols) & MLX5_FLEX_PROTO_ICMPV6)) {
+		esw_warn(esw->dev,
+			 "Can't create MLD flow group due to missing hardware ICMPv6 parsing support\n");
+		return NULL;
+	}
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return ERR_PTR(-ENOMEM);
+
+	MLX5_SET(create_flow_group_in, in, match_criteria_enable,
+		 MLX5_MATCH_OUTER_HEADERS | MLX5_MATCH_MISC_PARAMETERS_3);
+	match = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+
+	MLX5_SET_TO_ONES(fte_match_param, match, outer_headers.ip_version);
+	MLX5_SET_TO_ONES(fte_match_param, match, misc_parameters_3.icmpv6_type);
+
+	MLX5_SET(create_flow_group_in, in, start_flow_index,
+		 MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_IDX_FROM);
+	MLX5_SET(create_flow_group_in, in, end_flow_index,
+		 MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_IDX_TO);
+
+	fg = mlx5_create_flow_group(ingress_ft, in);
+	kvfree(in);
+	if (IS_ERR(fg))
+		esw_warn(esw->dev,
+			 "Failed to create MLD flow group for bridge ingress table (err=%pe)\n",
+			 fg);
+
+	return fg;
+}
+
+static int
+mlx5_esw_bridge_ingress_mcast_fgs_init(struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	struct mlx5_flow_table *ingress_ft = br_offloads->ingress_ft;
+	struct mlx5_eswitch *esw = br_offloads->esw;
+	struct mlx5_flow_group *igmp_fg, *mld_fg;
+
+	igmp_fg = mlx5_esw_bridge_ingress_igmp_fg_create(esw, ingress_ft);
+	if (IS_ERR(igmp_fg))
+		return PTR_ERR(igmp_fg);
+
+	mld_fg = mlx5_esw_bridge_ingress_mld_fg_create(esw, ingress_ft);
+	if (IS_ERR(mld_fg)) {
+		mlx5_destroy_flow_group(igmp_fg);
+		return PTR_ERR(mld_fg);
+	}
+
+	br_offloads->ingress_igmp_fg = igmp_fg;
+	br_offloads->ingress_mld_fg = mld_fg;
+	return 0;
+}
+
+static void
+mlx5_esw_bridge_ingress_mcast_fgs_cleanup(struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	if (br_offloads->ingress_mld_fg)
+		mlx5_destroy_flow_group(br_offloads->ingress_mld_fg);
+	br_offloads->ingress_mld_fg = NULL;
+	if (br_offloads->ingress_igmp_fg)
+		mlx5_destroy_flow_group(br_offloads->ingress_igmp_fg);
+	br_offloads->ingress_igmp_fg = NULL;
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_ingress_igmp_fh_create(struct mlx5_flow_table *ingress_ft,
+				       struct mlx5_flow_table *skip_ft)
+{
+	struct mlx5_flow_destination dest = {
+		.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE,
+		.ft = skip_ft,
+	};
+	struct mlx5_flow_act flow_act = {
+		.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+		.flags = FLOW_ACT_NO_APPEND,
+	};
+	struct mlx5_flow_spec *rule_spec;
+	struct mlx5_flow_handle *handle;
+
+	rule_spec = kvzalloc(sizeof(*rule_spec), GFP_KERNEL);
+	if (!rule_spec)
+		return ERR_PTR(-ENOMEM);
+
+	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+
+	MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria, outer_headers.ip_version);
+	MLX5_SET(fte_match_param, rule_spec->match_value, outer_headers.ip_version, 4);
+	MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria, outer_headers.ip_protocol);
+	MLX5_SET(fte_match_param, rule_spec->match_value, outer_headers.ip_protocol, IPPROTO_IGMP);
+
+	handle = mlx5_add_flow_rules(ingress_ft, rule_spec, &flow_act, &dest, 1);
+
+	kvfree(rule_spec);
+	return handle;
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_ingress_mld_fh_create(u8 type, struct mlx5_flow_table *ingress_ft,
+				      struct mlx5_flow_table *skip_ft)
+{
+	struct mlx5_flow_destination dest = {
+		.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE,
+		.ft = skip_ft,
+	};
+	struct mlx5_flow_act flow_act = {
+		.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+		.flags = FLOW_ACT_NO_APPEND,
+	};
+	struct mlx5_flow_spec *rule_spec;
+	struct mlx5_flow_handle *handle;
+
+	rule_spec = kvzalloc(sizeof(*rule_spec), GFP_KERNEL);
+	if (!rule_spec)
+		return ERR_PTR(-ENOMEM);
+
+	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS | MLX5_MATCH_MISC_PARAMETERS_3;
+
+	MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria, outer_headers.ip_version);
+	MLX5_SET(fte_match_param, rule_spec->match_value, outer_headers.ip_version, 6);
+	MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria, misc_parameters_3.icmpv6_type);
+	MLX5_SET(fte_match_param, rule_spec->match_value, misc_parameters_3.icmpv6_type, type);
+
+	handle = mlx5_add_flow_rules(ingress_ft, rule_spec, &flow_act, &dest, 1);
+
+	kvfree(rule_spec);
+	return handle;
+}
+
+static int
+mlx5_esw_bridge_ingress_mcast_fhs_create(struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	struct mlx5_flow_handle *igmp_handle, *mld_query_handle, *mld_report_handle,
+		*mld_done_handle;
+	struct mlx5_flow_table *ingress_ft = br_offloads->ingress_ft,
+		*skip_ft = br_offloads->skip_ft;
+	int err;
+
+	igmp_handle = mlx5_esw_bridge_ingress_igmp_fh_create(ingress_ft, skip_ft);
+	if (IS_ERR(igmp_handle))
+		return PTR_ERR(igmp_handle);
+
+	if (br_offloads->ingress_mld_fg) {
+		mld_query_handle = mlx5_esw_bridge_ingress_mld_fh_create(ICMPV6_MGM_QUERY,
+									 ingress_ft,
+									 skip_ft);
+		if (IS_ERR(mld_query_handle)) {
+			err = PTR_ERR(mld_query_handle);
+			goto err_mld_query;
+		}
+
+		mld_report_handle = mlx5_esw_bridge_ingress_mld_fh_create(ICMPV6_MGM_REPORT,
+									  ingress_ft,
+									  skip_ft);
+		if (IS_ERR(mld_report_handle)) {
+			err = PTR_ERR(mld_report_handle);
+			goto err_mld_report;
+		}
+
+		mld_done_handle = mlx5_esw_bridge_ingress_mld_fh_create(ICMPV6_MGM_REDUCTION,
+									ingress_ft,
+									skip_ft);
+		if (IS_ERR(mld_done_handle)) {
+			err = PTR_ERR(mld_done_handle);
+			goto err_mld_done;
+		}
+	} else {
+		mld_query_handle = NULL;
+		mld_report_handle = NULL;
+		mld_done_handle = NULL;
+	}
+
+	br_offloads->igmp_handle = igmp_handle;
+	br_offloads->mld_query_handle = mld_query_handle;
+	br_offloads->mld_report_handle = mld_report_handle;
+	br_offloads->mld_done_handle = mld_done_handle;
+
+	return 0;
+
+err_mld_done:
+	mlx5_del_flow_rules(mld_report_handle);
+err_mld_report:
+	mlx5_del_flow_rules(mld_query_handle);
+err_mld_query:
+	mlx5_del_flow_rules(igmp_handle);
+	return err;
+}
+
+static void
+mlx5_esw_bridge_ingress_mcast_fhs_cleanup(struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	if (br_offloads->mld_done_handle)
+		mlx5_del_flow_rules(br_offloads->mld_done_handle);
+	br_offloads->mld_done_handle = NULL;
+	if (br_offloads->mld_report_handle)
+		mlx5_del_flow_rules(br_offloads->mld_report_handle);
+	br_offloads->mld_report_handle = NULL;
+	if (br_offloads->mld_query_handle)
+		mlx5_del_flow_rules(br_offloads->mld_query_handle);
+	br_offloads->mld_query_handle = NULL;
+	if (br_offloads->igmp_handle)
+		mlx5_del_flow_rules(br_offloads->igmp_handle);
+	br_offloads->igmp_handle = NULL;
+}
+
+static int mlx5_esw_brige_mcast_global_enable(struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	int err;
+
+	if (br_offloads->ingress_igmp_fg)
+		return 0; /* already enabled by another bridge */
+
+	err = mlx5_esw_bridge_ingress_mcast_fgs_init(br_offloads);
+	if (err) {
+		esw_warn(br_offloads->esw->dev,
+			 "Failed to create global multicast flow groups (err=%d)\n",
+			 err);
+		return err;
+	}
+
+	err = mlx5_esw_bridge_ingress_mcast_fhs_create(br_offloads);
+	if (err) {
+		esw_warn(br_offloads->esw->dev,
+			 "Failed to create global multicast flows (err=%d)\n",
+			 err);
+		goto err_fhs;
+	}
+
+	return 0;
+
+err_fhs:
+	mlx5_esw_bridge_ingress_mcast_fgs_cleanup(br_offloads);
+	return err;
+}
+
+static void mlx5_esw_brige_mcast_global_disable(struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	struct mlx5_esw_bridge *br;
+
+	list_for_each_entry(br, &br_offloads->bridges, list) {
+		/* Ingress table is global, so only disable snooping when all
+		 * bridges on esw have multicast disabled.
+		 */
+		if (br->flags & MLX5_ESW_BRIDGE_MCAST_FLAG)
+			return;
+	}
+
+	mlx5_esw_bridge_ingress_mcast_fhs_cleanup(br_offloads);
+	mlx5_esw_bridge_ingress_mcast_fgs_cleanup(br_offloads);
+}
+
+int mlx5_esw_bridge_mcast_enable(struct mlx5_esw_bridge *bridge)
+{
+	int err;
+
+	err = mlx5_esw_brige_mcast_global_enable(bridge->br_offloads);
+	if (err)
+		return err;
+
+	bridge->flags |= MLX5_ESW_BRIDGE_MCAST_FLAG;
+	return 0;
+}
+
+void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge)
+{
+	bridge->flags &= ~MLX5_ESW_BRIDGE_MCAST_FLAG;
+	mlx5_esw_brige_mcast_global_disable(bridge->br_offloads);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
index b99761e73c1b..dbb935db1b3c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
@@ -12,11 +12,26 @@
 #include <linux/xarray.h>
 #include "fs_core.h"
 
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_SIZE 1
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_SIZE 3
 #define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE 131072
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE 524288
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_FROM 0
-#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO		\
-	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_UNTAGGED_GRP_SIZE			\
+	(524288 - MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_SIZE -		\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_SIZE)
+
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_IDX_FROM 0
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_IDX_FROM	\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_IGMP_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_IDX_FROM +	\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_FROM			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_MLD_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_SIZE - 1)
 #define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_FROM	\
 	(MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_GRP_IDX_TO + 1)
 #define MLX5_ESW_BRIDGE_INGRESS_TABLE_VLAN_FILTER_GRP_IDX_TO		\
@@ -74,6 +89,7 @@ enum {
 
 enum {
 	MLX5_ESW_BRIDGE_VLAN_FILTERING_FLAG = BIT(0),
+	MLX5_ESW_BRIDGE_MCAST_FLAG = BIT(1),
 };
 
 struct mlx5_esw_bridge_fdb_key {
@@ -145,4 +161,7 @@ struct mlx5_esw_bridge {
 	u16 vlan_proto;
 };
 
+int mlx5_esw_bridge_mcast_enable(struct mlx5_esw_bridge *bridge);
+void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge);
+
 #endif /* _MLX5_ESW_BRIDGE_PRIVATE_ */
-- 
cgit v1.2.3-70-g09d2


From 272ecfc92f6f141f3381da65c2966851bd77baae Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Wed, 22 Feb 2023 13:10:02 +0100
Subject: net/mlx5: Bridge, add per-port multicast replication tables

Multicast replication requires adding one more level of FDB_BR_OFFLOAD
priority flow tables. The new level is used for per-port multicast-specific
tables that have following flow groups structure (flow highest to lowest
priority):

- Flow group of size one that matches on source port metadata. This will
have a static single rule that prevent packets from being replicated to
their source port.

- Flow group of size one that matches all packets and forwards them to the
port that owns the table.

Initialize the table dynamically on all bridge ports when adding a port to
the bridge that has multicast enabled and on all existing bridge ports when
receiving multicast enable notification.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   |  14 +-
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c | 329 ++++++++++++++++++++-
 .../ethernet/mellanox/mlx5/core/esw/bridge_priv.h  |  28 ++
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c  |   2 +-
 4 files changed, 370 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index 35436aa9548d..4bc8c6fc394b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -61,7 +61,7 @@ mlx5_esw_bridge_pkt_reformat_vlan_pop_create(struct mlx5_eswitch *esw)
 	return mlx5_packet_reformat_alloc(esw->dev, &reformat_params, MLX5_FLOW_NAMESPACE_FDB);
 }
 
-static struct mlx5_flow_table *
+struct mlx5_flow_table *
 mlx5_esw_bridge_table_create(int max_fte, u32 level, struct mlx5_eswitch *esw)
 {
 	struct mlx5_flow_table_attr ft_attr = {};
@@ -1506,6 +1506,15 @@ static int mlx5_esw_bridge_vport_init(u16 vport_num, u16 esw_owner_vhca_id, u16
 	port->bridge = bridge;
 	port->flags |= flags;
 	xa_init(&port->vlans);
+
+	err = mlx5_esw_bridge_port_mcast_init(port);
+	if (err) {
+		esw_warn(esw->dev,
+			 "Failed to initialize port multicast (vport=%u,esw_owner_vhca_id=%u,err=%d)\n",
+			 port->vport_num, port->esw_owner_vhca_id, err);
+		goto err_port_mcast;
+	}
+
 	err = mlx5_esw_bridge_port_insert(port, br_offloads);
 	if (err) {
 		esw_warn(esw->dev,
@@ -1518,6 +1527,8 @@ static int mlx5_esw_bridge_vport_init(u16 vport_num, u16 esw_owner_vhca_id, u16
 	return 0;
 
 err_port_insert:
+	mlx5_esw_bridge_port_mcast_cleanup(port);
+err_port_mcast:
 	kvfree(port);
 	return err;
 }
@@ -1535,6 +1546,7 @@ static int mlx5_esw_bridge_vport_cleanup(struct mlx5_esw_bridge_offloads *br_off
 
 	trace_mlx5_esw_bridge_vport_cleanup(port);
 	mlx5_esw_bridge_port_vlans_flush(port, bridge);
+	mlx5_esw_bridge_port_mcast_cleanup(port);
 	mlx5_esw_bridge_port_erase(port, br_offloads);
 	kvfree(port);
 	mlx5_esw_bridge_put(br_offloads, bridge);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
index d5a89a86c9e8..4f54cb41ed19 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
@@ -1,10 +1,283 @@
 // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
 
+#include "lib/devcom.h"
 #include "bridge.h"
 #include "eswitch.h"
 #include "bridge_priv.h"
 
+static int mlx5_esw_bridge_port_mcast_fts_init(struct mlx5_esw_bridge_port *port,
+					       struct mlx5_esw_bridge *bridge)
+{
+	struct mlx5_eswitch *esw = bridge->br_offloads->esw;
+	struct mlx5_flow_table *mcast_ft;
+
+	mcast_ft = mlx5_esw_bridge_table_create(MLX5_ESW_BRIDGE_MCAST_TABLE_SIZE,
+						MLX5_ESW_BRIDGE_LEVEL_MCAST_TABLE,
+						esw);
+	if (IS_ERR(mcast_ft))
+		return PTR_ERR(mcast_ft);
+
+	port->mcast.ft = mcast_ft;
+	return 0;
+}
+
+static void mlx5_esw_bridge_port_mcast_fts_cleanup(struct mlx5_esw_bridge_port *port)
+{
+	if (port->mcast.ft)
+		mlx5_destroy_flow_table(port->mcast.ft);
+	port->mcast.ft = NULL;
+}
+
+static struct mlx5_flow_group *
+mlx5_esw_bridge_mcast_filter_fg_create(struct mlx5_eswitch *esw,
+				       struct mlx5_flow_table *mcast_ft)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_group *fg;
+	u32 *in, *match;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return ERR_PTR(-ENOMEM);
+
+	MLX5_SET(create_flow_group_in, in, match_criteria_enable, MLX5_MATCH_MISC_PARAMETERS_2);
+	match = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+
+	MLX5_SET(fte_match_param, match, misc_parameters_2.metadata_reg_c_0,
+		 mlx5_eswitch_get_vport_metadata_mask());
+
+	MLX5_SET(create_flow_group_in, in, start_flow_index,
+		 MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_FROM);
+	MLX5_SET(create_flow_group_in, in, end_flow_index,
+		 MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_TO);
+
+	fg = mlx5_create_flow_group(mcast_ft, in);
+	kvfree(in);
+	if (IS_ERR(fg))
+		esw_warn(esw->dev,
+			 "Failed to create filter flow group for bridge mcast table (err=%pe)\n",
+			 fg);
+
+	return fg;
+}
+
+static struct mlx5_flow_group *
+mlx5_esw_bridge_mcast_fwd_fg_create(struct mlx5_eswitch *esw,
+				    struct mlx5_flow_table *mcast_ft)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_group *fg;
+	u32 *in;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return ERR_PTR(-ENOMEM);
+
+	MLX5_SET(create_flow_group_in, in, start_flow_index,
+		 MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_FROM);
+	MLX5_SET(create_flow_group_in, in, end_flow_index,
+		 MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_TO);
+
+	fg = mlx5_create_flow_group(mcast_ft, in);
+	kvfree(in);
+	if (IS_ERR(fg))
+		esw_warn(esw->dev,
+			 "Failed to create forward flow group for bridge mcast table (err=%pe)\n",
+			 fg);
+
+	return fg;
+}
+
+static int mlx5_esw_bridge_port_mcast_fgs_init(struct mlx5_esw_bridge_port *port)
+{
+	struct mlx5_eswitch *esw = port->bridge->br_offloads->esw;
+	struct mlx5_flow_table *mcast_ft = port->mcast.ft;
+	struct mlx5_flow_group *fwd_fg, *filter_fg;
+	int err;
+
+	filter_fg = mlx5_esw_bridge_mcast_filter_fg_create(esw, mcast_ft);
+	if (IS_ERR(filter_fg))
+		return PTR_ERR(filter_fg);
+
+	fwd_fg = mlx5_esw_bridge_mcast_fwd_fg_create(esw, mcast_ft);
+	if (IS_ERR(fwd_fg)) {
+		err = PTR_ERR(fwd_fg);
+		goto err_fwd_fg;
+	}
+
+	port->mcast.filter_fg = filter_fg;
+	port->mcast.fwd_fg = fwd_fg;
+
+	return 0;
+
+err_fwd_fg:
+	mlx5_destroy_flow_group(filter_fg);
+	return err;
+}
+
+static void mlx5_esw_bridge_port_mcast_fgs_cleanup(struct mlx5_esw_bridge_port *port)
+{
+	if (port->mcast.fwd_fg)
+		mlx5_destroy_flow_group(port->mcast.fwd_fg);
+	port->mcast.fwd_fg = NULL;
+	if (port->mcast.filter_fg)
+		mlx5_destroy_flow_group(port->mcast.filter_fg);
+	port->mcast.filter_fg = NULL;
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_mcast_flow_with_esw_create(struct mlx5_esw_bridge_port *port,
+					   struct mlx5_eswitch *esw)
+{
+	struct mlx5_flow_act flow_act = {
+		.action = MLX5_FLOW_CONTEXT_ACTION_DROP,
+		.flags = FLOW_ACT_NO_APPEND,
+	};
+	struct mlx5_flow_spec *rule_spec;
+	struct mlx5_flow_handle *handle;
+
+	rule_spec = kvzalloc(sizeof(*rule_spec), GFP_KERNEL);
+	if (!rule_spec)
+		return ERR_PTR(-ENOMEM);
+
+	rule_spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+
+	MLX5_SET(fte_match_param, rule_spec->match_criteria,
+		 misc_parameters_2.metadata_reg_c_0, mlx5_eswitch_get_vport_metadata_mask());
+	MLX5_SET(fte_match_param, rule_spec->match_value, misc_parameters_2.metadata_reg_c_0,
+		 mlx5_eswitch_get_vport_metadata_for_match(esw, port->vport_num));
+
+	handle = mlx5_add_flow_rules(port->mcast.ft, rule_spec, &flow_act, NULL, 0);
+
+	kvfree(rule_spec);
+	return handle;
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_mcast_filter_flow_create(struct mlx5_esw_bridge_port *port)
+{
+	return mlx5_esw_bridge_mcast_flow_with_esw_create(port, port->bridge->br_offloads->esw);
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_mcast_filter_flow_peer_create(struct mlx5_esw_bridge_port *port)
+{
+	struct mlx5_devcom *devcom = port->bridge->br_offloads->esw->dev->priv.devcom;
+	static struct mlx5_flow_handle *handle;
+	struct mlx5_eswitch *peer_esw;
+
+	peer_esw = mlx5_devcom_get_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
+	if (!peer_esw)
+		return ERR_PTR(-ENODEV);
+
+	handle = mlx5_esw_bridge_mcast_flow_with_esw_create(port, peer_esw);
+
+	mlx5_devcom_release_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
+	return handle;
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_mcast_fwd_flow_create(struct mlx5_esw_bridge_port *port)
+{
+	struct mlx5_flow_act flow_act = {
+		.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+		.flags = FLOW_ACT_NO_APPEND,
+	};
+	struct mlx5_flow_destination dest = {
+		.type = MLX5_FLOW_DESTINATION_TYPE_VPORT,
+		.vport.num = port->vport_num,
+	};
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_flow_spec *rule_spec;
+	struct mlx5_flow_handle *handle;
+
+	rule_spec = kvzalloc(sizeof(*rule_spec), GFP_KERNEL);
+	if (!rule_spec)
+		return ERR_PTR(-ENOMEM);
+
+	if (MLX5_CAP_ESW_FLOWTABLE(bridge->br_offloads->esw->dev, flow_source) &&
+	    port->vport_num == MLX5_VPORT_UPLINK)
+		rule_spec->flow_context.flow_source =
+			MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;
+
+	if (MLX5_CAP_ESW(bridge->br_offloads->esw->dev, merged_eswitch)) {
+		dest.vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;
+		dest.vport.vhca_id = port->esw_owner_vhca_id;
+	}
+	handle = mlx5_add_flow_rules(port->mcast.ft, rule_spec, &flow_act, &dest, 1);
+
+	kvfree(rule_spec);
+	return handle;
+}
+
+static int mlx5_esw_bridge_port_mcast_fhs_init(struct mlx5_esw_bridge_port *port)
+{
+	struct mlx5_flow_handle *filter_handle, *fwd_handle;
+
+	filter_handle = (port->flags & MLX5_ESW_BRIDGE_PORT_FLAG_PEER) ?
+		mlx5_esw_bridge_mcast_filter_flow_peer_create(port) :
+		mlx5_esw_bridge_mcast_filter_flow_create(port);
+	if (IS_ERR(filter_handle))
+		return PTR_ERR(filter_handle);
+
+	fwd_handle = mlx5_esw_bridge_mcast_fwd_flow_create(port);
+	if (IS_ERR(fwd_handle)) {
+		mlx5_del_flow_rules(filter_handle);
+		return PTR_ERR(fwd_handle);
+	}
+
+	port->mcast.filter_handle = filter_handle;
+	port->mcast.fwd_handle = fwd_handle;
+
+	return 0;
+}
+
+static void mlx5_esw_bridge_port_mcast_fhs_cleanup(struct mlx5_esw_bridge_port *port)
+{
+	if (port->mcast.fwd_handle)
+		mlx5_del_flow_rules(port->mcast.fwd_handle);
+	port->mcast.fwd_handle = NULL;
+	if (port->mcast.filter_handle)
+		mlx5_del_flow_rules(port->mcast.filter_handle);
+	port->mcast.filter_handle = NULL;
+}
+
+int mlx5_esw_bridge_port_mcast_init(struct mlx5_esw_bridge_port *port)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	int err;
+
+	if (!(bridge->flags & MLX5_ESW_BRIDGE_MCAST_FLAG))
+		return 0;
+
+	err = mlx5_esw_bridge_port_mcast_fts_init(port, bridge);
+	if (err)
+		return err;
+
+	err = mlx5_esw_bridge_port_mcast_fgs_init(port);
+	if (err)
+		goto err_fgs;
+
+	err = mlx5_esw_bridge_port_mcast_fhs_init(port);
+	if (err)
+		goto err_fhs;
+	return err;
+
+err_fhs:
+	mlx5_esw_bridge_port_mcast_fgs_cleanup(port);
+err_fgs:
+	mlx5_esw_bridge_port_mcast_fts_cleanup(port);
+	return err;
+}
+
+void mlx5_esw_bridge_port_mcast_cleanup(struct mlx5_esw_bridge_port *port)
+{
+	mlx5_esw_bridge_port_mcast_fhs_cleanup(port);
+	mlx5_esw_bridge_port_mcast_fgs_cleanup(port);
+	mlx5_esw_bridge_port_mcast_fts_cleanup(port);
+}
+
 static struct mlx5_flow_group *
 mlx5_esw_bridge_ingress_igmp_fg_create(struct mlx5_eswitch *esw,
 				       struct mlx5_flow_table *ingress_ft)
@@ -251,6 +524,51 @@ mlx5_esw_bridge_ingress_mcast_fhs_cleanup(struct mlx5_esw_bridge_offloads *br_of
 	br_offloads->igmp_handle = NULL;
 }
 
+static int mlx5_esw_brige_mcast_init(struct mlx5_esw_bridge *bridge)
+{
+	struct mlx5_esw_bridge_offloads *br_offloads = bridge->br_offloads;
+	struct mlx5_esw_bridge_port *port, *failed;
+	unsigned long i;
+	int err;
+
+	xa_for_each(&br_offloads->ports, i, port) {
+		if (port->bridge != bridge)
+			continue;
+
+		err = mlx5_esw_bridge_port_mcast_init(port);
+		if (err) {
+			failed = port;
+			goto err_port;
+		}
+	}
+	return 0;
+
+err_port:
+	xa_for_each(&br_offloads->ports, i, port) {
+		if (port == failed)
+			break;
+		if (port->bridge != bridge)
+			continue;
+
+		mlx5_esw_bridge_port_mcast_cleanup(port);
+	}
+	return err;
+}
+
+static void mlx5_esw_brige_mcast_cleanup(struct mlx5_esw_bridge *bridge)
+{
+	struct mlx5_esw_bridge_offloads *br_offloads = bridge->br_offloads;
+	struct mlx5_esw_bridge_port *port;
+	unsigned long i;
+
+	xa_for_each(&br_offloads->ports, i, port) {
+		if (port->bridge != bridge)
+			continue;
+
+		mlx5_esw_bridge_port_mcast_cleanup(port);
+	}
+}
+
 static int mlx5_esw_brige_mcast_global_enable(struct mlx5_esw_bridge_offloads *br_offloads)
 {
 	int err;
@@ -306,11 +624,20 @@ int mlx5_esw_bridge_mcast_enable(struct mlx5_esw_bridge *bridge)
 		return err;
 
 	bridge->flags |= MLX5_ESW_BRIDGE_MCAST_FLAG;
-	return 0;
+
+	err = mlx5_esw_brige_mcast_init(bridge);
+	if (err) {
+		esw_warn(bridge->br_offloads->esw->dev, "Failed to enable multicast (err=%d)\n",
+			 err);
+		bridge->flags &= ~MLX5_ESW_BRIDGE_MCAST_FLAG;
+		mlx5_esw_brige_mcast_global_disable(bridge->br_offloads);
+	}
+	return err;
 }
 
 void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge)
 {
+	mlx5_esw_brige_mcast_cleanup(bridge);
 	bridge->flags &= ~MLX5_ESW_BRIDGE_MCAST_FLAG;
 	mlx5_esw_brige_mcast_global_disable(bridge->br_offloads);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
index dbb935db1b3c..7fdd719f363c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
@@ -81,9 +81,23 @@ static_assert(MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE == 524288);
 
 #define MLX5_ESW_BRIDGE_SKIP_TABLE_SIZE 0
 
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_SIZE 1
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_SIZE 1
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_FROM 0
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_TO		\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_FROM		\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_FROM +			\
+	 MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_SIZE - 1)
+
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_SIZE			\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_TO + 1)
 enum {
 	MLX5_ESW_BRIDGE_LEVEL_INGRESS_TABLE,
 	MLX5_ESW_BRIDGE_LEVEL_EGRESS_TABLE,
+	MLX5_ESW_BRIDGE_LEVEL_MCAST_TABLE,
 	MLX5_ESW_BRIDGE_LEVEL_SKIP_TABLE,
 };
 
@@ -138,6 +152,14 @@ struct mlx5_esw_bridge_port {
 	u16 flags;
 	struct mlx5_esw_bridge *bridge;
 	struct xarray vlans;
+	struct {
+		struct mlx5_flow_table *ft;
+		struct mlx5_flow_group *filter_fg;
+		struct mlx5_flow_group *fwd_fg;
+
+		struct mlx5_flow_handle *filter_handle;
+		struct mlx5_flow_handle *fwd_handle;
+	} mcast;
 };
 
 struct mlx5_esw_bridge {
@@ -161,6 +183,12 @@ struct mlx5_esw_bridge {
 	u16 vlan_proto;
 };
 
+struct mlx5_flow_table *mlx5_esw_bridge_table_create(int max_fte, u32 level,
+						     struct mlx5_eswitch *esw);
+
+int mlx5_esw_bridge_port_mcast_init(struct mlx5_esw_bridge_port *port);
+void mlx5_esw_bridge_port_mcast_cleanup(struct mlx5_esw_bridge_port *port);
+
 int mlx5_esw_bridge_mcast_enable(struct mlx5_esw_bridge *bridge);
 void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 8e3da9d4fe1c..19da02c41616 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2997,7 +2997,7 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
 		goto out_err;
 	}
 
-	maj_prio = fs_create_prio(&steering->fdb_root_ns->ns, FDB_BR_OFFLOAD, 3);
+	maj_prio = fs_create_prio(&steering->fdb_root_ns->ns, FDB_BR_OFFLOAD, 4);
 	if (IS_ERR(maj_prio)) {
 		err = PTR_ERR(maj_prio);
 		goto out_err;
-- 
cgit v1.2.3-70-g09d2


From b5e80625d1687c75e51b3dc9543a1b769d964250 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Wed, 22 Feb 2023 13:13:32 +0100
Subject: net/mlx5: Bridge, support multicast VLAN pop

When VLAN with 'untagged' flag is created on port also provision the
per-port multicast table rule to pop the VLAN during packet replication.
This functionality must be in per-port table because some subset of ports
that are member of multicast group can require just a match on VLAN (trunk
mode) while other subset can be configured to remove the VLAN tag from
packets received on the ports (access mode).

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   |  33 +++-
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c | 190 ++++++++++++++++++++-
 .../ethernet/mellanox/mlx5/core/esw/bridge_priv.h  |  22 ++-
 3 files changed, 236 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index 4bc8c6fc394b..52c976135397 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -1095,8 +1095,21 @@ mlx5_esw_bridge_vlan_push_mark_cleanup(struct mlx5_esw_bridge_vlan *vlan, struct
 }
 
 static int
-mlx5_esw_bridge_vlan_push_pop_create(u16 vlan_proto, u16 flags, struct mlx5_esw_bridge_vlan *vlan,
-				     struct mlx5_eswitch *esw)
+mlx5_esw_bridge_vlan_push_pop_fhs_create(u16 vlan_proto, struct mlx5_esw_bridge_port *port,
+					 struct mlx5_esw_bridge_vlan *vlan)
+{
+	return mlx5_esw_bridge_vlan_mcast_init(vlan_proto, port, vlan);
+}
+
+static void
+mlx5_esw_bridge_vlan_push_pop_fhs_cleanup(struct mlx5_esw_bridge_vlan *vlan)
+{
+	mlx5_esw_bridge_vlan_mcast_cleanup(vlan);
+}
+
+static int
+mlx5_esw_bridge_vlan_push_pop_create(u16 vlan_proto, u16 flags, struct mlx5_esw_bridge_port *port,
+				     struct mlx5_esw_bridge_vlan *vlan, struct mlx5_eswitch *esw)
 {
 	int err;
 
@@ -1114,10 +1127,16 @@ mlx5_esw_bridge_vlan_push_pop_create(u16 vlan_proto, u16 flags, struct mlx5_esw_
 		err = mlx5_esw_bridge_vlan_pop_create(vlan, esw);
 		if (err)
 			goto err_vlan_pop;
+
+		err = mlx5_esw_bridge_vlan_push_pop_fhs_create(vlan_proto, port, vlan);
+		if (err)
+			goto err_vlan_pop_fhs;
 	}
 
 	return 0;
 
+err_vlan_pop_fhs:
+	mlx5_esw_bridge_vlan_pop_cleanup(vlan, esw);
 err_vlan_pop:
 	if (vlan->pkt_mod_hdr_push_mark)
 		mlx5_esw_bridge_vlan_push_mark_cleanup(vlan, esw);
@@ -1142,7 +1161,7 @@ mlx5_esw_bridge_vlan_create(u16 vlan_proto, u16 vid, u16 flags, struct mlx5_esw_
 	vlan->flags = flags;
 	INIT_LIST_HEAD(&vlan->fdb_list);
 
-	err = mlx5_esw_bridge_vlan_push_pop_create(vlan_proto, flags, vlan, esw);
+	err = mlx5_esw_bridge_vlan_push_pop_create(vlan_proto, flags, port, vlan, esw);
 	if (err)
 		goto err_vlan_push_pop;
 
@@ -1154,6 +1173,8 @@ mlx5_esw_bridge_vlan_create(u16 vlan_proto, u16 vid, u16 flags, struct mlx5_esw_
 	return vlan;
 
 err_xa_insert:
+	if (vlan->mcast_handle)
+		mlx5_esw_bridge_vlan_push_pop_fhs_cleanup(vlan);
 	if (vlan->pkt_reformat_pop)
 		mlx5_esw_bridge_vlan_pop_cleanup(vlan, esw);
 	if (vlan->pkt_mod_hdr_push_mark)
@@ -1180,6 +1201,8 @@ static void mlx5_esw_bridge_vlan_flush(struct mlx5_esw_bridge_vlan *vlan,
 	list_for_each_entry_safe(entry, tmp, &vlan->fdb_list, vlan_list)
 		mlx5_esw_bridge_fdb_entry_notify_and_cleanup(entry, bridge);
 
+	if (vlan->mcast_handle)
+		mlx5_esw_bridge_vlan_push_pop_fhs_cleanup(vlan);
 	if (vlan->pkt_reformat_pop)
 		mlx5_esw_bridge_vlan_pop_cleanup(vlan, esw);
 	if (vlan->pkt_mod_hdr_push_mark)
@@ -1218,8 +1241,8 @@ static int mlx5_esw_bridge_port_vlans_recreate(struct mlx5_esw_bridge_port *port
 
 	xa_for_each(&port->vlans, i, vlan) {
 		mlx5_esw_bridge_vlan_flush(vlan, bridge);
-		err = mlx5_esw_bridge_vlan_push_pop_create(bridge->vlan_proto, vlan->flags, vlan,
-							   br_offloads->esw);
+		err = mlx5_esw_bridge_vlan_push_pop_create(bridge->vlan_proto, vlan->flags, port,
+							   vlan, br_offloads->esw);
 		if (err) {
 			esw_warn(br_offloads->esw->dev,
 				 "Failed to create VLAN=%u(proto=%x) push/pop actions (vport=%u,err=%d)\n",
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
index 4f54cb41ed19..99e2f9fc11a2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
@@ -62,6 +62,60 @@ mlx5_esw_bridge_mcast_filter_fg_create(struct mlx5_eswitch *esw,
 	return fg;
 }
 
+static struct mlx5_flow_group *
+mlx5_esw_bridge_mcast_vlan_proto_fg_create(unsigned int from, unsigned int to, u16 vlan_proto,
+					   struct mlx5_eswitch *esw,
+					   struct mlx5_flow_table *mcast_ft)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_group *fg;
+	u32 *in, *match;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return ERR_PTR(-ENOMEM);
+
+	MLX5_SET(create_flow_group_in, in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
+	match = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
+
+	if (vlan_proto == ETH_P_8021Q)
+		MLX5_SET_TO_ONES(fte_match_param, match, outer_headers.cvlan_tag);
+	else if (vlan_proto == ETH_P_8021AD)
+		MLX5_SET_TO_ONES(fte_match_param, match, outer_headers.svlan_tag);
+	MLX5_SET_TO_ONES(fte_match_param, match, outer_headers.first_vid);
+
+	MLX5_SET(create_flow_group_in, in, start_flow_index, from);
+	MLX5_SET(create_flow_group_in, in, end_flow_index, to);
+
+	fg = mlx5_create_flow_group(mcast_ft, in);
+	kvfree(in);
+	if (IS_ERR(fg))
+		esw_warn(esw->dev,
+			 "Failed to create VLAN(proto=%x) flow group for bridge mcast table (err=%pe)\n",
+			 vlan_proto, fg);
+
+	return fg;
+}
+
+static struct mlx5_flow_group *
+mlx5_esw_bridge_mcast_vlan_fg_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *mcast_ft)
+{
+	unsigned int from = MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_IDX_FROM;
+	unsigned int to = MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_IDX_TO;
+
+	return mlx5_esw_bridge_mcast_vlan_proto_fg_create(from, to, ETH_P_8021Q, esw, mcast_ft);
+}
+
+static struct mlx5_flow_group *
+mlx5_esw_bridge_mcast_qinq_fg_create(struct mlx5_eswitch *esw,
+				     struct mlx5_flow_table *mcast_ft)
+{
+	unsigned int from = MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_IDX_FROM;
+	unsigned int to = MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_IDX_TO;
+
+	return mlx5_esw_bridge_mcast_vlan_proto_fg_create(from, to, ETH_P_8021AD, esw, mcast_ft);
+}
+
 static struct mlx5_flow_group *
 mlx5_esw_bridge_mcast_fwd_fg_create(struct mlx5_eswitch *esw,
 				    struct mlx5_flow_table *mcast_ft)
@@ -91,15 +145,27 @@ mlx5_esw_bridge_mcast_fwd_fg_create(struct mlx5_eswitch *esw,
 
 static int mlx5_esw_bridge_port_mcast_fgs_init(struct mlx5_esw_bridge_port *port)
 {
+	struct mlx5_flow_group *fwd_fg, *qinq_fg, *vlan_fg, *filter_fg;
 	struct mlx5_eswitch *esw = port->bridge->br_offloads->esw;
 	struct mlx5_flow_table *mcast_ft = port->mcast.ft;
-	struct mlx5_flow_group *fwd_fg, *filter_fg;
 	int err;
 
 	filter_fg = mlx5_esw_bridge_mcast_filter_fg_create(esw, mcast_ft);
 	if (IS_ERR(filter_fg))
 		return PTR_ERR(filter_fg);
 
+	vlan_fg = mlx5_esw_bridge_mcast_vlan_fg_create(esw, mcast_ft);
+	if (IS_ERR(vlan_fg)) {
+		err = PTR_ERR(vlan_fg);
+		goto err_vlan_fg;
+	}
+
+	qinq_fg = mlx5_esw_bridge_mcast_qinq_fg_create(esw, mcast_ft);
+	if (IS_ERR(qinq_fg)) {
+		err = PTR_ERR(qinq_fg);
+		goto err_qinq_fg;
+	}
+
 	fwd_fg = mlx5_esw_bridge_mcast_fwd_fg_create(esw, mcast_ft);
 	if (IS_ERR(fwd_fg)) {
 		err = PTR_ERR(fwd_fg);
@@ -107,11 +173,17 @@ static int mlx5_esw_bridge_port_mcast_fgs_init(struct mlx5_esw_bridge_port *port
 	}
 
 	port->mcast.filter_fg = filter_fg;
+	port->mcast.vlan_fg = vlan_fg;
+	port->mcast.qinq_fg = qinq_fg;
 	port->mcast.fwd_fg = fwd_fg;
 
 	return 0;
 
 err_fwd_fg:
+	mlx5_destroy_flow_group(qinq_fg);
+err_qinq_fg:
+	mlx5_destroy_flow_group(vlan_fg);
+err_vlan_fg:
 	mlx5_destroy_flow_group(filter_fg);
 	return err;
 }
@@ -121,6 +193,12 @@ static void mlx5_esw_bridge_port_mcast_fgs_cleanup(struct mlx5_esw_bridge_port *
 	if (port->mcast.fwd_fg)
 		mlx5_destroy_flow_group(port->mcast.fwd_fg);
 	port->mcast.fwd_fg = NULL;
+	if (port->mcast.qinq_fg)
+		mlx5_destroy_flow_group(port->mcast.qinq_fg);
+	port->mcast.qinq_fg = NULL;
+	if (port->mcast.vlan_fg)
+		mlx5_destroy_flow_group(port->mcast.vlan_fg);
+	port->mcast.vlan_fg = NULL;
 	if (port->mcast.filter_fg)
 		mlx5_destroy_flow_group(port->mcast.filter_fg);
 	port->mcast.filter_fg = NULL;
@@ -177,6 +255,82 @@ mlx5_esw_bridge_mcast_filter_flow_peer_create(struct mlx5_esw_bridge_port *port)
 	return handle;
 }
 
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_mcast_vlan_flow_create(u16 vlan_proto, struct mlx5_esw_bridge_port *port,
+				       struct mlx5_esw_bridge_vlan *vlan)
+{
+	struct mlx5_flow_act flow_act = {
+		.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+		.flags = FLOW_ACT_NO_APPEND,
+	};
+	struct mlx5_flow_destination dest = {
+		.type = MLX5_FLOW_DESTINATION_TYPE_VPORT,
+		.vport.num = port->vport_num,
+	};
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_flow_spec *rule_spec;
+	struct mlx5_flow_handle *handle;
+
+	rule_spec = kvzalloc(sizeof(*rule_spec), GFP_KERNEL);
+	if (!rule_spec)
+		return ERR_PTR(-ENOMEM);
+
+	if (MLX5_CAP_ESW_FLOWTABLE(bridge->br_offloads->esw->dev, flow_source) &&
+	    port->vport_num == MLX5_VPORT_UPLINK)
+		rule_spec->flow_context.flow_source =
+			MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;
+	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+
+	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
+	flow_act.pkt_reformat = vlan->pkt_reformat_pop;
+
+	if (vlan_proto == ETH_P_8021Q) {
+		MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria,
+				 outer_headers.cvlan_tag);
+		MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_value,
+				 outer_headers.cvlan_tag);
+	} else if (vlan_proto == ETH_P_8021AD) {
+		MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria,
+				 outer_headers.svlan_tag);
+		MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_value,
+				 outer_headers.svlan_tag);
+	}
+	MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria, outer_headers.first_vid);
+	MLX5_SET(fte_match_param, rule_spec->match_value, outer_headers.first_vid, vlan->vid);
+
+	if (MLX5_CAP_ESW(bridge->br_offloads->esw->dev, merged_eswitch)) {
+		dest.vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;
+		dest.vport.vhca_id = port->esw_owner_vhca_id;
+	}
+	handle = mlx5_add_flow_rules(port->mcast.ft, rule_spec, &flow_act, &dest, 1);
+
+	kvfree(rule_spec);
+	return handle;
+}
+
+int mlx5_esw_bridge_vlan_mcast_init(u16 vlan_proto, struct mlx5_esw_bridge_port *port,
+				    struct mlx5_esw_bridge_vlan *vlan)
+{
+	struct mlx5_flow_handle *handle;
+
+	if (!(port->bridge->flags & MLX5_ESW_BRIDGE_MCAST_FLAG))
+		return 0;
+
+	handle = mlx5_esw_bridge_mcast_vlan_flow_create(vlan_proto, port, vlan);
+	if (IS_ERR(handle))
+		return PTR_ERR(handle);
+
+	vlan->mcast_handle = handle;
+	return 0;
+}
+
+void mlx5_esw_bridge_vlan_mcast_cleanup(struct mlx5_esw_bridge_vlan *vlan)
+{
+	if (vlan->mcast_handle)
+		mlx5_del_flow_rules(vlan->mcast_handle);
+	vlan->mcast_handle = NULL;
+}
+
 static struct mlx5_flow_handle *
 mlx5_esw_bridge_mcast_fwd_flow_create(struct mlx5_esw_bridge_port *port)
 {
@@ -214,6 +368,10 @@ mlx5_esw_bridge_mcast_fwd_flow_create(struct mlx5_esw_bridge_port *port)
 static int mlx5_esw_bridge_port_mcast_fhs_init(struct mlx5_esw_bridge_port *port)
 {
 	struct mlx5_flow_handle *filter_handle, *fwd_handle;
+	struct mlx5_esw_bridge_vlan *vlan, *failed;
+	unsigned long index;
+	int err;
+
 
 	filter_handle = (port->flags & MLX5_ESW_BRIDGE_PORT_FLAG_PEER) ?
 		mlx5_esw_bridge_mcast_filter_flow_peer_create(port) :
@@ -223,18 +381,44 @@ static int mlx5_esw_bridge_port_mcast_fhs_init(struct mlx5_esw_bridge_port *port
 
 	fwd_handle = mlx5_esw_bridge_mcast_fwd_flow_create(port);
 	if (IS_ERR(fwd_handle)) {
-		mlx5_del_flow_rules(filter_handle);
-		return PTR_ERR(fwd_handle);
+		err = PTR_ERR(fwd_handle);
+		goto err_fwd;
+	}
+
+	xa_for_each(&port->vlans, index, vlan) {
+		err = mlx5_esw_bridge_vlan_mcast_init(port->bridge->vlan_proto, port, vlan);
+		if (err) {
+			failed = vlan;
+			goto err_vlan;
+		}
 	}
 
 	port->mcast.filter_handle = filter_handle;
 	port->mcast.fwd_handle = fwd_handle;
 
 	return 0;
+
+err_vlan:
+	xa_for_each(&port->vlans, index, vlan) {
+		if (vlan == failed)
+			break;
+
+		mlx5_esw_bridge_vlan_mcast_cleanup(vlan);
+	}
+	mlx5_del_flow_rules(fwd_handle);
+err_fwd:
+	mlx5_del_flow_rules(filter_handle);
+	return err;
 }
 
 static void mlx5_esw_bridge_port_mcast_fhs_cleanup(struct mlx5_esw_bridge_port *port)
 {
+	struct mlx5_esw_bridge_vlan *vlan;
+	unsigned long index;
+
+	xa_for_each(&port->vlans, index, vlan)
+		mlx5_esw_bridge_vlan_mcast_cleanup(vlan);
+
 	if (port->mcast.fwd_handle)
 		mlx5_del_flow_rules(port->mcast.fwd_handle);
 	port->mcast.fwd_handle = NULL;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
index 7fdd719f363c..36ff32001ce8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
@@ -83,17 +83,31 @@ static_assert(MLX5_ESW_BRIDGE_EGRESS_TABLE_SIZE == 524288);
 
 #define MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_SIZE 1
 #define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_SIZE 1
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_SIZE 4095
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_SIZE MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_SIZE
 #define MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_FROM 0
 #define MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_TO		\
 	(MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_SIZE - 1)
-#define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_FROM		\
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_IDX_FROM		\
 	(MLX5_ESW_BRIDGE_MCAST_TABLE_FILTER_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_IDX_FROM		\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_VLAN_GRP_IDX_TO + 1)
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_IDX_TO			\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_IDX_FROM +		\
+	 MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_SIZE - 1)
+#define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_FROM		\
+	(MLX5_ESW_BRIDGE_MCAST_TABLE_QINQ_GRP_IDX_TO + 1)
 #define MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_TO			\
 	(MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_FROM +			\
 	 MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_SIZE - 1)
 
 #define MLX5_ESW_BRIDGE_MCAST_TABLE_SIZE			\
 	(MLX5_ESW_BRIDGE_MCAST_TABLE_FWD_GRP_IDX_TO + 1)
+static_assert(MLX5_ESW_BRIDGE_MCAST_TABLE_SIZE == 8192);
+
 enum {
 	MLX5_ESW_BRIDGE_LEVEL_INGRESS_TABLE,
 	MLX5_ESW_BRIDGE_LEVEL_EGRESS_TABLE,
@@ -144,6 +158,7 @@ struct mlx5_esw_bridge_vlan {
 	struct mlx5_pkt_reformat *pkt_reformat_push;
 	struct mlx5_pkt_reformat *pkt_reformat_pop;
 	struct mlx5_modify_hdr *pkt_mod_hdr_push_mark;
+	struct mlx5_flow_handle *mcast_handle;
 };
 
 struct mlx5_esw_bridge_port {
@@ -155,6 +170,8 @@ struct mlx5_esw_bridge_port {
 	struct {
 		struct mlx5_flow_table *ft;
 		struct mlx5_flow_group *filter_fg;
+		struct mlx5_flow_group *vlan_fg;
+		struct mlx5_flow_group *qinq_fg;
 		struct mlx5_flow_group *fwd_fg;
 
 		struct mlx5_flow_handle *filter_handle;
@@ -188,6 +205,9 @@ struct mlx5_flow_table *mlx5_esw_bridge_table_create(int max_fte, u32 level,
 
 int mlx5_esw_bridge_port_mcast_init(struct mlx5_esw_bridge_port *port);
 void mlx5_esw_bridge_port_mcast_cleanup(struct mlx5_esw_bridge_port *port);
+int mlx5_esw_bridge_vlan_mcast_init(u16 vlan_proto, struct mlx5_esw_bridge_port *port,
+				    struct mlx5_esw_bridge_vlan *vlan);
+void mlx5_esw_bridge_vlan_mcast_cleanup(struct mlx5_esw_bridge_vlan *vlan);
 
 int mlx5_esw_bridge_mcast_enable(struct mlx5_esw_bridge *bridge);
 void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge);
-- 
cgit v1.2.3-70-g09d2


From 70f0302b3f2053b64a6a653a76a20cb95438dc85 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Sat, 4 Feb 2023 07:58:54 +0100
Subject: net/mlx5: Bridge, implement mdb offload

Implement support for add/del SWITCHDEV_OBJ_ID_PORT_MDB events. For mdb
destination addresses configure egress table rules to replicate to per-port
multicast tables of all ports that are member of the multicast group as
illustrated by 'MDB1' rule in the following diagram:

                                                                                                                            +--------+--+
                                                                                    +---------------------------------------> Port 1 |  |
                                                                                    |                                       +-^------+--+
                                                                                    |                                         |
                                                                                    |                                         |
                                       +-----------------------------------------+  |     +---------------------------+       |
                                       | EGRESS table                            |  |  +--> PORT 1 multicast table    |       |
+----------------------------------+   +-----------------------------------------+  |  |  +---------------------------+       |
| INGRESS table                    |   |                                         |  |  |  |                           |       |
+----------------------------------+   | dst_mac=P1,vlan=X -> pop vlan, goto P1  +--+  |  | FG0:                      |       |
|                                  |   | dst_mac=P1,vlan=Y -> pop vlan, goto P1  |     |  | src_port=dst_port -> drop |       |
| src_mac=M1,vlan=X -> goto egress +---> dst_mac=P2,vlan=X -> pop vlan, goto P2  +--+  |  | FG1:                      |       |
| ...                              |   | dst_mac=P2,vlan=Y -> goto P2            |  |  |  | VLAN X -> pop, goto port  |       |
|                                  |   | dst_mac=MDB1,vlan=Y -> goto mcast P1,P2 +-----+  | ...                       |       |
+----------------------------------+   |                                         |  |  |  | VLAN Y -> pop, goto port  +-------+
                                       +-----------------------------------------+  |  |  | FG3:                      |
                                                                                    |  |  | matchall -> goto port     |
                                                                                    |  |  |                           |
                                                                                    |  |  +---------------------------+
                                                                                    |  |
                                                                                    |  |
                                                                                    |  |                                    +--------+--+
                                                                                    +---------------------------------------> Port 2 |  |
                                                                                       |                                    +-^------+--+
                                                                                       |                                      |
                                                                                       |                                      |
                                                                                       |  +---------------------------+       |
                                                                                       +--> PORT 2 multicast table    |       |
                                                                                          +---------------------------+       |
                                                                                          |                           |       |
                                                                                          | FG0:                      |       |
                                                                                          | src_port=dst_port -> drop |       |
                                                                                          | FG1:                      |       |
                                                                                          | VLAN X -> pop, goto port  |       |
                                                                                          | ...                       |       |
                                                                                          |                           |       |
                                                                                          | FG3:                      |       |
                                                                                          | matchall -> goto port     +-------+
                                                                                          |                           |
                                                                                          +---------------------------+

MDB is managed by extending mlx5 bridge to store an entry in
mlx5_esw_bridge->mdb_list linked list (used to iterate over all offloaded
MDBs) and mlx5_esw_bridge->mdb_ht hash table (used to lookup existing MDB
by MAC+VLAN). Every MDB entry can be attached to arbitrary amount of bridge
ports that are stored in mlx5_esw_bridge_mdb_entry->ports xarray in order
to allow both efficient lookup of the port and also iteration over all
ports that the entry is attached to. Every time MDB is attached/detached
to/from a port, the hardware rule is recreated with list of destinations
corresponding to all attached ports. When the entry is detached from the
last port it is removed from mdb and destroyed which means that the ports
xarray also acts as implicit reference counting mechanism.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/rep/bridge.c    |  12 +
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   |  75 +++++-
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.h   |   6 +
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c | 295 +++++++++++++++++++++
 .../ethernet/mellanox/mlx5/core/esw/bridge_priv.h  |  29 ++
 5 files changed, 413 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
index 9701eb3c5f1b..dd0dd3f028a3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
@@ -220,6 +220,7 @@ mlx5_esw_bridge_port_obj_add(struct net_device *dev,
 	struct netlink_ext_ack *extack = switchdev_notifier_info_to_extack(&port_obj_info->info);
 	const struct switchdev_obj *obj = port_obj_info->obj;
 	const struct switchdev_obj_port_vlan *vlan;
+	const struct switchdev_obj_port_mdb *mdb;
 	u16 vport_num, esw_owner_vhca_id;
 	int err;
 
@@ -235,6 +236,11 @@ mlx5_esw_bridge_port_obj_add(struct net_device *dev,
 		err = mlx5_esw_bridge_port_vlan_add(vport_num, esw_owner_vhca_id, vlan->vid,
 						    vlan->flags, br_offloads, extack);
 		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+		mdb = SWITCHDEV_OBJ_PORT_MDB(obj);
+		err = mlx5_esw_bridge_port_mdb_add(vport_num, esw_owner_vhca_id, mdb->addr,
+						   mdb->vid, br_offloads, extack);
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -248,6 +254,7 @@ mlx5_esw_bridge_port_obj_del(struct net_device *dev,
 {
 	const struct switchdev_obj *obj = port_obj_info->obj;
 	const struct switchdev_obj_port_vlan *vlan;
+	const struct switchdev_obj_port_mdb *mdb;
 	u16 vport_num, esw_owner_vhca_id;
 
 	if (!mlx5_esw_bridge_rep_vport_num_vhca_id_get(dev, br_offloads->esw, &vport_num,
@@ -261,6 +268,11 @@ mlx5_esw_bridge_port_obj_del(struct net_device *dev,
 		vlan = SWITCHDEV_OBJ_PORT_VLAN(obj);
 		mlx5_esw_bridge_port_vlan_del(vport_num, esw_owner_vhca_id, vlan->vid, br_offloads);
 		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+		mdb = SWITCHDEV_OBJ_PORT_MDB(obj);
+		mlx5_esw_bridge_port_mdb_del(vport_num, esw_owner_vhca_id, mdb->addr, mdb->vid,
+					     br_offloads);
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index 52c976135397..be4787539c6c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -840,6 +840,10 @@ static struct mlx5_esw_bridge *mlx5_esw_bridge_create(int ifindex,
 	if (err)
 		goto err_fdb_ht;
 
+	err = mlx5_esw_bridge_mdb_init(bridge);
+	if (err)
+		goto err_mdb_ht;
+
 	INIT_LIST_HEAD(&bridge->fdb_list);
 	bridge->ifindex = ifindex;
 	bridge->refcnt = 1;
@@ -849,6 +853,8 @@ static struct mlx5_esw_bridge *mlx5_esw_bridge_create(int ifindex,
 
 	return bridge;
 
+err_mdb_ht:
+	rhashtable_destroy(&bridge->fdb_ht);
 err_fdb_ht:
 	mlx5_esw_bridge_egress_table_cleanup(bridge);
 err_egress_tbl:
@@ -870,6 +876,7 @@ static void mlx5_esw_bridge_put(struct mlx5_esw_bridge_offloads *br_offloads,
 	mlx5_esw_bridge_egress_table_cleanup(bridge);
 	mlx5_esw_bridge_mcast_disable(bridge);
 	list_del(&bridge->list);
+	mlx5_esw_bridge_mdb_cleanup(bridge);
 	rhashtable_destroy(&bridge->fdb_ht);
 	kvfree(bridge);
 
@@ -909,7 +916,7 @@ static unsigned long mlx5_esw_bridge_port_key_from_data(u16 vport_num, u16 esw_o
 	return vport_num | (unsigned long)esw_owner_vhca_id << sizeof(vport_num) * BITS_PER_BYTE;
 }
 
-static unsigned long mlx5_esw_bridge_port_key(struct mlx5_esw_bridge_port *port)
+unsigned long mlx5_esw_bridge_port_key(struct mlx5_esw_bridge_port *port)
 {
 	return mlx5_esw_bridge_port_key_from_data(port->vport_num, port->esw_owner_vhca_id);
 }
@@ -1192,7 +1199,8 @@ static void mlx5_esw_bridge_vlan_erase(struct mlx5_esw_bridge_port *port,
 	xa_erase(&port->vlans, vlan->vid);
 }
 
-static void mlx5_esw_bridge_vlan_flush(struct mlx5_esw_bridge_vlan *vlan,
+static void mlx5_esw_bridge_vlan_flush(struct mlx5_esw_bridge_port *port,
+				       struct mlx5_esw_bridge_vlan *vlan,
 				       struct mlx5_esw_bridge *bridge)
 {
 	struct mlx5_eswitch *esw = bridge->br_offloads->esw;
@@ -1200,6 +1208,7 @@ static void mlx5_esw_bridge_vlan_flush(struct mlx5_esw_bridge_vlan *vlan,
 
 	list_for_each_entry_safe(entry, tmp, &vlan->fdb_list, vlan_list)
 		mlx5_esw_bridge_fdb_entry_notify_and_cleanup(entry, bridge);
+	mlx5_esw_bridge_port_mdb_vlan_flush(port, vlan);
 
 	if (vlan->mcast_handle)
 		mlx5_esw_bridge_vlan_push_pop_fhs_cleanup(vlan);
@@ -1216,7 +1225,7 @@ static void mlx5_esw_bridge_vlan_cleanup(struct mlx5_esw_bridge_port *port,
 					 struct mlx5_esw_bridge *bridge)
 {
 	trace_mlx5_esw_bridge_vlan_cleanup(vlan);
-	mlx5_esw_bridge_vlan_flush(vlan, bridge);
+	mlx5_esw_bridge_vlan_flush(port, vlan, bridge);
 	mlx5_esw_bridge_vlan_erase(port, vlan);
 	kvfree(vlan);
 }
@@ -1240,7 +1249,7 @@ static int mlx5_esw_bridge_port_vlans_recreate(struct mlx5_esw_bridge_port *port
 	int err;
 
 	xa_for_each(&port->vlans, i, vlan) {
-		mlx5_esw_bridge_vlan_flush(vlan, bridge);
+		mlx5_esw_bridge_vlan_flush(port, vlan, bridge);
 		err = mlx5_esw_bridge_vlan_push_pop_create(bridge->vlan_proto, vlan->flags, port,
 							   vlan, br_offloads->esw);
 		if (err) {
@@ -1450,6 +1459,7 @@ int mlx5_esw_bridge_vlan_filtering_set(u16 vport_num, u16 esw_owner_vhca_id, boo
 		return 0;
 
 	mlx5_esw_bridge_fdb_flush(bridge);
+	mlx5_esw_bridge_mdb_flush(bridge);
 	if (enable)
 		bridge->flags |= MLX5_ESW_BRIDGE_VLAN_FILTERING_FLAG;
 	else
@@ -1476,6 +1486,7 @@ int mlx5_esw_bridge_vlan_proto_set(u16 vport_num, u16 esw_owner_vhca_id, u16 pro
 	}
 
 	mlx5_esw_bridge_fdb_flush(bridge);
+	mlx5_esw_bridge_mdb_flush(bridge);
 	bridge->vlan_proto = proto;
 	mlx5_esw_bridge_vlans_recreate(bridge);
 
@@ -1792,6 +1803,62 @@ void mlx5_esw_bridge_update(struct mlx5_esw_bridge_offloads *br_offloads)
 	}
 }
 
+int mlx5_esw_bridge_port_mdb_add(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
+				 u16 vid, struct mlx5_esw_bridge_offloads *br_offloads,
+				 struct netlink_ext_ack *extack)
+{
+	struct mlx5_esw_bridge_vlan *vlan;
+	struct mlx5_esw_bridge_port *port;
+	struct mlx5_esw_bridge *bridge;
+	int err;
+
+	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!port) {
+		esw_warn(br_offloads->esw->dev,
+			 "Failed to lookup bridge port to add MDB (MAC=%pM,vport=%u)\n",
+			 addr, vport_num);
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "Failed to lookup bridge port to add MDB (MAC=%pM,vport=%u)\n",
+				       addr, vport_num);
+		return -EINVAL;
+	}
+
+	bridge = port->bridge;
+	if (bridge->flags & MLX5_ESW_BRIDGE_VLAN_FILTERING_FLAG && vid) {
+		vlan = mlx5_esw_bridge_vlan_lookup(vid, port);
+		if (!vlan) {
+			esw_warn(br_offloads->esw->dev,
+				 "Failed to lookup bridge port vlan metadata to create MDB (MAC=%pM,vid=%u,vport=%u)\n",
+				 addr, vid, vport_num);
+			NL_SET_ERR_MSG_FMT_MOD(extack,
+					       "Failed to lookup bridge port vlan metadata to create MDB (MAC=%pM,vid=%u,vport=%u)\n",
+					       addr, vid, vport_num);
+			return -EINVAL;
+		}
+	}
+
+	err = mlx5_esw_bridge_port_mdb_attach(port, addr, vid);
+	if (err) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Failed to add MDB (MAC=%pM,vid=%u,vport=%u)\n",
+				       addr, vid, vport_num);
+		return err;
+	}
+
+	return 0;
+}
+
+void mlx5_esw_bridge_port_mdb_del(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
+				  u16 vid, struct mlx5_esw_bridge_offloads *br_offloads)
+{
+	struct mlx5_esw_bridge_port *port;
+
+	port = mlx5_esw_bridge_port_lookup(vport_num, esw_owner_vhca_id, br_offloads);
+	if (!port)
+		return;
+
+	mlx5_esw_bridge_port_mdb_detach(port, addr, vid);
+}
+
 static void mlx5_esw_bridge_flush(struct mlx5_esw_bridge_offloads *br_offloads)
 {
 	struct mlx5_esw_bridge_port *port;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
index b18f137173d9..9cab66467289 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
@@ -79,4 +79,10 @@ int mlx5_esw_bridge_port_vlan_add(u16 vport_num, u16 esw_owner_vhca_id, u16 vid,
 void mlx5_esw_bridge_port_vlan_del(u16 vport_num, u16 esw_owner_vhca_id, u16 vid,
 				   struct mlx5_esw_bridge_offloads *br_offloads);
 
+int mlx5_esw_bridge_port_mdb_add(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
+				 u16 vid, struct mlx5_esw_bridge_offloads *br_offloads,
+				 struct netlink_ext_ack *extack);
+void mlx5_esw_bridge_port_mdb_del(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
+				  u16 vid, struct mlx5_esw_bridge_offloads *br_offloads);
+
 #endif /* __MLX5_ESW_BRIDGE_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
index 99e2f9fc11a2..d17fe6d374b5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
@@ -6,6 +6,300 @@
 #include "eswitch.h"
 #include "bridge_priv.h"
 
+static const struct rhashtable_params mdb_ht_params = {
+	.key_offset = offsetof(struct mlx5_esw_bridge_mdb_entry, key),
+	.key_len = sizeof(struct mlx5_esw_bridge_mdb_key),
+	.head_offset = offsetof(struct mlx5_esw_bridge_mdb_entry, ht_node),
+	.automatic_shrinking = true,
+};
+
+int mlx5_esw_bridge_mdb_init(struct mlx5_esw_bridge *bridge)
+{
+	INIT_LIST_HEAD(&bridge->mdb_list);
+	return rhashtable_init(&bridge->mdb_ht, &mdb_ht_params);
+}
+
+void mlx5_esw_bridge_mdb_cleanup(struct mlx5_esw_bridge *bridge)
+{
+	rhashtable_destroy(&bridge->mdb_ht);
+}
+
+static struct mlx5_esw_bridge_port *
+mlx5_esw_bridge_mdb_port_lookup(struct mlx5_esw_bridge_port *port,
+				struct mlx5_esw_bridge_mdb_entry *entry)
+{
+	return xa_load(&entry->ports, mlx5_esw_bridge_port_key(port));
+}
+
+static int mlx5_esw_bridge_mdb_port_insert(struct mlx5_esw_bridge_port *port,
+					   struct mlx5_esw_bridge_mdb_entry *entry)
+{
+	int err = xa_insert(&entry->ports, mlx5_esw_bridge_port_key(port), port, GFP_KERNEL);
+
+	if (!err)
+		entry->num_ports++;
+	return err;
+}
+
+static void mlx5_esw_bridge_mdb_port_remove(struct mlx5_esw_bridge_port *port,
+					    struct mlx5_esw_bridge_mdb_entry *entry)
+{
+	xa_erase(&entry->ports, mlx5_esw_bridge_port_key(port));
+	entry->num_ports--;
+}
+
+static struct mlx5_flow_handle *
+mlx5_esw_bridge_mdb_flow_create(u16 esw_owner_vhca_id, struct mlx5_esw_bridge_mdb_entry *entry,
+				struct mlx5_esw_bridge *bridge)
+{
+	struct mlx5_flow_act flow_act = {
+		.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST,
+		.flags = FLOW_ACT_NO_APPEND | FLOW_ACT_IGNORE_FLOW_LEVEL,
+	};
+	int num_dests = entry->num_ports, i = 0;
+	struct mlx5_flow_destination *dests;
+	struct mlx5_esw_bridge_port *port;
+	struct mlx5_flow_spec *rule_spec;
+	struct mlx5_flow_handle *handle;
+	u8 *dmac_v, *dmac_c;
+	unsigned long idx;
+
+	rule_spec = kvzalloc(sizeof(*rule_spec), GFP_KERNEL);
+	if (!rule_spec)
+		return ERR_PTR(-ENOMEM);
+
+	dests = kvcalloc(num_dests, sizeof(*dests), GFP_KERNEL);
+	if (!dests) {
+		kvfree(rule_spec);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	xa_for_each(&entry->ports, idx, port) {
+		dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+		dests[i].ft = port->mcast.ft;
+		i++;
+	}
+
+	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
+	dmac_v = MLX5_ADDR_OF(fte_match_param, rule_spec->match_value, outer_headers.dmac_47_16);
+	ether_addr_copy(dmac_v, entry->key.addr);
+	dmac_c = MLX5_ADDR_OF(fte_match_param, rule_spec->match_criteria, outer_headers.dmac_47_16);
+	eth_broadcast_addr(dmac_c);
+
+	if (entry->key.vid) {
+		if (bridge->vlan_proto == ETH_P_8021Q) {
+			MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria,
+					 outer_headers.cvlan_tag);
+			MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_value,
+					 outer_headers.cvlan_tag);
+		} else if (bridge->vlan_proto == ETH_P_8021AD) {
+			MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria,
+					 outer_headers.svlan_tag);
+			MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_value,
+					 outer_headers.svlan_tag);
+		}
+		MLX5_SET_TO_ONES(fte_match_param, rule_spec->match_criteria,
+				 outer_headers.first_vid);
+		MLX5_SET(fte_match_param, rule_spec->match_value, outer_headers.first_vid,
+			 entry->key.vid);
+	}
+
+	handle = mlx5_add_flow_rules(bridge->egress_ft, rule_spec, &flow_act, dests, num_dests);
+
+	kvfree(dests);
+	kvfree(rule_spec);
+	return handle;
+}
+
+static int
+mlx5_esw_bridge_port_mdb_offload(struct mlx5_esw_bridge_port *port,
+				 struct mlx5_esw_bridge_mdb_entry *entry)
+{
+	struct mlx5_flow_handle *handle;
+
+	handle = mlx5_esw_bridge_mdb_flow_create(port->esw_owner_vhca_id, entry, port->bridge);
+	if (entry->egress_handle) {
+		mlx5_del_flow_rules(entry->egress_handle);
+		entry->egress_handle = NULL;
+	}
+	if (IS_ERR(handle))
+		return PTR_ERR(handle);
+
+	entry->egress_handle = handle;
+	return 0;
+}
+
+static struct mlx5_esw_bridge_mdb_entry *
+mlx5_esw_bridge_mdb_lookup(struct mlx5_esw_bridge *bridge,
+			   const unsigned char *addr, u16 vid)
+{
+	struct mlx5_esw_bridge_mdb_key key = {};
+
+	ether_addr_copy(key.addr, addr);
+	key.vid = vid;
+	return rhashtable_lookup_fast(&bridge->mdb_ht, &key, mdb_ht_params);
+}
+
+static struct mlx5_esw_bridge_mdb_entry *
+mlx5_esw_bridge_port_mdb_entry_init(struct mlx5_esw_bridge_port *port,
+				    const unsigned char *addr, u16 vid)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_esw_bridge_mdb_entry *entry;
+	int err;
+
+	entry = kvzalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		return ERR_PTR(-ENOMEM);
+
+	ether_addr_copy(entry->key.addr, addr);
+	entry->key.vid = vid;
+	xa_init(&entry->ports);
+	err = rhashtable_insert_fast(&bridge->mdb_ht, &entry->ht_node, mdb_ht_params);
+	if (err)
+		goto err_ht_insert;
+
+	list_add(&entry->list, &bridge->mdb_list);
+
+	return entry;
+
+err_ht_insert:
+	xa_destroy(&entry->ports);
+	kvfree(entry);
+	return ERR_PTR(err);
+}
+
+static void mlx5_esw_bridge_port_mdb_entry_cleanup(struct mlx5_esw_bridge *bridge,
+						   struct mlx5_esw_bridge_mdb_entry *entry)
+{
+	if (entry->egress_handle)
+		mlx5_del_flow_rules(entry->egress_handle);
+	list_del(&entry->list);
+	rhashtable_remove_fast(&bridge->mdb_ht, &entry->ht_node, mdb_ht_params);
+	xa_destroy(&entry->ports);
+	kvfree(entry);
+}
+
+int mlx5_esw_bridge_port_mdb_attach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
+				    u16 vid)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_esw_bridge_mdb_entry *entry;
+	int err;
+
+	if (!(bridge->flags & MLX5_ESW_BRIDGE_MCAST_FLAG))
+		return -EOPNOTSUPP;
+
+	entry = mlx5_esw_bridge_mdb_lookup(bridge, addr, vid);
+	if (entry) {
+		if (mlx5_esw_bridge_mdb_port_lookup(port, entry)) {
+			esw_warn(bridge->br_offloads->esw->dev, "MDB attach entry is already attached to port (MAC=%pM,vid=%u,vport=%u)\n",
+				 addr, vid, port->vport_num);
+			return 0;
+		}
+	} else {
+		entry = mlx5_esw_bridge_port_mdb_entry_init(port, addr, vid);
+		if (IS_ERR(entry)) {
+			err = PTR_ERR(entry);
+			esw_warn(bridge->br_offloads->esw->dev, "MDB attach failed to init entry (MAC=%pM,vid=%u,vport=%u,err=%d)\n",
+				 addr, vid, port->vport_num, err);
+			return err;
+		}
+	}
+
+	err = mlx5_esw_bridge_mdb_port_insert(port, entry);
+	if (err) {
+		if (!entry->num_ports)
+			mlx5_esw_bridge_port_mdb_entry_cleanup(bridge, entry); /* new mdb entry */
+		esw_warn(bridge->br_offloads->esw->dev,
+			 "MDB attach failed to insert port (MAC=%pM,vid=%u,vport=%u,err=%d)\n",
+			 addr, vid, port->vport_num, err);
+		return err;
+	}
+
+	err = mlx5_esw_bridge_port_mdb_offload(port, entry);
+	if (err)
+		/* Single mdb can be used by multiple ports, so just log the
+		 * error and continue.
+		 */
+		esw_warn(bridge->br_offloads->esw->dev, "MDB attach failed to offload (MAC=%pM,vid=%u,vport=%u,err=%d)\n",
+			 addr, vid, port->vport_num, err);
+	return 0;
+}
+
+static void mlx5_esw_bridge_port_mdb_entry_detach(struct mlx5_esw_bridge_port *port,
+						  struct mlx5_esw_bridge_mdb_entry *entry)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	int err;
+
+	mlx5_esw_bridge_mdb_port_remove(port, entry);
+	if (!entry->num_ports) {
+		mlx5_esw_bridge_port_mdb_entry_cleanup(bridge, entry);
+		return;
+	}
+
+	err = mlx5_esw_bridge_port_mdb_offload(port, entry);
+	if (err)
+		/* Single mdb can be used by multiple ports, so just log the
+		 * error and continue.
+		 */
+		esw_warn(bridge->br_offloads->esw->dev, "MDB detach failed to offload (MAC=%pM,vid=%u,vport=%u)\n",
+			 entry->key.addr, entry->key.vid, port->vport_num);
+}
+
+void mlx5_esw_bridge_port_mdb_detach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
+				     u16 vid)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_esw_bridge_mdb_entry *entry;
+
+	entry = mlx5_esw_bridge_mdb_lookup(bridge, addr, vid);
+	if (!entry) {
+		esw_debug(bridge->br_offloads->esw->dev,
+			  "MDB detach entry not found (MAC=%pM,vid=%u,vport=%u)\n",
+			  addr, vid, port->vport_num);
+		return;
+	}
+
+	if (!mlx5_esw_bridge_mdb_port_lookup(port, entry)) {
+		esw_debug(bridge->br_offloads->esw->dev,
+			  "MDB detach entry not attached to the port (MAC=%pM,vid=%u,vport=%u)\n",
+			  addr, vid, port->vport_num);
+		return;
+	}
+
+	mlx5_esw_bridge_port_mdb_entry_detach(port, entry);
+}
+
+void mlx5_esw_bridge_port_mdb_vlan_flush(struct mlx5_esw_bridge_port *port,
+					 struct mlx5_esw_bridge_vlan *vlan)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_esw_bridge_mdb_entry *entry, *tmp;
+
+	list_for_each_entry_safe(entry, tmp, &bridge->mdb_list, list)
+		if (entry->key.vid == vlan->vid && mlx5_esw_bridge_mdb_port_lookup(port, entry))
+			mlx5_esw_bridge_port_mdb_entry_detach(port, entry);
+}
+
+static void mlx5_esw_bridge_port_mdb_flush(struct mlx5_esw_bridge_port *port)
+{
+	struct mlx5_esw_bridge *bridge = port->bridge;
+	struct mlx5_esw_bridge_mdb_entry *entry, *tmp;
+
+	list_for_each_entry_safe(entry, tmp, &bridge->mdb_list, list)
+		if (mlx5_esw_bridge_mdb_port_lookup(port, entry))
+			mlx5_esw_bridge_port_mdb_entry_detach(port, entry);
+}
+
+void mlx5_esw_bridge_mdb_flush(struct mlx5_esw_bridge *bridge)
+{
+	struct mlx5_esw_bridge_mdb_entry *entry, *tmp;
+
+	list_for_each_entry_safe(entry, tmp, &bridge->mdb_list, list)
+		mlx5_esw_bridge_port_mdb_entry_cleanup(bridge, entry);
+}
 static int mlx5_esw_bridge_port_mcast_fts_init(struct mlx5_esw_bridge_port *port,
 					       struct mlx5_esw_bridge *bridge)
 {
@@ -457,6 +751,7 @@ err_fgs:
 
 void mlx5_esw_bridge_port_mcast_cleanup(struct mlx5_esw_bridge_port *port)
 {
+	mlx5_esw_bridge_port_mdb_flush(port);
 	mlx5_esw_bridge_port_mcast_fhs_cleanup(port);
 	mlx5_esw_bridge_port_mcast_fgs_cleanup(port);
 	mlx5_esw_bridge_port_mcast_fts_cleanup(port);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
index 36ff32001ce8..849028f94be2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
@@ -125,6 +125,11 @@ struct mlx5_esw_bridge_fdb_key {
 	u16 vid;
 };
 
+struct mlx5_esw_bridge_mdb_key {
+	unsigned char addr[ETH_ALEN];
+	u16 vid;
+};
+
 enum {
 	MLX5_ESW_BRIDGE_FLAG_ADDED_BY_USER = BIT(0),
 	MLX5_ESW_BRIDGE_FLAG_PEER = BIT(1),
@@ -151,6 +156,16 @@ struct mlx5_esw_bridge_fdb_entry {
 	struct mlx5_flow_handle *filter_handle;
 };
 
+struct mlx5_esw_bridge_mdb_entry {
+	struct mlx5_esw_bridge_mdb_key key;
+	struct rhash_head ht_node;
+	struct list_head list;
+	struct xarray ports;
+	int num_ports;
+
+	struct mlx5_flow_handle *egress_handle;
+};
+
 struct mlx5_esw_bridge_vlan {
 	u16 vid;
 	u16 flags;
@@ -188,6 +203,9 @@ struct mlx5_esw_bridge {
 	struct list_head fdb_list;
 	struct rhashtable fdb_ht;
 
+	struct list_head mdb_list;
+	struct rhashtable mdb_ht;
+
 	struct mlx5_flow_table *egress_ft;
 	struct mlx5_flow_group *egress_vlan_fg;
 	struct mlx5_flow_group *egress_qinq_fg;
@@ -202,6 +220,7 @@ struct mlx5_esw_bridge {
 
 struct mlx5_flow_table *mlx5_esw_bridge_table_create(int max_fte, u32 level,
 						     struct mlx5_eswitch *esw);
+unsigned long mlx5_esw_bridge_port_key(struct mlx5_esw_bridge_port *port);
 
 int mlx5_esw_bridge_port_mcast_init(struct mlx5_esw_bridge_port *port);
 void mlx5_esw_bridge_port_mcast_cleanup(struct mlx5_esw_bridge_port *port);
@@ -212,4 +231,14 @@ void mlx5_esw_bridge_vlan_mcast_cleanup(struct mlx5_esw_bridge_vlan *vlan);
 int mlx5_esw_bridge_mcast_enable(struct mlx5_esw_bridge *bridge);
 void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge);
 
+int mlx5_esw_bridge_mdb_init(struct mlx5_esw_bridge *bridge);
+void mlx5_esw_bridge_mdb_cleanup(struct mlx5_esw_bridge *bridge);
+int mlx5_esw_bridge_port_mdb_attach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
+				    u16 vid);
+void mlx5_esw_bridge_port_mdb_detach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
+				     u16 vid);
+void mlx5_esw_bridge_port_mdb_vlan_flush(struct mlx5_esw_bridge_port *port,
+					 struct mlx5_esw_bridge_vlan *vlan);
+void mlx5_esw_bridge_mdb_flush(struct mlx5_esw_bridge *bridge);
+
 #endif /* _MLX5_ESW_BRIDGE_PRIVATE_ */
-- 
cgit v1.2.3-70-g09d2


From 55f3e740f7f697a2287447678c09d93238a6635a Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Tue, 14 Feb 2023 22:00:41 +0100
Subject: net/mlx5: Bridge, add tracepoints for multicast

Pass target struct net_device to mdb attach/detach handler in order to
expose the port name to the new tracepoints. Implemented following
tracepoints:

- Attach mdb to port.
- Detach mdb from port.

Usage example:

># cd /sys/kernel/debug/tracing
># echo mlx5:mlx5_esw_bridge_port_mdb_attach >> set_event
># cat trace
...
     kworker/0:0-19071   [000] ..... 259004.253848: mlx5_esw_bridge_port_mdb_attach: net_device=enp8s0f0_0 addr=33:33:ff:00:00:01 vid=0 num_ports=1 offloaded=1

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/rep/bridge.c    |  4 +--
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   | 14 +++++----
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.h   | 10 ++++---
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c | 12 +++++---
 .../ethernet/mellanox/mlx5/core/esw/bridge_priv.h  |  8 ++---
 .../mlx5/core/esw/diag/bridge_tracepoint.h         | 35 ++++++++++++++++++++++
 6 files changed, 63 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
index dd0dd3f028a3..fd191925ab4b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
@@ -238,7 +238,7 @@ mlx5_esw_bridge_port_obj_add(struct net_device *dev,
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_MDB:
 		mdb = SWITCHDEV_OBJ_PORT_MDB(obj);
-		err = mlx5_esw_bridge_port_mdb_add(vport_num, esw_owner_vhca_id, mdb->addr,
+		err = mlx5_esw_bridge_port_mdb_add(dev, vport_num, esw_owner_vhca_id, mdb->addr,
 						   mdb->vid, br_offloads, extack);
 		break;
 	default:
@@ -270,7 +270,7 @@ mlx5_esw_bridge_port_obj_del(struct net_device *dev,
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_MDB:
 		mdb = SWITCHDEV_OBJ_PORT_MDB(obj);
-		mlx5_esw_bridge_port_mdb_del(vport_num, esw_owner_vhca_id, mdb->addr, mdb->vid,
+		mlx5_esw_bridge_port_mdb_del(dev, vport_num, esw_owner_vhca_id, mdb->addr, mdb->vid,
 					     br_offloads);
 		break;
 	default:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index be4787539c6c..1ba03e219111 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -1803,8 +1803,9 @@ void mlx5_esw_bridge_update(struct mlx5_esw_bridge_offloads *br_offloads)
 	}
 }
 
-int mlx5_esw_bridge_port_mdb_add(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
-				 u16 vid, struct mlx5_esw_bridge_offloads *br_offloads,
+int mlx5_esw_bridge_port_mdb_add(struct net_device *dev, u16 vport_num, u16 esw_owner_vhca_id,
+				 const unsigned char *addr, u16 vid,
+				 struct mlx5_esw_bridge_offloads *br_offloads,
 				 struct netlink_ext_ack *extack)
 {
 	struct mlx5_esw_bridge_vlan *vlan;
@@ -1837,7 +1838,7 @@ int mlx5_esw_bridge_port_mdb_add(u16 vport_num, u16 esw_owner_vhca_id, const uns
 		}
 	}
 
-	err = mlx5_esw_bridge_port_mdb_attach(port, addr, vid);
+	err = mlx5_esw_bridge_port_mdb_attach(dev, port, addr, vid);
 	if (err) {
 		NL_SET_ERR_MSG_FMT_MOD(extack, "Failed to add MDB (MAC=%pM,vid=%u,vport=%u)\n",
 				       addr, vid, vport_num);
@@ -1847,8 +1848,9 @@ int mlx5_esw_bridge_port_mdb_add(u16 vport_num, u16 esw_owner_vhca_id, const uns
 	return 0;
 }
 
-void mlx5_esw_bridge_port_mdb_del(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
-				  u16 vid, struct mlx5_esw_bridge_offloads *br_offloads)
+void mlx5_esw_bridge_port_mdb_del(struct net_device *dev, u16 vport_num, u16 esw_owner_vhca_id,
+				  const unsigned char *addr, u16 vid,
+				  struct mlx5_esw_bridge_offloads *br_offloads)
 {
 	struct mlx5_esw_bridge_port *port;
 
@@ -1856,7 +1858,7 @@ void mlx5_esw_bridge_port_mdb_del(u16 vport_num, u16 esw_owner_vhca_id, const un
 	if (!port)
 		return;
 
-	mlx5_esw_bridge_port_mdb_detach(port, addr, vid);
+	mlx5_esw_bridge_port_mdb_detach(dev, port, addr, vid);
 }
 
 static void mlx5_esw_bridge_flush(struct mlx5_esw_bridge_offloads *br_offloads)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
index 9cab66467289..a9dd18c73d6a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.h
@@ -79,10 +79,12 @@ int mlx5_esw_bridge_port_vlan_add(u16 vport_num, u16 esw_owner_vhca_id, u16 vid,
 void mlx5_esw_bridge_port_vlan_del(u16 vport_num, u16 esw_owner_vhca_id, u16 vid,
 				   struct mlx5_esw_bridge_offloads *br_offloads);
 
-int mlx5_esw_bridge_port_mdb_add(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
-				 u16 vid, struct mlx5_esw_bridge_offloads *br_offloads,
+int mlx5_esw_bridge_port_mdb_add(struct net_device *dev, u16 vport_num, u16 esw_owner_vhca_id,
+				 const unsigned char *addr, u16 vid,
+				 struct mlx5_esw_bridge_offloads *br_offloads,
 				 struct netlink_ext_ack *extack);
-void mlx5_esw_bridge_port_mdb_del(u16 vport_num, u16 esw_owner_vhca_id, const unsigned char *addr,
-				  u16 vid, struct mlx5_esw_bridge_offloads *br_offloads);
+void mlx5_esw_bridge_port_mdb_del(struct net_device *dev, u16 vport_num, u16 esw_owner_vhca_id,
+				  const unsigned char *addr, u16 vid,
+				  struct mlx5_esw_bridge_offloads *br_offloads);
 
 #endif /* __MLX5_ESW_BRIDGE_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
index d17fe6d374b5..2eae594a5e80 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
@@ -5,6 +5,7 @@
 #include "bridge.h"
 #include "eswitch.h"
 #include "bridge_priv.h"
+#include "diag/bridge_tracepoint.h"
 
 static const struct rhashtable_params mdb_ht_params = {
 	.key_offset = offsetof(struct mlx5_esw_bridge_mdb_entry, key),
@@ -180,8 +181,8 @@ static void mlx5_esw_bridge_port_mdb_entry_cleanup(struct mlx5_esw_bridge *bridg
 	kvfree(entry);
 }
 
-int mlx5_esw_bridge_port_mdb_attach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
-				    u16 vid)
+int mlx5_esw_bridge_port_mdb_attach(struct net_device *dev, struct mlx5_esw_bridge_port *port,
+				    const unsigned char *addr, u16 vid)
 {
 	struct mlx5_esw_bridge *bridge = port->bridge;
 	struct mlx5_esw_bridge_mdb_entry *entry;
@@ -224,6 +225,8 @@ int mlx5_esw_bridge_port_mdb_attach(struct mlx5_esw_bridge_port *port, const uns
 		 */
 		esw_warn(bridge->br_offloads->esw->dev, "MDB attach failed to offload (MAC=%pM,vid=%u,vport=%u,err=%d)\n",
 			 addr, vid, port->vport_num, err);
+
+	trace_mlx5_esw_bridge_port_mdb_attach(dev, entry);
 	return 0;
 }
 
@@ -248,8 +251,8 @@ static void mlx5_esw_bridge_port_mdb_entry_detach(struct mlx5_esw_bridge_port *p
 			 entry->key.addr, entry->key.vid, port->vport_num);
 }
 
-void mlx5_esw_bridge_port_mdb_detach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
-				     u16 vid)
+void mlx5_esw_bridge_port_mdb_detach(struct net_device *dev, struct mlx5_esw_bridge_port *port,
+				     const unsigned char *addr, u16 vid)
 {
 	struct mlx5_esw_bridge *bridge = port->bridge;
 	struct mlx5_esw_bridge_mdb_entry *entry;
@@ -269,6 +272,7 @@ void mlx5_esw_bridge_port_mdb_detach(struct mlx5_esw_bridge_port *port, const un
 		return;
 	}
 
+	trace_mlx5_esw_bridge_port_mdb_detach(dev, entry);
 	mlx5_esw_bridge_port_mdb_entry_detach(port, entry);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
index 849028f94be2..c9595801bdb4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_priv.h
@@ -233,10 +233,10 @@ void mlx5_esw_bridge_mcast_disable(struct mlx5_esw_bridge *bridge);
 
 int mlx5_esw_bridge_mdb_init(struct mlx5_esw_bridge *bridge);
 void mlx5_esw_bridge_mdb_cleanup(struct mlx5_esw_bridge *bridge);
-int mlx5_esw_bridge_port_mdb_attach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
-				    u16 vid);
-void mlx5_esw_bridge_port_mdb_detach(struct mlx5_esw_bridge_port *port, const unsigned char *addr,
-				     u16 vid);
+int mlx5_esw_bridge_port_mdb_attach(struct net_device *dev, struct mlx5_esw_bridge_port *port,
+				    const unsigned char *addr, u16 vid);
+void mlx5_esw_bridge_port_mdb_detach(struct net_device *dev, struct mlx5_esw_bridge_port *port,
+				     const unsigned char *addr, u16 vid);
 void mlx5_esw_bridge_port_mdb_vlan_flush(struct mlx5_esw_bridge_port *port,
 					 struct mlx5_esw_bridge_vlan *vlan);
 void mlx5_esw_bridge_mdb_flush(struct mlx5_esw_bridge *bridge);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/bridge_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/bridge_tracepoint.h
index 51ac24e6ec3c..1808da214094 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/bridge_tracepoint.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/bridge_tracepoint.h
@@ -110,6 +110,41 @@ DEFINE_EVENT(mlx5_esw_bridge_port_template,
 	     TP_ARGS(port)
 	);
 
+DECLARE_EVENT_CLASS(mlx5_esw_bridge_mdb_port_change_template,
+		    TP_PROTO(const struct net_device *dev,
+			     const struct mlx5_esw_bridge_mdb_entry *mdb),
+		    TP_ARGS(dev, mdb),
+		    TP_STRUCT__entry(
+			    __array(char, dev_name, IFNAMSIZ)
+			    __array(unsigned char, addr, ETH_ALEN)
+			    __field(u16, vid)
+			    __field(int, num_ports)
+			    __field(bool, offloaded)),
+		    TP_fast_assign(
+			    strscpy(__entry->dev_name, netdev_name(dev), IFNAMSIZ);
+			    memcpy(__entry->addr, mdb->key.addr, ETH_ALEN);
+			    __entry->vid = mdb->key.vid;
+			    __entry->num_ports = mdb->num_ports;
+			    __entry->offloaded = mdb->egress_handle;),
+		    TP_printk("net_device=%s addr=%pM vid=%u num_ports=%d offloaded=%d",
+			      __entry->dev_name,
+			      __entry->addr,
+			      __entry->vid,
+			      __entry->num_ports,
+			      __entry->offloaded));
+
+DEFINE_EVENT(mlx5_esw_bridge_mdb_port_change_template,
+	     mlx5_esw_bridge_port_mdb_attach,
+	     TP_PROTO(const struct net_device *dev,
+		      const struct mlx5_esw_bridge_mdb_entry *mdb),
+	     TP_ARGS(dev, mdb));
+
+DEFINE_EVENT(mlx5_esw_bridge_mdb_port_change_template,
+	     mlx5_esw_bridge_port_mdb_detach,
+	     TP_PROTO(const struct net_device *dev,
+		      const struct mlx5_esw_bridge_mdb_entry *mdb),
+	     TP_ARGS(dev, mdb));
+
 #endif
 
 /* This part must be outside protection */
-- 
cgit v1.2.3-70-g09d2


From 9df839a711aee437390b16ee39cf0b5c1620be6a Mon Sep 17 00:00:00 2001
From: Parav Pandit <parav@mellanox.com>
Date: Thu, 23 Apr 2020 08:27:59 -0500
Subject: net/mlx5: Create a new profile for SFs

Create a new profile for SFs in order to disable the command cache.
Each function command cache consumes ~500KB of memory, when using a
large number of SFs this savings is notable on memory constarined
systems.

Use a new profile to provide for future differences between SFs and PFs.

The mr_cache not used for non-PF functions, so it is excluded from the
new profile.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Bodong Wang <bodong@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c           | 6 +++---
 drivers/net/ethernet/mellanox/mlx5/core/main.c          | 9 +++++++++
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h     | 1 +
 drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c | 2 +-
 include/linux/mlx5/driver.h                             | 1 +
 5 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index b00e33ed05e9..d53de39539a8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -1802,7 +1802,7 @@ static struct mlx5_cmd_msg *alloc_msg(struct mlx5_core_dev *dev, int in_size,
 	if (in_size <= 16)
 		goto cache_miss;
 
-	for (i = 0; i < MLX5_NUM_COMMAND_CACHES; i++) {
+	for (i = 0; i < dev->profile.num_cmd_caches; i++) {
 		ch = &cmd->cache[i];
 		if (in_size > ch->max_inbox_size)
 			continue;
@@ -2097,7 +2097,7 @@ static void destroy_msg_cache(struct mlx5_core_dev *dev)
 	struct mlx5_cmd_msg *n;
 	int i;
 
-	for (i = 0; i < MLX5_NUM_COMMAND_CACHES; i++) {
+	for (i = 0; i < dev->profile.num_cmd_caches; i++) {
 		ch = &dev->cmd.cache[i];
 		list_for_each_entry_safe(msg, n, &ch->head, list) {
 			list_del(&msg->list);
@@ -2127,7 +2127,7 @@ static void create_msg_cache(struct mlx5_core_dev *dev)
 	int k;
 
 	/* Initialize and fill the caches with initial entries */
-	for (k = 0; k < MLX5_NUM_COMMAND_CACHES; k++) {
+	for (k = 0; k < dev->profile.num_cmd_caches; k++) {
 		ch = &cmd->cache[k];
 		spin_lock_init(&ch->lock);
 		INIT_LIST_HEAD(&ch->head);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index f95df73d1089..a95d1218def9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -100,15 +100,19 @@ enum {
 static struct mlx5_profile profile[] = {
 	[0] = {
 		.mask           = 0,
+		.num_cmd_caches = MLX5_NUM_COMMAND_CACHES,
 	},
 	[1] = {
 		.mask		= MLX5_PROF_MASK_QP_SIZE,
 		.log_max_qp	= 12,
+		.num_cmd_caches = MLX5_NUM_COMMAND_CACHES,
+
 	},
 	[2] = {
 		.mask		= MLX5_PROF_MASK_QP_SIZE |
 				  MLX5_PROF_MASK_MR_CACHE,
 		.log_max_qp	= LOG_MAX_SUPPORTED_QPS,
+		.num_cmd_caches = MLX5_NUM_COMMAND_CACHES,
 		.mr_cache[0]	= {
 			.size	= 500,
 			.limit	= 250
@@ -174,6 +178,11 @@ static struct mlx5_profile profile[] = {
 			.limit	= 4
 		},
 	},
+	[3] = {
+		.mask		= MLX5_PROF_MASK_QP_SIZE,
+		.log_max_qp	= LOG_MAX_SUPPORTED_QPS,
+		.num_cmd_caches = 0,
+	},
 };
 
 static int wait_fw_init(struct mlx5_core_dev *dev, u32 max_wait_mili,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index be0785f83083..5eaab99678ee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -142,6 +142,7 @@ enum mlx5_semaphore_space_address {
 };
 
 #define MLX5_DEFAULT_PROF       2
+#define MLX5_SF_PROF		3
 
 static inline int mlx5_flexible_inlen(struct mlx5_core_dev *dev, size_t fixed,
 				      size_t item_size, size_t num_items,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
index a7377619ba6f..e2f26d0bc615 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c
@@ -28,7 +28,7 @@ static int mlx5_sf_dev_probe(struct auxiliary_device *adev, const struct auxilia
 	mdev->priv.adev_idx = adev->id;
 	sf_dev->mdev = mdev;
 
-	err = mlx5_mdev_init(mdev, MLX5_DEFAULT_PROF);
+	err = mlx5_mdev_init(mdev, MLX5_SF_PROF);
 	if (err) {
 		mlx5_core_warn(mdev, "mlx5_mdev_init on err=%d\n", err);
 		goto mdev_err;
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index f243bd10a5e1..135a3c8d8237 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -751,6 +751,7 @@ enum {
 struct mlx5_profile {
 	u64	mask;
 	u8	log_max_qp;
+	u8	num_cmd_caches;
 	struct {
 		int	size;
 		int	limit;
-- 
cgit v1.2.3-70-g09d2


From cee6484eddc1066f009cc82ad12a1739892172b8 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Wed, 1 Mar 2023 00:16:34 +0200
Subject: net/mlx5: DR, Set counter ID on the last STE for STEv1 TX

In STEv1 counter action can be set either by filling counter ID on STE, in
which case it is executed before other actions on this STE, or as a single
action, in which case it is executed in accordance with the actions order.
FW steering on STEv1 devices implements counter as counter ID on STE, and
this counter is set on the last STE.
Fix SMFS to be consistent with this behaviour - move TX counter to the
last STE, this way the counter will include all actions of the previous STEs
that might have changed packet headers length, e.g. encap, vlan push, etc.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
index 084145f18084..27cc6931bbde 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
@@ -604,9 +604,6 @@ void dr_ste_v1_set_actions_tx(struct mlx5dr_domain *dmn,
 			allow_modify_hdr = false;
 	}
 
-	if (action_type_set[DR_ACTION_TYP_CTR])
-		dr_ste_v1_set_counter_id(last_ste, attr->ctr_id);
-
 	if (action_type_set[DR_ACTION_TYP_MODIFY_HDR]) {
 		if (!allow_modify_hdr || action_sz < DR_STE_ACTION_DOUBLE_SZ) {
 			dr_ste_v1_arr_init_next_match(&last_ste, added_stes,
@@ -724,6 +721,10 @@ void dr_ste_v1_set_actions_tx(struct mlx5dr_domain *dmn,
 						  attr->range.max);
 	}
 
+	/* set counter ID on the last STE to adhere to DMFS behavior */
+	if (action_type_set[DR_ACTION_TYP_CTR])
+		dr_ste_v1_set_counter_id(last_ste, attr->ctr_id);
+
 	dr_ste_v1_set_hit_gvmi(last_ste, attr->hit_gvmi);
 	dr_ste_v1_set_hit_addr(last_ste, attr->final_icm_addr, 1);
 }
-- 
cgit v1.2.3-70-g09d2


From 9fa7f1de3dda0bc74c964b5127b3f22bd8ef8fd2 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:20:30 +0300
Subject: net/mlx5: Add mlx5_ifc bits for modify header argument

Add enum value for modify-header argument object and mlx5_bits
for the related capabilities.

Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 include/linux/mlx5/mlx5_ifc.h | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 02c628f4fe26..6c84bf6eec85 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -78,12 +78,15 @@ enum {
 
 enum {
 	MLX5_OBJ_TYPE_SW_ICM = 0x0008,
+	MLX5_OBJ_TYPE_HEADER_MODIFY_ARGUMENT  = 0x23,
 };
 
 enum {
 	MLX5_GENERAL_OBJ_TYPES_CAP_SW_ICM = (1ULL << MLX5_OBJ_TYPE_SW_ICM),
 	MLX5_GENERAL_OBJ_TYPES_CAP_GENEVE_TLV_OPT = (1ULL << 11),
 	MLX5_GENERAL_OBJ_TYPES_CAP_VIRTIO_NET_Q = (1ULL << 13),
+	MLX5_GENERAL_OBJ_TYPES_CAP_HEADER_MODIFY_ARGUMENT =
+		(1ULL << MLX5_OBJ_TYPE_HEADER_MODIFY_ARGUMENT),
 	MLX5_GENERAL_OBJ_TYPES_CAP_MACSEC_OFFLOAD = (1ULL << 39),
 };
 
@@ -321,6 +324,10 @@ enum {
 	MLX5_FT_NIC_TX_RDMA_2_NIC_TX = BIT(1),
 };
 
+enum {
+	MLX5_CMD_OP_MOD_UPDATE_HEADER_MODIFY_ARGUMENT = 0x1,
+};
+
 struct mlx5_ifc_flow_table_fields_supported_bits {
 	u8         outer_dmac[0x1];
 	u8         outer_smac[0x1];
@@ -1927,7 +1934,14 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8	   reserved_at_750[0x4];
 	u8	   max_dynamic_vf_msix_table_size[0xc];
 
-	u8	   reserved_at_760[0x20];
+	u8         reserved_at_760[0x3];
+	u8         log_max_num_header_modify_argument[0x5];
+	u8         reserved_at_768[0x4];
+	u8         log_header_modify_argument_granularity[0x4];
+	u8         reserved_at_770[0x3];
+	u8         log_header_modify_argument_max_alloc[0x5];
+	u8         reserved_at_778[0x8];
+
 	u8	   vhca_tunnel_commands[0x40];
 	u8         match_definer_format_supported[0x40];
 };
@@ -6361,6 +6375,18 @@ struct mlx5_ifc_general_obj_out_cmd_hdr_bits {
 	u8         reserved_at_60[0x20];
 };
 
+struct mlx5_ifc_modify_header_arg_bits {
+	u8         reserved_at_0[0x80];
+
+	u8         reserved_at_80[0x8];
+	u8         access_pd[0x18];
+};
+
+struct mlx5_ifc_create_modify_header_arg_in_bits {
+	struct mlx5_ifc_general_obj_in_cmd_hdr_bits hdr;
+	struct mlx5_ifc_modify_header_arg_bits arg;
+};
+
 struct mlx5_ifc_create_match_definer_in_bits {
 	struct mlx5_ifc_general_obj_in_cmd_hdr_bits general_obj_in_cmd_hdr;
 
-- 
cgit v1.2.3-70-g09d2


From 977c4a3e7c89d5b48fdc548773fb9ed3e6c50cf8 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:20:40 +0300
Subject: net/mlx5: Add new WQE for updating flow table

Add new WQE type: FLOW_TBL_ACCESS, which will be used for
writing modify header arguments.
This type has specific control segment and special data segment.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 include/linux/mlx5/device.h |  2 ++
 include/linux/mlx5/qp.h     | 10 ++++++++++
 2 files changed, 12 insertions(+)

diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index af4dd536a52c..e4aa147ab390 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -442,6 +442,8 @@ enum {
 
 	MLX5_OPCODE_UMR			= 0x25,
 
+	MLX5_OPCODE_FLOW_TBL_ACCESS	= 0x2c,
+
 	MLX5_OPCODE_ACCESS_ASO		= 0x2d,
 };
 
diff --git a/include/linux/mlx5/qp.h b/include/linux/mlx5/qp.h
index df55fbb65717..bd53cf4be7bd 100644
--- a/include/linux/mlx5/qp.h
+++ b/include/linux/mlx5/qp.h
@@ -499,6 +499,16 @@ struct mlx5_stride_block_ctrl_seg {
 	__be16		num_entries;
 };
 
+struct mlx5_wqe_flow_update_ctrl_seg {
+	__be32		flow_idx_update;
+	__be32		dest_handle;
+	u8		reserved0[40];
+};
+
+struct mlx5_wqe_header_modify_argument_update_seg {
+	u8		argument_list[64];
+};
+
 struct mlx5_core_qp {
 	struct mlx5_core_rsc_common	common; /* must be first */
 	void (*event)		(struct mlx5_core_qp *, int);
-- 
cgit v1.2.3-70-g09d2


From 1e5cc7369bb08808813d2795ae4adbd7591b31d0 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:20:49 +0300
Subject: net/mlx5: DR, Prepare sending new WQE type

The send engine should be ready to handle more opcodes
in addition to RDMA_WRITE/RDMA_READ.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_send.c | 60 ++++++++++++++--------
 1 file changed, 39 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index fd2d31cdbcf9..00bb65613300 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -18,7 +18,12 @@ struct dr_data_seg {
 	unsigned int send_flags;
 };
 
+enum send_info_type {
+	WRITE_ICM = 0,
+};
+
 struct postsend_info {
+	enum send_info_type type;
 	struct dr_data_seg write;
 	struct dr_data_seg read;
 	u64 remote_addr;
@@ -402,10 +407,12 @@ static void dr_rdma_segments(struct mlx5dr_qp *dr_qp, u64 remote_addr,
 
 static void dr_post_send(struct mlx5dr_qp *dr_qp, struct postsend_info *send_info)
 {
-	dr_rdma_segments(dr_qp, send_info->remote_addr, send_info->rkey,
-			 &send_info->write, MLX5_OPCODE_RDMA_WRITE, false);
-	dr_rdma_segments(dr_qp, send_info->remote_addr, send_info->rkey,
-			 &send_info->read, MLX5_OPCODE_RDMA_READ, true);
+	if (send_info->type == WRITE_ICM) {
+		dr_rdma_segments(dr_qp, send_info->remote_addr, send_info->rkey,
+				 &send_info->write, MLX5_OPCODE_RDMA_WRITE, false);
+		dr_rdma_segments(dr_qp, send_info->remote_addr, send_info->rkey,
+				 &send_info->read, MLX5_OPCODE_RDMA_READ, true);
+	}
 }
 
 /**
@@ -476,9 +483,26 @@ static int dr_handle_pending_wc(struct mlx5dr_domain *dmn,
 	return 0;
 }
 
-static void dr_fill_data_segs(struct mlx5dr_send_ring *send_ring,
-			      struct postsend_info *send_info)
+static void dr_fill_write_icm_segs(struct mlx5dr_domain *dmn,
+				   struct mlx5dr_send_ring *send_ring,
+				   struct postsend_info *send_info)
 {
+	u32 buff_offset;
+
+	if (send_info->write.length > dmn->info.max_inline_size) {
+		buff_offset = (send_ring->tx_head &
+			       (dmn->send_ring->signal_th - 1)) *
+			      send_ring->max_post_send_size;
+		/* Copy to ring mr */
+		memcpy(send_ring->buf + buff_offset,
+		       (void *)(uintptr_t)send_info->write.addr,
+		       send_info->write.length);
+		send_info->write.addr = (uintptr_t)send_ring->mr->dma_addr + buff_offset;
+		send_info->write.lkey = send_ring->mr->mkey;
+
+		send_ring->tx_head++;
+	}
+
 	send_ring->pending_wqe++;
 
 	if (send_ring->pending_wqe % send_ring->signal_th == 0)
@@ -496,11 +520,18 @@ static void dr_fill_data_segs(struct mlx5dr_send_ring *send_ring,
 		send_info->read.send_flags = 0;
 }
 
+static void dr_fill_data_segs(struct mlx5dr_domain *dmn,
+			      struct mlx5dr_send_ring *send_ring,
+			      struct postsend_info *send_info)
+{
+	if (send_info->type == WRITE_ICM)
+		dr_fill_write_icm_segs(dmn, send_ring, send_info);
+}
+
 static int dr_postsend_icm_data(struct mlx5dr_domain *dmn,
 				struct postsend_info *send_info)
 {
 	struct mlx5dr_send_ring *send_ring = dmn->send_ring;
-	u32 buff_offset;
 	int ret;
 
 	if (unlikely(dmn->mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR ||
@@ -517,20 +548,7 @@ static int dr_postsend_icm_data(struct mlx5dr_domain *dmn,
 	if (ret)
 		goto out_unlock;
 
-	if (send_info->write.length > dmn->info.max_inline_size) {
-		buff_offset = (send_ring->tx_head &
-			       (dmn->send_ring->signal_th - 1)) *
-			send_ring->max_post_send_size;
-		/* Copy to ring mr */
-		memcpy(send_ring->buf + buff_offset,
-		       (void *)(uintptr_t)send_info->write.addr,
-		       send_info->write.length);
-		send_info->write.addr = (uintptr_t)send_ring->mr->dma_addr + buff_offset;
-		send_info->write.lkey = send_ring->mr->mkey;
-	}
-
-	send_ring->tx_head++;
-	dr_fill_data_segs(send_ring, send_info);
+	dr_fill_data_segs(dmn, send_ring, send_info);
 	dr_post_send(send_ring->qp, send_info);
 
 out_unlock:
-- 
cgit v1.2.3-70-g09d2


From 108ff8215b55903545abafa198f83624a20f44c8 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:20:54 +0300
Subject: net/mlx5: DR, Add modify-header-pattern ICM pool

There is a new ICM area for that memory, so we need to handle it as we
did for the others ICM types.
The patch added that specific pool with its requirements and management.

Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |  2 +-
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |  6 +++
 .../mellanox/mlx5/core/steering/dr_domain.c        | 45 ++++++++++++++++++++--
 .../mellanox/mlx5/core/steering/dr_icm_pool.c      | 41 ++++++++++++++------
 .../ethernet/mellanox/mlx5/core/steering/dr_ptrn.c | 43 +++++++++++++++++++++
 .../mellanox/mlx5/core/steering/dr_types.h         | 11 ++++++
 6 files changed, 132 insertions(+), 16 deletions(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 4a009b720bee..39c2c8dc7e07 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -112,7 +112,7 @@ mlx5_core-$(CONFIG_MLX5_SW_STEERING) += steering/dr_domain.o steering/dr_table.o
 					steering/dr_ste_v2.o \
 					steering/dr_cmd.o steering/dr_fw.o \
 					steering/dr_action.o steering/fs_dr.o \
-					steering/dr_definer.o \
+					steering/dr_definer.o steering/dr_ptrn.o \
 					steering/dr_dbg.o lib/smfs.o
 #
 # SF device
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index 07b6a6dcb92f..229f3684100c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -200,6 +200,12 @@ int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev,
 	caps->hdr_modify_icm_addr =
 		MLX5_CAP64_DEV_MEM(mdev, header_modify_sw_icm_start_address);
 
+	caps->log_modify_pattern_icm_size =
+		MLX5_CAP_DEV_MEM(mdev, log_header_modify_pattern_sw_icm_size);
+
+	caps->hdr_modify_pattern_icm_addr =
+		MLX5_CAP64_DEV_MEM(mdev, header_modify_pattern_sw_icm_start_address);
+
 	caps->roce_min_src_udp = MLX5_CAP_ROCE(mdev, r_roce_min_src_udp_port);
 
 	caps->is_ecpf = mlx5_core_is_ecpf_esw_manager(mdev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
index 5b8bb2ca31e6..7a0381572c4c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
@@ -10,6 +10,33 @@
 	 ((dmn)->info.caps.dmn_type##_sw_owner_v2 &&	\
 	  (dmn)->info.caps.sw_format_ver <= MLX5_STEERING_FORMAT_CONNECTX_7))
 
+bool mlx5dr_domain_is_support_ptrn_arg(struct mlx5dr_domain *dmn)
+{
+	return false;
+}
+
+static int dr_domain_init_modify_header_resources(struct mlx5dr_domain *dmn)
+{
+	if (!mlx5dr_domain_is_support_ptrn_arg(dmn))
+		return 0;
+
+	dmn->ptrn_mgr = mlx5dr_ptrn_mgr_create(dmn);
+	if (!dmn->ptrn_mgr) {
+		mlx5dr_err(dmn, "Couldn't create ptrn_mgr\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void dr_domain_destroy_modify_header_resources(struct mlx5dr_domain *dmn)
+{
+	if (!mlx5dr_domain_is_support_ptrn_arg(dmn))
+		return;
+
+	mlx5dr_ptrn_mgr_destroy(dmn->ptrn_mgr);
+}
+
 static void dr_domain_init_csum_recalc_fts(struct mlx5dr_domain *dmn)
 {
 	/* Per vport cached FW FT for checksum recalculation, this
@@ -149,14 +176,22 @@ static int dr_domain_init_resources(struct mlx5dr_domain *dmn)
 		goto clean_uar;
 	}
 
+	ret = dr_domain_init_modify_header_resources(dmn);
+	if (ret) {
+		mlx5dr_err(dmn, "Couldn't create modify-header-resources\n");
+		goto clean_mem_resources;
+	}
+
 	ret = mlx5dr_send_ring_alloc(dmn);
 	if (ret) {
 		mlx5dr_err(dmn, "Couldn't create send-ring\n");
-		goto clean_mem_resources;
+		goto clean_modify_hdr;
 	}
 
 	return 0;
 
+clean_modify_hdr:
+	dr_domain_destroy_modify_header_resources(dmn);
 clean_mem_resources:
 	dr_domain_uninit_mem_resources(dmn);
 clean_uar:
@@ -170,6 +205,7 @@ clean_pd:
 static void dr_domain_uninit_resources(struct mlx5dr_domain *dmn)
 {
 	mlx5dr_send_ring_free(dmn, dmn->send_ring);
+	dr_domain_destroy_modify_header_resources(dmn);
 	dr_domain_uninit_mem_resources(dmn);
 	mlx5_put_uars_page(dmn->mdev, dmn->uar);
 	mlx5_core_dealloc_pd(dmn->mdev, dmn->pdn);
@@ -215,7 +251,7 @@ static int dr_domain_query_vport(struct mlx5dr_domain *dmn,
 	return 0;
 }
 
-static int dr_domain_query_esw_mngr(struct mlx5dr_domain *dmn)
+static int dr_domain_query_esw_mgr(struct mlx5dr_domain *dmn)
 {
 	return dr_domain_query_vport(dmn, 0, false,
 				     &dmn->info.caps.vports.esw_manager_caps);
@@ -321,7 +357,7 @@ static int dr_domain_query_fdb_caps(struct mlx5_core_dev *mdev,
 	 * vports (vport 0, VFs and SFs) will be queried dynamically.
 	 */
 
-	ret = dr_domain_query_esw_mngr(dmn);
+	ret = dr_domain_query_esw_mgr(dmn);
 	if (ret) {
 		mlx5dr_err(dmn, "Failed to query eswitch manager vport caps (err: %d)", ret);
 		goto free_vports_caps_xa;
@@ -435,6 +471,9 @@ mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type)
 	dmn->info.max_log_action_icm_sz = DR_CHUNK_SIZE_4K;
 	dmn->info.max_log_sw_icm_sz = min_t(u32, DR_CHUNK_SIZE_1024K,
 					    dmn->info.caps.log_icm_size);
+	dmn->info.max_log_modify_hdr_pattern_icm_sz =
+		min_t(u32, DR_CHUNK_SIZE_4K,
+		      dmn->info.caps.log_modify_pattern_icm_size);
 
 	if (!dmn->info.supp_sw_steering) {
 		mlx5dr_err(dmn, "SW steering is not supported\n");
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
index 3eb6719bc8eb..04fc170a6c16 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
@@ -107,9 +107,9 @@ static struct mlx5dr_icm_mr *
 dr_icm_pool_mr_create(struct mlx5dr_icm_pool *pool)
 {
 	struct mlx5_core_dev *mdev = pool->dmn->mdev;
-	enum mlx5_sw_icm_type dm_type;
+	enum mlx5_sw_icm_type dm_type = 0;
 	struct mlx5dr_icm_mr *icm_mr;
-	size_t log_align_base;
+	size_t log_align_base = 0;
 	int err;
 
 	icm_mr = kvzalloc(sizeof(*icm_mr), GFP_KERNEL);
@@ -121,14 +121,25 @@ dr_icm_pool_mr_create(struct mlx5dr_icm_pool *pool)
 	icm_mr->dm.length = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
 							       pool->icm_type);
 
-	if (pool->icm_type == DR_ICM_TYPE_STE) {
+	switch (pool->icm_type) {
+	case DR_ICM_TYPE_STE:
 		dm_type = MLX5_SW_ICM_TYPE_STEERING;
 		log_align_base = ilog2(icm_mr->dm.length);
-	} else {
+		break;
+	case DR_ICM_TYPE_MODIFY_ACTION:
 		dm_type = MLX5_SW_ICM_TYPE_HEADER_MODIFY;
 		/* Align base is 64B */
 		log_align_base = ilog2(DR_ICM_MODIFY_HDR_ALIGN_BASE);
+		break;
+	case DR_ICM_TYPE_MODIFY_HDR_PTRN:
+		dm_type = MLX5_SW_ICM_TYPE_HEADER_MODIFY_PATTERN;
+		/* Align base is 64B */
+		log_align_base = ilog2(DR_ICM_MODIFY_HDR_ALIGN_BASE);
+		break;
+	default:
+		WARN_ON(pool->icm_type);
 	}
+
 	icm_mr->dm.type = dm_type;
 
 	err = mlx5_dm_sw_icm_alloc(mdev, icm_mr->dm.type, icm_mr->dm.length,
@@ -493,27 +504,33 @@ struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
 					       enum mlx5dr_icm_type icm_type)
 {
 	u32 num_of_chunks, entry_size, max_hot_size;
-	enum mlx5dr_icm_chunk_size max_log_chunk_sz;
 	struct mlx5dr_icm_pool *pool;
 
-	if (icm_type == DR_ICM_TYPE_STE)
-		max_log_chunk_sz = dmn->info.max_log_sw_icm_sz;
-	else
-		max_log_chunk_sz = dmn->info.max_log_action_icm_sz;
-
 	pool = kvzalloc(sizeof(*pool), GFP_KERNEL);
 	if (!pool)
 		return NULL;
 
 	pool->dmn = dmn;
 	pool->icm_type = icm_type;
-	pool->max_log_chunk_sz = max_log_chunk_sz;
 	pool->chunks_kmem_cache = dmn->chunks_kmem_cache;
 
 	INIT_LIST_HEAD(&pool->buddy_mem_list);
-
 	mutex_init(&pool->mutex);
 
+	switch (icm_type) {
+	case DR_ICM_TYPE_STE:
+		pool->max_log_chunk_sz = dmn->info.max_log_sw_icm_sz;
+		break;
+	case DR_ICM_TYPE_MODIFY_ACTION:
+		pool->max_log_chunk_sz = dmn->info.max_log_action_icm_sz;
+		break;
+	case DR_ICM_TYPE_MODIFY_HDR_PTRN:
+		pool->max_log_chunk_sz = dmn->info.max_log_modify_hdr_pattern_icm_sz;
+		break;
+	default:
+		WARN_ON(icm_type);
+	}
+
 	entry_size = mlx5dr_icm_pool_dm_type_to_entry_size(pool->icm_type);
 
 	max_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c
new file mode 100644
index 000000000000..698e79d278bf
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c
@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+// Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+#include "dr_types.h"
+
+struct mlx5dr_ptrn_mgr {
+	struct mlx5dr_domain *dmn;
+	struct mlx5dr_icm_pool *ptrn_icm_pool;
+};
+
+struct mlx5dr_ptrn_mgr *mlx5dr_ptrn_mgr_create(struct mlx5dr_domain *dmn)
+{
+	struct mlx5dr_ptrn_mgr *mgr;
+
+	if (!mlx5dr_domain_is_support_ptrn_arg(dmn))
+		return NULL;
+
+	mgr = kzalloc(sizeof(*mgr), GFP_KERNEL);
+	if (!mgr)
+		return NULL;
+
+	mgr->dmn = dmn;
+	mgr->ptrn_icm_pool = mlx5dr_icm_pool_create(dmn, DR_ICM_TYPE_MODIFY_HDR_PTRN);
+	if (!mgr->ptrn_icm_pool) {
+		mlx5dr_err(dmn, "Couldn't get modify-header-pattern memory\n");
+		goto free_mgr;
+	}
+
+	return mgr;
+
+free_mgr:
+	kfree(mgr);
+	return NULL;
+}
+
+void mlx5dr_ptrn_mgr_destroy(struct mlx5dr_ptrn_mgr *mgr)
+{
+	if (!mgr)
+		return;
+
+	mlx5dr_icm_pool_destroy(mgr->ptrn_icm_pool);
+	kfree(mgr);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 2b769dcbd453..5b9faa714f42 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -26,6 +26,8 @@
 #define mlx5dr_info(dmn, arg...) mlx5_core_info((dmn)->mdev, ##arg)
 #define mlx5dr_dbg(dmn, arg...) mlx5_core_dbg((dmn)->mdev, ##arg)
 
+struct mlx5dr_ptrn_mgr;
+
 static inline bool dr_is_flex_parser_0_id(u8 parser_id)
 {
 	return parser_id <= DR_STE_MAX_FLEX_0_ID;
@@ -66,6 +68,7 @@ enum mlx5dr_icm_chunk_size {
 enum mlx5dr_icm_type {
 	DR_ICM_TYPE_STE,
 	DR_ICM_TYPE_MODIFY_ACTION,
+	DR_ICM_TYPE_MODIFY_HDR_PTRN,
 };
 
 static inline enum mlx5dr_icm_chunk_size
@@ -861,6 +864,8 @@ struct mlx5dr_cmd_caps {
 	u64 esw_tx_drop_address;
 	u32 log_icm_size;
 	u64 hdr_modify_icm_addr;
+	u32 log_modify_pattern_icm_size;
+	u64 hdr_modify_pattern_icm_addr;
 	u32 flex_protocols;
 	u8 flex_parser_id_icmp_dw0;
 	u8 flex_parser_id_icmp_dw1;
@@ -910,6 +915,7 @@ struct mlx5dr_domain_info {
 	u32 max_send_wr;
 	u32 max_log_sw_icm_sz;
 	u32 max_log_action_icm_sz;
+	u32 max_log_modify_hdr_pattern_icm_sz;
 	struct mlx5dr_domain_rx_tx rx;
 	struct mlx5dr_domain_rx_tx tx;
 	struct mlx5dr_cmd_caps caps;
@@ -928,6 +934,7 @@ struct mlx5dr_domain {
 	struct mlx5dr_send_info_pool *send_info_pool_tx;
 	struct kmem_cache *chunks_kmem_cache;
 	struct kmem_cache *htbls_kmem_cache;
+	struct mlx5dr_ptrn_mgr *ptrn_mgr;
 	struct mlx5dr_send_ring *send_ring;
 	struct mlx5dr_domain_info info;
 	struct xarray csum_fts_xa;
@@ -1526,4 +1533,8 @@ static inline bool mlx5dr_supp_match_ranges(struct mlx5_core_dev *dev)
 			(1ULL << MLX5_IFC_DEFINER_FORMAT_ID_SELECT));
 }
 
+bool mlx5dr_domain_is_support_ptrn_arg(struct mlx5dr_domain *dmn);
+struct mlx5dr_ptrn_mgr *mlx5dr_ptrn_mgr_create(struct mlx5dr_domain *dmn);
+void mlx5dr_ptrn_mgr_destroy(struct mlx5dr_ptrn_mgr *mgr);
+
 #endif  /* _DR_TYPES_H_ */
-- 
cgit v1.2.3-70-g09d2


From 18bb56ab4477b966e2974b96a5e3bf9ba052aac5 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Tue, 11 Apr 2023 04:35:41 +0200
Subject: net: dsa: mv88e6xxx: Correct cmode to PHY_INTERFACE_

The switch can either take the MAC or the PHY role in an MII or RMII
link. There are distinct PHY_INTERFACE_ macros for these two roles.
Correct the mapping so that the `REV` version is used for the PHY
role.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230411023541.2372609-1-andrew@lunn.ch
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 2e22dad77935..dc263cea205f 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -611,10 +611,10 @@ static void mv88e6185_phylink_get_caps(struct mv88e6xxx_chip *chip, int port,
 }
 
 static const u8 mv88e6xxx_phy_interface_modes[] = {
-	[MV88E6XXX_PORT_STS_CMODE_MII_PHY]	= PHY_INTERFACE_MODE_MII,
+	[MV88E6XXX_PORT_STS_CMODE_MII_PHY]	= PHY_INTERFACE_MODE_REVMII,
 	[MV88E6XXX_PORT_STS_CMODE_MII]		= PHY_INTERFACE_MODE_MII,
 	[MV88E6XXX_PORT_STS_CMODE_GMII]		= PHY_INTERFACE_MODE_GMII,
-	[MV88E6XXX_PORT_STS_CMODE_RMII_PHY]	= PHY_INTERFACE_MODE_RMII,
+	[MV88E6XXX_PORT_STS_CMODE_RMII_PHY]	= PHY_INTERFACE_MODE_REVRMII,
 	[MV88E6XXX_PORT_STS_CMODE_RMII]		= PHY_INTERFACE_MODE_RMII,
 	[MV88E6XXX_PORT_STS_CMODE_100BASEX]	= PHY_INTERFACE_MODE_100BASEX,
 	[MV88E6XXX_PORT_STS_CMODE_1000BASEX]	= PHY_INTERFACE_MODE_1000BASEX,
-- 
cgit v1.2.3-70-g09d2


From 9538ebce88ffa074202d592d468521995cb1e714 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 7 Apr 2023 17:14:50 +0300
Subject: net: dsa: add trace points for FDB/MDB operations

DSA performs non-trivial housekeeping of unicast and multicast addresses
on shared (CPU and DSA) ports, and puts a bit of pressure on higher
layers, requiring them to behave correctly (remove these addresses
exactly as many times as they were added). Otherwise, either addresses
linger around forever, or DSA returns -ENOENT complaining that entries
that were already deleted must be deleted again.

To aid debugging, introduce some trace points specifically for FDB and
MDB - that's where some of the bugs still are right now.

Some bugs I have seen were also due to race conditions, see:
- 630fd4822af2 ("net: dsa: flush switchdev workqueue on bridge join error path")
- a2614140dc0f ("net: dsa: mv88e6xxx: flush switchdev FDB workqueue before removing VLAN")

so it would be good to not disturb the timing too much, hence the choice
to use trace points vs regular dev_dbg().

I've had these for some time on my computer in a less polished form, and
they've proven useful. What I found most useful was to enable
CONFIG_BOOTTIME_TRACING, add "trace_event=dsa" to the kernel cmdline,
and run "cat /sys/kernel/debug/tracing/trace". This is to debug more
complex environments with network managers started by the init system,
things like that.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/Makefile |   6 +-
 net/dsa/switch.c |  61 +++++++++--
 net/dsa/trace.c  |  39 +++++++
 net/dsa/trace.h  | 329 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 423 insertions(+), 12 deletions(-)
 create mode 100644 net/dsa/trace.c
 create mode 100644 net/dsa/trace.h

diff --git a/net/dsa/Makefile b/net/dsa/Makefile
index 3835de286116..12e305824a96 100644
--- a/net/dsa/Makefile
+++ b/net/dsa/Makefile
@@ -16,7 +16,8 @@ dsa_core-y += \
 	slave.o \
 	switch.o \
 	tag.o \
-	tag_8021q.o
+	tag_8021q.o \
+	trace.o
 
 # tagging formats
 obj-$(CONFIG_NET_DSA_TAG_AR9331) += tag_ar9331.o
@@ -37,3 +38,6 @@ obj-$(CONFIG_NET_DSA_TAG_RZN1_A5PSW) += tag_rzn1_a5psw.o
 obj-$(CONFIG_NET_DSA_TAG_SJA1105) += tag_sja1105.o
 obj-$(CONFIG_NET_DSA_TAG_TRAILER) += tag_trailer.o
 obj-$(CONFIG_NET_DSA_TAG_XRS700X) += tag_xrs700x.o
+
+# for tracing framework to find trace.h
+CFLAGS_trace.o := -I$(src)
diff --git a/net/dsa/switch.c b/net/dsa/switch.c
index d5bc4bb7310d..ff1b5d980e37 100644
--- a/net/dsa/switch.c
+++ b/net/dsa/switch.c
@@ -18,6 +18,7 @@
 #include "slave.h"
 #include "switch.h"
 #include "tag_8021q.h"
+#include "trace.h"
 
 static unsigned int dsa_switch_fastest_ageing_time(struct dsa_switch *ds,
 						   unsigned int ageing_time)
@@ -164,14 +165,20 @@ static int dsa_port_do_mdb_add(struct dsa_port *dp,
 	int err = 0;
 
 	/* No need to bother with refcounting for user ports */
-	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
-		return ds->ops->port_mdb_add(ds, port, mdb, db);
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) {
+		err = ds->ops->port_mdb_add(ds, port, mdb, db);
+		trace_dsa_mdb_add_hw(dp, mdb->addr, mdb->vid, &db, err);
+
+		return err;
+	}
 
 	mutex_lock(&dp->addr_lists_lock);
 
 	a = dsa_mac_addr_find(&dp->mdbs, mdb->addr, mdb->vid, db);
 	if (a) {
 		refcount_inc(&a->refcount);
+		trace_dsa_mdb_add_bump(dp, mdb->addr, mdb->vid, &db,
+				       &a->refcount);
 		goto out;
 	}
 
@@ -182,6 +189,7 @@ static int dsa_port_do_mdb_add(struct dsa_port *dp,
 	}
 
 	err = ds->ops->port_mdb_add(ds, port, mdb, db);
+	trace_dsa_mdb_add_hw(dp, mdb->addr, mdb->vid, &db, err);
 	if (err) {
 		kfree(a);
 		goto out;
@@ -209,21 +217,30 @@ static int dsa_port_do_mdb_del(struct dsa_port *dp,
 	int err = 0;
 
 	/* No need to bother with refcounting for user ports */
-	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
-		return ds->ops->port_mdb_del(ds, port, mdb, db);
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) {
+		err = ds->ops->port_mdb_del(ds, port, mdb, db);
+		trace_dsa_mdb_del_hw(dp, mdb->addr, mdb->vid, &db, err);
+
+		return err;
+	}
 
 	mutex_lock(&dp->addr_lists_lock);
 
 	a = dsa_mac_addr_find(&dp->mdbs, mdb->addr, mdb->vid, db);
 	if (!a) {
+		trace_dsa_mdb_del_not_found(dp, mdb->addr, mdb->vid, &db);
 		err = -ENOENT;
 		goto out;
 	}
 
-	if (!refcount_dec_and_test(&a->refcount))
+	if (!refcount_dec_and_test(&a->refcount)) {
+		trace_dsa_mdb_del_drop(dp, mdb->addr, mdb->vid, &db,
+				       &a->refcount);
 		goto out;
+	}
 
 	err = ds->ops->port_mdb_del(ds, port, mdb, db);
+	trace_dsa_mdb_del_hw(dp, mdb->addr, mdb->vid, &db, err);
 	if (err) {
 		refcount_set(&a->refcount, 1);
 		goto out;
@@ -247,14 +264,19 @@ static int dsa_port_do_fdb_add(struct dsa_port *dp, const unsigned char *addr,
 	int err = 0;
 
 	/* No need to bother with refcounting for user ports */
-	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
-		return ds->ops->port_fdb_add(ds, port, addr, vid, db);
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) {
+		err = ds->ops->port_fdb_add(ds, port, addr, vid, db);
+		trace_dsa_fdb_add_hw(dp, addr, vid, &db, err);
+
+		return err;
+	}
 
 	mutex_lock(&dp->addr_lists_lock);
 
 	a = dsa_mac_addr_find(&dp->fdbs, addr, vid, db);
 	if (a) {
 		refcount_inc(&a->refcount);
+		trace_dsa_fdb_add_bump(dp, addr, vid, &db, &a->refcount);
 		goto out;
 	}
 
@@ -265,6 +287,7 @@ static int dsa_port_do_fdb_add(struct dsa_port *dp, const unsigned char *addr,
 	}
 
 	err = ds->ops->port_fdb_add(ds, port, addr, vid, db);
+	trace_dsa_fdb_add_hw(dp, addr, vid, &db, err);
 	if (err) {
 		kfree(a);
 		goto out;
@@ -291,21 +314,29 @@ static int dsa_port_do_fdb_del(struct dsa_port *dp, const unsigned char *addr,
 	int err = 0;
 
 	/* No need to bother with refcounting for user ports */
-	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
-		return ds->ops->port_fdb_del(ds, port, addr, vid, db);
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) {
+		err = ds->ops->port_fdb_del(ds, port, addr, vid, db);
+		trace_dsa_fdb_del_hw(dp, addr, vid, &db, err);
+
+		return err;
+	}
 
 	mutex_lock(&dp->addr_lists_lock);
 
 	a = dsa_mac_addr_find(&dp->fdbs, addr, vid, db);
 	if (!a) {
+		trace_dsa_fdb_del_not_found(dp, addr, vid, &db);
 		err = -ENOENT;
 		goto out;
 	}
 
-	if (!refcount_dec_and_test(&a->refcount))
+	if (!refcount_dec_and_test(&a->refcount)) {
+		trace_dsa_fdb_del_drop(dp, addr, vid, &db, &a->refcount);
 		goto out;
+	}
 
 	err = ds->ops->port_fdb_del(ds, port, addr, vid, db);
+	trace_dsa_fdb_del_hw(dp, addr, vid, &db, err);
 	if (err) {
 		refcount_set(&a->refcount, 1);
 		goto out;
@@ -332,6 +363,8 @@ static int dsa_switch_do_lag_fdb_add(struct dsa_switch *ds, struct dsa_lag *lag,
 	a = dsa_mac_addr_find(&lag->fdbs, addr, vid, db);
 	if (a) {
 		refcount_inc(&a->refcount);
+		trace_dsa_lag_fdb_add_bump(lag->dev, addr, vid, &db,
+					   &a->refcount);
 		goto out;
 	}
 
@@ -342,6 +375,7 @@ static int dsa_switch_do_lag_fdb_add(struct dsa_switch *ds, struct dsa_lag *lag,
 	}
 
 	err = ds->ops->lag_fdb_add(ds, *lag, addr, vid, db);
+	trace_dsa_lag_fdb_add_hw(lag->dev, addr, vid, &db, err);
 	if (err) {
 		kfree(a);
 		goto out;
@@ -370,14 +404,19 @@ static int dsa_switch_do_lag_fdb_del(struct dsa_switch *ds, struct dsa_lag *lag,
 
 	a = dsa_mac_addr_find(&lag->fdbs, addr, vid, db);
 	if (!a) {
+		trace_dsa_lag_fdb_del_not_found(lag->dev, addr, vid, &db);
 		err = -ENOENT;
 		goto out;
 	}
 
-	if (!refcount_dec_and_test(&a->refcount))
+	if (!refcount_dec_and_test(&a->refcount)) {
+		trace_dsa_lag_fdb_del_drop(lag->dev, addr, vid, &db,
+					   &a->refcount);
 		goto out;
+	}
 
 	err = ds->ops->lag_fdb_del(ds, *lag, addr, vid, db);
+	trace_dsa_lag_fdb_del_hw(lag->dev, addr, vid, &db, err);
 	if (err) {
 		refcount_set(&a->refcount, 1);
 		goto out;
diff --git a/net/dsa/trace.c b/net/dsa/trace.c
new file mode 100644
index 000000000000..1b107165d331
--- /dev/null
+++ b/net/dsa/trace.c
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* Copyright 2022-2023 NXP
+ */
+
+#define CREATE_TRACE_POINTS
+#include "trace.h"
+
+void dsa_db_print(const struct dsa_db *db, char buf[DSA_DB_BUFSIZ])
+{
+	switch (db->type) {
+	case DSA_DB_PORT:
+		sprintf(buf, "port %s", db->dp->name);
+		break;
+	case DSA_DB_LAG:
+		sprintf(buf, "lag %s id %d", db->lag.dev->name, db->lag.id);
+		break;
+	case DSA_DB_BRIDGE:
+		sprintf(buf, "bridge %s num %d", db->bridge.dev->name,
+			db->bridge.num);
+		break;
+	default:
+		sprintf(buf, "unknown");
+		break;
+	}
+}
+
+const char *dsa_port_kind(const struct dsa_port *dp)
+{
+	switch (dp->type) {
+	case DSA_PORT_TYPE_USER:
+		return "user";
+	case DSA_PORT_TYPE_CPU:
+		return "cpu";
+	case DSA_PORT_TYPE_DSA:
+		return "dsa";
+	default:
+		return "unused";
+	}
+}
diff --git a/net/dsa/trace.h b/net/dsa/trace.h
new file mode 100644
index 000000000000..42c8bbc7d472
--- /dev/null
+++ b/net/dsa/trace.h
@@ -0,0 +1,329 @@
+/* SPDX-License-Identifier: GPL-2.0
+ * Copyright 2022-2023 NXP
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM	dsa
+
+#if !defined(_NET_DSA_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _NET_DSA_TRACE_H
+
+#include <net/dsa.h>
+#include <linux/etherdevice.h>
+#include <linux/refcount.h>
+#include <linux/tracepoint.h>
+
+/* Enough to fit "bridge %s num %d" where num has 3 digits */
+#define DSA_DB_BUFSIZ	(IFNAMSIZ + 16)
+
+void dsa_db_print(const struct dsa_db *db, char buf[DSA_DB_BUFSIZ]);
+const char *dsa_port_kind(const struct dsa_port *dp);
+
+DECLARE_EVENT_CLASS(dsa_port_addr_op_hw,
+
+	TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, u16 vid,
+		 const struct dsa_db *db, int err),
+
+	TP_ARGS(dp, addr, vid, db, err),
+
+	TP_STRUCT__entry(
+		__string(dev, dev_name(dp->ds->dev))
+		__string(kind, dsa_port_kind(dp))
+		__field(int, port)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+		__field(int, err)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, dev_name(dp->ds->dev));
+		__assign_str(kind, dsa_port_kind(dp));
+		__entry->port = dp->index;
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+		__entry->err = err;
+	),
+
+	TP_printk("%s %s port %d addr %pM vid %u db \"%s\" err %d",
+		  __get_str(dev), __get_str(kind), __entry->port, __entry->addr,
+		  __entry->vid, __entry->db_buf, __entry->err)
+);
+
+/* Add unicast/multicast address to hardware, either on user ports
+ * (where no refcounting is kept), or on shared ports when the entry
+ * is first seen and its refcount is 1.
+ */
+DEFINE_EVENT(dsa_port_addr_op_hw, dsa_fdb_add_hw,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db, int err),
+	     TP_ARGS(dp, addr, vid, db, err));
+
+DEFINE_EVENT(dsa_port_addr_op_hw, dsa_mdb_add_hw,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db, int err),
+	     TP_ARGS(dp, addr, vid, db, err));
+
+/* Delete unicast/multicast address from hardware, either on user ports or
+ * when the refcount on shared ports reaches 0
+ */
+DEFINE_EVENT(dsa_port_addr_op_hw, dsa_fdb_del_hw,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db, int err),
+	     TP_ARGS(dp, addr, vid, db, err));
+
+DEFINE_EVENT(dsa_port_addr_op_hw, dsa_mdb_del_hw,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db, int err),
+	     TP_ARGS(dp, addr, vid, db, err));
+
+DECLARE_EVENT_CLASS(dsa_port_addr_op_refcount,
+
+	TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, u16 vid,
+		 const struct dsa_db *db, const refcount_t *refcount),
+
+	TP_ARGS(dp, addr, vid, db, refcount),
+
+	TP_STRUCT__entry(
+		__string(dev, dev_name(dp->ds->dev))
+		__string(kind, dsa_port_kind(dp))
+		__field(int, port)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+		__field(unsigned int, refcount)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, dev_name(dp->ds->dev));
+		__assign_str(kind, dsa_port_kind(dp));
+		__entry->port = dp->index;
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+		__entry->refcount = refcount_read(refcount);
+	),
+
+	TP_printk("%s %s port %d addr %pM vid %u db \"%s\" refcount %u",
+		  __get_str(dev), __get_str(kind), __entry->port, __entry->addr,
+		  __entry->vid, __entry->db_buf, __entry->refcount)
+);
+
+/* Bump the refcount of an existing unicast/multicast address on shared ports */
+DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_fdb_add_bump,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db,
+		      const refcount_t *refcount),
+	     TP_ARGS(dp, addr, vid, db, refcount));
+
+DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_mdb_add_bump,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db,
+		      const refcount_t *refcount),
+	     TP_ARGS(dp, addr, vid, db, refcount));
+
+/* Drop the refcount of a multicast address that we still keep on
+ * shared ports
+ */
+DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_fdb_del_drop,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db,
+		      const refcount_t *refcount),
+	     TP_ARGS(dp, addr, vid, db, refcount));
+
+DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_mdb_del_drop,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db,
+		      const refcount_t *refcount),
+	     TP_ARGS(dp, addr, vid, db, refcount));
+
+DECLARE_EVENT_CLASS(dsa_port_addr_del_not_found,
+
+	TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, u16 vid,
+		 const struct dsa_db *db),
+
+	TP_ARGS(dp, addr, vid, db),
+
+	TP_STRUCT__entry(
+		__string(dev, dev_name(dp->ds->dev))
+		__string(kind, dsa_port_kind(dp))
+		__field(int, port)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, dev_name(dp->ds->dev));
+		__assign_str(kind, dsa_port_kind(dp));
+		__entry->port = dp->index;
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+	),
+
+	TP_printk("%s %s port %d addr %pM vid %u db \"%s\"",
+		  __get_str(dev), __get_str(kind), __entry->port,
+		  __entry->addr, __entry->vid, __entry->db_buf)
+);
+
+/* Attempt to delete a unicast/multicast address on shared ports for which
+ * the delete operation was called more times than the addition
+ */
+DEFINE_EVENT(dsa_port_addr_del_not_found, dsa_fdb_del_not_found,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db),
+	     TP_ARGS(dp, addr, vid, db));
+
+DEFINE_EVENT(dsa_port_addr_del_not_found, dsa_mdb_del_not_found,
+	     TP_PROTO(const struct dsa_port *dp, const unsigned char *addr,
+		      u16 vid, const struct dsa_db *db),
+	     TP_ARGS(dp, addr, vid, db));
+
+TRACE_EVENT(dsa_lag_fdb_add_hw,
+
+	TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr,
+		 u16 vid, const struct dsa_db *db, int err),
+
+	TP_ARGS(lag_dev, addr, vid, db, err),
+
+	TP_STRUCT__entry(
+		__string(dev, lag_dev->name)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+		__field(int, err)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, lag_dev->name);
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+		__entry->err = err;
+	),
+
+	TP_printk("%s addr %pM vid %u db \"%s\" err %d",
+		  __get_str(dev), __entry->addr, __entry->vid,
+		  __entry->db_buf, __entry->err)
+);
+
+TRACE_EVENT(dsa_lag_fdb_add_bump,
+
+	TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr,
+		 u16 vid, const struct dsa_db *db, const refcount_t *refcount),
+
+	TP_ARGS(lag_dev, addr, vid, db, refcount),
+
+	TP_STRUCT__entry(
+		__string(dev, lag_dev->name)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+		__field(unsigned int, refcount)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, lag_dev->name);
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+		__entry->refcount = refcount_read(refcount);
+	),
+
+	TP_printk("%s addr %pM vid %u db \"%s\" refcount %u",
+		  __get_str(dev), __entry->addr, __entry->vid,
+		  __entry->db_buf, __entry->refcount)
+);
+
+TRACE_EVENT(dsa_lag_fdb_del_hw,
+
+	TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr,
+		 u16 vid, const struct dsa_db *db, int err),
+
+	TP_ARGS(lag_dev, addr, vid, db, err),
+
+	TP_STRUCT__entry(
+		__string(dev, lag_dev->name)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+		__field(int, err)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, lag_dev->name);
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+		__entry->err = err;
+	),
+
+	TP_printk("%s addr %pM vid %u db \"%s\" err %d",
+		  __get_str(dev), __entry->addr, __entry->vid,
+		  __entry->db_buf, __entry->err)
+);
+
+TRACE_EVENT(dsa_lag_fdb_del_drop,
+
+	TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr,
+		 u16 vid, const struct dsa_db *db, const refcount_t *refcount),
+
+	TP_ARGS(lag_dev, addr, vid, db, refcount),
+
+	TP_STRUCT__entry(
+		__string(dev, lag_dev->name)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+		__field(unsigned int, refcount)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, lag_dev->name);
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+		__entry->refcount = refcount_read(refcount);
+	),
+
+	TP_printk("%s addr %pM vid %u db \"%s\" refcount %u",
+		  __get_str(dev), __entry->addr, __entry->vid,
+		  __entry->db_buf, __entry->refcount)
+);
+
+TRACE_EVENT(dsa_lag_fdb_del_not_found,
+
+	TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr,
+		 u16 vid, const struct dsa_db *db),
+
+	TP_ARGS(lag_dev, addr, vid, db),
+
+	TP_STRUCT__entry(
+		__string(dev, lag_dev->name)
+		__array(unsigned char, addr, ETH_ALEN)
+		__field(u16, vid)
+		__array(char, db_buf, DSA_DB_BUFSIZ)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, lag_dev->name);
+		ether_addr_copy(__entry->addr, addr);
+		__entry->vid = vid;
+		dsa_db_print(db, __entry->db_buf);
+	),
+
+	TP_printk("%s addr %pM vid %u db \"%s\"",
+		  __get_str(dev), __entry->addr, __entry->vid, __entry->db_buf)
+);
+
+#endif /* _NET_DSA_TRACE_H */
+
+/* We don't want to use include/trace/events */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE	trace
+/* This part must be outside protection */
+#include <trace/define_trace.h>
-- 
cgit v1.2.3-70-g09d2


From 02020bd70fa6abcb1c2a8525ce7c1500dd4f44a8 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 7 Apr 2023 17:14:51 +0300
Subject: net: dsa: add trace points for VLAN operations

These are not as critical as the FDB/MDB trace points (I'm not aware of
outstanding VLAN related bugs), but maybe they are useful to somebody,
either debugging something or simply trying to learn more.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/switch.c |  24 ++++++++---
 net/dsa/trace.h  | 118 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 137 insertions(+), 5 deletions(-)

diff --git a/net/dsa/switch.c b/net/dsa/switch.c
index ff1b5d980e37..8c9a9f94b756 100644
--- a/net/dsa/switch.c
+++ b/net/dsa/switch.c
@@ -695,8 +695,12 @@ static int dsa_port_do_vlan_add(struct dsa_port *dp,
 	int err = 0;
 
 	/* No need to bother with refcounting for user ports. */
-	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
-		return ds->ops->port_vlan_add(ds, port, vlan, extack);
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) {
+		err = ds->ops->port_vlan_add(ds, port, vlan, extack);
+		trace_dsa_vlan_add_hw(dp, vlan, err);
+
+		return err;
+	}
 
 	/* No need to propagate on shared ports the existing VLANs that were
 	 * re-notified after just the flags have changed. This would cause a
@@ -711,6 +715,7 @@ static int dsa_port_do_vlan_add(struct dsa_port *dp,
 	v = dsa_vlan_find(&dp->vlans, vlan);
 	if (v) {
 		refcount_inc(&v->refcount);
+		trace_dsa_vlan_add_bump(dp, vlan, &v->refcount);
 		goto out;
 	}
 
@@ -721,6 +726,7 @@ static int dsa_port_do_vlan_add(struct dsa_port *dp,
 	}
 
 	err = ds->ops->port_vlan_add(ds, port, vlan, extack);
+	trace_dsa_vlan_add_hw(dp, vlan, err);
 	if (err) {
 		kfree(v);
 		goto out;
@@ -745,21 +751,29 @@ static int dsa_port_do_vlan_del(struct dsa_port *dp,
 	int err = 0;
 
 	/* No need to bother with refcounting for user ports */
-	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp)))
-		return ds->ops->port_vlan_del(ds, port, vlan);
+	if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) {
+		err = ds->ops->port_vlan_del(ds, port, vlan);
+		trace_dsa_vlan_del_hw(dp, vlan, err);
+
+		return err;
+	}
 
 	mutex_lock(&dp->vlans_lock);
 
 	v = dsa_vlan_find(&dp->vlans, vlan);
 	if (!v) {
+		trace_dsa_vlan_del_not_found(dp, vlan);
 		err = -ENOENT;
 		goto out;
 	}
 
-	if (!refcount_dec_and_test(&v->refcount))
+	if (!refcount_dec_and_test(&v->refcount)) {
+		trace_dsa_vlan_del_drop(dp, vlan, &v->refcount);
 		goto out;
+	}
 
 	err = ds->ops->port_vlan_del(ds, port, vlan);
+	trace_dsa_vlan_del_hw(dp, vlan, err);
 	if (err) {
 		refcount_set(&v->refcount, 1);
 		goto out;
diff --git a/net/dsa/trace.h b/net/dsa/trace.h
index 42c8bbc7d472..567f29a39707 100644
--- a/net/dsa/trace.h
+++ b/net/dsa/trace.h
@@ -9,7 +9,9 @@
 #define _NET_DSA_TRACE_H
 
 #include <net/dsa.h>
+#include <net/switchdev.h>
 #include <linux/etherdevice.h>
+#include <linux/if_bridge.h>
 #include <linux/refcount.h>
 #include <linux/tracepoint.h>
 
@@ -318,6 +320,122 @@ TRACE_EVENT(dsa_lag_fdb_del_not_found,
 		  __get_str(dev), __entry->addr, __entry->vid, __entry->db_buf)
 );
 
+DECLARE_EVENT_CLASS(dsa_vlan_op_hw,
+
+	TP_PROTO(const struct dsa_port *dp,
+		 const struct switchdev_obj_port_vlan *vlan, int err),
+
+	TP_ARGS(dp, vlan, err),
+
+	TP_STRUCT__entry(
+		__string(dev, dev_name(dp->ds->dev))
+		__string(kind, dsa_port_kind(dp))
+		__field(int, port)
+		__field(u16, vid)
+		__field(u16, flags)
+		__field(bool, changed)
+		__field(int, err)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, dev_name(dp->ds->dev));
+		__assign_str(kind, dsa_port_kind(dp));
+		__entry->port = dp->index;
+		__entry->vid = vlan->vid;
+		__entry->flags = vlan->flags;
+		__entry->changed = vlan->changed;
+		__entry->err = err;
+	),
+
+	TP_printk("%s %s port %d vid %u%s%s%s",
+		  __get_str(dev), __get_str(kind), __entry->port, __entry->vid,
+		  __entry->flags & BRIDGE_VLAN_INFO_PVID ? " pvid" : "",
+		  __entry->flags & BRIDGE_VLAN_INFO_UNTAGGED ? " untagged" : "",
+		  __entry->changed ? " (changed)" : "")
+);
+
+DEFINE_EVENT(dsa_vlan_op_hw, dsa_vlan_add_hw,
+	     TP_PROTO(const struct dsa_port *dp,
+		      const struct switchdev_obj_port_vlan *vlan, int err),
+	     TP_ARGS(dp, vlan, err));
+
+DEFINE_EVENT(dsa_vlan_op_hw, dsa_vlan_del_hw,
+	     TP_PROTO(const struct dsa_port *dp,
+		      const struct switchdev_obj_port_vlan *vlan, int err),
+	     TP_ARGS(dp, vlan, err));
+
+DECLARE_EVENT_CLASS(dsa_vlan_op_refcount,
+
+	TP_PROTO(const struct dsa_port *dp,
+		 const struct switchdev_obj_port_vlan *vlan,
+		 const refcount_t *refcount),
+
+	TP_ARGS(dp, vlan, refcount),
+
+	TP_STRUCT__entry(
+		__string(dev, dev_name(dp->ds->dev))
+		__string(kind, dsa_port_kind(dp))
+		__field(int, port)
+		__field(u16, vid)
+		__field(u16, flags)
+		__field(bool, changed)
+		__field(unsigned int, refcount)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, dev_name(dp->ds->dev));
+		__assign_str(kind, dsa_port_kind(dp));
+		__entry->port = dp->index;
+		__entry->vid = vlan->vid;
+		__entry->flags = vlan->flags;
+		__entry->changed = vlan->changed;
+		__entry->refcount = refcount_read(refcount);
+	),
+
+	TP_printk("%s %s port %d vid %u%s%s%s refcount %u",
+		  __get_str(dev), __get_str(kind), __entry->port, __entry->vid,
+		  __entry->flags & BRIDGE_VLAN_INFO_PVID ? " pvid" : "",
+		  __entry->flags & BRIDGE_VLAN_INFO_UNTAGGED ? " untagged" : "",
+		  __entry->changed ? " (changed)" : "", __entry->refcount)
+);
+
+DEFINE_EVENT(dsa_vlan_op_refcount, dsa_vlan_add_bump,
+	     TP_PROTO(const struct dsa_port *dp,
+		      const struct switchdev_obj_port_vlan *vlan,
+		      const refcount_t *refcount),
+	     TP_ARGS(dp, vlan, refcount));
+
+DEFINE_EVENT(dsa_vlan_op_refcount, dsa_vlan_del_drop,
+	     TP_PROTO(const struct dsa_port *dp,
+		      const struct switchdev_obj_port_vlan *vlan,
+		      const refcount_t *refcount),
+	     TP_ARGS(dp, vlan, refcount));
+
+TRACE_EVENT(dsa_vlan_del_not_found,
+
+	TP_PROTO(const struct dsa_port *dp,
+		 const struct switchdev_obj_port_vlan *vlan),
+
+	TP_ARGS(dp, vlan),
+
+	TP_STRUCT__entry(
+		__string(dev, dev_name(dp->ds->dev))
+		__string(kind, dsa_port_kind(dp))
+		__field(int, port)
+		__field(u16, vid)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev, dev_name(dp->ds->dev));
+		__assign_str(kind, dsa_port_kind(dp));
+		__entry->port = dp->index;
+		__entry->vid = vlan->vid;
+	),
+
+	TP_printk("%s %s port %d vid %u",
+		  __get_str(dev), __get_str(kind), __entry->port, __entry->vid)
+);
+
 #endif /* _NET_DSA_TRACE_H */
 
 /* We don't want to use include/trace/events */
-- 
cgit v1.2.3-70-g09d2


From ea449f7fa0bf3fcd02e04a770b9ff707bf5e8f96 Mon Sep 17 00:00:00 2001
From: Sebastian Reichel <sebastian.reichel@collabora.com>
Date: Fri, 7 Apr 2023 18:11:28 +0200
Subject: net: ethernet: stmmac: dwmac-rk: rework optional clock handling

The clock requesting code is quite repetitive. Fix this by requesting
the clocks via devm_clk_bulk_get_optional. The optional variant has been
used, since this is effectively what the old code did. The exact clocks
required depend on the platform and configuration. As a side effect
this change adds correct -EPROBE_DEFER handling.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Suggested-by: Andrew Lunn <andrew@lunn.ch>
Fixes: 7ad269ea1a2b ("GMAC: add driver for Rockchip RK3288 SoCs integrated GMAC")
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 183 ++++++++++---------------
 1 file changed, 70 insertions(+), 113 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
index 4b8fd11563e4..7ac9ca9b4935 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
@@ -39,6 +39,24 @@ struct rk_gmac_ops {
 	u32 regs[];
 };
 
+static const char * const rk_clocks[] = {
+	"aclk_mac", "pclk_mac", "mac_clk_tx", "clk_mac_speed",
+};
+
+static const char * const rk_rmii_clocks[] = {
+	"mac_clk_rx", "clk_mac_ref", "clk_mac_refout",
+};
+
+enum rk_clocks_index {
+	RK_ACLK_MAC = 0,
+	RK_PCLK_MAC,
+	RK_MAC_CLK_TX,
+	RK_CLK_MAC_SPEED,
+	RK_MAC_CLK_RX,
+	RK_CLK_MAC_REF,
+	RK_CLK_MAC_REFOUT,
+};
+
 struct rk_priv_data {
 	struct platform_device *pdev;
 	phy_interface_t phy_iface;
@@ -51,15 +69,9 @@ struct rk_priv_data {
 	bool clock_input;
 	bool integrated_phy;
 
+	struct clk_bulk_data *clks;
+	int num_clks;
 	struct clk *clk_mac;
-	struct clk *gmac_clkin;
-	struct clk *mac_clk_rx;
-	struct clk *mac_clk_tx;
-	struct clk *clk_mac_ref;
-	struct clk *clk_mac_refout;
-	struct clk *clk_mac_speed;
-	struct clk *aclk_mac;
-	struct clk *pclk_mac;
 	struct clk *clk_phy;
 
 	struct reset_control *phy_reset;
@@ -104,10 +116,11 @@ static void px30_set_to_rmii(struct rk_priv_data *bsp_priv)
 
 static void px30_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed)
 {
+	struct clk *clk_mac_speed = bsp_priv->clks[RK_CLK_MAC_SPEED].clk;
 	struct device *dev = &bsp_priv->pdev->dev;
 	int ret;
 
-	if (IS_ERR(bsp_priv->clk_mac_speed)) {
+	if (!clk_mac_speed) {
 		dev_err(dev, "%s: Missing clk_mac_speed clock\n", __func__);
 		return;
 	}
@@ -116,7 +129,7 @@ static void px30_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed)
 		regmap_write(bsp_priv->grf, PX30_GRF_GMAC_CON1,
 			     PX30_GMAC_SPEED_10M);
 
-		ret = clk_set_rate(bsp_priv->clk_mac_speed, 2500000);
+		ret = clk_set_rate(clk_mac_speed, 2500000);
 		if (ret)
 			dev_err(dev, "%s: set clk_mac_speed rate 2500000 failed: %d\n",
 				__func__, ret);
@@ -124,7 +137,7 @@ static void px30_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed)
 		regmap_write(bsp_priv->grf, PX30_GRF_GMAC_CON1,
 			     PX30_GMAC_SPEED_100M);
 
-		ret = clk_set_rate(bsp_priv->clk_mac_speed, 25000000);
+		ret = clk_set_rate(clk_mac_speed, 25000000);
 		if (ret)
 			dev_err(dev, "%s: set clk_mac_speed rate 25000000 failed: %d\n",
 				__func__, ret);
@@ -1066,6 +1079,7 @@ static void rk3568_set_to_rmii(struct rk_priv_data *bsp_priv)
 
 static void rk3568_set_gmac_speed(struct rk_priv_data *bsp_priv, int speed)
 {
+	struct clk *clk_mac_speed = bsp_priv->clks[RK_CLK_MAC_SPEED].clk;
 	struct device *dev = &bsp_priv->pdev->dev;
 	unsigned long rate;
 	int ret;
@@ -1085,7 +1099,7 @@ static void rk3568_set_gmac_speed(struct rk_priv_data *bsp_priv, int speed)
 		return;
 	}
 
-	ret = clk_set_rate(bsp_priv->clk_mac_speed, rate);
+	ret = clk_set_rate(clk_mac_speed, rate);
 	if (ret)
 		dev_err(dev, "%s: set clk_mac_speed rate %ld failed %d\n",
 			__func__, rate, ret);
@@ -1371,6 +1385,7 @@ static void rv1126_set_to_rmii(struct rk_priv_data *bsp_priv)
 
 static void rv1126_set_rgmii_speed(struct rk_priv_data *bsp_priv, int speed)
 {
+	struct clk *clk_mac_speed = bsp_priv->clks[RK_CLK_MAC_SPEED].clk;
 	struct device *dev = &bsp_priv->pdev->dev;
 	unsigned long rate;
 	int ret;
@@ -1390,7 +1405,7 @@ static void rv1126_set_rgmii_speed(struct rk_priv_data *bsp_priv, int speed)
 		return;
 	}
 
-	ret = clk_set_rate(bsp_priv->clk_mac_speed, rate);
+	ret = clk_set_rate(clk_mac_speed, rate);
 	if (ret)
 		dev_err(dev, "%s: set clk_mac_speed rate %ld failed %d\n",
 			__func__, rate, ret);
@@ -1398,6 +1413,7 @@ static void rv1126_set_rgmii_speed(struct rk_priv_data *bsp_priv, int speed)
 
 static void rv1126_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed)
 {
+	struct clk *clk_mac_speed = bsp_priv->clks[RK_CLK_MAC_SPEED].clk;
 	struct device *dev = &bsp_priv->pdev->dev;
 	unsigned long rate;
 	int ret;
@@ -1414,7 +1430,7 @@ static void rv1126_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed)
 		return;
 	}
 
-	ret = clk_set_rate(bsp_priv->clk_mac_speed, rate);
+	ret = clk_set_rate(clk_mac_speed, rate);
 	if (ret)
 		dev_err(dev, "%s: set clk_mac_speed rate %ld failed %d\n",
 			__func__, rate, ret);
@@ -1475,68 +1491,50 @@ static int rk_gmac_clk_init(struct plat_stmmacenet_data *plat)
 {
 	struct rk_priv_data *bsp_priv = plat->bsp_priv;
 	struct device *dev = &bsp_priv->pdev->dev;
-	int ret;
+	int phy_iface = bsp_priv->phy_iface;
+	int i, j, ret;
 
 	bsp_priv->clk_enabled = false;
 
-	bsp_priv->mac_clk_rx = devm_clk_get(dev, "mac_clk_rx");
-	if (IS_ERR(bsp_priv->mac_clk_rx))
-		dev_err(dev, "cannot get clock %s\n",
-			"mac_clk_rx");
-
-	bsp_priv->mac_clk_tx = devm_clk_get(dev, "mac_clk_tx");
-	if (IS_ERR(bsp_priv->mac_clk_tx))
-		dev_err(dev, "cannot get clock %s\n",
-			"mac_clk_tx");
+	bsp_priv->num_clks = ARRAY_SIZE(rk_clocks);
+	if (phy_iface == PHY_INTERFACE_MODE_RMII)
+		bsp_priv->num_clks += ARRAY_SIZE(rk_rmii_clocks);
 
-	bsp_priv->aclk_mac = devm_clk_get(dev, "aclk_mac");
-	if (IS_ERR(bsp_priv->aclk_mac))
-		dev_err(dev, "cannot get clock %s\n",
-			"aclk_mac");
+	bsp_priv->clks = devm_kcalloc(dev, bsp_priv->num_clks,
+				      sizeof(*bsp_priv->clks), GFP_KERNEL);
+	if (!bsp_priv->clks)
+		return -ENOMEM;
 
-	bsp_priv->pclk_mac = devm_clk_get(dev, "pclk_mac");
-	if (IS_ERR(bsp_priv->pclk_mac))
-		dev_err(dev, "cannot get clock %s\n",
-			"pclk_mac");
+	for (i = 0; i < ARRAY_SIZE(rk_clocks); i++)
+		bsp_priv->clks[i].id = rk_clocks[i];
 
-	bsp_priv->clk_mac = devm_clk_get(dev, "stmmaceth");
-	if (IS_ERR(bsp_priv->clk_mac))
-		dev_err(dev, "cannot get clock %s\n",
-			"stmmaceth");
-
-	if (bsp_priv->phy_iface == PHY_INTERFACE_MODE_RMII) {
-		bsp_priv->clk_mac_ref = devm_clk_get(dev, "clk_mac_ref");
-		if (IS_ERR(bsp_priv->clk_mac_ref))
-			dev_err(dev, "cannot get clock %s\n",
-				"clk_mac_ref");
-
-		if (!bsp_priv->clock_input) {
-			bsp_priv->clk_mac_refout =
-				devm_clk_get(dev, "clk_mac_refout");
-			if (IS_ERR(bsp_priv->clk_mac_refout))
-				dev_err(dev, "cannot get clock %s\n",
-					"clk_mac_refout");
-		}
+	if (phy_iface == PHY_INTERFACE_MODE_RMII) {
+		for (j = 0; j < ARRAY_SIZE(rk_rmii_clocks); j++)
+			bsp_priv->clks[i++].id = rk_rmii_clocks[j];
 	}
 
-	bsp_priv->clk_mac_speed = devm_clk_get(dev, "clk_mac_speed");
-	if (IS_ERR(bsp_priv->clk_mac_speed))
-		dev_err(dev, "cannot get clock %s\n", "clk_mac_speed");
+	ret = devm_clk_bulk_get_optional(dev, bsp_priv->num_clks,
+					 bsp_priv->clks);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get clocks\n");
+
+	/* "stmmaceth" will be enabled by the core */
+	bsp_priv->clk_mac = devm_clk_get(dev, "stmmaceth");
+	ret = PTR_ERR_OR_ZERO(bsp_priv->clk_mac);
+	if (ret)
+		return dev_err_probe(dev, ret, "Cannot get stmmaceth clock\n");
 
 	if (bsp_priv->clock_input) {
 		dev_info(dev, "clock input from PHY\n");
-	} else {
-		if (bsp_priv->phy_iface == PHY_INTERFACE_MODE_RMII)
-			clk_set_rate(bsp_priv->clk_mac, 50000000);
+	} else if (phy_iface == PHY_INTERFACE_MODE_RMII) {
+		clk_set_rate(bsp_priv->clk_mac, 50000000);
 	}
 
 	if (plat->phy_node && bsp_priv->integrated_phy) {
 		bsp_priv->clk_phy = of_clk_get(plat->phy_node, 0);
-		if (IS_ERR(bsp_priv->clk_phy)) {
-			ret = PTR_ERR(bsp_priv->clk_phy);
-			dev_err(dev, "Cannot get PHY clock: %d\n", ret);
-			return -EINVAL;
-		}
+		ret = PTR_ERR_OR_ZERO(bsp_priv->clk_phy);
+		if (ret)
+			return dev_err_probe(dev, ret, "Cannot get PHY clock\n");
 		clk_set_rate(bsp_priv->clk_phy, 50000000);
 	}
 
@@ -1545,77 +1543,36 @@ static int rk_gmac_clk_init(struct plat_stmmacenet_data *plat)
 
 static int gmac_clk_enable(struct rk_priv_data *bsp_priv, bool enable)
 {
-	int phy_iface = bsp_priv->phy_iface;
+	int ret;
 
 	if (enable) {
 		if (!bsp_priv->clk_enabled) {
-			if (phy_iface == PHY_INTERFACE_MODE_RMII) {
-				if (!IS_ERR(bsp_priv->mac_clk_rx))
-					clk_prepare_enable(
-						bsp_priv->mac_clk_rx);
-
-				if (!IS_ERR(bsp_priv->clk_mac_ref))
-					clk_prepare_enable(
-						bsp_priv->clk_mac_ref);
-
-				if (!IS_ERR(bsp_priv->clk_mac_refout))
-					clk_prepare_enable(
-						bsp_priv->clk_mac_refout);
-			}
-
-			if (!IS_ERR(bsp_priv->clk_phy))
-				clk_prepare_enable(bsp_priv->clk_phy);
-
-			if (!IS_ERR(bsp_priv->aclk_mac))
-				clk_prepare_enable(bsp_priv->aclk_mac);
-
-			if (!IS_ERR(bsp_priv->pclk_mac))
-				clk_prepare_enable(bsp_priv->pclk_mac);
-
-			if (!IS_ERR(bsp_priv->mac_clk_tx))
-				clk_prepare_enable(bsp_priv->mac_clk_tx);
+			ret = clk_bulk_prepare_enable(bsp_priv->num_clks,
+						      bsp_priv->clks);
+			if (ret)
+				return ret;
 
-			if (!IS_ERR(bsp_priv->clk_mac_speed))
-				clk_prepare_enable(bsp_priv->clk_mac_speed);
+			ret = clk_prepare_enable(bsp_priv->clk_phy);
+			if (ret)
+				return ret;
 
 			if (bsp_priv->ops && bsp_priv->ops->set_clock_selection)
 				bsp_priv->ops->set_clock_selection(bsp_priv,
 					       bsp_priv->clock_input, true);
 
-			/**
-			 * if (!IS_ERR(bsp_priv->clk_mac))
-			 *	clk_prepare_enable(bsp_priv->clk_mac);
-			 */
 			mdelay(5);
 			bsp_priv->clk_enabled = true;
 		}
 	} else {
 		if (bsp_priv->clk_enabled) {
-			if (phy_iface == PHY_INTERFACE_MODE_RMII) {
-				clk_disable_unprepare(bsp_priv->mac_clk_rx);
-
-				clk_disable_unprepare(bsp_priv->clk_mac_ref);
-
-				clk_disable_unprepare(bsp_priv->clk_mac_refout);
-			}
-
+			clk_bulk_disable_unprepare(bsp_priv->num_clks,
+						   bsp_priv->clks);
 			clk_disable_unprepare(bsp_priv->clk_phy);
 
-			clk_disable_unprepare(bsp_priv->aclk_mac);
-
-			clk_disable_unprepare(bsp_priv->pclk_mac);
-
-			clk_disable_unprepare(bsp_priv->mac_clk_tx);
-
-			clk_disable_unprepare(bsp_priv->clk_mac_speed);
-
 			if (bsp_priv->ops && bsp_priv->ops->set_clock_selection)
 				bsp_priv->ops->set_clock_selection(bsp_priv,
 					      bsp_priv->clock_input, false);
-			/**
-			 * if (!IS_ERR(bsp_priv->clk_mac))
-			 *	clk_disable_unprepare(bsp_priv->clk_mac);
-			 */
+
 			bsp_priv->clk_enabled = false;
 		}
 	}
-- 
cgit v1.2.3-70-g09d2


From db21973263f8c56750cb610f1d5e8bee00a513b9 Mon Sep 17 00:00:00 2001
From: Sebastian Reichel <sebastian.reichel@collabora.com>
Date: Fri, 7 Apr 2023 18:11:29 +0200
Subject: net: ethernet: stmmac: dwmac-rk: fix optional phy regulator handling

The usual devm_regulator_get() call already handles "optional"
regulators by returning a valid dummy and printing a warning
that the dummy regulator should be described properly. This
code open coded the same behaviour, but masked any errors that
are not -EPROBE_DEFER and is quite noisy.

This change effectively unmasks and propagates regulators errors
not involving -ENODEV, downgrades the error print to warning level
if no regulator is specified and captures the probe defer message
for /sys/kernel/debug/devices_deferred.

Fixes: 2e12f536635f ("net: stmmac: dwmac-rk: Use standard devicetree property for phy regulator")
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
index 7ac9ca9b4935..4ea31ccf24d0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
@@ -1586,9 +1586,6 @@ static int phy_power_on(struct rk_priv_data *bsp_priv, bool enable)
 	int ret;
 	struct device *dev = &bsp_priv->pdev->dev;
 
-	if (!ldo)
-		return 0;
-
 	if (enable) {
 		ret = regulator_enable(ldo);
 		if (ret)
@@ -1636,14 +1633,11 @@ static struct rk_priv_data *rk_gmac_setup(struct platform_device *pdev,
 		}
 	}
 
-	bsp_priv->regulator = devm_regulator_get_optional(dev, "phy");
+	bsp_priv->regulator = devm_regulator_get(dev, "phy");
 	if (IS_ERR(bsp_priv->regulator)) {
-		if (PTR_ERR(bsp_priv->regulator) == -EPROBE_DEFER) {
-			dev_err(dev, "phy regulator is not available yet, deferred probing\n");
-			return ERR_PTR(-EPROBE_DEFER);
-		}
-		dev_err(dev, "no regulator found\n");
-		bsp_priv->regulator = NULL;
+		ret = PTR_ERR(bsp_priv->regulator);
+		dev_err_probe(dev, ret, "failed to get phy regulator\n");
+		return ERR_PTR(ret);
 	}
 
 	ret = of_property_read_string(dev->of_node, "clock_in_out", &strings);
-- 
cgit v1.2.3-70-g09d2


From be690daa224ec41889bf8f293a5bcd8a3ea6a420 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Fri, 7 Apr 2023 11:45:39 -0700
Subject: ionic: Don't overwrite the cyclecounter bitmask

The driver was incorrectly overwriting the cyclecounter bitmask,
which was truncating it and not aligning to the hardware mask value.
This isn't causing any issues, but it's wrong. Fix this by not
constraining the cyclecounter/hardware mask.

Luckily, this seems to cause no issues, which is why this change
doesn't have a fixes tag and isn't being sent to net. However, if
any transformations from time->cycles are needed in the future,
this change will be needed.

Suggested-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/pensando/ionic/ionic_phc.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_phc.c b/drivers/net/ethernet/pensando/ionic/ionic_phc.c
index eac2f0e3576e..7505efdff8e9 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_phc.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_phc.c
@@ -579,11 +579,10 @@ void ionic_lif_alloc_phc(struct ionic_lif *lif)
 	diff |= diff >> 16;
 	diff |= diff >> 32;
 
-	/* constrain to the hardware bitmask, and use this as the bitmask */
+	/* constrain to the hardware bitmask */
 	diff &= phc->cc.mask;
-	phc->cc.mask = diff;
 
-	/* the wrap period is now defined by diff (or phc->cc.mask)
+	/* the wrap period is now defined by diff
 	 *
 	 * we will update the time basis at about 1/4 the wrap period, so
 	 * should not see a difference of more than +/- diff/4.
-- 
cgit v1.2.3-70-g09d2


From ed72bd5a6790a0c3747cb32b0427f921bd03bb71 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Sat, 8 Apr 2023 09:47:54 +0200
Subject: ksz884x: Remove unused functions

Remove unused functions.

These functions may have some value in documenting the
hardware. But that information may be accessed via SCM history.

Flagged by clang-16 with W=1.
No functional change intended.
Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/micrel/ksz884x.c | 294 ----------------------------------
 1 file changed, 294 deletions(-)

diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c
index f78e8ead8c36..c5aeeb964c17 100644
--- a/drivers/net/ethernet/micrel/ksz884x.c
+++ b/drivers/net/ethernet/micrel/ksz884x.c
@@ -1476,15 +1476,6 @@ static void hw_turn_on_intr(struct ksz_hw *hw, u32 bit)
 		hw_set_intr(hw, hw->intr_mask);
 }
 
-static inline void hw_ena_intr_bit(struct ksz_hw *hw, uint interrupt)
-{
-	u32 read_intr;
-
-	read_intr = readl(hw->io + KS884X_INTERRUPTS_ENABLE);
-	hw->intr_set = read_intr | interrupt;
-	writel(hw->intr_set, hw->io + KS884X_INTERRUPTS_ENABLE);
-}
-
 static inline void hw_read_intr(struct ksz_hw *hw, uint *status)
 {
 	*status = readl(hw->io + KS884X_INTERRUPTS_STATUS);
@@ -1853,29 +1844,6 @@ static void port_init_cnt(struct ksz_hw *hw, int port)
  * Port functions
  */
 
-/**
- * port_chk - check port register bits
- * @hw: 	The hardware instance.
- * @port:	The port index.
- * @offset:	The offset of the port register.
- * @bits:	The data bits to check.
- *
- * This function checks whether the specified bits of the port register are set
- * or not.
- *
- * Return 0 if the bits are not set.
- */
-static int port_chk(struct ksz_hw *hw, int port, int offset, u16 bits)
-{
-	u32 addr;
-	u16 data;
-
-	PORT_CTRL_ADDR(port, addr);
-	addr += offset;
-	data = readw(hw->io + addr);
-	return (data & bits) == bits;
-}
-
 /**
  * port_cfg - set port register bits
  * @hw: 	The hardware instance.
@@ -1902,53 +1870,6 @@ static void port_cfg(struct ksz_hw *hw, int port, int offset, u16 bits,
 	writew(data, hw->io + addr);
 }
 
-/**
- * port_chk_shift - check port bit
- * @hw: 	The hardware instance.
- * @port:	The port index.
- * @addr:	The offset of the register.
- * @shift:	Number of bits to shift.
- *
- * This function checks whether the specified port is set in the register or
- * not.
- *
- * Return 0 if the port is not set.
- */
-static int port_chk_shift(struct ksz_hw *hw, int port, u32 addr, int shift)
-{
-	u16 data;
-	u16 bit = 1 << port;
-
-	data = readw(hw->io + addr);
-	data >>= shift;
-	return (data & bit) == bit;
-}
-
-/**
- * port_cfg_shift - set port bit
- * @hw: 	The hardware instance.
- * @port:	The port index.
- * @addr:	The offset of the register.
- * @shift:	Number of bits to shift.
- * @set:	The flag indicating whether the port is to be set or not.
- *
- * This routine sets or resets the specified port in the register.
- */
-static void port_cfg_shift(struct ksz_hw *hw, int port, u32 addr, int shift,
-	int set)
-{
-	u16 data;
-	u16 bits = 1 << port;
-
-	data = readw(hw->io + addr);
-	bits <<= shift;
-	if (set)
-		data |= bits;
-	else
-		data &= ~bits;
-	writew(data, hw->io + addr);
-}
-
 /**
  * port_r8 - read byte from port register
  * @hw: 	The hardware instance.
@@ -2051,12 +1972,6 @@ static inline void port_cfg_broad_storm(struct ksz_hw *hw, int p, int set)
 		KS8842_PORT_CTRL_1_OFFSET, PORT_BROADCAST_STORM, set);
 }
 
-static inline int port_chk_broad_storm(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_BROADCAST_STORM);
-}
-
 /* Driver set switch broadcast storm protection at 10% rate. */
 #define BROADCAST_STORM_PROTECTION_RATE	10
 
@@ -2209,102 +2124,6 @@ static inline void port_cfg_back_pressure(struct ksz_hw *hw, int p, int set)
 		KS8842_PORT_CTRL_2_OFFSET, PORT_BACK_PRESSURE, set);
 }
 
-static inline void port_cfg_force_flow_ctrl(struct ksz_hw *hw, int p, int set)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_FORCE_FLOW_CTRL, set);
-}
-
-static inline int port_chk_back_pressure(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_BACK_PRESSURE);
-}
-
-static inline int port_chk_force_flow_ctrl(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_FORCE_FLOW_CTRL);
-}
-
-/* Spanning Tree */
-
-static inline void port_cfg_rx(struct ksz_hw *hw, int p, int set)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_RX_ENABLE, set);
-}
-
-static inline void port_cfg_tx(struct ksz_hw *hw, int p, int set)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_TX_ENABLE, set);
-}
-
-static inline void sw_cfg_fast_aging(struct ksz_hw *hw, int set)
-{
-	sw_cfg(hw, KS8842_SWITCH_CTRL_1_OFFSET, SWITCH_FAST_AGING, set);
-}
-
-static inline void sw_flush_dyn_mac_table(struct ksz_hw *hw)
-{
-	if (!(hw->overrides & FAST_AGING)) {
-		sw_cfg_fast_aging(hw, 1);
-		mdelay(1);
-		sw_cfg_fast_aging(hw, 0);
-	}
-}
-
-/* VLAN */
-
-static inline void port_cfg_ins_tag(struct ksz_hw *hw, int p, int insert)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_INSERT_TAG, insert);
-}
-
-static inline void port_cfg_rmv_tag(struct ksz_hw *hw, int p, int remove)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_REMOVE_TAG, remove);
-}
-
-static inline int port_chk_ins_tag(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_INSERT_TAG);
-}
-
-static inline int port_chk_rmv_tag(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_REMOVE_TAG);
-}
-
-static inline void port_cfg_dis_non_vid(struct ksz_hw *hw, int p, int set)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_DISCARD_NON_VID, set);
-}
-
-static inline void port_cfg_in_filter(struct ksz_hw *hw, int p, int set)
-{
-	port_cfg(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_INGRESS_VLAN_FILTER, set);
-}
-
-static inline int port_chk_dis_non_vid(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_DISCARD_NON_VID);
-}
-
-static inline int port_chk_in_filter(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_INGRESS_VLAN_FILTER);
-}
-
 /* Mirroring */
 
 static inline void port_cfg_mirror_sniffer(struct ksz_hw *hw, int p, int set)
@@ -2342,28 +2161,6 @@ static void sw_init_mirror(struct ksz_hw *hw)
 	sw_cfg_mirror_rx_tx(hw, 0);
 }
 
-static inline void sw_cfg_unk_def_deliver(struct ksz_hw *hw, int set)
-{
-	sw_cfg(hw, KS8842_SWITCH_CTRL_7_OFFSET,
-		SWITCH_UNK_DEF_PORT_ENABLE, set);
-}
-
-static inline int sw_cfg_chk_unk_def_deliver(struct ksz_hw *hw)
-{
-	return sw_chk(hw, KS8842_SWITCH_CTRL_7_OFFSET,
-		SWITCH_UNK_DEF_PORT_ENABLE);
-}
-
-static inline void sw_cfg_unk_def_port(struct ksz_hw *hw, int port, int set)
-{
-	port_cfg_shift(hw, port, KS8842_SWITCH_CTRL_7_OFFSET, 0, set);
-}
-
-static inline int sw_chk_unk_def_port(struct ksz_hw *hw, int port)
-{
-	return port_chk_shift(hw, port, KS8842_SWITCH_CTRL_7_OFFSET, 0);
-}
-
 /* Priority */
 
 static inline void port_cfg_diffserv(struct ksz_hw *hw, int p, int set)
@@ -2390,30 +2187,6 @@ static inline void port_cfg_prio(struct ksz_hw *hw, int p, int set)
 		KS8842_PORT_CTRL_1_OFFSET, PORT_PRIO_QUEUE_ENABLE, set);
 }
 
-static inline int port_chk_diffserv(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_DIFFSERV_ENABLE);
-}
-
-static inline int port_chk_802_1p(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_802_1P_ENABLE);
-}
-
-static inline int port_chk_replace_vid(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_2_OFFSET, PORT_USER_PRIORITY_CEILING);
-}
-
-static inline int port_chk_prio(struct ksz_hw *hw, int p)
-{
-	return port_chk(hw, p,
-		KS8842_PORT_CTRL_1_OFFSET, PORT_PRIO_QUEUE_ENABLE);
-}
-
 /**
  * sw_dis_diffserv - disable switch DiffServ priority
  * @hw: 	The hardware instance.
@@ -2613,23 +2386,6 @@ static void sw_cfg_port_base_vlan(struct ksz_hw *hw, int port, u8 member)
 	hw->ksz_switch->port_cfg[port].member = member;
 }
 
-/**
- * sw_get_addr - get the switch MAC address.
- * @hw: 	The hardware instance.
- * @mac_addr:	Buffer to store the MAC address.
- *
- * This function retrieves the MAC address of the switch.
- */
-static inline void sw_get_addr(struct ksz_hw *hw, u8 *mac_addr)
-{
-	int i;
-
-	for (i = 0; i < 6; i += 2) {
-		mac_addr[i] = readb(hw->io + KS8842_MAC_ADDR_0_OFFSET + i);
-		mac_addr[1 + i] = readb(hw->io + KS8842_MAC_ADDR_1_OFFSET + i);
-	}
-}
-
 /**
  * sw_set_addr - configure switch MAC address
  * @hw: 	The hardware instance.
@@ -2828,56 +2584,6 @@ static inline void hw_w_phy_ctrl(struct ksz_hw *hw, int phy, u16 data)
 	writew(data, hw->io + phy + KS884X_PHY_CTRL_OFFSET);
 }
 
-static inline void hw_r_phy_link_stat(struct ksz_hw *hw, int phy, u16 *data)
-{
-	*data = readw(hw->io + phy + KS884X_PHY_STATUS_OFFSET);
-}
-
-static inline void hw_r_phy_auto_neg(struct ksz_hw *hw, int phy, u16 *data)
-{
-	*data = readw(hw->io + phy + KS884X_PHY_AUTO_NEG_OFFSET);
-}
-
-static inline void hw_w_phy_auto_neg(struct ksz_hw *hw, int phy, u16 data)
-{
-	writew(data, hw->io + phy + KS884X_PHY_AUTO_NEG_OFFSET);
-}
-
-static inline void hw_r_phy_rem_cap(struct ksz_hw *hw, int phy, u16 *data)
-{
-	*data = readw(hw->io + phy + KS884X_PHY_REMOTE_CAP_OFFSET);
-}
-
-static inline void hw_r_phy_crossover(struct ksz_hw *hw, int phy, u16 *data)
-{
-	*data = readw(hw->io + phy + KS884X_PHY_CTRL_OFFSET);
-}
-
-static inline void hw_w_phy_crossover(struct ksz_hw *hw, int phy, u16 data)
-{
-	writew(data, hw->io + phy + KS884X_PHY_CTRL_OFFSET);
-}
-
-static inline void hw_r_phy_polarity(struct ksz_hw *hw, int phy, u16 *data)
-{
-	*data = readw(hw->io + phy + KS884X_PHY_PHY_CTRL_OFFSET);
-}
-
-static inline void hw_w_phy_polarity(struct ksz_hw *hw, int phy, u16 data)
-{
-	writew(data, hw->io + phy + KS884X_PHY_PHY_CTRL_OFFSET);
-}
-
-static inline void hw_r_phy_link_md(struct ksz_hw *hw, int phy, u16 *data)
-{
-	*data = readw(hw->io + phy + KS884X_PHY_LINK_MD_OFFSET);
-}
-
-static inline void hw_w_phy_link_md(struct ksz_hw *hw, int phy, u16 data)
-{
-	writew(data, hw->io + phy + KS884X_PHY_LINK_MD_OFFSET);
-}
-
 /**
  * hw_r_phy - read data from PHY register
  * @hw: 	The hardware instance.
-- 
cgit v1.2.3-70-g09d2


From 8b4d2f080afbd4280ecca0f4b3ceea943a7a86d0 Mon Sep 17 00:00:00 2001
From: Manikanta Pubbisetty <quic_mpubbise@quicinc.com>
Date: Thu, 23 Mar 2023 11:39:13 +0530
Subject: wifi: ath11k: Optimize 6 GHz scan time

Currently, time taken to scan all supported channels on WCN6750
is ~8 seconds and connection time is almost 10 seconds. WCN6750
supports three Wi-Fi bands (i.e., 2.4/5/6 GHz) and the numbers of
channels for scan come around ~100 channels (default case).
Since the chip doesn't have support for DBS (Dual Band Simultaneous),
scans cannot be parallelized resulting in longer scan times.

Among the 100 odd channels, ~60 channels are in 6 GHz band. Therefore,
optimizing the scan for 6 GHz channels will bring down the overall
scan time.

WCN6750 firmware has support to scan a 6 GHz channel based on co-located
AP information i.e., RNR IE which is found in the legacy 2.4/5 GHz scan
results. When a scan request with all supported channel list is enqueued
to the firmware, then based on WMI_SCAN_CHAN_FLAG_SCAN_ONLY_IF_RNR_FOUND
scan channel flag, firmware will scan only those 6 GHz channels for which
RNR IEs are found in the legacy scan results.

In the proposed design, based on NL80211_SCAN_FLAG_COLOCATED_6GHZ scan
flag, driver will set the WMI_SCAN_CHAN_FLAG_SCAN_ONLY_IF_RNR_FOUND flag
for non-PSC channels. Since there is high probability to find 6 GHz APs
on PSC channels, these channels are always scanned. Only non-PSC channels
are selectively scanned based on cached RNR information from the legacy
scan results.

If NL80211_SCAN_FLAG_COLOCATED_6GHZ is not set in the scan flags,
then scan will happen on all supported channels (default behavior).

With these optimizations, scan time is improved by 1.5-1.8 seconds on
WCN6750. Similar savings have been observed on WCN6855.

Tested-on: WCN6750 hw1.0 AHB WLAN.MSL.1.0.1-00887-QCAMSLSWPLZ-1
Tested-on: WCN6855 hw2.1 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.16

Signed-off-by: Manikanta Pubbisetty <quic_mpubbise@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230323060913.10097-1-quic_mpubbise@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 25 +++++++++++++++++++++++--
 drivers/net/wireless/ath/ath11k/wmi.h |  4 ++++
 2 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index cad832e0e6b8..c81b238d0d91 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3819,8 +3819,29 @@ static int ath11k_mac_op_hw_scan(struct ieee80211_hw *hw,
 			goto exit;
 		}
 
-		for (i = 0; i < arg->num_chan; i++)
-			arg->chan_list[i] = req->channels[i]->center_freq;
+		for (i = 0; i < arg->num_chan; i++) {
+			if (test_bit(WMI_TLV_SERVICE_SCAN_CONFIG_PER_CHANNEL,
+				     ar->ab->wmi_ab.svc_map)) {
+				arg->chan_list[i] =
+					u32_encode_bits(req->channels[i]->center_freq,
+							WMI_SCAN_CONFIG_PER_CHANNEL_MASK);
+
+				/* If NL80211_SCAN_FLAG_COLOCATED_6GHZ is set in scan
+				 * flags, then scan all PSC channels in 6 GHz band and
+				 * those non-PSC channels where RNR IE is found during
+				 * the legacy 2.4/5 GHz scan.
+				 * If NL80211_SCAN_FLAG_COLOCATED_6GHZ is not set,
+				 * then all channels in 6 GHz will be scanned.
+				 */
+				if (req->channels[i]->band == NL80211_BAND_6GHZ &&
+				    req->flags & NL80211_SCAN_FLAG_COLOCATED_6GHZ &&
+				    !cfg80211_channel_is_psc(req->channels[i]))
+					arg->chan_list[i] |=
+						WMI_SCAN_CH_FLAG_SCAN_ONLY_IF_RNR_FOUND;
+			} else {
+				arg->chan_list[i] = req->channels[i]->center_freq;
+			}
+		}
 	}
 
 	if (req->flags & NL80211_SCAN_FLAG_RANDOM_ADDR) {
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index b23b7a22bc9a..86b22f2cec12 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2100,6 +2100,7 @@ enum wmi_tlv_service {
 
 	/* The second 128 bits */
 	WMI_MAX_EXT_SERVICE = 256,
+	WMI_TLV_SERVICE_SCAN_CONFIG_PER_CHANNEL = 265,
 	WMI_TLV_SERVICE_REG_CC_EXT_EVENT_SUPPORT = 281,
 	WMI_TLV_SERVICE_BIOS_SAR_SUPPORT = 326,
 
@@ -3249,6 +3250,9 @@ struct  wmi_start_scan_cmd {
 #define WMI_SCAN_DWELL_MODE_SHIFT        21
 #define WMI_SCAN_FLAG_EXT_PASSIVE_SCAN_START_TIME_ENHANCE   0x00000800
 
+#define WMI_SCAN_CONFIG_PER_CHANNEL_MASK	GENMASK(19, 0)
+#define WMI_SCAN_CH_FLAG_SCAN_ONLY_IF_RNR_FOUND	BIT(20)
+
 enum {
 	WMI_SCAN_DWELL_MODE_DEFAULT      = 0,
 	WMI_SCAN_DWELL_MODE_CONSERVATIVE = 1,
-- 
cgit v1.2.3-70-g09d2


From 813968c24126cc5c8320cd5db0e262069a535063 Mon Sep 17 00:00:00 2001
From: Ganesh Babu Jothiram <quic_gjothira@quicinc.com>
Date: Fri, 24 Mar 2023 16:57:00 +0200
Subject: wifi: ath11k: Configure the FTM responder role using firmware
 capability flag

Fine Time Measurement(FTM) is offloaded feature to firmware.
Hence, the configuration of FTM responder role is done using
firmware capability flag instead of hw param.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Ganesh Babu Jothiram <quic_gjothira@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230317072034.8217-1-quic_gjothira@quicinc.com
---
 drivers/net/wireless/ath/ath11k/core.c | 8 --------
 drivers/net/wireless/ath/ath11k/hw.h   | 1 -
 drivers/net/wireless/ath/ath11k/mac.c  | 4 ++--
 3 files changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index 75fdbe4ef83a..498310e38d44 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -116,7 +116,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tcl_ring_retry = true,
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
-		.ftm_responder = true,
 	},
 	{
 		.hw_rev = ATH11K_HW_IPQ6018_HW10,
@@ -199,7 +198,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
 		.support_fw_mac_sequence = false,
-		.ftm_responder = true,
 	},
 	{
 		.name = "qca6390 hw2.0",
@@ -284,7 +282,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
 		.support_fw_mac_sequence = true,
-		.ftm_responder = false,
 	},
 	{
 		.name = "qcn9074 hw1.0",
@@ -366,7 +363,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
 		.support_fw_mac_sequence = false,
-		.ftm_responder = true,
 	},
 	{
 		.name = "wcn6855 hw2.0",
@@ -451,7 +447,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
 		.support_fw_mac_sequence = true,
-		.ftm_responder = false,
 	},
 	{
 		.name = "wcn6855 hw2.1",
@@ -534,7 +529,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
 		.support_fw_mac_sequence = true,
-		.ftm_responder = false,
 	},
 	{
 		.name = "wcn6750 hw1.0",
@@ -615,7 +609,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE_WCN6750,
 		.smp2p_wow_exit = true,
 		.support_fw_mac_sequence = true,
-		.ftm_responder = false,
 	},
 	{
 		.hw_rev = ATH11K_HW_IPQ5018_HW10,
@@ -695,7 +688,6 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.tx_ring_size = DP_TCL_DATA_RING_SIZE,
 		.smp2p_wow_exit = false,
 		.support_fw_mac_sequence = false,
-		.ftm_responder = true,
 	},
 };
 
diff --git a/drivers/net/wireless/ath/ath11k/hw.h b/drivers/net/wireless/ath/ath11k/hw.h
index 0be4e1232384..9f45d061d826 100644
--- a/drivers/net/wireless/ath/ath11k/hw.h
+++ b/drivers/net/wireless/ath/ath11k/hw.h
@@ -224,7 +224,6 @@ struct ath11k_hw_params {
 	u32 tx_ring_size;
 	bool smp2p_wow_exit;
 	bool support_fw_mac_sequence;
-	bool ftm_responder;
 };
 
 struct ath11k_hw_ops {
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index c81b238d0d91..7226d88c0c48 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3538,7 +3538,7 @@ static void ath11k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 
 	if (changed & BSS_CHANGED_FTM_RESPONDER &&
 	    arvif->ftm_responder != info->ftm_responder &&
-	    ar->ab->hw_params.ftm_responder &&
+	    test_bit(WMI_TLV_SERVICE_RTT, ar->ab->wmi_ab.svc_map) &&
 	    (vif->type == NL80211_IFTYPE_AP ||
 	     vif->type == NL80211_IFTYPE_MESH_POINT)) {
 		arvif->ftm_responder = info->ftm_responder;
@@ -9234,7 +9234,7 @@ static int __ath11k_mac_register(struct ath11k *ar)
 	wiphy_ext_feature_set(ar->hw->wiphy,
 			      NL80211_EXT_FEATURE_SET_SCAN_DWELL);
 
-	if (ab->hw_params.ftm_responder)
+	if (test_bit(WMI_TLV_SERVICE_RTT, ar->ab->wmi_ab.svc_map))
 		wiphy_ext_feature_set(ar->hw->wiphy,
 				      NL80211_EXT_FEATURE_ENABLE_FTM_RESPONDER);
 
-- 
cgit v1.2.3-70-g09d2


From 031ffa6c2cd305a57ccc6d610f2decd956b2e7f6 Mon Sep 17 00:00:00 2001
From: P Praneesh <quic_ppranees@quicinc.com>
Date: Fri, 24 Mar 2023 16:57:00 +0200
Subject: wifi: ath11k: fix rssi station dump not updated in QCN9074

In QCN9074, station dump signal values display default value which
is -95 dbm, since there is firmware header change for HAL_RX_MPDU_START
between QCN9074 and IPQ8074 which cause wrong peer_id fetch from msdu.
Fix this by updating hal_rx_mpdu_info with corresponding QCN9074 tlv
format.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.4.0.1-01695-QCAHKSWPL_SILICONZ-1

Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230320110312.20639-1-quic_ppranees@quicinc.com
---
 drivers/net/wireless/ath/ath11k/hal_rx.c | 10 +++++++++-
 drivers/net/wireless/ath/ath11k/hal_rx.h | 18 +++++++++++++++++-
 drivers/net/wireless/ath/ath11k/hw.c     | 27 ++++++++++++++++++---------
 drivers/net/wireless/ath/ath11k/hw.h     |  2 +-
 4 files changed, 45 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.c b/drivers/net/wireless/ath/ath11k/hal_rx.c
index 7f39c6fb7408..b9939c5a4ecd 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.c
@@ -865,6 +865,12 @@ ath11k_hal_rx_populate_mu_user_info(void *rx_tlv, struct hal_rx_mon_ppdu_info *p
 	ath11k_hal_rx_populate_byte_count(rx_tlv, ppdu_info, rx_user_status);
 }
 
+static u16 ath11k_hal_rx_mpduinfo_get_peerid(struct ath11k_base *ab,
+					     struct hal_rx_mpdu_info *mpdu_info)
+{
+	return ab->hw_params.hw_ops->mpdu_info_get_peerid(mpdu_info);
+}
+
 static enum hal_rx_mon_status
 ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
 				   struct hal_rx_mon_ppdu_info *ppdu_info,
@@ -1459,9 +1465,11 @@ ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
 		break;
 	}
 	case HAL_RX_MPDU_START: {
+		struct hal_rx_mpdu_info *mpdu_info =
+				(struct hal_rx_mpdu_info *)tlv_data;
 		u16 peer_id;
 
-		peer_id = ab->hw_params.hw_ops->mpdu_info_get_peerid(tlv_data);
+		peer_id = ath11k_hal_rx_mpduinfo_get_peerid(ab, mpdu_info);
 		if (peer_id)
 			ppdu_info->peer_id = peer_id;
 		break;
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.h b/drivers/net/wireless/ath/ath11k/hal_rx.h
index f6bae07abfd3..47e8208b22e1 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.h
@@ -405,7 +405,7 @@ struct hal_rx_phyrx_rssi_legacy_info {
 #define HAL_RX_MPDU_INFO_INFO0_PEERID_WCN6855	GENMASK(15, 0)
 #define HAL_RX_MPDU_INFO_INFO1_MPDU_LEN		GENMASK(13, 0)
 
-struct hal_rx_mpdu_info {
+struct hal_rx_mpdu_info_ipq8074 {
 	__le32 rsvd0;
 	__le32 info0;
 	__le32 rsvd1[11];
@@ -413,12 +413,28 @@ struct hal_rx_mpdu_info {
 	__le32 rsvd2[9];
 } __packed;
 
+struct hal_rx_mpdu_info_qcn9074 {
+	__le32 rsvd0[10];
+	__le32 info0;
+	__le32 rsvd1[2];
+	__le32 info1;
+	__le32 rsvd2[9];
+} __packed;
+
 struct hal_rx_mpdu_info_wcn6855 {
 	__le32 rsvd0[8];
 	__le32 info0;
 	__le32 rsvd1[14];
 } __packed;
 
+struct hal_rx_mpdu_info {
+	union {
+		struct hal_rx_mpdu_info_ipq8074 ipq8074;
+		struct hal_rx_mpdu_info_qcn9074 qcn9074;
+		struct hal_rx_mpdu_info_wcn6855 wcn6855;
+	} u;
+} __packed;
+
 #define HAL_RX_PPDU_END_DURATION	GENMASK(23, 0)
 struct hal_rx_ppdu_end_duration {
 	__le32 rsvd0[9];
diff --git a/drivers/net/wireless/ath/ath11k/hw.c b/drivers/net/wireless/ath/ath11k/hw.c
index 60ac215e0678..6b4355a68e26 100644
--- a/drivers/net/wireless/ath/ath11k/hw.c
+++ b/drivers/net/wireless/ath/ath11k/hw.c
@@ -835,26 +835,35 @@ static void ath11k_hw_ipq5018_reo_setup(struct ath11k_base *ab)
 			   ring_hash_map);
 }
 
-static u16 ath11k_hw_ipq8074_mpdu_info_get_peerid(u8 *tlv_data)
+static u16
+ath11k_hw_ipq8074_mpdu_info_get_peerid(struct hal_rx_mpdu_info *mpdu_info)
 {
 	u16 peer_id = 0;
-	struct hal_rx_mpdu_info *mpdu_info =
-		(struct hal_rx_mpdu_info *)tlv_data;
 
 	peer_id = FIELD_GET(HAL_RX_MPDU_INFO_INFO0_PEERID,
-			    __le32_to_cpu(mpdu_info->info0));
+			    __le32_to_cpu(mpdu_info->u.ipq8074.info0));
 
 	return peer_id;
 }
 
-static u16 ath11k_hw_wcn6855_mpdu_info_get_peerid(u8 *tlv_data)
+static u16
+ath11k_hw_qcn9074_mpdu_info_get_peerid(struct hal_rx_mpdu_info *mpdu_info)
+{
+	u16 peer_id = 0;
+
+	peer_id = FIELD_GET(HAL_RX_MPDU_INFO_INFO0_PEERID,
+			    __le32_to_cpu(mpdu_info->u.qcn9074.info0));
+
+	return peer_id;
+}
+
+static u16
+ath11k_hw_wcn6855_mpdu_info_get_peerid(struct hal_rx_mpdu_info *mpdu_info)
 {
 	u16 peer_id = 0;
-	struct hal_rx_mpdu_info_wcn6855 *mpdu_info =
-		(struct hal_rx_mpdu_info_wcn6855 *)tlv_data;
 
 	peer_id = FIELD_GET(HAL_RX_MPDU_INFO_INFO0_PEERID_WCN6855,
-			    __le32_to_cpu(mpdu_info->info0));
+			    __le32_to_cpu(mpdu_info->u.wcn6855.info0));
 	return peer_id;
 }
 
@@ -1042,7 +1051,7 @@ const struct ath11k_hw_ops qcn9074_ops = {
 	.rx_desc_get_attention = ath11k_hw_qcn9074_rx_desc_get_attention,
 	.rx_desc_get_msdu_payload = ath11k_hw_qcn9074_rx_desc_get_msdu_payload,
 	.reo_setup = ath11k_hw_ipq8074_reo_setup,
-	.mpdu_info_get_peerid = ath11k_hw_ipq8074_mpdu_info_get_peerid,
+	.mpdu_info_get_peerid = ath11k_hw_qcn9074_mpdu_info_get_peerid,
 	.rx_desc_mac_addr2_valid = ath11k_hw_ipq9074_rx_desc_mac_addr2_valid,
 	.rx_desc_mpdu_start_addr2 = ath11k_hw_ipq9074_rx_desc_mpdu_start_addr2,
 	.get_ring_selector = ath11k_hw_ipq8074_get_tcl_ring_selector,
diff --git a/drivers/net/wireless/ath/ath11k/hw.h b/drivers/net/wireless/ath/ath11k/hw.h
index 9f45d061d826..6a5dd2dbdb3a 100644
--- a/drivers/net/wireless/ath/ath11k/hw.h
+++ b/drivers/net/wireless/ath/ath11k/hw.h
@@ -263,7 +263,7 @@ struct ath11k_hw_ops {
 	struct rx_attention *(*rx_desc_get_attention)(struct hal_rx_desc *desc);
 	u8 *(*rx_desc_get_msdu_payload)(struct hal_rx_desc *desc);
 	void (*reo_setup)(struct ath11k_base *ab);
-	u16 (*mpdu_info_get_peerid)(u8 *tlv_data);
+	u16 (*mpdu_info_get_peerid)(struct hal_rx_mpdu_info *mpdu_info);
 	bool (*rx_desc_mac_addr2_valid)(struct hal_rx_desc *desc);
 	u8* (*rx_desc_mpdu_start_addr2)(struct hal_rx_desc *desc);
 	u32 (*get_ring_selector)(struct sk_buff *skb);
-- 
cgit v1.2.3-70-g09d2


From 447b0398a9cd41ca343dfd43e555af92d6214487 Mon Sep 17 00:00:00 2001
From: Bhagavathi Perumal S <quic_bperumal@quicinc.com>
Date: Fri, 24 Mar 2023 16:57:00 +0200
Subject: wifi: ath11k: Fix invalid management rx frame length issue

The WMI management rx event has multiple arrays of TLVs, however the common
WMI TLV parser won't handle multiple TLV tags of same type.
So the multiple array tags of WMI management rx TLV is parsed incorrectly
and the length calculated becomes wrong when the target sends multiple
array tags.

Add separate TLV parser to handle multiple arrays for WMI management rx
TLV. This fixes invalid length issue when the target sends multiple array
tags.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Bhagavathi Perumal S <quic_bperumal@quicinc.com>
Co-developed-by: Nagarajan Maran <quic_nmaran@quicinc.com>
Signed-off-by: Nagarajan Maran <quic_nmaran@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230320133840.30162-1-quic_nmaran@quicinc.com
---
 drivers/net/wireless/ath/ath11k/wmi.c | 45 +++++++++++++++++++++++++++--------
 1 file changed, 35 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 27f3fceb33c5..f338d4447f11 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -82,6 +82,12 @@ struct wmi_tlv_fw_stats_parse {
 	bool chain_rssi_done;
 };
 
+struct wmi_tlv_mgmt_rx_parse {
+	const struct wmi_mgmt_rx_hdr *fixed;
+	const u8 *frame_buf;
+	bool frame_buf_done;
+};
+
 static const struct wmi_tlv_policy wmi_tlv_policies[] = {
 	[WMI_TAG_ARRAY_BYTE]
 		= { .min_len = 0 },
@@ -5633,28 +5639,49 @@ static int ath11k_pull_vdev_stopped_param_tlv(struct ath11k_base *ab, struct sk_
 	return 0;
 }
 
+static int ath11k_wmi_tlv_mgmt_rx_parse(struct ath11k_base *ab,
+					u16 tag, u16 len,
+					const void *ptr, void *data)
+{
+	struct wmi_tlv_mgmt_rx_parse *parse = data;
+
+	switch (tag) {
+	case WMI_TAG_MGMT_RX_HDR:
+		parse->fixed = ptr;
+		break;
+	case WMI_TAG_ARRAY_BYTE:
+		if (!parse->frame_buf_done) {
+			parse->frame_buf = ptr;
+			parse->frame_buf_done = true;
+		}
+		break;
+	}
+	return 0;
+}
+
 static int ath11k_pull_mgmt_rx_params_tlv(struct ath11k_base *ab,
 					  struct sk_buff *skb,
 					  struct mgmt_rx_event_params *hdr)
 {
-	const void **tb;
+	struct wmi_tlv_mgmt_rx_parse parse = { };
 	const struct wmi_mgmt_rx_hdr *ev;
 	const u8 *frame;
 	int ret;
 
-	tb = ath11k_wmi_tlv_parse_alloc(ab, skb->data, skb->len, GFP_ATOMIC);
-	if (IS_ERR(tb)) {
-		ret = PTR_ERR(tb);
-		ath11k_warn(ab, "failed to parse tlv: %d\n", ret);
+	ret = ath11k_wmi_tlv_iter(ab, skb->data, skb->len,
+				  ath11k_wmi_tlv_mgmt_rx_parse,
+				  &parse);
+	if (ret) {
+		ath11k_warn(ab, "failed to parse mgmt rx tlv %d\n",
+			    ret);
 		return ret;
 	}
 
-	ev = tb[WMI_TAG_MGMT_RX_HDR];
-	frame = tb[WMI_TAG_ARRAY_BYTE];
+	ev = parse.fixed;
+	frame = parse.frame_buf;
 
 	if (!ev || !frame) {
 		ath11k_warn(ab, "failed to fetch mgmt rx hdr");
-		kfree(tb);
 		return -EPROTO;
 	}
 
@@ -5673,7 +5700,6 @@ static int ath11k_pull_mgmt_rx_params_tlv(struct ath11k_base *ab,
 
 	if (skb->len < (frame - skb->data) + hdr->buf_len) {
 		ath11k_warn(ab, "invalid length in mgmt rx hdr ev");
-		kfree(tb);
 		return -EPROTO;
 	}
 
@@ -5685,7 +5711,6 @@ static int ath11k_pull_mgmt_rx_params_tlv(struct ath11k_base *ab,
 
 	ath11k_ce_byte_swap(skb->data, hdr->buf_len);
 
-	kfree(tb);
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 756a7f90878f0866fd2fe167ef37e90b47326b96 Mon Sep 17 00:00:00 2001
From: P Praneesh <quic_ppranees@quicinc.com>
Date: Fri, 24 Mar 2023 16:57:01 +0200
Subject: wifi: ath11k: fix writing to unintended memory region

While initializing spectral, the magic value is getting written to the
invalid memory address leading to random boot-up crash. This occurs
due to the incorrect index increment in ath11k_dbring_fill_magic_value
function. Fix it by replacing the existing logic with memset32 to ensure
there is no invalid memory access.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.4.0.1-01838-QCAHKSWPL_SILICONZ-1

Fixes: d3d358efc553 ("ath11k: add spectral/CFR buffer validation support")
Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230321052900.16895-1-quic_ppranees@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dbring.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dbring.c b/drivers/net/wireless/ath/ath11k/dbring.c
index 2107ec05d14f..5536e8642331 100644
--- a/drivers/net/wireless/ath/ath11k/dbring.c
+++ b/drivers/net/wireless/ath/ath11k/dbring.c
@@ -26,13 +26,13 @@ int ath11k_dbring_validate_buffer(struct ath11k *ar, void *buffer, u32 size)
 static void ath11k_dbring_fill_magic_value(struct ath11k *ar,
 					   void *buffer, u32 size)
 {
-	u32 *temp;
-	int idx;
-
-	size = size >> 2;
+	/* memset32 function fills buffer payload with the ATH11K_DB_MAGIC_VALUE
+	 * and the variable size is expected to be the number of u32 values
+	 * to be stored, not the number of bytes.
+	 */
+	size = size / sizeof(u32);
 
-	for (idx = 0, temp = buffer; idx < size; idx++, temp++)
-		*temp++ = ATH11K_DB_MAGIC_VALUE;
+	memset32(buffer, ATH11K_DB_MAGIC_VALUE, size);
 }
 
 static int ath11k_dbring_bufs_replenish(struct ath11k *ar,
-- 
cgit v1.2.3-70-g09d2


From e89a51aedf380bc60219dc9afa96c36507060fb3 Mon Sep 17 00:00:00 2001
From: Manikanta Pubbisetty <quic_mpubbise@quicinc.com>
Date: Wed, 15 Mar 2023 21:48:17 +0530
Subject: wifi: ath11k: Send 11d scan start before WMI_START_SCAN_CMDID

Firmwares advertising the support of triggering 11d algorithm on the
scan results of a regular scan expects driver to send
WMI_11D_SCAN_START_CMDID before sending WMI_START_SCAN_CMDID.
Triggering 11d algorithm on the scan results of a normal scan helps
in completely avoiding a separate 11d scan for determining regdomain.
This indirectly helps in speeding up connections on station
interfaces on the chipsets supporting 11D scan.

To enable this feature, send WMI_11D_SCAN_START_CMDID just before
sending WMI_START_SCAN_CMDID if the firmware advertises
WMI_TLV_SERVICE_SUPPORT_11D_FOR_HOST_SCAN service flag.

WCN6750 & WCN6855 supports this feature.

Tested-on: WCN6750 hw1.0 AHB WLAN.MSL.1.0.1-01160-QCAMSLSWPLZ-1
Tested-on: WCN6855 hw2.1 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.23

Signed-off-by: Manikanta Pubbisetty <quic_mpubbise@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230315161817.29627-1-quic_mpubbise@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 12 ++++++++++++
 drivers/net/wireless/ath/ath11k/wmi.h |  1 +
 2 files changed, 13 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 7226d88c0c48..c0dc3e387aad 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -3755,6 +3755,18 @@ static int ath11k_mac_op_hw_scan(struct ieee80211_hw *hw,
 	int i;
 	u32 scan_timeout;
 
+	/* Firmwares advertising the support of triggering 11D algorithm
+	 * on the scan results of a regular scan expects driver to send
+	 * WMI_11D_SCAN_START_CMDID before sending WMI_START_SCAN_CMDID.
+	 * With this feature, separate 11D scan can be avoided since
+	 * regdomain can be determined with the scan results of the
+	 * regular scan.
+	 */
+	if (ar->state_11d == ATH11K_11D_PREPARING &&
+	    test_bit(WMI_TLV_SERVICE_SUPPORT_11D_FOR_HOST_SCAN,
+		     ar->ab->wmi_ab.svc_map))
+		ath11k_mac_11d_scan_start(ar, arvif->vdev_id);
+
 	mutex_lock(&ar->conf_mutex);
 
 	spin_lock_bh(&ar->data_lock);
diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h
index 86b22f2cec12..92fddb77669c 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.h
+++ b/drivers/net/wireless/ath/ath11k/wmi.h
@@ -2103,6 +2103,7 @@ enum wmi_tlv_service {
 	WMI_TLV_SERVICE_SCAN_CONFIG_PER_CHANNEL = 265,
 	WMI_TLV_SERVICE_REG_CC_EXT_EVENT_SUPPORT = 281,
 	WMI_TLV_SERVICE_BIOS_SAR_SUPPORT = 326,
+	WMI_TLV_SERVICE_SUPPORT_11D_FOR_HOST_SCAN = 357,
 
 	/* The third 128 bits */
 	WMI_MAX_EXT2_SERVICE = 384
-- 
cgit v1.2.3-70-g09d2


From f812e2a9f85d6bea78957ccb5197e4491316848b Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:26:09 +0800
Subject: wifi: ath11k: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230323112613.7550-1-cai.huoqing@linux.dev
---
 drivers/net/wireless/ath/ath11k/pci.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/pci.c b/drivers/net/wireless/ath/ath11k/pci.c
index 0aeef2948ff5..522d01e5e11c 100644
--- a/drivers/net/wireless/ath/ath11k/pci.c
+++ b/drivers/net/wireless/ath/ath11k/pci.c
@@ -540,7 +540,7 @@ static int ath11k_pci_claim(struct ath11k_pci *ab_pci, struct pci_dev *pdev)
 	if (!ab->mem) {
 		ath11k_err(ab, "failed to map pci bar %d\n", ATH11K_PCI_BAR_NUM);
 		ret = -EIO;
-		goto clear_master;
+		goto release_region;
 	}
 
 	ab->mem_ce = ab->mem;
@@ -548,8 +548,6 @@ static int ath11k_pci_claim(struct ath11k_pci *ab_pci, struct pci_dev *pdev)
 	ath11k_dbg(ab, ATH11K_DBG_BOOT, "boot pci_mem 0x%pK\n", ab->mem);
 	return 0;
 
-clear_master:
-	pci_clear_master(pdev);
 release_region:
 	pci_release_region(pdev, ATH11K_PCI_BAR_NUM);
 disable_device:
@@ -565,7 +563,6 @@ static void ath11k_pci_free_region(struct ath11k_pci *ab_pci)
 
 	pci_iounmap(pci_dev, ab->mem);
 	ab->mem = NULL;
-	pci_clear_master(pci_dev);
 	pci_release_region(pci_dev, ATH11K_PCI_BAR_NUM);
 	if (pci_is_enabled(pci_dev))
 		pci_disable_device(pci_dev);
-- 
cgit v1.2.3-70-g09d2


From 76008fc13b09a89e4251e6af5062043427babaa2 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:26:10 +0800
Subject: wifi: ath10k: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230323112613.7550-2-cai.huoqing@linux.dev
---
 drivers/net/wireless/ath/ath10k/pci.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 728d607289c3..a7f44f6335fb 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -3406,15 +3406,12 @@ static int ath10k_pci_claim(struct ath10k *ar)
 	if (!ar_pci->mem) {
 		ath10k_err(ar, "failed to iomap BAR%d\n", BAR_NUM);
 		ret = -EIO;
-		goto err_master;
+		goto err_region;
 	}
 
 	ath10k_dbg(ar, ATH10K_DBG_BOOT, "boot pci_mem 0x%pK\n", ar_pci->mem);
 	return 0;
 
-err_master:
-	pci_clear_master(pdev);
-
 err_region:
 	pci_release_region(pdev, BAR_NUM);
 
@@ -3431,7 +3428,6 @@ static void ath10k_pci_release(struct ath10k *ar)
 
 	pci_iounmap(pdev, ar_pci->mem);
 	pci_release_region(pdev, BAR_NUM);
-	pci_clear_master(pdev);
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-70-g09d2


From b9235aef84929e5330cb87125a6baf1cf7988983 Mon Sep 17 00:00:00 2001
From: Cai Huoqing <cai.huoqing@linux.dev>
Date: Thu, 23 Mar 2023 19:26:11 +0800
Subject: wifi: ath12k: Remove redundant pci_clear_master

Remove pci_clear_master to simplify the code,
the bus-mastering is also cleared in do_pci_disable_device,
like this:
./drivers/pci/pci.c:2197
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}.
And dev->is_busmaster is set to 0 in pci_disable_device.

Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230323112613.7550-3-cai.huoqing@linux.dev
---
 drivers/net/wireless/ath/ath12k/pci.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c
index 00b0080dbac3..9f174daf324c 100644
--- a/drivers/net/wireless/ath/ath12k/pci.c
+++ b/drivers/net/wireless/ath/ath12k/pci.c
@@ -755,14 +755,12 @@ static int ath12k_pci_claim(struct ath12k_pci *ab_pci, struct pci_dev *pdev)
 	if (!ab->mem) {
 		ath12k_err(ab, "failed to map pci bar %d\n", ATH12K_PCI_BAR_NUM);
 		ret = -EIO;
-		goto clear_master;
+		goto release_region;
 	}
 
 	ath12k_dbg(ab, ATH12K_DBG_BOOT, "boot pci_mem 0x%pK\n", ab->mem);
 	return 0;
 
-clear_master:
-	pci_clear_master(pdev);
 release_region:
 	pci_release_region(pdev, ATH12K_PCI_BAR_NUM);
 disable_device:
@@ -778,7 +776,6 @@ static void ath12k_pci_free_region(struct ath12k_pci *ab_pci)
 
 	pci_iounmap(pci_dev, ab->mem);
 	ab->mem = NULL;
-	pci_clear_master(pci_dev);
 	pci_release_region(pci_dev, ATH12K_PCI_BAR_NUM);
 	if (pci_is_enabled(pci_dev))
 		pci_disable_device(pci_dev);
-- 
cgit v1.2.3-70-g09d2


From 6a92566088b1a37c1cf2c4b6b5fb733dc5bdc6a6 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:21 +0200
Subject: wifi: rtw88: Clear RTW_FLAG_POWERON early in rtw_mac_power_switch()

The SDIO HCI implementation needs to know when the MAC is powered on.
This is needed because 32-bit register access has to be split into 4x
8-bit register access when the MAC is not fully powered on or while
powering off. When the MAC is powered on 32-bit register access can be
used to reduce the number of transfers but splitting into 4x 8-bit
register access still works in that case.

During the power on sequence is how RTW_FLAG_POWERON is only set when
the power on sequence has completed successfully. During power off
however RTW_FLAG_POWERON is set. This means that the upcoming SDIO HCI
implementation does not know that it has to use 4x 8-bit register
accessors. Clear the RTW_FLAG_POWERON flag early when powering off the
MAC so the whole power off sequence is processed with RTW_FLAG_POWERON
unset. This will make it possible to use the RTW_FLAG_POWERON flag in
the upcoming SDIO HCI implementation.

Note that a failure in rtw_pwr_seq_parser() while applying
chip->pwr_off_seq can theoretically result in the RTW_FLAG_POWERON
flag being cleared while the chip is still powered on. However,
depending on when the failure occurs in the power off sequence the
chip may be on or off. Even the original approach of clearing
RTW_FLAG_POWERON only when the power off sequence has been applied
successfully could end up in some corner case where the chip is
powered off but RTW_FLAG_POWERON was not cleared.

Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-2-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index f3a566cf979b..cfdfc8a2c836 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -273,6 +273,9 @@ static int rtw_mac_power_switch(struct rtw_dev *rtwdev, bool pwr_on)
 	if (pwr_on == cur_pwr)
 		return -EALREADY;
 
+	if (!pwr_on)
+		clear_bit(RTW_FLAG_POWERON, rtwdev->flags);
+
 	pwr_seq = pwr_on ? chip->pwr_on_seq : chip->pwr_off_seq;
 	ret = rtw_pwr_seq_parser(rtwdev, pwr_seq);
 	if (ret)
@@ -280,8 +283,6 @@ static int rtw_mac_power_switch(struct rtw_dev *rtwdev, bool pwr_on)
 
 	if (pwr_on)
 		set_bit(RTW_FLAG_POWERON, rtwdev->flags);
-	else
-		clear_bit(RTW_FLAG_POWERON, rtwdev->flags);
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 65371a3f14e73979958aea0db1e3bb456a296149 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:22 +0200
Subject: wifi: rtw88: sdio: Add HCI implementation for SDIO based chipsets

Add a sub-driver for SDIO based chipsets which implements the following
functionality:
- register accessors for 8, 16 and 32 bits for all states of the card
  (including usage of 4x 8 bit access for one 32 bit buffer if the card
  is not fully powered on yet - or if it's fully powered on then 1x 32
  bit access is used)
- checking whether there's space in the TX FIFO queue to transmit data
- transfers from the host to the device for actual network traffic,
  reserved pages (for firmware download) and H2C (host-to-card)
  transfers
- receiving data from the device
- deep power saving state

The transmit path is optimized so DMA-capable SDIO host controllers can
directly use the buffers provided because the buffer's physical
addresses are 8 byte aligned.

The receive path is prepared to support RX aggregation where the
chipset combines multiple MAC frames into one bigger buffer to reduce
SDIO transfer overhead.

Co-developed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-3-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/Kconfig  |    3 +
 drivers/net/wireless/realtek/rtw88/Makefile |    3 +
 drivers/net/wireless/realtek/rtw88/debug.h  |    1 +
 drivers/net/wireless/realtek/rtw88/mac.c    |    1 +
 drivers/net/wireless/realtek/rtw88/mac.h    |    1 -
 drivers/net/wireless/realtek/rtw88/reg.h    |   12 +
 drivers/net/wireless/realtek/rtw88/sdio.c   | 1394 +++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw88/sdio.h   |  178 ++++
 8 files changed, 1592 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/wireless/realtek/rtw88/sdio.c
 create mode 100644 drivers/net/wireless/realtek/rtw88/sdio.h

diff --git a/drivers/net/wireless/realtek/rtw88/Kconfig b/drivers/net/wireless/realtek/rtw88/Kconfig
index 651ab56d9c6b..cdf9cb478ee2 100644
--- a/drivers/net/wireless/realtek/rtw88/Kconfig
+++ b/drivers/net/wireless/realtek/rtw88/Kconfig
@@ -16,6 +16,9 @@ config RTW88_CORE
 config RTW88_PCI
 	tristate
 
+config RTW88_SDIO
+	tristate
+
 config RTW88_USB
 	tristate
 
diff --git a/drivers/net/wireless/realtek/rtw88/Makefile b/drivers/net/wireless/realtek/rtw88/Makefile
index fe7293ee87b4..892cad60ba31 100644
--- a/drivers/net/wireless/realtek/rtw88/Makefile
+++ b/drivers/net/wireless/realtek/rtw88/Makefile
@@ -59,5 +59,8 @@ rtw88_8821cu-objs		:= rtw8821cu.o
 obj-$(CONFIG_RTW88_PCI)		+= rtw88_pci.o
 rtw88_pci-objs			:= pci.o
 
+obj-$(CONFIG_RTW88_SDIO)	+= rtw88_sdio.o
+rtw88_sdio-objs			:= sdio.o
+
 obj-$(CONFIG_RTW88_USB)		+= rtw88_usb.o
 rtw88_usb-objs			:= usb.o
diff --git a/drivers/net/wireless/realtek/rtw88/debug.h b/drivers/net/wireless/realtek/rtw88/debug.h
index 066792dd96af..a9149c6c2b48 100644
--- a/drivers/net/wireless/realtek/rtw88/debug.h
+++ b/drivers/net/wireless/realtek/rtw88/debug.h
@@ -24,6 +24,7 @@ enum rtw_debug_mask {
 	RTW_DBG_ADAPTIVITY	= 0x00008000,
 	RTW_DBG_HW_SCAN		= 0x00010000,
 	RTW_DBG_STATE		= 0x00020000,
+	RTW_DBG_SDIO		= 0x00040000,
 
 	RTW_DBG_ALL		= 0xffffffff
 };
diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index cfdfc8a2c836..2fcba43a6f72 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -7,6 +7,7 @@
 #include "reg.h"
 #include "fw.h"
 #include "debug.h"
+#include "sdio.h"
 
 void rtw_set_channel_mac(struct rtw_dev *rtwdev, u8 channel, u8 bw,
 			 u8 primary_ch_idx)
diff --git a/drivers/net/wireless/realtek/rtw88/mac.h b/drivers/net/wireless/realtek/rtw88/mac.h
index 3172aa5ac4de..58c3dccc14bb 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.h
+++ b/drivers/net/wireless/realtek/rtw88/mac.h
@@ -7,7 +7,6 @@
 
 #define RTW_HW_PORT_NUM		5
 #define cut_version_to_mask(cut) (0x1 << ((cut) + 1))
-#define SDIO_LOCAL_OFFSET	0x10250000
 #define DDMA_POLLING_COUNT	1000
 #define C2H_PKT_BUF		256
 #define REPORT_BUF		128
diff --git a/drivers/net/wireless/realtek/rtw88/reg.h b/drivers/net/wireless/realtek/rtw88/reg.h
index 8852b24d6c2a..2a2ae2081f34 100644
--- a/drivers/net/wireless/realtek/rtw88/reg.h
+++ b/drivers/net/wireless/realtek/rtw88/reg.h
@@ -87,6 +87,7 @@
 #define BIT_LTE_MUX_CTRL_PATH	BIT(26)
 #define REG_HCI_OPT_CTRL	0x0074
 #define BIT_USB_SUS_DIS		BIT(8)
+#define BIT_SDIO_PAD_E5		BIT(18)
 
 #define REG_AFE_CTRL_4		0x0078
 #define BIT_CK320M_AFE_EN	BIT(4)
@@ -185,6 +186,9 @@
 	(((x) & BIT_MASK_TXDMA_VIQ_MAP) << BIT_SHIFT_TXDMA_VIQ_MAP)
 #define REG_TXDMA_PQ_MAP	0x010C
 #define BIT_RXDMA_ARBBW_EN	BIT(0)
+#define BIT_RXSHFT_EN		BIT(1)
+#define BIT_RXDMA_AGG_EN	BIT(2)
+#define BIT_TXDMA_BW_EN		BIT(3)
 #define BIT_SHIFT_TXDMA_BEQ_MAP	8
 #define BIT_MASK_TXDMA_BEQ_MAP	0x3
 #define BIT_TXDMA_BEQ_MAP(x)                                                   \
@@ -283,10 +287,18 @@
 #define REG_H2C_TAIL		0x0248
 #define REG_H2C_READ_ADDR	0x024C
 #define REG_H2C_INFO		0x0254
+#define REG_RXDMA_AGG_PG_TH	0x0280
+#define BIT_RXDMA_AGG_PG_TH	GENMASK(7, 0)
+#define BIT_DMA_AGG_TO_V1	GENMASK(15, 8)
+#define BIT_EN_PRE_CALC		BIT(29)
 #define REG_RXPKT_NUM		0x0284
 #define BIT_RXDMA_REQ		BIT(19)
 #define BIT_RW_RELEASE		BIT(18)
 #define BIT_RXDMA_IDLE		BIT(17)
+#define REG_RXDMA_STATUS	0x0288
+#define REG_RXDMA_DPR		0x028C
+#define REG_RXDMA_MODE		0x0290
+#define BIT_DMA_MODE		BIT(1)
 #define REG_RXPKTNUM		0x02B0
 
 #define REG_INT_MIG		0x0304
diff --git a/drivers/net/wireless/realtek/rtw88/sdio.c b/drivers/net/wireless/realtek/rtw88/sdio.c
new file mode 100644
index 000000000000..af0459a79899
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/sdio.c
@@ -0,0 +1,1394 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright (C) 2021 Martin Blumenstingl <martin.blumenstingl@googlemail.com>
+ * Copyright (C) 2021 Jernej Skrabec <jernej.skrabec@gmail.com>
+ *
+ * Based on rtw88/pci.c:
+ *   Copyright(c) 2018-2019  Realtek Corporation
+ */
+
+#include <linux/module.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/sdio_func.h>
+#include "main.h"
+#include "debug.h"
+#include "fw.h"
+#include "ps.h"
+#include "reg.h"
+#include "rx.h"
+#include "sdio.h"
+#include "tx.h"
+
+#define RTW_SDIO_INDIRECT_RW_RETRIES			50
+
+static bool rtw_sdio_is_bus_addr(u32 addr)
+{
+	return !!(addr & RTW_SDIO_BUS_MSK);
+}
+
+static bool rtw_sdio_bus_claim_needed(struct rtw_sdio *rtwsdio)
+{
+	return !rtwsdio->irq_thread ||
+	       rtwsdio->irq_thread != current;
+}
+
+static u32 rtw_sdio_to_bus_offset(struct rtw_dev *rtwdev, u32 addr)
+{
+	switch (addr & RTW_SDIO_BUS_MSK) {
+	case WLAN_IOREG_OFFSET:
+		addr &= WLAN_IOREG_REG_MSK;
+		addr |= FIELD_PREP(REG_SDIO_CMD_ADDR_MSK,
+				   REG_SDIO_CMD_ADDR_MAC_REG);
+		break;
+	case SDIO_LOCAL_OFFSET:
+		addr &= SDIO_LOCAL_REG_MSK;
+		addr |= FIELD_PREP(REG_SDIO_CMD_ADDR_MSK,
+				   REG_SDIO_CMD_ADDR_SDIO_REG);
+		break;
+	default:
+		rtw_warn(rtwdev, "Cannot convert addr 0x%08x to bus offset",
+			 addr);
+	}
+
+	return addr;
+}
+
+static bool rtw_sdio_use_memcpy_io(struct rtw_dev *rtwdev, u32 addr,
+				   u8 alignment)
+{
+	return IS_ALIGNED(addr, alignment) &&
+	       test_bit(RTW_FLAG_POWERON, rtwdev->flags);
+}
+
+static void rtw_sdio_writel(struct rtw_dev *rtwdev, u32 val, u32 addr,
+			    int *err_ret)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	u8 buf[4];
+	int i;
+
+	if (rtw_sdio_use_memcpy_io(rtwdev, addr, 4)) {
+		sdio_writel(rtwsdio->sdio_func, val, addr, err_ret);
+		return;
+	}
+
+	*(__le32 *)buf = cpu_to_le32(val);
+
+	for (i = 0; i < 4; i++) {
+		sdio_writeb(rtwsdio->sdio_func, buf[i], addr + i, err_ret);
+		if (*err_ret)
+			return;
+	}
+}
+
+static void rtw_sdio_writew(struct rtw_dev *rtwdev, u16 val, u32 addr,
+			    int *err_ret)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	u8 buf[2];
+	int i;
+
+	if (rtw_sdio_use_memcpy_io(rtwdev, addr, 2)) {
+		sdio_writew(rtwsdio->sdio_func, val, addr, err_ret);
+		return;
+	}
+
+	*(__le16 *)buf = cpu_to_le16(val);
+
+	for (i = 0; i < 2; i++) {
+		sdio_writeb(rtwsdio->sdio_func, buf[i], addr + i, err_ret);
+		if (*err_ret)
+			return;
+	}
+}
+
+static u32 rtw_sdio_readl(struct rtw_dev *rtwdev, u32 addr, int *err_ret)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	u8 buf[4];
+	int i;
+
+	if (rtw_sdio_use_memcpy_io(rtwdev, addr, 4))
+		return sdio_readl(rtwsdio->sdio_func, addr, err_ret);
+
+	for (i = 0; i < 4; i++) {
+		buf[i] = sdio_readb(rtwsdio->sdio_func, addr + i, err_ret);
+		if (*err_ret)
+			return 0;
+	}
+
+	return le32_to_cpu(*(__le32 *)buf);
+}
+
+static u16 rtw_sdio_readw(struct rtw_dev *rtwdev, u32 addr, int *err_ret)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	u8 buf[2];
+	int i;
+
+	if (rtw_sdio_use_memcpy_io(rtwdev, addr, 2))
+		return sdio_readw(rtwsdio->sdio_func, addr, err_ret);
+
+	for (i = 0; i < 2; i++) {
+		buf[i] = sdio_readb(rtwsdio->sdio_func, addr + i, err_ret);
+		if (*err_ret)
+			return 0;
+	}
+
+	return le16_to_cpu(*(__le16 *)buf);
+}
+
+static u32 rtw_sdio_to_io_address(struct rtw_dev *rtwdev, u32 addr,
+				  bool direct)
+{
+	if (!direct)
+		return addr;
+
+	if (!rtw_sdio_is_bus_addr(addr))
+		addr |= WLAN_IOREG_OFFSET;
+
+	return rtw_sdio_to_bus_offset(rtwdev, addr);
+}
+
+static bool rtw_sdio_use_direct_io(struct rtw_dev *rtwdev, u32 addr)
+{
+	return !rtw_sdio_is_sdio30_supported(rtwdev) ||
+		rtw_sdio_is_bus_addr(addr);
+}
+
+static int rtw_sdio_indirect_reg_cfg(struct rtw_dev *rtwdev, u32 addr, u32 cfg)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	unsigned int retry;
+	u32 reg_cfg;
+	int ret;
+	u8 tmp;
+
+	reg_cfg = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_CFG);
+
+	rtw_sdio_writel(rtwdev, addr | cfg | BIT_SDIO_INDIRECT_REG_CFG_UNK20,
+			reg_cfg, &ret);
+	if (ret)
+		return ret;
+
+	for (retry = 0; retry < RTW_SDIO_INDIRECT_RW_RETRIES; retry++) {
+		tmp = sdio_readb(rtwsdio->sdio_func, reg_cfg + 2, &ret);
+		if (!ret && (tmp & BIT(4)))
+			return 0;
+	}
+
+	return -ETIMEDOUT;
+}
+
+static u8 rtw_sdio_indirect_read8(struct rtw_dev *rtwdev, u32 addr,
+				  int *err_ret)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	u32 reg_data;
+
+	*err_ret = rtw_sdio_indirect_reg_cfg(rtwdev, addr,
+					     BIT_SDIO_INDIRECT_REG_CFG_READ);
+	if (*err_ret)
+		return 0;
+
+	reg_data = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_DATA);
+	return sdio_readb(rtwsdio->sdio_func, reg_data, err_ret);
+}
+
+static int rtw_sdio_indirect_read_bytes(struct rtw_dev *rtwdev, u32 addr,
+					u8 *buf, int count)
+{
+	int i, ret = 0;
+
+	for (i = 0; i < count; i++) {
+		buf[i] = rtw_sdio_indirect_read8(rtwdev, addr + i, &ret);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+static u16 rtw_sdio_indirect_read16(struct rtw_dev *rtwdev, u32 addr,
+				    int *err_ret)
+{
+	u32 reg_data;
+	u8 buf[2];
+
+	if (!IS_ALIGNED(addr, 2)) {
+		*err_ret = rtw_sdio_indirect_read_bytes(rtwdev, addr, buf, 2);
+		if (*err_ret)
+			return 0;
+
+		return le16_to_cpu(*(__le16 *)buf);
+	}
+
+	*err_ret = rtw_sdio_indirect_reg_cfg(rtwdev, addr,
+					     BIT_SDIO_INDIRECT_REG_CFG_READ);
+	if (*err_ret)
+		return 0;
+
+	reg_data = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_DATA);
+	return rtw_sdio_readw(rtwdev, reg_data, err_ret);
+}
+
+static u32 rtw_sdio_indirect_read32(struct rtw_dev *rtwdev, u32 addr,
+				    int *err_ret)
+{
+	u32 reg_data;
+	u8 buf[4];
+
+	if (!IS_ALIGNED(addr, 4)) {
+		*err_ret = rtw_sdio_indirect_read_bytes(rtwdev, addr, buf, 4);
+		if (*err_ret)
+			return 0;
+
+		return le32_to_cpu(*(__le32 *)buf);
+	}
+
+	*err_ret = rtw_sdio_indirect_reg_cfg(rtwdev, addr,
+					     BIT_SDIO_INDIRECT_REG_CFG_READ);
+	if (*err_ret)
+		return 0;
+
+	reg_data = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_DATA);
+	return rtw_sdio_readl(rtwdev, reg_data, err_ret);
+}
+
+static u8 rtw_sdio_read8(struct rtw_dev *rtwdev, u32 addr)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool direct, bus_claim;
+	int ret;
+	u8 val;
+
+	direct = rtw_sdio_use_direct_io(rtwdev, addr);
+	addr = rtw_sdio_to_io_address(rtwdev, addr, direct);
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	if (direct)
+		val = sdio_readb(rtwsdio->sdio_func, addr, &ret);
+	else
+		val = rtw_sdio_indirect_read8(rtwdev, addr, &ret);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev, "sdio read8 failed (0x%x): %d", addr, ret);
+
+	return val;
+}
+
+static u16 rtw_sdio_read16(struct rtw_dev *rtwdev, u32 addr)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool direct, bus_claim;
+	int ret;
+	u16 val;
+
+	direct = rtw_sdio_use_direct_io(rtwdev, addr);
+	addr = rtw_sdio_to_io_address(rtwdev, addr, direct);
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	if (direct)
+		val = rtw_sdio_readw(rtwdev, addr, &ret);
+	else
+		val = rtw_sdio_indirect_read16(rtwdev, addr, &ret);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev, "sdio read16 failed (0x%x): %d", addr, ret);
+
+	return val;
+}
+
+static u32 rtw_sdio_read32(struct rtw_dev *rtwdev, u32 addr)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool direct, bus_claim;
+	u32 val;
+	int ret;
+
+	direct = rtw_sdio_use_direct_io(rtwdev, addr);
+	addr = rtw_sdio_to_io_address(rtwdev, addr, direct);
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	if (direct)
+		val = rtw_sdio_readl(rtwdev, addr, &ret);
+	else
+		val = rtw_sdio_indirect_read32(rtwdev, addr, &ret);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev, "sdio read32 failed (0x%x): %d", addr, ret);
+
+	return val;
+}
+
+static void rtw_sdio_indirect_write8(struct rtw_dev *rtwdev, u8 val, u32 addr,
+				     int *err_ret)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	u32 reg_data;
+
+	reg_data = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_DATA);
+	sdio_writeb(rtwsdio->sdio_func, val, reg_data, err_ret);
+	if (*err_ret)
+		return;
+
+	*err_ret = rtw_sdio_indirect_reg_cfg(rtwdev, addr,
+					     BIT_SDIO_INDIRECT_REG_CFG_WRITE);
+}
+
+static void rtw_sdio_indirect_write16(struct rtw_dev *rtwdev, u16 val, u32 addr,
+				      int *err_ret)
+{
+	u32 reg_data;
+
+	if (!IS_ALIGNED(addr, 2)) {
+		addr = rtw_sdio_to_io_address(rtwdev, addr, true);
+		rtw_sdio_writew(rtwdev, val, addr, err_ret);
+		return;
+	}
+
+	reg_data = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_DATA);
+	rtw_sdio_writew(rtwdev, val, reg_data, err_ret);
+	if (*err_ret)
+		return;
+
+	*err_ret = rtw_sdio_indirect_reg_cfg(rtwdev, addr,
+					     BIT_SDIO_INDIRECT_REG_CFG_WRITE |
+					     BIT_SDIO_INDIRECT_REG_CFG_WORD);
+}
+
+static void rtw_sdio_indirect_write32(struct rtw_dev *rtwdev, u32 val,
+				      u32 addr, int *err_ret)
+{
+	u32 reg_data;
+
+	if (!IS_ALIGNED(addr, 4)) {
+		addr = rtw_sdio_to_io_address(rtwdev, addr, true);
+		rtw_sdio_writel(rtwdev, val, addr, err_ret);
+		return;
+	}
+
+	reg_data = rtw_sdio_to_bus_offset(rtwdev, REG_SDIO_INDIRECT_REG_DATA);
+	rtw_sdio_writel(rtwdev, val, reg_data, err_ret);
+
+	*err_ret = rtw_sdio_indirect_reg_cfg(rtwdev, addr,
+					     BIT_SDIO_INDIRECT_REG_CFG_WRITE |
+					     BIT_SDIO_INDIRECT_REG_CFG_DWORD);
+}
+
+static void rtw_sdio_write8(struct rtw_dev *rtwdev, u32 addr, u8 val)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool direct, bus_claim;
+	int ret;
+
+	direct = rtw_sdio_use_direct_io(rtwdev, addr);
+	addr = rtw_sdio_to_io_address(rtwdev, addr, direct);
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	if (direct)
+		sdio_writeb(rtwsdio->sdio_func, val, addr, &ret);
+	else
+		rtw_sdio_indirect_write8(rtwdev, val, addr, &ret);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev, "sdio write8 failed (0x%x): %d", addr, ret);
+}
+
+static void rtw_sdio_write16(struct rtw_dev *rtwdev, u32 addr, u16 val)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool direct, bus_claim;
+	int ret;
+
+	direct = rtw_sdio_use_direct_io(rtwdev, addr);
+	addr = rtw_sdio_to_io_address(rtwdev, addr, direct);
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	if (direct)
+		rtw_sdio_writew(rtwdev, val, addr, &ret);
+	else
+		rtw_sdio_indirect_write16(rtwdev, val, addr, &ret);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev, "sdio write16 failed (0x%x): %d", addr, ret);
+}
+
+static void rtw_sdio_write32(struct rtw_dev *rtwdev, u32 addr, u32 val)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool direct, bus_claim;
+	int ret;
+
+	direct = rtw_sdio_use_direct_io(rtwdev, addr);
+	addr = rtw_sdio_to_io_address(rtwdev, addr, direct);
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	if (direct)
+		rtw_sdio_writel(rtwdev, val, addr, &ret);
+	else
+		rtw_sdio_indirect_write32(rtwdev, val, addr, &ret);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev, "sdio write32 failed (0x%x): %d", addr, ret);
+}
+
+static u32 rtw_sdio_get_tx_addr(struct rtw_dev *rtwdev, size_t size,
+				enum rtw_tx_queue_type queue)
+{
+	u32 txaddr;
+
+	switch (queue) {
+	case RTW_TX_QUEUE_BCN:
+	case RTW_TX_QUEUE_H2C:
+	case RTW_TX_QUEUE_HI0:
+		txaddr = FIELD_PREP(REG_SDIO_CMD_ADDR_MSK,
+				    REG_SDIO_CMD_ADDR_TXFF_HIGH);
+		break;
+	case RTW_TX_QUEUE_VI:
+	case RTW_TX_QUEUE_VO:
+		txaddr = FIELD_PREP(REG_SDIO_CMD_ADDR_MSK,
+				    REG_SDIO_CMD_ADDR_TXFF_NORMAL);
+		break;
+	case RTW_TX_QUEUE_BE:
+	case RTW_TX_QUEUE_BK:
+		txaddr = FIELD_PREP(REG_SDIO_CMD_ADDR_MSK,
+				    REG_SDIO_CMD_ADDR_TXFF_LOW);
+		break;
+	case RTW_TX_QUEUE_MGMT:
+		txaddr = FIELD_PREP(REG_SDIO_CMD_ADDR_MSK,
+				    REG_SDIO_CMD_ADDR_TXFF_EXTRA);
+		break;
+	default:
+		rtw_warn(rtwdev, "Unsupported queue for TX addr: 0x%02x\n",
+			 queue);
+		return 0;
+	}
+
+	txaddr += DIV_ROUND_UP(size, 4);
+
+	return txaddr;
+};
+
+static int rtw_sdio_read_port(struct rtw_dev *rtwdev, u8 *buf, size_t count)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+	u32 rxaddr = rtwsdio->rx_addr++;
+	int ret;
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	ret = sdio_memcpy_fromio(rtwsdio->sdio_func, buf,
+				 RTW_SDIO_ADDR_RX_RX0FF_GEN(rxaddr), count);
+	if (ret)
+		rtw_warn(rtwdev,
+			 "Failed to read %zu byte(s) from SDIO port 0x%08x",
+			 count, rxaddr);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	return ret;
+}
+
+static int rtw_sdio_check_free_txpg(struct rtw_dev *rtwdev, u8 queue,
+				    size_t count)
+{
+	unsigned int pages_free, pages_needed;
+
+	if (rtw_chip_wcpu_11n(rtwdev)) {
+		u32 free_txpg;
+
+		free_txpg = rtw_sdio_read32(rtwdev, REG_SDIO_FREE_TXPG);
+
+		switch (queue) {
+		case RTW_TX_QUEUE_BCN:
+		case RTW_TX_QUEUE_H2C:
+		case RTW_TX_QUEUE_HI0:
+		case RTW_TX_QUEUE_MGMT:
+			/* high */
+			pages_free = free_txpg & 0xff;
+			break;
+		case RTW_TX_QUEUE_VI:
+		case RTW_TX_QUEUE_VO:
+			/* normal */
+			pages_free = (free_txpg >> 8) & 0xff;
+			break;
+		case RTW_TX_QUEUE_BE:
+		case RTW_TX_QUEUE_BK:
+			/* low */
+			pages_free = (free_txpg >> 16) & 0xff;
+			break;
+		default:
+			rtw_warn(rtwdev, "Unknown mapping for queue %u\n", queue);
+			return -EINVAL;
+		}
+
+		/* add the pages from the public queue */
+		pages_free += (free_txpg >> 24) & 0xff;
+	} else {
+		u32 free_txpg[3];
+
+		free_txpg[0] = rtw_sdio_read32(rtwdev, REG_SDIO_FREE_TXPG);
+		free_txpg[1] = rtw_sdio_read32(rtwdev, REG_SDIO_FREE_TXPG + 4);
+		free_txpg[2] = rtw_sdio_read32(rtwdev, REG_SDIO_FREE_TXPG + 8);
+
+		switch (queue) {
+		case RTW_TX_QUEUE_BCN:
+		case RTW_TX_QUEUE_H2C:
+		case RTW_TX_QUEUE_HI0:
+			/* high */
+			pages_free = free_txpg[0] & 0xfff;
+			break;
+		case RTW_TX_QUEUE_VI:
+		case RTW_TX_QUEUE_VO:
+			/* normal */
+			pages_free = (free_txpg[0] >> 16) & 0xfff;
+			break;
+		case RTW_TX_QUEUE_BE:
+		case RTW_TX_QUEUE_BK:
+			/* low */
+			pages_free = free_txpg[1] & 0xfff;
+			break;
+		case RTW_TX_QUEUE_MGMT:
+			/* extra */
+			pages_free = free_txpg[2] & 0xfff;
+			break;
+		default:
+			rtw_warn(rtwdev, "Unknown mapping for queue %u\n", queue);
+			return -EINVAL;
+		}
+
+		/* add the pages from the public queue */
+		pages_free += (free_txpg[1] >> 16) & 0xfff;
+	}
+
+	pages_needed = DIV_ROUND_UP(count, rtwdev->chip->page_size);
+
+	if (pages_needed > pages_free) {
+		rtw_dbg(rtwdev, RTW_DBG_SDIO,
+			"Not enough free pages (%u needed, %u free) in queue %u for %zu bytes\n",
+			pages_needed, pages_free, queue, count);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static int rtw_sdio_write_port(struct rtw_dev *rtwdev, struct sk_buff *skb,
+			       enum rtw_tx_queue_type queue)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool bus_claim;
+	size_t txsize;
+	u32 txaddr;
+	int ret;
+
+	txaddr = rtw_sdio_get_tx_addr(rtwdev, skb->len, queue);
+	if (!txaddr)
+		return -EINVAL;
+
+	txsize = sdio_align_size(rtwsdio->sdio_func, skb->len);
+
+	ret = rtw_sdio_check_free_txpg(rtwdev, queue, txsize);
+	if (ret)
+		return ret;
+
+	if (!IS_ALIGNED((unsigned long)skb->data, RTW_SDIO_DATA_PTR_ALIGN))
+		rtw_warn(rtwdev, "Got unaligned SKB in %s() for queue %u\n",
+			 __func__, queue);
+
+	bus_claim = rtw_sdio_bus_claim_needed(rtwsdio);
+
+	if (bus_claim)
+		sdio_claim_host(rtwsdio->sdio_func);
+
+	ret = sdio_memcpy_toio(rtwsdio->sdio_func, txaddr, skb->data, txsize);
+
+	if (bus_claim)
+		sdio_release_host(rtwsdio->sdio_func);
+
+	if (ret)
+		rtw_warn(rtwdev,
+			 "Failed to write %zu byte(s) to SDIO port 0x%08x",
+			 txsize, txaddr);
+
+	return ret;
+}
+
+static void rtw_sdio_init(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+
+	rtwsdio->irq_mask = REG_SDIO_HIMR_RX_REQUEST | REG_SDIO_HIMR_CPWM1;
+}
+
+static void rtw_sdio_enable_rx_aggregation(struct rtw_dev *rtwdev)
+{
+	u8 size, timeout;
+
+	if (rtw_chip_wcpu_11n(rtwdev)) {
+		size = 0x6;
+		timeout = 0x6;
+	} else {
+		size = 0xff;
+		timeout = 0x1;
+	}
+
+	/* Make the firmware honor the size limit configured below */
+	rtw_write32_set(rtwdev, REG_RXDMA_AGG_PG_TH, BIT_EN_PRE_CALC);
+
+	rtw_write8_set(rtwdev, REG_TXDMA_PQ_MAP, BIT_RXDMA_AGG_EN);
+
+	rtw_write16(rtwdev, REG_RXDMA_AGG_PG_TH,
+		    FIELD_PREP(BIT_RXDMA_AGG_PG_TH, size) |
+		    FIELD_PREP(BIT_DMA_AGG_TO_V1, timeout));
+
+	rtw_write8_set(rtwdev, REG_RXDMA_MODE, BIT_DMA_MODE);
+}
+
+static void rtw_sdio_enable_interrupt(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+
+	rtw_write32(rtwdev, REG_SDIO_HIMR, rtwsdio->irq_mask);
+}
+
+static void rtw_sdio_disable_interrupt(struct rtw_dev *rtwdev)
+{
+	rtw_write32(rtwdev, REG_SDIO_HIMR, 0x0);
+}
+
+static u8 rtw_sdio_get_tx_qsel(struct rtw_dev *rtwdev, struct sk_buff *skb,
+			       u8 queue)
+{
+	switch (queue) {
+	case RTW_TX_QUEUE_BCN:
+		return TX_DESC_QSEL_BEACON;
+	case RTW_TX_QUEUE_H2C:
+		return TX_DESC_QSEL_H2C;
+	case RTW_TX_QUEUE_MGMT:
+		if (rtw_chip_wcpu_11n(rtwdev))
+			return TX_DESC_QSEL_HIGH;
+		else
+			return TX_DESC_QSEL_MGMT;
+	case RTW_TX_QUEUE_HI0:
+		return TX_DESC_QSEL_HIGH;
+	default:
+		return skb->priority;
+	}
+}
+
+static int rtw_sdio_setup(struct rtw_dev *rtwdev)
+{
+	/* nothing to do */
+	return 0;
+}
+
+static int rtw_sdio_start(struct rtw_dev *rtwdev)
+{
+	rtw_sdio_enable_rx_aggregation(rtwdev);
+	rtw_sdio_enable_interrupt(rtwdev);
+
+	return 0;
+}
+
+static void rtw_sdio_stop(struct rtw_dev *rtwdev)
+{
+	rtw_sdio_disable_interrupt(rtwdev);
+}
+
+static void rtw_sdio_deep_ps_enter(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	bool tx_empty = true;
+	u8 queue;
+
+	if (!rtw_fw_feature_check(&rtwdev->fw, FW_FEATURE_TX_WAKE)) {
+		/* Deep PS state is not allowed to TX-DMA */
+		for (queue = 0; queue < RTK_MAX_TX_QUEUE_NUM; queue++) {
+			/* BCN queue is rsvd page, does not have DMA interrupt
+			 * H2C queue is managed by firmware
+			 */
+			if (queue == RTW_TX_QUEUE_BCN ||
+			    queue == RTW_TX_QUEUE_H2C)
+				continue;
+
+			/* check if there is any skb DMAing */
+			if (skb_queue_len(&rtwsdio->tx_queue[queue])) {
+				tx_empty = false;
+				break;
+			}
+		}
+	}
+
+	if (!tx_empty) {
+		rtw_dbg(rtwdev, RTW_DBG_PS,
+			"TX path not empty, cannot enter deep power save state\n");
+		return;
+	}
+
+	set_bit(RTW_FLAG_LEISURE_PS_DEEP, rtwdev->flags);
+	rtw_power_mode_change(rtwdev, true);
+}
+
+static void rtw_sdio_deep_ps_leave(struct rtw_dev *rtwdev)
+{
+	if (test_and_clear_bit(RTW_FLAG_LEISURE_PS_DEEP, rtwdev->flags))
+		rtw_power_mode_change(rtwdev, false);
+}
+
+static void rtw_sdio_deep_ps(struct rtw_dev *rtwdev, bool enter)
+{
+	if (enter && !test_bit(RTW_FLAG_LEISURE_PS_DEEP, rtwdev->flags))
+		rtw_sdio_deep_ps_enter(rtwdev);
+
+	if (!enter && test_bit(RTW_FLAG_LEISURE_PS_DEEP, rtwdev->flags))
+		rtw_sdio_deep_ps_leave(rtwdev);
+}
+
+static void rtw_sdio_tx_kick_off(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+
+	queue_work(rtwsdio->txwq, &rtwsdio->tx_handler_data->work);
+}
+
+static void rtw_sdio_link_ps(struct rtw_dev *rtwdev, bool enter)
+{
+	/* nothing to do */
+}
+
+static void rtw_sdio_interface_cfg(struct rtw_dev *rtwdev)
+{
+	u32 val;
+
+	rtw_read32(rtwdev, REG_SDIO_FREE_TXPG);
+
+	val = rtw_read32(rtwdev, REG_SDIO_TX_CTRL);
+	val &= 0xfff8;
+	rtw_write32(rtwdev, REG_SDIO_TX_CTRL, val);
+}
+
+static struct rtw_sdio_tx_data *rtw_sdio_get_tx_data(struct sk_buff *skb)
+{
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
+	BUILD_BUG_ON(sizeof(struct rtw_sdio_tx_data) >
+		     sizeof(info->status.status_driver_data));
+
+	return (struct rtw_sdio_tx_data *)info->status.status_driver_data;
+}
+
+static void rtw_sdio_tx_skb_prepare(struct rtw_dev *rtwdev,
+				    struct rtw_tx_pkt_info *pkt_info,
+				    struct sk_buff *skb,
+				    enum rtw_tx_queue_type queue)
+{
+	const struct rtw_chip_info *chip = rtwdev->chip;
+	unsigned long data_addr, aligned_addr;
+	size_t offset;
+	u8 *pkt_desc;
+
+	pkt_desc = skb_push(skb, chip->tx_pkt_desc_sz);
+
+	data_addr = (unsigned long)pkt_desc;
+	aligned_addr = ALIGN(data_addr, RTW_SDIO_DATA_PTR_ALIGN);
+
+	if (data_addr != aligned_addr) {
+		/* Ensure that the start of the pkt_desc is always aligned at
+		 * RTW_SDIO_DATA_PTR_ALIGN.
+		 */
+		offset = RTW_SDIO_DATA_PTR_ALIGN - (aligned_addr - data_addr);
+
+		pkt_desc = skb_push(skb, offset);
+
+		/* By inserting padding to align the start of the pkt_desc we
+		 * need to inform the firmware that the actual data starts at
+		 * a different offset than normal.
+		 */
+		pkt_info->offset += offset;
+	}
+
+	memset(pkt_desc, 0, chip->tx_pkt_desc_sz);
+
+	pkt_info->qsel = rtw_sdio_get_tx_qsel(rtwdev, skb, queue);
+
+	rtw_tx_fill_tx_desc(pkt_info, skb);
+	rtw_tx_fill_txdesc_checksum(rtwdev, pkt_info, pkt_desc);
+}
+
+static int rtw_sdio_write_data(struct rtw_dev *rtwdev,
+			       struct rtw_tx_pkt_info *pkt_info,
+			       struct sk_buff *skb,
+			       enum rtw_tx_queue_type queue)
+{
+	int ret;
+
+	rtw_sdio_tx_skb_prepare(rtwdev, pkt_info, skb, queue);
+
+	ret = rtw_sdio_write_port(rtwdev, skb, queue);
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
+static int rtw_sdio_write_data_rsvd_page(struct rtw_dev *rtwdev, u8 *buf,
+					 u32 size)
+{
+	struct rtw_tx_pkt_info pkt_info = {};
+	struct sk_buff *skb;
+
+	skb = rtw_tx_write_data_rsvd_page_get(rtwdev, &pkt_info, buf, size);
+	if (!skb)
+		return -ENOMEM;
+
+	return rtw_sdio_write_data(rtwdev, &pkt_info, skb, RTW_TX_QUEUE_BCN);
+}
+
+static int rtw_sdio_write_data_h2c(struct rtw_dev *rtwdev, u8 *buf, u32 size)
+{
+	struct rtw_tx_pkt_info pkt_info = {};
+	struct sk_buff *skb;
+
+	skb = rtw_tx_write_data_h2c_get(rtwdev, &pkt_info, buf, size);
+	if (!skb)
+		return -ENOMEM;
+
+	return rtw_sdio_write_data(rtwdev, &pkt_info, skb, RTW_TX_QUEUE_H2C);
+}
+
+static int rtw_sdio_tx_write(struct rtw_dev *rtwdev,
+			     struct rtw_tx_pkt_info *pkt_info,
+			     struct sk_buff *skb)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	enum rtw_tx_queue_type queue = rtw_tx_queue_mapping(skb);
+	struct rtw_sdio_tx_data *tx_data;
+
+	rtw_sdio_tx_skb_prepare(rtwdev, pkt_info, skb, queue);
+
+	tx_data = rtw_sdio_get_tx_data(skb);
+	tx_data->sn = pkt_info->sn;
+
+	skb_queue_tail(&rtwsdio->tx_queue[queue], skb);
+
+	return 0;
+}
+
+static void rtw_sdio_tx_err_isr(struct rtw_dev *rtwdev)
+{
+	u32 val = rtw_read32(rtwdev, REG_TXDMA_STATUS);
+
+	rtw_write32(rtwdev, REG_TXDMA_STATUS, val);
+}
+
+static void rtw_sdio_rx_skb(struct rtw_dev *rtwdev, struct sk_buff *skb,
+			    u32 pkt_offset, struct rtw_rx_pkt_stat *pkt_stat,
+			    struct ieee80211_rx_status *rx_status)
+{
+	*IEEE80211_SKB_RXCB(skb) = *rx_status;
+
+	if (pkt_stat->is_c2h) {
+		skb_put(skb, pkt_stat->pkt_len + pkt_offset);
+		rtw_fw_c2h_cmd_rx_irqsafe(rtwdev, pkt_offset, skb);
+		return;
+	}
+
+	skb_put(skb, pkt_stat->pkt_len);
+	skb_reserve(skb, pkt_offset);
+
+	rtw_rx_stats(rtwdev, pkt_stat->vif, skb);
+
+	ieee80211_rx_irqsafe(rtwdev->hw, skb);
+}
+
+static void rtw_sdio_rxfifo_recv(struct rtw_dev *rtwdev, u32 rx_len)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	const struct rtw_chip_info *chip = rtwdev->chip;
+	u32 pkt_desc_sz = chip->rx_pkt_desc_sz;
+	struct ieee80211_rx_status rx_status;
+	struct rtw_rx_pkt_stat pkt_stat;
+	struct sk_buff *skb, *split_skb;
+	u32 pkt_offset, curr_pkt_len;
+	size_t bufsz;
+	u8 *rx_desc;
+	int ret;
+
+	bufsz = sdio_align_size(rtwsdio->sdio_func, rx_len);
+
+	skb = dev_alloc_skb(bufsz);
+	if (!skb)
+		return;
+
+	ret = rtw_sdio_read_port(rtwdev, skb->data, bufsz);
+	if (ret) {
+		dev_kfree_skb_any(skb);
+		return;
+	}
+
+	while (true) {
+		rx_desc = skb->data;
+		chip->ops->query_rx_desc(rtwdev, rx_desc, &pkt_stat,
+					 &rx_status);
+		pkt_offset = pkt_desc_sz + pkt_stat.drv_info_sz +
+			     pkt_stat.shift;
+
+		curr_pkt_len = ALIGN(pkt_offset + pkt_stat.pkt_len,
+				     RTW_SDIO_DATA_PTR_ALIGN);
+
+		if ((curr_pkt_len + pkt_desc_sz) >= rx_len) {
+			/* Use the original skb (with it's adjusted offset)
+			 * when processing the last (or even the only) entry to
+			 * have it's memory freed automatically.
+			 */
+			rtw_sdio_rx_skb(rtwdev, skb, pkt_offset, &pkt_stat,
+					&rx_status);
+			break;
+		}
+
+		split_skb = dev_alloc_skb(curr_pkt_len);
+		if (!split_skb) {
+			rtw_sdio_rx_skb(rtwdev, skb, pkt_offset, &pkt_stat,
+					&rx_status);
+			break;
+		}
+
+		skb_copy_header(split_skb, skb);
+		memcpy(split_skb->data, skb->data, curr_pkt_len);
+
+		rtw_sdio_rx_skb(rtwdev, split_skb, pkt_offset, &pkt_stat,
+				&rx_status);
+
+		/* Move to the start of the next RX descriptor */
+		skb_reserve(skb, curr_pkt_len);
+		rx_len -= curr_pkt_len;
+	}
+}
+
+static void rtw_sdio_rx_isr(struct rtw_dev *rtwdev)
+{
+	u32 rx_len, total_rx_bytes = 0;
+
+	while (total_rx_bytes < SZ_64K) {
+		if (rtw_chip_wcpu_11n(rtwdev))
+			rx_len = rtw_read16(rtwdev, REG_SDIO_RX0_REQ_LEN);
+		else
+			rx_len = rtw_read32(rtwdev, REG_SDIO_RX0_REQ_LEN);
+
+		if (!rx_len)
+			break;
+
+		rtw_sdio_rxfifo_recv(rtwdev, rx_len);
+
+		total_rx_bytes += rx_len;
+	}
+}
+
+static void rtw_sdio_handle_interrupt(struct sdio_func *sdio_func)
+{
+	struct ieee80211_hw *hw = sdio_get_drvdata(sdio_func);
+	struct rtw_sdio *rtwsdio;
+	struct rtw_dev *rtwdev;
+	u32 hisr;
+
+	rtwdev = hw->priv;
+	rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+
+	rtwsdio->irq_thread = current;
+
+	hisr = rtw_read32(rtwdev, REG_SDIO_HISR);
+
+	if (hisr & REG_SDIO_HISR_TXERR)
+		rtw_sdio_tx_err_isr(rtwdev);
+	if (hisr & REG_SDIO_HISR_RX_REQUEST) {
+		hisr &= ~REG_SDIO_HISR_RX_REQUEST;
+		rtw_sdio_rx_isr(rtwdev);
+	}
+
+	rtw_write32(rtwdev, REG_SDIO_HISR, hisr);
+
+	rtwsdio->irq_thread = NULL;
+}
+
+static int __maybe_unused rtw_sdio_suspend(struct device *dev)
+{
+	struct sdio_func *func = dev_to_sdio_func(dev);
+	struct ieee80211_hw *hw = dev_get_drvdata(dev);
+	struct rtw_dev *rtwdev = hw->priv;
+	int ret;
+
+	ret = sdio_set_host_pm_flags(func, MMC_PM_KEEP_POWER);
+	if (ret)
+		rtw_err(rtwdev, "Failed to host PM flag MMC_PM_KEEP_POWER");
+
+	return ret;
+}
+
+static int __maybe_unused rtw_sdio_resume(struct device *dev)
+{
+	return 0;
+}
+
+SIMPLE_DEV_PM_OPS(rtw_sdio_pm_ops, rtw_sdio_suspend, rtw_sdio_resume);
+EXPORT_SYMBOL(rtw_sdio_pm_ops);
+
+static int rtw_sdio_claim(struct rtw_dev *rtwdev, struct sdio_func *sdio_func)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	int ret;
+
+	sdio_claim_host(sdio_func);
+
+	ret = sdio_enable_func(sdio_func);
+	if (ret) {
+		rtw_err(rtwdev, "Failed to enable SDIO func");
+		goto err_release_host;
+	}
+
+	ret = sdio_set_block_size(sdio_func, RTW_SDIO_BLOCK_SIZE);
+	if (ret) {
+		rtw_err(rtwdev, "Failed to set SDIO block size to 512");
+		goto err_disable_func;
+	}
+
+	rtwsdio->sdio_func = sdio_func;
+
+	rtwsdio->sdio3_bus_mode = mmc_card_uhs(sdio_func->card);
+
+	sdio_set_drvdata(sdio_func, rtwdev->hw);
+	SET_IEEE80211_DEV(rtwdev->hw, &sdio_func->dev);
+
+	sdio_release_host(sdio_func);
+
+	return 0;
+
+err_disable_func:
+	sdio_disable_func(sdio_func);
+err_release_host:
+	sdio_release_host(sdio_func);
+	return ret;
+}
+
+static void rtw_sdio_declaim(struct rtw_dev *rtwdev,
+			     struct sdio_func *sdio_func)
+{
+	sdio_claim_host(sdio_func);
+	sdio_disable_func(sdio_func);
+	sdio_release_host(sdio_func);
+}
+
+static struct rtw_hci_ops rtw_sdio_ops = {
+	.tx_write = rtw_sdio_tx_write,
+	.tx_kick_off = rtw_sdio_tx_kick_off,
+	.setup = rtw_sdio_setup,
+	.start = rtw_sdio_start,
+	.stop = rtw_sdio_stop,
+	.deep_ps = rtw_sdio_deep_ps,
+	.link_ps = rtw_sdio_link_ps,
+	.interface_cfg = rtw_sdio_interface_cfg,
+
+	.read8 = rtw_sdio_read8,
+	.read16 = rtw_sdio_read16,
+	.read32 = rtw_sdio_read32,
+	.write8 = rtw_sdio_write8,
+	.write16 = rtw_sdio_write16,
+	.write32 = rtw_sdio_write32,
+	.write_data_rsvd_page = rtw_sdio_write_data_rsvd_page,
+	.write_data_h2c = rtw_sdio_write_data_h2c,
+};
+
+static int rtw_sdio_request_irq(struct rtw_dev *rtwdev,
+				struct sdio_func *sdio_func)
+{
+	int ret;
+
+	sdio_claim_host(sdio_func);
+	ret = sdio_claim_irq(sdio_func, &rtw_sdio_handle_interrupt);
+	sdio_release_host(sdio_func);
+
+	if (ret) {
+		rtw_err(rtwdev, "failed to claim SDIO IRQ");
+		return ret;
+	}
+
+	return 0;
+}
+
+static void rtw_sdio_indicate_tx_status(struct rtw_dev *rtwdev,
+					struct sk_buff *skb)
+{
+	struct rtw_sdio_tx_data *tx_data = rtw_sdio_get_tx_data(skb);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_hw *hw = rtwdev->hw;
+
+	/* enqueue to wait for tx report */
+	if (info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS) {
+		rtw_tx_report_enqueue(rtwdev, skb, tx_data->sn);
+		return;
+	}
+
+	/* always ACK for others, then they won't be marked as drop */
+	ieee80211_tx_info_clear_status(info);
+	if (info->flags & IEEE80211_TX_CTL_NO_ACK)
+		info->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
+	else
+		info->flags |= IEEE80211_TX_STAT_ACK;
+
+	ieee80211_tx_status_irqsafe(hw, skb);
+}
+
+static void rtw_sdio_process_tx_queue(struct rtw_dev *rtwdev,
+				      enum rtw_tx_queue_type queue)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	struct sk_buff *skb;
+	int ret;
+
+	skb = skb_dequeue(&rtwsdio->tx_queue[queue]);
+	if (!skb)
+		return;
+
+	ret = rtw_sdio_write_port(rtwdev, skb, queue);
+	if (ret) {
+		skb_queue_head(&rtwsdio->tx_queue[queue], skb);
+		return;
+	}
+
+	if (queue <= RTW_TX_QUEUE_VO)
+		rtw_sdio_indicate_tx_status(rtwdev, skb);
+	else
+		dev_kfree_skb_any(skb);
+}
+
+static void rtw_sdio_tx_handler(struct work_struct *work)
+{
+	struct rtw_sdio_work_data *work_data =
+		container_of(work, struct rtw_sdio_work_data, work);
+	struct rtw_sdio *rtwsdio;
+	struct rtw_dev *rtwdev;
+	int limit, queue;
+
+	rtwdev = work_data->rtwdev;
+	rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+
+	if (!rtw_fw_feature_check(&rtwdev->fw, FW_FEATURE_TX_WAKE))
+		rtw_sdio_deep_ps_leave(rtwdev);
+
+	for (queue = RTK_MAX_TX_QUEUE_NUM - 1; queue >= 0; queue--) {
+		for (limit = 0; limit < 1000; limit++) {
+			rtw_sdio_process_tx_queue(rtwdev, queue);
+
+			if (skb_queue_empty(&rtwsdio->tx_queue[queue]))
+				break;
+		}
+	}
+}
+
+static void rtw_sdio_free_irq(struct rtw_dev *rtwdev,
+			      struct sdio_func *sdio_func)
+{
+	sdio_claim_host(sdio_func);
+	sdio_release_irq(sdio_func);
+	sdio_release_host(sdio_func);
+}
+
+static int rtw_sdio_init_tx(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	int i;
+
+	rtwsdio->txwq = create_singlethread_workqueue("rtw88_sdio: tx wq");
+	if (!rtwsdio->txwq) {
+		rtw_err(rtwdev, "failed to create TX work queue\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < RTK_MAX_TX_QUEUE_NUM; i++)
+		skb_queue_head_init(&rtwsdio->tx_queue[i]);
+	rtwsdio->tx_handler_data = kmalloc(sizeof(*rtwsdio->tx_handler_data),
+					   GFP_KERNEL);
+	if (!rtwsdio->tx_handler_data)
+		goto err_destroy_wq;
+
+	rtwsdio->tx_handler_data->rtwdev = rtwdev;
+	INIT_WORK(&rtwsdio->tx_handler_data->work, rtw_sdio_tx_handler);
+
+	return 0;
+
+err_destroy_wq:
+	destroy_workqueue(rtwsdio->txwq);
+	return -ENOMEM;
+}
+
+static void rtw_sdio_deinit_tx(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+	int i;
+
+	for (i = 0; i < RTK_MAX_TX_QUEUE_NUM; i++)
+		skb_queue_purge(&rtwsdio->tx_queue[i]);
+
+	flush_workqueue(rtwsdio->txwq);
+	destroy_workqueue(rtwsdio->txwq);
+	kfree(rtwsdio->tx_handler_data);
+}
+
+int rtw_sdio_probe(struct sdio_func *sdio_func,
+		   const struct sdio_device_id *id)
+{
+	struct ieee80211_hw *hw;
+	struct rtw_dev *rtwdev;
+	int drv_data_size;
+	int ret;
+
+	drv_data_size = sizeof(struct rtw_dev) + sizeof(struct rtw_sdio);
+	hw = ieee80211_alloc_hw(drv_data_size, &rtw_ops);
+	if (!hw) {
+		dev_err(&sdio_func->dev, "failed to allocate hw");
+		return -ENOMEM;
+	}
+
+	rtwdev = hw->priv;
+	rtwdev->hw = hw;
+	rtwdev->dev = &sdio_func->dev;
+	rtwdev->chip = (struct rtw_chip_info *)id->driver_data;
+	rtwdev->hci.ops = &rtw_sdio_ops;
+	rtwdev->hci.type = RTW_HCI_TYPE_SDIO;
+
+	ret = rtw_core_init(rtwdev);
+	if (ret)
+		goto err_release_hw;
+
+	rtw_dbg(rtwdev, RTW_DBG_SDIO,
+		"rtw88 SDIO probe: vendor=0x%04x device=%04x class=%02x",
+		id->vendor, id->device, id->class);
+
+	ret = rtw_sdio_claim(rtwdev, sdio_func);
+	if (ret) {
+		rtw_err(rtwdev, "failed to claim SDIO device");
+		goto err_deinit_core;
+	}
+
+	rtw_sdio_init(rtwdev);
+
+	ret = rtw_sdio_init_tx(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to init SDIO TX queue\n");
+		goto err_sdio_declaim;
+	}
+
+	ret = rtw_chip_info_setup(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to setup chip information");
+		goto err_destroy_txwq;
+	}
+
+	ret = rtw_sdio_request_irq(rtwdev, sdio_func);
+	if (ret)
+		goto err_destroy_txwq;
+
+	ret = rtw_register_hw(rtwdev, hw);
+	if (ret) {
+		rtw_err(rtwdev, "failed to register hw");
+		goto err_free_irq;
+	}
+
+	return 0;
+
+err_free_irq:
+	rtw_sdio_free_irq(rtwdev, sdio_func);
+err_destroy_txwq:
+	rtw_sdio_deinit_tx(rtwdev);
+err_sdio_declaim:
+	rtw_sdio_declaim(rtwdev, sdio_func);
+err_deinit_core:
+	rtw_core_deinit(rtwdev);
+err_release_hw:
+	ieee80211_free_hw(hw);
+
+	return ret;
+}
+EXPORT_SYMBOL(rtw_sdio_probe);
+
+void rtw_sdio_remove(struct sdio_func *sdio_func)
+{
+	struct ieee80211_hw *hw = sdio_get_drvdata(sdio_func);
+	struct rtw_dev *rtwdev;
+
+	if (!hw)
+		return;
+
+	rtwdev = hw->priv;
+
+	rtw_unregister_hw(rtwdev, hw);
+	rtw_sdio_disable_interrupt(rtwdev);
+	rtw_sdio_free_irq(rtwdev, sdio_func);
+	rtw_sdio_declaim(rtwdev, sdio_func);
+	rtw_sdio_deinit_tx(rtwdev);
+	rtw_core_deinit(rtwdev);
+	ieee80211_free_hw(hw);
+}
+EXPORT_SYMBOL(rtw_sdio_remove);
+
+void rtw_sdio_shutdown(struct device *dev)
+{
+	struct sdio_func *sdio_func = dev_to_sdio_func(dev);
+	const struct rtw_chip_info *chip;
+	struct ieee80211_hw *hw;
+	struct rtw_dev *rtwdev;
+
+	hw = sdio_get_drvdata(sdio_func);
+	if (!hw)
+		return;
+
+	rtwdev = hw->priv;
+	chip = rtwdev->chip;
+
+	if (chip->ops->shutdown)
+		chip->ops->shutdown(rtwdev);
+}
+EXPORT_SYMBOL(rtw_sdio_shutdown);
+
+MODULE_AUTHOR("Martin Blumenstingl");
+MODULE_AUTHOR("Jernej Skrabec");
+MODULE_DESCRIPTION("Realtek 802.11ac wireless SDIO driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/realtek/rtw88/sdio.h b/drivers/net/wireless/realtek/rtw88/sdio.h
new file mode 100644
index 000000000000..3c659ed180f0
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/sdio.h
@@ -0,0 +1,178 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright (C) 2021 Martin Blumenstingl <martin.blumenstingl@googlemail.com>
+ * Copyright (C) 2021 Jernej Skrabec <jernej.skrabec@gmail.com>
+ */
+
+#ifndef __REG_SDIO_H_
+#define __REG_SDIO_H_
+
+/* I/O bus domain address mapping */
+#define SDIO_LOCAL_OFFSET			0x10250000
+#define WLAN_IOREG_OFFSET			0x10260000
+#define FIRMWARE_FIFO_OFFSET			0x10270000
+#define TX_HIQ_OFFSET				0x10310000
+#define TX_MIQ_OFFSET				0x10320000
+#define TX_LOQ_OFFSET				0x10330000
+#define TX_EPQ_OFFSET				0x10350000
+#define RX_RX0FF_OFFSET				0x10340000
+
+#define RTW_SDIO_BUS_MSK			0xffff0000
+#define SDIO_LOCAL_REG_MSK			0x00000fff
+#define WLAN_IOREG_REG_MSK			0x0000ffff
+
+/* SDIO Tx Control */
+#define REG_SDIO_TX_CTRL			(SDIO_LOCAL_OFFSET + 0x0000)
+
+/*SDIO status timeout*/
+#define REG_SDIO_TIMEOUT			(SDIO_LOCAL_OFFSET + 0x0002)
+
+/* SDIO Host Interrupt Mask */
+#define REG_SDIO_HIMR				(SDIO_LOCAL_OFFSET + 0x0014)
+#define REG_SDIO_HIMR_RX_REQUEST		BIT(0)
+#define REG_SDIO_HIMR_AVAL			BIT(1)
+#define REG_SDIO_HIMR_TXERR			BIT(2)
+#define REG_SDIO_HIMR_RXERR			BIT(3)
+#define REG_SDIO_HIMR_TXFOVW			BIT(4)
+#define REG_SDIO_HIMR_RXFOVW			BIT(5)
+#define REG_SDIO_HIMR_TXBCNOK			BIT(6)
+#define REG_SDIO_HIMR_TXBCNERR			BIT(7)
+#define REG_SDIO_HIMR_BCNERLY_INT		BIT(16)
+#define REG_SDIO_HIMR_C2HCMD			BIT(17)
+#define REG_SDIO_HIMR_CPWM1			BIT(18)
+#define REG_SDIO_HIMR_CPWM2			BIT(19)
+#define REG_SDIO_HIMR_HSISR_IND			BIT(20)
+#define REG_SDIO_HIMR_GTINT3_IND		BIT(21)
+#define REG_SDIO_HIMR_GTINT4_IND		BIT(22)
+#define REG_SDIO_HIMR_PSTIMEOUT			BIT(23)
+#define REG_SDIO_HIMR_OCPINT			BIT(24)
+#define REG_SDIO_HIMR_ATIMEND			BIT(25)
+#define REG_SDIO_HIMR_ATIMEND_E			BIT(26)
+#define REG_SDIO_HIMR_CTWEND			BIT(27)
+/* the following two are RTL8188 SDIO Specific */
+#define REG_SDIO_HIMR_MCU_ERR			BIT(28)
+#define REG_SDIO_HIMR_TSF_BIT32_TOGGLE		BIT(29)
+
+/* SDIO Host Interrupt Service Routine */
+#define REG_SDIO_HISR				(SDIO_LOCAL_OFFSET + 0x0018)
+#define REG_SDIO_HISR_RX_REQUEST		BIT(0)
+#define REG_SDIO_HISR_AVAL			BIT(1)
+#define REG_SDIO_HISR_TXERR			BIT(2)
+#define REG_SDIO_HISR_RXERR			BIT(3)
+#define REG_SDIO_HISR_TXFOVW			BIT(4)
+#define REG_SDIO_HISR_RXFOVW			BIT(5)
+#define REG_SDIO_HISR_TXBCNOK			BIT(6)
+#define REG_SDIO_HISR_TXBCNERR			BIT(7)
+#define REG_SDIO_HISR_BCNERLY_INT		BIT(16)
+#define REG_SDIO_HISR_C2HCMD			BIT(17)
+#define REG_SDIO_HISR_CPWM1			BIT(18)
+#define REG_SDIO_HISR_CPWM2			BIT(19)
+#define REG_SDIO_HISR_HSISR_IND			BIT(20)
+#define REG_SDIO_HISR_GTINT3_IND		BIT(21)
+#define REG_SDIO_HISR_GTINT4_IND		BIT(22)
+#define REG_SDIO_HISR_PSTIMEOUT			BIT(23)
+#define REG_SDIO_HISR_OCPINT			BIT(24)
+#define REG_SDIO_HISR_ATIMEND			BIT(25)
+#define REG_SDIO_HISR_ATIMEND_E			BIT(26)
+#define REG_SDIO_HISR_CTWEND			BIT(27)
+/* the following two are RTL8188 SDIO Specific */
+#define REG_SDIO_HISR_MCU_ERR			BIT(28)
+#define REG_SDIO_HISR_TSF_BIT32_TOGGLE		BIT(29)
+
+/* HCI Current Power Mode */
+#define REG_SDIO_HCPWM				(SDIO_LOCAL_OFFSET + 0x0019)
+/* RXDMA Request Length */
+#define REG_SDIO_RX0_REQ_LEN			(SDIO_LOCAL_OFFSET + 0x001C)
+/* OQT Free Page */
+#define REG_SDIO_OQT_FREE_PG			(SDIO_LOCAL_OFFSET + 0x001E)
+/* Free Tx Buffer Page */
+#define REG_SDIO_FREE_TXPG			(SDIO_LOCAL_OFFSET + 0x0020)
+/* HCI Current Power Mode 1 */
+#define REG_SDIO_HCPWM1				(SDIO_LOCAL_OFFSET + 0x0024)
+/* HCI Current Power Mode 2 */
+#define REG_SDIO_HCPWM2				(SDIO_LOCAL_OFFSET + 0x0026)
+/* Free Tx Page Sequence */
+#define REG_SDIO_FREE_TXPG_SEQ			(SDIO_LOCAL_OFFSET + 0x0028)
+/* HTSF Information */
+#define REG_SDIO_HTSFR_INFO			(SDIO_LOCAL_OFFSET + 0x0030)
+#define REG_SDIO_HCPWM1_V2			(SDIO_LOCAL_OFFSET + 0x0038)
+/* H2C */
+#define REG_SDIO_H2C				(SDIO_LOCAL_OFFSET + 0x0060)
+/* HCI Request Power Mode 1 */
+#define REG_SDIO_HRPWM1				(SDIO_LOCAL_OFFSET + 0x0080)
+/* HCI Request Power Mode 2 */
+#define REG_SDIO_HRPWM2				(SDIO_LOCAL_OFFSET + 0x0082)
+/* HCI Power Save Clock */
+#define REG_SDIO_HPS_CLKR			(SDIO_LOCAL_OFFSET + 0x0084)
+/* SDIO HCI Suspend Control */
+#define REG_SDIO_HSUS_CTRL			(SDIO_LOCAL_OFFSET + 0x0086)
+#define BIT_HCI_SUS_REQ				BIT(0)
+#define BIT_HCI_RESUME_RDY			BIT(1)
+/* SDIO Host Extension Interrupt Mask Always */
+#define REG_SDIO_HIMR_ON			(SDIO_LOCAL_OFFSET + 0x0090)
+/* SDIO Host Extension Interrupt Status Always */
+#define REG_SDIO_HISR_ON			(SDIO_LOCAL_OFFSET + 0x0091)
+
+#define REG_SDIO_INDIRECT_REG_CFG		(SDIO_LOCAL_OFFSET + 0x0040)
+#define BIT_SDIO_INDIRECT_REG_CFG_WORD		BIT(16)
+#define BIT_SDIO_INDIRECT_REG_CFG_DWORD		BIT(17)
+#define BIT_SDIO_INDIRECT_REG_CFG_WRITE		BIT(18)
+#define BIT_SDIO_INDIRECT_REG_CFG_READ		BIT(19)
+#define BIT_SDIO_INDIRECT_REG_CFG_UNK20		BIT(20)
+#define REG_SDIO_INDIRECT_REG_DATA		(SDIO_LOCAL_OFFSET + 0x0044)
+
+/* Sdio Address for SDIO Local Reg, TRX FIFO, MAC Reg */
+#define REG_SDIO_CMD_ADDR_MSK			GENMASK(16, 13)
+#define REG_SDIO_CMD_ADDR_SDIO_REG		0
+#define REG_SDIO_CMD_ADDR_MAC_REG		8
+#define REG_SDIO_CMD_ADDR_TXFF_HIGH		4
+#define REG_SDIO_CMD_ADDR_TXFF_LOW		6
+#define REG_SDIO_CMD_ADDR_TXFF_NORMAL		5
+#define REG_SDIO_CMD_ADDR_TXFF_EXTRA		7
+#define REG_SDIO_CMD_ADDR_RXFF			7
+
+#define RTW_SDIO_BLOCK_SIZE			512
+#define RTW_SDIO_ADDR_RX_RX0FF_GEN(_id)		(0x0e000 | ((_id) & 0x3))
+
+#define RTW_SDIO_DATA_PTR_ALIGN			8
+
+struct sdio_func;
+struct sdio_device_id;
+
+struct rtw_sdio_tx_data {
+	u8 sn;
+};
+
+struct rtw_sdio_work_data {
+	struct work_struct work;
+	struct rtw_dev *rtwdev;
+};
+
+struct rtw_sdio {
+	struct sdio_func *sdio_func;
+
+	u32 irq_mask;
+	u8 rx_addr;
+	bool sdio3_bus_mode;
+
+	void *irq_thread;
+
+	struct workqueue_struct *txwq;
+	struct rtw_sdio_work_data *tx_handler_data;
+	struct sk_buff_head tx_queue[RTK_MAX_TX_QUEUE_NUM];
+};
+
+extern const struct dev_pm_ops rtw_sdio_pm_ops;
+
+int rtw_sdio_probe(struct sdio_func *sdio_func,
+		   const struct sdio_device_id *id);
+void rtw_sdio_remove(struct sdio_func *sdio_func);
+void rtw_sdio_shutdown(struct device *dev);
+
+static inline bool rtw_sdio_is_sdio30_supported(struct rtw_dev *rtwdev)
+{
+	struct rtw_sdio *rtwsdio = (struct rtw_sdio *)rtwdev->priv;
+
+	return rtwsdio->sdio3_bus_mode;
+}
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From b722e5b130bcaa8224c93c6625685d8276742cbd Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:23 +0200
Subject: wifi: rtw88: mac: Support SDIO specific bits in the power on sequence

Add the code specific to SDIO HCI in the MAC power on sequence. This is
based on the RTL8822BS and RTL8822CS vendor drivers.

Co-developed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-4-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c | 47 ++++++++++++++++++++++++++++----
 1 file changed, 42 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 2fcba43a6f72..44e07b61b9b9 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -61,6 +61,7 @@ EXPORT_SYMBOL(rtw_set_channel_mac);
 
 static int rtw_mac_pre_system_cfg(struct rtw_dev *rtwdev)
 {
+	unsigned int retry;
 	u32 value32;
 	u8 value8;
 
@@ -78,6 +79,28 @@ static int rtw_mac_pre_system_cfg(struct rtw_dev *rtwdev)
 	case RTW_HCI_TYPE_PCIE:
 		rtw_write32_set(rtwdev, REG_HCI_OPT_CTRL, BIT_USB_SUS_DIS);
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		rtw_write8_clr(rtwdev, REG_SDIO_HSUS_CTRL, BIT_HCI_SUS_REQ);
+
+		for (retry = 0; retry < RTW_PWR_POLLING_CNT; retry++) {
+			if (rtw_read8(rtwdev, REG_SDIO_HSUS_CTRL) & BIT_HCI_RESUME_RDY)
+				break;
+
+			usleep_range(10, 50);
+		}
+
+		if (retry == RTW_PWR_POLLING_CNT) {
+			rtw_err(rtwdev, "failed to poll REG_SDIO_HSUS_CTRL[1]");
+			return -ETIMEDOUT;
+		}
+
+		if (rtw_sdio_is_sdio30_supported(rtwdev))
+			rtw_write8_set(rtwdev, REG_HCI_OPT_CTRL + 2,
+				       BIT_SDIO_PAD_E5 >> 16);
+		else
+			rtw_write8_clr(rtwdev, REG_HCI_OPT_CTRL + 2,
+				       BIT_SDIO_PAD_E5 >> 16);
+		break;
 	case RTW_HCI_TYPE_USB:
 		break;
 	default:
@@ -249,6 +272,7 @@ static int rtw_mac_power_switch(struct rtw_dev *rtwdev, bool pwr_on)
 {
 	const struct rtw_chip_info *chip = rtwdev->chip;
 	const struct rtw_pwr_seq_cmd **pwr_seq;
+	u32 imr = 0;
 	u8 rpwm;
 	bool cur_pwr;
 	int ret;
@@ -274,18 +298,24 @@ static int rtw_mac_power_switch(struct rtw_dev *rtwdev, bool pwr_on)
 	if (pwr_on == cur_pwr)
 		return -EALREADY;
 
+	if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_SDIO) {
+		imr = rtw_read32(rtwdev, REG_SDIO_HIMR);
+		rtw_write32(rtwdev, REG_SDIO_HIMR, 0);
+	}
+
 	if (!pwr_on)
 		clear_bit(RTW_FLAG_POWERON, rtwdev->flags);
 
 	pwr_seq = pwr_on ? chip->pwr_on_seq : chip->pwr_off_seq;
 	ret = rtw_pwr_seq_parser(rtwdev, pwr_seq);
-	if (ret)
-		return ret;
 
-	if (pwr_on)
+	if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_SDIO)
+		rtw_write32(rtwdev, REG_SDIO_HIMR, imr);
+
+	if (!ret && pwr_on)
 		set_bit(RTW_FLAG_POWERON, rtwdev->flags);
 
-	return 0;
+	return ret;
 }
 
 static int __rtw_mac_init_system_cfg(struct rtw_dev *rtwdev)
@@ -456,6 +486,9 @@ static void download_firmware_reg_backup(struct rtw_dev *rtwdev,
 	rtw_write16(rtwdev, REG_FIFOPAGE_INFO_1, 0x200);
 	rtw_write32(rtwdev, REG_RQPN_CTRL_2, bckp[bckp_idx - 1].val);
 
+	if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_SDIO)
+		rtw_read32(rtwdev, REG_SDIO_FREE_TXPG);
+
 	/* Disable beacon related functions */
 	tmp = rtw_read8(rtwdev, REG_BCN_CTRL);
 	bckp[bckp_idx].len = 1;
@@ -1068,8 +1101,12 @@ static int txdma_queue_mapping(struct rtw_dev *rtwdev)
 	if (rtw_chip_wcpu_11ac(rtwdev))
 		rtw_write32(rtwdev, REG_H2CQ_CSR, BIT_H2CQ_FULL);
 
-	if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_USB)
+	if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_SDIO) {
+		rtw_read32(rtwdev, REG_SDIO_FREE_TXPG);
+		rtw_write32(rtwdev, REG_SDIO_TX_CTRL, 0);
+	} else if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_USB) {
 		rtw_write8_set(rtwdev, REG_TXDMA_PQ_MAP, BIT_RXDMA_ARBBW_EN);
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From a5d25f9ff91831f3a87aca96480a8b080546ccb8 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:24 +0200
Subject: wifi: rtw88: main: Add the {cpwm,rpwm}_addr for SDIO based chipsets

Initialize the rpwm_addr and cpwm_addr for power-saving support on SDIO
based chipsets.

Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-5-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/main.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index b2e78737bd5d..cdc4703ead5f 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -18,6 +18,7 @@
 #include "debug.h"
 #include "bf.h"
 #include "sar.h"
+#include "sdio.h"
 
 bool rtw_disable_lps_deep_mode;
 EXPORT_SYMBOL(rtw_disable_lps_deep_mode);
@@ -1785,6 +1786,10 @@ static int rtw_chip_parameter_setup(struct rtw_dev *rtwdev)
 		rtwdev->hci.rpwm_addr = 0x03d9;
 		rtwdev->hci.cpwm_addr = 0x03da;
 		break;
+	case RTW_HCI_TYPE_SDIO:
+		rtwdev->hci.rpwm_addr = REG_SDIO_HRPWM1;
+		rtwdev->hci.cpwm_addr = REG_SDIO_HCPWM1_V2;
+		break;
 	case RTW_HCI_TYPE_USB:
 		rtwdev->hci.rpwm_addr = 0xfe58;
 		rtwdev->hci.cpwm_addr = 0xfe57;
-- 
cgit v1.2.3-70-g09d2


From 02461d9368c59510ef51cc8a1db1f0f31cfbf9ad Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:25 +0200
Subject: wifi: rtw88: main: Reserve 8 bytes of extra TX headroom for SDIO
 cards

For SDIO host controllers with DMA support the TX buffer physical memory
address need to be aligned at an 8-byte boundary. Reserve 8 bytes of
extra TX headroom so we can align the data without re-allocating the
transmit buffer.

While here, also remove the TODO comment regarding extra headroom for
USB and SDIO. For SDIO the extra headroom is now handled and for USB it
was not needed so far.

Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-6-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/main.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index cdc4703ead5f..1cb553485cff 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -2163,9 +2163,11 @@ int rtw_register_hw(struct rtw_dev *rtwdev, struct ieee80211_hw *hw)
 	int max_tx_headroom = 0;
 	int ret;
 
-	/* TODO: USB & SDIO may need extra room? */
 	max_tx_headroom = rtwdev->chip->tx_pkt_desc_sz;
 
+	if (rtw_hci_type(rtwdev) == RTW_HCI_TYPE_SDIO)
+		max_tx_headroom += RTW_SDIO_DATA_PTR_ALIGN;
+
 	hw->extra_tx_headroom = max_tx_headroom;
 	hw->queues = IEEE80211_NUM_ACS;
 	hw->txq_data_size = sizeof(struct rtw_txq);
-- 
cgit v1.2.3-70-g09d2


From 7d6d2dd326a8a8d32091e9748f3428dd3be68367 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:26 +0200
Subject: mmc: sdio: add Realtek SDIO vendor ID and various wifi device IDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add the SDIO vendor ID for Realtek and some device IDs extracted from
their GPL vendor driver. This will be useful in the future when the
rtw88 driver gains support for these chips.

Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Reviewed-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-7-martin.blumenstingl@googlemail.com
---
 include/linux/mmc/sdio_ids.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/include/linux/mmc/sdio_ids.h b/include/linux/mmc/sdio_ids.h
index 0e4ef9c5127a..66f503ed2448 100644
--- a/include/linux/mmc/sdio_ids.h
+++ b/include/linux/mmc/sdio_ids.h
@@ -112,6 +112,15 @@
 #define SDIO_VENDOR_ID_MICROCHIP_WILC		0x0296
 #define SDIO_DEVICE_ID_MICROCHIP_WILC1000	0x5347
 
+#define SDIO_VENDOR_ID_REALTEK			0x024c
+#define SDIO_DEVICE_ID_REALTEK_RTW8723BS	0xb723
+#define SDIO_DEVICE_ID_REALTEK_RTW8821BS	0xb821
+#define SDIO_DEVICE_ID_REALTEK_RTW8822BS	0xb822
+#define SDIO_DEVICE_ID_REALTEK_RTW8821CS	0xc821
+#define SDIO_DEVICE_ID_REALTEK_RTW8822CS	0xc822
+#define SDIO_DEVICE_ID_REALTEK_RTW8723DS	0xd723
+#define SDIO_DEVICE_ID_REALTEK_RTW8821DS	0xd821
+
 #define SDIO_VENDOR_ID_SIANO			0x039a
 #define SDIO_DEVICE_ID_SIANO_NOVA_B0		0x0201
 #define SDIO_DEVICE_ID_SIANO_NICE		0x0202
-- 
cgit v1.2.3-70-g09d2


From 095e62dd74276dd801c04d9a2a9afcd922811218 Mon Sep 17 00:00:00 2001
From: Jernej Skrabec <jernej.skrabec@gmail.com>
Date: Wed, 5 Apr 2023 22:07:27 +0200
Subject: wifi: rtw88: Add support for the SDIO based RTL8822BS chipset

Wire up RTL8822BS chipset support using the new rtw88 SDIO HCI code as
well as the existing RTL8822B chipset code.

Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-8-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/Kconfig     | 11 ++++++++
 drivers/net/wireless/realtek/rtw88/Makefile    |  3 +++
 drivers/net/wireless/realtek/rtw88/rtw8822bs.c | 36 ++++++++++++++++++++++++++
 3 files changed, 50 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw88/rtw8822bs.c

diff --git a/drivers/net/wireless/realtek/rtw88/Kconfig b/drivers/net/wireless/realtek/rtw88/Kconfig
index cdf9cb478ee2..0cfc68dcc416 100644
--- a/drivers/net/wireless/realtek/rtw88/Kconfig
+++ b/drivers/net/wireless/realtek/rtw88/Kconfig
@@ -45,6 +45,17 @@ config RTW88_8822BE
 
 	  802.11ac PCIe wireless network adapter
 
+config RTW88_8822BS
+	tristate "Realtek 8822BS SDIO wireless network adapter"
+	depends on MMC
+	select RTW88_CORE
+	select RTW88_SDIO
+	select RTW88_8822B
+	help
+	  Select this option will enable support for 8822BS chipset
+
+	  802.11ac SDIO wireless network adapter
+
 config RTW88_8822BU
 	tristate "Realtek 8822BU USB wireless network adapter"
 	depends on USB
diff --git a/drivers/net/wireless/realtek/rtw88/Makefile b/drivers/net/wireless/realtek/rtw88/Makefile
index 892cad60ba31..2b8f4dd9707f 100644
--- a/drivers/net/wireless/realtek/rtw88/Makefile
+++ b/drivers/net/wireless/realtek/rtw88/Makefile
@@ -26,6 +26,9 @@ rtw88_8822b-objs		:= rtw8822b.o rtw8822b_table.o
 obj-$(CONFIG_RTW88_8822BE)	+= rtw88_8822be.o
 rtw88_8822be-objs		:= rtw8822be.o
 
+obj-$(CONFIG_RTW88_8822BS)	+= rtw88_8822bs.o
+rtw88_8822bs-objs		:= rtw8822bs.o
+
 obj-$(CONFIG_RTW88_8822BU)	+= rtw88_8822bu.o
 rtw88_8822bu-objs		:= rtw8822bu.o
 
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822bs.c b/drivers/net/wireless/realtek/rtw88/rtw8822bs.c
new file mode 100644
index 000000000000..31d8645f83bd
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822bs.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) Jernej Skrabec <jernej.skrabec@gmail.com>
+ */
+
+#include <linux/mmc/sdio_func.h>
+#include <linux/mmc/sdio_ids.h>
+#include <linux/module.h>
+#include "main.h"
+#include "rtw8822b.h"
+#include "sdio.h"
+
+static const struct sdio_device_id rtw_8822bs_id_table[] =  {
+	{
+		SDIO_DEVICE(SDIO_VENDOR_ID_REALTEK,
+			    SDIO_DEVICE_ID_REALTEK_RTW8822BS),
+		.driver_data = (kernel_ulong_t)&rtw8822b_hw_spec,
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(sdio, rtw_8822bs_id_table);
+
+static struct sdio_driver rtw_8822bs_driver = {
+	.name = "rtw_8822bs",
+	.probe = rtw_sdio_probe,
+	.remove = rtw_sdio_remove,
+	.id_table = rtw_8822bs_id_table,
+	.drv = {
+		.pm = &rtw_sdio_pm_ops,
+		.shutdown = rtw_sdio_shutdown,
+	}
+};
+module_sdio_driver(rtw_8822bs_driver);
+
+MODULE_AUTHOR("Jernej Skrabec <jernej.skrabec@gmail.com>");
+MODULE_DESCRIPTION("Realtek 802.11ac wireless 8822bs driver");
+MODULE_LICENSE("Dual BSD/GPL");
-- 
cgit v1.2.3-70-g09d2


From 6fdacb78f7999f5c14d9dae10d47de50959297d9 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:28 +0200
Subject: wifi: rtw88: Add support for the SDIO based RTL8822CS chipset

Wire up RTL8822CS chipset support using the new rtw88 SDIO HCI code as
well as the existing RTL8822C chipset code.

Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-9-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/Kconfig     | 11 ++++++++
 drivers/net/wireless/realtek/rtw88/Makefile    |  3 +++
 drivers/net/wireless/realtek/rtw88/rtw8822cs.c | 36 ++++++++++++++++++++++++++
 3 files changed, 50 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw88/rtw8822cs.c

diff --git a/drivers/net/wireless/realtek/rtw88/Kconfig b/drivers/net/wireless/realtek/rtw88/Kconfig
index 0cfc68dcc416..6b65da81127f 100644
--- a/drivers/net/wireless/realtek/rtw88/Kconfig
+++ b/drivers/net/wireless/realtek/rtw88/Kconfig
@@ -78,6 +78,17 @@ config RTW88_8822CE
 
 	  802.11ac PCIe wireless network adapter
 
+config RTW88_8822CS
+	tristate "Realtek 8822CS SDIO wireless network adapter"
+	depends on MMC
+	select RTW88_CORE
+	select RTW88_SDIO
+	select RTW88_8822C
+	help
+	  Select this option will enable support for 8822CS chipset
+
+	  802.11ac SDIO wireless network adapter
+
 config RTW88_8822CU
 	tristate "Realtek 8822CU USB wireless network adapter"
 	depends on USB
diff --git a/drivers/net/wireless/realtek/rtw88/Makefile b/drivers/net/wireless/realtek/rtw88/Makefile
index 2b8f4dd9707f..6105c2745bda 100644
--- a/drivers/net/wireless/realtek/rtw88/Makefile
+++ b/drivers/net/wireless/realtek/rtw88/Makefile
@@ -38,6 +38,9 @@ rtw88_8822c-objs		:= rtw8822c.o rtw8822c_table.o
 obj-$(CONFIG_RTW88_8822CE)	+= rtw88_8822ce.o
 rtw88_8822ce-objs		:= rtw8822ce.o
 
+obj-$(CONFIG_RTW88_8822CS)	+= rtw88_8822cs.o
+rtw88_8822cs-objs		:= rtw8822cs.o
+
 obj-$(CONFIG_RTW88_8822CU)	+= rtw88_8822cu.o
 rtw88_8822cu-objs		:= rtw8822cu.o
 
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822cs.c b/drivers/net/wireless/realtek/rtw88/rtw8822cs.c
new file mode 100644
index 000000000000..975e81c824f2
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822cs.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) Martin Blumenstingl <martin.blumenstingl@googlemail.com>
+ */
+
+#include <linux/mmc/sdio_func.h>
+#include <linux/mmc/sdio_ids.h>
+#include <linux/module.h>
+#include "main.h"
+#include "rtw8822c.h"
+#include "sdio.h"
+
+static const struct sdio_device_id rtw_8822cs_id_table[] =  {
+	{
+		SDIO_DEVICE(SDIO_VENDOR_ID_REALTEK,
+			    SDIO_DEVICE_ID_REALTEK_RTW8822CS),
+		.driver_data = (kernel_ulong_t)&rtw8822c_hw_spec,
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(sdio, rtw_8822cs_id_table);
+
+static struct sdio_driver rtw_8822cs_driver = {
+	.name = "rtw_8822cs",
+	.probe = rtw_sdio_probe,
+	.remove = rtw_sdio_remove,
+	.id_table = rtw_8822cs_id_table,
+	.drv = {
+		.pm = &rtw_sdio_pm_ops,
+		.shutdown = rtw_sdio_shutdown,
+	}
+};
+module_sdio_driver(rtw_8822cs_driver);
+
+MODULE_AUTHOR("Martin Blumenstingl <martin.blumenstingl@googlemail.com>");
+MODULE_DESCRIPTION("Realtek 802.11ac wireless 8822cs driver");
+MODULE_LICENSE("Dual BSD/GPL");
-- 
cgit v1.2.3-70-g09d2


From b2a777d68434375dc05a6fda5fec34a474bbf21f Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Wed, 5 Apr 2023 22:07:29 +0200
Subject: wifi: rtw88: Add support for the SDIO based RTL8821CS chipset

Wire up RTL8821CS chipset support using the new rtw88 SDIO HCI code as
well as the existing RTL8821C chipset code.

Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Tested-by: Chris Morgan <macromorgan@hotmail.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230405200729.632435-10-martin.blumenstingl@googlemail.com
---
 drivers/net/wireless/realtek/rtw88/Kconfig     | 11 ++++++++
 drivers/net/wireless/realtek/rtw88/Makefile    |  3 +++
 drivers/net/wireless/realtek/rtw88/rtw8821cs.c | 36 ++++++++++++++++++++++++++
 3 files changed, 50 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw88/rtw8821cs.c

diff --git a/drivers/net/wireless/realtek/rtw88/Kconfig b/drivers/net/wireless/realtek/rtw88/Kconfig
index 6b65da81127f..29eb2f8e0eb7 100644
--- a/drivers/net/wireless/realtek/rtw88/Kconfig
+++ b/drivers/net/wireless/realtek/rtw88/Kconfig
@@ -133,6 +133,17 @@ config RTW88_8821CE
 
 	  802.11ac PCIe wireless network adapter
 
+config RTW88_8821CS
+	tristate "Realtek 8821CS SDIO wireless network adapter"
+	depends on MMC
+	select RTW88_CORE
+	select RTW88_SDIO
+	select RTW88_8821C
+	help
+	  Select this option will enable support for 8821CS chipset
+
+	  802.11ac SDIO wireless network adapter
+
 config RTW88_8821CU
 	tristate "Realtek 8821CU USB wireless network adapter"
 	depends on USB
diff --git a/drivers/net/wireless/realtek/rtw88/Makefile b/drivers/net/wireless/realtek/rtw88/Makefile
index 6105c2745bda..82979b30ae8d 100644
--- a/drivers/net/wireless/realtek/rtw88/Makefile
+++ b/drivers/net/wireless/realtek/rtw88/Makefile
@@ -59,6 +59,9 @@ rtw88_8821c-objs		:= rtw8821c.o rtw8821c_table.o
 obj-$(CONFIG_RTW88_8821CE)	+= rtw88_8821ce.o
 rtw88_8821ce-objs		:= rtw8821ce.o
 
+obj-$(CONFIG_RTW88_8821CS)	+= rtw88_8821cs.o
+rtw88_8821cs-objs		:= rtw8821cs.o
+
 obj-$(CONFIG_RTW88_8821CU)	+= rtw88_8821cu.o
 rtw88_8821cu-objs		:= rtw8821cu.o
 
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821cs.c b/drivers/net/wireless/realtek/rtw88/rtw8821cs.c
new file mode 100644
index 000000000000..a359413369a4
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821cs.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) Martin Blumenstingl <martin.blumenstingl@googlemail.com>
+ */
+
+#include <linux/mmc/sdio_func.h>
+#include <linux/mmc/sdio_ids.h>
+#include <linux/module.h>
+#include "main.h"
+#include "rtw8821c.h"
+#include "sdio.h"
+
+static const struct sdio_device_id rtw_8821cs_id_table[] =  {
+	{
+		SDIO_DEVICE(SDIO_VENDOR_ID_REALTEK,
+			    SDIO_DEVICE_ID_REALTEK_RTW8821CS),
+		.driver_data = (kernel_ulong_t)&rtw8821c_hw_spec,
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(sdio, rtw_8821cs_id_table);
+
+static struct sdio_driver rtw_8821cs_driver = {
+	.name = "rtw_8821cs",
+	.probe = rtw_sdio_probe,
+	.remove = rtw_sdio_remove,
+	.id_table = rtw_8821cs_id_table,
+	.drv = {
+		.pm = &rtw_sdio_pm_ops,
+		.shutdown = rtw_sdio_shutdown,
+	}
+};
+module_sdio_driver(rtw_8821cs_driver);
+
+MODULE_AUTHOR("Martin Blumenstingl <martin.blumenstingl@googlemail.com>");
+MODULE_DESCRIPTION("Realtek 802.11ac wireless 8821cs driver");
+MODULE_LICENSE("Dual BSD/GPL");
-- 
cgit v1.2.3-70-g09d2


From 10fd5f70c397782a97f411f25bfb312ea92b55bc Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Wed, 12 Apr 2023 10:12:52 -0700
Subject: bpf: Handle NULL in bpf_local_storage_free.

During OOM bpf_local_storage_alloc() may fail to allocate 'storage' and
call to bpf_local_storage_free() with NULL pointer will cause a crash like:
[ 271718.917646] BUG: kernel NULL pointer dereference, address: 00000000000000a0
[ 271719.019620] RIP: 0010:call_rcu+0x2d/0x240
[ 271719.216274]  bpf_local_storage_alloc+0x19e/0x1e0
[ 271719.250121]  bpf_local_storage_update+0x33b/0x740

Fixes: 7e30a8477b0b ("bpf: Add bpf_local_storage_free()")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20230412171252.15635-1-alexei.starovoitov@gmail.com
---
 kernel/bpf/bpf_local_storage.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index dab2ff4c99d9..47d9948d768f 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -157,6 +157,9 @@ static void bpf_local_storage_free(struct bpf_local_storage *local_storage,
 				   struct bpf_local_storage_map *smap,
 				   bool bpf_ma, bool reuse_now)
 {
+	if (!local_storage)
+		return;
+
 	if (!bpf_ma) {
 		__bpf_local_storage_free(local_storage, reuse_now);
 		return;
-- 
cgit v1.2.3-70-g09d2


From 1d71283987c729dceccce834a864c27301ba155e Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Mon, 10 Apr 2023 23:16:31 -0500
Subject: bpf: Make bpf_cgroup_acquire() KF_RCU | KF_RET_NULL

struct cgroup is already an RCU-safe type in the verifier. We can
therefore update bpf_cgroup_acquire() to be KF_RCU | KF_RET_NULL, and
subsequently remove bpf_cgroup_kptr_get(). This patch does the first of
these by updating bpf_cgroup_acquire() to be KF_RCU | KF_RET_NULL, and
also updates selftests accordingly.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230411041633.179404-1-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c                               |  5 ++--
 .../selftests/bpf/progs/cgrp_kfunc_common.h        |  5 ++++
 .../selftests/bpf/progs/cgrp_kfunc_failure.c       | 35 ++++++++++++++++++----
 .../selftests/bpf/progs/cgrp_kfunc_success.c       |  5 +++-
 4 files changed, 40 insertions(+), 10 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index b6a5cda5bb59..71f0604bdc97 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2037,8 +2037,7 @@ __bpf_kfunc void bpf_task_release(struct task_struct *p)
  */
 __bpf_kfunc struct cgroup *bpf_cgroup_acquire(struct cgroup *cgrp)
 {
-	cgroup_get(cgrp);
-	return cgrp;
+	return cgroup_tryget(cgrp) ? cgrp : NULL;
 }
 
 /**
@@ -2314,7 +2313,7 @@ BTF_ID_FLAGS(func, bpf_rbtree_add)
 BTF_ID_FLAGS(func, bpf_rbtree_first, KF_RET_NULL)
 
 #ifdef CONFIG_CGROUPS
-BTF_ID_FLAGS(func, bpf_cgroup_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS)
+BTF_ID_FLAGS(func, bpf_cgroup_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_release, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_cgroup_ancestor, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
index d0b7cd0d09d7..b0e279f4652b 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
@@ -61,6 +61,11 @@ static inline int cgrps_kfunc_map_insert(struct cgroup *cgrp)
 	}
 
 	acquired = bpf_cgroup_acquire(cgrp);
+	if (!acquired) {
+		bpf_map_delete_elem(&__cgrps_kfunc_map, &id);
+		return -ENOENT;
+	}
+
 	old = bpf_kptr_xchg(&v->cgrp, acquired);
 	if (old) {
 		bpf_cgroup_release(old);
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
index 48b2034cadb3..49347f12de39 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
@@ -41,6 +41,23 @@ int BPF_PROG(cgrp_kfunc_acquire_untrusted, struct cgroup *cgrp, const char *path
 
 	/* Can't invoke bpf_cgroup_acquire() on an untrusted pointer. */
 	acquired = bpf_cgroup_acquire(v->cgrp);
+	if (acquired)
+		bpf_cgroup_release(acquired);
+
+	return 0;
+}
+
+SEC("tp_btf/cgroup_mkdir")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
+int BPF_PROG(cgrp_kfunc_acquire_no_null_check, struct cgroup *cgrp, const char *path)
+{
+	struct cgroup *acquired;
+
+	acquired = bpf_cgroup_acquire(cgrp);
+	/*
+	 * Can't invoke bpf_cgroup_release() without checking the return value
+	 * of bpf_cgroup_acquire().
+	 */
 	bpf_cgroup_release(acquired);
 
 	return 0;
@@ -54,7 +71,8 @@ int BPF_PROG(cgrp_kfunc_acquire_fp, struct cgroup *cgrp, const char *path)
 
 	/* Can't invoke bpf_cgroup_acquire() on a random frame pointer. */
 	acquired = bpf_cgroup_acquire((struct cgroup *)&stack_cgrp);
-	bpf_cgroup_release(acquired);
+	if (acquired)
+		bpf_cgroup_release(acquired);
 
 	return 0;
 }
@@ -67,7 +85,8 @@ int BPF_PROG(cgrp_kfunc_acquire_unsafe_kretprobe, struct cgroup *cgrp)
 
 	/* Can't acquire an untrusted struct cgroup * pointer. */
 	acquired = bpf_cgroup_acquire(cgrp);
-	bpf_cgroup_release(acquired);
+	if (acquired)
+		bpf_cgroup_release(acquired);
 
 	return 0;
 }
@@ -80,7 +99,8 @@ int BPF_PROG(cgrp_kfunc_acquire_trusted_walked, struct cgroup *cgrp, const char
 
 	/* Can't invoke bpf_cgroup_acquire() on a pointer obtained from walking a trusted cgroup. */
 	acquired = bpf_cgroup_acquire(cgrp->old_dom_cgrp);
-	bpf_cgroup_release(acquired);
+	if (acquired)
+		bpf_cgroup_release(acquired);
 
 	return 0;
 }
@@ -93,9 +113,8 @@ int BPF_PROG(cgrp_kfunc_acquire_null, struct cgroup *cgrp, const char *path)
 
 	/* Can't invoke bpf_cgroup_acquire() on a NULL pointer. */
 	acquired = bpf_cgroup_acquire(NULL);
-	if (!acquired)
-		return 0;
-	bpf_cgroup_release(acquired);
+	if (acquired)
+		bpf_cgroup_release(acquired);
 
 	return 0;
 }
@@ -137,6 +156,8 @@ int BPF_PROG(cgrp_kfunc_get_non_kptr_acquired, struct cgroup *cgrp, const char *
 	struct cgroup *kptr, *acquired;
 
 	acquired = bpf_cgroup_acquire(cgrp);
+	if (!acquired)
+		return 0;
 
 	/* Cannot use bpf_cgroup_kptr_get() on a non-map-value, even if the kptr was acquired. */
 	kptr = bpf_cgroup_kptr_get(&acquired);
@@ -256,6 +277,8 @@ int BPF_PROG(cgrp_kfunc_release_null, struct cgroup *cgrp, const char *path)
 		return -ENOENT;
 
 	acquired = bpf_cgroup_acquire(cgrp);
+	if (!acquired)
+		return -ENOENT;
 
 	old = bpf_kptr_xchg(&v->cgrp, acquired);
 
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
index 030aff700084..e9dbd1af05a7 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
@@ -38,7 +38,10 @@ int BPF_PROG(test_cgrp_acquire_release_argument, struct cgroup *cgrp, const char
 		return 0;
 
 	acquired = bpf_cgroup_acquire(cgrp);
-	bpf_cgroup_release(acquired);
+	if (!acquired)
+		err = 1;
+	else
+		bpf_cgroup_release(acquired);
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 6499fe6edc4fd5b91aed4d5cd84bd113e1c58d5f Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Mon, 10 Apr 2023 23:16:32 -0500
Subject: bpf: Remove bpf_cgroup_kptr_get() kfunc

Now that bpf_cgroup_acquire() is KF_RCU | KF_RET_NULL,
bpf_cgroup_kptr_get() is redundant. Let's remove it, and update
selftests to instead use bpf_cgroup_acquire() where appropriate. The
next patch will update the BPF documentation to not mention
bpf_cgroup_kptr_get().

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230411041633.179404-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c                               | 32 ----------
 .../selftests/bpf/progs/cgrp_kfunc_common.h        |  3 +-
 .../selftests/bpf/progs/cgrp_kfunc_failure.c       | 68 +++-------------------
 .../selftests/bpf/progs/cgrp_kfunc_success.c       | 10 ++--
 4 files changed, 14 insertions(+), 99 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 71f0604bdc97..f04e60a4847f 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2040,37 +2040,6 @@ __bpf_kfunc struct cgroup *bpf_cgroup_acquire(struct cgroup *cgrp)
 	return cgroup_tryget(cgrp) ? cgrp : NULL;
 }
 
-/**
- * bpf_cgroup_kptr_get - Acquire a reference on a struct cgroup kptr. A cgroup
- * kptr acquired by this kfunc which is not subsequently stored in a map, must
- * be released by calling bpf_cgroup_release().
- * @cgrpp: A pointer to a cgroup kptr on which a reference is being acquired.
- */
-__bpf_kfunc struct cgroup *bpf_cgroup_kptr_get(struct cgroup **cgrpp)
-{
-	struct cgroup *cgrp;
-
-	rcu_read_lock();
-	/* Another context could remove the cgroup from the map and release it
-	 * at any time, including after we've done the lookup above. This is
-	 * safe because we're in an RCU read region, so the cgroup is
-	 * guaranteed to remain valid until at least the rcu_read_unlock()
-	 * below.
-	 */
-	cgrp = READ_ONCE(*cgrpp);
-
-	if (cgrp && !cgroup_tryget(cgrp))
-		/* If the cgroup had been removed from the map and freed as
-		 * described above, cgroup_tryget() will return false. The
-		 * cgroup will be freed at some point after the current RCU gp
-		 * has ended, so just return NULL to the user.
-		 */
-		cgrp = NULL;
-	rcu_read_unlock();
-
-	return cgrp;
-}
-
 /**
  * bpf_cgroup_release - Release the reference acquired on a cgroup.
  * If this kfunc is invoked in an RCU read region, the cgroup is guaranteed to
@@ -2314,7 +2283,6 @@ BTF_ID_FLAGS(func, bpf_rbtree_first, KF_RET_NULL)
 
 #ifdef CONFIG_CGROUPS
 BTF_ID_FLAGS(func, bpf_cgroup_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
-BTF_ID_FLAGS(func, bpf_cgroup_kptr_get, KF_ACQUIRE | KF_KPTR_GET | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_release, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_cgroup_ancestor, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_cgroup_from_id, KF_ACQUIRE | KF_RET_NULL)
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
index b0e279f4652b..22914a70db54 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
@@ -21,10 +21,11 @@ struct hash_map {
 } __cgrps_kfunc_map SEC(".maps");
 
 struct cgroup *bpf_cgroup_acquire(struct cgroup *p) __ksym;
-struct cgroup *bpf_cgroup_kptr_get(struct cgroup **pp) __ksym;
 void bpf_cgroup_release(struct cgroup *p) __ksym;
 struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level) __ksym;
 struct cgroup *bpf_cgroup_from_id(u64 cgid) __ksym;
+void bpf_rcu_read_lock(void) __ksym;
+void bpf_rcu_read_unlock(void) __ksym;
 
 static inline struct __cgrps_kfunc_map_value *cgrps_kfunc_map_value_lookup(struct cgroup *cgrp)
 {
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
index 49347f12de39..0fa564a5cc5b 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c
@@ -133,59 +133,6 @@ int BPF_PROG(cgrp_kfunc_acquire_unreleased, struct cgroup *cgrp, const char *pat
 	return 0;
 }
 
-SEC("tp_btf/cgroup_mkdir")
-__failure __msg("arg#0 expected pointer to map value")
-int BPF_PROG(cgrp_kfunc_get_non_kptr_param, struct cgroup *cgrp, const char *path)
-{
-	struct cgroup *kptr;
-
-	/* Cannot use bpf_cgroup_kptr_get() on a non-kptr, even on a valid cgroup. */
-	kptr = bpf_cgroup_kptr_get(&cgrp);
-	if (!kptr)
-		return 0;
-
-	bpf_cgroup_release(kptr);
-
-	return 0;
-}
-
-SEC("tp_btf/cgroup_mkdir")
-__failure __msg("arg#0 expected pointer to map value")
-int BPF_PROG(cgrp_kfunc_get_non_kptr_acquired, struct cgroup *cgrp, const char *path)
-{
-	struct cgroup *kptr, *acquired;
-
-	acquired = bpf_cgroup_acquire(cgrp);
-	if (!acquired)
-		return 0;
-
-	/* Cannot use bpf_cgroup_kptr_get() on a non-map-value, even if the kptr was acquired. */
-	kptr = bpf_cgroup_kptr_get(&acquired);
-	bpf_cgroup_release(acquired);
-	if (!kptr)
-		return 0;
-
-	bpf_cgroup_release(kptr);
-
-	return 0;
-}
-
-SEC("tp_btf/cgroup_mkdir")
-__failure __msg("arg#0 expected pointer to map value")
-int BPF_PROG(cgrp_kfunc_get_null, struct cgroup *cgrp, const char *path)
-{
-	struct cgroup *kptr;
-
-	/* Cannot use bpf_cgroup_kptr_get() on a NULL pointer. */
-	kptr = bpf_cgroup_kptr_get(NULL);
-	if (!kptr)
-		return 0;
-
-	bpf_cgroup_release(kptr);
-
-	return 0;
-}
-
 SEC("tp_btf/cgroup_mkdir")
 __failure __msg("Unreleased reference")
 int BPF_PROG(cgrp_kfunc_xchg_unreleased, struct cgroup *cgrp, const char *path)
@@ -207,8 +154,8 @@ int BPF_PROG(cgrp_kfunc_xchg_unreleased, struct cgroup *cgrp, const char *path)
 }
 
 SEC("tp_btf/cgroup_mkdir")
-__failure __msg("Unreleased reference")
-int BPF_PROG(cgrp_kfunc_get_unreleased, struct cgroup *cgrp, const char *path)
+__failure __msg("must be referenced or trusted")
+int BPF_PROG(cgrp_kfunc_rcu_get_release, struct cgroup *cgrp, const char *path)
 {
 	struct cgroup *kptr;
 	struct __cgrps_kfunc_map_value *v;
@@ -217,11 +164,12 @@ int BPF_PROG(cgrp_kfunc_get_unreleased, struct cgroup *cgrp, const char *path)
 	if (!v)
 		return 0;
 
-	kptr = bpf_cgroup_kptr_get(&v->cgrp);
-	if (!kptr)
-		return 0;
-
-	/* Kptr acquired above is never released. */
+	bpf_rcu_read_lock();
+	kptr = v->cgrp;
+	if (kptr)
+		/* Can't release a cgroup kptr stored in a map. */
+		bpf_cgroup_release(kptr);
+	bpf_rcu_read_unlock();
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
index e9dbd1af05a7..5354455a01be 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_success.c
@@ -126,13 +126,11 @@ int BPF_PROG(test_cgrp_get_release, struct cgroup *cgrp, const char *path)
 		return 0;
 	}
 
-	kptr = bpf_cgroup_kptr_get(&v->cgrp);
-	if (!kptr) {
+	bpf_rcu_read_lock();
+	kptr = v->cgrp;
+	if (!kptr)
 		err = 3;
-		return 0;
-	}
-
-	bpf_cgroup_release(kptr);
+	bpf_rcu_read_unlock();
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From ec48599abee3a16fdc93c1c3c3e153a4f4d29420 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Mon, 10 Apr 2023 23:16:33 -0500
Subject: bpf,docs: Remove references to bpf_cgroup_kptr_get()

The bpf_cgroup_kptr_get() kfunc has been removed, and
bpf_cgroup_acquire() / bpf_cgroup_release() now have the same semantics
as bpf_task_acquire() / bpf_task_release(). This patch updates the BPF
documentation to reflect this.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230411041633.179404-3-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst | 68 --------------------------------------------
 1 file changed, 68 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index d8a16c4bef7f..3b42cfe12437 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -572,74 +572,6 @@ bpf_task_release() respectively, so we won't provide examples for them.
 
 ----
 
-You may also acquire a reference to a ``struct cgroup`` kptr that's already
-stored in a map using bpf_cgroup_kptr_get():
-
-.. kernel-doc:: kernel/bpf/helpers.c
-   :identifiers: bpf_cgroup_kptr_get
-
-Here's an example of how it can be used:
-
-.. code-block:: c
-
-	/* struct containing the struct task_struct kptr which is actually stored in the map. */
-	struct __cgroups_kfunc_map_value {
-		struct cgroup __kptr * cgroup;
-	};
-
-	/* The map containing struct __cgroups_kfunc_map_value entries. */
-	struct {
-		__uint(type, BPF_MAP_TYPE_HASH);
-		__type(key, int);
-		__type(value, struct __cgroups_kfunc_map_value);
-		__uint(max_entries, 1);
-	} __cgroups_kfunc_map SEC(".maps");
-
-	/* ... */
-
-	/**
-	 * A simple example tracepoint program showing how a
-	 * struct cgroup kptr that is stored in a map can
-	 * be acquired using the bpf_cgroup_kptr_get() kfunc.
-	 */
-	 SEC("tp_btf/cgroup_mkdir")
-	 int BPF_PROG(cgroup_kptr_get_example, struct cgroup *cgrp, const char *path)
-	 {
-		struct cgroup *kptr;
-		struct __cgroups_kfunc_map_value *v;
-		s32 id = cgrp->self.id;
-
-		/* Assume a cgroup kptr was previously stored in the map. */
-		v = bpf_map_lookup_elem(&__cgroups_kfunc_map, &id);
-		if (!v)
-			return -ENOENT;
-
-		/* Acquire a reference to the cgroup kptr that's already stored in the map. */
-		kptr = bpf_cgroup_kptr_get(&v->cgroup);
-		if (!kptr)
-			/* If no cgroup was present in the map, it's because
-			 * we're racing with another CPU that removed it with
-			 * bpf_kptr_xchg() between the bpf_map_lookup_elem()
-			 * above, and our call to bpf_cgroup_kptr_get().
-			 * bpf_cgroup_kptr_get() internally safely handles this
-			 * race, and will return NULL if the task is no longer
-			 * present in the map by the time we invoke the kfunc.
-			 */
-			return -EBUSY;
-
-		/* Free the reference we just took above. Note that the
-		 * original struct cgroup kptr is still in the map. It will
-		 * be freed either at a later time if another context deletes
-		 * it from the map, or automatically by the BPF subsystem if
-		 * it's still present when the map is destroyed.
-		 */
-		bpf_cgroup_release(kptr);
-
-		return 0;
-        }
-
-----
-
 Other kfuncs available for interacting with ``struct cgroup *`` objects are
 bpf_cgroup_ancestor() and bpf_cgroup_from_id(), allowing callers to access
 the ancestor of a cgroup and find a cgroup by its ID, respectively. Both
-- 
cgit v1.2.3-70-g09d2


From ed17aa92dc56b6d8883e4b7a8f1c6fbf5ed6cd29 Mon Sep 17 00:00:00 2001
From: Xin Liu <liuxin350@huawei.com>
Date: Thu, 6 Apr 2023 20:26:22 +0800
Subject: bpf, sockmap: fix deadlocks in the sockhash and sockmap

When huang uses sched_switch tracepoint, the tracepoint
does only one thing in the mounted ebpf program, which
deletes the fixed elements in sockhash ([0])

It seems that elements in sockhash are rarely actively
deleted by users or ebpf program. Therefore, we do not
pay much attention to their deletion. Compared with hash
maps, sockhash only provides spin_lock_bh protection.
This causes it to appear to have self-locking behavior
in the interrupt context.

  [0]:https://lore.kernel.org/all/CABcoxUayum5oOqFMMqAeWuS8+EzojquSOSyDA3J_2omY=2EeAg@mail.gmail.com/

Reported-by: Hsin-Wei Hung <hsinweih@uci.edu>
Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: Xin Liu <liuxin350@huawei.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/r/20230406122622.109978-1-liuxin350@huawei.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/core/sock_map.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index 7c189c2e2fbf..66305b7bf8b7 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -414,8 +414,9 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test,
 {
 	struct sock *sk;
 	int err = 0;
+	unsigned long flags;
 
-	raw_spin_lock_bh(&stab->lock);
+	raw_spin_lock_irqsave(&stab->lock, flags);
 	sk = *psk;
 	if (!sk_test || sk_test == sk)
 		sk = xchg(psk, NULL);
@@ -425,7 +426,7 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test,
 	else
 		err = -EINVAL;
 
-	raw_spin_unlock_bh(&stab->lock);
+	raw_spin_unlock_irqrestore(&stab->lock, flags);
 	return err;
 }
 
@@ -932,11 +933,12 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key)
 	struct bpf_shtab_bucket *bucket;
 	struct bpf_shtab_elem *elem;
 	int ret = -ENOENT;
+	unsigned long flags;
 
 	hash = sock_hash_bucket_hash(key, key_size);
 	bucket = sock_hash_select_bucket(htab, hash);
 
-	raw_spin_lock_bh(&bucket->lock);
+	raw_spin_lock_irqsave(&bucket->lock, flags);
 	elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size);
 	if (elem) {
 		hlist_del_rcu(&elem->node);
@@ -944,7 +946,7 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key)
 		sock_hash_free_elem(htab, elem);
 		ret = 0;
 	}
-	raw_spin_unlock_bh(&bucket->lock);
+	raw_spin_unlock_irqrestore(&bucket->lock, flags);
 	return ret;
 }
 
-- 
cgit v1.2.3-70-g09d2


From ac931d4cdec3df8b6eac3bc40a6871123021f078 Mon Sep 17 00:00:00 2001
From: Christian Ehrig <cehrig@cloudflare.com>
Date: Fri, 7 Apr 2023 15:38:53 +0200
Subject: ipip,ip_tunnel,sit: Add FOU support for externally controlled ipip
 devices

Today ipip devices in collect-metadata mode don't allow for sending FOU
or GUE encapsulated packets. This patch lifts the restriction by adding
a struct ip_tunnel_encap to the tunnel metadata.

On the egress path, the members of this struct can be set by the
bpf_skb_set_fou_encap kfunc via a BPF tc-hook. Instead of dropping packets
wishing to use additional UDP encapsulation, ip_md_tunnel_xmit now
evaluates the contents of this struct and adds the corresponding FOU or
GUE header. Furthermore, it is making sure that additional header bytes
are taken into account for PMTU discovery.

On the ingress path, an ipip device in collect-metadata mode will fill this
struct and a BPF tc-hook can obtain the information via a call to the
bpf_skb_get_fou_encap kfunc.

The minor change to ip_tunnel_encap, which now takes a pointer to
struct ip_tunnel_encap instead of struct ip_tunnel, allows us to control
FOU encap type and parameters on a per packet-level.

Signed-off-by: Christian Ehrig <cehrig@cloudflare.com>
Link: https://lore.kernel.org/r/cfea47de655d0f870248abf725932f851b53960a.1680874078.git.cehrig@cloudflare.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/net/ip_tunnels.h | 28 +++++++++++++++-------------
 net/ipv4/ip_tunnel.c     | 22 ++++++++++++++++++++--
 net/ipv4/ipip.c          |  1 +
 net/ipv6/sit.c           |  2 +-
 4 files changed, 37 insertions(+), 16 deletions(-)

diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index fca357679816..7912f53caae0 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -57,6 +57,13 @@ struct ip_tunnel_key {
 	__u8			flow_flags;
 };
 
+struct ip_tunnel_encap {
+	u16			type;
+	u16			flags;
+	__be16			sport;
+	__be16			dport;
+};
+
 /* Flags for ip_tunnel_info mode. */
 #define IP_TUNNEL_INFO_TX	0x01	/* represents tx tunnel parameters */
 #define IP_TUNNEL_INFO_IPV6	0x02	/* key contains IPv6 addresses */
@@ -66,9 +73,9 @@ struct ip_tunnel_key {
 #define IP_TUNNEL_OPTS_MAX					\
 	GENMASK((sizeof_field(struct ip_tunnel_info,		\
 			      options_len) * BITS_PER_BYTE) - 1, 0)
-
 struct ip_tunnel_info {
 	struct ip_tunnel_key	key;
+	struct ip_tunnel_encap	encap;
 #ifdef CONFIG_DST_CACHE
 	struct dst_cache	dst_cache;
 #endif
@@ -86,13 +93,6 @@ struct ip_tunnel_6rd_parm {
 };
 #endif
 
-struct ip_tunnel_encap {
-	u16			type;
-	u16			flags;
-	__be16			sport;
-	__be16			dport;
-};
-
 struct ip_tunnel_prl_entry {
 	struct ip_tunnel_prl_entry __rcu *next;
 	__be32				addr;
@@ -293,6 +293,7 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn,
 				   __be32 remote, __be32 local,
 				   __be32 key);
 
+void ip_tunnel_md_udp_encap(struct sk_buff *skb, struct ip_tunnel_info *info);
 int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 		  const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst,
 		  bool log_ecn_error);
@@ -371,22 +372,23 @@ static inline int ip_encap_hlen(struct ip_tunnel_encap *e)
 	return hlen;
 }
 
-static inline int ip_tunnel_encap(struct sk_buff *skb, struct ip_tunnel *t,
+static inline int ip_tunnel_encap(struct sk_buff *skb,
+				  struct ip_tunnel_encap *e,
 				  u8 *protocol, struct flowi4 *fl4)
 {
 	const struct ip_tunnel_encap_ops *ops;
 	int ret = -EINVAL;
 
-	if (t->encap.type == TUNNEL_ENCAP_NONE)
+	if (e->type == TUNNEL_ENCAP_NONE)
 		return 0;
 
-	if (t->encap.type >= MAX_IPTUN_ENCAP_OPS)
+	if (e->type >= MAX_IPTUN_ENCAP_OPS)
 		return -EINVAL;
 
 	rcu_read_lock();
-	ops = rcu_dereference(iptun_encaps[t->encap.type]);
+	ops = rcu_dereference(iptun_encaps[e->type]);
 	if (likely(ops && ops->build_header))
-		ret = ops->build_header(skb, &t->encap, protocol, fl4);
+		ret = ops->build_header(skb, e, protocol, fl4);
 	rcu_read_unlock();
 
 	return ret;
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index de90b09dfe78..add437f710fc 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -359,6 +359,20 @@ err_dev_set_mtu:
 	return ERR_PTR(err);
 }
 
+void ip_tunnel_md_udp_encap(struct sk_buff *skb, struct ip_tunnel_info *info)
+{
+	const struct iphdr *iph = ip_hdr(skb);
+	const struct udphdr *udph;
+
+	if (iph->protocol != IPPROTO_UDP)
+		return;
+
+	udph = (struct udphdr *)((__u8 *)iph + (iph->ihl << 2));
+	info->encap.sport = udph->source;
+	info->encap.dport = udph->dest;
+}
+EXPORT_SYMBOL(ip_tunnel_md_udp_encap);
+
 int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 		  const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst,
 		  bool log_ecn_error)
@@ -572,7 +586,11 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 			    tunnel_id_to_key32(key->tun_id), RT_TOS(tos),
 			    dev_net(dev), 0, skb->mark, skb_get_hash(skb),
 			    key->flow_flags);
-	if (tunnel->encap.type != TUNNEL_ENCAP_NONE)
+
+	if (!tunnel_hlen)
+		tunnel_hlen = ip_encap_hlen(&tun_info->encap);
+
+	if (ip_tunnel_encap(skb, &tun_info->encap, &proto, &fl4) < 0)
 		goto tx_error;
 
 	use_cache = ip_tunnel_dst_cache_usable(skb, tun_info);
@@ -732,7 +750,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 			    dev_net(dev), tunnel->parms.link,
 			    tunnel->fwmark, skb_get_hash(skb), 0);
 
-	if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0)
+	if (ip_tunnel_encap(skb, &tunnel->encap, &protocol, &fl4) < 0)
 		goto tx_error;
 
 	if (connected && md) {
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index abea77759b7e..27b8f83c6ea2 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -241,6 +241,7 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto)
 			tun_dst = ip_tun_rx_dst(skb, 0, 0, 0);
 			if (!tun_dst)
 				return 0;
+			ip_tunnel_md_udp_encap(skb, &tun_dst->u.tun_info);
 		}
 		skb_reset_mac_header(skb);
 
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 70d81bba5093..063560e2cb1a 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -1024,7 +1024,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb,
 		ttl = iph6->hop_limit;
 	tos = INET_ECN_encapsulate(tos, ipv6_get_dsfield(iph6));
 
-	if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0) {
+	if (ip_tunnel_encap(skb, &tunnel->encap, &protocol, &fl4) < 0) {
 		ip_rt_put(rt);
 		goto tx_error;
 	}
-- 
cgit v1.2.3-70-g09d2


From c50e96099edb134bf107fafc02715fbc4aa2277f Mon Sep 17 00:00:00 2001
From: Christian Ehrig <cehrig@cloudflare.com>
Date: Fri, 7 Apr 2023 15:38:54 +0200
Subject: bpf,fou: Add bpf_skb_{set,get}_fou_encap kfuncs

Add two new kfuncs that allow a BPF tc-hook, installed on an ipip
device in collect-metadata mode, to control FOU encap parameters on a
per-packet level. The set of kfuncs is registered with the fou module.

The bpf_skb_set_fou_encap kfunc is supposed to be used in tandem and after
a successful call to the bpf_skb_set_tunnel_key bpf-helper. UDP source and
destination ports can be controlled by passing a struct bpf_fou_encap. A
source port of zero will auto-assign a source port. enum bpf_fou_encap_type
is used to specify if the egress path should FOU or GUE encap the packet.

On the ingress path bpf_skb_get_fou_encap can be used to read UDP source
and destination ports from the receiver's point of view and allows for
packet multiplexing across different destination ports within a single
BPF program and ipip device.

Signed-off-by: Christian Ehrig <cehrig@cloudflare.com>
Link: https://lore.kernel.org/r/e17c94a646b63e78ce0dbf3f04b2c33dc948a32d.1680874078.git.cehrig@cloudflare.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/net/fou.h   |   2 +
 net/ipv4/Makefile   |   2 +-
 net/ipv4/fou_bpf.c  | 119 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/fou_core.c |   5 +++
 4 files changed, 127 insertions(+), 1 deletion(-)
 create mode 100644 net/ipv4/fou_bpf.c

diff --git a/include/net/fou.h b/include/net/fou.h
index 80f56e275b08..824eb4b231fd 100644
--- a/include/net/fou.h
+++ b/include/net/fou.h
@@ -17,4 +17,6 @@ int __fou_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 int __gue_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 		       u8 *protocol, __be16 *sport, int type);
 
+int register_fou_bpf(void);
+
 #endif
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 880277c9fd07..b18ba8ef93ad 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -26,7 +26,7 @@ obj-$(CONFIG_IP_MROUTE) += ipmr.o
 obj-$(CONFIG_IP_MROUTE_COMMON) += ipmr_base.o
 obj-$(CONFIG_NET_IPIP) += ipip.o
 gre-y := gre_demux.o
-fou-y := fou_core.o fou_nl.o
+fou-y := fou_core.o fou_nl.o fou_bpf.o
 obj-$(CONFIG_NET_FOU) += fou.o
 obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o
 obj-$(CONFIG_NET_IPGRE) += ip_gre.o
diff --git a/net/ipv4/fou_bpf.c b/net/ipv4/fou_bpf.c
new file mode 100644
index 000000000000..3760a14b6b57
--- /dev/null
+++ b/net/ipv4/fou_bpf.c
@@ -0,0 +1,119 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Unstable Fou Helpers for TC-BPF hook
+ *
+ * These are called from SCHED_CLS BPF programs. Note that it is
+ * allowed to break compatibility for these functions since the interface they
+ * are exposed through to BPF programs is explicitly unstable.
+ */
+
+#include <linux/bpf.h>
+#include <linux/btf_ids.h>
+
+#include <net/dst_metadata.h>
+#include <net/fou.h>
+
+struct bpf_fou_encap {
+	__be16 sport;
+	__be16 dport;
+};
+
+enum bpf_fou_encap_type {
+	FOU_BPF_ENCAP_FOU,
+	FOU_BPF_ENCAP_GUE,
+};
+
+__diag_push();
+__diag_ignore_all("-Wmissing-prototypes",
+		  "Global functions as their definitions will be in BTF");
+
+/* bpf_skb_set_fou_encap - Set FOU encap parameters
+ *
+ * This function allows for using GUE or FOU encapsulation together with an
+ * ipip device in collect-metadata mode.
+ *
+ * It is meant to be used in BPF tc-hooks and after a call to the
+ * bpf_skb_set_tunnel_key helper, responsible for setting IP addresses.
+ *
+ * Parameters:
+ * @skb_ctx	Pointer to ctx (__sk_buff) in TC program. Cannot be NULL
+ * @encap	Pointer to a `struct bpf_fou_encap` storing UDP src and
+ * 		dst ports. If sport is set to 0 the kernel will auto-assign a
+ * 		port. This is similar to using `encap-sport auto`.
+ * 		Cannot be NULL
+ * @type	Encapsulation type for the packet. Their definitions are
+ * 		specified in `enum bpf_fou_encap_type`
+ */
+__bpf_kfunc int bpf_skb_set_fou_encap(struct __sk_buff *skb_ctx,
+				      struct bpf_fou_encap *encap, int type)
+{
+	struct sk_buff *skb = (struct sk_buff *)skb_ctx;
+	struct ip_tunnel_info *info = skb_tunnel_info(skb);
+
+	if (unlikely(!encap))
+		return -EINVAL;
+
+	if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX)))
+		return -EINVAL;
+
+	switch (type) {
+	case FOU_BPF_ENCAP_FOU:
+		info->encap.type = TUNNEL_ENCAP_FOU;
+		break;
+	case FOU_BPF_ENCAP_GUE:
+		info->encap.type = TUNNEL_ENCAP_GUE;
+		break;
+	default:
+		info->encap.type = TUNNEL_ENCAP_NONE;
+	}
+
+	if (info->key.tun_flags & TUNNEL_CSUM)
+		info->encap.flags |= TUNNEL_ENCAP_FLAG_CSUM;
+
+	info->encap.sport = encap->sport;
+	info->encap.dport = encap->dport;
+
+	return 0;
+}
+
+/* bpf_skb_get_fou_encap - Get FOU encap parameters
+ *
+ * This function allows for reading encap metadata from a packet received
+ * on an ipip device in collect-metadata mode.
+ *
+ * Parameters:
+ * @skb_ctx	Pointer to ctx (__sk_buff) in TC program. Cannot be NULL
+ * @encap	Pointer to a struct bpf_fou_encap storing UDP source and
+ * 		destination port. Cannot be NULL
+ */
+__bpf_kfunc int bpf_skb_get_fou_encap(struct __sk_buff *skb_ctx,
+				      struct bpf_fou_encap *encap)
+{
+	struct sk_buff *skb = (struct sk_buff *)skb_ctx;
+	struct ip_tunnel_info *info = skb_tunnel_info(skb);
+
+	if (unlikely(!info))
+		return -EINVAL;
+
+	encap->sport = info->encap.sport;
+	encap->dport = info->encap.dport;
+
+	return 0;
+}
+
+__diag_pop()
+
+BTF_SET8_START(fou_kfunc_set)
+BTF_ID_FLAGS(func, bpf_skb_set_fou_encap)
+BTF_ID_FLAGS(func, bpf_skb_get_fou_encap)
+BTF_SET8_END(fou_kfunc_set)
+
+static const struct btf_kfunc_id_set fou_bpf_kfunc_set = {
+	.owner = THIS_MODULE,
+	.set   = &fou_kfunc_set,
+};
+
+int register_fou_bpf(void)
+{
+	return register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS,
+					 &fou_bpf_kfunc_set);
+}
diff --git a/net/ipv4/fou_core.c b/net/ipv4/fou_core.c
index cafec9b4eee0..0c41076e31ed 100644
--- a/net/ipv4/fou_core.c
+++ b/net/ipv4/fou_core.c
@@ -1236,10 +1236,15 @@ static int __init fou_init(void)
 	if (ret < 0)
 		goto unregister;
 
+	ret = register_fou_bpf();
+	if (ret < 0)
+		goto kfunc_failed;
+
 	ret = ip_tunnel_encap_add_fou_ops();
 	if (ret == 0)
 		return 0;
 
+kfunc_failed:
 	genl_unregister_family(&fou_nl_family);
 unregister:
 	unregister_pernet_device(&fou_net_ops);
-- 
cgit v1.2.3-70-g09d2


From d9688f898c08c8f96fb0e7879262877ffd319bfd Mon Sep 17 00:00:00 2001
From: Christian Ehrig <cehrig@cloudflare.com>
Date: Fri, 7 Apr 2023 15:38:55 +0200
Subject: selftests/bpf: Test FOU kfuncs for externally controlled ipip devices

Add tests for FOU and GUE encapsulation via the bpf_skb_{set,get}_fou_encap
kfuncs, using ipip devices in collect-metadata mode.

These tests make sure that we can successfully set and obtain FOU and GUE
encap parameters using ingress / egress BPF tc-hooks.

Signed-off-by: Christian Ehrig <cehrig@cloudflare.com>
Link: https://lore.kernel.org/r/040193566ddbdb0b53eb359f7ac7bbd316f338b5.1680874078.git.cehrig@cloudflare.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/test_tunnel.c | 153 ++++++++++++++++++++-
 .../testing/selftests/bpf/progs/test_tunnel_kern.c | 117 ++++++++++++++++
 2 files changed, 268 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_tunnel.c b/tools/testing/selftests/bpf/prog_tests/test_tunnel.c
index 47f1d482fe39..d149ab98798d 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_tunnel.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_tunnel.c
@@ -89,6 +89,9 @@
 #define IP6VXLAN_TUNL_DEV0 "ip6vxlan00"
 #define IP6VXLAN_TUNL_DEV1 "ip6vxlan11"
 
+#define IPIP_TUNL_DEV0 "ipip00"
+#define IPIP_TUNL_DEV1 "ipip11"
+
 #define PING_ARGS "-i 0.01 -c 3 -w 10 -q"
 
 static int config_device(void)
@@ -188,6 +191,79 @@ static void delete_ip6vxlan_tunnel(void)
 	SYS_NOFAIL("ip link delete dev %s", IP6VXLAN_TUNL_DEV1);
 }
 
+enum ipip_encap {
+	NONE	= 0,
+	FOU	= 1,
+	GUE	= 2,
+};
+
+static int set_ipip_encap(const char *ipproto, const char *type)
+{
+	SYS(fail, "ip -n at_ns0 fou add port 5555 %s", ipproto);
+	SYS(fail, "ip -n at_ns0 link set dev %s type ipip encap %s",
+	    IPIP_TUNL_DEV0, type);
+	SYS(fail, "ip -n at_ns0 link set dev %s type ipip encap-dport 5555",
+	    IPIP_TUNL_DEV0);
+
+	return 0;
+fail:
+	return -1;
+}
+
+static int add_ipip_tunnel(enum ipip_encap encap)
+{
+	int err;
+	const char *ipproto, *type;
+
+	switch (encap) {
+	case FOU:
+		ipproto = "ipproto 4";
+		type = "fou";
+		break;
+	case GUE:
+		ipproto = "gue";
+		type = ipproto;
+		break;
+	default:
+		ipproto = NULL;
+		type = ipproto;
+	}
+
+	/* at_ns0 namespace */
+	SYS(fail, "ip -n at_ns0 link add dev %s type ipip local %s remote %s",
+	    IPIP_TUNL_DEV0, IP4_ADDR_VETH0, IP4_ADDR1_VETH1);
+
+	if (type && ipproto) {
+		err = set_ipip_encap(ipproto, type);
+		if (!ASSERT_OK(err, "set_ipip_encap"))
+			goto fail;
+	}
+
+	SYS(fail, "ip -n at_ns0 link set dev %s up", IPIP_TUNL_DEV0);
+	SYS(fail, "ip -n at_ns0 addr add dev %s %s/24",
+	    IPIP_TUNL_DEV0, IP4_ADDR_TUNL_DEV0);
+
+	/* root namespace */
+	if (type && ipproto)
+		SYS(fail, "ip fou add port 5555 %s", ipproto);
+	SYS(fail, "ip link add dev %s type ipip external", IPIP_TUNL_DEV1);
+	SYS(fail, "ip link set dev %s up", IPIP_TUNL_DEV1);
+	SYS(fail, "ip addr add dev %s %s/24", IPIP_TUNL_DEV1,
+	    IP4_ADDR_TUNL_DEV1);
+
+	return 0;
+fail:
+	return -1;
+}
+
+static void delete_ipip_tunnel(void)
+{
+	SYS_NOFAIL("ip -n at_ns0 link delete dev %s", IPIP_TUNL_DEV0);
+	SYS_NOFAIL("ip -n at_ns0 fou del port 5555 2> /dev/null");
+	SYS_NOFAIL("ip link delete dev %s", IPIP_TUNL_DEV1);
+	SYS_NOFAIL("ip fou del port 5555 2> /dev/null");
+}
+
 static int test_ping(int family, const char *addr)
 {
 	SYS(fail, "%s %s %s > /dev/null", ping_command(family), PING_ARGS, addr);
@@ -386,10 +462,80 @@ done:
 		test_tunnel_kern__destroy(skel);
 }
 
-#define RUN_TEST(name)							\
+static void test_ipip_tunnel(enum ipip_encap encap)
+{
+	struct test_tunnel_kern *skel = NULL;
+	struct nstoken *nstoken;
+	int set_src_prog_fd, get_src_prog_fd;
+	int ifindex = -1;
+	int err;
+	DECLARE_LIBBPF_OPTS(bpf_tc_hook, tc_hook,
+			    .attach_point = BPF_TC_INGRESS);
+
+	/* add ipip tunnel */
+	err = add_ipip_tunnel(encap);
+	if (!ASSERT_OK(err, "add_ipip_tunnel"))
+		goto done;
+
+	/* load and attach bpf prog to tunnel dev tc hook point */
+	skel = test_tunnel_kern__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "test_tunnel_kern__open_and_load"))
+		goto done;
+	ifindex = if_nametoindex(IPIP_TUNL_DEV1);
+	if (!ASSERT_NEQ(ifindex, 0, "ipip11 ifindex"))
+		goto done;
+	tc_hook.ifindex = ifindex;
+
+	switch (encap) {
+	case FOU:
+		get_src_prog_fd = bpf_program__fd(
+			skel->progs.ipip_encap_get_tunnel);
+		set_src_prog_fd = bpf_program__fd(
+			skel->progs.ipip_fou_set_tunnel);
+		break;
+	case GUE:
+		get_src_prog_fd = bpf_program__fd(
+			skel->progs.ipip_encap_get_tunnel);
+		set_src_prog_fd = bpf_program__fd(
+			skel->progs.ipip_gue_set_tunnel);
+		break;
+	default:
+		get_src_prog_fd = bpf_program__fd(
+			skel->progs.ipip_get_tunnel);
+		set_src_prog_fd = bpf_program__fd(
+			skel->progs.ipip_set_tunnel);
+	}
+
+	if (!ASSERT_GE(set_src_prog_fd, 0, "bpf_program__fd"))
+		goto done;
+	if (!ASSERT_GE(get_src_prog_fd, 0, "bpf_program__fd"))
+		goto done;
+	if (attach_tc_prog(&tc_hook, get_src_prog_fd, set_src_prog_fd))
+		goto done;
+
+	/* ping from root namespace test */
+	err = test_ping(AF_INET, IP4_ADDR_TUNL_DEV0);
+	if (!ASSERT_OK(err, "test_ping"))
+		goto done;
+
+	/* ping from at_ns0 namespace test */
+	nstoken = open_netns("at_ns0");
+	err = test_ping(AF_INET, IP4_ADDR_TUNL_DEV1);
+	if (!ASSERT_OK(err, "test_ping"))
+		goto done;
+	close_netns(nstoken);
+
+done:
+	/* delete ipip tunnel */
+	delete_ipip_tunnel();
+	if (skel)
+		test_tunnel_kern__destroy(skel);
+}
+
+#define RUN_TEST(name, ...)						\
 	({								\
 		if (test__start_subtest(#name)) {			\
-			test_ ## name();				\
+			test_ ## name(__VA_ARGS__);			\
 		}							\
 	})
 
@@ -400,6 +546,9 @@ static void *test_tunnel_run_tests(void *arg)
 
 	RUN_TEST(vxlan_tunnel);
 	RUN_TEST(ip6vxlan_tunnel);
+	RUN_TEST(ipip_tunnel, NONE);
+	RUN_TEST(ipip_tunnel, FOU);
+	RUN_TEST(ipip_tunnel, GUE);
 
 	cleanup();
 
diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
index 9ab2d55ab7c0..f66af753bbbb 100644
--- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
@@ -52,6 +52,21 @@ struct vxlan_metadata {
 	__u32     gbp;
 };
 
+struct bpf_fou_encap {
+	__be16 sport;
+	__be16 dport;
+};
+
+enum bpf_fou_encap_type {
+	FOU_BPF_ENCAP_FOU,
+	FOU_BPF_ENCAP_GUE,
+};
+
+int bpf_skb_set_fou_encap(struct __sk_buff *skb_ctx,
+			  struct bpf_fou_encap *encap, int type) __ksym;
+int bpf_skb_get_fou_encap(struct __sk_buff *skb_ctx,
+			  struct bpf_fou_encap *encap) __ksym;
+
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
 	__uint(max_entries, 1);
@@ -749,6 +764,108 @@ int ipip_get_tunnel(struct __sk_buff *skb)
 	return TC_ACT_OK;
 }
 
+SEC("tc")
+int ipip_gue_set_tunnel(struct __sk_buff *skb)
+{
+	struct bpf_tunnel_key key = {};
+	struct bpf_fou_encap encap = {};
+	void *data = (void *)(long)skb->data;
+	struct iphdr *iph = data;
+	void *data_end = (void *)(long)skb->data_end;
+	int ret;
+
+	if (data + sizeof(*iph) > data_end) {
+		log_err(1);
+		return TC_ACT_SHOT;
+	}
+
+	key.tunnel_ttl = 64;
+	if (iph->protocol == IPPROTO_ICMP)
+		key.remote_ipv4 = 0xac100164; /* 172.16.1.100 */
+
+	ret = bpf_skb_set_tunnel_key(skb, &key, sizeof(key), 0);
+	if (ret < 0) {
+		log_err(ret);
+		return TC_ACT_SHOT;
+	}
+
+	encap.sport = 0;
+	encap.dport = bpf_htons(5555);
+
+	ret = bpf_skb_set_fou_encap(skb, &encap, FOU_BPF_ENCAP_GUE);
+	if (ret < 0) {
+		log_err(ret);
+		return TC_ACT_SHOT;
+	}
+
+	return TC_ACT_OK;
+}
+
+SEC("tc")
+int ipip_fou_set_tunnel(struct __sk_buff *skb)
+{
+	struct bpf_tunnel_key key = {};
+	struct bpf_fou_encap encap = {};
+	void *data = (void *)(long)skb->data;
+	struct iphdr *iph = data;
+	void *data_end = (void *)(long)skb->data_end;
+	int ret;
+
+	if (data + sizeof(*iph) > data_end) {
+		log_err(1);
+		return TC_ACT_SHOT;
+	}
+
+	key.tunnel_ttl = 64;
+	if (iph->protocol == IPPROTO_ICMP)
+		key.remote_ipv4 = 0xac100164; /* 172.16.1.100 */
+
+	ret = bpf_skb_set_tunnel_key(skb, &key, sizeof(key), 0);
+	if (ret < 0) {
+		log_err(ret);
+		return TC_ACT_SHOT;
+	}
+
+	encap.sport = 0;
+	encap.dport = bpf_htons(5555);
+
+	ret = bpf_skb_set_fou_encap(skb, &encap, FOU_BPF_ENCAP_FOU);
+	if (ret < 0) {
+		log_err(ret);
+		return TC_ACT_SHOT;
+	}
+
+	return TC_ACT_OK;
+}
+
+SEC("tc")
+int ipip_encap_get_tunnel(struct __sk_buff *skb)
+{
+	int ret;
+	struct bpf_tunnel_key key = {};
+	struct bpf_fou_encap encap = {};
+
+	ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key), 0);
+	if (ret < 0) {
+		log_err(ret);
+		return TC_ACT_SHOT;
+	}
+
+	ret = bpf_skb_get_fou_encap(skb, &encap);
+	if (ret < 0) {
+		log_err(ret);
+		return TC_ACT_SHOT;
+	}
+
+	if (bpf_ntohs(encap.dport) != 5555)
+		return TC_ACT_SHOT;
+
+	bpf_printk("%d remote ip 0x%x, sport %d, dport %d\n", ret,
+		   key.remote_ipv4, bpf_ntohs(encap.sport),
+		   bpf_ntohs(encap.dport));
+	return TC_ACT_OK;
+}
+
 SEC("tc")
 int ipip6_set_tunnel(struct __sk_buff *skb)
 {
-- 
cgit v1.2.3-70-g09d2


From d319f344561de23e810515d109c7278919bff7b0 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Mon, 10 Apr 2023 19:43:44 +0200
Subject: mm: Fix copy_from_user_nofault().

There are several issues with copy_from_user_nofault():

- access_ok() is designed for user context only and for that reason
it has WARN_ON_IN_IRQ() which triggers when bpf, kprobe, eprobe
and perf on ppc are calling it from irq.

- it's missing nmi_uaccess_okay() which is a nop on all architectures
except x86 where it's required.
The comment in arch/x86/mm/tlb.c explains the details why it's necessary.
Calling copy_from_user_nofault() from bpf, [ke]probe without this check is not safe.

- __copy_from_user_inatomic() under CONFIG_HARDENED_USERCOPY is calling
check_object_size()->__check_object_size()->check_heap_object()->find_vmap_area()->spin_lock()
which is not safe to do from bpf, [ke]probe and perf due to potential deadlock.

Fix all three issues. At the end the copy_from_user_nofault() becomes
equivalent to copy_from_user_nmi() from safety point of view with
a difference in the return value.

Reported-by: Hsin-Wei Hung <hsinweih@uci.edu>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Florian Lehner <dev@der-flo.net>
Tested-by: Hsin-Wei Hung <hsinweih@uci.edu>
Tested-by: Florian Lehner <dev@der-flo.net>
Link: https://lore.kernel.org/r/20230410174345.4376-2-dev@der-flo.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 mm/maccess.c  | 16 +++++++++++-----
 mm/usercopy.c |  2 +-
 2 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/mm/maccess.c b/mm/maccess.c
index 074f6b086671..518a25667323 100644
--- a/mm/maccess.c
+++ b/mm/maccess.c
@@ -5,6 +5,7 @@
 #include <linux/export.h>
 #include <linux/mm.h>
 #include <linux/uaccess.h>
+#include <asm/tlb.h>
 
 bool __weak copy_from_kernel_nofault_allowed(const void *unsafe_src,
 		size_t size)
@@ -113,11 +114,16 @@ Efault:
 long copy_from_user_nofault(void *dst, const void __user *src, size_t size)
 {
 	long ret = -EFAULT;
-	if (access_ok(src, size)) {
-		pagefault_disable();
-		ret = __copy_from_user_inatomic(dst, src, size);
-		pagefault_enable();
-	}
+
+	if (!__access_ok(src, size))
+		return ret;
+
+	if (!nmi_uaccess_okay())
+		return ret;
+
+	pagefault_disable();
+	ret = __copy_from_user_inatomic(dst, src, size);
+	pagefault_enable();
 
 	if (ret)
 		return -EFAULT;
diff --git a/mm/usercopy.c b/mm/usercopy.c
index 4c3164beacec..83c164aba6e0 100644
--- a/mm/usercopy.c
+++ b/mm/usercopy.c
@@ -173,7 +173,7 @@ static inline void check_heap_object(const void *ptr, unsigned long n,
 		return;
 	}
 
-	if (is_vmalloc_addr(ptr)) {
+	if (is_vmalloc_addr(ptr) && !pagefault_disabled()) {
 		struct vmap_area *area = find_vmap_area(addr);
 
 		if (!area)
-- 
cgit v1.2.3-70-g09d2


From 37f9b2a6c086bb28487a0682b8098f907861c4a1 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sun, 9 Apr 2023 17:02:04 +0200
Subject: net: ethernet: Add missing depends on MDIO_DEVRES

A number of MDIO drivers make use of devm_mdiobus_alloc_size(). This
is only available when CONFIG_MDIO_DEVRES is enabled. Add missing
depends or selects, depending on if there are circular dependencies or
not. This avoids linker errors, especially for randconfig builds.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20230409150204.2346231-1-andrew@lunn.ch
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/Kconfig       | 1 +
 drivers/net/ethernet/freescale/enetc/Kconfig | 1 +
 drivers/net/ethernet/marvell/Kconfig         | 1 +
 drivers/net/ethernet/qualcomm/Kconfig        | 1 +
 drivers/net/mdio/Kconfig                     | 3 +++
 5 files changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/freescale/Kconfig b/drivers/net/ethernet/freescale/Kconfig
index f1e80d6996ef..1c78f66a89da 100644
--- a/drivers/net/ethernet/freescale/Kconfig
+++ b/drivers/net/ethernet/freescale/Kconfig
@@ -71,6 +71,7 @@ config FSL_XGMAC_MDIO
 	tristate "Freescale XGMAC MDIO"
 	select PHYLIB
 	depends on OF
+	select MDIO_DEVRES
 	select OF_MDIO
 	help
 	  This driver supports the MDIO bus on the Fman 10G Ethernet MACs, and
diff --git a/drivers/net/ethernet/freescale/enetc/Kconfig b/drivers/net/ethernet/freescale/enetc/Kconfig
index 9bc099cf3cb1..4d75e6807e92 100644
--- a/drivers/net/ethernet/freescale/enetc/Kconfig
+++ b/drivers/net/ethernet/freescale/enetc/Kconfig
@@ -10,6 +10,7 @@ config FSL_ENETC_CORE
 config FSL_ENETC
 	tristate "ENETC PF driver"
 	depends on PCI_MSI
+	select MDIO_DEVRES
 	select FSL_ENETC_CORE
 	select FSL_ENETC_IERB
 	select FSL_ENETC_MDIO
diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index f58a1c0144ba..884d64114bff 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -34,6 +34,7 @@ config MV643XX_ETH
 config MVMDIO
 	tristate "Marvell MDIO interface support"
 	depends on HAS_IOMEM
+	select MDIO_DEVRES
 	select PHYLIB
 	help
 	  This driver supports the MDIO interface found in the network
diff --git a/drivers/net/ethernet/qualcomm/Kconfig b/drivers/net/ethernet/qualcomm/Kconfig
index a4434eb38950..9210ff360fdc 100644
--- a/drivers/net/ethernet/qualcomm/Kconfig
+++ b/drivers/net/ethernet/qualcomm/Kconfig
@@ -52,6 +52,7 @@ config QCOM_EMAC
 	depends on HAS_DMA && HAS_IOMEM
 	select CRC32
 	select PHYLIB
+	select MDIO_DEVRES
 	help
 	  This driver supports the Qualcomm Technologies, Inc. Gigabit
 	  Ethernet Media Access Controller (EMAC). The controller
diff --git a/drivers/net/mdio/Kconfig b/drivers/net/mdio/Kconfig
index 90309980686e..9ff2e6f22f3f 100644
--- a/drivers/net/mdio/Kconfig
+++ b/drivers/net/mdio/Kconfig
@@ -65,6 +65,7 @@ config MDIO_ASPEED
 	tristate "ASPEED MDIO bus controller"
 	depends on ARCH_ASPEED || COMPILE_TEST
 	depends on OF_MDIO && HAS_IOMEM
+	depends on MDIO_DEVRES
 	help
 	  This module provides a driver for the independent MDIO bus
 	  controllers found in the ASPEED AST2600 SoC. This is a driver for the
@@ -170,6 +171,7 @@ config MDIO_IPQ4019
 	tristate "Qualcomm IPQ4019 MDIO interface support"
 	depends on HAS_IOMEM && OF_MDIO
 	depends on COMMON_CLK
+	depends on MDIO_DEVRES
 	help
 	  This driver supports the MDIO interface found in Qualcomm
 	  IPQ40xx, IPQ60xx, IPQ807x and IPQ50xx series Soc-s.
@@ -178,6 +180,7 @@ config MDIO_IPQ8064
 	tristate "Qualcomm IPQ8064 MDIO interface support"
 	depends on HAS_IOMEM && OF_MDIO
 	depends on MFD_SYSCON
+	depends on MDIO_DEVRES
 	help
 	  This driver supports the MDIO interface found in the network
 	  interface units of the IPQ8064 SoC
-- 
cgit v1.2.3-70-g09d2


From 185367221503f6bc4745d944ed372c9dfef99fa2 Mon Sep 17 00:00:00 2001
From: Mika Westerberg <mika.westerberg@linux.intel.com>
Date: Tue, 11 Apr 2023 12:10:47 +0300
Subject: net: thunderbolt: Fix sparse warnings in tbnet_check_frame() and
 tbnet_poll()

Fixes the following warnings when the driver is built with sparse
checks enabled:

main.c:767:47: warning: restricted __le32 degrades to integer
main.c:775:47: warning: restricted __le16 degrades to integer
main.c:776:44: warning: restricted __le16 degrades to integer
main.c:876:40: warning: incorrect type in assignment (different base types)
main.c:876:40:    expected restricted __le32 [usertype] frame_size
main.c:876:40:    got unsigned int [assigned] [usertype] frame_size
main.c:877:41: warning: incorrect type in assignment (different base types)
main.c:877:41:    expected restricted __le32 [usertype] frame_count
main.c:877:41:    got unsigned int [usertype]
main.c:878:41: warning: incorrect type in assignment (different base types)
main.c:878:41:    expected restricted __le16 [usertype] frame_index
main.c:878:41:    got unsigned short [usertype]
main.c:879:38: warning: incorrect type in assignment (different base types)
main.c:879:38:    expected restricted __le16 [usertype] frame_id
main.c:879:38:    got unsigned short [usertype]
main.c:880:62: warning: restricted __le32 degrades to integer
main.c:880:35: warning: restricted __le16 degrades to integer

No functional changes intended.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/thunderbolt/main.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/thunderbolt/main.c b/drivers/net/thunderbolt/main.c
index 26ef3706445e..27f8573a2b6e 100644
--- a/drivers/net/thunderbolt/main.c
+++ b/drivers/net/thunderbolt/main.c
@@ -764,7 +764,7 @@ static bool tbnet_check_frame(struct tbnet *net, const struct tbnet_frame *tf,
 	 */
 	if (net->skb && net->rx_hdr.frame_count) {
 		/* Check the frame count fits the count field */
-		if (frame_count != net->rx_hdr.frame_count) {
+		if (frame_count != le32_to_cpu(net->rx_hdr.frame_count)) {
 			net->stats.rx_length_errors++;
 			return false;
 		}
@@ -772,8 +772,8 @@ static bool tbnet_check_frame(struct tbnet *net, const struct tbnet_frame *tf,
 		/* Check the frame identifiers are incremented correctly,
 		 * and id is matching.
 		 */
-		if (frame_index != net->rx_hdr.frame_index + 1 ||
-		    frame_id != net->rx_hdr.frame_id) {
+		if (frame_index != le16_to_cpu(net->rx_hdr.frame_index) + 1 ||
+		    frame_id != le16_to_cpu(net->rx_hdr.frame_id)) {
 			net->stats.rx_missed_errors++;
 			return false;
 		}
@@ -873,11 +873,12 @@ static int tbnet_poll(struct napi_struct *napi, int budget)
 					TBNET_RX_PAGE_SIZE - hdr_size);
 		}
 
-		net->rx_hdr.frame_size = frame_size;
-		net->rx_hdr.frame_count = le32_to_cpu(hdr->frame_count);
-		net->rx_hdr.frame_index = le16_to_cpu(hdr->frame_index);
-		net->rx_hdr.frame_id = le16_to_cpu(hdr->frame_id);
-		last = net->rx_hdr.frame_index == net->rx_hdr.frame_count - 1;
+		net->rx_hdr.frame_size = hdr->frame_size;
+		net->rx_hdr.frame_count = hdr->frame_count;
+		net->rx_hdr.frame_index = hdr->frame_index;
+		net->rx_hdr.frame_id = hdr->frame_id;
+		last = le16_to_cpu(net->rx_hdr.frame_index) ==
+		       le32_to_cpu(net->rx_hdr.frame_count) - 1;
 
 		rx_packets++;
 		net->stats.rx_bytes += frame_size;
-- 
cgit v1.2.3-70-g09d2


From 5bbec0adfa03899cbf91ba99d4fdb4975d30a631 Mon Sep 17 00:00:00 2001
From: Mika Westerberg <mika.westerberg@linux.intel.com>
Date: Tue, 11 Apr 2023 12:10:48 +0300
Subject: net: thunderbolt: Fix sparse warnings in tbnet_xmit_csum_and_map()

Fixes the following warning when the driver is built with sparse checks
enabled:

main.c:993:23: warning: incorrect type in initializer (different base types)
main.c:993:23:    expected restricted __wsum [usertype] wsum
main.c:993:23:    got restricted __be32 [usertype]

No functional changes intended.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/thunderbolt/main.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/thunderbolt/main.c b/drivers/net/thunderbolt/main.c
index 27f8573a2b6e..6a43ced74881 100644
--- a/drivers/net/thunderbolt/main.c
+++ b/drivers/net/thunderbolt/main.c
@@ -991,8 +991,10 @@ static bool tbnet_xmit_csum_and_map(struct tbnet *net, struct sk_buff *skb,
 {
 	struct thunderbolt_ip_frame_header *hdr = page_address(frames[0]->page);
 	struct device *dma_dev = tb_ring_dma_device(net->tx_ring.ring);
-	__wsum wsum = htonl(skb->len - skb_transport_offset(skb));
 	unsigned int i, len, offset = skb_transport_offset(skb);
+	/* Remove payload length from checksum */
+	u32 paylen = skb->len - skb_transport_offset(skb);
+	__wsum wsum = (__force __wsum)htonl(paylen);
 	__be16 protocol = skb->protocol;
 	void *data = skb->data;
 	void *dest = hdr + 1;
-- 
cgit v1.2.3-70-g09d2


From 9c60f2a4446cae273c62bbe865e141c4be816552 Mon Sep 17 00:00:00 2001
From: Mika Westerberg <mika.westerberg@linux.intel.com>
Date: Tue, 11 Apr 2023 12:10:49 +0300
Subject: net: thunderbolt: Fix typos in comments

Fix two typos in comments:

  blongs -> belongs
  UPD -> UDP

No functional changes.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/thunderbolt/main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/thunderbolt/main.c b/drivers/net/thunderbolt/main.c
index 6a43ced74881..0c1e8970ee58 100644
--- a/drivers/net/thunderbolt/main.c
+++ b/drivers/net/thunderbolt/main.c
@@ -148,7 +148,7 @@ struct tbnet_ring {
 /**
  * struct tbnet - ThunderboltIP network driver private data
  * @svc: XDomain service the driver is bound to
- * @xd: XDomain the service blongs to
+ * @xd: XDomain the service belongs to
  * @handler: ThunderboltIP configuration protocol handler
  * @dev: Networking device
  * @napi: NAPI structure for Rx polling
@@ -1030,7 +1030,7 @@ static bool tbnet_xmit_csum_and_map(struct tbnet *net, struct sk_buff *skb,
 	/* Data points on the beginning of packet.
 	 * Check is the checksum absolute place in the packet.
 	 * ipcso will update IP checksum.
-	 * tucso will update TCP/UPD checksum.
+	 * tucso will update TCP/UDP checksum.
 	 */
 	if (protocol == htons(ETH_P_IP)) {
 		__sum16 *ipcso = dest + ((void *)&(ip_hdr(skb)->check) - data);
-- 
cgit v1.2.3-70-g09d2


From f032d8a9c8b353806366cb3786512b80f478efab Mon Sep 17 00:00:00 2001
From: Ivan Vecera <ivecera@redhat.com>
Date: Tue, 11 Apr 2023 14:04:42 +0200
Subject: bnxt_en: Allow to set switchdev mode without existing VFs

Remove an inability of bnxt_en driver to set eswitch to switchdev
mode without existing VFs by:

1. Allow to set switchdev mode in bnxt_dl_eswitch_mode_set() so
   representors are created only when num_vfs > 0 otherwise just
   set bp->eswitch_mode
2. Do not automatically change bp->eswitch_mode during
   bnxt_vf_reps_create() and bnxt_vf_reps_destroy() calls so
   the eswitch mode is managed only by an user by devlink.
   Just set temporarily bp->eswitch_mode to legacy to avoid
   re-opening of representors during destroy.
3. Create representors in bnxt_sriov_enable() if current eswitch
   mode is switchdev one

Tested by this sequence:
1. Set PF interface up
2. Set PF's eswitch mode to switchdev
3. Created N VFs
4. Checked that N representors were created
5. Set eswitch mode to legacy
6. Checked that representors were deleted
7. Set eswitch mode back to switchdev
8. Checked that representors exist again for VFs
9. Deleted all VFs
10. Checked that all representors were deleted as well
11. Checked that current eswitch mode is still switchdev

Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Link: https://lore.kernel.org/r/20230411120443.126055-1-ivecera@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 16 ++++++++++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c   | 29 ++++++++++++++++---------
 drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h   |  6 +++++
 3 files changed, 41 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index 3ed3a2b3b3a9..dde327f2c57e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -825,8 +825,24 @@ static int bnxt_sriov_enable(struct bnxt *bp, int *num_vfs)
 	if (rc)
 		goto err_out2;
 
+	if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
+		return 0;
+
+	/* Create representors for VFs in switchdev mode */
+	devl_lock(bp->dl);
+	rc = bnxt_vf_reps_create(bp);
+	devl_unlock(bp->dl);
+	if (rc) {
+		netdev_info(bp->dev, "Cannot enable VFS as representors cannot be created\n");
+		goto err_out3;
+	}
+
 	return 0;
 
+err_out3:
+	/* Disable SR-IOV */
+	pci_disable_sriov(bp->pdev);
+
 err_out2:
 	/* Free the resources reserved for various VF's */
 	bnxt_hwrm_func_vf_resource_free(bp, *num_vfs);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
index fcc65890820a..2f1a1f2d2157 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
@@ -356,10 +356,15 @@ void bnxt_vf_reps_destroy(struct bnxt *bp)
 	/* un-publish cfa_code_map so that RX path can't see it anymore */
 	kfree(bp->cfa_code_map);
 	bp->cfa_code_map = NULL;
-	bp->eswitch_mode = DEVLINK_ESWITCH_MODE_LEGACY;
 
-	if (closed)
+	if (closed) {
+		/* Temporarily set legacy mode to avoid re-opening
+		 * representors and restore switchdev mode after that.
+		 */
+		bp->eswitch_mode = DEVLINK_ESWITCH_MODE_LEGACY;
 		bnxt_open_nic(bp, false, false);
+		bp->eswitch_mode = DEVLINK_ESWITCH_MODE_SWITCHDEV;
+	}
 	rtnl_unlock();
 
 	/* Need to call vf_reps_destroy() outside of rntl_lock
@@ -482,7 +487,7 @@ static void bnxt_vf_rep_netdev_init(struct bnxt *bp, struct bnxt_vf_rep *vf_rep,
 	dev->min_mtu = ETH_ZLEN;
 }
 
-static int bnxt_vf_reps_create(struct bnxt *bp)
+int bnxt_vf_reps_create(struct bnxt *bp)
 {
 	u16 *cfa_code_map = NULL, num_vfs = pci_num_vf(bp->pdev);
 	struct bnxt_vf_rep *vf_rep;
@@ -535,7 +540,6 @@ static int bnxt_vf_reps_create(struct bnxt *bp)
 
 	/* publish cfa_code_map only after all VF-reps have been initialized */
 	bp->cfa_code_map = cfa_code_map;
-	bp->eswitch_mode = DEVLINK_ESWITCH_MODE_SWITCHDEV;
 	netif_keep_dst(bp->dev);
 	return 0;
 
@@ -559,6 +563,7 @@ int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode,
 			     struct netlink_ext_ack *extack)
 {
 	struct bnxt *bp = bnxt_get_bp_from_dl(devlink);
+	int ret = 0;
 
 	if (bp->eswitch_mode == mode) {
 		netdev_info(bp->dev, "already in %s eswitch mode\n",
@@ -570,7 +575,7 @@ int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode,
 	switch (mode) {
 	case DEVLINK_ESWITCH_MODE_LEGACY:
 		bnxt_vf_reps_destroy(bp);
-		return 0;
+		break;
 
 	case DEVLINK_ESWITCH_MODE_SWITCHDEV:
 		if (bp->hwrm_spec_code < 0x10803) {
@@ -578,15 +583,19 @@ int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode,
 			return -ENOTSUPP;
 		}
 
-		if (pci_num_vf(bp->pdev) == 0) {
-			netdev_info(bp->dev, "Enable VFs before setting switchdev mode\n");
-			return -EPERM;
-		}
-		return bnxt_vf_reps_create(bp);
+		/* Create representors for existing VFs */
+		if (pci_num_vf(bp->pdev) > 0)
+			ret = bnxt_vf_reps_create(bp);
+		break;
 
 	default:
 		return -EINVAL;
 	}
+
+	if (!ret)
+		bp->eswitch_mode = mode;
+
+	return ret;
 }
 
 #endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h
index 5637a84884d7..33a965631d0b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.h
@@ -14,6 +14,7 @@
 
 #define	MAX_CFA_CODE			65536
 
+int bnxt_vf_reps_create(struct bnxt *bp);
 void bnxt_vf_reps_destroy(struct bnxt *bp);
 void bnxt_vf_reps_close(struct bnxt *bp);
 void bnxt_vf_reps_open(struct bnxt *bp);
@@ -37,6 +38,11 @@ int bnxt_dl_eswitch_mode_set(struct devlink *devlink, u16 mode,
 
 #else
 
+static inline int bnxt_vf_reps_create(struct bnxt *bp)
+{
+	return 0;
+}
+
 static inline void bnxt_vf_reps_close(struct bnxt *bp)
 {
 }
-- 
cgit v1.2.3-70-g09d2


From d554ba0ea03cbd1cc1455f6133d03cd9b7967ac6 Mon Sep 17 00:00:00 2001
From: Bhupesh Sharma <bhupesh.sharma@linaro.org>
Date: Tue, 11 Apr 2023 15:03:58 -0500
Subject: dt-bindings: net: snps,dwmac: Update interrupt-names

As commit fc191af1bb0d ("net: stmmac: platform: Fix misleading
interrupt error msg") noted, not every stmmac based platform
makes use of the 'eth_wake_irq' or 'eth_lpi' interrupts.

So, update the 'interrupt-names' inside 'snps,dwmac' YAML
bindings to reflect the same.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/snps,dwmac.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 74f2ddc12018..5a4737e969a3 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -105,7 +105,7 @@ properties:
     minItems: 1
     items:
       - const: macirq
-      - const: eth_wake_irq
+      - enum: [eth_wake_irq, eth_lpi]
       - const: eth_lpi
 
   clocks:
-- 
cgit v1.2.3-70-g09d2


From d70c215bdd17b92358e90c4e4d190e749a48d982 Mon Sep 17 00:00:00 2001
From: Bhupesh Sharma <bhupesh.sharma@linaro.org>
Date: Tue, 11 Apr 2023 15:03:59 -0500
Subject: dt-bindings: net: snps,dwmac: Add Qualcomm Ethernet ETHQOS
 compatibles

Add Qualcomm Ethernet ETHQOS compatible checks
in snps,dwmac YAML binding document.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/snps,dwmac.yaml | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 5a4737e969a3..1e7982704114 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -65,6 +65,8 @@ properties:
         - ingenic,x2000-mac
         - loongson,ls2k-dwmac
         - loongson,ls7a-dwmac
+        - qcom,qcs404-ethqos
+        - qcom,sm8150-ethqos
         - renesas,r9a06g032-gmac
         - renesas,rzn1-gmac
         - rockchip,px30-gmac
@@ -625,6 +627,8 @@ allOf:
               - ingenic,x1600-mac
               - ingenic,x1830-mac
               - ingenic,x2000-mac
+              - qcom,qcs404-ethqos
+              - qcom,sm8150-ethqos
               - snps,dwmac-4.00
               - snps,dwmac-4.10a
               - snps,dwmac-4.20a
-- 
cgit v1.2.3-70-g09d2


From 02e98ce3db149919d6aa35f873e5d74c1ac36640 Mon Sep 17 00:00:00 2001
From: Bhupesh Sharma <bhupesh.sharma@linaro.org>
Date: Tue, 11 Apr 2023 15:04:00 -0500
Subject: dt-bindings: net: qcom,ethqos: Convert bindings to yaml

Convert Qualcomm ETHQOS Ethernet devicetree binding to YAML.
In doing so add a new property for iommus since newer platforms support
using one, and without such make dtbs_check fails on them.

While at it, also update the MAINTAINERS file to point to the yaml
version of the bindings.

Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
[halaney: Remove duplicated properties, add MAINTAINERS and iommus]
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../devicetree/bindings/net/qcom,ethqos.txt        |  66 -------------
 .../devicetree/bindings/net/qcom,ethqos.yaml       | 110 +++++++++++++++++++++
 MAINTAINERS                                        |   2 +-
 3 files changed, 111 insertions(+), 67 deletions(-)
 delete mode 100644 Documentation/devicetree/bindings/net/qcom,ethqos.txt
 create mode 100644 Documentation/devicetree/bindings/net/qcom,ethqos.yaml

diff --git a/Documentation/devicetree/bindings/net/qcom,ethqos.txt b/Documentation/devicetree/bindings/net/qcom,ethqos.txt
deleted file mode 100644
index 1f5746849a71..000000000000
--- a/Documentation/devicetree/bindings/net/qcom,ethqos.txt
+++ /dev/null
@@ -1,66 +0,0 @@
-Qualcomm Ethernet ETHQOS device
-
-This documents dwmmac based ethernet device which supports Gigabit
-ethernet for version v2.3.0 onwards.
-
-This device has following properties:
-
-Required properties:
-
-- compatible: Should be one of:
-		"qcom,qcs404-ethqos"
-		"qcom,sm8150-ethqos"
-
-- reg: Address and length of the register set for the device
-
-- reg-names: Should contain register names "stmmaceth", "rgmii"
-
-- clocks: Should contain phandle to clocks
-
-- clock-names: Should contain clock names "stmmaceth", "pclk",
-		"ptp_ref", "rgmii"
-
-- interrupts: Should contain phandle to interrupts
-
-- interrupt-names: Should contain interrupt names "macirq", "eth_lpi"
-
-Rest of the properties are defined in stmmac.txt file in same directory
-
-
-Example:
-
-ethernet: ethernet@7a80000 {
-	compatible = "qcom,qcs404-ethqos";
-	reg = <0x07a80000 0x10000>,
-		<0x07a96000 0x100>;
-	reg-names = "stmmaceth", "rgmii";
-	clock-names = "stmmaceth", "pclk", "ptp_ref", "rgmii";
-	clocks = <&gcc GCC_ETH_AXI_CLK>,
-		<&gcc GCC_ETH_SLAVE_AHB_CLK>,
-		<&gcc GCC_ETH_PTP_CLK>,
-		<&gcc GCC_ETH_RGMII_CLK>;
-	interrupts = <GIC_SPI 56 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 55 IRQ_TYPE_LEVEL_HIGH>;
-	interrupt-names = "macirq", "eth_lpi";
-	snps,reset-gpio = <&tlmm 60 GPIO_ACTIVE_LOW>;
-	snps,reset-active-low;
-
-	snps,txpbl = <8>;
-	snps,rxpbl = <2>;
-	snps,aal;
-	snps,tso;
-
-	phy-handle = <&phy1>;
-	phy-mode = "rgmii";
-
-	mdio {
-		#address-cells = <0x1>;
-		#size-cells = <0x0>;
-		compatible = "snps,dwmac-mdio";
-		phy1: phy@4 {
-			device_type = "ethernet-phy";
-			reg = <0x4>;
-		};
-	};
-
-};
diff --git a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml
new file mode 100644
index 000000000000..88234a2010b1
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml
@@ -0,0 +1,110 @@
+# SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/qcom,ethqos.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Ethernet ETHQOS device
+
+maintainers:
+  - Bhupesh Sharma <bhupesh.sharma@linaro.org>
+
+description:
+  dwmmac based Qualcomm ethernet devices which support Gigabit
+  ethernet (version v2.3.0 and onwards).
+
+allOf:
+  - $ref: snps,dwmac.yaml#
+
+properties:
+  compatible:
+    enum:
+      - qcom,qcs404-ethqos
+      - qcom,sm8150-ethqos
+
+  reg:
+    maxItems: 2
+
+  reg-names:
+    items:
+      - const: stmmaceth
+      - const: rgmii
+
+  interrupts:
+    items:
+      - description: Combined signal for various interrupt events
+      - description: The interrupt that occurs when Rx exits the LPI state
+
+  interrupt-names:
+    items:
+      - const: macirq
+      - const: eth_lpi
+
+  clocks:
+    maxItems: 4
+
+  clock-names:
+    items:
+      - const: stmmaceth
+      - const: pclk
+      - const: ptp_ref
+      - const: rgmii
+
+  iommus:
+    maxItems: 1
+
+required:
+  - compatible
+  - clocks
+  - clock-names
+  - reg-names
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/clock/qcom,gcc-qcs404.h>
+    #include <dt-bindings/gpio/gpio.h>
+
+    ethernet: ethernet@7a80000 {
+      compatible = "qcom,qcs404-ethqos";
+      reg = <0x07a80000 0x10000>,
+            <0x07a96000 0x100>;
+      reg-names = "stmmaceth", "rgmii";
+      clock-names = "stmmaceth", "pclk", "ptp_ref", "rgmii";
+      clocks = <&gcc GCC_ETH_AXI_CLK>,
+               <&gcc GCC_ETH_SLAVE_AHB_CLK>,
+               <&gcc GCC_ETH_PTP_CLK>,
+               <&gcc GCC_ETH_RGMII_CLK>;
+      interrupts = <GIC_SPI 56 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 55 IRQ_TYPE_LEVEL_HIGH>;
+      interrupt-names = "macirq", "eth_lpi";
+
+      rx-fifo-depth = <4096>;
+      tx-fifo-depth = <4096>;
+
+      snps,tso;
+      snps,reset-gpio = <&tlmm 60 GPIO_ACTIVE_LOW>;
+      snps,reset-active-low;
+      snps,reset-delays-us = <0 10000 10000>;
+
+      pinctrl-names = "default";
+      pinctrl-0 = <&ethernet_defaults>;
+
+      phy-handle = <&phy1>;
+      phy-mode = "rgmii";
+      mdio {
+        #address-cells = <0x1>;
+        #size-cells = <0x0>;
+
+        compatible = "snps,dwmac-mdio";
+        phy1: phy@4 {
+          compatible = "ethernet-phy-ieee802.3-c22";
+          device_type = "ethernet-phy";
+          reg = <0x4>;
+
+          #phy-cells = <0>;
+        };
+      };
+    };
diff --git a/MAINTAINERS b/MAINTAINERS
index 0b19a3fb266c..fdcef63fa9a0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -17304,7 +17304,7 @@ M:	Vinod Koul <vkoul@kernel.org>
 R:	Bhupesh Sharma <bhupesh.sharma@linaro.org>
 L:	netdev@vger.kernel.org
 S:	Maintained
-F:	Documentation/devicetree/bindings/net/qcom,ethqos.txt
+F:	Documentation/devicetree/bindings/net/qcom,ethqos.yaml
 F:	drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
 
 QUALCOMM FASTRPC DRIVER
-- 
cgit v1.2.3-70-g09d2


From 25926a703ec155c99a28f6730b79e9da3acc0b10 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:01 -0500
Subject: dt-bindings: net: qcom,ethqos: Add Qualcomm sc8280xp compatibles

The sc8280xp has a new version of the ETHQOS hardware in it, EMAC v3.
Add a compatible for this.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/qcom,ethqos.yaml | 1 +
 Documentation/devicetree/bindings/net/snps,dwmac.yaml  | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml
index 88234a2010b1..60a38044fb19 100644
--- a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml
+++ b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml
@@ -20,6 +20,7 @@ properties:
   compatible:
     enum:
       - qcom,qcs404-ethqos
+      - qcom,sc8280xp-ethqos
       - qcom,sm8150-ethqos
 
   reg:
diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 1e7982704114..da311c1f2c88 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -66,6 +66,7 @@ properties:
         - loongson,ls2k-dwmac
         - loongson,ls7a-dwmac
         - qcom,qcs404-ethqos
+        - qcom,sc8280xp-ethqos
         - qcom,sm8150-ethqos
         - renesas,r9a06g032-gmac
         - renesas,rzn1-gmac
@@ -574,6 +575,7 @@ allOf:
               - ingenic,x1600-mac
               - ingenic,x1830-mac
               - ingenic,x2000-mac
+              - qcom,sc8280xp-ethqos
               - snps,dwmac-3.50a
               - snps,dwmac-4.10a
               - snps,dwmac-4.20a
@@ -628,6 +630,7 @@ allOf:
               - ingenic,x1830-mac
               - ingenic,x2000-mac
               - qcom,qcs404-ethqos
+              - qcom,sc8280xp-ethqos
               - qcom,sm8150-ethqos
               - snps,dwmac-4.00
               - snps,dwmac-4.10a
-- 
cgit v1.2.3-70-g09d2


From 7c6b942b81caec74fab53dac508e3f2541867d0a Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:02 -0500
Subject: net: stmmac: Remove unnecessary if statement brackets

The brackets are unnecessary, remove them to match the coding style
used in the kernel.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index 21aaa2730ac8..6807c4c1a0a2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -281,9 +281,8 @@ static int stmmac_mdio_read_c22(struct mii_bus *bus, int phyaddr, int phyreg)
 	value |= (phyreg << priv->hw->mii.reg_shift) & priv->hw->mii.reg_mask;
 	value |= (priv->clk_csr << priv->hw->mii.clk_csr_shift)
 		& priv->hw->mii.clk_csr_mask;
-	if (priv->plat->has_gmac4) {
+	if (priv->plat->has_gmac4)
 		value |= MII_GMAC4_READ;
-	}
 
 	data = stmmac_mdio_read(priv, data, value);
 
-- 
cgit v1.2.3-70-g09d2


From d638dcb52b099e9123bef2690ffe2e2bf9d39746 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:03 -0500
Subject: net: stmmac: Fix DMA typo

DAM is supposed to be DMA. Fix it to improve readability.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 54bb072aeb2d..4ad692c4116c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -242,7 +242,7 @@ struct stmmac_safety_stats {
 
 #define SF_DMA_MODE 1		/* DMA STORE-AND-FORWARD Operation Mode */
 
-/* DAM HW feature register fields */
+/* DMA HW feature register fields */
 #define DMA_HW_FEAT_MIISEL	0x00000001	/* 10/100 Mbps Support */
 #define DMA_HW_FEAT_GMIISEL	0x00000002	/* 1000 Mbps Support */
 #define DMA_HW_FEAT_HDSEL	0x00000004	/* Half-Duplex Support */
-- 
cgit v1.2.3-70-g09d2


From 0c3f3c4f4b15a2c105e1ca882d100048074a2865 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:04 -0500
Subject: net: stmmac: Remove some unnecessary void pointers

There's a few spots in the hardware interface where a void pointer is
used, but what's passed in and later cast out is always the same type.

Just use the proper type directly.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/chain_mode.c   | 10 +++-----
 drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c |  4 +--
 drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c |  8 +++---
 .../net/ethernet/stmicro/stmmac/dwxgmac2_descs.c   |  6 +++--
 drivers/net/ethernet/stmicro/stmmac/enh_desc.c     | 11 ++++----
 drivers/net/ethernet/stmicro/stmmac/hwif.h         | 30 ++++++++++++++--------
 drivers/net/ethernet/stmicro/stmmac/norm_desc.c    |  8 +++---
 drivers/net/ethernet/stmicro/stmmac/ring_mode.c    | 10 +++-----
 8 files changed, 47 insertions(+), 40 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index 2e8744ac6b91..fb55efd52240 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -14,9 +14,9 @@
 
 #include "stmmac.h"
 
-static int jumbo_frm(void *p, struct sk_buff *skb, int csum)
+static int jumbo_frm(struct stmmac_tx_queue *tx_q, struct sk_buff *skb,
+		     int csum)
 {
-	struct stmmac_tx_queue *tx_q = (struct stmmac_tx_queue *)p;
 	unsigned int nopaged_len = skb_headlen(skb);
 	struct stmmac_priv *priv = tx_q->priv_data;
 	unsigned int entry = tx_q->cur_tx;
@@ -125,9 +125,8 @@ static void init_dma_chain(void *des, dma_addr_t phy_addr,
 	}
 }
 
-static void refill_desc3(void *priv_ptr, struct dma_desc *p)
+static void refill_desc3(struct stmmac_rx_queue *rx_q, struct dma_desc *p)
 {
-	struct stmmac_rx_queue *rx_q = (struct stmmac_rx_queue *)priv_ptr;
 	struct stmmac_priv *priv = rx_q->priv_data;
 
 	if (priv->hwts_rx_en && !priv->extend_desc)
@@ -141,9 +140,8 @@ static void refill_desc3(void *priv_ptr, struct dma_desc *p)
 				      sizeof(struct dma_desc)));
 }
 
-static void clean_desc3(void *priv_ptr, struct dma_desc *p)
+static void clean_desc3(struct stmmac_tx_queue *tx_q, struct dma_desc *p)
 {
-	struct stmmac_tx_queue *tx_q = (struct stmmac_tx_queue *)priv_ptr;
 	struct stmmac_priv *priv = tx_q->priv_data;
 	unsigned int entry = tx_q->dirty_tx;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index 8f0d9bc7cab5..f6abc7bfd29d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -80,10 +80,10 @@ static void dwmac100_dump_dma_regs(void __iomem *ioaddr, u32 *reg_space)
 }
 
 /* DMA controller has two counters to track the number of the missed frames. */
-static void dwmac100_dma_diagnostic_fr(void *data, struct stmmac_extra_stats *x,
+static void dwmac100_dma_diagnostic_fr(struct net_device_stats *stats,
+				       struct stmmac_extra_stats *x,
 				       void __iomem *ioaddr)
 {
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 	u32 csr8 = readl(ioaddr + DMA_MISSED_FRAME_CTR);
 
 	if (unlikely(csr8)) {
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
index 8cc80b1db4cb..6a011d8633e8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
@@ -13,11 +13,11 @@
 #include "dwmac4.h"
 #include "dwmac4_descs.h"
 
-static int dwmac4_wrback_get_tx_status(void *data, struct stmmac_extra_stats *x,
+static int dwmac4_wrback_get_tx_status(struct net_device_stats *stats,
+				       struct stmmac_extra_stats *x,
 				       struct dma_desc *p,
 				       void __iomem *ioaddr)
 {
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int tdes3;
 	int ret = tx_done;
 
@@ -73,10 +73,10 @@ static int dwmac4_wrback_get_tx_status(void *data, struct stmmac_extra_stats *x,
 	return ret;
 }
 
-static int dwmac4_wrback_get_rx_status(void *data, struct stmmac_extra_stats *x,
+static int dwmac4_wrback_get_rx_status(struct net_device_stats *stats,
+				       struct stmmac_extra_stats *x,
 				       struct dma_desc *p)
 {
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int rdes1 = le32_to_cpu(p->des1);
 	unsigned int rdes2 = le32_to_cpu(p->des2);
 	unsigned int rdes3 = le32_to_cpu(p->des3);
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c
index b1f0c3984a09..13c347ee8be9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c
@@ -8,7 +8,8 @@
 #include "common.h"
 #include "dwxgmac2.h"
 
-static int dwxgmac2_get_tx_status(void *data, struct stmmac_extra_stats *x,
+static int dwxgmac2_get_tx_status(struct net_device_stats *stats,
+				  struct stmmac_extra_stats *x,
 				  struct dma_desc *p, void __iomem *ioaddr)
 {
 	unsigned int tdes3 = le32_to_cpu(p->des3);
@@ -22,7 +23,8 @@ static int dwxgmac2_get_tx_status(void *data, struct stmmac_extra_stats *x,
 	return ret;
 }
 
-static int dwxgmac2_get_rx_status(void *data, struct stmmac_extra_stats *x,
+static int dwxgmac2_get_rx_status(struct net_device_stats *stats,
+				  struct stmmac_extra_stats *x,
 				  struct dma_desc *p)
 {
 	unsigned int rdes3 = le32_to_cpu(p->des3);
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 1bcbbd724fb5..a91d8f13a931 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -12,10 +12,10 @@
 #include "common.h"
 #include "descs_com.h"
 
-static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
+static int enh_desc_get_tx_status(struct net_device_stats *stats,
+				  struct stmmac_extra_stats *x,
 				  struct dma_desc *p, void __iomem *ioaddr)
 {
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int tdes0 = le32_to_cpu(p->des0);
 	int ret = tx_done;
 
@@ -117,7 +117,8 @@ static int enh_desc_coe_rdes0(int ipc_err, int type, int payload_err)
 	return ret;
 }
 
-static void enh_desc_get_ext_status(void *data, struct stmmac_extra_stats *x,
+static void enh_desc_get_ext_status(struct net_device_stats *stats,
+				    struct stmmac_extra_stats *x,
 				    struct dma_extended_desc *p)
 {
 	unsigned int rdes0 = le32_to_cpu(p->basic.des0);
@@ -181,10 +182,10 @@ static void enh_desc_get_ext_status(void *data, struct stmmac_extra_stats *x,
 	}
 }
 
-static int enh_desc_get_rx_status(void *data, struct stmmac_extra_stats *x,
+static int enh_desc_get_rx_status(struct net_device_stats *stats,
+				  struct stmmac_extra_stats *x,
 				  struct dma_desc *p)
 {
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int rdes0 = le32_to_cpu(p->des0);
 	int ret = good_frame;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 1cc286b000b6..7f906ef3ea4f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -56,8 +56,9 @@ struct stmmac_desc_ops {
 	/* Last tx segment reports the transmit status */
 	int (*get_tx_ls)(struct dma_desc *p);
 	/* Return the transmit status looking at the TDES1 */
-	int (*tx_status)(void *data, struct stmmac_extra_stats *x,
-			struct dma_desc *p, void __iomem *ioaddr);
+	int (*tx_status)(struct net_device_stats *stats,
+			 struct stmmac_extra_stats *x,
+			 struct dma_desc *p, void __iomem *ioaddr);
 	/* Get the buffer size from the descriptor */
 	int (*get_tx_len)(struct dma_desc *p);
 	/* Handle extra events on specific interrupts hw dependent */
@@ -65,10 +66,12 @@ struct stmmac_desc_ops {
 	/* Get the receive frame size */
 	int (*get_rx_frame_len)(struct dma_desc *p, int rx_coe_type);
 	/* Return the reception status looking at the RDES1 */
-	int (*rx_status)(void *data, struct stmmac_extra_stats *x,
-			struct dma_desc *p);
-	void (*rx_extended_status)(void *data, struct stmmac_extra_stats *x,
-			struct dma_extended_desc *p);
+	int (*rx_status)(struct net_device_stats *stats,
+			 struct stmmac_extra_stats *x,
+			 struct dma_desc *p);
+	void (*rx_extended_status)(struct net_device_stats *stats,
+				   struct stmmac_extra_stats *x,
+				   struct dma_extended_desc *p);
 	/* Set tx timestamp enable bit */
 	void (*enable_tx_timestamp) (struct dma_desc *p);
 	/* get tx timestamp status */
@@ -185,8 +188,9 @@ struct stmmac_dma_ops {
 	void (*dma_tx_mode)(void __iomem *ioaddr, int mode, u32 channel,
 			    int fifosz, u8 qmode);
 	/* To track extra statistic (if supported) */
-	void (*dma_diagnostic_fr) (void *data, struct stmmac_extra_stats *x,
-				   void __iomem *ioaddr);
+	void (*dma_diagnostic_fr)(struct net_device_stats *stats,
+				  struct stmmac_extra_stats *x,
+				  void __iomem *ioaddr);
 	void (*enable_dma_transmission) (void __iomem *ioaddr);
 	void (*enable_dma_irq)(void __iomem *ioaddr, u32 chan,
 			       bool rx, bool tx);
@@ -533,16 +537,20 @@ struct stmmac_hwtimestamp {
 #define stmmac_timestamp_interrupt(__priv, __args...) \
 	stmmac_do_void_callback(__priv, ptp, timestamp_interrupt, __args)
 
+struct stmmac_tx_queue;
+struct stmmac_rx_queue;
+
 /* Helpers to manage the descriptors for chain and ring modes */
 struct stmmac_mode_ops {
 	void (*init) (void *des, dma_addr_t phy_addr, unsigned int size,
 		      unsigned int extend_desc);
 	unsigned int (*is_jumbo_frm) (int len, int ehn_desc);
-	int (*jumbo_frm)(void *priv, struct sk_buff *skb, int csum);
+	int (*jumbo_frm)(struct stmmac_tx_queue *tx_q, struct sk_buff *skb,
+			 int csum);
 	int (*set_16kib_bfsize)(int mtu);
 	void (*init_desc3)(struct dma_desc *p);
-	void (*refill_desc3) (void *priv, struct dma_desc *p);
-	void (*clean_desc3) (void *priv, struct dma_desc *p);
+	void (*refill_desc3)(struct stmmac_rx_queue *rx_q, struct dma_desc *p);
+	void (*clean_desc3)(struct stmmac_tx_queue *tx_q, struct dma_desc *p);
 };
 
 #define stmmac_mode_init(__priv, __args...) \
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index e3da4da242ee..350e6670a576 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -12,10 +12,10 @@
 #include "common.h"
 #include "descs_com.h"
 
-static int ndesc_get_tx_status(void *data, struct stmmac_extra_stats *x,
+static int ndesc_get_tx_status(struct net_device_stats *stats,
+			       struct stmmac_extra_stats *x,
 			       struct dma_desc *p, void __iomem *ioaddr)
 {
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int tdes0 = le32_to_cpu(p->des0);
 	unsigned int tdes1 = le32_to_cpu(p->des1);
 	int ret = tx_done;
@@ -70,12 +70,12 @@ static int ndesc_get_tx_len(struct dma_desc *p)
  * and, if required, updates the multicast statistics.
  * In case of success, it returns good_frame because the GMAC device
  * is supposed to be able to compute the csum in HW. */
-static int ndesc_get_rx_status(void *data, struct stmmac_extra_stats *x,
+static int ndesc_get_rx_status(struct net_device_stats *stats,
+			       struct stmmac_extra_stats *x,
 			       struct dma_desc *p)
 {
 	int ret = good_frame;
 	unsigned int rdes0 = le32_to_cpu(p->des0);
-	struct net_device_stats *stats = (struct net_device_stats *)data;
 
 	if (unlikely(rdes0 & RDES0_OWN))
 		return dma_own;
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index 2b5b17d8b8a0..d218412ca832 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -14,9 +14,9 @@
 
 #include "stmmac.h"
 
-static int jumbo_frm(void *p, struct sk_buff *skb, int csum)
+static int jumbo_frm(struct stmmac_tx_queue *tx_q, struct sk_buff *skb,
+		     int csum)
 {
-	struct stmmac_tx_queue *tx_q = (struct stmmac_tx_queue *)p;
 	unsigned int nopaged_len = skb_headlen(skb);
 	struct stmmac_priv *priv = tx_q->priv_data;
 	unsigned int entry = tx_q->cur_tx;
@@ -101,9 +101,8 @@ static unsigned int is_jumbo_frm(int len, int enh_desc)
 	return ret;
 }
 
-static void refill_desc3(void *priv_ptr, struct dma_desc *p)
+static void refill_desc3(struct stmmac_rx_queue *rx_q, struct dma_desc *p)
 {
-	struct stmmac_rx_queue *rx_q = priv_ptr;
 	struct stmmac_priv *priv = rx_q->priv_data;
 
 	/* Fill DES3 in case of RING mode */
@@ -117,9 +116,8 @@ static void init_desc3(struct dma_desc *p)
 	p->des3 = cpu_to_le32(le32_to_cpu(p->des2) + BUF_SIZE_8KiB);
 }
 
-static void clean_desc3(void *priv_ptr, struct dma_desc *p)
+static void clean_desc3(struct stmmac_tx_queue *tx_q, struct dma_desc *p)
 {
-	struct stmmac_tx_queue *tx_q = (struct stmmac_tx_queue *)priv_ptr;
 	struct stmmac_priv *priv = tx_q->priv_data;
 	unsigned int entry = tx_q->dirty_tx;
 
-- 
cgit v1.2.3-70-g09d2


From 1d84b487bc2d9061bd00203f571e5cb99fe0a312 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:05 -0500
Subject: net: stmmac: Pass stmmac_priv in some callbacks

Passing stmmac_priv to some of the callbacks allows hwif implementations
to grab some data that platforms can customize. Adjust the callbacks
accordingly in preparation of such a platform customization.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c  |  36 +++--
 .../net/ethernet/stmicro/stmmac/dwmac1000_core.c   |   3 +-
 .../net/ethernet/stmicro/stmmac/dwmac1000_dma.c    |  19 ++-
 drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c |  10 +-
 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c  |  14 +-
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c   |  44 ++++---
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h   |  38 ++++--
 drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c   |  38 ++++--
 drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h    |  22 ++--
 drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c    |  18 ++-
 .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c    |   9 +-
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c |  71 ++++++----
 drivers/net/ethernet/stmicro/stmmac/hwif.h         | 146 ++++++++++++---------
 13 files changed, 291 insertions(+), 177 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
index f834472599f7..c2c592ba0eb8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
@@ -304,7 +304,8 @@ static void sun8i_dwmac_dma_init(void __iomem *ioaddr,
 	writel(0x1FFFFFF, ioaddr + EMAC_INT_STA);
 }
 
-static void sun8i_dwmac_dma_init_rx(void __iomem *ioaddr,
+static void sun8i_dwmac_dma_init_rx(struct stmmac_priv *priv,
+				    void __iomem *ioaddr,
 				    struct stmmac_dma_cfg *dma_cfg,
 				    dma_addr_t dma_rx_phy, u32 chan)
 {
@@ -312,7 +313,8 @@ static void sun8i_dwmac_dma_init_rx(void __iomem *ioaddr,
 	writel(lower_32_bits(dma_rx_phy), ioaddr + EMAC_RX_DESC_LIST);
 }
 
-static void sun8i_dwmac_dma_init_tx(void __iomem *ioaddr,
+static void sun8i_dwmac_dma_init_tx(struct stmmac_priv *priv,
+				    void __iomem *ioaddr,
 				    struct stmmac_dma_cfg *dma_cfg,
 				    dma_addr_t dma_tx_phy, u32 chan)
 {
@@ -324,7 +326,8 @@ static void sun8i_dwmac_dma_init_tx(void __iomem *ioaddr,
  * Called from stmmac_dma_ops->dump_regs
  * Used for ethtool
  */
-static void sun8i_dwmac_dump_regs(void __iomem *ioaddr, u32 *reg_space)
+static void sun8i_dwmac_dump_regs(struct stmmac_priv *priv,
+				  void __iomem *ioaddr, u32 *reg_space)
 {
 	int i;
 
@@ -352,7 +355,8 @@ static void sun8i_dwmac_dump_mac_regs(struct mac_device_info *hw,
 	}
 }
 
-static void sun8i_dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan,
+static void sun8i_dwmac_enable_dma_irq(struct stmmac_priv *priv,
+				       void __iomem *ioaddr, u32 chan,
 				       bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + EMAC_INT_EN);
@@ -365,7 +369,8 @@ static void sun8i_dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan,
 	writel(value, ioaddr + EMAC_INT_EN);
 }
 
-static void sun8i_dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan,
+static void sun8i_dwmac_disable_dma_irq(struct stmmac_priv *priv,
+					void __iomem *ioaddr, u32 chan,
 					bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + EMAC_INT_EN);
@@ -378,7 +383,8 @@ static void sun8i_dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan,
 	writel(value, ioaddr + EMAC_INT_EN);
 }
 
-static void sun8i_dwmac_dma_start_tx(void __iomem *ioaddr, u32 chan)
+static void sun8i_dwmac_dma_start_tx(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 chan)
 {
 	u32 v;
 
@@ -398,7 +404,8 @@ static void sun8i_dwmac_enable_dma_transmission(void __iomem *ioaddr)
 	writel(v, ioaddr + EMAC_TX_CTL1);
 }
 
-static void sun8i_dwmac_dma_stop_tx(void __iomem *ioaddr, u32 chan)
+static void sun8i_dwmac_dma_stop_tx(struct stmmac_priv *priv,
+				    void __iomem *ioaddr, u32 chan)
 {
 	u32 v;
 
@@ -407,7 +414,8 @@ static void sun8i_dwmac_dma_stop_tx(void __iomem *ioaddr, u32 chan)
 	writel(v, ioaddr + EMAC_TX_CTL1);
 }
 
-static void sun8i_dwmac_dma_start_rx(void __iomem *ioaddr, u32 chan)
+static void sun8i_dwmac_dma_start_rx(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 chan)
 {
 	u32 v;
 
@@ -417,7 +425,8 @@ static void sun8i_dwmac_dma_start_rx(void __iomem *ioaddr, u32 chan)
 	writel(v, ioaddr + EMAC_RX_CTL1);
 }
 
-static void sun8i_dwmac_dma_stop_rx(void __iomem *ioaddr, u32 chan)
+static void sun8i_dwmac_dma_stop_rx(struct stmmac_priv *priv,
+				    void __iomem *ioaddr, u32 chan)
 {
 	u32 v;
 
@@ -426,7 +435,8 @@ static void sun8i_dwmac_dma_stop_rx(void __iomem *ioaddr, u32 chan)
 	writel(v, ioaddr + EMAC_RX_CTL1);
 }
 
-static int sun8i_dwmac_dma_interrupt(void __iomem *ioaddr,
+static int sun8i_dwmac_dma_interrupt(struct stmmac_priv *priv,
+				     void __iomem *ioaddr,
 				     struct stmmac_extra_stats *x, u32 chan,
 				     u32 dir)
 {
@@ -492,7 +502,8 @@ static int sun8i_dwmac_dma_interrupt(void __iomem *ioaddr,
 	return ret;
 }
 
-static void sun8i_dwmac_dma_operation_mode_rx(void __iomem *ioaddr, int mode,
+static void sun8i_dwmac_dma_operation_mode_rx(struct stmmac_priv *priv,
+					      void __iomem *ioaddr, int mode,
 					      u32 channel, int fifosz, u8 qmode)
 {
 	u32 v;
@@ -515,7 +526,8 @@ static void sun8i_dwmac_dma_operation_mode_rx(void __iomem *ioaddr, int mode,
 	writel(v, ioaddr + EMAC_RX_CTL1);
 }
 
-static void sun8i_dwmac_dma_operation_mode_tx(void __iomem *ioaddr, int mode,
+static void sun8i_dwmac_dma_operation_mode_tx(struct stmmac_priv *priv,
+					      void __iomem *ioaddr, int mode,
 					      u32 channel, int fifosz, u8 qmode)
 {
 	u32 v;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
index 0e00dd83d027..3927609abc44 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
@@ -414,7 +414,8 @@ static void dwmac1000_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv)
 	dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv);
 }
 
-static void dwmac1000_debug(void __iomem *ioaddr, struct stmmac_extra_stats *x,
+static void dwmac1000_debug(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    struct stmmac_extra_stats *x,
 			    u32 rx_queues, u32 tx_queues)
 {
 	u32 value = readl(ioaddr + GMAC_DEBUG);
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index f5581db0ba9b..daf79cdbd3ec 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -110,7 +110,8 @@ static void dwmac1000_dma_init(void __iomem *ioaddr,
 	writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
 }
 
-static void dwmac1000_dma_init_rx(void __iomem *ioaddr,
+static void dwmac1000_dma_init_rx(struct stmmac_priv *priv,
+				  void __iomem *ioaddr,
 				  struct stmmac_dma_cfg *dma_cfg,
 				  dma_addr_t dma_rx_phy, u32 chan)
 {
@@ -118,7 +119,8 @@ static void dwmac1000_dma_init_rx(void __iomem *ioaddr,
 	writel(lower_32_bits(dma_rx_phy), ioaddr + DMA_RCV_BASE_ADDR);
 }
 
-static void dwmac1000_dma_init_tx(void __iomem *ioaddr,
+static void dwmac1000_dma_init_tx(struct stmmac_priv *priv,
+				  void __iomem *ioaddr,
 				  struct stmmac_dma_cfg *dma_cfg,
 				  dma_addr_t dma_tx_phy, u32 chan)
 {
@@ -147,7 +149,8 @@ static u32 dwmac1000_configure_fc(u32 csr6, int rxfifosz)
 	return csr6;
 }
 
-static void dwmac1000_dma_operation_mode_rx(void __iomem *ioaddr, int mode,
+static void dwmac1000_dma_operation_mode_rx(struct stmmac_priv *priv,
+					    void __iomem *ioaddr, int mode,
 					    u32 channel, int fifosz, u8 qmode)
 {
 	u32 csr6 = readl(ioaddr + DMA_CONTROL);
@@ -175,7 +178,8 @@ static void dwmac1000_dma_operation_mode_rx(void __iomem *ioaddr, int mode,
 	writel(csr6, ioaddr + DMA_CONTROL);
 }
 
-static void dwmac1000_dma_operation_mode_tx(void __iomem *ioaddr, int mode,
+static void dwmac1000_dma_operation_mode_tx(struct stmmac_priv *priv,
+					    void __iomem *ioaddr, int mode,
 					    u32 channel, int fifosz, u8 qmode)
 {
 	u32 csr6 = readl(ioaddr + DMA_CONTROL);
@@ -208,7 +212,8 @@ static void dwmac1000_dma_operation_mode_tx(void __iomem *ioaddr, int mode,
 	writel(csr6, ioaddr + DMA_CONTROL);
 }
 
-static void dwmac1000_dump_dma_regs(void __iomem *ioaddr, u32 *reg_space)
+static void dwmac1000_dump_dma_regs(struct stmmac_priv *priv,
+				    void __iomem *ioaddr, u32 *reg_space)
 {
 	int i;
 
@@ -263,8 +268,8 @@ static int dwmac1000_get_hw_feature(void __iomem *ioaddr,
 	return 0;
 }
 
-static void dwmac1000_rx_watchdog(void __iomem *ioaddr, u32 riwt,
-				  u32 queue)
+static void dwmac1000_rx_watchdog(struct stmmac_priv *priv,
+				  void __iomem *ioaddr, u32 riwt, u32 queue)
 {
 	writel(riwt, ioaddr + DMA_RX_WATCHDOG);
 }
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index f6abc7bfd29d..1c32b1788f02 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -29,7 +29,7 @@ static void dwmac100_dma_init(void __iomem *ioaddr,
 	writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
 }
 
-static void dwmac100_dma_init_rx(void __iomem *ioaddr,
+static void dwmac100_dma_init_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
 				 struct stmmac_dma_cfg *dma_cfg,
 				 dma_addr_t dma_rx_phy, u32 chan)
 {
@@ -37,7 +37,7 @@ static void dwmac100_dma_init_rx(void __iomem *ioaddr,
 	writel(lower_32_bits(dma_rx_phy), ioaddr + DMA_RCV_BASE_ADDR);
 }
 
-static void dwmac100_dma_init_tx(void __iomem *ioaddr,
+static void dwmac100_dma_init_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
 				 struct stmmac_dma_cfg *dma_cfg,
 				 dma_addr_t dma_tx_phy, u32 chan)
 {
@@ -50,7 +50,8 @@ static void dwmac100_dma_init_tx(void __iomem *ioaddr,
  * The transmit threshold can be programmed by setting the TTC bits in the DMA
  * control register.
  */
-static void dwmac100_dma_operation_mode_tx(void __iomem *ioaddr, int mode,
+static void dwmac100_dma_operation_mode_tx(struct stmmac_priv *priv,
+					   void __iomem *ioaddr, int mode,
 					   u32 channel, int fifosz, u8 qmode)
 {
 	u32 csr6 = readl(ioaddr + DMA_CONTROL);
@@ -65,7 +66,8 @@ static void dwmac100_dma_operation_mode_tx(void __iomem *ioaddr, int mode,
 	writel(csr6, ioaddr + DMA_CONTROL);
 }
 
-static void dwmac100_dump_dma_regs(void __iomem *ioaddr, u32 *reg_space)
+static void dwmac100_dump_dma_regs(struct stmmac_priv *priv,
+				   void __iomem *ioaddr, u32 *reg_space)
 {
 	int i;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
index 36251ec2589c..f44180519638 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
@@ -198,7 +198,8 @@ static void dwmac4_prog_mtl_tx_algorithms(struct mac_device_info *hw,
 	writel(value, ioaddr + MTL_OPERATION_MODE);
 }
 
-static void dwmac4_set_mtl_tx_queue_weight(struct mac_device_info *hw,
+static void dwmac4_set_mtl_tx_queue_weight(struct stmmac_priv *priv,
+					   struct mac_device_info *hw,
 					   u32 weight, u32 queue)
 {
 	void __iomem *ioaddr = hw->pcsr;
@@ -227,7 +228,8 @@ static void dwmac4_map_mtl_dma(struct mac_device_info *hw, u32 queue, u32 chan)
 	}
 }
 
-static void dwmac4_config_cbs(struct mac_device_info *hw,
+static void dwmac4_config_cbs(struct stmmac_priv *priv,
+			      struct mac_device_info *hw,
 			      u32 send_slope, u32 idle_slope,
 			      u32 high_credit, u32 low_credit, u32 queue)
 {
@@ -253,7 +255,7 @@ static void dwmac4_config_cbs(struct mac_device_info *hw,
 	writel(value, ioaddr + MTL_SEND_SLP_CREDX_BASE_ADDR(queue));
 
 	/* configure idle slope (same register as tx weight) */
-	dwmac4_set_mtl_tx_queue_weight(hw, idle_slope, queue);
+	dwmac4_set_mtl_tx_queue_weight(priv, hw, idle_slope, queue);
 
 	/* configure high credit */
 	value = readl(ioaddr + MTL_HIGH_CREDX_BASE_ADDR(queue));
@@ -759,7 +761,8 @@ static void dwmac4_phystatus(void __iomem *ioaddr, struct stmmac_extra_stats *x)
 	}
 }
 
-static int dwmac4_irq_mtl_status(struct mac_device_info *hw, u32 chan)
+static int dwmac4_irq_mtl_status(struct stmmac_priv *priv,
+				 struct mac_device_info *hw, u32 chan)
 {
 	void __iomem *ioaddr = hw->pcsr;
 	u32 mtl_int_qx_status;
@@ -833,7 +836,8 @@ static int dwmac4_irq_status(struct mac_device_info *hw,
 	return ret;
 }
 
-static void dwmac4_debug(void __iomem *ioaddr, struct stmmac_extra_stats *x,
+static void dwmac4_debug(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 struct stmmac_extra_stats *x,
 			 u32 rx_queues, u32 tx_queues)
 {
 	u32 value;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
index d99fa028c646..6f255d12f60f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
@@ -68,7 +68,8 @@ static void dwmac4_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi)
 	writel(value, ioaddr + DMA_SYS_BUS_MODE);
 }
 
-static void dwmac4_dma_init_rx_chan(void __iomem *ioaddr,
+static void dwmac4_dma_init_rx_chan(struct stmmac_priv *priv,
+				    void __iomem *ioaddr,
 				    struct stmmac_dma_cfg *dma_cfg,
 				    dma_addr_t dma_rx_phy, u32 chan)
 {
@@ -86,7 +87,8 @@ static void dwmac4_dma_init_rx_chan(void __iomem *ioaddr,
 	writel(lower_32_bits(dma_rx_phy), ioaddr + DMA_CHAN_RX_BASE_ADDR(chan));
 }
 
-static void dwmac4_dma_init_tx_chan(void __iomem *ioaddr,
+static void dwmac4_dma_init_tx_chan(struct stmmac_priv *priv,
+				    void __iomem *ioaddr,
 				    struct stmmac_dma_cfg *dma_cfg,
 				    dma_addr_t dma_tx_phy, u32 chan)
 {
@@ -108,7 +110,8 @@ static void dwmac4_dma_init_tx_chan(void __iomem *ioaddr,
 	writel(lower_32_bits(dma_tx_phy), ioaddr + DMA_CHAN_TX_BASE_ADDR(chan));
 }
 
-static void dwmac4_dma_init_channel(void __iomem *ioaddr,
+static void dwmac4_dma_init_channel(struct stmmac_priv *priv,
+				    void __iomem *ioaddr,
 				    struct stmmac_dma_cfg *dma_cfg, u32 chan)
 {
 	u32 value;
@@ -124,7 +127,8 @@ static void dwmac4_dma_init_channel(void __iomem *ioaddr,
 	       ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-static void dwmac410_dma_init_channel(void __iomem *ioaddr,
+static void dwmac410_dma_init_channel(struct stmmac_priv *priv,
+				      void __iomem *ioaddr,
 				      struct stmmac_dma_cfg *dma_cfg, u32 chan)
 {
 	u32 value;
@@ -176,7 +180,8 @@ static void dwmac4_dma_init(void __iomem *ioaddr,
 
 }
 
-static void _dwmac4_dump_dma_regs(void __iomem *ioaddr, u32 channel,
+static void _dwmac4_dump_dma_regs(struct stmmac_priv *priv,
+				  void __iomem *ioaddr, u32 channel,
 				  u32 *reg_space)
 {
 	reg_space[DMA_CHAN_CONTROL(channel) / 4] =
@@ -215,20 +220,23 @@ static void _dwmac4_dump_dma_regs(void __iomem *ioaddr, u32 channel,
 		readl(ioaddr + DMA_CHAN_STATUS(channel));
 }
 
-static void dwmac4_dump_dma_regs(void __iomem *ioaddr, u32 *reg_space)
+static void dwmac4_dump_dma_regs(struct stmmac_priv *priv, void __iomem *ioaddr,
+				 u32 *reg_space)
 {
 	int i;
 
 	for (i = 0; i < DMA_CHANNEL_NB_MAX; i++)
-		_dwmac4_dump_dma_regs(ioaddr, i, reg_space);
+		_dwmac4_dump_dma_regs(priv, ioaddr, i, reg_space);
 }
 
-static void dwmac4_rx_watchdog(void __iomem *ioaddr, u32 riwt, u32 queue)
+static void dwmac4_rx_watchdog(struct stmmac_priv *priv, void __iomem *ioaddr,
+			       u32 riwt, u32 queue)
 {
 	writel(riwt, ioaddr + DMA_CHAN_RX_WATCHDOG(queue));
 }
 
-static void dwmac4_dma_rx_chan_op_mode(void __iomem *ioaddr, int mode,
+static void dwmac4_dma_rx_chan_op_mode(struct stmmac_priv *priv,
+				       void __iomem *ioaddr, int mode,
 				       u32 channel, int fifosz, u8 qmode)
 {
 	unsigned int rqs = fifosz / 256 - 1;
@@ -295,7 +303,8 @@ static void dwmac4_dma_rx_chan_op_mode(void __iomem *ioaddr, int mode,
 	writel(mtl_rx_op, ioaddr + MTL_CHAN_RX_OP_MODE(channel));
 }
 
-static void dwmac4_dma_tx_chan_op_mode(void __iomem *ioaddr, int mode,
+static void dwmac4_dma_tx_chan_op_mode(struct stmmac_priv *priv,
+				       void __iomem *ioaddr, int mode,
 				       u32 channel, int fifosz, u8 qmode)
 {
 	u32 mtl_tx_op = readl(ioaddr + MTL_CHAN_TX_OP_MODE(channel));
@@ -442,7 +451,8 @@ static int dwmac4_get_hw_feature(void __iomem *ioaddr,
 }
 
 /* Enable/disable TSO feature and set MSS */
-static void dwmac4_enable_tso(void __iomem *ioaddr, bool en, u32 chan)
+static void dwmac4_enable_tso(struct stmmac_priv *priv, void __iomem *ioaddr,
+			      bool en, u32 chan)
 {
 	u32 value;
 
@@ -459,7 +469,8 @@ static void dwmac4_enable_tso(void __iomem *ioaddr, bool en, u32 chan)
 	}
 }
 
-static void dwmac4_qmode(void __iomem *ioaddr, u32 channel, u8 qmode)
+static void dwmac4_qmode(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 channel, u8 qmode)
 {
 	u32 mtl_tx_op = readl(ioaddr + MTL_CHAN_TX_OP_MODE(channel));
 
@@ -472,7 +483,8 @@ static void dwmac4_qmode(void __iomem *ioaddr, u32 channel, u8 qmode)
 	writel(mtl_tx_op, ioaddr +  MTL_CHAN_TX_OP_MODE(channel));
 }
 
-static void dwmac4_set_bfsize(void __iomem *ioaddr, int bfsize, u32 chan)
+static void dwmac4_set_bfsize(struct stmmac_priv *priv, void __iomem *ioaddr,
+			      int bfsize, u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
 
@@ -482,7 +494,8 @@ static void dwmac4_set_bfsize(void __iomem *ioaddr, int bfsize, u32 chan)
 	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(chan));
 }
 
-static void dwmac4_enable_sph(void __iomem *ioaddr, bool en, u32 chan)
+static void dwmac4_enable_sph(struct stmmac_priv *priv, void __iomem *ioaddr,
+			      bool en, u32 chan)
 {
 	u32 value = readl(ioaddr + GMAC_EXT_CONFIG);
 
@@ -498,7 +511,8 @@ static void dwmac4_enable_sph(void __iomem *ioaddr, bool en, u32 chan)
 	writel(value, ioaddr + DMA_CHAN_CONTROL(chan));
 }
 
-static int dwmac4_enable_tbs(void __iomem *ioaddr, bool en, u32 chan)
+static int dwmac4_enable_tbs(struct stmmac_priv *priv, void __iomem *ioaddr,
+			     bool en, u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
index 9321879b599c..740c3bc8d9a0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
@@ -220,19 +220,31 @@
 #define DMA_CHAN0_DBG_STAT_RPS_SHIFT	8
 
 int dwmac4_dma_reset(void __iomem *ioaddr);
-void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
-void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
-void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
-void dwmac410_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
-void dwmac4_dma_start_tx(void __iomem *ioaddr, u32 chan);
-void dwmac4_dma_stop_tx(void __iomem *ioaddr, u32 chan);
-void dwmac4_dma_start_rx(void __iomem *ioaddr, u32 chan);
-void dwmac4_dma_stop_rx(void __iomem *ioaddr, u32 chan);
-int dwmac4_dma_interrupt(void __iomem *ioaddr,
+void dwmac4_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   u32 chan, bool rx, bool tx);
+void dwmac410_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			     u32 chan, bool rx, bool tx);
+void dwmac4_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 chan, bool rx, bool tx);
+void dwmac410_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			      u32 chan, bool rx, bool tx);
+void dwmac4_dma_start_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 chan);
+void dwmac4_dma_stop_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan);
+void dwmac4_dma_start_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 chan);
+void dwmac4_dma_stop_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan);
+int dwmac4_dma_interrupt(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 struct stmmac_extra_stats *x, u32 chan, u32 dir);
-void dwmac4_set_rx_ring_len(void __iomem *ioaddr, u32 len, u32 chan);
-void dwmac4_set_tx_ring_len(void __iomem *ioaddr, u32 len, u32 chan);
-void dwmac4_set_rx_tail_ptr(void __iomem *ioaddr, u32 tail_ptr, u32 chan);
-void dwmac4_set_tx_tail_ptr(void __iomem *ioaddr, u32 tail_ptr, u32 chan);
+void dwmac4_set_rx_ring_len(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 len, u32 chan);
+void dwmac4_set_tx_ring_len(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 len, u32 chan);
+void dwmac4_set_rx_tail_ptr(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 tail_ptr, u32 chan);
+void dwmac4_set_tx_tail_ptr(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 tail_ptr, u32 chan);
 
 #endif /* __DWMAC4_DMA_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
index d1c605777985..5e9c495aa03e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
@@ -25,17 +25,20 @@ int dwmac4_dma_reset(void __iomem *ioaddr)
 				 10000, 1000000);
 }
 
-void dwmac4_set_rx_tail_ptr(void __iomem *ioaddr, u32 tail_ptr, u32 chan)
+void dwmac4_set_rx_tail_ptr(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 tail_ptr, u32 chan)
 {
 	writel(tail_ptr, ioaddr + DMA_CHAN_RX_END_ADDR(chan));
 }
 
-void dwmac4_set_tx_tail_ptr(void __iomem *ioaddr, u32 tail_ptr, u32 chan)
+void dwmac4_set_tx_tail_ptr(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 tail_ptr, u32 chan)
 {
 	writel(tail_ptr, ioaddr + DMA_CHAN_TX_END_ADDR(chan));
 }
 
-void dwmac4_dma_start_tx(void __iomem *ioaddr, u32 chan)
+void dwmac4_dma_start_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
 
@@ -47,7 +50,8 @@ void dwmac4_dma_start_tx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + GMAC_CONFIG);
 }
 
-void dwmac4_dma_stop_tx(void __iomem *ioaddr, u32 chan)
+void dwmac4_dma_stop_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
 
@@ -55,7 +59,8 @@ void dwmac4_dma_stop_tx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(chan));
 }
 
-void dwmac4_dma_start_rx(void __iomem *ioaddr, u32 chan)
+void dwmac4_dma_start_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
 
@@ -68,7 +73,8 @@ void dwmac4_dma_start_rx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + GMAC_CONFIG);
 }
 
-void dwmac4_dma_stop_rx(void __iomem *ioaddr, u32 chan)
+void dwmac4_dma_stop_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
 
@@ -76,17 +82,20 @@ void dwmac4_dma_stop_rx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(chan));
 }
 
-void dwmac4_set_tx_ring_len(void __iomem *ioaddr, u32 len, u32 chan)
+void dwmac4_set_tx_ring_len(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 len, u32 chan)
 {
 	writel(len, ioaddr + DMA_CHAN_TX_RING_LEN(chan));
 }
 
-void dwmac4_set_rx_ring_len(void __iomem *ioaddr, u32 len, u32 chan)
+void dwmac4_set_rx_ring_len(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 len, u32 chan)
 {
 	writel(len, ioaddr + DMA_CHAN_RX_RING_LEN(chan));
 }
 
-void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+void dwmac4_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   u32 chan, bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
 
@@ -98,7 +107,8 @@ void dwmac4_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+void dwmac410_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			     u32 chan, bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
 
@@ -110,7 +120,8 @@ void dwmac410_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+void dwmac4_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 chan, bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
 
@@ -122,7 +133,8 @@ void dwmac4_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-void dwmac410_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+void dwmac410_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			      u32 chan, bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
 
@@ -134,7 +146,7 @@ void dwmac410_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
 }
 
-int dwmac4_dma_interrupt(void __iomem *ioaddr,
+int dwmac4_dma_interrupt(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 struct stmmac_extra_stats *x, u32 chan, u32 dir)
 {
 	u32 intr_status = readl(ioaddr + DMA_CHAN_STATUS(chan));
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
index acd70b9a3173..72672391675f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
@@ -153,14 +153,20 @@
 #define NUM_DWMAC4_DMA_REGS	27
 
 void dwmac_enable_dma_transmission(void __iomem *ioaddr);
-void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
-void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx);
-void dwmac_dma_start_tx(void __iomem *ioaddr, u32 chan);
-void dwmac_dma_stop_tx(void __iomem *ioaddr, u32 chan);
-void dwmac_dma_start_rx(void __iomem *ioaddr, u32 chan);
-void dwmac_dma_stop_rx(void __iomem *ioaddr, u32 chan);
-int dwmac_dma_interrupt(void __iomem *ioaddr, struct stmmac_extra_stats *x,
-			u32 chan, u32 dir);
+void dwmac_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			  u32 chan, bool rx, bool tx);
+void dwmac_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   u32 chan, bool rx, bool tx);
+void dwmac_dma_start_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan);
+void dwmac_dma_stop_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+		       u32 chan);
+void dwmac_dma_start_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan);
+void dwmac_dma_stop_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+		       u32 chan);
+int dwmac_dma_interrupt(struct stmmac_priv *priv, void __iomem *ioaddr,
+			struct stmmac_extra_stats *x, u32 chan, u32 dir);
 int dwmac_dma_reset(void __iomem *ioaddr);
 
 #endif /* __DWMAC_DMA_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
index 9b6138b11776..0b6f999a8305 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
@@ -32,7 +32,8 @@ void dwmac_enable_dma_transmission(void __iomem *ioaddr)
 	writel(1, ioaddr + DMA_XMT_POLL_DEMAND);
 }
 
-void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+void dwmac_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			  u32 chan, bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + DMA_INTR_ENA);
 
@@ -44,7 +45,8 @@ void dwmac_enable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 	writel(value, ioaddr + DMA_INTR_ENA);
 }
 
-void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
+void dwmac_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   u32 chan, bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + DMA_INTR_ENA);
 
@@ -56,28 +58,30 @@ void dwmac_disable_dma_irq(void __iomem *ioaddr, u32 chan, bool rx, bool tx)
 	writel(value, ioaddr + DMA_INTR_ENA);
 }
 
-void dwmac_dma_start_tx(void __iomem *ioaddr, u32 chan)
+void dwmac_dma_start_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value |= DMA_CONTROL_ST;
 	writel(value, ioaddr + DMA_CONTROL);
 }
 
-void dwmac_dma_stop_tx(void __iomem *ioaddr, u32 chan)
+void dwmac_dma_stop_tx(struct stmmac_priv *priv, void __iomem *ioaddr, u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value &= ~DMA_CONTROL_ST;
 	writel(value, ioaddr + DMA_CONTROL);
 }
 
-void dwmac_dma_start_rx(void __iomem *ioaddr, u32 chan)
+void dwmac_dma_start_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value |= DMA_CONTROL_SR;
 	writel(value, ioaddr + DMA_CONTROL);
 }
 
-void dwmac_dma_stop_rx(void __iomem *ioaddr, u32 chan)
+void dwmac_dma_stop_rx(struct stmmac_priv *priv, void __iomem *ioaddr, u32 chan)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value &= ~DMA_CONTROL_SR;
@@ -154,7 +158,7 @@ static void show_rx_process_state(unsigned int status)
 }
 #endif
 
-int dwmac_dma_interrupt(void __iomem *ioaddr,
+int dwmac_dma_interrupt(struct stmmac_priv *priv, void __iomem *ioaddr,
 			struct stmmac_extra_stats *x, u32 chan, u32 dir)
 {
 	int ret = 0;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
index c6c4d7948fe5..a0c2ef8bb0ac 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
@@ -187,7 +187,8 @@ static void dwxgmac2_prog_mtl_tx_algorithms(struct mac_device_info *hw,
 	}
 }
 
-static void dwxgmac2_set_mtl_tx_queue_weight(struct mac_device_info *hw,
+static void dwxgmac2_set_mtl_tx_queue_weight(struct stmmac_priv *priv,
+					     struct mac_device_info *hw,
 					     u32 weight, u32 queue)
 {
 	void __iomem *ioaddr = hw->pcsr;
@@ -212,7 +213,8 @@ static void dwxgmac2_map_mtl_to_dma(struct mac_device_info *hw, u32 queue,
 	writel(value, ioaddr + reg);
 }
 
-static void dwxgmac2_config_cbs(struct mac_device_info *hw,
+static void dwxgmac2_config_cbs(struct stmmac_priv *priv,
+				struct mac_device_info *hw,
 				u32 send_slope, u32 idle_slope,
 				u32 high_credit, u32 low_credit, u32 queue)
 {
@@ -276,7 +278,8 @@ static int dwxgmac2_host_irq_status(struct mac_device_info *hw,
 	return ret;
 }
 
-static int dwxgmac2_host_mtl_irq_status(struct mac_device_info *hw, u32 chan)
+static int dwxgmac2_host_mtl_irq_status(struct stmmac_priv *priv,
+					struct mac_device_info *hw, u32 chan)
 {
 	void __iomem *ioaddr = hw->pcsr;
 	int ret = 0;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
index 5e98355f422b..dfd53264e036 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c
@@ -33,7 +33,8 @@ static void dwxgmac2_dma_init(void __iomem *ioaddr,
 	writel(value, ioaddr + XGMAC_DMA_SYSBUS_MODE);
 }
 
-static void dwxgmac2_dma_init_chan(void __iomem *ioaddr,
+static void dwxgmac2_dma_init_chan(struct stmmac_priv *priv,
+				   void __iomem *ioaddr,
 				   struct stmmac_dma_cfg *dma_cfg, u32 chan)
 {
 	u32 value = readl(ioaddr + XGMAC_DMA_CH_CONTROL(chan));
@@ -45,7 +46,8 @@ static void dwxgmac2_dma_init_chan(void __iomem *ioaddr,
 	writel(XGMAC_DMA_INT_DEFAULT_EN, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
 }
 
-static void dwxgmac2_dma_init_rx_chan(void __iomem *ioaddr,
+static void dwxgmac2_dma_init_rx_chan(struct stmmac_priv *priv,
+				      void __iomem *ioaddr,
 				      struct stmmac_dma_cfg *dma_cfg,
 				      dma_addr_t phy, u32 chan)
 {
@@ -61,7 +63,8 @@ static void dwxgmac2_dma_init_rx_chan(void __iomem *ioaddr,
 	writel(lower_32_bits(phy), ioaddr + XGMAC_DMA_CH_RxDESC_LADDR(chan));
 }
 
-static void dwxgmac2_dma_init_tx_chan(void __iomem *ioaddr,
+static void dwxgmac2_dma_init_tx_chan(struct stmmac_priv *priv,
+				      void __iomem *ioaddr,
 				      struct stmmac_dma_cfg *dma_cfg,
 				      dma_addr_t phy, u32 chan)
 {
@@ -131,7 +134,8 @@ static void dwxgmac2_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi)
 	writel(XGMAC_RDPS, ioaddr + XGMAC_RX_EDMA_CTRL);
 }
 
-static void dwxgmac2_dma_dump_regs(void __iomem *ioaddr, u32 *reg_space)
+static void dwxgmac2_dma_dump_regs(struct stmmac_priv *priv,
+				   void __iomem *ioaddr, u32 *reg_space)
 {
 	int i;
 
@@ -139,8 +143,8 @@ static void dwxgmac2_dma_dump_regs(void __iomem *ioaddr, u32 *reg_space)
 		reg_space[i] = readl(ioaddr + i * 4);
 }
 
-static void dwxgmac2_dma_rx_mode(void __iomem *ioaddr, int mode,
-				 u32 channel, int fifosz, u8 qmode)
+static void dwxgmac2_dma_rx_mode(struct stmmac_priv *priv, void __iomem *ioaddr,
+				 int mode, u32 channel, int fifosz, u8 qmode)
 {
 	u32 value = readl(ioaddr + XGMAC_MTL_RXQ_OPMODE(channel));
 	unsigned int rqs = fifosz / 256 - 1;
@@ -205,8 +209,8 @@ static void dwxgmac2_dma_rx_mode(void __iomem *ioaddr, int mode,
 	writel(value | XGMAC_RXOIE, ioaddr + XGMAC_MTL_QINTEN(channel));
 }
 
-static void dwxgmac2_dma_tx_mode(void __iomem *ioaddr, int mode,
-				 u32 channel, int fifosz, u8 qmode)
+static void dwxgmac2_dma_tx_mode(struct stmmac_priv *priv, void __iomem *ioaddr,
+				 int mode, u32 channel, int fifosz, u8 qmode)
 {
 	u32 value = readl(ioaddr + XGMAC_MTL_TXQ_OPMODE(channel));
 	unsigned int tqs = fifosz / 256 - 1;
@@ -248,7 +252,8 @@ static void dwxgmac2_dma_tx_mode(void __iomem *ioaddr, int mode,
 	writel(value, ioaddr +  XGMAC_MTL_TXQ_OPMODE(channel));
 }
 
-static void dwxgmac2_enable_dma_irq(void __iomem *ioaddr, u32 chan,
+static void dwxgmac2_enable_dma_irq(struct stmmac_priv *priv,
+				    void __iomem *ioaddr, u32 chan,
 				    bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + XGMAC_DMA_CH_INT_EN(chan));
@@ -261,7 +266,8 @@ static void dwxgmac2_enable_dma_irq(void __iomem *ioaddr, u32 chan,
 	writel(value, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
 }
 
-static void dwxgmac2_disable_dma_irq(void __iomem *ioaddr, u32 chan,
+static void dwxgmac2_disable_dma_irq(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 chan,
 				     bool rx, bool tx)
 {
 	u32 value = readl(ioaddr + XGMAC_DMA_CH_INT_EN(chan));
@@ -274,7 +280,8 @@ static void dwxgmac2_disable_dma_irq(void __iomem *ioaddr, u32 chan,
 	writel(value, ioaddr + XGMAC_DMA_CH_INT_EN(chan));
 }
 
-static void dwxgmac2_dma_start_tx(void __iomem *ioaddr, u32 chan)
+static void dwxgmac2_dma_start_tx(struct stmmac_priv *priv,
+				  void __iomem *ioaddr, u32 chan)
 {
 	u32 value;
 
@@ -287,7 +294,8 @@ static void dwxgmac2_dma_start_tx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + XGMAC_TX_CONFIG);
 }
 
-static void dwxgmac2_dma_stop_tx(void __iomem *ioaddr, u32 chan)
+static void dwxgmac2_dma_stop_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
+				 u32 chan)
 {
 	u32 value;
 
@@ -300,7 +308,8 @@ static void dwxgmac2_dma_stop_tx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + XGMAC_TX_CONFIG);
 }
 
-static void dwxgmac2_dma_start_rx(void __iomem *ioaddr, u32 chan)
+static void dwxgmac2_dma_start_rx(struct stmmac_priv *priv,
+				  void __iomem *ioaddr, u32 chan)
 {
 	u32 value;
 
@@ -313,7 +322,8 @@ static void dwxgmac2_dma_start_rx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + XGMAC_RX_CONFIG);
 }
 
-static void dwxgmac2_dma_stop_rx(void __iomem *ioaddr, u32 chan)
+static void dwxgmac2_dma_stop_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
+				 u32 chan)
 {
 	u32 value;
 
@@ -322,7 +332,8 @@ static void dwxgmac2_dma_stop_rx(void __iomem *ioaddr, u32 chan)
 	writel(value, ioaddr + XGMAC_DMA_CH_RX_CONTROL(chan));
 }
 
-static int dwxgmac2_dma_interrupt(void __iomem *ioaddr,
+static int dwxgmac2_dma_interrupt(struct stmmac_priv *priv,
+				  void __iomem *ioaddr,
 				  struct stmmac_extra_stats *x, u32 chan,
 				  u32 dir)
 {
@@ -449,32 +460,38 @@ static int dwxgmac2_get_hw_feature(void __iomem *ioaddr,
 	return 0;
 }
 
-static void dwxgmac2_rx_watchdog(void __iomem *ioaddr, u32 riwt, u32 queue)
+static void dwxgmac2_rx_watchdog(struct stmmac_priv *priv, void __iomem *ioaddr,
+				 u32 riwt, u32 queue)
 {
 	writel(riwt & XGMAC_RWT, ioaddr + XGMAC_DMA_CH_Rx_WATCHDOG(queue));
 }
 
-static void dwxgmac2_set_rx_ring_len(void __iomem *ioaddr, u32 len, u32 chan)
+static void dwxgmac2_set_rx_ring_len(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 len, u32 chan)
 {
 	writel(len, ioaddr + XGMAC_DMA_CH_RxDESC_RING_LEN(chan));
 }
 
-static void dwxgmac2_set_tx_ring_len(void __iomem *ioaddr, u32 len, u32 chan)
+static void dwxgmac2_set_tx_ring_len(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 len, u32 chan)
 {
 	writel(len, ioaddr + XGMAC_DMA_CH_TxDESC_RING_LEN(chan));
 }
 
-static void dwxgmac2_set_rx_tail_ptr(void __iomem *ioaddr, u32 ptr, u32 chan)
+static void dwxgmac2_set_rx_tail_ptr(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 ptr, u32 chan)
 {
 	writel(ptr, ioaddr + XGMAC_DMA_CH_RxDESC_TAIL_LPTR(chan));
 }
 
-static void dwxgmac2_set_tx_tail_ptr(void __iomem *ioaddr, u32 ptr, u32 chan)
+static void dwxgmac2_set_tx_tail_ptr(struct stmmac_priv *priv,
+				     void __iomem *ioaddr, u32 ptr, u32 chan)
 {
 	writel(ptr, ioaddr + XGMAC_DMA_CH_TxDESC_TAIL_LPTR(chan));
 }
 
-static void dwxgmac2_enable_tso(void __iomem *ioaddr, bool en, u32 chan)
+static void dwxgmac2_enable_tso(struct stmmac_priv *priv, void __iomem *ioaddr,
+				bool en, u32 chan)
 {
 	u32 value = readl(ioaddr + XGMAC_DMA_CH_TX_CONTROL(chan));
 
@@ -486,7 +503,8 @@ static void dwxgmac2_enable_tso(void __iomem *ioaddr, bool en, u32 chan)
 	writel(value, ioaddr + XGMAC_DMA_CH_TX_CONTROL(chan));
 }
 
-static void dwxgmac2_qmode(void __iomem *ioaddr, u32 channel, u8 qmode)
+static void dwxgmac2_qmode(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   u32 channel, u8 qmode)
 {
 	u32 value = readl(ioaddr + XGMAC_MTL_TXQ_OPMODE(channel));
 	u32 flow = readl(ioaddr + XGMAC_RX_FLOW_CTRL);
@@ -503,7 +521,8 @@ static void dwxgmac2_qmode(void __iomem *ioaddr, u32 channel, u8 qmode)
 	writel(value, ioaddr +  XGMAC_MTL_TXQ_OPMODE(channel));
 }
 
-static void dwxgmac2_set_bfsize(void __iomem *ioaddr, int bfsize, u32 chan)
+static void dwxgmac2_set_bfsize(struct stmmac_priv *priv, void __iomem *ioaddr,
+				int bfsize, u32 chan)
 {
 	u32 value;
 
@@ -513,7 +532,8 @@ static void dwxgmac2_set_bfsize(void __iomem *ioaddr, int bfsize, u32 chan)
 	writel(value, ioaddr + XGMAC_DMA_CH_RX_CONTROL(chan));
 }
 
-static void dwxgmac2_enable_sph(void __iomem *ioaddr, bool en, u32 chan)
+static void dwxgmac2_enable_sph(struct stmmac_priv *priv, void __iomem *ioaddr,
+				bool en, u32 chan)
 {
 	u32 value = readl(ioaddr + XGMAC_RX_CONFIG);
 
@@ -529,7 +549,8 @@ static void dwxgmac2_enable_sph(void __iomem *ioaddr, bool en, u32 chan)
 	writel(value, ioaddr + XGMAC_DMA_CH_CONTROL(chan));
 }
 
-static int dwxgmac2_enable_tbs(void __iomem *ioaddr, bool en, u32 chan)
+static int dwxgmac2_enable_tbs(struct stmmac_priv *priv, void __iomem *ioaddr,
+			       bool en, u32 chan)
 {
 	u32 value = readl(ioaddr + XGMAC_DMA_CH_TX_CONTROL(chan));
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 7f906ef3ea4f..6ee7cf07cfd7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -26,6 +26,7 @@
 })
 
 struct stmmac_extra_stats;
+struct stmmac_priv;
 struct stmmac_safety_stats;
 struct dma_desc;
 struct dma_extended_desc;
@@ -171,109 +172,125 @@ struct stmmac_dma_ops {
 	int (*reset)(void __iomem *ioaddr);
 	void (*init)(void __iomem *ioaddr, struct stmmac_dma_cfg *dma_cfg,
 		     int atds);
-	void (*init_chan)(void __iomem *ioaddr,
+	void (*init_chan)(struct stmmac_priv *priv, void __iomem *ioaddr,
 			  struct stmmac_dma_cfg *dma_cfg, u32 chan);
-	void (*init_rx_chan)(void __iomem *ioaddr,
+	void (*init_rx_chan)(struct stmmac_priv *priv, void __iomem *ioaddr,
 			     struct stmmac_dma_cfg *dma_cfg,
 			     dma_addr_t phy, u32 chan);
-	void (*init_tx_chan)(void __iomem *ioaddr,
+	void (*init_tx_chan)(struct stmmac_priv *priv, void __iomem *ioaddr,
 			     struct stmmac_dma_cfg *dma_cfg,
 			     dma_addr_t phy, u32 chan);
 	/* Configure the AXI Bus Mode Register */
 	void (*axi)(void __iomem *ioaddr, struct stmmac_axi *axi);
 	/* Dump DMA registers */
-	void (*dump_regs)(void __iomem *ioaddr, u32 *reg_space);
-	void (*dma_rx_mode)(void __iomem *ioaddr, int mode, u32 channel,
-			    int fifosz, u8 qmode);
-	void (*dma_tx_mode)(void __iomem *ioaddr, int mode, u32 channel,
+	void (*dump_regs)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			  u32 *reg_space);
+	void (*dma_rx_mode)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    int mode, u32 channel,
 			    int fifosz, u8 qmode);
+	void (*dma_tx_mode)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    int mode, u32 channel, int fifosz, u8 qmode);
 	/* To track extra statistic (if supported) */
 	void (*dma_diagnostic_fr)(struct net_device_stats *stats,
 				  struct stmmac_extra_stats *x,
 				  void __iomem *ioaddr);
 	void (*enable_dma_transmission) (void __iomem *ioaddr);
-	void (*enable_dma_irq)(void __iomem *ioaddr, u32 chan,
-			       bool rx, bool tx);
-	void (*disable_dma_irq)(void __iomem *ioaddr, u32 chan,
-				bool rx, bool tx);
-	void (*start_tx)(void __iomem *ioaddr, u32 chan);
-	void (*stop_tx)(void __iomem *ioaddr, u32 chan);
-	void (*start_rx)(void __iomem *ioaddr, u32 chan);
-	void (*stop_rx)(void __iomem *ioaddr, u32 chan);
-	int (*dma_interrupt) (void __iomem *ioaddr,
-			      struct stmmac_extra_stats *x, u32 chan, u32 dir);
+	void (*enable_dma_irq)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			       u32 chan, bool rx, bool tx);
+	void (*disable_dma_irq)(struct stmmac_priv *priv, void __iomem *ioaddr,
+				u32 chan, bool rx, bool tx);
+	void (*start_tx)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 chan);
+	void (*stop_tx)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan);
+	void (*start_rx)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			 u32 chan);
+	void (*stop_rx)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			u32 chan);
+	int (*dma_interrupt)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			     struct stmmac_extra_stats *x, u32 chan, u32 dir);
 	/* If supported then get the optional core features */
 	int (*get_hw_feature)(void __iomem *ioaddr,
 			      struct dma_features *dma_cap);
 	/* Program the HW RX Watchdog */
-	void (*rx_watchdog)(void __iomem *ioaddr, u32 riwt, u32 queue);
-	void (*set_tx_ring_len)(void __iomem *ioaddr, u32 len, u32 chan);
-	void (*set_rx_ring_len)(void __iomem *ioaddr, u32 len, u32 chan);
-	void (*set_rx_tail_ptr)(void __iomem *ioaddr, u32 tail_ptr, u32 chan);
-	void (*set_tx_tail_ptr)(void __iomem *ioaddr, u32 tail_ptr, u32 chan);
-	void (*enable_tso)(void __iomem *ioaddr, bool en, u32 chan);
-	void (*qmode)(void __iomem *ioaddr, u32 channel, u8 qmode);
-	void (*set_bfsize)(void __iomem *ioaddr, int bfsize, u32 chan);
-	void (*enable_sph)(void __iomem *ioaddr, bool en, u32 chan);
-	int (*enable_tbs)(void __iomem *ioaddr, bool en, u32 chan);
+	void (*rx_watchdog)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			    u32 riwt, u32 queue);
+	void (*set_tx_ring_len)(struct stmmac_priv *priv, void __iomem *ioaddr,
+				u32 len, u32 chan);
+	void (*set_rx_ring_len)(struct stmmac_priv *priv, void __iomem *ioaddr,
+				u32 len, u32 chan);
+	void (*set_rx_tail_ptr)(struct stmmac_priv *priv, void __iomem *ioaddr,
+				u32 tail_ptr, u32 chan);
+	void (*set_tx_tail_ptr)(struct stmmac_priv *priv, void __iomem *ioaddr,
+				u32 tail_ptr, u32 chan);
+	void (*enable_tso)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   bool en, u32 chan);
+	void (*qmode)(struct stmmac_priv *priv, void __iomem *ioaddr,
+		      u32 channel, u8 qmode);
+	void (*set_bfsize)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   int bfsize, u32 chan);
+	void (*enable_sph)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			   bool en, u32 chan);
+	int (*enable_tbs)(struct stmmac_priv *priv, void __iomem *ioaddr,
+			  bool en, u32 chan);
 };
 
 #define stmmac_dma_init(__priv, __args...) \
 	stmmac_do_void_callback(__priv, dma, init, __args)
 #define stmmac_init_chan(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, init_chan, __args)
+	stmmac_do_void_callback(__priv, dma, init_chan, __priv, __args)
 #define stmmac_init_rx_chan(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, init_rx_chan, __args)
+	stmmac_do_void_callback(__priv, dma, init_rx_chan, __priv, __args)
 #define stmmac_init_tx_chan(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, init_tx_chan, __args)
+	stmmac_do_void_callback(__priv, dma, init_tx_chan, __priv, __args)
 #define stmmac_axi(__priv, __args...) \
 	stmmac_do_void_callback(__priv, dma, axi, __args)
 #define stmmac_dump_dma_regs(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, dump_regs, __args)
+	stmmac_do_void_callback(__priv, dma, dump_regs, __priv, __args)
 #define stmmac_dma_rx_mode(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, dma_rx_mode, __args)
+	stmmac_do_void_callback(__priv, dma, dma_rx_mode, __priv, __args)
 #define stmmac_dma_tx_mode(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, dma_tx_mode, __args)
+	stmmac_do_void_callback(__priv, dma, dma_tx_mode, __priv, __args)
 #define stmmac_dma_diagnostic_fr(__priv, __args...) \
 	stmmac_do_void_callback(__priv, dma, dma_diagnostic_fr, __args)
 #define stmmac_enable_dma_transmission(__priv, __args...) \
 	stmmac_do_void_callback(__priv, dma, enable_dma_transmission, __args)
 #define stmmac_enable_dma_irq(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, enable_dma_irq, __args)
+	stmmac_do_void_callback(__priv, dma, enable_dma_irq, __priv, __args)
 #define stmmac_disable_dma_irq(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, disable_dma_irq, __args)
+	stmmac_do_void_callback(__priv, dma, disable_dma_irq, __priv, __args)
 #define stmmac_start_tx(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, start_tx, __args)
+	stmmac_do_void_callback(__priv, dma, start_tx, __priv, __args)
 #define stmmac_stop_tx(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, stop_tx, __args)
+	stmmac_do_void_callback(__priv, dma, stop_tx, __priv, __args)
 #define stmmac_start_rx(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, start_rx, __args)
+	stmmac_do_void_callback(__priv, dma, start_rx, __priv, __args)
 #define stmmac_stop_rx(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, stop_rx, __args)
+	stmmac_do_void_callback(__priv, dma, stop_rx, __priv, __args)
 #define stmmac_dma_interrupt_status(__priv, __args...) \
-	stmmac_do_callback(__priv, dma, dma_interrupt, __args)
+	stmmac_do_callback(__priv, dma, dma_interrupt, __priv, __args)
 #define stmmac_get_hw_feature(__priv, __args...) \
 	stmmac_do_callback(__priv, dma, get_hw_feature, __args)
 #define stmmac_rx_watchdog(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, rx_watchdog, __args)
+	stmmac_do_void_callback(__priv, dma, rx_watchdog, __priv, __args)
 #define stmmac_set_tx_ring_len(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, set_tx_ring_len, __args)
+	stmmac_do_void_callback(__priv, dma, set_tx_ring_len, __priv, __args)
 #define stmmac_set_rx_ring_len(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, set_rx_ring_len, __args)
+	stmmac_do_void_callback(__priv, dma, set_rx_ring_len, __priv, __args)
 #define stmmac_set_rx_tail_ptr(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, set_rx_tail_ptr, __args)
+	stmmac_do_void_callback(__priv, dma, set_rx_tail_ptr, __priv, __args)
 #define stmmac_set_tx_tail_ptr(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, set_tx_tail_ptr, __args)
+	stmmac_do_void_callback(__priv, dma, set_tx_tail_ptr, __priv, __args)
 #define stmmac_enable_tso(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, enable_tso, __args)
+	stmmac_do_void_callback(__priv, dma, enable_tso, __priv, __args)
 #define stmmac_dma_qmode(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, qmode, __args)
+	stmmac_do_void_callback(__priv, dma, qmode, __priv, __args)
 #define stmmac_set_dma_bfsize(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, set_bfsize, __args)
+	stmmac_do_void_callback(__priv, dma, set_bfsize, __priv, __args)
 #define stmmac_enable_sph(__priv, __args...) \
-	stmmac_do_void_callback(__priv, dma, enable_sph, __args)
+	stmmac_do_void_callback(__priv, dma, enable_sph, __priv, __args)
 #define stmmac_enable_tbs(__priv, __args...) \
-	stmmac_do_callback(__priv, dma, enable_tbs, __args)
+	stmmac_do_callback(__priv, dma, enable_tbs, __priv, __args)
 
 struct mac_device_info;
 struct net_device;
@@ -305,21 +322,23 @@ struct stmmac_ops {
 	/* Program TX Algorithms */
 	void (*prog_mtl_tx_algorithms)(struct mac_device_info *hw, u32 tx_alg);
 	/* Set MTL TX queues weight */
-	void (*set_mtl_tx_queue_weight)(struct mac_device_info *hw,
+	void (*set_mtl_tx_queue_weight)(struct stmmac_priv *priv,
+					struct mac_device_info *hw,
 					u32 weight, u32 queue);
 	/* RX MTL queue to RX dma mapping */
 	void (*map_mtl_to_dma)(struct mac_device_info *hw, u32 queue, u32 chan);
 	/* Configure AV Algorithm */
-	void (*config_cbs)(struct mac_device_info *hw, u32 send_slope,
-			   u32 idle_slope, u32 high_credit, u32 low_credit,
-			   u32 queue);
+	void (*config_cbs)(struct stmmac_priv *priv, struct mac_device_info *hw,
+			   u32 send_slope, u32 idle_slope, u32 high_credit,
+			   u32 low_credit, u32 queue);
 	/* Dump MAC registers */
 	void (*dump_regs)(struct mac_device_info *hw, u32 *reg_space);
 	/* Handle extra events on specific interrupts hw dependent */
 	int (*host_irq_status)(struct mac_device_info *hw,
 			       struct stmmac_extra_stats *x);
 	/* Handle MTL interrupts */
-	int (*host_mtl_irq_status)(struct mac_device_info *hw, u32 chan);
+	int (*host_mtl_irq_status)(struct stmmac_priv *priv,
+				   struct mac_device_info *hw, u32 chan);
 	/* Multicast filter setting */
 	void (*set_filter)(struct mac_device_info *hw, struct net_device *dev);
 	/* Flow control setting */
@@ -339,8 +358,9 @@ struct stmmac_ops {
 	void (*set_eee_lpi_entry_timer)(struct mac_device_info *hw, int et);
 	void (*set_eee_timer)(struct mac_device_info *hw, int ls, int tw);
 	void (*set_eee_pls)(struct mac_device_info *hw, int link);
-	void (*debug)(void __iomem *ioaddr, struct stmmac_extra_stats *x,
-		      u32 rx_queues, u32 tx_queues);
+	void (*debug)(struct stmmac_priv *priv, void __iomem *ioaddr,
+		      struct stmmac_extra_stats *x, u32 rx_queues,
+		      u32 tx_queues);
 	/* PCS calls */
 	void (*pcs_ctrl_ane)(void __iomem *ioaddr, bool ane, bool srgmi_ral,
 			     bool loopback);
@@ -420,17 +440,17 @@ struct stmmac_ops {
 #define stmmac_prog_mtl_tx_algorithms(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, prog_mtl_tx_algorithms, __args)
 #define stmmac_set_mtl_tx_queue_weight(__priv, __args...) \
-	stmmac_do_void_callback(__priv, mac, set_mtl_tx_queue_weight, __args)
+	stmmac_do_void_callback(__priv, mac, set_mtl_tx_queue_weight, __priv, __args)
 #define stmmac_map_mtl_to_dma(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, map_mtl_to_dma, __args)
 #define stmmac_config_cbs(__priv, __args...) \
-	stmmac_do_void_callback(__priv, mac, config_cbs, __args)
+	stmmac_do_void_callback(__priv, mac, config_cbs, __priv, __args)
 #define stmmac_dump_mac_regs(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, dump_regs, __args)
 #define stmmac_host_irq_status(__priv, __args...) \
 	stmmac_do_callback(__priv, mac, host_irq_status, __args)
 #define stmmac_host_mtl_irq_status(__priv, __args...) \
-	stmmac_do_callback(__priv, mac, host_mtl_irq_status, __args)
+	stmmac_do_callback(__priv, mac, host_mtl_irq_status, __priv, __args)
 #define stmmac_set_filter(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, set_filter, __args)
 #define stmmac_flow_ctrl(__priv, __args...) \
@@ -452,11 +472,11 @@ struct stmmac_ops {
 #define stmmac_set_eee_pls(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, set_eee_pls, __args)
 #define stmmac_mac_debug(__priv, __args...) \
-	stmmac_do_void_callback(__priv, mac, debug, __args)
+	stmmac_do_void_callback(__priv, mac, debug, __priv, __args)
 #define stmmac_pcs_ctrl_ane(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, pcs_ctrl_ane, __args)
 #define stmmac_pcs_rane(__priv, __args...) \
-	stmmac_do_void_callback(__priv, mac, pcs_rane, __args)
+	stmmac_do_void_callback(__priv, mac, pcs_rane, __priv, __args)
 #define stmmac_pcs_get_adv_lp(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, pcs_get_adv_lp, __args)
 #define stmmac_safety_feat_config(__priv, __args...) \
@@ -504,8 +524,6 @@ struct stmmac_ops {
 #define stmmac_fpe_irq_status(__priv, __args...) \
 	stmmac_do_callback(__priv, mac, fpe_irq_status, __args)
 
-struct stmmac_priv;
-
 /* PTP and HW Timer helpers */
 struct stmmac_hwtimestamp {
 	void (*config_hw_tstamping) (void __iomem *ioaddr, u32 data);
-- 
cgit v1.2.3-70-g09d2


From 33719b57f52e5b761234373f98f55f4e036d61c9 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:06 -0500
Subject: net: stmmac: dwmac4: Allow platforms to specify some DMA/MTL offsets

Some platforms have dwmac4 implementations that have a different
address space layout than the default, resulting in the need to define
their own DMA/MTL offsets.

Extend the functions to allow a platform driver to indicate what its
addresses are, overriding the defaults.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac4.h      | 101 +++++++++++---
 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c |  36 +++--
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c  | 157 +++++++++++++---------
 drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h  |  54 +++++---
 drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c  |  67 +++++----
 include/linux/stmmac.h                            |  19 +++
 6 files changed, 293 insertions(+), 141 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
index ccd49346d3b3..4538f334df57 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h
@@ -336,14 +336,25 @@ enum power_event {
 
 #define MTL_CHAN_BASE_ADDR		0x00000d00
 #define MTL_CHAN_BASE_OFFSET		0x40
-#define MTL_CHANX_BASE_ADDR(x)		(MTL_CHAN_BASE_ADDR + \
-					(x * MTL_CHAN_BASE_OFFSET))
 
-#define MTL_CHAN_TX_OP_MODE(x)		MTL_CHANX_BASE_ADDR(x)
-#define MTL_CHAN_TX_DEBUG(x)		(MTL_CHANX_BASE_ADDR(x) + 0x8)
-#define MTL_CHAN_INT_CTRL(x)		(MTL_CHANX_BASE_ADDR(x) + 0x2c)
-#define MTL_CHAN_RX_OP_MODE(x)		(MTL_CHANX_BASE_ADDR(x) + 0x30)
-#define MTL_CHAN_RX_DEBUG(x)		(MTL_CHANX_BASE_ADDR(x) + 0x38)
+static inline u32 mtl_chanx_base_addr(const struct dwmac4_addrs *addrs,
+				      const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->mtl_chan + (x * addrs->mtl_chan_offset);
+	else
+		addr = MTL_CHAN_BASE_ADDR + (x * MTL_CHAN_BASE_OFFSET);
+
+	return addr;
+}
+
+#define MTL_CHAN_TX_OP_MODE(addrs, x)	mtl_chanx_base_addr(addrs, x)
+#define MTL_CHAN_TX_DEBUG(addrs, x)	(mtl_chanx_base_addr(addrs, x) + 0x8)
+#define MTL_CHAN_INT_CTRL(addrs, x)	(mtl_chanx_base_addr(addrs, x) + 0x2c)
+#define MTL_CHAN_RX_OP_MODE(addrs, x)	(mtl_chanx_base_addr(addrs, x) + 0x30)
+#define MTL_CHAN_RX_DEBUG(addrs, x)	(mtl_chanx_base_addr(addrs, x) + 0x38)
 
 #define MTL_OP_MODE_RSF			BIT(5)
 #define MTL_OP_MODE_TXQEN_MASK		GENMASK(3, 2)
@@ -388,8 +399,19 @@ enum power_event {
 /* MTL ETS Control register */
 #define MTL_ETS_CTRL_BASE_ADDR		0x00000d10
 #define MTL_ETS_CTRL_BASE_OFFSET	0x40
-#define MTL_ETSX_CTRL_BASE_ADDR(x)	(MTL_ETS_CTRL_BASE_ADDR + \
-					((x) * MTL_ETS_CTRL_BASE_OFFSET))
+
+static inline u32 mtl_etsx_ctrl_base_addr(const struct dwmac4_addrs *addrs,
+					  const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->mtl_ets_ctrl + (x * addrs->mtl_ets_ctrl_offset);
+	else
+		addr = MTL_ETS_CTRL_BASE_ADDR + (x * MTL_ETS_CTRL_BASE_OFFSET);
+
+	return addr;
+}
 
 #define MTL_ETS_CTRL_CC			BIT(3)
 #define MTL_ETS_CTRL_AVALG		BIT(2)
@@ -397,31 +419,76 @@ enum power_event {
 /* MTL Queue Quantum Weight */
 #define MTL_TXQ_WEIGHT_BASE_ADDR	0x00000d18
 #define MTL_TXQ_WEIGHT_BASE_OFFSET	0x40
-#define MTL_TXQX_WEIGHT_BASE_ADDR(x)	(MTL_TXQ_WEIGHT_BASE_ADDR + \
-					((x) * MTL_TXQ_WEIGHT_BASE_OFFSET))
+
+static inline u32 mtl_txqx_weight_base_addr(const struct dwmac4_addrs *addrs,
+					    const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->mtl_txq_weight + (x * addrs->mtl_txq_weight_offset);
+	else
+		addr = MTL_TXQ_WEIGHT_BASE_ADDR + (x * MTL_TXQ_WEIGHT_BASE_OFFSET);
+
+	return addr;
+}
+
 #define MTL_TXQ_WEIGHT_ISCQW_MASK	GENMASK(20, 0)
 
 /* MTL sendSlopeCredit register */
 #define MTL_SEND_SLP_CRED_BASE_ADDR	0x00000d1c
 #define MTL_SEND_SLP_CRED_OFFSET	0x40
-#define MTL_SEND_SLP_CREDX_BASE_ADDR(x)	(MTL_SEND_SLP_CRED_BASE_ADDR + \
-					((x) * MTL_SEND_SLP_CRED_OFFSET))
+
+static inline u32 mtl_send_slp_credx_base_addr(const struct dwmac4_addrs *addrs,
+					       const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->mtl_send_slp_cred + (x * addrs->mtl_send_slp_cred_offset);
+	else
+		addr = MTL_SEND_SLP_CRED_BASE_ADDR + (x * MTL_SEND_SLP_CRED_OFFSET);
+
+	return addr;
+}
 
 #define MTL_SEND_SLP_CRED_SSC_MASK	GENMASK(13, 0)
 
 /* MTL hiCredit register */
 #define MTL_HIGH_CRED_BASE_ADDR		0x00000d20
 #define MTL_HIGH_CRED_OFFSET		0x40
-#define MTL_HIGH_CREDX_BASE_ADDR(x)	(MTL_HIGH_CRED_BASE_ADDR + \
-					((x) * MTL_HIGH_CRED_OFFSET))
+
+static inline u32 mtl_high_credx_base_addr(const struct dwmac4_addrs *addrs,
+					   const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->mtl_high_cred + (x * addrs->mtl_high_cred_offset);
+	else
+		addr = MTL_HIGH_CRED_BASE_ADDR + (x * MTL_HIGH_CRED_OFFSET);
+
+	return addr;
+}
 
 #define MTL_HIGH_CRED_HC_MASK		GENMASK(28, 0)
 
 /* MTL loCredit register */
 #define MTL_LOW_CRED_BASE_ADDR		0x00000d24
 #define MTL_LOW_CRED_OFFSET		0x40
-#define MTL_LOW_CREDX_BASE_ADDR(x)	(MTL_LOW_CRED_BASE_ADDR + \
-					((x) * MTL_LOW_CRED_OFFSET))
+
+static inline u32 mtl_low_credx_base_addr(const struct dwmac4_addrs *addrs,
+					  const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->mtl_low_cred + (x * addrs->mtl_low_cred_offset);
+	else
+		addr = MTL_LOW_CRED_BASE_ADDR + (x * MTL_LOW_CRED_OFFSET);
+
+	return addr;
+}
 
 #define MTL_HIGH_CRED_LC_MASK		GENMASK(28, 0)
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
index f44180519638..afaec3fb9ab6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
@@ -202,12 +202,14 @@ static void dwmac4_set_mtl_tx_queue_weight(struct stmmac_priv *priv,
 					   struct mac_device_info *hw,
 					   u32 weight, u32 queue)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	void __iomem *ioaddr = hw->pcsr;
-	u32 value = readl(ioaddr + MTL_TXQX_WEIGHT_BASE_ADDR(queue));
+	u32 value = readl(ioaddr + mtl_txqx_weight_base_addr(dwmac4_addrs,
+							     queue));
 
 	value &= ~MTL_TXQ_WEIGHT_ISCQW_MASK;
 	value |= weight & MTL_TXQ_WEIGHT_ISCQW_MASK;
-	writel(value, ioaddr + MTL_TXQX_WEIGHT_BASE_ADDR(queue));
+	writel(value, ioaddr + mtl_txqx_weight_base_addr(dwmac4_addrs, queue));
 }
 
 static void dwmac4_map_mtl_dma(struct mac_device_info *hw, u32 queue, u32 chan)
@@ -233,6 +235,7 @@ static void dwmac4_config_cbs(struct stmmac_priv *priv,
 			      u32 send_slope, u32 idle_slope,
 			      u32 high_credit, u32 low_credit, u32 queue)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	void __iomem *ioaddr = hw->pcsr;
 	u32 value;
 
@@ -243,31 +246,33 @@ static void dwmac4_config_cbs(struct stmmac_priv *priv,
 	pr_debug("\tlow_credit: 0x%08x\n", low_credit);
 
 	/* enable AV algorithm */
-	value = readl(ioaddr + MTL_ETSX_CTRL_BASE_ADDR(queue));
+	value = readl(ioaddr + mtl_etsx_ctrl_base_addr(dwmac4_addrs, queue));
 	value |= MTL_ETS_CTRL_AVALG;
 	value |= MTL_ETS_CTRL_CC;
-	writel(value, ioaddr + MTL_ETSX_CTRL_BASE_ADDR(queue));
+	writel(value, ioaddr + mtl_etsx_ctrl_base_addr(dwmac4_addrs, queue));
 
 	/* configure send slope */
-	value = readl(ioaddr + MTL_SEND_SLP_CREDX_BASE_ADDR(queue));
+	value = readl(ioaddr + mtl_send_slp_credx_base_addr(dwmac4_addrs,
+							    queue));
 	value &= ~MTL_SEND_SLP_CRED_SSC_MASK;
 	value |= send_slope & MTL_SEND_SLP_CRED_SSC_MASK;
-	writel(value, ioaddr + MTL_SEND_SLP_CREDX_BASE_ADDR(queue));
+	writel(value, ioaddr + mtl_send_slp_credx_base_addr(dwmac4_addrs,
+							    queue));
 
 	/* configure idle slope (same register as tx weight) */
 	dwmac4_set_mtl_tx_queue_weight(priv, hw, idle_slope, queue);
 
 	/* configure high credit */
-	value = readl(ioaddr + MTL_HIGH_CREDX_BASE_ADDR(queue));
+	value = readl(ioaddr + mtl_high_credx_base_addr(dwmac4_addrs, queue));
 	value &= ~MTL_HIGH_CRED_HC_MASK;
 	value |= high_credit & MTL_HIGH_CRED_HC_MASK;
-	writel(value, ioaddr + MTL_HIGH_CREDX_BASE_ADDR(queue));
+	writel(value, ioaddr + mtl_high_credx_base_addr(dwmac4_addrs, queue));
 
 	/* configure high credit */
-	value = readl(ioaddr + MTL_LOW_CREDX_BASE_ADDR(queue));
+	value = readl(ioaddr + mtl_low_credx_base_addr(dwmac4_addrs, queue));
 	value &= ~MTL_HIGH_CRED_LC_MASK;
 	value |= low_credit & MTL_HIGH_CRED_LC_MASK;
-	writel(value, ioaddr + MTL_LOW_CREDX_BASE_ADDR(queue));
+	writel(value, ioaddr + mtl_low_credx_base_addr(dwmac4_addrs, queue));
 }
 
 static void dwmac4_dump_regs(struct mac_device_info *hw, u32 *reg_space)
@@ -764,6 +769,7 @@ static void dwmac4_phystatus(void __iomem *ioaddr, struct stmmac_extra_stats *x)
 static int dwmac4_irq_mtl_status(struct stmmac_priv *priv,
 				 struct mac_device_info *hw, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	void __iomem *ioaddr = hw->pcsr;
 	u32 mtl_int_qx_status;
 	int ret = 0;
@@ -773,12 +779,13 @@ static int dwmac4_irq_mtl_status(struct stmmac_priv *priv,
 	/* Check MTL Interrupt */
 	if (mtl_int_qx_status & MTL_INT_QX(chan)) {
 		/* read Queue x Interrupt status */
-		u32 status = readl(ioaddr + MTL_CHAN_INT_CTRL(chan));
+		u32 status = readl(ioaddr + MTL_CHAN_INT_CTRL(dwmac4_addrs,
+							      chan));
 
 		if (status & MTL_RX_OVERFLOW_INT) {
 			/*  clear Interrupt */
 			writel(status | MTL_RX_OVERFLOW_INT,
-			       ioaddr + MTL_CHAN_INT_CTRL(chan));
+			       ioaddr + MTL_CHAN_INT_CTRL(dwmac4_addrs, chan));
 			ret = CORE_IRQ_MTL_RX_OVERFLOW;
 		}
 	}
@@ -840,11 +847,12 @@ static void dwmac4_debug(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 struct stmmac_extra_stats *x,
 			 u32 rx_queues, u32 tx_queues)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value;
 	u32 queue;
 
 	for (queue = 0; queue < tx_queues; queue++) {
-		value = readl(ioaddr + MTL_CHAN_TX_DEBUG(queue));
+		value = readl(ioaddr + MTL_CHAN_TX_DEBUG(dwmac4_addrs, queue));
 
 		if (value & MTL_DEBUG_TXSTSFSTS)
 			x->mtl_tx_status_fifo_full++;
@@ -869,7 +877,7 @@ static void dwmac4_debug(struct stmmac_priv *priv, void __iomem *ioaddr,
 	}
 
 	for (queue = 0; queue < rx_queues; queue++) {
-		value = readl(ioaddr + MTL_CHAN_RX_DEBUG(queue));
+		value = readl(ioaddr + MTL_CHAN_RX_DEBUG(dwmac4_addrs, queue));
 
 		if (value & MTL_DEBUG_RXFSTS_MASK) {
 			u32 rxfsts = (value & MTL_DEBUG_RXFSTS_MASK)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
index 6f255d12f60f..84d3a8551b03 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c
@@ -13,6 +13,7 @@
 #include <linux/io.h>
 #include "dwmac4.h"
 #include "dwmac4_dma.h"
+#include "stmmac.h"
 
 static void dwmac4_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi)
 {
@@ -73,18 +74,20 @@ static void dwmac4_dma_init_rx_chan(struct stmmac_priv *priv,
 				    struct stmmac_dma_cfg *dma_cfg,
 				    dma_addr_t dma_rx_phy, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value;
 	u32 rxpbl = dma_cfg->rxpbl ?: dma_cfg->pbl;
 
-	value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	value = readl(ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 	value = value | (rxpbl << DMA_BUS_MODE_RPBL_SHIFT);
-	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 
 	if (IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT) && likely(dma_cfg->eame))
 		writel(upper_32_bits(dma_rx_phy),
-		       ioaddr + DMA_CHAN_RX_BASE_ADDR_HI(chan));
+		       ioaddr + DMA_CHAN_RX_BASE_ADDR_HI(dwmac4_addrs, chan));
 
-	writel(lower_32_bits(dma_rx_phy), ioaddr + DMA_CHAN_RX_BASE_ADDR(chan));
+	writel(lower_32_bits(dma_rx_phy),
+	       ioaddr + DMA_CHAN_RX_BASE_ADDR(dwmac4_addrs, chan));
 }
 
 static void dwmac4_dma_init_tx_chan(struct stmmac_priv *priv,
@@ -92,57 +95,61 @@ static void dwmac4_dma_init_tx_chan(struct stmmac_priv *priv,
 				    struct stmmac_dma_cfg *dma_cfg,
 				    dma_addr_t dma_tx_phy, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value;
 	u32 txpbl = dma_cfg->txpbl ?: dma_cfg->pbl;
 
-	value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 	value = value | (txpbl << DMA_BUS_MODE_PBL_SHIFT);
 
 	/* Enable OSP to get best performance */
 	value |= DMA_CONTROL_OSP;
 
-	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 
 	if (IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT) && likely(dma_cfg->eame))
 		writel(upper_32_bits(dma_tx_phy),
-		       ioaddr + DMA_CHAN_TX_BASE_ADDR_HI(chan));
+		       ioaddr + DMA_CHAN_TX_BASE_ADDR_HI(dwmac4_addrs, chan));
 
-	writel(lower_32_bits(dma_tx_phy), ioaddr + DMA_CHAN_TX_BASE_ADDR(chan));
+	writel(lower_32_bits(dma_tx_phy),
+	       ioaddr + DMA_CHAN_TX_BASE_ADDR(dwmac4_addrs, chan));
 }
 
 static void dwmac4_dma_init_channel(struct stmmac_priv *priv,
 				    void __iomem *ioaddr,
 				    struct stmmac_dma_cfg *dma_cfg, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value;
 
 	/* common channel control register config */
-	value = readl(ioaddr + DMA_CHAN_CONTROL(chan));
+	value = readl(ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, chan));
 	if (dma_cfg->pblx8)
 		value = value | DMA_BUS_MODE_PBL;
-	writel(value, ioaddr + DMA_CHAN_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, chan));
 
 	/* Mask interrupts by writing to CSR7 */
 	writel(DMA_CHAN_INTR_DEFAULT_MASK,
-	       ioaddr + DMA_CHAN_INTR_ENA(chan));
+	       ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 }
 
 static void dwmac410_dma_init_channel(struct stmmac_priv *priv,
 				      void __iomem *ioaddr,
 				      struct stmmac_dma_cfg *dma_cfg, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value;
 
 	/* common channel control register config */
-	value = readl(ioaddr + DMA_CHAN_CONTROL(chan));
+	value = readl(ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, chan));
 	if (dma_cfg->pblx8)
 		value = value | DMA_BUS_MODE_PBL;
 
-	writel(value, ioaddr + DMA_CHAN_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, chan));
 
 	/* Mask interrupts by writing to CSR7 */
 	writel(DMA_CHAN_INTR_DEFAULT_MASK_4_10,
-	       ioaddr + DMA_CHAN_INTR_ENA(chan));
+	       ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 }
 
 static void dwmac4_dma_init(void __iomem *ioaddr,
@@ -184,40 +191,46 @@ static void _dwmac4_dump_dma_regs(struct stmmac_priv *priv,
 				  void __iomem *ioaddr, u32 channel,
 				  u32 *reg_space)
 {
-	reg_space[DMA_CHAN_CONTROL(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_CONTROL(channel));
-	reg_space[DMA_CHAN_TX_CONTROL(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_TX_CONTROL(channel));
-	reg_space[DMA_CHAN_RX_CONTROL(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_RX_CONTROL(channel));
-	reg_space[DMA_CHAN_TX_BASE_ADDR(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_TX_BASE_ADDR(channel));
-	reg_space[DMA_CHAN_RX_BASE_ADDR(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_RX_BASE_ADDR(channel));
-	reg_space[DMA_CHAN_TX_END_ADDR(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_TX_END_ADDR(channel));
-	reg_space[DMA_CHAN_RX_END_ADDR(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_RX_END_ADDR(channel));
-	reg_space[DMA_CHAN_TX_RING_LEN(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_TX_RING_LEN(channel));
-	reg_space[DMA_CHAN_RX_RING_LEN(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_RX_RING_LEN(channel));
-	reg_space[DMA_CHAN_INTR_ENA(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_INTR_ENA(channel));
-	reg_space[DMA_CHAN_RX_WATCHDOG(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_RX_WATCHDOG(channel));
-	reg_space[DMA_CHAN_SLOT_CTRL_STATUS(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_SLOT_CTRL_STATUS(channel));
-	reg_space[DMA_CHAN_CUR_TX_DESC(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_CUR_TX_DESC(channel));
-	reg_space[DMA_CHAN_CUR_RX_DESC(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_CUR_RX_DESC(channel));
-	reg_space[DMA_CHAN_CUR_TX_BUF_ADDR(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_CUR_TX_BUF_ADDR(channel));
-	reg_space[DMA_CHAN_CUR_RX_BUF_ADDR(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_CUR_RX_BUF_ADDR(channel));
-	reg_space[DMA_CHAN_STATUS(channel) / 4] =
-		readl(ioaddr + DMA_CHAN_STATUS(channel));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	const struct dwmac4_addrs *default_addrs = NULL;
+
+	/* Purposely save the registers in the "normal" layout, regardless of
+	 * platform modifications, to keep reg_space size constant
+	 */
+	reg_space[DMA_CHAN_CONTROL(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_TX_CONTROL(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_RX_CONTROL(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_TX_BASE_ADDR(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_TX_BASE_ADDR(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_RX_BASE_ADDR(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_RX_BASE_ADDR(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_TX_END_ADDR(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_TX_END_ADDR(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_RX_END_ADDR(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_RX_END_ADDR(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_TX_RING_LEN(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_TX_RING_LEN(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_RX_RING_LEN(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_RX_RING_LEN(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_INTR_ENA(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_RX_WATCHDOG(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_RX_WATCHDOG(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_SLOT_CTRL_STATUS(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_SLOT_CTRL_STATUS(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_CUR_TX_DESC(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_CUR_TX_DESC(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_CUR_RX_DESC(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_CUR_RX_DESC(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_CUR_TX_BUF_ADDR(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_CUR_TX_BUF_ADDR(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_CUR_RX_BUF_ADDR(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_CUR_RX_BUF_ADDR(dwmac4_addrs, channel));
+	reg_space[DMA_CHAN_STATUS(default_addrs, channel) / 4] =
+		readl(ioaddr + DMA_CHAN_STATUS(dwmac4_addrs, channel));
 }
 
 static void dwmac4_dump_dma_regs(struct stmmac_priv *priv, void __iomem *ioaddr,
@@ -232,17 +245,20 @@ static void dwmac4_dump_dma_regs(struct stmmac_priv *priv, void __iomem *ioaddr,
 static void dwmac4_rx_watchdog(struct stmmac_priv *priv, void __iomem *ioaddr,
 			       u32 riwt, u32 queue)
 {
-	writel(riwt, ioaddr + DMA_CHAN_RX_WATCHDOG(queue));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	writel(riwt, ioaddr + DMA_CHAN_RX_WATCHDOG(dwmac4_addrs, queue));
 }
 
 static void dwmac4_dma_rx_chan_op_mode(struct stmmac_priv *priv,
 				       void __iomem *ioaddr, int mode,
 				       u32 channel, int fifosz, u8 qmode)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	unsigned int rqs = fifosz / 256 - 1;
 	u32 mtl_rx_op;
 
-	mtl_rx_op = readl(ioaddr + MTL_CHAN_RX_OP_MODE(channel));
+	mtl_rx_op = readl(ioaddr + MTL_CHAN_RX_OP_MODE(dwmac4_addrs, channel));
 
 	if (mode == SF_DMA_MODE) {
 		pr_debug("GMAC: enable RX store and forward mode\n");
@@ -300,14 +316,16 @@ static void dwmac4_dma_rx_chan_op_mode(struct stmmac_priv *priv,
 		mtl_rx_op |= rfa << MTL_OP_MODE_RFA_SHIFT;
 	}
 
-	writel(mtl_rx_op, ioaddr + MTL_CHAN_RX_OP_MODE(channel));
+	writel(mtl_rx_op, ioaddr + MTL_CHAN_RX_OP_MODE(dwmac4_addrs, channel));
 }
 
 static void dwmac4_dma_tx_chan_op_mode(struct stmmac_priv *priv,
 				       void __iomem *ioaddr, int mode,
 				       u32 channel, int fifosz, u8 qmode)
 {
-	u32 mtl_tx_op = readl(ioaddr + MTL_CHAN_TX_OP_MODE(channel));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 mtl_tx_op = readl(ioaddr + MTL_CHAN_TX_OP_MODE(dwmac4_addrs,
+							   channel));
 	unsigned int tqs = fifosz / 256 - 1;
 
 	if (mode == SF_DMA_MODE) {
@@ -353,7 +371,7 @@ static void dwmac4_dma_tx_chan_op_mode(struct stmmac_priv *priv,
 	mtl_tx_op &= ~MTL_OP_MODE_TQS_MASK;
 	mtl_tx_op |= tqs << MTL_OP_MODE_TQS_SHIFT;
 
-	writel(mtl_tx_op, ioaddr +  MTL_CHAN_TX_OP_MODE(channel));
+	writel(mtl_tx_op, ioaddr +  MTL_CHAN_TX_OP_MODE(dwmac4_addrs, channel));
 }
 
 static int dwmac4_get_hw_feature(void __iomem *ioaddr,
@@ -454,25 +472,28 @@ static int dwmac4_get_hw_feature(void __iomem *ioaddr,
 static void dwmac4_enable_tso(struct stmmac_priv *priv, void __iomem *ioaddr,
 			      bool en, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value;
 
 	if (en) {
 		/* enable TSO */
-		value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+		value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 		writel(value | DMA_CONTROL_TSE,
-		       ioaddr + DMA_CHAN_TX_CONTROL(chan));
+		       ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 	} else {
 		/* enable TSO */
-		value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+		value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 		writel(value & ~DMA_CONTROL_TSE,
-		       ioaddr + DMA_CHAN_TX_CONTROL(chan));
+		       ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 	}
 }
 
 static void dwmac4_qmode(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 u32 channel, u8 qmode)
 {
-	u32 mtl_tx_op = readl(ioaddr + MTL_CHAN_TX_OP_MODE(channel));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 mtl_tx_op = readl(ioaddr + MTL_CHAN_TX_OP_MODE(dwmac4_addrs,
+							   channel));
 
 	mtl_tx_op &= ~MTL_OP_MODE_TXQEN_MASK;
 	if (qmode != MTL_QUEUE_AVB)
@@ -480,50 +501,54 @@ static void dwmac4_qmode(struct stmmac_priv *priv, void __iomem *ioaddr,
 	else
 		mtl_tx_op |= MTL_OP_MODE_TXQEN_AV;
 
-	writel(mtl_tx_op, ioaddr +  MTL_CHAN_TX_OP_MODE(channel));
+	writel(mtl_tx_op, ioaddr +  MTL_CHAN_TX_OP_MODE(dwmac4_addrs, channel));
 }
 
 static void dwmac4_set_bfsize(struct stmmac_priv *priv, void __iomem *ioaddr,
 			      int bfsize, u32 chan)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 
 	value &= ~DMA_RBSZ_MASK;
 	value |= (bfsize << DMA_RBSZ_SHIFT) & DMA_RBSZ_MASK;
 
-	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 }
 
 static void dwmac4_enable_sph(struct stmmac_priv *priv, void __iomem *ioaddr,
 			      bool en, u32 chan)
 {
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
 	u32 value = readl(ioaddr + GMAC_EXT_CONFIG);
 
 	value &= ~GMAC_CONFIG_HDSMS;
 	value |= GMAC_CONFIG_HDSMS_256; /* Segment max 256 bytes */
 	writel(value, ioaddr + GMAC_EXT_CONFIG);
 
-	value = readl(ioaddr + DMA_CHAN_CONTROL(chan));
+	value = readl(ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, chan));
 	if (en)
 		value |= DMA_CONTROL_SPH;
 	else
 		value &= ~DMA_CONTROL_SPH;
-	writel(value, ioaddr + DMA_CHAN_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_CONTROL(dwmac4_addrs, chan));
 }
 
 static int dwmac4_enable_tbs(struct stmmac_priv *priv, void __iomem *ioaddr,
 			     bool en, u32 chan)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 
 	if (en)
 		value |= DMA_CONTROL_EDSE;
 	else
 		value &= ~DMA_CONTROL_EDSE;
 
-	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 
-	value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan)) & DMA_CONTROL_EDSE;
+	value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs,
+						   chan)) & DMA_CONTROL_EDSE;
 	if (en && !value)
 		return -EIO;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
index 740c3bc8d9a0..358e7dcb6a9a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h
@@ -95,29 +95,41 @@
 /* Following DMA defines are chanels oriented */
 #define DMA_CHAN_BASE_ADDR		0x00001100
 #define DMA_CHAN_BASE_OFFSET		0x80
-#define DMA_CHANX_BASE_ADDR(x)		(DMA_CHAN_BASE_ADDR + \
-					(x * DMA_CHAN_BASE_OFFSET))
+
+static inline u32 dma_chanx_base_addr(const struct dwmac4_addrs *addrs,
+				      const u32 x)
+{
+	u32 addr;
+
+	if (addrs)
+		addr = addrs->dma_chan + (x * addrs->dma_chan_offset);
+	else
+		addr = DMA_CHAN_BASE_ADDR + (x * DMA_CHAN_BASE_OFFSET);
+
+	return addr;
+}
+
 #define DMA_CHAN_REG_NUMBER		17
 
-#define DMA_CHAN_CONTROL(x)		DMA_CHANX_BASE_ADDR(x)
-#define DMA_CHAN_TX_CONTROL(x)		(DMA_CHANX_BASE_ADDR(x) + 0x4)
-#define DMA_CHAN_RX_CONTROL(x)		(DMA_CHANX_BASE_ADDR(x) + 0x8)
-#define DMA_CHAN_TX_BASE_ADDR_HI(x)	(DMA_CHANX_BASE_ADDR(x) + 0x10)
-#define DMA_CHAN_TX_BASE_ADDR(x)	(DMA_CHANX_BASE_ADDR(x) + 0x14)
-#define DMA_CHAN_RX_BASE_ADDR_HI(x)	(DMA_CHANX_BASE_ADDR(x) + 0x18)
-#define DMA_CHAN_RX_BASE_ADDR(x)	(DMA_CHANX_BASE_ADDR(x) + 0x1c)
-#define DMA_CHAN_TX_END_ADDR(x)		(DMA_CHANX_BASE_ADDR(x) + 0x20)
-#define DMA_CHAN_RX_END_ADDR(x)		(DMA_CHANX_BASE_ADDR(x) + 0x28)
-#define DMA_CHAN_TX_RING_LEN(x)		(DMA_CHANX_BASE_ADDR(x) + 0x2c)
-#define DMA_CHAN_RX_RING_LEN(x)		(DMA_CHANX_BASE_ADDR(x) + 0x30)
-#define DMA_CHAN_INTR_ENA(x)		(DMA_CHANX_BASE_ADDR(x) + 0x34)
-#define DMA_CHAN_RX_WATCHDOG(x)		(DMA_CHANX_BASE_ADDR(x) + 0x38)
-#define DMA_CHAN_SLOT_CTRL_STATUS(x)	(DMA_CHANX_BASE_ADDR(x) + 0x3c)
-#define DMA_CHAN_CUR_TX_DESC(x)		(DMA_CHANX_BASE_ADDR(x) + 0x44)
-#define DMA_CHAN_CUR_RX_DESC(x)		(DMA_CHANX_BASE_ADDR(x) + 0x4c)
-#define DMA_CHAN_CUR_TX_BUF_ADDR(x)	(DMA_CHANX_BASE_ADDR(x) + 0x54)
-#define DMA_CHAN_CUR_RX_BUF_ADDR(x)	(DMA_CHANX_BASE_ADDR(x) + 0x5c)
-#define DMA_CHAN_STATUS(x)		(DMA_CHANX_BASE_ADDR(x) + 0x60)
+#define DMA_CHAN_CONTROL(addrs, x)	dma_chanx_base_addr(addrs, x)
+#define DMA_CHAN_TX_CONTROL(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x4)
+#define DMA_CHAN_RX_CONTROL(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x8)
+#define DMA_CHAN_TX_BASE_ADDR_HI(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x10)
+#define DMA_CHAN_TX_BASE_ADDR(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x14)
+#define DMA_CHAN_RX_BASE_ADDR_HI(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x18)
+#define DMA_CHAN_RX_BASE_ADDR(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x1c)
+#define DMA_CHAN_TX_END_ADDR(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x20)
+#define DMA_CHAN_RX_END_ADDR(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x28)
+#define DMA_CHAN_TX_RING_LEN(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x2c)
+#define DMA_CHAN_RX_RING_LEN(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x30)
+#define DMA_CHAN_INTR_ENA(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x34)
+#define DMA_CHAN_RX_WATCHDOG(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x38)
+#define DMA_CHAN_SLOT_CTRL_STATUS(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x3c)
+#define DMA_CHAN_CUR_TX_DESC(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x44)
+#define DMA_CHAN_CUR_RX_DESC(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x4c)
+#define DMA_CHAN_CUR_TX_BUF_ADDR(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x54)
+#define DMA_CHAN_CUR_RX_BUF_ADDR(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x5c)
+#define DMA_CHAN_STATUS(addrs, x)	(dma_chanx_base_addr(addrs, x) + 0x60)
 
 /* DMA Control X */
 #define DMA_CONTROL_SPH			BIT(24)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
index 5e9c495aa03e..df41eac54058 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c
@@ -11,6 +11,7 @@
 #include "common.h"
 #include "dwmac4_dma.h"
 #include "dwmac4.h"
+#include "stmmac.h"
 
 int dwmac4_dma_reset(void __iomem *ioaddr)
 {
@@ -28,22 +29,27 @@ int dwmac4_dma_reset(void __iomem *ioaddr)
 void dwmac4_set_rx_tail_ptr(struct stmmac_priv *priv, void __iomem *ioaddr,
 			    u32 tail_ptr, u32 chan)
 {
-	writel(tail_ptr, ioaddr + DMA_CHAN_RX_END_ADDR(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	writel(tail_ptr, ioaddr + DMA_CHAN_RX_END_ADDR(dwmac4_addrs, chan));
 }
 
 void dwmac4_set_tx_tail_ptr(struct stmmac_priv *priv, void __iomem *ioaddr,
 			    u32 tail_ptr, u32 chan)
 {
-	writel(tail_ptr, ioaddr + DMA_CHAN_TX_END_ADDR(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	writel(tail_ptr, ioaddr + DMA_CHAN_TX_END_ADDR(dwmac4_addrs, chan));
 }
 
 void dwmac4_dma_start_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 u32 chan)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 
 	value |= DMA_CONTROL_ST;
-	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 
 	value = readl(ioaddr + GMAC_CONFIG);
 	value |= GMAC_CONFIG_TE;
@@ -53,20 +59,24 @@ void dwmac4_dma_start_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
 void dwmac4_dma_stop_tx(struct stmmac_priv *priv, void __iomem *ioaddr,
 			u32 chan)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	u32 value = readl(ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 
 	value &= ~DMA_CONTROL_ST;
-	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_TX_CONTROL(dwmac4_addrs, chan));
 }
 
 void dwmac4_dma_start_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 u32 chan)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 
 	value |= DMA_CONTROL_SR;
 
-	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 
 	value = readl(ioaddr + GMAC_CONFIG);
 	value |= GMAC_CONFIG_RE;
@@ -76,81 +86,91 @@ void dwmac4_dma_start_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
 void dwmac4_dma_stop_rx(struct stmmac_priv *priv, void __iomem *ioaddr,
 			u32 chan)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 
 	value &= ~DMA_CONTROL_SR;
-	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(chan));
+	writel(value, ioaddr + DMA_CHAN_RX_CONTROL(dwmac4_addrs, chan));
 }
 
 void dwmac4_set_tx_ring_len(struct stmmac_priv *priv, void __iomem *ioaddr,
 			    u32 len, u32 chan)
 {
-	writel(len, ioaddr + DMA_CHAN_TX_RING_LEN(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	writel(len, ioaddr + DMA_CHAN_TX_RING_LEN(dwmac4_addrs, chan));
 }
 
 void dwmac4_set_rx_ring_len(struct stmmac_priv *priv, void __iomem *ioaddr,
 			    u32 len, u32 chan)
 {
-	writel(len, ioaddr + DMA_CHAN_RX_RING_LEN(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+
+	writel(len, ioaddr + DMA_CHAN_RX_RING_LEN(dwmac4_addrs, chan));
 }
 
 void dwmac4_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
 			   u32 chan, bool rx, bool tx)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 
 	if (rx)
 		value |= DMA_CHAN_INTR_DEFAULT_RX;
 	if (tx)
 		value |= DMA_CHAN_INTR_DEFAULT_TX;
 
-	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 }
 
 void dwmac410_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
 			     u32 chan, bool rx, bool tx)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 
 	if (rx)
 		value |= DMA_CHAN_INTR_DEFAULT_RX_4_10;
 	if (tx)
 		value |= DMA_CHAN_INTR_DEFAULT_TX_4_10;
 
-	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 }
 
 void dwmac4_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
 			    u32 chan, bool rx, bool tx)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 
 	if (rx)
 		value &= ~DMA_CHAN_INTR_DEFAULT_RX;
 	if (tx)
 		value &= ~DMA_CHAN_INTR_DEFAULT_TX;
 
-	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 }
 
 void dwmac410_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr,
 			      u32 chan, bool rx, bool tx)
 {
-	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 value = readl(ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 
 	if (rx)
 		value &= ~DMA_CHAN_INTR_DEFAULT_RX_4_10;
 	if (tx)
 		value &= ~DMA_CHAN_INTR_DEFAULT_TX_4_10;
 
-	writel(value, ioaddr + DMA_CHAN_INTR_ENA(chan));
+	writel(value, ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 }
 
 int dwmac4_dma_interrupt(struct stmmac_priv *priv, void __iomem *ioaddr,
 			 struct stmmac_extra_stats *x, u32 chan, u32 dir)
 {
-	u32 intr_status = readl(ioaddr + DMA_CHAN_STATUS(chan));
-	u32 intr_en = readl(ioaddr + DMA_CHAN_INTR_ENA(chan));
+	const struct dwmac4_addrs *dwmac4_addrs = priv->plat->dwmac4_addrs;
+	u32 intr_status = readl(ioaddr + DMA_CHAN_STATUS(dwmac4_addrs, chan));
+	u32 intr_en = readl(ioaddr + DMA_CHAN_INTR_ENA(dwmac4_addrs, chan));
 	int ret = 0;
 
 	if (dir == DMA_DIR_RX)
@@ -195,7 +215,8 @@ int dwmac4_dma_interrupt(struct stmmac_priv *priv, void __iomem *ioaddr,
 	if (unlikely(intr_status & DMA_CHAN_STATUS_ERI))
 		x->rx_early_irq++;
 
-	writel(intr_status & intr_en, ioaddr + DMA_CHAN_STATUS(chan));
+	writel(intr_status & intr_en,
+	       ioaddr + DMA_CHAN_STATUS(dwmac4_addrs, chan));
 	return ret;
 }
 
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index dafa001e9e7a..225751a8fd8e 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -186,6 +186,24 @@ struct stmmac_safety_feature_cfg {
 	u32 tmouten;
 };
 
+/* Addresses that may be customized by a platform */
+struct dwmac4_addrs {
+	u32 dma_chan;
+	u32 dma_chan_offset;
+	u32 mtl_chan;
+	u32 mtl_chan_offset;
+	u32 mtl_ets_ctrl;
+	u32 mtl_ets_ctrl_offset;
+	u32 mtl_txq_weight;
+	u32 mtl_txq_weight_offset;
+	u32 mtl_send_slp_cred;
+	u32 mtl_send_slp_cred_offset;
+	u32 mtl_high_cred;
+	u32 mtl_high_cred_offset;
+	u32 mtl_low_cred;
+	u32 mtl_low_cred_offset;
+};
+
 struct plat_stmmacenet_data {
 	int bus_id;
 	int phy_addr;
@@ -274,5 +292,6 @@ struct plat_stmmacenet_data {
 	bool use_phy_wol;
 	bool sph_disable;
 	bool serdes_up_after_phy_linkup;
+	const struct dwmac4_addrs *dwmac4_addrs;
 };
 #endif
-- 
cgit v1.2.3-70-g09d2


From 164a9ebe97420cbe801140bb3de039695e09e205 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:07 -0500
Subject: net: stmmac: dwmac-qcom-ethqos: Respect phy-mode and TX delay

The driver currently sets a MAC TX delay of 2 ns no matter what the
phy-mode is. If the phy-mode indicates the phy is in charge of the
TX delay (rgmii-txid, rgmii-id), don't do it in the MAC.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c  | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index 32763566c214..abec6dd27992 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -279,6 +279,17 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos)
 
 static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 {
+	int phase_shift;
+	int phy_mode;
+
+	/* Determine if the PHY adds a 2 ns TX delay or the MAC handles it */
+	phy_mode = device_get_phy_mode(&ethqos->pdev->dev);
+	if (phy_mode == PHY_INTERFACE_MODE_RGMII_ID ||
+	    phy_mode == PHY_INTERFACE_MODE_RGMII_TXID)
+		phase_shift = 0;
+	else
+		phase_shift = RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN;
+
 	/* Disable loopback mode */
 	rgmii_updatel(ethqos, RGMII_CONFIG2_TX_TO_RX_LOOPBACK_EN,
 		      0, RGMII_IO_MACRO_CONFIG2);
@@ -300,9 +311,9 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 			      RGMII_CONFIG_PROG_SWAP, RGMII_IO_MACRO_CONFIG);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL,
 			      0, RGMII_IO_MACRO_CONFIG2);
+
 		rgmii_updatel(ethqos, RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN,
-			      RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN,
-			      RGMII_IO_MACRO_CONFIG2);
+			      phase_shift, RGMII_IO_MACRO_CONFIG2);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15,
 			      0, RGMII_IO_MACRO_CONFIG2);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
@@ -336,8 +347,7 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 		rgmii_updatel(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL,
 			      0, RGMII_IO_MACRO_CONFIG2);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN,
-			      RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN,
-			      RGMII_IO_MACRO_CONFIG2);
+			      phase_shift, RGMII_IO_MACRO_CONFIG2);
 		rgmii_updatel(ethqos, RGMII_CONFIG_MAX_SPD_PRG_2,
 			      BIT(6), RGMII_IO_MACRO_CONFIG);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15,
@@ -375,7 +385,7 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 		rgmii_updatel(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL,
 			      0, RGMII_IO_MACRO_CONFIG2);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN,
-			      0, RGMII_IO_MACRO_CONFIG2);
+			      phase_shift, RGMII_IO_MACRO_CONFIG2);
 		rgmii_updatel(ethqos, RGMII_CONFIG_MAX_SPD_PRG_9,
 			      BIT(12) | GENMASK(9, 8),
 			      RGMII_IO_MACRO_CONFIG);
-- 
cgit v1.2.3-70-g09d2


From 030f1d5972aaac8584a8ecad4d2739c6da4a20ea Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:08 -0500
Subject: net: stmmac: dwmac-qcom-ethqos: Use loopback_en for all speeds

It seems that this variable should be used for all speeds, not just
1000/100.

While at it refactor it slightly to be more readable, including fixing
the typo in the variable name.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c    | 36 ++++++++++------------
 1 file changed, 17 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index abec6dd27992..ec9e93147716 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -78,7 +78,7 @@ struct ethqos_emac_por {
 struct ethqos_emac_driver_data {
 	const struct ethqos_emac_por *por;
 	unsigned int num_por;
-	bool rgmii_config_looback_en;
+	bool rgmii_config_loopback_en;
 };
 
 struct qcom_ethqos {
@@ -91,7 +91,7 @@ struct qcom_ethqos {
 
 	const struct ethqos_emac_por *por;
 	unsigned int num_por;
-	bool rgmii_config_looback_en;
+	bool rgmii_config_loopback_en;
 };
 
 static int rgmii_readl(struct qcom_ethqos *ethqos, unsigned int offset)
@@ -183,7 +183,7 @@ static const struct ethqos_emac_por emac_v2_3_0_por[] = {
 static const struct ethqos_emac_driver_data emac_v2_3_0_data = {
 	.por = emac_v2_3_0_por,
 	.num_por = ARRAY_SIZE(emac_v2_3_0_por),
-	.rgmii_config_looback_en = true,
+	.rgmii_config_loopback_en = true,
 };
 
 static const struct ethqos_emac_por emac_v2_1_0_por[] = {
@@ -198,7 +198,7 @@ static const struct ethqos_emac_por emac_v2_1_0_por[] = {
 static const struct ethqos_emac_driver_data emac_v2_1_0_data = {
 	.por = emac_v2_1_0_por,
 	.num_por = ARRAY_SIZE(emac_v2_1_0_por),
-	.rgmii_config_looback_en = false,
+	.rgmii_config_loopback_en = false,
 };
 
 static int ethqos_dll_configure(struct qcom_ethqos *ethqos)
@@ -281,6 +281,7 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 {
 	int phase_shift;
 	int phy_mode;
+	int loopback;
 
 	/* Determine if the PHY adds a 2 ns TX delay or the MAC handles it */
 	phy_mode = device_get_phy_mode(&ethqos->pdev->dev);
@@ -294,6 +295,12 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 	rgmii_updatel(ethqos, RGMII_CONFIG2_TX_TO_RX_LOOPBACK_EN,
 		      0, RGMII_IO_MACRO_CONFIG2);
 
+	/* Determine if this platform wants loopback enabled after programming */
+	if (ethqos->rgmii_config_loopback_en)
+		loopback = RGMII_CONFIG_LOOPBACK_EN;
+	else
+		loopback = 0;
+
 	/* Select RGMII, write 0 to interface select */
 	rgmii_updatel(ethqos, RGMII_CONFIG_INTF_SEL,
 		      0, RGMII_IO_MACRO_CONFIG);
@@ -326,12 +333,8 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 		rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_DLY_EN,
 			      SDCC_DDR_CONFIG_PRG_DLY_EN,
 			      SDCC_HC_REG_DDR_CONFIG);
-		if (ethqos->rgmii_config_looback_en)
-			rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
-				      RGMII_CONFIG_LOOPBACK_EN, RGMII_IO_MACRO_CONFIG);
-		else
-			rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
-				      0, RGMII_IO_MACRO_CONFIG);
+		rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
+			      loopback, RGMII_IO_MACRO_CONFIG);
 		break;
 
 	case SPEED_100:
@@ -363,13 +366,8 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 		rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN,
 			      SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN,
 			      SDCC_HC_REG_DDR_CONFIG);
-		if (ethqos->rgmii_config_looback_en)
-			rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
-				      RGMII_CONFIG_LOOPBACK_EN, RGMII_IO_MACRO_CONFIG);
-		else
-			rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
-				      0, RGMII_IO_MACRO_CONFIG);
-
+		rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
+			      loopback, RGMII_IO_MACRO_CONFIG);
 		break;
 
 	case SPEED_10:
@@ -403,7 +401,7 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 			      SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN,
 			      SDCC_HC_REG_DDR_CONFIG);
 		rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN,
-			      RGMII_CONFIG_LOOPBACK_EN, RGMII_IO_MACRO_CONFIG);
+			      loopback, RGMII_IO_MACRO_CONFIG);
 		break;
 	default:
 		dev_err(&ethqos->pdev->dev,
@@ -548,7 +546,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev)
 	data = of_device_get_match_data(&pdev->dev);
 	ethqos->por = data->por;
 	ethqos->num_por = data->num_por;
-	ethqos->rgmii_config_looback_en = data->rgmii_config_looback_en;
+	ethqos->rgmii_config_loopback_en = data->rgmii_config_loopback_en;
 
 	ethqos->rgmii_clk = devm_clk_get(&pdev->dev, "rgmii");
 	if (IS_ERR(ethqos->rgmii_clk)) {
-- 
cgit v1.2.3-70-g09d2


From b68376191c69e621cba9cae4f9d5cead36cddb43 Mon Sep 17 00:00:00 2001
From: Andrew Halaney <ahalaney@redhat.com>
Date: Tue, 11 Apr 2023 15:04:09 -0500
Subject: net: stmmac: dwmac-qcom-ethqos: Add EMAC3 support

Add the new programming sequence needed for EMAC3 based platforms such
as the sc8280xp family.

Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c    | 122 +++++++++++++++++----
 1 file changed, 101 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index ec9e93147716..16a8c361283b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -11,6 +11,7 @@
 
 #define RGMII_IO_MACRO_CONFIG		0x0
 #define SDCC_HC_REG_DLL_CONFIG		0x4
+#define SDCC_TEST_CTL			0x8
 #define SDCC_HC_REG_DDR_CONFIG		0xC
 #define SDCC_HC_REG_DLL_CONFIG2		0x10
 #define SDC4_STATUS			0x14
@@ -49,6 +50,7 @@
 #define SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY	GENMASK(26, 21)
 #define SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_CODE	GENMASK(29, 27)
 #define SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN	BIT(30)
+#define SDCC_DDR_CONFIG_TCXO_CYCLES_CNT		GENMASK(11, 9)
 #define SDCC_DDR_CONFIG_PRG_RCLK_DLY		GENMASK(8, 0)
 
 /* SDCC_HC_REG_DLL_CONFIG2 fields */
@@ -79,6 +81,8 @@ struct ethqos_emac_driver_data {
 	const struct ethqos_emac_por *por;
 	unsigned int num_por;
 	bool rgmii_config_loopback_en;
+	bool has_emac3;
+	struct dwmac4_addrs dwmac4_addrs;
 };
 
 struct qcom_ethqos {
@@ -92,6 +96,7 @@ struct qcom_ethqos {
 	const struct ethqos_emac_por *por;
 	unsigned int num_por;
 	bool rgmii_config_loopback_en;
+	bool has_emac3;
 };
 
 static int rgmii_readl(struct qcom_ethqos *ethqos, unsigned int offset)
@@ -184,6 +189,7 @@ static const struct ethqos_emac_driver_data emac_v2_3_0_data = {
 	.por = emac_v2_3_0_por,
 	.num_por = ARRAY_SIZE(emac_v2_3_0_por),
 	.rgmii_config_loopback_en = true,
+	.has_emac3 = false,
 };
 
 static const struct ethqos_emac_por emac_v2_1_0_por[] = {
@@ -199,6 +205,39 @@ static const struct ethqos_emac_driver_data emac_v2_1_0_data = {
 	.por = emac_v2_1_0_por,
 	.num_por = ARRAY_SIZE(emac_v2_1_0_por),
 	.rgmii_config_loopback_en = false,
+	.has_emac3 = false,
+};
+
+static const struct ethqos_emac_por emac_v3_0_0_por[] = {
+	{ .offset = RGMII_IO_MACRO_CONFIG,	.value = 0x40c01343 },
+	{ .offset = SDCC_HC_REG_DLL_CONFIG,	.value = 0x2004642c },
+	{ .offset = SDCC_HC_REG_DDR_CONFIG,	.value = 0x80040800 },
+	{ .offset = SDCC_HC_REG_DLL_CONFIG2,	.value = 0x00200000 },
+	{ .offset = SDCC_USR_CTL,		.value = 0x00010800 },
+	{ .offset = RGMII_IO_MACRO_CONFIG2,	.value = 0x00002060 },
+};
+
+static const struct ethqos_emac_driver_data emac_v3_0_0_data = {
+	.por = emac_v3_0_0_por,
+	.num_por = ARRAY_SIZE(emac_v3_0_0_por),
+	.rgmii_config_loopback_en = false,
+	.has_emac3 = true,
+	.dwmac4_addrs = {
+		.dma_chan = 0x00008100,
+		.dma_chan_offset = 0x1000,
+		.mtl_chan = 0x00008000,
+		.mtl_chan_offset = 0x1000,
+		.mtl_ets_ctrl = 0x00008010,
+		.mtl_ets_ctrl_offset = 0x1000,
+		.mtl_txq_weight = 0x00008018,
+		.mtl_txq_weight_offset = 0x1000,
+		.mtl_send_slp_cred = 0x0000801c,
+		.mtl_send_slp_cred_offset = 0x1000,
+		.mtl_high_cred = 0x00008020,
+		.mtl_high_cred_offset = 0x1000,
+		.mtl_low_cred = 0x00008024,
+		.mtl_low_cred_offset = 0x1000,
+	},
 };
 
 static int ethqos_dll_configure(struct qcom_ethqos *ethqos)
@@ -222,11 +261,13 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos)
 	rgmii_updatel(ethqos, SDCC_DLL_CONFIG_DLL_EN,
 		      SDCC_DLL_CONFIG_DLL_EN, SDCC_HC_REG_DLL_CONFIG);
 
-	rgmii_updatel(ethqos, SDCC_DLL_MCLK_GATING_EN,
-		      0, SDCC_HC_REG_DLL_CONFIG);
+	if (!ethqos->has_emac3) {
+		rgmii_updatel(ethqos, SDCC_DLL_MCLK_GATING_EN,
+			      0, SDCC_HC_REG_DLL_CONFIG);
 
-	rgmii_updatel(ethqos, SDCC_DLL_CDR_FINE_PHASE,
-		      0, SDCC_HC_REG_DLL_CONFIG);
+		rgmii_updatel(ethqos, SDCC_DLL_CDR_FINE_PHASE,
+			      0, SDCC_HC_REG_DLL_CONFIG);
+	}
 
 	/* Wait for CK_OUT_EN clear */
 	do {
@@ -261,18 +302,20 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos)
 	rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_CAL_EN,
 		      SDCC_DLL_CONFIG2_DDR_CAL_EN, SDCC_HC_REG_DLL_CONFIG2);
 
-	rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DLL_CLOCK_DIS,
-		      0, SDCC_HC_REG_DLL_CONFIG2);
+	if (!ethqos->has_emac3) {
+		rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DLL_CLOCK_DIS,
+			      0, SDCC_HC_REG_DLL_CONFIG2);
 
-	rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_MCLK_FREQ_CALC,
-		      0x1A << 10, SDCC_HC_REG_DLL_CONFIG2);
+		rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_MCLK_FREQ_CALC,
+			      0x1A << 10, SDCC_HC_REG_DLL_CONFIG2);
 
-	rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SEL,
-		      BIT(2), SDCC_HC_REG_DLL_CONFIG2);
+		rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SEL,
+			      BIT(2), SDCC_HC_REG_DLL_CONFIG2);
 
-	rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW,
-		      SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW,
-		      SDCC_HC_REG_DLL_CONFIG2);
+		rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW,
+			      SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW,
+			      SDCC_HC_REG_DLL_CONFIG2);
+	}
 
 	return 0;
 }
@@ -327,9 +370,18 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 			      RGMII_CONFIG2_RX_PROG_SWAP,
 			      RGMII_IO_MACRO_CONFIG2);
 
-		/* Set PRG_RCLK_DLY to 57 for 1.8 ns delay */
-		rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_RCLK_DLY,
-			      57, SDCC_HC_REG_DDR_CONFIG);
+		/* PRG_RCLK_DLY = TCXO period * TCXO_CYCLES_CNT / 2 * RX delay ns,
+		 * in practice this becomes PRG_RCLK_DLY = 52 * 4 / 2 * RX delay ns
+		 */
+		if (ethqos->has_emac3) {
+			/* 0.9 ns */
+			rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_RCLK_DLY,
+				      115, SDCC_HC_REG_DDR_CONFIG);
+		} else {
+			/* 1.8 ns */
+			rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_RCLK_DLY,
+				      57, SDCC_HC_REG_DDR_CONFIG);
+		}
 		rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_DLY_EN,
 			      SDCC_DDR_CONFIG_PRG_DLY_EN,
 			      SDCC_HC_REG_DDR_CONFIG);
@@ -355,8 +407,15 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 			      BIT(6), RGMII_IO_MACRO_CONFIG);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15,
 			      0, RGMII_IO_MACRO_CONFIG2);
-		rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
-			      0, RGMII_IO_MACRO_CONFIG2);
+
+		if (ethqos->has_emac3)
+			rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
+				      RGMII_CONFIG2_RX_PROG_SWAP,
+				      RGMII_IO_MACRO_CONFIG2);
+		else
+			rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
+				      0, RGMII_IO_MACRO_CONFIG2);
+
 		/* Write 0x5 to PRG_RCLK_DLY_CODE */
 		rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_CODE,
 			      (BIT(29) | BIT(27)), SDCC_HC_REG_DDR_CONFIG);
@@ -389,8 +448,13 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos)
 			      RGMII_IO_MACRO_CONFIG);
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15,
 			      0, RGMII_IO_MACRO_CONFIG2);
-		rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
-			      0, RGMII_IO_MACRO_CONFIG2);
+		if (ethqos->has_emac3)
+			rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
+				      RGMII_CONFIG2_RX_PROG_SWAP,
+				      RGMII_IO_MACRO_CONFIG2);
+		else
+			rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP,
+				      0, RGMII_IO_MACRO_CONFIG2);
 		/* Write 0x5 to PRG_RCLK_DLY_CODE */
 		rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_CODE,
 			      (BIT(29) | BIT(27)), SDCC_HC_REG_DDR_CONFIG);
@@ -433,6 +497,17 @@ static int ethqos_configure(struct qcom_ethqos *ethqos)
 	rgmii_updatel(ethqos, SDCC_DLL_CONFIG_PDN,
 		      SDCC_DLL_CONFIG_PDN, SDCC_HC_REG_DLL_CONFIG);
 
+	if (ethqos->has_emac3) {
+		if (ethqos->speed == SPEED_1000) {
+			rgmii_writel(ethqos, 0x1800000, SDCC_TEST_CTL);
+			rgmii_writel(ethqos, 0x2C010800, SDCC_USR_CTL);
+			rgmii_writel(ethqos, 0xA001, SDCC_HC_REG_DLL_CONFIG2);
+		} else {
+			rgmii_writel(ethqos, 0x40010800, SDCC_USR_CTL);
+			rgmii_writel(ethqos, 0xA001, SDCC_HC_REG_DLL_CONFIG2);
+		}
+	}
+
 	/* Clear DLL_RST */
 	rgmii_updatel(ethqos, SDCC_DLL_CONFIG_DLL_RST, 0,
 		      SDCC_HC_REG_DLL_CONFIG);
@@ -452,7 +527,9 @@ static int ethqos_configure(struct qcom_ethqos *ethqos)
 			      SDCC_HC_REG_DLL_CONFIG);
 
 		/* Set USR_CTL bit 26 with mask of 3 bits */
-		rgmii_updatel(ethqos, GENMASK(26, 24), BIT(26), SDCC_USR_CTL);
+		if (!ethqos->has_emac3)
+			rgmii_updatel(ethqos, GENMASK(26, 24), BIT(26),
+				      SDCC_USR_CTL);
 
 		/* wait for DLL LOCK */
 		do {
@@ -547,6 +624,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev)
 	ethqos->por = data->por;
 	ethqos->num_por = data->num_por;
 	ethqos->rgmii_config_loopback_en = data->rgmii_config_loopback_en;
+	ethqos->has_emac3 = data->has_emac3;
 
 	ethqos->rgmii_clk = devm_clk_get(&pdev->dev, "rgmii");
 	if (IS_ERR(ethqos->rgmii_clk)) {
@@ -566,6 +644,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev)
 	plat_dat->fix_mac_speed = ethqos_fix_mac_speed;
 	plat_dat->dump_debug_regs = rgmii_dump;
 	plat_dat->has_gmac4 = 1;
+	plat_dat->dwmac4_addrs = &data->dwmac4_addrs;
 	plat_dat->pmt = 1;
 	plat_dat->tso_en = of_property_read_bool(np, "snps,tso");
 	if (of_device_is_compatible(np, "qcom,qcs404-ethqos"))
@@ -603,6 +682,7 @@ static int qcom_ethqos_remove(struct platform_device *pdev)
 
 static const struct of_device_id qcom_ethqos_match[] = {
 	{ .compatible = "qcom,qcs404-ethqos", .data = &emac_v2_3_0_data},
+	{ .compatible = "qcom,sc8280xp-ethqos", .data = &emac_v3_0_0_data},
 	{ .compatible = "qcom,sm8150-ethqos", .data = &emac_v2_1_0_data},
 	{ }
 };
-- 
cgit v1.2.3-70-g09d2


From 50762d9af307b1c466fe0e1441c7923975927d98 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 11 Apr 2023 18:50:36 -0700
Subject: net: docs: update the sample code in driver.rst

The sample code talks about single-queue devices and uses locks.
Update it to something resembling more modern code.
Make sure we mention use of READ_ONCE() / WRITE_ONCE().

Change the comment which talked about consumer on the xmit side.
AFAIU xmit is the producer and completions are a consumer.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/networking/driver.rst | 61 ++++++++++++++++---------------------
 1 file changed, 27 insertions(+), 34 deletions(-)

diff --git a/Documentation/networking/driver.rst b/Documentation/networking/driver.rst
index 4071f2c00f8b..4f5dfa9c022e 100644
--- a/Documentation/networking/driver.rst
+++ b/Documentation/networking/driver.rst
@@ -47,30 +47,43 @@ for a driver implementing scatter-gather this means:
 
 .. code-block:: c
 
+	static u32 drv_tx_avail(struct drv_ring *dr)
+	{
+		u32 used = READ_ONCE(dr->prod) - READ_ONCE(dr->cons);
+
+		return dr->tx_ring_size - (used & bp->tx_ring_mask);
+	}
+
 	static netdev_tx_t drv_hard_start_xmit(struct sk_buff *skb,
 					       struct net_device *dev)
 	{
 		struct drv *dp = netdev_priv(dev);
+		struct netdev_queue *txq;
+		struct drv_ring *dr;
+		int idx;
+
+		idx = skb_get_queue_mapping(skb);
+		dr = dp->tx_rings[idx];
+		txq = netdev_get_tx_queue(dev, idx);
 
-		lock_tx(dp);
 		//...
-		/* This is a hard error log it. */
-		if (TX_BUFFS_AVAIL(dp) <= (skb_shinfo(skb)->nr_frags + 1)) {
+		/* This should be a very rare race - log it. */
+		if (drv_tx_avail(dr) <= skb_shinfo(skb)->nr_frags + 1) {
 			netif_stop_queue(dev);
-			unlock_tx(dp);
-			printk(KERN_ERR PFX "%s: BUG! Tx Ring full when queue awake!\n",
-			       dev->name);
+			netdev_warn(dev, "Tx Ring full when queue awake!\n");
 			return NETDEV_TX_BUSY;
 		}
 
 		//... queue packet to card ...
-		//... update tx consumer index ...
 
-		if (TX_BUFFS_AVAIL(dp) <= (MAX_SKB_FRAGS + 1))
-			netif_stop_queue(dev);
+		netdev_tx_sent_queue(txq, skb->len);
+
+		//... update tx producer index using WRITE_ONCE() ...
+
+		if (!netif_txq_maybe_stop(txq, drv_tx_avail(dr),
+					  MAX_SKB_FRAGS + 1, 2 * MAX_SKB_FRAGS))
+			dr->stats.stopped++;
 
-		//...
-		unlock_tx(dp);
 		//...
 		return NETDEV_TX_OK;
 	}
@@ -79,30 +92,10 @@ And then at the end of your TX reclamation event handling:
 
 .. code-block:: c
 
-	if (netif_queue_stopped(dp->dev) &&
-	    TX_BUFFS_AVAIL(dp) > (MAX_SKB_FRAGS + 1))
-		netif_wake_queue(dp->dev);
-
-For a non-scatter-gather supporting card, the three tests simply become:
-
-.. code-block:: c
-
-		/* This is a hard error log it. */
-		if (TX_BUFFS_AVAIL(dp) <= 0)
-
-and:
-
-.. code-block:: c
-
-		if (TX_BUFFS_AVAIL(dp) == 0)
-
-and:
-
-.. code-block:: c
+	//... update tx consumer index using WRITE_ONCE() ...
 
-	if (netif_queue_stopped(dp->dev) &&
-	    TX_BUFFS_AVAIL(dp) > 0)
-		netif_wake_queue(dp->dev);
+	netif_txq_completed_wake(txq, cmpl_pkts, cmpl_bytes,
+				 drv_tx_avail(dr), 2 * MAX_SKB_FRAGS);
 
 Lockless queue stop / wake helper macros
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-- 
cgit v1.2.3-70-g09d2


From 36647b206c014a0bf3ab17bc88f2c840eefd796c Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 11 Apr 2023 18:50:37 -0700
Subject: bnxt: use READ_ONCE/WRITE_ONCE for ring indexes

Eric points out that we should make sure that ring index updates
are wrapped in the appropriate READ_ONCE/WRITE_ONCE macros.

Suggested-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 6 +++---
 drivers/net/ethernet/broadcom/bnxt/bnxt.h     | 9 ++++-----
 drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c | 6 +++---
 3 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index f7602d8d79e3..92289ab2f34a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -472,7 +472,7 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		prod = NEXT_TX(prod);
 		tx_push->doorbell =
 			cpu_to_le32(DB_KEY_TX_PUSH | DB_LONG_TX_PUSH | prod);
-		txr->tx_prod = prod;
+		WRITE_ONCE(txr->tx_prod, prod);
 
 		tx_buf->is_push = 1;
 		netdev_tx_sent_queue(txq, skb->len);
@@ -583,7 +583,7 @@ normal_tx:
 	wmb();
 
 	prod = NEXT_TX(prod);
-	txr->tx_prod = prod;
+	WRITE_ONCE(txr->tx_prod, prod);
 
 	if (!netdev_xmit_more() || netif_xmit_stopped(txq))
 		bnxt_txr_db_kick(bp, txr, prod);
@@ -688,7 +688,7 @@ next_tx_int:
 		dev_kfree_skb_any(skb);
 	}
 
-	txr->tx_cons = cons;
+	WRITE_ONCE(txr->tx_cons, cons);
 
 	__netif_txq_completed_wake(txq, nr_pkts, tx_bytes,
 				   bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 18cac98ba58e..080e73496066 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2231,13 +2231,12 @@ struct bnxt {
 #define SFF_MODULE_ID_QSFP28			0x11
 #define BNXT_MAX_PHY_I2C_RESP_SIZE		64
 
-static inline u32 bnxt_tx_avail(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
+static inline u32 bnxt_tx_avail(struct bnxt *bp,
+				const struct bnxt_tx_ring_info *txr)
 {
-	/* Tell compiler to fetch tx indices from memory. */
-	barrier();
+	u32 used = READ_ONCE(txr->tx_prod) - READ_ONCE(txr->tx_cons);
 
-	return bp->tx_ring_size -
-		((txr->tx_prod - txr->tx_cons) & bp->tx_ring_mask);
+	return bp->tx_ring_size - (used & bp->tx_ring_mask);
 }
 
 static inline void bnxt_writeq(struct bnxt *bp, u64 val,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
index 5843c93b1711..4efa5fe6972b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
@@ -64,7 +64,7 @@ struct bnxt_sw_tx_bd *bnxt_xmit_bd(struct bnxt *bp,
 		int frag_len;
 
 		prod = NEXT_TX(prod);
-		txr->tx_prod = prod;
+		WRITE_ONCE(txr->tx_prod, prod);
 
 		/* first fill up the first buffer */
 		frag_tx_buf = &txr->tx_buf_ring[prod];
@@ -94,7 +94,7 @@ struct bnxt_sw_tx_bd *bnxt_xmit_bd(struct bnxt *bp,
 	/* Sync TX BD */
 	wmb();
 	prod = NEXT_TX(prod);
-	txr->tx_prod = prod;
+	WRITE_ONCE(txr->tx_prod, prod);
 
 	return tx_buf;
 }
@@ -161,7 +161,7 @@ void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
 		}
 		tx_cons = NEXT_TX(tx_cons);
 	}
-	txr->tx_cons = tx_cons;
+	WRITE_ONCE(txr->tx_cons, tx_cons);
 	if (rx_doorbell_needed) {
 		tx_buf = &txr->tx_buf_ring[last_tx_cons];
 		bnxt_db_write(bp, &rxr->rx_db, tx_buf->rx_prod);
-- 
cgit v1.2.3-70-g09d2


From 9a714997386ba110154dc444cdd0279f966151b2 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 11 Apr 2023 18:50:38 -0700
Subject: mlx4: use READ_ONCE/WRITE_ONCE for ring indexes

Eric points out that we should make sure that ring index updates
are wrapped in the appropriate READ_ONCE/WRITE_ONCE macros.

Suggested-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_tx.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 2f79378fbf6e..65cb63f6c465 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -228,7 +228,9 @@ void mlx4_en_deactivate_tx_ring(struct mlx4_en_priv *priv,
 
 static inline bool mlx4_en_is_tx_ring_full(struct mlx4_en_tx_ring *ring)
 {
-	return ring->prod - ring->cons > ring->full_size;
+	u32 used = READ_ONCE(ring->prod) - READ_ONCE(ring->cons);
+
+	return used > ring->full_size;
 }
 
 static void mlx4_en_stamp_wqe(struct mlx4_en_priv *priv,
@@ -1083,7 +1085,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 			op_own |= cpu_to_be32(MLX4_WQE_CTRL_IIP);
 	}
 
-	ring->prod += nr_txbb;
+	WRITE_ONCE(ring->prod, ring->prod + nr_txbb);
 
 	/* If we used a bounce buffer then copy descriptor back into place */
 	if (unlikely(bounce))
@@ -1214,7 +1216,7 @@ netdev_tx_t mlx4_en_xmit_frame(struct mlx4_en_rx_ring *rx_ring,
 
 	rx_ring->xdp_tx++;
 
-	ring->prod += MLX4_EN_XDP_TX_NRTXBB;
+	WRITE_ONCE(ring->prod, ring->prod + MLX4_EN_XDP_TX_NRTXBB);
 
 	/* Ensure new descriptor hits memory
 	 * before setting ownership of this descriptor to HW
-- 
cgit v1.2.3-70-g09d2


From 5a674611d116a5fc28c5429beea2b78c6e2933ef Mon Sep 17 00:00:00 2001
From: Lorenz Bauer <lmb@isovalent.com>
Date: Thu, 13 Apr 2023 10:47:40 +0100
Subject: selftests/bpf: Fix use of uninitialized op_name in log tests

One of the test assertions uses an uninitialized op_name, which leads
to some headscratching if it fails. Use a string constant instead.

Fixes: b1a7a480a112 ("selftests/bpf: Add fixed vs rotating verifier log tests")
Signed-off-by: Lorenz Bauer <lmb@isovalent.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230413094740.18041-1-lmb@isovalent.com
---
 tools/testing/selftests/bpf/prog_tests/verifier_log.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier_log.c b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
index 475092a78deb..8337c6bc5b95 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier_log.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier_log.c
@@ -110,7 +110,7 @@ static void verif_log_subtest(const char *name, bool expect_load_error, int log_
 		}
 		if (!ASSERT_EQ(strlen(logs.buf), 24, "log_fixed_25"))
 			goto cleanup;
-		if (!ASSERT_STRNEQ(logs.buf, logs.reference, 24, op_name))
+		if (!ASSERT_STRNEQ(logs.buf, logs.reference, 24, "log_fixed_contents_25"))
 			goto cleanup;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From de6d014a09bf12a9a8959d60c0a1d4a41d394a89 Mon Sep 17 00:00:00 2001
From: Song Liu <song@kernel.org>
Date: Wed, 12 Apr 2023 14:04:21 -0700
Subject: selftests/bpf: Use read_perf_max_sample_freq() in perf_event_stackmap

Currently, perf_event sample period in perf_event_stackmap is set too low
that the test fails randomly. Fix this by using the max sample frequency,
from read_perf_max_sample_freq().

Move read_perf_max_sample_freq() to testing_helpers.c. Replace the CHECK()
with if-printf, as CHECK is not available in testing_helpers.c.

Fixes: 1da4864c2b20 ("selftests/bpf: Add callchain_stackid")
Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230412210423.900851-2-song@kernel.org
---
 .../selftests/bpf/prog_tests/perf_event_stackmap.c   |  3 ++-
 .../bpf/prog_tests/stacktrace_build_id_nmi.c         | 15 ---------------
 tools/testing/selftests/bpf/testing_helpers.c        | 20 ++++++++++++++++++++
 tools/testing/selftests/bpf/testing_helpers.h        |  2 ++
 4 files changed, 24 insertions(+), 16 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c b/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c
index 33144c9432ae..f4aad35afae1 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c
@@ -63,7 +63,8 @@ void test_perf_event_stackmap(void)
 			PERF_SAMPLE_BRANCH_NO_FLAGS |
 			PERF_SAMPLE_BRANCH_NO_CYCLES |
 			PERF_SAMPLE_BRANCH_CALL_STACK,
-		.sample_period = 5000,
+		.freq = 1,
+		.sample_freq = read_perf_max_sample_freq(),
 		.size = sizeof(struct perf_event_attr),
 	};
 	struct perf_event_stackmap *skel;
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
index 47558b0d7f66..5db9eec24b5b 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
@@ -2,21 +2,6 @@
 #include <test_progs.h>
 #include "test_stacktrace_build_id.skel.h"
 
-static __u64 read_perf_max_sample_freq(void)
-{
-	__u64 sample_freq = 5000; /* fallback to 5000 on error */
-	FILE *f;
-	__u32 duration = 0;
-
-	f = fopen("/proc/sys/kernel/perf_event_max_sample_rate", "r");
-	if (f == NULL)
-		return sample_freq;
-	CHECK(fscanf(f, "%llu", &sample_freq) != 1, "Get max sample rate",
-		  "return default value: 5000,err %d\n", -errno);
-	fclose(f);
-	return sample_freq;
-}
-
 void test_stacktrace_build_id_nmi(void)
 {
 	int control_map_fd, stackid_hmap_fd, stackmap_fd;
diff --git a/tools/testing/selftests/bpf/testing_helpers.c b/tools/testing/selftests/bpf/testing_helpers.c
index ecfea13f938b..0b5e0829e5be 100644
--- a/tools/testing/selftests/bpf/testing_helpers.c
+++ b/tools/testing/selftests/bpf/testing_helpers.c
@@ -229,3 +229,23 @@ int bpf_test_load_program(enum bpf_prog_type type, const struct bpf_insn *insns,
 
 	return bpf_prog_load(type, NULL, license, insns, insns_cnt, &opts);
 }
+
+__u64 read_perf_max_sample_freq(void)
+{
+	__u64 sample_freq = 5000; /* fallback to 5000 on error */
+	FILE *f;
+
+	f = fopen("/proc/sys/kernel/perf_event_max_sample_rate", "r");
+	if (f == NULL) {
+		printf("Failed to open /proc/sys/kernel/perf_event_max_sample_rate: err %d\n"
+		       "return default value: 5000\n", -errno);
+		return sample_freq;
+	}
+	if (fscanf(f, "%llu", &sample_freq) != 1) {
+		printf("Failed to parse /proc/sys/kernel/perf_event_max_sample_rate: err %d\n"
+		       "return default value: 5000\n", -errno);
+	}
+
+	fclose(f);
+	return sample_freq;
+}
diff --git a/tools/testing/selftests/bpf/testing_helpers.h b/tools/testing/selftests/bpf/testing_helpers.h
index 6ec00bf79cb5..eb8790f928e4 100644
--- a/tools/testing/selftests/bpf/testing_helpers.h
+++ b/tools/testing/selftests/bpf/testing_helpers.h
@@ -20,3 +20,5 @@ struct test_filter_set;
 int parse_test_list(const char *s,
 		    struct test_filter_set *test_set,
 		    bool is_glob_pattern);
+
+__u64 read_perf_max_sample_freq(void);
-- 
cgit v1.2.3-70-g09d2


From c1e07a80cf23d3a6e96172bc9a73bfa912a9fcbc Mon Sep 17 00:00:00 2001
From: Song Liu <song@kernel.org>
Date: Wed, 12 Apr 2023 14:04:22 -0700
Subject: selftests/bpf: Fix leaked bpf_link in get_stackid_cannot_attach

skel->links.oncpu is leaked in one case. This causes test perf_branches
fails when it runs after get_stackid_cannot_attach:

./test_progs -t get_stackid_cannot_attach,perf_branches
84      get_stackid_cannot_attach:OK
test_perf_branches_common:PASS:test_perf_branches_load 0 nsec
test_perf_branches_common:PASS:attach_perf_event 0 nsec
test_perf_branches_common:PASS:set_affinity 0 nsec
check_good_sample:FAIL:output not valid no valid sample from prog
146/1   perf_branches/perf_branches_hw:FAIL
146/2   perf_branches/perf_branches_no_hw:OK
146     perf_branches:FAIL

All error logs:
test_perf_branches_common:PASS:test_perf_branches_load 0 nsec
test_perf_branches_common:PASS:attach_perf_event 0 nsec
test_perf_branches_common:PASS:set_affinity 0 nsec
check_good_sample:FAIL:output not valid no valid sample from prog
146/1   perf_branches/perf_branches_hw:FAIL
146     perf_branches:FAIL
Summary: 1/1 PASSED, 0 SKIPPED, 1 FAILED

Fix this by adding the missing bpf_link__destroy().

Fixes: 346938e9380c ("selftests/bpf: Add get_stackid_cannot_attach")
Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230412210423.900851-3-song@kernel.org
---
 tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c b/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c
index 5308de1ed478..2715c68301f5 100644
--- a/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c
@@ -65,6 +65,7 @@ void test_get_stackid_cannot_attach(void)
 	skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu,
 							   pmu_fd);
 	ASSERT_OK_PTR(skel->links.oncpu, "attach_perf_event_callchain");
+	bpf_link__destroy(skel->links.oncpu);
 	close(pmu_fd);
 
 	/* add exclude_callchain_kernel, attach should fail */
-- 
cgit v1.2.3-70-g09d2


From 2995f9a8d427b9ff6f3cf4e85c0f9d4456ef324d Mon Sep 17 00:00:00 2001
From: Song Liu <song@kernel.org>
Date: Wed, 12 Apr 2023 14:04:23 -0700
Subject: selftests/bpf: Keep the loop in bpf_testmod_loop_test

Some compilers (for example clang-15) optimize bpf_testmod_loop_test and
remove the loop:

gcc version
(gdb) disassemble bpf_testmod_loop_test
Dump of assembler code for function bpf_testmod_loop_test:
   0x0000000000000570 <+0>:     callq  0x575 <bpf_testmod_loop_test+5>
   0x0000000000000575 <+5>:     xor    %eax,%eax
   0x0000000000000577 <+7>:     test   %edi,%edi
   0x0000000000000579 <+9>:     jle    0x587 <bpf_testmod_loop_test+23>
   0x000000000000057b <+11>:    xor    %edx,%edx
   0x000000000000057d <+13>:    add    %edx,%eax
   0x000000000000057f <+15>:    add    $0x1,%edx
   0x0000000000000582 <+18>:    cmp    %edx,%edi
   0x0000000000000584 <+20>:    jne    0x57d <bpf_testmod_loop_test+13>
   0x0000000000000586 <+22>:    retq
   0x0000000000000587 <+23>:    retq

clang-15 version
(gdb) disassemble bpf_testmod_loop_test
Dump of assembler code for function bpf_testmod_loop_test:
   0x0000000000000450 <+0>:     nopl   0x0(%rax,%rax,1)
   0x0000000000000455 <+5>:     test   %edi,%edi
   0x0000000000000457 <+7>:     jle    0x46b <bpf_testmod_loop_test+27>
   0x0000000000000459 <+9>:     lea    -0x1(%rdi),%eax
   0x000000000000045c <+12>:    lea    -0x2(%rdi),%ecx
   0x000000000000045f <+15>:    imul   %rax,%rcx
   0x0000000000000463 <+19>:    shr    %rcx
   0x0000000000000466 <+22>:    lea    -0x1(%rdi,%rcx,1),%eax
   0x000000000000046a <+26>:    retq
   0x000000000000046b <+27>:    xor    %eax,%eax
   0x000000000000046d <+29>:    retq

Note: The jne instruction is removed in clang-15 version.

Force the compile to keep the loop by making sum volatile.

Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230412210423.900851-4-song@kernel.org
---
 tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 7999476b9446..c5ad39bbe9af 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -130,7 +130,11 @@ bpf_testmod_test_btf_type_tag_percpu_2(struct bpf_testmod_btf_type_tag_3 *arg) {
 
 noinline int bpf_testmod_loop_test(int n)
 {
-	int i, sum = 0;
+	/* Make sum volatile, so smart compilers, such as clang, will not
+	 * optimize the code by removing the loop.
+	 */
+	volatile int sum = 0;
+	int i;
 
 	/* the primary goal of this test is to test LBR. Create a lot of
 	 * branches in the function, so we can catch it easily.
-- 
cgit v1.2.3-70-g09d2


From ee5059a64dbad4806a3c11babd0dbed5a5d04ead Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 12 Apr 2023 10:06:55 -0700
Subject: selftests/bpf: Remove stand-along test_verifier_log test binary

test_prog's prog_tests/verifier_log.c is superseding test_verifier_log
stand-alone test. It cover same checks and adds more, and is also
integrated into test_progs test runner.

Just remove test_verifier_log.c.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230412170655.1866831-1-andrii@kernel.org
---
 tools/testing/selftests/bpf/Makefile            |   2 +-
 tools/testing/selftests/bpf/test_verifier_log.c | 175 ------------------------
 2 files changed, 1 insertion(+), 176 deletions(-)
 delete mode 100644 tools/testing/selftests/bpf/test_verifier_log.c

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index b5ffdd89b86f..c49e5403ad0e 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -36,7 +36,7 @@ endif
 
 # Order correspond to 'make run_tests' order
 TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
-	test_verifier_log test_dev_cgroup \
+	test_dev_cgroup \
 	test_sock test_sockmap get_cgroup_id_user \
 	test_cgroup_storage \
 	test_tcpnotify_user test_sysctl \
diff --git a/tools/testing/selftests/bpf/test_verifier_log.c b/tools/testing/selftests/bpf/test_verifier_log.c
deleted file mode 100644
index 70feda97cee5..000000000000
--- a/tools/testing/selftests/bpf/test_verifier_log.c
+++ /dev/null
@@ -1,175 +0,0 @@
-#include <errno.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <unistd.h>
-#include <sys/time.h>
-
-#include <linux/bpf.h>
-#include <linux/filter.h>
-#include <linux/unistd.h>
-
-#include <bpf/bpf.h>
-
-#define LOG_SIZE (1 << 20)
-
-#define err(str...)	printf("ERROR: " str)
-
-static const struct bpf_insn code_sample[] = {
-	/* We need a few instructions to pass the min log length */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-		     BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-};
-
-static inline __u64 ptr_to_u64(const void *ptr)
-{
-	return (__u64) (unsigned long) ptr;
-}
-
-static int load(char *log, size_t log_len, int log_level)
-{
-	union bpf_attr attr;
-
-	bzero(&attr, sizeof(attr));
-	attr.prog_type = BPF_PROG_TYPE_SOCKET_FILTER;
-	attr.insn_cnt = (__u32)(sizeof(code_sample) / sizeof(struct bpf_insn));
-	attr.insns = ptr_to_u64(code_sample);
-	attr.license = ptr_to_u64("GPL");
-	attr.log_buf = ptr_to_u64(log);
-	attr.log_size = log_len;
-	attr.log_level = log_level;
-
-	return syscall(__NR_bpf, BPF_PROG_LOAD, &attr, sizeof(attr));
-}
-
-static void check_ret(int ret, int exp_errno)
-{
-	if (ret > 0) {
-		close(ret);
-		err("broken sample loaded successfully!?\n");
-		exit(1);
-	}
-
-	if (!ret || errno != exp_errno) {
-		err("Program load returned: ret:%d/errno:%d, expected ret:%d/errno:%d\n",
-		    ret, errno, -1, exp_errno);
-		exit(1);
-	}
-}
-
-static void check_ones(const char *buf, size_t len, const char *msg)
-{
-	while (len--)
-		if (buf[len] != 1) {
-			err("%s", msg);
-			exit(1);
-		}
-}
-
-static void test_log_good(char *log, size_t buf_len, size_t log_len,
-			  size_t exp_len, int exp_errno, const char *full_log)
-{
-	size_t len;
-	int ret;
-
-	memset(log, 1, buf_len);
-
-	ret = load(log, log_len, 1);
-	check_ret(ret, exp_errno);
-
-	len = strnlen(log, buf_len);
-	if (len == buf_len) {
-		err("verifier did not NULL terminate the log\n");
-		exit(1);
-	}
-	if (exp_len && len != exp_len) {
-		err("incorrect log length expected:%zd have:%zd\n",
-		    exp_len, len);
-		exit(1);
-	}
-
-	if (strchr(log, 1)) {
-		err("verifier leaked a byte through\n");
-		exit(1);
-	}
-
-	check_ones(log + len + 1, buf_len - len - 1,
-		   "verifier wrote bytes past NULL termination\n");
-
-	if (memcmp(full_log, log, LOG_SIZE)) {
-		err("log did not match expected output\n");
-		exit(1);
-	}
-}
-
-static void test_log_bad(char *log, size_t log_len, int log_level)
-{
-	int ret;
-
-	ret = load(log, log_len, log_level);
-	check_ret(ret, EINVAL);
-	if (log)
-		check_ones(log, LOG_SIZE,
-			   "verifier touched log with bad parameters\n");
-}
-
-int main(int argc, char **argv)
-{
-	char full_log[LOG_SIZE];
-	char log[LOG_SIZE];
-	size_t want_len;
-	int i;
-
-	memset(log, 1, LOG_SIZE);
-
-	/* Use libbpf 1.0 API mode */
-	libbpf_set_strict_mode(LIBBPF_STRICT_ALL);
-
-	/* Test incorrect attr */
-	printf("Test log_level 0...\n");
-	test_log_bad(log, LOG_SIZE, 0);
-
-	printf("Test log_size < 128...\n");
-	test_log_bad(log, 15, 1);
-
-	printf("Test log_buff = NULL...\n");
-	test_log_bad(NULL, LOG_SIZE, 1);
-
-	/* Test with log big enough */
-	printf("Test oversized buffer...\n");
-	test_log_good(full_log, LOG_SIZE, LOG_SIZE, 0, EACCES, full_log);
-
-	want_len = strlen(full_log);
-
-	printf("Test exact buffer...\n");
-	test_log_good(log, LOG_SIZE, want_len + 2, want_len, EACCES, full_log);
-
-	printf("Test undersized buffers...\n");
-	for (i = 0; i < 64; i++) {
-		full_log[want_len - i + 1] = 1;
-		full_log[want_len - i] = 0;
-
-		test_log_good(log, LOG_SIZE, want_len + 1 - i, want_len - i,
-			      ENOSPC, full_log);
-	}
-
-	printf("test_verifier_log: OK\n");
-	return 0;
-}
-- 
cgit v1.2.3-70-g09d2


From 0c5f48599bed55c13d6f979ea824554bdebdf2ad Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Mon, 10 Apr 2023 14:18:41 +0200
Subject: xsk: Simplify xp_aligned_validate_desc implementation

Perform the chunk boundary check like the page boundary check in
xp_desc_crosses_non_contig_pg(). This simplifies the implementation and
reduces the number of branches.

Signed-off-by: Kal Conley <kal.conley@dectris.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20230410121841.643254-1-kal.conley@dectris.com
---
 net/xdp/xsk_queue.h | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 66c6f57c9c44..76b574bffd4a 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -133,16 +133,12 @@ static inline bool xskq_cons_read_addr_unchecked(struct xsk_queue *q, u64 *addr)
 static inline bool xp_aligned_validate_desc(struct xsk_buff_pool *pool,
 					    struct xdp_desc *desc)
 {
-	u64 chunk, chunk_end;
+	u64 offset = desc->addr & (pool->chunk_size - 1);
 
-	chunk = xp_aligned_extract_addr(pool, desc->addr);
-	if (likely(desc->len)) {
-		chunk_end = xp_aligned_extract_addr(pool, desc->addr + desc->len - 1);
-		if (chunk != chunk_end)
-			return false;
-	}
+	if (offset + desc->len > pool->chunk_size)
+		return false;
 
-	if (chunk >= pool->addrs_cnt)
+	if (desc->addr >= pool->addrs_cnt)
 		return false;
 
 	if (desc->options)
-- 
cgit v1.2.3-70-g09d2


From 4099be372faf7b3616634dfe6994b81b1edf1906 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 11 Apr 2023 20:46:47 -0700
Subject: selftests/bpf: Fix compiler warnings in bpf_testmod for kfuncs

Add -Wmissing-prototypes ignore in bpf_testmod.c, similarly to what we
do in kernel code proper.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/oe-kbuild-all/202304080951.l14IDv3n-lkp@intel.com/
Link: https://lore.kernel.org/bpf/20230412034647.3968143-1-andrii@kernel.org
---
 tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index c5ad39bbe9af..fe847ebfb731 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -28,6 +28,10 @@ struct bpf_testmod_struct_arg_2 {
 	long b;
 };
 
+__diag_push();
+__diag_ignore_all("-Wmissing-prototypes",
+		  "Global functions as their definitions will be in bpf_testmod.ko BTF");
+
 noinline int
 bpf_testmod_test_struct_arg_1(struct bpf_testmod_struct_arg_2 a, int b, int c) {
 	bpf_testmod_test_struct_arg_result = a.a + a.b  + b + c;
@@ -175,6 +179,8 @@ noinline int bpf_testmod_fentry_test3(char a, int b, u64 c)
 	return a + b + c;
 }
 
+__diag_pop();
+
 int bpf_testmod_fentry_ok;
 
 noinline ssize_t
-- 
cgit v1.2.3-70-g09d2


From 1ba83f505c53d35eda892ac7f108ef1189da6fa8 Mon Sep 17 00:00:00 2001
From: Kal Conley <kal.conley@dectris.com>
Date: Tue, 11 Apr 2023 15:00:25 +0200
Subject: xsk: Elide base_addr comparison in xp_unaligned_validate_desc

Remove redundant (base_addr >= pool->addrs_cnt) comparison from the
conditional.

In particular, addr is computed as:

    addr = base_addr + offset

... where base_addr and offset are stored as 48-bit and 16-bit unsigned
integers, respectively. The above sum cannot overflow u64 since base_addr
has a maximum value of 0x0000ffffffffffff and offset has a maximum value
of 0xffff (implying a maximum sum of 0x000100000000fffe). Since overflow
is impossible, it follows that addr >= base_addr.

Now if (base_addr >= pool->addrs_cnt), then clearly:

    addr >= base_addr
         >= pool->addrs_cnt

Thus, (base_addr >= pool->addrs_cnt) implies (addr >= pool->addrs_cnt).
Subsequently, the former comparison is unnecessary in the conditional
since for any boolean expressions A and B, (A || B) && (A -> B) is
equivalent to B.

Signed-off-by: Kal Conley <kal.conley@dectris.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20230411130025.19704-1-kal.conley@dectris.com
---
 net/xdp/xsk_queue.h | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 76b574bffd4a..6d40a77fccbe 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -149,16 +149,12 @@ static inline bool xp_aligned_validate_desc(struct xsk_buff_pool *pool,
 static inline bool xp_unaligned_validate_desc(struct xsk_buff_pool *pool,
 					      struct xdp_desc *desc)
 {
-	u64 addr, base_addr;
-
-	base_addr = xp_unaligned_extract_addr(desc->addr);
-	addr = xp_unaligned_add_offset_to_addr(desc->addr);
+	u64 addr = xp_unaligned_add_offset_to_addr(desc->addr);
 
 	if (desc->len > pool->chunk_size)
 		return false;
 
-	if (base_addr >= pool->addrs_cnt || addr >= pool->addrs_cnt ||
-	    addr + desc->len > pool->addrs_cnt ||
+	if (addr >= pool->addrs_cnt || addr + desc->len > pool->addrs_cnt ||
 	    xp_desc_crosses_non_contig_pg(pool, addr, desc->len))
 		return false;
 
-- 
cgit v1.2.3-70-g09d2


From 23acb14af1914010dd0aae1bbb7fab28bf518b8e Mon Sep 17 00:00:00 2001
From: Hao Zeng <zenghao@kylinos.cn>
Date: Tue, 11 Apr 2023 16:43:49 +0800
Subject: samples/bpf: Fix fout leak in hbm's run_bpf_prog

Fix fout being fopen'ed but then not subsequently fclose'd. In the affected
branch, fout is otherwise going out of scope.

Signed-off-by: Hao Zeng <zenghao@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230411084349.1999628-1-zenghao@kylinos.cn
---
 samples/bpf/hbm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/samples/bpf/hbm.c b/samples/bpf/hbm.c
index ff58ec43f56a..6448b7826107 100644
--- a/samples/bpf/hbm.c
+++ b/samples/bpf/hbm.c
@@ -315,6 +315,7 @@ static int run_bpf_prog(char *prog, int cg_id)
 		fout = fopen(fname, "w");
 		fprintf(fout, "id:%d\n", cg_id);
 		fprintf(fout, "ERROR: Could not lookup queue_stats\n");
+		fclose(fout);
 	} else if (stats_flag && qstats.lastPacketTime >
 		   qstats.firstPacketTime) {
 		long long delta_us = (qstats.lastPacketTime -
-- 
cgit v1.2.3-70-g09d2


From 634c7b1bd08ca322537ab389f8cbd7bb543b5e45 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 10:44:01 +0300
Subject: wifi: iwlwifi: debug: fix crash in __iwl_err()

In __iwl_err(), if we rate-limit the message away, then
vaf.va is still NULL-initialized by the time we get to
the tracing code, which then crashes. When it doesn't
get rate-limited out, it's still wrong to reuse the old
args2 that was already printed, which is why we bother
making a copy in the first place.

Assign vaf.va properly to fix this.

Fixes: e5f1cc98cc1b ("iwlwifi: allow rate-limited error messages")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.e27134c6bcd4.Ib3894cd2ba7a5ad5e75912a7634f146ceaa569e2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-debug.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-debug.c b/drivers/net/wireless/intel/iwlwifi/iwl-debug.c
index ae4c2a3d63d5..3a3c13a41fc6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-debug.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-debug.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2005-2011, 2021 Intel Corporation
+ * Copyright (C) 2005-2011, 2021-2022 Intel Corporation
  */
 #include <linux/device.h>
 #include <linux/interrupt.h>
@@ -57,6 +57,7 @@ void __iwl_err(struct device *dev, enum iwl_err_mode mode, const char *fmt, ...)
 	default:
 		break;
 	}
+	vaf.va = &args;
 	trace_iwlwifi_err(&vaf);
 	va_end(args);
 }
-- 
cgit v1.2.3-70-g09d2


From 0cc6fb8a0c360de23606d013eb029a8c3e78bcc8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 10:44:02 +0300
Subject: wifi: iwlwifi: nvm-parse: enable 160/320 MHz for AP mode

Initially, 160/320 MHz in AP mode were not supported.
After testing, enable the wider bandwidths in AP mode
as well.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.ed04de3a2833.Ie3991179dfaf24880b96a0904a625dbf6b8fd579@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 923bbfc151dd..449a1922a215 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -876,16 +876,13 @@ iwl_nvm_fixup_sband_iftd(struct iwl_trans *trans,
 				       IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_MASK);
 		break;
 	case NL80211_BAND_6GHZ:
-		if (!is_ap || iwlwifi_mod_params.nvm_file)
-			iftype_data->eht_cap.eht_cap_elem.phy_cap_info[0] |=
-				IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ;
+		iftype_data->eht_cap.eht_cap_elem.phy_cap_info[0] |=
+			IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ;
 		fallthrough;
 	case NL80211_BAND_5GHZ:
 		iftype_data->he_cap.he_cap_elem.phy_cap_info[0] |=
-			IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G;
-		if (!is_ap || iwlwifi_mod_params.nvm_file)
-			iftype_data->he_cap.he_cap_elem.phy_cap_info[0] |=
-				IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G;
+			IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G |
+			IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G;
 		break;
 	default:
 		WARN_ON(1);
-- 
cgit v1.2.3-70-g09d2


From 413be839bfca9a4f6b32b92807ecdc2c2c182d94 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Thu, 13 Apr 2023 10:44:03 +0300
Subject: wifi: iwlwifi: add a validity check of queue_id in iwl_txq_reclaim

This function receives the queue id to reclaim packets from. Currently
we're passing to it the queue id we received from the FW in the flush
response. We don't do any check of this value and it might be invalid.
In such case we will refer to a pointer to a queue which might be NULL.
Fix this by adding a validity check of the queue id before using it.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.a9c3fd32bce7.I5fbdcf3b1b80eb96a907116c166f19dc0aae7cb8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/queue/tx.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/queue/tx.c b/drivers/net/wireless/intel/iwlwifi/queue/tx.c
index 726185d6fab8..d1c39c214f95 100644
--- a/drivers/net/wireless/intel/iwlwifi/queue/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/queue/tx.c
@@ -1554,14 +1554,18 @@ void iwl_txq_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		     struct sk_buff_head *skbs)
 {
 	struct iwl_txq *txq = trans->txqs.txq[txq_id];
-	int tfd_num = iwl_txq_get_cmd_index(txq, ssn);
-	int read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr);
-	int last_to_free;
+	int tfd_num, read_ptr, last_to_free;
 
 	/* This function is not meant to release cmd queue*/
 	if (WARN_ON(txq_id == trans->txqs.cmd.q_id))
 		return;
 
+	if (WARN_ON(!txq))
+		return;
+
+	tfd_num = iwl_txq_get_cmd_index(txq, ssn);
+	read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr);
+
 	spin_lock_bh(&txq->lock);
 
 	if (!test_bit(txq_id, trans->txqs.queue_used)) {
-- 
cgit v1.2.3-70-g09d2


From 0aa27be59e01591a2f5b11fc97fd395dad63c5d3 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Thu, 13 Apr 2023 10:44:04 +0300
Subject: wifi: iwlwifi: Update configuration for SO,SOF MAC and HR RF

update the device configuration for HR1 device for SO and SOF device.
QuZ device configuration is corrected to support specific CRF.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.86f08520323f.Ieccb50de47f877b85732000a0d67b645eeeb0c2a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 86b094683631..3112821dad78 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -995,6 +995,11 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_quz_a0_hr_b0, iwl_ax203_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_QUZ, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_quz_a0_hr_b0, iwl_ax201_name),
 
 /* QnJ with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
@@ -1090,7 +1095,7 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
-		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
@@ -1107,7 +1112,7 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
-		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-- 
cgit v1.2.3-70-g09d2


From 925c6a40e31db33ae22f35cd866786d3164481ac Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 10:44:05 +0300
Subject: wifi: iwlwifi: mvm: convert TID to FW value on queue remove

On queue remove, we should convert the TID value to the
firmware value (8 -> 15) just like we do on queue add.
Otherwise, the firmware will not be able to find the
correct queue to remove.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.6651077eaec3.Ia6868c8fc1a92063609bb057b6a618726712d0bb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c     | 6 +++++-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 78d4f186cd99..ad71233a2299 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -268,6 +268,9 @@ static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm, int sta_id,
 	int queue = *queueptr;
 	int ret = 0;
 
+	if (tid == IWL_MAX_TID_COUNT)
+		tid = IWL_MGMT_TID;
+
 	if (mvm->sta_remove_requires_queue_remove) {
 		u32 cmd_id = WIDE_ID(DATA_PATH_GROUP,
 				     SCD_QUEUE_CONFIG_CMD);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 50e224883af0..6b2690534f59 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -356,10 +356,14 @@ static int iwl_mvm_disable_txq(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 					     SCD_QUEUE_CONFIG_CMD);
 			struct iwl_scd_queue_cfg_cmd remove_cmd = {
 				.operation = cpu_to_le32(IWL_SCD_QUEUE_REMOVE),
-				.u.remove.tid = cpu_to_le32(tid),
 				.u.remove.sta_mask = cpu_to_le32(BIT(sta_id)),
 			};
 
+			if (tid == IWL_MAX_TID_COUNT)
+				tid = IWL_MGMT_TID;
+
+			remove_cmd.u.remove.tid = cpu_to_le32(tid);
+
 			ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0,
 						   sizeof(remove_cmd),
 						   &remove_cmd);
-- 
cgit v1.2.3-70-g09d2


From 4e17e156051be8c2223af7bc0267fb3d325d277d Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Thu, 13 Apr 2023 10:44:06 +0300
Subject: wifi: iwlwifi: mvm: use OFDM rate if IEEE80211_TX_CTL_NO_CCK_RATE is
 set

IEEE80211_TX_CTL_NO_CCK_RATE indicates that CCK rates should not be
used, but is ignored by the driver. Fix it.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.a322d18b5eb1.Icc46027a03f90feffb6fab49a5d82e54829d3dd9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 2c842938656d..9a6e7dd166c0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -343,7 +343,8 @@ static u32 iwl_mvm_get_tx_rate(struct iwl_mvm *mvm,
 	 * For non 2 GHZ band, remap mac80211 rate
 	 * indices into driver indices
 	 */
-	if (info->band != NL80211_BAND_2GHZ)
+	if (info->band != NL80211_BAND_2GHZ ||
+	    (info->flags & IEEE80211_TX_CTL_NO_CCK_RATE))
 		rate_idx += IWL_FIRST_OFDM_RATE;
 
 	/* For 2.4 GHZ band, check that there is no need to remap */
-- 
cgit v1.2.3-70-g09d2


From 72429d68a0e88a1a19bbc61ed4c35a3b6baba628 Mon Sep 17 00:00:00 2001
From: Golan Ben Ami <golan.ben.ami@intel.com>
Date: Thu, 13 Apr 2023 10:44:07 +0300
Subject: wifi: iwlwifi: mvm: enable bz hw checksum from c step

B step doesn't support full checksum yet, move to c step.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.697a9d74e84d.I6724874112692a04e29287cac9dad7140532557f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 9a6e7dd166c0..b3545c3273e5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -187,7 +187,7 @@ static u32 iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (mvm->trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_BZ ||
 	    (mvm->trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
 	     CSR_HW_REV_TYPE(mvm->trans->hw_rev) == IWL_CFG_MAC_TYPE_GL &&
-	     mvm->trans->hw_rev_step == SILICON_A_STEP))
+	     mvm->trans->hw_rev_step <= SILICON_B_STEP))
 		return iwl_mvm_tx_csum_pre_bz(mvm, skb, info, amsdu);
 	return iwl_mvm_tx_csum_bz(mvm, skb, amsdu);
 }
-- 
cgit v1.2.3-70-g09d2


From d34d11aea2d5d71a66f2f90507c3f7202b964175 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 10:44:08 +0300
Subject: wifi: iwlwifi: mvm: fix A-MSDU checks

Since Gl A-step devices use the old checksum hardware,
we shouldn't use the Bz code to check for A-MSDU
combining ability; fix that.

Fixes: ec18e7d4d20d ("wifi: iwlwifi: mvm: use old checksum for Bz A-step")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.8c445b943fee.Ibf772102ca712f59e2ee0cdd4c344011fcf445aa@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index aaa7e3c561a0..0ee5169de034 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -6129,7 +6129,10 @@ static bool iwl_mvm_mac_can_aggregate(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	if (mvm->trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ)
+	if (mvm->trans->trans_cfg->device_family > IWL_DEVICE_FAMILY_BZ ||
+	    (mvm->trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
+	     !(CSR_HW_REV_TYPE(mvm->trans->hw_rev) == IWL_CFG_MAC_TYPE_GL &&
+	       mvm->trans->hw_rev_step == SILICON_A_STEP)))
 		return iwl_mvm_tx_csum_bz(mvm, head, true) ==
 		       iwl_mvm_tx_csum_bz(mvm, skb, true);
 
-- 
cgit v1.2.3-70-g09d2


From a0c8ab93eb3edf82511502be71a825f74b632137 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 10:44:09 +0300
Subject: wifi: iwlwifi: mvm: refactor TX csum mode check

There are two modes now, and we have two places checking
that must be in sync. Refactor the logic into a new small
helper function.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.ef6246f4b73b.I44820ec095634dd0bba3007465cf25e4ce1c77c6@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  5 +----
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      | 13 +++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c       |  5 +----
 3 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 0ee5169de034..fc15636beec0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -6129,10 +6129,7 @@ static bool iwl_mvm_mac_can_aggregate(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	if (mvm->trans->trans_cfg->device_family > IWL_DEVICE_FAMILY_BZ ||
-	    (mvm->trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
-	     !(CSR_HW_REV_TYPE(mvm->trans->hw_rev) == IWL_CFG_MAC_TYPE_GL &&
-	       mvm->trans->hw_rev_step == SILICON_A_STEP)))
+	if (iwl_mvm_has_new_tx_csum(mvm))
 		return iwl_mvm_tx_csum_bz(mvm, head, true) ==
 		       iwl_mvm_tx_csum_bz(mvm, skb, true);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index e32ce876951f..76ac8a480712 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1522,6 +1522,19 @@ static inline bool iwl_mvm_is_ctdp_supported(struct iwl_mvm *mvm)
 			   IWL_UCODE_TLV_CAPA_CTDP_SUPPORT);
 }
 
+static inline bool iwl_mvm_has_new_tx_csum(struct iwl_mvm *mvm)
+{
+	if (mvm->trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_BZ)
+		return false;
+
+	if (mvm->trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
+	    CSR_HW_REV_TYPE(mvm->trans->hw_rev) == IWL_CFG_MAC_TYPE_GL &&
+	    mvm->trans->hw_rev_step <= SILICON_B_STEP)
+		return false;
+
+	return true;
+}
+
 extern const u8 iwl_mvm_ac_to_tx_fifo[];
 extern const u8 iwl_mvm_ac_to_gen2_tx_fifo[];
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index b3545c3273e5..51f21cbf7a20 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -184,10 +184,7 @@ static u32 iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 			   struct ieee80211_tx_info *info,
 			   bool amsdu)
 {
-	if (mvm->trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_BZ ||
-	    (mvm->trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
-	     CSR_HW_REV_TYPE(mvm->trans->hw_rev) == IWL_CFG_MAC_TYPE_GL &&
-	     mvm->trans->hw_rev_step <= SILICON_B_STEP))
+	if (!iwl_mvm_has_new_tx_csum(mvm))
 		return iwl_mvm_tx_csum_pre_bz(mvm, skb, info, amsdu);
 	return iwl_mvm_tx_csum_bz(mvm, skb, amsdu);
 }
-- 
cgit v1.2.3-70-g09d2


From cb75abcebc458efe3b329372e0a57c15f7bdaee0 Mon Sep 17 00:00:00 2001
From: Daniel Gabay <daniel.gabay@intel.com>
Date: Thu, 13 Apr 2023 10:44:10 +0300
Subject: wifi: iwlwifi: nvm: Update HE capabilities on 6GHz band for EHT
 device

Max A-MPDU length exponent shall be set to 2 for EHT capable
device on 6GHz band in order to support 4MB aggregation.

Update HE MAC capabilities accordingly for station and softap
interfaces.

This change requires to add another ieee80211_sband_iftype_data for
uhb since high/uhb are no longer the same.

Signed-off-by: Daniel Gabay <daniel.gabay@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.1eee32cfd199.I9c5ff3a6956d509137deca620814935149516fbc@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h |  5 ++---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c    | 10 ++++++++--
 2 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h
index baa643386018..0e8ca761d24b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h
@@ -47,13 +47,12 @@ struct iwl_nvm_data {
 	struct ieee80211_supported_band bands[NUM_NL80211_BANDS];
 
 	/*
-	 * iftype data for low (2.4 GHz) and high (5 and 6 GHz) bands,
-	 * we can use the same for 5 and 6 GHz bands because they have
-	 * the same data
+	 * iftype data for low (2.4 GHz) high (5 GHz) and uhb (6 GHz) bands
 	 */
 	struct {
 		struct ieee80211_sband_iftype_data low[2];
 		struct ieee80211_sband_iftype_data high[2];
+		struct ieee80211_sband_iftype_data uhb[2];
 	} iftd;
 
 	struct ieee80211_channel channels[];
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 449a1922a215..92a79df3e6d2 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -860,7 +860,10 @@ iwl_nvm_fixup_sband_iftd(struct iwl_trans *trans,
 	/* Advertise an A-MPDU exponent extension based on
 	 * operating band
 	 */
-	if (sband->band != NL80211_BAND_2GHZ)
+	if (sband->band == NL80211_BAND_6GHZ && iftype_data->eht_cap.has_eht)
+		iftype_data->he_cap.he_cap_elem.mac_cap_info[3] |=
+			IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_2;
+	else if (sband->band != NL80211_BAND_2GHZ)
 		iftype_data->he_cap.he_cap_elem.mac_cap_info[3] |=
 			IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_1;
 	else
@@ -996,15 +999,18 @@ static void iwl_init_he_hw_capab(struct iwl_trans *trans,
 
 	BUILD_BUG_ON(sizeof(data->iftd.low) != sizeof(iwl_he_eht_capa));
 	BUILD_BUG_ON(sizeof(data->iftd.high) != sizeof(iwl_he_eht_capa));
+	BUILD_BUG_ON(sizeof(data->iftd.uhb) != sizeof(iwl_he_eht_capa));
 
 	switch (sband->band) {
 	case NL80211_BAND_2GHZ:
 		iftype_data = data->iftd.low;
 		break;
 	case NL80211_BAND_5GHZ:
-	case NL80211_BAND_6GHZ:
 		iftype_data = data->iftd.high;
 		break;
+	case NL80211_BAND_6GHZ:
+		iftype_data = data->iftd.uhb;
+		break;
 	default:
 		WARN_ON(1);
 		return;
-- 
cgit v1.2.3-70-g09d2


From e305a408c5f9c8df06a58e09a23375fd257f081c Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Thu, 13 Apr 2023 10:44:11 +0300
Subject: wifi: iwlwifi: mvm: move function sequence

Need to move a function definition and actual changes
will be done in following commit.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.00a6c203712f.I7c099e5c1954f1daa5a5039b98149b6f081e46ae@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 210 ++++++++++++++--------------
 1 file changed, 105 insertions(+), 105 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 7fe733dcc748..a6b11bccb0c1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -467,111 +467,6 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 	return 0;
 }
 
-static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
-{
-	struct iwl_notification_wait init_wait;
-	struct iwl_nvm_access_complete_cmd nvm_complete = {};
-	struct iwl_init_extended_cfg_cmd init_cfg = {
-		.init_flags = cpu_to_le32(BIT(IWL_INIT_NVM)),
-	};
-	static const u16 init_complete[] = {
-		INIT_COMPLETE_NOTIF,
-	};
-	int ret;
-
-	if (mvm->trans->cfg->tx_with_siso_diversity)
-		init_cfg.init_flags |= cpu_to_le32(BIT(IWL_INIT_PHY));
-
-	lockdep_assert_held(&mvm->mutex);
-
-	mvm->rfkill_safe_init_done = false;
-
-	iwl_init_notification_wait(&mvm->notif_wait,
-				   &init_wait,
-				   init_complete,
-				   ARRAY_SIZE(init_complete),
-				   iwl_wait_init_complete,
-				   NULL);
-
-	iwl_dbg_tlv_time_point(&mvm->fwrt, IWL_FW_INI_TIME_POINT_EARLY, NULL);
-
-	/* Will also start the device */
-	ret = iwl_mvm_load_ucode_wait_alive(mvm, IWL_UCODE_REGULAR);
-	if (ret) {
-		IWL_ERR(mvm, "Failed to start RT ucode: %d\n", ret);
-		goto error;
-	}
-	iwl_dbg_tlv_time_point(&mvm->fwrt, IWL_FW_INI_TIME_POINT_AFTER_ALIVE,
-			       NULL);
-
-	/* Send init config command to mark that we are sending NVM access
-	 * commands
-	 */
-	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(SYSTEM_GROUP,
-						INIT_EXTENDED_CFG_CMD),
-				   CMD_SEND_IN_RFKILL,
-				   sizeof(init_cfg), &init_cfg);
-	if (ret) {
-		IWL_ERR(mvm, "Failed to run init config command: %d\n",
-			ret);
-		goto error;
-	}
-
-	/* Load NVM to NIC if needed */
-	if (mvm->nvm_file_name) {
-		ret = iwl_read_external_nvm(mvm->trans, mvm->nvm_file_name,
-					    mvm->nvm_sections);
-		if (ret)
-			goto error;
-		ret = iwl_mvm_load_nvm_to_nic(mvm);
-		if (ret)
-			goto error;
-	}
-
-	if (IWL_MVM_PARSE_NVM && !mvm->nvm_data) {
-		ret = iwl_nvm_init(mvm);
-		if (ret) {
-			IWL_ERR(mvm, "Failed to read NVM: %d\n", ret);
-			goto error;
-		}
-	}
-
-	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(REGULATORY_AND_NVM_GROUP,
-						NVM_ACCESS_COMPLETE),
-				   CMD_SEND_IN_RFKILL,
-				   sizeof(nvm_complete), &nvm_complete);
-	if (ret) {
-		IWL_ERR(mvm, "Failed to run complete NVM access: %d\n",
-			ret);
-		goto error;
-	}
-
-	/* We wait for the INIT complete notification */
-	ret = iwl_wait_notification(&mvm->notif_wait, &init_wait,
-				    MVM_UCODE_ALIVE_TIMEOUT);
-	if (ret)
-		return ret;
-
-	/* Read the NVM only at driver load time, no need to do this twice */
-	if (!IWL_MVM_PARSE_NVM && !mvm->nvm_data) {
-		mvm->nvm_data = iwl_get_nvm(mvm->trans, mvm->fw);
-		if (IS_ERR(mvm->nvm_data)) {
-			ret = PTR_ERR(mvm->nvm_data);
-			mvm->nvm_data = NULL;
-			IWL_ERR(mvm, "Failed to read NVM: %d\n", ret);
-			return ret;
-		}
-	}
-
-	mvm->rfkill_safe_init_done = true;
-
-	return 0;
-
-error:
-	iwl_remove_notification(&mvm->notif_wait, &init_wait);
-	return ret;
-}
-
 #ifdef CONFIG_ACPI
 static void iwl_mvm_phy_filter_init(struct iwl_mvm *mvm,
 				    struct iwl_phy_specific_cfg *phy_filters)
@@ -698,6 +593,111 @@ static int iwl_send_phy_cfg_cmd(struct iwl_mvm *mvm)
 	return iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0, cmd_size, &phy_cfg_cmd);
 }
 
+static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
+{
+	struct iwl_notification_wait init_wait;
+	struct iwl_nvm_access_complete_cmd nvm_complete = {};
+	struct iwl_init_extended_cfg_cmd init_cfg = {
+		.init_flags = cpu_to_le32(BIT(IWL_INIT_NVM)),
+	};
+	static const u16 init_complete[] = {
+		INIT_COMPLETE_NOTIF,
+	};
+	int ret;
+
+	if (mvm->trans->cfg->tx_with_siso_diversity)
+		init_cfg.init_flags |= cpu_to_le32(BIT(IWL_INIT_PHY));
+
+	lockdep_assert_held(&mvm->mutex);
+
+	mvm->rfkill_safe_init_done = false;
+
+	iwl_init_notification_wait(&mvm->notif_wait,
+				   &init_wait,
+				   init_complete,
+				   ARRAY_SIZE(init_complete),
+				   iwl_wait_init_complete,
+				   NULL);
+
+	iwl_dbg_tlv_time_point(&mvm->fwrt, IWL_FW_INI_TIME_POINT_EARLY, NULL);
+
+	/* Will also start the device */
+	ret = iwl_mvm_load_ucode_wait_alive(mvm, IWL_UCODE_REGULAR);
+	if (ret) {
+		IWL_ERR(mvm, "Failed to start RT ucode: %d\n", ret);
+		goto error;
+	}
+	iwl_dbg_tlv_time_point(&mvm->fwrt, IWL_FW_INI_TIME_POINT_AFTER_ALIVE,
+			       NULL);
+
+	/* Send init config command to mark that we are sending NVM access
+	 * commands
+	 */
+	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(SYSTEM_GROUP,
+						INIT_EXTENDED_CFG_CMD),
+				   CMD_SEND_IN_RFKILL,
+				   sizeof(init_cfg), &init_cfg);
+	if (ret) {
+		IWL_ERR(mvm, "Failed to run init config command: %d\n",
+			ret);
+		goto error;
+	}
+
+	/* Load NVM to NIC if needed */
+	if (mvm->nvm_file_name) {
+		ret = iwl_read_external_nvm(mvm->trans, mvm->nvm_file_name,
+					    mvm->nvm_sections);
+		if (ret)
+			goto error;
+		ret = iwl_mvm_load_nvm_to_nic(mvm);
+		if (ret)
+			goto error;
+	}
+
+	if (IWL_MVM_PARSE_NVM && !mvm->nvm_data) {
+		ret = iwl_nvm_init(mvm);
+		if (ret) {
+			IWL_ERR(mvm, "Failed to read NVM: %d\n", ret);
+			goto error;
+		}
+	}
+
+	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(REGULATORY_AND_NVM_GROUP,
+						NVM_ACCESS_COMPLETE),
+				   CMD_SEND_IN_RFKILL,
+				   sizeof(nvm_complete), &nvm_complete);
+	if (ret) {
+		IWL_ERR(mvm, "Failed to run complete NVM access: %d\n",
+			ret);
+		goto error;
+	}
+
+	/* We wait for the INIT complete notification */
+	ret = iwl_wait_notification(&mvm->notif_wait, &init_wait,
+				    MVM_UCODE_ALIVE_TIMEOUT);
+	if (ret)
+		return ret;
+
+	/* Read the NVM only at driver load time, no need to do this twice */
+	if (!IWL_MVM_PARSE_NVM && !mvm->nvm_data) {
+		mvm->nvm_data = iwl_get_nvm(mvm->trans, mvm->fw);
+		if (IS_ERR(mvm->nvm_data)) {
+			ret = PTR_ERR(mvm->nvm_data);
+			mvm->nvm_data = NULL;
+			IWL_ERR(mvm, "Failed to read NVM: %d\n", ret);
+			return ret;
+		}
+	}
+
+	mvm->rfkill_safe_init_done = true;
+
+	return 0;
+
+error:
+	iwl_remove_notification(&mvm->notif_wait, &init_wait);
+	return ret;
+}
+
 int iwl_run_init_mvm_ucode(struct iwl_mvm *mvm)
 {
 	struct iwl_notification_wait calib_wait;
-- 
cgit v1.2.3-70-g09d2


From d2ccc5c1526396aef5196e4dd2ec8538994e5072 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Thu, 13 Apr 2023 10:44:12 +0300
Subject: wifi: iwlwifi: Update init sequence if tx diversity supported

PHY configuration command need to be sent to FW if the tx diversity
with SISO is supported.
This need to be sent to get the init notification from FW.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.c2121c8694a7.Ibee3dd8765ef4b7504660fa228a7c7eff78920af@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index a6b11bccb0c1..521eded39028 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -672,6 +672,13 @@ static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
 		goto error;
 	}
 
+	ret = iwl_send_phy_cfg_cmd(mvm);
+	if (ret) {
+		IWL_ERR(mvm, "Failed to run PHY configuration: %d\n",
+			ret);
+		goto error;
+	}
+
 	/* We wait for the INIT complete notification */
 	ret = iwl_wait_notification(&mvm->notif_wait, &init_wait,
 				    MVM_UCODE_ALIVE_TIMEOUT);
@@ -1538,12 +1545,11 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 		ret = iwl_send_phy_db_data(mvm->phy_db);
 		if (ret)
 			goto error;
+		ret = iwl_send_phy_cfg_cmd(mvm);
+		if (ret)
+			goto error;
 	}
 
-	ret = iwl_send_phy_cfg_cmd(mvm);
-	if (ret)
-		goto error;
-
 	ret = iwl_mvm_send_bt_init_conf(mvm);
 	if (ret)
 		goto error;
-- 
cgit v1.2.3-70-g09d2


From 5caa82529eabff044c5f3eb254f3a7ff00e9170d Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Thu, 13 Apr 2023 10:44:13 +0300
Subject: wifi: iwlwifi: Update configurations for Bnj-a0 and specific rf
 devices

Correction in config data is done for loading the ucode.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.879b654c8d83.I7dbea9f411a0b6f47908c4ad6321c7e55cbeb636@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 3112821dad78..70f6c59eec5d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1235,7 +1235,7 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_NO_JACKET,
 		      iwl_cfg_bnj_a0_hr_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-- 
cgit v1.2.3-70-g09d2


From 277f56a141fc54ef7f9e09dba65fb2e12021411d Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Thu, 13 Apr 2023 10:44:14 +0300
Subject: wifi: iwlwifi: trans: don't trigger d3 interrupt twice

When the IPC registers are used for sleep control, setting
the IPC sleep bit already triggers an interrupt to the fw, so
there is no need to also set the doorbell. Setting also the
doorbell triggers the sleep interrupt twice which lead to
an assert.

Fixes: af08571d3925 ("iwlwifi: pcie: support Bz suspend/resume trigger")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.b5f2f6e44d38.I4cb5b6ad4914db47a714e731c4c8b4db679cabce@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 0a9af1ad1f20..40283fe622da 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1522,19 +1522,16 @@ static int iwl_pcie_d3_handshake(struct iwl_trans *trans, bool suspend)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int ret;
 
-	if (trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_AX210) {
+	if (trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_AX210)
 		iwl_write_umac_prph(trans, UREG_DOORBELL_TO_ISR6,
 				    suspend ? UREG_DOORBELL_TO_ISR6_SUSPEND :
 					      UREG_DOORBELL_TO_ISR6_RESUME);
-	} else if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) {
+	else if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ)
 		iwl_write32(trans, CSR_IPC_SLEEP_CONTROL,
 			    suspend ? CSR_IPC_SLEEP_CONTROL_SUSPEND :
 				      CSR_IPC_SLEEP_CONTROL_RESUME);
-		iwl_write_umac_prph(trans, UREG_DOORBELL_TO_ISR6,
-				    UREG_DOORBELL_TO_ISR6_SLEEP_CTRL);
-	} else {
+	else
 		return 0;
-	}
 
 	ret = wait_event_timeout(trans_pcie->sx_waitq,
 				 trans_pcie->sx_complete, 2 * HZ);
-- 
cgit v1.2.3-70-g09d2


From 217f3c52f00d3419ecdd38a99a7eceecb11679b2 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Thu, 13 Apr 2023 10:44:15 +0300
Subject: wifi: iwlwifi: mvm: don't set CHECKSUM_COMPLETE for unsupported
 protocols

On Bz devices, CHECKSUM_COMPLETE was set for unsupported protocols
which results in a warning. Fix it.

Fixes: b6f5b647f694 ("iwlwifi: mvm: handle RX checksum on Bz devices")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413102635.a2a35286f0ca.I50daa9445a6465514c44f5096c32adef64beba5f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 5d803e537b00..31e1d0ccdec5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -172,8 +172,7 @@ static int iwl_mvm_create_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 	 * Starting from Bz hardware, it calculates starting directly after
 	 * the MAC header, so that matches mac80211's expectation.
 	 */
-	if (skb->ip_summed == CHECKSUM_COMPLETE &&
-	    mvm->trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_BZ) {
+	if (skb->ip_summed == CHECKSUM_COMPLETE) {
 		struct {
 			u8 hdr[6];
 			__be16 type;
@@ -188,7 +187,7 @@ static int iwl_mvm_create_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 			      shdr->type != htons(ETH_P_PAE) &&
 			      shdr->type != htons(ETH_P_TDLS))))
 			skb->ip_summed = CHECKSUM_NONE;
-		else
+		else if (mvm->trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_BZ)
 			/* mac80211 assumes full CSUM including SNAP header */
 			skb_postpush_rcsum(skb, shdr, sizeof(*shdr));
 	}
-- 
cgit v1.2.3-70-g09d2


From 4fdeb847130229dc94befa241461669c7359776b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 31 Mar 2023 16:59:07 +0200
Subject: wifi: ieee80211: clean up public action codes

WLAN_PUBLIC_ACTION_FTM_RESPONSE is duplicated with
WLAN_PUB_ACTION_FTM, but that might better be called
WLAN_PUB_ACTION_FTM_RESPONSE; clean up here.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 2463bdd2a382..0583b2b0ce1f 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -9,7 +9,7 @@
  * Copyright (c) 2006, Michael Wu <flamingice@sourmilk.net>
  * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH
  * Copyright (c) 2016 - 2017 Intel Deutschland GmbH
- * Copyright (c) 2018 - 2022 Intel Corporation
+ * Copyright (c) 2018 - 2023 Intel Corporation
  */
 
 #ifndef LINUX_IEEE80211_H
@@ -3557,11 +3557,6 @@ enum ieee80211_unprotected_wnm_actioncode {
 	WLAN_UNPROTECTED_WNM_ACTION_TIMING_MEASUREMENT_RESPONSE = 1,
 };
 
-/* Public action codes */
-enum ieee80211_public_actioncode {
-	WLAN_PUBLIC_ACTION_FTM_RESPONSE = 33,
-};
-
 /* Security key length */
 enum ieee80211_key_len {
 	WLAN_KEY_LEN_WEP40 = 5,
@@ -3653,7 +3648,7 @@ enum ieee80211_pub_actioncode {
 	WLAN_PUB_ACTION_NETWORK_CHANNEL_CONTROL = 30,
 	WLAN_PUB_ACTION_WHITE_SPACE_MAP_ANN = 31,
 	WLAN_PUB_ACTION_FTM_REQUEST = 32,
-	WLAN_PUB_ACTION_FTM = 33,
+	WLAN_PUB_ACTION_FTM_RESPONSE = 33,
 	WLAN_PUB_ACTION_FILS_DISCOVERY = 34,
 };
 
@@ -4383,7 +4378,7 @@ static inline bool ieee80211_is_ftm(struct sk_buff *skb)
 		return false;
 
 	if (mgmt->u.action.u.ftm.action_code ==
-		WLAN_PUBLIC_ACTION_FTM_RESPONSE &&
+		WLAN_PUB_ACTION_FTM_RESPONSE &&
 	    skb->len >= offsetofend(typeof(*mgmt), u.action.u.ftm))
 		return true;
 
-- 
cgit v1.2.3-70-g09d2


From 2c9abe653bc5134eeab411c46dde008d8a1c37b0 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 31 Mar 2023 16:59:08 +0200
Subject: wifi: ieee80211: correctly mark FTM frames non-bufferable

The checks of whether or not a frame is bufferable were not
taking into account that some action frames aren't, such as
FTM. Check this, which requires some changes to the function
ieee80211_is_bufferable_mmpdu() since we need the whole skb
for the checks now.

Reviewed-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c |  7 ++--
 drivers/net/wireless/mediatek/mt76/tx.c     |  2 +-
 include/linux/ieee80211.h                   | 52 +++++++++++++++++++++--------
 net/mac80211/tx.c                           |  4 +--
 4 files changed, 45 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 51f21cbf7a20..478442e16d43 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -604,8 +604,9 @@ static void iwl_mvm_skb_prepare_status(struct sk_buff *skb,
 static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 				      struct iwl_mvm_vif_link_info *link,
 				      struct ieee80211_tx_info *info,
-				      struct ieee80211_hdr *hdr)
+				      struct sk_buff *skb)
 {
+	struct ieee80211_hdr *hdr = (void *)skb->data;
 	__le16 fc = hdr->frame_control;
 
 	switch (info->control.vif->type) {
@@ -622,7 +623,7 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 		 * reason 7 ("Class 3 frame received from nonassociated STA").
 		 */
 		if (ieee80211_is_mgmt(fc) &&
-		    (!ieee80211_is_bufferable_mmpdu(fc) ||
+		    (!ieee80211_is_bufferable_mmpdu(skb) ||
 		     ieee80211_is_deauth(fc) || ieee80211_is_disassoc(fc)))
 			return link->mgmt_queue;
 
@@ -755,7 +756,7 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 				sta_id = link->mcast_sta.sta_id;
 
 			queue = iwl_mvm_get_ctrl_vif_queue(mvm, link, &info,
-							   hdr);
+							   skb);
 		} else if (info.control.vif->type == NL80211_IFTYPE_MONITOR) {
 			queue = mvm->snif_queue;
 			sta_id = mvm->snif_sta.sta_id;
diff --git a/drivers/net/wireless/mediatek/mt76/tx.c b/drivers/net/wireless/mediatek/mt76/tx.c
index 1f309d05380a..3ad9742364ba 100644
--- a/drivers/net/wireless/mediatek/mt76/tx.c
+++ b/drivers/net/wireless/mediatek/mt76/tx.c
@@ -330,7 +330,7 @@ mt76_tx(struct mt76_phy *phy, struct ieee80211_sta *sta,
 	if ((dev->drv->drv_flags & MT_DRV_HW_MGMT_TXQ) &&
 	    !(info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP) &&
 	    !ieee80211_is_data(hdr->frame_control) &&
-	    !ieee80211_is_bufferable_mmpdu(hdr->frame_control)) {
+	    !ieee80211_is_bufferable_mmpdu(skb)) {
 		qid = MT_TXQ_PSD;
 	}
 
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 0583b2b0ce1f..c4cf296e7eaf 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -782,20 +782,6 @@ static inline bool ieee80211_is_any_nullfunc(__le16 fc)
 	return (ieee80211_is_nullfunc(fc) || ieee80211_is_qos_nullfunc(fc));
 }
 
-/**
- * ieee80211_is_bufferable_mmpdu - check if frame is bufferable MMPDU
- * @fc: frame control field in little-endian byteorder
- */
-static inline bool ieee80211_is_bufferable_mmpdu(__le16 fc)
-{
-	/* IEEE 802.11-2012, definition of "bufferable management frame";
-	 * note that this ignores the IBSS special case. */
-	return ieee80211_is_mgmt(fc) &&
-	       (ieee80211_is_action(fc) ||
-		ieee80211_is_disassoc(fc) ||
-		ieee80211_is_deauth(fc));
-}
-
 /**
  * ieee80211_is_first_frag - check if IEEE80211_SCTL_FRAG is not set
  * @seq_ctrl: frame sequence control bytes in little-endian byteorder
@@ -4132,6 +4118,44 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 		return hdr->addr1;
 }
 
+/**
+ * ieee80211_is_bufferable_mmpdu - check if frame is bufferable MMPDU
+ * @skb: the skb to check, starting with the 802.11 header
+ */
+static inline bool ieee80211_is_bufferable_mmpdu(struct sk_buff *skb)
+{
+	struct ieee80211_mgmt *mgmt = (void *)skb->data;
+	__le16 fc = mgmt->frame_control;
+
+	/*
+	 * IEEE 802.11 REVme D2.0 definition of bufferable MMPDU;
+	 * note that this ignores the IBSS special case.
+	 */
+	if (!ieee80211_is_mgmt(fc))
+		return false;
+
+	if (ieee80211_is_disassoc(fc) || ieee80211_is_deauth(fc))
+		return true;
+
+	if (!ieee80211_is_action(fc))
+		return false;
+
+	if (skb->len < offsetofend(typeof(*mgmt), u.action.u.ftm.action_code))
+		return true;
+
+	/* action frame - additionally check for non-bufferable FTM */
+
+	if (mgmt->u.action.category != WLAN_CATEGORY_PUBLIC &&
+	    mgmt->u.action.category != WLAN_CATEGORY_PROTECTED_DUAL_OF_ACTION)
+		return true;
+
+	if (mgmt->u.action.u.ftm.action_code == WLAN_PUB_ACTION_FTM_REQUEST ||
+	    mgmt->u.action.u.ftm.action_code == WLAN_PUB_ACTION_FTM_RESPONSE)
+		return false;
+
+	return true;
+}
+
 /**
  * _ieee80211_is_robust_mgmt_frame - check if frame is a robust management frame
  * @hdr: the frame (buffer must include at least the first octet of payload)
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index dfe6b9c9b29e..1a3327407552 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -488,7 +488,7 @@ ieee80211_tx_h_unicast_ps_buf(struct ieee80211_tx_data *tx)
 		int ac = skb_get_queue_mapping(tx->skb);
 
 		if (ieee80211_is_mgmt(hdr->frame_control) &&
-		    !ieee80211_is_bufferable_mmpdu(hdr->frame_control)) {
+		    !ieee80211_is_bufferable_mmpdu(tx->skb)) {
 			info->flags |= IEEE80211_TX_CTL_NO_PS_BUFFER;
 			return TX_CONTINUE;
 		}
@@ -1323,7 +1323,7 @@ static struct txq_info *ieee80211_get_txq(struct ieee80211_local *local,
 	if (!(info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP) &&
 	    unlikely(!ieee80211_is_data_present(hdr->frame_control))) {
 		if ((!ieee80211_is_mgmt(hdr->frame_control) ||
-		     ieee80211_is_bufferable_mmpdu(hdr->frame_control) ||
+		     ieee80211_is_bufferable_mmpdu(skb) ||
 		     vif->type == NL80211_IFTYPE_STATION) &&
 		    sta && sta->uploaded) {
 			/*
-- 
cgit v1.2.3-70-g09d2


From 0b75a1b1e42e07ae84e3a11d2368b418546e2bec Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 31 Mar 2023 16:59:16 +0200
Subject: wifi: mac80211: flush queues on STA removal

When we remove a station, we first make it unreachable,
then we (must) remove its keys, and then remove the
station itself. Depending on the hardware design, if
we have hardware crypto at all, frames still sitting
on hardware queues may then be transmitted without a
valid key, possibly unencrypted or with a fixed key.

Fix this by flushing the queues when removing stations
so this cannot happen.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/sta_info.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 941bda9141fa..ce7c3b997269 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -1294,6 +1294,14 @@ static void __sta_info_destroy_part2(struct sta_info *sta)
 		WARN_ON_ONCE(ret);
 	}
 
+	/* Flush queues before removing keys, as that might remove them
+	 * from hardware, and then depending on the offload method, any
+	 * frames sitting on hardware queues might be sent out without
+	 * any encryption at all.
+	 */
+	if (local->ops->set_key)
+		ieee80211_flush_queues(local, sta->sdata, false);
+
 	/* now keys can no longer be reached */
 	ieee80211_free_sta_keys(local, sta);
 
-- 
cgit v1.2.3-70-g09d2


From d00800a289c9349bb659a698cbd7bc04521dc927 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 31 Mar 2023 16:59:17 +0200
Subject: wifi: mac80211: add flush_sta method

Some drivers like iwlwifi might have per-STA queues, so we
may want to flush/drop just those queues rather than all
when removing a station. Add a separate method for that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h    |  6 ++++++
 net/mac80211/driver-ops.h | 15 +++++++++++++++
 net/mac80211/sta_info.c   |  8 ++++++--
 net/mac80211/trace.h      |  7 +++++++
 4 files changed, 34 insertions(+), 2 deletions(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 679421d37a42..a8dadbd83d95 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -3950,6 +3950,10 @@ struct ieee80211_prep_tx_info {
  *	Note that vif can be NULL.
  *	The callback can sleep.
  *
+ * @flush_sta: Flush or drop all pending frames from the hardware queue(s) for
+ *	the given station, as it's about to be removed.
+ *	The callback can sleep.
+ *
  * @channel_switch: Drivers that need (or want) to offload the channel
  *	switch operation for CSAs received from the AP may implement this
  *	callback. They must then call ieee80211_chswitch_done() to indicate
@@ -4415,6 +4419,8 @@ struct ieee80211_ops {
 #endif
 	void (*flush)(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		      u32 queues, bool drop);
+	void (*flush_sta)(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			  struct ieee80211_sta *sta);
 	void (*channel_switch)(struct ieee80211_hw *hw,
 			       struct ieee80211_vif *vif,
 			       struct ieee80211_channel_switch *ch_switch);
diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h
index 0bf208f5bbc5..45d3e53c7383 100644
--- a/net/mac80211/driver-ops.h
+++ b/net/mac80211/driver-ops.h
@@ -649,6 +649,21 @@ static inline void drv_flush(struct ieee80211_local *local,
 	trace_drv_return_void(local);
 }
 
+static inline void drv_flush_sta(struct ieee80211_local *local,
+				 struct ieee80211_sub_if_data *sdata,
+				 struct sta_info *sta)
+{
+	might_sleep();
+
+	if (sdata && !check_sdata_in_driver(sdata))
+		return;
+
+	trace_drv_flush_sta(local, sdata, &sta->sta);
+	if (local->ops->flush_sta)
+		local->ops->flush_sta(&local->hw, &sdata->vif, &sta->sta);
+	trace_drv_return_void(local);
+}
+
 static inline void drv_channel_switch(struct ieee80211_local *local,
 				      struct ieee80211_sub_if_data *sdata,
 				      struct ieee80211_channel_switch *ch_switch)
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index ce7c3b997269..1400512e0dde 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -1299,8 +1299,12 @@ static void __sta_info_destroy_part2(struct sta_info *sta)
 	 * frames sitting on hardware queues might be sent out without
 	 * any encryption at all.
 	 */
-	if (local->ops->set_key)
-		ieee80211_flush_queues(local, sta->sdata, false);
+	if (local->ops->set_key) {
+		if (local->ops->flush_sta)
+			drv_flush_sta(local, sta->sdata, sta);
+		else
+			ieee80211_flush_queues(local, sta->sdata, false);
+	}
 
 	/* now keys can no longer be reached */
 	ieee80211_free_sta_keys(local, sta);
diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h
index e0ccf5fe708a..de5d69f21306 100644
--- a/net/mac80211/trace.h
+++ b/net/mac80211/trace.h
@@ -1177,6 +1177,13 @@ TRACE_EVENT(drv_flush,
 	)
 );
 
+DEFINE_EVENT(sta_event, drv_flush_sta,
+	TP_PROTO(struct ieee80211_local *local,
+		 struct ieee80211_sub_if_data *sdata,
+		 struct ieee80211_sta *sta),
+	TP_ARGS(local, sdata, sta)
+);
+
 TRACE_EVENT(drv_channel_switch,
 	TP_PROTO(struct ieee80211_local *local,
 		 struct ieee80211_sub_if_data *sdata,
-- 
cgit v1.2.3-70-g09d2


From 8c5c2a4898e3d6bad86e29d471e023c8a19ba799 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Thu, 13 Apr 2023 20:28:42 +0200
Subject: bpf, sockmap: Revert buggy deadlock fix in the sockhash and sockmap

syzbot reported a splat and bisected it to recent commit ed17aa92dc56 ("bpf,
sockmap: fix deadlocks in the sockhash and sockmap"):

  [...]
  WARNING: CPU: 1 PID: 9280 at kernel/softirq.c:376 __local_bh_enable_ip+0xbe/0x130 kernel/softirq.c:376
  Modules linked in:
  CPU: 1 PID: 9280 Comm: syz-executor.1 Not tainted 6.2.0-syzkaller-13249-gd319f344561d #0
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/30/2023
  RIP: 0010:__local_bh_enable_ip+0xbe/0x130 kernel/softirq.c:376
  [...]
  Call Trace:
  <TASK>
  spin_unlock_bh include/linux/spinlock.h:395 [inline]
  sock_map_del_link+0x2ea/0x510 net/core/sock_map.c:165
  sock_map_unref+0xb0/0x1d0 net/core/sock_map.c:184
  sock_hash_delete_elem+0x1ec/0x2a0 net/core/sock_map.c:945
  map_delete_elem kernel/bpf/syscall.c:1536 [inline]
  __sys_bpf+0x2edc/0x53e0 kernel/bpf/syscall.c:5053
  __do_sys_bpf kernel/bpf/syscall.c:5166 [inline]
  __se_sys_bpf kernel/bpf/syscall.c:5164 [inline]
  __x64_sys_bpf+0x79/0xc0 kernel/bpf/syscall.c:5164
  do_syscall_x64 arch/x86/entry/common.c:50 [inline]
  do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
  entry_SYSCALL_64_after_hwframe+0x63/0xcd
  RIP: 0033:0x7fe8f7c8c169
  </TASK>
  [...]

Revert for now until we have a proper solution.

Fixes: ed17aa92dc56 ("bpf, sockmap: fix deadlocks in the sockhash and sockmap")
Reported-by: syzbot+49f6cef45247ff249498@syzkaller.appspotmail.com
Cc: Hsin-Wei Hung <hsinweih@uci.edu>
Cc: Xin Liu <liuxin350@huawei.com>
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/000000000000f1db9605f939720e@google.com/
---
 net/core/sock_map.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index 66305b7bf8b7..7c189c2e2fbf 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -414,9 +414,8 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test,
 {
 	struct sock *sk;
 	int err = 0;
-	unsigned long flags;
 
-	raw_spin_lock_irqsave(&stab->lock, flags);
+	raw_spin_lock_bh(&stab->lock);
 	sk = *psk;
 	if (!sk_test || sk_test == sk)
 		sk = xchg(psk, NULL);
@@ -426,7 +425,7 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test,
 	else
 		err = -EINVAL;
 
-	raw_spin_unlock_irqrestore(&stab->lock, flags);
+	raw_spin_unlock_bh(&stab->lock);
 	return err;
 }
 
@@ -933,12 +932,11 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key)
 	struct bpf_shtab_bucket *bucket;
 	struct bpf_shtab_elem *elem;
 	int ret = -ENOENT;
-	unsigned long flags;
 
 	hash = sock_hash_bucket_hash(key, key_size);
 	bucket = sock_hash_select_bucket(htab, hash);
 
-	raw_spin_lock_irqsave(&bucket->lock, flags);
+	raw_spin_lock_bh(&bucket->lock);
 	elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size);
 	if (elem) {
 		hlist_del_rcu(&elem->node);
@@ -946,7 +944,7 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key)
 		sock_hash_free_elem(htab, elem);
 		ret = 0;
 	}
-	raw_spin_unlock_irqrestore(&bucket->lock, flags);
+	raw_spin_unlock_bh(&bucket->lock);
 	return ret;
 }
 
-- 
cgit v1.2.3-70-g09d2


From c04135ab351badb1daf5868328581075ead691e0 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 13 Apr 2023 17:22:48 -0700
Subject: selftests/bpf: Fix merge conflict due to SYS() macro change.

Fix merge conflict between bpf/bpf-next trees due to change of arguments in SYS() macro.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
index c94eb63b7b77..498d3bdaa4b0 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
@@ -175,8 +175,8 @@ void test_xdp_do_redirect(void)
 		goto out;
 
 	/* Enable GRO */
-	SYS("ethtool -K veth_src gro on");
-	SYS("ethtool -K veth_dst gro on");
+	SYS(out, "ethtool -K veth_src gro on");
+	SYS(out, "ethtool -K veth_dst gro on");
 
 	err = bpf_xdp_query(ifindex_src, XDP_FLAGS_DRV_MODE, &query_opts);
 	if (!ASSERT_OK(err, "veth_src bpf_xdp_query gro on"))
-- 
cgit v1.2.3-70-g09d2


From 75860b52019cf93662c31596254599bf5f8df900 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 13 Apr 2023 19:54:17 -0700
Subject: selftests/bpf: Workaround for older vm_sockets.h.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some distros ship with older vm_sockets.h that doesn't have VMADDR_CID_LOCAL
which causes selftests build to fail:
/tmp/work/bpf/bpf/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c:261:18: error: ‘VMADDR_CID_LOCAL’ undeclared (first use in this function); did you mean ‘VMADDR_CID_HOST’?
    261 |  addr->svm_cid = VMADDR_CID_LOCAL;
        |                  ^~~~~~~~~~~~~~~~
        |                  VMADDR_CID_HOST

Workaround this issue by defining it on demand.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/sockmap_listen.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c b/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
index 8f09e1ea3ba7..141c1e5944ee 100644
--- a/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
+++ b/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
@@ -20,6 +20,11 @@
 #include <unistd.h>
 #include <linux/vm_sockets.h>
 
+/* workaround for older vm_sockets.h */
+#ifndef VMADDR_CID_LOCAL
+#define VMADDR_CID_LOCAL 1
+#endif
+
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
-- 
cgit v1.2.3-70-g09d2


From c11bd046485d7bf1ca200db0e7d0bdc4bafdd395 Mon Sep 17 00:00:00 2001
From: Yafang <laoar.shao@gmail.com>
Date: Thu, 13 Apr 2023 02:52:48 +0000
Subject: bpf: Add preempt_count_{sub,add} into btf id deny list

The recursion check in __bpf_prog_enter* and __bpf_prog_exit*
leave preempt_count_{sub,add} unprotected. When attaching trampoline to
them we get panic as follows,

[  867.843050] BUG: TASK stack guard page was hit at 0000000009d325cf (stack is 0000000046a46a15..00000000537e7b28)
[  867.843064] stack guard page: 0000 [#1] PREEMPT SMP NOPTI
[  867.843067] CPU: 8 PID: 11009 Comm: trace Kdump: loaded Not tainted 6.2.0+ #4
[  867.843100] Call Trace:
[  867.843101]  <TASK>
[  867.843104]  asm_exc_int3+0x3a/0x40
[  867.843108] RIP: 0010:preempt_count_sub+0x1/0xa0
[  867.843135]  __bpf_prog_enter_recur+0x17/0x90
[  867.843148]  bpf_trampoline_6442468108_0+0x2e/0x1000
[  867.843154]  ? preempt_count_sub+0x1/0xa0
[  867.843157]  preempt_count_sub+0x5/0xa0
[  867.843159]  ? migrate_enable+0xac/0xf0
[  867.843164]  __bpf_prog_exit_recur+0x2d/0x40
[  867.843168]  bpf_trampoline_6442468108_0+0x55/0x1000
...
[  867.843788]  preempt_count_sub+0x5/0xa0
[  867.843793]  ? migrate_enable+0xac/0xf0
[  867.843829]  __bpf_prog_exit_recur+0x2d/0x40
[  867.843837] BUG: IRQ stack guard page was hit at 0000000099bd8228 (stack is 00000000b23e2bc4..000000006d95af35)
[  867.843841] BUG: IRQ stack guard page was hit at 000000005ae07924 (stack is 00000000ffd69623..0000000014eb594c)
[  867.843843] BUG: IRQ stack guard page was hit at 00000000028320f0 (stack is 00000000034b6438..0000000078d1bcec)
[  867.843842]  bpf_trampoline_6442468108_0+0x55/0x1000
...

That is because in __bpf_prog_exit_recur, the preempt_count_{sub,add} are
called after prog->active is decreased.

Fixing this by adding these two functions into btf ids deny list.

Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Yafang <laoar.shao@gmail.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Jiri Olsa <olsajiri@gmail.com>
Acked-by: Hao Luo <haoluo@google.com>
Link: https://lore.kernel.org/r/20230413025248.79764-1-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d6db6de3e9ea..e3d5a7a2f428 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18582,6 +18582,10 @@ BTF_ID(func, migrate_enable)
 #if !defined CONFIG_PREEMPT_RCU && !defined CONFIG_TINY_RCU
 BTF_ID(func, rcu_read_unlock_strict)
 #endif
+#if defined(CONFIG_DEBUG_PREEMPT) || defined(CONFIG_TRACE_PREEMPT_TOGGLE)
+BTF_ID(func, preempt_count_add)
+BTF_ID(func, preempt_count_sub)
+#endif
 BTF_SET_END(btf_id_deny)
 
 static bool can_be_sleepable(struct bpf_prog *prog)
-- 
cgit v1.2.3-70-g09d2


From 1cf3bfc60f9836f44da951f58b6ae24680484b35 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Thu, 13 Apr 2023 01:06:32 +0200
Subject: bpf: Support 64-bit pointers to kfuncs

test_ksyms_module fails to emit a kfunc call targeting a module on
s390x, because the verifier stores the difference between kfunc
address and __bpf_call_base in bpf_insn.imm, which is s32, and modules
are roughly (1 << 42) bytes away from the kernel on s390x.

Fix by keeping BTF id in bpf_insn.imm for BPF_PSEUDO_KFUNC_CALLs,
and storing the absolute address in bpf_kfunc_desc.

Introduce bpf_jit_supports_far_kfunc_call() in order to limit this new
behavior to the s390x JIT. Otherwise other JITs need to be modified,
which is not desired.

Introduce bpf_get_kfunc_addr() instead of exposing both
find_kfunc_desc() and struct bpf_kfunc_desc.

In addition to sorting kfuncs by imm, also sort them by offset, in
order to handle conflicting imms from different modules. Do this on
all architectures in order to simplify code.

Factor out resolving specialized kfuncs (XPD and dynptr) from
fixup_kfunc_call(). This was required in the first place, because
fixup_kfunc_call() uses find_kfunc_desc(), which returns a const
pointer, so it's not possible to modify kfunc addr without stripping
const, which is not nice. It also removes repetition of code like:

	if (bpf_jit_supports_far_kfunc_call())
		desc->addr = func;
	else
		insn->imm = BPF_CALL_IMM(func);

and separates kfunc_desc_tab fixups from kfunc_call fixups.

Suggested-by: Jiri Olsa <olsajiri@gmail.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20230412230632.885985-1-iii@linux.ibm.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 arch/s390/net/bpf_jit_comp.c |   5 ++
 include/linux/bpf.h          |  10 ++++
 include/linux/filter.h       |   1 +
 kernel/bpf/core.c            |  11 ++++
 kernel/bpf/verifier.c        | 123 +++++++++++++++++++++++++++++--------------
 5 files changed, 110 insertions(+), 40 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index d0846ba818ee..7102e4b674a0 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -2001,6 +2001,11 @@ bool bpf_jit_supports_kfunc_call(void)
 	return true;
 }
 
+bool bpf_jit_supports_far_kfunc_call(void)
+{
+	return true;
+}
+
 int bpf_arch_text_poke(void *ip, enum bpf_text_poke_type t,
 		       void *old_addr, void *new_addr)
 {
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 2c6095bd7d69..88845aadc47d 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2295,6 +2295,9 @@ bool bpf_prog_has_kfunc_call(const struct bpf_prog *prog);
 const struct btf_func_model *
 bpf_jit_find_kfunc_model(const struct bpf_prog *prog,
 			 const struct bpf_insn *insn);
+int bpf_get_kfunc_addr(const struct bpf_prog *prog, u32 func_id,
+		       u16 btf_fd_idx, u8 **func_addr);
+
 struct bpf_core_ctx {
 	struct bpf_verifier_log *log;
 	const struct btf *btf;
@@ -2545,6 +2548,13 @@ bpf_jit_find_kfunc_model(const struct bpf_prog *prog,
 	return NULL;
 }
 
+static inline int
+bpf_get_kfunc_addr(const struct bpf_prog *prog, u32 func_id,
+		   u16 btf_fd_idx, u8 **func_addr)
+{
+	return -ENOTSUPP;
+}
+
 static inline bool unprivileged_ebpf_enabled(void)
 {
 	return false;
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 5364b0c52c1d..bbce89937fde 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -920,6 +920,7 @@ void bpf_jit_compile(struct bpf_prog *prog);
 bool bpf_jit_needs_zext(void);
 bool bpf_jit_supports_subprog_tailcalls(void);
 bool bpf_jit_supports_kfunc_call(void);
+bool bpf_jit_supports_far_kfunc_call(void);
 bool bpf_helper_changes_pkt_data(void *func);
 
 static inline bool bpf_dump_raw_ok(const struct cred *cred)
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index e2d256c82072..7421487422d4 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1187,6 +1187,7 @@ int bpf_jit_get_func_addr(const struct bpf_prog *prog,
 	s16 off = insn->off;
 	s32 imm = insn->imm;
 	u8 *addr;
+	int err;
 
 	*func_addr_fixed = insn->src_reg != BPF_PSEUDO_CALL;
 	if (!*func_addr_fixed) {
@@ -1201,6 +1202,11 @@ int bpf_jit_get_func_addr(const struct bpf_prog *prog,
 			addr = (u8 *)prog->aux->func[off]->bpf_func;
 		else
 			return -EINVAL;
+	} else if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL &&
+		   bpf_jit_supports_far_kfunc_call()) {
+		err = bpf_get_kfunc_addr(prog, insn->imm, insn->off, &addr);
+		if (err)
+			return err;
 	} else {
 		/* Address of a BPF helper call. Since part of the core
 		 * kernel, it's always at a fixed location. __bpf_call_base
@@ -2732,6 +2738,11 @@ bool __weak bpf_jit_supports_kfunc_call(void)
 	return false;
 }
 
+bool __weak bpf_jit_supports_far_kfunc_call(void)
+{
+	return false;
+}
+
 /* To execute LD_ABS/LD_IND instructions __bpf_prog_run() may call
  * skb_copy_bits(), so provide a weak definition of it for NET-less config.
  */
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index e3d5a7a2f428..4aa6d715e655 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -195,6 +195,8 @@ static void invalidate_non_owning_refs(struct bpf_verifier_env *env);
 static bool in_rbtree_lock_required_cb(struct bpf_verifier_env *env);
 static int ref_set_non_owning(struct bpf_verifier_env *env,
 			      struct bpf_reg_state *reg);
+static void specialize_kfunc(struct bpf_verifier_env *env,
+			     u32 func_id, u16 offset, unsigned long *addr);
 
 static bool bpf_map_ptr_poisoned(const struct bpf_insn_aux_data *aux)
 {
@@ -2374,6 +2376,7 @@ struct bpf_kfunc_desc {
 	u32 func_id;
 	s32 imm;
 	u16 offset;
+	unsigned long addr;
 };
 
 struct bpf_kfunc_btf {
@@ -2383,6 +2386,11 @@ struct bpf_kfunc_btf {
 };
 
 struct bpf_kfunc_desc_tab {
+	/* Sorted by func_id (BTF ID) and offset (fd_array offset) during
+	 * verification. JITs do lookups by bpf_insn, where func_id may not be
+	 * available, therefore at the end of verification do_misc_fixups()
+	 * sorts this by imm and offset.
+	 */
 	struct bpf_kfunc_desc descs[MAX_KFUNC_DESCS];
 	u32 nr_descs;
 };
@@ -2423,6 +2431,19 @@ find_kfunc_desc(const struct bpf_prog *prog, u32 func_id, u16 offset)
 		       sizeof(tab->descs[0]), kfunc_desc_cmp_by_id_off);
 }
 
+int bpf_get_kfunc_addr(const struct bpf_prog *prog, u32 func_id,
+		       u16 btf_fd_idx, u8 **func_addr)
+{
+	const struct bpf_kfunc_desc *desc;
+
+	desc = find_kfunc_desc(prog, func_id, btf_fd_idx);
+	if (!desc)
+		return -EFAULT;
+
+	*func_addr = (u8 *)desc->addr;
+	return 0;
+}
+
 static struct btf *__find_kfunc_desc_btf(struct bpf_verifier_env *env,
 					 s16 offset)
 {
@@ -2602,13 +2623,18 @@ static int add_kfunc_call(struct bpf_verifier_env *env, u32 func_id, s16 offset)
 			func_name);
 		return -EINVAL;
 	}
+	specialize_kfunc(env, func_id, offset, &addr);
 
-	call_imm = BPF_CALL_IMM(addr);
-	/* Check whether or not the relative offset overflows desc->imm */
-	if ((unsigned long)(s32)call_imm != call_imm) {
-		verbose(env, "address of kernel function %s is out of range\n",
-			func_name);
-		return -EINVAL;
+	if (bpf_jit_supports_far_kfunc_call()) {
+		call_imm = func_id;
+	} else {
+		call_imm = BPF_CALL_IMM(addr);
+		/* Check whether the relative offset overflows desc->imm */
+		if ((unsigned long)(s32)call_imm != call_imm) {
+			verbose(env, "address of kernel function %s is out of range\n",
+				func_name);
+			return -EINVAL;
+		}
 	}
 
 	if (bpf_dev_bound_kfunc_id(func_id)) {
@@ -2621,6 +2647,7 @@ static int add_kfunc_call(struct bpf_verifier_env *env, u32 func_id, s16 offset)
 	desc->func_id = func_id;
 	desc->imm = call_imm;
 	desc->offset = offset;
+	desc->addr = addr;
 	err = btf_distill_func_proto(&env->log, desc_btf,
 				     func_proto, func_name,
 				     &desc->func_model);
@@ -2630,19 +2657,19 @@ static int add_kfunc_call(struct bpf_verifier_env *env, u32 func_id, s16 offset)
 	return err;
 }
 
-static int kfunc_desc_cmp_by_imm(const void *a, const void *b)
+static int kfunc_desc_cmp_by_imm_off(const void *a, const void *b)
 {
 	const struct bpf_kfunc_desc *d0 = a;
 	const struct bpf_kfunc_desc *d1 = b;
 
-	if (d0->imm > d1->imm)
-		return 1;
-	else if (d0->imm < d1->imm)
-		return -1;
+	if (d0->imm != d1->imm)
+		return d0->imm < d1->imm ? -1 : 1;
+	if (d0->offset != d1->offset)
+		return d0->offset < d1->offset ? -1 : 1;
 	return 0;
 }
 
-static void sort_kfunc_descs_by_imm(struct bpf_prog *prog)
+static void sort_kfunc_descs_by_imm_off(struct bpf_prog *prog)
 {
 	struct bpf_kfunc_desc_tab *tab;
 
@@ -2651,7 +2678,7 @@ static void sort_kfunc_descs_by_imm(struct bpf_prog *prog)
 		return;
 
 	sort(tab->descs, tab->nr_descs, sizeof(tab->descs[0]),
-	     kfunc_desc_cmp_by_imm, NULL);
+	     kfunc_desc_cmp_by_imm_off, NULL);
 }
 
 bool bpf_prog_has_kfunc_call(const struct bpf_prog *prog)
@@ -2665,13 +2692,14 @@ bpf_jit_find_kfunc_model(const struct bpf_prog *prog,
 {
 	const struct bpf_kfunc_desc desc = {
 		.imm = insn->imm,
+		.offset = insn->off,
 	};
 	const struct bpf_kfunc_desc *res;
 	struct bpf_kfunc_desc_tab *tab;
 
 	tab = prog->aux->kfunc_tab;
 	res = bsearch(&desc, tab->descs, tab->nr_descs,
-		      sizeof(tab->descs[0]), kfunc_desc_cmp_by_imm);
+		      sizeof(tab->descs[0]), kfunc_desc_cmp_by_imm_off);
 
 	return res ? &res->func_model : NULL;
 }
@@ -17293,11 +17321,45 @@ static int fixup_call_args(struct bpf_verifier_env *env)
 	return err;
 }
 
+/* replace a generic kfunc with a specialized version if necessary */
+static void specialize_kfunc(struct bpf_verifier_env *env,
+			     u32 func_id, u16 offset, unsigned long *addr)
+{
+	struct bpf_prog *prog = env->prog;
+	bool seen_direct_write;
+	void *xdp_kfunc;
+	bool is_rdonly;
+
+	if (bpf_dev_bound_kfunc_id(func_id)) {
+		xdp_kfunc = bpf_dev_bound_resolve_kfunc(prog, func_id);
+		if (xdp_kfunc) {
+			*addr = (unsigned long)xdp_kfunc;
+			return;
+		}
+		/* fallback to default kfunc when not supported by netdev */
+	}
+
+	if (offset)
+		return;
+
+	if (func_id == special_kfunc_list[KF_bpf_dynptr_from_skb]) {
+		seen_direct_write = env->seen_direct_write;
+		is_rdonly = !may_access_direct_pkt_data(env, NULL, BPF_WRITE);
+
+		if (is_rdonly)
+			*addr = (unsigned long)bpf_dynptr_from_skb_rdonly;
+
+		/* restore env->seen_direct_write to its original value, since
+		 * may_access_direct_pkt_data mutates it
+		 */
+		env->seen_direct_write = seen_direct_write;
+	}
+}
+
 static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			    struct bpf_insn *insn_buf, int insn_idx, int *cnt)
 {
 	const struct bpf_kfunc_desc *desc;
-	void *xdp_kfunc;
 
 	if (!insn->imm) {
 		verbose(env, "invalid kernel function call not eliminated in verifier pass\n");
@@ -17306,18 +17368,9 @@ static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 
 	*cnt = 0;
 
-	if (bpf_dev_bound_kfunc_id(insn->imm)) {
-		xdp_kfunc = bpf_dev_bound_resolve_kfunc(env->prog, insn->imm);
-		if (xdp_kfunc) {
-			insn->imm = BPF_CALL_IMM(xdp_kfunc);
-			return 0;
-		}
-
-		/* fallback to default kfunc when not supported by netdev */
-	}
-
-	/* insn->imm has the btf func_id. Replace it with
-	 * an address (relative to __bpf_call_base).
+	/* insn->imm has the btf func_id. Replace it with an offset relative to
+	 * __bpf_call_base, unless the JIT needs to call functions that are
+	 * further than 32 bits away (bpf_jit_supports_far_kfunc_call()).
 	 */
 	desc = find_kfunc_desc(env->prog, insn->imm, insn->off);
 	if (!desc) {
@@ -17326,7 +17379,8 @@ static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		return -EFAULT;
 	}
 
-	insn->imm = desc->imm;
+	if (!bpf_jit_supports_far_kfunc_call())
+		insn->imm = BPF_CALL_IMM(desc->addr);
 	if (insn->off)
 		return 0;
 	if (desc->func_id == special_kfunc_list[KF_bpf_obj_new_impl]) {
@@ -17351,17 +17405,6 @@ static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		   desc->func_id == special_kfunc_list[KF_bpf_rdonly_cast]) {
 		insn_buf[0] = BPF_MOV64_REG(BPF_REG_0, BPF_REG_1);
 		*cnt = 1;
-	} else if (desc->func_id == special_kfunc_list[KF_bpf_dynptr_from_skb]) {
-		bool seen_direct_write = env->seen_direct_write;
-		bool is_rdonly = !may_access_direct_pkt_data(env, NULL, BPF_WRITE);
-
-		if (is_rdonly)
-			insn->imm = BPF_CALL_IMM(bpf_dynptr_from_skb_rdonly);
-
-		/* restore env->seen_direct_write to its original value, since
-		 * may_access_direct_pkt_data mutates it
-		 */
-		env->seen_direct_write = seen_direct_write;
 	}
 	return 0;
 }
@@ -17891,7 +17934,7 @@ patch_call_imm:
 		}
 	}
 
-	sort_kfunc_descs_by_imm(env->prog);
+	sort_kfunc_descs_by_imm_off(env->prog);
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 9ecd05794b8da1f6cfca4c3721a3b0fed2e21a82 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:30 +0300
Subject: net: mscc: ocelot: strengthen type of "u32 reg" in I/O accessors

The "u32 reg" argument that is passed to these functions is not a plain
address, but rather a driver-specific encoding of another enum
ocelot_target target in the upper bits, and an index into the
u32 ocelot->map[target][] array in the lower bits. That encoded value
takes the type "enum ocelot_reg" and is what is passed to these I/O
functions, so let's actually use that to prevent type confusion.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_io.c | 20 +++++++++++---------
 include/soc/mscc/ocelot.h             | 20 +++++++++++---------
 2 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_io.c b/drivers/net/ethernet/mscc/ocelot_io.c
index 2067382d0ee1..ddb96f32830d 100644
--- a/drivers/net/ethernet/mscc/ocelot_io.c
+++ b/drivers/net/ethernet/mscc/ocelot_io.c
@@ -10,8 +10,8 @@
 
 #include "ocelot.h"
 
-int __ocelot_bulk_read_ix(struct ocelot *ocelot, u32 reg, u32 offset, void *buf,
-			  int count)
+int __ocelot_bulk_read_ix(struct ocelot *ocelot, enum ocelot_reg reg,
+			  u32 offset, void *buf, int count)
 {
 	u16 target = reg >> TARGET_OFFSET;
 
@@ -23,7 +23,7 @@ int __ocelot_bulk_read_ix(struct ocelot *ocelot, u32 reg, u32 offset, void *buf,
 }
 EXPORT_SYMBOL_GPL(__ocelot_bulk_read_ix);
 
-u32 __ocelot_read_ix(struct ocelot *ocelot, u32 reg, u32 offset)
+u32 __ocelot_read_ix(struct ocelot *ocelot, enum ocelot_reg reg, u32 offset)
 {
 	u16 target = reg >> TARGET_OFFSET;
 	u32 val;
@@ -36,7 +36,8 @@ u32 __ocelot_read_ix(struct ocelot *ocelot, u32 reg, u32 offset)
 }
 EXPORT_SYMBOL_GPL(__ocelot_read_ix);
 
-void __ocelot_write_ix(struct ocelot *ocelot, u32 val, u32 reg, u32 offset)
+void __ocelot_write_ix(struct ocelot *ocelot, u32 val, enum ocelot_reg reg,
+		       u32 offset)
 {
 	u16 target = reg >> TARGET_OFFSET;
 
@@ -47,8 +48,8 @@ void __ocelot_write_ix(struct ocelot *ocelot, u32 val, u32 reg, u32 offset)
 }
 EXPORT_SYMBOL_GPL(__ocelot_write_ix);
 
-void __ocelot_rmw_ix(struct ocelot *ocelot, u32 val, u32 mask, u32 reg,
-		     u32 offset)
+void __ocelot_rmw_ix(struct ocelot *ocelot, u32 val, u32 mask,
+		     enum ocelot_reg reg, u32 offset)
 {
 	u16 target = reg >> TARGET_OFFSET;
 
@@ -60,7 +61,7 @@ void __ocelot_rmw_ix(struct ocelot *ocelot, u32 val, u32 mask, u32 reg,
 }
 EXPORT_SYMBOL_GPL(__ocelot_rmw_ix);
 
-u32 ocelot_port_readl(struct ocelot_port *port, u32 reg)
+u32 ocelot_port_readl(struct ocelot_port *port, enum ocelot_reg reg)
 {
 	struct ocelot *ocelot = port->ocelot;
 	u16 target = reg >> TARGET_OFFSET;
@@ -73,7 +74,7 @@ u32 ocelot_port_readl(struct ocelot_port *port, u32 reg)
 }
 EXPORT_SYMBOL_GPL(ocelot_port_readl);
 
-void ocelot_port_writel(struct ocelot_port *port, u32 val, u32 reg)
+void ocelot_port_writel(struct ocelot_port *port, u32 val, enum ocelot_reg reg)
 {
 	struct ocelot *ocelot = port->ocelot;
 	u16 target = reg >> TARGET_OFFSET;
@@ -84,7 +85,8 @@ void ocelot_port_writel(struct ocelot_port *port, u32 val, u32 reg)
 }
 EXPORT_SYMBOL_GPL(ocelot_port_writel);
 
-void ocelot_port_rmwl(struct ocelot_port *port, u32 val, u32 mask, u32 reg)
+void ocelot_port_rmwl(struct ocelot_port *port, u32 val, u32 mask,
+		      enum ocelot_reg reg)
 {
 	u32 cur = ocelot_port_readl(port, reg);
 
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index d757b5e26d26..277e6d1f2096 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -940,15 +940,17 @@ struct ocelot_policer {
 	__ocelot_target_write_ix(ocelot, target, val, reg, 0)
 
 /* I/O */
-u32 ocelot_port_readl(struct ocelot_port *port, u32 reg);
-void ocelot_port_writel(struct ocelot_port *port, u32 val, u32 reg);
-void ocelot_port_rmwl(struct ocelot_port *port, u32 val, u32 mask, u32 reg);
-int __ocelot_bulk_read_ix(struct ocelot *ocelot, u32 reg, u32 offset, void *buf,
-			  int count);
-u32 __ocelot_read_ix(struct ocelot *ocelot, u32 reg, u32 offset);
-void __ocelot_write_ix(struct ocelot *ocelot, u32 val, u32 reg, u32 offset);
-void __ocelot_rmw_ix(struct ocelot *ocelot, u32 val, u32 mask, u32 reg,
-		     u32 offset);
+u32 ocelot_port_readl(struct ocelot_port *port, enum ocelot_reg reg);
+void ocelot_port_writel(struct ocelot_port *port, u32 val, enum ocelot_reg reg);
+void ocelot_port_rmwl(struct ocelot_port *port, u32 val, u32 mask,
+		      enum ocelot_reg reg);
+int __ocelot_bulk_read_ix(struct ocelot *ocelot, enum ocelot_reg reg,
+			  u32 offset, void *buf, int count);
+u32 __ocelot_read_ix(struct ocelot *ocelot, enum ocelot_reg reg, u32 offset);
+void __ocelot_write_ix(struct ocelot *ocelot, u32 val, enum ocelot_reg reg,
+		       u32 offset);
+void __ocelot_rmw_ix(struct ocelot *ocelot, u32 val, u32 mask,
+		     enum ocelot_reg reg, u32 offset);
 u32 __ocelot_target_read_ix(struct ocelot *ocelot, enum ocelot_target target,
 			    u32 reg, u32 offset);
 void __ocelot_target_write_ix(struct ocelot *ocelot, enum ocelot_target target,
-- 
cgit v1.2.3-70-g09d2


From 40cd07cb42617f0e8b8597d0db288a5ce66621d9 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:31 +0300
Subject: net: mscc: ocelot: refactor enum ocelot_reg decoding to helper

ocelot_io.c duplicates the decoding of an enum ocelot_reg (which holds
an enum ocelot_target in the upper bits and an index into a regmap array
in the lower bits) 4 times.

We'd like to reuse that logic once more, from ocelot.c. In order to do
that, let's consolidate the existing 4 instances into a header
accessible both by ocelot.c as well as by ocelot_io.c.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot.h    |  9 +++++++++
 drivers/net/ethernet/mscc/ocelot_io.c | 30 ++++++++++++++++--------------
 2 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h
index e9a0179448bf..d920ca930690 100644
--- a/drivers/net/ethernet/mscc/ocelot.h
+++ b/drivers/net/ethernet/mscc/ocelot.h
@@ -74,6 +74,15 @@ struct ocelot_multicast {
 	struct ocelot_pgid *pgid;
 };
 
+static inline void ocelot_reg_to_target_addr(struct ocelot *ocelot,
+					     enum ocelot_reg reg,
+					     enum ocelot_target *target,
+					     u32 *addr)
+{
+	*target = reg >> TARGET_OFFSET;
+	*addr = ocelot->map[*target][reg & REG_MASK];
+}
+
 int ocelot_bridge_num_find(struct ocelot *ocelot,
 			   const struct net_device *bridge);
 
diff --git a/drivers/net/ethernet/mscc/ocelot_io.c b/drivers/net/ethernet/mscc/ocelot_io.c
index ddb96f32830d..3aa7dc29ebe1 100644
--- a/drivers/net/ethernet/mscc/ocelot_io.c
+++ b/drivers/net/ethernet/mscc/ocelot_io.c
@@ -13,25 +13,26 @@
 int __ocelot_bulk_read_ix(struct ocelot *ocelot, enum ocelot_reg reg,
 			  u32 offset, void *buf, int count)
 {
-	u16 target = reg >> TARGET_OFFSET;
+	enum ocelot_target target;
+	u32 addr;
 
+	ocelot_reg_to_target_addr(ocelot, reg, &target, &addr);
 	WARN_ON(!target);
 
-	return regmap_bulk_read(ocelot->targets[target],
-				ocelot->map[target][reg & REG_MASK] + offset,
+	return regmap_bulk_read(ocelot->targets[target], addr + offset,
 				buf, count);
 }
 EXPORT_SYMBOL_GPL(__ocelot_bulk_read_ix);
 
 u32 __ocelot_read_ix(struct ocelot *ocelot, enum ocelot_reg reg, u32 offset)
 {
-	u16 target = reg >> TARGET_OFFSET;
-	u32 val;
+	enum ocelot_target target;
+	u32 addr, val;
 
+	ocelot_reg_to_target_addr(ocelot, reg, &target, &addr);
 	WARN_ON(!target);
 
-	regmap_read(ocelot->targets[target],
-		    ocelot->map[target][reg & REG_MASK] + offset, &val);
+	regmap_read(ocelot->targets[target], addr + offset, &val);
 	return val;
 }
 EXPORT_SYMBOL_GPL(__ocelot_read_ix);
@@ -39,25 +40,26 @@ EXPORT_SYMBOL_GPL(__ocelot_read_ix);
 void __ocelot_write_ix(struct ocelot *ocelot, u32 val, enum ocelot_reg reg,
 		       u32 offset)
 {
-	u16 target = reg >> TARGET_OFFSET;
+	enum ocelot_target target;
+	u32 addr;
 
+	ocelot_reg_to_target_addr(ocelot, reg, &target, &addr);
 	WARN_ON(!target);
 
-	regmap_write(ocelot->targets[target],
-		     ocelot->map[target][reg & REG_MASK] + offset, val);
+	regmap_write(ocelot->targets[target], addr + offset, val);
 }
 EXPORT_SYMBOL_GPL(__ocelot_write_ix);
 
 void __ocelot_rmw_ix(struct ocelot *ocelot, u32 val, u32 mask,
 		     enum ocelot_reg reg, u32 offset)
 {
-	u16 target = reg >> TARGET_OFFSET;
+	enum ocelot_target target;
+	u32 addr;
 
+	ocelot_reg_to_target_addr(ocelot, reg, &target, &addr);
 	WARN_ON(!target);
 
-	regmap_update_bits(ocelot->targets[target],
-			   ocelot->map[target][reg & REG_MASK] + offset,
-			   mask, val);
+	regmap_update_bits(ocelot->targets[target], addr + offset, mask, val);
 }
 EXPORT_SYMBOL_GPL(__ocelot_rmw_ix);
 
-- 
cgit v1.2.3-70-g09d2


From 07de32655bb4e4518cdb8ce4e2e0d6ed49f2ceb0 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:32 +0300
Subject: net: mscc: ocelot: debugging print for statistics regions

To make it easier to debug future issues with statistics counters not
getting aggregated properly into regions, like what happened in commit
6acc72a43eac ("net: mscc: ocelot: fix stats region batching"), add some
dev_dbg() prints which show the regions that were dynamically
determined.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_stats.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/net/ethernet/mscc/ocelot_stats.c b/drivers/net/ethernet/mscc/ocelot_stats.c
index d0e6cd8dbe5c..b50d9d9f8023 100644
--- a/drivers/net/ethernet/mscc/ocelot_stats.c
+++ b/drivers/net/ethernet/mscc/ocelot_stats.c
@@ -925,6 +925,15 @@ static int ocelot_prepare_stats_regions(struct ocelot *ocelot)
 	}
 
 	list_for_each_entry(region, &ocelot->stats_regions, node) {
+		enum ocelot_target target;
+		u32 addr;
+
+		ocelot_reg_to_target_addr(ocelot, region->base, &target,
+					  &addr);
+
+		dev_dbg(ocelot->dev,
+			"region of %d contiguous counters starting with SYS:STAT:CNT[0x%03x]\n",
+			region->count, addr / 4);
 		region->buf = devm_kcalloc(ocelot->dev, region->count,
 					   sizeof(*region->buf), GFP_KERNEL);
 		if (!region->buf)
-- 
cgit v1.2.3-70-g09d2


From 93f0f93bbdb9ebf3d2831e8023cae4f31cb83cee Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:33 +0300
Subject: net: mscc: ocelot: remove blank line at the end of ocelot_stats.c

Commit a3bb8f521fd8 ("net: mscc: ocelot: remove unnecessary exposure of
stats structures") made an unnecessary change which was to add a new
line at the end of ocelot_stats.c. Remove it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_stats.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_stats.c b/drivers/net/ethernet/mscc/ocelot_stats.c
index b50d9d9f8023..99a14a942600 100644
--- a/drivers/net/ethernet/mscc/ocelot_stats.c
+++ b/drivers/net/ethernet/mscc/ocelot_stats.c
@@ -981,4 +981,3 @@ void ocelot_stats_deinit(struct ocelot *ocelot)
 	cancel_delayed_work(&ocelot->stats_work);
 	destroy_workqueue(ocelot->stats_queue);
 }
-
-- 
cgit v1.2.3-70-g09d2


From a9afc3e41c61688e26ba526a1f6c574608ae42d5 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:34 +0300
Subject: net: dsa: felix: remove confusing/incorrect comment from
 felix_setup()

That comment was written prior to knowing that what I was actually
seeing was a manifestation of the bug fixed in commit b4024c9e5c57
("felix: Fix initialization of ioremap resources").

There isn't any particular reason now why the hardware initialization is
done in felix_setup(), so just delete that comment to avoid spreading
misinformation.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/ocelot/felix.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 6dcebcfd71e7..80861ac090ae 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1550,11 +1550,6 @@ static int felix_connect_tag_protocol(struct dsa_switch *ds,
 	}
 }
 
-/* Hardware initialization done here so that we can allocate structures with
- * devm without fear of dsa_register_switch returning -EPROBE_DEFER and causing
- * us to allocate structures twice (leak memory) and map PCI memory twice
- * (which will not work).
- */
 static int felix_setup(struct dsa_switch *ds)
 {
 	struct ocelot *ocelot = ds->priv;
-- 
cgit v1.2.3-70-g09d2


From eae0b9d15ba6050852180449b932996bc3add48e Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:35 +0300
Subject: net: mscc: ocelot: strengthen type of "u32 reg" and "u32 base" in
 ocelot_stats.c

Use the specific enum ocelot_reg to make it clear that the region
registers are encoded and not plain addresses.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_stats.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_stats.c b/drivers/net/ethernet/mscc/ocelot_stats.c
index 99a14a942600..a381e326cb2b 100644
--- a/drivers/net/ethernet/mscc/ocelot_stats.c
+++ b/drivers/net/ethernet/mscc/ocelot_stats.c
@@ -145,7 +145,7 @@ enum ocelot_stat {
 };
 
 struct ocelot_stat_layout {
-	u32 reg;
+	enum ocelot_reg reg;
 	char name[ETH_GSTRING_LEN];
 };
 
@@ -257,7 +257,7 @@ struct ocelot_stat_layout {
 
 struct ocelot_stats_region {
 	struct list_head node;
-	u32 base;
+	enum ocelot_reg base;
 	enum ocelot_stat first_stat;
 	int count;
 	u32 *buf;
@@ -889,7 +889,7 @@ static int ocelot_prepare_stats_regions(struct ocelot *ocelot)
 {
 	struct ocelot_stats_region *region = NULL;
 	const struct ocelot_stat_layout *layout;
-	unsigned int last = 0;
+	enum ocelot_reg last = 0;
 	int i;
 
 	INIT_LIST_HEAD(&ocelot->stats_regions);
-- 
cgit v1.2.3-70-g09d2


From 6663c01eca1a250374a898f19b60ec01859a3d0b Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:36 +0300
Subject: net: mscc: ocelot: strengthen type of "int i" in ocelot_stats.c

The "int i" used to index the struct ocelot_stat_layout array actually
has a specific type: enum ocelot_stat. Use it, so that the WARN()
comment from ocelot_prepare_stats_regions() makes more sense.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_stats.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_stats.c b/drivers/net/ethernet/mscc/ocelot_stats.c
index a381e326cb2b..e82c9d9d0ad3 100644
--- a/drivers/net/ethernet/mscc/ocelot_stats.c
+++ b/drivers/net/ethernet/mscc/ocelot_stats.c
@@ -395,7 +395,7 @@ static void ocelot_check_stats_work(struct work_struct *work)
 void ocelot_get_strings(struct ocelot *ocelot, int port, u32 sset, u8 *data)
 {
 	const struct ocelot_stat_layout *layout;
-	int i;
+	enum ocelot_stat i;
 
 	if (sset != ETH_SS_STATS)
 		return;
@@ -442,7 +442,8 @@ out_unlock:
 int ocelot_get_sset_count(struct ocelot *ocelot, int port, int sset)
 {
 	const struct ocelot_stat_layout *layout;
-	int i, num_stats = 0;
+	enum ocelot_stat i;
+	int num_stats = 0;
 
 	if (sset != ETH_SS_STATS)
 		return -EOPNOTSUPP;
@@ -461,8 +462,8 @@ static void ocelot_port_ethtool_stats_cb(struct ocelot *ocelot, int port,
 					 void *priv)
 {
 	const struct ocelot_stat_layout *layout;
+	enum ocelot_stat i;
 	u64 *data = priv;
-	int i;
 
 	layout = ocelot_get_stats_layout(ocelot);
 
@@ -890,7 +891,7 @@ static int ocelot_prepare_stats_regions(struct ocelot *ocelot)
 	struct ocelot_stats_region *region = NULL;
 	const struct ocelot_stat_layout *layout;
 	enum ocelot_reg last = 0;
-	int i;
+	enum ocelot_stat i;
 
 	INIT_LIST_HEAD(&ocelot->stats_regions);
 
-- 
cgit v1.2.3-70-g09d2


From a291399e635415295860765954e247e83445e291 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Wed, 12 Apr 2023 15:47:37 +0300
Subject: net: mscc: ocelot: fix ineffective WARN_ON() in ocelot_stats.c

Since it is hopefully now clear that, since "last" and "layout[i].reg"
are enum types and not addresses, the existing WARN_ON() is ineffective
in checking that the _addresses_ are sorted in the proper order.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_stats.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_stats.c b/drivers/net/ethernet/mscc/ocelot_stats.c
index e82c9d9d0ad3..5c55197c7327 100644
--- a/drivers/net/ethernet/mscc/ocelot_stats.c
+++ b/drivers/net/ethernet/mscc/ocelot_stats.c
@@ -901,6 +901,17 @@ static int ocelot_prepare_stats_regions(struct ocelot *ocelot)
 		if (!layout[i].reg)
 			continue;
 
+		/* enum ocelot_stat must be kept sorted in the same order
+		 * as the addresses behind layout[i].reg in order to have
+		 * efficient bulking
+		 */
+		if (last) {
+			WARN(ocelot->map[SYS][last & REG_MASK] >= ocelot->map[SYS][layout[i].reg & REG_MASK],
+			     "reg 0x%x had address 0x%x but reg 0x%x has address 0x%x, bulking broken!",
+			     last, ocelot->map[SYS][last & REG_MASK],
+			     layout[i].reg, ocelot->map[SYS][layout[i].reg & REG_MASK]);
+		}
+
 		if (region && ocelot->map[SYS][layout[i].reg & REG_MASK] ==
 		    ocelot->map[SYS][last & REG_MASK] + 4) {
 			region->count++;
@@ -910,12 +921,6 @@ static int ocelot_prepare_stats_regions(struct ocelot *ocelot)
 			if (!region)
 				return -ENOMEM;
 
-			/* enum ocelot_stat must be kept sorted in the same
-			 * order as layout[i].reg in order to have efficient
-			 * bulking
-			 */
-			WARN_ON(last >= layout[i].reg);
-
 			region->base = layout[i].reg;
 			region->first_stat = i;
 			region->count = 1;
-- 
cgit v1.2.3-70-g09d2


From 4a1e885c6d143ff1b557ec7f3fc6ddf39c51502f Mon Sep 17 00:00:00 2001
From: Rong Tao <rongtao@cestc.cn>
Date: Wed, 12 Apr 2023 16:16:24 +0800
Subject: samples/bpf: sampleip: Replace PAGE_OFFSET with _text address

Macro PAGE_OFFSET(0xffff880000000000) in sampleip_user.c is inaccurate,
for example, in aarch64 architecture, this value depends on the
CONFIG_ARM64_VA_BITS compilation configuration, this value defaults to 48,
the corresponding PAGE_OFFSET is 0xffff800000000000, if we use the value
defined in sampleip_user.c, then all KSYMs obtained by sampleip are (user)

Symbol error due to PAGE_OFFSET error:

    $ sudo ./sampleip 1
    Sampling at 99 Hertz for 1 seconds. Ctrl-C also ends.
    ADDR                KSYM                             COUNT
    0xffff80000810ceb8  (user)                           1
    0xffffb28ec880      (user)                           1
    0xffff8000080c82b8  (user)                           1
    0xffffb23fed24      (user)                           1
    0xffffb28944fc      (user)                           1
    0xffff8000084628bc  (user)                           1
    0xffffb2a935c0      (user)                           1
    0xffff80000844677c  (user)                           1
    0xffff80000857a3a4  (user)                           1
    ...

A few examples of addresses in the CONFIG_ARM64_VA_BITS=48 environment in
the aarch64 environment:

    $ sudo head /proc/kallsyms
    ffff8000080a0000 T _text
    ffff8000080b0000 t gic_handle_irq
    ffff8000080b0000 T _stext
    ffff8000080b0000 T __irqentry_text_start
    ffff8000080b00b0 t gic_handle_irq
    ffff8000080b0230 t gic_handle_irq
    ffff8000080b03b4 T __irqentry_text_end
    ffff8000080b03b8 T __softirqentry_text_start
    ffff8000080b03c0 T __do_softirq
    ffff8000080b0718 T __entry_text_start

We just need to replace the PAGE_OFFSET with the address _text in
/proc/kallsyms to solve this problem:

    $ sudo ./sampleip 1
    Sampling at 99 Hertz for 1 seconds. Ctrl-C also ends.
    ADDR                KSYM                             COUNT
    0xffffb2892ab0      (user)                           1
    0xffffb2b1edfc      (user)                           1
    0xffff800008462834  __arm64_sys_ppoll                1
    0xffff8000084b87f4  eventfd_read                     1
    0xffffb28e6788      (user)                           1
    0xffff8000081e96d8  rcu_all_qs                       1
    0xffffb2ada878      (user)                           1
    ...

Signed-off-by: Rong Tao <rongtao@cestc.cn>
Link: https://lore.kernel.org/r/tencent_A0E82E0BEE925285F8156D540731DF805F05@qq.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 samples/bpf/sampleip_user.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/samples/bpf/sampleip_user.c b/samples/bpf/sampleip_user.c
index 921c505bb567..9283f47844fb 100644
--- a/samples/bpf/sampleip_user.c
+++ b/samples/bpf/sampleip_user.c
@@ -21,10 +21,10 @@
 #define DEFAULT_FREQ	99
 #define DEFAULT_SECS	5
 #define MAX_IPS		8192
-#define PAGE_OFFSET	0xffff880000000000
 
 static int map_fd;
 static int nr_cpus;
+static long _text_addr;
 
 static void usage(void)
 {
@@ -108,7 +108,7 @@ static void print_ip_map(int fd)
 	/* sort and print */
 	qsort(counts, max, sizeof(struct ipcount), count_cmp);
 	for (i = 0; i < max; i++) {
-		if (counts[i].ip > PAGE_OFFSET) {
+		if (counts[i].ip > _text_addr) {
 			sym = ksym_search(counts[i].ip);
 			if (!sym) {
 				printf("ksym not found. Is kallsyms loaded?\n");
@@ -169,6 +169,13 @@ int main(int argc, char **argv)
 		return 2;
 	}
 
+	/* used to determine whether the address is kernel space */
+	_text_addr = ksym_get_addr("_text");
+	if (!_text_addr) {
+		fprintf(stderr, "ERROR: no '_text' in /proc/kallsyms\n");
+		return 3;
+	}
+
 	/* create perf FDs for each CPU */
 	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
 	links = calloc(nr_cpus, sizeof(struct bpf_link *));
-- 
cgit v1.2.3-70-g09d2


From adee474a3b43dfab2d1f3b6a8ba7d052099f56a9 Mon Sep 17 00:00:00 2001
From: Harini Katakam <harini.katakam@amd.com>
Date: Tue, 11 Apr 2023 18:07:10 +0530
Subject: net: macb: Update gem PTP support check

There are currently two checks for PTP functionality - one on GEM
capability and another on the kernel config option. Combine them
into a single function as there's no use case where gem_has_ptp is
TRUE and MACB_USE_HWSTAMP is false.

Signed-off-by: Harini Katakam <harini.katakam@amd.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cadence/macb.h      | 2 +-
 drivers/net/ethernet/cadence/macb_main.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index c1fc91c97cee..07560803aa26 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -1363,7 +1363,7 @@ static inline bool macb_is_gem(struct macb *bp)
 
 static inline bool gem_has_ptp(struct macb *bp)
 {
-	return !!(bp->caps & MACB_CAPS_GEM_HAS_PTP);
+	return IS_ENABLED(CONFIG_MACB_USE_HWSTAMP) && (bp->caps & MACB_CAPS_GEM_HAS_PTP);
 }
 
 /**
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 541e4dda7950..df1a6756042e 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -3893,17 +3893,17 @@ static void macb_configure_caps(struct macb *bp,
 		dcfg = gem_readl(bp, DCFG2);
 		if ((dcfg & (GEM_BIT(RX_PKT_BUFF) | GEM_BIT(TX_PKT_BUFF))) == 0)
 			bp->caps |= MACB_CAPS_FIFO_MODE;
-#ifdef CONFIG_MACB_USE_HWSTAMP
 		if (gem_has_ptp(bp)) {
 			if (!GEM_BFEXT(TSU, gem_readl(bp, DCFG5)))
 				dev_err(&bp->pdev->dev,
 					"GEM doesn't support hardware ptp.\n");
 			else {
+#ifdef CONFIG_MACB_USE_HWSTAMP
 				bp->hw_dma_cap |= HW_DMA_CAP_PTP;
 				bp->ptp_info = &gem_ptp_info;
+#endif
 			}
 		}
-#endif
 	}
 
 	dev_dbg(&bp->pdev->dev, "Cadence caps 0x%08x\n", bp->caps);
-- 
cgit v1.2.3-70-g09d2


From ee4e92c26c60b7344b7261035683a37da5a6119b Mon Sep 17 00:00:00 2001
From: Harini Katakam <harini.katakam@xilinx.com>
Date: Tue, 11 Apr 2023 18:07:11 +0530
Subject: net: macb: Enable PTP unicast

Enable transmission and reception of PTP unicast packets by
updating PTP unicast config bit and setting current HW mac
address as allowed address in PTP unicast filter registers.

Signed-off-by: Harini Katakam <harini.katakam@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cadence/macb.h      |  4 ++++
 drivers/net/ethernet/cadence/macb_main.c | 13 +++++++++++--
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 07560803aa26..cfbdd0022764 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -95,6 +95,8 @@
 #define GEM_SA4B		0x00A0 /* Specific4 Bottom */
 #define GEM_SA4T		0x00A4 /* Specific4 Top */
 #define GEM_WOL			0x00b8 /* Wake on LAN */
+#define GEM_RXPTPUNI		0x00D4 /* PTP RX Unicast address */
+#define GEM_TXPTPUNI		0x00D8 /* PTP TX Unicast address */
 #define GEM_EFTSH		0x00e8 /* PTP Event Frame Transmitted Seconds Register 47:32 */
 #define GEM_EFRSH		0x00ec /* PTP Event Frame Received Seconds Register 47:32 */
 #define GEM_PEFTSH		0x00f0 /* PTP Peer Event Frame Transmitted Seconds Register 47:32 */
@@ -245,6 +247,8 @@
 #define MACB_TZQ_OFFSET		12 /* Transmit zero quantum pause frame */
 #define MACB_TZQ_SIZE		1
 #define MACB_SRTSM_OFFSET	15 /* Store Receive Timestamp to Memory */
+#define MACB_PTPUNI_OFFSET	20 /* PTP Unicast packet enable */
+#define MACB_PTPUNI_SIZE	1
 #define MACB_OSSMODE_OFFSET	24 /* Enable One Step Synchro Mode */
 #define MACB_OSSMODE_SIZE	1
 #define MACB_MIIONRGMII_OFFSET	28 /* MII Usage on RGMII Interface */
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index df1a6756042e..29a1199dad14 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -287,6 +287,11 @@ static void macb_set_hwaddr(struct macb *bp)
 	top = cpu_to_le16(*((u16 *)(bp->dev->dev_addr + 4)));
 	macb_or_gem_writel(bp, SA1T, top);
 
+	if (gem_has_ptp(bp)) {
+		gem_writel(bp, RXPTPUNI, bottom);
+		gem_writel(bp, TXPTPUNI, bottom);
+	}
+
 	/* Clear unused address register sets */
 	macb_or_gem_writel(bp, SA2B, 0);
 	macb_or_gem_writel(bp, SA2T, 0);
@@ -773,8 +778,12 @@ static void macb_mac_link_up(struct phylink_config *config,
 
 	spin_unlock_irqrestore(&bp->lock, flags);
 
-	/* Enable Rx and Tx */
-	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(RE) | MACB_BIT(TE));
+	/* Enable Rx and Tx; Enable PTP unicast */
+	ctrl = macb_readl(bp, NCR);
+	if (gem_has_ptp(bp))
+		ctrl |= MACB_BIT(PTPUNI);
+
+	macb_writel(bp, NCR, ctrl | MACB_BIT(RE) | MACB_BIT(TE));
 
 	netif_tx_wake_all_queues(ndev);
 }
-- 
cgit v1.2.3-70-g09d2


From 8c0d0fe04449b9ea2654bf435e51171fbf2a71ac Mon Sep 17 00:00:00 2001
From: Harini Katakam <harini.katakam@xilinx.com>
Date: Tue, 11 Apr 2023 18:07:12 +0530
Subject: net: macb: Optimize reading HW timestamp

The seconds input from BD (6 bits) just needs to be ORed with the
upper bits from timer in this function. Avoid addition operation
every single time. Seconds rollover handling is left untouched.

Signed-off-by: Harini Katakam <harini.katakam@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cadence/macb_ptp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c
index f962a95068a0..51d26fa190d7 100644
--- a/drivers/net/ethernet/cadence/macb_ptp.c
+++ b/drivers/net/ethernet/cadence/macb_ptp.c
@@ -258,6 +258,8 @@ static int gem_hw_timestamp(struct macb *bp, u32 dma_desc_ts_1,
 	 */
 	gem_tsu_get_time(&bp->ptp_clock_info, &tsu, NULL);
 
+	ts->tv_sec |= ((~GEM_DMA_SEC_MASK) & tsu.tv_sec);
+
 	/* If the top bit is set in the timestamp,
 	 * but not in 1588 timer, it has rolled over,
 	 * so subtract max size
@@ -266,8 +268,6 @@ static int gem_hw_timestamp(struct macb *bp, u32 dma_desc_ts_1,
 	    !(tsu.tv_sec & (GEM_DMA_SEC_TOP >> 1)))
 		ts->tv_sec -= GEM_DMA_SEC_TOP;
 
-	ts->tv_sec += ((~GEM_DMA_SEC_MASK) & tsu.tv_sec);
-
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 3ea31e66644b263f43850174847956fbbe2c28cf Mon Sep 17 00:00:00 2001
From: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Date: Wed, 12 Apr 2023 18:22:51 -0700
Subject: tools: ynl: Remove absolute paths to yaml files from ethtool testing
 tool

Absolute paths for the spec and schema files make the ethtool testing tool
unusable with freshly checked-out source trees. Replace absolute paths with
relative paths for both files in the Documentation/ directory.

Issue seen before the change

  Traceback (most recent call last):
    File "/home/binary-eater/Documents/mlx/linux/tools/net/ynl/./ethtool", line 424, in <module>
      main()
    File "/home/binary-eater/Documents/mlx/linux/tools/net/ynl/./ethtool", line 158, in main
      ynl = YnlFamily(spec, schema)
    File "/home/binary-eater/Documents/mlx/linux/tools/net/ynl/lib/ynl.py", line 342, in __init__
      super().__init__(def_path, schema)
    File "/home/binary-eater/Documents/mlx/linux/tools/net/ynl/lib/nlspec.py", line 333, in __init__
      with open(spec_path, "r") as stream:
  FileNotFoundError: [Errno 2] No such file or directory: '/usr/local/google/home/sdf/src/linux/Documentation/netlink/specs/ethtool.yaml'

Fixes: f3d07b02b2b8 ("tools: ynl: ethtool testing tool")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Link: https://lore.kernel.org/r/20230413012252.184434-1-rrameshbabu@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ethtool | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/net/ynl/ethtool b/tools/net/ynl/ethtool
index 5fb1d670693a..6c9f7e31250c 100755
--- a/tools/net/ynl/ethtool
+++ b/tools/net/ynl/ethtool
@@ -152,8 +152,8 @@ def main():
     global args
     args = parser.parse_args()
 
-    spec = '/usr/local/google/home/sdf/src/linux/Documentation/netlink/specs/ethtool.yaml'
-    schema = '/usr/local/google/home/sdf/src/linux/Documentation/netlink/genetlink-legacy.yaml'
+    spec = '../../../Documentation/netlink/specs/ethtool.yaml'
+    schema = '../../../Documentation/netlink/genetlink-legacy.yaml'
 
     ynl = YnlFamily(spec, schema)
 
-- 
cgit v1.2.3-70-g09d2


From 85a4abed155402d37fef9ea29836dbcdf816cb08 Mon Sep 17 00:00:00 2001
From: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Date: Wed, 12 Apr 2023 18:22:52 -0700
Subject: tools: ynl: Rename ethtool to ethtool.py

Make it explicit that this tool is not a drop-in replacement for ethtool.
This tool is intended for testing ethtool functionality implemented in the
kernel and should use a name that differentiates it from the ethtool
utility.

Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Link: https://lore.kernel.org/r/20230413012252.184434-2-rrameshbabu@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ethtool    | 424 -----------------------------------------------
 tools/net/ynl/ethtool.py | 424 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 424 insertions(+), 424 deletions(-)
 delete mode 100755 tools/net/ynl/ethtool
 create mode 100755 tools/net/ynl/ethtool.py

diff --git a/tools/net/ynl/ethtool b/tools/net/ynl/ethtool
deleted file mode 100755
index 6c9f7e31250c..000000000000
--- a/tools/net/ynl/ethtool
+++ /dev/null
@@ -1,424 +0,0 @@
-#!/usr/bin/env python3
-# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
-
-import argparse
-import json
-import pprint
-import sys
-import re
-
-from lib import YnlFamily
-
-def args_to_req(ynl, op_name, args, req):
-    """
-    Verify and convert command-line arguments to the ynl-compatible request.
-    """
-    valid_attrs = ynl.operation_do_attributes(op_name)
-    valid_attrs.remove('header') # not user-provided
-
-    if len(args) == 0:
-        print(f'no attributes, expected: {valid_attrs}')
-        sys.exit(1)
-
-    i = 0
-    while i < len(args):
-        attr = args[i]
-        if i + 1 >= len(args):
-            print(f'expected value for \'{attr}\'')
-            sys.exit(1)
-
-        if attr not in valid_attrs:
-            print(f'invalid attribute \'{attr}\', expected: {valid_attrs}')
-            sys.exit(1)
-
-        val = args[i+1]
-        i += 2
-
-        req[attr] = val
-
-def print_field(reply, *desc):
-    """
-    Pretty-print a set of fields from the reply. desc specifies the
-    fields and the optional type (bool/yn).
-    """
-    if len(desc) == 0:
-        return print_field(reply, *zip(reply.keys(), reply.keys()))
-
-    for spec in desc:
-        try:
-            field, name, tp = spec
-        except:
-            field, name = spec
-            tp = 'int'
-
-        value = reply.get(field, None)
-        if tp == 'yn':
-            value = 'yes' if value else 'no'
-        elif tp == 'bool' or isinstance(value, bool):
-            value = 'on' if value else 'off'
-        else:
-            value = 'n/a' if value is None else value
-
-        print(f'{name}: {value}')
-
-def print_speed(name, value):
-    """
-    Print out the speed-like strings from the value dict.
-    """
-    speed_re = re.compile(r'[0-9]+base[^/]+/.+')
-    speed = [ k for k, v in value.items() if v and speed_re.match(k) ]
-    print(f'{name}: {" ".join(speed)}')
-
-def doit(ynl, args, op_name):
-    """
-    Prepare request header, parse arguments and doit.
-    """
-    req = {
-        'header': {
-          'dev-name': args.device,
-        },
-    }
-
-    args_to_req(ynl, op_name, args.args, req)
-    ynl.do(op_name, req)
-
-def dumpit(ynl, args, op_name, extra = {}):
-    """
-    Prepare request header, parse arguments and dumpit (filtering out the
-    devices we're not interested in).
-    """
-    reply = ynl.dump(op_name, { 'header': {} } | extra)
-    if not reply:
-        return {}
-
-    for msg in reply:
-        if msg['header']['dev-name'] == args.device:
-            if args.json:
-                pprint.PrettyPrinter().pprint(msg)
-                sys.exit(0)
-            msg.pop('header', None)
-            return msg
-
-    print(f"Not supported for device {args.device}")
-    sys.exit(1)
-
-def bits_to_dict(attr):
-    """
-    Convert ynl-formatted bitmask to a dict of bit=value.
-    """
-    ret = {}
-    if 'bits' not in attr:
-        return dict()
-    if 'bit' not in attr['bits']:
-        return dict()
-    for bit in attr['bits']['bit']:
-        if bit['name'] == '':
-            continue
-        name = bit['name']
-        value = bit.get('value', False)
-        ret[name] = value
-    return ret
-
-def main():
-    parser = argparse.ArgumentParser(description='ethtool wannabe')
-    parser.add_argument('--json', action=argparse.BooleanOptionalAction)
-    parser.add_argument('--show-priv-flags', action=argparse.BooleanOptionalAction)
-    parser.add_argument('--set-priv-flags', action=argparse.BooleanOptionalAction)
-    parser.add_argument('--show-eee', action=argparse.BooleanOptionalAction)
-    parser.add_argument('--set-eee', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-a', '--show-pause', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-A', '--set-pause', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-c', '--show-coalesce', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-C', '--set-coalesce', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-g', '--show-ring', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-G', '--set-ring', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-k', '--show-features', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-K', '--set-features', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-l', '--show-channels', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-L', '--set-channels', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-T', '--show-time-stamping', action=argparse.BooleanOptionalAction)
-    parser.add_argument('-S', '--statistics', action=argparse.BooleanOptionalAction)
-    # TODO: --show-tunnels        tunnel-info-get
-    # TODO: --show-module         module-get
-    # TODO: --get-plca-cfg        plca-get
-    # TODO: --get-plca-status     plca-get-status
-    # TODO: --show-mm             mm-get
-    # TODO: --show-fec            fec-get
-    # TODO: --dump-module-eerpom  module-eeprom-get
-    # TODO:                       pse-get
-    # TODO:                       rss-get
-    parser.add_argument('device', metavar='device', type=str)
-    parser.add_argument('args', metavar='args', type=str, nargs='*')
-    global args
-    args = parser.parse_args()
-
-    spec = '../../../Documentation/netlink/specs/ethtool.yaml'
-    schema = '../../../Documentation/netlink/genetlink-legacy.yaml'
-
-    ynl = YnlFamily(spec, schema)
-
-    if args.set_priv_flags:
-        # TODO: parse the bitmask
-        print("not implemented")
-        return
-
-    if args.set_eee:
-        return doit(ynl, args, 'eee-set')
-
-    if args.set_pause:
-        return doit(ynl, args, 'pause-set')
-
-    if args.set_coalesce:
-        return doit(ynl, args, 'coalesce-set')
-
-    if args.set_features:
-        # TODO: parse the bitmask
-        print("not implemented")
-        return
-
-    if args.set_channels:
-        return doit(ynl, args, 'channels-set')
-
-    if args.set_ring:
-        return doit(ynl, args, 'rings-set')
-
-    if args.show_priv_flags:
-        flags = bits_to_dict(dumpit(ynl, args, 'privflags-get')['flags'])
-        print_field(flags)
-        return
-
-    if args.show_eee:
-        eee = dumpit(ynl, args, 'eee-get')
-        ours = bits_to_dict(eee['modes-ours'])
-        peer = bits_to_dict(eee['modes-peer'])
-
-        if 'enabled' in eee:
-            status = 'enabled' if eee['enabled'] else 'disabled'
-            if 'active' in eee and eee['active']:
-                status = status + ' - active'
-            else:
-                status = status + ' - inactive'
-        else:
-            status = 'not supported'
-
-        print(f'EEE status: {status}')
-        print_field(eee, ('tx-lpi-timer', 'Tx LPI'))
-        print_speed('Advertised EEE link modes', ours)
-        print_speed('Link partner advertised EEE link modes', peer)
-
-        return
-
-    if args.show_pause:
-        print_field(dumpit(ynl, args, 'pause-get'),
-                ('autoneg', 'Autonegotiate', 'bool'),
-                ('rx', 'RX', 'bool'),
-                ('tx', 'TX', 'bool'))
-        return
-
-    if args.show_coalesce:
-        print_field(dumpit(ynl, args, 'coalesce-get'))
-        return
-
-    if args.show_features:
-        reply = dumpit(ynl, args, 'features-get')
-        available = bits_to_dict(reply['hw'])
-        requested = bits_to_dict(reply['wanted']).keys()
-        active = bits_to_dict(reply['active']).keys()
-        never_changed = bits_to_dict(reply['nochange']).keys()
-
-        for f in sorted(available):
-            value = "off"
-            if f in active:
-                value = "on"
-
-            fixed = ""
-            if f not in available or f in never_changed:
-                fixed = " [fixed]"
-
-            req = ""
-            if f in requested:
-                if f in active:
-                    req = " [requested on]"
-                else:
-                    req = " [requested off]"
-
-            print(f'{f}: {value}{fixed}{req}')
-
-        return
-
-    if args.show_channels:
-        reply = dumpit(ynl, args, 'channels-get')
-        print(f'Channel parameters for {args.device}:')
-
-        print(f'Pre-set maximums:')
-        print_field(reply,
-            ('rx-max', 'RX'),
-            ('tx-max', 'TX'),
-            ('other-max', 'Other'),
-            ('combined-max', 'Combined'))
-
-        print(f'Current hardware settings:')
-        print_field(reply,
-            ('rx-count', 'RX'),
-            ('tx-count', 'TX'),
-            ('other-count', 'Other'),
-            ('combined-count', 'Combined'))
-
-        return
-
-    if args.show_ring:
-        reply = dumpit(ynl, args, 'channels-get')
-
-        print(f'Ring parameters for {args.device}:')
-
-        print(f'Pre-set maximums:')
-        print_field(reply,
-            ('rx-max', 'RX'),
-            ('rx-mini-max', 'RX Mini'),
-            ('rx-jumbo-max', 'RX Jumbo'),
-            ('tx-max', 'TX'))
-
-        print(f'Current hardware settings:')
-        print_field(reply,
-            ('rx', 'RX'),
-            ('rx-mini', 'RX Mini'),
-            ('rx-jumbo', 'RX Jumbo'),
-            ('tx', 'TX'))
-
-        print_field(reply,
-            ('rx-buf-len', 'RX Buf Len'),
-            ('cqe-size', 'CQE Size'),
-            ('tx-push', 'TX Push', 'bool'))
-
-        return
-
-    if args.statistics:
-        print(f'NIC statistics:')
-
-        # TODO: pass id?
-        strset = dumpit(ynl, args, 'strset-get')
-        pprint.PrettyPrinter().pprint(strset)
-
-        req = {
-          'groups': {
-            'size': 1,
-            'bits': {
-              'bit':
-                # TODO: support passing the bitmask
-                #[
-                  #{ 'name': 'eth-phy', 'value': True },
-                  { 'name': 'eth-mac', 'value': True },
-                  #{ 'name': 'eth-ctrl', 'value': True },
-                  #{ 'name': 'rmon', 'value': True },
-                #],
-            },
-          },
-        }
-
-        rsp = dumpit(ynl, args, 'stats-get', req)
-        pprint.PrettyPrinter().pprint(rsp)
-        return
-
-    if args.show_time_stamping:
-        tsinfo = dumpit(ynl, args, 'tsinfo-get')
-
-        print(f'Time stamping parameters for {args.device}:')
-
-        print('Capabilities:')
-        [print(f'\t{v}') for v in bits_to_dict(tsinfo['timestamping'])]
-
-        print(f'PTP Hardware Clock: {tsinfo["phc-index"]}')
-
-        print('Hardware Transmit Timestamp Modes:')
-        [print(f'\t{v}') for v in bits_to_dict(tsinfo['tx-types'])]
-
-        print('Hardware Receive Filter Modes:')
-        [print(f'\t{v}') for v in bits_to_dict(tsinfo['rx-filters'])]
-        return
-
-    print(f'Settings for {args.device}:')
-    linkmodes = dumpit(ynl, args, 'linkmodes-get')
-    ours = bits_to_dict(linkmodes['ours'])
-
-    supported_ports = ('TP',  'AUI', 'BNC', 'MII', 'FIBRE', 'Backplane')
-    ports = [ p for p in supported_ports if ours.get(p, False)]
-    print(f'Supported ports: [ {" ".join(ports)} ]')
-
-    print_speed('Supported link modes', ours)
-
-    print_field(ours, ('Pause', 'Supported pause frame use', 'yn'))
-    print_field(ours, ('Autoneg', 'Supports auto-negotiation', 'yn'))
-
-    supported_fec = ('None',  'PS', 'BASER', 'LLRS')
-    fec = [ p for p in supported_fec if ours.get(p, False)]
-    fec_str = " ".join(fec)
-    if len(fec) == 0:
-        fec_str = "Not reported"
-
-    print(f'Supported FEC modes: {fec_str}')
-
-    speed = 'Unknown!'
-    if linkmodes['speed'] > 0 and linkmodes['speed'] < 0xffffffff:
-        speed = f'{linkmodes["speed"]}Mb/s'
-    print(f'Speed: {speed}')
-
-    duplex_modes = {
-            0: 'Half',
-            1: 'Full',
-    }
-    duplex = duplex_modes.get(linkmodes["duplex"], None)
-    if not duplex:
-        duplex = f'Unknown! ({linkmodes["duplex"]})'
-    print(f'Duplex: {duplex}')
-
-    autoneg = "off"
-    if linkmodes.get("autoneg", 0) != 0:
-        autoneg = "on"
-    print(f'Auto-negotiation: {autoneg}')
-
-    ports = {
-            0: 'Twisted Pair',
-            1: 'AUI',
-            2: 'MII',
-            3: 'FIBRE',
-            4: 'BNC',
-            5: 'Directly Attached Copper',
-            0xef: 'None',
-    }
-    linkinfo = dumpit(ynl, args, 'linkinfo-get')
-    print(f'Port: {ports.get(linkinfo["port"], "Other")}')
-
-    print_field(linkinfo, ('phyaddr', 'PHYAD'))
-
-    transceiver = {
-            0: 'Internal',
-            1: 'External',
-    }
-    print(f'Transceiver: {transceiver.get(linkinfo["transceiver"], "Unknown")}')
-
-    mdix_ctrl = {
-            1: 'off',
-            2: 'on',
-    }
-    mdix = mdix_ctrl.get(linkinfo['tp-mdix-ctrl'], None)
-    if mdix:
-        mdix = mdix + ' (forced)'
-    else:
-        mdix = mdix_ctrl.get(linkinfo['tp-mdix'], 'Unknown (auto)')
-    print(f'MDI-X: {mdix}')
-
-    debug = dumpit(ynl, args, 'debug-get')
-    msgmask = bits_to_dict(debug.get("msgmask", [])).keys()
-    print(f'Current message level: {" ".join(msgmask)}')
-
-    linkstate = dumpit(ynl, args, 'linkstate-get')
-    detected_states = {
-            0: 'no',
-            1: 'yes',
-    }
-    # TODO: wol-get
-    detected = detected_states.get(linkstate['link'], 'unknown')
-    print(f'Link detected: {detected}')
-
-if __name__ == '__main__':
-    main()
diff --git a/tools/net/ynl/ethtool.py b/tools/net/ynl/ethtool.py
new file mode 100755
index 000000000000..6c9f7e31250c
--- /dev/null
+++ b/tools/net/ynl/ethtool.py
@@ -0,0 +1,424 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+import argparse
+import json
+import pprint
+import sys
+import re
+
+from lib import YnlFamily
+
+def args_to_req(ynl, op_name, args, req):
+    """
+    Verify and convert command-line arguments to the ynl-compatible request.
+    """
+    valid_attrs = ynl.operation_do_attributes(op_name)
+    valid_attrs.remove('header') # not user-provided
+
+    if len(args) == 0:
+        print(f'no attributes, expected: {valid_attrs}')
+        sys.exit(1)
+
+    i = 0
+    while i < len(args):
+        attr = args[i]
+        if i + 1 >= len(args):
+            print(f'expected value for \'{attr}\'')
+            sys.exit(1)
+
+        if attr not in valid_attrs:
+            print(f'invalid attribute \'{attr}\', expected: {valid_attrs}')
+            sys.exit(1)
+
+        val = args[i+1]
+        i += 2
+
+        req[attr] = val
+
+def print_field(reply, *desc):
+    """
+    Pretty-print a set of fields from the reply. desc specifies the
+    fields and the optional type (bool/yn).
+    """
+    if len(desc) == 0:
+        return print_field(reply, *zip(reply.keys(), reply.keys()))
+
+    for spec in desc:
+        try:
+            field, name, tp = spec
+        except:
+            field, name = spec
+            tp = 'int'
+
+        value = reply.get(field, None)
+        if tp == 'yn':
+            value = 'yes' if value else 'no'
+        elif tp == 'bool' or isinstance(value, bool):
+            value = 'on' if value else 'off'
+        else:
+            value = 'n/a' if value is None else value
+
+        print(f'{name}: {value}')
+
+def print_speed(name, value):
+    """
+    Print out the speed-like strings from the value dict.
+    """
+    speed_re = re.compile(r'[0-9]+base[^/]+/.+')
+    speed = [ k for k, v in value.items() if v and speed_re.match(k) ]
+    print(f'{name}: {" ".join(speed)}')
+
+def doit(ynl, args, op_name):
+    """
+    Prepare request header, parse arguments and doit.
+    """
+    req = {
+        'header': {
+          'dev-name': args.device,
+        },
+    }
+
+    args_to_req(ynl, op_name, args.args, req)
+    ynl.do(op_name, req)
+
+def dumpit(ynl, args, op_name, extra = {}):
+    """
+    Prepare request header, parse arguments and dumpit (filtering out the
+    devices we're not interested in).
+    """
+    reply = ynl.dump(op_name, { 'header': {} } | extra)
+    if not reply:
+        return {}
+
+    for msg in reply:
+        if msg['header']['dev-name'] == args.device:
+            if args.json:
+                pprint.PrettyPrinter().pprint(msg)
+                sys.exit(0)
+            msg.pop('header', None)
+            return msg
+
+    print(f"Not supported for device {args.device}")
+    sys.exit(1)
+
+def bits_to_dict(attr):
+    """
+    Convert ynl-formatted bitmask to a dict of bit=value.
+    """
+    ret = {}
+    if 'bits' not in attr:
+        return dict()
+    if 'bit' not in attr['bits']:
+        return dict()
+    for bit in attr['bits']['bit']:
+        if bit['name'] == '':
+            continue
+        name = bit['name']
+        value = bit.get('value', False)
+        ret[name] = value
+    return ret
+
+def main():
+    parser = argparse.ArgumentParser(description='ethtool wannabe')
+    parser.add_argument('--json', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--show-priv-flags', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--set-priv-flags', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--show-eee', action=argparse.BooleanOptionalAction)
+    parser.add_argument('--set-eee', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-a', '--show-pause', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-A', '--set-pause', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-c', '--show-coalesce', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-C', '--set-coalesce', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-g', '--show-ring', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-G', '--set-ring', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-k', '--show-features', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-K', '--set-features', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-l', '--show-channels', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-L', '--set-channels', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-T', '--show-time-stamping', action=argparse.BooleanOptionalAction)
+    parser.add_argument('-S', '--statistics', action=argparse.BooleanOptionalAction)
+    # TODO: --show-tunnels        tunnel-info-get
+    # TODO: --show-module         module-get
+    # TODO: --get-plca-cfg        plca-get
+    # TODO: --get-plca-status     plca-get-status
+    # TODO: --show-mm             mm-get
+    # TODO: --show-fec            fec-get
+    # TODO: --dump-module-eerpom  module-eeprom-get
+    # TODO:                       pse-get
+    # TODO:                       rss-get
+    parser.add_argument('device', metavar='device', type=str)
+    parser.add_argument('args', metavar='args', type=str, nargs='*')
+    global args
+    args = parser.parse_args()
+
+    spec = '../../../Documentation/netlink/specs/ethtool.yaml'
+    schema = '../../../Documentation/netlink/genetlink-legacy.yaml'
+
+    ynl = YnlFamily(spec, schema)
+
+    if args.set_priv_flags:
+        # TODO: parse the bitmask
+        print("not implemented")
+        return
+
+    if args.set_eee:
+        return doit(ynl, args, 'eee-set')
+
+    if args.set_pause:
+        return doit(ynl, args, 'pause-set')
+
+    if args.set_coalesce:
+        return doit(ynl, args, 'coalesce-set')
+
+    if args.set_features:
+        # TODO: parse the bitmask
+        print("not implemented")
+        return
+
+    if args.set_channels:
+        return doit(ynl, args, 'channels-set')
+
+    if args.set_ring:
+        return doit(ynl, args, 'rings-set')
+
+    if args.show_priv_flags:
+        flags = bits_to_dict(dumpit(ynl, args, 'privflags-get')['flags'])
+        print_field(flags)
+        return
+
+    if args.show_eee:
+        eee = dumpit(ynl, args, 'eee-get')
+        ours = bits_to_dict(eee['modes-ours'])
+        peer = bits_to_dict(eee['modes-peer'])
+
+        if 'enabled' in eee:
+            status = 'enabled' if eee['enabled'] else 'disabled'
+            if 'active' in eee and eee['active']:
+                status = status + ' - active'
+            else:
+                status = status + ' - inactive'
+        else:
+            status = 'not supported'
+
+        print(f'EEE status: {status}')
+        print_field(eee, ('tx-lpi-timer', 'Tx LPI'))
+        print_speed('Advertised EEE link modes', ours)
+        print_speed('Link partner advertised EEE link modes', peer)
+
+        return
+
+    if args.show_pause:
+        print_field(dumpit(ynl, args, 'pause-get'),
+                ('autoneg', 'Autonegotiate', 'bool'),
+                ('rx', 'RX', 'bool'),
+                ('tx', 'TX', 'bool'))
+        return
+
+    if args.show_coalesce:
+        print_field(dumpit(ynl, args, 'coalesce-get'))
+        return
+
+    if args.show_features:
+        reply = dumpit(ynl, args, 'features-get')
+        available = bits_to_dict(reply['hw'])
+        requested = bits_to_dict(reply['wanted']).keys()
+        active = bits_to_dict(reply['active']).keys()
+        never_changed = bits_to_dict(reply['nochange']).keys()
+
+        for f in sorted(available):
+            value = "off"
+            if f in active:
+                value = "on"
+
+            fixed = ""
+            if f not in available or f in never_changed:
+                fixed = " [fixed]"
+
+            req = ""
+            if f in requested:
+                if f in active:
+                    req = " [requested on]"
+                else:
+                    req = " [requested off]"
+
+            print(f'{f}: {value}{fixed}{req}')
+
+        return
+
+    if args.show_channels:
+        reply = dumpit(ynl, args, 'channels-get')
+        print(f'Channel parameters for {args.device}:')
+
+        print(f'Pre-set maximums:')
+        print_field(reply,
+            ('rx-max', 'RX'),
+            ('tx-max', 'TX'),
+            ('other-max', 'Other'),
+            ('combined-max', 'Combined'))
+
+        print(f'Current hardware settings:')
+        print_field(reply,
+            ('rx-count', 'RX'),
+            ('tx-count', 'TX'),
+            ('other-count', 'Other'),
+            ('combined-count', 'Combined'))
+
+        return
+
+    if args.show_ring:
+        reply = dumpit(ynl, args, 'channels-get')
+
+        print(f'Ring parameters for {args.device}:')
+
+        print(f'Pre-set maximums:')
+        print_field(reply,
+            ('rx-max', 'RX'),
+            ('rx-mini-max', 'RX Mini'),
+            ('rx-jumbo-max', 'RX Jumbo'),
+            ('tx-max', 'TX'))
+
+        print(f'Current hardware settings:')
+        print_field(reply,
+            ('rx', 'RX'),
+            ('rx-mini', 'RX Mini'),
+            ('rx-jumbo', 'RX Jumbo'),
+            ('tx', 'TX'))
+
+        print_field(reply,
+            ('rx-buf-len', 'RX Buf Len'),
+            ('cqe-size', 'CQE Size'),
+            ('tx-push', 'TX Push', 'bool'))
+
+        return
+
+    if args.statistics:
+        print(f'NIC statistics:')
+
+        # TODO: pass id?
+        strset = dumpit(ynl, args, 'strset-get')
+        pprint.PrettyPrinter().pprint(strset)
+
+        req = {
+          'groups': {
+            'size': 1,
+            'bits': {
+              'bit':
+                # TODO: support passing the bitmask
+                #[
+                  #{ 'name': 'eth-phy', 'value': True },
+                  { 'name': 'eth-mac', 'value': True },
+                  #{ 'name': 'eth-ctrl', 'value': True },
+                  #{ 'name': 'rmon', 'value': True },
+                #],
+            },
+          },
+        }
+
+        rsp = dumpit(ynl, args, 'stats-get', req)
+        pprint.PrettyPrinter().pprint(rsp)
+        return
+
+    if args.show_time_stamping:
+        tsinfo = dumpit(ynl, args, 'tsinfo-get')
+
+        print(f'Time stamping parameters for {args.device}:')
+
+        print('Capabilities:')
+        [print(f'\t{v}') for v in bits_to_dict(tsinfo['timestamping'])]
+
+        print(f'PTP Hardware Clock: {tsinfo["phc-index"]}')
+
+        print('Hardware Transmit Timestamp Modes:')
+        [print(f'\t{v}') for v in bits_to_dict(tsinfo['tx-types'])]
+
+        print('Hardware Receive Filter Modes:')
+        [print(f'\t{v}') for v in bits_to_dict(tsinfo['rx-filters'])]
+        return
+
+    print(f'Settings for {args.device}:')
+    linkmodes = dumpit(ynl, args, 'linkmodes-get')
+    ours = bits_to_dict(linkmodes['ours'])
+
+    supported_ports = ('TP',  'AUI', 'BNC', 'MII', 'FIBRE', 'Backplane')
+    ports = [ p for p in supported_ports if ours.get(p, False)]
+    print(f'Supported ports: [ {" ".join(ports)} ]')
+
+    print_speed('Supported link modes', ours)
+
+    print_field(ours, ('Pause', 'Supported pause frame use', 'yn'))
+    print_field(ours, ('Autoneg', 'Supports auto-negotiation', 'yn'))
+
+    supported_fec = ('None',  'PS', 'BASER', 'LLRS')
+    fec = [ p for p in supported_fec if ours.get(p, False)]
+    fec_str = " ".join(fec)
+    if len(fec) == 0:
+        fec_str = "Not reported"
+
+    print(f'Supported FEC modes: {fec_str}')
+
+    speed = 'Unknown!'
+    if linkmodes['speed'] > 0 and linkmodes['speed'] < 0xffffffff:
+        speed = f'{linkmodes["speed"]}Mb/s'
+    print(f'Speed: {speed}')
+
+    duplex_modes = {
+            0: 'Half',
+            1: 'Full',
+    }
+    duplex = duplex_modes.get(linkmodes["duplex"], None)
+    if not duplex:
+        duplex = f'Unknown! ({linkmodes["duplex"]})'
+    print(f'Duplex: {duplex}')
+
+    autoneg = "off"
+    if linkmodes.get("autoneg", 0) != 0:
+        autoneg = "on"
+    print(f'Auto-negotiation: {autoneg}')
+
+    ports = {
+            0: 'Twisted Pair',
+            1: 'AUI',
+            2: 'MII',
+            3: 'FIBRE',
+            4: 'BNC',
+            5: 'Directly Attached Copper',
+            0xef: 'None',
+    }
+    linkinfo = dumpit(ynl, args, 'linkinfo-get')
+    print(f'Port: {ports.get(linkinfo["port"], "Other")}')
+
+    print_field(linkinfo, ('phyaddr', 'PHYAD'))
+
+    transceiver = {
+            0: 'Internal',
+            1: 'External',
+    }
+    print(f'Transceiver: {transceiver.get(linkinfo["transceiver"], "Unknown")}')
+
+    mdix_ctrl = {
+            1: 'off',
+            2: 'on',
+    }
+    mdix = mdix_ctrl.get(linkinfo['tp-mdix-ctrl'], None)
+    if mdix:
+        mdix = mdix + ' (forced)'
+    else:
+        mdix = mdix_ctrl.get(linkinfo['tp-mdix'], 'Unknown (auto)')
+    print(f'MDI-X: {mdix}')
+
+    debug = dumpit(ynl, args, 'debug-get')
+    msgmask = bits_to_dict(debug.get("msgmask", [])).keys()
+    print(f'Current message level: {" ".join(msgmask)}')
+
+    linkstate = dumpit(ynl, args, 'linkstate-get')
+    detected_states = {
+            0: 'no',
+            1: 'yes',
+    }
+    # TODO: wol-get
+    detected = detected_states.get(linkstate['link'], 'unknown')
+    print(f'Link detected: {detected}')
+
+if __name__ == '__main__':
+    main()
-- 
cgit v1.2.3-70-g09d2


From d54151aa0f4b5c89561705a00d8a5ebb4230028c Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:49 +0300
Subject: net: ethtool: create and export ethtool_dev_mm_supported()

Create a wrapper over __ethtool_dev_mm_supported() which also calls
ethnl_ops_begin() and ethnl_ops_complete(). It can be used by other code
layers, such as tc, to make sure that preemptible TCs are supported
(this is true if an underlying MAC Merge layer exists).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool_netlink.h |  6 ++++++
 net/ethtool/mm.c                | 23 +++++++++++++++++++++++
 2 files changed, 29 insertions(+)

diff --git a/include/linux/ethtool_netlink.h b/include/linux/ethtool_netlink.h
index 17003b385756..fae0dfb9a9c8 100644
--- a/include/linux/ethtool_netlink.h
+++ b/include/linux/ethtool_netlink.h
@@ -39,6 +39,7 @@ void ethtool_aggregate_pause_stats(struct net_device *dev,
 				   struct ethtool_pause_stats *pause_stats);
 void ethtool_aggregate_rmon_stats(struct net_device *dev,
 				  struct ethtool_rmon_stats *rmon_stats);
+bool ethtool_dev_mm_supported(struct net_device *dev);
 
 #else
 static inline int ethnl_cable_test_alloc(struct phy_device *phydev, u8 cmd)
@@ -112,5 +113,10 @@ ethtool_aggregate_rmon_stats(struct net_device *dev,
 {
 }
 
+static inline bool ethtool_dev_mm_supported(struct net_device *dev)
+{
+	return false;
+}
+
 #endif /* IS_ENABLED(CONFIG_ETHTOOL_NETLINK) */
 #endif /* _LINUX_ETHTOOL_NETLINK_H_ */
diff --git a/net/ethtool/mm.c b/net/ethtool/mm.c
index fce3cc2734f9..e00d7d5cea7e 100644
--- a/net/ethtool/mm.c
+++ b/net/ethtool/mm.c
@@ -249,3 +249,26 @@ bool __ethtool_dev_mm_supported(struct net_device *dev)
 
 	return !ret;
 }
+
+bool ethtool_dev_mm_supported(struct net_device *dev)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	bool supported;
+	int ret;
+
+	ASSERT_RTNL();
+
+	if (!ops)
+		return false;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return false;
+
+	supported = __ethtool_dev_mm_supported(dev);
+
+	ethnl_ops_complete(dev);
+
+	return supported;
+}
+EXPORT_SYMBOL_GPL(ethtool_dev_mm_supported);
-- 
cgit v1.2.3-70-g09d2


From 3dd0c16ec93e2bf5212758baea34cfed90d9e6bd Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:50 +0300
Subject: net/sched: mqprio: simplify handling of nlattr portion of TCA_OPTIONS

In commit 4e8b86c06269 ("mqprio: Introduce new hardware offload mode and
shaper in mqprio"), the TCA_OPTIONS format of mqprio was extended to
contain a fixed portion (of size NLA_ALIGN(sizeof struct tc_mqprio_qopt))
and a variable portion of other nlattrs (in the TCA_MQPRIO_* type space)
following immediately afterwards.

In commit feb2cf3dcfb9 ("net/sched: mqprio: refactor nlattr parsing to a
separate function"), we've moved the nlattr handling to a smaller
function, but yet, a small parse_attr() still remains, and the larger
mqprio_parse_nlattr() still does not have access to the beginning, and
the length, of the TCA_OPTIONS region containing these other nlattrs.

In a future change, the mqprio qdisc will need to iterate through this
nlattr region to discover other attributes, so eliminate parse_attr()
and add 2 variables in mqprio_parse_nlattr() which hold the beginning
and the length of the nlattr range.

We avoid the need to memset when nlattr_opt_len has insufficient length
by pre-initializing the table "tb".

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/sch_mqprio.c | 32 +++++++++++++-------------------
 1 file changed, 13 insertions(+), 19 deletions(-)

diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index fdd6a6575a54..89e5dfb22db6 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -146,32 +146,26 @@ static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = {
 	[TCA_MQPRIO_MAX_RATE64]	= { .type = NLA_NESTED },
 };
 
-static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla,
-		      const struct nla_policy *policy, int len)
-{
-	int nested_len = nla_len(nla) - NLA_ALIGN(len);
-
-	if (nested_len >= nla_attr_size(0))
-		return nla_parse_deprecated(tb, maxtype,
-					    nla_data(nla) + NLA_ALIGN(len),
-					    nested_len, policy, NULL);
-
-	memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1));
-	return 0;
-}
-
+/* Parse the other netlink attributes that represent the payload of
+ * TCA_OPTIONS, which are appended right after struct tc_mqprio_qopt.
+ */
 static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 			       struct nlattr *opt)
 {
+	struct nlattr *nlattr_opt = nla_data(opt) + NLA_ALIGN(sizeof(*qopt));
+	int nlattr_opt_len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt));
 	struct mqprio_sched *priv = qdisc_priv(sch);
-	struct nlattr *tb[TCA_MQPRIO_MAX + 1];
+	struct nlattr *tb[TCA_MQPRIO_MAX + 1] = {};
 	struct nlattr *attr;
 	int i, rem, err;
 
-	err = parse_attr(tb, TCA_MQPRIO_MAX, opt, mqprio_policy,
-			 sizeof(*qopt));
-	if (err < 0)
-		return err;
+	if (nlattr_opt_len >= nla_attr_size(0)) {
+		err = nla_parse_deprecated(tb, TCA_MQPRIO_MAX, nlattr_opt,
+					   nlattr_opt_len, mqprio_policy,
+					   NULL);
+		if (err < 0)
+			return err;
+	}
 
 	if (!qopt->hw)
 		return -EINVAL;
-- 
cgit v1.2.3-70-g09d2


From 57f21bf85400abadac0cb2a4db5de1d663f8863f Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:51 +0300
Subject: net/sched: mqprio: add extack to mqprio_parse_nlattr()

Netlink attribute parsing in mqprio is a minesweeper game, with many
options having the possibility of being passed incorrectly and the user
being none the wiser.

Try to make errors less sour by giving user space some information
regarding what went wrong.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/sch_mqprio.c | 30 +++++++++++++++++++++++-------
 1 file changed, 23 insertions(+), 7 deletions(-)

diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 89e5dfb22db6..8e8151ca8307 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -150,7 +150,8 @@ static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = {
  * TCA_OPTIONS, which are appended right after struct tc_mqprio_qopt.
  */
 static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
-			       struct nlattr *opt)
+			       struct nlattr *opt,
+			       struct netlink_ext_ack *extack)
 {
 	struct nlattr *nlattr_opt = nla_data(opt) + NLA_ALIGN(sizeof(*qopt));
 	int nlattr_opt_len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt));
@@ -167,8 +168,11 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 			return err;
 	}
 
-	if (!qopt->hw)
+	if (!qopt->hw) {
+		NL_SET_ERR_MSG(extack,
+			       "mqprio TCA_OPTIONS can only contain netlink attributes in hardware mode");
 		return -EINVAL;
+	}
 
 	if (tb[TCA_MQPRIO_MODE]) {
 		priv->flags |= TC_MQPRIO_F_MODE;
@@ -181,13 +185,19 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 	}
 
 	if (tb[TCA_MQPRIO_MIN_RATE64]) {
-		if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE)
+		if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) {
+			NL_SET_ERR_MSG_ATTR(extack, tb[TCA_MQPRIO_MIN_RATE64],
+					    "min_rate accepted only when shaper is in bw_rlimit mode");
 			return -EINVAL;
+		}
 		i = 0;
 		nla_for_each_nested(attr, tb[TCA_MQPRIO_MIN_RATE64],
 				    rem) {
-			if (nla_type(attr) != TCA_MQPRIO_MIN_RATE64)
+			if (nla_type(attr) != TCA_MQPRIO_MIN_RATE64) {
+				NL_SET_ERR_MSG_ATTR(extack, attr,
+						    "Attribute type expected to be TCA_MQPRIO_MIN_RATE64");
 				return -EINVAL;
+			}
 			if (i >= qopt->num_tc)
 				break;
 			priv->min_rate[i] = nla_get_u64(attr);
@@ -197,13 +207,19 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 	}
 
 	if (tb[TCA_MQPRIO_MAX_RATE64]) {
-		if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE)
+		if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) {
+			NL_SET_ERR_MSG_ATTR(extack, tb[TCA_MQPRIO_MAX_RATE64],
+					    "max_rate accepted only when shaper is in bw_rlimit mode");
 			return -EINVAL;
+		}
 		i = 0;
 		nla_for_each_nested(attr, tb[TCA_MQPRIO_MAX_RATE64],
 				    rem) {
-			if (nla_type(attr) != TCA_MQPRIO_MAX_RATE64)
+			if (nla_type(attr) != TCA_MQPRIO_MAX_RATE64) {
+				NL_SET_ERR_MSG_ATTR(extack, attr,
+						    "Attribute type expected to be TCA_MQPRIO_MAX_RATE64");
 				return -EINVAL;
+			}
 			if (i >= qopt->num_tc)
 				break;
 			priv->max_rate[i] = nla_get_u64(attr);
@@ -252,7 +268,7 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt,
 
 	len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt));
 	if (len > 0) {
-		err = mqprio_parse_nlattr(sch, qopt, opt);
+		err = mqprio_parse_nlattr(sch, qopt, opt, extack);
 		if (err)
 			return err;
 	}
-- 
cgit v1.2.3-70-g09d2


From ab277d2084ba59956b177a7d417791a27361e283 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:52 +0300
Subject: net/sched: mqprio: add an extack message to mqprio_parse_opt()

Ferenc reports that a combination of poor iproute2 defaults and obscure
cases where the kernel returns -EINVAL make it difficult to understand
what is wrong with this command:

$ ip link add veth0 numtxqueues 8 numrxqueues 8 type veth peer name veth1
$ tc qdisc add dev veth0 root mqprio num_tc 8 map 0 1 2 3 4 5 6 7 \
        queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7
RTNETLINK answers: Invalid argument

Hopefully with this patch, the cause is clearer:

Error: Device does not support hardware offload.

The kernel was (and still is) rejecting this because iproute2 defaults
to "hw 1" if this command line option is not specified.

Link: https://lore.kernel.org/netdev/ede5e9a2f27bf83bfb86d3e8c4ca7b34093b99e2.camel@inf.elte.hu/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/sch_mqprio.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 8e8151ca8307..aae4d64dbf3f 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -133,8 +133,11 @@ static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt,
 	/* If ndo_setup_tc is not present then hardware doesn't support offload
 	 * and we should return an error.
 	 */
-	if (qopt->hw && !dev->netdev_ops->ndo_setup_tc)
+	if (qopt->hw && !dev->netdev_ops->ndo_setup_tc) {
+		NL_SET_ERR_MSG(extack,
+			       "Device does not support hardware offload");
 		return -EINVAL;
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From c54876cd5961ce0f8e74807f79a6739cd6b35ddf Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:53 +0300
Subject: net/sched: pass netlink extack to mqprio and taprio offload

With the multiplexed ndo_setup_tc() model which lacks a first-class
struct netlink_ext_ack * argument, the only way to pass the netlink
extended ACK message down to the device driver is to embed it within the
offload structure.

Do this for struct tc_mqprio_qopt_offload and struct tc_taprio_qopt_offload.

Since struct tc_taprio_qopt_offload also contains a tc_mqprio_qopt_offload
structure, and since device drivers might effectively reuse their mqprio
implementation for the mqprio portion of taprio, we make taprio set the
extack in both offload structures to point at the same netlink extack
message.

In fact, the taprio handling is a bit more tricky, for 2 reasons.

First is because the offload structure has a longer lifetime than the
extack structure. The driver is supposed to populate the extack
synchronously from ndo_setup_tc() and leave it alone afterwards.
To not have any use-after-free surprises, we zero out the extack pointer
when we leave taprio_enable_offload().

The second reason is because taprio does overwrite the extack message on
ndo_setup_tc() error. We need to switch to the weak form of setting an
extack message, which preserves a potential message set by the driver.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/pkt_sched.h |  2 ++
 net/sched/sch_mqprio.c  |  5 ++++-
 net/sched/sch_taprio.c  | 12 ++++++++++--
 3 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index bb0bd69fb655..b43ed4733455 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -166,6 +166,7 @@ struct tc_mqprio_caps {
 struct tc_mqprio_qopt_offload {
 	/* struct tc_mqprio_qopt must always be the first element */
 	struct tc_mqprio_qopt qopt;
+	struct netlink_ext_ack *extack;
 	u16 mode;
 	u16 shaper;
 	u32 flags;
@@ -193,6 +194,7 @@ struct tc_taprio_sched_entry {
 
 struct tc_taprio_qopt_offload {
 	struct tc_mqprio_qopt_offload mqprio;
+	struct netlink_ext_ack *extack;
 	u8 enable;
 	ktime_t base_time;
 	u64 cycle_time;
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index aae4d64dbf3f..67d77495c8fd 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -33,9 +33,12 @@ static int mqprio_enable_offload(struct Qdisc *sch,
 				 const struct tc_mqprio_qopt *qopt,
 				 struct netlink_ext_ack *extack)
 {
-	struct tc_mqprio_qopt_offload mqprio = {.qopt = *qopt};
 	struct mqprio_sched *priv = qdisc_priv(sch);
 	struct net_device *dev = qdisc_dev(sch);
+	struct tc_mqprio_qopt_offload mqprio = {
+		.qopt = *qopt,
+		.extack = extack,
+	};
 	int err, i;
 
 	switch (priv->mode) {
diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c
index 1f469861eae3..cbad43019172 100644
--- a/net/sched/sch_taprio.c
+++ b/net/sched/sch_taprio.c
@@ -1520,7 +1520,9 @@ static int taprio_enable_offload(struct net_device *dev,
 		return -ENOMEM;
 	}
 	offload->enable = 1;
+	offload->extack = extack;
 	mqprio_qopt_reconstruct(dev, &offload->mqprio.qopt);
+	offload->mqprio.extack = extack;
 	taprio_sched_to_offload(dev, sched, offload, &caps);
 
 	for (tc = 0; tc < TC_MAX_QUEUE; tc++)
@@ -1528,14 +1530,20 @@ static int taprio_enable_offload(struct net_device *dev,
 
 	err = ops->ndo_setup_tc(dev, TC_SETUP_QDISC_TAPRIO, offload);
 	if (err < 0) {
-		NL_SET_ERR_MSG(extack,
-			       "Device failed to setup taprio offload");
+		NL_SET_ERR_MSG_WEAK(extack,
+				    "Device failed to setup taprio offload");
 		goto done;
 	}
 
 	q->offloaded = true;
 
 done:
+	/* The offload structure may linger around via a reference taken by the
+	 * device driver, so clear up the netlink extack pointer so that the
+	 * driver isn't tempted to dereference data which stopped being valid
+	 */
+	offload->extack = NULL;
+	offload->mqprio.extack = NULL;
 	taprio_offload_free(offload);
 
 	return err;
-- 
cgit v1.2.3-70-g09d2


From f62af20bed2d9e824f51cfc97ff01bc261f40e58 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:54 +0300
Subject: net/sched: mqprio: allow per-TC user input of FP adminStatus

IEEE 802.1Q-2018 clause 6.7.2 Frame preemption specifies that each
packet priority can be assigned to a "frame preemption status" value of
either "express" or "preemptible". Express priorities are transmitted by
the local device through the eMAC, and preemptible priorities through
the pMAC (the concepts of eMAC and pMAC come from the 802.3 MAC Merge
layer).

The FP adminStatus is defined per packet priority, but 802.1Q clause
12.30.1.1.1 framePreemptionAdminStatus also says that:

| Priorities that all map to the same traffic class should be
| constrained to use the same value of preemption status.

It is impossible to ignore the cognitive dissonance in the standard
here, because it practically means that the FP adminStatus only takes
distinct values per traffic class, even though it is defined per
priority.

I can see no valid use case which is prevented by having the kernel take
the FP adminStatus as input per traffic class (what we do here).
In addition, this also enforces the above constraint by construction.
User space network managers which wish to expose FP adminStatus per
priority are free to do so; they must only observe the prio_tc_map of
the netdev (which presumably is also under their control, when
constructing the mqprio netlink attributes).

The reason for configuring frame preemption as a property of the Qdisc
layer is that the information about "preemptible TCs" is closest to the
place which handles the num_tc and prio_tc_map of the netdev. If the
UAPI would have been any other layer, it would be unclear what to do
with the FP information when num_tc collapses to 0. A key assumption is
that only mqprio/taprio change the num_tc and prio_tc_map of the netdev.
Not sure if that's a great assumption to make.

Having FP in tc-mqprio can be seen as an implementation of the use case
defined in 802.1Q Annex S.2 "Preemption used in isolation". There will
be a separate implementation of FP in tc-taprio, for the other use
cases.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/pkt_sched.h        |   1 +
 include/uapi/linux/pkt_sched.h |  16 ++++++
 net/sched/sch_mqprio.c         | 128 ++++++++++++++++++++++++++++++++++++++++-
 net/sched/sch_mqprio_lib.c     |  14 +++++
 net/sched/sch_mqprio_lib.h     |   2 +
 5 files changed, 160 insertions(+), 1 deletion(-)

diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index b43ed4733455..f436688b6efc 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -172,6 +172,7 @@ struct tc_mqprio_qopt_offload {
 	u32 flags;
 	u64 min_rate[TC_QOPT_MAX_QUEUE];
 	u64 max_rate[TC_QOPT_MAX_QUEUE];
+	unsigned long preemptible_tcs;
 };
 
 struct tc_taprio_caps {
diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index 000eec106856..b8d29be91b62 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -719,6 +719,11 @@ enum {
 
 #define __TC_MQPRIO_SHAPER_MAX (__TC_MQPRIO_SHAPER_MAX - 1)
 
+enum {
+	TC_FP_EXPRESS = 1,
+	TC_FP_PREEMPTIBLE = 2,
+};
+
 struct tc_mqprio_qopt {
 	__u8	num_tc;
 	__u8	prio_tc_map[TC_QOPT_BITMASK + 1];
@@ -732,12 +737,23 @@ struct tc_mqprio_qopt {
 #define TC_MQPRIO_F_MIN_RATE		0x4
 #define TC_MQPRIO_F_MAX_RATE		0x8
 
+enum {
+	TCA_MQPRIO_TC_ENTRY_UNSPEC,
+	TCA_MQPRIO_TC_ENTRY_INDEX,		/* u32 */
+	TCA_MQPRIO_TC_ENTRY_FP,			/* u32 */
+
+	/* add new constants above here */
+	__TCA_MQPRIO_TC_ENTRY_CNT,
+	TCA_MQPRIO_TC_ENTRY_MAX = (__TCA_MQPRIO_TC_ENTRY_CNT - 1)
+};
+
 enum {
 	TCA_MQPRIO_UNSPEC,
 	TCA_MQPRIO_MODE,
 	TCA_MQPRIO_SHAPER,
 	TCA_MQPRIO_MIN_RATE64,
 	TCA_MQPRIO_MAX_RATE64,
+	TCA_MQPRIO_TC_ENTRY,
 	__TCA_MQPRIO_MAX,
 };
 
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 67d77495c8fd..dc5a0ff50b14 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -5,6 +5,7 @@
  * Copyright (c) 2010 John Fastabend <john.r.fastabend@intel.com>
  */
 
+#include <linux/ethtool_netlink.h>
 #include <linux/types.h>
 #include <linux/slab.h>
 #include <linux/kernel.h>
@@ -27,6 +28,7 @@ struct mqprio_sched {
 	u32 flags;
 	u64 min_rate[TC_QOPT_MAX_QUEUE];
 	u64 max_rate[TC_QOPT_MAX_QUEUE];
+	u32 fp[TC_QOPT_MAX_QUEUE];
 };
 
 static int mqprio_enable_offload(struct Qdisc *sch,
@@ -63,6 +65,8 @@ static int mqprio_enable_offload(struct Qdisc *sch,
 		return -EINVAL;
 	}
 
+	mqprio_fp_to_offload(priv->fp, &mqprio);
+
 	err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_MQPRIO,
 					    &mqprio);
 	if (err)
@@ -145,13 +149,95 @@ static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt,
 	return 0;
 }
 
+static const struct
+nla_policy mqprio_tc_entry_policy[TCA_MQPRIO_TC_ENTRY_MAX + 1] = {
+	[TCA_MQPRIO_TC_ENTRY_INDEX]	= NLA_POLICY_MAX(NLA_U32,
+							 TC_QOPT_MAX_QUEUE),
+	[TCA_MQPRIO_TC_ENTRY_FP]	= NLA_POLICY_RANGE(NLA_U32,
+							   TC_FP_EXPRESS,
+							   TC_FP_PREEMPTIBLE),
+};
+
 static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = {
 	[TCA_MQPRIO_MODE]	= { .len = sizeof(u16) },
 	[TCA_MQPRIO_SHAPER]	= { .len = sizeof(u16) },
 	[TCA_MQPRIO_MIN_RATE64]	= { .type = NLA_NESTED },
 	[TCA_MQPRIO_MAX_RATE64]	= { .type = NLA_NESTED },
+	[TCA_MQPRIO_TC_ENTRY]	= { .type = NLA_NESTED },
 };
 
+static int mqprio_parse_tc_entry(u32 fp[TC_QOPT_MAX_QUEUE],
+				 struct nlattr *opt,
+				 unsigned long *seen_tcs,
+				 struct netlink_ext_ack *extack)
+{
+	struct nlattr *tb[TCA_MQPRIO_TC_ENTRY_MAX + 1];
+	int err, tc;
+
+	err = nla_parse_nested(tb, TCA_MQPRIO_TC_ENTRY_MAX, opt,
+			       mqprio_tc_entry_policy, extack);
+	if (err < 0)
+		return err;
+
+	if (NL_REQ_ATTR_CHECK(extack, opt, tb, TCA_MQPRIO_TC_ENTRY_INDEX)) {
+		NL_SET_ERR_MSG(extack, "TC entry index missing");
+		return -EINVAL;
+	}
+
+	tc = nla_get_u32(tb[TCA_MQPRIO_TC_ENTRY_INDEX]);
+	if (*seen_tcs & BIT(tc)) {
+		NL_SET_ERR_MSG_ATTR(extack, tb[TCA_MQPRIO_TC_ENTRY_INDEX],
+				    "Duplicate tc entry");
+		return -EINVAL;
+	}
+
+	*seen_tcs |= BIT(tc);
+
+	if (tb[TCA_MQPRIO_TC_ENTRY_FP])
+		fp[tc] = nla_get_u32(tb[TCA_MQPRIO_TC_ENTRY_FP]);
+
+	return 0;
+}
+
+static int mqprio_parse_tc_entries(struct Qdisc *sch, struct nlattr *nlattr_opt,
+				   int nlattr_opt_len,
+				   struct netlink_ext_ack *extack)
+{
+	struct mqprio_sched *priv = qdisc_priv(sch);
+	struct net_device *dev = qdisc_dev(sch);
+	bool have_preemption = false;
+	unsigned long seen_tcs = 0;
+	u32 fp[TC_QOPT_MAX_QUEUE];
+	struct nlattr *n;
+	int tc, rem;
+	int err = 0;
+
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++)
+		fp[tc] = priv->fp[tc];
+
+	nla_for_each_attr(n, nlattr_opt, nlattr_opt_len, rem) {
+		if (nla_type(n) != TCA_MQPRIO_TC_ENTRY)
+			continue;
+
+		err = mqprio_parse_tc_entry(fp, n, &seen_tcs, extack);
+		if (err)
+			goto out;
+	}
+
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) {
+		priv->fp[tc] = fp[tc];
+		if (fp[tc] == TC_FP_PREEMPTIBLE)
+			have_preemption = true;
+	}
+
+	if (have_preemption && !ethtool_dev_mm_supported(dev)) {
+		NL_SET_ERR_MSG(extack, "Device does not support preemption");
+		return -EOPNOTSUPP;
+	}
+out:
+	return err;
+}
+
 /* Parse the other netlink attributes that represent the payload of
  * TCA_OPTIONS, which are appended right after struct tc_mqprio_qopt.
  */
@@ -234,6 +320,13 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt,
 		priv->flags |= TC_MQPRIO_F_MAX_RATE;
 	}
 
+	if (tb[TCA_MQPRIO_TC_ENTRY]) {
+		err = mqprio_parse_tc_entries(sch, nlattr_opt, nlattr_opt_len,
+					      extack);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
@@ -247,7 +340,7 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt,
 	int i, err = -EOPNOTSUPP;
 	struct tc_mqprio_qopt *qopt = NULL;
 	struct tc_mqprio_caps caps;
-	int len;
+	int len, tc;
 
 	BUILD_BUG_ON(TC_MAX_QUEUE != TC_QOPT_MAX_QUEUE);
 	BUILD_BUG_ON(TC_BITMASK != TC_QOPT_BITMASK);
@@ -265,6 +358,9 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt,
 	if (!opt || nla_len(opt) < sizeof(*qopt))
 		return -EINVAL;
 
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++)
+		priv->fp[tc] = TC_FP_EXPRESS;
+
 	qdisc_offload_query_caps(dev, TC_SETUP_QDISC_MQPRIO,
 				 &caps, sizeof(caps));
 
@@ -415,6 +511,33 @@ nla_put_failure:
 	return -1;
 }
 
+static int mqprio_dump_tc_entries(struct mqprio_sched *priv,
+				  struct sk_buff *skb)
+{
+	struct nlattr *n;
+	int tc;
+
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) {
+		n = nla_nest_start(skb, TCA_MQPRIO_TC_ENTRY);
+		if (!n)
+			return -EMSGSIZE;
+
+		if (nla_put_u32(skb, TCA_MQPRIO_TC_ENTRY_INDEX, tc))
+			goto nla_put_failure;
+
+		if (nla_put_u32(skb, TCA_MQPRIO_TC_ENTRY_FP, priv->fp[tc]))
+			goto nla_put_failure;
+
+		nla_nest_end(skb, n);
+	}
+
+	return 0;
+
+nla_put_failure:
+	nla_nest_cancel(skb, n);
+	return -EMSGSIZE;
+}
+
 static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
 	struct net_device *dev = qdisc_dev(sch);
@@ -465,6 +588,9 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb)
 	    (dump_rates(priv, &opt, skb) != 0))
 		goto nla_put_failure;
 
+	if (mqprio_dump_tc_entries(priv, skb))
+		goto nla_put_failure;
+
 	return nla_nest_end(skb, nla);
 nla_put_failure:
 	nlmsg_trim(skb, nla);
diff --git a/net/sched/sch_mqprio_lib.c b/net/sched/sch_mqprio_lib.c
index c58a533b8ec5..83b3793c4012 100644
--- a/net/sched/sch_mqprio_lib.c
+++ b/net/sched/sch_mqprio_lib.c
@@ -114,4 +114,18 @@ void mqprio_qopt_reconstruct(struct net_device *dev, struct tc_mqprio_qopt *qopt
 }
 EXPORT_SYMBOL_GPL(mqprio_qopt_reconstruct);
 
+void mqprio_fp_to_offload(u32 fp[TC_QOPT_MAX_QUEUE],
+			  struct tc_mqprio_qopt_offload *mqprio)
+{
+	unsigned long preemptible_tcs = 0;
+	int tc;
+
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++)
+		if (fp[tc] == TC_FP_PREEMPTIBLE)
+			preemptible_tcs |= BIT(tc);
+
+	mqprio->preemptible_tcs = preemptible_tcs;
+}
+EXPORT_SYMBOL_GPL(mqprio_fp_to_offload);
+
 MODULE_LICENSE("GPL");
diff --git a/net/sched/sch_mqprio_lib.h b/net/sched/sch_mqprio_lib.h
index 63f725ab8761..079f597072e3 100644
--- a/net/sched/sch_mqprio_lib.h
+++ b/net/sched/sch_mqprio_lib.h
@@ -14,5 +14,7 @@ int mqprio_validate_qopt(struct net_device *dev, struct tc_mqprio_qopt *qopt,
 			 struct netlink_ext_ack *extack);
 void mqprio_qopt_reconstruct(struct net_device *dev,
 			     struct tc_mqprio_qopt *qopt);
+void mqprio_fp_to_offload(u32 fp[TC_QOPT_MAX_QUEUE],
+			  struct tc_mqprio_qopt_offload *mqprio);
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From a721c3e54b80e45cd9202e7fca29ef018bed9069 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:55 +0300
Subject: net/sched: taprio: allow per-TC user input of FP adminStatus

This is a duplication of the FP adminStatus logic introduced for
tc-mqprio. Offloading is done through the tc_mqprio_qopt_offload
structure embedded within tc_taprio_qopt_offload. So practically, if a
device driver is written to treat the mqprio portion of taprio just like
standalone mqprio, it gets unified handling of frame preemption.

I would have reused more code with taprio, but this is mostly netlink
attribute parsing, which is hard to transform into generic code without
having something that stinks as a result. We have the same variables
with the same semantics, just different nlattr type values
(TCA_MQPRIO_TC_ENTRY=5 vs TCA_TAPRIO_ATTR_TC_ENTRY=12;
TCA_MQPRIO_TC_ENTRY_FP=2 vs TCA_TAPRIO_TC_ENTRY_FP=3, etc) and
consequently, different policies for the nest.

Every time nla_parse_nested() is called, an on-stack table "tb" of
nlattr pointers is allocated statically, up to the maximum understood
nlattr type. That array size is hardcoded as a constant, but when
transforming this into a common parsing function, it would become either
a VLA (which the Linux kernel rightfully doesn't like) or a call to the
allocator.

Having FP adminStatus in tc-taprio can be seen as addressing the 802.1Q
Annex S.3 "Scheduling and preemption used in combination, no HOLD/RELEASE"
and S.4 "Scheduling and preemption used in combination with HOLD/RELEASE"
use cases. HOLD and RELEASE events are emitted towards the underlying
MAC Merge layer when the schedule hits a Set-And-Hold-MAC or a
Set-And-Release-MAC gate operation. So within the tc-taprio UAPI space,
one can distinguish between the 2 use cases by choosing whether to use
the TC_TAPRIO_CMD_SET_AND_HOLD and TC_TAPRIO_CMD_SET_AND_RELEASE gate
operations within the schedule, or just TC_TAPRIO_CMD_SET_GATES.

A small part of the change is dedicated to refactoring the max_sdu
nlattr parsing to put all logic under the "if" that tests for presence
of that nlattr.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/uapi/linux/pkt_sched.h |  1 +
 net/sched/sch_taprio.c         | 65 +++++++++++++++++++++++++++++++++---------
 2 files changed, 53 insertions(+), 13 deletions(-)

diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index b8d29be91b62..51a7addc56c6 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -1252,6 +1252,7 @@ enum {
 	TCA_TAPRIO_TC_ENTRY_UNSPEC,
 	TCA_TAPRIO_TC_ENTRY_INDEX,		/* u32 */
 	TCA_TAPRIO_TC_ENTRY_MAX_SDU,		/* u32 */
+	TCA_TAPRIO_TC_ENTRY_FP,			/* u32 */
 
 	/* add new constants above here */
 	__TCA_TAPRIO_TC_ENTRY_CNT,
diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c
index cbad43019172..76db9a10ef50 100644
--- a/net/sched/sch_taprio.c
+++ b/net/sched/sch_taprio.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/ethtool.h>
+#include <linux/ethtool_netlink.h>
 #include <linux/types.h>
 #include <linux/slab.h>
 #include <linux/kernel.h>
@@ -96,6 +97,7 @@ struct taprio_sched {
 	struct list_head taprio_list;
 	int cur_txq[TC_MAX_QUEUE];
 	u32 max_sdu[TC_MAX_QUEUE]; /* save info from the user */
+	u32 fp[TC_QOPT_MAX_QUEUE]; /* only for dump and offloading */
 	u32 txtime_delay;
 };
 
@@ -1002,6 +1004,9 @@ static const struct nla_policy entry_policy[TCA_TAPRIO_SCHED_ENTRY_MAX + 1] = {
 static const struct nla_policy taprio_tc_policy[TCA_TAPRIO_TC_ENTRY_MAX + 1] = {
 	[TCA_TAPRIO_TC_ENTRY_INDEX]	   = { .type = NLA_U32 },
 	[TCA_TAPRIO_TC_ENTRY_MAX_SDU]	   = { .type = NLA_U32 },
+	[TCA_TAPRIO_TC_ENTRY_FP]	   = NLA_POLICY_RANGE(NLA_U32,
+							      TC_FP_EXPRESS,
+							      TC_FP_PREEMPTIBLE),
 };
 
 static const struct nla_policy taprio_policy[TCA_TAPRIO_ATTR_MAX + 1] = {
@@ -1524,6 +1529,7 @@ static int taprio_enable_offload(struct net_device *dev,
 	mqprio_qopt_reconstruct(dev, &offload->mqprio.qopt);
 	offload->mqprio.extack = extack;
 	taprio_sched_to_offload(dev, sched, offload, &caps);
+	mqprio_fp_to_offload(q->fp, &offload->mqprio);
 
 	for (tc = 0; tc < TC_MAX_QUEUE; tc++)
 		offload->max_sdu[tc] = q->max_sdu[tc];
@@ -1671,13 +1677,14 @@ out:
 static int taprio_parse_tc_entry(struct Qdisc *sch,
 				 struct nlattr *opt,
 				 u32 max_sdu[TC_QOPT_MAX_QUEUE],
+				 u32 fp[TC_QOPT_MAX_QUEUE],
 				 unsigned long *seen_tcs,
 				 struct netlink_ext_ack *extack)
 {
 	struct nlattr *tb[TCA_TAPRIO_TC_ENTRY_MAX + 1] = { };
 	struct net_device *dev = qdisc_dev(sch);
-	u32 val = 0;
 	int err, tc;
+	u32 val;
 
 	err = nla_parse_nested(tb, TCA_TAPRIO_TC_ENTRY_MAX, opt,
 			       taprio_tc_policy, extack);
@@ -1702,15 +1709,18 @@ static int taprio_parse_tc_entry(struct Qdisc *sch,
 
 	*seen_tcs |= BIT(tc);
 
-	if (tb[TCA_TAPRIO_TC_ENTRY_MAX_SDU])
+	if (tb[TCA_TAPRIO_TC_ENTRY_MAX_SDU]) {
 		val = nla_get_u32(tb[TCA_TAPRIO_TC_ENTRY_MAX_SDU]);
+		if (val > dev->max_mtu) {
+			NL_SET_ERR_MSG_MOD(extack, "TC max SDU exceeds device max MTU");
+			return -ERANGE;
+		}
 
-	if (val > dev->max_mtu) {
-		NL_SET_ERR_MSG_MOD(extack, "TC max SDU exceeds device max MTU");
-		return -ERANGE;
+		max_sdu[tc] = val;
 	}
 
-	max_sdu[tc] = val;
+	if (tb[TCA_TAPRIO_TC_ENTRY_FP])
+		fp[tc] = nla_get_u32(tb[TCA_TAPRIO_TC_ENTRY_FP]);
 
 	return 0;
 }
@@ -1720,29 +1730,51 @@ static int taprio_parse_tc_entries(struct Qdisc *sch,
 				   struct netlink_ext_ack *extack)
 {
 	struct taprio_sched *q = qdisc_priv(sch);
+	struct net_device *dev = qdisc_dev(sch);
 	u32 max_sdu[TC_QOPT_MAX_QUEUE];
+	bool have_preemption = false;
 	unsigned long seen_tcs = 0;
+	u32 fp[TC_QOPT_MAX_QUEUE];
 	struct nlattr *n;
 	int tc, rem;
 	int err = 0;
 
-	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++)
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) {
 		max_sdu[tc] = q->max_sdu[tc];
+		fp[tc] = q->fp[tc];
+	}
 
 	nla_for_each_nested(n, opt, rem) {
 		if (nla_type(n) != TCA_TAPRIO_ATTR_TC_ENTRY)
 			continue;
 
-		err = taprio_parse_tc_entry(sch, n, max_sdu, &seen_tcs,
+		err = taprio_parse_tc_entry(sch, n, max_sdu, fp, &seen_tcs,
 					    extack);
 		if (err)
-			goto out;
+			return err;
 	}
 
-	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++)
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) {
 		q->max_sdu[tc] = max_sdu[tc];
+		q->fp[tc] = fp[tc];
+		if (fp[tc] != TC_FP_EXPRESS)
+			have_preemption = true;
+	}
+
+	if (have_preemption) {
+		if (!FULL_OFFLOAD_IS_ENABLED(q->flags)) {
+			NL_SET_ERR_MSG(extack,
+				       "Preemption only supported with full offload");
+			return -EOPNOTSUPP;
+		}
+
+		if (!ethtool_dev_mm_supported(dev)) {
+			NL_SET_ERR_MSG(extack,
+				       "Device does not support preemption");
+			return -EOPNOTSUPP;
+		}
+	}
 
-out:
 	return err;
 }
 
@@ -2023,7 +2055,7 @@ static int taprio_init(struct Qdisc *sch, struct nlattr *opt,
 {
 	struct taprio_sched *q = qdisc_priv(sch);
 	struct net_device *dev = qdisc_dev(sch);
-	int i;
+	int i, tc;
 
 	spin_lock_init(&q->current_entry_lock);
 
@@ -2080,6 +2112,9 @@ static int taprio_init(struct Qdisc *sch, struct nlattr *opt,
 		q->qdiscs[i] = qdisc;
 	}
 
+	for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++)
+		q->fp[tc] = TC_FP_EXPRESS;
+
 	taprio_detect_broken_mqprio(q);
 
 	return taprio_change(sch, opt, extack);
@@ -2223,6 +2258,7 @@ error_nest:
 }
 
 static int taprio_dump_tc_entries(struct sk_buff *skb,
+				  struct taprio_sched *q,
 				  struct sched_gate_list *sched)
 {
 	struct nlattr *n;
@@ -2240,6 +2276,9 @@ static int taprio_dump_tc_entries(struct sk_buff *skb,
 				sched->max_sdu[tc]))
 			goto nla_put_failure;
 
+		if (nla_put_u32(skb, TCA_TAPRIO_TC_ENTRY_FP, q->fp[tc]))
+			goto nla_put_failure;
+
 		nla_nest_end(skb, n);
 	}
 
@@ -2281,7 +2320,7 @@ static int taprio_dump(struct Qdisc *sch, struct sk_buff *skb)
 	    nla_put_u32(skb, TCA_TAPRIO_ATTR_TXTIME_DELAY, q->txtime_delay))
 		goto options_error;
 
-	if (oper && taprio_dump_tc_entries(skb, oper))
+	if (oper && taprio_dump_tc_entries(skb, q, oper))
 		goto options_error;
 
 	if (oper && dump_schedule(skb, oper))
-- 
cgit v1.2.3-70-g09d2


From 50764da37cbed7097898a49ae5ba8337205ab4ac Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:56 +0300
Subject: net: enetc: rename "mqprio" to "qopt"

To gain access to the larger encapsulating structure which has the type
tc_mqprio_qopt_offload, rename just the "qopt" field as "qopt".

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/enetc/enetc.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 2fc712b24d12..e0207b01ddd6 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -2644,12 +2644,13 @@ static void enetc_reset_tc_mqprio(struct net_device *ndev)
 
 int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
 {
+	struct tc_mqprio_qopt_offload *mqprio = type_data;
 	struct enetc_ndev_priv *priv = netdev_priv(ndev);
-	struct tc_mqprio_qopt *mqprio = type_data;
+	struct tc_mqprio_qopt *qopt = &mqprio->qopt;
 	struct enetc_hw *hw = &priv->si->hw;
 	int num_stack_tx_queues = 0;
-	u8 num_tc = mqprio->num_tc;
 	struct enetc_bdr *tx_ring;
+	u8 num_tc = qopt->num_tc;
 	int offset, count;
 	int err, tc, q;
 
@@ -2663,8 +2664,8 @@ int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
 		return err;
 
 	for (tc = 0; tc < num_tc; tc++) {
-		offset = mqprio->offset[tc];
-		count = mqprio->count[tc];
+		offset = qopt->offset[tc];
+		count = qopt->count[tc];
 		num_stack_tx_queues += count;
 
 		err = netdev_set_tc_queue(ndev, tc, count, offset);
-- 
cgit v1.2.3-70-g09d2


From 01e23b2b3bad2d6e996d546f69c6ee804d0fc92f Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 11 Apr 2023 21:01:57 +0300
Subject: net: enetc: add support for preemptible traffic classes

PFs which support the MAC Merge layer also have a set of 8 registers
called "Port traffic class N frame preemption register (PTC0FPR - PTC7FPR)".
Through these, a traffic class (group of TX rings of same dequeue
priority) can be mapped to the eMAC or to the pMAC.

There's nothing particularly spectacular here. We should probably only
commit the preemptible TCs to hardware once the MAC Merge layer became
active, but unlike Felix, we don't have an IRQ that notifies us of that.
We'd have to sleep for up to verifyTime (127 ms) to wait for a
resolution coming from the verification state machine; not only from the
ndo_setup_tc() code path, but also from enetc_mm_link_state_update().
Since it's relatively complicated and has a relatively small benefit,
I'm not doing it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/enetc/enetc.c    | 22 ++++++++++++++++++++++
 drivers/net/ethernet/freescale/enetc/enetc.h    |  1 +
 drivers/net/ethernet/freescale/enetc/enetc_hw.h |  4 ++++
 3 files changed, 27 insertions(+)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index e0207b01ddd6..41c194c1672d 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -25,6 +25,24 @@ void enetc_port_mac_wr(struct enetc_si *si, u32 reg, u32 val)
 }
 EXPORT_SYMBOL_GPL(enetc_port_mac_wr);
 
+void enetc_set_ptcfpr(struct enetc_hw *hw, unsigned long preemptible_tcs)
+{
+	u32 val;
+	int tc;
+
+	for (tc = 0; tc < 8; tc++) {
+		val = enetc_port_rd(hw, ENETC_PTCFPR(tc));
+
+		if (preemptible_tcs & BIT(tc))
+			val |= ENETC_PTCFPR_FPE;
+		else
+			val &= ~ENETC_PTCFPR_FPE;
+
+		enetc_port_wr(hw, ENETC_PTCFPR(tc), val);
+	}
+}
+EXPORT_SYMBOL_GPL(enetc_set_ptcfpr);
+
 static int enetc_num_stack_tx_queues(struct enetc_ndev_priv *priv)
 {
 	int num_tx_rings = priv->num_tx_rings;
@@ -2640,6 +2658,8 @@ static void enetc_reset_tc_mqprio(struct net_device *ndev)
 	}
 
 	enetc_debug_tx_ring_prios(priv);
+
+	enetc_set_ptcfpr(hw, 0);
 }
 
 int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
@@ -2694,6 +2714,8 @@ int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
 
 	enetc_debug_tx_ring_prios(priv);
 
+	enetc_set_ptcfpr(hw, mqprio->preemptible_tcs);
+
 	return 0;
 
 err_reset_tc:
diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h
index 8010f31cd10d..143078a9ef16 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc.h
@@ -486,6 +486,7 @@ static inline void enetc_cbd_free_data_mem(struct enetc_si *si, int size,
 
 void enetc_reset_ptcmsdur(struct enetc_hw *hw);
 void enetc_set_ptcmsdur(struct enetc_hw *hw, u32 *queue_max_sdu);
+void enetc_set_ptcfpr(struct enetc_hw *hw, unsigned long preemptible_tcs);
 
 #ifdef CONFIG_FSL_ENETC_QOS
 int enetc_qos_query_caps(struct net_device *ndev, void *type_data);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
index de2e0ee8cdcb..36bb2d6d5658 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
@@ -965,6 +965,10 @@ static inline u32 enetc_usecs_to_cycles(u32 usecs)
 	return (u32)div_u64(usecs * ENETC_CLK, 1000000ULL);
 }
 
+/* Port traffic class frame preemption register */
+#define ENETC_PTCFPR(n)			(0x1910 + (n) * 4) /* n = [0 ..7] */
+#define ENETC_PTCFPR_FPE		BIT(31)
+
 /* port time gating control register */
 #define ENETC_PTGCR			0x11a00
 #define ENETC_PTGCR_TGE			BIT(31)
-- 
cgit v1.2.3-70-g09d2


From ce518bc3e9ca342309995c9270c3ec4892963695 Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Wed, 12 Apr 2023 14:16:00 -0700
Subject: net: mana: Use napi_build_skb in RX path

Use napi_build_skb() instead of build_skb() to take advantage of the
NAPI percpu caches to obtain skbuff_head.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 492474b4d8aa..112c642dc89b 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -1188,7 +1188,7 @@ static void mana_post_pkt_rxq(struct mana_rxq *rxq)
 static struct sk_buff *mana_build_skb(void *buf_va, uint pkt_len,
 				      struct xdp_buff *xdp)
 {
-	struct sk_buff *skb = build_skb(buf_va, PAGE_SIZE);
+	struct sk_buff *skb = napi_build_skb(buf_va, PAGE_SIZE);
 
 	if (!skb)
 		return NULL;
-- 
cgit v1.2.3-70-g09d2


From a2917b23497e4205db32271e4e06e142a9f8a6aa Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Wed, 12 Apr 2023 14:16:01 -0700
Subject: net: mana: Refactor RX buffer allocation code to prepare for various
 MTU

Move out common buffer allocation code from mana_process_rx_cqe() and
mana_alloc_rx_wqe() to helper functions.
Refactor related variables so they can be changed in one place, and buffer
sizes are in sync.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 154 +++++++++++++++-----------
 include/net/mana/mana.h                       |   6 +-
 2 files changed, 91 insertions(+), 69 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 112c642dc89b..911954ff84ee 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -1282,14 +1282,64 @@ drop_xdp:
 	u64_stats_update_end(&rx_stats->syncp);
 
 drop:
-	WARN_ON_ONCE(rxq->xdp_save_page);
-	rxq->xdp_save_page = virt_to_page(buf_va);
+	WARN_ON_ONCE(rxq->xdp_save_va);
+	/* Save for reuse */
+	rxq->xdp_save_va = buf_va;
 
 	++ndev->stats.rx_dropped;
 
 	return;
 }
 
+static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
+			     dma_addr_t *da, bool is_napi)
+{
+	struct page *page;
+	void *va;
+
+	/* Reuse XDP dropped page if available */
+	if (rxq->xdp_save_va) {
+		va = rxq->xdp_save_va;
+		rxq->xdp_save_va = NULL;
+	} else {
+		page = dev_alloc_page();
+		if (!page)
+			return NULL;
+
+		va = page_to_virt(page);
+	}
+
+	*da = dma_map_single(dev, va + XDP_PACKET_HEADROOM, rxq->datasize,
+			     DMA_FROM_DEVICE);
+
+	if (dma_mapping_error(dev, *da)) {
+		put_page(virt_to_head_page(va));
+		return NULL;
+	}
+
+	return va;
+}
+
+/* Allocate frag for rx buffer, and save the old buf */
+static void mana_refill_rxoob(struct device *dev, struct mana_rxq *rxq,
+			      struct mana_recv_buf_oob *rxoob, void **old_buf)
+{
+	dma_addr_t da;
+	void *va;
+
+	va = mana_get_rxfrag(rxq, dev, &da, true);
+
+	if (!va)
+		return;
+
+	dma_unmap_single(dev, rxoob->sgl[0].address, rxq->datasize,
+			 DMA_FROM_DEVICE);
+	*old_buf = rxoob->buf_va;
+
+	rxoob->buf_va = va;
+	rxoob->sgl[0].address = da;
+}
+
 static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 				struct gdma_comp *cqe)
 {
@@ -1299,10 +1349,8 @@ static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	struct mana_recv_buf_oob *rxbuf_oob;
 	struct mana_port_context *apc;
 	struct device *dev = gc->dev;
-	void *new_buf, *old_buf;
-	struct page *new_page;
+	void *old_buf = NULL;
 	u32 curr, pktlen;
-	dma_addr_t da;
 
 	apc = netdev_priv(ndev);
 
@@ -1345,40 +1393,11 @@ static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	rxbuf_oob = &rxq->rx_oobs[curr];
 	WARN_ON_ONCE(rxbuf_oob->wqe_inf.wqe_size_in_bu != 1);
 
-	/* Reuse XDP dropped page if available */
-	if (rxq->xdp_save_page) {
-		new_page = rxq->xdp_save_page;
-		rxq->xdp_save_page = NULL;
-	} else {
-		new_page = alloc_page(GFP_ATOMIC);
-	}
-
-	if (new_page) {
-		da = dma_map_page(dev, new_page, XDP_PACKET_HEADROOM, rxq->datasize,
-				  DMA_FROM_DEVICE);
-
-		if (dma_mapping_error(dev, da)) {
-			__free_page(new_page);
-			new_page = NULL;
-		}
-	}
-
-	new_buf = new_page ? page_to_virt(new_page) : NULL;
-
-	if (new_buf) {
-		dma_unmap_page(dev, rxbuf_oob->buf_dma_addr, rxq->datasize,
-			       DMA_FROM_DEVICE);
-
-		old_buf = rxbuf_oob->buf_va;
-
-		/* refresh the rxbuf_oob with the new page */
-		rxbuf_oob->buf_va = new_buf;
-		rxbuf_oob->buf_dma_addr = da;
-		rxbuf_oob->sgl[0].address = rxbuf_oob->buf_dma_addr;
-	} else {
-		old_buf = NULL; /* drop the packet if no memory */
-	}
+	mana_refill_rxoob(dev, rxq, rxbuf_oob, &old_buf);
 
+	/* Unsuccessful refill will have old_buf == NULL.
+	 * In this case, mana_rx_skb() will drop the packet.
+	 */
 	mana_rx_skb(old_buf, oob, rxq);
 
 drop:
@@ -1659,8 +1678,8 @@ static void mana_destroy_rxq(struct mana_port_context *apc,
 
 	mana_deinit_cq(apc, &rxq->rx_cq);
 
-	if (rxq->xdp_save_page)
-		__free_page(rxq->xdp_save_page);
+	if (rxq->xdp_save_va)
+		put_page(virt_to_head_page(rxq->xdp_save_va));
 
 	for (i = 0; i < rxq->num_rx_buf; i++) {
 		rx_oob = &rxq->rx_oobs[i];
@@ -1668,10 +1687,10 @@ static void mana_destroy_rxq(struct mana_port_context *apc,
 		if (!rx_oob->buf_va)
 			continue;
 
-		dma_unmap_page(dev, rx_oob->buf_dma_addr, rxq->datasize,
-			       DMA_FROM_DEVICE);
+		dma_unmap_single(dev, rx_oob->sgl[0].address,
+				 rx_oob->sgl[0].size, DMA_FROM_DEVICE);
 
-		free_page((unsigned long)rx_oob->buf_va);
+		put_page(virt_to_head_page(rx_oob->buf_va));
 		rx_oob->buf_va = NULL;
 	}
 
@@ -1681,6 +1700,26 @@ static void mana_destroy_rxq(struct mana_port_context *apc,
 	kfree(rxq);
 }
 
+static int mana_fill_rx_oob(struct mana_recv_buf_oob *rx_oob, u32 mem_key,
+			    struct mana_rxq *rxq, struct device *dev)
+{
+	dma_addr_t da;
+	void *va;
+
+	va = mana_get_rxfrag(rxq, dev, &da, false);
+
+	if (!va)
+		return -ENOMEM;
+
+	rx_oob->buf_va = va;
+
+	rx_oob->sgl[0].address = da;
+	rx_oob->sgl[0].size = rxq->datasize;
+	rx_oob->sgl[0].mem_key = mem_key;
+
+	return 0;
+}
+
 #define MANA_WQE_HEADER_SIZE 16
 #define MANA_WQE_SGE_SIZE 16
 
@@ -1690,9 +1729,8 @@ static int mana_alloc_rx_wqe(struct mana_port_context *apc,
 	struct gdma_context *gc = apc->ac->gdma_dev->gdma_context;
 	struct mana_recv_buf_oob *rx_oob;
 	struct device *dev = gc->dev;
-	struct page *page;
-	dma_addr_t da;
 	u32 buf_idx;
+	int ret;
 
 	WARN_ON(rxq->datasize == 0 || rxq->datasize > PAGE_SIZE);
 
@@ -1703,25 +1741,12 @@ static int mana_alloc_rx_wqe(struct mana_port_context *apc,
 		rx_oob = &rxq->rx_oobs[buf_idx];
 		memset(rx_oob, 0, sizeof(*rx_oob));
 
-		page = alloc_page(GFP_KERNEL);
-		if (!page)
-			return -ENOMEM;
-
-		da = dma_map_page(dev, page, XDP_PACKET_HEADROOM, rxq->datasize,
-				  DMA_FROM_DEVICE);
-
-		if (dma_mapping_error(dev, da)) {
-			__free_page(page);
-			return -ENOMEM;
-		}
-
-		rx_oob->buf_va = page_to_virt(page);
-		rx_oob->buf_dma_addr = da;
-
 		rx_oob->num_sge = 1;
-		rx_oob->sgl[0].address = rx_oob->buf_dma_addr;
-		rx_oob->sgl[0].size = rxq->datasize;
-		rx_oob->sgl[0].mem_key = apc->ac->gdma_dev->gpa_mkey;
+
+		ret = mana_fill_rx_oob(rx_oob, apc->ac->gdma_dev->gpa_mkey, rxq,
+				       dev);
+		if (ret)
+			return ret;
 
 		rx_oob->wqe_req.sgl = rx_oob->sgl;
 		rx_oob->wqe_req.num_sge = rx_oob->num_sge;
@@ -1780,9 +1805,10 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	rxq->ndev = ndev;
 	rxq->num_rx_buf = RX_BUFFERS_PER_QUEUE;
 	rxq->rxq_idx = rxq_idx;
-	rxq->datasize = ALIGN(MAX_FRAME_SIZE, 64);
 	rxq->rxobj = INVALID_MANA_HANDLE;
 
+	rxq->datasize = ALIGN(ETH_FRAME_LEN, 64);
+
 	err = mana_alloc_rx_wqe(apc, rxq, &rq_size, &cq_size);
 	if (err)
 		goto out;
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index bb11a6535d80..037bcabf6b98 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -36,9 +36,6 @@ enum TRI_STATE {
 
 #define COMP_ENTRY_SIZE 64
 
-#define ADAPTER_MTU_SIZE 1500
-#define MAX_FRAME_SIZE (ADAPTER_MTU_SIZE + 14)
-
 #define RX_BUFFERS_PER_QUEUE 512
 
 #define MAX_SEND_BUFFERS_PER_QUEUE 256
@@ -282,7 +279,6 @@ struct mana_recv_buf_oob {
 	struct gdma_wqe_request wqe_req;
 
 	void *buf_va;
-	dma_addr_t buf_dma_addr;
 
 	/* SGL of the buffer going to be sent has part of the work request. */
 	u32 num_sge;
@@ -322,7 +318,7 @@ struct mana_rxq {
 
 	struct bpf_prog __rcu *bpf_prog;
 	struct xdp_rxq_info xdp_rxq;
-	struct page *xdp_save_page;
+	void *xdp_save_va; /* for reusing */
 	bool xdp_flush;
 	int xdp_rc; /* XDP redirect return code */
 
-- 
cgit v1.2.3-70-g09d2


From 2fbbd712baf1c60996554326728bbdbef5616e12 Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Wed, 12 Apr 2023 14:16:02 -0700
Subject: net: mana: Enable RX path to handle various MTU sizes

Update RX data path to allocate and use RX queue DMA buffers with
proper size based on potentially various MTU sizes.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 38 ++++++++++++++++++++-------
 include/net/mana/mana.h                       |  7 +++++
 2 files changed, 35 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 911954ff84ee..8e7fa6e9c3b5 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -1185,10 +1185,10 @@ static void mana_post_pkt_rxq(struct mana_rxq *rxq)
 	WARN_ON_ONCE(recv_buf_oob->wqe_inf.wqe_size_in_bu != 1);
 }
 
-static struct sk_buff *mana_build_skb(void *buf_va, uint pkt_len,
-				      struct xdp_buff *xdp)
+static struct sk_buff *mana_build_skb(struct mana_rxq *rxq, void *buf_va,
+				      uint pkt_len, struct xdp_buff *xdp)
 {
-	struct sk_buff *skb = napi_build_skb(buf_va, PAGE_SIZE);
+	struct sk_buff *skb = napi_build_skb(buf_va, rxq->alloc_size);
 
 	if (!skb)
 		return NULL;
@@ -1196,11 +1196,12 @@ static struct sk_buff *mana_build_skb(void *buf_va, uint pkt_len,
 	if (xdp->data_hard_start) {
 		skb_reserve(skb, xdp->data - xdp->data_hard_start);
 		skb_put(skb, xdp->data_end - xdp->data);
-	} else {
-		skb_reserve(skb, XDP_PACKET_HEADROOM);
-		skb_put(skb, pkt_len);
+		return skb;
 	}
 
+	skb_reserve(skb, rxq->headroom);
+	skb_put(skb, pkt_len);
+
 	return skb;
 }
 
@@ -1233,7 +1234,7 @@ static void mana_rx_skb(void *buf_va, struct mana_rxcomp_oob *cqe,
 	if (act != XDP_PASS && act != XDP_TX)
 		goto drop_xdp;
 
-	skb = mana_build_skb(buf_va, pkt_len, &xdp);
+	skb = mana_build_skb(rxq, buf_va, pkt_len, &xdp);
 
 	if (!skb)
 		goto drop;
@@ -1301,6 +1302,14 @@ static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 	if (rxq->xdp_save_va) {
 		va = rxq->xdp_save_va;
 		rxq->xdp_save_va = NULL;
+	} else if (rxq->alloc_size > PAGE_SIZE) {
+		if (is_napi)
+			va = napi_alloc_frag(rxq->alloc_size);
+		else
+			va = netdev_alloc_frag(rxq->alloc_size);
+
+		if (!va)
+			return NULL;
 	} else {
 		page = dev_alloc_page();
 		if (!page)
@@ -1309,7 +1318,7 @@ static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 		va = page_to_virt(page);
 	}
 
-	*da = dma_map_single(dev, va + XDP_PACKET_HEADROOM, rxq->datasize,
+	*da = dma_map_single(dev, va + rxq->headroom, rxq->datasize,
 			     DMA_FROM_DEVICE);
 
 	if (dma_mapping_error(dev, *da)) {
@@ -1732,7 +1741,7 @@ static int mana_alloc_rx_wqe(struct mana_port_context *apc,
 	u32 buf_idx;
 	int ret;
 
-	WARN_ON(rxq->datasize == 0 || rxq->datasize > PAGE_SIZE);
+	WARN_ON(rxq->datasize == 0);
 
 	*rxq_size = 0;
 	*cq_size = 0;
@@ -1788,6 +1797,7 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	struct gdma_dev *gd = apc->ac->gdma_dev;
 	struct mana_obj_spec wq_spec;
 	struct mana_obj_spec cq_spec;
+	unsigned int mtu = ndev->mtu;
 	struct gdma_queue_spec spec;
 	struct mana_cq *cq = NULL;
 	struct gdma_context *gc;
@@ -1807,7 +1817,15 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	rxq->rxq_idx = rxq_idx;
 	rxq->rxobj = INVALID_MANA_HANDLE;
 
-	rxq->datasize = ALIGN(ETH_FRAME_LEN, 64);
+	rxq->datasize = ALIGN(mtu + ETH_HLEN, 64);
+
+	if (mtu > MANA_XDP_MTU_MAX) {
+		rxq->alloc_size = mtu + MANA_RXBUF_PAD;
+		rxq->headroom = 0;
+	} else {
+		rxq->alloc_size = mtu + MANA_RXBUF_PAD + XDP_PACKET_HEADROOM;
+		rxq->headroom = XDP_PACKET_HEADROOM;
+	}
 
 	err = mana_alloc_rx_wqe(apc, rxq, &rq_size, &cq_size);
 	if (err)
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index 037bcabf6b98..fee99d704281 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -291,6 +291,11 @@ struct mana_recv_buf_oob {
 	struct gdma_posted_wqe_info wqe_inf;
 };
 
+#define MANA_RXBUF_PAD (SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) \
+			+ ETH_HLEN)
+
+#define MANA_XDP_MTU_MAX (PAGE_SIZE - MANA_RXBUF_PAD - XDP_PACKET_HEADROOM)
+
 struct mana_rxq {
 	struct gdma_queue *gdma_rq;
 	/* Cache the gdma receive queue id */
@@ -300,6 +305,8 @@ struct mana_rxq {
 	u32 rxq_idx;
 
 	u32 datasize;
+	u32 alloc_size;
+	u32 headroom;
 
 	mana_handle_t rxobj;
 
-- 
cgit v1.2.3-70-g09d2


From 80f6215b450eb8e92d8b1f117abf5ecf867f963e Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Wed, 12 Apr 2023 14:16:03 -0700
Subject: net: mana: Add support for jumbo frame

During probe, get the hardware-allowed max MTU by querying the device
configuration. Users can select MTU up to the device limit.
When XDP is in use, limit MTU settings so the buffer size is within
one page. And, when MTU is set to a too large value, XDP is not allowed
to run.
Also, to prevent changing MTU fails, and leaves the NIC in a bad state,
pre-allocate all buffers before starting the change. So in low memory
condition, it will return error, without affecting the NIC.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microsoft/mana/mana_bpf.c |  22 +--
 drivers/net/ethernet/microsoft/mana/mana_en.c  | 217 +++++++++++++++++++++++--
 include/net/mana/gdma.h                        |   4 +
 include/net/mana/mana.h                        |  14 ++
 4 files changed, 233 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_bpf.c b/drivers/net/ethernet/microsoft/mana/mana_bpf.c
index 3caea631229c..23b1521c0df9 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_bpf.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_bpf.c
@@ -133,12 +133,6 @@ out:
 	return act;
 }
 
-static unsigned int mana_xdp_fraglen(unsigned int len)
-{
-	return SKB_DATA_ALIGN(len) +
-	       SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
-}
-
 struct bpf_prog *mana_xdp_get(struct mana_port_context *apc)
 {
 	ASSERT_RTNL();
@@ -179,17 +173,18 @@ static int mana_xdp_set(struct net_device *ndev, struct bpf_prog *prog,
 {
 	struct mana_port_context *apc = netdev_priv(ndev);
 	struct bpf_prog *old_prog;
-	int buf_max;
+	struct gdma_context *gc;
+
+	gc = apc->ac->gdma_dev->gdma_context;
 
 	old_prog = mana_xdp_get(apc);
 
 	if (!old_prog && !prog)
 		return 0;
 
-	buf_max = XDP_PACKET_HEADROOM + mana_xdp_fraglen(ndev->mtu + ETH_HLEN);
-	if (prog && buf_max > PAGE_SIZE) {
-		netdev_err(ndev, "XDP: mtu:%u too large, buf_max:%u\n",
-			   ndev->mtu, buf_max);
+	if (prog && ndev->mtu > MANA_XDP_MTU_MAX) {
+		netdev_err(ndev, "XDP: mtu:%u too large, mtu_max:%lu\n",
+			   ndev->mtu, MANA_XDP_MTU_MAX);
 		NL_SET_ERR_MSG_MOD(extack, "XDP: mtu too large");
 
 		return -EOPNOTSUPP;
@@ -206,6 +201,11 @@ static int mana_xdp_set(struct net_device *ndev, struct bpf_prog *prog,
 	if (apc->port_is_up)
 		mana_chn_setxdp(apc, prog);
 
+	if (prog)
+		ndev->max_mtu = MANA_XDP_MTU_MAX;
+	else
+		ndev->max_mtu = gc->adapter_mtu - ETH_HLEN;
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 8e7fa6e9c3b5..cabecbfa1102 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -427,6 +427,192 @@ static u16 mana_select_queue(struct net_device *ndev, struct sk_buff *skb,
 	return txq;
 }
 
+/* Release pre-allocated RX buffers */
+static void mana_pre_dealloc_rxbufs(struct mana_port_context *mpc)
+{
+	struct device *dev;
+	int i;
+
+	dev = mpc->ac->gdma_dev->gdma_context->dev;
+
+	if (!mpc->rxbufs_pre)
+		goto out1;
+
+	if (!mpc->das_pre)
+		goto out2;
+
+	while (mpc->rxbpre_total) {
+		i = --mpc->rxbpre_total;
+		dma_unmap_single(dev, mpc->das_pre[i], mpc->rxbpre_datasize,
+				 DMA_FROM_DEVICE);
+		put_page(virt_to_head_page(mpc->rxbufs_pre[i]));
+	}
+
+	kfree(mpc->das_pre);
+	mpc->das_pre = NULL;
+
+out2:
+	kfree(mpc->rxbufs_pre);
+	mpc->rxbufs_pre = NULL;
+
+out1:
+	mpc->rxbpre_datasize = 0;
+	mpc->rxbpre_alloc_size = 0;
+	mpc->rxbpre_headroom = 0;
+}
+
+/* Get a buffer from the pre-allocated RX buffers */
+static void *mana_get_rxbuf_pre(struct mana_rxq *rxq, dma_addr_t *da)
+{
+	struct net_device *ndev = rxq->ndev;
+	struct mana_port_context *mpc;
+	void *va;
+
+	mpc = netdev_priv(ndev);
+
+	if (!mpc->rxbufs_pre || !mpc->das_pre || !mpc->rxbpre_total) {
+		netdev_err(ndev, "No RX pre-allocated bufs\n");
+		return NULL;
+	}
+
+	/* Check sizes to catch unexpected coding error */
+	if (mpc->rxbpre_datasize != rxq->datasize) {
+		netdev_err(ndev, "rxbpre_datasize mismatch: %u: %u\n",
+			   mpc->rxbpre_datasize, rxq->datasize);
+		return NULL;
+	}
+
+	if (mpc->rxbpre_alloc_size != rxq->alloc_size) {
+		netdev_err(ndev, "rxbpre_alloc_size mismatch: %u: %u\n",
+			   mpc->rxbpre_alloc_size, rxq->alloc_size);
+		return NULL;
+	}
+
+	if (mpc->rxbpre_headroom != rxq->headroom) {
+		netdev_err(ndev, "rxbpre_headroom mismatch: %u: %u\n",
+			   mpc->rxbpre_headroom, rxq->headroom);
+		return NULL;
+	}
+
+	mpc->rxbpre_total--;
+
+	*da = mpc->das_pre[mpc->rxbpre_total];
+	va = mpc->rxbufs_pre[mpc->rxbpre_total];
+	mpc->rxbufs_pre[mpc->rxbpre_total] = NULL;
+
+	/* Deallocate the array after all buffers are gone */
+	if (!mpc->rxbpre_total)
+		mana_pre_dealloc_rxbufs(mpc);
+
+	return va;
+}
+
+/* Get RX buffer's data size, alloc size, XDP headroom based on MTU */
+static void mana_get_rxbuf_cfg(int mtu, u32 *datasize, u32 *alloc_size,
+			       u32 *headroom)
+{
+	if (mtu > MANA_XDP_MTU_MAX)
+		*headroom = 0; /* no support for XDP */
+	else
+		*headroom = XDP_PACKET_HEADROOM;
+
+	*alloc_size = mtu + MANA_RXBUF_PAD + *headroom;
+
+	*datasize = ALIGN(mtu + ETH_HLEN, MANA_RX_DATA_ALIGN);
+}
+
+static int mana_pre_alloc_rxbufs(struct mana_port_context *mpc, int new_mtu)
+{
+	struct device *dev;
+	struct page *page;
+	dma_addr_t da;
+	int num_rxb;
+	void *va;
+	int i;
+
+	mana_get_rxbuf_cfg(new_mtu, &mpc->rxbpre_datasize,
+			   &mpc->rxbpre_alloc_size, &mpc->rxbpre_headroom);
+
+	dev = mpc->ac->gdma_dev->gdma_context->dev;
+
+	num_rxb = mpc->num_queues * RX_BUFFERS_PER_QUEUE;
+
+	WARN(mpc->rxbufs_pre, "mana rxbufs_pre exists\n");
+	mpc->rxbufs_pre = kmalloc_array(num_rxb, sizeof(void *), GFP_KERNEL);
+	if (!mpc->rxbufs_pre)
+		goto error;
+
+	mpc->das_pre = kmalloc_array(num_rxb, sizeof(dma_addr_t), GFP_KERNEL);
+	if (!mpc->das_pre)
+		goto error;
+
+	mpc->rxbpre_total = 0;
+
+	for (i = 0; i < num_rxb; i++) {
+		if (mpc->rxbpre_alloc_size > PAGE_SIZE) {
+			va = netdev_alloc_frag(mpc->rxbpre_alloc_size);
+			if (!va)
+				goto error;
+		} else {
+			page = dev_alloc_page();
+			if (!page)
+				goto error;
+
+			va = page_to_virt(page);
+		}
+
+		da = dma_map_single(dev, va + mpc->rxbpre_headroom,
+				    mpc->rxbpre_datasize, DMA_FROM_DEVICE);
+
+		if (dma_mapping_error(dev, da)) {
+			put_page(virt_to_head_page(va));
+			goto error;
+		}
+
+		mpc->rxbufs_pre[i] = va;
+		mpc->das_pre[i] = da;
+		mpc->rxbpre_total = i + 1;
+	}
+
+	return 0;
+
+error:
+	mana_pre_dealloc_rxbufs(mpc);
+	return -ENOMEM;
+}
+
+static int mana_change_mtu(struct net_device *ndev, int new_mtu)
+{
+	struct mana_port_context *mpc = netdev_priv(ndev);
+	unsigned int old_mtu = ndev->mtu;
+	int err;
+
+	/* Pre-allocate buffers to prevent failure in mana_attach later */
+	err = mana_pre_alloc_rxbufs(mpc, new_mtu);
+	if (err) {
+		netdev_err(ndev, "Insufficient memory for new MTU\n");
+		return err;
+	}
+
+	err = mana_detach(ndev, false);
+	if (err) {
+		netdev_err(ndev, "mana_detach failed: %d\n", err);
+		goto out;
+	}
+
+	ndev->mtu = new_mtu;
+
+	err = mana_attach(ndev);
+	if (err) {
+		netdev_err(ndev, "mana_attach failed: %d\n", err);
+		ndev->mtu = old_mtu;
+	}
+
+out:
+	mana_pre_dealloc_rxbufs(mpc);
+	return err;
+}
+
 static const struct net_device_ops mana_devops = {
 	.ndo_open		= mana_open,
 	.ndo_stop		= mana_close,
@@ -436,6 +622,7 @@ static const struct net_device_ops mana_devops = {
 	.ndo_get_stats64	= mana_get_stats64,
 	.ndo_bpf		= mana_bpf,
 	.ndo_xdp_xmit		= mana_xdp_xmit,
+	.ndo_change_mtu		= mana_change_mtu,
 };
 
 static void mana_cleanup_port_context(struct mana_port_context *apc)
@@ -625,6 +812,9 @@ static int mana_query_device_cfg(struct mana_context *ac, u32 proto_major_ver,
 
 	mana_gd_init_req_hdr(&req.hdr, MANA_QUERY_DEV_CONFIG,
 			     sizeof(req), sizeof(resp));
+
+	req.hdr.resp.msg_version = GDMA_MESSAGE_V2;
+
 	req.proto_major_ver = proto_major_ver;
 	req.proto_minor_ver = proto_minor_ver;
 	req.proto_micro_ver = proto_micro_ver;
@@ -647,6 +837,11 @@ static int mana_query_device_cfg(struct mana_context *ac, u32 proto_major_ver,
 
 	*max_num_vports = resp.max_num_vports;
 
+	if (resp.hdr.response.msg_version == GDMA_MESSAGE_V2)
+		gc->adapter_mtu = resp.adapter_mtu;
+	else
+		gc->adapter_mtu = ETH_FRAME_LEN;
+
 	return 0;
 }
 
@@ -1712,10 +1907,14 @@ static void mana_destroy_rxq(struct mana_port_context *apc,
 static int mana_fill_rx_oob(struct mana_recv_buf_oob *rx_oob, u32 mem_key,
 			    struct mana_rxq *rxq, struct device *dev)
 {
+	struct mana_port_context *mpc = netdev_priv(rxq->ndev);
 	dma_addr_t da;
 	void *va;
 
-	va = mana_get_rxfrag(rxq, dev, &da, false);
+	if (mpc->rxbufs_pre)
+		va = mana_get_rxbuf_pre(rxq, &da);
+	else
+		va = mana_get_rxfrag(rxq, dev, &da, false);
 
 	if (!va)
 		return -ENOMEM;
@@ -1797,7 +1996,6 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	struct gdma_dev *gd = apc->ac->gdma_dev;
 	struct mana_obj_spec wq_spec;
 	struct mana_obj_spec cq_spec;
-	unsigned int mtu = ndev->mtu;
 	struct gdma_queue_spec spec;
 	struct mana_cq *cq = NULL;
 	struct gdma_context *gc;
@@ -1817,15 +2015,8 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	rxq->rxq_idx = rxq_idx;
 	rxq->rxobj = INVALID_MANA_HANDLE;
 
-	rxq->datasize = ALIGN(mtu + ETH_HLEN, 64);
-
-	if (mtu > MANA_XDP_MTU_MAX) {
-		rxq->alloc_size = mtu + MANA_RXBUF_PAD;
-		rxq->headroom = 0;
-	} else {
-		rxq->alloc_size = mtu + MANA_RXBUF_PAD + XDP_PACKET_HEADROOM;
-		rxq->headroom = XDP_PACKET_HEADROOM;
-	}
+	mana_get_rxbuf_cfg(ndev->mtu, &rxq->datasize, &rxq->alloc_size,
+			   &rxq->headroom);
 
 	err = mana_alloc_rx_wqe(apc, rxq, &rq_size, &cq_size);
 	if (err)
@@ -2238,8 +2429,8 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
 	ndev->netdev_ops = &mana_devops;
 	ndev->ethtool_ops = &mana_ethtool_ops;
 	ndev->mtu = ETH_DATA_LEN;
-	ndev->max_mtu = ndev->mtu;
-	ndev->min_mtu = ndev->mtu;
+	ndev->max_mtu = gc->adapter_mtu - ETH_HLEN;
+	ndev->min_mtu = ETH_MIN_MTU;
 	ndev->needed_headroom = MANA_HEADROOM;
 	ndev->dev_port = port_idx;
 	SET_NETDEV_DEV(ndev, gc->dev);
diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
index 56189e4252da..96c120160f15 100644
--- a/include/net/mana/gdma.h
+++ b/include/net/mana/gdma.h
@@ -145,6 +145,7 @@ struct gdma_general_req {
 }; /* HW DATA */
 
 #define GDMA_MESSAGE_V1 1
+#define GDMA_MESSAGE_V2 2
 
 struct gdma_general_resp {
 	struct gdma_resp_hdr hdr;
@@ -354,6 +355,9 @@ struct gdma_context {
 	struct gdma_resource	msix_resource;
 	struct gdma_irq_context	*irq_contexts;
 
+	/* L2 MTU */
+	u16 adapter_mtu;
+
 	/* This maps a CQ index to the queue structure. */
 	unsigned int		max_num_cqs;
 	struct gdma_queue	**cq_table;
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index fee99d704281..cd386aa7c7cc 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -37,6 +37,7 @@ enum TRI_STATE {
 #define COMP_ENTRY_SIZE 64
 
 #define RX_BUFFERS_PER_QUEUE 512
+#define MANA_RX_DATA_ALIGN 64
 
 #define MAX_SEND_BUFFERS_PER_QUEUE 256
 
@@ -390,6 +391,14 @@ struct mana_port_context {
 	/* This points to an array of num_queues of RQ pointers. */
 	struct mana_rxq **rxqs;
 
+	/* pre-allocated rx buffer array */
+	void **rxbufs_pre;
+	dma_addr_t *das_pre;
+	int rxbpre_total;
+	u32 rxbpre_datasize;
+	u32 rxbpre_alloc_size;
+	u32 rxbpre_headroom;
+
 	struct bpf_prog *bpf_prog;
 
 	/* Create num_queues EQs, SQs, SQ-CQs, RQs and RQ-CQs, respectively. */
@@ -489,6 +498,11 @@ struct mana_query_device_cfg_resp {
 	u16 max_num_vports;
 	u16 reserved;
 	u32 max_num_eqs;
+
+	/* response v2: */
+	u16 adapter_mtu;
+	u16 reserved2;
+	u32 reserved3;
 }; /* HW DATA */
 
 /* Query vPort Configuration */
-- 
cgit v1.2.3-70-g09d2


From eaaa4e92397981c65dae7b9a38425b8fa984d674 Mon Sep 17 00:00:00 2001
From: Arseniy Krasnov <avkrasnov@sberdevices.ru>
Date: Thu, 13 Apr 2023 12:17:19 +0300
Subject: vsock/loopback: don't disable irqs for queue access

This replaces 'skb_queue_tail()' with 'virtio_vsock_skb_queue_tail()'.
The first one uses 'spin_lock_irqsave()', second uses 'spin_lock_bh()'.
There is no need to disable interrupts in the loopback transport as
there is no access to the queue with skbs from interrupt context. Both
virtio and vhost transports work in the same way.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/vmw_vsock/vsock_loopback.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/vmw_vsock/vsock_loopback.c b/net/vmw_vsock/vsock_loopback.c
index e3afc0c866f5..5c6360df1f31 100644
--- a/net/vmw_vsock/vsock_loopback.c
+++ b/net/vmw_vsock/vsock_loopback.c
@@ -31,8 +31,7 @@ static int vsock_loopback_send_pkt(struct sk_buff *skb)
 	struct vsock_loopback *vsock = &the_vsock_loopback;
 	int len = skb->len;
 
-	skb_queue_tail(&vsock->pkt_queue, skb);
-
+	virtio_vsock_skb_queue_tail(&vsock->pkt_queue, skb);
 	queue_work(vsock->workqueue, &vsock->pkt_work);
 
 	return len;
-- 
cgit v1.2.3-70-g09d2


From c39ef2130491b2382f2c81774c825527d4bdf4e2 Mon Sep 17 00:00:00 2001
From: Kevin Brodsky <kevin.brodsky@arm.com>
Date: Thu, 13 Apr 2023 12:47:03 +0100
Subject: net: Ensure ->msg_control_user is used for user buffers

Since commit 1f466e1f15cf ("net: cleanly handle kernel vs user
buffers for ->msg_control"), pointers to user buffers should be
stored in struct msghdr::msg_control_user, instead of the
msg_control field.  Most users of msg_control have already been
converted (where user buffers are involved), but not all of them.

This patch attempts to address the remaining cases. An exception is
made for null checks, as it should be safe to use msg_control
unconditionally for that purpose.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Kevin Brodsky <kevin.brodsky@arm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/compat.c   | 12 ++++++------
 net/core/scm.c |  9 ++++++---
 net/ipv4/tcp.c |  4 ++--
 3 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/net/compat.c b/net/compat.c
index 161b7bea1f62..000a2e054d4c 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -113,7 +113,7 @@ int get_compat_msghdr(struct msghdr *kmsg,
 
 #define CMSG_COMPAT_FIRSTHDR(msg)			\
 	(((msg)->msg_controllen) >= sizeof(struct compat_cmsghdr) ?	\
-	 (struct compat_cmsghdr __user *)((msg)->msg_control) :		\
+	 (struct compat_cmsghdr __user *)((msg)->msg_control_user) :	\
 	 (struct compat_cmsghdr __user *)NULL)
 
 #define CMSG_COMPAT_OK(ucmlen, ucmsg, mhdr) \
@@ -126,7 +126,7 @@ static inline struct compat_cmsghdr __user *cmsg_compat_nxthdr(struct msghdr *ms
 		struct compat_cmsghdr __user *cmsg, int cmsg_len)
 {
 	char __user *ptr = (char __user *)cmsg + CMSG_COMPAT_ALIGN(cmsg_len);
-	if ((unsigned long)(ptr + 1 - (char __user *)msg->msg_control) >
+	if ((unsigned long)(ptr + 1 - (char __user *)msg->msg_control_user) >
 			msg->msg_controllen)
 		return NULL;
 	return (struct compat_cmsghdr __user *)ptr;
@@ -225,7 +225,7 @@ Efault:
 
 int put_cmsg_compat(struct msghdr *kmsg, int level, int type, int len, void *data)
 {
-	struct compat_cmsghdr __user *cm = (struct compat_cmsghdr __user *) kmsg->msg_control;
+	struct compat_cmsghdr __user *cm = (struct compat_cmsghdr __user *) kmsg->msg_control_user;
 	struct compat_cmsghdr cmhdr;
 	struct old_timeval32 ctv;
 	struct old_timespec32 cts[3];
@@ -274,7 +274,7 @@ int put_cmsg_compat(struct msghdr *kmsg, int level, int type, int len, void *dat
 	cmlen = CMSG_COMPAT_SPACE(len);
 	if (kmsg->msg_controllen < cmlen)
 		cmlen = kmsg->msg_controllen;
-	kmsg->msg_control += cmlen;
+	kmsg->msg_control_user += cmlen;
 	kmsg->msg_controllen -= cmlen;
 	return 0;
 }
@@ -289,7 +289,7 @@ static int scm_max_fds_compat(struct msghdr *msg)
 void scm_detach_fds_compat(struct msghdr *msg, struct scm_cookie *scm)
 {
 	struct compat_cmsghdr __user *cm =
-		(struct compat_cmsghdr __user *)msg->msg_control;
+		(struct compat_cmsghdr __user *)msg->msg_control_user;
 	unsigned int o_flags = (msg->msg_flags & MSG_CMSG_CLOEXEC) ? O_CLOEXEC : 0;
 	int fdmax = min_t(int, scm_max_fds_compat(msg), scm->fp->count);
 	int __user *cmsg_data = CMSG_COMPAT_DATA(cm);
@@ -313,7 +313,7 @@ void scm_detach_fds_compat(struct msghdr *msg, struct scm_cookie *scm)
 			cmlen = CMSG_COMPAT_SPACE(i * sizeof(int));
 			if (msg->msg_controllen < cmlen)
 				cmlen = msg->msg_controllen;
-			msg->msg_control += cmlen;
+			msg->msg_control_user += cmlen;
 			msg->msg_controllen -= cmlen;
 		}
 	}
diff --git a/net/core/scm.c b/net/core/scm.c
index acb7d776fa6e..3cd7dd377e53 100644
--- a/net/core/scm.c
+++ b/net/core/scm.c
@@ -250,7 +250,10 @@ int put_cmsg(struct msghdr * msg, int level, int type, int len, void *data)
 	}
 
 	cmlen = min(CMSG_SPACE(len), msg->msg_controllen);
-	msg->msg_control += cmlen;
+	if (msg->msg_control_is_user)
+		msg->msg_control_user += cmlen;
+	else
+		msg->msg_control += cmlen;
 	msg->msg_controllen -= cmlen;
 	return 0;
 
@@ -299,7 +302,7 @@ static int scm_max_fds(struct msghdr *msg)
 void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm)
 {
 	struct cmsghdr __user *cm =
-		(__force struct cmsghdr __user *)msg->msg_control;
+		(__force struct cmsghdr __user *)msg->msg_control_user;
 	unsigned int o_flags = (msg->msg_flags & MSG_CMSG_CLOEXEC) ? O_CLOEXEC : 0;
 	int fdmax = min_t(int, scm_max_fds(msg), scm->fp->count);
 	int __user *cmsg_data = CMSG_USER_DATA(cm);
@@ -332,7 +335,7 @@ void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm)
 			cmlen = CMSG_SPACE(i * sizeof(int));
 			if (msg->msg_controllen < cmlen)
 				cmlen = msg->msg_controllen;
-			msg->msg_control += cmlen;
+			msg->msg_control_user += cmlen;
 			msg->msg_controllen -= cmlen;
 		}
 	}
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index fd68d49490f2..20db115c38c4 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2165,7 +2165,7 @@ static void tcp_zc_finalize_rx_tstamp(struct sock *sk,
 	struct msghdr cmsg_dummy;
 
 	msg_control_addr = (unsigned long)zc->msg_control;
-	cmsg_dummy.msg_control = (void *)msg_control_addr;
+	cmsg_dummy.msg_control_user = (void __user *)msg_control_addr;
 	cmsg_dummy.msg_controllen =
 		(__kernel_size_t)zc->msg_controllen;
 	cmsg_dummy.msg_flags = in_compat_syscall()
@@ -2176,7 +2176,7 @@ static void tcp_zc_finalize_rx_tstamp(struct sock *sk,
 	    zc->msg_controllen == cmsg_dummy.msg_controllen) {
 		tcp_recv_timestamp(&cmsg_dummy, sk, tss);
 		zc->msg_control = (__u64)
-			((uintptr_t)cmsg_dummy.msg_control);
+			((uintptr_t)cmsg_dummy.msg_control_user);
 		zc->msg_controllen =
 			(__u64)cmsg_dummy.msg_controllen;
 		zc->msg_flags = (__u32)cmsg_dummy.msg_flags;
-- 
cgit v1.2.3-70-g09d2


From 60daf8d40b80ccbd593930235aea9ee82ea8dbc2 Mon Sep 17 00:00:00 2001
From: Kevin Brodsky <kevin.brodsky@arm.com>
Date: Thu, 13 Apr 2023 12:47:04 +0100
Subject: net/compat: Update msg_control_is_user when setting a kernel pointer

cmsghdr_from_user_compat_to_kern() is an unusual case w.r.t. how
the kmsg->msg_control* fields are used. The input struct msghdr
holds a pointer to a user buffer, i.e. ksmg->msg_control_user is
active. However, upon success, a kernel pointer is stored in
kmsg->msg_control. kmsg->msg_control_is_user should therefore be
updated accordingly.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Kevin Brodsky <kevin.brodsky@arm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/compat.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/compat.c b/net/compat.c
index 000a2e054d4c..6564720f32b7 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -211,6 +211,7 @@ int cmsghdr_from_user_compat_to_kern(struct msghdr *kmsg, struct sock *sk,
 		goto Einval;
 
 	/* Ok, looks like we made it.  Hook it up and return success. */
+	kmsg->msg_control_is_user = false;
 	kmsg->msg_control = kcmsg_base;
 	kmsg->msg_controllen = kcmlen;
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From b6d85cf5bd1433c5dd6bf6bb3a176537184c630c Mon Sep 17 00:00:00 2001
From: Kevin Brodsky <kevin.brodsky@arm.com>
Date: Thu, 13 Apr 2023 12:47:05 +0100
Subject: net/ipv6: Initialise msg_control_is_user

do_ipv6_setsockopt() makes use of struct msghdr::msg_control in the
IPV6_2292PKTOPTIONS case. Make sure to initialise
msg_control_is_user accordingly.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Kevin Brodsky <kevin.brodsky@arm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ipv6_sockglue.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 2917dd8d198c..ae818ff46224 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -716,6 +716,7 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 			goto done;
 
 		msg.msg_controllen = optlen;
+		msg.msg_control_is_user = false;
 		msg.msg_control = (void *)(opt+1);
 		ipc6.opt = opt;
 
-- 
cgit v1.2.3-70-g09d2


From 5e31b3df86ec6fbb925eee77fe2c450099c61dff Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Thu, 13 Apr 2023 21:40:21 +0300
Subject: wifi: iwlwifi: dbg: print pc register data once fw dump occurred

Add debug print for different FW program counter details of
different CPU. Program counter pc details will be read from
TLV during init.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.862790d336a9.I34e2ea05a79e8b2552f7f221bacf3af0166cb9c0@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dump.c   | 11 +++++++++++
 drivers/net/wireless/intel/iwlwifi/fw/file.h   |  1 +
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c   |  9 +++++++++
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h | 16 ++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c    | 10 ++++++++++
 5 files changed, 47 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dump.c b/drivers/net/wireless/intel/iwlwifi/fw/dump.c
index 59ed321bcc27..f86f7b4baa18 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dump.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dump.c
@@ -484,6 +484,9 @@ static void iwl_fwrt_dump_fseq_regs(struct iwl_fw_runtime *fwrt)
 
 void iwl_fwrt_dump_error_logs(struct iwl_fw_runtime *fwrt)
 {
+	struct iwl_pc_data *pc_data;
+	u8 count;
+
 	if (!test_bit(STATUS_DEVICE_ENABLED, &fwrt->trans->status)) {
 		IWL_ERR(fwrt,
 			"DEVICE_ENABLED bit is not set. Aborting dump.\n");
@@ -502,6 +505,14 @@ void iwl_fwrt_dump_error_logs(struct iwl_fw_runtime *fwrt)
 		iwl_fwrt_dump_rcm_error_log(fwrt, 1);
 	iwl_fwrt_dump_iml_error_log(fwrt);
 	iwl_fwrt_dump_fseq_regs(fwrt);
+	if (fwrt->trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_22000) {
+		pc_data = fwrt->trans->dbg.pc_data;
+		for (count = 0; count < fwrt->trans->dbg.num_pc;
+		     count++, pc_data++)
+			IWL_ERR(fwrt, "%s: 0x%x\n",
+				pc_data->pc_name,
+				pc_data->pc_address);
+	}
 
 	if (fwrt->trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) {
 		u32 scratch = iwl_read32(fwrt->trans, CSR_FUNC_SCRATCH);
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index f2eefca6e260..a0260a1fa099 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -101,6 +101,7 @@ enum iwl_ucode_tlv_type {
 
 	IWL_UCODE_TLV_SEC_TABLE_ADDR		= 66,
 	IWL_UCODE_TLV_D3_KEK_KCK_ADDR		= 67,
+	IWL_UCODE_TLV_CURRENT_PC		= 68,
 
 	IWL_UCODE_TLV_FW_NUM_STATIONS		= IWL_UCODE_TLV_CONST_BASE + 0,
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 4c977ba9cd85..75d52e6cf988 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -127,6 +127,7 @@ static void iwl_dealloc_ucode(struct iwl_drv *drv)
 	kfree(drv->fw.iml);
 	kfree(drv->fw.ucode_capa.cmd_versions);
 	kfree(drv->fw.phy_integration_ver);
+	kfree(drv->trans->dbg.pc_data);
 
 	for (i = 0; i < IWL_UCODE_TYPE_MAX; i++)
 		iwl_free_fw_img(drv, drv->fw.img + i);
@@ -1232,6 +1233,14 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 			iwl_drv_set_dump_exclude(drv, tlv_type,
 						 tlv_data, tlv_len);
 			break;
+		case IWL_UCODE_TLV_CURRENT_PC:
+			if (tlv_len < sizeof(struct iwl_pc_data))
+				goto invalid_tlv_len;
+			drv->trans->dbg.num_pc =
+				tlv_len / sizeof(struct iwl_pc_data);
+			drv->trans->dbg.pc_data =
+				kmemdup(tlv_data, tlv_len, GFP_KERNEL);
+			break;
 		default:
 			IWL_DEBUG_INFO(drv, "unknown TLV: %d\n", tlv_type);
 			break;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index dd277a4fa8dd..74ff4788dc83 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -748,6 +748,18 @@ struct iwl_imr_data {
 	__le64 imr_base_addr;
 };
 
+#define IWL_TRANS_CURRENT_PC_NAME_MAX_BYTES      32
+
+/**
+ * struct iwl_pc_data - program counter details
+ * @pc_name: cpu name
+ * @pc_address: cpu program counter
+ */
+struct iwl_pc_data {
+	u8  pc_name[IWL_TRANS_CURRENT_PC_NAME_MAX_BYTES];
+	u32 pc_address;
+};
+
 /**
  * struct iwl_trans_debug - transport debug related data
  *
@@ -777,6 +789,8 @@ struct iwl_imr_data {
  * @ucode_preset: preset based on ucode
  * @dump_file_name_ext: dump file name extension
  * @dump_file_name_ext_valid: dump file name extension if valid or not
+ * @num_pc: number of program counter for cpu
+ * @pc_data: details of the program counter
  */
 struct iwl_trans_debug {
 	u8 n_dest_reg;
@@ -817,6 +831,8 @@ struct iwl_trans_debug {
 	struct iwl_imr_data imr_data;
 	u8 dump_file_name_ext[IWL_FW_INI_MAX_NAME];
 	bool dump_file_name_ext_valid;
+	u32 num_pc;
+	struct iwl_pc_data *pc_data;
 };
 
 struct iwl_dma_ptr {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 521eded39028..3f5cc6bf3f1c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -321,6 +321,8 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 	static const u16 alive_cmd[] = { UCODE_ALIVE_NTFY };
 	bool run_in_rfkill =
 		ucode_type == IWL_UCODE_INIT || iwl_mvm_has_unified_ucode(mvm);
+	u8 count;
+	struct iwl_pc_data *pc_data;
 
 	if (ucode_type == IWL_UCODE_REGULAR &&
 	    iwl_fw_dbg_conf_usniffer(mvm->fw, FW_DBG_START_FROM_ALIVE) &&
@@ -393,6 +395,14 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 
 		/* LMAC/UMAC PC info */
 		if (trans->trans_cfg->device_family >=
+					IWL_DEVICE_FAMILY_22000) {
+			pc_data = trans->dbg.pc_data;
+			for (count = 0; count < trans->dbg.num_pc;
+			     count++, pc_data++)
+				IWL_ERR(mvm, "%s: 0x%x\n",
+					pc_data->pc_name,
+					pc_data->pc_address);
+		} else if (trans->trans_cfg->device_family >=
 					IWL_DEVICE_FAMILY_9000) {
 			IWL_ERR(mvm, "UMAC PC: 0x%x\n",
 				iwl_read_umac_prph(trans,
-- 
cgit v1.2.3-70-g09d2


From 8e5a26360cbe29b896b6758518280d41c3704d43 Mon Sep 17 00:00:00 2001
From: Ayala Beker <ayala.beker@intel.com>
Date: Thu, 13 Apr 2023 21:40:22 +0300
Subject: wifi: iwlwifi: mvm: don't drop unencrypted MCAST frames

MCAST frames are filtered out by the driver if we are not
authorized yet.
Fix it to filter out only protected frames.

Fixes: 147eb05f24e6 ("iwlwifi: mvm: always tell the firmware to accept MCAST frames in BSS")
Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.9cedcc27db60.I8fb7057981392660da482dd215e85c15946d3f4b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index e08dca8d2baa..b38b24246675 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -383,9 +383,10 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi,
 		 * Don't even try to decrypt a MCAST frame that was received
 		 * before the managed vif is authorized, we'd fail anyway.
 		 */
-		if (vif->type == NL80211_IFTYPE_STATION &&
+		if (is_multicast_ether_addr(hdr->addr1) &&
+		    vif->type == NL80211_IFTYPE_STATION &&
 		    !mvmvif->authorized &&
-		    is_multicast_ether_addr(hdr->addr1)) {
+		    ieee80211_has_protected(hdr->frame_control)) {
 			IWL_DEBUG_DROP(mvm, "MCAST before the vif is authorized\n");
 			kfree_skb(skb);
 			rcu_read_unlock();
-- 
cgit v1.2.3-70-g09d2


From 1843676ad075e6dd24a7a201171382c810e84b84 Mon Sep 17 00:00:00 2001
From: Alon Giladi <alon.giladi@intel.com>
Date: Thu, 13 Apr 2023 21:40:23 +0300
Subject: wifi: iwlwifi: acpi: support modules with high antenna gain

Starting from version 4 of ANT_GAIN_CMD verification of ppag table
values is done by the FW. Driver has to send the complete table as
it appears in the BIOS. Make this change.

Signed-off-by: Alon Giladi <alon.giladi@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.a965365e2336.I1081f00ae309d9225d0529fa2da6ba7e34a8bd4f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/acpi.c    | 41 ++++++++++++++++---------
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h |  1 +
 2 files changed, 28 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c
index a02e5a67b706..5f4a51310add 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c
@@ -1006,8 +1006,10 @@ int iwl_acpi_get_ppag_table(struct iwl_fw_runtime *fwrt)
 	union acpi_object *wifi_pkg, *data, *flags;
 	int i, j, ret, tbl_rev, num_sub_bands = 0;
 	int idx = 2;
+	u8 cmd_ver;
 
 	fwrt->ppag_flags = 0;
+	fwrt->ppag_table_valid = false;
 
 	data = iwl_acpi_get_object(fwrt->dev, ACPI_PPAG_METHOD);
 	if (IS_ERR(data))
@@ -1054,8 +1056,15 @@ read_table:
 	}
 
 	fwrt->ppag_flags = flags->integer.value & ACPI_PPAG_MASK;
-
-	if (!fwrt->ppag_flags) {
+	cmd_ver = iwl_fw_lookup_cmd_ver(fwrt->fw,
+					WIDE_ID(PHY_OPS_GROUP,
+						PER_PLATFORM_ANT_GAIN_CMD),
+					IWL_FW_CMD_VER_UNKNOWN);
+	if (cmd_ver == IWL_FW_CMD_VER_UNKNOWN) {
+		ret = -EINVAL;
+		goto out_free;
+	}
+	if (!fwrt->ppag_flags && cmd_ver <= 3) {
 		ret = 0;
 		goto out_free;
 	}
@@ -1076,21 +1085,22 @@ read_table:
 			}
 
 			fwrt->ppag_chains[i].subbands[j] = ent->integer.value;
-
+			/* from ver 4 the fw deals with out of range values */
+			if (cmd_ver >= 4)
+				continue;
 			if ((j == 0 &&
 				(fwrt->ppag_chains[i].subbands[j] > ACPI_PPAG_MAX_LB ||
 				 fwrt->ppag_chains[i].subbands[j] < ACPI_PPAG_MIN_LB)) ||
 				(j != 0 &&
 				(fwrt->ppag_chains[i].subbands[j] > ACPI_PPAG_MAX_HB ||
 				fwrt->ppag_chains[i].subbands[j] < ACPI_PPAG_MIN_HB))) {
-					fwrt->ppag_flags = 0;
 					ret = -EINVAL;
 					goto out_free;
 				}
 		}
 	}
 
-
+	fwrt->ppag_table_valid = true;
 	ret = 0;
 
 out_free:
@@ -1115,19 +1125,22 @@ int iwl_read_ppag_table(struct iwl_fw_runtime *fwrt, union iwl_ppag_table_cmd *c
                 IWL_DEBUG_RADIO(fwrt,
                                 "PPAG capability not supported by FW, command not sent.\n");
                 return -EINVAL;
-        }
-        if (!fwrt->ppag_flags) {
-                IWL_DEBUG_RADIO(fwrt, "PPAG not enabled, command not sent.\n");
-                return -EINVAL;
-        }
+	}
+
+	cmd_ver = iwl_fw_lookup_cmd_ver(fwrt->fw,
+					WIDE_ID(PHY_OPS_GROUP,
+						PER_PLATFORM_ANT_GAIN_CMD),
+					IWL_FW_CMD_VER_UNKNOWN);
+	if (!fwrt->ppag_table_valid || (cmd_ver <= 3 && !fwrt->ppag_flags)) {
+		IWL_DEBUG_RADIO(fwrt, "PPAG not enabled, command not sent.\n");
+		return -EINVAL;
+	}
 
         /* The 'flags' field is the same in v1 and in v2 so we can just
          * use v1 to access it.
          */
         cmd->v1.flags = cpu_to_le32(fwrt->ppag_flags);
-        cmd_ver = iwl_fw_lookup_cmd_ver(fwrt->fw,
-                                        WIDE_ID(PHY_OPS_GROUP, PER_PLATFORM_ANT_GAIN_CMD),
-                                        IWL_FW_CMD_VER_UNKNOWN);
+
 	if (cmd_ver == 1) {
                 num_sub_bands = IWL_NUM_SUB_BANDS_V1;
                 gain = cmd->v1.gain[0];
@@ -1138,7 +1151,7 @@ int iwl_read_ppag_table(struct iwl_fw_runtime *fwrt, union iwl_ppag_table_cmd *c
                                         fwrt->ppag_ver);
                         cmd->v1.flags &= cpu_to_le32(IWL_PPAG_ETSI_MASK);
 		}
-	} else if (cmd_ver == 2 || cmd_ver == 3) {
+	} else if (cmd_ver >= 2 && cmd_ver <= 4) {
                 num_sub_bands = IWL_NUM_SUB_BANDS_V2;
                 gain = cmd->v2.gain[0];
                 *cmd_size = sizeof(cmd->v2);
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index a59cf4d9567c..df689a9b7e2c 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -165,6 +165,7 @@ struct iwl_fw_runtime {
 	struct iwl_ppag_chain ppag_chains[IWL_NUM_CHAIN_LIMITS];
 	u32 ppag_flags;
 	u32 ppag_ver;
+	bool ppag_table_valid;
 	struct iwl_sar_offset_mapping_cmd sgom_table;
 	bool sgom_enabled;
 	u8 reduced_power_flags;
-- 
cgit v1.2.3-70-g09d2


From 72c20e60966cb8429875a58ddbb99110f9252b9d Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Thu, 13 Apr 2023 21:40:24 +0300
Subject: wifi: iwlwifi: mvm: fix shift-out-of-bounds

The aux queue is initialized to IWL_MVM_INVALID_QUEUE. This is
later used for a bitmask of the queue, which results in a
shift-out-of-bounds. Fix it.

Fixes: b85f7ebb2497 ("wifi: iwlwifi: mvm: avoid UB shift of snif_queue")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.c41a33c32898.Idc15f9eed005345a4137c28ef62efd80a405fad0@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 6b2690534f59..3e5bed885bd9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2231,11 +2231,13 @@ static int iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx,
 int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
 {
 	int ret;
+	u32 qmask = mvm->aux_queue == IWL_MVM_INVALID_QUEUE ? 0 :
+		BIT(mvm->aux_queue);
 
 	lockdep_assert_held(&mvm->mutex);
 
 	/* Allocate aux station and assign to it the aux queue */
-	ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue),
+	ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, qmask,
 				       NL80211_IFTYPE_UNSPECIFIED,
 				       IWL_STA_AUX_ACTIVITY);
 	if (ret)
-- 
cgit v1.2.3-70-g09d2


From 784d4a4258b6dfc953b9f109cfb203535e0b0796 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 21:40:25 +0300
Subject: wifi: iwlwifi: mvm: request limiting to 8 MSDUs per A-MSDU

Due to hardware limitations when splitting up A-MSDUs, it
cannot handle an arbitrary number of MSDUs per A-MSDU. Set
the bits to ask transmitters to send only 8 MSDUs in one
A-MSDU.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.e6c518f2724f.I15c13d13b38289edbcd64f67d57cf18f6026457a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index fc15636beec0..ac5c4799fa7e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -227,14 +227,18 @@ int iwl_mvm_init_fw_regd(struct iwl_mvm *mvm)
 static const u8 he_if_types_ext_capa_sta[] = {
 	 [0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
 	 [2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT,
-	 [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+	 [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF |
+	       WLAN_EXT_CAPA8_MAX_MSDU_IN_AMSDU_LSB,
+	 [8] = WLAN_EXT_CAPA9_MAX_MSDU_IN_AMSDU_MSB,
 };
 
 static const u8 tm_if_types_ext_capa_sta[] = {
 	 [0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
 	 [2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT |
 	       WLAN_EXT_CAPA3_TIMING_MEASUREMENT_SUPPORT,
-	 [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+	 [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF |
+	       WLAN_EXT_CAPA8_MAX_MSDU_IN_AMSDU_LSB,
+	 [8] = WLAN_EXT_CAPA9_MAX_MSDU_IN_AMSDU_MSB,
 	 [9] = WLAN_EXT_CAPA10_TWT_REQUESTER_SUPPORT,
 };
 
-- 
cgit v1.2.3-70-g09d2


From 26c680b7319dba5da2ff743a6b18c9e4bba0ba04 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Thu, 13 Apr 2023 21:40:26 +0300
Subject: wifi: iwlwifi: mvm: make HLTK configuration for PASN station optional

When secure LTF is not used, the HLTK is not derived during the PASN
authentication. Make the HLTK optional when adding a PASN station.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.a57f7f3b59bd.Ifa88afb51a2516156153ac321d81556e295036c2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 25 +++++++++++++++++++---
 .../net/wireless/intel/iwlwifi/mvm/ftm-responder.c | 19 ++++++++++++----
 2 files changed, 37 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 379da4bec5dd..3963a0d4ed04 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -25,6 +25,10 @@ struct iwl_mvm_smooth_entry {
 	u64 host_time;
 };
 
+enum iwl_mvm_pasn_flags {
+	IWL_MVM_PASN_FLAG_HAS_HLTK = BIT(0),
+};
+
 struct iwl_mvm_ftm_pasn_entry {
 	struct list_head list;
 	u8 addr[ETH_ALEN];
@@ -33,6 +37,7 @@ struct iwl_mvm_ftm_pasn_entry {
 	u8 cipher;
 	u8 tx_pn[IEEE80211_CCMP_PN_LEN];
 	u8 rx_pn[IEEE80211_CCMP_PN_LEN];
+	u32 flags;
 };
 
 int iwl_mvm_ftm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
@@ -79,14 +84,24 @@ int iwl_mvm_ftm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		rcu_read_unlock();
 	}
 
-	if (tk_len != expected_tk_len || hltk_len != sizeof(pasn->hltk)) {
+	if (tk_len != expected_tk_len ||
+	    (hltk_len && hltk_len != sizeof(pasn->hltk))) {
 		IWL_ERR(mvm, "Invalid key length: tk_len=%u hltk_len=%u\n",
 			tk_len, hltk_len);
 		goto out;
 	}
 
+	if (!expected_tk_len && !hltk_len) {
+		IWL_ERR(mvm, "TK and HLTK not set\n");
+		goto out;
+	}
+
 	memcpy(pasn->addr, addr, sizeof(pasn->addr));
-	memcpy(pasn->hltk, hltk, sizeof(pasn->hltk));
+
+	if (hltk_len) {
+		memcpy(pasn->hltk, hltk, sizeof(pasn->hltk));
+		pasn->flags |= IWL_MVM_PASN_FLAG_HAS_HLTK;
+	}
 
 	if (tk && tk_len)
 		memcpy(pasn->tk, tk, sizeof(pasn->tk));
@@ -691,7 +706,11 @@ iwl_mvm_ftm_set_secured_ranging(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			continue;
 
 		target->cipher = entry->cipher;
-		memcpy(target->hltk, entry->hltk, sizeof(target->hltk));
+
+		if (entry->flags & IWL_MVM_PASN_FLAG_HAS_HLTK)
+			memcpy(target->hltk, entry->hltk, sizeof(target->hltk));
+		else
+			memset(target->hltk, 0, sizeof(target->hltk));
 
 		if (vif->cfg.assoc &&
 		    !memcmp(vif->bss_conf.bssid, target->bssid,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
index c37d793d6a4a..1b6fb73ddfc7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
@@ -317,6 +317,8 @@ int iwl_mvm_ftm_respoder_add_pasn_sta(struct iwl_mvm *mvm,
 		.addr = addr,
 		.hltk = hltk,
 	};
+	struct iwl_mvm_pasn_hltk_data *hltk_data_ptr = NULL;
+
 	u8 cmd_ver = iwl_fw_lookup_cmd_ver(mvm->fw,
 					   WIDE_ID(LOCATION_GROUP, TOF_RESPONDER_DYN_CONFIG_CMD),
 					   2);
@@ -328,12 +330,21 @@ int iwl_mvm_ftm_respoder_add_pasn_sta(struct iwl_mvm *mvm,
 		return -ENOTSUPP;
 	}
 
-	hltk_data.cipher = iwl_mvm_cipher_to_location_cipher(cipher);
-	if (hltk_data.cipher == IWL_LOCATION_CIPHER_INVALID) {
-		IWL_ERR(mvm, "invalid cipher: %u\n", cipher);
+	if ((!hltk || !hltk_len) && (!tk || !tk_len)) {
+		IWL_ERR(mvm, "TK and HLTK not set\n");
 		return -EINVAL;
 	}
 
+	if (hltk && hltk_len) {
+		hltk_data.cipher = iwl_mvm_cipher_to_location_cipher(cipher);
+		if (hltk_data.cipher == IWL_LOCATION_CIPHER_INVALID) {
+			IWL_ERR(mvm, "invalid cipher: %u\n", cipher);
+			return -EINVAL;
+		}
+
+		hltk_data_ptr = &hltk_data;
+	}
+
 	if (tk && tk_len) {
 		sta = kzalloc(sizeof(*sta), GFP_KERNEL);
 		if (!sta)
@@ -350,7 +361,7 @@ int iwl_mvm_ftm_respoder_add_pasn_sta(struct iwl_mvm *mvm,
 		list_add_tail(&sta->list, &mvm->resp_pasn_list);
 	}
 
-	ret = iwl_mvm_ftm_responder_dyn_cfg_v3(mvm, vif, NULL, &hltk_data);
+	ret = iwl_mvm_ftm_responder_dyn_cfg_v3(mvm, vif, NULL, hltk_data_ptr);
 	if (ret && sta)
 		iwl_mvm_resp_del_pasn_sta(mvm, vif, sta);
 
-- 
cgit v1.2.3-70-g09d2


From 905d50ddbc83bef0d7f3386e7f3472b0324b405b Mon Sep 17 00:00:00 2001
From: Haim Dreyfuss <haim.dreyfuss@intel.com>
Date: Thu, 13 Apr 2023 21:40:27 +0300
Subject: wifi: iwlwifi: mvm: support wowlan info notification version 2

As part of version 2 we don't need to have wake_packet_bufsize
and wake_packet_length. The first one is already calculated by the driver,
the latter is sent as part of the wake packet notification.

Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.3b53213b10d4.Ibf2f15aca614def2d262dd267d1aad65931b58f1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/d3.h | 37 ++++++++++++++++++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c    | 35 +++++++++++++++++++++++-
 2 files changed, 69 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h b/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h
index df0833890e55..8a613e150a02 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h
@@ -767,7 +767,7 @@ struct iwl_wowlan_status_v12 {
 } __packed; /* WOWLAN_STATUSES_RSP_API_S_VER_12 */
 
 /**
- * struct iwl_wowlan_info_notif - WoWLAN information notification
+ * struct iwl_wowlan_info_notif_v1 - WoWLAN information notification
  * @gtk: GTK data
  * @igtk: IGTK data
  * @replay_ctr: GTK rekey replay counter
@@ -785,7 +785,7 @@ struct iwl_wowlan_status_v12 {
  * @station_id: station id
  * @reserved2: reserved
  */
-struct iwl_wowlan_info_notif {
+struct iwl_wowlan_info_notif_v1 {
 	struct iwl_wowlan_gtk_status_v3 gtk[WOWLAN_GTK_KEYS_NUM];
 	struct iwl_wowlan_igtk_status igtk[WOWLAN_IGTK_KEYS_NUM];
 	__le64 replay_ctr;
@@ -803,6 +803,39 @@ struct iwl_wowlan_info_notif {
 	u8 reserved2[2];
 } __packed; /* WOWLAN_INFO_NTFY_API_S_VER_1 */
 
+/**
+ * struct iwl_wowlan_info_notif - WoWLAN information notification
+ * @gtk: GTK data
+ * @igtk: IGTK data
+ * @replay_ctr: GTK rekey replay counter
+ * @pattern_number: number of the matched patterns
+ * @reserved1: reserved
+ * @qos_seq_ctr: QoS sequence counters to use next
+ * @wakeup_reasons: wakeup reasons, see &enum iwl_wowlan_wakeup_reason
+ * @num_of_gtk_rekeys: number of GTK rekeys
+ * @transmitted_ndps: number of transmitted neighbor discovery packets
+ * @received_beacons: number of received beacons
+ * @tid_tear_down: bit mask of tids whose BA sessions were closed
+ *	in suspend state
+ * @station_id: station id
+ * @reserved2: reserved
+ */
+struct iwl_wowlan_info_notif {
+	struct iwl_wowlan_gtk_status_v3 gtk[WOWLAN_GTK_KEYS_NUM];
+	struct iwl_wowlan_igtk_status igtk[WOWLAN_IGTK_KEYS_NUM];
+	__le64 replay_ctr;
+	__le16 pattern_number;
+	__le16 reserved1;
+	__le16 qos_seq_ctr[8];
+	__le32 wakeup_reasons;
+	__le32 num_of_gtk_rekeys;
+	__le32 transmitted_ndps;
+	__le32 received_beacons;
+	u8 tid_tear_down;
+	u8 station_id;
+	u8 reserved2[2];
+} __packed; /* WOWLAN_INFO_NTFY_API_S_VER_2 */
+
 /**
  * struct iwl_wowlan_wake_pkt_notif - WoWLAN wake packet notification
  * @wake_packet_length: wakeup packet length
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index 40adf789c8fc..bf80b1d093a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -2017,6 +2017,12 @@ static void iwl_mvm_parse_wowlan_info_notif(struct iwl_mvm *mvm,
 {
 	u32 i;
 
+	if (!data) {
+		IWL_ERR(mvm, "iwl_wowlan_info_notif data is NULL\n");
+		status = NULL;
+		return;
+	}
+
 	if (len < sizeof(*data)) {
 		IWL_ERR(mvm, "Invalid WoWLAN info notification!\n");
 		status = NULL;
@@ -2705,10 +2711,33 @@ static bool iwl_mvm_wait_d3_notif(struct iwl_notif_wait_data *notif_wait,
 	struct iwl_d3_data *d3_data = data;
 	u32 len;
 	int ret;
+	int wowlan_info_ver = iwl_fw_lookup_notif_ver(mvm->fw,
+						      PROT_OFFLOAD_GROUP,
+						      WOWLAN_INFO_NOTIFICATION,
+						      IWL_FW_CMD_VER_UNKNOWN);
+
 
 	switch (WIDE_ID(pkt->hdr.group_id, pkt->hdr.cmd)) {
 	case WIDE_ID(PROT_OFFLOAD_GROUP, WOWLAN_INFO_NOTIFICATION): {
-		struct iwl_wowlan_info_notif *notif = (void *)pkt->data;
+		struct iwl_wowlan_info_notif *notif;
+
+		if (wowlan_info_ver < 2) {
+			struct iwl_wowlan_info_notif_v1 *notif_v1 = (void *)pkt->data;
+
+			notif = kmemdup(notif_v1,
+					offsetofend(struct iwl_wowlan_info_notif,
+						    received_beacons),
+					GFP_ATOMIC);
+
+			if (!notif)
+				return false;
+
+			notif->tid_tear_down = notif_v1->tid_tear_down;
+			notif->station_id = notif_v1->station_id;
+
+		} else {
+			notif = (void *)pkt->data;
+		}
 
 		if (d3_data->notif_received & IWL_D3_NOTIF_WOWLAN_INFO) {
 			/* We might get two notifications due to dual bss */
@@ -2721,6 +2750,10 @@ static bool iwl_mvm_wait_d3_notif(struct iwl_notif_wait_data *notif_wait,
 		len = iwl_rx_packet_payload_len(pkt);
 		iwl_mvm_parse_wowlan_info_notif(mvm, notif, d3_data->status,
 						len);
+
+		if (wowlan_info_ver < 2)
+			kfree(notif);
+
 		if (d3_data->status &&
 		    d3_data->status->wakeup_reasons & IWL_WOWLAN_WAKEUP_REASON_HAS_WAKEUP_PKT)
 			/* We are supposed to get also wake packet notif */
-- 
cgit v1.2.3-70-g09d2


From fb40cd9d914c8a1b4acc56edaa96046c5f4f1669 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Thu, 13 Apr 2023 21:40:28 +0300
Subject: wifi: iwlwifi: mvm: cleanup beacon_inject_active during hw restart

beacon_inject_active turns true and false via debugfs when we inject a
beacon. During the time of the beacon injection we can't configure the
FW with a beacon cmd. If we have a hw restart during the beacon injection
then in the recovery flow we will not be able to configure the beacon
cmd to the FW. Fix this by cleaning up this variable after an hw restart.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.b021bce5e162.Ia5a0a0b5d8734f63077ceaac936176a345f8d3d1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index ac5c4799fa7e..d6f29aa2a66f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1065,6 +1065,9 @@ static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm)
 	mvm->rx_ba_sessions = 0;
 	mvm->fwrt.dump.conf = FW_DBG_INVALID;
 	mvm->monitor_on = false;
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+	mvm->beacon_inject_active = false;
+#endif
 
 	/* keep statistics ticking */
 	iwl_mvm_accu_radio_stats(mvm);
-- 
cgit v1.2.3-70-g09d2


From d48304326e5a34cc8be236d084dc2003d11f4c16 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 21:40:29 +0300
Subject: wifi: iwlwifi: mvm: add DSM_FUNC_ENABLE_6E value to debugfs

In order to simplify checking that this value was set up
correctly, add it to debugfs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.c9a787710e1e.If509b2e510e22d62f3eefb2c54ff1645e331dc23@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 527daaf46f96..affaed4d13fa 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -340,6 +340,26 @@ static ssize_t iwl_dbgfs_sar_geo_profile_read(struct file *file,
 
 	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
 }
+
+static ssize_t iwl_dbgfs_wifi_6e_enable_read(struct file *file,
+					     char __user *user_buf,
+					     size_t count, loff_t *ppos)
+{
+	struct iwl_mvm *mvm = file->private_data;
+	int err, pos;
+	char buf[12];
+	u32 value;
+
+	err = iwl_acpi_get_dsm_u32(mvm->fwrt.dev, 0,
+				   DSM_FUNC_ENABLE_6E,
+				   &iwl_guid, &value);
+	if (err)
+		return err;
+
+	pos = sprintf(buf, "0x%08x\n", value);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
+}
 #endif
 
 static ssize_t iwl_dbgfs_stations_read(struct file *file, char __user *user_buf,
@@ -1898,6 +1918,7 @@ MVM_DEBUGFS_READ_FILE_OPS(uapsd_noagg_bssids);
 
 #ifdef CONFIG_ACPI
 MVM_DEBUGFS_READ_FILE_OPS(sar_geo_profile);
+MVM_DEBUGFS_READ_FILE_OPS(wifi_6e_enable);
 #endif
 
 MVM_DEBUGFS_READ_WRITE_STA_FILE_OPS(amsdu_len, 16);
@@ -2092,6 +2113,7 @@ void iwl_mvm_dbgfs_register(struct iwl_mvm *mvm)
 	MVM_DEBUGFS_ADD_FILE(tas_get_status, mvm->debugfs_dir, 0400);
 #ifdef CONFIG_ACPI
 	MVM_DEBUGFS_ADD_FILE(sar_geo_profile, mvm->debugfs_dir, 0400);
+	MVM_DEBUGFS_ADD_FILE(wifi_6e_enable, mvm->debugfs_dir, 0400);
 #endif
 	MVM_DEBUGFS_ADD_FILE(he_sniffer_params, mvm->debugfs_dir, 0600);
 
-- 
cgit v1.2.3-70-g09d2


From 70582b826365d27b0be2c7cd0ebe85d01b438d2b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 13 Apr 2023 21:40:30 +0300
Subject: wifi: iwlwifi: pcie: work around ROM bug on AX210 integrated

On 22000 and AX210 devices, there's a ROM bug that causes it to
set invalid LTR settings. On 22000 and AX210 non-integrated we
can fix up these settings from the driver (as done in the code
here), but on AX210 integrated these registers are not available
to the driver.

Attempt to work around the issue by spinning while the IML is
being loaded, the IML will then reprogram the LTR values itself
after it's loaded, so only the brief IML load (which the ROM is
doing) is affected.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.aaa0a4339984.If08da23e960b6236f8c05c06fc8b26041ac89f1e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-csr.h       |  2 +
 .../net/wireless/intel/iwlwifi/pcie/trans-gen2.c   | 72 ++++++++++++++++++++--
 2 files changed, 69 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index bece76b1a514..52479516c76b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -102,6 +102,8 @@
 #define CSR_LTR_LONG_VAL_AD_SNOOP_VAL		0x000003ff
 #define CSR_LTR_LONG_VAL_AD_SCALE_USEC		2
 
+#define CSR_LTR_LAST_MSG			(CSR_BASE + 0x0DC)
+
 /* GIO Chicken Bits (PCI Express bus link power management) */
 #define CSR_GIO_CHICKEN_BITS    (CSR_BASE+0x100)
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index 1e263154e9eb..45b63e3f5df3 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -350,7 +350,7 @@ void iwl_trans_pcie_gen2_fw_alive(struct iwl_trans *trans, u32 scd_addr)
 	mutex_unlock(&trans_pcie->mutex);
 }
 
-static void iwl_pcie_set_ltr(struct iwl_trans *trans)
+static bool iwl_pcie_set_ltr(struct iwl_trans *trans)
 {
 	u32 ltr_val = CSR_LTR_LONG_VAL_AD_NO_SNOOP_REQ |
 		      u32_encode_bits(CSR_LTR_LONG_VAL_AD_SCALE_USEC,
@@ -371,18 +371,77 @@ static void iwl_pcie_set_ltr(struct iwl_trans *trans)
 	     trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_22000) &&
 	    !trans->trans_cfg->integrated) {
 		iwl_write32(trans, CSR_LTR_LONG_VAL_AD, ltr_val);
-	} else if (trans->trans_cfg->integrated &&
-		   trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_22000) {
+		return true;
+	}
+
+	if (trans->trans_cfg->integrated &&
+	    trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_22000) {
 		iwl_write_prph(trans, HPM_MAC_LTR_CSR, HPM_MAC_LRT_ENABLE_ALL);
 		iwl_write_prph(trans, HPM_UMAC_LTR, ltr_val);
+		return true;
+	}
+
+	if (trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_AX210) {
+		/* First clear the interrupt, just in case */
+		iwl_write32(trans, CSR_MSIX_HW_INT_CAUSES_AD,
+			    MSIX_HW_INT_CAUSES_REG_IML);
+		/* In this case, unfortunately the same ROM bug exists in the
+		 * device (not setting LTR correctly), but we don't have control
+		 * over the settings from the host due to some hardware security
+		 * features. The only workaround we've been able to come up with
+		 * so far is to try to keep the CPU and device busy by polling
+		 * it and the IML (image loader) completed interrupt.
+		 */
+		return false;
+	}
+
+	/* nothing needs to be done on other devices */
+	return true;
+}
+
+static void iwl_pcie_spin_for_iml(struct iwl_trans *trans)
+{
+/* in practice, this seems to complete in around 20-30ms at most, wait 100 */
+#define IML_WAIT_TIMEOUT	(HZ / 10)
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	unsigned long end_time = jiffies + IML_WAIT_TIMEOUT;
+	u32 value, loops = 0;
+	bool irq = false;
+
+	if (WARN_ON(!trans_pcie->iml))
+		return;
+
+	value = iwl_read32(trans, CSR_LTR_LAST_MSG);
+	IWL_DEBUG_INFO(trans, "Polling for IML load - CSR_LTR_LAST_MSG=0x%x\n",
+		       value);
+
+	while (time_before(jiffies, end_time)) {
+		if (iwl_read32(trans, CSR_MSIX_HW_INT_CAUSES_AD) &
+				MSIX_HW_INT_CAUSES_REG_IML) {
+			irq = true;
+			break;
+		}
+		/* Keep the CPU and device busy. */
+		value = iwl_read32(trans, CSR_LTR_LAST_MSG);
+		loops++;
 	}
+
+	IWL_DEBUG_INFO(trans,
+		       "Polled for IML load: irq=%d, loops=%d, CSR_LTR_LAST_MSG=0x%x\n",
+		       irq, loops, value);
+
+	/* We don't fail here even if we timed out - maybe we get lucky and the
+	 * interrupt comes in later (and we get alive from firmware) and then
+	 * we're all happy - but if not we'll fail on alive timeout or get some
+	 * other error out.
+	 */
 }
 
 int iwl_trans_pcie_gen2_start_fw(struct iwl_trans *trans,
 				 const struct fw_img *fw, bool run_in_rfkill)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	bool hw_rfkill;
+	bool hw_rfkill, keep_ram_busy;
 	int ret;
 
 	/* This may fail if AMT took ownership of the device */
@@ -443,7 +502,7 @@ int iwl_trans_pcie_gen2_start_fw(struct iwl_trans *trans,
 	if (ret)
 		goto out;
 
-	iwl_pcie_set_ltr(trans);
+	keep_ram_busy = !iwl_pcie_set_ltr(trans);
 
 	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) {
 		iwl_write32(trans, CSR_FUNC_SCRATCH, CSR_FUNC_SCRATCH_INIT_VALUE);
@@ -455,6 +514,9 @@ int iwl_trans_pcie_gen2_start_fw(struct iwl_trans *trans,
 		iwl_write_prph(trans, UREG_CPU_INIT_RUN, 1);
 	}
 
+	if (keep_ram_busy)
+		iwl_pcie_spin_for_iml(trans);
+
 	/* re-check RF-Kill state since we may have missed the interrupt */
 	hw_rfkill = iwl_pcie_check_hw_rf_kill(trans);
 	if (hw_rfkill && !run_in_rfkill)
-- 
cgit v1.2.3-70-g09d2


From 74aae25013ecae39a133148c36c1009b85b06a79 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Thu, 13 Apr 2023 21:40:31 +0300
Subject: wifi: iwlwifi: Fix the duplicate dump name

Driver is using the dump name generated first time in all
generated dump in case of NMI.
The validity of dump name should be with each dump created
and need to reset after each use.

Fixes: 834f920ef34b ("wifi: iwlwifi: yoyo: Add driver defined dump file name")
Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.6c987f820557.I924f2eae1eef2fe82c7a23be566551653d46f729@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index ca97f2fcb693..51cdbc13fbcd 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -2345,6 +2345,8 @@ static u32 iwl_dump_ini_file_name_info(struct iwl_fw_runtime *fwrt,
 	/* add the dump file name extension tlv to the list */
 	list_add_tail(&entry->list, list);
 
+	fwrt->trans->dbg.dump_file_name_ext_valid = false;
+
 	return entry->size;
 }
 
-- 
cgit v1.2.3-70-g09d2


From b655b9a9f8467684cfa8906713d33b71ea8c8f54 Mon Sep 17 00:00:00 2001
From: Daniel Gabay <daniel.gabay@intel.com>
Date: Thu, 13 Apr 2023 21:40:32 +0300
Subject: wifi: iwlwifi: pcie: fix possible NULL pointer dereference

It is possible that iwl_pci_probe() will fail and free the trans,
then afterwards iwl_pci_remove() will be called and crash by trying
to access trans which is already freed, fix it.

iwlwifi 0000:01:00.0: Detected crf-id 0xa5a5a5a2, cnv-id 0xa5a5a5a2
		      wfpm id 0xa5a5a5a2
iwlwifi 0000:01:00.0: Can't find a correct rfid for crf id 0x5a2
...
BUG: kernel NULL pointer dereference, address: 0000000000000028
...
RIP: 0010:iwl_pci_remove+0x12/0x30 [iwlwifi]
pci_device_remove+0x3e/0xb0
device_release_driver_internal+0x103/0x1f0
driver_detach+0x4c/0x90
bus_remove_driver+0x5c/0xd0
driver_unregister+0x31/0x50
pci_unregister_driver+0x40/0x90
iwl_pci_unregister_driver+0x15/0x20 [iwlwifi]
__exit_compat+0x9/0x98 [iwlwifi]
__x64_sys_delete_module+0x147/0x260

Signed-off-by: Daniel Gabay <daniel.gabay@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.082f6e21341b.I0db21d7fa9a828d571ca886713bd0b5d0b6e1e5c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 70f6c59eec5d..a60329079167 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1733,6 +1733,9 @@ static void iwl_pci_remove(struct pci_dev *pdev)
 {
 	struct iwl_trans *trans = pci_get_drvdata(pdev);
 
+	if (!trans)
+		return;
+
 	iwl_drv_stop(trans->drv);
 
 	iwl_trans_pcie_free(trans);
-- 
cgit v1.2.3-70-g09d2


From 11195ab0d6f3202cf7af1a4c69570f59c377d8ad Mon Sep 17 00:00:00 2001
From: Daniel Gabay <daniel.gabay@intel.com>
Date: Thu, 13 Apr 2023 21:40:33 +0300
Subject: wifi: iwlwifi: yoyo: skip dump correctly on hw error

When NIC is in a bad state, reading data will return 28 bits as
0xa5a5a5a and the lowest 4 bits are not fixed value.

Mask these bits in a few places to skip the dump correctly.

Fixes: 89639e06d0f3 ("iwlwifi: yoyo: support for new DBGI_SRAM region")
Signed-off-by: Daniel Gabay <daniel.gabay@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.df6c0663179d.I36d8487b2419c6fefa65e5514855d94327c3b1eb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index 51cdbc13fbcd..d5b51496bd0b 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -1038,7 +1038,7 @@ iwl_dump_ini_prph_mac_iter(struct iwl_fw_runtime *fwrt,
 	range->range_data_size = reg->dev_addr.size;
 	for (i = 0; i < le32_to_cpu(reg->dev_addr.size); i += 4) {
 		prph_val = iwl_read_prph(fwrt->trans, addr + i);
-		if (prph_val == 0x5a5a5a5a)
+		if ((prph_val & ~0xf) == 0xa5a5a5a0)
 			return -EBUSY;
 		*val++ = cpu_to_le32(prph_val);
 	}
@@ -1562,7 +1562,7 @@ iwl_dump_ini_dbgi_sram_iter(struct iwl_fw_runtime *fwrt,
 		prph_data = iwl_read_prph_no_grab(fwrt->trans, (i % 2) ?
 					  DBGI_SRAM_TARGET_ACCESS_RDATA_MSB :
 					  DBGI_SRAM_TARGET_ACCESS_RDATA_LSB);
-		if (prph_data == 0x5a5a5a5a) {
+		if ((prph_data & ~0xf) == 0xa5a5a5a0) {
 			iwl_trans_release_nic_access(fwrt->trans);
 			return -EBUSY;
 		}
-- 
cgit v1.2.3-70-g09d2


From ba30415118eee374a08b39a0460a1d1e52c24a25 Mon Sep 17 00:00:00 2001
From: Daniel Gabay <daniel.gabay@intel.com>
Date: Thu, 13 Apr 2023 21:40:34 +0300
Subject: wifi: iwlwifi: yoyo: Fix possible division by zero

Don't allow buffer allocation TLV with zero req_size since it
leads later to division by zero in iwl_dbg_tlv_alloc_fragments().
Also, NPK/SRAM locations are allowed to have zero buffer req_size,
don't discard them.

Fixes: a9248de42464 ("iwlwifi: dbg_ini: add TLV allocation new API support")
Signed-off-by: Daniel Gabay <daniel.gabay@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.5d6688ed74d8.I5c2f3a882b50698b708d54f4524dc5bdf11e3d32@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 87366b70b17f..546737320a57 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -138,6 +138,12 @@ static int iwl_dbg_tlv_alloc_buf_alloc(struct iwl_trans *trans,
 	    alloc_id != IWL_FW_INI_ALLOCATION_ID_DBGC1)
 		goto err;
 
+	if (buf_location == IWL_FW_INI_LOCATION_DRAM_PATH &&
+	    alloc->req_size == 0) {
+		IWL_ERR(trans, "WRT: Invalid DRAM buffer allocation requested size (0)\n");
+		return -EINVAL;
+	}
+
 	trans->dbg.fw_mon_cfg[alloc_id] = *alloc;
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 20f8cb7dcbfe0d52f6939467ee6a1b6f92fc0adf Mon Sep 17 00:00:00 2001
From: Ayala Beker <ayala.beker@intel.com>
Date: Thu, 13 Apr 2023 21:40:35 +0300
Subject: wifi: iwlwifi: mvm: scan legacy bands and UHB channels with same
 antenna

In case of UHB scan that follows a scan on legacy bands,
consider both scan commands as part of the same scan cycle,
and thus configure them to use the same antenna configuration.

Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230413213309.fd582c423ad8.I35239f94cb3ee1642d16936199c336a07ec2df8f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/scan.h | 3 +++
 drivers/net/wireless/intel/iwlwifi/fw/file.h     | 1 +
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c    | 5 +++++
 3 files changed, 9 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
index 7ba0e3409199..ec96ba053a5c 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
@@ -709,10 +709,13 @@ enum iwl_umac_scan_general_flags_v2 {
  *     should be aware of a P2P GO operation on the 2GHz band.
  * @IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_RESPECT_P2P_GO_HB: scan event scheduling
  *     should be aware of a P2P GO operation on the 5GHz or 6GHz band.
+ * @IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_DONT_TOGGLE_ANT: don't toggle between
+ *     valid antennas, and use the same antenna as in previous scan
  */
 enum iwl_umac_scan_general_params_flags2 {
 	IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_RESPECT_P2P_GO_LB = BIT(0),
 	IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_RESPECT_P2P_GO_HB = BIT(1),
+	IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_DONT_TOGGLE_ANT   = BIT(2),
 };
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index a0260a1fa099..34ef106a2274 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -459,6 +459,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_SYNCED_TIME			= (__force iwl_ucode_tlv_capa_t)106,
 	IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM        = (__force iwl_ucode_tlv_capa_t)108,
 	IWL_UCODE_TLV_CAPA_BIGTK_TX_SUPPORT		= (__force iwl_ucode_tlv_capa_t)109,
+	IWL_UCODE_TLV_CAPA_SCAN_DONT_TOGGLE_ANT         = (__force iwl_ucode_tlv_capa_t)111,
 
 #ifdef __CHECKER__
 	/* sparse says it cannot increment the previous enum member */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 07045092c717..7749f15b7f14 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -2081,6 +2081,11 @@ static u8 iwl_mvm_scan_umac_flags2(struct iwl_mvm *mvm,
 				IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_RESPECT_P2P_GO_HB;
 	}
 
+	if (params->scan_6ghz &&
+	    fw_has_capa(&mvm->fw->ucode_capa,
+			IWL_UCODE_TLV_CAPA_SCAN_DONT_TOGGLE_ANT))
+		flags |= IWL_UMAC_SCAN_GEN_PARAMS_FLAGS2_DONT_TOGGLE_ANT;
+
 	return flags;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 729040295f96b8060b1a3bdd35f767d266508502 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Fri, 14 Apr 2023 13:11:52 +0300
Subject: wifi: iwlwifi: Add RF Step Type for BZ device

Add RF step id handling and handle for BZ device.
Read CNV Aux register and uses for no OTP case.
Add missing rf support for Bz/Bnj device and correct/add
the mapping for rf-type if OTP not present.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.8f2e2fff7bbc.Iee5554178bc5f134dcc28591db0968e619afbdca@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  |  47 ++-
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |   4 +
 drivers/net/wireless/intel/iwlwifi/iwl-csr.h    |   2 +
 drivers/net/wireless/intel/iwlwifi/iwl-prph.h   |   7 +
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h  |   5 +-
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   | 375 +++++++++++++-----------
 6 files changed, 255 insertions(+), 185 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 0b10b34a77a8..032928631ddc 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -51,6 +51,7 @@
 #define IWL_MA_A_MR_A_FW_PRE		"iwlwifi-ma-a0-mr-a0-"
 #define IWL_MA_A_FM_A_FW_PRE		"iwlwifi-ma-a0-fm-a0-"
 #define IWL_SNJ_A_MR_A_FW_PRE		"iwlwifi-SoSnj-a0-mr-a0-"
+#define IWL_BZ_A_HR_A_FW_PRE		"iwlwifi-bz-a0-hr-b0-"
 #define IWL_BZ_A_HR_B_FW_PRE		"iwlwifi-bz-a0-hr-b0-"
 #define IWL_BZ_A_GF_A_FW_PRE		"iwlwifi-bz-a0-gf-a0-"
 #define IWL_BZ_A_GF4_A_FW_PRE		"iwlwifi-bz-a0-gf4-a0-"
@@ -69,7 +70,9 @@
 #define IWL_BNJ_B_GF_A_FW_PRE		"iwlwifi-BzBnj-b0-gf-a0-"
 #define IWL_BNJ_A_GF4_A_FW_PRE		"iwlwifi-BzBnj-a0-gf4-a0-"
 #define IWL_BNJ_B_GF4_A_FW_PRE		"iwlwifi-BzBnj-b0-gf4-a0-"
+#define IWL_BNJ_A_HR_A_FW_PRE		"iwlwifi-BzBnj-a0-hr-b0-"
 #define IWL_BNJ_A_HR_B_FW_PRE		"iwlwifi-BzBnj-a0-hr-b0-"
+#define IWL_BNJ_B_HR_A_FW_PRE		"iwlwifi-BzBnj-b0-hr-b0-"
 #define IWL_BNJ_B_HR_B_FW_PRE		"iwlwifi-BzBnj-b0-hr-b0-"
 #define IWL_BNJ_B_FM_B_FW_PRE		"iwlwifi-BzBnj-b0-fm-b0-"
 
@@ -118,6 +121,8 @@
 	IWL_MA_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_SNJ_A_MR_A_MODULE_FIRMWARE(api) \
 	IWL_SNJ_A_MR_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_BZ_A_HR_A_MODULE_FIRMWARE(api) \
+	IWL_BZ_A_HR_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_HR_B_MODULE_FIRMWARE(api) \
 	IWL_BZ_A_HR_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_GF_A_MODULE_FIRMWARE(api) \
@@ -127,17 +132,17 @@
 #define IWL_BZ_A_MR_A_MODULE_FIRMWARE(api) \
 	IWL_BZ_A_MR_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_FM_A_MODULE_FIRMWARE(api) \
-		IWL_BZ_A_FM_A_FW_PRE __stringify(api) ".ucode"
+	IWL_BZ_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_FM4_A_MODULE_FIRMWARE(api) \
-		IWL_BZ_A_FM4_A_FW_PRE __stringify(api) ".ucode"
+	IWL_BZ_A_FM4_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_FM_B_MODULE_FIRMWARE(api) \
-		IWL_BZ_A_FM_B_FW_PRE __stringify(api) ".ucode"
+	IWL_BZ_A_FM_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_FM4_B_MODULE_FIRMWARE(api) \
-		IWL_BZ_A_FM4_B_FW_PRE __stringify(api) ".ucode"
+	IWL_BZ_A_FM4_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_GL_A_FM_A_MODULE_FIRMWARE(api) \
-		IWL_GL_A_FM_A_FW_PRE __stringify(api) ".ucode"
+	IWL_GL_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_GL_B_FM_B_MODULE_FIRMWARE(api) \
-		IWL_GL_B_FM_B_FW_PRE __stringify(api) ".ucode"
+	IWL_GL_B_FM_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_FM_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_FM4_A_MODULE_FIRMWARE(api) \
@@ -152,8 +157,12 @@
 	IWL_BNJ_A_GF4_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_B_GF4_A_MODULE_FIRMWARE(api) \
 	IWL_BNJ_B_GF4_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_BNJ_A_HR_A_MODULE_FIRMWARE(api) \
+	IWL_BNJ_A_HR_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_A_HR_B_MODULE_FIRMWARE(api) \
 	IWL_BNJ_A_HR_B_FW_PRE __stringify(api) ".ucode"
+#define IWL_BNJ_B_HR_A_MODULE_FIRMWARE(api) \
+	IWL_BNJ_B_HR_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_B_HR_B_MODULE_FIRMWARE(api) \
 	IWL_BNJ_B_HR_B_FW_PRE __stringify(api) ".ucode"
 #define IWL_BNJ_B_FM_B_MODULE_FIRMWARE(api) \
@@ -928,6 +937,14 @@ const struct iwl_cfg iwl_cfg_quz_a0_hr_b0 = {
 	.num_rbds = IWL_NUM_RBDS_22000_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bz_a0_hr_a0 = {
+	.fw_name_pre = IWL_BZ_A_HR_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS_BZ | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bz_a0_hr_b0 = {
 	.fw_name_pre = IWL_BZ_A_HR_B_FW_PRE,
 	.uhb_supported = true,
@@ -1072,6 +1089,14 @@ const struct iwl_cfg iwl_cfg_bnj_b0_gf4_a0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bnj_a0_hr_a0 = {
+	.fw_name_pre = IWL_BNJ_A_HR_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0 = {
 	.fw_name_pre = IWL_BNJ_A_HR_B_FW_PRE,
 	.uhb_supported = true,
@@ -1080,6 +1105,14 @@ const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_bnj_b0_hr_a0 = {
+	.fw_name_pre = IWL_BNJ_B_HR_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_BZ,
+	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_bnj_b0_hr_b0 = {
 	.fw_name_pre = IWL_BNJ_B_HR_B_FW_PRE,
 	.uhb_supported = true,
@@ -1117,6 +1150,7 @@ MODULE_FIRMWARE(IWL_MA_A_GF4_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_MA_A_MR_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_MA_A_FM_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_SNJ_A_MR_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BZ_A_HR_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
@@ -1132,6 +1166,7 @@ MODULE_FIRMWARE(IWL_BNJ_B_GF_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_B_GF4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_A_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_BNJ_B_HR_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BNJ_B_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_FM4_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_FM4_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 9b7b6fca2b1b..c4ac951de19f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -469,6 +469,7 @@ struct iwl_dev_info {
 	u16 mac_type;
 	u16 rf_type;
 	u8 mac_step;
+	u8 rf_step;
 	u8 rf_id;
 	u8 no_160;
 	u8 cores;
@@ -644,6 +645,7 @@ extern const struct iwl_cfg iwl_cfg_snj_a0_ms_a0;
 extern const struct iwl_cfg iwl_cfg_so_a0_hr_a0;
 extern const struct iwl_cfg iwl_cfg_so_a0_ms_a0;
 extern const struct iwl_cfg iwl_cfg_quz_a0_hr_b0;
+extern const struct iwl_cfg iwl_cfg_bz_a0_hr_a0;
 extern const struct iwl_cfg iwl_cfg_bz_a0_hr_b0;
 extern const struct iwl_cfg iwl_cfg_bz_a0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bz_a0_gf4_a0;
@@ -661,7 +663,9 @@ extern const struct iwl_cfg iwl_cfg_bnj_a0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_gf_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_gf4_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_gf4_a0;
+extern const struct iwl_cfg iwl_cfg_bnj_a0_hr_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_a0_hr_b0;
+extern const struct iwl_cfg iwl_cfg_bnj_b0_hr_a0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_hr_b0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_fm_b0;
 extern const struct iwl_cfg iwl_cfg_bnj_b0_fm4_b0;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index 52479516c76b..587368a0ad4a 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -311,6 +311,8 @@ enum {
 	SILICON_A_STEP = 0,
 	SILICON_B_STEP,
 	SILICON_C_STEP,
+	SILICON_D_STEP,
+	SILICON_E_STEP,
 	SILICON_Z_STEP = 0xf,
 };
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
index 62ce116d3783..0dfe00eae05d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
@@ -350,6 +350,11 @@
 #define WFPM_OTP_CFG1_ADDR		0x00a03098
 #define WFPM_OTP_CFG1_IS_JACKET_BIT	BIT(4)
 #define WFPM_OTP_CFG1_IS_CDB_BIT	BIT(5)
+#define WFPM_OTP_BZ_BNJ_JACKET_BIT	5
+#define WFPM_OTP_BZ_BNJ_CDB_BIT		4
+#define WFPM_OTP_CFG1_IS_JACKET(_val)   (((_val) & 0x00000020) >> WFPM_OTP_BZ_BNJ_JACKET_BIT)
+#define WFPM_OTP_CFG1_IS_CDB(_val)      (((_val) & 0x00000010) >> WFPM_OTP_BZ_BNJ_CDB_BIT)
+
 
 #define WFPM_GP2			0xA030B4
 
@@ -445,6 +450,8 @@ enum {
 #define REG_CRF_ID_TYPE_GF_TC			0xF08
 #define REG_CRF_ID_TYPE_MR			0x810
 #define REG_CRF_ID_TYPE_FM			0x910
+#define REG_CRF_ID_TYPE_FMI			0x930
+#define REG_CRF_ID_TYPE_FMR			0x900
 
 #define HPM_DEBUG			0xA03440
 #define PERSISTENCE_BIT			BIT(12)
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 74ff4788dc83..9f1228b5a384 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -997,7 +997,7 @@ struct iwl_trans_txqs {
  *	0 indicates that frag SKBs (NETIF_F_SG) aren't supported.
  * @hw_rf_id a u32 with the device RF ID
  * @hw_crf_id a u32 with the device CRF ID
- * @hw_cdb_id a u32 with the device CDB ID
+ * @hw_wfpm_id a u32 with the device wfpm ID
  * @hw_id: a u32 with the ID of the device / sub-device.
  *	Set during transport allocation.
  * @hw_id_str: a string with info about HW ID. Set during transport allocation.
@@ -1040,7 +1040,8 @@ struct iwl_trans {
 	u32 hw_rev_step;
 	u32 hw_rf_id;
 	u32 hw_crf_id;
-	u32 hw_cdb_id;
+	u32 hw_cnv_id;
+	u32 hw_wfpm_id;
 	u32 hw_id;
 	char hw_id_str[52];
 	u32 sku_id[3];
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index a60329079167..fb0c2c0f236f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -513,16 +513,17 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 MODULE_DEVICE_TABLE(pci, iwl_hw_card_ids);
 
 #define _IWL_DEV_INFO(_device, _subdevice, _mac_type, _mac_step, _rf_type, \
-		      _rf_id, _no_160, _cores, _cdb, _jacket, _cfg, _name) \
-	{ .device = (_device), .subdevice = (_subdevice), .cfg = &(_cfg),  \
-	  .name = _name, .mac_type = _mac_type, .rf_type = _rf_type,	   \
-	  .no_160 = _no_160, .cores = _cores, .rf_id = _rf_id,		   \
+		      _rf_id, _rf_step, _no_160, _cores, _cdb, _jacket, _cfg, \
+		      _name) \
+	{ .device = (_device), .subdevice = (_subdevice), .cfg = &(_cfg), \
+	  .name = _name, .mac_type = _mac_type, .rf_type = _rf_type, .rf_step = _rf_step, \
+	  .no_160 = _no_160, .cores = _cores, .rf_id = _rf_id, \
 	  .mac_step = _mac_step, .cdb = _cdb, .jacket = _jacket }
 
 #define IWL_DEV_INFO(_device, _subdevice, _cfg, _name) \
-	_IWL_DEV_INFO(_device, _subdevice, IWL_CFG_ANY, IWL_CFG_ANY,	   \
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,  \
-		      IWL_CFG_ANY, IWL_CFG_ANY, _cfg, _name)
+	_IWL_DEV_INFO(_device, _subdevice, IWL_CFG_ANY, IWL_CFG_ANY, \
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, \
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, _cfg, _name)
 
 static const struct iwl_dev_info iwl_dev_info_table[] = {
 #if IS_ENABLED(CONFIG_IWLMVM)
@@ -693,87 +694,87 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_2ac_cfg_soc, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_2ac_cfg_soc, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_2ac_cfg_soc, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_2ac_cfg_soc, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_2ac_cfg_soc, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_2ac_cfg_soc, iwl9560_name),
 
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9461_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT_GNSS, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9270_160_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT_GNSS, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9270_name),
 
 	_IWL_DEV_INFO(0x271B, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_TH1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_TH1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9162_160_name),
 	_IWL_DEV_INFO(0x271B, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_TH1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_TH1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9162_name),
 
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9260_160_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9260_2ac_cfg, iwl9260_name),
 
@@ -781,176 +782,176 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 	/* Qu B step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
 	/* Qu C step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
 	/* QuZ */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
 	/* QnJ */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl9560_qnj_b0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
@@ -958,402 +959,383 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 	/* Qu B step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_qu_b0_hr1_b0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_qu_b0_hr_b0, iwl_ax203_name),
 
 	/* Qu C step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_qu_c0_hr1_b0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_qu_c0_hr_b0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_qu_c0_hr_b0, iwl_ax201_name),
 
 	/* QuZ */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_quz_a0_hr1_b0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_quz_a0_hr_b0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_quz_a0_hr_b0, iwl_ax201_name),
 
 /* QnJ with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_qnj_b0_hr_b0_cfg, iwl_ax201_name),
 
 /* SnJ with Jf */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_jf_b0, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_jf_b0, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_jf_b0, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_jf_b0, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_jf_b0, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_jf_b0, iwl9560_name),
 
 /* SnJ with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_hr_b0, iwl_ax101_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_hr_b0, iwl_ax201_name),
 
 /* Ma */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_hr_b0, iwl_ax201_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_gf_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_gf4_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_mr_a0, iwl_ax221_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_fm_a0, iwl_ax231_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_mr_a0, iwl_ax221_name),
 
 /* So with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax201_name),
 
 /* So-F with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax201_name),
 
 /* So-F with Gf */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax211_2ax_cfg_so_gf_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_ANY,
 		      iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_name),
 
 /* Bz */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_bz_a0_hr_a0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bz_a0_hr_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bz_a0_gf_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bz_a0_gf4_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bz_a0_mr_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_BZ, SILICON_A_STEP,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_A_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bz_a0_fm_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_A_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_NO_JACKET,
 		      iwl_cfg_bz_a0_fm4_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_B_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bz_a0_fm_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_B_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bz_a0_fm4_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_A_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_NO_JACKET,
 		      iwl_cfg_gl_a0_fm_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_B_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_NO_JACKET,
 		      iwl_cfg_gl_b0_fm_b0, iwl_bz_name),
 
 /* BZ Z step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, SILICON_Z_STEP,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bz_z0_gf_a0, iwl_bz_name),
 
 /* BNJ */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_A_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_fm_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_B_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_b0_fm_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_A_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_fm4_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, SILICON_B_STEP,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_b0_fm4_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_gf_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_b0_gf_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_a0_gf4_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_IS_JACKET,
 		      iwl_cfg_bnj_b0_gf4_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_NO_JACKET,
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_bnj_a0_hr_a0, iwl_bz_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, SILICON_A_STEP,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_bnj_a0_hr_b0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
-		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_NO_JACKET,
-		      iwl_cfg_bnj_b0_hr_b0, iwl_bz_name),
-
-/* SoF with JF2 */
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
-		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
-		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_160_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
-		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
-		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_name),
-
-/* SoF with JF */
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
-		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
-		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_160_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
-		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
-		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_160_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
-		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
-		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_name),
+		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_bnj_b0_hr_a0, iwl_bz_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
-		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
-		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_name),
+		      IWL_CFG_MAC_TYPE_GL, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_bnj_b0_hr_b0, iwl_bz_name),
 
 /* SoF with JF2 */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_name),
 
 /* SoF with JF */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_name),
 
 /* So with GF */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax211_2ax_cfg_so_gf_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_ANY,
 		      iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_name),
 
 /* So with JF2 */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF,
+		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_name),
 
 /* So with JF */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV,
+		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_name),
 
@@ -1361,22 +1343,22 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 /* For now we use the same FW as MR, but this will change in the future. */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_ms_a0, iwl_ax204_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_so_a0_ms_a0, iwl_ax204_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_ms_a0, iwl_ax204_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SNJ, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_MS, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_ms_a0, iwl_ax204_name)
 
@@ -1411,8 +1393,16 @@ static int get_crf_id(struct iwl_trans *iwl_trans)
 	/* Read crf info */
 	iwl_trans->hw_crf_id = iwl_read_prph_no_grab(iwl_trans, sd_reg_ver_addr);
 
+	/* Read cnv info */
+	iwl_trans->hw_cnv_id =
+		iwl_read_prph_no_grab(iwl_trans, CNVI_AUX_MISC_CHIP);
+
 	/* Read cdb info (also contains the jacket info if needed in the future */
-	iwl_trans->hw_cdb_id = iwl_read_umac_prph_no_grab(iwl_trans, WFPM_OTP_CFG1_ADDR);
+	iwl_trans->hw_wfpm_id =
+		iwl_read_umac_prph_no_grab(iwl_trans, WFPM_OTP_CFG1_ADDR);
+	IWL_INFO(iwl_trans, "Detected crf-id 0x%x, cnv-id 0x%x wfpm id 0x%x\n",
+		 iwl_trans->hw_crf_id, iwl_trans->hw_cnv_id,
+		 iwl_trans->hw_wfpm_id);
 
 	iwl_trans_release_nic_access(iwl_trans);
 
@@ -1428,7 +1418,11 @@ static int map_crf_id(struct iwl_trans *iwl_trans)
 {
 	int ret = 0;
 	u32 val = iwl_trans->hw_crf_id;
-	u32 cdb = iwl_trans->hw_cdb_id;
+	u32 step_id = REG_CRF_ID_STEP(val);
+	u32 slave_id = REG_CRF_ID_SLAVE(val);
+	u32 jacket_id_cnv  = REG_CRF_ID_SLAVE(iwl_trans->hw_cnv_id);
+	u32 jacket_id_wfpm  = WFPM_OTP_CFG1_IS_JACKET(iwl_trans->hw_wfpm_id);
+	u32 cdb_id_wfpm  = WFPM_OTP_CFG1_IS_CDB(iwl_trans->hw_wfpm_id);
 
 	/* Map between crf id to rf id */
 	switch (REG_CRF_ID_TYPE(val)) {
@@ -1438,9 +1432,12 @@ static int map_crf_id(struct iwl_trans *iwl_trans)
 	case REG_CRF_ID_TYPE_JF_2:
 		iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_JF2 << 12);
 		break;
-	case REG_CRF_ID_TYPE_HR_NONE_CDB:
+	case REG_CRF_ID_TYPE_HR_NONE_CDB_1X1:
 		iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_HR1 << 12);
 		break;
+	case REG_CRF_ID_TYPE_HR_NONE_CDB:
+		iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_HR2 << 12);
+		break;
 	case REG_CRF_ID_TYPE_HR_CDB:
 		iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_HR2 << 12);
 		break;
@@ -1450,27 +1447,43 @@ static int map_crf_id(struct iwl_trans *iwl_trans)
 	case REG_CRF_ID_TYPE_MR:
 		iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_MR << 12);
 		break;
-		case REG_CRF_ID_TYPE_FM:
-			iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_FM << 12);
-			break;
+	case REG_CRF_ID_TYPE_FM:
+	case REG_CRF_ID_TYPE_FMI:
+	case REG_CRF_ID_TYPE_FMR:
+		iwl_trans->hw_rf_id = (IWL_CFG_RF_TYPE_FM << 12);
+		break;
 	default:
 		ret = -EIO;
 		IWL_ERR(iwl_trans,
-			"Can find a correct rfid for crf id 0x%x\n",
+			"Can't find a correct rfid for crf id 0x%x\n",
 			REG_CRF_ID_TYPE(val));
 		goto out;
 
 	}
 
+	/* Set Step-id */
+	iwl_trans->hw_rf_id |= (step_id << 8);
+
 	/* Set CDB capabilities */
-	if (cdb & BIT(4)) {
+	if (cdb_id_wfpm || slave_id) {
 		iwl_trans->hw_rf_id += BIT(28);
 		IWL_INFO(iwl_trans, "Adding cdb to rf id\n");
 	}
 
-	IWL_INFO(iwl_trans, "Detected RF 0x%x from crf id 0x%x\n",
-		 iwl_trans->hw_rf_id, REG_CRF_ID_TYPE(val));
+	/* Set Jacket capabilities */
+	if (jacket_id_wfpm || jacket_id_cnv) {
+		iwl_trans->hw_rf_id += BIT(29);
+		IWL_INFO(iwl_trans, "Adding jacket to rf id\n");
+	}
 
+	IWL_INFO(iwl_trans,
+		 "Detected rf-type 0x%x step-id 0x%x slave-id 0x%x from crf id 0x%x\n",
+		 REG_CRF_ID_TYPE(val), step_id, slave_id, iwl_trans->hw_rf_id);
+	IWL_INFO(iwl_trans,
+		 "Detected cdb-id 0x%x jacket-id 0x%x from wfpm id 0x%x\n",
+		 cdb_id_wfpm, jacket_id_wfpm, iwl_trans->hw_wfpm_id);
+	IWL_INFO(iwl_trans, "Detected jacket-id 0x%x from cnvi id 0x%x\n",
+		 jacket_id_cnv, iwl_trans->hw_cnv_id);
 
 out:
 	return ret;
@@ -1481,8 +1494,8 @@ out:
 
 static const struct iwl_dev_info *
 iwl_pci_find_dev_info(u16 device, u16 subsystem_device,
-		      u16 mac_type, u8 mac_step,
-		      u16 rf_type, u8 cdb, u8 jacket, u8 rf_id, u8 no_160, u8 cores)
+		      u16 mac_type, u8 mac_step, u16 rf_type, u8 cdb,
+		      u8 jacket, u8 rf_id, u8 no_160, u8 cores, u8 rf_step)
 {
 	int num_devices = ARRAY_SIZE(iwl_dev_info_table);
 	int i;
@@ -1533,6 +1546,10 @@ iwl_pci_find_dev_info(u16 device, u16 subsystem_device,
 		    dev_info->cores != cores)
 			continue;
 
+		if (dev_info->rf_step != (u8)IWL_CFG_ANY &&
+		    dev_info->rf_step != rf_step)
+			continue;
+
 		return dev_info;
 	}
 
@@ -1604,6 +1621,10 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_free_trans;
 	}
 
+	IWL_INFO(iwl_trans, "PCI dev %04x/%04x, rev=0x%x, rfid=0x%x\n",
+		 pdev->device, pdev->subsystem_device,
+		 iwl_trans->hw_rev, iwl_trans->hw_rf_id);
+
 	dev_info = iwl_pci_find_dev_info(pdev->device, pdev->subsystem_device,
 					 CSR_HW_REV_TYPE(iwl_trans->hw_rev),
 					 iwl_trans->hw_rev_step,
@@ -1612,8 +1633,8 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 					 CSR_HW_RFID_IS_JACKET(iwl_trans->hw_rf_id),
 					 IWL_SUBDEVICE_RF_ID(pdev->subsystem_device),
 					 IWL_SUBDEVICE_NO_160(pdev->subsystem_device),
-					 IWL_SUBDEVICE_CORES(pdev->subsystem_device));
-
+					 IWL_SUBDEVICE_CORES(pdev->subsystem_device),
+					 CSR_HW_RFID_STEP(iwl_trans->hw_rf_id));
 	if (dev_info) {
 		iwl_trans->cfg = dev_info->cfg;
 		iwl_trans->name = dev_info->name;
-- 
cgit v1.2.3-70-g09d2


From c30a2a64788b3d617a9c5d96adb76c68b0862e5f Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Fri, 14 Apr 2023 13:11:53 +0300
Subject: wifi: iwlwifi: add a new PCI device ID for BZ device

Add support for a new PCI device ID 0x272b once registering with PCIe.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.56342664110d.I5aa6f2858fdcf69fdea4f1a873115a48bd43764e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index fb0c2c0f236f..31b575f35410 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -504,6 +504,7 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
 
 /* Bz devices */
 	{IWL_PCI_DEVICE(0x2727, PCI_ANY_ID, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0x272b, PCI_ANY_ID, iwl_bz_trans_cfg)},
 	{IWL_PCI_DEVICE(0xA840, PCI_ANY_ID, iwl_bz_trans_cfg)},
 	{IWL_PCI_DEVICE(0x7740, PCI_ANY_ID, iwl_bz_trans_cfg)},
 #endif /* CONFIG_IWLMVM */
-- 
cgit v1.2.3-70-g09d2


From 7a41d58936b2cd32197cfe4a583a9c7d3b4dc7b0 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Fri, 14 Apr 2023 13:11:54 +0300
Subject: wifi: iwlwifi: mvm: enable new MLD FW API

Enable the new FW API with MLD based on FW TLV.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.5dc66b08539e.I1c65024c879346b0349e1e431d36ec2b5fd85dd7@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/file.h |  1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 20 ++------------------
 2 files changed, 3 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index 34ef106a2274..b00174c9ebce 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -459,6 +459,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_SYNCED_TIME			= (__force iwl_ucode_tlv_capa_t)106,
 	IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM        = (__force iwl_ucode_tlv_capa_t)108,
 	IWL_UCODE_TLV_CAPA_BIGTK_TX_SUPPORT		= (__force iwl_ucode_tlv_capa_t)109,
+	IWL_UCODE_TLV_CAPA_MLD_API_SUPPORT		= (__force iwl_ucode_tlv_capa_t)110,
 	IWL_UCODE_TLV_CAPA_SCAN_DONT_TOGGLE_ANT         = (__force iwl_ucode_tlv_capa_t)111,
 
 #ifdef __CHECKER__
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 76ac8a480712..42e67df13a1e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1401,24 +1401,8 @@ static inline bool iwl_mvm_has_new_rx_api(struct iwl_mvm *mvm)
 
 static inline bool iwl_mvm_has_mld_api(const struct iwl_fw *fw)
 {
-	return (iwl_fw_lookup_cmd_ver(fw, LINK_CONFIG_CMD,
-				      IWL_FW_CMD_VER_UNKNOWN) !=
-				IWL_FW_CMD_VER_UNKNOWN) &&
-		(iwl_fw_lookup_cmd_ver(fw, MAC_CONFIG_CMD,
-				       IWL_FW_CMD_VER_UNKNOWN) !=
-				IWL_FW_CMD_VER_UNKNOWN) &&
-		(iwl_fw_lookup_cmd_ver(fw, STA_CONFIG_CMD,
-				       IWL_FW_CMD_VER_UNKNOWN) !=
-				IWL_FW_CMD_VER_UNKNOWN) &&
-		(iwl_fw_lookup_cmd_ver(fw, AUX_STA_CMD,
-				       IWL_FW_CMD_VER_UNKNOWN) !=
-				IWL_FW_CMD_VER_UNKNOWN) &&
-		(iwl_fw_lookup_cmd_ver(fw, STA_REMOVE_CMD,
-				       IWL_FW_CMD_VER_UNKNOWN) !=
-				IWL_FW_CMD_VER_UNKNOWN) &&
-		(iwl_fw_lookup_cmd_ver(fw, STA_DISABLE_TX_CMD,
-				       IWL_FW_CMD_VER_UNKNOWN) !=
-				IWL_FW_CMD_VER_UNKNOWN);
+	return fw_has_capa(&fw->ucode_capa,
+			   IWL_UCODE_TLV_CAPA_MLD_API_SUPPORT);
 }
 
 static inline bool iwl_mvm_has_new_tx_api(struct iwl_mvm *mvm)
-- 
cgit v1.2.3-70-g09d2


From 41c1b0b5aa4518ed3b7a97f35164dec298072aff Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Fri, 14 Apr 2023 13:11:55 +0300
Subject: wifi: iwlwifi: Fix spelling mistake "upto" -> "up to"

There is a spelling mistake in a IWL_ERR error message. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.a00965734d02.I760cb43829b3d7a0f567f0958059f2fae0aa8a93@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 75d52e6cf988..042c06a22f5f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -895,7 +895,7 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 				drv->fw.img[IWL_UCODE_WOWLAN].is_dual_cpus =
 					true;
 			} else if ((num_of_cpus > 2) || (num_of_cpus < 1)) {
-				IWL_ERR(drv, "Driver support upto 2 CPUs\n");
+				IWL_ERR(drv, "Driver support up to 2 CPUs\n");
 				return -EINVAL;
 			}
 			break;
-- 
cgit v1.2.3-70-g09d2


From 5f10705eea400212c20c00c5dddd53978383c51c Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Fri, 14 Apr 2023 13:11:56 +0300
Subject: wifi: iwlwifi: mvm: Fix spelling mistake "Gerenal" -> "General"

There is a spelling mistake in a IWL_DEBUG_SCAN debug message. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.eac55cbef8f1.I15cc5d3f1489759bf915bfb3cbbb72b67b18f8b6@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 7749f15b7f14..7ac9bdbfe194 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -2297,7 +2297,7 @@ iwl_mvm_scan_umac_fill_general_p_v11(struct iwl_mvm *mvm,
 
 	iwl_mvm_scan_umac_dwell_v11(mvm, gp, params);
 
-	IWL_DEBUG_SCAN(mvm, "Gerenal: flags=0x%x, flags2=0x%x\n",
+	IWL_DEBUG_SCAN(mvm, "General: flags=0x%x, flags2=0x%x\n",
 		       gen_flags, gen_flags2);
 
 	gp->flags = cpu_to_le16(gen_flags);
-- 
cgit v1.2.3-70-g09d2


From 11e94d2bcd88dea5d9ce99555b6b172f5232d3e2 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Fri, 14 Apr 2023 13:11:57 +0300
Subject: wifi: iwlwifi: mvm: initialize seq variable

Clang static analysis reports this issue
d3.c:567:22: warning: The left operand of '>' is
  a garbage value
  if (seq.tkip.iv32 > cur_rx_iv32)
      ~~~~~~~~~~~~~ ^

seq is never initialized. Call ieee80211_get_key_rx_seq() to
initialize seq.

Fixes: 0419e5e672d6 ("iwlwifi: mvm: d3: separate TKIP data from key iteration")
Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.6dd372f84f93.If1f708c90e6424a935b4eba3917dfb7582e0dd0a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index bf80b1d093a2..798a7a1299e1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -564,6 +564,7 @@ static void iwl_mvm_wowlan_get_tkip_data(struct ieee80211_hw *hw,
 		}
 
 		for (i = 0; i < IWL_NUM_RSC; i++) {
+			ieee80211_get_key_rx_seq(key, i, &seq);
 			/* wrapping isn't allowed, AP must rekey */
 			if (seq.tkip.iv32 > cur_rx_iv32)
 				cur_rx_iv32 = seq.tkip.iv32;
-- 
cgit v1.2.3-70-g09d2


From 8ce437dd5b2e4adef13aa4ecce07392f9966b1ab Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Fri, 14 Apr 2023 13:11:58 +0300
Subject: wifi: iwlwifi: fw: move memset before early return

Clang static analysis reports this representative issue
dbg.c:1455:6: warning: Branch condition evaluates to
a garbage value
  if (!rxf_data.size)
       ^~~~~~~~~~~~~~

This check depends on iwl_ini_get_rxf_data() to clear
rxf_data but the function can return early without
doing the clear.  So move the memset before the early
return.

Fixes: cc9b6012d34b ("iwlwifi: yoyo: use hweight_long instead of bit manipulating")
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.872a7175f1ff.I33802a77a91998276992b088fbe25f61c87c33ac@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index d5b51496bd0b..d9faaae01abd 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -1388,13 +1388,13 @@ static void iwl_ini_get_rxf_data(struct iwl_fw_runtime *fwrt,
 	if (!data)
 		return;
 
+	memset(data, 0, sizeof(*data));
+
 	/* make sure only one bit is set in only one fid */
 	if (WARN_ONCE(hweight_long(fid1) + hweight_long(fid2) != 1,
 		      "fid1=%x, fid2=%x\n", fid1, fid2))
 		return;
 
-	memset(data, 0, sizeof(*data));
-
 	if (fid1) {
 		fifo_idx = ffs(fid1) - 1;
 		if (WARN_ONCE(fifo_idx >= MAX_NUM_LMAC, "fifo_idx=%d\n",
-- 
cgit v1.2.3-70-g09d2


From 58d1b717879bfeabe09b35e41ad667c79933eb2e Mon Sep 17 00:00:00 2001
From: Hyunwoo Kim <imv4bel@gmail.com>
Date: Fri, 14 Apr 2023 13:11:59 +0300
Subject: wifi: iwlwifi: pcie: Fix integer overflow in iwl_write_to_user_buf

An integer overflow occurs in the iwl_write_to_user_buf() function,
which is called by the iwl_dbgfs_monitor_data_read() function.

static bool iwl_write_to_user_buf(char __user *user_buf, ssize_t count,
				  void *buf, ssize_t *size,
				  ssize_t *bytes_copied)
{
	int buf_size_left = count - *bytes_copied;

	buf_size_left = buf_size_left - (buf_size_left % sizeof(u32));
	if (*size > buf_size_left)
		*size = buf_size_left;

If the user passes a SIZE_MAX value to the "ssize_t count" parameter,
the ssize_t count parameter is assigned to "int buf_size_left".
Then compare "*size" with "buf_size_left" . Here, "buf_size_left" is a
negative number, so "*size" is assigned "buf_size_left" and goes into
the third argument of the copy_to_user function, causing a heap overflow.

This is not a security vulnerability because iwl_dbgfs_monitor_data_read()
is a debugfs operation with 0400 privileges.

Signed-off-by: Hyunwoo Kim <imv4bel@gmail.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.2d80ace81532.Iecfba549e0e0be21bbb0324675392e42e75bd5ad@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 40283fe622da..1b32a4035d88 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -2860,7 +2860,7 @@ static bool iwl_write_to_user_buf(char __user *user_buf, ssize_t count,
 				  void *buf, ssize_t *size,
 				  ssize_t *bytes_copied)
 {
-	int buf_size_left = count - *bytes_copied;
+	ssize_t buf_size_left = count - *bytes_copied;
 
 	buf_size_left = buf_size_left - (buf_size_left % sizeof(u32));
 	if (*size > buf_size_left)
-- 
cgit v1.2.3-70-g09d2


From 0c9a8f9084be45525d445bf9e55ccbb152a16e1f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 14 Apr 2023 13:12:00 +0300
Subject: wifi: iwlwifi: mvm: track AP STA pointer and use it for MFP

Instead of relying on the AP STA ID and dereferencing our
firmware -> mac80211 STA array when we really only need
the STA, not the per-link information (and might not have
it, e.g. for a pairwise key installation where this is
relevant), keep track of the AP STA as a pointer to the
mac80211 STA, protected by the mutex. Then use it here in
the key code instead.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.7db01b1508de.I88e6d252cdb0a6e4581d7c5fd7cbc037b4fd40ae@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  8 +++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c  | 13 ++++---------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  2 ++
 3 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d6f29aa2a66f..1670c2cef4c3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3669,6 +3669,7 @@ iwl_mvm_sta_state_notexist_to_none(struct iwl_mvm *mvm,
 				   struct ieee80211_sta *sta,
 				   struct iwl_mvm_sta_state_ops *callbacks)
 {
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	unsigned int i;
 	int ret;
 
@@ -3704,6 +3705,9 @@ iwl_mvm_sta_state_notexist_to_none(struct iwl_mvm *mvm,
 	}
 	ieee80211_sta_recalc_aggregates(sta);
 
+	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
+		mvmvif->ap_sta = sta;
+
 	return 0;
 }
 
@@ -3931,8 +3935,10 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw,
 		ret = 0;
 	} else if (old_state == IEEE80211_STA_NONE &&
 		   new_state == IEEE80211_STA_NOTEXIST) {
-		if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
+		if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls) {
 			iwl_mvm_stop_session_protection(mvm, vif);
+			mvmvif->ap_sta = NULL;
+		}
 		ret = callbacks->rm_sta(mvm, vif, sta);
 		if (sta->tdls) {
 			iwl_mvm_recalc_tdls_state(mvm, vif, false);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index f4785c0a0b84..9ec1c505002f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -41,6 +41,8 @@ static u32 iwl_mvm_get_sec_flags(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	u32 flags = 0;
 
+	lockdep_assert_held(&mvm->mutex);
+
 	if (!(keyconf->flags & IEEE80211_KEY_FLAG_PAIRWISE))
 		flags |= IWL_SEC_KEY_FLAG_MCAST_KEY;
 
@@ -68,18 +70,11 @@ static u32 iwl_mvm_get_sec_flags(struct iwl_mvm *mvm,
 		break;
 	}
 
-	rcu_read_lock();
-	if (!sta && vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA) {
-		u8 sta_id = mvmvif->deflink.ap_sta_id;
-
-		sta = rcu_dereference_check(mvm->fw_id_to_mac_id[sta_id],
-					    lockdep_is_held(&mvm->mutex));
-	}
+	if (!sta && vif->type == NL80211_IFTYPE_STATION)
+		sta = mvmvif->ap_sta;
 
 	if (!IS_ERR_OR_NULL(sta) && sta->mfp)
 		flags |= IWL_SEC_KEY_FLAG_MFP;
-	rcu_read_unlock();
 
 	return flags;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 42e67df13a1e..f435260af519 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -434,6 +434,8 @@ struct iwl_mvm_vif {
 	/* TCP Checksum Offload */
 	netdev_features_t features;
 
+	struct ieee80211_sta *ap_sta;
+
 	/* we can only have 2 GTK + 2 IGTK active at a time */
 	struct ieee80211_key_conf *ap_early_keys[4];
 
-- 
cgit v1.2.3-70-g09d2


From 7035b5baae21be4af3d4432bd83e695c8523c2b0 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 14 Apr 2023 13:12:01 +0300
Subject: wifi: iwlwifi: mvm: make iwl_mvm_mac_ctxt_send_beacon() static

We never needed this to be non-static, that was just an
artifact of the development process. Make it static.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.3c018c15f948.I87fd1aee288f33312a7206492608002d4e8e213e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 8 ++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      | 4 ----
 2 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 82fad042a281..962d8c286fd8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -1102,10 +1102,10 @@ static int iwl_mvm_mac_ctxt_send_beacon_v9(struct iwl_mvm *mvm,
 						sizeof(beacon_cmd));
 }
 
-int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm,
-				 struct ieee80211_vif *vif,
-				 struct sk_buff *beacon,
-				 struct ieee80211_bss_conf *link_conf)
+static int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					struct sk_buff *beacon,
+					struct ieee80211_bss_conf *link_conf)
 {
 	if (WARN_ON(!beacon))
 		return -EINVAL;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index f435260af519..a4f8b5a04c64 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1780,10 +1780,6 @@ int iwl_mvm_mac_ctxt_remove(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
 int iwl_mvm_mac_ctxt_beacon_changed(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif,
 				    struct ieee80211_bss_conf *link_conf);
-int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm,
-				 struct ieee80211_vif *vif,
-				 struct sk_buff *beacon,
-				 struct ieee80211_bss_conf *link_conf);
 int iwl_mvm_mac_ctxt_send_beacon_cmd(struct iwl_mvm *mvm,
 				     struct sk_buff *beacon,
 				     void *data, int len);
-- 
cgit v1.2.3-70-g09d2


From d066a530af8e1833c7ea2cef7784004700c85f79 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 14 Apr 2023 13:12:02 +0300
Subject: wifi: iwlwifi: mvm: fix ptk_pn memory leak

If adding a key to firmware fails we leak the allocated ptk_pn.
This shouldn't happen in practice, but we should still fix it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.99446ffd02bc.I82a2ad6ec1395f188e0a1677cc619e3fcb1feac9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 1670c2cef4c3..540d6e3e30d4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4107,7 +4107,7 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_sta *mvmsta = NULL;
-	struct iwl_mvm_key_pn *ptk_pn;
+	struct iwl_mvm_key_pn *ptk_pn = NULL;
 	int keyidx = key->keyidx;
 	u32 sec_key_id = WIDE_ID(DATA_PATH_GROUP, SEC_KEY_CMD);
 	u8 sec_key_ver = iwl_fw_lookup_cmd_ver(mvm->fw, sec_key_id, 0);
@@ -4265,6 +4265,10 @@ static int __iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 		if (ret) {
 			IWL_WARN(mvm, "set key failed\n");
 			key->hw_key_idx = STA_KEY_IDX_INVALID;
+			if (ptk_pn) {
+				RCU_INIT_POINTER(mvmsta->ptk_pn[keyidx], NULL);
+				kfree(ptk_pn);
+			}
 			/*
 			 * can't add key for RX, but we don't need it
 			 * in the device for TX so still return 0,
-- 
cgit v1.2.3-70-g09d2


From aea99650f7316b43a064311bb79064e397ed237c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 14 Apr 2023 13:12:03 +0300
Subject: wifi: iwlwifi: mvm: set STA mask for keys in MLO

Implement the full STA mask and selecting the correct link
for key installation.

While at it, catch errors if this function returns a bad
zero station mask, rather than waiting for the firmware to
crash on it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.cedae2f21829.Iae07b736c3109d085ad5b74ec8282ce45020da39@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c | 69 ++++++++++++++++++++----
 1 file changed, 60 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index 9ec1c505002f..7c417b39aca4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -14,23 +14,68 @@ static u32 iwl_mvm_get_sec_sta_mask(struct iwl_mvm *mvm,
 				    struct ieee80211_key_conf *keyconf)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif_link_info *link_info = &mvmvif->deflink;
+	struct iwl_mvm_link_sta *link_sta;
+	struct iwl_mvm_sta *mvmsta;
+	u32 result = 0;
+	int link_id;
 
+	lockdep_assert_held(&mvm->mutex);
+
+	if (keyconf->link_id >= 0) {
+		link_info = mvmvif->link[keyconf->link_id];
+		if (!link_info)
+			return 0;
+	}
+
+	/* AP group keys are per link and should be on the mcast STA */
 	if (vif->type == NL80211_IFTYPE_AP &&
 	    !(keyconf->flags & IEEE80211_KEY_FLAG_PAIRWISE))
-		return BIT(mvmvif->deflink.mcast_sta.sta_id);
+		return BIT(link_info->mcast_sta.sta_id);
+
+	/* for client mode use the AP STA also for group keys */
+	if (!sta && vif->type == NL80211_IFTYPE_STATION)
+		sta = mvmvif->ap_sta;
+
+	/* During remove the STA was removed and the group keys come later
+	 * (which sounds like a bad sequence, but remember that to mac80211 the
+	 * group keys have no sta pointer), so we don't have a STA now.
+	 * Since this happens for group keys only, just use the link_info as
+	 * the group keys are per link; make sure that is the case by checking
+	 * we do have a link_id or are not doing MLO.
+	 * Of course the same can be done during add as well, but we must do
+	 * it during remove, since we don't have the mvmvif->ap_sta pointer.
+	 */
+	if (!sta && (keyconf->link_id >= 0 || !vif->valid_links))
+		return BIT(link_info->ap_sta_id);
 
-	if (sta) {
-		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
+	/* this shouldn't happen now */
+	if (!sta)
+		return 0;
 
+	mvmsta = iwl_mvm_sta_from_mac80211(sta);
+
+	/* it's easy when the STA is not an MLD */
+	if (!sta->valid_links)
 		return BIT(mvmsta->deflink.sta_id);
-	}
 
-	if (vif->type == NL80211_IFTYPE_STATION &&
-	    mvmvif->deflink.ap_sta_id != IWL_MVM_INVALID_STA)
-		return BIT(mvmvif->deflink.ap_sta_id);
+	/* but if it is an MLD, get the mask of all the FW STAs it has ... */
+	for (link_id = 0; link_id < ARRAY_SIZE(mvmsta->link); link_id++) {
+		/* unless we have a specific link in mind (GTK on client) */
+		if (keyconf->link_id >= 0 &&
+		    keyconf->link_id != link_id)
+			continue;
+
+		link_sta =
+			rcu_dereference_protected(mvmsta->link[link_id],
+						  lockdep_is_held(&mvm->mutex));
+		if (!link_sta)
+			continue;
+
+		result |= BIT(link_sta->sta_id);
+	}
 
-	/* invalid */
-	return 0;
+	return result;
 }
 
 static u32 iwl_mvm_get_sec_flags(struct iwl_mvm *mvm,
@@ -113,6 +158,9 @@ int iwl_mvm_sec_key_add(struct iwl_mvm *mvm,
 	if (WARN_ON(keyconf->keylen > sizeof(cmd.u.add.key)))
 		return -EINVAL;
 
+	if (WARN_ON(!sta_mask))
+		return -EINVAL;
+
 	if (keyconf->cipher == WLAN_CIPHER_SUITE_WEP40 ||
 	    keyconf->cipher == WLAN_CIPHER_SUITE_WEP104)
 		memcpy(cmd.u.add.key + IWL_SEC_WEP_KEY_OFFSET, keyconf->key,
@@ -159,6 +207,9 @@ static int _iwl_mvm_sec_key_del(struct iwl_mvm *mvm,
 	u32 key_flags = iwl_mvm_get_sec_flags(mvm, vif, sta, keyconf);
 	int ret;
 
+	if (WARN_ON(!sta_mask))
+		return -EINVAL;
+
 	ret = __iwl_mvm_sec_key_del(mvm, sta_mask, key_flags, keyconf->keyidx,
 				    flags);
 	if (ret)
-- 
cgit v1.2.3-70-g09d2


From d3f9cd61b11a3ba9be1ddb27a6a4a2aceb8afe5b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 14 Apr 2023 13:12:04 +0300
Subject: wifi: iwlwifi: mvm: validate station properly in flush

If we have MLO, then the deflink.ap_sta_id cannot be used.
However, we can use the new mvmvif->ap_sta pointer for the
validation instead.

Also don't do it multiple times for different FW instances
of the same AP STA (TDLS STAs are only on a single link).

Note that this isn't really working right yet since the
underlying flush code hasn't been updated yet to know of
multiple link STAs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.7d747df44d65.Ie5392859fab4cfb73c20b49bfee2caadef5fd8ec@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 540d6e3e30d4..718b5ba6ec92 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -5605,6 +5605,7 @@ void iwl_mvm_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	struct iwl_mvm_vif *mvmvif;
 	struct iwl_mvm_sta *mvmsta;
 	struct ieee80211_sta *sta;
+	bool ap_sta_done = false;
 	int i;
 	u32 msk = 0;
 
@@ -5633,8 +5634,14 @@ void iwl_mvm_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		if (mvmsta->vif != vif)
 			continue;
 
+		if (sta == mvmvif->ap_sta) {
+			if (ap_sta_done)
+				continue;
+			ap_sta_done = true;
+		}
+
 		/* make sure only TDLS peers or the AP are flushed */
-		WARN_ON_ONCE(i != mvmvif->deflink.ap_sta_id && !sta->tdls);
+		WARN_ON_ONCE(sta != mvmvif->ap_sta && !sta->tdls);
 
 		if (drop) {
 			if (iwl_mvm_flush_sta(mvm, mvmsta, false))
-- 
cgit v1.2.3-70-g09d2


From 7f11d17f0b0805d68f3d2e116f2e7ff0c655fe39 Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Fri, 14 Apr 2023 13:12:05 +0300
Subject: wifi: iwlwifi: mvm: Fix setting the rate for non station cases

The setting of the rate used for non station cases did not take into
consideration the interface type etc., thus when probe responses
are transmitted on P2P Device interface etc. CCK rates were used
which is not allowed.

Modify the code so the non station case would consider the
interface type etc. For HWs/FWs that do not support rate control,
preserve the previous behavior, i.e., take the rate from the
skb metadata and adjust as needed.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.fa0e4d42205d.Iab4784560c77b965fec070d4d0a1ec84171d94ff@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 10 +++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |  3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c       | 26 +++++++++++------------
 3 files changed, 22 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 962d8c286fd8..20db9fc2e61a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -870,9 +870,9 @@ static u32 iwl_mvm_find_ie_offset(u8 *beacon, u8 eid, u32 frame_size)
 	return ie - beacon;
 }
 
-static u8 iwl_mvm_mac_ctxt_get_lowest_rate(struct iwl_mvm *mvm,
-					   struct ieee80211_tx_info *info,
-					   struct ieee80211_vif *vif)
+u8 iwl_mvm_mac_ctxt_get_lowest_rate(struct iwl_mvm *mvm,
+				    struct ieee80211_tx_info *info,
+				    struct ieee80211_vif *vif)
 {
 	struct ieee80211_supported_band *sband;
 	unsigned long basic = vif->bss_conf.basic_rates;
@@ -892,7 +892,9 @@ static u8 iwl_mvm_mac_ctxt_get_lowest_rate(struct iwl_mvm *mvm,
 		}
 	}
 
-	if (info->band == NL80211_BAND_2GHZ && !vif->p2p) {
+	if (info->band == NL80211_BAND_2GHZ && !vif->p2p &&
+	    vif->type != NL80211_IFTYPE_P2P_DEVICE &&
+	    !(info->flags & IEEE80211_TX_CTL_NO_CCK_RATE)) {
 		if (lowest_cck != IWL_RATE_COUNT)
 			rate = lowest_cck;
 		else if (lowest_ofdm != IWL_RATE_COUNT)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index a4f8b5a04c64..7a8b4be8dcf2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1786,6 +1786,9 @@ int iwl_mvm_mac_ctxt_send_beacon_cmd(struct iwl_mvm *mvm,
 u8 iwl_mvm_mac_ctxt_get_beacon_rate(struct iwl_mvm *mvm,
 				    struct ieee80211_tx_info *info,
 				    struct ieee80211_vif *vif);
+u8 iwl_mvm_mac_ctxt_get_lowest_rate(struct iwl_mvm *mvm,
+				    struct ieee80211_tx_info *info,
+				    struct ieee80211_vif *vif);
 u16 iwl_mvm_mac_ctxt_get_beacon_flags(const struct iwl_fw *fw,
 				      u8 rate_idx);
 void iwl_mvm_mac_ctxt_set_tim(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 478442e16d43..8bdc98daca5c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -329,23 +329,23 @@ static u32 iwl_mvm_get_tx_rate(struct iwl_mvm *mvm,
 			  sta ? iwl_mvm_sta_from_mac80211(sta)->sta_state : -1);
 
 		rate_idx = info->control.rates[0].idx;
+
+		/* For non 2 GHZ band, remap mac80211 rate indices into driver
+		 * indices.
+		 */
+		if (info->band != NL80211_BAND_2GHZ ||
+		    (info->flags & IEEE80211_TX_CTL_NO_CCK_RATE))
+			rate_idx += IWL_FIRST_OFDM_RATE;
+
+		/* For 2.4 GHZ band, check that there is no need to remap */
+		BUILD_BUG_ON(IWL_FIRST_CCK_RATE != 0);
 	}
 
 	/* if the rate isn't a well known legacy rate, take the lowest one */
 	if (rate_idx < 0 || rate_idx >= IWL_RATE_COUNT_LEGACY)
-		rate_idx = rate_lowest_index(
-				&mvm->nvm_data->bands[info->band], sta);
-
-	/*
-	 * For non 2 GHZ band, remap mac80211 rate
-	 * indices into driver indices
-	 */
-	if (info->band != NL80211_BAND_2GHZ ||
-	    (info->flags & IEEE80211_TX_CTL_NO_CCK_RATE))
-		rate_idx += IWL_FIRST_OFDM_RATE;
-
-	/* For 2.4 GHZ band, check that there is no need to remap */
-	BUILD_BUG_ON(IWL_FIRST_CCK_RATE != 0);
+		rate_idx = iwl_mvm_mac_ctxt_get_lowest_rate(mvm,
+							    info,
+							    info->control.vif);
 
 	/* Get PLCP rate for tx_cmd->rate_n_flags */
 	rate_plcp = iwl_mvm_mac80211_idx_to_hwrate(mvm->fw, rate_idx);
-- 
cgit v1.2.3-70-g09d2


From cdc419e9e7f3cbfd17a11cb3bce726903fada546 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 14 Apr 2023 13:12:06 +0300
Subject: wifi: iwlwifi: mvm: tx: remove misleading if statement

The if statement here is misleading, we return zero anyway
since we just checked the 'ret' variable, simplify the code
to remove the condition entirely.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230414130637.57c1eb58e655.I1b47a7771cd66306931089c150c6b5b240bdcba5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 8bdc98daca5c..09787b8a1c4c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -1255,8 +1255,7 @@ int iwl_mvm_tx_skb_sta(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (ret)
 		return ret;
 
-	if (WARN_ON(skb_queue_empty(&mpdus_skbs)))
-		return ret;
+	WARN_ON(skb_queue_empty(&mpdus_skbs));
 
 	while (!skb_queue_empty(&mpdus_skbs)) {
 		skb = __skb_dequeue(&mpdus_skbs);
-- 
cgit v1.2.3-70-g09d2


From d2b6da242454ee886729aad7020dfda7995bc26b Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Tue, 11 Apr 2023 20:48:28 +0800
Subject: wifi: rtw89: 8852c: add beacon filter and CQM support

Adding this supports beacon filter and connection quality monitor.
To make host CPU wake up less, let firmware perform signal
monitoring and beacon processing, then notify driver upon signal
changes or beacon loss.

This feature needs firmware 0.27.56 or newer to support it.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230411124832.14965-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c     |  11 ++-
 drivers/net/wireless/realtek/rtw89/core.h     |   1 +
 drivers/net/wireless/realtek/rtw89/fw.c       | 101 ++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/fw.h       |  63 ++++++++++++++++
 drivers/net/wireless/realtek/rtw89/mac.c      |  59 +++++++++++++++
 drivers/net/wireless/realtek/rtw89/mac.h      |   1 +
 drivers/net/wireless/realtek/rtw89/mac80211.c |   8 ++
 7 files changed, 243 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 56a13be2e283..3e1d9dd6637d 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -1457,6 +1457,7 @@ static void rtw89_vif_rx_stats_iter(void *data, u8 *mac,
 	struct rtw89_rx_desc_info *desc_info = iter_data->desc_info;
 	struct sk_buff *skb = iter_data->skb;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	struct rtw89_rx_phy_ppdu *phy_ppdu = iter_data->phy_ppdu;
 	const u8 *bssid = iter_data->bssid;
 
 	if (rtwdev->scanning &&
@@ -1475,8 +1476,11 @@ static void rtw89_vif_rx_stats_iter(void *data, u8 *mac,
 	if (!ether_addr_equal(vif->bss_conf.bssid, bssid))
 		return;
 
-	if (ieee80211_is_beacon(hdr->frame_control))
+	if (ieee80211_is_beacon(hdr->frame_control)) {
+		if (vif->type == NL80211_IFTYPE_STATION)
+			rtw89_fw_h2c_rssi_offload(rtwdev, phy_ppdu);
 		pkt_stat->beacon_nr++;
+	}
 
 	if (!ether_addr_equal(vif->addr, hdr->addr1))
 		return;
@@ -2539,6 +2543,9 @@ int rtw89_core_sta_disassoc(struct rtw89_dev *rtwdev,
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
 	struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
 
+	if (vif->type == NL80211_IFTYPE_STATION)
+		rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, false);
+
 	rtwdev->total_sta_assoc--;
 	if (sta->tdls)
 		rtwvif->tdls_peer--;
@@ -3415,6 +3422,8 @@ static int rtw89_core_register_hw(struct rtw89_dev *rtwdev)
 	ieee80211_hw_set(hw, SINGLE_SCAN_ON_ALL_BANDS);
 	ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID);
 	ieee80211_hw_set(hw, WANT_MONITOR_VIF);
+	if (RTW89_CHK_FW_FEATURE(BEACON_FILTER, &rtwdev->fw))
+		ieee80211_hw_set(hw, CONNECTION_MONITOR);
 
 	hw->wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) |
 				     BIT(NL80211_IFTYPE_AP) |
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 40fb18b613d9..f81c098a7a89 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3231,6 +3231,7 @@ enum rtw89_fw_feature {
 	RTW89_FW_FEATURE_NO_PACKET_DROP,
 	RTW89_FW_FEATURE_NO_DEEP_PS,
 	RTW89_FW_FEATURE_NO_LPS_PG,
+	RTW89_FW_FEATURE_BEACON_FILTER,
 };
 
 struct rtw89_fw_suit {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 5fa6863d36b3..032afbaeb71d 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -266,6 +266,7 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 34, 0, TX_WAKE),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 36, 0, SCAN_OFFLOAD),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 40, 0, CRASH_TRIGGER),
+	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 56, 10, BEACON_FILTER),
 };
 
 static void rtw89_fw_recognize_features(struct rtw89_dev *rtwdev)
@@ -1737,6 +1738,106 @@ fail:
 	return ret;
 }
 
+int rtw89_fw_h2c_set_bcn_fltr_cfg(struct rtw89_dev *rtwdev,
+				  struct ieee80211_vif *vif,
+				  bool connect)
+{
+	struct rtw89_vif *rtwvif = vif_to_rtwvif_safe(vif);
+	struct ieee80211_bss_conf *bss_conf = vif ? &vif->bss_conf : NULL;
+	struct rtw89_h2c_bcnfltr *h2c;
+	u32 len = sizeof(*h2c);
+	struct sk_buff *skb;
+	int ret;
+
+	if (!RTW89_CHK_FW_FEATURE(BEACON_FILTER, &rtwdev->fw))
+		return -EINVAL;
+
+	if (!rtwvif || !bss_conf || rtwvif->net_type != RTW89_NET_TYPE_INFRA)
+		return -EINVAL;
+
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len);
+	if (!skb) {
+		rtw89_err(rtwdev, "failed to alloc skb for h2c bcn filter\n");
+		return -ENOMEM;
+	}
+
+	skb_put(skb, len);
+	h2c = (struct rtw89_h2c_bcnfltr *)skb->data;
+
+	h2c->w0 = le32_encode_bits(connect, RTW89_H2C_BCNFLTR_W0_MON_RSSI) |
+		  le32_encode_bits(connect, RTW89_H2C_BCNFLTR_W0_MON_BCN) |
+		  le32_encode_bits(connect, RTW89_H2C_BCNFLTR_W0_MON_EN) |
+		  le32_encode_bits(RTW89_BCN_FLTR_OFFLOAD_MODE_DEFAULT,
+				   RTW89_H2C_BCNFLTR_W0_MODE) |
+		  le32_encode_bits(RTW89_BCN_LOSS_CNT, RTW89_H2C_BCNFLTR_W0_BCN_LOSS_CNT) |
+		  le32_encode_bits(bss_conf->cqm_rssi_hyst, RTW89_H2C_BCNFLTR_W0_RSSI_HYST) |
+		  le32_encode_bits(bss_conf->cqm_rssi_thold + MAX_RSSI,
+				   RTW89_H2C_BCNFLTR_W0_RSSI_THRESHOLD) |
+		  le32_encode_bits(rtwvif->mac_id, RTW89_H2C_BCNFLTR_W0_MAC_ID);
+
+	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
+			      H2C_CAT_MAC, H2C_CL_MAC_FW_OFLD,
+			      H2C_FUNC_CFG_BCNFLTR, 0, 1, len);
+
+	ret = rtw89_h2c_tx(rtwdev, skb, false);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to send h2c\n");
+		goto fail;
+	}
+
+	return 0;
+fail:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
+int rtw89_fw_h2c_rssi_offload(struct rtw89_dev *rtwdev,
+			      struct rtw89_rx_phy_ppdu *phy_ppdu)
+{
+	struct rtw89_h2c_ofld_rssi *h2c;
+	u32 len = sizeof(*h2c);
+	struct sk_buff *skb;
+	s8 rssi;
+	int ret;
+
+	if (!RTW89_CHK_FW_FEATURE(BEACON_FILTER, &rtwdev->fw))
+		return -EINVAL;
+
+	if (!phy_ppdu)
+		return -EINVAL;
+
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len);
+	if (!skb) {
+		rtw89_err(rtwdev, "failed to alloc skb for h2c rssi\n");
+		return -ENOMEM;
+	}
+
+	rssi = phy_ppdu->rssi_avg >> RSSI_FACTOR;
+	skb_put(skb, len);
+	h2c = (struct rtw89_h2c_ofld_rssi *)skb->data;
+
+	h2c->w0 = le32_encode_bits(phy_ppdu->mac_id, RTW89_H2C_OFLD_RSSI_W0_MACID) |
+		  le32_encode_bits(1, RTW89_H2C_OFLD_RSSI_W0_NUM);
+	h2c->w1 = le32_encode_bits(rssi, RTW89_H2C_OFLD_RSSI_W1_VAL);
+
+	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
+			      H2C_CAT_MAC, H2C_CL_MAC_FW_OFLD,
+			      H2C_FUNC_OFLD_RSSI, 0, 1, len);
+
+	ret = rtw89_h2c_tx(rtwdev, skb, false);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to send h2c\n");
+		goto fail;
+	}
+
+	return 0;
+fail:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
 #define H2C_RA_LEN 16
 int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index c3c67ddf61a2..4d8d961f7b23 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -162,6 +162,27 @@ enum rtw89_p2pps_action {
 	RTW89_P2P_ACT_TERMINATE = 3,
 };
 
+enum rtw89_bcn_fltr_offload_mode {
+	RTW89_BCN_FLTR_OFFLOAD_MODE_0 = 0,
+	RTW89_BCN_FLTR_OFFLOAD_MODE_1,
+	RTW89_BCN_FLTR_OFFLOAD_MODE_2,
+	RTW89_BCN_FLTR_OFFLOAD_MODE_3,
+
+	RTW89_BCN_FLTR_OFFLOAD_MODE_DEFAULT = RTW89_BCN_FLTR_OFFLOAD_MODE_0,
+};
+
+enum rtw89_bcn_fltr_type {
+	RTW89_BCN_FLTR_BEACON_LOSS,
+	RTW89_BCN_FLTR_RSSI,
+	RTW89_BCN_FLTR_NOTIFY,
+};
+
+enum rtw89_bcn_fltr_rssi_event {
+	RTW89_BCN_FLTR_RSSI_NOT_CHANGED,
+	RTW89_BCN_FLTR_RSSI_HIGH,
+	RTW89_BCN_FLTR_RSSI_LOW,
+};
+
 #define FWDL_SECTION_MAX_NUM 10
 #define FWDL_SECTION_CHKSUM_LEN	8
 #define FWDL_SECTION_PER_PKT_LEN 2020
@@ -216,6 +237,8 @@ struct rtw89_h2creg_sch_tx_en {
 #define RTW89_SCAN_LIST_LIMIT \
 		((RTW89_H2C_MAX_SIZE / RTW89_MAC_CHINFO_SIZE) - RTW89_SCAN_LIST_GUARD)
 
+#define RTW89_BCN_LOSS_CNT 10
+
 struct rtw89_mac_chinfo {
 	u8 period;
 	u8 dwell_time;
@@ -3317,6 +3340,17 @@ static inline struct rtw89_fw_c2h_attr *RTW89_SKB_C2H_CB(struct sk_buff *skb)
 #define RTW89_GET_MAC_C2H_REV_ACK_H2C_SEQ(c2h) \
 	le32_get_bits(*((const __le32 *)(c2h) + 2), GENMASK(23, 16))
 
+struct rtw89_c2h_mac_bcnfltr_rpt {
+	__le32 w0;
+	__le32 w1;
+	__le32 w2;
+} __packed;
+
+#define RTW89_C2H_MAC_BCNFLTR_RPT_W2_MACID GENMASK(7, 0)
+#define RTW89_C2H_MAC_BCNFLTR_RPT_W2_TYPE GENMASK(9, 8)
+#define RTW89_C2H_MAC_BCNFLTR_RPT_W2_EVENT GENMASK(11, 10)
+#define RTW89_C2H_MAC_BCNFLTR_RPT_W2_MA GENMASK(23, 16)
+
 #define RTW89_GET_PHY_C2H_RA_RPT_MACID(c2h) \
 	le32_get_bits(*((const __le32 *)(c2h) + 2), GENMASK(15, 0))
 #define RTW89_GET_PHY_C2H_RA_RPT_RETRY_RATIO(c2h) \
@@ -3410,6 +3444,28 @@ static_assert(sizeof(struct rtw89_mac_mcc_tsf_rpt) <= RTW89_COMPLETION_BUF_SIZE)
 #define RTW89_GET_MAC_C2H_MCC_STATUS_RPT_TSF_HIGH(c2h) \
 	le32_get_bits(*((const __le32 *)(c2h) + 4), GENMASK(31, 0))
 
+struct rtw89_h2c_bcnfltr {
+	__le32 w0;
+} __packed;
+
+#define RTW89_H2C_BCNFLTR_W0_MON_RSSI BIT(0)
+#define RTW89_H2C_BCNFLTR_W0_MON_BCN BIT(1)
+#define RTW89_H2C_BCNFLTR_W0_MON_EN BIT(2)
+#define RTW89_H2C_BCNFLTR_W0_MODE GENMASK(4, 3)
+#define RTW89_H2C_BCNFLTR_W0_BCN_LOSS_CNT GENMASK(11, 8)
+#define RTW89_H2C_BCNFLTR_W0_RSSI_HYST GENMASK(15, 12)
+#define RTW89_H2C_BCNFLTR_W0_RSSI_THRESHOLD GENMASK(23, 16)
+#define RTW89_H2C_BCNFLTR_W0_MAC_ID GENMASK(31, 24)
+
+struct rtw89_h2c_ofld_rssi {
+	__le32 w0;
+	__le32 w1;
+} __packed;
+
+#define RTW89_H2C_OFLD_RSSI_W0_MACID GENMASK(7, 0)
+#define RTW89_H2C_OFLD_RSSI_W0_NUM GENMASK(15, 8)
+#define RTW89_H2C_OFLD_RSSI_W1_VAL GENMASK(7, 0)
+
 #define RTW89_FW_HDR_SIZE 32
 #define RTW89_FW_SECTION_HDR_SIZE 16
 
@@ -3537,6 +3593,8 @@ struct rtw89_fw_h2c_rf_reg_info {
 #define H2C_FUNC_ADD_SCANOFLD_CH	0x16
 #define H2C_FUNC_SCANOFLD		0x17
 #define H2C_FUNC_PKT_DROP		0x1b
+#define H2C_FUNC_CFG_BCNFLTR		0x1e
+#define H2C_FUNC_OFLD_RSSI		0x1f
 
 /* CLASS 10 - Security CAM */
 #define H2C_CL_MAC_SEC_CAM		0xa
@@ -3637,6 +3695,11 @@ int rtw89_fw_h2c_macid_pause(struct rtw89_dev *rtwdev, u8 sh, u8 grp,
 int rtw89_fw_h2c_set_edca(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
 			  u8 ac, u32 val);
 int rtw89_fw_h2c_set_ofld_cfg(struct rtw89_dev *rtwdev);
+int rtw89_fw_h2c_set_bcn_fltr_cfg(struct rtw89_dev *rtwdev,
+				  struct ieee80211_vif *vif,
+				  bool connect);
+int rtw89_fw_h2c_rssi_offload(struct rtw89_dev *rtwdev,
+			      struct rtw89_rx_phy_ppdu *phy_ppdu);
 int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi);
 int rtw89_fw_h2c_cxdrv_init(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_role(struct rtw89_dev *rtwdev);
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index d0e138f8b880..367a7f95401c 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4237,6 +4237,64 @@ rtw89_mac_c2h_scanofld_rsp(struct rtw89_dev *rtwdev, struct sk_buff *c2h,
 	}
 }
 
+static void
+rtw89_mac_bcn_fltr_rpt(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+		       struct sk_buff *skb)
+{
+	struct ieee80211_vif *vif = rtwvif_to_vif_safe(rtwvif);
+	enum nl80211_cqm_rssi_threshold_event nl_event;
+	const struct rtw89_c2h_mac_bcnfltr_rpt *c2h =
+		(const struct rtw89_c2h_mac_bcnfltr_rpt *)skb->data;
+	u8 type, event, mac_id;
+	s8 sig;
+
+	type = le32_get_bits(c2h->w2, RTW89_C2H_MAC_BCNFLTR_RPT_W2_TYPE);
+	sig = le32_get_bits(c2h->w2, RTW89_C2H_MAC_BCNFLTR_RPT_W2_MA) - MAX_RSSI;
+	event = le32_get_bits(c2h->w2, RTW89_C2H_MAC_BCNFLTR_RPT_W2_EVENT);
+	mac_id = le32_get_bits(c2h->w2, RTW89_C2H_MAC_BCNFLTR_RPT_W2_MACID);
+
+	if (mac_id != rtwvif->mac_id)
+		return;
+
+	rtw89_debug(rtwdev, RTW89_DBG_FW,
+		    "C2H bcnfltr rpt macid: %d, type: %d, ma: %d, event: %d\n",
+		    mac_id, type, sig, event);
+
+	switch (type) {
+	case RTW89_BCN_FLTR_BEACON_LOSS:
+		if (!rtwdev->scanning)
+			ieee80211_connection_loss(vif);
+		else
+			rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true);
+		return;
+	case RTW89_BCN_FLTR_NOTIFY:
+		nl_event = NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH;
+		break;
+	case RTW89_BCN_FLTR_RSSI:
+		if (event == RTW89_BCN_FLTR_RSSI_LOW)
+			nl_event = NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW;
+		else if (event == RTW89_BCN_FLTR_RSSI_HIGH)
+			nl_event = NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH;
+		else
+			return;
+		break;
+	default:
+		return;
+	}
+
+	ieee80211_cqm_rssi_notify(vif, nl_event, sig, GFP_KERNEL);
+}
+
+static void
+rtw89_mac_c2h_bcn_fltr_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h,
+			   u32 len)
+{
+	struct rtw89_vif *rtwvif;
+
+	rtw89_for_each_rtwvif(rtwdev, rtwvif)
+		rtw89_mac_bcn_fltr_rpt(rtwdev, rtwvif, c2h);
+}
+
 static void
 rtw89_mac_c2h_rec_ack(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len)
 {
@@ -4457,6 +4515,7 @@ void (* const rtw89_mac_c2h_ofld_handler[])(struct rtw89_dev *rtwdev,
 	[RTW89_MAC_C2H_FUNC_MACID_PAUSE] = rtw89_mac_c2h_macid_pause,
 	[RTW89_MAC_C2H_FUNC_SCANOFLD_RSP] = rtw89_mac_c2h_scanofld_rsp,
 	[RTW89_MAC_C2H_FUNC_TSF32_TOGL_RPT] = rtw89_mac_c2h_tsf32_toggle_rpt,
+	[RTW89_MAC_C2H_FUNC_BCNFLTR_RPT] = rtw89_mac_c2h_bcn_fltr_rpt,
 };
 
 static
diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h
index 8064d3953d7f..899c45c6774e 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.h
+++ b/drivers/net/wireless/realtek/rtw89/mac.h
@@ -359,6 +359,7 @@ enum rtw89_mac_c2h_ofld_func {
 	RTW89_MAC_C2H_FUNC_MACID_PAUSE,
 	RTW89_MAC_C2H_FUNC_TSF32_TOGL_RPT = 0x6,
 	RTW89_MAC_C2H_FUNC_SCANOFLD_RSP = 0x9,
+	RTW89_MAC_C2H_FUNC_BCNFLTR_RPT = 0xd,
 	RTW89_MAC_C2H_FUNC_OFLD_MAX,
 };
 
diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index 367a7bf319da..629b32dafecb 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -114,6 +114,11 @@ static int rtw89_ops_add_interface(struct ieee80211_hw *hw,
 		    vif->addr, vif->type, vif->p2p);
 
 	mutex_lock(&rtwdev->mutex);
+
+	if (RTW89_CHK_FW_FEATURE(BEACON_FILTER, &rtwdev->fw))
+		vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER |
+				     IEEE80211_VIF_SUPPORTS_CQM_RSSI;
+
 	rtwvif->rtwdev = rtwdev;
 	list_add_tail(&rtwvif->list, &rtwdev->rtwvifs_list);
 	INIT_WORK(&rtwvif->update_beacon_work, rtw89_core_update_beacon_work);
@@ -425,6 +430,9 @@ static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_P2P_PS)
 		rtw89_process_p2p_ps(rtwdev, vif);
 
+	if (changed & BSS_CHANGED_CQM)
+		rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true);
+
 	mutex_unlock(&rtwdev->mutex);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 1ae5ca615285d5d4f72d1de464716d85dffef19f Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Tue, 11 Apr 2023 20:48:29 +0800
Subject: wifi: rtw89: add function to wait for completion of TX skbs

Allocate a per-skb completion to track those skbs we are interested in
and wait for them to complete transmission with TX status.

Normally, the completion object is freed by wait side, but it could be
timeout result that complete side should free the object instead. Add a
owner field with RCU to determine which side should free the object.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230411124832.14965-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 31 ++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/core.h | 40 +++++++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/pci.c  |  5 ++++
 drivers/net/wireless/realtek/rtw89/pci.h  |  4 ++--
 4 files changed, 78 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 3e1d9dd6637d..aa28204cbcd8 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -867,6 +867,37 @@ void rtw89_core_tx_kick_off(struct rtw89_dev *rtwdev, u8 qsel)
 	rtw89_hci_tx_kick_off(rtwdev, ch_dma);
 }
 
+int rtw89_core_tx_kick_off_and_wait(struct rtw89_dev *rtwdev, struct sk_buff *skb,
+				    int qsel, unsigned int timeout)
+{
+	struct rtw89_tx_skb_data *skb_data = RTW89_TX_SKB_CB(skb);
+	struct rtw89_tx_wait_info *wait;
+	unsigned long time_left;
+	int ret = 0;
+
+	wait = kzalloc(sizeof(*wait), GFP_KERNEL);
+	if (!wait) {
+		rtw89_core_tx_kick_off(rtwdev, qsel);
+		return 0;
+	}
+
+	init_completion(&wait->completion);
+	rcu_assign_pointer(skb_data->wait, wait);
+
+	rtw89_core_tx_kick_off(rtwdev, qsel);
+	time_left = wait_for_completion_timeout(&wait->completion,
+						msecs_to_jiffies(timeout));
+	if (time_left == 0)
+		ret = -ETIMEDOUT;
+	else if (!wait->tx_done)
+		ret = -EAGAIN;
+
+	rcu_assign_pointer(skb_data->wait, NULL);
+	kfree_rcu(wait, rcu_head);
+
+	return ret;
+}
+
 int rtw89_h2c_tx(struct rtw89_dev *rtwdev,
 		 struct sk_buff *skb, bool fwdl)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index f81c098a7a89..d94d26397724 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -2623,6 +2623,17 @@ struct rtw89_phy_rate_pattern {
 	bool enable;
 };
 
+struct rtw89_tx_wait_info {
+	struct rcu_head rcu_head;
+	struct completion completion;
+	bool tx_done;
+};
+
+struct rtw89_tx_skb_data {
+	struct rtw89_tx_wait_info __rcu *wait;
+	u8 hci_priv[];
+};
+
 #define RTW89_P2P_MAX_NOA_NUM 2
 
 struct rtw89_vif {
@@ -4179,6 +4190,14 @@ static inline void rtw89_hci_clear(struct rtw89_dev *rtwdev, struct pci_dev *pde
 		rtwdev->hci.ops->clear(rtwdev, pdev);
 }
 
+static inline
+struct rtw89_tx_skb_data *RTW89_TX_SKB_CB(struct sk_buff *skb)
+{
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
+	return (struct rtw89_tx_skb_data *)info->status.status_driver_data;
+}
+
 static inline u8 rtw89_read8(struct rtw89_dev *rtwdev, u32 addr)
 {
 	return rtwdev->hci.ops->read8(rtwdev, addr);
@@ -4822,11 +4841,32 @@ static inline struct sk_buff *rtw89_alloc_skb_for_rx(struct rtw89_dev *rtwdev,
 	return dev_alloc_skb(length);
 }
 
+static inline void rtw89_core_tx_wait_complete(struct rtw89_dev *rtwdev,
+					       struct rtw89_tx_skb_data *skb_data,
+					       bool tx_done)
+{
+	struct rtw89_tx_wait_info *wait;
+
+	rcu_read_lock();
+
+	wait = rcu_dereference(skb_data->wait);
+	if (!wait)
+		goto out;
+
+	wait->tx_done = tx_done;
+	complete(&wait->completion);
+
+out:
+	rcu_read_unlock();
+}
+
 int rtw89_core_tx_write(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta, struct sk_buff *skb, int *qsel);
 int rtw89_h2c_tx(struct rtw89_dev *rtwdev,
 		 struct sk_buff *skb, bool fwdl);
 void rtw89_core_tx_kick_off(struct rtw89_dev *rtwdev, u8 qsel);
+int rtw89_core_tx_kick_off_and_wait(struct rtw89_dev *rtwdev, struct sk_buff *skb,
+				    int qsel, unsigned int timeout);
 void rtw89_core_fill_txdesc(struct rtw89_dev *rtwdev,
 			    struct rtw89_tx_desc_info *desc_info,
 			    void *txdesc);
diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index 68f0fed6d31e..fbe4d11ca920 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -364,8 +364,11 @@ static void rtw89_pci_tx_status(struct rtw89_dev *rtwdev,
 				struct rtw89_pci_tx_ring *tx_ring,
 				struct sk_buff *skb, u8 tx_status)
 {
+	struct rtw89_tx_skb_data *skb_data = RTW89_TX_SKB_CB(skb);
 	struct ieee80211_tx_info *info;
 
+	rtw89_core_tx_wait_complete(rtwdev, skb_data, tx_status == RTW89_TX_DONE);
+
 	info = IEEE80211_SKB_CB(skb);
 	ieee80211_tx_info_clear_status(info);
 
@@ -1203,6 +1206,7 @@ static int rtw89_pci_txwd_submit(struct rtw89_dev *rtwdev,
 	struct pci_dev *pdev = rtwpci->pdev;
 	struct sk_buff *skb = tx_req->skb;
 	struct rtw89_pci_tx_data *tx_data = RTW89_PCI_TX_SKB_CB(skb);
+	struct rtw89_tx_skb_data *skb_data = RTW89_TX_SKB_CB(skb);
 	bool en_wd_info = desc_info->en_wd_info;
 	u32 txwd_len;
 	u32 txwp_len;
@@ -1218,6 +1222,7 @@ static int rtw89_pci_txwd_submit(struct rtw89_dev *rtwdev,
 	}
 
 	tx_data->dma = dma;
+	rcu_assign_pointer(skb_data->wait, NULL);
 
 	txwp_len = sizeof(*txwp_info);
 	txwd_len = chip->txwd_body_size;
diff --git a/drivers/net/wireless/realtek/rtw89/pci.h b/drivers/net/wireless/realtek/rtw89/pci.h
index 1e19740db8c5..0e4bd210b100 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.h
+++ b/drivers/net/wireless/realtek/rtw89/pci.h
@@ -1004,9 +1004,9 @@ rtw89_pci_rxbd_increase(struct rtw89_pci_rx_ring *rx_ring, u32 cnt)
 
 static inline struct rtw89_pci_tx_data *RTW89_PCI_TX_SKB_CB(struct sk_buff *skb)
 {
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct rtw89_tx_skb_data *data = RTW89_TX_SKB_CB(skb);
 
-	return (struct rtw89_pci_tx_data *)info->status.status_driver_data;
+	return (struct rtw89_pci_tx_data *)data->hci_priv;
 }
 
 static inline struct rtw89_pci_tx_bd_32 *
-- 
cgit v1.2.3-70-g09d2


From a0e97ae3f3320a246a79db7372fc23a38556014e Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Tue, 11 Apr 2023 20:48:30 +0800
Subject: wifi: rtw89: add ieee80211::remain_on_channel ops

Add support of remain on channel ops. Since channel context is
required to enable multi-channel concurrent(MCC) and the current
ROC in mac80211 don't support more than 1 channel context, add this
to let P2P and other protocols relying on this work as expected.
The off-channel duration and cancel timing is purely controlled by
upper layers.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230411124832.14965-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/chan.c     |  35 +++++
 drivers/net/wireless/realtek/rtw89/chan.h     |   3 +
 drivers/net/wireless/realtek/rtw89/core.c     | 212 +++++++++++++++++++++++++-
 drivers/net/wireless/realtek/rtw89/core.h     |  30 ++++
 drivers/net/wireless/realtek/rtw89/mac.c      |   5 +-
 drivers/net/wireless/realtek/rtw89/mac80211.c |  80 +++++++++-
 drivers/net/wireless/realtek/rtw89/ps.h       |  16 ++
 7 files changed, 378 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/chan.c b/drivers/net/wireless/realtek/rtw89/chan.c
index 90596806bc93..4663db4ce2f6 100644
--- a/drivers/net/wireless/realtek/rtw89/chan.c
+++ b/drivers/net/wireless/realtek/rtw89/chan.c
@@ -141,6 +141,38 @@ void rtw89_config_entity_chandef(struct rtw89_dev *rtwdev,
 	__rtw89_config_entity_chandef(rtwdev, idx, chandef, true);
 }
 
+void rtw89_config_roc_chandef(struct rtw89_dev *rtwdev,
+			      enum rtw89_sub_entity_idx idx,
+			      const struct cfg80211_chan_def *chandef)
+{
+	struct rtw89_hal *hal = &rtwdev->hal;
+	enum rtw89_sub_entity_idx cur;
+
+	if (chandef) {
+		cur = atomic_cmpxchg(&hal->roc_entity_idx,
+				     RTW89_SUB_ENTITY_IDLE, idx);
+		if (cur != RTW89_SUB_ENTITY_IDLE) {
+			rtw89_debug(rtwdev, RTW89_DBG_TXRX,
+				    "ROC still processing on entity %d\n", idx);
+			return;
+		}
+
+		hal->roc_chandef = *chandef;
+	} else {
+		cur = atomic_cmpxchg(&hal->roc_entity_idx, idx,
+				     RTW89_SUB_ENTITY_IDLE);
+		if (cur == idx)
+			return;
+
+		if (cur == RTW89_SUB_ENTITY_IDLE)
+			rtw89_debug(rtwdev, RTW89_DBG_TXRX,
+				    "ROC already finished on entity %d\n", idx);
+		else
+			rtw89_debug(rtwdev, RTW89_DBG_TXRX,
+				    "ROC is processing on entity %d\n", cur);
+	}
+}
+
 static void rtw89_config_default_chandef(struct rtw89_dev *rtwdev)
 {
 	struct cfg80211_chan_def chandef = {0};
@@ -154,6 +186,7 @@ void rtw89_entity_init(struct rtw89_dev *rtwdev)
 	struct rtw89_hal *hal = &rtwdev->hal;
 
 	bitmap_zero(hal->entity_map, NUM_OF_RTW89_SUB_ENTITY);
+	atomic_set(&hal->roc_entity_idx, RTW89_SUB_ENTITY_IDLE);
 	rtw89_config_default_chandef(rtwdev);
 }
 
@@ -229,6 +262,8 @@ void rtw89_chanctx_ops_remove(struct rtw89_dev *rtwdev,
 			rtwvif->sub_entity_idx = RTW89_SUB_ENTITY_0;
 	}
 
+	atomic_cmpxchg(&hal->roc_entity_idx, roll, RTW89_SUB_ENTITY_0);
+
 	drop = roll;
 
 out:
diff --git a/drivers/net/wireless/realtek/rtw89/chan.h b/drivers/net/wireless/realtek/rtw89/chan.h
index ecbd4503bead..bdf369db5041 100644
--- a/drivers/net/wireless/realtek/rtw89/chan.h
+++ b/drivers/net/wireless/realtek/rtw89/chan.h
@@ -45,6 +45,9 @@ bool rtw89_assign_entity_chan(struct rtw89_dev *rtwdev,
 void rtw89_config_entity_chandef(struct rtw89_dev *rtwdev,
 				 enum rtw89_sub_entity_idx idx,
 				 const struct cfg80211_chan_def *chandef);
+void rtw89_config_roc_chandef(struct rtw89_dev *rtwdev,
+			      enum rtw89_sub_entity_idx idx,
+			      const struct cfg80211_chan_def *chandef);
 void rtw89_entity_init(struct rtw89_dev *rtwdev);
 enum rtw89_entity_mode rtw89_entity_recalc(struct rtw89_dev *rtwdev);
 int rtw89_chanctx_ops_add(struct rtw89_dev *rtwdev,
diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index aa28204cbcd8..30a5ab5e48b1 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -2014,6 +2014,18 @@ static void rtw89_core_free_sta_pending_forbid_ba(struct rtw89_dev *rtwdev,
 	spin_unlock_bh(&rtwdev->ba_lock);
 }
 
+static void rtw89_core_free_sta_pending_roc_tx(struct rtw89_dev *rtwdev,
+					       struct ieee80211_sta *sta)
+{
+	struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
+	struct sk_buff *skb, *tmp;
+
+	skb_queue_walk_safe(&rtwsta->roc_queue, skb, tmp) {
+		skb_unlink(skb, &rtwsta->roc_queue);
+		dev_kfree_skb_any(skb);
+	}
+}
+
 static void rtw89_core_stop_tx_ba_session(struct rtw89_dev *rtwdev,
 					  struct rtw89_txq *rtwtxq)
 {
@@ -2153,6 +2165,7 @@ static void rtw89_core_txq_schedule(struct rtw89_dev *rtwdev, u8 ac, bool *reinv
 {
 	struct ieee80211_hw *hw = rtwdev->hw;
 	struct ieee80211_txq *txq;
+	struct rtw89_vif *rtwvif;
 	struct rtw89_txq *rtwtxq;
 	unsigned long frame_cnt;
 	unsigned long byte_cnt;
@@ -2162,6 +2175,12 @@ static void rtw89_core_txq_schedule(struct rtw89_dev *rtwdev, u8 ac, bool *reinv
 	ieee80211_txq_schedule_start(hw, ac);
 	while ((txq = ieee80211_next_txq(hw, ac))) {
 		rtwtxq = (struct rtw89_txq *)txq->drv_priv;
+		rtwvif = (struct rtw89_vif *)txq->vif->drv_priv;
+
+		if (rtwvif->offchan) {
+			ieee80211_return_txq(hw, txq, true);
+			continue;
+		}
 		tx_resource = rtw89_check_and_reclaim_tx_resource(rtwdev, txq->tid);
 		sched_txq = false;
 
@@ -2230,6 +2249,187 @@ static void rtw89_forbid_ba_work(struct work_struct *w)
 	spin_unlock_bh(&rtwdev->ba_lock);
 }
 
+static void rtw89_core_sta_pending_tx_iter(void *data,
+					   struct ieee80211_sta *sta)
+{
+	struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
+	struct rtw89_vif *rtwvif_target = data, *rtwvif = rtwsta->rtwvif;
+	struct rtw89_dev *rtwdev = rtwvif->rtwdev;
+	struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif);
+	struct sk_buff *skb, *tmp;
+	int qsel, ret;
+
+	if (rtwvif->sub_entity_idx != rtwvif_target->sub_entity_idx)
+		return;
+
+	if (skb_queue_len(&rtwsta->roc_queue) == 0)
+		return;
+
+	skb_queue_walk_safe(&rtwsta->roc_queue, skb, tmp) {
+		skb_unlink(skb, &rtwsta->roc_queue);
+
+		ret = rtw89_core_tx_write(rtwdev, vif, sta, skb, &qsel);
+		if (ret) {
+			rtw89_warn(rtwdev, "pending tx failed with %d\n", ret);
+			dev_kfree_skb_any(skb);
+		} else {
+			rtw89_core_tx_kick_off(rtwdev, qsel);
+		}
+	}
+}
+
+static void rtw89_core_handle_sta_pending_tx(struct rtw89_dev *rtwdev,
+					     struct rtw89_vif *rtwvif)
+{
+	ieee80211_iterate_stations_atomic(rtwdev->hw,
+					  rtw89_core_sta_pending_tx_iter,
+					  rtwvif);
+}
+
+static int rtw89_core_send_nullfunc(struct rtw89_dev *rtwdev,
+				    struct rtw89_vif *rtwvif, bool qos, bool ps)
+{
+	struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif);
+	struct ieee80211_sta *sta;
+	struct ieee80211_hdr *hdr;
+	struct sk_buff *skb;
+	int ret, qsel;
+
+	if (vif->type != NL80211_IFTYPE_STATION || !vif->cfg.assoc)
+		return 0;
+
+	rcu_read_lock();
+	sta = ieee80211_find_sta(vif, vif->bss_conf.bssid);
+	if (!sta) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	skb = ieee80211_nullfunc_get(rtwdev->hw, vif, -1, qos);
+	if (!skb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	hdr = (struct ieee80211_hdr *)skb->data;
+	if (ps)
+		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
+
+	ret = rtw89_core_tx_write(rtwdev, vif, sta, skb, &qsel);
+	if (ret) {
+		rtw89_warn(rtwdev, "nullfunc transmit failed: %d\n", ret);
+		dev_kfree_skb_any(skb);
+		goto out;
+	}
+
+	rcu_read_unlock();
+
+	return rtw89_core_tx_kick_off_and_wait(rtwdev, skb, qsel,
+					       RTW89_ROC_TX_TIMEOUT);
+out:
+	rcu_read_unlock();
+
+	return ret;
+}
+
+void rtw89_roc_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
+{
+	struct ieee80211_hw *hw = rtwdev->hw;
+	struct rtw89_roc *roc = &rtwvif->roc;
+	struct cfg80211_chan_def roc_chan;
+	struct rtw89_vif *tmp;
+	int ret;
+
+	lockdep_assert_held(&rtwdev->mutex);
+
+	ieee80211_queue_delayed_work(hw, &rtwvif->roc.roc_work,
+				     msecs_to_jiffies(rtwvif->roc.duration));
+
+	rtw89_leave_ips_by_hwflags(rtwdev);
+	rtw89_leave_lps(rtwdev);
+
+	ret = rtw89_core_send_nullfunc(rtwdev, rtwvif, true, true);
+	if (ret)
+		rtw89_debug(rtwdev, RTW89_DBG_TXRX,
+			    "roc send null-1 failed: %d\n", ret);
+
+	rtw89_for_each_rtwvif(rtwdev, tmp)
+		if (tmp->sub_entity_idx == rtwvif->sub_entity_idx)
+			tmp->offchan = true;
+
+	cfg80211_chandef_create(&roc_chan, &roc->chan, NL80211_CHAN_NO_HT);
+	rtw89_config_roc_chandef(rtwdev, rtwvif->sub_entity_idx, &roc_chan);
+	rtw89_set_channel(rtwdev);
+	rtw89_write32_clr(rtwdev,
+			  rtw89_mac_reg_by_idx(R_AX_RX_FLTR_OPT, RTW89_MAC_0),
+			  B_AX_A_UC_CAM_MATCH | B_AX_A_BC_CAM_MATCH);
+
+	ieee80211_ready_on_channel(hw);
+}
+
+void rtw89_roc_end(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
+{
+	struct ieee80211_hw *hw = rtwdev->hw;
+	struct rtw89_roc *roc = &rtwvif->roc;
+	struct rtw89_vif *tmp;
+	int ret;
+
+	lockdep_assert_held(&rtwdev->mutex);
+
+	ieee80211_remain_on_channel_expired(hw);
+
+	rtw89_leave_ips_by_hwflags(rtwdev);
+	rtw89_leave_lps(rtwdev);
+
+	rtw89_write32_mask(rtwdev,
+			   rtw89_mac_reg_by_idx(R_AX_RX_FLTR_OPT, RTW89_MAC_0),
+			   B_AX_RX_FLTR_CFG_MASK,
+			   rtwdev->hal.rx_fltr);
+
+	roc->state = RTW89_ROC_IDLE;
+	rtw89_config_roc_chandef(rtwdev, rtwvif->sub_entity_idx, NULL);
+	rtw89_set_channel(rtwdev);
+	ret = rtw89_core_send_nullfunc(rtwdev, rtwvif, true, false);
+	if (ret)
+		rtw89_debug(rtwdev, RTW89_DBG_TXRX,
+			    "roc send null-0 failed: %d\n", ret);
+
+	rtw89_for_each_rtwvif(rtwdev, tmp)
+		if (tmp->sub_entity_idx == rtwvif->sub_entity_idx)
+			tmp->offchan = false;
+
+	rtw89_core_handle_sta_pending_tx(rtwdev, rtwvif);
+	queue_work(rtwdev->txq_wq, &rtwdev->txq_work);
+
+	if (hw->conf.flags & IEEE80211_CONF_IDLE)
+		ieee80211_queue_delayed_work(hw, &roc->roc_work,
+					     RTW89_ROC_IDLE_TIMEOUT);
+}
+
+void rtw89_roc_work(struct work_struct *work)
+{
+	struct rtw89_vif *rtwvif = container_of(work, struct rtw89_vif,
+						roc.roc_work.work);
+	struct rtw89_dev *rtwdev = rtwvif->rtwdev;
+	struct rtw89_roc *roc = &rtwvif->roc;
+
+	mutex_lock(&rtwdev->mutex);
+
+	switch (roc->state) {
+	case RTW89_ROC_IDLE:
+		rtw89_enter_ips_by_hwflags(rtwdev);
+		break;
+	case RTW89_ROC_MGMT:
+	case RTW89_ROC_NORMAL:
+		rtw89_roc_end(rtwdev, rtwvif);
+		break;
+	default:
+		break;
+	}
+
+	mutex_unlock(&rtwdev->mutex);
+}
+
 static enum rtw89_tfc_lv rtw89_get_traffic_level(struct rtw89_dev *rtwdev,
 						 u32 throughput, u64 cnt)
 {
@@ -2299,6 +2499,9 @@ static void rtw89_vif_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwv
 	    rtwvif->tdls_peer)
 		return;
 
+	if (rtwvif->offchan)
+		return;
+
 	if (rtwvif->stats.tx_tfc_lv == RTW89_TFC_IDLE &&
 	    rtwvif->stats.rx_tfc_lv == RTW89_TFC_IDLE)
 		rtw89_enter_lps(rtwdev, rtwvif);
@@ -2528,6 +2731,7 @@ int rtw89_core_sta_add(struct rtw89_dev *rtwdev,
 	rtwsta->rtwvif = rtwvif;
 	rtwsta->prev_rssi = 0;
 	INIT_LIST_HEAD(&rtwsta->ba_cam_list);
+	skb_queue_head_init(&rtwsta->roc_queue);
 
 	for (i = 0; i < ARRAY_SIZE(sta->txq); i++)
 		rtw89_core_txq_init(rtwdev, sta->txq[i]);
@@ -2597,6 +2801,8 @@ int rtw89_core_sta_disconnect(struct rtw89_dev *rtwdev,
 	rtw89_mac_bf_disassoc(rtwdev, vif, sta);
 	rtw89_core_free_sta_pending_ba(rtwdev, sta);
 	rtw89_core_free_sta_pending_forbid_ba(rtwdev, sta);
+	rtw89_core_free_sta_pending_roc_tx(rtwdev, sta);
+
 	if (vif->type == NL80211_IFTYPE_AP || sta->tdls)
 		rtw89_cam_deinit_addr_cam(rtwdev, &rtwsta->addr_cam);
 	if (sta->tdls)
@@ -3480,6 +3686,7 @@ static int rtw89_core_register_hw(struct rtw89_dev *rtwdev)
 	hw->wiphy->tid_config_support.peer |= BIT(NL80211_TID_CONFIG_ATTR_AMPDU_CTRL);
 	hw->wiphy->tid_config_support.vif |= BIT(NL80211_TID_CONFIG_ATTR_AMSDU_CTRL);
 	hw->wiphy->tid_config_support.peer |= BIT(NL80211_TID_CONFIG_ATTR_AMSDU_CTRL);
+	hw->wiphy->max_remain_on_channel_duration = 1000;
 
 	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 
@@ -3563,7 +3770,8 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 		goto err;
 
 	no_chanctx = chip->support_chanctx_num == 0 ||
-		     !(early_feat_map & BIT(RTW89_FW_FEATURE_SCAN_OFFLOAD));
+		     !(early_feat_map & BIT(RTW89_FW_FEATURE_SCAN_OFFLOAD)) ||
+		     !(early_feat_map & BIT(RTW89_FW_FEATURE_BEACON_FILTER));
 
 	if (no_chanctx) {
 		ops->add_chanctx = NULL;
@@ -3571,6 +3779,8 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 		ops->change_chanctx = NULL;
 		ops->assign_vif_chanctx = NULL;
 		ops->unassign_vif_chanctx = NULL;
+		ops->remain_on_channel = NULL;
+		ops->cancel_remain_on_channel = NULL;
 	}
 
 	driver_data_size = sizeof(struct rtw89_dev) + bus_data_size;
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index d94d26397724..39083a0e6ce3 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -569,6 +569,7 @@ enum rtw89_sub_entity_idx {
 	RTW89_SUB_ENTITY_0 = 0,
 
 	NUM_OF_RTW89_SUB_ENTITY,
+	RTW89_SUB_ENTITY_IDLE = NUM_OF_RTW89_SUB_ENTITY,
 };
 
 enum rtw89_rf_path {
@@ -2597,6 +2598,7 @@ struct rtw89_sta {
 	struct rtw89_addr_cam_entry addr_cam; /* AP mode or TDLS peer only */
 	struct rtw89_bssid_cam_entry bssid_cam; /* TDLS peer only */
 	struct list_head ba_cam_list;
+	struct sk_buff_head roc_queue;
 
 	bool use_cfg_mask;
 	struct cfg80211_bitrate_mask mask;
@@ -2634,11 +2636,28 @@ struct rtw89_tx_skb_data {
 	u8 hci_priv[];
 };
 
+#define RTW89_ROC_IDLE_TIMEOUT 500
+#define RTW89_ROC_TX_TIMEOUT 30
+enum rtw89_roc_state {
+	RTW89_ROC_IDLE,
+	RTW89_ROC_NORMAL,
+	RTW89_ROC_MGMT,
+};
+
+struct rtw89_roc {
+	struct ieee80211_channel chan;
+	struct delayed_work roc_work;
+	enum ieee80211_roc_type type;
+	enum rtw89_roc_state state;
+	int duration;
+};
+
 #define RTW89_P2P_MAX_NOA_NUM 2
 
 struct rtw89_vif {
 	struct list_head list;
 	struct rtw89_dev *rtwdev;
+	struct rtw89_roc roc;
 	enum rtw89_sub_entity_idx sub_entity_idx;
 
 	u8 mac_id;
@@ -2654,6 +2673,7 @@ struct rtw89_vif {
 	u8 bcn_hit_cond;
 	u8 hit_rule;
 	u8 last_noa_nr;
+	bool offchan;
 	bool trigger;
 	bool lsig_txop;
 	u8 tgt_ind;
@@ -3370,9 +3390,11 @@ struct rtw89_hal {
 	bool tx_path_diversity;
 	bool support_cckpd;
 	bool support_igi;
+	atomic_t roc_entity_idx;
 
 	DECLARE_BITMAP(entity_map, NUM_OF_RTW89_SUB_ENTITY);
 	struct rtw89_sub_entity sub[NUM_OF_RTW89_SUB_ENTITY];
+	struct cfg80211_chan_def roc_chandef;
 
 	bool entity_active;
 	enum rtw89_entity_mode entity_mode;
@@ -4035,6 +4057,7 @@ struct rtw89_dev {
 	struct delayed_work coex_rfk_chk_work;
 	struct delayed_work cfo_track_work;
 	struct delayed_work forbid_ba_work;
+	struct delayed_work roc_work;
 	struct rtw89_ppdu_sts_info ppdu_sts;
 	u8 total_sta_assoc;
 	bool scanning;
@@ -4550,6 +4573,10 @@ const struct cfg80211_chan_def *rtw89_chandef_get(struct rtw89_dev *rtwdev,
 						  enum rtw89_sub_entity_idx idx)
 {
 	struct rtw89_hal *hal = &rtwdev->hal;
+	enum rtw89_sub_entity_idx roc_idx = atomic_read(&hal->roc_entity_idx);
+
+	if (roc_idx == idx)
+		return &hal->roc_chandef;
 
 	return &hal->sub[idx].chandef;
 }
@@ -4936,6 +4963,9 @@ void rtw89_complete_cond(struct rtw89_wait_info *wait, unsigned int cond,
 int rtw89_core_start(struct rtw89_dev *rtwdev);
 void rtw89_core_stop(struct rtw89_dev *rtwdev);
 void rtw89_core_update_beacon_work(struct work_struct *work);
+void rtw89_roc_work(struct work_struct *work);
+void rtw89_roc_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
+void rtw89_roc_end(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
 void rtw89_core_scan_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
 			   const u8 *mac_addr, bool hw_scan);
 void rtw89_core_scan_complete(struct rtw89_dev *rtwdev,
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 367a7f95401c..54887835ac15 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4193,6 +4193,9 @@ rtw89_mac_c2h_scanofld_rsp(struct rtw89_dev *rtwdev, struct sk_buff *c2h,
 	u16 chan;
 	int ret;
 
+	if (!rtwvif)
+		return;
+
 	tx_fail = RTW89_GET_MAC_C2H_SCANOFLD_TX_FAIL(c2h->data);
 	status = RTW89_GET_MAC_C2H_SCANOFLD_STATUS(c2h->data);
 	chan = RTW89_GET_MAC_C2H_SCANOFLD_PRI_CH(c2h->data);
@@ -4262,7 +4265,7 @@ rtw89_mac_bcn_fltr_rpt(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
 
 	switch (type) {
 	case RTW89_BCN_FLTR_BEACON_LOSS:
-		if (!rtwdev->scanning)
+		if (!rtwdev->scanning && !rtwvif->offchan)
 			ieee80211_connection_loss(vif);
 		else
 			rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true);
diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index 629b32dafecb..b059aa8d88db 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -23,9 +23,19 @@ static void rtw89_ops_tx(struct ieee80211_hw *hw,
 	struct rtw89_dev *rtwdev = hw->priv;
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_vif *vif = info->control.vif;
+	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
 	struct ieee80211_sta *sta = control->sta;
+	u32 flags = IEEE80211_SKB_CB(skb)->flags;
 	int ret, qsel;
 
+	if (rtwvif->offchan && !(flags & IEEE80211_TX_CTL_TX_OFFCHAN) && sta) {
+		struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
+
+		rtw89_debug(rtwdev, RTW89_DBG_TXRX, "ops_tx during offchan\n");
+		skb_queue_tail(&rtwsta->roc_queue, skb);
+		return;
+	}
+
 	ret = rtw89_core_tx_write(rtwdev, vif, sta, skb, &qsel);
 	if (ret) {
 		rtw89_err(rtwdev, "failed to transmit skb: %d\n", ret);
@@ -115,13 +125,18 @@ static int rtw89_ops_add_interface(struct ieee80211_hw *hw,
 
 	mutex_lock(&rtwdev->mutex);
 
+	rtw89_leave_ips_by_hwflags(rtwdev);
+
 	if (RTW89_CHK_FW_FEATURE(BEACON_FILTER, &rtwdev->fw))
 		vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER |
 				     IEEE80211_VIF_SUPPORTS_CQM_RSSI;
 
 	rtwvif->rtwdev = rtwdev;
+	rtwvif->roc.state = RTW89_ROC_IDLE;
+	rtwvif->offchan = false;
 	list_add_tail(&rtwvif->list, &rtwdev->rtwvifs_list);
 	INIT_WORK(&rtwvif->update_beacon_work, rtw89_core_update_beacon_work);
+	INIT_DELAYED_WORK(&rtwvif->roc.roc_work, rtw89_roc_work);
 	rtw89_leave_ps_mode(rtwdev);
 
 	rtw89_traffic_stats_init(rtwdev, &rtwvif->stats);
@@ -168,6 +183,7 @@ static void rtw89_ops_remove_interface(struct ieee80211_hw *hw,
 		    vif->addr, vif->type, vif->p2p);
 
 	cancel_work_sync(&rtwvif->update_beacon_work);
+	cancel_delayed_work_sync(&rtwvif->roc.roc_work);
 
 	mutex_lock(&rtwdev->mutex);
 	rtw89_leave_ps_mode(rtwdev);
@@ -175,6 +191,8 @@ static void rtw89_ops_remove_interface(struct ieee80211_hw *hw,
 	rtw89_mac_remove_vif(rtwdev, rtwvif);
 	rtw89_core_release_bit_map(rtwdev->hw_port, rtwvif->port);
 	list_del_init(&rtwvif->list);
+	rtw89_enter_ips_by_hwflags(rtwdev);
+
 	mutex_unlock(&rtwdev->mutex);
 }
 
@@ -803,12 +821,13 @@ static int rtw89_ops_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			     struct ieee80211_scan_request *req)
 {
 	struct rtw89_dev *rtwdev = hw->priv;
+	struct rtw89_vif *rtwvif = vif_to_rtwvif_safe(vif);
 	int ret = 0;
 
 	if (!RTW89_CHK_FW_FEATURE(SCAN_OFFLOAD, &rtwdev->fw))
 		return 1;
 
-	if (rtwdev->scanning)
+	if (rtwdev->scanning || rtwvif->offchan)
 		return -EBUSY;
 
 	mutex_lock(&rtwdev->mutex);
@@ -911,6 +930,63 @@ static void rtw89_ops_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	mutex_unlock(&rtwdev->mutex);
 }
 
+static int rtw89_ops_remain_on_channel(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_channel *chan,
+				       int duration,
+				       enum ieee80211_roc_type type)
+{
+	struct rtw89_dev *rtwdev = hw->priv;
+	struct rtw89_vif *rtwvif = vif_to_rtwvif_safe(vif);
+	struct rtw89_roc *roc = &rtwvif->roc;
+
+	if (!vif)
+		return -EINVAL;
+
+	mutex_lock(&rtwdev->mutex);
+
+	if (roc->state != RTW89_ROC_IDLE) {
+		mutex_unlock(&rtwdev->mutex);
+		return -EBUSY;
+	}
+
+	if (rtwdev->scanning)
+		rtw89_hw_scan_abort(rtwdev, vif);
+
+	if (type == IEEE80211_ROC_TYPE_MGMT_TX)
+		roc->state = RTW89_ROC_MGMT;
+	else
+		roc->state = RTW89_ROC_NORMAL;
+
+	roc->duration = duration;
+	roc->chan = *chan;
+	roc->type = type;
+
+	rtw89_roc_start(rtwdev, rtwvif);
+
+	mutex_unlock(&rtwdev->mutex);
+
+	return 0;
+}
+
+static int rtw89_ops_cancel_remain_on_channel(struct ieee80211_hw *hw,
+					      struct ieee80211_vif *vif)
+{
+	struct rtw89_dev *rtwdev = hw->priv;
+	struct rtw89_vif *rtwvif = vif_to_rtwvif_safe(vif);
+
+	if (!rtwvif)
+		return -EINVAL;
+
+	cancel_delayed_work_sync(&rtwvif->roc.roc_work);
+
+	mutex_lock(&rtwdev->mutex);
+	rtw89_roc_end(rtwdev, rtwvif);
+	mutex_unlock(&rtwdev->mutex);
+
+	return 0;
+}
+
 static void rtw89_set_tid_config_iter(void *data, struct ieee80211_sta *sta)
 {
 	struct cfg80211_tid_config *tid_config = data;
@@ -1022,6 +1098,8 @@ const struct ieee80211_ops rtw89_ops = {
 	.change_chanctx		= rtw89_ops_change_chanctx,
 	.assign_vif_chanctx	= rtw89_ops_assign_vif_chanctx,
 	.unassign_vif_chanctx	= rtw89_ops_unassign_vif_chanctx,
+	.remain_on_channel		= rtw89_ops_remain_on_channel,
+	.cancel_remain_on_channel	= rtw89_ops_cancel_remain_on_channel,
 	.set_sar_specs		= rtw89_ops_set_sar_specs,
 	.sta_rc_update		= rtw89_ops_sta_rc_update,
 	.set_tid_config		= rtw89_ops_set_tid_config,
diff --git a/drivers/net/wireless/realtek/rtw89/ps.h b/drivers/net/wireless/realtek/rtw89/ps.h
index 6ac1f7ea5339..c9e29d92fc1b 100644
--- a/drivers/net/wireless/realtek/rtw89/ps.h
+++ b/drivers/net/wireless/realtek/rtw89/ps.h
@@ -15,4 +15,20 @@ void rtw89_leave_ips(struct rtw89_dev *rtwdev);
 void rtw89_set_coex_ctrl_lps(struct rtw89_dev *rtwdev, bool btc_ctrl);
 void rtw89_process_p2p_ps(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif);
 
+static inline void rtw89_leave_ips_by_hwflags(struct rtw89_dev *rtwdev)
+{
+	struct ieee80211_hw *hw = rtwdev->hw;
+
+	if (hw->conf.flags & IEEE80211_CONF_IDLE)
+		rtw89_leave_ips(rtwdev);
+}
+
+static inline void rtw89_enter_ips_by_hwflags(struct rtw89_dev *rtwdev)
+{
+	struct ieee80211_hw *hw = rtwdev->hw;
+
+	if (hw->conf.flags & IEEE80211_CONF_IDLE)
+		rtw89_enter_ips(rtwdev);
+}
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 6cfb6cc20a61aa17bdb5440459a3503a885ee913 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Tue, 11 Apr 2023 20:48:31 +0800
Subject: wifi: rtw89: add flag check for power state

Use POWER_ON flag to make sure power on/off is symmetric. Since both
remain_on_channel and hw_scan both alter the power state, this makes
sure that we don't enter/leave IPS mode twice.
Also, replace IPS related functions with inline function that does
similar logic so we can track it more easily.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230411124832.14965-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 7 +++----
 drivers/net/wireless/realtek/rtw89/ps.c   | 6 ++++++
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 30a5ab5e48b1..72124d6c2968 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -2207,8 +2207,7 @@ static void rtw89_ips_work(struct work_struct *work)
 	struct rtw89_dev *rtwdev = container_of(work, struct rtw89_dev,
 						ips_work);
 	mutex_lock(&rtwdev->mutex);
-	if (rtwdev->hw->conf.flags & IEEE80211_CONF_IDLE)
-		rtw89_enter_ips(rtwdev);
+	rtw89_enter_ips_by_hwflags(rtwdev);
 	mutex_unlock(&rtwdev->mutex);
 }
 
@@ -3501,8 +3500,8 @@ void rtw89_core_scan_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
 
 	rtwdev->scanning = true;
 	rtw89_leave_lps(rtwdev);
-	if (hw_scan && (rtwdev->hw->conf.flags & IEEE80211_CONF_IDLE))
-		rtw89_leave_ips(rtwdev);
+	if (hw_scan)
+		rtw89_leave_ips_by_hwflags(rtwdev);
 
 	ether_addr_copy(rtwvif->mac_addr, mac_addr);
 	rtw89_btc_ntfy_scan_start(rtwdev, RTW89_PHY_0, chan->band_type);
diff --git a/drivers/net/wireless/realtek/rtw89/ps.c b/drivers/net/wireless/realtek/rtw89/ps.c
index 40498812205e..cf72861c7adc 100644
--- a/drivers/net/wireless/realtek/rtw89/ps.c
+++ b/drivers/net/wireless/realtek/rtw89/ps.c
@@ -155,6 +155,9 @@ void rtw89_enter_ips(struct rtw89_dev *rtwdev)
 
 	set_bit(RTW89_FLAG_INACTIVE_PS, rtwdev->flags);
 
+	if (!test_bit(RTW89_FLAG_POWERON, rtwdev->flags))
+		return;
+
 	rtw89_for_each_rtwvif(rtwdev, rtwvif)
 		rtw89_mac_vif_deinit(rtwdev, rtwvif);
 
@@ -166,6 +169,9 @@ void rtw89_leave_ips(struct rtw89_dev *rtwdev)
 	struct rtw89_vif *rtwvif;
 	int ret;
 
+	if (test_bit(RTW89_FLAG_POWERON, rtwdev->flags))
+		return;
+
 	ret = rtw89_core_start(rtwdev);
 	if (ret)
 		rtw89_err(rtwdev, "failed to leave idle state\n");
-- 
cgit v1.2.3-70-g09d2


From c5280e5f6763c216e16feec10c63b6b32106ddb7 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Tue, 11 Apr 2023 20:48:32 +0800
Subject: wifi: rtw89: fix authentication fail during scan

We used to store operating channel info after associated. However, scan
might happen before that. Without switching back to operating channel,
authentication or association might fail. Therefore, we switch back to
operating channel when the scanning vif's BSSID is non-zero, which
implies connected or during attempt to connect.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230411124832.14965-6-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 032afbaeb71d..d3348f9bcc2a 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -3199,7 +3199,7 @@ static void rtw89_hw_scan_add_chan(struct rtw89_dev *rtwdev, int chan_type,
 }
 
 static int rtw89_hw_scan_add_chan_list(struct rtw89_dev *rtwdev,
-				       struct rtw89_vif *rtwvif)
+				       struct rtw89_vif *rtwvif, bool connected)
 {
 	struct cfg80211_scan_request *req = rtwvif->scan_req;
 	struct rtw89_mac_chinfo	*ch_info, *tmp;
@@ -3243,7 +3243,7 @@ static int rtw89_hw_scan_add_chan_list(struct rtw89_dev *rtwdev,
 			type = RTW89_CHAN_ACTIVE;
 		rtw89_hw_scan_add_chan(rtwdev, type, req->n_ssids, ch_info);
 
-		if (rtwvif->net_type != RTW89_NET_TYPE_NO_LINK &&
+		if (connected &&
 		    off_chan_time + ch_info->period > RTW89_OFF_CHAN_TIME) {
 			tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
 			if (!tmp) {
@@ -3276,7 +3276,7 @@ out:
 }
 
 static int rtw89_hw_scan_prehandle(struct rtw89_dev *rtwdev,
-				   struct rtw89_vif *rtwvif)
+				   struct rtw89_vif *rtwvif, bool connected)
 {
 	int ret;
 
@@ -3285,7 +3285,7 @@ static int rtw89_hw_scan_prehandle(struct rtw89_dev *rtwdev,
 		rtw89_err(rtwdev, "Update probe request failed\n");
 		goto out;
 	}
-	ret = rtw89_hw_scan_add_chan_list(rtwdev, rtwvif);
+	ret = rtw89_hw_scan_add_chan_list(rtwdev, rtwvif, connected);
 out:
 	return ret;
 }
@@ -3363,16 +3363,19 @@ int rtw89_hw_scan_offload(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 {
 	struct rtw89_scan_option opt = {0};
 	struct rtw89_vif *rtwvif;
+	bool connected;
 	int ret = 0;
 
 	rtwvif = vif ? (struct rtw89_vif *)vif->drv_priv : NULL;
 	if (!rtwvif)
 		return -EINVAL;
 
+	/* This variable implies connected or during attempt to connect */
+	connected = !is_zero_ether_addr(rtwvif->bssid);
 	opt.enable = enable;
-	opt.target_ch_mode = rtwvif->net_type != RTW89_NET_TYPE_NO_LINK;
+	opt.target_ch_mode = connected;
 	if (enable) {
-		ret = rtw89_hw_scan_prehandle(rtwdev, rtwvif);
+		ret = rtw89_hw_scan_prehandle(rtwdev, rtwvif, connected);
 		if (ret)
 			goto out;
 	}
-- 
cgit v1.2.3-70-g09d2


From 639ec6d63588b7e7c2dfae1df447daf243fb8342 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Mon, 20 Mar 2023 21:06:04 +0800
Subject: wifi: rtw89: fw: use generic flow to set/check features

In early feature bitmap obtained from rtw89_early_fw_feature_recognize(),
the bits needed to check get increased. It's more friendly to work with
RTW89_CHK_FW_FEATURE(). So, we concentrate the flow of iterating FW feature
configures and calling RTW89_SET_FW_FEATURE() for various uses. And then,
we adjust rtw89_early_fw_feature_recognize() for RTW89_CHK_FW_FEATURE().

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320130606.20777-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c |  8 +++---
 drivers/net/wireless/realtek/rtw89/fw.c   | 41 +++++++++++++++----------------
 drivers/net/wireless/realtek/rtw89/fw.h   |  2 +-
 3 files changed, 25 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 72124d6c2968..4fd3416bc2d4 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3754,23 +3754,23 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 					   u32 bus_data_size,
 					   const struct rtw89_chip_info *chip)
 {
+	struct rtw89_fw_info early_fw = {};
 	const struct firmware *firmware;
 	struct ieee80211_hw *hw;
 	struct rtw89_dev *rtwdev;
 	struct ieee80211_ops *ops;
 	u32 driver_data_size;
-	u32 early_feat_map = 0;
 	bool no_chanctx;
 
-	firmware = rtw89_early_fw_feature_recognize(device, chip, &early_feat_map);
+	firmware = rtw89_early_fw_feature_recognize(device, chip, &early_fw);
 
 	ops = kmemdup(&rtw89_ops, sizeof(rtw89_ops), GFP_KERNEL);
 	if (!ops)
 		goto err;
 
 	no_chanctx = chip->support_chanctx_num == 0 ||
-		     !(early_feat_map & BIT(RTW89_FW_FEATURE_SCAN_OFFLOAD)) ||
-		     !(early_feat_map & BIT(RTW89_FW_FEATURE_BEACON_FILTER));
+		     !RTW89_CHK_FW_FEATURE(SCAN_OFFLOAD, &early_fw) ||
+		     !RTW89_CHK_FW_FEATURE(BEACON_FILTER, &early_fw);
 
 	if (no_chanctx) {
 		ops->add_chanctx = NULL;
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index d3348f9bcc2a..d9bd4339a59e 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -269,38 +269,45 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 56, 10, BEACON_FILTER),
 };
 
+static void rtw89_fw_iterate_feature_cfg(struct rtw89_fw_info *fw,
+					 const struct rtw89_chip_info *chip,
+					 u32 ver_code)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(fw_feat_tbl); i++) {
+		const struct __fw_feat_cfg *ent = &fw_feat_tbl[i];
+
+		if (chip->chip_id != ent->chip_id)
+			continue;
+
+		if (ent->cond(ver_code, ent->ver_code))
+			RTW89_SET_FW_FEATURE(ent->feature, fw);
+	}
+}
+
 static void rtw89_fw_recognize_features(struct rtw89_dev *rtwdev)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
-	const struct __fw_feat_cfg *ent;
 	const struct rtw89_fw_suit *fw_suit;
 	u32 suit_ver_code;
-	int i;
 
 	fw_suit = rtw89_fw_suit_get(rtwdev, RTW89_FW_NORMAL);
 	suit_ver_code = RTW89_FW_SUIT_VER_CODE(fw_suit);
 
-	for (i = 0; i < ARRAY_SIZE(fw_feat_tbl); i++) {
-		ent = &fw_feat_tbl[i];
-		if (chip->chip_id != ent->chip_id)
-			continue;
-
-		if (ent->cond(suit_ver_code, ent->ver_code))
-			RTW89_SET_FW_FEATURE(ent->feature, &rtwdev->fw);
-	}
+	rtw89_fw_iterate_feature_cfg(&rtwdev->fw, chip, suit_ver_code);
 }
 
 const struct firmware *
 rtw89_early_fw_feature_recognize(struct device *device,
 				 const struct rtw89_chip_info *chip,
-				 u32 *early_feat_map)
+				 struct rtw89_fw_info *early_fw)
 {
 	union rtw89_compat_fw_hdr buf = {};
 	const struct firmware *firmware;
 	bool full_req = false;
 	u32 ver_code;
 	int ret;
-	int i;
 
 	/* If SECURITY_LOADPIN_ENFORCE is enabled, reading partial files will
 	 * be denied (-EPERM). Then, we don't get right firmware things as
@@ -329,15 +336,7 @@ rtw89_early_fw_feature_recognize(struct device *device,
 	if (!ver_code)
 		goto out;
 
-	for (i = 0; i < ARRAY_SIZE(fw_feat_tbl); i++) {
-		const struct __fw_feat_cfg *ent = &fw_feat_tbl[i];
-
-		if (chip->chip_id != ent->chip_id)
-			continue;
-
-		if (ent->cond(ver_code, ent->ver_code))
-			*early_feat_map |= BIT(ent->feature);
-	}
+	rtw89_fw_iterate_feature_cfg(early_fw, chip, ver_code);
 
 out:
 	if (full_req)
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index 4d8d961f7b23..c5c7279ccc23 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -3658,7 +3658,7 @@ int rtw89_fw_recognize(struct rtw89_dev *rtwdev);
 const struct firmware *
 rtw89_early_fw_feature_recognize(struct device *device,
 				 const struct rtw89_chip_info *chip,
-				 u32 *early_feat_map);
+				 struct rtw89_fw_info *early_fw);
 int rtw89_fw_download(struct rtw89_dev *rtwdev, enum rtw89_fw_type type);
 int rtw89_load_firmware(struct rtw89_dev *rtwdev);
 void rtw89_unload_firmware(struct rtw89_dev *rtwdev);
-- 
cgit v1.2.3-70-g09d2


From b80ad23a8f2e0b63384cadc0aa2c1135b1dc03eb Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Mon, 20 Mar 2023 21:06:05 +0800
Subject: wifi: rtw89: use schedule_work to request firmware

Since we are going to load more than one firmware and some are not
presented or optional, using asynchronous API request_firmware_nowait()
will become complicated. Also, we want to use firmware_request_nowarn()
to avoid warning messages when loading optional files. So, use
schedule_work to be simpler.

To abstract loading a firmware or file, define a struct rtw89_fw_req_info
containing a struct firmware and a completion to ensure this firmware is
loaded completely.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320130606.20777-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 17 ++++----
 drivers/net/wireless/realtek/rtw89/core.h |  8 +++-
 drivers/net/wireless/realtek/rtw89/fw.c   | 68 ++++++++++++++-----------------
 drivers/net/wireless/realtek/rtw89/fw.h   |  2 +-
 4 files changed, 45 insertions(+), 50 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 4fd3416bc2d4..8e79064c3a7b 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3423,7 +3423,6 @@ void rtw89_core_stop(struct rtw89_dev *rtwdev)
 int rtw89_core_init(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
-	int ret;
 	u8 band;
 
 	INIT_LIST_HEAD(&rtwdev->ba_list);
@@ -3457,6 +3456,8 @@ int rtw89_core_init(struct rtw89_dev *rtwdev)
 
 	INIT_WORK(&rtwdev->c2h_work, rtw89_fw_c2h_work);
 	INIT_WORK(&rtwdev->ips_work, rtw89_ips_work);
+	INIT_WORK(&rtwdev->load_firmware_work, rtw89_load_firmware_work);
+
 	skb_queue_head_init(&rtwdev->c2h_queue);
 	rtw89_core_ppdu_sts_init(rtwdev);
 	rtw89_traffic_stats_init(rtwdev, &rtwdev->stats);
@@ -3468,12 +3469,10 @@ int rtw89_core_init(struct rtw89_dev *rtwdev)
 	INIT_WORK(&btc->dhcp_notify_work, rtw89_btc_ntfy_dhcp_packet_work);
 	INIT_WORK(&btc->icmp_notify_work, rtw89_btc_ntfy_icmp_packet_work);
 
-	ret = rtw89_load_firmware(rtwdev);
-	if (ret) {
-		rtw89_warn(rtwdev, "no firmware loaded\n");
-		destroy_workqueue(rtwdev->txq_wq);
-		return ret;
-	}
+	init_completion(&rtwdev->fw.req.completion);
+
+	schedule_work(&rtwdev->load_firmware_work);
+
 	rtw89_ser_init(rtwdev);
 	rtw89_entity_init(rtwdev);
 
@@ -3792,7 +3791,7 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 	rtwdev->dev = device;
 	rtwdev->ops = ops;
 	rtwdev->chip = chip;
-	rtwdev->fw.firmware = firmware;
+	rtwdev->fw.req.firmware = firmware;
 
 	rtw89_debug(rtwdev, RTW89_DBG_FW, "probe driver %s chanctx\n",
 		    no_chanctx ? "without" : "with");
@@ -3809,7 +3808,7 @@ EXPORT_SYMBOL(rtw89_alloc_ieee80211_hw);
 void rtw89_free_ieee80211_hw(struct rtw89_dev *rtwdev)
 {
 	kfree(rtwdev->ops);
-	release_firmware(rtwdev->fw.firmware);
+	release_firmware(rtwdev->fw.req.firmware);
 	ieee80211_free_hw(rtwdev->hw);
 }
 EXPORT_SYMBOL(rtw89_free_ieee80211_hw);
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 39083a0e6ce3..bd5578ca1636 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3297,10 +3297,13 @@ struct rtw89_fw_suit {
 			  GET_FW_HDR_SUBVERSION(fw_hdr),	\
 			  GET_FW_HDR_SUBINDEX(fw_hdr))
 
-struct rtw89_fw_info {
+struct rtw89_fw_req_info {
 	const struct firmware *firmware;
-	struct rtw89_dev *rtwdev;
 	struct completion completion;
+};
+
+struct rtw89_fw_info {
+	struct rtw89_fw_req_info req;
 	u8 h2c_seq;
 	u8 rec_seq;
 	u8 h2c_counter;
@@ -4018,6 +4021,7 @@ struct rtw89_dev {
 	struct sk_buff_head c2h_queue;
 	struct work_struct c2h_work;
 	struct work_struct ips_work;
+	struct work_struct load_firmware_work;
 
 	struct list_head early_h2c_list;
 
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index d9bd4339a59e..9edc421e176c 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -155,8 +155,9 @@ int rtw89_mfw_recognize(struct rtw89_dev *rtwdev, enum rtw89_fw_type type,
 			struct rtw89_fw_suit *fw_suit, bool nowarn)
 {
 	struct rtw89_fw_info *fw_info = &rtwdev->fw;
-	const u8 *mfw = fw_info->firmware->data;
-	u32 mfw_len = fw_info->firmware->size;
+	const struct firmware *firmware = fw_info->req.firmware;
+	const u8 *mfw = firmware->data;
+	u32 mfw_len = firmware->size;
 	const struct rtw89_mfw_hdr *mfw_hdr = (const struct rtw89_mfw_hdr *)mfw;
 	const struct rtw89_mfw_info *mfw_info;
 	int i;
@@ -631,67 +632,58 @@ int rtw89_wait_firmware_completion(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_fw_info *fw = &rtwdev->fw;
 
-	wait_for_completion(&fw->completion);
-	if (!fw->firmware)
+	wait_for_completion(&fw->req.completion);
+	if (!fw->req.firmware)
 		return -EINVAL;
 
 	return 0;
 }
 
-static void rtw89_load_firmware_cb(const struct firmware *firmware, void *context)
+static int rtw89_load_firmware_req(struct rtw89_dev *rtwdev,
+				   struct rtw89_fw_req_info *req,
+				   const char *fw_name, bool nowarn)
 {
-	struct rtw89_fw_info *fw = context;
-	struct rtw89_dev *rtwdev = fw->rtwdev;
-
-	if (!firmware || !firmware->data) {
-		rtw89_err(rtwdev, "failed to request firmware\n");
-		complete_all(&fw->completion);
-		return;
-	}
-
-	fw->firmware = firmware;
-	complete_all(&fw->completion);
-}
-
-int rtw89_load_firmware(struct rtw89_dev *rtwdev)
-{
-	struct rtw89_fw_info *fw = &rtwdev->fw;
-	const char *fw_name = rtwdev->chip->fw_name;
 	int ret;
 
-	fw->rtwdev = rtwdev;
-	init_completion(&fw->completion);
-
-	if (fw->firmware) {
+	if (req->firmware) {
 		rtw89_debug(rtwdev, RTW89_DBG_FW,
 			    "full firmware has been early requested\n");
-		complete_all(&fw->completion);
+		complete_all(&req->completion);
 		return 0;
 	}
 
-	ret = request_firmware_nowait(THIS_MODULE, true, fw_name, rtwdev->dev,
-				      GFP_KERNEL, fw, rtw89_load_firmware_cb);
-	if (ret) {
-		rtw89_err(rtwdev, "failed to async firmware request\n");
-		return ret;
-	}
+	if (nowarn)
+		ret = firmware_request_nowarn(&req->firmware, fw_name, rtwdev->dev);
+	else
+		ret = request_firmware(&req->firmware, fw_name, rtwdev->dev);
 
-	return 0;
+	complete_all(&req->completion);
+
+	return ret;
+}
+
+void rtw89_load_firmware_work(struct work_struct *work)
+{
+	struct rtw89_dev *rtwdev =
+		container_of(work, struct rtw89_dev, load_firmware_work);
+	const char *fw_name = rtwdev->chip->fw_name;
+
+	rtw89_load_firmware_req(rtwdev, &rtwdev->fw.req, fw_name, false);
 }
 
 void rtw89_unload_firmware(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_fw_info *fw = &rtwdev->fw;
 
-	rtw89_wait_firmware_completion(rtwdev);
+	cancel_work_sync(&rtwdev->load_firmware_work);
 
-	if (fw->firmware) {
-		release_firmware(fw->firmware);
+	if (fw->req.firmware) {
+		release_firmware(fw->req.firmware);
 
 		/* assign NULL back in case rtw89_free_ieee80211_hw()
 		 * try to release the same one again.
 		 */
-		fw->firmware = NULL;
+		fw->req.firmware = NULL;
 	}
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index c5c7279ccc23..661e08ae7bc8 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -3660,7 +3660,7 @@ rtw89_early_fw_feature_recognize(struct device *device,
 				 const struct rtw89_chip_info *chip,
 				 struct rtw89_fw_info *early_fw);
 int rtw89_fw_download(struct rtw89_dev *rtwdev, enum rtw89_fw_type type);
-int rtw89_load_firmware(struct rtw89_dev *rtwdev);
+void rtw89_load_firmware_work(struct work_struct *work);
 void rtw89_unload_firmware(struct rtw89_dev *rtwdev);
 int rtw89_wait_firmware_completion(struct rtw89_dev *rtwdev);
 void rtw89_h2c_pkt_set_hdr(struct rtw89_dev *rtwdev, struct sk_buff *skb,
-- 
cgit v1.2.3-70-g09d2


From ffde7f3476a6dfd5856dc2af207dd0f950b10122 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Mon, 20 Mar 2023 21:06:06 +0800
Subject: wifi: rtw89: add firmware format version to backward compatible with
 older drivers

In the discuss threads [1] [2], new firmware format break user space
because older drivers can't recognize new firmware format. To avoid this,
the new format will be named rtw89/rtw8852b_fw-1.bin and only new driver
try to load it. Old drivers only load original and understandable firmware
rtw89/rtw8852b_fw.bin.

More, new driver will be still backward compatible with old firmware, so
original firmware can be used by new driver.

If there is newer firmware format is introduced, rtw89/rtw8852b_fw-2.bin
will be given. The same rules will be applied like above. So, we will have
firmware like below in linux-firmware in the future.

  rtw89/rtw8852b_fw-2.bin
  rtw89/rtw8852b_fw-1.bin
  rtw89/rtw8852b_fw.bin

After this patch, MODULE_FIRMWARE() of 8852A/B/C become
  rtw89/rtw8852a_fw.bin
  rtw89/rtw8852b_fw-1.bin
  rtw89/rtw8852c_fw.bin

[1] https://lore.kernel.org/linux-wireless/df1ce994-3368-a57e-7078-8bdcccf4a1fd@gmail.com/T/#m24cb43be31a762d0ea70bf07f27ae96c59f6931b
[2] https://bugzilla.kernel.org/show_bug.cgi?id=217207

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230320130606.20777-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c     |  4 +++-
 drivers/net/wireless/realtek/rtw89/core.h     |  4 +++-
 drivers/net/wireless/realtek/rtw89/fw.c       | 33 ++++++++++++++++++++-------
 drivers/net/wireless/realtek/rtw89/fw.h       | 12 +++++++++-
 drivers/net/wireless/realtek/rtw89/rtw8852a.c | 10 ++++++--
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 10 ++++++--
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 10 ++++++--
 7 files changed, 66 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 8e79064c3a7b..79bce458d3a2 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3759,9 +3759,10 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 	struct rtw89_dev *rtwdev;
 	struct ieee80211_ops *ops;
 	u32 driver_data_size;
+	int fw_format = -1;
 	bool no_chanctx;
 
-	firmware = rtw89_early_fw_feature_recognize(device, chip, &early_fw);
+	firmware = rtw89_early_fw_feature_recognize(device, chip, &early_fw, &fw_format);
 
 	ops = kmemdup(&rtw89_ops, sizeof(rtw89_ops), GFP_KERNEL);
 	if (!ops)
@@ -3792,6 +3793,7 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 	rtwdev->ops = ops;
 	rtwdev->chip = chip;
 	rtwdev->fw.req.firmware = firmware;
+	rtwdev->fw.fw_format = fw_format;
 
 	rtw89_debug(rtwdev, RTW89_DBG_FW, "probe driver %s chanctx\n",
 		    no_chanctx ? "without" : "with");
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index bd5578ca1636..be937a420c70 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3080,7 +3080,8 @@ struct rtw89_phy_ul_tb_info {
 struct rtw89_chip_info {
 	enum rtw89_core_chip_id chip_id;
 	const struct rtw89_chip_ops *ops;
-	const char *fw_name;
+	const char *fw_basename;
+	u8 fw_format_max;
 	bool try_ce_fw;
 	u32 fifo_size;
 	u32 dle_scc_rsvd_size;
@@ -3304,6 +3305,7 @@ struct rtw89_fw_req_info {
 
 struct rtw89_fw_info {
 	struct rtw89_fw_req_info req;
+	int fw_format;
 	u8 h2c_seq;
 	u8 rec_seq;
 	u8 h2c_counter;
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 9edc421e176c..239dfb33eed5 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -302,11 +302,14 @@ static void rtw89_fw_recognize_features(struct rtw89_dev *rtwdev)
 const struct firmware *
 rtw89_early_fw_feature_recognize(struct device *device,
 				 const struct rtw89_chip_info *chip,
-				 struct rtw89_fw_info *early_fw)
+				 struct rtw89_fw_info *early_fw,
+				 int *used_fw_format)
 {
 	union rtw89_compat_fw_hdr buf = {};
 	const struct firmware *firmware;
 	bool full_req = false;
+	char fw_name[64];
+	int fw_format;
 	u32 ver_code;
 	int ret;
 
@@ -317,12 +320,22 @@ rtw89_early_fw_feature_recognize(struct device *device,
 	if (IS_ENABLED(CONFIG_SECURITY_LOADPIN_ENFORCE))
 		full_req = true;
 
-	if (full_req)
-		ret = request_firmware(&firmware, chip->fw_name, device);
-	else
-		ret = request_partial_firmware_into_buf(&firmware, chip->fw_name,
-							device, &buf, sizeof(buf),
-							0);
+	for (fw_format = chip->fw_format_max; fw_format >= 0; fw_format--) {
+		rtw89_fw_get_filename(fw_name, sizeof(fw_name),
+				      chip->fw_basename, fw_format);
+
+		if (full_req)
+			ret = request_firmware(&firmware, fw_name, device);
+		else
+			ret = request_partial_firmware_into_buf(&firmware, fw_name,
+								device, &buf, sizeof(buf),
+								0);
+		if (!ret) {
+			dev_info(device, "loaded firmware %s\n", fw_name);
+			*used_fw_format = fw_format;
+			break;
+		}
+	}
 
 	if (ret) {
 		dev_err(device, "failed to early request firmware: %d\n", ret);
@@ -666,7 +679,11 @@ void rtw89_load_firmware_work(struct work_struct *work)
 {
 	struct rtw89_dev *rtwdev =
 		container_of(work, struct rtw89_dev, load_firmware_work);
-	const char *fw_name = rtwdev->chip->fw_name;
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+	char fw_name[64];
+
+	rtw89_fw_get_filename(fw_name, sizeof(fw_name),
+			      chip->fw_basename, rtwdev->fw.fw_format);
 
 	rtw89_load_firmware_req(rtwdev, &rtwdev->fw.req, fw_name, false);
 }
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index 661e08ae7bc8..a2ff173e5544 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -3515,6 +3515,15 @@ static inline u32 rtw89_compat_fw_hdr_ver_code(const void *fw_buf)
 		return RTW89_FW_HDR_VER_CODE(&compat->fw_hdr);
 }
 
+static inline void rtw89_fw_get_filename(char *buf, size_t size,
+					 const char *fw_basename, int fw_format)
+{
+	if (fw_format <= 0)
+		snprintf(buf, size, "%s.bin", fw_basename);
+	else
+		snprintf(buf, size, "%s-%d.bin", fw_basename, fw_format);
+}
+
 #define RTW89_H2C_RF_PAGE_SIZE 500
 #define RTW89_H2C_RF_PAGE_NUM 3
 struct rtw89_fw_h2c_rf_reg_info {
@@ -3658,7 +3667,8 @@ int rtw89_fw_recognize(struct rtw89_dev *rtwdev);
 const struct firmware *
 rtw89_early_fw_feature_recognize(struct device *device,
 				 const struct rtw89_chip_info *chip,
-				 struct rtw89_fw_info *early_fw);
+				 struct rtw89_fw_info *early_fw,
+				 int *used_fw_format);
 int rtw89_fw_download(struct rtw89_dev *rtwdev, enum rtw89_fw_type type);
 void rtw89_load_firmware_work(struct work_struct *work);
 void rtw89_unload_firmware(struct rtw89_dev *rtwdev);
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index 46907748778b..caaaa9f37d6b 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -12,6 +12,11 @@
 #include "rtw8852a_table.h"
 #include "txrx.h"
 
+#define RTW8852A_FW_FORMAT_MAX 0
+#define RTW8852A_FW_BASENAME "rtw89/rtw8852a_fw"
+#define RTW8852A_MODULE_FIRMWARE \
+	RTW8852A_FW_BASENAME ".bin"
+
 static const struct rtw89_hfc_ch_cfg rtw8852a_hfc_chcfg_pcie[] = {
 	{128, 1896, grp_0}, /* ACH 0 */
 	{128, 1896, grp_0}, /* ACH 1 */
@@ -2059,7 +2064,8 @@ static const struct rtw89_chip_ops rtw8852a_chip_ops = {
 const struct rtw89_chip_info rtw8852a_chip_info = {
 	.chip_id		= RTL8852A,
 	.ops			= &rtw8852a_chip_ops,
-	.fw_name		= "rtw89/rtw8852a_fw.bin",
+	.fw_basename		= RTW8852A_FW_BASENAME,
+	.fw_format_max		= RTW8852A_FW_FORMAT_MAX,
 	.try_ce_fw		= false,
 	.fifo_size		= 458752,
 	.dle_scc_rsvd_size	= 0,
@@ -2156,7 +2162,7 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 };
 EXPORT_SYMBOL(rtw8852a_chip_info);
 
-MODULE_FIRMWARE("rtw89/rtw8852a_fw.bin");
+MODULE_FIRMWARE(RTW8852A_MODULE_FIRMWARE);
 MODULE_AUTHOR("Realtek Corporation");
 MODULE_DESCRIPTION("Realtek 802.11ax wireless 8852A driver");
 MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index bae80984cfd5..6a3a28d729db 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -12,6 +12,11 @@
 #include "rtw8852b_table.h"
 #include "txrx.h"
 
+#define RTW8852B_FW_FORMAT_MAX 1
+#define RTW8852B_FW_BASENAME "rtw89/rtw8852b_fw"
+#define RTW8852B_MODULE_FIRMWARE \
+	RTW8852B_FW_BASENAME "-" __stringify(RTW8852B_FW_FORMAT_MAX) ".bin"
+
 static const struct rtw89_hfc_ch_cfg rtw8852b_hfc_chcfg_pcie[] = {
 	{5, 343, grp_0}, /* ACH 0 */
 	{5, 343, grp_0}, /* ACH 1 */
@@ -2480,7 +2485,8 @@ static const struct rtw89_chip_ops rtw8852b_chip_ops = {
 const struct rtw89_chip_info rtw8852b_chip_info = {
 	.chip_id		= RTL8852B,
 	.ops			= &rtw8852b_chip_ops,
-	.fw_name		= "rtw89/rtw8852b_fw.bin",
+	.fw_basename		= RTW8852B_FW_BASENAME,
+	.fw_format_max		= RTW8852B_FW_FORMAT_MAX,
 	.try_ce_fw		= true,
 	.fifo_size		= 196608,
 	.dle_scc_rsvd_size	= 98304,
@@ -2576,7 +2582,7 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 };
 EXPORT_SYMBOL(rtw8852b_chip_info);
 
-MODULE_FIRMWARE("rtw89/rtw8852b_fw.bin");
+MODULE_FIRMWARE(RTW8852B_MODULE_FIRMWARE);
 MODULE_AUTHOR("Realtek Corporation");
 MODULE_DESCRIPTION("Realtek 802.11ax wireless 8852B driver");
 MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index ba728fca9188..c5147d965160 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -13,6 +13,11 @@
 #include "rtw8852c_table.h"
 #include "util.h"
 
+#define RTW8852C_FW_FORMAT_MAX 0
+#define RTW8852C_FW_BASENAME "rtw89/rtw8852c_fw"
+#define RTW8852C_MODULE_FIRMWARE \
+	RTW8852C_FW_BASENAME ".bin"
+
 static const struct rtw89_hfc_ch_cfg rtw8852c_hfc_chcfg_pcie[] = {
 	{13, 1614, grp_0}, /* ACH 0 */
 	{13, 1614, grp_0}, /* ACH 1 */
@@ -2791,7 +2796,8 @@ static const struct rtw89_chip_ops rtw8852c_chip_ops = {
 const struct rtw89_chip_info rtw8852c_chip_info = {
 	.chip_id		= RTL8852C,
 	.ops			= &rtw8852c_chip_ops,
-	.fw_name		= "rtw89/rtw8852c_fw.bin",
+	.fw_basename		= RTW8852C_FW_BASENAME,
+	.fw_format_max		= RTW8852C_FW_FORMAT_MAX,
 	.try_ce_fw		= false,
 	.fifo_size		= 458752,
 	.dle_scc_rsvd_size	= 0,
@@ -2892,7 +2898,7 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 };
 EXPORT_SYMBOL(rtw8852c_chip_info);
 
-MODULE_FIRMWARE("rtw89/rtw8852c_fw.bin");
+MODULE_FIRMWARE(RTW8852C_MODULE_FIRMWARE);
 MODULE_AUTHOR("Realtek Corporation");
 MODULE_DESCRIPTION("Realtek 802.11ax wireless 8852C driver");
 MODULE_LICENSE("Dual BSD/GPL");
-- 
cgit v1.2.3-70-g09d2


From 5395482afabb61cc980c9e78f013dd31cf212568 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Thu, 30 Mar 2023 16:03:31 +0800
Subject: wifi: rtw89: support parameter tables by RFE type

One chip can have different RFE (RF front end) types which we will judge
at runtime. And, different RFE types may use different RF parameter tables.
Though we didn't really meet this case previously, we are going to meet it
on upcoming chip RTL8851B. So, this commit handles parameter tables for
runtime RFE type.

We now encapsulate rtw89_txpwr_rule_<2/5/6>ghz tables into rtw89_rfe_parms.
Then, each chip defines its default parameter tables, and if needed, it can
configure extra parameter tables by RFE type. Finally we determine runtime
parameter tables by RFE type if one is configured. Otherwise, we use the
default parameter tables.

For now, we just move all settings under default parameter tables. We will
configure parameter tables by RFE types in separate commits afterwards.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330080331.37155-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c          | 23 ++++++++
 drivers/net/wireless/realtek/rtw89/core.h          | 55 ++++++++++++++-----
 drivers/net/wireless/realtek/rtw89/phy.c           | 64 +++++++++++++---------
 drivers/net/wireless/realtek/rtw89/rtw8852a.c      |  6 +-
 .../net/wireless/realtek/rtw89/rtw8852a_table.c    | 15 +++++
 .../net/wireless/realtek/rtw89/rtw8852a_table.h    | 11 +---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c      |  6 +-
 .../net/wireless/realtek/rtw89/rtw8852b_table.c    | 15 +++++
 .../net/wireless/realtek/rtw89/rtw8852b_table.h    | 11 +---
 drivers/net/wireless/realtek/rtw89/rtw8852c.c      |  8 +--
 .../net/wireless/realtek/rtw89/rtw8852c_table.c    | 21 +++++++
 .../net/wireless/realtek/rtw89/rtw8852c_table.h    | 16 +-----
 12 files changed, 161 insertions(+), 90 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 79bce458d3a2..d91876eb2520 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3557,6 +3557,28 @@ static void rtw89_core_setup_phycap(struct rtw89_dev *rtwdev)
 		rtwdev->chip->chip_id == RTL8852A && rtwdev->hal.cv <= CHIP_CBV;
 }
 
+static void rtw89_core_setup_rfe_parms(struct rtw89_dev *rtwdev)
+{
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+	const struct rtw89_rfe_parms_conf *conf = chip->rfe_parms_conf;
+	struct rtw89_efuse *efuse = &rtwdev->efuse;
+	u8 rfe_type = efuse->rfe_type;
+
+	if (!conf)
+		goto out;
+
+	while (conf->rfe_parms) {
+		if (rfe_type == conf->rfe_type) {
+			rtwdev->rfe_parms = conf->rfe_parms;
+			return;
+		}
+		conf++;
+	}
+
+out:
+	rtwdev->rfe_parms = chip->dflt_parms;
+}
+
 static int rtw89_chip_efuse_info_setup(struct rtw89_dev *rtwdev)
 {
 	int ret;
@@ -3578,6 +3600,7 @@ static int rtw89_chip_efuse_info_setup(struct rtw89_dev *rtwdev)
 		return ret;
 
 	rtw89_core_setup_phycap(rtwdev);
+	rtw89_core_setup_rfe_parms(rtwdev);
 
 	rtw89_mac_pwr_off(rtwdev);
 
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index be937a420c70..e0c8b3a4c43f 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -2990,6 +2990,41 @@ struct rtw89_txpwr_table {
 		     const struct rtw89_txpwr_table *tbl);
 };
 
+struct rtw89_txpwr_rule_2ghz {
+	const s8 (*lmt)[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
+		       [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+		       [RTW89_REGD_NUM][RTW89_2G_CH_NUM];
+	const s8 (*lmt_ru)[RTW89_RU_NUM][RTW89_NTX_NUM]
+			  [RTW89_REGD_NUM][RTW89_2G_CH_NUM];
+};
+
+struct rtw89_txpwr_rule_5ghz {
+	const s8 (*lmt)[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
+		       [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+		       [RTW89_REGD_NUM][RTW89_5G_CH_NUM];
+	const s8 (*lmt_ru)[RTW89_RU_NUM][RTW89_NTX_NUM]
+			  [RTW89_REGD_NUM][RTW89_5G_CH_NUM];
+};
+
+struct rtw89_txpwr_rule_6ghz {
+	const s8 (*lmt)[RTW89_6G_BW_NUM][RTW89_NTX_NUM]
+		       [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+		       [RTW89_REGD_NUM][RTW89_6G_CH_NUM];
+	const s8 (*lmt_ru)[RTW89_RU_NUM][RTW89_NTX_NUM]
+			  [RTW89_REGD_NUM][RTW89_6G_CH_NUM];
+};
+
+struct rtw89_rfe_parms {
+	struct rtw89_txpwr_rule_2ghz rule_2ghz;
+	struct rtw89_txpwr_rule_5ghz rule_5ghz;
+	struct rtw89_txpwr_rule_6ghz rule_6ghz;
+};
+
+struct rtw89_rfe_parms_conf {
+	const struct rtw89_rfe_parms *rfe_parms;
+	u8 rfe_type;
+};
+
 struct rtw89_page_regs {
 	u32 hci_fc_ctrl;
 	u32 ch_page_ctrl;
@@ -3127,21 +3162,10 @@ struct rtw89_chip_info {
 	const struct rtw89_phy_dig_gain_table *dig_table;
 	const struct rtw89_dig_regs *dig_regs;
 	const struct rtw89_phy_tssi_dbw_table *tssi_dbw_table;
-	const s8 (*txpwr_lmt_2g)[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
-				[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-				[RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-	const s8 (*txpwr_lmt_5g)[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
-				[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-				[RTW89_REGD_NUM][RTW89_5G_CH_NUM];
-	const s8 (*txpwr_lmt_6g)[RTW89_6G_BW_NUM][RTW89_NTX_NUM]
-				[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-				[RTW89_REGD_NUM][RTW89_6G_CH_NUM];
-	const s8 (*txpwr_lmt_ru_2g)[RTW89_RU_NUM][RTW89_NTX_NUM]
-				   [RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-	const s8 (*txpwr_lmt_ru_5g)[RTW89_RU_NUM][RTW89_NTX_NUM]
-				   [RTW89_REGD_NUM][RTW89_5G_CH_NUM];
-	const s8 (*txpwr_lmt_ru_6g)[RTW89_RU_NUM][RTW89_NTX_NUM]
-				   [RTW89_REGD_NUM][RTW89_6G_CH_NUM];
+
+	/* NULL if no rfe-specific, or a null-terminated array by rfe_parms */
+	const struct rtw89_rfe_parms_conf *rfe_parms_conf;
+	const struct rtw89_rfe_parms *dflt_parms;
 
 	u8 txpwr_factor_rf;
 	u8 txpwr_factor_mac;
@@ -3992,6 +4016,7 @@ struct rtw89_dev {
 	struct rtw89_hw_scan_info scan_info;
 	const struct rtw89_chip_info *chip;
 	const struct rtw89_pci_info *pci_info;
+	const struct rtw89_rfe_parms *rfe_parms;
 	struct rtw89_hal hal;
 	struct rtw89_mcc_info mcc;
 	struct rtw89_mac_info mac;
diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index cb0f6cc51d6b..53f0d964b7ae 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -1617,29 +1617,35 @@ static u8 rtw89_channel_to_idx(struct rtw89_dev *rtwdev, u8 band, u8 channel)
 s8 rtw89_phy_read_txpwr_limit(struct rtw89_dev *rtwdev, u8 band,
 			      u8 bw, u8 ntx, u8 rs, u8 bf, u8 ch)
 {
-	const struct rtw89_chip_info *chip = rtwdev->chip;
+	const struct rtw89_rfe_parms *rfe_parms = rtwdev->rfe_parms;
+	const struct rtw89_txpwr_rule_2ghz *rule_2ghz = &rfe_parms->rule_2ghz;
+	const struct rtw89_txpwr_rule_5ghz *rule_5ghz = &rfe_parms->rule_5ghz;
+	const struct rtw89_txpwr_rule_6ghz *rule_6ghz = &rfe_parms->rule_6ghz;
 	u8 ch_idx = rtw89_channel_to_idx(rtwdev, band, ch);
 	u8 regd = rtw89_regd_get(rtwdev, band);
 	s8 lmt = 0, sar;
 
 	switch (band) {
 	case RTW89_BAND_2G:
-		lmt = (*chip->txpwr_lmt_2g)[bw][ntx][rs][bf][regd][ch_idx];
-		if (!lmt)
-			lmt = (*chip->txpwr_lmt_2g)[bw][ntx][rs][bf]
-						   [RTW89_WW][ch_idx];
+		lmt = (*rule_2ghz->lmt)[bw][ntx][rs][bf][regd][ch_idx];
+		if (lmt)
+			break;
+
+		lmt = (*rule_2ghz->lmt)[bw][ntx][rs][bf][RTW89_WW][ch_idx];
 		break;
 	case RTW89_BAND_5G:
-		lmt = (*chip->txpwr_lmt_5g)[bw][ntx][rs][bf][regd][ch_idx];
-		if (!lmt)
-			lmt = (*chip->txpwr_lmt_5g)[bw][ntx][rs][bf]
-						   [RTW89_WW][ch_idx];
+		lmt = (*rule_5ghz->lmt)[bw][ntx][rs][bf][regd][ch_idx];
+		if (lmt)
+			break;
+
+		lmt = (*rule_5ghz->lmt)[bw][ntx][rs][bf][RTW89_WW][ch_idx];
 		break;
 	case RTW89_BAND_6G:
-		lmt = (*chip->txpwr_lmt_6g)[bw][ntx][rs][bf][regd][ch_idx];
-		if (!lmt)
-			lmt = (*chip->txpwr_lmt_6g)[bw][ntx][rs][bf]
-						   [RTW89_WW][ch_idx];
+		lmt = (*rule_6ghz->lmt)[bw][ntx][rs][bf][regd][ch_idx];
+		if (lmt)
+			break;
+
+		lmt = (*rule_6ghz->lmt)[bw][ntx][rs][bf][RTW89_WW][ch_idx];
 		break;
 	default:
 		rtw89_warn(rtwdev, "unknown band type: %d\n", band);
@@ -1862,29 +1868,35 @@ void rtw89_phy_fill_txpwr_limit(struct rtw89_dev *rtwdev,
 static s8 rtw89_phy_read_txpwr_limit_ru(struct rtw89_dev *rtwdev, u8 band,
 					u8 ru, u8 ntx, u8 ch)
 {
-	const struct rtw89_chip_info *chip = rtwdev->chip;
+	const struct rtw89_rfe_parms *rfe_parms = rtwdev->rfe_parms;
+	const struct rtw89_txpwr_rule_2ghz *rule_2ghz = &rfe_parms->rule_2ghz;
+	const struct rtw89_txpwr_rule_5ghz *rule_5ghz = &rfe_parms->rule_5ghz;
+	const struct rtw89_txpwr_rule_6ghz *rule_6ghz = &rfe_parms->rule_6ghz;
 	u8 ch_idx = rtw89_channel_to_idx(rtwdev, band, ch);
 	u8 regd = rtw89_regd_get(rtwdev, band);
 	s8 lmt_ru = 0, sar;
 
 	switch (band) {
 	case RTW89_BAND_2G:
-		lmt_ru = (*chip->txpwr_lmt_ru_2g)[ru][ntx][regd][ch_idx];
-		if (!lmt_ru)
-			lmt_ru = (*chip->txpwr_lmt_ru_2g)[ru][ntx]
-							 [RTW89_WW][ch_idx];
+		lmt_ru = (*rule_2ghz->lmt_ru)[ru][ntx][regd][ch_idx];
+		if (lmt_ru)
+			break;
+
+		lmt_ru = (*rule_2ghz->lmt_ru)[ru][ntx][RTW89_WW][ch_idx];
 		break;
 	case RTW89_BAND_5G:
-		lmt_ru = (*chip->txpwr_lmt_ru_5g)[ru][ntx][regd][ch_idx];
-		if (!lmt_ru)
-			lmt_ru = (*chip->txpwr_lmt_ru_5g)[ru][ntx]
-							 [RTW89_WW][ch_idx];
+		lmt_ru = (*rule_5ghz->lmt_ru)[ru][ntx][regd][ch_idx];
+		if (lmt_ru)
+			break;
+
+		lmt_ru = (*rule_5ghz->lmt_ru)[ru][ntx][RTW89_WW][ch_idx];
 		break;
 	case RTW89_BAND_6G:
-		lmt_ru = (*chip->txpwr_lmt_ru_6g)[ru][ntx][regd][ch_idx];
-		if (!lmt_ru)
-			lmt_ru = (*chip->txpwr_lmt_ru_6g)[ru][ntx]
-							 [RTW89_WW][ch_idx];
+		lmt_ru = (*rule_6ghz->lmt_ru)[ru][ntx][regd][ch_idx];
+		if (lmt_ru)
+			break;
+
+		lmt_ru = (*rule_6ghz->lmt_ru)[ru][ntx][RTW89_WW][ch_idx];
 		break;
 	default:
 		rtw89_warn(rtwdev, "unknown band type: %d\n", band);
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index caaaa9f37d6b..499625338a64 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -2085,10 +2085,8 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 				   &rtw89_8852a_phy_radiob_table,},
 	.nctl_table		= &rtw89_8852a_phy_nctl_table,
 	.byr_table		= &rtw89_8852a_byr_table,
-	.txpwr_lmt_2g		= &rtw89_8852a_txpwr_lmt_2g,
-	.txpwr_lmt_5g		= &rtw89_8852a_txpwr_lmt_5g,
-	.txpwr_lmt_ru_2g	= &rtw89_8852a_txpwr_lmt_ru_2g,
-	.txpwr_lmt_ru_5g	= &rtw89_8852a_txpwr_lmt_ru_5g,
+	.dflt_parms		= &rtw89_8852a_dflt_parms,
+	.rfe_parms_conf		= NULL,
 	.txpwr_factor_rf	= 2,
 	.txpwr_factor_mac	= 1,
 	.dig_table		= &rtw89_8852a_phy_dig_table,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a_table.c b/drivers/net/wireless/realtek/rtw89/rtw8852a_table.c
index 320bcd4852c6..be54194558ff 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a_table.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a_table.c
@@ -43377,6 +43377,7 @@ static const s8 _txpwr_track_delta_swingidx_2g_cck_a_p[] = {
 	0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5,
 	 6, 6, 6, 7, 7, 7, 8, 8, 8, 9, 9, 9, 10, 10};
 
+static
 const s8 rtw89_8852a_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
@@ -45566,6 +45567,7 @@ const s8 rtw89_8852a_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
 	[1][1][2][1][RTW89_UK][13] = 127,
 };
 
+static
 const s8 rtw89_8852a_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
@@ -47898,6 +47900,7 @@ const s8 rtw89_8852a_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 	[2][1][2][1][RTW89_UK][41] = 40,
 };
 
+static
 const s8 rtw89_8852a_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = 32,
@@ -48994,6 +48997,7 @@ const s8 rtw89_8852a_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
 	[2][1][RTW89_UK][13] = 127,
 };
 
+static
 const s8 rtw89_8852a_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = 22,
@@ -51043,3 +51047,14 @@ const struct rtw89_phy_dig_gain_table rtw89_8852a_phy_dig_table = {
 	.cfg_lna_a = &rtw89_8852a_lna_gain_a_table,
 	.cfg_tia_a = &rtw89_8852a_tia_gain_a_table
 };
+
+const struct rtw89_rfe_parms rtw89_8852a_dflt_parms = {
+	.rule_2ghz = {
+		.lmt = &rtw89_8852a_txpwr_lmt_2g,
+		.lmt_ru = &rtw89_8852a_txpwr_lmt_ru_2g,
+	},
+	.rule_5ghz = {
+		.lmt = &rtw89_8852a_txpwr_lmt_5g,
+		.lmt_ru = &rtw89_8852a_txpwr_lmt_ru_5g,
+	},
+};
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a_table.h b/drivers/net/wireless/realtek/rtw89/rtw8852a_table.h
index 913796506286..41c379b1044d 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a_table.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a_table.h
@@ -14,15 +14,6 @@ extern const struct rtw89_phy_table rtw89_8852a_phy_nctl_table;
 extern const struct rtw89_txpwr_table rtw89_8852a_byr_table;
 extern const struct rtw89_phy_dig_gain_table rtw89_8852a_phy_dig_table;
 extern const struct rtw89_txpwr_track_cfg rtw89_8852a_trk_cfg;
-extern const s8 rtw89_8852a_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-extern const s8 rtw89_8852a_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_5G_CH_NUM];
-extern const s8 rtw89_8852a_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-extern const s8 rtw89_8852a_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_5G_CH_NUM];
+extern const struct rtw89_rfe_parms rtw89_8852a_dflt_parms;
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 6a3a28d729db..55b83aad2410 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2506,10 +2506,8 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 				   &rtw89_8852b_phy_radiob_table,},
 	.nctl_table		= &rtw89_8852b_phy_nctl_table,
 	.byr_table		= &rtw89_8852b_byr_table,
-	.txpwr_lmt_2g		= &rtw89_8852b_txpwr_lmt_2g,
-	.txpwr_lmt_5g		= &rtw89_8852b_txpwr_lmt_5g,
-	.txpwr_lmt_ru_2g	= &rtw89_8852b_txpwr_lmt_ru_2g,
-	.txpwr_lmt_ru_5g	= &rtw89_8852b_txpwr_lmt_ru_5g,
+	.dflt_parms		= &rtw89_8852b_dflt_parms,
+	.rfe_parms_conf		= NULL,
 	.txpwr_factor_rf	= 2,
 	.txpwr_factor_mac	= 1,
 	.dig_table		= NULL,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_table.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_table.c
index a6734965361f..904cdb9e56fa 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_table.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_table.c
@@ -14706,6 +14706,7 @@ const u8 rtw89_8852b_tx_shape[RTW89_BAND_MAX][RTW89_RS_TX_SHAPE_NUM]
 	[1][1][RTW89_UKRAINE] = 0,
 };
 
+static
 const s8 rtw89_8852b_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
@@ -16895,6 +16896,7 @@ const s8 rtw89_8852b_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
 	[1][1][2][1][RTW89_UK][13] = 127,
 };
 
+static
 const s8 rtw89_8852b_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
@@ -19539,6 +19541,7 @@ const s8 rtw89_8852b_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 	[2][1][2][1][RTW89_UK][49] = 127,
 };
 
+static
 const s8 rtw89_8852b_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = 32,
@@ -20635,6 +20638,7 @@ const s8 rtw89_8852b_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
 	[2][1][RTW89_UK][13] = 127,
 };
 
+static
 const s8 rtw89_8852b_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = 24,
@@ -22875,3 +22879,14 @@ const struct rtw89_txpwr_track_cfg rtw89_8852b_trk_cfg = {
 	.delta_swingidx_2g_cck_a_n = _txpwr_track_delta_swingidx_2g_cck_a_n,
 	.delta_swingidx_2g_cck_a_p = _txpwr_track_delta_swingidx_2g_cck_a_p,
 };
+
+const struct rtw89_rfe_parms rtw89_8852b_dflt_parms = {
+	.rule_2ghz = {
+		.lmt = &rtw89_8852b_txpwr_lmt_2g,
+		.lmt_ru = &rtw89_8852b_txpwr_lmt_ru_2g,
+	},
+	.rule_5ghz = {
+		.lmt = &rtw89_8852b_txpwr_lmt_5g,
+		.lmt_ru = &rtw89_8852b_txpwr_lmt_ru_5g,
+	},
+};
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_table.h b/drivers/net/wireless/realtek/rtw89/rtw8852b_table.h
index 114337ac9fb0..5f4161496a58 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_table.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_table.h
@@ -16,15 +16,6 @@ extern const struct rtw89_txpwr_table rtw89_8852b_byr_table;
 extern const struct rtw89_txpwr_track_cfg rtw89_8852b_trk_cfg;
 extern const u8 rtw89_8852b_tx_shape[RTW89_BAND_MAX][RTW89_RS_TX_SHAPE_NUM]
 				    [RTW89_REGD_NUM];
-extern const s8 rtw89_8852b_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-extern const s8 rtw89_8852b_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_5G_CH_NUM];
-extern const s8 rtw89_8852b_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-extern const s8 rtw89_8852b_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_5G_CH_NUM];
+extern const struct rtw89_rfe_parms rtw89_8852b_dflt_parms;
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index c5147d965160..2be9e89b1965 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2817,12 +2817,8 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 				   &rtw89_8852c_phy_radioa_table,},
 	.nctl_table		= &rtw89_8852c_phy_nctl_table,
 	.byr_table		= &rtw89_8852c_byr_table,
-	.txpwr_lmt_2g		= &rtw89_8852c_txpwr_lmt_2g,
-	.txpwr_lmt_5g		= &rtw89_8852c_txpwr_lmt_5g,
-	.txpwr_lmt_6g		= &rtw89_8852c_txpwr_lmt_6g,
-	.txpwr_lmt_ru_2g	= &rtw89_8852c_txpwr_lmt_ru_2g,
-	.txpwr_lmt_ru_5g	= &rtw89_8852c_txpwr_lmt_ru_5g,
-	.txpwr_lmt_ru_6g	= &rtw89_8852c_txpwr_lmt_ru_6g,
+	.dflt_parms		= &rtw89_8852c_dflt_parms,
+	.rfe_parms_conf		= NULL,
 	.txpwr_factor_rf	= 2,
 	.txpwr_factor_mac	= 1,
 	.dig_table		= NULL,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c_table.c b/drivers/net/wireless/realtek/rtw89/rtw8852c_table.c
index 96c264a057ff..7011e5a6f8fd 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c_table.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c_table.c
@@ -28590,6 +28590,7 @@ const u8 rtw89_8852c_tx_shape[RTW89_BAND_MAX][RTW89_RS_TX_SHAPE_NUM]
 	[2][1][RTW89_KCC] = 0,
 };
 
+static
 const s8 rtw89_8852c_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
@@ -30107,6 +30108,7 @@ const s8 rtw89_8852c_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
 	[1][1][2][1][RTW89_UK][13] = 127,
 };
 
+static
 const s8 rtw89_8852c_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
@@ -32020,6 +32022,7 @@ const s8 rtw89_8852c_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 	[3][1][2][1][RTW89_UK][45] = 127,
 };
 
+static
 const s8 rtw89_8852c_txpwr_lmt_6g[RTW89_6G_BW_NUM][RTW89_NTX_NUM]
 				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
 				 [RTW89_REGD_NUM][RTW89_6G_CH_NUM] = {
@@ -33977,6 +33980,7 @@ const s8 rtw89_8852c_txpwr_lmt_6g[RTW89_6G_BW_NUM][RTW89_NTX_NUM]
 	[3][1][2][1][RTW89_KCC][112] = 127,
 };
 
+static
 const s8 rtw89_8852c_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = 32,
@@ -34737,6 +34741,7 @@ const s8 rtw89_8852c_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
 	[2][1][RTW89_UK][13] = 127,
 };
 
+static
 const s8 rtw89_8852c_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = 16,
@@ -36253,6 +36258,7 @@ const s8 rtw89_8852c_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
 	[2][1][RTW89_UK][52] = 127,
 };
 
+static
 const s8 rtw89_8852c_txpwr_lmt_ru_6g[RTW89_RU_NUM][RTW89_NTX_NUM]
 				    [RTW89_REGD_NUM][RTW89_6G_CH_NUM] = {
 	[0][0][RTW89_WW][0] = -16,
@@ -37472,3 +37478,18 @@ const struct rtw89_phy_tssi_dbw_table rtw89_8852c_tssi_dbw_table = {
 	.data[RTW89_TSSI_BANDEDGE_MID] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
 	.data[RTW89_TSSI_BANDEDGE_HIGH] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
 };
+
+const struct rtw89_rfe_parms rtw89_8852c_dflt_parms = {
+	.rule_2ghz = {
+		.lmt = &rtw89_8852c_txpwr_lmt_2g,
+		.lmt_ru = &rtw89_8852c_txpwr_lmt_ru_2g,
+	},
+	.rule_5ghz = {
+		.lmt = &rtw89_8852c_txpwr_lmt_5g,
+		.lmt_ru = &rtw89_8852c_txpwr_lmt_ru_5g,
+	},
+	.rule_6ghz = {
+		.lmt = &rtw89_8852c_txpwr_lmt_6g,
+		.lmt_ru = &rtw89_8852c_txpwr_lmt_ru_6g,
+	},
+};
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c_table.h b/drivers/net/wireless/realtek/rtw89/rtw8852c_table.h
index 7d71a92e2d27..6da1849fb1fa 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c_table.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c_table.h
@@ -17,20 +17,6 @@ extern const struct rtw89_phy_tssi_dbw_table rtw89_8852c_tssi_dbw_table;
 extern const struct rtw89_txpwr_track_cfg rtw89_8852c_trk_cfg;
 extern const u8 rtw89_8852c_tx_shape[RTW89_BAND_MAX][RTW89_RS_TX_SHAPE_NUM]
 				    [RTW89_REGD_NUM];
-extern const s8 rtw89_8852c_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-extern const s8 rtw89_8852c_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_5G_CH_NUM];
-extern const s8 rtw89_8852c_txpwr_lmt_6g[RTW89_6G_BW_NUM][RTW89_NTX_NUM]
-					[RTW89_RS_LMT_NUM][RTW89_BF_NUM]
-					[RTW89_REGD_NUM][RTW89_6G_CH_NUM];
-extern const s8 rtw89_8852c_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_2G_CH_NUM];
-extern const s8 rtw89_8852c_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_5G_CH_NUM];
-extern const s8 rtw89_8852c_txpwr_lmt_ru_6g[RTW89_RU_NUM][RTW89_NTX_NUM]
-					   [RTW89_REGD_NUM][RTW89_6G_CH_NUM];
+extern const struct rtw89_rfe_parms rtw89_8852c_dflt_parms;
 
 #endif
-- 
cgit v1.2.3-70-g09d2


From 9f9882dbe2eecdb27f5f3832f215679747f94d8f Mon Sep 17 00:00:00 2001
From: Eric Huang <echuang@realtek.com>
Date: Thu, 30 Mar 2023 21:23:52 +0800
Subject: wifi: rtw89: use hardware CFO to improve performance

Turn on hardware CFO (central frequency offset) compensation based on IC
capability, and improve digital CFO compensation accuracy by using
more fixed points number.

Signed-off-by: Eric Huang <echuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330132352.13647-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h     |  3 +++
 drivers/net/wireless/realtek/rtw89/phy.c      | 28 +++++++++++++++++++--------
 drivers/net/wireless/realtek/rtw89/rtw8852a.c |  3 ++-
 drivers/net/wireless/realtek/rtw89/rtw8852b.c |  3 ++-
 drivers/net/wireless/realtek/rtw89/rtw8852c.c |  3 ++-
 5 files changed, 29 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index e0c8b3a4c43f..7d0433be5572 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3202,6 +3202,7 @@ struct rtw89_chip_info {
 	struct rtw89_reg_def c2h_counter_reg;
 	const struct rtw89_page_regs *page_regs;
 	bool cfo_src_fd;
+	bool cfo_hw_comp;
 	const struct rtw89_reg_def *dcfo_comp;
 	u8 dcfo_comp_sft;
 	const struct rtw89_imr_info *imr_info;
@@ -3684,6 +3685,8 @@ struct rtw89_cfo_tracking_info {
 	s32 cfo_avg_pre;
 	s32 cfo_avg[CFO_TRACK_MAX_USER];
 	s32 pre_cfo_avg[CFO_TRACK_MAX_USER];
+	s32 dcfo_avg;
+	s32 dcfo_avg_pre;
 	u32 packet_count;
 	u32 packet_count_pre;
 	s32 residual_cfo_acc;
diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index 53f0d964b7ae..c7e906123416 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -2417,7 +2417,6 @@ static void rtw89_dcfo_comp(struct rtw89_dev *rtwdev, s32 curr_cfo)
 	bool is_linked = rtwdev->total_sta_assoc > 0;
 	s32 cfo_avg_312;
 	s32 dcfo_comp_val;
-	u8 dcfo_comp_sft = rtwdev->chip->dcfo_comp_sft;
 	int sign;
 
 	if (!is_linked) {
@@ -2430,8 +2429,8 @@ static void rtw89_dcfo_comp(struct rtw89_dev *rtwdev, s32 curr_cfo)
 		return;
 	dcfo_comp_val = rtw89_phy_read32_mask(rtwdev, R_DCFO, B_DCFO);
 	sign = curr_cfo > 0 ? 1 : -1;
-	cfo_avg_312 = (curr_cfo << dcfo_comp_sft) / 5 + sign * dcfo_comp_val;
-	rtw89_debug(rtwdev, RTW89_DBG_CFO, "DCFO: avg_cfo=%d\n", cfo_avg_312);
+	cfo_avg_312 = curr_cfo / 625 + sign * dcfo_comp_val;
+	rtw89_debug(rtwdev, RTW89_DBG_CFO, "avg_cfo_312=%d step\n", cfo_avg_312);
 	if (rtwdev->chip->chip_id == RTL8852A && rtwdev->hal.cv == CHIP_CBV)
 		cfo_avg_312 = -cfo_avg_312;
 	rtw89_phy_set_phy_regs(rtwdev, dcfo_comp->addr, dcfo_comp->mask,
@@ -2440,9 +2439,16 @@ static void rtw89_dcfo_comp(struct rtw89_dev *rtwdev, s32 curr_cfo)
 
 static void rtw89_dcfo_comp_init(struct rtw89_dev *rtwdev)
 {
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+
 	rtw89_phy_set_phy_regs(rtwdev, R_DCFO_OPT, B_DCFO_OPT_EN, 1);
 	rtw89_phy_set_phy_regs(rtwdev, R_DCFO_WEIGHT, B_DCFO_WEIGHT_MSK, 8);
-	rtw89_write32_clr(rtwdev, R_AX_PWR_UL_CTRL2, B_AX_PWR_UL_CFO_MASK);
+
+	if (chip->cfo_hw_comp)
+		rtw89_write32_mask(rtwdev, R_AX_PWR_UL_CTRL2,
+				   B_AX_PWR_UL_CFO_MASK, 0x6);
+	else
+		rtw89_write32_clr(rtwdev, R_AX_PWR_UL_CTRL2, B_AX_PWR_UL_CFO_MASK);
 }
 
 static void rtw89_phy_cfo_init(struct rtw89_dev *rtwdev)
@@ -2512,6 +2518,7 @@ static void rtw89_phy_cfo_crystal_cap_adjust(struct rtw89_dev *rtwdev,
 
 static s32 rtw89_phy_average_cfo_calc(struct rtw89_dev *rtwdev)
 {
+	const struct rtw89_chip_info *chip = rtwdev->chip;
 	struct rtw89_cfo_tracking_info *cfo = &rtwdev->cfo_tracking;
 	s32 cfo_khz_all = 0;
 	s32 cfo_cnt_all = 0;
@@ -2528,6 +2535,8 @@ static s32 rtw89_phy_average_cfo_calc(struct rtw89_dev *rtwdev)
 		cfo_cnt_all += cfo->cfo_cnt[i];
 		cfo_all_avg = phy_div(cfo_khz_all, cfo_cnt_all);
 		cfo->pre_cfo_avg[i] = cfo->cfo_avg[i];
+		cfo->dcfo_avg = phy_div(cfo_khz_all << chip->dcfo_comp_sft,
+					cfo_cnt_all);
 	}
 	rtw89_debug(rtwdev, RTW89_DBG_CFO,
 		    "CFO track for macid = %d\n", i);
@@ -2654,7 +2663,9 @@ static void rtw89_phy_cfo_dm(struct rtw89_dev *rtwdev)
 	s32 new_cfo = 0;
 	bool x_cap_update = false;
 	u8 pre_x_cap = cfo->crystal_cap;
+	u8 dcfo_comp_sft = rtwdev->chip->dcfo_comp_sft;
 
+	cfo->dcfo_avg = 0;
 	rtw89_debug(rtwdev, RTW89_DBG_CFO, "CFO:total_sta_assoc=%d\n",
 		    rtwdev->total_sta_assoc);
 	if (rtwdev->total_sta_assoc == 0) {
@@ -2696,18 +2707,19 @@ static void rtw89_phy_cfo_dm(struct rtw89_dev *rtwdev)
 
 	rtw89_phy_cfo_crystal_cap_adjust(rtwdev, new_cfo);
 	cfo->cfo_avg_pre = new_cfo;
+	cfo->dcfo_avg_pre = cfo->dcfo_avg;
 	x_cap_update =  cfo->crystal_cap != pre_x_cap;
 	rtw89_debug(rtwdev, RTW89_DBG_CFO, "Xcap_up=%d\n", x_cap_update);
 	rtw89_debug(rtwdev, RTW89_DBG_CFO, "Xcap: D:%x C:%x->%x, ofst=%d\n",
 		    cfo->def_x_cap, pre_x_cap, cfo->crystal_cap,
 		    cfo->x_cap_ofst);
 	if (x_cap_update) {
-		if (new_cfo > 0)
-			new_cfo -= CFO_SW_COMP_FINE_TUNE;
+		if (cfo->dcfo_avg > 0)
+			cfo->dcfo_avg -= CFO_SW_COMP_FINE_TUNE << dcfo_comp_sft;
 		else
-			new_cfo += CFO_SW_COMP_FINE_TUNE;
+			cfo->dcfo_avg += CFO_SW_COMP_FINE_TUNE << dcfo_comp_sft;
 	}
-	rtw89_dcfo_comp(rtwdev, new_cfo);
+	rtw89_dcfo_comp(rtwdev, cfo->dcfo_avg);
 	rtw89_phy_cfo_statistics_reset(rtwdev);
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index 499625338a64..eb7a8340f633 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -2147,8 +2147,9 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.page_regs		= &rtw8852a_page_regs,
 	.cfo_src_fd		= false,
+	.cfo_hw_comp            = false,
 	.dcfo_comp		= &rtw8852a_dcfo_comp,
-	.dcfo_comp_sft		= 3,
+	.dcfo_comp_sft		= 10,
 	.imr_info		= &rtw8852a_imr_info,
 	.rrsr_cfgs		= &rtw8852a_rrsr_cfgs,
 	.bss_clr_map_reg	= R_BSS_CLR_MAP,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 55b83aad2410..7ac7aa997a96 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2568,8 +2568,9 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 	.c2h_regs		= rtw8852b_c2h_regs,
 	.page_regs		= &rtw8852b_page_regs,
 	.cfo_src_fd		= true,
+	.cfo_hw_comp		= true,
 	.dcfo_comp		= &rtw8852b_dcfo_comp,
-	.dcfo_comp_sft		= 3,
+	.dcfo_comp_sft		= 10,
 	.imr_info		= &rtw8852b_imr_info,
 	.rrsr_cfgs		= &rtw8852b_rrsr_cfgs,
 	.bss_clr_map_reg	= R_BSS_CLR_MAP_V1,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 2be9e89b1965..011b1957b712 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2881,8 +2881,9 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 	.c2h_regs		= rtw8852c_c2h_regs,
 	.page_regs		= &rtw8852c_page_regs,
 	.cfo_src_fd		= false,
+	.cfo_hw_comp            = false,
 	.dcfo_comp		= &rtw8852c_dcfo_comp,
-	.dcfo_comp_sft		= 5,
+	.dcfo_comp_sft		= 12,
 	.imr_info		= &rtw8852c_imr_info,
 	.rrsr_cfgs		= &rtw8852c_rrsr_cfgs,
 	.bss_clr_map_reg	= R_BSS_CLR_MAP,
-- 
cgit v1.2.3-70-g09d2


From a6fb2bb84654dde55fab94251c9119b6917d098d Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 30 Mar 2023 21:33:21 +0800
Subject: wifi: rtw89: read version of analog hardware

The chip contains digital and analog parts, and each of them has its own
version number. This is used by BT coexistence mechanism to make strategy
decision for different analog version.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330133324.19538-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 10 ++++++++++
 drivers/net/wireless/realtek/rtw89/core.h |  2 ++
 drivers/net/wireless/realtek/rtw89/mac.h  |  1 +
 3 files changed, 13 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index d91876eb2520..d6204e07a5b6 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3535,6 +3535,8 @@ void rtw89_core_scan_complete(struct rtw89_dev *rtwdev,
 static void rtw89_read_chip_ver(struct rtw89_dev *rtwdev)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
+	int ret;
+	u8 val;
 	u8 cv;
 
 	cv = rtw89_read32_mask(rtwdev, R_AX_SYS_CFG1, B_AX_CHIP_VER_MASK);
@@ -3546,6 +3548,14 @@ static void rtw89_read_chip_ver(struct rtw89_dev *rtwdev)
 	}
 
 	rtwdev->hal.cv = cv;
+
+	if (chip->chip_id == RTL8852B || chip->chip_id == RTL8851B) {
+		ret = rtw89_mac_read_xtal_si(rtwdev, XTAL_SI_CV, &val);
+		if (!ret)
+			return;
+
+		rtwdev->hal.acv = u8_get_bits(val, XTAL_SI_ACV_MASK);
+	}
 }
 
 static void rtw89_core_setup_phycap(struct rtw89_dev *rtwdev)
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 7d0433be5572..eab0abb9d213 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -108,6 +108,7 @@ enum rtw89_core_chip_id {
 	RTL8852A,
 	RTL8852B,
 	RTL8852C,
+	RTL8851B,
 };
 
 enum rtw89_cv {
@@ -3412,6 +3413,7 @@ struct rtw89_sub_entity {
 struct rtw89_hal {
 	u32 rx_fltr;
 	u8 cv;
+	u8 acv;
 	u32 sw_amsdu_max_size;
 	u32 antenna_tx;
 	u32 antenna_rx;
diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h
index 899c45c6774e..a9ea3459ed82 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.h
+++ b/drivers/net/wireless/realtek/rtw89/mac.h
@@ -1117,6 +1117,7 @@ enum rtw89_mac_xtal_si_offset {
 	XTAL_SI_XTAL_XMD_4 = 0x26,
 #define XTAL_SI_LPS_CAP		GENMASK(3, 0)
 	XTAL_SI_CV = 0x41,
+#define XTAL_SI_ACV_MASK	GENMASK(3, 0)
 	XTAL_SI_LOW_ADDR = 0x62,
 #define XTAL_SI_LOW_ADDR_MASK	GENMASK(7, 0)
 	XTAL_SI_CTRL = 0x63,
-- 
cgit v1.2.3-70-g09d2


From d5289b2d69a777d24686d7ee8264a55761dc9f93 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 30 Mar 2023 21:33:22 +0800
Subject: wifi: rtw89: 8851b: fix TX path to path A for one RF path chip

For two RF paths chips, we normally set path B as main path by default.
8851B has single one RF path, so set TX path to A and set mapping of
path B to 0.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330133324.19538-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 239dfb33eed5..c22e282849ff 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -1161,9 +1161,18 @@ fail:
 static void __rtw89_fw_h2c_set_tx_path(struct rtw89_dev *rtwdev,
 				       struct sk_buff *skb)
 {
+	const struct rtw89_chip_info *chip = rtwdev->chip;
 	struct rtw89_hal *hal = &rtwdev->hal;
-	u8 ntx_path = hal->antenna_tx ? hal->antenna_tx : RF_B;
-	u8 map_b = hal->antenna_tx == RF_AB ? 1 : 0;
+	u8 ntx_path;
+	u8 map_b;
+
+	if (chip->rf_path_num == 1) {
+		ntx_path = RF_A;
+		map_b = 0;
+	} else {
+		ntx_path = hal->antenna_tx ? hal->antenna_tx : RF_B;
+		map_b = hal->antenna_tx == RF_AB ? 1 : 0;
+	}
 
 	SET_CMC_TBL_NTX_PATH_EN(skb->data, ntx_path);
 	SET_CMC_TBL_PATH_MAP_A(skb->data, 0);
-- 
cgit v1.2.3-70-g09d2


From 5c3afcba545cc820ba7911cecdf3abb05ea5e0df Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 30 Mar 2023 21:33:23 +0800
Subject: wifi: rtw89: mac: update MAC settings to support 8851b

Many settings of 8851B are the same as 8852B or 8852A, like DLE (Data link
engine), security engine and so on. Update them according to hardware
design.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330133324.19538-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/mac.c | 44 ++++++++++++++++++++++----------
 drivers/net/wireless/realtek/rtw89/reg.h |  2 ++
 2 files changed, 33 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 54887835ac15..385e2a0dfea5 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -1584,12 +1584,15 @@ static void dle_func_en(struct rtw89_dev *rtwdev, bool enable)
 
 static void dle_clk_en(struct rtw89_dev *rtwdev, bool enable)
 {
-	if (enable)
-		rtw89_write32_set(rtwdev, R_AX_DMAC_CLK_EN,
-				  B_AX_DLE_WDE_CLK_EN | B_AX_DLE_PLE_CLK_EN);
-	else
-		rtw89_write32_clr(rtwdev, R_AX_DMAC_CLK_EN,
-				  B_AX_DLE_WDE_CLK_EN | B_AX_DLE_PLE_CLK_EN);
+	u32 val = B_AX_DLE_WDE_CLK_EN | B_AX_DLE_PLE_CLK_EN;
+
+	if (enable) {
+		if (rtwdev->chip->chip_id == RTL8851B)
+			val |= B_AX_AXIDMA_CLK_EN;
+		rtw89_write32_set(rtwdev, R_AX_DMAC_CLK_EN, val);
+	} else {
+		rtw89_write32_clr(rtwdev, R_AX_DMAC_CLK_EN, val);
+	}
 }
 
 static int dle_mix_cfg(struct rtw89_dev *rtwdev, const struct rtw89_dle_mem *cfg)
@@ -1854,7 +1857,8 @@ static int preload_init(struct rtw89_dev *rtwdev, enum rtw89_mac_idx mac_idx,
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
 
-	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B || !is_qta_poh(rtwdev))
+	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
+	    chip->chip_id == RTL8851B || !is_qta_poh(rtwdev))
 		return 0;
 
 	return preload_init_set(rtwdev, mac_idx, mode);
@@ -1890,7 +1894,8 @@ static void _patch_ss2f_path(struct rtw89_dev *rtwdev)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
 
-	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B)
+	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
+	    chip->chip_id == RTL8851B)
 		return;
 
 	rtw89_write32_mask(rtwdev, R_AX_SS2FINFO_PATH, B_AX_SS_DEST_QUEUE_MASK,
@@ -1959,7 +1964,8 @@ static int sec_eng_init(struct rtw89_dev *rtwdev)
 	/* init TX encryption */
 	val |= (B_AX_SEC_TX_ENC | B_AX_SEC_RX_DEC);
 	val |= (B_AX_MC_DEC | B_AX_BC_DEC);
-	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B)
+	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
+	    chip->chip_id == RTL8851B)
 		val &= ~B_AX_TX_PARTIAL_MODE;
 	rtw89_write32(rtwdev, R_AX_SEC_ENG_CTRL, val);
 
@@ -2065,7 +2071,7 @@ static int scheduler_init(struct rtw89_dev *rtwdev, u8 mac_idx)
 		rtw89_write32_mask(rtwdev, reg, B_AX_SIFS_MACTXEN_T1_MASK,
 				   SIFS_MACTXEN_T1);
 
-	if (rtwdev->chip->chip_id == RTL8852B) {
+	if (rtwdev->chip->chip_id == RTL8852B || rtwdev->chip->chip_id == RTL8851B) {
 		reg = rtw89_mac_reg_by_idx(R_AX_SCH_EXT_CTRL, mac_idx);
 		rtw89_write32_set(rtwdev, reg, B_AX_PORT_RST_TSF_ADV);
 	}
@@ -3364,8 +3370,15 @@ static int rtw89_mac_trx_init(struct rtw89_dev *rtwdev)
 
 static void rtw89_disable_fw_watchdog(struct rtw89_dev *rtwdev)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	u32 val32;
 
+	if (chip_id == RTL8852B || chip_id == RTL8851B) {
+		rtw89_write32_clr(rtwdev, R_AX_PLATFORM_ENABLE, B_AX_APB_WRAP_EN);
+		rtw89_write32_set(rtwdev, R_AX_PLATFORM_ENABLE, B_AX_APB_WRAP_EN);
+		return;
+	}
+
 	rtw89_mac_mem_write(rtwdev, R_AX_WDT_CTRL,
 			    WDT_CTRL_ALL_DIS, RTW89_MAC_MEM_CPU_LOCAL);
 
@@ -3450,7 +3463,10 @@ static int rtw89_mac_dmac_pre_init(struct rtw89_dev *rtwdev)
 		      B_AX_PKT_BUF_EN;
 	rtw89_write32(rtwdev, R_AX_DMAC_FUNC_EN, val);
 
-	val = B_AX_DISPATCHER_CLK_EN;
+	if (chip_id == RTL8851B)
+		val = B_AX_DISPATCHER_CLK_EN | B_AX_AXIDMA_CLK_EN;
+	else
+		val = B_AX_DISPATCHER_CLK_EN;
 	rtw89_write32(rtwdev, R_AX_DMAC_CLK_EN, val);
 
 	if (chip_id != RTL8852C)
@@ -4691,11 +4707,13 @@ int rtw89_mac_coex_init(struct rtw89_dev *rtwdev, const struct rtw89_mac_ax_coex
 	int ret;
 
 	rtw89_write8_set(rtwdev, R_AX_GPIO_MUXCFG, B_AX_ENBT);
-	rtw89_write8_set(rtwdev, R_AX_BTC_FUNC_EN, B_AX_PTA_WL_TX_EN);
+	if (rtwdev->chip->chip_id != RTL8851B)
+		rtw89_write8_set(rtwdev, R_AX_BTC_FUNC_EN, B_AX_PTA_WL_TX_EN);
 	rtw89_write8_set(rtwdev, R_AX_BT_COEX_CFG_2 + 1, B_AX_GNT_BT_POLARITY >> 8);
 	rtw89_write8_set(rtwdev, R_AX_CSR_MODE, B_AX_STATIS_BT_EN | B_AX_WL_ACT_MSK);
 	rtw89_write8_set(rtwdev, R_AX_CSR_MODE + 2, B_AX_BT_CNT_RST >> 16);
-	rtw89_write8_clr(rtwdev, R_AX_TRXPTCL_RESP_0 + 3, B_AX_RSP_CHK_BTCCA >> 24);
+	if (rtwdev->chip->chip_id != RTL8851B)
+		rtw89_write8_clr(rtwdev, R_AX_TRXPTCL_RESP_0 + 3, B_AX_RSP_CHK_BTCCA >> 24);
 
 	val16 = rtw89_read16(rtwdev, R_AX_CCA_CFG_0);
 	val16 = (val16 | B_AX_BTCCA_EN) & ~B_AX_BTCCA_BRK_TXOP_EN;
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 37a1777a25a3..7e466ebaec2c 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -129,6 +129,7 @@
 
 #define R_AX_PLATFORM_ENABLE 0x0088
 #define B_AX_AXIDMA_EN BIT(3)
+#define B_AX_APB_WRAP_EN BIT(2)
 #define B_AX_WCPU_EN BIT(1)
 #define B_AX_PLATFORM_EN BIT(0)
 
@@ -488,6 +489,7 @@
 #define B_AX_DISPATCHER_CLK_EN BIT(18)
 #define B_AX_BBRPT_CLK_EN BIT(17)
 #define B_AX_MAC_SEC_CLK_EN BIT(16)
+#define B_AX_AXIDMA_CLK_EN BIT(9)
 
 #define PCI_LTR_IDLE_TIMER_1US 0
 #define PCI_LTR_IDLE_TIMER_10US 1
-- 
cgit v1.2.3-70-g09d2


From 2a6d518dedcbc8dc6e666c1a68700945bca174b8 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 30 Mar 2023 21:33:24 +0800
Subject: wifi: rtw89: pci: update PCI related settings to support 8851B

Many settings of 8851B are like 8852A or 8852B. Change them to proper
settings as hardware design.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230330133324.19538-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/pci.c | 33 ++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index fbe4d11ca920..70b4754667c9 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -1917,9 +1917,10 @@ __get_target(struct rtw89_dev *rtwdev, u16 *target, enum rtw89_pcie_phy phy_rate
 
 static int rtw89_pci_autok_x(struct rtw89_dev *rtwdev)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	int ret;
 
-	if (rtwdev->chip->chip_id != RTL8852B)
+	if (chip_id != RTL8852B && chip_id != RTL8851B)
 		return 0;
 
 	ret = rtw89_write16_mdio_mask(rtwdev, RAC_REG_FLD_0, BAC_AUTOK_N_MASK,
@@ -1929,13 +1930,14 @@ static int rtw89_pci_autok_x(struct rtw89_dev *rtwdev)
 
 static int rtw89_pci_auto_refclk_cal(struct rtw89_dev *rtwdev, bool autook_en)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	enum rtw89_pcie_phy phy_rate;
 	u16 val16, mgn_set, div_set, tar;
 	u8 val8, bdr_ori;
 	bool l1_flag = false;
 	int ret = 0;
 
-	if (rtwdev->chip->chip_id != RTL8852B)
+	if (chip_id != RTL8852B && chip_id != RTL8851B)
 		return 0;
 
 	ret = rtw89_pci_read_config_byte(rtwdev, RTW89_PCIE_PHY_RATE, &val8);
@@ -2112,7 +2114,9 @@ static void rtw89_pci_rxdma_prefth(struct rtw89_dev *rtwdev)
 
 static void rtw89_pci_l1off_pwroff(struct rtw89_dev *rtwdev)
 {
-	if (rtwdev->chip->chip_id != RTL8852A && rtwdev->chip->chip_id != RTL8852B)
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
+
+	if (chip_id != RTL8852A && chip_id != RTL8852B && chip_id != RTL8851B)
 		return;
 
 	rtw89_write32_clr(rtwdev, R_AX_PCIE_PS_CTRL, B_AX_L1OFF_PWR_OFF_EN);
@@ -2140,7 +2144,9 @@ static u32 rtw89_pci_l2_rxen_lat(struct rtw89_dev *rtwdev)
 
 static void rtw89_pci_aphy_pwrcut(struct rtw89_dev *rtwdev)
 {
-	if (rtwdev->chip->chip_id != RTL8852A && rtwdev->chip->chip_id != RTL8852B)
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
+
+	if (chip_id != RTL8852A && chip_id != RTL8852B && chip_id != RTL8851B)
 		return;
 
 	rtw89_write32_clr(rtwdev, R_AX_SYS_PW_CTRL, B_AX_PSUS_OFF_CAPC_EN);
@@ -2148,8 +2154,9 @@ static void rtw89_pci_aphy_pwrcut(struct rtw89_dev *rtwdev)
 
 static void rtw89_pci_hci_ldo(struct rtw89_dev *rtwdev)
 {
-	if (rtwdev->chip->chip_id == RTL8852A ||
-	    rtwdev->chip->chip_id == RTL8852B) {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
+
+	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
 		rtw89_write32_set(rtwdev, R_AX_SYS_SDIO_CTRL,
 				  B_AX_PCIE_DIS_L2_CTRL_LDO_HCI);
 		rtw89_write32_clr(rtwdev, R_AX_SYS_SDIO_CTRL,
@@ -2162,7 +2169,9 @@ static void rtw89_pci_hci_ldo(struct rtw89_dev *rtwdev)
 
 static int rtw89_pci_dphy_delay(struct rtw89_dev *rtwdev)
 {
-	if (rtwdev->chip->chip_id != RTL8852B)
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
+
+	if (chip_id != RTL8852B && chip_id != RTL8851B)
 		return 0;
 
 	return rtw89_write16_mdio_mask(rtwdev, RAC_REG_REV2, BAC_CMU_EN_DLY_MASK,
@@ -3402,7 +3411,7 @@ static void rtw89_pci_clkreq_set(struct rtw89_dev *rtwdev, bool enable)
 	if (ret)
 		rtw89_err(rtwdev, "failed to set CLKREQ Delay\n");
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
 		if (enable)
 			ret = rtw89_pci_config_byte_set(rtwdev,
 							RTW89_PCIE_L1_CTRL,
@@ -3447,7 +3456,7 @@ static void rtw89_pci_aspm_set(struct rtw89_dev *rtwdev, bool enable)
 	if (ret)
 		rtw89_err(rtwdev, "failed to read ASPM Delay\n");
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
 		if (enable)
 			ret = rtw89_pci_config_byte_set(rtwdev,
 							RTW89_PCIE_L1_CTRL,
@@ -3527,7 +3536,7 @@ static void rtw89_pci_l1ss_set(struct rtw89_dev *rtwdev, bool enable)
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	int ret;
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
 		if (enable)
 			ret = rtw89_pci_config_byte_set(rtwdev,
 							RTW89_PCIE_TIMER_CTRL,
@@ -3726,7 +3735,7 @@ static int __maybe_unused rtw89_pci_suspend(struct device *dev)
 	rtw89_write32_set(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
 	rtw89_write32_set(rtwdev, R_AX_RSV_CTRL, B_AX_R_DIS_PRST);
 	rtw89_write32_clr(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
 		rtw89_write32_clr(rtwdev, R_AX_SYS_SDIO_CTRL,
 				  B_AX_PCIE_DIS_L2_CTRL_LDO_HCI);
 		rtw89_write32_set(rtwdev, R_AX_PCIE_INIT_CFG1,
@@ -3760,7 +3769,7 @@ static int __maybe_unused rtw89_pci_resume(struct device *dev)
 	rtw89_write32_set(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
 	rtw89_write32_clr(rtwdev, R_AX_RSV_CTRL, B_AX_R_DIS_PRST);
 	rtw89_write32_clr(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
 		rtw89_write32_set(rtwdev, R_AX_SYS_SDIO_CTRL,
 				  B_AX_PCIE_DIS_L2_CTRL_LDO_HCI);
 		rtw89_write32_clr(rtwdev, R_AX_PCIE_INIT_CFG1,
-- 
cgit v1.2.3-70-g09d2


From 108bdaaa8bc7eafa2cf4aa3b883fdadca0d2ce45 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Sat, 1 Apr 2023 22:25:46 +0800
Subject: wifi: rtw89: 8851b: add BB and RF tables (1 of 2)

These tables contain BB and RF parameters that driver will load them into
registers. It also contains TX power according to country, band, rate and
so on. Increasing thermal can cause TX power degraded, so power tracking
tables are defined to compensate TX power.

Internal version of these tables:
 - HALBB_029_106_15 (V17)
 - HALRF_029_00_089
   * Radio A 0x22
   * NCTL 0x5

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230401142548.55466-2-pkshih@realtek.com
---
 .../net/wireless/realtek/rtw89/rtw8851b_table.c    | 6766 ++++++++++++++++++++
 .../net/wireless/realtek/rtw89/rtw8851b_table.h    |   21 +
 2 files changed, 6787 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8851b_table.c
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8851b_table.h

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c b/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c
new file mode 100644
index 000000000000..e964b32e4cc4
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c
@@ -0,0 +1,6766 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2022-2023  Realtek Corporation
+ */
+
+#include "phy.h"
+#include "reg.h"
+#include "rtw8851b_table.h"
+
+static const struct rtw89_reg2_def rtw89_8851b_phy_bb_regs[] = {
+	{0x704, 0x601E0500},
+	{0x4000, 0x00000000},
+	{0x4004, 0xCA014000},
+	{0x4008, 0xC751D4F0},
+	{0x400C, 0x44511475},
+	{0x4010, 0x00000000},
+	{0x4014, 0x00000000},
+	{0x47BC, 0x00000380},
+	{0x4018, 0x4F4C084B},
+	{0x401C, 0x084A4E52},
+	{0x4020, 0x4D504E4B},
+	{0x4024, 0x4F4C0849},
+	{0x4028, 0x08484C50},
+	{0x402C, 0x4C50504C},
+	{0x4030, 0x5454084A},
+	{0x4034, 0x084B5654},
+	{0x4038, 0x6A6C605A},
+	{0x403C, 0x4C4C084C},
+	{0x4040, 0x084B4E4D},
+	{0x4044, 0x4E4C4B4B},
+	{0x4048, 0x4B4B084A},
+	{0x404C, 0x084A4E4C},
+	{0x4050, 0x514F4C4A},
+	{0x4054, 0x524E084A},
+	{0x4058, 0x084A5154},
+	{0x405C, 0x53555554},
+	{0x4060, 0x45450845},
+	{0x4064, 0x08454144},
+	{0x4068, 0x40434445},
+	{0x406C, 0x44450845},
+	{0x4070, 0x08444043},
+	{0x4074, 0x42434444},
+	{0x4078, 0x46450844},
+	{0x407C, 0x08444843},
+	{0x4080, 0x4B4E4A47},
+	{0x4084, 0x4F4C084B},
+	{0x4088, 0x084A4E52},
+	{0x408C, 0x4D504E4B},
+	{0x4090, 0x4F4C0849},
+	{0x4094, 0x08484C50},
+	{0x4098, 0x4C50504C},
+	{0x409C, 0x5454084A},
+	{0x40A0, 0x084B5654},
+	{0x40A4, 0x6A6C605A},
+	{0x40A8, 0x4C4C084C},
+	{0x40AC, 0x084B4E4D},
+	{0x40B0, 0x4E4C4B4B},
+	{0x40B4, 0x4B4B084A},
+	{0x40B8, 0x084A4E4C},
+	{0x40BC, 0x514F4C4A},
+	{0x40C0, 0x524E084A},
+	{0x40C4, 0x084A5154},
+	{0x40C8, 0x53555554},
+	{0x40CC, 0x45450845},
+	{0x40D0, 0x08454144},
+	{0x40D4, 0x40434445},
+	{0x40D8, 0x44450845},
+	{0x40DC, 0x08444043},
+	{0x40E0, 0x42434444},
+	{0x40E4, 0x46450844},
+	{0x40E8, 0x08444843},
+	{0x40EC, 0x4B4E4A47},
+	{0x40F0, 0x00000000},
+	{0x4A38, 0x00000000},
+	{0x40F4, 0x00000006},
+	{0x40F8, 0x00000000},
+	{0x40FC, 0x8C30C30C},
+	{0x4100, 0x4C30C30C},
+	{0x4104, 0x0C30C30C},
+	{0x4108, 0x0C30C30C},
+	{0x410C, 0x0C30C30C},
+	{0x4110, 0x0C30C30C},
+	{0x4114, 0x28A28A28},
+	{0x4118, 0x28A28A28},
+	{0x411C, 0x28A28A28},
+	{0x4120, 0x28A28A28},
+	{0x4124, 0x28A28A28},
+	{0x4128, 0x28A28A28},
+	{0x412C, 0x06666666},
+	{0x4130, 0x33333333},
+	{0x4134, 0x33333333},
+	{0x4138, 0x33333333},
+	{0x413C, 0x00000031},
+	{0x4140, 0x5100600A},
+	{0x4144, 0x18363113},
+	{0x4148, 0x1D976DDC},
+	{0x414C, 0x1C072DD7},
+	{0x4150, 0x1127CDF4},
+	{0x4154, 0x1E37BDF1},
+	{0x4158, 0x1FB7F1D6},
+	{0x415C, 0x1EA7DDF9},
+	{0x4160, 0x1FE445DD},
+	{0x4164, 0x1F97F1FE},
+	{0x4168, 0x1FF781ED},
+	{0x416C, 0x1FA7F5FE},
+	{0x4170, 0x1E07B913},
+	{0x4174, 0x1FD7FDFF},
+	{0x4178, 0x1E17B9FA},
+	{0x417C, 0x19A66914},
+	{0x4180, 0x10F65598},
+	{0x4184, 0x14A5A111},
+	{0x4188, 0x1D3765DB},
+	{0x418C, 0x17C685CA},
+	{0x4190, 0x1107C5F3},
+	{0x4194, 0x1B5785EB},
+	{0x4198, 0x1F97ED8F},
+	{0x419C, 0x1BC7A5F3},
+	{0x41A0, 0x1FE43595},
+	{0x41A4, 0x1EB7D9FC},
+	{0x41A8, 0x1FE65DBE},
+	{0x41AC, 0x1EC7D9FC},
+	{0x41B0, 0x1976FCFF},
+	{0x41B4, 0x1F77F5FF},
+	{0x41B8, 0x1976FDEC},
+	{0x41BC, 0x198664EF},
+	{0x41C0, 0x11062D93},
+	{0x41C4, 0x10C4E910},
+	{0x41C8, 0x1CA759DB},
+	{0x41CC, 0x1335A9B5},
+	{0x41D0, 0x1097B9F3},
+	{0x41D4, 0x17B72DE1},
+	{0x41D8, 0x1F67ED42},
+	{0x41DC, 0x18074DE9},
+	{0x41E0, 0x1FD40547},
+	{0x41E4, 0x1D57ADF9},
+	{0x41E8, 0x1FE52182},
+	{0x41EC, 0x1D67B1F9},
+	{0x41F0, 0x14860CE1},
+	{0x41F4, 0x1EC7E9FE},
+	{0x41F8, 0x14860DD6},
+	{0x41FC, 0x195664C7},
+	{0x4200, 0x0005E58A},
+	{0x4204, 0x00000000},
+	{0x4208, 0x00000000},
+	{0x420C, 0x7A000000},
+	{0x4210, 0x0F9F3D7A},
+	{0x4214, 0x0040817C},
+	{0x4218, 0x00E10204},
+	{0x421C, 0x227D94CD},
+	{0x4220, 0x08028A28},
+	{0x4224, 0x00000200},
+	{0x4228, 0x04688000},
+	{0x47C0, 0x00000001},
+	{0x4A48, 0x00000002},
+	{0x4B04, 0x00000000},
+	{0x4B08, 0x00000000},
+	{0x422C, 0x0060B002},
+	{0x4230, 0x9A8249A8},
+	{0x4234, 0x26A1469E},
+	{0x4238, 0x2099A824},
+	{0x423C, 0x2359461C},
+	{0x4240, 0x1631A675},
+	{0x4244, 0x2C6B1D63},
+	{0x4248, 0x0000000E},
+	{0x424C, 0x00000001},
+	{0x4250, 0x00000001},
+	{0x4254, 0x00000000},
+	{0x4258, 0x00000000},
+	{0x425C, 0x00000000},
+	{0x4260, 0x0020000C},
+	{0x4A30, 0x00000000},
+	{0x4264, 0x00000000},
+	{0x4268, 0x00000000},
+	{0x426C, 0x0418317C},
+	{0x4270, 0x2B33135C},
+	{0x4274, 0x00000002},
+	{0x4278, 0x00000000},
+	{0x427C, 0x00000000},
+	{0x4280, 0x00000000},
+	{0x4284, 0x00000000},
+	{0x4288, 0x00000000},
+	{0x428C, 0x00000000},
+	{0x4290, 0x00000000},
+	{0x4294, 0x00000000},
+	{0x4298, 0x00000000},
+	{0x429C, 0x84026000},
+	{0x42A0, 0x0051AC20},
+	{0x4A24, 0x0010C040},
+	{0x42A4, 0x02024008},
+	{0x42A8, 0x00000000},
+	{0x42AC, 0x00000000},
+	{0x42B0, 0x22CE803C},
+	{0x42B4, 0xD8000000},
+	{0x42B8, 0x596FD67E},
+	{0x42BC, 0x7D67D67D},
+	{0x42C0, 0x7D67D65B},
+	{0x42C4, 0x28029F59},
+	{0x42C8, 0x00280280},
+	{0x4AF4, 0x00000000},
+	{0x42CC, 0x00000000},
+	{0x42D0, 0x00000000},
+	{0x42D4, 0x00000003},
+	{0x4AF8, 0x00280000},
+	{0x42D8, 0x00000001},
+	{0x42DC, 0x69AEC800},
+	{0x42E0, 0x8B4CD3D1},
+	{0x42E4, 0xC514534F},
+	{0x42E8, 0x85145145},
+	{0x42EC, 0x45145145},
+	{0x42F0, 0x05145145},
+	{0x42F4, 0x05145145},
+	{0x42F8, 0x05145145},
+	{0x42FC, 0x17659145},
+	{0x4300, 0x176DD5D9},
+	{0x4304, 0x0F65765B},
+	{0x4308, 0x0F3CF3CF},
+	{0x430C, 0x0F3CF3CF},
+	{0x4310, 0x0F3CF3CF},
+	{0x4314, 0x0F3CF3CF},
+	{0x4318, 0x0F3CF3CF},
+	{0x431C, 0x0F3CF3CF},
+	{0x4320, 0x0F3CF3CF},
+	{0x4324, 0x0F44F351},
+	{0x4328, 0x192D7547},
+	{0x432C, 0x0F5CF5CF},
+	{0x4330, 0x051593D9},
+	{0x4334, 0x05145145},
+	{0x4338, 0x05145145},
+	{0x433C, 0x05145145},
+	{0x4340, 0x05145145},
+	{0x4344, 0x05145145},
+	{0x4348, 0x19545145},
+	{0x434C, 0x1B65B5DB},
+	{0x4350, 0x1965965B},
+	{0x4354, 0x0F3CF3CF},
+	{0x4358, 0x0F3CF3CF},
+	{0x435C, 0x0F3CF1CF},
+	{0x4360, 0x0F3CF3CF},
+	{0x4364, 0x0F3CF3CF},
+	{0x4368, 0x0F3CF3CF},
+	{0x436C, 0x0F3CF3CF},
+	{0x4370, 0x0934D2CF},
+	{0x4374, 0x112CB3CF},
+	{0x4378, 0x9777A777},
+	{0x437C, 0xBB7BAC95},
+	{0x4380, 0xB667B889},
+	{0x4384, 0x7B9B8899},
+	{0x4388, 0x7A5567C8},
+	{0x438C, 0x2278CCCC},
+	{0x4390, 0x7C222222},
+	{0x4394, 0x0000029B},
+	{0x4398, 0x001CCCCC},
+	{0x4AAC, 0xCCCCC88C},
+	{0x4AB0, 0x0000AACC},
+	{0x439C, 0x00000000},
+	{0x43A0, 0x00000008},
+	{0x43A4, 0x00000000},
+	{0x43A8, 0x00000000},
+	{0x43AC, 0x00000000},
+	{0x43B0, 0x10000000},
+	{0x43B4, 0x00401001},
+	{0x43B8, 0x00061003},
+	{0x43BC, 0x000024D8},
+	{0x43C0, 0x00000000},
+	{0x43C4, 0x10000020},
+	{0x43C8, 0x20000200},
+	{0x43CC, 0x00000000},
+	{0x43D0, 0x04000000},
+	{0x43D4, 0x44000100},
+	{0x43D8, 0x60804060},
+	{0x43DC, 0x44204210},
+	{0x43E0, 0x82108082},
+	{0x43E4, 0x82108402},
+	{0x43E8, 0xC8082108},
+	{0x43EC, 0xC8202084},
+	{0x43F0, 0x44208208},
+	{0x43F4, 0x84108204},
+	{0x43F8, 0xD0108104},
+	{0x43FC, 0xF8210108},
+	{0x4400, 0x6431E930},
+	{0x4404, 0x02309468},
+	{0x4408, 0x10C61C22},
+	{0x440C, 0x02109469},
+	{0x4410, 0x10C61C22},
+	{0x4414, 0x00041049},
+	{0x4A4C, 0x00060581},
+	{0x4418, 0x00000000},
+	{0x441C, 0x00000000},
+	{0x4420, 0xEC000000},
+	{0x4424, 0xB0200020},
+	{0x4428, 0x00001FF0},
+	{0x4AC8, 0x00000001},
+	{0x4B0C, 0x00000000},
+	{0x4CDC, 0x00000000},
+	{0x442C, 0x00000000},
+	{0x4430, 0x00000000},
+	{0x4434, 0x00000000},
+	{0x4438, 0x00000000},
+	{0x443C, 0x190642D0},
+	{0x4440, 0xA80668A0},
+	{0x4444, 0x60900820},
+	{0x4448, 0x9F28518C},
+	{0x444C, 0x32488A62},
+	{0x4450, 0x9C6E36DC},
+	{0x4454, 0x0000F52B},
+	{0x4A34, 0x00000007},
+	{0x4CE0, 0x68120000},
+	{0x4CE4, 0x1A0681E0},
+	{0x4CE8, 0x94060180},
+	{0x4CEC, 0x000603FF},
+	{0x4CF0, 0xA0502000},
+	{0x4CF4, 0x00001000},
+	{0x4D00, 0x00000044},
+	{0x4B14, 0x00000000},
+	{0x4458, 0x00000000},
+	{0x445C, 0x4801442E},
+	{0x4460, 0x0051A0FA},
+	{0x4B18, 0x0000011F},
+	{0x4B1C, 0x0000011F},
+	{0x4464, 0x00000000},
+	{0x4468, 0x00000000},
+	{0x446C, 0x00000000},
+	{0x4470, 0x00000000},
+	{0x4474, 0x00000000},
+	{0x4478, 0x00000000},
+	{0x447C, 0x00000000},
+	{0x4480, 0x2A0A6040},
+	{0x4484, 0x0A0A6829},
+	{0x4488, 0x00000004},
+	{0x448C, 0x00000000},
+	{0x4490, 0x80000000},
+	{0x4494, 0x10000000},
+	{0x4498, 0xE0000000},
+	{0x4A28, 0x000ED877},
+	{0x4AB4, 0x00000000},
+	{0x4B20, 0x00000000},
+	{0x4B24, 0x00000000},
+	{0x4B28, 0x00000000},
+	{0x4B2C, 0x00000000},
+	{0x449C, 0x0000001E},
+	{0x44A0, 0x02B2C394},
+	{0x44A4, 0x00000400},
+	{0x4A2C, 0x0050240E},
+	{0x4B30, 0x7FFFFD20},
+	{0x4B34, 0x920823FF},
+	{0x4B38, 0x7FFFFFFF},
+	{0x4B3C, 0x01773773},
+	{0x44A8, 0x00000001},
+	{0x44AC, 0x000190C0},
+	{0x44B0, 0x00000000},
+	{0x44B4, 0x00000000},
+	{0x44B8, 0x00000000},
+	{0x44BC, 0x00000000},
+	{0x44C0, 0x00000000},
+	{0x44C4, 0x00000000},
+	{0x44C8, 0x00000000},
+	{0x44CC, 0x00000000},
+	{0x44D0, 0x00000000},
+	{0x44D4, 0x00000000},
+	{0x44D8, 0x00000000},
+	{0x44DC, 0x00000000},
+	{0x44E0, 0x00000000},
+	{0x44E4, 0x00000000},
+	{0x44E8, 0x00000000},
+	{0x44EC, 0x00000000},
+	{0x44F0, 0x00000000},
+	{0x44F4, 0x00000000},
+	{0x44F8, 0x00000000},
+	{0x44FC, 0x00000000},
+	{0x4500, 0x00000000},
+	{0x4504, 0x00000000},
+	{0x4508, 0x00000000},
+	{0x450C, 0x00000000},
+	{0x4510, 0x00000000},
+	{0x4514, 0x00000000},
+	{0x4518, 0x00000000},
+	{0x451C, 0x00000000},
+	{0x4520, 0x00000000},
+	{0x4524, 0x00000000},
+	{0x4528, 0x00000000},
+	{0x452C, 0x00000000},
+	{0x4530, 0x4E830171},
+	{0x4534, 0x00000870},
+	{0x4538, 0x000000FF},
+	{0x453C, 0x00000000},
+	{0x4540, 0x00000000},
+	{0x4544, 0x00000000},
+	{0x4548, 0x00000000},
+	{0x454C, 0x00000000},
+	{0x4550, 0x00000000},
+	{0x4554, 0x00000000},
+	{0x4558, 0x00000000},
+	{0x455C, 0x00000000},
+	{0x4560, 0x40000000},
+	{0x4564, 0x40000000},
+	{0x4568, 0x00000000},
+	{0x456C, 0x20000000},
+	{0x4570, 0x04F040BB},
+	{0x4574, 0x000E53FF},
+	{0x4578, 0x000205CB},
+	{0x457C, 0x00200000},
+	{0x4580, 0x00000040},
+	{0x4584, 0x00000000},
+	{0x4588, 0x00000017},
+	{0x458C, 0x30000000},
+	{0x4590, 0x00000000},
+	{0x4594, 0x00000000},
+	{0x4598, 0x00000001},
+	{0x459C, 0x0003FE00},
+	{0x45A0, 0x00000086},
+	{0x45A4, 0x00000000},
+	{0x45A8, 0xC00001C0},
+	{0x45AC, 0x78038000},
+	{0x45B0, 0x8000004A},
+	{0x45B4, 0x04094800},
+	{0x45B8, 0x00280002},
+	{0x45BC, 0x06748790},
+	{0x45C0, 0x80000000},
+	{0x45C4, 0x00000000},
+	{0x45C8, 0x00000000},
+	{0x45CC, 0x00558670},
+	{0x45D0, 0x002883F0},
+	{0x45D4, 0x00090120},
+	{0x45D8, 0x00000000},
+	{0x45E0, 0xA3A6D3C4},
+	{0x45E4, 0xAB27B126},
+	{0x45E8, 0x00006778},
+	{0x45F4, 0x000001B5},
+	{0x45EC, 0x11110F0A},
+	{0x45F0, 0x00000003},
+	{0x4A0C, 0x0000000A},
+	{0x45F8, 0x0058BC3F},
+	{0x45FC, 0x00000003},
+	{0x462C, 0x00000020},
+	{0x4600, 0x000003D9},
+	{0x45F0, 0x00000004},
+	{0x4604, 0x002B1CB0},
+	{0x4A50, 0xC0000000},
+	{0x4A54, 0x00001000},
+	{0x4A58, 0x00000000},
+	{0x4A18, 0x00000024},
+	{0x4608, 0x00000001},
+	{0x460C, 0x00000000},
+	{0x4A10, 0x00000001},
+	{0x4610, 0x00000001},
+	{0x4614, 0x16E5298F},
+	{0x4618, 0x18C6294A},
+	{0x461C, 0x0E06318A},
+	{0x4620, 0x0E539CE5},
+	{0x4624, 0x00019287},
+	{0x4A14, 0x000000BF},
+	{0x4628, 0x00000001},
+	{0x4630, 0x000001AA},
+	{0x4A18, 0x00001900},
+	{0x4A1C, 0x000002A6},
+	{0x4634, 0x000000A3},
+	{0x4A20, 0x00000086},
+	{0x4638, 0x00045656},
+	{0x49F8, 0x00000000},
+	{0x463C, 0x00000000},
+	{0x4640, 0x00000000},
+	{0x4644, 0x00C8CC00},
+	{0x4648, 0xC400B6B6},
+	{0x464C, 0xDC400FC0},
+	{0x4A44, 0x00000000},
+	{0x4A8C, 0x00000110},
+	{0x4BC4, 0x00000001},
+	{0x4650, 0x08882550},
+	{0x4654, 0x08CC2660},
+	{0x4658, 0x09102660},
+	{0x465C, 0x00000154},
+	{0x45DC, 0xC39E38E8},
+	{0x4660, 0x452607E6},
+	{0x4664, 0x6750DC65},
+	{0x4668, 0xF3F0F1ED},
+	{0x466C, 0x30141506},
+	{0x4670, 0x2C2B2B2B},
+	{0x4674, 0x2C2C2C2C},
+	{0x4678, 0xDDB738E8},
+	{0x467C, 0x543618FB},
+	{0x4680, 0x4F31DC6F},
+	{0x4684, 0xFBEBDA00},
+	{0x4688, 0x1A10FF04},
+	{0x468C, 0x282A3000},
+	{0x4690, 0x2A29292A},
+	{0x4694, 0x04FA2A2A},
+	{0x4698, 0xEE0F04D1},
+	{0x469C, 0x99E91436},
+	{0x46A0, 0x0701E79E},
+	{0x46A4, 0x08D77CFF},
+	{0x46A8, 0x321AFF14},
+	{0x46AC, 0x60313447},
+	{0x46B0, 0x63666666},
+	{0x46B4, 0x35374425},
+	{0x46B8, 0x35883042},
+	{0x46BC, 0x5177C252},
+	{0x4720, 0x7FFFFD63},
+	{0x4724, 0xB58D11FF},
+	{0x4728, 0x47FFFFFF},
+	{0x472C, 0x0E7893B6},
+	{0x4730, 0xE0391201},
+	{0x4734, 0x00000020},
+	{0x4738, 0x8325C500},
+	{0x473C, 0x00000B7F},
+	{0x46C0, 0x00000000},
+	{0x46C4, 0x00000000},
+	{0x46C8, 0x00000219},
+	{0x4BDC, 0x00002020},
+	{0x46CC, 0x00000000},
+	{0x46D0, 0x00000000},
+	{0x4A3C, 0x00000002},
+	{0x46D4, 0x00000001},
+	{0x46D8, 0x00000001},
+	{0x46DC, 0x00000000},
+	{0x46E0, 0x00000000},
+	{0x46E4, 0x00000151},
+	{0x46E8, 0x00000498},
+	{0x46EC, 0x00000498},
+	{0x46F0, 0x00000000},
+	{0x46F4, 0x00000000},
+	{0x46F8, 0x00001146},
+	{0x46FC, 0x00000000},
+	{0x4700, 0x00000000},
+	{0x4704, 0x00C8CC00},
+	{0x4708, 0xC400B6B6},
+	{0x470C, 0xDC400FC0},
+	{0x4A90, 0x00000110},
+	{0x4B10, 0x00000000},
+	{0x4BE0, 0x00000001},
+	{0x4710, 0x08882550},
+	{0x4714, 0x08CC2660},
+	{0x4718, 0x09102660},
+	{0x471C, 0x00000154},
+	{0x4740, 0xC69F38E8},
+	{0x4744, 0x462709E9},
+	{0x4748, 0x6750DC67},
+	{0x474C, 0xF3F0F1ED},
+	{0x4750, 0x30141506},
+	{0x4754, 0x2C2B2B2B},
+	{0x4758, 0x2C2C2C2C},
+	{0x475C, 0xE0B738E8},
+	{0x4760, 0x52381BFE},
+	{0x4764, 0x5031DC6C},
+	{0x4768, 0xFBEBDA00},
+	{0x476C, 0x1A10FF04},
+	{0x4770, 0x282A3000},
+	{0x4774, 0x2A29292A},
+	{0x4778, 0x04FA2A2A},
+	{0x477C, 0xEE0F04D1},
+	{0x47C4, 0x00000000},
+	{0x47C8, 0xA32103FE},
+	{0x47CC, 0xB20A5328},
+	{0x47D0, 0xC686314F},
+	{0x47D4, 0x000004D7},
+	{0x4BFC, 0x00000000},
+	{0x4C00, 0x0C442416},
+	{0x4C04, 0x00000000},
+	{0x47D8, 0x009B902A},
+	{0x47DC, 0x009B902A},
+	{0x47E0, 0x98682C18},
+	{0x47E4, 0x6318C4C1},
+	{0x47E8, 0x6248C631},
+	{0x47EC, 0x922A8253},
+	{0x47F0, 0x00000005},
+	{0x47F4, 0x00001759},
+	{0x47F8, 0x4BB01800},
+	{0x47FC, 0x831408BE},
+	{0x4A84, 0x000000E9},
+	{0x4C08, 0x0F801404},
+	{0x4C0C, 0x00A2B404},
+	{0x4800, 0x9ABBCACB},
+	{0x4804, 0x56867578},
+	{0x4808, 0xBCCBBB13},
+	{0x480C, 0x7889989B},
+	{0x4810, 0xBBB0F455},
+	{0x4814, 0x777BBBBB},
+	{0x4818, 0x15277777},
+	{0x481C, 0x27039CE9},
+	{0x4820, 0x42424432},
+	{0x4824, 0x36058342},
+	{0x4828, 0x00000006},
+	{0x482C, 0x00000005},
+	{0x4830, 0x00000005},
+	{0x4834, 0xC7013016},
+	{0x4838, 0x84413016},
+	{0x483C, 0x84413016},
+	{0x4840, 0x8C413016},
+	{0x4844, 0x8C40B028},
+	{0x4848, 0x3140B028},
+	{0x484C, 0x2940B028},
+	{0x4850, 0x8440B028},
+	{0x4854, 0x2318C610},
+	{0x4858, 0x45344753},
+	{0x485C, 0x236A6A88},
+	{0x4860, 0xAC8DF814},
+	{0x4864, 0x08877ACB},
+	{0x4868, 0x000107AA},
+	{0x4A94, 0x00000000},
+	{0x486C, 0xBCEB4A14},
+	{0x4870, 0x000A3A4A},
+	{0x4874, 0xBCEB4A14},
+	{0x4878, 0x000A3A4A},
+	{0x487C, 0xBCBDBD85},
+	{0x4880, 0x0CABB99A},
+	{0x4884, 0x38384242},
+	{0x4888, 0x0086102E},
+	{0x488C, 0xCA24C82A},
+	{0x4AFC, 0x00000000},
+	{0x4C14, 0x0000349D},
+	{0x4CF8, 0x00000007},
+	{0x4890, 0x00008A62},
+	{0x4894, 0x00000008},
+	{0x4958, 0x80040000},
+	{0x495C, 0x80040000},
+	{0x4960, 0xFE800000},
+	{0x4964, 0x834C0000},
+	{0x4968, 0x00000000},
+	{0x496C, 0x00000000},
+	{0x4970, 0x00000000},
+	{0x4974, 0x00000000},
+	{0x4978, 0x00000000},
+	{0x497C, 0x00000000},
+	{0x4980, 0x40000000},
+	{0x4984, 0x00000000},
+	{0x4988, 0x00000000},
+	{0x498C, 0x00000000},
+	{0x4990, 0x00000000},
+	{0x4994, 0x04065800},
+	{0x4998, 0x02004080},
+	{0x499C, 0x0E1E3E05},
+	{0x49A0, 0x0A163068},
+	{0x49A4, 0x00206040},
+	{0x49A8, 0x02020202},
+	{0x49AC, 0x00002020},
+	{0x49B0, 0xF8F8F418},
+	{0x49B4, 0xF8E8F8F8},
+	{0x49B8, 0xF80808E8},
+	{0x4A00, 0xF8F8FA00},
+	{0x4A04, 0xFAFAFAF8},
+	{0x4A08, 0xFAFAFAFA},
+	{0x49BC, 0x00000000},
+	{0x49C0, 0x800C562D},
+	{0x49C4, 0x00000101},
+	{0x49C8, 0x00000000},
+	{0x49CC, 0x00000000},
+	{0x49D0, 0x00000000},
+	{0x49D4, 0x00000000},
+	{0x49D8, 0x00000000},
+	{0x49DC, 0x00000000},
+	{0x49E0, 0x00000000},
+	{0x49E4, 0x00000000},
+	{0x49E8, 0x00000000},
+	{0x49EC, 0x00000000},
+	{0x4C28, 0x00000000},
+	{0x4C2C, 0x00000000},
+	{0x4C30, 0x00000000},
+	{0x4C34, 0x00000000},
+	{0x4C38, 0x00000000},
+	{0x4C3C, 0x00000000},
+	{0x4C40, 0x00000000},
+	{0x4C44, 0x01C0C832},
+	{0x4C48, 0x03207032},
+	{0x4C4C, 0x0320701C},
+	{0x4C50, 0x03207032},
+	{0x4C54, 0x01C0C81C},
+	{0x4C58, 0x00A0281C},
+	{0x4C5C, 0x0320C80A},
+	{0x4C60, 0x00A0C832},
+	{0x4C64, 0x01C0C832},
+	{0x4C68, 0x03207032},
+	{0x4C6C, 0x0320701C},
+	{0x4C70, 0x03207032},
+	{0x4C74, 0x01C0C81C},
+	{0x4C78, 0x00A0281C},
+	{0x4C7C, 0x0321A80A},
+	{0x4C80, 0x0320C86A},
+	{0x4C84, 0x12B02832},
+	{0x4C88, 0x12B3292B},
+	{0x4C8C, 0x0CA4ACCA},
+	{0x4C90, 0x12B4AC6A},
+	{0x4C94, 0x0CA4ACCA},
+	{0x4C98, 0x06A3292B},
+	{0x4C9C, 0x06A0280A},
+	{0x4CA0, 0x0CA0286A},
+	{0x4CA4, 0x0CA1A8CA},
+	{0x4CA8, 0x06A3286A},
+	{0x4CAC, 0x0000000A},
+	{0x4CB0, 0x01209C27},
+	{0x4CB4, 0x02704800},
+	{0x4CB8, 0x02704812},
+	{0x4CBC, 0x00004827},
+	{0x4CC0, 0x01209C12},
+	{0x4CC4, 0x00000012},
+	{0x4CC8, 0x02718000},
+	{0x4CCC, 0x02709C60},
+	{0x4CD0, 0x00000027},
+	{0x4CD4, 0x00000000},
+	{0x4CD8, 0x0000014A},
+	{0x994, 0x00000010},
+	{0x904, 0x00000005},
+	{0x708, 0x00000000},
+	{0x884, 0x0043F01D},
+	{0x710, 0xEF810000},
+	{0x718, 0x1333233F},
+	{0x604, 0x041E1E1E},
+	{0x714, 0x00010000},
+	{0x586C, 0x000000F0},
+	{0x586C, 0x000000E0},
+	{0x586C, 0x000000D0},
+	{0x586C, 0x000000C0},
+	{0x586C, 0x000000B0},
+	{0x586C, 0x000000A0},
+	{0x586C, 0x00000090},
+	{0x586C, 0x00000080},
+	{0x586C, 0x00000070},
+	{0x586C, 0x00000060},
+	{0x586C, 0x00000050},
+	{0x586C, 0x00000040},
+	{0x586C, 0x00000030},
+	{0x586C, 0x00000020},
+	{0x586C, 0x00000010},
+	{0x586C, 0x00000000},
+	{0xC0D4, 0xABA41460},
+	{0xC0D8, 0xC43A7E87},
+	{0xC0DC, 0x30C194B8},
+	{0xC0E0, 0x75008138},
+	{0xC0E4, 0x0000272B},
+	{0xC0E8, 0x000A0C81},
+	{0xC0EC, 0x00030003},
+	{0xC0F0, 0x00000024},
+	{0xC0C4, 0x005E3A00},
+	{0xC004, 0x45800000},
+	{0xC024, 0x45800000},
+	{0x334, 0xFFFFFFFF},
+	{0x33C, 0x55000000},
+	{0x340, 0x00005555},
+	{0x724, 0x00111200},
+	{0x5868, 0xA9550000},
+	{0x5870, 0x33221100},
+	{0x5874, 0x77665544},
+	{0x5878, 0xBBAA9988},
+	{0x587C, 0xFFEEDDCC},
+	{0x5880, 0x76543210},
+	{0x5884, 0xFEDCBA98},
+	{0x5888, 0x00000000},
+	{0x588C, 0x00000000},
+	{0x5894, 0x00000008},
+	{0x650, 0x00200888},
+	{0x710, 0xF3810000},
+	{0x020, 0x0000F381},
+	{0x024, 0x0000F381},
+	{0x000, 0xC580801E},
+	{0x980, 0x10002250},
+	{0x988, 0x3C3C4107},
+	{0x994, 0x00000010},
+	{0x000, 0x0580801F},
+	{0x240C, 0x00000000},
+	{0x640, 0x210A141E},
+	{0x640, 0x2114141E},
+	{0x640, 0x2114141E},
+	{0x644, 0x3414283C},
+	{0x644, 0x3425283C},
+	{0x644, 0x3426283C},
+	{0x2640, 0x140A141E},
+	{0x2640, 0x1414141E},
+	{0x2640, 0x1414141E},
+	{0x2644, 0x3414283C},
+	{0x2644, 0x3425283C},
+	{0x2644, 0x3425183C},
+	{0x2300, 0x02748790},
+	{0x2304, 0x00558670},
+	{0x2308, 0x002883F0},
+	{0x230C, 0x00090120},
+	{0x2310, 0x00000000},
+	{0x2314, 0x06000000},
+	{0x2318, 0x00000000},
+	{0x231C, 0x00000000},
+	{0x2320, 0x03020100},
+	{0x2324, 0x07060504},
+	{0x2328, 0x0B0A0908},
+	{0x232C, 0x0F0E0D0C},
+	{0x2330, 0x13121110},
+	{0x2334, 0x17161514},
+	{0x2338, 0x0C700022},
+	{0x233C, 0x0A0529D0},
+	{0x2340, 0x000529D0},
+	{0x2344, 0x0006318A},
+	{0x2348, 0xB7E6318A},
+	{0x234C, 0x80039C00},
+	{0x2350, 0x80039C00},
+	{0x2354, 0x0005298F},
+	{0x2358, 0x0015296E},
+	{0x235C, 0x0C07FC31},
+	{0x2360, 0x0219AAAE},
+	{0x2364, 0xE4F624C3},
+	{0x2368, 0x53626F15},
+	{0x236C, 0x48000000},
+	{0x2370, 0x48000000},
+	{0x2374, 0x07540000},
+	{0x2378, 0x202401B9},
+	{0x237C, 0x00F7000E},
+	{0x2380, 0x0F0A1111},
+	{0x2384, 0x30D9000F},
+	{0x2388, 0x0200EA02},
+	{0x238C, 0x003CB061},
+	{0x2390, 0x69C00000},
+	{0x2394, 0x00000000},
+	{0x2398, 0x000000F0},
+	{0x239C, 0x0001FFFF},
+	{0x23A0, 0x00C80064},
+	{0x23A4, 0x0190012C},
+	{0x23A8, 0x001917BE},
+	{0x23AC, 0x0B30880C},
+	{0x23B0, 0x9281CE00},
+	{0x23B4, 0x7F027C00},
+	{0x704, 0x601E0502},
+	{0x5600, 0x00000000},
+	{0x5604, 0x802D2721},
+	{0x5610, 0x00201020},
+	{0x5618, 0x00801008},
+	{0x5624, 0x0808081E},
+	{0x562C, 0x0000081D},
+	{0x5634, 0x3D2EE000},
+	{0x5638, 0x0001AC42},
+	{0x5640, 0x3D6EF000},
+	{0x5644, 0x0001AC3E},
+	{0x566C, 0x00210005},
+	{0x5680, 0x20500010},
+	{0x5684, 0x00020001},
+	{0x56A0, 0x0034C000},
+	{0x56BC, 0x04000000},
+	{0x56C0, 0x00000688},
+	{0x56C4, 0x00000010},
+	{0x56C8, 0x0E800400},
+	{0x56CC, 0x01E400FF},
+	{0x5800, 0x003F807F},
+	{0x5810, 0x59008400},
+	{0x5814, 0x201AF000},
+	{0x5818, 0x182C18E8},
+	{0x581C, 0x3DD80280},
+	{0x5820, 0x80000080},
+	{0x5828, 0x023F8121},
+	{0x5830, 0x023F8121},
+	{0x5838, 0x003F8121},
+	{0x5840, 0x023F8121},
+	{0x5848, 0x023F8121},
+	{0x5850, 0x023F8121},
+	{0x5858, 0x003F7121},
+	{0x5860, 0x023F7121},
+	{0x5864, 0x1A1801FF},
+	{0x5868, 0xA9A90002},
+	{0x5880, 0x77777777},
+	{0x5884, 0x77777777},
+	{0x5894, 0x01080604},
+	{0x5898, 0x00000000},
+	{0x589C, 0x00000000},
+	{0x58A0, 0x000000FE},
+	{0x58B0, 0x00000800},
+	{0x58BC, 0x07A7807F},
+	{0x58C0, 0x007E0000},
+	{0x58C4, 0x0003FFFF},
+	{0x58D4, 0x7401FE00},
+	{0x58D8, 0x8008016C},
+	{0x58DC, 0xC000807F},
+	{0x58E4, 0x3000881F},
+	{0x58E8, 0x00000003},
+	{0x58F0, 0x400401FF},
+	{0x58F4, 0x80000000},
+	{0x58F8, 0xC0000000},
+	{0x58FC, 0x00000000},
+	{0x700, 0x40000030},
+	{0x704, 0x601E0502},
+	{0x704, 0x601E0500},
+	{0x704, 0x601E0502},
+	{0x20FC, 0x00000000},
+	{0x20F8, 0x00000000},
+	{0x20F0, 0x00000000},
+	{0x9C0, 0x00000001},
+	{0x9C0, 0x00000000},
+	{0x9C0, 0x00000001},
+	{0x9C0, 0x00000000},
+	{0x4AE8, 0x00000744},
+	{0x4AD4, 0x00000040},
+	{0x4AE4, 0x0079E99E},
+	{0x4BC8, 0xFBD5B89F},
+	{0x4BCC, 0x99563918},
+	{0x4BD0, 0x12EED5B8},
+	{0x4BD4, 0x6F7D542F},
+	{0x4BD8, 0x0000001D},
+	{0x300, 0xF30CE31C},
+	{0x304, 0x13EF1F19},
+	{0x308, 0x0C0CF3F3},
+	{0x30C, 0x0CE30C0C},
+	{0x310, 0x80496000},
+	{0x314, 0x0041E000},
+	{0x318, 0x20022042},
+	{0x31C, 0x20448009},
+	{0x320, 0x00010031},
+	{0x324, 0xE000E000},
+	{0x328, 0xE000E000},
+	{0x32C, 0xE0008049},
+	{0x12BC, 0x10104041},
+	{0x12C0, 0x13311111},
+	{0x12E4, 0x30D52A68},
+	{0x010, 0x0005FFFF},
+	{0x028, 0x0000F381},
+	{0x02C, 0x0000F381},
+	{0x620, 0x00141230},
+	{0x70C, 0x00000020},
+	{0x720, 0x20000000},
+	{0x730, 0x00000002},
+	{0x738, 0x004100C0},
+	{0x73C, 0x00000002},
+	{0x748, 0x01000002},
+	{0x74C, 0x00000001},
+	{0xA08, 0x00007800},
+	{0xC14, 0x25010000},
+	{0xC3C, 0x2840E1BF},
+	{0xC40, 0x00000000},
+	{0xC44, 0x00000007},
+	{0xC48, 0x410E4000},
+	{0xC54, 0x1EE14368},
+	{0xC58, 0x41000000},
+	{0xC5C, 0x80558000},
+	{0xC60, 0x017FFFF2},
+	{0xC64, 0x0010A130},
+	{0xC68, 0x90000050},
+	{0xC6C, 0x10201021},
+	{0xC70, 0x071B0660},
+	{0xC74, 0x00000000},
+	{0xC78, 0x80000000},
+	{0xC7C, 0x0020BFE0},
+	{0xC88, 0xC2AC8000},
+	{0xC8C, 0x02F2FC08},
+	{0xD00, 0x77777777},
+	{0xD04, 0xBBBBBBBB},
+	{0xD08, 0xBBBBBBBB},
+	{0xD0C, 0x000B2070},
+	{0xD10, 0x20110FFF},
+	{0xD18, 0x50009800},
+	{0xD20, 0x01900000},
+	{0xD30, 0x03FF8000},
+	{0xD40, 0xF64FA0F7},
+	{0xD44, 0x0401463F},
+	{0xD48, 0x0003FF7F},
+	{0xD4C, 0x00000000},
+	{0xD50, 0xF64FA0F7},
+	{0xD54, 0x04100437},
+	{0xD58, 0x0000FF7F},
+	{0xD5C, 0x00000000},
+	{0xD60, 0x00000000},
+	{0xD64, 0x00000000},
+	{0xD70, 0x00000015},
+	{0xD78, 0x00000001},
+	{0xD7C, 0x001D050E},
+	{0xD80, 0x00000100},
+	{0xD84, 0x00006607},
+	{0xD90, 0x000003FF},
+	{0xD94, 0x00000000},
+	{0xD98, 0x0000003F},
+	{0xD9C, 0x00000000},
+	{0xDA0, 0x000003FE},
+	{0xDA4, 0x00000000},
+	{0xDA8, 0x0000003F},
+	{0xDAC, 0x00000000},
+	{0xDD4, 0x00000000},
+	{0x1010, 0x00000000},
+	{0x2000, 0x50BBBF04},
+	{0x2008, 0x000FFFFF},
+	{0x5800, 0x03FF807F},
+	{0x5804, 0x04237040},
+	{0x5808, 0x04237040},
+	{0x5818, 0x082C1800},
+	{0x624, 0x0101030A},
+	{0x241C, 0x00000001},
+	{0xC0F8, 0x00000001},
+	{0x35C, 0x000004C4},
+	{0x1200, 0x00010142},
+	{0x120C, 0x00012233},
+	{0x1210, 0x8049E304},
+	{0x12A0, 0x49107056},
+	{0x12A8, 0x33337025},
+	{0x12AC, 0x12333121},
+	{0x12B8, 0x30020000},
+	{0x0F0, 0x00000001},
+	{0x0F4, 0x00000011},
+	{0x0F8, 0x20230307},
+};
+
+static const struct rtw89_reg2_def rtw89_8851b_phy_bb_reg_gain[] = {
+	{0xF00100FF, 0x00000000},
+	{0xF00200FF, 0x00000001},
+	{0xF00300FF, 0x00000002},
+	{0xF00400FF, 0x00000003},
+	{0xF00500FF, 0x00000004},
+	{0xF00600FF, 0x00000005},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x000, 0x13F6D7B6},
+	{0x001, 0x00725132},
+	{0x002, 0x00005A38},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x000, 0x13F6D7B6},
+	{0x001, 0x00725132},
+	{0x002, 0x00005A38},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x000, 0x13F6D7B6},
+	{0x001, 0x00725132},
+	{0x002, 0x00005A38},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x000, 0x19FADCBC},
+	{0x001, 0x007A5A3A},
+	{0x002, 0x00005838},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x000, 0x19FADCBC},
+	{0x001, 0x007A5A3A},
+	{0x002, 0x00005838},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x000, 0x19FADCBC},
+	{0x001, 0x007A5A3A},
+	{0x002, 0x00005838},
+	{0xA0000000, 0x00000000},
+	{0x000, 0x13F6D7B6},
+	{0x001, 0x00725132},
+	{0x002, 0x00005A38},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0xA0000000, 0x00000000},
+	{0x100, 0x1BFEE0B7},
+	{0x101, 0x006C5238},
+	{0x102, 0x00005031},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10000, 0x19F8D8C1},
+	{0x10001, 0x006F4F31},
+	{0x10002, 0x00006F58},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10000, 0x19F8D8C1},
+	{0x10001, 0x006F4F31},
+	{0x10002, 0x00006F58},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10000, 0x19F8D8C1},
+	{0x10001, 0x006F4F31},
+	{0x10002, 0x00006F58},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10000, 0x1DF8DAC1},
+	{0x10001, 0x00755437},
+	{0x10002, 0x00007058},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10000, 0x1DF8DAC1},
+	{0x10001, 0x00755437},
+	{0x10002, 0x00007058},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10000, 0x1DF8DAC1},
+	{0x10001, 0x00755437},
+	{0x10002, 0x00007058},
+	{0xA0000000, 0x00000000},
+	{0x10000, 0x19F8D8C1},
+	{0x10001, 0x006F4F31},
+	{0x10002, 0x00006F58},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0xA0000000, 0x00000000},
+	{0x10100, 0x09E9C69F},
+	{0x10101, 0x00674627},
+	{0x10102, 0x00006750},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20000, 0x1AF0D2B8},
+	{0x20001, 0x00755334},
+	{0x20002, 0x00006F58},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20000, 0x1AF0D2B8},
+	{0x20001, 0x00755334},
+	{0x20002, 0x00006F58},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20000, 0x1AF0D2B8},
+	{0x20001, 0x00755334},
+	{0x20002, 0x00006F58},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20000, 0x1D00E2C8},
+	{0x20001, 0x00775336},
+	{0x20002, 0x00006D58},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20000, 0x1D00E2C8},
+	{0x20001, 0x00775336},
+	{0x20002, 0x00006D58},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20000, 0x1D00E2C8},
+	{0x20001, 0x00775336},
+	{0x20002, 0x00006D58},
+	{0xA0000000, 0x00000000},
+	{0x20000, 0x1AF0D2B8},
+	{0x20001, 0x00755334},
+	{0x20002, 0x00006F58},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0xA0000000, 0x00000000},
+	{0x20100, 0x07E9C6A0},
+	{0x20101, 0x00674728},
+	{0x20102, 0x00006850},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30000, 0x15EED2B6},
+	{0x30001, 0x006F4D2F},
+	{0x30002, 0x00006F58},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30000, 0x15EED2B6},
+	{0x30001, 0x006F4D2F},
+	{0x30002, 0x00006F58},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30000, 0x15EED2B6},
+	{0x30001, 0x006F4D2F},
+	{0x30002, 0x00006F58},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30000, 0x1F00E2C6},
+	{0x30001, 0x00795739},
+	{0x30002, 0x00006F58},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30000, 0x1F00E2C6},
+	{0x30001, 0x00795739},
+	{0x30002, 0x00006F58},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30000, 0x1F00E2C6},
+	{0x30001, 0x00795739},
+	{0x30002, 0x00006F58},
+	{0xA0000000, 0x00000000},
+	{0x30000, 0x15EED2B6},
+	{0x30001, 0x006F4D2F},
+	{0x30002, 0x00006F58},
+	{0xB0000000, 0x00000000},
+	{0x800100ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0x900200ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0x900300ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0x900400ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0x900500ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0x900600ff, 0x00000000}, {0x40000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0xA0000000, 0x00000000},
+	{0x30100, 0x06E9C69F},
+	{0x30101, 0x00654527},
+	{0x30102, 0x00006750},
+	{0xB0000000, 0x00000000},
+	{0x1000000, 0x000000F4},
+	{0x1000010, 0x000000F8},
+	{0x1000011, 0x0000F8F8},
+	{0x1000100, 0x000000F8},
+	{0x1000110, 0x00000000},
+	{0x1000111, 0x00000000},
+	{0x1010000, 0x000000F4},
+	{0x1010010, 0x000000F8},
+	{0x1010011, 0x0000F8F8},
+	{0x1010020, 0x000000F8},
+	{0x1010021, 0x0808E8E8},
+	{0x1010029, 0x0000F8F8},
+	{0x1010100, 0x000000F4},
+	{0x1010110, 0x000000F8},
+	{0x1010111, 0x0000F8F8},
+	{0x1010120, 0x000000F8},
+	{0x1010121, 0x0808E8E8},
+	{0x1010129, 0x0000F8F8},
+	{0x1020000, 0x000000F4},
+	{0x1020010, 0x000000F8},
+	{0x1020011, 0x0000F8F8},
+	{0x1020020, 0x000000F8},
+	{0x1020021, 0x0808E8E8},
+	{0x1020029, 0x0000F8F8},
+	{0x1020100, 0x000000F4},
+	{0x1020110, 0x000000F8},
+	{0x1020111, 0x0000F8F8},
+	{0x1020120, 0x000000F8},
+	{0x1020121, 0x0808E8E8},
+	{0x1020129, 0x0000F8F8},
+	{0x1030000, 0x000000F4},
+	{0x1030010, 0x000000F8},
+	{0x1030011, 0x0000F8F8},
+	{0x1030020, 0x000000F8},
+	{0x1030021, 0x0808E8E8},
+	{0x1030029, 0x0000F8F8},
+	{0x1030100, 0x000000F4},
+	{0x1030110, 0x000000F8},
+	{0x1030111, 0x0000F8F8},
+	{0x1030120, 0x000000F8},
+	{0x1030121, 0x0808E8E8},
+	{0x1030129, 0x0000F8F8},
+};
+
+static const struct rtw89_reg2_def rtw89_8851b_phy_radioa_regs[] = {
+	{0xF0010000, 0x00000000},
+	{0xF0020000, 0x00000001},
+	{0xF0030000, 0x00000002},
+	{0x000, 0x00030000},
+	{0x018, 0x00013124},
+	{0x0EF, 0x00080000},
+	{0x033, 0x00000008},
+	{0x03E, 0x00000110},
+	{0x03F, 0x0000D39C},
+	{0x033, 0x0000000C},
+	{0x03E, 0x00000110},
+	{0x03F, 0x0000F79E},
+	{0x0EF, 0x00000000},
+	{0x01B, 0x00003A40},
+	{0x08F, 0x000C170E},
+	{0x08E, 0x00005160},
+	{0x002, 0x00000600},
+	{0x0EE, 0x00000002},
+	{0x033, 0x00000002},
+	{0x03F, 0x0000003F},
+	{0x033, 0x00000003},
+	{0x03F, 0x0000003F},
+	{0x033, 0x00000004},
+	{0x03F, 0x0000003F},
+	{0x033, 0x00000005},
+	{0x03F, 0x0000003F},
+	{0x033, 0x00000006},
+	{0x03F, 0x0000003F},
+	{0x033, 0x00000007},
+	{0x03F, 0x0000003F},
+	{0x033, 0x00000008},
+	{0x03F, 0x0000003F},
+	{0x033, 0x0000000C},
+	{0x03F, 0x0000003F},
+	{0x033, 0x0000000D},
+	{0x03F, 0x0000003F},
+	{0x033, 0x0000000E},
+	{0x03F, 0x0000003F},
+	{0x0EE, 0x00000000},
+	{0x0EF, 0x00004000},
+	{0x033, 0x00000007},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00000707},
+	{0x033, 0x00000006},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00000704},
+	{0x033, 0x00000005},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00020500},
+	{0x033, 0x00000004},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00010404},
+	{0x033, 0x00000003},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00099B04},
+	{0x033, 0x00000002},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00092B04},
+	{0x033, 0x00000001},
+	{0x03E, 0x00000000},
+	{0x03F, 0x000B3204},
+	{0x033, 0x00000000},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00003000},
+	{0x033, 0x00000017},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00000787},
+	{0x033, 0x00000016},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00000784},
+	{0x033, 0x00000015},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00020580},
+	{0x033, 0x00000014},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00010484},
+	{0x033, 0x00000013},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00099B84},
+	{0x033, 0x00000012},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00092B84},
+	{0x033, 0x00000011},
+	{0x03E, 0x00000000},
+	{0x03F, 0x000B3284},
+	{0x033, 0x00000010},
+	{0x03E, 0x00000000},
+	{0x03F, 0x00003080},
+	{0x0EF, 0x00000000},
+	{0x0EE, 0x00000010},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000003},
+	{0x033, 0x00000007},
+	{0x03F, 0x00000003},
+	{0x0EE, 0x00000000},
+	{0x0EF, 0x00001000},
+	{0x033, 0x00000000},
+	{0x03F, 0x00000034},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000037},
+	{0x033, 0x00000002},
+	{0x03F, 0x00000034},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000024},
+	{0x033, 0x00000004},
+	{0x03F, 0x00000037},
+	{0x033, 0x00000005},
+	{0x03F, 0x00000027},
+	{0x0EF, 0x00000000},
+	{0x0EC, 0x00000400},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000022},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000022},
+	{0x033, 0x00000009},
+	{0x03F, 0x00000022},
+	{0x0EC, 0x00000000},
+	{0x0EC, 0x00000004},
+	{0x033, 0x00000000},
+	{0x03F, 0x000000AE},
+	{0x033, 0x00000001},
+	{0x03F, 0x0000008C},
+	{0x033, 0x00000002},
+	{0x03F, 0x0000006A},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000048},
+	{0x033, 0x00000004},
+	{0x03F, 0x00000026},
+	{0x033, 0x00000005},
+	{0x03F, 0x00000004},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000002},
+	{0x033, 0x00000007},
+	{0x03F, 0x00000000},
+	{0x0EC, 0x00000000},
+	{0x0EF, 0x00008000},
+	{0x033, 0x00000007},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x00000006},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x00000005},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001DB0},
+	{0x033, 0x00000004},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001CB0},
+	{0x033, 0x00000003},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001BB0},
+	{0x033, 0x00000002},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001AB0},
+	{0x033, 0x00000001},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D9BC},
+	{0x033, 0x00000000},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D4BC},
+	{0x033, 0x00000017},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x00000016},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x00000015},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001DB0},
+	{0x033, 0x00000014},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001CB0},
+	{0x033, 0x00000013},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001BB0},
+	{0x033, 0x00000012},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001AB0},
+	{0x033, 0x00000011},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D9BC},
+	{0x033, 0x00000010},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D4BC},
+	{0x033, 0x00000027},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x00000026},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x00000025},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001DB0},
+	{0x033, 0x00000024},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001CB0},
+	{0x033, 0x00000023},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001BB0},
+	{0x033, 0x00000022},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001AB0},
+	{0x033, 0x00000021},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D9BC},
+	{0x033, 0x00000020},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D4BC},
+	{0x033, 0x0000000E},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x0000000D},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001DB0},
+	{0x033, 0x0000000C},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001CB0},
+	{0x033, 0x0000000B},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00091BB0},
+	{0x033, 0x0000000A},
+	{0x03E, 0x00000003},
+	{0x03F, 0x000A9AB0},
+	{0x033, 0x00000009},
+	{0x03E, 0x00000003},
+	{0x03F, 0x000BD9BC},
+	{0x033, 0x00000008},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0009D4BC},
+	{0x033, 0x0000001E},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x0000001D},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001DB0},
+	{0x033, 0x0000001C},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001CB0},
+	{0x033, 0x0000001B},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00091BB0},
+	{0x033, 0x0000001A},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00099AB0},
+	{0x033, 0x00000019},
+	{0x03E, 0x00000003},
+	{0x03F, 0x000AD9BC},
+	{0x033, 0x00000018},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0009D4BC},
+	{0x033, 0x0000002E},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001FB0},
+	{0x033, 0x0000002D},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001DB0},
+	{0x033, 0x0000002C},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001CB0},
+	{0x033, 0x0000002B},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00001BB0},
+	{0x033, 0x0000002A},
+	{0x03E, 0x00000003},
+	{0x03F, 0x00009AB0},
+	{0x033, 0x00000029},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0009D9BC},
+	{0x033, 0x00000028},
+	{0x03E, 0x00000003},
+	{0x03F, 0x0000D4BC},
+	{0x0EF, 0x00000000},
+	{0x0EF, 0x00002000},
+	{0x033, 0x00000000},
+	{0x03F, 0x00000005},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000004},
+	{0x033, 0x00000002},
+	{0x03F, 0x00000004},
+	{0x033, 0x00000004},
+	{0x03F, 0x0000000C},
+	{0x033, 0x00000005},
+	{0x03F, 0x00000003},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000003},
+	{0x0EF, 0x00000000},
+	{0x06C, 0x00038085},
+	{0x06D, 0x00000D6B},
+	{0x06E, 0x0001FB89},
+	{0x06F, 0x00097B99},
+	{0x069, 0x00008040},
+	{0x0EF, 0x00000200},
+	{0x033, 0x00000004},
+	{0x03F, 0x000008FF},
+	{0x033, 0x00000005},
+	{0x03F, 0x000004F2},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000217},
+	{0x033, 0x00000007},
+	{0x03F, 0x00000131},
+	{0x0EF, 0x00000000},
+	{0x0EF, 0x00000400},
+	{0x033, 0x00000004},
+	{0x03F, 0x000004F7},
+	{0x033, 0x00000005},
+	{0x03F, 0x000004F7},
+	{0x033, 0x00000006},
+	{0x03F, 0x000004F2},
+	{0x033, 0x00000007},
+	{0x03F, 0x00000117},
+	{0x0EF, 0x00000000},
+	{0x043, 0x00005000},
+	{0x036, 0x000147D0},
+	{0x0B0, 0x0008677C},
+	{0x0B1, 0x00012920},
+	{0x0BB, 0x000EF000},
+	{0x0CB, 0x000A9594},
+	{0x0CC, 0x000C36D2},
+	{0x0CD, 0x00024923},
+	{0x0CE, 0x00020180},
+	{0x0CF, 0x00000000},
+	{0x0D5, 0x0006E27A},
+	{0x0D8, 0x00000044},
+	{0x0D9, 0x00000007},
+	{0x0DD, 0x00000020},
+	{0x0E3, 0x0000002C},
+	{0x0B7, 0x0000000C},
+	{0x0E1, 0x000080C0},
+	{0x0E4, 0x00000380},
+	{0x0ED, 0x00002000},
+	{0x033, 0x00000001},
+	{0x03D, 0x000A6094},
+	{0x03E, 0x00003449},
+	{0x03F, 0x00000001},
+	{0x033, 0x00000003},
+	{0x03D, 0x000AA094},
+	{0x03E, 0x00003449},
+	{0x03F, 0x00000001},
+	{0x0ED, 0x00000000},
+	{0x0ED, 0x00000100},
+	{0x033, 0x00000000},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000001},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000002},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000003},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000004},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000005},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000007},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000008},
+	{0x03F, 0x0000007F},
+	{0x033, 0x00000009},
+	{0x03F, 0x0000007F},
+	{0x0ED, 0x00000000},
+	{0x0ED, 0x00000080},
+	{0x033, 0x00000000},
+	{0x03E, 0x000007E1},
+	{0x03F, 0x0001F87F},
+	{0x033, 0x00000010},
+	{0x03E, 0x000007E1},
+	{0x03F, 0x0001F87F},
+	{0x033, 0x00000030},
+	{0x03E, 0x000007E1},
+	{0x03F, 0x0001F87F},
+	{0x033, 0x00000040},
+	{0x03E, 0x000007E1},
+	{0x03F, 0x0001F87F},
+	{0x033, 0x00000050},
+	{0x03E, 0x000007E1},
+	{0x03F, 0x0001F87F},
+	{0x033, 0x00000070},
+	{0x03E, 0x000007E1},
+	{0x03F, 0x0001F87F},
+	{0x0ED, 0x00000000},
+	{0x0ED, 0x00000004},
+	{0x033, 0x00000000},
+	{0x03F, 0x00008420},
+	{0x0ED, 0x00000000},
+	{0x018, 0x00011108},
+	{0x0B9, 0x00000000},
+	{0x0B9, 0x00000000},
+	{0x0B9, 0x00000200},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0FF, 0x00000000},
+	{0x0B9, 0x00000000},
+	{0x018, 0x00013124},
+	{0x05A, 0x0006808F},
+	{0x0ED, 0x00000008},
+	{0x033, 0x00000001},
+	{0x03F, 0x0000000F},
+	{0x0ED, 0x00000000},
+	{0x000, 0x00020000},
+	{0x018, 0x00010124},
+	{0x0EE, 0x00000800},
+	{0x033, 0x00000004},
+	{0x03F, 0x00000002},
+	{0x033, 0x00000005},
+	{0x03F, 0x00000003},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000006},
+	{0x033, 0x00000007},
+	{0x03F, 0x00000007},
+	{0x0EE, 0x00000000},
+	{0x0EE, 0x00001000},
+	{0x033, 0x00000008},
+	{0x03F, 0x00000000},
+	{0x033, 0x00000009},
+	{0x03F, 0x00000001},
+	{0x033, 0x0000000A},
+	{0x03F, 0x00000003},
+	{0x033, 0x0000000B},
+	{0x03F, 0x00000103},
+	{0x033, 0x0000000C},
+	{0x03F, 0x00000107},
+	{0x033, 0x0000000D},
+	{0x03F, 0x00000207},
+	{0x033, 0x0000000E},
+	{0x03F, 0x00000307},
+	{0x033, 0x0000000F},
+	{0x03F, 0x00000307},
+	{0x0EE, 0x00000000},
+	{0x0EE, 0x00000200},
+	{0x033, 0x00000004},
+	{0x03F, 0x00000000},
+	{0x033, 0x00000005},
+	{0x03F, 0x00000001},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000002},
+	{0x033, 0x00000007},
+	{0x03F, 0x00000003},
+	{0x0EE, 0x00000000},
+	{0x011, 0x00014062},
+	{0x0EF, 0x00000010},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000DF3},
+	{0x033, 0x00000002},
+	{0x03F, 0x00000DF3},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000A83},
+	{0x033, 0x00000004},
+	{0x03F, 0x00000A83},
+	{0x033, 0x00000005},
+	{0x03F, 0x00000643},
+	{0x033, 0x00000006},
+	{0x03F, 0x00000643},
+	{0x0EF, 0x00000000},
+	{0x0EF, 0x00000100},
+	{0x033, 0x00000001},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000002},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000003},
+	{0x03F, 0x0001B5A9},
+	{0x033, 0x00000004},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000005},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000006},
+	{0x03F, 0x0001B589},
+	{0x033, 0x00000007},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000008},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000009},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000000A},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000000B},
+	{0x03F, 0x0001B5A9},
+	{0x033, 0x0000000C},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000000D},
+	{0x03F, 0x0001B5A9},
+	{0x033, 0x0000000E},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000000F},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000010},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000011},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000012},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000013},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000014},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000015},
+	{0x03F, 0x0001B589},
+	{0x033, 0x00000016},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000017},
+	{0x03F, 0x0001B5A9},
+	{0x033, 0x00000018},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000019},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000001A},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000001B},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000001C},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000001D},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000001E},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x0000001F},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000020},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000021},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000022},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000023},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000024},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000025},
+	{0x03F, 0x0001B5A8},
+	{0x033, 0x00000022},
+	{0x03F, 0x0001B5A8},
+	{0x0EF, 0x00000000},
+	{0x0EF, 0x00000040},
+	{0x033, 0x00000001},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000002},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000003},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000004},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000005},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000006},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000007},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000008},
+	{0x03F, 0x000002C5},
+	{0x033, 0x00000009},
+	{0x03F, 0x000002C5},
+	{0x033, 0x0000000A},
+	{0x03F, 0x000002C5},
+	{0x033, 0x0000000B},
+	{0x03F, 0x000002C5},
+	{0x0EF, 0x00000000},
+	{0x059, 0x00050033},
+	{0x061, 0x0005F48A},
+	{0x062, 0x00077435},
+	{0x063, 0x000F80A2},
+	{0x065, 0x00018F22},
+	{0x067, 0x00008060},
+	{0x07E, 0x0009780B},
+	{0x0EE, 0x00000004},
+	{0x033, 0x0000000B},
+	{0x03F, 0x0000000B},
+	{0x033, 0x0000000C},
+	{0x03F, 0x00000012},
+	{0x033, 0x0000000D},
+	{0x03F, 0x00000019},
+	{0x033, 0x0000000F},
+	{0x03F, 0x0000000B},
+	{0x033, 0x00000010},
+	{0x03F, 0x00000012},
+	{0x033, 0x00000011},
+	{0x03F, 0x00000019},
+	{0x03F, 0x00000000},
+	{0x0EE, 0x00000000},
+	{0x0EE, 0x00000800},
+	{0x033, 0x00000000},
+	{0x03F, 0x00000001},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000002},
+	{0x033, 0x00000002},
+	{0x03F, 0x00000003},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000007},
+	{0x0EE, 0x00000000},
+	{0x0EE, 0x00001000},
+	{0x033, 0x00000000},
+	{0x03F, 0x00003000},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000000},
+	{0x033, 0x00000002},
+	{0x03F, 0x00000001},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000003},
+	{0x033, 0x00000004},
+	{0x03F, 0x00000007},
+	{0x033, 0x00000005},
+	{0x03F, 0x0000000F},
+	{0x033, 0x00000006},
+	{0x03F, 0x0000010F},
+	{0x033, 0x00000007},
+	{0x03F, 0x0000030F},
+	{0x0EE, 0x00000000},
+	{0x0EE, 0x00000200},
+	{0x033, 0x00000000},
+	{0x03F, 0x00000004},
+	{0x033, 0x00000001},
+	{0x03F, 0x00000005},
+	{0x033, 0x00000002},
+	{0x03F, 0x00000006},
+	{0x033, 0x00000003},
+	{0x03F, 0x00000007},
+	{0x0EE, 0x00000000},
+	{0x0EF, 0x00000080},
+	{0x033, 0x00000004},
+	{0x03E, 0x0000001D},
+	{0x03F, 0x0001A241},
+	{0x033, 0x00000005},
+	{0x03E, 0x0000001D},
+	{0x03F, 0x0001A241},
+	{0x033, 0x00000006},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000007},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000008},
+	{0x03E, 0x0000001D},
+	{0x03F, 0x0001A241},
+	{0x033, 0x00000009},
+	{0x03E, 0x0001A241},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000000A},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000000B},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000000C},
+	{0x03E, 0x0000001D},
+	{0x03F, 0x0001A241},
+	{0x033, 0x0000000D},
+	{0x03E, 0x0000001D},
+	{0x03F, 0x0001A241},
+	{0x033, 0x0000000E},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000000F},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000010},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000011},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000012},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000013},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000014},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000015},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000016},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000017},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000018},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000019},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x0000001A},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000001B},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000001C},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x0000001D},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x0000001E},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000001F},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000020},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000021},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x000199C1},
+	{0x033, 0x00000022},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000023},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000024},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000025},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000026},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000027},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000028},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000029},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x0000002A},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000002B},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000002C},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x0000002D},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x0000002E},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000002F},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000030},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000031},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000032},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000033},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000034},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000035},
+	{0x03E, 0x0000001C},
+	{0x03F, 0x0001E141},
+	{0x033, 0x00000036},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000037},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000038},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x00000039},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C1},
+	{0x033, 0x0000003A},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C3},
+	{0x033, 0x0000003B},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C3},
+	{0x033, 0x0000003C},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C3},
+	{0x033, 0x0000003D},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C3},
+	{0x033, 0x0000003E},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C3},
+	{0x033, 0x0000003F},
+	{0x03E, 0x0000001B},
+	{0x03F, 0x0001C3C3},
+	{0x0EF, 0x00000000},
+	{0x051, 0x0003D368},
+	{0x052, 0x000A3338},
+	{0x053, 0x000688AF},
+	{0x054, 0x00012C04},
+	{0x058, 0x00084221},
+	{0x05B, 0x000EB000},
+	{0x100EE, 0x00002000},
+	{0x10030, 0x000000F9},
+	{0x10030, 0x000004F6},
+	{0x10030, 0x000008F3},
+	{0x10030, 0x00000CF0},
+	{0x10030, 0x000010ED},
+	{0x10030, 0x000014EA},
+	{0x10030, 0x000018E7},
+	{0x10030, 0x00001CE4},
+	{0x10030, 0x000020E1},
+	{0x10030, 0x000024A4},
+	{0x10030, 0x000028A1},
+	{0x10030, 0x00002C9E},
+	{0x10030, 0x0000309B},
+	{0x10030, 0x0000341E},
+	{0x10030, 0x0000381B},
+	{0x10030, 0x00003C18},
+	{0x10030, 0x00004015},
+	{0x10030, 0x000200BC},
+	{0x10030, 0x000204B9},
+	{0x10030, 0x000208B6},
+	{0x10030, 0x00020CB3},
+	{0x10030, 0x000210B0},
+	{0x10030, 0x000214AD},
+	{0x10030, 0x0002186C},
+	{0x10030, 0x00021C69},
+	{0x10030, 0x00022066},
+	{0x10030, 0x00022426},
+	{0x10030, 0x00022823},
+	{0x10030, 0x00022C20},
+	{0x10030, 0x0002301D},
+	{0x10030, 0x0002341A},
+	{0x10030, 0x00023817},
+	{0x10030, 0x00023C14},
+	{0x10030, 0x00024011},
+	{0x10030, 0x000280BC},
+	{0x10030, 0x000284B9},
+	{0x10030, 0x000288B6},
+	{0x10030, 0x00028CB3},
+	{0x10030, 0x000290B0},
+	{0x10030, 0x000294AD},
+	{0x10030, 0x0002986C},
+	{0x10030, 0x00029C69},
+	{0x10030, 0x0002A066},
+	{0x10030, 0x0002A426},
+	{0x10030, 0x0002A823},
+	{0x10030, 0x0002AC20},
+	{0x10030, 0x0002B01D},
+	{0x10030, 0x0002B41A},
+	{0x10030, 0x0002B817},
+	{0x10030, 0x0002BC14},
+	{0x10030, 0x0002C011},
+	{0x10030, 0x000300BC},
+	{0x10030, 0x000304B9},
+	{0x10030, 0x000308B6},
+	{0x10030, 0x00030CB3},
+	{0x10030, 0x000310B0},
+	{0x10030, 0x000314AD},
+	{0x10030, 0x0003186C},
+	{0x10030, 0x00031C69},
+	{0x10030, 0x00032066},
+	{0x10030, 0x00032426},
+	{0x10030, 0x00032823},
+	{0x10030, 0x00032C20},
+	{0x10030, 0x0003301D},
+	{0x10030, 0x0003341A},
+	{0x10030, 0x00033817},
+	{0x10030, 0x00033C14},
+	{0x10030, 0x00034011},
+	{0x100EE, 0x00000000},
+	{0x100EE, 0x00004000},
+	{0x10030, 0x000201EF},
+	{0x10030, 0x000205E9},
+	{0x10030, 0x000209E3},
+	{0x10030, 0x00020DDD},
+	{0x10030, 0x000211D7},
+	{0x10030, 0x000215D1},
+	{0x10030, 0x00021919},
+	{0x10030, 0x00021D13},
+	{0x10030, 0x000220D9},
+	{0x10030, 0x000224D3},
+	{0x10030, 0x00022899},
+	{0x10030, 0x00022C93},
+	{0x10030, 0x00023059},
+	{0x10030, 0x00023453},
+	{0x10030, 0x00023819},
+	{0x10030, 0x00023C13},
+	{0x10030, 0x0002400D},
+	{0x10030, 0x00024407},
+	{0x10030, 0x000281EF},
+	{0x10030, 0x000285E9},
+	{0x10030, 0x000289E3},
+	{0x10030, 0x00028DDD},
+	{0x10030, 0x000291D7},
+	{0x10030, 0x000295D1},
+	{0x10030, 0x00029919},
+	{0x10030, 0x00029D13},
+	{0x10030, 0x0002A0D9},
+	{0x10030, 0x0002A4D3},
+	{0x10030, 0x0002A899},
+	{0x10030, 0x0002AC93},
+	{0x10030, 0x0002B059},
+	{0x10030, 0x0002B453},
+	{0x10030, 0x0002B819},
+	{0x10030, 0x0002BC13},
+	{0x10030, 0x0002C00D},
+	{0x10030, 0x0002C407},
+	{0x10030, 0x000301EF},
+	{0x10030, 0x000305E9},
+	{0x10030, 0x000309E3},
+	{0x10030, 0x00030DDD},
+	{0x10030, 0x000311D7},
+	{0x10030, 0x000315D1},
+	{0x10030, 0x00031919},
+	{0x10030, 0x00031D13},
+	{0x10030, 0x000320D9},
+	{0x10030, 0x000324D3},
+	{0x10030, 0x00032899},
+	{0x10030, 0x00032C93},
+	{0x10030, 0x00033059},
+	{0x10030, 0x00033453},
+	{0x10030, 0x00033819},
+	{0x10030, 0x00033C13},
+	{0x10030, 0x0003400D},
+	{0x10030, 0x00034407},
+	{0x100EE, 0x00000000},
+	{0x100EE, 0x00004000},
+	{0x10030, 0x000001EF},
+	{0x10030, 0x000005E9},
+	{0x10030, 0x000009E3},
+	{0x10030, 0x00000DDD},
+	{0x10030, 0x000011A5},
+	{0x10030, 0x0000159F},
+	{0x10030, 0x00001965},
+	{0x10030, 0x00001D5F},
+	{0x10030, 0x00002125},
+	{0x10030, 0x0000251F},
+	{0x10030, 0x000028E5},
+	{0x10030, 0x00002CDF},
+	{0x10030, 0x000030A5},
+	{0x10030, 0x0000349F},
+	{0x10030, 0x00003865},
+	{0x10030, 0x00003C5F},
+	{0x10030, 0x00004025},
+	{0x10030, 0x0000441F},
+	{0x100EE, 0x00000000},
+	{0x0EF, 0x00000008},
+	{0x033, 0x00000000},
+	{0x03F, 0x00000004},
+	{0x0EF, 0x00000000},
+	{0x005, 0x00000001},
+	{0x10005, 0x00000001},
+	{0x0FE, 0x00000022},
+};
+
+static const struct rtw89_reg2_def rtw89_8851b_phy_nctl_regs[] = {
+	{0x8000, 0x00000008},
+	{0x8008, 0x00000000},
+	{0x8004, 0xe8862b66},
+	{0x800c, 0x78000000},
+	{0x8010, 0x88015000},
+	{0x8014, 0x80010100},
+	{0x8018, 0x10010100},
+	{0x801c, 0xa210bc00},
+	{0x8020, 0x000403e0},
+	{0x8024, 0x00072160},
+	{0x8028, 0x00180e00},
+	{0x8030, 0x400000c0},
+	{0x8034, 0x11000830},
+	{0x8038, 0x40000000},
+	{0x803c, 0x00000008},
+	{0x8040, 0x00000046},
+	{0x8044, 0x0010001f},
+	{0x8048, 0x00000003},
+	{0x804c, 0x420840e0},
+	{0x8050, 0xce08cce0},
+	{0x8054, 0x420840e0},
+	{0x8058, 0xce08cce0},
+	{0x805c, 0x150c0b02},
+	{0x8060, 0x150c0b02},
+	{0x8064, 0x2aa00047},
+	{0x8074, 0x80000000},
+	{0x807c, 0x000000ee},
+	{0x8088, 0x80000000},
+	{0x808c, 0x00000000},
+	{0x80b0, 0x00000000},
+	{0x80cc, 0x00000000},
+	{0x80d0, 0x00000000},
+	{0x80ec, 0x00000002},
+	{0x8098, 0x0000ff00},
+	{0x8070, 0x00e80000},
+	{0x80b0, 0xffe00fff},
+	{0x809c, 0x0000001f},
+	{0x80b8, 0x00002000},
+	{0x80bc, 0x00050033},
+	{0xa400, 0x00000000},
+	{0xa404, 0x00000180},
+	{0xa408, 0x000001af},
+	{0xa40c, 0x000001e3},
+	{0xa410, 0x00000220},
+	{0xa414, 0x00000262},
+	{0xa418, 0x000002ac},
+	{0xa41c, 0x0000035e},
+	{0xa420, 0x000003c7},
+	{0xa424, 0x0000043d},
+	{0xa428, 0x000004c1},
+	{0xa42c, 0x00000556},
+	{0xa430, 0x000005fc},
+	{0xa434, 0x000006b7},
+	{0xa438, 0x00000789},
+	{0xa43c, 0x00000875},
+	{0xa440, 0x0000011f},
+	{0x8104, 0x00000000},
+	{0x810c, 0x00000000},
+	{0x8110, 0x00000000},
+	{0x8114, 0x00000000},
+	{0x8120, 0x10010000},
+	{0x8124, 0x00000000},
+	{0x8128, 0x00000200},
+	{0x812c, 0x0000c000},
+	{0x8130, 0x40000000},
+	{0x8138, 0x40000000},
+	{0x813c, 0x40000000},
+	{0x8140, 0x00000000},
+	{0x8144, 0x0b040b03},
+	{0x8148, 0x07020b04},
+	{0x814c, 0x07020b04},
+	{0x8150, 0xe4e40000},
+	{0x8158, 0xffffffff},
+	{0x815c, 0xffffffff},
+	{0x8160, 0xffffffff},
+	{0x8164, 0xffffffff},
+	{0x8168, 0xffffffff},
+	{0x816c, 0x1fffffff},
+	{0x81cc, 0x00000000},
+	{0x81dc, 0x00000002},
+	{0x81e0, 0x00000000},
+	{0x81e4, 0x00000001},
+	{0x81a0, 0x00000000},
+	{0x81ac, 0x3fc20400},
+	{0x81b0, 0x3f914100},
+	{0x81bc, 0x0000005b},
+	{0x81c0, 0x0000005b},
+	{0x81b4, 0x01e0f078},
+	{0x81b8, 0x01e0f078},
+	{0x81f0, 0x0000f078},
+	{0x81d8, 0x00000001},
+	{0x9500, 0x00000000},
+	{0x9504, 0x00000000},
+	{0x9508, 0x00000000},
+	{0x950c, 0x00000000},
+	{0x9510, 0x00000000},
+	{0x9514, 0x00000000},
+	{0x9518, 0x00000000},
+	{0x951c, 0x00000000},
+	{0x9520, 0x00000000},
+	{0x9524, 0x00000000},
+	{0x9528, 0x00000000},
+	{0x952c, 0x00000000},
+	{0x9530, 0x00000000},
+	{0x9534, 0x00000000},
+	{0x9538, 0x00000000},
+	{0x953c, 0x00000000},
+	{0x9540, 0x04000000},
+	{0x9544, 0x00000000},
+	{0x9548, 0x00000000},
+	{0x954c, 0x00000000},
+	{0x9550, 0x00000000},
+	{0x9554, 0x00000000},
+	{0x9558, 0x00000000},
+	{0x955c, 0x00000000},
+	{0x9560, 0x00000000},
+	{0x9564, 0x00000000},
+	{0x9568, 0x00000000},
+	{0x956c, 0x00000000},
+	{0x9570, 0x00000000},
+	{0x9574, 0x00000000},
+	{0x9578, 0x00000000},
+	{0x957c, 0x00000000},
+	{0x9580, 0x00000000},
+	{0x9584, 0x04000000},
+	{0x9588, 0x00000000},
+	{0x958c, 0x00000000},
+	{0x9590, 0x00000000},
+	{0x9594, 0x00000000},
+	{0x9598, 0x00000000},
+	{0x959c, 0x00000000},
+	{0x95a0, 0x00000000},
+	{0x95a4, 0x00000000},
+	{0x95a8, 0x00000000},
+	{0x95ac, 0x00000000},
+	{0x95b0, 0x00000000},
+	{0x95b4, 0x00000000},
+	{0x95b8, 0x00000000},
+	{0x95bc, 0x00000000},
+	{0x95c0, 0x00000000},
+	{0x95c4, 0x00000000},
+	{0x95c8, 0x04000000},
+	{0x95cc, 0x00000000},
+	{0x95d0, 0x00000000},
+	{0x95d4, 0x00000000},
+	{0x95d8, 0x00000000},
+	{0x95dc, 0x00000000},
+	{0x95e0, 0x00000000},
+	{0x95e4, 0x00000000},
+	{0x95e8, 0x00000000},
+	{0x95ec, 0x00000000},
+	{0x95f0, 0x00000000},
+	{0x95f4, 0x00000000},
+	{0x95f8, 0x00000000},
+	{0x95fc, 0x00000000},
+	{0x9600, 0x00000000},
+	{0x9604, 0x00000000},
+	{0x9608, 0x00000000},
+	{0x960c, 0x04000000},
+	{0x9610, 0x00000000},
+	{0x9614, 0x00000000},
+	{0x9618, 0x00000000},
+	{0x961c, 0x00000000},
+	{0x9620, 0x00000000},
+	{0x9624, 0x00000000},
+	{0x9628, 0x00000000},
+	{0x962c, 0x00000000},
+	{0x9630, 0x00000000},
+	{0x9634, 0x00000000},
+	{0x9638, 0x00000000},
+	{0x963c, 0x00000000},
+	{0x9640, 0x00000000},
+	{0x9644, 0x00000000},
+	{0x9648, 0x00000000},
+	{0x964c, 0x00000000},
+	{0x9650, 0x04000000},
+	{0x9654, 0x00000000},
+	{0x9658, 0x00000000},
+	{0x965c, 0x00000000},
+	{0x9660, 0x00000000},
+	{0x9664, 0x00000000},
+	{0x9668, 0x00000000},
+	{0x966c, 0x00000000},
+	{0x9670, 0x00000000},
+	{0x9674, 0x00000000},
+	{0x9678, 0x00000000},
+	{0x967c, 0x00000000},
+	{0x9680, 0x00000000},
+	{0x9684, 0x00000000},
+	{0x9688, 0x00000000},
+	{0x968c, 0x00000000},
+	{0x9690, 0x00000000},
+	{0x9694, 0x04000000},
+	{0x9698, 0x00000000},
+	{0x969c, 0x00000000},
+	{0x96a0, 0x00000000},
+	{0x96a4, 0x00000000},
+	{0x96a8, 0x00000000},
+	{0x96ac, 0x00000000},
+	{0x96b0, 0x00000000},
+	{0x96b4, 0x00000000},
+	{0x96b8, 0x00000000},
+	{0x96bc, 0x00000000},
+	{0x96c0, 0x00000000},
+	{0x96c4, 0x00000000},
+	{0x96c8, 0x00000000},
+	{0x96cc, 0x00000000},
+	{0x96d0, 0x00000000},
+	{0x96d4, 0x00000000},
+	{0x96d8, 0x04000000},
+	{0x96dc, 0x00000000},
+	{0x96e0, 0x00000000},
+	{0x96e4, 0x00000000},
+	{0x96e8, 0x00000000},
+	{0x96ec, 0x00000000},
+	{0x96f0, 0x00000000},
+	{0x96f4, 0x00000000},
+	{0x96f8, 0x00000000},
+	{0x96fc, 0x00000000},
+	{0x9700, 0x00000000},
+	{0x9704, 0x00000000},
+	{0x9708, 0x00000000},
+	{0x970c, 0x00000000},
+	{0x9710, 0x00000000},
+	{0x9714, 0x00000000},
+	{0x9718, 0x00000000},
+	{0x971c, 0x04000000},
+	{0x9720, 0x00000000},
+	{0x9724, 0x00000000},
+	{0x9728, 0x00000000},
+	{0x972c, 0x00000000},
+	{0x9730, 0x00000000},
+	{0x9734, 0x00000000},
+	{0x9738, 0x00000000},
+	{0x973c, 0x00000000},
+	{0x9740, 0x00000000},
+	{0x9744, 0x00000000},
+	{0x9748, 0x00000000},
+	{0x974c, 0x00000000},
+	{0x9750, 0x00000000},
+	{0x9754, 0x00000000},
+	{0x9758, 0x00000000},
+	{0x975c, 0x00000000},
+	{0x9760, 0x04000000},
+	{0x9764, 0x00000000},
+	{0x9768, 0x00000000},
+	{0x976c, 0x00000000},
+	{0x9770, 0x00000000},
+	{0x9774, 0x00000000},
+	{0x9778, 0x00000000},
+	{0x977c, 0x00000000},
+	{0x9780, 0x00000000},
+	{0x9784, 0x00000000},
+	{0x9788, 0x00000000},
+	{0x978c, 0x00000000},
+	{0x9790, 0x00000000},
+	{0x9794, 0x00000000},
+	{0x9798, 0x00000000},
+	{0x979c, 0x00000000},
+	{0x97a0, 0x00000000},
+	{0x97a4, 0x04000000},
+	{0x97a8, 0x00000000},
+	{0x97ac, 0x00000000},
+	{0x97b0, 0x00000000},
+	{0x97b4, 0x00000000},
+	{0x97b8, 0x00000000},
+	{0x97bc, 0x00000000},
+	{0x97c0, 0x00000000},
+	{0x97c4, 0x00000000},
+	{0x97c8, 0x00000000},
+	{0x97cc, 0x00000000},
+	{0x97d0, 0x00000000},
+	{0x97d4, 0x00000000},
+	{0x97d8, 0x00000000},
+	{0x97dc, 0x00000000},
+	{0x97e0, 0x00000000},
+	{0x97e4, 0x00000000},
+	{0x97e8, 0x04000000},
+	{0x97ec, 0x00000000},
+	{0x97f0, 0x00000000},
+	{0x97f4, 0x00000000},
+	{0x97f8, 0x00000000},
+	{0x97fc, 0x00000000},
+	{0x9800, 0x00000000},
+	{0x9804, 0x00000000},
+	{0x9808, 0x00000000},
+	{0x980c, 0x00000000},
+	{0x9810, 0x00000000},
+	{0x9814, 0x00000000},
+	{0x9818, 0x00000000},
+	{0x981c, 0x00000000},
+	{0x9820, 0x00000000},
+	{0x9824, 0x00000000},
+	{0x9828, 0x00000000},
+	{0x982c, 0x04000000},
+	{0x81d8, 0x00000000},
+	{0xb104, 0x2b251f19},
+	{0xb108, 0x433d3731},
+	{0xb10c, 0x5b554f49},
+	{0xb110, 0x736d6761},
+	{0xb114, 0x7f7f7f79},
+	{0xb118, 0x120f7f7f},
+	{0xb11c, 0x1e1b1815},
+	{0xb120, 0x2a272421},
+	{0xb124, 0x3633302d},
+	{0xb128, 0x3f3f3c39},
+	{0xb12c, 0x3f3f3f3f},
+	{0x8088, 0x00000110},
+	{0x8000, 0x00000008},
+	{0x8080, 0x00000005},
+	{0x8500, 0x80000008},
+	{0x8504, 0x43000004},
+	{0x8508, 0x4b044a00},
+	{0x850c, 0x40098604},
+	{0x8510, 0x0004e01f},
+	{0x8514, 0x74104b00},
+	{0x8518, 0x000021e0},
+	{0x851c, 0x74301658},
+	{0x8520, 0x43800004},
+	{0x8524, 0x4c000007},
+	{0x8528, 0x43000004},
+	{0x852c, 0x56030007},
+	{0x8530, 0x57000004},
+	{0x8534, 0x400042fe},
+	{0x8538, 0x50554200},
+	{0x853c, 0xb4183000},
+	{0x8540, 0xe537a50f},
+	{0x8544, 0xf12bf02b},
+	{0x8548, 0xf32bf22b},
+	{0x854c, 0xf62bf42b},
+	{0x8550, 0xf82bf72b},
+	{0x8554, 0xfa2bf92b},
+	{0x8558, 0xfd2bfc2b},
+	{0x855c, 0xe537fe2b},
+	{0x8560, 0xf12af02a},
+	{0x8564, 0xf32af22a},
+	{0x8568, 0xf52af42a},
+	{0x856c, 0x000bf62a},
+	{0x8570, 0xf028a511},
+	{0x8574, 0xf228f128},
+	{0x8578, 0xf428f328},
+	{0x857c, 0xf628f528},
+	{0x8580, 0xf828f728},
+	{0x8584, 0xfa28f928},
+	{0x8588, 0xfc28fb28},
+	{0x858c, 0xfe28fd28},
+	{0x8590, 0xf028ff28},
+	{0x8594, 0xf228f128},
+	{0x8598, 0x30750001},
+	{0x859c, 0x30753075},
+	{0x85a0, 0x30b63097},
+	{0x85a4, 0x30be30bb},
+	{0x85a8, 0x30d930cc},
+	{0x85ac, 0x316d30e6},
+	{0x85b0, 0x3189317f},
+	{0x85b4, 0x31d23193},
+	{0x85b8, 0x31e43210},
+	{0x85bc, 0x31e831dd},
+	{0x85c0, 0x322831e1},
+	{0x85c4, 0x323c3232},
+	{0x85c8, 0x32503246},
+	{0x85cc, 0x3264325a},
+	{0x85d0, 0x3278326e},
+	{0x85d4, 0x32983285},
+	{0x85d8, 0x32aa32a6},
+	{0x85dc, 0x330b32f3},
+	{0x85e0, 0x333f330c},
+	{0x85e4, 0x334c3341},
+	{0x85e8, 0xe35e0001},
+	{0x85ec, 0x20887410},
+	{0x85f0, 0x140f0200},
+	{0x85f4, 0x02002098},
+	{0x85f8, 0x7430140f},
+	{0x85fc, 0x5b10e39c},
+	{0x8600, 0x20807410},
+	{0x8604, 0x140f0000},
+	{0x8608, 0x56015507},
+	{0x860c, 0x7410e382},
+	{0x8610, 0x02002088},
+	{0x8614, 0x5517140f},
+	{0x8618, 0xe34ee382},
+	{0x861c, 0x468e7508},
+	{0x8620, 0xe0ace38c},
+	{0x8624, 0x5500f0e2},
+	{0x8628, 0x5501e37e},
+	{0x862c, 0x5b10f1de},
+	{0x8630, 0x20907410},
+	{0x8634, 0x140f0000},
+	{0x8638, 0xe3825507},
+	{0x863c, 0x20987410},
+	{0x8640, 0x140f0200},
+	{0x8644, 0xe3825517},
+	{0x8648, 0x46967509},
+	{0x864c, 0xe0ace38c},
+	{0x8650, 0xe37e5500},
+	{0x8654, 0x00015501},
+	{0x8658, 0x4d000007},
+	{0x865c, 0x74200004},
+	{0x8660, 0x57005710},
+	{0x8664, 0x9700140f},
+	{0x8668, 0x00017430},
+	{0x866c, 0xe39ce35e},
+	{0x8670, 0xe52a0bbd},
+	{0x8674, 0xe36a0001},
+	{0x8678, 0x0001e3c4},
+	{0x867c, 0x55005b30},
+	{0x8680, 0x46500005},
+	{0x8684, 0x74000004},
+	{0x8688, 0x1658e37e},
+	{0x868c, 0x74305501},
+	{0x8690, 0x46100005},
+	{0x8694, 0x00010004},
+	{0x8698, 0x30f8e35e},
+	{0x869c, 0xe52a0023},
+	{0x86a0, 0x54ed0002},
+	{0x86a4, 0x00230baa},
+	{0x86a8, 0x0002e52a},
+	{0x86ac, 0xe356e3e4},
+	{0x86b0, 0xe35e0001},
+	{0x86b4, 0x002230f3},
+	{0x86b8, 0x0002e52a},
+	{0x86bc, 0x0baa54ec},
+	{0x86c0, 0xe52a0022},
+	{0x86c4, 0xe3e40002},
+	{0x86c8, 0x0001e356},
+	{0x86cc, 0x0baae35e},
+	{0x86d0, 0xe3e430ec},
+	{0x86d4, 0x0001e356},
+	{0x86d8, 0x6d0f6c67},
+	{0x86dc, 0xe52ae39c},
+	{0x86e0, 0xe39c6c8b},
+	{0x86e4, 0x0bace52a},
+	{0x86e8, 0x6d0f6cb3},
+	{0x86ec, 0xe52ae39c},
+	{0x86f0, 0x6cdb0bad},
+	{0x86f4, 0xe39c6d0f},
+	{0x86f8, 0x6cf5e52a},
+	{0x86fc, 0xe39c6d0f},
+	{0x8700, 0x6c0be52a},
+	{0x8704, 0xe39c6d00},
+	{0x8708, 0x6c25e52a},
+	{0x870c, 0xe52ae39c},
+	{0x8710, 0x6c4df8c6},
+	{0x8714, 0xe52ae39c},
+	{0x8718, 0x6c75f9cf},
+	{0x871c, 0xe52ae39c},
+	{0x8720, 0xe39c6c99},
+	{0x8724, 0xfad6e52a},
+	{0x8728, 0x21e87410},
+	{0x872c, 0x6e670009},
+	{0x8730, 0xe3c46f0f},
+	{0x8734, 0x7410e52f},
+	{0x8738, 0x000b21e8},
+	{0x873c, 0xe3c46e8b},
+	{0x8740, 0x7410e52f},
+	{0x8744, 0x000d21e8},
+	{0x8748, 0x6f0f6eb3},
+	{0x874c, 0xe52fe3c4},
+	{0x8750, 0xfe07ff08},
+	{0x8754, 0x21e87410},
+	{0x8758, 0x6ec7000e},
+	{0x875c, 0xe52fe3c4},
+	{0x8760, 0x21e87410},
+	{0x8764, 0x6edb000f},
+	{0x8768, 0xe3c46f0f},
+	{0x876c, 0x7410e52f},
+	{0x8770, 0x001021e8},
+	{0x8774, 0xe3c46eef},
+	{0x8778, 0xff03e52f},
+	{0x877c, 0xe52ffe02},
+	{0x8780, 0x21e87410},
+	{0x8784, 0x6e110013},
+	{0x8788, 0xe3c46f00},
+	{0x878c, 0xff03e52f},
+	{0x8790, 0xe52ffe02},
+	{0x8794, 0x21e87410},
+	{0x8798, 0x6e250014},
+	{0x879c, 0xe52fe3c4},
+	{0x87a0, 0xff08fc24},
+	{0x87a4, 0x7410fe07},
+	{0x87a8, 0x001521e8},
+	{0x87ac, 0xe3c46e39},
+	{0x87b0, 0x7410e52f},
+	{0x87b4, 0x001621e8},
+	{0x87b8, 0xe3c46e4d},
+	{0x87bc, 0xfd27e52f},
+	{0x87c0, 0x21e87410},
+	{0x87c4, 0x6e750018},
+	{0x87c8, 0xe52fe3c4},
+	{0x87cc, 0x21e87410},
+	{0x87d0, 0x6e99001a},
+	{0x87d4, 0xe52fe3c4},
+	{0x87d8, 0xe36afe24},
+	{0x87dc, 0x63404380},
+	{0x87e0, 0x43006880},
+	{0x87e4, 0x31300bac},
+	{0x87e8, 0xe52f0022},
+	{0x87ec, 0x54ec0002},
+	{0x87f0, 0x00220baa},
+	{0x87f4, 0x0002e52f},
+	{0x87f8, 0xe362e3e4},
+	{0x87fc, 0xe36a0001},
+	{0x8800, 0x63404380},
+	{0x8804, 0x43006881},
+	{0x8808, 0x31210baa},
+	{0x880c, 0xe362e3e4},
+	{0x8810, 0xe36a0001},
+	{0x8814, 0x63414380},
+	{0x8818, 0x43006882},
+	{0x881c, 0x31140baa},
+	{0x8820, 0xe362e3e4},
+	{0x8824, 0x00040001},
+	{0x8828, 0x000742fc},
+	{0x882c, 0x00046001},
+	{0x8830, 0x00074200},
+	{0x8834, 0x62006220},
+	{0x8838, 0x55010004},
+	{0x883c, 0x66055b40},
+	{0x8840, 0x62000007},
+	{0x8844, 0xe40e6300},
+	{0x8848, 0x09000004},
+	{0x884c, 0x0b400a01},
+	{0x8850, 0x0e010d00},
+	{0x8854, 0x00040032},
+	{0x8858, 0x42fb950b},
+	{0x885c, 0x4d040007},
+	{0x8860, 0x42000004},
+	{0x8864, 0x00074380},
+	{0x8868, 0x00044d01},
+	{0x886c, 0x00074300},
+	{0x8870, 0x05a30562},
+	{0x8874, 0xe40e961f},
+	{0x8878, 0xe37e0004},
+	{0x887c, 0x06a20007},
+	{0x8880, 0xe40e07a3},
+	{0x8884, 0xe37e0004},
+	{0x8888, 0x0002e3fe},
+	{0x888c, 0x4380e406},
+	{0x8890, 0x4d000007},
+	{0x8894, 0x43000004},
+	{0x8898, 0x000742fe},
+	{0x889c, 0x00044d00},
+	{0x88a0, 0x00014200},
+	{0x88a4, 0x42fc0004},
+	{0x88a8, 0x60030007},
+	{0x88ac, 0x42000004},
+	{0x88b0, 0x00073199},
+	{0x88b4, 0x07a306a2},
+	{0x88b8, 0xe1eb31c5},
+	{0x88bc, 0xe1fee1f9},
+	{0x88c0, 0xe1eb0001},
+	{0x88c4, 0x0001e1fe},
+	{0x88c8, 0xe1f9e1f2},
+	{0x88cc, 0x0001e1fe},
+	{0x88d0, 0xe1fee1f2},
+	{0x88d4, 0x00040001},
+	{0x88d8, 0x000742fc},
+	{0x88dc, 0x00046003},
+	{0x88e0, 0x00014200},
+	{0x88e4, 0x42fc0004},
+	{0x88e8, 0x60010007},
+	{0x88ec, 0x42000004},
+	{0x88f0, 0x00070001},
+	{0x88f4, 0x62006220},
+	{0x88f8, 0x0001e406},
+	{0x88fc, 0x63000007},
+	{0x8900, 0x09000004},
+	{0x8904, 0x0e010a00},
+	{0x8908, 0x00070032},
+	{0x890c, 0xe40e06a2},
+	{0x8910, 0x0002e41a},
+	{0x8914, 0x000742fe},
+	{0x8918, 0x00044d00},
+	{0x891c, 0x00014200},
+	{0x8920, 0x77000005},
+	{0x8924, 0x52000007},
+	{0x8928, 0x42fe0004},
+	{0x892c, 0x60000007},
+	{0x8930, 0x42000004},
+	{0x8934, 0x60004380},
+	{0x8938, 0x62016100},
+	{0x893c, 0x68046310},
+	{0x8940, 0x41000005},
+	{0x8944, 0x00075500},
+	{0x8948, 0x00045c02},
+	{0x894c, 0x00014300},
+	{0x8950, 0x6c060005},
+	{0x8954, 0xe2aae298},
+	{0x8958, 0xe42ae285},
+	{0x895c, 0xe432e2f3},
+	{0x8960, 0x0001e30c},
+	{0x8964, 0x0005e285},
+	{0x8968, 0xe2986c06},
+	{0x896c, 0xe42ae4a9},
+	{0x8970, 0xe432e2f3},
+	{0x8974, 0x0001e30c},
+	{0x8978, 0x6c000005},
+	{0x897c, 0xe2aae298},
+	{0x8980, 0xe445e285},
+	{0x8984, 0xe44de2f3},
+	{0x8988, 0x0001e30c},
+	{0x898c, 0x0005e285},
+	{0x8990, 0xe2986c00},
+	{0x8994, 0xe445e4a9},
+	{0x8998, 0xe44de2f3},
+	{0x899c, 0x0001e30c},
+	{0x89a0, 0x6c040005},
+	{0x89a4, 0xe2aae298},
+	{0x89a8, 0xe460e285},
+	{0x89ac, 0xe468e2f3},
+	{0x89b0, 0x0001e30c},
+	{0x89b4, 0x0005e285},
+	{0x89b8, 0xe2986c04},
+	{0x89bc, 0xe460e4a9},
+	{0x89c0, 0xe468e2f3},
+	{0x89c4, 0x0001e30c},
+	{0x89c8, 0x6c020005},
+	{0x89cc, 0xe2aae298},
+	{0x89d0, 0xe47be285},
+	{0x89d4, 0xe483e2f3},
+	{0x89d8, 0x0001e30c},
+	{0x89dc, 0x0005e285},
+	{0x89e0, 0xe2986c02},
+	{0x89e4, 0xe47be4a9},
+	{0x89e8, 0xe483e2f3},
+	{0x89ec, 0x0001e30c},
+	{0x89f0, 0x43800004},
+	{0x89f4, 0x610a6008},
+	{0x89f8, 0x63ce6200},
+	{0x89fc, 0x60800006},
+	{0x8a00, 0x00047f00},
+	{0x8a04, 0xe4e04300},
+	{0x8a08, 0x00070001},
+	{0x8a0c, 0x4d015500},
+	{0x8a10, 0x74200004},
+	{0x8a14, 0x57107711},
+	{0x8a18, 0x140f5700},
+	{0x8a1c, 0x00077430},
+	{0x8a20, 0x00044d00},
+	{0x8a24, 0x00074380},
+	{0x8a28, 0x00047200},
+	{0x8a2c, 0x00014300},
+	{0x8a30, 0x74200004},
+	{0x8a34, 0x77000005},
+	{0x8a38, 0x73887e07},
+	{0x8a3c, 0x8f007380},
+	{0x8a40, 0x0004140f},
+	{0x8a44, 0x00057430},
+	{0x8a48, 0x00017300},
+	{0x8a4c, 0x0005e496},
+	{0x8a50, 0x00017300},
+	{0x8a54, 0x43800004},
+	{0x8a58, 0x0006b103},
+	{0x8a5c, 0x91037cdb},
+	{0x8a60, 0x40db0007},
+	{0x8a64, 0x43000004},
+	{0x8a68, 0x0005e496},
+	{0x8a6c, 0x00067380},
+	{0x8a70, 0x60025d01},
+	{0x8a74, 0xe4ba6200},
+	{0x8a78, 0x73000005},
+	{0x8a7c, 0x76080007},
+	{0x8a80, 0x00047578},
+	{0x8a84, 0x00074380},
+	{0x8a88, 0x5e005e01},
+	{0x8a8c, 0x0006140a},
+	{0x8a90, 0x7f006380},
+	{0x8a94, 0x00076080},
+	{0x8a98, 0x4e204c3f},
+	{0x8a9c, 0x73047280},
+	{0x8aa0, 0x140a7300},
+	{0x8aa4, 0x00044d20},
+	{0x8aa8, 0x00064300},
+	{0x8aac, 0x00077402},
+	{0x8ab0, 0x40004001},
+	{0x8ab4, 0x0006ab00},
+	{0x8ab8, 0x00077404},
+	{0x8abc, 0x40004001},
+	{0x8ac0, 0x140aab00},
+	{0x8ac4, 0x43800004},
+	{0x8ac8, 0x52800007},
+	{0x8acc, 0x140a5200},
+	{0x8ad0, 0x4d004c00},
+	{0x8ad4, 0x00064e00},
+	{0x8ad8, 0x63006080},
+	{0x8adc, 0x43000004},
+	{0x8ae0, 0x76000007},
+	{0x8ae4, 0x00040001},
+	{0x8ae8, 0xb1034380},
+	{0x8aec, 0x7cdb0006},
+	{0x8af0, 0x00079103},
+	{0x8af4, 0x000440db},
+	{0x8af8, 0xe4964300},
+	{0x8afc, 0xe4ba7e03},
+	{0x8b00, 0x43800004},
+	{0x8b04, 0x0006b103},
+	{0x8b08, 0x91037c5b},
+	{0x8b0c, 0x405b0007},
+	{0x8b10, 0x43000004},
+	{0x8b14, 0x00010001},
+	{0x8b18, 0x43800004},
+	{0x8b1c, 0x4e200007},
+	{0x8b20, 0x63800006},
+	{0x8b24, 0x5f807cdb},
+	{0x8b28, 0x43000004},
+	{0x8b2c, 0x76080007},
+	{0x8b30, 0x00057560},
+	{0x8b34, 0x00047380},
+	{0x8b38, 0x0005420e},
+	{0x8b3c, 0x14c86c01},
+	{0x8b40, 0x6c001432},
+	{0x8b44, 0x42000004},
+	{0x8b48, 0x43800004},
+	{0x8b4c, 0x5f000006},
+	{0x8b50, 0x73010007},
+	{0x8b54, 0x00047300},
+	{0x8b58, 0x0007420f},
+	{0x8b5c, 0x52005280},
+	{0x8b60, 0x0004140a},
+	{0x8b64, 0x00064200},
+	{0x8b68, 0x7c5b6300},
+	{0x8b6c, 0x4e000007},
+	{0x8b70, 0x43000004},
+	{0x8b74, 0x73000005},
+	{0x8b78, 0x76000007},
+	{0x8b7c, 0xe4c30001},
+	{0x8b80, 0x00040001},
+	{0x8b84, 0x60004380},
+	{0x8b88, 0x62016100},
+	{0x8b8c, 0x00066310},
+	{0x8b90, 0x00046000},
+	{0x8b94, 0x00014300},
+	{0x8b98, 0x0001e4e0},
+	{0x8b9c, 0x4e004f02},
+	{0x8ba0, 0x52015302},
+	{0x8ba4, 0x140f0001},
+	{0x8ba8, 0x00019700},
+	{0x8bac, 0x65014380},
+	{0x8bb0, 0x79007800},
+	{0x8bb4, 0x7b407a00},
+	{0x8bb8, 0x00014300},
+	{0x8bbc, 0x65004380},
+	{0x8bc0, 0x00014300},
+	{0x8bc4, 0x64014380},
+	{0x8bc8, 0x7d007c00},
+	{0x8bcc, 0x7f407e00},
+	{0x8bd0, 0x00014300},
+	{0x8bd4, 0x64004380},
+	{0x8bd8, 0x00014300},
+	{0x8bdc, 0x7b004380},
+	{0x8be0, 0x79007a04},
+	{0x8be4, 0x43007802},
+	{0x8be8, 0x33825509},
+	{0x8bec, 0x43800001},
+	{0x8bf0, 0x7a007b40},
+	{0x8bf4, 0x55194300},
+	{0x8bf8, 0x00013382},
+	{0x8bfc, 0x74007401},
+	{0x8c00, 0x00018e00},
+	{0x8c04, 0x52300007},
+	{0x8c08, 0x74310004},
+	{0x8c0c, 0x8e007430},
+	{0x8c10, 0x52200007},
+	{0x8c14, 0x00010004},
+	{0x8c18, 0x57005702},
+	{0x8c1c, 0x00018e00},
+	{0x8c20, 0x57425740},
+	{0x8c24, 0x8e005740},
+	{0x8c28, 0x00015700},
+	{0x8c2c, 0x561042ef},
+	{0x8c30, 0x42005600},
+	{0x8c34, 0x00018c00},
+	{0x8c38, 0xe3a75b20},
+	{0x8c3c, 0x54005480},
+	{0x8c40, 0x54005481},
+	{0x8c44, 0x54005482},
+	{0x8c48, 0xbf1ae3ac},
+	{0x8c4c, 0xe36e300b},
+	{0x8c50, 0xe390e377},
+	{0x8c54, 0x0001e523},
+	{0x8c58, 0x54c054bf},
+	{0x8c5c, 0x54c154a3},
+	{0x8c60, 0x4c1854a4},
+	{0x8c64, 0xbf091402},
+	{0x8c68, 0x54a454c2},
+	{0x8c6c, 0xbf051402},
+	{0x8c70, 0x54a354c1},
+	{0x8c74, 0xbf011402},
+	{0x8c78, 0x54dfe534},
+	{0x8c7c, 0x54bf0001},
+	{0x8c80, 0x050a54e5},
+	{0x8c84, 0x000154df},
+	{0x8c88, 0x00071657},
+	{0x8c8c, 0x00044c80},
+	{0x8c90, 0x43807430},
+	{0x8c94, 0x7e007f40},
+	{0x8c98, 0x7c027d00},
+	{0x8c9c, 0x5b404300},
+	{0x8ca0, 0x5c015501},
+	{0x8ca4, 0x5480e396},
+	{0x8ca8, 0x54815400},
+	{0x8cac, 0x54825400},
+	{0x8cb0, 0x00075400},
+	{0x8cb4, 0x00044c00},
+	{0x8cb8, 0xe3ac7410},
+	{0x8cbc, 0x300bbfe1},
+	{0x8cc0, 0x56005610},
+	{0x8cc4, 0x00018c00},
+	{0x8cc8, 0x57005704},
+	{0x8ccc, 0xa7038e00},
+	{0x8cd0, 0x33f0aff7},
+	{0x8cd4, 0xaf034019},
+	{0x8cd8, 0x33f0402b},
+	{0x8cdc, 0x33df402b},
+	{0x8ce0, 0x57005708},
+	{0x8ce4, 0x57818e00},
+	{0x8ce8, 0x8e005780},
+	{0x8cec, 0x00074380},
+	{0x8cf0, 0x5c005c01},
+	{0x8cf4, 0x00041403},
+	{0x8cf8, 0x00014300},
+	{0x8cfc, 0x0007427f},
+	{0x8d00, 0x62006280},
+	{0x8d04, 0x00049200},
+	{0x8d08, 0x00014200},
+	{0x8d0c, 0x0007427f},
+	{0x8d10, 0x63146394},
+	{0x8d14, 0x00049200},
+	{0x8d18, 0x00014200},
+	{0x8d1c, 0x42fe0004},
+	{0x8d20, 0x4d010007},
+	{0x8d24, 0x42000004},
+	{0x8d28, 0x140f7420},
+	{0x8d2c, 0x57005710},
+	{0x8d30, 0x0001141f},
+	{0x8d34, 0x42fe0004},
+	{0x8d38, 0x4d010007},
+	{0x8d3c, 0x42000004},
+	{0x8d40, 0x140f7420},
+	{0x8d44, 0x000742bf},
+	{0x8d48, 0x62006240},
+	{0x8d4c, 0x0004141f},
+	{0x8d50, 0x00014200},
+	{0x8d54, 0x5d060006},
+	{0x8d58, 0x61046003},
+	{0x8d5c, 0x00056201},
+	{0x8d60, 0x00017310},
+	{0x8d64, 0x43800004},
+	{0x8d68, 0x5e010007},
+	{0x8d6c, 0x140a5e00},
+	{0x8d70, 0x0006b103},
+	{0x8d74, 0x91037f07},
+	{0x8d78, 0x43070007},
+	{0x8d7c, 0x5c000006},
+	{0x8d80, 0x5e035d02},
+	{0x8d84, 0x43000004},
+	{0x8d88, 0x00060001},
+	{0x8d8c, 0x60005d04},
+	{0x8d90, 0x62016104},
+	{0x8d94, 0x73100005},
+	{0x8d98, 0x00040001},
+	{0x8d9c, 0x00074380},
+	{0x8da0, 0x5e005e01},
+	{0x8da4, 0xb103140a},
+	{0x8da8, 0x7fc60006},
+	{0x8dac, 0x00079103},
+	{0x8db0, 0x000643c6},
+	{0x8db4, 0x5d025c00},
+	{0x8db8, 0x00045e03},
+	{0x8dbc, 0x00014300},
+	{0x8dc0, 0x5d040006},
+	{0x8dc4, 0x61046000},
+	{0x8dc8, 0x00056201},
+	{0x8dcc, 0x00017310},
+	{0x8dd0, 0x43800004},
+	{0x8dd4, 0x5e010007},
+	{0x8dd8, 0x140a5e00},
+	{0x8ddc, 0x0006b103},
+	{0x8de0, 0x91037fc6},
+	{0x8de4, 0x43c60007},
+	{0x8de8, 0x5c000006},
+	{0x8dec, 0x5e035d02},
+	{0x8df0, 0x43000004},
+	{0x8df4, 0x00060001},
+	{0x8df8, 0x60025d00},
+	{0x8dfc, 0x62016100},
+	{0x8e00, 0x73000005},
+	{0x8e04, 0x00040001},
+	{0x8e08, 0x00074380},
+	{0x8e0c, 0x5e005e01},
+	{0x8e10, 0xb103140a},
+	{0x8e14, 0x7fc00006},
+	{0x8e18, 0x00079103},
+	{0x8e1c, 0x000643c0},
+	{0x8e20, 0x5d025c00},
+	{0x8e24, 0x00045e03},
+	{0x8e28, 0x00014300},
+	{0x8e2c, 0x7e020005},
+	{0x8e30, 0x42f70004},
+	{0x8e34, 0x6c080005},
+	{0x8e38, 0x42700004},
+	{0x8e3c, 0x73810005},
+	{0x8e40, 0x93007380},
+	{0x8e44, 0x42f70004},
+	{0x8e48, 0x6c000005},
+	{0x8e4c, 0x42000004},
+	{0x8e50, 0x00040001},
+	{0x8e54, 0x00074380},
+	{0x8e58, 0x73007304},
+	{0x8e5c, 0x72401405},
+	{0x8e60, 0x43000004},
+	{0x8e64, 0x74040006},
+	{0x8e68, 0x40010007},
+	{0x8e6c, 0xab004000},
+	{0x8e70, 0x0001140f},
+	{0x8e74, 0x140ae517},
+	{0x8e78, 0x140ae4c3},
+	{0x8e7c, 0x0001e51e},
+	{0x8e80, 0xe4c3e517},
+	{0x8e84, 0x00040001},
+	{0x8e88, 0x00047410},
+	{0x8e8c, 0x42f04380},
+	{0x8e90, 0x62080007},
+	{0x8e94, 0x24206301},
+	{0x8e98, 0x14c80000},
+	{0x8e9c, 0x00002428},
+	{0x8ea0, 0x1a4215f4},
+	{0x8ea4, 0x6300000b},
+	{0x8ea8, 0x42000004},
+	{0x8eac, 0x74304300},
+	{0x8eb0, 0x4380140f},
+	{0x8eb4, 0x73080007},
+	{0x8eb8, 0x00047300},
+	{0x8ebc, 0x00014300},
+	{0x8ec0, 0x4bf00007},
+	{0x8ec4, 0x490b4a8f},
+	{0x8ec8, 0x4a8e48f1},
+	{0x8ecc, 0x48a5490a},
+	{0x8ed0, 0x49094a8d},
+	{0x8ed4, 0x4a8c487d},
+	{0x8ed8, 0x48754908},
+	{0x8edc, 0x49074a8b},
+	{0x8ee0, 0x4a8a4889},
+	{0x8ee4, 0x48b74906},
+	{0x8ee8, 0x49054a89},
+	{0x8eec, 0x4a8848fc},
+	{0x8ef0, 0x48564905},
+	{0x8ef4, 0x49044a87},
+	{0x8ef8, 0x4a8648c1},
+	{0x8efc, 0x483d4904},
+	{0x8f00, 0x49034a85},
+	{0x8f04, 0x4a8448c7},
+	{0x8f08, 0x485e4903},
+	{0x8f0c, 0x49024a83},
+	{0x8f10, 0x4a8248ac},
+	{0x8f14, 0x48624902},
+	{0x8f18, 0x49024a81},
+	{0x8f1c, 0x4a804820},
+	{0x8f20, 0x48004900},
+	{0x8f24, 0x49014a90},
+	{0x8f28, 0x4a10481f},
+	{0x8f2c, 0x00060001},
+	{0x8f30, 0x5f005f80},
+	{0x8f34, 0x00059900},
+	{0x8f38, 0x00017300},
+	{0x8f3c, 0x63800006},
+	{0x8f40, 0x98006300},
+	{0x8f44, 0x549f0001},
+	{0x8f48, 0x5c015400},
+	{0x8f4c, 0x540054df},
+	{0x8f50, 0x00015c02},
+	{0x8f54, 0x07145c01},
+	{0x8f58, 0x5c025400},
+	{0x8f5c, 0x5c020001},
+	{0x8f60, 0x54000714},
+	{0x8f64, 0x00015c01},
+	{0x8f68, 0x4c184c98},
+	{0x8f6c, 0x00080001},
+	{0x8f70, 0x5c020004},
+	{0x8f74, 0x09017430},
+	{0x8f78, 0x0ba60c01},
+	{0x8f7c, 0x77800005},
+	{0x8f80, 0x52200007},
+	{0x8f84, 0x43800004},
+	{0x8f88, 0x610a6008},
+	{0x8f8c, 0x63c26200},
+	{0x8f90, 0x5c000007},
+	{0x8f94, 0x43000004},
+	{0x8f98, 0x00000001},
+	{0x8080, 0x00000004},
+	{0x8080, 0x00000000},
+	{0x8088, 0x00000000},
+};
+
+static const struct rtw89_txpwr_byrate_cfg rtw89_8851b_txpwr_byrate[] = {
+	{ 0, 0, 0, 0, 4, 0x50505050, },
+	{ 0, 0, 1, 0, 4, 0x54585858, },
+	{ 0, 0, 1, 4, 4, 0x44484c50, },
+	{ 0, 0, 2, 0, 4, 0x50545858, },
+	{ 0, 0, 2, 4, 4, 0x4044484c, },
+	{ 0, 0, 2, 8, 4, 0x3034383c, },
+	{ 0, 0, 3, 0, 4, 0x50505050, },
+	{ 0, 1, 2, 0, 4, 0x50545858, },
+	{ 0, 1, 2, 4, 4, 0x4044484c, },
+	{ 0, 1, 2, 8, 4, 0x3034383c, },
+	{ 0, 1, 3, 0, 4, 0x50505050, },
+	{ 0, 0, 4, 1, 4, 0x00000000, },
+	{ 0, 0, 4, 0, 1, 0x00000000, },
+	{ 1, 0, 1, 0, 4, 0x58585858, },
+	{ 1, 0, 1, 4, 4, 0x484c5054, },
+	{ 1, 0, 2, 0, 4, 0x54585858, },
+	{ 1, 0, 2, 4, 4, 0x44484c50, },
+	{ 1, 0, 2, 8, 4, 0x34383c40, },
+	{ 1, 0, 3, 0, 4, 0x40404040, },
+	{ 1, 1, 2, 0, 4, 0x54585858, },
+	{ 1, 1, 2, 4, 4, 0x44484c50, },
+	{ 1, 1, 2, 8, 4, 0x34383c40, },
+	{ 1, 1, 3, 0, 4, 0x48484848, },
+	{ 1, 0, 4, 0, 4, 0x00000000, },
+	{ 2, 0, 1, 0, 4, 0x40404040, },
+	{ 2, 0, 1, 4, 4, 0x383c4040, },
+	{ 2, 0, 2, 0, 4, 0x40404040, },
+	{ 2, 0, 2, 4, 4, 0x34383c40, },
+	{ 2, 0, 2, 8, 4, 0x24282c30, },
+	{ 2, 0, 3, 0, 4, 0x40404040, },
+	{ 2, 1, 2, 0, 4, 0x40404040, },
+	{ 2, 1, 2, 4, 4, 0x34383c40, },
+	{ 2, 1, 2, 8, 4, 0x24282c30, },
+	{ 2, 1, 3, 0, 4, 0x40404040, },
+	{ 2, 0, 4, 0, 4, 0x00000000, },
+};
+
+static const s8 _txpwr_track_delta_swingidx_5ga_n[][DELTA_SWINGIDX_SIZE] = {
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0,
+	 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1},
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1,
+	 1, 2, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4},
+};
+
+static const s8 _txpwr_track_delta_swingidx_5ga_p[][DELTA_SWINGIDX_SIZE] = {
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1,
+	 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2,
+	 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2},
+};
+
+static const s8 _txpwr_track_delta_swingidx_2ga_n[] = {
+	0, 0, 0, 0, -1, -1, -1, -2, -2, -2, -2, -3, -3, -3, -3, -3,
+	 -4, -4, -4, -4, -4, -5, -5, -5, -5, -5, -5, -6, -6, -6
+};
+
+static const s8 _txpwr_track_delta_swingidx_2ga_p[] = {
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	 0, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 4
+};
+
+static const s8 _txpwr_track_delta_swingidx_2g_cck_a_n[] = {
+	0, 0, 0, 0, -1, -1, -1, -2, -2, -2, -2, -3, -3, -3, -3, -3,
+	 -4, -4, -4, -4, -4, -5, -5, -5, -5, -5, -5, -6, -6, -6
+};
+
+static const s8 _txpwr_track_delta_swingidx_2g_cck_a_p[] = {
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	 0, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 4
+};
+
+const u8 rtw89_8851b_tx_shape[RTW89_BAND_MAX][RTW89_RS_TX_SHAPE_NUM]
+			     [RTW89_REGD_NUM] = {
+	[0][0][RTW89_ACMA] = 0,
+	[0][0][RTW89_CN] = 0,
+	[0][0][RTW89_ETSI] = 0,
+	[0][0][RTW89_FCC] = 1,
+	[0][0][RTW89_IC] = 1,
+	[0][0][RTW89_KCC] = 0,
+	[0][0][RTW89_MKK] = 0,
+	[0][0][RTW89_UK] = 0,
+	[0][1][RTW89_ACMA] = 0,
+	[0][1][RTW89_CN] = 0,
+	[0][1][RTW89_ETSI] = 0,
+	[0][1][RTW89_FCC] = 3,
+	[0][1][RTW89_IC] = 3,
+	[0][1][RTW89_KCC] = 0,
+	[0][1][RTW89_MKK] = 0,
+	[0][1][RTW89_UK] = 0,
+	[1][1][RTW89_ACMA] = 0,
+	[1][1][RTW89_CN] = 0,
+	[1][1][RTW89_ETSI] = 0,
+	[1][1][RTW89_FCC] = 3,
+	[1][1][RTW89_IC] = 3,
+	[1][1][RTW89_KCC] = 0,
+	[1][1][RTW89_MKK] = 0,
+	[1][1][RTW89_UK] = 0,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_2g[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
+				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+				 [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
+	[0][0][0][0][RTW89_WW][0] = 58,
+	[0][0][0][0][RTW89_WW][1] = 58,
+	[0][0][0][0][RTW89_WW][2] = 58,
+	[0][0][0][0][RTW89_WW][3] = 58,
+	[0][0][0][0][RTW89_WW][4] = 58,
+	[0][0][0][0][RTW89_WW][5] = 58,
+	[0][0][0][0][RTW89_WW][6] = 58,
+	[0][0][0][0][RTW89_WW][7] = 58,
+	[0][0][0][0][RTW89_WW][8] = 58,
+	[0][0][0][0][RTW89_WW][9] = 58,
+	[0][0][0][0][RTW89_WW][10] = 58,
+	[0][0][0][0][RTW89_WW][11] = 58,
+	[0][0][0][0][RTW89_WW][12] = 52,
+	[0][0][0][0][RTW89_WW][13] = 76,
+	[0][1][0][0][RTW89_WW][0] = 0,
+	[0][1][0][0][RTW89_WW][1] = 0,
+	[0][1][0][0][RTW89_WW][2] = 0,
+	[0][1][0][0][RTW89_WW][3] = 0,
+	[0][1][0][0][RTW89_WW][4] = 0,
+	[0][1][0][0][RTW89_WW][5] = 0,
+	[0][1][0][0][RTW89_WW][6] = 0,
+	[0][1][0][0][RTW89_WW][7] = 0,
+	[0][1][0][0][RTW89_WW][8] = 0,
+	[0][1][0][0][RTW89_WW][9] = 0,
+	[0][1][0][0][RTW89_WW][10] = 0,
+	[0][1][0][0][RTW89_WW][11] = 0,
+	[0][1][0][0][RTW89_WW][12] = 0,
+	[0][1][0][0][RTW89_WW][13] = 0,
+	[1][0][0][0][RTW89_WW][0] = 0,
+	[1][0][0][0][RTW89_WW][1] = 0,
+	[1][0][0][0][RTW89_WW][2] = 58,
+	[1][0][0][0][RTW89_WW][3] = 58,
+	[1][0][0][0][RTW89_WW][4] = 58,
+	[1][0][0][0][RTW89_WW][5] = 58,
+	[1][0][0][0][RTW89_WW][6] = 58,
+	[1][0][0][0][RTW89_WW][7] = 58,
+	[1][0][0][0][RTW89_WW][8] = 58,
+	[1][0][0][0][RTW89_WW][9] = 58,
+	[1][0][0][0][RTW89_WW][10] = 58,
+	[1][0][0][0][RTW89_WW][11] = 0,
+	[1][0][0][0][RTW89_WW][12] = 0,
+	[1][0][0][0][RTW89_WW][13] = 0,
+	[1][1][0][0][RTW89_WW][0] = 0,
+	[1][1][0][0][RTW89_WW][1] = 0,
+	[1][1][0][0][RTW89_WW][2] = 0,
+	[1][1][0][0][RTW89_WW][3] = 0,
+	[1][1][0][0][RTW89_WW][4] = 0,
+	[1][1][0][0][RTW89_WW][5] = 0,
+	[1][1][0][0][RTW89_WW][6] = 0,
+	[1][1][0][0][RTW89_WW][7] = 0,
+	[1][1][0][0][RTW89_WW][8] = 0,
+	[1][1][0][0][RTW89_WW][9] = 0,
+	[1][1][0][0][RTW89_WW][10] = 0,
+	[1][1][0][0][RTW89_WW][11] = 0,
+	[1][1][0][0][RTW89_WW][12] = 0,
+	[1][1][0][0][RTW89_WW][13] = 0,
+	[0][0][1][0][RTW89_WW][0] = 58,
+	[0][0][1][0][RTW89_WW][1] = 60,
+	[0][0][1][0][RTW89_WW][2] = 60,
+	[0][0][1][0][RTW89_WW][3] = 60,
+	[0][0][1][0][RTW89_WW][4] = 60,
+	[0][0][1][0][RTW89_WW][5] = 60,
+	[0][0][1][0][RTW89_WW][6] = 60,
+	[0][0][1][0][RTW89_WW][7] = 60,
+	[0][0][1][0][RTW89_WW][8] = 60,
+	[0][0][1][0][RTW89_WW][9] = 60,
+	[0][0][1][0][RTW89_WW][10] = 60,
+	[0][0][1][0][RTW89_WW][11] = 60,
+	[0][0][1][0][RTW89_WW][12] = 58,
+	[0][0][1][0][RTW89_WW][13] = 0,
+	[0][1][1][0][RTW89_WW][0] = 0,
+	[0][1][1][0][RTW89_WW][1] = 0,
+	[0][1][1][0][RTW89_WW][2] = 0,
+	[0][1][1][0][RTW89_WW][3] = 0,
+	[0][1][1][0][RTW89_WW][4] = 0,
+	[0][1][1][0][RTW89_WW][5] = 0,
+	[0][1][1][0][RTW89_WW][6] = 0,
+	[0][1][1][0][RTW89_WW][7] = 0,
+	[0][1][1][0][RTW89_WW][8] = 0,
+	[0][1][1][0][RTW89_WW][9] = 0,
+	[0][1][1][0][RTW89_WW][10] = 0,
+	[0][1][1][0][RTW89_WW][11] = 0,
+	[0][1][1][0][RTW89_WW][12] = 0,
+	[0][1][1][0][RTW89_WW][13] = 0,
+	[0][0][2][0][RTW89_WW][0] = 60,
+	[0][0][2][0][RTW89_WW][1] = 60,
+	[0][0][2][0][RTW89_WW][2] = 60,
+	[0][0][2][0][RTW89_WW][3] = 60,
+	[0][0][2][0][RTW89_WW][4] = 60,
+	[0][0][2][0][RTW89_WW][5] = 60,
+	[0][0][2][0][RTW89_WW][6] = 60,
+	[0][0][2][0][RTW89_WW][7] = 60,
+	[0][0][2][0][RTW89_WW][8] = 60,
+	[0][0][2][0][RTW89_WW][9] = 60,
+	[0][0][2][0][RTW89_WW][10] = 60,
+	[0][0][2][0][RTW89_WW][11] = 60,
+	[0][0][2][0][RTW89_WW][12] = 60,
+	[0][0][2][0][RTW89_WW][13] = 0,
+	[0][1][2][0][RTW89_WW][0] = 0,
+	[0][1][2][0][RTW89_WW][1] = 0,
+	[0][1][2][0][RTW89_WW][2] = 0,
+	[0][1][2][0][RTW89_WW][3] = 0,
+	[0][1][2][0][RTW89_WW][4] = 0,
+	[0][1][2][0][RTW89_WW][5] = 0,
+	[0][1][2][0][RTW89_WW][6] = 0,
+	[0][1][2][0][RTW89_WW][7] = 0,
+	[0][1][2][0][RTW89_WW][8] = 0,
+	[0][1][2][0][RTW89_WW][9] = 0,
+	[0][1][2][0][RTW89_WW][10] = 0,
+	[0][1][2][0][RTW89_WW][11] = 0,
+	[0][1][2][0][RTW89_WW][12] = 0,
+	[0][1][2][0][RTW89_WW][13] = 0,
+	[0][1][2][1][RTW89_WW][0] = 0,
+	[0][1][2][1][RTW89_WW][1] = 0,
+	[0][1][2][1][RTW89_WW][2] = 0,
+	[0][1][2][1][RTW89_WW][3] = 0,
+	[0][1][2][1][RTW89_WW][4] = 0,
+	[0][1][2][1][RTW89_WW][5] = 0,
+	[0][1][2][1][RTW89_WW][6] = 0,
+	[0][1][2][1][RTW89_WW][7] = 0,
+	[0][1][2][1][RTW89_WW][8] = 0,
+	[0][1][2][1][RTW89_WW][9] = 0,
+	[0][1][2][1][RTW89_WW][10] = 0,
+	[0][1][2][1][RTW89_WW][11] = 0,
+	[0][1][2][1][RTW89_WW][12] = 0,
+	[0][1][2][1][RTW89_WW][13] = 0,
+	[1][0][2][0][RTW89_WW][0] = 0,
+	[1][0][2][0][RTW89_WW][1] = 0,
+	[1][0][2][0][RTW89_WW][2] = 58,
+	[1][0][2][0][RTW89_WW][3] = 58,
+	[1][0][2][0][RTW89_WW][4] = 58,
+	[1][0][2][0][RTW89_WW][5] = 58,
+	[1][0][2][0][RTW89_WW][6] = 58,
+	[1][0][2][0][RTW89_WW][7] = 58,
+	[1][0][2][0][RTW89_WW][8] = 58,
+	[1][0][2][0][RTW89_WW][9] = 58,
+	[1][0][2][0][RTW89_WW][10] = 58,
+	[1][0][2][0][RTW89_WW][11] = 0,
+	[1][0][2][0][RTW89_WW][12] = 0,
+	[1][0][2][0][RTW89_WW][13] = 0,
+	[1][1][2][0][RTW89_WW][0] = 0,
+	[1][1][2][0][RTW89_WW][1] = 0,
+	[1][1][2][0][RTW89_WW][2] = 0,
+	[1][1][2][0][RTW89_WW][3] = 0,
+	[1][1][2][0][RTW89_WW][4] = 0,
+	[1][1][2][0][RTW89_WW][5] = 0,
+	[1][1][2][0][RTW89_WW][6] = 0,
+	[1][1][2][0][RTW89_WW][7] = 0,
+	[1][1][2][0][RTW89_WW][8] = 0,
+	[1][1][2][0][RTW89_WW][9] = 0,
+	[1][1][2][0][RTW89_WW][10] = 0,
+	[1][1][2][0][RTW89_WW][11] = 0,
+	[1][1][2][0][RTW89_WW][12] = 0,
+	[1][1][2][0][RTW89_WW][13] = 0,
+	[1][1][2][1][RTW89_WW][0] = 0,
+	[1][1][2][1][RTW89_WW][1] = 0,
+	[1][1][2][1][RTW89_WW][2] = 0,
+	[1][1][2][1][RTW89_WW][3] = 0,
+	[1][1][2][1][RTW89_WW][4] = 0,
+	[1][1][2][1][RTW89_WW][5] = 0,
+	[1][1][2][1][RTW89_WW][6] = 0,
+	[1][1][2][1][RTW89_WW][7] = 0,
+	[1][1][2][1][RTW89_WW][8] = 0,
+	[1][1][2][1][RTW89_WW][9] = 0,
+	[1][1][2][1][RTW89_WW][10] = 0,
+	[1][1][2][1][RTW89_WW][11] = 0,
+	[1][1][2][1][RTW89_WW][12] = 0,
+	[1][1][2][1][RTW89_WW][13] = 0,
+	[0][0][0][0][RTW89_FCC][0] = 84,
+	[0][0][0][0][RTW89_ETSI][0] = 58,
+	[0][0][0][0][RTW89_MKK][0] = 68,
+	[0][0][0][0][RTW89_IC][0] = 84,
+	[0][0][0][0][RTW89_KCC][0] = 68,
+	[0][0][0][0][RTW89_ACMA][0] = 58,
+	[0][0][0][0][RTW89_CN][0] = 60,
+	[0][0][0][0][RTW89_UK][0] = 58,
+	[0][0][0][0][RTW89_FCC][1] = 84,
+	[0][0][0][0][RTW89_ETSI][1] = 58,
+	[0][0][0][0][RTW89_MKK][1] = 68,
+	[0][0][0][0][RTW89_IC][1] = 84,
+	[0][0][0][0][RTW89_KCC][1] = 68,
+	[0][0][0][0][RTW89_ACMA][1] = 58,
+	[0][0][0][0][RTW89_CN][1] = 60,
+	[0][0][0][0][RTW89_UK][1] = 58,
+	[0][0][0][0][RTW89_FCC][2] = 84,
+	[0][0][0][0][RTW89_ETSI][2] = 58,
+	[0][0][0][0][RTW89_MKK][2] = 68,
+	[0][0][0][0][RTW89_IC][2] = 84,
+	[0][0][0][0][RTW89_KCC][2] = 68,
+	[0][0][0][0][RTW89_ACMA][2] = 58,
+	[0][0][0][0][RTW89_CN][2] = 60,
+	[0][0][0][0][RTW89_UK][2] = 58,
+	[0][0][0][0][RTW89_FCC][3] = 84,
+	[0][0][0][0][RTW89_ETSI][3] = 58,
+	[0][0][0][0][RTW89_MKK][3] = 68,
+	[0][0][0][0][RTW89_IC][3] = 84,
+	[0][0][0][0][RTW89_KCC][3] = 68,
+	[0][0][0][0][RTW89_ACMA][3] = 58,
+	[0][0][0][0][RTW89_CN][3] = 60,
+	[0][0][0][0][RTW89_UK][3] = 58,
+	[0][0][0][0][RTW89_FCC][4] = 84,
+	[0][0][0][0][RTW89_ETSI][4] = 58,
+	[0][0][0][0][RTW89_MKK][4] = 68,
+	[0][0][0][0][RTW89_IC][4] = 84,
+	[0][0][0][0][RTW89_KCC][4] = 68,
+	[0][0][0][0][RTW89_ACMA][4] = 58,
+	[0][0][0][0][RTW89_CN][4] = 60,
+	[0][0][0][0][RTW89_UK][4] = 58,
+	[0][0][0][0][RTW89_FCC][5] = 84,
+	[0][0][0][0][RTW89_ETSI][5] = 58,
+	[0][0][0][0][RTW89_MKK][5] = 68,
+	[0][0][0][0][RTW89_IC][5] = 84,
+	[0][0][0][0][RTW89_KCC][5] = 68,
+	[0][0][0][0][RTW89_ACMA][5] = 58,
+	[0][0][0][0][RTW89_CN][5] = 60,
+	[0][0][0][0][RTW89_UK][5] = 58,
+	[0][0][0][0][RTW89_FCC][6] = 84,
+	[0][0][0][0][RTW89_ETSI][6] = 58,
+	[0][0][0][0][RTW89_MKK][6] = 68,
+	[0][0][0][0][RTW89_IC][6] = 84,
+	[0][0][0][0][RTW89_KCC][6] = 68,
+	[0][0][0][0][RTW89_ACMA][6] = 58,
+	[0][0][0][0][RTW89_CN][6] = 60,
+	[0][0][0][0][RTW89_UK][6] = 58,
+	[0][0][0][0][RTW89_FCC][7] = 84,
+	[0][0][0][0][RTW89_ETSI][7] = 58,
+	[0][0][0][0][RTW89_MKK][7] = 68,
+	[0][0][0][0][RTW89_IC][7] = 84,
+	[0][0][0][0][RTW89_KCC][7] = 68,
+	[0][0][0][0][RTW89_ACMA][7] = 58,
+	[0][0][0][0][RTW89_CN][7] = 60,
+	[0][0][0][0][RTW89_UK][7] = 58,
+	[0][0][0][0][RTW89_FCC][8] = 84,
+	[0][0][0][0][RTW89_ETSI][8] = 58,
+	[0][0][0][0][RTW89_MKK][8] = 68,
+	[0][0][0][0][RTW89_IC][8] = 84,
+	[0][0][0][0][RTW89_KCC][8] = 68,
+	[0][0][0][0][RTW89_ACMA][8] = 58,
+	[0][0][0][0][RTW89_CN][8] = 60,
+	[0][0][0][0][RTW89_UK][8] = 58,
+	[0][0][0][0][RTW89_FCC][9] = 84,
+	[0][0][0][0][RTW89_ETSI][9] = 58,
+	[0][0][0][0][RTW89_MKK][9] = 68,
+	[0][0][0][0][RTW89_IC][9] = 84,
+	[0][0][0][0][RTW89_KCC][9] = 68,
+	[0][0][0][0][RTW89_ACMA][9] = 58,
+	[0][0][0][0][RTW89_CN][9] = 60,
+	[0][0][0][0][RTW89_UK][9] = 58,
+	[0][0][0][0][RTW89_FCC][10] = 82,
+	[0][0][0][0][RTW89_ETSI][10] = 58,
+	[0][0][0][0][RTW89_MKK][10] = 68,
+	[0][0][0][0][RTW89_IC][10] = 82,
+	[0][0][0][0][RTW89_KCC][10] = 68,
+	[0][0][0][0][RTW89_ACMA][10] = 58,
+	[0][0][0][0][RTW89_CN][10] = 60,
+	[0][0][0][0][RTW89_UK][10] = 58,
+	[0][0][0][0][RTW89_FCC][11] = 62,
+	[0][0][0][0][RTW89_ETSI][11] = 58,
+	[0][0][0][0][RTW89_MKK][11] = 68,
+	[0][0][0][0][RTW89_IC][11] = 62,
+	[0][0][0][0][RTW89_KCC][11] = 68,
+	[0][0][0][0][RTW89_ACMA][11] = 58,
+	[0][0][0][0][RTW89_CN][11] = 60,
+	[0][0][0][0][RTW89_UK][11] = 58,
+	[0][0][0][0][RTW89_FCC][12] = 52,
+	[0][0][0][0][RTW89_ETSI][12] = 58,
+	[0][0][0][0][RTW89_MKK][12] = 68,
+	[0][0][0][0][RTW89_IC][12] = 52,
+	[0][0][0][0][RTW89_KCC][12] = 68,
+	[0][0][0][0][RTW89_ACMA][12] = 58,
+	[0][0][0][0][RTW89_CN][12] = 60,
+	[0][0][0][0][RTW89_UK][12] = 58,
+	[0][0][0][0][RTW89_FCC][13] = 127,
+	[0][0][0][0][RTW89_ETSI][13] = 127,
+	[0][0][0][0][RTW89_MKK][13] = 76,
+	[0][0][0][0][RTW89_IC][13] = 127,
+	[0][0][0][0][RTW89_KCC][13] = 127,
+	[0][0][0][0][RTW89_ACMA][13] = 127,
+	[0][0][0][0][RTW89_CN][13] = 127,
+	[0][0][0][0][RTW89_UK][13] = 127,
+	[0][1][0][0][RTW89_FCC][0] = 127,
+	[0][1][0][0][RTW89_ETSI][0] = 127,
+	[0][1][0][0][RTW89_MKK][0] = 127,
+	[0][1][0][0][RTW89_IC][0] = 127,
+	[0][1][0][0][RTW89_KCC][0] = 127,
+	[0][1][0][0][RTW89_ACMA][0] = 127,
+	[0][1][0][0][RTW89_CN][0] = 127,
+	[0][1][0][0][RTW89_UK][0] = 127,
+	[0][1][0][0][RTW89_FCC][1] = 127,
+	[0][1][0][0][RTW89_ETSI][1] = 127,
+	[0][1][0][0][RTW89_MKK][1] = 127,
+	[0][1][0][0][RTW89_IC][1] = 127,
+	[0][1][0][0][RTW89_KCC][1] = 127,
+	[0][1][0][0][RTW89_ACMA][1] = 127,
+	[0][1][0][0][RTW89_CN][1] = 127,
+	[0][1][0][0][RTW89_UK][1] = 127,
+	[0][1][0][0][RTW89_FCC][2] = 127,
+	[0][1][0][0][RTW89_ETSI][2] = 127,
+	[0][1][0][0][RTW89_MKK][2] = 127,
+	[0][1][0][0][RTW89_IC][2] = 127,
+	[0][1][0][0][RTW89_KCC][2] = 127,
+	[0][1][0][0][RTW89_ACMA][2] = 127,
+	[0][1][0][0][RTW89_CN][2] = 127,
+	[0][1][0][0][RTW89_UK][2] = 127,
+	[0][1][0][0][RTW89_FCC][3] = 127,
+	[0][1][0][0][RTW89_ETSI][3] = 127,
+	[0][1][0][0][RTW89_MKK][3] = 127,
+	[0][1][0][0][RTW89_IC][3] = 127,
+	[0][1][0][0][RTW89_KCC][3] = 127,
+	[0][1][0][0][RTW89_ACMA][3] = 127,
+	[0][1][0][0][RTW89_CN][3] = 127,
+	[0][1][0][0][RTW89_UK][3] = 127,
+	[0][1][0][0][RTW89_FCC][4] = 127,
+	[0][1][0][0][RTW89_ETSI][4] = 127,
+	[0][1][0][0][RTW89_MKK][4] = 127,
+	[0][1][0][0][RTW89_IC][4] = 127,
+	[0][1][0][0][RTW89_KCC][4] = 127,
+	[0][1][0][0][RTW89_ACMA][4] = 127,
+	[0][1][0][0][RTW89_CN][4] = 127,
+	[0][1][0][0][RTW89_UK][4] = 127,
+	[0][1][0][0][RTW89_FCC][5] = 127,
+	[0][1][0][0][RTW89_ETSI][5] = 127,
+	[0][1][0][0][RTW89_MKK][5] = 127,
+	[0][1][0][0][RTW89_IC][5] = 127,
+	[0][1][0][0][RTW89_KCC][5] = 127,
+	[0][1][0][0][RTW89_ACMA][5] = 127,
+	[0][1][0][0][RTW89_CN][5] = 127,
+	[0][1][0][0][RTW89_UK][5] = 127,
+	[0][1][0][0][RTW89_FCC][6] = 127,
+	[0][1][0][0][RTW89_ETSI][6] = 127,
+	[0][1][0][0][RTW89_MKK][6] = 127,
+	[0][1][0][0][RTW89_IC][6] = 127,
+	[0][1][0][0][RTW89_KCC][6] = 127,
+	[0][1][0][0][RTW89_ACMA][6] = 127,
+	[0][1][0][0][RTW89_CN][6] = 127,
+	[0][1][0][0][RTW89_UK][6] = 127,
+	[0][1][0][0][RTW89_FCC][7] = 127,
+	[0][1][0][0][RTW89_ETSI][7] = 127,
+	[0][1][0][0][RTW89_MKK][7] = 127,
+	[0][1][0][0][RTW89_IC][7] = 127,
+	[0][1][0][0][RTW89_KCC][7] = 127,
+	[0][1][0][0][RTW89_ACMA][7] = 127,
+	[0][1][0][0][RTW89_CN][7] = 127,
+	[0][1][0][0][RTW89_UK][7] = 127,
+	[0][1][0][0][RTW89_FCC][8] = 127,
+	[0][1][0][0][RTW89_ETSI][8] = 127,
+	[0][1][0][0][RTW89_MKK][8] = 127,
+	[0][1][0][0][RTW89_IC][8] = 127,
+	[0][1][0][0][RTW89_KCC][8] = 127,
+	[0][1][0][0][RTW89_ACMA][8] = 127,
+	[0][1][0][0][RTW89_CN][8] = 127,
+	[0][1][0][0][RTW89_UK][8] = 127,
+	[0][1][0][0][RTW89_FCC][9] = 127,
+	[0][1][0][0][RTW89_ETSI][9] = 127,
+	[0][1][0][0][RTW89_MKK][9] = 127,
+	[0][1][0][0][RTW89_IC][9] = 127,
+	[0][1][0][0][RTW89_KCC][9] = 127,
+	[0][1][0][0][RTW89_ACMA][9] = 127,
+	[0][1][0][0][RTW89_CN][9] = 127,
+	[0][1][0][0][RTW89_UK][9] = 127,
+	[0][1][0][0][RTW89_FCC][10] = 127,
+	[0][1][0][0][RTW89_ETSI][10] = 127,
+	[0][1][0][0][RTW89_MKK][10] = 127,
+	[0][1][0][0][RTW89_IC][10] = 127,
+	[0][1][0][0][RTW89_KCC][10] = 127,
+	[0][1][0][0][RTW89_ACMA][10] = 127,
+	[0][1][0][0][RTW89_CN][10] = 127,
+	[0][1][0][0][RTW89_UK][10] = 127,
+	[0][1][0][0][RTW89_FCC][11] = 127,
+	[0][1][0][0][RTW89_ETSI][11] = 127,
+	[0][1][0][0][RTW89_MKK][11] = 127,
+	[0][1][0][0][RTW89_IC][11] = 127,
+	[0][1][0][0][RTW89_KCC][11] = 127,
+	[0][1][0][0][RTW89_ACMA][11] = 127,
+	[0][1][0][0][RTW89_CN][11] = 127,
+	[0][1][0][0][RTW89_UK][11] = 127,
+	[0][1][0][0][RTW89_FCC][12] = 127,
+	[0][1][0][0][RTW89_ETSI][12] = 127,
+	[0][1][0][0][RTW89_MKK][12] = 127,
+	[0][1][0][0][RTW89_IC][12] = 127,
+	[0][1][0][0][RTW89_KCC][12] = 127,
+	[0][1][0][0][RTW89_ACMA][12] = 127,
+	[0][1][0][0][RTW89_CN][12] = 127,
+	[0][1][0][0][RTW89_UK][12] = 127,
+	[0][1][0][0][RTW89_FCC][13] = 127,
+	[0][1][0][0][RTW89_ETSI][13] = 127,
+	[0][1][0][0][RTW89_MKK][13] = 127,
+	[0][1][0][0][RTW89_IC][13] = 127,
+	[0][1][0][0][RTW89_KCC][13] = 127,
+	[0][1][0][0][RTW89_ACMA][13] = 127,
+	[0][1][0][0][RTW89_CN][13] = 127,
+	[0][1][0][0][RTW89_UK][13] = 127,
+	[1][0][0][0][RTW89_FCC][0] = 127,
+	[1][0][0][0][RTW89_ETSI][0] = 127,
+	[1][0][0][0][RTW89_MKK][0] = 127,
+	[1][0][0][0][RTW89_IC][0] = 127,
+	[1][0][0][0][RTW89_KCC][0] = 127,
+	[1][0][0][0][RTW89_ACMA][0] = 127,
+	[1][0][0][0][RTW89_CN][0] = 127,
+	[1][0][0][0][RTW89_UK][0] = 127,
+	[1][0][0][0][RTW89_FCC][1] = 127,
+	[1][0][0][0][RTW89_ETSI][1] = 127,
+	[1][0][0][0][RTW89_MKK][1] = 127,
+	[1][0][0][0][RTW89_IC][1] = 127,
+	[1][0][0][0][RTW89_KCC][1] = 127,
+	[1][0][0][0][RTW89_ACMA][1] = 127,
+	[1][0][0][0][RTW89_CN][1] = 127,
+	[1][0][0][0][RTW89_UK][1] = 127,
+	[1][0][0][0][RTW89_FCC][2] = 127,
+	[1][0][0][0][RTW89_ETSI][2] = 58,
+	[1][0][0][0][RTW89_MKK][2] = 70,
+	[1][0][0][0][RTW89_IC][2] = 127,
+	[1][0][0][0][RTW89_KCC][2] = 68,
+	[1][0][0][0][RTW89_ACMA][2] = 58,
+	[1][0][0][0][RTW89_CN][2] = 60,
+	[1][0][0][0][RTW89_UK][2] = 58,
+	[1][0][0][0][RTW89_FCC][3] = 127,
+	[1][0][0][0][RTW89_ETSI][3] = 58,
+	[1][0][0][0][RTW89_MKK][3] = 76,
+	[1][0][0][0][RTW89_IC][3] = 127,
+	[1][0][0][0][RTW89_KCC][3] = 68,
+	[1][0][0][0][RTW89_ACMA][3] = 58,
+	[1][0][0][0][RTW89_CN][3] = 60,
+	[1][0][0][0][RTW89_UK][3] = 58,
+	[1][0][0][0][RTW89_FCC][4] = 127,
+	[1][0][0][0][RTW89_ETSI][4] = 58,
+	[1][0][0][0][RTW89_MKK][4] = 76,
+	[1][0][0][0][RTW89_IC][4] = 127,
+	[1][0][0][0][RTW89_KCC][4] = 68,
+	[1][0][0][0][RTW89_ACMA][4] = 58,
+	[1][0][0][0][RTW89_CN][4] = 60,
+	[1][0][0][0][RTW89_UK][4] = 58,
+	[1][0][0][0][RTW89_FCC][5] = 127,
+	[1][0][0][0][RTW89_ETSI][5] = 58,
+	[1][0][0][0][RTW89_MKK][5] = 76,
+	[1][0][0][0][RTW89_IC][5] = 127,
+	[1][0][0][0][RTW89_KCC][5] = 68,
+	[1][0][0][0][RTW89_ACMA][5] = 58,
+	[1][0][0][0][RTW89_CN][5] = 60,
+	[1][0][0][0][RTW89_UK][5] = 58,
+	[1][0][0][0][RTW89_FCC][6] = 127,
+	[1][0][0][0][RTW89_ETSI][6] = 58,
+	[1][0][0][0][RTW89_MKK][6] = 76,
+	[1][0][0][0][RTW89_IC][6] = 127,
+	[1][0][0][0][RTW89_KCC][6] = 68,
+	[1][0][0][0][RTW89_ACMA][6] = 58,
+	[1][0][0][0][RTW89_CN][6] = 60,
+	[1][0][0][0][RTW89_UK][6] = 58,
+	[1][0][0][0][RTW89_FCC][7] = 127,
+	[1][0][0][0][RTW89_ETSI][7] = 58,
+	[1][0][0][0][RTW89_MKK][7] = 76,
+	[1][0][0][0][RTW89_IC][7] = 127,
+	[1][0][0][0][RTW89_KCC][7] = 68,
+	[1][0][0][0][RTW89_ACMA][7] = 58,
+	[1][0][0][0][RTW89_CN][7] = 60,
+	[1][0][0][0][RTW89_UK][7] = 58,
+	[1][0][0][0][RTW89_FCC][8] = 127,
+	[1][0][0][0][RTW89_ETSI][8] = 58,
+	[1][0][0][0][RTW89_MKK][8] = 76,
+	[1][0][0][0][RTW89_IC][8] = 127,
+	[1][0][0][0][RTW89_KCC][8] = 68,
+	[1][0][0][0][RTW89_ACMA][8] = 58,
+	[1][0][0][0][RTW89_CN][8] = 60,
+	[1][0][0][0][RTW89_UK][8] = 58,
+	[1][0][0][0][RTW89_FCC][9] = 127,
+	[1][0][0][0][RTW89_ETSI][9] = 58,
+	[1][0][0][0][RTW89_MKK][9] = 76,
+	[1][0][0][0][RTW89_IC][9] = 127,
+	[1][0][0][0][RTW89_KCC][9] = 68,
+	[1][0][0][0][RTW89_ACMA][9] = 58,
+	[1][0][0][0][RTW89_CN][9] = 60,
+	[1][0][0][0][RTW89_UK][9] = 58,
+	[1][0][0][0][RTW89_FCC][10] = 127,
+	[1][0][0][0][RTW89_ETSI][10] = 58,
+	[1][0][0][0][RTW89_MKK][10] = 66,
+	[1][0][0][0][RTW89_IC][10] = 127,
+	[1][0][0][0][RTW89_KCC][10] = 68,
+	[1][0][0][0][RTW89_ACMA][10] = 58,
+	[1][0][0][0][RTW89_CN][10] = 60,
+	[1][0][0][0][RTW89_UK][10] = 58,
+	[1][0][0][0][RTW89_FCC][11] = 127,
+	[1][0][0][0][RTW89_ETSI][11] = 127,
+	[1][0][0][0][RTW89_MKK][11] = 127,
+	[1][0][0][0][RTW89_IC][11] = 127,
+	[1][0][0][0][RTW89_KCC][11] = 127,
+	[1][0][0][0][RTW89_ACMA][11] = 127,
+	[1][0][0][0][RTW89_CN][11] = 127,
+	[1][0][0][0][RTW89_UK][11] = 127,
+	[1][0][0][0][RTW89_FCC][12] = 127,
+	[1][0][0][0][RTW89_ETSI][12] = 127,
+	[1][0][0][0][RTW89_MKK][12] = 127,
+	[1][0][0][0][RTW89_IC][12] = 127,
+	[1][0][0][0][RTW89_KCC][12] = 127,
+	[1][0][0][0][RTW89_ACMA][12] = 127,
+	[1][0][0][0][RTW89_CN][12] = 127,
+	[1][0][0][0][RTW89_UK][12] = 127,
+	[1][0][0][0][RTW89_FCC][13] = 127,
+	[1][0][0][0][RTW89_ETSI][13] = 127,
+	[1][0][0][0][RTW89_MKK][13] = 127,
+	[1][0][0][0][RTW89_IC][13] = 127,
+	[1][0][0][0][RTW89_KCC][13] = 127,
+	[1][0][0][0][RTW89_ACMA][13] = 127,
+	[1][0][0][0][RTW89_CN][13] = 127,
+	[1][0][0][0][RTW89_UK][13] = 127,
+	[1][1][0][0][RTW89_FCC][0] = 127,
+	[1][1][0][0][RTW89_ETSI][0] = 127,
+	[1][1][0][0][RTW89_MKK][0] = 127,
+	[1][1][0][0][RTW89_IC][0] = 127,
+	[1][1][0][0][RTW89_KCC][0] = 127,
+	[1][1][0][0][RTW89_ACMA][0] = 127,
+	[1][1][0][0][RTW89_CN][0] = 127,
+	[1][1][0][0][RTW89_UK][0] = 127,
+	[1][1][0][0][RTW89_FCC][1] = 127,
+	[1][1][0][0][RTW89_ETSI][1] = 127,
+	[1][1][0][0][RTW89_MKK][1] = 127,
+	[1][1][0][0][RTW89_IC][1] = 127,
+	[1][1][0][0][RTW89_KCC][1] = 127,
+	[1][1][0][0][RTW89_ACMA][1] = 127,
+	[1][1][0][0][RTW89_CN][1] = 127,
+	[1][1][0][0][RTW89_UK][1] = 127,
+	[1][1][0][0][RTW89_FCC][2] = 127,
+	[1][1][0][0][RTW89_ETSI][2] = 127,
+	[1][1][0][0][RTW89_MKK][2] = 127,
+	[1][1][0][0][RTW89_IC][2] = 127,
+	[1][1][0][0][RTW89_KCC][2] = 127,
+	[1][1][0][0][RTW89_ACMA][2] = 127,
+	[1][1][0][0][RTW89_CN][2] = 127,
+	[1][1][0][0][RTW89_UK][2] = 127,
+	[1][1][0][0][RTW89_FCC][3] = 127,
+	[1][1][0][0][RTW89_ETSI][3] = 127,
+	[1][1][0][0][RTW89_MKK][3] = 127,
+	[1][1][0][0][RTW89_IC][3] = 127,
+	[1][1][0][0][RTW89_KCC][3] = 127,
+	[1][1][0][0][RTW89_ACMA][3] = 127,
+	[1][1][0][0][RTW89_CN][3] = 127,
+	[1][1][0][0][RTW89_UK][3] = 127,
+	[1][1][0][0][RTW89_FCC][4] = 127,
+	[1][1][0][0][RTW89_ETSI][4] = 127,
+	[1][1][0][0][RTW89_MKK][4] = 127,
+	[1][1][0][0][RTW89_IC][4] = 127,
+	[1][1][0][0][RTW89_KCC][4] = 127,
+	[1][1][0][0][RTW89_ACMA][4] = 127,
+	[1][1][0][0][RTW89_CN][4] = 127,
+	[1][1][0][0][RTW89_UK][4] = 127,
+	[1][1][0][0][RTW89_FCC][5] = 127,
+	[1][1][0][0][RTW89_ETSI][5] = 127,
+	[1][1][0][0][RTW89_MKK][5] = 127,
+	[1][1][0][0][RTW89_IC][5] = 127,
+	[1][1][0][0][RTW89_KCC][5] = 127,
+	[1][1][0][0][RTW89_ACMA][5] = 127,
+	[1][1][0][0][RTW89_CN][5] = 127,
+	[1][1][0][0][RTW89_UK][5] = 127,
+	[1][1][0][0][RTW89_FCC][6] = 127,
+	[1][1][0][0][RTW89_ETSI][6] = 127,
+	[1][1][0][0][RTW89_MKK][6] = 127,
+	[1][1][0][0][RTW89_IC][6] = 127,
+	[1][1][0][0][RTW89_KCC][6] = 127,
+	[1][1][0][0][RTW89_ACMA][6] = 127,
+	[1][1][0][0][RTW89_CN][6] = 127,
+	[1][1][0][0][RTW89_UK][6] = 127,
+	[1][1][0][0][RTW89_FCC][7] = 127,
+	[1][1][0][0][RTW89_ETSI][7] = 127,
+	[1][1][0][0][RTW89_MKK][7] = 127,
+	[1][1][0][0][RTW89_IC][7] = 127,
+	[1][1][0][0][RTW89_KCC][7] = 127,
+	[1][1][0][0][RTW89_ACMA][7] = 127,
+	[1][1][0][0][RTW89_CN][7] = 127,
+	[1][1][0][0][RTW89_UK][7] = 127,
+	[1][1][0][0][RTW89_FCC][8] = 127,
+	[1][1][0][0][RTW89_ETSI][8] = 127,
+	[1][1][0][0][RTW89_MKK][8] = 127,
+	[1][1][0][0][RTW89_IC][8] = 127,
+	[1][1][0][0][RTW89_KCC][8] = 127,
+	[1][1][0][0][RTW89_ACMA][8] = 127,
+	[1][1][0][0][RTW89_CN][8] = 127,
+	[1][1][0][0][RTW89_UK][8] = 127,
+	[1][1][0][0][RTW89_FCC][9] = 127,
+	[1][1][0][0][RTW89_ETSI][9] = 127,
+	[1][1][0][0][RTW89_MKK][9] = 127,
+	[1][1][0][0][RTW89_IC][9] = 127,
+	[1][1][0][0][RTW89_KCC][9] = 127,
+	[1][1][0][0][RTW89_ACMA][9] = 127,
+	[1][1][0][0][RTW89_CN][9] = 127,
+	[1][1][0][0][RTW89_UK][9] = 127,
+	[1][1][0][0][RTW89_FCC][10] = 127,
+	[1][1][0][0][RTW89_ETSI][10] = 127,
+	[1][1][0][0][RTW89_MKK][10] = 127,
+	[1][1][0][0][RTW89_IC][10] = 127,
+	[1][1][0][0][RTW89_KCC][10] = 127,
+	[1][1][0][0][RTW89_ACMA][10] = 127,
+	[1][1][0][0][RTW89_CN][10] = 127,
+	[1][1][0][0][RTW89_UK][10] = 127,
+	[1][1][0][0][RTW89_FCC][11] = 127,
+	[1][1][0][0][RTW89_ETSI][11] = 127,
+	[1][1][0][0][RTW89_MKK][11] = 127,
+	[1][1][0][0][RTW89_IC][11] = 127,
+	[1][1][0][0][RTW89_KCC][11] = 127,
+	[1][1][0][0][RTW89_ACMA][11] = 127,
+	[1][1][0][0][RTW89_CN][11] = 127,
+	[1][1][0][0][RTW89_UK][11] = 127,
+	[1][1][0][0][RTW89_FCC][12] = 127,
+	[1][1][0][0][RTW89_ETSI][12] = 127,
+	[1][1][0][0][RTW89_MKK][12] = 127,
+	[1][1][0][0][RTW89_IC][12] = 127,
+	[1][1][0][0][RTW89_KCC][12] = 127,
+	[1][1][0][0][RTW89_ACMA][12] = 127,
+	[1][1][0][0][RTW89_CN][12] = 127,
+	[1][1][0][0][RTW89_UK][12] = 127,
+	[1][1][0][0][RTW89_FCC][13] = 127,
+	[1][1][0][0][RTW89_ETSI][13] = 127,
+	[1][1][0][0][RTW89_MKK][13] = 127,
+	[1][1][0][0][RTW89_IC][13] = 127,
+	[1][1][0][0][RTW89_KCC][13] = 127,
+	[1][1][0][0][RTW89_ACMA][13] = 127,
+	[1][1][0][0][RTW89_CN][13] = 127,
+	[1][1][0][0][RTW89_UK][13] = 127,
+	[0][0][1][0][RTW89_FCC][0] = 80,
+	[0][0][1][0][RTW89_ETSI][0] = 58,
+	[0][0][1][0][RTW89_MKK][0] = 72,
+	[0][0][1][0][RTW89_IC][0] = 80,
+	[0][0][1][0][RTW89_KCC][0] = 78,
+	[0][0][1][0][RTW89_ACMA][0] = 58,
+	[0][0][1][0][RTW89_CN][0] = 60,
+	[0][0][1][0][RTW89_UK][0] = 58,
+	[0][0][1][0][RTW89_FCC][1] = 80,
+	[0][0][1][0][RTW89_ETSI][1] = 60,
+	[0][0][1][0][RTW89_MKK][1] = 74,
+	[0][0][1][0][RTW89_IC][1] = 80,
+	[0][0][1][0][RTW89_KCC][1] = 78,
+	[0][0][1][0][RTW89_ACMA][1] = 60,
+	[0][0][1][0][RTW89_CN][1] = 60,
+	[0][0][1][0][RTW89_UK][1] = 60,
+	[0][0][1][0][RTW89_FCC][2] = 84,
+	[0][0][1][0][RTW89_ETSI][2] = 60,
+	[0][0][1][0][RTW89_MKK][2] = 74,
+	[0][0][1][0][RTW89_IC][2] = 84,
+	[0][0][1][0][RTW89_KCC][2] = 78,
+	[0][0][1][0][RTW89_ACMA][2] = 60,
+	[0][0][1][0][RTW89_CN][2] = 60,
+	[0][0][1][0][RTW89_UK][2] = 60,
+	[0][0][1][0][RTW89_FCC][3] = 84,
+	[0][0][1][0][RTW89_ETSI][3] = 60,
+	[0][0][1][0][RTW89_MKK][3] = 74,
+	[0][0][1][0][RTW89_IC][3] = 84,
+	[0][0][1][0][RTW89_KCC][3] = 78,
+	[0][0][1][0][RTW89_ACMA][3] = 60,
+	[0][0][1][0][RTW89_CN][3] = 60,
+	[0][0][1][0][RTW89_UK][3] = 60,
+	[0][0][1][0][RTW89_FCC][4] = 84,
+	[0][0][1][0][RTW89_ETSI][4] = 60,
+	[0][0][1][0][RTW89_MKK][4] = 74,
+	[0][0][1][0][RTW89_IC][4] = 84,
+	[0][0][1][0][RTW89_KCC][4] = 76,
+	[0][0][1][0][RTW89_ACMA][4] = 60,
+	[0][0][1][0][RTW89_CN][4] = 60,
+	[0][0][1][0][RTW89_UK][4] = 60,
+	[0][0][1][0][RTW89_FCC][5] = 84,
+	[0][0][1][0][RTW89_ETSI][5] = 60,
+	[0][0][1][0][RTW89_MKK][5] = 74,
+	[0][0][1][0][RTW89_IC][5] = 84,
+	[0][0][1][0][RTW89_KCC][5] = 76,
+	[0][0][1][0][RTW89_ACMA][5] = 60,
+	[0][0][1][0][RTW89_CN][5] = 60,
+	[0][0][1][0][RTW89_UK][5] = 60,
+	[0][0][1][0][RTW89_FCC][6] = 84,
+	[0][0][1][0][RTW89_ETSI][6] = 60,
+	[0][0][1][0][RTW89_MKK][6] = 74,
+	[0][0][1][0][RTW89_IC][6] = 84,
+	[0][0][1][0][RTW89_KCC][6] = 76,
+	[0][0][1][0][RTW89_ACMA][6] = 60,
+	[0][0][1][0][RTW89_CN][6] = 60,
+	[0][0][1][0][RTW89_UK][6] = 60,
+	[0][0][1][0][RTW89_FCC][7] = 84,
+	[0][0][1][0][RTW89_ETSI][7] = 60,
+	[0][0][1][0][RTW89_MKK][7] = 74,
+	[0][0][1][0][RTW89_IC][7] = 84,
+	[0][0][1][0][RTW89_KCC][7] = 76,
+	[0][0][1][0][RTW89_ACMA][7] = 60,
+	[0][0][1][0][RTW89_CN][7] = 60,
+	[0][0][1][0][RTW89_UK][7] = 60,
+	[0][0][1][0][RTW89_FCC][8] = 80,
+	[0][0][1][0][RTW89_ETSI][8] = 60,
+	[0][0][1][0][RTW89_MKK][8] = 74,
+	[0][0][1][0][RTW89_IC][8] = 80,
+	[0][0][1][0][RTW89_KCC][8] = 76,
+	[0][0][1][0][RTW89_ACMA][8] = 60,
+	[0][0][1][0][RTW89_CN][8] = 60,
+	[0][0][1][0][RTW89_UK][8] = 60,
+	[0][0][1][0][RTW89_FCC][9] = 76,
+	[0][0][1][0][RTW89_ETSI][9] = 60,
+	[0][0][1][0][RTW89_MKK][9] = 74,
+	[0][0][1][0][RTW89_IC][9] = 76,
+	[0][0][1][0][RTW89_KCC][9] = 74,
+	[0][0][1][0][RTW89_ACMA][9] = 60,
+	[0][0][1][0][RTW89_CN][9] = 60,
+	[0][0][1][0][RTW89_UK][9] = 60,
+	[0][0][1][0][RTW89_FCC][10] = 76,
+	[0][0][1][0][RTW89_ETSI][10] = 60,
+	[0][0][1][0][RTW89_MKK][10] = 74,
+	[0][0][1][0][RTW89_IC][10] = 76,
+	[0][0][1][0][RTW89_KCC][10] = 74,
+	[0][0][1][0][RTW89_ACMA][10] = 60,
+	[0][0][1][0][RTW89_CN][10] = 60,
+	[0][0][1][0][RTW89_UK][10] = 60,
+	[0][0][1][0][RTW89_FCC][11] = 68,
+	[0][0][1][0][RTW89_ETSI][11] = 60,
+	[0][0][1][0][RTW89_MKK][11] = 74,
+	[0][0][1][0][RTW89_IC][11] = 68,
+	[0][0][1][0][RTW89_KCC][11] = 74,
+	[0][0][1][0][RTW89_ACMA][11] = 60,
+	[0][0][1][0][RTW89_CN][11] = 60,
+	[0][0][1][0][RTW89_UK][11] = 60,
+	[0][0][1][0][RTW89_FCC][12] = 64,
+	[0][0][1][0][RTW89_ETSI][12] = 58,
+	[0][0][1][0][RTW89_MKK][12] = 70,
+	[0][0][1][0][RTW89_IC][12] = 64,
+	[0][0][1][0][RTW89_KCC][12] = 74,
+	[0][0][1][0][RTW89_ACMA][12] = 58,
+	[0][0][1][0][RTW89_CN][12] = 60,
+	[0][0][1][0][RTW89_UK][12] = 58,
+	[0][0][1][0][RTW89_FCC][13] = 127,
+	[0][0][1][0][RTW89_ETSI][13] = 127,
+	[0][0][1][0][RTW89_MKK][13] = 127,
+	[0][0][1][0][RTW89_IC][13] = 127,
+	[0][0][1][0][RTW89_KCC][13] = 127,
+	[0][0][1][0][RTW89_ACMA][13] = 127,
+	[0][0][1][0][RTW89_CN][13] = 127,
+	[0][0][1][0][RTW89_UK][13] = 127,
+	[0][1][1][0][RTW89_FCC][0] = 127,
+	[0][1][1][0][RTW89_ETSI][0] = 127,
+	[0][1][1][0][RTW89_MKK][0] = 127,
+	[0][1][1][0][RTW89_IC][0] = 127,
+	[0][1][1][0][RTW89_KCC][0] = 127,
+	[0][1][1][0][RTW89_ACMA][0] = 127,
+	[0][1][1][0][RTW89_CN][0] = 127,
+	[0][1][1][0][RTW89_UK][0] = 127,
+	[0][1][1][0][RTW89_FCC][1] = 127,
+	[0][1][1][0][RTW89_ETSI][1] = 127,
+	[0][1][1][0][RTW89_MKK][1] = 127,
+	[0][1][1][0][RTW89_IC][1] = 127,
+	[0][1][1][0][RTW89_KCC][1] = 127,
+	[0][1][1][0][RTW89_ACMA][1] = 127,
+	[0][1][1][0][RTW89_CN][1] = 127,
+	[0][1][1][0][RTW89_UK][1] = 127,
+	[0][1][1][0][RTW89_FCC][2] = 127,
+	[0][1][1][0][RTW89_ETSI][2] = 127,
+	[0][1][1][0][RTW89_MKK][2] = 127,
+	[0][1][1][0][RTW89_IC][2] = 127,
+	[0][1][1][0][RTW89_KCC][2] = 127,
+	[0][1][1][0][RTW89_ACMA][2] = 127,
+	[0][1][1][0][RTW89_CN][2] = 127,
+	[0][1][1][0][RTW89_UK][2] = 127,
+	[0][1][1][0][RTW89_FCC][3] = 127,
+	[0][1][1][0][RTW89_ETSI][3] = 127,
+	[0][1][1][0][RTW89_MKK][3] = 127,
+	[0][1][1][0][RTW89_IC][3] = 127,
+	[0][1][1][0][RTW89_KCC][3] = 127,
+	[0][1][1][0][RTW89_ACMA][3] = 127,
+	[0][1][1][0][RTW89_CN][3] = 127,
+	[0][1][1][0][RTW89_UK][3] = 127,
+	[0][1][1][0][RTW89_FCC][4] = 127,
+	[0][1][1][0][RTW89_ETSI][4] = 127,
+	[0][1][1][0][RTW89_MKK][4] = 127,
+	[0][1][1][0][RTW89_IC][4] = 127,
+	[0][1][1][0][RTW89_KCC][4] = 127,
+	[0][1][1][0][RTW89_ACMA][4] = 127,
+	[0][1][1][0][RTW89_CN][4] = 127,
+	[0][1][1][0][RTW89_UK][4] = 127,
+	[0][1][1][0][RTW89_FCC][5] = 127,
+	[0][1][1][0][RTW89_ETSI][5] = 127,
+	[0][1][1][0][RTW89_MKK][5] = 127,
+	[0][1][1][0][RTW89_IC][5] = 127,
+	[0][1][1][0][RTW89_KCC][5] = 127,
+	[0][1][1][0][RTW89_ACMA][5] = 127,
+	[0][1][1][0][RTW89_CN][5] = 127,
+	[0][1][1][0][RTW89_UK][5] = 127,
+	[0][1][1][0][RTW89_FCC][6] = 127,
+	[0][1][1][0][RTW89_ETSI][6] = 127,
+	[0][1][1][0][RTW89_MKK][6] = 127,
+	[0][1][1][0][RTW89_IC][6] = 127,
+	[0][1][1][0][RTW89_KCC][6] = 127,
+	[0][1][1][0][RTW89_ACMA][6] = 127,
+	[0][1][1][0][RTW89_CN][6] = 127,
+	[0][1][1][0][RTW89_UK][6] = 127,
+	[0][1][1][0][RTW89_FCC][7] = 127,
+	[0][1][1][0][RTW89_ETSI][7] = 127,
+	[0][1][1][0][RTW89_MKK][7] = 127,
+	[0][1][1][0][RTW89_IC][7] = 127,
+	[0][1][1][0][RTW89_KCC][7] = 127,
+	[0][1][1][0][RTW89_ACMA][7] = 127,
+	[0][1][1][0][RTW89_CN][7] = 127,
+	[0][1][1][0][RTW89_UK][7] = 127,
+	[0][1][1][0][RTW89_FCC][8] = 127,
+	[0][1][1][0][RTW89_ETSI][8] = 127,
+	[0][1][1][0][RTW89_MKK][8] = 127,
+	[0][1][1][0][RTW89_IC][8] = 127,
+	[0][1][1][0][RTW89_KCC][8] = 127,
+	[0][1][1][0][RTW89_ACMA][8] = 127,
+	[0][1][1][0][RTW89_CN][8] = 127,
+	[0][1][1][0][RTW89_UK][8] = 127,
+	[0][1][1][0][RTW89_FCC][9] = 127,
+	[0][1][1][0][RTW89_ETSI][9] = 127,
+	[0][1][1][0][RTW89_MKK][9] = 127,
+	[0][1][1][0][RTW89_IC][9] = 127,
+	[0][1][1][0][RTW89_KCC][9] = 127,
+	[0][1][1][0][RTW89_ACMA][9] = 127,
+	[0][1][1][0][RTW89_CN][9] = 127,
+	[0][1][1][0][RTW89_UK][9] = 127,
+	[0][1][1][0][RTW89_FCC][10] = 127,
+	[0][1][1][0][RTW89_ETSI][10] = 127,
+	[0][1][1][0][RTW89_MKK][10] = 127,
+	[0][1][1][0][RTW89_IC][10] = 127,
+	[0][1][1][0][RTW89_KCC][10] = 127,
+	[0][1][1][0][RTW89_ACMA][10] = 127,
+	[0][1][1][0][RTW89_CN][10] = 127,
+	[0][1][1][0][RTW89_UK][10] = 127,
+	[0][1][1][0][RTW89_FCC][11] = 127,
+	[0][1][1][0][RTW89_ETSI][11] = 127,
+	[0][1][1][0][RTW89_MKK][11] = 127,
+	[0][1][1][0][RTW89_IC][11] = 127,
+	[0][1][1][0][RTW89_KCC][11] = 127,
+	[0][1][1][0][RTW89_ACMA][11] = 127,
+	[0][1][1][0][RTW89_CN][11] = 127,
+	[0][1][1][0][RTW89_UK][11] = 127,
+	[0][1][1][0][RTW89_FCC][12] = 127,
+	[0][1][1][0][RTW89_ETSI][12] = 127,
+	[0][1][1][0][RTW89_MKK][12] = 127,
+	[0][1][1][0][RTW89_IC][12] = 127,
+	[0][1][1][0][RTW89_KCC][12] = 127,
+	[0][1][1][0][RTW89_ACMA][12] = 127,
+	[0][1][1][0][RTW89_CN][12] = 127,
+	[0][1][1][0][RTW89_UK][12] = 127,
+	[0][1][1][0][RTW89_FCC][13] = 127,
+	[0][1][1][0][RTW89_ETSI][13] = 127,
+	[0][1][1][0][RTW89_MKK][13] = 127,
+	[0][1][1][0][RTW89_IC][13] = 127,
+	[0][1][1][0][RTW89_KCC][13] = 127,
+	[0][1][1][0][RTW89_ACMA][13] = 127,
+	[0][1][1][0][RTW89_CN][13] = 127,
+	[0][1][1][0][RTW89_UK][13] = 127,
+	[0][0][2][0][RTW89_FCC][0] = 78,
+	[0][0][2][0][RTW89_ETSI][0] = 60,
+	[0][0][2][0][RTW89_MKK][0] = 72,
+	[0][0][2][0][RTW89_IC][0] = 78,
+	[0][0][2][0][RTW89_KCC][0] = 78,
+	[0][0][2][0][RTW89_ACMA][0] = 60,
+	[0][0][2][0][RTW89_CN][0] = 60,
+	[0][0][2][0][RTW89_UK][0] = 60,
+	[0][0][2][0][RTW89_FCC][1] = 78,
+	[0][0][2][0][RTW89_ETSI][1] = 60,
+	[0][0][2][0][RTW89_MKK][1] = 78,
+	[0][0][2][0][RTW89_IC][1] = 78,
+	[0][0][2][0][RTW89_KCC][1] = 78,
+	[0][0][2][0][RTW89_ACMA][1] = 60,
+	[0][0][2][0][RTW89_CN][1] = 60,
+	[0][0][2][0][RTW89_UK][1] = 60,
+	[0][0][2][0][RTW89_FCC][2] = 82,
+	[0][0][2][0][RTW89_ETSI][2] = 60,
+	[0][0][2][0][RTW89_MKK][2] = 78,
+	[0][0][2][0][RTW89_IC][2] = 82,
+	[0][0][2][0][RTW89_KCC][2] = 78,
+	[0][0][2][0][RTW89_ACMA][2] = 60,
+	[0][0][2][0][RTW89_CN][2] = 60,
+	[0][0][2][0][RTW89_UK][2] = 60,
+	[0][0][2][0][RTW89_FCC][3] = 82,
+	[0][0][2][0][RTW89_ETSI][3] = 60,
+	[0][0][2][0][RTW89_MKK][3] = 78,
+	[0][0][2][0][RTW89_IC][3] = 82,
+	[0][0][2][0][RTW89_KCC][3] = 78,
+	[0][0][2][0][RTW89_ACMA][3] = 60,
+	[0][0][2][0][RTW89_CN][3] = 60,
+	[0][0][2][0][RTW89_UK][3] = 60,
+	[0][0][2][0][RTW89_FCC][4] = 82,
+	[0][0][2][0][RTW89_ETSI][4] = 60,
+	[0][0][2][0][RTW89_MKK][4] = 78,
+	[0][0][2][0][RTW89_IC][4] = 82,
+	[0][0][2][0][RTW89_KCC][4] = 78,
+	[0][0][2][0][RTW89_ACMA][4] = 60,
+	[0][0][2][0][RTW89_CN][4] = 60,
+	[0][0][2][0][RTW89_UK][4] = 60,
+	[0][0][2][0][RTW89_FCC][5] = 82,
+	[0][0][2][0][RTW89_ETSI][5] = 60,
+	[0][0][2][0][RTW89_MKK][5] = 78,
+	[0][0][2][0][RTW89_IC][5] = 82,
+	[0][0][2][0][RTW89_KCC][5] = 78,
+	[0][0][2][0][RTW89_ACMA][5] = 60,
+	[0][0][2][0][RTW89_CN][5] = 60,
+	[0][0][2][0][RTW89_UK][5] = 60,
+	[0][0][2][0][RTW89_FCC][6] = 82,
+	[0][0][2][0][RTW89_ETSI][6] = 60,
+	[0][0][2][0][RTW89_MKK][6] = 78,
+	[0][0][2][0][RTW89_IC][6] = 82,
+	[0][0][2][0][RTW89_KCC][6] = 78,
+	[0][0][2][0][RTW89_ACMA][6] = 60,
+	[0][0][2][0][RTW89_CN][6] = 60,
+	[0][0][2][0][RTW89_UK][6] = 60,
+	[0][0][2][0][RTW89_FCC][7] = 82,
+	[0][0][2][0][RTW89_ETSI][7] = 60,
+	[0][0][2][0][RTW89_MKK][7] = 78,
+	[0][0][2][0][RTW89_IC][7] = 82,
+	[0][0][2][0][RTW89_KCC][7] = 78,
+	[0][0][2][0][RTW89_ACMA][7] = 60,
+	[0][0][2][0][RTW89_CN][7] = 60,
+	[0][0][2][0][RTW89_UK][7] = 60,
+	[0][0][2][0][RTW89_FCC][8] = 80,
+	[0][0][2][0][RTW89_ETSI][8] = 60,
+	[0][0][2][0][RTW89_MKK][8] = 78,
+	[0][0][2][0][RTW89_IC][8] = 80,
+	[0][0][2][0][RTW89_KCC][8] = 78,
+	[0][0][2][0][RTW89_ACMA][8] = 60,
+	[0][0][2][0][RTW89_CN][8] = 60,
+	[0][0][2][0][RTW89_UK][8] = 60,
+	[0][0][2][0][RTW89_FCC][9] = 76,
+	[0][0][2][0][RTW89_ETSI][9] = 60,
+	[0][0][2][0][RTW89_MKK][9] = 78,
+	[0][0][2][0][RTW89_IC][9] = 76,
+	[0][0][2][0][RTW89_KCC][9] = 78,
+	[0][0][2][0][RTW89_ACMA][9] = 60,
+	[0][0][2][0][RTW89_CN][9] = 60,
+	[0][0][2][0][RTW89_UK][9] = 60,
+	[0][0][2][0][RTW89_FCC][10] = 76,
+	[0][0][2][0][RTW89_ETSI][10] = 60,
+	[0][0][2][0][RTW89_MKK][10] = 78,
+	[0][0][2][0][RTW89_IC][10] = 76,
+	[0][0][2][0][RTW89_KCC][10] = 78,
+	[0][0][2][0][RTW89_ACMA][10] = 60,
+	[0][0][2][0][RTW89_CN][10] = 60,
+	[0][0][2][0][RTW89_UK][10] = 60,
+	[0][0][2][0][RTW89_FCC][11] = 70,
+	[0][0][2][0][RTW89_ETSI][11] = 60,
+	[0][0][2][0][RTW89_MKK][11] = 78,
+	[0][0][2][0][RTW89_IC][11] = 70,
+	[0][0][2][0][RTW89_KCC][11] = 78,
+	[0][0][2][0][RTW89_ACMA][11] = 60,
+	[0][0][2][0][RTW89_CN][11] = 60,
+	[0][0][2][0][RTW89_UK][11] = 60,
+	[0][0][2][0][RTW89_FCC][12] = 70,
+	[0][0][2][0][RTW89_ETSI][12] = 60,
+	[0][0][2][0][RTW89_MKK][12] = 70,
+	[0][0][2][0][RTW89_IC][12] = 70,
+	[0][0][2][0][RTW89_KCC][12] = 78,
+	[0][0][2][0][RTW89_ACMA][12] = 60,
+	[0][0][2][0][RTW89_CN][12] = 60,
+	[0][0][2][0][RTW89_UK][12] = 60,
+	[0][0][2][0][RTW89_FCC][13] = 127,
+	[0][0][2][0][RTW89_ETSI][13] = 127,
+	[0][0][2][0][RTW89_MKK][13] = 127,
+	[0][0][2][0][RTW89_IC][13] = 127,
+	[0][0][2][0][RTW89_KCC][13] = 127,
+	[0][0][2][0][RTW89_ACMA][13] = 127,
+	[0][0][2][0][RTW89_CN][13] = 127,
+	[0][0][2][0][RTW89_UK][13] = 127,
+	[0][1][2][0][RTW89_FCC][0] = 127,
+	[0][1][2][0][RTW89_ETSI][0] = 127,
+	[0][1][2][0][RTW89_MKK][0] = 127,
+	[0][1][2][0][RTW89_IC][0] = 127,
+	[0][1][2][0][RTW89_KCC][0] = 127,
+	[0][1][2][0][RTW89_ACMA][0] = 127,
+	[0][1][2][0][RTW89_CN][0] = 127,
+	[0][1][2][0][RTW89_UK][0] = 127,
+	[0][1][2][0][RTW89_FCC][1] = 127,
+	[0][1][2][0][RTW89_ETSI][1] = 127,
+	[0][1][2][0][RTW89_MKK][1] = 127,
+	[0][1][2][0][RTW89_IC][1] = 127,
+	[0][1][2][0][RTW89_KCC][1] = 127,
+	[0][1][2][0][RTW89_ACMA][1] = 127,
+	[0][1][2][0][RTW89_CN][1] = 127,
+	[0][1][2][0][RTW89_UK][1] = 127,
+	[0][1][2][0][RTW89_FCC][2] = 127,
+	[0][1][2][0][RTW89_ETSI][2] = 127,
+	[0][1][2][0][RTW89_MKK][2] = 127,
+	[0][1][2][0][RTW89_IC][2] = 127,
+	[0][1][2][0][RTW89_KCC][2] = 127,
+	[0][1][2][0][RTW89_ACMA][2] = 127,
+	[0][1][2][0][RTW89_CN][2] = 127,
+	[0][1][2][0][RTW89_UK][2] = 127,
+	[0][1][2][0][RTW89_FCC][3] = 127,
+	[0][1][2][0][RTW89_ETSI][3] = 127,
+	[0][1][2][0][RTW89_MKK][3] = 127,
+	[0][1][2][0][RTW89_IC][3] = 127,
+	[0][1][2][0][RTW89_KCC][3] = 127,
+	[0][1][2][0][RTW89_ACMA][3] = 127,
+	[0][1][2][0][RTW89_CN][3] = 127,
+	[0][1][2][0][RTW89_UK][3] = 127,
+	[0][1][2][0][RTW89_FCC][4] = 127,
+	[0][1][2][0][RTW89_ETSI][4] = 127,
+	[0][1][2][0][RTW89_MKK][4] = 127,
+	[0][1][2][0][RTW89_IC][4] = 127,
+	[0][1][2][0][RTW89_KCC][4] = 127,
+	[0][1][2][0][RTW89_ACMA][4] = 127,
+	[0][1][2][0][RTW89_CN][4] = 127,
+	[0][1][2][0][RTW89_UK][4] = 127,
+	[0][1][2][0][RTW89_FCC][5] = 127,
+	[0][1][2][0][RTW89_ETSI][5] = 127,
+	[0][1][2][0][RTW89_MKK][5] = 127,
+	[0][1][2][0][RTW89_IC][5] = 127,
+	[0][1][2][0][RTW89_KCC][5] = 127,
+	[0][1][2][0][RTW89_ACMA][5] = 127,
+	[0][1][2][0][RTW89_CN][5] = 127,
+	[0][1][2][0][RTW89_UK][5] = 127,
+	[0][1][2][0][RTW89_FCC][6] = 127,
+	[0][1][2][0][RTW89_ETSI][6] = 127,
+	[0][1][2][0][RTW89_MKK][6] = 127,
+	[0][1][2][0][RTW89_IC][6] = 127,
+	[0][1][2][0][RTW89_KCC][6] = 127,
+	[0][1][2][0][RTW89_ACMA][6] = 127,
+	[0][1][2][0][RTW89_CN][6] = 127,
+	[0][1][2][0][RTW89_UK][6] = 127,
+	[0][1][2][0][RTW89_FCC][7] = 127,
+	[0][1][2][0][RTW89_ETSI][7] = 127,
+	[0][1][2][0][RTW89_MKK][7] = 127,
+	[0][1][2][0][RTW89_IC][7] = 127,
+	[0][1][2][0][RTW89_KCC][7] = 127,
+	[0][1][2][0][RTW89_ACMA][7] = 127,
+	[0][1][2][0][RTW89_CN][7] = 127,
+	[0][1][2][0][RTW89_UK][7] = 127,
+	[0][1][2][0][RTW89_FCC][8] = 127,
+	[0][1][2][0][RTW89_ETSI][8] = 127,
+	[0][1][2][0][RTW89_MKK][8] = 127,
+	[0][1][2][0][RTW89_IC][8] = 127,
+	[0][1][2][0][RTW89_KCC][8] = 127,
+	[0][1][2][0][RTW89_ACMA][8] = 127,
+	[0][1][2][0][RTW89_CN][8] = 127,
+	[0][1][2][0][RTW89_UK][8] = 127,
+	[0][1][2][0][RTW89_FCC][9] = 127,
+	[0][1][2][0][RTW89_ETSI][9] = 127,
+	[0][1][2][0][RTW89_MKK][9] = 127,
+	[0][1][2][0][RTW89_IC][9] = 127,
+	[0][1][2][0][RTW89_KCC][9] = 127,
+	[0][1][2][0][RTW89_ACMA][9] = 127,
+	[0][1][2][0][RTW89_CN][9] = 127,
+	[0][1][2][0][RTW89_UK][9] = 127,
+	[0][1][2][0][RTW89_FCC][10] = 127,
+	[0][1][2][0][RTW89_ETSI][10] = 127,
+	[0][1][2][0][RTW89_MKK][10] = 127,
+	[0][1][2][0][RTW89_IC][10] = 127,
+	[0][1][2][0][RTW89_KCC][10] = 127,
+	[0][1][2][0][RTW89_ACMA][10] = 127,
+	[0][1][2][0][RTW89_CN][10] = 127,
+	[0][1][2][0][RTW89_UK][10] = 127,
+	[0][1][2][0][RTW89_FCC][11] = 127,
+	[0][1][2][0][RTW89_ETSI][11] = 127,
+	[0][1][2][0][RTW89_MKK][11] = 127,
+	[0][1][2][0][RTW89_IC][11] = 127,
+	[0][1][2][0][RTW89_KCC][11] = 127,
+	[0][1][2][0][RTW89_ACMA][11] = 127,
+	[0][1][2][0][RTW89_CN][11] = 127,
+	[0][1][2][0][RTW89_UK][11] = 127,
+	[0][1][2][0][RTW89_FCC][12] = 127,
+	[0][1][2][0][RTW89_ETSI][12] = 127,
+	[0][1][2][0][RTW89_MKK][12] = 127,
+	[0][1][2][0][RTW89_IC][12] = 127,
+	[0][1][2][0][RTW89_KCC][12] = 127,
+	[0][1][2][0][RTW89_ACMA][12] = 127,
+	[0][1][2][0][RTW89_CN][12] = 127,
+	[0][1][2][0][RTW89_UK][12] = 127,
+	[0][1][2][0][RTW89_FCC][13] = 127,
+	[0][1][2][0][RTW89_ETSI][13] = 127,
+	[0][1][2][0][RTW89_MKK][13] = 127,
+	[0][1][2][0][RTW89_IC][13] = 127,
+	[0][1][2][0][RTW89_KCC][13] = 127,
+	[0][1][2][0][RTW89_ACMA][13] = 127,
+	[0][1][2][0][RTW89_CN][13] = 127,
+	[0][1][2][0][RTW89_UK][13] = 127,
+	[0][1][2][1][RTW89_FCC][0] = 127,
+	[0][1][2][1][RTW89_ETSI][0] = 127,
+	[0][1][2][1][RTW89_MKK][0] = 127,
+	[0][1][2][1][RTW89_IC][0] = 127,
+	[0][1][2][1][RTW89_KCC][0] = 127,
+	[0][1][2][1][RTW89_ACMA][0] = 127,
+	[0][1][2][1][RTW89_CN][0] = 127,
+	[0][1][2][1][RTW89_UK][0] = 127,
+	[0][1][2][1][RTW89_FCC][1] = 127,
+	[0][1][2][1][RTW89_ETSI][1] = 127,
+	[0][1][2][1][RTW89_MKK][1] = 127,
+	[0][1][2][1][RTW89_IC][1] = 127,
+	[0][1][2][1][RTW89_KCC][1] = 127,
+	[0][1][2][1][RTW89_ACMA][1] = 127,
+	[0][1][2][1][RTW89_CN][1] = 127,
+	[0][1][2][1][RTW89_UK][1] = 127,
+	[0][1][2][1][RTW89_FCC][2] = 127,
+	[0][1][2][1][RTW89_ETSI][2] = 127,
+	[0][1][2][1][RTW89_MKK][2] = 127,
+	[0][1][2][1][RTW89_IC][2] = 127,
+	[0][1][2][1][RTW89_KCC][2] = 127,
+	[0][1][2][1][RTW89_ACMA][2] = 127,
+	[0][1][2][1][RTW89_CN][2] = 127,
+	[0][1][2][1][RTW89_UK][2] = 127,
+	[0][1][2][1][RTW89_FCC][3] = 127,
+	[0][1][2][1][RTW89_ETSI][3] = 127,
+	[0][1][2][1][RTW89_MKK][3] = 127,
+	[0][1][2][1][RTW89_IC][3] = 127,
+	[0][1][2][1][RTW89_KCC][3] = 127,
+	[0][1][2][1][RTW89_ACMA][3] = 127,
+	[0][1][2][1][RTW89_CN][3] = 127,
+	[0][1][2][1][RTW89_UK][3] = 127,
+	[0][1][2][1][RTW89_FCC][4] = 127,
+	[0][1][2][1][RTW89_ETSI][4] = 127,
+	[0][1][2][1][RTW89_MKK][4] = 127,
+	[0][1][2][1][RTW89_IC][4] = 127,
+	[0][1][2][1][RTW89_KCC][4] = 127,
+	[0][1][2][1][RTW89_ACMA][4] = 127,
+	[0][1][2][1][RTW89_CN][4] = 127,
+	[0][1][2][1][RTW89_UK][4] = 127,
+	[0][1][2][1][RTW89_FCC][5] = 127,
+	[0][1][2][1][RTW89_ETSI][5] = 127,
+	[0][1][2][1][RTW89_MKK][5] = 127,
+	[0][1][2][1][RTW89_IC][5] = 127,
+	[0][1][2][1][RTW89_KCC][5] = 127,
+	[0][1][2][1][RTW89_ACMA][5] = 127,
+	[0][1][2][1][RTW89_CN][5] = 127,
+	[0][1][2][1][RTW89_UK][5] = 127,
+	[0][1][2][1][RTW89_FCC][6] = 127,
+	[0][1][2][1][RTW89_ETSI][6] = 127,
+	[0][1][2][1][RTW89_MKK][6] = 127,
+	[0][1][2][1][RTW89_IC][6] = 127,
+	[0][1][2][1][RTW89_KCC][6] = 127,
+	[0][1][2][1][RTW89_ACMA][6] = 127,
+	[0][1][2][1][RTW89_CN][6] = 127,
+	[0][1][2][1][RTW89_UK][6] = 127,
+	[0][1][2][1][RTW89_FCC][7] = 127,
+	[0][1][2][1][RTW89_ETSI][7] = 127,
+	[0][1][2][1][RTW89_MKK][7] = 127,
+	[0][1][2][1][RTW89_IC][7] = 127,
+	[0][1][2][1][RTW89_KCC][7] = 127,
+	[0][1][2][1][RTW89_ACMA][7] = 127,
+	[0][1][2][1][RTW89_CN][7] = 127,
+	[0][1][2][1][RTW89_UK][7] = 127,
+	[0][1][2][1][RTW89_FCC][8] = 127,
+	[0][1][2][1][RTW89_ETSI][8] = 127,
+	[0][1][2][1][RTW89_MKK][8] = 127,
+	[0][1][2][1][RTW89_IC][8] = 127,
+	[0][1][2][1][RTW89_KCC][8] = 127,
+	[0][1][2][1][RTW89_ACMA][8] = 127,
+	[0][1][2][1][RTW89_CN][8] = 127,
+	[0][1][2][1][RTW89_UK][8] = 127,
+	[0][1][2][1][RTW89_FCC][9] = 127,
+	[0][1][2][1][RTW89_ETSI][9] = 127,
+	[0][1][2][1][RTW89_MKK][9] = 127,
+	[0][1][2][1][RTW89_IC][9] = 127,
+	[0][1][2][1][RTW89_KCC][9] = 127,
+	[0][1][2][1][RTW89_ACMA][9] = 127,
+	[0][1][2][1][RTW89_CN][9] = 127,
+	[0][1][2][1][RTW89_UK][9] = 127,
+	[0][1][2][1][RTW89_FCC][10] = 127,
+	[0][1][2][1][RTW89_ETSI][10] = 127,
+	[0][1][2][1][RTW89_MKK][10] = 127,
+	[0][1][2][1][RTW89_IC][10] = 127,
+	[0][1][2][1][RTW89_KCC][10] = 127,
+	[0][1][2][1][RTW89_ACMA][10] = 127,
+	[0][1][2][1][RTW89_CN][10] = 127,
+	[0][1][2][1][RTW89_UK][10] = 127,
+	[0][1][2][1][RTW89_FCC][11] = 127,
+	[0][1][2][1][RTW89_ETSI][11] = 127,
+	[0][1][2][1][RTW89_MKK][11] = 127,
+	[0][1][2][1][RTW89_IC][11] = 127,
+	[0][1][2][1][RTW89_KCC][11] = 127,
+	[0][1][2][1][RTW89_ACMA][11] = 127,
+	[0][1][2][1][RTW89_CN][11] = 127,
+	[0][1][2][1][RTW89_UK][11] = 127,
+	[0][1][2][1][RTW89_FCC][12] = 127,
+	[0][1][2][1][RTW89_ETSI][12] = 127,
+	[0][1][2][1][RTW89_MKK][12] = 127,
+	[0][1][2][1][RTW89_IC][12] = 127,
+	[0][1][2][1][RTW89_KCC][12] = 127,
+	[0][1][2][1][RTW89_ACMA][12] = 127,
+	[0][1][2][1][RTW89_CN][12] = 127,
+	[0][1][2][1][RTW89_UK][12] = 127,
+	[0][1][2][1][RTW89_FCC][13] = 127,
+	[0][1][2][1][RTW89_ETSI][13] = 127,
+	[0][1][2][1][RTW89_MKK][13] = 127,
+	[0][1][2][1][RTW89_IC][13] = 127,
+	[0][1][2][1][RTW89_KCC][13] = 127,
+	[0][1][2][1][RTW89_ACMA][13] = 127,
+	[0][1][2][1][RTW89_CN][13] = 127,
+	[0][1][2][1][RTW89_UK][13] = 127,
+	[1][0][2][0][RTW89_FCC][0] = 127,
+	[1][0][2][0][RTW89_ETSI][0] = 127,
+	[1][0][2][0][RTW89_MKK][0] = 127,
+	[1][0][2][0][RTW89_IC][0] = 127,
+	[1][0][2][0][RTW89_KCC][0] = 127,
+	[1][0][2][0][RTW89_ACMA][0] = 127,
+	[1][0][2][0][RTW89_CN][0] = 127,
+	[1][0][2][0][RTW89_UK][0] = 127,
+	[1][0][2][0][RTW89_FCC][1] = 127,
+	[1][0][2][0][RTW89_ETSI][1] = 127,
+	[1][0][2][0][RTW89_MKK][1] = 127,
+	[1][0][2][0][RTW89_IC][1] = 127,
+	[1][0][2][0][RTW89_KCC][1] = 127,
+	[1][0][2][0][RTW89_ACMA][1] = 127,
+	[1][0][2][0][RTW89_CN][1] = 127,
+	[1][0][2][0][RTW89_UK][1] = 127,
+	[1][0][2][0][RTW89_FCC][2] = 72,
+	[1][0][2][0][RTW89_ETSI][2] = 58,
+	[1][0][2][0][RTW89_MKK][2] = 80,
+	[1][0][2][0][RTW89_IC][2] = 72,
+	[1][0][2][0][RTW89_KCC][2] = 80,
+	[1][0][2][0][RTW89_ACMA][2] = 58,
+	[1][0][2][0][RTW89_CN][2] = 60,
+	[1][0][2][0][RTW89_UK][2] = 58,
+	[1][0][2][0][RTW89_FCC][3] = 72,
+	[1][0][2][0][RTW89_ETSI][3] = 58,
+	[1][0][2][0][RTW89_MKK][3] = 80,
+	[1][0][2][0][RTW89_IC][3] = 72,
+	[1][0][2][0][RTW89_KCC][3] = 80,
+	[1][0][2][0][RTW89_ACMA][3] = 58,
+	[1][0][2][0][RTW89_CN][3] = 60,
+	[1][0][2][0][RTW89_UK][3] = 58,
+	[1][0][2][0][RTW89_FCC][4] = 76,
+	[1][0][2][0][RTW89_ETSI][4] = 58,
+	[1][0][2][0][RTW89_MKK][4] = 80,
+	[1][0][2][0][RTW89_IC][4] = 76,
+	[1][0][2][0][RTW89_KCC][4] = 80,
+	[1][0][2][0][RTW89_ACMA][4] = 58,
+	[1][0][2][0][RTW89_CN][4] = 60,
+	[1][0][2][0][RTW89_UK][4] = 58,
+	[1][0][2][0][RTW89_FCC][5] = 78,
+	[1][0][2][0][RTW89_ETSI][5] = 58,
+	[1][0][2][0][RTW89_MKK][5] = 80,
+	[1][0][2][0][RTW89_IC][5] = 78,
+	[1][0][2][0][RTW89_KCC][5] = 80,
+	[1][0][2][0][RTW89_ACMA][5] = 58,
+	[1][0][2][0][RTW89_CN][5] = 60,
+	[1][0][2][0][RTW89_UK][5] = 58,
+	[1][0][2][0][RTW89_FCC][6] = 78,
+	[1][0][2][0][RTW89_ETSI][6] = 58,
+	[1][0][2][0][RTW89_MKK][6] = 78,
+	[1][0][2][0][RTW89_IC][6] = 78,
+	[1][0][2][0][RTW89_KCC][6] = 80,
+	[1][0][2][0][RTW89_ACMA][6] = 58,
+	[1][0][2][0][RTW89_CN][6] = 60,
+	[1][0][2][0][RTW89_UK][6] = 58,
+	[1][0][2][0][RTW89_FCC][7] = 78,
+	[1][0][2][0][RTW89_ETSI][7] = 58,
+	[1][0][2][0][RTW89_MKK][7] = 80,
+	[1][0][2][0][RTW89_IC][7] = 78,
+	[1][0][2][0][RTW89_KCC][7] = 80,
+	[1][0][2][0][RTW89_ACMA][7] = 58,
+	[1][0][2][0][RTW89_CN][7] = 60,
+	[1][0][2][0][RTW89_UK][7] = 58,
+	[1][0][2][0][RTW89_FCC][8] = 78,
+	[1][0][2][0][RTW89_ETSI][8] = 58,
+	[1][0][2][0][RTW89_MKK][8] = 80,
+	[1][0][2][0][RTW89_IC][8] = 78,
+	[1][0][2][0][RTW89_KCC][8] = 78,
+	[1][0][2][0][RTW89_ACMA][8] = 58,
+	[1][0][2][0][RTW89_CN][8] = 60,
+	[1][0][2][0][RTW89_UK][8] = 58,
+	[1][0][2][0][RTW89_FCC][9] = 76,
+	[1][0][2][0][RTW89_ETSI][9] = 58,
+	[1][0][2][0][RTW89_MKK][9] = 80,
+	[1][0][2][0][RTW89_IC][9] = 76,
+	[1][0][2][0][RTW89_KCC][9] = 78,
+	[1][0][2][0][RTW89_ACMA][9] = 58,
+	[1][0][2][0][RTW89_CN][9] = 60,
+	[1][0][2][0][RTW89_UK][9] = 58,
+	[1][0][2][0][RTW89_FCC][10] = 70,
+	[1][0][2][0][RTW89_ETSI][10] = 58,
+	[1][0][2][0][RTW89_MKK][10] = 78,
+	[1][0][2][0][RTW89_IC][10] = 70,
+	[1][0][2][0][RTW89_KCC][10] = 78,
+	[1][0][2][0][RTW89_ACMA][10] = 58,
+	[1][0][2][0][RTW89_CN][10] = 60,
+	[1][0][2][0][RTW89_UK][10] = 58,
+	[1][0][2][0][RTW89_FCC][11] = 127,
+	[1][0][2][0][RTW89_ETSI][11] = 127,
+	[1][0][2][0][RTW89_MKK][11] = 127,
+	[1][0][2][0][RTW89_IC][11] = 127,
+	[1][0][2][0][RTW89_KCC][11] = 127,
+	[1][0][2][0][RTW89_ACMA][11] = 127,
+	[1][0][2][0][RTW89_CN][11] = 127,
+	[1][0][2][0][RTW89_UK][11] = 127,
+	[1][0][2][0][RTW89_FCC][12] = 127,
+	[1][0][2][0][RTW89_ETSI][12] = 127,
+	[1][0][2][0][RTW89_MKK][12] = 127,
+	[1][0][2][0][RTW89_IC][12] = 127,
+	[1][0][2][0][RTW89_KCC][12] = 127,
+	[1][0][2][0][RTW89_ACMA][12] = 127,
+	[1][0][2][0][RTW89_CN][12] = 127,
+	[1][0][2][0][RTW89_UK][12] = 127,
+	[1][0][2][0][RTW89_FCC][13] = 127,
+	[1][0][2][0][RTW89_ETSI][13] = 127,
+	[1][0][2][0][RTW89_MKK][13] = 127,
+	[1][0][2][0][RTW89_IC][13] = 127,
+	[1][0][2][0][RTW89_KCC][13] = 127,
+	[1][0][2][0][RTW89_ACMA][13] = 127,
+	[1][0][2][0][RTW89_CN][13] = 127,
+	[1][0][2][0][RTW89_UK][13] = 127,
+	[1][1][2][0][RTW89_FCC][0] = 127,
+	[1][1][2][0][RTW89_ETSI][0] = 127,
+	[1][1][2][0][RTW89_MKK][0] = 127,
+	[1][1][2][0][RTW89_IC][0] = 127,
+	[1][1][2][0][RTW89_KCC][0] = 127,
+	[1][1][2][0][RTW89_ACMA][0] = 127,
+	[1][1][2][0][RTW89_CN][0] = 127,
+	[1][1][2][0][RTW89_UK][0] = 127,
+	[1][1][2][0][RTW89_FCC][1] = 127,
+	[1][1][2][0][RTW89_ETSI][1] = 127,
+	[1][1][2][0][RTW89_MKK][1] = 127,
+	[1][1][2][0][RTW89_IC][1] = 127,
+	[1][1][2][0][RTW89_KCC][1] = 127,
+	[1][1][2][0][RTW89_ACMA][1] = 127,
+	[1][1][2][0][RTW89_CN][1] = 127,
+	[1][1][2][0][RTW89_UK][1] = 127,
+	[1][1][2][0][RTW89_FCC][2] = 127,
+	[1][1][2][0][RTW89_ETSI][2] = 127,
+	[1][1][2][0][RTW89_MKK][2] = 127,
+	[1][1][2][0][RTW89_IC][2] = 127,
+	[1][1][2][0][RTW89_KCC][2] = 127,
+	[1][1][2][0][RTW89_ACMA][2] = 127,
+	[1][1][2][0][RTW89_CN][2] = 127,
+	[1][1][2][0][RTW89_UK][2] = 127,
+	[1][1][2][0][RTW89_FCC][3] = 127,
+	[1][1][2][0][RTW89_ETSI][3] = 127,
+	[1][1][2][0][RTW89_MKK][3] = 127,
+	[1][1][2][0][RTW89_IC][3] = 127,
+	[1][1][2][0][RTW89_KCC][3] = 127,
+	[1][1][2][0][RTW89_ACMA][3] = 127,
+	[1][1][2][0][RTW89_CN][3] = 127,
+	[1][1][2][0][RTW89_UK][3] = 127,
+	[1][1][2][0][RTW89_FCC][4] = 127,
+	[1][1][2][0][RTW89_ETSI][4] = 127,
+	[1][1][2][0][RTW89_MKK][4] = 127,
+	[1][1][2][0][RTW89_IC][4] = 127,
+	[1][1][2][0][RTW89_KCC][4] = 127,
+	[1][1][2][0][RTW89_ACMA][4] = 127,
+	[1][1][2][0][RTW89_CN][4] = 127,
+	[1][1][2][0][RTW89_UK][4] = 127,
+	[1][1][2][0][RTW89_FCC][5] = 127,
+	[1][1][2][0][RTW89_ETSI][5] = 127,
+	[1][1][2][0][RTW89_MKK][5] = 127,
+	[1][1][2][0][RTW89_IC][5] = 127,
+	[1][1][2][0][RTW89_KCC][5] = 127,
+	[1][1][2][0][RTW89_ACMA][5] = 127,
+	[1][1][2][0][RTW89_CN][5] = 127,
+	[1][1][2][0][RTW89_UK][5] = 127,
+	[1][1][2][0][RTW89_FCC][6] = 127,
+	[1][1][2][0][RTW89_ETSI][6] = 127,
+	[1][1][2][0][RTW89_MKK][6] = 127,
+	[1][1][2][0][RTW89_IC][6] = 127,
+	[1][1][2][0][RTW89_KCC][6] = 127,
+	[1][1][2][0][RTW89_ACMA][6] = 127,
+	[1][1][2][0][RTW89_CN][6] = 127,
+	[1][1][2][0][RTW89_UK][6] = 127,
+	[1][1][2][0][RTW89_FCC][7] = 127,
+	[1][1][2][0][RTW89_ETSI][7] = 127,
+	[1][1][2][0][RTW89_MKK][7] = 127,
+	[1][1][2][0][RTW89_IC][7] = 127,
+	[1][1][2][0][RTW89_KCC][7] = 127,
+	[1][1][2][0][RTW89_ACMA][7] = 127,
+	[1][1][2][0][RTW89_CN][7] = 127,
+	[1][1][2][0][RTW89_UK][7] = 127,
+	[1][1][2][0][RTW89_FCC][8] = 127,
+	[1][1][2][0][RTW89_ETSI][8] = 127,
+	[1][1][2][0][RTW89_MKK][8] = 127,
+	[1][1][2][0][RTW89_IC][8] = 127,
+	[1][1][2][0][RTW89_KCC][8] = 127,
+	[1][1][2][0][RTW89_ACMA][8] = 127,
+	[1][1][2][0][RTW89_CN][8] = 127,
+	[1][1][2][0][RTW89_UK][8] = 127,
+	[1][1][2][0][RTW89_FCC][9] = 127,
+	[1][1][2][0][RTW89_ETSI][9] = 127,
+	[1][1][2][0][RTW89_MKK][9] = 127,
+	[1][1][2][0][RTW89_IC][9] = 127,
+	[1][1][2][0][RTW89_KCC][9] = 127,
+	[1][1][2][0][RTW89_ACMA][9] = 127,
+	[1][1][2][0][RTW89_CN][9] = 127,
+	[1][1][2][0][RTW89_UK][9] = 127,
+	[1][1][2][0][RTW89_FCC][10] = 127,
+	[1][1][2][0][RTW89_ETSI][10] = 127,
+	[1][1][2][0][RTW89_MKK][10] = 127,
+	[1][1][2][0][RTW89_IC][10] = 127,
+	[1][1][2][0][RTW89_KCC][10] = 127,
+	[1][1][2][0][RTW89_ACMA][10] = 127,
+	[1][1][2][0][RTW89_CN][10] = 127,
+	[1][1][2][0][RTW89_UK][10] = 127,
+	[1][1][2][0][RTW89_FCC][11] = 127,
+	[1][1][2][0][RTW89_ETSI][11] = 127,
+	[1][1][2][0][RTW89_MKK][11] = 127,
+	[1][1][2][0][RTW89_IC][11] = 127,
+	[1][1][2][0][RTW89_KCC][11] = 127,
+	[1][1][2][0][RTW89_ACMA][11] = 127,
+	[1][1][2][0][RTW89_CN][11] = 127,
+	[1][1][2][0][RTW89_UK][11] = 127,
+	[1][1][2][0][RTW89_FCC][12] = 127,
+	[1][1][2][0][RTW89_ETSI][12] = 127,
+	[1][1][2][0][RTW89_MKK][12] = 127,
+	[1][1][2][0][RTW89_IC][12] = 127,
+	[1][1][2][0][RTW89_KCC][12] = 127,
+	[1][1][2][0][RTW89_ACMA][12] = 127,
+	[1][1][2][0][RTW89_CN][12] = 127,
+	[1][1][2][0][RTW89_UK][12] = 127,
+	[1][1][2][0][RTW89_FCC][13] = 127,
+	[1][1][2][0][RTW89_ETSI][13] = 127,
+	[1][1][2][0][RTW89_MKK][13] = 127,
+	[1][1][2][0][RTW89_IC][13] = 127,
+	[1][1][2][0][RTW89_KCC][13] = 127,
+	[1][1][2][0][RTW89_ACMA][13] = 127,
+	[1][1][2][0][RTW89_CN][13] = 127,
+	[1][1][2][0][RTW89_UK][13] = 127,
+	[1][1][2][1][RTW89_FCC][0] = 127,
+	[1][1][2][1][RTW89_ETSI][0] = 127,
+	[1][1][2][1][RTW89_MKK][0] = 127,
+	[1][1][2][1][RTW89_IC][0] = 127,
+	[1][1][2][1][RTW89_KCC][0] = 127,
+	[1][1][2][1][RTW89_ACMA][0] = 127,
+	[1][1][2][1][RTW89_CN][0] = 127,
+	[1][1][2][1][RTW89_UK][0] = 127,
+	[1][1][2][1][RTW89_FCC][1] = 127,
+	[1][1][2][1][RTW89_ETSI][1] = 127,
+	[1][1][2][1][RTW89_MKK][1] = 127,
+	[1][1][2][1][RTW89_IC][1] = 127,
+	[1][1][2][1][RTW89_KCC][1] = 127,
+	[1][1][2][1][RTW89_ACMA][1] = 127,
+	[1][1][2][1][RTW89_CN][1] = 127,
+	[1][1][2][1][RTW89_UK][1] = 127,
+	[1][1][2][1][RTW89_FCC][2] = 127,
+	[1][1][2][1][RTW89_ETSI][2] = 127,
+	[1][1][2][1][RTW89_MKK][2] = 127,
+	[1][1][2][1][RTW89_IC][2] = 127,
+	[1][1][2][1][RTW89_KCC][2] = 127,
+	[1][1][2][1][RTW89_ACMA][2] = 127,
+	[1][1][2][1][RTW89_CN][2] = 127,
+	[1][1][2][1][RTW89_UK][2] = 127,
+	[1][1][2][1][RTW89_FCC][3] = 127,
+	[1][1][2][1][RTW89_ETSI][3] = 127,
+	[1][1][2][1][RTW89_MKK][3] = 127,
+	[1][1][2][1][RTW89_IC][3] = 127,
+	[1][1][2][1][RTW89_KCC][3] = 127,
+	[1][1][2][1][RTW89_ACMA][3] = 127,
+	[1][1][2][1][RTW89_CN][3] = 127,
+	[1][1][2][1][RTW89_UK][3] = 127,
+	[1][1][2][1][RTW89_FCC][4] = 127,
+	[1][1][2][1][RTW89_ETSI][4] = 127,
+	[1][1][2][1][RTW89_MKK][4] = 127,
+	[1][1][2][1][RTW89_IC][4] = 127,
+	[1][1][2][1][RTW89_KCC][4] = 127,
+	[1][1][2][1][RTW89_ACMA][4] = 127,
+	[1][1][2][1][RTW89_CN][4] = 127,
+	[1][1][2][1][RTW89_UK][4] = 127,
+	[1][1][2][1][RTW89_FCC][5] = 127,
+	[1][1][2][1][RTW89_ETSI][5] = 127,
+	[1][1][2][1][RTW89_MKK][5] = 127,
+	[1][1][2][1][RTW89_IC][5] = 127,
+	[1][1][2][1][RTW89_KCC][5] = 127,
+	[1][1][2][1][RTW89_ACMA][5] = 127,
+	[1][1][2][1][RTW89_CN][5] = 127,
+	[1][1][2][1][RTW89_UK][5] = 127,
+	[1][1][2][1][RTW89_FCC][6] = 127,
+	[1][1][2][1][RTW89_ETSI][6] = 127,
+	[1][1][2][1][RTW89_MKK][6] = 127,
+	[1][1][2][1][RTW89_IC][6] = 127,
+	[1][1][2][1][RTW89_KCC][6] = 127,
+	[1][1][2][1][RTW89_ACMA][6] = 127,
+	[1][1][2][1][RTW89_CN][6] = 127,
+	[1][1][2][1][RTW89_UK][6] = 127,
+	[1][1][2][1][RTW89_FCC][7] = 127,
+	[1][1][2][1][RTW89_ETSI][7] = 127,
+	[1][1][2][1][RTW89_MKK][7] = 127,
+	[1][1][2][1][RTW89_IC][7] = 127,
+	[1][1][2][1][RTW89_KCC][7] = 127,
+	[1][1][2][1][RTW89_ACMA][7] = 127,
+	[1][1][2][1][RTW89_CN][7] = 127,
+	[1][1][2][1][RTW89_UK][7] = 127,
+	[1][1][2][1][RTW89_FCC][8] = 127,
+	[1][1][2][1][RTW89_ETSI][8] = 127,
+	[1][1][2][1][RTW89_MKK][8] = 127,
+	[1][1][2][1][RTW89_IC][8] = 127,
+	[1][1][2][1][RTW89_KCC][8] = 127,
+	[1][1][2][1][RTW89_ACMA][8] = 127,
+	[1][1][2][1][RTW89_CN][8] = 127,
+	[1][1][2][1][RTW89_UK][8] = 127,
+	[1][1][2][1][RTW89_FCC][9] = 127,
+	[1][1][2][1][RTW89_ETSI][9] = 127,
+	[1][1][2][1][RTW89_MKK][9] = 127,
+	[1][1][2][1][RTW89_IC][9] = 127,
+	[1][1][2][1][RTW89_KCC][9] = 127,
+	[1][1][2][1][RTW89_ACMA][9] = 127,
+	[1][1][2][1][RTW89_CN][9] = 127,
+	[1][1][2][1][RTW89_UK][9] = 127,
+	[1][1][2][1][RTW89_FCC][10] = 127,
+	[1][1][2][1][RTW89_ETSI][10] = 127,
+	[1][1][2][1][RTW89_MKK][10] = 127,
+	[1][1][2][1][RTW89_IC][10] = 127,
+	[1][1][2][1][RTW89_KCC][10] = 127,
+	[1][1][2][1][RTW89_ACMA][10] = 127,
+	[1][1][2][1][RTW89_CN][10] = 127,
+	[1][1][2][1][RTW89_UK][10] = 127,
+	[1][1][2][1][RTW89_FCC][11] = 127,
+	[1][1][2][1][RTW89_ETSI][11] = 127,
+	[1][1][2][1][RTW89_MKK][11] = 127,
+	[1][1][2][1][RTW89_IC][11] = 127,
+	[1][1][2][1][RTW89_KCC][11] = 127,
+	[1][1][2][1][RTW89_ACMA][11] = 127,
+	[1][1][2][1][RTW89_CN][11] = 127,
+	[1][1][2][1][RTW89_UK][11] = 127,
+	[1][1][2][1][RTW89_FCC][12] = 127,
+	[1][1][2][1][RTW89_ETSI][12] = 127,
+	[1][1][2][1][RTW89_MKK][12] = 127,
+	[1][1][2][1][RTW89_IC][12] = 127,
+	[1][1][2][1][RTW89_KCC][12] = 127,
+	[1][1][2][1][RTW89_ACMA][12] = 127,
+	[1][1][2][1][RTW89_CN][12] = 127,
+	[1][1][2][1][RTW89_UK][12] = 127,
+	[1][1][2][1][RTW89_FCC][13] = 127,
+	[1][1][2][1][RTW89_ETSI][13] = 127,
+	[1][1][2][1][RTW89_MKK][13] = 127,
+	[1][1][2][1][RTW89_IC][13] = 127,
+	[1][1][2][1][RTW89_KCC][13] = 127,
+	[1][1][2][1][RTW89_ACMA][13] = 127,
+	[1][1][2][1][RTW89_CN][13] = 127,
+	[1][1][2][1][RTW89_UK][13] = 127,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
+				 [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+				 [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
+	[0][0][1][0][RTW89_WW][0] = 58,
+	[0][0][1][0][RTW89_WW][2] = 58,
+	[0][0][1][0][RTW89_WW][4] = 58,
+	[0][0][1][0][RTW89_WW][6] = 50,
+	[0][0][1][0][RTW89_WW][8] = 58,
+	[0][0][1][0][RTW89_WW][10] = 58,
+	[0][0][1][0][RTW89_WW][12] = 58,
+	[0][0][1][0][RTW89_WW][14] = 58,
+	[0][0][1][0][RTW89_WW][15] = 58,
+	[0][0][1][0][RTW89_WW][17] = 60,
+	[0][0][1][0][RTW89_WW][19] = 60,
+	[0][0][1][0][RTW89_WW][21] = 60,
+	[0][0][1][0][RTW89_WW][23] = 60,
+	[0][0][1][0][RTW89_WW][25] = 60,
+	[0][0][1][0][RTW89_WW][27] = 60,
+	[0][0][1][0][RTW89_WW][29] = 60,
+	[0][0][1][0][RTW89_WW][31] = 60,
+	[0][0][1][0][RTW89_WW][33] = 60,
+	[0][0][1][0][RTW89_WW][35] = 60,
+	[0][0][1][0][RTW89_WW][37] = 74,
+	[0][0][1][0][RTW89_WW][38] = 30,
+	[0][0][1][0][RTW89_WW][40] = 30,
+	[0][0][1][0][RTW89_WW][42] = 30,
+	[0][0][1][0][RTW89_WW][44] = 30,
+	[0][0][1][0][RTW89_WW][46] = 30,
+	[0][0][1][0][RTW89_WW][48] = 72,
+	[0][0][1][0][RTW89_WW][50] = 72,
+	[0][0][1][0][RTW89_WW][52] = 72,
+	[0][1][1][0][RTW89_WW][0] = 0,
+	[0][1][1][0][RTW89_WW][2] = 0,
+	[0][1][1][0][RTW89_WW][4] = 0,
+	[0][1][1][0][RTW89_WW][6] = 0,
+	[0][1][1][0][RTW89_WW][8] = 0,
+	[0][1][1][0][RTW89_WW][10] = 0,
+	[0][1][1][0][RTW89_WW][12] = 0,
+	[0][1][1][0][RTW89_WW][14] = 0,
+	[0][1][1][0][RTW89_WW][15] = 0,
+	[0][1][1][0][RTW89_WW][17] = 0,
+	[0][1][1][0][RTW89_WW][19] = 0,
+	[0][1][1][0][RTW89_WW][21] = 0,
+	[0][1][1][0][RTW89_WW][23] = 0,
+	[0][1][1][0][RTW89_WW][25] = 0,
+	[0][1][1][0][RTW89_WW][27] = 0,
+	[0][1][1][0][RTW89_WW][29] = 0,
+	[0][1][1][0][RTW89_WW][31] = 0,
+	[0][1][1][0][RTW89_WW][33] = 0,
+	[0][1][1][0][RTW89_WW][35] = 0,
+	[0][1][1][0][RTW89_WW][37] = 0,
+	[0][1][1][0][RTW89_WW][38] = 0,
+	[0][1][1][0][RTW89_WW][40] = 0,
+	[0][1][1][0][RTW89_WW][42] = 0,
+	[0][1][1][0][RTW89_WW][44] = 0,
+	[0][1][1][0][RTW89_WW][46] = 0,
+	[0][1][1][0][RTW89_WW][48] = 0,
+	[0][1][1][0][RTW89_WW][50] = 0,
+	[0][1][1][0][RTW89_WW][52] = 0,
+	[0][0][2][0][RTW89_WW][0] = 62,
+	[0][0][2][0][RTW89_WW][2] = 62,
+	[0][0][2][0][RTW89_WW][4] = 62,
+	[0][0][2][0][RTW89_WW][6] = 54,
+	[0][0][2][0][RTW89_WW][8] = 62,
+	[0][0][2][0][RTW89_WW][10] = 62,
+	[0][0][2][0][RTW89_WW][12] = 62,
+	[0][0][2][0][RTW89_WW][14] = 62,
+	[0][0][2][0][RTW89_WW][15] = 60,
+	[0][0][2][0][RTW89_WW][17] = 62,
+	[0][0][2][0][RTW89_WW][19] = 62,
+	[0][0][2][0][RTW89_WW][21] = 62,
+	[0][0][2][0][RTW89_WW][23] = 62,
+	[0][0][2][0][RTW89_WW][25] = 62,
+	[0][0][2][0][RTW89_WW][27] = 62,
+	[0][0][2][0][RTW89_WW][29] = 62,
+	[0][0][2][0][RTW89_WW][31] = 62,
+	[0][0][2][0][RTW89_WW][33] = 62,
+	[0][0][2][0][RTW89_WW][35] = 62,
+	[0][0][2][0][RTW89_WW][37] = 76,
+	[0][0][2][0][RTW89_WW][38] = 30,
+	[0][0][2][0][RTW89_WW][40] = 30,
+	[0][0][2][0][RTW89_WW][42] = 30,
+	[0][0][2][0][RTW89_WW][44] = 30,
+	[0][0][2][0][RTW89_WW][46] = 30,
+	[0][0][2][0][RTW89_WW][48] = 74,
+	[0][0][2][0][RTW89_WW][50] = 76,
+	[0][0][2][0][RTW89_WW][52] = 76,
+	[0][1][2][0][RTW89_WW][0] = 0,
+	[0][1][2][0][RTW89_WW][2] = 0,
+	[0][1][2][0][RTW89_WW][4] = 0,
+	[0][1][2][0][RTW89_WW][6] = 0,
+	[0][1][2][0][RTW89_WW][8] = 0,
+	[0][1][2][0][RTW89_WW][10] = 0,
+	[0][1][2][0][RTW89_WW][12] = 0,
+	[0][1][2][0][RTW89_WW][14] = 0,
+	[0][1][2][0][RTW89_WW][15] = 0,
+	[0][1][2][0][RTW89_WW][17] = 0,
+	[0][1][2][0][RTW89_WW][19] = 0,
+	[0][1][2][0][RTW89_WW][21] = 0,
+	[0][1][2][0][RTW89_WW][23] = 0,
+	[0][1][2][0][RTW89_WW][25] = 0,
+	[0][1][2][0][RTW89_WW][27] = 0,
+	[0][1][2][0][RTW89_WW][29] = 0,
+	[0][1][2][0][RTW89_WW][31] = 0,
+	[0][1][2][0][RTW89_WW][33] = 0,
+	[0][1][2][0][RTW89_WW][35] = 0,
+	[0][1][2][0][RTW89_WW][37] = 0,
+	[0][1][2][0][RTW89_WW][38] = 0,
+	[0][1][2][0][RTW89_WW][40] = 0,
+	[0][1][2][0][RTW89_WW][42] = 0,
+	[0][1][2][0][RTW89_WW][44] = 0,
+	[0][1][2][0][RTW89_WW][46] = 0,
+	[0][1][2][0][RTW89_WW][48] = 0,
+	[0][1][2][0][RTW89_WW][50] = 0,
+	[0][1][2][0][RTW89_WW][52] = 0,
+	[0][1][2][1][RTW89_WW][0] = 0,
+	[0][1][2][1][RTW89_WW][2] = 0,
+	[0][1][2][1][RTW89_WW][4] = 0,
+	[0][1][2][1][RTW89_WW][6] = 0,
+	[0][1][2][1][RTW89_WW][8] = 0,
+	[0][1][2][1][RTW89_WW][10] = 0,
+	[0][1][2][1][RTW89_WW][12] = 0,
+	[0][1][2][1][RTW89_WW][14] = 0,
+	[0][1][2][1][RTW89_WW][15] = 0,
+	[0][1][2][1][RTW89_WW][17] = 0,
+	[0][1][2][1][RTW89_WW][19] = 0,
+	[0][1][2][1][RTW89_WW][21] = 0,
+	[0][1][2][1][RTW89_WW][23] = 0,
+	[0][1][2][1][RTW89_WW][25] = 0,
+	[0][1][2][1][RTW89_WW][27] = 0,
+	[0][1][2][1][RTW89_WW][29] = 0,
+	[0][1][2][1][RTW89_WW][31] = 0,
+	[0][1][2][1][RTW89_WW][33] = 0,
+	[0][1][2][1][RTW89_WW][35] = 0,
+	[0][1][2][1][RTW89_WW][37] = 0,
+	[0][1][2][1][RTW89_WW][38] = 0,
+	[0][1][2][1][RTW89_WW][40] = 0,
+	[0][1][2][1][RTW89_WW][42] = 0,
+	[0][1][2][1][RTW89_WW][44] = 0,
+	[0][1][2][1][RTW89_WW][46] = 0,
+	[0][1][2][1][RTW89_WW][48] = 0,
+	[0][1][2][1][RTW89_WW][50] = 0,
+	[0][1][2][1][RTW89_WW][52] = 0,
+	[1][0][2][0][RTW89_WW][1] = 64,
+	[1][0][2][0][RTW89_WW][5] = 62,
+	[1][0][2][0][RTW89_WW][9] = 64,
+	[1][0][2][0][RTW89_WW][13] = 64,
+	[1][0][2][0][RTW89_WW][16] = 66,
+	[1][0][2][0][RTW89_WW][20] = 66,
+	[1][0][2][0][RTW89_WW][24] = 66,
+	[1][0][2][0][RTW89_WW][28] = 66,
+	[1][0][2][0][RTW89_WW][32] = 66,
+	[1][0][2][0][RTW89_WW][36] = 76,
+	[1][0][2][0][RTW89_WW][39] = 30,
+	[1][0][2][0][RTW89_WW][43] = 30,
+	[1][0][2][0][RTW89_WW][47] = 84,
+	[1][0][2][0][RTW89_WW][51] = 84,
+	[1][1][2][0][RTW89_WW][1] = 0,
+	[1][1][2][0][RTW89_WW][5] = 0,
+	[1][1][2][0][RTW89_WW][9] = 0,
+	[1][1][2][0][RTW89_WW][13] = 0,
+	[1][1][2][0][RTW89_WW][16] = 0,
+	[1][1][2][0][RTW89_WW][20] = 0,
+	[1][1][2][0][RTW89_WW][24] = 0,
+	[1][1][2][0][RTW89_WW][28] = 0,
+	[1][1][2][0][RTW89_WW][32] = 0,
+	[1][1][2][0][RTW89_WW][36] = 0,
+	[1][1][2][0][RTW89_WW][39] = 0,
+	[1][1][2][0][RTW89_WW][43] = 0,
+	[1][1][2][0][RTW89_WW][47] = 0,
+	[1][1][2][0][RTW89_WW][51] = 0,
+	[1][1][2][1][RTW89_WW][1] = 0,
+	[1][1][2][1][RTW89_WW][5] = 0,
+	[1][1][2][1][RTW89_WW][9] = 0,
+	[1][1][2][1][RTW89_WW][13] = 0,
+	[1][1][2][1][RTW89_WW][16] = 0,
+	[1][1][2][1][RTW89_WW][20] = 0,
+	[1][1][2][1][RTW89_WW][24] = 0,
+	[1][1][2][1][RTW89_WW][28] = 0,
+	[1][1][2][1][RTW89_WW][32] = 0,
+	[1][1][2][1][RTW89_WW][36] = 0,
+	[1][1][2][1][RTW89_WW][39] = 0,
+	[1][1][2][1][RTW89_WW][43] = 0,
+	[1][1][2][1][RTW89_WW][47] = 0,
+	[1][1][2][1][RTW89_WW][51] = 0,
+	[2][0][2][0][RTW89_WW][3] = 62,
+	[2][0][2][0][RTW89_WW][11] = 62,
+	[2][0][2][0][RTW89_WW][18] = 64,
+	[2][0][2][0][RTW89_WW][26] = 64,
+	[2][0][2][0][RTW89_WW][34] = 72,
+	[2][0][2][0][RTW89_WW][41] = 30,
+	[2][0][2][0][RTW89_WW][49] = 74,
+	[2][1][2][0][RTW89_WW][3] = 0,
+	[2][1][2][0][RTW89_WW][11] = 0,
+	[2][1][2][0][RTW89_WW][18] = 0,
+	[2][1][2][0][RTW89_WW][26] = 0,
+	[2][1][2][0][RTW89_WW][34] = 0,
+	[2][1][2][0][RTW89_WW][41] = 0,
+	[2][1][2][0][RTW89_WW][49] = 0,
+	[2][1][2][1][RTW89_WW][3] = 0,
+	[2][1][2][1][RTW89_WW][11] = 0,
+	[2][1][2][1][RTW89_WW][18] = 0,
+	[2][1][2][1][RTW89_WW][26] = 0,
+	[2][1][2][1][RTW89_WW][34] = 0,
+	[2][1][2][1][RTW89_WW][41] = 0,
+	[2][1][2][1][RTW89_WW][49] = 0,
+	[3][0][2][0][RTW89_WW][7] = 58,
+	[3][0][2][0][RTW89_WW][22] = 58,
+	[3][0][2][0][RTW89_WW][45] = 0,
+	[3][1][2][0][RTW89_WW][7] = 0,
+	[3][1][2][0][RTW89_WW][22] = 0,
+	[3][1][2][0][RTW89_WW][45] = 0,
+	[3][1][2][1][RTW89_WW][7] = 0,
+	[3][1][2][1][RTW89_WW][22] = 0,
+	[3][1][2][1][RTW89_WW][45] = 0,
+	[0][0][1][0][RTW89_FCC][0] = 80,
+	[0][0][1][0][RTW89_ETSI][0] = 58,
+	[0][0][1][0][RTW89_MKK][0] = 60,
+	[0][0][1][0][RTW89_IC][0] = 62,
+	[0][0][1][0][RTW89_KCC][0] = 74,
+	[0][0][1][0][RTW89_ACMA][0] = 58,
+	[0][0][1][0][RTW89_CN][0] = 60,
+	[0][0][1][0][RTW89_UK][0] = 58,
+	[0][0][1][0][RTW89_FCC][2] = 82,
+	[0][0][1][0][RTW89_ETSI][2] = 58,
+	[0][0][1][0][RTW89_MKK][2] = 60,
+	[0][0][1][0][RTW89_IC][2] = 62,
+	[0][0][1][0][RTW89_KCC][2] = 74,
+	[0][0][1][0][RTW89_ACMA][2] = 58,
+	[0][0][1][0][RTW89_CN][2] = 60,
+	[0][0][1][0][RTW89_UK][2] = 58,
+	[0][0][1][0][RTW89_FCC][4] = 82,
+	[0][0][1][0][RTW89_ETSI][4] = 58,
+	[0][0][1][0][RTW89_MKK][4] = 60,
+	[0][0][1][0][RTW89_IC][4] = 62,
+	[0][0][1][0][RTW89_KCC][4] = 74,
+	[0][0][1][0][RTW89_ACMA][4] = 58,
+	[0][0][1][0][RTW89_CN][4] = 60,
+	[0][0][1][0][RTW89_UK][4] = 58,
+	[0][0][1][0][RTW89_FCC][6] = 82,
+	[0][0][1][0][RTW89_ETSI][6] = 58,
+	[0][0][1][0][RTW89_MKK][6] = 60,
+	[0][0][1][0][RTW89_IC][6] = 62,
+	[0][0][1][0][RTW89_KCC][6] = 50,
+	[0][0][1][0][RTW89_ACMA][6] = 58,
+	[0][0][1][0][RTW89_CN][6] = 60,
+	[0][0][1][0][RTW89_UK][6] = 58,
+	[0][0][1][0][RTW89_FCC][8] = 82,
+	[0][0][1][0][RTW89_ETSI][8] = 58,
+	[0][0][1][0][RTW89_MKK][8] = 60,
+	[0][0][1][0][RTW89_IC][8] = 64,
+	[0][0][1][0][RTW89_KCC][8] = 74,
+	[0][0][1][0][RTW89_ACMA][8] = 58,
+	[0][0][1][0][RTW89_CN][8] = 60,
+	[0][0][1][0][RTW89_UK][8] = 58,
+	[0][0][1][0][RTW89_FCC][10] = 82,
+	[0][0][1][0][RTW89_ETSI][10] = 58,
+	[0][0][1][0][RTW89_MKK][10] = 60,
+	[0][0][1][0][RTW89_IC][10] = 64,
+	[0][0][1][0][RTW89_KCC][10] = 74,
+	[0][0][1][0][RTW89_ACMA][10] = 58,
+	[0][0][1][0][RTW89_CN][10] = 60,
+	[0][0][1][0][RTW89_UK][10] = 58,
+	[0][0][1][0][RTW89_FCC][12] = 82,
+	[0][0][1][0][RTW89_ETSI][12] = 58,
+	[0][0][1][0][RTW89_MKK][12] = 60,
+	[0][0][1][0][RTW89_IC][12] = 64,
+	[0][0][1][0][RTW89_KCC][12] = 76,
+	[0][0][1][0][RTW89_ACMA][12] = 58,
+	[0][0][1][0][RTW89_CN][12] = 60,
+	[0][0][1][0][RTW89_UK][12] = 58,
+	[0][0][1][0][RTW89_FCC][14] = 78,
+	[0][0][1][0][RTW89_ETSI][14] = 58,
+	[0][0][1][0][RTW89_MKK][14] = 60,
+	[0][0][1][0][RTW89_IC][14] = 64,
+	[0][0][1][0][RTW89_KCC][14] = 76,
+	[0][0][1][0][RTW89_ACMA][14] = 58,
+	[0][0][1][0][RTW89_CN][14] = 60,
+	[0][0][1][0][RTW89_UK][14] = 58,
+	[0][0][1][0][RTW89_FCC][15] = 78,
+	[0][0][1][0][RTW89_ETSI][15] = 58,
+	[0][0][1][0][RTW89_MKK][15] = 78,
+	[0][0][1][0][RTW89_IC][15] = 78,
+	[0][0][1][0][RTW89_KCC][15] = 78,
+	[0][0][1][0][RTW89_ACMA][15] = 58,
+	[0][0][1][0][RTW89_CN][15] = 127,
+	[0][0][1][0][RTW89_UK][15] = 58,
+	[0][0][1][0][RTW89_FCC][17] = 82,
+	[0][0][1][0][RTW89_ETSI][17] = 60,
+	[0][0][1][0][RTW89_MKK][17] = 78,
+	[0][0][1][0][RTW89_IC][17] = 82,
+	[0][0][1][0][RTW89_KCC][17] = 78,
+	[0][0][1][0][RTW89_ACMA][17] = 60,
+	[0][0][1][0][RTW89_CN][17] = 127,
+	[0][0][1][0][RTW89_UK][17] = 60,
+	[0][0][1][0][RTW89_FCC][19] = 82,
+	[0][0][1][0][RTW89_ETSI][19] = 60,
+	[0][0][1][0][RTW89_MKK][19] = 78,
+	[0][0][1][0][RTW89_IC][19] = 82,
+	[0][0][1][0][RTW89_KCC][19] = 78,
+	[0][0][1][0][RTW89_ACMA][19] = 60,
+	[0][0][1][0][RTW89_CN][19] = 127,
+	[0][0][1][0][RTW89_UK][19] = 60,
+	[0][0][1][0][RTW89_FCC][21] = 82,
+	[0][0][1][0][RTW89_ETSI][21] = 60,
+	[0][0][1][0][RTW89_MKK][21] = 78,
+	[0][0][1][0][RTW89_IC][21] = 82,
+	[0][0][1][0][RTW89_KCC][21] = 78,
+	[0][0][1][0][RTW89_ACMA][21] = 60,
+	[0][0][1][0][RTW89_CN][21] = 127,
+	[0][0][1][0][RTW89_UK][21] = 60,
+	[0][0][1][0][RTW89_FCC][23] = 82,
+	[0][0][1][0][RTW89_ETSI][23] = 60,
+	[0][0][1][0][RTW89_MKK][23] = 78,
+	[0][0][1][0][RTW89_IC][23] = 82,
+	[0][0][1][0][RTW89_KCC][23] = 78,
+	[0][0][1][0][RTW89_ACMA][23] = 60,
+	[0][0][1][0][RTW89_CN][23] = 127,
+	[0][0][1][0][RTW89_UK][23] = 60,
+	[0][0][1][0][RTW89_FCC][25] = 82,
+	[0][0][1][0][RTW89_ETSI][25] = 60,
+	[0][0][1][0][RTW89_MKK][25] = 78,
+	[0][0][1][0][RTW89_IC][25] = 127,
+	[0][0][1][0][RTW89_KCC][25] = 78,
+	[0][0][1][0][RTW89_ACMA][25] = 127,
+	[0][0][1][0][RTW89_CN][25] = 127,
+	[0][0][1][0][RTW89_UK][25] = 60,
+	[0][0][1][0][RTW89_FCC][27] = 82,
+	[0][0][1][0][RTW89_ETSI][27] = 60,
+	[0][0][1][0][RTW89_MKK][27] = 78,
+	[0][0][1][0][RTW89_IC][27] = 127,
+	[0][0][1][0][RTW89_KCC][27] = 78,
+	[0][0][1][0][RTW89_ACMA][27] = 127,
+	[0][0][1][0][RTW89_CN][27] = 127,
+	[0][0][1][0][RTW89_UK][27] = 60,
+	[0][0][1][0][RTW89_FCC][29] = 82,
+	[0][0][1][0][RTW89_ETSI][29] = 60,
+	[0][0][1][0][RTW89_MKK][29] = 78,
+	[0][0][1][0][RTW89_IC][29] = 127,
+	[0][0][1][0][RTW89_KCC][29] = 78,
+	[0][0][1][0][RTW89_ACMA][29] = 127,
+	[0][0][1][0][RTW89_CN][29] = 127,
+	[0][0][1][0][RTW89_UK][29] = 60,
+	[0][0][1][0][RTW89_FCC][31] = 82,
+	[0][0][1][0][RTW89_ETSI][31] = 60,
+	[0][0][1][0][RTW89_MKK][31] = 78,
+	[0][0][1][0][RTW89_IC][31] = 82,
+	[0][0][1][0][RTW89_KCC][31] = 74,
+	[0][0][1][0][RTW89_ACMA][31] = 60,
+	[0][0][1][0][RTW89_CN][31] = 127,
+	[0][0][1][0][RTW89_UK][31] = 60,
+	[0][0][1][0][RTW89_FCC][33] = 82,
+	[0][0][1][0][RTW89_ETSI][33] = 60,
+	[0][0][1][0][RTW89_MKK][33] = 78,
+	[0][0][1][0][RTW89_IC][33] = 82,
+	[0][0][1][0][RTW89_KCC][33] = 74,
+	[0][0][1][0][RTW89_ACMA][33] = 60,
+	[0][0][1][0][RTW89_CN][33] = 127,
+	[0][0][1][0][RTW89_UK][33] = 60,
+	[0][0][1][0][RTW89_FCC][35] = 72,
+	[0][0][1][0][RTW89_ETSI][35] = 60,
+	[0][0][1][0][RTW89_MKK][35] = 78,
+	[0][0][1][0][RTW89_IC][35] = 72,
+	[0][0][1][0][RTW89_KCC][35] = 74,
+	[0][0][1][0][RTW89_ACMA][35] = 60,
+	[0][0][1][0][RTW89_CN][35] = 127,
+	[0][0][1][0][RTW89_UK][35] = 60,
+	[0][0][1][0][RTW89_FCC][37] = 82,
+	[0][0][1][0][RTW89_ETSI][37] = 127,
+	[0][0][1][0][RTW89_MKK][37] = 78,
+	[0][0][1][0][RTW89_IC][37] = 82,
+	[0][0][1][0][RTW89_KCC][37] = 74,
+	[0][0][1][0][RTW89_ACMA][37] = 78,
+	[0][0][1][0][RTW89_CN][37] = 127,
+	[0][0][1][0][RTW89_UK][37] = 78,
+	[0][0][1][0][RTW89_FCC][38] = 82,
+	[0][0][1][0][RTW89_ETSI][38] = 30,
+	[0][0][1][0][RTW89_MKK][38] = 127,
+	[0][0][1][0][RTW89_IC][38] = 82,
+	[0][0][1][0][RTW89_KCC][38] = 70,
+	[0][0][1][0][RTW89_ACMA][38] = 78,
+	[0][0][1][0][RTW89_CN][38] = 78,
+	[0][0][1][0][RTW89_UK][38] = 58,
+	[0][0][1][0][RTW89_FCC][40] = 82,
+	[0][0][1][0][RTW89_ETSI][40] = 30,
+	[0][0][1][0][RTW89_MKK][40] = 127,
+	[0][0][1][0][RTW89_IC][40] = 82,
+	[0][0][1][0][RTW89_KCC][40] = 76,
+	[0][0][1][0][RTW89_ACMA][40] = 78,
+	[0][0][1][0][RTW89_CN][40] = 78,
+	[0][0][1][0][RTW89_UK][40] = 58,
+	[0][0][1][0][RTW89_FCC][42] = 82,
+	[0][0][1][0][RTW89_ETSI][42] = 30,
+	[0][0][1][0][RTW89_MKK][42] = 127,
+	[0][0][1][0][RTW89_IC][42] = 82,
+	[0][0][1][0][RTW89_KCC][42] = 76,
+	[0][0][1][0][RTW89_ACMA][42] = 78,
+	[0][0][1][0][RTW89_CN][42] = 78,
+	[0][0][1][0][RTW89_UK][42] = 58,
+	[0][0][1][0][RTW89_FCC][44] = 82,
+	[0][0][1][0][RTW89_ETSI][44] = 30,
+	[0][0][1][0][RTW89_MKK][44] = 127,
+	[0][0][1][0][RTW89_IC][44] = 82,
+	[0][0][1][0][RTW89_KCC][44] = 76,
+	[0][0][1][0][RTW89_ACMA][44] = 78,
+	[0][0][1][0][RTW89_CN][44] = 78,
+	[0][0][1][0][RTW89_UK][44] = 58,
+	[0][0][1][0][RTW89_FCC][46] = 82,
+	[0][0][1][0][RTW89_ETSI][46] = 30,
+	[0][0][1][0][RTW89_MKK][46] = 127,
+	[0][0][1][0][RTW89_IC][46] = 82,
+	[0][0][1][0][RTW89_KCC][46] = 76,
+	[0][0][1][0][RTW89_ACMA][46] = 78,
+	[0][0][1][0][RTW89_CN][46] = 78,
+	[0][0][1][0][RTW89_UK][46] = 58,
+	[0][0][1][0][RTW89_FCC][48] = 72,
+	[0][0][1][0][RTW89_ETSI][48] = 127,
+	[0][0][1][0][RTW89_MKK][48] = 127,
+	[0][0][1][0][RTW89_IC][48] = 127,
+	[0][0][1][0][RTW89_KCC][48] = 127,
+	[0][0][1][0][RTW89_ACMA][48] = 127,
+	[0][0][1][0][RTW89_CN][48] = 127,
+	[0][0][1][0][RTW89_UK][48] = 127,
+	[0][0][1][0][RTW89_FCC][50] = 72,
+	[0][0][1][0][RTW89_ETSI][50] = 127,
+	[0][0][1][0][RTW89_MKK][50] = 127,
+	[0][0][1][0][RTW89_IC][50] = 127,
+	[0][0][1][0][RTW89_KCC][50] = 127,
+	[0][0][1][0][RTW89_ACMA][50] = 127,
+	[0][0][1][0][RTW89_CN][50] = 127,
+	[0][0][1][0][RTW89_UK][50] = 127,
+	[0][0][1][0][RTW89_FCC][52] = 72,
+	[0][0][1][0][RTW89_ETSI][52] = 127,
+	[0][0][1][0][RTW89_MKK][52] = 127,
+	[0][0][1][0][RTW89_IC][52] = 127,
+	[0][0][1][0][RTW89_KCC][52] = 127,
+	[0][0][1][0][RTW89_ACMA][52] = 127,
+	[0][0][1][0][RTW89_CN][52] = 127,
+	[0][0][1][0][RTW89_UK][52] = 127,
+	[0][1][1][0][RTW89_FCC][0] = 127,
+	[0][1][1][0][RTW89_ETSI][0] = 127,
+	[0][1][1][0][RTW89_MKK][0] = 127,
+	[0][1][1][0][RTW89_IC][0] = 127,
+	[0][1][1][0][RTW89_KCC][0] = 127,
+	[0][1][1][0][RTW89_ACMA][0] = 127,
+	[0][1][1][0][RTW89_CN][0] = 127,
+	[0][1][1][0][RTW89_UK][0] = 127,
+	[0][1][1][0][RTW89_FCC][2] = 127,
+	[0][1][1][0][RTW89_ETSI][2] = 127,
+	[0][1][1][0][RTW89_MKK][2] = 127,
+	[0][1][1][0][RTW89_IC][2] = 127,
+	[0][1][1][0][RTW89_KCC][2] = 127,
+	[0][1][1][0][RTW89_ACMA][2] = 127,
+	[0][1][1][0][RTW89_CN][2] = 127,
+	[0][1][1][0][RTW89_UK][2] = 127,
+	[0][1][1][0][RTW89_FCC][4] = 127,
+	[0][1][1][0][RTW89_ETSI][4] = 127,
+	[0][1][1][0][RTW89_MKK][4] = 127,
+	[0][1][1][0][RTW89_IC][4] = 127,
+	[0][1][1][0][RTW89_KCC][4] = 127,
+	[0][1][1][0][RTW89_ACMA][4] = 127,
+	[0][1][1][0][RTW89_CN][4] = 127,
+	[0][1][1][0][RTW89_UK][4] = 127,
+	[0][1][1][0][RTW89_FCC][6] = 127,
+	[0][1][1][0][RTW89_ETSI][6] = 127,
+	[0][1][1][0][RTW89_MKK][6] = 127,
+	[0][1][1][0][RTW89_IC][6] = 127,
+	[0][1][1][0][RTW89_KCC][6] = 127,
+	[0][1][1][0][RTW89_ACMA][6] = 127,
+	[0][1][1][0][RTW89_CN][6] = 127,
+	[0][1][1][0][RTW89_UK][6] = 127,
+	[0][1][1][0][RTW89_FCC][8] = 127,
+	[0][1][1][0][RTW89_ETSI][8] = 127,
+	[0][1][1][0][RTW89_MKK][8] = 127,
+	[0][1][1][0][RTW89_IC][8] = 127,
+	[0][1][1][0][RTW89_KCC][8] = 127,
+	[0][1][1][0][RTW89_ACMA][8] = 127,
+	[0][1][1][0][RTW89_CN][8] = 127,
+	[0][1][1][0][RTW89_UK][8] = 127,
+	[0][1][1][0][RTW89_FCC][10] = 127,
+	[0][1][1][0][RTW89_ETSI][10] = 127,
+	[0][1][1][0][RTW89_MKK][10] = 127,
+	[0][1][1][0][RTW89_IC][10] = 127,
+	[0][1][1][0][RTW89_KCC][10] = 127,
+	[0][1][1][0][RTW89_ACMA][10] = 127,
+	[0][1][1][0][RTW89_CN][10] = 127,
+	[0][1][1][0][RTW89_UK][10] = 127,
+	[0][1][1][0][RTW89_FCC][12] = 127,
+	[0][1][1][0][RTW89_ETSI][12] = 127,
+	[0][1][1][0][RTW89_MKK][12] = 127,
+	[0][1][1][0][RTW89_IC][12] = 127,
+	[0][1][1][0][RTW89_KCC][12] = 127,
+	[0][1][1][0][RTW89_ACMA][12] = 127,
+	[0][1][1][0][RTW89_CN][12] = 127,
+	[0][1][1][0][RTW89_UK][12] = 127,
+	[0][1][1][0][RTW89_FCC][14] = 127,
+	[0][1][1][0][RTW89_ETSI][14] = 127,
+	[0][1][1][0][RTW89_MKK][14] = 127,
+	[0][1][1][0][RTW89_IC][14] = 127,
+	[0][1][1][0][RTW89_KCC][14] = 127,
+	[0][1][1][0][RTW89_ACMA][14] = 127,
+	[0][1][1][0][RTW89_CN][14] = 127,
+	[0][1][1][0][RTW89_UK][14] = 127,
+	[0][1][1][0][RTW89_FCC][15] = 127,
+	[0][1][1][0][RTW89_ETSI][15] = 127,
+	[0][1][1][0][RTW89_MKK][15] = 127,
+	[0][1][1][0][RTW89_IC][15] = 127,
+	[0][1][1][0][RTW89_KCC][15] = 127,
+	[0][1][1][0][RTW89_ACMA][15] = 127,
+	[0][1][1][0][RTW89_CN][15] = 127,
+	[0][1][1][0][RTW89_UK][15] = 127,
+	[0][1][1][0][RTW89_FCC][17] = 127,
+	[0][1][1][0][RTW89_ETSI][17] = 127,
+	[0][1][1][0][RTW89_MKK][17] = 127,
+	[0][1][1][0][RTW89_IC][17] = 127,
+	[0][1][1][0][RTW89_KCC][17] = 127,
+	[0][1][1][0][RTW89_ACMA][17] = 127,
+	[0][1][1][0][RTW89_CN][17] = 127,
+	[0][1][1][0][RTW89_UK][17] = 127,
+	[0][1][1][0][RTW89_FCC][19] = 127,
+	[0][1][1][0][RTW89_ETSI][19] = 127,
+	[0][1][1][0][RTW89_MKK][19] = 127,
+	[0][1][1][0][RTW89_IC][19] = 127,
+	[0][1][1][0][RTW89_KCC][19] = 127,
+	[0][1][1][0][RTW89_ACMA][19] = 127,
+	[0][1][1][0][RTW89_CN][19] = 127,
+	[0][1][1][0][RTW89_UK][19] = 127,
+	[0][1][1][0][RTW89_FCC][21] = 127,
+	[0][1][1][0][RTW89_ETSI][21] = 127,
+	[0][1][1][0][RTW89_MKK][21] = 127,
+	[0][1][1][0][RTW89_IC][21] = 127,
+	[0][1][1][0][RTW89_KCC][21] = 127,
+	[0][1][1][0][RTW89_ACMA][21] = 127,
+	[0][1][1][0][RTW89_CN][21] = 127,
+	[0][1][1][0][RTW89_UK][21] = 127,
+	[0][1][1][0][RTW89_FCC][23] = 127,
+	[0][1][1][0][RTW89_ETSI][23] = 127,
+	[0][1][1][0][RTW89_MKK][23] = 127,
+	[0][1][1][0][RTW89_IC][23] = 127,
+	[0][1][1][0][RTW89_KCC][23] = 127,
+	[0][1][1][0][RTW89_ACMA][23] = 127,
+	[0][1][1][0][RTW89_CN][23] = 127,
+	[0][1][1][0][RTW89_UK][23] = 127,
+	[0][1][1][0][RTW89_FCC][25] = 127,
+	[0][1][1][0][RTW89_ETSI][25] = 127,
+	[0][1][1][0][RTW89_MKK][25] = 127,
+	[0][1][1][0][RTW89_IC][25] = 127,
+	[0][1][1][0][RTW89_KCC][25] = 127,
+	[0][1][1][0][RTW89_ACMA][25] = 127,
+	[0][1][1][0][RTW89_CN][25] = 127,
+	[0][1][1][0][RTW89_UK][25] = 127,
+	[0][1][1][0][RTW89_FCC][27] = 127,
+	[0][1][1][0][RTW89_ETSI][27] = 127,
+	[0][1][1][0][RTW89_MKK][27] = 127,
+	[0][1][1][0][RTW89_IC][27] = 127,
+	[0][1][1][0][RTW89_KCC][27] = 127,
+	[0][1][1][0][RTW89_ACMA][27] = 127,
+	[0][1][1][0][RTW89_CN][27] = 127,
+	[0][1][1][0][RTW89_UK][27] = 127,
+	[0][1][1][0][RTW89_FCC][29] = 127,
+	[0][1][1][0][RTW89_ETSI][29] = 127,
+	[0][1][1][0][RTW89_MKK][29] = 127,
+	[0][1][1][0][RTW89_IC][29] = 127,
+	[0][1][1][0][RTW89_KCC][29] = 127,
+	[0][1][1][0][RTW89_ACMA][29] = 127,
+	[0][1][1][0][RTW89_CN][29] = 127,
+	[0][1][1][0][RTW89_UK][29] = 127,
+	[0][1][1][0][RTW89_FCC][31] = 127,
+	[0][1][1][0][RTW89_ETSI][31] = 127,
+	[0][1][1][0][RTW89_MKK][31] = 127,
+	[0][1][1][0][RTW89_IC][31] = 127,
+	[0][1][1][0][RTW89_KCC][31] = 127,
+	[0][1][1][0][RTW89_ACMA][31] = 127,
+	[0][1][1][0][RTW89_CN][31] = 127,
+	[0][1][1][0][RTW89_UK][31] = 127,
+	[0][1][1][0][RTW89_FCC][33] = 127,
+	[0][1][1][0][RTW89_ETSI][33] = 127,
+	[0][1][1][0][RTW89_MKK][33] = 127,
+	[0][1][1][0][RTW89_IC][33] = 127,
+	[0][1][1][0][RTW89_KCC][33] = 127,
+	[0][1][1][0][RTW89_ACMA][33] = 127,
+	[0][1][1][0][RTW89_CN][33] = 127,
+	[0][1][1][0][RTW89_UK][33] = 127,
+	[0][1][1][0][RTW89_FCC][35] = 127,
+	[0][1][1][0][RTW89_ETSI][35] = 127,
+	[0][1][1][0][RTW89_MKK][35] = 127,
+	[0][1][1][0][RTW89_IC][35] = 127,
+	[0][1][1][0][RTW89_KCC][35] = 127,
+	[0][1][1][0][RTW89_ACMA][35] = 127,
+	[0][1][1][0][RTW89_CN][35] = 127,
+	[0][1][1][0][RTW89_UK][35] = 127,
+	[0][1][1][0][RTW89_FCC][37] = 127,
+	[0][1][1][0][RTW89_ETSI][37] = 127,
+	[0][1][1][0][RTW89_MKK][37] = 127,
+	[0][1][1][0][RTW89_IC][37] = 127,
+	[0][1][1][0][RTW89_KCC][37] = 127,
+	[0][1][1][0][RTW89_ACMA][37] = 127,
+	[0][1][1][0][RTW89_CN][37] = 127,
+	[0][1][1][0][RTW89_UK][37] = 127,
+	[0][1][1][0][RTW89_FCC][38] = 127,
+	[0][1][1][0][RTW89_ETSI][38] = 127,
+	[0][1][1][0][RTW89_MKK][38] = 127,
+	[0][1][1][0][RTW89_IC][38] = 127,
+	[0][1][1][0][RTW89_KCC][38] = 127,
+	[0][1][1][0][RTW89_ACMA][38] = 127,
+	[0][1][1][0][RTW89_CN][38] = 127,
+	[0][1][1][0][RTW89_UK][38] = 127,
+	[0][1][1][0][RTW89_FCC][40] = 127,
+	[0][1][1][0][RTW89_ETSI][40] = 127,
+	[0][1][1][0][RTW89_MKK][40] = 127,
+	[0][1][1][0][RTW89_IC][40] = 127,
+	[0][1][1][0][RTW89_KCC][40] = 127,
+	[0][1][1][0][RTW89_ACMA][40] = 127,
+	[0][1][1][0][RTW89_CN][40] = 127,
+	[0][1][1][0][RTW89_UK][40] = 127,
+	[0][1][1][0][RTW89_FCC][42] = 127,
+	[0][1][1][0][RTW89_ETSI][42] = 127,
+	[0][1][1][0][RTW89_MKK][42] = 127,
+	[0][1][1][0][RTW89_IC][42] = 127,
+	[0][1][1][0][RTW89_KCC][42] = 127,
+	[0][1][1][0][RTW89_ACMA][42] = 127,
+	[0][1][1][0][RTW89_CN][42] = 127,
+	[0][1][1][0][RTW89_UK][42] = 127,
+	[0][1][1][0][RTW89_FCC][44] = 127,
+	[0][1][1][0][RTW89_ETSI][44] = 127,
+	[0][1][1][0][RTW89_MKK][44] = 127,
+	[0][1][1][0][RTW89_IC][44] = 127,
+	[0][1][1][0][RTW89_KCC][44] = 127,
+	[0][1][1][0][RTW89_ACMA][44] = 127,
+	[0][1][1][0][RTW89_CN][44] = 127,
+	[0][1][1][0][RTW89_UK][44] = 127,
+	[0][1][1][0][RTW89_FCC][46] = 127,
+	[0][1][1][0][RTW89_ETSI][46] = 127,
+	[0][1][1][0][RTW89_MKK][46] = 127,
+	[0][1][1][0][RTW89_IC][46] = 127,
+	[0][1][1][0][RTW89_KCC][46] = 127,
+	[0][1][1][0][RTW89_ACMA][46] = 127,
+	[0][1][1][0][RTW89_CN][46] = 127,
+	[0][1][1][0][RTW89_UK][46] = 127,
+	[0][1][1][0][RTW89_FCC][48] = 127,
+	[0][1][1][0][RTW89_ETSI][48] = 127,
+	[0][1][1][0][RTW89_MKK][48] = 127,
+	[0][1][1][0][RTW89_IC][48] = 127,
+	[0][1][1][0][RTW89_KCC][48] = 127,
+	[0][1][1][0][RTW89_ACMA][48] = 127,
+	[0][1][1][0][RTW89_CN][48] = 127,
+	[0][1][1][0][RTW89_UK][48] = 127,
+	[0][1][1][0][RTW89_FCC][50] = 127,
+	[0][1][1][0][RTW89_ETSI][50] = 127,
+	[0][1][1][0][RTW89_MKK][50] = 127,
+	[0][1][1][0][RTW89_IC][50] = 127,
+	[0][1][1][0][RTW89_KCC][50] = 127,
+	[0][1][1][0][RTW89_ACMA][50] = 127,
+	[0][1][1][0][RTW89_CN][50] = 127,
+	[0][1][1][0][RTW89_UK][50] = 127,
+	[0][1][1][0][RTW89_FCC][52] = 127,
+	[0][1][1][0][RTW89_ETSI][52] = 127,
+	[0][1][1][0][RTW89_MKK][52] = 127,
+	[0][1][1][0][RTW89_IC][52] = 127,
+	[0][1][1][0][RTW89_KCC][52] = 127,
+	[0][1][1][0][RTW89_ACMA][52] = 127,
+	[0][1][1][0][RTW89_CN][52] = 127,
+	[0][1][1][0][RTW89_UK][52] = 127,
+	[0][0][2][0][RTW89_FCC][0] = 78,
+	[0][0][2][0][RTW89_ETSI][0] = 62,
+	[0][0][2][0][RTW89_MKK][0] = 62,
+	[0][0][2][0][RTW89_IC][0] = 64,
+	[0][0][2][0][RTW89_KCC][0] = 76,
+	[0][0][2][0][RTW89_ACMA][0] = 62,
+	[0][0][2][0][RTW89_CN][0] = 62,
+	[0][0][2][0][RTW89_UK][0] = 62,
+	[0][0][2][0][RTW89_FCC][2] = 82,
+	[0][0][2][0][RTW89_ETSI][2] = 62,
+	[0][0][2][0][RTW89_MKK][2] = 62,
+	[0][0][2][0][RTW89_IC][2] = 64,
+	[0][0][2][0][RTW89_KCC][2] = 76,
+	[0][0][2][0][RTW89_ACMA][2] = 62,
+	[0][0][2][0][RTW89_CN][2] = 62,
+	[0][0][2][0][RTW89_UK][2] = 62,
+	[0][0][2][0][RTW89_FCC][4] = 82,
+	[0][0][2][0][RTW89_ETSI][4] = 62,
+	[0][0][2][0][RTW89_MKK][4] = 62,
+	[0][0][2][0][RTW89_IC][4] = 64,
+	[0][0][2][0][RTW89_KCC][4] = 76,
+	[0][0][2][0][RTW89_ACMA][4] = 62,
+	[0][0][2][0][RTW89_CN][4] = 62,
+	[0][0][2][0][RTW89_UK][4] = 62,
+	[0][0][2][0][RTW89_FCC][6] = 82,
+	[0][0][2][0][RTW89_ETSI][6] = 62,
+	[0][0][2][0][RTW89_MKK][6] = 62,
+	[0][0][2][0][RTW89_IC][6] = 64,
+	[0][0][2][0][RTW89_KCC][6] = 54,
+	[0][0][2][0][RTW89_ACMA][6] = 62,
+	[0][0][2][0][RTW89_CN][6] = 62,
+	[0][0][2][0][RTW89_UK][6] = 62,
+	[0][0][2][0][RTW89_FCC][8] = 82,
+	[0][0][2][0][RTW89_ETSI][8] = 62,
+	[0][0][2][0][RTW89_MKK][8] = 62,
+	[0][0][2][0][RTW89_IC][8] = 64,
+	[0][0][2][0][RTW89_KCC][8] = 76,
+	[0][0][2][0][RTW89_ACMA][8] = 62,
+	[0][0][2][0][RTW89_CN][8] = 62,
+	[0][0][2][0][RTW89_UK][8] = 62,
+	[0][0][2][0][RTW89_FCC][10] = 82,
+	[0][0][2][0][RTW89_ETSI][10] = 62,
+	[0][0][2][0][RTW89_MKK][10] = 62,
+	[0][0][2][0][RTW89_IC][10] = 64,
+	[0][0][2][0][RTW89_KCC][10] = 76,
+	[0][0][2][0][RTW89_ACMA][10] = 62,
+	[0][0][2][0][RTW89_CN][10] = 62,
+	[0][0][2][0][RTW89_UK][10] = 62,
+	[0][0][2][0][RTW89_FCC][12] = 82,
+	[0][0][2][0][RTW89_ETSI][12] = 62,
+	[0][0][2][0][RTW89_MKK][12] = 62,
+	[0][0][2][0][RTW89_IC][12] = 64,
+	[0][0][2][0][RTW89_KCC][12] = 78,
+	[0][0][2][0][RTW89_ACMA][12] = 62,
+	[0][0][2][0][RTW89_CN][12] = 62,
+	[0][0][2][0][RTW89_UK][12] = 62,
+	[0][0][2][0][RTW89_FCC][14] = 76,
+	[0][0][2][0][RTW89_ETSI][14] = 62,
+	[0][0][2][0][RTW89_MKK][14] = 62,
+	[0][0][2][0][RTW89_IC][14] = 64,
+	[0][0][2][0][RTW89_KCC][14] = 78,
+	[0][0][2][0][RTW89_ACMA][14] = 62,
+	[0][0][2][0][RTW89_CN][14] = 62,
+	[0][0][2][0][RTW89_UK][14] = 62,
+	[0][0][2][0][RTW89_FCC][15] = 76,
+	[0][0][2][0][RTW89_ETSI][15] = 60,
+	[0][0][2][0][RTW89_MKK][15] = 78,
+	[0][0][2][0][RTW89_IC][15] = 76,
+	[0][0][2][0][RTW89_KCC][15] = 78,
+	[0][0][2][0][RTW89_ACMA][15] = 60,
+	[0][0][2][0][RTW89_CN][15] = 127,
+	[0][0][2][0][RTW89_UK][15] = 60,
+	[0][0][2][0][RTW89_FCC][17] = 82,
+	[0][0][2][0][RTW89_ETSI][17] = 62,
+	[0][0][2][0][RTW89_MKK][17] = 78,
+	[0][0][2][0][RTW89_IC][17] = 82,
+	[0][0][2][0][RTW89_KCC][17] = 78,
+	[0][0][2][0][RTW89_ACMA][17] = 62,
+	[0][0][2][0][RTW89_CN][17] = 127,
+	[0][0][2][0][RTW89_UK][17] = 62,
+	[0][0][2][0][RTW89_FCC][19] = 82,
+	[0][0][2][0][RTW89_ETSI][19] = 62,
+	[0][0][2][0][RTW89_MKK][19] = 78,
+	[0][0][2][0][RTW89_IC][19] = 82,
+	[0][0][2][0][RTW89_KCC][19] = 78,
+	[0][0][2][0][RTW89_ACMA][19] = 62,
+	[0][0][2][0][RTW89_CN][19] = 127,
+	[0][0][2][0][RTW89_UK][19] = 62,
+	[0][0][2][0][RTW89_FCC][21] = 82,
+	[0][0][2][0][RTW89_ETSI][21] = 62,
+	[0][0][2][0][RTW89_MKK][21] = 78,
+	[0][0][2][0][RTW89_IC][21] = 82,
+	[0][0][2][0][RTW89_KCC][21] = 78,
+	[0][0][2][0][RTW89_ACMA][21] = 62,
+	[0][0][2][0][RTW89_CN][21] = 127,
+	[0][0][2][0][RTW89_UK][21] = 62,
+	[0][0][2][0][RTW89_FCC][23] = 82,
+	[0][0][2][0][RTW89_ETSI][23] = 62,
+	[0][0][2][0][RTW89_MKK][23] = 78,
+	[0][0][2][0][RTW89_IC][23] = 82,
+	[0][0][2][0][RTW89_KCC][23] = 78,
+	[0][0][2][0][RTW89_ACMA][23] = 62,
+	[0][0][2][0][RTW89_CN][23] = 127,
+	[0][0][2][0][RTW89_UK][23] = 62,
+	[0][0][2][0][RTW89_FCC][25] = 82,
+	[0][0][2][0][RTW89_ETSI][25] = 62,
+	[0][0][2][0][RTW89_MKK][25] = 78,
+	[0][0][2][0][RTW89_IC][25] = 127,
+	[0][0][2][0][RTW89_KCC][25] = 78,
+	[0][0][2][0][RTW89_ACMA][25] = 127,
+	[0][0][2][0][RTW89_CN][25] = 127,
+	[0][0][2][0][RTW89_UK][25] = 62,
+	[0][0][2][0][RTW89_FCC][27] = 82,
+	[0][0][2][0][RTW89_ETSI][27] = 62,
+	[0][0][2][0][RTW89_MKK][27] = 78,
+	[0][0][2][0][RTW89_IC][27] = 127,
+	[0][0][2][0][RTW89_KCC][27] = 78,
+	[0][0][2][0][RTW89_ACMA][27] = 127,
+	[0][0][2][0][RTW89_CN][27] = 127,
+	[0][0][2][0][RTW89_UK][27] = 62,
+	[0][0][2][0][RTW89_FCC][29] = 82,
+	[0][0][2][0][RTW89_ETSI][29] = 62,
+	[0][0][2][0][RTW89_MKK][29] = 78,
+	[0][0][2][0][RTW89_IC][29] = 127,
+	[0][0][2][0][RTW89_KCC][29] = 78,
+	[0][0][2][0][RTW89_ACMA][29] = 127,
+	[0][0][2][0][RTW89_CN][29] = 127,
+	[0][0][2][0][RTW89_UK][29] = 62,
+	[0][0][2][0][RTW89_FCC][31] = 82,
+	[0][0][2][0][RTW89_ETSI][31] = 62,
+	[0][0][2][0][RTW89_MKK][31] = 78,
+	[0][0][2][0][RTW89_IC][31] = 82,
+	[0][0][2][0][RTW89_KCC][31] = 74,
+	[0][0][2][0][RTW89_ACMA][31] = 62,
+	[0][0][2][0][RTW89_CN][31] = 127,
+	[0][0][2][0][RTW89_UK][31] = 62,
+	[0][0][2][0][RTW89_FCC][33] = 82,
+	[0][0][2][0][RTW89_ETSI][33] = 62,
+	[0][0][2][0][RTW89_MKK][33] = 78,
+	[0][0][2][0][RTW89_IC][33] = 82,
+	[0][0][2][0][RTW89_KCC][33] = 74,
+	[0][0][2][0][RTW89_ACMA][33] = 62,
+	[0][0][2][0][RTW89_CN][33] = 127,
+	[0][0][2][0][RTW89_UK][33] = 62,
+	[0][0][2][0][RTW89_FCC][35] = 72,
+	[0][0][2][0][RTW89_ETSI][35] = 62,
+	[0][0][2][0][RTW89_MKK][35] = 78,
+	[0][0][2][0][RTW89_IC][35] = 72,
+	[0][0][2][0][RTW89_KCC][35] = 74,
+	[0][0][2][0][RTW89_ACMA][35] = 62,
+	[0][0][2][0][RTW89_CN][35] = 127,
+	[0][0][2][0][RTW89_UK][35] = 62,
+	[0][0][2][0][RTW89_FCC][37] = 82,
+	[0][0][2][0][RTW89_ETSI][37] = 127,
+	[0][0][2][0][RTW89_MKK][37] = 78,
+	[0][0][2][0][RTW89_IC][37] = 82,
+	[0][0][2][0][RTW89_KCC][37] = 76,
+	[0][0][2][0][RTW89_ACMA][37] = 78,
+	[0][0][2][0][RTW89_CN][37] = 127,
+	[0][0][2][0][RTW89_UK][37] = 78,
+	[0][0][2][0][RTW89_FCC][38] = 82,
+	[0][0][2][0][RTW89_ETSI][38] = 30,
+	[0][0][2][0][RTW89_MKK][38] = 127,
+	[0][0][2][0][RTW89_IC][38] = 82,
+	[0][0][2][0][RTW89_KCC][38] = 66,
+	[0][0][2][0][RTW89_ACMA][38] = 78,
+	[0][0][2][0][RTW89_CN][38] = 78,
+	[0][0][2][0][RTW89_UK][38] = 60,
+	[0][0][2][0][RTW89_FCC][40] = 82,
+	[0][0][2][0][RTW89_ETSI][40] = 30,
+	[0][0][2][0][RTW89_MKK][40] = 127,
+	[0][0][2][0][RTW89_IC][40] = 82,
+	[0][0][2][0][RTW89_KCC][40] = 74,
+	[0][0][2][0][RTW89_ACMA][40] = 78,
+	[0][0][2][0][RTW89_CN][40] = 78,
+	[0][0][2][0][RTW89_UK][40] = 60,
+	[0][0][2][0][RTW89_FCC][42] = 82,
+	[0][0][2][0][RTW89_ETSI][42] = 30,
+	[0][0][2][0][RTW89_MKK][42] = 127,
+	[0][0][2][0][RTW89_IC][42] = 82,
+	[0][0][2][0][RTW89_KCC][42] = 74,
+	[0][0][2][0][RTW89_ACMA][42] = 78,
+	[0][0][2][0][RTW89_CN][42] = 78,
+	[0][0][2][0][RTW89_UK][42] = 60,
+	[0][0][2][0][RTW89_FCC][44] = 82,
+	[0][0][2][0][RTW89_ETSI][44] = 30,
+	[0][0][2][0][RTW89_MKK][44] = 127,
+	[0][0][2][0][RTW89_IC][44] = 82,
+	[0][0][2][0][RTW89_KCC][44] = 74,
+	[0][0][2][0][RTW89_ACMA][44] = 78,
+	[0][0][2][0][RTW89_CN][44] = 78,
+	[0][0][2][0][RTW89_UK][44] = 60,
+	[0][0][2][0][RTW89_FCC][46] = 82,
+	[0][0][2][0][RTW89_ETSI][46] = 30,
+	[0][0][2][0][RTW89_MKK][46] = 127,
+	[0][0][2][0][RTW89_IC][46] = 82,
+	[0][0][2][0][RTW89_KCC][46] = 74,
+	[0][0][2][0][RTW89_ACMA][46] = 78,
+	[0][0][2][0][RTW89_CN][46] = 78,
+	[0][0][2][0][RTW89_UK][46] = 60,
+	[0][0][2][0][RTW89_FCC][48] = 74,
+	[0][0][2][0][RTW89_ETSI][48] = 127,
+	[0][0][2][0][RTW89_MKK][48] = 127,
+	[0][0][2][0][RTW89_IC][48] = 127,
+	[0][0][2][0][RTW89_KCC][48] = 127,
+	[0][0][2][0][RTW89_ACMA][48] = 127,
+	[0][0][2][0][RTW89_CN][48] = 127,
+	[0][0][2][0][RTW89_UK][48] = 127,
+	[0][0][2][0][RTW89_FCC][50] = 76,
+	[0][0][2][0][RTW89_ETSI][50] = 127,
+	[0][0][2][0][RTW89_MKK][50] = 127,
+	[0][0][2][0][RTW89_IC][50] = 127,
+	[0][0][2][0][RTW89_KCC][50] = 127,
+	[0][0][2][0][RTW89_ACMA][50] = 127,
+	[0][0][2][0][RTW89_CN][50] = 127,
+	[0][0][2][0][RTW89_UK][50] = 127,
+	[0][0][2][0][RTW89_FCC][52] = 76,
+	[0][0][2][0][RTW89_ETSI][52] = 127,
+	[0][0][2][0][RTW89_MKK][52] = 127,
+	[0][0][2][0][RTW89_IC][52] = 127,
+	[0][0][2][0][RTW89_KCC][52] = 127,
+	[0][0][2][0][RTW89_ACMA][52] = 127,
+	[0][0][2][0][RTW89_CN][52] = 127,
+	[0][0][2][0][RTW89_UK][52] = 127,
+	[0][1][2][0][RTW89_FCC][0] = 127,
+	[0][1][2][0][RTW89_ETSI][0] = 127,
+	[0][1][2][0][RTW89_MKK][0] = 127,
+	[0][1][2][0][RTW89_IC][0] = 127,
+	[0][1][2][0][RTW89_KCC][0] = 127,
+	[0][1][2][0][RTW89_ACMA][0] = 127,
+	[0][1][2][0][RTW89_CN][0] = 127,
+	[0][1][2][0][RTW89_UK][0] = 127,
+	[0][1][2][0][RTW89_FCC][2] = 127,
+	[0][1][2][0][RTW89_ETSI][2] = 127,
+	[0][1][2][0][RTW89_MKK][2] = 127,
+	[0][1][2][0][RTW89_IC][2] = 127,
+	[0][1][2][0][RTW89_KCC][2] = 127,
+	[0][1][2][0][RTW89_ACMA][2] = 127,
+	[0][1][2][0][RTW89_CN][2] = 127,
+	[0][1][2][0][RTW89_UK][2] = 127,
+	[0][1][2][0][RTW89_FCC][4] = 127,
+	[0][1][2][0][RTW89_ETSI][4] = 127,
+	[0][1][2][0][RTW89_MKK][4] = 127,
+	[0][1][2][0][RTW89_IC][4] = 127,
+	[0][1][2][0][RTW89_KCC][4] = 127,
+	[0][1][2][0][RTW89_ACMA][4] = 127,
+	[0][1][2][0][RTW89_CN][4] = 127,
+	[0][1][2][0][RTW89_UK][4] = 127,
+	[0][1][2][0][RTW89_FCC][6] = 127,
+	[0][1][2][0][RTW89_ETSI][6] = 127,
+	[0][1][2][0][RTW89_MKK][6] = 127,
+	[0][1][2][0][RTW89_IC][6] = 127,
+	[0][1][2][0][RTW89_KCC][6] = 127,
+	[0][1][2][0][RTW89_ACMA][6] = 127,
+	[0][1][2][0][RTW89_CN][6] = 127,
+	[0][1][2][0][RTW89_UK][6] = 127,
+	[0][1][2][0][RTW89_FCC][8] = 127,
+	[0][1][2][0][RTW89_ETSI][8] = 127,
+	[0][1][2][0][RTW89_MKK][8] = 127,
+	[0][1][2][0][RTW89_IC][8] = 127,
+	[0][1][2][0][RTW89_KCC][8] = 127,
+	[0][1][2][0][RTW89_ACMA][8] = 127,
+	[0][1][2][0][RTW89_CN][8] = 127,
+	[0][1][2][0][RTW89_UK][8] = 127,
+	[0][1][2][0][RTW89_FCC][10] = 127,
+	[0][1][2][0][RTW89_ETSI][10] = 127,
+	[0][1][2][0][RTW89_MKK][10] = 127,
+	[0][1][2][0][RTW89_IC][10] = 127,
+	[0][1][2][0][RTW89_KCC][10] = 127,
+	[0][1][2][0][RTW89_ACMA][10] = 127,
+	[0][1][2][0][RTW89_CN][10] = 127,
+	[0][1][2][0][RTW89_UK][10] = 127,
+	[0][1][2][0][RTW89_FCC][12] = 127,
+	[0][1][2][0][RTW89_ETSI][12] = 127,
+	[0][1][2][0][RTW89_MKK][12] = 127,
+	[0][1][2][0][RTW89_IC][12] = 127,
+	[0][1][2][0][RTW89_KCC][12] = 127,
+	[0][1][2][0][RTW89_ACMA][12] = 127,
+	[0][1][2][0][RTW89_CN][12] = 127,
+	[0][1][2][0][RTW89_UK][12] = 127,
+	[0][1][2][0][RTW89_FCC][14] = 127,
+	[0][1][2][0][RTW89_ETSI][14] = 127,
+	[0][1][2][0][RTW89_MKK][14] = 127,
+	[0][1][2][0][RTW89_IC][14] = 127,
+	[0][1][2][0][RTW89_KCC][14] = 127,
+	[0][1][2][0][RTW89_ACMA][14] = 127,
+	[0][1][2][0][RTW89_CN][14] = 127,
+	[0][1][2][0][RTW89_UK][14] = 127,
+	[0][1][2][0][RTW89_FCC][15] = 127,
+	[0][1][2][0][RTW89_ETSI][15] = 127,
+	[0][1][2][0][RTW89_MKK][15] = 127,
+	[0][1][2][0][RTW89_IC][15] = 127,
+	[0][1][2][0][RTW89_KCC][15] = 127,
+	[0][1][2][0][RTW89_ACMA][15] = 127,
+	[0][1][2][0][RTW89_CN][15] = 127,
+	[0][1][2][0][RTW89_UK][15] = 127,
+	[0][1][2][0][RTW89_FCC][17] = 127,
+	[0][1][2][0][RTW89_ETSI][17] = 127,
+	[0][1][2][0][RTW89_MKK][17] = 127,
+	[0][1][2][0][RTW89_IC][17] = 127,
+	[0][1][2][0][RTW89_KCC][17] = 127,
+	[0][1][2][0][RTW89_ACMA][17] = 127,
+	[0][1][2][0][RTW89_CN][17] = 127,
+	[0][1][2][0][RTW89_UK][17] = 127,
+	[0][1][2][0][RTW89_FCC][19] = 127,
+	[0][1][2][0][RTW89_ETSI][19] = 127,
+	[0][1][2][0][RTW89_MKK][19] = 127,
+	[0][1][2][0][RTW89_IC][19] = 127,
+	[0][1][2][0][RTW89_KCC][19] = 127,
+	[0][1][2][0][RTW89_ACMA][19] = 127,
+	[0][1][2][0][RTW89_CN][19] = 127,
+	[0][1][2][0][RTW89_UK][19] = 127,
+	[0][1][2][0][RTW89_FCC][21] = 127,
+	[0][1][2][0][RTW89_ETSI][21] = 127,
+	[0][1][2][0][RTW89_MKK][21] = 127,
+	[0][1][2][0][RTW89_IC][21] = 127,
+	[0][1][2][0][RTW89_KCC][21] = 127,
+	[0][1][2][0][RTW89_ACMA][21] = 127,
+	[0][1][2][0][RTW89_CN][21] = 127,
+	[0][1][2][0][RTW89_UK][21] = 127,
+	[0][1][2][0][RTW89_FCC][23] = 127,
+	[0][1][2][0][RTW89_ETSI][23] = 127,
+	[0][1][2][0][RTW89_MKK][23] = 127,
+	[0][1][2][0][RTW89_IC][23] = 127,
+	[0][1][2][0][RTW89_KCC][23] = 127,
+	[0][1][2][0][RTW89_ACMA][23] = 127,
+	[0][1][2][0][RTW89_CN][23] = 127,
+	[0][1][2][0][RTW89_UK][23] = 127,
+	[0][1][2][0][RTW89_FCC][25] = 127,
+	[0][1][2][0][RTW89_ETSI][25] = 127,
+	[0][1][2][0][RTW89_MKK][25] = 127,
+	[0][1][2][0][RTW89_IC][25] = 127,
+	[0][1][2][0][RTW89_KCC][25] = 127,
+	[0][1][2][0][RTW89_ACMA][25] = 127,
+	[0][1][2][0][RTW89_CN][25] = 127,
+	[0][1][2][0][RTW89_UK][25] = 127,
+	[0][1][2][0][RTW89_FCC][27] = 127,
+	[0][1][2][0][RTW89_ETSI][27] = 127,
+	[0][1][2][0][RTW89_MKK][27] = 127,
+	[0][1][2][0][RTW89_IC][27] = 127,
+	[0][1][2][0][RTW89_KCC][27] = 127,
+	[0][1][2][0][RTW89_ACMA][27] = 127,
+	[0][1][2][0][RTW89_CN][27] = 127,
+	[0][1][2][0][RTW89_UK][27] = 127,
+	[0][1][2][0][RTW89_FCC][29] = 127,
+	[0][1][2][0][RTW89_ETSI][29] = 127,
+	[0][1][2][0][RTW89_MKK][29] = 127,
+	[0][1][2][0][RTW89_IC][29] = 127,
+	[0][1][2][0][RTW89_KCC][29] = 127,
+	[0][1][2][0][RTW89_ACMA][29] = 127,
+	[0][1][2][0][RTW89_CN][29] = 127,
+	[0][1][2][0][RTW89_UK][29] = 127,
+	[0][1][2][0][RTW89_FCC][31] = 127,
+	[0][1][2][0][RTW89_ETSI][31] = 127,
+	[0][1][2][0][RTW89_MKK][31] = 127,
+	[0][1][2][0][RTW89_IC][31] = 127,
+	[0][1][2][0][RTW89_KCC][31] = 127,
+	[0][1][2][0][RTW89_ACMA][31] = 127,
+	[0][1][2][0][RTW89_CN][31] = 127,
+	[0][1][2][0][RTW89_UK][31] = 127,
+	[0][1][2][0][RTW89_FCC][33] = 127,
+	[0][1][2][0][RTW89_ETSI][33] = 127,
+	[0][1][2][0][RTW89_MKK][33] = 127,
+	[0][1][2][0][RTW89_IC][33] = 127,
+	[0][1][2][0][RTW89_KCC][33] = 127,
+	[0][1][2][0][RTW89_ACMA][33] = 127,
+	[0][1][2][0][RTW89_CN][33] = 127,
+	[0][1][2][0][RTW89_UK][33] = 127,
+	[0][1][2][0][RTW89_FCC][35] = 127,
+	[0][1][2][0][RTW89_ETSI][35] = 127,
+	[0][1][2][0][RTW89_MKK][35] = 127,
+	[0][1][2][0][RTW89_IC][35] = 127,
+	[0][1][2][0][RTW89_KCC][35] = 127,
+	[0][1][2][0][RTW89_ACMA][35] = 127,
+	[0][1][2][0][RTW89_CN][35] = 127,
+	[0][1][2][0][RTW89_UK][35] = 127,
+	[0][1][2][0][RTW89_FCC][37] = 127,
+	[0][1][2][0][RTW89_ETSI][37] = 127,
+	[0][1][2][0][RTW89_MKK][37] = 127,
+	[0][1][2][0][RTW89_IC][37] = 127,
+	[0][1][2][0][RTW89_KCC][37] = 127,
+	[0][1][2][0][RTW89_ACMA][37] = 127,
+	[0][1][2][0][RTW89_CN][37] = 127,
+	[0][1][2][0][RTW89_UK][37] = 127,
+	[0][1][2][0][RTW89_FCC][38] = 127,
+	[0][1][2][0][RTW89_ETSI][38] = 127,
+	[0][1][2][0][RTW89_MKK][38] = 127,
+	[0][1][2][0][RTW89_IC][38] = 127,
+	[0][1][2][0][RTW89_KCC][38] = 127,
+	[0][1][2][0][RTW89_ACMA][38] = 127,
+	[0][1][2][0][RTW89_CN][38] = 127,
+	[0][1][2][0][RTW89_UK][38] = 127,
+	[0][1][2][0][RTW89_FCC][40] = 127,
+	[0][1][2][0][RTW89_ETSI][40] = 127,
+	[0][1][2][0][RTW89_MKK][40] = 127,
+	[0][1][2][0][RTW89_IC][40] = 127,
+	[0][1][2][0][RTW89_KCC][40] = 127,
+	[0][1][2][0][RTW89_ACMA][40] = 127,
+	[0][1][2][0][RTW89_CN][40] = 127,
+	[0][1][2][0][RTW89_UK][40] = 127,
+	[0][1][2][0][RTW89_FCC][42] = 127,
+	[0][1][2][0][RTW89_ETSI][42] = 127,
+	[0][1][2][0][RTW89_MKK][42] = 127,
+	[0][1][2][0][RTW89_IC][42] = 127,
+	[0][1][2][0][RTW89_KCC][42] = 127,
+	[0][1][2][0][RTW89_ACMA][42] = 127,
+	[0][1][2][0][RTW89_CN][42] = 127,
+	[0][1][2][0][RTW89_UK][42] = 127,
+	[0][1][2][0][RTW89_FCC][44] = 127,
+	[0][1][2][0][RTW89_ETSI][44] = 127,
+	[0][1][2][0][RTW89_MKK][44] = 127,
+	[0][1][2][0][RTW89_IC][44] = 127,
+	[0][1][2][0][RTW89_KCC][44] = 127,
+	[0][1][2][0][RTW89_ACMA][44] = 127,
+	[0][1][2][0][RTW89_CN][44] = 127,
+	[0][1][2][0][RTW89_UK][44] = 127,
+	[0][1][2][0][RTW89_FCC][46] = 127,
+	[0][1][2][0][RTW89_ETSI][46] = 127,
+	[0][1][2][0][RTW89_MKK][46] = 127,
+	[0][1][2][0][RTW89_IC][46] = 127,
+	[0][1][2][0][RTW89_KCC][46] = 127,
+	[0][1][2][0][RTW89_ACMA][46] = 127,
+	[0][1][2][0][RTW89_CN][46] = 127,
+	[0][1][2][0][RTW89_UK][46] = 127,
+	[0][1][2][0][RTW89_FCC][48] = 127,
+	[0][1][2][0][RTW89_ETSI][48] = 127,
+	[0][1][2][0][RTW89_MKK][48] = 127,
+	[0][1][2][0][RTW89_IC][48] = 127,
+	[0][1][2][0][RTW89_KCC][48] = 127,
+	[0][1][2][0][RTW89_ACMA][48] = 127,
+	[0][1][2][0][RTW89_CN][48] = 127,
+	[0][1][2][0][RTW89_UK][48] = 127,
+	[0][1][2][0][RTW89_FCC][50] = 127,
+	[0][1][2][0][RTW89_ETSI][50] = 127,
+	[0][1][2][0][RTW89_MKK][50] = 127,
+	[0][1][2][0][RTW89_IC][50] = 127,
+	[0][1][2][0][RTW89_KCC][50] = 127,
+	[0][1][2][0][RTW89_ACMA][50] = 127,
+	[0][1][2][0][RTW89_CN][50] = 127,
+	[0][1][2][0][RTW89_UK][50] = 127,
+	[0][1][2][0][RTW89_FCC][52] = 127,
+	[0][1][2][0][RTW89_ETSI][52] = 127,
+	[0][1][2][0][RTW89_MKK][52] = 127,
+	[0][1][2][0][RTW89_IC][52] = 127,
+	[0][1][2][0][RTW89_KCC][52] = 127,
+	[0][1][2][0][RTW89_ACMA][52] = 127,
+	[0][1][2][0][RTW89_CN][52] = 127,
+	[0][1][2][0][RTW89_UK][52] = 127,
+	[0][1][2][1][RTW89_FCC][0] = 127,
+	[0][1][2][1][RTW89_ETSI][0] = 127,
+	[0][1][2][1][RTW89_MKK][0] = 127,
+	[0][1][2][1][RTW89_IC][0] = 127,
+	[0][1][2][1][RTW89_KCC][0] = 127,
+	[0][1][2][1][RTW89_ACMA][0] = 127,
+	[0][1][2][1][RTW89_CN][0] = 127,
+	[0][1][2][1][RTW89_UK][0] = 127,
+	[0][1][2][1][RTW89_FCC][2] = 127,
+	[0][1][2][1][RTW89_ETSI][2] = 127,
+	[0][1][2][1][RTW89_MKK][2] = 127,
+	[0][1][2][1][RTW89_IC][2] = 127,
+	[0][1][2][1][RTW89_KCC][2] = 127,
+	[0][1][2][1][RTW89_ACMA][2] = 127,
+	[0][1][2][1][RTW89_CN][2] = 127,
+	[0][1][2][1][RTW89_UK][2] = 127,
+	[0][1][2][1][RTW89_FCC][4] = 127,
+	[0][1][2][1][RTW89_ETSI][4] = 127,
+	[0][1][2][1][RTW89_MKK][4] = 127,
+	[0][1][2][1][RTW89_IC][4] = 127,
+	[0][1][2][1][RTW89_KCC][4] = 127,
+	[0][1][2][1][RTW89_ACMA][4] = 127,
+	[0][1][2][1][RTW89_CN][4] = 127,
+	[0][1][2][1][RTW89_UK][4] = 127,
+	[0][1][2][1][RTW89_FCC][6] = 127,
+	[0][1][2][1][RTW89_ETSI][6] = 127,
+	[0][1][2][1][RTW89_MKK][6] = 127,
+	[0][1][2][1][RTW89_IC][6] = 127,
+	[0][1][2][1][RTW89_KCC][6] = 127,
+	[0][1][2][1][RTW89_ACMA][6] = 127,
+	[0][1][2][1][RTW89_CN][6] = 127,
+	[0][1][2][1][RTW89_UK][6] = 127,
+	[0][1][2][1][RTW89_FCC][8] = 127,
+	[0][1][2][1][RTW89_ETSI][8] = 127,
+	[0][1][2][1][RTW89_MKK][8] = 127,
+	[0][1][2][1][RTW89_IC][8] = 127,
+	[0][1][2][1][RTW89_KCC][8] = 127,
+	[0][1][2][1][RTW89_ACMA][8] = 127,
+	[0][1][2][1][RTW89_CN][8] = 127,
+	[0][1][2][1][RTW89_UK][8] = 127,
+	[0][1][2][1][RTW89_FCC][10] = 127,
+	[0][1][2][1][RTW89_ETSI][10] = 127,
+	[0][1][2][1][RTW89_MKK][10] = 127,
+	[0][1][2][1][RTW89_IC][10] = 127,
+	[0][1][2][1][RTW89_KCC][10] = 127,
+	[0][1][2][1][RTW89_ACMA][10] = 127,
+	[0][1][2][1][RTW89_CN][10] = 127,
+	[0][1][2][1][RTW89_UK][10] = 127,
+	[0][1][2][1][RTW89_FCC][12] = 127,
+	[0][1][2][1][RTW89_ETSI][12] = 127,
+	[0][1][2][1][RTW89_MKK][12] = 127,
+	[0][1][2][1][RTW89_IC][12] = 127,
+	[0][1][2][1][RTW89_KCC][12] = 127,
+	[0][1][2][1][RTW89_ACMA][12] = 127,
+	[0][1][2][1][RTW89_CN][12] = 127,
+	[0][1][2][1][RTW89_UK][12] = 127,
+	[0][1][2][1][RTW89_FCC][14] = 127,
+	[0][1][2][1][RTW89_ETSI][14] = 127,
+	[0][1][2][1][RTW89_MKK][14] = 127,
+	[0][1][2][1][RTW89_IC][14] = 127,
+	[0][1][2][1][RTW89_KCC][14] = 127,
+	[0][1][2][1][RTW89_ACMA][14] = 127,
+	[0][1][2][1][RTW89_CN][14] = 127,
+	[0][1][2][1][RTW89_UK][14] = 127,
+	[0][1][2][1][RTW89_FCC][15] = 127,
+	[0][1][2][1][RTW89_ETSI][15] = 127,
+	[0][1][2][1][RTW89_MKK][15] = 127,
+	[0][1][2][1][RTW89_IC][15] = 127,
+	[0][1][2][1][RTW89_KCC][15] = 127,
+	[0][1][2][1][RTW89_ACMA][15] = 127,
+	[0][1][2][1][RTW89_CN][15] = 127,
+	[0][1][2][1][RTW89_UK][15] = 127,
+	[0][1][2][1][RTW89_FCC][17] = 127,
+	[0][1][2][1][RTW89_ETSI][17] = 127,
+	[0][1][2][1][RTW89_MKK][17] = 127,
+	[0][1][2][1][RTW89_IC][17] = 127,
+	[0][1][2][1][RTW89_KCC][17] = 127,
+	[0][1][2][1][RTW89_ACMA][17] = 127,
+	[0][1][2][1][RTW89_CN][17] = 127,
+	[0][1][2][1][RTW89_UK][17] = 127,
+	[0][1][2][1][RTW89_FCC][19] = 127,
+	[0][1][2][1][RTW89_ETSI][19] = 127,
+	[0][1][2][1][RTW89_MKK][19] = 127,
+	[0][1][2][1][RTW89_IC][19] = 127,
+	[0][1][2][1][RTW89_KCC][19] = 127,
+	[0][1][2][1][RTW89_ACMA][19] = 127,
+	[0][1][2][1][RTW89_CN][19] = 127,
+	[0][1][2][1][RTW89_UK][19] = 127,
+	[0][1][2][1][RTW89_FCC][21] = 127,
+	[0][1][2][1][RTW89_ETSI][21] = 127,
+	[0][1][2][1][RTW89_MKK][21] = 127,
+	[0][1][2][1][RTW89_IC][21] = 127,
+	[0][1][2][1][RTW89_KCC][21] = 127,
+	[0][1][2][1][RTW89_ACMA][21] = 127,
+	[0][1][2][1][RTW89_CN][21] = 127,
+	[0][1][2][1][RTW89_UK][21] = 127,
+	[0][1][2][1][RTW89_FCC][23] = 127,
+	[0][1][2][1][RTW89_ETSI][23] = 127,
+	[0][1][2][1][RTW89_MKK][23] = 127,
+	[0][1][2][1][RTW89_IC][23] = 127,
+	[0][1][2][1][RTW89_KCC][23] = 127,
+	[0][1][2][1][RTW89_ACMA][23] = 127,
+	[0][1][2][1][RTW89_CN][23] = 127,
+	[0][1][2][1][RTW89_UK][23] = 127,
+	[0][1][2][1][RTW89_FCC][25] = 127,
+	[0][1][2][1][RTW89_ETSI][25] = 127,
+	[0][1][2][1][RTW89_MKK][25] = 127,
+	[0][1][2][1][RTW89_IC][25] = 127,
+	[0][1][2][1][RTW89_KCC][25] = 127,
+	[0][1][2][1][RTW89_ACMA][25] = 127,
+	[0][1][2][1][RTW89_CN][25] = 127,
+	[0][1][2][1][RTW89_UK][25] = 127,
+	[0][1][2][1][RTW89_FCC][27] = 127,
+	[0][1][2][1][RTW89_ETSI][27] = 127,
+	[0][1][2][1][RTW89_MKK][27] = 127,
+	[0][1][2][1][RTW89_IC][27] = 127,
+	[0][1][2][1][RTW89_KCC][27] = 127,
+	[0][1][2][1][RTW89_ACMA][27] = 127,
+	[0][1][2][1][RTW89_CN][27] = 127,
+	[0][1][2][1][RTW89_UK][27] = 127,
+	[0][1][2][1][RTW89_FCC][29] = 127,
+	[0][1][2][1][RTW89_ETSI][29] = 127,
+	[0][1][2][1][RTW89_MKK][29] = 127,
+	[0][1][2][1][RTW89_IC][29] = 127,
+	[0][1][2][1][RTW89_KCC][29] = 127,
+	[0][1][2][1][RTW89_ACMA][29] = 127,
+	[0][1][2][1][RTW89_CN][29] = 127,
+	[0][1][2][1][RTW89_UK][29] = 127,
+	[0][1][2][1][RTW89_FCC][31] = 127,
+	[0][1][2][1][RTW89_ETSI][31] = 127,
+	[0][1][2][1][RTW89_MKK][31] = 127,
+	[0][1][2][1][RTW89_IC][31] = 127,
+	[0][1][2][1][RTW89_KCC][31] = 127,
+	[0][1][2][1][RTW89_ACMA][31] = 127,
+	[0][1][2][1][RTW89_CN][31] = 127,
+	[0][1][2][1][RTW89_UK][31] = 127,
+	[0][1][2][1][RTW89_FCC][33] = 127,
+	[0][1][2][1][RTW89_ETSI][33] = 127,
+	[0][1][2][1][RTW89_MKK][33] = 127,
+	[0][1][2][1][RTW89_IC][33] = 127,
+	[0][1][2][1][RTW89_KCC][33] = 127,
+	[0][1][2][1][RTW89_ACMA][33] = 127,
+	[0][1][2][1][RTW89_CN][33] = 127,
+	[0][1][2][1][RTW89_UK][33] = 127,
+	[0][1][2][1][RTW89_FCC][35] = 127,
+	[0][1][2][1][RTW89_ETSI][35] = 127,
+	[0][1][2][1][RTW89_MKK][35] = 127,
+	[0][1][2][1][RTW89_IC][35] = 127,
+	[0][1][2][1][RTW89_KCC][35] = 127,
+	[0][1][2][1][RTW89_ACMA][35] = 127,
+	[0][1][2][1][RTW89_CN][35] = 127,
+	[0][1][2][1][RTW89_UK][35] = 127,
+	[0][1][2][1][RTW89_FCC][37] = 127,
+	[0][1][2][1][RTW89_ETSI][37] = 127,
+	[0][1][2][1][RTW89_MKK][37] = 127,
+	[0][1][2][1][RTW89_IC][37] = 127,
+	[0][1][2][1][RTW89_KCC][37] = 127,
+	[0][1][2][1][RTW89_ACMA][37] = 127,
+	[0][1][2][1][RTW89_CN][37] = 127,
+	[0][1][2][1][RTW89_UK][37] = 127,
+	[0][1][2][1][RTW89_FCC][38] = 127,
+	[0][1][2][1][RTW89_ETSI][38] = 127,
+	[0][1][2][1][RTW89_MKK][38] = 127,
+	[0][1][2][1][RTW89_IC][38] = 127,
+	[0][1][2][1][RTW89_KCC][38] = 127,
+	[0][1][2][1][RTW89_ACMA][38] = 127,
+	[0][1][2][1][RTW89_CN][38] = 127,
+	[0][1][2][1][RTW89_UK][38] = 127,
+	[0][1][2][1][RTW89_FCC][40] = 127,
+	[0][1][2][1][RTW89_ETSI][40] = 127,
+	[0][1][2][1][RTW89_MKK][40] = 127,
+	[0][1][2][1][RTW89_IC][40] = 127,
+	[0][1][2][1][RTW89_KCC][40] = 127,
+	[0][1][2][1][RTW89_ACMA][40] = 127,
+	[0][1][2][1][RTW89_CN][40] = 127,
+	[0][1][2][1][RTW89_UK][40] = 127,
+	[0][1][2][1][RTW89_FCC][42] = 127,
+	[0][1][2][1][RTW89_ETSI][42] = 127,
+	[0][1][2][1][RTW89_MKK][42] = 127,
+	[0][1][2][1][RTW89_IC][42] = 127,
+	[0][1][2][1][RTW89_KCC][42] = 127,
+	[0][1][2][1][RTW89_ACMA][42] = 127,
+	[0][1][2][1][RTW89_CN][42] = 127,
+	[0][1][2][1][RTW89_UK][42] = 127,
+	[0][1][2][1][RTW89_FCC][44] = 127,
+	[0][1][2][1][RTW89_ETSI][44] = 127,
+	[0][1][2][1][RTW89_MKK][44] = 127,
+	[0][1][2][1][RTW89_IC][44] = 127,
+	[0][1][2][1][RTW89_KCC][44] = 127,
+	[0][1][2][1][RTW89_ACMA][44] = 127,
+	[0][1][2][1][RTW89_CN][44] = 127,
+	[0][1][2][1][RTW89_UK][44] = 127,
+	[0][1][2][1][RTW89_FCC][46] = 127,
+	[0][1][2][1][RTW89_ETSI][46] = 127,
+	[0][1][2][1][RTW89_MKK][46] = 127,
+	[0][1][2][1][RTW89_IC][46] = 127,
+	[0][1][2][1][RTW89_KCC][46] = 127,
+	[0][1][2][1][RTW89_ACMA][46] = 127,
+	[0][1][2][1][RTW89_CN][46] = 127,
+	[0][1][2][1][RTW89_UK][46] = 127,
+	[0][1][2][1][RTW89_FCC][48] = 127,
+	[0][1][2][1][RTW89_ETSI][48] = 127,
+	[0][1][2][1][RTW89_MKK][48] = 127,
+	[0][1][2][1][RTW89_IC][48] = 127,
+	[0][1][2][1][RTW89_KCC][48] = 127,
+	[0][1][2][1][RTW89_ACMA][48] = 127,
+	[0][1][2][1][RTW89_CN][48] = 127,
+	[0][1][2][1][RTW89_UK][48] = 127,
+	[0][1][2][1][RTW89_FCC][50] = 127,
+	[0][1][2][1][RTW89_ETSI][50] = 127,
+	[0][1][2][1][RTW89_MKK][50] = 127,
+	[0][1][2][1][RTW89_IC][50] = 127,
+	[0][1][2][1][RTW89_KCC][50] = 127,
+	[0][1][2][1][RTW89_ACMA][50] = 127,
+	[0][1][2][1][RTW89_CN][50] = 127,
+	[0][1][2][1][RTW89_UK][50] = 127,
+	[0][1][2][1][RTW89_FCC][52] = 127,
+	[0][1][2][1][RTW89_ETSI][52] = 127,
+	[0][1][2][1][RTW89_MKK][52] = 127,
+	[0][1][2][1][RTW89_IC][52] = 127,
+	[0][1][2][1][RTW89_KCC][52] = 127,
+	[0][1][2][1][RTW89_ACMA][52] = 127,
+	[0][1][2][1][RTW89_CN][52] = 127,
+	[0][1][2][1][RTW89_UK][52] = 127,
+	[1][0][2][0][RTW89_FCC][1] = 68,
+	[1][0][2][0][RTW89_ETSI][1] = 64,
+	[1][0][2][0][RTW89_MKK][1] = 64,
+	[1][0][2][0][RTW89_IC][1] = 64,
+	[1][0][2][0][RTW89_KCC][1] = 74,
+	[1][0][2][0][RTW89_ACMA][1] = 64,
+	[1][0][2][0][RTW89_CN][1] = 64,
+	[1][0][2][0][RTW89_UK][1] = 64,
+	[1][0][2][0][RTW89_FCC][5] = 82,
+	[1][0][2][0][RTW89_ETSI][5] = 64,
+	[1][0][2][0][RTW89_MKK][5] = 62,
+	[1][0][2][0][RTW89_IC][5] = 64,
+	[1][0][2][0][RTW89_KCC][5] = 66,
+	[1][0][2][0][RTW89_ACMA][5] = 64,
+	[1][0][2][0][RTW89_CN][5] = 64,
+	[1][0][2][0][RTW89_UK][5] = 64,
+	[1][0][2][0][RTW89_FCC][9] = 82,
+	[1][0][2][0][RTW89_ETSI][9] = 64,
+	[1][0][2][0][RTW89_MKK][9] = 64,
+	[1][0][2][0][RTW89_IC][9] = 64,
+	[1][0][2][0][RTW89_KCC][9] = 78,
+	[1][0][2][0][RTW89_ACMA][9] = 64,
+	[1][0][2][0][RTW89_CN][9] = 64,
+	[1][0][2][0][RTW89_UK][9] = 64,
+	[1][0][2][0][RTW89_FCC][13] = 66,
+	[1][0][2][0][RTW89_ETSI][13] = 64,
+	[1][0][2][0][RTW89_MKK][13] = 64,
+	[1][0][2][0][RTW89_IC][13] = 64,
+	[1][0][2][0][RTW89_KCC][13] = 72,
+	[1][0][2][0][RTW89_ACMA][13] = 64,
+	[1][0][2][0][RTW89_CN][13] = 64,
+	[1][0][2][0][RTW89_UK][13] = 64,
+	[1][0][2][0][RTW89_FCC][16] = 66,
+	[1][0][2][0][RTW89_ETSI][16] = 66,
+	[1][0][2][0][RTW89_MKK][16] = 80,
+	[1][0][2][0][RTW89_IC][16] = 66,
+	[1][0][2][0][RTW89_KCC][16] = 74,
+	[1][0][2][0][RTW89_ACMA][16] = 66,
+	[1][0][2][0][RTW89_CN][16] = 127,
+	[1][0][2][0][RTW89_UK][16] = 66,
+	[1][0][2][0][RTW89_FCC][20] = 80,
+	[1][0][2][0][RTW89_ETSI][20] = 66,
+	[1][0][2][0][RTW89_MKK][20] = 80,
+	[1][0][2][0][RTW89_IC][20] = 80,
+	[1][0][2][0][RTW89_KCC][20] = 74,
+	[1][0][2][0][RTW89_ACMA][20] = 66,
+	[1][0][2][0][RTW89_CN][20] = 127,
+	[1][0][2][0][RTW89_UK][20] = 66,
+	[1][0][2][0][RTW89_FCC][24] = 80,
+	[1][0][2][0][RTW89_ETSI][24] = 66,
+	[1][0][2][0][RTW89_MKK][24] = 80,
+	[1][0][2][0][RTW89_IC][24] = 127,
+	[1][0][2][0][RTW89_KCC][24] = 74,
+	[1][0][2][0][RTW89_ACMA][24] = 127,
+	[1][0][2][0][RTW89_CN][24] = 127,
+	[1][0][2][0][RTW89_UK][24] = 66,
+	[1][0][2][0][RTW89_FCC][28] = 80,
+	[1][0][2][0][RTW89_ETSI][28] = 66,
+	[1][0][2][0][RTW89_MKK][28] = 80,
+	[1][0][2][0][RTW89_IC][28] = 127,
+	[1][0][2][0][RTW89_KCC][28] = 74,
+	[1][0][2][0][RTW89_ACMA][28] = 127,
+	[1][0][2][0][RTW89_CN][28] = 127,
+	[1][0][2][0][RTW89_UK][28] = 66,
+	[1][0][2][0][RTW89_FCC][32] = 76,
+	[1][0][2][0][RTW89_ETSI][32] = 66,
+	[1][0][2][0][RTW89_MKK][32] = 80,
+	[1][0][2][0][RTW89_IC][32] = 76,
+	[1][0][2][0][RTW89_KCC][32] = 78,
+	[1][0][2][0][RTW89_ACMA][32] = 66,
+	[1][0][2][0][RTW89_CN][32] = 127,
+	[1][0][2][0][RTW89_UK][32] = 66,
+	[1][0][2][0][RTW89_FCC][36] = 80,
+	[1][0][2][0][RTW89_ETSI][36] = 127,
+	[1][0][2][0][RTW89_MKK][36] = 80,
+	[1][0][2][0][RTW89_IC][36] = 80,
+	[1][0][2][0][RTW89_KCC][36] = 76,
+	[1][0][2][0][RTW89_ACMA][36] = 78,
+	[1][0][2][0][RTW89_CN][36] = 127,
+	[1][0][2][0][RTW89_UK][36] = 80,
+	[1][0][2][0][RTW89_FCC][39] = 84,
+	[1][0][2][0][RTW89_ETSI][39] = 30,
+	[1][0][2][0][RTW89_MKK][39] = 127,
+	[1][0][2][0][RTW89_IC][39] = 84,
+	[1][0][2][0][RTW89_KCC][39] = 68,
+	[1][0][2][0][RTW89_ACMA][39] = 80,
+	[1][0][2][0][RTW89_CN][39] = 70,
+	[1][0][2][0][RTW89_UK][39] = 64,
+	[1][0][2][0][RTW89_FCC][43] = 84,
+	[1][0][2][0][RTW89_ETSI][43] = 30,
+	[1][0][2][0][RTW89_MKK][43] = 127,
+	[1][0][2][0][RTW89_IC][43] = 84,
+	[1][0][2][0][RTW89_KCC][43] = 78,
+	[1][0][2][0][RTW89_ACMA][43] = 80,
+	[1][0][2][0][RTW89_CN][43] = 80,
+	[1][0][2][0][RTW89_UK][43] = 64,
+	[1][0][2][0][RTW89_FCC][47] = 84,
+	[1][0][2][0][RTW89_ETSI][47] = 127,
+	[1][0][2][0][RTW89_MKK][47] = 127,
+	[1][0][2][0][RTW89_IC][47] = 127,
+	[1][0][2][0][RTW89_KCC][47] = 127,
+	[1][0][2][0][RTW89_ACMA][47] = 127,
+	[1][0][2][0][RTW89_CN][47] = 127,
+	[1][0][2][0][RTW89_UK][47] = 127,
+	[1][0][2][0][RTW89_FCC][51] = 84,
+	[1][0][2][0][RTW89_ETSI][51] = 127,
+	[1][0][2][0][RTW89_MKK][51] = 127,
+	[1][0][2][0][RTW89_IC][51] = 127,
+	[1][0][2][0][RTW89_KCC][51] = 127,
+	[1][0][2][0][RTW89_ACMA][51] = 127,
+	[1][0][2][0][RTW89_CN][51] = 127,
+	[1][0][2][0][RTW89_UK][51] = 127,
+	[1][1][2][0][RTW89_FCC][1] = 127,
+	[1][1][2][0][RTW89_ETSI][1] = 127,
+	[1][1][2][0][RTW89_MKK][1] = 127,
+	[1][1][2][0][RTW89_IC][1] = 127,
+	[1][1][2][0][RTW89_KCC][1] = 127,
+	[1][1][2][0][RTW89_ACMA][1] = 127,
+	[1][1][2][0][RTW89_CN][1] = 127,
+	[1][1][2][0][RTW89_UK][1] = 127,
+	[1][1][2][0][RTW89_FCC][5] = 127,
+	[1][1][2][0][RTW89_ETSI][5] = 127,
+	[1][1][2][0][RTW89_MKK][5] = 127,
+	[1][1][2][0][RTW89_IC][5] = 127,
+	[1][1][2][0][RTW89_KCC][5] = 127,
+	[1][1][2][0][RTW89_ACMA][5] = 127,
+	[1][1][2][0][RTW89_CN][5] = 127,
+	[1][1][2][0][RTW89_UK][5] = 127,
+	[1][1][2][0][RTW89_FCC][9] = 127,
+	[1][1][2][0][RTW89_ETSI][9] = 127,
+	[1][1][2][0][RTW89_MKK][9] = 127,
+	[1][1][2][0][RTW89_IC][9] = 127,
+	[1][1][2][0][RTW89_KCC][9] = 127,
+	[1][1][2][0][RTW89_ACMA][9] = 127,
+	[1][1][2][0][RTW89_CN][9] = 127,
+	[1][1][2][0][RTW89_UK][9] = 127,
+	[1][1][2][0][RTW89_FCC][13] = 127,
+	[1][1][2][0][RTW89_ETSI][13] = 127,
+	[1][1][2][0][RTW89_MKK][13] = 127,
+	[1][1][2][0][RTW89_IC][13] = 127,
+	[1][1][2][0][RTW89_KCC][13] = 127,
+	[1][1][2][0][RTW89_ACMA][13] = 127,
+	[1][1][2][0][RTW89_CN][13] = 127,
+	[1][1][2][0][RTW89_UK][13] = 127,
+	[1][1][2][0][RTW89_FCC][16] = 127,
+	[1][1][2][0][RTW89_ETSI][16] = 127,
+	[1][1][2][0][RTW89_MKK][16] = 127,
+	[1][1][2][0][RTW89_IC][16] = 127,
+	[1][1][2][0][RTW89_KCC][16] = 127,
+	[1][1][2][0][RTW89_ACMA][16] = 127,
+	[1][1][2][0][RTW89_CN][16] = 127,
+	[1][1][2][0][RTW89_UK][16] = 127,
+	[1][1][2][0][RTW89_FCC][20] = 127,
+	[1][1][2][0][RTW89_ETSI][20] = 127,
+	[1][1][2][0][RTW89_MKK][20] = 127,
+	[1][1][2][0][RTW89_IC][20] = 127,
+	[1][1][2][0][RTW89_KCC][20] = 127,
+	[1][1][2][0][RTW89_ACMA][20] = 127,
+	[1][1][2][0][RTW89_CN][20] = 127,
+	[1][1][2][0][RTW89_UK][20] = 127,
+	[1][1][2][0][RTW89_FCC][24] = 127,
+	[1][1][2][0][RTW89_ETSI][24] = 127,
+	[1][1][2][0][RTW89_MKK][24] = 127,
+	[1][1][2][0][RTW89_IC][24] = 127,
+	[1][1][2][0][RTW89_KCC][24] = 127,
+	[1][1][2][0][RTW89_ACMA][24] = 127,
+	[1][1][2][0][RTW89_CN][24] = 127,
+	[1][1][2][0][RTW89_UK][24] = 127,
+	[1][1][2][0][RTW89_FCC][28] = 127,
+	[1][1][2][0][RTW89_ETSI][28] = 127,
+	[1][1][2][0][RTW89_MKK][28] = 127,
+	[1][1][2][0][RTW89_IC][28] = 127,
+	[1][1][2][0][RTW89_KCC][28] = 127,
+	[1][1][2][0][RTW89_ACMA][28] = 127,
+	[1][1][2][0][RTW89_CN][28] = 127,
+	[1][1][2][0][RTW89_UK][28] = 127,
+	[1][1][2][0][RTW89_FCC][32] = 127,
+	[1][1][2][0][RTW89_ETSI][32] = 127,
+	[1][1][2][0][RTW89_MKK][32] = 127,
+	[1][1][2][0][RTW89_IC][32] = 127,
+	[1][1][2][0][RTW89_KCC][32] = 127,
+	[1][1][2][0][RTW89_ACMA][32] = 127,
+	[1][1][2][0][RTW89_CN][32] = 127,
+	[1][1][2][0][RTW89_UK][32] = 127,
+	[1][1][2][0][RTW89_FCC][36] = 127,
+	[1][1][2][0][RTW89_ETSI][36] = 127,
+	[1][1][2][0][RTW89_MKK][36] = 127,
+	[1][1][2][0][RTW89_IC][36] = 127,
+	[1][1][2][0][RTW89_KCC][36] = 127,
+	[1][1][2][0][RTW89_ACMA][36] = 127,
+	[1][1][2][0][RTW89_CN][36] = 127,
+	[1][1][2][0][RTW89_UK][36] = 127,
+	[1][1][2][0][RTW89_FCC][39] = 127,
+	[1][1][2][0][RTW89_ETSI][39] = 127,
+	[1][1][2][0][RTW89_MKK][39] = 127,
+	[1][1][2][0][RTW89_IC][39] = 127,
+	[1][1][2][0][RTW89_KCC][39] = 127,
+	[1][1][2][0][RTW89_ACMA][39] = 127,
+	[1][1][2][0][RTW89_CN][39] = 127,
+	[1][1][2][0][RTW89_UK][39] = 127,
+	[1][1][2][0][RTW89_FCC][43] = 127,
+	[1][1][2][0][RTW89_ETSI][43] = 127,
+	[1][1][2][0][RTW89_MKK][43] = 127,
+	[1][1][2][0][RTW89_IC][43] = 127,
+	[1][1][2][0][RTW89_KCC][43] = 127,
+	[1][1][2][0][RTW89_ACMA][43] = 127,
+	[1][1][2][0][RTW89_CN][43] = 127,
+	[1][1][2][0][RTW89_UK][43] = 127,
+	[1][1][2][0][RTW89_FCC][47] = 127,
+	[1][1][2][0][RTW89_ETSI][47] = 127,
+	[1][1][2][0][RTW89_MKK][47] = 127,
+	[1][1][2][0][RTW89_IC][47] = 127,
+	[1][1][2][0][RTW89_KCC][47] = 127,
+	[1][1][2][0][RTW89_ACMA][47] = 127,
+	[1][1][2][0][RTW89_CN][47] = 127,
+	[1][1][2][0][RTW89_UK][47] = 127,
+	[1][1][2][0][RTW89_FCC][51] = 127,
+	[1][1][2][0][RTW89_ETSI][51] = 127,
+	[1][1][2][0][RTW89_MKK][51] = 127,
+	[1][1][2][0][RTW89_IC][51] = 127,
+	[1][1][2][0][RTW89_KCC][51] = 127,
+	[1][1][2][0][RTW89_ACMA][51] = 127,
+	[1][1][2][0][RTW89_CN][51] = 127,
+	[1][1][2][0][RTW89_UK][51] = 127,
+	[1][1][2][1][RTW89_FCC][1] = 127,
+	[1][1][2][1][RTW89_ETSI][1] = 127,
+	[1][1][2][1][RTW89_MKK][1] = 127,
+	[1][1][2][1][RTW89_IC][1] = 127,
+	[1][1][2][1][RTW89_KCC][1] = 127,
+	[1][1][2][1][RTW89_ACMA][1] = 127,
+	[1][1][2][1][RTW89_CN][1] = 127,
+	[1][1][2][1][RTW89_UK][1] = 127,
+	[1][1][2][1][RTW89_FCC][5] = 127,
+	[1][1][2][1][RTW89_ETSI][5] = 127,
+	[1][1][2][1][RTW89_MKK][5] = 127,
+	[1][1][2][1][RTW89_IC][5] = 127,
+	[1][1][2][1][RTW89_KCC][5] = 127,
+	[1][1][2][1][RTW89_ACMA][5] = 127,
+	[1][1][2][1][RTW89_CN][5] = 127,
+	[1][1][2][1][RTW89_UK][5] = 127,
+	[1][1][2][1][RTW89_FCC][9] = 127,
+	[1][1][2][1][RTW89_ETSI][9] = 127,
+	[1][1][2][1][RTW89_MKK][9] = 127,
+	[1][1][2][1][RTW89_IC][9] = 127,
+	[1][1][2][1][RTW89_KCC][9] = 127,
+	[1][1][2][1][RTW89_ACMA][9] = 127,
+	[1][1][2][1][RTW89_CN][9] = 127,
+	[1][1][2][1][RTW89_UK][9] = 127,
+	[1][1][2][1][RTW89_FCC][13] = 127,
+	[1][1][2][1][RTW89_ETSI][13] = 127,
+	[1][1][2][1][RTW89_MKK][13] = 127,
+	[1][1][2][1][RTW89_IC][13] = 127,
+	[1][1][2][1][RTW89_KCC][13] = 127,
+	[1][1][2][1][RTW89_ACMA][13] = 127,
+	[1][1][2][1][RTW89_CN][13] = 127,
+	[1][1][2][1][RTW89_UK][13] = 127,
+	[1][1][2][1][RTW89_FCC][16] = 127,
+	[1][1][2][1][RTW89_ETSI][16] = 127,
+	[1][1][2][1][RTW89_MKK][16] = 127,
+	[1][1][2][1][RTW89_IC][16] = 127,
+	[1][1][2][1][RTW89_KCC][16] = 127,
+	[1][1][2][1][RTW89_ACMA][16] = 127,
+	[1][1][2][1][RTW89_CN][16] = 127,
+	[1][1][2][1][RTW89_UK][16] = 127,
+	[1][1][2][1][RTW89_FCC][20] = 127,
+	[1][1][2][1][RTW89_ETSI][20] = 127,
+	[1][1][2][1][RTW89_MKK][20] = 127,
+	[1][1][2][1][RTW89_IC][20] = 127,
+	[1][1][2][1][RTW89_KCC][20] = 127,
+	[1][1][2][1][RTW89_ACMA][20] = 127,
+	[1][1][2][1][RTW89_CN][20] = 127,
+	[1][1][2][1][RTW89_UK][20] = 127,
+	[1][1][2][1][RTW89_FCC][24] = 127,
+	[1][1][2][1][RTW89_ETSI][24] = 127,
+	[1][1][2][1][RTW89_MKK][24] = 127,
+	[1][1][2][1][RTW89_IC][24] = 127,
+	[1][1][2][1][RTW89_KCC][24] = 127,
+	[1][1][2][1][RTW89_ACMA][24] = 127,
+	[1][1][2][1][RTW89_CN][24] = 127,
+	[1][1][2][1][RTW89_UK][24] = 127,
+	[1][1][2][1][RTW89_FCC][28] = 127,
+	[1][1][2][1][RTW89_ETSI][28] = 127,
+	[1][1][2][1][RTW89_MKK][28] = 127,
+	[1][1][2][1][RTW89_IC][28] = 127,
+	[1][1][2][1][RTW89_KCC][28] = 127,
+	[1][1][2][1][RTW89_ACMA][28] = 127,
+	[1][1][2][1][RTW89_CN][28] = 127,
+	[1][1][2][1][RTW89_UK][28] = 127,
+	[1][1][2][1][RTW89_FCC][32] = 127,
+	[1][1][2][1][RTW89_ETSI][32] = 127,
+	[1][1][2][1][RTW89_MKK][32] = 127,
+	[1][1][2][1][RTW89_IC][32] = 127,
+	[1][1][2][1][RTW89_KCC][32] = 127,
+	[1][1][2][1][RTW89_ACMA][32] = 127,
+	[1][1][2][1][RTW89_CN][32] = 127,
+	[1][1][2][1][RTW89_UK][32] = 127,
+	[1][1][2][1][RTW89_FCC][36] = 127,
+	[1][1][2][1][RTW89_ETSI][36] = 127,
+	[1][1][2][1][RTW89_MKK][36] = 127,
+	[1][1][2][1][RTW89_IC][36] = 127,
+	[1][1][2][1][RTW89_KCC][36] = 127,
+	[1][1][2][1][RTW89_ACMA][36] = 127,
+	[1][1][2][1][RTW89_CN][36] = 127,
+	[1][1][2][1][RTW89_UK][36] = 127,
+	[1][1][2][1][RTW89_FCC][39] = 127,
+	[1][1][2][1][RTW89_ETSI][39] = 127,
+	[1][1][2][1][RTW89_MKK][39] = 127,
+	[1][1][2][1][RTW89_IC][39] = 127,
+	[1][1][2][1][RTW89_KCC][39] = 127,
+	[1][1][2][1][RTW89_ACMA][39] = 127,
+	[1][1][2][1][RTW89_CN][39] = 127,
+	[1][1][2][1][RTW89_UK][39] = 127,
+	[1][1][2][1][RTW89_FCC][43] = 127,
+	[1][1][2][1][RTW89_ETSI][43] = 127,
+	[1][1][2][1][RTW89_MKK][43] = 127,
+	[1][1][2][1][RTW89_IC][43] = 127,
+	[1][1][2][1][RTW89_KCC][43] = 127,
+	[1][1][2][1][RTW89_ACMA][43] = 127,
+	[1][1][2][1][RTW89_CN][43] = 127,
+	[1][1][2][1][RTW89_UK][43] = 127,
+	[1][1][2][1][RTW89_FCC][47] = 127,
+	[1][1][2][1][RTW89_ETSI][47] = 127,
+	[1][1][2][1][RTW89_MKK][47] = 127,
+	[1][1][2][1][RTW89_IC][47] = 127,
+	[1][1][2][1][RTW89_KCC][47] = 127,
+	[1][1][2][1][RTW89_ACMA][47] = 127,
+	[1][1][2][1][RTW89_CN][47] = 127,
+	[1][1][2][1][RTW89_UK][47] = 127,
+	[1][1][2][1][RTW89_FCC][51] = 127,
+	[1][1][2][1][RTW89_ETSI][51] = 127,
+	[1][1][2][1][RTW89_MKK][51] = 127,
+	[1][1][2][1][RTW89_IC][51] = 127,
+	[1][1][2][1][RTW89_KCC][51] = 127,
+	[1][1][2][1][RTW89_ACMA][51] = 127,
+	[1][1][2][1][RTW89_CN][51] = 127,
+	[1][1][2][1][RTW89_UK][51] = 127,
+	[2][0][2][0][RTW89_FCC][3] = 76,
+	[2][0][2][0][RTW89_ETSI][3] = 64,
+	[2][0][2][0][RTW89_MKK][3] = 62,
+	[2][0][2][0][RTW89_IC][3] = 64,
+	[2][0][2][0][RTW89_KCC][3] = 72,
+	[2][0][2][0][RTW89_ACMA][3] = 64,
+	[2][0][2][0][RTW89_CN][3] = 64,
+	[2][0][2][0][RTW89_UK][3] = 64,
+	[2][0][2][0][RTW89_FCC][11] = 64,
+	[2][0][2][0][RTW89_ETSI][11] = 64,
+	[2][0][2][0][RTW89_MKK][11] = 64,
+	[2][0][2][0][RTW89_IC][11] = 62,
+	[2][0][2][0][RTW89_KCC][11] = 72,
+	[2][0][2][0][RTW89_ACMA][11] = 64,
+	[2][0][2][0][RTW89_CN][11] = 64,
+	[2][0][2][0][RTW89_UK][11] = 64,
+	[2][0][2][0][RTW89_FCC][18] = 66,
+	[2][0][2][0][RTW89_ETSI][18] = 64,
+	[2][0][2][0][RTW89_MKK][18] = 72,
+	[2][0][2][0][RTW89_IC][18] = 66,
+	[2][0][2][0][RTW89_KCC][18] = 72,
+	[2][0][2][0][RTW89_ACMA][18] = 64,
+	[2][0][2][0][RTW89_CN][18] = 127,
+	[2][0][2][0][RTW89_UK][18] = 64,
+	[2][0][2][0][RTW89_FCC][26] = 76,
+	[2][0][2][0][RTW89_ETSI][26] = 64,
+	[2][0][2][0][RTW89_MKK][26] = 72,
+	[2][0][2][0][RTW89_IC][26] = 127,
+	[2][0][2][0][RTW89_KCC][26] = 72,
+	[2][0][2][0][RTW89_ACMA][26] = 127,
+	[2][0][2][0][RTW89_CN][26] = 127,
+	[2][0][2][0][RTW89_UK][26] = 64,
+	[2][0][2][0][RTW89_FCC][34] = 76,
+	[2][0][2][0][RTW89_ETSI][34] = 127,
+	[2][0][2][0][RTW89_MKK][34] = 72,
+	[2][0][2][0][RTW89_IC][34] = 76,
+	[2][0][2][0][RTW89_KCC][34] = 72,
+	[2][0][2][0][RTW89_ACMA][34] = 72,
+	[2][0][2][0][RTW89_CN][34] = 127,
+	[2][0][2][0][RTW89_UK][34] = 72,
+	[2][0][2][0][RTW89_FCC][41] = 76,
+	[2][0][2][0][RTW89_ETSI][41] = 30,
+	[2][0][2][0][RTW89_MKK][41] = 127,
+	[2][0][2][0][RTW89_IC][41] = 76,
+	[2][0][2][0][RTW89_KCC][41] = 64,
+	[2][0][2][0][RTW89_ACMA][41] = 72,
+	[2][0][2][0][RTW89_CN][41] = 72,
+	[2][0][2][0][RTW89_UK][41] = 64,
+	[2][0][2][0][RTW89_FCC][49] = 74,
+	[2][0][2][0][RTW89_ETSI][49] = 127,
+	[2][0][2][0][RTW89_MKK][49] = 127,
+	[2][0][2][0][RTW89_IC][49] = 127,
+	[2][0][2][0][RTW89_KCC][49] = 127,
+	[2][0][2][0][RTW89_ACMA][49] = 127,
+	[2][0][2][0][RTW89_CN][49] = 127,
+	[2][0][2][0][RTW89_UK][49] = 127,
+	[2][1][2][0][RTW89_FCC][3] = 127,
+	[2][1][2][0][RTW89_ETSI][3] = 127,
+	[2][1][2][0][RTW89_MKK][3] = 127,
+	[2][1][2][0][RTW89_IC][3] = 127,
+	[2][1][2][0][RTW89_KCC][3] = 127,
+	[2][1][2][0][RTW89_ACMA][3] = 127,
+	[2][1][2][0][RTW89_CN][3] = 127,
+	[2][1][2][0][RTW89_UK][3] = 127,
+	[2][1][2][0][RTW89_FCC][11] = 127,
+	[2][1][2][0][RTW89_ETSI][11] = 127,
+	[2][1][2][0][RTW89_MKK][11] = 127,
+	[2][1][2][0][RTW89_IC][11] = 127,
+	[2][1][2][0][RTW89_KCC][11] = 127,
+	[2][1][2][0][RTW89_ACMA][11] = 127,
+	[2][1][2][0][RTW89_CN][11] = 127,
+	[2][1][2][0][RTW89_UK][11] = 127,
+	[2][1][2][0][RTW89_FCC][18] = 127,
+	[2][1][2][0][RTW89_ETSI][18] = 127,
+	[2][1][2][0][RTW89_MKK][18] = 127,
+	[2][1][2][0][RTW89_IC][18] = 127,
+	[2][1][2][0][RTW89_KCC][18] = 127,
+	[2][1][2][0][RTW89_ACMA][18] = 127,
+	[2][1][2][0][RTW89_CN][18] = 127,
+	[2][1][2][0][RTW89_UK][18] = 127,
+	[2][1][2][0][RTW89_FCC][26] = 127,
+	[2][1][2][0][RTW89_ETSI][26] = 127,
+	[2][1][2][0][RTW89_MKK][26] = 127,
+	[2][1][2][0][RTW89_IC][26] = 127,
+	[2][1][2][0][RTW89_KCC][26] = 127,
+	[2][1][2][0][RTW89_ACMA][26] = 127,
+	[2][1][2][0][RTW89_CN][26] = 127,
+	[2][1][2][0][RTW89_UK][26] = 127,
+	[2][1][2][0][RTW89_FCC][34] = 127,
+	[2][1][2][0][RTW89_ETSI][34] = 127,
+	[2][1][2][0][RTW89_MKK][34] = 127,
+	[2][1][2][0][RTW89_IC][34] = 127,
+	[2][1][2][0][RTW89_KCC][34] = 127,
+	[2][1][2][0][RTW89_ACMA][34] = 127,
+	[2][1][2][0][RTW89_CN][34] = 127,
+	[2][1][2][0][RTW89_UK][34] = 127,
+	[2][1][2][0][RTW89_FCC][41] = 127,
+	[2][1][2][0][RTW89_ETSI][41] = 127,
+	[2][1][2][0][RTW89_MKK][41] = 127,
+	[2][1][2][0][RTW89_IC][41] = 127,
+	[2][1][2][0][RTW89_KCC][41] = 127,
+	[2][1][2][0][RTW89_ACMA][41] = 127,
+	[2][1][2][0][RTW89_CN][41] = 127,
+	[2][1][2][0][RTW89_UK][41] = 127,
+	[2][1][2][0][RTW89_FCC][49] = 127,
+	[2][1][2][0][RTW89_ETSI][49] = 127,
+	[2][1][2][0][RTW89_MKK][49] = 127,
+	[2][1][2][0][RTW89_IC][49] = 127,
+	[2][1][2][0][RTW89_KCC][49] = 127,
+	[2][1][2][0][RTW89_ACMA][49] = 127,
+	[2][1][2][0][RTW89_CN][49] = 127,
+	[2][1][2][0][RTW89_UK][49] = 127,
+	[2][1][2][1][RTW89_FCC][3] = 127,
+	[2][1][2][1][RTW89_ETSI][3] = 127,
+	[2][1][2][1][RTW89_MKK][3] = 127,
+	[2][1][2][1][RTW89_IC][3] = 127,
+	[2][1][2][1][RTW89_KCC][3] = 127,
+	[2][1][2][1][RTW89_ACMA][3] = 127,
+	[2][1][2][1][RTW89_CN][3] = 127,
+	[2][1][2][1][RTW89_UK][3] = 127,
+	[2][1][2][1][RTW89_FCC][11] = 127,
+	[2][1][2][1][RTW89_ETSI][11] = 127,
+	[2][1][2][1][RTW89_MKK][11] = 127,
+	[2][1][2][1][RTW89_IC][11] = 127,
+	[2][1][2][1][RTW89_KCC][11] = 127,
+	[2][1][2][1][RTW89_ACMA][11] = 127,
+	[2][1][2][1][RTW89_CN][11] = 127,
+	[2][1][2][1][RTW89_UK][11] = 127,
+	[2][1][2][1][RTW89_FCC][18] = 127,
+	[2][1][2][1][RTW89_ETSI][18] = 127,
+	[2][1][2][1][RTW89_MKK][18] = 127,
+	[2][1][2][1][RTW89_IC][18] = 127,
+	[2][1][2][1][RTW89_KCC][18] = 127,
+	[2][1][2][1][RTW89_ACMA][18] = 127,
+	[2][1][2][1][RTW89_CN][18] = 127,
+	[2][1][2][1][RTW89_UK][18] = 127,
+	[2][1][2][1][RTW89_FCC][26] = 127,
+	[2][1][2][1][RTW89_ETSI][26] = 127,
+	[2][1][2][1][RTW89_MKK][26] = 127,
+	[2][1][2][1][RTW89_IC][26] = 127,
+	[2][1][2][1][RTW89_KCC][26] = 127,
+	[2][1][2][1][RTW89_ACMA][26] = 127,
+	[2][1][2][1][RTW89_CN][26] = 127,
+	[2][1][2][1][RTW89_UK][26] = 127,
+	[2][1][2][1][RTW89_FCC][34] = 127,
+	[2][1][2][1][RTW89_ETSI][34] = 127,
+	[2][1][2][1][RTW89_MKK][34] = 127,
+	[2][1][2][1][RTW89_IC][34] = 127,
+	[2][1][2][1][RTW89_KCC][34] = 127,
+	[2][1][2][1][RTW89_ACMA][34] = 127,
+	[2][1][2][1][RTW89_CN][34] = 127,
+	[2][1][2][1][RTW89_UK][34] = 127,
+	[2][1][2][1][RTW89_FCC][41] = 127,
+	[2][1][2][1][RTW89_ETSI][41] = 127,
+	[2][1][2][1][RTW89_MKK][41] = 127,
+	[2][1][2][1][RTW89_IC][41] = 127,
+	[2][1][2][1][RTW89_KCC][41] = 127,
+	[2][1][2][1][RTW89_ACMA][41] = 127,
+	[2][1][2][1][RTW89_CN][41] = 127,
+	[2][1][2][1][RTW89_UK][41] = 127,
+	[2][1][2][1][RTW89_FCC][49] = 127,
+	[2][1][2][1][RTW89_ETSI][49] = 127,
+	[2][1][2][1][RTW89_MKK][49] = 127,
+	[2][1][2][1][RTW89_IC][49] = 127,
+	[2][1][2][1][RTW89_KCC][49] = 127,
+	[2][1][2][1][RTW89_ACMA][49] = 127,
+	[2][1][2][1][RTW89_CN][49] = 127,
+	[2][1][2][1][RTW89_UK][49] = 127,
+	[3][0][2][0][RTW89_FCC][7] = 127,
+	[3][0][2][0][RTW89_ETSI][7] = 127,
+	[3][0][2][0][RTW89_MKK][7] = 127,
+	[3][0][2][0][RTW89_IC][7] = 127,
+	[3][0][2][0][RTW89_KCC][7] = 127,
+	[3][0][2][0][RTW89_ACMA][7] = 127,
+	[3][0][2][0][RTW89_CN][7] = 58,
+	[3][0][2][0][RTW89_UK][7] = 127,
+	[3][0][2][0][RTW89_FCC][22] = 127,
+	[3][0][2][0][RTW89_ETSI][22] = 127,
+	[3][0][2][0][RTW89_MKK][22] = 127,
+	[3][0][2][0][RTW89_IC][22] = 127,
+	[3][0][2][0][RTW89_KCC][22] = 127,
+	[3][0][2][0][RTW89_ACMA][22] = 127,
+	[3][0][2][0][RTW89_CN][22] = 58,
+	[3][0][2][0][RTW89_UK][22] = 127,
+	[3][0][2][0][RTW89_FCC][45] = 127,
+	[3][0][2][0][RTW89_ETSI][45] = 127,
+	[3][0][2][0][RTW89_MKK][45] = 127,
+	[3][0][2][0][RTW89_IC][45] = 127,
+	[3][0][2][0][RTW89_KCC][45] = 127,
+	[3][0][2][0][RTW89_ACMA][45] = 127,
+	[3][0][2][0][RTW89_CN][45] = 127,
+	[3][0][2][0][RTW89_UK][45] = 127,
+	[3][1][2][0][RTW89_FCC][7] = 127,
+	[3][1][2][0][RTW89_ETSI][7] = 127,
+	[3][1][2][0][RTW89_MKK][7] = 127,
+	[3][1][2][0][RTW89_IC][7] = 127,
+	[3][1][2][0][RTW89_KCC][7] = 127,
+	[3][1][2][0][RTW89_ACMA][7] = 127,
+	[3][1][2][0][RTW89_CN][7] = 127,
+	[3][1][2][0][RTW89_UK][7] = 127,
+	[3][1][2][0][RTW89_FCC][22] = 127,
+	[3][1][2][0][RTW89_ETSI][22] = 127,
+	[3][1][2][0][RTW89_MKK][22] = 127,
+	[3][1][2][0][RTW89_IC][22] = 127,
+	[3][1][2][0][RTW89_KCC][22] = 127,
+	[3][1][2][0][RTW89_ACMA][22] = 127,
+	[3][1][2][0][RTW89_CN][22] = 127,
+	[3][1][2][0][RTW89_UK][22] = 127,
+	[3][1][2][0][RTW89_FCC][45] = 127,
+	[3][1][2][0][RTW89_ETSI][45] = 127,
+	[3][1][2][0][RTW89_MKK][45] = 127,
+	[3][1][2][0][RTW89_IC][45] = 127,
+	[3][1][2][0][RTW89_KCC][45] = 127,
+	[3][1][2][0][RTW89_ACMA][45] = 127,
+	[3][1][2][0][RTW89_CN][45] = 127,
+	[3][1][2][0][RTW89_UK][45] = 127,
+	[3][1][2][1][RTW89_FCC][7] = 127,
+	[3][1][2][1][RTW89_ETSI][7] = 127,
+	[3][1][2][1][RTW89_MKK][7] = 127,
+	[3][1][2][1][RTW89_IC][7] = 127,
+	[3][1][2][1][RTW89_KCC][7] = 127,
+	[3][1][2][1][RTW89_ACMA][7] = 127,
+	[3][1][2][1][RTW89_CN][7] = 127,
+	[3][1][2][1][RTW89_UK][7] = 127,
+	[3][1][2][1][RTW89_FCC][22] = 127,
+	[3][1][2][1][RTW89_ETSI][22] = 127,
+	[3][1][2][1][RTW89_MKK][22] = 127,
+	[3][1][2][1][RTW89_IC][22] = 127,
+	[3][1][2][1][RTW89_KCC][22] = 127,
+	[3][1][2][1][RTW89_ACMA][22] = 127,
+	[3][1][2][1][RTW89_CN][22] = 127,
+	[3][1][2][1][RTW89_UK][22] = 127,
+	[3][1][2][1][RTW89_FCC][45] = 127,
+	[3][1][2][1][RTW89_ETSI][45] = 127,
+	[3][1][2][1][RTW89_MKK][45] = 127,
+	[3][1][2][1][RTW89_IC][45] = 127,
+	[3][1][2][1][RTW89_KCC][45] = 127,
+	[3][1][2][1][RTW89_ACMA][45] = 127,
+	[3][1][2][1][RTW89_CN][45] = 127,
+	[3][1][2][1][RTW89_UK][45] = 127,
+};
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b_table.h b/drivers/net/wireless/realtek/rtw89/rtw8851b_table.h
new file mode 100644
index 000000000000..f2e673ba39c8
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b_table.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2022-2023  Realtek Corporation
+ */
+
+#ifndef __RTW89_8851B_TABLE_H__
+#define __RTW89_8851B_TABLE_H__
+
+#include "core.h"
+
+extern const struct rtw89_phy_table rtw89_8851b_phy_bb_table;
+extern const struct rtw89_phy_table rtw89_8851b_phy_bb_gain_table;
+extern const struct rtw89_phy_table rtw89_8851b_phy_radioa_table;
+extern const struct rtw89_phy_table rtw89_8851b_phy_nctl_table;
+extern const struct rtw89_txpwr_table rtw89_8851b_byr_table;
+extern const struct rtw89_txpwr_track_cfg rtw89_8851b_trk_cfg;
+extern const u8 rtw89_8851b_tx_shape[RTW89_BAND_MAX][RTW89_RS_TX_SHAPE_NUM]
+				    [RTW89_REGD_NUM];
+extern const struct rtw89_rfe_parms rtw89_8851b_dflt_parms;
+extern const struct rtw89_rfe_parms_conf rtw89_8851b_rfe_parms_conf[];
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From cf4917cf0ab828a4b520daa1708e8572b810417a Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Sat, 1 Apr 2023 22:25:47 +0800
Subject: wifi: rtw89: 8851b: add BB and RF tables (2 of 2)

These tables contain BB and RF parameters that driver will load them into
registers. It also contains TX power according to country, band, rate and
so on. Increasing thermal can cause TX power degraded, so power tracking
tables are defined to compensate TX power.

Internal version of these tables:
 - HALBB_029_106_15 (V17)
 - HALRF_029_00_089
   * Radio A 0x22
   * NCTL 0x5

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230401142548.55466-3-pkshih@realtek.com
---
 .../net/wireless/realtek/rtw89/rtw8851b_table.c    | 8058 ++++++++++++++++++++
 1 file changed, 8058 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c b/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c
index e964b32e4cc4..bb724140df4f 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b_table.c
@@ -6764,3 +6764,8061 @@ const s8 rtw89_8851b_txpwr_lmt_5g[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
 	[3][1][2][1][RTW89_CN][45] = 127,
 	[3][1][2][1][RTW89_UK][45] = 127,
 };
+
+static
+const s8 rtw89_8851b_txpwr_lmt_ru_2g[RTW89_RU_NUM][RTW89_NTX_NUM]
+				    [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
+	[0][0][RTW89_WW][0] = 30,
+	[0][0][RTW89_WW][1] = 30,
+	[0][0][RTW89_WW][2] = 30,
+	[0][0][RTW89_WW][3] = 30,
+	[0][0][RTW89_WW][4] = 30,
+	[0][0][RTW89_WW][5] = 30,
+	[0][0][RTW89_WW][6] = 30,
+	[0][0][RTW89_WW][7] = 30,
+	[0][0][RTW89_WW][8] = 30,
+	[0][0][RTW89_WW][9] = 30,
+	[0][0][RTW89_WW][10] = 30,
+	[0][0][RTW89_WW][11] = 30,
+	[0][0][RTW89_WW][12] = 30,
+	[0][0][RTW89_WW][13] = 0,
+	[0][1][RTW89_WW][0] = 20,
+	[0][1][RTW89_WW][1] = 22,
+	[0][1][RTW89_WW][2] = 22,
+	[0][1][RTW89_WW][3] = 22,
+	[0][1][RTW89_WW][4] = 22,
+	[0][1][RTW89_WW][5] = 22,
+	[0][1][RTW89_WW][6] = 22,
+	[0][1][RTW89_WW][7] = 22,
+	[0][1][RTW89_WW][8] = 22,
+	[0][1][RTW89_WW][9] = 22,
+	[0][1][RTW89_WW][10] = 22,
+	[0][1][RTW89_WW][11] = 22,
+	[0][1][RTW89_WW][12] = 20,
+	[0][1][RTW89_WW][13] = 0,
+	[1][0][RTW89_WW][0] = 42,
+	[1][0][RTW89_WW][1] = 42,
+	[1][0][RTW89_WW][2] = 42,
+	[1][0][RTW89_WW][3] = 42,
+	[1][0][RTW89_WW][4] = 42,
+	[1][0][RTW89_WW][5] = 42,
+	[1][0][RTW89_WW][6] = 42,
+	[1][0][RTW89_WW][7] = 42,
+	[1][0][RTW89_WW][8] = 42,
+	[1][0][RTW89_WW][9] = 42,
+	[1][0][RTW89_WW][10] = 42,
+	[1][0][RTW89_WW][11] = 42,
+	[1][0][RTW89_WW][12] = 34,
+	[1][0][RTW89_WW][13] = 0,
+	[1][1][RTW89_WW][0] = 32,
+	[1][1][RTW89_WW][1] = 32,
+	[1][1][RTW89_WW][2] = 32,
+	[1][1][RTW89_WW][3] = 32,
+	[1][1][RTW89_WW][4] = 32,
+	[1][1][RTW89_WW][5] = 32,
+	[1][1][RTW89_WW][6] = 32,
+	[1][1][RTW89_WW][7] = 32,
+	[1][1][RTW89_WW][8] = 32,
+	[1][1][RTW89_WW][9] = 32,
+	[1][1][RTW89_WW][10] = 32,
+	[1][1][RTW89_WW][11] = 32,
+	[1][1][RTW89_WW][12] = 32,
+	[1][1][RTW89_WW][13] = 0,
+	[2][0][RTW89_WW][0] = 54,
+	[2][0][RTW89_WW][1] = 54,
+	[2][0][RTW89_WW][2] = 54,
+	[2][0][RTW89_WW][3] = 54,
+	[2][0][RTW89_WW][4] = 54,
+	[2][0][RTW89_WW][5] = 54,
+	[2][0][RTW89_WW][6] = 54,
+	[2][0][RTW89_WW][7] = 54,
+	[2][0][RTW89_WW][8] = 54,
+	[2][0][RTW89_WW][9] = 54,
+	[2][0][RTW89_WW][10] = 54,
+	[2][0][RTW89_WW][11] = 54,
+	[2][0][RTW89_WW][12] = 34,
+	[2][0][RTW89_WW][13] = 0,
+	[2][1][RTW89_WW][0] = 44,
+	[2][1][RTW89_WW][1] = 44,
+	[2][1][RTW89_WW][2] = 44,
+	[2][1][RTW89_WW][3] = 44,
+	[2][1][RTW89_WW][4] = 44,
+	[2][1][RTW89_WW][5] = 44,
+	[2][1][RTW89_WW][6] = 44,
+	[2][1][RTW89_WW][7] = 44,
+	[2][1][RTW89_WW][8] = 44,
+	[2][1][RTW89_WW][9] = 44,
+	[2][1][RTW89_WW][10] = 44,
+	[2][1][RTW89_WW][11] = 44,
+	[2][1][RTW89_WW][12] = 42,
+	[2][1][RTW89_WW][13] = 0,
+	[0][0][RTW89_FCC][0] = 62,
+	[0][0][RTW89_ETSI][0] = 30,
+	[0][0][RTW89_MKK][0] = 40,
+	[0][0][RTW89_IC][0] = 62,
+	[0][0][RTW89_KCC][0] = 46,
+	[0][0][RTW89_ACMA][0] = 30,
+	[0][0][RTW89_CN][0] = 32,
+	[0][0][RTW89_UK][0] = 30,
+	[0][0][RTW89_FCC][1] = 62,
+	[0][0][RTW89_ETSI][1] = 30,
+	[0][0][RTW89_MKK][1] = 44,
+	[0][0][RTW89_IC][1] = 62,
+	[0][0][RTW89_KCC][1] = 46,
+	[0][0][RTW89_ACMA][1] = 30,
+	[0][0][RTW89_CN][1] = 32,
+	[0][0][RTW89_UK][1] = 30,
+	[0][0][RTW89_FCC][2] = 66,
+	[0][0][RTW89_ETSI][2] = 30,
+	[0][0][RTW89_MKK][2] = 44,
+	[0][0][RTW89_IC][2] = 66,
+	[0][0][RTW89_KCC][2] = 46,
+	[0][0][RTW89_ACMA][2] = 30,
+	[0][0][RTW89_CN][2] = 32,
+	[0][0][RTW89_UK][2] = 30,
+	[0][0][RTW89_FCC][3] = 70,
+	[0][0][RTW89_ETSI][3] = 30,
+	[0][0][RTW89_MKK][3] = 44,
+	[0][0][RTW89_IC][3] = 70,
+	[0][0][RTW89_KCC][3] = 46,
+	[0][0][RTW89_ACMA][3] = 30,
+	[0][0][RTW89_CN][3] = 32,
+	[0][0][RTW89_UK][3] = 30,
+	[0][0][RTW89_FCC][4] = 70,
+	[0][0][RTW89_ETSI][4] = 30,
+	[0][0][RTW89_MKK][4] = 44,
+	[0][0][RTW89_IC][4] = 70,
+	[0][0][RTW89_KCC][4] = 48,
+	[0][0][RTW89_ACMA][4] = 30,
+	[0][0][RTW89_CN][4] = 32,
+	[0][0][RTW89_UK][4] = 30,
+	[0][0][RTW89_FCC][5] = 84,
+	[0][0][RTW89_ETSI][5] = 30,
+	[0][0][RTW89_MKK][5] = 44,
+	[0][0][RTW89_IC][5] = 84,
+	[0][0][RTW89_KCC][5] = 48,
+	[0][0][RTW89_ACMA][5] = 30,
+	[0][0][RTW89_CN][5] = 32,
+	[0][0][RTW89_UK][5] = 30,
+	[0][0][RTW89_FCC][6] = 66,
+	[0][0][RTW89_ETSI][6] = 30,
+	[0][0][RTW89_MKK][6] = 44,
+	[0][0][RTW89_IC][6] = 66,
+	[0][0][RTW89_KCC][6] = 48,
+	[0][0][RTW89_ACMA][6] = 30,
+	[0][0][RTW89_CN][6] = 32,
+	[0][0][RTW89_UK][6] = 30,
+	[0][0][RTW89_FCC][7] = 66,
+	[0][0][RTW89_ETSI][7] = 30,
+	[0][0][RTW89_MKK][7] = 44,
+	[0][0][RTW89_IC][7] = 66,
+	[0][0][RTW89_KCC][7] = 48,
+	[0][0][RTW89_ACMA][7] = 30,
+	[0][0][RTW89_CN][7] = 32,
+	[0][0][RTW89_UK][7] = 30,
+	[0][0][RTW89_FCC][8] = 62,
+	[0][0][RTW89_ETSI][8] = 30,
+	[0][0][RTW89_MKK][8] = 44,
+	[0][0][RTW89_IC][8] = 62,
+	[0][0][RTW89_KCC][8] = 48,
+	[0][0][RTW89_ACMA][8] = 30,
+	[0][0][RTW89_CN][8] = 32,
+	[0][0][RTW89_UK][8] = 30,
+	[0][0][RTW89_FCC][9] = 58,
+	[0][0][RTW89_ETSI][9] = 30,
+	[0][0][RTW89_MKK][9] = 44,
+	[0][0][RTW89_IC][9] = 58,
+	[0][0][RTW89_KCC][9] = 44,
+	[0][0][RTW89_ACMA][9] = 30,
+	[0][0][RTW89_CN][9] = 32,
+	[0][0][RTW89_UK][9] = 30,
+	[0][0][RTW89_FCC][10] = 58,
+	[0][0][RTW89_ETSI][10] = 30,
+	[0][0][RTW89_MKK][10] = 44,
+	[0][0][RTW89_IC][10] = 58,
+	[0][0][RTW89_KCC][10] = 44,
+	[0][0][RTW89_ACMA][10] = 30,
+	[0][0][RTW89_CN][10] = 32,
+	[0][0][RTW89_UK][10] = 30,
+	[0][0][RTW89_FCC][11] = 54,
+	[0][0][RTW89_ETSI][11] = 30,
+	[0][0][RTW89_MKK][11] = 44,
+	[0][0][RTW89_IC][11] = 54,
+	[0][0][RTW89_KCC][11] = 44,
+	[0][0][RTW89_ACMA][11] = 30,
+	[0][0][RTW89_CN][11] = 32,
+	[0][0][RTW89_UK][11] = 30,
+	[0][0][RTW89_FCC][12] = 36,
+	[0][0][RTW89_ETSI][12] = 30,
+	[0][0][RTW89_MKK][12] = 40,
+	[0][0][RTW89_IC][12] = 36,
+	[0][0][RTW89_KCC][12] = 44,
+	[0][0][RTW89_ACMA][12] = 30,
+	[0][0][RTW89_CN][12] = 32,
+	[0][0][RTW89_UK][12] = 30,
+	[0][0][RTW89_FCC][13] = 127,
+	[0][0][RTW89_ETSI][13] = 127,
+	[0][0][RTW89_MKK][13] = 127,
+	[0][0][RTW89_IC][13] = 127,
+	[0][0][RTW89_KCC][13] = 127,
+	[0][0][RTW89_ACMA][13] = 127,
+	[0][0][RTW89_CN][13] = 127,
+	[0][0][RTW89_UK][13] = 127,
+	[0][1][RTW89_FCC][0] = 127,
+	[0][1][RTW89_ETSI][0] = 127,
+	[0][1][RTW89_MKK][0] = 127,
+	[0][1][RTW89_IC][0] = 127,
+	[0][1][RTW89_KCC][0] = 127,
+	[0][1][RTW89_ACMA][0] = 127,
+	[0][1][RTW89_CN][0] = 20,
+	[0][1][RTW89_UK][0] = 127,
+	[0][1][RTW89_FCC][1] = 127,
+	[0][1][RTW89_ETSI][1] = 127,
+	[0][1][RTW89_MKK][1] = 127,
+	[0][1][RTW89_IC][1] = 127,
+	[0][1][RTW89_KCC][1] = 127,
+	[0][1][RTW89_ACMA][1] = 127,
+	[0][1][RTW89_CN][1] = 22,
+	[0][1][RTW89_UK][1] = 127,
+	[0][1][RTW89_FCC][2] = 127,
+	[0][1][RTW89_ETSI][2] = 127,
+	[0][1][RTW89_MKK][2] = 127,
+	[0][1][RTW89_IC][2] = 127,
+	[0][1][RTW89_KCC][2] = 127,
+	[0][1][RTW89_ACMA][2] = 127,
+	[0][1][RTW89_CN][2] = 22,
+	[0][1][RTW89_UK][2] = 127,
+	[0][1][RTW89_FCC][3] = 127,
+	[0][1][RTW89_ETSI][3] = 127,
+	[0][1][RTW89_MKK][3] = 127,
+	[0][1][RTW89_IC][3] = 127,
+	[0][1][RTW89_KCC][3] = 127,
+	[0][1][RTW89_ACMA][3] = 127,
+	[0][1][RTW89_CN][3] = 22,
+	[0][1][RTW89_UK][3] = 127,
+	[0][1][RTW89_FCC][4] = 127,
+	[0][1][RTW89_ETSI][4] = 127,
+	[0][1][RTW89_MKK][4] = 127,
+	[0][1][RTW89_IC][4] = 127,
+	[0][1][RTW89_KCC][4] = 127,
+	[0][1][RTW89_ACMA][4] = 127,
+	[0][1][RTW89_CN][4] = 22,
+	[0][1][RTW89_UK][4] = 127,
+	[0][1][RTW89_FCC][5] = 127,
+	[0][1][RTW89_ETSI][5] = 127,
+	[0][1][RTW89_MKK][5] = 127,
+	[0][1][RTW89_IC][5] = 127,
+	[0][1][RTW89_KCC][5] = 127,
+	[0][1][RTW89_ACMA][5] = 127,
+	[0][1][RTW89_CN][5] = 22,
+	[0][1][RTW89_UK][5] = 127,
+	[0][1][RTW89_FCC][6] = 127,
+	[0][1][RTW89_ETSI][6] = 127,
+	[0][1][RTW89_MKK][6] = 127,
+	[0][1][RTW89_IC][6] = 127,
+	[0][1][RTW89_KCC][6] = 127,
+	[0][1][RTW89_ACMA][6] = 127,
+	[0][1][RTW89_CN][6] = 22,
+	[0][1][RTW89_UK][6] = 127,
+	[0][1][RTW89_FCC][7] = 127,
+	[0][1][RTW89_ETSI][7] = 127,
+	[0][1][RTW89_MKK][7] = 127,
+	[0][1][RTW89_IC][7] = 127,
+	[0][1][RTW89_KCC][7] = 127,
+	[0][1][RTW89_ACMA][7] = 127,
+	[0][1][RTW89_CN][7] = 22,
+	[0][1][RTW89_UK][7] = 127,
+	[0][1][RTW89_FCC][8] = 127,
+	[0][1][RTW89_ETSI][8] = 127,
+	[0][1][RTW89_MKK][8] = 127,
+	[0][1][RTW89_IC][8] = 127,
+	[0][1][RTW89_KCC][8] = 127,
+	[0][1][RTW89_ACMA][8] = 127,
+	[0][1][RTW89_CN][8] = 22,
+	[0][1][RTW89_UK][8] = 127,
+	[0][1][RTW89_FCC][9] = 127,
+	[0][1][RTW89_ETSI][9] = 127,
+	[0][1][RTW89_MKK][9] = 127,
+	[0][1][RTW89_IC][9] = 127,
+	[0][1][RTW89_KCC][9] = 127,
+	[0][1][RTW89_ACMA][9] = 127,
+	[0][1][RTW89_CN][9] = 22,
+	[0][1][RTW89_UK][9] = 127,
+	[0][1][RTW89_FCC][10] = 127,
+	[0][1][RTW89_ETSI][10] = 127,
+	[0][1][RTW89_MKK][10] = 127,
+	[0][1][RTW89_IC][10] = 127,
+	[0][1][RTW89_KCC][10] = 127,
+	[0][1][RTW89_ACMA][10] = 127,
+	[0][1][RTW89_CN][10] = 22,
+	[0][1][RTW89_UK][10] = 127,
+	[0][1][RTW89_FCC][11] = 127,
+	[0][1][RTW89_ETSI][11] = 127,
+	[0][1][RTW89_MKK][11] = 127,
+	[0][1][RTW89_IC][11] = 127,
+	[0][1][RTW89_KCC][11] = 127,
+	[0][1][RTW89_ACMA][11] = 127,
+	[0][1][RTW89_CN][11] = 22,
+	[0][1][RTW89_UK][11] = 127,
+	[0][1][RTW89_FCC][12] = 127,
+	[0][1][RTW89_ETSI][12] = 127,
+	[0][1][RTW89_MKK][12] = 127,
+	[0][1][RTW89_IC][12] = 127,
+	[0][1][RTW89_KCC][12] = 127,
+	[0][1][RTW89_ACMA][12] = 127,
+	[0][1][RTW89_CN][12] = 20,
+	[0][1][RTW89_UK][12] = 127,
+	[0][1][RTW89_FCC][13] = 127,
+	[0][1][RTW89_ETSI][13] = 127,
+	[0][1][RTW89_MKK][13] = 127,
+	[0][1][RTW89_IC][13] = 127,
+	[0][1][RTW89_KCC][13] = 127,
+	[0][1][RTW89_ACMA][13] = 127,
+	[0][1][RTW89_CN][13] = 127,
+	[0][1][RTW89_UK][13] = 127,
+	[1][0][RTW89_FCC][0] = 70,
+	[1][0][RTW89_ETSI][0] = 42,
+	[1][0][RTW89_MKK][0] = 52,
+	[1][0][RTW89_IC][0] = 70,
+	[1][0][RTW89_KCC][0] = 56,
+	[1][0][RTW89_ACMA][0] = 42,
+	[1][0][RTW89_CN][0] = 42,
+	[1][0][RTW89_UK][0] = 42,
+	[1][0][RTW89_FCC][1] = 70,
+	[1][0][RTW89_ETSI][1] = 42,
+	[1][0][RTW89_MKK][1] = 52,
+	[1][0][RTW89_IC][1] = 70,
+	[1][0][RTW89_KCC][1] = 56,
+	[1][0][RTW89_ACMA][1] = 42,
+	[1][0][RTW89_CN][1] = 44,
+	[1][0][RTW89_UK][1] = 42,
+	[1][0][RTW89_FCC][2] = 74,
+	[1][0][RTW89_ETSI][2] = 42,
+	[1][0][RTW89_MKK][2] = 52,
+	[1][0][RTW89_IC][2] = 74,
+	[1][0][RTW89_KCC][2] = 56,
+	[1][0][RTW89_ACMA][2] = 42,
+	[1][0][RTW89_CN][2] = 44,
+	[1][0][RTW89_UK][2] = 42,
+	[1][0][RTW89_FCC][3] = 76,
+	[1][0][RTW89_ETSI][3] = 42,
+	[1][0][RTW89_MKK][3] = 52,
+	[1][0][RTW89_IC][3] = 76,
+	[1][0][RTW89_KCC][3] = 56,
+	[1][0][RTW89_ACMA][3] = 42,
+	[1][0][RTW89_CN][3] = 44,
+	[1][0][RTW89_UK][3] = 42,
+	[1][0][RTW89_FCC][4] = 76,
+	[1][0][RTW89_ETSI][4] = 42,
+	[1][0][RTW89_MKK][4] = 52,
+	[1][0][RTW89_IC][4] = 76,
+	[1][0][RTW89_KCC][4] = 56,
+	[1][0][RTW89_ACMA][4] = 42,
+	[1][0][RTW89_CN][4] = 44,
+	[1][0][RTW89_UK][4] = 42,
+	[1][0][RTW89_FCC][5] = 82,
+	[1][0][RTW89_ETSI][5] = 42,
+	[1][0][RTW89_MKK][5] = 52,
+	[1][0][RTW89_IC][5] = 82,
+	[1][0][RTW89_KCC][5] = 56,
+	[1][0][RTW89_ACMA][5] = 42,
+	[1][0][RTW89_CN][5] = 44,
+	[1][0][RTW89_UK][5] = 42,
+	[1][0][RTW89_FCC][6] = 74,
+	[1][0][RTW89_ETSI][6] = 42,
+	[1][0][RTW89_MKK][6] = 52,
+	[1][0][RTW89_IC][6] = 74,
+	[1][0][RTW89_KCC][6] = 56,
+	[1][0][RTW89_ACMA][6] = 42,
+	[1][0][RTW89_CN][6] = 44,
+	[1][0][RTW89_UK][6] = 42,
+	[1][0][RTW89_FCC][7] = 74,
+	[1][0][RTW89_ETSI][7] = 42,
+	[1][0][RTW89_MKK][7] = 52,
+	[1][0][RTW89_IC][7] = 74,
+	[1][0][RTW89_KCC][7] = 56,
+	[1][0][RTW89_ACMA][7] = 42,
+	[1][0][RTW89_CN][7] = 44,
+	[1][0][RTW89_UK][7] = 42,
+	[1][0][RTW89_FCC][8] = 74,
+	[1][0][RTW89_ETSI][8] = 42,
+	[1][0][RTW89_MKK][8] = 52,
+	[1][0][RTW89_IC][8] = 74,
+	[1][0][RTW89_KCC][8] = 56,
+	[1][0][RTW89_ACMA][8] = 42,
+	[1][0][RTW89_CN][8] = 44,
+	[1][0][RTW89_UK][8] = 42,
+	[1][0][RTW89_FCC][9] = 70,
+	[1][0][RTW89_ETSI][9] = 42,
+	[1][0][RTW89_MKK][9] = 52,
+	[1][0][RTW89_IC][9] = 70,
+	[1][0][RTW89_KCC][9] = 58,
+	[1][0][RTW89_ACMA][9] = 42,
+	[1][0][RTW89_CN][9] = 44,
+	[1][0][RTW89_UK][9] = 42,
+	[1][0][RTW89_FCC][10] = 70,
+	[1][0][RTW89_ETSI][10] = 42,
+	[1][0][RTW89_MKK][10] = 52,
+	[1][0][RTW89_IC][10] = 70,
+	[1][0][RTW89_KCC][10] = 58,
+	[1][0][RTW89_ACMA][10] = 42,
+	[1][0][RTW89_CN][10] = 44,
+	[1][0][RTW89_UK][10] = 42,
+	[1][0][RTW89_FCC][11] = 66,
+	[1][0][RTW89_ETSI][11] = 42,
+	[1][0][RTW89_MKK][11] = 52,
+	[1][0][RTW89_IC][11] = 66,
+	[1][0][RTW89_KCC][11] = 58,
+	[1][0][RTW89_ACMA][11] = 42,
+	[1][0][RTW89_CN][11] = 44,
+	[1][0][RTW89_UK][11] = 42,
+	[1][0][RTW89_FCC][12] = 34,
+	[1][0][RTW89_ETSI][12] = 42,
+	[1][0][RTW89_MKK][12] = 52,
+	[1][0][RTW89_IC][12] = 34,
+	[1][0][RTW89_KCC][12] = 58,
+	[1][0][RTW89_ACMA][12] = 42,
+	[1][0][RTW89_CN][12] = 42,
+	[1][0][RTW89_UK][12] = 42,
+	[1][0][RTW89_FCC][13] = 127,
+	[1][0][RTW89_ETSI][13] = 127,
+	[1][0][RTW89_MKK][13] = 127,
+	[1][0][RTW89_IC][13] = 127,
+	[1][0][RTW89_KCC][13] = 127,
+	[1][0][RTW89_ACMA][13] = 127,
+	[1][0][RTW89_CN][13] = 127,
+	[1][0][RTW89_UK][13] = 127,
+	[1][1][RTW89_FCC][0] = 127,
+	[1][1][RTW89_ETSI][0] = 127,
+	[1][1][RTW89_MKK][0] = 127,
+	[1][1][RTW89_IC][0] = 127,
+	[1][1][RTW89_KCC][0] = 127,
+	[1][1][RTW89_ACMA][0] = 127,
+	[1][1][RTW89_CN][0] = 32,
+	[1][1][RTW89_UK][0] = 127,
+	[1][1][RTW89_FCC][1] = 127,
+	[1][1][RTW89_ETSI][1] = 127,
+	[1][1][RTW89_MKK][1] = 127,
+	[1][1][RTW89_IC][1] = 127,
+	[1][1][RTW89_KCC][1] = 127,
+	[1][1][RTW89_ACMA][1] = 127,
+	[1][1][RTW89_CN][1] = 32,
+	[1][1][RTW89_UK][1] = 127,
+	[1][1][RTW89_FCC][2] = 127,
+	[1][1][RTW89_ETSI][2] = 127,
+	[1][1][RTW89_MKK][2] = 127,
+	[1][1][RTW89_IC][2] = 127,
+	[1][1][RTW89_KCC][2] = 127,
+	[1][1][RTW89_ACMA][2] = 127,
+	[1][1][RTW89_CN][2] = 32,
+	[1][1][RTW89_UK][2] = 127,
+	[1][1][RTW89_FCC][3] = 127,
+	[1][1][RTW89_ETSI][3] = 127,
+	[1][1][RTW89_MKK][3] = 127,
+	[1][1][RTW89_IC][3] = 127,
+	[1][1][RTW89_KCC][3] = 127,
+	[1][1][RTW89_ACMA][3] = 127,
+	[1][1][RTW89_CN][3] = 32,
+	[1][1][RTW89_UK][3] = 127,
+	[1][1][RTW89_FCC][4] = 127,
+	[1][1][RTW89_ETSI][4] = 127,
+	[1][1][RTW89_MKK][4] = 127,
+	[1][1][RTW89_IC][4] = 127,
+	[1][1][RTW89_KCC][4] = 127,
+	[1][1][RTW89_ACMA][4] = 127,
+	[1][1][RTW89_CN][4] = 32,
+	[1][1][RTW89_UK][4] = 127,
+	[1][1][RTW89_FCC][5] = 127,
+	[1][1][RTW89_ETSI][5] = 127,
+	[1][1][RTW89_MKK][5] = 127,
+	[1][1][RTW89_IC][5] = 127,
+	[1][1][RTW89_KCC][5] = 127,
+	[1][1][RTW89_ACMA][5] = 127,
+	[1][1][RTW89_CN][5] = 32,
+	[1][1][RTW89_UK][5] = 127,
+	[1][1][RTW89_FCC][6] = 127,
+	[1][1][RTW89_ETSI][6] = 127,
+	[1][1][RTW89_MKK][6] = 127,
+	[1][1][RTW89_IC][6] = 127,
+	[1][1][RTW89_KCC][6] = 127,
+	[1][1][RTW89_ACMA][6] = 127,
+	[1][1][RTW89_CN][6] = 32,
+	[1][1][RTW89_UK][6] = 127,
+	[1][1][RTW89_FCC][7] = 127,
+	[1][1][RTW89_ETSI][7] = 127,
+	[1][1][RTW89_MKK][7] = 127,
+	[1][1][RTW89_IC][7] = 127,
+	[1][1][RTW89_KCC][7] = 127,
+	[1][1][RTW89_ACMA][7] = 127,
+	[1][1][RTW89_CN][7] = 32,
+	[1][1][RTW89_UK][7] = 127,
+	[1][1][RTW89_FCC][8] = 127,
+	[1][1][RTW89_ETSI][8] = 127,
+	[1][1][RTW89_MKK][8] = 127,
+	[1][1][RTW89_IC][8] = 127,
+	[1][1][RTW89_KCC][8] = 127,
+	[1][1][RTW89_ACMA][8] = 127,
+	[1][1][RTW89_CN][8] = 32,
+	[1][1][RTW89_UK][8] = 127,
+	[1][1][RTW89_FCC][9] = 127,
+	[1][1][RTW89_ETSI][9] = 127,
+	[1][1][RTW89_MKK][9] = 127,
+	[1][1][RTW89_IC][9] = 127,
+	[1][1][RTW89_KCC][9] = 127,
+	[1][1][RTW89_ACMA][9] = 127,
+	[1][1][RTW89_CN][9] = 32,
+	[1][1][RTW89_UK][9] = 127,
+	[1][1][RTW89_FCC][10] = 127,
+	[1][1][RTW89_ETSI][10] = 127,
+	[1][1][RTW89_MKK][10] = 127,
+	[1][1][RTW89_IC][10] = 127,
+	[1][1][RTW89_KCC][10] = 127,
+	[1][1][RTW89_ACMA][10] = 127,
+	[1][1][RTW89_CN][10] = 32,
+	[1][1][RTW89_UK][10] = 127,
+	[1][1][RTW89_FCC][11] = 127,
+	[1][1][RTW89_ETSI][11] = 127,
+	[1][1][RTW89_MKK][11] = 127,
+	[1][1][RTW89_IC][11] = 127,
+	[1][1][RTW89_KCC][11] = 127,
+	[1][1][RTW89_ACMA][11] = 127,
+	[1][1][RTW89_CN][11] = 32,
+	[1][1][RTW89_UK][11] = 127,
+	[1][1][RTW89_FCC][12] = 127,
+	[1][1][RTW89_ETSI][12] = 127,
+	[1][1][RTW89_MKK][12] = 127,
+	[1][1][RTW89_IC][12] = 127,
+	[1][1][RTW89_KCC][12] = 127,
+	[1][1][RTW89_ACMA][12] = 127,
+	[1][1][RTW89_CN][12] = 32,
+	[1][1][RTW89_UK][12] = 127,
+	[1][1][RTW89_FCC][13] = 127,
+	[1][1][RTW89_ETSI][13] = 127,
+	[1][1][RTW89_MKK][13] = 127,
+	[1][1][RTW89_IC][13] = 127,
+	[1][1][RTW89_KCC][13] = 127,
+	[1][1][RTW89_ACMA][13] = 127,
+	[1][1][RTW89_CN][13] = 127,
+	[1][1][RTW89_UK][13] = 127,
+	[2][0][RTW89_FCC][0] = 76,
+	[2][0][RTW89_ETSI][0] = 54,
+	[2][0][RTW89_MKK][0] = 64,
+	[2][0][RTW89_IC][0] = 76,
+	[2][0][RTW89_KCC][0] = 68,
+	[2][0][RTW89_ACMA][0] = 54,
+	[2][0][RTW89_CN][0] = 56,
+	[2][0][RTW89_UK][0] = 54,
+	[2][0][RTW89_FCC][1] = 76,
+	[2][0][RTW89_ETSI][1] = 54,
+	[2][0][RTW89_MKK][1] = 64,
+	[2][0][RTW89_IC][1] = 76,
+	[2][0][RTW89_KCC][1] = 68,
+	[2][0][RTW89_ACMA][1] = 54,
+	[2][0][RTW89_CN][1] = 56,
+	[2][0][RTW89_UK][1] = 54,
+	[2][0][RTW89_FCC][2] = 78,
+	[2][0][RTW89_ETSI][2] = 54,
+	[2][0][RTW89_MKK][2] = 64,
+	[2][0][RTW89_IC][2] = 78,
+	[2][0][RTW89_KCC][2] = 68,
+	[2][0][RTW89_ACMA][2] = 54,
+	[2][0][RTW89_CN][2] = 56,
+	[2][0][RTW89_UK][2] = 54,
+	[2][0][RTW89_FCC][3] = 78,
+	[2][0][RTW89_ETSI][3] = 54,
+	[2][0][RTW89_MKK][3] = 64,
+	[2][0][RTW89_IC][3] = 78,
+	[2][0][RTW89_KCC][3] = 68,
+	[2][0][RTW89_ACMA][3] = 54,
+	[2][0][RTW89_CN][3] = 56,
+	[2][0][RTW89_UK][3] = 54,
+	[2][0][RTW89_FCC][4] = 78,
+	[2][0][RTW89_ETSI][4] = 54,
+	[2][0][RTW89_MKK][4] = 64,
+	[2][0][RTW89_IC][4] = 78,
+	[2][0][RTW89_KCC][4] = 68,
+	[2][0][RTW89_ACMA][4] = 54,
+	[2][0][RTW89_CN][4] = 56,
+	[2][0][RTW89_UK][4] = 54,
+	[2][0][RTW89_FCC][5] = 82,
+	[2][0][RTW89_ETSI][5] = 54,
+	[2][0][RTW89_MKK][5] = 64,
+	[2][0][RTW89_IC][5] = 82,
+	[2][0][RTW89_KCC][5] = 68,
+	[2][0][RTW89_ACMA][5] = 54,
+	[2][0][RTW89_CN][5] = 56,
+	[2][0][RTW89_UK][5] = 54,
+	[2][0][RTW89_FCC][6] = 74,
+	[2][0][RTW89_ETSI][6] = 54,
+	[2][0][RTW89_MKK][6] = 64,
+	[2][0][RTW89_IC][6] = 74,
+	[2][0][RTW89_KCC][6] = 68,
+	[2][0][RTW89_ACMA][6] = 54,
+	[2][0][RTW89_CN][6] = 56,
+	[2][0][RTW89_UK][6] = 54,
+	[2][0][RTW89_FCC][7] = 74,
+	[2][0][RTW89_ETSI][7] = 54,
+	[2][0][RTW89_MKK][7] = 64,
+	[2][0][RTW89_IC][7] = 74,
+	[2][0][RTW89_KCC][7] = 68,
+	[2][0][RTW89_ACMA][7] = 54,
+	[2][0][RTW89_CN][7] = 56,
+	[2][0][RTW89_UK][7] = 54,
+	[2][0][RTW89_FCC][8] = 74,
+	[2][0][RTW89_ETSI][8] = 54,
+	[2][0][RTW89_MKK][8] = 64,
+	[2][0][RTW89_IC][8] = 74,
+	[2][0][RTW89_KCC][8] = 68,
+	[2][0][RTW89_ACMA][8] = 54,
+	[2][0][RTW89_CN][8] = 56,
+	[2][0][RTW89_UK][8] = 54,
+	[2][0][RTW89_FCC][9] = 72,
+	[2][0][RTW89_ETSI][9] = 54,
+	[2][0][RTW89_MKK][9] = 64,
+	[2][0][RTW89_IC][9] = 72,
+	[2][0][RTW89_KCC][9] = 68,
+	[2][0][RTW89_ACMA][9] = 54,
+	[2][0][RTW89_CN][9] = 56,
+	[2][0][RTW89_UK][9] = 54,
+	[2][0][RTW89_FCC][10] = 72,
+	[2][0][RTW89_ETSI][10] = 54,
+	[2][0][RTW89_MKK][10] = 64,
+	[2][0][RTW89_IC][10] = 72,
+	[2][0][RTW89_KCC][10] = 68,
+	[2][0][RTW89_ACMA][10] = 54,
+	[2][0][RTW89_CN][10] = 56,
+	[2][0][RTW89_UK][10] = 54,
+	[2][0][RTW89_FCC][11] = 64,
+	[2][0][RTW89_ETSI][11] = 54,
+	[2][0][RTW89_MKK][11] = 64,
+	[2][0][RTW89_IC][11] = 64,
+	[2][0][RTW89_KCC][11] = 68,
+	[2][0][RTW89_ACMA][11] = 54,
+	[2][0][RTW89_CN][11] = 56,
+	[2][0][RTW89_UK][11] = 54,
+	[2][0][RTW89_FCC][12] = 34,
+	[2][0][RTW89_ETSI][12] = 54,
+	[2][0][RTW89_MKK][12] = 64,
+	[2][0][RTW89_IC][12] = 34,
+	[2][0][RTW89_KCC][12] = 68,
+	[2][0][RTW89_ACMA][12] = 54,
+	[2][0][RTW89_CN][12] = 56,
+	[2][0][RTW89_UK][12] = 54,
+	[2][0][RTW89_FCC][13] = 127,
+	[2][0][RTW89_ETSI][13] = 127,
+	[2][0][RTW89_MKK][13] = 127,
+	[2][0][RTW89_IC][13] = 127,
+	[2][0][RTW89_KCC][13] = 127,
+	[2][0][RTW89_ACMA][13] = 127,
+	[2][0][RTW89_CN][13] = 127,
+	[2][0][RTW89_UK][13] = 127,
+	[2][1][RTW89_FCC][0] = 127,
+	[2][1][RTW89_ETSI][0] = 127,
+	[2][1][RTW89_MKK][0] = 127,
+	[2][1][RTW89_IC][0] = 127,
+	[2][1][RTW89_KCC][0] = 127,
+	[2][1][RTW89_ACMA][0] = 127,
+	[2][1][RTW89_CN][0] = 44,
+	[2][1][RTW89_UK][0] = 127,
+	[2][1][RTW89_FCC][1] = 127,
+	[2][1][RTW89_ETSI][1] = 127,
+	[2][1][RTW89_MKK][1] = 127,
+	[2][1][RTW89_IC][1] = 127,
+	[2][1][RTW89_KCC][1] = 127,
+	[2][1][RTW89_ACMA][1] = 127,
+	[2][1][RTW89_CN][1] = 44,
+	[2][1][RTW89_UK][1] = 127,
+	[2][1][RTW89_FCC][2] = 127,
+	[2][1][RTW89_ETSI][2] = 127,
+	[2][1][RTW89_MKK][2] = 127,
+	[2][1][RTW89_IC][2] = 127,
+	[2][1][RTW89_KCC][2] = 127,
+	[2][1][RTW89_ACMA][2] = 127,
+	[2][1][RTW89_CN][2] = 44,
+	[2][1][RTW89_UK][2] = 127,
+	[2][1][RTW89_FCC][3] = 127,
+	[2][1][RTW89_ETSI][3] = 127,
+	[2][1][RTW89_MKK][3] = 127,
+	[2][1][RTW89_IC][3] = 127,
+	[2][1][RTW89_KCC][3] = 127,
+	[2][1][RTW89_ACMA][3] = 127,
+	[2][1][RTW89_CN][3] = 44,
+	[2][1][RTW89_UK][3] = 127,
+	[2][1][RTW89_FCC][4] = 127,
+	[2][1][RTW89_ETSI][4] = 127,
+	[2][1][RTW89_MKK][4] = 127,
+	[2][1][RTW89_IC][4] = 127,
+	[2][1][RTW89_KCC][4] = 127,
+	[2][1][RTW89_ACMA][4] = 127,
+	[2][1][RTW89_CN][4] = 44,
+	[2][1][RTW89_UK][4] = 127,
+	[2][1][RTW89_FCC][5] = 127,
+	[2][1][RTW89_ETSI][5] = 127,
+	[2][1][RTW89_MKK][5] = 127,
+	[2][1][RTW89_IC][5] = 127,
+	[2][1][RTW89_KCC][5] = 127,
+	[2][1][RTW89_ACMA][5] = 127,
+	[2][1][RTW89_CN][5] = 44,
+	[2][1][RTW89_UK][5] = 127,
+	[2][1][RTW89_FCC][6] = 127,
+	[2][1][RTW89_ETSI][6] = 127,
+	[2][1][RTW89_MKK][6] = 127,
+	[2][1][RTW89_IC][6] = 127,
+	[2][1][RTW89_KCC][6] = 127,
+	[2][1][RTW89_ACMA][6] = 127,
+	[2][1][RTW89_CN][6] = 44,
+	[2][1][RTW89_UK][6] = 127,
+	[2][1][RTW89_FCC][7] = 127,
+	[2][1][RTW89_ETSI][7] = 127,
+	[2][1][RTW89_MKK][7] = 127,
+	[2][1][RTW89_IC][7] = 127,
+	[2][1][RTW89_KCC][7] = 127,
+	[2][1][RTW89_ACMA][7] = 127,
+	[2][1][RTW89_CN][7] = 44,
+	[2][1][RTW89_UK][7] = 127,
+	[2][1][RTW89_FCC][8] = 127,
+	[2][1][RTW89_ETSI][8] = 127,
+	[2][1][RTW89_MKK][8] = 127,
+	[2][1][RTW89_IC][8] = 127,
+	[2][1][RTW89_KCC][8] = 127,
+	[2][1][RTW89_ACMA][8] = 127,
+	[2][1][RTW89_CN][8] = 44,
+	[2][1][RTW89_UK][8] = 127,
+	[2][1][RTW89_FCC][9] = 127,
+	[2][1][RTW89_ETSI][9] = 127,
+	[2][1][RTW89_MKK][9] = 127,
+	[2][1][RTW89_IC][9] = 127,
+	[2][1][RTW89_KCC][9] = 127,
+	[2][1][RTW89_ACMA][9] = 127,
+	[2][1][RTW89_CN][9] = 44,
+	[2][1][RTW89_UK][9] = 127,
+	[2][1][RTW89_FCC][10] = 127,
+	[2][1][RTW89_ETSI][10] = 127,
+	[2][1][RTW89_MKK][10] = 127,
+	[2][1][RTW89_IC][10] = 127,
+	[2][1][RTW89_KCC][10] = 127,
+	[2][1][RTW89_ACMA][10] = 127,
+	[2][1][RTW89_CN][10] = 44,
+	[2][1][RTW89_UK][10] = 127,
+	[2][1][RTW89_FCC][11] = 127,
+	[2][1][RTW89_ETSI][11] = 127,
+	[2][1][RTW89_MKK][11] = 127,
+	[2][1][RTW89_IC][11] = 127,
+	[2][1][RTW89_KCC][11] = 127,
+	[2][1][RTW89_ACMA][11] = 127,
+	[2][1][RTW89_CN][11] = 44,
+	[2][1][RTW89_UK][11] = 127,
+	[2][1][RTW89_FCC][12] = 127,
+	[2][1][RTW89_ETSI][12] = 127,
+	[2][1][RTW89_MKK][12] = 127,
+	[2][1][RTW89_IC][12] = 127,
+	[2][1][RTW89_KCC][12] = 127,
+	[2][1][RTW89_ACMA][12] = 127,
+	[2][1][RTW89_CN][12] = 42,
+	[2][1][RTW89_UK][12] = 127,
+	[2][1][RTW89_FCC][13] = 127,
+	[2][1][RTW89_ETSI][13] = 127,
+	[2][1][RTW89_MKK][13] = 127,
+	[2][1][RTW89_IC][13] = 127,
+	[2][1][RTW89_KCC][13] = 127,
+	[2][1][RTW89_ACMA][13] = 127,
+	[2][1][RTW89_CN][13] = 127,
+	[2][1][RTW89_UK][13] = 127,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_ru_5g[RTW89_RU_NUM][RTW89_NTX_NUM]
+				    [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
+	[0][0][RTW89_WW][0] = 16,
+	[0][0][RTW89_WW][2] = 16,
+	[0][0][RTW89_WW][4] = 16,
+	[0][0][RTW89_WW][6] = 16,
+	[0][0][RTW89_WW][8] = 16,
+	[0][0][RTW89_WW][10] = 16,
+	[0][0][RTW89_WW][12] = 16,
+	[0][0][RTW89_WW][14] = 16,
+	[0][0][RTW89_WW][15] = 24,
+	[0][0][RTW89_WW][17] = 24,
+	[0][0][RTW89_WW][19] = 24,
+	[0][0][RTW89_WW][21] = 24,
+	[0][0][RTW89_WW][23] = 24,
+	[0][0][RTW89_WW][25] = 24,
+	[0][0][RTW89_WW][27] = 24,
+	[0][0][RTW89_WW][29] = 24,
+	[0][0][RTW89_WW][31] = 24,
+	[0][0][RTW89_WW][33] = 24,
+	[0][0][RTW89_WW][35] = 24,
+	[0][0][RTW89_WW][37] = 44,
+	[0][0][RTW89_WW][38] = 24,
+	[0][0][RTW89_WW][40] = 24,
+	[0][0][RTW89_WW][42] = 24,
+	[0][0][RTW89_WW][44] = 24,
+	[0][0][RTW89_WW][46] = 24,
+	[0][0][RTW89_WW][48] = 42,
+	[0][0][RTW89_WW][50] = 42,
+	[0][0][RTW89_WW][52] = 40,
+	[0][1][RTW89_WW][0] = 4,
+	[0][1][RTW89_WW][2] = 4,
+	[0][1][RTW89_WW][4] = 4,
+	[0][1][RTW89_WW][6] = 4,
+	[0][1][RTW89_WW][8] = 4,
+	[0][1][RTW89_WW][10] = 4,
+	[0][1][RTW89_WW][12] = 4,
+	[0][1][RTW89_WW][14] = 4,
+	[0][1][RTW89_WW][15] = 0,
+	[0][1][RTW89_WW][17] = 0,
+	[0][1][RTW89_WW][19] = 0,
+	[0][1][RTW89_WW][21] = 0,
+	[0][1][RTW89_WW][23] = 0,
+	[0][1][RTW89_WW][25] = 0,
+	[0][1][RTW89_WW][27] = 0,
+	[0][1][RTW89_WW][29] = 0,
+	[0][1][RTW89_WW][31] = 0,
+	[0][1][RTW89_WW][33] = 0,
+	[0][1][RTW89_WW][35] = 0,
+	[0][1][RTW89_WW][37] = 0,
+	[0][1][RTW89_WW][38] = 42,
+	[0][1][RTW89_WW][40] = 42,
+	[0][1][RTW89_WW][42] = 42,
+	[0][1][RTW89_WW][44] = 42,
+	[0][1][RTW89_WW][46] = 42,
+	[0][1][RTW89_WW][48] = 0,
+	[0][1][RTW89_WW][50] = 0,
+	[0][1][RTW89_WW][52] = 0,
+	[1][0][RTW89_WW][0] = 26,
+	[1][0][RTW89_WW][2] = 26,
+	[1][0][RTW89_WW][4] = 26,
+	[1][0][RTW89_WW][6] = 26,
+	[1][0][RTW89_WW][8] = 26,
+	[1][0][RTW89_WW][10] = 26,
+	[1][0][RTW89_WW][12] = 26,
+	[1][0][RTW89_WW][14] = 26,
+	[1][0][RTW89_WW][15] = 34,
+	[1][0][RTW89_WW][17] = 34,
+	[1][0][RTW89_WW][19] = 34,
+	[1][0][RTW89_WW][21] = 34,
+	[1][0][RTW89_WW][23] = 34,
+	[1][0][RTW89_WW][25] = 34,
+	[1][0][RTW89_WW][27] = 34,
+	[1][0][RTW89_WW][29] = 34,
+	[1][0][RTW89_WW][31] = 34,
+	[1][0][RTW89_WW][33] = 34,
+	[1][0][RTW89_WW][35] = 34,
+	[1][0][RTW89_WW][37] = 54,
+	[1][0][RTW89_WW][38] = 28,
+	[1][0][RTW89_WW][40] = 28,
+	[1][0][RTW89_WW][42] = 28,
+	[1][0][RTW89_WW][44] = 28,
+	[1][0][RTW89_WW][46] = 28,
+	[1][0][RTW89_WW][48] = 52,
+	[1][0][RTW89_WW][50] = 52,
+	[1][0][RTW89_WW][52] = 52,
+	[1][1][RTW89_WW][0] = 14,
+	[1][1][RTW89_WW][2] = 14,
+	[1][1][RTW89_WW][4] = 14,
+	[1][1][RTW89_WW][6] = 14,
+	[1][1][RTW89_WW][8] = 14,
+	[1][1][RTW89_WW][10] = 14,
+	[1][1][RTW89_WW][12] = 14,
+	[1][1][RTW89_WW][14] = 14,
+	[1][1][RTW89_WW][15] = 0,
+	[1][1][RTW89_WW][17] = 0,
+	[1][1][RTW89_WW][19] = 0,
+	[1][1][RTW89_WW][21] = 0,
+	[1][1][RTW89_WW][23] = 0,
+	[1][1][RTW89_WW][25] = 0,
+	[1][1][RTW89_WW][27] = 0,
+	[1][1][RTW89_WW][29] = 0,
+	[1][1][RTW89_WW][31] = 0,
+	[1][1][RTW89_WW][33] = 0,
+	[1][1][RTW89_WW][35] = 0,
+	[1][1][RTW89_WW][37] = 0,
+	[1][1][RTW89_WW][38] = 54,
+	[1][1][RTW89_WW][40] = 54,
+	[1][1][RTW89_WW][42] = 54,
+	[1][1][RTW89_WW][44] = 54,
+	[1][1][RTW89_WW][46] = 54,
+	[1][1][RTW89_WW][48] = 0,
+	[1][1][RTW89_WW][50] = 0,
+	[1][1][RTW89_WW][52] = 0,
+	[2][0][RTW89_WW][0] = 40,
+	[2][0][RTW89_WW][2] = 40,
+	[2][0][RTW89_WW][4] = 40,
+	[2][0][RTW89_WW][6] = 40,
+	[2][0][RTW89_WW][8] = 40,
+	[2][0][RTW89_WW][10] = 40,
+	[2][0][RTW89_WW][12] = 40,
+	[2][0][RTW89_WW][14] = 40,
+	[2][0][RTW89_WW][15] = 46,
+	[2][0][RTW89_WW][17] = 46,
+	[2][0][RTW89_WW][19] = 46,
+	[2][0][RTW89_WW][21] = 46,
+	[2][0][RTW89_WW][23] = 46,
+	[2][0][RTW89_WW][25] = 46,
+	[2][0][RTW89_WW][27] = 46,
+	[2][0][RTW89_WW][29] = 46,
+	[2][0][RTW89_WW][31] = 46,
+	[2][0][RTW89_WW][33] = 46,
+	[2][0][RTW89_WW][35] = 46,
+	[2][0][RTW89_WW][37] = 66,
+	[2][0][RTW89_WW][38] = 28,
+	[2][0][RTW89_WW][40] = 28,
+	[2][0][RTW89_WW][42] = 28,
+	[2][0][RTW89_WW][44] = 28,
+	[2][0][RTW89_WW][46] = 28,
+	[2][0][RTW89_WW][48] = 64,
+	[2][0][RTW89_WW][50] = 64,
+	[2][0][RTW89_WW][52] = 60,
+	[2][1][RTW89_WW][0] = 28,
+	[2][1][RTW89_WW][2] = 28,
+	[2][1][RTW89_WW][4] = 28,
+	[2][1][RTW89_WW][6] = 28,
+	[2][1][RTW89_WW][8] = 28,
+	[2][1][RTW89_WW][10] = 28,
+	[2][1][RTW89_WW][12] = 28,
+	[2][1][RTW89_WW][14] = 28,
+	[2][1][RTW89_WW][15] = 0,
+	[2][1][RTW89_WW][17] = 0,
+	[2][1][RTW89_WW][19] = 0,
+	[2][1][RTW89_WW][21] = 0,
+	[2][1][RTW89_WW][23] = 0,
+	[2][1][RTW89_WW][25] = 0,
+	[2][1][RTW89_WW][27] = 0,
+	[2][1][RTW89_WW][29] = 0,
+	[2][1][RTW89_WW][31] = 0,
+	[2][1][RTW89_WW][33] = 0,
+	[2][1][RTW89_WW][35] = 0,
+	[2][1][RTW89_WW][37] = 0,
+	[2][1][RTW89_WW][38] = 56,
+	[2][1][RTW89_WW][40] = 56,
+	[2][1][RTW89_WW][42] = 56,
+	[2][1][RTW89_WW][44] = 56,
+	[2][1][RTW89_WW][46] = 56,
+	[2][1][RTW89_WW][48] = 0,
+	[2][1][RTW89_WW][50] = 0,
+	[2][1][RTW89_WW][52] = 0,
+	[0][0][RTW89_FCC][0] = 52,
+	[0][0][RTW89_ETSI][0] = 24,
+	[0][0][RTW89_MKK][0] = 26,
+	[0][0][RTW89_IC][0] = 28,
+	[0][0][RTW89_KCC][0] = 42,
+	[0][0][RTW89_ACMA][0] = 24,
+	[0][0][RTW89_CN][0] = 16,
+	[0][0][RTW89_UK][0] = 24,
+	[0][0][RTW89_FCC][2] = 54,
+	[0][0][RTW89_ETSI][2] = 24,
+	[0][0][RTW89_MKK][2] = 26,
+	[0][0][RTW89_IC][2] = 28,
+	[0][0][RTW89_KCC][2] = 42,
+	[0][0][RTW89_ACMA][2] = 24,
+	[0][0][RTW89_CN][2] = 16,
+	[0][0][RTW89_UK][2] = 24,
+	[0][0][RTW89_FCC][4] = 52,
+	[0][0][RTW89_ETSI][4] = 24,
+	[0][0][RTW89_MKK][4] = 26,
+	[0][0][RTW89_IC][4] = 28,
+	[0][0][RTW89_KCC][4] = 42,
+	[0][0][RTW89_ACMA][4] = 24,
+	[0][0][RTW89_CN][4] = 16,
+	[0][0][RTW89_UK][4] = 24,
+	[0][0][RTW89_FCC][6] = 52,
+	[0][0][RTW89_ETSI][6] = 24,
+	[0][0][RTW89_MKK][6] = 26,
+	[0][0][RTW89_IC][6] = 28,
+	[0][0][RTW89_KCC][6] = 18,
+	[0][0][RTW89_ACMA][6] = 24,
+	[0][0][RTW89_CN][6] = 16,
+	[0][0][RTW89_UK][6] = 24,
+	[0][0][RTW89_FCC][8] = 52,
+	[0][0][RTW89_ETSI][8] = 24,
+	[0][0][RTW89_MKK][8] = 26,
+	[0][0][RTW89_IC][8] = 52,
+	[0][0][RTW89_KCC][8] = 42,
+	[0][0][RTW89_ACMA][8] = 24,
+	[0][0][RTW89_CN][8] = 16,
+	[0][0][RTW89_UK][8] = 24,
+	[0][0][RTW89_FCC][10] = 52,
+	[0][0][RTW89_ETSI][10] = 24,
+	[0][0][RTW89_MKK][10] = 26,
+	[0][0][RTW89_IC][10] = 52,
+	[0][0][RTW89_KCC][10] = 42,
+	[0][0][RTW89_ACMA][10] = 24,
+	[0][0][RTW89_CN][10] = 16,
+	[0][0][RTW89_UK][10] = 24,
+	[0][0][RTW89_FCC][12] = 56,
+	[0][0][RTW89_ETSI][12] = 24,
+	[0][0][RTW89_MKK][12] = 26,
+	[0][0][RTW89_IC][12] = 56,
+	[0][0][RTW89_KCC][12] = 44,
+	[0][0][RTW89_ACMA][12] = 24,
+	[0][0][RTW89_CN][12] = 16,
+	[0][0][RTW89_UK][12] = 24,
+	[0][0][RTW89_FCC][14] = 56,
+	[0][0][RTW89_ETSI][14] = 24,
+	[0][0][RTW89_MKK][14] = 26,
+	[0][0][RTW89_IC][14] = 56,
+	[0][0][RTW89_KCC][14] = 44,
+	[0][0][RTW89_ACMA][14] = 24,
+	[0][0][RTW89_CN][14] = 16,
+	[0][0][RTW89_UK][14] = 24,
+	[0][0][RTW89_FCC][15] = 54,
+	[0][0][RTW89_ETSI][15] = 24,
+	[0][0][RTW89_MKK][15] = 46,
+	[0][0][RTW89_IC][15] = 54,
+	[0][0][RTW89_KCC][15] = 44,
+	[0][0][RTW89_ACMA][15] = 24,
+	[0][0][RTW89_CN][15] = 127,
+	[0][0][RTW89_UK][15] = 24,
+	[0][0][RTW89_FCC][17] = 54,
+	[0][0][RTW89_ETSI][17] = 24,
+	[0][0][RTW89_MKK][17] = 50,
+	[0][0][RTW89_IC][17] = 54,
+	[0][0][RTW89_KCC][17] = 44,
+	[0][0][RTW89_ACMA][17] = 24,
+	[0][0][RTW89_CN][17] = 127,
+	[0][0][RTW89_UK][17] = 24,
+	[0][0][RTW89_FCC][19] = 54,
+	[0][0][RTW89_ETSI][19] = 24,
+	[0][0][RTW89_MKK][19] = 50,
+	[0][0][RTW89_IC][19] = 54,
+	[0][0][RTW89_KCC][19] = 44,
+	[0][0][RTW89_ACMA][19] = 24,
+	[0][0][RTW89_CN][19] = 127,
+	[0][0][RTW89_UK][19] = 24,
+	[0][0][RTW89_FCC][21] = 54,
+	[0][0][RTW89_ETSI][21] = 24,
+	[0][0][RTW89_MKK][21] = 50,
+	[0][0][RTW89_IC][21] = 54,
+	[0][0][RTW89_KCC][21] = 44,
+	[0][0][RTW89_ACMA][21] = 24,
+	[0][0][RTW89_CN][21] = 127,
+	[0][0][RTW89_UK][21] = 24,
+	[0][0][RTW89_FCC][23] = 54,
+	[0][0][RTW89_ETSI][23] = 24,
+	[0][0][RTW89_MKK][23] = 50,
+	[0][0][RTW89_IC][23] = 54,
+	[0][0][RTW89_KCC][23] = 44,
+	[0][0][RTW89_ACMA][23] = 24,
+	[0][0][RTW89_CN][23] = 127,
+	[0][0][RTW89_UK][23] = 24,
+	[0][0][RTW89_FCC][25] = 54,
+	[0][0][RTW89_ETSI][25] = 24,
+	[0][0][RTW89_MKK][25] = 50,
+	[0][0][RTW89_IC][25] = 127,
+	[0][0][RTW89_KCC][25] = 44,
+	[0][0][RTW89_ACMA][25] = 127,
+	[0][0][RTW89_CN][25] = 127,
+	[0][0][RTW89_UK][25] = 24,
+	[0][0][RTW89_FCC][27] = 54,
+	[0][0][RTW89_ETSI][27] = 24,
+	[0][0][RTW89_MKK][27] = 50,
+	[0][0][RTW89_IC][27] = 127,
+	[0][0][RTW89_KCC][27] = 42,
+	[0][0][RTW89_ACMA][27] = 127,
+	[0][0][RTW89_CN][27] = 127,
+	[0][0][RTW89_UK][27] = 24,
+	[0][0][RTW89_FCC][29] = 54,
+	[0][0][RTW89_ETSI][29] = 24,
+	[0][0][RTW89_MKK][29] = 50,
+	[0][0][RTW89_IC][29] = 127,
+	[0][0][RTW89_KCC][29] = 42,
+	[0][0][RTW89_ACMA][29] = 127,
+	[0][0][RTW89_CN][29] = 127,
+	[0][0][RTW89_UK][29] = 24,
+	[0][0][RTW89_FCC][31] = 54,
+	[0][0][RTW89_ETSI][31] = 24,
+	[0][0][RTW89_MKK][31] = 50,
+	[0][0][RTW89_IC][31] = 56,
+	[0][0][RTW89_KCC][31] = 42,
+	[0][0][RTW89_ACMA][31] = 24,
+	[0][0][RTW89_CN][31] = 127,
+	[0][0][RTW89_UK][31] = 24,
+	[0][0][RTW89_FCC][33] = 56,
+	[0][0][RTW89_ETSI][33] = 24,
+	[0][0][RTW89_MKK][33] = 50,
+	[0][0][RTW89_IC][33] = 56,
+	[0][0][RTW89_KCC][33] = 42,
+	[0][0][RTW89_ACMA][33] = 24,
+	[0][0][RTW89_CN][33] = 127,
+	[0][0][RTW89_UK][33] = 24,
+	[0][0][RTW89_FCC][35] = 56,
+	[0][0][RTW89_ETSI][35] = 24,
+	[0][0][RTW89_MKK][35] = 50,
+	[0][0][RTW89_IC][35] = 56,
+	[0][0][RTW89_KCC][35] = 42,
+	[0][0][RTW89_ACMA][35] = 24,
+	[0][0][RTW89_CN][35] = 127,
+	[0][0][RTW89_UK][35] = 24,
+	[0][0][RTW89_FCC][37] = 86,
+	[0][0][RTW89_ETSI][37] = 127,
+	[0][0][RTW89_MKK][37] = 46,
+	[0][0][RTW89_IC][37] = 86,
+	[0][0][RTW89_KCC][37] = 44,
+	[0][0][RTW89_ACMA][37] = 50,
+	[0][0][RTW89_CN][37] = 127,
+	[0][0][RTW89_UK][37] = 52,
+	[0][0][RTW89_FCC][38] = 68,
+	[0][0][RTW89_ETSI][38] = 28,
+	[0][0][RTW89_MKK][38] = 127,
+	[0][0][RTW89_IC][38] = 68,
+	[0][0][RTW89_KCC][38] = 44,
+	[0][0][RTW89_ACMA][38] = 84,
+	[0][0][RTW89_CN][38] = 54,
+	[0][0][RTW89_UK][38] = 24,
+	[0][0][RTW89_FCC][40] = 68,
+	[0][0][RTW89_ETSI][40] = 28,
+	[0][0][RTW89_MKK][40] = 127,
+	[0][0][RTW89_IC][40] = 68,
+	[0][0][RTW89_KCC][40] = 44,
+	[0][0][RTW89_ACMA][40] = 84,
+	[0][0][RTW89_CN][40] = 54,
+	[0][0][RTW89_UK][40] = 24,
+	[0][0][RTW89_FCC][42] = 70,
+	[0][0][RTW89_ETSI][42] = 28,
+	[0][0][RTW89_MKK][42] = 127,
+	[0][0][RTW89_IC][42] = 70,
+	[0][0][RTW89_KCC][42] = 44,
+	[0][0][RTW89_ACMA][42] = 84,
+	[0][0][RTW89_CN][42] = 54,
+	[0][0][RTW89_UK][42] = 24,
+	[0][0][RTW89_FCC][44] = 62,
+	[0][0][RTW89_ETSI][44] = 28,
+	[0][0][RTW89_MKK][44] = 127,
+	[0][0][RTW89_IC][44] = 62,
+	[0][0][RTW89_KCC][44] = 44,
+	[0][0][RTW89_ACMA][44] = 84,
+	[0][0][RTW89_CN][44] = 54,
+	[0][0][RTW89_UK][44] = 24,
+	[0][0][RTW89_FCC][46] = 62,
+	[0][0][RTW89_ETSI][46] = 28,
+	[0][0][RTW89_MKK][46] = 127,
+	[0][0][RTW89_IC][46] = 62,
+	[0][0][RTW89_KCC][46] = 44,
+	[0][0][RTW89_ACMA][46] = 84,
+	[0][0][RTW89_CN][46] = 54,
+	[0][0][RTW89_UK][46] = 24,
+	[0][0][RTW89_FCC][48] = 42,
+	[0][0][RTW89_ETSI][48] = 127,
+	[0][0][RTW89_MKK][48] = 127,
+	[0][0][RTW89_IC][48] = 127,
+	[0][0][RTW89_KCC][48] = 127,
+	[0][0][RTW89_ACMA][48] = 127,
+	[0][0][RTW89_CN][48] = 127,
+	[0][0][RTW89_UK][48] = 127,
+	[0][0][RTW89_FCC][50] = 42,
+	[0][0][RTW89_ETSI][50] = 127,
+	[0][0][RTW89_MKK][50] = 127,
+	[0][0][RTW89_IC][50] = 127,
+	[0][0][RTW89_KCC][50] = 127,
+	[0][0][RTW89_ACMA][50] = 127,
+	[0][0][RTW89_CN][50] = 127,
+	[0][0][RTW89_UK][50] = 127,
+	[0][0][RTW89_FCC][52] = 40,
+	[0][0][RTW89_ETSI][52] = 127,
+	[0][0][RTW89_MKK][52] = 127,
+	[0][0][RTW89_IC][52] = 127,
+	[0][0][RTW89_KCC][52] = 127,
+	[0][0][RTW89_ACMA][52] = 127,
+	[0][0][RTW89_CN][52] = 127,
+	[0][0][RTW89_UK][52] = 127,
+	[0][1][RTW89_FCC][0] = 127,
+	[0][1][RTW89_ETSI][0] = 127,
+	[0][1][RTW89_MKK][0] = 127,
+	[0][1][RTW89_IC][0] = 127,
+	[0][1][RTW89_KCC][0] = 127,
+	[0][1][RTW89_ACMA][0] = 127,
+	[0][1][RTW89_CN][0] = 4,
+	[0][1][RTW89_UK][0] = 127,
+	[0][1][RTW89_FCC][2] = 127,
+	[0][1][RTW89_ETSI][2] = 127,
+	[0][1][RTW89_MKK][2] = 127,
+	[0][1][RTW89_IC][2] = 127,
+	[0][1][RTW89_KCC][2] = 127,
+	[0][1][RTW89_ACMA][2] = 127,
+	[0][1][RTW89_CN][2] = 4,
+	[0][1][RTW89_UK][2] = 127,
+	[0][1][RTW89_FCC][4] = 127,
+	[0][1][RTW89_ETSI][4] = 127,
+	[0][1][RTW89_MKK][4] = 127,
+	[0][1][RTW89_IC][4] = 127,
+	[0][1][RTW89_KCC][4] = 127,
+	[0][1][RTW89_ACMA][4] = 127,
+	[0][1][RTW89_CN][4] = 4,
+	[0][1][RTW89_UK][4] = 127,
+	[0][1][RTW89_FCC][6] = 127,
+	[0][1][RTW89_ETSI][6] = 127,
+	[0][1][RTW89_MKK][6] = 127,
+	[0][1][RTW89_IC][6] = 127,
+	[0][1][RTW89_KCC][6] = 127,
+	[0][1][RTW89_ACMA][6] = 127,
+	[0][1][RTW89_CN][6] = 4,
+	[0][1][RTW89_UK][6] = 127,
+	[0][1][RTW89_FCC][8] = 127,
+	[0][1][RTW89_ETSI][8] = 127,
+	[0][1][RTW89_MKK][8] = 127,
+	[0][1][RTW89_IC][8] = 127,
+	[0][1][RTW89_KCC][8] = 127,
+	[0][1][RTW89_ACMA][8] = 127,
+	[0][1][RTW89_CN][8] = 4,
+	[0][1][RTW89_UK][8] = 127,
+	[0][1][RTW89_FCC][10] = 127,
+	[0][1][RTW89_ETSI][10] = 127,
+	[0][1][RTW89_MKK][10] = 127,
+	[0][1][RTW89_IC][10] = 127,
+	[0][1][RTW89_KCC][10] = 127,
+	[0][1][RTW89_ACMA][10] = 127,
+	[0][1][RTW89_CN][10] = 4,
+	[0][1][RTW89_UK][10] = 127,
+	[0][1][RTW89_FCC][12] = 127,
+	[0][1][RTW89_ETSI][12] = 127,
+	[0][1][RTW89_MKK][12] = 127,
+	[0][1][RTW89_IC][12] = 127,
+	[0][1][RTW89_KCC][12] = 127,
+	[0][1][RTW89_ACMA][12] = 127,
+	[0][1][RTW89_CN][12] = 4,
+	[0][1][RTW89_UK][12] = 127,
+	[0][1][RTW89_FCC][14] = 127,
+	[0][1][RTW89_ETSI][14] = 127,
+	[0][1][RTW89_MKK][14] = 127,
+	[0][1][RTW89_IC][14] = 127,
+	[0][1][RTW89_KCC][14] = 127,
+	[0][1][RTW89_ACMA][14] = 127,
+	[0][1][RTW89_CN][14] = 4,
+	[0][1][RTW89_UK][14] = 127,
+	[0][1][RTW89_FCC][15] = 127,
+	[0][1][RTW89_ETSI][15] = 127,
+	[0][1][RTW89_MKK][15] = 127,
+	[0][1][RTW89_IC][15] = 127,
+	[0][1][RTW89_KCC][15] = 127,
+	[0][1][RTW89_ACMA][15] = 127,
+	[0][1][RTW89_CN][15] = 127,
+	[0][1][RTW89_UK][15] = 127,
+	[0][1][RTW89_FCC][17] = 127,
+	[0][1][RTW89_ETSI][17] = 127,
+	[0][1][RTW89_MKK][17] = 127,
+	[0][1][RTW89_IC][17] = 127,
+	[0][1][RTW89_KCC][17] = 127,
+	[0][1][RTW89_ACMA][17] = 127,
+	[0][1][RTW89_CN][17] = 127,
+	[0][1][RTW89_UK][17] = 127,
+	[0][1][RTW89_FCC][19] = 127,
+	[0][1][RTW89_ETSI][19] = 127,
+	[0][1][RTW89_MKK][19] = 127,
+	[0][1][RTW89_IC][19] = 127,
+	[0][1][RTW89_KCC][19] = 127,
+	[0][1][RTW89_ACMA][19] = 127,
+	[0][1][RTW89_CN][19] = 127,
+	[0][1][RTW89_UK][19] = 127,
+	[0][1][RTW89_FCC][21] = 127,
+	[0][1][RTW89_ETSI][21] = 127,
+	[0][1][RTW89_MKK][21] = 127,
+	[0][1][RTW89_IC][21] = 127,
+	[0][1][RTW89_KCC][21] = 127,
+	[0][1][RTW89_ACMA][21] = 127,
+	[0][1][RTW89_CN][21] = 127,
+	[0][1][RTW89_UK][21] = 127,
+	[0][1][RTW89_FCC][23] = 127,
+	[0][1][RTW89_ETSI][23] = 127,
+	[0][1][RTW89_MKK][23] = 127,
+	[0][1][RTW89_IC][23] = 127,
+	[0][1][RTW89_KCC][23] = 127,
+	[0][1][RTW89_ACMA][23] = 127,
+	[0][1][RTW89_CN][23] = 127,
+	[0][1][RTW89_UK][23] = 127,
+	[0][1][RTW89_FCC][25] = 127,
+	[0][1][RTW89_ETSI][25] = 127,
+	[0][1][RTW89_MKK][25] = 127,
+	[0][1][RTW89_IC][25] = 127,
+	[0][1][RTW89_KCC][25] = 127,
+	[0][1][RTW89_ACMA][25] = 127,
+	[0][1][RTW89_CN][25] = 127,
+	[0][1][RTW89_UK][25] = 127,
+	[0][1][RTW89_FCC][27] = 127,
+	[0][1][RTW89_ETSI][27] = 127,
+	[0][1][RTW89_MKK][27] = 127,
+	[0][1][RTW89_IC][27] = 127,
+	[0][1][RTW89_KCC][27] = 127,
+	[0][1][RTW89_ACMA][27] = 127,
+	[0][1][RTW89_CN][27] = 127,
+	[0][1][RTW89_UK][27] = 127,
+	[0][1][RTW89_FCC][29] = 127,
+	[0][1][RTW89_ETSI][29] = 127,
+	[0][1][RTW89_MKK][29] = 127,
+	[0][1][RTW89_IC][29] = 127,
+	[0][1][RTW89_KCC][29] = 127,
+	[0][1][RTW89_ACMA][29] = 127,
+	[0][1][RTW89_CN][29] = 127,
+	[0][1][RTW89_UK][29] = 127,
+	[0][1][RTW89_FCC][31] = 127,
+	[0][1][RTW89_ETSI][31] = 127,
+	[0][1][RTW89_MKK][31] = 127,
+	[0][1][RTW89_IC][31] = 127,
+	[0][1][RTW89_KCC][31] = 127,
+	[0][1][RTW89_ACMA][31] = 127,
+	[0][1][RTW89_CN][31] = 127,
+	[0][1][RTW89_UK][31] = 127,
+	[0][1][RTW89_FCC][33] = 127,
+	[0][1][RTW89_ETSI][33] = 127,
+	[0][1][RTW89_MKK][33] = 127,
+	[0][1][RTW89_IC][33] = 127,
+	[0][1][RTW89_KCC][33] = 127,
+	[0][1][RTW89_ACMA][33] = 127,
+	[0][1][RTW89_CN][33] = 127,
+	[0][1][RTW89_UK][33] = 127,
+	[0][1][RTW89_FCC][35] = 127,
+	[0][1][RTW89_ETSI][35] = 127,
+	[0][1][RTW89_MKK][35] = 127,
+	[0][1][RTW89_IC][35] = 127,
+	[0][1][RTW89_KCC][35] = 127,
+	[0][1][RTW89_ACMA][35] = 127,
+	[0][1][RTW89_CN][35] = 127,
+	[0][1][RTW89_UK][35] = 127,
+	[0][1][RTW89_FCC][37] = 127,
+	[0][1][RTW89_ETSI][37] = 127,
+	[0][1][RTW89_MKK][37] = 127,
+	[0][1][RTW89_IC][37] = 127,
+	[0][1][RTW89_KCC][37] = 127,
+	[0][1][RTW89_ACMA][37] = 127,
+	[0][1][RTW89_CN][37] = 127,
+	[0][1][RTW89_UK][37] = 127,
+	[0][1][RTW89_FCC][38] = 127,
+	[0][1][RTW89_ETSI][38] = 127,
+	[0][1][RTW89_MKK][38] = 127,
+	[0][1][RTW89_IC][38] = 127,
+	[0][1][RTW89_KCC][38] = 127,
+	[0][1][RTW89_ACMA][38] = 127,
+	[0][1][RTW89_CN][38] = 42,
+	[0][1][RTW89_UK][38] = 127,
+	[0][1][RTW89_FCC][40] = 127,
+	[0][1][RTW89_ETSI][40] = 127,
+	[0][1][RTW89_MKK][40] = 127,
+	[0][1][RTW89_IC][40] = 127,
+	[0][1][RTW89_KCC][40] = 127,
+	[0][1][RTW89_ACMA][40] = 127,
+	[0][1][RTW89_CN][40] = 42,
+	[0][1][RTW89_UK][40] = 127,
+	[0][1][RTW89_FCC][42] = 127,
+	[0][1][RTW89_ETSI][42] = 127,
+	[0][1][RTW89_MKK][42] = 127,
+	[0][1][RTW89_IC][42] = 127,
+	[0][1][RTW89_KCC][42] = 127,
+	[0][1][RTW89_ACMA][42] = 127,
+	[0][1][RTW89_CN][42] = 42,
+	[0][1][RTW89_UK][42] = 127,
+	[0][1][RTW89_FCC][44] = 127,
+	[0][1][RTW89_ETSI][44] = 127,
+	[0][1][RTW89_MKK][44] = 127,
+	[0][1][RTW89_IC][44] = 127,
+	[0][1][RTW89_KCC][44] = 127,
+	[0][1][RTW89_ACMA][44] = 127,
+	[0][1][RTW89_CN][44] = 42,
+	[0][1][RTW89_UK][44] = 127,
+	[0][1][RTW89_FCC][46] = 127,
+	[0][1][RTW89_ETSI][46] = 127,
+	[0][1][RTW89_MKK][46] = 127,
+	[0][1][RTW89_IC][46] = 127,
+	[0][1][RTW89_KCC][46] = 127,
+	[0][1][RTW89_ACMA][46] = 127,
+	[0][1][RTW89_CN][46] = 42,
+	[0][1][RTW89_UK][46] = 127,
+	[0][1][RTW89_FCC][48] = 127,
+	[0][1][RTW89_ETSI][48] = 127,
+	[0][1][RTW89_MKK][48] = 127,
+	[0][1][RTW89_IC][48] = 127,
+	[0][1][RTW89_KCC][48] = 127,
+	[0][1][RTW89_ACMA][48] = 127,
+	[0][1][RTW89_CN][48] = 127,
+	[0][1][RTW89_UK][48] = 127,
+	[0][1][RTW89_FCC][50] = 127,
+	[0][1][RTW89_ETSI][50] = 127,
+	[0][1][RTW89_MKK][50] = 127,
+	[0][1][RTW89_IC][50] = 127,
+	[0][1][RTW89_KCC][50] = 127,
+	[0][1][RTW89_ACMA][50] = 127,
+	[0][1][RTW89_CN][50] = 127,
+	[0][1][RTW89_UK][50] = 127,
+	[0][1][RTW89_FCC][52] = 127,
+	[0][1][RTW89_ETSI][52] = 127,
+	[0][1][RTW89_MKK][52] = 127,
+	[0][1][RTW89_IC][52] = 127,
+	[0][1][RTW89_KCC][52] = 127,
+	[0][1][RTW89_ACMA][52] = 127,
+	[0][1][RTW89_CN][52] = 127,
+	[0][1][RTW89_UK][52] = 127,
+	[1][0][RTW89_FCC][0] = 64,
+	[1][0][RTW89_ETSI][0] = 34,
+	[1][0][RTW89_MKK][0] = 38,
+	[1][0][RTW89_IC][0] = 38,
+	[1][0][RTW89_KCC][0] = 52,
+	[1][0][RTW89_ACMA][0] = 34,
+	[1][0][RTW89_CN][0] = 26,
+	[1][0][RTW89_UK][0] = 34,
+	[1][0][RTW89_FCC][2] = 66,
+	[1][0][RTW89_ETSI][2] = 34,
+	[1][0][RTW89_MKK][2] = 38,
+	[1][0][RTW89_IC][2] = 38,
+	[1][0][RTW89_KCC][2] = 52,
+	[1][0][RTW89_ACMA][2] = 34,
+	[1][0][RTW89_CN][2] = 26,
+	[1][0][RTW89_UK][2] = 34,
+	[1][0][RTW89_FCC][4] = 62,
+	[1][0][RTW89_ETSI][4] = 34,
+	[1][0][RTW89_MKK][4] = 36,
+	[1][0][RTW89_IC][4] = 38,
+	[1][0][RTW89_KCC][4] = 52,
+	[1][0][RTW89_ACMA][4] = 34,
+	[1][0][RTW89_CN][4] = 26,
+	[1][0][RTW89_UK][4] = 34,
+	[1][0][RTW89_FCC][6] = 62,
+	[1][0][RTW89_ETSI][6] = 34,
+	[1][0][RTW89_MKK][6] = 36,
+	[1][0][RTW89_IC][6] = 38,
+	[1][0][RTW89_KCC][6] = 32,
+	[1][0][RTW89_ACMA][6] = 34,
+	[1][0][RTW89_CN][6] = 26,
+	[1][0][RTW89_UK][6] = 34,
+	[1][0][RTW89_FCC][8] = 62,
+	[1][0][RTW89_ETSI][8] = 34,
+	[1][0][RTW89_MKK][8] = 38,
+	[1][0][RTW89_IC][8] = 62,
+	[1][0][RTW89_KCC][8] = 52,
+	[1][0][RTW89_ACMA][8] = 34,
+	[1][0][RTW89_CN][8] = 26,
+	[1][0][RTW89_UK][8] = 34,
+	[1][0][RTW89_FCC][10] = 62,
+	[1][0][RTW89_ETSI][10] = 34,
+	[1][0][RTW89_MKK][10] = 38,
+	[1][0][RTW89_IC][10] = 62,
+	[1][0][RTW89_KCC][10] = 52,
+	[1][0][RTW89_ACMA][10] = 34,
+	[1][0][RTW89_CN][10] = 26,
+	[1][0][RTW89_UK][10] = 34,
+	[1][0][RTW89_FCC][12] = 62,
+	[1][0][RTW89_ETSI][12] = 34,
+	[1][0][RTW89_MKK][12] = 38,
+	[1][0][RTW89_IC][12] = 62,
+	[1][0][RTW89_KCC][12] = 54,
+	[1][0][RTW89_ACMA][12] = 34,
+	[1][0][RTW89_CN][12] = 26,
+	[1][0][RTW89_UK][12] = 34,
+	[1][0][RTW89_FCC][14] = 64,
+	[1][0][RTW89_ETSI][14] = 34,
+	[1][0][RTW89_MKK][14] = 38,
+	[1][0][RTW89_IC][14] = 64,
+	[1][0][RTW89_KCC][14] = 54,
+	[1][0][RTW89_ACMA][14] = 34,
+	[1][0][RTW89_CN][14] = 26,
+	[1][0][RTW89_UK][14] = 34,
+	[1][0][RTW89_FCC][15] = 62,
+	[1][0][RTW89_ETSI][15] = 34,
+	[1][0][RTW89_MKK][15] = 58,
+	[1][0][RTW89_IC][15] = 62,
+	[1][0][RTW89_KCC][15] = 54,
+	[1][0][RTW89_ACMA][15] = 34,
+	[1][0][RTW89_CN][15] = 127,
+	[1][0][RTW89_UK][15] = 34,
+	[1][0][RTW89_FCC][17] = 62,
+	[1][0][RTW89_ETSI][17] = 34,
+	[1][0][RTW89_MKK][17] = 58,
+	[1][0][RTW89_IC][17] = 62,
+	[1][0][RTW89_KCC][17] = 54,
+	[1][0][RTW89_ACMA][17] = 34,
+	[1][0][RTW89_CN][17] = 127,
+	[1][0][RTW89_UK][17] = 34,
+	[1][0][RTW89_FCC][19] = 64,
+	[1][0][RTW89_ETSI][19] = 34,
+	[1][0][RTW89_MKK][19] = 58,
+	[1][0][RTW89_IC][19] = 64,
+	[1][0][RTW89_KCC][19] = 54,
+	[1][0][RTW89_ACMA][19] = 34,
+	[1][0][RTW89_CN][19] = 127,
+	[1][0][RTW89_UK][19] = 34,
+	[1][0][RTW89_FCC][21] = 64,
+	[1][0][RTW89_ETSI][21] = 34,
+	[1][0][RTW89_MKK][21] = 58,
+	[1][0][RTW89_IC][21] = 64,
+	[1][0][RTW89_KCC][21] = 54,
+	[1][0][RTW89_ACMA][21] = 34,
+	[1][0][RTW89_CN][21] = 127,
+	[1][0][RTW89_UK][21] = 34,
+	[1][0][RTW89_FCC][23] = 64,
+	[1][0][RTW89_ETSI][23] = 34,
+	[1][0][RTW89_MKK][23] = 58,
+	[1][0][RTW89_IC][23] = 64,
+	[1][0][RTW89_KCC][23] = 54,
+	[1][0][RTW89_ACMA][23] = 34,
+	[1][0][RTW89_CN][23] = 127,
+	[1][0][RTW89_UK][23] = 34,
+	[1][0][RTW89_FCC][25] = 64,
+	[1][0][RTW89_ETSI][25] = 34,
+	[1][0][RTW89_MKK][25] = 58,
+	[1][0][RTW89_IC][25] = 127,
+	[1][0][RTW89_KCC][25] = 54,
+	[1][0][RTW89_ACMA][25] = 127,
+	[1][0][RTW89_CN][25] = 127,
+	[1][0][RTW89_UK][25] = 34,
+	[1][0][RTW89_FCC][27] = 64,
+	[1][0][RTW89_ETSI][27] = 34,
+	[1][0][RTW89_MKK][27] = 58,
+	[1][0][RTW89_IC][27] = 127,
+	[1][0][RTW89_KCC][27] = 54,
+	[1][0][RTW89_ACMA][27] = 127,
+	[1][0][RTW89_CN][27] = 127,
+	[1][0][RTW89_UK][27] = 34,
+	[1][0][RTW89_FCC][29] = 64,
+	[1][0][RTW89_ETSI][29] = 34,
+	[1][0][RTW89_MKK][29] = 58,
+	[1][0][RTW89_IC][29] = 127,
+	[1][0][RTW89_KCC][29] = 54,
+	[1][0][RTW89_ACMA][29] = 127,
+	[1][0][RTW89_CN][29] = 127,
+	[1][0][RTW89_UK][29] = 34,
+	[1][0][RTW89_FCC][31] = 64,
+	[1][0][RTW89_ETSI][31] = 34,
+	[1][0][RTW89_MKK][31] = 58,
+	[1][0][RTW89_IC][31] = 64,
+	[1][0][RTW89_KCC][31] = 54,
+	[1][0][RTW89_ACMA][31] = 34,
+	[1][0][RTW89_CN][31] = 127,
+	[1][0][RTW89_UK][31] = 34,
+	[1][0][RTW89_FCC][33] = 64,
+	[1][0][RTW89_ETSI][33] = 34,
+	[1][0][RTW89_MKK][33] = 58,
+	[1][0][RTW89_IC][33] = 64,
+	[1][0][RTW89_KCC][33] = 54,
+	[1][0][RTW89_ACMA][33] = 34,
+	[1][0][RTW89_CN][33] = 127,
+	[1][0][RTW89_UK][33] = 34,
+	[1][0][RTW89_FCC][35] = 64,
+	[1][0][RTW89_ETSI][35] = 34,
+	[1][0][RTW89_MKK][35] = 58,
+	[1][0][RTW89_IC][35] = 64,
+	[1][0][RTW89_KCC][35] = 54,
+	[1][0][RTW89_ACMA][35] = 34,
+	[1][0][RTW89_CN][35] = 127,
+	[1][0][RTW89_UK][35] = 34,
+	[1][0][RTW89_FCC][37] = 78,
+	[1][0][RTW89_ETSI][37] = 127,
+	[1][0][RTW89_MKK][37] = 56,
+	[1][0][RTW89_IC][37] = 78,
+	[1][0][RTW89_KCC][37] = 54,
+	[1][0][RTW89_ACMA][37] = 62,
+	[1][0][RTW89_CN][37] = 127,
+	[1][0][RTW89_UK][37] = 62,
+	[1][0][RTW89_FCC][38] = 82,
+	[1][0][RTW89_ETSI][38] = 28,
+	[1][0][RTW89_MKK][38] = 127,
+	[1][0][RTW89_IC][38] = 82,
+	[1][0][RTW89_KCC][38] = 54,
+	[1][0][RTW89_ACMA][38] = 84,
+	[1][0][RTW89_CN][38] = 66,
+	[1][0][RTW89_UK][38] = 34,
+	[1][0][RTW89_FCC][40] = 82,
+	[1][0][RTW89_ETSI][40] = 28,
+	[1][0][RTW89_MKK][40] = 127,
+	[1][0][RTW89_IC][40] = 82,
+	[1][0][RTW89_KCC][40] = 54,
+	[1][0][RTW89_ACMA][40] = 84,
+	[1][0][RTW89_CN][40] = 66,
+	[1][0][RTW89_UK][40] = 34,
+	[1][0][RTW89_FCC][42] = 78,
+	[1][0][RTW89_ETSI][42] = 28,
+	[1][0][RTW89_MKK][42] = 127,
+	[1][0][RTW89_IC][42] = 78,
+	[1][0][RTW89_KCC][42] = 54,
+	[1][0][RTW89_ACMA][42] = 84,
+	[1][0][RTW89_CN][42] = 66,
+	[1][0][RTW89_UK][42] = 34,
+	[1][0][RTW89_FCC][44] = 82,
+	[1][0][RTW89_ETSI][44] = 28,
+	[1][0][RTW89_MKK][44] = 127,
+	[1][0][RTW89_IC][44] = 82,
+	[1][0][RTW89_KCC][44] = 54,
+	[1][0][RTW89_ACMA][44] = 84,
+	[1][0][RTW89_CN][44] = 66,
+	[1][0][RTW89_UK][44] = 34,
+	[1][0][RTW89_FCC][46] = 82,
+	[1][0][RTW89_ETSI][46] = 28,
+	[1][0][RTW89_MKK][46] = 127,
+	[1][0][RTW89_IC][46] = 82,
+	[1][0][RTW89_KCC][46] = 54,
+	[1][0][RTW89_ACMA][46] = 84,
+	[1][0][RTW89_CN][46] = 66,
+	[1][0][RTW89_UK][46] = 34,
+	[1][0][RTW89_FCC][48] = 52,
+	[1][0][RTW89_ETSI][48] = 127,
+	[1][0][RTW89_MKK][48] = 127,
+	[1][0][RTW89_IC][48] = 127,
+	[1][0][RTW89_KCC][48] = 127,
+	[1][0][RTW89_ACMA][48] = 127,
+	[1][0][RTW89_CN][48] = 127,
+	[1][0][RTW89_UK][48] = 127,
+	[1][0][RTW89_FCC][50] = 52,
+	[1][0][RTW89_ETSI][50] = 127,
+	[1][0][RTW89_MKK][50] = 127,
+	[1][0][RTW89_IC][50] = 127,
+	[1][0][RTW89_KCC][50] = 127,
+	[1][0][RTW89_ACMA][50] = 127,
+	[1][0][RTW89_CN][50] = 127,
+	[1][0][RTW89_UK][50] = 127,
+	[1][0][RTW89_FCC][52] = 52,
+	[1][0][RTW89_ETSI][52] = 127,
+	[1][0][RTW89_MKK][52] = 127,
+	[1][0][RTW89_IC][52] = 127,
+	[1][0][RTW89_KCC][52] = 127,
+	[1][0][RTW89_ACMA][52] = 127,
+	[1][0][RTW89_CN][52] = 127,
+	[1][0][RTW89_UK][52] = 127,
+	[1][1][RTW89_FCC][0] = 127,
+	[1][1][RTW89_ETSI][0] = 127,
+	[1][1][RTW89_MKK][0] = 127,
+	[1][1][RTW89_IC][0] = 127,
+	[1][1][RTW89_KCC][0] = 127,
+	[1][1][RTW89_ACMA][0] = 127,
+	[1][1][RTW89_CN][0] = 14,
+	[1][1][RTW89_UK][0] = 127,
+	[1][1][RTW89_FCC][2] = 127,
+	[1][1][RTW89_ETSI][2] = 127,
+	[1][1][RTW89_MKK][2] = 127,
+	[1][1][RTW89_IC][2] = 127,
+	[1][1][RTW89_KCC][2] = 127,
+	[1][1][RTW89_ACMA][2] = 127,
+	[1][1][RTW89_CN][2] = 14,
+	[1][1][RTW89_UK][2] = 127,
+	[1][1][RTW89_FCC][4] = 127,
+	[1][1][RTW89_ETSI][4] = 127,
+	[1][1][RTW89_MKK][4] = 127,
+	[1][1][RTW89_IC][4] = 127,
+	[1][1][RTW89_KCC][4] = 127,
+	[1][1][RTW89_ACMA][4] = 127,
+	[1][1][RTW89_CN][4] = 14,
+	[1][1][RTW89_UK][4] = 127,
+	[1][1][RTW89_FCC][6] = 127,
+	[1][1][RTW89_ETSI][6] = 127,
+	[1][1][RTW89_MKK][6] = 127,
+	[1][1][RTW89_IC][6] = 127,
+	[1][1][RTW89_KCC][6] = 127,
+	[1][1][RTW89_ACMA][6] = 127,
+	[1][1][RTW89_CN][6] = 14,
+	[1][1][RTW89_UK][6] = 127,
+	[1][1][RTW89_FCC][8] = 127,
+	[1][1][RTW89_ETSI][8] = 127,
+	[1][1][RTW89_MKK][8] = 127,
+	[1][1][RTW89_IC][8] = 127,
+	[1][1][RTW89_KCC][8] = 127,
+	[1][1][RTW89_ACMA][8] = 127,
+	[1][1][RTW89_CN][8] = 14,
+	[1][1][RTW89_UK][8] = 127,
+	[1][1][RTW89_FCC][10] = 127,
+	[1][1][RTW89_ETSI][10] = 127,
+	[1][1][RTW89_MKK][10] = 127,
+	[1][1][RTW89_IC][10] = 127,
+	[1][1][RTW89_KCC][10] = 127,
+	[1][1][RTW89_ACMA][10] = 127,
+	[1][1][RTW89_CN][10] = 14,
+	[1][1][RTW89_UK][10] = 127,
+	[1][1][RTW89_FCC][12] = 127,
+	[1][1][RTW89_ETSI][12] = 127,
+	[1][1][RTW89_MKK][12] = 127,
+	[1][1][RTW89_IC][12] = 127,
+	[1][1][RTW89_KCC][12] = 127,
+	[1][1][RTW89_ACMA][12] = 127,
+	[1][1][RTW89_CN][12] = 14,
+	[1][1][RTW89_UK][12] = 127,
+	[1][1][RTW89_FCC][14] = 127,
+	[1][1][RTW89_ETSI][14] = 127,
+	[1][1][RTW89_MKK][14] = 127,
+	[1][1][RTW89_IC][14] = 127,
+	[1][1][RTW89_KCC][14] = 127,
+	[1][1][RTW89_ACMA][14] = 127,
+	[1][1][RTW89_CN][14] = 14,
+	[1][1][RTW89_UK][14] = 127,
+	[1][1][RTW89_FCC][15] = 127,
+	[1][1][RTW89_ETSI][15] = 127,
+	[1][1][RTW89_MKK][15] = 127,
+	[1][1][RTW89_IC][15] = 127,
+	[1][1][RTW89_KCC][15] = 127,
+	[1][1][RTW89_ACMA][15] = 127,
+	[1][1][RTW89_CN][15] = 127,
+	[1][1][RTW89_UK][15] = 127,
+	[1][1][RTW89_FCC][17] = 127,
+	[1][1][RTW89_ETSI][17] = 127,
+	[1][1][RTW89_MKK][17] = 127,
+	[1][1][RTW89_IC][17] = 127,
+	[1][1][RTW89_KCC][17] = 127,
+	[1][1][RTW89_ACMA][17] = 127,
+	[1][1][RTW89_CN][17] = 127,
+	[1][1][RTW89_UK][17] = 127,
+	[1][1][RTW89_FCC][19] = 127,
+	[1][1][RTW89_ETSI][19] = 127,
+	[1][1][RTW89_MKK][19] = 127,
+	[1][1][RTW89_IC][19] = 127,
+	[1][1][RTW89_KCC][19] = 127,
+	[1][1][RTW89_ACMA][19] = 127,
+	[1][1][RTW89_CN][19] = 127,
+	[1][1][RTW89_UK][19] = 127,
+	[1][1][RTW89_FCC][21] = 127,
+	[1][1][RTW89_ETSI][21] = 127,
+	[1][1][RTW89_MKK][21] = 127,
+	[1][1][RTW89_IC][21] = 127,
+	[1][1][RTW89_KCC][21] = 127,
+	[1][1][RTW89_ACMA][21] = 127,
+	[1][1][RTW89_CN][21] = 127,
+	[1][1][RTW89_UK][21] = 127,
+	[1][1][RTW89_FCC][23] = 127,
+	[1][1][RTW89_ETSI][23] = 127,
+	[1][1][RTW89_MKK][23] = 127,
+	[1][1][RTW89_IC][23] = 127,
+	[1][1][RTW89_KCC][23] = 127,
+	[1][1][RTW89_ACMA][23] = 127,
+	[1][1][RTW89_CN][23] = 127,
+	[1][1][RTW89_UK][23] = 127,
+	[1][1][RTW89_FCC][25] = 127,
+	[1][1][RTW89_ETSI][25] = 127,
+	[1][1][RTW89_MKK][25] = 127,
+	[1][1][RTW89_IC][25] = 127,
+	[1][1][RTW89_KCC][25] = 127,
+	[1][1][RTW89_ACMA][25] = 127,
+	[1][1][RTW89_CN][25] = 127,
+	[1][1][RTW89_UK][25] = 127,
+	[1][1][RTW89_FCC][27] = 127,
+	[1][1][RTW89_ETSI][27] = 127,
+	[1][1][RTW89_MKK][27] = 127,
+	[1][1][RTW89_IC][27] = 127,
+	[1][1][RTW89_KCC][27] = 127,
+	[1][1][RTW89_ACMA][27] = 127,
+	[1][1][RTW89_CN][27] = 127,
+	[1][1][RTW89_UK][27] = 127,
+	[1][1][RTW89_FCC][29] = 127,
+	[1][1][RTW89_ETSI][29] = 127,
+	[1][1][RTW89_MKK][29] = 127,
+	[1][1][RTW89_IC][29] = 127,
+	[1][1][RTW89_KCC][29] = 127,
+	[1][1][RTW89_ACMA][29] = 127,
+	[1][1][RTW89_CN][29] = 127,
+	[1][1][RTW89_UK][29] = 127,
+	[1][1][RTW89_FCC][31] = 127,
+	[1][1][RTW89_ETSI][31] = 127,
+	[1][1][RTW89_MKK][31] = 127,
+	[1][1][RTW89_IC][31] = 127,
+	[1][1][RTW89_KCC][31] = 127,
+	[1][1][RTW89_ACMA][31] = 127,
+	[1][1][RTW89_CN][31] = 127,
+	[1][1][RTW89_UK][31] = 127,
+	[1][1][RTW89_FCC][33] = 127,
+	[1][1][RTW89_ETSI][33] = 127,
+	[1][1][RTW89_MKK][33] = 127,
+	[1][1][RTW89_IC][33] = 127,
+	[1][1][RTW89_KCC][33] = 127,
+	[1][1][RTW89_ACMA][33] = 127,
+	[1][1][RTW89_CN][33] = 127,
+	[1][1][RTW89_UK][33] = 127,
+	[1][1][RTW89_FCC][35] = 127,
+	[1][1][RTW89_ETSI][35] = 127,
+	[1][1][RTW89_MKK][35] = 127,
+	[1][1][RTW89_IC][35] = 127,
+	[1][1][RTW89_KCC][35] = 127,
+	[1][1][RTW89_ACMA][35] = 127,
+	[1][1][RTW89_CN][35] = 127,
+	[1][1][RTW89_UK][35] = 127,
+	[1][1][RTW89_FCC][37] = 127,
+	[1][1][RTW89_ETSI][37] = 127,
+	[1][1][RTW89_MKK][37] = 127,
+	[1][1][RTW89_IC][37] = 127,
+	[1][1][RTW89_KCC][37] = 127,
+	[1][1][RTW89_ACMA][37] = 127,
+	[1][1][RTW89_CN][37] = 127,
+	[1][1][RTW89_UK][37] = 127,
+	[1][1][RTW89_FCC][38] = 127,
+	[1][1][RTW89_ETSI][38] = 127,
+	[1][1][RTW89_MKK][38] = 127,
+	[1][1][RTW89_IC][38] = 127,
+	[1][1][RTW89_KCC][38] = 127,
+	[1][1][RTW89_ACMA][38] = 127,
+	[1][1][RTW89_CN][38] = 54,
+	[1][1][RTW89_UK][38] = 127,
+	[1][1][RTW89_FCC][40] = 127,
+	[1][1][RTW89_ETSI][40] = 127,
+	[1][1][RTW89_MKK][40] = 127,
+	[1][1][RTW89_IC][40] = 127,
+	[1][1][RTW89_KCC][40] = 127,
+	[1][1][RTW89_ACMA][40] = 127,
+	[1][1][RTW89_CN][40] = 54,
+	[1][1][RTW89_UK][40] = 127,
+	[1][1][RTW89_FCC][42] = 127,
+	[1][1][RTW89_ETSI][42] = 127,
+	[1][1][RTW89_MKK][42] = 127,
+	[1][1][RTW89_IC][42] = 127,
+	[1][1][RTW89_KCC][42] = 127,
+	[1][1][RTW89_ACMA][42] = 127,
+	[1][1][RTW89_CN][42] = 54,
+	[1][1][RTW89_UK][42] = 127,
+	[1][1][RTW89_FCC][44] = 127,
+	[1][1][RTW89_ETSI][44] = 127,
+	[1][1][RTW89_MKK][44] = 127,
+	[1][1][RTW89_IC][44] = 127,
+	[1][1][RTW89_KCC][44] = 127,
+	[1][1][RTW89_ACMA][44] = 127,
+	[1][1][RTW89_CN][44] = 54,
+	[1][1][RTW89_UK][44] = 127,
+	[1][1][RTW89_FCC][46] = 127,
+	[1][1][RTW89_ETSI][46] = 127,
+	[1][1][RTW89_MKK][46] = 127,
+	[1][1][RTW89_IC][46] = 127,
+	[1][1][RTW89_KCC][46] = 127,
+	[1][1][RTW89_ACMA][46] = 127,
+	[1][1][RTW89_CN][46] = 54,
+	[1][1][RTW89_UK][46] = 127,
+	[1][1][RTW89_FCC][48] = 127,
+	[1][1][RTW89_ETSI][48] = 127,
+	[1][1][RTW89_MKK][48] = 127,
+	[1][1][RTW89_IC][48] = 127,
+	[1][1][RTW89_KCC][48] = 127,
+	[1][1][RTW89_ACMA][48] = 127,
+	[1][1][RTW89_CN][48] = 127,
+	[1][1][RTW89_UK][48] = 127,
+	[1][1][RTW89_FCC][50] = 127,
+	[1][1][RTW89_ETSI][50] = 127,
+	[1][1][RTW89_MKK][50] = 127,
+	[1][1][RTW89_IC][50] = 127,
+	[1][1][RTW89_KCC][50] = 127,
+	[1][1][RTW89_ACMA][50] = 127,
+	[1][1][RTW89_CN][50] = 127,
+	[1][1][RTW89_UK][50] = 127,
+	[1][1][RTW89_FCC][52] = 127,
+	[1][1][RTW89_ETSI][52] = 127,
+	[1][1][RTW89_MKK][52] = 127,
+	[1][1][RTW89_IC][52] = 127,
+	[1][1][RTW89_KCC][52] = 127,
+	[1][1][RTW89_ACMA][52] = 127,
+	[1][1][RTW89_CN][52] = 127,
+	[1][1][RTW89_UK][52] = 127,
+	[2][0][RTW89_FCC][0] = 78,
+	[2][0][RTW89_ETSI][0] = 46,
+	[2][0][RTW89_MKK][0] = 48,
+	[2][0][RTW89_IC][0] = 50,
+	[2][0][RTW89_KCC][0] = 64,
+	[2][0][RTW89_ACMA][0] = 46,
+	[2][0][RTW89_CN][0] = 40,
+	[2][0][RTW89_UK][0] = 46,
+	[2][0][RTW89_FCC][2] = 74,
+	[2][0][RTW89_ETSI][2] = 46,
+	[2][0][RTW89_MKK][2] = 48,
+	[2][0][RTW89_IC][2] = 48,
+	[2][0][RTW89_KCC][2] = 64,
+	[2][0][RTW89_ACMA][2] = 46,
+	[2][0][RTW89_CN][2] = 40,
+	[2][0][RTW89_UK][2] = 46,
+	[2][0][RTW89_FCC][4] = 74,
+	[2][0][RTW89_ETSI][4] = 46,
+	[2][0][RTW89_MKK][4] = 48,
+	[2][0][RTW89_IC][4] = 48,
+	[2][0][RTW89_KCC][4] = 64,
+	[2][0][RTW89_ACMA][4] = 46,
+	[2][0][RTW89_CN][4] = 40,
+	[2][0][RTW89_UK][4] = 46,
+	[2][0][RTW89_FCC][6] = 74,
+	[2][0][RTW89_ETSI][6] = 46,
+	[2][0][RTW89_MKK][6] = 48,
+	[2][0][RTW89_IC][6] = 48,
+	[2][0][RTW89_KCC][6] = 40,
+	[2][0][RTW89_ACMA][6] = 46,
+	[2][0][RTW89_CN][6] = 40,
+	[2][0][RTW89_UK][6] = 46,
+	[2][0][RTW89_FCC][8] = 74,
+	[2][0][RTW89_ETSI][8] = 46,
+	[2][0][RTW89_MKK][8] = 48,
+	[2][0][RTW89_IC][8] = 64,
+	[2][0][RTW89_KCC][8] = 66,
+	[2][0][RTW89_ACMA][8] = 46,
+	[2][0][RTW89_CN][8] = 40,
+	[2][0][RTW89_UK][8] = 46,
+	[2][0][RTW89_FCC][10] = 74,
+	[2][0][RTW89_ETSI][10] = 46,
+	[2][0][RTW89_MKK][10] = 48,
+	[2][0][RTW89_IC][10] = 64,
+	[2][0][RTW89_KCC][10] = 66,
+	[2][0][RTW89_ACMA][10] = 46,
+	[2][0][RTW89_CN][10] = 40,
+	[2][0][RTW89_UK][10] = 46,
+	[2][0][RTW89_FCC][12] = 74,
+	[2][0][RTW89_ETSI][12] = 46,
+	[2][0][RTW89_MKK][12] = 48,
+	[2][0][RTW89_IC][12] = 64,
+	[2][0][RTW89_KCC][12] = 64,
+	[2][0][RTW89_ACMA][12] = 46,
+	[2][0][RTW89_CN][12] = 40,
+	[2][0][RTW89_UK][12] = 46,
+	[2][0][RTW89_FCC][14] = 80,
+	[2][0][RTW89_ETSI][14] = 46,
+	[2][0][RTW89_MKK][14] = 48,
+	[2][0][RTW89_IC][14] = 64,
+	[2][0][RTW89_KCC][14] = 64,
+	[2][0][RTW89_ACMA][14] = 46,
+	[2][0][RTW89_CN][14] = 40,
+	[2][0][RTW89_UK][14] = 46,
+	[2][0][RTW89_FCC][15] = 72,
+	[2][0][RTW89_ETSI][15] = 46,
+	[2][0][RTW89_MKK][15] = 70,
+	[2][0][RTW89_IC][15] = 72,
+	[2][0][RTW89_KCC][15] = 66,
+	[2][0][RTW89_ACMA][15] = 46,
+	[2][0][RTW89_CN][15] = 127,
+	[2][0][RTW89_UK][15] = 46,
+	[2][0][RTW89_FCC][17] = 72,
+	[2][0][RTW89_ETSI][17] = 46,
+	[2][0][RTW89_MKK][17] = 70,
+	[2][0][RTW89_IC][17] = 72,
+	[2][0][RTW89_KCC][17] = 66,
+	[2][0][RTW89_ACMA][17] = 46,
+	[2][0][RTW89_CN][17] = 127,
+	[2][0][RTW89_UK][17] = 46,
+	[2][0][RTW89_FCC][19] = 70,
+	[2][0][RTW89_ETSI][19] = 46,
+	[2][0][RTW89_MKK][19] = 70,
+	[2][0][RTW89_IC][19] = 70,
+	[2][0][RTW89_KCC][19] = 66,
+	[2][0][RTW89_ACMA][19] = 46,
+	[2][0][RTW89_CN][19] = 127,
+	[2][0][RTW89_UK][19] = 46,
+	[2][0][RTW89_FCC][21] = 70,
+	[2][0][RTW89_ETSI][21] = 46,
+	[2][0][RTW89_MKK][21] = 70,
+	[2][0][RTW89_IC][21] = 70,
+	[2][0][RTW89_KCC][21] = 66,
+	[2][0][RTW89_ACMA][21] = 46,
+	[2][0][RTW89_CN][21] = 127,
+	[2][0][RTW89_UK][21] = 46,
+	[2][0][RTW89_FCC][23] = 70,
+	[2][0][RTW89_ETSI][23] = 46,
+	[2][0][RTW89_MKK][23] = 70,
+	[2][0][RTW89_IC][23] = 70,
+	[2][0][RTW89_KCC][23] = 66,
+	[2][0][RTW89_ACMA][23] = 46,
+	[2][0][RTW89_CN][23] = 127,
+	[2][0][RTW89_UK][23] = 46,
+	[2][0][RTW89_FCC][25] = 70,
+	[2][0][RTW89_ETSI][25] = 46,
+	[2][0][RTW89_MKK][25] = 70,
+	[2][0][RTW89_IC][25] = 127,
+	[2][0][RTW89_KCC][25] = 66,
+	[2][0][RTW89_ACMA][25] = 127,
+	[2][0][RTW89_CN][25] = 127,
+	[2][0][RTW89_UK][25] = 46,
+	[2][0][RTW89_FCC][27] = 70,
+	[2][0][RTW89_ETSI][27] = 46,
+	[2][0][RTW89_MKK][27] = 70,
+	[2][0][RTW89_IC][27] = 127,
+	[2][0][RTW89_KCC][27] = 64,
+	[2][0][RTW89_ACMA][27] = 127,
+	[2][0][RTW89_CN][27] = 127,
+	[2][0][RTW89_UK][27] = 46,
+	[2][0][RTW89_FCC][29] = 70,
+	[2][0][RTW89_ETSI][29] = 46,
+	[2][0][RTW89_MKK][29] = 70,
+	[2][0][RTW89_IC][29] = 127,
+	[2][0][RTW89_KCC][29] = 64,
+	[2][0][RTW89_ACMA][29] = 127,
+	[2][0][RTW89_CN][29] = 127,
+	[2][0][RTW89_UK][29] = 46,
+	[2][0][RTW89_FCC][31] = 70,
+	[2][0][RTW89_ETSI][31] = 46,
+	[2][0][RTW89_MKK][31] = 70,
+	[2][0][RTW89_IC][31] = 70,
+	[2][0][RTW89_KCC][31] = 64,
+	[2][0][RTW89_ACMA][31] = 46,
+	[2][0][RTW89_CN][31] = 127,
+	[2][0][RTW89_UK][31] = 46,
+	[2][0][RTW89_FCC][33] = 70,
+	[2][0][RTW89_ETSI][33] = 46,
+	[2][0][RTW89_MKK][33] = 70,
+	[2][0][RTW89_IC][33] = 70,
+	[2][0][RTW89_KCC][33] = 64,
+	[2][0][RTW89_ACMA][33] = 46,
+	[2][0][RTW89_CN][33] = 127,
+	[2][0][RTW89_UK][33] = 46,
+	[2][0][RTW89_FCC][35] = 70,
+	[2][0][RTW89_ETSI][35] = 46,
+	[2][0][RTW89_MKK][35] = 70,
+	[2][0][RTW89_IC][35] = 70,
+	[2][0][RTW89_KCC][35] = 64,
+	[2][0][RTW89_ACMA][35] = 46,
+	[2][0][RTW89_CN][35] = 127,
+	[2][0][RTW89_UK][35] = 46,
+	[2][0][RTW89_FCC][37] = 84,
+	[2][0][RTW89_ETSI][37] = 127,
+	[2][0][RTW89_MKK][37] = 68,
+	[2][0][RTW89_IC][37] = 84,
+	[2][0][RTW89_KCC][37] = 66,
+	[2][0][RTW89_ACMA][37] = 74,
+	[2][0][RTW89_CN][37] = 127,
+	[2][0][RTW89_UK][37] = 74,
+	[2][0][RTW89_FCC][38] = 84,
+	[2][0][RTW89_ETSI][38] = 28,
+	[2][0][RTW89_MKK][38] = 127,
+	[2][0][RTW89_IC][38] = 84,
+	[2][0][RTW89_KCC][38] = 64,
+	[2][0][RTW89_ACMA][38] = 84,
+	[2][0][RTW89_CN][38] = 68,
+	[2][0][RTW89_UK][38] = 46,
+	[2][0][RTW89_FCC][40] = 84,
+	[2][0][RTW89_ETSI][40] = 28,
+	[2][0][RTW89_MKK][40] = 127,
+	[2][0][RTW89_IC][40] = 84,
+	[2][0][RTW89_KCC][40] = 64,
+	[2][0][RTW89_ACMA][40] = 84,
+	[2][0][RTW89_CN][40] = 68,
+	[2][0][RTW89_UK][40] = 46,
+	[2][0][RTW89_FCC][42] = 80,
+	[2][0][RTW89_ETSI][42] = 28,
+	[2][0][RTW89_MKK][42] = 127,
+	[2][0][RTW89_IC][42] = 80,
+	[2][0][RTW89_KCC][42] = 66,
+	[2][0][RTW89_ACMA][42] = 84,
+	[2][0][RTW89_CN][42] = 68,
+	[2][0][RTW89_UK][42] = 46,
+	[2][0][RTW89_FCC][44] = 82,
+	[2][0][RTW89_ETSI][44] = 28,
+	[2][0][RTW89_MKK][44] = 127,
+	[2][0][RTW89_IC][44] = 82,
+	[2][0][RTW89_KCC][44] = 66,
+	[2][0][RTW89_ACMA][44] = 84,
+	[2][0][RTW89_CN][44] = 68,
+	[2][0][RTW89_UK][44] = 46,
+	[2][0][RTW89_FCC][46] = 82,
+	[2][0][RTW89_ETSI][46] = 28,
+	[2][0][RTW89_MKK][46] = 127,
+	[2][0][RTW89_IC][46] = 82,
+	[2][0][RTW89_KCC][46] = 66,
+	[2][0][RTW89_ACMA][46] = 84,
+	[2][0][RTW89_CN][46] = 68,
+	[2][0][RTW89_UK][46] = 46,
+	[2][0][RTW89_FCC][48] = 64,
+	[2][0][RTW89_ETSI][48] = 127,
+	[2][0][RTW89_MKK][48] = 127,
+	[2][0][RTW89_IC][48] = 127,
+	[2][0][RTW89_KCC][48] = 127,
+	[2][0][RTW89_ACMA][48] = 127,
+	[2][0][RTW89_CN][48] = 127,
+	[2][0][RTW89_UK][48] = 127,
+	[2][0][RTW89_FCC][50] = 64,
+	[2][0][RTW89_ETSI][50] = 127,
+	[2][0][RTW89_MKK][50] = 127,
+	[2][0][RTW89_IC][50] = 127,
+	[2][0][RTW89_KCC][50] = 127,
+	[2][0][RTW89_ACMA][50] = 127,
+	[2][0][RTW89_CN][50] = 127,
+	[2][0][RTW89_UK][50] = 127,
+	[2][0][RTW89_FCC][52] = 60,
+	[2][0][RTW89_ETSI][52] = 127,
+	[2][0][RTW89_MKK][52] = 127,
+	[2][0][RTW89_IC][52] = 127,
+	[2][0][RTW89_KCC][52] = 127,
+	[2][0][RTW89_ACMA][52] = 127,
+	[2][0][RTW89_CN][52] = 127,
+	[2][0][RTW89_UK][52] = 127,
+	[2][1][RTW89_FCC][0] = 127,
+	[2][1][RTW89_ETSI][0] = 127,
+	[2][1][RTW89_MKK][0] = 127,
+	[2][1][RTW89_IC][0] = 127,
+	[2][1][RTW89_KCC][0] = 127,
+	[2][1][RTW89_ACMA][0] = 127,
+	[2][1][RTW89_CN][0] = 28,
+	[2][1][RTW89_UK][0] = 127,
+	[2][1][RTW89_FCC][2] = 127,
+	[2][1][RTW89_ETSI][2] = 127,
+	[2][1][RTW89_MKK][2] = 127,
+	[2][1][RTW89_IC][2] = 127,
+	[2][1][RTW89_KCC][2] = 127,
+	[2][1][RTW89_ACMA][2] = 127,
+	[2][1][RTW89_CN][2] = 28,
+	[2][1][RTW89_UK][2] = 127,
+	[2][1][RTW89_FCC][4] = 127,
+	[2][1][RTW89_ETSI][4] = 127,
+	[2][1][RTW89_MKK][4] = 127,
+	[2][1][RTW89_IC][4] = 127,
+	[2][1][RTW89_KCC][4] = 127,
+	[2][1][RTW89_ACMA][4] = 127,
+	[2][1][RTW89_CN][4] = 28,
+	[2][1][RTW89_UK][4] = 127,
+	[2][1][RTW89_FCC][6] = 127,
+	[2][1][RTW89_ETSI][6] = 127,
+	[2][1][RTW89_MKK][6] = 127,
+	[2][1][RTW89_IC][6] = 127,
+	[2][1][RTW89_KCC][6] = 127,
+	[2][1][RTW89_ACMA][6] = 127,
+	[2][1][RTW89_CN][6] = 28,
+	[2][1][RTW89_UK][6] = 127,
+	[2][1][RTW89_FCC][8] = 127,
+	[2][1][RTW89_ETSI][8] = 127,
+	[2][1][RTW89_MKK][8] = 127,
+	[2][1][RTW89_IC][8] = 127,
+	[2][1][RTW89_KCC][8] = 127,
+	[2][1][RTW89_ACMA][8] = 127,
+	[2][1][RTW89_CN][8] = 28,
+	[2][1][RTW89_UK][8] = 127,
+	[2][1][RTW89_FCC][10] = 127,
+	[2][1][RTW89_ETSI][10] = 127,
+	[2][1][RTW89_MKK][10] = 127,
+	[2][1][RTW89_IC][10] = 127,
+	[2][1][RTW89_KCC][10] = 127,
+	[2][1][RTW89_ACMA][10] = 127,
+	[2][1][RTW89_CN][10] = 28,
+	[2][1][RTW89_UK][10] = 127,
+	[2][1][RTW89_FCC][12] = 127,
+	[2][1][RTW89_ETSI][12] = 127,
+	[2][1][RTW89_MKK][12] = 127,
+	[2][1][RTW89_IC][12] = 127,
+	[2][1][RTW89_KCC][12] = 127,
+	[2][1][RTW89_ACMA][12] = 127,
+	[2][1][RTW89_CN][12] = 28,
+	[2][1][RTW89_UK][12] = 127,
+	[2][1][RTW89_FCC][14] = 127,
+	[2][1][RTW89_ETSI][14] = 127,
+	[2][1][RTW89_MKK][14] = 127,
+	[2][1][RTW89_IC][14] = 127,
+	[2][1][RTW89_KCC][14] = 127,
+	[2][1][RTW89_ACMA][14] = 127,
+	[2][1][RTW89_CN][14] = 28,
+	[2][1][RTW89_UK][14] = 127,
+	[2][1][RTW89_FCC][15] = 127,
+	[2][1][RTW89_ETSI][15] = 127,
+	[2][1][RTW89_MKK][15] = 127,
+	[2][1][RTW89_IC][15] = 127,
+	[2][1][RTW89_KCC][15] = 127,
+	[2][1][RTW89_ACMA][15] = 127,
+	[2][1][RTW89_CN][15] = 127,
+	[2][1][RTW89_UK][15] = 127,
+	[2][1][RTW89_FCC][17] = 127,
+	[2][1][RTW89_ETSI][17] = 127,
+	[2][1][RTW89_MKK][17] = 127,
+	[2][1][RTW89_IC][17] = 127,
+	[2][1][RTW89_KCC][17] = 127,
+	[2][1][RTW89_ACMA][17] = 127,
+	[2][1][RTW89_CN][17] = 127,
+	[2][1][RTW89_UK][17] = 127,
+	[2][1][RTW89_FCC][19] = 127,
+	[2][1][RTW89_ETSI][19] = 127,
+	[2][1][RTW89_MKK][19] = 127,
+	[2][1][RTW89_IC][19] = 127,
+	[2][1][RTW89_KCC][19] = 127,
+	[2][1][RTW89_ACMA][19] = 127,
+	[2][1][RTW89_CN][19] = 127,
+	[2][1][RTW89_UK][19] = 127,
+	[2][1][RTW89_FCC][21] = 127,
+	[2][1][RTW89_ETSI][21] = 127,
+	[2][1][RTW89_MKK][21] = 127,
+	[2][1][RTW89_IC][21] = 127,
+	[2][1][RTW89_KCC][21] = 127,
+	[2][1][RTW89_ACMA][21] = 127,
+	[2][1][RTW89_CN][21] = 127,
+	[2][1][RTW89_UK][21] = 127,
+	[2][1][RTW89_FCC][23] = 127,
+	[2][1][RTW89_ETSI][23] = 127,
+	[2][1][RTW89_MKK][23] = 127,
+	[2][1][RTW89_IC][23] = 127,
+	[2][1][RTW89_KCC][23] = 127,
+	[2][1][RTW89_ACMA][23] = 127,
+	[2][1][RTW89_CN][23] = 127,
+	[2][1][RTW89_UK][23] = 127,
+	[2][1][RTW89_FCC][25] = 127,
+	[2][1][RTW89_ETSI][25] = 127,
+	[2][1][RTW89_MKK][25] = 127,
+	[2][1][RTW89_IC][25] = 127,
+	[2][1][RTW89_KCC][25] = 127,
+	[2][1][RTW89_ACMA][25] = 127,
+	[2][1][RTW89_CN][25] = 127,
+	[2][1][RTW89_UK][25] = 127,
+	[2][1][RTW89_FCC][27] = 127,
+	[2][1][RTW89_ETSI][27] = 127,
+	[2][1][RTW89_MKK][27] = 127,
+	[2][1][RTW89_IC][27] = 127,
+	[2][1][RTW89_KCC][27] = 127,
+	[2][1][RTW89_ACMA][27] = 127,
+	[2][1][RTW89_CN][27] = 127,
+	[2][1][RTW89_UK][27] = 127,
+	[2][1][RTW89_FCC][29] = 127,
+	[2][1][RTW89_ETSI][29] = 127,
+	[2][1][RTW89_MKK][29] = 127,
+	[2][1][RTW89_IC][29] = 127,
+	[2][1][RTW89_KCC][29] = 127,
+	[2][1][RTW89_ACMA][29] = 127,
+	[2][1][RTW89_CN][29] = 127,
+	[2][1][RTW89_UK][29] = 127,
+	[2][1][RTW89_FCC][31] = 127,
+	[2][1][RTW89_ETSI][31] = 127,
+	[2][1][RTW89_MKK][31] = 127,
+	[2][1][RTW89_IC][31] = 127,
+	[2][1][RTW89_KCC][31] = 127,
+	[2][1][RTW89_ACMA][31] = 127,
+	[2][1][RTW89_CN][31] = 127,
+	[2][1][RTW89_UK][31] = 127,
+	[2][1][RTW89_FCC][33] = 127,
+	[2][1][RTW89_ETSI][33] = 127,
+	[2][1][RTW89_MKK][33] = 127,
+	[2][1][RTW89_IC][33] = 127,
+	[2][1][RTW89_KCC][33] = 127,
+	[2][1][RTW89_ACMA][33] = 127,
+	[2][1][RTW89_CN][33] = 127,
+	[2][1][RTW89_UK][33] = 127,
+	[2][1][RTW89_FCC][35] = 127,
+	[2][1][RTW89_ETSI][35] = 127,
+	[2][1][RTW89_MKK][35] = 127,
+	[2][1][RTW89_IC][35] = 127,
+	[2][1][RTW89_KCC][35] = 127,
+	[2][1][RTW89_ACMA][35] = 127,
+	[2][1][RTW89_CN][35] = 127,
+	[2][1][RTW89_UK][35] = 127,
+	[2][1][RTW89_FCC][37] = 127,
+	[2][1][RTW89_ETSI][37] = 127,
+	[2][1][RTW89_MKK][37] = 127,
+	[2][1][RTW89_IC][37] = 127,
+	[2][1][RTW89_KCC][37] = 127,
+	[2][1][RTW89_ACMA][37] = 127,
+	[2][1][RTW89_CN][37] = 127,
+	[2][1][RTW89_UK][37] = 127,
+	[2][1][RTW89_FCC][38] = 127,
+	[2][1][RTW89_ETSI][38] = 127,
+	[2][1][RTW89_MKK][38] = 127,
+	[2][1][RTW89_IC][38] = 127,
+	[2][1][RTW89_KCC][38] = 127,
+	[2][1][RTW89_ACMA][38] = 127,
+	[2][1][RTW89_CN][38] = 56,
+	[2][1][RTW89_UK][38] = 127,
+	[2][1][RTW89_FCC][40] = 127,
+	[2][1][RTW89_ETSI][40] = 127,
+	[2][1][RTW89_MKK][40] = 127,
+	[2][1][RTW89_IC][40] = 127,
+	[2][1][RTW89_KCC][40] = 127,
+	[2][1][RTW89_ACMA][40] = 127,
+	[2][1][RTW89_CN][40] = 56,
+	[2][1][RTW89_UK][40] = 127,
+	[2][1][RTW89_FCC][42] = 127,
+	[2][1][RTW89_ETSI][42] = 127,
+	[2][1][RTW89_MKK][42] = 127,
+	[2][1][RTW89_IC][42] = 127,
+	[2][1][RTW89_KCC][42] = 127,
+	[2][1][RTW89_ACMA][42] = 127,
+	[2][1][RTW89_CN][42] = 56,
+	[2][1][RTW89_UK][42] = 127,
+	[2][1][RTW89_FCC][44] = 127,
+	[2][1][RTW89_ETSI][44] = 127,
+	[2][1][RTW89_MKK][44] = 127,
+	[2][1][RTW89_IC][44] = 127,
+	[2][1][RTW89_KCC][44] = 127,
+	[2][1][RTW89_ACMA][44] = 127,
+	[2][1][RTW89_CN][44] = 56,
+	[2][1][RTW89_UK][44] = 127,
+	[2][1][RTW89_FCC][46] = 127,
+	[2][1][RTW89_ETSI][46] = 127,
+	[2][1][RTW89_MKK][46] = 127,
+	[2][1][RTW89_IC][46] = 127,
+	[2][1][RTW89_KCC][46] = 127,
+	[2][1][RTW89_ACMA][46] = 127,
+	[2][1][RTW89_CN][46] = 56,
+	[2][1][RTW89_UK][46] = 127,
+	[2][1][RTW89_FCC][48] = 127,
+	[2][1][RTW89_ETSI][48] = 127,
+	[2][1][RTW89_MKK][48] = 127,
+	[2][1][RTW89_IC][48] = 127,
+	[2][1][RTW89_KCC][48] = 127,
+	[2][1][RTW89_ACMA][48] = 127,
+	[2][1][RTW89_CN][48] = 127,
+	[2][1][RTW89_UK][48] = 127,
+	[2][1][RTW89_FCC][50] = 127,
+	[2][1][RTW89_ETSI][50] = 127,
+	[2][1][RTW89_MKK][50] = 127,
+	[2][1][RTW89_IC][50] = 127,
+	[2][1][RTW89_KCC][50] = 127,
+	[2][1][RTW89_ACMA][50] = 127,
+	[2][1][RTW89_CN][50] = 127,
+	[2][1][RTW89_UK][50] = 127,
+	[2][1][RTW89_FCC][52] = 127,
+	[2][1][RTW89_ETSI][52] = 127,
+	[2][1][RTW89_MKK][52] = 127,
+	[2][1][RTW89_IC][52] = 127,
+	[2][1][RTW89_KCC][52] = 127,
+	[2][1][RTW89_ACMA][52] = 127,
+	[2][1][RTW89_CN][52] = 127,
+	[2][1][RTW89_UK][52] = 127,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_2g_type2[RTW89_2G_BW_NUM][RTW89_NTX_NUM]
+				       [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+				       [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
+	[0][0][0][0][RTW89_WW][0] = 58,
+	[0][0][0][0][RTW89_WW][1] = 58,
+	[0][0][0][0][RTW89_WW][2] = 58,
+	[0][0][0][0][RTW89_WW][3] = 58,
+	[0][0][0][0][RTW89_WW][4] = 58,
+	[0][0][0][0][RTW89_WW][5] = 58,
+	[0][0][0][0][RTW89_WW][6] = 58,
+	[0][0][0][0][RTW89_WW][7] = 58,
+	[0][0][0][0][RTW89_WW][8] = 58,
+	[0][0][0][0][RTW89_WW][9] = 58,
+	[0][0][0][0][RTW89_WW][10] = 58,
+	[0][0][0][0][RTW89_WW][11] = 58,
+	[0][0][0][0][RTW89_WW][12] = 52,
+	[0][0][0][0][RTW89_WW][13] = 76,
+	[0][1][0][0][RTW89_WW][0] = 0,
+	[0][1][0][0][RTW89_WW][1] = 0,
+	[0][1][0][0][RTW89_WW][2] = 0,
+	[0][1][0][0][RTW89_WW][3] = 0,
+	[0][1][0][0][RTW89_WW][4] = 0,
+	[0][1][0][0][RTW89_WW][5] = 0,
+	[0][1][0][0][RTW89_WW][6] = 0,
+	[0][1][0][0][RTW89_WW][7] = 0,
+	[0][1][0][0][RTW89_WW][8] = 0,
+	[0][1][0][0][RTW89_WW][9] = 0,
+	[0][1][0][0][RTW89_WW][10] = 0,
+	[0][1][0][0][RTW89_WW][11] = 0,
+	[0][1][0][0][RTW89_WW][12] = 0,
+	[0][1][0][0][RTW89_WW][13] = 0,
+	[1][0][0][0][RTW89_WW][0] = 0,
+	[1][0][0][0][RTW89_WW][1] = 0,
+	[1][0][0][0][RTW89_WW][2] = 58,
+	[1][0][0][0][RTW89_WW][3] = 58,
+	[1][0][0][0][RTW89_WW][4] = 58,
+	[1][0][0][0][RTW89_WW][5] = 58,
+	[1][0][0][0][RTW89_WW][6] = 58,
+	[1][0][0][0][RTW89_WW][7] = 58,
+	[1][0][0][0][RTW89_WW][8] = 58,
+	[1][0][0][0][RTW89_WW][9] = 58,
+	[1][0][0][0][RTW89_WW][10] = 58,
+	[1][0][0][0][RTW89_WW][11] = 0,
+	[1][0][0][0][RTW89_WW][12] = 0,
+	[1][0][0][0][RTW89_WW][13] = 0,
+	[1][1][0][0][RTW89_WW][0] = 0,
+	[1][1][0][0][RTW89_WW][1] = 0,
+	[1][1][0][0][RTW89_WW][2] = 0,
+	[1][1][0][0][RTW89_WW][3] = 0,
+	[1][1][0][0][RTW89_WW][4] = 0,
+	[1][1][0][0][RTW89_WW][5] = 0,
+	[1][1][0][0][RTW89_WW][6] = 0,
+	[1][1][0][0][RTW89_WW][7] = 0,
+	[1][1][0][0][RTW89_WW][8] = 0,
+	[1][1][0][0][RTW89_WW][9] = 0,
+	[1][1][0][0][RTW89_WW][10] = 0,
+	[1][1][0][0][RTW89_WW][11] = 0,
+	[1][1][0][0][RTW89_WW][12] = 0,
+	[1][1][0][0][RTW89_WW][13] = 0,
+	[0][0][1][0][RTW89_WW][0] = 58,
+	[0][0][1][0][RTW89_WW][1] = 60,
+	[0][0][1][0][RTW89_WW][2] = 60,
+	[0][0][1][0][RTW89_WW][3] = 60,
+	[0][0][1][0][RTW89_WW][4] = 60,
+	[0][0][1][0][RTW89_WW][5] = 60,
+	[0][0][1][0][RTW89_WW][6] = 60,
+	[0][0][1][0][RTW89_WW][7] = 60,
+	[0][0][1][0][RTW89_WW][8] = 60,
+	[0][0][1][0][RTW89_WW][9] = 60,
+	[0][0][1][0][RTW89_WW][10] = 60,
+	[0][0][1][0][RTW89_WW][11] = 60,
+	[0][0][1][0][RTW89_WW][12] = 58,
+	[0][0][1][0][RTW89_WW][13] = 0,
+	[0][1][1][0][RTW89_WW][0] = 0,
+	[0][1][1][0][RTW89_WW][1] = 0,
+	[0][1][1][0][RTW89_WW][2] = 0,
+	[0][1][1][0][RTW89_WW][3] = 0,
+	[0][1][1][0][RTW89_WW][4] = 0,
+	[0][1][1][0][RTW89_WW][5] = 0,
+	[0][1][1][0][RTW89_WW][6] = 0,
+	[0][1][1][0][RTW89_WW][7] = 0,
+	[0][1][1][0][RTW89_WW][8] = 0,
+	[0][1][1][0][RTW89_WW][9] = 0,
+	[0][1][1][0][RTW89_WW][10] = 0,
+	[0][1][1][0][RTW89_WW][11] = 0,
+	[0][1][1][0][RTW89_WW][12] = 0,
+	[0][1][1][0][RTW89_WW][13] = 0,
+	[0][0][2][0][RTW89_WW][0] = 60,
+	[0][0][2][0][RTW89_WW][1] = 60,
+	[0][0][2][0][RTW89_WW][2] = 60,
+	[0][0][2][0][RTW89_WW][3] = 60,
+	[0][0][2][0][RTW89_WW][4] = 60,
+	[0][0][2][0][RTW89_WW][5] = 60,
+	[0][0][2][0][RTW89_WW][6] = 60,
+	[0][0][2][0][RTW89_WW][7] = 60,
+	[0][0][2][0][RTW89_WW][8] = 60,
+	[0][0][2][0][RTW89_WW][9] = 60,
+	[0][0][2][0][RTW89_WW][10] = 60,
+	[0][0][2][0][RTW89_WW][11] = 60,
+	[0][0][2][0][RTW89_WW][12] = 60,
+	[0][0][2][0][RTW89_WW][13] = 0,
+	[0][1][2][0][RTW89_WW][0] = 0,
+	[0][1][2][0][RTW89_WW][1] = 0,
+	[0][1][2][0][RTW89_WW][2] = 0,
+	[0][1][2][0][RTW89_WW][3] = 0,
+	[0][1][2][0][RTW89_WW][4] = 0,
+	[0][1][2][0][RTW89_WW][5] = 0,
+	[0][1][2][0][RTW89_WW][6] = 0,
+	[0][1][2][0][RTW89_WW][7] = 0,
+	[0][1][2][0][RTW89_WW][8] = 0,
+	[0][1][2][0][RTW89_WW][9] = 0,
+	[0][1][2][0][RTW89_WW][10] = 0,
+	[0][1][2][0][RTW89_WW][11] = 0,
+	[0][1][2][0][RTW89_WW][12] = 0,
+	[0][1][2][0][RTW89_WW][13] = 0,
+	[0][1][2][1][RTW89_WW][0] = 0,
+	[0][1][2][1][RTW89_WW][1] = 0,
+	[0][1][2][1][RTW89_WW][2] = 0,
+	[0][1][2][1][RTW89_WW][3] = 0,
+	[0][1][2][1][RTW89_WW][4] = 0,
+	[0][1][2][1][RTW89_WW][5] = 0,
+	[0][1][2][1][RTW89_WW][6] = 0,
+	[0][1][2][1][RTW89_WW][7] = 0,
+	[0][1][2][1][RTW89_WW][8] = 0,
+	[0][1][2][1][RTW89_WW][9] = 0,
+	[0][1][2][1][RTW89_WW][10] = 0,
+	[0][1][2][1][RTW89_WW][11] = 0,
+	[0][1][2][1][RTW89_WW][12] = 0,
+	[0][1][2][1][RTW89_WW][13] = 0,
+	[1][0][2][0][RTW89_WW][0] = 0,
+	[1][0][2][0][RTW89_WW][1] = 0,
+	[1][0][2][0][RTW89_WW][2] = 58,
+	[1][0][2][0][RTW89_WW][3] = 58,
+	[1][0][2][0][RTW89_WW][4] = 58,
+	[1][0][2][0][RTW89_WW][5] = 58,
+	[1][0][2][0][RTW89_WW][6] = 58,
+	[1][0][2][0][RTW89_WW][7] = 58,
+	[1][0][2][0][RTW89_WW][8] = 58,
+	[1][0][2][0][RTW89_WW][9] = 58,
+	[1][0][2][0][RTW89_WW][10] = 58,
+	[1][0][2][0][RTW89_WW][11] = 0,
+	[1][0][2][0][RTW89_WW][12] = 0,
+	[1][0][2][0][RTW89_WW][13] = 0,
+	[1][1][2][0][RTW89_WW][0] = 0,
+	[1][1][2][0][RTW89_WW][1] = 0,
+	[1][1][2][0][RTW89_WW][2] = 0,
+	[1][1][2][0][RTW89_WW][3] = 0,
+	[1][1][2][0][RTW89_WW][4] = 0,
+	[1][1][2][0][RTW89_WW][5] = 0,
+	[1][1][2][0][RTW89_WW][6] = 0,
+	[1][1][2][0][RTW89_WW][7] = 0,
+	[1][1][2][0][RTW89_WW][8] = 0,
+	[1][1][2][0][RTW89_WW][9] = 0,
+	[1][1][2][0][RTW89_WW][10] = 0,
+	[1][1][2][0][RTW89_WW][11] = 0,
+	[1][1][2][0][RTW89_WW][12] = 0,
+	[1][1][2][0][RTW89_WW][13] = 0,
+	[1][1][2][1][RTW89_WW][0] = 0,
+	[1][1][2][1][RTW89_WW][1] = 0,
+	[1][1][2][1][RTW89_WW][2] = 0,
+	[1][1][2][1][RTW89_WW][3] = 0,
+	[1][1][2][1][RTW89_WW][4] = 0,
+	[1][1][2][1][RTW89_WW][5] = 0,
+	[1][1][2][1][RTW89_WW][6] = 0,
+	[1][1][2][1][RTW89_WW][7] = 0,
+	[1][1][2][1][RTW89_WW][8] = 0,
+	[1][1][2][1][RTW89_WW][9] = 0,
+	[1][1][2][1][RTW89_WW][10] = 0,
+	[1][1][2][1][RTW89_WW][11] = 0,
+	[1][1][2][1][RTW89_WW][12] = 0,
+	[1][1][2][1][RTW89_WW][13] = 0,
+	[0][0][0][0][RTW89_FCC][0] = 82,
+	[0][0][0][0][RTW89_ETSI][0] = 58,
+	[0][0][0][0][RTW89_MKK][0] = 68,
+	[0][0][0][0][RTW89_IC][0] = 82,
+	[0][0][0][0][RTW89_KCC][0] = 68,
+	[0][0][0][0][RTW89_ACMA][0] = 58,
+	[0][0][0][0][RTW89_CN][0] = 60,
+	[0][0][0][0][RTW89_UK][0] = 58,
+	[0][0][0][0][RTW89_FCC][1] = 82,
+	[0][0][0][0][RTW89_ETSI][1] = 58,
+	[0][0][0][0][RTW89_MKK][1] = 68,
+	[0][0][0][0][RTW89_IC][1] = 82,
+	[0][0][0][0][RTW89_KCC][1] = 68,
+	[0][0][0][0][RTW89_ACMA][1] = 58,
+	[0][0][0][0][RTW89_CN][1] = 60,
+	[0][0][0][0][RTW89_UK][1] = 58,
+	[0][0][0][0][RTW89_FCC][2] = 82,
+	[0][0][0][0][RTW89_ETSI][2] = 58,
+	[0][0][0][0][RTW89_MKK][2] = 68,
+	[0][0][0][0][RTW89_IC][2] = 82,
+	[0][0][0][0][RTW89_KCC][2] = 68,
+	[0][0][0][0][RTW89_ACMA][2] = 58,
+	[0][0][0][0][RTW89_CN][2] = 60,
+	[0][0][0][0][RTW89_UK][2] = 58,
+	[0][0][0][0][RTW89_FCC][3] = 82,
+	[0][0][0][0][RTW89_ETSI][3] = 58,
+	[0][0][0][0][RTW89_MKK][3] = 68,
+	[0][0][0][0][RTW89_IC][3] = 82,
+	[0][0][0][0][RTW89_KCC][3] = 68,
+	[0][0][0][0][RTW89_ACMA][3] = 58,
+	[0][0][0][0][RTW89_CN][3] = 60,
+	[0][0][0][0][RTW89_UK][3] = 58,
+	[0][0][0][0][RTW89_FCC][4] = 82,
+	[0][0][0][0][RTW89_ETSI][4] = 58,
+	[0][0][0][0][RTW89_MKK][4] = 68,
+	[0][0][0][0][RTW89_IC][4] = 82,
+	[0][0][0][0][RTW89_KCC][4] = 68,
+	[0][0][0][0][RTW89_ACMA][4] = 58,
+	[0][0][0][0][RTW89_CN][4] = 60,
+	[0][0][0][0][RTW89_UK][4] = 58,
+	[0][0][0][0][RTW89_FCC][5] = 82,
+	[0][0][0][0][RTW89_ETSI][5] = 58,
+	[0][0][0][0][RTW89_MKK][5] = 68,
+	[0][0][0][0][RTW89_IC][5] = 82,
+	[0][0][0][0][RTW89_KCC][5] = 68,
+	[0][0][0][0][RTW89_ACMA][5] = 58,
+	[0][0][0][0][RTW89_CN][5] = 60,
+	[0][0][0][0][RTW89_UK][5] = 58,
+	[0][0][0][0][RTW89_FCC][6] = 82,
+	[0][0][0][0][RTW89_ETSI][6] = 58,
+	[0][0][0][0][RTW89_MKK][6] = 68,
+	[0][0][0][0][RTW89_IC][6] = 82,
+	[0][0][0][0][RTW89_KCC][6] = 68,
+	[0][0][0][0][RTW89_ACMA][6] = 58,
+	[0][0][0][0][RTW89_CN][6] = 60,
+	[0][0][0][0][RTW89_UK][6] = 58,
+	[0][0][0][0][RTW89_FCC][7] = 82,
+	[0][0][0][0][RTW89_ETSI][7] = 58,
+	[0][0][0][0][RTW89_MKK][7] = 68,
+	[0][0][0][0][RTW89_IC][7] = 82,
+	[0][0][0][0][RTW89_KCC][7] = 68,
+	[0][0][0][0][RTW89_ACMA][7] = 58,
+	[0][0][0][0][RTW89_CN][7] = 60,
+	[0][0][0][0][RTW89_UK][7] = 58,
+	[0][0][0][0][RTW89_FCC][8] = 82,
+	[0][0][0][0][RTW89_ETSI][8] = 58,
+	[0][0][0][0][RTW89_MKK][8] = 68,
+	[0][0][0][0][RTW89_IC][8] = 82,
+	[0][0][0][0][RTW89_KCC][8] = 68,
+	[0][0][0][0][RTW89_ACMA][8] = 58,
+	[0][0][0][0][RTW89_CN][8] = 60,
+	[0][0][0][0][RTW89_UK][8] = 58,
+	[0][0][0][0][RTW89_FCC][9] = 82,
+	[0][0][0][0][RTW89_ETSI][9] = 58,
+	[0][0][0][0][RTW89_MKK][9] = 68,
+	[0][0][0][0][RTW89_IC][9] = 82,
+	[0][0][0][0][RTW89_KCC][9] = 68,
+	[0][0][0][0][RTW89_ACMA][9] = 58,
+	[0][0][0][0][RTW89_CN][9] = 60,
+	[0][0][0][0][RTW89_UK][9] = 58,
+	[0][0][0][0][RTW89_FCC][10] = 80,
+	[0][0][0][0][RTW89_ETSI][10] = 58,
+	[0][0][0][0][RTW89_MKK][10] = 68,
+	[0][0][0][0][RTW89_IC][10] = 80,
+	[0][0][0][0][RTW89_KCC][10] = 68,
+	[0][0][0][0][RTW89_ACMA][10] = 58,
+	[0][0][0][0][RTW89_CN][10] = 60,
+	[0][0][0][0][RTW89_UK][10] = 58,
+	[0][0][0][0][RTW89_FCC][11] = 60,
+	[0][0][0][0][RTW89_ETSI][11] = 58,
+	[0][0][0][0][RTW89_MKK][11] = 68,
+	[0][0][0][0][RTW89_IC][11] = 60,
+	[0][0][0][0][RTW89_KCC][11] = 68,
+	[0][0][0][0][RTW89_ACMA][11] = 58,
+	[0][0][0][0][RTW89_CN][11] = 60,
+	[0][0][0][0][RTW89_UK][11] = 58,
+	[0][0][0][0][RTW89_FCC][12] = 52,
+	[0][0][0][0][RTW89_ETSI][12] = 58,
+	[0][0][0][0][RTW89_MKK][12] = 68,
+	[0][0][0][0][RTW89_IC][12] = 52,
+	[0][0][0][0][RTW89_KCC][12] = 68,
+	[0][0][0][0][RTW89_ACMA][12] = 58,
+	[0][0][0][0][RTW89_CN][12] = 60,
+	[0][0][0][0][RTW89_UK][12] = 58,
+	[0][0][0][0][RTW89_FCC][13] = 127,
+	[0][0][0][0][RTW89_ETSI][13] = 127,
+	[0][0][0][0][RTW89_MKK][13] = 76,
+	[0][0][0][0][RTW89_IC][13] = 127,
+	[0][0][0][0][RTW89_KCC][13] = 127,
+	[0][0][0][0][RTW89_ACMA][13] = 127,
+	[0][0][0][0][RTW89_CN][13] = 127,
+	[0][0][0][0][RTW89_UK][13] = 127,
+	[0][1][0][0][RTW89_FCC][0] = 127,
+	[0][1][0][0][RTW89_ETSI][0] = 127,
+	[0][1][0][0][RTW89_MKK][0] = 127,
+	[0][1][0][0][RTW89_IC][0] = 127,
+	[0][1][0][0][RTW89_KCC][0] = 127,
+	[0][1][0][0][RTW89_ACMA][0] = 127,
+	[0][1][0][0][RTW89_CN][0] = 127,
+	[0][1][0][0][RTW89_UK][0] = 127,
+	[0][1][0][0][RTW89_FCC][1] = 127,
+	[0][1][0][0][RTW89_ETSI][1] = 127,
+	[0][1][0][0][RTW89_MKK][1] = 127,
+	[0][1][0][0][RTW89_IC][1] = 127,
+	[0][1][0][0][RTW89_KCC][1] = 127,
+	[0][1][0][0][RTW89_ACMA][1] = 127,
+	[0][1][0][0][RTW89_CN][1] = 127,
+	[0][1][0][0][RTW89_UK][1] = 127,
+	[0][1][0][0][RTW89_FCC][2] = 127,
+	[0][1][0][0][RTW89_ETSI][2] = 127,
+	[0][1][0][0][RTW89_MKK][2] = 127,
+	[0][1][0][0][RTW89_IC][2] = 127,
+	[0][1][0][0][RTW89_KCC][2] = 127,
+	[0][1][0][0][RTW89_ACMA][2] = 127,
+	[0][1][0][0][RTW89_CN][2] = 127,
+	[0][1][0][0][RTW89_UK][2] = 127,
+	[0][1][0][0][RTW89_FCC][3] = 127,
+	[0][1][0][0][RTW89_ETSI][3] = 127,
+	[0][1][0][0][RTW89_MKK][3] = 127,
+	[0][1][0][0][RTW89_IC][3] = 127,
+	[0][1][0][0][RTW89_KCC][3] = 127,
+	[0][1][0][0][RTW89_ACMA][3] = 127,
+	[0][1][0][0][RTW89_CN][3] = 127,
+	[0][1][0][0][RTW89_UK][3] = 127,
+	[0][1][0][0][RTW89_FCC][4] = 127,
+	[0][1][0][0][RTW89_ETSI][4] = 127,
+	[0][1][0][0][RTW89_MKK][4] = 127,
+	[0][1][0][0][RTW89_IC][4] = 127,
+	[0][1][0][0][RTW89_KCC][4] = 127,
+	[0][1][0][0][RTW89_ACMA][4] = 127,
+	[0][1][0][0][RTW89_CN][4] = 127,
+	[0][1][0][0][RTW89_UK][4] = 127,
+	[0][1][0][0][RTW89_FCC][5] = 127,
+	[0][1][0][0][RTW89_ETSI][5] = 127,
+	[0][1][0][0][RTW89_MKK][5] = 127,
+	[0][1][0][0][RTW89_IC][5] = 127,
+	[0][1][0][0][RTW89_KCC][5] = 127,
+	[0][1][0][0][RTW89_ACMA][5] = 127,
+	[0][1][0][0][RTW89_CN][5] = 127,
+	[0][1][0][0][RTW89_UK][5] = 127,
+	[0][1][0][0][RTW89_FCC][6] = 127,
+	[0][1][0][0][RTW89_ETSI][6] = 127,
+	[0][1][0][0][RTW89_MKK][6] = 127,
+	[0][1][0][0][RTW89_IC][6] = 127,
+	[0][1][0][0][RTW89_KCC][6] = 127,
+	[0][1][0][0][RTW89_ACMA][6] = 127,
+	[0][1][0][0][RTW89_CN][6] = 127,
+	[0][1][0][0][RTW89_UK][6] = 127,
+	[0][1][0][0][RTW89_FCC][7] = 127,
+	[0][1][0][0][RTW89_ETSI][7] = 127,
+	[0][1][0][0][RTW89_MKK][7] = 127,
+	[0][1][0][0][RTW89_IC][7] = 127,
+	[0][1][0][0][RTW89_KCC][7] = 127,
+	[0][1][0][0][RTW89_ACMA][7] = 127,
+	[0][1][0][0][RTW89_CN][7] = 127,
+	[0][1][0][0][RTW89_UK][7] = 127,
+	[0][1][0][0][RTW89_FCC][8] = 127,
+	[0][1][0][0][RTW89_ETSI][8] = 127,
+	[0][1][0][0][RTW89_MKK][8] = 127,
+	[0][1][0][0][RTW89_IC][8] = 127,
+	[0][1][0][0][RTW89_KCC][8] = 127,
+	[0][1][0][0][RTW89_ACMA][8] = 127,
+	[0][1][0][0][RTW89_CN][8] = 127,
+	[0][1][0][0][RTW89_UK][8] = 127,
+	[0][1][0][0][RTW89_FCC][9] = 127,
+	[0][1][0][0][RTW89_ETSI][9] = 127,
+	[0][1][0][0][RTW89_MKK][9] = 127,
+	[0][1][0][0][RTW89_IC][9] = 127,
+	[0][1][0][0][RTW89_KCC][9] = 127,
+	[0][1][0][0][RTW89_ACMA][9] = 127,
+	[0][1][0][0][RTW89_CN][9] = 127,
+	[0][1][0][0][RTW89_UK][9] = 127,
+	[0][1][0][0][RTW89_FCC][10] = 127,
+	[0][1][0][0][RTW89_ETSI][10] = 127,
+	[0][1][0][0][RTW89_MKK][10] = 127,
+	[0][1][0][0][RTW89_IC][10] = 127,
+	[0][1][0][0][RTW89_KCC][10] = 127,
+	[0][1][0][0][RTW89_ACMA][10] = 127,
+	[0][1][0][0][RTW89_CN][10] = 127,
+	[0][1][0][0][RTW89_UK][10] = 127,
+	[0][1][0][0][RTW89_FCC][11] = 127,
+	[0][1][0][0][RTW89_ETSI][11] = 127,
+	[0][1][0][0][RTW89_MKK][11] = 127,
+	[0][1][0][0][RTW89_IC][11] = 127,
+	[0][1][0][0][RTW89_KCC][11] = 127,
+	[0][1][0][0][RTW89_ACMA][11] = 127,
+	[0][1][0][0][RTW89_CN][11] = 127,
+	[0][1][0][0][RTW89_UK][11] = 127,
+	[0][1][0][0][RTW89_FCC][12] = 127,
+	[0][1][0][0][RTW89_ETSI][12] = 127,
+	[0][1][0][0][RTW89_MKK][12] = 127,
+	[0][1][0][0][RTW89_IC][12] = 127,
+	[0][1][0][0][RTW89_KCC][12] = 127,
+	[0][1][0][0][RTW89_ACMA][12] = 127,
+	[0][1][0][0][RTW89_CN][12] = 127,
+	[0][1][0][0][RTW89_UK][12] = 127,
+	[0][1][0][0][RTW89_FCC][13] = 127,
+	[0][1][0][0][RTW89_ETSI][13] = 127,
+	[0][1][0][0][RTW89_MKK][13] = 127,
+	[0][1][0][0][RTW89_IC][13] = 127,
+	[0][1][0][0][RTW89_KCC][13] = 127,
+	[0][1][0][0][RTW89_ACMA][13] = 127,
+	[0][1][0][0][RTW89_CN][13] = 127,
+	[0][1][0][0][RTW89_UK][13] = 127,
+	[1][0][0][0][RTW89_FCC][0] = 127,
+	[1][0][0][0][RTW89_ETSI][0] = 127,
+	[1][0][0][0][RTW89_MKK][0] = 127,
+	[1][0][0][0][RTW89_IC][0] = 127,
+	[1][0][0][0][RTW89_KCC][0] = 127,
+	[1][0][0][0][RTW89_ACMA][0] = 127,
+	[1][0][0][0][RTW89_CN][0] = 127,
+	[1][0][0][0][RTW89_UK][0] = 127,
+	[1][0][0][0][RTW89_FCC][1] = 127,
+	[1][0][0][0][RTW89_ETSI][1] = 127,
+	[1][0][0][0][RTW89_MKK][1] = 127,
+	[1][0][0][0][RTW89_IC][1] = 127,
+	[1][0][0][0][RTW89_KCC][1] = 127,
+	[1][0][0][0][RTW89_ACMA][1] = 127,
+	[1][0][0][0][RTW89_CN][1] = 127,
+	[1][0][0][0][RTW89_UK][1] = 127,
+	[1][0][0][0][RTW89_FCC][2] = 127,
+	[1][0][0][0][RTW89_ETSI][2] = 58,
+	[1][0][0][0][RTW89_MKK][2] = 70,
+	[1][0][0][0][RTW89_IC][2] = 127,
+	[1][0][0][0][RTW89_KCC][2] = 68,
+	[1][0][0][0][RTW89_ACMA][2] = 58,
+	[1][0][0][0][RTW89_CN][2] = 60,
+	[1][0][0][0][RTW89_UK][2] = 58,
+	[1][0][0][0][RTW89_FCC][3] = 127,
+	[1][0][0][0][RTW89_ETSI][3] = 58,
+	[1][0][0][0][RTW89_MKK][3] = 76,
+	[1][0][0][0][RTW89_IC][3] = 127,
+	[1][0][0][0][RTW89_KCC][3] = 68,
+	[1][0][0][0][RTW89_ACMA][3] = 58,
+	[1][0][0][0][RTW89_CN][3] = 60,
+	[1][0][0][0][RTW89_UK][3] = 58,
+	[1][0][0][0][RTW89_FCC][4] = 127,
+	[1][0][0][0][RTW89_ETSI][4] = 58,
+	[1][0][0][0][RTW89_MKK][4] = 76,
+	[1][0][0][0][RTW89_IC][4] = 127,
+	[1][0][0][0][RTW89_KCC][4] = 68,
+	[1][0][0][0][RTW89_ACMA][4] = 58,
+	[1][0][0][0][RTW89_CN][4] = 60,
+	[1][0][0][0][RTW89_UK][4] = 58,
+	[1][0][0][0][RTW89_FCC][5] = 127,
+	[1][0][0][0][RTW89_ETSI][5] = 58,
+	[1][0][0][0][RTW89_MKK][5] = 76,
+	[1][0][0][0][RTW89_IC][5] = 127,
+	[1][0][0][0][RTW89_KCC][5] = 68,
+	[1][0][0][0][RTW89_ACMA][5] = 58,
+	[1][0][0][0][RTW89_CN][5] = 60,
+	[1][0][0][0][RTW89_UK][5] = 58,
+	[1][0][0][0][RTW89_FCC][6] = 127,
+	[1][0][0][0][RTW89_ETSI][6] = 58,
+	[1][0][0][0][RTW89_MKK][6] = 76,
+	[1][0][0][0][RTW89_IC][6] = 127,
+	[1][0][0][0][RTW89_KCC][6] = 68,
+	[1][0][0][0][RTW89_ACMA][6] = 58,
+	[1][0][0][0][RTW89_CN][6] = 60,
+	[1][0][0][0][RTW89_UK][6] = 58,
+	[1][0][0][0][RTW89_FCC][7] = 127,
+	[1][0][0][0][RTW89_ETSI][7] = 58,
+	[1][0][0][0][RTW89_MKK][7] = 76,
+	[1][0][0][0][RTW89_IC][7] = 127,
+	[1][0][0][0][RTW89_KCC][7] = 68,
+	[1][0][0][0][RTW89_ACMA][7] = 58,
+	[1][0][0][0][RTW89_CN][7] = 60,
+	[1][0][0][0][RTW89_UK][7] = 58,
+	[1][0][0][0][RTW89_FCC][8] = 127,
+	[1][0][0][0][RTW89_ETSI][8] = 58,
+	[1][0][0][0][RTW89_MKK][8] = 76,
+	[1][0][0][0][RTW89_IC][8] = 127,
+	[1][0][0][0][RTW89_KCC][8] = 68,
+	[1][0][0][0][RTW89_ACMA][8] = 58,
+	[1][0][0][0][RTW89_CN][8] = 60,
+	[1][0][0][0][RTW89_UK][8] = 58,
+	[1][0][0][0][RTW89_FCC][9] = 127,
+	[1][0][0][0][RTW89_ETSI][9] = 58,
+	[1][0][0][0][RTW89_MKK][9] = 76,
+	[1][0][0][0][RTW89_IC][9] = 127,
+	[1][0][0][0][RTW89_KCC][9] = 68,
+	[1][0][0][0][RTW89_ACMA][9] = 58,
+	[1][0][0][0][RTW89_CN][9] = 60,
+	[1][0][0][0][RTW89_UK][9] = 58,
+	[1][0][0][0][RTW89_FCC][10] = 127,
+	[1][0][0][0][RTW89_ETSI][10] = 58,
+	[1][0][0][0][RTW89_MKK][10] = 66,
+	[1][0][0][0][RTW89_IC][10] = 127,
+	[1][0][0][0][RTW89_KCC][10] = 68,
+	[1][0][0][0][RTW89_ACMA][10] = 58,
+	[1][0][0][0][RTW89_CN][10] = 60,
+	[1][0][0][0][RTW89_UK][10] = 58,
+	[1][0][0][0][RTW89_FCC][11] = 127,
+	[1][0][0][0][RTW89_ETSI][11] = 127,
+	[1][0][0][0][RTW89_MKK][11] = 127,
+	[1][0][0][0][RTW89_IC][11] = 127,
+	[1][0][0][0][RTW89_KCC][11] = 127,
+	[1][0][0][0][RTW89_ACMA][11] = 127,
+	[1][0][0][0][RTW89_CN][11] = 127,
+	[1][0][0][0][RTW89_UK][11] = 127,
+	[1][0][0][0][RTW89_FCC][12] = 127,
+	[1][0][0][0][RTW89_ETSI][12] = 127,
+	[1][0][0][0][RTW89_MKK][12] = 127,
+	[1][0][0][0][RTW89_IC][12] = 127,
+	[1][0][0][0][RTW89_KCC][12] = 127,
+	[1][0][0][0][RTW89_ACMA][12] = 127,
+	[1][0][0][0][RTW89_CN][12] = 127,
+	[1][0][0][0][RTW89_UK][12] = 127,
+	[1][0][0][0][RTW89_FCC][13] = 127,
+	[1][0][0][0][RTW89_ETSI][13] = 127,
+	[1][0][0][0][RTW89_MKK][13] = 127,
+	[1][0][0][0][RTW89_IC][13] = 127,
+	[1][0][0][0][RTW89_KCC][13] = 127,
+	[1][0][0][0][RTW89_ACMA][13] = 127,
+	[1][0][0][0][RTW89_CN][13] = 127,
+	[1][0][0][0][RTW89_UK][13] = 127,
+	[1][1][0][0][RTW89_FCC][0] = 127,
+	[1][1][0][0][RTW89_ETSI][0] = 127,
+	[1][1][0][0][RTW89_MKK][0] = 127,
+	[1][1][0][0][RTW89_IC][0] = 127,
+	[1][1][0][0][RTW89_KCC][0] = 127,
+	[1][1][0][0][RTW89_ACMA][0] = 127,
+	[1][1][0][0][RTW89_CN][0] = 127,
+	[1][1][0][0][RTW89_UK][0] = 127,
+	[1][1][0][0][RTW89_FCC][1] = 127,
+	[1][1][0][0][RTW89_ETSI][1] = 127,
+	[1][1][0][0][RTW89_MKK][1] = 127,
+	[1][1][0][0][RTW89_IC][1] = 127,
+	[1][1][0][0][RTW89_KCC][1] = 127,
+	[1][1][0][0][RTW89_ACMA][1] = 127,
+	[1][1][0][0][RTW89_CN][1] = 127,
+	[1][1][0][0][RTW89_UK][1] = 127,
+	[1][1][0][0][RTW89_FCC][2] = 127,
+	[1][1][0][0][RTW89_ETSI][2] = 127,
+	[1][1][0][0][RTW89_MKK][2] = 127,
+	[1][1][0][0][RTW89_IC][2] = 127,
+	[1][1][0][0][RTW89_KCC][2] = 127,
+	[1][1][0][0][RTW89_ACMA][2] = 127,
+	[1][1][0][0][RTW89_CN][2] = 127,
+	[1][1][0][0][RTW89_UK][2] = 127,
+	[1][1][0][0][RTW89_FCC][3] = 127,
+	[1][1][0][0][RTW89_ETSI][3] = 127,
+	[1][1][0][0][RTW89_MKK][3] = 127,
+	[1][1][0][0][RTW89_IC][3] = 127,
+	[1][1][0][0][RTW89_KCC][3] = 127,
+	[1][1][0][0][RTW89_ACMA][3] = 127,
+	[1][1][0][0][RTW89_CN][3] = 127,
+	[1][1][0][0][RTW89_UK][3] = 127,
+	[1][1][0][0][RTW89_FCC][4] = 127,
+	[1][1][0][0][RTW89_ETSI][4] = 127,
+	[1][1][0][0][RTW89_MKK][4] = 127,
+	[1][1][0][0][RTW89_IC][4] = 127,
+	[1][1][0][0][RTW89_KCC][4] = 127,
+	[1][1][0][0][RTW89_ACMA][4] = 127,
+	[1][1][0][0][RTW89_CN][4] = 127,
+	[1][1][0][0][RTW89_UK][4] = 127,
+	[1][1][0][0][RTW89_FCC][5] = 127,
+	[1][1][0][0][RTW89_ETSI][5] = 127,
+	[1][1][0][0][RTW89_MKK][5] = 127,
+	[1][1][0][0][RTW89_IC][5] = 127,
+	[1][1][0][0][RTW89_KCC][5] = 127,
+	[1][1][0][0][RTW89_ACMA][5] = 127,
+	[1][1][0][0][RTW89_CN][5] = 127,
+	[1][1][0][0][RTW89_UK][5] = 127,
+	[1][1][0][0][RTW89_FCC][6] = 127,
+	[1][1][0][0][RTW89_ETSI][6] = 127,
+	[1][1][0][0][RTW89_MKK][6] = 127,
+	[1][1][0][0][RTW89_IC][6] = 127,
+	[1][1][0][0][RTW89_KCC][6] = 127,
+	[1][1][0][0][RTW89_ACMA][6] = 127,
+	[1][1][0][0][RTW89_CN][6] = 127,
+	[1][1][0][0][RTW89_UK][6] = 127,
+	[1][1][0][0][RTW89_FCC][7] = 127,
+	[1][1][0][0][RTW89_ETSI][7] = 127,
+	[1][1][0][0][RTW89_MKK][7] = 127,
+	[1][1][0][0][RTW89_IC][7] = 127,
+	[1][1][0][0][RTW89_KCC][7] = 127,
+	[1][1][0][0][RTW89_ACMA][7] = 127,
+	[1][1][0][0][RTW89_CN][7] = 127,
+	[1][1][0][0][RTW89_UK][7] = 127,
+	[1][1][0][0][RTW89_FCC][8] = 127,
+	[1][1][0][0][RTW89_ETSI][8] = 127,
+	[1][1][0][0][RTW89_MKK][8] = 127,
+	[1][1][0][0][RTW89_IC][8] = 127,
+	[1][1][0][0][RTW89_KCC][8] = 127,
+	[1][1][0][0][RTW89_ACMA][8] = 127,
+	[1][1][0][0][RTW89_CN][8] = 127,
+	[1][1][0][0][RTW89_UK][8] = 127,
+	[1][1][0][0][RTW89_FCC][9] = 127,
+	[1][1][0][0][RTW89_ETSI][9] = 127,
+	[1][1][0][0][RTW89_MKK][9] = 127,
+	[1][1][0][0][RTW89_IC][9] = 127,
+	[1][1][0][0][RTW89_KCC][9] = 127,
+	[1][1][0][0][RTW89_ACMA][9] = 127,
+	[1][1][0][0][RTW89_CN][9] = 127,
+	[1][1][0][0][RTW89_UK][9] = 127,
+	[1][1][0][0][RTW89_FCC][10] = 127,
+	[1][1][0][0][RTW89_ETSI][10] = 127,
+	[1][1][0][0][RTW89_MKK][10] = 127,
+	[1][1][0][0][RTW89_IC][10] = 127,
+	[1][1][0][0][RTW89_KCC][10] = 127,
+	[1][1][0][0][RTW89_ACMA][10] = 127,
+	[1][1][0][0][RTW89_CN][10] = 127,
+	[1][1][0][0][RTW89_UK][10] = 127,
+	[1][1][0][0][RTW89_FCC][11] = 127,
+	[1][1][0][0][RTW89_ETSI][11] = 127,
+	[1][1][0][0][RTW89_MKK][11] = 127,
+	[1][1][0][0][RTW89_IC][11] = 127,
+	[1][1][0][0][RTW89_KCC][11] = 127,
+	[1][1][0][0][RTW89_ACMA][11] = 127,
+	[1][1][0][0][RTW89_CN][11] = 127,
+	[1][1][0][0][RTW89_UK][11] = 127,
+	[1][1][0][0][RTW89_FCC][12] = 127,
+	[1][1][0][0][RTW89_ETSI][12] = 127,
+	[1][1][0][0][RTW89_MKK][12] = 127,
+	[1][1][0][0][RTW89_IC][12] = 127,
+	[1][1][0][0][RTW89_KCC][12] = 127,
+	[1][1][0][0][RTW89_ACMA][12] = 127,
+	[1][1][0][0][RTW89_CN][12] = 127,
+	[1][1][0][0][RTW89_UK][12] = 127,
+	[1][1][0][0][RTW89_FCC][13] = 127,
+	[1][1][0][0][RTW89_ETSI][13] = 127,
+	[1][1][0][0][RTW89_MKK][13] = 127,
+	[1][1][0][0][RTW89_IC][13] = 127,
+	[1][1][0][0][RTW89_KCC][13] = 127,
+	[1][1][0][0][RTW89_ACMA][13] = 127,
+	[1][1][0][0][RTW89_CN][13] = 127,
+	[1][1][0][0][RTW89_UK][13] = 127,
+	[0][0][1][0][RTW89_FCC][0] = 78,
+	[0][0][1][0][RTW89_ETSI][0] = 58,
+	[0][0][1][0][RTW89_MKK][0] = 72,
+	[0][0][1][0][RTW89_IC][0] = 78,
+	[0][0][1][0][RTW89_KCC][0] = 76,
+	[0][0][1][0][RTW89_ACMA][0] = 58,
+	[0][0][1][0][RTW89_CN][0] = 60,
+	[0][0][1][0][RTW89_UK][0] = 58,
+	[0][0][1][0][RTW89_FCC][1] = 78,
+	[0][0][1][0][RTW89_ETSI][1] = 60,
+	[0][0][1][0][RTW89_MKK][1] = 74,
+	[0][0][1][0][RTW89_IC][1] = 78,
+	[0][0][1][0][RTW89_KCC][1] = 76,
+	[0][0][1][0][RTW89_ACMA][1] = 60,
+	[0][0][1][0][RTW89_CN][1] = 60,
+	[0][0][1][0][RTW89_UK][1] = 60,
+	[0][0][1][0][RTW89_FCC][2] = 80,
+	[0][0][1][0][RTW89_ETSI][2] = 60,
+	[0][0][1][0][RTW89_MKK][2] = 74,
+	[0][0][1][0][RTW89_IC][2] = 80,
+	[0][0][1][0][RTW89_KCC][2] = 76,
+	[0][0][1][0][RTW89_ACMA][2] = 60,
+	[0][0][1][0][RTW89_CN][2] = 60,
+	[0][0][1][0][RTW89_UK][2] = 60,
+	[0][0][1][0][RTW89_FCC][3] = 80,
+	[0][0][1][0][RTW89_ETSI][3] = 60,
+	[0][0][1][0][RTW89_MKK][3] = 74,
+	[0][0][1][0][RTW89_IC][3] = 80,
+	[0][0][1][0][RTW89_KCC][3] = 76,
+	[0][0][1][0][RTW89_ACMA][3] = 60,
+	[0][0][1][0][RTW89_CN][3] = 60,
+	[0][0][1][0][RTW89_UK][3] = 60,
+	[0][0][1][0][RTW89_FCC][4] = 80,
+	[0][0][1][0][RTW89_ETSI][4] = 60,
+	[0][0][1][0][RTW89_MKK][4] = 74,
+	[0][0][1][0][RTW89_IC][4] = 80,
+	[0][0][1][0][RTW89_KCC][4] = 76,
+	[0][0][1][0][RTW89_ACMA][4] = 60,
+	[0][0][1][0][RTW89_CN][4] = 60,
+	[0][0][1][0][RTW89_UK][4] = 60,
+	[0][0][1][0][RTW89_FCC][5] = 80,
+	[0][0][1][0][RTW89_ETSI][5] = 60,
+	[0][0][1][0][RTW89_MKK][5] = 74,
+	[0][0][1][0][RTW89_IC][5] = 80,
+	[0][0][1][0][RTW89_KCC][5] = 76,
+	[0][0][1][0][RTW89_ACMA][5] = 60,
+	[0][0][1][0][RTW89_CN][5] = 60,
+	[0][0][1][0][RTW89_UK][5] = 60,
+	[0][0][1][0][RTW89_FCC][6] = 80,
+	[0][0][1][0][RTW89_ETSI][6] = 60,
+	[0][0][1][0][RTW89_MKK][6] = 74,
+	[0][0][1][0][RTW89_IC][6] = 80,
+	[0][0][1][0][RTW89_KCC][6] = 76,
+	[0][0][1][0][RTW89_ACMA][6] = 60,
+	[0][0][1][0][RTW89_CN][6] = 60,
+	[0][0][1][0][RTW89_UK][6] = 60,
+	[0][0][1][0][RTW89_FCC][7] = 80,
+	[0][0][1][0][RTW89_ETSI][7] = 60,
+	[0][0][1][0][RTW89_MKK][7] = 74,
+	[0][0][1][0][RTW89_IC][7] = 80,
+	[0][0][1][0][RTW89_KCC][7] = 76,
+	[0][0][1][0][RTW89_ACMA][7] = 60,
+	[0][0][1][0][RTW89_CN][7] = 60,
+	[0][0][1][0][RTW89_UK][7] = 60,
+	[0][0][1][0][RTW89_FCC][8] = 80,
+	[0][0][1][0][RTW89_ETSI][8] = 60,
+	[0][0][1][0][RTW89_MKK][8] = 74,
+	[0][0][1][0][RTW89_IC][8] = 80,
+	[0][0][1][0][RTW89_KCC][8] = 76,
+	[0][0][1][0][RTW89_ACMA][8] = 60,
+	[0][0][1][0][RTW89_CN][8] = 60,
+	[0][0][1][0][RTW89_UK][8] = 60,
+	[0][0][1][0][RTW89_FCC][9] = 76,
+	[0][0][1][0][RTW89_ETSI][9] = 60,
+	[0][0][1][0][RTW89_MKK][9] = 74,
+	[0][0][1][0][RTW89_IC][9] = 76,
+	[0][0][1][0][RTW89_KCC][9] = 74,
+	[0][0][1][0][RTW89_ACMA][9] = 60,
+	[0][0][1][0][RTW89_CN][9] = 60,
+	[0][0][1][0][RTW89_UK][9] = 60,
+	[0][0][1][0][RTW89_FCC][10] = 76,
+	[0][0][1][0][RTW89_ETSI][10] = 60,
+	[0][0][1][0][RTW89_MKK][10] = 74,
+	[0][0][1][0][RTW89_IC][10] = 76,
+	[0][0][1][0][RTW89_KCC][10] = 74,
+	[0][0][1][0][RTW89_ACMA][10] = 60,
+	[0][0][1][0][RTW89_CN][10] = 60,
+	[0][0][1][0][RTW89_UK][10] = 60,
+	[0][0][1][0][RTW89_FCC][11] = 68,
+	[0][0][1][0][RTW89_ETSI][11] = 60,
+	[0][0][1][0][RTW89_MKK][11] = 74,
+	[0][0][1][0][RTW89_IC][11] = 68,
+	[0][0][1][0][RTW89_KCC][11] = 74,
+	[0][0][1][0][RTW89_ACMA][11] = 60,
+	[0][0][1][0][RTW89_CN][11] = 60,
+	[0][0][1][0][RTW89_UK][11] = 60,
+	[0][0][1][0][RTW89_FCC][12] = 64,
+	[0][0][1][0][RTW89_ETSI][12] = 58,
+	[0][0][1][0][RTW89_MKK][12] = 70,
+	[0][0][1][0][RTW89_IC][12] = 64,
+	[0][0][1][0][RTW89_KCC][12] = 74,
+	[0][0][1][0][RTW89_ACMA][12] = 58,
+	[0][0][1][0][RTW89_CN][12] = 60,
+	[0][0][1][0][RTW89_UK][12] = 58,
+	[0][0][1][0][RTW89_FCC][13] = 127,
+	[0][0][1][0][RTW89_ETSI][13] = 127,
+	[0][0][1][0][RTW89_MKK][13] = 127,
+	[0][0][1][0][RTW89_IC][13] = 127,
+	[0][0][1][0][RTW89_KCC][13] = 127,
+	[0][0][1][0][RTW89_ACMA][13] = 127,
+	[0][0][1][0][RTW89_CN][13] = 127,
+	[0][0][1][0][RTW89_UK][13] = 127,
+	[0][1][1][0][RTW89_FCC][0] = 127,
+	[0][1][1][0][RTW89_ETSI][0] = 127,
+	[0][1][1][0][RTW89_MKK][0] = 127,
+	[0][1][1][0][RTW89_IC][0] = 127,
+	[0][1][1][0][RTW89_KCC][0] = 127,
+	[0][1][1][0][RTW89_ACMA][0] = 127,
+	[0][1][1][0][RTW89_CN][0] = 127,
+	[0][1][1][0][RTW89_UK][0] = 127,
+	[0][1][1][0][RTW89_FCC][1] = 127,
+	[0][1][1][0][RTW89_ETSI][1] = 127,
+	[0][1][1][0][RTW89_MKK][1] = 127,
+	[0][1][1][0][RTW89_IC][1] = 127,
+	[0][1][1][0][RTW89_KCC][1] = 127,
+	[0][1][1][0][RTW89_ACMA][1] = 127,
+	[0][1][1][0][RTW89_CN][1] = 127,
+	[0][1][1][0][RTW89_UK][1] = 127,
+	[0][1][1][0][RTW89_FCC][2] = 127,
+	[0][1][1][0][RTW89_ETSI][2] = 127,
+	[0][1][1][0][RTW89_MKK][2] = 127,
+	[0][1][1][0][RTW89_IC][2] = 127,
+	[0][1][1][0][RTW89_KCC][2] = 127,
+	[0][1][1][0][RTW89_ACMA][2] = 127,
+	[0][1][1][0][RTW89_CN][2] = 127,
+	[0][1][1][0][RTW89_UK][2] = 127,
+	[0][1][1][0][RTW89_FCC][3] = 127,
+	[0][1][1][0][RTW89_ETSI][3] = 127,
+	[0][1][1][0][RTW89_MKK][3] = 127,
+	[0][1][1][0][RTW89_IC][3] = 127,
+	[0][1][1][0][RTW89_KCC][3] = 127,
+	[0][1][1][0][RTW89_ACMA][3] = 127,
+	[0][1][1][0][RTW89_CN][3] = 127,
+	[0][1][1][0][RTW89_UK][3] = 127,
+	[0][1][1][0][RTW89_FCC][4] = 127,
+	[0][1][1][0][RTW89_ETSI][4] = 127,
+	[0][1][1][0][RTW89_MKK][4] = 127,
+	[0][1][1][0][RTW89_IC][4] = 127,
+	[0][1][1][0][RTW89_KCC][4] = 127,
+	[0][1][1][0][RTW89_ACMA][4] = 127,
+	[0][1][1][0][RTW89_CN][4] = 127,
+	[0][1][1][0][RTW89_UK][4] = 127,
+	[0][1][1][0][RTW89_FCC][5] = 127,
+	[0][1][1][0][RTW89_ETSI][5] = 127,
+	[0][1][1][0][RTW89_MKK][5] = 127,
+	[0][1][1][0][RTW89_IC][5] = 127,
+	[0][1][1][0][RTW89_KCC][5] = 127,
+	[0][1][1][0][RTW89_ACMA][5] = 127,
+	[0][1][1][0][RTW89_CN][5] = 127,
+	[0][1][1][0][RTW89_UK][5] = 127,
+	[0][1][1][0][RTW89_FCC][6] = 127,
+	[0][1][1][0][RTW89_ETSI][6] = 127,
+	[0][1][1][0][RTW89_MKK][6] = 127,
+	[0][1][1][0][RTW89_IC][6] = 127,
+	[0][1][1][0][RTW89_KCC][6] = 127,
+	[0][1][1][0][RTW89_ACMA][6] = 127,
+	[0][1][1][0][RTW89_CN][6] = 127,
+	[0][1][1][0][RTW89_UK][6] = 127,
+	[0][1][1][0][RTW89_FCC][7] = 127,
+	[0][1][1][0][RTW89_ETSI][7] = 127,
+	[0][1][1][0][RTW89_MKK][7] = 127,
+	[0][1][1][0][RTW89_IC][7] = 127,
+	[0][1][1][0][RTW89_KCC][7] = 127,
+	[0][1][1][0][RTW89_ACMA][7] = 127,
+	[0][1][1][0][RTW89_CN][7] = 127,
+	[0][1][1][0][RTW89_UK][7] = 127,
+	[0][1][1][0][RTW89_FCC][8] = 127,
+	[0][1][1][0][RTW89_ETSI][8] = 127,
+	[0][1][1][0][RTW89_MKK][8] = 127,
+	[0][1][1][0][RTW89_IC][8] = 127,
+	[0][1][1][0][RTW89_KCC][8] = 127,
+	[0][1][1][0][RTW89_ACMA][8] = 127,
+	[0][1][1][0][RTW89_CN][8] = 127,
+	[0][1][1][0][RTW89_UK][8] = 127,
+	[0][1][1][0][RTW89_FCC][9] = 127,
+	[0][1][1][0][RTW89_ETSI][9] = 127,
+	[0][1][1][0][RTW89_MKK][9] = 127,
+	[0][1][1][0][RTW89_IC][9] = 127,
+	[0][1][1][0][RTW89_KCC][9] = 127,
+	[0][1][1][0][RTW89_ACMA][9] = 127,
+	[0][1][1][0][RTW89_CN][9] = 127,
+	[0][1][1][0][RTW89_UK][9] = 127,
+	[0][1][1][0][RTW89_FCC][10] = 127,
+	[0][1][1][0][RTW89_ETSI][10] = 127,
+	[0][1][1][0][RTW89_MKK][10] = 127,
+	[0][1][1][0][RTW89_IC][10] = 127,
+	[0][1][1][0][RTW89_KCC][10] = 127,
+	[0][1][1][0][RTW89_ACMA][10] = 127,
+	[0][1][1][0][RTW89_CN][10] = 127,
+	[0][1][1][0][RTW89_UK][10] = 127,
+	[0][1][1][0][RTW89_FCC][11] = 127,
+	[0][1][1][0][RTW89_ETSI][11] = 127,
+	[0][1][1][0][RTW89_MKK][11] = 127,
+	[0][1][1][0][RTW89_IC][11] = 127,
+	[0][1][1][0][RTW89_KCC][11] = 127,
+	[0][1][1][0][RTW89_ACMA][11] = 127,
+	[0][1][1][0][RTW89_CN][11] = 127,
+	[0][1][1][0][RTW89_UK][11] = 127,
+	[0][1][1][0][RTW89_FCC][12] = 127,
+	[0][1][1][0][RTW89_ETSI][12] = 127,
+	[0][1][1][0][RTW89_MKK][12] = 127,
+	[0][1][1][0][RTW89_IC][12] = 127,
+	[0][1][1][0][RTW89_KCC][12] = 127,
+	[0][1][1][0][RTW89_ACMA][12] = 127,
+	[0][1][1][0][RTW89_CN][12] = 127,
+	[0][1][1][0][RTW89_UK][12] = 127,
+	[0][1][1][0][RTW89_FCC][13] = 127,
+	[0][1][1][0][RTW89_ETSI][13] = 127,
+	[0][1][1][0][RTW89_MKK][13] = 127,
+	[0][1][1][0][RTW89_IC][13] = 127,
+	[0][1][1][0][RTW89_KCC][13] = 127,
+	[0][1][1][0][RTW89_ACMA][13] = 127,
+	[0][1][1][0][RTW89_CN][13] = 127,
+	[0][1][1][0][RTW89_UK][13] = 127,
+	[0][0][2][0][RTW89_FCC][0] = 78,
+	[0][0][2][0][RTW89_ETSI][0] = 60,
+	[0][0][2][0][RTW89_MKK][0] = 72,
+	[0][0][2][0][RTW89_IC][0] = 78,
+	[0][0][2][0][RTW89_KCC][0] = 76,
+	[0][0][2][0][RTW89_ACMA][0] = 60,
+	[0][0][2][0][RTW89_CN][0] = 60,
+	[0][0][2][0][RTW89_UK][0] = 60,
+	[0][0][2][0][RTW89_FCC][1] = 78,
+	[0][0][2][0][RTW89_ETSI][1] = 60,
+	[0][0][2][0][RTW89_MKK][1] = 76,
+	[0][0][2][0][RTW89_IC][1] = 78,
+	[0][0][2][0][RTW89_KCC][1] = 76,
+	[0][0][2][0][RTW89_ACMA][1] = 60,
+	[0][0][2][0][RTW89_CN][1] = 60,
+	[0][0][2][0][RTW89_UK][1] = 60,
+	[0][0][2][0][RTW89_FCC][2] = 80,
+	[0][0][2][0][RTW89_ETSI][2] = 60,
+	[0][0][2][0][RTW89_MKK][2] = 76,
+	[0][0][2][0][RTW89_IC][2] = 80,
+	[0][0][2][0][RTW89_KCC][2] = 76,
+	[0][0][2][0][RTW89_ACMA][2] = 60,
+	[0][0][2][0][RTW89_CN][2] = 60,
+	[0][0][2][0][RTW89_UK][2] = 60,
+	[0][0][2][0][RTW89_FCC][3] = 80,
+	[0][0][2][0][RTW89_ETSI][3] = 60,
+	[0][0][2][0][RTW89_MKK][3] = 76,
+	[0][0][2][0][RTW89_IC][3] = 80,
+	[0][0][2][0][RTW89_KCC][3] = 76,
+	[0][0][2][0][RTW89_ACMA][3] = 60,
+	[0][0][2][0][RTW89_CN][3] = 60,
+	[0][0][2][0][RTW89_UK][3] = 60,
+	[0][0][2][0][RTW89_FCC][4] = 80,
+	[0][0][2][0][RTW89_ETSI][4] = 60,
+	[0][0][2][0][RTW89_MKK][4] = 76,
+	[0][0][2][0][RTW89_IC][4] = 80,
+	[0][0][2][0][RTW89_KCC][4] = 76,
+	[0][0][2][0][RTW89_ACMA][4] = 60,
+	[0][0][2][0][RTW89_CN][4] = 60,
+	[0][0][2][0][RTW89_UK][4] = 60,
+	[0][0][2][0][RTW89_FCC][5] = 80,
+	[0][0][2][0][RTW89_ETSI][5] = 60,
+	[0][0][2][0][RTW89_MKK][5] = 76,
+	[0][0][2][0][RTW89_IC][5] = 80,
+	[0][0][2][0][RTW89_KCC][5] = 76,
+	[0][0][2][0][RTW89_ACMA][5] = 60,
+	[0][0][2][0][RTW89_CN][5] = 60,
+	[0][0][2][0][RTW89_UK][5] = 60,
+	[0][0][2][0][RTW89_FCC][6] = 80,
+	[0][0][2][0][RTW89_ETSI][6] = 60,
+	[0][0][2][0][RTW89_MKK][6] = 76,
+	[0][0][2][0][RTW89_IC][6] = 80,
+	[0][0][2][0][RTW89_KCC][6] = 76,
+	[0][0][2][0][RTW89_ACMA][6] = 60,
+	[0][0][2][0][RTW89_CN][6] = 60,
+	[0][0][2][0][RTW89_UK][6] = 60,
+	[0][0][2][0][RTW89_FCC][7] = 80,
+	[0][0][2][0][RTW89_ETSI][7] = 60,
+	[0][0][2][0][RTW89_MKK][7] = 76,
+	[0][0][2][0][RTW89_IC][7] = 80,
+	[0][0][2][0][RTW89_KCC][7] = 76,
+	[0][0][2][0][RTW89_ACMA][7] = 60,
+	[0][0][2][0][RTW89_CN][7] = 60,
+	[0][0][2][0][RTW89_UK][7] = 60,
+	[0][0][2][0][RTW89_FCC][8] = 78,
+	[0][0][2][0][RTW89_ETSI][8] = 60,
+	[0][0][2][0][RTW89_MKK][8] = 76,
+	[0][0][2][0][RTW89_IC][8] = 78,
+	[0][0][2][0][RTW89_KCC][8] = 76,
+	[0][0][2][0][RTW89_ACMA][8] = 60,
+	[0][0][2][0][RTW89_CN][8] = 60,
+	[0][0][2][0][RTW89_UK][8] = 60,
+	[0][0][2][0][RTW89_FCC][9] = 74,
+	[0][0][2][0][RTW89_ETSI][9] = 60,
+	[0][0][2][0][RTW89_MKK][9] = 76,
+	[0][0][2][0][RTW89_IC][9] = 74,
+	[0][0][2][0][RTW89_KCC][9] = 76,
+	[0][0][2][0][RTW89_ACMA][9] = 60,
+	[0][0][2][0][RTW89_CN][9] = 60,
+	[0][0][2][0][RTW89_UK][9] = 60,
+	[0][0][2][0][RTW89_FCC][10] = 74,
+	[0][0][2][0][RTW89_ETSI][10] = 60,
+	[0][0][2][0][RTW89_MKK][10] = 76,
+	[0][0][2][0][RTW89_IC][10] = 74,
+	[0][0][2][0][RTW89_KCC][10] = 76,
+	[0][0][2][0][RTW89_ACMA][10] = 60,
+	[0][0][2][0][RTW89_CN][10] = 60,
+	[0][0][2][0][RTW89_UK][10] = 60,
+	[0][0][2][0][RTW89_FCC][11] = 68,
+	[0][0][2][0][RTW89_ETSI][11] = 60,
+	[0][0][2][0][RTW89_MKK][11] = 76,
+	[0][0][2][0][RTW89_IC][11] = 68,
+	[0][0][2][0][RTW89_KCC][11] = 76,
+	[0][0][2][0][RTW89_ACMA][11] = 60,
+	[0][0][2][0][RTW89_CN][11] = 60,
+	[0][0][2][0][RTW89_UK][11] = 60,
+	[0][0][2][0][RTW89_FCC][12] = 68,
+	[0][0][2][0][RTW89_ETSI][12] = 60,
+	[0][0][2][0][RTW89_MKK][12] = 70,
+	[0][0][2][0][RTW89_IC][12] = 68,
+	[0][0][2][0][RTW89_KCC][12] = 76,
+	[0][0][2][0][RTW89_ACMA][12] = 60,
+	[0][0][2][0][RTW89_CN][12] = 60,
+	[0][0][2][0][RTW89_UK][12] = 60,
+	[0][0][2][0][RTW89_FCC][13] = 127,
+	[0][0][2][0][RTW89_ETSI][13] = 127,
+	[0][0][2][0][RTW89_MKK][13] = 127,
+	[0][0][2][0][RTW89_IC][13] = 127,
+	[0][0][2][0][RTW89_KCC][13] = 127,
+	[0][0][2][0][RTW89_ACMA][13] = 127,
+	[0][0][2][0][RTW89_CN][13] = 127,
+	[0][0][2][0][RTW89_UK][13] = 127,
+	[0][1][2][0][RTW89_FCC][0] = 127,
+	[0][1][2][0][RTW89_ETSI][0] = 127,
+	[0][1][2][0][RTW89_MKK][0] = 127,
+	[0][1][2][0][RTW89_IC][0] = 127,
+	[0][1][2][0][RTW89_KCC][0] = 127,
+	[0][1][2][0][RTW89_ACMA][0] = 127,
+	[0][1][2][0][RTW89_CN][0] = 127,
+	[0][1][2][0][RTW89_UK][0] = 127,
+	[0][1][2][0][RTW89_FCC][1] = 127,
+	[0][1][2][0][RTW89_ETSI][1] = 127,
+	[0][1][2][0][RTW89_MKK][1] = 127,
+	[0][1][2][0][RTW89_IC][1] = 127,
+	[0][1][2][0][RTW89_KCC][1] = 127,
+	[0][1][2][0][RTW89_ACMA][1] = 127,
+	[0][1][2][0][RTW89_CN][1] = 127,
+	[0][1][2][0][RTW89_UK][1] = 127,
+	[0][1][2][0][RTW89_FCC][2] = 127,
+	[0][1][2][0][RTW89_ETSI][2] = 127,
+	[0][1][2][0][RTW89_MKK][2] = 127,
+	[0][1][2][0][RTW89_IC][2] = 127,
+	[0][1][2][0][RTW89_KCC][2] = 127,
+	[0][1][2][0][RTW89_ACMA][2] = 127,
+	[0][1][2][0][RTW89_CN][2] = 127,
+	[0][1][2][0][RTW89_UK][2] = 127,
+	[0][1][2][0][RTW89_FCC][3] = 127,
+	[0][1][2][0][RTW89_ETSI][3] = 127,
+	[0][1][2][0][RTW89_MKK][3] = 127,
+	[0][1][2][0][RTW89_IC][3] = 127,
+	[0][1][2][0][RTW89_KCC][3] = 127,
+	[0][1][2][0][RTW89_ACMA][3] = 127,
+	[0][1][2][0][RTW89_CN][3] = 127,
+	[0][1][2][0][RTW89_UK][3] = 127,
+	[0][1][2][0][RTW89_FCC][4] = 127,
+	[0][1][2][0][RTW89_ETSI][4] = 127,
+	[0][1][2][0][RTW89_MKK][4] = 127,
+	[0][1][2][0][RTW89_IC][4] = 127,
+	[0][1][2][0][RTW89_KCC][4] = 127,
+	[0][1][2][0][RTW89_ACMA][4] = 127,
+	[0][1][2][0][RTW89_CN][4] = 127,
+	[0][1][2][0][RTW89_UK][4] = 127,
+	[0][1][2][0][RTW89_FCC][5] = 127,
+	[0][1][2][0][RTW89_ETSI][5] = 127,
+	[0][1][2][0][RTW89_MKK][5] = 127,
+	[0][1][2][0][RTW89_IC][5] = 127,
+	[0][1][2][0][RTW89_KCC][5] = 127,
+	[0][1][2][0][RTW89_ACMA][5] = 127,
+	[0][1][2][0][RTW89_CN][5] = 127,
+	[0][1][2][0][RTW89_UK][5] = 127,
+	[0][1][2][0][RTW89_FCC][6] = 127,
+	[0][1][2][0][RTW89_ETSI][6] = 127,
+	[0][1][2][0][RTW89_MKK][6] = 127,
+	[0][1][2][0][RTW89_IC][6] = 127,
+	[0][1][2][0][RTW89_KCC][6] = 127,
+	[0][1][2][0][RTW89_ACMA][6] = 127,
+	[0][1][2][0][RTW89_CN][6] = 127,
+	[0][1][2][0][RTW89_UK][6] = 127,
+	[0][1][2][0][RTW89_FCC][7] = 127,
+	[0][1][2][0][RTW89_ETSI][7] = 127,
+	[0][1][2][0][RTW89_MKK][7] = 127,
+	[0][1][2][0][RTW89_IC][7] = 127,
+	[0][1][2][0][RTW89_KCC][7] = 127,
+	[0][1][2][0][RTW89_ACMA][7] = 127,
+	[0][1][2][0][RTW89_CN][7] = 127,
+	[0][1][2][0][RTW89_UK][7] = 127,
+	[0][1][2][0][RTW89_FCC][8] = 127,
+	[0][1][2][0][RTW89_ETSI][8] = 127,
+	[0][1][2][0][RTW89_MKK][8] = 127,
+	[0][1][2][0][RTW89_IC][8] = 127,
+	[0][1][2][0][RTW89_KCC][8] = 127,
+	[0][1][2][0][RTW89_ACMA][8] = 127,
+	[0][1][2][0][RTW89_CN][8] = 127,
+	[0][1][2][0][RTW89_UK][8] = 127,
+	[0][1][2][0][RTW89_FCC][9] = 127,
+	[0][1][2][0][RTW89_ETSI][9] = 127,
+	[0][1][2][0][RTW89_MKK][9] = 127,
+	[0][1][2][0][RTW89_IC][9] = 127,
+	[0][1][2][0][RTW89_KCC][9] = 127,
+	[0][1][2][0][RTW89_ACMA][9] = 127,
+	[0][1][2][0][RTW89_CN][9] = 127,
+	[0][1][2][0][RTW89_UK][9] = 127,
+	[0][1][2][0][RTW89_FCC][10] = 127,
+	[0][1][2][0][RTW89_ETSI][10] = 127,
+	[0][1][2][0][RTW89_MKK][10] = 127,
+	[0][1][2][0][RTW89_IC][10] = 127,
+	[0][1][2][0][RTW89_KCC][10] = 127,
+	[0][1][2][0][RTW89_ACMA][10] = 127,
+	[0][1][2][0][RTW89_CN][10] = 127,
+	[0][1][2][0][RTW89_UK][10] = 127,
+	[0][1][2][0][RTW89_FCC][11] = 127,
+	[0][1][2][0][RTW89_ETSI][11] = 127,
+	[0][1][2][0][RTW89_MKK][11] = 127,
+	[0][1][2][0][RTW89_IC][11] = 127,
+	[0][1][2][0][RTW89_KCC][11] = 127,
+	[0][1][2][0][RTW89_ACMA][11] = 127,
+	[0][1][2][0][RTW89_CN][11] = 127,
+	[0][1][2][0][RTW89_UK][11] = 127,
+	[0][1][2][0][RTW89_FCC][12] = 127,
+	[0][1][2][0][RTW89_ETSI][12] = 127,
+	[0][1][2][0][RTW89_MKK][12] = 127,
+	[0][1][2][0][RTW89_IC][12] = 127,
+	[0][1][2][0][RTW89_KCC][12] = 127,
+	[0][1][2][0][RTW89_ACMA][12] = 127,
+	[0][1][2][0][RTW89_CN][12] = 127,
+	[0][1][2][0][RTW89_UK][12] = 127,
+	[0][1][2][0][RTW89_FCC][13] = 127,
+	[0][1][2][0][RTW89_ETSI][13] = 127,
+	[0][1][2][0][RTW89_MKK][13] = 127,
+	[0][1][2][0][RTW89_IC][13] = 127,
+	[0][1][2][0][RTW89_KCC][13] = 127,
+	[0][1][2][0][RTW89_ACMA][13] = 127,
+	[0][1][2][0][RTW89_CN][13] = 127,
+	[0][1][2][0][RTW89_UK][13] = 127,
+	[0][1][2][1][RTW89_FCC][0] = 127,
+	[0][1][2][1][RTW89_ETSI][0] = 127,
+	[0][1][2][1][RTW89_MKK][0] = 127,
+	[0][1][2][1][RTW89_IC][0] = 127,
+	[0][1][2][1][RTW89_KCC][0] = 127,
+	[0][1][2][1][RTW89_ACMA][0] = 127,
+	[0][1][2][1][RTW89_CN][0] = 127,
+	[0][1][2][1][RTW89_UK][0] = 127,
+	[0][1][2][1][RTW89_FCC][1] = 127,
+	[0][1][2][1][RTW89_ETSI][1] = 127,
+	[0][1][2][1][RTW89_MKK][1] = 127,
+	[0][1][2][1][RTW89_IC][1] = 127,
+	[0][1][2][1][RTW89_KCC][1] = 127,
+	[0][1][2][1][RTW89_ACMA][1] = 127,
+	[0][1][2][1][RTW89_CN][1] = 127,
+	[0][1][2][1][RTW89_UK][1] = 127,
+	[0][1][2][1][RTW89_FCC][2] = 127,
+	[0][1][2][1][RTW89_ETSI][2] = 127,
+	[0][1][2][1][RTW89_MKK][2] = 127,
+	[0][1][2][1][RTW89_IC][2] = 127,
+	[0][1][2][1][RTW89_KCC][2] = 127,
+	[0][1][2][1][RTW89_ACMA][2] = 127,
+	[0][1][2][1][RTW89_CN][2] = 127,
+	[0][1][2][1][RTW89_UK][2] = 127,
+	[0][1][2][1][RTW89_FCC][3] = 127,
+	[0][1][2][1][RTW89_ETSI][3] = 127,
+	[0][1][2][1][RTW89_MKK][3] = 127,
+	[0][1][2][1][RTW89_IC][3] = 127,
+	[0][1][2][1][RTW89_KCC][3] = 127,
+	[0][1][2][1][RTW89_ACMA][3] = 127,
+	[0][1][2][1][RTW89_CN][3] = 127,
+	[0][1][2][1][RTW89_UK][3] = 127,
+	[0][1][2][1][RTW89_FCC][4] = 127,
+	[0][1][2][1][RTW89_ETSI][4] = 127,
+	[0][1][2][1][RTW89_MKK][4] = 127,
+	[0][1][2][1][RTW89_IC][4] = 127,
+	[0][1][2][1][RTW89_KCC][4] = 127,
+	[0][1][2][1][RTW89_ACMA][4] = 127,
+	[0][1][2][1][RTW89_CN][4] = 127,
+	[0][1][2][1][RTW89_UK][4] = 127,
+	[0][1][2][1][RTW89_FCC][5] = 127,
+	[0][1][2][1][RTW89_ETSI][5] = 127,
+	[0][1][2][1][RTW89_MKK][5] = 127,
+	[0][1][2][1][RTW89_IC][5] = 127,
+	[0][1][2][1][RTW89_KCC][5] = 127,
+	[0][1][2][1][RTW89_ACMA][5] = 127,
+	[0][1][2][1][RTW89_CN][5] = 127,
+	[0][1][2][1][RTW89_UK][5] = 127,
+	[0][1][2][1][RTW89_FCC][6] = 127,
+	[0][1][2][1][RTW89_ETSI][6] = 127,
+	[0][1][2][1][RTW89_MKK][6] = 127,
+	[0][1][2][1][RTW89_IC][6] = 127,
+	[0][1][2][1][RTW89_KCC][6] = 127,
+	[0][1][2][1][RTW89_ACMA][6] = 127,
+	[0][1][2][1][RTW89_CN][6] = 127,
+	[0][1][2][1][RTW89_UK][6] = 127,
+	[0][1][2][1][RTW89_FCC][7] = 127,
+	[0][1][2][1][RTW89_ETSI][7] = 127,
+	[0][1][2][1][RTW89_MKK][7] = 127,
+	[0][1][2][1][RTW89_IC][7] = 127,
+	[0][1][2][1][RTW89_KCC][7] = 127,
+	[0][1][2][1][RTW89_ACMA][7] = 127,
+	[0][1][2][1][RTW89_CN][7] = 127,
+	[0][1][2][1][RTW89_UK][7] = 127,
+	[0][1][2][1][RTW89_FCC][8] = 127,
+	[0][1][2][1][RTW89_ETSI][8] = 127,
+	[0][1][2][1][RTW89_MKK][8] = 127,
+	[0][1][2][1][RTW89_IC][8] = 127,
+	[0][1][2][1][RTW89_KCC][8] = 127,
+	[0][1][2][1][RTW89_ACMA][8] = 127,
+	[0][1][2][1][RTW89_CN][8] = 127,
+	[0][1][2][1][RTW89_UK][8] = 127,
+	[0][1][2][1][RTW89_FCC][9] = 127,
+	[0][1][2][1][RTW89_ETSI][9] = 127,
+	[0][1][2][1][RTW89_MKK][9] = 127,
+	[0][1][2][1][RTW89_IC][9] = 127,
+	[0][1][2][1][RTW89_KCC][9] = 127,
+	[0][1][2][1][RTW89_ACMA][9] = 127,
+	[0][1][2][1][RTW89_CN][9] = 127,
+	[0][1][2][1][RTW89_UK][9] = 127,
+	[0][1][2][1][RTW89_FCC][10] = 127,
+	[0][1][2][1][RTW89_ETSI][10] = 127,
+	[0][1][2][1][RTW89_MKK][10] = 127,
+	[0][1][2][1][RTW89_IC][10] = 127,
+	[0][1][2][1][RTW89_KCC][10] = 127,
+	[0][1][2][1][RTW89_ACMA][10] = 127,
+	[0][1][2][1][RTW89_CN][10] = 127,
+	[0][1][2][1][RTW89_UK][10] = 127,
+	[0][1][2][1][RTW89_FCC][11] = 127,
+	[0][1][2][1][RTW89_ETSI][11] = 127,
+	[0][1][2][1][RTW89_MKK][11] = 127,
+	[0][1][2][1][RTW89_IC][11] = 127,
+	[0][1][2][1][RTW89_KCC][11] = 127,
+	[0][1][2][1][RTW89_ACMA][11] = 127,
+	[0][1][2][1][RTW89_CN][11] = 127,
+	[0][1][2][1][RTW89_UK][11] = 127,
+	[0][1][2][1][RTW89_FCC][12] = 127,
+	[0][1][2][1][RTW89_ETSI][12] = 127,
+	[0][1][2][1][RTW89_MKK][12] = 127,
+	[0][1][2][1][RTW89_IC][12] = 127,
+	[0][1][2][1][RTW89_KCC][12] = 127,
+	[0][1][2][1][RTW89_ACMA][12] = 127,
+	[0][1][2][1][RTW89_CN][12] = 127,
+	[0][1][2][1][RTW89_UK][12] = 127,
+	[0][1][2][1][RTW89_FCC][13] = 127,
+	[0][1][2][1][RTW89_ETSI][13] = 127,
+	[0][1][2][1][RTW89_MKK][13] = 127,
+	[0][1][2][1][RTW89_IC][13] = 127,
+	[0][1][2][1][RTW89_KCC][13] = 127,
+	[0][1][2][1][RTW89_ACMA][13] = 127,
+	[0][1][2][1][RTW89_CN][13] = 127,
+	[0][1][2][1][RTW89_UK][13] = 127,
+	[1][0][2][0][RTW89_FCC][0] = 127,
+	[1][0][2][0][RTW89_ETSI][0] = 127,
+	[1][0][2][0][RTW89_MKK][0] = 127,
+	[1][0][2][0][RTW89_IC][0] = 127,
+	[1][0][2][0][RTW89_KCC][0] = 127,
+	[1][0][2][0][RTW89_ACMA][0] = 127,
+	[1][0][2][0][RTW89_CN][0] = 127,
+	[1][0][2][0][RTW89_UK][0] = 127,
+	[1][0][2][0][RTW89_FCC][1] = 127,
+	[1][0][2][0][RTW89_ETSI][1] = 127,
+	[1][0][2][0][RTW89_MKK][1] = 127,
+	[1][0][2][0][RTW89_IC][1] = 127,
+	[1][0][2][0][RTW89_KCC][1] = 127,
+	[1][0][2][0][RTW89_ACMA][1] = 127,
+	[1][0][2][0][RTW89_CN][1] = 127,
+	[1][0][2][0][RTW89_UK][1] = 127,
+	[1][0][2][0][RTW89_FCC][2] = 70,
+	[1][0][2][0][RTW89_ETSI][2] = 58,
+	[1][0][2][0][RTW89_MKK][2] = 76,
+	[1][0][2][0][RTW89_IC][2] = 70,
+	[1][0][2][0][RTW89_KCC][2] = 76,
+	[1][0][2][0][RTW89_ACMA][2] = 58,
+	[1][0][2][0][RTW89_CN][2] = 60,
+	[1][0][2][0][RTW89_UK][2] = 58,
+	[1][0][2][0][RTW89_FCC][3] = 70,
+	[1][0][2][0][RTW89_ETSI][3] = 58,
+	[1][0][2][0][RTW89_MKK][3] = 76,
+	[1][0][2][0][RTW89_IC][3] = 70,
+	[1][0][2][0][RTW89_KCC][3] = 76,
+	[1][0][2][0][RTW89_ACMA][3] = 58,
+	[1][0][2][0][RTW89_CN][3] = 60,
+	[1][0][2][0][RTW89_UK][3] = 58,
+	[1][0][2][0][RTW89_FCC][4] = 74,
+	[1][0][2][0][RTW89_ETSI][4] = 58,
+	[1][0][2][0][RTW89_MKK][4] = 76,
+	[1][0][2][0][RTW89_IC][4] = 74,
+	[1][0][2][0][RTW89_KCC][4] = 76,
+	[1][0][2][0][RTW89_ACMA][4] = 58,
+	[1][0][2][0][RTW89_CN][4] = 60,
+	[1][0][2][0][RTW89_UK][4] = 58,
+	[1][0][2][0][RTW89_FCC][5] = 76,
+	[1][0][2][0][RTW89_ETSI][5] = 58,
+	[1][0][2][0][RTW89_MKK][5] = 76,
+	[1][0][2][0][RTW89_IC][5] = 76,
+	[1][0][2][0][RTW89_KCC][5] = 76,
+	[1][0][2][0][RTW89_ACMA][5] = 58,
+	[1][0][2][0][RTW89_CN][5] = 60,
+	[1][0][2][0][RTW89_UK][5] = 58,
+	[1][0][2][0][RTW89_FCC][6] = 76,
+	[1][0][2][0][RTW89_ETSI][6] = 58,
+	[1][0][2][0][RTW89_MKK][6] = 76,
+	[1][0][2][0][RTW89_IC][6] = 76,
+	[1][0][2][0][RTW89_KCC][6] = 76,
+	[1][0][2][0][RTW89_ACMA][6] = 58,
+	[1][0][2][0][RTW89_CN][6] = 60,
+	[1][0][2][0][RTW89_UK][6] = 58,
+	[1][0][2][0][RTW89_FCC][7] = 76,
+	[1][0][2][0][RTW89_ETSI][7] = 58,
+	[1][0][2][0][RTW89_MKK][7] = 76,
+	[1][0][2][0][RTW89_IC][7] = 76,
+	[1][0][2][0][RTW89_KCC][7] = 76,
+	[1][0][2][0][RTW89_ACMA][7] = 58,
+	[1][0][2][0][RTW89_CN][7] = 60,
+	[1][0][2][0][RTW89_UK][7] = 58,
+	[1][0][2][0][RTW89_FCC][8] = 78,
+	[1][0][2][0][RTW89_ETSI][8] = 58,
+	[1][0][2][0][RTW89_MKK][8] = 76,
+	[1][0][2][0][RTW89_IC][8] = 78,
+	[1][0][2][0][RTW89_KCC][8] = 76,
+	[1][0][2][0][RTW89_ACMA][8] = 58,
+	[1][0][2][0][RTW89_CN][8] = 60,
+	[1][0][2][0][RTW89_UK][8] = 58,
+	[1][0][2][0][RTW89_FCC][9] = 74,
+	[1][0][2][0][RTW89_ETSI][9] = 58,
+	[1][0][2][0][RTW89_MKK][9] = 76,
+	[1][0][2][0][RTW89_IC][9] = 74,
+	[1][0][2][0][RTW89_KCC][9] = 76,
+	[1][0][2][0][RTW89_ACMA][9] = 58,
+	[1][0][2][0][RTW89_CN][9] = 60,
+	[1][0][2][0][RTW89_UK][9] = 58,
+	[1][0][2][0][RTW89_FCC][10] = 68,
+	[1][0][2][0][RTW89_ETSI][10] = 58,
+	[1][0][2][0][RTW89_MKK][10] = 76,
+	[1][0][2][0][RTW89_IC][10] = 68,
+	[1][0][2][0][RTW89_KCC][10] = 76,
+	[1][0][2][0][RTW89_ACMA][10] = 58,
+	[1][0][2][0][RTW89_CN][10] = 60,
+	[1][0][2][0][RTW89_UK][10] = 58,
+	[1][0][2][0][RTW89_FCC][11] = 127,
+	[1][0][2][0][RTW89_ETSI][11] = 127,
+	[1][0][2][0][RTW89_MKK][11] = 127,
+	[1][0][2][0][RTW89_IC][11] = 127,
+	[1][0][2][0][RTW89_KCC][11] = 127,
+	[1][0][2][0][RTW89_ACMA][11] = 127,
+	[1][0][2][0][RTW89_CN][11] = 127,
+	[1][0][2][0][RTW89_UK][11] = 127,
+	[1][0][2][0][RTW89_FCC][12] = 127,
+	[1][0][2][0][RTW89_ETSI][12] = 127,
+	[1][0][2][0][RTW89_MKK][12] = 127,
+	[1][0][2][0][RTW89_IC][12] = 127,
+	[1][0][2][0][RTW89_KCC][12] = 127,
+	[1][0][2][0][RTW89_ACMA][12] = 127,
+	[1][0][2][0][RTW89_CN][12] = 127,
+	[1][0][2][0][RTW89_UK][12] = 127,
+	[1][0][2][0][RTW89_FCC][13] = 127,
+	[1][0][2][0][RTW89_ETSI][13] = 127,
+	[1][0][2][0][RTW89_MKK][13] = 127,
+	[1][0][2][0][RTW89_IC][13] = 127,
+	[1][0][2][0][RTW89_KCC][13] = 127,
+	[1][0][2][0][RTW89_ACMA][13] = 127,
+	[1][0][2][0][RTW89_CN][13] = 127,
+	[1][0][2][0][RTW89_UK][13] = 127,
+	[1][1][2][0][RTW89_FCC][0] = 127,
+	[1][1][2][0][RTW89_ETSI][0] = 127,
+	[1][1][2][0][RTW89_MKK][0] = 127,
+	[1][1][2][0][RTW89_IC][0] = 127,
+	[1][1][2][0][RTW89_KCC][0] = 127,
+	[1][1][2][0][RTW89_ACMA][0] = 127,
+	[1][1][2][0][RTW89_CN][0] = 127,
+	[1][1][2][0][RTW89_UK][0] = 127,
+	[1][1][2][0][RTW89_FCC][1] = 127,
+	[1][1][2][0][RTW89_ETSI][1] = 127,
+	[1][1][2][0][RTW89_MKK][1] = 127,
+	[1][1][2][0][RTW89_IC][1] = 127,
+	[1][1][2][0][RTW89_KCC][1] = 127,
+	[1][1][2][0][RTW89_ACMA][1] = 127,
+	[1][1][2][0][RTW89_CN][1] = 127,
+	[1][1][2][0][RTW89_UK][1] = 127,
+	[1][1][2][0][RTW89_FCC][2] = 127,
+	[1][1][2][0][RTW89_ETSI][2] = 127,
+	[1][1][2][0][RTW89_MKK][2] = 127,
+	[1][1][2][0][RTW89_IC][2] = 127,
+	[1][1][2][0][RTW89_KCC][2] = 127,
+	[1][1][2][0][RTW89_ACMA][2] = 127,
+	[1][1][2][0][RTW89_CN][2] = 127,
+	[1][1][2][0][RTW89_UK][2] = 127,
+	[1][1][2][0][RTW89_FCC][3] = 127,
+	[1][1][2][0][RTW89_ETSI][3] = 127,
+	[1][1][2][0][RTW89_MKK][3] = 127,
+	[1][1][2][0][RTW89_IC][3] = 127,
+	[1][1][2][0][RTW89_KCC][3] = 127,
+	[1][1][2][0][RTW89_ACMA][3] = 127,
+	[1][1][2][0][RTW89_CN][3] = 127,
+	[1][1][2][0][RTW89_UK][3] = 127,
+	[1][1][2][0][RTW89_FCC][4] = 127,
+	[1][1][2][0][RTW89_ETSI][4] = 127,
+	[1][1][2][0][RTW89_MKK][4] = 127,
+	[1][1][2][0][RTW89_IC][4] = 127,
+	[1][1][2][0][RTW89_KCC][4] = 127,
+	[1][1][2][0][RTW89_ACMA][4] = 127,
+	[1][1][2][0][RTW89_CN][4] = 127,
+	[1][1][2][0][RTW89_UK][4] = 127,
+	[1][1][2][0][RTW89_FCC][5] = 127,
+	[1][1][2][0][RTW89_ETSI][5] = 127,
+	[1][1][2][0][RTW89_MKK][5] = 127,
+	[1][1][2][0][RTW89_IC][5] = 127,
+	[1][1][2][0][RTW89_KCC][5] = 127,
+	[1][1][2][0][RTW89_ACMA][5] = 127,
+	[1][1][2][0][RTW89_CN][5] = 127,
+	[1][1][2][0][RTW89_UK][5] = 127,
+	[1][1][2][0][RTW89_FCC][6] = 127,
+	[1][1][2][0][RTW89_ETSI][6] = 127,
+	[1][1][2][0][RTW89_MKK][6] = 127,
+	[1][1][2][0][RTW89_IC][6] = 127,
+	[1][1][2][0][RTW89_KCC][6] = 127,
+	[1][1][2][0][RTW89_ACMA][6] = 127,
+	[1][1][2][0][RTW89_CN][6] = 127,
+	[1][1][2][0][RTW89_UK][6] = 127,
+	[1][1][2][0][RTW89_FCC][7] = 127,
+	[1][1][2][0][RTW89_ETSI][7] = 127,
+	[1][1][2][0][RTW89_MKK][7] = 127,
+	[1][1][2][0][RTW89_IC][7] = 127,
+	[1][1][2][0][RTW89_KCC][7] = 127,
+	[1][1][2][0][RTW89_ACMA][7] = 127,
+	[1][1][2][0][RTW89_CN][7] = 127,
+	[1][1][2][0][RTW89_UK][7] = 127,
+	[1][1][2][0][RTW89_FCC][8] = 127,
+	[1][1][2][0][RTW89_ETSI][8] = 127,
+	[1][1][2][0][RTW89_MKK][8] = 127,
+	[1][1][2][0][RTW89_IC][8] = 127,
+	[1][1][2][0][RTW89_KCC][8] = 127,
+	[1][1][2][0][RTW89_ACMA][8] = 127,
+	[1][1][2][0][RTW89_CN][8] = 127,
+	[1][1][2][0][RTW89_UK][8] = 127,
+	[1][1][2][0][RTW89_FCC][9] = 127,
+	[1][1][2][0][RTW89_ETSI][9] = 127,
+	[1][1][2][0][RTW89_MKK][9] = 127,
+	[1][1][2][0][RTW89_IC][9] = 127,
+	[1][1][2][0][RTW89_KCC][9] = 127,
+	[1][1][2][0][RTW89_ACMA][9] = 127,
+	[1][1][2][0][RTW89_CN][9] = 127,
+	[1][1][2][0][RTW89_UK][9] = 127,
+	[1][1][2][0][RTW89_FCC][10] = 127,
+	[1][1][2][0][RTW89_ETSI][10] = 127,
+	[1][1][2][0][RTW89_MKK][10] = 127,
+	[1][1][2][0][RTW89_IC][10] = 127,
+	[1][1][2][0][RTW89_KCC][10] = 127,
+	[1][1][2][0][RTW89_ACMA][10] = 127,
+	[1][1][2][0][RTW89_CN][10] = 127,
+	[1][1][2][0][RTW89_UK][10] = 127,
+	[1][1][2][0][RTW89_FCC][11] = 127,
+	[1][1][2][0][RTW89_ETSI][11] = 127,
+	[1][1][2][0][RTW89_MKK][11] = 127,
+	[1][1][2][0][RTW89_IC][11] = 127,
+	[1][1][2][0][RTW89_KCC][11] = 127,
+	[1][1][2][0][RTW89_ACMA][11] = 127,
+	[1][1][2][0][RTW89_CN][11] = 127,
+	[1][1][2][0][RTW89_UK][11] = 127,
+	[1][1][2][0][RTW89_FCC][12] = 127,
+	[1][1][2][0][RTW89_ETSI][12] = 127,
+	[1][1][2][0][RTW89_MKK][12] = 127,
+	[1][1][2][0][RTW89_IC][12] = 127,
+	[1][1][2][0][RTW89_KCC][12] = 127,
+	[1][1][2][0][RTW89_ACMA][12] = 127,
+	[1][1][2][0][RTW89_CN][12] = 127,
+	[1][1][2][0][RTW89_UK][12] = 127,
+	[1][1][2][0][RTW89_FCC][13] = 127,
+	[1][1][2][0][RTW89_ETSI][13] = 127,
+	[1][1][2][0][RTW89_MKK][13] = 127,
+	[1][1][2][0][RTW89_IC][13] = 127,
+	[1][1][2][0][RTW89_KCC][13] = 127,
+	[1][1][2][0][RTW89_ACMA][13] = 127,
+	[1][1][2][0][RTW89_CN][13] = 127,
+	[1][1][2][0][RTW89_UK][13] = 127,
+	[1][1][2][1][RTW89_FCC][0] = 127,
+	[1][1][2][1][RTW89_ETSI][0] = 127,
+	[1][1][2][1][RTW89_MKK][0] = 127,
+	[1][1][2][1][RTW89_IC][0] = 127,
+	[1][1][2][1][RTW89_KCC][0] = 127,
+	[1][1][2][1][RTW89_ACMA][0] = 127,
+	[1][1][2][1][RTW89_CN][0] = 127,
+	[1][1][2][1][RTW89_UK][0] = 127,
+	[1][1][2][1][RTW89_FCC][1] = 127,
+	[1][1][2][1][RTW89_ETSI][1] = 127,
+	[1][1][2][1][RTW89_MKK][1] = 127,
+	[1][1][2][1][RTW89_IC][1] = 127,
+	[1][1][2][1][RTW89_KCC][1] = 127,
+	[1][1][2][1][RTW89_ACMA][1] = 127,
+	[1][1][2][1][RTW89_CN][1] = 127,
+	[1][1][2][1][RTW89_UK][1] = 127,
+	[1][1][2][1][RTW89_FCC][2] = 127,
+	[1][1][2][1][RTW89_ETSI][2] = 127,
+	[1][1][2][1][RTW89_MKK][2] = 127,
+	[1][1][2][1][RTW89_IC][2] = 127,
+	[1][1][2][1][RTW89_KCC][2] = 127,
+	[1][1][2][1][RTW89_ACMA][2] = 127,
+	[1][1][2][1][RTW89_CN][2] = 127,
+	[1][1][2][1][RTW89_UK][2] = 127,
+	[1][1][2][1][RTW89_FCC][3] = 127,
+	[1][1][2][1][RTW89_ETSI][3] = 127,
+	[1][1][2][1][RTW89_MKK][3] = 127,
+	[1][1][2][1][RTW89_IC][3] = 127,
+	[1][1][2][1][RTW89_KCC][3] = 127,
+	[1][1][2][1][RTW89_ACMA][3] = 127,
+	[1][1][2][1][RTW89_CN][3] = 127,
+	[1][1][2][1][RTW89_UK][3] = 127,
+	[1][1][2][1][RTW89_FCC][4] = 127,
+	[1][1][2][1][RTW89_ETSI][4] = 127,
+	[1][1][2][1][RTW89_MKK][4] = 127,
+	[1][1][2][1][RTW89_IC][4] = 127,
+	[1][1][2][1][RTW89_KCC][4] = 127,
+	[1][1][2][1][RTW89_ACMA][4] = 127,
+	[1][1][2][1][RTW89_CN][4] = 127,
+	[1][1][2][1][RTW89_UK][4] = 127,
+	[1][1][2][1][RTW89_FCC][5] = 127,
+	[1][1][2][1][RTW89_ETSI][5] = 127,
+	[1][1][2][1][RTW89_MKK][5] = 127,
+	[1][1][2][1][RTW89_IC][5] = 127,
+	[1][1][2][1][RTW89_KCC][5] = 127,
+	[1][1][2][1][RTW89_ACMA][5] = 127,
+	[1][1][2][1][RTW89_CN][5] = 127,
+	[1][1][2][1][RTW89_UK][5] = 127,
+	[1][1][2][1][RTW89_FCC][6] = 127,
+	[1][1][2][1][RTW89_ETSI][6] = 127,
+	[1][1][2][1][RTW89_MKK][6] = 127,
+	[1][1][2][1][RTW89_IC][6] = 127,
+	[1][1][2][1][RTW89_KCC][6] = 127,
+	[1][1][2][1][RTW89_ACMA][6] = 127,
+	[1][1][2][1][RTW89_CN][6] = 127,
+	[1][1][2][1][RTW89_UK][6] = 127,
+	[1][1][2][1][RTW89_FCC][7] = 127,
+	[1][1][2][1][RTW89_ETSI][7] = 127,
+	[1][1][2][1][RTW89_MKK][7] = 127,
+	[1][1][2][1][RTW89_IC][7] = 127,
+	[1][1][2][1][RTW89_KCC][7] = 127,
+	[1][1][2][1][RTW89_ACMA][7] = 127,
+	[1][1][2][1][RTW89_CN][7] = 127,
+	[1][1][2][1][RTW89_UK][7] = 127,
+	[1][1][2][1][RTW89_FCC][8] = 127,
+	[1][1][2][1][RTW89_ETSI][8] = 127,
+	[1][1][2][1][RTW89_MKK][8] = 127,
+	[1][1][2][1][RTW89_IC][8] = 127,
+	[1][1][2][1][RTW89_KCC][8] = 127,
+	[1][1][2][1][RTW89_ACMA][8] = 127,
+	[1][1][2][1][RTW89_CN][8] = 127,
+	[1][1][2][1][RTW89_UK][8] = 127,
+	[1][1][2][1][RTW89_FCC][9] = 127,
+	[1][1][2][1][RTW89_ETSI][9] = 127,
+	[1][1][2][1][RTW89_MKK][9] = 127,
+	[1][1][2][1][RTW89_IC][9] = 127,
+	[1][1][2][1][RTW89_KCC][9] = 127,
+	[1][1][2][1][RTW89_ACMA][9] = 127,
+	[1][1][2][1][RTW89_CN][9] = 127,
+	[1][1][2][1][RTW89_UK][9] = 127,
+	[1][1][2][1][RTW89_FCC][10] = 127,
+	[1][1][2][1][RTW89_ETSI][10] = 127,
+	[1][1][2][1][RTW89_MKK][10] = 127,
+	[1][1][2][1][RTW89_IC][10] = 127,
+	[1][1][2][1][RTW89_KCC][10] = 127,
+	[1][1][2][1][RTW89_ACMA][10] = 127,
+	[1][1][2][1][RTW89_CN][10] = 127,
+	[1][1][2][1][RTW89_UK][10] = 127,
+	[1][1][2][1][RTW89_FCC][11] = 127,
+	[1][1][2][1][RTW89_ETSI][11] = 127,
+	[1][1][2][1][RTW89_MKK][11] = 127,
+	[1][1][2][1][RTW89_IC][11] = 127,
+	[1][1][2][1][RTW89_KCC][11] = 127,
+	[1][1][2][1][RTW89_ACMA][11] = 127,
+	[1][1][2][1][RTW89_CN][11] = 127,
+	[1][1][2][1][RTW89_UK][11] = 127,
+	[1][1][2][1][RTW89_FCC][12] = 127,
+	[1][1][2][1][RTW89_ETSI][12] = 127,
+	[1][1][2][1][RTW89_MKK][12] = 127,
+	[1][1][2][1][RTW89_IC][12] = 127,
+	[1][1][2][1][RTW89_KCC][12] = 127,
+	[1][1][2][1][RTW89_ACMA][12] = 127,
+	[1][1][2][1][RTW89_CN][12] = 127,
+	[1][1][2][1][RTW89_UK][12] = 127,
+	[1][1][2][1][RTW89_FCC][13] = 127,
+	[1][1][2][1][RTW89_ETSI][13] = 127,
+	[1][1][2][1][RTW89_MKK][13] = 127,
+	[1][1][2][1][RTW89_IC][13] = 127,
+	[1][1][2][1][RTW89_KCC][13] = 127,
+	[1][1][2][1][RTW89_ACMA][13] = 127,
+	[1][1][2][1][RTW89_CN][13] = 127,
+	[1][1][2][1][RTW89_UK][13] = 127,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_5g_type2[RTW89_5G_BW_NUM][RTW89_NTX_NUM]
+				       [RTW89_RS_LMT_NUM][RTW89_BF_NUM]
+				       [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
+	[0][0][1][0][RTW89_WW][0] = 58,
+	[0][0][1][0][RTW89_WW][2] = 58,
+	[0][0][1][0][RTW89_WW][4] = 58,
+	[0][0][1][0][RTW89_WW][6] = 50,
+	[0][0][1][0][RTW89_WW][8] = 58,
+	[0][0][1][0][RTW89_WW][10] = 58,
+	[0][0][1][0][RTW89_WW][12] = 58,
+	[0][0][1][0][RTW89_WW][14] = 58,
+	[0][0][1][0][RTW89_WW][15] = 58,
+	[0][0][1][0][RTW89_WW][17] = 60,
+	[0][0][1][0][RTW89_WW][19] = 60,
+	[0][0][1][0][RTW89_WW][21] = 60,
+	[0][0][1][0][RTW89_WW][23] = 60,
+	[0][0][1][0][RTW89_WW][25] = 60,
+	[0][0][1][0][RTW89_WW][27] = 60,
+	[0][0][1][0][RTW89_WW][29] = 60,
+	[0][0][1][0][RTW89_WW][31] = 60,
+	[0][0][1][0][RTW89_WW][33] = 60,
+	[0][0][1][0][RTW89_WW][35] = 60,
+	[0][0][1][0][RTW89_WW][37] = 74,
+	[0][0][1][0][RTW89_WW][38] = 30,
+	[0][0][1][0][RTW89_WW][40] = 30,
+	[0][0][1][0][RTW89_WW][42] = 30,
+	[0][0][1][0][RTW89_WW][44] = 30,
+	[0][0][1][0][RTW89_WW][46] = 30,
+	[0][0][1][0][RTW89_WW][48] = 72,
+	[0][0][1][0][RTW89_WW][50] = 72,
+	[0][0][1][0][RTW89_WW][52] = 72,
+	[0][1][1][0][RTW89_WW][0] = 0,
+	[0][1][1][0][RTW89_WW][2] = 0,
+	[0][1][1][0][RTW89_WW][4] = 0,
+	[0][1][1][0][RTW89_WW][6] = 0,
+	[0][1][1][0][RTW89_WW][8] = 0,
+	[0][1][1][0][RTW89_WW][10] = 0,
+	[0][1][1][0][RTW89_WW][12] = 0,
+	[0][1][1][0][RTW89_WW][14] = 0,
+	[0][1][1][0][RTW89_WW][15] = 0,
+	[0][1][1][0][RTW89_WW][17] = 0,
+	[0][1][1][0][RTW89_WW][19] = 0,
+	[0][1][1][0][RTW89_WW][21] = 0,
+	[0][1][1][0][RTW89_WW][23] = 0,
+	[0][1][1][0][RTW89_WW][25] = 0,
+	[0][1][1][0][RTW89_WW][27] = 0,
+	[0][1][1][0][RTW89_WW][29] = 0,
+	[0][1][1][0][RTW89_WW][31] = 0,
+	[0][1][1][0][RTW89_WW][33] = 0,
+	[0][1][1][0][RTW89_WW][35] = 0,
+	[0][1][1][0][RTW89_WW][37] = 0,
+	[0][1][1][0][RTW89_WW][38] = 0,
+	[0][1][1][0][RTW89_WW][40] = 0,
+	[0][1][1][0][RTW89_WW][42] = 0,
+	[0][1][1][0][RTW89_WW][44] = 0,
+	[0][1][1][0][RTW89_WW][46] = 0,
+	[0][1][1][0][RTW89_WW][48] = 0,
+	[0][1][1][0][RTW89_WW][50] = 0,
+	[0][1][1][0][RTW89_WW][52] = 0,
+	[0][0][2][0][RTW89_WW][0] = 62,
+	[0][0][2][0][RTW89_WW][2] = 62,
+	[0][0][2][0][RTW89_WW][4] = 62,
+	[0][0][2][0][RTW89_WW][6] = 54,
+	[0][0][2][0][RTW89_WW][8] = 62,
+	[0][0][2][0][RTW89_WW][10] = 62,
+	[0][0][2][0][RTW89_WW][12] = 62,
+	[0][0][2][0][RTW89_WW][14] = 62,
+	[0][0][2][0][RTW89_WW][15] = 60,
+	[0][0][2][0][RTW89_WW][17] = 62,
+	[0][0][2][0][RTW89_WW][19] = 62,
+	[0][0][2][0][RTW89_WW][21] = 62,
+	[0][0][2][0][RTW89_WW][23] = 62,
+	[0][0][2][0][RTW89_WW][25] = 62,
+	[0][0][2][0][RTW89_WW][27] = 62,
+	[0][0][2][0][RTW89_WW][29] = 62,
+	[0][0][2][0][RTW89_WW][31] = 62,
+	[0][0][2][0][RTW89_WW][33] = 62,
+	[0][0][2][0][RTW89_WW][35] = 62,
+	[0][0][2][0][RTW89_WW][37] = 74,
+	[0][0][2][0][RTW89_WW][38] = 30,
+	[0][0][2][0][RTW89_WW][40] = 30,
+	[0][0][2][0][RTW89_WW][42] = 30,
+	[0][0][2][0][RTW89_WW][44] = 30,
+	[0][0][2][0][RTW89_WW][46] = 30,
+	[0][0][2][0][RTW89_WW][48] = 74,
+	[0][0][2][0][RTW89_WW][50] = 74,
+	[0][0][2][0][RTW89_WW][52] = 74,
+	[0][1][2][0][RTW89_WW][0] = 0,
+	[0][1][2][0][RTW89_WW][2] = 0,
+	[0][1][2][0][RTW89_WW][4] = 0,
+	[0][1][2][0][RTW89_WW][6] = 0,
+	[0][1][2][0][RTW89_WW][8] = 0,
+	[0][1][2][0][RTW89_WW][10] = 0,
+	[0][1][2][0][RTW89_WW][12] = 0,
+	[0][1][2][0][RTW89_WW][14] = 0,
+	[0][1][2][0][RTW89_WW][15] = 0,
+	[0][1][2][0][RTW89_WW][17] = 0,
+	[0][1][2][0][RTW89_WW][19] = 0,
+	[0][1][2][0][RTW89_WW][21] = 0,
+	[0][1][2][0][RTW89_WW][23] = 0,
+	[0][1][2][0][RTW89_WW][25] = 0,
+	[0][1][2][0][RTW89_WW][27] = 0,
+	[0][1][2][0][RTW89_WW][29] = 0,
+	[0][1][2][0][RTW89_WW][31] = 0,
+	[0][1][2][0][RTW89_WW][33] = 0,
+	[0][1][2][0][RTW89_WW][35] = 0,
+	[0][1][2][0][RTW89_WW][37] = 0,
+	[0][1][2][0][RTW89_WW][38] = 0,
+	[0][1][2][0][RTW89_WW][40] = 0,
+	[0][1][2][0][RTW89_WW][42] = 0,
+	[0][1][2][0][RTW89_WW][44] = 0,
+	[0][1][2][0][RTW89_WW][46] = 0,
+	[0][1][2][0][RTW89_WW][48] = 0,
+	[0][1][2][0][RTW89_WW][50] = 0,
+	[0][1][2][0][RTW89_WW][52] = 0,
+	[0][1][2][1][RTW89_WW][0] = 0,
+	[0][1][2][1][RTW89_WW][2] = 0,
+	[0][1][2][1][RTW89_WW][4] = 0,
+	[0][1][2][1][RTW89_WW][6] = 0,
+	[0][1][2][1][RTW89_WW][8] = 0,
+	[0][1][2][1][RTW89_WW][10] = 0,
+	[0][1][2][1][RTW89_WW][12] = 0,
+	[0][1][2][1][RTW89_WW][14] = 0,
+	[0][1][2][1][RTW89_WW][15] = 0,
+	[0][1][2][1][RTW89_WW][17] = 0,
+	[0][1][2][1][RTW89_WW][19] = 0,
+	[0][1][2][1][RTW89_WW][21] = 0,
+	[0][1][2][1][RTW89_WW][23] = 0,
+	[0][1][2][1][RTW89_WW][25] = 0,
+	[0][1][2][1][RTW89_WW][27] = 0,
+	[0][1][2][1][RTW89_WW][29] = 0,
+	[0][1][2][1][RTW89_WW][31] = 0,
+	[0][1][2][1][RTW89_WW][33] = 0,
+	[0][1][2][1][RTW89_WW][35] = 0,
+	[0][1][2][1][RTW89_WW][37] = 0,
+	[0][1][2][1][RTW89_WW][38] = 0,
+	[0][1][2][1][RTW89_WW][40] = 0,
+	[0][1][2][1][RTW89_WW][42] = 0,
+	[0][1][2][1][RTW89_WW][44] = 0,
+	[0][1][2][1][RTW89_WW][46] = 0,
+	[0][1][2][1][RTW89_WW][48] = 0,
+	[0][1][2][1][RTW89_WW][50] = 0,
+	[0][1][2][1][RTW89_WW][52] = 0,
+	[1][0][2][0][RTW89_WW][1] = 64,
+	[1][0][2][0][RTW89_WW][5] = 62,
+	[1][0][2][0][RTW89_WW][9] = 64,
+	[1][0][2][0][RTW89_WW][13] = 64,
+	[1][0][2][0][RTW89_WW][16] = 66,
+	[1][0][2][0][RTW89_WW][20] = 66,
+	[1][0][2][0][RTW89_WW][24] = 66,
+	[1][0][2][0][RTW89_WW][28] = 66,
+	[1][0][2][0][RTW89_WW][32] = 66,
+	[1][0][2][0][RTW89_WW][36] = 76,
+	[1][0][2][0][RTW89_WW][39] = 30,
+	[1][0][2][0][RTW89_WW][43] = 30,
+	[1][0][2][0][RTW89_WW][47] = 80,
+	[1][0][2][0][RTW89_WW][51] = 80,
+	[1][1][2][0][RTW89_WW][1] = 0,
+	[1][1][2][0][RTW89_WW][5] = 0,
+	[1][1][2][0][RTW89_WW][9] = 0,
+	[1][1][2][0][RTW89_WW][13] = 0,
+	[1][1][2][0][RTW89_WW][16] = 0,
+	[1][1][2][0][RTW89_WW][20] = 0,
+	[1][1][2][0][RTW89_WW][24] = 0,
+	[1][1][2][0][RTW89_WW][28] = 0,
+	[1][1][2][0][RTW89_WW][32] = 0,
+	[1][1][2][0][RTW89_WW][36] = 0,
+	[1][1][2][0][RTW89_WW][39] = 0,
+	[1][1][2][0][RTW89_WW][43] = 0,
+	[1][1][2][0][RTW89_WW][47] = 0,
+	[1][1][2][0][RTW89_WW][51] = 0,
+	[1][1][2][1][RTW89_WW][1] = 0,
+	[1][1][2][1][RTW89_WW][5] = 0,
+	[1][1][2][1][RTW89_WW][9] = 0,
+	[1][1][2][1][RTW89_WW][13] = 0,
+	[1][1][2][1][RTW89_WW][16] = 0,
+	[1][1][2][1][RTW89_WW][20] = 0,
+	[1][1][2][1][RTW89_WW][24] = 0,
+	[1][1][2][1][RTW89_WW][28] = 0,
+	[1][1][2][1][RTW89_WW][32] = 0,
+	[1][1][2][1][RTW89_WW][36] = 0,
+	[1][1][2][1][RTW89_WW][39] = 0,
+	[1][1][2][1][RTW89_WW][43] = 0,
+	[1][1][2][1][RTW89_WW][47] = 0,
+	[1][1][2][1][RTW89_WW][51] = 0,
+	[2][0][2][0][RTW89_WW][3] = 62,
+	[2][0][2][0][RTW89_WW][11] = 62,
+	[2][0][2][0][RTW89_WW][18] = 64,
+	[2][0][2][0][RTW89_WW][26] = 64,
+	[2][0][2][0][RTW89_WW][34] = 68,
+	[2][0][2][0][RTW89_WW][41] = 30,
+	[2][0][2][0][RTW89_WW][49] = 72,
+	[2][1][2][0][RTW89_WW][3] = 0,
+	[2][1][2][0][RTW89_WW][11] = 0,
+	[2][1][2][0][RTW89_WW][18] = 0,
+	[2][1][2][0][RTW89_WW][26] = 0,
+	[2][1][2][0][RTW89_WW][34] = 0,
+	[2][1][2][0][RTW89_WW][41] = 0,
+	[2][1][2][0][RTW89_WW][49] = 0,
+	[2][1][2][1][RTW89_WW][3] = 0,
+	[2][1][2][1][RTW89_WW][11] = 0,
+	[2][1][2][1][RTW89_WW][18] = 0,
+	[2][1][2][1][RTW89_WW][26] = 0,
+	[2][1][2][1][RTW89_WW][34] = 0,
+	[2][1][2][1][RTW89_WW][41] = 0,
+	[2][1][2][1][RTW89_WW][49] = 0,
+	[3][0][2][0][RTW89_WW][7] = 58,
+	[3][0][2][0][RTW89_WW][22] = 58,
+	[3][0][2][0][RTW89_WW][45] = 0,
+	[3][1][2][0][RTW89_WW][7] = 0,
+	[3][1][2][0][RTW89_WW][22] = 0,
+	[3][1][2][0][RTW89_WW][45] = 0,
+	[3][1][2][1][RTW89_WW][7] = 0,
+	[3][1][2][1][RTW89_WW][22] = 0,
+	[3][1][2][1][RTW89_WW][45] = 0,
+	[0][0][1][0][RTW89_FCC][0] = 78,
+	[0][0][1][0][RTW89_ETSI][0] = 58,
+	[0][0][1][0][RTW89_MKK][0] = 60,
+	[0][0][1][0][RTW89_IC][0] = 62,
+	[0][0][1][0][RTW89_KCC][0] = 74,
+	[0][0][1][0][RTW89_ACMA][0] = 58,
+	[0][0][1][0][RTW89_CN][0] = 60,
+	[0][0][1][0][RTW89_UK][0] = 58,
+	[0][0][1][0][RTW89_FCC][2] = 78,
+	[0][0][1][0][RTW89_ETSI][2] = 58,
+	[0][0][1][0][RTW89_MKK][2] = 60,
+	[0][0][1][0][RTW89_IC][2] = 62,
+	[0][0][1][0][RTW89_KCC][2] = 74,
+	[0][0][1][0][RTW89_ACMA][2] = 58,
+	[0][0][1][0][RTW89_CN][2] = 60,
+	[0][0][1][0][RTW89_UK][2] = 58,
+	[0][0][1][0][RTW89_FCC][4] = 78,
+	[0][0][1][0][RTW89_ETSI][4] = 58,
+	[0][0][1][0][RTW89_MKK][4] = 60,
+	[0][0][1][0][RTW89_IC][4] = 62,
+	[0][0][1][0][RTW89_KCC][4] = 74,
+	[0][0][1][0][RTW89_ACMA][4] = 58,
+	[0][0][1][0][RTW89_CN][4] = 60,
+	[0][0][1][0][RTW89_UK][4] = 58,
+	[0][0][1][0][RTW89_FCC][6] = 78,
+	[0][0][1][0][RTW89_ETSI][6] = 58,
+	[0][0][1][0][RTW89_MKK][6] = 60,
+	[0][0][1][0][RTW89_IC][6] = 62,
+	[0][0][1][0][RTW89_KCC][6] = 50,
+	[0][0][1][0][RTW89_ACMA][6] = 58,
+	[0][0][1][0][RTW89_CN][6] = 60,
+	[0][0][1][0][RTW89_UK][6] = 58,
+	[0][0][1][0][RTW89_FCC][8] = 78,
+	[0][0][1][0][RTW89_ETSI][8] = 58,
+	[0][0][1][0][RTW89_MKK][8] = 60,
+	[0][0][1][0][RTW89_IC][8] = 62,
+	[0][0][1][0][RTW89_KCC][8] = 74,
+	[0][0][1][0][RTW89_ACMA][8] = 58,
+	[0][0][1][0][RTW89_CN][8] = 60,
+	[0][0][1][0][RTW89_UK][8] = 58,
+	[0][0][1][0][RTW89_FCC][10] = 78,
+	[0][0][1][0][RTW89_ETSI][10] = 58,
+	[0][0][1][0][RTW89_MKK][10] = 60,
+	[0][0][1][0][RTW89_IC][10] = 64,
+	[0][0][1][0][RTW89_KCC][10] = 74,
+	[0][0][1][0][RTW89_ACMA][10] = 58,
+	[0][0][1][0][RTW89_CN][10] = 60,
+	[0][0][1][0][RTW89_UK][10] = 58,
+	[0][0][1][0][RTW89_FCC][12] = 78,
+	[0][0][1][0][RTW89_ETSI][12] = 58,
+	[0][0][1][0][RTW89_MKK][12] = 60,
+	[0][0][1][0][RTW89_IC][12] = 64,
+	[0][0][1][0][RTW89_KCC][12] = 74,
+	[0][0][1][0][RTW89_ACMA][12] = 58,
+	[0][0][1][0][RTW89_CN][12] = 60,
+	[0][0][1][0][RTW89_UK][12] = 58,
+	[0][0][1][0][RTW89_FCC][14] = 76,
+	[0][0][1][0][RTW89_ETSI][14] = 58,
+	[0][0][1][0][RTW89_MKK][14] = 60,
+	[0][0][1][0][RTW89_IC][14] = 62,
+	[0][0][1][0][RTW89_KCC][14] = 74,
+	[0][0][1][0][RTW89_ACMA][14] = 58,
+	[0][0][1][0][RTW89_CN][14] = 60,
+	[0][0][1][0][RTW89_UK][14] = 58,
+	[0][0][1][0][RTW89_FCC][15] = 76,
+	[0][0][1][0][RTW89_ETSI][15] = 58,
+	[0][0][1][0][RTW89_MKK][15] = 74,
+	[0][0][1][0][RTW89_IC][15] = 76,
+	[0][0][1][0][RTW89_KCC][15] = 74,
+	[0][0][1][0][RTW89_ACMA][15] = 58,
+	[0][0][1][0][RTW89_CN][15] = 127,
+	[0][0][1][0][RTW89_UK][15] = 58,
+	[0][0][1][0][RTW89_FCC][17] = 78,
+	[0][0][1][0][RTW89_ETSI][17] = 60,
+	[0][0][1][0][RTW89_MKK][17] = 74,
+	[0][0][1][0][RTW89_IC][17] = 78,
+	[0][0][1][0][RTW89_KCC][17] = 74,
+	[0][0][1][0][RTW89_ACMA][17] = 60,
+	[0][0][1][0][RTW89_CN][17] = 127,
+	[0][0][1][0][RTW89_UK][17] = 60,
+	[0][0][1][0][RTW89_FCC][19] = 78,
+	[0][0][1][0][RTW89_ETSI][19] = 60,
+	[0][0][1][0][RTW89_MKK][19] = 74,
+	[0][0][1][0][RTW89_IC][19] = 78,
+	[0][0][1][0][RTW89_KCC][19] = 74,
+	[0][0][1][0][RTW89_ACMA][19] = 60,
+	[0][0][1][0][RTW89_CN][19] = 127,
+	[0][0][1][0][RTW89_UK][19] = 60,
+	[0][0][1][0][RTW89_FCC][21] = 78,
+	[0][0][1][0][RTW89_ETSI][21] = 60,
+	[0][0][1][0][RTW89_MKK][21] = 74,
+	[0][0][1][0][RTW89_IC][21] = 78,
+	[0][0][1][0][RTW89_KCC][21] = 74,
+	[0][0][1][0][RTW89_ACMA][21] = 60,
+	[0][0][1][0][RTW89_CN][21] = 127,
+	[0][0][1][0][RTW89_UK][21] = 60,
+	[0][0][1][0][RTW89_FCC][23] = 78,
+	[0][0][1][0][RTW89_ETSI][23] = 60,
+	[0][0][1][0][RTW89_MKK][23] = 74,
+	[0][0][1][0][RTW89_IC][23] = 78,
+	[0][0][1][0][RTW89_KCC][23] = 74,
+	[0][0][1][0][RTW89_ACMA][23] = 60,
+	[0][0][1][0][RTW89_CN][23] = 127,
+	[0][0][1][0][RTW89_UK][23] = 60,
+	[0][0][1][0][RTW89_FCC][25] = 78,
+	[0][0][1][0][RTW89_ETSI][25] = 60,
+	[0][0][1][0][RTW89_MKK][25] = 74,
+	[0][0][1][0][RTW89_IC][25] = 127,
+	[0][0][1][0][RTW89_KCC][25] = 74,
+	[0][0][1][0][RTW89_ACMA][25] = 127,
+	[0][0][1][0][RTW89_CN][25] = 127,
+	[0][0][1][0][RTW89_UK][25] = 60,
+	[0][0][1][0][RTW89_FCC][27] = 78,
+	[0][0][1][0][RTW89_ETSI][27] = 60,
+	[0][0][1][0][RTW89_MKK][27] = 74,
+	[0][0][1][0][RTW89_IC][27] = 127,
+	[0][0][1][0][RTW89_KCC][27] = 74,
+	[0][0][1][0][RTW89_ACMA][27] = 127,
+	[0][0][1][0][RTW89_CN][27] = 127,
+	[0][0][1][0][RTW89_UK][27] = 60,
+	[0][0][1][0][RTW89_FCC][29] = 78,
+	[0][0][1][0][RTW89_ETSI][29] = 60,
+	[0][0][1][0][RTW89_MKK][29] = 74,
+	[0][0][1][0][RTW89_IC][29] = 127,
+	[0][0][1][0][RTW89_KCC][29] = 74,
+	[0][0][1][0][RTW89_ACMA][29] = 127,
+	[0][0][1][0][RTW89_CN][29] = 127,
+	[0][0][1][0][RTW89_UK][29] = 60,
+	[0][0][1][0][RTW89_FCC][31] = 78,
+	[0][0][1][0][RTW89_ETSI][31] = 60,
+	[0][0][1][0][RTW89_MKK][31] = 74,
+	[0][0][1][0][RTW89_IC][31] = 78,
+	[0][0][1][0][RTW89_KCC][31] = 74,
+	[0][0][1][0][RTW89_ACMA][31] = 60,
+	[0][0][1][0][RTW89_CN][31] = 127,
+	[0][0][1][0][RTW89_UK][31] = 60,
+	[0][0][1][0][RTW89_FCC][33] = 78,
+	[0][0][1][0][RTW89_ETSI][33] = 60,
+	[0][0][1][0][RTW89_MKK][33] = 74,
+	[0][0][1][0][RTW89_IC][33] = 78,
+	[0][0][1][0][RTW89_KCC][33] = 74,
+	[0][0][1][0][RTW89_ACMA][33] = 60,
+	[0][0][1][0][RTW89_CN][33] = 127,
+	[0][0][1][0][RTW89_UK][33] = 60,
+	[0][0][1][0][RTW89_FCC][35] = 70,
+	[0][0][1][0][RTW89_ETSI][35] = 60,
+	[0][0][1][0][RTW89_MKK][35] = 74,
+	[0][0][1][0][RTW89_IC][35] = 70,
+	[0][0][1][0][RTW89_KCC][35] = 74,
+	[0][0][1][0][RTW89_ACMA][35] = 60,
+	[0][0][1][0][RTW89_CN][35] = 127,
+	[0][0][1][0][RTW89_UK][35] = 60,
+	[0][0][1][0][RTW89_FCC][37] = 78,
+	[0][0][1][0][RTW89_ETSI][37] = 127,
+	[0][0][1][0][RTW89_MKK][37] = 74,
+	[0][0][1][0][RTW89_IC][37] = 78,
+	[0][0][1][0][RTW89_KCC][37] = 74,
+	[0][0][1][0][RTW89_ACMA][37] = 74,
+	[0][0][1][0][RTW89_CN][37] = 127,
+	[0][0][1][0][RTW89_UK][37] = 74,
+	[0][0][1][0][RTW89_FCC][38] = 78,
+	[0][0][1][0][RTW89_ETSI][38] = 30,
+	[0][0][1][0][RTW89_MKK][38] = 127,
+	[0][0][1][0][RTW89_IC][38] = 78,
+	[0][0][1][0][RTW89_KCC][38] = 70,
+	[0][0][1][0][RTW89_ACMA][38] = 74,
+	[0][0][1][0][RTW89_CN][38] = 74,
+	[0][0][1][0][RTW89_UK][38] = 58,
+	[0][0][1][0][RTW89_FCC][40] = 78,
+	[0][0][1][0][RTW89_ETSI][40] = 30,
+	[0][0][1][0][RTW89_MKK][40] = 127,
+	[0][0][1][0][RTW89_IC][40] = 78,
+	[0][0][1][0][RTW89_KCC][40] = 74,
+	[0][0][1][0][RTW89_ACMA][40] = 74,
+	[0][0][1][0][RTW89_CN][40] = 74,
+	[0][0][1][0][RTW89_UK][40] = 58,
+	[0][0][1][0][RTW89_FCC][42] = 78,
+	[0][0][1][0][RTW89_ETSI][42] = 30,
+	[0][0][1][0][RTW89_MKK][42] = 127,
+	[0][0][1][0][RTW89_IC][42] = 78,
+	[0][0][1][0][RTW89_KCC][42] = 74,
+	[0][0][1][0][RTW89_ACMA][42] = 74,
+	[0][0][1][0][RTW89_CN][42] = 74,
+	[0][0][1][0][RTW89_UK][42] = 58,
+	[0][0][1][0][RTW89_FCC][44] = 78,
+	[0][0][1][0][RTW89_ETSI][44] = 30,
+	[0][0][1][0][RTW89_MKK][44] = 127,
+	[0][0][1][0][RTW89_IC][44] = 78,
+	[0][0][1][0][RTW89_KCC][44] = 74,
+	[0][0][1][0][RTW89_ACMA][44] = 74,
+	[0][0][1][0][RTW89_CN][44] = 74,
+	[0][0][1][0][RTW89_UK][44] = 58,
+	[0][0][1][0][RTW89_FCC][46] = 78,
+	[0][0][1][0][RTW89_ETSI][46] = 30,
+	[0][0][1][0][RTW89_MKK][46] = 127,
+	[0][0][1][0][RTW89_IC][46] = 78,
+	[0][0][1][0][RTW89_KCC][46] = 74,
+	[0][0][1][0][RTW89_ACMA][46] = 74,
+	[0][0][1][0][RTW89_CN][46] = 74,
+	[0][0][1][0][RTW89_UK][46] = 58,
+	[0][0][1][0][RTW89_FCC][48] = 72,
+	[0][0][1][0][RTW89_ETSI][48] = 127,
+	[0][0][1][0][RTW89_MKK][48] = 127,
+	[0][0][1][0][RTW89_IC][48] = 127,
+	[0][0][1][0][RTW89_KCC][48] = 127,
+	[0][0][1][0][RTW89_ACMA][48] = 127,
+	[0][0][1][0][RTW89_CN][48] = 127,
+	[0][0][1][0][RTW89_UK][48] = 127,
+	[0][0][1][0][RTW89_FCC][50] = 72,
+	[0][0][1][0][RTW89_ETSI][50] = 127,
+	[0][0][1][0][RTW89_MKK][50] = 127,
+	[0][0][1][0][RTW89_IC][50] = 127,
+	[0][0][1][0][RTW89_KCC][50] = 127,
+	[0][0][1][0][RTW89_ACMA][50] = 127,
+	[0][0][1][0][RTW89_CN][50] = 127,
+	[0][0][1][0][RTW89_UK][50] = 127,
+	[0][0][1][0][RTW89_FCC][52] = 72,
+	[0][0][1][0][RTW89_ETSI][52] = 127,
+	[0][0][1][0][RTW89_MKK][52] = 127,
+	[0][0][1][0][RTW89_IC][52] = 127,
+	[0][0][1][0][RTW89_KCC][52] = 127,
+	[0][0][1][0][RTW89_ACMA][52] = 127,
+	[0][0][1][0][RTW89_CN][52] = 127,
+	[0][0][1][0][RTW89_UK][52] = 127,
+	[0][1][1][0][RTW89_FCC][0] = 127,
+	[0][1][1][0][RTW89_ETSI][0] = 127,
+	[0][1][1][0][RTW89_MKK][0] = 127,
+	[0][1][1][0][RTW89_IC][0] = 127,
+	[0][1][1][0][RTW89_KCC][0] = 127,
+	[0][1][1][0][RTW89_ACMA][0] = 127,
+	[0][1][1][0][RTW89_CN][0] = 127,
+	[0][1][1][0][RTW89_UK][0] = 127,
+	[0][1][1][0][RTW89_FCC][2] = 127,
+	[0][1][1][0][RTW89_ETSI][2] = 127,
+	[0][1][1][0][RTW89_MKK][2] = 127,
+	[0][1][1][0][RTW89_IC][2] = 127,
+	[0][1][1][0][RTW89_KCC][2] = 127,
+	[0][1][1][0][RTW89_ACMA][2] = 127,
+	[0][1][1][0][RTW89_CN][2] = 127,
+	[0][1][1][0][RTW89_UK][2] = 127,
+	[0][1][1][0][RTW89_FCC][4] = 127,
+	[0][1][1][0][RTW89_ETSI][4] = 127,
+	[0][1][1][0][RTW89_MKK][4] = 127,
+	[0][1][1][0][RTW89_IC][4] = 127,
+	[0][1][1][0][RTW89_KCC][4] = 127,
+	[0][1][1][0][RTW89_ACMA][4] = 127,
+	[0][1][1][0][RTW89_CN][4] = 127,
+	[0][1][1][0][RTW89_UK][4] = 127,
+	[0][1][1][0][RTW89_FCC][6] = 127,
+	[0][1][1][0][RTW89_ETSI][6] = 127,
+	[0][1][1][0][RTW89_MKK][6] = 127,
+	[0][1][1][0][RTW89_IC][6] = 127,
+	[0][1][1][0][RTW89_KCC][6] = 127,
+	[0][1][1][0][RTW89_ACMA][6] = 127,
+	[0][1][1][0][RTW89_CN][6] = 127,
+	[0][1][1][0][RTW89_UK][6] = 127,
+	[0][1][1][0][RTW89_FCC][8] = 127,
+	[0][1][1][0][RTW89_ETSI][8] = 127,
+	[0][1][1][0][RTW89_MKK][8] = 127,
+	[0][1][1][0][RTW89_IC][8] = 127,
+	[0][1][1][0][RTW89_KCC][8] = 127,
+	[0][1][1][0][RTW89_ACMA][8] = 127,
+	[0][1][1][0][RTW89_CN][8] = 127,
+	[0][1][1][0][RTW89_UK][8] = 127,
+	[0][1][1][0][RTW89_FCC][10] = 127,
+	[0][1][1][0][RTW89_ETSI][10] = 127,
+	[0][1][1][0][RTW89_MKK][10] = 127,
+	[0][1][1][0][RTW89_IC][10] = 127,
+	[0][1][1][0][RTW89_KCC][10] = 127,
+	[0][1][1][0][RTW89_ACMA][10] = 127,
+	[0][1][1][0][RTW89_CN][10] = 127,
+	[0][1][1][0][RTW89_UK][10] = 127,
+	[0][1][1][0][RTW89_FCC][12] = 127,
+	[0][1][1][0][RTW89_ETSI][12] = 127,
+	[0][1][1][0][RTW89_MKK][12] = 127,
+	[0][1][1][0][RTW89_IC][12] = 127,
+	[0][1][1][0][RTW89_KCC][12] = 127,
+	[0][1][1][0][RTW89_ACMA][12] = 127,
+	[0][1][1][0][RTW89_CN][12] = 127,
+	[0][1][1][0][RTW89_UK][12] = 127,
+	[0][1][1][0][RTW89_FCC][14] = 127,
+	[0][1][1][0][RTW89_ETSI][14] = 127,
+	[0][1][1][0][RTW89_MKK][14] = 127,
+	[0][1][1][0][RTW89_IC][14] = 127,
+	[0][1][1][0][RTW89_KCC][14] = 127,
+	[0][1][1][0][RTW89_ACMA][14] = 127,
+	[0][1][1][0][RTW89_CN][14] = 127,
+	[0][1][1][0][RTW89_UK][14] = 127,
+	[0][1][1][0][RTW89_FCC][15] = 127,
+	[0][1][1][0][RTW89_ETSI][15] = 127,
+	[0][1][1][0][RTW89_MKK][15] = 127,
+	[0][1][1][0][RTW89_IC][15] = 127,
+	[0][1][1][0][RTW89_KCC][15] = 127,
+	[0][1][1][0][RTW89_ACMA][15] = 127,
+	[0][1][1][0][RTW89_CN][15] = 127,
+	[0][1][1][0][RTW89_UK][15] = 127,
+	[0][1][1][0][RTW89_FCC][17] = 127,
+	[0][1][1][0][RTW89_ETSI][17] = 127,
+	[0][1][1][0][RTW89_MKK][17] = 127,
+	[0][1][1][0][RTW89_IC][17] = 127,
+	[0][1][1][0][RTW89_KCC][17] = 127,
+	[0][1][1][0][RTW89_ACMA][17] = 127,
+	[0][1][1][0][RTW89_CN][17] = 127,
+	[0][1][1][0][RTW89_UK][17] = 127,
+	[0][1][1][0][RTW89_FCC][19] = 127,
+	[0][1][1][0][RTW89_ETSI][19] = 127,
+	[0][1][1][0][RTW89_MKK][19] = 127,
+	[0][1][1][0][RTW89_IC][19] = 127,
+	[0][1][1][0][RTW89_KCC][19] = 127,
+	[0][1][1][0][RTW89_ACMA][19] = 127,
+	[0][1][1][0][RTW89_CN][19] = 127,
+	[0][1][1][0][RTW89_UK][19] = 127,
+	[0][1][1][0][RTW89_FCC][21] = 127,
+	[0][1][1][0][RTW89_ETSI][21] = 127,
+	[0][1][1][0][RTW89_MKK][21] = 127,
+	[0][1][1][0][RTW89_IC][21] = 127,
+	[0][1][1][0][RTW89_KCC][21] = 127,
+	[0][1][1][0][RTW89_ACMA][21] = 127,
+	[0][1][1][0][RTW89_CN][21] = 127,
+	[0][1][1][0][RTW89_UK][21] = 127,
+	[0][1][1][0][RTW89_FCC][23] = 127,
+	[0][1][1][0][RTW89_ETSI][23] = 127,
+	[0][1][1][0][RTW89_MKK][23] = 127,
+	[0][1][1][0][RTW89_IC][23] = 127,
+	[0][1][1][0][RTW89_KCC][23] = 127,
+	[0][1][1][0][RTW89_ACMA][23] = 127,
+	[0][1][1][0][RTW89_CN][23] = 127,
+	[0][1][1][0][RTW89_UK][23] = 127,
+	[0][1][1][0][RTW89_FCC][25] = 127,
+	[0][1][1][0][RTW89_ETSI][25] = 127,
+	[0][1][1][0][RTW89_MKK][25] = 127,
+	[0][1][1][0][RTW89_IC][25] = 127,
+	[0][1][1][0][RTW89_KCC][25] = 127,
+	[0][1][1][0][RTW89_ACMA][25] = 127,
+	[0][1][1][0][RTW89_CN][25] = 127,
+	[0][1][1][0][RTW89_UK][25] = 127,
+	[0][1][1][0][RTW89_FCC][27] = 127,
+	[0][1][1][0][RTW89_ETSI][27] = 127,
+	[0][1][1][0][RTW89_MKK][27] = 127,
+	[0][1][1][0][RTW89_IC][27] = 127,
+	[0][1][1][0][RTW89_KCC][27] = 127,
+	[0][1][1][0][RTW89_ACMA][27] = 127,
+	[0][1][1][0][RTW89_CN][27] = 127,
+	[0][1][1][0][RTW89_UK][27] = 127,
+	[0][1][1][0][RTW89_FCC][29] = 127,
+	[0][1][1][0][RTW89_ETSI][29] = 127,
+	[0][1][1][0][RTW89_MKK][29] = 127,
+	[0][1][1][0][RTW89_IC][29] = 127,
+	[0][1][1][0][RTW89_KCC][29] = 127,
+	[0][1][1][0][RTW89_ACMA][29] = 127,
+	[0][1][1][0][RTW89_CN][29] = 127,
+	[0][1][1][0][RTW89_UK][29] = 127,
+	[0][1][1][0][RTW89_FCC][31] = 127,
+	[0][1][1][0][RTW89_ETSI][31] = 127,
+	[0][1][1][0][RTW89_MKK][31] = 127,
+	[0][1][1][0][RTW89_IC][31] = 127,
+	[0][1][1][0][RTW89_KCC][31] = 127,
+	[0][1][1][0][RTW89_ACMA][31] = 127,
+	[0][1][1][0][RTW89_CN][31] = 127,
+	[0][1][1][0][RTW89_UK][31] = 127,
+	[0][1][1][0][RTW89_FCC][33] = 127,
+	[0][1][1][0][RTW89_ETSI][33] = 127,
+	[0][1][1][0][RTW89_MKK][33] = 127,
+	[0][1][1][0][RTW89_IC][33] = 127,
+	[0][1][1][0][RTW89_KCC][33] = 127,
+	[0][1][1][0][RTW89_ACMA][33] = 127,
+	[0][1][1][0][RTW89_CN][33] = 127,
+	[0][1][1][0][RTW89_UK][33] = 127,
+	[0][1][1][0][RTW89_FCC][35] = 127,
+	[0][1][1][0][RTW89_ETSI][35] = 127,
+	[0][1][1][0][RTW89_MKK][35] = 127,
+	[0][1][1][0][RTW89_IC][35] = 127,
+	[0][1][1][0][RTW89_KCC][35] = 127,
+	[0][1][1][0][RTW89_ACMA][35] = 127,
+	[0][1][1][0][RTW89_CN][35] = 127,
+	[0][1][1][0][RTW89_UK][35] = 127,
+	[0][1][1][0][RTW89_FCC][37] = 127,
+	[0][1][1][0][RTW89_ETSI][37] = 127,
+	[0][1][1][0][RTW89_MKK][37] = 127,
+	[0][1][1][0][RTW89_IC][37] = 127,
+	[0][1][1][0][RTW89_KCC][37] = 127,
+	[0][1][1][0][RTW89_ACMA][37] = 127,
+	[0][1][1][0][RTW89_CN][37] = 127,
+	[0][1][1][0][RTW89_UK][37] = 127,
+	[0][1][1][0][RTW89_FCC][38] = 127,
+	[0][1][1][0][RTW89_ETSI][38] = 127,
+	[0][1][1][0][RTW89_MKK][38] = 127,
+	[0][1][1][0][RTW89_IC][38] = 127,
+	[0][1][1][0][RTW89_KCC][38] = 127,
+	[0][1][1][0][RTW89_ACMA][38] = 127,
+	[0][1][1][0][RTW89_CN][38] = 127,
+	[0][1][1][0][RTW89_UK][38] = 127,
+	[0][1][1][0][RTW89_FCC][40] = 127,
+	[0][1][1][0][RTW89_ETSI][40] = 127,
+	[0][1][1][0][RTW89_MKK][40] = 127,
+	[0][1][1][0][RTW89_IC][40] = 127,
+	[0][1][1][0][RTW89_KCC][40] = 127,
+	[0][1][1][0][RTW89_ACMA][40] = 127,
+	[0][1][1][0][RTW89_CN][40] = 127,
+	[0][1][1][0][RTW89_UK][40] = 127,
+	[0][1][1][0][RTW89_FCC][42] = 127,
+	[0][1][1][0][RTW89_ETSI][42] = 127,
+	[0][1][1][0][RTW89_MKK][42] = 127,
+	[0][1][1][0][RTW89_IC][42] = 127,
+	[0][1][1][0][RTW89_KCC][42] = 127,
+	[0][1][1][0][RTW89_ACMA][42] = 127,
+	[0][1][1][0][RTW89_CN][42] = 127,
+	[0][1][1][0][RTW89_UK][42] = 127,
+	[0][1][1][0][RTW89_FCC][44] = 127,
+	[0][1][1][0][RTW89_ETSI][44] = 127,
+	[0][1][1][0][RTW89_MKK][44] = 127,
+	[0][1][1][0][RTW89_IC][44] = 127,
+	[0][1][1][0][RTW89_KCC][44] = 127,
+	[0][1][1][0][RTW89_ACMA][44] = 127,
+	[0][1][1][0][RTW89_CN][44] = 127,
+	[0][1][1][0][RTW89_UK][44] = 127,
+	[0][1][1][0][RTW89_FCC][46] = 127,
+	[0][1][1][0][RTW89_ETSI][46] = 127,
+	[0][1][1][0][RTW89_MKK][46] = 127,
+	[0][1][1][0][RTW89_IC][46] = 127,
+	[0][1][1][0][RTW89_KCC][46] = 127,
+	[0][1][1][0][RTW89_ACMA][46] = 127,
+	[0][1][1][0][RTW89_CN][46] = 127,
+	[0][1][1][0][RTW89_UK][46] = 127,
+	[0][1][1][0][RTW89_FCC][48] = 127,
+	[0][1][1][0][RTW89_ETSI][48] = 127,
+	[0][1][1][0][RTW89_MKK][48] = 127,
+	[0][1][1][0][RTW89_IC][48] = 127,
+	[0][1][1][0][RTW89_KCC][48] = 127,
+	[0][1][1][0][RTW89_ACMA][48] = 127,
+	[0][1][1][0][RTW89_CN][48] = 127,
+	[0][1][1][0][RTW89_UK][48] = 127,
+	[0][1][1][0][RTW89_FCC][50] = 127,
+	[0][1][1][0][RTW89_ETSI][50] = 127,
+	[0][1][1][0][RTW89_MKK][50] = 127,
+	[0][1][1][0][RTW89_IC][50] = 127,
+	[0][1][1][0][RTW89_KCC][50] = 127,
+	[0][1][1][0][RTW89_ACMA][50] = 127,
+	[0][1][1][0][RTW89_CN][50] = 127,
+	[0][1][1][0][RTW89_UK][50] = 127,
+	[0][1][1][0][RTW89_FCC][52] = 127,
+	[0][1][1][0][RTW89_ETSI][52] = 127,
+	[0][1][1][0][RTW89_MKK][52] = 127,
+	[0][1][1][0][RTW89_IC][52] = 127,
+	[0][1][1][0][RTW89_KCC][52] = 127,
+	[0][1][1][0][RTW89_ACMA][52] = 127,
+	[0][1][1][0][RTW89_CN][52] = 127,
+	[0][1][1][0][RTW89_UK][52] = 127,
+	[0][0][2][0][RTW89_FCC][0] = 76,
+	[0][0][2][0][RTW89_ETSI][0] = 62,
+	[0][0][2][0][RTW89_MKK][0] = 62,
+	[0][0][2][0][RTW89_IC][0] = 64,
+	[0][0][2][0][RTW89_KCC][0] = 74,
+	[0][0][2][0][RTW89_ACMA][0] = 62,
+	[0][0][2][0][RTW89_CN][0] = 62,
+	[0][0][2][0][RTW89_UK][0] = 62,
+	[0][0][2][0][RTW89_FCC][2] = 78,
+	[0][0][2][0][RTW89_ETSI][2] = 62,
+	[0][0][2][0][RTW89_MKK][2] = 62,
+	[0][0][2][0][RTW89_IC][2] = 64,
+	[0][0][2][0][RTW89_KCC][2] = 74,
+	[0][0][2][0][RTW89_ACMA][2] = 62,
+	[0][0][2][0][RTW89_CN][2] = 62,
+	[0][0][2][0][RTW89_UK][2] = 62,
+	[0][0][2][0][RTW89_FCC][4] = 78,
+	[0][0][2][0][RTW89_ETSI][4] = 62,
+	[0][0][2][0][RTW89_MKK][4] = 62,
+	[0][0][2][0][RTW89_IC][4] = 64,
+	[0][0][2][0][RTW89_KCC][4] = 74,
+	[0][0][2][0][RTW89_ACMA][4] = 62,
+	[0][0][2][0][RTW89_CN][4] = 62,
+	[0][0][2][0][RTW89_UK][4] = 62,
+	[0][0][2][0][RTW89_FCC][6] = 78,
+	[0][0][2][0][RTW89_ETSI][6] = 62,
+	[0][0][2][0][RTW89_MKK][6] = 62,
+	[0][0][2][0][RTW89_IC][6] = 64,
+	[0][0][2][0][RTW89_KCC][6] = 54,
+	[0][0][2][0][RTW89_ACMA][6] = 62,
+	[0][0][2][0][RTW89_CN][6] = 62,
+	[0][0][2][0][RTW89_UK][6] = 62,
+	[0][0][2][0][RTW89_FCC][8] = 78,
+	[0][0][2][0][RTW89_ETSI][8] = 62,
+	[0][0][2][0][RTW89_MKK][8] = 62,
+	[0][0][2][0][RTW89_IC][8] = 64,
+	[0][0][2][0][RTW89_KCC][8] = 74,
+	[0][0][2][0][RTW89_ACMA][8] = 62,
+	[0][0][2][0][RTW89_CN][8] = 62,
+	[0][0][2][0][RTW89_UK][8] = 62,
+	[0][0][2][0][RTW89_FCC][10] = 78,
+	[0][0][2][0][RTW89_ETSI][10] = 62,
+	[0][0][2][0][RTW89_MKK][10] = 62,
+	[0][0][2][0][RTW89_IC][10] = 64,
+	[0][0][2][0][RTW89_KCC][10] = 74,
+	[0][0][2][0][RTW89_ACMA][10] = 62,
+	[0][0][2][0][RTW89_CN][10] = 62,
+	[0][0][2][0][RTW89_UK][10] = 62,
+	[0][0][2][0][RTW89_FCC][12] = 78,
+	[0][0][2][0][RTW89_ETSI][12] = 62,
+	[0][0][2][0][RTW89_MKK][12] = 62,
+	[0][0][2][0][RTW89_IC][12] = 64,
+	[0][0][2][0][RTW89_KCC][12] = 74,
+	[0][0][2][0][RTW89_ACMA][12] = 62,
+	[0][0][2][0][RTW89_CN][12] = 62,
+	[0][0][2][0][RTW89_UK][12] = 62,
+	[0][0][2][0][RTW89_FCC][14] = 74,
+	[0][0][2][0][RTW89_ETSI][14] = 62,
+	[0][0][2][0][RTW89_MKK][14] = 62,
+	[0][0][2][0][RTW89_IC][14] = 64,
+	[0][0][2][0][RTW89_KCC][14] = 74,
+	[0][0][2][0][RTW89_ACMA][14] = 62,
+	[0][0][2][0][RTW89_CN][14] = 62,
+	[0][0][2][0][RTW89_UK][14] = 62,
+	[0][0][2][0][RTW89_FCC][15] = 74,
+	[0][0][2][0][RTW89_ETSI][15] = 60,
+	[0][0][2][0][RTW89_MKK][15] = 74,
+	[0][0][2][0][RTW89_IC][15] = 74,
+	[0][0][2][0][RTW89_KCC][15] = 74,
+	[0][0][2][0][RTW89_ACMA][15] = 60,
+	[0][0][2][0][RTW89_CN][15] = 127,
+	[0][0][2][0][RTW89_UK][15] = 60,
+	[0][0][2][0][RTW89_FCC][17] = 78,
+	[0][0][2][0][RTW89_ETSI][17] = 62,
+	[0][0][2][0][RTW89_MKK][17] = 74,
+	[0][0][2][0][RTW89_IC][17] = 78,
+	[0][0][2][0][RTW89_KCC][17] = 74,
+	[0][0][2][0][RTW89_ACMA][17] = 62,
+	[0][0][2][0][RTW89_CN][17] = 127,
+	[0][0][2][0][RTW89_UK][17] = 62,
+	[0][0][2][0][RTW89_FCC][19] = 78,
+	[0][0][2][0][RTW89_ETSI][19] = 62,
+	[0][0][2][0][RTW89_MKK][19] = 74,
+	[0][0][2][0][RTW89_IC][19] = 78,
+	[0][0][2][0][RTW89_KCC][19] = 74,
+	[0][0][2][0][RTW89_ACMA][19] = 62,
+	[0][0][2][0][RTW89_CN][19] = 127,
+	[0][0][2][0][RTW89_UK][19] = 62,
+	[0][0][2][0][RTW89_FCC][21] = 78,
+	[0][0][2][0][RTW89_ETSI][21] = 62,
+	[0][0][2][0][RTW89_MKK][21] = 74,
+	[0][0][2][0][RTW89_IC][21] = 78,
+	[0][0][2][0][RTW89_KCC][21] = 74,
+	[0][0][2][0][RTW89_ACMA][21] = 62,
+	[0][0][2][0][RTW89_CN][21] = 127,
+	[0][0][2][0][RTW89_UK][21] = 62,
+	[0][0][2][0][RTW89_FCC][23] = 78,
+	[0][0][2][0][RTW89_ETSI][23] = 62,
+	[0][0][2][0][RTW89_MKK][23] = 74,
+	[0][0][2][0][RTW89_IC][23] = 78,
+	[0][0][2][0][RTW89_KCC][23] = 74,
+	[0][0][2][0][RTW89_ACMA][23] = 62,
+	[0][0][2][0][RTW89_CN][23] = 127,
+	[0][0][2][0][RTW89_UK][23] = 62,
+	[0][0][2][0][RTW89_FCC][25] = 78,
+	[0][0][2][0][RTW89_ETSI][25] = 62,
+	[0][0][2][0][RTW89_MKK][25] = 74,
+	[0][0][2][0][RTW89_IC][25] = 127,
+	[0][0][2][0][RTW89_KCC][25] = 74,
+	[0][0][2][0][RTW89_ACMA][25] = 127,
+	[0][0][2][0][RTW89_CN][25] = 127,
+	[0][0][2][0][RTW89_UK][25] = 62,
+	[0][0][2][0][RTW89_FCC][27] = 78,
+	[0][0][2][0][RTW89_ETSI][27] = 62,
+	[0][0][2][0][RTW89_MKK][27] = 74,
+	[0][0][2][0][RTW89_IC][27] = 127,
+	[0][0][2][0][RTW89_KCC][27] = 74,
+	[0][0][2][0][RTW89_ACMA][27] = 127,
+	[0][0][2][0][RTW89_CN][27] = 127,
+	[0][0][2][0][RTW89_UK][27] = 62,
+	[0][0][2][0][RTW89_FCC][29] = 78,
+	[0][0][2][0][RTW89_ETSI][29] = 62,
+	[0][0][2][0][RTW89_MKK][29] = 74,
+	[0][0][2][0][RTW89_IC][29] = 127,
+	[0][0][2][0][RTW89_KCC][29] = 74,
+	[0][0][2][0][RTW89_ACMA][29] = 127,
+	[0][0][2][0][RTW89_CN][29] = 127,
+	[0][0][2][0][RTW89_UK][29] = 62,
+	[0][0][2][0][RTW89_FCC][31] = 78,
+	[0][0][2][0][RTW89_ETSI][31] = 62,
+	[0][0][2][0][RTW89_MKK][31] = 74,
+	[0][0][2][0][RTW89_IC][31] = 78,
+	[0][0][2][0][RTW89_KCC][31] = 74,
+	[0][0][2][0][RTW89_ACMA][31] = 62,
+	[0][0][2][0][RTW89_CN][31] = 127,
+	[0][0][2][0][RTW89_UK][31] = 62,
+	[0][0][2][0][RTW89_FCC][33] = 78,
+	[0][0][2][0][RTW89_ETSI][33] = 62,
+	[0][0][2][0][RTW89_MKK][33] = 74,
+	[0][0][2][0][RTW89_IC][33] = 78,
+	[0][0][2][0][RTW89_KCC][33] = 74,
+	[0][0][2][0][RTW89_ACMA][33] = 62,
+	[0][0][2][0][RTW89_CN][33] = 127,
+	[0][0][2][0][RTW89_UK][33] = 62,
+	[0][0][2][0][RTW89_FCC][35] = 72,
+	[0][0][2][0][RTW89_ETSI][35] = 62,
+	[0][0][2][0][RTW89_MKK][35] = 74,
+	[0][0][2][0][RTW89_IC][35] = 72,
+	[0][0][2][0][RTW89_KCC][35] = 74,
+	[0][0][2][0][RTW89_ACMA][35] = 62,
+	[0][0][2][0][RTW89_CN][35] = 127,
+	[0][0][2][0][RTW89_UK][35] = 62,
+	[0][0][2][0][RTW89_FCC][37] = 78,
+	[0][0][2][0][RTW89_ETSI][37] = 127,
+	[0][0][2][0][RTW89_MKK][37] = 74,
+	[0][0][2][0][RTW89_IC][37] = 78,
+	[0][0][2][0][RTW89_KCC][37] = 74,
+	[0][0][2][0][RTW89_ACMA][37] = 74,
+	[0][0][2][0][RTW89_CN][37] = 127,
+	[0][0][2][0][RTW89_UK][37] = 74,
+	[0][0][2][0][RTW89_FCC][38] = 78,
+	[0][0][2][0][RTW89_ETSI][38] = 30,
+	[0][0][2][0][RTW89_MKK][38] = 127,
+	[0][0][2][0][RTW89_IC][38] = 78,
+	[0][0][2][0][RTW89_KCC][38] = 66,
+	[0][0][2][0][RTW89_ACMA][38] = 74,
+	[0][0][2][0][RTW89_CN][38] = 74,
+	[0][0][2][0][RTW89_UK][38] = 60,
+	[0][0][2][0][RTW89_FCC][40] = 78,
+	[0][0][2][0][RTW89_ETSI][40] = 30,
+	[0][0][2][0][RTW89_MKK][40] = 127,
+	[0][0][2][0][RTW89_IC][40] = 78,
+	[0][0][2][0][RTW89_KCC][40] = 74,
+	[0][0][2][0][RTW89_ACMA][40] = 74,
+	[0][0][2][0][RTW89_CN][40] = 74,
+	[0][0][2][0][RTW89_UK][40] = 60,
+	[0][0][2][0][RTW89_FCC][42] = 78,
+	[0][0][2][0][RTW89_ETSI][42] = 30,
+	[0][0][2][0][RTW89_MKK][42] = 127,
+	[0][0][2][0][RTW89_IC][42] = 78,
+	[0][0][2][0][RTW89_KCC][42] = 74,
+	[0][0][2][0][RTW89_ACMA][42] = 74,
+	[0][0][2][0][RTW89_CN][42] = 74,
+	[0][0][2][0][RTW89_UK][42] = 60,
+	[0][0][2][0][RTW89_FCC][44] = 78,
+	[0][0][2][0][RTW89_ETSI][44] = 30,
+	[0][0][2][0][RTW89_MKK][44] = 127,
+	[0][0][2][0][RTW89_IC][44] = 78,
+	[0][0][2][0][RTW89_KCC][44] = 74,
+	[0][0][2][0][RTW89_ACMA][44] = 74,
+	[0][0][2][0][RTW89_CN][44] = 74,
+	[0][0][2][0][RTW89_UK][44] = 60,
+	[0][0][2][0][RTW89_FCC][46] = 78,
+	[0][0][2][0][RTW89_ETSI][46] = 30,
+	[0][0][2][0][RTW89_MKK][46] = 127,
+	[0][0][2][0][RTW89_IC][46] = 78,
+	[0][0][2][0][RTW89_KCC][46] = 74,
+	[0][0][2][0][RTW89_ACMA][46] = 74,
+	[0][0][2][0][RTW89_CN][46] = 74,
+	[0][0][2][0][RTW89_UK][46] = 60,
+	[0][0][2][0][RTW89_FCC][48] = 74,
+	[0][0][2][0][RTW89_ETSI][48] = 127,
+	[0][0][2][0][RTW89_MKK][48] = 127,
+	[0][0][2][0][RTW89_IC][48] = 127,
+	[0][0][2][0][RTW89_KCC][48] = 127,
+	[0][0][2][0][RTW89_ACMA][48] = 127,
+	[0][0][2][0][RTW89_CN][48] = 127,
+	[0][0][2][0][RTW89_UK][48] = 127,
+	[0][0][2][0][RTW89_FCC][50] = 74,
+	[0][0][2][0][RTW89_ETSI][50] = 127,
+	[0][0][2][0][RTW89_MKK][50] = 127,
+	[0][0][2][0][RTW89_IC][50] = 127,
+	[0][0][2][0][RTW89_KCC][50] = 127,
+	[0][0][2][0][RTW89_ACMA][50] = 127,
+	[0][0][2][0][RTW89_CN][50] = 127,
+	[0][0][2][0][RTW89_UK][50] = 127,
+	[0][0][2][0][RTW89_FCC][52] = 74,
+	[0][0][2][0][RTW89_ETSI][52] = 127,
+	[0][0][2][0][RTW89_MKK][52] = 127,
+	[0][0][2][0][RTW89_IC][52] = 127,
+	[0][0][2][0][RTW89_KCC][52] = 127,
+	[0][0][2][0][RTW89_ACMA][52] = 127,
+	[0][0][2][0][RTW89_CN][52] = 127,
+	[0][0][2][0][RTW89_UK][52] = 127,
+	[0][1][2][0][RTW89_FCC][0] = 127,
+	[0][1][2][0][RTW89_ETSI][0] = 127,
+	[0][1][2][0][RTW89_MKK][0] = 127,
+	[0][1][2][0][RTW89_IC][0] = 127,
+	[0][1][2][0][RTW89_KCC][0] = 127,
+	[0][1][2][0][RTW89_ACMA][0] = 127,
+	[0][1][2][0][RTW89_CN][0] = 127,
+	[0][1][2][0][RTW89_UK][0] = 127,
+	[0][1][2][0][RTW89_FCC][2] = 127,
+	[0][1][2][0][RTW89_ETSI][2] = 127,
+	[0][1][2][0][RTW89_MKK][2] = 127,
+	[0][1][2][0][RTW89_IC][2] = 127,
+	[0][1][2][0][RTW89_KCC][2] = 127,
+	[0][1][2][0][RTW89_ACMA][2] = 127,
+	[0][1][2][0][RTW89_CN][2] = 127,
+	[0][1][2][0][RTW89_UK][2] = 127,
+	[0][1][2][0][RTW89_FCC][4] = 127,
+	[0][1][2][0][RTW89_ETSI][4] = 127,
+	[0][1][2][0][RTW89_MKK][4] = 127,
+	[0][1][2][0][RTW89_IC][4] = 127,
+	[0][1][2][0][RTW89_KCC][4] = 127,
+	[0][1][2][0][RTW89_ACMA][4] = 127,
+	[0][1][2][0][RTW89_CN][4] = 127,
+	[0][1][2][0][RTW89_UK][4] = 127,
+	[0][1][2][0][RTW89_FCC][6] = 127,
+	[0][1][2][0][RTW89_ETSI][6] = 127,
+	[0][1][2][0][RTW89_MKK][6] = 127,
+	[0][1][2][0][RTW89_IC][6] = 127,
+	[0][1][2][0][RTW89_KCC][6] = 127,
+	[0][1][2][0][RTW89_ACMA][6] = 127,
+	[0][1][2][0][RTW89_CN][6] = 127,
+	[0][1][2][0][RTW89_UK][6] = 127,
+	[0][1][2][0][RTW89_FCC][8] = 127,
+	[0][1][2][0][RTW89_ETSI][8] = 127,
+	[0][1][2][0][RTW89_MKK][8] = 127,
+	[0][1][2][0][RTW89_IC][8] = 127,
+	[0][1][2][0][RTW89_KCC][8] = 127,
+	[0][1][2][0][RTW89_ACMA][8] = 127,
+	[0][1][2][0][RTW89_CN][8] = 127,
+	[0][1][2][0][RTW89_UK][8] = 127,
+	[0][1][2][0][RTW89_FCC][10] = 127,
+	[0][1][2][0][RTW89_ETSI][10] = 127,
+	[0][1][2][0][RTW89_MKK][10] = 127,
+	[0][1][2][0][RTW89_IC][10] = 127,
+	[0][1][2][0][RTW89_KCC][10] = 127,
+	[0][1][2][0][RTW89_ACMA][10] = 127,
+	[0][1][2][0][RTW89_CN][10] = 127,
+	[0][1][2][0][RTW89_UK][10] = 127,
+	[0][1][2][0][RTW89_FCC][12] = 127,
+	[0][1][2][0][RTW89_ETSI][12] = 127,
+	[0][1][2][0][RTW89_MKK][12] = 127,
+	[0][1][2][0][RTW89_IC][12] = 127,
+	[0][1][2][0][RTW89_KCC][12] = 127,
+	[0][1][2][0][RTW89_ACMA][12] = 127,
+	[0][1][2][0][RTW89_CN][12] = 127,
+	[0][1][2][0][RTW89_UK][12] = 127,
+	[0][1][2][0][RTW89_FCC][14] = 127,
+	[0][1][2][0][RTW89_ETSI][14] = 127,
+	[0][1][2][0][RTW89_MKK][14] = 127,
+	[0][1][2][0][RTW89_IC][14] = 127,
+	[0][1][2][0][RTW89_KCC][14] = 127,
+	[0][1][2][0][RTW89_ACMA][14] = 127,
+	[0][1][2][0][RTW89_CN][14] = 127,
+	[0][1][2][0][RTW89_UK][14] = 127,
+	[0][1][2][0][RTW89_FCC][15] = 127,
+	[0][1][2][0][RTW89_ETSI][15] = 127,
+	[0][1][2][0][RTW89_MKK][15] = 127,
+	[0][1][2][0][RTW89_IC][15] = 127,
+	[0][1][2][0][RTW89_KCC][15] = 127,
+	[0][1][2][0][RTW89_ACMA][15] = 127,
+	[0][1][2][0][RTW89_CN][15] = 127,
+	[0][1][2][0][RTW89_UK][15] = 127,
+	[0][1][2][0][RTW89_FCC][17] = 127,
+	[0][1][2][0][RTW89_ETSI][17] = 127,
+	[0][1][2][0][RTW89_MKK][17] = 127,
+	[0][1][2][0][RTW89_IC][17] = 127,
+	[0][1][2][0][RTW89_KCC][17] = 127,
+	[0][1][2][0][RTW89_ACMA][17] = 127,
+	[0][1][2][0][RTW89_CN][17] = 127,
+	[0][1][2][0][RTW89_UK][17] = 127,
+	[0][1][2][0][RTW89_FCC][19] = 127,
+	[0][1][2][0][RTW89_ETSI][19] = 127,
+	[0][1][2][0][RTW89_MKK][19] = 127,
+	[0][1][2][0][RTW89_IC][19] = 127,
+	[0][1][2][0][RTW89_KCC][19] = 127,
+	[0][1][2][0][RTW89_ACMA][19] = 127,
+	[0][1][2][0][RTW89_CN][19] = 127,
+	[0][1][2][0][RTW89_UK][19] = 127,
+	[0][1][2][0][RTW89_FCC][21] = 127,
+	[0][1][2][0][RTW89_ETSI][21] = 127,
+	[0][1][2][0][RTW89_MKK][21] = 127,
+	[0][1][2][0][RTW89_IC][21] = 127,
+	[0][1][2][0][RTW89_KCC][21] = 127,
+	[0][1][2][0][RTW89_ACMA][21] = 127,
+	[0][1][2][0][RTW89_CN][21] = 127,
+	[0][1][2][0][RTW89_UK][21] = 127,
+	[0][1][2][0][RTW89_FCC][23] = 127,
+	[0][1][2][0][RTW89_ETSI][23] = 127,
+	[0][1][2][0][RTW89_MKK][23] = 127,
+	[0][1][2][0][RTW89_IC][23] = 127,
+	[0][1][2][0][RTW89_KCC][23] = 127,
+	[0][1][2][0][RTW89_ACMA][23] = 127,
+	[0][1][2][0][RTW89_CN][23] = 127,
+	[0][1][2][0][RTW89_UK][23] = 127,
+	[0][1][2][0][RTW89_FCC][25] = 127,
+	[0][1][2][0][RTW89_ETSI][25] = 127,
+	[0][1][2][0][RTW89_MKK][25] = 127,
+	[0][1][2][0][RTW89_IC][25] = 127,
+	[0][1][2][0][RTW89_KCC][25] = 127,
+	[0][1][2][0][RTW89_ACMA][25] = 127,
+	[0][1][2][0][RTW89_CN][25] = 127,
+	[0][1][2][0][RTW89_UK][25] = 127,
+	[0][1][2][0][RTW89_FCC][27] = 127,
+	[0][1][2][0][RTW89_ETSI][27] = 127,
+	[0][1][2][0][RTW89_MKK][27] = 127,
+	[0][1][2][0][RTW89_IC][27] = 127,
+	[0][1][2][0][RTW89_KCC][27] = 127,
+	[0][1][2][0][RTW89_ACMA][27] = 127,
+	[0][1][2][0][RTW89_CN][27] = 127,
+	[0][1][2][0][RTW89_UK][27] = 127,
+	[0][1][2][0][RTW89_FCC][29] = 127,
+	[0][1][2][0][RTW89_ETSI][29] = 127,
+	[0][1][2][0][RTW89_MKK][29] = 127,
+	[0][1][2][0][RTW89_IC][29] = 127,
+	[0][1][2][0][RTW89_KCC][29] = 127,
+	[0][1][2][0][RTW89_ACMA][29] = 127,
+	[0][1][2][0][RTW89_CN][29] = 127,
+	[0][1][2][0][RTW89_UK][29] = 127,
+	[0][1][2][0][RTW89_FCC][31] = 127,
+	[0][1][2][0][RTW89_ETSI][31] = 127,
+	[0][1][2][0][RTW89_MKK][31] = 127,
+	[0][1][2][0][RTW89_IC][31] = 127,
+	[0][1][2][0][RTW89_KCC][31] = 127,
+	[0][1][2][0][RTW89_ACMA][31] = 127,
+	[0][1][2][0][RTW89_CN][31] = 127,
+	[0][1][2][0][RTW89_UK][31] = 127,
+	[0][1][2][0][RTW89_FCC][33] = 127,
+	[0][1][2][0][RTW89_ETSI][33] = 127,
+	[0][1][2][0][RTW89_MKK][33] = 127,
+	[0][1][2][0][RTW89_IC][33] = 127,
+	[0][1][2][0][RTW89_KCC][33] = 127,
+	[0][1][2][0][RTW89_ACMA][33] = 127,
+	[0][1][2][0][RTW89_CN][33] = 127,
+	[0][1][2][0][RTW89_UK][33] = 127,
+	[0][1][2][0][RTW89_FCC][35] = 127,
+	[0][1][2][0][RTW89_ETSI][35] = 127,
+	[0][1][2][0][RTW89_MKK][35] = 127,
+	[0][1][2][0][RTW89_IC][35] = 127,
+	[0][1][2][0][RTW89_KCC][35] = 127,
+	[0][1][2][0][RTW89_ACMA][35] = 127,
+	[0][1][2][0][RTW89_CN][35] = 127,
+	[0][1][2][0][RTW89_UK][35] = 127,
+	[0][1][2][0][RTW89_FCC][37] = 127,
+	[0][1][2][0][RTW89_ETSI][37] = 127,
+	[0][1][2][0][RTW89_MKK][37] = 127,
+	[0][1][2][0][RTW89_IC][37] = 127,
+	[0][1][2][0][RTW89_KCC][37] = 127,
+	[0][1][2][0][RTW89_ACMA][37] = 127,
+	[0][1][2][0][RTW89_CN][37] = 127,
+	[0][1][2][0][RTW89_UK][37] = 127,
+	[0][1][2][0][RTW89_FCC][38] = 127,
+	[0][1][2][0][RTW89_ETSI][38] = 127,
+	[0][1][2][0][RTW89_MKK][38] = 127,
+	[0][1][2][0][RTW89_IC][38] = 127,
+	[0][1][2][0][RTW89_KCC][38] = 127,
+	[0][1][2][0][RTW89_ACMA][38] = 127,
+	[0][1][2][0][RTW89_CN][38] = 127,
+	[0][1][2][0][RTW89_UK][38] = 127,
+	[0][1][2][0][RTW89_FCC][40] = 127,
+	[0][1][2][0][RTW89_ETSI][40] = 127,
+	[0][1][2][0][RTW89_MKK][40] = 127,
+	[0][1][2][0][RTW89_IC][40] = 127,
+	[0][1][2][0][RTW89_KCC][40] = 127,
+	[0][1][2][0][RTW89_ACMA][40] = 127,
+	[0][1][2][0][RTW89_CN][40] = 127,
+	[0][1][2][0][RTW89_UK][40] = 127,
+	[0][1][2][0][RTW89_FCC][42] = 127,
+	[0][1][2][0][RTW89_ETSI][42] = 127,
+	[0][1][2][0][RTW89_MKK][42] = 127,
+	[0][1][2][0][RTW89_IC][42] = 127,
+	[0][1][2][0][RTW89_KCC][42] = 127,
+	[0][1][2][0][RTW89_ACMA][42] = 127,
+	[0][1][2][0][RTW89_CN][42] = 127,
+	[0][1][2][0][RTW89_UK][42] = 127,
+	[0][1][2][0][RTW89_FCC][44] = 127,
+	[0][1][2][0][RTW89_ETSI][44] = 127,
+	[0][1][2][0][RTW89_MKK][44] = 127,
+	[0][1][2][0][RTW89_IC][44] = 127,
+	[0][1][2][0][RTW89_KCC][44] = 127,
+	[0][1][2][0][RTW89_ACMA][44] = 127,
+	[0][1][2][0][RTW89_CN][44] = 127,
+	[0][1][2][0][RTW89_UK][44] = 127,
+	[0][1][2][0][RTW89_FCC][46] = 127,
+	[0][1][2][0][RTW89_ETSI][46] = 127,
+	[0][1][2][0][RTW89_MKK][46] = 127,
+	[0][1][2][0][RTW89_IC][46] = 127,
+	[0][1][2][0][RTW89_KCC][46] = 127,
+	[0][1][2][0][RTW89_ACMA][46] = 127,
+	[0][1][2][0][RTW89_CN][46] = 127,
+	[0][1][2][0][RTW89_UK][46] = 127,
+	[0][1][2][0][RTW89_FCC][48] = 127,
+	[0][1][2][0][RTW89_ETSI][48] = 127,
+	[0][1][2][0][RTW89_MKK][48] = 127,
+	[0][1][2][0][RTW89_IC][48] = 127,
+	[0][1][2][0][RTW89_KCC][48] = 127,
+	[0][1][2][0][RTW89_ACMA][48] = 127,
+	[0][1][2][0][RTW89_CN][48] = 127,
+	[0][1][2][0][RTW89_UK][48] = 127,
+	[0][1][2][0][RTW89_FCC][50] = 127,
+	[0][1][2][0][RTW89_ETSI][50] = 127,
+	[0][1][2][0][RTW89_MKK][50] = 127,
+	[0][1][2][0][RTW89_IC][50] = 127,
+	[0][1][2][0][RTW89_KCC][50] = 127,
+	[0][1][2][0][RTW89_ACMA][50] = 127,
+	[0][1][2][0][RTW89_CN][50] = 127,
+	[0][1][2][0][RTW89_UK][50] = 127,
+	[0][1][2][0][RTW89_FCC][52] = 127,
+	[0][1][2][0][RTW89_ETSI][52] = 127,
+	[0][1][2][0][RTW89_MKK][52] = 127,
+	[0][1][2][0][RTW89_IC][52] = 127,
+	[0][1][2][0][RTW89_KCC][52] = 127,
+	[0][1][2][0][RTW89_ACMA][52] = 127,
+	[0][1][2][0][RTW89_CN][52] = 127,
+	[0][1][2][0][RTW89_UK][52] = 127,
+	[0][1][2][1][RTW89_FCC][0] = 127,
+	[0][1][2][1][RTW89_ETSI][0] = 127,
+	[0][1][2][1][RTW89_MKK][0] = 127,
+	[0][1][2][1][RTW89_IC][0] = 127,
+	[0][1][2][1][RTW89_KCC][0] = 127,
+	[0][1][2][1][RTW89_ACMA][0] = 127,
+	[0][1][2][1][RTW89_CN][0] = 127,
+	[0][1][2][1][RTW89_UK][0] = 127,
+	[0][1][2][1][RTW89_FCC][2] = 127,
+	[0][1][2][1][RTW89_ETSI][2] = 127,
+	[0][1][2][1][RTW89_MKK][2] = 127,
+	[0][1][2][1][RTW89_IC][2] = 127,
+	[0][1][2][1][RTW89_KCC][2] = 127,
+	[0][1][2][1][RTW89_ACMA][2] = 127,
+	[0][1][2][1][RTW89_CN][2] = 127,
+	[0][1][2][1][RTW89_UK][2] = 127,
+	[0][1][2][1][RTW89_FCC][4] = 127,
+	[0][1][2][1][RTW89_ETSI][4] = 127,
+	[0][1][2][1][RTW89_MKK][4] = 127,
+	[0][1][2][1][RTW89_IC][4] = 127,
+	[0][1][2][1][RTW89_KCC][4] = 127,
+	[0][1][2][1][RTW89_ACMA][4] = 127,
+	[0][1][2][1][RTW89_CN][4] = 127,
+	[0][1][2][1][RTW89_UK][4] = 127,
+	[0][1][2][1][RTW89_FCC][6] = 127,
+	[0][1][2][1][RTW89_ETSI][6] = 127,
+	[0][1][2][1][RTW89_MKK][6] = 127,
+	[0][1][2][1][RTW89_IC][6] = 127,
+	[0][1][2][1][RTW89_KCC][6] = 127,
+	[0][1][2][1][RTW89_ACMA][6] = 127,
+	[0][1][2][1][RTW89_CN][6] = 127,
+	[0][1][2][1][RTW89_UK][6] = 127,
+	[0][1][2][1][RTW89_FCC][8] = 127,
+	[0][1][2][1][RTW89_ETSI][8] = 127,
+	[0][1][2][1][RTW89_MKK][8] = 127,
+	[0][1][2][1][RTW89_IC][8] = 127,
+	[0][1][2][1][RTW89_KCC][8] = 127,
+	[0][1][2][1][RTW89_ACMA][8] = 127,
+	[0][1][2][1][RTW89_CN][8] = 127,
+	[0][1][2][1][RTW89_UK][8] = 127,
+	[0][1][2][1][RTW89_FCC][10] = 127,
+	[0][1][2][1][RTW89_ETSI][10] = 127,
+	[0][1][2][1][RTW89_MKK][10] = 127,
+	[0][1][2][1][RTW89_IC][10] = 127,
+	[0][1][2][1][RTW89_KCC][10] = 127,
+	[0][1][2][1][RTW89_ACMA][10] = 127,
+	[0][1][2][1][RTW89_CN][10] = 127,
+	[0][1][2][1][RTW89_UK][10] = 127,
+	[0][1][2][1][RTW89_FCC][12] = 127,
+	[0][1][2][1][RTW89_ETSI][12] = 127,
+	[0][1][2][1][RTW89_MKK][12] = 127,
+	[0][1][2][1][RTW89_IC][12] = 127,
+	[0][1][2][1][RTW89_KCC][12] = 127,
+	[0][1][2][1][RTW89_ACMA][12] = 127,
+	[0][1][2][1][RTW89_CN][12] = 127,
+	[0][1][2][1][RTW89_UK][12] = 127,
+	[0][1][2][1][RTW89_FCC][14] = 127,
+	[0][1][2][1][RTW89_ETSI][14] = 127,
+	[0][1][2][1][RTW89_MKK][14] = 127,
+	[0][1][2][1][RTW89_IC][14] = 127,
+	[0][1][2][1][RTW89_KCC][14] = 127,
+	[0][1][2][1][RTW89_ACMA][14] = 127,
+	[0][1][2][1][RTW89_CN][14] = 127,
+	[0][1][2][1][RTW89_UK][14] = 127,
+	[0][1][2][1][RTW89_FCC][15] = 127,
+	[0][1][2][1][RTW89_ETSI][15] = 127,
+	[0][1][2][1][RTW89_MKK][15] = 127,
+	[0][1][2][1][RTW89_IC][15] = 127,
+	[0][1][2][1][RTW89_KCC][15] = 127,
+	[0][1][2][1][RTW89_ACMA][15] = 127,
+	[0][1][2][1][RTW89_CN][15] = 127,
+	[0][1][2][1][RTW89_UK][15] = 127,
+	[0][1][2][1][RTW89_FCC][17] = 127,
+	[0][1][2][1][RTW89_ETSI][17] = 127,
+	[0][1][2][1][RTW89_MKK][17] = 127,
+	[0][1][2][1][RTW89_IC][17] = 127,
+	[0][1][2][1][RTW89_KCC][17] = 127,
+	[0][1][2][1][RTW89_ACMA][17] = 127,
+	[0][1][2][1][RTW89_CN][17] = 127,
+	[0][1][2][1][RTW89_UK][17] = 127,
+	[0][1][2][1][RTW89_FCC][19] = 127,
+	[0][1][2][1][RTW89_ETSI][19] = 127,
+	[0][1][2][1][RTW89_MKK][19] = 127,
+	[0][1][2][1][RTW89_IC][19] = 127,
+	[0][1][2][1][RTW89_KCC][19] = 127,
+	[0][1][2][1][RTW89_ACMA][19] = 127,
+	[0][1][2][1][RTW89_CN][19] = 127,
+	[0][1][2][1][RTW89_UK][19] = 127,
+	[0][1][2][1][RTW89_FCC][21] = 127,
+	[0][1][2][1][RTW89_ETSI][21] = 127,
+	[0][1][2][1][RTW89_MKK][21] = 127,
+	[0][1][2][1][RTW89_IC][21] = 127,
+	[0][1][2][1][RTW89_KCC][21] = 127,
+	[0][1][2][1][RTW89_ACMA][21] = 127,
+	[0][1][2][1][RTW89_CN][21] = 127,
+	[0][1][2][1][RTW89_UK][21] = 127,
+	[0][1][2][1][RTW89_FCC][23] = 127,
+	[0][1][2][1][RTW89_ETSI][23] = 127,
+	[0][1][2][1][RTW89_MKK][23] = 127,
+	[0][1][2][1][RTW89_IC][23] = 127,
+	[0][1][2][1][RTW89_KCC][23] = 127,
+	[0][1][2][1][RTW89_ACMA][23] = 127,
+	[0][1][2][1][RTW89_CN][23] = 127,
+	[0][1][2][1][RTW89_UK][23] = 127,
+	[0][1][2][1][RTW89_FCC][25] = 127,
+	[0][1][2][1][RTW89_ETSI][25] = 127,
+	[0][1][2][1][RTW89_MKK][25] = 127,
+	[0][1][2][1][RTW89_IC][25] = 127,
+	[0][1][2][1][RTW89_KCC][25] = 127,
+	[0][1][2][1][RTW89_ACMA][25] = 127,
+	[0][1][2][1][RTW89_CN][25] = 127,
+	[0][1][2][1][RTW89_UK][25] = 127,
+	[0][1][2][1][RTW89_FCC][27] = 127,
+	[0][1][2][1][RTW89_ETSI][27] = 127,
+	[0][1][2][1][RTW89_MKK][27] = 127,
+	[0][1][2][1][RTW89_IC][27] = 127,
+	[0][1][2][1][RTW89_KCC][27] = 127,
+	[0][1][2][1][RTW89_ACMA][27] = 127,
+	[0][1][2][1][RTW89_CN][27] = 127,
+	[0][1][2][1][RTW89_UK][27] = 127,
+	[0][1][2][1][RTW89_FCC][29] = 127,
+	[0][1][2][1][RTW89_ETSI][29] = 127,
+	[0][1][2][1][RTW89_MKK][29] = 127,
+	[0][1][2][1][RTW89_IC][29] = 127,
+	[0][1][2][1][RTW89_KCC][29] = 127,
+	[0][1][2][1][RTW89_ACMA][29] = 127,
+	[0][1][2][1][RTW89_CN][29] = 127,
+	[0][1][2][1][RTW89_UK][29] = 127,
+	[0][1][2][1][RTW89_FCC][31] = 127,
+	[0][1][2][1][RTW89_ETSI][31] = 127,
+	[0][1][2][1][RTW89_MKK][31] = 127,
+	[0][1][2][1][RTW89_IC][31] = 127,
+	[0][1][2][1][RTW89_KCC][31] = 127,
+	[0][1][2][1][RTW89_ACMA][31] = 127,
+	[0][1][2][1][RTW89_CN][31] = 127,
+	[0][1][2][1][RTW89_UK][31] = 127,
+	[0][1][2][1][RTW89_FCC][33] = 127,
+	[0][1][2][1][RTW89_ETSI][33] = 127,
+	[0][1][2][1][RTW89_MKK][33] = 127,
+	[0][1][2][1][RTW89_IC][33] = 127,
+	[0][1][2][1][RTW89_KCC][33] = 127,
+	[0][1][2][1][RTW89_ACMA][33] = 127,
+	[0][1][2][1][RTW89_CN][33] = 127,
+	[0][1][2][1][RTW89_UK][33] = 127,
+	[0][1][2][1][RTW89_FCC][35] = 127,
+	[0][1][2][1][RTW89_ETSI][35] = 127,
+	[0][1][2][1][RTW89_MKK][35] = 127,
+	[0][1][2][1][RTW89_IC][35] = 127,
+	[0][1][2][1][RTW89_KCC][35] = 127,
+	[0][1][2][1][RTW89_ACMA][35] = 127,
+	[0][1][2][1][RTW89_CN][35] = 127,
+	[0][1][2][1][RTW89_UK][35] = 127,
+	[0][1][2][1][RTW89_FCC][37] = 127,
+	[0][1][2][1][RTW89_ETSI][37] = 127,
+	[0][1][2][1][RTW89_MKK][37] = 127,
+	[0][1][2][1][RTW89_IC][37] = 127,
+	[0][1][2][1][RTW89_KCC][37] = 127,
+	[0][1][2][1][RTW89_ACMA][37] = 127,
+	[0][1][2][1][RTW89_CN][37] = 127,
+	[0][1][2][1][RTW89_UK][37] = 127,
+	[0][1][2][1][RTW89_FCC][38] = 127,
+	[0][1][2][1][RTW89_ETSI][38] = 127,
+	[0][1][2][1][RTW89_MKK][38] = 127,
+	[0][1][2][1][RTW89_IC][38] = 127,
+	[0][1][2][1][RTW89_KCC][38] = 127,
+	[0][1][2][1][RTW89_ACMA][38] = 127,
+	[0][1][2][1][RTW89_CN][38] = 127,
+	[0][1][2][1][RTW89_UK][38] = 127,
+	[0][1][2][1][RTW89_FCC][40] = 127,
+	[0][1][2][1][RTW89_ETSI][40] = 127,
+	[0][1][2][1][RTW89_MKK][40] = 127,
+	[0][1][2][1][RTW89_IC][40] = 127,
+	[0][1][2][1][RTW89_KCC][40] = 127,
+	[0][1][2][1][RTW89_ACMA][40] = 127,
+	[0][1][2][1][RTW89_CN][40] = 127,
+	[0][1][2][1][RTW89_UK][40] = 127,
+	[0][1][2][1][RTW89_FCC][42] = 127,
+	[0][1][2][1][RTW89_ETSI][42] = 127,
+	[0][1][2][1][RTW89_MKK][42] = 127,
+	[0][1][2][1][RTW89_IC][42] = 127,
+	[0][1][2][1][RTW89_KCC][42] = 127,
+	[0][1][2][1][RTW89_ACMA][42] = 127,
+	[0][1][2][1][RTW89_CN][42] = 127,
+	[0][1][2][1][RTW89_UK][42] = 127,
+	[0][1][2][1][RTW89_FCC][44] = 127,
+	[0][1][2][1][RTW89_ETSI][44] = 127,
+	[0][1][2][1][RTW89_MKK][44] = 127,
+	[0][1][2][1][RTW89_IC][44] = 127,
+	[0][1][2][1][RTW89_KCC][44] = 127,
+	[0][1][2][1][RTW89_ACMA][44] = 127,
+	[0][1][2][1][RTW89_CN][44] = 127,
+	[0][1][2][1][RTW89_UK][44] = 127,
+	[0][1][2][1][RTW89_FCC][46] = 127,
+	[0][1][2][1][RTW89_ETSI][46] = 127,
+	[0][1][2][1][RTW89_MKK][46] = 127,
+	[0][1][2][1][RTW89_IC][46] = 127,
+	[0][1][2][1][RTW89_KCC][46] = 127,
+	[0][1][2][1][RTW89_ACMA][46] = 127,
+	[0][1][2][1][RTW89_CN][46] = 127,
+	[0][1][2][1][RTW89_UK][46] = 127,
+	[0][1][2][1][RTW89_FCC][48] = 127,
+	[0][1][2][1][RTW89_ETSI][48] = 127,
+	[0][1][2][1][RTW89_MKK][48] = 127,
+	[0][1][2][1][RTW89_IC][48] = 127,
+	[0][1][2][1][RTW89_KCC][48] = 127,
+	[0][1][2][1][RTW89_ACMA][48] = 127,
+	[0][1][2][1][RTW89_CN][48] = 127,
+	[0][1][2][1][RTW89_UK][48] = 127,
+	[0][1][2][1][RTW89_FCC][50] = 127,
+	[0][1][2][1][RTW89_ETSI][50] = 127,
+	[0][1][2][1][RTW89_MKK][50] = 127,
+	[0][1][2][1][RTW89_IC][50] = 127,
+	[0][1][2][1][RTW89_KCC][50] = 127,
+	[0][1][2][1][RTW89_ACMA][50] = 127,
+	[0][1][2][1][RTW89_CN][50] = 127,
+	[0][1][2][1][RTW89_UK][50] = 127,
+	[0][1][2][1][RTW89_FCC][52] = 127,
+	[0][1][2][1][RTW89_ETSI][52] = 127,
+	[0][1][2][1][RTW89_MKK][52] = 127,
+	[0][1][2][1][RTW89_IC][52] = 127,
+	[0][1][2][1][RTW89_KCC][52] = 127,
+	[0][1][2][1][RTW89_ACMA][52] = 127,
+	[0][1][2][1][RTW89_CN][52] = 127,
+	[0][1][2][1][RTW89_UK][52] = 127,
+	[1][0][2][0][RTW89_FCC][1] = 66,
+	[1][0][2][0][RTW89_ETSI][1] = 64,
+	[1][0][2][0][RTW89_MKK][1] = 64,
+	[1][0][2][0][RTW89_IC][1] = 64,
+	[1][0][2][0][RTW89_KCC][1] = 74,
+	[1][0][2][0][RTW89_ACMA][1] = 64,
+	[1][0][2][0][RTW89_CN][1] = 64,
+	[1][0][2][0][RTW89_UK][1] = 64,
+	[1][0][2][0][RTW89_FCC][5] = 80,
+	[1][0][2][0][RTW89_ETSI][5] = 64,
+	[1][0][2][0][RTW89_MKK][5] = 62,
+	[1][0][2][0][RTW89_IC][5] = 64,
+	[1][0][2][0][RTW89_KCC][5] = 66,
+	[1][0][2][0][RTW89_ACMA][5] = 64,
+	[1][0][2][0][RTW89_CN][5] = 64,
+	[1][0][2][0][RTW89_UK][5] = 64,
+	[1][0][2][0][RTW89_FCC][9] = 80,
+	[1][0][2][0][RTW89_ETSI][9] = 64,
+	[1][0][2][0][RTW89_MKK][9] = 64,
+	[1][0][2][0][RTW89_IC][9] = 64,
+	[1][0][2][0][RTW89_KCC][9] = 76,
+	[1][0][2][0][RTW89_ACMA][9] = 64,
+	[1][0][2][0][RTW89_CN][9] = 64,
+	[1][0][2][0][RTW89_UK][9] = 64,
+	[1][0][2][0][RTW89_FCC][13] = 64,
+	[1][0][2][0][RTW89_ETSI][13] = 64,
+	[1][0][2][0][RTW89_MKK][13] = 64,
+	[1][0][2][0][RTW89_IC][13] = 64,
+	[1][0][2][0][RTW89_KCC][13] = 72,
+	[1][0][2][0][RTW89_ACMA][13] = 64,
+	[1][0][2][0][RTW89_CN][13] = 64,
+	[1][0][2][0][RTW89_UK][13] = 64,
+	[1][0][2][0][RTW89_FCC][16] = 66,
+	[1][0][2][0][RTW89_ETSI][16] = 66,
+	[1][0][2][0][RTW89_MKK][16] = 76,
+	[1][0][2][0][RTW89_IC][16] = 66,
+	[1][0][2][0][RTW89_KCC][16] = 74,
+	[1][0][2][0][RTW89_ACMA][16] = 66,
+	[1][0][2][0][RTW89_CN][16] = 127,
+	[1][0][2][0][RTW89_UK][16] = 66,
+	[1][0][2][0][RTW89_FCC][20] = 80,
+	[1][0][2][0][RTW89_ETSI][20] = 66,
+	[1][0][2][0][RTW89_MKK][20] = 76,
+	[1][0][2][0][RTW89_IC][20] = 80,
+	[1][0][2][0][RTW89_KCC][20] = 74,
+	[1][0][2][0][RTW89_ACMA][20] = 66,
+	[1][0][2][0][RTW89_CN][20] = 127,
+	[1][0][2][0][RTW89_UK][20] = 66,
+	[1][0][2][0][RTW89_FCC][24] = 80,
+	[1][0][2][0][RTW89_ETSI][24] = 66,
+	[1][0][2][0][RTW89_MKK][24] = 76,
+	[1][0][2][0][RTW89_IC][24] = 127,
+	[1][0][2][0][RTW89_KCC][24] = 74,
+	[1][0][2][0][RTW89_ACMA][24] = 127,
+	[1][0][2][0][RTW89_CN][24] = 127,
+	[1][0][2][0][RTW89_UK][24] = 66,
+	[1][0][2][0][RTW89_FCC][28] = 80,
+	[1][0][2][0][RTW89_ETSI][28] = 66,
+	[1][0][2][0][RTW89_MKK][28] = 76,
+	[1][0][2][0][RTW89_IC][28] = 127,
+	[1][0][2][0][RTW89_KCC][28] = 74,
+	[1][0][2][0][RTW89_ACMA][28] = 127,
+	[1][0][2][0][RTW89_CN][28] = 127,
+	[1][0][2][0][RTW89_UK][28] = 66,
+	[1][0][2][0][RTW89_FCC][32] = 74,
+	[1][0][2][0][RTW89_ETSI][32] = 66,
+	[1][0][2][0][RTW89_MKK][32] = 76,
+	[1][0][2][0][RTW89_IC][32] = 74,
+	[1][0][2][0][RTW89_KCC][32] = 76,
+	[1][0][2][0][RTW89_ACMA][32] = 66,
+	[1][0][2][0][RTW89_CN][32] = 127,
+	[1][0][2][0][RTW89_UK][32] = 66,
+	[1][0][2][0][RTW89_FCC][36] = 78,
+	[1][0][2][0][RTW89_ETSI][36] = 127,
+	[1][0][2][0][RTW89_MKK][36] = 76,
+	[1][0][2][0][RTW89_IC][36] = 78,
+	[1][0][2][0][RTW89_KCC][36] = 76,
+	[1][0][2][0][RTW89_ACMA][36] = 76,
+	[1][0][2][0][RTW89_CN][36] = 127,
+	[1][0][2][0][RTW89_UK][36] = 76,
+	[1][0][2][0][RTW89_FCC][39] = 80,
+	[1][0][2][0][RTW89_ETSI][39] = 30,
+	[1][0][2][0][RTW89_MKK][39] = 127,
+	[1][0][2][0][RTW89_IC][39] = 80,
+	[1][0][2][0][RTW89_KCC][39] = 68,
+	[1][0][2][0][RTW89_ACMA][39] = 76,
+	[1][0][2][0][RTW89_CN][39] = 70,
+	[1][0][2][0][RTW89_UK][39] = 64,
+	[1][0][2][0][RTW89_FCC][43] = 80,
+	[1][0][2][0][RTW89_ETSI][43] = 30,
+	[1][0][2][0][RTW89_MKK][43] = 127,
+	[1][0][2][0][RTW89_IC][43] = 80,
+	[1][0][2][0][RTW89_KCC][43] = 76,
+	[1][0][2][0][RTW89_ACMA][43] = 76,
+	[1][0][2][0][RTW89_CN][43] = 76,
+	[1][0][2][0][RTW89_UK][43] = 64,
+	[1][0][2][0][RTW89_FCC][47] = 80,
+	[1][0][2][0][RTW89_ETSI][47] = 127,
+	[1][0][2][0][RTW89_MKK][47] = 127,
+	[1][0][2][0][RTW89_IC][47] = 127,
+	[1][0][2][0][RTW89_KCC][47] = 127,
+	[1][0][2][0][RTW89_ACMA][47] = 127,
+	[1][0][2][0][RTW89_CN][47] = 127,
+	[1][0][2][0][RTW89_UK][47] = 127,
+	[1][0][2][0][RTW89_FCC][51] = 80,
+	[1][0][2][0][RTW89_ETSI][51] = 127,
+	[1][0][2][0][RTW89_MKK][51] = 127,
+	[1][0][2][0][RTW89_IC][51] = 127,
+	[1][0][2][0][RTW89_KCC][51] = 127,
+	[1][0][2][0][RTW89_ACMA][51] = 127,
+	[1][0][2][0][RTW89_CN][51] = 127,
+	[1][0][2][0][RTW89_UK][51] = 127,
+	[1][1][2][0][RTW89_FCC][1] = 127,
+	[1][1][2][0][RTW89_ETSI][1] = 127,
+	[1][1][2][0][RTW89_MKK][1] = 127,
+	[1][1][2][0][RTW89_IC][1] = 127,
+	[1][1][2][0][RTW89_KCC][1] = 127,
+	[1][1][2][0][RTW89_ACMA][1] = 127,
+	[1][1][2][0][RTW89_CN][1] = 127,
+	[1][1][2][0][RTW89_UK][1] = 127,
+	[1][1][2][0][RTW89_FCC][5] = 127,
+	[1][1][2][0][RTW89_ETSI][5] = 127,
+	[1][1][2][0][RTW89_MKK][5] = 127,
+	[1][1][2][0][RTW89_IC][5] = 127,
+	[1][1][2][0][RTW89_KCC][5] = 127,
+	[1][1][2][0][RTW89_ACMA][5] = 127,
+	[1][1][2][0][RTW89_CN][5] = 127,
+	[1][1][2][0][RTW89_UK][5] = 127,
+	[1][1][2][0][RTW89_FCC][9] = 127,
+	[1][1][2][0][RTW89_ETSI][9] = 127,
+	[1][1][2][0][RTW89_MKK][9] = 127,
+	[1][1][2][0][RTW89_IC][9] = 127,
+	[1][1][2][0][RTW89_KCC][9] = 127,
+	[1][1][2][0][RTW89_ACMA][9] = 127,
+	[1][1][2][0][RTW89_CN][9] = 127,
+	[1][1][2][0][RTW89_UK][9] = 127,
+	[1][1][2][0][RTW89_FCC][13] = 127,
+	[1][1][2][0][RTW89_ETSI][13] = 127,
+	[1][1][2][0][RTW89_MKK][13] = 127,
+	[1][1][2][0][RTW89_IC][13] = 127,
+	[1][1][2][0][RTW89_KCC][13] = 127,
+	[1][1][2][0][RTW89_ACMA][13] = 127,
+	[1][1][2][0][RTW89_CN][13] = 127,
+	[1][1][2][0][RTW89_UK][13] = 127,
+	[1][1][2][0][RTW89_FCC][16] = 127,
+	[1][1][2][0][RTW89_ETSI][16] = 127,
+	[1][1][2][0][RTW89_MKK][16] = 127,
+	[1][1][2][0][RTW89_IC][16] = 127,
+	[1][1][2][0][RTW89_KCC][16] = 127,
+	[1][1][2][0][RTW89_ACMA][16] = 127,
+	[1][1][2][0][RTW89_CN][16] = 127,
+	[1][1][2][0][RTW89_UK][16] = 127,
+	[1][1][2][0][RTW89_FCC][20] = 127,
+	[1][1][2][0][RTW89_ETSI][20] = 127,
+	[1][1][2][0][RTW89_MKK][20] = 127,
+	[1][1][2][0][RTW89_IC][20] = 127,
+	[1][1][2][0][RTW89_KCC][20] = 127,
+	[1][1][2][0][RTW89_ACMA][20] = 127,
+	[1][1][2][0][RTW89_CN][20] = 127,
+	[1][1][2][0][RTW89_UK][20] = 127,
+	[1][1][2][0][RTW89_FCC][24] = 127,
+	[1][1][2][0][RTW89_ETSI][24] = 127,
+	[1][1][2][0][RTW89_MKK][24] = 127,
+	[1][1][2][0][RTW89_IC][24] = 127,
+	[1][1][2][0][RTW89_KCC][24] = 127,
+	[1][1][2][0][RTW89_ACMA][24] = 127,
+	[1][1][2][0][RTW89_CN][24] = 127,
+	[1][1][2][0][RTW89_UK][24] = 127,
+	[1][1][2][0][RTW89_FCC][28] = 127,
+	[1][1][2][0][RTW89_ETSI][28] = 127,
+	[1][1][2][0][RTW89_MKK][28] = 127,
+	[1][1][2][0][RTW89_IC][28] = 127,
+	[1][1][2][0][RTW89_KCC][28] = 127,
+	[1][1][2][0][RTW89_ACMA][28] = 127,
+	[1][1][2][0][RTW89_CN][28] = 127,
+	[1][1][2][0][RTW89_UK][28] = 127,
+	[1][1][2][0][RTW89_FCC][32] = 127,
+	[1][1][2][0][RTW89_ETSI][32] = 127,
+	[1][1][2][0][RTW89_MKK][32] = 127,
+	[1][1][2][0][RTW89_IC][32] = 127,
+	[1][1][2][0][RTW89_KCC][32] = 127,
+	[1][1][2][0][RTW89_ACMA][32] = 127,
+	[1][1][2][0][RTW89_CN][32] = 127,
+	[1][1][2][0][RTW89_UK][32] = 127,
+	[1][1][2][0][RTW89_FCC][36] = 127,
+	[1][1][2][0][RTW89_ETSI][36] = 127,
+	[1][1][2][0][RTW89_MKK][36] = 127,
+	[1][1][2][0][RTW89_IC][36] = 127,
+	[1][1][2][0][RTW89_KCC][36] = 127,
+	[1][1][2][0][RTW89_ACMA][36] = 127,
+	[1][1][2][0][RTW89_CN][36] = 127,
+	[1][1][2][0][RTW89_UK][36] = 127,
+	[1][1][2][0][RTW89_FCC][39] = 127,
+	[1][1][2][0][RTW89_ETSI][39] = 127,
+	[1][1][2][0][RTW89_MKK][39] = 127,
+	[1][1][2][0][RTW89_IC][39] = 127,
+	[1][1][2][0][RTW89_KCC][39] = 127,
+	[1][1][2][0][RTW89_ACMA][39] = 127,
+	[1][1][2][0][RTW89_CN][39] = 127,
+	[1][1][2][0][RTW89_UK][39] = 127,
+	[1][1][2][0][RTW89_FCC][43] = 127,
+	[1][1][2][0][RTW89_ETSI][43] = 127,
+	[1][1][2][0][RTW89_MKK][43] = 127,
+	[1][1][2][0][RTW89_IC][43] = 127,
+	[1][1][2][0][RTW89_KCC][43] = 127,
+	[1][1][2][0][RTW89_ACMA][43] = 127,
+	[1][1][2][0][RTW89_CN][43] = 127,
+	[1][1][2][0][RTW89_UK][43] = 127,
+	[1][1][2][0][RTW89_FCC][47] = 127,
+	[1][1][2][0][RTW89_ETSI][47] = 127,
+	[1][1][2][0][RTW89_MKK][47] = 127,
+	[1][1][2][0][RTW89_IC][47] = 127,
+	[1][1][2][0][RTW89_KCC][47] = 127,
+	[1][1][2][0][RTW89_ACMA][47] = 127,
+	[1][1][2][0][RTW89_CN][47] = 127,
+	[1][1][2][0][RTW89_UK][47] = 127,
+	[1][1][2][0][RTW89_FCC][51] = 127,
+	[1][1][2][0][RTW89_ETSI][51] = 127,
+	[1][1][2][0][RTW89_MKK][51] = 127,
+	[1][1][2][0][RTW89_IC][51] = 127,
+	[1][1][2][0][RTW89_KCC][51] = 127,
+	[1][1][2][0][RTW89_ACMA][51] = 127,
+	[1][1][2][0][RTW89_CN][51] = 127,
+	[1][1][2][0][RTW89_UK][51] = 127,
+	[1][1][2][1][RTW89_FCC][1] = 127,
+	[1][1][2][1][RTW89_ETSI][1] = 127,
+	[1][1][2][1][RTW89_MKK][1] = 127,
+	[1][1][2][1][RTW89_IC][1] = 127,
+	[1][1][2][1][RTW89_KCC][1] = 127,
+	[1][1][2][1][RTW89_ACMA][1] = 127,
+	[1][1][2][1][RTW89_CN][1] = 127,
+	[1][1][2][1][RTW89_UK][1] = 127,
+	[1][1][2][1][RTW89_FCC][5] = 127,
+	[1][1][2][1][RTW89_ETSI][5] = 127,
+	[1][1][2][1][RTW89_MKK][5] = 127,
+	[1][1][2][1][RTW89_IC][5] = 127,
+	[1][1][2][1][RTW89_KCC][5] = 127,
+	[1][1][2][1][RTW89_ACMA][5] = 127,
+	[1][1][2][1][RTW89_CN][5] = 127,
+	[1][1][2][1][RTW89_UK][5] = 127,
+	[1][1][2][1][RTW89_FCC][9] = 127,
+	[1][1][2][1][RTW89_ETSI][9] = 127,
+	[1][1][2][1][RTW89_MKK][9] = 127,
+	[1][1][2][1][RTW89_IC][9] = 127,
+	[1][1][2][1][RTW89_KCC][9] = 127,
+	[1][1][2][1][RTW89_ACMA][9] = 127,
+	[1][1][2][1][RTW89_CN][9] = 127,
+	[1][1][2][1][RTW89_UK][9] = 127,
+	[1][1][2][1][RTW89_FCC][13] = 127,
+	[1][1][2][1][RTW89_ETSI][13] = 127,
+	[1][1][2][1][RTW89_MKK][13] = 127,
+	[1][1][2][1][RTW89_IC][13] = 127,
+	[1][1][2][1][RTW89_KCC][13] = 127,
+	[1][1][2][1][RTW89_ACMA][13] = 127,
+	[1][1][2][1][RTW89_CN][13] = 127,
+	[1][1][2][1][RTW89_UK][13] = 127,
+	[1][1][2][1][RTW89_FCC][16] = 127,
+	[1][1][2][1][RTW89_ETSI][16] = 127,
+	[1][1][2][1][RTW89_MKK][16] = 127,
+	[1][1][2][1][RTW89_IC][16] = 127,
+	[1][1][2][1][RTW89_KCC][16] = 127,
+	[1][1][2][1][RTW89_ACMA][16] = 127,
+	[1][1][2][1][RTW89_CN][16] = 127,
+	[1][1][2][1][RTW89_UK][16] = 127,
+	[1][1][2][1][RTW89_FCC][20] = 127,
+	[1][1][2][1][RTW89_ETSI][20] = 127,
+	[1][1][2][1][RTW89_MKK][20] = 127,
+	[1][1][2][1][RTW89_IC][20] = 127,
+	[1][1][2][1][RTW89_KCC][20] = 127,
+	[1][1][2][1][RTW89_ACMA][20] = 127,
+	[1][1][2][1][RTW89_CN][20] = 127,
+	[1][1][2][1][RTW89_UK][20] = 127,
+	[1][1][2][1][RTW89_FCC][24] = 127,
+	[1][1][2][1][RTW89_ETSI][24] = 127,
+	[1][1][2][1][RTW89_MKK][24] = 127,
+	[1][1][2][1][RTW89_IC][24] = 127,
+	[1][1][2][1][RTW89_KCC][24] = 127,
+	[1][1][2][1][RTW89_ACMA][24] = 127,
+	[1][1][2][1][RTW89_CN][24] = 127,
+	[1][1][2][1][RTW89_UK][24] = 127,
+	[1][1][2][1][RTW89_FCC][28] = 127,
+	[1][1][2][1][RTW89_ETSI][28] = 127,
+	[1][1][2][1][RTW89_MKK][28] = 127,
+	[1][1][2][1][RTW89_IC][28] = 127,
+	[1][1][2][1][RTW89_KCC][28] = 127,
+	[1][1][2][1][RTW89_ACMA][28] = 127,
+	[1][1][2][1][RTW89_CN][28] = 127,
+	[1][1][2][1][RTW89_UK][28] = 127,
+	[1][1][2][1][RTW89_FCC][32] = 127,
+	[1][1][2][1][RTW89_ETSI][32] = 127,
+	[1][1][2][1][RTW89_MKK][32] = 127,
+	[1][1][2][1][RTW89_IC][32] = 127,
+	[1][1][2][1][RTW89_KCC][32] = 127,
+	[1][1][2][1][RTW89_ACMA][32] = 127,
+	[1][1][2][1][RTW89_CN][32] = 127,
+	[1][1][2][1][RTW89_UK][32] = 127,
+	[1][1][2][1][RTW89_FCC][36] = 127,
+	[1][1][2][1][RTW89_ETSI][36] = 127,
+	[1][1][2][1][RTW89_MKK][36] = 127,
+	[1][1][2][1][RTW89_IC][36] = 127,
+	[1][1][2][1][RTW89_KCC][36] = 127,
+	[1][1][2][1][RTW89_ACMA][36] = 127,
+	[1][1][2][1][RTW89_CN][36] = 127,
+	[1][1][2][1][RTW89_UK][36] = 127,
+	[1][1][2][1][RTW89_FCC][39] = 127,
+	[1][1][2][1][RTW89_ETSI][39] = 127,
+	[1][1][2][1][RTW89_MKK][39] = 127,
+	[1][1][2][1][RTW89_IC][39] = 127,
+	[1][1][2][1][RTW89_KCC][39] = 127,
+	[1][1][2][1][RTW89_ACMA][39] = 127,
+	[1][1][2][1][RTW89_CN][39] = 127,
+	[1][1][2][1][RTW89_UK][39] = 127,
+	[1][1][2][1][RTW89_FCC][43] = 127,
+	[1][1][2][1][RTW89_ETSI][43] = 127,
+	[1][1][2][1][RTW89_MKK][43] = 127,
+	[1][1][2][1][RTW89_IC][43] = 127,
+	[1][1][2][1][RTW89_KCC][43] = 127,
+	[1][1][2][1][RTW89_ACMA][43] = 127,
+	[1][1][2][1][RTW89_CN][43] = 127,
+	[1][1][2][1][RTW89_UK][43] = 127,
+	[1][1][2][1][RTW89_FCC][47] = 127,
+	[1][1][2][1][RTW89_ETSI][47] = 127,
+	[1][1][2][1][RTW89_MKK][47] = 127,
+	[1][1][2][1][RTW89_IC][47] = 127,
+	[1][1][2][1][RTW89_KCC][47] = 127,
+	[1][1][2][1][RTW89_ACMA][47] = 127,
+	[1][1][2][1][RTW89_CN][47] = 127,
+	[1][1][2][1][RTW89_UK][47] = 127,
+	[1][1][2][1][RTW89_FCC][51] = 127,
+	[1][1][2][1][RTW89_ETSI][51] = 127,
+	[1][1][2][1][RTW89_MKK][51] = 127,
+	[1][1][2][1][RTW89_IC][51] = 127,
+	[1][1][2][1][RTW89_KCC][51] = 127,
+	[1][1][2][1][RTW89_ACMA][51] = 127,
+	[1][1][2][1][RTW89_CN][51] = 127,
+	[1][1][2][1][RTW89_UK][51] = 127,
+	[2][0][2][0][RTW89_FCC][3] = 72,
+	[2][0][2][0][RTW89_ETSI][3] = 64,
+	[2][0][2][0][RTW89_MKK][3] = 62,
+	[2][0][2][0][RTW89_IC][3] = 64,
+	[2][0][2][0][RTW89_KCC][3] = 68,
+	[2][0][2][0][RTW89_ACMA][3] = 64,
+	[2][0][2][0][RTW89_CN][3] = 64,
+	[2][0][2][0][RTW89_UK][3] = 64,
+	[2][0][2][0][RTW89_FCC][11] = 62,
+	[2][0][2][0][RTW89_ETSI][11] = 64,
+	[2][0][2][0][RTW89_MKK][11] = 64,
+	[2][0][2][0][RTW89_IC][11] = 62,
+	[2][0][2][0][RTW89_KCC][11] = 68,
+	[2][0][2][0][RTW89_ACMA][11] = 64,
+	[2][0][2][0][RTW89_CN][11] = 64,
+	[2][0][2][0][RTW89_UK][11] = 64,
+	[2][0][2][0][RTW89_FCC][18] = 66,
+	[2][0][2][0][RTW89_ETSI][18] = 64,
+	[2][0][2][0][RTW89_MKK][18] = 68,
+	[2][0][2][0][RTW89_IC][18] = 66,
+	[2][0][2][0][RTW89_KCC][18] = 68,
+	[2][0][2][0][RTW89_ACMA][18] = 64,
+	[2][0][2][0][RTW89_CN][18] = 127,
+	[2][0][2][0][RTW89_UK][18] = 64,
+	[2][0][2][0][RTW89_FCC][26] = 72,
+	[2][0][2][0][RTW89_ETSI][26] = 64,
+	[2][0][2][0][RTW89_MKK][26] = 68,
+	[2][0][2][0][RTW89_IC][26] = 127,
+	[2][0][2][0][RTW89_KCC][26] = 68,
+	[2][0][2][0][RTW89_ACMA][26] = 127,
+	[2][0][2][0][RTW89_CN][26] = 127,
+	[2][0][2][0][RTW89_UK][26] = 64,
+	[2][0][2][0][RTW89_FCC][34] = 72,
+	[2][0][2][0][RTW89_ETSI][34] = 127,
+	[2][0][2][0][RTW89_MKK][34] = 68,
+	[2][0][2][0][RTW89_IC][34] = 72,
+	[2][0][2][0][RTW89_KCC][34] = 68,
+	[2][0][2][0][RTW89_ACMA][34] = 68,
+	[2][0][2][0][RTW89_CN][34] = 127,
+	[2][0][2][0][RTW89_UK][34] = 68,
+	[2][0][2][0][RTW89_FCC][41] = 72,
+	[2][0][2][0][RTW89_ETSI][41] = 30,
+	[2][0][2][0][RTW89_MKK][41] = 127,
+	[2][0][2][0][RTW89_IC][41] = 72,
+	[2][0][2][0][RTW89_KCC][41] = 64,
+	[2][0][2][0][RTW89_ACMA][41] = 68,
+	[2][0][2][0][RTW89_CN][41] = 68,
+	[2][0][2][0][RTW89_UK][41] = 64,
+	[2][0][2][0][RTW89_FCC][49] = 72,
+	[2][0][2][0][RTW89_ETSI][49] = 127,
+	[2][0][2][0][RTW89_MKK][49] = 127,
+	[2][0][2][0][RTW89_IC][49] = 127,
+	[2][0][2][0][RTW89_KCC][49] = 127,
+	[2][0][2][0][RTW89_ACMA][49] = 127,
+	[2][0][2][0][RTW89_CN][49] = 127,
+	[2][0][2][0][RTW89_UK][49] = 127,
+	[2][1][2][0][RTW89_FCC][3] = 127,
+	[2][1][2][0][RTW89_ETSI][3] = 127,
+	[2][1][2][0][RTW89_MKK][3] = 127,
+	[2][1][2][0][RTW89_IC][3] = 127,
+	[2][1][2][0][RTW89_KCC][3] = 127,
+	[2][1][2][0][RTW89_ACMA][3] = 127,
+	[2][1][2][0][RTW89_CN][3] = 127,
+	[2][1][2][0][RTW89_UK][3] = 127,
+	[2][1][2][0][RTW89_FCC][11] = 127,
+	[2][1][2][0][RTW89_ETSI][11] = 127,
+	[2][1][2][0][RTW89_MKK][11] = 127,
+	[2][1][2][0][RTW89_IC][11] = 127,
+	[2][1][2][0][RTW89_KCC][11] = 127,
+	[2][1][2][0][RTW89_ACMA][11] = 127,
+	[2][1][2][0][RTW89_CN][11] = 127,
+	[2][1][2][0][RTW89_UK][11] = 127,
+	[2][1][2][0][RTW89_FCC][18] = 127,
+	[2][1][2][0][RTW89_ETSI][18] = 127,
+	[2][1][2][0][RTW89_MKK][18] = 127,
+	[2][1][2][0][RTW89_IC][18] = 127,
+	[2][1][2][0][RTW89_KCC][18] = 127,
+	[2][1][2][0][RTW89_ACMA][18] = 127,
+	[2][1][2][0][RTW89_CN][18] = 127,
+	[2][1][2][0][RTW89_UK][18] = 127,
+	[2][1][2][0][RTW89_FCC][26] = 127,
+	[2][1][2][0][RTW89_ETSI][26] = 127,
+	[2][1][2][0][RTW89_MKK][26] = 127,
+	[2][1][2][0][RTW89_IC][26] = 127,
+	[2][1][2][0][RTW89_KCC][26] = 127,
+	[2][1][2][0][RTW89_ACMA][26] = 127,
+	[2][1][2][0][RTW89_CN][26] = 127,
+	[2][1][2][0][RTW89_UK][26] = 127,
+	[2][1][2][0][RTW89_FCC][34] = 127,
+	[2][1][2][0][RTW89_ETSI][34] = 127,
+	[2][1][2][0][RTW89_MKK][34] = 127,
+	[2][1][2][0][RTW89_IC][34] = 127,
+	[2][1][2][0][RTW89_KCC][34] = 127,
+	[2][1][2][0][RTW89_ACMA][34] = 127,
+	[2][1][2][0][RTW89_CN][34] = 127,
+	[2][1][2][0][RTW89_UK][34] = 127,
+	[2][1][2][0][RTW89_FCC][41] = 127,
+	[2][1][2][0][RTW89_ETSI][41] = 127,
+	[2][1][2][0][RTW89_MKK][41] = 127,
+	[2][1][2][0][RTW89_IC][41] = 127,
+	[2][1][2][0][RTW89_KCC][41] = 127,
+	[2][1][2][0][RTW89_ACMA][41] = 127,
+	[2][1][2][0][RTW89_CN][41] = 127,
+	[2][1][2][0][RTW89_UK][41] = 127,
+	[2][1][2][0][RTW89_FCC][49] = 127,
+	[2][1][2][0][RTW89_ETSI][49] = 127,
+	[2][1][2][0][RTW89_MKK][49] = 127,
+	[2][1][2][0][RTW89_IC][49] = 127,
+	[2][1][2][0][RTW89_KCC][49] = 127,
+	[2][1][2][0][RTW89_ACMA][49] = 127,
+	[2][1][2][0][RTW89_CN][49] = 127,
+	[2][1][2][0][RTW89_UK][49] = 127,
+	[2][1][2][1][RTW89_FCC][3] = 127,
+	[2][1][2][1][RTW89_ETSI][3] = 127,
+	[2][1][2][1][RTW89_MKK][3] = 127,
+	[2][1][2][1][RTW89_IC][3] = 127,
+	[2][1][2][1][RTW89_KCC][3] = 127,
+	[2][1][2][1][RTW89_ACMA][3] = 127,
+	[2][1][2][1][RTW89_CN][3] = 127,
+	[2][1][2][1][RTW89_UK][3] = 127,
+	[2][1][2][1][RTW89_FCC][11] = 127,
+	[2][1][2][1][RTW89_ETSI][11] = 127,
+	[2][1][2][1][RTW89_MKK][11] = 127,
+	[2][1][2][1][RTW89_IC][11] = 127,
+	[2][1][2][1][RTW89_KCC][11] = 127,
+	[2][1][2][1][RTW89_ACMA][11] = 127,
+	[2][1][2][1][RTW89_CN][11] = 127,
+	[2][1][2][1][RTW89_UK][11] = 127,
+	[2][1][2][1][RTW89_FCC][18] = 127,
+	[2][1][2][1][RTW89_ETSI][18] = 127,
+	[2][1][2][1][RTW89_MKK][18] = 127,
+	[2][1][2][1][RTW89_IC][18] = 127,
+	[2][1][2][1][RTW89_KCC][18] = 127,
+	[2][1][2][1][RTW89_ACMA][18] = 127,
+	[2][1][2][1][RTW89_CN][18] = 127,
+	[2][1][2][1][RTW89_UK][18] = 127,
+	[2][1][2][1][RTW89_FCC][26] = 127,
+	[2][1][2][1][RTW89_ETSI][26] = 127,
+	[2][1][2][1][RTW89_MKK][26] = 127,
+	[2][1][2][1][RTW89_IC][26] = 127,
+	[2][1][2][1][RTW89_KCC][26] = 127,
+	[2][1][2][1][RTW89_ACMA][26] = 127,
+	[2][1][2][1][RTW89_CN][26] = 127,
+	[2][1][2][1][RTW89_UK][26] = 127,
+	[2][1][2][1][RTW89_FCC][34] = 127,
+	[2][1][2][1][RTW89_ETSI][34] = 127,
+	[2][1][2][1][RTW89_MKK][34] = 127,
+	[2][1][2][1][RTW89_IC][34] = 127,
+	[2][1][2][1][RTW89_KCC][34] = 127,
+	[2][1][2][1][RTW89_ACMA][34] = 127,
+	[2][1][2][1][RTW89_CN][34] = 127,
+	[2][1][2][1][RTW89_UK][34] = 127,
+	[2][1][2][1][RTW89_FCC][41] = 127,
+	[2][1][2][1][RTW89_ETSI][41] = 127,
+	[2][1][2][1][RTW89_MKK][41] = 127,
+	[2][1][2][1][RTW89_IC][41] = 127,
+	[2][1][2][1][RTW89_KCC][41] = 127,
+	[2][1][2][1][RTW89_ACMA][41] = 127,
+	[2][1][2][1][RTW89_CN][41] = 127,
+	[2][1][2][1][RTW89_UK][41] = 127,
+	[2][1][2][1][RTW89_FCC][49] = 127,
+	[2][1][2][1][RTW89_ETSI][49] = 127,
+	[2][1][2][1][RTW89_MKK][49] = 127,
+	[2][1][2][1][RTW89_IC][49] = 127,
+	[2][1][2][1][RTW89_KCC][49] = 127,
+	[2][1][2][1][RTW89_ACMA][49] = 127,
+	[2][1][2][1][RTW89_CN][49] = 127,
+	[2][1][2][1][RTW89_UK][49] = 127,
+	[3][0][2][0][RTW89_FCC][7] = 127,
+	[3][0][2][0][RTW89_ETSI][7] = 127,
+	[3][0][2][0][RTW89_MKK][7] = 127,
+	[3][0][2][0][RTW89_IC][7] = 127,
+	[3][0][2][0][RTW89_KCC][7] = 127,
+	[3][0][2][0][RTW89_ACMA][7] = 127,
+	[3][0][2][0][RTW89_CN][7] = 58,
+	[3][0][2][0][RTW89_UK][7] = 127,
+	[3][0][2][0][RTW89_FCC][22] = 127,
+	[3][0][2][0][RTW89_ETSI][22] = 127,
+	[3][0][2][0][RTW89_MKK][22] = 127,
+	[3][0][2][0][RTW89_IC][22] = 127,
+	[3][0][2][0][RTW89_KCC][22] = 127,
+	[3][0][2][0][RTW89_ACMA][22] = 127,
+	[3][0][2][0][RTW89_CN][22] = 58,
+	[3][0][2][0][RTW89_UK][22] = 127,
+	[3][0][2][0][RTW89_FCC][45] = 127,
+	[3][0][2][0][RTW89_ETSI][45] = 127,
+	[3][0][2][0][RTW89_MKK][45] = 127,
+	[3][0][2][0][RTW89_IC][45] = 127,
+	[3][0][2][0][RTW89_KCC][45] = 127,
+	[3][0][2][0][RTW89_ACMA][45] = 127,
+	[3][0][2][0][RTW89_CN][45] = 127,
+	[3][0][2][0][RTW89_UK][45] = 127,
+	[3][1][2][0][RTW89_FCC][7] = 127,
+	[3][1][2][0][RTW89_ETSI][7] = 127,
+	[3][1][2][0][RTW89_MKK][7] = 127,
+	[3][1][2][0][RTW89_IC][7] = 127,
+	[3][1][2][0][RTW89_KCC][7] = 127,
+	[3][1][2][0][RTW89_ACMA][7] = 127,
+	[3][1][2][0][RTW89_CN][7] = 127,
+	[3][1][2][0][RTW89_UK][7] = 127,
+	[3][1][2][0][RTW89_FCC][22] = 127,
+	[3][1][2][0][RTW89_ETSI][22] = 127,
+	[3][1][2][0][RTW89_MKK][22] = 127,
+	[3][1][2][0][RTW89_IC][22] = 127,
+	[3][1][2][0][RTW89_KCC][22] = 127,
+	[3][1][2][0][RTW89_ACMA][22] = 127,
+	[3][1][2][0][RTW89_CN][22] = 127,
+	[3][1][2][0][RTW89_UK][22] = 127,
+	[3][1][2][0][RTW89_FCC][45] = 127,
+	[3][1][2][0][RTW89_ETSI][45] = 127,
+	[3][1][2][0][RTW89_MKK][45] = 127,
+	[3][1][2][0][RTW89_IC][45] = 127,
+	[3][1][2][0][RTW89_KCC][45] = 127,
+	[3][1][2][0][RTW89_ACMA][45] = 127,
+	[3][1][2][0][RTW89_CN][45] = 127,
+	[3][1][2][0][RTW89_UK][45] = 127,
+	[3][1][2][1][RTW89_FCC][7] = 127,
+	[3][1][2][1][RTW89_ETSI][7] = 127,
+	[3][1][2][1][RTW89_MKK][7] = 127,
+	[3][1][2][1][RTW89_IC][7] = 127,
+	[3][1][2][1][RTW89_KCC][7] = 127,
+	[3][1][2][1][RTW89_ACMA][7] = 127,
+	[3][1][2][1][RTW89_CN][7] = 127,
+	[3][1][2][1][RTW89_UK][7] = 127,
+	[3][1][2][1][RTW89_FCC][22] = 127,
+	[3][1][2][1][RTW89_ETSI][22] = 127,
+	[3][1][2][1][RTW89_MKK][22] = 127,
+	[3][1][2][1][RTW89_IC][22] = 127,
+	[3][1][2][1][RTW89_KCC][22] = 127,
+	[3][1][2][1][RTW89_ACMA][22] = 127,
+	[3][1][2][1][RTW89_CN][22] = 127,
+	[3][1][2][1][RTW89_UK][22] = 127,
+	[3][1][2][1][RTW89_FCC][45] = 127,
+	[3][1][2][1][RTW89_ETSI][45] = 127,
+	[3][1][2][1][RTW89_MKK][45] = 127,
+	[3][1][2][1][RTW89_IC][45] = 127,
+	[3][1][2][1][RTW89_KCC][45] = 127,
+	[3][1][2][1][RTW89_ACMA][45] = 127,
+	[3][1][2][1][RTW89_CN][45] = 127,
+	[3][1][2][1][RTW89_UK][45] = 127,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_ru_2g_type2[RTW89_RU_NUM][RTW89_NTX_NUM]
+					  [RTW89_REGD_NUM][RTW89_2G_CH_NUM] = {
+	[0][0][RTW89_WW][0] = 30,
+	[0][0][RTW89_WW][1] = 30,
+	[0][0][RTW89_WW][2] = 30,
+	[0][0][RTW89_WW][3] = 30,
+	[0][0][RTW89_WW][4] = 30,
+	[0][0][RTW89_WW][5] = 30,
+	[0][0][RTW89_WW][6] = 30,
+	[0][0][RTW89_WW][7] = 30,
+	[0][0][RTW89_WW][8] = 30,
+	[0][0][RTW89_WW][9] = 30,
+	[0][0][RTW89_WW][10] = 30,
+	[0][0][RTW89_WW][11] = 30,
+	[0][0][RTW89_WW][12] = 30,
+	[0][0][RTW89_WW][13] = 0,
+	[0][1][RTW89_WW][0] = 20,
+	[0][1][RTW89_WW][1] = 22,
+	[0][1][RTW89_WW][2] = 22,
+	[0][1][RTW89_WW][3] = 22,
+	[0][1][RTW89_WW][4] = 22,
+	[0][1][RTW89_WW][5] = 22,
+	[0][1][RTW89_WW][6] = 22,
+	[0][1][RTW89_WW][7] = 22,
+	[0][1][RTW89_WW][8] = 22,
+	[0][1][RTW89_WW][9] = 22,
+	[0][1][RTW89_WW][10] = 22,
+	[0][1][RTW89_WW][11] = 22,
+	[0][1][RTW89_WW][12] = 20,
+	[0][1][RTW89_WW][13] = 0,
+	[1][0][RTW89_WW][0] = 42,
+	[1][0][RTW89_WW][1] = 42,
+	[1][0][RTW89_WW][2] = 42,
+	[1][0][RTW89_WW][3] = 42,
+	[1][0][RTW89_WW][4] = 42,
+	[1][0][RTW89_WW][5] = 42,
+	[1][0][RTW89_WW][6] = 42,
+	[1][0][RTW89_WW][7] = 42,
+	[1][0][RTW89_WW][8] = 42,
+	[1][0][RTW89_WW][9] = 42,
+	[1][0][RTW89_WW][10] = 42,
+	[1][0][RTW89_WW][11] = 42,
+	[1][0][RTW89_WW][12] = 34,
+	[1][0][RTW89_WW][13] = 0,
+	[1][1][RTW89_WW][0] = 32,
+	[1][1][RTW89_WW][1] = 32,
+	[1][1][RTW89_WW][2] = 32,
+	[1][1][RTW89_WW][3] = 32,
+	[1][1][RTW89_WW][4] = 32,
+	[1][1][RTW89_WW][5] = 32,
+	[1][1][RTW89_WW][6] = 32,
+	[1][1][RTW89_WW][7] = 32,
+	[1][1][RTW89_WW][8] = 32,
+	[1][1][RTW89_WW][9] = 32,
+	[1][1][RTW89_WW][10] = 32,
+	[1][1][RTW89_WW][11] = 32,
+	[1][1][RTW89_WW][12] = 32,
+	[1][1][RTW89_WW][13] = 0,
+	[2][0][RTW89_WW][0] = 54,
+	[2][0][RTW89_WW][1] = 54,
+	[2][0][RTW89_WW][2] = 54,
+	[2][0][RTW89_WW][3] = 54,
+	[2][0][RTW89_WW][4] = 54,
+	[2][0][RTW89_WW][5] = 54,
+	[2][0][RTW89_WW][6] = 54,
+	[2][0][RTW89_WW][7] = 54,
+	[2][0][RTW89_WW][8] = 54,
+	[2][0][RTW89_WW][9] = 54,
+	[2][0][RTW89_WW][10] = 54,
+	[2][0][RTW89_WW][11] = 54,
+	[2][0][RTW89_WW][12] = 34,
+	[2][0][RTW89_WW][13] = 0,
+	[2][1][RTW89_WW][0] = 44,
+	[2][1][RTW89_WW][1] = 44,
+	[2][1][RTW89_WW][2] = 44,
+	[2][1][RTW89_WW][3] = 44,
+	[2][1][RTW89_WW][4] = 44,
+	[2][1][RTW89_WW][5] = 44,
+	[2][1][RTW89_WW][6] = 44,
+	[2][1][RTW89_WW][7] = 44,
+	[2][1][RTW89_WW][8] = 44,
+	[2][1][RTW89_WW][9] = 44,
+	[2][1][RTW89_WW][10] = 44,
+	[2][1][RTW89_WW][11] = 44,
+	[2][1][RTW89_WW][12] = 42,
+	[2][1][RTW89_WW][13] = 0,
+	[0][0][RTW89_FCC][0] = 60,
+	[0][0][RTW89_ETSI][0] = 30,
+	[0][0][RTW89_MKK][0] = 40,
+	[0][0][RTW89_IC][0] = 60,
+	[0][0][RTW89_KCC][0] = 46,
+	[0][0][RTW89_ACMA][0] = 30,
+	[0][0][RTW89_CN][0] = 32,
+	[0][0][RTW89_UK][0] = 30,
+	[0][0][RTW89_FCC][1] = 60,
+	[0][0][RTW89_ETSI][1] = 30,
+	[0][0][RTW89_MKK][1] = 44,
+	[0][0][RTW89_IC][1] = 60,
+	[0][0][RTW89_KCC][1] = 46,
+	[0][0][RTW89_ACMA][1] = 30,
+	[0][0][RTW89_CN][1] = 32,
+	[0][0][RTW89_UK][1] = 30,
+	[0][0][RTW89_FCC][2] = 64,
+	[0][0][RTW89_ETSI][2] = 30,
+	[0][0][RTW89_MKK][2] = 44,
+	[0][0][RTW89_IC][2] = 64,
+	[0][0][RTW89_KCC][2] = 46,
+	[0][0][RTW89_ACMA][2] = 30,
+	[0][0][RTW89_CN][2] = 32,
+	[0][0][RTW89_UK][2] = 30,
+	[0][0][RTW89_FCC][3] = 68,
+	[0][0][RTW89_ETSI][3] = 30,
+	[0][0][RTW89_MKK][3] = 44,
+	[0][0][RTW89_IC][3] = 68,
+	[0][0][RTW89_KCC][3] = 46,
+	[0][0][RTW89_ACMA][3] = 30,
+	[0][0][RTW89_CN][3] = 32,
+	[0][0][RTW89_UK][3] = 30,
+	[0][0][RTW89_FCC][4] = 68,
+	[0][0][RTW89_ETSI][4] = 30,
+	[0][0][RTW89_MKK][4] = 44,
+	[0][0][RTW89_IC][4] = 68,
+	[0][0][RTW89_KCC][4] = 48,
+	[0][0][RTW89_ACMA][4] = 30,
+	[0][0][RTW89_CN][4] = 32,
+	[0][0][RTW89_UK][4] = 30,
+	[0][0][RTW89_FCC][5] = 82,
+	[0][0][RTW89_ETSI][5] = 30,
+	[0][0][RTW89_MKK][5] = 44,
+	[0][0][RTW89_IC][5] = 82,
+	[0][0][RTW89_KCC][5] = 48,
+	[0][0][RTW89_ACMA][5] = 30,
+	[0][0][RTW89_CN][5] = 32,
+	[0][0][RTW89_UK][5] = 30,
+	[0][0][RTW89_FCC][6] = 64,
+	[0][0][RTW89_ETSI][6] = 30,
+	[0][0][RTW89_MKK][6] = 44,
+	[0][0][RTW89_IC][6] = 64,
+	[0][0][RTW89_KCC][6] = 48,
+	[0][0][RTW89_ACMA][6] = 30,
+	[0][0][RTW89_CN][6] = 32,
+	[0][0][RTW89_UK][6] = 30,
+	[0][0][RTW89_FCC][7] = 64,
+	[0][0][RTW89_ETSI][7] = 30,
+	[0][0][RTW89_MKK][7] = 44,
+	[0][0][RTW89_IC][7] = 64,
+	[0][0][RTW89_KCC][7] = 48,
+	[0][0][RTW89_ACMA][7] = 30,
+	[0][0][RTW89_CN][7] = 32,
+	[0][0][RTW89_UK][7] = 30,
+	[0][0][RTW89_FCC][8] = 60,
+	[0][0][RTW89_ETSI][8] = 30,
+	[0][0][RTW89_MKK][8] = 44,
+	[0][0][RTW89_IC][8] = 60,
+	[0][0][RTW89_KCC][8] = 48,
+	[0][0][RTW89_ACMA][8] = 30,
+	[0][0][RTW89_CN][8] = 32,
+	[0][0][RTW89_UK][8] = 30,
+	[0][0][RTW89_FCC][9] = 56,
+	[0][0][RTW89_ETSI][9] = 30,
+	[0][0][RTW89_MKK][9] = 44,
+	[0][0][RTW89_IC][9] = 56,
+	[0][0][RTW89_KCC][9] = 44,
+	[0][0][RTW89_ACMA][9] = 30,
+	[0][0][RTW89_CN][9] = 32,
+	[0][0][RTW89_UK][9] = 30,
+	[0][0][RTW89_FCC][10] = 56,
+	[0][0][RTW89_ETSI][10] = 30,
+	[0][0][RTW89_MKK][10] = 44,
+	[0][0][RTW89_IC][10] = 56,
+	[0][0][RTW89_KCC][10] = 44,
+	[0][0][RTW89_ACMA][10] = 30,
+	[0][0][RTW89_CN][10] = 32,
+	[0][0][RTW89_UK][10] = 30,
+	[0][0][RTW89_FCC][11] = 54,
+	[0][0][RTW89_ETSI][11] = 30,
+	[0][0][RTW89_MKK][11] = 44,
+	[0][0][RTW89_IC][11] = 54,
+	[0][0][RTW89_KCC][11] = 44,
+	[0][0][RTW89_ACMA][11] = 30,
+	[0][0][RTW89_CN][11] = 32,
+	[0][0][RTW89_UK][11] = 30,
+	[0][0][RTW89_FCC][12] = 34,
+	[0][0][RTW89_ETSI][12] = 30,
+	[0][0][RTW89_MKK][12] = 40,
+	[0][0][RTW89_IC][12] = 34,
+	[0][0][RTW89_KCC][12] = 44,
+	[0][0][RTW89_ACMA][12] = 30,
+	[0][0][RTW89_CN][12] = 32,
+	[0][0][RTW89_UK][12] = 30,
+	[0][0][RTW89_FCC][13] = 127,
+	[0][0][RTW89_ETSI][13] = 127,
+	[0][0][RTW89_MKK][13] = 127,
+	[0][0][RTW89_IC][13] = 127,
+	[0][0][RTW89_KCC][13] = 127,
+	[0][0][RTW89_ACMA][13] = 127,
+	[0][0][RTW89_CN][13] = 127,
+	[0][0][RTW89_UK][13] = 127,
+	[0][1][RTW89_FCC][0] = 127,
+	[0][1][RTW89_ETSI][0] = 127,
+	[0][1][RTW89_MKK][0] = 127,
+	[0][1][RTW89_IC][0] = 127,
+	[0][1][RTW89_KCC][0] = 127,
+	[0][1][RTW89_ACMA][0] = 127,
+	[0][1][RTW89_CN][0] = 20,
+	[0][1][RTW89_UK][0] = 127,
+	[0][1][RTW89_FCC][1] = 127,
+	[0][1][RTW89_ETSI][1] = 127,
+	[0][1][RTW89_MKK][1] = 127,
+	[0][1][RTW89_IC][1] = 127,
+	[0][1][RTW89_KCC][1] = 127,
+	[0][1][RTW89_ACMA][1] = 127,
+	[0][1][RTW89_CN][1] = 22,
+	[0][1][RTW89_UK][1] = 127,
+	[0][1][RTW89_FCC][2] = 127,
+	[0][1][RTW89_ETSI][2] = 127,
+	[0][1][RTW89_MKK][2] = 127,
+	[0][1][RTW89_IC][2] = 127,
+	[0][1][RTW89_KCC][2] = 127,
+	[0][1][RTW89_ACMA][2] = 127,
+	[0][1][RTW89_CN][2] = 22,
+	[0][1][RTW89_UK][2] = 127,
+	[0][1][RTW89_FCC][3] = 127,
+	[0][1][RTW89_ETSI][3] = 127,
+	[0][1][RTW89_MKK][3] = 127,
+	[0][1][RTW89_IC][3] = 127,
+	[0][1][RTW89_KCC][3] = 127,
+	[0][1][RTW89_ACMA][3] = 127,
+	[0][1][RTW89_CN][3] = 22,
+	[0][1][RTW89_UK][3] = 127,
+	[0][1][RTW89_FCC][4] = 127,
+	[0][1][RTW89_ETSI][4] = 127,
+	[0][1][RTW89_MKK][4] = 127,
+	[0][1][RTW89_IC][4] = 127,
+	[0][1][RTW89_KCC][4] = 127,
+	[0][1][RTW89_ACMA][4] = 127,
+	[0][1][RTW89_CN][4] = 22,
+	[0][1][RTW89_UK][4] = 127,
+	[0][1][RTW89_FCC][5] = 127,
+	[0][1][RTW89_ETSI][5] = 127,
+	[0][1][RTW89_MKK][5] = 127,
+	[0][1][RTW89_IC][5] = 127,
+	[0][1][RTW89_KCC][5] = 127,
+	[0][1][RTW89_ACMA][5] = 127,
+	[0][1][RTW89_CN][5] = 22,
+	[0][1][RTW89_UK][5] = 127,
+	[0][1][RTW89_FCC][6] = 127,
+	[0][1][RTW89_ETSI][6] = 127,
+	[0][1][RTW89_MKK][6] = 127,
+	[0][1][RTW89_IC][6] = 127,
+	[0][1][RTW89_KCC][6] = 127,
+	[0][1][RTW89_ACMA][6] = 127,
+	[0][1][RTW89_CN][6] = 22,
+	[0][1][RTW89_UK][6] = 127,
+	[0][1][RTW89_FCC][7] = 127,
+	[0][1][RTW89_ETSI][7] = 127,
+	[0][1][RTW89_MKK][7] = 127,
+	[0][1][RTW89_IC][7] = 127,
+	[0][1][RTW89_KCC][7] = 127,
+	[0][1][RTW89_ACMA][7] = 127,
+	[0][1][RTW89_CN][7] = 22,
+	[0][1][RTW89_UK][7] = 127,
+	[0][1][RTW89_FCC][8] = 127,
+	[0][1][RTW89_ETSI][8] = 127,
+	[0][1][RTW89_MKK][8] = 127,
+	[0][1][RTW89_IC][8] = 127,
+	[0][1][RTW89_KCC][8] = 127,
+	[0][1][RTW89_ACMA][8] = 127,
+	[0][1][RTW89_CN][8] = 22,
+	[0][1][RTW89_UK][8] = 127,
+	[0][1][RTW89_FCC][9] = 127,
+	[0][1][RTW89_ETSI][9] = 127,
+	[0][1][RTW89_MKK][9] = 127,
+	[0][1][RTW89_IC][9] = 127,
+	[0][1][RTW89_KCC][9] = 127,
+	[0][1][RTW89_ACMA][9] = 127,
+	[0][1][RTW89_CN][9] = 22,
+	[0][1][RTW89_UK][9] = 127,
+	[0][1][RTW89_FCC][10] = 127,
+	[0][1][RTW89_ETSI][10] = 127,
+	[0][1][RTW89_MKK][10] = 127,
+	[0][1][RTW89_IC][10] = 127,
+	[0][1][RTW89_KCC][10] = 127,
+	[0][1][RTW89_ACMA][10] = 127,
+	[0][1][RTW89_CN][10] = 22,
+	[0][1][RTW89_UK][10] = 127,
+	[0][1][RTW89_FCC][11] = 127,
+	[0][1][RTW89_ETSI][11] = 127,
+	[0][1][RTW89_MKK][11] = 127,
+	[0][1][RTW89_IC][11] = 127,
+	[0][1][RTW89_KCC][11] = 127,
+	[0][1][RTW89_ACMA][11] = 127,
+	[0][1][RTW89_CN][11] = 22,
+	[0][1][RTW89_UK][11] = 127,
+	[0][1][RTW89_FCC][12] = 127,
+	[0][1][RTW89_ETSI][12] = 127,
+	[0][1][RTW89_MKK][12] = 127,
+	[0][1][RTW89_IC][12] = 127,
+	[0][1][RTW89_KCC][12] = 127,
+	[0][1][RTW89_ACMA][12] = 127,
+	[0][1][RTW89_CN][12] = 20,
+	[0][1][RTW89_UK][12] = 127,
+	[0][1][RTW89_FCC][13] = 127,
+	[0][1][RTW89_ETSI][13] = 127,
+	[0][1][RTW89_MKK][13] = 127,
+	[0][1][RTW89_IC][13] = 127,
+	[0][1][RTW89_KCC][13] = 127,
+	[0][1][RTW89_ACMA][13] = 127,
+	[0][1][RTW89_CN][13] = 127,
+	[0][1][RTW89_UK][13] = 127,
+	[1][0][RTW89_FCC][0] = 70,
+	[1][0][RTW89_ETSI][0] = 42,
+	[1][0][RTW89_MKK][0] = 52,
+	[1][0][RTW89_IC][0] = 70,
+	[1][0][RTW89_KCC][0] = 56,
+	[1][0][RTW89_ACMA][0] = 42,
+	[1][0][RTW89_CN][0] = 42,
+	[1][0][RTW89_UK][0] = 42,
+	[1][0][RTW89_FCC][1] = 70,
+	[1][0][RTW89_ETSI][1] = 42,
+	[1][0][RTW89_MKK][1] = 52,
+	[1][0][RTW89_IC][1] = 70,
+	[1][0][RTW89_KCC][1] = 56,
+	[1][0][RTW89_ACMA][1] = 42,
+	[1][0][RTW89_CN][1] = 44,
+	[1][0][RTW89_UK][1] = 42,
+	[1][0][RTW89_FCC][2] = 74,
+	[1][0][RTW89_ETSI][2] = 42,
+	[1][0][RTW89_MKK][2] = 52,
+	[1][0][RTW89_IC][2] = 74,
+	[1][0][RTW89_KCC][2] = 56,
+	[1][0][RTW89_ACMA][2] = 42,
+	[1][0][RTW89_CN][2] = 44,
+	[1][0][RTW89_UK][2] = 42,
+	[1][0][RTW89_FCC][3] = 76,
+	[1][0][RTW89_ETSI][3] = 42,
+	[1][0][RTW89_MKK][3] = 52,
+	[1][0][RTW89_IC][3] = 76,
+	[1][0][RTW89_KCC][3] = 56,
+	[1][0][RTW89_ACMA][3] = 42,
+	[1][0][RTW89_CN][3] = 44,
+	[1][0][RTW89_UK][3] = 42,
+	[1][0][RTW89_FCC][4] = 76,
+	[1][0][RTW89_ETSI][4] = 42,
+	[1][0][RTW89_MKK][4] = 52,
+	[1][0][RTW89_IC][4] = 76,
+	[1][0][RTW89_KCC][4] = 56,
+	[1][0][RTW89_ACMA][4] = 42,
+	[1][0][RTW89_CN][4] = 44,
+	[1][0][RTW89_UK][4] = 42,
+	[1][0][RTW89_FCC][5] = 82,
+	[1][0][RTW89_ETSI][5] = 42,
+	[1][0][RTW89_MKK][5] = 52,
+	[1][0][RTW89_IC][5] = 82,
+	[1][0][RTW89_KCC][5] = 56,
+	[1][0][RTW89_ACMA][5] = 42,
+	[1][0][RTW89_CN][5] = 44,
+	[1][0][RTW89_UK][5] = 42,
+	[1][0][RTW89_FCC][6] = 72,
+	[1][0][RTW89_ETSI][6] = 42,
+	[1][0][RTW89_MKK][6] = 52,
+	[1][0][RTW89_IC][6] = 72,
+	[1][0][RTW89_KCC][6] = 56,
+	[1][0][RTW89_ACMA][6] = 42,
+	[1][0][RTW89_CN][6] = 44,
+	[1][0][RTW89_UK][6] = 42,
+	[1][0][RTW89_FCC][7] = 72,
+	[1][0][RTW89_ETSI][7] = 42,
+	[1][0][RTW89_MKK][7] = 52,
+	[1][0][RTW89_IC][7] = 72,
+	[1][0][RTW89_KCC][7] = 56,
+	[1][0][RTW89_ACMA][7] = 42,
+	[1][0][RTW89_CN][7] = 44,
+	[1][0][RTW89_UK][7] = 42,
+	[1][0][RTW89_FCC][8] = 72,
+	[1][0][RTW89_ETSI][8] = 42,
+	[1][0][RTW89_MKK][8] = 52,
+	[1][0][RTW89_IC][8] = 72,
+	[1][0][RTW89_KCC][8] = 56,
+	[1][0][RTW89_ACMA][8] = 42,
+	[1][0][RTW89_CN][8] = 44,
+	[1][0][RTW89_UK][8] = 42,
+	[1][0][RTW89_FCC][9] = 68,
+	[1][0][RTW89_ETSI][9] = 42,
+	[1][0][RTW89_MKK][9] = 52,
+	[1][0][RTW89_IC][9] = 68,
+	[1][0][RTW89_KCC][9] = 58,
+	[1][0][RTW89_ACMA][9] = 42,
+	[1][0][RTW89_CN][9] = 44,
+	[1][0][RTW89_UK][9] = 42,
+	[1][0][RTW89_FCC][10] = 68,
+	[1][0][RTW89_ETSI][10] = 42,
+	[1][0][RTW89_MKK][10] = 52,
+	[1][0][RTW89_IC][10] = 68,
+	[1][0][RTW89_KCC][10] = 58,
+	[1][0][RTW89_ACMA][10] = 42,
+	[1][0][RTW89_CN][10] = 44,
+	[1][0][RTW89_UK][10] = 42,
+	[1][0][RTW89_FCC][11] = 66,
+	[1][0][RTW89_ETSI][11] = 42,
+	[1][0][RTW89_MKK][11] = 52,
+	[1][0][RTW89_IC][11] = 66,
+	[1][0][RTW89_KCC][11] = 58,
+	[1][0][RTW89_ACMA][11] = 42,
+	[1][0][RTW89_CN][11] = 44,
+	[1][0][RTW89_UK][11] = 42,
+	[1][0][RTW89_FCC][12] = 34,
+	[1][0][RTW89_ETSI][12] = 42,
+	[1][0][RTW89_MKK][12] = 52,
+	[1][0][RTW89_IC][12] = 34,
+	[1][0][RTW89_KCC][12] = 58,
+	[1][0][RTW89_ACMA][12] = 42,
+	[1][0][RTW89_CN][12] = 42,
+	[1][0][RTW89_UK][12] = 42,
+	[1][0][RTW89_FCC][13] = 127,
+	[1][0][RTW89_ETSI][13] = 127,
+	[1][0][RTW89_MKK][13] = 127,
+	[1][0][RTW89_IC][13] = 127,
+	[1][0][RTW89_KCC][13] = 127,
+	[1][0][RTW89_ACMA][13] = 127,
+	[1][0][RTW89_CN][13] = 127,
+	[1][0][RTW89_UK][13] = 127,
+	[1][1][RTW89_FCC][0] = 127,
+	[1][1][RTW89_ETSI][0] = 127,
+	[1][1][RTW89_MKK][0] = 127,
+	[1][1][RTW89_IC][0] = 127,
+	[1][1][RTW89_KCC][0] = 127,
+	[1][1][RTW89_ACMA][0] = 127,
+	[1][1][RTW89_CN][0] = 32,
+	[1][1][RTW89_UK][0] = 127,
+	[1][1][RTW89_FCC][1] = 127,
+	[1][1][RTW89_ETSI][1] = 127,
+	[1][1][RTW89_MKK][1] = 127,
+	[1][1][RTW89_IC][1] = 127,
+	[1][1][RTW89_KCC][1] = 127,
+	[1][1][RTW89_ACMA][1] = 127,
+	[1][1][RTW89_CN][1] = 32,
+	[1][1][RTW89_UK][1] = 127,
+	[1][1][RTW89_FCC][2] = 127,
+	[1][1][RTW89_ETSI][2] = 127,
+	[1][1][RTW89_MKK][2] = 127,
+	[1][1][RTW89_IC][2] = 127,
+	[1][1][RTW89_KCC][2] = 127,
+	[1][1][RTW89_ACMA][2] = 127,
+	[1][1][RTW89_CN][2] = 32,
+	[1][1][RTW89_UK][2] = 127,
+	[1][1][RTW89_FCC][3] = 127,
+	[1][1][RTW89_ETSI][3] = 127,
+	[1][1][RTW89_MKK][3] = 127,
+	[1][1][RTW89_IC][3] = 127,
+	[1][1][RTW89_KCC][3] = 127,
+	[1][1][RTW89_ACMA][3] = 127,
+	[1][1][RTW89_CN][3] = 32,
+	[1][1][RTW89_UK][3] = 127,
+	[1][1][RTW89_FCC][4] = 127,
+	[1][1][RTW89_ETSI][4] = 127,
+	[1][1][RTW89_MKK][4] = 127,
+	[1][1][RTW89_IC][4] = 127,
+	[1][1][RTW89_KCC][4] = 127,
+	[1][1][RTW89_ACMA][4] = 127,
+	[1][1][RTW89_CN][4] = 32,
+	[1][1][RTW89_UK][4] = 127,
+	[1][1][RTW89_FCC][5] = 127,
+	[1][1][RTW89_ETSI][5] = 127,
+	[1][1][RTW89_MKK][5] = 127,
+	[1][1][RTW89_IC][5] = 127,
+	[1][1][RTW89_KCC][5] = 127,
+	[1][1][RTW89_ACMA][5] = 127,
+	[1][1][RTW89_CN][5] = 32,
+	[1][1][RTW89_UK][5] = 127,
+	[1][1][RTW89_FCC][6] = 127,
+	[1][1][RTW89_ETSI][6] = 127,
+	[1][1][RTW89_MKK][6] = 127,
+	[1][1][RTW89_IC][6] = 127,
+	[1][1][RTW89_KCC][6] = 127,
+	[1][1][RTW89_ACMA][6] = 127,
+	[1][1][RTW89_CN][6] = 32,
+	[1][1][RTW89_UK][6] = 127,
+	[1][1][RTW89_FCC][7] = 127,
+	[1][1][RTW89_ETSI][7] = 127,
+	[1][1][RTW89_MKK][7] = 127,
+	[1][1][RTW89_IC][7] = 127,
+	[1][1][RTW89_KCC][7] = 127,
+	[1][1][RTW89_ACMA][7] = 127,
+	[1][1][RTW89_CN][7] = 32,
+	[1][1][RTW89_UK][7] = 127,
+	[1][1][RTW89_FCC][8] = 127,
+	[1][1][RTW89_ETSI][8] = 127,
+	[1][1][RTW89_MKK][8] = 127,
+	[1][1][RTW89_IC][8] = 127,
+	[1][1][RTW89_KCC][8] = 127,
+	[1][1][RTW89_ACMA][8] = 127,
+	[1][1][RTW89_CN][8] = 32,
+	[1][1][RTW89_UK][8] = 127,
+	[1][1][RTW89_FCC][9] = 127,
+	[1][1][RTW89_ETSI][9] = 127,
+	[1][1][RTW89_MKK][9] = 127,
+	[1][1][RTW89_IC][9] = 127,
+	[1][1][RTW89_KCC][9] = 127,
+	[1][1][RTW89_ACMA][9] = 127,
+	[1][1][RTW89_CN][9] = 32,
+	[1][1][RTW89_UK][9] = 127,
+	[1][1][RTW89_FCC][10] = 127,
+	[1][1][RTW89_ETSI][10] = 127,
+	[1][1][RTW89_MKK][10] = 127,
+	[1][1][RTW89_IC][10] = 127,
+	[1][1][RTW89_KCC][10] = 127,
+	[1][1][RTW89_ACMA][10] = 127,
+	[1][1][RTW89_CN][10] = 32,
+	[1][1][RTW89_UK][10] = 127,
+	[1][1][RTW89_FCC][11] = 127,
+	[1][1][RTW89_ETSI][11] = 127,
+	[1][1][RTW89_MKK][11] = 127,
+	[1][1][RTW89_IC][11] = 127,
+	[1][1][RTW89_KCC][11] = 127,
+	[1][1][RTW89_ACMA][11] = 127,
+	[1][1][RTW89_CN][11] = 32,
+	[1][1][RTW89_UK][11] = 127,
+	[1][1][RTW89_FCC][12] = 127,
+	[1][1][RTW89_ETSI][12] = 127,
+	[1][1][RTW89_MKK][12] = 127,
+	[1][1][RTW89_IC][12] = 127,
+	[1][1][RTW89_KCC][12] = 127,
+	[1][1][RTW89_ACMA][12] = 127,
+	[1][1][RTW89_CN][12] = 32,
+	[1][1][RTW89_UK][12] = 127,
+	[1][1][RTW89_FCC][13] = 127,
+	[1][1][RTW89_ETSI][13] = 127,
+	[1][1][RTW89_MKK][13] = 127,
+	[1][1][RTW89_IC][13] = 127,
+	[1][1][RTW89_KCC][13] = 127,
+	[1][1][RTW89_ACMA][13] = 127,
+	[1][1][RTW89_CN][13] = 127,
+	[1][1][RTW89_UK][13] = 127,
+	[2][0][RTW89_FCC][0] = 74,
+	[2][0][RTW89_ETSI][0] = 54,
+	[2][0][RTW89_MKK][0] = 64,
+	[2][0][RTW89_IC][0] = 74,
+	[2][0][RTW89_KCC][0] = 68,
+	[2][0][RTW89_ACMA][0] = 54,
+	[2][0][RTW89_CN][0] = 56,
+	[2][0][RTW89_UK][0] = 54,
+	[2][0][RTW89_FCC][1] = 74,
+	[2][0][RTW89_ETSI][1] = 54,
+	[2][0][RTW89_MKK][1] = 64,
+	[2][0][RTW89_IC][1] = 74,
+	[2][0][RTW89_KCC][1] = 68,
+	[2][0][RTW89_ACMA][1] = 54,
+	[2][0][RTW89_CN][1] = 56,
+	[2][0][RTW89_UK][1] = 54,
+	[2][0][RTW89_FCC][2] = 76,
+	[2][0][RTW89_ETSI][2] = 54,
+	[2][0][RTW89_MKK][2] = 64,
+	[2][0][RTW89_IC][2] = 76,
+	[2][0][RTW89_KCC][2] = 68,
+	[2][0][RTW89_ACMA][2] = 54,
+	[2][0][RTW89_CN][2] = 56,
+	[2][0][RTW89_UK][2] = 54,
+	[2][0][RTW89_FCC][3] = 76,
+	[2][0][RTW89_ETSI][3] = 54,
+	[2][0][RTW89_MKK][3] = 64,
+	[2][0][RTW89_IC][3] = 76,
+	[2][0][RTW89_KCC][3] = 68,
+	[2][0][RTW89_ACMA][3] = 54,
+	[2][0][RTW89_CN][3] = 56,
+	[2][0][RTW89_UK][3] = 54,
+	[2][0][RTW89_FCC][4] = 76,
+	[2][0][RTW89_ETSI][4] = 54,
+	[2][0][RTW89_MKK][4] = 64,
+	[2][0][RTW89_IC][4] = 76,
+	[2][0][RTW89_KCC][4] = 68,
+	[2][0][RTW89_ACMA][4] = 54,
+	[2][0][RTW89_CN][4] = 56,
+	[2][0][RTW89_UK][4] = 54,
+	[2][0][RTW89_FCC][5] = 80,
+	[2][0][RTW89_ETSI][5] = 54,
+	[2][0][RTW89_MKK][5] = 64,
+	[2][0][RTW89_IC][5] = 80,
+	[2][0][RTW89_KCC][5] = 68,
+	[2][0][RTW89_ACMA][5] = 54,
+	[2][0][RTW89_CN][5] = 56,
+	[2][0][RTW89_UK][5] = 54,
+	[2][0][RTW89_FCC][6] = 72,
+	[2][0][RTW89_ETSI][6] = 54,
+	[2][0][RTW89_MKK][6] = 64,
+	[2][0][RTW89_IC][6] = 72,
+	[2][0][RTW89_KCC][6] = 68,
+	[2][0][RTW89_ACMA][6] = 54,
+	[2][0][RTW89_CN][6] = 56,
+	[2][0][RTW89_UK][6] = 54,
+	[2][0][RTW89_FCC][7] = 72,
+	[2][0][RTW89_ETSI][7] = 54,
+	[2][0][RTW89_MKK][7] = 64,
+	[2][0][RTW89_IC][7] = 72,
+	[2][0][RTW89_KCC][7] = 68,
+	[2][0][RTW89_ACMA][7] = 54,
+	[2][0][RTW89_CN][7] = 56,
+	[2][0][RTW89_UK][7] = 54,
+	[2][0][RTW89_FCC][8] = 72,
+	[2][0][RTW89_ETSI][8] = 54,
+	[2][0][RTW89_MKK][8] = 64,
+	[2][0][RTW89_IC][8] = 72,
+	[2][0][RTW89_KCC][8] = 68,
+	[2][0][RTW89_ACMA][8] = 54,
+	[2][0][RTW89_CN][8] = 56,
+	[2][0][RTW89_UK][8] = 54,
+	[2][0][RTW89_FCC][9] = 70,
+	[2][0][RTW89_ETSI][9] = 54,
+	[2][0][RTW89_MKK][9] = 64,
+	[2][0][RTW89_IC][9] = 70,
+	[2][0][RTW89_KCC][9] = 68,
+	[2][0][RTW89_ACMA][9] = 54,
+	[2][0][RTW89_CN][9] = 56,
+	[2][0][RTW89_UK][9] = 54,
+	[2][0][RTW89_FCC][10] = 70,
+	[2][0][RTW89_ETSI][10] = 54,
+	[2][0][RTW89_MKK][10] = 64,
+	[2][0][RTW89_IC][10] = 70,
+	[2][0][RTW89_KCC][10] = 68,
+	[2][0][RTW89_ACMA][10] = 54,
+	[2][0][RTW89_CN][10] = 56,
+	[2][0][RTW89_UK][10] = 54,
+	[2][0][RTW89_FCC][11] = 62,
+	[2][0][RTW89_ETSI][11] = 54,
+	[2][0][RTW89_MKK][11] = 64,
+	[2][0][RTW89_IC][11] = 62,
+	[2][0][RTW89_KCC][11] = 68,
+	[2][0][RTW89_ACMA][11] = 54,
+	[2][0][RTW89_CN][11] = 56,
+	[2][0][RTW89_UK][11] = 54,
+	[2][0][RTW89_FCC][12] = 34,
+	[2][0][RTW89_ETSI][12] = 54,
+	[2][0][RTW89_MKK][12] = 64,
+	[2][0][RTW89_IC][12] = 34,
+	[2][0][RTW89_KCC][12] = 68,
+	[2][0][RTW89_ACMA][12] = 54,
+	[2][0][RTW89_CN][12] = 56,
+	[2][0][RTW89_UK][12] = 54,
+	[2][0][RTW89_FCC][13] = 127,
+	[2][0][RTW89_ETSI][13] = 127,
+	[2][0][RTW89_MKK][13] = 127,
+	[2][0][RTW89_IC][13] = 127,
+	[2][0][RTW89_KCC][13] = 127,
+	[2][0][RTW89_ACMA][13] = 127,
+	[2][0][RTW89_CN][13] = 127,
+	[2][0][RTW89_UK][13] = 127,
+	[2][1][RTW89_FCC][0] = 127,
+	[2][1][RTW89_ETSI][0] = 127,
+	[2][1][RTW89_MKK][0] = 127,
+	[2][1][RTW89_IC][0] = 127,
+	[2][1][RTW89_KCC][0] = 127,
+	[2][1][RTW89_ACMA][0] = 127,
+	[2][1][RTW89_CN][0] = 44,
+	[2][1][RTW89_UK][0] = 127,
+	[2][1][RTW89_FCC][1] = 127,
+	[2][1][RTW89_ETSI][1] = 127,
+	[2][1][RTW89_MKK][1] = 127,
+	[2][1][RTW89_IC][1] = 127,
+	[2][1][RTW89_KCC][1] = 127,
+	[2][1][RTW89_ACMA][1] = 127,
+	[2][1][RTW89_CN][1] = 44,
+	[2][1][RTW89_UK][1] = 127,
+	[2][1][RTW89_FCC][2] = 127,
+	[2][1][RTW89_ETSI][2] = 127,
+	[2][1][RTW89_MKK][2] = 127,
+	[2][1][RTW89_IC][2] = 127,
+	[2][1][RTW89_KCC][2] = 127,
+	[2][1][RTW89_ACMA][2] = 127,
+	[2][1][RTW89_CN][2] = 44,
+	[2][1][RTW89_UK][2] = 127,
+	[2][1][RTW89_FCC][3] = 127,
+	[2][1][RTW89_ETSI][3] = 127,
+	[2][1][RTW89_MKK][3] = 127,
+	[2][1][RTW89_IC][3] = 127,
+	[2][1][RTW89_KCC][3] = 127,
+	[2][1][RTW89_ACMA][3] = 127,
+	[2][1][RTW89_CN][3] = 44,
+	[2][1][RTW89_UK][3] = 127,
+	[2][1][RTW89_FCC][4] = 127,
+	[2][1][RTW89_ETSI][4] = 127,
+	[2][1][RTW89_MKK][4] = 127,
+	[2][1][RTW89_IC][4] = 127,
+	[2][1][RTW89_KCC][4] = 127,
+	[2][1][RTW89_ACMA][4] = 127,
+	[2][1][RTW89_CN][4] = 44,
+	[2][1][RTW89_UK][4] = 127,
+	[2][1][RTW89_FCC][5] = 127,
+	[2][1][RTW89_ETSI][5] = 127,
+	[2][1][RTW89_MKK][5] = 127,
+	[2][1][RTW89_IC][5] = 127,
+	[2][1][RTW89_KCC][5] = 127,
+	[2][1][RTW89_ACMA][5] = 127,
+	[2][1][RTW89_CN][5] = 44,
+	[2][1][RTW89_UK][5] = 127,
+	[2][1][RTW89_FCC][6] = 127,
+	[2][1][RTW89_ETSI][6] = 127,
+	[2][1][RTW89_MKK][6] = 127,
+	[2][1][RTW89_IC][6] = 127,
+	[2][1][RTW89_KCC][6] = 127,
+	[2][1][RTW89_ACMA][6] = 127,
+	[2][1][RTW89_CN][6] = 44,
+	[2][1][RTW89_UK][6] = 127,
+	[2][1][RTW89_FCC][7] = 127,
+	[2][1][RTW89_ETSI][7] = 127,
+	[2][1][RTW89_MKK][7] = 127,
+	[2][1][RTW89_IC][7] = 127,
+	[2][1][RTW89_KCC][7] = 127,
+	[2][1][RTW89_ACMA][7] = 127,
+	[2][1][RTW89_CN][7] = 44,
+	[2][1][RTW89_UK][7] = 127,
+	[2][1][RTW89_FCC][8] = 127,
+	[2][1][RTW89_ETSI][8] = 127,
+	[2][1][RTW89_MKK][8] = 127,
+	[2][1][RTW89_IC][8] = 127,
+	[2][1][RTW89_KCC][8] = 127,
+	[2][1][RTW89_ACMA][8] = 127,
+	[2][1][RTW89_CN][8] = 44,
+	[2][1][RTW89_UK][8] = 127,
+	[2][1][RTW89_FCC][9] = 127,
+	[2][1][RTW89_ETSI][9] = 127,
+	[2][1][RTW89_MKK][9] = 127,
+	[2][1][RTW89_IC][9] = 127,
+	[2][1][RTW89_KCC][9] = 127,
+	[2][1][RTW89_ACMA][9] = 127,
+	[2][1][RTW89_CN][9] = 44,
+	[2][1][RTW89_UK][9] = 127,
+	[2][1][RTW89_FCC][10] = 127,
+	[2][1][RTW89_ETSI][10] = 127,
+	[2][1][RTW89_MKK][10] = 127,
+	[2][1][RTW89_IC][10] = 127,
+	[2][1][RTW89_KCC][10] = 127,
+	[2][1][RTW89_ACMA][10] = 127,
+	[2][1][RTW89_CN][10] = 44,
+	[2][1][RTW89_UK][10] = 127,
+	[2][1][RTW89_FCC][11] = 127,
+	[2][1][RTW89_ETSI][11] = 127,
+	[2][1][RTW89_MKK][11] = 127,
+	[2][1][RTW89_IC][11] = 127,
+	[2][1][RTW89_KCC][11] = 127,
+	[2][1][RTW89_ACMA][11] = 127,
+	[2][1][RTW89_CN][11] = 44,
+	[2][1][RTW89_UK][11] = 127,
+	[2][1][RTW89_FCC][12] = 127,
+	[2][1][RTW89_ETSI][12] = 127,
+	[2][1][RTW89_MKK][12] = 127,
+	[2][1][RTW89_IC][12] = 127,
+	[2][1][RTW89_KCC][12] = 127,
+	[2][1][RTW89_ACMA][12] = 127,
+	[2][1][RTW89_CN][12] = 42,
+	[2][1][RTW89_UK][12] = 127,
+	[2][1][RTW89_FCC][13] = 127,
+	[2][1][RTW89_ETSI][13] = 127,
+	[2][1][RTW89_MKK][13] = 127,
+	[2][1][RTW89_IC][13] = 127,
+	[2][1][RTW89_KCC][13] = 127,
+	[2][1][RTW89_ACMA][13] = 127,
+	[2][1][RTW89_CN][13] = 127,
+	[2][1][RTW89_UK][13] = 127,
+};
+
+static
+const s8 rtw89_8851b_txpwr_lmt_ru_5g_type2[RTW89_RU_NUM][RTW89_NTX_NUM]
+					  [RTW89_REGD_NUM][RTW89_5G_CH_NUM] = {
+	[0][0][RTW89_WW][0] = 16,
+	[0][0][RTW89_WW][2] = 16,
+	[0][0][RTW89_WW][4] = 16,
+	[0][0][RTW89_WW][6] = 16,
+	[0][0][RTW89_WW][8] = 16,
+	[0][0][RTW89_WW][10] = 16,
+	[0][0][RTW89_WW][12] = 16,
+	[0][0][RTW89_WW][14] = 16,
+	[0][0][RTW89_WW][15] = 24,
+	[0][0][RTW89_WW][17] = 24,
+	[0][0][RTW89_WW][19] = 24,
+	[0][0][RTW89_WW][21] = 24,
+	[0][0][RTW89_WW][23] = 24,
+	[0][0][RTW89_WW][25] = 24,
+	[0][0][RTW89_WW][27] = 24,
+	[0][0][RTW89_WW][29] = 24,
+	[0][0][RTW89_WW][31] = 24,
+	[0][0][RTW89_WW][33] = 24,
+	[0][0][RTW89_WW][35] = 24,
+	[0][0][RTW89_WW][37] = 44,
+	[0][0][RTW89_WW][38] = 24,
+	[0][0][RTW89_WW][40] = 24,
+	[0][0][RTW89_WW][42] = 24,
+	[0][0][RTW89_WW][44] = 24,
+	[0][0][RTW89_WW][46] = 24,
+	[0][0][RTW89_WW][48] = 40,
+	[0][0][RTW89_WW][50] = 42,
+	[0][0][RTW89_WW][52] = 38,
+	[0][1][RTW89_WW][0] = 4,
+	[0][1][RTW89_WW][2] = 4,
+	[0][1][RTW89_WW][4] = 4,
+	[0][1][RTW89_WW][6] = 4,
+	[0][1][RTW89_WW][8] = 4,
+	[0][1][RTW89_WW][10] = 4,
+	[0][1][RTW89_WW][12] = 4,
+	[0][1][RTW89_WW][14] = 4,
+	[0][1][RTW89_WW][15] = 0,
+	[0][1][RTW89_WW][17] = 0,
+	[0][1][RTW89_WW][19] = 0,
+	[0][1][RTW89_WW][21] = 0,
+	[0][1][RTW89_WW][23] = 0,
+	[0][1][RTW89_WW][25] = 0,
+	[0][1][RTW89_WW][27] = 0,
+	[0][1][RTW89_WW][29] = 0,
+	[0][1][RTW89_WW][31] = 0,
+	[0][1][RTW89_WW][33] = 0,
+	[0][1][RTW89_WW][35] = 0,
+	[0][1][RTW89_WW][37] = 0,
+	[0][1][RTW89_WW][38] = 42,
+	[0][1][RTW89_WW][40] = 42,
+	[0][1][RTW89_WW][42] = 42,
+	[0][1][RTW89_WW][44] = 42,
+	[0][1][RTW89_WW][46] = 42,
+	[0][1][RTW89_WW][48] = 0,
+	[0][1][RTW89_WW][50] = 0,
+	[0][1][RTW89_WW][52] = 0,
+	[1][0][RTW89_WW][0] = 26,
+	[1][0][RTW89_WW][2] = 26,
+	[1][0][RTW89_WW][4] = 26,
+	[1][0][RTW89_WW][6] = 26,
+	[1][0][RTW89_WW][8] = 26,
+	[1][0][RTW89_WW][10] = 26,
+	[1][0][RTW89_WW][12] = 26,
+	[1][0][RTW89_WW][14] = 26,
+	[1][0][RTW89_WW][15] = 34,
+	[1][0][RTW89_WW][17] = 34,
+	[1][0][RTW89_WW][19] = 34,
+	[1][0][RTW89_WW][21] = 34,
+	[1][0][RTW89_WW][23] = 34,
+	[1][0][RTW89_WW][25] = 34,
+	[1][0][RTW89_WW][27] = 34,
+	[1][0][RTW89_WW][29] = 34,
+	[1][0][RTW89_WW][31] = 34,
+	[1][0][RTW89_WW][33] = 34,
+	[1][0][RTW89_WW][35] = 34,
+	[1][0][RTW89_WW][37] = 54,
+	[1][0][RTW89_WW][38] = 28,
+	[1][0][RTW89_WW][40] = 28,
+	[1][0][RTW89_WW][42] = 28,
+	[1][0][RTW89_WW][44] = 28,
+	[1][0][RTW89_WW][46] = 28,
+	[1][0][RTW89_WW][48] = 52,
+	[1][0][RTW89_WW][50] = 52,
+	[1][0][RTW89_WW][52] = 50,
+	[1][1][RTW89_WW][0] = 14,
+	[1][1][RTW89_WW][2] = 14,
+	[1][1][RTW89_WW][4] = 14,
+	[1][1][RTW89_WW][6] = 14,
+	[1][1][RTW89_WW][8] = 14,
+	[1][1][RTW89_WW][10] = 14,
+	[1][1][RTW89_WW][12] = 14,
+	[1][1][RTW89_WW][14] = 14,
+	[1][1][RTW89_WW][15] = 0,
+	[1][1][RTW89_WW][17] = 0,
+	[1][1][RTW89_WW][19] = 0,
+	[1][1][RTW89_WW][21] = 0,
+	[1][1][RTW89_WW][23] = 0,
+	[1][1][RTW89_WW][25] = 0,
+	[1][1][RTW89_WW][27] = 0,
+	[1][1][RTW89_WW][29] = 0,
+	[1][1][RTW89_WW][31] = 0,
+	[1][1][RTW89_WW][33] = 0,
+	[1][1][RTW89_WW][35] = 0,
+	[1][1][RTW89_WW][37] = 0,
+	[1][1][RTW89_WW][38] = 54,
+	[1][1][RTW89_WW][40] = 54,
+	[1][1][RTW89_WW][42] = 54,
+	[1][1][RTW89_WW][44] = 54,
+	[1][1][RTW89_WW][46] = 54,
+	[1][1][RTW89_WW][48] = 0,
+	[1][1][RTW89_WW][50] = 0,
+	[1][1][RTW89_WW][52] = 0,
+	[2][0][RTW89_WW][0] = 40,
+	[2][0][RTW89_WW][2] = 40,
+	[2][0][RTW89_WW][4] = 40,
+	[2][0][RTW89_WW][6] = 40,
+	[2][0][RTW89_WW][8] = 40,
+	[2][0][RTW89_WW][10] = 40,
+	[2][0][RTW89_WW][12] = 40,
+	[2][0][RTW89_WW][14] = 40,
+	[2][0][RTW89_WW][15] = 46,
+	[2][0][RTW89_WW][17] = 46,
+	[2][0][RTW89_WW][19] = 46,
+	[2][0][RTW89_WW][21] = 46,
+	[2][0][RTW89_WW][23] = 46,
+	[2][0][RTW89_WW][25] = 46,
+	[2][0][RTW89_WW][27] = 46,
+	[2][0][RTW89_WW][29] = 46,
+	[2][0][RTW89_WW][31] = 46,
+	[2][0][RTW89_WW][33] = 46,
+	[2][0][RTW89_WW][35] = 46,
+	[2][0][RTW89_WW][37] = 66,
+	[2][0][RTW89_WW][38] = 28,
+	[2][0][RTW89_WW][40] = 28,
+	[2][0][RTW89_WW][42] = 28,
+	[2][0][RTW89_WW][44] = 28,
+	[2][0][RTW89_WW][46] = 28,
+	[2][0][RTW89_WW][48] = 62,
+	[2][0][RTW89_WW][50] = 62,
+	[2][0][RTW89_WW][52] = 60,
+	[2][1][RTW89_WW][0] = 28,
+	[2][1][RTW89_WW][2] = 28,
+	[2][1][RTW89_WW][4] = 28,
+	[2][1][RTW89_WW][6] = 28,
+	[2][1][RTW89_WW][8] = 28,
+	[2][1][RTW89_WW][10] = 28,
+	[2][1][RTW89_WW][12] = 28,
+	[2][1][RTW89_WW][14] = 28,
+	[2][1][RTW89_WW][15] = 0,
+	[2][1][RTW89_WW][17] = 0,
+	[2][1][RTW89_WW][19] = 0,
+	[2][1][RTW89_WW][21] = 0,
+	[2][1][RTW89_WW][23] = 0,
+	[2][1][RTW89_WW][25] = 0,
+	[2][1][RTW89_WW][27] = 0,
+	[2][1][RTW89_WW][29] = 0,
+	[2][1][RTW89_WW][31] = 0,
+	[2][1][RTW89_WW][33] = 0,
+	[2][1][RTW89_WW][35] = 0,
+	[2][1][RTW89_WW][37] = 0,
+	[2][1][RTW89_WW][38] = 56,
+	[2][1][RTW89_WW][40] = 56,
+	[2][1][RTW89_WW][42] = 56,
+	[2][1][RTW89_WW][44] = 56,
+	[2][1][RTW89_WW][46] = 56,
+	[2][1][RTW89_WW][48] = 0,
+	[2][1][RTW89_WW][50] = 0,
+	[2][1][RTW89_WW][52] = 0,
+	[0][0][RTW89_FCC][0] = 50,
+	[0][0][RTW89_ETSI][0] = 24,
+	[0][0][RTW89_MKK][0] = 26,
+	[0][0][RTW89_IC][0] = 28,
+	[0][0][RTW89_KCC][0] = 42,
+	[0][0][RTW89_ACMA][0] = 24,
+	[0][0][RTW89_CN][0] = 16,
+	[0][0][RTW89_UK][0] = 24,
+	[0][0][RTW89_FCC][2] = 54,
+	[0][0][RTW89_ETSI][2] = 24,
+	[0][0][RTW89_MKK][2] = 26,
+	[0][0][RTW89_IC][2] = 28,
+	[0][0][RTW89_KCC][2] = 42,
+	[0][0][RTW89_ACMA][2] = 24,
+	[0][0][RTW89_CN][2] = 16,
+	[0][0][RTW89_UK][2] = 24,
+	[0][0][RTW89_FCC][4] = 50,
+	[0][0][RTW89_ETSI][4] = 24,
+	[0][0][RTW89_MKK][4] = 26,
+	[0][0][RTW89_IC][4] = 28,
+	[0][0][RTW89_KCC][4] = 42,
+	[0][0][RTW89_ACMA][4] = 24,
+	[0][0][RTW89_CN][4] = 16,
+	[0][0][RTW89_UK][4] = 24,
+	[0][0][RTW89_FCC][6] = 50,
+	[0][0][RTW89_ETSI][6] = 24,
+	[0][0][RTW89_MKK][6] = 26,
+	[0][0][RTW89_IC][6] = 28,
+	[0][0][RTW89_KCC][6] = 18,
+	[0][0][RTW89_ACMA][6] = 24,
+	[0][0][RTW89_CN][6] = 16,
+	[0][0][RTW89_UK][6] = 24,
+	[0][0][RTW89_FCC][8] = 52,
+	[0][0][RTW89_ETSI][8] = 24,
+	[0][0][RTW89_MKK][8] = 26,
+	[0][0][RTW89_IC][8] = 52,
+	[0][0][RTW89_KCC][8] = 42,
+	[0][0][RTW89_ACMA][8] = 24,
+	[0][0][RTW89_CN][8] = 16,
+	[0][0][RTW89_UK][8] = 24,
+	[0][0][RTW89_FCC][10] = 52,
+	[0][0][RTW89_ETSI][10] = 24,
+	[0][0][RTW89_MKK][10] = 26,
+	[0][0][RTW89_IC][10] = 52,
+	[0][0][RTW89_KCC][10] = 42,
+	[0][0][RTW89_ACMA][10] = 24,
+	[0][0][RTW89_CN][10] = 16,
+	[0][0][RTW89_UK][10] = 24,
+	[0][0][RTW89_FCC][12] = 56,
+	[0][0][RTW89_ETSI][12] = 24,
+	[0][0][RTW89_MKK][12] = 26,
+	[0][0][RTW89_IC][12] = 56,
+	[0][0][RTW89_KCC][12] = 44,
+	[0][0][RTW89_ACMA][12] = 24,
+	[0][0][RTW89_CN][12] = 16,
+	[0][0][RTW89_UK][12] = 24,
+	[0][0][RTW89_FCC][14] = 56,
+	[0][0][RTW89_ETSI][14] = 24,
+	[0][0][RTW89_MKK][14] = 26,
+	[0][0][RTW89_IC][14] = 56,
+	[0][0][RTW89_KCC][14] = 44,
+	[0][0][RTW89_ACMA][14] = 24,
+	[0][0][RTW89_CN][14] = 16,
+	[0][0][RTW89_UK][14] = 24,
+	[0][0][RTW89_FCC][15] = 52,
+	[0][0][RTW89_ETSI][15] = 24,
+	[0][0][RTW89_MKK][15] = 46,
+	[0][0][RTW89_IC][15] = 52,
+	[0][0][RTW89_KCC][15] = 44,
+	[0][0][RTW89_ACMA][15] = 24,
+	[0][0][RTW89_CN][15] = 127,
+	[0][0][RTW89_UK][15] = 24,
+	[0][0][RTW89_FCC][17] = 52,
+	[0][0][RTW89_ETSI][17] = 24,
+	[0][0][RTW89_MKK][17] = 50,
+	[0][0][RTW89_IC][17] = 52,
+	[0][0][RTW89_KCC][17] = 44,
+	[0][0][RTW89_ACMA][17] = 24,
+	[0][0][RTW89_CN][17] = 127,
+	[0][0][RTW89_UK][17] = 24,
+	[0][0][RTW89_FCC][19] = 52,
+	[0][0][RTW89_ETSI][19] = 24,
+	[0][0][RTW89_MKK][19] = 50,
+	[0][0][RTW89_IC][19] = 52,
+	[0][0][RTW89_KCC][19] = 44,
+	[0][0][RTW89_ACMA][19] = 24,
+	[0][0][RTW89_CN][19] = 127,
+	[0][0][RTW89_UK][19] = 24,
+	[0][0][RTW89_FCC][21] = 52,
+	[0][0][RTW89_ETSI][21] = 24,
+	[0][0][RTW89_MKK][21] = 50,
+	[0][0][RTW89_IC][21] = 52,
+	[0][0][RTW89_KCC][21] = 44,
+	[0][0][RTW89_ACMA][21] = 24,
+	[0][0][RTW89_CN][21] = 127,
+	[0][0][RTW89_UK][21] = 24,
+	[0][0][RTW89_FCC][23] = 52,
+	[0][0][RTW89_ETSI][23] = 24,
+	[0][0][RTW89_MKK][23] = 50,
+	[0][0][RTW89_IC][23] = 52,
+	[0][0][RTW89_KCC][23] = 44,
+	[0][0][RTW89_ACMA][23] = 24,
+	[0][0][RTW89_CN][23] = 127,
+	[0][0][RTW89_UK][23] = 24,
+	[0][0][RTW89_FCC][25] = 52,
+	[0][0][RTW89_ETSI][25] = 24,
+	[0][0][RTW89_MKK][25] = 50,
+	[0][0][RTW89_IC][25] = 127,
+	[0][0][RTW89_KCC][25] = 44,
+	[0][0][RTW89_ACMA][25] = 127,
+	[0][0][RTW89_CN][25] = 127,
+	[0][0][RTW89_UK][25] = 24,
+	[0][0][RTW89_FCC][27] = 52,
+	[0][0][RTW89_ETSI][27] = 24,
+	[0][0][RTW89_MKK][27] = 50,
+	[0][0][RTW89_IC][27] = 127,
+	[0][0][RTW89_KCC][27] = 42,
+	[0][0][RTW89_ACMA][27] = 127,
+	[0][0][RTW89_CN][27] = 127,
+	[0][0][RTW89_UK][27] = 24,
+	[0][0][RTW89_FCC][29] = 52,
+	[0][0][RTW89_ETSI][29] = 24,
+	[0][0][RTW89_MKK][29] = 50,
+	[0][0][RTW89_IC][29] = 127,
+	[0][0][RTW89_KCC][29] = 42,
+	[0][0][RTW89_ACMA][29] = 127,
+	[0][0][RTW89_CN][29] = 127,
+	[0][0][RTW89_UK][29] = 24,
+	[0][0][RTW89_FCC][31] = 52,
+	[0][0][RTW89_ETSI][31] = 24,
+	[0][0][RTW89_MKK][31] = 50,
+	[0][0][RTW89_IC][31] = 56,
+	[0][0][RTW89_KCC][31] = 42,
+	[0][0][RTW89_ACMA][31] = 24,
+	[0][0][RTW89_CN][31] = 127,
+	[0][0][RTW89_UK][31] = 24,
+	[0][0][RTW89_FCC][33] = 56,
+	[0][0][RTW89_ETSI][33] = 24,
+	[0][0][RTW89_MKK][33] = 50,
+	[0][0][RTW89_IC][33] = 56,
+	[0][0][RTW89_KCC][33] = 42,
+	[0][0][RTW89_ACMA][33] = 24,
+	[0][0][RTW89_CN][33] = 127,
+	[0][0][RTW89_UK][33] = 24,
+	[0][0][RTW89_FCC][35] = 56,
+	[0][0][RTW89_ETSI][35] = 24,
+	[0][0][RTW89_MKK][35] = 50,
+	[0][0][RTW89_IC][35] = 56,
+	[0][0][RTW89_KCC][35] = 42,
+	[0][0][RTW89_ACMA][35] = 24,
+	[0][0][RTW89_CN][35] = 127,
+	[0][0][RTW89_UK][35] = 24,
+	[0][0][RTW89_FCC][37] = 84,
+	[0][0][RTW89_ETSI][37] = 127,
+	[0][0][RTW89_MKK][37] = 46,
+	[0][0][RTW89_IC][37] = 84,
+	[0][0][RTW89_KCC][37] = 44,
+	[0][0][RTW89_ACMA][37] = 50,
+	[0][0][RTW89_CN][37] = 127,
+	[0][0][RTW89_UK][37] = 52,
+	[0][0][RTW89_FCC][38] = 68,
+	[0][0][RTW89_ETSI][38] = 28,
+	[0][0][RTW89_MKK][38] = 127,
+	[0][0][RTW89_IC][38] = 68,
+	[0][0][RTW89_KCC][38] = 44,
+	[0][0][RTW89_ACMA][38] = 84,
+	[0][0][RTW89_CN][38] = 54,
+	[0][0][RTW89_UK][38] = 24,
+	[0][0][RTW89_FCC][40] = 68,
+	[0][0][RTW89_ETSI][40] = 28,
+	[0][0][RTW89_MKK][40] = 127,
+	[0][0][RTW89_IC][40] = 68,
+	[0][0][RTW89_KCC][40] = 44,
+	[0][0][RTW89_ACMA][40] = 84,
+	[0][0][RTW89_CN][40] = 54,
+	[0][0][RTW89_UK][40] = 24,
+	[0][0][RTW89_FCC][42] = 70,
+	[0][0][RTW89_ETSI][42] = 28,
+	[0][0][RTW89_MKK][42] = 127,
+	[0][0][RTW89_IC][42] = 70,
+	[0][0][RTW89_KCC][42] = 44,
+	[0][0][RTW89_ACMA][42] = 84,
+	[0][0][RTW89_CN][42] = 54,
+	[0][0][RTW89_UK][42] = 24,
+	[0][0][RTW89_FCC][44] = 62,
+	[0][0][RTW89_ETSI][44] = 28,
+	[0][0][RTW89_MKK][44] = 127,
+	[0][0][RTW89_IC][44] = 62,
+	[0][0][RTW89_KCC][44] = 44,
+	[0][0][RTW89_ACMA][44] = 84,
+	[0][0][RTW89_CN][44] = 54,
+	[0][0][RTW89_UK][44] = 24,
+	[0][0][RTW89_FCC][46] = 62,
+	[0][0][RTW89_ETSI][46] = 28,
+	[0][0][RTW89_MKK][46] = 127,
+	[0][0][RTW89_IC][46] = 62,
+	[0][0][RTW89_KCC][46] = 44,
+	[0][0][RTW89_ACMA][46] = 84,
+	[0][0][RTW89_CN][46] = 54,
+	[0][0][RTW89_UK][46] = 24,
+	[0][0][RTW89_FCC][48] = 40,
+	[0][0][RTW89_ETSI][48] = 127,
+	[0][0][RTW89_MKK][48] = 127,
+	[0][0][RTW89_IC][48] = 127,
+	[0][0][RTW89_KCC][48] = 127,
+	[0][0][RTW89_ACMA][48] = 127,
+	[0][0][RTW89_CN][48] = 127,
+	[0][0][RTW89_UK][48] = 127,
+	[0][0][RTW89_FCC][50] = 42,
+	[0][0][RTW89_ETSI][50] = 127,
+	[0][0][RTW89_MKK][50] = 127,
+	[0][0][RTW89_IC][50] = 127,
+	[0][0][RTW89_KCC][50] = 127,
+	[0][0][RTW89_ACMA][50] = 127,
+	[0][0][RTW89_CN][50] = 127,
+	[0][0][RTW89_UK][50] = 127,
+	[0][0][RTW89_FCC][52] = 38,
+	[0][0][RTW89_ETSI][52] = 127,
+	[0][0][RTW89_MKK][52] = 127,
+	[0][0][RTW89_IC][52] = 127,
+	[0][0][RTW89_KCC][52] = 127,
+	[0][0][RTW89_ACMA][52] = 127,
+	[0][0][RTW89_CN][52] = 127,
+	[0][0][RTW89_UK][52] = 127,
+	[0][1][RTW89_FCC][0] = 127,
+	[0][1][RTW89_ETSI][0] = 127,
+	[0][1][RTW89_MKK][0] = 127,
+	[0][1][RTW89_IC][0] = 127,
+	[0][1][RTW89_KCC][0] = 127,
+	[0][1][RTW89_ACMA][0] = 127,
+	[0][1][RTW89_CN][0] = 4,
+	[0][1][RTW89_UK][0] = 127,
+	[0][1][RTW89_FCC][2] = 127,
+	[0][1][RTW89_ETSI][2] = 127,
+	[0][1][RTW89_MKK][2] = 127,
+	[0][1][RTW89_IC][2] = 127,
+	[0][1][RTW89_KCC][2] = 127,
+	[0][1][RTW89_ACMA][2] = 127,
+	[0][1][RTW89_CN][2] = 4,
+	[0][1][RTW89_UK][2] = 127,
+	[0][1][RTW89_FCC][4] = 127,
+	[0][1][RTW89_ETSI][4] = 127,
+	[0][1][RTW89_MKK][4] = 127,
+	[0][1][RTW89_IC][4] = 127,
+	[0][1][RTW89_KCC][4] = 127,
+	[0][1][RTW89_ACMA][4] = 127,
+	[0][1][RTW89_CN][4] = 4,
+	[0][1][RTW89_UK][4] = 127,
+	[0][1][RTW89_FCC][6] = 127,
+	[0][1][RTW89_ETSI][6] = 127,
+	[0][1][RTW89_MKK][6] = 127,
+	[0][1][RTW89_IC][6] = 127,
+	[0][1][RTW89_KCC][6] = 127,
+	[0][1][RTW89_ACMA][6] = 127,
+	[0][1][RTW89_CN][6] = 4,
+	[0][1][RTW89_UK][6] = 127,
+	[0][1][RTW89_FCC][8] = 127,
+	[0][1][RTW89_ETSI][8] = 127,
+	[0][1][RTW89_MKK][8] = 127,
+	[0][1][RTW89_IC][8] = 127,
+	[0][1][RTW89_KCC][8] = 127,
+	[0][1][RTW89_ACMA][8] = 127,
+	[0][1][RTW89_CN][8] = 4,
+	[0][1][RTW89_UK][8] = 127,
+	[0][1][RTW89_FCC][10] = 127,
+	[0][1][RTW89_ETSI][10] = 127,
+	[0][1][RTW89_MKK][10] = 127,
+	[0][1][RTW89_IC][10] = 127,
+	[0][1][RTW89_KCC][10] = 127,
+	[0][1][RTW89_ACMA][10] = 127,
+	[0][1][RTW89_CN][10] = 4,
+	[0][1][RTW89_UK][10] = 127,
+	[0][1][RTW89_FCC][12] = 127,
+	[0][1][RTW89_ETSI][12] = 127,
+	[0][1][RTW89_MKK][12] = 127,
+	[0][1][RTW89_IC][12] = 127,
+	[0][1][RTW89_KCC][12] = 127,
+	[0][1][RTW89_ACMA][12] = 127,
+	[0][1][RTW89_CN][12] = 4,
+	[0][1][RTW89_UK][12] = 127,
+	[0][1][RTW89_FCC][14] = 127,
+	[0][1][RTW89_ETSI][14] = 127,
+	[0][1][RTW89_MKK][14] = 127,
+	[0][1][RTW89_IC][14] = 127,
+	[0][1][RTW89_KCC][14] = 127,
+	[0][1][RTW89_ACMA][14] = 127,
+	[0][1][RTW89_CN][14] = 4,
+	[0][1][RTW89_UK][14] = 127,
+	[0][1][RTW89_FCC][15] = 127,
+	[0][1][RTW89_ETSI][15] = 127,
+	[0][1][RTW89_MKK][15] = 127,
+	[0][1][RTW89_IC][15] = 127,
+	[0][1][RTW89_KCC][15] = 127,
+	[0][1][RTW89_ACMA][15] = 127,
+	[0][1][RTW89_CN][15] = 127,
+	[0][1][RTW89_UK][15] = 127,
+	[0][1][RTW89_FCC][17] = 127,
+	[0][1][RTW89_ETSI][17] = 127,
+	[0][1][RTW89_MKK][17] = 127,
+	[0][1][RTW89_IC][17] = 127,
+	[0][1][RTW89_KCC][17] = 127,
+	[0][1][RTW89_ACMA][17] = 127,
+	[0][1][RTW89_CN][17] = 127,
+	[0][1][RTW89_UK][17] = 127,
+	[0][1][RTW89_FCC][19] = 127,
+	[0][1][RTW89_ETSI][19] = 127,
+	[0][1][RTW89_MKK][19] = 127,
+	[0][1][RTW89_IC][19] = 127,
+	[0][1][RTW89_KCC][19] = 127,
+	[0][1][RTW89_ACMA][19] = 127,
+	[0][1][RTW89_CN][19] = 127,
+	[0][1][RTW89_UK][19] = 127,
+	[0][1][RTW89_FCC][21] = 127,
+	[0][1][RTW89_ETSI][21] = 127,
+	[0][1][RTW89_MKK][21] = 127,
+	[0][1][RTW89_IC][21] = 127,
+	[0][1][RTW89_KCC][21] = 127,
+	[0][1][RTW89_ACMA][21] = 127,
+	[0][1][RTW89_CN][21] = 127,
+	[0][1][RTW89_UK][21] = 127,
+	[0][1][RTW89_FCC][23] = 127,
+	[0][1][RTW89_ETSI][23] = 127,
+	[0][1][RTW89_MKK][23] = 127,
+	[0][1][RTW89_IC][23] = 127,
+	[0][1][RTW89_KCC][23] = 127,
+	[0][1][RTW89_ACMA][23] = 127,
+	[0][1][RTW89_CN][23] = 127,
+	[0][1][RTW89_UK][23] = 127,
+	[0][1][RTW89_FCC][25] = 127,
+	[0][1][RTW89_ETSI][25] = 127,
+	[0][1][RTW89_MKK][25] = 127,
+	[0][1][RTW89_IC][25] = 127,
+	[0][1][RTW89_KCC][25] = 127,
+	[0][1][RTW89_ACMA][25] = 127,
+	[0][1][RTW89_CN][25] = 127,
+	[0][1][RTW89_UK][25] = 127,
+	[0][1][RTW89_FCC][27] = 127,
+	[0][1][RTW89_ETSI][27] = 127,
+	[0][1][RTW89_MKK][27] = 127,
+	[0][1][RTW89_IC][27] = 127,
+	[0][1][RTW89_KCC][27] = 127,
+	[0][1][RTW89_ACMA][27] = 127,
+	[0][1][RTW89_CN][27] = 127,
+	[0][1][RTW89_UK][27] = 127,
+	[0][1][RTW89_FCC][29] = 127,
+	[0][1][RTW89_ETSI][29] = 127,
+	[0][1][RTW89_MKK][29] = 127,
+	[0][1][RTW89_IC][29] = 127,
+	[0][1][RTW89_KCC][29] = 127,
+	[0][1][RTW89_ACMA][29] = 127,
+	[0][1][RTW89_CN][29] = 127,
+	[0][1][RTW89_UK][29] = 127,
+	[0][1][RTW89_FCC][31] = 127,
+	[0][1][RTW89_ETSI][31] = 127,
+	[0][1][RTW89_MKK][31] = 127,
+	[0][1][RTW89_IC][31] = 127,
+	[0][1][RTW89_KCC][31] = 127,
+	[0][1][RTW89_ACMA][31] = 127,
+	[0][1][RTW89_CN][31] = 127,
+	[0][1][RTW89_UK][31] = 127,
+	[0][1][RTW89_FCC][33] = 127,
+	[0][1][RTW89_ETSI][33] = 127,
+	[0][1][RTW89_MKK][33] = 127,
+	[0][1][RTW89_IC][33] = 127,
+	[0][1][RTW89_KCC][33] = 127,
+	[0][1][RTW89_ACMA][33] = 127,
+	[0][1][RTW89_CN][33] = 127,
+	[0][1][RTW89_UK][33] = 127,
+	[0][1][RTW89_FCC][35] = 127,
+	[0][1][RTW89_ETSI][35] = 127,
+	[0][1][RTW89_MKK][35] = 127,
+	[0][1][RTW89_IC][35] = 127,
+	[0][1][RTW89_KCC][35] = 127,
+	[0][1][RTW89_ACMA][35] = 127,
+	[0][1][RTW89_CN][35] = 127,
+	[0][1][RTW89_UK][35] = 127,
+	[0][1][RTW89_FCC][37] = 127,
+	[0][1][RTW89_ETSI][37] = 127,
+	[0][1][RTW89_MKK][37] = 127,
+	[0][1][RTW89_IC][37] = 127,
+	[0][1][RTW89_KCC][37] = 127,
+	[0][1][RTW89_ACMA][37] = 127,
+	[0][1][RTW89_CN][37] = 127,
+	[0][1][RTW89_UK][37] = 127,
+	[0][1][RTW89_FCC][38] = 127,
+	[0][1][RTW89_ETSI][38] = 127,
+	[0][1][RTW89_MKK][38] = 127,
+	[0][1][RTW89_IC][38] = 127,
+	[0][1][RTW89_KCC][38] = 127,
+	[0][1][RTW89_ACMA][38] = 127,
+	[0][1][RTW89_CN][38] = 42,
+	[0][1][RTW89_UK][38] = 127,
+	[0][1][RTW89_FCC][40] = 127,
+	[0][1][RTW89_ETSI][40] = 127,
+	[0][1][RTW89_MKK][40] = 127,
+	[0][1][RTW89_IC][40] = 127,
+	[0][1][RTW89_KCC][40] = 127,
+	[0][1][RTW89_ACMA][40] = 127,
+	[0][1][RTW89_CN][40] = 42,
+	[0][1][RTW89_UK][40] = 127,
+	[0][1][RTW89_FCC][42] = 127,
+	[0][1][RTW89_ETSI][42] = 127,
+	[0][1][RTW89_MKK][42] = 127,
+	[0][1][RTW89_IC][42] = 127,
+	[0][1][RTW89_KCC][42] = 127,
+	[0][1][RTW89_ACMA][42] = 127,
+	[0][1][RTW89_CN][42] = 42,
+	[0][1][RTW89_UK][42] = 127,
+	[0][1][RTW89_FCC][44] = 127,
+	[0][1][RTW89_ETSI][44] = 127,
+	[0][1][RTW89_MKK][44] = 127,
+	[0][1][RTW89_IC][44] = 127,
+	[0][1][RTW89_KCC][44] = 127,
+	[0][1][RTW89_ACMA][44] = 127,
+	[0][1][RTW89_CN][44] = 42,
+	[0][1][RTW89_UK][44] = 127,
+	[0][1][RTW89_FCC][46] = 127,
+	[0][1][RTW89_ETSI][46] = 127,
+	[0][1][RTW89_MKK][46] = 127,
+	[0][1][RTW89_IC][46] = 127,
+	[0][1][RTW89_KCC][46] = 127,
+	[0][1][RTW89_ACMA][46] = 127,
+	[0][1][RTW89_CN][46] = 42,
+	[0][1][RTW89_UK][46] = 127,
+	[0][1][RTW89_FCC][48] = 127,
+	[0][1][RTW89_ETSI][48] = 127,
+	[0][1][RTW89_MKK][48] = 127,
+	[0][1][RTW89_IC][48] = 127,
+	[0][1][RTW89_KCC][48] = 127,
+	[0][1][RTW89_ACMA][48] = 127,
+	[0][1][RTW89_CN][48] = 127,
+	[0][1][RTW89_UK][48] = 127,
+	[0][1][RTW89_FCC][50] = 127,
+	[0][1][RTW89_ETSI][50] = 127,
+	[0][1][RTW89_MKK][50] = 127,
+	[0][1][RTW89_IC][50] = 127,
+	[0][1][RTW89_KCC][50] = 127,
+	[0][1][RTW89_ACMA][50] = 127,
+	[0][1][RTW89_CN][50] = 127,
+	[0][1][RTW89_UK][50] = 127,
+	[0][1][RTW89_FCC][52] = 127,
+	[0][1][RTW89_ETSI][52] = 127,
+	[0][1][RTW89_MKK][52] = 127,
+	[0][1][RTW89_IC][52] = 127,
+	[0][1][RTW89_KCC][52] = 127,
+	[0][1][RTW89_ACMA][52] = 127,
+	[0][1][RTW89_CN][52] = 127,
+	[0][1][RTW89_UK][52] = 127,
+	[1][0][RTW89_FCC][0] = 64,
+	[1][0][RTW89_ETSI][0] = 34,
+	[1][0][RTW89_MKK][0] = 38,
+	[1][0][RTW89_IC][0] = 38,
+	[1][0][RTW89_KCC][0] = 52,
+	[1][0][RTW89_ACMA][0] = 34,
+	[1][0][RTW89_CN][0] = 26,
+	[1][0][RTW89_UK][0] = 34,
+	[1][0][RTW89_FCC][2] = 66,
+	[1][0][RTW89_ETSI][2] = 34,
+	[1][0][RTW89_MKK][2] = 38,
+	[1][0][RTW89_IC][2] = 38,
+	[1][0][RTW89_KCC][2] = 52,
+	[1][0][RTW89_ACMA][2] = 34,
+	[1][0][RTW89_CN][2] = 26,
+	[1][0][RTW89_UK][2] = 34,
+	[1][0][RTW89_FCC][4] = 60,
+	[1][0][RTW89_ETSI][4] = 34,
+	[1][0][RTW89_MKK][4] = 36,
+	[1][0][RTW89_IC][4] = 38,
+	[1][0][RTW89_KCC][4] = 52,
+	[1][0][RTW89_ACMA][4] = 34,
+	[1][0][RTW89_CN][4] = 26,
+	[1][0][RTW89_UK][4] = 34,
+	[1][0][RTW89_FCC][6] = 60,
+	[1][0][RTW89_ETSI][6] = 34,
+	[1][0][RTW89_MKK][6] = 36,
+	[1][0][RTW89_IC][6] = 38,
+	[1][0][RTW89_KCC][6] = 32,
+	[1][0][RTW89_ACMA][6] = 34,
+	[1][0][RTW89_CN][6] = 26,
+	[1][0][RTW89_UK][6] = 34,
+	[1][0][RTW89_FCC][8] = 62,
+	[1][0][RTW89_ETSI][8] = 34,
+	[1][0][RTW89_MKK][8] = 38,
+	[1][0][RTW89_IC][8] = 62,
+	[1][0][RTW89_KCC][8] = 52,
+	[1][0][RTW89_ACMA][8] = 34,
+	[1][0][RTW89_CN][8] = 26,
+	[1][0][RTW89_UK][8] = 34,
+	[1][0][RTW89_FCC][10] = 62,
+	[1][0][RTW89_ETSI][10] = 34,
+	[1][0][RTW89_MKK][10] = 38,
+	[1][0][RTW89_IC][10] = 62,
+	[1][0][RTW89_KCC][10] = 52,
+	[1][0][RTW89_ACMA][10] = 34,
+	[1][0][RTW89_CN][10] = 26,
+	[1][0][RTW89_UK][10] = 34,
+	[1][0][RTW89_FCC][12] = 62,
+	[1][0][RTW89_ETSI][12] = 34,
+	[1][0][RTW89_MKK][12] = 38,
+	[1][0][RTW89_IC][12] = 62,
+	[1][0][RTW89_KCC][12] = 54,
+	[1][0][RTW89_ACMA][12] = 34,
+	[1][0][RTW89_CN][12] = 26,
+	[1][0][RTW89_UK][12] = 34,
+	[1][0][RTW89_FCC][14] = 62,
+	[1][0][RTW89_ETSI][14] = 34,
+	[1][0][RTW89_MKK][14] = 38,
+	[1][0][RTW89_IC][14] = 62,
+	[1][0][RTW89_KCC][14] = 54,
+	[1][0][RTW89_ACMA][14] = 34,
+	[1][0][RTW89_CN][14] = 26,
+	[1][0][RTW89_UK][14] = 34,
+	[1][0][RTW89_FCC][15] = 60,
+	[1][0][RTW89_ETSI][15] = 34,
+	[1][0][RTW89_MKK][15] = 58,
+	[1][0][RTW89_IC][15] = 60,
+	[1][0][RTW89_KCC][15] = 54,
+	[1][0][RTW89_ACMA][15] = 34,
+	[1][0][RTW89_CN][15] = 127,
+	[1][0][RTW89_UK][15] = 34,
+	[1][0][RTW89_FCC][17] = 60,
+	[1][0][RTW89_ETSI][17] = 34,
+	[1][0][RTW89_MKK][17] = 58,
+	[1][0][RTW89_IC][17] = 60,
+	[1][0][RTW89_KCC][17] = 54,
+	[1][0][RTW89_ACMA][17] = 34,
+	[1][0][RTW89_CN][17] = 127,
+	[1][0][RTW89_UK][17] = 34,
+	[1][0][RTW89_FCC][19] = 62,
+	[1][0][RTW89_ETSI][19] = 34,
+	[1][0][RTW89_MKK][19] = 58,
+	[1][0][RTW89_IC][19] = 62,
+	[1][0][RTW89_KCC][19] = 54,
+	[1][0][RTW89_ACMA][19] = 34,
+	[1][0][RTW89_CN][19] = 127,
+	[1][0][RTW89_UK][19] = 34,
+	[1][0][RTW89_FCC][21] = 62,
+	[1][0][RTW89_ETSI][21] = 34,
+	[1][0][RTW89_MKK][21] = 58,
+	[1][0][RTW89_IC][21] = 62,
+	[1][0][RTW89_KCC][21] = 54,
+	[1][0][RTW89_ACMA][21] = 34,
+	[1][0][RTW89_CN][21] = 127,
+	[1][0][RTW89_UK][21] = 34,
+	[1][0][RTW89_FCC][23] = 62,
+	[1][0][RTW89_ETSI][23] = 34,
+	[1][0][RTW89_MKK][23] = 58,
+	[1][0][RTW89_IC][23] = 62,
+	[1][0][RTW89_KCC][23] = 54,
+	[1][0][RTW89_ACMA][23] = 34,
+	[1][0][RTW89_CN][23] = 127,
+	[1][0][RTW89_UK][23] = 34,
+	[1][0][RTW89_FCC][25] = 62,
+	[1][0][RTW89_ETSI][25] = 34,
+	[1][0][RTW89_MKK][25] = 58,
+	[1][0][RTW89_IC][25] = 127,
+	[1][0][RTW89_KCC][25] = 54,
+	[1][0][RTW89_ACMA][25] = 127,
+	[1][0][RTW89_CN][25] = 127,
+	[1][0][RTW89_UK][25] = 34,
+	[1][0][RTW89_FCC][27] = 62,
+	[1][0][RTW89_ETSI][27] = 34,
+	[1][0][RTW89_MKK][27] = 58,
+	[1][0][RTW89_IC][27] = 127,
+	[1][0][RTW89_KCC][27] = 54,
+	[1][0][RTW89_ACMA][27] = 127,
+	[1][0][RTW89_CN][27] = 127,
+	[1][0][RTW89_UK][27] = 34,
+	[1][0][RTW89_FCC][29] = 62,
+	[1][0][RTW89_ETSI][29] = 34,
+	[1][0][RTW89_MKK][29] = 58,
+	[1][0][RTW89_IC][29] = 127,
+	[1][0][RTW89_KCC][29] = 54,
+	[1][0][RTW89_ACMA][29] = 127,
+	[1][0][RTW89_CN][29] = 127,
+	[1][0][RTW89_UK][29] = 34,
+	[1][0][RTW89_FCC][31] = 62,
+	[1][0][RTW89_ETSI][31] = 34,
+	[1][0][RTW89_MKK][31] = 58,
+	[1][0][RTW89_IC][31] = 64,
+	[1][0][RTW89_KCC][31] = 54,
+	[1][0][RTW89_ACMA][31] = 34,
+	[1][0][RTW89_CN][31] = 127,
+	[1][0][RTW89_UK][31] = 34,
+	[1][0][RTW89_FCC][33] = 64,
+	[1][0][RTW89_ETSI][33] = 34,
+	[1][0][RTW89_MKK][33] = 58,
+	[1][0][RTW89_IC][33] = 64,
+	[1][0][RTW89_KCC][33] = 54,
+	[1][0][RTW89_ACMA][33] = 34,
+	[1][0][RTW89_CN][33] = 127,
+	[1][0][RTW89_UK][33] = 34,
+	[1][0][RTW89_FCC][35] = 64,
+	[1][0][RTW89_ETSI][35] = 34,
+	[1][0][RTW89_MKK][35] = 58,
+	[1][0][RTW89_IC][35] = 64,
+	[1][0][RTW89_KCC][35] = 54,
+	[1][0][RTW89_ACMA][35] = 34,
+	[1][0][RTW89_CN][35] = 127,
+	[1][0][RTW89_UK][35] = 34,
+	[1][0][RTW89_FCC][37] = 76,
+	[1][0][RTW89_ETSI][37] = 127,
+	[1][0][RTW89_MKK][37] = 56,
+	[1][0][RTW89_IC][37] = 76,
+	[1][0][RTW89_KCC][37] = 54,
+	[1][0][RTW89_ACMA][37] = 62,
+	[1][0][RTW89_CN][37] = 127,
+	[1][0][RTW89_UK][37] = 62,
+	[1][0][RTW89_FCC][38] = 82,
+	[1][0][RTW89_ETSI][38] = 28,
+	[1][0][RTW89_MKK][38] = 127,
+	[1][0][RTW89_IC][38] = 82,
+	[1][0][RTW89_KCC][38] = 54,
+	[1][0][RTW89_ACMA][38] = 84,
+	[1][0][RTW89_CN][38] = 66,
+	[1][0][RTW89_UK][38] = 34,
+	[1][0][RTW89_FCC][40] = 82,
+	[1][0][RTW89_ETSI][40] = 28,
+	[1][0][RTW89_MKK][40] = 127,
+	[1][0][RTW89_IC][40] = 82,
+	[1][0][RTW89_KCC][40] = 54,
+	[1][0][RTW89_ACMA][40] = 84,
+	[1][0][RTW89_CN][40] = 66,
+	[1][0][RTW89_UK][40] = 34,
+	[1][0][RTW89_FCC][42] = 78,
+	[1][0][RTW89_ETSI][42] = 28,
+	[1][0][RTW89_MKK][42] = 127,
+	[1][0][RTW89_IC][42] = 78,
+	[1][0][RTW89_KCC][42] = 54,
+	[1][0][RTW89_ACMA][42] = 84,
+	[1][0][RTW89_CN][42] = 66,
+	[1][0][RTW89_UK][42] = 34,
+	[1][0][RTW89_FCC][44] = 82,
+	[1][0][RTW89_ETSI][44] = 28,
+	[1][0][RTW89_MKK][44] = 127,
+	[1][0][RTW89_IC][44] = 82,
+	[1][0][RTW89_KCC][44] = 54,
+	[1][0][RTW89_ACMA][44] = 84,
+	[1][0][RTW89_CN][44] = 66,
+	[1][0][RTW89_UK][44] = 34,
+	[1][0][RTW89_FCC][46] = 82,
+	[1][0][RTW89_ETSI][46] = 28,
+	[1][0][RTW89_MKK][46] = 127,
+	[1][0][RTW89_IC][46] = 82,
+	[1][0][RTW89_KCC][46] = 54,
+	[1][0][RTW89_ACMA][46] = 84,
+	[1][0][RTW89_CN][46] = 66,
+	[1][0][RTW89_UK][46] = 34,
+	[1][0][RTW89_FCC][48] = 52,
+	[1][0][RTW89_ETSI][48] = 127,
+	[1][0][RTW89_MKK][48] = 127,
+	[1][0][RTW89_IC][48] = 127,
+	[1][0][RTW89_KCC][48] = 127,
+	[1][0][RTW89_ACMA][48] = 127,
+	[1][0][RTW89_CN][48] = 127,
+	[1][0][RTW89_UK][48] = 127,
+	[1][0][RTW89_FCC][50] = 52,
+	[1][0][RTW89_ETSI][50] = 127,
+	[1][0][RTW89_MKK][50] = 127,
+	[1][0][RTW89_IC][50] = 127,
+	[1][0][RTW89_KCC][50] = 127,
+	[1][0][RTW89_ACMA][50] = 127,
+	[1][0][RTW89_CN][50] = 127,
+	[1][0][RTW89_UK][50] = 127,
+	[1][0][RTW89_FCC][52] = 50,
+	[1][0][RTW89_ETSI][52] = 127,
+	[1][0][RTW89_MKK][52] = 127,
+	[1][0][RTW89_IC][52] = 127,
+	[1][0][RTW89_KCC][52] = 127,
+	[1][0][RTW89_ACMA][52] = 127,
+	[1][0][RTW89_CN][52] = 127,
+	[1][0][RTW89_UK][52] = 127,
+	[1][1][RTW89_FCC][0] = 127,
+	[1][1][RTW89_ETSI][0] = 127,
+	[1][1][RTW89_MKK][0] = 127,
+	[1][1][RTW89_IC][0] = 127,
+	[1][1][RTW89_KCC][0] = 127,
+	[1][1][RTW89_ACMA][0] = 127,
+	[1][1][RTW89_CN][0] = 14,
+	[1][1][RTW89_UK][0] = 127,
+	[1][1][RTW89_FCC][2] = 127,
+	[1][1][RTW89_ETSI][2] = 127,
+	[1][1][RTW89_MKK][2] = 127,
+	[1][1][RTW89_IC][2] = 127,
+	[1][1][RTW89_KCC][2] = 127,
+	[1][1][RTW89_ACMA][2] = 127,
+	[1][1][RTW89_CN][2] = 14,
+	[1][1][RTW89_UK][2] = 127,
+	[1][1][RTW89_FCC][4] = 127,
+	[1][1][RTW89_ETSI][4] = 127,
+	[1][1][RTW89_MKK][4] = 127,
+	[1][1][RTW89_IC][4] = 127,
+	[1][1][RTW89_KCC][4] = 127,
+	[1][1][RTW89_ACMA][4] = 127,
+	[1][1][RTW89_CN][4] = 14,
+	[1][1][RTW89_UK][4] = 127,
+	[1][1][RTW89_FCC][6] = 127,
+	[1][1][RTW89_ETSI][6] = 127,
+	[1][1][RTW89_MKK][6] = 127,
+	[1][1][RTW89_IC][6] = 127,
+	[1][1][RTW89_KCC][6] = 127,
+	[1][1][RTW89_ACMA][6] = 127,
+	[1][1][RTW89_CN][6] = 14,
+	[1][1][RTW89_UK][6] = 127,
+	[1][1][RTW89_FCC][8] = 127,
+	[1][1][RTW89_ETSI][8] = 127,
+	[1][1][RTW89_MKK][8] = 127,
+	[1][1][RTW89_IC][8] = 127,
+	[1][1][RTW89_KCC][8] = 127,
+	[1][1][RTW89_ACMA][8] = 127,
+	[1][1][RTW89_CN][8] = 14,
+	[1][1][RTW89_UK][8] = 127,
+	[1][1][RTW89_FCC][10] = 127,
+	[1][1][RTW89_ETSI][10] = 127,
+	[1][1][RTW89_MKK][10] = 127,
+	[1][1][RTW89_IC][10] = 127,
+	[1][1][RTW89_KCC][10] = 127,
+	[1][1][RTW89_ACMA][10] = 127,
+	[1][1][RTW89_CN][10] = 14,
+	[1][1][RTW89_UK][10] = 127,
+	[1][1][RTW89_FCC][12] = 127,
+	[1][1][RTW89_ETSI][12] = 127,
+	[1][1][RTW89_MKK][12] = 127,
+	[1][1][RTW89_IC][12] = 127,
+	[1][1][RTW89_KCC][12] = 127,
+	[1][1][RTW89_ACMA][12] = 127,
+	[1][1][RTW89_CN][12] = 14,
+	[1][1][RTW89_UK][12] = 127,
+	[1][1][RTW89_FCC][14] = 127,
+	[1][1][RTW89_ETSI][14] = 127,
+	[1][1][RTW89_MKK][14] = 127,
+	[1][1][RTW89_IC][14] = 127,
+	[1][1][RTW89_KCC][14] = 127,
+	[1][1][RTW89_ACMA][14] = 127,
+	[1][1][RTW89_CN][14] = 14,
+	[1][1][RTW89_UK][14] = 127,
+	[1][1][RTW89_FCC][15] = 127,
+	[1][1][RTW89_ETSI][15] = 127,
+	[1][1][RTW89_MKK][15] = 127,
+	[1][1][RTW89_IC][15] = 127,
+	[1][1][RTW89_KCC][15] = 127,
+	[1][1][RTW89_ACMA][15] = 127,
+	[1][1][RTW89_CN][15] = 127,
+	[1][1][RTW89_UK][15] = 127,
+	[1][1][RTW89_FCC][17] = 127,
+	[1][1][RTW89_ETSI][17] = 127,
+	[1][1][RTW89_MKK][17] = 127,
+	[1][1][RTW89_IC][17] = 127,
+	[1][1][RTW89_KCC][17] = 127,
+	[1][1][RTW89_ACMA][17] = 127,
+	[1][1][RTW89_CN][17] = 127,
+	[1][1][RTW89_UK][17] = 127,
+	[1][1][RTW89_FCC][19] = 127,
+	[1][1][RTW89_ETSI][19] = 127,
+	[1][1][RTW89_MKK][19] = 127,
+	[1][1][RTW89_IC][19] = 127,
+	[1][1][RTW89_KCC][19] = 127,
+	[1][1][RTW89_ACMA][19] = 127,
+	[1][1][RTW89_CN][19] = 127,
+	[1][1][RTW89_UK][19] = 127,
+	[1][1][RTW89_FCC][21] = 127,
+	[1][1][RTW89_ETSI][21] = 127,
+	[1][1][RTW89_MKK][21] = 127,
+	[1][1][RTW89_IC][21] = 127,
+	[1][1][RTW89_KCC][21] = 127,
+	[1][1][RTW89_ACMA][21] = 127,
+	[1][1][RTW89_CN][21] = 127,
+	[1][1][RTW89_UK][21] = 127,
+	[1][1][RTW89_FCC][23] = 127,
+	[1][1][RTW89_ETSI][23] = 127,
+	[1][1][RTW89_MKK][23] = 127,
+	[1][1][RTW89_IC][23] = 127,
+	[1][1][RTW89_KCC][23] = 127,
+	[1][1][RTW89_ACMA][23] = 127,
+	[1][1][RTW89_CN][23] = 127,
+	[1][1][RTW89_UK][23] = 127,
+	[1][1][RTW89_FCC][25] = 127,
+	[1][1][RTW89_ETSI][25] = 127,
+	[1][1][RTW89_MKK][25] = 127,
+	[1][1][RTW89_IC][25] = 127,
+	[1][1][RTW89_KCC][25] = 127,
+	[1][1][RTW89_ACMA][25] = 127,
+	[1][1][RTW89_CN][25] = 127,
+	[1][1][RTW89_UK][25] = 127,
+	[1][1][RTW89_FCC][27] = 127,
+	[1][1][RTW89_ETSI][27] = 127,
+	[1][1][RTW89_MKK][27] = 127,
+	[1][1][RTW89_IC][27] = 127,
+	[1][1][RTW89_KCC][27] = 127,
+	[1][1][RTW89_ACMA][27] = 127,
+	[1][1][RTW89_CN][27] = 127,
+	[1][1][RTW89_UK][27] = 127,
+	[1][1][RTW89_FCC][29] = 127,
+	[1][1][RTW89_ETSI][29] = 127,
+	[1][1][RTW89_MKK][29] = 127,
+	[1][1][RTW89_IC][29] = 127,
+	[1][1][RTW89_KCC][29] = 127,
+	[1][1][RTW89_ACMA][29] = 127,
+	[1][1][RTW89_CN][29] = 127,
+	[1][1][RTW89_UK][29] = 127,
+	[1][1][RTW89_FCC][31] = 127,
+	[1][1][RTW89_ETSI][31] = 127,
+	[1][1][RTW89_MKK][31] = 127,
+	[1][1][RTW89_IC][31] = 127,
+	[1][1][RTW89_KCC][31] = 127,
+	[1][1][RTW89_ACMA][31] = 127,
+	[1][1][RTW89_CN][31] = 127,
+	[1][1][RTW89_UK][31] = 127,
+	[1][1][RTW89_FCC][33] = 127,
+	[1][1][RTW89_ETSI][33] = 127,
+	[1][1][RTW89_MKK][33] = 127,
+	[1][1][RTW89_IC][33] = 127,
+	[1][1][RTW89_KCC][33] = 127,
+	[1][1][RTW89_ACMA][33] = 127,
+	[1][1][RTW89_CN][33] = 127,
+	[1][1][RTW89_UK][33] = 127,
+	[1][1][RTW89_FCC][35] = 127,
+	[1][1][RTW89_ETSI][35] = 127,
+	[1][1][RTW89_MKK][35] = 127,
+	[1][1][RTW89_IC][35] = 127,
+	[1][1][RTW89_KCC][35] = 127,
+	[1][1][RTW89_ACMA][35] = 127,
+	[1][1][RTW89_CN][35] = 127,
+	[1][1][RTW89_UK][35] = 127,
+	[1][1][RTW89_FCC][37] = 127,
+	[1][1][RTW89_ETSI][37] = 127,
+	[1][1][RTW89_MKK][37] = 127,
+	[1][1][RTW89_IC][37] = 127,
+	[1][1][RTW89_KCC][37] = 127,
+	[1][1][RTW89_ACMA][37] = 127,
+	[1][1][RTW89_CN][37] = 127,
+	[1][1][RTW89_UK][37] = 127,
+	[1][1][RTW89_FCC][38] = 127,
+	[1][1][RTW89_ETSI][38] = 127,
+	[1][1][RTW89_MKK][38] = 127,
+	[1][1][RTW89_IC][38] = 127,
+	[1][1][RTW89_KCC][38] = 127,
+	[1][1][RTW89_ACMA][38] = 127,
+	[1][1][RTW89_CN][38] = 54,
+	[1][1][RTW89_UK][38] = 127,
+	[1][1][RTW89_FCC][40] = 127,
+	[1][1][RTW89_ETSI][40] = 127,
+	[1][1][RTW89_MKK][40] = 127,
+	[1][1][RTW89_IC][40] = 127,
+	[1][1][RTW89_KCC][40] = 127,
+	[1][1][RTW89_ACMA][40] = 127,
+	[1][1][RTW89_CN][40] = 54,
+	[1][1][RTW89_UK][40] = 127,
+	[1][1][RTW89_FCC][42] = 127,
+	[1][1][RTW89_ETSI][42] = 127,
+	[1][1][RTW89_MKK][42] = 127,
+	[1][1][RTW89_IC][42] = 127,
+	[1][1][RTW89_KCC][42] = 127,
+	[1][1][RTW89_ACMA][42] = 127,
+	[1][1][RTW89_CN][42] = 54,
+	[1][1][RTW89_UK][42] = 127,
+	[1][1][RTW89_FCC][44] = 127,
+	[1][1][RTW89_ETSI][44] = 127,
+	[1][1][RTW89_MKK][44] = 127,
+	[1][1][RTW89_IC][44] = 127,
+	[1][1][RTW89_KCC][44] = 127,
+	[1][1][RTW89_ACMA][44] = 127,
+	[1][1][RTW89_CN][44] = 54,
+	[1][1][RTW89_UK][44] = 127,
+	[1][1][RTW89_FCC][46] = 127,
+	[1][1][RTW89_ETSI][46] = 127,
+	[1][1][RTW89_MKK][46] = 127,
+	[1][1][RTW89_IC][46] = 127,
+	[1][1][RTW89_KCC][46] = 127,
+	[1][1][RTW89_ACMA][46] = 127,
+	[1][1][RTW89_CN][46] = 54,
+	[1][1][RTW89_UK][46] = 127,
+	[1][1][RTW89_FCC][48] = 127,
+	[1][1][RTW89_ETSI][48] = 127,
+	[1][1][RTW89_MKK][48] = 127,
+	[1][1][RTW89_IC][48] = 127,
+	[1][1][RTW89_KCC][48] = 127,
+	[1][1][RTW89_ACMA][48] = 127,
+	[1][1][RTW89_CN][48] = 127,
+	[1][1][RTW89_UK][48] = 127,
+	[1][1][RTW89_FCC][50] = 127,
+	[1][1][RTW89_ETSI][50] = 127,
+	[1][1][RTW89_MKK][50] = 127,
+	[1][1][RTW89_IC][50] = 127,
+	[1][1][RTW89_KCC][50] = 127,
+	[1][1][RTW89_ACMA][50] = 127,
+	[1][1][RTW89_CN][50] = 127,
+	[1][1][RTW89_UK][50] = 127,
+	[1][1][RTW89_FCC][52] = 127,
+	[1][1][RTW89_ETSI][52] = 127,
+	[1][1][RTW89_MKK][52] = 127,
+	[1][1][RTW89_IC][52] = 127,
+	[1][1][RTW89_KCC][52] = 127,
+	[1][1][RTW89_ACMA][52] = 127,
+	[1][1][RTW89_CN][52] = 127,
+	[1][1][RTW89_UK][52] = 127,
+	[2][0][RTW89_FCC][0] = 76,
+	[2][0][RTW89_ETSI][0] = 46,
+	[2][0][RTW89_MKK][0] = 48,
+	[2][0][RTW89_IC][0] = 50,
+	[2][0][RTW89_KCC][0] = 64,
+	[2][0][RTW89_ACMA][0] = 46,
+	[2][0][RTW89_CN][0] = 40,
+	[2][0][RTW89_UK][0] = 46,
+	[2][0][RTW89_FCC][2] = 72,
+	[2][0][RTW89_ETSI][2] = 46,
+	[2][0][RTW89_MKK][2] = 48,
+	[2][0][RTW89_IC][2] = 48,
+	[2][0][RTW89_KCC][2] = 64,
+	[2][0][RTW89_ACMA][2] = 46,
+	[2][0][RTW89_CN][2] = 40,
+	[2][0][RTW89_UK][2] = 46,
+	[2][0][RTW89_FCC][4] = 74,
+	[2][0][RTW89_ETSI][4] = 46,
+	[2][0][RTW89_MKK][4] = 48,
+	[2][0][RTW89_IC][4] = 48,
+	[2][0][RTW89_KCC][4] = 64,
+	[2][0][RTW89_ACMA][4] = 46,
+	[2][0][RTW89_CN][4] = 40,
+	[2][0][RTW89_UK][4] = 46,
+	[2][0][RTW89_FCC][6] = 74,
+	[2][0][RTW89_ETSI][6] = 46,
+	[2][0][RTW89_MKK][6] = 48,
+	[2][0][RTW89_IC][6] = 48,
+	[2][0][RTW89_KCC][6] = 40,
+	[2][0][RTW89_ACMA][6] = 46,
+	[2][0][RTW89_CN][6] = 40,
+	[2][0][RTW89_UK][6] = 46,
+	[2][0][RTW89_FCC][8] = 72,
+	[2][0][RTW89_ETSI][8] = 46,
+	[2][0][RTW89_MKK][8] = 48,
+	[2][0][RTW89_IC][8] = 64,
+	[2][0][RTW89_KCC][8] = 66,
+	[2][0][RTW89_ACMA][8] = 46,
+	[2][0][RTW89_CN][8] = 40,
+	[2][0][RTW89_UK][8] = 46,
+	[2][0][RTW89_FCC][10] = 72,
+	[2][0][RTW89_ETSI][10] = 46,
+	[2][0][RTW89_MKK][10] = 48,
+	[2][0][RTW89_IC][10] = 64,
+	[2][0][RTW89_KCC][10] = 66,
+	[2][0][RTW89_ACMA][10] = 46,
+	[2][0][RTW89_CN][10] = 40,
+	[2][0][RTW89_UK][10] = 46,
+	[2][0][RTW89_FCC][12] = 74,
+	[2][0][RTW89_ETSI][12] = 46,
+	[2][0][RTW89_MKK][12] = 48,
+	[2][0][RTW89_IC][12] = 64,
+	[2][0][RTW89_KCC][12] = 64,
+	[2][0][RTW89_ACMA][12] = 46,
+	[2][0][RTW89_CN][12] = 40,
+	[2][0][RTW89_UK][12] = 46,
+	[2][0][RTW89_FCC][14] = 80,
+	[2][0][RTW89_ETSI][14] = 46,
+	[2][0][RTW89_MKK][14] = 48,
+	[2][0][RTW89_IC][14] = 64,
+	[2][0][RTW89_KCC][14] = 64,
+	[2][0][RTW89_ACMA][14] = 46,
+	[2][0][RTW89_CN][14] = 40,
+	[2][0][RTW89_UK][14] = 46,
+	[2][0][RTW89_FCC][15] = 72,
+	[2][0][RTW89_ETSI][15] = 46,
+	[2][0][RTW89_MKK][15] = 70,
+	[2][0][RTW89_IC][15] = 72,
+	[2][0][RTW89_KCC][15] = 66,
+	[2][0][RTW89_ACMA][15] = 46,
+	[2][0][RTW89_CN][15] = 127,
+	[2][0][RTW89_UK][15] = 46,
+	[2][0][RTW89_FCC][17] = 72,
+	[2][0][RTW89_ETSI][17] = 46,
+	[2][0][RTW89_MKK][17] = 70,
+	[2][0][RTW89_IC][17] = 72,
+	[2][0][RTW89_KCC][17] = 66,
+	[2][0][RTW89_ACMA][17] = 46,
+	[2][0][RTW89_CN][17] = 127,
+	[2][0][RTW89_UK][17] = 46,
+	[2][0][RTW89_FCC][19] = 68,
+	[2][0][RTW89_ETSI][19] = 46,
+	[2][0][RTW89_MKK][19] = 70,
+	[2][0][RTW89_IC][19] = 68,
+	[2][0][RTW89_KCC][19] = 66,
+	[2][0][RTW89_ACMA][19] = 46,
+	[2][0][RTW89_CN][19] = 127,
+	[2][0][RTW89_UK][19] = 46,
+	[2][0][RTW89_FCC][21] = 68,
+	[2][0][RTW89_ETSI][21] = 46,
+	[2][0][RTW89_MKK][21] = 70,
+	[2][0][RTW89_IC][21] = 68,
+	[2][0][RTW89_KCC][21] = 66,
+	[2][0][RTW89_ACMA][21] = 46,
+	[2][0][RTW89_CN][21] = 127,
+	[2][0][RTW89_UK][21] = 46,
+	[2][0][RTW89_FCC][23] = 68,
+	[2][0][RTW89_ETSI][23] = 46,
+	[2][0][RTW89_MKK][23] = 70,
+	[2][0][RTW89_IC][23] = 68,
+	[2][0][RTW89_KCC][23] = 66,
+	[2][0][RTW89_ACMA][23] = 46,
+	[2][0][RTW89_CN][23] = 127,
+	[2][0][RTW89_UK][23] = 46,
+	[2][0][RTW89_FCC][25] = 68,
+	[2][0][RTW89_ETSI][25] = 46,
+	[2][0][RTW89_MKK][25] = 70,
+	[2][0][RTW89_IC][25] = 127,
+	[2][0][RTW89_KCC][25] = 66,
+	[2][0][RTW89_ACMA][25] = 127,
+	[2][0][RTW89_CN][25] = 127,
+	[2][0][RTW89_UK][25] = 46,
+	[2][0][RTW89_FCC][27] = 68,
+	[2][0][RTW89_ETSI][27] = 46,
+	[2][0][RTW89_MKK][27] = 70,
+	[2][0][RTW89_IC][27] = 127,
+	[2][0][RTW89_KCC][27] = 64,
+	[2][0][RTW89_ACMA][27] = 127,
+	[2][0][RTW89_CN][27] = 127,
+	[2][0][RTW89_UK][27] = 46,
+	[2][0][RTW89_FCC][29] = 68,
+	[2][0][RTW89_ETSI][29] = 46,
+	[2][0][RTW89_MKK][29] = 70,
+	[2][0][RTW89_IC][29] = 127,
+	[2][0][RTW89_KCC][29] = 64,
+	[2][0][RTW89_ACMA][29] = 127,
+	[2][0][RTW89_CN][29] = 127,
+	[2][0][RTW89_UK][29] = 46,
+	[2][0][RTW89_FCC][31] = 68,
+	[2][0][RTW89_ETSI][31] = 46,
+	[2][0][RTW89_MKK][31] = 70,
+	[2][0][RTW89_IC][31] = 70,
+	[2][0][RTW89_KCC][31] = 64,
+	[2][0][RTW89_ACMA][31] = 46,
+	[2][0][RTW89_CN][31] = 127,
+	[2][0][RTW89_UK][31] = 46,
+	[2][0][RTW89_FCC][33] = 70,
+	[2][0][RTW89_ETSI][33] = 46,
+	[2][0][RTW89_MKK][33] = 70,
+	[2][0][RTW89_IC][33] = 70,
+	[2][0][RTW89_KCC][33] = 64,
+	[2][0][RTW89_ACMA][33] = 46,
+	[2][0][RTW89_CN][33] = 127,
+	[2][0][RTW89_UK][33] = 46,
+	[2][0][RTW89_FCC][35] = 70,
+	[2][0][RTW89_ETSI][35] = 46,
+	[2][0][RTW89_MKK][35] = 70,
+	[2][0][RTW89_IC][35] = 70,
+	[2][0][RTW89_KCC][35] = 64,
+	[2][0][RTW89_ACMA][35] = 46,
+	[2][0][RTW89_CN][35] = 127,
+	[2][0][RTW89_UK][35] = 46,
+	[2][0][RTW89_FCC][37] = 84,
+	[2][0][RTW89_ETSI][37] = 127,
+	[2][0][RTW89_MKK][37] = 68,
+	[2][0][RTW89_IC][37] = 84,
+	[2][0][RTW89_KCC][37] = 66,
+	[2][0][RTW89_ACMA][37] = 74,
+	[2][0][RTW89_CN][37] = 127,
+	[2][0][RTW89_UK][37] = 74,
+	[2][0][RTW89_FCC][38] = 84,
+	[2][0][RTW89_ETSI][38] = 28,
+	[2][0][RTW89_MKK][38] = 127,
+	[2][0][RTW89_IC][38] = 84,
+	[2][0][RTW89_KCC][38] = 64,
+	[2][0][RTW89_ACMA][38] = 84,
+	[2][0][RTW89_CN][38] = 68,
+	[2][0][RTW89_UK][38] = 46,
+	[2][0][RTW89_FCC][40] = 84,
+	[2][0][RTW89_ETSI][40] = 28,
+	[2][0][RTW89_MKK][40] = 127,
+	[2][0][RTW89_IC][40] = 84,
+	[2][0][RTW89_KCC][40] = 64,
+	[2][0][RTW89_ACMA][40] = 84,
+	[2][0][RTW89_CN][40] = 68,
+	[2][0][RTW89_UK][40] = 46,
+	[2][0][RTW89_FCC][42] = 78,
+	[2][0][RTW89_ETSI][42] = 28,
+	[2][0][RTW89_MKK][42] = 127,
+	[2][0][RTW89_IC][42] = 78,
+	[2][0][RTW89_KCC][42] = 66,
+	[2][0][RTW89_ACMA][42] = 84,
+	[2][0][RTW89_CN][42] = 68,
+	[2][0][RTW89_UK][42] = 46,
+	[2][0][RTW89_FCC][44] = 80,
+	[2][0][RTW89_ETSI][44] = 28,
+	[2][0][RTW89_MKK][44] = 127,
+	[2][0][RTW89_IC][44] = 80,
+	[2][0][RTW89_KCC][44] = 66,
+	[2][0][RTW89_ACMA][44] = 84,
+	[2][0][RTW89_CN][44] = 68,
+	[2][0][RTW89_UK][44] = 46,
+	[2][0][RTW89_FCC][46] = 80,
+	[2][0][RTW89_ETSI][46] = 28,
+	[2][0][RTW89_MKK][46] = 127,
+	[2][0][RTW89_IC][46] = 80,
+	[2][0][RTW89_KCC][46] = 66,
+	[2][0][RTW89_ACMA][46] = 84,
+	[2][0][RTW89_CN][46] = 68,
+	[2][0][RTW89_UK][46] = 46,
+	[2][0][RTW89_FCC][48] = 62,
+	[2][0][RTW89_ETSI][48] = 127,
+	[2][0][RTW89_MKK][48] = 127,
+	[2][0][RTW89_IC][48] = 127,
+	[2][0][RTW89_KCC][48] = 127,
+	[2][0][RTW89_ACMA][48] = 127,
+	[2][0][RTW89_CN][48] = 127,
+	[2][0][RTW89_UK][48] = 127,
+	[2][0][RTW89_FCC][50] = 62,
+	[2][0][RTW89_ETSI][50] = 127,
+	[2][0][RTW89_MKK][50] = 127,
+	[2][0][RTW89_IC][50] = 127,
+	[2][0][RTW89_KCC][50] = 127,
+	[2][0][RTW89_ACMA][50] = 127,
+	[2][0][RTW89_CN][50] = 127,
+	[2][0][RTW89_UK][50] = 127,
+	[2][0][RTW89_FCC][52] = 60,
+	[2][0][RTW89_ETSI][52] = 127,
+	[2][0][RTW89_MKK][52] = 127,
+	[2][0][RTW89_IC][52] = 127,
+	[2][0][RTW89_KCC][52] = 127,
+	[2][0][RTW89_ACMA][52] = 127,
+	[2][0][RTW89_CN][52] = 127,
+	[2][0][RTW89_UK][52] = 127,
+	[2][1][RTW89_FCC][0] = 127,
+	[2][1][RTW89_ETSI][0] = 127,
+	[2][1][RTW89_MKK][0] = 127,
+	[2][1][RTW89_IC][0] = 127,
+	[2][1][RTW89_KCC][0] = 127,
+	[2][1][RTW89_ACMA][0] = 127,
+	[2][1][RTW89_CN][0] = 28,
+	[2][1][RTW89_UK][0] = 127,
+	[2][1][RTW89_FCC][2] = 127,
+	[2][1][RTW89_ETSI][2] = 127,
+	[2][1][RTW89_MKK][2] = 127,
+	[2][1][RTW89_IC][2] = 127,
+	[2][1][RTW89_KCC][2] = 127,
+	[2][1][RTW89_ACMA][2] = 127,
+	[2][1][RTW89_CN][2] = 28,
+	[2][1][RTW89_UK][2] = 127,
+	[2][1][RTW89_FCC][4] = 127,
+	[2][1][RTW89_ETSI][4] = 127,
+	[2][1][RTW89_MKK][4] = 127,
+	[2][1][RTW89_IC][4] = 127,
+	[2][1][RTW89_KCC][4] = 127,
+	[2][1][RTW89_ACMA][4] = 127,
+	[2][1][RTW89_CN][4] = 28,
+	[2][1][RTW89_UK][4] = 127,
+	[2][1][RTW89_FCC][6] = 127,
+	[2][1][RTW89_ETSI][6] = 127,
+	[2][1][RTW89_MKK][6] = 127,
+	[2][1][RTW89_IC][6] = 127,
+	[2][1][RTW89_KCC][6] = 127,
+	[2][1][RTW89_ACMA][6] = 127,
+	[2][1][RTW89_CN][6] = 28,
+	[2][1][RTW89_UK][6] = 127,
+	[2][1][RTW89_FCC][8] = 127,
+	[2][1][RTW89_ETSI][8] = 127,
+	[2][1][RTW89_MKK][8] = 127,
+	[2][1][RTW89_IC][8] = 127,
+	[2][1][RTW89_KCC][8] = 127,
+	[2][1][RTW89_ACMA][8] = 127,
+	[2][1][RTW89_CN][8] = 28,
+	[2][1][RTW89_UK][8] = 127,
+	[2][1][RTW89_FCC][10] = 127,
+	[2][1][RTW89_ETSI][10] = 127,
+	[2][1][RTW89_MKK][10] = 127,
+	[2][1][RTW89_IC][10] = 127,
+	[2][1][RTW89_KCC][10] = 127,
+	[2][1][RTW89_ACMA][10] = 127,
+	[2][1][RTW89_CN][10] = 28,
+	[2][1][RTW89_UK][10] = 127,
+	[2][1][RTW89_FCC][12] = 127,
+	[2][1][RTW89_ETSI][12] = 127,
+	[2][1][RTW89_MKK][12] = 127,
+	[2][1][RTW89_IC][12] = 127,
+	[2][1][RTW89_KCC][12] = 127,
+	[2][1][RTW89_ACMA][12] = 127,
+	[2][1][RTW89_CN][12] = 28,
+	[2][1][RTW89_UK][12] = 127,
+	[2][1][RTW89_FCC][14] = 127,
+	[2][1][RTW89_ETSI][14] = 127,
+	[2][1][RTW89_MKK][14] = 127,
+	[2][1][RTW89_IC][14] = 127,
+	[2][1][RTW89_KCC][14] = 127,
+	[2][1][RTW89_ACMA][14] = 127,
+	[2][1][RTW89_CN][14] = 28,
+	[2][1][RTW89_UK][14] = 127,
+	[2][1][RTW89_FCC][15] = 127,
+	[2][1][RTW89_ETSI][15] = 127,
+	[2][1][RTW89_MKK][15] = 127,
+	[2][1][RTW89_IC][15] = 127,
+	[2][1][RTW89_KCC][15] = 127,
+	[2][1][RTW89_ACMA][15] = 127,
+	[2][1][RTW89_CN][15] = 127,
+	[2][1][RTW89_UK][15] = 127,
+	[2][1][RTW89_FCC][17] = 127,
+	[2][1][RTW89_ETSI][17] = 127,
+	[2][1][RTW89_MKK][17] = 127,
+	[2][1][RTW89_IC][17] = 127,
+	[2][1][RTW89_KCC][17] = 127,
+	[2][1][RTW89_ACMA][17] = 127,
+	[2][1][RTW89_CN][17] = 127,
+	[2][1][RTW89_UK][17] = 127,
+	[2][1][RTW89_FCC][19] = 127,
+	[2][1][RTW89_ETSI][19] = 127,
+	[2][1][RTW89_MKK][19] = 127,
+	[2][1][RTW89_IC][19] = 127,
+	[2][1][RTW89_KCC][19] = 127,
+	[2][1][RTW89_ACMA][19] = 127,
+	[2][1][RTW89_CN][19] = 127,
+	[2][1][RTW89_UK][19] = 127,
+	[2][1][RTW89_FCC][21] = 127,
+	[2][1][RTW89_ETSI][21] = 127,
+	[2][1][RTW89_MKK][21] = 127,
+	[2][1][RTW89_IC][21] = 127,
+	[2][1][RTW89_KCC][21] = 127,
+	[2][1][RTW89_ACMA][21] = 127,
+	[2][1][RTW89_CN][21] = 127,
+	[2][1][RTW89_UK][21] = 127,
+	[2][1][RTW89_FCC][23] = 127,
+	[2][1][RTW89_ETSI][23] = 127,
+	[2][1][RTW89_MKK][23] = 127,
+	[2][1][RTW89_IC][23] = 127,
+	[2][1][RTW89_KCC][23] = 127,
+	[2][1][RTW89_ACMA][23] = 127,
+	[2][1][RTW89_CN][23] = 127,
+	[2][1][RTW89_UK][23] = 127,
+	[2][1][RTW89_FCC][25] = 127,
+	[2][1][RTW89_ETSI][25] = 127,
+	[2][1][RTW89_MKK][25] = 127,
+	[2][1][RTW89_IC][25] = 127,
+	[2][1][RTW89_KCC][25] = 127,
+	[2][1][RTW89_ACMA][25] = 127,
+	[2][1][RTW89_CN][25] = 127,
+	[2][1][RTW89_UK][25] = 127,
+	[2][1][RTW89_FCC][27] = 127,
+	[2][1][RTW89_ETSI][27] = 127,
+	[2][1][RTW89_MKK][27] = 127,
+	[2][1][RTW89_IC][27] = 127,
+	[2][1][RTW89_KCC][27] = 127,
+	[2][1][RTW89_ACMA][27] = 127,
+	[2][1][RTW89_CN][27] = 127,
+	[2][1][RTW89_UK][27] = 127,
+	[2][1][RTW89_FCC][29] = 127,
+	[2][1][RTW89_ETSI][29] = 127,
+	[2][1][RTW89_MKK][29] = 127,
+	[2][1][RTW89_IC][29] = 127,
+	[2][1][RTW89_KCC][29] = 127,
+	[2][1][RTW89_ACMA][29] = 127,
+	[2][1][RTW89_CN][29] = 127,
+	[2][1][RTW89_UK][29] = 127,
+	[2][1][RTW89_FCC][31] = 127,
+	[2][1][RTW89_ETSI][31] = 127,
+	[2][1][RTW89_MKK][31] = 127,
+	[2][1][RTW89_IC][31] = 127,
+	[2][1][RTW89_KCC][31] = 127,
+	[2][1][RTW89_ACMA][31] = 127,
+	[2][1][RTW89_CN][31] = 127,
+	[2][1][RTW89_UK][31] = 127,
+	[2][1][RTW89_FCC][33] = 127,
+	[2][1][RTW89_ETSI][33] = 127,
+	[2][1][RTW89_MKK][33] = 127,
+	[2][1][RTW89_IC][33] = 127,
+	[2][1][RTW89_KCC][33] = 127,
+	[2][1][RTW89_ACMA][33] = 127,
+	[2][1][RTW89_CN][33] = 127,
+	[2][1][RTW89_UK][33] = 127,
+	[2][1][RTW89_FCC][35] = 127,
+	[2][1][RTW89_ETSI][35] = 127,
+	[2][1][RTW89_MKK][35] = 127,
+	[2][1][RTW89_IC][35] = 127,
+	[2][1][RTW89_KCC][35] = 127,
+	[2][1][RTW89_ACMA][35] = 127,
+	[2][1][RTW89_CN][35] = 127,
+	[2][1][RTW89_UK][35] = 127,
+	[2][1][RTW89_FCC][37] = 127,
+	[2][1][RTW89_ETSI][37] = 127,
+	[2][1][RTW89_MKK][37] = 127,
+	[2][1][RTW89_IC][37] = 127,
+	[2][1][RTW89_KCC][37] = 127,
+	[2][1][RTW89_ACMA][37] = 127,
+	[2][1][RTW89_CN][37] = 127,
+	[2][1][RTW89_UK][37] = 127,
+	[2][1][RTW89_FCC][38] = 127,
+	[2][1][RTW89_ETSI][38] = 127,
+	[2][1][RTW89_MKK][38] = 127,
+	[2][1][RTW89_IC][38] = 127,
+	[2][1][RTW89_KCC][38] = 127,
+	[2][1][RTW89_ACMA][38] = 127,
+	[2][1][RTW89_CN][38] = 56,
+	[2][1][RTW89_UK][38] = 127,
+	[2][1][RTW89_FCC][40] = 127,
+	[2][1][RTW89_ETSI][40] = 127,
+	[2][1][RTW89_MKK][40] = 127,
+	[2][1][RTW89_IC][40] = 127,
+	[2][1][RTW89_KCC][40] = 127,
+	[2][1][RTW89_ACMA][40] = 127,
+	[2][1][RTW89_CN][40] = 56,
+	[2][1][RTW89_UK][40] = 127,
+	[2][1][RTW89_FCC][42] = 127,
+	[2][1][RTW89_ETSI][42] = 127,
+	[2][1][RTW89_MKK][42] = 127,
+	[2][1][RTW89_IC][42] = 127,
+	[2][1][RTW89_KCC][42] = 127,
+	[2][1][RTW89_ACMA][42] = 127,
+	[2][1][RTW89_CN][42] = 56,
+	[2][1][RTW89_UK][42] = 127,
+	[2][1][RTW89_FCC][44] = 127,
+	[2][1][RTW89_ETSI][44] = 127,
+	[2][1][RTW89_MKK][44] = 127,
+	[2][1][RTW89_IC][44] = 127,
+	[2][1][RTW89_KCC][44] = 127,
+	[2][1][RTW89_ACMA][44] = 127,
+	[2][1][RTW89_CN][44] = 56,
+	[2][1][RTW89_UK][44] = 127,
+	[2][1][RTW89_FCC][46] = 127,
+	[2][1][RTW89_ETSI][46] = 127,
+	[2][1][RTW89_MKK][46] = 127,
+	[2][1][RTW89_IC][46] = 127,
+	[2][1][RTW89_KCC][46] = 127,
+	[2][1][RTW89_ACMA][46] = 127,
+	[2][1][RTW89_CN][46] = 56,
+	[2][1][RTW89_UK][46] = 127,
+	[2][1][RTW89_FCC][48] = 127,
+	[2][1][RTW89_ETSI][48] = 127,
+	[2][1][RTW89_MKK][48] = 127,
+	[2][1][RTW89_IC][48] = 127,
+	[2][1][RTW89_KCC][48] = 127,
+	[2][1][RTW89_ACMA][48] = 127,
+	[2][1][RTW89_CN][48] = 127,
+	[2][1][RTW89_UK][48] = 127,
+	[2][1][RTW89_FCC][50] = 127,
+	[2][1][RTW89_ETSI][50] = 127,
+	[2][1][RTW89_MKK][50] = 127,
+	[2][1][RTW89_IC][50] = 127,
+	[2][1][RTW89_KCC][50] = 127,
+	[2][1][RTW89_ACMA][50] = 127,
+	[2][1][RTW89_CN][50] = 127,
+	[2][1][RTW89_UK][50] = 127,
+	[2][1][RTW89_FCC][52] = 127,
+	[2][1][RTW89_ETSI][52] = 127,
+	[2][1][RTW89_MKK][52] = 127,
+	[2][1][RTW89_IC][52] = 127,
+	[2][1][RTW89_KCC][52] = 127,
+	[2][1][RTW89_ACMA][52] = 127,
+	[2][1][RTW89_CN][52] = 127,
+	[2][1][RTW89_UK][52] = 127,
+};
+
+const struct rtw89_phy_table rtw89_8851b_phy_bb_table = {
+	.regs		= rtw89_8851b_phy_bb_regs,
+	.n_regs		= ARRAY_SIZE(rtw89_8851b_phy_bb_regs),
+	.rf_path	= 0, /* don't care */
+};
+
+const struct rtw89_phy_table rtw89_8851b_phy_bb_gain_table = {
+	.regs		= rtw89_8851b_phy_bb_reg_gain,
+	.n_regs		= ARRAY_SIZE(rtw89_8851b_phy_bb_reg_gain),
+	.rf_path	= 0, /* don't care */
+};
+
+const struct rtw89_phy_table rtw89_8851b_phy_radioa_table = {
+	.regs		= rtw89_8851b_phy_radioa_regs,
+	.n_regs		= ARRAY_SIZE(rtw89_8851b_phy_radioa_regs),
+	.rf_path	= RF_PATH_A,
+	.config		= rtw89_phy_config_rf_reg_v1,
+};
+
+const struct rtw89_phy_table rtw89_8851b_phy_nctl_table = {
+	.regs		= rtw89_8851b_phy_nctl_regs,
+	.n_regs		= ARRAY_SIZE(rtw89_8851b_phy_nctl_regs),
+	.rf_path	= 0, /* don't care */
+};
+
+const struct rtw89_txpwr_table rtw89_8851b_byr_table = {
+	.data = rtw89_8851b_txpwr_byrate,
+	.size = ARRAY_SIZE(rtw89_8851b_txpwr_byrate),
+	.load = rtw89_phy_load_txpwr_byrate,
+};
+
+const struct rtw89_txpwr_track_cfg rtw89_8851b_trk_cfg = {
+	.delta_swingidx_5ga_n = _txpwr_track_delta_swingidx_5ga_n,
+	.delta_swingidx_5ga_p = _txpwr_track_delta_swingidx_5ga_p,
+	.delta_swingidx_2ga_n = _txpwr_track_delta_swingidx_2ga_n,
+	.delta_swingidx_2ga_p = _txpwr_track_delta_swingidx_2ga_p,
+	.delta_swingidx_2g_cck_a_n = _txpwr_track_delta_swingidx_2g_cck_a_n,
+	.delta_swingidx_2g_cck_a_p = _txpwr_track_delta_swingidx_2g_cck_a_p,
+};
+
+const struct rtw89_rfe_parms rtw89_8851b_dflt_parms = {
+	.rule_2ghz = {
+		.lmt = &rtw89_8851b_txpwr_lmt_2g,
+		.lmt_ru = &rtw89_8851b_txpwr_lmt_ru_2g,
+	},
+	.rule_5ghz = {
+		.lmt = &rtw89_8851b_txpwr_lmt_5g,
+		.lmt_ru = &rtw89_8851b_txpwr_lmt_ru_5g,
+	},
+};
+
+static const struct rtw89_rfe_parms rtw89_8851b_rfe_parms_type2 = {
+	.rule_2ghz = {
+		.lmt = &rtw89_8851b_txpwr_lmt_2g_type2,
+		.lmt_ru = &rtw89_8851b_txpwr_lmt_ru_2g_type2,
+	},
+	.rule_5ghz = {
+		.lmt = &rtw89_8851b_txpwr_lmt_5g_type2,
+		.lmt_ru = &rtw89_8851b_txpwr_lmt_ru_5g_type2,
+	},
+};
+
+const struct rtw89_rfe_parms_conf rtw89_8851b_rfe_parms_conf[] = {
+	{
+		.rfe_parms = &rtw89_8851b_rfe_parms_type2,
+		.rfe_type = 2,
+	},
+	{},
+};
-- 
cgit v1.2.3-70-g09d2


From 8c36cf0df434e17c9c39a42f87319a6aaca89b05 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Sat, 1 Apr 2023 22:25:48 +0800
Subject: wifi: rtw89: 8851b: add tables for RFK

These tables are used by RF calibrations to assist to configure PHY and
RF registers.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230401142548.55466-4-pkshih@realtek.com
---
 .../wireless/realtek/rtw89/rtw8851b_rfk_table.c    | 534 +++++++++++++++++++++
 .../wireless/realtek/rtw89/rtw8851b_rfk_table.h    |  38 ++
 2 files changed, 572 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.c
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.h

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.c b/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.c
new file mode 100644
index 000000000000..0abf7978ccab
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.c
@@ -0,0 +1,534 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2022-2023  Realtek Corporation
+ */
+
+#include "rtw8851b_rfk_table.h"
+
+static const struct rtw89_reg5_def rtw8851b_dadck_setup_defs[] = {
+	RTW89_DECL_RFK_WM(0xc210, 0x003fc000, 0x80),
+	RTW89_DECL_RFK_WM(0xc224, 0x003fc000, 0x80),
+	RTW89_DECL_RFK_WM(0xc0f8, 0x30000000, 0x3),
+	RTW89_DECL_RFK_WM(0x12b8, BIT(30), 0x1),
+	RTW89_DECL_RFK_WM(0x030c, 0x1f000000, 0x1f),
+	RTW89_DECL_RFK_WM(0x032c, 0xc0000000, 0x0),
+	RTW89_DECL_RFK_WM(0x032c, BIT(22), 0x0),
+	RTW89_DECL_RFK_WM(0x032c, BIT(22), 0x1),
+	RTW89_DECL_RFK_WM(0x032c, BIT(16), 0x0),
+	RTW89_DECL_RFK_WM(0x032c, BIT(20), 0x1),
+	RTW89_DECL_RFK_WM(0x030c, 0x0f000000, 0x3),
+	RTW89_DECL_RFK_WM(0xc0f4, BIT(2), 0x0),
+	RTW89_DECL_RFK_WM(0xc0f4, BIT(4), 0x0),
+	RTW89_DECL_RFK_WM(0xc0f4, BIT(11), 0x1),
+	RTW89_DECL_RFK_WM(0xc0f4, BIT(11), 0x0),
+	RTW89_DECL_RFK_DELAY(1),
+	RTW89_DECL_RFK_WM(0xc0f4, 0x300, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_dadck_setup_defs);
+
+static const struct rtw89_reg5_def rtw8851b_dadck_post_defs[] = {
+	RTW89_DECL_RFK_WM(0x032c, BIT(16), 0x1),
+	RTW89_DECL_RFK_WM(0x032c, BIT(20), 0x0),
+	RTW89_DECL_RFK_WM(0x030c, 0x1f000000, 0xc),
+	RTW89_DECL_RFK_WM(0x032c, 0xc0000000, 0x1),
+	RTW89_DECL_RFK_WM(0x12b8, BIT(30), 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_dadck_post_defs);
+
+static const struct rtw89_reg5_def rtw8851b_dack_s0_1_defs[] = {
+	RTW89_DECL_RFK_WM(0x12a0, BIT(15), 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x7000, 0x3),
+	RTW89_DECL_RFK_WM(0x12b8, BIT(30), 0x1),
+	RTW89_DECL_RFK_WM(0x030c, BIT(28), 0x1),
+	RTW89_DECL_RFK_WM(0x032c, 0x80000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_dack_s0_1_defs);
+
+static const struct rtw89_reg5_def rtw8851b_dack_s0_2_defs[] = {
+	RTW89_DECL_RFK_WM(0xc004, BIT(0), 0x0),
+	RTW89_DECL_RFK_WM(0x12a0, BIT(15), 0x0),
+	RTW89_DECL_RFK_WM(0x12a0, 0x7000, 0x7),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_dack_s0_2_defs);
+
+static const struct rtw89_reg5_def rtw8851b_dack_manual_off_defs[] = {
+	RTW89_DECL_RFK_WM(0xc0f8, 0x30000000, 0x0),
+	RTW89_DECL_RFK_WM(0xc210, BIT(0), 0x0),
+	RTW89_DECL_RFK_WM(0xc224, BIT(0), 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_dack_manual_off_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_rxclk_80_defs[] = {
+	RTW89_DECL_RFK_WM(0x20fc, 0xffff0000, 0x0101),
+	RTW89_DECL_RFK_WM(0x5670, 0x00002000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00080000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00070000, 0x2),
+	RTW89_DECL_RFK_WM(0x5670, 0x60000000, 0x1),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00000780, 0x8),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00007800, 0x2),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x0c000000, 0x2),
+	RTW89_DECL_RFK_WM(0xc0d8, 0x000001e0, 0x5),
+	RTW89_DECL_RFK_WM(0xc0c4, 0x003e0000, 0xf),
+	RTW89_DECL_RFK_WM(0xc0ec, 0x00006000, 0x0),
+	RTW89_DECL_RFK_WM(0x12b8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x0f),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x03),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0001),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0041),
+	RTW89_DECL_RFK_WM(0x20fc, 0xffff0000, 0x1101),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_rxclk_80_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_rxclk_others_defs[] = {
+	RTW89_DECL_RFK_WM(0x20fc, 0xffff0000, 0x0101),
+	RTW89_DECL_RFK_WM(0x5670, 0x00002000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00080000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00070000, 0x2),
+	RTW89_DECL_RFK_WM(0x5670, 0x60000000, 0x0),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00000780, 0x8),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00007800, 0x2),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x0c000000, 0x2),
+	RTW89_DECL_RFK_WM(0xc0d8, 0x000001e0, 0x5),
+	RTW89_DECL_RFK_WM(0xc0c4, 0x003e0000, 0xf),
+	RTW89_DECL_RFK_WM(0xc0ec, 0x00006000, 0x2),
+	RTW89_DECL_RFK_WM(0x12b8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x0f),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x03),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0001),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0041),
+	RTW89_DECL_RFK_WM(0x20fc, 0xffff0000, 0x1101),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_rxclk_others_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_txk_2ghz_defs[] = {
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x51, 0x80000, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x51, 0x00800, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x52, 0x00800, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x55, 0x0001f, 0x4),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0xef, 0x00004, 0x1),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x00, 0xffff0, 0x403e),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x11, 0x00003, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x11, 0x00070, 0x6),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x11, 0x1f000, 0x10),
+	RTW89_DECL_RFK_DELAY(1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_txk_2ghz_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_txk_5ghz_defs[] = {
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x60, 0x00007, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x55, 0x0001f, 0x4),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0xef, 0x00004, 0x1),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x00, 0xffff0, 0x403e),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x11, 0x00003, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x11, 0x00070, 0x7),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x11, 0x1f000, 0x7),
+	RTW89_DECL_RFK_DELAY(1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_txk_5ghz_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_afebb_restore_defs[] = {
+	RTW89_DECL_RFK_WM(0x12b8, 0x40000000, 0x0),
+	RTW89_DECL_RFK_WM(0x20fc, 0x00010000, 0x1),
+	RTW89_DECL_RFK_WM(0x20fc, 0x00100000, 0x0),
+	RTW89_DECL_RFK_WM(0x20fc, 0x01000000, 0x1),
+	RTW89_DECL_RFK_WM(0x20fc, 0x10000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5670, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x12a0, 0x000ff000, 0x00),
+	RTW89_DECL_RFK_WM(0x20fc, 0x00010000, 0x0),
+	RTW89_DECL_RFK_WM(0x20fc, 0x01000000, 0x0),
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x10005, 0x00001, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_afebb_restore_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_macbb_defs[] = {
+	RTW89_DECL_RFK_WRF(RF_PATH_A, 0x10005, 0x00001, 0x0),
+	RTW89_DECL_RFK_WM(0x20fc, 0x00010000, 0x1),
+	RTW89_DECL_RFK_WM(0x20fc, 0x00100000, 0x0),
+	RTW89_DECL_RFK_WM(0x20fc, 0x01000000, 0x1),
+	RTW89_DECL_RFK_WM(0x20fc, 0x10000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5670, MASKDWORD, 0xf801fffd),
+	RTW89_DECL_RFK_WM(0x5670, 0x00004000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00008000, 0x1),
+	RTW89_DECL_RFK_WM(0x5670, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00007000, 0x7),
+	RTW89_DECL_RFK_WM(0x5670, 0x00002000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00080000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00070000, 0x3),
+	RTW89_DECL_RFK_WM(0x5670, 0x60000000, 0x2),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00000780, 0x9),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00007800, 0x1),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x0c000000, 0x0),
+	RTW89_DECL_RFK_WM(0xc0d8, 0x000001e0, 0x3),
+	RTW89_DECL_RFK_WM(0xc0c4, 0x003e0000, 0xa),
+	RTW89_DECL_RFK_WM(0xc0ec, 0x00006000, 0x0),
+	RTW89_DECL_RFK_WM(0xc0e8, 0x00000040, 0x1),
+	RTW89_DECL_RFK_WM(0x12b8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x1f),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x13),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0001),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0041),
+	RTW89_DECL_RFK_WM(0x20fc, 0x00100000, 0x1),
+	RTW89_DECL_RFK_WM(0x20fc, 0x10000000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_macbb_defs);
+
+static const struct rtw89_reg5_def rtw8851b_iqk_bb_afe_defs[] = {
+	RTW89_DECL_RFK_WM(0x5670, 0x00004000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00008000, 0x1),
+	RTW89_DECL_RFK_WM(0x5670, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00007000, 0x7),
+	RTW89_DECL_RFK_WM(0x5670, 0x00002000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00080000, 0x1),
+	RTW89_DECL_RFK_WM(0x12a0, 0x00070000, 0x3),
+	RTW89_DECL_RFK_WM(0x5670, 0x60000000, 0x2),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00000780, 0x9),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x00007800, 0x1),
+	RTW89_DECL_RFK_WM(0xc0d4, 0x0c000000, 0x0),
+	RTW89_DECL_RFK_WM(0xc0d8, 0x000001e0, 0x3),
+	RTW89_DECL_RFK_WM(0xc0c4, 0x003e0000, 0xa),
+	RTW89_DECL_RFK_WM(0xc0ec, 0x00006000, 0x0),
+	RTW89_DECL_RFK_WM(0xc0e8, 0x00000040, 0x1),
+	RTW89_DECL_RFK_WM(0x12b8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x030c, MASKBYTE3, 0x1f),
+	RTW89_DECL_RFK_WM(0x030c, MASKBYTE3, 0x13),
+	RTW89_DECL_RFK_WM(0x032c, MASKHWORD, 0x0001),
+	RTW89_DECL_RFK_WM(0x032c, MASKHWORD, 0x0041),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_iqk_bb_afe_defs);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_sys_defs[] = {
+	RTW89_DECL_RFK_WM(0x12bc, 0x000ffff0, 0xb5b5),
+	RTW89_DECL_RFK_WM(0x32bc, 0x000ffff0, 0xb5b5),
+	RTW89_DECL_RFK_WM(0x0300, 0xff000000, 0x16),
+	RTW89_DECL_RFK_WM(0x0304, 0x0000ffff, 0x1f19),
+	RTW89_DECL_RFK_WM(0x0308, 0xff000000, 0x1c),
+	RTW89_DECL_RFK_WM(0x0314, 0xffff0000, 0x2041),
+	RTW89_DECL_RFK_WM(0x0318, 0xffffffff, 0x20012041),
+	RTW89_DECL_RFK_WM(0x0324, 0xffff0000, 0x2001),
+	RTW89_DECL_RFK_WM(0x0020, 0x00006000, 0x3),
+	RTW89_DECL_RFK_WM(0x0024, 0x00006000, 0x3),
+	RTW89_DECL_RFK_WM(0x0704, 0xffff0000, 0x601e),
+	RTW89_DECL_RFK_WM(0x2704, 0xffff0000, 0x601e),
+	RTW89_DECL_RFK_WM(0x0700, 0xf0000000, 0x4),
+	RTW89_DECL_RFK_WM(0x2700, 0xf0000000, 0x4),
+	RTW89_DECL_RFK_WM(0x0650, 0x3c000000, 0x0),
+	RTW89_DECL_RFK_WM(0x2650, 0x3c000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_sys_defs);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_sys_a_defs_2g[] = {
+	RTW89_DECL_RFK_WM(0x120c, 0x000000ff, 0x33),
+	RTW89_DECL_RFK_WM(0x12c0, 0x0ff00000, 0x33),
+	RTW89_DECL_RFK_WM(0x58f8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5814, 0x20000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_sys_a_defs_2g);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_sys_a_defs_5g[] = {
+	RTW89_DECL_RFK_WM(0x120c, 0x000000ff, 0x44),
+	RTW89_DECL_RFK_WM(0x12c0, 0x0ff00000, 0x44),
+	RTW89_DECL_RFK_WM(0x58f8, 0x40000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5814, 0x20000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_sys_a_defs_5g);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_init_txpwr_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x566c, 0x00001000, 0x0),
+	RTW89_DECL_RFK_WM(0x5800, 0xffffffff, 0x003f807f),
+	RTW89_DECL_RFK_WM(0x580c, 0x0000007f, 0x40),
+	RTW89_DECL_RFK_WM(0x580c, 0x0fffff00, 0x00040),
+	RTW89_DECL_RFK_WM(0x5810, 0xffffffff, 0x59010000),
+	RTW89_DECL_RFK_WM(0x5814, 0x01ffffff, 0x026d000),
+	RTW89_DECL_RFK_WM(0x5814, 0xf8000000, 0x00),
+	RTW89_DECL_RFK_WM(0x5818, 0x00ffffff, 0x2c18e8),
+	RTW89_DECL_RFK_WM(0x5818, 0x07000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5818, 0xf0000000, 0x0),
+	RTW89_DECL_RFK_WM(0x581c, 0x3fffffff, 0x3dc80280),
+	RTW89_DECL_RFK_WM(0x5820, 0xffffffff, 0x00000080),
+	RTW89_DECL_RFK_WM(0x58e8, 0x0000003f, 0x04),
+	RTW89_DECL_RFK_WM(0x580c, 0x10000000, 0x1),
+	RTW89_DECL_RFK_WM(0x580c, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5834, 0x3fffffff, 0x000115f2),
+	RTW89_DECL_RFK_WM(0x5838, 0x7fffffff, 0x0000121),
+	RTW89_DECL_RFK_WM(0x5854, 0x3fffffff, 0x000115f2),
+	RTW89_DECL_RFK_WM(0x5858, 0x7fffffff, 0x0000121),
+	RTW89_DECL_RFK_WM(0x5860, 0x80000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5864, 0x07ffffff, 0x00801ff),
+	RTW89_DECL_RFK_WM(0x5898, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x589c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x58a4, 0x000000ff, 0x16),
+	RTW89_DECL_RFK_WM(0x58b0, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x58b4, 0x7fffffff, 0x0a002000),
+	RTW89_DECL_RFK_WM(0x58b8, 0x7fffffff, 0x00007628),
+	RTW89_DECL_RFK_WM(0x58bc, 0x07ffffff, 0x7a7807f),
+	RTW89_DECL_RFK_WM(0x58c0, 0xfffe0000, 0x003f),
+	RTW89_DECL_RFK_WM(0x58c4, 0xffffffff, 0x0003ffff),
+	RTW89_DECL_RFK_WM(0x58c8, 0x00ffffff, 0x000000),
+	RTW89_DECL_RFK_WM(0x58c8, 0xf0000000, 0x0),
+	RTW89_DECL_RFK_WM(0x58cc, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x58d0, 0x07ffffff, 0x2008101),
+	RTW89_DECL_RFK_WM(0x58d4, 0x000000ff, 0x00),
+	RTW89_DECL_RFK_WM(0x58d4, 0x0003fe00, 0x0ff),
+	RTW89_DECL_RFK_WM(0x58d4, 0x07fc0000, 0x100),
+	RTW89_DECL_RFK_WM(0x58d8, 0xffffffff, 0x8008016c),
+	RTW89_DECL_RFK_WM(0x58dc, 0x0001ffff, 0x0807f),
+	RTW89_DECL_RFK_WM(0x58dc, 0xfff00000, 0x800),
+	RTW89_DECL_RFK_WM(0x58f0, 0x0003ffff, 0x001ff),
+	RTW89_DECL_RFK_WM(0x58f4, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x58f8, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_init_txpwr_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_init_txpwr_he_tb_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x58a0, MASKDWORD, 0x000000fe),
+	RTW89_DECL_RFK_WM(0x58e4, 0x0000007f, 0x1f),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_init_txpwr_he_tb_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_dck_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x580c, 0x0fff0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5814, 0x00001000, 0x1),
+	RTW89_DECL_RFK_WM(0x5814, 0x00002000, 0x1),
+	RTW89_DECL_RFK_WM(0x5814, 0x00004000, 0x1),
+	RTW89_DECL_RFK_WM(0x5814, 0x00038000, 0x3),
+	RTW89_DECL_RFK_WM(0x5814, 0x003c0000, 0x5),
+	RTW89_DECL_RFK_WM(0x5814, 0x18000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_dck_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_dac_gain_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x58b0, 0x00000fff, 0x000),
+	RTW89_DECL_RFK_WM(0x5a00, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a04, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a08, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a0c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a10, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a14, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a18, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a1c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a20, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a24, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a28, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a2c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a30, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a34, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a38, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a3c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a40, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a44, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a48, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a4c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a50, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a54, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a58, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a5c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a60, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a64, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a68, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a6c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a70, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a74, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a78, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a7c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a80, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a84, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a88, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a8c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a90, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a94, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a98, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a9c, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aa0, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aa4, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aa8, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aac, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ab0, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ab4, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ab8, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5abc, MASKDWORD, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ac0, MASKDWORD, 0x00000000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_dac_gain_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_slope_a_defs_2g[] = {
+	RTW89_DECL_RFK_WM(0x5608, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x560c, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x5610, 0x07ffffff, 0x0200e08),
+	RTW89_DECL_RFK_WM(0x5614, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x5618, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x561c, 0x000001ff, 0x007),
+	RTW89_DECL_RFK_WM(0x561c, 0xffff0000, 0x0808),
+	RTW89_DECL_RFK_WM(0x5620, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x5624, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x5628, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x562c, 0x0000ffff, 0x0808),
+	RTW89_DECL_RFK_WM(0x581c, 0x00100000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_slope_a_defs_2g);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_slope_a_defs_5g[] = {
+	RTW89_DECL_RFK_WM(0x5608, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x560c, 0x07ffffff, 0x0341a08),
+	RTW89_DECL_RFK_WM(0x5610, 0x07ffffff, 0x0201417),
+	RTW89_DECL_RFK_WM(0x5614, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x5618, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x561c, 0x000001ff, 0x008),
+	RTW89_DECL_RFK_WM(0x561c, 0xffff0000, 0x0808),
+	RTW89_DECL_RFK_WM(0x5620, 0xffffffff, 0x0e0e0808),
+	RTW89_DECL_RFK_WM(0x5624, 0xffffffff, 0x08080d18),
+	RTW89_DECL_RFK_WM(0x5628, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x562c, 0x0000ffff, 0x0808),
+	RTW89_DECL_RFK_WM(0x581c, 0x00100000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_slope_a_defs_5g);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_align_a_2g_defs[] = {
+	RTW89_DECL_RFK_WM(0x5604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5600, 0x3fffffff, 0x000000),
+	RTW89_DECL_RFK_WM(0x5604, 0x003fffff, 0x2d2400),
+	RTW89_DECL_RFK_WM(0x5630, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5634, 0x000003ff, 0x000),
+	RTW89_DECL_RFK_WM(0x5634, 0x000ffc00, 0x000),
+	RTW89_DECL_RFK_WM(0x5634, 0x3ff00000, 0x3fa),
+	RTW89_DECL_RFK_WM(0x5638, 0x000003ff, 0x02e),
+	RTW89_DECL_RFK_WM(0x5638, 0x000ffc00, 0x09c),
+	RTW89_DECL_RFK_WM(0x563c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5640, 0x3fffffff, 0x3fb00000),
+	RTW89_DECL_RFK_WM(0x5644, 0x000003ff, 0x02f),
+	RTW89_DECL_RFK_WM(0x5644, 0x000ffc00, 0x09c),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_align_a_2g_defs);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_align_a_5g_defs[] = {
+	RTW89_DECL_RFK_WM(0x5604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5600, 0x3fffffff, 0x000000),
+	RTW89_DECL_RFK_WM(0x5604, 0x003fffff, 0x3b2d24),
+	RTW89_DECL_RFK_WM(0x5630, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5634, 0x000003ff, 0x000),
+	RTW89_DECL_RFK_WM(0x5634, 0x000ffc00, 0x3cb),
+	RTW89_DECL_RFK_WM(0x5634, 0x3ff00000, 0x030),
+	RTW89_DECL_RFK_WM(0x5638, 0x000003ff, 0x73),
+	RTW89_DECL_RFK_WM(0x5638, 0x000ffc00, 0xd4),
+	RTW89_DECL_RFK_WM(0x563c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_align_a_5g_defs);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_slope_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x5820, 0x80000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5818, 0x10000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5814, 0x00000800, 0x1),
+	RTW89_DECL_RFK_WM(0x581c, 0x20000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5820, 0x0000f000, 0xf),
+	RTW89_DECL_RFK_WM(0x581c, 0x000003ff, 0x280),
+	RTW89_DECL_RFK_WM(0x581c, 0x000ffc00, 0x200),
+	RTW89_DECL_RFK_WM(0x58b8, 0x007f0000, 0x00),
+	RTW89_DECL_RFK_WM(0x58b8, 0x7f000000, 0x00),
+	RTW89_DECL_RFK_WM(0x58b4, 0x7f000000, 0x0a),
+	RTW89_DECL_RFK_WM(0x58b8, 0x0000007f, 0x28),
+	RTW89_DECL_RFK_WM(0x58b8, 0x00007f00, 0x76),
+	RTW89_DECL_RFK_WM(0x5810, 0x20000000, 0x0),
+	RTW89_DECL_RFK_WM(0x580c, 0x10000000, 0x1),
+	RTW89_DECL_RFK_WM(0x580c, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5834, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x5834, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5838, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5838, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x5854, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x5854, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5858, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5858, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x5824, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x5824, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5828, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5828, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x582c, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x582c, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5830, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5830, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x583c, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x583c, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5840, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5840, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x5844, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x5844, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5848, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5848, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x584c, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x584c, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5850, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5850, 0x003ff000, 0x000),
+	RTW89_DECL_RFK_WM(0x585c, 0x0003ffff, 0x115f2),
+	RTW89_DECL_RFK_WM(0x585c, 0x3ffc0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5860, 0x00000fff, 0x121),
+	RTW89_DECL_RFK_WM(0x5860, 0x003ff000, 0x000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_slope_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_track_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x5820, 0x80000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5818, 0x10000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5814, 0x00000800, 0x0),
+	RTW89_DECL_RFK_WM(0x581c, 0x20000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5864, 0x000003ff, 0x1ff),
+	RTW89_DECL_RFK_WM(0x5864, 0x000ffc00, 0x200),
+	RTW89_DECL_RFK_WM(0x5820, 0x00000fff, 0x080),
+	RTW89_DECL_RFK_WM(0x5814, 0x01000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_track_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_tssi_mv_avg_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x58e4, 0x00003800, 0x1),
+	RTW89_DECL_RFK_WM(0x58e4, 0x00004000, 0x0),
+	RTW89_DECL_RFK_WM(0x58e4, 0x00008000, 0x1),
+	RTW89_DECL_RFK_WM(0x58e4, 0x000f0000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_tssi_mv_avg_defs_a);
+
+static const struct rtw89_reg5_def rtw8851b_nctl_post_defs[] = {
+	RTW89_DECL_RFK_WM(0x5864, 0x18000000, 0x3),
+	RTW89_DECL_RFK_WM(0x7864, 0x18000000, 0x3),
+	RTW89_DECL_RFK_WM(0x030c, 0xff000000, 0x13),
+	RTW89_DECL_RFK_WM(0x032c, 0xffff0000, 0x0041),
+	RTW89_DECL_RFK_WM(0x12b8, 0x10000000, 0x1),
+	RTW89_DECL_RFK_WM(0x2008, 0x01ffffff, 0x00fffff),
+	RTW89_DECL_RFK_WM(0x0c60, 0x00000003, 0x3),
+	RTW89_DECL_RFK_WM(0x0c6c, 0x00000001, 0x1),
+	RTW89_DECL_RFK_WM(0x58ac, 0x08000000, 0x1),
+	RTW89_DECL_RFK_WM(0x78ac, 0x08000000, 0x1),
+	RTW89_DECL_RFK_WM(0x0730, 0x00003800, 0x7),
+	RTW89_DECL_RFK_WM(0x2730, 0x00003800, 0x7),
+	RTW89_DECL_RFK_WM(0x0c7c, 0x00e00000, 0x1),
+	RTW89_DECL_RFK_WM(0x58c0, 0x0001ffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x78c0, 0x0001ffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x58fc, 0x3f000000, 0x00),
+	RTW89_DECL_RFK_WM(0x78fc, 0x3f000000, 0x00),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8851b_nctl_post_defs);
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.h b/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.h
new file mode 100644
index 000000000000..febfbecb691c
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk_table.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2022-2023  Realtek Corporation
+ */
+
+#ifndef __RTW89_8851B_RFK_TABLE_H__
+#define __RTW89_8851B_RFK_TABLE_H__
+
+#include "phy.h"
+
+extern const struct rtw89_rfk_tbl rtw8851b_dadck_setup_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_dadck_post_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_dack_s0_1_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_dack_s0_2_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_dack_manual_off_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_rxclk_80_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_rxclk_others_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_txk_2ghz_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_txk_5ghz_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_afebb_restore_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_bb_afe_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_iqk_macbb_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_sys_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_sys_a_defs_2g_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_sys_a_defs_5g_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_init_txpwr_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_init_txpwr_he_tb_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_dck_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_dac_gain_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_slope_a_defs_2g_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_slope_a_defs_5g_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_align_a_2g_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_align_a_5g_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_slope_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_track_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_tssi_mv_avg_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8851b_nctl_post_defs_tbl;
+
+#endif
-- 
cgit v1.2.3-70-g09d2


From d33fc8d0368c180fe2338bfae4f5367a66a719f4 Mon Sep 17 00:00:00 2001
From: Eric Huang <echuang@realtek.com>
Date: Thu, 6 Apr 2023 15:28:41 +0800
Subject: wifi: rtw89: correct 5 MHz mask setting

Use primary channel index to determine which 5 MHz mask should be enable.
This mask is used to prevent noise from channel edge to effect CCA
threshold in wide bandwidth (>= 40 MHZ).

Fixes: 1b00e9236a71 ("rtw89: 8852c: add set channel of BB part")
Fixes: 6b0698984eb0 ("wifi: rtw89: 8852b: add chip_ops::set_channel")
Cc: stable@vger.kernel.org
Signed-off-by: Eric Huang <echuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230406072841.8308-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 9 +++++----
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 9 +++++----
 2 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 7ac7aa997a96..d7ba513b4f7a 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -1289,7 +1289,7 @@ static void rtw8852b_ctrl_cck_en(struct rtw89_dev *rtwdev, bool cck_en)
 static void rtw8852b_5m_mask(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
 			     enum rtw89_phy_idx phy_idx)
 {
-	u8 pri_ch = chan->primary_channel;
+	u8 pri_ch = chan->pri_ch_idx;
 	bool mask_5m_low;
 	bool mask_5m_en;
 
@@ -1297,12 +1297,13 @@ static void rtw8852b_5m_mask(struct rtw89_dev *rtwdev, const struct rtw89_chan *
 	case RTW89_CHANNEL_WIDTH_40:
 		/* Prich=1: Mask 5M High, Prich=2: Mask 5M Low */
 		mask_5m_en = true;
-		mask_5m_low = pri_ch == 2;
+		mask_5m_low = pri_ch == RTW89_SC_20_LOWER;
 		break;
 	case RTW89_CHANNEL_WIDTH_80:
 		/* Prich=3: Mask 5M High, Prich=4: Mask 5M Low, Else: Disable */
-		mask_5m_en = pri_ch == 3 || pri_ch == 4;
-		mask_5m_low = pri_ch == 4;
+		mask_5m_en = pri_ch == RTW89_SC_20_UPMOST ||
+			     pri_ch == RTW89_SC_20_LOWEST;
+		mask_5m_low = pri_ch == RTW89_SC_20_LOWEST;
 		break;
 	default:
 		mask_5m_en = false;
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 011b1957b712..6d8fa8acfc5e 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -1380,18 +1380,19 @@ static void rtw8852c_5m_mask(struct rtw89_dev *rtwdev,
 			     const struct rtw89_chan *chan,
 			     enum rtw89_phy_idx phy_idx)
 {
-	u8 pri_ch = chan->primary_channel;
+	u8 pri_ch = chan->pri_ch_idx;
 	bool mask_5m_low;
 	bool mask_5m_en;
 
 	switch (chan->band_width) {
 	case RTW89_CHANNEL_WIDTH_40:
 		mask_5m_en = true;
-		mask_5m_low = pri_ch == 2;
+		mask_5m_low = pri_ch == RTW89_SC_20_LOWER;
 		break;
 	case RTW89_CHANNEL_WIDTH_80:
-		mask_5m_en = ((pri_ch == 3) || (pri_ch == 4));
-		mask_5m_low = pri_ch == 4;
+		mask_5m_en = pri_ch == RTW89_SC_20_UPMOST ||
+			     pri_ch == RTW89_SC_20_LOWEST;
+		mask_5m_low = pri_ch == RTW89_SC_20_LOWEST;
 		break;
 	default:
 		mask_5m_en = false;
-- 
cgit v1.2.3-70-g09d2


From 8551844d2c5b28b6809a45eb4f5d6e931838fd04 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 6 Apr 2023 17:30:09 +0800
Subject: wifi: rtw89: fix crash due to null pointer of sta in AP mode

In AP mode, 'sta' could be NULL if sending broadcast/multicast packets,
so we should check before accessing, or it causes crash:

  BUG: kernel NULL pointer dereference, address: 0000000000000004
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 0 P4D 0
  Oops: 0000 [#1] PREEMPT SMP PTI
  CPU: 2 PID: 92 Comm: kworker/u33:0 Tainted: G           OE
  Workqueue: rtw89_tx_wq rtw89_core_txq_work [rtw89_core]
  RIP: 0010:rtw89_core_tx_update_desc_info+0x2cc/0x7d0 [rtw89_core]
  Code: e2 01 41 be 04 00 00 00 41 8b 84 c4 0c 01 00 00 75 0d 45 31 f6 ...
  RSP: 0018:ffffb4cf807afce0 EFLAGS: 00010297
  RAX: 0000000000000001 RBX: ffffb4cf807afd48 RCX: 0000000000000000
  RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000001
  RBP: ffffb4cf807afd30 R08: ffff9b28c1e59808 R09: ffff9b28c0297100
  R10: 00000000052cf7c4 R11: 00000000052cf7c4 R12: ffff9b28c1602040
  R13: ffff9b28c07b3000 R14: 0000000000000004 R15: 0000000000000000
  FS:  0000000000000000(0000) GS:ffff9b2a73280000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000004 CR3: 00000001ca410003 CR4: 00000000000606e0
  Call Trace:
   <TASK>
   rtw89_core_tx_write+0x7c/0x100 [rtw89_core]
   rtw89_core_txq_work+0x1b4/0x530 [rtw89_core]
   process_one_work+0x222/0x3f0
   worker_thread+0x50/0x3f0
   kthread+0x16b/0x190
   ? rescuer_thread+0x3a0/0x3a0
   ? set_kthread_struct+0x50/0x50
   ret_from_fork+0x22/0x30
   </TASK>

Fixes: e5307c9cd7ee ("wifi: rtw89: set data lowest rate according to AP supported rate")
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230406093009.5869-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index d6204e07a5b6..d144903c60b3 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -707,7 +707,7 @@ static u16 rtw89_core_get_data_rate(struct rtw89_dev *rtwdev,
 	else
 		lowest_rate = RTW89_HW_RATE_OFDM6;
 
-	if (!sta->deflink.supp_rates[chan->band_type])
+	if (!sta || !sta->deflink.supp_rates[chan->band_type])
 		return lowest_rate;
 
 	return __ffs(sta->deflink.supp_rates[chan->band_type]) + lowest_rate;
-- 
cgit v1.2.3-70-g09d2


From cc4cffc3c142d57df48c07851862444e1d33bdaa Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Fri, 7 Apr 2023 22:37:52 +0200
Subject: wifi: brcmfmac: add Cypress 43439 SDIO ids

Add SDIO ids for use with the muRata 1YN (Cypress CYW43439).
The odd thing about this is that the previous 1YN populated
on M.2 card for evaluation purposes had BRCM SDIO vendor ID,
while the chip populated on real hardware has a Cypress one.
The device ID also differs between the two devices. But they
are both 43439 otherwise, so add the IDs for both.

On-device 1YN (43439), the new one, chip label reads "1YN":
```
/sys/.../mmc_host/mmc2/mmc2:0001 # cat vendor device
0x04b4
0xbd3d
```

EA M.2 evaluation board 1YN (43439), the old one, chip label reads "1YN ES1.4":
```
/sys/.../mmc_host/mmc0/mmc0:0001/# cat vendor device
0x02d0
0xa9a6
```

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230407203752.128539-1-marex@denx.de
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c | 9 ++++++++-
 include/linux/mmc/sdio_ids.h                              | 5 ++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index b7c918f241c9..f5dc3bb11b64 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -965,6 +965,12 @@ out:
 		.driver_data = BRCMF_FWVENDOR_ ## fw_vend \
 	}
 
+#define CYW_SDIO_DEVICE(dev_id, fw_vend) \
+	{ \
+		SDIO_DEVICE(SDIO_VENDOR_ID_CYPRESS, dev_id), \
+		.driver_data = BRCMF_FWVENDOR_ ## fw_vend \
+	}
+
 /* devices we support, null terminated */
 static const struct sdio_device_id brcmf_sdmmc_ids[] = {
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43143, WCC),
@@ -979,6 +985,7 @@ static const struct sdio_device_id brcmf_sdmmc_ids[] = {
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4335_4339, WCC),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4339, WCC),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43430, WCC),
+	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43439, WCC),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4345, WCC),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43455, WCC),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4354, WCC),
@@ -986,9 +993,9 @@ static const struct sdio_device_id brcmf_sdmmc_ids[] = {
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4359, WCC),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_CYPRESS_4373, CYW),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_CYPRESS_43012, CYW),
-	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_CYPRESS_43439, CYW),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_CYPRESS_43752, CYW),
 	BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_CYPRESS_89359, CYW),
+	CYW_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_CYPRESS_43439, CYW),
 	{ /* end: all zeroes */ }
 };
 MODULE_DEVICE_TABLE(sdio, brcmf_sdmmc_ids);
diff --git a/include/linux/mmc/sdio_ids.h b/include/linux/mmc/sdio_ids.h
index 66f503ed2448..c653accdc7fd 100644
--- a/include/linux/mmc/sdio_ids.h
+++ b/include/linux/mmc/sdio_ids.h
@@ -74,10 +74,13 @@
 #define SDIO_DEVICE_ID_BROADCOM_43362		0xa962
 #define SDIO_DEVICE_ID_BROADCOM_43364		0xa9a4
 #define SDIO_DEVICE_ID_BROADCOM_43430		0xa9a6
-#define SDIO_DEVICE_ID_BROADCOM_CYPRESS_43439	0xa9af
+#define SDIO_DEVICE_ID_BROADCOM_43439		0xa9af
 #define SDIO_DEVICE_ID_BROADCOM_43455		0xa9bf
 #define SDIO_DEVICE_ID_BROADCOM_CYPRESS_43752	0xaae8
 
+#define SDIO_VENDOR_ID_CYPRESS			0x04b4
+#define SDIO_DEVICE_ID_BROADCOM_CYPRESS_43439	0xbd3d
+
 #define SDIO_VENDOR_ID_MARVELL			0x02df
 #define SDIO_DEVICE_ID_MARVELL_LIBERTAS		0x9103
 #define SDIO_DEVICE_ID_MARVELL_8688_WLAN	0x9104
-- 
cgit v1.2.3-70-g09d2


From a5be45ea459371ad70eb1508b3fd4b731cdc57c7 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 10 Apr 2023 18:35:45 +0300
Subject: wifi: rtl8xxxu: Clean up some messy ifs

Add some new members to rtl8xxxu_fileops and use them instead of
checking priv->rtl_chip.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/e24a5534-6e33-cfb9-0634-0caf4646513f@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h   |  5 +++++
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c |  1 +
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c |  5 +++++
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c |  9 ++++++++
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c |  3 +++
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c  | 26 ++++++----------------
 6 files changed, 30 insertions(+), 19 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 9d48c69ffece..39fee07917e7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1923,6 +1923,11 @@ struct rtl8xxxu_fileops {
 	u8 has_tx_report:1;
 	u8 gen2_thermal_meter:1;
 	u8 needs_full_init:1;
+	u8 init_reg_rxfltmap:1;
+	u8 init_reg_pkt_life_time:1;
+	u8 init_reg_hmtfr:1;
+	u8 ampdu_max_time;
+	u8 ustime_tsf_edca;
 	u32 adda_1t_init;
 	u32 adda_1t_path_on;
 	u32 adda_2t_path_on_a;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
index 6a82ec47568e..af8436070ba7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
@@ -1883,6 +1883,7 @@ struct rtl8xxxu_fileops rtl8188eu_fops = {
 	.rx_desc_size = sizeof(struct rtl8xxxu_rxdesc16),
 	.tx_desc_size = sizeof(struct rtl8xxxu_txdesc32),
 	.has_tx_report = 1,
+	.init_reg_pkt_life_time = 1,
 	.gen2_thermal_meter = 1,
 	.adda_1t_init = 0x0b1b25a0,
 	.adda_1t_path_on = 0x0bdb25a0,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
index 82dee1fed477..dfb250adb168 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
@@ -1746,6 +1746,11 @@ struct rtl8xxxu_fileops rtl8188fu_fops = {
 	.has_tx_report = 1,
 	.gen2_thermal_meter = 1,
 	.needs_full_init = 1,
+	.init_reg_rxfltmap = 1,
+	.init_reg_pkt_life_time = 1,
+	.init_reg_hmtfr = 1,
+	.ampdu_max_time = 0x70,
+	.ustime_tsf_edca = 0x28,
 	.adda_1t_init = 0x03c00014,
 	.adda_1t_path_on = 0x03c00014,
 	.trxff_boundary = 0x3f7f,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c
index 920466e39604..22d4704dd31e 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8710b.c
@@ -1865,6 +1865,15 @@ struct rtl8xxxu_fileops rtl8710bu_fops = {
 	.has_tx_report = 1,
 	.gen2_thermal_meter = 1,
 	.needs_full_init = 1,
+	.init_reg_rxfltmap = 1,
+	.init_reg_pkt_life_time = 1,
+	.init_reg_hmtfr = 1,
+	.ampdu_max_time = 0x5e,
+	/*
+	 * The RTL8710BU vendor driver uses 0x50 here and it works fine,
+	 * but in rtl8xxxu 0x50 causes slow upload and random packet loss. Why?
+	 */
+	.ustime_tsf_edca = 0x28,
 	.adda_1t_init = 0x03c00016,
 	.adda_1t_path_on = 0x03c00016,
 	.trxff_boundary = 0x3f7f,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
index d99538eb8398..c31c2b52ac77 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
@@ -1741,6 +1741,9 @@ struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.has_tx_report = 1,
 	.gen2_thermal_meter = 1,
 	.needs_full_init = 1,
+	.init_reg_hmtfr = 1,
+	.ampdu_max_time = 0x5e,
+	.ustime_tsf_edca = 0x50,
 	.adda_1t_init = 0x01c00014,
 	.adda_1t_path_on = 0x01c00014,
 	.adda_2t_path_on_a = 0x01c00014,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index c152b228606f..62dd53a57659 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -1916,7 +1916,7 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 	/*
 	 * Init H2C command
 	 */
-	if (priv->rtl_chip == RTL8723B || priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B)
+	if (priv->fops->init_reg_hmtfr)
 		rtl8xxxu_write8(priv, REG_HMTFR, 0x0f);
 exit:
 	return ret;
@@ -3864,11 +3864,8 @@ void rtl8xxxu_init_burst(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_HT_SINGLE_AMPDU_8723B, val8);
 
 	rtl8xxxu_write16(priv, REG_MAX_AGGR_NUM, 0x0c14);
-	if (priv->rtl_chip == RTL8723B || priv->rtl_chip == RTL8710B)
-		val8 = 0x5e;
-	else if (priv->rtl_chip == RTL8188F)
-		val8 = 0x70; /* 0x5e would make it very slow */
-	rtl8xxxu_write8(priv, REG_AMPDU_MAX_TIME_8723B, val8);
+	rtl8xxxu_write8(priv, REG_AMPDU_MAX_TIME_8723B,
+			priv->fops->ampdu_max_time);
 	rtl8xxxu_write32(priv, REG_AGGLEN_LMT, 0xffffffff);
 	rtl8xxxu_write8(priv, REG_RX_PKT_LIMIT, 0x18);
 	rtl8xxxu_write8(priv, REG_PIFS, 0x00);
@@ -3876,16 +3873,8 @@ void rtl8xxxu_init_burst(struct rtl8xxxu_priv *priv)
 		rtl8xxxu_write8(priv, REG_FWHW_TXQ_CTRL, FWHW_TXQ_CTRL_AMPDU_RETRY);
 		rtl8xxxu_write32(priv, REG_FAST_EDCA_CTRL, 0x03086666);
 	}
-	/*
-	 * The RTL8710BU vendor driver uses 0x50 here and it works fine,
-	 * but in rtl8xxxu 0x50 causes slow upload and random packet loss. Why?
-	 */
-	if (priv->rtl_chip == RTL8723B)
-		val8 = 0x50;
-	else if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B)
-		val8 = 0x28; /* 0x50 would make the upload slow */
-	rtl8xxxu_write8(priv, REG_USTIME_TSF_8723B, val8);
-	rtl8xxxu_write8(priv, REG_USTIME_EDCA, val8);
+	rtl8xxxu_write8(priv, REG_USTIME_TSF_8723B, priv->fops->ustime_tsf_edca);
+	rtl8xxxu_write8(priv, REG_USTIME_EDCA, priv->fops->ustime_tsf_edca);
 
 	/* to prevent mac is reseted by bus. */
 	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL);
@@ -4102,7 +4091,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		RCR_APPEND_PHYSTAT | RCR_APPEND_ICV | RCR_APPEND_MIC;
 	rtl8xxxu_write32(priv, REG_RCR, val32);
 
-	if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8710B) {
+	if (fops->init_reg_rxfltmap) {
 		/* Accept all data frames */
 		rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0xffff);
 
@@ -4187,8 +4176,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (fops->init_aggregation)
 		fops->init_aggregation(priv);
 
-	if (priv->rtl_chip == RTL8188F || priv->rtl_chip == RTL8188E ||
-	    priv->rtl_chip == RTL8710B) {
+	if (fops->init_reg_pkt_life_time) {
 		rtl8xxxu_write16(priv, REG_PKT_VO_VI_LIFE_TIME, 0x0400); /* unit: 256us. 256ms */
 		rtl8xxxu_write16(priv, REG_PKT_BE_BK_LIFE_TIME, 0x0400); /* unit: 256us. 256ms */
 	}
-- 
cgit v1.2.3-70-g09d2


From b9c3379dda1435110440bf9d6d230b3067e8f1d6 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 10 Apr 2023 18:36:48 +0300
Subject: wifi: rtl8xxxu: Support devices with 5-6 out endpoints

Handle them the same way as the devices with 3-4 USB out endpoints.
This is needed for the RTL8192FU.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/ef4bc66d-84f5-4021-efd7-1787d097519c@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h      | 2 +-
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 39fee07917e7..82a0290ccb29 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -27,7 +27,7 @@
 #define RTL8XXXU_MAX_REG_POLL		500
 #define	USB_INTR_CONTENT_LENGTH		56
 
-#define RTL8XXXU_OUT_ENDPOINTS		4
+#define RTL8XXXU_OUT_ENDPOINTS		6
 
 #define REALTEK_USB_READ		0xc0
 #define REALTEK_USB_WRITE		0x40
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 62dd53a57659..6106b47d0c37 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -1663,6 +1663,8 @@ int rtl8xxxu_config_endpoints_no_sie(struct rtl8xxxu_priv *priv)
 	struct device *dev = &priv->udev->dev;
 
 	switch (priv->nr_out_eps) {
+	case 6:
+	case 5:
 	case 4:
 	case 3:
 		priv->ep_tx_low_queue = 1;
-- 
cgit v1.2.3-70-g09d2


From 666f4ab26c2c7c5e1aa963dc5d8623df1b015d06 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Mon, 10 Apr 2023 18:27:01 -0500
Subject: bcma: Add explicit of_device.h include

bcma/main.c uses of_dma_configure() which is declared in of_device.h.
of_device.h gets implicitly included by of_platform.h, but that is going
to be removed soon.

Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230410232701.1561613-1-robh@kernel.org
---
 drivers/bcma/main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bcma/main.c b/drivers/bcma/main.c
index 5e438f74ee4c..7061d3ee836a 100644
--- a/drivers/bcma/main.c
+++ b/drivers/bcma/main.c
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
+#include <linux/of_device.h>
 #include <linux/of_platform.h>
 
 MODULE_DESCRIPTION("Broadcom's specific AMBA driver");
-- 
cgit v1.2.3-70-g09d2


From b47dddc624efbc6c31afe0f4ee415ccbdeb0e4ce Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Sun, 6 Nov 2022 12:50:41 +0200
Subject: net/mlx5: DR, Move ACTION_CACHE_LINE_SIZE macro to header

Move ACTION_CACHE_LINE_SIZE macro to header to be used by
the pattern functions as well.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c | 10 ++++------
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h  |  1 +
 2 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index ee104cf04392..724e4d9e70ac 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -1365,8 +1365,6 @@ out_err:
 	return -EINVAL;
 }
 
-#define ACTION_CACHE_LINE_SIZE 64
-
 static int
 dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
 				 u8 reformat_param_0, u8 reformat_param_1,
@@ -1403,13 +1401,13 @@ dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
 	}
 	case DR_ACTION_TYP_TNL_L3_TO_L2:
 	{
-		u8 hw_actions[ACTION_CACHE_LINE_SIZE] = {};
+		u8 hw_actions[DR_ACTION_CACHE_LINE_SIZE] = {};
 		int ret;
 
 		ret = mlx5dr_ste_set_action_decap_l3_list(dmn->ste_ctx,
 							  data, data_sz,
 							  hw_actions,
-							  ACTION_CACHE_LINE_SIZE,
+							  DR_ACTION_CACHE_LINE_SIZE,
 							  &action->rewrite->num_of_actions);
 		if (ret) {
 			mlx5dr_dbg(dmn, "Failed creating decap l3 action list\n");
@@ -1427,7 +1425,7 @@ dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
 		action->rewrite->index = (mlx5dr_icm_pool_get_chunk_icm_addr
 					  (action->rewrite->chunk) -
 					 dmn->info.caps.hdr_modify_icm_addr) /
-					 ACTION_CACHE_LINE_SIZE;
+					 DR_ACTION_CACHE_LINE_SIZE;
 
 		ret = mlx5dr_send_postsend_action(dmn, action);
 		if (ret) {
@@ -2006,7 +2004,7 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 	action->rewrite->num_of_actions = num_hw_actions;
 	action->rewrite->index = (mlx5dr_icm_pool_get_chunk_icm_addr(chunk) -
 				  dmn->info.caps.hdr_modify_icm_addr) /
-				  ACTION_CACHE_LINE_SIZE;
+				  DR_ACTION_CACHE_LINE_SIZE;
 
 	ret = mlx5dr_send_postsend_action(dmn, action);
 	if (ret)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 5b9faa714f42..7453fc6df494 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -21,6 +21,7 @@
 #define DR_NUM_OF_FLEX_PARSERS 8
 #define DR_STE_MAX_FLEX_0_ID 3
 #define DR_STE_MAX_FLEX_1_ID 7
+#define DR_ACTION_CACHE_LINE_SIZE 64
 
 #define mlx5dr_err(dmn, arg...) mlx5_core_err((dmn)->mdev, ##arg)
 #define mlx5dr_info(dmn, arg...) mlx5_core_info((dmn)->mdev, ##arg)
-- 
cgit v1.2.3-70-g09d2


From da5d0027d666697f28af65fd0d975e9c7e2327c6 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Sun, 6 Nov 2022 12:52:23 +0200
Subject: net/mlx5: DR, Add cache for modify header pattern

Starting with ConnectX-6 Dx, we use new design of modify_header FW object.
The current modify_header object allows for having only limited number
of FW objects, so the new design of pattern and argument allows pattern
reuse, saving memory, and having a large number of modify_header objects.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_ptrn.c | 198 +++++++++++++++++++++
 .../ethernet/mellanox/mlx5/core/steering/dr_send.c |  20 +++
 .../mellanox/mlx5/core/steering/dr_types.h         |  18 ++
 3 files changed, 236 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c
index 698e79d278bf..13e06a6a6b22 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c
@@ -2,12 +2,198 @@
 // Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 
 #include "dr_types.h"
+#include "mlx5_ifc_dr_ste_v1.h"
+
+enum dr_ptrn_modify_hdr_action_id {
+	DR_PTRN_MODIFY_HDR_ACTION_ID_NOP = 0x00,
+	DR_PTRN_MODIFY_HDR_ACTION_ID_COPY = 0x05,
+	DR_PTRN_MODIFY_HDR_ACTION_ID_SET = 0x06,
+	DR_PTRN_MODIFY_HDR_ACTION_ID_ADD = 0x07,
+	DR_PTRN_MODIFY_HDR_ACTION_ID_INSERT_INLINE = 0x0a,
+};
 
 struct mlx5dr_ptrn_mgr {
 	struct mlx5dr_domain *dmn;
 	struct mlx5dr_icm_pool *ptrn_icm_pool;
+	/* cache for modify_header ptrn */
+	struct list_head ptrn_list;
+	struct mutex modify_hdr_mutex; /* protect the pattern cache */
 };
 
+/* Cache structure and functions */
+static bool dr_ptrn_compare_modify_hdr(size_t cur_num_of_actions,
+				       __be64 cur_hw_actions[],
+				       size_t num_of_actions,
+				       __be64 hw_actions[])
+{
+	int i;
+
+	if (cur_num_of_actions != num_of_actions)
+		return false;
+
+	for (i = 0; i < num_of_actions; i++) {
+		u8 action_id =
+			MLX5_GET(ste_double_action_set_v1, &hw_actions[i], action_id);
+
+		if (action_id == DR_PTRN_MODIFY_HDR_ACTION_ID_COPY) {
+			if (hw_actions[i] != cur_hw_actions[i])
+				return false;
+		} else {
+			if ((__force __be32)hw_actions[i] !=
+			    (__force __be32)cur_hw_actions[i])
+				return false;
+		}
+	}
+
+	return true;
+}
+
+static struct mlx5dr_ptrn_obj *
+dr_ptrn_find_cached_pattern(struct mlx5dr_ptrn_mgr *mgr,
+			    size_t num_of_actions,
+			    __be64 hw_actions[])
+{
+	struct mlx5dr_ptrn_obj *cached_pattern;
+	struct mlx5dr_ptrn_obj *tmp;
+
+	list_for_each_entry_safe(cached_pattern, tmp, &mgr->ptrn_list, list) {
+		if (dr_ptrn_compare_modify_hdr(cached_pattern->num_of_actions,
+					       (__be64 *)cached_pattern->data,
+					       num_of_actions,
+					       hw_actions)) {
+			/* Put this pattern in the head of the list,
+			 * as we will probably use it more.
+			 */
+			list_del_init(&cached_pattern->list);
+			list_add(&cached_pattern->list, &mgr->ptrn_list);
+			return cached_pattern;
+		}
+	}
+
+	return NULL;
+}
+
+static struct mlx5dr_ptrn_obj *
+dr_ptrn_alloc_pattern(struct mlx5dr_ptrn_mgr *mgr,
+		      u16 num_of_actions, u8 *data)
+{
+	struct mlx5dr_ptrn_obj *pattern;
+	struct mlx5dr_icm_chunk *chunk;
+	u32 chunk_size;
+	u32 index;
+
+	chunk_size = ilog2(num_of_actions);
+	/* HW modify action index granularity is at least 64B */
+	chunk_size = max_t(u32, chunk_size, DR_CHUNK_SIZE_8);
+
+	chunk = mlx5dr_icm_alloc_chunk(mgr->ptrn_icm_pool, chunk_size);
+	if (!chunk)
+		return NULL;
+
+	index = (mlx5dr_icm_pool_get_chunk_icm_addr(chunk) -
+		 mgr->dmn->info.caps.hdr_modify_pattern_icm_addr) /
+		DR_ACTION_CACHE_LINE_SIZE;
+
+	pattern = kzalloc(sizeof(*pattern), GFP_KERNEL);
+	if (!pattern)
+		goto free_chunk;
+
+	pattern->data = kzalloc(num_of_actions * DR_MODIFY_ACTION_SIZE *
+				sizeof(*pattern->data), GFP_KERNEL);
+	if (!pattern->data)
+		goto free_pattern;
+
+	memcpy(pattern->data, data, num_of_actions * DR_MODIFY_ACTION_SIZE);
+	pattern->chunk = chunk;
+	pattern->index = index;
+	pattern->num_of_actions = num_of_actions;
+
+	list_add(&pattern->list, &mgr->ptrn_list);
+	refcount_set(&pattern->refcount, 1);
+
+	return pattern;
+
+free_pattern:
+	kfree(pattern);
+free_chunk:
+	mlx5dr_icm_free_chunk(chunk);
+	return NULL;
+}
+
+static void
+dr_ptrn_free_pattern(struct mlx5dr_ptrn_obj *pattern)
+{
+	list_del(&pattern->list);
+	mlx5dr_icm_free_chunk(pattern->chunk);
+	kfree(pattern->data);
+	kfree(pattern);
+}
+
+struct mlx5dr_ptrn_obj *
+mlx5dr_ptrn_cache_get_pattern(struct mlx5dr_ptrn_mgr *mgr,
+			      u16 num_of_actions,
+			      u8 *data)
+{
+	struct mlx5dr_ptrn_obj *pattern;
+	u64 *hw_actions;
+	u8 action_id;
+	int i;
+
+	mutex_lock(&mgr->modify_hdr_mutex);
+	pattern = dr_ptrn_find_cached_pattern(mgr,
+					      num_of_actions,
+					      (__be64 *)data);
+	if (!pattern) {
+		/* Alloc and add new pattern to cache */
+		pattern = dr_ptrn_alloc_pattern(mgr, num_of_actions, data);
+		if (!pattern)
+			goto out_unlock;
+
+		hw_actions = (u64 *)pattern->data;
+		/* Here we mask the pattern data to create a valid pattern
+		 * since we do an OR operation between the arg and pattern
+		 */
+		for (i = 0; i < num_of_actions; i++) {
+			action_id = MLX5_GET(ste_double_action_set_v1, &hw_actions[i], action_id);
+
+			if (action_id == DR_PTRN_MODIFY_HDR_ACTION_ID_SET ||
+			    action_id == DR_PTRN_MODIFY_HDR_ACTION_ID_ADD ||
+			    action_id == DR_PTRN_MODIFY_HDR_ACTION_ID_INSERT_INLINE)
+				MLX5_SET(ste_double_action_set_v1, &hw_actions[i], inline_data, 0);
+		}
+
+		if (mlx5dr_send_postsend_pattern(mgr->dmn, pattern->chunk,
+						 num_of_actions, pattern->data)) {
+			refcount_dec(&pattern->refcount);
+			goto free_pattern;
+		}
+	} else {
+		refcount_inc(&pattern->refcount);
+	}
+
+	mutex_unlock(&mgr->modify_hdr_mutex);
+
+	return pattern;
+
+free_pattern:
+	dr_ptrn_free_pattern(pattern);
+out_unlock:
+	mutex_unlock(&mgr->modify_hdr_mutex);
+	return NULL;
+}
+
+void
+mlx5dr_ptrn_cache_put_pattern(struct mlx5dr_ptrn_mgr *mgr,
+			      struct mlx5dr_ptrn_obj *pattern)
+{
+	mutex_lock(&mgr->modify_hdr_mutex);
+
+	if (refcount_dec_and_test(&pattern->refcount))
+		dr_ptrn_free_pattern(pattern);
+
+	mutex_unlock(&mgr->modify_hdr_mutex);
+}
+
 struct mlx5dr_ptrn_mgr *mlx5dr_ptrn_mgr_create(struct mlx5dr_domain *dmn)
 {
 	struct mlx5dr_ptrn_mgr *mgr;
@@ -26,6 +212,7 @@ struct mlx5dr_ptrn_mgr *mlx5dr_ptrn_mgr_create(struct mlx5dr_domain *dmn)
 		goto free_mgr;
 	}
 
+	INIT_LIST_HEAD(&mgr->ptrn_list);
 	return mgr;
 
 free_mgr:
@@ -35,9 +222,20 @@ free_mgr:
 
 void mlx5dr_ptrn_mgr_destroy(struct mlx5dr_ptrn_mgr *mgr)
 {
+	struct mlx5dr_ptrn_obj *pattern;
+	struct mlx5dr_ptrn_obj *tmp;
+
 	if (!mgr)
 		return;
 
+	WARN_ON(!list_empty(&mgr->ptrn_list));
+
+	list_for_each_entry_safe(pattern, tmp, &mgr->ptrn_list, list) {
+		list_del(&pattern->list);
+		kfree(pattern->data);
+		kfree(pattern);
+	}
+
 	mlx5dr_icm_pool_destroy(mgr->ptrn_icm_pool);
 	kfree(mgr);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index 00bb65613300..78756840d263 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -754,6 +754,26 @@ int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn,
 	return dr_postsend_icm_data(dmn, &send_info);
 }
 
+int mlx5dr_send_postsend_pattern(struct mlx5dr_domain *dmn,
+				 struct mlx5dr_icm_chunk *chunk,
+				 u16 num_of_actions,
+				 u8 *data)
+{
+	struct postsend_info send_info = {};
+	int ret;
+
+	send_info.write.addr = (uintptr_t)data;
+	send_info.write.length = num_of_actions * DR_MODIFY_ACTION_SIZE;
+	send_info.remote_addr = mlx5dr_icm_pool_get_chunk_mr_addr(chunk);
+	send_info.rkey = mlx5dr_icm_pool_get_chunk_rkey(chunk);
+
+	ret = dr_postsend_icm_data(dmn, &send_info);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
 static int dr_modify_qp_rst2init(struct mlx5_core_dev *mdev,
 				 struct mlx5dr_qp *dr_qp,
 				 int port)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 7453fc6df494..097f1f389b76 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1002,6 +1002,15 @@ struct mlx5dr_ste_action_modify_field {
 	u8 l4_type;
 };
 
+struct mlx5dr_ptrn_obj {
+	struct mlx5dr_icm_chunk *chunk;
+	u8 *data;
+	u16 num_of_actions;
+	u32 index;
+	refcount_t refcount;
+	struct list_head list;
+};
+
 struct mlx5dr_action_rewrite {
 	struct mlx5dr_domain *dmn;
 	struct mlx5dr_icm_chunk *chunk;
@@ -1011,6 +1020,7 @@ struct mlx5dr_action_rewrite {
 	u8 allow_rx:1;
 	u8 allow_tx:1;
 	u8 modify_ttl:1;
+	struct mlx5dr_ptrn_obj *ptrn;
 };
 
 struct mlx5dr_action_reformat {
@@ -1448,6 +1458,10 @@ int mlx5dr_send_postsend_formatted_htbl(struct mlx5dr_domain *dmn,
 					bool update_hw_ste);
 int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn,
 				struct mlx5dr_action *action);
+int mlx5dr_send_postsend_pattern(struct mlx5dr_domain *dmn,
+				 struct mlx5dr_icm_chunk *chunk,
+				 u16 num_of_actions,
+				 u8 *data);
 
 int mlx5dr_send_info_pool_create(struct mlx5dr_domain *dmn);
 void mlx5dr_send_info_pool_destroy(struct mlx5dr_domain *dmn);
@@ -1537,5 +1551,9 @@ static inline bool mlx5dr_supp_match_ranges(struct mlx5_core_dev *dev)
 bool mlx5dr_domain_is_support_ptrn_arg(struct mlx5dr_domain *dmn);
 struct mlx5dr_ptrn_mgr *mlx5dr_ptrn_mgr_create(struct mlx5dr_domain *dmn);
 void mlx5dr_ptrn_mgr_destroy(struct mlx5dr_ptrn_mgr *mgr);
+struct mlx5dr_ptrn_obj *mlx5dr_ptrn_cache_get_pattern(struct mlx5dr_ptrn_mgr *mgr,
+						      u16 num_of_actions, u8 *data);
+void mlx5dr_ptrn_cache_put_pattern(struct mlx5dr_ptrn_mgr *mgr,
+				   struct mlx5dr_ptrn_obj *pattern);
 
 #endif  /* _DR_TYPES_H_ */
-- 
cgit v1.2.3-70-g09d2


From 2533e726f4725b00a43b2d087480c7f2f414b56d Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:21:05 +0300
Subject: net/mlx5: DR, Split chunk allocation to HW-dependent ways

This way we are able to allocate chunk for modify_headers from 2 types:
STEv0 that is allocated from the action area, and STEv1 that is allocating
the chunks from the special area for patterns.

Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c        | 22 ++-------
 .../ethernet/mellanox/mlx5/core/steering/dr_ste.c  | 57 ++++++++++++++++++++++
 .../ethernet/mellanox/mlx5/core/steering/dr_ste.h  |  2 +
 .../mellanox/mlx5/core/steering/dr_ste_v1.c        | 29 +++++++++++
 .../mellanox/mlx5/core/steering/dr_ste_v1.h        |  2 +
 .../mellanox/mlx5/core/steering/dr_ste_v2.c        |  2 +
 .../mellanox/mlx5/core/steering/dr_types.h         |  2 +
 7 files changed, 98 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 724e4d9e70ac..732a4002eab5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -1961,7 +1961,6 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 					  __be64 actions[],
 					  struct mlx5dr_action *action)
 {
-	struct mlx5dr_icm_chunk *chunk;
 	u32 max_hw_actions;
 	u32 num_hw_actions;
 	u32 num_sw_actions;
@@ -1978,15 +1977,9 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 		return -EINVAL;
 	}
 
-	chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool, DR_CHUNK_SIZE_16);
-	if (!chunk)
-		return -ENOMEM;
-
 	hw_actions = kcalloc(1, max_hw_actions * DR_MODIFY_ACTION_SIZE, GFP_KERNEL);
-	if (!hw_actions) {
-		ret = -ENOMEM;
-		goto free_chunk;
-	}
+	if (!hw_actions)
+		return -ENOMEM;
 
 	ret = dr_actions_convert_modify_header(action,
 					       max_hw_actions,
@@ -1998,15 +1991,11 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 	if (ret)
 		goto free_hw_actions;
 
-	action->rewrite->chunk = chunk;
 	action->rewrite->modify_ttl = modify_ttl;
 	action->rewrite->data = (u8 *)hw_actions;
 	action->rewrite->num_of_actions = num_hw_actions;
-	action->rewrite->index = (mlx5dr_icm_pool_get_chunk_icm_addr(chunk) -
-				  dmn->info.caps.hdr_modify_icm_addr) /
-				  DR_ACTION_CACHE_LINE_SIZE;
 
-	ret = mlx5dr_send_postsend_action(dmn, action);
+	ret = mlx5dr_ste_alloc_modify_hdr(action);
 	if (ret)
 		goto free_hw_actions;
 
@@ -2014,8 +2003,6 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 
 free_hw_actions:
 	kfree(hw_actions);
-free_chunk:
-	mlx5dr_icm_free_chunk(chunk);
 	return ret;
 }
 
@@ -2171,8 +2158,7 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 		refcount_dec(&action->reformat->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_MODIFY_HDR:
-		mlx5dr_icm_free_chunk(action->rewrite->chunk);
-		kfree(action->rewrite->data);
+		mlx5dr_ste_free_modify_hdr(action);
 		refcount_dec(&action->rewrite->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_SAMPLER:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c
index 1e15f605df6e..9413aaf51251 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c
@@ -633,6 +633,63 @@ int mlx5dr_ste_set_action_decap_l3_list(struct mlx5dr_ste_ctx *ste_ctx,
 						 used_hw_action_num);
 }
 
+static int
+dr_ste_alloc_modify_hdr_chunk(struct mlx5dr_action *action)
+{
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
+	u32 chunk_size;
+	int ret;
+
+	chunk_size = ilog2(roundup_pow_of_two(action->rewrite->num_of_actions));
+
+	/* HW modify action index granularity is at least 64B */
+	chunk_size = max_t(u32, chunk_size, DR_CHUNK_SIZE_8);
+
+	action->rewrite->chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool,
+							chunk_size);
+	if (!action->rewrite->chunk)
+		return -ENOMEM;
+
+	action->rewrite->index = (mlx5dr_icm_pool_get_chunk_icm_addr(action->rewrite->chunk) -
+				  dmn->info.caps.hdr_modify_icm_addr) /
+				 DR_ACTION_CACHE_LINE_SIZE;
+
+	ret = mlx5dr_send_postsend_action(action->rewrite->dmn, action);
+	if (ret)
+		goto free_chunk;
+
+	return 0;
+
+free_chunk:
+	mlx5dr_icm_free_chunk(action->rewrite->chunk);
+	return -ENOMEM;
+}
+
+static void dr_ste_free_modify_hdr_chunk(struct mlx5dr_action *action)
+{
+	mlx5dr_icm_free_chunk(action->rewrite->chunk);
+}
+
+int mlx5dr_ste_alloc_modify_hdr(struct mlx5dr_action *action)
+{
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
+
+	if (mlx5dr_domain_is_support_ptrn_arg(dmn))
+		return dmn->ste_ctx->alloc_modify_hdr_chunk(action);
+
+	return dr_ste_alloc_modify_hdr_chunk(action);
+}
+
+void mlx5dr_ste_free_modify_hdr(struct mlx5dr_action *action)
+{
+	struct mlx5dr_domain *dmn = action->rewrite->dmn;
+
+	if (mlx5dr_domain_is_support_ptrn_arg(dmn))
+		return dmn->ste_ctx->dealloc_modify_hdr_chunk(action);
+
+	return dr_ste_free_modify_hdr_chunk(action);
+}
+
 static int dr_ste_build_pre_check_spec(struct mlx5dr_domain *dmn,
 				       struct mlx5dr_match_spec *spec)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h
index 7075142bcfb6..54a6619c3ecb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h
@@ -195,6 +195,8 @@ struct mlx5dr_ste_ctx {
 					u8 *hw_action,
 					u32 hw_action_sz,
 					u16 *used_hw_action_num);
+	int (*alloc_modify_hdr_chunk)(struct mlx5dr_action *action);
+	void (*dealloc_modify_hdr_chunk)(struct mlx5dr_action *action);
 
 	/* Send */
 	void (*prepare_for_postsend)(u8 *hw_ste_p, u32 ste_size);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
index 27cc6931bbde..cf8508139f55 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
@@ -2176,6 +2176,32 @@ dr_ste_v1_build_tnl_gtpu_flex_parser_1_init(struct mlx5dr_ste_build *sb,
 	sb->ste_build_tag_func = &dr_ste_v1_build_tnl_gtpu_flex_parser_1_tag;
 }
 
+int dr_ste_v1_alloc_modify_hdr_ptrn_arg(struct mlx5dr_action *action)
+{
+	struct mlx5dr_ptrn_mgr *ptrn_mgr;
+
+	ptrn_mgr = action->rewrite->dmn->ptrn_mgr;
+	if (!ptrn_mgr)
+		return -EOPNOTSUPP;
+
+	action->rewrite->ptrn =
+		mlx5dr_ptrn_cache_get_pattern(ptrn_mgr,
+					      action->rewrite->num_of_actions,
+					      action->rewrite->data);
+	if (!action->rewrite->ptrn) {
+		mlx5dr_err(action->rewrite->dmn, "Failed to get pattern\n");
+		return -EAGAIN;
+	}
+
+	return 0;
+}
+
+void dr_ste_v1_free_modify_hdr_ptrn_arg(struct mlx5dr_action *action)
+{
+	mlx5dr_ptrn_cache_put_pattern(action->rewrite->dmn->ptrn_mgr,
+				      action->rewrite->ptrn);
+}
+
 static struct mlx5dr_ste_ctx ste_ctx_v1 = {
 	/* Builders */
 	.build_eth_l2_src_dst_init	= &dr_ste_v1_build_eth_l2_src_dst_init,
@@ -2232,6 +2258,9 @@ static struct mlx5dr_ste_ctx ste_ctx_v1 = {
 	.set_action_add			= &dr_ste_v1_set_action_add,
 	.set_action_copy		= &dr_ste_v1_set_action_copy,
 	.set_action_decap_l3_list	= &dr_ste_v1_set_action_decap_l3_list,
+	.alloc_modify_hdr_chunk		= &dr_ste_v1_alloc_modify_hdr_ptrn_arg,
+	.dealloc_modify_hdr_chunk	= &dr_ste_v1_free_modify_hdr_ptrn_arg,
+
 	/* Send */
 	.prepare_for_postsend		= &dr_ste_v1_prepare_for_postsend,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h
index b5c0f0f8392f..e2fc69867088 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h
@@ -31,6 +31,8 @@ void dr_ste_v1_set_action_copy(u8 *d_action, u8 dst_hw_field, u8 dst_shifter,
 			       u8 dst_len, u8 src_hw_field, u8 src_shifter);
 int dr_ste_v1_set_action_decap_l3_list(void *data, u32 data_sz, u8 *hw_action,
 				       u32 hw_action_sz, u16 *used_hw_action_num);
+int dr_ste_v1_alloc_modify_hdr_ptrn_arg(struct mlx5dr_action *action);
+void dr_ste_v1_free_modify_hdr_ptrn_arg(struct mlx5dr_action *action);
 void dr_ste_v1_build_eth_l2_src_dst_init(struct mlx5dr_ste_build *sb,
 					 struct mlx5dr_match_param *mask);
 void dr_ste_v1_build_eth_l3_ipv6_dst_init(struct mlx5dr_ste_build *sb,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c
index cf1a3c9a1cf4..808b013cf48c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c
@@ -221,6 +221,8 @@ static struct mlx5dr_ste_ctx ste_ctx_v2 = {
 	.set_action_add			= &dr_ste_v1_set_action_add,
 	.set_action_copy		= &dr_ste_v1_set_action_copy,
 	.set_action_decap_l3_list	= &dr_ste_v1_set_action_decap_l3_list,
+	.alloc_modify_hdr_chunk		= &dr_ste_v1_alloc_modify_hdr_ptrn_arg,
+	.dealloc_modify_hdr_chunk	= &dr_ste_v1_free_modify_hdr_ptrn_arg,
 
 	/* Send */
 	.prepare_for_postsend		= &dr_ste_v1_prepare_for_postsend,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 097f1f389b76..a1c549fed9ca 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -335,6 +335,8 @@ int mlx5dr_ste_set_action_decap_l3_list(struct mlx5dr_ste_ctx *ste_ctx,
 					u8 *hw_action,
 					u32 hw_action_sz,
 					u16 *used_hw_action_num);
+int mlx5dr_ste_alloc_modify_hdr(struct mlx5dr_action *action);
+void mlx5dr_ste_free_modify_hdr(struct mlx5dr_action *action);
 
 const struct mlx5dr_ste_action_modify_field *
 mlx5dr_ste_conv_modify_hdr_sw_field(struct mlx5dr_ste_ctx *ste_ctx, u16 sw_field);
-- 
cgit v1.2.3-70-g09d2


From b7ba743a2f1c7d14f281c6d1a2877984008e28e9 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:21:10 +0300
Subject: net/mlx5: DR, Check for modify_header_argument device capabilities

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c   | 11 +++++++++++
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h |  3 +++
 2 files changed, 14 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index 229f3684100c..e2cbc2b5bc27 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -132,6 +132,17 @@ int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev,
 
 	caps->isolate_vl_tc = MLX5_CAP_GEN(mdev, isolate_vl_tc_new);
 
+	caps->support_modify_argument =
+		MLX5_CAP_GEN_64(mdev, general_obj_types) &
+		MLX5_GENERAL_OBJ_TYPES_CAP_HEADER_MODIFY_ARGUMENT;
+
+	if (caps->support_modify_argument) {
+		caps->log_header_modify_argument_granularity =
+			MLX5_CAP_GEN(mdev, log_header_modify_argument_granularity);
+		caps->log_header_modify_argument_max_alloc =
+			MLX5_CAP_GEN(mdev, log_header_modify_argument_max_alloc);
+	}
+
 	/* geneve_tlv_option_0_exist is the indication of
 	 * STE support for lookup type flex_parser_ok
 	 */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index a1c549fed9ca..9187e9d6ea54 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -896,6 +896,9 @@ struct mlx5dr_cmd_caps {
 	struct mlx5dr_vports vports;
 	bool prio_tag_required;
 	struct mlx5dr_roce_cap roce_caps;
+	u16 log_header_modify_argument_granularity;
+	u16 log_header_modify_argument_max_alloc;
+	bool support_modify_argument;
 	u8 is_ecpf:1;
 	u8 isolate_vl_tc:1;
 };
-- 
cgit v1.2.3-70-g09d2


From de69696b6eeef3c0704fdb3bcd473ddb4757130e Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:18:29 +0300
Subject: net/mlx5: DR, Add create/destroy for modify-header-argument general
 object

Add functions for creation/destruction of the new type of general object.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  | 43 ++++++++++++++++++++++
 .../mellanox/mlx5/core/steering/dr_types.h         |  6 +++
 2 files changed, 49 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index e2cbc2b5bc27..3835ba3f4dda 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -693,6 +693,49 @@ int mlx5dr_cmd_query_gid(struct mlx5_core_dev *mdev, u8 vhca_port_num,
 	return 0;
 }
 
+int mlx5dr_cmd_create_modify_header_arg(struct mlx5_core_dev *dev,
+					u16 log_obj_range, u32 pd,
+					u32 *obj_id)
+{
+	u32 in[MLX5_ST_SZ_DW(create_modify_header_arg_in)] = {};
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {};
+	void *attr;
+	int ret;
+
+	attr = MLX5_ADDR_OF(create_modify_header_arg_in, in, hdr);
+	MLX5_SET(general_obj_in_cmd_hdr, attr, opcode,
+		 MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, attr, obj_type,
+		 MLX5_OBJ_TYPE_HEADER_MODIFY_ARGUMENT);
+	MLX5_SET(general_obj_in_cmd_hdr, attr,
+		 op_param.create.log_obj_range, log_obj_range);
+
+	attr = MLX5_ADDR_OF(create_modify_header_arg_in, in, arg);
+	MLX5_SET(modify_header_arg, attr, access_pd, pd);
+
+	ret = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+	if (ret)
+		return ret;
+
+	*obj_id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return 0;
+}
+
+void mlx5dr_cmd_destroy_modify_header_arg(struct mlx5_core_dev *dev,
+					  u32 obj_id)
+{
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {};
+	u32 in[MLX5_ST_SZ_DW(general_obj_in_cmd_hdr)] = {};
+
+	MLX5_SET(general_obj_in_cmd_hdr, in, opcode,
+		 MLX5_CMD_OP_DESTROY_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, in, obj_type,
+		 MLX5_OBJ_TYPE_HEADER_MODIFY_ARGUMENT);
+	MLX5_SET(general_obj_in_cmd_hdr, in, obj_id, obj_id);
+
+	mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
 static int mlx5dr_cmd_set_extended_dest(struct mlx5_core_dev *dev,
 					struct mlx5dr_cmd_fte_info *fte,
 					bool *extended_dest)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 9187e9d6ea54..0075e2c7a441 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1357,6 +1357,12 @@ struct mlx5dr_cmd_gid_attr {
 int mlx5dr_cmd_query_gid(struct mlx5_core_dev *mdev, u8 vhca_port_num,
 			 u16 index, struct mlx5dr_cmd_gid_attr *attr);
 
+int mlx5dr_cmd_create_modify_header_arg(struct mlx5_core_dev *dev,
+					u16 log_obj_range, u32 pd,
+					u32 *obj_id);
+void mlx5dr_cmd_destroy_modify_header_arg(struct mlx5_core_dev *dev,
+					  u32 obj_id);
+
 struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
 					       enum mlx5dr_icm_type icm_type);
 void mlx5dr_icm_pool_destroy(struct mlx5dr_icm_pool *pool);
-- 
cgit v1.2.3-70-g09d2


From 4605fc0a2b652070f45b0a17967c2cb11a9e5a6e Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 8 Nov 2022 14:27:03 +0200
Subject: net/mlx5: DR, Add support for writing modify header argument

The accelerated modify header arguments are written in the HW area
with special WQE and specific data format.
New function was added to support writing of new argument type.
Note that GTA WQE is larger than READ and WRITE, so the queue
management logic was updated to support this.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_send.c | 167 ++++++++++++++++++---
 .../mellanox/mlx5/core/steering/dr_types.h         |   3 +
 2 files changed, 150 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index 78756840d263..d7c7363f9096 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -20,6 +20,7 @@ struct dr_data_seg {
 
 enum send_info_type {
 	WRITE_ICM = 0,
+	GTA_ARG   = 1,
 };
 
 struct postsend_info {
@@ -269,6 +270,7 @@ static struct mlx5dr_qp *dr_create_rc_qp(struct mlx5_core_dev *mdev,
 	dr_qp->rq.wqe_cnt = 4;
 	dr_qp->sq.pc = 0;
 	dr_qp->sq.cc = 0;
+	dr_qp->sq.head = 0;
 	dr_qp->sq.wqe_cnt = roundup_pow_of_two(attr->max_send_wr);
 
 	MLX5_SET(qpc, temp_qpc, log_rq_stride, ilog2(MLX5_SEND_WQE_DS) - 4);
@@ -367,39 +369,113 @@ static void dr_cmd_notify_hw(struct mlx5dr_qp *dr_qp, void *ctrl)
 	mlx5_write64(ctrl, dr_qp->uar->map + MLX5_BF_OFFSET);
 }
 
-static void dr_rdma_segments(struct mlx5dr_qp *dr_qp, u64 remote_addr,
-			     u32 rkey, struct dr_data_seg *data_seg,
-			     u32 opcode, bool notify_hw)
+static void
+dr_rdma_handle_flow_access_arg_segments(struct mlx5_wqe_ctrl_seg *wq_ctrl,
+					u32 remote_addr,
+					struct dr_data_seg *data_seg,
+					int *size)
 {
-	struct mlx5_wqe_raddr_seg *wq_raddr;
-	struct mlx5_wqe_ctrl_seg *wq_ctrl;
-	struct mlx5_wqe_data_seg *wq_dseg;
-	unsigned int size;
-	unsigned int idx;
+	struct mlx5_wqe_header_modify_argument_update_seg *wq_arg_seg;
+	struct mlx5_wqe_flow_update_ctrl_seg *wq_flow_seg;
 
-	size = sizeof(*wq_ctrl) / 16 + sizeof(*wq_dseg) / 16 +
-		sizeof(*wq_raddr) / 16;
+	wq_ctrl->general_id = cpu_to_be32(remote_addr);
+	wq_flow_seg = (void *)(wq_ctrl + 1);
 
-	idx = dr_qp->sq.pc & (dr_qp->sq.wqe_cnt - 1);
+	/* mlx5_wqe_flow_update_ctrl_seg - all reserved */
+	memset(wq_flow_seg, 0, sizeof(*wq_flow_seg));
+	wq_arg_seg = (void *)(wq_flow_seg + 1);
+
+	memcpy(wq_arg_seg->argument_list,
+	       (void *)(uintptr_t)data_seg->addr,
+	       data_seg->length);
+
+	*size = (sizeof(*wq_ctrl) +      /* WQE ctrl segment */
+		 sizeof(*wq_flow_seg) +  /* WQE flow update ctrl seg - reserved */
+		 sizeof(*wq_arg_seg)) /  /* WQE hdr modify arg seg - data */
+		MLX5_SEND_WQE_DS;
+}
+
+static void
+dr_rdma_handle_icm_write_segments(struct mlx5_wqe_ctrl_seg *wq_ctrl,
+				  u64 remote_addr,
+				  u32 rkey,
+				  struct dr_data_seg *data_seg,
+				  unsigned int *size)
+{
+	struct mlx5_wqe_raddr_seg *wq_raddr;
+	struct mlx5_wqe_data_seg *wq_dseg;
 
-	wq_ctrl = mlx5_wq_cyc_get_wqe(&dr_qp->wq.sq, idx);
-	wq_ctrl->imm = 0;
-	wq_ctrl->fm_ce_se = (data_seg->send_flags) ?
-		MLX5_WQE_CTRL_CQ_UPDATE : 0;
-	wq_ctrl->opmod_idx_opcode = cpu_to_be32(((dr_qp->sq.pc & 0xffff) << 8) |
-						opcode);
-	wq_ctrl->qpn_ds = cpu_to_be32(size | dr_qp->qpn << 8);
 	wq_raddr = (void *)(wq_ctrl + 1);
+
 	wq_raddr->raddr = cpu_to_be64(remote_addr);
 	wq_raddr->rkey = cpu_to_be32(rkey);
 	wq_raddr->reserved = 0;
 
 	wq_dseg = (void *)(wq_raddr + 1);
+
 	wq_dseg->byte_count = cpu_to_be32(data_seg->length);
 	wq_dseg->lkey = cpu_to_be32(data_seg->lkey);
 	wq_dseg->addr = cpu_to_be64(data_seg->addr);
 
-	dr_qp->sq.wqe_head[idx] = dr_qp->sq.pc++;
+	*size = (sizeof(*wq_ctrl) +    /* WQE ctrl segment */
+		 sizeof(*wq_dseg) +    /* WQE data segment */
+		 sizeof(*wq_raddr)) /  /* WQE remote addr segment */
+		MLX5_SEND_WQE_DS;
+}
+
+static void dr_set_ctrl_seg(struct mlx5_wqe_ctrl_seg *wq_ctrl,
+			    struct dr_data_seg *data_seg)
+{
+	wq_ctrl->signature = 0;
+	wq_ctrl->rsvd[0] = 0;
+	wq_ctrl->rsvd[1] = 0;
+	wq_ctrl->fm_ce_se = data_seg->send_flags & IB_SEND_SIGNALED ?
+				MLX5_WQE_CTRL_CQ_UPDATE : 0;
+	wq_ctrl->imm = 0;
+}
+
+static void dr_rdma_segments(struct mlx5dr_qp *dr_qp, u64 remote_addr,
+			     u32 rkey, struct dr_data_seg *data_seg,
+			     u32 opcode, bool notify_hw)
+{
+	struct mlx5_wqe_ctrl_seg *wq_ctrl;
+	int opcode_mod = 0;
+	unsigned int size;
+	unsigned int idx;
+
+	idx = dr_qp->sq.pc & (dr_qp->sq.wqe_cnt - 1);
+
+	wq_ctrl = mlx5_wq_cyc_get_wqe(&dr_qp->wq.sq, idx);
+	dr_set_ctrl_seg(wq_ctrl, data_seg);
+
+	switch (opcode) {
+	case MLX5_OPCODE_RDMA_READ:
+	case MLX5_OPCODE_RDMA_WRITE:
+		dr_rdma_handle_icm_write_segments(wq_ctrl, remote_addr,
+						  rkey, data_seg, &size);
+		break;
+	case MLX5_OPCODE_FLOW_TBL_ACCESS:
+		opcode_mod = MLX5_CMD_OP_MOD_UPDATE_HEADER_MODIFY_ARGUMENT;
+		dr_rdma_handle_flow_access_arg_segments(wq_ctrl, remote_addr,
+							data_seg, &size);
+		break;
+	default:
+		WARN(true, "illegal opcode %d", opcode);
+		return;
+	}
+
+	/* --------------------------------------------------------
+	 * |opcode_mod (8 bit)|wqe_index (16 bits)| opcod (8 bits)|
+	 * --------------------------------------------------------
+	 */
+	wq_ctrl->opmod_idx_opcode =
+		cpu_to_be32((opcode_mod << 24) |
+			    ((dr_qp->sq.pc & 0xffff) << 8) |
+			    opcode);
+	wq_ctrl->qpn_ds = cpu_to_be32(size | dr_qp->qpn << 8);
+
+	dr_qp->sq.pc += DIV_ROUND_UP(size * 16, MLX5_SEND_WQE_BB);
+	dr_qp->sq.wqe_head[idx] = dr_qp->sq.head++;
 
 	if (notify_hw)
 		dr_cmd_notify_hw(dr_qp, wq_ctrl);
@@ -412,7 +488,11 @@ static void dr_post_send(struct mlx5dr_qp *dr_qp, struct postsend_info *send_inf
 				 &send_info->write, MLX5_OPCODE_RDMA_WRITE, false);
 		dr_rdma_segments(dr_qp, send_info->remote_addr, send_info->rkey,
 				 &send_info->read, MLX5_OPCODE_RDMA_READ, true);
+	} else { /* GTA_ARG */
+		dr_rdma_segments(dr_qp, send_info->remote_addr, send_info->rkey,
+				 &send_info->write, MLX5_OPCODE_FLOW_TBL_ACCESS, true);
 	}
+
 }
 
 /**
@@ -478,11 +558,23 @@ static int dr_handle_pending_wc(struct mlx5dr_domain *dmn,
 		} else if (ne == 1) {
 			send_ring->pending_wqe -= send_ring->signal_th;
 		}
-	} while (is_drain && send_ring->pending_wqe);
+	} while (ne == 1 ||
+		 (is_drain && send_ring->pending_wqe  >= send_ring->signal_th));
 
 	return 0;
 }
 
+static void dr_fill_write_args_segs(struct mlx5dr_send_ring *send_ring,
+				    struct postsend_info *send_info)
+{
+	send_ring->pending_wqe++;
+
+	if (send_ring->pending_wqe % send_ring->signal_th == 0)
+		send_info->write.send_flags |= IB_SEND_SIGNALED;
+	else
+		send_info->write.send_flags = 0;
+}
+
 static void dr_fill_write_icm_segs(struct mlx5dr_domain *dmn,
 				   struct mlx5dr_send_ring *send_ring,
 				   struct postsend_info *send_info)
@@ -526,6 +618,8 @@ static void dr_fill_data_segs(struct mlx5dr_domain *dmn,
 {
 	if (send_info->type == WRITE_ICM)
 		dr_fill_write_icm_segs(dmn, send_ring, send_info);
+	else /* args */
+		dr_fill_write_args_segs(send_ring, send_info);
 }
 
 static int dr_postsend_icm_data(struct mlx5dr_domain *dmn,
@@ -774,6 +868,39 @@ int mlx5dr_send_postsend_pattern(struct mlx5dr_domain *dmn,
 	return 0;
 }
 
+int mlx5dr_send_postsend_args(struct mlx5dr_domain *dmn, u64 arg_id,
+			      u16 num_of_actions, u8 *actions_data)
+{
+	int data_len, iter = 0, cur_sent;
+	u64 addr;
+	int ret;
+
+	addr = (uintptr_t)actions_data;
+	data_len = num_of_actions * DR_MODIFY_ACTION_SIZE;
+
+	do {
+		struct postsend_info send_info = {};
+
+		send_info.type = GTA_ARG;
+		send_info.write.addr = addr;
+		cur_sent = min_t(u32, data_len, DR_ACTION_CACHE_LINE_SIZE);
+		send_info.write.length = cur_sent;
+		send_info.write.lkey = 0;
+		send_info.remote_addr = arg_id + iter;
+
+		ret = dr_postsend_icm_data(dmn, &send_info);
+		if (ret)
+			goto out;
+
+		iter++;
+		addr += cur_sent;
+		data_len -= cur_sent;
+	} while (data_len > 0);
+
+out:
+	return ret;
+}
+
 static int dr_modify_qp_rst2init(struct mlx5_core_dev *mdev,
 				 struct mlx5dr_qp *dr_qp,
 				 int port)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 0075e2c7a441..7b35f78a84a2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1397,6 +1397,7 @@ struct mlx5dr_qp {
 	struct mlx5_wq_ctrl wq_ctrl;
 	u32 qpn;
 	struct {
+		unsigned int head;
 		unsigned int pc;
 		unsigned int cc;
 		unsigned int size;
@@ -1473,6 +1474,8 @@ int mlx5dr_send_postsend_pattern(struct mlx5dr_domain *dmn,
 				 struct mlx5dr_icm_chunk *chunk,
 				 u16 num_of_actions,
 				 u8 *data);
+int mlx5dr_send_postsend_args(struct mlx5dr_domain *dmn, u64 arg_id,
+			      u16 num_of_actions, u8 *actions_data);
 
 int mlx5dr_send_info_pool_create(struct mlx5dr_domain *dmn);
 void mlx5dr_send_info_pool_destroy(struct mlx5dr_domain *dmn);
-- 
cgit v1.2.3-70-g09d2


From 7d7c9453d679fe55d72d63bacb3b639cd963ebc0 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 15 Nov 2022 00:33:25 +0200
Subject: net/mlx5: DR, Read ICM memory into dedicated buffer

Instead of using the write buffer for reading we will use a dedicated
buffer only for reading ICM memory.
Due to the new support for args, we can have a case with pending_wc
being odd number, and with reading into the same write buffer, it is
possible to overwrite next write on the same slot.
For example:
pending_wc is 17 so the buffer for write is:
   | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
and we have requests as follows:
   r wr wr wr wr wr wr wr wr
Now, the first read will be written into the last write because we use
the same buffer for read and write, before it was written to the HW and
we will have a wrong data in the ICM area.

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_send.c  | 21 ++++++++++++++++-----
 .../ethernet/mellanox/mlx5/core/steering/dr_types.h |  5 +----
 2 files changed, 17 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index d7c7363f9096..d052d469d4df 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -602,9 +602,10 @@ static void dr_fill_write_icm_segs(struct mlx5dr_domain *dmn,
 
 	send_ring->pending_wqe++;
 	send_info->read.length = send_info->write.length;
-	/* Read into the same write area */
-	send_info->read.addr = (uintptr_t)send_info->write.addr;
-	send_info->read.lkey = send_ring->mr->mkey;
+
+	/* Read into dedicated sync buffer */
+	send_info->read.addr = (uintptr_t)send_ring->sync_mr->dma_addr;
+	send_info->read.lkey = send_ring->sync_mr->mkey;
 
 	if (send_ring->pending_wqe % send_ring->signal_th == 0)
 		send_info->read.send_flags = IB_SEND_SIGNALED;
@@ -1288,16 +1289,25 @@ int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn)
 		goto free_mem;
 	}
 
+	dmn->send_ring->sync_buff = kzalloc(dmn->send_ring->max_post_send_size,
+					    GFP_KERNEL);
+	if (!dmn->send_ring->sync_buff) {
+		ret = -ENOMEM;
+		goto clean_mr;
+	}
+
 	dmn->send_ring->sync_mr = dr_reg_mr(dmn->mdev,
 					    dmn->pdn, dmn->send_ring->sync_buff,
-					    MIN_READ_SYNC);
+					    dmn->send_ring->max_post_send_size);
 	if (!dmn->send_ring->sync_mr) {
 		ret = -ENOMEM;
-		goto clean_mr;
+		goto free_sync_mem;
 	}
 
 	return 0;
 
+free_sync_mem:
+	kfree(dmn->send_ring->sync_buff);
 clean_mr:
 	dr_dereg_mr(dmn->mdev, dmn->send_ring->mr);
 free_mem:
@@ -1320,6 +1330,7 @@ void mlx5dr_send_ring_free(struct mlx5dr_domain *dmn,
 	dr_dereg_mr(dmn->mdev, send_ring->sync_mr);
 	dr_dereg_mr(dmn->mdev, send_ring->mr);
 	kfree(send_ring->buf);
+	kfree(send_ring->sync_buff);
 	kfree(send_ring);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 7b35f78a84a2..81d7ac6d6258 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1429,9 +1429,6 @@ struct mlx5dr_mr {
 	size_t size;
 };
 
-#define MAX_SEND_CQE		64
-#define MIN_READ_SYNC		64
-
 struct mlx5dr_send_ring {
 	struct mlx5dr_cq *cq;
 	struct mlx5dr_qp *qp;
@@ -1446,7 +1443,7 @@ struct mlx5dr_send_ring {
 	u32 tx_head;
 	void *buf;
 	u32 buf_size;
-	u8 sync_buff[MIN_READ_SYNC];
+	u8 *sync_buff;
 	struct mlx5dr_mr *sync_mr;
 	spinlock_t lock; /* Protect the data path of the send ring */
 	bool err_state; /* send_ring is not usable in err state */
-- 
cgit v1.2.3-70-g09d2


From 17dc71c336aac381f59ba541cf85fb0c192d1c1c Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 15 Nov 2022 00:11:38 +0200
Subject: net/mlx5: DR, Fix QP continuous allocation

When allocating a QP we allocate an RQ and an SQ, the RQ is stored first
in memory and followed by the SQ.
This allocation is not physically continiuos - it may span across different
physical pages. SW Steering code always writes in pairs: 1BB write + 1BB read,
or 2 continuous BBs of GTA WQE.

This lead to an issue where RQ allocation was 4x16 which is equal to 1 WQE BB,
causing 1 BB offset in the page and splitting the GTA WQE between different
physical pages.

The solution was to create the RQ with a even number of BBs and to have the
RQ aligned to a page.

Signed-off-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index d052d469d4df..4a5ae86e2b62 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -267,7 +267,7 @@ static struct mlx5dr_qp *dr_create_rc_qp(struct mlx5_core_dev *mdev,
 
 	dr_qp->rq.pc = 0;
 	dr_qp->rq.cc = 0;
-	dr_qp->rq.wqe_cnt = 4;
+	dr_qp->rq.wqe_cnt = 256;
 	dr_qp->sq.pc = 0;
 	dr_qp->sq.cc = 0;
 	dr_qp->sq.head = 0;
-- 
cgit v1.2.3-70-g09d2


From 608d4f1769d80824d0b02eca73243d377468ee68 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Mon, 27 Mar 2023 23:26:55 +0300
Subject: net/mlx5: DR, Add modify header arg pool mechanism

Added new mechanism for handling arguments for modify-header action.
The new action "accelerated modify-header" asks for the arguments from
separated area from the pattern, this area accessed via general objects.
Handling of these object is done via the pool-manager struct.

When the new header patterns are supported, while loading the domain,
a few pools for argument creations will be created. The requests for
allocating/deallocating arg objects are done via the pool manager API.

Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
 .../ethernet/mellanox/mlx5/core/steering/dr_arg.c  | 273 +++++++++++++++++++++
 .../mellanox/mlx5/core/steering/dr_domain.c        |  12 +
 .../mellanox/mlx5/core/steering/dr_types.h         |  18 ++
 4 files changed, 304 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_arg.c

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 39c2c8dc7e07..ca3c66cd47ec 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -113,7 +113,7 @@ mlx5_core-$(CONFIG_MLX5_SW_STEERING) += steering/dr_domain.o steering/dr_table.o
 					steering/dr_cmd.o steering/dr_fw.o \
 					steering/dr_action.o steering/fs_dr.o \
 					steering/dr_definer.o steering/dr_ptrn.o \
-					steering/dr_dbg.o lib/smfs.o
+					steering/dr_arg.o steering/dr_dbg.o lib/smfs.o
 #
 # SF device
 #
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_arg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_arg.c
new file mode 100644
index 000000000000..01ed6442095d
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_arg.c
@@ -0,0 +1,273 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+// Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+#include "dr_types.h"
+
+#define DR_ICM_MODIFY_HDR_GRANULARITY_4K 12
+
+/* modify-header arg pool */
+enum dr_arg_chunk_size {
+	DR_ARG_CHUNK_SIZE_1,
+	DR_ARG_CHUNK_SIZE_MIN = DR_ARG_CHUNK_SIZE_1, /* keep updated when changing */
+	DR_ARG_CHUNK_SIZE_2,
+	DR_ARG_CHUNK_SIZE_3,
+	DR_ARG_CHUNK_SIZE_4,
+	DR_ARG_CHUNK_SIZE_MAX,
+};
+
+/* argument pool area */
+struct dr_arg_pool {
+	enum dr_arg_chunk_size log_chunk_size;
+	struct mlx5dr_domain *dmn;
+	struct list_head free_list;
+	struct mutex mutex; /* protect arg pool */
+};
+
+struct mlx5dr_arg_mgr {
+	struct mlx5dr_domain *dmn;
+	struct dr_arg_pool *pools[DR_ARG_CHUNK_SIZE_MAX];
+};
+
+static int dr_arg_pool_alloc_objs(struct dr_arg_pool *pool)
+{
+	struct mlx5dr_arg_obj *arg_obj, *tmp_arg;
+	struct list_head cur_list;
+	u16 object_range;
+	int num_of_objects;
+	u32 obj_id = 0;
+	int i, ret;
+
+	INIT_LIST_HEAD(&cur_list);
+
+	object_range =
+		pool->dmn->info.caps.log_header_modify_argument_granularity;
+
+	object_range =
+		max_t(u32, pool->dmn->info.caps.log_header_modify_argument_granularity,
+		      DR_ICM_MODIFY_HDR_GRANULARITY_4K);
+	object_range =
+		min_t(u32, pool->dmn->info.caps.log_header_modify_argument_max_alloc,
+		      object_range);
+
+	if (pool->log_chunk_size > object_range) {
+		mlx5dr_err(pool->dmn, "Required chunk size (%d) is not supported\n",
+			   pool->log_chunk_size);
+		return -ENOMEM;
+	}
+
+	num_of_objects = (1 << (object_range - pool->log_chunk_size));
+	/* Only one devx object per range */
+	ret = mlx5dr_cmd_create_modify_header_arg(pool->dmn->mdev,
+						  object_range,
+						  pool->dmn->pdn,
+						  &obj_id);
+	if (ret) {
+		mlx5dr_err(pool->dmn, "failed allocating object with range: %d:\n",
+			   object_range);
+		return -EAGAIN;
+	}
+
+	for (i = 0; i < num_of_objects; i++) {
+		arg_obj = kzalloc(sizeof(*arg_obj), GFP_KERNEL);
+		if (!arg_obj) {
+			ret = -ENOMEM;
+			goto clean_arg_obj;
+		}
+
+		arg_obj->log_chunk_size = pool->log_chunk_size;
+
+		list_add_tail(&arg_obj->list_node, &cur_list);
+
+		arg_obj->obj_id = obj_id;
+		arg_obj->obj_offset = i * (1 << pool->log_chunk_size);
+	}
+	list_splice_tail_init(&cur_list, &pool->free_list);
+
+	return 0;
+
+clean_arg_obj:
+	mlx5dr_cmd_destroy_modify_header_arg(pool->dmn->mdev, obj_id);
+	list_for_each_entry_safe(arg_obj, tmp_arg, &cur_list, list_node) {
+		list_del(&arg_obj->list_node);
+		kfree(arg_obj);
+	}
+	return ret;
+}
+
+static struct mlx5dr_arg_obj *dr_arg_pool_get_arg_obj(struct dr_arg_pool *pool)
+{
+	struct mlx5dr_arg_obj *arg_obj = NULL;
+	int ret;
+
+	mutex_lock(&pool->mutex);
+	if (list_empty(&pool->free_list)) {
+		ret = dr_arg_pool_alloc_objs(pool);
+		if (ret)
+			goto out;
+	}
+
+	arg_obj = list_first_entry_or_null(&pool->free_list,
+					   struct mlx5dr_arg_obj,
+					   list_node);
+	WARN(!arg_obj, "couldn't get dr arg obj from pool");
+
+	if (arg_obj)
+		list_del_init(&arg_obj->list_node);
+
+out:
+	mutex_unlock(&pool->mutex);
+	return arg_obj;
+}
+
+static void dr_arg_pool_put_arg_obj(struct dr_arg_pool *pool,
+				    struct mlx5dr_arg_obj *arg_obj)
+{
+	mutex_lock(&pool->mutex);
+	list_add(&arg_obj->list_node, &pool->free_list);
+	mutex_unlock(&pool->mutex);
+}
+
+static struct dr_arg_pool *dr_arg_pool_create(struct mlx5dr_domain *dmn,
+					      enum dr_arg_chunk_size chunk_size)
+{
+	struct dr_arg_pool *pool;
+
+	pool = kzalloc(sizeof(*pool), GFP_KERNEL);
+	if (!pool)
+		return NULL;
+
+	pool->dmn = dmn;
+
+	INIT_LIST_HEAD(&pool->free_list);
+	mutex_init(&pool->mutex);
+
+	pool->log_chunk_size = chunk_size;
+	if (dr_arg_pool_alloc_objs(pool))
+		goto free_pool;
+
+	return pool;
+
+free_pool:
+	kfree(pool);
+
+	return NULL;
+}
+
+static void dr_arg_pool_destroy(struct dr_arg_pool *pool)
+{
+	struct mlx5dr_arg_obj *arg_obj, *tmp_arg;
+
+	list_for_each_entry_safe(arg_obj, tmp_arg, &pool->free_list, list_node) {
+		list_del(&arg_obj->list_node);
+		if (!arg_obj->obj_offset) /* the first in range */
+			mlx5dr_cmd_destroy_modify_header_arg(pool->dmn->mdev, arg_obj->obj_id);
+		kfree(arg_obj);
+	}
+
+	mutex_destroy(&pool->mutex);
+	kfree(pool);
+}
+
+static enum dr_arg_chunk_size dr_arg_get_chunk_size(u16 num_of_actions)
+{
+	if (num_of_actions <= 8)
+		return DR_ARG_CHUNK_SIZE_1;
+	if (num_of_actions <= 16)
+		return DR_ARG_CHUNK_SIZE_2;
+	if (num_of_actions <= 32)
+		return DR_ARG_CHUNK_SIZE_3;
+	if (num_of_actions <= 64)
+		return DR_ARG_CHUNK_SIZE_4;
+
+	return DR_ARG_CHUNK_SIZE_MAX;
+}
+
+u32 mlx5dr_arg_get_obj_id(struct mlx5dr_arg_obj *arg_obj)
+{
+	return (arg_obj->obj_id + arg_obj->obj_offset);
+}
+
+struct mlx5dr_arg_obj *mlx5dr_arg_get_obj(struct mlx5dr_arg_mgr *mgr,
+					  u16 num_of_actions,
+					  u8 *data)
+{
+	u32 size = dr_arg_get_chunk_size(num_of_actions);
+	struct mlx5dr_arg_obj *arg_obj;
+	int ret;
+
+	if (size >= DR_ARG_CHUNK_SIZE_MAX)
+		return NULL;
+
+	arg_obj = dr_arg_pool_get_arg_obj(mgr->pools[size]);
+	if (!arg_obj) {
+		mlx5dr_err(mgr->dmn, "Failed allocating args object for modify header\n");
+		return NULL;
+	}
+
+	/* write it into the hw */
+	ret = mlx5dr_send_postsend_args(mgr->dmn,
+					mlx5dr_arg_get_obj_id(arg_obj),
+					num_of_actions, data);
+	if (ret) {
+		mlx5dr_err(mgr->dmn, "Failed writing args object\n");
+		goto put_obj;
+	}
+
+	return arg_obj;
+
+put_obj:
+	mlx5dr_arg_put_obj(mgr, arg_obj);
+	return NULL;
+}
+
+void mlx5dr_arg_put_obj(struct mlx5dr_arg_mgr *mgr,
+			struct mlx5dr_arg_obj *arg_obj)
+{
+	dr_arg_pool_put_arg_obj(mgr->pools[arg_obj->log_chunk_size], arg_obj);
+}
+
+struct mlx5dr_arg_mgr*
+mlx5dr_arg_mgr_create(struct mlx5dr_domain *dmn)
+{
+	struct mlx5dr_arg_mgr *pool_mgr;
+	int i;
+
+	if (!mlx5dr_domain_is_support_ptrn_arg(dmn))
+		return NULL;
+
+	pool_mgr = kzalloc(sizeof(*pool_mgr), GFP_KERNEL);
+	if (!pool_mgr)
+		return NULL;
+
+	pool_mgr->dmn = dmn;
+
+	for (i = 0; i < DR_ARG_CHUNK_SIZE_MAX; i++) {
+		pool_mgr->pools[i] = dr_arg_pool_create(dmn, i);
+		if (!pool_mgr->pools[i])
+			goto clean_pools;
+	}
+
+	return pool_mgr;
+
+clean_pools:
+	for (i--; i >= 0; i--)
+		dr_arg_pool_destroy(pool_mgr->pools[i]);
+
+	kfree(pool_mgr);
+	return NULL;
+}
+
+void mlx5dr_arg_mgr_destroy(struct mlx5dr_arg_mgr *mgr)
+{
+	struct dr_arg_pool **pools;
+	int i;
+
+	if (!mgr)
+		return;
+
+	pools = mgr->pools;
+	for (i = 0; i < DR_ARG_CHUNK_SIZE_MAX; i++)
+		dr_arg_pool_destroy(pools[i]);
+
+	kfree(mgr);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
index 7a0381572c4c..c4545daf179b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
@@ -26,7 +26,18 @@ static int dr_domain_init_modify_header_resources(struct mlx5dr_domain *dmn)
 		return -ENOMEM;
 	}
 
+	/* create argument pool */
+	dmn->arg_mgr = mlx5dr_arg_mgr_create(dmn);
+	if (!dmn->arg_mgr) {
+		mlx5dr_err(dmn, "Couldn't create arg_mgr\n");
+		goto free_modify_header_pattern;
+	}
+
 	return 0;
+
+free_modify_header_pattern:
+	mlx5dr_ptrn_mgr_destroy(dmn->ptrn_mgr);
+	return -ENOMEM;
 }
 
 static void dr_domain_destroy_modify_header_resources(struct mlx5dr_domain *dmn)
@@ -34,6 +45,7 @@ static void dr_domain_destroy_modify_header_resources(struct mlx5dr_domain *dmn)
 	if (!mlx5dr_domain_is_support_ptrn_arg(dmn))
 		return;
 
+	mlx5dr_arg_mgr_destroy(dmn->arg_mgr);
 	mlx5dr_ptrn_mgr_destroy(dmn->ptrn_mgr);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 81d7ac6d6258..e102ceb20e01 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -28,6 +28,8 @@
 #define mlx5dr_dbg(dmn, arg...) mlx5_core_dbg((dmn)->mdev, ##arg)
 
 struct mlx5dr_ptrn_mgr;
+struct mlx5dr_arg_mgr;
+struct mlx5dr_arg_obj;
 
 static inline bool dr_is_flex_parser_0_id(u8 parser_id)
 {
@@ -941,6 +943,7 @@ struct mlx5dr_domain {
 	struct kmem_cache *chunks_kmem_cache;
 	struct kmem_cache *htbls_kmem_cache;
 	struct mlx5dr_ptrn_mgr *ptrn_mgr;
+	struct mlx5dr_arg_mgr *arg_mgr;
 	struct mlx5dr_send_ring *send_ring;
 	struct mlx5dr_domain_info info;
 	struct xarray csum_fts_xa;
@@ -1016,6 +1019,13 @@ struct mlx5dr_ptrn_obj {
 	struct list_head list;
 };
 
+struct mlx5dr_arg_obj {
+	u32 obj_id;
+	u32 obj_offset;
+	struct list_head list_node;
+	u32 log_chunk_size;
+};
+
 struct mlx5dr_action_rewrite {
 	struct mlx5dr_domain *dmn;
 	struct mlx5dr_icm_chunk *chunk;
@@ -1566,5 +1576,13 @@ struct mlx5dr_ptrn_obj *mlx5dr_ptrn_cache_get_pattern(struct mlx5dr_ptrn_mgr *mg
 						      u16 num_of_actions, u8 *data);
 void mlx5dr_ptrn_cache_put_pattern(struct mlx5dr_ptrn_mgr *mgr,
 				   struct mlx5dr_ptrn_obj *pattern);
+struct mlx5dr_arg_mgr *mlx5dr_arg_mgr_create(struct mlx5dr_domain *dmn);
+void mlx5dr_arg_mgr_destroy(struct mlx5dr_arg_mgr *mgr);
+struct mlx5dr_arg_obj *mlx5dr_arg_get_obj(struct mlx5dr_arg_mgr *mgr,
+					  u16 num_of_actions,
+					  u8 *data);
+void mlx5dr_arg_put_obj(struct mlx5dr_arg_mgr *mgr,
+			struct mlx5dr_arg_obj *arg_obj);
+u32 mlx5dr_arg_get_obj_id(struct mlx5dr_arg_obj *arg_obj);
 
 #endif  /* _DR_TYPES_H_ */
-- 
cgit v1.2.3-70-g09d2


From 0caebadda57b7f1740406b054ddff36cedf0bf39 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Mon, 7 Nov 2022 02:49:22 +0200
Subject: net/mlx5: DR, Add modify header argument pointer to actions
 attributes

While building the actions, add the pointer of the arguments for
accelerated modify list action into the action's attributes.
This will be used later on while building the specific STE
for this action.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c        | 26 +++++++++++++++++-----
 .../mellanox/mlx5/core/steering/dr_ste_v1.c        | 19 +++++++++++++++-
 .../mellanox/mlx5/core/steering/dr_types.h         |  4 ++++
 3 files changed, 42 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 732a4002eab5..98e3d4f572eb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -819,14 +819,28 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 		case DR_ACTION_TYP_TNL_L2_TO_L2:
 			break;
 		case DR_ACTION_TYP_TNL_L3_TO_L2:
-			attr.decap_index = action->rewrite->index;
-			attr.decap_actions = action->rewrite->num_of_actions;
-			attr.decap_with_vlan =
-				attr.decap_actions == WITH_VLAN_NUM_HW_ACTIONS;
+			if (action->rewrite->ptrn && action->rewrite->arg) {
+				attr.decap_index = mlx5dr_arg_get_obj_id(action->rewrite->arg);
+				attr.decap_actions = action->rewrite->ptrn->num_of_actions;
+				attr.decap_pat_idx = action->rewrite->ptrn->index;
+			} else {
+				attr.decap_index = action->rewrite->index;
+				attr.decap_actions = action->rewrite->num_of_actions;
+				attr.decap_with_vlan =
+					attr.decap_actions == WITH_VLAN_NUM_HW_ACTIONS;
+				attr.decap_pat_idx = MLX5DR_INVALID_PATTERN_INDEX;
+			}
 			break;
 		case DR_ACTION_TYP_MODIFY_HDR:
-			attr.modify_index = action->rewrite->index;
-			attr.modify_actions = action->rewrite->num_of_actions;
+			if (action->rewrite->ptrn && action->rewrite->arg) {
+				attr.modify_index = mlx5dr_arg_get_obj_id(action->rewrite->arg);
+				attr.modify_actions = action->rewrite->ptrn->num_of_actions;
+				attr.modify_pat_idx = action->rewrite->ptrn->index;
+			} else {
+				attr.modify_index = action->rewrite->index;
+				attr.modify_actions = action->rewrite->num_of_actions;
+				attr.modify_pat_idx = MLX5DR_INVALID_PATTERN_INDEX;
+			}
 			if (action->rewrite->modify_ttl)
 				dr_action_modify_ttl_adjust(dmn, &attr, rx_rule,
 							    &recalc_cs_required);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
index cf8508139f55..3d04ac08be77 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
@@ -2179,27 +2179,44 @@ dr_ste_v1_build_tnl_gtpu_flex_parser_1_init(struct mlx5dr_ste_build *sb,
 int dr_ste_v1_alloc_modify_hdr_ptrn_arg(struct mlx5dr_action *action)
 {
 	struct mlx5dr_ptrn_mgr *ptrn_mgr;
+	int ret;
 
 	ptrn_mgr = action->rewrite->dmn->ptrn_mgr;
 	if (!ptrn_mgr)
 		return -EOPNOTSUPP;
 
+	action->rewrite->arg = mlx5dr_arg_get_obj(action->rewrite->dmn->arg_mgr,
+						  action->rewrite->num_of_actions,
+						  action->rewrite->data);
+	if (!action->rewrite->arg) {
+		mlx5dr_err(action->rewrite->dmn, "Failed allocating args for modify header\n");
+		return -EAGAIN;
+	}
+
 	action->rewrite->ptrn =
 		mlx5dr_ptrn_cache_get_pattern(ptrn_mgr,
 					      action->rewrite->num_of_actions,
 					      action->rewrite->data);
 	if (!action->rewrite->ptrn) {
 		mlx5dr_err(action->rewrite->dmn, "Failed to get pattern\n");
-		return -EAGAIN;
+		ret = -EAGAIN;
+		goto put_arg;
 	}
 
 	return 0;
+
+put_arg:
+	mlx5dr_arg_put_obj(action->rewrite->dmn->arg_mgr,
+			   action->rewrite->arg);
+	return ret;
 }
 
 void dr_ste_v1_free_modify_hdr_ptrn_arg(struct mlx5dr_action *action)
 {
 	mlx5dr_ptrn_cache_put_pattern(action->rewrite->dmn->ptrn_mgr,
 				      action->rewrite->ptrn);
+	mlx5dr_arg_put_obj(action->rewrite->dmn->arg_mgr,
+			   action->rewrite->arg);
 }
 
 static struct mlx5dr_ste_ctx ste_ctx_v1 = {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index e102ceb20e01..3ffda3d302e0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -261,11 +261,14 @@ u64 mlx5dr_ste_get_mr_addr(struct mlx5dr_ste *ste);
 struct list_head *mlx5dr_ste_get_miss_list(struct mlx5dr_ste *ste);
 
 #define MLX5DR_MAX_VLANS 2
+#define MLX5DR_INVALID_PATTERN_INDEX 0xffffffff
 
 struct mlx5dr_ste_actions_attr {
 	u32	modify_index;
+	u32	modify_pat_idx;
 	u16	modify_actions;
 	u32	decap_index;
+	u32	decap_pat_idx;
 	u16	decap_actions;
 	u8	decap_with_vlan:1;
 	u64	final_icm_addr;
@@ -1036,6 +1039,7 @@ struct mlx5dr_action_rewrite {
 	u8 allow_tx:1;
 	u8 modify_ttl:1;
 	struct mlx5dr_ptrn_obj *ptrn;
+	struct mlx5dr_arg_obj *arg;
 };
 
 struct mlx5dr_action_reformat {
-- 
cgit v1.2.3-70-g09d2


From 62e40c8568251c65026712a90a9a486dc5d06d56 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 8 Nov 2022 11:56:11 +0200
Subject: net/mlx5: DR, Apply new accelerated modify action and decapl3

If there is support for pattern/args, use the new accelerated modify
header action for modify header and decap L3 actions.
Otherwise fall back to the old modify-header implementation.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_ste_v1.c        | 51 +++++++++++++++++++---
 .../mlx5/core/steering/mlx5_ifc_dr_ste_v1.h        |  2 +-
 2 files changed, 47 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
index 3d04ac08be77..d2d312454564 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
@@ -495,21 +495,59 @@ static void dr_ste_v1_set_rx_decap(u8 *hw_ste_p, u8 *s_action)
 	dr_ste_v1_set_reparse(hw_ste_p);
 }
 
-static void dr_ste_v1_set_rewrite_actions(u8 *hw_ste_p,
-					  u8 *s_action,
-					  u16 num_of_actions,
-					  u32 re_write_index)
+static void dr_ste_v1_set_accelerated_rewrite_actions(u8 *hw_ste_p,
+						      u8 *d_action,
+						      u16 num_of_actions,
+						      u32 rewrite_pattern,
+						      u32 rewrite_args)
+{
+	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+		 action_id, DR_STE_V1_ACTION_ID_ACCELERATED_LIST);
+	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+		 modify_actions_pattern_pointer, rewrite_pattern);
+	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+		 number_of_modify_actions, num_of_actions);
+	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+		 modify_actions_argument_pointer, rewrite_args);
+
+	dr_ste_v1_set_reparse(hw_ste_p);
+}
+
+static void dr_ste_v1_set_basic_rewrite_actions(u8 *hw_ste_p,
+						u8 *s_action,
+						u16 num_of_actions,
+						u32 rewrite_index)
 {
 	MLX5_SET(ste_single_action_modify_list_v1, s_action, action_id,
 		 DR_STE_V1_ACTION_ID_MODIFY_LIST);
 	MLX5_SET(ste_single_action_modify_list_v1, s_action, num_of_modify_actions,
 		 num_of_actions);
 	MLX5_SET(ste_single_action_modify_list_v1, s_action, modify_actions_ptr,
-		 re_write_index);
+		 rewrite_index);
 
 	dr_ste_v1_set_reparse(hw_ste_p);
 }
 
+static void dr_ste_v1_set_rewrite_actions(u8 *hw_ste_p,
+					  u8 *action,
+					  u16 num_of_actions,
+					  u32 rewrite_pattern,
+					  u32 rewrite_args)
+{
+	if (rewrite_pattern != MLX5DR_INVALID_PATTERN_INDEX)
+		return dr_ste_v1_set_accelerated_rewrite_actions(hw_ste_p,
+								 action,
+								 num_of_actions,
+								 rewrite_pattern,
+								 rewrite_args);
+
+	/* fall back to the code that doesn't support accelerated modify header */
+	return dr_ste_v1_set_basic_rewrite_actions(hw_ste_p,
+						   action,
+						   num_of_actions,
+						   rewrite_args);
+}
+
 static void dr_ste_v1_set_aso_flow_meter(u8 *d_action,
 					 u32 object_id,
 					 u32 offset,
@@ -614,6 +652,7 @@ void dr_ste_v1_set_actions_tx(struct mlx5dr_domain *dmn,
 		}
 		dr_ste_v1_set_rewrite_actions(last_ste, action,
 					      attr->modify_actions,
+					      attr->modify_pat_idx,
 					      attr->modify_index);
 		action_sz -= DR_STE_ACTION_DOUBLE_SZ;
 		action += DR_STE_ACTION_DOUBLE_SZ;
@@ -744,6 +783,7 @@ void dr_ste_v1_set_actions_rx(struct mlx5dr_domain *dmn,
 	if (action_type_set[DR_ACTION_TYP_TNL_L3_TO_L2]) {
 		dr_ste_v1_set_rewrite_actions(last_ste, action,
 					      attr->decap_actions,
+					      attr->decap_pat_idx,
 					      attr->decap_index);
 		action_sz -= DR_STE_ACTION_DOUBLE_SZ;
 		action += DR_STE_ACTION_DOUBLE_SZ;
@@ -799,6 +839,7 @@ void dr_ste_v1_set_actions_rx(struct mlx5dr_domain *dmn,
 		}
 		dr_ste_v1_set_rewrite_actions(last_ste, action,
 					      attr->modify_actions,
+					      attr->modify_pat_idx,
 					      attr->modify_index);
 		action_sz -= DR_STE_ACTION_DOUBLE_SZ;
 		action += DR_STE_ACTION_DOUBLE_SZ;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h
index 790a17d6207f..ca3b0f1453a7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h
@@ -100,7 +100,7 @@ struct mlx5_ifc_ste_double_action_insert_with_ptr_v1_bits {
 	u8         pointer[0x20];
 };
 
-struct mlx5_ifc_ste_double_action_modify_action_list_v1_bits {
+struct mlx5_ifc_ste_double_action_accelerated_modify_action_list_v1_bits {
 	u8         action_id[0x8];
 	u8         modify_actions_pattern_pointer[0x18];
 
-- 
cgit v1.2.3-70-g09d2


From 947e258537ea43098d778b3315bde4fe1a1a10a3 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 15 Nov 2022 02:36:07 +0200
Subject: net/mlx5: DR, Support decap L3 action using pattern / arg mechanism

Use the new accelerated action for decap L3 on RX side:
use the mechanism of pattern and argument same as in
modify-header action.

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c         | 21 +++++----------------
 1 file changed, 5 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 98e3d4f572eb..8f8f0a0b38fd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -1428,23 +1428,12 @@ dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
 			return ret;
 		}
 
-		action->rewrite->chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool,
-								DR_CHUNK_SIZE_8);
-		if (!action->rewrite->chunk) {
-			mlx5dr_dbg(dmn, "Failed allocating modify header chunk\n");
-			return -ENOMEM;
-		}
-
-		action->rewrite->data = (void *)hw_actions;
-		action->rewrite->index = (mlx5dr_icm_pool_get_chunk_icm_addr
-					  (action->rewrite->chunk) -
-					 dmn->info.caps.hdr_modify_icm_addr) /
-					 DR_ACTION_CACHE_LINE_SIZE;
+		action->rewrite->data = hw_actions;
+		action->rewrite->dmn = dmn;
 
-		ret = mlx5dr_send_postsend_action(dmn, action);
+		ret = mlx5dr_ste_alloc_modify_hdr(action);
 		if (ret) {
-			mlx5dr_dbg(dmn, "Writing decap l3 actions to ICM failed\n");
-			mlx5dr_icm_free_chunk(action->rewrite->chunk);
+			mlx5dr_dbg(dmn, "Failed preparing reformat data\n");
 			return ret;
 		}
 		return 0;
@@ -2161,7 +2150,7 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 		refcount_dec(&action->reformat->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_TNL_L3_TO_L2:
-		mlx5dr_icm_free_chunk(action->rewrite->chunk);
+		mlx5dr_ste_free_modify_hdr(action);
 		refcount_dec(&action->rewrite->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_L2_TO_TNL_L2:
-- 
cgit v1.2.3-70-g09d2


From 40ff097f2503869fe5d96f796cf84ad4b17eaa9b Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 15 Nov 2022 18:19:39 +0200
Subject: net/mlx5: DR, Modify header action of size 1 optimization

Set modify header action of size 1 directly on the STE for supporting
devices, thus reducing number of hops and cache misses.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c        | 37 ++++++++++++++-------
 .../mellanox/mlx5/core/steering/dr_ste_v1.c        | 38 ++++++++++++++--------
 .../mellanox/mlx5/core/steering/dr_types.h         |  2 ++
 3 files changed, 52 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 8f8f0a0b38fd..0eb9a8d7f282 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -832,14 +832,20 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 			}
 			break;
 		case DR_ACTION_TYP_MODIFY_HDR:
-			if (action->rewrite->ptrn && action->rewrite->arg) {
-				attr.modify_index = mlx5dr_arg_get_obj_id(action->rewrite->arg);
-				attr.modify_actions = action->rewrite->ptrn->num_of_actions;
-				attr.modify_pat_idx = action->rewrite->ptrn->index;
-			} else {
-				attr.modify_index = action->rewrite->index;
+			if (action->rewrite->single_action_opt) {
 				attr.modify_actions = action->rewrite->num_of_actions;
-				attr.modify_pat_idx = MLX5DR_INVALID_PATTERN_INDEX;
+				attr.single_modify_action = action->rewrite->data;
+			} else {
+				if (action->rewrite->ptrn && action->rewrite->arg) {
+					attr.modify_index =
+						mlx5dr_arg_get_obj_id(action->rewrite->arg);
+					attr.modify_actions = action->rewrite->ptrn->num_of_actions;
+					attr.modify_pat_idx = action->rewrite->ptrn->index;
+				} else {
+					attr.modify_index = action->rewrite->index;
+					attr.modify_actions = action->rewrite->num_of_actions;
+					attr.modify_pat_idx = MLX5DR_INVALID_PATTERN_INDEX;
+				}
 			}
 			if (action->rewrite->modify_ttl)
 				dr_action_modify_ttl_adjust(dmn, &attr, rx_rule,
@@ -1998,9 +2004,15 @@ static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
 	action->rewrite->data = (u8 *)hw_actions;
 	action->rewrite->num_of_actions = num_hw_actions;
 
-	ret = mlx5dr_ste_alloc_modify_hdr(action);
-	if (ret)
-		goto free_hw_actions;
+	if (num_hw_actions == 1 &&
+	    dmn->info.caps.sw_format_ver >= MLX5_STEERING_FORMAT_CONNECTX_6DX) {
+		action->rewrite->single_action_opt = true;
+	} else {
+		action->rewrite->single_action_opt = false;
+		ret = mlx5dr_ste_alloc_modify_hdr(action);
+		if (ret)
+			goto free_hw_actions;
+	}
 
 	return 0;
 
@@ -2151,6 +2163,7 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 		break;
 	case DR_ACTION_TYP_TNL_L3_TO_L2:
 		mlx5dr_ste_free_modify_hdr(action);
+		kfree(action->rewrite->data);
 		refcount_dec(&action->rewrite->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_L2_TO_TNL_L2:
@@ -2161,7 +2174,9 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action)
 		refcount_dec(&action->reformat->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_MODIFY_HDR:
-		mlx5dr_ste_free_modify_hdr(action);
+		if (!action->rewrite->single_action_opt)
+			mlx5dr_ste_free_modify_hdr(action);
+		kfree(action->rewrite->data);
 		refcount_dec(&action->rewrite->dmn->refcount);
 		break;
 	case DR_ACTION_TYP_SAMPLER:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
index d2d312454564..4c0704ad166b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c
@@ -499,16 +499,21 @@ static void dr_ste_v1_set_accelerated_rewrite_actions(u8 *hw_ste_p,
 						      u8 *d_action,
 						      u16 num_of_actions,
 						      u32 rewrite_pattern,
-						      u32 rewrite_args)
+						      u32 rewrite_args,
+						      u8 *action_data)
 {
-	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
-		 action_id, DR_STE_V1_ACTION_ID_ACCELERATED_LIST);
-	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
-		 modify_actions_pattern_pointer, rewrite_pattern);
-	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
-		 number_of_modify_actions, num_of_actions);
-	MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
-		 modify_actions_argument_pointer, rewrite_args);
+	if (action_data) {
+		memcpy(d_action, action_data, DR_MODIFY_ACTION_SIZE);
+	} else {
+		MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+			 action_id, DR_STE_V1_ACTION_ID_ACCELERATED_LIST);
+		MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+			 modify_actions_pattern_pointer, rewrite_pattern);
+		MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+			 number_of_modify_actions, num_of_actions);
+		MLX5_SET(ste_double_action_accelerated_modify_action_list_v1, d_action,
+			 modify_actions_argument_pointer, rewrite_args);
+	}
 
 	dr_ste_v1_set_reparse(hw_ste_p);
 }
@@ -532,14 +537,16 @@ static void dr_ste_v1_set_rewrite_actions(u8 *hw_ste_p,
 					  u8 *action,
 					  u16 num_of_actions,
 					  u32 rewrite_pattern,
-					  u32 rewrite_args)
+					  u32 rewrite_args,
+					  u8 *action_data)
 {
 	if (rewrite_pattern != MLX5DR_INVALID_PATTERN_INDEX)
 		return dr_ste_v1_set_accelerated_rewrite_actions(hw_ste_p,
 								 action,
 								 num_of_actions,
 								 rewrite_pattern,
-								 rewrite_args);
+								 rewrite_args,
+								 action_data);
 
 	/* fall back to the code that doesn't support accelerated modify header */
 	return dr_ste_v1_set_basic_rewrite_actions(hw_ste_p,
@@ -653,7 +660,8 @@ void dr_ste_v1_set_actions_tx(struct mlx5dr_domain *dmn,
 		dr_ste_v1_set_rewrite_actions(last_ste, action,
 					      attr->modify_actions,
 					      attr->modify_pat_idx,
-					      attr->modify_index);
+					      attr->modify_index,
+					      attr->single_modify_action);
 		action_sz -= DR_STE_ACTION_DOUBLE_SZ;
 		action += DR_STE_ACTION_DOUBLE_SZ;
 		allow_encap = false;
@@ -784,7 +792,8 @@ void dr_ste_v1_set_actions_rx(struct mlx5dr_domain *dmn,
 		dr_ste_v1_set_rewrite_actions(last_ste, action,
 					      attr->decap_actions,
 					      attr->decap_pat_idx,
-					      attr->decap_index);
+					      attr->decap_index,
+					      NULL);
 		action_sz -= DR_STE_ACTION_DOUBLE_SZ;
 		action += DR_STE_ACTION_DOUBLE_SZ;
 		allow_modify_hdr = false;
@@ -840,7 +849,8 @@ void dr_ste_v1_set_actions_rx(struct mlx5dr_domain *dmn,
 		dr_ste_v1_set_rewrite_actions(last_ste, action,
 					      attr->modify_actions,
 					      attr->modify_pat_idx,
-					      attr->modify_index);
+					      attr->modify_index,
+					      attr->single_modify_action);
 		action_sz -= DR_STE_ACTION_DOUBLE_SZ;
 		action += DR_STE_ACTION_DOUBLE_SZ;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 3ffda3d302e0..37b7b1a79f93 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -267,6 +267,7 @@ struct mlx5dr_ste_actions_attr {
 	u32	modify_index;
 	u32	modify_pat_idx;
 	u16	modify_actions;
+	u8	*single_modify_action;
 	u32	decap_index;
 	u32	decap_pat_idx;
 	u16	decap_actions;
@@ -1035,6 +1036,7 @@ struct mlx5dr_action_rewrite {
 	u8 *data;
 	u16 num_of_actions;
 	u32 index;
+	u8 single_action_opt:1;
 	u8 allow_rx:1;
 	u8 allow_tx:1;
 	u8 modify_ttl:1;
-- 
cgit v1.2.3-70-g09d2


From a21e52bb8f372b98855d931ca87e43e3c958c04f Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Sun, 1 Jan 2023 00:20:53 +0200
Subject: net/mlx5: DR, Add support for the pattern/arg parameters in debug
 dump

Support the pattern/args-based MODIFY_HDR and TNL_L3_TO_L2 actions in dbg dump

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/steering/dr_dbg.c  | 30 +++++++++++++++++++---
 1 file changed, 27 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
index db81d881d38e..1ff8bde90e1e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
@@ -140,10 +140,31 @@ dr_dump_rule_action_mem(struct seq_file *file, const u64 rule_id,
 			   action->flow_tag->flow_tag);
 		break;
 	case DR_ACTION_TYP_MODIFY_HDR:
-		seq_printf(file, "%d,0x%llx,0x%llx,0x%x\n",
+	{
+		struct mlx5dr_ptrn_obj *ptrn = action->rewrite->ptrn;
+		struct mlx5dr_arg_obj *arg = action->rewrite->arg;
+		u8 *rewrite_data = action->rewrite->data;
+		bool ptrn_arg;
+		int i;
+
+		ptrn_arg = !action->rewrite->single_action_opt && ptrn && arg;
+
+		seq_printf(file, "%d,0x%llx,0x%llx,0x%x,%d,0x%x,0x%x,0x%x",
 			   DR_DUMP_REC_TYPE_ACTION_MODIFY_HDR, action_id,
-			   rule_id, action->rewrite->index);
+			   rule_id, action->rewrite->index,
+			   action->rewrite->single_action_opt,
+			   action->rewrite->num_of_actions,
+			   ptrn_arg ? ptrn->index : 0,
+			   ptrn_arg ? mlx5dr_arg_get_obj_id(arg) : 0);
+
+		for (i = 0; i < action->rewrite->num_of_actions; i++) {
+			seq_printf(file, ",0x%016llx",
+				   be64_to_cpu(((__be64 *)rewrite_data)[i]));
+		}
+
+		seq_puts(file, "\n");
 		break;
+	}
 	case DR_ACTION_TYP_VPORT:
 		seq_printf(file, "%d,0x%llx,0x%llx,0x%x\n",
 			   DR_DUMP_REC_TYPE_ACTION_VPORT, action_id, rule_id,
@@ -157,7 +178,10 @@ dr_dump_rule_action_mem(struct seq_file *file, const u64 rule_id,
 	case DR_ACTION_TYP_TNL_L3_TO_L2:
 		seq_printf(file, "%d,0x%llx,0x%llx,0x%x\n",
 			   DR_DUMP_REC_TYPE_ACTION_DECAP_L3, action_id,
-			   rule_id, action->rewrite->index);
+			   rule_id,
+			   (action->rewrite->ptrn && action->rewrite->arg) ?
+			   mlx5dr_arg_get_obj_id(action->rewrite->arg) :
+			   action->rewrite->index);
 		break;
 	case DR_ACTION_TYP_L2_TO_TNL_L2:
 		seq_printf(file, "%d,0x%llx,0x%llx,0x%x\n",
-- 
cgit v1.2.3-70-g09d2


From 220ae987838c893fe11e46a3e3994a549f203daa Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 30 Aug 2022 01:18:59 +0300
Subject: net/mlx5: DR, Enable patterns and arguments for supporting devices

Check if patterns and arguments for modify header action
are supported and enable them accordingly.

Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
index c4545daf179b..9a2dfe6ebe31 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c
@@ -12,7 +12,8 @@
 
 bool mlx5dr_domain_is_support_ptrn_arg(struct mlx5dr_domain *dmn)
 {
-	return false;
+	return dmn->info.caps.sw_format_ver >= MLX5_STEERING_FORMAT_CONNECTX_6DX &&
+	       dmn->info.caps.support_modify_argument;
 }
 
 static int dr_domain_init_modify_header_resources(struct mlx5dr_domain *dmn)
-- 
cgit v1.2.3-70-g09d2


From b07a2d97ba5ef154fe736aa510e43a3299eee5f8 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 12 Apr 2023 21:26:03 -0700
Subject: net: skb: plumb napi state thru skb freeing paths

We maintain a NAPI-local cache of skbs which is fed by napi_consume_skb().
Going forward we will also try to cache head and data pages.
Plumb the "are we in a normal NAPI context" information thru
deeper into the freeing path, up to skb_release_data() and
skb_free_head()/skb_pp_recycle(). The "not normal NAPI context"
comes from netpoll which passes budget of 0 to try to reap
the Tx completions but not perform any Rx.

Use "bool napi_safe" rather than bare "int budget",
the further we get from NAPI the more confusing the budget
argument may seem (particularly whether 0 or MAX is the
correct value to pass in when not in NAPI).

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Tested-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/skbuff.c | 38 ++++++++++++++++++++------------------
 1 file changed, 20 insertions(+), 18 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 78238a13dbcf..2b5a98c5cb49 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -839,7 +839,7 @@ static void skb_clone_fraglist(struct sk_buff *skb)
 		skb_get(list);
 }
 
-static bool skb_pp_recycle(struct sk_buff *skb, void *data)
+static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe)
 {
 	if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)
 		return false;
@@ -856,12 +856,12 @@ static void skb_kfree_head(void *head, unsigned int end_offset)
 		kfree(head);
 }
 
-static void skb_free_head(struct sk_buff *skb)
+static void skb_free_head(struct sk_buff *skb, bool napi_safe)
 {
 	unsigned char *head = skb->head;
 
 	if (skb->head_frag) {
-		if (skb_pp_recycle(skb, head))
+		if (skb_pp_recycle(skb, head, napi_safe))
 			return;
 		skb_free_frag(head);
 	} else {
@@ -869,7 +869,8 @@ static void skb_free_head(struct sk_buff *skb)
 	}
 }
 
-static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason)
+static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason,
+			     bool napi_safe)
 {
 	struct skb_shared_info *shinfo = skb_shinfo(skb);
 	int i;
@@ -894,7 +895,7 @@ free_head:
 	if (shinfo->frag_list)
 		kfree_skb_list_reason(shinfo->frag_list, reason);
 
-	skb_free_head(skb);
+	skb_free_head(skb, napi_safe);
 exit:
 	/* When we clone an SKB we copy the reycling bit. The pp_recycle
 	 * bit is only set on the head though, so in order to avoid races
@@ -955,11 +956,12 @@ void skb_release_head_state(struct sk_buff *skb)
 }
 
 /* Free everything but the sk_buff shell. */
-static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason)
+static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason,
+			    bool napi_safe)
 {
 	skb_release_head_state(skb);
 	if (likely(skb->head))
-		skb_release_data(skb, reason);
+		skb_release_data(skb, reason, napi_safe);
 }
 
 /**
@@ -973,7 +975,7 @@ static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason)
 
 void __kfree_skb(struct sk_buff *skb)
 {
-	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED);
+	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, false);
 	kfree_skbmem(skb);
 }
 EXPORT_SYMBOL(__kfree_skb);
@@ -1027,7 +1029,7 @@ static void kfree_skb_add_bulk(struct sk_buff *skb,
 		return;
 	}
 
-	skb_release_all(skb, reason);
+	skb_release_all(skb, reason, false);
 	sa->skb_array[sa->skb_count++] = skb;
 
 	if (unlikely(sa->skb_count == KFREE_SKB_BULK_SIZE)) {
@@ -1201,7 +1203,7 @@ EXPORT_SYMBOL(consume_skb);
 void __consume_stateless_skb(struct sk_buff *skb)
 {
 	trace_consume_skb(skb, __builtin_return_address(0));
-	skb_release_data(skb, SKB_CONSUMED);
+	skb_release_data(skb, SKB_CONSUMED, false);
 	kfree_skbmem(skb);
 }
 
@@ -1226,7 +1228,7 @@ static void napi_skb_cache_put(struct sk_buff *skb)
 
 void __kfree_skb_defer(struct sk_buff *skb)
 {
-	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED);
+	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, true);
 	napi_skb_cache_put(skb);
 }
 
@@ -1264,7 +1266,7 @@ void napi_consume_skb(struct sk_buff *skb, int budget)
 		return;
 	}
 
-	skb_release_all(skb, SKB_CONSUMED);
+	skb_release_all(skb, SKB_CONSUMED, !!budget);
 	napi_skb_cache_put(skb);
 }
 EXPORT_SYMBOL(napi_consume_skb);
@@ -1395,7 +1397,7 @@ EXPORT_SYMBOL_GPL(alloc_skb_for_msg);
  */
 struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src)
 {
-	skb_release_all(dst, SKB_CONSUMED);
+	skb_release_all(dst, SKB_CONSUMED, false);
 	return __skb_clone(dst, src);
 }
 EXPORT_SYMBOL_GPL(skb_morph);
@@ -2018,9 +2020,9 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
 		if (skb_has_frag_list(skb))
 			skb_clone_fraglist(skb);
 
-		skb_release_data(skb, SKB_CONSUMED);
+		skb_release_data(skb, SKB_CONSUMED, false);
 	} else {
-		skb_free_head(skb);
+		skb_free_head(skb, false);
 	}
 	off = (data + nhead) - skb->head;
 
@@ -6389,12 +6391,12 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off,
 			skb_frag_ref(skb, i);
 		if (skb_has_frag_list(skb))
 			skb_clone_fraglist(skb);
-		skb_release_data(skb, SKB_CONSUMED);
+		skb_release_data(skb, SKB_CONSUMED, false);
 	} else {
 		/* we can reuse existing recount- all we did was
 		 * relocate values
 		 */
-		skb_free_head(skb);
+		skb_free_head(skb, false);
 	}
 
 	skb->head = data;
@@ -6529,7 +6531,7 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off,
 		skb_kfree_head(data, size);
 		return -ENOMEM;
 	}
-	skb_release_data(skb, SKB_CONSUMED);
+	skb_release_data(skb, SKB_CONSUMED, false);
 
 	skb->head = data;
 	skb->head_frag = 0;
-- 
cgit v1.2.3-70-g09d2


From 8c48eea3adf3119e0a3fc57bd31f6966f26ee784 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 12 Apr 2023 21:26:04 -0700
Subject: page_pool: allow caching from safely localized NAPI

Recent patches to mlx5 mentioned a regression when moving from
driver local page pool to only using the generic page pool code.
Page pool has two recycling paths (1) direct one, which runs in
safe NAPI context (basically consumer context, so producing
can be lockless); and (2) via a ptr_ring, which takes a spin
lock because the freeing can happen from any CPU; producer
and consumer may run concurrently.

Since the page pool code was added, Eric introduced a revised version
of deferred skb freeing. TCP skbs are now usually returned to the CPU
which allocated them, and freed in softirq context. This places the
freeing (producing of pages back to the pool) enticingly close to
the allocation (consumer).

If we can prove that we're freeing in the same softirq context in which
the consumer NAPI will run - lockless use of the cache is perfectly fine,
no need for the lock.

Let drivers link the page pool to a NAPI instance. If the NAPI instance
is scheduled on the same CPU on which we're freeing - place the pages
in the direct cache.

With that and patched bnxt (XDP enabled to engage the page pool, sigh,
bnxt really needs page pool work :() I see a 2.6% perf boost with
a TCP stream test (app on a different physical core than softirq).

The CPU use of relevant functions decreases as expected:

  page_pool_refill_alloc_cache   1.17% -> 0%
  _raw_spin_lock                 2.41% -> 0.98%

Only consider lockless path to be safe when NAPI is scheduled
- in practice this should cover majority if not all of steady state
workloads. It's usually the NAPI kicking in that causes the skb flush.

The main case we'll miss out on is when application runs on the same
CPU as NAPI. In that case we don't use the deferred skb free path.

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/page_pool.rst |  1 +
 include/linux/netdevice.h              |  3 +++
 include/linux/skbuff.h                 | 20 +++++++++++++-------
 include/net/page_pool.h                |  3 ++-
 net/core/dev.c                         |  3 +++
 net/core/page_pool.c                   | 15 +++++++++++++--
 net/core/skbuff.c                      |  4 ++--
 7 files changed, 37 insertions(+), 12 deletions(-)

diff --git a/Documentation/networking/page_pool.rst b/Documentation/networking/page_pool.rst
index 30f1344e7cca..873efd97f822 100644
--- a/Documentation/networking/page_pool.rst
+++ b/Documentation/networking/page_pool.rst
@@ -165,6 +165,7 @@ Registration
     pp_params.pool_size = DESC_NUM;
     pp_params.nid = NUMA_NO_NODE;
     pp_params.dev = priv->dev;
+    pp_params.napi = napi; /* only if locking is tied to NAPI */
     pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
     page_pool = page_pool_create(&pp_params);
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 96d27d558b0c..203c0df2046c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -360,8 +360,11 @@ struct napi_struct {
 	unsigned long		gro_bitmask;
 	int			(*poll)(struct napi_struct *, int);
 #ifdef CONFIG_NETPOLL
+	/* CPU actively polling if netpoll is configured */
 	int			poll_owner;
 #endif
+	/* CPU on which NAPI has been scheduled for processing */
+	int			list_owner;
 	struct net_device	*dev;
 	struct gro_list		gro_hash[GRO_HASH_BUCKETS];
 	struct sk_buff		*skb;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 494a23a976b0..a823ec3aa326 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3386,6 +3386,18 @@ static inline void skb_frag_ref(struct sk_buff *skb, int f)
 	__skb_frag_ref(&skb_shinfo(skb)->frags[f]);
 }
 
+static inline void
+napi_frag_unref(skb_frag_t *frag, bool recycle, bool napi_safe)
+{
+	struct page *page = skb_frag_page(frag);
+
+#ifdef CONFIG_PAGE_POOL
+	if (recycle && page_pool_return_skb_page(page, napi_safe))
+		return;
+#endif
+	put_page(page);
+}
+
 /**
  * __skb_frag_unref - release a reference on a paged fragment.
  * @frag: the paged fragment
@@ -3396,13 +3408,7 @@ static inline void skb_frag_ref(struct sk_buff *skb, int f)
  */
 static inline void __skb_frag_unref(skb_frag_t *frag, bool recycle)
 {
-	struct page *page = skb_frag_page(frag);
-
-#ifdef CONFIG_PAGE_POOL
-	if (recycle && page_pool_return_skb_page(page))
-		return;
-#endif
-	put_page(page);
+	napi_frag_unref(frag, recycle, false);
 }
 
 /**
diff --git a/include/net/page_pool.h b/include/net/page_pool.h
index ddfa0b328677..91b808dade82 100644
--- a/include/net/page_pool.h
+++ b/include/net/page_pool.h
@@ -77,6 +77,7 @@ struct page_pool_params {
 	unsigned int	pool_size;
 	int		nid;  /* Numa node id to allocate from pages from */
 	struct device	*dev; /* device, for DMA pre-mapping purposes */
+	struct napi_struct *napi; /* Sole consumer of pages, otherwise NULL */
 	enum dma_data_direction dma_dir; /* DMA mapping direction */
 	unsigned int	max_len; /* max DMA sync memory size */
 	unsigned int	offset;  /* DMA addr offset */
@@ -239,7 +240,7 @@ inline enum dma_data_direction page_pool_get_dma_dir(struct page_pool *pool)
 	return pool->p.dma_dir;
 }
 
-bool page_pool_return_skb_page(struct page *page);
+bool page_pool_return_skb_page(struct page *page, bool napi_safe);
 
 struct page_pool *page_pool_create(const struct page_pool_params *params);
 
diff --git a/net/core/dev.c b/net/core/dev.c
index c7f13742b56c..8aea68275172 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4359,6 +4359,7 @@ static inline void ____napi_schedule(struct softnet_data *sd,
 	}
 
 	list_add_tail(&napi->poll_list, &sd->poll_list);
+	WRITE_ONCE(napi->list_owner, smp_processor_id());
 	/* If not called from net_rx_action()
 	 * we have to raise NET_RX_SOFTIRQ.
 	 */
@@ -6069,6 +6070,7 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
 		list_del_init(&n->poll_list);
 		local_irq_restore(flags);
 	}
+	WRITE_ONCE(n->list_owner, -1);
 
 	val = READ_ONCE(n->state);
 	do {
@@ -6384,6 +6386,7 @@ void netif_napi_add_weight(struct net_device *dev, struct napi_struct *napi,
 #ifdef CONFIG_NETPOLL
 	napi->poll_owner = -1;
 #endif
+	napi->list_owner = -1;
 	set_bit(NAPI_STATE_SCHED, &napi->state);
 	set_bit(NAPI_STATE_NPSVC, &napi->state);
 	list_add_rcu(&napi->dev_list, &dev->napi_list);
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 193c18799865..2f6bf422ed30 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -19,6 +19,7 @@
 #include <linux/mm.h> /* for put_page() */
 #include <linux/poison.h>
 #include <linux/ethtool.h>
+#include <linux/netdevice.h>
 
 #include <trace/events/page_pool.h>
 
@@ -874,9 +875,11 @@ void page_pool_update_nid(struct page_pool *pool, int new_nid)
 }
 EXPORT_SYMBOL(page_pool_update_nid);
 
-bool page_pool_return_skb_page(struct page *page)
+bool page_pool_return_skb_page(struct page *page, bool napi_safe)
 {
+	struct napi_struct *napi;
 	struct page_pool *pp;
+	bool allow_direct;
 
 	page = compound_head(page);
 
@@ -892,12 +895,20 @@ bool page_pool_return_skb_page(struct page *page)
 
 	pp = page->pp;
 
+	/* Allow direct recycle if we have reasons to believe that we are
+	 * in the same context as the consumer would run, so there's
+	 * no possible race.
+	 */
+	napi = pp->p.napi;
+	allow_direct = napi_safe && napi &&
+		READ_ONCE(napi->list_owner) == smp_processor_id();
+
 	/* Driver set this to memory recycling info. Reset it on recycle.
 	 * This will *not* work for NIC using a split-page memory model.
 	 * The page will be returned to the pool here regardless of the
 	 * 'flipped' fragment being in use or not.
 	 */
-	page_pool_put_full_page(pp, page, false);
+	page_pool_put_full_page(pp, page, allow_direct);
 
 	return true;
 }
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 2b5a98c5cb49..ef81452759be 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -843,7 +843,7 @@ static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe)
 {
 	if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)
 		return false;
-	return page_pool_return_skb_page(virt_to_page(data));
+	return page_pool_return_skb_page(virt_to_page(data), napi_safe);
 }
 
 static void skb_kfree_head(void *head, unsigned int end_offset)
@@ -889,7 +889,7 @@ static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason,
 	}
 
 	for (i = 0; i < shinfo->nr_frags; i++)
-		__skb_frag_unref(&shinfo->frags[i], skb->pp_recycle);
+		napi_frag_unref(&shinfo->frags[i], skb->pp_recycle, napi_safe);
 
 free_head:
 	if (shinfo->frag_list)
-- 
cgit v1.2.3-70-g09d2


From 294e39e0d03449f32b0723d3fd99ab5c23881c05 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 12 Apr 2023 21:26:05 -0700
Subject: bnxt: hook NAPIs to page pools

bnxt has 1:1 mapping of page pools and NAPIs, so it's safe
to hoook them up together.

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Tested-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 92289ab2f34a..8fb9d1bbe56f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3211,6 +3211,7 @@ static int bnxt_alloc_rx_page_pool(struct bnxt *bp,
 
 	pp.pool_size = bp->rx_ring_size;
 	pp.nid = dev_to_node(&bp->pdev->dev);
+	pp.napi = &rxr->bnapi->napi;
 	pp.dev = &bp->pdev->dev;
 	pp.dma_dir = DMA_BIDIRECTIONAL;
 
-- 
cgit v1.2.3-70-g09d2


From cd2a8079014aced27da9b2e669784f31680f1351 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:03 -0700
Subject: bpf: Remove btf_field_offs, use btf_record's fields instead

The btf_field_offs struct contains (offset, size) for btf_record fields,
sorted by offset. btf_field_offs is always used in conjunction with
btf_record, which has btf_field 'fields' array with (offset, type), the
latter of which btf_field_offs' size is derived from via
btf_field_type_size.

This patch adds a size field to struct btf_field and sorts btf_record's
fields by offset, making it possible to get rid of btf_field_offs. Less
data duplication and less code complexity results.

Since btf_field_offs' lifetime closely followed the btf_record used to
populate it, most complexity wins are from removal of initialization
code like:

  if (btf_record_successfully_initialized) {
    foffs = btf_parse_field_offs(rec);
    if (IS_ERR_OR_NULL(foffs))
      // free the btf_record and return err
  }

Other changes in this patch are pretty mechanical:

  * foffs->field_off[i] -> rec->fields[i].offset
  * foffs->field_sz[i] -> rec->fields[i].size
  * Sort rec->fields in btf_parse_fields before returning
    * It's possible that this is necessary independently of other
      changes in this patch. btf_record_find in syscall.c expects
      btf_record's fields to be sorted by offset, yet there's no
      explicit sorting of them before this patch, record's fields are
      populated in the order they're read from BTF struct definition.
      BTF docs don't say anything about the sortedness of struct fields.
  * All functions taking struct btf_field_offs * input now instead take
    struct btf_record *. All callsites of these functions already have
    access to the correct btf_record.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-2-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h     | 44 ++++++++++-------------
 include/linux/btf.h     |  2 --
 kernel/bpf/btf.c        | 93 +++++++++++--------------------------------------
 kernel/bpf/helpers.c    |  2 +-
 kernel/bpf/map_in_map.c | 15 --------
 kernel/bpf/syscall.c    | 17 ++-------
 6 files changed, 43 insertions(+), 130 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 88845aadc47d..7888ed497432 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -210,6 +210,7 @@ struct btf_field_graph_root {
 
 struct btf_field {
 	u32 offset;
+	u32 size;
 	enum btf_field_type type;
 	union {
 		struct btf_field_kptr kptr;
@@ -225,12 +226,6 @@ struct btf_record {
 	struct btf_field fields[];
 };
 
-struct btf_field_offs {
-	u32 cnt;
-	u32 field_off[BTF_FIELDS_MAX];
-	u8 field_sz[BTF_FIELDS_MAX];
-};
-
 struct bpf_map {
 	/* The first two cachelines with read-mostly members of which some
 	 * are also accessed in fast-path (e.g. ops, max_entries).
@@ -257,7 +252,6 @@ struct bpf_map {
 	struct obj_cgroup *objcg;
 #endif
 	char name[BPF_OBJ_NAME_LEN];
-	struct btf_field_offs *field_offs;
 	/* The 3rd and 4th cacheline with misc members to avoid false sharing
 	 * particularly with refcounting.
 	 */
@@ -360,14 +354,14 @@ static inline bool btf_record_has_field(const struct btf_record *rec, enum btf_f
 	return rec->field_mask & type;
 }
 
-static inline void bpf_obj_init(const struct btf_field_offs *foffs, void *obj)
+static inline void bpf_obj_init(const struct btf_record *rec, void *obj)
 {
 	int i;
 
-	if (!foffs)
+	if (IS_ERR_OR_NULL(rec))
 		return;
-	for (i = 0; i < foffs->cnt; i++)
-		memset(obj + foffs->field_off[i], 0, foffs->field_sz[i]);
+	for (i = 0; i < rec->cnt; i++)
+		memset(obj + rec->fields[i].offset, 0, rec->fields[i].size);
 }
 
 /* 'dst' must be a temporary buffer and should not point to memory that is being
@@ -379,7 +373,7 @@ static inline void bpf_obj_init(const struct btf_field_offs *foffs, void *obj)
  */
 static inline void check_and_init_map_value(struct bpf_map *map, void *dst)
 {
-	bpf_obj_init(map->field_offs, dst);
+	bpf_obj_init(map->record, dst);
 }
 
 /* memcpy that is used with 8-byte aligned pointers, power-of-8 size and
@@ -399,14 +393,14 @@ static inline void bpf_long_memcpy(void *dst, const void *src, u32 size)
 }
 
 /* copy everything but bpf_spin_lock, bpf_timer, and kptrs. There could be one of each. */
-static inline void bpf_obj_memcpy(struct btf_field_offs *foffs,
+static inline void bpf_obj_memcpy(struct btf_record *rec,
 				  void *dst, void *src, u32 size,
 				  bool long_memcpy)
 {
 	u32 curr_off = 0;
 	int i;
 
-	if (likely(!foffs)) {
+	if (IS_ERR_OR_NULL(rec)) {
 		if (long_memcpy)
 			bpf_long_memcpy(dst, src, round_up(size, 8));
 		else
@@ -414,49 +408,49 @@ static inline void bpf_obj_memcpy(struct btf_field_offs *foffs,
 		return;
 	}
 
-	for (i = 0; i < foffs->cnt; i++) {
-		u32 next_off = foffs->field_off[i];
+	for (i = 0; i < rec->cnt; i++) {
+		u32 next_off = rec->fields[i].offset;
 		u32 sz = next_off - curr_off;
 
 		memcpy(dst + curr_off, src + curr_off, sz);
-		curr_off += foffs->field_sz[i] + sz;
+		curr_off += rec->fields[i].size + sz;
 	}
 	memcpy(dst + curr_off, src + curr_off, size - curr_off);
 }
 
 static inline void copy_map_value(struct bpf_map *map, void *dst, void *src)
 {
-	bpf_obj_memcpy(map->field_offs, dst, src, map->value_size, false);
+	bpf_obj_memcpy(map->record, dst, src, map->value_size, false);
 }
 
 static inline void copy_map_value_long(struct bpf_map *map, void *dst, void *src)
 {
-	bpf_obj_memcpy(map->field_offs, dst, src, map->value_size, true);
+	bpf_obj_memcpy(map->record, dst, src, map->value_size, true);
 }
 
-static inline void bpf_obj_memzero(struct btf_field_offs *foffs, void *dst, u32 size)
+static inline void bpf_obj_memzero(struct btf_record *rec, void *dst, u32 size)
 {
 	u32 curr_off = 0;
 	int i;
 
-	if (likely(!foffs)) {
+	if (IS_ERR_OR_NULL(rec)) {
 		memset(dst, 0, size);
 		return;
 	}
 
-	for (i = 0; i < foffs->cnt; i++) {
-		u32 next_off = foffs->field_off[i];
+	for (i = 0; i < rec->cnt; i++) {
+		u32 next_off = rec->fields[i].offset;
 		u32 sz = next_off - curr_off;
 
 		memset(dst + curr_off, 0, sz);
-		curr_off += foffs->field_sz[i] + sz;
+		curr_off += rec->fields[i].size + sz;
 	}
 	memset(dst + curr_off, 0, size - curr_off);
 }
 
 static inline void zero_map_value(struct bpf_map *map, void *dst)
 {
-	bpf_obj_memzero(map->field_offs, dst, map->value_size);
+	bpf_obj_memzero(map->record, dst, map->value_size);
 }
 
 void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 495250162422..813227bff58a 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -113,7 +113,6 @@ struct btf_id_dtor_kfunc {
 struct btf_struct_meta {
 	u32 btf_id;
 	struct btf_record *record;
-	struct btf_field_offs *field_offs;
 };
 
 struct btf_struct_metas {
@@ -207,7 +206,6 @@ int btf_find_timer(const struct btf *btf, const struct btf_type *t);
 struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type *t,
 				    u32 field_mask, u32 value_size);
 int btf_check_and_fixup_fields(const struct btf *btf, struct btf_record *rec);
-struct btf_field_offs *btf_parse_field_offs(struct btf_record *rec);
 bool btf_type_is_void(const struct btf_type *t);
 s32 btf_find_by_name_kind(const struct btf *btf, const char *name, u8 kind);
 const struct btf_type *btf_type_skip_modifiers(const struct btf *btf,
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 2c2d1fb9f410..f3c998feeccb 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -1666,10 +1666,8 @@ static void btf_struct_metas_free(struct btf_struct_metas *tab)
 
 	if (!tab)
 		return;
-	for (i = 0; i < tab->cnt; i++) {
+	for (i = 0; i < tab->cnt; i++)
 		btf_record_free(tab->types[i].record);
-		kfree(tab->types[i].field_offs);
-	}
 	kfree(tab);
 }
 
@@ -3700,12 +3698,24 @@ static int btf_parse_rb_root(const struct btf *btf, struct btf_field *field,
 					    __alignof__(struct bpf_rb_node));
 }
 
+static int btf_field_cmp(const void *_a, const void *_b, const void *priv)
+{
+	const struct btf_field *a = (const struct btf_field *)_a;
+	const struct btf_field *b = (const struct btf_field *)_b;
+
+	if (a->offset < b->offset)
+		return -1;
+	else if (a->offset > b->offset)
+		return 1;
+	return 0;
+}
+
 struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type *t,
 				    u32 field_mask, u32 value_size)
 {
 	struct btf_field_info info_arr[BTF_FIELDS_MAX];
+	u32 next_off = 0, field_type_size;
 	struct btf_record *rec;
-	u32 next_off = 0;
 	int ret, i, cnt;
 
 	ret = btf_find_field(btf, t, field_mask, info_arr, ARRAY_SIZE(info_arr));
@@ -3725,7 +3735,8 @@ struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type
 	rec->spin_lock_off = -EINVAL;
 	rec->timer_off = -EINVAL;
 	for (i = 0; i < cnt; i++) {
-		if (info_arr[i].off + btf_field_type_size(info_arr[i].type) > value_size) {
+		field_type_size = btf_field_type_size(info_arr[i].type);
+		if (info_arr[i].off + field_type_size > value_size) {
 			WARN_ONCE(1, "verifier bug off %d size %d", info_arr[i].off, value_size);
 			ret = -EFAULT;
 			goto end;
@@ -3734,11 +3745,12 @@ struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type
 			ret = -EEXIST;
 			goto end;
 		}
-		next_off = info_arr[i].off + btf_field_type_size(info_arr[i].type);
+		next_off = info_arr[i].off + field_type_size;
 
 		rec->field_mask |= info_arr[i].type;
 		rec->fields[i].offset = info_arr[i].off;
 		rec->fields[i].type = info_arr[i].type;
+		rec->fields[i].size = field_type_size;
 
 		switch (info_arr[i].type) {
 		case BPF_SPIN_LOCK:
@@ -3808,6 +3820,9 @@ struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type
 		goto end;
 	}
 
+	sort_r(rec->fields, rec->cnt, sizeof(struct btf_field), btf_field_cmp,
+	       NULL, rec);
+
 	return rec;
 end:
 	btf_record_free(rec);
@@ -3889,61 +3904,6 @@ int btf_check_and_fixup_fields(const struct btf *btf, struct btf_record *rec)
 	return 0;
 }
 
-static int btf_field_offs_cmp(const void *_a, const void *_b, const void *priv)
-{
-	const u32 a = *(const u32 *)_a;
-	const u32 b = *(const u32 *)_b;
-
-	if (a < b)
-		return -1;
-	else if (a > b)
-		return 1;
-	return 0;
-}
-
-static void btf_field_offs_swap(void *_a, void *_b, int size, const void *priv)
-{
-	struct btf_field_offs *foffs = (void *)priv;
-	u32 *off_base = foffs->field_off;
-	u32 *a = _a, *b = _b;
-	u8 *sz_a, *sz_b;
-
-	sz_a = foffs->field_sz + (a - off_base);
-	sz_b = foffs->field_sz + (b - off_base);
-
-	swap(*a, *b);
-	swap(*sz_a, *sz_b);
-}
-
-struct btf_field_offs *btf_parse_field_offs(struct btf_record *rec)
-{
-	struct btf_field_offs *foffs;
-	u32 i, *off;
-	u8 *sz;
-
-	BUILD_BUG_ON(ARRAY_SIZE(foffs->field_off) != ARRAY_SIZE(foffs->field_sz));
-	if (IS_ERR_OR_NULL(rec))
-		return NULL;
-
-	foffs = kzalloc(sizeof(*foffs), GFP_KERNEL | __GFP_NOWARN);
-	if (!foffs)
-		return ERR_PTR(-ENOMEM);
-
-	off = foffs->field_off;
-	sz = foffs->field_sz;
-	for (i = 0; i < rec->cnt; i++) {
-		off[i] = rec->fields[i].offset;
-		sz[i] = btf_field_type_size(rec->fields[i].type);
-	}
-	foffs->cnt = rec->cnt;
-
-	if (foffs->cnt == 1)
-		return foffs;
-	sort_r(foffs->field_off, foffs->cnt, sizeof(foffs->field_off[0]),
-	       btf_field_offs_cmp, btf_field_offs_swap, foffs);
-	return foffs;
-}
-
 static void __btf_struct_show(const struct btf *btf, const struct btf_type *t,
 			      u32 type_id, void *data, u8 bits_offset,
 			      struct btf_show *show)
@@ -5386,7 +5346,6 @@ btf_parse_struct_metas(struct bpf_verifier_log *log, struct btf *btf)
 	for (i = 1; i < n; i++) {
 		struct btf_struct_metas *new_tab;
 		const struct btf_member *member;
-		struct btf_field_offs *foffs;
 		struct btf_struct_meta *type;
 		struct btf_record *record;
 		const struct btf_type *t;
@@ -5428,17 +5387,7 @@ btf_parse_struct_metas(struct bpf_verifier_log *log, struct btf *btf)
 			ret = PTR_ERR_OR_ZERO(record) ?: -EFAULT;
 			goto free;
 		}
-		foffs = btf_parse_field_offs(record);
-		/* We need the field_offs to be valid for a valid record,
-		 * either both should be set or both should be unset.
-		 */
-		if (IS_ERR_OR_NULL(foffs)) {
-			btf_record_free(record);
-			ret = -EFAULT;
-			goto free;
-		}
 		type->record = record;
-		type->field_offs = foffs;
 		tab->cnt++;
 	}
 	return tab;
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index f04e60a4847f..e989b6460147 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1893,7 +1893,7 @@ __bpf_kfunc void *bpf_obj_new_impl(u64 local_type_id__k, void *meta__ign)
 	if (!p)
 		return NULL;
 	if (meta)
-		bpf_obj_init(meta->field_offs, p);
+		bpf_obj_init(meta->record, p);
 	return p;
 }
 
diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c
index 38136ec4e095..2c5c64c2a53b 100644
--- a/kernel/bpf/map_in_map.c
+++ b/kernel/bpf/map_in_map.c
@@ -56,18 +56,6 @@ struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 		ret = PTR_ERR(inner_map_meta->record);
 		goto free;
 	}
-	if (inner_map_meta->record) {
-		struct btf_field_offs *field_offs;
-		/* If btf_record is !IS_ERR_OR_NULL, then field_offs is always
-		 * valid.
-		 */
-		field_offs = kmemdup(inner_map->field_offs, sizeof(*inner_map->field_offs), GFP_KERNEL | __GFP_NOWARN);
-		if (!field_offs) {
-			ret = -ENOMEM;
-			goto free_rec;
-		}
-		inner_map_meta->field_offs = field_offs;
-	}
 	/* Note: We must use the same BTF, as we also used btf_record_dup above
 	 * which relies on BTF being same for both maps, as some members like
 	 * record->fields.list_head have pointers like value_rec pointing into
@@ -88,8 +76,6 @@ struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 
 	fdput(f);
 	return inner_map_meta;
-free_rec:
-	btf_record_free(inner_map_meta->record);
 free:
 	kfree(inner_map_meta);
 put:
@@ -99,7 +85,6 @@ put:
 
 void bpf_map_meta_free(struct bpf_map *map_meta)
 {
-	kfree(map_meta->field_offs);
 	bpf_map_free_record(map_meta);
 	btf_put(map_meta->btf);
 	kfree(map_meta);
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 6d575505f89c..c08b7933bf8f 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -717,14 +717,13 @@ void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 static void bpf_map_free_deferred(struct work_struct *work)
 {
 	struct bpf_map *map = container_of(work, struct bpf_map, work);
-	struct btf_field_offs *foffs = map->field_offs;
 	struct btf_record *rec = map->record;
 
 	security_bpf_map_free(map);
 	bpf_map_release_memcg(map);
 	/* implementation dependent freeing */
 	map->ops->map_free(map);
-	/* Delay freeing of field_offs and btf_record for maps, as map_free
+	/* Delay freeing of btf_record for maps, as map_free
 	 * callback usually needs access to them. It is better to do it here
 	 * than require each callback to do the free itself manually.
 	 *
@@ -733,7 +732,6 @@ static void bpf_map_free_deferred(struct work_struct *work)
 	 * eventually calls bpf_map_free_meta, since inner_map_meta is only a
 	 * template bpf_map struct used during verification.
 	 */
-	kfree(foffs);
 	btf_record_free(rec);
 }
 
@@ -1125,7 +1123,6 @@ free_map_tab:
 static int map_create(union bpf_attr *attr)
 {
 	int numa_node = bpf_map_attr_numa_node(attr);
-	struct btf_field_offs *foffs;
 	struct bpf_map *map;
 	int f_flags;
 	int err;
@@ -1205,17 +1202,9 @@ static int map_create(union bpf_attr *attr)
 			attr->btf_vmlinux_value_type_id;
 	}
 
-
-	foffs = btf_parse_field_offs(map->record);
-	if (IS_ERR(foffs)) {
-		err = PTR_ERR(foffs);
-		goto free_map;
-	}
-	map->field_offs = foffs;
-
 	err = security_bpf_map_alloc(map);
 	if (err)
-		goto free_map_field_offs;
+		goto free_map;
 
 	err = bpf_map_alloc_id(map);
 	if (err)
@@ -1239,8 +1228,6 @@ static int map_create(union bpf_attr *attr)
 
 free_map_sec:
 	security_bpf_map_free(map);
-free_map_field_offs:
-	kfree(map->field_offs);
 free_map:
 	btf_put(map->btf);
 	map->ops->map_free(map);
-- 
cgit v1.2.3-70-g09d2


From d54730b50bae1f3119bd686d551d66f0fcc387ca Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:04 -0700
Subject: bpf: Introduce opaque bpf_refcount struct and add btf_record plumbing

A 'struct bpf_refcount' is added to the set of opaque uapi/bpf.h types
meant for use in BPF programs. Similarly to other opaque types like
bpf_spin_lock and bpf_rbtree_node, the verifier needs to know where in
user-defined struct types a bpf_refcount can be located, so necessary
btf_record plumbing is added to enable this. bpf_refcount is sized to
hold a refcount_t.

Similarly to bpf_spin_lock, the offset of a bpf_refcount is cached in
btf_record as refcount_off in addition to being in the field array.
Caching refcount_off makes sense for this field because further patches
in the series will modify functions that take local kptrs (e.g.
bpf_obj_drop) to change their behavior if the type they're operating on
is refcounted. So enabling fast "is this type refcounted?" checks is
desirable.

No such verifier behavior changes are introduced in this patch, just
logic to recognize 'struct bpf_refcount' in btf_record.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-3-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h            |  8 ++++++++
 include/uapi/linux/bpf.h       |  4 ++++
 kernel/bpf/btf.c               | 12 +++++++++++-
 kernel/bpf/syscall.c           |  6 +++++-
 tools/include/uapi/linux/bpf.h |  4 ++++
 5 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 7888ed497432..be44d765b7a4 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -187,6 +187,7 @@ enum btf_field_type {
 	BPF_RB_NODE    = (1 << 7),
 	BPF_GRAPH_NODE_OR_ROOT = BPF_LIST_NODE | BPF_LIST_HEAD |
 				 BPF_RB_NODE | BPF_RB_ROOT,
+	BPF_REFCOUNT   = (1 << 8),
 };
 
 typedef void (*btf_dtor_kfunc_t)(void *);
@@ -223,6 +224,7 @@ struct btf_record {
 	u32 field_mask;
 	int spin_lock_off;
 	int timer_off;
+	int refcount_off;
 	struct btf_field fields[];
 };
 
@@ -293,6 +295,8 @@ static inline const char *btf_field_type_name(enum btf_field_type type)
 		return "bpf_rb_root";
 	case BPF_RB_NODE:
 		return "bpf_rb_node";
+	case BPF_REFCOUNT:
+		return "bpf_refcount";
 	default:
 		WARN_ON_ONCE(1);
 		return "unknown";
@@ -317,6 +321,8 @@ static inline u32 btf_field_type_size(enum btf_field_type type)
 		return sizeof(struct bpf_rb_root);
 	case BPF_RB_NODE:
 		return sizeof(struct bpf_rb_node);
+	case BPF_REFCOUNT:
+		return sizeof(struct bpf_refcount);
 	default:
 		WARN_ON_ONCE(1);
 		return 0;
@@ -341,6 +347,8 @@ static inline u32 btf_field_type_align(enum btf_field_type type)
 		return __alignof__(struct bpf_rb_root);
 	case BPF_RB_NODE:
 		return __alignof__(struct bpf_rb_node);
+	case BPF_REFCOUNT:
+		return __alignof__(struct bpf_refcount);
 	default:
 		WARN_ON_ONCE(1);
 		return 0;
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 3823100b7934..4b20a7269bee 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -6985,6 +6985,10 @@ struct bpf_rb_node {
 	__u64 :64;
 } __attribute__((aligned(8)));
 
+struct bpf_refcount {
+	__u32 :32;
+} __attribute__((aligned(4)));
+
 struct bpf_sysctl {
 	__u32	write;		/* Sysctl is being read (= 0) or written (= 1).
 				 * Allows 1,2,4-byte read, but no write.
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index f3c998feeccb..14889fd5ba8e 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3391,6 +3391,7 @@ static int btf_get_field_type(const char *name, u32 field_mask, u32 *seen_mask,
 	field_mask_test_name(BPF_LIST_NODE, "bpf_list_node");
 	field_mask_test_name(BPF_RB_ROOT,   "bpf_rb_root");
 	field_mask_test_name(BPF_RB_NODE,   "bpf_rb_node");
+	field_mask_test_name(BPF_REFCOUNT,  "bpf_refcount");
 
 	/* Only return BPF_KPTR when all other types with matchable names fail */
 	if (field_mask & BPF_KPTR) {
@@ -3439,6 +3440,7 @@ static int btf_find_struct_field(const struct btf *btf,
 		case BPF_TIMER:
 		case BPF_LIST_NODE:
 		case BPF_RB_NODE:
+		case BPF_REFCOUNT:
 			ret = btf_find_struct(btf, member_type, off, sz, field_type,
 					      idx < info_cnt ? &info[idx] : &tmp);
 			if (ret < 0)
@@ -3504,6 +3506,7 @@ static int btf_find_datasec_var(const struct btf *btf, const struct btf_type *t,
 		case BPF_TIMER:
 		case BPF_LIST_NODE:
 		case BPF_RB_NODE:
+		case BPF_REFCOUNT:
 			ret = btf_find_struct(btf, var_type, off, sz, field_type,
 					      idx < info_cnt ? &info[idx] : &tmp);
 			if (ret < 0)
@@ -3734,6 +3737,7 @@ struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type
 
 	rec->spin_lock_off = -EINVAL;
 	rec->timer_off = -EINVAL;
+	rec->refcount_off = -EINVAL;
 	for (i = 0; i < cnt; i++) {
 		field_type_size = btf_field_type_size(info_arr[i].type);
 		if (info_arr[i].off + field_type_size > value_size) {
@@ -3763,6 +3767,11 @@ struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type
 			/* Cache offset for faster lookup at runtime */
 			rec->timer_off = rec->fields[i].offset;
 			break;
+		case BPF_REFCOUNT:
+			WARN_ON_ONCE(rec->refcount_off >= 0);
+			/* Cache offset for faster lookup at runtime */
+			rec->refcount_off = rec->fields[i].offset;
+			break;
 		case BPF_KPTR_UNREF:
 		case BPF_KPTR_REF:
 			ret = btf_parse_kptr(btf, &rec->fields[i], &info_arr[i]);
@@ -5308,6 +5317,7 @@ static const char *alloc_obj_fields[] = {
 	"bpf_list_node",
 	"bpf_rb_root",
 	"bpf_rb_node",
+	"bpf_refcount",
 };
 
 static struct btf_struct_metas *
@@ -5381,7 +5391,7 @@ btf_parse_struct_metas(struct bpf_verifier_log *log, struct btf *btf)
 		type = &tab->types[tab->cnt];
 		type->btf_id = i;
 		record = btf_parse_fields(btf, t, BPF_SPIN_LOCK | BPF_LIST_HEAD | BPF_LIST_NODE |
-						  BPF_RB_ROOT | BPF_RB_NODE, t->size);
+						  BPF_RB_ROOT | BPF_RB_NODE | BPF_REFCOUNT, t->size);
 		/* The record cannot be unset, treat it as an error if so */
 		if (IS_ERR_OR_NULL(record)) {
 			ret = PTR_ERR_OR_ZERO(record) ?: -EFAULT;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index c08b7933bf8f..28eac7434d32 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -552,6 +552,7 @@ void btf_record_free(struct btf_record *rec)
 		case BPF_RB_NODE:
 		case BPF_SPIN_LOCK:
 		case BPF_TIMER:
+		case BPF_REFCOUNT:
 			/* Nothing to release */
 			break;
 		default:
@@ -599,6 +600,7 @@ struct btf_record *btf_record_dup(const struct btf_record *rec)
 		case BPF_RB_NODE:
 		case BPF_SPIN_LOCK:
 		case BPF_TIMER:
+		case BPF_REFCOUNT:
 			/* Nothing to acquire */
 			break;
 		default:
@@ -705,6 +707,7 @@ void bpf_obj_free_fields(const struct btf_record *rec, void *obj)
 			break;
 		case BPF_LIST_NODE:
 		case BPF_RB_NODE:
+		case BPF_REFCOUNT:
 			break;
 		default:
 			WARN_ON_ONCE(1);
@@ -1032,7 +1035,7 @@ static int map_check_btf(struct bpf_map *map, const struct btf *btf,
 
 	map->record = btf_parse_fields(btf, value_type,
 				       BPF_SPIN_LOCK | BPF_TIMER | BPF_KPTR | BPF_LIST_HEAD |
-				       BPF_RB_ROOT,
+				       BPF_RB_ROOT | BPF_REFCOUNT,
 				       map->value_size);
 	if (!IS_ERR_OR_NULL(map->record)) {
 		int i;
@@ -1071,6 +1074,7 @@ static int map_check_btf(struct bpf_map *map, const struct btf *btf,
 				break;
 			case BPF_KPTR_UNREF:
 			case BPF_KPTR_REF:
+			case BPF_REFCOUNT:
 				if (map->map_type != BPF_MAP_TYPE_HASH &&
 				    map->map_type != BPF_MAP_TYPE_PERCPU_HASH &&
 				    map->map_type != BPF_MAP_TYPE_LRU_HASH &&
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 3823100b7934..4b20a7269bee 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -6985,6 +6985,10 @@ struct bpf_rb_node {
 	__u64 :64;
 } __attribute__((aligned(8)));
 
+struct bpf_refcount {
+	__u32 :32;
+} __attribute__((aligned(4)));
+
 struct bpf_sysctl {
 	__u32	write;		/* Sysctl is being read (= 0) or written (= 1).
 				 * Allows 1,2,4-byte read, but no write.
-- 
cgit v1.2.3-70-g09d2


From 1512217c47f0e8ea076dd0e67262e5a668a78f01 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:05 -0700
Subject: bpf: Support refcounted local kptrs in existing semantics

A local kptr is considered 'refcounted' when it is of a type that has a
bpf_refcount field. When such a kptr is created, its refcount should be
initialized to 1; when destroyed, the object should be free'd only if a
refcount decr results in 0 refcount.

Existing logic always frees the underlying memory when destroying a
local kptr, and 0-initializes all btf_record fields. This patch adds
checks for "is local kptr refcounted?" and new logic for that case in
the appropriate places.

This patch focuses on changing existing semantics and thus conspicuously
does _not_ provide a way for BPF programs in increment refcount. That
follows later in the series.

__bpf_obj_drop_impl is modified to do the right thing when it sees a
refcounted type. Container types for graph nodes (list, tree, stashed in
map) are migrated to use __bpf_obj_drop_impl as a destructor for their
nodes instead of each having custom destruction code in their _free
paths. Now that "drop" isn't a synonym for "free" when the type is
refcounted it makes sense to centralize this logic.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-4-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  |  3 +++
 kernel/bpf/helpers.c | 21 +++++++++++++--------
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index be44d765b7a4..b065de2cfcea 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -370,6 +370,9 @@ static inline void bpf_obj_init(const struct btf_record *rec, void *obj)
 		return;
 	for (i = 0; i < rec->cnt; i++)
 		memset(obj + rec->fields[i].offset, 0, rec->fields[i].size);
+
+	if (rec->refcount_off >= 0)
+		refcount_set((refcount_t *)(obj + rec->refcount_off), 1);
 }
 
 /* 'dst' must be a temporary buffer and should not point to memory that is being
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index e989b6460147..e2dbd9644e5c 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1798,6 +1798,8 @@ bpf_base_func_proto(enum bpf_func_id func_id)
 	}
 }
 
+void __bpf_obj_drop_impl(void *p, const struct btf_record *rec);
+
 void bpf_list_head_free(const struct btf_field *field, void *list_head,
 			struct bpf_spin_lock *spin_lock)
 {
@@ -1828,13 +1830,8 @@ unlock:
 		/* The contained type can also have resources, including a
 		 * bpf_list_head which needs to be freed.
 		 */
-		bpf_obj_free_fields(field->graph_root.value_rec, obj);
-		/* bpf_mem_free requires migrate_disable(), since we can be
-		 * called from map free path as well apart from BPF program (as
-		 * part of map ops doing bpf_obj_free_fields).
-		 */
 		migrate_disable();
-		bpf_mem_free(&bpf_global_ma, obj);
+		__bpf_obj_drop_impl(obj, field->graph_root.value_rec);
 		migrate_enable();
 	}
 }
@@ -1871,10 +1868,9 @@ void bpf_rb_root_free(const struct btf_field *field, void *rb_root,
 		obj = pos;
 		obj -= field->graph_root.node_offset;
 
-		bpf_obj_free_fields(field->graph_root.value_rec, obj);
 
 		migrate_disable();
-		bpf_mem_free(&bpf_global_ma, obj);
+		__bpf_obj_drop_impl(obj, field->graph_root.value_rec);
 		migrate_enable();
 	}
 }
@@ -1897,8 +1893,17 @@ __bpf_kfunc void *bpf_obj_new_impl(u64 local_type_id__k, void *meta__ign)
 	return p;
 }
 
+/* Must be called under migrate_disable(), as required by bpf_mem_free */
 void __bpf_obj_drop_impl(void *p, const struct btf_record *rec)
 {
+	if (rec && rec->refcount_off >= 0 &&
+	    !refcount_dec_and_test((refcount_t *)(p + rec->refcount_off))) {
+		/* Object is refcounted and refcount_dec didn't result in 0
+		 * refcount. Return without freeing the object
+		 */
+		return;
+	}
+
 	if (rec)
 		bpf_obj_free_fields(rec, p);
 	bpf_mem_free(&bpf_global_ma, p);
-- 
cgit v1.2.3-70-g09d2


From 7c50b1cb76aca4540aa917db5f2a302acddcadff Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:06 -0700
Subject: bpf: Add bpf_refcount_acquire kfunc

Currently, BPF programs can interact with the lifetime of refcounted
local kptrs in the following ways:

  bpf_obj_new  - Initialize refcount to 1 as part of new object creation
  bpf_obj_drop - Decrement refcount and free object if it's 0
  collection add - Pass ownership to the collection. No change to
                   refcount but collection is responsible for
		   bpf_obj_dropping it

In order to be able to add a refcounted local kptr to multiple
collections we need to be able to increment the refcount and acquire a
new owning reference. This patch adds a kfunc, bpf_refcount_acquire,
implementing such an operation.

bpf_refcount_acquire takes a refcounted local kptr and returns a new
owning reference to the same underlying memory as the input. The input
can be either owning or non-owning. To reinforce why this is safe,
consider the following code snippets:

  struct node *n = bpf_obj_new(typeof(*n)); // A
  struct node *m = bpf_refcount_acquire(n); // B

In the above snippet, n will be alive with refcount=1 after (A), and
since nothing changes that state before (B), it's obviously safe. If
n is instead added to some rbtree, we can still safely refcount_acquire
it:

  struct node *n = bpf_obj_new(typeof(*n));
  struct node *m;

  bpf_spin_lock(&glock);
  bpf_rbtree_add(&groot, &n->node, less);   // A
  m = bpf_refcount_acquire(n);              // B
  bpf_spin_unlock(&glock);

In the above snippet, after (A) n is a non-owning reference, and after
(B) m is an owning reference pointing to the same memory as n. Although
n has no ownership of that memory's lifetime, it's guaranteed to be
alive until the end of the critical section, and n would be clobbered if
we were past the end of the critical section, so it's safe to bump
refcount.

Implementation details:

* From verifier's perspective, bpf_refcount_acquire handling is similar
  to bpf_obj_new and bpf_obj_drop. Like the former, it returns a new
  owning reference matching input type, although like the latter, type
  can be inferred from concrete kptr input. Verifier changes in
  {check,fixup}_kfunc_call and check_kfunc_args are largely copied from
  aforementioned functions' verifier changes.

* An exception to the above is the new KF_ARG_PTR_TO_REFCOUNTED_KPTR
  arg, indicated by new "__refcounted_kptr" kfunc arg suffix. This is
  necessary in order to handle both owning and non-owning input without
  adding special-casing to "__alloc" arg handling. Also a convenient
  place to confirm that input type has bpf_refcount field.

* The implemented kfunc is actually bpf_refcount_acquire_impl, with
  'hidden' second arg that the verifier sets to the type's struct_meta
  in fixup_kfunc_call.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-5-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c                           | 15 ++++++
 kernel/bpf/verifier.c                          | 74 ++++++++++++++++++++++----
 tools/testing/selftests/bpf/bpf_experimental.h | 13 +++++
 3 files changed, 91 insertions(+), 11 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index e2dbd9644e5c..57ff8a60222c 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1917,6 +1917,20 @@ __bpf_kfunc void bpf_obj_drop_impl(void *p__alloc, void *meta__ign)
 	__bpf_obj_drop_impl(p, meta ? meta->record : NULL);
 }
 
+__bpf_kfunc void *bpf_refcount_acquire_impl(void *p__refcounted_kptr, void *meta__ign)
+{
+	struct btf_struct_meta *meta = meta__ign;
+	struct bpf_refcount *ref;
+
+	/* Could just cast directly to refcount_t *, but need some code using
+	 * bpf_refcount type so that it is emitted in vmlinux BTF
+	 */
+	ref = (struct bpf_refcount *)p__refcounted_kptr + meta->record->refcount_off;
+
+	refcount_inc((refcount_t *)ref);
+	return (void *)p__refcounted_kptr;
+}
+
 static void __bpf_list_add(struct bpf_list_node *node, struct bpf_list_head *head, bool tail)
 {
 	struct list_head *n = (void *)node, *h = (void *)head;
@@ -2276,6 +2290,7 @@ BTF_ID_FLAGS(func, crash_kexec, KF_DESTRUCTIVE)
 #endif
 BTF_ID_FLAGS(func, bpf_obj_new_impl, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_obj_drop_impl, KF_RELEASE)
+BTF_ID_FLAGS(func, bpf_refcount_acquire_impl, KF_ACQUIRE)
 BTF_ID_FLAGS(func, bpf_list_push_front)
 BTF_ID_FLAGS(func, bpf_list_push_back)
 BTF_ID_FLAGS(func, bpf_list_pop_front, KF_ACQUIRE | KF_RET_NULL)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 4aa6d715e655..29e106f7ccaa 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -273,6 +273,11 @@ struct bpf_call_arg_meta {
 	struct btf_field *kptr_field;
 };
 
+struct btf_and_id {
+	struct btf *btf;
+	u32 btf_id;
+};
+
 struct bpf_kfunc_call_arg_meta {
 	/* In parameters */
 	struct btf *btf;
@@ -291,10 +296,10 @@ struct bpf_kfunc_call_arg_meta {
 		u64 value;
 		bool found;
 	} arg_constant;
-	struct {
-		struct btf *btf;
-		u32 btf_id;
-	} arg_obj_drop;
+	union {
+		struct btf_and_id arg_obj_drop;
+		struct btf_and_id arg_refcount_acquire;
+	};
 	struct {
 		struct btf_field *field;
 	} arg_list_head;
@@ -9403,6 +9408,11 @@ static bool is_kfunc_arg_uninit(const struct btf *btf, const struct btf_param *a
 	return __kfunc_param_match_suffix(btf, arg, "__uninit");
 }
 
+static bool is_kfunc_arg_refcounted_kptr(const struct btf *btf, const struct btf_param *arg)
+{
+	return __kfunc_param_match_suffix(btf, arg, "__refcounted_kptr");
+}
+
 static bool is_kfunc_arg_scalar_with_name(const struct btf *btf,
 					  const struct btf_param *arg,
 					  const char *name)
@@ -9542,15 +9552,16 @@ static u32 *reg2btf_ids[__BPF_REG_TYPE_MAX] = {
 
 enum kfunc_ptr_arg_type {
 	KF_ARG_PTR_TO_CTX,
-	KF_ARG_PTR_TO_ALLOC_BTF_ID,  /* Allocated object */
-	KF_ARG_PTR_TO_KPTR,	     /* PTR_TO_KPTR but type specific */
+	KF_ARG_PTR_TO_ALLOC_BTF_ID,    /* Allocated object */
+	KF_ARG_PTR_TO_REFCOUNTED_KPTR, /* Refcounted local kptr */
+	KF_ARG_PTR_TO_KPTR,	       /* PTR_TO_KPTR but type specific */
 	KF_ARG_PTR_TO_DYNPTR,
 	KF_ARG_PTR_TO_ITER,
 	KF_ARG_PTR_TO_LIST_HEAD,
 	KF_ARG_PTR_TO_LIST_NODE,
-	KF_ARG_PTR_TO_BTF_ID,	     /* Also covers reg2btf_ids conversions */
+	KF_ARG_PTR_TO_BTF_ID,	       /* Also covers reg2btf_ids conversions */
 	KF_ARG_PTR_TO_MEM,
-	KF_ARG_PTR_TO_MEM_SIZE,	     /* Size derived from next argument, skip it */
+	KF_ARG_PTR_TO_MEM_SIZE,	       /* Size derived from next argument, skip it */
 	KF_ARG_PTR_TO_CALLBACK,
 	KF_ARG_PTR_TO_RB_ROOT,
 	KF_ARG_PTR_TO_RB_NODE,
@@ -9559,6 +9570,7 @@ enum kfunc_ptr_arg_type {
 enum special_kfunc_type {
 	KF_bpf_obj_new_impl,
 	KF_bpf_obj_drop_impl,
+	KF_bpf_refcount_acquire_impl,
 	KF_bpf_list_push_front,
 	KF_bpf_list_push_back,
 	KF_bpf_list_pop_front,
@@ -9579,6 +9591,7 @@ enum special_kfunc_type {
 BTF_SET_START(special_kfunc_set)
 BTF_ID(func, bpf_obj_new_impl)
 BTF_ID(func, bpf_obj_drop_impl)
+BTF_ID(func, bpf_refcount_acquire_impl)
 BTF_ID(func, bpf_list_push_front)
 BTF_ID(func, bpf_list_push_back)
 BTF_ID(func, bpf_list_pop_front)
@@ -9597,6 +9610,7 @@ BTF_SET_END(special_kfunc_set)
 BTF_ID_LIST(special_kfunc_list)
 BTF_ID(func, bpf_obj_new_impl)
 BTF_ID(func, bpf_obj_drop_impl)
+BTF_ID(func, bpf_refcount_acquire_impl)
 BTF_ID(func, bpf_list_push_front)
 BTF_ID(func, bpf_list_push_back)
 BTF_ID(func, bpf_list_pop_front)
@@ -9649,6 +9663,9 @@ get_kfunc_ptr_arg_type(struct bpf_verifier_env *env,
 	if (is_kfunc_arg_alloc_obj(meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_ALLOC_BTF_ID;
 
+	if (is_kfunc_arg_refcounted_kptr(meta->btf, &args[argno]))
+		return KF_ARG_PTR_TO_REFCOUNTED_KPTR;
+
 	if (is_kfunc_arg_kptr_get(meta, argno)) {
 		if (!btf_type_is_ptr(ref_t)) {
 			verbose(env, "arg#0 BTF type must be a double pointer for kptr_get kfunc\n");
@@ -9952,7 +9969,8 @@ static bool is_bpf_rbtree_api_kfunc(u32 btf_id)
 
 static bool is_bpf_graph_api_kfunc(u32 btf_id)
 {
-	return is_bpf_list_api_kfunc(btf_id) || is_bpf_rbtree_api_kfunc(btf_id);
+	return is_bpf_list_api_kfunc(btf_id) || is_bpf_rbtree_api_kfunc(btf_id) ||
+	       btf_id == special_kfunc_list[KF_bpf_refcount_acquire_impl];
 }
 
 static bool is_callback_calling_kfunc(u32 btf_id)
@@ -10171,6 +10189,7 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 	const char *func_name = meta->func_name, *ref_tname;
 	const struct btf *btf = meta->btf;
 	const struct btf_param *args;
+	struct btf_record *rec;
 	u32 i, nargs;
 	int ret;
 
@@ -10306,6 +10325,7 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 		case KF_ARG_PTR_TO_MEM:
 		case KF_ARG_PTR_TO_MEM_SIZE:
 		case KF_ARG_PTR_TO_CALLBACK:
+		case KF_ARG_PTR_TO_REFCOUNTED_KPTR:
 			/* Trusted by default */
 			break;
 		default:
@@ -10523,6 +10543,26 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 		case KF_ARG_PTR_TO_CALLBACK:
 			meta->subprogno = reg->subprogno;
 			break;
+		case KF_ARG_PTR_TO_REFCOUNTED_KPTR:
+			if (!type_is_ptr_alloc_obj(reg->type) && !type_is_non_owning_ref(reg->type)) {
+				verbose(env, "arg#%d is neither owning or non-owning ref\n", i);
+				return -EINVAL;
+			}
+
+			rec = reg_btf_record(reg);
+			if (!rec) {
+				verbose(env, "verifier internal error: Couldn't find btf_record\n");
+				return -EFAULT;
+			}
+
+			if (rec->refcount_off < 0) {
+				verbose(env, "arg#%d doesn't point to a type with bpf_refcount field\n", i);
+				return -EINVAL;
+			}
+
+			meta->arg_refcount_acquire.btf = reg->btf;
+			meta->arg_refcount_acquire.btf_id = reg->btf_id;
+			break;
 		}
 	}
 
@@ -10699,7 +10739,9 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 
 	if (is_kfunc_acquire(&meta) && !btf_type_is_struct_ptr(meta.btf, t)) {
 		/* Only exception is bpf_obj_new_impl */
-		if (meta.btf != btf_vmlinux || meta.func_id != special_kfunc_list[KF_bpf_obj_new_impl]) {
+		if (meta.btf != btf_vmlinux ||
+		    (meta.func_id != special_kfunc_list[KF_bpf_obj_new_impl] &&
+		     meta.func_id != special_kfunc_list[KF_bpf_refcount_acquire_impl])) {
 			verbose(env, "acquire kernel function does not return PTR_TO_BTF_ID\n");
 			return -EINVAL;
 		}
@@ -10747,6 +10789,15 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 				insn_aux->obj_new_size = ret_t->size;
 				insn_aux->kptr_struct_meta =
 					btf_find_struct_meta(ret_btf, ret_btf_id);
+			} else if (meta.func_id == special_kfunc_list[KF_bpf_refcount_acquire_impl]) {
+				mark_reg_known_zero(env, regs, BPF_REG_0);
+				regs[BPF_REG_0].type = PTR_TO_BTF_ID | MEM_ALLOC;
+				regs[BPF_REG_0].btf = meta.arg_refcount_acquire.btf;
+				regs[BPF_REG_0].btf_id = meta.arg_refcount_acquire.btf_id;
+
+				insn_aux->kptr_struct_meta =
+					btf_find_struct_meta(meta.arg_refcount_acquire.btf,
+							     meta.arg_refcount_acquire.btf_id);
 			} else if (meta.func_id == special_kfunc_list[KF_bpf_list_pop_front] ||
 				   meta.func_id == special_kfunc_list[KF_bpf_list_pop_back]) {
 				struct btf_field *field = meta.arg_list_head.field;
@@ -17393,7 +17444,8 @@ static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		insn_buf[2] = addr[1];
 		insn_buf[3] = *insn;
 		*cnt = 4;
-	} else if (desc->func_id == special_kfunc_list[KF_bpf_obj_drop_impl]) {
+	} else if (desc->func_id == special_kfunc_list[KF_bpf_obj_drop_impl] ||
+		   desc->func_id == special_kfunc_list[KF_bpf_refcount_acquire_impl]) {
 		struct btf_struct_meta *kptr_struct_meta = env->insn_aux_data[insn_idx].kptr_struct_meta;
 		struct bpf_insn addr[2] = { BPF_LD_IMM64(BPF_REG_2, (long)kptr_struct_meta) };
 
diff --git a/tools/testing/selftests/bpf/bpf_experimental.h b/tools/testing/selftests/bpf/bpf_experimental.h
index dbd2c729781a..619afcab2ab0 100644
--- a/tools/testing/selftests/bpf/bpf_experimental.h
+++ b/tools/testing/selftests/bpf/bpf_experimental.h
@@ -37,6 +37,19 @@ extern void bpf_obj_drop_impl(void *kptr, void *meta) __ksym;
 /* Convenience macro to wrap over bpf_obj_drop_impl */
 #define bpf_obj_drop(kptr) bpf_obj_drop_impl(kptr, NULL)
 
+/* Description
+ *	Increment the refcount on a refcounted local kptr, turning the
+ *	non-owning reference input into an owning reference in the process.
+ *
+ *	The 'meta' parameter is a hidden argument that is ignored.
+ * Returns
+ *	An owning reference to the object pointed to by 'kptr'
+ */
+extern void *bpf_refcount_acquire_impl(void *kptr, void *meta) __ksym;
+
+/* Convenience macro to wrap over bpf_refcount_acquire_impl */
+#define bpf_refcount_acquire(kptr) bpf_refcount_acquire_impl(kptr, NULL)
+
 /* Description
  *	Add a new entry to the beginning of the BPF linked list.
  * Returns
-- 
cgit v1.2.3-70-g09d2


From d2dcc67df910dd85253a701b6a5b747f955d28f5 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:07 -0700
Subject: bpf: Migrate bpf_rbtree_add and bpf_list_push_{front,back} to
 possibly fail

Consider this code snippet:

  struct node {
    long key;
    bpf_list_node l;
    bpf_rb_node r;
    bpf_refcount ref;
  }

  int some_bpf_prog(void *ctx)
  {
    struct node *n = bpf_obj_new(/*...*/), *m;

    bpf_spin_lock(&glock);

    bpf_rbtree_add(&some_tree, &n->r, /* ... */);
    m = bpf_refcount_acquire(n);
    bpf_rbtree_add(&other_tree, &m->r, /* ... */);

    bpf_spin_unlock(&glock);

    /* ... */
  }

After bpf_refcount_acquire, n and m point to the same underlying memory,
and that node's bpf_rb_node field is being used by the some_tree insert,
so overwriting it as a result of the second insert is an error. In order
to properly support refcounted nodes, the rbtree and list insert
functions must be allowed to fail. This patch adds such support.

The kfuncs bpf_rbtree_add, bpf_list_push_{front,back} are modified to
return an int indicating success/failure, with 0 -> success, nonzero ->
failure.

bpf_obj_drop on failure
=======================

Currently the only reason an insert can fail is the example above: the
bpf_{list,rb}_node is already in use. When such a failure occurs, the
insert kfuncs will bpf_obj_drop the input node. This allows the insert
operations to logically fail without changing their verifier owning ref
behavior, namely the unconditional release_reference of the input
owning ref.

With insert that always succeeds, ownership of the node is always passed
to the collection, since the node always ends up in the collection.

With a possibly-failed insert w/ bpf_obj_drop, ownership of the node
is always passed either to the collection (success), or to bpf_obj_drop
(failure). Regardless, it's correct to continue unconditionally
releasing the input owning ref, as something is always taking ownership
from the calling program on insert.

Keeping owning ref behavior unchanged results in a nice default UX for
insert functions that can fail. If the program's reaction to a failed
insert is "fine, just get rid of this owning ref for me and let me go
on with my business", then there's no reason to check for failure since
that's default behavior. e.g.:

  long important_failures = 0;

  int some_bpf_prog(void *ctx)
  {
    struct node *n, *m, *o; /* all bpf_obj_new'd */

    bpf_spin_lock(&glock);
    bpf_rbtree_add(&some_tree, &n->node, /* ... */);
    bpf_rbtree_add(&some_tree, &m->node, /* ... */);
    if (bpf_rbtree_add(&some_tree, &o->node, /* ... */)) {
      important_failures++;
    }
    bpf_spin_unlock(&glock);
  }

If we instead chose to pass ownership back to the program on failed
insert - by returning NULL on success or an owning ref on failure -
programs would always have to do something with the returned ref on
failure. The most likely action is probably "I'll just get rid of this
owning ref and go about my business", which ideally would look like:

  if (n = bpf_rbtree_add(&some_tree, &n->node, /* ... */))
    bpf_obj_drop(n);

But bpf_obj_drop isn't allowed in a critical section and inserts must
occur within one, so in reality error handling would become a
hard-to-parse mess.

For refcounted nodes, we can replicate the "pass ownership back to
program on failure" logic with this patch's semantics, albeit in an ugly
way:

  struct node *n = bpf_obj_new(/* ... */), *m;

  bpf_spin_lock(&glock);

  m = bpf_refcount_acquire(n);
  if (bpf_rbtree_add(&some_tree, &n->node, /* ... */)) {
    /* Do something with m */
  }

  bpf_spin_unlock(&glock);
  bpf_obj_drop(m);

bpf_refcount_acquire is used to simulate "return owning ref on failure".
This should be an uncommon occurrence, though.

Addition of two verifier-fixup'd args to collection inserts
===========================================================

The actual bpf_obj_drop kfunc is
bpf_obj_drop_impl(void *, struct btf_struct_meta *), with bpf_obj_drop
macro populating the second arg with 0 and the verifier later filling in
the arg during insn fixup.

Because bpf_rbtree_add and bpf_list_push_{front,back} now might do
bpf_obj_drop, these kfuncs need a btf_struct_meta parameter that can be
passed to bpf_obj_drop_impl.

Similarly, because the 'node' param to those insert functions is the
bpf_{list,rb}_node within the node type, and bpf_obj_drop expects a
pointer to the beginning of the node, the insert functions need to be
able to find the beginning of the node struct. A second
verifier-populated param is necessary: the offset of {list,rb}_node within the
node type.

These two new params allow the insert kfuncs to correctly call
__bpf_obj_drop_impl:

  beginning_of_node = bpf_rb_node_ptr - offset
  if (already_inserted)
    __bpf_obj_drop_impl(beginning_of_node, btf_struct_meta->record);

Similarly to other kfuncs with "hidden" verifier-populated params, the
insert functions are renamed with _impl prefix and a macro is provided
for common usage. For example, bpf_rbtree_add kfunc is now
bpf_rbtree_add_impl and bpf_rbtree_add is now a macro which sets
"hidden" args to 0.

Due to the two new args BPF progs will need to be recompiled to work
with the new _impl kfuncs.

This patch also rewrites the "hidden argument" explanation to more
directly say why the BPF program writer doesn't need to populate the
arguments with anything meaningful.

How does this new logic affect non-owning references?
=====================================================

Currently, non-owning refs are valid until the end of the critical
section in which they're created. We can make this guarantee because, if
a non-owning ref exists, the referent was added to some collection. The
collection will drop() its nodes when it goes away, but it can't go away
while our program is accessing it, so that's not a problem. If the
referent is removed from the collection in the same CS that it was added
in, it can't be bpf_obj_drop'd until after CS end. Those are the only
two ways to free the referent's memory and neither can happen until
after the non-owning ref's lifetime ends.

On first glance, having these collection insert functions potentially
bpf_obj_drop their input seems like it breaks the "can't be
bpf_obj_drop'd until after CS end" line of reasoning. But we care about
the memory not being _freed_ until end of CS end, and a previous patch
in the series modified bpf_obj_drop such that it doesn't free refcounted
nodes until refcount == 0. So the statement can be more accurately
rewritten as "can't be free'd until after CS end".

We can prove that this rewritten statement holds for any non-owning
reference produced by collection insert functions:

* If the input to the insert function is _not_ refcounted
  * We have an owning reference to the input, and can conclude it isn't
    in any collection
    * Inserting a node in a collection turns owning refs into
      non-owning, and since our input type isn't refcounted, there's no
      way to obtain additional owning refs to the same underlying
      memory
  * Because our node isn't in any collection, the insert operation
    cannot fail, so bpf_obj_drop will not execute
  * If bpf_obj_drop is guaranteed not to execute, there's no risk of
    memory being free'd

* Otherwise, the input to the insert function is refcounted
  * If the insert operation fails due to the node's list_head or rb_root
    already being in some collection, there was some previous successful
    insert which passed refcount to the collection
  * We have an owning reference to the input, it must have been
    acquired via bpf_refcount_acquire, which bumped the refcount
  * refcount must be >= 2 since there's a valid owning reference and the
    node is already in a collection
  * Insert triggering bpf_obj_drop will decr refcount to >= 1, never
    resulting in a free

So although we may do bpf_obj_drop during the critical section, this
will never result in memory being free'd, and no changes to non-owning
ref logic are needed in this patch.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-6-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_verifier.h                   |  7 ++-
 kernel/bpf/helpers.c                           | 65 +++++++++++++++------
 kernel/bpf/verifier.c                          | 78 ++++++++++++++++++--------
 tools/testing/selftests/bpf/bpf_experimental.h | 49 ++++++++++++----
 4 files changed, 148 insertions(+), 51 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index f03852b89d28..3dd29a53b711 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -464,7 +464,12 @@ struct bpf_insn_aux_data {
 		 */
 		struct bpf_loop_inline_state loop_inline_state;
 	};
-	u64 obj_new_size; /* remember the size of type passed to bpf_obj_new to rewrite R1 */
+	union {
+		/* remember the size of type passed to bpf_obj_new to rewrite R1 */
+		u64 obj_new_size;
+		/* remember the offset of node field within type to rewrite */
+		u64 insert_off;
+	};
 	struct btf_struct_meta *kptr_struct_meta;
 	u64 map_key_state; /* constant (32 bit) key tracking for maps */
 	int ctx_field_size; /* the ctx field size for load insn, maybe 0 */
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 57ff8a60222c..5067f8d46872 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1931,7 +1931,8 @@ __bpf_kfunc void *bpf_refcount_acquire_impl(void *p__refcounted_kptr, void *meta
 	return (void *)p__refcounted_kptr;
 }
 
-static void __bpf_list_add(struct bpf_list_node *node, struct bpf_list_head *head, bool tail)
+static int __bpf_list_add(struct bpf_list_node *node, struct bpf_list_head *head,
+			  bool tail, struct btf_record *rec, u64 off)
 {
 	struct list_head *n = (void *)node, *h = (void *)head;
 
@@ -1939,17 +1940,35 @@ static void __bpf_list_add(struct bpf_list_node *node, struct bpf_list_head *hea
 		INIT_LIST_HEAD(h);
 	if (unlikely(!n->next))
 		INIT_LIST_HEAD(n);
+	if (!list_empty(n)) {
+		/* Only called from BPF prog, no need to migrate_disable */
+		__bpf_obj_drop_impl(n - off, rec);
+		return -EINVAL;
+	}
+
 	tail ? list_add_tail(n, h) : list_add(n, h);
+
+	return 0;
 }
 
-__bpf_kfunc void bpf_list_push_front(struct bpf_list_head *head, struct bpf_list_node *node)
+__bpf_kfunc int bpf_list_push_front_impl(struct bpf_list_head *head,
+					 struct bpf_list_node *node,
+					 void *meta__ign, u64 off)
 {
-	return __bpf_list_add(node, head, false);
+	struct btf_struct_meta *meta = meta__ign;
+
+	return __bpf_list_add(node, head, false,
+			      meta ? meta->record : NULL, off);
 }
 
-__bpf_kfunc void bpf_list_push_back(struct bpf_list_head *head, struct bpf_list_node *node)
+__bpf_kfunc int bpf_list_push_back_impl(struct bpf_list_head *head,
+					struct bpf_list_node *node,
+					void *meta__ign, u64 off)
 {
-	return __bpf_list_add(node, head, true);
+	struct btf_struct_meta *meta = meta__ign;
+
+	return __bpf_list_add(node, head, true,
+			      meta ? meta->record : NULL, off);
 }
 
 static struct bpf_list_node *__bpf_list_del(struct bpf_list_head *head, bool tail)
@@ -1989,14 +2008,23 @@ __bpf_kfunc struct bpf_rb_node *bpf_rbtree_remove(struct bpf_rb_root *root,
 /* Need to copy rbtree_add_cached's logic here because our 'less' is a BPF
  * program
  */
-static void __bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
-			     void *less)
+static int __bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
+			    void *less, struct btf_record *rec, u64 off)
 {
 	struct rb_node **link = &((struct rb_root_cached *)root)->rb_root.rb_node;
+	struct rb_node *parent = NULL, *n = (struct rb_node *)node;
 	bpf_callback_t cb = (bpf_callback_t)less;
-	struct rb_node *parent = NULL;
 	bool leftmost = true;
 
+	if (!n->__rb_parent_color)
+		RB_CLEAR_NODE(n);
+
+	if (!RB_EMPTY_NODE(n)) {
+		/* Only called from BPF prog, no need to migrate_disable */
+		__bpf_obj_drop_impl(n - off, rec);
+		return -EINVAL;
+	}
+
 	while (*link) {
 		parent = *link;
 		if (cb((uintptr_t)node, (uintptr_t)parent, 0, 0, 0)) {
@@ -2007,15 +2035,18 @@ static void __bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
 		}
 	}
 
-	rb_link_node((struct rb_node *)node, parent, link);
-	rb_insert_color_cached((struct rb_node *)node,
-			       (struct rb_root_cached *)root, leftmost);
+	rb_link_node(n, parent, link);
+	rb_insert_color_cached(n, (struct rb_root_cached *)root, leftmost);
+	return 0;
 }
 
-__bpf_kfunc void bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
-				bool (less)(struct bpf_rb_node *a, const struct bpf_rb_node *b))
+__bpf_kfunc int bpf_rbtree_add_impl(struct bpf_rb_root *root, struct bpf_rb_node *node,
+				    bool (less)(struct bpf_rb_node *a, const struct bpf_rb_node *b),
+				    void *meta__ign, u64 off)
 {
-	__bpf_rbtree_add(root, node, (void *)less);
+	struct btf_struct_meta *meta = meta__ign;
+
+	return __bpf_rbtree_add(root, node, (void *)less, meta ? meta->record : NULL, off);
 }
 
 __bpf_kfunc struct bpf_rb_node *bpf_rbtree_first(struct bpf_rb_root *root)
@@ -2291,14 +2322,14 @@ BTF_ID_FLAGS(func, crash_kexec, KF_DESTRUCTIVE)
 BTF_ID_FLAGS(func, bpf_obj_new_impl, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_obj_drop_impl, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_refcount_acquire_impl, KF_ACQUIRE)
-BTF_ID_FLAGS(func, bpf_list_push_front)
-BTF_ID_FLAGS(func, bpf_list_push_back)
+BTF_ID_FLAGS(func, bpf_list_push_front_impl)
+BTF_ID_FLAGS(func, bpf_list_push_back_impl)
 BTF_ID_FLAGS(func, bpf_list_pop_front, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_list_pop_back, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_release, KF_RELEASE)
 BTF_ID_FLAGS(func, bpf_rbtree_remove, KF_ACQUIRE)
-BTF_ID_FLAGS(func, bpf_rbtree_add)
+BTF_ID_FLAGS(func, bpf_rbtree_add_impl)
 BTF_ID_FLAGS(func, bpf_rbtree_first, KF_RET_NULL)
 
 #ifdef CONFIG_CGROUPS
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 29e106f7ccaa..736cb7cec0bd 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -8500,10 +8500,10 @@ static int set_rbtree_add_callback_state(struct bpf_verifier_env *env,
 					 struct bpf_func_state *callee,
 					 int insn_idx)
 {
-	/* void bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
+	/* void bpf_rbtree_add_impl(struct bpf_rb_root *root, struct bpf_rb_node *node,
 	 *                     bool (less)(struct bpf_rb_node *a, const struct bpf_rb_node *b));
 	 *
-	 * 'struct bpf_rb_node *node' arg to bpf_rbtree_add is the same PTR_TO_BTF_ID w/ offset
+	 * 'struct bpf_rb_node *node' arg to bpf_rbtree_add_impl is the same PTR_TO_BTF_ID w/ offset
 	 * that 'less' callback args will be receiving. However, 'node' arg was release_reference'd
 	 * by this point, so look at 'root'
 	 */
@@ -9571,8 +9571,8 @@ enum special_kfunc_type {
 	KF_bpf_obj_new_impl,
 	KF_bpf_obj_drop_impl,
 	KF_bpf_refcount_acquire_impl,
-	KF_bpf_list_push_front,
-	KF_bpf_list_push_back,
+	KF_bpf_list_push_front_impl,
+	KF_bpf_list_push_back_impl,
 	KF_bpf_list_pop_front,
 	KF_bpf_list_pop_back,
 	KF_bpf_cast_to_kern_ctx,
@@ -9580,7 +9580,7 @@ enum special_kfunc_type {
 	KF_bpf_rcu_read_lock,
 	KF_bpf_rcu_read_unlock,
 	KF_bpf_rbtree_remove,
-	KF_bpf_rbtree_add,
+	KF_bpf_rbtree_add_impl,
 	KF_bpf_rbtree_first,
 	KF_bpf_dynptr_from_skb,
 	KF_bpf_dynptr_from_xdp,
@@ -9592,14 +9592,14 @@ BTF_SET_START(special_kfunc_set)
 BTF_ID(func, bpf_obj_new_impl)
 BTF_ID(func, bpf_obj_drop_impl)
 BTF_ID(func, bpf_refcount_acquire_impl)
-BTF_ID(func, bpf_list_push_front)
-BTF_ID(func, bpf_list_push_back)
+BTF_ID(func, bpf_list_push_front_impl)
+BTF_ID(func, bpf_list_push_back_impl)
 BTF_ID(func, bpf_list_pop_front)
 BTF_ID(func, bpf_list_pop_back)
 BTF_ID(func, bpf_cast_to_kern_ctx)
 BTF_ID(func, bpf_rdonly_cast)
 BTF_ID(func, bpf_rbtree_remove)
-BTF_ID(func, bpf_rbtree_add)
+BTF_ID(func, bpf_rbtree_add_impl)
 BTF_ID(func, bpf_rbtree_first)
 BTF_ID(func, bpf_dynptr_from_skb)
 BTF_ID(func, bpf_dynptr_from_xdp)
@@ -9611,8 +9611,8 @@ BTF_ID_LIST(special_kfunc_list)
 BTF_ID(func, bpf_obj_new_impl)
 BTF_ID(func, bpf_obj_drop_impl)
 BTF_ID(func, bpf_refcount_acquire_impl)
-BTF_ID(func, bpf_list_push_front)
-BTF_ID(func, bpf_list_push_back)
+BTF_ID(func, bpf_list_push_front_impl)
+BTF_ID(func, bpf_list_push_back_impl)
 BTF_ID(func, bpf_list_pop_front)
 BTF_ID(func, bpf_list_pop_back)
 BTF_ID(func, bpf_cast_to_kern_ctx)
@@ -9620,7 +9620,7 @@ BTF_ID(func, bpf_rdonly_cast)
 BTF_ID(func, bpf_rcu_read_lock)
 BTF_ID(func, bpf_rcu_read_unlock)
 BTF_ID(func, bpf_rbtree_remove)
-BTF_ID(func, bpf_rbtree_add)
+BTF_ID(func, bpf_rbtree_add_impl)
 BTF_ID(func, bpf_rbtree_first)
 BTF_ID(func, bpf_dynptr_from_skb)
 BTF_ID(func, bpf_dynptr_from_xdp)
@@ -9954,15 +9954,15 @@ static int check_reg_allocation_locked(struct bpf_verifier_env *env, struct bpf_
 
 static bool is_bpf_list_api_kfunc(u32 btf_id)
 {
-	return btf_id == special_kfunc_list[KF_bpf_list_push_front] ||
-	       btf_id == special_kfunc_list[KF_bpf_list_push_back] ||
+	return btf_id == special_kfunc_list[KF_bpf_list_push_front_impl] ||
+	       btf_id == special_kfunc_list[KF_bpf_list_push_back_impl] ||
 	       btf_id == special_kfunc_list[KF_bpf_list_pop_front] ||
 	       btf_id == special_kfunc_list[KF_bpf_list_pop_back];
 }
 
 static bool is_bpf_rbtree_api_kfunc(u32 btf_id)
 {
-	return btf_id == special_kfunc_list[KF_bpf_rbtree_add] ||
+	return btf_id == special_kfunc_list[KF_bpf_rbtree_add_impl] ||
 	       btf_id == special_kfunc_list[KF_bpf_rbtree_remove] ||
 	       btf_id == special_kfunc_list[KF_bpf_rbtree_first];
 }
@@ -9975,7 +9975,7 @@ static bool is_bpf_graph_api_kfunc(u32 btf_id)
 
 static bool is_callback_calling_kfunc(u32 btf_id)
 {
-	return btf_id == special_kfunc_list[KF_bpf_rbtree_add];
+	return btf_id == special_kfunc_list[KF_bpf_rbtree_add_impl];
 }
 
 static bool is_rbtree_lock_required_kfunc(u32 btf_id)
@@ -10016,12 +10016,12 @@ static bool check_kfunc_is_graph_node_api(struct bpf_verifier_env *env,
 
 	switch (node_field_type) {
 	case BPF_LIST_NODE:
-		ret = (kfunc_btf_id == special_kfunc_list[KF_bpf_list_push_front] ||
-		       kfunc_btf_id == special_kfunc_list[KF_bpf_list_push_back]);
+		ret = (kfunc_btf_id == special_kfunc_list[KF_bpf_list_push_front_impl] ||
+		       kfunc_btf_id == special_kfunc_list[KF_bpf_list_push_back_impl]);
 		break;
 	case BPF_RB_NODE:
 		ret = (kfunc_btf_id == special_kfunc_list[KF_bpf_rbtree_remove] ||
-		       kfunc_btf_id == special_kfunc_list[KF_bpf_rbtree_add]);
+		       kfunc_btf_id == special_kfunc_list[KF_bpf_rbtree_add_impl]);
 		break;
 	default:
 		verbose(env, "verifier internal error: unexpected graph node argument type %s\n",
@@ -10702,10 +10702,11 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		}
 	}
 
-	if (meta.func_id == special_kfunc_list[KF_bpf_list_push_front] ||
-	    meta.func_id == special_kfunc_list[KF_bpf_list_push_back] ||
-	    meta.func_id == special_kfunc_list[KF_bpf_rbtree_add]) {
+	if (meta.func_id == special_kfunc_list[KF_bpf_list_push_front_impl] ||
+	    meta.func_id == special_kfunc_list[KF_bpf_list_push_back_impl] ||
+	    meta.func_id == special_kfunc_list[KF_bpf_rbtree_add_impl]) {
 		release_ref_obj_id = regs[BPF_REG_2].ref_obj_id;
+		insn_aux->insert_off = regs[BPF_REG_2].off;
 		err = ref_convert_owning_non_owning(env, release_ref_obj_id);
 		if (err) {
 			verbose(env, "kfunc %s#%d conversion of owning ref to non-owning failed\n",
@@ -10721,7 +10722,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		}
 	}
 
-	if (meta.func_id == special_kfunc_list[KF_bpf_rbtree_add]) {
+	if (meta.func_id == special_kfunc_list[KF_bpf_rbtree_add_impl]) {
 		err = __check_func_call(env, insn, insn_idx_p, meta.subprogno,
 					set_rbtree_add_callback_state);
 		if (err) {
@@ -14764,7 +14765,7 @@ static bool regs_exact(const struct bpf_reg_state *rold,
 		       const struct bpf_reg_state *rcur,
 		       struct bpf_id_pair *idmap)
 {
-	return memcmp(rold, rcur, offsetof(struct bpf_reg_state, id)) == 0 && 
+	return memcmp(rold, rcur, offsetof(struct bpf_reg_state, id)) == 0 &&
 	       check_ids(rold->id, rcur->id, idmap) &&
 	       check_ids(rold->ref_obj_id, rcur->ref_obj_id, idmap);
 }
@@ -17407,6 +17408,23 @@ static void specialize_kfunc(struct bpf_verifier_env *env,
 	}
 }
 
+static void __fixup_collection_insert_kfunc(struct bpf_insn_aux_data *insn_aux,
+					    u16 struct_meta_reg,
+					    u16 node_offset_reg,
+					    struct bpf_insn *insn,
+					    struct bpf_insn *insn_buf,
+					    int *cnt)
+{
+	struct btf_struct_meta *kptr_struct_meta = insn_aux->kptr_struct_meta;
+	struct bpf_insn addr[2] = { BPF_LD_IMM64(struct_meta_reg, (long)kptr_struct_meta) };
+
+	insn_buf[0] = addr[0];
+	insn_buf[1] = addr[1];
+	insn_buf[2] = BPF_MOV64_IMM(node_offset_reg, insn_aux->insert_off);
+	insn_buf[3] = *insn;
+	*cnt = 4;
+}
+
 static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			    struct bpf_insn *insn_buf, int insn_idx, int *cnt)
 {
@@ -17453,6 +17471,20 @@ static int fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 		insn_buf[1] = addr[1];
 		insn_buf[2] = *insn;
 		*cnt = 3;
+	} else if (desc->func_id == special_kfunc_list[KF_bpf_list_push_back_impl] ||
+		   desc->func_id == special_kfunc_list[KF_bpf_list_push_front_impl] ||
+		   desc->func_id == special_kfunc_list[KF_bpf_rbtree_add_impl]) {
+		int struct_meta_reg = BPF_REG_3;
+		int node_offset_reg = BPF_REG_4;
+
+		/* rbtree_add has extra 'less' arg, so args-to-fixup are in diff regs */
+		if (desc->func_id == special_kfunc_list[KF_bpf_rbtree_add_impl]) {
+			struct_meta_reg = BPF_REG_4;
+			node_offset_reg = BPF_REG_5;
+		}
+
+		__fixup_collection_insert_kfunc(&env->insn_aux_data[insn_idx], struct_meta_reg,
+						node_offset_reg, insn, insn_buf, cnt);
 	} else if (desc->func_id == special_kfunc_list[KF_bpf_cast_to_kern_ctx] ||
 		   desc->func_id == special_kfunc_list[KF_bpf_rdonly_cast]) {
 		insn_buf[0] = BPF_MOV64_REG(BPF_REG_0, BPF_REG_1);
diff --git a/tools/testing/selftests/bpf/bpf_experimental.h b/tools/testing/selftests/bpf/bpf_experimental.h
index 619afcab2ab0..209811b1993a 100644
--- a/tools/testing/selftests/bpf/bpf_experimental.h
+++ b/tools/testing/selftests/bpf/bpf_experimental.h
@@ -14,7 +14,8 @@
  *	type ID of a struct in program BTF.
  *
  *	The 'local_type_id' parameter must be a known constant.
- *	The 'meta' parameter is a hidden argument that is ignored.
+ *	The 'meta' parameter is rewritten by the verifier, no need for BPF
+ *	program to set it.
  * Returns
  *	A pointer to an object of the type corresponding to the passed in
  *	'local_type_id', or NULL on failure.
@@ -28,7 +29,8 @@ extern void *bpf_obj_new_impl(__u64 local_type_id, void *meta) __ksym;
  *	Free an allocated object. All fields of the object that require
  *	destruction will be destructed before the storage is freed.
  *
- *	The 'meta' parameter is a hidden argument that is ignored.
+ *	The 'meta' parameter is rewritten by the verifier, no need for BPF
+ *	program to set it.
  * Returns
  *	Void.
  */
@@ -41,7 +43,8 @@ extern void bpf_obj_drop_impl(void *kptr, void *meta) __ksym;
  *	Increment the refcount on a refcounted local kptr, turning the
  *	non-owning reference input into an owning reference in the process.
  *
- *	The 'meta' parameter is a hidden argument that is ignored.
+ *	The 'meta' parameter is rewritten by the verifier, no need for BPF
+ *	program to set it.
  * Returns
  *	An owning reference to the object pointed to by 'kptr'
  */
@@ -52,17 +55,35 @@ extern void *bpf_refcount_acquire_impl(void *kptr, void *meta) __ksym;
 
 /* Description
  *	Add a new entry to the beginning of the BPF linked list.
+ *
+ *	The 'meta' and 'off' parameters are rewritten by the verifier, no need
+ *	for BPF programs to set them
  * Returns
- *	Void.
+ *	0 if the node was successfully added
+ *	-EINVAL if the node wasn't added because it's already in a list
  */
-extern void bpf_list_push_front(struct bpf_list_head *head, struct bpf_list_node *node) __ksym;
+extern int bpf_list_push_front_impl(struct bpf_list_head *head,
+				    struct bpf_list_node *node,
+				    void *meta, __u64 off) __ksym;
+
+/* Convenience macro to wrap over bpf_list_push_front_impl */
+#define bpf_list_push_front(head, node) bpf_list_push_front_impl(head, node, NULL, 0)
 
 /* Description
  *	Add a new entry to the end of the BPF linked list.
+ *
+ *	The 'meta' and 'off' parameters are rewritten by the verifier, no need
+ *	for BPF programs to set them
  * Returns
- *	Void.
+ *	0 if the node was successfully added
+ *	-EINVAL if the node wasn't added because it's already in a list
  */
-extern void bpf_list_push_back(struct bpf_list_head *head, struct bpf_list_node *node) __ksym;
+extern int bpf_list_push_back_impl(struct bpf_list_head *head,
+				   struct bpf_list_node *node,
+				   void *meta, __u64 off) __ksym;
+
+/* Convenience macro to wrap over bpf_list_push_back_impl */
+#define bpf_list_push_back(head, node) bpf_list_push_back_impl(head, node, NULL, 0)
 
 /* Description
  *	Remove the entry at the beginning of the BPF linked list.
@@ -88,11 +109,19 @@ extern struct bpf_rb_node *bpf_rbtree_remove(struct bpf_rb_root *root,
 
 /* Description
  *	Add 'node' to rbtree with root 'root' using comparator 'less'
+ *
+ *	The 'meta' and 'off' parameters are rewritten by the verifier, no need
+ *	for BPF programs to set them
  * Returns
- *	Nothing
+ *	0 if the node was successfully added
+ *	-EINVAL if the node wasn't added because it's already in a tree
  */
-extern void bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
-			   bool (less)(struct bpf_rb_node *a, const struct bpf_rb_node *b)) __ksym;
+extern int bpf_rbtree_add_impl(struct bpf_rb_root *root, struct bpf_rb_node *node,
+			       bool (less)(struct bpf_rb_node *a, const struct bpf_rb_node *b),
+			       void *meta, __u64 off) __ksym;
+
+/* Convenience macro to wrap over bpf_rbtree_add_impl */
+#define bpf_rbtree_add(head, node, less) bpf_rbtree_add_impl(head, node, less, NULL, 0)
 
 /* Description
  *	Return the first (leftmost) node in input tree
-- 
cgit v1.2.3-70-g09d2


From de67ba3968fa1455e8020b21e5ccc2bb48b9a852 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:08 -0700
Subject: selftests/bpf: Modify linked_list tests to work with macro-ified
 inserts

The linked_list tests use macros and function pointers to reduce code
duplication. Earlier in the series, bpf_list_push_{front,back} were
modified to be macros, expanding to invoke actual kfuncs
bpf_list_push_{front,back}_impl. Due to this change, a code snippet
like:

  void (*p)(void *, void *) = (void *)&bpf_list_##op;
  p(hexpr, nexpr);

meant to do bpf_list_push_{front,back}(hexpr, nexpr), will no longer
work as it's no longer valid to do &bpf_list_push_{front,back} since
they're no longer functions.

This patch fixes issues of this type, along with two other minor changes
- one improvement and one fix - both related to the node argument to
list_push_{front,back}.

  * The fix: migration of list_push tests away from (void *, void *)
    func ptr uncovered that some tests were incorrectly passing pointer
    to node, not pointer to struct bpf_list_node within the node. This
    patch fixes such issues (CHECK(..., f) -> CHECK(..., &f->node))

  * The improvement: In linked_list tests, the struct foo type has two
    list_node fields: node and node2, at byte offsets 0 and 40 within
    the struct, respectively. Currently node is used in ~all tests
    involving struct foo and lists. The verifier needs to do some work
    to account for the offset of bpf_list_node within the node type, so
    using node2 instead of node exercises that logic more in the tests.
    This patch migrates linked_list tests to use node2 instead of node.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-7-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/linked_list.c |  6 +-
 tools/testing/selftests/bpf/progs/linked_list.c    | 34 ++++----
 tools/testing/selftests/bpf/progs/linked_list.h    |  4 +-
 .../testing/selftests/bpf/progs/linked_list_fail.c | 96 ++++++++++++----------
 4 files changed, 73 insertions(+), 67 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/linked_list.c b/tools/testing/selftests/bpf/prog_tests/linked_list.c
index 0ed8132ce1c3..872e4bd500fd 100644
--- a/tools/testing/selftests/bpf/prog_tests/linked_list.c
+++ b/tools/testing/selftests/bpf/prog_tests/linked_list.c
@@ -84,11 +84,11 @@ static struct {
 	{ "double_push_back", "arg#1 expected pointer to allocated object" },
 	{ "no_node_value_type", "bpf_list_node not found at offset=0" },
 	{ "incorrect_value_type",
-	  "operation on bpf_list_head expects arg#1 bpf_list_node at offset=0 in struct foo, "
+	  "operation on bpf_list_head expects arg#1 bpf_list_node at offset=40 in struct foo, "
 	  "but arg is at offset=0 in struct bar" },
 	{ "incorrect_node_var_off", "variable ptr_ access var_off=(0x0; 0xffffffff) disallowed" },
-	{ "incorrect_node_off1", "bpf_list_node not found at offset=1" },
-	{ "incorrect_node_off2", "arg#1 offset=40, but expected bpf_list_node at offset=0 in struct foo" },
+	{ "incorrect_node_off1", "bpf_list_node not found at offset=41" },
+	{ "incorrect_node_off2", "arg#1 offset=0, but expected bpf_list_node at offset=40 in struct foo" },
 	{ "no_head_type", "bpf_list_head not found at offset=0" },
 	{ "incorrect_head_var_off1", "R1 doesn't have constant offset" },
 	{ "incorrect_head_var_off2", "variable ptr_ access var_off=(0x0; 0xffffffff) disallowed" },
diff --git a/tools/testing/selftests/bpf/progs/linked_list.c b/tools/testing/selftests/bpf/progs/linked_list.c
index 53ded51a3abb..57440a554304 100644
--- a/tools/testing/selftests/bpf/progs/linked_list.c
+++ b/tools/testing/selftests/bpf/progs/linked_list.c
@@ -25,7 +25,7 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 	n = bpf_list_pop_front(head);
 	bpf_spin_unlock(lock);
 	if (n) {
-		bpf_obj_drop(container_of(n, struct foo, node));
+		bpf_obj_drop(container_of(n, struct foo, node2));
 		bpf_obj_drop(f);
 		return 3;
 	}
@@ -34,7 +34,7 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 	n = bpf_list_pop_back(head);
 	bpf_spin_unlock(lock);
 	if (n) {
-		bpf_obj_drop(container_of(n, struct foo, node));
+		bpf_obj_drop(container_of(n, struct foo, node2));
 		bpf_obj_drop(f);
 		return 4;
 	}
@@ -42,7 +42,7 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 
 	bpf_spin_lock(lock);
 	f->data = 42;
-	bpf_list_push_front(head, &f->node);
+	bpf_list_push_front(head, &f->node2);
 	bpf_spin_unlock(lock);
 	if (leave_in_map)
 		return 0;
@@ -51,7 +51,7 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 	bpf_spin_unlock(lock);
 	if (!n)
 		return 5;
-	f = container_of(n, struct foo, node);
+	f = container_of(n, struct foo, node2);
 	if (f->data != 42) {
 		bpf_obj_drop(f);
 		return 6;
@@ -59,14 +59,14 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 
 	bpf_spin_lock(lock);
 	f->data = 13;
-	bpf_list_push_front(head, &f->node);
+	bpf_list_push_front(head, &f->node2);
 	bpf_spin_unlock(lock);
 	bpf_spin_lock(lock);
 	n = bpf_list_pop_front(head);
 	bpf_spin_unlock(lock);
 	if (!n)
 		return 7;
-	f = container_of(n, struct foo, node);
+	f = container_of(n, struct foo, node2);
 	if (f->data != 13) {
 		bpf_obj_drop(f);
 		return 8;
@@ -77,7 +77,7 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 	n = bpf_list_pop_front(head);
 	bpf_spin_unlock(lock);
 	if (n) {
-		bpf_obj_drop(container_of(n, struct foo, node));
+		bpf_obj_drop(container_of(n, struct foo, node2));
 		return 9;
 	}
 
@@ -85,7 +85,7 @@ int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool l
 	n = bpf_list_pop_back(head);
 	bpf_spin_unlock(lock);
 	if (n) {
-		bpf_obj_drop(container_of(n, struct foo, node));
+		bpf_obj_drop(container_of(n, struct foo, node2));
 		return 10;
 	}
 	return 0;
@@ -119,8 +119,8 @@ int list_push_pop_multiple(struct bpf_spin_lock *lock, struct bpf_list_head *hea
 		f[i + 1]->data = i + 1;
 
 		bpf_spin_lock(lock);
-		bpf_list_push_front(head, &f[i]->node);
-		bpf_list_push_front(head, &f[i + 1]->node);
+		bpf_list_push_front(head, &f[i]->node2);
+		bpf_list_push_front(head, &f[i + 1]->node2);
 		bpf_spin_unlock(lock);
 	}
 
@@ -130,13 +130,13 @@ int list_push_pop_multiple(struct bpf_spin_lock *lock, struct bpf_list_head *hea
 		bpf_spin_unlock(lock);
 		if (!n)
 			return 3;
-		pf = container_of(n, struct foo, node);
+		pf = container_of(n, struct foo, node2);
 		if (pf->data != (ARRAY_SIZE(f) - i - 1)) {
 			bpf_obj_drop(pf);
 			return 4;
 		}
 		bpf_spin_lock(lock);
-		bpf_list_push_back(head, &pf->node);
+		bpf_list_push_back(head, &pf->node2);
 		bpf_spin_unlock(lock);
 	}
 
@@ -149,7 +149,7 @@ int list_push_pop_multiple(struct bpf_spin_lock *lock, struct bpf_list_head *hea
 		bpf_spin_unlock(lock);
 		if (!n)
 			return 5;
-		pf = container_of(n, struct foo, node);
+		pf = container_of(n, struct foo, node2);
 		if (pf->data != i) {
 			bpf_obj_drop(pf);
 			return 6;
@@ -160,7 +160,7 @@ int list_push_pop_multiple(struct bpf_spin_lock *lock, struct bpf_list_head *hea
 	n = bpf_list_pop_back(head);
 	bpf_spin_unlock(lock);
 	if (n) {
-		bpf_obj_drop(container_of(n, struct foo, node));
+		bpf_obj_drop(container_of(n, struct foo, node2));
 		return 7;
 	}
 
@@ -168,7 +168,7 @@ int list_push_pop_multiple(struct bpf_spin_lock *lock, struct bpf_list_head *hea
 	n = bpf_list_pop_front(head);
 	bpf_spin_unlock(lock);
 	if (n) {
-		bpf_obj_drop(container_of(n, struct foo, node));
+		bpf_obj_drop(container_of(n, struct foo, node2));
 		return 8;
 	}
 	return 0;
@@ -199,7 +199,7 @@ int list_in_list(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool le
 
 	bpf_spin_lock(lock);
 	f->data = 42;
-	bpf_list_push_front(head, &f->node);
+	bpf_list_push_front(head, &f->node2);
 	bpf_spin_unlock(lock);
 
 	if (leave_in_map)
@@ -210,7 +210,7 @@ int list_in_list(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool le
 	bpf_spin_unlock(lock);
 	if (!n)
 		return 4;
-	f = container_of(n, struct foo, node);
+	f = container_of(n, struct foo, node2);
 	if (f->data != 42) {
 		bpf_obj_drop(f);
 		return 5;
diff --git a/tools/testing/selftests/bpf/progs/linked_list.h b/tools/testing/selftests/bpf/progs/linked_list.h
index 3fb2412552fc..c0f3609a7ffa 100644
--- a/tools/testing/selftests/bpf/progs/linked_list.h
+++ b/tools/testing/selftests/bpf/progs/linked_list.h
@@ -22,7 +22,7 @@ struct foo {
 struct map_value {
 	struct bpf_spin_lock lock;
 	int data;
-	struct bpf_list_head head __contains(foo, node);
+	struct bpf_list_head head __contains(foo, node2);
 };
 
 struct array_map {
@@ -50,7 +50,7 @@ struct {
 #define private(name) SEC(".bss." #name) __hidden __attribute__((aligned(8)))
 
 private(A) struct bpf_spin_lock glock;
-private(A) struct bpf_list_head ghead __contains(foo, node);
+private(A) struct bpf_list_head ghead __contains(foo, node2);
 private(B) struct bpf_spin_lock glock2;
 
 #endif
diff --git a/tools/testing/selftests/bpf/progs/linked_list_fail.c b/tools/testing/selftests/bpf/progs/linked_list_fail.c
index 41978b46f58e..f4c63daba229 100644
--- a/tools/testing/selftests/bpf/progs/linked_list_fail.c
+++ b/tools/testing/selftests/bpf/progs/linked_list_fail.c
@@ -73,22 +73,21 @@ CHECK(inner_map, pop_back, &iv->head);
 	int test##_missing_lock_##op(void *ctx)				\
 	{								\
 		INIT;							\
-		void (*p)(void *, void *) = (void *)&bpf_list_##op;	\
-		p(hexpr, nexpr);					\
+		bpf_list_##op(hexpr, nexpr);				\
 		return 0;						\
 	}
 
-CHECK(kptr, push_front, &f->head, b);
-CHECK(kptr, push_back, &f->head, b);
+CHECK(kptr, push_front, &f->head, &b->node);
+CHECK(kptr, push_back, &f->head, &b->node);
 
-CHECK(global, push_front, &ghead, f);
-CHECK(global, push_back, &ghead, f);
+CHECK(global, push_front, &ghead, &f->node2);
+CHECK(global, push_back, &ghead, &f->node2);
 
-CHECK(map, push_front, &v->head, f);
-CHECK(map, push_back, &v->head, f);
+CHECK(map, push_front, &v->head, &f->node2);
+CHECK(map, push_back, &v->head, &f->node2);
 
-CHECK(inner_map, push_front, &iv->head, f);
-CHECK(inner_map, push_back, &iv->head, f);
+CHECK(inner_map, push_front, &iv->head, &f->node2);
+CHECK(inner_map, push_back, &iv->head, &f->node2);
 
 #undef CHECK
 
@@ -135,32 +134,31 @@ CHECK_OP(pop_back);
 	int test##_incorrect_lock_##op(void *ctx)			\
 	{								\
 		INIT;							\
-		void (*p)(void *, void*) = (void *)&bpf_list_##op;	\
 		bpf_spin_lock(lexpr);					\
-		p(hexpr, nexpr);					\
+		bpf_list_##op(hexpr, nexpr);				\
 		return 0;						\
 	}
 
 #define CHECK_OP(op)							\
-	CHECK(kptr_kptr, op, &f1->lock, &f2->head, b);			\
-	CHECK(kptr_global, op, &f1->lock, &ghead, f);			\
-	CHECK(kptr_map, op, &f1->lock, &v->head, f);			\
-	CHECK(kptr_inner_map, op, &f1->lock, &iv->head, f);		\
+	CHECK(kptr_kptr, op, &f1->lock, &f2->head, &b->node);		\
+	CHECK(kptr_global, op, &f1->lock, &ghead, &f->node2);		\
+	CHECK(kptr_map, op, &f1->lock, &v->head, &f->node2);		\
+	CHECK(kptr_inner_map, op, &f1->lock, &iv->head, &f->node2);	\
 									\
-	CHECK(global_global, op, &glock2, &ghead, f);			\
-	CHECK(global_kptr, op, &glock, &f1->head, b);			\
-	CHECK(global_map, op, &glock, &v->head, f);			\
-	CHECK(global_inner_map, op, &glock, &iv->head, f);		\
+	CHECK(global_global, op, &glock2, &ghead, &f->node2);		\
+	CHECK(global_kptr, op, &glock, &f1->head, &b->node);		\
+	CHECK(global_map, op, &glock, &v->head, &f->node2);		\
+	CHECK(global_inner_map, op, &glock, &iv->head, &f->node2);	\
 									\
-	CHECK(map_map, op, &v->lock, &v2->head, f);			\
-	CHECK(map_kptr, op, &v->lock, &f2->head, b);			\
-	CHECK(map_global, op, &v->lock, &ghead, f);			\
-	CHECK(map_inner_map, op, &v->lock, &iv->head, f);		\
+	CHECK(map_map, op, &v->lock, &v2->head, &f->node2);		\
+	CHECK(map_kptr, op, &v->lock, &f2->head, &b->node);		\
+	CHECK(map_global, op, &v->lock, &ghead, &f->node2);		\
+	CHECK(map_inner_map, op, &v->lock, &iv->head, &f->node2);	\
 									\
-	CHECK(inner_map_inner_map, op, &iv->lock, &iv2->head, f);	\
-	CHECK(inner_map_kptr, op, &iv->lock, &f2->head, b);		\
-	CHECK(inner_map_global, op, &iv->lock, &ghead, f);		\
-	CHECK(inner_map_map, op, &iv->lock, &v->head, f);
+	CHECK(inner_map_inner_map, op, &iv->lock, &iv2->head, &f->node2);\
+	CHECK(inner_map_kptr, op, &iv->lock, &f2->head, &b->node);	\
+	CHECK(inner_map_global, op, &iv->lock, &ghead, &f->node2);	\
+	CHECK(inner_map_map, op, &iv->lock, &v->head, &f->node2);
 
 CHECK_OP(push_front);
 CHECK_OP(push_back);
@@ -340,7 +338,7 @@ int direct_read_node(void *ctx)
 	f = bpf_obj_new(typeof(*f));
 	if (!f)
 		return 0;
-	return *(int *)&f->node;
+	return *(int *)&f->node2;
 }
 
 SEC("?tc")
@@ -351,12 +349,12 @@ int direct_write_node(void *ctx)
 	f = bpf_obj_new(typeof(*f));
 	if (!f)
 		return 0;
-	*(int *)&f->node = 0;
+	*(int *)&f->node2 = 0;
 	return 0;
 }
 
 static __always_inline
-int use_after_unlock(void (*op)(void *head, void *node))
+int use_after_unlock(bool push_front)
 {
 	struct foo *f;
 
@@ -365,7 +363,10 @@ int use_after_unlock(void (*op)(void *head, void *node))
 		return 0;
 	bpf_spin_lock(&glock);
 	f->data = 42;
-	op(&ghead, &f->node);
+	if (push_front)
+		bpf_list_push_front(&ghead, &f->node2);
+	else
+		bpf_list_push_back(&ghead, &f->node2);
 	bpf_spin_unlock(&glock);
 
 	return f->data;
@@ -374,17 +375,17 @@ int use_after_unlock(void (*op)(void *head, void *node))
 SEC("?tc")
 int use_after_unlock_push_front(void *ctx)
 {
-	return use_after_unlock((void *)bpf_list_push_front);
+	return use_after_unlock(true);
 }
 
 SEC("?tc")
 int use_after_unlock_push_back(void *ctx)
 {
-	return use_after_unlock((void *)bpf_list_push_back);
+	return use_after_unlock(false);
 }
 
 static __always_inline
-int list_double_add(void (*op)(void *head, void *node))
+int list_double_add(bool push_front)
 {
 	struct foo *f;
 
@@ -392,8 +393,13 @@ int list_double_add(void (*op)(void *head, void *node))
 	if (!f)
 		return 0;
 	bpf_spin_lock(&glock);
-	op(&ghead, &f->node);
-	op(&ghead, &f->node);
+	if (push_front) {
+		bpf_list_push_front(&ghead, &f->node2);
+		bpf_list_push_front(&ghead, &f->node2);
+	} else {
+		bpf_list_push_back(&ghead, &f->node2);
+		bpf_list_push_back(&ghead, &f->node2);
+	}
 	bpf_spin_unlock(&glock);
 
 	return 0;
@@ -402,13 +408,13 @@ int list_double_add(void (*op)(void *head, void *node))
 SEC("?tc")
 int double_push_front(void *ctx)
 {
-	return list_double_add((void *)bpf_list_push_front);
+	return list_double_add(true);
 }
 
 SEC("?tc")
 int double_push_back(void *ctx)
 {
-	return list_double_add((void *)bpf_list_push_back);
+	return list_double_add(false);
 }
 
 SEC("?tc")
@@ -450,7 +456,7 @@ int incorrect_node_var_off(struct __sk_buff *ctx)
 	if (!f)
 		return 0;
 	bpf_spin_lock(&glock);
-	bpf_list_push_front(&ghead, (void *)&f->node + ctx->protocol);
+	bpf_list_push_front(&ghead, (void *)&f->node2 + ctx->protocol);
 	bpf_spin_unlock(&glock);
 
 	return 0;
@@ -465,7 +471,7 @@ int incorrect_node_off1(void *ctx)
 	if (!f)
 		return 0;
 	bpf_spin_lock(&glock);
-	bpf_list_push_front(&ghead, (void *)&f->node + 1);
+	bpf_list_push_front(&ghead, (void *)&f->node2 + 1);
 	bpf_spin_unlock(&glock);
 
 	return 0;
@@ -480,7 +486,7 @@ int incorrect_node_off2(void *ctx)
 	if (!f)
 		return 0;
 	bpf_spin_lock(&glock);
-	bpf_list_push_front(&ghead, &f->node2);
+	bpf_list_push_front(&ghead, &f->node);
 	bpf_spin_unlock(&glock);
 
 	return 0;
@@ -510,7 +516,7 @@ int incorrect_head_var_off1(struct __sk_buff *ctx)
 	if (!f)
 		return 0;
 	bpf_spin_lock(&glock);
-	bpf_list_push_front((void *)&ghead + ctx->protocol, &f->node);
+	bpf_list_push_front((void *)&ghead + ctx->protocol, &f->node2);
 	bpf_spin_unlock(&glock);
 
 	return 0;
@@ -525,7 +531,7 @@ int incorrect_head_var_off2(struct __sk_buff *ctx)
 	if (!f)
 		return 0;
 	bpf_spin_lock(&glock);
-	bpf_list_push_front((void *)&f->head + ctx->protocol, &f->node);
+	bpf_list_push_front((void *)&f->head + ctx->protocol, &f->node2);
 	bpf_spin_unlock(&glock);
 
 	return 0;
@@ -563,7 +569,7 @@ int incorrect_head_off2(void *ctx)
 		return 0;
 
 	bpf_spin_lock(&glock);
-	bpf_list_push_front((void *)&ghead + 1, &f->node);
+	bpf_list_push_front((void *)&ghead + 1, &f->node2);
 	bpf_spin_unlock(&glock);
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 404ad75a36fb1a1008e9fe803aa7d0212df9e240 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:09 -0700
Subject: bpf: Migrate bpf_rbtree_remove to possibly fail

This patch modifies bpf_rbtree_remove to account for possible failure
due to the input rb_node already not being in any collection.
The function can now return NULL, and does when the aforementioned
scenario occurs. As before, on successful removal an owning reference to
the removed node is returned.

Adding KF_RET_NULL to bpf_rbtree_remove's kfunc flags - now KF_RET_NULL |
KF_ACQUIRE - provides the desired verifier semantics:

  * retval must be checked for NULL before use
  * if NULL, retval's ref_obj_id is released
  * retval is a "maybe acquired" owning ref, not a non-owning ref,
    so it will live past end of critical section (bpf_spin_unlock), and
    thus can be checked for NULL after the end of the CS

BPF programs must add checks
============================

This does change bpf_rbtree_remove's verifier behavior. BPF program
writers will need to add NULL checks to their programs, but the
resulting UX looks natural:

  bpf_spin_lock(&glock);

  n = bpf_rbtree_first(&ghead);
  if (!n) { /* ... */}
  res = bpf_rbtree_remove(&ghead, &n->node);

  bpf_spin_unlock(&glock);

  if (!res)  /* Newly-added check after this patch */
    return 1;

  n = container_of(res, /* ... */);
  /* Do something else with n */
  bpf_obj_drop(n);
  return 0;

The "if (!res)" check above is the only addition necessary for the above
program to pass verification after this patch.

bpf_rbtree_remove no longer clobbers non-owning refs
====================================================

An issue arises when bpf_rbtree_remove fails, though. Consider this
example:

  struct node_data {
    long key;
    struct bpf_list_node l;
    struct bpf_rb_node r;
    struct bpf_refcount ref;
  };

  long failed_sum;

  void bpf_prog()
  {
    struct node_data *n = bpf_obj_new(/* ... */);
    struct bpf_rb_node *res;
    n->key = 10;

    bpf_spin_lock(&glock);

    bpf_list_push_back(&some_list, &n->l); /* n is now a non-owning ref */
    res = bpf_rbtree_remove(&some_tree, &n->r, /* ... */);
    if (!res)
      failed_sum += n->key;  /* not possible */

    bpf_spin_unlock(&glock);
    /* if (res) { do something useful and drop } ... */
  }

The bpf_rbtree_remove in this example will always fail. Similarly to
bpf_spin_unlock, bpf_rbtree_remove is a non-owning reference
invalidation point. The verifier clobbers all non-owning refs after a
bpf_rbtree_remove call, so the "failed_sum += n->key" line will fail
verification, and in fact there's no good way to get information about
the node which failed to add after the invalidation. This patch removes
non-owning reference invalidation from bpf_rbtree_remove to allow the
above usecase to pass verification. The logic for why this is now
possible is as follows:

Before this series, bpf_rbtree_add couldn't fail and thus assumed that
its input, a non-owning reference, was in the tree. But it's easy to
construct an example where two non-owning references pointing to the same
underlying memory are acquired and passed to rbtree_remove one after
another (see rbtree_api_release_aliasing in
selftests/bpf/progs/rbtree_fail.c).

So it was necessary to clobber non-owning refs to prevent this
case and, more generally, to enforce "non-owning ref is definitely
in some collection" invariant. This series removes that invariant and
the failure / runtime checking added in this patch provide a clean way
to deal with the aliasing issue - just fail to remove.

Because the aliasing issue prevented by clobbering non-owning refs is no
longer an issue, this patch removes the invalidate_non_owning_refs
call from verifier handling of bpf_rbtree_remove. Note that
bpf_spin_unlock - the other caller of invalidate_non_owning_refs -
clobbers non-owning refs for a different reason, so its clobbering
behavior remains unchanged.

No BPF program changes are necessary for programs to remain valid as a
result of this clobbering change. A valid program before this patch
passed verification with its non-owning refs having shorter (or equal)
lifetimes due to more aggressive clobbering.

Also, update existing tests to check bpf_rbtree_remove retval for NULL
where necessary, and move rbtree_api_release_aliasing from
progs/rbtree_fail.c to progs/rbtree.c since it's now expected to pass
verification.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-8-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c                                   | 21 +----
 kernel/bpf/helpers.c                               |  8 +-
 kernel/bpf/verifier.c                              |  3 -
 .../testing/selftests/bpf/prog_tests/linked_list.c | 90 ++++++++++++++--------
 tools/testing/selftests/bpf/prog_tests/rbtree.c    | 25 ++++++
 tools/testing/selftests/bpf/progs/rbtree.c         | 74 +++++++++++++++++-
 tools/testing/selftests/bpf/progs/rbtree_fail.c    | 77 +++++++-----------
 7 files changed, 191 insertions(+), 107 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 14889fd5ba8e..027f9f8a3551 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3805,25 +3805,8 @@ struct btf_record *btf_parse_fields(const struct btf *btf, const struct btf_type
 		goto end;
 	}
 
-	/* need collection identity for non-owning refs before allowing this
-	 *
-	 * Consider a node type w/ both list and rb_node fields:
-	 *   struct node {
-	 *     struct bpf_list_node l;
-	 *     struct bpf_rb_node r;
-	 *   }
-	 *
-	 * Used like so:
-	 *   struct node *n = bpf_obj_new(....);
-	 *   bpf_list_push_front(&list_head, &n->l);
-	 *   bpf_rbtree_remove(&rb_root, &n->r);
-	 *
-	 * It should not be possible to rbtree_remove the node since it hasn't
-	 * been added to a tree. But push_front converts n to a non-owning
-	 * reference, and rbtree_remove accepts the non-owning reference to
-	 * a type w/ bpf_rb_node field.
-	 */
-	if (btf_record_has_field(rec, BPF_LIST_NODE) &&
+	if (rec->refcount_off < 0 &&
+	    btf_record_has_field(rec, BPF_LIST_NODE) &&
 	    btf_record_has_field(rec, BPF_RB_NODE)) {
 		ret = -EINVAL;
 		goto end;
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 5067f8d46872..1835df333287 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2000,6 +2000,12 @@ __bpf_kfunc struct bpf_rb_node *bpf_rbtree_remove(struct bpf_rb_root *root,
 	struct rb_root_cached *r = (struct rb_root_cached *)root;
 	struct rb_node *n = (struct rb_node *)node;
 
+	if (!n->__rb_parent_color)
+		RB_CLEAR_NODE(n);
+
+	if (RB_EMPTY_NODE(n))
+		return NULL;
+
 	rb_erase_cached(n, r);
 	RB_CLEAR_NODE(n);
 	return (struct bpf_rb_node *)n;
@@ -2328,7 +2334,7 @@ BTF_ID_FLAGS(func, bpf_list_pop_front, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_list_pop_back, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_task_release, KF_RELEASE)
-BTF_ID_FLAGS(func, bpf_rbtree_remove, KF_ACQUIRE)
+BTF_ID_FLAGS(func, bpf_rbtree_remove, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_rbtree_add_impl)
 BTF_ID_FLAGS(func, bpf_rbtree_first, KF_RET_NULL)
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 736cb7cec0bd..6a41b69a424e 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -10922,9 +10922,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			ref_set_non_owning(env, &regs[BPF_REG_0]);
 		}
 
-		if (meta.func_id == special_kfunc_list[KF_bpf_rbtree_remove])
-			invalidate_non_owning_refs(env);
-
 		if (reg_may_point_to_spin_lock(&regs[BPF_REG_0]) && !regs[BPF_REG_0].id)
 			regs[BPF_REG_0].id = ++env->id_gen;
 	} else if (btf_type_is_void(t)) {
diff --git a/tools/testing/selftests/bpf/prog_tests/linked_list.c b/tools/testing/selftests/bpf/prog_tests/linked_list.c
index 872e4bd500fd..f63309fd0e28 100644
--- a/tools/testing/selftests/bpf/prog_tests/linked_list.c
+++ b/tools/testing/selftests/bpf/prog_tests/linked_list.c
@@ -266,6 +266,59 @@ end:
 	return NULL;
 }
 
+static void list_and_rb_node_same_struct(bool refcount_field)
+{
+	int bpf_rb_node_btf_id, bpf_refcount_btf_id, foo_btf_id;
+	struct btf *btf;
+	int id, err;
+
+	btf = init_btf();
+	if (!ASSERT_OK_PTR(btf, "init_btf"))
+		return;
+
+	bpf_rb_node_btf_id = btf__add_struct(btf, "bpf_rb_node", 24);
+	if (!ASSERT_GT(bpf_rb_node_btf_id, 0, "btf__add_struct bpf_rb_node"))
+		return;
+
+	if (refcount_field) {
+		bpf_refcount_btf_id = btf__add_struct(btf, "bpf_refcount", 4);
+		if (!ASSERT_GT(bpf_refcount_btf_id, 0, "btf__add_struct bpf_refcount"))
+			return;
+	}
+
+	id = btf__add_struct(btf, "bar", refcount_field ? 44 : 40);
+	if (!ASSERT_GT(id, 0, "btf__add_struct bar"))
+		return;
+	err = btf__add_field(btf, "a", LIST_NODE, 0, 0);
+	if (!ASSERT_OK(err, "btf__add_field bar::a"))
+		return;
+	err = btf__add_field(btf, "c", bpf_rb_node_btf_id, 128, 0);
+	if (!ASSERT_OK(err, "btf__add_field bar::c"))
+		return;
+	if (refcount_field) {
+		err = btf__add_field(btf, "ref", bpf_refcount_btf_id, 320, 0);
+		if (!ASSERT_OK(err, "btf__add_field bar::ref"))
+			return;
+	}
+
+	foo_btf_id = btf__add_struct(btf, "foo", 20);
+	if (!ASSERT_GT(foo_btf_id, 0, "btf__add_struct foo"))
+		return;
+	err = btf__add_field(btf, "a", LIST_HEAD, 0, 0);
+	if (!ASSERT_OK(err, "btf__add_field foo::a"))
+		return;
+	err = btf__add_field(btf, "b", SPIN_LOCK, 128, 0);
+	if (!ASSERT_OK(err, "btf__add_field foo::b"))
+		return;
+	id = btf__add_decl_tag(btf, "contains:bar:a", foo_btf_id, 0);
+	if (!ASSERT_GT(id, 0, "btf__add_decl_tag contains:bar:a"))
+		return;
+
+	err = btf__load_into_kernel(btf);
+	ASSERT_EQ(err, refcount_field ? 0 : -EINVAL, "check btf");
+	btf__free(btf);
+}
+
 static void test_btf(void)
 {
 	struct btf *btf = NULL;
@@ -717,39 +770,12 @@ static void test_btf(void)
 	}
 
 	while (test__start_subtest("btf: list_node and rb_node in same struct")) {
-		btf = init_btf();
-		if (!ASSERT_OK_PTR(btf, "init_btf"))
-			break;
-
-		id = btf__add_struct(btf, "bpf_rb_node", 24);
-		if (!ASSERT_EQ(id, 5, "btf__add_struct bpf_rb_node"))
-			break;
-		id = btf__add_struct(btf, "bar", 40);
-		if (!ASSERT_EQ(id, 6, "btf__add_struct bar"))
-			break;
-		err = btf__add_field(btf, "a", LIST_NODE, 0, 0);
-		if (!ASSERT_OK(err, "btf__add_field bar::a"))
-			break;
-		err = btf__add_field(btf, "c", 5, 128, 0);
-		if (!ASSERT_OK(err, "btf__add_field bar::c"))
-			break;
-
-		id = btf__add_struct(btf, "foo", 20);
-		if (!ASSERT_EQ(id, 7, "btf__add_struct foo"))
-			break;
-		err = btf__add_field(btf, "a", LIST_HEAD, 0, 0);
-		if (!ASSERT_OK(err, "btf__add_field foo::a"))
-			break;
-		err = btf__add_field(btf, "b", SPIN_LOCK, 128, 0);
-		if (!ASSERT_OK(err, "btf__add_field foo::b"))
-			break;
-		id = btf__add_decl_tag(btf, "contains:bar:a", 7, 0);
-		if (!ASSERT_EQ(id, 8, "btf__add_decl_tag contains:bar:a"))
-			break;
+		list_and_rb_node_same_struct(true);
+		break;
+	}
 
-		err = btf__load_into_kernel(btf);
-		ASSERT_EQ(err, -EINVAL, "check btf");
-		btf__free(btf);
+	while (test__start_subtest("btf: list_node and rb_node in same struct, no bpf_refcount")) {
+		list_and_rb_node_same_struct(false);
 		break;
 	}
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/rbtree.c b/tools/testing/selftests/bpf/prog_tests/rbtree.c
index 156fa95c42f6..e9300c96607d 100644
--- a/tools/testing/selftests/bpf/prog_tests/rbtree.c
+++ b/tools/testing/selftests/bpf/prog_tests/rbtree.c
@@ -77,6 +77,29 @@ static void test_rbtree_first_and_remove(void)
 	rbtree__destroy(skel);
 }
 
+static void test_rbtree_api_release_aliasing(void)
+{
+	LIBBPF_OPTS(bpf_test_run_opts, opts,
+		    .data_in = &pkt_v4,
+		    .data_size_in = sizeof(pkt_v4),
+		    .repeat = 1,
+	);
+	struct rbtree *skel;
+	int ret;
+
+	skel = rbtree__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "rbtree__open_and_load"))
+		return;
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.rbtree_api_release_aliasing), &opts);
+	ASSERT_OK(ret, "rbtree_api_release_aliasing");
+	ASSERT_OK(opts.retval, "rbtree_api_release_aliasing retval");
+	ASSERT_EQ(skel->data->first_data[0], 42, "rbtree_api_release_aliasing first rbtree_remove()");
+	ASSERT_EQ(skel->data->first_data[1], -1, "rbtree_api_release_aliasing second rbtree_remove()");
+
+	rbtree__destroy(skel);
+}
+
 void test_rbtree_success(void)
 {
 	if (test__start_subtest("rbtree_add_nodes"))
@@ -85,6 +108,8 @@ void test_rbtree_success(void)
 		test_rbtree_add_and_remove();
 	if (test__start_subtest("rbtree_first_and_remove"))
 		test_rbtree_first_and_remove();
+	if (test__start_subtest("rbtree_api_release_aliasing"))
+		test_rbtree_api_release_aliasing();
 }
 
 #define BTF_FAIL_TEST(suffix)									\
diff --git a/tools/testing/selftests/bpf/progs/rbtree.c b/tools/testing/selftests/bpf/progs/rbtree.c
index 4c90aa6abddd..b09f4fffe57c 100644
--- a/tools/testing/selftests/bpf/progs/rbtree.c
+++ b/tools/testing/selftests/bpf/progs/rbtree.c
@@ -93,9 +93,11 @@ long rbtree_add_and_remove(void *ctx)
 	res = bpf_rbtree_remove(&groot, &n->node);
 	bpf_spin_unlock(&glock);
 
+	if (!res)
+		return 1;
+
 	n = container_of(res, struct node_data, node);
 	removed_key = n->key;
-
 	bpf_obj_drop(n);
 
 	return 0;
@@ -148,9 +150,11 @@ long rbtree_first_and_remove(void *ctx)
 	res = bpf_rbtree_remove(&groot, &o->node);
 	bpf_spin_unlock(&glock);
 
+	if (!res)
+		return 5;
+
 	o = container_of(res, struct node_data, node);
 	removed_key = o->key;
-
 	bpf_obj_drop(o);
 
 	bpf_spin_lock(&glock);
@@ -173,4 +177,70 @@ err_out:
 	return 1;
 }
 
+SEC("tc")
+long rbtree_api_release_aliasing(void *ctx)
+{
+	struct node_data *n, *m, *o;
+	struct bpf_rb_node *res, *res2;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return 1;
+	n->key = 41;
+	n->data = 42;
+
+	bpf_spin_lock(&glock);
+	bpf_rbtree_add(&groot, &n->node, less);
+	bpf_spin_unlock(&glock);
+
+	bpf_spin_lock(&glock);
+
+	/* m and o point to the same node,
+	 * but verifier doesn't know this
+	 */
+	res = bpf_rbtree_first(&groot);
+	if (!res)
+		goto err_out;
+	o = container_of(res, struct node_data, node);
+
+	res = bpf_rbtree_first(&groot);
+	if (!res)
+		goto err_out;
+	m = container_of(res, struct node_data, node);
+
+	res = bpf_rbtree_remove(&groot, &m->node);
+	/* Retval of previous remove returns an owning reference to m,
+	 * which is the same node non-owning ref o is pointing at.
+	 * We can safely try to remove o as the second rbtree_remove will
+	 * return NULL since the node isn't in a tree.
+	 *
+	 * Previously we relied on the verifier type system + rbtree_remove
+	 * invalidating non-owning refs to ensure that rbtree_remove couldn't
+	 * fail, but now rbtree_remove does runtime checking so we no longer
+	 * invalidate non-owning refs after remove.
+	 */
+	res2 = bpf_rbtree_remove(&groot, &o->node);
+
+	bpf_spin_unlock(&glock);
+
+	if (res) {
+		o = container_of(res, struct node_data, node);
+		first_data[0] = o->data;
+		bpf_obj_drop(o);
+	}
+	if (res2) {
+		/* The second remove fails, so res2 is null and this doesn't
+		 * execute
+		 */
+		m = container_of(res2, struct node_data, node);
+		first_data[1] = m->data;
+		bpf_obj_drop(m);
+	}
+	return 0;
+
+err_out:
+	bpf_spin_unlock(&glock);
+	return 1;
+}
+
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/rbtree_fail.c b/tools/testing/selftests/bpf/progs/rbtree_fail.c
index 46d7d18a218f..3fecf1c6dfe5 100644
--- a/tools/testing/selftests/bpf/progs/rbtree_fail.c
+++ b/tools/testing/selftests/bpf/progs/rbtree_fail.c
@@ -105,7 +105,7 @@ long rbtree_api_remove_unadded_node(void *ctx)
 }
 
 SEC("?tc")
-__failure __msg("Unreleased reference id=2 alloc_insn=10")
+__failure __msg("Unreleased reference id=3 alloc_insn=10")
 long rbtree_api_remove_no_drop(void *ctx)
 {
 	struct bpf_rb_node *res;
@@ -118,11 +118,13 @@ long rbtree_api_remove_no_drop(void *ctx)
 
 	res = bpf_rbtree_remove(&groot, res);
 
-	n = container_of(res, struct node_data, node);
-	__sink(n);
+	if (res) {
+		n = container_of(res, struct node_data, node);
+		__sink(n);
+	}
 	bpf_spin_unlock(&glock);
 
-	/* bpf_obj_drop(n) is missing here */
+	/* if (res) { bpf_obj_drop(n); } is missing here */
 	return 0;
 
 unlock_err:
@@ -150,35 +152,36 @@ long rbtree_api_add_to_multiple_trees(void *ctx)
 }
 
 SEC("?tc")
-__failure __msg("rbtree_remove node input must be non-owning ref")
-long rbtree_api_add_release_unlock_escape(void *ctx)
+__failure __msg("dereference of modified ptr_or_null_ ptr R2 off=16 disallowed")
+long rbtree_api_use_unchecked_remove_retval(void *ctx)
 {
-	struct node_data *n;
-
-	n = bpf_obj_new(typeof(*n));
-	if (!n)
-		return 1;
+	struct bpf_rb_node *res;
 
 	bpf_spin_lock(&glock);
-	bpf_rbtree_add(&groot, &n->node, less);
+
+	res = bpf_rbtree_first(&groot);
+	if (!res)
+		goto err_out;
+	res = bpf_rbtree_remove(&groot, res);
+
 	bpf_spin_unlock(&glock);
 
 	bpf_spin_lock(&glock);
-	/* After add() in previous critical section, n should be
-	 * release_on_unlock and released after previous spin_unlock,
-	 * so should not be possible to use it here
-	 */
-	bpf_rbtree_remove(&groot, &n->node);
+	/* Must check res for NULL before using in rbtree_add below */
+	bpf_rbtree_add(&groot, res, less);
 	bpf_spin_unlock(&glock);
 	return 0;
+
+err_out:
+	bpf_spin_unlock(&glock);
+	return 1;
 }
 
 SEC("?tc")
 __failure __msg("rbtree_remove node input must be non-owning ref")
-long rbtree_api_release_aliasing(void *ctx)
+long rbtree_api_add_release_unlock_escape(void *ctx)
 {
-	struct node_data *n, *m, *o;
-	struct bpf_rb_node *res;
+	struct node_data *n;
 
 	n = bpf_obj_new(typeof(*n));
 	if (!n)
@@ -189,37 +192,11 @@ long rbtree_api_release_aliasing(void *ctx)
 	bpf_spin_unlock(&glock);
 
 	bpf_spin_lock(&glock);
-
-	/* m and o point to the same node,
-	 * but verifier doesn't know this
-	 */
-	res = bpf_rbtree_first(&groot);
-	if (!res)
-		return 1;
-	o = container_of(res, struct node_data, node);
-
-	res = bpf_rbtree_first(&groot);
-	if (!res)
-		return 1;
-	m = container_of(res, struct node_data, node);
-
-	bpf_rbtree_remove(&groot, &m->node);
-	/* This second remove shouldn't be possible. Retval of previous
-	 * remove returns owning reference to m, which is the same
-	 * node o's non-owning ref is pointing at
-	 *
-	 * In order to preserve property
-	 *   * owning ref must not be in rbtree
-	 *   * non-owning ref must be in rbtree
-	 *
-	 * o's ref must be invalidated after previous remove. Otherwise
-	 * we'd have non-owning ref to node that isn't in rbtree, and
-	 * verifier wouldn't be able to use type system to prevent remove
-	 * of ref that already isn't in any tree. Would have to do runtime
-	 * checks in that case.
+	/* After add() in previous critical section, n should be
+	 * release_on_unlock and released after previous spin_unlock,
+	 * so should not be possible to use it here
 	 */
-	bpf_rbtree_remove(&groot, &o->node);
-
+	bpf_rbtree_remove(&groot, &n->node);
 	bpf_spin_unlock(&glock);
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 3e81740a90626024a9d9c6f9bfa3d36204dafefb Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:10 -0700
Subject: bpf: Centralize btf_field-specific initialization logic

All btf_fields in an object are 0-initialized by memset in
bpf_obj_init. This might not be a valid initial state for some field
types, in which case kfuncs that use the type will properly initialize
their input if it's been 0-initialized. Some BPF graph collection types
and kfuncs do this: bpf_list_{head,node} and bpf_rb_node.

An earlier patch in this series added the bpf_refcount field, for which
the 0 state indicates that the refcounted object should be free'd.
bpf_obj_init treats this field specially, setting refcount to 1 instead
of relying on scattered "refcount is 0? Must have just been initialized,
let's set to 1" logic in kfuncs.

This patch extends this treatment to list and rbtree field types,
allowing most scattered initialization logic in kfuncs to be removed.

Note that bpf_{list_head,rb_root} may be inside a BPF map, in which case
they'll be 0-initialized without passing through the newly-added logic,
so scattered initialization logic must remain for these collection root
types.

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-9-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h  | 33 +++++++++++++++++++++++++++++----
 kernel/bpf/helpers.c | 14 ++++++--------
 2 files changed, 35 insertions(+), 12 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index b065de2cfcea..18b592fde896 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -355,6 +355,34 @@ static inline u32 btf_field_type_align(enum btf_field_type type)
 	}
 }
 
+static inline void bpf_obj_init_field(const struct btf_field *field, void *addr)
+{
+	memset(addr, 0, field->size);
+
+	switch (field->type) {
+	case BPF_REFCOUNT:
+		refcount_set((refcount_t *)addr, 1);
+		break;
+	case BPF_RB_NODE:
+		RB_CLEAR_NODE((struct rb_node *)addr);
+		break;
+	case BPF_LIST_HEAD:
+	case BPF_LIST_NODE:
+		INIT_LIST_HEAD((struct list_head *)addr);
+		break;
+	case BPF_RB_ROOT:
+		/* RB_ROOT_CACHED 0-inits, no need to do anything after memset */
+	case BPF_SPIN_LOCK:
+	case BPF_TIMER:
+	case BPF_KPTR_UNREF:
+	case BPF_KPTR_REF:
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		return;
+	}
+}
+
 static inline bool btf_record_has_field(const struct btf_record *rec, enum btf_field_type type)
 {
 	if (IS_ERR_OR_NULL(rec))
@@ -369,10 +397,7 @@ static inline void bpf_obj_init(const struct btf_record *rec, void *obj)
 	if (IS_ERR_OR_NULL(rec))
 		return;
 	for (i = 0; i < rec->cnt; i++)
-		memset(obj + rec->fields[i].offset, 0, rec->fields[i].size);
-
-	if (rec->refcount_off >= 0)
-		refcount_set((refcount_t *)(obj + rec->refcount_off), 1);
+		bpf_obj_init_field(&rec->fields[i], obj + rec->fields[i].offset);
 }
 
 /* 'dst' must be a temporary buffer and should not point to memory that is being
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 1835df333287..00e5fb0682ac 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1936,10 +1936,11 @@ static int __bpf_list_add(struct bpf_list_node *node, struct bpf_list_head *head
 {
 	struct list_head *n = (void *)node, *h = (void *)head;
 
+	/* If list_head was 0-initialized by map, bpf_obj_init_field wasn't
+	 * called on its fields, so init here
+	 */
 	if (unlikely(!h->next))
 		INIT_LIST_HEAD(h);
-	if (unlikely(!n->next))
-		INIT_LIST_HEAD(n);
 	if (!list_empty(n)) {
 		/* Only called from BPF prog, no need to migrate_disable */
 		__bpf_obj_drop_impl(n - off, rec);
@@ -1975,6 +1976,9 @@ static struct bpf_list_node *__bpf_list_del(struct bpf_list_head *head, bool tai
 {
 	struct list_head *n, *h = (void *)head;
 
+	/* If list_head was 0-initialized by map, bpf_obj_init_field wasn't
+	 * called on its fields, so init here
+	 */
 	if (unlikely(!h->next))
 		INIT_LIST_HEAD(h);
 	if (list_empty(h))
@@ -2000,9 +2004,6 @@ __bpf_kfunc struct bpf_rb_node *bpf_rbtree_remove(struct bpf_rb_root *root,
 	struct rb_root_cached *r = (struct rb_root_cached *)root;
 	struct rb_node *n = (struct rb_node *)node;
 
-	if (!n->__rb_parent_color)
-		RB_CLEAR_NODE(n);
-
 	if (RB_EMPTY_NODE(n))
 		return NULL;
 
@@ -2022,9 +2023,6 @@ static int __bpf_rbtree_add(struct bpf_rb_root *root, struct bpf_rb_node *node,
 	bpf_callback_t cb = (bpf_callback_t)less;
 	bool leftmost = true;
 
-	if (!n->__rb_parent_color)
-		RB_CLEAR_NODE(n);
-
 	if (!RB_EMPTY_NODE(n)) {
 		/* Only called from BPF prog, no need to migrate_disable */
 		__bpf_obj_drop_impl(n - off, rec);
-- 
cgit v1.2.3-70-g09d2


From 6147f15131e2df544a5449815f456da48c0c88e7 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Sat, 15 Apr 2023 13:18:11 -0700
Subject: selftests/bpf: Add refcounted_kptr tests

Test refcounted local kptr functionality added in previous patches in
the series.

Usecases which pass verification:

* Add refcounted local kptr to both tree and list. Then, read and -
  possibly, depending on test variant - delete from tree, then list.
  * Also test doing read-and-maybe-delete in opposite order
* Stash a refcounted local kptr in a map_value, then add it to a
  rbtree. Read from both, possibly deleting after tree read.
* Add refcounted local kptr to both tree and list. Then, try reading and
  deleting twice from one of the collections.
* bpf_refcount_acquire of just-added non-owning ref should work, as
  should bpf_refcount_acquire of owning ref just out of bpf_obj_new

Usecases which fail verification:

* The simple successful bpf_refcount_acquire cases from above should
  both fail to verify if the newly-acquired owning ref is not dropped

Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230415201811.343116-10-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/prog_tests/refcounted_kptr.c     |  18 +
 .../testing/selftests/bpf/progs/refcounted_kptr.c  | 406 +++++++++++++++++++++
 .../selftests/bpf/progs/refcounted_kptr_fail.c     |  72 ++++
 3 files changed, 496 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c
 create mode 100644 tools/testing/selftests/bpf/progs/refcounted_kptr.c
 create mode 100644 tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c

diff --git a/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c b/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c
new file mode 100644
index 000000000000..2ab23832062d
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c
@@ -0,0 +1,18 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <test_progs.h>
+#include <network_helpers.h>
+
+#include "refcounted_kptr.skel.h"
+#include "refcounted_kptr_fail.skel.h"
+
+void test_refcounted_kptr(void)
+{
+	RUN_TESTS(refcounted_kptr);
+}
+
+void test_refcounted_kptr_fail(void)
+{
+	RUN_TESTS(refcounted_kptr_fail);
+}
diff --git a/tools/testing/selftests/bpf/progs/refcounted_kptr.c b/tools/testing/selftests/bpf/progs/refcounted_kptr.c
new file mode 100644
index 000000000000..1d348a225140
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/refcounted_kptr.c
@@ -0,0 +1,406 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
+
+#include <vmlinux.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
+#include "bpf_misc.h"
+#include "bpf_experimental.h"
+
+struct node_data {
+	long key;
+	long list_data;
+	struct bpf_rb_node r;
+	struct bpf_list_node l;
+	struct bpf_refcount ref;
+};
+
+struct map_value {
+	struct node_data __kptr *node;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} stashed_nodes SEC(".maps");
+
+struct node_acquire {
+	long key;
+	long data;
+	struct bpf_rb_node node;
+	struct bpf_refcount refcount;
+};
+
+#define private(name) SEC(".bss." #name) __hidden __attribute__((aligned(8)))
+private(A) struct bpf_spin_lock lock;
+private(A) struct bpf_rb_root root __contains(node_data, r);
+private(A) struct bpf_list_head head __contains(node_data, l);
+
+private(B) struct bpf_spin_lock alock;
+private(B) struct bpf_rb_root aroot __contains(node_acquire, node);
+
+static bool less(struct bpf_rb_node *node_a, const struct bpf_rb_node *node_b)
+{
+	struct node_data *a;
+	struct node_data *b;
+
+	a = container_of(node_a, struct node_data, r);
+	b = container_of(node_b, struct node_data, r);
+
+	return a->key < b->key;
+}
+
+static bool less_a(struct bpf_rb_node *a, const struct bpf_rb_node *b)
+{
+	struct node_acquire *node_a;
+	struct node_acquire *node_b;
+
+	node_a = container_of(a, struct node_acquire, node);
+	node_b = container_of(b, struct node_acquire, node);
+
+	return node_a->key < node_b->key;
+}
+
+static long __insert_in_tree_and_list(struct bpf_list_head *head,
+				      struct bpf_rb_root *root,
+				      struct bpf_spin_lock *lock)
+{
+	struct node_data *n, *m;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return -1;
+
+	m = bpf_refcount_acquire(n);
+	m->key = 123;
+	m->list_data = 456;
+
+	bpf_spin_lock(lock);
+	if (bpf_rbtree_add(root, &n->r, less)) {
+		/* Failure to insert - unexpected */
+		bpf_spin_unlock(lock);
+		bpf_obj_drop(m);
+		return -2;
+	}
+	bpf_spin_unlock(lock);
+
+	bpf_spin_lock(lock);
+	if (bpf_list_push_front(head, &m->l)) {
+		/* Failure to insert - unexpected */
+		bpf_spin_unlock(lock);
+		return -3;
+	}
+	bpf_spin_unlock(lock);
+	return 0;
+}
+
+static long __stash_map_insert_tree(int idx, int val, struct bpf_rb_root *root,
+				    struct bpf_spin_lock *lock)
+{
+	struct map_value *mapval;
+	struct node_data *n, *m;
+
+	mapval = bpf_map_lookup_elem(&stashed_nodes, &idx);
+	if (!mapval)
+		return -1;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return -2;
+
+	n->key = val;
+	m = bpf_refcount_acquire(n);
+
+	n = bpf_kptr_xchg(&mapval->node, n);
+	if (n) {
+		bpf_obj_drop(n);
+		bpf_obj_drop(m);
+		return -3;
+	}
+
+	bpf_spin_lock(lock);
+	if (bpf_rbtree_add(root, &m->r, less)) {
+		/* Failure to insert - unexpected */
+		bpf_spin_unlock(lock);
+		return -4;
+	}
+	bpf_spin_unlock(lock);
+	return 0;
+}
+
+static long __read_from_tree(struct bpf_rb_root *root,
+			     struct bpf_spin_lock *lock,
+			     bool remove_from_tree)
+{
+	struct bpf_rb_node *rb;
+	struct node_data *n;
+	long res = -99;
+
+	bpf_spin_lock(lock);
+
+	rb = bpf_rbtree_first(root);
+	if (!rb) {
+		bpf_spin_unlock(lock);
+		return -1;
+	}
+
+	n = container_of(rb, struct node_data, r);
+	res = n->key;
+
+	if (!remove_from_tree) {
+		bpf_spin_unlock(lock);
+		return res;
+	}
+
+	rb = bpf_rbtree_remove(root, rb);
+	bpf_spin_unlock(lock);
+	if (!rb)
+		return -2;
+	n = container_of(rb, struct node_data, r);
+	bpf_obj_drop(n);
+	return res;
+}
+
+static long __read_from_list(struct bpf_list_head *head,
+			     struct bpf_spin_lock *lock,
+			     bool remove_from_list)
+{
+	struct bpf_list_node *l;
+	struct node_data *n;
+	long res = -99;
+
+	bpf_spin_lock(lock);
+
+	l = bpf_list_pop_front(head);
+	if (!l) {
+		bpf_spin_unlock(lock);
+		return -1;
+	}
+
+	n = container_of(l, struct node_data, l);
+	res = n->list_data;
+
+	if (!remove_from_list) {
+		if (bpf_list_push_back(head, &n->l)) {
+			bpf_spin_unlock(lock);
+			return -2;
+		}
+	}
+
+	bpf_spin_unlock(lock);
+
+	if (remove_from_list)
+		bpf_obj_drop(n);
+	return res;
+}
+
+static long __read_from_unstash(int idx)
+{
+	struct node_data *n = NULL;
+	struct map_value *mapval;
+	long val = -99;
+
+	mapval = bpf_map_lookup_elem(&stashed_nodes, &idx);
+	if (!mapval)
+		return -1;
+
+	n = bpf_kptr_xchg(&mapval->node, n);
+	if (!n)
+		return -2;
+
+	val = n->key;
+	bpf_obj_drop(n);
+	return val;
+}
+
+#define INSERT_READ_BOTH(rem_tree, rem_list, desc)			\
+SEC("tc")								\
+__description(desc)							\
+__success __retval(579)							\
+long insert_and_remove_tree_##rem_tree##_list_##rem_list(void *ctx)	\
+{									\
+	long err, tree_data, list_data;					\
+									\
+	err = __insert_in_tree_and_list(&head, &root, &lock);		\
+	if (err)							\
+		return err;						\
+									\
+	err = __read_from_tree(&root, &lock, rem_tree);			\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		tree_data = err;					\
+									\
+	err = __read_from_list(&head, &lock, rem_list);			\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		list_data = err;					\
+									\
+	return tree_data + list_data;					\
+}
+
+/* After successful insert of struct node_data into both collections:
+ *   - it should have refcount = 2
+ *   - removing / not removing the node_data from a collection after
+ *     reading should have no effect on ability to read / remove from
+ *     the other collection
+ */
+INSERT_READ_BOTH(true, true, "insert_read_both: remove from tree + list");
+INSERT_READ_BOTH(false, false, "insert_read_both: remove from neither");
+INSERT_READ_BOTH(true, false, "insert_read_both: remove from tree");
+INSERT_READ_BOTH(false, true, "insert_read_both: remove from list");
+
+#undef INSERT_READ_BOTH
+#define INSERT_READ_BOTH(rem_tree, rem_list, desc)			\
+SEC("tc")								\
+__description(desc)							\
+__success __retval(579)							\
+long insert_and_remove_lf_tree_##rem_tree##_list_##rem_list(void *ctx)	\
+{									\
+	long err, tree_data, list_data;					\
+									\
+	err = __insert_in_tree_and_list(&head, &root, &lock);		\
+	if (err)							\
+		return err;						\
+									\
+	err = __read_from_list(&head, &lock, rem_list);			\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		list_data = err;					\
+									\
+	err = __read_from_tree(&root, &lock, rem_tree);			\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		tree_data = err;					\
+									\
+	return tree_data + list_data;					\
+}
+
+/* Similar to insert_read_both, but list data is read and possibly removed
+ * first
+ *
+ * Results should be no different than reading and possibly removing rbtree
+ * node first
+ */
+INSERT_READ_BOTH(true, true, "insert_read_both_list_first: remove from tree + list");
+INSERT_READ_BOTH(false, false, "insert_read_both_list_first: remove from neither");
+INSERT_READ_BOTH(true, false, "insert_read_both_list_first: remove from tree");
+INSERT_READ_BOTH(false, true, "insert_read_both_list_first: remove from list");
+
+#define INSERT_DOUBLE_READ_AND_DEL(read_fn, read_root, desc)		\
+SEC("tc")								\
+__description(desc)							\
+__success __retval(-1)							\
+long insert_double_##read_fn##_and_del_##read_root(void *ctx)		\
+{									\
+	long err, list_data;						\
+									\
+	err = __insert_in_tree_and_list(&head, &root, &lock);		\
+	if (err)							\
+		return err;						\
+									\
+	err = read_fn(&read_root, &lock, true);				\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		list_data = err;					\
+									\
+	err = read_fn(&read_root, &lock, true);				\
+	if (err < 0)							\
+		return err;						\
+									\
+	return err + list_data;						\
+}
+
+/* Insert into both tree and list, then try reading-and-removing from either twice
+ *
+ * The second read-and-remove should fail on read step since the node has
+ * already been removed
+ */
+INSERT_DOUBLE_READ_AND_DEL(__read_from_tree, root, "insert_double_del: 2x read-and-del from tree");
+INSERT_DOUBLE_READ_AND_DEL(__read_from_list, head, "insert_double_del: 2x read-and-del from list");
+
+#define INSERT_STASH_READ(rem_tree, desc)				\
+SEC("tc")								\
+__description(desc)							\
+__success __retval(84)							\
+long insert_rbtree_and_stash__del_tree_##rem_tree(void *ctx)		\
+{									\
+	long err, tree_data, map_data;					\
+									\
+	err = __stash_map_insert_tree(0, 42, &root, &lock);		\
+	if (err)							\
+		return err;						\
+									\
+	err = __read_from_tree(&root, &lock, rem_tree);			\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		tree_data = err;					\
+									\
+	err = __read_from_unstash(0);					\
+	if (err < 0)							\
+		return err;						\
+	else								\
+		map_data = err;						\
+									\
+	return tree_data + map_data;					\
+}
+
+/* Stash a refcounted node in map_val, insert same node into tree, then try
+ * reading data from tree then unstashed map_val, possibly removing from tree
+ *
+ * Removing from tree should have no effect on map_val kptr validity
+ */
+INSERT_STASH_READ(true, "insert_stash_read: remove from tree");
+INSERT_STASH_READ(false, "insert_stash_read: don't remove from tree");
+
+SEC("tc")
+__success
+long rbtree_refcounted_node_ref_escapes(void *ctx)
+{
+	struct node_acquire *n, *m;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return 1;
+
+	bpf_spin_lock(&alock);
+	bpf_rbtree_add(&aroot, &n->node, less_a);
+	m = bpf_refcount_acquire(n);
+	bpf_spin_unlock(&alock);
+
+	m->key = 2;
+	bpf_obj_drop(m);
+	return 0;
+}
+
+SEC("tc")
+__success
+long rbtree_refcounted_node_ref_escapes_owning_input(void *ctx)
+{
+	struct node_acquire *n, *m;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return 1;
+
+	m = bpf_refcount_acquire(n);
+	m->key = 2;
+
+	bpf_spin_lock(&alock);
+	bpf_rbtree_add(&aroot, &n->node, less_a);
+	bpf_spin_unlock(&alock);
+
+	bpf_obj_drop(m);
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c b/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c
new file mode 100644
index 000000000000..efcb308f80ad
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c
@@ -0,0 +1,72 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_core_read.h>
+#include "bpf_experimental.h"
+#include "bpf_misc.h"
+
+struct node_acquire {
+	long key;
+	long data;
+	struct bpf_rb_node node;
+	struct bpf_refcount refcount;
+};
+
+#define private(name) SEC(".data." #name) __hidden __attribute__((aligned(8)))
+private(A) struct bpf_spin_lock glock;
+private(A) struct bpf_rb_root groot __contains(node_acquire, node);
+
+static bool less(struct bpf_rb_node *a, const struct bpf_rb_node *b)
+{
+	struct node_acquire *node_a;
+	struct node_acquire *node_b;
+
+	node_a = container_of(a, struct node_acquire, node);
+	node_b = container_of(b, struct node_acquire, node);
+
+	return node_a->key < node_b->key;
+}
+
+SEC("?tc")
+__failure __msg("Unreleased reference id=3 alloc_insn=21")
+long rbtree_refcounted_node_ref_escapes(void *ctx)
+{
+	struct node_acquire *n, *m;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return 1;
+
+	bpf_spin_lock(&glock);
+	bpf_rbtree_add(&groot, &n->node, less);
+	/* m becomes an owning ref but is never drop'd or added to a tree */
+	m = bpf_refcount_acquire(n);
+	bpf_spin_unlock(&glock);
+
+	m->key = 2;
+	return 0;
+}
+
+SEC("?tc")
+__failure __msg("Unreleased reference id=3 alloc_insn=9")
+long rbtree_refcounted_node_ref_escapes_owning_input(void *ctx)
+{
+	struct node_acquire *n, *m;
+
+	n = bpf_obj_new(typeof(*n));
+	if (!n)
+		return 1;
+
+	/* m becomes an owning ref but is never drop'd or added to a tree */
+	m = bpf_refcount_acquire(n);
+	m->key = 2;
+
+	bpf_spin_lock(&glock);
+	bpf_rbtree_add(&groot, &n->node, less);
+	bpf_spin_unlock(&glock);
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 09b501d905217a38f03c0f07d5a66e0b5c8c1644 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Sun, 16 Apr 2023 03:49:26 -0500
Subject: bpf: Remove bpf_kfunc_call_test_kptr_get() test kfunc

We've managed to improve the UX for kptrs significantly over the last 9
months. All of the prior main use cases, struct bpf_cpumask *, struct
task_struct *, and struct cgroup *, have all been updated to be
synchronized mainly using RCU. In other words, their KF_ACQUIRE kfunc
calls are all KF_RCU, and the pointers themselves are MEM_RCU and can be
accessed in an RCU read region in BPF.

In a follow-on change, we'll be removing the KF_KPTR_GET kfunc flag.
This patch prepares for that by removing the
bpf_kfunc_call_test_kptr_get() kfunc, and all associated selftests.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230416084928.326135-2-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/bpf/test_run.c                                | 12 ----
 tools/testing/selftests/bpf/progs/map_kptr.c      | 40 ++----------
 tools/testing/selftests/bpf/progs/map_kptr_fail.c | 78 -----------------------
 tools/testing/selftests/bpf/verifier/map_kptr.c   | 27 --------
 4 files changed, 5 insertions(+), 152 deletions(-)

diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 0b9bd9b39990..f170e8a17974 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -679,17 +679,6 @@ __bpf_kfunc void bpf_kfunc_call_int_mem_release(int *p)
 {
 }
 
-__bpf_kfunc struct prog_test_ref_kfunc *
-bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **pp, int a, int b)
-{
-	struct prog_test_ref_kfunc *p = READ_ONCE(*pp);
-
-	if (!p)
-		return NULL;
-	refcount_inc(&p->cnt);
-	return p;
-}
-
 struct prog_test_pass1 {
 	int x0;
 	struct {
@@ -804,7 +793,6 @@ BTF_ID_FLAGS(func, bpf_kfunc_call_test_get_rdwr_mem, KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_get_rdonly_mem, KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_acq_rdonly_mem, KF_ACQUIRE | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_kfunc_call_int_mem_release, KF_RELEASE)
-BTF_ID_FLAGS(func, bpf_kfunc_call_test_kptr_get, KF_ACQUIRE | KF_RET_NULL | KF_KPTR_GET)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_pass_ctx)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_pass1)
 BTF_ID_FLAGS(func, bpf_kfunc_call_test_pass2)
diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c
index dae5dab1bbf7..d7150041e5d1 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr.c
@@ -115,8 +115,6 @@ DEFINE_MAP_OF_MAP(BPF_MAP_TYPE_HASH_OF_MAPS, hash_malloc_map, hash_of_hash_mallo
 DEFINE_MAP_OF_MAP(BPF_MAP_TYPE_HASH_OF_MAPS, lru_hash_map, hash_of_lru_hash_maps);
 
 extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
-extern struct prog_test_ref_kfunc *
-bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **p, int a, int b) __ksym;
 extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
 void bpf_kfunc_call_test_ref(struct prog_test_ref_kfunc *p) __ksym;
 
@@ -187,25 +185,10 @@ static void test_kptr_ref(struct map_value *v)
 	bpf_kfunc_call_test_release(p);
 }
 
-static void test_kptr_get(struct map_value *v)
-{
-	struct prog_test_ref_kfunc *p;
-
-	p = bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0);
-	if (!p)
-		return;
-	if (p->a + p->b > 100) {
-		bpf_kfunc_call_test_release(p);
-		return;
-	}
-	bpf_kfunc_call_test_release(p);
-}
-
 static void test_kptr(struct map_value *v)
 {
 	test_kptr_unref(v);
 	test_kptr_ref(v);
-	test_kptr_get(v);
 }
 
 SEC("tc")
@@ -338,38 +321,25 @@ int test_map_kptr_ref_pre(struct map_value *v)
 	if (p_st->cnt.refs.counter != ref)
 		return 4;
 
-	p = bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0);
-	if (!p)
-		return 5;
-	ref++;
-	if (p_st->cnt.refs.counter != ref) {
-		ret = 6;
-		goto end;
-	}
-	bpf_kfunc_call_test_release(p);
-	ref--;
-	if (p_st->cnt.refs.counter != ref)
-		return 7;
-
 	p = bpf_kptr_xchg(&v->ref_ptr, NULL);
 	if (!p)
-		return 8;
+		return 5;
 	bpf_kfunc_call_test_release(p);
 	ref--;
 	if (p_st->cnt.refs.counter != ref)
-		return 9;
+		return 6;
 
 	p = bpf_kfunc_call_test_acquire(&arg);
 	if (!p)
-		return 10;
+		return 7;
 	ref++;
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
-		ret = 11;
+		ret = 8;
 		goto end;
 	}
 	if (p_st->cnt.refs.counter != ref)
-		return 12;
+		return 9;
 	/* Leave in map */
 
 	return 0;
diff --git a/tools/testing/selftests/bpf/progs/map_kptr_fail.c b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
index 15bf3127dba3..da8c724f839b 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr_fail.c
@@ -21,8 +21,6 @@ struct array_map {
 
 extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
 extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
-extern struct prog_test_ref_kfunc *
-bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **p, int a, int b) __ksym;
 
 SEC("?tc")
 __failure __msg("kptr access size must be BPF_DW")
@@ -220,67 +218,6 @@ int reject_kptr_xchg_on_unref(struct __sk_buff *ctx)
 	return 0;
 }
 
-SEC("?tc")
-__failure __msg("arg#0 expected pointer to map value")
-int reject_kptr_get_no_map_val(struct __sk_buff *ctx)
-{
-	bpf_kfunc_call_test_kptr_get((void *)&ctx, 0, 0);
-	return 0;
-}
-
-SEC("?tc")
-__failure __msg("arg#0 expected pointer to map value")
-int reject_kptr_get_no_null_map_val(struct __sk_buff *ctx)
-{
-	bpf_kfunc_call_test_kptr_get(bpf_map_lookup_elem(&array_map, &(int){0}), 0, 0);
-	return 0;
-}
-
-SEC("?tc")
-__failure __msg("arg#0 no referenced kptr at map value offset=0")
-int reject_kptr_get_no_kptr(struct __sk_buff *ctx)
-{
-	struct map_value *v;
-	int key = 0;
-
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v)
-		return 0;
-
-	bpf_kfunc_call_test_kptr_get((void *)v, 0, 0);
-	return 0;
-}
-
-SEC("?tc")
-__failure __msg("arg#0 no referenced kptr at map value offset=8")
-int reject_kptr_get_on_unref(struct __sk_buff *ctx)
-{
-	struct map_value *v;
-	int key = 0;
-
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v)
-		return 0;
-
-	bpf_kfunc_call_test_kptr_get(&v->unref_ptr, 0, 0);
-	return 0;
-}
-
-SEC("?tc")
-__failure __msg("kernel function bpf_kfunc_call_test_kptr_get args#0")
-int reject_kptr_get_bad_type_match(struct __sk_buff *ctx)
-{
-	struct map_value *v;
-	int key = 0;
-
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v)
-		return 0;
-
-	bpf_kfunc_call_test_kptr_get((void *)&v->ref_memb_ptr, 0, 0);
-	return 0;
-}
-
 SEC("?tc")
 __failure __msg("R1 type=rcu_ptr_or_null_ expected=percpu_ptr_")
 int mark_ref_as_untrusted_or_null(struct __sk_buff *ctx)
@@ -428,21 +365,6 @@ int kptr_xchg_ref_state(struct __sk_buff *ctx)
 	return 0;
 }
 
-SEC("?tc")
-__failure __msg("Unreleased reference id=3 alloc_insn=")
-int kptr_get_ref_state(struct __sk_buff *ctx)
-{
-	struct map_value *v;
-	int key = 0;
-
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v)
-		return 0;
-
-	bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0);
-	return 0;
-}
-
 SEC("?tc")
 __failure __msg("Possibly NULL pointer passed to helper arg2")
 int kptr_xchg_possibly_null(struct __sk_buff *ctx)
diff --git a/tools/testing/selftests/bpf/verifier/map_kptr.c b/tools/testing/selftests/bpf/verifier/map_kptr.c
index d775ccb01989..a0cfc06d75bc 100644
--- a/tools/testing/selftests/bpf/verifier/map_kptr.c
+++ b/tools/testing/selftests/bpf/verifier/map_kptr.c
@@ -288,33 +288,6 @@
 	.result = REJECT,
 	.errstr = "off=0 kptr isn't referenced kptr",
 },
-{
-	"map_kptr: unref: bpf_kfunc_call_test_kptr_get rejected",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_LD_MAP_FD(BPF_REG_6, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ST_MEM(BPF_W, BPF_REG_2, 0, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_kptr = { 1 },
-	.result = REJECT,
-	.errstr = "arg#0 no referenced kptr at map value offset=0",
-	.fixup_kfunc_btf_id = {
-		{ "bpf_kfunc_call_test_kptr_get", 13 },
-	}
-},
 /* Tests for referenced PTR_TO_BTF_ID */
 {
 	"map_kptr: ref: loaded pointer marked as untrusted",
-- 
cgit v1.2.3-70-g09d2


From 7b4ddf3920d247c2949073b9c274301c8131332a Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Sun, 16 Apr 2023 03:49:27 -0500
Subject: bpf: Remove KF_KPTR_GET kfunc flag

We've managed to improve the UX for kptrs significantly over the last 9
months. All of the existing use cases which previously had KF_KPTR_GET
kfuncs (struct bpf_cpumask *, struct task_struct *, and struct cgroup *)
have all been updated to be synchronized using RCU. In other words,
their KF_KPTR_GET kfuncs have been removed in favor of KF_RCU |
KF_ACQUIRE kfuncs, with the pointers themselves also being readable from
maps in an RCU read region thanks to the types being RCU safe.

While KF_KPTR_GET was a logical starting point for kptrs, it's become
clear that they're not the correct abstraction. KF_KPTR_GET is a flag
that essentially does nothing other than enforcing that the argument to
a function is a pointer to a referenced kptr map value. At first glance,
that's a useful thing to guarantee to a kfunc. It gives kfuncs the
ability to try and acquire a reference on that kptr without requiring
the BPF prog to do something like this:

struct kptr_type *in_map, *new = NULL;

in_map = bpf_kptr_xchg(&map->value, NULL);
if (in_map) {
        new = bpf_kptr_type_acquire(in_map);
        in_map = bpf_kptr_xchg(&map->value, in_map);
        if (in_map)
                bpf_kptr_type_release(in_map);
}

That's clearly a pretty ugly (and racy) UX, and if using KF_KPTR_GET is
the only alternative, it's better than nothing. However, the problem
with any KF_KPTR_GET kfunc lies in the fact that it always requires some
kind of synchronization in order to safely do an opportunistic acquire
of the kptr in the map. This is because a BPF program running on another
CPU could do a bpf_kptr_xchg() on that map value, and free the kptr
after it's been read by the KF_KPTR_GET kfunc. For example, the
now-removed bpf_task_kptr_get() kfunc did the following:

struct task_struct *bpf_task_kptr_get(struct task_struct **pp)
{
            struct task_struct *p;

        rcu_read_lock();
        p = READ_ONCE(*pp);
        /* If p is non-NULL, it could still be freed by another CPU,
         * so we have to do an opportunistic refcount_inc_not_zero()
         * and return NULL if the task will be freed after the
         * current RCU read region.
         */
        |f (p && !refcount_inc_not_zero(&p->rcu_users))
                p = NULL;
        rcu_read_unlock();

        return p;
}

In other words, the kfunc uses RCU to ensure that the task remains valid
after it's been peeked from the map. However, this is completely
redundant with just defining a KF_RCU kfunc that itself does a
refcount_inc_not_zero(), which is exactly what bpf_task_acquire() now
does.

So, the question of whether KF_KPTR_GET is useful is actually, "Are
there any synchronization mechanisms / safety flags that are required by
certain kptrs, but which are not provided by the verifier to kfuncs?"
The answer to that question today is "No", because every kptr we
currently care about is RCU protected.

Even if the answer ever became "yes", the proper way to support that
referenced kptr type would be to add support for whatever
synchronization mechanism it requires in the verifier, rather than
giving kfuncs a flag that says, "Here's a pointer to a referenced kptr
in a map, do whatever you need to do."

With all that said -- so as to allow us to consolidate the kfunc API,
and simplify the verifier a bit, this patch removes KF_KPTR_GET, and all
relevant logic from the verifier.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230416084928.326135-3-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/btf.h   |  1 -
 kernel/bpf/verifier.c | 65 ---------------------------------------------------
 2 files changed, 66 deletions(-)

diff --git a/include/linux/btf.h b/include/linux/btf.h
index 813227bff58a..508199e38415 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -18,7 +18,6 @@
 #define KF_ACQUIRE	(1 << 0) /* kfunc is an acquire function */
 #define KF_RELEASE	(1 << 1) /* kfunc is a release function */
 #define KF_RET_NULL	(1 << 2) /* kfunc returns a pointer that may be NULL */
-#define KF_KPTR_GET	(1 << 3) /* kfunc returns reference to a kptr */
 /* Trusted arguments are those which are guaranteed to be valid when passed to
  * the kfunc. It is used to enforce that pointers obtained from either acquire
  * kfuncs, or from the main kernel on a tracepoint or struct_ops callback
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 6a41b69a424e..5dae11ee01c3 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -9339,11 +9339,6 @@ static bool is_kfunc_rcu(struct bpf_kfunc_call_arg_meta *meta)
 	return meta->kfunc_flags & KF_RCU;
 }
 
-static bool is_kfunc_arg_kptr_get(struct bpf_kfunc_call_arg_meta *meta, int arg)
-{
-	return arg == 0 && (meta->kfunc_flags & KF_KPTR_GET);
-}
-
 static bool __kfunc_param_match_suffix(const struct btf *btf,
 				       const struct btf_param *arg,
 				       const char *suffix)
@@ -9554,7 +9549,6 @@ enum kfunc_ptr_arg_type {
 	KF_ARG_PTR_TO_CTX,
 	KF_ARG_PTR_TO_ALLOC_BTF_ID,    /* Allocated object */
 	KF_ARG_PTR_TO_REFCOUNTED_KPTR, /* Refcounted local kptr */
-	KF_ARG_PTR_TO_KPTR,	       /* PTR_TO_KPTR but type specific */
 	KF_ARG_PTR_TO_DYNPTR,
 	KF_ARG_PTR_TO_ITER,
 	KF_ARG_PTR_TO_LIST_HEAD,
@@ -9666,21 +9660,6 @@ get_kfunc_ptr_arg_type(struct bpf_verifier_env *env,
 	if (is_kfunc_arg_refcounted_kptr(meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_REFCOUNTED_KPTR;
 
-	if (is_kfunc_arg_kptr_get(meta, argno)) {
-		if (!btf_type_is_ptr(ref_t)) {
-			verbose(env, "arg#0 BTF type must be a double pointer for kptr_get kfunc\n");
-			return -EINVAL;
-		}
-		ref_t = btf_type_by_id(meta->btf, ref_t->type);
-		ref_tname = btf_name_by_offset(meta->btf, ref_t->name_off);
-		if (!btf_type_is_struct(ref_t)) {
-			verbose(env, "kernel function %s args#0 pointer type %s %s is not supported\n",
-				meta->func_name, btf_type_str(ref_t), ref_tname);
-			return -EINVAL;
-		}
-		return KF_ARG_PTR_TO_KPTR;
-	}
-
 	if (is_kfunc_arg_dynptr(meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_DYNPTR;
 
@@ -9794,40 +9773,6 @@ static int process_kf_arg_ptr_to_btf_id(struct bpf_verifier_env *env,
 	return 0;
 }
 
-static int process_kf_arg_ptr_to_kptr(struct bpf_verifier_env *env,
-				      struct bpf_reg_state *reg,
-				      const struct btf_type *ref_t,
-				      const char *ref_tname,
-				      struct bpf_kfunc_call_arg_meta *meta,
-				      int argno)
-{
-	struct btf_field *kptr_field;
-
-	/* check_func_arg_reg_off allows var_off for
-	 * PTR_TO_MAP_VALUE, but we need fixed offset to find
-	 * off_desc.
-	 */
-	if (!tnum_is_const(reg->var_off)) {
-		verbose(env, "arg#0 must have constant offset\n");
-		return -EINVAL;
-	}
-
-	kptr_field = btf_record_find(reg->map_ptr->record, reg->off + reg->var_off.value, BPF_KPTR);
-	if (!kptr_field || kptr_field->type != BPF_KPTR_REF) {
-		verbose(env, "arg#0 no referenced kptr at map value offset=%llu\n",
-			reg->off + reg->var_off.value);
-		return -EINVAL;
-	}
-
-	if (!btf_struct_ids_match(&env->log, meta->btf, ref_t->type, 0, kptr_field->kptr.btf,
-				  kptr_field->kptr.btf_id, true)) {
-		verbose(env, "kernel function %s args#%d expected pointer to %s %s\n",
-			meta->func_name, argno, btf_type_str(ref_t), ref_tname);
-		return -EINVAL;
-	}
-	return 0;
-}
-
 static int ref_set_non_owning(struct bpf_verifier_env *env, struct bpf_reg_state *reg)
 {
 	struct bpf_verifier_state *state = env->cur_state;
@@ -10315,7 +10260,6 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 			/* Trusted arguments have the same offset checks as release arguments */
 			arg_type |= OBJ_RELEASE;
 			break;
-		case KF_ARG_PTR_TO_KPTR:
 		case KF_ARG_PTR_TO_DYNPTR:
 		case KF_ARG_PTR_TO_ITER:
 		case KF_ARG_PTR_TO_LIST_HEAD:
@@ -10368,15 +10312,6 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 				meta->arg_obj_drop.btf_id = reg->btf_id;
 			}
 			break;
-		case KF_ARG_PTR_TO_KPTR:
-			if (reg->type != PTR_TO_MAP_VALUE) {
-				verbose(env, "arg#0 expected pointer to map value\n");
-				return -EINVAL;
-			}
-			ret = process_kf_arg_ptr_to_kptr(env, reg, ref_t, ref_tname, meta, i);
-			if (ret < 0)
-				return ret;
-			break;
 		case KF_ARG_PTR_TO_DYNPTR:
 		{
 			enum bpf_arg_type dynptr_arg_type = ARG_PTR_TO_DYNPTR;
-- 
cgit v1.2.3-70-g09d2


From 530474e6d044d07b179dc4d3392fb853c47446d0 Mon Sep 17 00:00:00 2001
From: David Vernet <void@manifault.com>
Date: Sun, 16 Apr 2023 03:49:28 -0500
Subject: bpf,docs: Remove KF_KPTR_GET from documentation

A prior patch removed KF_KPTR_GET from the kernel. Now that it's no
longer accessible to kfunc authors, this patch removes it from the BPF
kfunc documentation.

Signed-off-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20230416084928.326135-4-void@manifault.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/kfuncs.rst | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index 3b42cfe12437..ea2516374d92 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -184,16 +184,7 @@ in. All copies of the pointer being released are invalidated as a result of
 invoking kfunc with this flag. KF_RELEASE kfuncs automatically receive the
 protection afforded by the KF_TRUSTED_ARGS flag described below.
 
-2.4.4 KF_KPTR_GET flag
-----------------------
-
-The KF_KPTR_GET flag is used to indicate that the kfunc takes the first argument
-as a pointer to kptr, safely increments the refcount of the object it points to,
-and returns a reference to the user. The rest of the arguments may be normal
-arguments of a kfunc. The KF_KPTR_GET flag should be used in conjunction with
-KF_ACQUIRE and KF_RET_NULL flags.
-
-2.4.5 KF_TRUSTED_ARGS flag
+2.4.4 KF_TRUSTED_ARGS flag
 --------------------------
 
 The KF_TRUSTED_ARGS flag is used for kfuncs taking pointer arguments. It
@@ -205,7 +196,7 @@ exception described below).
 There are two types of pointers to kernel objects which are considered "valid":
 
 1. Pointers which are passed as tracepoint or struct_ops callback arguments.
-2. Pointers which were returned from a KF_ACQUIRE or KF_KPTR_GET kfunc.
+2. Pointers which were returned from a KF_ACQUIRE kfunc.
 
 Pointers to non-BTF objects (e.g. scalar pointers) may also be passed to
 KF_TRUSTED_ARGS kfuncs, and may have a non-zero offset.
@@ -232,13 +223,13 @@ In other words, you must:
 2. Specify the type and name of the trusted nested field. This field must match
    the field in the original type definition exactly.
 
-2.4.6 KF_SLEEPABLE flag
+2.4.5 KF_SLEEPABLE flag
 -----------------------
 
 The KF_SLEEPABLE flag is used for kfuncs that may sleep. Such kfuncs can only
 be called by sleepable BPF programs (BPF_F_SLEEPABLE).
 
-2.4.7 KF_DESTRUCTIVE flag
+2.4.6 KF_DESTRUCTIVE flag
 --------------------------
 
 The KF_DESTRUCTIVE flag is used to indicate functions calling which is
@@ -247,7 +238,7 @@ rebooting or panicking. Due to this additional restrictions apply to these
 calls. At the moment they only require CAP_SYS_BOOT capability, but more can be
 added later.
 
-2.4.8 KF_RCU flag
+2.4.7 KF_RCU flag
 -----------------
 
 The KF_RCU flag is a weaker version of KF_TRUSTED_ARGS. The kfuncs marked with
@@ -260,7 +251,7 @@ also be KF_RET_NULL.
 
 .. _KF_deprecated_flag:
 
-2.4.9 KF_DEPRECATED flag
+2.4.8 KF_DEPRECATED flag
 ------------------------
 
 The KF_DEPRECATED flag is used for kfuncs which are scheduled to be
-- 
cgit v1.2.3-70-g09d2


From c6d6ef3ee3b6118dae6b32816e43c790f81079b2 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Fri, 14 Apr 2023 10:26:59 +0200
Subject: net: phy: micrel: Fix PTP_PF_PEROUT for lan8841

If the 1PPS output was enabled and then lan8841 was configured to be a
follower, then target clock which is used to generate the 1PPS was not
configure correctly. The problem was that for each adjustments of the
time, also the nanosecond part of the target clock was changed.
Therefore the initial nanosecond part of the target clock was changed.
The issue can be observed if both the leader and the follower are
generating 1PPS and see that their PPS are not aligned even if the time
is allined.
The fix consists of not modifying the nanosecond part of the target
clock when adjusting the time. In this way the 1PPS get also aligned.

Fixes: e4ed8ba08e3f ("net: phy: micrel: Add support for PTP_PF_PEROUT for lan8841")
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/micrel.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 2184b1e859ae..602e5a46b796 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -3761,7 +3761,7 @@ static int lan8841_ptp_update_target(struct kszphy_ptp_priv *ptp_priv,
 				     const struct timespec64 *ts)
 {
 	return lan8841_ptp_set_target(ptp_priv, LAN8841_EVENT_A,
-				      ts->tv_sec + LAN8841_BUFFER_TIME, ts->tv_nsec);
+				      ts->tv_sec + LAN8841_BUFFER_TIME, 0);
 }
 
 #define LAN8841_PTP_LTC_TARGET_RELOAD_SEC_HI(event)	((event) == LAN8841_EVENT_A ? 282 : 292)
-- 
cgit v1.2.3-70-g09d2


From 74cc26f416b9ed88af300393a6f06e0765ebde8b Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Fri, 14 Apr 2023 09:17:48 -0400
Subject: selftests: openvswitch: add interface support

Includes an associated test to generate netns and connect
interfaces, with the option to include packet tracing.

This will be used in the future when flow support is added
for additional test cases.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/net/openvswitch/openvswitch.sh       |  55 ++++++++++
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 118 +++++++++++++++++++--
 2 files changed, 163 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index 7ce46700a3ae..18383b0b7b9c 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -70,6 +70,49 @@ ovs_add_dp () {
 	on_exit "ovs_sbx $sbxname python3 $ovs_base/ovs-dpctl.py del-dp $1;"
 }
 
+ovs_add_if () {
+	info "Adding IF to DP: br:$2 if:$3"
+	ovs_sbx "$1" python3 $ovs_base/ovs-dpctl.py add-if "$2" "$3" || return 1
+}
+
+ovs_del_if () {
+	info "Deleting IF from DP: br:$2 if:$3"
+	ovs_sbx "$1" python3 $ovs_base/ovs-dpctl.py del-if "$2" "$3" || return 1
+}
+
+ovs_netns_spawn_daemon() {
+	sbx=$1
+	shift
+	netns=$1
+	shift
+	info "spawning cmd: $*"
+	ip netns exec $netns $*  >> $ovs_dir/stdout  2>> $ovs_dir/stderr &
+	pid=$!
+	ovs_sbx "$sbx" on_exit "kill -TERM $pid 2>/dev/null"
+}
+
+ovs_add_netns_and_veths () {
+	info "Adding netns attached: sbx:$1 dp:$2 {$3, $4, $5}"
+	ovs_sbx "$1" ip netns add "$3" || return 1
+	on_exit "ovs_sbx $1 ip netns del $3"
+	ovs_sbx "$1" ip link add "$4" type veth peer name "$5" || return 1
+	on_exit "ovs_sbx $1 ip link del $4 >/dev/null 2>&1"
+	ovs_sbx "$1" ip link set "$4" up || return 1
+	ovs_sbx "$1" ip link set "$5" netns "$3" || return 1
+	ovs_sbx "$1" ip netns exec "$3" ip link set "$5" up || return 1
+
+	if [ "$6" != "" ]; then
+		ovs_sbx "$1" ip netns exec "$3" ip addr add "$6" dev "$5" \
+		    || return 1
+	fi
+
+	ovs_add_if "$1" "$2" "$4" || return 1
+	[ $TRACING -eq 1 ] && ovs_netns_spawn_daemon "$1" "$ns" \
+			tcpdump -i any -s 65535
+
+	return 0
+}
+
 usage() {
 	echo
 	echo "$0 [OPTIONS] [TEST]..."
@@ -101,6 +144,18 @@ test_netlink_checks () {
 		return 1
 	fi
 
+	ovs_add_netns_and_veths "test_netlink_checks" nv0 left left0 l0 || \
+	    return 1
+	ovs_add_netns_and_veths "test_netlink_checks" nv0 right right0 r0 || \
+	    return 1
+	[ $(python3 $ovs_base/ovs-dpctl.py show nv0 | grep port | \
+	    wc -l) == 3 ] || \
+	      return 1
+	ovs_del_if "test_netlink_checks" nv0 right0 || return 1
+	[ $(python3 $ovs_base/ovs-dpctl.py show nv0 | grep port | \
+	    wc -l) == 2 ] || \
+	      return 1
+
 	return 0
 }
 
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 5d467d1993cb..626013dfd020 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -50,7 +50,6 @@ class ovs_dp_msg(genlmsg):
 
 
 class OvsDatapath(GenericNetlinkSocket):
-
     OVS_DP_F_VPORT_PIDS = 1 << 1
     OVS_DP_F_DISPATCH_UPCALL_PER_CPU = 1 << 3
 
@@ -170,6 +169,12 @@ class OvsDatapath(GenericNetlinkSocket):
 
 
 class OvsVport(GenericNetlinkSocket):
+    OVS_VPORT_TYPE_NETDEV = 1
+    OVS_VPORT_TYPE_INTERNAL = 2
+    OVS_VPORT_TYPE_GRE = 3
+    OVS_VPORT_TYPE_VXLAN = 4
+    OVS_VPORT_TYPE_GENEVE = 5
+
     class ovs_vport_msg(ovs_dp_msg):
         nla_map = (
             ("OVS_VPORT_ATTR_UNSPEC", "none"),
@@ -197,17 +202,30 @@ class OvsVport(GenericNetlinkSocket):
             )
 
     def type_to_str(vport_type):
-        if vport_type == 1:
+        if vport_type == OvsVport.OVS_VPORT_TYPE_NETDEV:
             return "netdev"
-        elif vport_type == 2:
+        elif vport_type == OvsVport.OVS_VPORT_TYPE_INTERNAL:
             return "internal"
-        elif vport_type == 3:
+        elif vport_type == OvsVport.OVS_VPORT_TYPE_GRE:
             return "gre"
-        elif vport_type == 4:
+        elif vport_type == OvsVport.OVS_VPORT_TYPE_VXLAN:
             return "vxlan"
-        elif vport_type == 5:
+        elif vport_type == OvsVport.OVS_VPORT_TYPE_GENEVE:
             return "geneve"
-        return "unknown:%d" % vport_type
+        raise ValueError("Unknown vport type:%d" % vport_type)
+
+    def str_to_type(vport_type):
+        if vport_type == "netdev":
+            return OvsVport.OVS_VPORT_TYPE_NETDEV
+        elif vport_type == "internal":
+            return OvsVport.OVS_VPORT_TYPE_INTERNAL
+        elif vport_type == "gre":
+            return OvsVport.OVS_VPORT_TYPE_INTERNAL
+        elif vport_type == "vxlan":
+            return OvsVport.OVS_VPORT_TYPE_VXLAN
+        elif vport_type == "geneve":
+            return OvsVport.OVS_VPORT_TYPE_GENEVE
+        raise ValueError("Unknown vport type: '%s'" % vport_type)
 
     def __init__(self):
         GenericNetlinkSocket.__init__(self)
@@ -238,8 +256,51 @@ class OvsVport(GenericNetlinkSocket):
                 raise ne
         return reply
 
+    def attach(self, dpindex, vport_ifname, ptype):
+        msg = OvsVport.ovs_vport_msg()
+
+        msg["cmd"] = OVS_VPORT_CMD_NEW
+        msg["version"] = OVS_DATAPATH_VERSION
+        msg["reserved"] = 0
+        msg["dpifindex"] = dpindex
+        port_type = OvsVport.str_to_type(ptype)
+
+        msg["attrs"].append(["OVS_VPORT_ATTR_TYPE", port_type])
+        msg["attrs"].append(["OVS_VPORT_ATTR_NAME", vport_ifname])
+        msg["attrs"].append(["OVS_VPORT_ATTR_UPCALL_PID", [self.pid]])
+
+        try:
+            reply = self.nlm_request(
+                msg, msg_type=self.prid, msg_flags=NLM_F_REQUEST | NLM_F_ACK
+            )
+            reply = reply[0]
+        except NetlinkError as ne:
+            raise ne
+        return reply
+
+    def detach(self, dpindex, vport_ifname):
+        msg = OvsVport.ovs_vport_msg()
+
+        msg["cmd"] = OVS_VPORT_CMD_DEL
+        msg["version"] = OVS_DATAPATH_VERSION
+        msg["reserved"] = 0
+        msg["dpifindex"] = dpindex
+        msg["attrs"].append(["OVS_VPORT_ATTR_NAME", vport_ifname])
 
-def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB()):
+        try:
+            reply = self.nlm_request(
+                msg, msg_type=self.prid, msg_flags=NLM_F_REQUEST | NLM_F_ACK
+            )
+            reply = reply[0]
+        except NetlinkError as ne:
+            if ne.code == errno.ENODEV:
+                reply = None
+            else:
+                raise ne
+        return reply
+
+
+def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB(), vpl=OvsVport()):
     dp_name = dp_lookup_rep.get_attr("OVS_DP_ATTR_NAME")
     base_stats = dp_lookup_rep.get_attr("OVS_DP_ATTR_STATS")
     megaflow_stats = dp_lookup_rep.get_attr("OVS_DP_ATTR_MEGAFLOW_STATS")
@@ -265,7 +326,6 @@ def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB()):
         print("  features: 0x%X" % user_features)
 
     # port print out
-    vpl = OvsVport()
     for iface in ndb.interfaces:
         rep = vpl.info(iface.ifname, ifindex)
         if rep is not None:
@@ -312,9 +372,25 @@ def main(argv):
     deldpcmd = subparsers.add_parser("del-dp")
     deldpcmd.add_argument("deldp", help="Datapath Name")
 
+    addifcmd = subparsers.add_parser("add-if")
+    addifcmd.add_argument("dpname", help="Datapath Name")
+    addifcmd.add_argument("addif", help="Interface name for adding")
+    addifcmd.add_argument(
+        "-t",
+        "--ptype",
+        type=str,
+        default="netdev",
+        choices=["netdev", "internal"],
+        help="Interface type (default netdev)",
+    )
+    delifcmd = subparsers.add_parser("del-if")
+    delifcmd.add_argument("dpname", help="Datapath Name")
+    delifcmd.add_argument("delif", help="Interface name for adding")
+
     args = parser.parse_args()
 
     ovsdp = OvsDatapath()
+    ovsvp = OvsVport()
     ndb = NDB()
 
     if hasattr(args, "showdp"):
@@ -328,7 +404,7 @@ def main(argv):
 
             if rep is not None:
                 found = True
-                print_ovsdp_full(rep, iface.index, ndb)
+                print_ovsdp_full(rep, iface.index, ndb, ovsvp)
 
         if not found:
             msg = "No DP found"
@@ -343,6 +419,28 @@ def main(argv):
             print("DP '%s' added" % args.adddp)
     elif hasattr(args, "deldp"):
         ovsdp.destroy(args.deldp)
+    elif hasattr(args, "addif"):
+        rep = ovsdp.info(args.dpname, 0)
+        if rep is None:
+            print("DP '%s' not found." % args.dpname)
+            return 1
+        rep = ovsvp.attach(rep["dpifindex"], args.addif, args.ptype)
+        msg = "vport '%s'" % args.addif
+        if rep and rep["header"]["error"] is None:
+            msg += " added."
+        else:
+            msg += " failed to add."
+    elif hasattr(args, "delif"):
+        rep = ovsdp.info(args.dpname, 0)
+        if rep is None:
+            print("DP '%s' not found." % args.dpname)
+            return 1
+        rep = ovsvp.detach(rep["dpifindex"], args.delif)
+        msg = "vport '%s'" % args.delif
+        if rep and rep["header"]["error"] is None:
+            msg += " removed."
+        else:
+            msg += " failed to remove."
 
     return 0
 
-- 
cgit v1.2.3-70-g09d2


From e52b07aa1a54fcb66461149c5185a815c1c60340 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Fri, 14 Apr 2023 09:17:49 -0400
Subject: selftests: openvswitch: add flow dump support

Add a basic set of fields to print in a 'dpflow' format.  This will be
used by future commits to check for flow fields after parsing, as
well as verifying the flow fields pushed into the kernel from
userspace.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 1026 ++++++++++++++++++++
 1 file changed, 1026 insertions(+)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 626013dfd020..21b1b8deda7d 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -6,15 +6,21 @@
 
 import argparse
 import errno
+import ipaddress
+import logging
 import sys
+import time
 
 try:
     from pyroute2 import NDB
 
+    from pyroute2.netlink import NLA_F_NESTED
     from pyroute2.netlink import NLM_F_ACK
+    from pyroute2.netlink import NLM_F_DUMP
     from pyroute2.netlink import NLM_F_REQUEST
     from pyroute2.netlink import genlmsg
     from pyroute2.netlink import nla
+    from pyroute2.netlink import nlmsg_atoms
     from pyroute2.netlink.exceptions import NetlinkError
     from pyroute2.netlink.generic import GenericNetlinkSocket
 except ModuleNotFoundError:
@@ -40,6 +46,36 @@ OVS_VPORT_CMD_DEL = 2
 OVS_VPORT_CMD_GET = 3
 OVS_VPORT_CMD_SET = 4
 
+OVS_FLOW_CMD_NEW = 1
+OVS_FLOW_CMD_DEL = 2
+OVS_FLOW_CMD_GET = 3
+OVS_FLOW_CMD_SET = 4
+
+
+def macstr(mac):
+    outstr = ":".join(["%02X" % i for i in mac])
+    return outstr
+
+
+def convert_mac(mac_str, mask=False):
+    if mac_str is None or mac_str == "":
+        mac_str = "00:00:00:00:00:00"
+    if mask is True and mac_str != "00:00:00:00:00:00":
+        mac_str = "FF:FF:FF:FF:FF:FF"
+    mac_split = mac_str.split(":")
+    ret = bytearray([int(i, 16) for i in mac_split])
+    return bytes(ret)
+
+
+def convert_ipv4(ip, mask=False):
+    if ip is None:
+        ip = 0
+    if mask is True:
+        if ip != 0:
+            ip = int(ipaddress.IPv4Address(ip)) & 0xFFFFFFFF
+
+    return int(ipaddress.IPv4Address(ip))
+
 
 class ovs_dp_msg(genlmsg):
     # include the OVS version
@@ -49,6 +85,847 @@ class ovs_dp_msg(genlmsg):
     fields = genlmsg.fields + (("dpifindex", "I"),)
 
 
+class ovsactions(nla):
+    nla_flags = NLA_F_NESTED
+
+    nla_map = (
+        ("OVS_ACTION_ATTR_UNSPEC", "none"),
+        ("OVS_ACTION_ATTR_OUTPUT", "uint32"),
+        ("OVS_ACTION_ATTR_USERSPACE", "userspace"),
+        ("OVS_ACTION_ATTR_SET", "none"),
+        ("OVS_ACTION_ATTR_PUSH_VLAN", "none"),
+        ("OVS_ACTION_ATTR_POP_VLAN", "flag"),
+        ("OVS_ACTION_ATTR_SAMPLE", "none"),
+        ("OVS_ACTION_ATTR_RECIRC", "uint32"),
+        ("OVS_ACTION_ATTR_HASH", "none"),
+        ("OVS_ACTION_ATTR_PUSH_MPLS", "none"),
+        ("OVS_ACTION_ATTR_POP_MPLS", "flag"),
+        ("OVS_ACTION_ATTR_SET_MASKED", "none"),
+        ("OVS_ACTION_ATTR_CT", "ctact"),
+        ("OVS_ACTION_ATTR_TRUNC", "uint32"),
+        ("OVS_ACTION_ATTR_PUSH_ETH", "none"),
+        ("OVS_ACTION_ATTR_POP_ETH", "flag"),
+        ("OVS_ACTION_ATTR_CT_CLEAR", "flag"),
+        ("OVS_ACTION_ATTR_PUSH_NSH", "none"),
+        ("OVS_ACTION_ATTR_POP_NSH", "flag"),
+        ("OVS_ACTION_ATTR_METER", "none"),
+        ("OVS_ACTION_ATTR_CLONE", "none"),
+        ("OVS_ACTION_ATTR_CHECK_PKT_LEN", "none"),
+        ("OVS_ACTION_ATTR_ADD_MPLS", "none"),
+        ("OVS_ACTION_ATTR_DEC_TTL", "none"),
+    )
+
+    class ctact(nla):
+        nla_flags = NLA_F_NESTED
+
+        nla_map = (
+            ("OVS_CT_ATTR_NONE", "none"),
+            ("OVS_CT_ATTR_COMMIT", "flag"),
+            ("OVS_CT_ATTR_ZONE", "uint16"),
+            ("OVS_CT_ATTR_MARK", "none"),
+            ("OVS_CT_ATTR_LABELS", "none"),
+            ("OVS_CT_ATTR_HELPER", "asciiz"),
+            ("OVS_CT_ATTR_NAT", "natattr"),
+            ("OVS_CT_ATTR_FORCE_COMMIT", "flag"),
+            ("OVS_CT_ATTR_EVENTMASK", "uint32"),
+            ("OVS_CT_ATTR_TIMEOUT", "asciiz"),
+        )
+
+        class natattr(nla):
+            nla_flags = NLA_F_NESTED
+
+            nla_map = (
+                ("OVS_NAT_ATTR_NONE", "none"),
+                ("OVS_NAT_ATTR_SRC", "flag"),
+                ("OVS_NAT_ATTR_DST", "flag"),
+                ("OVS_NAT_ATTR_IP_MIN", "ipaddr"),
+                ("OVS_NAT_ATTR_IP_MAX", "ipaddr"),
+                ("OVS_NAT_ATTR_PROTO_MIN", "uint16"),
+                ("OVS_NAT_ATTR_PROTO_MAX", "uint16"),
+                ("OVS_NAT_ATTR_PERSISTENT", "flag"),
+                ("OVS_NAT_ATTR_PROTO_HASH", "flag"),
+                ("OVS_NAT_ATTR_PROTO_RANDOM", "flag"),
+            )
+
+            def dpstr(self, more=False):
+                print_str = "nat("
+
+                if self.get_attr("OVS_NAT_ATTR_SRC"):
+                    print_str += "src"
+                elif self.get_attr("OVS_NAT_ATTR_DST"):
+                    print_str += "dst"
+                else:
+                    print_str += "XXX-unknown-nat"
+
+                if self.get_attr("OVS_NAT_ATTR_IP_MIN") or self.get_attr(
+                    "OVS_NAT_ATTR_IP_MAX"
+                ):
+                    if self.get_attr("OVS_NAT_ATTR_IP_MIN"):
+                        print_str += "=%s," % str(
+                            self.get_attr("OVS_NAT_ATTR_IP_MIN")
+                        )
+
+                    if self.get_attr("OVS_NAT_ATTR_IP_MAX"):
+                        print_str += "-%s," % str(
+                            self.get_attr("OVS_NAT_ATTR_IP_MAX")
+                        )
+                else:
+                    print_str += ","
+
+                if self.get_attr("OVS_NAT_ATTR_PROTO_MIN"):
+                    print_str += "proto_min=%d," % self.get_attr(
+                        "OVS_NAT_ATTR_PROTO_MIN"
+                    )
+
+                if self.get_attr("OVS_NAT_ATTR_PROTO_MAX"):
+                    print_str += "proto_max=%d," % self.get_attr(
+                        "OVS_NAT_ATTR_PROTO_MAX"
+                    )
+
+                if self.get_attr("OVS_NAT_ATTR_PERSISTENT"):
+                    print_str += "persistent,"
+                if self.get_attr("OVS_NAT_ATTR_HASH"):
+                    print_str += "hash,"
+                if self.get_attr("OVS_NAT_ATTR_RANDOM"):
+                    print_str += "random"
+                print_str += ")"
+                return print_str
+
+        def dpstr(self, more=False):
+            print_str = "ct("
+
+            if self.get_attr("OVS_CT_ATTR_COMMIT") is not None:
+                print_str += "commit,"
+            if self.get_attr("OVS_CT_ATTR_ZONE") is not None:
+                print_str += "zone=%d," % self.get_attr("OVS_CT_ATTR_ZONE")
+            if self.get_attr("OVS_CT_ATTR_HELPER") is not None:
+                print_str += "helper=%s," % self.get_attr("OVS_CT_ATTR_HELPER")
+            if self.get_attr("OVS_CT_ATTR_NAT") is not None:
+                print_str += self.get_attr("OVS_CT_ATTR_NAT").dpstr(more)
+                print_str += ","
+            if self.get_attr("OVS_CT_ATTR_FORCE_COMMIT") is not None:
+                print_str += "force,"
+            if self.get_attr("OVS_CT_ATTR_EVENTMASK") is not None:
+                print_str += "emask=0x%X," % self.get_attr(
+                    "OVS_CT_ATTR_EVENTMASK"
+                )
+            if self.get_attr("OVS_CT_ATTR_TIMEOUT") is not None:
+                print_str += "timeout=%s" % self.get_attr(
+                    "OVS_CT_ATTR_TIMEOUT"
+                )
+            print_str += ")"
+            return print_str
+
+    class userspace(nla):
+        nla_flags = NLA_F_NESTED
+
+        nla_map = (
+            ("OVS_USERSPACE_ATTR_UNUSED", "none"),
+            ("OVS_USERSPACE_ATTR_PID", "uint32"),
+            ("OVS_USERSPACE_ATTR_USERDATA", "array(uint8)"),
+            ("OVS_USERSPACE_ATTR_EGRESS_TUN_PORT", "uint32"),
+        )
+
+        def dpstr(self, more=False):
+            print_str = "userspace("
+            if self.get_attr("OVS_USERSPACE_ATTR_PID") is not None:
+                print_str += "pid=%d," % self.get_attr(
+                    "OVS_USERSPACE_ATTR_PID"
+                )
+            if self.get_attr("OVS_USERSPACE_ATTR_USERDATA") is not None:
+                print_str += "userdata="
+                for f in self.get_attr("OVS_USERSPACE_ATTR_USERDATA"):
+                    print_str += "%x." % f
+            if self.get_attr("OVS_USERSPACE_ATTR_TUN_PORT") is not None:
+                print_str += "egress_tun_port=%d" % self.get_attr(
+                    "OVS_USERSPACE_ATTR_TUN_PORT"
+                )
+            print_str += ")"
+            return print_str
+
+    def dpstr(self, more=False):
+        print_str = ""
+
+        for field in self.nla_map:
+            if field[1] == "none" or self.get_attr(field[0]) is None:
+                continue
+            if print_str != "":
+                print_str += ","
+
+            if field[1] == "uint32":
+                if field[0] == "OVS_ACTION_ATTR_OUTPUT":
+                    print_str += "%d" % int(self.get_attr(field[0]))
+                elif field[0] == "OVS_ACTION_ATTR_RECIRC":
+                    print_str += "recirc(0x%x)" % int(self.get_attr(field[0]))
+                elif field[0] == "OVS_ACTION_ATTR_TRUNC":
+                    print_str += "trunc(%d)" % int(self.get_attr(field[0]))
+            elif field[1] == "flag":
+                if field[0] == "OVS_ACTION_ATTR_CT_CLEAR":
+                    print_str += "ct_clear"
+                elif field[0] == "OVS_ACTION_ATTR_POP_VLAN":
+                    print_str += "pop_vlan"
+                elif field[0] == "OVS_ACTION_ATTR_POP_ETH":
+                    print_str += "pop_eth"
+                elif field[0] == "OVS_ACTION_ATTR_POP_NSH":
+                    print_str += "pop_nsh"
+                elif field[0] == "OVS_ACTION_ATTR_POP_MPLS":
+                    print_str += "pop_mpls"
+            else:
+                datum = self.get_attr(field[0])
+                print_str += datum.dpstr(more)
+
+        return print_str
+
+
+class ovskey(nla):
+    nla_flags = NLA_F_NESTED
+    nla_map = (
+        ("OVS_KEY_ATTR_UNSPEC", "none"),
+        ("OVS_KEY_ATTR_ENCAP", "none"),
+        ("OVS_KEY_ATTR_PRIORITY", "uint32"),
+        ("OVS_KEY_ATTR_IN_PORT", "uint32"),
+        ("OVS_KEY_ATTR_ETHERNET", "ethaddr"),
+        ("OVS_KEY_ATTR_VLAN", "uint16"),
+        ("OVS_KEY_ATTR_ETHERTYPE", "be16"),
+        ("OVS_KEY_ATTR_IPV4", "ovs_key_ipv4"),
+        ("OVS_KEY_ATTR_IPV6", "ovs_key_ipv6"),
+        ("OVS_KEY_ATTR_TCP", "ovs_key_tcp"),
+        ("OVS_KEY_ATTR_UDP", "ovs_key_udp"),
+        ("OVS_KEY_ATTR_ICMP", "ovs_key_icmp"),
+        ("OVS_KEY_ATTR_ICMPV6", "ovs_key_icmpv6"),
+        ("OVS_KEY_ATTR_ARP", "ovs_key_arp"),
+        ("OVS_KEY_ATTR_ND", "ovs_key_nd"),
+        ("OVS_KEY_ATTR_SKB_MARK", "uint32"),
+        ("OVS_KEY_ATTR_TUNNEL", "none"),
+        ("OVS_KEY_ATTR_SCTP", "ovs_key_sctp"),
+        ("OVS_KEY_ATTR_TCP_FLAGS", "be16"),
+        ("OVS_KEY_ATTR_DP_HASH", "uint32"),
+        ("OVS_KEY_ATTR_RECIRC_ID", "uint32"),
+        ("OVS_KEY_ATTR_MPLS", "array(ovs_key_mpls)"),
+        ("OVS_KEY_ATTR_CT_STATE", "uint32"),
+        ("OVS_KEY_ATTR_CT_ZONE", "uint16"),
+        ("OVS_KEY_ATTR_CT_MARK", "uint32"),
+        ("OVS_KEY_ATTR_CT_LABELS", "none"),
+        ("OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV4", "ovs_key_ct_tuple_ipv4"),
+        ("OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV6", "ovs_key_ct_tuple_ipv6"),
+        ("OVS_KEY_ATTR_NSH", "none"),
+        ("OVS_KEY_ATTR_PACKET_TYPE", "none"),
+        ("OVS_KEY_ATTR_ND_EXTENSIONS", "none"),
+        ("OVS_KEY_ATTR_TUNNEL_INFO", "none"),
+        ("OVS_KEY_ATTR_IPV6_EXTENSIONS", "none"),
+    )
+
+    class ovs_key_proto(nla):
+        fields = (
+            ("src", "!H"),
+            ("dst", "!H"),
+        )
+
+        fields_map = (
+            ("src", "src", "%d", lambda x: int(x) if x is not None else 0),
+            ("dst", "dst", "%d", lambda x: int(x) if x is not None else 0),
+        )
+
+        def __init__(
+            self,
+            protostr,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            self.proto_str = protostr
+            nla.__init__(
+                self,
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+        def dpstr(self, masked=None, more=False):
+            outstr = self.proto_str + "("
+            first = False
+            for f in self.fields_map:
+                if first:
+                    outstr += ","
+                if masked is None:
+                    outstr += "%s=" % f[0]
+                    if isinstance(f[2], str):
+                        outstr += f[2] % self[f[1]]
+                    else:
+                        outstr += f[2](self[f[1]])
+                    first = True
+                elif more or f[3](masked[f[1]]) != 0:
+                    outstr += "%s=" % f[0]
+                    if isinstance(f[2], str):
+                        outstr += f[2] % self[f[1]]
+                    else:
+                        outstr += f[2](self[f[1]])
+                    outstr += "/"
+                    if isinstance(f[2], str):
+                        outstr += f[2] % masked[f[1]]
+                    else:
+                        outstr += f[2](masked[f[1]])
+                    first = True
+            outstr += ")"
+            return outstr
+
+    class ethaddr(ovs_key_proto):
+        fields = (
+            ("src", "!6s"),
+            ("dst", "!6s"),
+        )
+
+        fields_map = (
+            (
+                "src",
+                "src",
+                macstr,
+                lambda x: int.from_bytes(x, "big"),
+                convert_mac,
+            ),
+            (
+                "dst",
+                "dst",
+                macstr,
+                lambda x: int.from_bytes(x, "big"),
+                convert_mac,
+            ),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "eth",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_ipv4(ovs_key_proto):
+        fields = (
+            ("src", "!I"),
+            ("dst", "!I"),
+            ("proto", "B"),
+            ("tos", "B"),
+            ("ttl", "B"),
+            ("frag", "B"),
+        )
+
+        fields_map = (
+            (
+                "src",
+                "src",
+                lambda x: str(ipaddress.IPv4Address(x)),
+                int,
+                convert_ipv4,
+            ),
+            (
+                "dst",
+                "dst",
+                lambda x: str(ipaddress.IPv4Address(x)),
+                int,
+                convert_ipv4,
+            ),
+            ("proto", "proto", "%d", lambda x: int(x) if x is not None else 0),
+            ("tos", "tos", "%d", lambda x: int(x) if x is not None else 0),
+            ("ttl", "ttl", "%d", lambda x: int(x) if x is not None else 0),
+            ("frag", "frag", "%d", lambda x: int(x) if x is not None else 0),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "ipv4",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_ipv6(ovs_key_proto):
+        fields = (
+            ("src", "!16s"),
+            ("dst", "!16s"),
+            ("label", "!I"),
+            ("proto", "B"),
+            ("tclass", "B"),
+            ("hlimit", "B"),
+            ("frag", "B"),
+        )
+
+        fields_map = (
+            (
+                "src",
+                "src",
+                lambda x: str(ipaddress.IPv6Address(x)),
+                lambda x: int.from_bytes(x, "big"),
+                lambda x: ipaddress.IPv6Address(x),
+            ),
+            (
+                "dst",
+                "dst",
+                lambda x: str(ipaddress.IPv6Address(x)),
+                lambda x: int.from_bytes(x, "big"),
+                lambda x: ipaddress.IPv6Address(x),
+            ),
+            ("label", "label", "%d", int),
+            ("proto", "proto", "%d", int),
+            ("tclass", "tclass", "%d", int),
+            ("hlimit", "hlimit", "%d", int),
+            ("frag", "frag", "%d", int),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "ipv6",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_tcp(ovs_key_proto):
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "tcp",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_udp(ovs_key_proto):
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "udp",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_sctp(ovs_key_proto):
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "sctp",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_icmp(ovs_key_proto):
+        fields = (
+            ("type", "B"),
+            ("code", "B"),
+        )
+
+        fields_map = (
+            ("type", "type", "%d", int),
+            ("code", "code", "%d", int),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "icmp",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_icmpv6(ovs_key_icmp):
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "icmpv6",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_arp(ovs_key_proto):
+        fields = (
+            ("sip", "!I"),
+            ("tip", "!I"),
+            ("op", "!H"),
+            ("sha", "!6s"),
+            ("tha", "!6s"),
+            ("pad", "xx"),
+        )
+
+        fields_map = (
+            (
+                "sip",
+                "sip",
+                lambda x: str(ipaddress.IPv4Address(x)),
+                int,
+                convert_ipv4,
+            ),
+            (
+                "tip",
+                "tip",
+                lambda x: str(ipaddress.IPv4Address(x)),
+                int,
+                convert_ipv4,
+            ),
+            ("op", "op", "%d", lambda x: int(x) if x is not None else 0),
+            (
+                "sha",
+                "sha",
+                macstr,
+                lambda x: int.from_bytes(x, "big"),
+                convert_mac,
+            ),
+            (
+                "tha",
+                "tha",
+                macstr,
+                lambda x: int.from_bytes(x, "big"),
+                convert_mac,
+            ),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "arp",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_nd(ovs_key_proto):
+        fields = (
+            ("target", "!16s"),
+            ("sll", "!6s"),
+            ("tll", "!6s"),
+        )
+
+        fields_map = (
+            (
+                "target",
+                "target",
+                lambda x: str(ipaddress.IPv6Address(x)),
+                lambda x: int.from_bytes(x, "big"),
+            ),
+            ("sll", "sll", macstr, lambda x: int.from_bytes(x, "big")),
+            ("tll", "tll", macstr, lambda x: int.from_bytes(x, "big")),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "nd",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_ct_tuple_ipv4(ovs_key_proto):
+        fields = (
+            ("src", "!I"),
+            ("dst", "!I"),
+            ("tp_src", "!H"),
+            ("tp_dst", "!H"),
+            ("proto", "B"),
+        )
+
+        fields_map = (
+            (
+                "src",
+                "src",
+                lambda x: str(ipaddress.IPv4Address(x)),
+                int,
+            ),
+            (
+                "dst",
+                "dst",
+                lambda x: str(ipaddress.IPv6Address(x)),
+                int,
+            ),
+            ("tp_src", "tp_src", "%d", int),
+            ("tp_dst", "tp_dst", "%d", int),
+            ("proto", "proto", "%d", int),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "ct_tuple4",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_ct_tuple_ipv6(nla):
+        fields = (
+            ("src", "!16s"),
+            ("dst", "!16s"),
+            ("tp_src", "!H"),
+            ("tp_dst", "!H"),
+            ("proto", "B"),
+        )
+
+        fields_map = (
+            (
+                "src",
+                "src",
+                lambda x: str(ipaddress.IPv6Address(x)),
+                lambda x: int.from_bytes(x, "big", convertmac),
+            ),
+            (
+                "dst",
+                "dst",
+                lambda x: str(ipaddress.IPv6Address(x)),
+                lambda x: int.from_bytes(x, "big"),
+            ),
+            ("tp_src", "tp_src", "%d", int),
+            ("tp_dst", "tp_dst", "%d", int),
+            ("proto", "proto", "%d", int),
+        )
+
+        def __init__(
+            self,
+            data=None,
+            offset=None,
+            parent=None,
+            length=None,
+            init=None,
+        ):
+            ovskey.ovs_key_proto.__init__(
+                self,
+                "ct_tuple6",
+                data=data,
+                offset=offset,
+                parent=parent,
+                length=length,
+                init=init,
+            )
+
+    class ovs_key_mpls(nla):
+        fields = (("lse", ">I"),)
+
+    def dpstr(self, mask=None, more=False):
+        print_str = ""
+
+        for field in (
+            (
+                "OVS_KEY_ATTR_PRIORITY",
+                "skb_priority",
+                "%d",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_SKB_MARK",
+                "skb_mark",
+                "%d",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_RECIRC_ID",
+                "recirc_id",
+                "0x%08X",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_DP_HASH",
+                "dp_hash",
+                "0x%08X",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_CT_STATE",
+                "ct_state",
+                "0x%04x",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_CT_ZONE",
+                "ct_zone",
+                "0x%04x",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_CT_MARK",
+                "ct_mark",
+                "0x%08x",
+                lambda x: False,
+                True,
+            ),
+            (
+                "OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV4",
+                None,
+                None,
+                False,
+                False,
+            ),
+            (
+                "OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV6",
+                None,
+                None,
+                False,
+                False,
+            ),
+            (
+                "OVS_KEY_ATTR_IN_PORT",
+                "in_port",
+                "%d",
+                lambda x: True,
+                True,
+            ),
+            ("OVS_KEY_ATTR_ETHERNET", None, None, False, False),
+            (
+                "OVS_KEY_ATTR_ETHERTYPE",
+                "eth_type",
+                "0x%04x",
+                lambda x: int(x) == 0xFFFF,
+                True,
+            ),
+            ("OVS_KEY_ATTR_IPV4", None, None, False, False),
+            ("OVS_KEY_ATTR_IPV6", None, None, False, False),
+            ("OVS_KEY_ATTR_ARP", None, None, False, False),
+            ("OVS_KEY_ATTR_TCP", None, None, False, False),
+            (
+                "OVS_KEY_ATTR_TCP_FLAGS",
+                "tcp_flags",
+                "0x%04x",
+                lambda x: False,
+                True,
+            ),
+            ("OVS_KEY_ATTR_UDP", None, None, False, False),
+            ("OVS_KEY_ATTR_SCTP", None, None, False, False),
+            ("OVS_KEY_ATTR_ICMP", None, None, False, False),
+            ("OVS_KEY_ATTR_ICMPV6", None, None, False, False),
+            ("OVS_KEY_ATTR_ND", None, None, False, False),
+        ):
+            v = self.get_attr(field[0])
+            if v is not None:
+                m = None if mask is None else mask.get_attr(field[0])
+                if field[4] is False:
+                    print_str += v.dpstr(m, more)
+                    print_str += ","
+                else:
+                    if m is None or field[3](m):
+                        print_str += field[1] + "("
+                        print_str += field[2] % v
+                        print_str += "),"
+                    elif more or m != 0:
+                        print_str += field[1] + "("
+                        print_str += (field[2] % v) + "/" + (field[2] % m)
+                        print_str += "),"
+
+        return print_str
+
+
 class OvsDatapath(GenericNetlinkSocket):
     OVS_DP_F_VPORT_PIDS = 1 << 1
     OVS_DP_F_DISPATCH_UPCALL_PER_CPU = 1 << 3
@@ -300,6 +1177,135 @@ class OvsVport(GenericNetlinkSocket):
         return reply
 
 
+class OvsFlow(GenericNetlinkSocket):
+    class ovs_flow_msg(ovs_dp_msg):
+        nla_map = (
+            ("OVS_FLOW_ATTR_UNSPEC", "none"),
+            ("OVS_FLOW_ATTR_KEY", "ovskey"),
+            ("OVS_FLOW_ATTR_ACTIONS", "ovsactions"),
+            ("OVS_FLOW_ATTR_STATS", "flowstats"),
+            ("OVS_FLOW_ATTR_TCP_FLAGS", "uint8"),
+            ("OVS_FLOW_ATTR_USED", "uint64"),
+            ("OVS_FLOW_ATTR_CLEAR", "none"),
+            ("OVS_FLOW_ATTR_MASK", "ovskey"),
+            ("OVS_FLOW_ATTR_PROBE", "none"),
+            ("OVS_FLOW_ATTR_UFID", "array(uint32)"),
+            ("OVS_FLOW_ATTR_UFID_FLAGS", "uint32"),
+        )
+
+        class flowstats(nla):
+            fields = (
+                ("packets", "=Q"),
+                ("bytes", "=Q"),
+            )
+
+        def dpstr(self, more=False):
+            ufid = self.get_attr("OVS_FLOW_ATTR_UFID")
+            ufid_str = ""
+            if ufid is not None:
+                ufid_str = (
+                    "ufid:{:08x}-{:04x}-{:04x}-{:04x}-{:04x}{:08x}".format(
+                        ufid[0],
+                        ufid[1] >> 16,
+                        ufid[1] & 0xFFFF,
+                        ufid[2] >> 16,
+                        ufid[2] & 0,
+                        ufid[3],
+                    )
+                )
+
+            key_field = self.get_attr("OVS_FLOW_ATTR_KEY")
+            keymsg = None
+            if key_field is not None:
+                keymsg = key_field
+
+            mask_field = self.get_attr("OVS_FLOW_ATTR_MASK")
+            maskmsg = None
+            if mask_field is not None:
+                maskmsg = mask_field
+
+            acts_field = self.get_attr("OVS_FLOW_ATTR_ACTIONS")
+            actsmsg = None
+            if acts_field is not None:
+                actsmsg = acts_field
+
+            print_str = ""
+
+            if more:
+                print_str += ufid_str + ","
+
+            if keymsg is not None:
+                print_str += keymsg.dpstr(maskmsg, more)
+
+            stats = self.get_attr("OVS_FLOW_ATTR_STATS")
+            if stats is None:
+                print_str += " packets:0, bytes:0,"
+            else:
+                print_str += " packets:%d, bytes:%d," % (
+                    stats["packets"],
+                    stats["bytes"],
+                )
+
+            used = self.get_attr("OVS_FLOW_ATTR_USED")
+            print_str += " used:"
+            if used is None:
+                print_str += "never,"
+            else:
+                used_time = int(used)
+                cur_time_sec = time.clock_gettime(time.CLOCK_MONOTONIC)
+                used_time = (cur_time_sec * 1000) - used_time
+                print_str += "{}s,".format(used_time / 1000)
+
+            print_str += " actions:"
+            if (
+                actsmsg is None
+                or "attrs" not in actsmsg
+                or len(actsmsg["attrs"]) == 0
+            ):
+                print_str += "drop"
+            else:
+                print_str += actsmsg.dpstr(more)
+
+            return print_str
+
+    def __init__(self):
+        GenericNetlinkSocket.__init__(self)
+
+        self.bind(OVS_FLOW_FAMILY, OvsFlow.ovs_flow_msg)
+
+    def dump(self, dpifindex, flowspec=None):
+        """
+        Returns a list of messages containing flows.
+
+        dpifindex should be a valid datapath obtained by calling
+        into the OvsDatapath lookup
+
+        flowpsec is a string which represents a flow in the dpctl
+        format.
+        """
+        msg = OvsFlow.ovs_flow_msg()
+
+        msg["cmd"] = OVS_FLOW_CMD_GET
+        msg["version"] = OVS_DATAPATH_VERSION
+        msg["reserved"] = 0
+        msg["dpifindex"] = dpifindex
+
+        msg_flags = NLM_F_REQUEST | NLM_F_ACK
+        if flowspec is None:
+            msg_flags |= NLM_F_DUMP
+        rep = None
+
+        try:
+            rep = self.nlm_request(
+                msg,
+                msg_type=self.prid,
+                msg_flags=msg_flags,
+            )
+        except NetlinkError as ne:
+            raise ne
+        return rep
+
+
 def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB(), vpl=OvsVport()):
     dp_name = dp_lookup_rep.get_attr("OVS_DP_ATTR_NAME")
     base_stats = dp_lookup_rep.get_attr("OVS_DP_ATTR_STATS")
@@ -340,12 +1346,16 @@ def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB(), vpl=OvsVport()):
 
 
 def main(argv):
+    nlmsg_atoms.ovskey = ovskey
+    nlmsg_atoms.ovsactions = ovsactions
+
     parser = argparse.ArgumentParser()
     parser.add_argument(
         "-v",
         "--verbose",
         action="count",
         help="Increment 'verbose' output counter.",
+        default=0,
     )
     subparsers = parser.add_subparsers()
 
@@ -387,10 +1397,18 @@ def main(argv):
     delifcmd.add_argument("dpname", help="Datapath Name")
     delifcmd.add_argument("delif", help="Interface name for adding")
 
+    dumpflcmd = subparsers.add_parser("dump-flows")
+    dumpflcmd.add_argument("dumpdp", help="Datapath Name")
+
     args = parser.parse_args()
 
+    if args.verbose > 0:
+        if args.verbose > 1:
+            logging.basicConfig(level=logging.DEBUG)
+
     ovsdp = OvsDatapath()
     ovsvp = OvsVport()
+    ovsflow = OvsFlow()
     ndb = NDB()
 
     if hasattr(args, "showdp"):
@@ -441,6 +1459,14 @@ def main(argv):
             msg += " removed."
         else:
             msg += " failed to remove."
+    elif hasattr(args, "dumpdp"):
+        rep = ovsdp.info(args.dumpdp, 0)
+        if rep is None:
+            print("DP '%s' not found." % args.dumpdp)
+            return 1
+        rep = ovsflow.dump(rep["dpifindex"])
+        for flow in rep:
+            print(flow.dpstr(True if args.verbose > 0 else False))
 
     return 0
 
-- 
cgit v1.2.3-70-g09d2


From 9feac87b673c63e2de9aaf21bbf46cd9c4158a97 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Fri, 14 Apr 2023 09:17:50 -0400
Subject: selftests: openvswitch: add support for upcall testing

The upcall socket interface can be exercised now to make sure that
future feature adjustments to the field can maintain backwards
compatibility.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/net/openvswitch/openvswitch.sh       |  38 +++++-
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 138 +++++++++++++++++++--
 2 files changed, 165 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index 18383b0b7b9c..3117a4be0cd0 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -11,7 +11,8 @@ VERBOSE=0
 TRACING=0
 
 tests="
-	netlink_checks				ovsnl: validate netlink attrs and settings"
+	netlink_checks				ovsnl: validate netlink attrs and settings
+	upcall_interfaces			ovs: test the upcall interfaces"
 
 info() {
     [ $VERBOSE = 0 ] || echo $*
@@ -72,7 +73,15 @@ ovs_add_dp () {
 
 ovs_add_if () {
 	info "Adding IF to DP: br:$2 if:$3"
-	ovs_sbx "$1" python3 $ovs_base/ovs-dpctl.py add-if "$2" "$3" || return 1
+	if [ "$4" != "-u" ]; then
+		ovs_sbx "$1" python3 $ovs_base/ovs-dpctl.py add-if "$2" "$3" \
+		    || return 1
+	else
+		python3 $ovs_base/ovs-dpctl.py add-if \
+		    -u "$2" "$3" >$ovs_dir/$3.out 2>$ovs_dir/$3.err &
+		pid=$!
+		on_exit "ovs_sbx $1 kill -TERM $pid 2>/dev/null"
+	fi
 }
 
 ovs_del_if () {
@@ -106,7 +115,12 @@ ovs_add_netns_and_veths () {
 		    || return 1
 	fi
 
-	ovs_add_if "$1" "$2" "$4" || return 1
+	if [ "$7" != "-u" ]; then
+		ovs_add_if "$1" "$2" "$4" || return 1
+	else
+		ovs_add_if "$1" "$2" "$4" -u || return 1
+	fi
+
 	[ $TRACING -eq 1 ] && ovs_netns_spawn_daemon "$1" "$ns" \
 			tcpdump -i any -s 65535
 
@@ -159,6 +173,24 @@ test_netlink_checks () {
 	return 0
 }
 
+test_upcall_interfaces() {
+	sbx_add "test_upcall_interfaces" || return 1
+
+	info "setting up new DP"
+	ovs_add_dp "test_upcall_interfaces" ui0 -V 2:1 || return 1
+
+	ovs_add_netns_and_veths "test_upcall_interfaces" ui0 upc left0 l0 \
+	    172.31.110.1/24 -u || return 1
+
+	sleep 1
+	info "sending arping"
+	ip netns exec upc arping -I l0 172.31.110.20 -c 1 \
+	    >$ovs_dir/arping.stdout 2>$ovs_dir/arping.stderr
+
+	grep -E "MISS upcall\[0/yes\]: .*arp\(sip=172.31.110.1,tip=172.31.110.20,op=1,sha=" $ovs_dir/left0.out >/dev/null 2>&1 || return 1
+	return 0
+}
+
 run_test() {
 	(
 	tname="$1"
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 21b1b8deda7d..1c8b36bc15d4 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -8,6 +8,8 @@ import argparse
 import errno
 import ipaddress
 import logging
+import multiprocessing
+import struct
 import sys
 import time
 
@@ -926,6 +928,51 @@ class ovskey(nla):
         return print_str
 
 
+class OvsPacket(GenericNetlinkSocket):
+    OVS_PACKET_CMD_MISS = 1  # Flow table miss
+    OVS_PACKET_CMD_ACTION = 2  # USERSPACE action
+    OVS_PACKET_CMD_EXECUTE = 3  # Apply actions to packet
+
+    class ovs_packet_msg(ovs_dp_msg):
+        nla_map = (
+            ("OVS_PACKET_ATTR_UNSPEC", "none"),
+            ("OVS_PACKET_ATTR_PACKET", "array(uint8)"),
+            ("OVS_PACKET_ATTR_KEY", "ovskey"),
+            ("OVS_PACKET_ATTR_ACTIONS", "ovsactions"),
+            ("OVS_PACKET_ATTR_USERDATA", "none"),
+            ("OVS_PACKET_ATTR_EGRESS_TUN_KEY", "none"),
+            ("OVS_PACKET_ATTR_UNUSED1", "none"),
+            ("OVS_PACKET_ATTR_UNUSED2", "none"),
+            ("OVS_PACKET_ATTR_PROBE", "none"),
+            ("OVS_PACKET_ATTR_MRU", "uint16"),
+            ("OVS_PACKET_ATTR_LEN", "uint32"),
+            ("OVS_PACKET_ATTR_HASH", "uint64"),
+        )
+
+    def __init__(self):
+        GenericNetlinkSocket.__init__(self)
+        self.bind(OVS_PACKET_FAMILY, OvsPacket.ovs_packet_msg)
+
+    def upcall_handler(self, up=None):
+        print("listening on upcall packet handler:", self.epid)
+        while True:
+            try:
+                msgs = self.get()
+                for msg in msgs:
+                    if not up:
+                        continue
+                    if msg["cmd"] == OvsPacket.OVS_PACKET_CMD_MISS:
+                        up.miss(msg)
+                    elif msg["cmd"] == OvsPacket.OVS_PACKET_CMD_ACTION:
+                        up.action(msg)
+                    elif msg["cmd"] == OvsPacket.OVS_PACKET_CMD_EXECUTE:
+                        up.execute(msg)
+                    else:
+                        print("Unkonwn cmd: %d" % msg["cmd"])
+            except NetlinkError as ne:
+                raise ne
+
+
 class OvsDatapath(GenericNetlinkSocket):
     OVS_DP_F_VPORT_PIDS = 1 << 1
     OVS_DP_F_DISPATCH_UPCALL_PER_CPU = 1 << 3
@@ -989,7 +1036,9 @@ class OvsDatapath(GenericNetlinkSocket):
 
         return reply
 
-    def create(self, dpname, shouldUpcall=False, versionStr=None):
+    def create(
+        self, dpname, shouldUpcall=False, versionStr=None, p=OvsPacket()
+    ):
         msg = OvsDatapath.dp_cmd_msg()
         msg["cmd"] = OVS_DP_CMD_NEW
         if versionStr is None:
@@ -1004,11 +1053,18 @@ class OvsDatapath(GenericNetlinkSocket):
         if versionStr is not None and versionStr.find(":") != -1:
             dpfeatures = int(versionStr.split(":")[1], 0)
         else:
-            dpfeatures = OvsDatapath.OVS_DP_F_VPORT_PIDS
-
+            if versionStr is None or versionStr.find(":") == -1:
+                dpfeatures |= OvsDatapath.OVS_DP_F_DISPATCH_UPCALL_PER_CPU
+                dpfeatures &= ~OvsDatapath.OVS_DP_F_VPORT_PIDS
+
+            nproc = multiprocessing.cpu_count()
+            procarray = []
+            for i in range(1, nproc):
+                procarray += [int(p.epid)]
+            msg["attrs"].append(["OVS_DP_ATTR_UPCALL_PID", procarray])
         msg["attrs"].append(["OVS_DP_ATTR_USER_FEATURES", dpfeatures])
         if not shouldUpcall:
-            msg["attrs"].append(["OVS_DP_ATTR_UPCALL_PID", 0])
+            msg["attrs"].append(["OVS_DP_ATTR_UPCALL_PID", [0]])
 
         try:
             reply = self.nlm_request(
@@ -1104,9 +1160,10 @@ class OvsVport(GenericNetlinkSocket):
             return OvsVport.OVS_VPORT_TYPE_GENEVE
         raise ValueError("Unknown vport type: '%s'" % vport_type)
 
-    def __init__(self):
+    def __init__(self, packet=OvsPacket()):
         GenericNetlinkSocket.__init__(self)
         self.bind(OVS_VPORT_FAMILY, OvsVport.ovs_vport_msg)
+        self.upcall_packet = packet
 
     def info(self, vport_name, dpifindex=0, portno=None):
         msg = OvsVport.ovs_vport_msg()
@@ -1144,7 +1201,37 @@ class OvsVport(GenericNetlinkSocket):
 
         msg["attrs"].append(["OVS_VPORT_ATTR_TYPE", port_type])
         msg["attrs"].append(["OVS_VPORT_ATTR_NAME", vport_ifname])
-        msg["attrs"].append(["OVS_VPORT_ATTR_UPCALL_PID", [self.pid]])
+        msg["attrs"].append(
+            ["OVS_VPORT_ATTR_UPCALL_PID", [self.upcall_packet.epid]]
+        )
+
+        try:
+            reply = self.nlm_request(
+                msg, msg_type=self.prid, msg_flags=NLM_F_REQUEST | NLM_F_ACK
+            )
+            reply = reply[0]
+        except NetlinkError as ne:
+            if ne.code == errno.EEXIST:
+                reply = None
+            else:
+                raise ne
+        return reply
+
+    def reset_upcall(self, dpindex, vport_ifname, p=None):
+        msg = OvsVport.ovs_vport_msg()
+
+        msg["cmd"] = OVS_VPORT_CMD_SET
+        msg["version"] = OVS_DATAPATH_VERSION
+        msg["reserved"] = 0
+        msg["dpifindex"] = dpindex
+        msg["attrs"].append(["OVS_VPORT_ATTR_NAME", vport_ifname])
+
+        if p == None:
+            p = self.upcall_packet
+        else:
+            self.upcall_packet = p
+
+        msg["attrs"].append(["OVS_VPORT_ATTR_UPCALL_PID", [p.epid]])
 
         try:
             reply = self.nlm_request(
@@ -1176,6 +1263,9 @@ class OvsVport(GenericNetlinkSocket):
                 raise ne
         return reply
 
+    def upcall_handler(self, handler=None):
+        self.upcall_packet.upcall_handler(handler)
+
 
 class OvsFlow(GenericNetlinkSocket):
     class ovs_flow_msg(ovs_dp_msg):
@@ -1305,6 +1395,24 @@ class OvsFlow(GenericNetlinkSocket):
             raise ne
         return rep
 
+    def miss(self, packetmsg):
+        seq = packetmsg["header"]["sequence_number"]
+        keystr = "(none)"
+        key_field = packetmsg.get_attr("OVS_PACKET_ATTR_KEY")
+        if key_field is not None:
+            keystr = key_field.dpstr(None, True)
+
+        pktdata = packetmsg.get_attr("OVS_PACKET_ATTR_PACKET")
+        pktpres = "yes" if pktdata is not None else "no"
+
+        print("MISS upcall[%d/%s]: %s" % (seq, pktpres, keystr), flush=True)
+
+    def execute(self, packetmsg):
+        print("userspace execute command")
+
+    def action(self, packetmsg):
+        print("userspace action command")
+
 
 def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB(), vpl=OvsVport()):
     dp_name = dp_lookup_rep.get_attr("OVS_DP_ATTR_NAME")
@@ -1385,6 +1493,12 @@ def main(argv):
     addifcmd = subparsers.add_parser("add-if")
     addifcmd.add_argument("dpname", help="Datapath Name")
     addifcmd.add_argument("addif", help="Interface name for adding")
+    addifcmd.add_argument(
+        "-u",
+        "--upcall",
+        action="store_true",
+        help="Leave open a reader for upcalls",
+    )
     addifcmd.add_argument(
         "-t",
         "--ptype",
@@ -1406,8 +1520,9 @@ def main(argv):
         if args.verbose > 1:
             logging.basicConfig(level=logging.DEBUG)
 
+    ovspk = OvsPacket()
     ovsdp = OvsDatapath()
-    ovsvp = OvsVport()
+    ovsvp = OvsVport(ovspk)
     ovsflow = OvsFlow()
     ndb = NDB()
 
@@ -1430,11 +1545,13 @@ def main(argv):
                 msg += ":'%s'" % args.showdp
             print(msg)
     elif hasattr(args, "adddp"):
-        rep = ovsdp.create(args.adddp, args.upcall, args.versioning)
+        rep = ovsdp.create(args.adddp, args.upcall, args.versioning, ovspk)
         if rep is None:
             print("DP '%s' already exists" % args.adddp)
         else:
             print("DP '%s' added" % args.adddp)
+        if args.upcall:
+            ovspk.upcall_handler(ovsflow)
     elif hasattr(args, "deldp"):
         ovsdp.destroy(args.deldp)
     elif hasattr(args, "addif"):
@@ -1442,12 +1559,17 @@ def main(argv):
         if rep is None:
             print("DP '%s' not found." % args.dpname)
             return 1
+        dpindex = rep["dpifindex"]
         rep = ovsvp.attach(rep["dpifindex"], args.addif, args.ptype)
         msg = "vport '%s'" % args.addif
         if rep and rep["header"]["error"] is None:
             msg += " added."
         else:
             msg += " failed to add."
+        if args.upcall:
+            if rep is None:
+                rep = ovsvp.reset_upcall(dpindex, args.addif, ovspk)
+            ovsvp.upcall_handler(ovsflow)
     elif hasattr(args, "delif"):
         rep = ovsdp.info(args.dpname, 0)
         if rep is None:
-- 
cgit v1.2.3-70-g09d2


From 7a486c443c89bd949f7a64e0040f704e02710b3c Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 14 Apr 2023 16:08:00 +0200
Subject: mptcp: drop unneeded argument

After commit 3a236aef280e ("mptcp: refactor passive socket initialization"),
every mptcp_pm_fully_established() call is always invoked with a
GFP_ATOMIC argument. We can then drop it.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/options.c  | 2 +-
 net/mptcp/pm.c       | 4 ++--
 net/mptcp/protocol.h | 2 +-
 net/mptcp/subflow.c  | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 355f798d575a..cd3b885c8faa 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -1001,7 +1001,7 @@ check_notify:
 		clear_3rdack_retransmission(ssk);
 		mptcp_pm_subflow_established(msk);
 	} else {
-		mptcp_pm_fully_established(msk, ssk, GFP_ATOMIC);
+		mptcp_pm_fully_established(msk, ssk);
 	}
 	return true;
 
diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 70f0ced3ca86..78c924506e83 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -126,7 +126,7 @@ static bool mptcp_pm_schedule_work(struct mptcp_sock *msk,
 	return true;
 }
 
-void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk, gfp_t gfp)
+void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk)
 {
 	struct mptcp_pm_data *pm = &msk->pm;
 	bool announce = false;
@@ -150,7 +150,7 @@ void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk,
 	spin_unlock_bh(&pm->lock);
 
 	if (announce)
-		mptcp_event(MPTCP_EVENT_ESTABLISHED, msk, ssk, gfp);
+		mptcp_event(MPTCP_EVENT_ESTABLISHED, msk, ssk, GFP_ATOMIC);
 }
 
 void mptcp_pm_connection_closed(struct mptcp_sock *msk)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index e1310bc113be..a9eb0e428a6b 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -782,7 +782,7 @@ bool mptcp_pm_addr_families_match(const struct sock *sk,
 void mptcp_pm_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk);
 void mptcp_pm_nl_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk);
 void mptcp_pm_new_connection(struct mptcp_sock *msk, const struct sock *ssk, int server_side);
-void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk, gfp_t gfp);
+void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk);
 bool mptcp_pm_allow_new_subflow(struct mptcp_sock *msk);
 void mptcp_pm_connection_closed(struct mptcp_sock *msk);
 void mptcp_pm_subflow_established(struct mptcp_sock *msk);
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index f46d8f6c40aa..80bbe96c0694 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -850,7 +850,7 @@ create_child:
 			 */
 			if (mp_opt.suboptions & OPTION_MPTCP_MPC_ACK) {
 				mptcp_subflow_fully_established(ctx, &mp_opt);
-				mptcp_pm_fully_established(owner, child, GFP_ATOMIC);
+				mptcp_pm_fully_established(owner, child);
 				ctx->pm_notified = 1;
 			}
 		} else if (ctx->mp_join) {
-- 
cgit v1.2.3-70-g09d2


From 617612316953093bc859890e405e1b550c27d840 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 14 Apr 2023 16:08:01 +0200
Subject: mptcp: avoid unneeded __mptcp_nmpc_socket() usage

In a few spots, the mptcp code invokes the __mptcp_nmpc_socket() helper
multiple times under the same socket lock scope. Additionally, in such
places, the socket status ensures that there is no MP capable handshake
running.

Under the above condition we can replace the later __mptcp_nmpc_socket()
helper invocation with direct access to the msk->subflow pointer and
better document such access is not supposed to fail with WARN().

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index e6cb36784a68..9cdcfdb44aee 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -3143,7 +3143,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 	struct socket *listener;
 	struct sock *newsk;
 
-	listener = __mptcp_nmpc_socket(msk);
+	listener = msk->subflow;
 	if (WARN_ON_ONCE(!listener)) {
 		*err = -EINVAL;
 		return NULL;
@@ -3363,7 +3363,7 @@ static int mptcp_get_port(struct sock *sk, unsigned short snum)
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	struct socket *ssock;
 
-	ssock = __mptcp_nmpc_socket(msk);
+	ssock = msk->subflow;
 	pr_debug("msk=%p, subflow=%p", msk, ssock);
 	if (WARN_ON_ONCE(!ssock))
 		return -EINVAL;
@@ -3709,7 +3709,10 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock,
 
 	pr_debug("msk=%p", msk);
 
-	ssock = __mptcp_nmpc_socket(msk);
+	/* buggy applications can call accept on socket states other then LISTEN
+	 * but no need to allocate the first subflow just to error out.
+	 */
+	ssock = msk->subflow;
 	if (!ssock)
 		return -EINVAL;
 
-- 
cgit v1.2.3-70-g09d2


From a2702a076e73787db660fb8fd07d26a1a3108358 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 14 Apr 2023 16:08:02 +0200
Subject: mptcp: move fastopen subflow check inside mptcp_sendmsg_fastopen()

So that we can avoid a bunch of check in fastpath. Additionally we
can specialize such check according to the specific fastopen method
- defer_connect vs MSG_FASTOPEN.

The latter bits will simplify the next patches.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 9cdcfdb44aee..22e073b373af 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1662,13 +1662,27 @@ static void mptcp_set_nospace(struct sock *sk)
 
 static int mptcp_disconnect(struct sock *sk, int flags);
 
-static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msghdr *msg,
+static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
 				  size_t len, int *copied_syn)
 {
 	unsigned int saved_flags = msg->msg_flags;
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct sock *ssk;
 	int ret;
 
+	/* on flags based fastopen the mptcp is supposed to create the
+	 * first subflow right now. Otherwise we are in the defer_connect
+	 * path, and the first subflow must be already present.
+	 * Since the defer_connect flag is cleared after the first succsful
+	 * fastopen attempt, no need to check for additional subflow status.
+	 */
+	if (msg->msg_flags & MSG_FASTOPEN && !__mptcp_nmpc_socket(msk))
+		return -EINVAL;
+	if (!msk->first)
+		return -EINVAL;
+
+	ssk = msk->first;
+
 	lock_sock(ssk);
 	msg->msg_flags |= MSG_DONTWAIT;
 	msk->connect_flags = O_NONBLOCK;
@@ -1691,6 +1705,7 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msgh
 	} else if (ret && ret != -EINPROGRESS) {
 		mptcp_disconnect(sk, 0);
 	}
+	inet_sk(sk)->defer_connect = 0;
 
 	return ret;
 }
@@ -1699,7 +1714,6 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	struct page_frag *pfrag;
-	struct socket *ssock;
 	size_t copied = 0;
 	int ret = 0;
 	long timeo;
@@ -1709,12 +1723,10 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	lock_sock(sk);
 
-	ssock = __mptcp_nmpc_socket(msk);
-	if (unlikely(ssock && (inet_sk(ssock->sk)->defer_connect ||
-			       msg->msg_flags & MSG_FASTOPEN))) {
+	if (unlikely(inet_sk(sk)->defer_connect || msg->msg_flags & MSG_FASTOPEN)) {
 		int copied_syn = 0;
 
-		ret = mptcp_sendmsg_fastopen(sk, ssock->sk, msg, len, &copied_syn);
+		ret = mptcp_sendmsg_fastopen(sk, msg, len, &copied_syn);
 		copied += copied_syn;
 		if (ret == -EINPROGRESS && copied_syn > 0)
 			goto out;
-- 
cgit v1.2.3-70-g09d2


From ddb1a072f858704b3555876877ca38c5b103a215 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 14 Apr 2023 16:08:03 +0200
Subject: mptcp: move first subflow allocation at mpc access time

In the long run this will simplify the mptcp code and will
allow for more consistent behavior. Move the first subflow
allocation out of the sock->init ops into the __mptcp_nmpc_socket()
helper.

Since the first subflow creation can now happen after the first
setsockopt() we additionally need to invoke mptcp_sockopt_sync()
on it.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/pm_netlink.c |  4 ++--
 net/mptcp/protocol.c   | 61 ++++++++++++++++++++++++++++++--------------------
 net/mptcp/protocol.h   |  2 +-
 net/mptcp/sockopt.c    | 24 +++++++++++---------
 4 files changed, 54 insertions(+), 37 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 1c42bebca39e..bc343dab5e3f 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1035,8 +1035,8 @@ static int mptcp_pm_nl_create_listen_socket(struct sock *sk,
 	lock_sock(newsk);
 	ssock = __mptcp_nmpc_socket(mptcp_sk(newsk));
 	release_sock(newsk);
-	if (!ssock)
-		return -EINVAL;
+	if (IS_ERR(ssock))
+		return PTR_ERR(ssock);
 
 	mptcp_info2sockaddr(&entry->addr, &addr, entry->addr.family);
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 22e073b373af..a676ac1bb9f1 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -49,18 +49,6 @@ static void __mptcp_check_send_data_fin(struct sock *sk);
 DEFINE_PER_CPU(struct mptcp_delegated_action, mptcp_delegated_actions);
 static struct net_device mptcp_napi_dev;
 
-/* If msk has an initial subflow socket, and the MP_CAPABLE handshake has not
- * completed yet or has failed, return the subflow socket.
- * Otherwise return NULL.
- */
-struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk)
-{
-	if (!msk->subflow || READ_ONCE(msk->can_ack))
-		return NULL;
-
-	return msk->subflow;
-}
-
 /* Returns end sequence number of the receiver's advertised window */
 static u64 mptcp_wnd_end(const struct mptcp_sock *msk)
 {
@@ -116,6 +104,31 @@ static int __mptcp_socket_create(struct mptcp_sock *msk)
 	return 0;
 }
 
+/* If the MPC handshake is not started, returns the first subflow,
+ * eventually allocating it.
+ */
+struct socket *__mptcp_nmpc_socket(struct mptcp_sock *msk)
+{
+	struct sock *sk = (struct sock *)msk;
+	int ret;
+
+	if (!((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
+		return ERR_PTR(-EINVAL);
+
+	if (!msk->subflow) {
+		if (msk->first)
+			return ERR_PTR(-EINVAL);
+
+		ret = __mptcp_socket_create(msk);
+		if (ret)
+			return ERR_PTR(ret);
+
+		mptcp_sockopt_sync(msk, msk->first);
+	}
+
+	return msk->subflow;
+}
+
 static void mptcp_drop(struct sock *sk, struct sk_buff *skb)
 {
 	sk_drops_add(sk, skb);
@@ -1667,6 +1680,7 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
 {
 	unsigned int saved_flags = msg->msg_flags;
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	struct socket *ssock;
 	struct sock *ssk;
 	int ret;
 
@@ -1676,8 +1690,11 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
 	 * Since the defer_connect flag is cleared after the first succsful
 	 * fastopen attempt, no need to check for additional subflow status.
 	 */
-	if (msg->msg_flags & MSG_FASTOPEN && !__mptcp_nmpc_socket(msk))
-		return -EINVAL;
+	if (msg->msg_flags & MSG_FASTOPEN) {
+		ssock = __mptcp_nmpc_socket(msk);
+		if (IS_ERR(ssock))
+			return PTR_ERR(ssock);
+	}
 	if (!msk->first)
 		return -EINVAL;
 
@@ -2740,10 +2757,6 @@ static int mptcp_init_sock(struct sock *sk)
 	if (unlikely(!net->mib.mptcp_statistics) && !mptcp_mib_alloc(net))
 		return -ENOMEM;
 
-	ret = __mptcp_socket_create(mptcp_sk(sk));
-	if (ret)
-		return ret;
-
 	set_bit(SOCK_CUSTOM_SOCKOPT, &sk->sk_socket->flags);
 
 	/* fetch the ca name; do it outside __mptcp_init_sock(), so that clone will
@@ -3563,8 +3576,8 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	int err = -EINVAL;
 
 	ssock = __mptcp_nmpc_socket(msk);
-	if (!ssock)
-		return -EINVAL;
+	if (IS_ERR(ssock))
+		return PTR_ERR(ssock);
 
 	mptcp_token_destroy(msk);
 	inet_sk_state_store(sk, TCP_SYN_SENT);
@@ -3652,8 +3665,8 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 
 	lock_sock(sock->sk);
 	ssock = __mptcp_nmpc_socket(msk);
-	if (!ssock) {
-		err = -EINVAL;
+	if (IS_ERR(ssock)) {
+		err = PTR_ERR(ssock);
 		goto unlock;
 	}
 
@@ -3689,8 +3702,8 @@ static int mptcp_listen(struct socket *sock, int backlog)
 
 	lock_sock(sk);
 	ssock = __mptcp_nmpc_socket(msk);
-	if (!ssock) {
-		err = -EINVAL;
+	if (IS_ERR(ssock)) {
+		err = PTR_ERR(ssock);
 		goto unlock;
 	}
 
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index a9eb0e428a6b..21eda9cd0c52 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -627,7 +627,7 @@ void mptcp_close_ssk(struct sock *sk, struct sock *ssk,
 void __mptcp_subflow_send_ack(struct sock *ssk);
 void mptcp_subflow_reset(struct sock *ssk);
 void mptcp_sock_graft(struct sock *sk, struct socket *parent);
-struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk);
+struct socket *__mptcp_nmpc_socket(struct mptcp_sock *msk);
 bool __mptcp_close(struct sock *sk, long timeout);
 void mptcp_cancel_work(struct sock *sk);
 void mptcp_set_owner_r(struct sk_buff *skb, struct sock *sk);
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c
index b655cebda0f3..d4258869ac48 100644
--- a/net/mptcp/sockopt.c
+++ b/net/mptcp/sockopt.c
@@ -301,9 +301,9 @@ static int mptcp_setsockopt_sol_socket(struct mptcp_sock *msk, int optname,
 	case SO_BINDTOIFINDEX:
 		lock_sock(sk);
 		ssock = __mptcp_nmpc_socket(msk);
-		if (!ssock) {
+		if (IS_ERR(ssock)) {
 			release_sock(sk);
-			return -EINVAL;
+			return PTR_ERR(ssock);
 		}
 
 		ret = sock_setsockopt(ssock, SOL_SOCKET, optname, optval, optlen);
@@ -396,9 +396,9 @@ static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname,
 	case IPV6_FREEBIND:
 		lock_sock(sk);
 		ssock = __mptcp_nmpc_socket(msk);
-		if (!ssock) {
+		if (IS_ERR(ssock)) {
 			release_sock(sk);
-			return -EINVAL;
+			return PTR_ERR(ssock);
 		}
 
 		ret = tcp_setsockopt(ssock->sk, SOL_IPV6, optname, optval, optlen);
@@ -693,9 +693,9 @@ static int mptcp_setsockopt_sol_ip_set_transparent(struct mptcp_sock *msk, int o
 	lock_sock(sk);
 
 	ssock = __mptcp_nmpc_socket(msk);
-	if (!ssock) {
+	if (IS_ERR(ssock)) {
 		release_sock(sk);
-		return -EINVAL;
+		return PTR_ERR(ssock);
 	}
 
 	issk = inet_sk(ssock->sk);
@@ -762,13 +762,15 @@ static int mptcp_setsockopt_first_sf_only(struct mptcp_sock *msk, int level, int
 {
 	struct sock *sk = (struct sock *)msk;
 	struct socket *sock;
-	int ret = -EINVAL;
+	int ret;
 
 	/* Limit to first subflow, before the connection establishment */
 	lock_sock(sk);
 	sock = __mptcp_nmpc_socket(msk);
-	if (!sock)
+	if (IS_ERR(sock)) {
+		ret = PTR_ERR(sock);
 		goto unlock;
+	}
 
 	ret = tcp_setsockopt(sock->sk, level, optname, optval, optlen);
 
@@ -861,7 +863,7 @@ static int mptcp_getsockopt_first_sf_only(struct mptcp_sock *msk, int level, int
 {
 	struct sock *sk = (struct sock *)msk;
 	struct socket *ssock;
-	int ret = -EINVAL;
+	int ret;
 	struct sock *ssk;
 
 	lock_sock(sk);
@@ -872,8 +874,10 @@ static int mptcp_getsockopt_first_sf_only(struct mptcp_sock *msk, int level, int
 	}
 
 	ssock = __mptcp_nmpc_socket(msk);
-	if (!ssock)
+	if (IS_ERR(ssock)) {
+		ret = PTR_ERR(ssock);
 		goto out;
+	}
 
 	ret = tcp_getsockopt(ssock->sk, level, optname, optval, optlen);
 
-- 
cgit v1.2.3-70-g09d2


From 8d547809a5d74aacf022d1df7a508c631088aaec Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 14 Apr 2023 16:08:04 +0200
Subject: mptcp: fastclose msk when cleaning unaccepted sockets

When cleaning up unaccepted mptcp socket still laying inside
the listener queue at listener close time, such sockets will
go through a regular close, waiting for a timeout before
shutting down the subflows.

There is no need to keep the kernel resources in use for
such a possibly long time: short-circuit to fast-close.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/protocol.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index a676ac1bb9f1..1926b81a9538 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2953,10 +2953,13 @@ bool __mptcp_close(struct sock *sk, long timeout)
 		goto cleanup;
 	}
 
-	if (mptcp_check_readable(msk)) {
-		/* the msk has read data, do the MPTCP equivalent of TCP reset */
+	if (mptcp_check_readable(msk) || timeout < 0) {
+		/* If the msk has read data, or the caller explicitly ask it,
+		 * do the MPTCP equivalent of TCP reset, aka MPTCP fastclose
+		 */
 		inet_sk_state_store(sk, TCP_CLOSE);
 		mptcp_do_fastclose(sk);
+		timeout = 0;
 	} else if (mptcp_close_state(sk)) {
 		__mptcp_wr_shutdown(sk);
 	}
-- 
cgit v1.2.3-70-g09d2


From aa5887dca2d236fc50000e27023d4d78dce3af30 Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliang.tang@suse.com>
Date: Fri, 14 Apr 2023 17:47:06 +0200
Subject: mptcp: make userspace_pm_append_new_local_addr static

mptcp_userspace_pm_append_new_local_addr() has always exclusively been
used in pm_userspace.c since its introduction in
commit 4638de5aefe5 ("mptcp: handle local addrs announced by userspace PMs").

So make it static.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/pm_userspace.c | 4 ++--
 net/mptcp/protocol.h     | 2 --
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index a02d3cbf2a1b..27a275805c06 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -25,8 +25,8 @@ void mptcp_free_local_addr_list(struct mptcp_sock *msk)
 	}
 }
 
-int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
-					     struct mptcp_pm_addr_entry *entry)
+static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
+						    struct mptcp_pm_addr_entry *entry)
 {
 	DECLARE_BITMAP(id_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
 	struct mptcp_pm_addr_entry *match = NULL;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 21eda9cd0c52..d91b852c2405 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -830,8 +830,6 @@ int mptcp_pm_remove_subflow(struct mptcp_sock *msk, const struct mptcp_rm_list *
 void mptcp_pm_remove_addrs_and_subflows(struct mptcp_sock *msk,
 					struct list_head *rm_list);
 
-int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
-					     struct mptcp_pm_addr_entry *entry);
 void mptcp_free_local_addr_list(struct mptcp_sock *msk);
 int mptcp_nl_cmd_announce(struct sk_buff *skb, struct genl_info *info);
 int mptcp_nl_cmd_remove(struct sk_buff *skb, struct genl_info *info);
-- 
cgit v1.2.3-70-g09d2


From c3d713409b53c0db59887230341973d9b1f73bd4 Mon Sep 17 00:00:00 2001
From: Matthieu Baerts <matthieu.baerts@tessares.net>
Date: Fri, 14 Apr 2023 17:47:07 +0200
Subject: MAINTAINERS: add git trees for MPTCP

This will help occasional developers to find our git repo without having
to look at our wiki.

Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index b8b275e27cdb..1c09473685b1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -14617,6 +14617,8 @@ L:	mptcp@lists.linux.dev
 S:	Maintained
 W:	https://github.com/multipath-tcp/mptcp_net-next/wiki
 B:	https://github.com/multipath-tcp/mptcp_net-next/issues
+T:	git https://github.com/multipath-tcp/mptcp_net-next.git export-net
+T:	git https://github.com/multipath-tcp/mptcp_net-next.git export
 F:	Documentation/networking/mptcp-sysctl.rst
 F:	include/net/mptcp.h
 F:	include/trace/events/mptcp.h
-- 
cgit v1.2.3-70-g09d2


From ce395d0e3ad585839245a1f6a43cc144617a0163 Mon Sep 17 00:00:00 2001
From: Matthieu Baerts <matthieu.baerts@tessares.net>
Date: Fri, 14 Apr 2023 17:47:08 +0200
Subject: mptcp: remove unused 'remaining' variable

In some functions, 'remaining' variable was given in argument and/or set
but never read.

  net/mptcp/options.c:779:3: warning: Value stored to 'remaining' is never
  read [clang-analyzer-deadcode.DeadStores].

  net/mptcp/options.c:547:3: warning: Value stored to 'remaining' is never
  read [clang-analyzer-deadcode.DeadStores].

The issue has been reported internally by Alibaba CI.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Suggested-by: Mat Martineau <martineau@kernel.org>
Co-developed-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mptcp/options.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index cd3b885c8faa..19a01b6566f1 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -442,7 +442,6 @@ static void clear_3rdack_retransmission(struct sock *sk)
 static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
 					 bool snd_data_fin_enable,
 					 unsigned int *size,
-					 unsigned int remaining,
 					 struct mptcp_out_options *opts)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
@@ -556,7 +555,6 @@ static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow,
 static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 					  bool snd_data_fin_enable,
 					  unsigned int *size,
-					  unsigned int remaining,
 					  struct mptcp_out_options *opts)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
@@ -580,7 +578,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 			opts->ext_copy = *mpext;
 		}
 
-		remaining -= map_size;
 		dss_size = map_size;
 		if (skb && snd_data_fin_enable)
 			mptcp_write_data_fin(subflow, skb, &opts->ext_copy);
@@ -851,9 +848,9 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 	}
 
 	snd_data_fin = mptcp_data_fin_enabled(msk);
-	if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, remaining, opts))
+	if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, opts))
 		ret = true;
-	else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, remaining, opts)) {
+	else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, opts)) {
 		unsigned int mp_fail_size;
 
 		ret = true;
-- 
cgit v1.2.3-70-g09d2


From 0a85264e48b642d360720589fdb837a3643fb9b0 Mon Sep 17 00:00:00 2001
From: Matthieu Baerts <matthieu.baerts@tessares.net>
Date: Fri, 14 Apr 2023 17:47:09 +0200
Subject: selftests: mptcp: remove duplicated entries in usage

mptcp_connect tool was printing some duplicated entries when showing how
to use it: -j -l -r

While at it, I also:

 - moved the very few entries that were not sorted,

 - added -R that was missing since
   commit 8a4b910d005d ("mptcp: selftests: add rcvbuf set option"),

 - removed the -u parameter that has been removed in
   commit f730b65c9d85 ("selftests: mptcp: try to set mptcp ulp mode in different sk states").

No need to backport this, it is just an internal tool used by our
selftests. The help menu is mainly useful for MPTCP kernel devs.

Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/mptcp/mptcp_connect.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c
index b25a31445ded..c7f9ebeebc2c 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_connect.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c
@@ -106,8 +106,8 @@ static struct cfg_sockopt_types cfg_sockopt_types;
 static void die_usage(void)
 {
 	fprintf(stderr, "Usage: mptcp_connect [-6] [-c cmsg] [-f offset] [-i file] [-I num] [-j] [-l] "
-		"[-m mode] [-M mark] [-o option] [-p port] [-P mode] [-j] [-l] [-r num] "
-		"[-s MPTCP|TCP] [-S num] [-r num] [-t num] [-T num] [-u] [-w sec] connect_address\n");
+		"[-m mode] [-M mark] [-o option] [-p port] [-P mode] [-r num] [-R num] "
+		"[-s MPTCP|TCP] [-S num] [-t num] [-T num] [-w sec] connect_address\n");
 	fprintf(stderr, "\t-6 use ipv6\n");
 	fprintf(stderr, "\t-c cmsg -- test cmsg type <cmsg>\n");
 	fprintf(stderr, "\t-f offset -- stop the I/O after receiving and sending the specified amount "
@@ -126,13 +126,13 @@ static void die_usage(void)
 	fprintf(stderr, "\t-p num -- use port num\n");
 	fprintf(stderr,
 		"\t-P [saveWithPeek|saveAfterPeek] -- save data with/after MSG_PEEK form tcp socket\n");
-	fprintf(stderr, "\t-t num -- set poll timeout to num\n");
-	fprintf(stderr, "\t-T num -- set expected runtime to num ms\n");
 	fprintf(stderr, "\t-r num -- enable slow mode, limiting each write to num bytes "
 		"-- for remove addr tests\n");
 	fprintf(stderr, "\t-R num -- set SO_RCVBUF to num\n");
 	fprintf(stderr, "\t-s [MPTCP|TCP] -- use mptcp(default) or tcp sockets\n");
 	fprintf(stderr, "\t-S num -- set SO_SNDBUF to num\n");
+	fprintf(stderr, "\t-t num -- set poll timeout to num\n");
+	fprintf(stderr, "\t-T num -- set expected runtime to num ms\n");
 	fprintf(stderr, "\t-w num -- wait num sec before closing the socket\n");
 	exit(1);
 }
-- 
cgit v1.2.3-70-g09d2


From 0fcd72df8847d3a62eb34a084862157ce0564a94 Mon Sep 17 00:00:00 2001
From: Matthieu Baerts <matthieu.baerts@tessares.net>
Date: Fri, 14 Apr 2023 17:47:10 +0200
Subject: selftests: mptcp: join: fix ShellCheck warnings

Most of the code had an issue according to ShellCheck.

That's mainly due to the fact it incorrectly believes most of the code
was unreachable because it's invoked by variable name, see how the
"tests" array is used.

Once SC2317 has been ignored, three small warnings were still visible:

 - SC2155: Declare and assign separately to avoid masking return values.

 - SC2046: Quote this to prevent word splitting: can be ignored because
   "ip netns pids" can display more than one pid.

 - SC2166: Prefer [ p ] || [ q ] as [ p -o q ] is not well defined.

This probably didn't fix any actual issues but it might help spotting
new interesting warnings reported by ShellCheck as just before,
ShellCheck was reporting issues for most lines making it a bit useless.

Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index fafd19ec7e1f..26310c17b4c6 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -6,6 +6,10 @@
 # address all other issues detected by shellcheck.
 #shellcheck disable=SC2086
 
+# ShellCheck incorrectly believes that most of the code here is unreachable
+# because it's invoked by variable name, see how the "tests" array is used
+#shellcheck disable=SC2317
+
 ret=0
 sin=""
 sinfail=""
@@ -371,8 +375,9 @@ check_transfer()
 
 	local line
 	if [ -n "$bytes" ]; then
+		local out_size
 		# when truncating we must check the size explicitly
-		local out_size=$(wc -c $out | awk '{print $1}')
+		out_size=$(wc -c $out | awk '{print $1}')
 		if [ $out_size -ne $bytes ]; then
 			echo "[ FAIL ] $what output file has wrong size ($out_size, $bytes)"
 			fail_test
@@ -500,6 +505,7 @@ kill_events_pids()
 
 kill_tests_wait()
 {
+	#shellcheck disable=SC2046
 	kill -SIGUSR1 $(ip netns pids $ns2) $(ip netns pids $ns1)
 	wait
 }
@@ -1703,7 +1709,7 @@ chk_subflow_nr()
 
 	cnt1=$(ss -N $ns1 -tOni | grep -c token)
 	cnt2=$(ss -N $ns2 -tOni | grep -c token)
-	if [ "$cnt1" != "$subflow_nr" -o "$cnt2" != "$subflow_nr" ]; then
+	if [ "$cnt1" != "$subflow_nr" ] || [ "$cnt2" != "$subflow_nr" ]; then
 		echo "[fail] got $cnt1:$cnt2 subflows expected $subflow_nr"
 		fail_test
 		dump_stats=1
-- 
cgit v1.2.3-70-g09d2


From bd4b28189469492df2b962d737842c311ce2659c Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 14 Apr 2023 17:21:15 -0400
Subject: sctp: delete the obsolete code for the host name address param

In the latest RFC9260, the Host Name Address param has been deprecated.
For INIT chunk:

  Note 3: An INIT chunk MUST NOT contain the Host Name Address
  parameter.  The receiver of an INIT chunk containing a Host Name
  Address parameter MUST send an ABORT chunk and MAY include an
  "Unresolvable Address" error cause.

For Supported Address Types:

  The value indicating the Host Name Address parameter MUST NOT be
  used when sending this parameter and MUST be ignored when receiving
  this parameter.

Currently Linux SCTP doesn't really support Host Name Address param,
but only saves some flag and print debug info, which actually won't
even be triggered due to the verification in sctp_verify_param().
This patch is to delete those dead code.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  1 -
 net/sctp/sm_make_chunk.c   | 10 +---------
 net/sctp/socket.c          |  2 +-
 3 files changed, 2 insertions(+), 11 deletions(-)

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index a0933efd93c3..070c9458fff4 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1711,7 +1711,6 @@ struct sctp_association {
 		__u16	ecn_capable:1,      /* Can peer do ECN? */
 			ipv4_address:1,     /* Peer understands IPv4 addresses? */
 			ipv6_address:1,     /* Peer understands IPv6 addresses? */
-			hostname_address:1, /* Peer understands DNS addresses? */
 			asconf_capable:1,   /* Does peer support ADDIP? */
 			prsctp_capable:1,   /* Can peer do PR-SCTP? */
 			reconf_capable:1,   /* Can peer do RE-CONFIG? */
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index c7503fd64915..c8f4ec5d5f98 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -2207,7 +2207,7 @@ static enum sctp_ierror sctp_verify_param(struct net *net,
 		break;
 
 	case SCTP_PARAM_HOST_NAME_ADDRESS:
-		/* Tell the peer, we won't support this param.  */
+		/* This param has been Deprecated, send ABORT.  */
 		sctp_process_hn_param(asoc, param, chunk, err_chunk);
 		retval = SCTP_IERROR_ABORT;
 		break;
@@ -2589,10 +2589,6 @@ do_addr_param:
 		asoc->cookie_life = ktime_add_ms(asoc->cookie_life, stale);
 		break;
 
-	case SCTP_PARAM_HOST_NAME_ADDRESS:
-		pr_debug("%s: unimplemented SCTP_HOST_NAME_ADDRESS\n", __func__);
-		break;
-
 	case SCTP_PARAM_SUPPORTED_ADDRESS_TYPES:
 		/* Turn off the default values first so we'll know which
 		 * ones are really set by the peer.
@@ -2624,10 +2620,6 @@ do_addr_param:
 					asoc->peer.ipv6_address = 1;
 				break;
 
-			case SCTP_PARAM_HOST_NAME_ADDRESS:
-				asoc->peer.hostname_address = 1;
-				break;
-
 			default: /* Just ignore anything else.  */
 				break;
 			}
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 218e0982c370..079e726909b4 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -5195,7 +5195,7 @@ int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc,
 	mask = asoc->peer.ecn_capable << 1;
 	mask = (mask | asoc->peer.ipv4_address) << 1;
 	mask = (mask | asoc->peer.ipv6_address) << 1;
-	mask = (mask | asoc->peer.hostname_address) << 1;
+	mask = mask << 1;
 	mask = (mask | asoc->peer.asconf_capable) << 1;
 	mask = (mask | asoc->peer.prsctp_capable) << 1;
 	mask = (mask | asoc->peer.auth_capable);
-- 
cgit v1.2.3-70-g09d2


From ab4f1e28c941b2554456bfe091db0caf25c33044 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Fri, 14 Apr 2023 17:21:16 -0400
Subject: sctp: add intl_capable and reconf_capable in ss peer_capable

There are two new peer capables have been added since sctp_diag was
introduced into SCTP. When dumping the peer capables, these two new
peer capables should also be included. To not break the old capables,
reconf_capable takes the old hostname_address bit, and intl_capable
uses the higher available bit in sctpi_peer_capable.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sctp/socket.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 079e726909b4..cda8c2874691 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -5192,10 +5192,11 @@ int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc,
 	info->sctpi_peer_rwnd = asoc->peer.rwnd;
 	info->sctpi_peer_tag = asoc->c.peer_vtag;
 
-	mask = asoc->peer.ecn_capable << 1;
+	mask = asoc->peer.intl_capable << 1;
+	mask = (mask | asoc->peer.ecn_capable) << 1;
 	mask = (mask | asoc->peer.ipv4_address) << 1;
 	mask = (mask | asoc->peer.ipv6_address) << 1;
-	mask = mask << 1;
+	mask = (mask | asoc->peer.reconf_capable) << 1;
 	mask = (mask | asoc->peer.asconf_capable) << 1;
 	mask = (mask | asoc->peer.prsctp_capable) << 1;
 	mask = (mask | asoc->peer.auth_capable);
-- 
cgit v1.2.3-70-g09d2


From f8f9c31129c4ebb098dba45a76e4ccf65dc7e6c8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:25 +0300
Subject: wifi: iwlwifi: nvm-parse: add full BW UL MU-MIMO support

For some devices, in client mode, we support full bandwidth
uplink multi-user MIMO. Add the necessary capability.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.88b6aac6c876.Ibf2a7a38d7a172c371c347c9d4441e8c656cefe3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 92a79df3e6d2..7dcb1c3ab728 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -938,6 +938,10 @@ iwl_nvm_fixup_sband_iftd(struct iwl_trans *trans,
 		}
 	}
 
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210 && !is_ap)
+		iftype_data->he_cap.he_cap_elem.phy_cap_info[2] |=
+			IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO;
+
 	switch (CSR_HW_RFID_TYPE(trans->hw_rf_id)) {
 	case IWL_CFG_RF_TYPE_GF:
 	case IWL_CFG_RF_TYPE_MR:
-- 
cgit v1.2.3-70-g09d2


From 5af2bb3168db6b0af9988eb25cccf2e3bc4455e2 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Sun, 16 Apr 2023 15:47:26 +0300
Subject: wifi: iwlwifi: call napi_synchronize() before freeing rx/tx queues

When rx/tx queues are being freed, on a different CPU there could be
still rx flow running. Call napi_synchronize() to prevent such a race.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Co-developed-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Link: https://lore.kernel.org/r/20230416154301.5171ee44dcc1.Iff18718540da412e084e7d8266447d40730600ed@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h   |  1 +
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c         | 18 +++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c |  1 +
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c      |  1 +
 4 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index f7e4f868363d..69b95ad5993b 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -497,6 +497,7 @@ int iwl_pcie_rx_stop(struct iwl_trans *trans);
 void iwl_pcie_rx_free(struct iwl_trans *trans);
 void iwl_pcie_free_rbs_pool(struct iwl_trans *trans);
 void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq);
+void iwl_pcie_rx_napi_sync(struct iwl_trans *trans);
 void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
 			    struct iwl_rxq *rxq);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 9c9f87fe8377..0d7890f99a5f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2003-2014, 2018-2022 Intel Corporation
+ * Copyright (C) 2003-2014, 2018-2023 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -1053,6 +1053,22 @@ static int iwl_pcie_napi_poll_msix(struct napi_struct *napi, int budget)
 	return ret;
 }
 
+void iwl_pcie_rx_napi_sync(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int i;
+
+	if (unlikely(!trans_pcie->rxq))
+		return;
+
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
+
+		if (rxq && rxq->napi.poll)
+			napi_synchronize(&rxq->napi);
+	}
+}
+
 static int _iwl_pcie_rx_init(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index 45b63e3f5df3..73b395841ca8 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -156,6 +156,7 @@ void _iwl_trans_pcie_gen2_stop_device(struct iwl_trans *trans)
 	if (test_and_clear_bit(STATUS_DEVICE_ENABLED, &trans->status)) {
 		IWL_DEBUG_INFO(trans,
 			       "DEVICE_ENABLED bit was set and is now cleared\n");
+		iwl_pcie_rx_napi_sync(trans);
 		iwl_txq_gen2_tx_free(trans);
 		iwl_pcie_rx_stop(trans);
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 1b32a4035d88..c5446b82c6c7 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1260,6 +1260,7 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans)
 	if (test_and_clear_bit(STATUS_DEVICE_ENABLED, &trans->status)) {
 		IWL_DEBUG_INFO(trans,
 			       "DEVICE_ENABLED bit was set and is now cleared\n");
+		iwl_pcie_rx_napi_sync(trans);
 		iwl_pcie_tx_stop(trans);
 		iwl_pcie_rx_stop(trans);
 
-- 
cgit v1.2.3-70-g09d2


From 0af637b5719fee09d006e83e1eecd235f7bc62f3 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:27 +0300
Subject: wifi: iwlwifi: mvm: fix getting lowest TX rate for MLO

In iwl_mvm_mac_ctxt_get_lowest_rate() we were still accessing
vif->bss_conf without any multi-link provisions, and also the
info->band, both of which isn't valid in MLO.

Fix the code to look at the correct link. In case of EAPOL
transmissions for the initial 4-way-HS, look up the correct
link here as well, and warn if multiple are active.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.d892f68d3bcd.I7d6927abeea5c3899db225391dbc6a5c77805e80@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 31 +++++++++++++++++++++--
 1 file changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 20db9fc2e61a..389eef453d17 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -874,13 +874,40 @@ u8 iwl_mvm_mac_ctxt_get_lowest_rate(struct iwl_mvm *mvm,
 				    struct ieee80211_tx_info *info,
 				    struct ieee80211_vif *vif)
 {
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct ieee80211_supported_band *sband;
 	unsigned long basic = vif->bss_conf.basic_rates;
 	u16 lowest_cck = IWL_RATE_COUNT, lowest_ofdm = IWL_RATE_COUNT;
+	u32 link_id = u32_get_bits(info->control.flags,
+				   IEEE80211_TX_CTRL_MLO_LINK);
+	u8 band = info->band;
 	u8 rate;
 	u32 i;
 
-	sband = mvm->hw->wiphy->bands[info->band];
+	if (link_id == IEEE80211_LINK_UNSPECIFIED && vif->valid_links) {
+		for (i = 0; i < ARRAY_SIZE(mvmvif->link); i++) {
+			if (!mvmvif->link[i])
+				continue;
+			/* shouldn't do this when >1 link is active */
+			WARN_ON_ONCE(link_id != IEEE80211_LINK_UNSPECIFIED);
+			link_id = i;
+		}
+	}
+
+	if (link_id < IEEE80211_LINK_UNSPECIFIED) {
+		struct ieee80211_bss_conf *link_conf;
+
+		rcu_read_lock();
+		link_conf = rcu_dereference(vif->link_conf[link_id]);
+		if (link_conf) {
+			basic = link_conf->basic_rates;
+			if (link_conf->chandef.chan)
+				band = link_conf->chandef.chan->band;
+		}
+		rcu_read_unlock();
+	}
+
+	sband = mvm->hw->wiphy->bands[band];
 	for_each_set_bit(i, &basic, BITS_PER_LONG) {
 		u16 hw = sband->bitrates[i].hw_value;
 
@@ -892,7 +919,7 @@ u8 iwl_mvm_mac_ctxt_get_lowest_rate(struct iwl_mvm *mvm,
 		}
 	}
 
-	if (info->band == NL80211_BAND_2GHZ && !vif->p2p &&
+	if (band == NL80211_BAND_2GHZ && !vif->p2p &&
 	    vif->type != NL80211_IFTYPE_P2P_DEVICE &&
 	    !(info->flags & IEEE80211_TX_CTL_NO_CCK_RATE)) {
 		if (lowest_cck != IWL_RATE_COUNT)
-- 
cgit v1.2.3-70-g09d2


From 69e1089316271f7651a3a7ce3b3509405f5b2b8a Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Sun, 16 Apr 2023 15:47:28 +0300
Subject: wifi: iwlwifi: mvm: Fix _iwl_mvm_get_scan_type()

The usage of the 'dtim_period' value was wrong, as it is only
a multiplier of the beacon interval, and thus, beacon interval should
also be considered. Fix it.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.e08293d6cace.I25f8cea3189472bd714676ca38b121d7c60fb9d9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 7ac9bdbfe194..38b68827a4bd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -240,8 +240,9 @@ iwl_mvm_scan_type _iwl_mvm_get_scan_type(struct iwl_mvm *mvm,
 		 * set all scan requests as fast-balance scan
 		 */
 		if (vif && vif->type == NL80211_IFTYPE_STATION &&
-		    vif->bss_conf.dtim_period < 220 &&
-		    data.is_dcm_with_p2p_go)
+		    data.is_dcm_with_p2p_go &&
+		    ((vif->bss_conf.beacon_int *
+		      vif->bss_conf.dtim_period) < 220))
 			return IWL_SCAN_TYPE_FAST_BALANCE;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From bb7fcb37c9004ce7e296f7bb4e56ce65119b83b4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:29 +0300
Subject: wifi: iwlwifi: mvm: properly implement HE AP support

The firmware split the HE support field into HE and "pseudo HE",
the latter is really for AP and doesn't implement trigger frame
handling for example.

Use the new field for AP mode.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.34dbfefe2a49.I0e39cd35dbe03ff9209b26733746479eae1c8966@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h |  4 +++-
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c    | 14 ++++++++++++--
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index e236d1b0aae2..0bafa3e21cb5 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -278,6 +278,7 @@ enum iwl_mac_config_filter_flags {
  * @reserved_for_local_mld_addr: reserved
  * @filter_flags: combination of &enum iwl_mac_config_filter_flags
  * @he_support: does this MAC support HE
+ * @he_ap_support: HE AP enabled, "pseudo HE", no trigger frame handling
  * @eht_support: does this MAC support EHT. Requires he_support
  * @nic_not_ack_enabled: mark that the NIC doesn't support receiving
  *	ACK-enabled AGG, (i.e. both BACK and non-BACK frames in single AGG).
@@ -296,7 +297,8 @@ struct iwl_mac_config_cmd {
 	u8 local_mld_addr[6];
 	__le16 reserved_for_local_mld_addr;
 	__le32 filter_flags;
-	__le32 he_support;
+	__le16 he_support;
+	__le16 he_ap_support;
 	__le32 eht_support;
 	__le32 nic_not_ack_enabled;
 	/* MAC_CONTEXT_CONFIG_SPECIFIC_DATA_API_U_VER_1 */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
index ab0ba85936b4..d73ab1712e0e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
@@ -4,6 +4,16 @@
  */
 #include "mvm.h"
 
+static void iwl_mvm_mld_set_he_support(struct iwl_mvm *mvm,
+				       struct ieee80211_vif *vif,
+				       struct iwl_mac_config_cmd *cmd)
+{
+	if (vif->type == NL80211_IFTYPE_AP)
+		cmd->he_ap_support = cpu_to_le16(1);
+	else
+		cmd->he_support = cpu_to_le16(1);
+}
+
 static void iwl_mvm_mld_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 					    struct ieee80211_vif *vif,
 					    struct iwl_mac_config_cmd *cmd,
@@ -41,7 +51,7 @@ static void iwl_mvm_mld_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 	 * and enable both when we have MLO.
 	 */
 	if (vif->valid_links) {
-		cmd->he_support = cpu_to_le32(1);
+		iwl_mvm_mld_set_he_support(mvm, vif, cmd);
 		cmd->eht_support = cpu_to_le32(1);
 		return;
 	}
@@ -53,7 +63,7 @@ static void iwl_mvm_mld_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 			continue;
 
 		if (link_conf->he_support)
-			cmd->he_support = cpu_to_le32(1);
+			iwl_mvm_mld_set_he_support(mvm, vif, cmd);
 
 		/* it's not reasonable to have EHT without HE and FW API doesn't
 		 * support it. Ignore EHT in this case.
-- 
cgit v1.2.3-70-g09d2


From 66a588bff29e61345df6d281c56d9864b84eb8cc Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:30 +0300
Subject: wifi: iwlwifi: mvm: factor out iwl_mvm_sta_fw_id_mask()

We are going to need this in more places than just the
key code, so factor out the functionality of getting
the FW station ID mask (filtered to a specific link if
needed) to a separate function that can now be called
both under RCU and mvm->mutex protection.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.eff92b93025d.I2c50290a0537d5db3d3460f4d57c78a4712ffb75@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c | 33 ++-------------------
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 37 ++++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h     |  2 ++
 3 files changed, 42 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index 7c417b39aca4..a8ab35473924 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -15,10 +15,6 @@ static u32 iwl_mvm_get_sec_sta_mask(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_vif_link_info *link_info = &mvmvif->deflink;
-	struct iwl_mvm_link_sta *link_sta;
-	struct iwl_mvm_sta *mvmsta;
-	u32 result = 0;
-	int link_id;
 
 	lockdep_assert_held(&mvm->mutex);
 
@@ -49,33 +45,10 @@ static u32 iwl_mvm_get_sec_sta_mask(struct iwl_mvm *mvm,
 	if (!sta && (keyconf->link_id >= 0 || !vif->valid_links))
 		return BIT(link_info->ap_sta_id);
 
-	/* this shouldn't happen now */
-	if (!sta)
-		return 0;
+	/* STA should be non-NULL now, but iwl_mvm_sta_fw_id_mask() checks */
 
-	mvmsta = iwl_mvm_sta_from_mac80211(sta);
-
-	/* it's easy when the STA is not an MLD */
-	if (!sta->valid_links)
-		return BIT(mvmsta->deflink.sta_id);
-
-	/* but if it is an MLD, get the mask of all the FW STAs it has ... */
-	for (link_id = 0; link_id < ARRAY_SIZE(mvmsta->link); link_id++) {
-		/* unless we have a specific link in mind (GTK on client) */
-		if (keyconf->link_id >= 0 &&
-		    keyconf->link_id != link_id)
-			continue;
-
-		link_sta =
-			rcu_dereference_protected(mvmsta->link[link_id],
-						  lockdep_is_held(&mvm->mutex));
-		if (!link_sta)
-			continue;
-
-		result |= BIT(link_sta->sta_id);
-	}
-
-	return result;
+	/* pass link_id to filter by it if not -1 (GTK on client) */
+	return iwl_mvm_sta_fw_id_mask(mvm, sta, keyconf->link_id);
 }
 
 static u32 iwl_mvm_get_sec_flags(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index ad71233a2299..c5d25772159b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -4,6 +4,43 @@
  */
 #include "mvm.h"
 #include "time-sync.h"
+#include "sta.h"
+
+u32 iwl_mvm_sta_fw_id_mask(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			   int filter_link_id)
+{
+	struct iwl_mvm_sta *mvmsta;
+	unsigned int link_id;
+	u32 result = 0;
+
+	if (!sta)
+		return 0;
+
+	mvmsta = iwl_mvm_sta_from_mac80211(sta);
+
+	/* it's easy when the STA is not an MLD */
+	if (!sta->valid_links)
+		return BIT(mvmsta->deflink.sta_id);
+
+	/* but if it is an MLD, get the mask of all the FW STAs it has ... */
+	for (link_id = 0; link_id < ARRAY_SIZE(mvmsta->link); link_id++) {
+		struct iwl_mvm_link_sta *link_sta;
+
+		/* unless we have a specific link in mind */
+		if (filter_link_id >= 0 && link_id != filter_link_id)
+			continue;
+
+		link_sta =
+			rcu_dereference_check(mvmsta->link[link_id],
+					      lockdep_is_held(&mvm->mutex));
+		if (!link_sta)
+			continue;
+
+		result |= BIT(link_sta->sta_id);
+	}
+
+	return result;
+}
 
 static int iwl_mvm_mld_send_sta_cmd(struct iwl_mvm *mvm,
 				    struct iwl_mvm_sta_cfg_cmd *cmd)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 7b9e91935aa0..698b9c014cd3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -642,6 +642,8 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif,
 				 struct ieee80211_sta *sta,
 				 u16 old_links, u16 new_links);
+u32 iwl_mvm_sta_fw_id_mask(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			   int filter_link_id);
 
 /* Queues */
 void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
-- 
cgit v1.2.3-70-g09d2


From 85eb75c34eadb892d89e506619c6c49b194a392b Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Sun, 16 Apr 2023 15:47:31 +0300
Subject: wifi: iwlwifi: mvm: avoid iterating over an un-initialized list

The initiator smooth list is initialized only in iwl_mvm_up(), but is
cleared in iwl_mvm_mac_stop. This may result in iterating over the
list before it was initialized in case iwl_mvm_up() failed early.
Fix it by moving the list initialization to an earlier stage.

Fixes: b68bd2e3143a ("iwlwifi: mvm: Add FTM initiator RTT smoothing logic")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.b50b1fe9a576.Ie348ffae110612d2e252ac120a3ba0aea063b1b6@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c  | 2 --
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 2 ++
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 3f5cc6bf3f1c..b35c96cf7ad2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1727,8 +1727,6 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	iwl_mvm_tas_init(mvm);
 	iwl_mvm_leds_sync(mvm);
 
-	iwl_mvm_ftm_initiator_smooth_config(mvm);
-
 	if (fw_has_capa(&mvm->fw->ucode_capa,
 			IWL_UCODE_TLV_CAPA_RFIM_SUPPORT)) {
 		if (iwl_mvm_eval_dsm_rfi(mvm) == DSM_VALUE_RFI_ENABLE)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 6f082059025a..8bb865e708dc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1367,6 +1367,8 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	else
 		memset(&mvm->rx_stats, 0, sizeof(struct mvm_statistics_rx));
 
+	iwl_mvm_ftm_initiator_smooth_config(mvm);
+
 	iwl_mvm_init_time_sync(&mvm->time_sync);
 
 	mvm->debugfs_dir = dbgfs_dir;
-- 
cgit v1.2.3-70-g09d2


From 6f2c5f38a7917a04a43494eda48aa71da97484e3 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:32 +0300
Subject: wifi: iwlwifi: mvm: use correct sta mask to remove queue

When we remove a queue we need to use the currently active
firmware stations in the mask, not the deflink one. Fix that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.01cdd2153418.I176d54f2d869f51b3707d056adb96455cf885f93@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index c5d25772159b..819e05a6f651 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -299,7 +299,7 @@ int iwl_mvm_mld_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id)
 				       IWL_MAX_TID_COUNT, NULL);
 }
 
-static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm, int sta_id,
+static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm, u32 sta_mask,
 				   u16 *queueptr, u8 tid)
 {
 	int queue = *queueptr;
@@ -314,7 +314,7 @@ static int iwl_mvm_mld_disable_txq(struct iwl_mvm *mvm, int sta_id,
 		struct iwl_scd_queue_cfg_cmd remove_cmd = {
 			.operation = cpu_to_le32(IWL_SCD_QUEUE_REMOVE),
 			.u.remove.tid = cpu_to_le32(tid),
-			.u.remove.sta_mask = cpu_to_le32(BIT(sta_id)),
+			.u.remove.sta_mask = cpu_to_le32(sta_mask),
 		};
 
 		ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0,
@@ -344,7 +344,7 @@ static int iwl_mvm_mld_rm_int_sta(struct iwl_mvm *mvm,
 	if (flush)
 		iwl_mvm_flush_sta(mvm, int_sta, true);
 
-	iwl_mvm_mld_disable_txq(mvm, int_sta->sta_id, queuptr, tid);
+	iwl_mvm_mld_disable_txq(mvm, BIT(int_sta->sta_id), queuptr, tid);
 
 	ret = iwl_mvm_mld_rm_sta_from_fw(mvm, int_sta->sta_id);
 	if (ret)
@@ -760,6 +760,7 @@ static void iwl_mvm_mld_disable_sta_queues(struct iwl_mvm *mvm,
 					   struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	u32 sta_mask = iwl_mvm_sta_fw_id_mask(mvm, sta, -1);
 	int i;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -768,7 +769,7 @@ static void iwl_mvm_mld_disable_sta_queues(struct iwl_mvm *mvm,
 		if (mvm_sta->tid_data[i].txq_id == IWL_MVM_INVALID_QUEUE)
 			continue;
 
-		iwl_mvm_mld_disable_txq(mvm, mvm_sta->deflink.sta_id,
+		iwl_mvm_mld_disable_txq(mvm, sta_mask,
 					&mvm_sta->tid_data[i].txq_id, i);
 		mvm_sta->tid_data[i].txq_id = IWL_MVM_INVALID_QUEUE;
 	}
-- 
cgit v1.2.3-70-g09d2


From ef3ed33dfc8f0f1c81ca103e6b68b4f77ee0ab65 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Sun, 16 Apr 2023 15:47:33 +0300
Subject: wifi: iwlwifi: bump FW API to 77 for AX devices

Start supporting API version 77 for AX devices.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.e522ccefe354.If7628363fafeb7687163103e734206915c445197@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 032928631ddc..b98f0ff02362 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_22000_UCODE_API_MAX	75
+#define IWL_22000_UCODE_API_MAX	77
 
 /* Lowest firmware API version supported */
 #define IWL_22000_UCODE_API_MIN	39
-- 
cgit v1.2.3-70-g09d2


From 9aa3856d9b6f3f8bf085b5bbb5f3703578bf452b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:34 +0300
Subject: wifi: iwlwifi: mvm: track station mask for BAIDs

When we have MLO connections, a BAID applies to multiple
firmware stations. Track the station mask instead of the
station ID, getting rid of a few more deflink cases and
preparing for handling link switching for BAIDs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.c08523808c34.I719b7bba499648d1495ed3e3a90889d4732ef15d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h  |  4 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 34 +++++++++++++++++----------
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c  | 23 ++++++++++--------
 3 files changed, 36 insertions(+), 25 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 7a8b4be8dcf2..29a87d1ba343 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -680,7 +680,7 @@ __aligned(roundup_pow_of_two(sizeof(struct _iwl_mvm_reorder_buf_entry)))
 
 /**
  * struct iwl_mvm_baid_data - BA session data
- * @sta_id: station id
+ * @sta_mask: current station mask for the BAID
  * @tid: tid of the session
  * @baid baid of the session
  * @timeout: the timeout set in the addba request
@@ -694,7 +694,7 @@ __aligned(roundup_pow_of_two(sizeof(struct _iwl_mvm_reorder_buf_entry)))
  */
 struct iwl_mvm_baid_data {
 	struct rcu_head rcu_head;
-	u8 sta_id;
+	u32 sta_mask;
 	u8 tid;
 	u8 baid;
 	u16 timeout;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 31e1d0ccdec5..9dbf14fa0ca7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -705,7 +705,7 @@ void iwl_mvm_reorder_timer_expired(struct timer_list *t)
 	if (expired) {
 		struct ieee80211_sta *sta;
 		struct iwl_mvm_sta *mvmsta;
-		u8 sta_id = baid_data->sta_id;
+		u8 sta_id = ffs(baid_data->sta_mask) - 1;
 
 		rcu_read_lock();
 		sta = rcu_dereference(buf->mvm->fw_id_to_mac_id[sta_id]);
@@ -740,6 +740,7 @@ static void iwl_mvm_del_ba(struct iwl_mvm *mvm, int queue,
 	struct ieee80211_sta *sta;
 	struct iwl_mvm_reorder_buffer *reorder_buf;
 	u8 baid = data->baid;
+	u32 sta_id;
 
 	if (WARN_ONCE(baid >= IWL_MAX_BAID, "invalid BAID: %x\n", baid))
 		return;
@@ -750,7 +751,9 @@ static void iwl_mvm_del_ba(struct iwl_mvm *mvm, int queue,
 	if (WARN_ON_ONCE(!ba_data))
 		goto out;
 
-	sta = rcu_dereference(mvm->fw_id_to_mac_id[ba_data->sta_id]);
+	/* pick any STA ID to find the pointer */
+	sta_id = ffs(ba_data->sta_mask) - 1;
+	sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]);
 	if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta)))
 		goto out;
 
@@ -777,6 +780,7 @@ static void iwl_mvm_release_frames_from_notif(struct iwl_mvm *mvm,
 	struct ieee80211_sta *sta;
 	struct iwl_mvm_reorder_buffer *reorder_buf;
 	struct iwl_mvm_baid_data *ba_data;
+	u32 sta_id;
 
 	IWL_DEBUG_HT(mvm, "Frame release notification for BAID %u, NSSN %d\n",
 		     baid, nssn);
@@ -794,7 +798,9 @@ static void iwl_mvm_release_frames_from_notif(struct iwl_mvm *mvm,
 		goto out;
 	}
 
-	sta = rcu_dereference(mvm->fw_id_to_mac_id[ba_data->sta_id]);
+	/* pick any STA ID to find the pointer */
+	sta_id = ffs(ba_data->sta_mask) - 1;
+	sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]);
 	if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta)))
 		goto out;
 
@@ -935,7 +941,6 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
 {
 	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
 	struct ieee80211_hdr *hdr = (void *)skb_mac_header(skb);
-	struct iwl_mvm_sta *mvm_sta;
 	struct iwl_mvm_baid_data *baid_data;
 	struct iwl_mvm_reorder_buffer *buffer;
 	struct sk_buff *tail;
@@ -947,6 +952,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
 	u8 sub_frame_idx = desc->amsdu_info &
 			   IWL_RX_MPDU_AMSDU_SUBFRAME_IDX_MASK;
 	struct iwl_mvm_reorder_buf_entry *entries;
+	u32 sta_mask;
 	int index;
 	u16 nssn, sn;
 	u8 baid;
@@ -969,8 +975,6 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
 		      "Got valid BAID without a valid station assigned\n"))
 		return false;
 
-	mvm_sta = iwl_mvm_sta_from_mac80211(sta);
-
 	/* not a data packet or a bar */
 	if (!ieee80211_is_back_req(hdr->frame_control) &&
 	    (!ieee80211_is_data_qos(hdr->frame_control) ||
@@ -988,11 +992,14 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
 		return false;
 	}
 
+	rcu_read_lock();
+	sta_mask = iwl_mvm_sta_fw_id_mask(mvm, sta, -1);
+	rcu_read_unlock();
+
 	if (WARN(tid != baid_data->tid ||
-		 mvm_sta->deflink.sta_id != baid_data->sta_id,
-		 "baid 0x%x is mapped to sta:%d tid:%d, but was received for sta:%d tid:%d\n",
-		 baid, baid_data->sta_id, baid_data->tid, mvm_sta->deflink.sta_id,
-		 tid))
+		 !(sta_mask & baid_data->sta_mask),
+		 "baid 0x%x is mapped to sta_mask:0x%x tid:%d, but was received for sta_mask:0x%x tid:%d\n",
+		 baid, baid_data->sta_mask, baid_data->tid, sta_mask, tid))
 		return false;
 
 	nssn = reorder & IWL_RX_MPDU_REORDER_NSSN_MASK;
@@ -2776,9 +2783,10 @@ void iwl_mvm_rx_bar_frame_release(struct iwl_mvm *mvm, struct napi_struct *napi,
 		goto out;
 	}
 
-	if (WARN(tid != baid_data->tid || sta_id != baid_data->sta_id,
-		 "baid 0x%x is mapped to sta:%d tid:%d, but BAR release received for sta:%d tid:%d\n",
-		 baid, baid_data->sta_id, baid_data->tid, sta_id,
+	if (WARN(tid != baid_data->tid || sta_id > IWL_MVM_STATION_COUNT_MAX ||
+		 !(baid_data->sta_mask & BIT(sta_id)),
+		 "baid 0x%x is mapped to sta_mask:0x%x tid:%d, but BAR release received for sta:%d tid:%d\n",
+		 baid, baid_data->sta_mask, baid_data->tid, sta_id,
 		 tid))
 		goto out;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 3e5bed885bd9..d323d119c334 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -251,6 +251,7 @@ static void iwl_mvm_rx_agg_session_expired(struct timer_list *t)
 	struct ieee80211_sta *sta;
 	struct iwl_mvm_sta *mvm_sta;
 	unsigned long timeout;
+	unsigned int sta_id;
 
 	rcu_read_lock();
 
@@ -269,7 +270,8 @@ static void iwl_mvm_rx_agg_session_expired(struct timer_list *t)
 	}
 
 	/* Timer expired */
-	sta = rcu_dereference(ba_data->mvm->fw_id_to_mac_id[ba_data->sta_id]);
+	sta_id = ffs(ba_data->sta_mask) - 1; /* don't care which one */
+	sta = rcu_dereference(ba_data->mvm->fw_id_to_mac_id[sta_id]);
 
 	/*
 	 * sta should be valid unless the following happens:
@@ -2756,10 +2758,11 @@ static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
 }
 
 static int iwl_mvm_fw_baid_op_sta(struct iwl_mvm *mvm,
-				  struct iwl_mvm_sta *mvm_sta,
+				  struct ieee80211_sta *sta,
 				  bool start, int tid, u16 ssn,
 				  u16 buf_size)
 {
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_add_sta_cmd cmd = {
 		.mac_id_n_color = cpu_to_le32(mvm_sta->mac_id_n_color),
 		.sta_id = mvm_sta->deflink.sta_id,
@@ -2804,7 +2807,7 @@ static int iwl_mvm_fw_baid_op_sta(struct iwl_mvm *mvm,
 }
 
 static int iwl_mvm_fw_baid_op_cmd(struct iwl_mvm *mvm,
-				  struct iwl_mvm_sta *mvm_sta,
+				  struct ieee80211_sta *sta,
 				  bool start, int tid, u16 ssn,
 				  u16 buf_size, int baid)
 {
@@ -2819,7 +2822,7 @@ static int iwl_mvm_fw_baid_op_cmd(struct iwl_mvm *mvm,
 
 	if (start) {
 		cmd.alloc.sta_id_mask =
-			cpu_to_le32(BIT(mvm_sta->deflink.sta_id));
+			cpu_to_le32(iwl_mvm_sta_fw_id_mask(mvm, sta, -1));
 		cmd.alloc.tid = tid;
 		cmd.alloc.ssn = cpu_to_le16(ssn);
 		cmd.alloc.win_size = cpu_to_le16(buf_size);
@@ -2829,7 +2832,7 @@ static int iwl_mvm_fw_baid_op_cmd(struct iwl_mvm *mvm,
 		BUILD_BUG_ON(sizeof(cmd.remove_v1) > sizeof(cmd.remove));
 	} else {
 		cmd.remove.sta_id_mask =
-			cpu_to_le32(BIT(mvm_sta->deflink.sta_id));
+			cpu_to_le32(iwl_mvm_sta_fw_id_mask(mvm, sta, -1));
 		cmd.remove.tid = cpu_to_le32(tid);
 	}
 
@@ -2852,16 +2855,16 @@ static int iwl_mvm_fw_baid_op_cmd(struct iwl_mvm *mvm,
 	return baid;
 }
 
-static int iwl_mvm_fw_baid_op(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvm_sta,
+static int iwl_mvm_fw_baid_op(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			      bool start, int tid, u16 ssn, u16 buf_size,
 			      int baid)
 {
 	if (fw_has_capa(&mvm->fw->ucode_capa,
 			IWL_UCODE_TLV_CAPA_BAID_ML_SUPPORT))
-		return iwl_mvm_fw_baid_op_cmd(mvm, mvm_sta, start,
+		return iwl_mvm_fw_baid_op_cmd(mvm, sta, start,
 					      tid, ssn, buf_size, baid);
 
-	return iwl_mvm_fw_baid_op_sta(mvm, mvm_sta, start,
+	return iwl_mvm_fw_baid_op_sta(mvm, sta, start,
 				      tid, ssn, buf_size);
 }
 
@@ -2931,7 +2934,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 
 	/* Don't send command to remove (start=0) BAID during restart */
 	if (start || !test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status))
-		baid = iwl_mvm_fw_baid_op(mvm, mvm_sta, start, tid, ssn, buf_size,
+		baid = iwl_mvm_fw_baid_op(mvm, sta, start, tid, ssn, buf_size,
 					  baid);
 
 	if (baid < 0) {
@@ -2953,7 +2956,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			    iwl_mvm_rx_agg_session_expired, 0);
 		baid_data->mvm = mvm;
 		baid_data->tid = tid;
-		baid_data->sta_id = mvm_sta->deflink.sta_id;
+		baid_data->sta_mask = iwl_mvm_sta_fw_id_mask(mvm, sta, -1);
 
 		mvm_sta->tid_to_baid[tid] = baid;
 		if (timeout)
-- 
cgit v1.2.3-70-g09d2


From 7a243c6b680694caca0d0d330eeb17ef8c2b2cd2 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:35 +0300
Subject: wifi: iwlwifi: mvm: implement BAID link switching

When we switch station links, also add the code to switch
BAIDs from one station mask to the new one.

To do so, refactor the switching code a bit to have common
code for all the needed switches; will add keys next.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.40654afce24f.I0e35151f69e7513be53ddb8f008e9ab48278c352@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 80 +++++++++++++++++++++---
 1 file changed, 70 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 819e05a6f651..36d83e22f0af 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -911,11 +911,12 @@ void iwl_mvm_mld_modify_all_sta_disable_tx(struct iwl_mvm *mvm,
 	rcu_read_unlock();
 }
 
-static int iwl_mvm_mld_update_sta_queue(struct iwl_mvm *mvm,
-					struct iwl_mvm_sta *mvm_sta,
-					u32 old_sta_mask,
-					u32 new_sta_mask)
+static int iwl_mvm_mld_update_sta_queues(struct iwl_mvm *mvm,
+					 struct ieee80211_sta *sta,
+					 u32 old_sta_mask,
+					 u32 new_sta_mask)
 {
+	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_scd_queue_cfg_cmd cmd = {
 		.operation = cpu_to_le32(IWL_SCD_QUEUE_MODIFY),
 		.u.modify.old_sta_mask = cpu_to_le32(old_sta_mask),
@@ -951,6 +952,63 @@ static int iwl_mvm_mld_update_sta_queue(struct iwl_mvm *mvm,
 	return 0;
 }
 
+static int iwl_mvm_mld_update_sta_baids(struct iwl_mvm *mvm,
+					u32 old_sta_mask,
+					u32 new_sta_mask)
+{
+	struct iwl_rx_baid_cfg_cmd cmd = {
+		.action = cpu_to_le32(IWL_RX_BAID_ACTION_MODIFY),
+		.modify.old_sta_id_mask = cpu_to_le32(old_sta_mask),
+		.modify.new_sta_id_mask = cpu_to_le32(new_sta_mask),
+	};
+	u32 cmd_id = WIDE_ID(DATA_PATH_GROUP, RX_BAID_ALLOCATION_CONFIG_CMD);
+	int baid;
+
+	BUILD_BUG_ON(sizeof(struct iwl_rx_baid_cfg_resp) != sizeof(baid));
+
+	for (baid = 0; baid < ARRAY_SIZE(mvm->baid_map); baid++) {
+		struct iwl_mvm_baid_data *data;
+		int ret;
+
+		data = rcu_dereference_protected(mvm->baid_map[baid],
+						 lockdep_is_held(&mvm->mutex));
+		if (!data)
+			continue;
+
+		if (!(data->sta_mask & old_sta_mask))
+			continue;
+
+		WARN_ONCE(data->sta_mask != old_sta_mask,
+			  "BAID data for %d corrupted - expected 0x%x found 0x%x\n",
+			  baid, old_sta_mask, data->sta_mask);
+
+		cmd.modify.tid = cpu_to_le32(data->tid);
+
+		ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0, sizeof(cmd), &cmd);
+		data->sta_mask = new_sta_mask;
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int iwl_mvm_mld_update_sta_resources(struct iwl_mvm *mvm,
+					    struct ieee80211_sta *sta,
+					    u32 old_sta_mask,
+					    u32 new_sta_mask)
+{
+	int ret;
+
+	ret = iwl_mvm_mld_update_sta_queues(mvm, sta,
+					    old_sta_mask,
+					    new_sta_mask);
+	if (ret)
+		return ret;
+
+	return iwl_mvm_mld_update_sta_baids(mvm, old_sta_mask, new_sta_mask);
+}
+
 int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif,
 				 struct ieee80211_sta *sta,
@@ -987,9 +1045,10 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 	}
 
 	if (sta_mask_to_rem) {
-		ret = iwl_mvm_mld_update_sta_queue(mvm, mvm_sta,
-						   current_sta_mask,
-						   current_sta_mask & ~sta_mask_to_rem);
+		ret = iwl_mvm_mld_update_sta_resources(mvm, sta,
+						       current_sta_mask,
+						       current_sta_mask &
+							~sta_mask_to_rem);
 		if (WARN_ON(ret))
 			goto err;
 
@@ -1064,9 +1123,10 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 	}
 
 	if (sta_mask_added) {
-		ret = iwl_mvm_mld_update_sta_queue(mvm, mvm_sta,
-						   current_sta_mask,
-						   current_sta_mask | sta_mask_added);
+		ret = iwl_mvm_mld_update_sta_resources(mvm, sta,
+						       current_sta_mask,
+						       current_sta_mask |
+							sta_mask_added);
 		if (WARN_ON(ret))
 			goto err;
 	}
-- 
cgit v1.2.3-70-g09d2


From 8642ddb2a363cab424fd6975165007f308de75ba Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:36 +0300
Subject: wifi: iwlwifi: mvm: implement key link switching

Implement switching keys from one set of firmware station IDs
to another set, during link switch.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.c6a777dd5e47.I693f7fd7c52fe8b51a58af69d45488511367f49e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c | 53 ++++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 11 ++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h     |  5 +++
 3 files changed, 67 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
index a8ab35473924..8853821b3716 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-key.c
@@ -97,6 +97,59 @@ static u32 iwl_mvm_get_sec_flags(struct iwl_mvm *mvm,
 	return flags;
 }
 
+struct iwl_mvm_sta_key_update_data {
+	struct ieee80211_sta *sta;
+	u32 old_sta_mask;
+	u32 new_sta_mask;
+	int err;
+};
+
+static void iwl_mvm_mld_update_sta_key(struct ieee80211_hw *hw,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_sta *sta,
+				       struct ieee80211_key_conf *key,
+				       void *_data)
+{
+	u32 cmd_id = WIDE_ID(DATA_PATH_GROUP, SEC_KEY_CMD);
+	struct iwl_mvm_sta_key_update_data *data = _data;
+	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct iwl_sec_key_cmd cmd = {
+		.action = cpu_to_le32(FW_CTXT_ACTION_MODIFY),
+		.u.modify.old_sta_mask = cpu_to_le32(data->old_sta_mask),
+		.u.modify.new_sta_mask = cpu_to_le32(data->new_sta_mask),
+		.u.modify.key_id = cpu_to_le32(key->keyidx),
+		.u.modify.key_flags =
+			cpu_to_le32(iwl_mvm_get_sec_flags(mvm, vif, sta, key)),
+	};
+	int err;
+
+	/* only need to do this for pairwise keys (link_id == -1) */
+	if (sta != data->sta || key->link_id >= 0)
+		return;
+
+	err = iwl_mvm_send_cmd_pdu(mvm, cmd_id, CMD_ASYNC, sizeof(cmd), &cmd);
+
+	if (err)
+		data->err = err;
+}
+
+int iwl_mvm_mld_update_sta_keys(struct iwl_mvm *mvm,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta,
+				u32 old_sta_mask,
+				u32 new_sta_mask)
+{
+	struct iwl_mvm_sta_key_update_data data = {
+		.sta = sta,
+		.old_sta_mask = old_sta_mask,
+		.new_sta_mask = new_sta_mask,
+	};
+
+	ieee80211_iter_keys_rcu(mvm->hw, vif, iwl_mvm_mld_update_sta_key,
+				&data);
+	return data.err;
+}
+
 static int __iwl_mvm_sec_key_del(struct iwl_mvm *mvm, u32 sta_mask,
 				 u32 key_flags, u32 keyidx, u32 flags)
 {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 36d83e22f0af..65436736f87f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -994,6 +994,7 @@ static int iwl_mvm_mld_update_sta_baids(struct iwl_mvm *mvm,
 }
 
 static int iwl_mvm_mld_update_sta_resources(struct iwl_mvm *mvm,
+					    struct ieee80211_vif *vif,
 					    struct ieee80211_sta *sta,
 					    u32 old_sta_mask,
 					    u32 new_sta_mask)
@@ -1006,6 +1007,12 @@ static int iwl_mvm_mld_update_sta_resources(struct iwl_mvm *mvm,
 	if (ret)
 		return ret;
 
+	ret = iwl_mvm_mld_update_sta_keys(mvm, vif, sta,
+					  old_sta_mask,
+					  new_sta_mask);
+	if (ret)
+		return ret;
+
 	return iwl_mvm_mld_update_sta_baids(mvm, old_sta_mask, new_sta_mask);
 }
 
@@ -1045,7 +1052,7 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 	}
 
 	if (sta_mask_to_rem) {
-		ret = iwl_mvm_mld_update_sta_resources(mvm, sta,
+		ret = iwl_mvm_mld_update_sta_resources(mvm, vif, sta,
 						       current_sta_mask,
 						       current_sta_mask &
 							~sta_mask_to_rem);
@@ -1123,7 +1130,7 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 	}
 
 	if (sta_mask_added) {
-		ret = iwl_mvm_mld_update_sta_resources(mvm, sta,
+		ret = iwl_mvm_mld_update_sta_resources(mvm, vif, sta,
 						       current_sta_mask,
 						       current_sta_mask |
 							sta_mask_added);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 29a87d1ba343..203eb7233c77 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2332,6 +2332,11 @@ void iwl_mvm_sec_key_remove_ap(struct iwl_mvm *mvm,
 			       struct ieee80211_vif *vif,
 			       struct iwl_mvm_vif_link_info *link,
 			       unsigned int link_id);
+int iwl_mvm_mld_update_sta_keys(struct iwl_mvm *mvm,
+				struct ieee80211_vif *vif,
+				struct ieee80211_sta *sta,
+				u32 old_sta_mask,
+				u32 new_sta_mask);
 
 int iwl_rfi_send_config_cmd(struct iwl_mvm *mvm,
 			    struct iwl_rfi_lut_entry *rfi_table);
-- 
cgit v1.2.3-70-g09d2


From d9bfd5a06448c175ba3a333e13168fb2622e9dd8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 16 Apr 2023 15:47:37 +0300
Subject: wifi: iwlwifi: mvm: allow number of beacons from FW

Newer firmware images have a TLV advertising how many
beacons they support, use that to permit adding more
links in AP mode (FW needs to support at least as many
links as beacons).

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.0d7522533557.Ic6b5992e94446c35cb0f3add019defa6e7aded2a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/file.h          | 1 +
 drivers/net/wireless/intel/iwlwifi/fw/img.h           | 1 +
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c          | 7 +++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 5 ++++-
 4 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h
index b00174c9ebce..cddf09d6be1c 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/file.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h
@@ -104,6 +104,7 @@ enum iwl_ucode_tlv_type {
 	IWL_UCODE_TLV_CURRENT_PC		= 68,
 
 	IWL_UCODE_TLV_FW_NUM_STATIONS		= IWL_UCODE_TLV_CONST_BASE + 0,
+	IWL_UCODE_TLV_FW_NUM_BEACONS		= IWL_UCODE_TLV_CONST_BASE + 2,
 
 	IWL_UCODE_TLV_TYPE_DEBUG_INFO		= IWL_UCODE_TLV_DEBUG_BASE + 0,
 	IWL_UCODE_TLV_TYPE_BUFFER_ALLOCATION	= IWL_UCODE_TLV_DEBUG_BASE + 1,
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/img.h b/drivers/net/wireless/intel/iwlwifi/fw/img.h
index f5c4d93d1033..8d0d58d61892 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/img.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/img.h
@@ -51,6 +51,7 @@ struct iwl_ucode_capabilities {
 	u32 error_log_addr;
 	u32 error_log_size;
 	u32 num_stations;
+	u32 num_beacons;
 	unsigned long _api[BITS_TO_LONGS(NUM_IWL_UCODE_TLV_API)];
 	unsigned long _capa[BITS_TO_LONGS(NUM_IWL_UCODE_TLV_CAPA)];
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 042c06a22f5f..34feb4d29adc 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -1155,6 +1155,12 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 			capa->num_stations =
 				le32_to_cpup((const __le32 *)tlv_data);
 			break;
+		case IWL_UCODE_TLV_FW_NUM_BEACONS:
+			if (tlv_len != sizeof(u32))
+				goto invalid_tlv_len;
+			capa->num_beacons =
+				le32_to_cpup((const __le32 *)tlv_data);
+			break;
 		case IWL_UCODE_TLV_UMAC_DEBUG_ADDRS: {
 			const struct iwl_umac_debug_addrs *dbg_ptrs =
 				(const void *)tlv_data;
@@ -1415,6 +1421,7 @@ static void iwl_req_fw_callback(const struct firmware *ucode_raw, void *context)
 			IWL_DEFAULT_STANDARD_PHY_CALIBRATE_TBL_SIZE;
 	fw->ucode_capa.n_scan_channels = IWL_DEFAULT_SCAN_CHANNELS;
 	fw->ucode_capa.num_stations = IWL_MVM_STATION_COUNT_MAX;
+	fw->ucode_capa.num_beacons = 1;
 	/* dump all fw memory areas by default */
 	fw->dbg.dump_mask = 0xffffffff;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 203f2513e7ea..99ef12c530ea 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -882,7 +882,10 @@ iwl_mvm_mld_change_vif_links(struct ieee80211_hw *hw,
 				n_active++;
 		}
 
-		if (n_active > 1)
+		if (vif->type == NL80211_IFTYPE_AP &&
+		    n_active > mvm->fw->ucode_capa.num_beacons)
+			return -EOPNOTSUPP;
+		else if (n_active > 1)
 			return -EOPNOTSUPP;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 28965ec0b5d9112585f725660e2ff13218505ace Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 16 Apr 2023 15:47:38 +0300
Subject: wifi: iwlwifi: make the loop for card preparation effective

Since we didn't reset t to 0, only the first iteration of the loop
did checked the ready bit several times.
From the second iteration and on, we just tested the bit once and
continued to the next iteration.

Reported-and-tested-by: Lorenzo Zolfanelli <lorenzo@zolfa.nl>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216452
Fixes: 289e5501c314 ("iwlwifi: fix the preparation of the card")
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.615b683ab9c8.Ic52c3229d3345b0064fa34263293db095d88daf8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index c5446b82c6c7..b281850fbf7a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -599,7 +599,6 @@ static int iwl_pcie_set_hw_ready(struct iwl_trans *trans)
 int iwl_pcie_prepare_card_hw(struct iwl_trans *trans)
 {
 	int ret;
-	int t = 0;
 	int iter;
 
 	IWL_DEBUG_INFO(trans, "iwl_trans_prepare_card_hw enter\n");
@@ -616,6 +615,8 @@ int iwl_pcie_prepare_card_hw(struct iwl_trans *trans)
 	usleep_range(1000, 2000);
 
 	for (iter = 0; iter < 10; iter++) {
+		int t = 0;
+
 		/* If HW is not ready, prepare the conditions to check again */
 		iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
 			    CSR_HW_IF_CONFIG_REG_PREPARE);
-- 
cgit v1.2.3-70-g09d2


From 9ec71b52cc6b72ca13258f942b16f3207e3d6502 Mon Sep 17 00:00:00 2001
From: Golan Ben Ami <golan.ben.ami@intel.com>
Date: Sun, 16 Apr 2023 15:47:39 +0300
Subject: wifi: iwlwifi: move debug buffer allocation failure to info verbosity

This is noising the kernel log and customers asked to hush it down.
We can live with this message in "info" verbosity.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230416154301.0a7f557aa2a0.If9db474b63242b1bfaed659aa174b678ae8dc196@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 546737320a57..898d5dcf1012 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -803,7 +803,7 @@ static void iwl_dbg_tlv_update_drams(struct iwl_fw_runtime *fwrt)
 		if (!ret)
 			dram_alloc = true;
 		else
-			IWL_WARN(fwrt,
+			IWL_INFO(fwrt,
 				 "WRT: Failed to set DRAM buffer for alloc id %d, ret=%d\n",
 				 i, ret);
 	}
-- 
cgit v1.2.3-70-g09d2


From 99676a5766412f3936c55b9d18565d248e5463ee Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Mon, 17 Apr 2023 09:26:41 +0200
Subject: net: lan966x: Fix lan966x_ifh_get

From time to time, it was observed that the nanosecond part of the
received timestamp, which is extracted from the IFH, it was actually
bigger than 1 second. So then when actually calculating the full
received timestamp, based on the nanosecond part from IFH and the second
part which is read from HW, it was actually wrong.

The issue seems to be inside the function lan966x_ifh_get, which
extracts information from an IFH(which is an byte array) and returns the
value in a u64. When extracting the timestamp value from the IFH, which
starts at bit 192 and have the size of 32 bits, then if the most
significant bit was set in the timestamp, then this bit was extended
then the return value became 0xffffffff... . And the reason of this is
because constants without any postfix are treated as signed longs and
that is the reason why '1 << 31' becomes 0xffffffff80000000.
This is fixed by adding the postfix 'ULL' to 1.

Fixes: fd7627833ddf ("net: lan966x: Stop using packing library")
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/microchip/lan966x/lan966x_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
index 9be6462f1cc5..2b6e046e1d10 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
@@ -605,7 +605,7 @@ static u64 lan966x_ifh_get(u8 *ifh, size_t pos, size_t length)
 			v = ifh[IFH_LEN_BYTES - (j / 8) - 1];
 
 		if (v & (1 << k))
-			val |= (1 << i);
+			val |= (1ULL << i);
 	}
 
 	return val;
-- 
cgit v1.2.3-70-g09d2


From 6863ad915d32990aec79e59db9d2a21a5abedf97 Mon Sep 17 00:00:00 2001
From: Chin-Yen Lee <timlee@realtek.com>
Date: Mon, 10 Apr 2023 13:34:37 +0800
Subject: wifi: rtw89: support WoWLAN mode for 8852be

To support WoWLAN mode for 8852be, we add one PLE quota setting and
WoWLAN stub, which shows that supported WLAN events include receiving
magic packet, rekey packet and deauth packet, and disconnecting from AP.

Signed-off-by: Chin-Yen Lee <timlee@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230410053438.10682-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/mac.c      |  2 ++
 drivers/net/wireless/realtek/rtw89/mac.h      |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 16 ++++++++++++++++
 3 files changed, 19 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 385e2a0dfea5..ce657dda5e35 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -1473,6 +1473,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = {
 	.ple_qt58 = {147, 0, 16, 20, 157, 13, 229, 0, 172, 14, 24, 0,},
 	/* 8852A PCIE WOW */
 	.ple_qt_52a_wow = {264, 0, 32, 20, 64, 13, 1005, 0, 64, 128, 120,},
+	/* 8852B PCIE WOW */
+	.ple_qt_52b_wow = {147, 0, 16, 20, 157, 13, 133, 0, 172, 14, 24, 0,},
 };
 EXPORT_SYMBOL(rtw89_mac_size);
 
diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h
index a9ea3459ed82..06cbad4b8d62 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.h
+++ b/drivers/net/wireless/realtek/rtw89/mac.h
@@ -816,6 +816,7 @@ struct rtw89_mac_size_set {
 	const struct rtw89_ple_quota ple_qt47;
 	const struct rtw89_ple_quota ple_qt58;
 	const struct rtw89_ple_quota ple_qt_52a_wow;
+	const struct rtw89_ple_quota ple_qt_52b_wow;
 };
 
 extern const struct rtw89_mac_size_set rtw89_mac_size;
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index d7ba513b4f7a..e93d745f6fb9 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -53,6 +53,10 @@ static const struct rtw89_dle_mem rtw8852b_dle_mem_pcie[] = {
 			   &rtw89_mac_size.ple_size6, &rtw89_mac_size.wde_qt6,
 			   &rtw89_mac_size.wde_qt6, &rtw89_mac_size.ple_qt18,
 			   &rtw89_mac_size.ple_qt58},
+	[RTW89_QTA_WOW] = {RTW89_QTA_WOW, &rtw89_mac_size.wde_size6,
+			   &rtw89_mac_size.ple_size6, &rtw89_mac_size.wde_qt6,
+			   &rtw89_mac_size.wde_qt6, &rtw89_mac_size.ple_qt18,
+			   &rtw89_mac_size.ple_qt_52b_wow},
 	[RTW89_QTA_DLFW] = {RTW89_QTA_DLFW, &rtw89_mac_size.wde_size9,
 			    &rtw89_mac_size.ple_size8, &rtw89_mac_size.wde_qt4,
 			    &rtw89_mac_size.wde_qt4, &rtw89_mac_size.ple_qt13,
@@ -2483,6 +2487,15 @@ static const struct rtw89_chip_ops rtw8852b_chip_ops = {
 	.btc_set_policy		= rtw89_btc_set_policy_v1,
 };
 
+#ifdef CONFIG_PM
+static const struct wiphy_wowlan_support rtw_wowlan_stub_8852b = {
+	.flags = WIPHY_WOWLAN_MAGIC_PKT | WIPHY_WOWLAN_DISCONNECT,
+	.n_patterns = RTW89_MAX_PATTERN_NUM,
+	.pattern_max_len = RTW89_MAX_PATTERN_SIZE,
+	.pattern_min_len = 1,
+};
+#endif
+
 const struct rtw89_chip_info rtw8852b_chip_info = {
 	.chip_id		= RTL8852B,
 	.ops			= &rtw8852b_chip_ops,
@@ -2579,6 +2592,9 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 				  BIT(RTW89_DMA_ACH6) | BIT(RTW89_DMA_ACH7) |
 				  BIT(RTW89_DMA_B1MG) | BIT(RTW89_DMA_B1HI),
 	.edcca_lvl_reg		= R_SEG0R_EDCCA_LVL_V1,
+#ifdef CONFIG_PM
+	.wowlan_stub		= &rtw_wowlan_stub_8852b,
+#endif
 };
 EXPORT_SYMBOL(rtw8852b_chip_info);
 
-- 
cgit v1.2.3-70-g09d2


From deb1b2aed763828b54a5be9be76c3a43c295f9f0 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Mon, 10 Apr 2023 13:34:38 +0800
Subject: wifi: rtw89: fix power save function in WoWLAN mode

In WoWLAN Mode, it's expected that WiFi chip could enter power save mode
only after all setting is finished, but current wow_enter_lps function
break the rule and may lead to WoWLAN function fail in low probability,
so fix it.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230410053438.10682-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 2 +-
 drivers/net/wireless/realtek/rtw89/ps.c   | 6 ++++--
 drivers/net/wireless/realtek/rtw89/ps.h   | 3 ++-
 drivers/net/wireless/realtek/rtw89/wow.c  | 2 +-
 4 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index d144903c60b3..8459a7bfa32f 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -2503,7 +2503,7 @@ static void rtw89_vif_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwv
 
 	if (rtwvif->stats.tx_tfc_lv == RTW89_TFC_IDLE &&
 	    rtwvif->stats.rx_tfc_lv == RTW89_TFC_IDLE)
-		rtw89_enter_lps(rtwdev, rtwvif);
+		rtw89_enter_lps(rtwdev, rtwvif, true);
 }
 
 static void rtw89_enter_lps_track(struct rtw89_dev *rtwdev)
diff --git a/drivers/net/wireless/realtek/rtw89/ps.c b/drivers/net/wireless/realtek/rtw89/ps.c
index cf72861c7adc..fa94335f699a 100644
--- a/drivers/net/wireless/realtek/rtw89/ps.c
+++ b/drivers/net/wireless/realtek/rtw89/ps.c
@@ -114,7 +114,8 @@ void rtw89_leave_ps_mode(struct rtw89_dev *rtwdev)
 	__rtw89_leave_ps_mode(rtwdev);
 }
 
-void rtw89_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
+void rtw89_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+		     bool ps_mode)
 {
 	lockdep_assert_held(&rtwdev->mutex);
 
@@ -122,7 +123,8 @@ void rtw89_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 		return;
 
 	__rtw89_enter_lps(rtwdev, rtwvif->mac_id);
-	__rtw89_enter_ps_mode(rtwdev, rtwvif);
+	if (ps_mode)
+		__rtw89_enter_ps_mode(rtwdev, rtwvif);
 }
 
 static void rtw89_leave_lps_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
diff --git a/drivers/net/wireless/realtek/rtw89/ps.h b/drivers/net/wireless/realtek/rtw89/ps.h
index c9e29d92fc1b..73c008db0426 100644
--- a/drivers/net/wireless/realtek/rtw89/ps.h
+++ b/drivers/net/wireless/realtek/rtw89/ps.h
@@ -5,7 +5,8 @@
 #ifndef __RTW89_PS_H_
 #define __RTW89_PS_H_
 
-void rtw89_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
+void rtw89_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+		     bool ps_mode);
 void rtw89_leave_lps(struct rtw89_dev *rtwdev);
 void __rtw89_leave_ps_mode(struct rtw89_dev *rtwdev);
 void __rtw89_enter_ps_mode(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c
index 7cff9c1d8d37..2ca8abb70f11 100644
--- a/drivers/net/wireless/realtek/rtw89/wow.c
+++ b/drivers/net/wireless/realtek/rtw89/wow.c
@@ -30,7 +30,7 @@ static void rtw89_wow_enter_lps(struct rtw89_dev *rtwdev)
 	struct ieee80211_vif *wow_vif = rtwdev->wow.wow_vif;
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)wow_vif->drv_priv;
 
-	rtw89_enter_lps(rtwdev, rtwvif);
+	rtw89_enter_lps(rtwdev, rtwvif, false);
 }
 
 static void rtw89_wow_leave_lps(struct rtw89_dev *rtwdev)
-- 
cgit v1.2.3-70-g09d2


From 36ef71db559f6a0dc2d7a9af8edb16c387d18ce2 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 12 Apr 2023 09:28:28 +0800
Subject: wifi: rtw89: coex: Enable Wi-Fi RX gain control for free run solution

When Wi-Fi & Bluetooth are both busy at the same time, Wi-Fi need to
enable RX gain to protect Wi-Fi RX RF ability. Without this configure
the interference from Bluetooth will bring a big impact to Wi-Fi RX.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230412012831.10519-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 4 ++--
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index e93d745f6fb9..c9a9f28b056c 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -332,7 +332,7 @@ static const struct rtw89_btc_rf_trx_para rtw89_btc_8852b_rf_ul[] = {
 	{255, 0, 0, 7}, /* 2 ->reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 3- >reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 4 ->reserved for shared-antenna */
-	{255, 0, 0, 7}, /* the below id is for non-shared-antenna free-run */
+	{255, 1, 0, 7}, /* the below id is for non-shared-antenna free-run */
 	{6, 1, 0, 7},
 	{13, 1, 0, 7},
 	{13, 1, 0, 7}
@@ -344,7 +344,7 @@ static const struct rtw89_btc_rf_trx_para rtw89_btc_8852b_rf_dl[] = {
 	{255, 0, 0, 7}, /* 2 ->reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 3- >reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 4 ->reserved for shared-antenna */
-	{255, 0, 0, 7}, /* the below id is for non-shared-antenna free-run */
+	{255, 1, 0, 7}, /* the below id is for non-shared-antenna free-run */
 	{255, 1, 0, 7},
 	{255, 1, 0, 7},
 	{255, 1, 0, 7}
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 6d8fa8acfc5e..3bff0502d6c9 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2542,7 +2542,7 @@ static const struct rtw89_btc_rf_trx_para rtw89_btc_8852c_rf_ul[] = {
 	{255, 0, 0, 7}, /* 2 ->reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 3- >reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 4 ->reserved for shared-antenna */
-	{255, 0, 0, 7}, /* the below id is for non-shared-antenna free-run */
+	{255, 1, 0, 7}, /* the below id is for non-shared-antenna free-run */
 	{6, 1, 0, 7},
 	{13, 1, 0, 7},
 	{13, 1, 0, 7}
@@ -2554,7 +2554,7 @@ static const struct rtw89_btc_rf_trx_para rtw89_btc_8852c_rf_dl[] = {
 	{255, 0, 0, 7}, /* 2 ->reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 3- >reserved for shared-antenna */
 	{255, 0, 0, 7}, /* 4 ->reserved for shared-antenna */
-	{255, 0, 0, 7}, /* the below id is for non-shared-antenna free-run */
+	{255, 1, 0, 7}, /* the below id is for non-shared-antenna free-run */
 	{255, 1, 0, 7},
 	{255, 1, 0, 7},
 	{255, 1, 0, 7}
-- 
cgit v1.2.3-70-g09d2


From 9fde30562840837e4e2138bc3f88dba2b9963d98 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 12 Apr 2023 09:28:29 +0800
Subject: wifi: rtw89: coex: Add path control register to monitor list

Chips use similar hardware for path control, but could different
path/antenna configuration. Add these register to monitor, if there are
wrong settings, these register can help to debug.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230412012831.10519-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 4 +++-
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 3 +++
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index c9a9f28b056c..ea4a4dc63312 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -364,7 +364,9 @@ static const struct rtw89_btc_fbtc_mreg rtw89_btc_8852b_mon_reg[] = {
 	RTW89_DEF_FBTC_MREG(REG_MAC, 4, 0xd200),
 	RTW89_DEF_FBTC_MREG(REG_MAC, 4, 0xd220),
 	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x980),
-	RTW89_DEF_FBTC_MREG(REG_BT_MODEM, 4, 0x178),
+	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x4738),
+	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x4688),
+	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x4694),
 };
 
 static const u8 rtw89_btc_8852b_wl_rssi_thres[BTC_WL_RSSI_THMAX] = {70, 60, 50, 40};
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 3bff0502d6c9..31e1650404b7 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2576,6 +2576,9 @@ static const struct rtw89_btc_fbtc_mreg rtw89_btc_8852c_mon_reg[] = {
 	RTW89_DEF_FBTC_MREG(REG_MAC, 4, 0xd200),
 	RTW89_DEF_FBTC_MREG(REG_MAC, 4, 0xd220),
 	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x980),
+	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x4aa4),
+	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x4778),
+	RTW89_DEF_FBTC_MREG(REG_BB, 4, 0x476c),
 };
 
 static
-- 
cgit v1.2.3-70-g09d2


From 2380a220316fc2e753014b1862ed579796d29ac5 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 12 Apr 2023 09:28:30 +0800
Subject: wifi: rtw89: coex: Update function to get BT RSSI and hardware
 counter

Correct Bluetooth RSSI count method. The 6dB is the gap between hardware
packet sampled value and real RSSI value.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230412012831.10519-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852a.c | 3 ++-
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 3 ++-
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 3 ++-
 3 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index eb7a8340f633..d7930efd89b7 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -1832,7 +1832,8 @@ rtw8852a_btc_set_wl_txpwr_ctrl(struct rtw89_dev *rtwdev, u32 txpwr_val)
 static
 s8 rtw8852a_btc_get_bt_rssi(struct rtw89_dev *rtwdev, s8 val)
 {
-	return clamp_t(s8, val, -100, 0) + 100;
+	/* +6 for compensate offset */
+	return clamp_t(s8, val + 6, -100, 0) + 100;
 }
 
 static struct rtw89_btc_rf_trx_para rtw89_btc_8852a_rf_ul[] = {
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index ea4a4dc63312..eaa2ea0586bc 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2279,7 +2279,8 @@ do {								\
 static
 s8 rtw8852b_btc_get_bt_rssi(struct rtw89_dev *rtwdev, s8 val)
 {
-	return clamp_t(s8, val, -100, 0) + 100;
+	/* +6 for compensate offset */
+	return clamp_t(s8, val + 6, -100, 0) + 100;
 }
 
 static
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 31e1650404b7..ceb819a62efc 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2533,7 +2533,8 @@ do {								\
 static
 s8 rtw8852c_btc_get_bt_rssi(struct rtw89_dev *rtwdev, s8 val)
 {
-	return clamp_t(s8, val, -100, 0) + 100;
+	/* +6 for compensate offset */
+	return clamp_t(s8, val + 6, -100, 0) + 100;
 }
 
 static const struct rtw89_btc_rf_trx_para rtw89_btc_8852c_rf_ul[] = {
-- 
cgit v1.2.3-70-g09d2


From c0fea064b2647a5069e2c234fb44286c8a205993 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Wed, 12 Apr 2023 09:28:31 +0800
Subject: wifi: rtw89: coex: send more hardware module info to firmware for
 8851B

8851B has various hardware module types, so BT coexistence in firmware
needs these information to make decision. Add them to make 8851B work
well.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230412012831.10519-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h |   5 +-
 drivers/net/wireless/realtek/rtw89/fw.c   |  64 +++++++++--------
 drivers/net/wireless/realtek/rtw89/fw.h   | 113 ++++++++++--------------------
 3 files changed, 75 insertions(+), 107 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index eab0abb9d213..d1229c864d98 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -959,6 +959,8 @@ struct rtw89_btc_ant_info {
 
 	u8 single_pos: 1;/* Single antenna at S0 or S1 */
 	u8 diversity: 1;
+	u8 btg_pos: 2;
+	u8 stream_cnt: 4;
 };
 
 enum rtw89_tfc_dir {
@@ -1415,8 +1417,9 @@ struct rtw89_btc_module {
 	u8 bt_solo: 1;
 	u8 bt_pos: 1;
 	u8 switch_type: 1;
+	u8 wa_type: 3;
 
-	u8 rsvd;
+	u8 kt_ver_adie;
 };
 
 #define RTW89_BTC_DM_MAXSTEP 30
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index c22e282849ff..2d648653c292 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -1924,8 +1924,6 @@ fail:
 	return ret;
 }
 
-#define H2C_LEN_CXDRVHDR 2
-#define H2C_LEN_CXDRVINFO_INIT (12 + H2C_LEN_CXDRVHDR)
 int rtw89_fw_h2c_cxdrv_init(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
@@ -1933,44 +1931,52 @@ int rtw89_fw_h2c_cxdrv_init(struct rtw89_dev *rtwdev)
 	struct rtw89_btc_init_info *init_info = &dm->init_info;
 	struct rtw89_btc_module *module = &init_info->module;
 	struct rtw89_btc_ant_info *ant = &module->ant;
+	struct rtw89_h2c_cxinit *h2c;
+	u32 len = sizeof(*h2c);
 	struct sk_buff *skb;
-	u8 *cmd;
 	int ret;
 
-	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, H2C_LEN_CXDRVINFO_INIT);
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len);
 	if (!skb) {
 		rtw89_err(rtwdev, "failed to alloc skb for h2c cxdrv_init\n");
 		return -ENOMEM;
 	}
-	skb_put(skb, H2C_LEN_CXDRVINFO_INIT);
-	cmd = skb->data;
-
-	RTW89_SET_FWCMD_CXHDR_TYPE(cmd, CXDRVINFO_INIT);
-	RTW89_SET_FWCMD_CXHDR_LEN(cmd, H2C_LEN_CXDRVINFO_INIT - H2C_LEN_CXDRVHDR);
-
-	RTW89_SET_FWCMD_CXINIT_ANT_TYPE(cmd, ant->type);
-	RTW89_SET_FWCMD_CXINIT_ANT_NUM(cmd, ant->num);
-	RTW89_SET_FWCMD_CXINIT_ANT_ISO(cmd, ant->isolation);
-	RTW89_SET_FWCMD_CXINIT_ANT_POS(cmd, ant->single_pos);
-	RTW89_SET_FWCMD_CXINIT_ANT_DIVERSITY(cmd, ant->diversity);
-
-	RTW89_SET_FWCMD_CXINIT_MOD_RFE(cmd, module->rfe_type);
-	RTW89_SET_FWCMD_CXINIT_MOD_CV(cmd, module->cv);
-	RTW89_SET_FWCMD_CXINIT_MOD_BT_SOLO(cmd, module->bt_solo);
-	RTW89_SET_FWCMD_CXINIT_MOD_BT_POS(cmd, module->bt_pos);
-	RTW89_SET_FWCMD_CXINIT_MOD_SW_TYPE(cmd, module->switch_type);
-
-	RTW89_SET_FWCMD_CXINIT_WL_GCH(cmd, init_info->wl_guard_ch);
-	RTW89_SET_FWCMD_CXINIT_WL_ONLY(cmd, init_info->wl_only);
-	RTW89_SET_FWCMD_CXINIT_WL_INITOK(cmd, init_info->wl_init_ok);
-	RTW89_SET_FWCMD_CXINIT_DBCC_EN(cmd, init_info->dbcc_en);
-	RTW89_SET_FWCMD_CXINIT_CX_OTHER(cmd, init_info->cx_other);
-	RTW89_SET_FWCMD_CXINIT_BT_ONLY(cmd, init_info->bt_only);
+	skb_put(skb, len);
+	h2c = (struct rtw89_h2c_cxinit *)skb->data;
+
+	h2c->hdr.type = CXDRVINFO_INIT;
+	h2c->hdr.len = len - H2C_LEN_CXDRVHDR;
+
+	h2c->ant_type = ant->type;
+	h2c->ant_num = ant->num;
+	h2c->ant_iso = ant->isolation;
+	h2c->ant_info =
+		u8_encode_bits(ant->single_pos, RTW89_H2C_CXINIT_ANT_INFO_POS) |
+		u8_encode_bits(ant->diversity, RTW89_H2C_CXINIT_ANT_INFO_DIVERSITY) |
+		u8_encode_bits(ant->btg_pos, RTW89_H2C_CXINIT_ANT_INFO_BTG_POS) |
+		u8_encode_bits(ant->stream_cnt, RTW89_H2C_CXINIT_ANT_INFO_STREAM_CNT);
+
+	h2c->mod_rfe = module->rfe_type;
+	h2c->mod_cv = module->cv;
+	h2c->mod_info =
+		u8_encode_bits(module->bt_solo, RTW89_H2C_CXINIT_MOD_INFO_BT_SOLO) |
+		u8_encode_bits(module->bt_pos, RTW89_H2C_CXINIT_MOD_INFO_BT_POS) |
+		u8_encode_bits(module->switch_type, RTW89_H2C_CXINIT_MOD_INFO_SW_TYPE) |
+		u8_encode_bits(module->wa_type, RTW89_H2C_CXINIT_MOD_INFO_WA_TYPE);
+	h2c->mod_adie_kt = module->kt_ver_adie;
+	h2c->wl_gch = init_info->wl_guard_ch;
+
+	h2c->info =
+		u8_encode_bits(init_info->wl_only, RTW89_H2C_CXINIT_INFO_WL_ONLY) |
+		u8_encode_bits(init_info->wl_init_ok, RTW89_H2C_CXINIT_INFO_WL_INITOK) |
+		u8_encode_bits(init_info->dbcc_en, RTW89_H2C_CXINIT_INFO_DBCC_EN) |
+		u8_encode_bits(init_info->cx_other, RTW89_H2C_CXINIT_INFO_CX_OTHER) |
+		u8_encode_bits(init_info->bt_only, RTW89_H2C_CXINIT_INFO_BT_ONLY);
 
 	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
 			      H2C_CAT_OUTSRC, BTFC_SET,
 			      SET_DRV_INFO, 0, 0,
-			      H2C_LEN_CXDRVINFO_INIT);
+			      len);
 
 	ret = rtw89_h2c_tx(rtwdev, skb, false);
 	if (ret) {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index a2ff173e5544..aef703f44a70 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -2197,85 +2197,44 @@ static inline void RTW89_SET_FWCMD_CXHDR_LEN(void *cmd, u8 val)
 	u8p_replace_bits((u8 *)(cmd) + 1, val, GENMASK(7, 0));
 }
 
-static inline void RTW89_SET_FWCMD_CXINIT_ANT_TYPE(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 2, val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_ANT_NUM(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 3, val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_ANT_ISO(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 4, val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_ANT_POS(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 5, val, BIT(0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_ANT_DIVERSITY(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 5, val, BIT(1));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_MOD_RFE(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 6, val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_MOD_CV(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 7, val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_MOD_BT_SOLO(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 8, val, BIT(0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_MOD_BT_POS(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 8, val, BIT(1));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_MOD_SW_TYPE(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 8, val, BIT(2));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_WL_GCH(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 10, val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_WL_ONLY(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 11, val, BIT(0));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_WL_INITOK(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 11, val, BIT(1));
-}
-
-static inline void RTW89_SET_FWCMD_CXINIT_DBCC_EN(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 11, val, BIT(2));
-}
+struct rtw89_h2c_cxhdr {
+	u8 type;
+	u8 len;
+} __packed;
 
-static inline void RTW89_SET_FWCMD_CXINIT_CX_OTHER(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 11, val, BIT(3));
-}
+#define H2C_LEN_CXDRVHDR sizeof(struct rtw89_h2c_cxhdr)
+
+struct rtw89_h2c_cxinit {
+	struct rtw89_h2c_cxhdr hdr;
+	u8 ant_type;
+	u8 ant_num;
+	u8 ant_iso;
+	u8 ant_info;
+	u8 mod_rfe;
+	u8 mod_cv;
+	u8 mod_info;
+	u8 mod_adie_kt;
+	u8 wl_gch;
+	u8 info;
+	u8 rsvd;
+	u8 rsvd1;
+} __packed;
 
-static inline void RTW89_SET_FWCMD_CXINIT_BT_ONLY(void *cmd, u8 val)
-{
-	u8p_replace_bits((u8 *)(cmd) + 11, val, BIT(4));
-}
+#define RTW89_H2C_CXINIT_ANT_INFO_POS BIT(0)
+#define RTW89_H2C_CXINIT_ANT_INFO_DIVERSITY BIT(1)
+#define RTW89_H2C_CXINIT_ANT_INFO_BTG_POS GENMASK(3, 2)
+#define RTW89_H2C_CXINIT_ANT_INFO_STREAM_CNT GENMASK(7, 4)
+
+#define RTW89_H2C_CXINIT_MOD_INFO_BT_SOLO BIT(0)
+#define RTW89_H2C_CXINIT_MOD_INFO_BT_POS BIT(1)
+#define RTW89_H2C_CXINIT_MOD_INFO_SW_TYPE BIT(2)
+#define RTW89_H2C_CXINIT_MOD_INFO_WA_TYPE GENMASK(5, 3)
+
+#define RTW89_H2C_CXINIT_INFO_WL_ONLY BIT(0)
+#define RTW89_H2C_CXINIT_INFO_WL_INITOK BIT(1)
+#define RTW89_H2C_CXINIT_INFO_DBCC_EN BIT(2)
+#define RTW89_H2C_CXINIT_INFO_CX_OTHER BIT(3)
+#define RTW89_H2C_CXINIT_INFO_BT_ONLY BIT(4)
 
 static inline void RTW89_SET_FWCMD_CXROLE_CONNECT_CNT(void *cmd, u8 val)
 {
-- 
cgit v1.2.3-70-g09d2


From e579e943bac3b52f69a25738fcbd8be945f72689 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Sat, 15 Apr 2023 11:48:55 +0800
Subject: wifi: rtw89: prohibit enter IPS during HW scan

Mac80211 core may ask driver to change to idle mode during HW scan,
then H2C command for HW scan will send failed since chip is in idle
mode. Therefore, We check the SCANNING flag before entering IPS to
prevent this behavior.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230415034900.15679-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/mac80211.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index b059aa8d88db..b5cbfc15ebad 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -105,7 +105,8 @@ static int rtw89_ops_config(struct ieee80211_hw *hw, u32 changed)
 	}
 
 	if ((changed & IEEE80211_CONF_CHANGE_IDLE) &&
-	    (hw->conf.flags & IEEE80211_CONF_IDLE))
+	    (hw->conf.flags & IEEE80211_CONF_IDLE) &&
+	    !rtwdev->scanning)
 		rtw89_enter_ips(rtwdev);
 
 	mutex_unlock(&rtwdev->mutex);
-- 
cgit v1.2.3-70-g09d2


From e7399db231d07f1e5a4179f100ccd0106fdbee03 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Sat, 15 Apr 2023 11:48:56 +0800
Subject: wifi: rtw89: refine scan function after chanctx

Since we can get the current channel definition each interface maps to,
remove store_op function that is no longer required to make things simple.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230415034900.15679-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c     |  9 +++++
 drivers/net/wireless/realtek/rtw89/core.h     |  7 ++--
 drivers/net/wireless/realtek/rtw89/fw.c       | 47 ++++++++-------------------
 drivers/net/wireless/realtek/rtw89/fw.h       |  1 -
 drivers/net/wireless/realtek/rtw89/mac.c      | 14 +++++---
 drivers/net/wireless/realtek/rtw89/mac80211.c |  1 -
 6 files changed, 35 insertions(+), 44 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 8459a7bfa32f..8672bfc4801c 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -360,6 +360,15 @@ void rtw89_set_channel(struct rtw89_dev *rtwdev)
 	rtw89_set_entity_state(rtwdev, true);
 }
 
+void rtw89_get_channel(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+		       struct rtw89_chan *chan)
+{
+	const struct cfg80211_chan_def *chandef;
+
+	chandef = rtw89_chandef_get(rtwdev, rtwvif->sub_entity_idx);
+	rtw89_get_channel_params(chandef, chan);
+}
+
 static enum rtw89_core_tx_type
 rtw89_core_get_tx_type(struct rtw89_dev *rtwdev,
 		       struct sk_buff *skb)
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index d1229c864d98..6df386a38fb4 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3936,10 +3936,7 @@ struct rtw89_early_h2c {
 struct rtw89_hw_scan_info {
 	struct ieee80211_vif *scanning_vif;
 	struct list_head pkt_list[NUM_NL80211_BANDS];
-	u8 op_pri_ch;
-	u8 op_chan;
-	u8 op_bw;
-	u8 op_band;
+	struct rtw89_chan op_chan;
 	u32 last_chan_idx;
 };
 
@@ -4981,6 +4978,8 @@ void rtw89_free_ieee80211_hw(struct rtw89_dev *rtwdev);
 void rtw89_core_set_chip_txpwr(struct rtw89_dev *rtwdev);
 void rtw89_get_default_chandef(struct cfg80211_chan_def *chandef);
 void rtw89_set_channel(struct rtw89_dev *rtwdev);
+void rtw89_get_channel(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+		       struct rtw89_chan *chan);
 u8 rtw89_core_acquire_bit_map(unsigned long *addr, unsigned long size);
 void rtw89_core_release_bit_map(unsigned long *addr, u8 bit);
 void rtw89_core_release_all_bits_map(unsigned long *addr, unsigned int nbits);
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 2d648653c292..1ea56e2fd02f 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2551,7 +2551,7 @@ int rtw89_fw_h2c_scan_offload(struct rtw89_dev *rtwdev,
 			      struct rtw89_scan_option *option,
 			      struct rtw89_vif *rtwvif)
 {
-	struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info;
+	struct rtw89_chan *op = &rtwdev->scan_info.op_chan;
 	struct sk_buff *skb;
 	u8 *cmd;
 	int ret;
@@ -2573,13 +2573,11 @@ int rtw89_fw_h2c_scan_offload(struct rtw89_dev *rtwdev,
 	RTW89_SET_FWCMD_SCANOFLD_START_MODE(cmd, RTW89_SCAN_IMMEDIATE);
 	RTW89_SET_FWCMD_SCANOFLD_SCAN_TYPE(cmd, RTW89_SCAN_ONCE);
 	if (option->target_ch_mode) {
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BW(cmd, scan_info->op_bw);
+		RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BW(cmd, op->band_width);
 		RTW89_SET_FWCMD_SCANOFLD_TARGET_PRI_CH(cmd,
-						       scan_info->op_pri_ch);
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_CENTRAL_CH(cmd,
-							   scan_info->op_chan);
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BAND(cmd,
-							scan_info->op_band);
+						       op->primary_channel);
+		RTW89_SET_FWCMD_SCANOFLD_TARGET_CENTRAL_CH(cmd, op->channel);
+		RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BAND(cmd, op->band_type);
 	}
 
 	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
@@ -3158,6 +3156,7 @@ static void rtw89_hw_scan_add_chan(struct rtw89_dev *rtwdev, int chan_type,
 	struct ieee80211_vif *vif = rtwdev->scan_info.scanning_vif;
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
 	struct cfg80211_scan_request *req = rtwvif->scan_req;
+	struct rtw89_chan *op = &rtwdev->scan_info.op_chan;
 	struct rtw89_pktofld_info *info;
 	u8 band, probe_count = 0;
 	int ret;
@@ -3201,10 +3200,10 @@ static void rtw89_hw_scan_add_chan(struct rtw89_dev *rtwdev, int chan_type,
 
 	switch (chan_type) {
 	case RTW89_CHAN_OPERATE:
-		ch_info->central_ch = scan_info->op_chan;
-		ch_info->pri_ch = scan_info->op_pri_ch;
-		ch_info->ch_band = scan_info->op_band;
-		ch_info->bw = scan_info->op_bw;
+		ch_info->central_ch = op->channel;
+		ch_info->pri_ch = op->primary_channel;
+		ch_info->ch_band = op->band_type;
+		ch_info->bw = op->band_width;
 		ch_info->tx_null = true;
 		ch_info->num_pkt = 0;
 		break;
@@ -3321,6 +3320,7 @@ void rtw89_hw_scan_start(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 	u32 rx_fltr = rtwdev->hal.rx_fltr;
 	u8 mac_addr[ETH_ALEN];
 
+	rtw89_get_channel(rtwdev, rtwvif, &rtwdev->scan_info.op_chan);
 	rtwdev->scan_info.scanning_vif = vif;
 	rtwdev->scan_info.last_chan_idx = 0;
 	rtwvif->scan_ies = &scan_req->ies;
@@ -3346,6 +3346,7 @@ void rtw89_hw_scan_start(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 void rtw89_hw_scan_complete(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 			    bool aborted)
 {
+	struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info;
 	struct cfg80211_scan_info info = {
 		.aborted = aborted,
 	};
@@ -3367,11 +3368,9 @@ void rtw89_hw_scan_complete(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 	rtwvif = (struct rtw89_vif *)vif->drv_priv;
 	rtwvif->scan_req = NULL;
 	rtwvif->scan_ies = NULL;
-	rtwdev->scan_info.last_chan_idx = 0;
-	rtwdev->scan_info.scanning_vif = NULL;
+	scan_info->last_chan_idx = 0;
+	scan_info->scanning_vif = NULL;
 
-	if (rtwvif->net_type != RTW89_NET_TYPE_NO_LINK)
-		rtw89_store_op_chan(rtwdev, false);
 	rtw89_set_channel(rtwdev);
 }
 
@@ -3407,24 +3406,6 @@ out:
 	return ret;
 }
 
-void rtw89_store_op_chan(struct rtw89_dev *rtwdev, bool backup)
-{
-	struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info;
-	const struct rtw89_chan *cur = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
-	struct rtw89_chan new;
-
-	if (backup) {
-		scan_info->op_pri_ch = cur->primary_channel;
-		scan_info->op_chan = cur->channel;
-		scan_info->op_bw = cur->band_width;
-		scan_info->op_band = cur->band_type;
-	} else {
-		rtw89_chan_create(&new, scan_info->op_chan, scan_info->op_pri_ch,
-				  scan_info->op_band, scan_info->op_bw);
-		rtw89_assign_entity_chan(rtwdev, RTW89_SUB_ENTITY_0, &new);
-	}
-}
-
 #define H2C_FW_CPU_EXCEPTION_LEN 4
 #define H2C_FW_CPU_EXCEPTION_TYPE_DEF 0x5566
 int rtw89_fw_h2c_trigger_cpu_exception(struct rtw89_dev *rtwdev)
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index aef703f44a70..0277caf6ae8a 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -3713,7 +3713,6 @@ int rtw89_fw_msg_reg(struct rtw89_dev *rtwdev,
 		     struct rtw89_mac_c2h_info *c2h_info);
 int rtw89_fw_h2c_fw_log(struct rtw89_dev *rtwdev, bool enable);
 void rtw89_fw_st_dbg_dump(struct rtw89_dev *rtwdev);
-void rtw89_store_op_chan(struct rtw89_dev *rtwdev, bool backup);
 void rtw89_hw_scan_start(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 			 struct ieee80211_scan_request *req);
 void rtw89_hw_scan_complete(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index ce657dda5e35..9b47d80b900b 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4194,9 +4194,9 @@ rtw89_mac_c2h_macid_pause(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len
 
 static bool rtw89_is_op_chan(struct rtw89_dev *rtwdev, u8 band, u8 channel)
 {
-	struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info;
+	const struct rtw89_chan *op = &rtwdev->scan_info.op_chan;
 
-	return band == scan_info->op_band && channel == scan_info->op_pri_ch;
+	return band == op->band_type && channel == op->primary_channel;
 }
 
 static void
@@ -4246,11 +4246,15 @@ rtw89_mac_c2h_scanofld_rsp(struct rtw89_dev *rtwdev, struct sk_buff *c2h,
 		}
 		break;
 	case RTW89_SCAN_ENTER_CH_NOTIFY:
-		rtw89_chan_create(&new, chan, chan, band, RTW89_CHANNEL_WIDTH_20);
-		rtw89_assign_entity_chan(rtwdev, RTW89_SUB_ENTITY_0, &new);
 		if (rtw89_is_op_chan(rtwdev, band, chan)) {
-			rtw89_store_op_chan(rtwdev, false);
+			rtw89_assign_entity_chan(rtwdev, rtwvif->sub_entity_idx,
+						 &rtwdev->scan_info.op_chan);
 			ieee80211_wake_queues(rtwdev->hw);
+		} else {
+			rtw89_chan_create(&new, chan, chan, band,
+					  RTW89_CHANNEL_WIDTH_20);
+			rtw89_assign_entity_chan(rtwdev, rtwvif->sub_entity_idx,
+						 &new);
 		}
 		break;
 	default:
diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index b5cbfc15ebad..ee4588b61b8f 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -418,7 +418,6 @@ static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw,
 			rtw89_chip_cfg_txpwr_ul_tb_offset(rtwdev, vif);
 			rtw89_mac_port_update(rtwdev, rtwvif);
 			rtw89_mac_set_he_obss_narrow_bw_ru(rtwdev, vif);
-			rtw89_store_op_chan(rtwdev, true);
 		} else {
 			/* Abort ongoing scan if cancel_scan isn't issued
 			 * when disconnected by peer
-- 
cgit v1.2.3-70-g09d2


From 8b048bd5ddf700c72734c4a2a79ecdf082273edb Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Sat, 15 Apr 2023 11:48:57 +0800
Subject: wifi: rtw89: use struct instead of macros to set H2C command of
 hardware scan

Remove macros that set H2C data. Instead, use struct and
le32_encode_bits() with mask definition to make it clean.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230415034900.15679-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c |  45 +++++++------
 drivers/net/wireless/realtek/rtw89/fw.h | 114 +++++++-------------------------
 2 files changed, 51 insertions(+), 108 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 1ea56e2fd02f..f55db06f75ef 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2546,44 +2546,51 @@ fail:
 	return ret;
 }
 
-#define H2C_LEN_SCAN_OFFLOAD 28
 int rtw89_fw_h2c_scan_offload(struct rtw89_dev *rtwdev,
 			      struct rtw89_scan_option *option,
 			      struct rtw89_vif *rtwvif)
 {
 	struct rtw89_chan *op = &rtwdev->scan_info.op_chan;
+	struct rtw89_h2c_scanofld *h2c;
+	u32 len = sizeof(*h2c);
 	struct sk_buff *skb;
-	u8 *cmd;
 	int ret;
 
-	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, H2C_LEN_SCAN_OFFLOAD);
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len);
 	if (!skb) {
 		rtw89_err(rtwdev, "failed to alloc skb for h2c scan offload\n");
 		return -ENOMEM;
 	}
-	skb_put(skb, H2C_LEN_SCAN_OFFLOAD);
-	cmd = skb->data;
+	skb_put(skb, len);
+	h2c = (struct rtw89_h2c_scanofld *)skb->data;
+
+	h2c->w0 = le32_encode_bits(rtwvif->mac_id, RTW89_H2C_SCANOFLD_W0_MACID) |
+		  le32_encode_bits(rtwvif->port, RTW89_H2C_SCANOFLD_W0_PORT_ID) |
+		  le32_encode_bits(RTW89_PHY_0, RTW89_H2C_SCANOFLD_W0_BAND) |
+		  le32_encode_bits(option->enable, RTW89_H2C_SCANOFLD_W0_OPERATION);
+
+	h2c->w1 = le32_encode_bits(true, RTW89_H2C_SCANOFLD_W1_NOTIFY_END) |
+		  le32_encode_bits(option->target_ch_mode,
+				   RTW89_H2C_SCANOFLD_W1_TARGET_CH_MODE) |
+		  le32_encode_bits(RTW89_SCAN_IMMEDIATE,
+				   RTW89_H2C_SCANOFLD_W1_START_MODE) |
+		  le32_encode_bits(RTW89_SCAN_ONCE, RTW89_H2C_SCANOFLD_W1_SCAN_TYPE);
 
-	RTW89_SET_FWCMD_SCANOFLD_MACID(cmd, rtwvif->mac_id);
-	RTW89_SET_FWCMD_SCANOFLD_PORT_ID(cmd, rtwvif->port);
-	RTW89_SET_FWCMD_SCANOFLD_BAND(cmd, RTW89_PHY_0);
-	RTW89_SET_FWCMD_SCANOFLD_OPERATION(cmd, option->enable);
-	RTW89_SET_FWCMD_SCANOFLD_NOTIFY_END(cmd, true);
-	RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_MODE(cmd, option->target_ch_mode);
-	RTW89_SET_FWCMD_SCANOFLD_START_MODE(cmd, RTW89_SCAN_IMMEDIATE);
-	RTW89_SET_FWCMD_SCANOFLD_SCAN_TYPE(cmd, RTW89_SCAN_ONCE);
 	if (option->target_ch_mode) {
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BW(cmd, op->band_width);
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_PRI_CH(cmd,
-						       op->primary_channel);
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_CENTRAL_CH(cmd, op->channel);
-		RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BAND(cmd, op->band_type);
+		h2c->w1 |= le32_encode_bits(op->band_width,
+					    RTW89_H2C_SCANOFLD_W1_TARGET_CH_BW) |
+			   le32_encode_bits(op->primary_channel,
+					    RTW89_H2C_SCANOFLD_W1_TARGET_PRI_CH) |
+			   le32_encode_bits(op->channel,
+					    RTW89_H2C_SCANOFLD_W1_TARGET_CENTRAL_CH);
+		h2c->w0 |= le32_encode_bits(op->band_type,
+					    RTW89_H2C_SCANOFLD_W0_TARGET_CH_BAND);
 	}
 
 	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
 			      H2C_CAT_MAC, H2C_CL_MAC_FW_OFLD,
 			      H2C_FUNC_SCANOFLD, 1, 1,
-			      H2C_LEN_SCAN_OFFLOAD);
+			      len);
 
 	ret = rtw89_h2c_tx(rtwdev, skb, false);
 	if (ret) {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index 0277caf6ae8a..b878a8869e18 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -2731,96 +2731,32 @@ static inline void RTW89_SET_FWCMD_CHINFO_POWER_IDX(void *cmd, u32 val)
 	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 16), val, GENMASK(15, 0));
 }
 
-static inline void RTW89_SET_FWCMD_SCANOFLD_MACID(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd)), val, GENMASK(7, 0));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_NORM_CY(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd)), val, GENMASK(15, 8));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_PORT_ID(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd)), val, GENMASK(18, 16));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_BAND(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd)), val, BIT(19));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_OPERATION(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd)), val, GENMASK(21, 20));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BAND(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd)), val, GENMASK(23, 22));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_NOTIFY_END(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, BIT(0));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_MODE(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, BIT(1));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_START_MODE(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, BIT(2));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_SCAN_TYPE(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, GENMASK(4, 3));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_TARGET_CH_BW(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, GENMASK(7, 5));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_TARGET_PRI_CH(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, GENMASK(15, 8));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_TARGET_CENTRAL_CH(void *cmd,
-							      u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, GENMASK(23, 16));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_PROBE_REQ_PKT_ID(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 4), val, GENMASK(31, 24));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_NORM_PD(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 8), val, GENMASK(15, 0));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_SLOW_PD(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 8), val, GENMASK(23, 16));
-}
-
-static inline void RTW89_SET_FWCMD_SCANOFLD_TSF_HIGH(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 12), val, GENMASK(31, 0));
-}
+struct rtw89_h2c_scanofld {
+	__le32 w0;
+	__le32 w1;
+	__le32 w2;
+	__le32 tsf_high;
+	__le32 tsf_low;
+	__le32 w5;
+	__le32 w6;
+} __packed;
 
-static inline void RTW89_SET_FWCMD_SCANOFLD_TSF_SLOW(void *cmd, u32 val)
-{
-	le32p_replace_bits((__le32 *)((u8 *)(cmd) + 16), val, GENMASK(31, 0));
-}
+#define RTW89_H2C_SCANOFLD_W0_MACID GENMASK(7, 0)
+#define RTW89_H2C_SCANOFLD_W0_NORM_CY GENMASK(15, 8)
+#define RTW89_H2C_SCANOFLD_W0_PORT_ID GENMASK(18, 16)
+#define RTW89_H2C_SCANOFLD_W0_BAND BIT(19)
+#define RTW89_H2C_SCANOFLD_W0_OPERATION GENMASK(21, 20)
+#define RTW89_H2C_SCANOFLD_W0_TARGET_CH_BAND GENMASK(23, 22)
+#define RTW89_H2C_SCANOFLD_W1_NOTIFY_END BIT(0)
+#define RTW89_H2C_SCANOFLD_W1_TARGET_CH_MODE BIT(1)
+#define RTW89_H2C_SCANOFLD_W1_START_MODE BIT(2)
+#define RTW89_H2C_SCANOFLD_W1_SCAN_TYPE GENMASK(4, 3)
+#define RTW89_H2C_SCANOFLD_W1_TARGET_CH_BW GENMASK(7, 5)
+#define RTW89_H2C_SCANOFLD_W1_TARGET_PRI_CH GENMASK(15, 8)
+#define RTW89_H2C_SCANOFLD_W1_TARGET_CENTRAL_CH GENMASK(23, 16)
+#define RTW89_H2C_SCANOFLD_W1_PROBE_REQ_PKT_ID GENMASK(31, 24)
+#define RTW89_H2C_SCANOFLD_W2_NORM_PD GENMASK(15, 0)
+#define RTW89_H2C_SCANOFLD_W2_SLOW_PD GENMASK(23, 16)
 
 static inline void RTW89_SET_FWCMD_P2P_MACID(void *cmd, u32 val)
 {
-- 
cgit v1.2.3-70-g09d2


From ac83f380905591beecfe5b29a9ef811e35a3aa8d Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Sat, 15 Apr 2023 11:48:58 +0800
Subject: wifi: rtw89: update statistics to FW for fine-tuning performance

Since firmware can't have proper statistics, driver update the
statistics periodically to firmware to assist in tuning performance.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230415034900.15679-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c |  4 ++-
 drivers/net/wireless/realtek/rtw89/fw.c   | 41 +++++++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/fw.h   | 10 ++++++++
 3 files changed, 54 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 8672bfc4801c..8bbc05571104 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -2494,8 +2494,10 @@ static bool rtw89_traffic_stats_track(struct rtw89_dev *rtwdev)
 	bool tfc_changed;
 
 	tfc_changed = rtw89_traffic_stats_calc(rtwdev, &rtwdev->stats);
-	rtw89_for_each_rtwvif(rtwdev, rtwvif)
+	rtw89_for_each_rtwvif(rtwdev, rtwvif) {
 		rtw89_traffic_stats_calc(rtwdev, &rtwvif->stats);
+		rtw89_fw_h2c_tp_offload(rtwdev, rtwvif);
+	}
 
 	return tfc_changed;
 }
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index f55db06f75ef..b9b675bf9d05 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -1855,6 +1855,47 @@ fail:
 	return ret;
 }
 
+int rtw89_fw_h2c_tp_offload(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
+{
+	struct rtw89_traffic_stats *stats = &rtwvif->stats;
+	struct rtw89_h2c_ofld *h2c;
+	u32 len = sizeof(*h2c);
+	struct sk_buff *skb;
+	int ret;
+
+	if (rtwvif->net_type != RTW89_NET_TYPE_INFRA)
+		return -EINVAL;
+
+	skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len);
+	if (!skb) {
+		rtw89_err(rtwdev, "failed to alloc skb for h2c tp\n");
+		return -ENOMEM;
+	}
+
+	skb_put(skb, len);
+	h2c = (struct rtw89_h2c_ofld *)skb->data;
+
+	h2c->w0 = le32_encode_bits(rtwvif->mac_id, RTW89_H2C_OFLD_W0_MAC_ID) |
+		  le32_encode_bits(stats->tx_throughput, RTW89_H2C_OFLD_W0_TX_TP) |
+		  le32_encode_bits(stats->rx_throughput, RTW89_H2C_OFLD_W0_RX_TP);
+
+	rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C,
+			      H2C_CAT_MAC, H2C_CL_MAC_FW_OFLD,
+			      H2C_FUNC_OFLD_TP, 0, 1, len);
+
+	ret = rtw89_h2c_tx(rtwdev, skb, false);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to send h2c\n");
+		goto fail;
+	}
+
+	return 0;
+fail:
+	dev_kfree_skb_any(skb);
+
+	return ret;
+}
+
 #define H2C_RA_LEN 16
 int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index b878a8869e18..675f85c41471 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -3361,6 +3361,14 @@ struct rtw89_h2c_ofld_rssi {
 #define RTW89_H2C_OFLD_RSSI_W0_NUM GENMASK(15, 8)
 #define RTW89_H2C_OFLD_RSSI_W1_VAL GENMASK(7, 0)
 
+struct rtw89_h2c_ofld {
+	__le32 w0;
+} __packed;
+
+#define RTW89_H2C_OFLD_W0_MAC_ID GENMASK(7, 0)
+#define RTW89_H2C_OFLD_W0_TX_TP GENMASK(17, 8)
+#define RTW89_H2C_OFLD_W0_RX_TP GENMASK(27, 18)
+
 #define RTW89_FW_HDR_SIZE 32
 #define RTW89_FW_SECTION_HDR_SIZE 16
 
@@ -3499,6 +3507,7 @@ struct rtw89_fw_h2c_rf_reg_info {
 #define H2C_FUNC_PKT_DROP		0x1b
 #define H2C_FUNC_CFG_BCNFLTR		0x1e
 #define H2C_FUNC_OFLD_RSSI		0x1f
+#define H2C_FUNC_OFLD_TP		0x20
 
 /* CLASS 10 - Security CAM */
 #define H2C_CL_MAC_SEC_CAM		0xa
@@ -3605,6 +3614,7 @@ int rtw89_fw_h2c_set_bcn_fltr_cfg(struct rtw89_dev *rtwdev,
 				  bool connect);
 int rtw89_fw_h2c_rssi_offload(struct rtw89_dev *rtwdev,
 			      struct rtw89_rx_phy_ppdu *phy_ppdu);
+int rtw89_fw_h2c_tp_offload(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
 int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi);
 int rtw89_fw_h2c_cxdrv_init(struct rtw89_dev *rtwdev);
 int rtw89_fw_h2c_cxdrv_role(struct rtw89_dev *rtwdev);
-- 
cgit v1.2.3-70-g09d2


From 982a91642708bb55246a7b169cb573866260124c Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Sat, 15 Apr 2023 11:50:15 +0800
Subject: wifi: rtw89: Disallow power save with multiple stations

Power saving for more than one station is not supported currently.
Disallow entering PS mode when we have more than one associated
stations.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230415035016.15788-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 8bbc05571104..a589f0902a92 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -2509,6 +2509,9 @@ static void rtw89_vif_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwv
 	    rtwvif->tdls_peer)
 		return;
 
+	if (rtwdev->total_sta_assoc > 1)
+		return;
+
 	if (rtwvif->offchan)
 		return;
 
-- 
cgit v1.2.3-70-g09d2


From f22c0bffe8d9528ace89a853c6065b79dcb88c43 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Sat, 15 Apr 2023 11:50:16 +0800
Subject: wifi: rtw89: add support of concurrent mode

Add iface_combination declaration to enable concurrent mode. Only two
interfaces under same frequency is supported currently. We limit the
role combination to be STA + P2P or STA + AP only for now until new
feature is requested.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230415035016.15788-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index a589f0902a92..7fc0a26a4d73 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -156,6 +156,28 @@ static struct ieee80211_rate rtw89_bitrates[] = {
 	{ .bitrate = 540, .hw_value = 0x0b, },
 };
 
+static const struct ieee80211_iface_limit rtw89_iface_limits[] = {
+	{
+		.max = 1,
+		.types = BIT(NL80211_IFTYPE_STATION),
+	},
+	{
+		.max = 1,
+		.types = BIT(NL80211_IFTYPE_P2P_CLIENT) |
+			 BIT(NL80211_IFTYPE_P2P_GO) |
+			 BIT(NL80211_IFTYPE_AP),
+	},
+};
+
+static const struct ieee80211_iface_combination rtw89_iface_combs[] = {
+	{
+		.limits = rtw89_iface_limits,
+		.n_limits = ARRAY_SIZE(rtw89_iface_limits),
+		.max_interfaces = 2,
+		.num_different_channels = 1,
+	}
+};
+
 bool rtw89_ra_report_to_bitrate(struct rtw89_dev *rtwdev, u8 rpt_rate, u16 *bitrate)
 {
 	struct ieee80211_rate rate;
@@ -3834,6 +3856,9 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 	if (!hw)
 		goto err;
 
+	hw->wiphy->iface_combinations = rtw89_iface_combs;
+	hw->wiphy->n_iface_combinations = ARRAY_SIZE(rtw89_iface_combs);
+
 	rtwdev = hw->priv;
 	rtwdev->hw = hw;
 	rtwdev->dev = device;
-- 
cgit v1.2.3-70-g09d2


From 5c690db63b45c6c4c4932b13173af71df369dba5 Mon Sep 17 00:00:00 2001
From: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Date: Tue, 28 Mar 2023 12:41:50 +0530
Subject: wifi: ath11k: Disable Spectral scan upon removing interface

Host might receive spectral events during interface
down sequence and this might create below errors.

failed to handle dma buf release event -22
failed to handle dma buf release event -22

Fix this by disabling spectral config during remove interface.

Tested-on: IPQ5018 hw1.0 AHB WLAN.HK.2.6.0.1-00861-QCAHKSWPL_SILICONZ-1

Signed-off-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230328071150.29645-1-quic_tamizhr@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index c0dc3e387aad..4447fd4bd68f 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -6685,6 +6685,11 @@ static void ath11k_mac_op_remove_interface(struct ieee80211_hw *hw,
 	ath11k_dbg(ab, ATH11K_DBG_MAC, "mac remove interface (vdev %d)\n",
 		   arvif->vdev_id);
 
+	ret = ath11k_spectral_vif_stop(arvif);
+	if (ret)
+		ath11k_warn(ab, "failed to stop spectral for vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+
 	if (arvif->vdev_type == WMI_VDEV_TYPE_STA)
 		ath11k_mac_11d_scan_stop(ar);
 
-- 
cgit v1.2.3-70-g09d2


From abf57d84973ce1abcb67504ac0df8aea1fe09a76 Mon Sep 17 00:00:00 2001
From: Youghandhar Chintala <quic_youghand@quicinc.com>
Date: Tue, 28 Mar 2023 17:04:55 +0530
Subject: wifi: ath11k: enable SAR support on WCN6750

Currently, SAR is enabled only on WCN6855, enable this for WCN6750 too. This
functionality gets triggered, when the user space application calls
NL80211_CMD_SET_SAR_SPECS.

Tested-on: WCN6750 hw1.0 AHB WLAN.MSL.1.0.1-00887-QCAMSLSWPLZ-1

Signed-off-by: Youghandhar Chintala <quic_youghand@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230328113455.11252-1-quic_youghand@quicinc.com
---
 drivers/net/wireless/ath/ath11k/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index 498310e38d44..b1b90bd34d67 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -593,7 +593,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.current_cc_support = true,
 		.dbr_debug_support = false,
 		.global_reset = false,
-		.bios_sar_capa = NULL,
+		.bios_sar_capa = &ath11k_hw_sar_capa_wcn6855,
 		.m3_fw_support = false,
 		.fixed_bdf_addr = false,
 		.fixed_mem_region = false,
-- 
cgit v1.2.3-70-g09d2


From 06c58473969239e00d76b683edd511952060ca56 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Thu, 30 Mar 2023 16:37:18 +0200
Subject: wifi: ath11k: pci: Add more MODULE_FIRMWARE() entries

As there are a few more models supported by the driver, let's add the
missing MODULE_FIRMWARE() entries for them.  The lack of them resulted
in the missing device enablement on some systems, such as the
installation image of openSUSE.

While we are at it, use the wildcard instead of listing each firmware
files individually for each.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230330143718.19511-1-tiwai@suse.de
---
 drivers/net/wireless/ath/ath11k/pci.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/pci.c b/drivers/net/wireless/ath/ath11k/pci.c
index 522d01e5e11c..7b33731a50ee 100644
--- a/drivers/net/wireless/ath/ath11k/pci.c
+++ b/drivers/net/wireless/ath/ath11k/pci.c
@@ -1036,7 +1036,8 @@ module_exit(ath11k_pci_exit);
 MODULE_DESCRIPTION("Driver support for Qualcomm Technologies 802.11ax WLAN PCIe devices");
 MODULE_LICENSE("Dual BSD/GPL");
 
-/* QCA639x 2.0 firmware files */
-MODULE_FIRMWARE(ATH11K_FW_DIR "/QCA6390/hw2.0/" ATH11K_BOARD_API2_FILE);
-MODULE_FIRMWARE(ATH11K_FW_DIR "/QCA6390/hw2.0/" ATH11K_AMSS_FILE);
-MODULE_FIRMWARE(ATH11K_FW_DIR "/QCA6390/hw2.0/" ATH11K_M3_FILE);
+/* firmware files */
+MODULE_FIRMWARE(ATH11K_FW_DIR "/QCA6390/hw2.0/*");
+MODULE_FIRMWARE(ATH11K_FW_DIR "/QCN9074/hw1.0/*");
+MODULE_FIRMWARE(ATH11K_FW_DIR "/WCN6855/hw2.0/*");
+MODULE_FIRMWARE(ATH11K_FW_DIR "/WCN6855/hw2.1/*");
-- 
cgit v1.2.3-70-g09d2


From a87a9110ac0dcbfd9458b6665c141fa1c16a669d Mon Sep 17 00:00:00 2001
From: Kalle Valo <quic_kvalo@quicinc.com>
Date: Wed, 5 Apr 2023 12:04:25 +0300
Subject: wifi: ath11k: print a warning when crypto_alloc_shash() fails

Christoph reported that ath11k failed to initialise when michael_mic.ko
module was not installed. To make it easier to notice that case print a
warning when crypto_alloc_shash() fails.

Compile tested only.

Reported-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/all/20221130133016.GC3055@lst.de/
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230405090425.1351-1-kvalo@kernel.org
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 99859b59138e..d22e5349c551 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -3106,8 +3106,11 @@ int ath11k_peer_rx_frag_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id
 	int i;
 
 	tfm = crypto_alloc_shash("michael_mic", 0, 0);
-	if (IS_ERR(tfm))
+	if (IS_ERR(tfm)) {
+		ath11k_warn(ab, "failed to allocate michael_mic shash: %ld\n",
+			    PTR_ERR(tfm));
 		return PTR_ERR(tfm);
+	}
 
 	spin_lock_bh(&ab->base_lock);
 
-- 
cgit v1.2.3-70-g09d2


From a06bfb3c9f69f303692cdae87bc0899d2ae8b2a6 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Tue, 4 Apr 2023 00:11:54 +0530
Subject: wifi: ath11k: Ignore frags from uninitialized peer in dp.

When max virtual ap interfaces are configured in all the bands with
ACS and hostapd restart is done every 60s, a crash is observed at
random times.
In this certain scenario, a fragmented packet is received for
self peer, for which rx_tid and rx_frags are not initialized in
datapath. While handling this fragment, crash is observed as the
rx_frag list is uninitialised and when we walk in
ath11k_dp_rx_h_sort_frags, skb null leads to exception.

To address this, before processing received fragments we check
dp_setup_done flag is set to ensure that peer has completed its
dp peer setup for fragment queue, else ignore processing the
fragments.

Call trace:
  ath11k_dp_process_rx_err+0x550/0x1084 [ath11k]
  ath11k_dp_service_srng+0x70/0x370 [ath11k]
  0xffffffc009693a04
  __napi_poll+0x30/0xa4
  net_rx_action+0x118/0x270
  __do_softirq+0x10c/0x244
  irq_exit+0x64/0xb4
  __handle_domain_irq+0x88/0xac
  gic_handle_irq+0x74/0xbc
  el1_irq+0xf0/0x1c0
  arch_cpu_idle+0x10/0x18
  do_idle+0x104/0x248
  cpu_startup_entry+0x20/0x64
  rest_init+0xd0/0xdc
  arch_call_rest_init+0xc/0x14
  start_kernel+0x480/0x4b8
  Code: f9400281 f94066a2 91405021 b94a0023 (f9406401)

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Nagarajan Maran <quic_nmaran@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403184155.8670-2-quic_nmaran@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp.c    | 4 +++-
 drivers/net/wireless/ath/ath11k/dp_rx.c | 8 ++++++++
 drivers/net/wireless/ath/ath11k/peer.h  | 1 +
 3 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
index f5156a7fbdd7..d070bcb3fe24 100644
--- a/drivers/net/wireless/ath/ath11k/dp.c
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -36,6 +36,7 @@ void ath11k_dp_peer_cleanup(struct ath11k *ar, int vdev_id, const u8 *addr)
 	}
 
 	ath11k_peer_rx_tid_cleanup(ar, peer);
+	peer->dp_setup_done = false;
 	crypto_free_shash(peer->tfm_mmic);
 	spin_unlock_bh(&ab->base_lock);
 }
@@ -72,7 +73,8 @@ int ath11k_dp_peer_setup(struct ath11k *ar, int vdev_id, const u8 *addr)
 	ret = ath11k_peer_rx_frag_setup(ar, addr, vdev_id);
 	if (ret) {
 		ath11k_warn(ab, "failed to setup rx defrag context\n");
-		return ret;
+		tid--;
+		goto peer_clean;
 	}
 
 	/* TODO: Setup other peer specific resource used in data path */
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index d22e5349c551..ab6b18de6371 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -3130,6 +3130,7 @@ int ath11k_peer_rx_frag_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id
 	}
 
 	peer->tfm_mmic = tfm;
+	peer->dp_setup_done = true;
 	spin_unlock_bh(&ab->base_lock);
 
 	return 0;
@@ -3575,6 +3576,13 @@ static int ath11k_dp_rx_frag_h_mpdu(struct ath11k *ar,
 		ret = -ENOENT;
 		goto out_unlock;
 	}
+	if (!peer->dp_setup_done) {
+		ath11k_warn(ab, "The peer %pM [%d] has uninitialized datapath\n",
+			    peer->addr, peer_id);
+		ret = -ENOENT;
+		goto out_unlock;
+	}
+
 	rx_tid = &peer->rx_tid[tid];
 
 	if ((!skb_queue_empty(&rx_tid->rx_frags) && seqno != rx_tid->cur_sn) ||
diff --git a/drivers/net/wireless/ath/ath11k/peer.h b/drivers/net/wireless/ath/ath11k/peer.h
index 6dd17bafe3a0..9bd385d0a38c 100644
--- a/drivers/net/wireless/ath/ath11k/peer.h
+++ b/drivers/net/wireless/ath/ath11k/peer.h
@@ -35,6 +35,7 @@ struct ath11k_peer {
 	u16 sec_type;
 	u16 sec_type_grp;
 	bool is_authorized;
+	bool dp_setup_done;
 };
 
 void ath11k_peer_unmap_event(struct ath11k_base *ab, u16 peer_id);
-- 
cgit v1.2.3-70-g09d2


From 41e02bf4ae32cf2ac47b08b4caaa9c1a032e4ce7 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Tue, 4 Apr 2023 00:11:55 +0530
Subject: wifi: ath11k: fix undefined behavior with __fls in dp

"__fls" would have an undefined behavior if the argument is passed
as "0". Hence, added changes to handle the same.

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Nagarajan Maran <quic_nmaran@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403184155.8670-3-quic_nmaran@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index ab6b18de6371..847da167a41c 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -3598,7 +3598,7 @@ static int ath11k_dp_rx_frag_h_mpdu(struct ath11k *ar,
 		goto out_unlock;
 	}
 
-	if (frag_no > __fls(rx_tid->rx_frag_bitmap))
+	if (!rx_tid->rx_frag_bitmap || (frag_no > __fls(rx_tid->rx_frag_bitmap)))
 		__skb_queue_tail(&rx_tid->rx_frags, msdu);
 	else
 		ath11k_dp_rx_h_sort_frags(ar, &rx_tid->rx_frags, msdu);
-- 
cgit v1.2.3-70-g09d2


From e04e4b6e01e71978d5df92a5f25b472e1827b46b Mon Sep 17 00:00:00 2001
From: Jonas Jelonek <jelonek.jonas@gmail.com>
Date: Thu, 30 Mar 2023 15:21:59 +0200
Subject: wifi: ath9k: fix per-packet TX-power cap for TPC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix incorrect usage of plain rate_idx as index into the max (power) per
rate lookup table.

For transmit power control (TPC), the ath9k driver maintains internal
tables (in struct ath_hw) to store the max allowed power level per rate.
They are used to limit a given TX-power according to regulatory and user
limits in the TX-path per packet. The tables are filled in a predefined
order, starting with values for CCK + OFDM rates and followed by the
values for MCS rates. Thus, the maximum power levels for MCS do not
start at index 0 in the table but are shifted by a fixed value.

The TX-power limiting in ath_get_rate_txpower currently does not apply
this shift, thus retrieves the incorrect maximum power level for a given
rate. In particular for MCS rates, the maximum power levels for CCK/OFDM
rates were used, e.g. maximum power for OFDM 0 was used for MCS 0. If
STBC is used, the power is mostly limited to 0 because the STBC table
is zeroed for legacy CCK/OFDM rates. Encountered this during testing of
our work-in-progress TPC per packet for ath9k.
This only has an effect when TPC is enabled in ath9k (tpc_enabled in
struct ath_hw) which defaults to false. In this case it has a
significant impact on the used TX-power, throughput + RSSI. Otherwise
the affected code is just skipped and TX-power is limited with the
hardware registers only. This patch fixes this table lookup.

Tested on OpenWrt (kernel 5.15.98, but backported ath9k driver) with
small desk setup using ath9k chips AR9280 and AR9580. Cap of TX-power is
working properly for all rates now, throughput and RSSI as expected,
equal to as if TPC was disabled.
Compile-tested with latest 6.3 kernel + allyesconfig.

Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230330132159.758088-1-jelonek.jonas@gmail.com
---
 drivers/net/wireless/ath/ath9k/xmit.c | 30 ++++++++++++++++++++++++------
 1 file changed, 24 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/xmit.c b/drivers/net/wireless/ath/ath9k/xmit.c
index ef9a8e0b75e6..f6f2ab7a63ff 100644
--- a/drivers/net/wireless/ath/ath9k/xmit.c
+++ b/drivers/net/wireless/ath/ath9k/xmit.c
@@ -34,6 +34,12 @@
 #define NUM_SYMBOLS_PER_USEC(_usec) (_usec >> 2)
 #define NUM_SYMBOLS_PER_USEC_HALFGI(_usec) (((_usec*5)-4)/18)
 
+/* Shifts in ar5008_phy.c and ar9003_phy.c are equal for all revisions */
+#define ATH9K_PWRTBL_11NA_OFDM_SHIFT    0
+#define ATH9K_PWRTBL_11NG_OFDM_SHIFT    4
+#define ATH9K_PWRTBL_11NA_HT_SHIFT      8
+#define ATH9K_PWRTBL_11NG_HT_SHIFT      12
+
 
 static u16 bits_per_symbol[][2] = {
 	/* 20MHz 40MHz */
@@ -1169,13 +1175,14 @@ void ath_update_max_aggr_framelen(struct ath_softc *sc, int queue, int txop)
 }
 
 static u8 ath_get_rate_txpower(struct ath_softc *sc, struct ath_buf *bf,
-			       u8 rateidx, bool is_40, bool is_cck)
+			       u8 rateidx, bool is_40, bool is_cck, bool is_mcs)
 {
 	u8 max_power;
 	struct sk_buff *skb;
 	struct ath_frame_info *fi;
 	struct ieee80211_tx_info *info;
 	struct ath_hw *ah = sc->sc_ah;
+	bool is_2ghz, is_5ghz, use_stbc;
 
 	if (sc->tx99_state || !ah->tpc_enabled)
 		return MAX_RATE_POWER;
@@ -1184,6 +1191,19 @@ static u8 ath_get_rate_txpower(struct ath_softc *sc, struct ath_buf *bf,
 	fi = get_frame_info(skb);
 	info = IEEE80211_SKB_CB(skb);
 
+	is_2ghz = info->band == NL80211_BAND_2GHZ;
+	is_5ghz = info->band == NL80211_BAND_5GHZ;
+	use_stbc = is_mcs && rateidx < 8 && (info->flags &
+					     IEEE80211_TX_CTL_STBC);
+
+	if (is_mcs)
+		rateidx += is_5ghz ? ATH9K_PWRTBL_11NA_HT_SHIFT
+				   : ATH9K_PWRTBL_11NG_HT_SHIFT;
+	else if (is_2ghz && !is_cck)
+		rateidx += ATH9K_PWRTBL_11NG_OFDM_SHIFT;
+	else
+		rateidx += ATH9K_PWRTBL_11NA_OFDM_SHIFT;
+
 	if (!AR_SREV_9300_20_OR_LATER(ah)) {
 		int txpower = fi->tx_power;
 
@@ -1193,10 +1213,8 @@ static u8 ath_get_rate_txpower(struct ath_softc *sc, struct ath_buf *bf,
 			u16 eeprom_rev = ah->eep_ops->get_eeprom_rev(ah);
 
 			if (eeprom_rev >= AR5416_EEP_MINOR_VER_2) {
-				bool is_2ghz;
 				struct modal_eep_header *pmodal;
 
-				is_2ghz = info->band == NL80211_BAND_2GHZ;
 				pmodal = &eep->modalHeader[is_2ghz];
 				power_ht40delta = pmodal->ht40PowerIncForPdadc;
 			} else {
@@ -1229,7 +1247,7 @@ static u8 ath_get_rate_txpower(struct ath_softc *sc, struct ath_buf *bf,
 		if (!max_power && !AR_SREV_9280_20_OR_LATER(ah))
 			max_power = 1;
 	} else if (!bf->bf_state.bfs_paprd) {
-		if (rateidx < 8 && (info->flags & IEEE80211_TX_CTL_STBC))
+		if (use_stbc)
 			max_power = min_t(u8, ah->tx_power_stbc[rateidx],
 					  fi->tx_power);
 		else
@@ -1319,7 +1337,7 @@ static void ath_buf_set_rate(struct ath_softc *sc, struct ath_buf *bf,
 			}
 
 			info->txpower[i] = ath_get_rate_txpower(sc, bf, rix,
-								is_40, false);
+								is_40, false, true);
 			continue;
 		}
 
@@ -1350,7 +1368,7 @@ static void ath_buf_set_rate(struct ath_softc *sc, struct ath_buf *bf,
 
 		is_cck = IS_CCK_RATE(info->rates[i].Rate);
 		info->txpower[i] = ath_get_rate_txpower(sc, bf, rix, false,
-							is_cck);
+							is_cck, false);
 	}
 
 	/* For AR5416 - RTS cannot be followed by a frame larger than 8K */
-- 
cgit v1.2.3-70-g09d2


From 532f0482fc577a78c0086893ab39acb406ac3e65 Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Sat, 4 Feb 2023 14:53:04 +0800
Subject: wifi: mt76: remove redundent MCU_UNI_CMD_* definitions

clear redundent definitions only

Fixes: 5b55b6da982c ("wifi: mt76: mt7921: add unified ROC cmd/event support")
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index a5e6ee4daf92..40a99e0caded 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -967,9 +967,6 @@ enum {
 	DEV_INFO_MAX_NUM
 };
 
-#define MCU_UNI_CMD_EVENT                       BIT(1)
-#define MCU_UNI_CMD_UNSOLICITED_EVENT           BIT(2)
-
 /* event table */
 enum {
 	MCU_EVENT_TARGET_ADDRESS_LEN = 0x01,
-- 
cgit v1.2.3-70-g09d2


From 64822bdba456a145f7cb4c66d9939bf42c64ae62 Mon Sep 17 00:00:00 2001
From: Frank Wunderlich <frank-w@public-files.de>
Date: Tue, 7 Feb 2023 14:35:04 +0100
Subject: dt-bindings: mt76: add active-low property for led

LEDs can be in low-active mode, driver already supports it, but
documentation is missing. Add documentation for the dt property.

Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml b/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml
index 7d526ff53fb7..67b63f119f64 100644
--- a/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml
+++ b/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml
@@ -111,6 +111,11 @@ properties:
     $ref: /schemas/leds/common.yaml#
     additionalProperties: false
     properties:
+      led-active-low:
+        description:
+          LED is enabled with ground signal.
+        type: boolean
+
       led-sources:
         maxItems: 1
 
-- 
cgit v1.2.3-70-g09d2


From fcc51acfebb85dbc3ab1bea3ce4997d7c0a3a38d Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Sat, 11 Feb 2023 09:01:58 +0800
Subject: wifi: mt76: mt7921: fix wrong command to set STA channel

Should not use AND operator to check vif type NL80211_IFTYPE_MONITOR, and
that will cause we go into sniffer command for both STA and MONITOR
mode. However, the sniffer command would set channel properly (with some
extra options), the STA mode still works even if using the wrong
command.

Fix vif type check to make sure we using the right command to update
channel.

Fixes: 914189af23b8 ("wifi: mt76: mt7921: fix channel switch fail in monitor mode")
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index 75eaf86c6a78..a72964e7a807 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -1695,7 +1695,7 @@ static void mt7921_ctx_iter(void *priv, u8 *mac,
 	if (ctx != mvif->ctx)
 		return;
 
-	if (vif->type & NL80211_IFTYPE_MONITOR)
+	if (vif->type == NL80211_IFTYPE_MONITOR)
 		mt7921_mcu_config_sniffer(mvif, ctx);
 	else
 		mt76_connac_mcu_uni_set_chctx(mvif->phy->mt76, &mvif->mt76, ctx);
-- 
cgit v1.2.3-70-g09d2


From 9270270d62191b7549296721e8d5f3dc0df01563 Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Tue, 14 Feb 2023 10:49:57 +0800
Subject: wifi: mt76: mt7921: fix PCI DMA hang after reboot

mt7921 just stop some workers and clean up chip status before reboot.
In stress test, there are working activities still running at the period
of .shutdown callback and that would cause some hosts cannot recover
DMA after reboot. To avoid the floating state in reboot, we use
mt7921_pci_remove() to fully deinit all resources.

Fixes: f23a0cea8bd6 ("wifi: mt76: mt7921e: add pci .shutdown() support")
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index cb72ded37256..1a8a54a46dcc 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -509,17 +509,7 @@ failed:
 
 static void mt7921_pci_shutdown(struct pci_dev *pdev)
 {
-	struct mt76_dev *mdev = pci_get_drvdata(pdev);
-	struct mt7921_dev *dev = container_of(mdev, struct mt7921_dev, mt76);
-	struct mt76_connac_pm *pm = &dev->pm;
-
-	cancel_delayed_work_sync(&pm->ps_work);
-	cancel_work_sync(&pm->wake_work);
-
-	/* chip cleanup before reboot */
-	mt7921_mcu_drv_pmctrl(dev);
-	mt7921_dma_cleanup(dev);
-	mt7921_wfsys_reset(dev);
+	mt7921_pci_remove(pdev);
 }
 
 static DEFINE_SIMPLE_DEV_PM_OPS(mt7921_pm_ops, mt7921_pci_suspend, mt7921_pci_resume);
-- 
cgit v1.2.3-70-g09d2


From df5e5bfdb1d52c5524f21e34488a37f3fe2badad Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 15 Feb 2023 13:56:50 +0800
Subject: wifi: mt76: mt7996: Remove unneeded semicolon

./drivers/net/wireless/mediatek/mt76/mt7996/mcu.c:3136:3-4: Unneeded semicolon

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4059
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index dbe30832fd88..8ad51cbfdbe8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -3133,7 +3133,7 @@ int mt7996_mcu_get_chip_config(struct mt7996_dev *dev, u32 *cap)
 			break;
 		default:
 			break;
-		};
+		}
 
 		buf += le16_to_cpu(tlv->len);
 	}
-- 
cgit v1.2.3-70-g09d2


From cdc215c2c8d74b3c8886650e979b47f16c1f7f92 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Thu, 16 Feb 2023 15:15:05 +0300
Subject: wifi: mt76: mt7915: unlock on error in mt7915_thermal_temp_store()

Drop the lock before returning -EINVAL.

Fixes: ecaccdae7a7e ("wifi: mt76: mt7915: rework mt7915_thermal_temp_store()")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7915/init.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
index 5e288116b1b0..4f3efc942a4d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
@@ -89,6 +89,7 @@ static ssize_t mt7915_thermal_temp_store(struct device *dev,
 	     val < phy->throttle_temp[MT7915_CRIT_TEMP_IDX])) {
 		dev_err(phy->dev->mt76.dev,
 			"temp1_max shall be greater than temp1_crit.");
+		mutex_unlock(&phy->dev->mt76.mutex);
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 63a3724632464559d1e98d90d2c63d43ec6ef6f5 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sat, 18 Feb 2023 01:46:54 +0800
Subject: wifi: mt76: mt7996: fix radiotap bitfield

The 11be generation's radiotap bitfields were wrongly copy-and-pasted
from 11ax driver, so fix them accordingly.

Fixes: 98686cd21624 ("wifi: mt76: mt7996: add driver for MediaTek Wi-Fi 7 (802.11be) devices")
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 54 +++++++++++++------------
 drivers/net/wireless/mediatek/mt76/mt7996/mac.h | 41 +++++++++----------
 2 files changed, 47 insertions(+), 48 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index c9a9f0e31771..3c3506c7c87a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -260,12 +260,9 @@ mt7996_mac_decode_he_radiotap_ru(struct mt76_rx_status *status,
 				 struct ieee80211_radiotap_he *he,
 				 __le32 *rxv)
 {
-	u32 ru_h, ru_l;
-	u8 ru, offs = 0;
+	u32 ru, offs = 0;
 
-	ru_l = le32_get_bits(rxv[0], MT_PRXV_HE_RU_ALLOC_L);
-	ru_h = le32_get_bits(rxv[1], MT_PRXV_HE_RU_ALLOC_H);
-	ru = (u8)(ru_l | ru_h << 4);
+	ru = le32_get_bits(rxv[0], MT_PRXV_HE_RU_ALLOC);
 
 	status->bw = RATE_INFO_BW_HE_RU;
 
@@ -330,18 +327,23 @@ mt7996_mac_decode_he_mu_radiotap(struct sk_buff *skb, __le32 *rxv)
 
 	he_mu->flags2 |= MU_PREP(FLAGS2_BW_FROM_SIG_A_BW, status->bw) |
 			 MU_PREP(FLAGS2_SIG_B_SYMS_USERS,
-				 le32_get_bits(rxv[2], MT_CRXV_HE_NUM_USER));
+				 le32_get_bits(rxv[4], MT_CRXV_HE_NUM_USER));
 
-	he_mu->ru_ch1[0] = le32_get_bits(rxv[3], MT_CRXV_HE_RU0);
+	he_mu->ru_ch1[0] = le32_get_bits(rxv[16], MT_CRXV_HE_RU0) & 0xff;
 
 	if (status->bw >= RATE_INFO_BW_40) {
 		he_mu->flags1 |= HE_BITS(MU_FLAGS1_CH2_RU_KNOWN);
-		he_mu->ru_ch2[0] = le32_get_bits(rxv[3], MT_CRXV_HE_RU1);
+		he_mu->ru_ch2[0] = le32_get_bits(rxv[16], MT_CRXV_HE_RU1) & 0xff;
 	}
 
 	if (status->bw >= RATE_INFO_BW_80) {
-		he_mu->ru_ch1[1] = le32_get_bits(rxv[3], MT_CRXV_HE_RU2);
-		he_mu->ru_ch2[1] = le32_get_bits(rxv[3], MT_CRXV_HE_RU3);
+		u32 ru_h, ru_l;
+
+		he_mu->ru_ch1[1] = le32_get_bits(rxv[16], MT_CRXV_HE_RU2) & 0xff;
+
+		ru_l = le32_get_bits(rxv[16], MT_CRXV_HE_RU3_L);
+		ru_h = le32_get_bits(rxv[17], MT_CRXV_HE_RU3_H) & 0x7;
+		he_mu->ru_ch2[1] = (u8)(ru_l | ru_h << 4);
 	}
 }
 
@@ -364,23 +366,23 @@ mt7996_mac_decode_he_radiotap(struct sk_buff *skb, __le32 *rxv, u8 mode)
 			 HE_BITS(DATA2_TXOP_KNOWN),
 	};
 	struct ieee80211_radiotap_he *he = NULL;
-	u32 ltf_size = le32_get_bits(rxv[2], MT_CRXV_HE_LTF_SIZE) + 1;
+	u32 ltf_size = le32_get_bits(rxv[4], MT_CRXV_HE_LTF_SIZE) + 1;
 
 	status->flag |= RX_FLAG_RADIOTAP_HE;
 
 	he = skb_push(skb, sizeof(known));
 	memcpy(he, &known, sizeof(known));
 
-	he->data3 = HE_PREP(DATA3_BSS_COLOR, BSS_COLOR, rxv[14]) |
-		    HE_PREP(DATA3_LDPC_XSYMSEG, LDPC_EXT_SYM, rxv[2]);
-	he->data4 = HE_PREP(DATA4_SU_MU_SPTL_REUSE, SR_MASK, rxv[11]);
-	he->data5 = HE_PREP(DATA5_PE_DISAMBIG, PE_DISAMBIG, rxv[2]) |
+	he->data3 = HE_PREP(DATA3_BSS_COLOR, BSS_COLOR, rxv[9]) |
+		    HE_PREP(DATA3_LDPC_XSYMSEG, LDPC_EXT_SYM, rxv[4]);
+	he->data4 = HE_PREP(DATA4_SU_MU_SPTL_REUSE, SR_MASK, rxv[13]);
+	he->data5 = HE_PREP(DATA5_PE_DISAMBIG, PE_DISAMBIG, rxv[5]) |
 		    le16_encode_bits(ltf_size,
 				     IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE);
 	if (le32_to_cpu(rxv[0]) & MT_PRXV_TXBF)
 		he->data5 |= HE_BITS(DATA5_TXBF);
-	he->data6 = HE_PREP(DATA6_TXOP, TXOP_DUR, rxv[14]) |
-		    HE_PREP(DATA6_DOPPLER, DOPPLER, rxv[14]);
+	he->data6 = HE_PREP(DATA6_TXOP, TXOP_DUR, rxv[9]) |
+		    HE_PREP(DATA6_DOPPLER, DOPPLER, rxv[9]);
 
 	switch (mode) {
 	case MT_PHY_TYPE_HE_SU:
@@ -389,22 +391,22 @@ mt7996_mac_decode_he_radiotap(struct sk_buff *skb, __le32 *rxv, u8 mode)
 			     HE_BITS(DATA1_BEAM_CHANGE_KNOWN) |
 			     HE_BITS(DATA1_BW_RU_ALLOC_KNOWN);
 
-		he->data3 |= HE_PREP(DATA3_BEAM_CHANGE, BEAM_CHNG, rxv[14]) |
-			     HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
+		he->data3 |= HE_PREP(DATA3_BEAM_CHANGE, BEAM_CHNG, rxv[8]) |
+			     HE_PREP(DATA3_UL_DL, UPLINK, rxv[5]);
 		break;
 	case MT_PHY_TYPE_HE_EXT_SU:
 		he->data1 |= HE_BITS(DATA1_FORMAT_EXT_SU) |
 			     HE_BITS(DATA1_UL_DL_KNOWN) |
 			     HE_BITS(DATA1_BW_RU_ALLOC_KNOWN);
 
-		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
+		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[5]);
 		break;
 	case MT_PHY_TYPE_HE_MU:
 		he->data1 |= HE_BITS(DATA1_FORMAT_MU) |
 			     HE_BITS(DATA1_UL_DL_KNOWN);
 
-		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
-		he->data4 |= HE_PREP(DATA4_MU_STA_ID, MU_AID, rxv[7]);
+		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[5]);
+		he->data4 |= HE_PREP(DATA4_MU_STA_ID, MU_AID, rxv[8]);
 
 		mt7996_mac_decode_he_radiotap_ru(status, he, rxv);
 		mt7996_mac_decode_he_mu_radiotap(skb, rxv);
@@ -415,10 +417,10 @@ mt7996_mac_decode_he_radiotap(struct sk_buff *skb, __le32 *rxv, u8 mode)
 			     HE_BITS(DATA1_SPTL_REUSE3_KNOWN) |
 			     HE_BITS(DATA1_SPTL_REUSE4_KNOWN);
 
-		he->data4 |= HE_PREP(DATA4_TB_SPTL_REUSE1, SR_MASK, rxv[11]) |
-			     HE_PREP(DATA4_TB_SPTL_REUSE2, SR1_MASK, rxv[11]) |
-			     HE_PREP(DATA4_TB_SPTL_REUSE3, SR2_MASK, rxv[11]) |
-			     HE_PREP(DATA4_TB_SPTL_REUSE4, SR3_MASK, rxv[11]);
+		he->data4 |= HE_PREP(DATA4_TB_SPTL_REUSE1, SR_MASK, rxv[13]) |
+			     HE_PREP(DATA4_TB_SPTL_REUSE2, SR1_MASK, rxv[13]) |
+			     HE_PREP(DATA4_TB_SPTL_REUSE3, SR2_MASK, rxv[13]) |
+			     HE_PREP(DATA4_TB_SPTL_REUSE4, SR3_MASK, rxv[13]);
 
 		mt7996_mac_decode_he_radiotap_ru(status, he, rxv);
 		break;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
index 27184cbac619..2cc218f735d8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
@@ -102,8 +102,7 @@ enum rx_pkt_type {
 #define MT_PRXV_NSTS			GENMASK(10, 7)
 #define MT_PRXV_TXBF			BIT(11)
 #define MT_PRXV_HT_AD_CODE		BIT(12)
-#define MT_PRXV_HE_RU_ALLOC_L		GENMASK(31, 28)
-#define MT_PRXV_HE_RU_ALLOC_H		GENMASK(3, 0)
+#define MT_PRXV_HE_RU_ALLOC		GENMASK(30, 22)
 #define MT_PRXV_RCPI3			GENMASK(31, 24)
 #define MT_PRXV_RCPI2			GENMASK(23, 16)
 #define MT_PRXV_RCPI1			GENMASK(15, 8)
@@ -113,34 +112,32 @@ enum rx_pkt_type {
 #define MT_PRXV_TX_MODE			GENMASK(14, 11)
 #define MT_PRXV_FRAME_MODE		GENMASK(2, 0)
 #define MT_PRXV_DCM			BIT(5)
-#define MT_PRXV_NUM_RX			BIT(8, 6)
 
 /* C-RXV */
-#define MT_CRXV_HT_STBC			GENMASK(1, 0)
-#define MT_CRXV_TX_MODE			GENMASK(7, 4)
-#define MT_CRXV_FRAME_MODE		GENMASK(10, 8)
-#define MT_CRXV_HT_SHORT_GI		GENMASK(14, 13)
-#define MT_CRXV_HE_LTF_SIZE		GENMASK(18, 17)
-#define MT_CRXV_HE_LDPC_EXT_SYM		BIT(20)
-#define MT_CRXV_HE_PE_DISAMBIG		BIT(23)
-#define MT_CRXV_HE_NUM_USER		GENMASK(30, 24)
-#define MT_CRXV_HE_UPLINK		BIT(31)
-#define MT_CRXV_HE_RU0			GENMASK(7, 0)
-#define MT_CRXV_HE_RU1			GENMASK(15, 8)
-#define MT_CRXV_HE_RU2			GENMASK(23, 16)
-#define MT_CRXV_HE_RU3			GENMASK(31, 24)
-
-#define MT_CRXV_HE_MU_AID		GENMASK(30, 20)
+#define MT_CRXV_HE_NUM_USER		GENMASK(26, 20)
+#define MT_CRXV_HE_LTF_SIZE		GENMASK(28, 27)
+#define MT_CRXV_HE_LDPC_EXT_SYM		BIT(30)
+
+#define MT_CRXV_HE_PE_DISAMBIG		BIT(1)
+#define MT_CRXV_HE_UPLINK		BIT(2)
+
+#define MT_CRXV_HE_MU_AID		GENMASK(27, 17)
+#define MT_CRXV_HE_BEAM_CHNG		BIT(29)
+
+#define MT_CRXV_HE_DOPPLER		BIT(0)
+#define MT_CRXV_HE_BSS_COLOR		GENMASK(15, 10)
+#define MT_CRXV_HE_TXOP_DUR		GENMASK(19, 17)
 
 #define MT_CRXV_HE_SR_MASK		GENMASK(11, 8)
 #define MT_CRXV_HE_SR1_MASK		GENMASK(16, 12)
 #define MT_CRXV_HE_SR2_MASK             GENMASK(20, 17)
 #define MT_CRXV_HE_SR3_MASK             GENMASK(24, 21)
 
-#define MT_CRXV_HE_BSS_COLOR		GENMASK(5, 0)
-#define MT_CRXV_HE_TXOP_DUR		GENMASK(12, 6)
-#define MT_CRXV_HE_BEAM_CHNG		BIT(13)
-#define MT_CRXV_HE_DOPPLER		BIT(16)
+#define MT_CRXV_HE_RU0			GENMASK(8, 0)
+#define MT_CRXV_HE_RU1			GENMASK(17, 9)
+#define MT_CRXV_HE_RU2			GENMASK(26, 18)
+#define MT_CRXV_HE_RU3_L		GENMASK(31, 27)
+#define MT_CRXV_HE_RU3_H		GENMASK(3, 0)
 
 enum tx_header_format {
 	MT_HDR_FORMAT_802_3,
-- 
cgit v1.2.3-70-g09d2


From c278a64a9375b5410a25d3e17317c36ee9e2fd02 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sat, 18 Feb 2023 01:47:32 +0800
Subject: wifi: mt76: dynamic channel bandwidth changes in AP mode

Allow AP to change channel width for 40Mhz intolerant STA on the
2.4 GHz band.

Signed-off-by: Himanshu Goyal <himanshu.goyal@mediatek.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mac80211.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c
index 34abf70f44af..87902f4b8e12 100644
--- a/drivers/net/wireless/mediatek/mt76/mac80211.c
+++ b/drivers/net/wireless/mediatek/mt76/mac80211.c
@@ -418,7 +418,8 @@ mt76_phy_init(struct mt76_phy *phy, struct ieee80211_hw *hw)
 	SET_IEEE80211_DEV(hw, dev->dev);
 	SET_IEEE80211_PERM_ADDR(hw, phy->macaddr);
 
-	wiphy->features |= NL80211_FEATURE_ACTIVE_MONITOR;
+	wiphy->features |= NL80211_FEATURE_ACTIVE_MONITOR |
+			   NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE;
 	wiphy->flags |= WIPHY_FLAG_HAS_CHANNEL_SWITCH |
 			WIPHY_FLAG_SUPPORTS_TDLS |
 			WIPHY_FLAG_AP_UAPSD;
-- 
cgit v1.2.3-70-g09d2


From 90fb69212c60e26ef70ed0e8532b116c7649ac88 Mon Sep 17 00:00:00 2001
From: Lorenz Brun <lorenz@brun.one>
Date: Sat, 18 Feb 2023 12:29:45 +0100
Subject: wifi: mt76: mt7915: expose device tree match table

On MT7986 the WiFi driver currently does not get automatically loaded,
requiring manual modprobing because the device tree compatibles are not
exported into metadata.

Add the missing MODULE_DEVICE_TABLE macro to fix this.

Fixes: 99ad32a4ca3a2 ("mt76: mt7915: add support for MT7986")
Signed-off-by: Lorenz Brun <lorenz@brun.one>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7915/soc.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/soc.c b/drivers/net/wireless/mediatek/mt76/mt7915/soc.c
index 2ac0a0f2859c..32c137066e7f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/soc.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/soc.c
@@ -1239,6 +1239,8 @@ static const struct of_device_id mt7986_wmac_of_match[] = {
 	{},
 };
 
+MODULE_DEVICE_TABLE(of, mt7986_wmac_of_match);
+
 struct platform_driver mt7986_wmac_driver = {
 	.driver = {
 		.name = "mt7986-wmac",
-- 
cgit v1.2.3-70-g09d2


From db1a5a6c69e3f2dc05984f5472195bf011254d92 Mon Sep 17 00:00:00 2001
From: Sujuan Chen <sujuan.chen@mediatek.com>
Date: Wed, 22 Feb 2023 21:01:35 +0800
Subject: wifi: mt76: mt7915: add dev->hif2 support for mt7916 WED device

Enable two PCIe interfaces (dev->hif2) support for mt7916 when WED is enabled.

Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7915/dma.c  | 10 ++++++++--
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c  |  4 +++-
 drivers/net/wireless/mediatek/mt76/mt7915/mmio.c | 10 +++++-----
 3 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/dma.c b/drivers/net/wireless/mediatek/mt76/mt7915/dma.c
index abe17dac9996..43a5456d4b97 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/dma.c
@@ -87,8 +87,14 @@ static void mt7915_dma_config(struct mt7915_dev *dev)
 				   MT7916_RXQ_BAND0);
 			RXQ_CONFIG(MT_RXQ_MCU_WA, WFDMA0, MT_INT_WED_RX_DONE_WA_MT7916,
 				   MT7916_RXQ_MCU_WA);
-			RXQ_CONFIG(MT_RXQ_BAND1, WFDMA0, MT_INT_WED_RX_DONE_BAND1_MT7916,
-				   MT7916_RXQ_BAND1);
+			if (dev->hif2)
+				RXQ_CONFIG(MT_RXQ_BAND1, WFDMA0,
+					   MT_INT_RX_DONE_BAND1_MT7916,
+					   MT7916_RXQ_BAND1);
+			else
+				RXQ_CONFIG(MT_RXQ_BAND1, WFDMA0,
+					   MT_INT_WED_RX_DONE_BAND1_MT7916,
+					   MT7916_RXQ_BAND1);
 			RXQ_CONFIG(MT_RXQ_MAIN_WA, WFDMA0, MT_INT_WED_RX_DONE_WA_MAIN_MT7916,
 				   MT7916_RXQ_MCU_WA_MAIN);
 			TXQ_CONFIG(0, WFDMA0, MT_INT_WED_TX_DONE_BAND0,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
index 5545a8bdf1d0..16f09ead307d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
@@ -2370,7 +2370,9 @@ int mt7915_mcu_init_firmware(struct mt7915_dev *dev)
 	if (ret)
 		return ret;
 
-	if (mtk_wed_device_active(&dev->mt76.mmio.wed) && is_mt7915(&dev->mt76))
+	if ((mtk_wed_device_active(&dev->mt76.mmio.wed) &&
+	     is_mt7915(&dev->mt76)) ||
+	    !mtk_wed_get_rx_capa(&dev->mt76.mmio.wed))
 		mt7915_mcu_wa_cmd(dev, MCU_WA_PARAM_CMD(CAPABILITY), 0, 0, 0);
 
 	ret = mt7915_mcu_set_mwds(dev, 1);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
index 225a19604d3e..6f0c0e2ae045 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
@@ -989,13 +989,13 @@ irqreturn_t mt7915_irq_handler(int irq, void *dev_instance)
 	struct mt7915_dev *dev = dev_instance;
 	struct mtk_wed_device *wed = &dev->mt76.mmio.wed;
 
-	if (mtk_wed_device_active(wed)) {
+	if (mtk_wed_device_active(wed))
 		mtk_wed_device_irq_set_mask(wed, 0);
-	} else {
+	else
 		mt76_wr(dev, MT_INT_MASK_CSR, 0);
-		if (dev->hif2)
-			mt76_wr(dev, MT_INT1_MASK_CSR, 0);
-	}
+
+	if (dev->hif2)
+		mt76_wr(dev, MT_INT1_MASK_CSR, 0);
 
 	if (!test_bit(MT76_STATE_INITIALIZED, &dev->mphy.state))
 		return IRQ_NONE;
-- 
cgit v1.2.3-70-g09d2


From 9c97df11dfe68b548a47744ba9fa6beddcfba2bc Mon Sep 17 00:00:00 2001
From: Howard Hsu <howard-yh.hsu@mediatek.com>
Date: Thu, 23 Feb 2023 14:24:57 +0800
Subject: wifi: mt76: mt7915: rework init flow in mt7915_thermal_init()

If kernel do not enable CONFIG_HWMON, it may cause thermal
initialization to be done with temperature value 0 and then can not
transmit. This commit fixes it by setting trigger/restore temperature
before checking CONFIG_HWMON.

Fixes: 7d12b38ab6f6 ("wifi: mt76: mt7915: call mt7915_mcu_set_thermal_throttling() only after init_work")
Signed-off-by: Howard Hsu <howard-yh.hsu@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7915/init.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
index 4f3efc942a4d..71ccefd39cb2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
@@ -203,6 +203,10 @@ static int mt7915_thermal_init(struct mt7915_phy *phy)
 			phy->cdev = cdev;
 	}
 
+	/* initialize critical/maximum high temperature */
+	phy->throttle_temp[MT7915_CRIT_TEMP_IDX] = MT7915_CRIT_TEMP;
+	phy->throttle_temp[MT7915_MAX_TEMP_IDX] = MT7915_MAX_TEMP;
+
 	if (!IS_REACHABLE(CONFIG_HWMON))
 		return 0;
 
@@ -211,10 +215,6 @@ static int mt7915_thermal_init(struct mt7915_phy *phy)
 	if (IS_ERR(hwmon))
 		return PTR_ERR(hwmon);
 
-	/* initialize critical/maximum high temperature */
-	phy->throttle_temp[MT7915_CRIT_TEMP_IDX] = MT7915_CRIT_TEMP;
-	phy->throttle_temp[MT7915_MAX_TEMP_IDX] = MT7915_MAX_TEMP;
-
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From e4d2b8bcac11c23536771ff38e536c7462b213f5 Mon Sep 17 00:00:00 2001
From: Peter Chiu <chui-hao.chiu@mediatek.com>
Date: Wed, 22 Feb 2023 14:27:29 +0800
Subject: wifi: mt76: drop the incorrect scatter and gather frame

The scatter and gather frame may be incorrect because WED and WO may
send frames to host driver interleaved.

Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/dma.c | 6 ++++--
 drivers/net/wireless/mediatek/mt76/dma.h | 1 +
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c
index da281cd1d36f..738d7e1569bf 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/dma.c
@@ -402,8 +402,8 @@ mt76_dma_get_buf(struct mt76_dev *dev, struct mt76_queue *q, int idx,
 		*info = le32_to_cpu(desc->info);
 
 	if (mt76_queue_is_wed_rx(q)) {
-		u32 token = FIELD_GET(MT_DMA_CTL_TOKEN,
-				      le32_to_cpu(desc->buf1));
+		u32 buf1 = le32_to_cpu(desc->buf1);
+		u32 token = FIELD_GET(MT_DMA_CTL_TOKEN, buf1);
 		struct mt76_txwi_cache *t = mt76_rx_token_release(dev, token);
 
 		if (!t)
@@ -424,6 +424,8 @@ mt76_dma_get_buf(struct mt76_dev *dev, struct mt76_queue *q, int idx,
 
 			*drop = !!(ctrl & (MT_DMA_CTL_TO_HOST_A |
 					   MT_DMA_CTL_DROP));
+
+			*drop |= !!(buf1 & MT_DMA_CTL_WO_DROP);
 		}
 	} else {
 		buf = e->buf;
diff --git a/drivers/net/wireless/mediatek/mt76/dma.h b/drivers/net/wireless/mediatek/mt76/dma.h
index 4b9bc7f462b8..1b090d78cd05 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.h
+++ b/drivers/net/wireless/mediatek/mt76/dma.h
@@ -19,6 +19,7 @@
 #define MT_DMA_CTL_TO_HOST_A		BIT(12)
 #define MT_DMA_CTL_DROP			BIT(14)
 #define MT_DMA_CTL_TOKEN		GENMASK(31, 16)
+#define MT_DMA_CTL_WO_DROP		BIT(8)
 
 #define MT_DMA_PPE_CPU_REASON		GENMASK(15, 11)
 #define MT_DMA_PPE_ENTRY		GENMASK(30, 16)
-- 
cgit v1.2.3-70-g09d2


From 9e47dd9f64a47ae00ca0123017584c37209ee900 Mon Sep 17 00:00:00 2001
From: Kang Chen <void0red@gmail.com>
Date: Mon, 27 Feb 2023 22:48:23 +0800
Subject: wifi: mt76: handle failure of vzalloc in mt7615_coredump_work

vzalloc may fails, dump might be null and will cause
illegal address access later.

Link: https://lore.kernel.org/all/Y%2Fy5Asxw3T3m4jCw@lore-desk
Fixes: d2bf7959d9c0 ("mt76: mt7663: introduce coredump support")
Signed-off-by: Kang Chen <void0red@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/mac.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
index a95602473359..6d6b46d7bd1a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
@@ -2380,7 +2380,7 @@ void mt7615_coredump_work(struct work_struct *work)
 			break;
 
 		skb_pull(skb, sizeof(struct mt7615_mcu_rxd));
-		if (data + skb->len - dump > MT76_CONNAC_COREDUMP_SZ) {
+		if (!dump || data + skb->len - dump > MT76_CONNAC_COREDUMP_SZ) {
 			dev_kfree_skb(skb);
 			continue;
 		}
@@ -2390,6 +2390,8 @@ void mt7615_coredump_work(struct work_struct *work)
 
 		dev_kfree_skb(skb);
 	}
-	dev_coredumpv(dev->mt76.dev, dump, MT76_CONNAC_COREDUMP_SZ,
-		      GFP_KERNEL);
+
+	if (dump)
+		dev_coredumpv(dev->mt76.dev, dump, MT76_CONNAC_COREDUMP_SZ,
+			      GFP_KERNEL);
 }
-- 
cgit v1.2.3-70-g09d2


From 5b8ccdfb943f6a03c676d2ea816dd38c149e920b Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Fri, 14 Apr 2023 14:10:54 +0200
Subject: wifi: mt76: add missing locking to protect against concurrent
 rx/status calls

According to the documentation, ieee80211_rx_list must not run concurrently
with ieee80211_tx_status (or its variants).

Cc: stable@vger.kernel.org
Fixes: 88046b2c9f6d ("mt76: add support for reporting tx status with skb")
Reported-by: Brian Coverstone <brian@mainsequence.net>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/dma.c         | 2 ++
 drivers/net/wireless/mediatek/mt76/mt7603/mac.c  | 5 ++++-
 drivers/net/wireless/mediatek/mt76/mt7615/mac.c  | 5 ++++-
 drivers/net/wireless/mediatek/mt76/mt76x02_mac.c | 5 ++++-
 drivers/net/wireless/mediatek/mt76/tx.c          | 4 ++++
 5 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c
index 738d7e1569bf..6c0174313d09 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/dma.c
@@ -578,7 +578,9 @@ free:
 free_skb:
 	status.skb = tx_info.skb;
 	hw = mt76_tx_status_get_hw(dev, tx_info.skb);
+	spin_lock_bh(&dev->rx_lock);
 	ieee80211_tx_status_ext(hw, &status);
+	spin_unlock_bh(&dev->rx_lock);
 
 	return ret;
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/mac.c b/drivers/net/wireless/mediatek/mt76/mt7603/mac.c
index 70a7f84af028..12e0af52082a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7603/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7603/mac.c
@@ -1279,8 +1279,11 @@ void mt7603_mac_add_txs(struct mt7603_dev *dev, void *data)
 	if (wcidx >= MT7603_WTBL_STA || !sta)
 		goto out;
 
-	if (mt7603_fill_txs(dev, msta, &info, txs_data))
+	if (mt7603_fill_txs(dev, msta, &info, txs_data)) {
+		spin_lock_bh(&dev->mt76.rx_lock);
 		ieee80211_tx_status_noskb(mt76_hw(dev), sta, &info);
+		spin_unlock_bh(&dev->mt76.rx_lock);
+	}
 
 out:
 	rcu_read_unlock();
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
index 6d6b46d7bd1a..ce216bad3b73 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
@@ -1558,8 +1558,11 @@ static void mt7615_mac_add_txs(struct mt7615_dev *dev, void *data)
 	if (wcid->phy_idx && dev->mt76.phys[MT_BAND1])
 		mphy = dev->mt76.phys[MT_BAND1];
 
-	if (mt7615_fill_txs(dev, msta, &info, txs_data))
+	if (mt7615_fill_txs(dev, msta, &info, txs_data)) {
+		spin_lock_bh(&dev->mt76.rx_lock);
 		ieee80211_tx_status_noskb(mphy->hw, sta, &info);
+		spin_unlock_bh(&dev->mt76.rx_lock);
+	}
 
 out:
 	rcu_read_unlock();
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
index d3f74473e6fb..3e41d809ade3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
@@ -631,8 +631,11 @@ void mt76x02_send_tx_status(struct mt76x02_dev *dev,
 
 	mt76_tx_status_unlock(mdev, &list);
 
-	if (!status.skb)
+	if (!status.skb) {
+		spin_lock_bh(&dev->mt76.rx_lock);
 		ieee80211_tx_status_ext(mt76_hw(dev), &status);
+		spin_unlock_bh(&dev->mt76.rx_lock);
+	}
 
 	if (!len)
 		goto out;
diff --git a/drivers/net/wireless/mediatek/mt76/tx.c b/drivers/net/wireless/mediatek/mt76/tx.c
index 3ad9742364ba..72b3ec715e47 100644
--- a/drivers/net/wireless/mediatek/mt76/tx.c
+++ b/drivers/net/wireless/mediatek/mt76/tx.c
@@ -77,7 +77,9 @@ mt76_tx_status_unlock(struct mt76_dev *dev, struct sk_buff_head *list)
 		}
 
 		hw = mt76_tx_status_get_hw(dev, skb);
+		spin_lock_bh(&dev->rx_lock);
 		ieee80211_tx_status_ext(hw, &status);
+		spin_unlock_bh(&dev->rx_lock);
 	}
 	rcu_read_unlock();
 }
@@ -263,7 +265,9 @@ void __mt76_tx_complete_skb(struct mt76_dev *dev, u16 wcid_idx, struct sk_buff *
 	if (cb->pktid < MT_PACKET_ID_FIRST) {
 		hw = mt76_tx_status_get_hw(dev, skb);
 		status.sta = wcid_to_sta(wcid);
+		spin_lock_bh(&dev->rx_lock);
 		ieee80211_tx_status_ext(hw, &status);
+		spin_unlock_bh(&dev->rx_lock);
 		goto out;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 1a1ff6c3b338a6e9ef8d187fbae3d8f823a300ce Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sat, 18 Feb 2023 01:50:38 +0800
Subject: wifi: mt76: connac: refresh tx session timer for WED device

There's no keep-alive packet being received by mac80211 stack when WED
is enabled, which leads to tx BA session timeout. This patch calls
ieee80211_refresh_tx_agg_session_timer() to refresh timer according
to tx status reporting.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
index aed4ee95fb2e..8d316d3aab8f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
@@ -582,6 +582,17 @@ bool mt76_connac2_mac_fill_txs(struct mt76_dev *dev, struct mt76_wcid *wcid,
 			le32_get_bits(txs_data[6], MT_TXS6_MPDU_FAIL_CNT);
 		stats->tx_retries +=
 			le32_get_bits(txs_data[7], MT_TXS7_MPDU_RETRY_CNT);
+
+		if (wcid->sta) {
+			struct ieee80211_sta *sta;
+			u8 tid;
+
+			sta = container_of((void *)wcid, struct ieee80211_sta,
+					   drv_priv);
+			tid = FIELD_GET(MT_TXS0_TID, txs);
+
+			ieee80211_refresh_tx_agg_session_timer(sta, tid);
+		}
 	}
 
 	txrate = FIELD_GET(MT_TXS0_TX_RATE, txs);
-- 
cgit v1.2.3-70-g09d2


From 885f7af7e54498e7168fa9e7c2f029f81988f933 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Thu, 9 Feb 2023 10:22:59 +0800
Subject: wifi: mt76: mt7915: remove mt7915_mcu_beacon_check_caps()

Those capabilities have been added into ieee80211_bss_conf.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7915/main.c   |   1 -
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c    | 107 +++------------------
 drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h |  14 ---
 3 files changed, 12 insertions(+), 110 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/main.c b/drivers/net/wireless/mediatek/mt76/mt7915/main.c
index 3bbccbdfc5eb..2ada2806de66 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/main.c
@@ -269,7 +269,6 @@ static int mt7915_add_interface(struct ieee80211_hw *hw,
 	vif->offload_flags |= IEEE80211_OFFLOAD_ENCAP_4ADDR;
 
 	mt7915_init_bitrate_mask(vif);
-	memset(&mvif->cap, -1, sizeof(mvif->cap));
 
 	mt7915_mcu_add_bss_info(phy, vif, true);
 	mt7915_mcu_add_sta(dev, vif, NULL, true);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
index 16f09ead307d..7432b1eb396d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
@@ -706,7 +706,6 @@ static void
 mt7915_mcu_sta_he_tlv(struct sk_buff *skb, struct ieee80211_sta *sta,
 		      struct ieee80211_vif *vif)
 {
-	struct mt7915_vif *mvif = (struct mt7915_vif *)vif->drv_priv;
 	struct ieee80211_he_cap_elem *elem = &sta->deflink.he_cap.he_cap_elem;
 	struct ieee80211_he_mcs_nss_supp mcs_map;
 	struct sta_rec_he *he;
@@ -740,7 +739,7 @@ mt7915_mcu_sta_he_tlv(struct sk_buff *skb, struct ieee80211_sta *sta,
 	     IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_RU_MAPPING_IN_5G))
 		cap |= STA_REC_HE_CAP_BW20_RU242_SUPPORT;
 
-	if (mvif->cap.he_ldpc &&
+	if (vif->bss_conf.he_ldpc &&
 	    (elem->phy_cap_info[1] &
 	     IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD))
 		cap |= STA_REC_HE_CAP_LDPC;
@@ -849,7 +848,6 @@ static void
 mt7915_mcu_sta_muru_tlv(struct mt7915_dev *dev, struct sk_buff *skb,
 			struct ieee80211_sta *sta, struct ieee80211_vif *vif)
 {
-	struct mt7915_vif *mvif = (struct mt7915_vif *)vif->drv_priv;
 	struct ieee80211_he_cap_elem *elem = &sta->deflink.he_cap.he_cap_elem;
 	struct sta_rec_muru *muru;
 	struct tlv *tlv;
@@ -862,9 +860,9 @@ mt7915_mcu_sta_muru_tlv(struct mt7915_dev *dev, struct sk_buff *skb,
 
 	muru = (struct sta_rec_muru *)tlv;
 
-	muru->cfg.mimo_dl_en = mvif->cap.he_mu_ebfer ||
-			       mvif->cap.vht_mu_ebfer ||
-			       mvif->cap.vht_mu_ebfee;
+	muru->cfg.mimo_dl_en = vif->bss_conf.he_mu_beamformer ||
+			       vif->bss_conf.vht_mu_beamformer ||
+			       vif->bss_conf.vht_mu_beamformee;
 	if (!is_mt7915(&dev->mt76))
 		muru->cfg.mimo_ul_en = true;
 	muru->cfg.ofdma_dl_en = true;
@@ -997,8 +995,8 @@ mt7915_mcu_sta_wtbl_tlv(struct mt7915_dev *dev, struct sk_buff *skb,
 	mt76_connac_mcu_wtbl_hdr_trans_tlv(skb, vif, wcid, tlv, wtbl_hdr);
 	if (sta)
 		mt76_connac_mcu_wtbl_ht_tlv(&dev->mt76, skb, sta, tlv,
-					    wtbl_hdr, mvif->cap.ht_ldpc,
-					    mvif->cap.vht_ldpc);
+					    wtbl_hdr, vif->bss_conf.ht_ldpc,
+					    vif->bss_conf.vht_ldpc);
 
 	return 0;
 }
@@ -1007,7 +1005,6 @@ static inline bool
 mt7915_is_ebf_supported(struct mt7915_phy *phy, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta, bool bfee)
 {
-	struct mt7915_vif *mvif = (struct mt7915_vif *)vif->drv_priv;
 	int tx_ant = hweight8(phy->mt76->chainmask) - 1;
 
 	if (vif->type != NL80211_IFTYPE_STATION &&
@@ -1021,10 +1018,10 @@ mt7915_is_ebf_supported(struct mt7915_phy *phy, struct ieee80211_vif *vif,
 		struct ieee80211_he_cap_elem *pe = &sta->deflink.he_cap.he_cap_elem;
 
 		if (bfee)
-			return mvif->cap.he_su_ebfee &&
+			return vif->bss_conf.he_su_beamformee &&
 			       HE_PHY(CAP3_SU_BEAMFORMER, pe->phy_cap_info[3]);
 		else
-			return mvif->cap.he_su_ebfer &&
+			return vif->bss_conf.he_su_beamformer &&
 			       HE_PHY(CAP4_SU_BEAMFORMEE, pe->phy_cap_info[4]);
 	}
 
@@ -1032,10 +1029,10 @@ mt7915_is_ebf_supported(struct mt7915_phy *phy, struct ieee80211_vif *vif,
 		u32 cap = sta->deflink.vht_cap.cap;
 
 		if (bfee)
-			return mvif->cap.vht_su_ebfee &&
+			return vif->bss_conf.vht_su_beamformee &&
 			       (cap & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE);
 		else
-			return mvif->cap.vht_su_ebfer &&
+			return vif->bss_conf.vht_su_beamformer &&
 			       (cap & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE);
 	}
 
@@ -1530,7 +1527,7 @@ mt7915_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb, struct mt7915_dev *dev,
 			cap |= STA_CAP_TX_STBC;
 		if (sta->deflink.ht_cap.cap & IEEE80211_HT_CAP_RX_STBC)
 			cap |= STA_CAP_RX_STBC;
-		if (mvif->cap.ht_ldpc &&
+		if (vif->bss_conf.ht_ldpc &&
 		    (sta->deflink.ht_cap.cap & IEEE80211_HT_CAP_LDPC_CODING))
 			cap |= STA_CAP_LDPC;
 
@@ -1556,7 +1553,7 @@ mt7915_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb, struct mt7915_dev *dev,
 			cap |= STA_CAP_VHT_TX_STBC;
 		if (sta->deflink.vht_cap.cap & IEEE80211_VHT_CAP_RXSTBC_1)
 			cap |= STA_CAP_VHT_RX_STBC;
-		if (mvif->cap.vht_ldpc &&
+		if (vif->bss_conf.vht_ldpc &&
 		    (sta->deflink.vht_cap.cap & IEEE80211_VHT_CAP_RXLDPC))
 			cap |= STA_CAP_VHT_LDPC;
 
@@ -1875,84 +1872,6 @@ mt7915_mcu_beacon_cont(struct mt7915_dev *dev, struct ieee80211_vif *vif,
 	memcpy(buf + MT_TXD_SIZE, skb->data, skb->len);
 }
 
-static void
-mt7915_mcu_beacon_check_caps(struct mt7915_phy *phy, struct ieee80211_vif *vif,
-			     struct sk_buff *skb)
-{
-	struct mt7915_vif *mvif = (struct mt7915_vif *)vif->drv_priv;
-	struct mt7915_vif_cap *vc = &mvif->cap;
-	const struct ieee80211_he_cap_elem *he;
-	const struct ieee80211_vht_cap *vht;
-	const struct ieee80211_ht_cap *ht;
-	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)skb->data;
-	const u8 *ie;
-	u32 len, bc;
-
-	/* Check missing configuration options to allow AP mode in mac80211
-	 * to remain in sync with hostapd settings, and get a subset of
-	 * beacon and hardware capabilities.
-	 */
-	if (WARN_ON_ONCE(skb->len <= (mgmt->u.beacon.variable - skb->data)))
-		return;
-
-	memset(vc, 0, sizeof(*vc));
-
-	len = skb->len - (mgmt->u.beacon.variable - skb->data);
-
-	ie = cfg80211_find_ie(WLAN_EID_HT_CAPABILITY, mgmt->u.beacon.variable,
-			      len);
-	if (ie && ie[1] >= sizeof(*ht)) {
-		ht = (void *)(ie + 2);
-		vc->ht_ldpc = !!(le16_to_cpu(ht->cap_info) &
-				 IEEE80211_HT_CAP_LDPC_CODING);
-	}
-
-	ie = cfg80211_find_ie(WLAN_EID_VHT_CAPABILITY, mgmt->u.beacon.variable,
-			      len);
-	if (ie && ie[1] >= sizeof(*vht)) {
-		u32 pc = phy->mt76->sband_5g.sband.vht_cap.cap;
-
-		vht = (void *)(ie + 2);
-		bc = le32_to_cpu(vht->vht_cap_info);
-
-		vc->vht_ldpc = !!(bc & IEEE80211_VHT_CAP_RXLDPC);
-		vc->vht_su_ebfer =
-			(bc & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE);
-		vc->vht_su_ebfee =
-			(bc & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE);
-		vc->vht_mu_ebfer =
-			(bc & IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE);
-		vc->vht_mu_ebfee =
-			(bc & IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE);
-	}
-
-	ie = cfg80211_find_ext_ie(WLAN_EID_EXT_HE_CAPABILITY,
-				  mgmt->u.beacon.variable, len);
-	if (ie && ie[1] >= sizeof(*he) + 1) {
-		const struct ieee80211_sta_he_cap *pc =
-			mt76_connac_get_he_phy_cap(phy->mt76, vif);
-		const struct ieee80211_he_cap_elem *pe = &pc->he_cap_elem;
-
-		he = (void *)(ie + 3);
-
-		vc->he_ldpc =
-			HE_PHY(CAP1_LDPC_CODING_IN_PAYLOAD, pe->phy_cap_info[1]);
-		vc->he_su_ebfer =
-			HE_PHY(CAP3_SU_BEAMFORMER, he->phy_cap_info[3]) &&
-			HE_PHY(CAP3_SU_BEAMFORMER, pe->phy_cap_info[3]);
-		vc->he_su_ebfee =
-			HE_PHY(CAP4_SU_BEAMFORMEE, he->phy_cap_info[4]) &&
-			HE_PHY(CAP4_SU_BEAMFORMEE, pe->phy_cap_info[4]);
-		vc->he_mu_ebfer =
-			HE_PHY(CAP4_MU_BEAMFORMER, he->phy_cap_info[4]) &&
-			HE_PHY(CAP4_MU_BEAMFORMER, pe->phy_cap_info[4]);
-	}
-}
-
 static void
 mt7915_mcu_beacon_inband_discov(struct mt7915_dev *dev, struct ieee80211_vif *vif,
 				struct sk_buff *rskb, struct bss_info_bcn *bcn,
@@ -2063,8 +1982,6 @@ int mt7915_mcu_add_beacon(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	info = IEEE80211_SKB_CB(skb);
 	info->hw_queue = FIELD_PREP(MT_TX_HW_QUEUE_PHY, ext_phy);
 
-	mt7915_mcu_beacon_check_caps(phy, vif, skb);
-
 	mt7915_mcu_beacon_cntdwn(vif, rskb, skb, bcn, &offs);
 	mt7915_mcu_beacon_mbss(rskb, skb, vif, bcn, &offs);
 	mt7915_mcu_beacon_cont(dev, vif, rskb, skb, bcn, &offs);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
index 3cbfb9b6a305..e4016cef9447 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
@@ -147,23 +147,9 @@ struct mt7915_sta {
 	} twt;
 };
 
-struct mt7915_vif_cap {
-	bool ht_ldpc:1;
-	bool vht_ldpc:1;
-	bool he_ldpc:1;
-	bool vht_su_ebfer:1;
-	bool vht_su_ebfee:1;
-	bool vht_mu_ebfer:1;
-	bool vht_mu_ebfee:1;
-	bool he_su_ebfer:1;
-	bool he_su_ebfee:1;
-	bool he_mu_ebfer:1;
-};
-
 struct mt7915_vif {
 	struct mt76_vif mt76; /* must be first */
 
-	struct mt7915_vif_cap cap;
 	struct mt7915_sta sta;
 	struct mt7915_phy *phy;
 
-- 
cgit v1.2.3-70-g09d2


From dda423dd65c3e977e545306de3f51031103ce610 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Thu, 9 Feb 2023 10:23:00 +0800
Subject: wifi: mt76: mt7996: remove mt7996_mcu_beacon_check_caps()

Those capabilities have been added into ieee80211_bss_conf.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/main.c   |   1 -
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    | 115 ++-------------------
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  16 ---
 3 files changed, 8 insertions(+), 124 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 3e4da0350d96..9d67fcaf2428 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -220,7 +220,6 @@ static int mt7996_add_interface(struct ieee80211_hw *hw,
 	vif->offload_flags |= IEEE80211_OFFLOAD_ENCAP_4ADDR;
 
 	mt7996_init_bitrate_mask(vif);
-	memset(&mvif->cap, -1, sizeof(mvif->cap));
 
 	mt7996_mcu_add_bss_info(phy, vif, true);
 	mt7996_mcu_add_sta(dev, vif, NULL, true);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 8ad51cbfdbe8..c1e6f7d1660c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -1053,7 +1053,6 @@ static inline bool
 mt7996_is_ebf_supported(struct mt7996_phy *phy, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta, bool bfee)
 {
-	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
 	int sts = hweight16(phy->mt76->chainmask);
 
 	if (vif->type != NL80211_IFTYPE_STATION &&
@@ -1068,10 +1067,10 @@ mt7996_is_ebf_supported(struct mt7996_phy *phy, struct ieee80211_vif *vif,
 		struct ieee80211_eht_cap_elem_fixed *pe = &pc->eht_cap_elem;
 
 		if (bfee)
-			return mvif->cap.eht_su_ebfee &&
+			return vif->bss_conf.eht_su_beamformee &&
 			       EHT_PHY(CAP0_SU_BEAMFORMEE, pe->phy_cap_info[0]);
 		else
-			return mvif->cap.eht_su_ebfer &&
+			return vif->bss_conf.eht_su_beamformer &&
 			       EHT_PHY(CAP0_SU_BEAMFORMER, pe->phy_cap_info[0]);
 	}
 
@@ -1079,10 +1078,10 @@ mt7996_is_ebf_supported(struct mt7996_phy *phy, struct ieee80211_vif *vif,
 		struct ieee80211_he_cap_elem *pe = &sta->deflink.he_cap.he_cap_elem;
 
 		if (bfee)
-			return mvif->cap.he_su_ebfee &&
+			return vif->bss_conf.he_su_beamformee &&
 			       HE_PHY(CAP3_SU_BEAMFORMER, pe->phy_cap_info[3]);
 		else
-			return mvif->cap.he_su_ebfer &&
+			return vif->bss_conf.he_su_beamformer &&
 			       HE_PHY(CAP4_SU_BEAMFORMEE, pe->phy_cap_info[4]);
 	}
 
@@ -1090,10 +1089,10 @@ mt7996_is_ebf_supported(struct mt7996_phy *phy, struct ieee80211_vif *vif,
 		u32 cap = sta->deflink.vht_cap.cap;
 
 		if (bfee)
-			return mvif->cap.vht_su_ebfee &&
+			return vif->bss_conf.vht_su_beamformee &&
 			       (cap & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE);
 		else
-			return mvif->cap.vht_su_ebfer &&
+			return vif->bss_conf.vht_su_beamformer &&
 			       (cap & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE);
 	}
 
@@ -1572,7 +1571,7 @@ mt7996_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb, struct mt7996_dev *dev,
 			cap |= STA_CAP_TX_STBC;
 		if (sta->deflink.ht_cap.cap & IEEE80211_HT_CAP_RX_STBC)
 			cap |= STA_CAP_RX_STBC;
-		if (mvif->cap.ht_ldpc &&
+		if (vif->bss_conf.ht_ldpc &&
 		    (sta->deflink.ht_cap.cap & IEEE80211_HT_CAP_LDPC_CODING))
 			cap |= STA_CAP_LDPC;
 
@@ -1598,7 +1597,7 @@ mt7996_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb, struct mt7996_dev *dev,
 			cap |= STA_CAP_VHT_TX_STBC;
 		if (sta->deflink.vht_cap.cap & IEEE80211_VHT_CAP_RXSTBC_1)
 			cap |= STA_CAP_VHT_RX_STBC;
-		if (mvif->cap.vht_ldpc &&
+		if (vif->bss_conf.vht_ldpc &&
 		    (sta->deflink.vht_cap.cap & IEEE80211_VHT_CAP_RXLDPC))
 			cap |= STA_CAP_VHT_LDPC;
 
@@ -1911,102 +1910,6 @@ mt7996_mcu_beacon_cont(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 	memcpy(buf + MT_TXD_SIZE, skb->data, skb->len);
 }
 
-static void
-mt7996_mcu_beacon_check_caps(struct mt7996_phy *phy, struct ieee80211_vif *vif,
-			     struct sk_buff *skb)
-{
-	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
-	struct mt7996_vif_cap *vc = &mvif->cap;
-	const struct ieee80211_eht_cap_elem_fixed *eht;
-	const struct ieee80211_he_cap_elem *he;
-	const struct ieee80211_vht_cap *vht;
-	const struct ieee80211_ht_cap *ht;
-	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)skb->data;
-	const u8 *ie;
-	u32 len, bc;
-
-	/* Check missing configuration options to allow AP mode in mac80211
-	 * to remain in sync with hostapd settings, and get a subset of
-	 * beacon and hardware capabilities.
-	 */
-	if (WARN_ON_ONCE(skb->len <= (mgmt->u.beacon.variable - skb->data)))
-		return;
-
-	memset(vc, 0, sizeof(*vc));
-
-	len = skb->len - (mgmt->u.beacon.variable - skb->data);
-
-	ie = cfg80211_find_ie(WLAN_EID_HT_CAPABILITY, mgmt->u.beacon.variable,
-			      len);
-	if (ie && ie[1] >= sizeof(*ht)) {
-		ht = (void *)(ie + 2);
-		vc->ht_ldpc |= !!(le16_to_cpu(ht->cap_info) &
-				  IEEE80211_HT_CAP_LDPC_CODING);
-	}
-
-	ie = cfg80211_find_ie(WLAN_EID_VHT_CAPABILITY, mgmt->u.beacon.variable,
-			      len);
-	if (ie && ie[1] >= sizeof(*vht)) {
-		u32 pc = phy->mt76->sband_5g.sband.vht_cap.cap;
-
-		vht = (void *)(ie + 2);
-		bc = le32_to_cpu(vht->vht_cap_info);
-
-		vc->vht_ldpc |= !!(bc & IEEE80211_VHT_CAP_RXLDPC);
-		vc->vht_su_ebfer =
-			(bc & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE);
-		vc->vht_su_ebfee =
-			(bc & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE);
-		vc->vht_mu_ebfer =
-			(bc & IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE);
-		vc->vht_mu_ebfee =
-			(bc & IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE) &&
-			(pc & IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE);
-	}
-
-	ie = cfg80211_find_ext_ie(WLAN_EID_EXT_HE_CAPABILITY,
-				  mgmt->u.beacon.variable, len);
-	if (ie && ie[1] >= sizeof(*he) + 1) {
-		const struct ieee80211_sta_he_cap *pc =
-			mt76_connac_get_he_phy_cap(phy->mt76, vif);
-		const struct ieee80211_he_cap_elem *pe = &pc->he_cap_elem;
-
-		he = (void *)(ie + 3);
-
-		vc->he_ldpc =
-			HE_PHY(CAP1_LDPC_CODING_IN_PAYLOAD, pe->phy_cap_info[1]);
-		vc->he_su_ebfer =
-			HE_PHY(CAP3_SU_BEAMFORMER, he->phy_cap_info[3]) &&
-			HE_PHY(CAP3_SU_BEAMFORMER, pe->phy_cap_info[3]);
-		vc->he_su_ebfee =
-			HE_PHY(CAP4_SU_BEAMFORMEE, he->phy_cap_info[4]) &&
-			HE_PHY(CAP4_SU_BEAMFORMEE, pe->phy_cap_info[4]);
-		vc->he_mu_ebfer =
-			HE_PHY(CAP4_MU_BEAMFORMER, he->phy_cap_info[4]) &&
-			HE_PHY(CAP4_MU_BEAMFORMER, pe->phy_cap_info[4]);
-	}
-
-	ie = cfg80211_find_ext_ie(WLAN_EID_EXT_EHT_CAPABILITY,
-				  mgmt->u.beacon.variable, len);
-	if (ie && ie[1] >= sizeof(*eht) + 1) {
-		const struct ieee80211_sta_eht_cap *pc =
-			mt76_connac_get_eht_phy_cap(phy->mt76, vif);
-		const struct ieee80211_eht_cap_elem_fixed *pe = &pc->eht_cap_elem;
-
-		eht = (void *)(ie + 3);
-
-		vc->eht_su_ebfer =
-			EHT_PHY(CAP0_SU_BEAMFORMER, eht->phy_cap_info[0]) &&
-			EHT_PHY(CAP0_SU_BEAMFORMER, pe->phy_cap_info[0]);
-		vc->eht_su_ebfee =
-			EHT_PHY(CAP0_SU_BEAMFORMEE, eht->phy_cap_info[0]) &&
-			EHT_PHY(CAP0_SU_BEAMFORMEE, pe->phy_cap_info[0]);
-	}
-}
-
 int mt7996_mcu_add_beacon(struct ieee80211_hw *hw,
 			  struct ieee80211_vif *vif, int en)
 {
@@ -2045,8 +1948,6 @@ int mt7996_mcu_add_beacon(struct ieee80211_hw *hw,
 	info = IEEE80211_SKB_CB(skb);
 	info->hw_queue |= FIELD_PREP(MT_TX_HW_QUEUE_PHY, phy->mt76->band_idx);
 
-	mt7996_mcu_beacon_check_caps(phy, vif, skb);
-
 	mt7996_mcu_beacon_cont(dev, vif, rskb, skb, bcn, &offs);
 	/* TODO: subtag - 11v MBSSID */
 	mt7996_mcu_beacon_cntdwn(vif, rskb, skb, &offs);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 018dfd2b36b0..24572dd55fe0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -112,25 +112,9 @@ struct mt7996_sta {
 	} twt;
 };
 
-struct mt7996_vif_cap {
-	bool ht_ldpc:1;
-	bool vht_ldpc:1;
-	bool he_ldpc:1;
-	bool vht_su_ebfer:1;
-	bool vht_su_ebfee:1;
-	bool vht_mu_ebfer:1;
-	bool vht_mu_ebfee:1;
-	bool he_su_ebfer:1;
-	bool he_su_ebfee:1;
-	bool he_mu_ebfer:1;
-	bool eht_su_ebfer:1;
-	bool eht_su_ebfee:1;
-};
-
 struct mt7996_vif {
 	struct mt76_vif mt76; /* must be first */
 
-	struct mt7996_vif_cap cap;
 	struct mt7996_sta sta;
 	struct mt7996_phy *phy;
 
-- 
cgit v1.2.3-70-g09d2


From 021af945997ffaeaa37c9673d71afad7cde6bdef Mon Sep 17 00:00:00 2001
From: Shayne Chen <shayne.chen@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:52 +0800
Subject: wifi: mt76: mt7996: add eht rx rate support

Add support to report eht rx rate.
Note that extended fields for eht in mt76_rx_status will make the struct
size exceed the cb size, so make nss and band share the same u8.

Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mac80211.c   | 11 ++++++++---
 drivers/net/wireless/mediatek/mt76/mt76.h       | 18 ++++++++++++++----
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c |  9 +++++----
 3 files changed, 27 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c
index 87902f4b8e12..e53166fcdc9d 100644
--- a/drivers/net/wireless/mediatek/mt76/mac80211.c
+++ b/drivers/net/wireless/mediatek/mt76/mac80211.c
@@ -1067,9 +1067,14 @@ mt76_rx_convert(struct mt76_dev *dev, struct sk_buff *skb,
 	status->enc_flags = mstat.enc_flags;
 	status->encoding = mstat.encoding;
 	status->bw = mstat.bw;
-	status->he_ru = mstat.he_ru;
-	status->he_gi = mstat.he_gi;
-	status->he_dcm = mstat.he_dcm;
+	if (status->encoding == RX_ENC_EHT) {
+		status->eht.ru = mstat.eht.ru;
+		status->eht.gi = mstat.eht.gi;
+	} else {
+		status->he_ru = mstat.he_ru;
+		status->he_gi = mstat.he_gi;
+		status->he_dcm = mstat.he_dcm;
+	}
 	status->rate_idx = mstat.rate_idx;
 	status->nss = mstat.nss;
 	status->band = mstat.band;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index 183b0fc5a2d4..c3d1313e2f20 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -621,12 +621,22 @@ struct mt76_rx_status {
 	u16 freq;
 	u32 flag;
 	u8 enc_flags;
-	u8 encoding:2, bw:3, he_ru:3;
-	u8 he_gi:2, he_dcm:1;
+	u8 encoding:3, bw:4;
+	union {
+		struct {
+			u8 he_ru:3;
+			u8 he_gi:2;
+			u8 he_dcm:1;
+		};
+		struct {
+			u8 ru:4;
+			u8 gi:2;
+		} eht;
+	};
+
 	u8 amsdu:1, first_amsdu:1, last_amsdu:1;
 	u8 rate_idx;
-	u8 nss;
-	u8 band;
+	u8 nss:5, band:3;
 	s8 signal;
 	u8 chains;
 	s8 chain_signal[IEEE80211_MAX_CHAINS];
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index 3c3506c7c87a..d811b4e01800 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -572,11 +572,12 @@ mt7996_mac_fill_rx_rate(struct mt7996_dev *dev,
 	case MT_PHY_TYPE_EHT_SU:
 	case MT_PHY_TYPE_EHT_TRIG:
 	case MT_PHY_TYPE_EHT_MU:
-		/* TODO: currently report rx rate with HE rate */
 		status->nss = nss;
-		status->encoding = RX_ENC_HE;
-		bw = min_t(int, bw, IEEE80211_STA_RX_BW_160);
-		i = min_t(int, i & 0xf, 11);
+		status->encoding = RX_ENC_EHT;
+		i &= GENMASK(3, 0);
+
+		if (gi <= NL80211_RATE_INFO_EHT_GI_3_2)
+			status->eht.gi = gi;
 		break;
 	default:
 		return -EINVAL;
-- 
cgit v1.2.3-70-g09d2


From d0b6f86fdbefa62fd4ad2acd1aea6c45f9b518ba Mon Sep 17 00:00:00 2001
From: Shayne Chen <shayne.chen@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:53 +0800
Subject: wifi: mt76: mt7996: let non-bufferable MMPDUs use correct hw queue

Pass qid into mt7996_mac_write_txwi() to let the tx descriptor of
non-bufferable MMPDUs be filled with correct hw queue index.

Fixes: 98686cd21624 ("wifi: mt76: mt7996: add driver for MediaTek Wi-Fi 7 (802.11be) devices")
Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    | 11 ++++++-----
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    |  6 +++---
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  5 +++--
 3 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index d811b4e01800..f1414a448023 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -982,8 +982,9 @@ mt7996_mac_write_txwi_80211(struct mt7996_dev *dev, __le32 *txwi,
 }
 
 void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
-			   struct sk_buff *skb, struct mt76_wcid *wcid, int pid,
-			   struct ieee80211_key_conf *key, u32 changed)
+			   struct sk_buff *skb, struct mt76_wcid *wcid,
+			   struct ieee80211_key_conf *key, int pid,
+			   enum mt76_txq_id qid, u32 changed)
 {
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_vif *vif = info->control.vif;
@@ -1014,7 +1015,7 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 	} else if (beacon) {
 		p_fmt = MT_TX_TYPE_FW;
 		q_idx = MT_LMAC_BCN0;
-	} else if (skb_get_queue_mapping(skb) >= MT_TXQ_PSD) {
+	} else if (qid >= MT_TXQ_PSD) {
 		p_fmt = MT_TX_TYPE_CT;
 		q_idx = MT_LMAC_ALTX0;
 	} else {
@@ -1123,8 +1124,8 @@ int mt7996_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 	memset(txwi_ptr, 0, MT_TXD_SIZE);
 	/* Transmit non qos data by 802.11 header and need to fill txd by host*/
 	if (!is_8023 || pid >= MT_PACKET_ID_FIRST)
-		mt7996_mac_write_txwi(dev, txwi_ptr, tx_info->skb, wcid, pid,
-				      key, 0);
+		mt7996_mac_write_txwi(dev, txwi_ptr, tx_info->skb, wcid, key,
+				      pid, qid, 0);
 
 	txp = (struct mt76_connac_txp_common *)(txwi + MT_TXD_SIZE);
 	for (i = 0; i < nbuf; i++) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index c1e6f7d1660c..bb95fdda0eba 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -1905,8 +1905,9 @@ mt7996_mcu_beacon_cont(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 	}
 
 	buf = (u8 *)bcn + sizeof(*bcn) - MAX_BEACON_SIZE;
-	mt7996_mac_write_txwi(dev, (__le32 *)buf, skb, wcid, 0, NULL,
+	mt7996_mac_write_txwi(dev, (__le32 *)buf, skb, wcid, NULL, 0, 0,
 			      BSS_CHANGED_BEACON);
+
 	memcpy(buf + MT_TXD_SIZE, skb->data, skb->len);
 }
 
@@ -2016,8 +2017,7 @@ int mt7996_mcu_beacon_inband_discov(struct mt7996_dev *dev,
 
 	buf = (u8 *)tlv + sizeof(*discov) - MAX_INBAND_FRAME_SIZE;
 
-	mt7996_mac_write_txwi(dev, (__le32 *)buf, skb, wcid, 0, NULL,
-			      changed);
+	mt7996_mac_write_txwi(dev, (__le32 *)buf, skb, wcid, NULL, 0, 0, changed);
 
 	memcpy(buf + MT_TXD_SIZE, skb->data, skb->len);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 24572dd55fe0..569d9f6ee1d0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -471,8 +471,9 @@ void mt7996_mac_enable_nf(struct mt7996_dev *dev, u8 band);
 void mt7996_mac_enable_rtscts(struct mt7996_dev *dev,
 			      struct ieee80211_vif *vif, bool enable);
 void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
-			   struct sk_buff *skb, struct mt76_wcid *wcid, int pid,
-			   struct ieee80211_key_conf *key, u32 changed);
+			   struct sk_buff *skb, struct mt76_wcid *wcid,
+			   struct ieee80211_key_conf *key, int pid,
+			   enum mt76_txq_id qid, u32 changed);
 void mt7996_mac_set_timing(struct mt7996_phy *phy);
 int mt7996_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta);
-- 
cgit v1.2.3-70-g09d2


From c04ab57dab559c519bb882995beeb6d95a0a21d3 Mon Sep 17 00:00:00 2001
From: Shayne Chen <shayne.chen@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:54 +0800
Subject: wifi: mt76: mt7996: remove unused eeprom band selection

Tri-band chipsets support one band on each a-die, so remove the unused
definition of eeprom band selection.

Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c | 4 ----
 drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h | 1 -
 2 files changed, 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c
index 2e48c5a40f81..544b6c6f1ea3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c
@@ -138,10 +138,6 @@ static int mt7996_eeprom_parse_band_config(struct mt7996_phy *phy)
 	case MT_EE_BAND_SEL_6GHZ:
 		phy->mt76->cap.has_6ghz = true;
 		break;
-	case MT_EE_BAND_SEL_5GHZ_6GHZ:
-		phy->mt76->cap.has_5ghz = true;
-		phy->mt76->cap.has_6ghz = true;
-		break;
 	default:
 		ret = -EINVAL;
 		break;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h
index 8da599e0abea..5571881f639b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h
@@ -46,7 +46,6 @@ enum mt7996_eeprom_band {
 	MT_EE_BAND_SEL_2GHZ,
 	MT_EE_BAND_SEL_5GHZ,
 	MT_EE_BAND_SEL_6GHZ,
-	MT_EE_BAND_SEL_5GHZ_6GHZ,
 };
 
 static inline int
-- 
cgit v1.2.3-70-g09d2


From 8b14ce24a0297175bc4ebdf26d45a22b5a33847f Mon Sep 17 00:00:00 2001
From: Peter Chiu <chui-hao.chiu@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:55 +0800
Subject: wifi: mt76: mt7996: fix pointer calculation in ie countdown event

Fix the tail and data pointers. The rxd->len in mt7996_mcu_rxd does not
include the length of general rxd. It only includes the length of
firmware event rxd. Use skb->length to get the correct length.

Fixes: 98686cd21624 ("wifi: mt76: mt7996: add driver for MediaTek Wi-Fi 7 (802.11be) devices")
Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com>
Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index bb95fdda0eba..29ff4c191a3c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -422,7 +422,8 @@ mt7996_mcu_ie_countdown(struct mt7996_dev *dev, struct sk_buff *skb)
 	if (hdr->band && dev->mt76.phys[hdr->band])
 		mphy = dev->mt76.phys[hdr->band];
 
-	tail = skb->data + le16_to_cpu(rxd->len);
+	tail = skb->data + skb->len;
+	data += sizeof(struct header);
 	while (data + sizeof(struct tlv) < tail && le16_to_cpu(tlv->len)) {
 		switch (le16_to_cpu(tlv->tag)) {
 		case UNI_EVENT_IE_COUNTDOWN_CSA:
-- 
cgit v1.2.3-70-g09d2


From 4e029000fc9f503f01fb91f19d341e65bcc6c3c2 Mon Sep 17 00:00:00 2001
From: Peter Chiu <chui-hao.chiu@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:56 +0800
Subject: wifi: mt76: mt7996: init mpdu density cap

Init mpdu density based on the hardware capability to prevent rx drop.

Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com>
Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/init.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index 946da93eed32..de94e151c47b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -196,10 +196,13 @@ mt7996_init_wiphy(struct ieee80211_hw *hw)
 
 	hw->max_tx_fragments = 4;
 
-	if (phy->mt76->cap.has_2ghz)
+	if (phy->mt76->cap.has_2ghz) {
 		phy->mt76->sband_2g.sband.ht_cap.cap |=
 			IEEE80211_HT_CAP_LDPC_CODING |
 			IEEE80211_HT_CAP_MAX_AMSDU;
+		phy->mt76->sband_2g.sband.ht_cap.ampdu_density =
+			IEEE80211_HT_MPDU_DENSITY_2;
+	}
 
 	if (phy->mt76->cap.has_5ghz) {
 		phy->mt76->sband_5g.sband.ht_cap.cap |=
@@ -211,6 +214,8 @@ mt7996_init_wiphy(struct ieee80211_hw *hw)
 			IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK |
 			IEEE80211_VHT_CAP_SHORT_GI_160 |
 			IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ;
+		phy->mt76->sband_5g.sband.ht_cap.ampdu_density =
+			IEEE80211_HT_MPDU_DENSITY_1;
 	}
 
 	mt76_set_stream_caps(phy->mt76, true);
@@ -689,7 +694,7 @@ mt7996_init_he_caps(struct mt7996_phy *phy, enum nl80211_band band,
 		u16 cap = IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS |
 			  IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS;
 
-		cap |= u16_encode_bits(IEEE80211_HT_MPDU_DENSITY_2,
+		cap |= u16_encode_bits(IEEE80211_HT_MPDU_DENSITY_0_5,
 				       IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START) |
 		       u16_encode_bits(IEEE80211_VHT_MAX_AMPDU_1024K,
 				       IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP) |
-- 
cgit v1.2.3-70-g09d2


From 6784b1785e214b0df43a6c9c37495b39c7473223 Mon Sep 17 00:00:00 2001
From: Peter Chiu <chui-hao.chiu@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:57 +0800
Subject: wifi: mt76: mt7996: remove mt7996_mcu_set_pm()

Currently using BSS_INFO_PS command will sometimes cause packet drop in
hw rx queue.
Temporarily remove this function until finding the cause.

Signed-off-by: Peter Chiu <chui-hao.chiu@mediatek.com>
Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/main.c   |  8 -------
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    | 26 ----------------------
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  1 -
 3 files changed, 35 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 9d67fcaf2428..33e76a97f46b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -52,10 +52,6 @@ static int mt7996_start(struct ieee80211_hw *hw)
 
 	set_bit(MT76_STATE_RUNNING, &phy->mt76->state);
 
-	ieee80211_iterate_interfaces(dev->mt76.hw,
-				     IEEE80211_IFACE_ITER_RESUME_ALL,
-				     mt7996_mcu_set_pm, dev->mt76.hw);
-
 	ieee80211_queue_delayed_work(hw, &phy->mt76->mac_work,
 				     MT7996_WATCHDOG_TIME);
 
@@ -79,10 +75,6 @@ static void mt7996_stop(struct ieee80211_hw *hw)
 
 	clear_bit(MT76_STATE_RUNNING, &phy->mt76->state);
 
-	ieee80211_iterate_interfaces(dev->mt76.hw,
-				     IEEE80211_IFACE_ITER_RESUME_ALL,
-				     mt7996_mcu_set_pm, dev->mt76.hw);
-
 	mutex_unlock(&dev->mt76.mutex);
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 29ff4c191a3c..2a66ac546513 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -3478,32 +3478,6 @@ int mt7996_mcu_twt_agrt_update(struct mt7996_dev *dev,
 				 &req, sizeof(req), true);
 }
 
-void mt7996_mcu_set_pm(void *priv, u8 *mac, struct ieee80211_vif *vif)
-{
-#define EXIT_PM_STATE	0
-#define ENTER_PM_STATE	1
-	struct ieee80211_hw *hw = priv;
-	struct mt7996_dev *dev = mt7996_hw_dev(hw);
-	struct mt7996_phy *phy = mt7996_hw_phy(hw);
-	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
-	struct bss_power_save *ps;
-	struct sk_buff *skb;
-	struct tlv *tlv;
-	bool running = test_bit(MT76_STATE_RUNNING, &phy->mt76->state);
-
-	skb = __mt7996_mcu_alloc_bss_req(&dev->mt76, &mvif->mt76,
-					 MT7996_BSS_UPDATE_MAX_SIZE);
-	if (IS_ERR(skb))
-		return;
-
-	tlv = mt7996_mcu_add_uni_tlv(skb, UNI_BSS_INFO_PS, sizeof(*ps));
-	ps = (struct bss_power_save *)tlv;
-	ps->profile = running ? EXIT_PM_STATE : ENTER_PM_STATE;
-
-	mt76_mcu_skb_send_msg(&dev->mt76, skb,
-			      MCU_WMWA_UNI_CMD(BSS_INFO_UPDATE), true);
-}
-
 int mt7996_mcu_set_rts_thresh(struct mt7996_phy *phy, u32 val)
 {
 	struct {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 569d9f6ee1d0..5463ce77a1a0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -416,7 +416,6 @@ int mt7996_mcu_set_pulse_th(struct mt7996_dev *dev,
 int mt7996_mcu_set_radar_th(struct mt7996_dev *dev, int index,
 			    const struct mt7996_dfs_pattern *pattern);
 int mt7996_mcu_set_radio_en(struct mt7996_phy *phy, bool enable);
-void mt7996_mcu_set_pm(void *priv, u8 *mac, struct ieee80211_vif *vif);
 int mt7996_mcu_set_rts_thresh(struct mt7996_phy *phy, u32 val);
 int mt7996_mcu_get_chan_mib_info(struct mt7996_phy *phy, bool chan_switch);
 int mt7996_mcu_rdd_cmd(struct mt7996_dev *dev, int cmd, u8 index,
-- 
cgit v1.2.3-70-g09d2


From 72fc0df3006ce5c109f9c68f0724e44c47b4ec7b Mon Sep 17 00:00:00 2001
From: StanleyYP Wang <StanleyYP.Wang@mediatek.com>
Date: Fri, 3 Mar 2023 16:35:58 +0800
Subject: wifi: mt76: mt7996: fix eeprom tx path bitfields

Swap the tx path bitfields of band1 and band2 to read correct setting.

Fixes: 98686cd21624 ("wifi: mt76: mt7996: add driver for MediaTek Wi-Fi 7 (802.11be) devices")
Signed-off-by: StanleyYP Wang <StanleyYP.Wang@mediatek.com>
Signed-off-by: Shayne Chen <shayne.chen@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h
index 5571881f639b..0c749774f6b1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h
@@ -31,11 +31,11 @@ enum mt7996_eeprom_field {
 #define MT_EE_WIFI_CONF2_BAND_SEL		GENMASK(2, 0)
 
 #define MT_EE_WIFI_CONF1_TX_PATH_BAND0		GENMASK(5, 3)
-#define MT_EE_WIFI_CONF2_TX_PATH_BAND1		GENMASK(5, 3)
-#define MT_EE_WIFI_CONF2_TX_PATH_BAND2		GENMASK(2, 0)
+#define MT_EE_WIFI_CONF2_TX_PATH_BAND1		GENMASK(2, 0)
+#define MT_EE_WIFI_CONF2_TX_PATH_BAND2		GENMASK(5, 3)
 #define MT_EE_WIFI_CONF4_STREAM_NUM_BAND0	GENMASK(5, 3)
-#define MT_EE_WIFI_CONF5_STREAM_NUM_BAND1	GENMASK(5, 3)
-#define MT_EE_WIFI_CONF5_STREAM_NUM_BAND2	GENMASK(2, 0)
+#define MT_EE_WIFI_CONF5_STREAM_NUM_BAND1	GENMASK(2, 0)
+#define MT_EE_WIFI_CONF5_STREAM_NUM_BAND2	GENMASK(5, 3)
 
 #define MT_EE_RATE_DELTA_MASK			GENMASK(5, 0)
 #define MT_EE_RATE_DELTA_SIGN			BIT(6)
-- 
cgit v1.2.3-70-g09d2


From a1ec7e6237c58c887e34ce4c02a9ec88ebe1c766 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 6 Mar 2023 18:50:29 +0100
Subject: wifi: mt76: mt7921: introduce mt7921_get_mac80211_ops utility routine

Since the fw offload capability check is shared between pci,usb and sdio
devices, move it in common init code and reduce code duplication.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/init.c   | 29 ++++++++++++++++++++--
 drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c    | 19 ++------------
 drivers/net/wireless/mediatek/mt76/mt7921/sdio.c   | 20 ++-------------
 drivers/net/wireless/mediatek/mt76/mt7921/usb.c    | 19 ++------------
 5 files changed, 35 insertions(+), 55 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
index 80c71acfe159..33681fbb2fe8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
@@ -169,7 +169,8 @@ mt7921_mac_init_band(struct mt7921_dev *dev, u8 band)
 	mt76_rmw(dev, MT_WTBLOFF_TOP_RSCR(band), mask, set);
 }
 
-u8 mt7921_check_offload_capability(struct device *dev, const char *fw_wm)
+static u8
+mt7921_get_offload_capability(struct device *dev, const char *fw_wm)
 {
 	struct mt7921_fw_features *features = NULL;
 	const struct mt76_connac2_fw_trailer *hdr;
@@ -220,7 +221,31 @@ out:
 
 	return features ? features->data : 0;
 }
-EXPORT_SYMBOL_GPL(mt7921_check_offload_capability);
+
+struct ieee80211_ops *
+mt7921_get_mac80211_ops(struct device *dev, void *drv_data, u8 *fw_features)
+{
+	struct ieee80211_ops *ops;
+
+	ops = devm_kmemdup(dev, &mt7921_ops, sizeof(mt7921_ops), GFP_KERNEL);
+	if (!ops)
+		return NULL;
+
+	*fw_features = mt7921_get_offload_capability(dev, drv_data);
+	if (!(*fw_features & MT7921_FW_CAP_CNM)) {
+		ops->remain_on_channel = NULL;
+		ops->cancel_remain_on_channel = NULL;
+		ops->add_chanctx = NULL;
+		ops->remove_chanctx = NULL;
+		ops->change_chanctx = NULL;
+		ops->assign_vif_chanctx = NULL;
+		ops->unassign_vif_chanctx = NULL;
+		ops->mgd_prepare_tx = NULL;
+		ops->mgd_complete_tx = NULL;
+	}
+	return ops;
+}
+EXPORT_SYMBOL_GPL(mt7921_get_mac80211_ops);
 
 int mt7921_mac_init(struct mt7921_dev *dev)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
index 1af70dac723b..b1dceeafa54e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
@@ -593,5 +593,6 @@ int mt7921_mcu_set_roc(struct mt7921_phy *phy, struct mt7921_vif *vif,
 		       enum mt7921_roc_req type, u8 token_id);
 int mt7921_mcu_abort_roc(struct mt7921_phy *phy, struct mt7921_vif *vif,
 			 u8 token_id);
-u8 mt7921_check_offload_capability(struct device *dev, const char *fw_wm);
+struct ieee80211_ops *mt7921_get_mac80211_ops(struct device *dev,
+					      void *drv_data, u8 *fw_features);
 #endif
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 1a8a54a46dcc..b520cd258170 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -256,7 +256,6 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 		.drv_own = mt7921e_mcu_drv_pmctrl,
 		.fw_own = mt7921e_mcu_fw_pmctrl,
 	};
-
 	struct ieee80211_ops *ops;
 	struct mt76_bus_ops *bus_ops;
 	struct mt7921_dev *dev;
@@ -285,27 +284,13 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 	if (mt7921_disable_aspm)
 		mt76_pci_disable_aspm(pdev);
 
-	features = mt7921_check_offload_capability(&pdev->dev, (const char *)
-						   id->driver_data);
-	ops = devm_kmemdup(&pdev->dev, &mt7921_ops, sizeof(mt7921_ops),
-			   GFP_KERNEL);
+	ops = mt7921_get_mac80211_ops(&pdev->dev, (void *)id->driver_data,
+				      &features);
 	if (!ops) {
 		ret = -ENOMEM;
 		goto err_free_pci_vec;
 	}
 
-	if (!(features & MT7921_FW_CAP_CNM)) {
-		ops->remain_on_channel = NULL;
-		ops->cancel_remain_on_channel = NULL;
-		ops->add_chanctx = NULL;
-		ops->remove_chanctx = NULL;
-		ops->change_chanctx = NULL;
-		ops->assign_vif_chanctx = NULL;
-		ops->unassign_vif_chanctx = NULL;
-		ops->mgd_prepare_tx = NULL;
-		ops->mgd_complete_tx = NULL;
-	}
-
 	mdev = mt76_alloc_device(&pdev->dev, sizeof(*dev), ops, &drv_ops);
 	if (!mdev) {
 		ret = -ENOMEM;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
index 8ce4252b8ae7..584921fb25b5 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
@@ -122,33 +122,17 @@ static int mt7921s_probe(struct sdio_func *func,
 		.drv_own = mt7921s_mcu_drv_pmctrl,
 		.fw_own = mt7921s_mcu_fw_pmctrl,
 	};
-
 	struct ieee80211_ops *ops;
 	struct mt7921_dev *dev;
 	struct mt76_dev *mdev;
 	u8 features;
 	int ret;
 
-	features = mt7921_check_offload_capability(&func->dev, (const char *)
-						   id->driver_data);
-
-	ops = devm_kmemdup(&func->dev, &mt7921_ops, sizeof(mt7921_ops),
-			   GFP_KERNEL);
+	ops = mt7921_get_mac80211_ops(&func->dev, (void *)id->driver_data,
+				      &features);
 	if (!ops)
 		return -ENOMEM;
 
-	if (!(features & MT7921_FW_CAP_CNM)) {
-		ops->remain_on_channel = NULL;
-		ops->cancel_remain_on_channel = NULL;
-		ops->add_chanctx = NULL;
-		ops->remove_chanctx = NULL;
-		ops->change_chanctx = NULL;
-		ops->assign_vif_chanctx = NULL;
-		ops->unassign_vif_chanctx = NULL;
-		ops->mgd_prepare_tx = NULL;
-		ops->mgd_complete_tx = NULL;
-	}
-
 	mdev = mt76_alloc_device(&func->dev, sizeof(*dev), ops, &drv_ops);
 	if (!mdev)
 		return -ENOMEM;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
index 8fef09ed29c9..376bf89a70d8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
@@ -210,27 +210,12 @@ static int mt7921u_probe(struct usb_interface *usb_intf,
 	u8 features;
 	int ret;
 
-	features = mt7921_check_offload_capability(&usb_intf->dev, (const char *)
-						   id->driver_info);
-	ops = devm_kmemdup(&usb_intf->dev, &mt7921_ops, sizeof(mt7921_ops),
-			   GFP_KERNEL);
+	ops = mt7921_get_mac80211_ops(&usb_intf->dev, (void *)id->driver_info,
+				      &features);
 	if (!ops)
 		return -ENOMEM;
 
-	if (!(features & MT7921_FW_CAP_CNM)) {
-		ops->remain_on_channel = NULL;
-		ops->cancel_remain_on_channel = NULL;
-		ops->add_chanctx = NULL;
-		ops->remove_chanctx = NULL;
-		ops->change_chanctx = NULL;
-		ops->assign_vif_chanctx = NULL;
-		ops->unassign_vif_chanctx = NULL;
-		ops->mgd_prepare_tx = NULL;
-		ops->mgd_complete_tx = NULL;
-	}
-
 	ops->stop = mt7921u_stop;
-
 	mdev = mt76_alloc_device(&usb_intf->dev, sizeof(*dev), ops, &drv_ops);
 	if (!mdev)
 		return -ENOMEM;
-- 
cgit v1.2.3-70-g09d2


From a5af1481a85679a9812f728ee5cf64ee7edbe669 Mon Sep 17 00:00:00 2001
From: Neil Chen <yn.chen@mediatek.com>
Date: Tue, 7 Mar 2023 19:22:49 +0800
Subject: wifi: mt76: mt7921: use driver flags rather than mac80211 flags to
 mcu

FIF_* flags from mac80211 is not ABI. mt7921 should not pass it into mcu
directly. Remap FIF_* to driver defined flags as mcu command input.

Fixes: c222f77fd421 ("wifi: mt76: mt7921: fix rx filter incorrect by drv/fw inconsistent")
Signed-off-by: Neil Chen <yn.chen@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/main.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index a72964e7a807..f1d537a7705a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -703,10 +703,25 @@ static void mt7921_configure_filter(struct ieee80211_hw *hw,
 				    unsigned int *total_flags,
 				    u64 multicast)
 {
+#define MT7921_FILTER_FCSFAIL    BIT(2)
+#define MT7921_FILTER_CONTROL    BIT(5)
+#define MT7921_FILTER_OTHER_BSS  BIT(6)
+#define MT7921_FILTER_ENABLE     BIT(31)
+
 	struct mt7921_dev *dev = mt7921_hw_dev(hw);
+	u32 flags = MT7921_FILTER_ENABLE;
+
+#define MT7921_FILTER(_fif, _type) do {			\
+		if (*total_flags & (_fif))		\
+			flags |= MT7921_FILTER_##_type;	\
+	} while (0)
+
+	MT7921_FILTER(FIF_FCSFAIL, FCSFAIL);
+	MT7921_FILTER(FIF_CONTROL, CONTROL);
+	MT7921_FILTER(FIF_OTHER_BSS, OTHER_BSS);
 
 	mt7921_mutex_acquire(dev);
-	mt7921_mcu_set_rxfilter(dev, *total_flags, 0, 0);
+	mt7921_mcu_set_rxfilter(dev, flags, 0, 0);
 	mt7921_mutex_release(dev);
 
 	*total_flags &= (FIF_OTHER_BSS | FIF_FCSFAIL | FIF_CONTROL);
-- 
cgit v1.2.3-70-g09d2


From 09d4d6da1b65d09414e7bce61459593f3c80ead1 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 29 Mar 2023 14:57:58 -0500
Subject: wifi: mt76: mt7921e: Set memory space enable in PCI_COMMAND if unset

When the BIOS has been configured for Fast Boot, systems with mt7921e
have non-functional wifi.  Turning on Fast boot caused both bus master
enable and memory space enable bits in PCI_COMMAND not to get configured.

The mt7921 driver already sets bus master enable, but explicitly check
and set memory access enable as well to fix this problem.

Tested-by: Anson Tsao <anson.tsao@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Acked-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index b520cd258170..974412b6f76a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -262,6 +262,7 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 	struct mt76_dev *mdev;
 	u8 features;
 	int ret;
+	u16 cmd;
 
 	ret = pcim_enable_device(pdev);
 	if (ret)
@@ -271,6 +272,11 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 	if (ret)
 		return ret;
 
+	pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+	if (!(cmd & PCI_COMMAND_MEMORY)) {
+		cmd |= PCI_COMMAND_MEMORY;
+		pci_write_config_word(pdev, PCI_COMMAND, cmd);
+	}
 	pci_set_master(pdev);
 
 	ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES);
-- 
cgit v1.2.3-70-g09d2


From ec193b41cc56077e481af4abca2e00e298fcf798 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 3 Apr 2023 12:33:55 +0200
Subject: wifi: mt76: move irq_tasklet in mt76_dev struct

irq_tasklet struct is used by most of the drivers (e.g. mt7915, mt7921,
mt7615, mt7663 and mt7996) so move it in common code.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76.h            | 1 +
 drivers/net/wireless/mediatek/mt76/mt7615/mmio.c     | 6 +++---
 drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h   | 4 +---
 drivers/net/wireless/mediatek/mt76/mt7615/pci.c      | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7915/init.c     | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7915/mac.c      | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7915/mmio.c     | 6 +++---
 drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h   | 3 +--
 drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h   | 3 +--
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c      | 8 ++++----
 drivers/net/wireless/mediatek/mt76/mt7996/init.c     | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c      | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mmio.c     | 6 +++---
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h   | 3 +--
 15 files changed, 24 insertions(+), 28 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index c3d1313e2f20..6b07b8fafec2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -788,6 +788,7 @@ struct mt76_dev {
 	spinlock_t rx_lock;
 	struct napi_struct napi[__MT_RXQ_MAX];
 	struct sk_buff_head rx_skb[__MT_RXQ_MAX];
+	struct tasklet_struct irq_tasklet;
 
 	struct list_head txwi_cache;
 	struct list_head rxwi_cache;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
index 83173efb56dc..e703400aa551 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
@@ -80,14 +80,14 @@ static irqreturn_t mt7615_irq_handler(int irq, void *dev_instance)
 	if (!test_bit(MT76_STATE_INITIALIZED, &dev->mphy.state))
 		return IRQ_NONE;
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 
 	return IRQ_HANDLED;
 }
 
 static void mt7615_irq_tasklet(struct tasklet_struct *t)
 {
-	struct mt7615_dev *dev = from_tasklet(dev, t, irq_tasklet);
+	struct mt7615_dev *dev = from_tasklet(dev, t, mt76.irq_tasklet);
 	u32 intr, mask = 0, tx_mcu_mask = mt7615_tx_mcu_int_mask(dev);
 	u32 mcu_int;
 
@@ -202,7 +202,7 @@ int mt7615_mmio_probe(struct device *pdev, void __iomem *mem_base,
 
 	dev = container_of(mdev, struct mt7615_dev, mt76);
 	mt76_mmio_init(&dev->mt76, mem_base);
-	tasklet_setup(&dev->irq_tasklet, mt7615_irq_tasklet);
+	tasklet_setup(&mdev->irq_tasklet, mt7615_irq_tasklet);
 
 	dev->reg_map = map;
 	dev->ops = ops;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
index 43591b4c1d9a..fbf73f8fda1b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
@@ -245,8 +245,6 @@ struct mt7615_dev {
 	};
 
 	const struct mt76_bus_ops *bus_ops;
-	struct tasklet_struct irq_tasklet;
-
 	struct mt7615_phy phy;
 	u64 omac_mask;
 
@@ -416,7 +414,7 @@ static inline void mt7615_irq_enable(struct mt7615_dev *dev, u32 mask)
 {
 	mt76_set_irq_mask(&dev->mt76, 0, 0, mask);
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 }
 
 static inline bool mt7615_firmware_offload(struct mt7615_dev *dev)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/pci.c b/drivers/net/wireless/mediatek/mt76/mt7615/pci.c
index b808248943ea..9f43e673518b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/pci.c
@@ -94,7 +94,7 @@ static int mt7615_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 	mt76_for_each_q_rx(mdev, i) {
 		napi_disable(&mdev->napi[i]);
 	}
-	tasklet_kill(&dev->irq_tasklet);
+	tasklet_kill(&mdev->irq_tasklet);
 
 	mt7615_dma_reset(dev);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c b/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c
index 0680e002b981..f607eee3fb47 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c
@@ -122,7 +122,7 @@ void mt7615_unregister_device(struct mt7615_dev *dev)
 
 	mt7615_tx_token_put(dev);
 	mt7615_dma_cleanup(dev);
-	tasklet_disable(&dev->irq_tasklet);
+	tasklet_disable(&dev->mt76.irq_tasklet);
 
 	mt76_free_device(&dev->mt76);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
index 71ccefd39cb2..0d52c30ff8fd 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
@@ -1180,7 +1180,7 @@ static void mt7915_stop_hardware(struct mt7915_dev *dev)
 	mt7915_mcu_exit(dev);
 	mt7915_tx_token_put(dev);
 	mt7915_dma_cleanup(dev);
-	tasklet_disable(&dev->irq_tasklet);
+	tasklet_disable(&dev->mt76.irq_tasklet);
 
 	if (is_mt7986(&dev->mt76))
 		mt7986_wmac_disable(dev);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c
index 97ca55d283fb..3c156c1d589b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c
@@ -1627,7 +1627,7 @@ void mt7915_mac_reset_work(struct work_struct *work)
 	}
 	local_bh_enable();
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 
 	mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_RESET_DONE);
 	mt7915_wait_reset_state(dev, MT_MCU_CMD_NORMAL_STATE);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
index 6f0c0e2ae045..49f0410384c0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
@@ -916,7 +916,7 @@ static void mt7915_rx_poll_complete(struct mt76_dev *mdev,
 /* TODO: support 2/4/6/8 MSI-X vectors */
 static void mt7915_irq_tasklet(struct tasklet_struct *t)
 {
-	struct mt7915_dev *dev = from_tasklet(dev, t, irq_tasklet);
+	struct mt7915_dev *dev = from_tasklet(dev, t, mt76.irq_tasklet);
 	struct mtk_wed_device *wed = &dev->mt76.mmio.wed;
 	u32 intr, intr1, mask;
 
@@ -1000,7 +1000,7 @@ irqreturn_t mt7915_irq_handler(int irq, void *dev_instance)
 	if (!test_bit(MT76_STATE_INITIALIZED, &dev->mphy.state))
 		return IRQ_NONE;
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 
 	return IRQ_HANDLED;
 }
@@ -1041,7 +1041,7 @@ struct mt7915_dev *mt7915_mmio_probe(struct device *pdev,
 	if (ret)
 		goto error;
 
-	tasklet_setup(&dev->irq_tasklet, mt7915_irq_tasklet);
+	tasklet_setup(&mdev->irq_tasklet, mt7915_irq_tasklet);
 
 	return dev;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
index e4016cef9447..530cdcef13c9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
@@ -294,7 +294,6 @@ struct mt7915_dev {
 	u32 wfdma_mask;
 
 	const struct mt76_bus_ops *bus_ops;
-	struct tasklet_struct irq_tasklet;
 	struct mt7915_phy phy;
 
 	/* monitor rx chain configured channel */
@@ -567,7 +566,7 @@ static inline void mt7915_irq_enable(struct mt7915_dev *dev, u32 mask)
 	else
 		mt76_set_irq_mask(&dev->mt76, 0, 0, mask);
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 }
 
 static inline void mt7915_irq_disable(struct mt7915_dev *dev, u32 mask)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
index b1dceeafa54e..aeb58ad169e6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
@@ -287,7 +287,6 @@ struct mt7921_dev {
 
 	const struct mt76_bus_ops *bus_ops;
 	struct mt7921_phy phy;
-	struct tasklet_struct irq_tasklet;
 
 	struct work_struct reset_work;
 	bool hw_full_reset:1;
@@ -395,7 +394,7 @@ static inline void mt7921_irq_enable(struct mt7921_dev *dev, u32 mask)
 {
 	mt76_set_irq_mask(&dev->mt76, 0, 0, mask);
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 }
 
 static inline u32
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 974412b6f76a..020ee0e49f78 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -50,7 +50,7 @@ static irqreturn_t mt7921_irq_handler(int irq, void *dev_instance)
 	if (!test_bit(MT76_STATE_INITIALIZED, &dev->mphy.state))
 		return IRQ_NONE;
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 
 	return IRQ_HANDLED;
 }
@@ -122,7 +122,7 @@ static void mt7921e_unregister_device(struct mt7921_dev *dev)
 	mt7921_wfsys_reset(dev);
 	skb_queue_purge(&dev->mt76.mcu.res_q);
 
-	tasklet_disable(&dev->irq_tasklet);
+	tasklet_disable(&dev->mt76.irq_tasklet);
 }
 
 static u32 __mt7921_reg_addr(struct mt7921_dev *dev, u32 addr)
@@ -309,7 +309,7 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 	dev->fw_features = features;
 	dev->hif_ops = &mt7921_pcie_ops;
 	mt76_mmio_init(&dev->mt76, pcim_iomap_table(pdev)[0]);
-	tasklet_init(&dev->irq_tasklet, mt7921_irq_tasklet, (unsigned long)dev);
+	tasklet_init(&mdev->irq_tasklet, mt7921_irq_tasklet, (unsigned long)dev);
 
 	dev->phy.dev = dev;
 	dev->phy.mt76 = &dev->mt76.phy;
@@ -421,7 +421,7 @@ static int mt7921_pci_suspend(struct device *device)
 	mt76_wr(dev, MT_WFDMA0_HOST_INT_ENA, 0);
 	mt76_wr(dev, MT_PCIE_MAC_INT_ENABLE, 0x0);
 	synchronize_irq(pdev->irq);
-	tasklet_kill(&dev->irq_tasklet);
+	tasklet_kill(&mdev->irq_tasklet);
 
 	err = mt7921_mcu_fw_pmctrl(dev);
 	if (err)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index de94e151c47b..539e5cc4aa2b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -902,7 +902,7 @@ void mt7996_unregister_device(struct mt7996_dev *dev)
 	mt7996_mcu_exit(dev);
 	mt7996_tx_token_put(dev);
 	mt7996_dma_cleanup(dev);
-	tasklet_disable(&dev->irq_tasklet);
+	tasklet_disable(&dev->mt76.irq_tasklet);
 
 	mt76_free_device(&dev->mt76);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index f1414a448023..18a997a79526 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -1883,7 +1883,7 @@ void mt7996_mac_reset_work(struct work_struct *work)
 	}
 	local_bh_enable();
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 
 	mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_RESET_DONE);
 	mt7996_wait_reset_state(dev, MT_MCU_CMD_NORMAL_STATE);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
index 902370a2a639..65fa1260e80b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
@@ -251,7 +251,7 @@ static void mt7996_rx_poll_complete(struct mt76_dev *mdev,
 /* TODO: support 2/4/6/8 MSI-X vectors */
 static void mt7996_irq_tasklet(struct tasklet_struct *t)
 {
-	struct mt7996_dev *dev = from_tasklet(dev, t, irq_tasklet);
+	struct mt7996_dev *dev = from_tasklet(dev, t, mt76.irq_tasklet);
 	u32 i, intr, mask, intr1;
 
 	mt76_wr(dev, MT_INT_MASK_CSR, 0);
@@ -308,7 +308,7 @@ irqreturn_t mt7996_irq_handler(int irq, void *dev_instance)
 	if (!test_bit(MT76_STATE_INITIALIZED, &dev->mphy.state))
 		return IRQ_NONE;
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 
 	return IRQ_HANDLED;
 }
@@ -349,7 +349,7 @@ struct mt7996_dev *mt7996_mmio_probe(struct device *pdev,
 	if (ret)
 		goto error;
 
-	tasklet_setup(&dev->irq_tasklet, mt7996_irq_tasklet);
+	tasklet_setup(&mdev->irq_tasklet, mt7996_irq_tasklet);
 
 	mt76_wr(dev, MT_INT_MASK_CSR, 0);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 5463ce77a1a0..4cdc3cc6102d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -222,7 +222,6 @@ struct mt7996_dev {
 	u32 q_wfdma_mask;
 
 	const struct mt76_bus_ops *bus_ops;
-	struct tasklet_struct irq_tasklet;
 	struct mt7996_phy phy;
 
 	/* monitor rx chain configured channel */
@@ -451,7 +450,7 @@ static inline void mt7996_irq_enable(struct mt7996_dev *dev, u32 mask)
 	else
 		mt76_set_irq_mask(&dev->mt76, 0, 0, mask);
 
-	tasklet_schedule(&dev->irq_tasklet);
+	tasklet_schedule(&dev->mt76.irq_tasklet);
 }
 
 static inline void mt7996_irq_disable(struct mt7996_dev *dev, u32 mask)
-- 
cgit v1.2.3-70-g09d2


From 4fc441569c9b18eb87d8cd20f197d9d31a84d453 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 3 Apr 2023 12:33:56 +0200
Subject: wifi: mt76: add mt76_connac_irq_enable utility routine

Most of connac based drivers (mt7921, mt7615, mt7663) share the same
code to enable interrupts. Move it in mt76_connac module.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/dma.c    |  5 +++--
 drivers/net/wireless/mediatek/mt76/mt7615/mmio.c   |  4 +---
 drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h |  7 -------
 drivers/net/wireless/mediatek/mt76/mt76_connac.h   |  6 ++++++
 drivers/net/wireless/mediatek/mt76/mt7921/dma.c    |  8 ++++----
 drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h |  7 -------
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c    | 13 ++++++-------
 7 files changed, 20 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/dma.c b/drivers/net/wireless/mediatek/mt76/mt7615/dma.c
index f1914431ff7f..0ce01ccc5dce 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/dma.c
@@ -76,7 +76,8 @@ static int mt7615_poll_tx(struct napi_struct *napi, int budget)
 
 	mt76_queue_tx_cleanup(dev, dev->mt76.q_mcu[MT_MCUQ_WM], false);
 	if (napi_complete(napi))
-		mt7615_irq_enable(dev, mt7615_tx_mcu_int_mask(dev));
+		mt76_connac_irq_enable(&dev->mt76,
+				       mt7615_tx_mcu_int_mask(dev));
 
 	mt76_connac_pm_unref(&dev->mphy, &dev->pm);
 
@@ -297,7 +298,7 @@ int mt7615_dma_init(struct mt7615_dev *dev)
 	else
 	    mask |= MT_INT_MCU_CMD;
 
-	mt7615_irq_enable(dev, mask);
+	mt76_connac_irq_enable(&dev->mt76, mask);
 
 	mt7615_dma_start(dev);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
index e703400aa551..cb10878dac7e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
@@ -66,9 +66,7 @@ const u32 mt7663e_reg_map[] = {
 static void
 mt7615_rx_poll_complete(struct mt76_dev *mdev, enum mt76_rxq_id q)
 {
-	struct mt7615_dev *dev = container_of(mdev, struct mt7615_dev, mt76);
-
-	mt7615_irq_enable(dev, MT_INT_RX_DONE(q));
+	mt76_connac_irq_enable(mdev, MT_INT_RX_DONE(q));
 }
 
 static irqreturn_t mt7615_irq_handler(int irq, void *dev_instance)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
index fbf73f8fda1b..ff910a80a6b4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
@@ -410,13 +410,6 @@ void mt7615_mcu_rx_event(struct mt7615_dev *dev, struct sk_buff *skb);
 int mt7615_mcu_rdd_send_pattern(struct mt7615_dev *dev);
 int mt7615_mcu_fw_log_2_host(struct mt7615_dev *dev, u8 ctrl);
 
-static inline void mt7615_irq_enable(struct mt7615_dev *dev, u32 mask)
-{
-	mt76_set_irq_mask(&dev->mt76, 0, 0, mask);
-
-	tasklet_schedule(&dev->mt76.irq_tasklet);
-}
-
 static inline bool mt7615_firmware_offload(struct mt7615_dev *dev)
 {
 	return dev->fw_ver > MT7615_FIRMWARE_V2;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
index b339c50bff20..84974dfe1953 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
@@ -279,6 +279,12 @@ static inline u8 mt76_connac_spe_idx(u8 antenna_mask)
 	return ant_to_spe[antenna_mask];
 }
 
+static inline void mt76_connac_irq_enable(struct mt76_dev *dev, u32 mask)
+{
+	mt76_set_irq_mask(dev, 0, 0, mask);
+	tasklet_schedule(&dev->irq_tasklet);
+}
+
 int mt76_connac_pm_wake(struct mt76_phy *phy, struct mt76_connac_pm *pm);
 void mt76_connac_power_save_sched(struct mt76_phy *phy,
 				  struct mt76_connac_pm *pm);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
index d1f10f6d9adc..7b33881a4b54 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
@@ -19,7 +19,7 @@ static int mt7921_poll_tx(struct napi_struct *napi, int budget)
 
 	mt76_connac_tx_cleanup(&dev->mt76);
 	if (napi_complete(napi))
-		mt7921_irq_enable(dev, MT_INT_TX_DONE_ALL);
+		mt76_connac_irq_enable(&dev->mt76, MT_INT_TX_DONE_ALL);
 	mt76_connac_pm_unref(&dev->mphy, &dev->pm);
 
 	return 0;
@@ -123,9 +123,9 @@ static int mt7921_dma_enable(struct mt7921_dev *dev)
 	mt76_set(dev, MT_WFDMA_DUMMY_CR, MT_WFDMA_NEED_REINIT);
 
 	/* enable interrupts for TX/RX rings */
-	mt7921_irq_enable(dev,
-			  MT_INT_RX_DONE_ALL | MT_INT_TX_DONE_ALL |
-			  MT_INT_MCU_CMD);
+	mt76_connac_irq_enable(&dev->mt76,
+			       MT_INT_RX_DONE_ALL | MT_INT_TX_DONE_ALL |
+			       MT_INT_MCU_CMD);
 	mt76_set(dev, MT_MCU2HOST_SW_INT_ENA, MT_MCU_CMD_WAKE_RX_PCIE);
 
 	return 0;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
index aeb58ad169e6..098230039e9a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
@@ -390,13 +390,6 @@ void mt7921_mcu_rx_event(struct mt7921_dev *dev, struct sk_buff *skb);
 int mt7921_mcu_set_rxfilter(struct mt7921_dev *dev, u32 fif,
 			    u8 bit_op, u32 bit_map);
 
-static inline void mt7921_irq_enable(struct mt7921_dev *dev, u32 mask)
-{
-	mt76_set_irq_mask(&dev->mt76, 0, 0, mask);
-
-	tasklet_schedule(&dev->mt76.irq_tasklet);
-}
-
 static inline u32
 mt7921_reg_map_l1(struct mt7921_dev *dev, u32 addr)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 020ee0e49f78..17ba2b4ffec7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -31,14 +31,12 @@ MODULE_PARM_DESC(disable_aspm, "disable PCI ASPM support");
 static void
 mt7921_rx_poll_complete(struct mt76_dev *mdev, enum mt76_rxq_id q)
 {
-	struct mt7921_dev *dev = container_of(mdev, struct mt7921_dev, mt76);
-
 	if (q == MT_RXQ_MAIN)
-		mt7921_irq_enable(dev, MT_INT_RX_DONE_DATA);
+		mt76_connac_irq_enable(mdev, MT_INT_RX_DONE_DATA);
 	else if (q == MT_RXQ_MCU_WA)
-		mt7921_irq_enable(dev, MT_INT_RX_DONE_WM2);
+		mt76_connac_irq_enable(mdev, MT_INT_RX_DONE_WM2);
 	else
-		mt7921_irq_enable(dev, MT_INT_RX_DONE_WM);
+		mt76_connac_irq_enable(mdev, MT_INT_RX_DONE_WM);
 }
 
 static irqreturn_t mt7921_irq_handler(int irq, void *dev_instance)
@@ -465,8 +463,9 @@ static int mt7921_pci_resume(struct device *device)
 
 	/* enable interrupt */
 	mt76_wr(dev, MT_PCIE_MAC_INT_ENABLE, 0xff);
-	mt7921_irq_enable(dev, MT_INT_RX_DONE_ALL | MT_INT_TX_DONE_ALL |
-			  MT_INT_MCU_CMD);
+	mt76_connac_irq_enable(&dev->mt76,
+			       MT_INT_RX_DONE_ALL | MT_INT_TX_DONE_ALL |
+			       MT_INT_MCU_CMD);
 	mt76_set(dev, MT_MCU2HOST_SW_INT_ENA, MT_MCU_CMD_WAKE_RX_PCIE);
 
 	/* put dma enabled */
-- 
cgit v1.2.3-70-g09d2


From f28c3139a9073a0f3161c4f144440b455124041c Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 3 Apr 2023 12:33:57 +0200
Subject: wifi: mt76: get rid of unused sta_ps callbacks

sta_ps callback is just an empty stub for most of the drivers,
so get rid of them.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mac80211.c      | 3 ++-
 drivers/net/wireless/mediatek/mt76/mt7615/mac.c    | 5 -----
 drivers/net/wireless/mediatek/mt76/mt7615/mmio.c   | 1 -
 drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h | 1 -
 drivers/net/wireless/mediatek/mt76/mt7615/sdio.c   | 1 -
 drivers/net/wireless/mediatek/mt76/mt7615/usb.c    | 1 -
 drivers/net/wireless/mediatek/mt76/mt7915/mac.c    | 4 ----
 drivers/net/wireless/mediatek/mt76/mt7915/mmio.c   | 1 -
 drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h | 1 -
 drivers/net/wireless/mediatek/mt76/mt7921/mac.c    | 5 -----
 drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h | 1 -
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c    | 1 -
 drivers/net/wireless/mediatek/mt76/mt7921/sdio.c   | 1 -
 drivers/net/wireless/mediatek/mt76/mt7921/usb.c    | 1 -
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    | 4 ----
 drivers/net/wireless/mediatek/mt76/mt7996/mmio.c   | 1 -
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h | 1 -
 17 files changed, 2 insertions(+), 31 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c
index e53166fcdc9d..467afef98ba2 100644
--- a/drivers/net/wireless/mediatek/mt76/mac80211.c
+++ b/drivers/net/wireless/mediatek/mt76/mac80211.c
@@ -1309,7 +1309,8 @@ mt76_check_sta(struct mt76_dev *dev, struct sk_buff *skb)
 	if (ps)
 		set_bit(MT_WCID_FLAG_PS, &wcid->flags);
 
-	dev->drv->sta_ps(dev, sta, ps);
+	if (dev->drv->sta_ps)
+		dev->drv->sta_ps(dev, sta, ps);
 
 	if (!ps)
 		clear_bit(MT_WCID_FLAG_PS, &wcid->flags);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
index ce216bad3b73..d90378a30d15 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
@@ -655,11 +655,6 @@ static int mt7615_mac_fill_rx(struct mt7615_dev *dev, struct sk_buff *skb)
 	return 0;
 }
 
-void mt7615_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps)
-{
-}
-EXPORT_SYMBOL_GPL(mt7615_sta_ps);
-
 static u16
 mt7615_mac_tx_rate_val(struct mt7615_dev *dev,
 		       struct mt76_phy *mphy,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
index cb10878dac7e..ac036a072439 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c
@@ -179,7 +179,6 @@ int mt7615_mmio_probe(struct device *pdev, void __iomem *mem_base,
 		.rx_check = mt7615_rx_check,
 		.rx_skb = mt7615_queue_rx_skb,
 		.rx_poll_complete = mt7615_rx_poll_complete,
-		.sta_ps = mt7615_sta_ps,
 		.sta_add = mt7615_mac_sta_add,
 		.sta_remove = mt7615_mac_sta_remove,
 		.update_survey = mt7615_update_channel,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
index ff910a80a6b4..932cdad7e7d4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
@@ -511,7 +511,6 @@ void mt7615_tx_token_put(struct mt7615_dev *dev);
 bool mt7615_rx_check(struct mt76_dev *mdev, void *data, int len);
 void mt7615_queue_rx_skb(struct mt76_dev *mdev, enum mt76_rxq_id q,
 			 struct sk_buff *skb, u32 *info);
-void mt7615_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps);
 int mt7615_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta);
 void mt7615_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c b/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c
index 304212f5f8da..fc547a0031ea 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c
@@ -84,7 +84,6 @@ static int mt7663s_probe(struct sdio_func *func,
 		.tx_status_data = mt7663_usb_sdio_tx_status_data,
 		.rx_skb = mt7615_queue_rx_skb,
 		.rx_check = mt7615_rx_check,
-		.sta_ps = mt7615_sta_ps,
 		.sta_add = mt7615_mac_sta_add,
 		.sta_remove = mt7615_mac_sta_remove,
 		.update_survey = mt7615_update_channel,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/usb.c b/drivers/net/wireless/mediatek/mt76/mt7615/usb.c
index f2d651d7adff..04963b9f7498 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/usb.c
@@ -120,7 +120,6 @@ static int mt7663u_probe(struct usb_interface *usb_intf,
 		.tx_status_data = mt7663_usb_sdio_tx_status_data,
 		.rx_skb = mt7615_queue_rx_skb,
 		.rx_check = mt7615_rx_check,
-		.sta_ps = mt7615_sta_ps,
 		.sta_add = mt7615_mac_sta_add,
 		.sta_remove = mt7615_mac_sta_remove,
 		.update_survey = mt7615_update_channel,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c
index 3c156c1d589b..7df8d95fc3fb 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c
@@ -73,10 +73,6 @@ static struct mt76_wcid *mt7915_rx_get_wcid(struct mt7915_dev *dev,
 	return &sta->vif->sta.wcid;
 }
 
-void mt7915_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps)
-{
-}
-
 bool mt7915_mac_wtbl_update(struct mt7915_dev *dev, int idx, u32 mask)
 {
 	mt76_rmw(dev, MT_WTBL_UPDATE, MT_WTBL_UPDATE_WLAN_IDX,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
index 49f0410384c0..45f3558bf31c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c
@@ -1022,7 +1022,6 @@ struct mt7915_dev *mt7915_mmio_probe(struct device *pdev,
 		.rx_skb = mt7915_queue_rx_skb,
 		.rx_check = mt7915_rx_check,
 		.rx_poll_complete = mt7915_rx_poll_complete,
-		.sta_ps = mt7915_sta_ps,
 		.sta_add = mt7915_mac_sta_add,
 		.sta_remove = mt7915_mac_sta_remove,
 		.update_survey = mt7915_update_channel,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
index 530cdcef13c9..b3ead3530740 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h
@@ -616,7 +616,6 @@ void mt7915_tx_token_put(struct mt7915_dev *dev);
 void mt7915_queue_rx_skb(struct mt76_dev *mdev, enum mt76_rxq_id q,
 			 struct sk_buff *skb, u32 *info);
 bool mt7915_rx_check(struct mt76_dev *mdev, void *data, int len);
-void mt7915_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps);
 void mt7915_stats_work(struct work_struct *work);
 int mt76_dfs_start_rdd(struct mt7915_dev *dev, bool force);
 int mt7915_dfs_init_radar_detector(struct mt7915_phy *phy);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
index 557c20190c2b..2eeea43f6bf6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
@@ -32,11 +32,6 @@ static struct mt76_wcid *mt7921_rx_get_wcid(struct mt7921_dev *dev,
 	return &sta->vif->sta.wcid;
 }
 
-void mt7921_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps)
-{
-}
-EXPORT_SYMBOL_GPL(mt7921_sta_ps);
-
 bool mt7921_mac_wtbl_update(struct mt7921_dev *dev, int idx, u32 mask)
 {
 	mt76_rmw(dev, MT_WTBL_UPDATE, MT_WTBL_UPDATE_WLAN_IDX,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
index 098230039e9a..f3155d45a774 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
@@ -470,7 +470,6 @@ void mt7921_tx_token_put(struct mt7921_dev *dev);
 bool mt7921_rx_check(struct mt76_dev *mdev, void *data, int len);
 void mt7921_queue_rx_skb(struct mt76_dev *mdev, enum mt76_rxq_id q,
 			 struct sk_buff *skb, u32 *info);
-void mt7921_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps);
 void mt7921_stats_work(struct work_struct *work);
 void mt7921_set_stream_he_caps(struct mt7921_phy *phy);
 void mt7921_update_channel(struct mt76_phy *mphy);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 17ba2b4ffec7..35ea76c796a3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -241,7 +241,6 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 		.rx_check = mt7921_rx_check,
 		.rx_skb = mt7921_queue_rx_skb,
 		.rx_poll_complete = mt7921_rx_poll_complete,
-		.sta_ps = mt7921_sta_ps,
 		.sta_add = mt7921_mac_sta_add,
 		.sta_assoc = mt7921_mac_sta_assoc,
 		.sta_remove = mt7921_mac_sta_remove,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
index 584921fb25b5..fe396f195dd7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
@@ -99,7 +99,6 @@ static int mt7921s_probe(struct sdio_func *func,
 		.tx_status_data = mt7921_usb_sdio_tx_status_data,
 		.rx_skb = mt7921_queue_rx_skb,
 		.rx_check = mt7921_rx_check,
-		.sta_ps = mt7921_sta_ps,
 		.sta_add = mt7921_mac_sta_add,
 		.sta_assoc = mt7921_mac_sta_assoc,
 		.sta_remove = mt7921_mac_sta_remove,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
index 376bf89a70d8..014476f8ad53 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
@@ -183,7 +183,6 @@ static int mt7921u_probe(struct usb_interface *usb_intf,
 		.tx_status_data = mt7921_usb_sdio_tx_status_data,
 		.rx_skb = mt7921_queue_rx_skb,
 		.rx_check = mt7921_rx_check,
-		.sta_ps = mt7921_sta_ps,
 		.sta_add = mt7921_mac_sta_add,
 		.sta_assoc = mt7921_mac_sta_assoc,
 		.sta_remove = mt7921_mac_sta_remove,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index 18a997a79526..6ce75243e40d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -78,10 +78,6 @@ static struct mt76_wcid *mt7996_rx_get_wcid(struct mt7996_dev *dev,
 	return &sta->vif->sta.wcid;
 }
 
-void mt7996_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps)
-{
-}
-
 bool mt7996_mac_wtbl_update(struct mt7996_dev *dev, int idx, u32 mask)
 {
 	mt76_rmw(dev, MT_WTBL_UPDATE, MT_WTBL_UPDATE_WLAN_IDX,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
index 65fa1260e80b..fc77641ce7bf 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
@@ -330,7 +330,6 @@ struct mt7996_dev *mt7996_mmio_probe(struct device *pdev,
 		.rx_skb = mt7996_queue_rx_skb,
 		.rx_check = mt7996_rx_check,
 		.rx_poll_complete = mt7996_rx_poll_complete,
-		.sta_ps = mt7996_sta_ps,
 		.sta_add = mt7996_mac_sta_add,
 		.sta_remove = mt7996_mac_sta_remove,
 		.update_survey = mt7996_update_channel,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 4cdc3cc6102d..dc77fa4e351e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -495,7 +495,6 @@ void mt7996_tx_token_put(struct mt7996_dev *dev);
 void mt7996_queue_rx_skb(struct mt76_dev *mdev, enum mt76_rxq_id q,
 			 struct sk_buff *skb, u32 *info);
 bool mt7996_rx_check(struct mt76_dev *mdev, void *data, int len);
-void mt7996_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps);
 void mt7996_stats_work(struct work_struct *work);
 int mt76_dfs_start_rdd(struct mt7996_dev *dev, bool force);
 int mt7996_dfs_init_radar_detector(struct mt7996_phy *phy);
-- 
cgit v1.2.3-70-g09d2


From 6a8b899df1562a46a8c55cebc7d35508a24300d3 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 3 Apr 2023 12:33:58 +0200
Subject: wifi: mt76: add mt76_connac_gen_ppe_thresh utility routine

gen_ppe_thresh routine is shared between mt7615, mt7915, mt7921 and
mt7996 so move it in mt76_connac module.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac.h   |  1 +
 .../net/wireless/mediatek/mt76/mt76_connac_mac.c   | 21 ++++++++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt7915/init.c   | 23 +---------------------
 drivers/net/wireless/mediatek/mt76/mt7921/main.c   | 23 +---------------------
 drivers/net/wireless/mediatek/mt76/mt7996/init.c   | 23 +---------------------
 5 files changed, 25 insertions(+), 66 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
index 84974dfe1953..fc8e999a431a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
@@ -359,6 +359,7 @@ mt76_connac_mutex_release(struct mt76_dev *dev, struct mt76_connac_pm *pm)
 	mutex_unlock(&dev->mutex);
 }
 
+void mt76_connac_gen_ppe_thresh(u8 *he_ppet, int nss);
 int mt76_connac_init_tx_queues(struct mt76_phy *phy, int idx, int n_desc,
 			       int ring_base, u32 flags);
 void mt76_connac_write_hw_txp(struct mt76_dev *dev,
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
index 8d316d3aab8f..bd4ce2c7742d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
@@ -9,6 +9,27 @@
 #define HE_PREP(f, m, v)	le16_encode_bits(le32_get_bits(v, MT_CRXV_HE_##m),\
 						 IEEE80211_RADIOTAP_HE_##f)
 
+void mt76_connac_gen_ppe_thresh(u8 *he_ppet, int nss)
+{
+	static const u8 ppet16_ppet8_ru3_ru0[] = { 0x1c, 0xc7, 0x71 };
+	u8 i, ppet_bits, ppet_size, ru_bit_mask = 0x7; /* HE80 */
+
+	he_ppet[0] = FIELD_PREP(IEEE80211_PPE_THRES_NSS_MASK, nss - 1) |
+		     FIELD_PREP(IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK,
+				ru_bit_mask);
+
+	ppet_bits = IEEE80211_PPE_THRES_INFO_PPET_SIZE *
+		    nss * hweight8(ru_bit_mask) * 2;
+	ppet_size = DIV_ROUND_UP(ppet_bits, 8);
+
+	for (i = 0; i < ppet_size - 1; i++)
+		he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3];
+
+	he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3] &
+			 (0xff >> (8 - (ppet_bits - 1) % 8));
+}
+EXPORT_SYMBOL_GPL(mt76_connac_gen_ppe_thresh);
+
 int mt76_connac_pm_wake(struct mt76_phy *phy, struct mt76_connac_pm *pm)
 {
 	struct mt76_dev *dev = phy->dev;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
index 0d52c30ff8fd..538ec0f8e630 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
@@ -931,27 +931,6 @@ mt7915_set_stream_he_txbf_caps(struct mt7915_phy *phy,
 	}
 }
 
-static void
-mt7915_gen_ppe_thresh(u8 *he_ppet, int nss)
-{
-	u8 i, ppet_bits, ppet_size, ru_bit_mask = 0x7; /* HE80 */
-	static const u8 ppet16_ppet8_ru3_ru0[] = {0x1c, 0xc7, 0x71};
-
-	he_ppet[0] = FIELD_PREP(IEEE80211_PPE_THRES_NSS_MASK, nss - 1) |
-		     FIELD_PREP(IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK,
-				ru_bit_mask);
-
-	ppet_bits = IEEE80211_PPE_THRES_INFO_PPET_SIZE *
-		    nss * hweight8(ru_bit_mask) * 2;
-	ppet_size = DIV_ROUND_UP(ppet_bits, 8);
-
-	for (i = 0; i < ppet_size - 1; i++)
-		he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3];
-
-	he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3] &
-			 (0xff >> (8 - (ppet_bits - 1) % 8));
-}
-
 static int
 mt7915_init_he_caps(struct mt7915_phy *phy, enum nl80211_band band,
 		    struct ieee80211_sband_iftype_data *data)
@@ -1101,7 +1080,7 @@ mt7915_init_he_caps(struct mt7915_phy *phy, enum nl80211_band band,
 		memset(he_cap->ppe_thres, 0, sizeof(he_cap->ppe_thres));
 		if (he_cap_elem->phy_cap_info[6] &
 		    IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
-			mt7915_gen_ppe_thresh(he_cap->ppe_thres, nss);
+			mt76_connac_gen_ppe_thresh(he_cap->ppe_thres, nss);
 		} else {
 			he_cap_elem->phy_cap_info[9] |=
 				u8_encode_bits(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index f1d537a7705a..0c9a472bc81a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -9,27 +9,6 @@
 #include "mt7921.h"
 #include "mcu.h"
 
-static void
-mt7921_gen_ppe_thresh(u8 *he_ppet, int nss)
-{
-	u8 i, ppet_bits, ppet_size, ru_bit_mask = 0x7; /* HE80 */
-	static const u8 ppet16_ppet8_ru3_ru0[] = {0x1c, 0xc7, 0x71};
-
-	he_ppet[0] = FIELD_PREP(IEEE80211_PPE_THRES_NSS_MASK, nss - 1) |
-		     FIELD_PREP(IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK,
-				ru_bit_mask);
-
-	ppet_bits = IEEE80211_PPE_THRES_INFO_PPET_SIZE *
-		    nss * hweight8(ru_bit_mask) * 2;
-	ppet_size = DIV_ROUND_UP(ppet_bits, 8);
-
-	for (i = 0; i < ppet_size - 1; i++)
-		he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3];
-
-	he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3] &
-			 (0xff >> (8 - (ppet_bits - 1) % 8));
-}
-
 static int
 mt7921_init_he_caps(struct mt7921_phy *phy, enum nl80211_band band,
 		    struct ieee80211_sband_iftype_data *data)
@@ -168,7 +147,7 @@ mt7921_init_he_caps(struct mt7921_phy *phy, enum nl80211_band band,
 		memset(he_cap->ppe_thres, 0, sizeof(he_cap->ppe_thres));
 		if (he_cap_elem->phy_cap_info[6] &
 		    IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
-			mt7921_gen_ppe_thresh(he_cap->ppe_thres, nss);
+			mt76_connac_gen_ppe_thresh(he_cap->ppe_thres, nss);
 		} else {
 			he_cap_elem->phy_cap_info[9] |=
 				u8_encode_bits(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index 539e5cc4aa2b..ddea5f48816d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -557,27 +557,6 @@ mt7996_set_stream_he_txbf_caps(struct mt7996_phy *phy,
 	elem->phy_cap_info[7] |= c;
 }
 
-static void
-mt7996_gen_ppe_thresh(u8 *he_ppet, int nss)
-{
-	u8 i, ppet_bits, ppet_size, ru_bit_mask = 0x7; /* HE80 */
-	static const u8 ppet16_ppet8_ru3_ru0[] = {0x1c, 0xc7, 0x71};
-
-	he_ppet[0] = FIELD_PREP(IEEE80211_PPE_THRES_NSS_MASK, nss - 1) |
-		     FIELD_PREP(IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK,
-				ru_bit_mask);
-
-	ppet_bits = IEEE80211_PPE_THRES_INFO_PPET_SIZE *
-		    nss * hweight8(ru_bit_mask) * 2;
-	ppet_size = DIV_ROUND_UP(ppet_bits, 8);
-
-	for (i = 0; i < ppet_size - 1; i++)
-		he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3];
-
-	he_ppet[i + 1] = ppet16_ppet8_ru3_ru0[i % 3] &
-			 (0xff >> (8 - (ppet_bits - 1) % 8));
-}
-
 static void
 mt7996_init_he_caps(struct mt7996_phy *phy, enum nl80211_band band,
 		    struct ieee80211_sband_iftype_data *data,
@@ -683,7 +662,7 @@ mt7996_init_he_caps(struct mt7996_phy *phy, enum nl80211_band band,
 	memset(he_cap->ppe_thres, 0, sizeof(he_cap->ppe_thres));
 	if (he_cap_elem->phy_cap_info[6] &
 	    IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
-		mt7996_gen_ppe_thresh(he_cap->ppe_thres, nss);
+		mt76_connac_gen_ppe_thresh(he_cap->ppe_thres, nss);
 	} else {
 		he_cap_elem->phy_cap_info[9] |=
 			u8_encode_bits(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US,
-- 
cgit v1.2.3-70-g09d2


From 50cc972dde9647e43e3c1e64a41e0553e92e6a37 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 3 Apr 2023 12:33:59 +0200
Subject: wifi: mt76: mt7921: get rid of eeprom.h

eeprom.h is mostly empty for mt7921, so get rid of it.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt7921/debugfs.c    |  1 -
 drivers/net/wireless/mediatek/mt76/mt7921/eeprom.h | 30 ----------------------
 drivers/net/wireless/mediatek/mt76/mt7921/init.c   |  1 -
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c    |  1 -
 drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h | 11 ++++++++
 5 files changed, 11 insertions(+), 33 deletions(-)
 delete mode 100644 drivers/net/wireless/mediatek/mt76/mt7921/eeprom.h

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c
index 29d8883268f6..d6b6edba2fec 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c
@@ -2,7 +2,6 @@
 /* Copyright (C) 2020 MediaTek Inc. */
 
 #include "mt7921.h"
-#include "eeprom.h"
 
 static int
 mt7921_reg_set(void *data, u64 val)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7921/eeprom.h
deleted file mode 100644
index 4b647278eb30..000000000000
--- a/drivers/net/wireless/mediatek/mt76/mt7921/eeprom.h
+++ /dev/null
@@ -1,30 +0,0 @@
-/* SPDX-License-Identifier: ISC */
-/* Copyright (C) 2020 MediaTek Inc. */
-
-#ifndef __MT7921_EEPROM_H
-#define __MT7921_EEPROM_H
-
-#include "mt7921.h"
-
-enum mt7921_eeprom_field {
-	MT_EE_CHIP_ID =		0x000,
-	MT_EE_VERSION =		0x002,
-	MT_EE_MAC_ADDR =	0x004,
-	MT_EE_WIFI_CONF =	0x07c,
-	MT_EE_HW_TYPE =		0x55b,
-	__MT_EE_MAX =		0x9ff
-};
-
-#define MT_EE_WIFI_CONF_TX_MASK			BIT(0)
-#define MT_EE_WIFI_CONF_BAND_SEL		GENMASK(3, 2)
-
-#define MT_EE_HW_TYPE_ENCAP			BIT(0)
-
-enum mt7921_eeprom_band {
-	MT_EE_NA,
-	MT_EE_5GHZ,
-	MT_EE_2GHZ,
-	MT_EE_DUAL_BAND,
-};
-
-#endif
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
index 33681fbb2fe8..f419f908dda2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
@@ -6,7 +6,6 @@
 #include "mt7921.h"
 #include "mac.h"
 #include "mcu.h"
-#include "eeprom.h"
 
 static const struct ieee80211_iface_limit if_limits[] = {
 	{
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
index c5e7ad06f877..601d5aadb4f4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
@@ -5,7 +5,6 @@
 #include <linux/firmware.h>
 #include "mt7921.h"
 #include "mt7921_trace.h"
-#include "eeprom.h"
 #include "mcu.h"
 #include "mac.h"
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
index f3155d45a774..149acb1662d5 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h
@@ -266,6 +266,17 @@ struct mt7921_phy {
 	bool roc_grant;
 };
 
+enum mt7921_eeprom_field {
+	MT_EE_CHIP_ID =		0x000,
+	MT_EE_VERSION =		0x002,
+	MT_EE_MAC_ADDR =	0x004,
+	MT_EE_WIFI_CONF =	0x07c,
+	MT_EE_HW_TYPE =		0x55b,
+	__MT_EE_MAX =		0x9ff
+};
+
+#define MT_EE_HW_TYPE_ENCAP			BIT(0)
+
 #define mt7921_init_reset(dev)		((dev)->hif_ops->init_reset(dev))
 #define mt7921_dev_reset(dev)		((dev)->hif_ops->reset(dev))
 #define mt7921_mcu_init(dev)		((dev)->hif_ops->mcu_init(dev))
-- 
cgit v1.2.3-70-g09d2


From 140efef36a91c1d7fad7f89c375fba8df1e4459f Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Mon, 3 Apr 2023 12:34:00 +0200
Subject: wifi: mt76: move shared mac definitions in mt76_connac2_mac.h

Move some mac shared definitions between mt7996, mt7921 and mt7915 in
mt76_connac2_mac.h.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/mac.h    | 12 -----
 drivers/net/wireless/mediatek/mt76/mt76_connac.h   | 14 ++++++
 .../net/wireless/mediatek/mt76/mt76_connac2_mac.h  | 22 +++++++++
 drivers/net/wireless/mediatek/mt76/mt7915/mac.h    | 33 +-------------
 drivers/net/wireless/mediatek/mt76/mt7921/dma.c    |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/init.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/mac.c    | 13 +++++-
 drivers/net/wireless/mediatek/mt76/mt7921/mac.h    | 53 ----------------------
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c    |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c    |  2 +-
 .../net/wireless/mediatek/mt76/mt7921/pci_mac.c    |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/sdio.c   |  2 +-
 .../net/wireless/mediatek/mt76/mt7921/sdio_mac.c   |  2 +-
 .../net/wireless/mediatek/mt76/mt7921/sdio_mcu.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/usb.c    |  2 +-
 .../net/wireless/mediatek/mt76/mt7921/usb_mac.c    |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mac.h    | 12 -----
 17 files changed, 59 insertions(+), 120 deletions(-)
 delete mode 100644 drivers/net/wireless/mediatek/mt76/mt7921/mac.h

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.h b/drivers/net/wireless/mediatek/mt76/mt7615/mac.h
index 880c9f74a7f1..d08fbe64c262 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.h
@@ -19,18 +19,6 @@
 #define MT_RXD0_NORMAL_GROUP_3		BIT(27)
 #define MT_RXD0_NORMAL_GROUP_4		BIT(28)
 
-enum rx_pkt_type {
-	PKT_TYPE_TXS,
-	PKT_TYPE_TXRXV,
-	PKT_TYPE_NORMAL,
-	PKT_TYPE_RX_DUP_RFB,
-	PKT_TYPE_RX_TMR,
-	PKT_TYPE_RETRIEVE,
-	PKT_TYPE_TXRX_NOTIFY,
-	PKT_TYPE_RX_EVENT,
-	PKT_TYPE_NORMAL_MCU,
-};
-
 #define MT_RXD1_NORMAL_BSSID		GENMASK(31, 26)
 #define MT_RXD1_NORMAL_PAYLOAD_FORMAT	GENMASK(25, 24)
 #define MT_RXD1_FIRST_AMSDU_FRAME	GENMASK(1, 0)
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
index fc8e999a431a..15653b274f83 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
@@ -6,6 +6,20 @@
 
 #include "mt76.h"
 
+enum rx_pkt_type {
+	PKT_TYPE_TXS,
+	PKT_TYPE_TXRXV,
+	PKT_TYPE_NORMAL,
+	PKT_TYPE_RX_DUP_RFB,
+	PKT_TYPE_RX_TMR,
+	PKT_TYPE_RETRIEVE,
+	PKT_TYPE_TXRX_NOTIFY,
+	PKT_TYPE_RX_EVENT,
+	PKT_TYPE_NORMAL_MCU,
+	PKT_TYPE_RX_FW_MONITOR	= 0x0c,
+	PKT_TYPE_TXRX_NOTIFY_V0	= 0x18,
+};
+
 #define MT76_CONNAC_SCAN_IE_LEN			600
 #define MT76_CONNAC_MAX_NUM_SCHED_SCAN_INTERVAL	 10
 #define MT76_CONNAC_MAX_TIME_SCHED_SCAN_INTERVAL U16_MAX
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h
index f33171bcd343..a5ec0f631385 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h
@@ -32,6 +32,16 @@ enum {
 	MT_LMAC_PSMP0,
 };
 
+#define MT_TX_FREE_MSDU_CNT		GENMASK(9, 0)
+#define MT_TX_FREE_WLAN_ID		GENMASK(23, 14)
+#define MT_TX_FREE_LATENCY		GENMASK(12, 0)
+/* 0: success, others: dropped */
+#define MT_TX_FREE_STATUS		GENMASK(14, 13)
+#define MT_TX_FREE_MSDU_ID		GENMASK(30, 16)
+#define MT_TX_FREE_PAIR			BIT(31)
+/* will support this field in further revision */
+#define MT_TX_FREE_RATE			GENMASK(13, 0)
+
 #define MT_TXD0_Q_IDX			GENMASK(31, 25)
 #define MT_TXD0_PKT_FMT			GENMASK(24, 23)
 #define MT_TXD0_ETH_TYPE_OFFSET		GENMASK(22, 16)
@@ -166,6 +176,15 @@ enum {
 
 #define MT_TXS7_MPDU_RETRY_CNT		GENMASK(31, 23)
 
+/* RXD DW0 */
+#define MT_RXD0_LENGTH			GENMASK(15, 0)
+#define MT_RXD0_PKT_FLAG                GENMASK(19, 16)
+#define MT_RXD0_PKT_TYPE		GENMASK(31, 27)
+
+#define MT_RXD0_NORMAL_ETH_TYPE_OFS	GENMASK(22, 16)
+#define MT_RXD0_NORMAL_IP_SUM		BIT(23)
+#define MT_RXD0_NORMAL_UDP_TCP_SUM	BIT(24)
+
 /* RXD DW1 */
 #define MT_RXD1_NORMAL_WLAN_IDX		GENMASK(9, 0)
 #define MT_RXD1_NORMAL_GROUP_1		BIT(11)
@@ -308,6 +327,9 @@ enum {
 #define MT_CRXV_FOE_HI		GENMASK(6, 0)
 #define MT_CRXV_FOE_SHIFT	13
 
+#define MT_CT_PARSE_LEN			72
+#define MT_CT_DMA_BUF_NUM		2
+
 #define MT_CT_INFO_APPLY_TXD		BIT(0)
 #define MT_CT_INFO_COPY_HOST_TXD_ALL	BIT(1)
 #define MT_CT_INFO_MGMT_FRAME		BIT(2)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mac.h b/drivers/net/wireless/mediatek/mt76/mt7915/mac.h
index 6fa9c79f3e5f..ce94f87e2042 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mac.h
@@ -6,43 +6,12 @@
 
 #include "../mt76_connac2_mac.h"
 
-#define MT_CT_PARSE_LEN			72
-#define MT_CT_DMA_BUF_NUM		2
-
-#define MT_RXD0_LENGTH			GENMASK(15, 0)
-#define MT_RXD0_PKT_TYPE		GENMASK(31, 27)
-
-#define MT_RXD0_NORMAL_ETH_TYPE_OFS	GENMASK(22, 16)
-#define MT_RXD0_NORMAL_IP_SUM		BIT(23)
-#define MT_RXD0_NORMAL_UDP_TCP_SUM	BIT(24)
-
-enum rx_pkt_type {
-	PKT_TYPE_TXS,
-	PKT_TYPE_TXRXV,
-	PKT_TYPE_NORMAL,
-	PKT_TYPE_RX_DUP_RFB,
-	PKT_TYPE_RX_TMR,
-	PKT_TYPE_RETRIEVE,
-	PKT_TYPE_TXRX_NOTIFY,
-	PKT_TYPE_RX_EVENT,
-	PKT_TYPE_RX_FW_MONITOR = 0x0c,
-	PKT_TYPE_TXRX_NOTIFY_V0 = 0x18,
-};
-
 #define MT_TX_FREE_VER			GENMASK(18, 16)
-#define MT_TX_FREE_MSDU_CNT		GENMASK(9, 0)
-#define MT_TX_FREE_MSDU_CNT_V0	GENMASK(6, 0)
-#define MT_TX_FREE_WLAN_ID		GENMASK(23, 14)
-#define MT_TX_FREE_LATENCY		GENMASK(12, 0)
+#define MT_TX_FREE_MSDU_CNT_V0		GENMASK(6, 0)
 /* 0: success, others: dropped */
-#define MT_TX_FREE_MSDU_ID		GENMASK(30, 16)
-#define MT_TX_FREE_PAIR			BIT(31)
 #define MT_TX_FREE_MPDU_HEADER		BIT(30)
 #define MT_TX_FREE_MSDU_ID_V3		GENMASK(14, 0)
 
-/* will support this field in further revision */
-#define MT_TX_FREE_RATE			GENMASK(13, 0)
-
 #define MT_TXS5_F0_FINAL_MPDU		BIT(31)
 #define MT_TXS5_F0_QOS			BIT(30)
 #define MT_TXS5_F0_TX_COUNT		GENMASK(29, 25)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
index 7b33881a4b54..1d85c0c10049 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
@@ -3,7 +3,7 @@
 
 #include "mt7921.h"
 #include "../dma.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 
 static int mt7921_poll_tx(struct napi_struct *napi, int budget)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
index f419f908dda2..e5fdf2df8c36 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
@@ -4,7 +4,7 @@
 #include <linux/etherdevice.h>
 #include <linux/firmware.h>
 #include "mt7921.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 #include "mcu.h"
 
 static const struct ieee80211_iface_limit if_limits[] = {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
index 2eeea43f6bf6..1675bf520481 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
@@ -6,9 +6,20 @@
 #include <linux/timekeeping.h>
 #include "mt7921.h"
 #include "../dma.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 #include "mcu.h"
 
+#define MT_WTBL_TXRX_CAP_RATE_OFFSET	7
+#define MT_WTBL_TXRX_RATE_G2_HE		24
+#define MT_WTBL_TXRX_RATE_G2		12
+
+#define MT_WTBL_AC0_CTT_OFFSET		20
+
+static u32 mt7921_mac_wtbl_lmac_addr(int idx, u8 offset)
+{
+	return MT_WTBL_LMAC_OFFS(idx, 0) + offset * 4;
+}
+
 static struct mt76_wcid *mt7921_rx_get_wcid(struct mt7921_dev *dev,
 					    u16 idx, bool unicast)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.h b/drivers/net/wireless/mediatek/mt76/mt7921/mac.h
deleted file mode 100644
index 8afec600364f..000000000000
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.h
+++ /dev/null
@@ -1,53 +0,0 @@
-/* SPDX-License-Identifier: ISC */
-/* Copyright (C) 2020 MediaTek Inc. */
-
-#ifndef __MT7921_MAC_H
-#define __MT7921_MAC_H
-
-#include "../mt76_connac2_mac.h"
-
-#define MT_CT_PARSE_LEN			72
-#define MT_CT_DMA_BUF_NUM		2
-
-#define MT_RXD0_LENGTH			GENMASK(15, 0)
-#define MT_RXD0_PKT_FLAG                GENMASK(19, 16)
-#define MT_RXD0_PKT_TYPE		GENMASK(31, 27)
-
-#define MT_RXD0_NORMAL_ETH_TYPE_OFS	GENMASK(22, 16)
-#define MT_RXD0_NORMAL_IP_SUM		BIT(23)
-#define MT_RXD0_NORMAL_UDP_TCP_SUM	BIT(24)
-
-enum rx_pkt_type {
-	PKT_TYPE_TXS,
-	PKT_TYPE_TXRXV,
-	PKT_TYPE_NORMAL,
-	PKT_TYPE_RX_DUP_RFB,
-	PKT_TYPE_RX_TMR,
-	PKT_TYPE_RETRIEVE,
-	PKT_TYPE_TXRX_NOTIFY,
-	PKT_TYPE_RX_EVENT,
-	PKT_TYPE_NORMAL_MCU,
-};
-
-#define MT_TX_FREE_MSDU_CNT		GENMASK(9, 0)
-#define MT_TX_FREE_WLAN_ID		GENMASK(23, 14)
-#define MT_TX_FREE_LATENCY		GENMASK(12, 0)
-/* 0: success, others: dropped */
-#define MT_TX_FREE_STATUS		GENMASK(14, 13)
-#define MT_TX_FREE_MSDU_ID		GENMASK(30, 16)
-#define MT_TX_FREE_PAIR			BIT(31)
-/* will support this field in further revision */
-#define MT_TX_FREE_RATE			GENMASK(13, 0)
-
-#define MT_WTBL_TXRX_CAP_RATE_OFFSET	7
-#define MT_WTBL_TXRX_RATE_G2_HE		24
-#define MT_WTBL_TXRX_RATE_G2		12
-
-#define MT_WTBL_AC0_CTT_OFFSET		20
-
-static inline u32 mt7921_mac_wtbl_lmac_addr(int idx, u8 offset)
-{
-	return MT_WTBL_LMAC_OFFS(idx, 0) + offset * 4;
-}
-
-#endif
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
index 601d5aadb4f4..04925e66f15b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
@@ -6,7 +6,7 @@
 #include "mt7921.h"
 #include "mt7921_trace.h"
 #include "mcu.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 
 #define MT_STA_BFER			BIT(0)
 #define MT_STA_BFEE			BIT(1)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 35ea76c796a3..40bea17cc440 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -8,7 +8,7 @@
 #include <linux/pci.h>
 
 #include "mt7921.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 #include "mcu.h"
 #include "../trace.h"
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
index 8dd60408b117..6053a2556c20 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
@@ -3,7 +3,7 @@
 
 #include "mt7921.h"
 #include "../dma.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 
 int mt7921e_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 			   enum mt76_txq_id qid, struct mt76_wcid *wcid,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
index fe396f195dd7..a77a309c0d60 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c
@@ -13,7 +13,7 @@
 
 #include "mt7921.h"
 #include "../sdio.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 #include "mcu.h"
 
 static const struct sdio_device_id mt7921s_table[] = {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c
index 1b3adb3d91e8..cff9925c41ea 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c
@@ -4,7 +4,7 @@
 #include <linux/iopoll.h>
 #include <linux/mmc/sdio_func.h>
 #include "mt7921.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 #include "../sdio.h"
 
 static void mt7921s_enable_irq(struct mt76_dev *dev)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c
index 5c1489766d9f..177679ce1c80 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c
@@ -8,7 +8,7 @@
 
 #include "mt7921.h"
 #include "../sdio.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 #include "mcu.h"
 #include "regs.h"
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
index 014476f8ad53..46118421ef28 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
@@ -10,7 +10,7 @@
 
 #include "mt7921.h"
 #include "mcu.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 
 static const struct usb_device_id mt7921u_device_table[] = {
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x0e8d, 0x7961, 0xff, 0xff, 0xff),
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb_mac.c
index efbd3954c883..50eb6e7fd6b5 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/usb_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb_mac.c
@@ -10,7 +10,7 @@
 
 #include "mt7921.h"
 #include "mcu.h"
-#include "mac.h"
+#include "../mt76_connac2_mac.h"
 
 static u32 mt7921u_uhw_rr(struct mt76_dev *dev, u32 addr)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
index 2cc218f735d8..0318e16fcfab 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
@@ -20,18 +20,6 @@
 #define MT_RXD0_SW_PKT_TYPE_MAP		0x380F
 #define MT_RXD0_SW_PKT_TYPE_FRAME	0x3801
 
-enum rx_pkt_type {
-	PKT_TYPE_TXS,
-	PKT_TYPE_TXRXV,
-	PKT_TYPE_NORMAL,
-	PKT_TYPE_RX_DUP_RFB,
-	PKT_TYPE_RX_TMR,
-	PKT_TYPE_RETRIEVE,
-	PKT_TYPE_TXRX_NOTIFY,
-	PKT_TYPE_RX_EVENT,
-	PKT_TYPE_RX_FW_MONITOR = 0x0c,
-};
-
 /* RXD DW1 */
 #define MT_RXD1_NORMAL_WLAN_IDX		GENMASK(11, 0)
 #define MT_RXD1_NORMAL_GROUP_1		BIT(16)
-- 
cgit v1.2.3-70-g09d2


From c397fc1e6365a2a9e5540a85b2c1d4ea412aa0e2 Mon Sep 17 00:00:00 2001
From: Quan Zhou <quan.zhou@mediatek.com>
Date: Sat, 18 Mar 2023 15:41:12 +0800
Subject: wifi: mt76: mt7921e: fix probe timeout after reboot

In system warm reboot scene, due to the polling timeout(now 1000us)
is too short to wait dma idle in time, it may make driver probe fail
with error code -ETIMEDOUT. Meanwhile, we also found the dma may take
around 70ms to enter idle state. Change the polling idle timeout to
100ms to avoid the probabilistic probe fail.

Tested pass with 5000 times warm reboot on x86 platform.

[4.477496] pci 0000:01:00.0: attach allowed to drvr mt7921e [internal device]
[4.478306] mt7921e 0000:01:00.0: ASIC revision: 79610010
[4.480063] mt7921e: probe of 0000:01:00.0 failed with error -110

Fixes: 0a1059d0f060 ("mt76: mt7921: move mt7921_dma_reset in dma.c")
Signed-off-by: Quan Zhou <quan.zhou@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/dma.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
index 1d85c0c10049..28207172c4ef 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
@@ -90,9 +90,9 @@ static int mt7921_dma_disable(struct mt7921_dev *dev, bool force)
 		   MT_WFDMA0_GLO_CFG_OMIT_RX_INFO |
 		   MT_WFDMA0_GLO_CFG_OMIT_RX_INFO_PFET2);
 
-	if (!mt76_poll(dev, MT_WFDMA0_GLO_CFG,
-		       MT_WFDMA0_GLO_CFG_TX_DMA_BUSY |
-		       MT_WFDMA0_GLO_CFG_RX_DMA_BUSY, 0, 1000))
+	if (!mt76_poll_msec_tick(dev, MT_WFDMA0_GLO_CFG,
+				 MT_WFDMA0_GLO_CFG_TX_DMA_BUSY |
+				 MT_WFDMA0_GLO_CFG_RX_DMA_BUSY, 0, 100, 1))
 		return -ETIMEDOUT;
 
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 23792cedaff02351b57bddd8957fc917fa88f2e0 Mon Sep 17 00:00:00 2001
From: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Date: Thu, 23 Mar 2023 21:26:12 +0800
Subject: wifi: mt76: fix 6GHz high channel not be scanned

mt76 scan command only support 64 channels currently. If the
channel count is larger than 64(for 2+5+6GHz), some channels will
not be scanned. Hence change the scan type to full channel scan
in case of the command cannot include proper list for chip.

Fixes: 399090ef9605 ("mt76: mt76_connac: move hw_scan and sched_scan routine in mt76_connac_mcu module")
Reported-by: Ben Greear <greearb@candelatech.com>
Tested-by: Isaac Konikoff <konikofi@candelatech.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index 008ece1b16f8..cb27885b5da7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -1678,8 +1678,16 @@ int mt76_connac_mcu_hw_scan(struct mt76_phy *phy, struct ieee80211_vif *vif,
 	req->channel_min_dwell_time = cpu_to_le16(duration);
 	req->channel_dwell_time = cpu_to_le16(duration);
 
-	req->channels_num = min_t(u8, sreq->n_channels, 32);
-	req->ext_channels_num = min_t(u8, ext_channels_num, 32);
+	if (sreq->n_channels == 0 || sreq->n_channels > 64) {
+		req->channel_type = 0;
+		req->channels_num = 0;
+		req->ext_channels_num = 0;
+	} else {
+		req->channel_type = 4;
+		req->channels_num = min_t(u8, sreq->n_channels, 32);
+		req->ext_channels_num = min_t(u8, ext_channels_num, 32);
+	}
+
 	for (i = 0; i < req->channels_num + req->ext_channels_num; i++) {
 		if (i >= 32)
 			chan = &req->ext_channels[i - 32];
@@ -1699,7 +1707,6 @@ int mt76_connac_mcu_hw_scan(struct mt76_phy *phy, struct ieee80211_vif *vif,
 		}
 		chan->channel_num = scan_list[i]->hw_value;
 	}
-	req->channel_type = sreq->n_channels ? 4 : 0;
 
 	if (sreq->ie_len > 0) {
 		memcpy(req->ies, sreq->ie, sreq->ie_len);
-- 
cgit v1.2.3-70-g09d2


From 268079896060b224cb0a42eb6ca0d0a0836c46bb Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Fri, 31 Mar 2023 06:10:17 +0800
Subject: wifi: mt76: mt7915: drop redundant prefix of mt7915_txpower_puts()

Just a cosmetic patch to drop redundant prefix for txpower ouput text.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt7915/debugfs.c    | 36 +++++++++++-----------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c
index 5a46813a59ea..879884ead660 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c
@@ -958,10 +958,10 @@ mt7915_xmit_queues_show(struct seq_file *file, void *data)
 
 DEFINE_SHOW_ATTRIBUTE(mt7915_xmit_queues);
 
-#define mt7915_txpower_puts(prefix, rate)					\
+#define mt7915_txpower_puts(rate)						\
 ({										\
-	len += scnprintf(buf + len, sz - len, "%-16s:", #prefix " (tmac)");	\
-	for (i = 0; i < mt7915_sku_group_len[rate]; i++, offs++)		\
+	len += scnprintf(buf + len, sz - len, "%-16s:", #rate " (TMAC)");	\
+	for (i = 0; i < mt7915_sku_group_len[SKU_##rate]; i++, offs++)		\
 		len += scnprintf(buf + len, sz - len, " %6d", txpwr[offs]);	\
 	len += scnprintf(buf + len, sz - len, "\n");				\
 })
@@ -1004,41 +1004,41 @@ mt7915_rate_txpower_get(struct file *file, char __user *user_buf,
 			 phy != &dev->phy, phy->mt76->chandef.chan->hw_value);
 	len += scnprintf(buf + len, sz - len, "%-16s  %6s %6s %6s %6s\n",
 			 " ", "1m", "2m", "5m", "11m");
-	mt7915_txpower_puts(CCK, SKU_CCK);
+	mt7915_txpower_puts(CCK);
 
 	len += scnprintf(buf + len, sz - len,
 			 "%-16s  %6s %6s %6s %6s %6s %6s %6s %6s\n",
 			 " ", "6m", "9m", "12m", "18m", "24m", "36m", "48m",
 			 "54m");
-	mt7915_txpower_puts(OFDM, SKU_OFDM);
+	mt7915_txpower_puts(OFDM);
 
 	len += scnprintf(buf + len, sz - len,
 			 "%-16s  %6s %6s %6s %6s %6s %6s %6s %6s\n",
 			 " ", "mcs0", "mcs1", "mcs2", "mcs3", "mcs4",
 			 "mcs5", "mcs6", "mcs7");
-	mt7915_txpower_puts(HT20, SKU_HT_BW20);
+	mt7915_txpower_puts(HT_BW20);
 
 	len += scnprintf(buf + len, sz - len,
 			 "%-16s  %6s %6s %6s %6s %6s %6s %6s %6s %6s\n",
 			 " ", "mcs0", "mcs1", "mcs2", "mcs3", "mcs4", "mcs5",
 			 "mcs6", "mcs7", "mcs32");
-	mt7915_txpower_puts(HT40, SKU_HT_BW40);
+	mt7915_txpower_puts(HT_BW40);
 
 	len += scnprintf(buf + len, sz - len,
 			 "%-16s  %6s %6s %6s %6s %6s %6s %6s %6s %6s %6s %6s %6s\n",
 			 " ", "mcs0", "mcs1", "mcs2", "mcs3", "mcs4", "mcs5",
 			 "mcs6", "mcs7", "mcs8", "mcs9", "mcs10", "mcs11");
-	mt7915_txpower_puts(VHT20, SKU_VHT_BW20);
-	mt7915_txpower_puts(VHT40, SKU_VHT_BW40);
-	mt7915_txpower_puts(VHT80, SKU_VHT_BW80);
-	mt7915_txpower_puts(VHT160, SKU_VHT_BW160);
-	mt7915_txpower_puts(HE26, SKU_HE_RU26);
-	mt7915_txpower_puts(HE52, SKU_HE_RU52);
-	mt7915_txpower_puts(HE106, SKU_HE_RU106);
-	mt7915_txpower_puts(HE242, SKU_HE_RU242);
-	mt7915_txpower_puts(HE484, SKU_HE_RU484);
-	mt7915_txpower_puts(HE996, SKU_HE_RU996);
-	mt7915_txpower_puts(HE996x2, SKU_HE_RU2x996);
+	mt7915_txpower_puts(VHT_BW20);
+	mt7915_txpower_puts(VHT_BW40);
+	mt7915_txpower_puts(VHT_BW80);
+	mt7915_txpower_puts(VHT_BW160);
+	mt7915_txpower_puts(HE_RU26);
+	mt7915_txpower_puts(HE_RU52);
+	mt7915_txpower_puts(HE_RU106);
+	mt7915_txpower_puts(HE_RU242);
+	mt7915_txpower_puts(HE_RU484);
+	mt7915_txpower_puts(HE_RU996);
+	mt7915_txpower_puts(HE_RU2x996);
 
 	reg = is_mt7915(&dev->mt76) ? MT_WF_PHY_TPC_CTRL_STAT(band) :
 	      MT_WF_PHY_TPC_CTRL_STAT_MT7916(band);
-- 
cgit v1.2.3-70-g09d2


From 03eb52dd78cab08f13925aeec8315fbdbcba3253 Mon Sep 17 00:00:00 2001
From: Reese Russell <git@qrsnap.io>
Date: Tue, 4 Apr 2023 00:35:12 -0700
Subject: wifi: mt76: mt7921: add Netgear AXE3000 (A8000) support

Issue: Though the Netgear AXE3000 (A8000) is based on the mt7921
chipset because of the unique USB VID:PID combination this device
does not initialize/register. Thus making it not plug and play.

Fix: Adds support for the Netgear AXE3000 (A8000) based on the Mediatek
mt7921au chipset. The method of action is adding the USD VID/PID
pair to the mt7921u_device_table[] array.

Notes: A retail sample of the Netgear AXE3000 (A8000) yeilds the following
from lsusb D 0846:9060 NetGear, Inc. Wireless_Device. This pair
0846:9060 VID:PID has been reported by other users on Github.

Signed-off-by: Reese Russell <git@qrsnap.io>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/usb.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
index 46118421ef28..9b05e4577cdf 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
@@ -18,6 +18,9 @@ static const struct usb_device_id mt7921u_device_table[] = {
 	/* Comfast CF-952AX */
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x3574, 0x6211, 0xff, 0xff, 0xff),
 		.driver_info = (kernel_ulong_t)MT7921_FIRMWARE_WM },
+	/* Netgear, Inc. [A8000,AXE3000] */
+	{ USB_DEVICE_AND_INTERFACE_INFO(0x0846, 0x9060, 0xff, 0xff, 0xff),
+		.driver_info = (kernel_ulong_t)MT7921_FIRMWARE_WM },
 	{ },
 };
 
-- 
cgit v1.2.3-70-g09d2


From 2631c5b6ef9d7c6707e020def6947464c8aa6f92 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Thu, 6 Apr 2023 08:32:12 -0600
Subject: wifi: mt76: Replace zero-length array with flexible-array member

Zero-length arrays are deprecated [1] and have to be replaced by C99
flexible-array members.

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE routines
on memcpy() and help to make progress towards globally enabling
-fstrict-flex-arrays=3 [2]

Link: https://github.com/KSPP/linux/issues/78 [1]
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [2]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 40a99e0caded..9dbfe26e87e0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -127,7 +127,7 @@ struct mt76_connac2_mcu_rxd {
 	u8 rsv1[2];
 	u8 s2d_index;
 
-	u8 tlv[0];
+	u8 tlv[];
 };
 
 struct mt76_connac2_patch_hdr {
-- 
cgit v1.2.3-70-g09d2


From 6d6793cef6a491b8f6db5f40ef3334411293da32 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Fri, 17 Mar 2023 14:56:39 -0600
Subject: wifi: mt76: mt7921: Replace fake flex-arrays with flexible-array
 members
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Address the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.c:266:25: warning: array subscript 0 is outside array bounds of ‘struct mt7921_asar_dyn_limit_v2[0]’ [-Warray-bounds=]
drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.c:263:25: warning: array subscript 0 is outside array bounds of ‘struct mt7921_asar_dyn_limit[0]’ [-Warray-bounds=]
drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.c:223:28: warning: array subscript <unknown> is outside array bounds of ‘struct mt7921_asar_geo_limit_v2[0]’ [-Warray-bounds=]
drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.c:220:28: warning: array subscript <unknown> is outside array bounds of ‘struct mt7921_asar_geo_limit[0]’ [-Warray-bounds=]
drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.c:334:37: warning: array subscript i is outside array bounds of ‘u8[0]’ {aka ‘unsigned char[]’} [-Warray-bounds=]

Notice that the DECLARE_FLEX_ARRAY() helper allows for flexible-array
members in unions.

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/272
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.h b/drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.h
index 35268b0890ad..6f2c4a572572 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/acpi_sar.h
@@ -24,7 +24,7 @@ struct mt7921_asar_dyn {
 	u8 names[4];
 	u8 enable;
 	u8 nr_tbl;
-	struct mt7921_asar_dyn_limit tbl[0];
+	DECLARE_FLEX_ARRAY(struct mt7921_asar_dyn_limit, tbl);
 } __packed;
 
 struct mt7921_asar_dyn_limit_v2 {
@@ -37,7 +37,7 @@ struct mt7921_asar_dyn_v2 {
 	u8 enable;
 	u8 rsvd;
 	u8 nr_tbl;
-	struct mt7921_asar_dyn_limit_v2 tbl[0];
+	DECLARE_FLEX_ARRAY(struct mt7921_asar_dyn_limit_v2, tbl);
 } __packed;
 
 struct mt7921_asar_geo_band {
@@ -55,7 +55,7 @@ struct mt7921_asar_geo {
 	u8 names[4];
 	u8 version;
 	u8 nr_tbl;
-	struct mt7921_asar_geo_limit tbl[0];
+	DECLARE_FLEX_ARRAY(struct mt7921_asar_geo_limit, tbl);
 } __packed;
 
 struct mt7921_asar_geo_limit_v2 {
@@ -69,7 +69,7 @@ struct mt7921_asar_geo_v2 {
 	u8 version;
 	u8 rsvd;
 	u8 nr_tbl;
-	struct mt7921_asar_geo_limit_v2 tbl[0];
+	DECLARE_FLEX_ARRAY(struct mt7921_asar_geo_limit_v2, tbl);
 } __packed;
 
 struct mt7921_asar_cl {
@@ -85,7 +85,7 @@ struct mt7921_asar_fg {
 	u8 rsvd;
 	u8 nr_flag;
 	u8 rsvd1;
-	u8 flag[0];
+	u8 flag[];
 } __packed;
 
 struct mt7921_acpi_sar {
-- 
cgit v1.2.3-70-g09d2


From 1b83d17ccece7db5e0ad4602969c1146f1381c3d Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 8 Mar 2023 05:50:59 +0800
Subject: wifi: mt76: mt7921: enable p2p support

Introduce p2p-go/p2p-client support to mt7921 driver

CONNECTION_P2P_GC/GO is not supported with the current firmware
so we added mt76_dev to mt76_connac_mcu_sta_basic_tlv signature to
use CONNECTION_INFRA_STA/AP instead for p2p-client and p2p-go
respectively to make it work.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/mcu.c      |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c |  8 ++++----
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c      |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7921/init.c     | 10 +++++++---
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c      |  4 ++--
 6 files changed, 18 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
index eea398c79a98..195fe1094d9b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
@@ -861,7 +861,8 @@ mt7615_mcu_wtbl_sta_add(struct mt7615_phy *phy, struct ieee80211_vif *vif,
 		else
 			mvif->sta_added = true;
 	}
-	mt76_connac_mcu_sta_basic_tlv(sskb, vif, sta, enable, new_entry);
+	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, sskb, vif, sta, enable,
+				      new_entry);
 	if (enable && sta)
 		mt76_connac_mcu_sta_tlv(phy->mt76, sskb, sta, vif, 0,
 					MT76_STA_INFO_STATE_ASSOC);
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index cb27885b5da7..0f0a519f956f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -363,7 +363,7 @@ void mt76_connac_mcu_bss_omac_tlv(struct sk_buff *skb,
 }
 EXPORT_SYMBOL_GPL(mt76_connac_mcu_bss_omac_tlv);
 
-void mt76_connac_mcu_sta_basic_tlv(struct sk_buff *skb,
+void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 				   struct ieee80211_vif *vif,
 				   struct ieee80211_sta *sta,
 				   bool enable, bool newly)
@@ -394,7 +394,7 @@ void mt76_connac_mcu_sta_basic_tlv(struct sk_buff *skb,
 	switch (vif->type) {
 	case NL80211_IFTYPE_MESH_POINT:
 	case NL80211_IFTYPE_AP:
-		if (vif->p2p)
+		if (vif->p2p && !is_mt7921(dev))
 			conn_type = CONNECTION_P2P_GC;
 		else
 			conn_type = CONNECTION_INFRA_STA;
@@ -402,7 +402,7 @@ void mt76_connac_mcu_sta_basic_tlv(struct sk_buff *skb,
 		basic->aid = cpu_to_le16(sta->aid);
 		break;
 	case NL80211_IFTYPE_STATION:
-		if (vif->p2p)
+		if (vif->p2p && !is_mt7921(dev))
 			conn_type = CONNECTION_P2P_GO;
 		else
 			conn_type = CONNECTION_INFRA_AP;
@@ -1029,7 +1029,7 @@ int mt76_connac_mcu_sta_cmd(struct mt76_phy *phy,
 		return PTR_ERR(skb);
 
 	if (info->sta || !info->offload_fw)
-		mt76_connac_mcu_sta_basic_tlv(skb, info->vif, info->sta,
+		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif, info->sta,
 					      info->enable, info->newly);
 	if (info->sta && info->enable)
 		mt76_connac_mcu_sta_tlv(phy, skb, info->sta,
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 9dbfe26e87e0..ac7d5817f345 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1776,7 +1776,7 @@ mt76_connac_mcu_add_tlv(struct sk_buff *skb, int tag, int len)
 
 int mt76_connac_mcu_set_channel_domain(struct mt76_phy *phy);
 int mt76_connac_mcu_set_vif_ps(struct mt76_dev *dev, struct ieee80211_vif *vif);
-void mt76_connac_mcu_sta_basic_tlv(struct sk_buff *skb,
+void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 				   struct ieee80211_vif *vif,
 				   struct ieee80211_sta *sta, bool enable,
 				   bool newly);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
index 7432b1eb396d..9fcb22fa1f97 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
@@ -1654,8 +1654,8 @@ int mt7915_mcu_add_sta(struct mt7915_dev *dev, struct ieee80211_vif *vif,
 		return PTR_ERR(skb);
 
 	/* starec basic */
-	mt76_connac_mcu_sta_basic_tlv(skb, vif, sta, enable,
-			!rcu_access_pointer(dev->mt76.wcid[msta->wcid.idx]));
+	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, skb, vif, sta, enable,
+				      !rcu_access_pointer(dev->mt76.wcid[msta->wcid.idx]));
 	if (!enable)
 		goto out;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
index e5fdf2df8c36..5955c7c5b91a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
@@ -31,11 +31,13 @@ static const struct ieee80211_iface_combination if_comb[] = {
 static const struct ieee80211_iface_limit if_limits_chanctx[] = {
 	{
 		.max = 2,
-		.types = BIT(NL80211_IFTYPE_STATION),
+		.types = BIT(NL80211_IFTYPE_STATION) |
+			 BIT(NL80211_IFTYPE_P2P_CLIENT)
 	},
 	{
 		.max = 1,
-		.types = BIT(NL80211_IFTYPE_AP),
+		.types = BIT(NL80211_IFTYPE_AP) |
+			 BIT(NL80211_IFTYPE_P2P_GO)
 	}
 };
 
@@ -99,7 +101,9 @@ mt7921_init_wiphy(struct ieee80211_hw *hw)
 	wiphy->flags &= ~(WIPHY_FLAG_IBSS_RSN | WIPHY_FLAG_4ADDR_AP |
 			  WIPHY_FLAG_4ADDR_STATION);
 	wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) |
-				 BIT(NL80211_IFTYPE_AP);
+				 BIT(NL80211_IFTYPE_AP) |
+				 BIT(NL80211_IFTYPE_P2P_CLIENT) |
+				 BIT(NL80211_IFTYPE_P2P_GO);
 	wiphy->max_remain_on_channel_duration = 5000;
 	wiphy->max_scan_ie_len = MT76_CONNAC_SCAN_IE_LEN;
 	wiphy->max_scan_ssids = 4;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 2a66ac546513..016853033c1f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -1694,8 +1694,8 @@ int mt7996_mcu_add_sta(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 		return PTR_ERR(skb);
 
 	/* starec basic */
-	mt76_connac_mcu_sta_basic_tlv(skb, vif, sta, enable,
-			!rcu_access_pointer(dev->mt76.wcid[msta->wcid.idx]));
+	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, skb, vif, sta, enable,
+				      !rcu_access_pointer(dev->mt76.wcid[msta->wcid.idx]));
 	if (!enable)
 		goto out;
 
-- 
cgit v1.2.3-70-g09d2


From 27015b6fbcca836c6dbf196afc266e068af4aeec Mon Sep 17 00:00:00 2001
From: Bo Jiao <Bo.Jiao@mediatek.com>
Date: Wed, 5 Apr 2023 07:24:37 +0800
Subject: wifi: mt76: mt7996: enable full system reset support

Add mt7996_reset() and refactor mt7996_mac_reset_work() to support
full system recovery.

Signed-off-by: Bo Jiao <Bo.Jiao@mediatek.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/dma.c    |  64 +++++
 drivers/net/wireless/mediatek/mt76/mt7996/init.c   |  11 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    | 284 +++++++++++++++++----
 drivers/net/wireless/mediatek/mt76/mt7996/main.c   |  18 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    |  22 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mmio.c   |   7 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  17 +-
 drivers/net/wireless/mediatek/mt76/mt7996/regs.h   |  13 +-
 8 files changed, 365 insertions(+), 71 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c
index c09fe4274935..534143465d9b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c
@@ -352,6 +352,70 @@ int mt7996_dma_init(struct mt7996_dev *dev)
 	return 0;
 }
 
+void mt7996_dma_reset(struct mt7996_dev *dev, bool force)
+{
+	struct mt76_phy *phy2 = dev->mt76.phys[MT_BAND1];
+	struct mt76_phy *phy3 = dev->mt76.phys[MT_BAND2];
+	u32 hif1_ofs = MT_WFDMA0_PCIE1(0) - MT_WFDMA0(0);
+	int i;
+
+	mt76_clear(dev, MT_WFDMA0_GLO_CFG,
+		   MT_WFDMA0_GLO_CFG_TX_DMA_EN |
+		   MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+
+	if (dev->hif2)
+		mt76_clear(dev, MT_WFDMA0_GLO_CFG + hif1_ofs,
+			   MT_WFDMA0_GLO_CFG_TX_DMA_EN |
+			   MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+
+	usleep_range(1000, 2000);
+
+	for (i = 0; i < __MT_TXQ_MAX; i++) {
+		mt76_queue_tx_cleanup(dev, dev->mphy.q_tx[i], true);
+		if (phy2)
+			mt76_queue_tx_cleanup(dev, phy2->q_tx[i], true);
+		if (phy3)
+			mt76_queue_tx_cleanup(dev, phy3->q_tx[i], true);
+	}
+
+	for (i = 0; i < __MT_MCUQ_MAX; i++)
+		mt76_queue_tx_cleanup(dev, dev->mt76.q_mcu[i], true);
+
+	mt76_for_each_q_rx(&dev->mt76, i)
+		mt76_queue_rx_cleanup(dev, &dev->mt76.q_rx[i]);
+
+	mt76_tx_status_check(&dev->mt76, true);
+
+	/* reset wfsys */
+	if (force)
+		mt7996_wfsys_reset(dev);
+
+	mt7996_dma_disable(dev, force);
+
+	/* reset hw queues */
+	for (i = 0; i < __MT_TXQ_MAX; i++) {
+		mt76_queue_reset(dev, dev->mphy.q_tx[i]);
+		if (phy2)
+			mt76_queue_reset(dev, phy2->q_tx[i]);
+		if (phy3)
+			mt76_queue_reset(dev, phy3->q_tx[i]);
+	}
+
+	for (i = 0; i < __MT_MCUQ_MAX; i++)
+		mt76_queue_reset(dev, dev->mt76.q_mcu[i]);
+
+	mt76_for_each_q_rx(&dev->mt76, i) {
+		mt76_queue_reset(dev, &dev->mt76.q_rx[i]);
+	}
+
+	mt76_tx_status_check(&dev->mt76, true);
+
+	mt76_for_each_q_rx(&dev->mt76, i)
+		mt76_queue_rx_reset(dev, i);
+
+	mt7996_dma_enable(dev);
+}
+
 void mt7996_dma_cleanup(struct mt7996_dev *dev)
 {
 	mt7996_dma_disable(dev, true);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index ddea5f48816d..6de9a0405b74 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -99,9 +99,8 @@ static void mt7996_led_set_brightness(struct led_classdev *led_cdev,
 		mt7996_led_set_config(led_cdev, 0xff, 0);
 }
 
-static void
-mt7996_init_txpower(struct mt7996_dev *dev,
-		    struct ieee80211_supported_band *sband)
+void mt7996_init_txpower(struct mt7996_dev *dev,
+			 struct ieee80211_supported_band *sband)
 {
 	int i, nss = hweight8(dev->mphy.antenna_mask);
 	int nss_delta = mt76_tx_power_nss_delta(nss);
@@ -255,7 +254,7 @@ mt7996_mac_init_band(struct mt7996_dev *dev, u8 band)
 	mt76_rmw(dev, MT_WTBLOFF_RSCR(band), mask, set);
 }
 
-static void mt7996_mac_init(struct mt7996_dev *dev)
+void mt7996_mac_init(struct mt7996_dev *dev)
 {
 #define HIF_TXD_V2_1	4
 	int i;
@@ -289,7 +288,7 @@ static void mt7996_mac_init(struct mt7996_dev *dev)
 		mt7996_mac_init_band(dev, i);
 }
 
-static int mt7996_txbf_init(struct mt7996_dev *dev)
+int mt7996_txbf_init(struct mt7996_dev *dev)
 {
 	int ret;
 
@@ -870,6 +869,8 @@ int mt7996_register_device(struct mt7996_dev *dev)
 	if (ret)
 		return ret;
 
+	dev->recovery.hw_init_done = true;
+
 	return mt7996_init_debugfs(&dev->phy);
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index 6ce75243e40d..d526d6f53ace 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -1704,7 +1704,7 @@ mt7996_wait_reset_state(struct mt7996_dev *dev, u32 state)
 	bool ret;
 
 	ret = wait_event_timeout(dev->reset_wait,
-				 (READ_ONCE(dev->reset_state) & state),
+				 (READ_ONCE(dev->recovery.state) & state),
 				 MT7996_RESET_TIMEOUT);
 
 	WARN(!ret, "Timeout waiting for MCU reset state %x\n", state);
@@ -1753,68 +1753,207 @@ mt7996_update_beacons(struct mt7996_dev *dev)
 					    mt7996_update_vif_beacon, phy3->hw);
 }
 
-static void
-mt7996_dma_reset(struct mt7996_dev *dev)
+void mt7996_tx_token_put(struct mt7996_dev *dev)
 {
-	struct mt76_phy *phy2 = dev->mt76.phys[MT_BAND1];
-	struct mt76_phy *phy3 = dev->mt76.phys[MT_BAND2];
-	u32 hif1_ofs = MT_WFDMA0_PCIE1(0) - MT_WFDMA0(0);
-	int i;
+	struct mt76_txwi_cache *txwi;
+	int id;
+
+	spin_lock_bh(&dev->mt76.token_lock);
+	idr_for_each_entry(&dev->mt76.token, txwi, id) {
+		mt7996_txwi_free(dev, txwi, NULL, NULL);
+		dev->mt76.token_count--;
+	}
+	spin_unlock_bh(&dev->mt76.token_lock);
+	idr_destroy(&dev->mt76.token);
+}
+
+static int
+mt7996_mac_restart(struct mt7996_dev *dev)
+{
+	struct mt7996_phy *phy2, *phy3;
+	struct mt76_dev *mdev = &dev->mt76;
+	int i, ret;
 
-	mt76_clear(dev, MT_WFDMA0_GLO_CFG,
-		   MT_WFDMA0_GLO_CFG_TX_DMA_EN |
-		   MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+	phy2 = mt7996_phy2(dev);
+	phy3 = mt7996_phy3(dev);
 
-	if (dev->hif2)
-		mt76_clear(dev, MT_WFDMA0_GLO_CFG + hif1_ofs,
-			   MT_WFDMA0_GLO_CFG_TX_DMA_EN |
-			   MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+	if (dev->hif2) {
+		mt76_wr(dev, MT_INT1_MASK_CSR, 0x0);
+		mt76_wr(dev, MT_INT1_SOURCE_CSR, ~0);
+	}
 
-	usleep_range(1000, 2000);
+	if (dev_is_pci(mdev->dev)) {
+		mt76_wr(dev, MT_PCIE_MAC_INT_ENABLE, 0x0);
+		if (dev->hif2)
+			mt76_wr(dev, MT_PCIE1_MAC_INT_ENABLE, 0x0);
+	}
 
-	for (i = 0; i < __MT_TXQ_MAX; i++) {
-		mt76_queue_tx_cleanup(dev, dev->mphy.q_tx[i], true);
-		if (phy2)
-			mt76_queue_tx_cleanup(dev, phy2->q_tx[i], true);
-		if (phy3)
-			mt76_queue_tx_cleanup(dev, phy3->q_tx[i], true);
+	set_bit(MT76_RESET, &dev->mphy.state);
+	set_bit(MT76_MCU_RESET, &dev->mphy.state);
+	wake_up(&dev->mt76.mcu.wait);
+	if (phy2) {
+		set_bit(MT76_RESET, &phy2->mt76->state);
+		set_bit(MT76_MCU_RESET, &phy2->mt76->state);
+	}
+	if (phy3) {
+		set_bit(MT76_RESET, &phy3->mt76->state);
+		set_bit(MT76_MCU_RESET, &phy3->mt76->state);
 	}
 
-	for (i = 0; i < __MT_MCUQ_MAX; i++)
-		mt76_queue_tx_cleanup(dev, dev->mt76.q_mcu[i], true);
+	/* lock/unlock all queues to ensure that no tx is pending */
+	mt76_txq_schedule_all(&dev->mphy);
+	if (phy2)
+		mt76_txq_schedule_all(phy2->mt76);
+	if (phy3)
+		mt76_txq_schedule_all(phy3->mt76);
 
-	mt76_for_each_q_rx(&dev->mt76, i)
-		mt76_queue_rx_reset(dev, i);
+	/* disable all tx/rx napi */
+	mt76_worker_disable(&dev->mt76.tx_worker);
+	mt76_for_each_q_rx(mdev, i) {
+		if (mdev->q_rx[i].ndesc)
+			napi_disable(&dev->mt76.napi[i]);
+	}
+	napi_disable(&dev->mt76.tx_napi);
+
+	/* token reinit */
+	mt7996_tx_token_put(dev);
+	idr_init(&dev->mt76.token);
+
+	mt7996_dma_reset(dev, true);
+
+	local_bh_disable();
+	mt76_for_each_q_rx(mdev, i) {
+		if (mdev->q_rx[i].ndesc) {
+			napi_enable(&dev->mt76.napi[i]);
+			napi_schedule(&dev->mt76.napi[i]);
+		}
+	}
+	local_bh_enable();
+	clear_bit(MT76_MCU_RESET, &dev->mphy.state);
+	clear_bit(MT76_STATE_MCU_RUNNING, &dev->mphy.state);
+
+	mt76_wr(dev, MT_INT_MASK_CSR, dev->mt76.mmio.irqmask);
+	mt76_wr(dev, MT_INT_SOURCE_CSR, ~0);
+	if (dev->hif2) {
+		mt76_wr(dev, MT_INT1_MASK_CSR, dev->mt76.mmio.irqmask);
+		mt76_wr(dev, MT_INT1_SOURCE_CSR, ~0);
+	}
+	if (dev_is_pci(mdev->dev)) {
+		mt76_wr(dev, MT_PCIE_MAC_INT_ENABLE, 0xff);
+		if (dev->hif2)
+			mt76_wr(dev, MT_PCIE1_MAC_INT_ENABLE, 0xff);
+	}
+
+	/* load firmware */
+	ret = mt7996_mcu_init_firmware(dev);
+	if (ret)
+		goto out;
+
+	/* set the necessary init items */
+	ret = mt7996_mcu_set_eeprom(dev);
+	if (ret)
+		goto out;
+
+	mt7996_mac_init(dev);
+	mt7996_init_txpower(dev, &dev->mphy.sband_2g.sband);
+	mt7996_init_txpower(dev, &dev->mphy.sband_5g.sband);
+	mt7996_init_txpower(dev, &dev->mphy.sband_6g.sband);
+	ret = mt7996_txbf_init(dev);
+
+	if (test_bit(MT76_STATE_RUNNING, &dev->mphy.state)) {
+		ret = mt7996_run(dev->mphy.hw);
+		if (ret)
+			goto out;
+	}
 
-	mt76_tx_status_check(&dev->mt76, true);
+	if (phy2 && test_bit(MT76_STATE_RUNNING, &phy2->mt76->state)) {
+		ret = mt7996_run(phy2->mt76->hw);
+		if (ret)
+			goto out;
+	}
 
-	/* re-init prefetch settings after reset */
-	mt7996_dma_prefetch(dev);
+	if (phy3 && test_bit(MT76_STATE_RUNNING, &phy3->mt76->state)) {
+		ret = mt7996_run(phy3->mt76->hw);
+		if (ret)
+			goto out;
+	}
+
+out:
+	/* reset done */
+	clear_bit(MT76_RESET, &dev->mphy.state);
+	if (phy2)
+		clear_bit(MT76_RESET, &phy2->mt76->state);
+	if (phy3)
+		clear_bit(MT76_RESET, &phy3->mt76->state);
 
-	mt76_set(dev, MT_WFDMA0_GLO_CFG,
-		 MT_WFDMA0_GLO_CFG_TX_DMA_EN | MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+	local_bh_disable();
+	napi_enable(&dev->mt76.tx_napi);
+	napi_schedule(&dev->mt76.tx_napi);
+	local_bh_enable();
 
-	if (dev->hif2)
-		mt76_set(dev, MT_WFDMA0_GLO_CFG + hif1_ofs,
-			 MT_WFDMA0_GLO_CFG_TX_DMA_EN |
-			 MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+	mt76_worker_enable(&dev->mt76.tx_worker);
+	return ret;
 }
 
-void mt7996_tx_token_put(struct mt7996_dev *dev)
+static void
+mt7996_mac_full_reset(struct mt7996_dev *dev)
 {
-	struct mt76_txwi_cache *txwi;
-	int id;
+	struct mt7996_phy *phy2, *phy3;
+	int i;
 
-	spin_lock_bh(&dev->mt76.token_lock);
-	idr_for_each_entry(&dev->mt76.token, txwi, id) {
-		mt7996_txwi_free(dev, txwi, NULL, NULL);
-		dev->mt76.token_count--;
+	phy2 = mt7996_phy2(dev);
+	phy3 = mt7996_phy3(dev);
+	dev->recovery.hw_full_reset = true;
+
+	wake_up(&dev->mt76.mcu.wait);
+	ieee80211_stop_queues(mt76_hw(dev));
+	if (phy2)
+		ieee80211_stop_queues(phy2->mt76->hw);
+	if (phy3)
+		ieee80211_stop_queues(phy3->mt76->hw);
+
+	cancel_delayed_work_sync(&dev->mphy.mac_work);
+	if (phy2)
+		cancel_delayed_work_sync(&phy2->mt76->mac_work);
+	if (phy3)
+		cancel_delayed_work_sync(&phy3->mt76->mac_work);
+
+	mutex_lock(&dev->mt76.mutex);
+	for (i = 0; i < 10; i++) {
+		if (!mt7996_mac_restart(dev))
+			break;
 	}
-	spin_unlock_bh(&dev->mt76.token_lock);
-	idr_destroy(&dev->mt76.token);
+	mutex_unlock(&dev->mt76.mutex);
+
+	if (i == 10)
+		dev_err(dev->mt76.dev, "chip full reset failed\n");
+
+	ieee80211_restart_hw(mt76_hw(dev));
+	if (phy2)
+		ieee80211_restart_hw(phy2->mt76->hw);
+	if (phy3)
+		ieee80211_restart_hw(phy3->mt76->hw);
+
+	ieee80211_wake_queues(mt76_hw(dev));
+	if (phy2)
+		ieee80211_wake_queues(phy2->mt76->hw);
+	if (phy3)
+		ieee80211_wake_queues(phy3->mt76->hw);
+
+	dev->recovery.hw_full_reset = false;
+	ieee80211_queue_delayed_work(mt76_hw(dev),
+				     &dev->mphy.mac_work,
+				     MT7996_WATCHDOG_TIME);
+	if (phy2)
+		ieee80211_queue_delayed_work(phy2->mt76->hw,
+					     &phy2->mt76->mac_work,
+					     MT7996_WATCHDOG_TIME);
+	if (phy3)
+		ieee80211_queue_delayed_work(phy3->mt76->hw,
+					     &phy3->mt76->mac_work,
+					     MT7996_WATCHDOG_TIME);
 }
 
-/* system error recovery */
 void mt7996_mac_reset_work(struct work_struct *work)
 {
 	struct mt7996_phy *phy2, *phy3;
@@ -1825,9 +1964,36 @@ void mt7996_mac_reset_work(struct work_struct *work)
 	phy2 = mt7996_phy2(dev);
 	phy3 = mt7996_phy3(dev);
 
-	if (!(READ_ONCE(dev->reset_state) & MT_MCU_CMD_STOP_DMA))
+	/* chip full reset */
+	if (dev->recovery.restart) {
+		/* disable WA/WM WDT */
+		mt76_clear(dev, MT_WFDMA0_MCU_HOST_INT_ENA,
+			   MT_MCU_CMD_WDT_MASK);
+
+		if (READ_ONCE(dev->recovery.state) & MT_MCU_CMD_WA_WDT)
+			dev->recovery.wa_reset_count++;
+		else
+			dev->recovery.wm_reset_count++;
+
+		mt7996_mac_full_reset(dev);
+
+		/* enable mcu irq */
+		mt7996_irq_enable(dev, MT_INT_MCU_CMD);
+		mt7996_irq_disable(dev, 0);
+
+		/* enable WA/WM WDT */
+		mt76_set(dev, MT_WFDMA0_MCU_HOST_INT_ENA, MT_MCU_CMD_WDT_MASK);
+
+		dev->recovery.state = MT_MCU_CMD_NORMAL_STATE;
+		dev->recovery.restart = false;
 		return;
+	}
 
+	if (!(READ_ONCE(dev->recovery.state) & MT_MCU_CMD_STOP_DMA))
+		return;
+
+	dev_info(dev->mt76.dev,"\n%s L1 SER recovery start.",
+		 wiphy_name(dev->mt76.hw->wiphy));
 	ieee80211_stop_queues(mt76_hw(dev));
 	if (phy2)
 		ieee80211_stop_queues(phy2->mt76->hw);
@@ -1856,7 +2022,7 @@ void mt7996_mac_reset_work(struct work_struct *work)
 	mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_DMA_STOPPED);
 
 	if (mt7996_wait_reset_state(dev, MT_MCU_CMD_RESET_DONE)) {
-		mt7996_dma_reset(dev);
+		mt7996_dma_reset(dev, false);
 
 		mt7996_tx_token_put(dev);
 		idr_init(&dev->mt76.token);
@@ -1911,6 +2077,32 @@ void mt7996_mac_reset_work(struct work_struct *work)
 		ieee80211_queue_delayed_work(phy3->mt76->hw,
 					     &phy3->mt76->mac_work,
 					     MT7996_WATCHDOG_TIME);
+	dev_info(dev->mt76.dev,"\n%s L1 SER recovery completed.",
+		 wiphy_name(dev->mt76.hw->wiphy));
+}
+
+void mt7996_reset(struct mt7996_dev *dev)
+{
+	if (!dev->recovery.hw_init_done)
+		return;
+
+	if (dev->recovery.hw_full_reset)
+		return;
+
+	/* wm/wa exception: do full recovery */
+	if (READ_ONCE(dev->recovery.state) & MT_MCU_CMD_WDT_MASK) {
+		dev->recovery.restart = true;
+		dev_info(dev->mt76.dev,
+			 "%s indicated firmware crash, attempting recovery\n",
+			 wiphy_name(dev->mt76.hw->wiphy));
+
+		mt7996_irq_disable(dev, MT_INT_MCU_CMD);
+		queue_work(dev->mt76.wq, &dev->reset_work);
+		return;
+	}
+
+	queue_work(dev->mt76.wq, &dev->reset_work);
+	wake_up(&dev->reset_wait);
 }
 
 void mt7996_mac_update_stats(struct mt7996_phy *phy)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 33e76a97f46b..7ddfcf68ebd1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -22,17 +22,13 @@ static bool mt7996_dev_running(struct mt7996_dev *dev)
 	return phy && test_bit(MT76_STATE_RUNNING, &phy->mt76->state);
 }
 
-static int mt7996_start(struct ieee80211_hw *hw)
+int mt7996_run(struct ieee80211_hw *hw)
 {
 	struct mt7996_dev *dev = mt7996_hw_dev(hw);
 	struct mt7996_phy *phy = mt7996_hw_phy(hw);
 	bool running;
 	int ret;
 
-	flush_work(&dev->init_work);
-
-	mutex_lock(&dev->mt76.mutex);
-
 	running = mt7996_dev_running(dev);
 	if (!running) {
 		ret = mt7996_mcu_set_hdr_trans(dev, true);
@@ -59,6 +55,18 @@ static int mt7996_start(struct ieee80211_hw *hw)
 		mt7996_mac_reset_counters(phy);
 
 out:
+	return ret;
+}
+
+static int mt7996_start(struct ieee80211_hw *hw)
+{
+	struct mt7996_dev *dev = mt7996_hw_dev(hw);
+	int ret;
+
+	flush_work(&dev->init_work);
+
+	mutex_lock(&dev->mt76.mutex);
+	ret = mt7996_run(hw);
 	mutex_unlock(&dev->mt76.mutex);
 
 	return ret;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 016853033c1f..7a73f1a474fa 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -2425,17 +2425,10 @@ mt7996_mcu_init_rx_airtime(struct mt7996_dev *dev)
 				     MCU_WM_UNI_CMD(VOW), true);
 }
 
-int mt7996_mcu_init(struct mt7996_dev *dev)
+int mt7996_mcu_init_firmware(struct mt7996_dev *dev)
 {
-	static const struct mt76_mcu_ops mt7996_mcu_ops = {
-		.headroom = sizeof(struct mt76_connac2_mcu_txd), /* reuse */
-		.mcu_skb_send_msg = mt7996_mcu_send_message,
-		.mcu_parse_response = mt7996_mcu_parse_response,
-	};
 	int ret;
 
-	dev->mt76.mcu_ops = &mt7996_mcu_ops;
-
 	/* force firmware operation mode into normal state,
 	 * which should be set before firmware download stage.
 	 */
@@ -2476,6 +2469,19 @@ int mt7996_mcu_init(struct mt7996_dev *dev)
 				 MCU_WA_PARAM_RED, 0, 0);
 }
 
+int mt7996_mcu_init(struct mt7996_dev *dev)
+{
+	static const struct mt76_mcu_ops mt7996_mcu_ops = {
+		.headroom = sizeof(struct mt76_connac2_mcu_txd), /* reuse */
+		.mcu_skb_send_msg = mt7996_mcu_send_message,
+		.mcu_parse_response = mt7996_mcu_parse_response,
+	};
+
+	dev->mt76.mcu_ops = &mt7996_mcu_ops;
+
+	return mt7996_mcu_init_firmware(dev);
+}
+
 void mt7996_mcu_exit(struct mt7996_dev *dev)
 {
 	mt7996_mcu_restart(&dev->mt76);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
index fc77641ce7bf..8d6ab919d086 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
@@ -289,10 +289,9 @@ static void mt7996_irq_tasklet(struct tasklet_struct *t)
 		u32 val = mt76_rr(dev, MT_MCU_CMD);
 
 		mt76_wr(dev, MT_MCU_CMD, val);
-		if (val & MT_MCU_CMD_ERROR_MASK) {
-			dev->reset_state = val;
-			ieee80211_queue_work(mt76_hw(dev), &dev->reset_work);
-			wake_up(&dev->reset_wait);
+		if (val & (MT_MCU_CMD_ERROR_MASK | MT_MCU_CMD_WDT_MASK)) {
+			dev->recovery.state = val;
+			mt7996_reset(dev);
 		}
 	}
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index dc77fa4e351e..1a780167efd0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -236,7 +236,14 @@ struct mt7996_dev {
 	struct work_struct rc_work;
 	struct work_struct reset_work;
 	wait_queue_head_t reset_wait;
-	u32 reset_state;
+	struct {
+		u32 state;
+		u32 wa_reset_count;
+		u32 wm_reset_count;
+		bool hw_full_reset:1;
+		bool hw_init_done:1;
+		bool restart:1;
+	} recovery;
 
 	struct list_head sta_rc_list;
 	struct list_head sta_poll_list;
@@ -369,9 +376,16 @@ int mt7996_eeprom_get_target_power(struct mt7996_dev *dev,
 				   struct ieee80211_channel *chan);
 s8 mt7996_eeprom_get_power_delta(struct mt7996_dev *dev, int band);
 int mt7996_dma_init(struct mt7996_dev *dev);
+void mt7996_dma_reset(struct mt7996_dev *dev, bool force);
 void mt7996_dma_prefetch(struct mt7996_dev *dev);
 void mt7996_dma_cleanup(struct mt7996_dev *dev);
+void mt7996_init_txpower(struct mt7996_dev *dev,
+			 struct ieee80211_supported_band *sband);
+int mt7996_txbf_init(struct mt7996_dev *dev);
+void mt7996_reset(struct mt7996_dev *dev);
+int mt7996_run(struct ieee80211_hw *hw);
 int mt7996_mcu_init(struct mt7996_dev *dev);
+int mt7996_mcu_init_firmware(struct mt7996_dev *dev);
 int mt7996_mcu_twt_agrt_update(struct mt7996_dev *dev,
 			       struct mt7996_vif *mvif,
 			       struct mt7996_twt_flow *flow,
@@ -461,6 +475,7 @@ static inline void mt7996_irq_disable(struct mt7996_dev *dev, u32 mask)
 		mt76_set_irq_mask(&dev->mt76, MT_INT_MASK_CSR, mask, 0);
 }
 
+void mt7996_mac_init(struct mt7996_dev *dev);
 u32 mt7996_mac_wtbl_lmac_addr(struct mt7996_dev *dev, u16 wcid, u8 dw);
 bool mt7996_mac_wtbl_update(struct mt7996_dev *dev, int idx, u32 mask);
 void mt7996_mac_reset_counters(struct mt7996_phy *phy);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
index 7a28cae34e34..5ba74b49d65d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
@@ -317,6 +317,8 @@ enum base_rev {
 #define MT_WFDMA0_RX_INT_PCIE_SEL		MT_WFDMA0(0x154)
 #define MT_WFDMA0_RX_INT_SEL_RING3		BIT(3)
 
+#define MT_WFDMA0_MCU_HOST_INT_ENA		MT_WFDMA0(0x1f4)
+
 #define MT_WFDMA0_GLO_CFG			MT_WFDMA0(0x208)
 #define MT_WFDMA0_GLO_CFG_TX_DMA_EN		BIT(0)
 #define MT_WFDMA0_GLO_CFG_RX_DMA_EN		BIT(2)
@@ -444,6 +446,10 @@ enum base_rev {
 #define MT_MCU_CMD_NORMAL_STATE			BIT(5)
 #define MT_MCU_CMD_ERROR_MASK			GENMASK(5, 1)
 
+#define MT_MCU_CMD_WA_WDT			BIT(31)
+#define MT_MCU_CMD_WM_WDT			BIT(30)
+#define MT_MCU_CMD_WDT_MASK			GENMASK(31, 30)
+
 /* l1/l2 remap */
 #define MT_HIF_REMAP_L1				0x155024
 #define MT_HIF_REMAP_L1_MASK			GENMASK(31, 16)
@@ -468,7 +474,10 @@ enum base_rev {
 #define MT_INFRA_MCU_END			0x7c3fffff
 
 /* FW MODE SYNC */
-#define MT_SWDEF_MODE				0x9143c
+#define MT_SWDEF_BASE				0x00401400
+
+#define MT_SWDEF(ofs)				(MT_SWDEF_BASE + (ofs))
+#define MT_SWDEF_MODE				MT_SWDEF(0x3c)
 #define MT_SWDEF_NORMAL_MODE			0
 
 /* LED */
@@ -506,7 +515,7 @@ enum base_rev {
 #define MT_TOP_MISC_FW_STATE			GENMASK(2, 0)
 
 #define MT_HW_REV				0x70010204
-#define MT_WF_SUBSYS_RST			0x70002600
+#define MT_WF_SUBSYS_RST			0x70028600
 
 /* PCIE MAC */
 #define MT_PCIE_MAC_BASE			0x74030000
-- 
cgit v1.2.3-70-g09d2


From 672662f0742bf51584b871939703e3deb4beec27 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Wed, 5 Apr 2023 07:24:38 +0800
Subject: wifi: mt76: mt7996: add full system reset knobs into debugfs

Add testing points into debugfs to trigger firmware assert and enable
full system recovery. Also rename knob "fw_ser" to a clear-cut name
"sys_recovery".

Co-developed-by: Bo Jiao <Bo.Jiao@mediatek.com>
Signed-off-by: Bo Jiao <Bo.Jiao@mediatek.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt76_connac_mcu.h   |   1 +
 .../net/wireless/mediatek/mt76/mt7996/debugfs.c    | 149 ++++++++++++++++++---
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    |  16 +++
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.h    |  28 ++--
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |   1 +
 drivers/net/wireless/mediatek/mt76/mt7996/regs.h   |  13 ++
 6 files changed, 178 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index ac7d5817f345..08b031def3ef 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1221,6 +1221,7 @@ enum {
 	MCU_UNI_CMD_VOW = 0x37,
 	MCU_UNI_CMD_RRO = 0x57,
 	MCU_UNI_CMD_OFFCH_SCAN_CTRL = 0x58,
+	MCU_UNI_CMD_ASSERT_DUMP = 0x6f,
 };
 
 enum {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c
index 9c5e9ac1c335..513ab4ba41c9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c
@@ -48,12 +48,12 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_implicit_txbf, mt7996_implicit_txbf_get,
 
 /* test knob of system error recovery */
 static ssize_t
-mt7996_fw_ser_set(struct file *file, const char __user *user_buf,
-		  size_t count, loff_t *ppos)
+mt7996_sys_recovery_set(struct file *file, const char __user *user_buf,
+			size_t count, loff_t *ppos)
 {
 	struct mt7996_phy *phy = file->private_data;
 	struct mt7996_dev *dev = phy->dev;
-	u8 band_idx = phy->mt76->band_idx;
+	bool band = phy->mt76->band_idx;
 	char buf[16];
 	int ret = 0;
 	u16 val;
@@ -73,17 +73,47 @@ mt7996_fw_ser_set(struct file *file, const char __user *user_buf,
 		return -EINVAL;
 
 	switch (val) {
-	case SER_SET_RECOVER_L1:
-	case SER_SET_RECOVER_L2:
-	case SER_SET_RECOVER_L3_RX_ABORT:
-	case SER_SET_RECOVER_L3_TX_ABORT:
-	case SER_SET_RECOVER_L3_TX_DISABLE:
-	case SER_SET_RECOVER_L3_BF:
-		ret = mt7996_mcu_set_ser(dev, SER_ENABLE, BIT(val), band_idx);
+	/*
+	 * 0: grab firmware current SER state.
+	 * 1: trigger & enable system error L1 recovery.
+	 * 2: trigger & enable system error L2 recovery.
+	 * 3: trigger & enable system error L3 rx abort.
+	 * 4: trigger & enable system error L3 tx abort
+	 * 5: trigger & enable system error L3 tx disable.
+	 * 6: trigger & enable system error L3 bf recovery.
+	 * 7: trigger & enable system error L4 mdp recovery.
+	 * 8: trigger & enable system error full recovery.
+	 * 9: trigger firmware crash.
+	 */
+	case UNI_CMD_SER_QUERY:
+		ret = mt7996_mcu_set_ser(dev, UNI_CMD_SER_QUERY, 0, band);
+		break;
+	case UNI_CMD_SER_SET_RECOVER_L1:
+	case UNI_CMD_SER_SET_RECOVER_L2:
+	case UNI_CMD_SER_SET_RECOVER_L3_RX_ABORT:
+	case UNI_CMD_SER_SET_RECOVER_L3_TX_ABORT:
+	case UNI_CMD_SER_SET_RECOVER_L3_TX_DISABLE:
+	case UNI_CMD_SER_SET_RECOVER_L3_BF:
+	case UNI_CMD_SER_SET_RECOVER_L4_MDP:
+		ret = mt7996_mcu_set_ser(dev, UNI_CMD_SER_SET, BIT(val), band);
 		if (ret)
 			return ret;
 
-		ret = mt7996_mcu_set_ser(dev, SER_RECOVER, val, band_idx);
+		ret = mt7996_mcu_set_ser(dev, UNI_CMD_SER_TRIGGER, val, band);
+		break;
+
+	/* enable full chip reset */
+	case UNI_CMD_SER_SET_RECOVER_FULL:
+		mt76_set(dev, MT_WFDMA0_MCU_HOST_INT_ENA, MT_MCU_CMD_WDT_MASK);
+		dev->recovery.state |= MT_MCU_CMD_WDT_MASK;
+		mt7996_reset(dev);
+		break;
+
+	/* WARNING: trigger firmware crash */
+	case UNI_CMD_SER_SET_SYSTEM_ASSERT:
+		ret = mt7996_mcu_trigger_assert(dev);
+		if (ret)
+			return ret;
 		break;
 	default:
 		break;
@@ -92,9 +122,97 @@ mt7996_fw_ser_set(struct file *file, const char __user *user_buf,
 	return ret ? ret : count;
 }
 
-static const struct file_operations mt7996_fw_ser_ops = {
-	.write = mt7996_fw_ser_set,
-	/* TODO: ser read */
+static ssize_t
+mt7996_sys_recovery_get(struct file *file, char __user *user_buf,
+			size_t count, loff_t *ppos)
+{
+	struct mt7996_phy *phy = file->private_data;
+	struct mt7996_dev *dev = phy->dev;
+	char *buff;
+	int desc = 0;
+	ssize_t ret;
+	static const size_t bufsz = 1024;
+
+	buff = kmalloc(bufsz, GFP_KERNEL);
+	if (!buff)
+		return -ENOMEM;
+
+	/* HELP */
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "Please echo the correct value ...\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "0: grab firmware transient SER state\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "1: trigger system error L1 recovery\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "2: trigger system error L2 recovery\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "3: trigger system error L3 rx abort\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "4: trigger system error L3 tx abort\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "5: trigger system error L3 tx disable\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "6: trigger system error L3 bf recovery\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "7: trigger system error L4 mdp recovery\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "8: trigger system error full recovery\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "9: trigger firmware crash\n");
+
+	/* SER statistics */
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "\nlet's dump firmware SER statistics...\n");
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_STATUS        = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_SER_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_PLE_ERR       = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_PLE_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_PLE_ERR_1     = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_PLE1_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_PLE_ERR_AMSDU = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_PLE_AMSDU_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_PSE_ERR       = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_PSE_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_PSE_ERR_1     = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_PSE1_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_LMAC_WISR6_B0 = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_LAMC_WISR6_BN0_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_LMAC_WISR6_B1 = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_LAMC_WISR6_BN1_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_LMAC_WISR6_B2 = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_LAMC_WISR6_BN2_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_LMAC_WISR7_B0 = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_LAMC_WISR7_BN0_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_LMAC_WISR7_B1 = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_LAMC_WISR7_BN1_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "::E  R , SER_LMAC_WISR7_B2 = 0x%08x\n",
+			  mt76_rr(dev, MT_SWDEF_LAMC_WISR7_BN2_STATS));
+	desc += scnprintf(buff + desc, bufsz - desc,
+			  "\nSYS_RESET_COUNT: WM %d, WA %d\n",
+			  dev->recovery.wm_reset_count,
+			  dev->recovery.wa_reset_count);
+
+	ret = simple_read_from_buffer(user_buf, count, ppos, buff, desc);
+	kfree(buff);
+	return ret;
+}
+
+static const struct file_operations mt7996_sys_recovery_ops = {
+	.write = mt7996_sys_recovery_set,
+	.read = mt7996_sys_recovery_get,
 	.open = simple_open,
 	.llseek = default_llseek,
 };
@@ -674,6 +792,8 @@ int mt7996_init_debugfs(struct mt7996_phy *phy)
 	debugfs_create_file("xmit-queues", 0400, dir, phy,
 			    &mt7996_xmit_queues_fops);
 	debugfs_create_file("tx_stats", 0400, dir, phy, &mt7996_tx_stats_fops);
+	debugfs_create_file("sys_recovery", 0600, dir, phy,
+			    &mt7996_sys_recovery_ops);
 	debugfs_create_file("fw_debug_wm", 0600, dir, dev, &fops_fw_debug_wm);
 	debugfs_create_file("fw_debug_wa", 0600, dir, dev, &fops_fw_debug_wa);
 	debugfs_create_file("fw_debug_bin", 0600, dir, dev, &fops_fw_debug_bin);
@@ -684,7 +804,6 @@ int mt7996_init_debugfs(struct mt7996_phy *phy)
 			    &fops_implicit_txbf);
 	debugfs_create_devm_seqfile(dev->mt76.dev, "twt_stats", dir,
 				    mt7996_twt_stats);
-	debugfs_create_file("fw_ser", 0600, dir, phy, &mt7996_fw_ser_ops);
 	debugfs_create_file("rf_regval", 0600, dir, dev, &fops_rf_regval);
 
 	if (phy->mt76->cap.has_5ghz) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 7a73f1a474fa..c99605a1ee6b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -3615,6 +3615,22 @@ int mt7996_mcu_rf_regval(struct mt7996_dev *dev, u32 regidx, u32 *val, bool set)
 	return 0;
 }
 
+int mt7996_mcu_trigger_assert(struct mt7996_dev *dev)
+{
+	struct {
+		__le16 tag;
+		__le16 len;
+		u8 enable;
+		u8 rsv[3];
+	} __packed req = {
+		.len = cpu_to_le16(sizeof(req) - 4),
+		.enable = true,
+	};
+
+	return mt76_mcu_send_msg(&dev->mt76, MCU_WM_UNI_CMD(ASSERT_DUMP),
+				 &req, sizeof(req), false);
+}
+
 int mt7996_mcu_set_rro(struct mt7996_dev *dev, u16 tag, u8 val)
 {
 	struct {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h
index dd0c5ac52703..6f92b9a7d86d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h
@@ -648,23 +648,21 @@ enum {
 };
 
 enum {
-	UNI_CMD_SER_QUERY = 0x0,
-	UNI_CMD_SER_SET = 0x2,
-	UNI_CMD_SER_TRIGGER = 0x3,
-};
-
-enum {
-	SER_QUERY,
+	UNI_CMD_SER_QUERY,
 	/* recovery */
-	SER_SET_RECOVER_L1,
-	SER_SET_RECOVER_L2,
-	SER_SET_RECOVER_L3_RX_ABORT,
-	SER_SET_RECOVER_L3_TX_ABORT,
-	SER_SET_RECOVER_L3_TX_DISABLE,
-	SER_SET_RECOVER_L3_BF,
+	UNI_CMD_SER_SET_RECOVER_L1,
+	UNI_CMD_SER_SET_RECOVER_L2,
+	UNI_CMD_SER_SET_RECOVER_L3_RX_ABORT,
+	UNI_CMD_SER_SET_RECOVER_L3_TX_ABORT,
+	UNI_CMD_SER_SET_RECOVER_L3_TX_DISABLE,
+	UNI_CMD_SER_SET_RECOVER_L3_BF,
+	UNI_CMD_SER_SET_RECOVER_L4_MDP,
+	UNI_CMD_SER_SET_RECOVER_FULL,
+	UNI_CMD_SER_SET_SYSTEM_ASSERT,
 	/* action */
-	SER_ENABLE = 2,
-	SER_RECOVER
+	UNI_CMD_SER_ENABLE = 1,
+	UNI_CMD_SER_SET,
+	UNI_CMD_SER_TRIGGER
 };
 
 enum {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 1a780167efd0..bc399e8d0dd3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -441,6 +441,7 @@ int mt7996_mcu_set_rro(struct mt7996_dev *dev, u16 tag, u8 val);
 int mt7996_mcu_wa_cmd(struct mt7996_dev *dev, int cmd, u32 a1, u32 a2, u32 a3);
 int mt7996_mcu_fw_log_2_host(struct mt7996_dev *dev, u8 type, u8 ctrl);
 int mt7996_mcu_fw_dbg_ctrl(struct mt7996_dev *dev, u32 module, u8 level);
+int mt7996_mcu_trigger_assert(struct mt7996_dev *dev);
 void mt7996_mcu_rx_event(struct mt7996_dev *dev, struct sk_buff *skb);
 void mt7996_mcu_exit(struct mt7996_dev *dev);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
index 5ba74b49d65d..fb9b8dec9189 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
@@ -480,6 +480,19 @@ enum base_rev {
 #define MT_SWDEF_MODE				MT_SWDEF(0x3c)
 #define MT_SWDEF_NORMAL_MODE			0
 
+#define MT_SWDEF_SER_STATS			MT_SWDEF(0x040)
+#define MT_SWDEF_PLE_STATS			MT_SWDEF(0x044)
+#define MT_SWDEF_PLE1_STATS			MT_SWDEF(0x048)
+#define MT_SWDEF_PLE_AMSDU_STATS		MT_SWDEF(0x04c)
+#define MT_SWDEF_PSE_STATS			MT_SWDEF(0x050)
+#define MT_SWDEF_PSE1_STATS			MT_SWDEF(0x054)
+#define MT_SWDEF_LAMC_WISR6_BN0_STATS		MT_SWDEF(0x058)
+#define MT_SWDEF_LAMC_WISR6_BN1_STATS		MT_SWDEF(0x05c)
+#define MT_SWDEF_LAMC_WISR6_BN2_STATS		MT_SWDEF(0x060)
+#define MT_SWDEF_LAMC_WISR7_BN0_STATS		MT_SWDEF(0x064)
+#define MT_SWDEF_LAMC_WISR7_BN1_STATS		MT_SWDEF(0x068)
+#define MT_SWDEF_LAMC_WISR7_BN2_STATS		MT_SWDEF(0x06c)
+
 /* LED */
 #define MT_LED_TOP_BASE				0x18013000
 #define MT_LED_PHYS(_n)				(MT_LED_TOP_BASE + (_n))
-- 
cgit v1.2.3-70-g09d2


From 878161d5d4a469a6ef7f3fb4fe9f676bc508ee99 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Wed, 5 Apr 2023 07:24:39 +0800
Subject: wifi: mt76: mt7996: enable coredump support

Host triggered and catastrophic event triggered firmware core dumping
for basic firmware issues triage, including state reporting, function
calltrace and MCU memory dump.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/Kconfig  |   1 +
 drivers/net/wireless/mediatek/mt76/mt7996/Makefile |   2 +
 .../net/wireless/mediatek/mt76/mt7996/coredump.c   | 268 +++++++++++++++++++++
 .../net/wireless/mediatek/mt76/mt7996/coredump.h   |  97 ++++++++
 drivers/net/wireless/mediatek/mt76/mt7996/init.c   |  10 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    |  72 +++++-
 drivers/net/wireless/mediatek/mt76/mt7996/mmio.c   |   8 +
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  22 ++
 drivers/net/wireless/mediatek/mt76/mt7996/regs.h   |  18 ++
 9 files changed, 496 insertions(+), 2 deletions(-)
 create mode 100644 drivers/net/wireless/mediatek/mt76/mt7996/coredump.c
 create mode 100644 drivers/net/wireless/mediatek/mt76/mt7996/coredump.h

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig
index 79fb47a73c91..1afa2f662e47 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig
@@ -2,6 +2,7 @@
 config MT7996E
 	tristate "MediaTek MT7996 (PCIe) support"
 	select MT76_CONNAC_LIB
+	select WANT_DEV_COREDUMP
 	select RELAY
 	depends on MAC80211
 	depends on PCI
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Makefile b/drivers/net/wireless/mediatek/mt76/mt7996/Makefile
index bcb9a3c53149..07c8b555c1ac 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/Makefile
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/Makefile
@@ -4,3 +4,5 @@ obj-$(CONFIG_MT7996E) += mt7996e.o
 
 mt7996e-y := pci.o init.o dma.o eeprom.o main.o mcu.o mac.o \
 	     debugfs.o mmio.o
+
+mt7996e-$(CONFIG_DEV_COREDUMP) += coredump.o
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c
new file mode 100644
index 000000000000..ccab0d7b9be4
--- /dev/null
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c
@@ -0,0 +1,268 @@
+// SPDX-License-Identifier: ISC
+/* Copyright (C) 2023 MediaTek Inc. */
+
+#include <linux/devcoredump.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/utsname.h>
+#include "coredump.h"
+
+static bool coredump_memdump;
+module_param(coredump_memdump, bool, 0644);
+MODULE_PARM_DESC(coredump_memdump, "Optional ability to dump firmware memory");
+
+static const struct mt7996_mem_region mt7996_mem_regions[] = {
+	{
+		.start = 0x00800000,
+		.len = 0x0004ffff,
+		.name = "ULM0",
+	},
+	{
+		.start = 0x00900000,
+		.len = 0x00037fff,
+		.name = "ULM1",
+	},
+	{
+		.start = 0x02200000,
+		.len = 0x0003ffff,
+		.name = "ULM2",
+	},
+	{
+		.start = 0x00400000,
+		.len = 0x00067fff,
+		.name = "SRAM",
+	},
+	{
+		.start = 0xe0000000,
+		.len = 0x0015ffff,
+		.name = "CRAM0",
+	},
+	{
+		.start = 0xe0160000,
+		.len = 0x0011bfff,
+		.name = "CRAM1",
+	},
+};
+
+const struct mt7996_mem_region*
+mt7996_coredump_get_mem_layout(struct mt7996_dev *dev, u32 *num)
+{
+	switch (mt76_chip(&dev->mt76)) {
+	case 0x7990:
+	case 0x7991:
+		*num = ARRAY_SIZE(mt7996_mem_regions);
+		return &mt7996_mem_regions[0];
+	default:
+		return NULL;
+	}
+}
+
+static int mt7996_coredump_get_mem_size(struct mt7996_dev *dev)
+{
+	const struct mt7996_mem_region *mem_region;
+	size_t size = 0;
+	u32 num;
+	int i;
+
+	mem_region = mt7996_coredump_get_mem_layout(dev, &num);
+	if (!mem_region)
+		return 0;
+
+	for (i = 0; i < num; i++) {
+		size += mem_region->len;
+		mem_region++;
+	}
+
+	/* reserve space for the headers */
+	size += num * sizeof(struct mt7996_mem_hdr);
+	/* make sure it is aligned 4 bytes for debug message print out */
+	size = ALIGN(size, 4);
+
+	return size;
+}
+
+struct mt7996_crash_data *mt7996_coredump_new(struct mt7996_dev *dev)
+{
+	struct mt7996_crash_data *crash_data = dev->coredump.crash_data;
+
+	lockdep_assert_held(&dev->dump_mutex);
+
+	if (coredump_memdump &&
+	    !mt76_poll_msec(dev, MT_FW_DUMP_STATE, 0x3, 0x2, 500))
+		return NULL;
+
+	guid_gen(&crash_data->guid);
+	ktime_get_real_ts64(&crash_data->timestamp);
+
+	return crash_data;
+}
+
+static void
+mt7996_coredump_fw_state(struct mt7996_dev *dev, struct mt7996_coredump *dump,
+			 bool *exception)
+{
+	u32 count;
+
+	count = mt76_rr(dev, MT_FW_ASSERT_CNT);
+
+	/* normal mode: driver can manually trigger assert for detail info */
+	if (!count)
+		strscpy(dump->fw_state, "normal", sizeof(dump->fw_state));
+	else
+		strscpy(dump->fw_state, "exception", sizeof(dump->fw_state));
+
+	*exception = !!count;
+}
+
+static void
+mt7996_coredump_fw_stack(struct mt7996_dev *dev, struct mt7996_coredump *dump,
+			 bool exception)
+{
+	u32 oldest, i, idx;
+
+	strscpy(dump->pc_current, "program counter", sizeof(dump->pc_current));
+
+	/* 0: WM PC log output */
+	mt76_wr(dev, MT_CONN_DBG_CTL_OUT_SEL, 0);
+	/* choose 33th PC log buffer to read current PC index */
+	mt76_wr(dev, MT_CONN_DBG_CTL_PC_LOG_SEL, 0x3f);
+
+	/* read current PC */
+	dump->pc_stack[0] = mt76_rr(dev, MT_CONN_DBG_CTL_PC_LOG);
+
+	/* stop call stack record */
+	if (!exception) {
+		mt76_clear(dev, MT_MCU_WM_EXCP_PC_CTRL, BIT(0));
+		mt76_clear(dev, MT_MCU_WM_EXCP_LR_CTRL, BIT(0));
+	}
+
+	oldest = (u32)mt76_get_field(dev, MT_MCU_WM_EXCP_PC_CTRL,
+				     GENMASK(20, 16)) + 2;
+	for (i = 0; i < 16; i++) {
+		idx = ((oldest + 2 * i + 1) % 32);
+		dump->pc_stack[i + 1] =
+			mt76_rr(dev, MT_MCU_WM_EXCP_PC_LOG + idx * 4);
+	}
+
+	oldest = (u32)mt76_get_field(dev, MT_MCU_WM_EXCP_LR_CTRL,
+				     GENMASK(20, 16)) + 2;
+	for (i = 0; i < 16; i++) {
+		idx = ((oldest + 2 * i + 1) % 32);
+		dump->lr_stack[i] =
+			mt76_rr(dev, MT_MCU_WM_EXCP_LR_LOG + idx * 4);
+	}
+
+	/* start call stack record */
+	if (!exception) {
+		mt76_set(dev, MT_MCU_WM_EXCP_PC_CTRL, BIT(0));
+		mt76_set(dev, MT_MCU_WM_EXCP_LR_CTRL, BIT(0));
+	}
+}
+
+static struct mt7996_coredump *mt7996_coredump_build(struct mt7996_dev *dev)
+{
+	struct mt7996_crash_data *crash_data = dev->coredump.crash_data;
+	struct mt7996_coredump *dump;
+	struct mt7996_coredump_mem *dump_mem;
+	size_t len, sofar = 0, hdr_len = sizeof(*dump);
+	unsigned char *buf;
+	bool exception;
+
+	len = hdr_len;
+
+	if (coredump_memdump && crash_data->memdump_buf_len)
+		len += sizeof(*dump_mem) + crash_data->memdump_buf_len;
+
+	sofar += hdr_len;
+
+	/* this is going to get big when we start dumping memory and such,
+	 * so go ahead and use vmalloc.
+	 */
+	buf = vzalloc(len);
+	if (!buf)
+		return NULL;
+
+	mutex_lock(&dev->dump_mutex);
+
+	dump = (struct mt7996_coredump *)(buf);
+	dump->len = len;
+
+	/* plain text */
+	strscpy(dump->magic, "mt76-crash-dump", sizeof(dump->magic));
+	strscpy(dump->kernel, init_utsname()->release, sizeof(dump->kernel));
+	strscpy(dump->fw_ver, dev->mt76.hw->wiphy->fw_version,
+		sizeof(dump->fw_ver));
+
+	guid_copy(&dump->guid, &crash_data->guid);
+	dump->tv_sec = crash_data->timestamp.tv_sec;
+	dump->tv_nsec = crash_data->timestamp.tv_nsec;
+	dump->device_id = mt76_chip(&dev->mt76);
+
+	mt7996_coredump_fw_state(dev, dump, &exception);
+	mt7996_coredump_fw_stack(dev, dump, exception);
+
+	/* gather memory content */
+	dump_mem = (struct mt7996_coredump_mem *)(buf + sofar);
+	dump_mem->len = crash_data->memdump_buf_len;
+	if (coredump_memdump && crash_data->memdump_buf_len)
+		memcpy(dump_mem->data, crash_data->memdump_buf,
+		       crash_data->memdump_buf_len);
+
+	mutex_unlock(&dev->dump_mutex);
+
+	return dump;
+}
+
+int mt7996_coredump_submit(struct mt7996_dev *dev)
+{
+	struct mt7996_coredump *dump;
+
+	dump = mt7996_coredump_build(dev);
+	if (!dump) {
+		dev_warn(dev->mt76.dev, "no crash dump data found\n");
+		return -ENODATA;
+	}
+
+	dev_coredumpv(dev->mt76.dev, dump, dump->len, GFP_KERNEL);
+
+	return 0;
+}
+
+int mt7996_coredump_register(struct mt7996_dev *dev)
+{
+	struct mt7996_crash_data *crash_data;
+
+	crash_data = vzalloc(sizeof(*dev->coredump.crash_data));
+	if (!crash_data)
+		return -ENOMEM;
+
+	dev->coredump.crash_data = crash_data;
+
+	if (coredump_memdump) {
+		crash_data->memdump_buf_len = mt7996_coredump_get_mem_size(dev);
+		if (!crash_data->memdump_buf_len)
+			/* no memory content */
+			return 0;
+
+		crash_data->memdump_buf = vzalloc(crash_data->memdump_buf_len);
+		if (!crash_data->memdump_buf) {
+			vfree(crash_data);
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+void mt7996_coredump_unregister(struct mt7996_dev *dev)
+{
+	if (dev->coredump.crash_data->memdump_buf) {
+		vfree(dev->coredump.crash_data->memdump_buf);
+		dev->coredump.crash_data->memdump_buf = NULL;
+		dev->coredump.crash_data->memdump_buf_len = 0;
+	}
+
+	vfree(dev->coredump.crash_data);
+	dev->coredump.crash_data = NULL;
+}
+
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h
new file mode 100644
index 000000000000..af2ba219b1b5
--- /dev/null
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h
@@ -0,0 +1,97 @@
+/* SPDX-License-Identifier: ISC */
+/* Copyright (C) 2023 MediaTek Inc. */
+
+#ifndef _COREDUMP_H_
+#define _COREDUMP_H_
+
+#include "mt7996.h"
+
+struct mt7996_coredump {
+	char magic[16];
+
+	u32 len;
+
+	guid_t guid;
+
+	/* time-of-day stamp */
+	u64 tv_sec;
+	/* time-of-day stamp, nano-seconds */
+	u64 tv_nsec;
+	/* kernel version */
+	char kernel[64];
+	/* firmware version */
+	char fw_ver[ETHTOOL_FWVERS_LEN];
+
+	u32 device_id;
+
+	/* exception state */
+	char fw_state[12];
+
+	/* program counters */
+	char pc_current[16];
+	u32 pc_stack[17];
+	/* link registers */
+	u32 lr_stack[16];
+
+	/* memory content */
+	u8 data[];
+} __packed;
+
+struct mt7996_coredump_mem {
+	u32 len;
+	u8 data[];
+} __packed;
+
+struct mt7996_mem_hdr {
+	u32 start;
+	u32 len;
+	u8 data[];
+};
+
+struct mt7996_mem_region {
+	u32 start;
+	size_t len;
+
+	const char *name;
+};
+
+#ifdef CONFIG_DEV_COREDUMP
+
+const struct mt7996_mem_region *
+mt7996_coredump_get_mem_layout(struct mt7996_dev *dev, u32 *num);
+struct mt7996_crash_data *mt7996_coredump_new(struct mt7996_dev *dev);
+int mt7996_coredump_submit(struct mt7996_dev *dev);
+int mt7996_coredump_register(struct mt7996_dev *dev);
+void mt7996_coredump_unregister(struct mt7996_dev *dev);
+
+#else /* CONFIG_DEV_COREDUMP */
+
+static inline const struct mt7996_mem_region *
+mt7996_coredump_get_mem_layout(struct mt7996_dev *dev, u32 *num)
+{
+	return NULL;
+}
+
+static inline int mt7996_coredump_submit(struct mt7996_dev *dev)
+{
+	return 0;
+}
+
+static inline struct
+mt7996_crash_data *mt7996_coredump_new(struct mt7996_dev *dev)
+{
+	return NULL;
+}
+
+static inline int mt7996_coredump_register(struct mt7996_dev *dev)
+{
+	return 0;
+}
+
+static inline void mt7996_coredump_unregister(struct mt7996_dev *dev)
+{
+}
+
+#endif /* CONFIG_DEV_COREDUMP */
+
+#endif /* _COREDUMP_H_ */
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index 6de9a0405b74..19cdd0625c40 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -8,6 +8,7 @@
 #include "mt7996.h"
 #include "mac.h"
 #include "mcu.h"
+#include "coredump.h"
 #include "eeprom.h"
 
 static const struct ieee80211_iface_limit if_limits[] = {
@@ -841,6 +842,8 @@ int mt7996_register_device(struct mt7996_dev *dev)
 
 	init_waitqueue_head(&dev->reset_wait);
 	INIT_WORK(&dev->reset_work, mt7996_mac_reset_work);
+	INIT_WORK(&dev->dump_work, mt7996_mac_dump_work);
+	mutex_init(&dev->dump_mutex);
 
 	ret = mt7996_init_hardware(dev);
 	if (ret)
@@ -871,13 +874,18 @@ int mt7996_register_device(struct mt7996_dev *dev)
 
 	dev->recovery.hw_init_done = true;
 
-	return mt7996_init_debugfs(&dev->phy);
+	ret = mt7996_init_debugfs(&dev->phy);
+	if (ret)
+		return ret;
+
+	return mt7996_coredump_register(dev);
 }
 
 void mt7996_unregister_device(struct mt7996_dev *dev)
 {
 	mt7996_unregister_phy(mt7996_phy3(dev), MT_BAND2);
 	mt7996_unregister_phy(mt7996_phy2(dev), MT_BAND1);
+	mt7996_coredump_unregister(dev);
 	mt76_unregister_device(&dev->mt76);
 	mt7996_mcu_exit(dev);
 	mt7996_tx_token_put(dev);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index d526d6f53ace..1e9452c33e47 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -5,6 +5,7 @@
 
 #include <linux/etherdevice.h>
 #include <linux/timekeeping.h>
+#include "coredump.h"
 #include "mt7996.h"
 #include "../dma.h"
 #include "mac.h"
@@ -2081,6 +2082,75 @@ void mt7996_mac_reset_work(struct work_struct *work)
 		 wiphy_name(dev->mt76.hw->wiphy));
 }
 
+/* firmware coredump */
+void mt7996_mac_dump_work(struct work_struct *work)
+{
+	const struct mt7996_mem_region *mem_region;
+	struct mt7996_crash_data *crash_data;
+	struct mt7996_dev *dev;
+	struct mt7996_mem_hdr *hdr;
+	size_t buf_len;
+	int i;
+	u32 num;
+	u8 *buf;
+
+	dev = container_of(work, struct mt7996_dev, dump_work);
+
+	mutex_lock(&dev->dump_mutex);
+
+	crash_data = mt7996_coredump_new(dev);
+	if (!crash_data) {
+		mutex_unlock(&dev->dump_mutex);
+		goto skip_coredump;
+	}
+
+	mem_region = mt7996_coredump_get_mem_layout(dev, &num);
+	if (!mem_region || !crash_data->memdump_buf_len) {
+		mutex_unlock(&dev->dump_mutex);
+		goto skip_memdump;
+	}
+
+	buf = crash_data->memdump_buf;
+	buf_len = crash_data->memdump_buf_len;
+
+	/* dumping memory content... */
+	memset(buf, 0, buf_len);
+	for (i = 0; i < num; i++) {
+		if (mem_region->len > buf_len) {
+			dev_warn(dev->mt76.dev, "%s len %zu is too large\n",
+				 mem_region->name, mem_region->len);
+			break;
+		}
+
+		/* reserve space for the header */
+		hdr = (void *)buf;
+		buf += sizeof(*hdr);
+		buf_len -= sizeof(*hdr);
+
+		mt7996_memcpy_fromio(dev, buf, mem_region->start,
+				     mem_region->len);
+
+		hdr->start = mem_region->start;
+		hdr->len = mem_region->len;
+
+		if (!mem_region->len)
+			/* note: the header remains, just with zero length */
+			break;
+
+		buf += mem_region->len;
+		buf_len -= mem_region->len;
+
+		mem_region++;
+	}
+
+	mutex_unlock(&dev->dump_mutex);
+
+skip_memdump:
+	mt7996_coredump_submit(dev);
+skip_coredump:
+	queue_work(dev->mt76.wq, &dev->reset_work);
+}
+
 void mt7996_reset(struct mt7996_dev *dev)
 {
 	if (!dev->recovery.hw_init_done)
@@ -2097,7 +2167,7 @@ void mt7996_reset(struct mt7996_dev *dev)
 			 wiphy_name(dev->mt76.hw->wiphy));
 
 		mt7996_irq_disable(dev, MT_INT_MCU_CMD);
-		queue_work(dev->mt76.wq, &dev->reset_work);
+		queue_work(dev->mt76.wq, &dev->dump_work);
 		return;
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
index 8d6ab919d086..879b0e90b52e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
@@ -162,6 +162,14 @@ static u32 __mt7996_reg_addr(struct mt7996_dev *dev, u32 addr)
 	return mt7996_reg_map_l2(dev, addr);
 }
 
+void mt7996_memcpy_fromio(struct mt7996_dev *dev, void *buf, u32 offset,
+			  size_t len)
+{
+	u32 addr = __mt7996_reg_addr(dev, offset);
+
+	memcpy_fromio(buf, dev->mt76.mmio.regs + addr, len);
+}
+
 static u32 mt7996_rr(struct mt76_dev *mdev, u32 offset)
 {
 	struct mt7996_dev *dev = container_of(mdev, struct mt7996_dev, mt76);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index bc399e8d0dd3..4001cf12b671 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -176,6 +176,15 @@ struct mib_stats {
 	u32 tx_amsdu_cnt;
 };
 
+/* crash-dump */
+struct mt7996_crash_data {
+	guid_t guid;
+	struct timespec64 timestamp;
+
+	u8 *memdump_buf;
+	size_t memdump_buf_len;
+};
+
 struct mt7996_hif {
 	struct list_head list;
 
@@ -234,6 +243,7 @@ struct mt7996_dev {
 
 	struct work_struct init_work;
 	struct work_struct rc_work;
+	struct work_struct dump_work;
 	struct work_struct reset_work;
 	wait_queue_head_t reset_wait;
 	struct {
@@ -245,6 +255,14 @@ struct mt7996_dev {
 		bool restart:1;
 	} recovery;
 
+	/* protects coredump data */
+	struct mutex dump_mutex;
+#ifdef CONFIG_DEV_COREDUMP
+	struct {
+		struct mt7996_crash_data *crash_data;
+	} coredump;
+#endif
+
 	struct list_head sta_rc_list;
 	struct list_head sta_poll_list;
 	struct list_head twt_list;
@@ -476,6 +494,9 @@ static inline void mt7996_irq_disable(struct mt7996_dev *dev, u32 mask)
 		mt76_set_irq_mask(&dev->mt76, MT_INT_MASK_CSR, mask, 0);
 }
 
+void mt7996_memcpy_fromio(struct mt7996_dev *dev, void *buf, u32 offset,
+			  size_t len);
+
 void mt7996_mac_init(struct mt7996_dev *dev);
 u32 mt7996_mac_wtbl_lmac_addr(struct mt7996_dev *dev, u16 wcid, u8 dw);
 bool mt7996_mac_wtbl_update(struct mt7996_dev *dev, int idx, u32 mask);
@@ -495,6 +516,7 @@ void mt7996_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta);
 void mt7996_mac_work(struct work_struct *work);
 void mt7996_mac_reset_work(struct work_struct *work);
+void mt7996_mac_dump_work(struct work_struct *work);
 void mt7996_mac_sta_rc_work(struct work_struct *work);
 void mt7996_mac_update_stats(struct mt7996_phy *phy);
 void mt7996_mac_twt_teardown_flow(struct mt7996_dev *dev,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
index fb9b8dec9189..75f856205559 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
@@ -474,6 +474,9 @@ enum base_rev {
 #define MT_INFRA_MCU_END			0x7c3fffff
 
 /* FW MODE SYNC */
+#define MT_FW_ASSERT_CNT			0x02208274
+#define MT_FW_DUMP_STATE			0x02209e90
+
 #define MT_SWDEF_BASE				0x00401400
 
 #define MT_SWDEF(ofs)				(MT_SWDEF_BASE + (ofs))
@@ -508,6 +511,13 @@ enum base_rev {
 
 #define MT_LED_EN(_n)				MT_LED_PHYS(0x40 + ((_n) * 4))
 
+/* CONN DBG */
+#define MT_CONN_DBG_CTL_BASE			0x18023000
+#define MT_CONN_DBG_CTL(ofs)			(MT_CONN_DBG_CTL_BASE + (ofs))
+#define MT_CONN_DBG_CTL_OUT_SEL			MT_CONN_DBG_CTL(0x604)
+#define MT_CONN_DBG_CTL_PC_LOG_SEL		MT_CONN_DBG_CTL(0x60c)
+#define MT_CONN_DBG_CTL_PC_LOG			MT_CONN_DBG_CTL(0x610)
+
 #define MT_LED_GPIO_MUX2			0x70005058 /* GPIO 18 */
 #define MT_LED_GPIO_MUX3			0x7000505C /* GPIO 26 */
 #define MT_LED_GPIO_SEL_MASK			GENMASK(11, 8)
@@ -561,4 +571,12 @@ enum base_rev {
 #define MT_WF_PHYRX_CSD_BAND_RXTD12_IRPI_SW_CLR_ONLY	BIT(18)
 #define MT_WF_PHYRX_CSD_BAND_RXTD12_IRPI_SW_CLR		BIT(29)
 
+/* CONN MCU EXCP CON */
+#define MT_MCU_WM_EXCP_BASE			0x89050000
+#define MT_MCU_WM_EXCP(ofs)			(MT_MCU_WM_EXCP_BASE + (ofs))
+#define MT_MCU_WM_EXCP_PC_CTRL			MT_MCU_WM_EXCP(0x100)
+#define MT_MCU_WM_EXCP_PC_LOG			MT_MCU_WM_EXCP(0x104)
+#define MT_MCU_WM_EXCP_LR_CTRL			MT_MCU_WM_EXCP(0x200)
+#define MT_MCU_WM_EXCP_LR_LOG			MT_MCU_WM_EXCP(0x204)
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 61d1f54533496711e06fcfd42b93c5ded9e27c7a Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Wed, 5 Apr 2023 12:38:53 +0200
Subject: wifi: mt76: move mcu_uni_event and mcu_reg_event in common code

mcu_uni_event and mcu_reg_event structs are shared between mt7921 and
mt7615 drivers, so move them in connac lib.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/mcu.c      |  8 ++++----
 drivers/net/wireless/mediatek/mt76/mt7615/mcu.h      | 11 -----------
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h | 11 +++++++++++
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c      |  8 ++++----
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.h      | 11 -----------
 5 files changed, 19 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
index 195fe1094d9b..8d745c9730c7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
@@ -163,16 +163,16 @@ int mt7615_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 		   cmd == MCU_UNI_CMD(HIF_CTRL) ||
 		   cmd == MCU_UNI_CMD(OFFLOAD) ||
 		   cmd == MCU_UNI_CMD(SUSPEND)) {
-		struct mt7615_mcu_uni_event *event;
+		struct mt76_connac_mcu_uni_event *event;
 
 		skb_pull(skb, sizeof(*rxd));
-		event = (struct mt7615_mcu_uni_event *)skb->data;
+		event = (struct mt76_connac_mcu_uni_event *)skb->data;
 		ret = le32_to_cpu(event->status);
 	} else if (cmd == MCU_CE_QUERY(REG_READ)) {
-		struct mt7615_mcu_reg_event *event;
+		struct mt76_connac_mcu_reg_event *event;
 
 		skb_pull(skb, sizeof(*rxd));
-		event = (struct mt7615_mcu_reg_event *)skb->data;
+		event = (struct mt76_connac_mcu_reg_event *)skb->data;
 		ret = (int)le32_to_cpu(event->val);
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h
index 615956acc6b5..8e9604be0792 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h
@@ -206,17 +206,6 @@ enum {
 	MCU_ATE_SET_TX_POWER_CONTROL = 0x15,
 };
 
-struct mt7615_mcu_uni_event {
-	u8 cid;
-	u8 pad[3];
-	__le32 status; /* 0: success, others: fail */
-} __packed;
-
-struct mt7615_mcu_reg_event {
-	__le32 reg;
-	__le32 val;
-} __packed;
-
 struct mt7615_roc_tlv {
 	u8 bss_idx;
 	u8 token;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 08b031def3ef..ca1ce97a6d2f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1690,6 +1690,17 @@ struct mt76_connac_config {
 	u8 data[320];
 } __packed;
 
+struct mt76_connac_mcu_uni_event {
+	u8 cid;
+	u8 pad[3];
+	__le32 status; /* 0: success, others: fail */
+} __packed;
+
+struct mt76_connac_mcu_reg_event {
+	__le32 reg;
+	__le32 val;
+} __packed;
+
 static inline enum mcu_cipher_type
 mt76_connac_mcu_get_cipher(int cipher)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
index 04925e66f15b..259b6ea88be4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
@@ -67,19 +67,19 @@ int mt7921_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 		   cmd == MCU_UNI_CMD(HIF_CTRL) ||
 		   cmd == MCU_UNI_CMD(OFFLOAD) ||
 		   cmd == MCU_UNI_CMD(SUSPEND)) {
-		struct mt7921_mcu_uni_event *event;
+		struct mt76_connac_mcu_uni_event *event;
 
 		skb_pull(skb, sizeof(*rxd));
-		event = (struct mt7921_mcu_uni_event *)skb->data;
+		event = (struct mt76_connac_mcu_uni_event *)skb->data;
 		ret = le32_to_cpu(event->status);
 		/* skip invalid event */
 		if (mcu_cmd != event->cid)
 			ret = -EAGAIN;
 	} else if (cmd == MCU_CE_QUERY(REG_READ)) {
-		struct mt7921_mcu_reg_event *event;
+		struct mt76_connac_mcu_reg_event *event;
 
 		skb_pull(skb, sizeof(*rxd));
-		event = (struct mt7921_mcu_reg_event *)skb->data;
+		event = (struct mt76_connac_mcu_reg_event *)skb->data;
 		ret = (int)le32_to_cpu(event->val);
 	} else {
 		skb_pull(skb, sizeof(struct mt76_connac2_mcu_rxd));
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h
index 96dc870fd35e..9b0aa3b70f0e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h
@@ -50,22 +50,11 @@ struct mt7921_mcu_eeprom_info {
 #define MT_RA_RATE_DCM_EN		BIT(4)
 #define MT_RA_RATE_BW			GENMASK(14, 13)
 
-struct mt7921_mcu_uni_event {
-	u8 cid;
-	u8 pad[3];
-	__le32 status; /* 0: success, others: fail */
-} __packed;
-
 enum {
 	MT_EBF = BIT(0),	/* explicit beamforming */
 	MT_IBF = BIT(1)		/* implicit beamforming */
 };
 
-struct mt7921_mcu_reg_event {
-	__le32 reg;
-	__le32 val;
-} __packed;
-
 struct mt7921_mcu_ant_id_config {
 	u8 ant_id[4];
 } __packed;
-- 
cgit v1.2.3-70-g09d2


From 12db28c3ef31f719bd18fa186a40bb152e6a527c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Apr 2023 05:23:11 +0800
Subject: mt76: mt7921: fix kernel panic by accessing unallocated eeprom.data

The MT7921 driver no longer uses eeprom.data, but the relevant code has not
been removed completely since
commit 16d98b548365 ("mt76: mt7921: rely on mcu_get_nic_capability").
This could result in potential invalid memory access.

To fix the kernel panic issue in mt7921, it is necessary to avoid accessing
unallocated eeprom.data which can lead to invalid memory access.

Furthermore, it is possible to entirely eliminate the
mt7921_mcu_parse_eeprom function and solely depend on
mt7921_mcu_parse_response to divide the RxD header.

[2.702735] BUG: kernel NULL pointer dereference, address: 0000000000000550
[2.702740] #PF: supervisor write access in kernel mode
[2.702741] #PF: error_code(0x0002) - not-present page
[2.702743] PGD 0 P4D 0
[2.702747] Oops: 0002 [#1] PREEMPT SMP NOPTI
[2.702755] RIP: 0010:mt7921_mcu_parse_response+0x147/0x170 [mt7921_common]
[2.702758] RSP: 0018:ffffae7c00fef828 EFLAGS: 00010286
[2.702760] RAX: ffffa367f57be024 RBX: ffffa367cc7bf500 RCX: 0000000000000000
[2.702762] RDX: 0000000000000550 RSI: 0000000000000000 RDI: ffffa367cc7bf500
[2.702763] RBP: ffffae7c00fef840 R08: ffffa367cb167000 R09: 0000000000000005
[2.702764] R10: 0000000000000000 R11: ffffffffc04702e4 R12: ffffa367e8329f40
[2.702766] R13: 0000000000000000 R14: 0000000000000001 R15: ffffa367e8329f40
[2.702768] FS:  000079ee6cf20c40(0000) GS:ffffa36b2f940000(0000) knlGS:0000000000000000
[2.702769] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[2.702775] CR2: 0000000000000550 CR3: 00000001233c6004 CR4: 0000000000770ee0
[2.702776] PKRU: 55555554
[2.702777] Call Trace:
[2.702782]  mt76_mcu_skb_send_and_get_msg+0xc3/0x11e [mt76 <HASH:1bc4 5>]
[2.702785]  mt7921_run_firmware+0x241/0x853 [mt7921_common <HASH:6a2f 6>]
[2.702789]  mt7921e_mcu_init+0x2b/0x56 [mt7921e <HASH:d290 7>]
[2.702792]  mt7921_register_device+0x2eb/0x5a5 [mt7921_common <HASH:6a2f 6>]
[2.702795]  ? mt7921_irq_tasklet+0x1d4/0x1d4 [mt7921e <HASH:d290 7>]
[2.702797]  mt7921_pci_probe+0x2d6/0x319 [mt7921e <HASH:d290 7>]
[2.702799]  pci_device_probe+0x9f/0x12a

Fixes: 16d98b548365 ("mt76: mt7921: rely on mcu_get_nic_capability")
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 20 --------------------
 1 file changed, 20 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
index 259b6ea88be4..c69ce6df4956 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
@@ -15,24 +15,6 @@ static bool mt7921_disable_clc;
 module_param_named(disable_clc, mt7921_disable_clc, bool, 0644);
 MODULE_PARM_DESC(disable_clc, "disable CLC support");
 
-static int
-mt7921_mcu_parse_eeprom(struct mt76_dev *dev, struct sk_buff *skb)
-{
-	struct mt7921_mcu_eeprom_info *res;
-	u8 *buf;
-
-	if (!skb)
-		return -EINVAL;
-
-	skb_pull(skb, sizeof(struct mt76_connac2_mcu_rxd));
-
-	res = (struct mt7921_mcu_eeprom_info *)skb->data;
-	buf = dev->eeprom.data + le32_to_cpu(res->addr);
-	memcpy(buf, res->data, 16);
-
-	return 0;
-}
-
 int mt7921_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 			      struct sk_buff *skb, int seq)
 {
@@ -59,8 +41,6 @@ int mt7921_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 	} else if (cmd == MCU_EXT_CMD(THERMAL_CTRL)) {
 		skb_pull(skb, sizeof(*rxd) + 4);
 		ret = le32_to_cpu(*(__le32 *)skb->data);
-	} else if (cmd == MCU_EXT_CMD(EFUSE_ACCESS)) {
-		ret = mt7921_mcu_parse_eeprom(mdev, skb);
 	} else if (cmd == MCU_UNI_CMD(DEV_INFO_UPDATE) ||
 		   cmd == MCU_UNI_CMD(BSS_INFO_UPDATE) ||
 		   cmd == MCU_UNI_CMD(STA_REC_UPDATE) ||
-- 
cgit v1.2.3-70-g09d2


From 5c47cdebbaeb7724df6f9f46917c93e53f791547 Mon Sep 17 00:00:00 2001
From: Jiefeng Li <jiefeng_li@hust.edu.cn>
Date: Wed, 12 Apr 2023 14:22:34 +0800
Subject: wifi: mt76: mt7921: fix missing unwind goto in `mt7921u_probe`

`mt7921u_dma_init` can only return zero or negative number according to its
definition. When it returns non-zero number, there exists an error and this
function should handle this error rather than return directly.

Fixes: 0d2afe09fad5 ("mt76: mt7921: add mt7921u driver")
Signed-off-by: Jiefeng Li <jiefeng_li@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/usb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
index 9b05e4577cdf..1f302c430339 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c
@@ -259,7 +259,7 @@ static int mt7921u_probe(struct usb_interface *usb_intf,
 
 	ret = mt7921u_dma_init(dev, false);
 	if (ret)
-		return ret;
+		goto error;
 
 	hw = mt76_hw(dev);
 	/* check hw sg support in order to enable AMSDU */
-- 
cgit v1.2.3-70-g09d2


From 87714bf6ed1589813e473db5471e6e9857755764 Mon Sep 17 00:00:00 2001
From: Quan Zhou <quan.zhou@mediatek.com>
Date: Thu, 13 Apr 2023 05:11:13 +0800
Subject: wifi: mt76: mt7921e: improve reliability of dma reset

The hardware team has advised the driver that it is necessary to first put
WFDMA into an idle state before resetting the WFDMA. Otherwise, the WFDMA
may enter an unknown state where it cannot be polled with the right state
successfully. To ensure that the DMA can work properly while a stressful
cold reboot test was being made, we have reordered the programming sequence
in the driver based on the hardware team's guidance.

The patch would modify the WFDMA disabling flow from

"DMA reset -> disabling DMASHDL -> disabling WFDMA -> polling and waiting
until DMA idle" to "disabling WFDMA -> polling and waiting for DMA idle ->
disabling DMASHDL -> DMA reset.

Where he polling and waiting until WFDMA is idle is coordinated with the
operation of disabling WFDMA. Even while WFDMA is being disabled, it can
still handle Tx/Rx requests. The additional polling allows sufficient time
for WFDMA to process the last T/Rx request. When the idle state of WFDMA is
reached, it is a reliable indication that DMASHDL is also idle to ensure it
is safe to disable it and perform the DMA reset.

Fixes: 0a1059d0f060 ("mt76: mt7921: move mt7921_dma_reset in dma.c")
Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Co-developed-by: Wang Zhao <wang.zhao@mediatek.com>
Signed-off-by: Wang Zhao <wang.zhao@mediatek.com>
Signed-off-by: Quan Zhou <quan.zhou@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/dma.c | 36 ++++++++++++++-----------
 1 file changed, 20 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
index 28207172c4ef..f0a80c2b476a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/dma.c
@@ -66,22 +66,6 @@ static void mt7921_dma_prefetch(struct mt7921_dev *dev)
 
 static int mt7921_dma_disable(struct mt7921_dev *dev, bool force)
 {
-	if (force) {
-		/* reset */
-		mt76_clear(dev, MT_WFDMA0_RST,
-			   MT_WFDMA0_RST_DMASHDL_ALL_RST |
-			   MT_WFDMA0_RST_LOGIC_RST);
-
-		mt76_set(dev, MT_WFDMA0_RST,
-			 MT_WFDMA0_RST_DMASHDL_ALL_RST |
-			 MT_WFDMA0_RST_LOGIC_RST);
-	}
-
-	/* disable dmashdl */
-	mt76_clear(dev, MT_WFDMA0_GLO_CFG_EXT0,
-		   MT_WFDMA0_CSR_TX_DMASHDL_ENABLE);
-	mt76_set(dev, MT_DMASHDL_SW_CONTROL, MT_DMASHDL_DMASHDL_BYPASS);
-
 	/* disable WFDMA0 */
 	mt76_clear(dev, MT_WFDMA0_GLO_CFG,
 		   MT_WFDMA0_GLO_CFG_TX_DMA_EN | MT_WFDMA0_GLO_CFG_RX_DMA_EN |
@@ -95,6 +79,22 @@ static int mt7921_dma_disable(struct mt7921_dev *dev, bool force)
 				 MT_WFDMA0_GLO_CFG_RX_DMA_BUSY, 0, 100, 1))
 		return -ETIMEDOUT;
 
+	/* disable dmashdl */
+	mt76_clear(dev, MT_WFDMA0_GLO_CFG_EXT0,
+		   MT_WFDMA0_CSR_TX_DMASHDL_ENABLE);
+	mt76_set(dev, MT_DMASHDL_SW_CONTROL, MT_DMASHDL_DMASHDL_BYPASS);
+
+	if (force) {
+		/* reset */
+		mt76_clear(dev, MT_WFDMA0_RST,
+			   MT_WFDMA0_RST_DMASHDL_ALL_RST |
+			   MT_WFDMA0_RST_LOGIC_RST);
+
+		mt76_set(dev, MT_WFDMA0_RST,
+			 MT_WFDMA0_RST_DMASHDL_ALL_RST |
+			 MT_WFDMA0_RST_LOGIC_RST);
+	}
+
 	return 0;
 }
 
@@ -301,6 +301,10 @@ void mt7921_dma_cleanup(struct mt7921_dev *dev)
 		   MT_WFDMA0_GLO_CFG_OMIT_RX_INFO |
 		   MT_WFDMA0_GLO_CFG_OMIT_RX_INFO_PFET2);
 
+	mt76_poll_msec_tick(dev, MT_WFDMA0_GLO_CFG,
+			    MT_WFDMA0_GLO_CFG_TX_DMA_BUSY |
+			    MT_WFDMA0_GLO_CFG_RX_DMA_BUSY, 0, 100, 1);
+
 	/* reset */
 	mt76_clear(dev, MT_WFDMA0_RST,
 		   MT_WFDMA0_RST_DMASHDL_ALL_RST |
-- 
cgit v1.2.3-70-g09d2


From 3d78c46423c6567ed25ca033e086865b1b4d5ae1 Mon Sep 17 00:00:00 2001
From: Quan Zhou <quan.zhou@mediatek.com>
Date: Thu, 13 Apr 2023 05:11:14 +0800
Subject: wifi: mt76: mt7921e: stop chip reset worker in unregister hook

If the chip reset worker is triggered during the remove process, the chip
DMA may not be properly pushed back to the idle state. This can lead to
corruption of the DMA flow due to the chip reset. Therefore, it is
necessary to stop the chip reset before the DMA is finalized.

To avoid resetting the chip after the reset worker is cancelled, use
__mt7921_mcu_drv_pmctrl() instead of mt7921_mcu_drv_pmctrl(). It is safe to
ignore the pm mutex because the pm worker and wake worker have already been
cancelled.

Fixes: 033ae79b3830 ("mt76: mt7921: refactor init.c to be bus independent")
Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Co-developed-by: Wang Zhao <wang.zhao@mediatek.com>
Signed-off-by: Wang Zhao <wang.zhao@mediatek.com>
Signed-off-by: Quan Zhou <quan.zhou@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 40bea17cc440..1c727870bbdb 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -113,9 +113,10 @@ static void mt7921e_unregister_device(struct mt7921_dev *dev)
 		napi_disable(&dev->mt76.napi[i]);
 	cancel_delayed_work_sync(&pm->ps_work);
 	cancel_work_sync(&pm->wake_work);
+	cancel_work_sync(&dev->reset_work);
 
 	mt7921_tx_token_put(dev);
-	mt7921_mcu_drv_pmctrl(dev);
+	__mt7921_mcu_drv_pmctrl(dev);
 	mt7921_dma_cleanup(dev);
 	mt7921_wfsys_reset(dev);
 	skb_queue_purge(&dev->mt76.mcu.res_q);
-- 
cgit v1.2.3-70-g09d2


From 3d2892e05086d09aecf14ea64b2debbf495e313c Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Fri, 14 Apr 2023 04:23:29 +0800
Subject: wifi: mt76: connac: fix txd multicast rate setting

The vif->bss_conf.mcast_rate should be applied to multicast data frame
only.

Fixes: 182071cdd594 ("mt76: connac: move connac2_mac_write_txwi in mt76_connac module")
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
index bd4ce2c7742d..a51a2d4ff2f8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
@@ -558,7 +558,8 @@ void mt76_connac2_mac_write_txwi(struct mt76_dev *dev, __le32 *txwi,
 	if (txwi[2] & cpu_to_le32(MT_TXD2_FIX_RATE)) {
 		/* Fixed rata is available just for 802.11 txd */
 		struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
-		bool multicast = is_multicast_ether_addr(hdr->addr1);
+		bool multicast = ieee80211_is_data(hdr->frame_control) &&
+				 is_multicast_ether_addr(hdr->addr1);
 		u16 rate = mt76_connac2_mac_tx_rate_val(mphy, vif, beacon,
 							multicast);
 		u32 val = MT_TXD6_FIXED_BW;
-- 
cgit v1.2.3-70-g09d2


From 9c54548bc9e46217c551ed69fd2f14f4cb5155eb Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Fri, 14 Apr 2023 04:23:30 +0800
Subject: wifi: mt76: connac: add nss calculation into
 mt76_connac2_mac_tx_rate_val()

Take nss calculation into account since this function always wrongly
returns 0.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt76_connac_mac.c   | 43 ++++++++++++++--------
 1 file changed, 27 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
index a51a2d4ff2f8..ee0fbfcd07d6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c
@@ -288,11 +288,29 @@ int mt76_connac_init_tx_queues(struct mt76_phy *phy, int idx, int n_desc,
 }
 EXPORT_SYMBOL_GPL(mt76_connac_init_tx_queues);
 
+#define __bitrate_mask_check(_mcs, _mode)				\
+({									\
+	u8 i = 0;							\
+	for (nss = 0; i < ARRAY_SIZE(mask->control[band]._mcs); i++) {	\
+		if (!mask->control[band]._mcs[i])			\
+			continue;					\
+		if (hweight16(mask->control[band]._mcs[i]) == 1) {	\
+			mode = MT_PHY_TYPE_##_mode;			\
+			rateidx = ffs(mask->control[band]._mcs[i]) - 1;	\
+			if (mode == MT_PHY_TYPE_HT)			\
+				rateidx += 8 * i;			\
+			else						\
+				nss = i + 1;				\
+			goto out;					\
+		}							\
+	}								\
+})
+
 u16 mt76_connac2_mac_tx_rate_val(struct mt76_phy *mphy,
 				 struct ieee80211_vif *vif,
 				 bool beacon, bool mcast)
 {
-	u8 mode = 0, band = mphy->chandef.chan->band;
+	u8 nss = 0, mode = 0, band = mphy->chandef.chan->band;
 	int rateidx = 0, mcast_rate;
 
 	if (!vif)
@@ -307,19 +325,12 @@ u16 mt76_connac2_mac_tx_rate_val(struct mt76_phy *mphy,
 		struct cfg80211_bitrate_mask *mask;
 
 		mask = &vif->bss_conf.beacon_tx_rate;
-		if (hweight16(mask->control[band].he_mcs[0]) == 1) {
-			rateidx = ffs(mask->control[band].he_mcs[0]) - 1;
-			mode = MT_PHY_TYPE_HE_SU;
-			goto out;
-		} else if (hweight16(mask->control[band].vht_mcs[0]) == 1) {
-			rateidx = ffs(mask->control[band].vht_mcs[0]) - 1;
-			mode = MT_PHY_TYPE_VHT;
-			goto out;
-		} else if (hweight8(mask->control[band].ht_mcs[0]) == 1) {
-			rateidx = ffs(mask->control[band].ht_mcs[0]) - 1;
-			mode = MT_PHY_TYPE_HT;
-			goto out;
-		} else if (hweight32(mask->control[band].legacy) == 1) {
+
+		__bitrate_mask_check(he_mcs, HE_SU);
+		__bitrate_mask_check(vht_mcs, VHT);
+		__bitrate_mask_check(ht_mcs, HT);
+
+		if (hweight32(mask->control[band].legacy) == 1) {
 			rateidx = ffs(mask->control[band].legacy) - 1;
 			goto legacy;
 		}
@@ -335,9 +346,9 @@ legacy:
 	rateidx = mt76_calculate_default_rate(mphy, rateidx);
 	mode = rateidx >> 8;
 	rateidx &= GENMASK(7, 0);
-
 out:
-	return FIELD_PREP(MT_TX_RATE_IDX, rateidx) |
+	return FIELD_PREP(MT_TX_RATE_NSS, nss) |
+	       FIELD_PREP(MT_TX_RATE_IDX, rateidx) |
 	       FIELD_PREP(MT_TX_RATE_MODE, mode);
 }
 EXPORT_SYMBOL_GPL(mt76_connac2_mac_tx_rate_val);
-- 
cgit v1.2.3-70-g09d2


From 59e498a3289f685261c076b998a8a2f8a516874f Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Mon, 17 Apr 2023 15:49:15 +0200
Subject: bpf: Set skb redirect and from_ingress info in __bpf_tx_skb
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There are some use-cases where it is desirable to use bpf_redirect()
in combination with ifb device, which currently is not supported, for
example, around filtering inbound traffic with BPF to then push it to
ifb which holds the qdisc for shaping in contrast to doing that on the
egress device.

Toke mentions the following case related to OpenWrt:

   Because there's not always a single egress on the other side. These are
   mainly home routers, which tend to have one or more WiFi devices bridged
   to one or more ethernet ports on the LAN side, and a single upstream WAN
   port. And the objective is to control the total amount of traffic going
   over the WAN link (in both directions), to deal with bufferbloat in the
   ISP network (which is sadly still all too prevalent).

   In this setup, the traffic can be split arbitrarily between the links
   on the LAN side, and the only "single bottleneck" is the WAN link. So we
   install both egress and ingress shapers on this, configured to something
   like 95-98% of the true link bandwidth, thus moving the queues into the
   qdisc layer in the router. It's usually necessary to set the ingress
   bandwidth shaper a bit lower than the egress due to being "downstream"
   of the bottleneck link, but it does work surprisingly well.

   We usually use something like a matchall filter to put all ingress
   traffic on the ifb, so doing the redirect from BPF has not been an
   immediate requirement thus far. However, it does seem a bit odd that
   this is not possible, and we do have a BPF-based filter that layers on
   top of this kind of setup, which currently uses u32 as the ingress
   filter and so it could presumably be improved to use BPF instead if
   that was available.

Reported-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reported-by: Yafang Shao <laoar.shao@gmail.com>
Reported-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://git.openwrt.org/?p=project/qosify.git;a=blob;f=README
Link: https://lore.kernel.org/bpf/875y9yzbuy.fsf@toke.dk
Link: https://lore.kernel.org/r/8cebc8b2b6e967e10cbafe2ffd6795050e74accd.1681739137.git.daniel@iogearbox.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/skbuff.h | 9 +++++++++
 net/core/filter.c      | 1 +
 2 files changed, 10 insertions(+)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 494a23a976b0..9ff2e3d57329 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -5041,6 +5041,15 @@ static inline void skb_reset_redirect(struct sk_buff *skb)
 	skb->redirected = 0;
 }
 
+static inline void skb_set_redirected_noclear(struct sk_buff *skb,
+					      bool from_ingress)
+{
+	skb->redirected = 1;
+#ifdef CONFIG_NET_REDIRECT
+	skb->from_ingress = from_ingress;
+#endif
+}
+
 static inline bool skb_csum_is_sctp(struct sk_buff *skb)
 {
 	return skb->csum_not_inet;
diff --git a/net/core/filter.c b/net/core/filter.c
index df0df59814ae..44fb997434ad 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2122,6 +2122,7 @@ static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb)
 	}
 
 	skb->dev = dev;
+	skb_set_redirected_noclear(skb, skb_at_tc_ingress(skb));
 	skb_clear_tstamp(skb);
 
 	dev_xmit_recursion_inc();
-- 
cgit v1.2.3-70-g09d2


From 69a8c792cd9518071dc801bb110e0f2210d9f958 Mon Sep 17 00:00:00 2001
From: Sean Young <sean@mess.org>
Date: Mon, 17 Apr 2023 09:17:48 +0100
Subject: bpf: lirc program type should not require SYS_CAP_ADMIN

Make it possible to load lirc program type with just CAP_BPF. There is
nothing exceptional about lirc programs that means they require
SYS_CAP_ADMIN.

In order to attach or detach a lirc program type you need permission to
open /dev/lirc0; if you have permission to do that, you can alter all
sorts of lirc receiving options. Changing the IR protocol decoder is no
different.

Right now on a typical distribution /dev/lirc devices are only
read/write by root. Ideally we would make them group read/write like
other devices so that local users can use them without becoming root.

Signed-off-by: Sean Young <sean@mess.org>
Link: https://lore.kernel.org/r/ZD0ArKpwnDBJZsrE@gofer.mess.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/syscall.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 28eac7434d32..bcf1a1920ddd 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2454,7 +2454,6 @@ static bool is_net_admin_prog_type(enum bpf_prog_type prog_type)
 	case BPF_PROG_TYPE_LWT_SEG6LOCAL:
 	case BPF_PROG_TYPE_SK_SKB:
 	case BPF_PROG_TYPE_SK_MSG:
-	case BPF_PROG_TYPE_LIRC_MODE2:
 	case BPF_PROG_TYPE_FLOW_DISSECTOR:
 	case BPF_PROG_TYPE_CGROUP_DEVICE:
 	case BPF_PROG_TYPE_CGROUP_SOCK:
-- 
cgit v1.2.3-70-g09d2


From 3be49f79555ee975acb4ad8b5de06fa4351264aa Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Mon, 17 Apr 2023 15:21:34 -0700
Subject: bpf: Improve verifier u32 scalar equality checking

In [1], I tried to remove bpf-specific codes to prevent certain
llvm optimizations, and add llvm TTI (target transform info) hooks
to prevent those optimizations. During this process, I found
if I enable llvm SimplifyCFG:shouldFoldTwoEntryPHINode
transformation, I will hit the following verification failure with selftests:

  ...
  8: (18) r1 = 0xffffc900001b2230       ; R1_w=map_value(off=560,ks=4,vs=564,imm=0)
  10: (61) r1 = *(u32 *)(r1 +0)         ; R1_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
  ; if (skb->tstamp == EGRESS_ENDHOST_MAGIC)
  11: (79) r2 = *(u64 *)(r6 +152)       ; R2_w=scalar() R6=ctx(off=0,imm=0)
  ; if (skb->tstamp == EGRESS_ENDHOST_MAGIC)
  12: (55) if r2 != 0xb9fbeef goto pc+10        ; R2_w=195018479
  13: (bc) w2 = w1                      ; R1_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff)) R2_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
  ; if (test < __NR_TESTS)
  14: (a6) if w1 < 0x9 goto pc+1 16: R0=2 R1_w=scalar(umax=8,var_off=(0x0; 0xf)) R2_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff)) R6=ctx(off=0,imm=0) R10=fp0
  ;
  16: (27) r2 *= 28                     ; R2_w=scalar(umax=120259084260,var_off=(0x0; 0x1ffffffffc),s32_max=2147483644,u32_max=-4)
  17: (18) r3 = 0xffffc900001b2118      ; R3_w=map_value(off=280,ks=4,vs=564,imm=0)
  19: (0f) r3 += r2                     ; R2_w=scalar(umax=120259084260,var_off=(0x0; 0x1ffffffffc),s32_max=2147483644,u32_max=-4) R3_w=map_value(off=280,ks=4,vs=564,umax=120259084260,var_off=(0x0; 0x1ffffffffc),s32_max=2147483644,u32_max=-4)
  20: (61) r2 = *(u32 *)(r3 +0)
  R3 unbounded memory access, make sure to bounds check any such access
  processed 97 insns (limit 1000000) max_states_per_insn 1 total_states 10 peak_states 10 mark_read 6
  -- END PROG LOAD LOG --
  libbpf: prog 'ingress_fwdns_prio100': failed to load: -13
  libbpf: failed to load object 'test_tc_dtime'
  libbpf: failed to load BPF skeleton 'test_tc_dtime': -13
  ...

At insn 14, with condition 'w1 < 9', register r1 is changed from an arbitrary
u32 value to `scalar(umax=8,var_off=(0x0; 0xf))`. Register r2, however, remains
as an arbitrary u32 value. Current verifier won't claim r1/r2 equality if
the previous mov is alu32 ('w2 = w1').

If r1 upper 32bit value is not 0, we indeed cannot clamin r1/r2 equality
after 'w2 = w1'. But in this particular case, we know r1 upper 32bit value
is 0, so it is safe to claim r1/r2 equality. This patch exactly did this.
For a 32bit subreg mov, if the src register upper 32bit is 0,
it is okay to claim equality between src and dst registers.

With this patch, the above verification sequence becomes

  ...
  8: (18) r1 = 0xffffc9000048e230       ; R1_w=map_value(off=560,ks=4,vs=564,imm=0)
  10: (61) r1 = *(u32 *)(r1 +0)         ; R1_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
  ; if (skb->tstamp == EGRESS_ENDHOST_MAGIC)
  11: (79) r2 = *(u64 *)(r6 +152)       ; R2_w=scalar() R6=ctx(off=0,imm=0)
  ; if (skb->tstamp == EGRESS_ENDHOST_MAGIC)
  12: (55) if r2 != 0xb9fbeef goto pc+10        ; R2_w=195018479
  13: (bc) w2 = w1                      ; R1_w=scalar(id=6,umax=4294967295,var_off=(0x0; 0xffffffff)) R2_w=scalar(id=6,umax=4294967295,var_off=(0x0; 0xffffffff))
  ; if (test < __NR_TESTS)
  14: (a6) if w1 < 0x9 goto pc+1        ; R1_w=scalar(id=6,umin=9,umax=4294967295,var_off=(0x0; 0xffffffff))
  ...
  from 14 to 16: R0=2 R1_w=scalar(id=6,umax=8,var_off=(0x0; 0xf)) R2_w=scalar(id=6,umax=8,var_off=(0x0; 0xf)) R6=ctx(off=0,imm=0) R10=fp0
  16: (27) r2 *= 28                     ; R2_w=scalar(umax=224,var_off=(0x0; 0xfc))
  17: (18) r3 = 0xffffc9000048e118      ; R3_w=map_value(off=280,ks=4,vs=564,imm=0)
  19: (0f) r3 += r2
  20: (61) r2 = *(u32 *)(r3 +0)         ; R2_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff)) R3_w=map_value(off=280,ks=4,vs=564,umax=224,var_off=(0x0; 0xfc),s32_max=252,u32_max=252)
  ...

and eventually the bpf program can be verified successfully.

  [1] https://reviews.llvm.org/D147968

Signed-off-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20230417222134.359714-1-yhs@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5dae11ee01c3..1e05355facdc 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -12421,12 +12421,17 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn)
 						insn->src_reg);
 					return -EACCES;
 				} else if (src_reg->type == SCALAR_VALUE) {
+					bool is_src_reg_u32 = src_reg->umax_value <= U32_MAX;
+
+					if (is_src_reg_u32 && !src_reg->id)
+						src_reg->id = ++env->id_gen;
 					copy_register_state(dst_reg, src_reg);
-					/* Make sure ID is cleared otherwise
+					/* Make sure ID is cleared if src_reg is not in u32 range otherwise
 					 * dst_reg min/max could be incorrectly
 					 * propagated into src_reg by find_equal_scalars()
 					 */
-					dst_reg->id = 0;
+					if (!is_src_reg_u32)
+						dst_reg->id = 0;
 					dst_reg->live |= REG_LIVE_WRITTEN;
 					dst_reg->subreg_def = env->insn_idx + 1;
 				} else {
-- 
cgit v1.2.3-70-g09d2


From 49859de997c3115b85544bce6b6ceab60a7fabc4 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yhs@fb.com>
Date: Mon, 17 Apr 2023 15:21:39 -0700
Subject: selftests/bpf: Add a selftest for checking subreg equality

Add a selftest to ensure subreg equality if source register
upper 32bit is 0. Without previous patch, the test will
fail verification.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20230417222139.360607-1-yhs@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../selftests/bpf/progs/verifier_reg_equal.c       | 58 ++++++++++++++++++++++
 2 files changed, 60 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_reg_equal.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 73dff693d411..25bc8958dbfe 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -31,6 +31,7 @@
 #include "verifier_meta_access.skel.h"
 #include "verifier_raw_stack.skel.h"
 #include "verifier_raw_tp_writable.skel.h"
+#include "verifier_reg_equal.skel.h"
 #include "verifier_ringbuf.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
@@ -95,6 +96,7 @@ void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
+void test_verifier_reg_equal(void)            { RUN(verifier_reg_equal); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_reg_equal.c b/tools/testing/selftests/bpf/progs/verifier_reg_equal.c
new file mode 100644
index 000000000000..dc1d8c30fb0e
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_reg_equal.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("socket")
+__description("check w reg equal if r reg upper32 bits 0")
+__success
+__naked void subreg_equality_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	*(u64 *)(r10 - 8) = r0;				\
+	r2 = *(u32 *)(r10 - 8);				\
+	/* At this point upper 4-bytes of r2 are 0,	\
+	 * thus insn w3 = w2 should propagate reg id,	\
+	 * and w2 < 9 comparison would also propagate	\
+	 * the range for r3.				\
+	 */						\
+	w3 = w2;					\
+	if w2 < 9 goto l0_%=;				\
+	exit;						\
+l0_%=:	if r3 < 9 goto l1_%=;				\
+	/* r1 read is illegal at this point */		\
+	r0 -= r1;					\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check w reg not equal if r reg upper32 bits not 0")
+__failure __msg("R1 !read_ok")
+__naked void subreg_equality_2(void)
+{
+	asm volatile ("					\
+	call %[bpf_ktime_get_ns];			\
+	r2 = r0;					\
+	/* Upper 4-bytes of r2 may not be 0, thus insn	\
+	 * w3 = w2 should not propagate reg id,	and	\
+	 * w2 < 9 comparison should not propagate	\
+	 * the range for r3 either.			\
+	 */						\
+	w3 = w2;					\
+	if w2 < 9 goto l0_%=;				\
+	exit;						\
+l0_%=:	if r3 < 9 goto l1_%=;				\
+	/* r1 read is illegal at this point */		\
+	r0 -= r1;					\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From 1210af3b99561bbe140af8d9a6b16d1bd0ba3fda Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:19 +0300
Subject: net/mlx5e: Add IPsec packet offload tunnel bits

Extend packet reformat types and flow table capabilities with
IPsec packet offload tunnel bits.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/mlx5/mlx5_ifc.h | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 6c84bf6eec85..20d00e09b168 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -463,9 +463,11 @@ struct mlx5_ifc_flow_table_prop_layout_bits {
 	u8         max_ft_level[0x8];
 
 	u8         reformat_add_esp_trasport[0x1];
-	u8         reserved_at_41[0x2];
+	u8         reformat_l2_to_l3_esp_tunnel[0x1];
+	u8         reserved_at_42[0x1];
 	u8         reformat_del_esp_trasport[0x1];
-	u8         reserved_at_44[0x2];
+	u8         reformat_l3_esp_tunnel_to_l2[0x1];
+	u8         reserved_at_45[0x1];
 	u8         execute_aso[0x1];
 	u8         reserved_at_47[0x19];
 
@@ -6630,7 +6632,9 @@ enum mlx5_reformat_ctx_type {
 	MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2 = 0x3,
 	MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL = 0x4,
 	MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV4 = 0x5,
+	MLX5_REFORMAT_TYPE_L2_TO_L3_ESP_TUNNEL = 0x6,
 	MLX5_REFORMAT_TYPE_DEL_ESP_TRANSPORT = 0x8,
+	MLX5_REFORMAT_TYPE_L3_ESP_TUNNEL_TO_L2 = 0x9,
 	MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV6 = 0xb,
 	MLX5_REFORMAT_TYPE_INSERT_HDR = 0xf,
 	MLX5_REFORMAT_TYPE_REMOVE_HDR = 0x10,
-- 
cgit v1.2.3-70-g09d2


From 1c80e949292f2ec490828b538cb9d490edf8f224 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:20 +0300
Subject: net/mlx5e: Check IPsec packet offload tunnel capabilities

Validate tunnel mode support for IPsec packet offload.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h         | 1 +
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c | 6 ++++++
 2 files changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 52890d7dce6b..bb89e18b17b4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -107,6 +107,7 @@ enum mlx5_ipsec_cap {
 	MLX5_IPSEC_CAP_PACKET_OFFLOAD	= 1 << 2,
 	MLX5_IPSEC_CAP_ROCE             = 1 << 3,
 	MLX5_IPSEC_CAP_PRIO             = 1 << 4,
+	MLX5_IPSEC_CAP_TUNNEL           = 1 << 5,
 };
 
 struct mlx5e_priv;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
index 5fddb86bb35e..df90e19066bc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c
@@ -48,6 +48,12 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev)
 		if (MLX5_CAP_FLOWTABLE_NIC_TX(mdev, ignore_flow_level) &&
 		    MLX5_CAP_FLOWTABLE_NIC_RX(mdev, ignore_flow_level))
 			caps |= MLX5_IPSEC_CAP_PRIO;
+
+		if (MLX5_CAP_FLOWTABLE_NIC_TX(mdev,
+					      reformat_l2_to_l3_esp_tunnel) &&
+		    MLX5_CAP_FLOWTABLE_NIC_RX(mdev,
+					      reformat_l3_esp_tunnel_to_l2))
+			caps |= MLX5_IPSEC_CAP_TUNNEL;
 	}
 
 	if (mlx5_get_roce_state(mdev) &&
-- 
cgit v1.2.3-70-g09d2


From 006adbc6de9f6290a542949af8f834c17e851890 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:21 +0300
Subject: net/mlx5e: Configure IPsec SA tables to support tunnel mode

Create SA flow steering tables both for RX and TX with tunnel reformat
property. This allows to add and delete extra headers needed for tunnel
mode.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 23 ++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index b47794d4146e..060be020ca64 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -118,7 +118,7 @@ static void ipsec_chains_put_table(struct mlx5_fs_chains *chains, u32 prio)
 
 static struct mlx5_flow_table *ipsec_ft_create(struct mlx5_flow_namespace *ns,
 					       int level, int prio,
-					       int max_num_groups)
+					       int max_num_groups, u32 flags)
 {
 	struct mlx5_flow_table_attr ft_attr = {};
 
@@ -127,6 +127,7 @@ static struct mlx5_flow_table *ipsec_ft_create(struct mlx5_flow_namespace *ns,
 	ft_attr.max_fte = NUM_IPSEC_FTE;
 	ft_attr.level = level;
 	ft_attr.prio = prio;
+	ft_attr.flags = flags;
 
 	return mlx5_create_auto_grouped_flow_table(ns, &ft_attr);
 }
@@ -267,6 +268,7 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	struct mlx5_flow_destination default_dest;
 	struct mlx5_flow_destination dest[2];
 	struct mlx5_flow_table *ft;
+	u32 flags = 0;
 	int err;
 
 	default_dest = mlx5_ttc_get_default_dest(ttc, family2tt(family));
@@ -277,7 +279,7 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 		return err;
 
 	ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_ESP_FT_ERR_LEVEL,
-			     MLX5E_NIC_PRIO, 1);
+			     MLX5E_NIC_PRIO, 1, 0);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_fs_ft_status;
@@ -300,8 +302,10 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 		goto err_add;
 
 	/* Create FT */
-	ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_ESP_FT_LEVEL, MLX5E_NIC_PRIO,
-			     2);
+	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_TUNNEL)
+		flags = MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT;
+	ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_ESP_FT_LEVEL, MLX5E_NIC_PRIO, 2,
+			     flags);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_fs_ft;
@@ -327,7 +331,7 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	}
 
 	ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_POL_FT_LEVEL, MLX5E_NIC_PRIO,
-			     2);
+			     2, 0);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_pol_ft;
@@ -511,9 +515,10 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 {
 	struct mlx5_flow_destination dest = {};
 	struct mlx5_flow_table *ft;
+	u32 flags = 0;
 	int err;
 
-	ft = ipsec_ft_create(tx->ns, 2, 0, 1);
+	ft = ipsec_ft_create(tx->ns, 2, 0, 1, 0);
 	if (IS_ERR(ft))
 		return PTR_ERR(ft);
 	tx->ft.status = ft;
@@ -522,7 +527,9 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 	if (err)
 		goto err_status_rule;
 
-	ft = ipsec_ft_create(tx->ns, 1, 0, 4);
+	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_TUNNEL)
+		flags = MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT;
+	ft = ipsec_ft_create(tx->ns, 1, 0, 4, flags);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_sa_ft;
@@ -541,7 +548,7 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 		goto connect_roce;
 	}
 
-	ft = ipsec_ft_create(tx->ns, 0, 0, 2);
+	ft = ipsec_ft_create(tx->ns, 0, 0, 2, 0);
 	if (IS_ERR(ft)) {
 		err = PTR_ERR(ft);
 		goto err_pol_ft;
-- 
cgit v1.2.3-70-g09d2


From 6480a3b6c90a8d2b765f802a4f30829bd539599a Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:22 +0300
Subject: net/mlx5e: Prepare IPsec packet reformat code for tunnel mode

Refactor setup_pkt_reformat() function to accommodate future extension
to support tunnel mode.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   |  1 +
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  2 +-
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 81 ++++++++++++++++------
 3 files changed, 63 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index def01bfde610..359da277c03a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -297,6 +297,7 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	attrs->upspec.sport = ntohs(x->sel.sport);
 	attrs->upspec.sport_mask = ntohs(x->sel.sport_mask);
 	attrs->upspec.proto = x->sel.proto;
+	attrs->mode = x->props.mode;
 
 	mlx5e_ipsec_init_limits(sa_entry, attrs);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index bb89e18b17b4..ae525420a492 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -77,7 +77,7 @@ struct mlx5_replay_esn {
 
 struct mlx5_accel_esp_xfrm_attrs {
 	u32   spi;
-	u32   flags;
+	u32   mode;
 	struct aes_gcm_keymat aes_gcm;
 
 	union {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 060be020ca64..6a1ed4114054 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -10,6 +10,7 @@
 #include "lib/fs_chains.h"
 
 #define NUM_IPSEC_FTE BIT(15)
+#define MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_SIZE 16
 
 struct mlx5e_ipsec_fc {
 	struct mlx5_fc *cnt;
@@ -836,40 +837,80 @@ static int setup_modify_header(struct mlx5_core_dev *mdev, u32 val, u8 dir,
 	return 0;
 }
 
+static int
+setup_pkt_transport_reformat(struct mlx5_accel_esp_xfrm_attrs *attrs,
+			     struct mlx5_pkt_reformat_params *reformat_params)
+{
+	u8 *reformatbf;
+	__be32 spi;
+
+	switch (attrs->dir) {
+	case XFRM_DEV_OFFLOAD_IN:
+		reformat_params->type = MLX5_REFORMAT_TYPE_DEL_ESP_TRANSPORT;
+		break;
+	case XFRM_DEV_OFFLOAD_OUT:
+		if (attrs->family == AF_INET)
+			reformat_params->type =
+				MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV4;
+		else
+			reformat_params->type =
+				MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV6;
+
+		reformatbf = kzalloc(MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_SIZE,
+				     GFP_KERNEL);
+		if (!reformatbf)
+			return -ENOMEM;
+
+		/* convert to network format */
+		spi = htonl(attrs->spi);
+		memcpy(reformatbf, &spi, sizeof(spi));
+
+		reformat_params->param_0 = attrs->authsize;
+		reformat_params->size =
+			MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_SIZE;
+		reformat_params->data = reformatbf;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int setup_pkt_reformat(struct mlx5_core_dev *mdev,
 			      struct mlx5_accel_esp_xfrm_attrs *attrs,
 			      struct mlx5_flow_act *flow_act)
 {
-	enum mlx5_flow_namespace_type ns_type = MLX5_FLOW_NAMESPACE_EGRESS;
 	struct mlx5_pkt_reformat_params reformat_params = {};
 	struct mlx5_pkt_reformat *pkt_reformat;
-	u8 reformatbf[16] = {};
-	__be32 spi;
+	enum mlx5_flow_namespace_type ns_type;
+	int ret;
 
-	if (attrs->dir == XFRM_DEV_OFFLOAD_IN) {
-		reformat_params.type = MLX5_REFORMAT_TYPE_DEL_ESP_TRANSPORT;
+	switch (attrs->dir) {
+	case XFRM_DEV_OFFLOAD_IN:
 		ns_type = MLX5_FLOW_NAMESPACE_KERNEL;
-		goto cmd;
+		break;
+	case XFRM_DEV_OFFLOAD_OUT:
+		ns_type = MLX5_FLOW_NAMESPACE_EGRESS;
+		break;
+	default:
+		return -EINVAL;
 	}
 
-	if (attrs->family == AF_INET)
-		reformat_params.type =
-			MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV4;
-	else
-		reformat_params.type =
-			MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV6;
-
-	/* convert to network format */
-	spi = htonl(attrs->spi);
-	memcpy(reformatbf, &spi, 4);
+	switch (attrs->mode) {
+	case XFRM_MODE_TRANSPORT:
+		ret = setup_pkt_transport_reformat(attrs, &reformat_params);
+		break;
+	default:
+		ret = -EINVAL;
+	}
 
-	reformat_params.param_0 = attrs->authsize;
-	reformat_params.size = sizeof(reformatbf);
-	reformat_params.data = &reformatbf;
+	if (ret)
+		return ret;
 
-cmd:
 	pkt_reformat =
 		mlx5_packet_reformat_alloc(mdev, &reformat_params, ns_type);
+	kfree(reformat_params.data);
 	if (IS_ERR(pkt_reformat))
 		return PTR_ERR(pkt_reformat);
 
-- 
cgit v1.2.3-70-g09d2


From 37a417ca911a52828aa84340484e4ccd33b7baae Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:23 +0300
Subject: net/mlx5e: Support IPsec RX packet offload in tunnel mode

Extend mlx5 driver with logic to support IPsec RX packet offload
in tunnel mode.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 36 ++++++++++++++++
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  2 +
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 50 ++++++++++++++++++++++
 3 files changed, 88 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 359da277c03a..7c55b37c1c01 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -242,6 +242,41 @@ static void mlx5e_ipsec_init_limits(struct mlx5e_ipsec_sa_entry *sa_entry,
 	attrs->lft.numb_rounds_soft = (u64)n;
 }
 
+static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
+				  struct mlx5_accel_esp_xfrm_attrs *attrs)
+{
+	struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry);
+	struct xfrm_state *x = sa_entry->x;
+	struct net_device *netdev;
+	struct neighbour *n;
+	u8 addr[ETH_ALEN];
+
+	if (attrs->mode != XFRM_MODE_TUNNEL &&
+	    attrs->type != XFRM_DEV_OFFLOAD_PACKET)
+		return;
+
+	netdev = x->xso.real_dev;
+
+	mlx5_query_mac_address(mdev, addr);
+	switch (attrs->dir) {
+	case XFRM_DEV_OFFLOAD_IN:
+		ether_addr_copy(attrs->dmac, addr);
+		n = neigh_lookup(&arp_tbl, &attrs->saddr.a4, netdev);
+		if (!n) {
+			n = neigh_create(&arp_tbl, &attrs->saddr.a4, netdev);
+			if (IS_ERR(n))
+				return;
+			neigh_event_send(n, NULL);
+		}
+		neigh_ha_snapshot(addr, n, netdev);
+		ether_addr_copy(attrs->smac, addr);
+		break;
+	default:
+		return;
+	}
+	neigh_release(n);
+}
+
 void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 					struct mlx5_accel_esp_xfrm_attrs *attrs)
 {
@@ -300,6 +335,7 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	attrs->mode = x->props.mode;
 
 	mlx5e_ipsec_init_limits(sa_entry, attrs);
+	mlx5e_ipsec_init_macs(sa_entry, attrs);
 }
 
 static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index ae525420a492..77384ffa4451 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -99,6 +99,8 @@ struct mlx5_accel_esp_xfrm_attrs {
 	u32 authsize;
 	u32 reqid;
 	struct mlx5_ipsec_lft lft;
+	u8 smac[ETH_ALEN];
+	u8 dmac[ETH_ALEN];
 };
 
 enum mlx5_ipsec_cap {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 6a1ed4114054..001d7c3add6a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -837,6 +837,53 @@ static int setup_modify_header(struct mlx5_core_dev *mdev, u32 val, u8 dir,
 	return 0;
 }
 
+static int
+setup_pkt_tunnel_reformat(struct mlx5_core_dev *mdev,
+			  struct mlx5_accel_esp_xfrm_attrs *attrs,
+			  struct mlx5_pkt_reformat_params *reformat_params)
+{
+	struct ethhdr *eth_hdr;
+	char *reformatbf;
+	size_t bfflen;
+
+	bfflen = sizeof(*eth_hdr);
+
+	reformatbf = kzalloc(bfflen, GFP_KERNEL);
+	if (!reformatbf)
+		return -ENOMEM;
+
+	eth_hdr = (struct ethhdr *)reformatbf;
+	switch (attrs->family) {
+	case AF_INET:
+		eth_hdr->h_proto = htons(ETH_P_IP);
+		break;
+	case AF_INET6:
+		eth_hdr->h_proto = htons(ETH_P_IPV6);
+		break;
+	default:
+		goto free_reformatbf;
+	}
+
+	ether_addr_copy(eth_hdr->h_dest, attrs->dmac);
+	ether_addr_copy(eth_hdr->h_source, attrs->smac);
+
+	switch (attrs->dir) {
+	case XFRM_DEV_OFFLOAD_IN:
+		reformat_params->type = MLX5_REFORMAT_TYPE_L3_ESP_TUNNEL_TO_L2;
+		break;
+	default:
+		goto free_reformatbf;
+	}
+
+	reformat_params->size = bfflen;
+	reformat_params->data = reformatbf;
+	return 0;
+
+free_reformatbf:
+	kfree(reformatbf);
+	return -EINVAL;
+}
+
 static int
 setup_pkt_transport_reformat(struct mlx5_accel_esp_xfrm_attrs *attrs,
 			     struct mlx5_pkt_reformat_params *reformat_params)
@@ -901,6 +948,9 @@ static int setup_pkt_reformat(struct mlx5_core_dev *mdev,
 	case XFRM_MODE_TRANSPORT:
 		ret = setup_pkt_transport_reformat(attrs, &reformat_params);
 		break;
+	case XFRM_MODE_TUNNEL:
+		ret = setup_pkt_tunnel_reformat(mdev, attrs, &reformat_params);
+		break;
 	default:
 		ret = -EINVAL;
 	}
-- 
cgit v1.2.3-70-g09d2


From efbd31c4d844e55526587cc3820ca8a11d7db733 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:24 +0300
Subject: net/mlx5e: Support IPsec TX packet offload in tunnel mode

Extend mlx5 driver with logic to support IPsec TX packet offload
in tunnel mode.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 12 +++++
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 52 ++++++++++++++++++++++
 2 files changed, 64 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 7c55b37c1c01..36f3ffd54355 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -271,6 +271,18 @@ static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
 		neigh_ha_snapshot(addr, n, netdev);
 		ether_addr_copy(attrs->smac, addr);
 		break;
+	case XFRM_DEV_OFFLOAD_OUT:
+		ether_addr_copy(attrs->smac, addr);
+		n = neigh_lookup(&arp_tbl, &attrs->daddr.a4, netdev);
+		if (!n) {
+			n = neigh_create(&arp_tbl, &attrs->daddr.a4, netdev);
+			if (IS_ERR(n))
+				return;
+			neigh_event_send(n, NULL);
+		}
+		neigh_ha_snapshot(addr, n, netdev);
+		ether_addr_copy(attrs->dmac, addr);
+		break;
 	default:
 		return;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 001d7c3add6a..4c800b54d8b6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -11,6 +11,7 @@
 
 #define NUM_IPSEC_FTE BIT(15)
 #define MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_SIZE 16
+#define IPSEC_TUNNEL_DEFAULT_TTL 0x40
 
 struct mlx5e_ipsec_fc {
 	struct mlx5_fc *cnt;
@@ -842,12 +843,31 @@ setup_pkt_tunnel_reformat(struct mlx5_core_dev *mdev,
 			  struct mlx5_accel_esp_xfrm_attrs *attrs,
 			  struct mlx5_pkt_reformat_params *reformat_params)
 {
+	struct ip_esp_hdr *esp_hdr;
+	struct ipv6hdr *ipv6hdr;
 	struct ethhdr *eth_hdr;
+	struct iphdr *iphdr;
 	char *reformatbf;
 	size_t bfflen;
+	void *hdr;
 
 	bfflen = sizeof(*eth_hdr);
 
+	if (attrs->dir == XFRM_DEV_OFFLOAD_OUT) {
+		bfflen += sizeof(*esp_hdr) + 8;
+
+		switch (attrs->family) {
+		case AF_INET:
+			bfflen += sizeof(*iphdr);
+			break;
+		case AF_INET6:
+			bfflen += sizeof(*ipv6hdr);
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
 	reformatbf = kzalloc(bfflen, GFP_KERNEL);
 	if (!reformatbf)
 		return -ENOMEM;
@@ -871,6 +891,38 @@ setup_pkt_tunnel_reformat(struct mlx5_core_dev *mdev,
 	case XFRM_DEV_OFFLOAD_IN:
 		reformat_params->type = MLX5_REFORMAT_TYPE_L3_ESP_TUNNEL_TO_L2;
 		break;
+	case XFRM_DEV_OFFLOAD_OUT:
+		reformat_params->type = MLX5_REFORMAT_TYPE_L2_TO_L3_ESP_TUNNEL;
+		reformat_params->param_0 = attrs->authsize;
+
+		hdr = reformatbf + sizeof(*eth_hdr);
+		switch (attrs->family) {
+		case AF_INET:
+			iphdr = (struct iphdr *)hdr;
+			memcpy(&iphdr->saddr, &attrs->saddr.a4, 4);
+			memcpy(&iphdr->daddr, &attrs->daddr.a4, 4);
+			iphdr->version = 4;
+			iphdr->ihl = 5;
+			iphdr->ttl = IPSEC_TUNNEL_DEFAULT_TTL;
+			iphdr->protocol = IPPROTO_ESP;
+			hdr += sizeof(*iphdr);
+			break;
+		case AF_INET6:
+			ipv6hdr = (struct ipv6hdr *)hdr;
+			memcpy(&ipv6hdr->saddr, &attrs->saddr.a6, 16);
+			memcpy(&ipv6hdr->daddr, &attrs->daddr.a6, 16);
+			ipv6hdr->nexthdr = IPPROTO_ESP;
+			ipv6hdr->version = 6;
+			ipv6hdr->hop_limit = IPSEC_TUNNEL_DEFAULT_TTL;
+			hdr += sizeof(*ipv6hdr);
+			break;
+		default:
+			goto free_reformatbf;
+		}
+
+		esp_hdr = (struct ip_esp_hdr *)hdr;
+		esp_hdr->spi = htonl(attrs->spi);
+		break;
 	default:
 		goto free_reformatbf;
 	}
-- 
cgit v1.2.3-70-g09d2


From 4c24272b4e2befca6ad1409c3c9aaa16c24b1099 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:25 +0300
Subject: net/mlx5e: Listen to ARP events to update IPsec L2 headers in tunnel
 mode

In IPsec packet offload mode all header manipulations are performed by
hardware, which is responsible to add/remove L2 header with source and
destinations MACs.

CX-7 devices don't support offload of in-kernel routing functionality,
as such HW needs external help to fill other side MAC as it isn't
available for HW.

As a solution, let's listen to neigh ARP updates and reconfigure IPsec
rules on the fly once new MAC data information arrives.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 132 +++++++++++++++++++--
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |   5 +
 2 files changed, 130 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 36f3ffd54355..b64281fd4142 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -35,12 +35,14 @@
 #include <crypto/aead.h>
 #include <linux/inetdevice.h>
 #include <linux/netdevice.h>
+#include <net/netevent.h>
 
 #include "en.h"
 #include "ipsec.h"
 #include "ipsec_rxtx.h"
 
 #define MLX5_IPSEC_RESCHED msecs_to_jiffies(1000)
+#define MLX5E_IPSEC_TUNNEL_SA XA_MARK_1
 
 static struct mlx5e_ipsec_sa_entry *to_ipsec_sa_entry(struct xfrm_state *x)
 {
@@ -251,7 +253,7 @@ static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	struct neighbour *n;
 	u8 addr[ETH_ALEN];
 
-	if (attrs->mode != XFRM_MODE_TUNNEL &&
+	if (attrs->mode != XFRM_MODE_TUNNEL ||
 	    attrs->type != XFRM_DEV_OFFLOAD_PACKET)
 		return;
 
@@ -267,6 +269,8 @@ static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
 			if (IS_ERR(n))
 				return;
 			neigh_event_send(n, NULL);
+			attrs->drop = true;
+			break;
 		}
 		neigh_ha_snapshot(addr, n, netdev);
 		ether_addr_copy(attrs->smac, addr);
@@ -279,6 +283,8 @@ static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
 			if (IS_ERR(n))
 				return;
 			neigh_event_send(n, NULL);
+			attrs->drop = true;
+			break;
 		}
 		neigh_ha_snapshot(addr, n, netdev);
 		ether_addr_copy(attrs->dmac, addr);
@@ -507,34 +513,81 @@ static void mlx5e_ipsec_set_esn_ops(struct mlx5e_ipsec_sa_entry *sa_entry)
 	sa_entry->set_iv_op = mlx5e_ipsec_set_iv;
 }
 
+static void mlx5e_ipsec_handle_netdev_event(struct work_struct *_work)
+{
+	struct mlx5e_ipsec_work *work =
+		container_of(_work, struct mlx5e_ipsec_work, work);
+	struct mlx5e_ipsec_sa_entry *sa_entry = work->sa_entry;
+	struct mlx5e_ipsec_netevent_data *data = work->data;
+	struct mlx5_accel_esp_xfrm_attrs *attrs;
+
+	attrs = &sa_entry->attrs;
+
+	switch (attrs->dir) {
+	case XFRM_DEV_OFFLOAD_IN:
+		ether_addr_copy(attrs->smac, data->addr);
+		break;
+	case XFRM_DEV_OFFLOAD_OUT:
+		ether_addr_copy(attrs->dmac, data->addr);
+		break;
+	default:
+		WARN_ON_ONCE(true);
+	}
+	attrs->drop = false;
+	mlx5e_accel_ipsec_fs_modify(sa_entry);
+}
+
 static int mlx5_ipsec_create_work(struct mlx5e_ipsec_sa_entry *sa_entry)
 {
 	struct xfrm_state *x = sa_entry->x;
 	struct mlx5e_ipsec_work *work;
+	void *data = NULL;
 
 	switch (x->xso.type) {
 	case XFRM_DEV_OFFLOAD_CRYPTO:
 		if (!(x->props.flags & XFRM_STATE_ESN))
 			return 0;
 		break;
+	case XFRM_DEV_OFFLOAD_PACKET:
+		if (x->props.mode != XFRM_MODE_TUNNEL)
+			return 0;
+		break;
 	default:
-		return 0;
+		break;
 	}
 
 	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (!work)
 		return -ENOMEM;
 
-	work->data = kzalloc(sizeof(*sa_entry), GFP_KERNEL);
-	if (!work->data) {
-		kfree(work);
-		return -ENOMEM;
+	switch (x->xso.type) {
+	case XFRM_DEV_OFFLOAD_CRYPTO:
+		data = kzalloc(sizeof(*sa_entry), GFP_KERNEL);
+		if (!data)
+			goto free_work;
+
+		INIT_WORK(&work->work, mlx5e_ipsec_modify_state);
+		break;
+	case XFRM_DEV_OFFLOAD_PACKET:
+		data = kzalloc(sizeof(struct mlx5e_ipsec_netevent_data),
+			       GFP_KERNEL);
+		if (!data)
+			goto free_work;
+
+		INIT_WORK(&work->work, mlx5e_ipsec_handle_netdev_event);
+		break;
+	default:
+		break;
 	}
 
-	INIT_WORK(&work->work, mlx5e_ipsec_modify_state);
+	work->data = data;
 	work->sa_entry = sa_entry;
 	sa_entry->work = work;
 	return 0;
+
+free_work:
+	kfree(work);
+	return -ENOMEM;
 }
 
 static int mlx5e_ipsec_create_dwork(struct mlx5e_ipsec_sa_entry *sa_entry)
@@ -629,6 +682,12 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 	if (sa_entry->dwork)
 		queue_delayed_work(ipsec->wq, &sa_entry->dwork->dwork,
 				   MLX5_IPSEC_RESCHED);
+
+	if (x->xso.type == XFRM_DEV_OFFLOAD_PACKET &&
+	    x->props.mode == XFRM_MODE_TUNNEL)
+		xa_set_mark(&ipsec->sadb, sa_entry->ipsec_obj_id,
+			    MLX5E_IPSEC_TUNNEL_SA);
+
 out:
 	x->xso.offload_handle = (unsigned long)sa_entry;
 	return 0;
@@ -651,6 +710,7 @@ err_xfrm:
 static void mlx5e_xfrm_del_state(struct xfrm_state *x)
 {
 	struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x);
+	struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs;
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
 	struct mlx5e_ipsec_sa_entry *old;
 
@@ -659,6 +719,12 @@ static void mlx5e_xfrm_del_state(struct xfrm_state *x)
 
 	old = xa_erase_bh(&ipsec->sadb, sa_entry->ipsec_obj_id);
 	WARN_ON(old != sa_entry);
+
+	if (attrs->mode == XFRM_MODE_TUNNEL &&
+	    attrs->type == XFRM_DEV_OFFLOAD_PACKET)
+		/* Make sure that no ARP requests are running in parallel */
+		flush_workqueue(ipsec->wq);
+
 }
 
 static void mlx5e_xfrm_free_state(struct xfrm_state *x)
@@ -683,6 +749,46 @@ sa_entry_free:
 	kfree(sa_entry);
 }
 
+static int mlx5e_ipsec_netevent_event(struct notifier_block *nb,
+				      unsigned long event, void *ptr)
+{
+	struct mlx5_accel_esp_xfrm_attrs *attrs;
+	struct mlx5e_ipsec_netevent_data *data;
+	struct mlx5e_ipsec_sa_entry *sa_entry;
+	struct mlx5e_ipsec *ipsec;
+	struct neighbour *n = ptr;
+	struct net_device *netdev;
+	struct xfrm_state *x;
+	unsigned long idx;
+
+	if (event != NETEVENT_NEIGH_UPDATE || !(n->nud_state & NUD_VALID))
+		return NOTIFY_DONE;
+
+	ipsec = container_of(nb, struct mlx5e_ipsec, netevent_nb);
+	xa_for_each_marked(&ipsec->sadb, idx, sa_entry, MLX5E_IPSEC_TUNNEL_SA) {
+		attrs = &sa_entry->attrs;
+
+		if (attrs->family == AF_INET) {
+			if (!neigh_key_eq32(n, &attrs->saddr.a4) &&
+			    !neigh_key_eq32(n, &attrs->daddr.a4))
+				continue;
+		} else {
+			if (!neigh_key_eq128(n, &attrs->saddr.a4) &&
+			    !neigh_key_eq128(n, &attrs->daddr.a4))
+				continue;
+		}
+
+		x = sa_entry->x;
+		netdev = x->xso.real_dev;
+		data = sa_entry->work->data;
+
+		neigh_ha_snapshot(data->addr, n, netdev);
+		queue_work(ipsec->wq, &sa_entry->work->work);
+	}
+
+	return NOTIFY_DONE;
+}
+
 void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 {
 	struct mlx5e_ipsec *ipsec;
@@ -711,6 +817,13 @@ void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 			goto err_aso;
 	}
 
+	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_TUNNEL) {
+		ipsec->netevent_nb.notifier_call = mlx5e_ipsec_netevent_event;
+		ret = register_netevent_notifier(&ipsec->netevent_nb);
+		if (ret)
+			goto clear_aso;
+	}
+
 	ret = mlx5e_accel_ipsec_fs_init(ipsec);
 	if (ret)
 		goto err_fs_init;
@@ -721,6 +834,9 @@ void mlx5e_ipsec_init(struct mlx5e_priv *priv)
 	return;
 
 err_fs_init:
+	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_TUNNEL)
+		unregister_netevent_notifier(&ipsec->netevent_nb);
+clear_aso:
 	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD)
 		mlx5e_ipsec_aso_cleanup(ipsec);
 err_aso:
@@ -739,6 +855,8 @@ void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv)
 		return;
 
 	mlx5e_accel_ipsec_fs_cleanup(ipsec);
+	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_TUNNEL)
+		unregister_netevent_notifier(&ipsec->netevent_nb);
 	if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD)
 		mlx5e_ipsec_aso_cleanup(ipsec);
 	destroy_workqueue(ipsec->wq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 77384ffa4451..d06c896eadb6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -144,6 +144,10 @@ struct mlx5e_ipsec_work {
 	void *data;
 };
 
+struct mlx5e_ipsec_netevent_data {
+	u8 addr[ETH_ALEN];
+};
+
 struct mlx5e_ipsec_dwork {
 	struct delayed_work dwork;
 	struct mlx5e_ipsec_sa_entry *sa_entry;
@@ -169,6 +173,7 @@ struct mlx5e_ipsec {
 	struct mlx5e_ipsec_tx *tx;
 	struct mlx5e_ipsec_aso *aso;
 	struct notifier_block nb;
+	struct notifier_block netevent_nb;
 	struct mlx5_ipsec_fs *roce;
 };
 
-- 
cgit v1.2.3-70-g09d2


From acc109291a028a9d9880dd44a9df8e79f2fb854c Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:26 +0300
Subject: net/mlx5: Allow blocking encap changes in eswitch

Existing eswitch encap option enables header encapsulation. Unfortunately
currently available hardware isn't able to perform double encapsulation,
which can happen once IPsec packet offload tunnel mode is used together
with encap mode set to BASIC.

So as a solution for misconfiguration, provide an option to block encap
changes, which will be used for IPsec packet offload.

Reviewed-by: Emeel Hakim <ehakim@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  | 14 +++++++
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 48 ++++++++++++++++++++++
 2 files changed, 62 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 19e9a77c4633..e9d68fdf68f5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -263,6 +263,7 @@ struct mlx5_esw_offload {
 	const struct mlx5_eswitch_rep_ops *rep_ops[NUM_REP_TYPES];
 	u8 inline_mode;
 	atomic64_t num_flows;
+	u64 num_block_encap;
 	enum devlink_eswitch_encap_mode encap;
 	struct ida vport_metadata_ida;
 	unsigned int host_number; /* ECPF supports one external host */
@@ -748,6 +749,9 @@ void mlx5_eswitch_offloads_destroy_single_fdb(struct mlx5_eswitch *master_esw,
 					      struct mlx5_eswitch *slave_esw);
 int mlx5_eswitch_reload_reps(struct mlx5_eswitch *esw);
 
+bool mlx5_eswitch_block_encap(struct mlx5_core_dev *dev);
+void mlx5_eswitch_unblock_encap(struct mlx5_core_dev *dev);
+
 static inline int mlx5_eswitch_num_vfs(struct mlx5_eswitch *esw)
 {
 	if (mlx5_esw_allowed(esw))
@@ -761,6 +765,7 @@ mlx5_eswitch_get_slow_fdb(struct mlx5_eswitch *esw)
 {
 	return esw->fdb_table.offloads.slow_fdb;
 }
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
@@ -805,6 +810,15 @@ mlx5_eswitch_reload_reps(struct mlx5_eswitch *esw)
 {
 	return 0;
 }
+
+static inline bool mlx5_eswitch_block_encap(struct mlx5_core_dev *dev)
+{
+	return true;
+}
+
+static inline void mlx5_eswitch_unblock_encap(struct mlx5_core_dev *dev)
+{
+}
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif /* __MLX5_ESWITCH_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 48036dfddd5e..b6e2709c1371 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -3586,6 +3586,47 @@ int mlx5_devlink_eswitch_inline_mode_get(struct devlink *devlink, u8 *mode)
 	return err;
 }
 
+bool mlx5_eswitch_block_encap(struct mlx5_core_dev *dev)
+{
+	struct devlink *devlink = priv_to_devlink(dev);
+	struct mlx5_eswitch *esw;
+
+	devl_lock(devlink);
+	esw = mlx5_devlink_eswitch_get(devlink);
+	if (IS_ERR(esw)) {
+		devl_unlock(devlink);
+		/* Failure means no eswitch => not possible to change encap */
+		return true;
+	}
+
+	down_write(&esw->mode_lock);
+	if (esw->mode != MLX5_ESWITCH_LEGACY &&
+	    esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) {
+		up_write(&esw->mode_lock);
+		devl_unlock(devlink);
+		return false;
+	}
+
+	esw->offloads.num_block_encap++;
+	up_write(&esw->mode_lock);
+	devl_unlock(devlink);
+	return true;
+}
+
+void mlx5_eswitch_unblock_encap(struct mlx5_core_dev *dev)
+{
+	struct devlink *devlink = priv_to_devlink(dev);
+	struct mlx5_eswitch *esw;
+
+	esw = mlx5_devlink_eswitch_get(devlink);
+	if (IS_ERR(esw))
+		return;
+
+	down_write(&esw->mode_lock);
+	esw->offloads.num_block_encap--;
+	up_write(&esw->mode_lock);
+}
+
 int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink,
 					enum devlink_eswitch_encap_mode encap,
 					struct netlink_ext_ack *extack)
@@ -3627,6 +3668,13 @@ int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink,
 		goto unlock;
 	}
 
+	if (esw->offloads.num_block_encap) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Can't set encapsulation when IPsec SA and/or policies are configured");
+		err = -EOPNOTSUPP;
+		goto unlock;
+	}
+
 	esw_destroy_offloads_fdb_tables(esw);
 
 	esw->offloads.encap = encap;
-- 
cgit v1.2.3-70-g09d2


From 146c196b60e49737a08dd2cb12cc1a9683a36074 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:27 +0300
Subject: net/mlx5e: Create IPsec table with tunnel support only when encap is
 disabled

Current hardware doesn't support double encapsulation which is
happening when IPsec packet offload tunnel mode is configured
together with eswitch encap option.

Any user attempt to add new SA/policy after he/she sets encap mode, will
generate the following FW syndrome:

 mlx5_core 0000:08:00.0: mlx5_cmd_out_err:803:(pid 1904): CREATE_FLOW_TABLE(0x930) op_mod(0x0) failed,
 status bad parameter(0x3), syndrome (0xa43321), err(-22)

Make sure that we block encap changes before creating flow steering tables.
This is applicable only for packet offload in tunnel mode, while packet
offload in transport mode and crypto offload, don't have such limitation
as they don't perform encapsulation.

Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   |  8 ++++++
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h   |  1 +
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c         | 33 ++++++++++++++++++++--
 3 files changed, 39 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index b64281fd4142..0bda5a91bff6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -668,6 +668,14 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x,
 	if (err)
 		goto err_hw_ctx;
 
+	if (x->props.mode == XFRM_MODE_TUNNEL &&
+	    x->xso.type == XFRM_DEV_OFFLOAD_PACKET &&
+	    !mlx5e_ipsec_fs_tunnel_enabled(sa_entry)) {
+		NL_SET_ERR_MSG_MOD(extack, "Packet offload tunnel mode is disabled due to encap settings");
+		err = -EINVAL;
+		goto err_add_rule;
+	}
+
 	/* We use *_bh() variant because xfrm_timer_handler(), which runs
 	 * in softirq context, can reach our state delete logic and we need
 	 * xa_erase_bh() there.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index d06c896eadb6..f7f7c09d2b32 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -251,6 +251,7 @@ void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry);
 int mlx5e_accel_ipsec_fs_add_pol(struct mlx5e_ipsec_pol_entry *pol_entry);
 void mlx5e_accel_ipsec_fs_del_pol(struct mlx5e_ipsec_pol_entry *pol_entry);
 void mlx5e_accel_ipsec_fs_modify(struct mlx5e_ipsec_sa_entry *sa_entry);
+bool mlx5e_ipsec_fs_tunnel_enabled(struct mlx5e_ipsec_sa_entry *sa_entry);
 
 int mlx5_ipsec_create_sa_ctx(struct mlx5e_ipsec_sa_entry *sa_entry);
 void mlx5_ipsec_free_sa_ctx(struct mlx5e_ipsec_sa_entry *sa_entry);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 4c800b54d8b6..5a8fcd30fcb1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -4,6 +4,7 @@
 #include <linux/netdevice.h>
 #include "en.h"
 #include "en/fs.h"
+#include "eswitch.h"
 #include "ipsec.h"
 #include "fs_core.h"
 #include "lib/ipsec_fs_roce.h"
@@ -38,6 +39,7 @@ struct mlx5e_ipsec_rx {
 	struct mlx5e_ipsec_rule status;
 	struct mlx5e_ipsec_fc *fc;
 	struct mlx5_fs_chains *chains;
+	u8 allow_tunnel_mode : 1;
 };
 
 struct mlx5e_ipsec_tx {
@@ -47,6 +49,7 @@ struct mlx5e_ipsec_tx {
 	struct mlx5_flow_namespace *ns;
 	struct mlx5e_ipsec_fc *fc;
 	struct mlx5_fs_chains *chains;
+	u8 allow_tunnel_mode : 1;
 };
 
 /* IPsec RX flow steering */
@@ -254,7 +257,8 @@ static void rx_destroy(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 	mlx5_del_flow_rules(rx->sa.rule);
 	mlx5_destroy_flow_group(rx->sa.group);
 	mlx5_destroy_flow_table(rx->ft.sa);
-
+	if (rx->allow_tunnel_mode)
+		mlx5_eswitch_unblock_encap(mdev);
 	mlx5_del_flow_rules(rx->status.rule);
 	mlx5_modify_header_dealloc(mdev, rx->status.modify_hdr);
 	mlx5_destroy_flow_table(rx->ft.status);
@@ -305,6 +309,8 @@ static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec,
 
 	/* Create FT */
 	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_TUNNEL)
+		rx->allow_tunnel_mode = mlx5_eswitch_block_encap(mdev);
+	if (rx->allow_tunnel_mode)
 		flags = MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT;
 	ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_ESP_FT_LEVEL, MLX5E_NIC_PRIO, 2,
 			     flags);
@@ -362,6 +368,8 @@ err_pol_ft:
 err_fs:
 	mlx5_destroy_flow_table(rx->ft.sa);
 err_fs_ft:
+	if (rx->allow_tunnel_mode)
+		mlx5_eswitch_unblock_encap(mdev);
 	mlx5_del_flow_rules(rx->status.rule);
 	mlx5_modify_header_dealloc(mdev, rx->status.modify_hdr);
 err_add:
@@ -496,7 +504,8 @@ err_rule:
 }
 
 /* IPsec TX flow steering */
-static void tx_destroy(struct mlx5e_ipsec_tx *tx, struct mlx5_ipsec_fs *roce)
+static void tx_destroy(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
+		       struct mlx5_ipsec_fs *roce)
 {
 	mlx5_ipsec_fs_roce_tx_destroy(roce);
 	if (tx->chains) {
@@ -508,6 +517,8 @@ static void tx_destroy(struct mlx5e_ipsec_tx *tx, struct mlx5_ipsec_fs *roce)
 	}
 
 	mlx5_destroy_flow_table(tx->ft.sa);
+	if (tx->allow_tunnel_mode)
+		mlx5_eswitch_unblock_encap(mdev);
 	mlx5_del_flow_rules(tx->status.rule);
 	mlx5_destroy_flow_table(tx->ft.status);
 }
@@ -530,6 +541,8 @@ static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx,
 		goto err_status_rule;
 
 	if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_TUNNEL)
+		tx->allow_tunnel_mode = mlx5_eswitch_block_encap(mdev);
+	if (tx->allow_tunnel_mode)
 		flags = MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT;
 	ft = ipsec_ft_create(tx->ns, 1, 0, 4, flags);
 	if (IS_ERR(ft)) {
@@ -581,6 +594,8 @@ err_roce:
 err_pol_ft:
 	mlx5_destroy_flow_table(tx->ft.sa);
 err_sa_ft:
+	if (tx->allow_tunnel_mode)
+		mlx5_eswitch_unblock_encap(mdev);
 	mlx5_del_flow_rules(tx->status.rule);
 err_status_rule:
 	mlx5_destroy_flow_table(tx->ft.status);
@@ -609,7 +624,7 @@ static void tx_put(struct mlx5e_ipsec *ipsec, struct mlx5e_ipsec_tx *tx)
 	if (--tx->ft.refcnt)
 		return;
 
-	tx_destroy(tx, ipsec->roce);
+	tx_destroy(ipsec->mdev, tx, ipsec->roce);
 }
 
 static struct mlx5_flow_table *tx_ft_get_policy(struct mlx5_core_dev *mdev,
@@ -1603,3 +1618,15 @@ void mlx5e_accel_ipsec_fs_modify(struct mlx5e_ipsec_sa_entry *sa_entry)
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 	memcpy(sa_entry, &sa_entry_shadow, sizeof(*sa_entry));
 }
+
+bool mlx5e_ipsec_fs_tunnel_enabled(struct mlx5e_ipsec_sa_entry *sa_entry)
+{
+	struct mlx5e_ipsec_rx *rx =
+		ipsec_rx(sa_entry->ipsec, sa_entry->attrs.family);
+	struct mlx5e_ipsec_tx *tx = sa_entry->ipsec->tx;
+
+	if (sa_entry->attrs.dir == XFRM_DEV_OFFLOAD_OUT)
+		return tx->allow_tunnel_mode;
+
+	return rx->allow_tunnel_mode;
+}
-- 
cgit v1.2.3-70-g09d2


From c941da23aaf065a893248943cedb27b0f492b292 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 13 Apr 2023 15:29:28 +0300
Subject: net/mlx5e: Accept tunnel mode for IPsec packet offload

Open mlx5 driver to accept IPsec tunnel mode.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 0bda5a91bff6..5fd609d1120e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -422,6 +422,11 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 		return -EINVAL;
 	}
 
+	if (x->props.mode != XFRM_MODE_TRANSPORT && x->props.mode != XFRM_MODE_TUNNEL) {
+		NL_SET_ERR_MSG_MOD(extack, "Only transport and tunnel xfrm states may be offloaded");
+		return -EINVAL;
+	}
+
 	switch (x->xso.type) {
 	case XFRM_DEV_OFFLOAD_CRYPTO:
 		if (!(mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_CRYPTO)) {
@@ -429,11 +434,6 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 			return -EINVAL;
 		}
 
-		if (x->props.mode != XFRM_MODE_TRANSPORT &&
-		    x->props.mode != XFRM_MODE_TUNNEL) {
-			NL_SET_ERR_MSG_MOD(extack, "Only transport and tunnel xfrm states may be offloaded");
-			return -EINVAL;
-		}
 		break;
 	case XFRM_DEV_OFFLOAD_PACKET:
 		if (!(mlx5_ipsec_device_caps(mdev) &
@@ -442,8 +442,9 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 			return -EINVAL;
 		}
 
-		if (x->props.mode != XFRM_MODE_TRANSPORT) {
-			NL_SET_ERR_MSG_MOD(extack, "Only transport xfrm states may be offloaded in packet mode");
+		if (x->props.mode == XFRM_MODE_TUNNEL &&
+		    !(mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_TUNNEL)) {
+			NL_SET_ERR_MSG_MOD(extack, "Packet offload is not supported for tunnel mode");
 			return -EINVAL;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 5b24324a907c996faf47ea3969499e04beea9c4f Mon Sep 17 00:00:00 2001
From: Song Yoong Siang <yoong.siang.song@intel.com>
Date: Sat, 15 Apr 2023 14:45:01 +0800
Subject: net: stmmac: introduce wrapper for struct xdp_buff

Introduce struct stmmac_xdp_buff as a preparation to support XDP Rx
metadata via kfuncs.

Signed-off-by: Song Yoong Siang <yoong.siang.song@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac.h      |  4 ++++
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 18 +++++++++---------
 2 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 3d15e1e92e18..ac8ccf851708 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -92,6 +92,10 @@ struct stmmac_rx_buffer {
 	dma_addr_t sec_addr;
 };
 
+struct stmmac_xdp_buff {
+	struct xdp_buff xdp;
+};
+
 struct stmmac_rx_queue {
 	u32 rx_count_frames;
 	u32 queue_index;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index d7fcab057032..10b9f8912bb2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -5190,7 +5190,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue)
 	enum dma_data_direction dma_dir;
 	unsigned int desc_size;
 	struct sk_buff *skb = NULL;
-	struct xdp_buff xdp;
+	struct stmmac_xdp_buff ctx;
 	int xdp_status = 0;
 	int buf_sz;
 
@@ -5311,17 +5311,17 @@ read_again:
 			dma_sync_single_for_cpu(priv->device, buf->addr,
 						buf1_len, dma_dir);
 
-			xdp_init_buff(&xdp, buf_sz, &rx_q->xdp_rxq);
-			xdp_prepare_buff(&xdp, page_address(buf->page),
+			xdp_init_buff(&ctx.xdp, buf_sz, &rx_q->xdp_rxq);
+			xdp_prepare_buff(&ctx.xdp, page_address(buf->page),
 					 buf->page_offset, buf1_len, false);
 
-			pre_len = xdp.data_end - xdp.data_hard_start -
+			pre_len = ctx.xdp.data_end - ctx.xdp.data_hard_start -
 				  buf->page_offset;
-			skb = stmmac_xdp_run_prog(priv, &xdp);
+			skb = stmmac_xdp_run_prog(priv, &ctx.xdp);
 			/* Due xdp_adjust_tail: DMA sync for_device
 			 * cover max len CPU touch
 			 */
-			sync_len = xdp.data_end - xdp.data_hard_start -
+			sync_len = ctx.xdp.data_end - ctx.xdp.data_hard_start -
 				   buf->page_offset;
 			sync_len = max(sync_len, pre_len);
 
@@ -5331,7 +5331,7 @@ read_again:
 
 				if (xdp_res & STMMAC_XDP_CONSUMED) {
 					page_pool_put_page(rx_q->page_pool,
-							   virt_to_head_page(xdp.data),
+							   virt_to_head_page(ctx.xdp.data),
 							   sync_len, true);
 					buf->page = NULL;
 					priv->dev->stats.rx_dropped++;
@@ -5359,7 +5359,7 @@ read_again:
 
 		if (!skb) {
 			/* XDP program may expand or reduce tail */
-			buf1_len = xdp.data_end - xdp.data;
+			buf1_len = ctx.xdp.data_end - ctx.xdp.data;
 
 			skb = napi_alloc_skb(&ch->rx_napi, buf1_len);
 			if (!skb) {
@@ -5369,7 +5369,7 @@ read_again:
 			}
 
 			/* XDP program may adjust header */
-			skb_copy_to_linear_data(skb, xdp.data, buf1_len);
+			skb_copy_to_linear_data(skb, ctx.xdp.data, buf1_len);
 			skb_put(skb, buf1_len);
 
 			/* Data payload copied into SKB, page ready for recycle */
-- 
cgit v1.2.3-70-g09d2


From e3f9c3e348406f7016ded85aed119edfcd2de2fa Mon Sep 17 00:00:00 2001
From: Song Yoong Siang <yoong.siang.song@intel.com>
Date: Sat, 15 Apr 2023 14:45:02 +0800
Subject: net: stmmac: add Rx HWTS metadata to XDP receive pkt

Add receive hardware timestamp metadata support via kfunc to XDP receive
packets.

Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Song Yoong Siang <yoong.siang.song@intel.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac.h      |  3 ++
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 40 ++++++++++++++++++++++-
 2 files changed, 42 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index ac8ccf851708..07ea5ab0a60b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -94,6 +94,9 @@ struct stmmac_rx_buffer {
 
 struct stmmac_xdp_buff {
 	struct xdp_buff xdp;
+	struct stmmac_priv *priv;
+	struct dma_desc *desc;
+	struct dma_desc *ndesc;
 };
 
 struct stmmac_rx_queue {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 10b9f8912bb2..2bfcc5347d6a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -5313,10 +5313,15 @@ read_again:
 
 			xdp_init_buff(&ctx.xdp, buf_sz, &rx_q->xdp_rxq);
 			xdp_prepare_buff(&ctx.xdp, page_address(buf->page),
-					 buf->page_offset, buf1_len, false);
+					 buf->page_offset, buf1_len, true);
 
 			pre_len = ctx.xdp.data_end - ctx.xdp.data_hard_start -
 				  buf->page_offset;
+
+			ctx.priv = priv;
+			ctx.desc = p;
+			ctx.ndesc = np;
+
 			skb = stmmac_xdp_run_prog(priv, &ctx.xdp);
 			/* Due xdp_adjust_tail: DMA sync for_device
 			 * cover max len CPU touch
@@ -7060,6 +7065,37 @@ void stmmac_fpe_handshake(struct stmmac_priv *priv, bool enable)
 	}
 }
 
+static int stmmac_xdp_rx_timestamp(const struct xdp_md *_ctx, u64 *timestamp)
+{
+	const struct stmmac_xdp_buff *ctx = (void *)_ctx;
+	struct dma_desc *desc_contains_ts = ctx->desc;
+	struct stmmac_priv *priv = ctx->priv;
+	struct dma_desc *ndesc = ctx->ndesc;
+	struct dma_desc *desc = ctx->desc;
+	u64 ns = 0;
+
+	if (!priv->hwts_rx_en)
+		return -ENODATA;
+
+	/* For GMAC4, the valid timestamp is from CTX next desc. */
+	if (priv->plat->has_gmac4 || priv->plat->has_xgmac)
+		desc_contains_ts = ndesc;
+
+	/* Check if timestamp is available */
+	if (stmmac_get_rx_timestamp_status(priv, desc, ndesc, priv->adv_ts)) {
+		stmmac_get_timestamp(priv, desc_contains_ts, priv->adv_ts, &ns);
+		ns -= priv->plat->cdc_error_adj;
+		*timestamp = ns_to_ktime(ns);
+		return 0;
+	}
+
+	return -ENODATA;
+}
+
+static const struct xdp_metadata_ops stmmac_xdp_metadata_ops = {
+	.xmo_rx_timestamp		= stmmac_xdp_rx_timestamp,
+};
+
 /**
  * stmmac_dvr_probe
  * @device: device pointer
@@ -7167,6 +7203,8 @@ int stmmac_dvr_probe(struct device *device,
 
 	ndev->netdev_ops = &stmmac_netdev_ops;
 
+	ndev->xdp_metadata_ops = &stmmac_xdp_metadata_ops;
+
 	ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 			    NETIF_F_RXCSUM;
 	ndev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
-- 
cgit v1.2.3-70-g09d2


From 9570df353309d010f6afc7146c88a135ee3d3f8a Mon Sep 17 00:00:00 2001
From: Song Yoong Siang <yoong.siang.song@intel.com>
Date: Sat, 15 Apr 2023 14:45:03 +0800
Subject: net: stmmac: add Rx HWTS metadata to XDP ZC receive pkt

Add receive hardware timestamp metadata support via kfunc to XDP Zero Copy
receive packets.

Signed-off-by: Song Yoong Siang <yoong.siang.song@intel.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 2bfcc5347d6a..c0e90fda572a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1614,6 +1614,12 @@ static int stmmac_alloc_rx_buffers_zc(struct stmmac_priv *priv,
 	struct stmmac_rx_queue *rx_q = &dma_conf->rx_queue[queue];
 	int i;
 
+	/* struct stmmac_xdp_buff is using cb field (maximum size of 24 bytes)
+	 * in struct xdp_buff_xsk to stash driver specific information. Thus,
+	 * use this macro to make sure no size violations.
+	 */
+	XSK_CHECK_PRIV_TYPE(struct stmmac_xdp_buff);
+
 	for (i = 0; i < dma_conf->dma_rx_size; i++) {
 		struct stmmac_rx_buffer *buf;
 		dma_addr_t dma_addr;
@@ -4998,6 +5004,16 @@ static bool stmmac_rx_refill_zc(struct stmmac_priv *priv, u32 queue, u32 budget)
 	return ret;
 }
 
+static struct stmmac_xdp_buff *xsk_buff_to_stmmac_ctx(struct xdp_buff *xdp)
+{
+	/* In XDP zero copy data path, xdp field in struct xdp_buff_xsk is used
+	 * to represent incoming packet, whereas cb field in the same structure
+	 * is used to store driver specific info. Thus, struct stmmac_xdp_buff
+	 * is laid on top of xdp and cb fields of struct xdp_buff_xsk.
+	 */
+	return (struct stmmac_xdp_buff *)xdp;
+}
+
 static int stmmac_rx_zc(struct stmmac_priv *priv, int limit, u32 queue)
 {
 	struct stmmac_rx_queue *rx_q = &priv->dma_conf.rx_queue[queue];
@@ -5027,6 +5043,7 @@ static int stmmac_rx_zc(struct stmmac_priv *priv, int limit, u32 queue)
 	}
 	while (count < limit) {
 		struct stmmac_rx_buffer *buf;
+		struct stmmac_xdp_buff *ctx;
 		unsigned int buf1_len = 0;
 		struct dma_desc *np, *p;
 		int entry;
@@ -5112,6 +5129,11 @@ read_again:
 			goto read_again;
 		}
 
+		ctx = xsk_buff_to_stmmac_ctx(buf->xdp);
+		ctx->priv = priv;
+		ctx->desc = p;
+		ctx->ndesc = np;
+
 		/* XDP ZC Frame only support primary buffers for now */
 		buf1_len = stmmac_rx_buf1_len(priv, p, status, len);
 		len += buf1_len;
-- 
cgit v1.2.3-70-g09d2


From 15f93f46f31232da863316769182c699e364c45f Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:45 +0300
Subject: net: mscc: ocelot: export a single ocelot_mm_irq()

When the switch emits an IRQ, we don't know what caused it, and we
iterate through all ports to check the MAC Merge status.

Move that iteration inside the ocelot lib; we will change the locking in
a future change and it would be good to encapsulate that lock completely
within the ocelot lib.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c |  5 +----
 drivers/net/ethernet/mscc/ocelot_mm.c  | 12 ++++++++++--
 include/soc/mscc/ocelot.h              |  2 +-
 3 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index dddb28984bdf..478893c06f56 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2610,12 +2610,9 @@ static const struct felix_info felix_info_vsc9959 = {
 static irqreturn_t felix_irq_handler(int irq, void *data)
 {
 	struct ocelot *ocelot = (struct ocelot *)data;
-	int port;
 
 	ocelot_get_txtstamp(ocelot);
-
-	for (port = 0; port < ocelot->num_phys_ports; port++)
-		ocelot_port_mm_irq(ocelot, port);
+	ocelot_mm_irq(ocelot);
 
 	return IRQ_HANDLED;
 }
diff --git a/drivers/net/ethernet/mscc/ocelot_mm.c b/drivers/net/ethernet/mscc/ocelot_mm.c
index 0a8f21ae23f0..ddaf1fb05e48 100644
--- a/drivers/net/ethernet/mscc/ocelot_mm.c
+++ b/drivers/net/ethernet/mscc/ocelot_mm.c
@@ -49,7 +49,7 @@ static enum ethtool_mm_verify_status ocelot_mm_verify_status(u32 val)
 	}
 }
 
-void ocelot_port_mm_irq(struct ocelot *ocelot, int port)
+static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 {
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
 	struct ocelot_mm_state *mm = &ocelot->mm[port];
@@ -91,7 +91,15 @@ void ocelot_port_mm_irq(struct ocelot *ocelot, int port)
 
 	mutex_unlock(&mm->lock);
 }
-EXPORT_SYMBOL_GPL(ocelot_port_mm_irq);
+
+void ocelot_mm_irq(struct ocelot *ocelot)
+{
+	int port;
+
+	for (port = 0; port < ocelot->num_phys_ports; port++)
+		ocelot_mm_update_port_status(ocelot, port);
+}
+EXPORT_SYMBOL_GPL(ocelot_mm_irq);
 
 int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 		       struct ethtool_mm_cfg *cfg,
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 277e6d1f2096..eb8e3935375d 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -1148,7 +1148,7 @@ int ocelot_vcap_policer_add(struct ocelot *ocelot, u32 pol_ix,
 			    struct ocelot_policer *pol);
 int ocelot_vcap_policer_del(struct ocelot *ocelot, u32 pol_ix);
 
-void ocelot_port_mm_irq(struct ocelot *ocelot, int port);
+void ocelot_mm_irq(struct ocelot *ocelot);
 int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 		       struct ethtool_mm_cfg *cfg,
 		       struct netlink_ext_ack *extack);
-- 
cgit v1.2.3-70-g09d2


From 3ff468ef987e38740de9ca0a811c55e11bfb2141 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:46 +0300
Subject: net: mscc: ocelot: remove struct ocelot_mm_state :: lock

Unfortunately, the workarounds for the hardware bugs make it pointless
to keep fine-grained locking for the MAC Merge state of each port.

Our vsc9959_cut_through_fwd() implementation requires
ocelot->fwd_domain_lock to be held, in order to serialize with changes
to the bridging domains and to port speed changes (which affect which
ports can be cut-through). Simultaneously, the traffic classes which can
be cut-through cannot be preemptible at the same time, and this will
depend on the MAC Merge layer state (which changes from threaded
interrupt context).

Since vsc9959_cut_through_fwd() would have to hold the mm->lock of all
ports for a correct and race-free implementation with respect to
ocelot_mm_irq(), in practice it means that any time a port's mm->lock is
held, it would potentially block holders of ocelot->fwd_domain_lock.

In the interest of simple locking rules, make all MAC Merge layer state
changes (and preemptible traffic class changes) be serialized by the
ocelot->fwd_domain_lock.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_mm.c | 20 ++++++++------------
 include/soc/mscc/ocelot.h             |  1 -
 2 files changed, 8 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_mm.c b/drivers/net/ethernet/mscc/ocelot_mm.c
index ddaf1fb05e48..d2df47e6f8f6 100644
--- a/drivers/net/ethernet/mscc/ocelot_mm.c
+++ b/drivers/net/ethernet/mscc/ocelot_mm.c
@@ -56,8 +56,6 @@ static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 	enum ethtool_mm_verify_status verify_status;
 	u32 val;
 
-	mutex_lock(&mm->lock);
-
 	val = ocelot_port_readl(ocelot_port, DEV_MM_STATUS);
 
 	verify_status = ocelot_mm_verify_status(val);
@@ -88,16 +86,18 @@ static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 	}
 
 	ocelot_port_writel(ocelot_port, val, DEV_MM_STATUS);
-
-	mutex_unlock(&mm->lock);
 }
 
 void ocelot_mm_irq(struct ocelot *ocelot)
 {
 	int port;
 
+	mutex_lock(&ocelot->fwd_domain_lock);
+
 	for (port = 0; port < ocelot->num_phys_ports; port++)
 		ocelot_mm_update_port_status(ocelot, port);
+
+	mutex_unlock(&ocelot->fwd_domain_lock);
 }
 EXPORT_SYMBOL_GPL(ocelot_mm_irq);
 
@@ -107,14 +107,11 @@ int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 {
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
 	u32 mm_enable = 0, verify_disable = 0, add_frag_size;
-	struct ocelot_mm_state *mm;
 	int err;
 
 	if (!ocelot->mm_supported)
 		return -EOPNOTSUPP;
 
-	mm = &ocelot->mm[port];
-
 	err = ethtool_mm_frag_size_min_to_add(cfg->tx_min_frag_size,
 					      &add_frag_size, extack);
 	if (err)
@@ -129,7 +126,7 @@ int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 	if (!cfg->verify_enabled)
 		verify_disable = DEV_MM_CONFIG_VERIF_CONFIG_PRM_VERIFY_DIS;
 
-	mutex_lock(&mm->lock);
+	mutex_lock(&ocelot->fwd_domain_lock);
 
 	ocelot_port_rmwl(ocelot_port, mm_enable,
 			 DEV_MM_CONFIG_ENABLE_CONFIG_MM_TX_ENA |
@@ -148,7 +145,7 @@ int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 		       QSYS_PREEMPTION_CFG,
 		       port);
 
-	mutex_unlock(&mm->lock);
+	mutex_unlock(&ocelot->fwd_domain_lock);
 
 	return 0;
 }
@@ -166,7 +163,7 @@ int ocelot_port_get_mm(struct ocelot *ocelot, int port,
 
 	mm = &ocelot->mm[port];
 
-	mutex_lock(&mm->lock);
+	mutex_lock(&ocelot->fwd_domain_lock);
 
 	val = ocelot_port_readl(ocelot_port, DEV_MM_ENABLE_CONFIG);
 	state->pmac_enabled = !!(val & DEV_MM_CONFIG_ENABLE_CONFIG_MM_RX_ENA);
@@ -185,7 +182,7 @@ int ocelot_port_get_mm(struct ocelot *ocelot, int port,
 	state->verify_status = mm->verify_status;
 	state->tx_active = mm->tx_active;
 
-	mutex_unlock(&mm->lock);
+	mutex_unlock(&ocelot->fwd_domain_lock);
 
 	return 0;
 }
@@ -209,7 +206,6 @@ int ocelot_mm_init(struct ocelot *ocelot)
 		u32 val;
 
 		mm = &ocelot->mm[port];
-		mutex_init(&mm->lock);
 		ocelot_port = ocelot->ports[port];
 
 		/* Update initial status variable for the
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index eb8e3935375d..9599be6a0a39 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -744,7 +744,6 @@ struct ocelot_mirror {
 };
 
 struct ocelot_mm_state {
-	struct mutex lock;
 	enum ethtool_mm_verify_status verify_status;
 	bool tx_active;
 };
-- 
cgit v1.2.3-70-g09d2


From 7bf4a5b071e59f48de8d39dfde07a3a65e7f6488 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:47 +0300
Subject: net: mscc: ocelot: optimize ocelot_mm_irq()

The MAC Merge IRQ of all ports is shared with the PTP TX timestamp IRQ
of all ports, which means that currently, when a PTP TX timestamp is
generated, felix_irq_handler() also polls for the MAC Merge layer status
of all ports, looking for changes. This makes the kernel do more work,
and under certain circumstances may make ptp4l require a
tx_timestamp_timeout argument higher than before.

Changes to the MAC Merge layer status are only to be expected under
certain conditions - its TX direction needs to be enabled - so we can
check early if that is the case, and omit register access otherwise.

Make ocelot_mm_update_port_status() skip register access if
mm->tx_enabled is unset, and also call it once more, outside IRQ
context, from ocelot_port_set_mm(), when mm->tx_enabled transitions from
true to false, because an IRQ is also expected in that case.

Also, a port may have its MAC Merge layer enabled but it may not have
generated the interrupt. In that case, there's no point in writing to
DEV_MM_STATUS to acknowledge that IRQ. We can reduce the number of
register writes per port with MM enabled by keeping an "ack" variable
which writes the "write-one-to-clear" bits. Those are 3 in number:
PRMPT_ACTIVE_STICKY, UNEXP_RX_PFRM_STICKY and UNEXP_TX_PFRM_STICKY.
The other fields in DEV_MM_STATUS are read-only and it doesn't matter
what is written to them, so writing zero is just fine.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_mm.c | 30 ++++++++++++++++++++++++++++--
 include/soc/mscc/ocelot.h             |  1 +
 2 files changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot_mm.c b/drivers/net/ethernet/mscc/ocelot_mm.c
index d2df47e6f8f6..ce6429d46814 100644
--- a/drivers/net/ethernet/mscc/ocelot_mm.c
+++ b/drivers/net/ethernet/mscc/ocelot_mm.c
@@ -54,7 +54,10 @@ static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
 	struct ocelot_mm_state *mm = &ocelot->mm[port];
 	enum ethtool_mm_verify_status verify_status;
-	u32 val;
+	u32 val, ack = 0;
+
+	if (!mm->tx_enabled)
+		return;
 
 	val = ocelot_port_readl(ocelot_port, DEV_MM_STATUS);
 
@@ -71,21 +74,28 @@ static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 
 		dev_dbg(ocelot->dev, "Port %d TX preemption %s\n",
 			port, mm->tx_active ? "active" : "inactive");
+
+		ack |= DEV_MM_STAT_MM_STATUS_PRMPT_ACTIVE_STICKY;
 	}
 
 	if (val & DEV_MM_STAT_MM_STATUS_UNEXP_RX_PFRM_STICKY) {
 		dev_err(ocelot->dev,
 			"Unexpected P-frame received on port %d while verification was unsuccessful or not yet verified\n",
 			port);
+
+		ack |= DEV_MM_STAT_MM_STATUS_UNEXP_RX_PFRM_STICKY;
 	}
 
 	if (val & DEV_MM_STAT_MM_STATUS_UNEXP_TX_PFRM_STICKY) {
 		dev_err(ocelot->dev,
 			"Unexpected P-frame requested to be transmitted on port %d while verification was unsuccessful or not yet verified, or MM_TX_ENA=0\n",
 			port);
+
+		ack |= DEV_MM_STAT_MM_STATUS_UNEXP_TX_PFRM_STICKY;
 	}
 
-	ocelot_port_writel(ocelot_port, val, DEV_MM_STATUS);
+	if (ack)
+		ocelot_port_writel(ocelot_port, ack, DEV_MM_STATUS);
 }
 
 void ocelot_mm_irq(struct ocelot *ocelot)
@@ -107,11 +117,14 @@ int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 {
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
 	u32 mm_enable = 0, verify_disable = 0, add_frag_size;
+	struct ocelot_mm_state *mm;
 	int err;
 
 	if (!ocelot->mm_supported)
 		return -EOPNOTSUPP;
 
+	mm = &ocelot->mm[port];
+
 	err = ethtool_mm_frag_size_min_to_add(cfg->tx_min_frag_size,
 					      &add_frag_size, extack);
 	if (err)
@@ -145,6 +158,19 @@ int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 		       QSYS_PREEMPTION_CFG,
 		       port);
 
+	/* The switch will emit an IRQ when TX is disabled, to notify that it
+	 * has become inactive. We optimize ocelot_mm_update_port_status() to
+	 * not bother processing MM IRQs at all for ports with TX disabled,
+	 * but we need to ACK this IRQ now, while mm->tx_enabled is still set,
+	 * otherwise we get an IRQ storm.
+	 */
+	if (mm->tx_enabled && !cfg->tx_enabled) {
+		ocelot_mm_update_port_status(ocelot, port);
+		WARN_ON(mm->tx_active);
+	}
+
+	mm->tx_enabled = cfg->tx_enabled;
+
 	mutex_unlock(&ocelot->fwd_domain_lock);
 
 	return 0;
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 9599be6a0a39..ee8d43dc5c06 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -745,6 +745,7 @@ struct ocelot_mirror {
 
 struct ocelot_mm_state {
 	enum ethtool_mm_verify_status verify_status;
+	bool tx_enabled;
 	bool tx_active;
 };
 
-- 
cgit v1.2.3-70-g09d2


From bddd96dd80777ec178770a953fdbb310feec29b4 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:48 +0300
Subject: net: mscc: ocelot: don't rely on cached verify_status in
 ocelot_port_get_mm()

ocelot_mm_update_port_status() updates mm->verify_status, but when the
verification state of a port changes, an IRQ isn't emitted, but rather,
only when the verification state reaches one of the final states (like
DISABLED, FAILED, SUCCEEDED) - things that would affect mm->tx_active,
which is what the IRQ *is* actually emitted for.

That is to say, user space may miss reports of an intermediary MAC Merge
verification state (like from INITIAL to VERIFYING), unless there was an
IRQ notifying the driver of the change in mm->tx_active as well.

This is not a huge deal, but for reliable reporting to user space, let's
call ocelot_mm_update_port_status() synchronously from
ocelot_port_get_mm(), which makes user space see the current MM status.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot_mm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mscc/ocelot_mm.c b/drivers/net/ethernet/mscc/ocelot_mm.c
index ce6429d46814..3e458f72f645 100644
--- a/drivers/net/ethernet/mscc/ocelot_mm.c
+++ b/drivers/net/ethernet/mscc/ocelot_mm.c
@@ -205,6 +205,7 @@ int ocelot_port_get_mm(struct ocelot *ocelot, int port,
 	state->tx_min_frag_size = ethtool_mm_frag_size_add_to_min(add_frag_size);
 	state->rx_min_frag_size = ETH_ZLEN;
 
+	ocelot_mm_update_port_status(ocelot, port);
 	state->verify_status = mm->verify_status;
 	state->tx_active = mm->tx_active;
 
-- 
cgit v1.2.3-70-g09d2


From aac80140dc31963d818a65a522c2e2da81979857 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:49 +0300
Subject: net: mscc: ocelot: add support for mqprio offload

This doesn't apply anything to hardware and in general doesn't do
anything that the software variant doesn't do, except for checking that
there isn't more than 1 TXQ per TC (TXQs for a DSA switch are a dubious
concept anyway). The reason we add this is to be able to parse one more
field added to struct tc_mqprio_qopt_offload, namely preemptible_tcs.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c |  9 ++++++
 drivers/net/ethernet/mscc/ocelot.c     | 50 ++++++++++++++++++++++++++++++++++
 include/soc/mscc/ocelot.h              |  4 +++
 3 files changed, 63 insertions(+)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 478893c06f56..66ec2740e3cb 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -1612,6 +1612,13 @@ static int vsc9959_qos_port_cbs_set(struct dsa_switch *ds, int port,
 static int vsc9959_qos_query_caps(struct tc_query_caps_base *base)
 {
 	switch (base->type) {
+	case TC_SETUP_QDISC_MQPRIO: {
+		struct tc_mqprio_caps *caps = base->caps;
+
+		caps->validate_queue_counts = true;
+
+		return 0;
+	}
 	case TC_SETUP_QDISC_TAPRIO: {
 		struct tc_taprio_caps *caps = base->caps;
 
@@ -1635,6 +1642,8 @@ static int vsc9959_port_setup_tc(struct dsa_switch *ds, int port,
 		return vsc9959_qos_query_caps(type_data);
 	case TC_SETUP_QDISC_TAPRIO:
 		return vsc9959_qos_port_tas_set(ocelot, port, type_data);
+	case TC_SETUP_QDISC_MQPRIO:
+		return ocelot_port_mqprio(ocelot, port, type_data);
 	case TC_SETUP_QDISC_CBS:
 		return vsc9959_qos_port_cbs_set(ds, port, type_data);
 	default:
diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 1502bb2c8ea7..8dc5fb1bc61b 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -8,6 +8,7 @@
 #include <linux/if_bridge.h>
 #include <linux/iopoll.h>
 #include <linux/phy/phy.h>
+#include <net/pkt_sched.h>
 #include <soc/mscc/ocelot_hsio.h>
 #include <soc/mscc/ocelot_vcap.h>
 #include "ocelot.h"
@@ -2699,6 +2700,55 @@ void ocelot_port_mirror_del(struct ocelot *ocelot, int from, bool ingress)
 }
 EXPORT_SYMBOL_GPL(ocelot_port_mirror_del);
 
+static void ocelot_port_reset_mqprio(struct ocelot *ocelot, int port)
+{
+	struct net_device *dev = ocelot->ops->port_to_netdev(ocelot, port);
+
+	netdev_reset_tc(dev);
+}
+
+int ocelot_port_mqprio(struct ocelot *ocelot, int port,
+		       struct tc_mqprio_qopt_offload *mqprio)
+{
+	struct net_device *dev = ocelot->ops->port_to_netdev(ocelot, port);
+	struct netlink_ext_ack *extack = mqprio->extack;
+	struct tc_mqprio_qopt *qopt = &mqprio->qopt;
+	int num_tc = qopt->num_tc;
+	int tc, err;
+
+	if (!num_tc) {
+		ocelot_port_reset_mqprio(ocelot, port);
+		return 0;
+	}
+
+	err = netdev_set_num_tc(dev, num_tc);
+	if (err)
+		return err;
+
+	for (tc = 0; tc < num_tc; tc++) {
+		if (qopt->count[tc] != 1) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Only one TXQ per TC supported");
+			return -EINVAL;
+		}
+
+		err = netdev_set_tc_queue(dev, tc, 1, qopt->offset[tc]);
+		if (err)
+			goto err_reset_tc;
+	}
+
+	err = netif_set_real_num_tx_queues(dev, num_tc);
+	if (err)
+		goto err_reset_tc;
+
+	return 0;
+
+err_reset_tc:
+	ocelot_port_reset_mqprio(ocelot, port);
+	return err;
+}
+EXPORT_SYMBOL_GPL(ocelot_port_mqprio);
+
 void ocelot_init_port(struct ocelot *ocelot, int port)
 {
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index ee8d43dc5c06..9596c79e9223 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -11,6 +11,8 @@
 #include <linux/regmap.h>
 #include <net/dsa.h>
 
+struct tc_mqprio_qopt_offload;
+
 /* Port Group IDs (PGID) are masks of destination ports.
  *
  * For L2 forwarding, the switch performs 3 lookups in the PGID table for each
@@ -1154,6 +1156,8 @@ int ocelot_port_set_mm(struct ocelot *ocelot, int port,
 		       struct netlink_ext_ack *extack);
 int ocelot_port_get_mm(struct ocelot *ocelot, int port,
 		       struct ethtool_mm_state *state);
+int ocelot_port_mqprio(struct ocelot *ocelot, int port,
+		       struct tc_mqprio_qopt_offload *mqprio);
 
 #if IS_ENABLED(CONFIG_BRIDGE_MRP)
 int ocelot_mrp_add(struct ocelot *ocelot, int port,
-- 
cgit v1.2.3-70-g09d2


From a1ca9f8b07d865224a6e4b1c41692f10bd95a311 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:50 +0300
Subject: net: dsa: felix: act upon the mqprio qopt in taprio offload

The mqprio queue configuration can appear either through
TC_SETUP_QDISC_MQPRIO or through TC_SETUP_QDISC_TAPRIO. Make sure both
are treated in the same way.

Code does nothing new for now (except for rejecting multiple TXQs per
TC, which is a useless concept with DSA switches).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c | 22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 66ec2740e3cb..e055b3980ccc 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -1424,6 +1424,7 @@ static int vsc9959_qos_port_tas_set(struct ocelot *ocelot, int port,
 	mutex_lock(&ocelot->tas_lock);
 
 	if (!taprio->enable) {
+		ocelot_port_mqprio(ocelot, port, &taprio->mqprio);
 		ocelot_rmw_rix(ocelot, 0, QSYS_TAG_CONFIG_ENABLE,
 			       QSYS_TAG_CONFIG, port);
 
@@ -1436,15 +1437,19 @@ static int vsc9959_qos_port_tas_set(struct ocelot *ocelot, int port,
 		return 0;
 	}
 
+	ret = ocelot_port_mqprio(ocelot, port, &taprio->mqprio);
+	if (ret)
+		goto err_unlock;
+
 	if (taprio->cycle_time > NSEC_PER_SEC ||
 	    taprio->cycle_time_extension >= NSEC_PER_SEC) {
 		ret = -EINVAL;
-		goto err;
+		goto err_reset_tc;
 	}
 
 	if (taprio->num_entries > VSC9959_TAS_GCL_ENTRY_MAX) {
 		ret = -ERANGE;
-		goto err;
+		goto err_reset_tc;
 	}
 
 	/* Enable guard band. The switch will schedule frames without taking
@@ -1468,7 +1473,7 @@ static int vsc9959_qos_port_tas_set(struct ocelot *ocelot, int port,
 	val = ocelot_read(ocelot, QSYS_PARAM_STATUS_REG_8);
 	if (val & QSYS_PARAM_STATUS_REG_8_CONFIG_PENDING) {
 		ret = -EBUSY;
-		goto err;
+		goto err_reset_tc;
 	}
 
 	ocelot_rmw_rix(ocelot,
@@ -1503,12 +1508,19 @@ static int vsc9959_qos_port_tas_set(struct ocelot *ocelot, int port,
 				 !(val & QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE),
 				 10, 100000);
 	if (ret)
-		goto err;
+		goto err_reset_tc;
 
 	ocelot_port->taprio = taprio_offload_get(taprio);
 	vsc9959_tas_guard_bands_update(ocelot, port);
 
-err:
+	mutex_unlock(&ocelot->tas_lock);
+
+	return 0;
+
+err_reset_tc:
+	taprio->mqprio.qopt.num_tc = 0;
+	ocelot_port_mqprio(ocelot, port, &taprio->mqprio);
+err_unlock:
 	mutex_unlock(&ocelot->tas_lock);
 
 	return ret;
-- 
cgit v1.2.3-70-g09d2


From 403ffc2c34de5297d007e0e169bf022094d444c2 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Apr 2023 20:05:51 +0300
Subject: net: mscc: ocelot: add support for preemptible traffic classes

In order to not transmit (preemptible) frames which will be received by
the link partner as corrupted (because it doesn't support FP), the
hardware requires the driver to program the QSYS_PREEMPTION_CFG_P_QUEUES
register only after the MAC Merge layer becomes active (verification
succeeds, or was disabled).

There are some cases when FP is known (through experimentation) to be
broken. Give priority to FP over cut-through switching, and disable FP
for known broken link modes.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c |  7 +++--
 drivers/net/ethernet/mscc/ocelot.c     | 10 ++++++-
 drivers/net/ethernet/mscc/ocelot.h     |  3 ++
 drivers/net/ethernet/mscc/ocelot_mm.c  | 54 ++++++++++++++++++++++++++++++++++
 include/soc/mscc/ocelot.h              |  3 ++
 5 files changed, 74 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index e055b3980ccc..cfb3faeaa5bf 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2519,6 +2519,7 @@ static void vsc9959_cut_through_fwd(struct ocelot *ocelot)
 
 	for (port = 0; port < ocelot->num_phys_ports; port++) {
 		struct ocelot_port *ocelot_port = ocelot->ports[port];
+		struct ocelot_mm_state *mm = &ocelot->mm[port];
 		int min_speed = ocelot_port->speed;
 		unsigned long mask = 0;
 		u32 tmp, val = 0;
@@ -2559,10 +2560,12 @@ static void vsc9959_cut_through_fwd(struct ocelot *ocelot)
 
 		/* Enable cut-through forwarding for all traffic classes that
 		 * don't have oversized dropping enabled, since this check is
-		 * bypassed in cut-through mode.
+		 * bypassed in cut-through mode. Also exclude preemptible
+		 * traffic classes, since these would hang the port for some
+		 * reason, if sent as cut-through.
 		 */
 		if (ocelot_port->speed == min_speed) {
-			val = GENMASK(7, 0);
+			val = GENMASK(7, 0) & ~mm->active_preemptible_tcs;
 
 			for (tc = 0; tc < OCELOT_NUM_TC; tc++)
 				if (vsc9959_port_qmaxsdu_get(ocelot, port, tc))
diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 8dc5fb1bc61b..1f5f00b30441 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -1006,7 +1006,12 @@ void ocelot_phylink_mac_link_up(struct ocelot *ocelot, int port,
 	 */
 	if (ocelot->ops->cut_through_fwd) {
 		mutex_lock(&ocelot->fwd_domain_lock);
-		ocelot->ops->cut_through_fwd(ocelot);
+		/* Workaround for hardware bug - FP doesn't work
+		 * at all link speeds for all PHY modes. The function
+		 * below also calls ocelot->ops->cut_through_fwd(),
+		 * so we don't need to do it twice.
+		 */
+		ocelot_port_update_active_preemptible_tcs(ocelot, port);
 		mutex_unlock(&ocelot->fwd_domain_lock);
 	}
 
@@ -2705,6 +2710,7 @@ static void ocelot_port_reset_mqprio(struct ocelot *ocelot, int port)
 	struct net_device *dev = ocelot->ops->port_to_netdev(ocelot, port);
 
 	netdev_reset_tc(dev);
+	ocelot_port_change_fp(ocelot, port, 0);
 }
 
 int ocelot_port_mqprio(struct ocelot *ocelot, int port,
@@ -2741,6 +2747,8 @@ int ocelot_port_mqprio(struct ocelot *ocelot, int port,
 	if (err)
 		goto err_reset_tc;
 
+	ocelot_port_change_fp(ocelot, port, mqprio->preemptible_tcs);
+
 	return 0;
 
 err_reset_tc:
diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h
index d920ca930690..14440a3b04c3 100644
--- a/drivers/net/ethernet/mscc/ocelot.h
+++ b/drivers/net/ethernet/mscc/ocelot.h
@@ -119,6 +119,9 @@ int ocelot_stats_init(struct ocelot *ocelot);
 void ocelot_stats_deinit(struct ocelot *ocelot);
 
 int ocelot_mm_init(struct ocelot *ocelot);
+void ocelot_port_change_fp(struct ocelot *ocelot, int port,
+			   unsigned long preemptible_tcs);
+void ocelot_port_update_active_preemptible_tcs(struct ocelot *ocelot, int port);
 
 extern struct notifier_block ocelot_netdevice_nb;
 extern struct notifier_block ocelot_switchdev_nb;
diff --git a/drivers/net/ethernet/mscc/ocelot_mm.c b/drivers/net/ethernet/mscc/ocelot_mm.c
index 3e458f72f645..fb3145118d68 100644
--- a/drivers/net/ethernet/mscc/ocelot_mm.c
+++ b/drivers/net/ethernet/mscc/ocelot_mm.c
@@ -49,6 +49,59 @@ static enum ethtool_mm_verify_status ocelot_mm_verify_status(u32 val)
 	}
 }
 
+void ocelot_port_update_active_preemptible_tcs(struct ocelot *ocelot, int port)
+{
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
+	struct ocelot_mm_state *mm = &ocelot->mm[port];
+	u32 val = 0;
+
+	lockdep_assert_held(&ocelot->fwd_domain_lock);
+
+	/* Only commit preemptible TCs when MAC Merge is active.
+	 * On NXP LS1028A, when using QSGMII, the port hangs if transmitting
+	 * preemptible frames at any other link speed than gigabit, so avoid
+	 * preemption at lower speeds in this PHY mode.
+	 */
+	if ((ocelot_port->phy_mode != PHY_INTERFACE_MODE_QSGMII ||
+	     ocelot_port->speed == SPEED_1000) && mm->tx_active)
+		val = mm->preemptible_tcs;
+
+	/* Cut through switching doesn't work for preemptible priorities,
+	 * so first make sure it is disabled.
+	 */
+	mm->active_preemptible_tcs = val;
+	ocelot->ops->cut_through_fwd(ocelot);
+
+	dev_dbg(ocelot->dev,
+		"port %d %s/%s, MM TX %s, preemptible TCs 0x%x, active 0x%x\n",
+		port, phy_modes(ocelot_port->phy_mode),
+		phy_speed_to_str(ocelot_port->speed),
+		mm->tx_active ? "active" : "inactive", mm->preemptible_tcs,
+		mm->active_preemptible_tcs);
+
+	ocelot_rmw_rix(ocelot, QSYS_PREEMPTION_CFG_P_QUEUES(val),
+		       QSYS_PREEMPTION_CFG_P_QUEUES_M,
+		       QSYS_PREEMPTION_CFG, port);
+}
+
+void ocelot_port_change_fp(struct ocelot *ocelot, int port,
+			   unsigned long preemptible_tcs)
+{
+	struct ocelot_mm_state *mm = &ocelot->mm[port];
+
+	mutex_lock(&ocelot->fwd_domain_lock);
+
+	if (mm->preemptible_tcs == preemptible_tcs)
+		goto out_unlock;
+
+	mm->preemptible_tcs = preemptible_tcs;
+
+	ocelot_port_update_active_preemptible_tcs(ocelot, port);
+
+out_unlock:
+	mutex_unlock(&ocelot->fwd_domain_lock);
+}
+
 static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 {
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
@@ -74,6 +127,7 @@ static void ocelot_mm_update_port_status(struct ocelot *ocelot, int port)
 
 		dev_dbg(ocelot->dev, "Port %d TX preemption %s\n",
 			port, mm->tx_active ? "active" : "inactive");
+		ocelot_port_update_active_preemptible_tcs(ocelot, port);
 
 		ack |= DEV_MM_STAT_MM_STATUS_PRMPT_ACTIVE_STICKY;
 	}
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 9596c79e9223..cb8fbb241879 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -749,6 +749,8 @@ struct ocelot_mm_state {
 	enum ethtool_mm_verify_status verify_status;
 	bool tx_enabled;
 	bool tx_active;
+	u8 preemptible_tcs;
+	u8 active_preemptible_tcs;
 };
 
 struct ocelot_port;
@@ -1158,6 +1160,7 @@ int ocelot_port_get_mm(struct ocelot *ocelot, int port,
 		       struct ethtool_mm_state *state);
 int ocelot_port_mqprio(struct ocelot *ocelot, int port,
 		       struct tc_mqprio_qopt_offload *mqprio);
+void ocelot_port_update_preemptible_tcs(struct ocelot *ocelot, int port);
 
 #if IS_ENABLED(CONFIG_BRIDGE_MRP)
 int ocelot_mrp_add(struct ocelot *ocelot, int port,
-- 
cgit v1.2.3-70-g09d2


From cb18e5595df7f4e01d0dd4a0f9c4e71b68ae351e Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 17 Apr 2023 11:36:15 +0200
Subject: net: add macro netif_subqueue_completed_wake

Add netif_subqueue_completed_wake, complementing the subqueue versions
netif_subqueue_try_stop and netif_subqueue_maybe_stop.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/netdev_queues.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
index b26fdb441e39..d68b0a483431 100644
--- a/include/net/netdev_queues.h
+++ b/include/net/netdev_queues.h
@@ -160,4 +160,14 @@ netdev_txq_completed_mb(struct netdev_queue *dev_queue,
 		netif_txq_maybe_stop(txq, get_desc, stop_thrs, start_thrs); \
 	})
 
+#define netif_subqueue_completed_wake(dev, idx, pkts, bytes,		\
+				      get_desc, start_thrs)		\
+	({								\
+		struct netdev_queue *txq;				\
+									\
+		txq = netdev_get_tx_queue(dev, idx);			\
+		netif_txq_completed_wake(txq, pkts, bytes,		\
+					 get_desc, start_thrs);		\
+	})
+
 #endif
-- 
cgit v1.2.3-70-g09d2


From 8624e9bbef641731e1b305c71187cb5dd63673ab Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 17 Apr 2023 11:37:20 +0200
Subject: r8169: use new macro netif_subqueue_maybe_stop in rtl8169_start_xmit

Use new net core macro netif_subqueue_maybe_stop in the start_xmit path
to simplify the code. Whilst at it, set the tx queue start threshold to
twice the stop threshold. Before values were the same, resulting in
stopping/starting the queue more often than needed.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/realtek/r8169_main.c | 39 +++++++++----------------------
 1 file changed, 11 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 9f8357bbc8a4..fff44d46bce1 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -30,6 +30,7 @@
 #include <linux/ipv6.h>
 #include <asm/unaligned.h>
 #include <net/ip6_checksum.h>
+#include <net/netdev_queues.h>
 
 #include "r8169.h"
 #include "r8169_firmware.h"
@@ -68,6 +69,8 @@
 #define NUM_RX_DESC	256	/* Number of Rx descriptor registers */
 #define R8169_TX_RING_BYTES	(NUM_TX_DESC * sizeof(struct TxDesc))
 #define R8169_RX_RING_BYTES	(NUM_RX_DESC * sizeof(struct RxDesc))
+#define R8169_TX_STOP_THRS	(MAX_SKB_FRAGS + 1)
+#define R8169_TX_START_THRS	(2 * R8169_TX_STOP_THRS)
 
 #define OCP_STD_PHY_BASE	0xa400
 
@@ -4162,13 +4165,9 @@ static bool rtl8169_tso_csum_v2(struct rtl8169_private *tp,
 	return true;
 }
 
-static bool rtl_tx_slots_avail(struct rtl8169_private *tp)
+static unsigned int rtl_tx_slots_avail(struct rtl8169_private *tp)
 {
-	unsigned int slots_avail = READ_ONCE(tp->dirty_tx) + NUM_TX_DESC
-					- READ_ONCE(tp->cur_tx);
-
-	/* A skbuff with nr_frags needs nr_frags+1 entries in the tx queue */
-	return slots_avail > MAX_SKB_FRAGS;
+	return READ_ONCE(tp->dirty_tx) + NUM_TX_DESC - READ_ONCE(tp->cur_tx);
 }
 
 /* Versions RTL8102e and from RTL8168c onwards support csum_v2 */
@@ -4245,27 +4244,10 @@ static netdev_tx_t rtl8169_start_xmit(struct sk_buff *skb,
 
 	WRITE_ONCE(tp->cur_tx, tp->cur_tx + frags + 1);
 
-	stop_queue = !rtl_tx_slots_avail(tp);
-	if (unlikely(stop_queue)) {
-		/* Avoid wrongly optimistic queue wake-up: rtl_tx thread must
-		 * not miss a ring update when it notices a stopped queue.
-		 */
-		smp_wmb();
-		netif_stop_queue(dev);
-		/* Sync with rtl_tx:
-		 * - publish queue status and cur_tx ring index (write barrier)
-		 * - refresh dirty_tx ring index (read barrier).
-		 * May the current thread have a pessimistic view of the ring
-		 * status and forget to wake up queue, a racing rtl_tx thread
-		 * can't.
-		 */
-		smp_mb__after_atomic();
-		if (rtl_tx_slots_avail(tp))
-			netif_start_queue(dev);
-		door_bell = true;
-	}
-
-	if (door_bell)
+	stop_queue = !netif_subqueue_maybe_stop(dev, 0, rtl_tx_slots_avail(tp),
+						R8169_TX_STOP_THRS,
+						R8169_TX_START_THRS);
+	if (door_bell || stop_queue)
 		rtl8169_doorbell(tp);
 
 	return NETDEV_TX_OK;
@@ -4400,7 +4382,8 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp,
 		 * ring status.
 		 */
 		smp_store_mb(tp->dirty_tx, dirty_tx);
-		if (netif_queue_stopped(dev) && rtl_tx_slots_avail(tp))
+		if (netif_queue_stopped(dev) &&
+		    rtl_tx_slots_avail(tp) >= R8169_TX_START_THRS)
 			netif_wake_queue(dev);
 		/*
 		 * 8168 hack: TxPoll requests are lost when the Tx packets are
-- 
cgit v1.2.3-70-g09d2


From 1a31ae00482cb6b271ae62a72c7f0a0ad8f038fc Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 17 Apr 2023 11:38:36 +0200
Subject: r8169: use new macro netif_subqueue_completed_wake in the tx cleanup
 path

Use new net core macro netif_subqueue_completed_wake to simplify
the code of the tx cleanup path.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/realtek/r8169_main.c | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index fff44d46bce1..a7e376e7e689 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -4371,20 +4371,12 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp,
 	}
 
 	if (tp->dirty_tx != dirty_tx) {
-		netdev_completed_queue(dev, pkts_compl, bytes_compl);
 		dev_sw_netstats_tx_add(dev, pkts_compl, bytes_compl);
+		WRITE_ONCE(tp->dirty_tx, dirty_tx);
 
-		/* Sync with rtl8169_start_xmit:
-		 * - publish dirty_tx ring index (write barrier)
-		 * - refresh cur_tx ring index and queue status (read barrier)
-		 * May the current thread miss the stopped queue condition,
-		 * a racing xmit thread can only have a right view of the
-		 * ring status.
-		 */
-		smp_store_mb(tp->dirty_tx, dirty_tx);
-		if (netif_queue_stopped(dev) &&
-		    rtl_tx_slots_avail(tp) >= R8169_TX_START_THRS)
-			netif_wake_queue(dev);
+		netif_subqueue_completed_wake(dev, 0, pkts_compl, bytes_compl,
+					      rtl_tx_slots_avail(tp),
+					      R8169_TX_START_THRS);
 		/*
 		 * 8168 hack: TxPoll requests are lost when the Tx packets are
 		 * too close. Let's kick an extra TxPoll request when a burst
-- 
cgit v1.2.3-70-g09d2


From 13f9351180aaa6cd745303339d2a1968fb5cc47d Mon Sep 17 00:00:00 2001
From: Emil Renner Berthing <kernel@esmil.dk>
Date: Mon, 17 Apr 2023 18:02:46 +0800
Subject: dt-bindings: net: snps,dwmac: Add dwmac-5.20 version

Add dwmac-5.20 IP version to snps.dwmac.yaml

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Emil Renner Berthing <kernel@esmil.dk>
Signed-off-by: Samin Guo <samin.guo@starfivetech.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/snps,dwmac.yaml | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index da311c1f2c88..e5ca96bcbae0 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -30,6 +30,7 @@ select:
           - snps,dwmac-4.10a
           - snps,dwmac-4.20a
           - snps,dwmac-5.10a
+          - snps,dwmac-5.20
           - snps,dwxgmac
           - snps,dwxgmac-2.10
 
@@ -90,6 +91,7 @@ properties:
         - snps,dwmac-4.10a
         - snps,dwmac-4.20a
         - snps,dwmac-5.10a
+        - snps,dwmac-5.20
         - snps,dwxgmac
         - snps,dwxgmac-2.10
 
@@ -579,6 +581,7 @@ allOf:
               - snps,dwmac-3.50a
               - snps,dwmac-4.10a
               - snps,dwmac-4.20a
+              - snps,dwmac-5.20
               - snps,dwxgmac
               - snps,dwxgmac-2.10
               - st,spear600-gmac
@@ -636,6 +639,7 @@ allOf:
               - snps,dwmac-4.10a
               - snps,dwmac-4.20a
               - snps,dwmac-5.10a
+              - snps,dwmac-5.20
               - snps,dwxgmac
               - snps,dwxgmac-2.10
               - st,spear600-gmac
-- 
cgit v1.2.3-70-g09d2


From 65a1d72f0c7cd0ef5ba0fc13224a44bc84e91dc4 Mon Sep 17 00:00:00 2001
From: Emil Renner Berthing <kernel@esmil.dk>
Date: Mon, 17 Apr 2023 18:02:47 +0800
Subject: net: stmmac: platform: Add snps,dwmac-5.20 IP compatible string

Add "snps,dwmac-5.20" compatible string for 5.20 version that can avoid
to define some platform data in the glue layer.

Tested-by: Tommaso Merciai <tomm.merciai@gmail.com>
Signed-off-by: Emil Renner Berthing <kernel@esmil.dk>
Signed-off-by: Samin Guo <samin.guo@starfivetech.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
index 067a40fe0a23..eb0b2898daa3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
@@ -519,7 +519,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac)
 	if (of_device_is_compatible(np, "snps,dwmac-4.00") ||
 	    of_device_is_compatible(np, "snps,dwmac-4.10a") ||
 	    of_device_is_compatible(np, "snps,dwmac-4.20a") ||
-	    of_device_is_compatible(np, "snps,dwmac-5.10a")) {
+	    of_device_is_compatible(np, "snps,dwmac-5.10a") ||
+	    of_device_is_compatible(np, "snps,dwmac-5.20")) {
 		plat->has_gmac4 = 1;
 		plat->has_gmac = 0;
 		plat->pmt = 1;
-- 
cgit v1.2.3-70-g09d2


From 843f603762a5453e10999d89f2adea6a3cfe8735 Mon Sep 17 00:00:00 2001
From: Samin Guo <samin.guo@starfivetech.com>
Date: Mon, 17 Apr 2023 18:02:48 +0800
Subject: dt-bindings: net: snps,dwmac: Add 'ahb' reset/reset-name

According to:
stmmac_platform.c: stmmac_probe_config_dt
stmmac_main.c: stmmac_dvr_probe

dwmac controller may require one (stmmaceth) or two (stmmaceth+ahb)
reset signals, and the maxItems of resets/reset-names is going to be 2.

The gmac of Starfive Jh7110 SOC must have two resets.
it uses snps,dwmac-5.20 IP.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Samin Guo <samin.guo@starfivetech.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/snps,dwmac.yaml | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index e5ca96bcbae0..71151478885a 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -136,12 +136,16 @@ properties:
         - ptp_ref
 
   resets:
-    maxItems: 1
-    description:
-      MAC Reset signal.
+    minItems: 1
+    items:
+      - description: GMAC stmmaceth reset
+      - description: AHB reset
 
   reset-names:
-    const: stmmaceth
+    minItems: 1
+    items:
+      - const: stmmaceth
+      - const: ahb
 
   power-domains:
     maxItems: 1
-- 
cgit v1.2.3-70-g09d2


From b76eaf7d7ede35adbbb914cb2e9223a250448e52 Mon Sep 17 00:00:00 2001
From: Yanhong Wang <yanhong.wang@starfivetech.com>
Date: Mon, 17 Apr 2023 18:02:49 +0800
Subject: dt-bindings: net: Add support StarFive dwmac

Add documentation to describe StarFive dwmac driver(GMAC).

Signed-off-by: Yanhong Wang <yanhong.wang@starfivetech.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Samin Guo <samin.guo@starfivetech.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../devicetree/bindings/net/snps,dwmac.yaml        |   1 +
 .../bindings/net/starfive,jh7110-dwmac.yaml        | 144 +++++++++++++++++++++
 MAINTAINERS                                        |   6 +
 3 files changed, 151 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 71151478885a..363b3e3ea3a6 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -94,6 +94,7 @@ properties:
         - snps,dwmac-5.20
         - snps,dwxgmac
         - snps,dwxgmac-2.10
+        - starfive,jh7110-dwmac
 
   reg:
     minItems: 1
diff --git a/Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml b/Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml
new file mode 100644
index 000000000000..5e7cfbbebce6
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml
@@ -0,0 +1,144 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+# Copyright (C) 2022 StarFive Technology Co., Ltd.
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/starfive,jh7110-dwmac.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: StarFive JH7110 DWMAC glue layer
+
+maintainers:
+  - Emil Renner Berthing <kernel@esmil.dk>
+  - Samin Guo <samin.guo@starfivetech.com>
+
+select:
+  properties:
+    compatible:
+      contains:
+        enum:
+          - starfive,jh7110-dwmac
+  required:
+    - compatible
+
+properties:
+  compatible:
+    items:
+      - enum:
+          - starfive,jh7110-dwmac
+      - const: snps,dwmac-5.20
+
+  reg:
+    maxItems: 1
+
+  clocks:
+    items:
+      - description: GMAC main clock
+      - description: GMAC AHB clock
+      - description: PTP clock
+      - description: TX clock
+      - description: GTX clock
+
+  clock-names:
+    items:
+      - const: stmmaceth
+      - const: pclk
+      - const: ptp_ref
+      - const: tx
+      - const: gtx
+
+  interrupts:
+    minItems: 3
+    maxItems: 3
+
+  interrupt-names:
+    minItems: 3
+    maxItems: 3
+
+  resets:
+    items:
+      - description: MAC Reset signal.
+      - description: AHB Reset signal.
+
+  reset-names:
+    items:
+      - const: stmmaceth
+      - const: ahb
+
+  starfive,tx-use-rgmii-clk:
+    description:
+      Tx clock is provided by external rgmii clock.
+    type: boolean
+
+  starfive,syscon:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    items:
+      - items:
+          - description: phandle to syscon that configures phy mode
+          - description: Offset of phy mode selection
+          - description: Shift of phy mode selection
+    description:
+      A phandle to syscon with two arguments that configure phy mode.
+      The argument one is the offset of phy mode selection, the
+      argument two is the shift of phy mode selection.
+
+required:
+  - compatible
+  - reg
+  - clocks
+  - clock-names
+  - interrupts
+  - interrupt-names
+  - resets
+  - reset-names
+
+allOf:
+  - $ref: snps,dwmac.yaml#
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    ethernet@16030000 {
+        compatible = "starfive,jh7110-dwmac", "snps,dwmac-5.20";
+        reg = <0x16030000 0x10000>;
+        clocks = <&clk 3>, <&clk 2>, <&clk 109>,
+                 <&clk 6>, <&clk 111>;
+        clock-names = "stmmaceth", "pclk", "ptp_ref",
+                      "tx", "gtx";
+        resets = <&rst 1>, <&rst 2>;
+        reset-names = "stmmaceth", "ahb";
+        interrupts = <7>, <6>, <5>;
+        interrupt-names = "macirq", "eth_wake_irq", "eth_lpi";
+        phy-mode = "rgmii-id";
+        snps,multicast-filter-bins = <64>;
+        snps,perfect-filter-entries = <8>;
+        rx-fifo-depth = <2048>;
+        tx-fifo-depth = <2048>;
+        snps,fixed-burst;
+        snps,no-pbl-x8;
+        snps,tso;
+        snps,force_thresh_dma_mode;
+        snps,axi-config = <&stmmac_axi_setup>;
+        snps,en-tx-lpi-clockgating;
+        snps,txpbl = <16>;
+        snps,rxpbl = <16>;
+        starfive,syscon = <&aon_syscon 0xc 0x12>;
+        phy-handle = <&phy0>;
+
+        mdio {
+            #address-cells = <1>;
+            #size-cells = <0>;
+            compatible = "snps,dwmac-mdio";
+
+            phy0: ethernet-phy@0 {
+                reg = <0>;
+            };
+        };
+
+        stmmac_axi_setup: stmmac-axi-config {
+            snps,lpi_en;
+            snps,wr_osr_lmt = <4>;
+            snps,rd_osr_lmt = <4>;
+            snps,blen = <256 128 64 32 0 0 0>;
+        };
+    };
diff --git a/MAINTAINERS b/MAINTAINERS
index 1c09473685b1..4301fe86ad35 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -19928,6 +19928,12 @@ M:	Emil Renner Berthing <kernel@esmil.dk>
 S:	Maintained
 F:	arch/riscv/boot/dts/starfive/
 
+STARFIVE DWMAC GLUE LAYER
+M:	Emil Renner Berthing <kernel@esmil.dk>
+M:	Samin Guo <samin.guo@starfivetech.com>
+S:	Maintained
+F:	Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml
+
 STARFIVE JH7100 CLOCK DRIVERS
 M:	Emil Renner Berthing <kernel@esmil.dk>
 S:	Maintained
-- 
cgit v1.2.3-70-g09d2


From 4bd3bb7b452690a939c3f327104a557e1b9876b7 Mon Sep 17 00:00:00 2001
From: Samin Guo <samin.guo@starfivetech.com>
Date: Mon, 17 Apr 2023 18:02:50 +0800
Subject: net: stmmac: Add glue layer for StarFive JH7110 SoC

This adds StarFive dwmac driver support on the StarFive JH7110 SoC.

Tested-by: Tommaso Merciai <tomm.merciai@gmail.com>
Co-developed-by: Emil Renner Berthing <kernel@esmil.dk>
Signed-off-by: Emil Renner Berthing <kernel@esmil.dk>
Signed-off-by: Samin Guo <samin.guo@starfivetech.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 MAINTAINERS                                        |   1 +
 drivers/net/ethernet/stmicro/stmmac/Kconfig        |  12 ++
 drivers/net/ethernet/stmicro/stmmac/Makefile       |   1 +
 .../net/ethernet/stmicro/stmmac/dwmac-starfive.c   | 123 +++++++++++++++++++++
 4 files changed, 137 insertions(+)
 create mode 100644 drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 4301fe86ad35..4fc57dfd5fd0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -19933,6 +19933,7 @@ M:	Emil Renner Berthing <kernel@esmil.dk>
 M:	Samin Guo <samin.guo@starfivetech.com>
 S:	Maintained
 F:	Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml
+F:	drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c
 
 STARFIVE JH7100 CLOCK DRIVERS
 M:	Emil Renner Berthing <kernel@esmil.dk>
diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig
index f77511fe4e87..5f5a997f21f3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/Kconfig
+++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig
@@ -165,6 +165,18 @@ config DWMAC_SOCFPGA
 	  for the stmmac device driver. This driver is used for
 	  arria5 and cyclone5 FPGA SoCs.
 
+config DWMAC_STARFIVE
+	tristate "StarFive dwmac support"
+	depends on OF && (ARCH_STARFIVE || COMPILE_TEST)
+	select MFD_SYSCON
+	default m if ARCH_STARFIVE
+	help
+	  Support for ethernet controllers on StarFive RISC-V SoCs
+
+	  This selects the StarFive platform specific glue layer support for
+	  the stmmac device driver. This driver is used for StarFive JH7110
+	  ethernet controller.
+
 config DWMAC_STI
 	tristate "STi GMAC support"
 	default ARCH_STI
diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile
index 057e4bab5c08..8738fdbb4b2d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/Makefile
+++ b/drivers/net/ethernet/stmicro/stmmac/Makefile
@@ -23,6 +23,7 @@ obj-$(CONFIG_DWMAC_OXNAS)	+= dwmac-oxnas.o
 obj-$(CONFIG_DWMAC_QCOM_ETHQOS)	+= dwmac-qcom-ethqos.o
 obj-$(CONFIG_DWMAC_ROCKCHIP)	+= dwmac-rk.o
 obj-$(CONFIG_DWMAC_SOCFPGA)	+= dwmac-altr-socfpga.o
+obj-$(CONFIG_DWMAC_STARFIVE)	+= dwmac-starfive.o
 obj-$(CONFIG_DWMAC_STI)		+= dwmac-sti.o
 obj-$(CONFIG_DWMAC_STM32)	+= dwmac-stm32.o
 obj-$(CONFIG_DWMAC_SUNXI)	+= dwmac-sunxi.o
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c
new file mode 100644
index 000000000000..4963d4008485
--- /dev/null
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c
@@ -0,0 +1,123 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * StarFive DWMAC platform driver
+ *
+ * Copyright (C) 2021 Emil Renner Berthing <kernel@esmil.dk>
+ * Copyright (C) 2022 StarFive Technology Co., Ltd.
+ *
+ */
+
+#include <linux/mfd/syscon.h>
+#include <linux/of_device.h>
+#include <linux/regmap.h>
+
+#include "stmmac_platform.h"
+
+struct starfive_dwmac {
+	struct device *dev;
+	struct clk *clk_tx;
+};
+
+static void starfive_dwmac_fix_mac_speed(void *priv, unsigned int speed)
+{
+	struct starfive_dwmac *dwmac = priv;
+	unsigned long rate;
+	int err;
+
+	rate = clk_get_rate(dwmac->clk_tx);
+
+	switch (speed) {
+	case SPEED_1000:
+		rate = 125000000;
+		break;
+	case SPEED_100:
+		rate = 25000000;
+		break;
+	case SPEED_10:
+		rate = 2500000;
+		break;
+	default:
+		dev_err(dwmac->dev, "invalid speed %u\n", speed);
+		break;
+	}
+
+	err = clk_set_rate(dwmac->clk_tx, rate);
+	if (err)
+		dev_err(dwmac->dev, "failed to set tx rate %lu\n", rate);
+}
+
+static int starfive_dwmac_probe(struct platform_device *pdev)
+{
+	struct plat_stmmacenet_data *plat_dat;
+	struct stmmac_resources stmmac_res;
+	struct starfive_dwmac *dwmac;
+	struct clk *clk_gtx;
+	int err;
+
+	err = stmmac_get_platform_resources(pdev, &stmmac_res);
+	if (err)
+		return dev_err_probe(&pdev->dev, err,
+				     "failed to get resources\n");
+
+	plat_dat = stmmac_probe_config_dt(pdev, stmmac_res.mac);
+	if (IS_ERR(plat_dat))
+		return dev_err_probe(&pdev->dev, PTR_ERR(plat_dat),
+				     "dt configuration failed\n");
+
+	dwmac = devm_kzalloc(&pdev->dev, sizeof(*dwmac), GFP_KERNEL);
+	if (!dwmac)
+		return -ENOMEM;
+
+	dwmac->clk_tx = devm_clk_get_enabled(&pdev->dev, "tx");
+	if (IS_ERR(dwmac->clk_tx))
+		return dev_err_probe(&pdev->dev, PTR_ERR(dwmac->clk_tx),
+				     "error getting tx clock\n");
+
+	clk_gtx = devm_clk_get_enabled(&pdev->dev, "gtx");
+	if (IS_ERR(clk_gtx))
+		return dev_err_probe(&pdev->dev, PTR_ERR(clk_gtx),
+				     "error getting gtx clock\n");
+
+	/* Generally, the rgmii_tx clock is provided by the internal clock,
+	 * which needs to match the corresponding clock frequency according
+	 * to different speeds. If the rgmii_tx clock is provided by the
+	 * external rgmii_rxin, there is no need to configure the clock
+	 * internally, because rgmii_rxin will be adaptively adjusted.
+	 */
+	if (!device_property_read_bool(&pdev->dev, "starfive,tx-use-rgmii-clk"))
+		plat_dat->fix_mac_speed = starfive_dwmac_fix_mac_speed;
+
+	dwmac->dev = &pdev->dev;
+	plat_dat->bsp_priv = dwmac;
+	plat_dat->dma_cfg->dche = true;
+
+	err = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res);
+	if (err) {
+		stmmac_remove_config_dt(pdev, plat_dat);
+		return err;
+	}
+
+	return 0;
+}
+
+static const struct of_device_id starfive_dwmac_match[] = {
+	{ .compatible = "starfive,jh7110-dwmac"	},
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, starfive_dwmac_match);
+
+static struct platform_driver starfive_dwmac_driver = {
+	.probe  = starfive_dwmac_probe,
+	.remove = stmmac_pltfr_remove,
+	.driver = {
+		.name = "starfive-dwmac",
+		.pm = &stmmac_pltfr_pm_ops,
+		.of_match_table = starfive_dwmac_match,
+	},
+};
+module_platform_driver(starfive_dwmac_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("StarFive DWMAC platform driver");
+MODULE_AUTHOR("Emil Renner Berthing <kernel@esmil.dk>");
+MODULE_AUTHOR("Samin Guo <samin.guo@starfivetech.com>");
-- 
cgit v1.2.3-70-g09d2


From b4a5afa51ceecab128b301da3896f0d0110ec347 Mon Sep 17 00:00:00 2001
From: Samin Guo <samin.guo@starfivetech.com>
Date: Mon, 17 Apr 2023 18:02:51 +0800
Subject: net: stmmac: dwmac-starfive: Add phy interface settings

dwmac supports multiple modess. When working under rmii and rgmii,
you need to set different phy interfaces.

According to the dwmac document, when working in rmii, it needs to be
set to 0x4, and rgmii needs to be set to 0x1.

The phy interface needs to be set in syscon, the format is as follows:
starfive,syscon: <&syscon, offset, shift>

Tested-by: Tommaso Merciai <tomm.merciai@gmail.com>
Signed-off-by: Samin Guo <samin.guo@starfivetech.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../net/ethernet/stmicro/stmmac/dwmac-starfive.c   | 48 ++++++++++++++++++++++
 1 file changed, 48 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c
index 4963d4008485..4f51a7889642 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c
@@ -13,6 +13,10 @@
 
 #include "stmmac_platform.h"
 
+#define STARFIVE_DWMAC_PHY_INFT_RGMII	0x1
+#define STARFIVE_DWMAC_PHY_INFT_RMII	0x4
+#define STARFIVE_DWMAC_PHY_INFT_FIELD	0x7U
+
 struct starfive_dwmac {
 	struct device *dev;
 	struct clk *clk_tx;
@@ -46,6 +50,46 @@ static void starfive_dwmac_fix_mac_speed(void *priv, unsigned int speed)
 		dev_err(dwmac->dev, "failed to set tx rate %lu\n", rate);
 }
 
+static int starfive_dwmac_set_mode(struct plat_stmmacenet_data *plat_dat)
+{
+	struct starfive_dwmac *dwmac = plat_dat->bsp_priv;
+	struct regmap *regmap;
+	unsigned int args[2];
+	unsigned int mode;
+	int err;
+
+	switch (plat_dat->interface) {
+	case PHY_INTERFACE_MODE_RMII:
+		mode = STARFIVE_DWMAC_PHY_INFT_RMII;
+		break;
+
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+		mode = STARFIVE_DWMAC_PHY_INFT_RGMII;
+		break;
+
+	default:
+		dev_err(dwmac->dev, "unsupported interface %d\n",
+			plat_dat->interface);
+		return -EINVAL;
+	}
+
+	regmap = syscon_regmap_lookup_by_phandle_args(dwmac->dev->of_node,
+						      "starfive,syscon",
+						      2, args);
+	if (IS_ERR(regmap))
+		return dev_err_probe(dwmac->dev, PTR_ERR(regmap), "getting the regmap failed\n");
+
+	/* args[0]:offset  args[1]: shift */
+	err = regmap_update_bits(regmap, args[0],
+				 STARFIVE_DWMAC_PHY_INFT_FIELD << args[1],
+				 mode << args[1]);
+	if (err)
+		return dev_err_probe(dwmac->dev, err, "error setting phy mode\n");
+
+	return 0;
+}
+
 static int starfive_dwmac_probe(struct platform_device *pdev)
 {
 	struct plat_stmmacenet_data *plat_dat;
@@ -91,6 +135,10 @@ static int starfive_dwmac_probe(struct platform_device *pdev)
 	plat_dat->bsp_priv = dwmac;
 	plat_dat->dma_cfg->dche = true;
 
+	err = starfive_dwmac_set_mode(plat_dat);
+	if (err)
+		return err;
+
 	err = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res);
 	if (err) {
 		stmmac_remove_config_dt(pdev, plat_dat);
-- 
cgit v1.2.3-70-g09d2


From 3ec6697ec974391ae3e72c89a347945b82b7268c Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Mon, 17 Apr 2023 11:41:20 +0300
Subject: wifi: iwlwifi: mvm: adopt the latest firmware API

The firmware no longer wants the beacon template inside the MAC command.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.92aed4180a06.I277efa343c88081cb3fc890dcbeae3161cdffe16@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h | 10 ----------
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c    |  7 -------
 2 files changed, 17 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index 0bafa3e21cb5..74f2efbad34e 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -223,15 +223,6 @@ struct iwl_mac_client_data {
 	__le32 ctwin;
 } __packed; /* MAC_CONTEXT_CONFIG_CLIENT_DATA_API_S_VER_1 */
 
-/**
- * struct iwl_mac_go_ibss_data - configuration data for GO and IBSS MAC context
- *
- * @beacon_template: beacon template ID
- */
-struct iwl_mac_go_ibss_data {
-	__le32 beacon_template;
-} __packed; /* MAC_CONTEXT_CONFIG_GO_IBSS_DATA_API_S_VER_1 */
-
 /**
  * struct iwl_mac_p2p_dev_data  - configuration data for P2P device MAC context
  *
@@ -304,7 +295,6 @@ struct iwl_mac_config_cmd {
 	/* MAC_CONTEXT_CONFIG_SPECIFIC_DATA_API_U_VER_1 */
 	union {
 		struct iwl_mac_client_data client;
-		struct iwl_mac_go_ibss_data go_ibss;
 		struct iwl_mac_p2p_dev_data p2p_dev;
 	};
 } __packed; /* MAC_CONTEXT_CONFIG_CMD_API_S_VER_1 */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
index d73ab1712e0e..1717fb52dc12 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac.c
@@ -167,7 +167,6 @@ static int iwl_mvm_mld_mac_ctxt_cmd_ibss(struct iwl_mvm *mvm,
 					 struct ieee80211_vif *vif,
 					 u32 action)
 {
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mac_config_cmd cmd = {};
 
 	WARN_ON(vif->type != NL80211_IFTYPE_ADHOC);
@@ -178,9 +177,6 @@ static int iwl_mvm_mld_mac_ctxt_cmd_ibss(struct iwl_mvm *mvm,
 				       MAC_CFG_FILTER_ACCEPT_PROBE_REQ |
 				       MAC_CFG_FILTER_ACCEPT_GRP);
 
-	/* TODO: Assumes that the beacon id == mac context id */
-	cmd.go_ibss.beacon_template = cpu_to_le32(mvmvif->id);
-
 	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
 }
 
@@ -220,9 +216,6 @@ static int iwl_mvm_mld_mac_ctxt_cmd_ap_go(struct iwl_mvm *mvm,
 						 MAC_CFG_FILTER_ACCEPT_PROBE_REQ,
 						 MAC_CFG_FILTER_ACCEPT_BEACON);
 
-	/* TODO: Assume that the beacon id == mac context id */
-	cmd.go_ibss.beacon_template = cpu_to_le32(mvmvif->id);
-
 	return iwl_mvm_mld_mac_ctxt_send_cmd(mvm, &cmd);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 95a35ec7b9dde6dc90b0876b9b25fd2e0c31162b Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Mon, 17 Apr 2023 11:41:21 +0300
Subject: wifi: iwlwifi: mvm: update mac id management

The restriction where MAC ID 0 could be used only for the
managed/IBSS vif is not required when using the new MLO FW API.
Update the driver.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.e4355615da92.Iba934ccf8589c3c27a25a390dc5e938312889b45@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 389eef453d17..cc90f2884cff 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -225,16 +225,20 @@ int iwl_mvm_mac_ctxt_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	 * that we should share it with another interface.
 	 */
 
-	/* Currently, MAC ID 0 should be used only for the managed/IBSS vif */
-	switch (vif->type) {
-	case NL80211_IFTYPE_ADHOC:
-		break;
-	case NL80211_IFTYPE_STATION:
-		if (!vif->p2p)
+	/* MAC ID 0 should be used only for the managed/IBSS vif with non-MLO
+	 * FW API
+	 */
+	if (!mvm->mld_api_is_used) {
+		switch (vif->type) {
+		case NL80211_IFTYPE_ADHOC:
 			break;
-		fallthrough;
-	default:
-		__clear_bit(0, data.available_mac_ids);
+		case NL80211_IFTYPE_STATION:
+			if (!vif->p2p)
+				break;
+			fallthrough;
+		default:
+			__clear_bit(0, data.available_mac_ids);
+		}
 	}
 
 	ieee80211_iterate_active_interfaces_atomic(
-- 
cgit v1.2.3-70-g09d2


From f1fec51cda70f1ac83b55f1f7292ee09318c3c0d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:22 +0300
Subject: wifi: iwlwifi: mvm: use BSSID when building probe requests

A long time, ago in commit 818965d39177 ("cfg80211: Allow a
scan request for a specific BSSID"), the stack started passing
the BSSID that should be scanned for. Use it in iwlwifi for the
intended optimisation, and to also allow the use of this for an
implementation sending multi-link probe requests.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.0be90360cad7.If279c28079a1db34280a824cee7c3f6545fd8b9e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 38b68827a4bd..1e20f9538640 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -98,6 +98,7 @@ struct iwl_mvm_scan_params {
 	bool scan_6ghz;
 	bool enable_6ghz_passive;
 	bool respect_p2p_go, respect_p2p_go_hb;
+	u8 bssid[ETH_ALEN] __aligned(2);
 };
 
 static inline void *iwl_mvm_get_scan_req_umac_data(struct iwl_mvm *mvm)
@@ -760,7 +761,7 @@ iwl_mvm_build_scan_probe(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	frame->frame_control = cpu_to_le16(IEEE80211_STYPE_PROBE_REQ);
 	eth_broadcast_addr(frame->da);
-	eth_broadcast_addr(frame->bssid);
+	ether_addr_copy(frame->bssid, params->bssid);
 	frame->seq_ctrl = 0;
 
 	pos = frame->u.probe_req.variable;
@@ -2801,6 +2802,7 @@ int iwl_mvm_reg_scan_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	params.pass_all = true;
 	params.n_match_sets = 0;
 	params.match_sets = NULL;
+	ether_addr_copy(params.bssid, req->bssid);
 
 	params.scan_plans = &scan_plan;
 	params.n_scan_plans = 1;
@@ -2894,6 +2896,7 @@ int iwl_mvm_sched_scan_start(struct iwl_mvm *mvm,
 	params.pass_all =  iwl_mvm_scan_pass_all(mvm, req);
 	params.n_match_sets = req->n_match_sets;
 	params.match_sets = req->match_sets;
+	eth_broadcast_addr(params.bssid);
 	if (!req->n_scan_plans)
 		return -EINVAL;
 
-- 
cgit v1.2.3-70-g09d2


From d16b96b5fd88b75db6fbbe88a8d5981557329581 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:23 +0300
Subject: wifi: iwlwifi: mvm: allow NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT

We can allow NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT since we
just use the elements from mac80211.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.88b1a70365fd.If5030437707ab67e2146291c1517a9b0e31d01ab@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 718b5ba6ec92..b086229f339d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -413,6 +413,8 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 
 	wiphy_ext_feature_set(hw->wiphy,
 			      NL80211_EXT_FEATURE_BEACON_RATE_LEGACY);
+	wiphy_ext_feature_set(hw->wiphy,
+			      NL80211_EXT_FEATURE_SCAN_MIN_PREQ_CONTENT);
 
 	if (fw_has_capa(&mvm->fw->ucode_capa,
 			IWL_UCODE_TLV_CAPA_FTM_CALIBRATED)) {
-- 
cgit v1.2.3-70-g09d2


From 84f650e6323cfdb44aaa94adbab7480435427cf6 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:24 +0300
Subject: wifi: iwlwifi: mvm: remove per-STA MFP setting

When we first add the STA before sending an authentication
frame this is false. However, in some cases such as FT or
certainly in MLO link switch scenarios, it will be true.

If it happens to be true, it causes a firmware assert (in
LMAC, 0x2528), because the same bit is used by the firmware
for tracking this setting as well as for tracking if the
management key has been installed, and then we get this
assert from the firmware when installing the MFP key as it
thinks it has already been installed.

Remove the setting for now, until the firmware disentangles
the two. We should be able to set it, in fact we should be
setting it speculatively before authentication/association,
to avoid processing management frames that should have been
protected/encrypted before the key is set.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.adbeb0e0bfed.I400d0ee3721dc4f294313be992d277ba4d9c88d9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 65436736f87f..fe362f621777 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -427,7 +427,6 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	struct iwl_mvm_sta_cfg_cmd cmd = {
 		.sta_id = cpu_to_le32(mvm_link_sta->sta_id),
 		.station_type = cpu_to_le32(mvm_sta->sta_type),
-		.mfp = cpu_to_le32(sta->mfp),
 	};
 	u32 agg_size = 0, mpdu_dens = 0;
 
-- 
cgit v1.2.3-70-g09d2


From a705a78281cad4665e4092a09f8c485e81b47882 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:25 +0300
Subject: wifi: iwlwifi: mvm: fix iwl_mvm_sta_rc_update for MLO

When this is called it might crash due to the use of
deflink's phy context, update all links instead. It
really shouldn't be called right now though, but it's
better to have safer code until we update this to be
with a link parameter.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.fd47d0de8319.I87c5e5bcb2fadd70acc32021eed394fc1eea12a4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index b086229f339d..9d5d2c6625a9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3585,9 +3585,9 @@ static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
 
 	for_each_mvm_vif_valid_link(mvmvif, link_id) {
 		struct ieee80211_bss_conf *conf =
-			link_conf_dereference_protected(vif, link_id);
+			link_conf_dereference_check(vif, link_id);
 		struct ieee80211_link_sta *link_sta =
-			link_sta_dereference_protected(sta, link_id);
+			link_sta_dereference_check(sta, link_id);
 
 		if (!conf || !link_sta || !mvmvif->link[link_id]->phy_ctxt)
 			continue;
@@ -3983,15 +3983,11 @@ void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta, u32 changed)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
 	if (changed & (IEEE80211_RC_BW_CHANGED |
 		       IEEE80211_RC_SUPP_RATES_CHANGED |
 		       IEEE80211_RC_NSS_CHANGED))
-		iwl_mvm_rs_rate_init(mvm, sta,
-				     &vif->bss_conf, &sta->deflink,
-				     mvmvif->deflink.phy_ctxt->channel->band,
-				     true);
+		iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, true);
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
 	    changed & IEEE80211_RC_NSS_CHANGED)
-- 
cgit v1.2.3-70-g09d2


From c45217bd3f2e01f89f2afe6ee28151df12420f65 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:26 +0300
Subject: wifi: iwlwifi: mvm: only clients can be 20MHz-only

Since only clients to an AP can be 20MHz-only STAs, adjust the
code to not make the use of EHT capabilities depend on only the
bandwidth, but also the type of interface.

Fixes: 701404f1091d ("wifi: iwlwifi: rs: add support for parsing max MCS per NSS/BW in 11be")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.a3a4c931e4a3.I693a07f4d88044c889eee04793883a83bc5ee362@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c    | 17 +++++++++++------
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c       |  9 ++++++---
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h       |  8 ++++++--
 4 files changed, 24 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 9d5d2c6625a9..8808bb3983a4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3592,7 +3592,7 @@ static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
 		if (!conf || !link_sta || !mvmvif->link[link_id]->phy_ctxt)
 			continue;
 
-		iwl_mvm_rs_rate_init(mvm, sta, conf, link_sta,
+		iwl_mvm_rs_rate_init(mvm, vif, sta, conf, link_sta,
 				     mvmvif->link[link_id]->phy_ctxt->channel->band,
 				     update);
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index c8ba2fe3e4a2..43d26a09cfce 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -283,7 +283,8 @@ rs_fw_rs_mcs2eht_mcs(enum IWL_TLC_MCS_PER_BW bw,
 }
 
 static void
-rs_fw_eht_set_enabled_rates(const struct ieee80211_link_sta *link_sta,
+rs_fw_eht_set_enabled_rates(struct ieee80211_vif *vif,
+			    const struct ieee80211_link_sta *link_sta,
 			    struct ieee80211_supported_band *sband,
 			    struct iwl_tlc_config_cmd_v4 *cmd)
 {
@@ -299,7 +300,8 @@ rs_fw_eht_set_enabled_rates(const struct ieee80211_link_sta *link_sta,
 	struct ieee80211_eht_mcs_nss_supp_20mhz_only mcs_tx_20;
 
 	/* peer is 20Mhz only */
-	if (!(link_sta->he_cap.he_cap_elem.phy_cap_info[0] &
+	if (vif->type == NL80211_IFTYPE_AP &&
+	    !(link_sta->he_cap.he_cap_elem.phy_cap_info[0] &
 	      IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK_ALL)) {
 		mcs_rx_20 = eht_rx_mcs->only_20mhz;
 	} else {
@@ -361,7 +363,8 @@ rs_fw_eht_set_enabled_rates(const struct ieee80211_link_sta *link_sta,
 		       sizeof(cmd->ht_rates[IWL_TLC_NSS_2]));
 }
 
-static void rs_fw_set_supp_rates(struct ieee80211_link_sta *link_sta,
+static void rs_fw_set_supp_rates(struct ieee80211_vif *vif,
+				 struct ieee80211_link_sta *link_sta,
 				 struct ieee80211_supported_band *sband,
 				 struct iwl_tlc_config_cmd_v4 *cmd)
 {
@@ -383,7 +386,7 @@ static void rs_fw_set_supp_rates(struct ieee80211_link_sta *link_sta,
 	/* HT/VHT rates */
 	if (link_sta->eht_cap.has_eht) {
 		cmd->mode = IWL_TLC_MNG_MODE_EHT;
-		rs_fw_eht_set_enabled_rates(link_sta, sband, cmd);
+		rs_fw_eht_set_enabled_rates(vif, link_sta, sband, cmd);
 	} else if (he_cap->has_he) {
 		cmd->mode = IWL_TLC_MNG_MODE_HE;
 		rs_fw_he_set_enabled_rates(link_sta, sband, cmd);
@@ -557,7 +560,9 @@ u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta,
 	return 0;
 }
 
-void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+void rs_fw_rate_init(struct iwl_mvm *mvm,
+		     struct ieee80211_vif *vif,
+		     struct ieee80211_sta *sta,
 		     struct ieee80211_bss_conf *link_conf,
 		     struct ieee80211_link_sta *link_sta,
 		     enum nl80211_band band, bool update)
@@ -601,7 +606,7 @@ void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	iwl_mvm_reset_frame_stats(mvm);
 #endif
-	rs_fw_set_supp_rates(link_sta, sband, &cfg_cmd);
+	rs_fw_set_supp_rates(vif, link_sta, sband, &cfg_cmd);
 
 	/*
 	 * since TLC offload works with one mode we can assume
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index ab82965bc0f4..020de09b13f2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -3015,7 +3015,7 @@ static void rs_drv_rate_update(void *mvm_r,
 	for (tid = 0; tid < IWL_MAX_TID_COUNT; tid++)
 		ieee80211_stop_tx_ba_session(sta, tid);
 
-	iwl_mvm_rs_rate_init(mvm, sta,
+	iwl_mvm_rs_rate_init(mvm, mvmsta->vif, sta,
 			     &mvmsta->vif->bss_conf, &sta->deflink,
 			     sband->band, true);
 }
@@ -4101,13 +4101,16 @@ static const struct rate_control_ops rs_mvm_ops_drv = {
 	.capa = RATE_CTRL_CAPA_VHT_EXT_NSS_BW,
 };
 
-void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm,
+			  struct ieee80211_vif *vif,
+			  struct ieee80211_sta *sta,
 			  struct ieee80211_bss_conf *link_conf,
 			  struct ieee80211_link_sta *link_sta,
 			  enum nl80211_band band, bool update)
 {
 	if (iwl_mvm_has_tlc_offload(mvm)) {
-		rs_fw_rate_init(mvm, sta, link_conf, link_sta, band, update);
+		rs_fw_rate_init(mvm, vif, sta, link_conf,
+				link_sta, band, update);
 	} else {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index f99603b0f693..bbc05c3f13bf 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -394,7 +394,9 @@ struct iwl_lq_sta {
 				   ((_c) << RS_DRV_DATA_LQ_COLOR_POS)))
 
 /* Initialize station's rate scaling information after adding station */
-void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm,
+			  struct ieee80211_vif *vif,
+			  struct ieee80211_sta *sta,
 			  struct ieee80211_bss_conf *link_conf,
 			  struct ieee80211_link_sta *link_sta,
 			  enum nl80211_band band, bool update);
@@ -433,7 +435,9 @@ void iwl_mvm_reset_frame_stats(struct iwl_mvm *mvm);
 #endif
 
 void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta);
-void rs_fw_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+void rs_fw_rate_init(struct iwl_mvm *mvm,
+		     struct ieee80211_vif *vif,
+		     struct ieee80211_sta *sta,
 		     struct ieee80211_bss_conf *link_conf,
 		     struct ieee80211_link_sta *link_sta,
 		     enum nl80211_band band, bool update);
-- 
cgit v1.2.3-70-g09d2


From 8884730eab8773325579c4e8202b6647a42a1b94 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:27 +0300
Subject: wifi: iwlwifi: mvm: rs-fw: properly access sband->iftype_data

We need to look up the correct version of this struct depending
on the interface type, rather than just checking if the pointer
is non-NULL. Fix that.

Fixes: befebbb30af0 ("iwlwifi: rs: consider LDPC capability in case of HE")
Fixes: b009cf71a982 ("iwlwifi: mvm: only enable HE DCM if we also support TX")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.1fd54368a02c.Ie4db5e8ae224d9a4a63b528da5d63e1b957b9cef@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 43d26a09cfce..2b52d3aa82ed 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -61,12 +61,14 @@ static u8 rs_fw_sgi_cw_support(struct ieee80211_link_sta *link_sta)
 }
 
 static u16 rs_fw_get_config_flags(struct iwl_mvm *mvm,
+				  struct ieee80211_vif *vif,
 				  struct ieee80211_link_sta *link_sta,
 				  struct ieee80211_supported_band *sband)
 {
 	struct ieee80211_sta_ht_cap *ht_cap = &link_sta->ht_cap;
 	struct ieee80211_sta_vht_cap *vht_cap = &link_sta->vht_cap;
 	struct ieee80211_sta_he_cap *he_cap = &link_sta->he_cap;
+	const struct ieee80211_sta_he_cap *sband_he_cap;
 	bool vht_ena = vht_cap->vht_supported;
 	u16 flags = 0;
 
@@ -92,17 +94,19 @@ static u16 rs_fw_get_config_flags(struct iwl_mvm *mvm,
 	    IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD))
 		flags |= IWL_TLC_MNG_CFG_FLAGS_LDPC_MSK;
 
-	if (sband->iftype_data && sband->iftype_data->he_cap.has_he &&
-	    !(sband->iftype_data->he_cap.he_cap_elem.phy_cap_info[1] &
-	     IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD))
+	sband_he_cap = ieee80211_get_he_iftype_cap(sband,
+						   ieee80211_vif_type_p2p(vif));
+	if (sband_he_cap &&
+	    !(sband_he_cap->he_cap_elem.phy_cap_info[1] &
+			IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD))
 		flags &= ~IWL_TLC_MNG_CFG_FLAGS_LDPC_MSK;
 
 	if (he_cap->has_he &&
 	    (he_cap->he_cap_elem.phy_cap_info[3] &
 	     IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_MASK &&
-	     sband->iftype_data &&
-	     sband->iftype_data->he_cap.he_cap_elem.phy_cap_info[3] &
-	     IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_MASK))
+	     sband_he_cap &&
+	     sband_he_cap->he_cap_elem.phy_cap_info[3] &
+			IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_MASK))
 		flags |= IWL_TLC_MNG_CFG_FLAGS_HE_DCM_NSS_1_MSK;
 
 	return flags;
@@ -576,9 +580,8 @@ void rs_fw_rate_init(struct iwl_mvm *mvm,
 	struct iwl_lq_sta_rs_fw *lq_sta;
 	struct iwl_tlc_config_cmd_v4 cfg_cmd = {
 		.max_ch_width = update ?
-			rs_fw_bw_from_sta_bw(link_sta) :
-			IWL_TLC_MNG_CH_WIDTH_20MHZ,
-		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, link_sta,
+			rs_fw_bw_from_sta_bw(link_sta) : IWL_TLC_MNG_CH_WIDTH_20MHZ,
+		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, vif, link_sta,
 							    sband)),
 		.chains = rs_fw_set_active_chains(iwl_mvm_get_valid_tx_ant(mvm)),
 		.sgi_ch_width_supp = rs_fw_sgi_cw_support(link_sta),
-- 
cgit v1.2.3-70-g09d2


From 9371ac0dfc13132aa57ab9078d5d43dda8114ac4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:28 +0300
Subject: wifi: iwlwifi: mvm: initialize per-link STA ratescale data

When allocating a new link station, initialize the ratescaling
data for it. To do that, refactor the initialization code out
into a new iwl_mvm_rs_add_sta_link() function.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.441b6a8c60fd.I34f1b3555c25aaa22cc34d1112fc3b6393a20b7c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c |  2 ++
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c   | 35 ++++++++++++++----------
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h      |  5 ++++
 3 files changed, 28 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index fe362f621777..0bfdf4462755 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -1126,6 +1126,8 @@ int iwl_mvm_mld_update_sta_links(struct iwl_mvm *mvm,
 			goto err;
 
 		link_sta_added_to_fw |= BIT(link_id);
+
+		iwl_mvm_rs_add_sta_link(mvm, mvm_sta_link);
 	}
 
 	if (sta_mask_added) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 2b52d3aa82ed..9a6e5684b10c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -679,6 +679,26 @@ int rs_fw_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 	return 0;
 }
 
+void iwl_mvm_rs_add_sta_link(struct iwl_mvm *mvm,
+			     struct iwl_mvm_link_sta *link_sta)
+{
+	struct iwl_lq_sta_rs_fw *lq_sta;
+
+	lq_sta = &link_sta->lq_sta.rs_fw;
+
+	lq_sta->pers.drv = mvm;
+	lq_sta->pers.sta_id = link_sta->sta_id;
+	lq_sta->pers.chains = 0;
+	memset(lq_sta->pers.chain_signal, 0,
+	       sizeof(lq_sta->pers.chain_signal));
+	lq_sta->pers.last_rssi = S8_MIN;
+	lq_sta->last_rate_n_flags = 0;
+
+#ifdef CONFIG_MAC80211_DEBUGFS
+	lq_sta->pers.dbg_fixed_rate = 0;
+#endif
+}
+
 void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta)
 {
 	unsigned int link_id;
@@ -686,25 +706,12 @@ void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta)
 	IWL_DEBUG_RATE(mvm, "create station rate scale window\n");
 
 	for (link_id = 0; link_id < ARRAY_SIZE(mvmsta->link); link_id++) {
-		struct iwl_lq_sta_rs_fw *lq_sta;
 		struct iwl_mvm_link_sta *link =
 			rcu_dereference_protected(mvmsta->link[link_id],
 						  lockdep_is_held(&mvm->mutex));
 		if (!link)
 			continue;
 
-		lq_sta = &link->lq_sta.rs_fw;
-
-		lq_sta->pers.drv = mvm;
-		lq_sta->pers.sta_id = link->sta_id;
-		lq_sta->pers.chains = 0;
-		memset(lq_sta->pers.chain_signal, 0,
-		       sizeof(lq_sta->pers.chain_signal));
-		lq_sta->pers.last_rssi = S8_MIN;
-		lq_sta->last_rate_n_flags = 0;
-
-#ifdef CONFIG_MAC80211_DEBUGFS
-		lq_sta->pers.dbg_fixed_rate = 0;
-#endif
+		iwl_mvm_rs_add_sta_link(mvm, link);
 	}
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index bbc05c3f13bf..c08271a5e3a9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -434,7 +434,12 @@ int iwl_mvm_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 void iwl_mvm_reset_frame_stats(struct iwl_mvm *mvm);
 #endif
 
+struct iwl_mvm_link_sta;
+
 void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta);
+void iwl_mvm_rs_add_sta_link(struct iwl_mvm *mvm,
+			     struct iwl_mvm_link_sta *link_sta);
+
 void rs_fw_rate_init(struct iwl_mvm *mvm,
 		     struct ieee80211_vif *vif,
 		     struct ieee80211_sta *sta,
-- 
cgit v1.2.3-70-g09d2


From 15d4183425a6282bd673b6cdd198ec9335503e62 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:29 +0300
Subject: wifi: iwlwifi: mvm: remove RS rate init update argument

Track instead whether or not the station was authorized, that's
clearer than trying to indicate in the code whether or not the
full bandwidth should be used via an 'update' argument.

While at it, give rs_fw_rate_init() the iwl_mvm_ prefix.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.92bd8d36e311.I1877a109104d5ffeaaad6a623e89f0c44decc38e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 19 +++++++++++--------
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c    | 14 +++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/rs.c       |  8 ++++----
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h       | 14 +++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h      |  2 ++
 5 files changed, 31 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 8808bb3983a4..ac63af549416 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3577,8 +3577,7 @@ static int iwl_mvm_mac_sta_state(struct ieee80211_hw *hw,
  */
 static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
 					   struct ieee80211_vif *vif,
-					   struct ieee80211_sta *sta,
-					   bool update)
+					   struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	unsigned int link_id;
@@ -3593,8 +3592,7 @@ static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
 			continue;
 
 		iwl_mvm_rs_rate_init(mvm, vif, sta, conf, link_sta,
-				     mvmvif->link[link_id]->phy_ctxt->channel->band,
-				     update);
+				     mvmvif->link[link_id]->phy_ctxt->channel->band);
 	}
 }
 
@@ -3766,7 +3764,7 @@ iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw,
 	}
 
 out:
-	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, false);
+	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta);
 
 	return callbacks->update_sta(mvm, vif, sta);
 }
@@ -3799,7 +3797,9 @@ iwl_mvm_sta_state_assoc_to_authorized(struct iwl_mvm *mvm,
 		iwl_mvm_mei_host_associated(mvm, vif, mvm_sta);
 	}
 
-	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, true);
+	mvm_sta->authorized = true;
+
+	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta);
 
 	return 0;
 }
@@ -3811,14 +3811,17 @@ iwl_mvm_sta_state_authorized_to_assoc(struct iwl_mvm *mvm,
 				      struct iwl_mvm_sta_state_ops *callbacks)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
 
+	mvmsta->authorized = false;
+
 	/* once we move into assoc state, need to update rate scale to
 	 * disable using wide bandwidth
 	 */
-	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, false);
+	iwl_mvm_rs_rate_init_all_links(mvm, vif, sta);
 
 	if (!sta->tdls) {
 		/* Set this but don't call iwl_mvm_mac_ctxt_changed()
@@ -3987,7 +3990,7 @@ void iwl_mvm_sta_rc_update(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	if (changed & (IEEE80211_RC_BW_CHANGED |
 		       IEEE80211_RC_SUPP_RATES_CHANGED |
 		       IEEE80211_RC_NSS_CHANGED))
-		iwl_mvm_rs_rate_init_all_links(mvm, vif, sta, true);
+		iwl_mvm_rs_rate_init_all_links(mvm, vif, sta);
 
 	if (vif->type == NL80211_IFTYPE_STATION &&
 	    changed & IEEE80211_RC_NSS_CHANGED)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 9a6e5684b10c..c3a00bfbeef2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -564,12 +564,12 @@ u16 rs_fw_get_max_amsdu_len(struct ieee80211_sta *sta,
 	return 0;
 }
 
-void rs_fw_rate_init(struct iwl_mvm *mvm,
-		     struct ieee80211_vif *vif,
-		     struct ieee80211_sta *sta,
-		     struct ieee80211_bss_conf *link_conf,
-		     struct ieee80211_link_sta *link_sta,
-		     enum nl80211_band band, bool update)
+void iwl_mvm_rs_fw_rate_init(struct iwl_mvm *mvm,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_sta *sta,
+			     struct ieee80211_bss_conf *link_conf,
+			     struct ieee80211_link_sta *link_sta,
+			     enum nl80211_band band)
 {
 	struct ieee80211_hw *hw = mvm->hw;
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
@@ -579,7 +579,7 @@ void rs_fw_rate_init(struct iwl_mvm *mvm,
 	struct iwl_mvm_link_sta *mvm_link_sta;
 	struct iwl_lq_sta_rs_fw *lq_sta;
 	struct iwl_tlc_config_cmd_v4 cfg_cmd = {
-		.max_ch_width = update ?
+		.max_ch_width = mvmsta->authorized ?
 			rs_fw_bw_from_sta_bw(link_sta) : IWL_TLC_MNG_CH_WIDTH_20MHZ,
 		.flags = cpu_to_le16(rs_fw_get_config_flags(mvm, vif, link_sta,
 							    sband)),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 020de09b13f2..a4c1e3bf4ff1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -3017,7 +3017,7 @@ static void rs_drv_rate_update(void *mvm_r,
 
 	iwl_mvm_rs_rate_init(mvm, mvmsta->vif, sta,
 			     &mvmsta->vif->bss_conf, &sta->deflink,
-			     sband->band, true);
+			     sband->band);
 }
 
 static void __iwl_mvm_rs_tx_status(struct iwl_mvm *mvm,
@@ -4106,11 +4106,11 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm,
 			  struct ieee80211_sta *sta,
 			  struct ieee80211_bss_conf *link_conf,
 			  struct ieee80211_link_sta *link_sta,
-			  enum nl80211_band band, bool update)
+			  enum nl80211_band band)
 {
 	if (iwl_mvm_has_tlc_offload(mvm)) {
-		rs_fw_rate_init(mvm, vif, sta, link_conf,
-				link_sta, band, update);
+		iwl_mvm_rs_fw_rate_init(mvm, vif, sta, link_conf,
+					link_sta, band);
 	} else {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index c08271a5e3a9..1ca375a5cf6b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -399,7 +399,7 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm,
 			  struct ieee80211_sta *sta,
 			  struct ieee80211_bss_conf *link_conf,
 			  struct ieee80211_link_sta *link_sta,
-			  enum nl80211_band band, bool update);
+			  enum nl80211_band band);
 
 /* Notify RS about Tx status */
 void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
@@ -440,12 +440,12 @@ void iwl_mvm_rs_add_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta);
 void iwl_mvm_rs_add_sta_link(struct iwl_mvm *mvm,
 			     struct iwl_mvm_link_sta *link_sta);
 
-void rs_fw_rate_init(struct iwl_mvm *mvm,
-		     struct ieee80211_vif *vif,
-		     struct ieee80211_sta *sta,
-		     struct ieee80211_bss_conf *link_conf,
-		     struct ieee80211_link_sta *link_sta,
-		     enum nl80211_band band, bool update);
+void iwl_mvm_rs_fw_rate_init(struct iwl_mvm *mvm,
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_sta *sta,
+			     struct ieee80211_bss_conf *link_conf,
+			     struct ieee80211_link_sta *link_sta,
+			     enum nl80211_band band);
 int rs_fw_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 			bool enable);
 void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 698b9c014cd3..a61d4f88125f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -356,6 +356,7 @@ struct iwl_mvm_link_sta {
  * @tid_disable_agg: bitmap: if bit(tid) is set, the fw won't send ampdus for
  *	tid.
  * @sta_type: station type
+ * @authorized: indicates station is authorized
  * @sta_state: station state according to enum %ieee80211_sta_state
  * @bt_reduced_txpower: is reduced tx power enabled for this station
  * @next_status_eosp: the next reclaimed packet is a PS-Poll response and
@@ -409,6 +410,7 @@ struct iwl_mvm_sta {
 	enum ieee80211_sta_state sta_state;
 	bool bt_reduced_txpower;
 	bool next_status_eosp;
+	bool authorized;
 	spinlock_t lock;
 	struct iwl_mvm_tid_data tid_data[IWL_MAX_TID_COUNT + 1];
 	u8 tid_to_baid[IWL_MAX_TID_COUNT];
-- 
cgit v1.2.3-70-g09d2


From b2bc600cced23762d4e97db8989b18772145604f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:30 +0300
Subject: wifi: iwlwifi: fix iwl_mvm_max_amsdu_size() for MLO

For MLO, we cannot use vif->bss_conf.chandef.chan->band, since
that will lead to a NULL-ptr dereference as bss_conf isn't used.
However, in case of real MLO, we also need to take both LMACs
into account if they exist, since the station might be active
on both LMACs at the same time.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.3588afc85d79.I11592893bbc191b9548518b8bd782de568a9f848@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 37 ++++++++++++++++++++++++++---
 1 file changed, 34 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 09787b8a1c4c..10d7178f1071 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -804,10 +804,11 @@ unsigned int iwl_mvm_max_amsdu_size(struct iwl_mvm *mvm,
 				    struct ieee80211_sta *sta, unsigned int tid)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	enum nl80211_band band = mvmsta->vif->bss_conf.chandef.chan->band;
 	u8 ac = tid_to_mac80211_ac[tid];
+	enum nl80211_band band;
 	unsigned int txf;
-	int lmac = iwl_mvm_get_lmac_id(mvm->fw, band);
+	unsigned int val;
+	int lmac;
 
 	/* For HE redirect to trigger based fifos */
 	if (sta->deflink.he_cap.has_he && !WARN_ON(!iwl_mvm_has_new_tx_api(mvm)))
@@ -821,7 +822,37 @@ unsigned int iwl_mvm_max_amsdu_size(struct iwl_mvm *mvm,
 	 * We also want to have the start of the next packet inside the
 	 * fifo to be able to send bursts.
 	 */
-	return min_t(unsigned int, mvmsta->max_amsdu_len,
+	val = mvmsta->max_amsdu_len;
+
+	if (hweight16(sta->valid_links) <= 1) {
+		if (sta->valid_links) {
+			struct ieee80211_bss_conf *link_conf;
+			unsigned int link = ffs(sta->valid_links) - 1;
+
+			rcu_read_lock();
+			link_conf = rcu_dereference(mvmsta->vif->link_conf[link]);
+			if (WARN_ON(!link_conf))
+				band = NL80211_BAND_2GHZ;
+			else
+				band = link_conf->chandef.chan->band;
+			rcu_read_unlock();
+		} else {
+			band = mvmsta->vif->bss_conf.chandef.chan->band;
+		}
+
+		lmac = iwl_mvm_get_lmac_id(mvm->fw, band);
+	} else if (fw_has_capa(&mvm->fw->ucode_capa,
+			       IWL_UCODE_TLV_CAPA_CDB_SUPPORT)) {
+		/* for real MLO restrict to both LMACs if they exist */
+		lmac = IWL_LMAC_5G_INDEX;
+		val = min_t(unsigned int, val,
+			    mvm->fwrt.smem_cfg.lmac[lmac].txfifo_size[txf] - 256);
+		lmac = IWL_LMAC_24G_INDEX;
+	} else {
+		lmac = IWL_LMAC_24G_INDEX;
+	}
+
+	return min_t(unsigned int, val,
 		     mvm->fwrt.smem_cfg.lmac[lmac].txfifo_size[txf] - 256);
 }
 
-- 
cgit v1.2.3-70-g09d2


From d2d0468f60cda38c275f7d1cc5955d60eebad43b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:31 +0300
Subject: wifi: iwlwifi: mvm: configure TLC on link activation

If the AP station already exists on link activation (which
means we're during link switch), configure the TLC in FW so
we can immediately transmit once the link is activated.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.1a93de9b7c1f.I42022f24bbe3572f5a082da8c99794ae14281875@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 24 ++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 99ef12c530ea..fbc2d5ed1006 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -256,6 +256,30 @@ __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 	if (ret)
 		goto out;
 
+	/* Initialize rate control for the AP station, since we might be
+	 * doing a link switch here - we cannot initialize it before since
+	 * this needs the phy context assigned (and in FW?), and we cannot
+	 * do it later because it needs to be initialized as soon as we're
+	 * able to TX on the link, i.e. when active.
+	 *
+	 * Firmware restart isn't quite correct yet for MLO, but we don't
+	 * need to do it in that case anyway since it will happen from the
+	 * normal station state callback.
+	 */
+	if (mvmvif->ap_sta &&
+	    !test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
+		struct ieee80211_link_sta *link_sta;
+
+		rcu_read_lock();
+		link_sta = rcu_dereference(mvmvif->ap_sta->link[link_id]);
+
+		if (!WARN_ON_ONCE(!link_sta))
+			iwl_mvm_rs_rate_init(mvm, vif, mvmvif->ap_sta,
+					     link_conf, link_sta,
+					     phy_ctxt->channel->band);
+		rcu_read_unlock();
+	}
+
 	/* then activate */
 	ret = iwl_mvm_link_changed(mvm, vif, link_conf,
 				   LINK_CONTEXT_MODIFY_ACTIVE |
-- 
cgit v1.2.3-70-g09d2


From 8939a18ce1d7efcaff8801ab16dd33eaeec4c91c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:32 +0300
Subject: wifi: iwlwifi: mvm: add MLO support to SF - use sta pointer

For MLO support, use the pointer to the peer MLD instead of the default
link station id, since the default link is only used for non-MLO cases.
Using the default link sta id is meaningless for MLO. Also remove the
rcu protected section since we now avoid the lookup based on sta id.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.40cef48c0592.I8dd666d603d6e1854033e5369c70e78d9303d236@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sf.c | 29 ++++++++---------------------
 1 file changed, 8 insertions(+), 21 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
index 7c5f41e40e7a..98f330fcf678 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c
@@ -8,7 +8,7 @@
 /* For counting bound interfaces */
 struct iwl_mvm_active_iface_iterator_data {
 	struct ieee80211_vif *ignore_vif;
-	u8 sta_vif_ap_sta_id;
+	struct ieee80211_sta *sta_vif_ap_sta;
 	enum iwl_sf_state sta_vif_state;
 	u32 num_active_macs;
 };
@@ -30,7 +30,7 @@ static void iwl_mvm_bound_iface_iterator(void *_data, u8 *mac,
 	data->num_active_macs++;
 
 	if (vif->type == NL80211_IFTYPE_STATION) {
-		data->sta_vif_ap_sta_id = mvmvif->deflink.ap_sta_id;
+		data->sta_vif_ap_sta = mvmvif->ap_sta;
 		if (vif->cfg.assoc)
 			data->sta_vif_state = SF_FULL_ON;
 		else
@@ -172,13 +172,12 @@ static void iwl_mvm_fill_sf_command(struct iwl_mvm *mvm,
 	}
 }
 
-static int iwl_mvm_sf_config(struct iwl_mvm *mvm, u8 sta_id,
+static int iwl_mvm_sf_config(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			     enum iwl_sf_state new_state)
 {
 	struct iwl_sf_cfg_cmd sf_cmd = {
 		.state = cpu_to_le32(new_state),
 	};
-	struct ieee80211_sta *sta;
 	int ret = 0;
 
 	if (mvm->cfg->disable_dummy_notification)
@@ -196,20 +195,12 @@ static int iwl_mvm_sf_config(struct iwl_mvm *mvm, u8 sta_id,
 		iwl_mvm_fill_sf_command(mvm, &sf_cmd, NULL);
 		break;
 	case SF_FULL_ON:
-		if (sta_id == IWL_MVM_INVALID_STA) {
+		if (!sta) {
 			IWL_ERR(mvm,
 				"No station: Cannot switch SF to FULL_ON\n");
 			return -EINVAL;
 		}
-		rcu_read_lock();
-		sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]);
-		if (IS_ERR_OR_NULL(sta)) {
-			IWL_ERR(mvm, "Invalid station id\n");
-			rcu_read_unlock();
-			return -EINVAL;
-		}
 		iwl_mvm_fill_sf_command(mvm, &sf_cmd, sta);
-		rcu_read_unlock();
 		break;
 	case SF_INIT_OFF:
 		iwl_mvm_fill_sf_command(mvm, &sf_cmd, NULL);
@@ -237,13 +228,12 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 		      bool remove_vif)
 {
 	enum iwl_sf_state new_state;
-	u8 sta_id = IWL_MVM_INVALID_STA;
 	struct iwl_mvm_vif *mvmvif = NULL;
 	struct iwl_mvm_active_iface_iterator_data data = {
 		.ignore_vif = changed_vif,
 		.sta_vif_state = SF_UNINIT,
-		.sta_vif_ap_sta_id = IWL_MVM_INVALID_STA,
 	};
+	struct ieee80211_sta *sta = NULL;
 
 	/*
 	 * Ignore the call if we are in HW Restart flow, or if the handled
@@ -273,7 +263,7 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 			 * and we filled the relevant data during iteration
 			 */
 			new_state = data.sta_vif_state;
-			sta_id = data.sta_vif_ap_sta_id;
+			sta = data.sta_vif_ap_sta;
 		} else {
 			if (WARN_ON(!changed_vif))
 				return -EINVAL;
@@ -282,7 +272,7 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 			} else if (changed_vif->cfg.assoc &&
 				   changed_vif->bss_conf.dtim_period) {
 				mvmvif = iwl_mvm_vif_from_mac80211(changed_vif);
-				sta_id = mvmvif->deflink.ap_sta_id;
+				sta = mvmvif->ap_sta;
 				new_state = SF_FULL_ON;
 			} else {
 				new_state = SF_INIT_OFF;
@@ -294,8 +284,5 @@ int iwl_mvm_sf_update(struct iwl_mvm *mvm, struct ieee80211_vif *changed_vif,
 		new_state = SF_UNINIT;
 	}
 
-	/* For MLO it's ok to use deflink->sta_id as it's needed only to get
-	 * a pointer to mac80211 sta
-	 */
-	return iwl_mvm_sf_config(mvm, sta_id, new_state);
+	return iwl_mvm_sf_config(mvm, sta, new_state);
 }
-- 
cgit v1.2.3-70-g09d2


From 13513cec93ac9902d0b896976d8bab3758a9881c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 17 Apr 2023 11:41:33 +0300
Subject: wifi: iwlwifi: mvm: check firmware response size

Check the firmware response size for responses to the
memory read/write command in debugfs before using it.

Fixes: 2b55f43f8e47 ("iwlwifi: mvm: Add mem debugfs entry")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.0d56fcaf68ee.I70e9571f3ed7263929b04f8fabad23c9b999e4ea@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index affaed4d13fa..84a488538427 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -1961,6 +1961,11 @@ static ssize_t iwl_dbgfs_mem_read(struct file *file, char __user *user_buf,
 	if (ret < 0)
 		return ret;
 
+	if (iwl_rx_packet_payload_len(hcmd.resp_pkt) < sizeof(*rsp)) {
+		ret = -EIO;
+		goto out;
+	}
+
 	rsp = (void *)hcmd.resp_pkt->data;
 	if (le32_to_cpu(rsp->status) != DEBUG_MEM_STATUS_SUCCESS) {
 		ret = -ENXIO;
@@ -2037,6 +2042,11 @@ static ssize_t iwl_dbgfs_mem_write(struct file *file,
 	if (ret < 0)
 		return ret;
 
+	if (iwl_rx_packet_payload_len(hcmd.resp_pkt) < sizeof(*rsp)) {
+		ret = -EIO;
+		goto out;
+	}
+
 	rsp = (void *)hcmd.resp_pkt->data;
 	if (rsp->status != DEBUG_MEM_STATUS_SUCCESS) {
 		ret = -ENXIO;
-- 
cgit v1.2.3-70-g09d2


From f25ee51452dfe156c515adefba4dafbc10366ce5 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Mon, 17 Apr 2023 11:41:34 +0300
Subject: wifi: iwlwifi: bump FW API to 78 for AX devices

Start supporting API version 78 for AX devices.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230417113648.5a4dcbf5a2c1.I125808566fe892ee0865e392bf1b1872daafe8ad@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index b98f0ff02362..532d14f61253 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_22000_UCODE_API_MAX	77
+#define IWL_22000_UCODE_API_MAX	78
 
 /* Lowest firmware API version supported */
 #define IWL_22000_UCODE_API_MIN	39
-- 
cgit v1.2.3-70-g09d2


From 9ae708f00161e1d789268fa9cc05bf6bec2af474 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Mon, 17 Apr 2023 15:37:51 +0200
Subject: wifi: mac80211: remove ieee80211_tx_status_8023

It is unused and should not be used. In order to avoid limitations in
4-address mode, the driver should always use ieee80211_tx_status_ext for
802.3 frames with a valid sta pointer.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://lore.kernel.org/r/20230417133751.79160-1-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h | 20 --------------------
 net/mac80211/status.c  | 24 ------------------------
 2 files changed, 44 deletions(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index a8dadbd83d95..ac0370e76874 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -5215,26 +5215,6 @@ static inline void ieee80211_tx_status_ni(struct ieee80211_hw *hw,
 void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
 				 struct sk_buff *skb);
 
-/**
- * ieee80211_tx_status_8023 - transmit status callback for 802.3 frame format
- *
- * Call this function for all transmitted data frames after their transmit
- * completion. This callback should only be called for data frames which
- * are using driver's (or hardware's) offload capability of encap/decap
- * 802.11 frames.
- *
- * This function may not be called in IRQ context. Calls to this function
- * for a single hardware must be synchronized against each other and all
- * calls in the same tx status family.
- *
- * @hw: the hardware the frame was transmitted by
- * @vif: the interface for which the frame was transmitted
- * @skb: the frame that was transmitted, owned by mac80211 after this call
- */
-void ieee80211_tx_status_8023(struct ieee80211_hw *hw,
-			       struct ieee80211_vif *vif,
-			       struct sk_buff *skb);
-
 /**
  * ieee80211_report_low_ack - report non-responding station
  *
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 3f9ddd7f04b6..2b13a52ce96c 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -1244,30 +1244,6 @@ void ieee80211_tx_rate_update(struct ieee80211_hw *hw,
 }
 EXPORT_SYMBOL(ieee80211_tx_rate_update);
 
-void ieee80211_tx_status_8023(struct ieee80211_hw *hw,
-			      struct ieee80211_vif *vif,
-			      struct sk_buff *skb)
-{
-	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_tx_status status = {
-		.skb = skb,
-		.info = IEEE80211_SKB_CB(skb),
-	};
-	struct sta_info *sta;
-
-	sdata = vif_to_sdata(vif);
-
-	rcu_read_lock();
-
-	if (!ieee80211_lookup_ra_sta(sdata, skb, &sta) && !IS_ERR(sta))
-		status.sta = &sta->sta;
-
-	ieee80211_tx_status_ext(hw, &status);
-
-	rcu_read_unlock();
-}
-EXPORT_SYMBOL(ieee80211_tx_status_8023);
-
 void ieee80211_report_low_ack(struct ieee80211_sta *pubsta, u32 num_packets)
 {
 	struct sta_info *sta = container_of(pubsta, struct sta_info, sta);
-- 
cgit v1.2.3-70-g09d2


From 3055ddd654f65ebdd7860b4dfecd7e49e54c01b7 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 17 Apr 2023 17:21:43 -0700
Subject: libbpf: misc internal libbpf clean ups around log fixup

Normalize internal constants, field names, and comments related to log
fixup. Also add explicit `ext_idx` alias for relocation where relocation
is pointing to extern description for additional information.

No functional changes, just a clean up before subsequent additions.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230418002148.3255690-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 49cd304ae3bc..a382ed3586bd 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -333,6 +333,7 @@ struct reloc_desc {
 		struct {
 			int map_idx;
 			int sym_off;
+			int ext_idx;
 		};
 	};
 };
@@ -4042,7 +4043,7 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 		else
 			reloc_desc->type = RELO_EXTERN_LD64;
 		reloc_desc->insn_idx = insn_idx;
-		reloc_desc->sym_off = i; /* sym_off stores extern index */
+		reloc_desc->ext_idx = i;
 		return 0;
 	}
 
@@ -5811,8 +5812,8 @@ out:
 }
 
 /* base map load ldimm64 special constant, used also for log fixup logic */
-#define MAP_LDIMM64_POISON_BASE 2001000000
-#define MAP_LDIMM64_POISON_PFX "200100"
+#define POISON_LDIMM64_MAP_BASE 2001000000
+#define POISON_LDIMM64_MAP_PFX "200100"
 
 static void poison_map_ldimm64(struct bpf_program *prog, int relo_idx,
 			       int insn_idx, struct bpf_insn *insn,
@@ -5834,7 +5835,7 @@ static void poison_map_ldimm64(struct bpf_program *prog, int relo_idx,
 		 * invalid func unknown#2001000123
 		 * where lower 123 is map index into obj->maps[] array
 		 */
-		insn->imm = MAP_LDIMM64_POISON_BASE + map_idx;
+		insn->imm = POISON_LDIMM64_MAP_BASE + map_idx;
 
 		insn++;
 	}
@@ -5885,7 +5886,7 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 			}
 			break;
 		case RELO_EXTERN_LD64:
-			ext = &obj->externs[relo->sym_off];
+			ext = &obj->externs[relo->ext_idx];
 			if (ext->type == EXT_KCFG) {
 				if (obj->gen_loader) {
 					insn[0].src_reg = BPF_PSEUDO_MAP_IDX_VALUE;
@@ -5907,7 +5908,7 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 			}
 			break;
 		case RELO_EXTERN_CALL:
-			ext = &obj->externs[relo->sym_off];
+			ext = &obj->externs[relo->ext_idx];
 			insn[0].src_reg = BPF_PSEUDO_KFUNC_CALL;
 			if (ext->is_set) {
 				insn[0].imm = ext->ksym.kernel_btf_id;
@@ -7022,13 +7023,13 @@ static void fixup_log_missing_map_load(struct bpf_program *prog,
 				       char *buf, size_t buf_sz, size_t log_sz,
 				       char *line1, char *line2, char *line3)
 {
-	/* Expected log for failed and not properly guarded CO-RE relocation:
+	/* Expected log for failed and not properly guarded map reference:
 	 * line1 -> 123: (85) call unknown#2001000345
 	 * line2 -> invalid func unknown#2001000345
 	 * line3 -> <anything else or end of buffer>
 	 *
 	 * "123" is the index of the instruction that was poisoned.
-	 * "345" in "2001000345" are map index in obj->maps to fetch map name.
+	 * "345" in "2001000345" is a map index in obj->maps to fetch map name.
 	 */
 	struct bpf_object *obj = prog->obj;
 	const struct bpf_map *map;
@@ -7038,7 +7039,7 @@ static void fixup_log_missing_map_load(struct bpf_program *prog,
 	if (sscanf(line1, "%d: (%*d) call unknown#%d\n", &insn_idx, &map_idx) != 2)
 		return;
 
-	map_idx -= MAP_LDIMM64_POISON_BASE;
+	map_idx -= POISON_LDIMM64_MAP_BASE;
 	if (map_idx < 0 || map_idx >= obj->nr_maps)
 		return;
 	map = &obj->maps[map_idx];
@@ -7070,20 +7071,21 @@ static void fixup_verifier_log(struct bpf_program *prog, char *buf, size_t buf_s
 		if (!cur_line)
 			return;
 
-		/* failed CO-RE relocation case */
 		if (str_has_pfx(cur_line, "invalid func unknown#195896080\n")) {
 			prev_line = find_prev_line(buf, cur_line);
 			if (!prev_line)
 				continue;
 
+			/* failed CO-RE relocation case */
 			fixup_log_failed_core_relo(prog, buf, buf_sz, log_sz,
 						   prev_line, cur_line, next_line);
 			return;
-		} else if (str_has_pfx(cur_line, "invalid func unknown#"MAP_LDIMM64_POISON_PFX)) {
+		} else if (str_has_pfx(cur_line, "invalid func unknown#"POISON_LDIMM64_MAP_PFX)) {
 			prev_line = find_prev_line(buf, cur_line);
 			if (!prev_line)
 				continue;
 
+			/* reference to uncreated BPF map */
 			fixup_log_missing_map_load(prog, buf, buf_sz, log_sz,
 						   prev_line, cur_line, next_line);
 			return;
@@ -7098,7 +7100,7 @@ static int bpf_program_record_relos(struct bpf_program *prog)
 
 	for (i = 0; i < prog->nr_reloc; i++) {
 		struct reloc_desc *relo = &prog->reloc_desc[i];
-		struct extern_desc *ext = &obj->externs[relo->sym_off];
+		struct extern_desc *ext = &obj->externs[relo->ext_idx];
 		int kind;
 
 		switch (relo->type) {
-- 
cgit v1.2.3-70-g09d2


From f709160d1724a4be469b985dcaadaa2aa8d5ce94 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 17 Apr 2023 17:21:44 -0700
Subject: libbpf: report vmlinux vs module name when dealing with ksyms

Currently libbpf always reports "kernel" as a source of ksym BTF type,
which is ambiguous given ksym's BTF can come from either vmlinux or
kernel module BTFs. Make this explicit and log module name, if used BTF
is from kernel module.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230418002148.3255690-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index a382ed3586bd..0a11563067b3 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -7538,8 +7538,9 @@ static int bpf_object__resolve_ksym_func_btf_id(struct bpf_object *obj,
 	ret = bpf_core_types_are_compat(obj->btf, local_func_proto_id,
 					kern_btf, kfunc_proto_id);
 	if (ret <= 0) {
-		pr_warn("extern (func ksym) '%s': func_proto [%d] incompatible with kernel [%d]\n",
-			ext->name, local_func_proto_id, kfunc_proto_id);
+		pr_warn("extern (func ksym) '%s': func_proto [%d] incompatible with %s [%d]\n",
+			ext->name, local_func_proto_id,
+			mod_btf ? mod_btf->name : "vmlinux", kfunc_proto_id);
 		return -EINVAL;
 	}
 
@@ -7573,8 +7574,8 @@ static int bpf_object__resolve_ksym_func_btf_id(struct bpf_object *obj,
 	 * {kernel_btf_id, kernel_btf_obj_fd} -> fixup ld_imm64.
 	 */
 	ext->ksym.kernel_btf_obj_fd = mod_btf ? mod_btf->fd : 0;
-	pr_debug("extern (func ksym) '%s': resolved to kernel [%d]\n",
-		 ext->name, kfunc_id);
+	pr_debug("extern (func ksym) '%s': resolved to %s [%d]\n",
+		 ext->name, mod_btf ? mod_btf->name : "vmlinux", kfunc_id);
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 05b6f766b25c455f2349df7714f0f0d3ac37fe46 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 17 Apr 2023 17:21:45 -0700
Subject: libbpf: improve handling of unresolved kfuncs

Currently, libbpf leaves `call #0` instruction for __weak unresolved
kfuncs, which might lead to a confusing verifier log situations, where
invalid `call #0` will be treated as successfully validated.

We can do better. Libbpf already has an established mechanism of
poisoning instructions that failed some form of resolution (e.g., CO-RE
relocation and BPF map set to not be auto-created). Libbpf doesn't fail
them outright to allow users to guard them through other means, and as
long as BPF verifier can prove that such poisoned instructions cannot be
ever reached, this doesn't consistute an invalid BPF program. If user
didn't guard such code, libbpf will extract few pieces of information to
tie such poisoned instructions back to additional information about what
entitity wasn't resolved (e.g., BPF map name, or CO-RE relocation
information).

__weak unresolved kfuncs fit this model well, so this patch extends
libbpf with poisioning and log fixup logic for kfunc calls.

Note, this poisoning is done only for kfunc *calls*, not kfunc address
resolution (ldimm64 instructions). The former cannot be ever valid, if
reached, so it's safe to poison them. The latter is a valid mechanism to
check if __weak kfunc ksym was resolved, and do necessary guarding and
work arounds based on this result, supported in most recent kernels. As
such, libbpf keeps such ldimm64 instructions as loading zero, never
poisoning them.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230418002148.3255690-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf.c | 72 +++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 69 insertions(+), 3 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 0a11563067b3..2600d8384252 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -5841,6 +5841,30 @@ static void poison_map_ldimm64(struct bpf_program *prog, int relo_idx,
 	}
 }
 
+/* unresolved kfunc call special constant, used also for log fixup logic */
+#define POISON_CALL_KFUNC_BASE 2002000000
+#define POISON_CALL_KFUNC_PFX "2002"
+
+static void poison_kfunc_call(struct bpf_program *prog, int relo_idx,
+			      int insn_idx, struct bpf_insn *insn,
+			      int ext_idx, const struct extern_desc *ext)
+{
+	pr_debug("prog '%s': relo #%d: poisoning insn #%d that calls kfunc '%s'\n",
+		 prog->name, relo_idx, insn_idx, ext->name);
+
+	/* we turn kfunc call into invalid helper call with identifiable constant */
+	insn->code = BPF_JMP | BPF_CALL;
+	insn->dst_reg = 0;
+	insn->src_reg = 0;
+	insn->off = 0;
+	/* if this instruction is reachable (not a dead code),
+	 * verifier will complain with something like:
+	 * invalid func unknown#2001000123
+	 * where lower 123 is extern index into obj->externs[] array
+	 */
+	insn->imm = POISON_CALL_KFUNC_BASE + ext_idx;
+}
+
 /* Relocate data references within program code:
  *  - map references;
  *  - global variable references;
@@ -5913,9 +5937,9 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 			if (ext->is_set) {
 				insn[0].imm = ext->ksym.kernel_btf_id;
 				insn[0].off = ext->ksym.btf_fd_idx;
-			} else { /* unresolved weak kfunc */
-				insn[0].imm = 0;
-				insn[0].off = 0;
+			} else { /* unresolved weak kfunc call */
+				poison_kfunc_call(prog, i, relo->insn_idx, insn,
+						  relo->ext_idx, ext);
 			}
 			break;
 		case RELO_SUBPROG_ADDR:
@@ -7052,6 +7076,39 @@ static void fixup_log_missing_map_load(struct bpf_program *prog,
 	patch_log(buf, buf_sz, log_sz, line1, line3 - line1, patch);
 }
 
+static void fixup_log_missing_kfunc_call(struct bpf_program *prog,
+					 char *buf, size_t buf_sz, size_t log_sz,
+					 char *line1, char *line2, char *line3)
+{
+	/* Expected log for failed and not properly guarded kfunc call:
+	 * line1 -> 123: (85) call unknown#2002000345
+	 * line2 -> invalid func unknown#2002000345
+	 * line3 -> <anything else or end of buffer>
+	 *
+	 * "123" is the index of the instruction that was poisoned.
+	 * "345" in "2002000345" is an extern index in obj->externs to fetch kfunc name.
+	 */
+	struct bpf_object *obj = prog->obj;
+	const struct extern_desc *ext;
+	int insn_idx, ext_idx;
+	char patch[128];
+
+	if (sscanf(line1, "%d: (%*d) call unknown#%d\n", &insn_idx, &ext_idx) != 2)
+		return;
+
+	ext_idx -= POISON_CALL_KFUNC_BASE;
+	if (ext_idx < 0 || ext_idx >= obj->nr_extern)
+		return;
+	ext = &obj->externs[ext_idx];
+
+	snprintf(patch, sizeof(patch),
+		 "%d: <invalid kfunc call>\n"
+		 "kfunc '%s' is referenced but wasn't resolved\n",
+		 insn_idx, ext->name);
+
+	patch_log(buf, buf_sz, log_sz, line1, line3 - line1, patch);
+}
+
 static void fixup_verifier_log(struct bpf_program *prog, char *buf, size_t buf_sz)
 {
 	/* look for familiar error patterns in last N lines of the log */
@@ -7089,6 +7146,15 @@ static void fixup_verifier_log(struct bpf_program *prog, char *buf, size_t buf_s
 			fixup_log_missing_map_load(prog, buf, buf_sz, log_sz,
 						   prev_line, cur_line, next_line);
 			return;
+		} else if (str_has_pfx(cur_line, "invalid func unknown#"POISON_CALL_KFUNC_PFX)) {
+			prev_line = find_prev_line(buf, cur_line);
+			if (!prev_line)
+				continue;
+
+			/* reference to unresolved kfunc */
+			fixup_log_missing_kfunc_call(prog, buf, buf_sz, log_sz,
+						     prev_line, cur_line, next_line);
+			return;
 		}
 	}
 }
-- 
cgit v1.2.3-70-g09d2


From 30bbfe3236b01b81ed5f4a91cd9d87a236b182e3 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 17 Apr 2023 17:21:46 -0700
Subject: selftests/bpf: add missing __weak kfunc log fixup test

Add test validating that libbpf correctly poisons and reports __weak
unresolved kfuncs in post-processed verifier log.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230418002148.3255690-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/log_fixup.c | 31 ++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/test_log_fixup.c | 10 +++++++
 2 files changed, 41 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/log_fixup.c b/tools/testing/selftests/bpf/prog_tests/log_fixup.c
index bc27170bdeb0..dba71d98a227 100644
--- a/tools/testing/selftests/bpf/prog_tests/log_fixup.c
+++ b/tools/testing/selftests/bpf/prog_tests/log_fixup.c
@@ -135,6 +135,35 @@ cleanup:
 	test_log_fixup__destroy(skel);
 }
 
+static void missing_kfunc(void)
+{
+	char log_buf[8 * 1024];
+	struct test_log_fixup* skel;
+	int err;
+
+	skel = test_log_fixup__open();
+	if (!ASSERT_OK_PTR(skel, "skel_open"))
+		return;
+
+	bpf_program__set_autoload(skel->progs.use_missing_kfunc, true);
+	bpf_program__set_log_buf(skel->progs.use_missing_kfunc, log_buf, sizeof(log_buf));
+
+	err = test_log_fixup__load(skel);
+	if (!ASSERT_ERR(err, "load_fail"))
+		goto cleanup;
+
+	ASSERT_HAS_SUBSTR(log_buf,
+			  "0: <invalid kfunc call>\n"
+			  "kfunc 'bpf_nonexistent_kfunc' is referenced but wasn't resolved\n",
+			  "log_buf");
+
+	if (env.verbosity > VERBOSE_NONE)
+		printf("LOG:   \n=================\n%s=================\n", log_buf);
+
+cleanup:
+	test_log_fixup__destroy(skel);
+}
+
 void test_log_fixup(void)
 {
 	if (test__start_subtest("bad_core_relo_trunc_none"))
@@ -147,4 +176,6 @@ void test_log_fixup(void)
 		bad_core_relo_subprog();
 	if (test__start_subtest("missing_map"))
 		missing_map();
+	if (test__start_subtest("missing_kfunc"))
+		missing_kfunc();
 }
diff --git a/tools/testing/selftests/bpf/progs/test_log_fixup.c b/tools/testing/selftests/bpf/progs/test_log_fixup.c
index 60450cb0e72e..1bd48feaaa42 100644
--- a/tools/testing/selftests/bpf/progs/test_log_fixup.c
+++ b/tools/testing/selftests/bpf/progs/test_log_fixup.c
@@ -61,4 +61,14 @@ int use_missing_map(const void *ctx)
 	return value != NULL;
 }
 
+extern int bpf_nonexistent_kfunc(void) __ksym __weak;
+
+SEC("?raw_tp/sys_enter")
+int use_missing_kfunc(const void *ctx)
+{
+	bpf_nonexistent_kfunc();
+
+	return 0;
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From c5e64741670883a5b35d42f3125d611f5a4aa14f Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 17 Apr 2023 17:21:47 -0700
Subject: libbpf: move bpf_for(), bpf_for_each(), and bpf_repeat() into
 bpf_helpers.h

To make it easier for bleeding-edge BPF applications, such as sched_ext,
to utilize open-coded iterators, move bpf_for(), bpf_for_each(), and
bpf_repeat() macros from selftests/bpf-internal bpf_misc.h helper, to
libbpf-provided bpf_helpers.h header.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230418002148.3255690-6-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/bpf_helpers.h                  | 103 +++++++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/bpf_misc.h | 103 ---------------------------
 2 files changed, 103 insertions(+), 103 deletions(-)

diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index e7e1a8acc299..525dec66c129 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -291,4 +291,107 @@ enum libbpf_tristate {
 /* Helper macro to print out debug messages */
 #define bpf_printk(fmt, args...) ___bpf_pick_printk(args)(fmt, ##args)
 
+struct bpf_iter_num;
+
+extern int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end) __ksym;
+extern int *bpf_iter_num_next(struct bpf_iter_num *it) __ksym;
+extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
+
+#ifndef bpf_for_each
+/* bpf_for_each(iter_type, cur_elem, args...) provides generic construct for
+ * using BPF open-coded iterators without having to write mundane explicit
+ * low-level loop logic. Instead, it provides for()-like generic construct
+ * that can be used pretty naturally. E.g., for some hypothetical cgroup
+ * iterator, you'd write:
+ *
+ * struct cgroup *cg, *parent_cg = <...>;
+ *
+ * bpf_for_each(cgroup, cg, parent_cg, CG_ITER_CHILDREN) {
+ *     bpf_printk("Child cgroup id = %d", cg->cgroup_id);
+ *     if (cg->cgroup_id == 123)
+ *         break;
+ * }
+ *
+ * I.e., it looks almost like high-level for each loop in other languages,
+ * supports continue/break, and is verifiable by BPF verifier.
+ *
+ * For iterating integers, the difference betwen bpf_for_each(num, i, N, M)
+ * and bpf_for(i, N, M) is in that bpf_for() provides additional proof to
+ * verifier that i is in [N, M) range, and in bpf_for_each() case i is `int
+ * *`, not just `int`. So for integers bpf_for() is more convenient.
+ *
+ * Note: this macro relies on C99 feature of allowing to declare variables
+ * inside for() loop, bound to for() loop lifetime. It also utilizes GCC
+ * extension: __attribute__((cleanup(<func>))), supported by both GCC and
+ * Clang.
+ */
+#define bpf_for_each(type, cur, args...) for (							\
+	/* initialize and define destructor */							\
+	struct bpf_iter_##type ___it __attribute__((aligned(8), /* enforce, just in case */,	\
+						    cleanup(bpf_iter_##type##_destroy))),	\
+	/* ___p pointer is just to call bpf_iter_##type##_new() *once* to init ___it */		\
+			       *___p __attribute__((unused)) = (				\
+					bpf_iter_##type##_new(&___it, ##args),			\
+	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
+	/* for bpf_iter_##type##_destroy() when used from cleanup() attribute */		\
+					(void)bpf_iter_##type##_destroy, (void *)0);		\
+	/* iteration and termination check */							\
+	(((cur) = bpf_iter_##type##_next(&___it)));						\
+)
+#endif /* bpf_for_each */
+
+#ifndef bpf_for
+/* bpf_for(i, start, end) implements a for()-like looping construct that sets
+ * provided integer variable *i* to values starting from *start* through,
+ * but not including, *end*. It also proves to BPF verifier that *i* belongs
+ * to range [start, end), so this can be used for accessing arrays without
+ * extra checks.
+ *
+ * Note: *start* and *end* are assumed to be expressions with no side effects
+ * and whose values do not change throughout bpf_for() loop execution. They do
+ * not have to be statically known or constant, though.
+ *
+ * Note: similarly to bpf_for_each(), it relies on C99 feature of declaring for()
+ * loop bound variables and cleanup attribute, supported by GCC and Clang.
+ */
+#define bpf_for(i, start, end) for (								\
+	/* initialize and define destructor */							\
+	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
+						 cleanup(bpf_iter_num_destroy))),		\
+	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
+			    *___p __attribute__((unused)) = (					\
+				bpf_iter_num_new(&___it, (start), (end)),			\
+	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
+	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
+				(void)bpf_iter_num_destroy, (void *)0);				\
+	({											\
+		/* iteration step */								\
+		int *___t = bpf_iter_num_next(&___it);						\
+		/* termination and bounds check */						\
+		(___t && ((i) = *___t, (i) >= (start) && (i) < (end)));				\
+	});											\
+)
+#endif /* bpf_for */
+
+#ifndef bpf_repeat
+/* bpf_repeat(N) performs N iterations without exposing iteration number
+ *
+ * Note: similarly to bpf_for_each(), it relies on C99 feature of declaring for()
+ * loop bound variables and cleanup attribute, supported by GCC and Clang.
+ */
+#define bpf_repeat(N) for (									\
+	/* initialize and define destructor */							\
+	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
+						 cleanup(bpf_iter_num_destroy))),		\
+	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
+			    *___p __attribute__((unused)) = (					\
+				bpf_iter_num_new(&___it, 0, (N)),				\
+	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
+	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
+				(void)bpf_iter_num_destroy, (void *)0);				\
+	bpf_iter_num_next(&___it);								\
+	/* nothing here  */									\
+)
+#endif /* bpf_repeat */
+
 #endif
diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 6e3b4903c541..3b307de8dab9 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -121,107 +121,4 @@
 /* make it look to compiler like value is read and written */
 #define __sink(expr) asm volatile("" : "+g"(expr))
 
-struct bpf_iter_num;
-
-extern int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end) __ksym;
-extern int *bpf_iter_num_next(struct bpf_iter_num *it) __ksym;
-extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
-
-#ifndef bpf_for_each
-/* bpf_for_each(iter_type, cur_elem, args...) provides generic construct for
- * using BPF open-coded iterators without having to write mundane explicit
- * low-level loop logic. Instead, it provides for()-like generic construct
- * that can be used pretty naturally. E.g., for some hypothetical cgroup
- * iterator, you'd write:
- *
- * struct cgroup *cg, *parent_cg = <...>;
- *
- * bpf_for_each(cgroup, cg, parent_cg, CG_ITER_CHILDREN) {
- *     bpf_printk("Child cgroup id = %d", cg->cgroup_id);
- *     if (cg->cgroup_id == 123)
- *         break;
- * }
- *
- * I.e., it looks almost like high-level for each loop in other languages,
- * supports continue/break, and is verifiable by BPF verifier.
- *
- * For iterating integers, the difference betwen bpf_for_each(num, i, N, M)
- * and bpf_for(i, N, M) is in that bpf_for() provides additional proof to
- * verifier that i is in [N, M) range, and in bpf_for_each() case i is `int
- * *`, not just `int`. So for integers bpf_for() is more convenient.
- *
- * Note: this macro relies on C99 feature of allowing to declare variables
- * inside for() loop, bound to for() loop lifetime. It also utilizes GCC
- * extension: __attribute__((cleanup(<func>))), supported by both GCC and
- * Clang.
- */
-#define bpf_for_each(type, cur, args...) for (							\
-	/* initialize and define destructor */							\
-	struct bpf_iter_##type ___it __attribute__((aligned(8), /* enforce, just in case */,	\
-						    cleanup(bpf_iter_##type##_destroy))),	\
-	/* ___p pointer is just to call bpf_iter_##type##_new() *once* to init ___it */		\
-			       *___p __attribute__((unused)) = (				\
-					bpf_iter_##type##_new(&___it, ##args),			\
-	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
-	/* for bpf_iter_##type##_destroy() when used from cleanup() attribute */		\
-					(void)bpf_iter_##type##_destroy, (void *)0);		\
-	/* iteration and termination check */							\
-	(((cur) = bpf_iter_##type##_next(&___it)));						\
-)
-#endif /* bpf_for_each */
-
-#ifndef bpf_for
-/* bpf_for(i, start, end) implements a for()-like looping construct that sets
- * provided integer variable *i* to values starting from *start* through,
- * but not including, *end*. It also proves to BPF verifier that *i* belongs
- * to range [start, end), so this can be used for accessing arrays without
- * extra checks.
- *
- * Note: *start* and *end* are assumed to be expressions with no side effects
- * and whose values do not change throughout bpf_for() loop execution. They do
- * not have to be statically known or constant, though.
- *
- * Note: similarly to bpf_for_each(), it relies on C99 feature of declaring for()
- * loop bound variables and cleanup attribute, supported by GCC and Clang.
- */
-#define bpf_for(i, start, end) for (								\
-	/* initialize and define destructor */							\
-	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
-						 cleanup(bpf_iter_num_destroy))),		\
-	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
-			    *___p __attribute__((unused)) = (					\
-				bpf_iter_num_new(&___it, (start), (end)),			\
-	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
-	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
-				(void)bpf_iter_num_destroy, (void *)0);				\
-	({											\
-		/* iteration step */								\
-		int *___t = bpf_iter_num_next(&___it);						\
-		/* termination and bounds check */						\
-		(___t && ((i) = *___t, (i) >= (start) && (i) < (end)));				\
-	});											\
-)
-#endif /* bpf_for */
-
-#ifndef bpf_repeat
-/* bpf_repeat(N) performs N iterations without exposing iteration number
- *
- * Note: similarly to bpf_for_each(), it relies on C99 feature of declaring for()
- * loop bound variables and cleanup attribute, supported by GCC and Clang.
- */
-#define bpf_repeat(N) for (									\
-	/* initialize and define destructor */							\
-	struct bpf_iter_num ___it __attribute__((aligned(8), /* enforce, just in case */	\
-						 cleanup(bpf_iter_num_destroy))),		\
-	/* ___p pointer is necessary to call bpf_iter_num_new() *once* to init ___it */		\
-			    *___p __attribute__((unused)) = (					\
-				bpf_iter_num_new(&___it, 0, (N)),				\
-	/* this is a workaround for Clang bug: it currently doesn't emit BTF */			\
-	/* for bpf_iter_num_destroy() when used from cleanup() attribute */			\
-				(void)bpf_iter_num_destroy, (void *)0);				\
-	bpf_iter_num_next(&___it);								\
-	/* nothing here  */									\
-)
-#endif /* bpf_repeat */
-
 #endif
-- 
cgit v1.2.3-70-g09d2


From 94dccba7952072ce448a3278c66405fbb2a44ec5 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 17 Apr 2023 17:21:48 -0700
Subject: libbpf: mark bpf_iter_num_{new,next,destroy} as __weak

Mark bpf_iter_num_{new,next,destroy}() kfuncs declared for
bpf_for()/bpf_repeat() macros as __weak to allow users to feature-detect
their presence and guard bpf_for()/bpf_repeat() loops accordingly for
backwards compatibility with old kernels.

Now that libbpf supports kfunc calls poisoning and better reporting of
unresolved (but called) kfuncs, declaring number iterator kfuncs in
bpf_helpers.h won't degrade user experience and won't cause unnecessary
kernel feature dependencies.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230418002148.3255690-7-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/bpf_helpers.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index 525dec66c129..929a3baca8ef 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -293,9 +293,9 @@ enum libbpf_tristate {
 
 struct bpf_iter_num;
 
-extern int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end) __ksym;
-extern int *bpf_iter_num_next(struct bpf_iter_num *it) __ksym;
-extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __ksym;
+extern int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end) __weak __ksym;
+extern int *bpf_iter_num_next(struct bpf_iter_num *it) __weak __ksym;
+extern void bpf_iter_num_destroy(struct bpf_iter_num *it) __weak __ksym;
 
 #ifndef bpf_for_each
 /* bpf_for_each(iter_type, cur_elem, args...) provides generic construct for
-- 
cgit v1.2.3-70-g09d2


From 980f0799a15c75403f1f9284a32b6056b9660144 Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Tue, 18 Apr 2023 11:48:41 +0800
Subject: bonding: add software tx timestamping support

Currently, bonding only obtain the timestamp (ts) information of
the active slave, which is available only for modes 1, 5, and 6.
For other modes, bonding only has software rx timestamping support.

However, some users who use modes such as LACP also want tx timestamp
support. To address this issue, let's check the ts information of each
slave. If all slaves support tx timestamping, we can enable tx
timestamping support for the bond.

Add a note that the get_ts_info may be called with RCU, or rtnl or
reference on the device in ethtool.h>

Suggested-by: Miroslav Lichvar <mlichvar@redhat.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Link: https://lore.kernel.org/r/20230418034841.2566262-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/bonding/bond_main.c | 30 ++++++++++++++++++++++++++++++
 include/linux/ethtool.h         |  1 +
 2 files changed, 31 insertions(+)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 8cc9a74789b7..db7e650d9ebb 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5696,9 +5696,13 @@ static int bond_ethtool_get_ts_info(struct net_device *bond_dev,
 				    struct ethtool_ts_info *info)
 {
 	struct bonding *bond = netdev_priv(bond_dev);
+	struct ethtool_ts_info ts_info;
 	const struct ethtool_ops *ops;
 	struct net_device *real_dev;
+	bool sw_tx_support = false;
 	struct phy_device *phydev;
+	struct list_head *iter;
+	struct slave *slave;
 	int ret = 0;
 
 	rcu_read_lock();
@@ -5717,10 +5721,36 @@ static int bond_ethtool_get_ts_info(struct net_device *bond_dev,
 			ret = ops->get_ts_info(real_dev, info);
 			goto out;
 		}
+	} else {
+		/* Check if all slaves support software tx timestamping */
+		rcu_read_lock();
+		bond_for_each_slave_rcu(bond, slave, iter) {
+			ret = -1;
+			ops = slave->dev->ethtool_ops;
+			phydev = slave->dev->phydev;
+
+			if (phy_has_tsinfo(phydev))
+				ret = phy_ts_info(phydev, &ts_info);
+			else if (ops->get_ts_info)
+				ret = ops->get_ts_info(slave->dev, &ts_info);
+
+			if (!ret && (ts_info.so_timestamping & SOF_TIMESTAMPING_TX_SOFTWARE)) {
+				sw_tx_support = true;
+				continue;
+			}
+
+			sw_tx_support = false;
+			break;
+		}
+		rcu_read_unlock();
 	}
 
+	ret = 0;
 	info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
 				SOF_TIMESTAMPING_SOFTWARE;
+	if (sw_tx_support)
+		info->so_timestamping |= SOF_TIMESTAMPING_TX_SOFTWARE;
+
 	info->phc_index = -1;
 
 out:
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 798d35890118..62b61527bcc4 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -711,6 +711,7 @@ struct ethtool_mm_stats {
  * @get_dump_data: Get dump data.
  * @set_dump: Set dump specific flags to the device.
  * @get_ts_info: Get the time stamping and PTP hardware clock capabilities.
+ *	It may be called with RCU, or rtnl or reference on the device.
  *	Drivers supporting transmit time stamps in software should set this to
  *	ethtool_op_get_ts_info().
  * @get_module_info: Get the size and type of the eeprom contained within
-- 
cgit v1.2.3-70-g09d2


From 6b2c6e4a938fece9b539c8085f21d17c5e6eb9de Mon Sep 17 00:00:00 2001
From: Corinna Vinschen <vinschen@redhat.com>
Date: Mon, 17 Apr 2023 21:28:45 +0200
Subject: net: stmmac: propagate feature flags to vlan

stmmac_dev_probe doesn't propagate feature flags to VLANs.  So features
like offloading don't correspond with the general features and it's not
possible to manipulate features via ethtool -K to affect VLANs.

Propagate feature flags to vlan features.  Drop TSO feature because
it does not work on VLANs yet.

Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Link: https://lore.kernel.org/r/20230417192845.590034-1-vinschen@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index c0e90fda572a..5280256a638f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -7313,6 +7313,10 @@ int stmmac_dvr_probe(struct device *device,
 	if (priv->dma_cap.rssen && priv->plat->rss_en)
 		ndev->features |= NETIF_F_RXHASH;
 
+	ndev->vlan_features |= ndev->features;
+	/* TSO doesn't work on VLANs yet */
+	ndev->vlan_features &= ~NETIF_F_TSO;
+
 	/* MTU range: 46 - hw-specific max */
 	ndev->min_mtu = ETH_ZLEN - ETH_HLEN;
 	if (priv->plat->has_xgmac)
-- 
cgit v1.2.3-70-g09d2


From 33d74c8ff5ced854bc0ab6f94a778464dd2b905b Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 17 Apr 2023 22:55:25 +0200
Subject: net: mscc: ocelot: remove incompatible prototypes

The types for the register argument changed recently, but there are
still incompatible prototypes that got left behind, and gcc-13 warns
about these:

In file included from drivers/net/ethernet/mscc/ocelot.c:13:
drivers/net/ethernet/mscc/ocelot.h:97:5: error: conflicting types for 'ocelot_port_readl' due to enum/integer mismatch; have 'u32(struct ocelot_port *, u32)' {aka 'unsigned int(struct ocelot_port *, unsigned int)'} [-Werror=enum-int-mismatch]
   97 | u32 ocelot_port_readl(struct ocelot_port *port, u32 reg);
      |     ^~~~~~~~~~~~~~~~~

Just remove the two prototypes, and rely on the copy in the global
header.

Fixes: 9ecd05794b8d ("net: mscc: ocelot: strengthen type of "u32 reg" in I/O accessors")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20230417205531.1880657-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mscc/ocelot.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h
index 14440a3b04c3..87f2055c242c 100644
--- a/drivers/net/ethernet/mscc/ocelot.h
+++ b/drivers/net/ethernet/mscc/ocelot.h
@@ -94,9 +94,6 @@ int ocelot_mact_forget(struct ocelot *ocelot,
 struct net_device *ocelot_port_to_netdev(struct ocelot *ocelot, int port);
 int ocelot_netdev_to_port(struct net_device *dev);
 
-u32 ocelot_port_readl(struct ocelot_port *port, u32 reg);
-void ocelot_port_writel(struct ocelot_port *port, u32 val, u32 reg);
-
 int ocelot_probe_port(struct ocelot *ocelot, int port, struct regmap *target,
 		      struct device_node *portnp);
 void ocelot_release_port(struct ocelot_port *ocelot_port);
-- 
cgit v1.2.3-70-g09d2


From 14cac662235ea097b6ff97f558ce5066292806b9 Mon Sep 17 00:00:00 2001
From: Alain Volmat <avolmat@me.com>
Date: Sun, 16 Apr 2023 21:55:23 +0200
Subject: net: ethernet: stmmac: dwmac-sti: remove stih415/stih416/stid127

Remove no more supported platforms (stih415/stih416 and stid127)

Signed-off-by: Alain Volmat <avolmat@me.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20230416195523.61075-1-avolmat@me.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c | 60 +------------------------
 1 file changed, 1 insertion(+), 59 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c
index be3b1ebc06ab..465ce66ef9c1 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c
@@ -35,7 +35,7 @@
 #define IS_PHY_IF_MODE_GBIT(iface)	(IS_PHY_IF_MODE_RGMII(iface) || \
 					 iface == PHY_INTERFACE_MODE_GMII)
 
-/* STiH4xx register definitions (STiH415/STiH416/STiH407/STiH410 families)
+/* STiH4xx register definitions (STiH407/STiH410 families)
  *
  * Below table summarizes the clock requirement and clock sources for
  * supported phy interface modes with link speeds.
@@ -75,27 +75,6 @@
 #define STIH4XX_ETH_SEL_INTERNAL_NOTEXT_PHYCLK	BIT(7)
 #define STIH4XX_ETH_SEL_TXCLK_NOT_CLK125	BIT(6)
 
-/* STiD127 register definitions
- *-----------------------
- * src	 |BIT(6)| BIT(7)|
- *-----------------------
- * MII   |  1	|   n/a	|
- *-----------------------
- * RMII  |  n/a	|   1	|
- * clkgen|	|	|
- *-----------------------
- * RMII  |  n/a	|   0	|
- * phyclk|	|	|
- *-----------------------
- * RGMII |  1	|  n/a	|
- * clkgen|	|	|
- *-----------------------
- */
-
-#define STID127_RETIME_SRC_MASK			GENMASK(7, 6)
-#define STID127_ETH_SEL_INTERNAL_NOTEXT_PHYCLK	BIT(7)
-#define STID127_ETH_SEL_INTERNAL_NOTEXT_TXCLK	BIT(6)
-
 #define ENMII_MASK	GENMASK(5, 5)
 #define ENMII		BIT(5)
 #define EN_MASK		GENMASK(1, 1)
@@ -194,36 +173,6 @@ static void stih4xx_fix_retime_src(void *priv, u32 spd)
 			   stih4xx_tx_retime_val[src]);
 }
 
-static void stid127_fix_retime_src(void *priv, u32 spd)
-{
-	struct sti_dwmac *dwmac = priv;
-	u32 reg = dwmac->ctrl_reg;
-	u32 freq = 0;
-	u32 val = 0;
-
-	if (dwmac->interface == PHY_INTERFACE_MODE_MII) {
-		val = STID127_ETH_SEL_INTERNAL_NOTEXT_TXCLK;
-	} else if (dwmac->interface == PHY_INTERFACE_MODE_RMII) {
-		if (!dwmac->ext_phyclk) {
-			val = STID127_ETH_SEL_INTERNAL_NOTEXT_PHYCLK;
-			freq = DWMAC_50MHZ;
-		}
-	} else if (IS_PHY_IF_MODE_RGMII(dwmac->interface)) {
-		val = STID127_ETH_SEL_INTERNAL_NOTEXT_TXCLK;
-		if (spd == SPEED_1000)
-			freq = DWMAC_125MHZ;
-		else if (spd == SPEED_100)
-			freq = DWMAC_25MHZ;
-		else if (spd == SPEED_10)
-			freq = DWMAC_2_5MHZ;
-	}
-
-	if (freq)
-		clk_set_rate(dwmac->clk, freq);
-
-	regmap_update_bits(dwmac->regmap, reg, STID127_RETIME_SRC_MASK, val);
-}
-
 static int sti_dwmac_set_mode(struct sti_dwmac *dwmac)
 {
 	struct regmap *regmap = dwmac->regmap;
@@ -408,14 +357,7 @@ static const struct sti_dwmac_of_data stih4xx_dwmac_data = {
 	.fix_retime_src = stih4xx_fix_retime_src,
 };
 
-static const struct sti_dwmac_of_data stid127_dwmac_data = {
-	.fix_retime_src = stid127_fix_retime_src,
-};
-
 static const struct of_device_id sti_dwmac_match[] = {
-	{ .compatible = "st,stih415-dwmac", .data = &stih4xx_dwmac_data},
-	{ .compatible = "st,stih416-dwmac", .data = &stih4xx_dwmac_data},
-	{ .compatible = "st,stid127-dwmac", .data = &stid127_dwmac_data},
 	{ .compatible = "st,stih407-dwmac", .data = &stih4xx_dwmac_data},
 	{ }
 };
-- 
cgit v1.2.3-70-g09d2


From ef16799640865f937719f0771c93be5dca18adc6 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Tue, 18 Apr 2023 15:25:46 +0200
Subject: wifi: iwlwifi: dvm: Fix memcpy: detected field-spanning write
 backtrace

A received TKIP key may be up to 32 bytes because it may contain
MIC rx/tx keys too. These are not used by iwl and copying these
over overflows the iwl_keyinfo.key field.

Add a check to not copy more data to iwl_keyinfo.key then will fit.

This fixes backtraces like this one:

 memcpy: detected field-spanning write (size 32) of single field "sta_cmd.key.key" at drivers/net/wireless/intel/iwlwifi/dvm/sta.c:1103 (size 16)
 WARNING: CPU: 1 PID: 946 at drivers/net/wireless/intel/iwlwifi/dvm/sta.c:1103 iwlagn_send_sta_key+0x375/0x390 [iwldvm]
 <snip>
 Hardware name: Dell Inc. Latitude E6430/0H3MT5, BIOS A21 05/08/2017
 RIP: 0010:iwlagn_send_sta_key+0x375/0x390 [iwldvm]
 <snip>
 Call Trace:
  <TASK>
  iwl_set_dynamic_key+0x1f0/0x220 [iwldvm]
  iwlagn_mac_set_key+0x1e4/0x280 [iwldvm]
  drv_set_key+0xa4/0x1b0 [mac80211]
  ieee80211_key_enable_hw_accel+0xa8/0x2d0 [mac80211]
  ieee80211_key_replace+0x22d/0x8e0 [mac80211]
 <snip>

Link: https://www.alionet.org/index.php?topic=1469.0
Link: https://lore.kernel.org/linux-wireless/20230218191056.never.374-kees@kernel.org/
Link: https://lore.kernel.org/linux-wireless/68760035-7f75-1b23-e355-bfb758a87d83@redhat.com/
Cc: Kees Cook <keescook@chromium.org>
Suggested-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/dvm/sta.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/sta.c b/drivers/net/wireless/intel/iwlwifi/dvm/sta.c
index cef43cf80620..8b01ab986cb1 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/sta.c
@@ -1081,6 +1081,7 @@ static int iwlagn_send_sta_key(struct iwl_priv *priv,
 {
 	__le16 key_flags;
 	struct iwl_addsta_cmd sta_cmd;
+	size_t to_copy;
 	int i;
 
 	spin_lock_bh(&priv->sta_lock);
@@ -1100,7 +1101,9 @@ static int iwlagn_send_sta_key(struct iwl_priv *priv,
 		sta_cmd.key.tkip_rx_tsc_byte2 = tkip_iv32;
 		for (i = 0; i < 5; i++)
 			sta_cmd.key.tkip_rx_ttak[i] = cpu_to_le16(tkip_p1k[i]);
-		memcpy(sta_cmd.key.key, keyconf->key, keyconf->keylen);
+		/* keyconf may contain MIC rx/tx keys which iwl does not use */
+		to_copy = min_t(size_t, sizeof(sta_cmd.key.key), keyconf->keylen);
+		memcpy(sta_cmd.key.key, keyconf->key, to_copy);
 		break;
 	case WLAN_CIPHER_SUITE_WEP104:
 		key_flags |= STA_KEY_FLG_KEY_SIZE_MSK;
-- 
cgit v1.2.3-70-g09d2


From aebc62d336fed235e35d8522a004866c78dfa5c4 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:49 +0300
Subject: net/mlx5e: Move XDP struct and enum to XDP header

Move struct mlx5e_xdp_info and enum mlx5e_xdp_xmit_mode from the generic
en.h to the XDP header, where they belong.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h     | 35 ------------------------
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h | 35 ++++++++++++++++++++++++
 2 files changed, 35 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index ba615b74bb8e..3f5463d42a1e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -475,41 +475,6 @@ struct mlx5e_txqsq {
 	cqe_ts_to_ns               ptp_cyc2time;
 } ____cacheline_aligned_in_smp;
 
-/* XDP packets can be transmitted in different ways. On completion, we need to
- * distinguish between them to clean up things in a proper way.
- */
-enum mlx5e_xdp_xmit_mode {
-	/* An xdp_frame was transmitted due to either XDP_REDIRECT from another
-	 * device or XDP_TX from an XSK RQ. The frame has to be unmapped and
-	 * returned.
-	 */
-	MLX5E_XDP_XMIT_MODE_FRAME,
-
-	/* The xdp_frame was created in place as a result of XDP_TX from a
-	 * regular RQ. No DMA remapping happened, and the page belongs to us.
-	 */
-	MLX5E_XDP_XMIT_MODE_PAGE,
-
-	/* No xdp_frame was created at all, the transmit happened from a UMEM
-	 * page. The UMEM Completion Ring producer pointer has to be increased.
-	 */
-	MLX5E_XDP_XMIT_MODE_XSK,
-};
-
-struct mlx5e_xdp_info {
-	enum mlx5e_xdp_xmit_mode mode;
-	union {
-		struct {
-			struct xdp_frame *xdpf;
-			dma_addr_t dma_addr;
-		} frame;
-		struct {
-			struct mlx5e_rq *rq;
-			struct page *page;
-		} page;
-	};
-};
-
 struct mlx5e_xmit_data {
 	dma_addr_t  dma_addr;
 	void       *data;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
index 10bcfa6f88c1..8208692035f8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
@@ -50,6 +50,41 @@ struct mlx5e_xdp_buff {
 	struct mlx5e_rq *rq;
 };
 
+/* XDP packets can be transmitted in different ways. On completion, we need to
+ * distinguish between them to clean up things in a proper way.
+ */
+enum mlx5e_xdp_xmit_mode {
+	/* An xdp_frame was transmitted due to either XDP_REDIRECT from another
+	 * device or XDP_TX from an XSK RQ. The frame has to be unmapped and
+	 * returned.
+	 */
+	MLX5E_XDP_XMIT_MODE_FRAME,
+
+	/* The xdp_frame was created in place as a result of XDP_TX from a
+	 * regular RQ. No DMA remapping happened, and the page belongs to us.
+	 */
+	MLX5E_XDP_XMIT_MODE_PAGE,
+
+	/* No xdp_frame was created at all, the transmit happened from a UMEM
+	 * page. The UMEM Completion Ring producer pointer has to be increased.
+	 */
+	MLX5E_XDP_XMIT_MODE_XSK,
+};
+
+struct mlx5e_xdp_info {
+	enum mlx5e_xdp_xmit_mode mode;
+	union {
+		struct {
+			struct xdp_frame *xdpf;
+			dma_addr_t dma_addr;
+		} frame;
+		struct {
+			struct mlx5e_rq *rq;
+			struct page *page;
+		} page;
+	};
+};
+
 struct mlx5e_xsk_param;
 int mlx5e_xdp_max_mtu(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk);
 bool mlx5e_xdp_handle(struct mlx5e_rq *rq,
-- 
cgit v1.2.3-70-g09d2


From e32654f198897b0323e52fbd96a38606d0289c35 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:50 +0300
Subject: net/mlx5e: Move struct mlx5e_xmit_data to datapath header

Move TX datapath struct from the generic en.h to the datapath txrx.h
header, where it belongs.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      | 7 +------
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h | 6 ++++++
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 3f5463d42a1e..479979318c50 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -475,12 +475,6 @@ struct mlx5e_txqsq {
 	cqe_ts_to_ns               ptp_cyc2time;
 } ____cacheline_aligned_in_smp;
 
-struct mlx5e_xmit_data {
-	dma_addr_t  dma_addr;
-	void       *data;
-	u32         len;
-};
-
 struct mlx5e_xdp_info_fifo {
 	struct mlx5e_xdp_info *xi;
 	u32 *cc;
@@ -489,6 +483,7 @@ struct mlx5e_xdp_info_fifo {
 };
 
 struct mlx5e_xdpsq;
+struct mlx5e_xmit_data;
 typedef int (*mlx5e_fp_xmit_xdp_frame_check)(struct mlx5e_xdpsq *);
 typedef bool (*mlx5e_fp_xmit_xdp_frame)(struct mlx5e_xdpsq *,
 					struct mlx5e_xmit_data *,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 651be7aaf7d5..6f7ebedda279 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -77,6 +77,12 @@ static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config)
 }
 
 /* TX */
+struct mlx5e_xmit_data {
+	dma_addr_t  dma_addr;
+	void       *data;
+	u32         len;
+};
+
 netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
 bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget);
 void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq);
-- 
cgit v1.2.3-70-g09d2


From eb9b9fdcafe29f022260dc43bd001b85937d9f83 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:51 +0300
Subject: net/mlx5e: Introduce extended version for mlx5e_xmit_data

Introduce struct mlx5e_xmit_data_frags to be used for non-linear xmit
buffers. Let it include sinfo pointer.

Take one bit from the len field to indicate if the descriptor has
fragments and can be casted-up into the extended version.

Zero-init to make sure has_frags, and potentially future fields, are
zero when not explicitly assigned.

Another field will be added in a downstream patch to indicate and point
to dma addresses of the different frags, for redirect-in requests.

This simplifies the mlx5e_xmit_xdp_frame/mlx5e_xmit_xdp_frame_mpwqe
functions params.

Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  1 -
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h  |  8 ++-
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c   | 63 ++++++++++++----------
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h   |  2 -
 .../net/ethernet/mellanox/mlx5/core/en/xsk/tx.c    |  4 +-
 5 files changed, 44 insertions(+), 34 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 479979318c50..386f5a498e52 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -487,7 +487,6 @@ struct mlx5e_xmit_data;
 typedef int (*mlx5e_fp_xmit_xdp_frame_check)(struct mlx5e_xdpsq *);
 typedef bool (*mlx5e_fp_xmit_xdp_frame)(struct mlx5e_xdpsq *,
 					struct mlx5e_xmit_data *,
-					struct skb_shared_info *,
 					int);
 
 struct mlx5e_xdpsq {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 6f7ebedda279..1302f52db883 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -80,7 +80,13 @@ static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config)
 struct mlx5e_xmit_data {
 	dma_addr_t  dma_addr;
 	void       *data;
-	u32         len;
+	u32         len : 31;
+	u32         has_frags : 1;
+};
+
+struct mlx5e_xmit_data_frags {
+	struct mlx5e_xmit_data xd;
+	struct skb_shared_info *sinfo;
 };
 
 netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index c8b532cea7d1..3e7ebf0f0f01 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -61,8 +61,8 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 		    struct xdp_buff *xdp)
 {
 	struct page *page = virt_to_page(xdp->data);
-	struct skb_shared_info *sinfo = NULL;
-	struct mlx5e_xmit_data xdptxd;
+	struct mlx5e_xmit_data_frags xdptxdf = {};
+	struct mlx5e_xmit_data *xdptxd;
 	struct mlx5e_xdp_info xdpi;
 	struct xdp_frame *xdpf;
 	dma_addr_t dma_addr;
@@ -72,8 +72,10 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	if (unlikely(!xdpf))
 		return false;
 
-	xdptxd.data = xdpf->data;
-	xdptxd.len  = xdpf->len;
+	xdptxd = &xdptxdf.xd;
+	xdptxd->data = xdpf->data;
+	xdptxd->len  = xdpf->len;
+	xdptxd->has_frags = xdp_frame_has_frags(xdpf);
 
 	if (xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL) {
 		/* The xdp_buff was in the UMEM and was copied into a newly
@@ -90,19 +92,22 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 
 		xdpi.mode = MLX5E_XDP_XMIT_MODE_FRAME;
 
-		dma_addr = dma_map_single(sq->pdev, xdptxd.data, xdptxd.len,
+		if (unlikely(xdptxd->has_frags))
+			return false;
+
+		dma_addr = dma_map_single(sq->pdev, xdptxd->data, xdptxd->len,
 					  DMA_TO_DEVICE);
 		if (dma_mapping_error(sq->pdev, dma_addr)) {
 			xdp_return_frame(xdpf);
 			return false;
 		}
 
-		xdptxd.dma_addr     = dma_addr;
+		xdptxd->dma_addr = dma_addr;
 		xdpi.frame.xdpf     = xdpf;
 		xdpi.frame.dma_addr = dma_addr;
 
 		if (unlikely(!INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
-					      mlx5e_xmit_xdp_frame, sq, &xdptxd, NULL, 0)))
+					      mlx5e_xmit_xdp_frame, sq, xdptxd, 0)))
 			return false;
 
 		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
@@ -119,13 +124,13 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	xdpi.page.rq = rq;
 
 	dma_addr = page_pool_get_dma_addr(page) + (xdpf->data - (void *)xdpf);
-	dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd.len, DMA_BIDIRECTIONAL);
+	dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd->len, DMA_BIDIRECTIONAL);
 
-	if (unlikely(xdp_frame_has_frags(xdpf))) {
-		sinfo = xdp_get_shared_info_from_frame(xdpf);
+	if (unlikely(xdptxd->has_frags)) {
+		xdptxdf.sinfo = xdp_get_shared_info_from_frame(xdpf);
 
-		for (i = 0; i < sinfo->nr_frags; i++) {
-			skb_frag_t *frag = &sinfo->frags[i];
+		for (i = 0; i < xdptxdf.sinfo->nr_frags; i++) {
+			skb_frag_t *frag = &xdptxdf.sinfo->frags[i];
 			dma_addr_t addr;
 			u32 len;
 
@@ -137,18 +142,18 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 		}
 	}
 
-	xdptxd.dma_addr = dma_addr;
+	xdptxd->dma_addr = dma_addr;
 
 	if (unlikely(!INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
-				      mlx5e_xmit_xdp_frame, sq, &xdptxd, sinfo, 0)))
+				      mlx5e_xmit_xdp_frame, sq, xdptxd, 0)))
 		return false;
 
 	xdpi.page.page = page;
 	mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
 
-	if (unlikely(xdp_frame_has_frags(xdpf))) {
-		for (i = 0; i < sinfo->nr_frags; i++) {
-			skb_frag_t *frag = &sinfo->frags[i];
+	if (unlikely(xdptxd->has_frags)) {
+		for (i = 0; i < xdptxdf.sinfo->nr_frags; i++) {
+			skb_frag_t *frag = &xdptxdf.sinfo->frags[i];
 
 			xdpi.page.page = skb_frag_page(frag);
 			mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
@@ -381,23 +386,23 @@ INDIRECT_CALLABLE_SCOPE int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq
 
 INDIRECT_CALLABLE_SCOPE bool
 mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
-		     struct skb_shared_info *sinfo, int check_result);
+		     int check_result);
 
 INDIRECT_CALLABLE_SCOPE bool
 mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
-			   struct skb_shared_info *sinfo, int check_result)
+			   int check_result)
 {
 	struct mlx5e_tx_mpwqe *session = &sq->mpwqe;
 	struct mlx5e_xdpsq_stats *stats = sq->stats;
 
-	if (unlikely(sinfo)) {
+	if (unlikely(xdptxd->has_frags)) {
 		/* MPWQE is enabled, but a multi-buffer packet is queued for
 		 * transmission. MPWQE can't send fragmented packets, so close
 		 * the current session and fall back to a regular WQE.
 		 */
 		if (unlikely(sq->mpwqe.wqe))
 			mlx5e_xdp_mpwqe_complete(sq);
-		return mlx5e_xmit_xdp_frame(sq, xdptxd, sinfo, 0);
+		return mlx5e_xmit_xdp_frame(sq, xdptxd, 0);
 	}
 
 	if (unlikely(xdptxd->len > sq->hw_mtu)) {
@@ -446,8 +451,10 @@ INDIRECT_CALLABLE_SCOPE int mlx5e_xmit_xdp_frame_check(struct mlx5e_xdpsq *sq)
 
 INDIRECT_CALLABLE_SCOPE bool
 mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
-		     struct skb_shared_info *sinfo, int check_result)
+		     int check_result)
 {
+	struct mlx5e_xmit_data_frags *xdptxdf =
+		container_of(xdptxd, struct mlx5e_xmit_data_frags, xd);
 	struct mlx5_wq_cyc       *wq   = &sq->wq;
 	struct mlx5_wqe_ctrl_seg *cseg;
 	struct mlx5_wqe_data_seg *dseg;
@@ -476,9 +483,9 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 	if (!check_result) {
 		int stop_room = 1;
 
-		if (unlikely(sinfo)) {
-			ds_cnt += sinfo->nr_frags;
-			num_frags = sinfo->nr_frags;
+		if (unlikely(xdptxd->has_frags)) {
+			ds_cnt += xdptxdf->sinfo->nr_frags;
+			num_frags = xdptxdf->sinfo->nr_frags;
 			num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS);
 			/* Assuming MLX5_CAP_GEN(mdev, max_wqe_sz_sq) is big
 			 * enough to hold all fragments.
@@ -529,7 +536,7 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 		dseg->lkey = sq->mkey_be;
 
 		for (i = 0; i < num_frags; i++) {
-			skb_frag_t *frag = &sinfo->frags[i];
+			skb_frag_t *frag = &xdptxdf->sinfo->frags[i];
 			dma_addr_t addr;
 
 			addr = page_pool_get_dma_addr(skb_frag_page(frag)) +
@@ -718,7 +725,7 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 
 	for (i = 0; i < n; i++) {
 		struct xdp_frame *xdpf = frames[i];
-		struct mlx5e_xmit_data xdptxd;
+		struct mlx5e_xmit_data xdptxd = {};
 		struct mlx5e_xdp_info xdpi;
 		bool ret;
 
@@ -735,7 +742,7 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		xdpi.frame.dma_addr = xdptxd.dma_addr;
 
 		ret = INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
-				      mlx5e_xmit_xdp_frame, sq, &xdptxd, NULL, 0);
+				      mlx5e_xmit_xdp_frame, sq, &xdptxd, 0);
 		if (unlikely(!ret)) {
 			dma_unmap_single(sq->pdev, xdptxd.dma_addr,
 					 xdptxd.len, DMA_TO_DEVICE);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
index 8208692035f8..8e97c68d11f4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
@@ -101,11 +101,9 @@ extern const struct xdp_metadata_ops mlx5e_xdp_metadata_ops;
 
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq,
 							  struct mlx5e_xmit_data *xdptxd,
-							  struct skb_shared_info *sinfo,
 							  int check_result));
 INDIRECT_CALLABLE_DECLARE(bool mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq,
 						    struct mlx5e_xmit_data *xdptxd,
-						    struct skb_shared_info *sinfo,
 						    int check_result));
 INDIRECT_CALLABLE_DECLARE(int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq *sq));
 INDIRECT_CALLABLE_DECLARE(int mlx5e_xmit_xdp_frame_check(struct mlx5e_xdpsq *sq));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
index 367a9505ca4f..b370a4daddfd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
@@ -61,7 +61,6 @@ static void mlx5e_xsk_tx_post_err(struct mlx5e_xdpsq *sq,
 bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget)
 {
 	struct xsk_buff_pool *pool = sq->xsk_pool;
-	struct mlx5e_xmit_data xdptxd;
 	struct mlx5e_xdp_info xdpi;
 	bool work_done = true;
 	bool flush = false;
@@ -73,6 +72,7 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget)
 						   mlx5e_xmit_xdp_frame_check_mpwqe,
 						   mlx5e_xmit_xdp_frame_check,
 						   sq);
+		struct mlx5e_xmit_data xdptxd = {};
 		struct xdp_desc desc;
 		bool ret;
 
@@ -97,7 +97,7 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget)
 		xsk_buff_raw_dma_sync_for_device(pool, xdptxd.dma_addr, xdptxd.len);
 
 		ret = INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
-				      mlx5e_xmit_xdp_frame, sq, &xdptxd, NULL,
+				      mlx5e_xmit_xdp_frame, sq, &xdptxd,
 				      check_result);
 		if (unlikely(!ret)) {
 			if (sq->mpwqe.wqe)
-- 
cgit v1.2.3-70-g09d2


From 3a48ba12b4e471a9a3cf5fcff5146632ae6453d5 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:52 +0300
Subject: net/mlx5e: XDP, Remove doubtful unlikely calls

It is not likely nor unlikely that the xdp buff has fragments, it
depends on the program loaded and size of the packet received.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 3e7ebf0f0f01..dcae2d4e2c03 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -126,7 +126,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	dma_addr = page_pool_get_dma_addr(page) + (xdpf->data - (void *)xdpf);
 	dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd->len, DMA_BIDIRECTIONAL);
 
-	if (unlikely(xdptxd->has_frags)) {
+	if (xdptxd->has_frags) {
 		xdptxdf.sinfo = xdp_get_shared_info_from_frame(xdpf);
 
 		for (i = 0; i < xdptxdf.sinfo->nr_frags; i++) {
@@ -151,7 +151,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	xdpi.page.page = page;
 	mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
 
-	if (unlikely(xdptxd->has_frags)) {
+	if (xdptxd->has_frags) {
 		for (i = 0; i < xdptxdf.sinfo->nr_frags; i++) {
 			skb_frag_t *frag = &xdptxdf.sinfo->frags[i];
 
@@ -395,7 +395,7 @@ mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptx
 	struct mlx5e_tx_mpwqe *session = &sq->mpwqe;
 	struct mlx5e_xdpsq_stats *stats = sq->stats;
 
-	if (unlikely(xdptxd->has_frags)) {
+	if (xdptxd->has_frags) {
 		/* MPWQE is enabled, but a multi-buffer packet is queued for
 		 * transmission. MPWQE can't send fragmented packets, so close
 		 * the current session and fall back to a regular WQE.
@@ -483,7 +483,7 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 	if (!check_result) {
 		int stop_room = 1;
 
-		if (unlikely(xdptxd->has_frags)) {
+		if (xdptxd->has_frags) {
 			ds_cnt += xdptxdf->sinfo->nr_frags;
 			num_frags = xdptxdf->sinfo->nr_frags;
 			num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS);
@@ -525,7 +525,7 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_SEND);
 
-	if (unlikely(test_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state))) {
+	if (test_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state)) {
 		u8 num_pkts = 1 + num_frags;
 		int i;
 
-- 
cgit v1.2.3-70-g09d2


From 3f734b8c594bd378b0266cd98d6ce12b459da1d4 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:53 +0300
Subject: net/mlx5e: XDP, Use multiple single-entry objects in xdpi_fifo

Here we fix the current wi->num_pkts abuse, as it was used to indicate
multiple xdpi entries in the xdpi_fifo.

Instead, reduce mlx5e_xdp_info to the size of a single field, making it
a union of unions. Per packet, use as many instances as needed to
provide the information needed at the time of completion.

The sequence of xdpi instances pushed is well defined, derived by the
xmit_mode.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c   | 95 +++++++++++++++-------
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h   | 38 ++++++---
 .../net/ethernet/mellanox/mlx5/core/en/xsk/tx.c    |  8 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  8 +-
 5 files changed, 101 insertions(+), 50 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 386f5a498e52..0e15afbe1673 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -476,7 +476,7 @@ struct mlx5e_txqsq {
 } ____cacheline_aligned_in_smp;
 
 struct mlx5e_xdp_info_fifo {
-	struct mlx5e_xdp_info *xi;
+	union mlx5e_xdp_info *xi;
 	u32 *cc;
 	u32 *pc;
 	u32 mask;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index dcae2d4e2c03..5dab9012dc2a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -63,7 +63,6 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	struct page *page = virt_to_page(xdp->data);
 	struct mlx5e_xmit_data_frags xdptxdf = {};
 	struct mlx5e_xmit_data *xdptxd;
-	struct mlx5e_xdp_info xdpi;
 	struct xdp_frame *xdpf;
 	dma_addr_t dma_addr;
 	int i;
@@ -90,8 +89,6 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 		 */
 		__set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags); /* non-atomic */
 
-		xdpi.mode = MLX5E_XDP_XMIT_MODE_FRAME;
-
 		if (unlikely(xdptxd->has_frags))
 			return false;
 
@@ -103,14 +100,18 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 		}
 
 		xdptxd->dma_addr = dma_addr;
-		xdpi.frame.xdpf     = xdpf;
-		xdpi.frame.dma_addr = dma_addr;
 
 		if (unlikely(!INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
 					      mlx5e_xmit_xdp_frame, sq, xdptxd, 0)))
 			return false;
 
-		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
+		/* xmit_mode == MLX5E_XDP_XMIT_MODE_FRAME */
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .mode = MLX5E_XDP_XMIT_MODE_FRAME });
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .frame.xdpf = xdpf });
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .frame.dma_addr = dma_addr });
 		return true;
 	}
 
@@ -120,9 +121,6 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 	 * mode.
 	 */
 
-	xdpi.mode = MLX5E_XDP_XMIT_MODE_PAGE;
-	xdpi.page.rq = rq;
-
 	dma_addr = page_pool_get_dma_addr(page) + (xdpf->data - (void *)xdpf);
 	dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd->len, DMA_BIDIRECTIONAL);
 
@@ -148,16 +146,28 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 				      mlx5e_xmit_xdp_frame, sq, xdptxd, 0)))
 		return false;
 
-	xdpi.page.page = page;
-	mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
+	/* xmit_mode == MLX5E_XDP_XMIT_MODE_PAGE */
+	mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+			     (union mlx5e_xdp_info) { .mode = MLX5E_XDP_XMIT_MODE_PAGE });
 
 	if (xdptxd->has_frags) {
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info)
+				     { .page.num = 1 + xdptxdf.sinfo->nr_frags });
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .page.page = page });
 		for (i = 0; i < xdptxdf.sinfo->nr_frags; i++) {
 			skb_frag_t *frag = &xdptxdf.sinfo->frags[i];
 
-			xdpi.page.page = skb_frag_page(frag);
-			mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
+			mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+					     (union mlx5e_xdp_info)
+					     { .page.page = skb_frag_page(frag) });
 		}
+	} else {
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .page.num = 1 });
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .page.page = page });
 	}
 
 	return true;
@@ -526,7 +536,6 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_SEND);
 
 	if (test_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state)) {
-		u8 num_pkts = 1 + num_frags;
 		int i;
 
 		memset(&cseg->trailer, 0, sizeof(cseg->trailer));
@@ -552,7 +561,7 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 
 		sq->db.wqe_info[pi] = (struct mlx5e_xdp_wqe_info) {
 			.num_wqebbs = num_wqebbs,
-			.num_pkts = num_pkts,
+			.num_pkts = 1,
 		};
 
 		sq->pc += num_wqebbs;
@@ -577,20 +586,46 @@ static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq,
 	u16 i;
 
 	for (i = 0; i < wi->num_pkts; i++) {
-		struct mlx5e_xdp_info xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+		union mlx5e_xdp_info xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
 
 		switch (xdpi.mode) {
-		case MLX5E_XDP_XMIT_MODE_FRAME:
+		case MLX5E_XDP_XMIT_MODE_FRAME: {
 			/* XDP_TX from the XSK RQ and XDP_REDIRECT */
-			dma_unmap_single(sq->pdev, xdpi.frame.dma_addr,
-					 xdpi.frame.xdpf->len, DMA_TO_DEVICE);
-			xdp_return_frame_bulk(xdpi.frame.xdpf, bq);
+			struct xdp_frame *xdpf;
+			dma_addr_t dma_addr;
+
+			xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+			xdpf = xdpi.frame.xdpf;
+			xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+			dma_addr = xdpi.frame.dma_addr;
+
+			dma_unmap_single(sq->pdev, dma_addr,
+					 xdpf->len, DMA_TO_DEVICE);
+			xdp_return_frame_bulk(xdpf, bq);
 			break;
-		case MLX5E_XDP_XMIT_MODE_PAGE:
+		}
+		case MLX5E_XDP_XMIT_MODE_PAGE: {
 			/* XDP_TX from the regular RQ */
-			page_pool_put_defragged_page(xdpi.page.rq->page_pool,
-						     xdpi.page.page, -1, true);
+			u8 num, n = 0;
+
+			xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+			num = xdpi.page.num;
+
+			do {
+				struct page *page;
+
+				xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+				page = xdpi.page.page;
+
+				/* No need to check ((page->pp_magic & ~0x3UL) == PP_SIGNATURE)
+				 * as we know this is a page_pool page.
+				 */
+				page_pool_put_defragged_page(page->pp,
+							     page, -1, true);
+			} while (++n < num);
+
 			break;
+		}
 		case MLX5E_XDP_XMIT_MODE_XSK:
 			/* AF_XDP send */
 			(*xsk_frames)++;
@@ -726,7 +761,6 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 	for (i = 0; i < n; i++) {
 		struct xdp_frame *xdpf = frames[i];
 		struct mlx5e_xmit_data xdptxd = {};
-		struct mlx5e_xdp_info xdpi;
 		bool ret;
 
 		xdptxd.data = xdpf->data;
@@ -737,10 +771,6 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		if (unlikely(dma_mapping_error(sq->pdev, xdptxd.dma_addr)))
 			break;
 
-		xdpi.mode           = MLX5E_XDP_XMIT_MODE_FRAME;
-		xdpi.frame.xdpf     = xdpf;
-		xdpi.frame.dma_addr = xdptxd.dma_addr;
-
 		ret = INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
 				      mlx5e_xmit_xdp_frame, sq, &xdptxd, 0);
 		if (unlikely(!ret)) {
@@ -748,7 +778,14 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 					 xdptxd.len, DMA_TO_DEVICE);
 			break;
 		}
-		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
+
+		/* xmit_mode == MLX5E_XDP_XMIT_MODE_FRAME */
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .mode = MLX5E_XDP_XMIT_MODE_FRAME });
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .frame.xdpf = xdpf });
+		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+				     (union mlx5e_xdp_info) { .frame.dma_addr = xdptxd.dma_addr });
 		nxmit++;
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
index 8e97c68d11f4..9e8e6184f9e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h
@@ -71,18 +71,30 @@ enum mlx5e_xdp_xmit_mode {
 	MLX5E_XDP_XMIT_MODE_XSK,
 };
 
-struct mlx5e_xdp_info {
+/* xmit_mode entry is pushed to the fifo per packet, followed by multiple
+ * entries, as follows:
+ *
+ * MLX5E_XDP_XMIT_MODE_FRAME:
+ *    xdpf, dma_addr_1, dma_addr_2, ... , dma_addr_num.
+ *    'num' is derived from xdpf.
+ *
+ * MLX5E_XDP_XMIT_MODE_PAGE:
+ *    num, page_1, page_2, ... , page_num.
+ *
+ * MLX5E_XDP_XMIT_MODE_XSK:
+ *    none.
+ */
+union mlx5e_xdp_info {
 	enum mlx5e_xdp_xmit_mode mode;
 	union {
-		struct {
-			struct xdp_frame *xdpf;
-			dma_addr_t dma_addr;
-		} frame;
-		struct {
-			struct mlx5e_rq *rq;
-			struct page *page;
-		} page;
-	};
+		struct xdp_frame *xdpf;
+		dma_addr_t dma_addr;
+	} frame;
+	union {
+		struct mlx5e_rq *rq;
+		u8 num;
+		struct page *page;
+	} page;
 };
 
 struct mlx5e_xsk_param;
@@ -212,14 +224,14 @@ mlx5e_xdp_mpwqe_add_dseg(struct mlx5e_xdpsq *sq,
 
 static inline void
 mlx5e_xdpi_fifo_push(struct mlx5e_xdp_info_fifo *fifo,
-		     struct mlx5e_xdp_info *xi)
+		     union mlx5e_xdp_info xi)
 {
 	u32 i = (*fifo->pc)++ & fifo->mask;
 
-	fifo->xi[i] = *xi;
+	fifo->xi[i] = xi;
 }
 
-static inline struct mlx5e_xdp_info
+static inline union mlx5e_xdp_info
 mlx5e_xdpi_fifo_pop(struct mlx5e_xdp_info_fifo *fifo)
 {
 	return fifo->xi[(*fifo->cc)++ & fifo->mask];
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
index b370a4daddfd..597f319d4770 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
@@ -44,7 +44,7 @@ int mlx5e_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags)
  * same.
  */
 static void mlx5e_xsk_tx_post_err(struct mlx5e_xdpsq *sq,
-				  struct mlx5e_xdp_info *xdpi)
+				  union mlx5e_xdp_info *xdpi)
 {
 	u16 pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc);
 	struct mlx5e_xdp_wqe_info *wi = &sq->db.wqe_info[pi];
@@ -54,14 +54,14 @@ static void mlx5e_xsk_tx_post_err(struct mlx5e_xdpsq *sq,
 	wi->num_pkts = 1;
 
 	nopwqe = mlx5e_post_nop(&sq->wq, sq->sqn, &sq->pc);
-	mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, xdpi);
+	mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, *xdpi);
 	sq->doorbell_cseg = &nopwqe->ctrl;
 }
 
 bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget)
 {
 	struct xsk_buff_pool *pool = sq->xsk_pool;
-	struct mlx5e_xdp_info xdpi;
+	union mlx5e_xdp_info xdpi;
 	bool work_done = true;
 	bool flush = false;
 
@@ -105,7 +105,7 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget)
 
 			mlx5e_xsk_tx_post_err(sq, &xdpi);
 		} else {
-			mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, &xdpi);
+			mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, xdpi);
 		}
 
 		flush = true;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index ec72743b64e2..0b5aafaefe4c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1300,17 +1300,19 @@ static int mlx5e_alloc_xdpsq_fifo(struct mlx5e_xdpsq *sq, int numa)
 {
 	struct mlx5e_xdp_info_fifo *xdpi_fifo = &sq->db.xdpi_fifo;
 	int wq_sz        = mlx5_wq_cyc_get_size(&sq->wq);
-	int dsegs_per_wq = wq_sz * MLX5_SEND_WQEBB_NUM_DS;
+	int entries = wq_sz * MLX5_SEND_WQEBB_NUM_DS * 2; /* upper bound for maximum num of
+							   * entries of all xmit_modes.
+							   */
 	size_t size;
 
-	size = array_size(sizeof(*xdpi_fifo->xi), dsegs_per_wq);
+	size = array_size(sizeof(*xdpi_fifo->xi), entries);
 	xdpi_fifo->xi = kvzalloc_node(size, GFP_KERNEL, numa);
 	if (!xdpi_fifo->xi)
 		return -ENOMEM;
 
 	xdpi_fifo->pc   = &sq->xdpi_fifo_pc;
 	xdpi_fifo->cc   = &sq->xdpi_fifo_cc;
-	xdpi_fifo->mask = dsegs_per_wq - 1;
+	xdpi_fifo->mask = entries - 1;
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From c1783e74fc042adc28907ac7229b01ed20da31f9 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:54 +0300
Subject: net/mlx5e: XDP, Add support for multi-buffer XDP redirect-in

Handle multi-buffer XDP redirect-in requests coming through
mlx5e_xdp_xmit.

Extend struct mlx5e_xmit_data_frags with an additional dma_arr field, to
point to the fragments dma mapping, as they cannot be retrieved via the
page_pool_get_dma_addr() function.

Push a dma_addr xdpi instance per each fragment, and use them in the
completion flow to dma_unmap the frags.

Finally, remove the restriction in mlx5e_open_xdpsq, and set the flag in
xdp_features.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c  | 82 ++++++++++++++++++++---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  9 +--
 3 files changed, 75 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 1302f52db883..47381e949f1f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -87,6 +87,7 @@ struct mlx5e_xmit_data {
 struct mlx5e_xmit_data_frags {
 	struct mlx5e_xmit_data xd;
 	struct skb_shared_info *sinfo;
+	dma_addr_t *dma_arr;
 };
 
 netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 5dab9012dc2a..c266d073e2f2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -126,6 +126,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq,
 
 	if (xdptxd->has_frags) {
 		xdptxdf.sinfo = xdp_get_shared_info_from_frame(xdpf);
+		xdptxdf.dma_arr = NULL;
 
 		for (i = 0; i < xdptxdf.sinfo->nr_frags; i++) {
 			skb_frag_t *frag = &xdptxdf.sinfo->frags[i];
@@ -548,7 +549,8 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 			skb_frag_t *frag = &xdptxdf->sinfo->frags[i];
 			dma_addr_t addr;
 
-			addr = page_pool_get_dma_addr(skb_frag_page(frag)) +
+			addr = xdptxdf->dma_arr ? xdptxdf->dma_arr[i] :
+				page_pool_get_dma_addr(skb_frag_page(frag)) +
 				skb_frag_off(frag);
 
 			dseg++;
@@ -601,6 +603,21 @@ static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq,
 
 			dma_unmap_single(sq->pdev, dma_addr,
 					 xdpf->len, DMA_TO_DEVICE);
+			if (xdp_frame_has_frags(xdpf)) {
+				struct skb_shared_info *sinfo;
+				int j;
+
+				sinfo = xdp_get_shared_info_from_frame(xdpf);
+				for (j = 0; j < sinfo->nr_frags; j++) {
+					skb_frag_t *frag = &sinfo->frags[j];
+
+					xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo);
+					dma_addr = xdpi.frame.dma_addr;
+
+					dma_unmap_single(sq->pdev, dma_addr,
+							 skb_frag_size(frag), DMA_TO_DEVICE);
+				}
+			}
 			xdp_return_frame_bulk(xdpf, bq);
 			break;
 		}
@@ -759,23 +776,57 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 	sq = &priv->channels.c[sq_num]->xdpsq;
 
 	for (i = 0; i < n; i++) {
+		struct mlx5e_xmit_data_frags xdptxdf = {};
 		struct xdp_frame *xdpf = frames[i];
-		struct mlx5e_xmit_data xdptxd = {};
+		dma_addr_t dma_arr[MAX_SKB_FRAGS];
+		struct mlx5e_xmit_data *xdptxd;
 		bool ret;
 
-		xdptxd.data = xdpf->data;
-		xdptxd.len = xdpf->len;
-		xdptxd.dma_addr = dma_map_single(sq->pdev, xdptxd.data,
-						 xdptxd.len, DMA_TO_DEVICE);
+		xdptxd = &xdptxdf.xd;
+		xdptxd->data = xdpf->data;
+		xdptxd->len = xdpf->len;
+		xdptxd->has_frags = xdp_frame_has_frags(xdpf);
+		xdptxd->dma_addr = dma_map_single(sq->pdev, xdptxd->data,
+						  xdptxd->len, DMA_TO_DEVICE);
 
-		if (unlikely(dma_mapping_error(sq->pdev, xdptxd.dma_addr)))
+		if (unlikely(dma_mapping_error(sq->pdev, xdptxd->dma_addr)))
 			break;
 
+		if (xdptxd->has_frags) {
+			int j;
+
+			xdptxdf.sinfo = xdp_get_shared_info_from_frame(xdpf);
+			xdptxdf.dma_arr = dma_arr;
+			for (j = 0; j < xdptxdf.sinfo->nr_frags; j++) {
+				skb_frag_t *frag = &xdptxdf.sinfo->frags[j];
+
+				dma_arr[j] = dma_map_single(sq->pdev, skb_frag_address(frag),
+							    skb_frag_size(frag), DMA_TO_DEVICE);
+
+				if (!dma_mapping_error(sq->pdev, dma_arr[j]))
+					continue;
+				/* mapping error */
+				while (--j >= 0)
+					dma_unmap_single(sq->pdev, dma_arr[j],
+							 skb_frag_size(&xdptxdf.sinfo->frags[j]),
+							 DMA_TO_DEVICE);
+				goto out;
+			}
+		}
+
 		ret = INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe,
-				      mlx5e_xmit_xdp_frame, sq, &xdptxd, 0);
+				      mlx5e_xmit_xdp_frame, sq, xdptxd, 0);
 		if (unlikely(!ret)) {
-			dma_unmap_single(sq->pdev, xdptxd.dma_addr,
-					 xdptxd.len, DMA_TO_DEVICE);
+			int j;
+
+			dma_unmap_single(sq->pdev, xdptxd->dma_addr,
+					 xdptxd->len, DMA_TO_DEVICE);
+			if (!xdptxd->has_frags)
+				break;
+			for (j = 0; j < xdptxdf.sinfo->nr_frags; j++)
+				dma_unmap_single(sq->pdev, dma_arr[j],
+						 skb_frag_size(&xdptxdf.sinfo->frags[j]),
+						 DMA_TO_DEVICE);
 			break;
 		}
 
@@ -785,10 +836,19 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
 				     (union mlx5e_xdp_info) { .frame.xdpf = xdpf });
 		mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
-				     (union mlx5e_xdp_info) { .frame.dma_addr = xdptxd.dma_addr });
+				     (union mlx5e_xdp_info) { .frame.dma_addr = xdptxd->dma_addr });
+		if (xdptxd->has_frags) {
+			int j;
+
+			for (j = 0; j < xdptxdf.sinfo->nr_frags; j++)
+				mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo,
+						     (union mlx5e_xdp_info)
+						     { .frame.dma_addr = dma_arr[j] });
+		}
 		nxmit++;
 	}
 
+out:
 	if (flags & XDP_XMIT_FLUSH) {
 		if (sq->mpwqe.wqe)
 			mlx5e_xdp_mpwqe_complete(sq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 0b5aafaefe4c..ccf7bb136f50 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1862,11 +1862,7 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params,
 	csp.min_inline_mode = sq->min_inline_mode;
 	set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state);
 
-	/* Don't enable multi buffer on XDP_REDIRECT SQ, as it's not yet
-	 * supported by upstream, and there is no defined trigger to allow
-	 * transmitting redirected multi-buffer frames.
-	 */
-	if (param->is_xdp_mb && !is_redirect)
+	if (param->is_xdp_mb)
 		set_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state);
 
 	err = mlx5e_create_sq_rdy(c->mdev, param, &csp, 0, &sq->sqn);
@@ -4068,7 +4064,8 @@ void mlx5e_set_xdp_feature(struct net_device *netdev)
 
 	val = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
 	      NETDEV_XDP_ACT_XSK_ZEROCOPY |
-	      NETDEV_XDP_ACT_NDO_XMIT;
+	      NETDEV_XDP_ACT_NDO_XMIT |
+	      NETDEV_XDP_ACT_NDO_XMIT_SG;
 	if (params->rq_wq_type == MLX5_WQ_TYPE_CYCLIC)
 		val |= NETDEV_XDP_ACT_RX_SG;
 	xdp_set_features_flag(netdev, val);
-- 
cgit v1.2.3-70-g09d2


From 7fc06dd2ae7b42fc5093f76878b77c74006b4fd5 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:55 +0300
Subject: net/mlx5e: XDP, Improve Striding RQ check with XDP

Non-linear mem scheme of Striding RQ does not yet support XDP at this
point. Take the check where it belongs, inside the params validation
function mlx5e_params_validate_xdp().

Reviewed-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 23 +++++++++--------------
 1 file changed, 9 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index ccf7bb136f50..faae443770bb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4261,8 +4261,16 @@ static bool mlx5e_params_validate_xdp(struct net_device *netdev,
 	/* No XSK params: AF_XDP can't be enabled yet at the point of setting
 	 * the XDP program.
 	 */
-	is_linear = mlx5e_rx_is_linear_skb(mdev, params, NULL);
+	is_linear = params->rq_wq_type == MLX5_WQ_TYPE_CYCLIC ?
+		mlx5e_rx_is_linear_skb(mdev, params, NULL) :
+		mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL);
 
+	/* XDP affects striding RQ parameters. Block XDP if striding RQ won't be
+	 * supported with the new parameters: if PAGE_SIZE is bigger than
+	 * MLX5_MPWQE_LOG_STRIDE_SZ_MAX, striding RQ can't be used, even though
+	 * the MTU is small enough for the linear mode, because XDP uses strides
+	 * of PAGE_SIZE on regular RQs.
+	 */
 	if (!is_linear && params->rq_wq_type != MLX5_WQ_TYPE_CYCLIC) {
 		netdev_warn(netdev, "XDP is not allowed with striding RQ and MTU(%d) > %d\n",
 			    params->sw_mtu,
@@ -4817,19 +4825,6 @@ static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog)
 	new_params = priv->channels.params;
 	new_params.xdp_prog = prog;
 
-	/* XDP affects striding RQ parameters. Block XDP if striding RQ won't be
-	 * supported with the new parameters: if PAGE_SIZE is bigger than
-	 * MLX5_MPWQE_LOG_STRIDE_SZ_MAX, striding RQ can't be used, even though
-	 * the MTU is small enough for the linear mode, because XDP uses strides
-	 * of PAGE_SIZE on regular RQs.
-	 */
-	if (reset && MLX5E_GET_PFLAG(&new_params, MLX5E_PFLAG_RX_STRIDING_RQ)) {
-		/* Checking for regular RQs here; XSK RQs were checked on XSK bind. */
-		err = mlx5e_mpwrq_validate_regular(priv->mdev, &new_params);
-		if (err)
-			goto unlock;
-	}
-
 	old_prog = priv->channels.params.xdp_prog;
 
 	err = mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, reset);
-- 
cgit v1.2.3-70-g09d2


From abd3f84ecab3ab90161c9ba49052ac4ca33a0ed3 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:56 +0300
Subject: net/mlx5e: XDP, Let XDP checker function get the params as input

Change mlx5e_xdp_allowed() so it gets the params structure with the
xdp_prog applied, rather than creating a local copy based on the current
params in priv.

This reduces the amount of memory on the stack, and acts on the exact
params instance that's about to be applied.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index faae443770bb..6a278901b40b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4773,20 +4773,15 @@ static void mlx5e_tx_timeout(struct net_device *dev, unsigned int txqueue)
 	queue_work(priv->wq, &priv->tx_timeout_work);
 }
 
-static int mlx5e_xdp_allowed(struct mlx5e_priv *priv, struct bpf_prog *prog)
+static int mlx5e_xdp_allowed(struct net_device *netdev, struct mlx5_core_dev *mdev,
+			     struct mlx5e_params *params)
 {
-	struct net_device *netdev = priv->netdev;
-	struct mlx5e_params new_params;
-
-	if (priv->channels.params.packet_merge.type != MLX5E_PACKET_MERGE_NONE) {
+	if (params->packet_merge.type != MLX5E_PACKET_MERGE_NONE) {
 		netdev_warn(netdev, "can't set XDP while HW-GRO/LRO is on, disable them first\n");
 		return -EINVAL;
 	}
 
-	new_params = priv->channels.params;
-	new_params.xdp_prog = prog;
-
-	if (!mlx5e_params_validate_xdp(netdev, priv->mdev, &new_params))
+	if (!mlx5e_params_validate_xdp(netdev, mdev, params))
 		return -EINVAL;
 
 	return 0;
@@ -4813,8 +4808,11 @@ static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog)
 
 	mutex_lock(&priv->state_lock);
 
+	new_params = priv->channels.params;
+	new_params.xdp_prog = prog;
+
 	if (prog) {
-		err = mlx5e_xdp_allowed(priv, prog);
+		err = mlx5e_xdp_allowed(netdev, priv->mdev, &new_params);
 		if (err)
 			goto unlock;
 	}
@@ -4822,9 +4820,6 @@ static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog)
 	/* no need for full reset when exchanging programs */
 	reset = (!priv->channels.params.xdp_prog || !prog);
 
-	new_params = priv->channels.params;
-	new_params.xdp_prog = prog;
-
 	old_prog = priv->channels.params.xdp_prog;
 
 	err = mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, reset);
-- 
cgit v1.2.3-70-g09d2


From 20409abe52b1de8cfdc0565d89b7bc6d159dba35 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:57 +0300
Subject: net/mlx5e: XDP, Consider large muti-buffer packets in Striding RQ
 params calculations

Function mlx5e_rx_get_linear_stride_sz() returns PAGE_SIZE immediately
in case an XDP program is attached. The more accurate formula is
ALIGN(sz, PAGE_SIZE), to prevent two packets from residing on the same
page.

The assumption behind the current code is that sz <= PAGE_SIZE holds for
all cases with XDP program set.

This is true because it is being called from:
- 3 times from Striding RQ flows, in which XDP is not supported for such
  large packets.
- 1 time from Legacy RQ flow, under the condition
  mlx5e_rx_is_linear_skb().

No functional change here, just removing the implied assumption in
preparation for supporting XDP multi-buffer in Striding RQ.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/params.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 31f3c6e51d9e..196862e67af3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -253,17 +253,20 @@ static u32 mlx5e_rx_get_linear_stride_sz(struct mlx5_core_dev *mdev,
 					 struct mlx5e_xsk_param *xsk,
 					 bool mpwqe)
 {
+	u32 sz;
+
 	/* XSK frames are mapped as individual pages, because frames may come in
 	 * an arbitrary order from random locations in the UMEM.
 	 */
 	if (xsk)
 		return mpwqe ? 1 << mlx5e_mpwrq_page_shift(mdev, xsk) : PAGE_SIZE;
 
-	/* XDP in mlx5e doesn't support multiple packets per page. */
-	if (params->xdp_prog)
-		return PAGE_SIZE;
+	sz = roundup_pow_of_two(mlx5e_rx_get_linear_sz_skb(params, false));
 
-	return roundup_pow_of_two(mlx5e_rx_get_linear_sz_skb(params, false));
+	/* XDP in mlx5e doesn't support multiple packets per page.
+	 * Do not assume sz <= PAGE_SIZE if params->xdp_prog is set.
+	 */
+	return params->xdp_prog && sz < PAGE_SIZE ? PAGE_SIZE : sz;
 }
 
 static u8 mlx5e_mpwqe_log_pkts_per_wqe(struct mlx5_core_dev *mdev,
-- 
cgit v1.2.3-70-g09d2


From 124d0d8daf85ddf106a25ab0647e41b956e5cd38 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:58 +0300
Subject: net/mlx5e: XDP, Remove un-established assumptions on XDP buffer

Remove the assumption of non-zero linear length in the XDP xmit
function, used to serve both internal XDP_TX operations as well as
redirected-in requests.

Do not apply the MLX5E_XDP_MIN_INLINE check unless necessary.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c  | 36 ++++++++++++++---------
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  4 ---
 2 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index c266d073e2f2..d89f934570ee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -477,18 +477,26 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 	u16 ds_cnt, inline_hdr_sz;
 	u8 num_wqebbs = 1;
 	int num_frags = 0;
+	bool inline_ok;
+	bool linear;
 	u16 pi;
 
 	struct mlx5e_xdpsq_stats *stats = sq->stats;
 
-	if (unlikely(dma_len < MLX5E_XDP_MIN_INLINE || sq->hw_mtu < dma_len)) {
+	inline_ok = sq->min_inline_mode == MLX5_INLINE_MODE_NONE ||
+		dma_len >= MLX5E_XDP_MIN_INLINE;
+
+	if (unlikely(!inline_ok || sq->hw_mtu < dma_len)) {
 		stats->err++;
 		return false;
 	}
 
-	ds_cnt = MLX5E_TX_WQE_EMPTY_DS_COUNT + 1;
+	inline_hdr_sz = 0;
 	if (sq->min_inline_mode != MLX5_INLINE_MODE_NONE)
-		ds_cnt++;
+		inline_hdr_sz = MLX5E_XDP_MIN_INLINE;
+
+	linear = !!(dma_len - inline_hdr_sz);
+	ds_cnt = MLX5E_TX_WQE_EMPTY_DS_COUNT + linear + !!inline_hdr_sz;
 
 	/* check_result must be 0 if sinfo is passed. */
 	if (!check_result) {
@@ -517,22 +525,23 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 	eseg = &wqe->eth;
 	dseg = wqe->data;
 
-	inline_hdr_sz = 0;
-
 	/* copy the inline part if required */
-	if (sq->min_inline_mode != MLX5_INLINE_MODE_NONE) {
+	if (inline_hdr_sz) {
 		memcpy(eseg->inline_hdr.start, xdptxd->data, sizeof(eseg->inline_hdr.start));
 		memcpy(dseg, xdptxd->data + sizeof(eseg->inline_hdr.start),
-		       MLX5E_XDP_MIN_INLINE - sizeof(eseg->inline_hdr.start));
-		dma_len  -= MLX5E_XDP_MIN_INLINE;
-		dma_addr += MLX5E_XDP_MIN_INLINE;
-		inline_hdr_sz = MLX5E_XDP_MIN_INLINE;
+		       inline_hdr_sz - sizeof(eseg->inline_hdr.start));
+		dma_len  -= inline_hdr_sz;
+		dma_addr += inline_hdr_sz;
 		dseg++;
 	}
 
 	/* write the dma part */
-	dseg->addr       = cpu_to_be64(dma_addr);
-	dseg->byte_count = cpu_to_be32(dma_len);
+	if (linear) {
+		dseg->addr       = cpu_to_be64(dma_addr);
+		dseg->byte_count = cpu_to_be32(dma_len);
+		dseg->lkey       = sq->mkey_be;
+		dseg++;
+	}
 
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_SEND);
 
@@ -543,7 +552,6 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 		memset(eseg, 0, sizeof(*eseg) - sizeof(eseg->trailer));
 
 		eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz);
-		dseg->lkey = sq->mkey_be;
 
 		for (i = 0; i < num_frags; i++) {
 			skb_frag_t *frag = &xdptxdf->sinfo->frags[i];
@@ -553,10 +561,10 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 				page_pool_get_dma_addr(skb_frag_page(frag)) +
 				skb_frag_off(frag);
 
-			dseg++;
 			dseg->addr = cpu_to_be64(addr);
 			dseg->byte_count = cpu_to_be32(skb_frag_size(frag));
 			dseg->lkey = sq->mkey_be;
+			dseg++;
 		}
 
 		cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 6a278901b40b..a95ce206391b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1886,7 +1886,6 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params,
 			struct mlx5e_tx_wqe      *wqe  = mlx5_wq_cyc_get_wqe(&sq->wq, i);
 			struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl;
 			struct mlx5_wqe_eth_seg  *eseg = &wqe->eth;
-			struct mlx5_wqe_data_seg *dseg;
 
 			sq->db.wqe_info[i] = (struct mlx5e_xdp_wqe_info) {
 				.num_wqebbs = 1,
@@ -1895,9 +1894,6 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params,
 
 			cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
 			eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz);
-
-			dseg = (struct mlx5_wqe_data_seg *)cseg + (ds_cnt - 1);
-			dseg->lkey = sq->mkey_be;
 		}
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 63abf14e13ed09ead1d979949963fe08ced3a77c Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:18:59 +0300
Subject: net/mlx5e: XDP, Allow non-linear single-segment frames in XDP TX
 MPWQE

Under a few restrictions, TX MPWQE feature can serve multiple TX packets
in a single TX descriptor. It requires each of the packets to have a
single scatter entry / segment.

Today we allow only linear frames to use this feature, although there's
no real problem with non-linear ones where the whole packet reside in
the first fragment.

Expand the XDP TX MPWQE feature support to include such frames. This is
in preparation for the downstream patch, in which we will generate such
non-linear frames.

Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c | 35 ++++++++++++++++++------
 1 file changed, 26 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index d89f934570ee..f0e6095809fa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -405,18 +405,35 @@ mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptx
 {
 	struct mlx5e_tx_mpwqe *session = &sq->mpwqe;
 	struct mlx5e_xdpsq_stats *stats = sq->stats;
+	struct mlx5e_xmit_data *p = xdptxd;
+	struct mlx5e_xmit_data tmp;
 
 	if (xdptxd->has_frags) {
-		/* MPWQE is enabled, but a multi-buffer packet is queued for
-		 * transmission. MPWQE can't send fragmented packets, so close
-		 * the current session and fall back to a regular WQE.
-		 */
-		if (unlikely(sq->mpwqe.wqe))
-			mlx5e_xdp_mpwqe_complete(sq);
-		return mlx5e_xmit_xdp_frame(sq, xdptxd, 0);
+		struct mlx5e_xmit_data_frags *xdptxdf =
+			container_of(xdptxd, struct mlx5e_xmit_data_frags, xd);
+
+		if (!!xdptxd->len + xdptxdf->sinfo->nr_frags > 1) {
+			/* MPWQE is enabled, but a multi-buffer packet is queued for
+			 * transmission. MPWQE can't send fragmented packets, so close
+			 * the current session and fall back to a regular WQE.
+			 */
+			if (unlikely(sq->mpwqe.wqe))
+				mlx5e_xdp_mpwqe_complete(sq);
+			return mlx5e_xmit_xdp_frame(sq, xdptxd, 0);
+		}
+		if (!xdptxd->len) {
+			skb_frag_t *frag = &xdptxdf->sinfo->frags[0];
+
+			tmp.data = skb_frag_address(frag);
+			tmp.len = skb_frag_size(frag);
+			tmp.dma_addr = xdptxdf->dma_arr ? xdptxdf->dma_arr[0] :
+				page_pool_get_dma_addr(skb_frag_page(frag)) +
+				skb_frag_off(frag);
+			p = &tmp;
+		}
 	}
 
-	if (unlikely(xdptxd->len > sq->hw_mtu)) {
+	if (unlikely(p->len > sq->hw_mtu)) {
 		stats->err++;
 		return false;
 	}
@@ -434,7 +451,7 @@ mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptx
 		mlx5e_xdp_mpwqe_session_start(sq);
 	}
 
-	mlx5e_xdp_mpwqe_add_dseg(sq, xdptxd, stats);
+	mlx5e_xdp_mpwqe_add_dseg(sq, p, stats);
 
 	if (unlikely(mlx5e_xdp_mpwqe_is_full(session, sq->max_sq_mpw_wqebbs)))
 		mlx5e_xdp_mpwqe_complete(sq);
-- 
cgit v1.2.3-70-g09d2


From 27602319e32862717c15c32283a727dc5bed695d Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:19:00 +0300
Subject: net/mlx5e: RX, Take shared info fragment addition into a function

Introduce mlx5e_add_skb_shared_info_frag(), a function dedicated for
adding a fragment into a struct skb_shared_info object.

Use it in the Legacy RQ flow. Similar usage will be added in a
downstream patch by the corresponding Striding RQ flow.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 56 ++++++++++++++-----------
 1 file changed, 31 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 1049805571c6..1118327f6467 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -471,6 +471,35 @@ static int mlx5e_refill_rx_wqes(struct mlx5e_rq *rq, u16 ix, int wqe_bulk)
 	return i;
 }
 
+static void
+mlx5e_add_skb_shared_info_frag(struct mlx5e_rq *rq, struct skb_shared_info *sinfo,
+			       struct xdp_buff *xdp, struct mlx5e_frag_page *frag_page,
+			       u32 frag_offset, u32 len)
+{
+	skb_frag_t *frag;
+
+	dma_addr_t addr = page_pool_get_dma_addr(frag_page->page);
+
+	dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len, rq->buff.map_dir);
+	if (!xdp_buff_has_frags(xdp)) {
+		/* Init on the first fragment to avoid cold cache access
+		 * when possible.
+		 */
+		sinfo->nr_frags = 0;
+		sinfo->xdp_frags_size = 0;
+		xdp_buff_set_frags_flag(xdp);
+	}
+
+	frag = &sinfo->frags[sinfo->nr_frags++];
+	__skb_frag_set_page(frag, frag_page->page);
+	skb_frag_off_set(frag, frag_offset);
+	skb_frag_size_set(frag, len);
+
+	if (page_is_pfmemalloc(frag_page->page))
+		xdp_buff_set_frag_pfmemalloc(xdp);
+	sinfo->xdp_frags_size += len;
+}
+
 static inline void
 mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb,
 		   struct page *page, u32 frag_offset, u32 len,
@@ -1694,35 +1723,12 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	wi++;
 
 	while (cqe_bcnt) {
-		skb_frag_t *frag;
-
 		frag_page = wi->frag_page;
 
 		frag_consumed_bytes = min_t(u32, frag_info->frag_size, cqe_bcnt);
 
-		addr = page_pool_get_dma_addr(frag_page->page);
-		dma_sync_single_for_cpu(rq->pdev, addr + wi->offset,
-					frag_consumed_bytes, rq->buff.map_dir);
-
-		if (!xdp_buff_has_frags(&mxbuf.xdp)) {
-			/* Init on the first fragment to avoid cold cache access
-			 * when possible.
-			 */
-			sinfo->nr_frags = 0;
-			sinfo->xdp_frags_size = 0;
-			xdp_buff_set_frags_flag(&mxbuf.xdp);
-		}
-
-		frag = &sinfo->frags[sinfo->nr_frags++];
-
-		__skb_frag_set_page(frag, frag_page->page);
-		skb_frag_off_set(frag, wi->offset);
-		skb_frag_size_set(frag, frag_consumed_bytes);
-
-		if (page_is_pfmemalloc(frag_page->page))
-			xdp_buff_set_frag_pfmemalloc(&mxbuf.xdp);
-
-		sinfo->xdp_frags_size += frag_consumed_bytes;
+		mlx5e_add_skb_shared_info_frag(rq, sinfo, &mxbuf.xdp, frag_page,
+					       wi->offset, frag_consumed_bytes);
 		truesize += frag_info->frag_stride;
 
 		cqe_bcnt -= frag_consumed_bytes;
-- 
cgit v1.2.3-70-g09d2


From 221c8c7ad7a2d414d6d64b5e5bbf7723ae23d3f5 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:19:01 +0300
Subject: net/mlx5e: RX, Generalize mlx5e_fill_mxbuf()

Make the function more generic. Let it get an additional frame_sz
parameter instead of deriving it from the RQ struct.

No functional change here, just a preparation for a downstream patch.

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 1118327f6467..a2c4b3df5757 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1630,10 +1630,10 @@ struct sk_buff *mlx5e_build_linear_skb(struct mlx5e_rq *rq, void *va,
 }
 
 static void mlx5e_fill_mxbuf(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe,
-			     void *va, u16 headroom, u32 len,
+			     void *va, u16 headroom, u32 frame_sz, u32 len,
 			     struct mlx5e_xdp_buff *mxbuf)
 {
-	xdp_init_buff(&mxbuf->xdp, rq->buff.frame0_sz, &rq->xdp_rxq);
+	xdp_init_buff(&mxbuf->xdp, frame_sz, &rq->xdp_rxq);
 	xdp_prepare_buff(&mxbuf->xdp, va, headroom, len, true);
 	mxbuf->cqe = cqe;
 	mxbuf->rq = rq;
@@ -1666,7 +1666,8 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi,
 		struct mlx5e_xdp_buff mxbuf;
 
 		net_prefetchw(va); /* xdp_frame data area */
-		mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, cqe_bcnt, &mxbuf);
+		mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, rq->buff.frame0_sz,
+				 cqe_bcnt, &mxbuf);
 		if (mlx5e_xdp_handle(rq, prog, &mxbuf))
 			return NULL; /* page/packet was consumed by XDP */
 
@@ -1714,7 +1715,8 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	net_prefetchw(va); /* xdp_frame data area */
 	net_prefetch(va + rx_headroom);
 
-	mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, frag_consumed_bytes, &mxbuf);
+	mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, rq->buff.frame0_sz,
+			 frag_consumed_bytes, &mxbuf);
 	sinfo = xdp_get_shared_info_from_buff(&mxbuf.xdp);
 	truesize = 0;
 
@@ -2042,7 +2044,8 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 		struct mlx5e_xdp_buff mxbuf;
 
 		net_prefetchw(va); /* xdp_frame data area */
-		mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, cqe_bcnt, &mxbuf);
+		mlx5e_fill_mxbuf(rq, cqe, va, rx_headroom, rq->buff.frame0_sz,
+				 cqe_bcnt, &mxbuf);
 		if (mlx5e_xdp_handle(rq, prog, &mxbuf)) {
 			if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags))
 				__set_bit(page_idx, wi->skip_release_bitmap); /* non-atomic */
-- 
cgit v1.2.3-70-g09d2


From 2cb0e27d43b4eec70e20103cb09055303f70b6d9 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:19:02 +0300
Subject: net/mlx5e: RX, Prepare non-linear striding RQ for XDP multi-buffer
 support

In preparation for supporting XDP multi-buffer in striding RQ, use
xdp_buff struct to describe the packet. Make its skb_shared_info collide
the one of the allocated SKB, then add the fragments using the xdp_buff
API.

Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 51 +++++++++++++++++++++++--
 1 file changed, 47 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index a2c4b3df5757..2e99bef49dd6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1977,10 +1977,17 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 	struct mlx5e_frag_page *frag_page = &wi->alloc_units.frag_pages[page_idx];
 	u16 headlen = min_t(u16, MLX5E_RX_MAX_HEAD, cqe_bcnt);
 	struct mlx5e_frag_page *head_page = frag_page;
-	u32 frag_offset    = head_offset + headlen;
-	u32 byte_cnt       = cqe_bcnt - headlen;
+	u32 frag_offset    = head_offset;
+	u32 byte_cnt       = cqe_bcnt;
+	struct skb_shared_info *sinfo;
+	struct mlx5e_xdp_buff mxbuf;
+	unsigned int truesize = 0;
 	struct sk_buff *skb;
+	u32 linear_frame_sz;
+	u16 linear_data_len;
 	dma_addr_t addr;
+	u16 linear_hr;
+	void *va;
 
 	skb = napi_alloc_skb(rq->cq.napi,
 			     ALIGN(MLX5E_RX_MAX_HEAD, sizeof(long)));
@@ -1989,16 +1996,52 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 		return NULL;
 	}
 
+	va = skb->head;
 	net_prefetchw(skb->data);
 
-	/* Non-linear mode, hence non-XSK, which always uses PAGE_SIZE. */
+	frag_offset += headlen;
+	byte_cnt -= headlen;
+	linear_hr = skb_headroom(skb);
+	linear_data_len = headlen;
+	linear_frame_sz = MLX5_SKB_FRAG_SZ(skb_end_offset(skb));
 	if (unlikely(frag_offset >= PAGE_SIZE)) {
 		frag_page++;
 		frag_offset -= PAGE_SIZE;
 	}
 
 	skb_mark_for_recycle(skb);
-	mlx5e_fill_skb_data(skb, rq, frag_page, byte_cnt, frag_offset);
+	mlx5e_fill_mxbuf(rq, cqe, va, linear_hr, linear_frame_sz, linear_data_len, &mxbuf);
+	net_prefetch(mxbuf.xdp.data);
+
+	sinfo = xdp_get_shared_info_from_buff(&mxbuf.xdp);
+
+	while (byte_cnt) {
+		/* Non-linear mode, hence non-XSK, which always uses PAGE_SIZE. */
+		u32 pg_consumed_bytes = min_t(u32, PAGE_SIZE - frag_offset, byte_cnt);
+
+		if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state))
+			truesize += pg_consumed_bytes;
+		else
+			truesize += ALIGN(pg_consumed_bytes, BIT(rq->mpwqe.log_stride_sz));
+
+		mlx5e_add_skb_shared_info_frag(rq, sinfo, &mxbuf.xdp, frag_page, frag_offset,
+					       pg_consumed_bytes);
+		byte_cnt -= pg_consumed_bytes;
+		frag_offset = 0;
+		frag_page++;
+	}
+	if (xdp_buff_has_frags(&mxbuf.xdp)) {
+		struct mlx5e_frag_page *pagep;
+
+		xdp_update_skb_shared_info(skb, sinfo->nr_frags,
+					   sinfo->xdp_frags_size, truesize,
+					   xdp_buff_is_frag_pfmemalloc(&mxbuf.xdp));
+
+		pagep = frag_page - sinfo->nr_frags;
+		do
+			pagep->frags++;
+		while (++pagep < frag_page);
+	}
 	/* copy header */
 	addr = page_pool_get_dma_addr(head_page->page);
 	mlx5e_copy_skb_header(rq, skb, head_page->page, addr,
-- 
cgit v1.2.3-70-g09d2


From f52ac7028bec22e925c8fece4f21641eb13b4d6f Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Mon, 17 Apr 2023 15:19:03 +0300
Subject: net/mlx5e: RX, Add XDP multi-buffer support in Striding RQ

Here we add support for multi-buffer XDP handling in Striding RQ, which
is our default out-of-the-box RQ type. Before this series, loading such
an XDP program would fail, until you switch to the legacy RQ (by
unsetting the rx_striding_rq priv-flag).

To overcome the lack of headroom and tailroom between the strides, we
allocate a side page to be used for the descriptor (xdp_buff / skb) and
the linear part. When an XDP program is attached, we structure the
xdp_buff so that it contains no data in the linear part, and the whole
packet resides in the fragments.

In case of XDP_PASS, where an SKB still needs to be created, we copy up
to 256 bytes to its linear part, to match the current behavior, and
satisfy functions that assume finding the packet headers in the SKB
linear part (like eth_type_trans).

Performance testing:

Packet rate test, 64 bytes, 32 channels, MTU 9000 bytes.
CPU: Intel(R) Xeon(R) Platinum 8380 CPU @ 2.30GHz.
NIC: ConnectX-6 Dx, at 100 Gbps.

+----------+-------------+-------------+---------+
| Test     | Legacy RQ   | Striding RQ | Speedup |
+----------+-------------+-------------+---------+
| XDP_DROP | 101,615,544 | 117,191,020 | +15%    |
+----------+-------------+-------------+---------+
| XDP_TX   |  95,608,169 | 117,043,422 | +22%    |
+----------+-------------+-------------+---------+

Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |   1 +
 .../net/ethernet/mellanox/mlx5/core/en/params.c    |  21 +++-
 .../net/ethernet/mellanox/mlx5/core/en/params.h    |   3 +
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  37 +++---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 135 +++++++++++++++------
 5 files changed, 138 insertions(+), 59 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 0e15afbe1673..b8987a404d75 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -587,6 +587,7 @@ union mlx5e_alloc_units {
 struct mlx5e_mpw_info {
 	u16 consumed_strides;
 	DECLARE_BITMAP(skip_release_bitmap, MLX5_MPWRQ_MAX_PAGES_PER_WQE);
+	struct mlx5e_frag_page linear_page;
 	union mlx5e_alloc_units alloc_units;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 196862e67af3..ef546ed8b4d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -323,6 +323,20 @@ static bool mlx5e_verify_rx_mpwqe_strides(struct mlx5_core_dev *mdev,
 	return log_num_strides >= MLX5_MPWQE_LOG_NUM_STRIDES_BASE;
 }
 
+bool mlx5e_verify_params_rx_mpwqe_strides(struct mlx5_core_dev *mdev,
+					  struct mlx5e_params *params,
+					  struct mlx5e_xsk_param *xsk)
+{
+	u8 log_wqe_num_of_strides = mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk);
+	u8 log_wqe_stride_size = mlx5e_mpwqe_get_log_stride_size(mdev, params, xsk);
+	enum mlx5e_mpwrq_umr_mode umr_mode = mlx5e_mpwrq_umr_mode(mdev, xsk);
+	u8 page_shift = mlx5e_mpwrq_page_shift(mdev, xsk);
+
+	return mlx5e_verify_rx_mpwqe_strides(mdev, log_wqe_stride_size,
+					     log_wqe_num_of_strides,
+					     page_shift, umr_mode);
+}
+
 bool mlx5e_rx_mpwqe_is_linear_skb(struct mlx5_core_dev *mdev,
 				  struct mlx5e_params *params,
 				  struct mlx5e_xsk_param *xsk)
@@ -405,6 +419,10 @@ u8 mlx5e_mpwqe_get_log_stride_size(struct mlx5_core_dev *mdev,
 	if (mlx5e_rx_mpwqe_is_linear_skb(mdev, params, xsk))
 		return order_base_2(mlx5e_rx_get_linear_stride_sz(mdev, params, xsk, true));
 
+	/* XDP in mlx5e doesn't support multiple packets per page. */
+	if (params->xdp_prog)
+		return PAGE_SHIFT;
+
 	return MLX5_MPWRQ_DEF_LOG_STRIDE_SZ(mdev);
 }
 
@@ -575,9 +593,6 @@ int mlx5e_mpwrq_validate_regular(struct mlx5_core_dev *mdev, struct mlx5e_params
 	if (!mlx5e_check_fragmented_striding_rq_cap(mdev, page_shift, umr_mode))
 		return -EOPNOTSUPP;
 
-	if (params->xdp_prog && !mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL))
-		return -EINVAL;
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
index c9be6eb88012..a5d20f6d6d9c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
@@ -153,6 +153,9 @@ int mlx5e_build_channel_param(struct mlx5_core_dev *mdev,
 
 u16 mlx5e_calc_sq_stop_room(struct mlx5_core_dev *mdev, struct mlx5e_params *params);
 int mlx5e_validate_params(struct mlx5_core_dev *mdev, struct mlx5e_params *params);
+bool mlx5e_verify_params_rx_mpwqe_strides(struct mlx5_core_dev *mdev,
+					  struct mlx5e_params *params,
+					  struct mlx5e_xsk_param *xsk);
 
 static inline void mlx5e_params_print_info(struct mlx5_core_dev *mdev,
 					   struct mlx5e_params *params)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index a95ce206391b..7eb1eeb115ca 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -803,6 +803,9 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		pool_size = rq->mpwqe.pages_per_wqe <<
 			mlx5e_mpwqe_get_log_rq_size(mdev, params, xsk);
 
+		if (!mlx5e_rx_mpwqe_is_linear_skb(mdev, params, xsk) && params->xdp_prog)
+			pool_size *= 2; /* additional page per packet for the linear part */
+
 		rq->mpwqe.log_stride_sz = mlx5e_mpwqe_get_log_stride_size(mdev, params, xsk);
 		rq->mpwqe.num_strides =
 			BIT(mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk));
@@ -4060,10 +4063,9 @@ void mlx5e_set_xdp_feature(struct net_device *netdev)
 
 	val = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
 	      NETDEV_XDP_ACT_XSK_ZEROCOPY |
+	      NETDEV_XDP_ACT_RX_SG |
 	      NETDEV_XDP_ACT_NDO_XMIT |
 	      NETDEV_XDP_ACT_NDO_XMIT_SG;
-	if (params->rq_wq_type == MLX5_WQ_TYPE_CYCLIC)
-		val |= NETDEV_XDP_ACT_RX_SG;
 	xdp_set_features_flag(netdev, val);
 }
 
@@ -4261,23 +4263,20 @@ static bool mlx5e_params_validate_xdp(struct net_device *netdev,
 		mlx5e_rx_is_linear_skb(mdev, params, NULL) :
 		mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL);
 
-	/* XDP affects striding RQ parameters. Block XDP if striding RQ won't be
-	 * supported with the new parameters: if PAGE_SIZE is bigger than
-	 * MLX5_MPWQE_LOG_STRIDE_SZ_MAX, striding RQ can't be used, even though
-	 * the MTU is small enough for the linear mode, because XDP uses strides
-	 * of PAGE_SIZE on regular RQs.
-	 */
-	if (!is_linear && params->rq_wq_type != MLX5_WQ_TYPE_CYCLIC) {
-		netdev_warn(netdev, "XDP is not allowed with striding RQ and MTU(%d) > %d\n",
-			    params->sw_mtu,
-			    mlx5e_xdp_max_mtu(params, NULL));
-		return false;
-	}
-	if (!is_linear && !params->xdp_prog->aux->xdp_has_frags) {
-		netdev_warn(netdev, "MTU(%d) > %d, too big for an XDP program not aware of multi buffer\n",
-			    params->sw_mtu,
-			    mlx5e_xdp_max_mtu(params, NULL));
-		return false;
+	if (!is_linear) {
+		if (!params->xdp_prog->aux->xdp_has_frags) {
+			netdev_warn(netdev, "MTU(%d) > %d, too big for an XDP program not aware of multi buffer\n",
+				    params->sw_mtu,
+				    mlx5e_xdp_max_mtu(params, NULL));
+			return false;
+		}
+		if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ &&
+		    !mlx5e_verify_params_rx_mpwqe_strides(mdev, params, NULL)) {
+			netdev_warn(netdev, "XDP is not allowed with striding RQ and MTU(%d) > %d\n",
+				    params->sw_mtu,
+				    mlx5e_xdp_max_mtu(params, NULL));
+			return false;
+		}
 	}
 
 	return true;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 2e99bef49dd6..a8c2ae389d6c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1982,36 +1982,51 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 	struct skb_shared_info *sinfo;
 	struct mlx5e_xdp_buff mxbuf;
 	unsigned int truesize = 0;
+	struct bpf_prog *prog;
 	struct sk_buff *skb;
 	u32 linear_frame_sz;
 	u16 linear_data_len;
-	dma_addr_t addr;
 	u16 linear_hr;
 	void *va;
 
-	skb = napi_alloc_skb(rq->cq.napi,
-			     ALIGN(MLX5E_RX_MAX_HEAD, sizeof(long)));
-	if (unlikely(!skb)) {
-		rq->stats->buff_alloc_err++;
-		return NULL;
-	}
-
-	va = skb->head;
-	net_prefetchw(skb->data);
+	prog = rcu_dereference(rq->xdp_prog);
 
-	frag_offset += headlen;
-	byte_cnt -= headlen;
-	linear_hr = skb_headroom(skb);
-	linear_data_len = headlen;
-	linear_frame_sz = MLX5_SKB_FRAG_SZ(skb_end_offset(skb));
-	if (unlikely(frag_offset >= PAGE_SIZE)) {
-		frag_page++;
-		frag_offset -= PAGE_SIZE;
+	if (prog) {
+		/* area for bpf_xdp_[store|load]_bytes */
+		net_prefetchw(page_address(frag_page->page) + frag_offset);
+		if (unlikely(mlx5e_page_alloc_fragmented(rq, &wi->linear_page))) {
+			rq->stats->buff_alloc_err++;
+			return NULL;
+		}
+		va = page_address(wi->linear_page.page);
+		net_prefetchw(va); /* xdp_frame data area */
+		linear_hr = XDP_PACKET_HEADROOM;
+		linear_data_len = 0;
+		linear_frame_sz = MLX5_SKB_FRAG_SZ(linear_hr + MLX5E_RX_MAX_HEAD);
+	} else {
+		skb = napi_alloc_skb(rq->cq.napi,
+				     ALIGN(MLX5E_RX_MAX_HEAD, sizeof(long)));
+		if (unlikely(!skb)) {
+			rq->stats->buff_alloc_err++;
+			return NULL;
+		}
+		skb_mark_for_recycle(skb);
+		va = skb->head;
+		net_prefetchw(va); /* xdp_frame data area */
+		net_prefetchw(skb->data);
+
+		frag_offset += headlen;
+		byte_cnt -= headlen;
+		linear_hr = skb_headroom(skb);
+		linear_data_len = headlen;
+		linear_frame_sz = MLX5_SKB_FRAG_SZ(skb_end_offset(skb));
+		if (unlikely(frag_offset >= PAGE_SIZE)) {
+			frag_page++;
+			frag_offset -= PAGE_SIZE;
+		}
 	}
 
-	skb_mark_for_recycle(skb);
 	mlx5e_fill_mxbuf(rq, cqe, va, linear_hr, linear_frame_sz, linear_data_len, &mxbuf);
-	net_prefetch(mxbuf.xdp.data);
 
 	sinfo = xdp_get_shared_info_from_buff(&mxbuf.xdp);
 
@@ -2030,25 +2045,71 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 		frag_offset = 0;
 		frag_page++;
 	}
-	if (xdp_buff_has_frags(&mxbuf.xdp)) {
-		struct mlx5e_frag_page *pagep;
 
-		xdp_update_skb_shared_info(skb, sinfo->nr_frags,
-					   sinfo->xdp_frags_size, truesize,
-					   xdp_buff_is_frag_pfmemalloc(&mxbuf.xdp));
+	if (prog) {
+		if (mlx5e_xdp_handle(rq, prog, &mxbuf)) {
+			if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) {
+				int i;
+
+				for (i = 0; i < sinfo->nr_frags; i++)
+					/* non-atomic */
+					__set_bit(page_idx + i, wi->skip_release_bitmap);
+				return NULL;
+			}
+			mlx5e_page_release_fragmented(rq, &wi->linear_page);
+			return NULL; /* page/packet was consumed by XDP */
+		}
+
+		skb = mlx5e_build_linear_skb(rq, mxbuf.xdp.data_hard_start,
+					     linear_frame_sz,
+					     mxbuf.xdp.data - mxbuf.xdp.data_hard_start, 0,
+					     mxbuf.xdp.data - mxbuf.xdp.data_meta);
+		if (unlikely(!skb)) {
+			mlx5e_page_release_fragmented(rq, &wi->linear_page);
+			return NULL;
+		}
 
-		pagep = frag_page - sinfo->nr_frags;
-		do
-			pagep->frags++;
-		while (++pagep < frag_page);
-	}
-	/* copy header */
-	addr = page_pool_get_dma_addr(head_page->page);
-	mlx5e_copy_skb_header(rq, skb, head_page->page, addr,
-			      head_offset, head_offset, headlen);
-	/* skb linear part was allocated with headlen and aligned to long */
-	skb->tail += headlen;
-	skb->len  += headlen;
+		skb_mark_for_recycle(skb);
+		wi->linear_page.frags++;
+		mlx5e_page_release_fragmented(rq, &wi->linear_page);
+
+		if (xdp_buff_has_frags(&mxbuf.xdp)) {
+			struct mlx5e_frag_page *pagep;
+
+			/* sinfo->nr_frags is reset by build_skb, calculate again. */
+			xdp_update_skb_shared_info(skb, frag_page - head_page,
+						   sinfo->xdp_frags_size, truesize,
+						   xdp_buff_is_frag_pfmemalloc(&mxbuf.xdp));
+
+			pagep = head_page;
+			do
+				pagep->frags++;
+			while (++pagep < frag_page);
+		}
+		__pskb_pull_tail(skb, headlen);
+	} else {
+		dma_addr_t addr;
+
+		if (xdp_buff_has_frags(&mxbuf.xdp)) {
+			struct mlx5e_frag_page *pagep;
+
+			xdp_update_skb_shared_info(skb, sinfo->nr_frags,
+						   sinfo->xdp_frags_size, truesize,
+						   xdp_buff_is_frag_pfmemalloc(&mxbuf.xdp));
+
+			pagep = frag_page - sinfo->nr_frags;
+			do
+				pagep->frags++;
+			while (++pagep < frag_page);
+		}
+		/* copy header */
+		addr = page_pool_get_dma_addr(head_page->page);
+		mlx5e_copy_skb_header(rq, skb, head_page->page, addr,
+				      head_offset, head_offset, headlen);
+		/* skb linear part was allocated with headlen and aligned to long */
+		skb->tail += headlen;
+		skb->len  += headlen;
+	}
 
 	return skb;
 }
-- 
cgit v1.2.3-70-g09d2


From 2cdaa3eefed83082923cf219c8b6a314e622da74 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 18 Apr 2023 23:31:26 +0200
Subject: netfilter: conntrack: restore IPS_CONFIRMED out of
 nf_conntrack_hash_check_insert()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

e6d57e9ff0ae ("netfilter: conntrack: fix rmmod double-free race")
consolidates IPS_CONFIRMED bit set in nf_conntrack_hash_check_insert().
However, this breaks ctnetlink:

 # conntrack -I -p tcp --timeout 123 --src 1.2.3.4 --dst 5.6.7.8 --state ESTABLISHED --sport 1 --dport 4 -u SEEN_REPLY
 conntrack v1.4.6 (conntrack-tools): Operation failed: Device or resource busy

This is a partial revert of the aforementioned commit to restore
IPS_CONFIRMED.

Fixes: e6d57e9ff0ae ("netfilter: conntrack: fix rmmod double-free race")
Reported-by: Stéphane Graber <stgraber@stgraber.org>
Tested-by: Stéphane Graber <stgraber@stgraber.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_conntrack_bpf.c     | 1 +
 net/netfilter/nf_conntrack_core.c    | 1 -
 net/netfilter/nf_conntrack_netlink.c | 3 +++
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c
index cd99e6dc1f35..34913521c385 100644
--- a/net/netfilter/nf_conntrack_bpf.c
+++ b/net/netfilter/nf_conntrack_bpf.c
@@ -381,6 +381,7 @@ __bpf_kfunc struct nf_conn *bpf_ct_insert_entry(struct nf_conn___init *nfct_i)
 	struct nf_conn *nfct = (struct nf_conn *)nfct_i;
 	int err;
 
+	nfct->status |= IPS_CONFIRMED;
 	err = nf_conntrack_hash_check_insert(nfct);
 	if (err < 0) {
 		nf_conntrack_free(nfct);
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index c6a6a6099b4e..7ba6ab9b54b5 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -932,7 +932,6 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct)
 		goto out;
 	}
 
-	ct->status |= IPS_CONFIRMED;
 	smp_wmb();
 	/* The caller holds a reference to this object */
 	refcount_set(&ct->ct_general.use, 2);
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index bfc3aaa2c872..d3ee18854698 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -2316,6 +2316,9 @@ ctnetlink_create_conntrack(struct net *net,
 	nfct_seqadj_ext_add(ct);
 	nfct_synproxy_ext_add(ct);
 
+	/* we must add conntrack extensions before confirmation. */
+	ct->status |= IPS_CONFIRMED;
+
 	if (cda[CTA_STATUS]) {
 		err = ctnetlink_change_status(ct, cda);
 		if (err < 0)
-- 
cgit v1.2.3-70-g09d2


From 15ee62e73705df447971613de4fa660dd71ed40e Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Fri, 14 Apr 2023 04:23:31 +0800
Subject: wifi: mt76: mt7996: enable BSS_CHANGED_BASIC_RATES support

The connac3 removes fixed rate fields to reduce txd size and introduces
global rate tables (64 entries) for rate setting. Driver needs to fill
the corresponding idx in MT_TXD6_TX_RATE while tx, and push mt76_rate
into predifined table at bootup stage so that  mvif->basic_rates_idx
can immediately switch out once setting changes.

spe_idx is also needed for fixed rate frames, and will be updated by
future patches.

Note that all table entries are shared across driver and firmware
(i.e.TxBF), hence adding MT7996_BASIC_RATES_TBL to reflect mapping
status.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/init.c   | 16 +++++++++++
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    | 32 ++++++++++------------
 drivers/net/wireless/mediatek/mt76/mt7996/mac.h    |  7 ++---
 drivers/net/wireless/mediatek/mt76/mt7996/main.c   | 29 ++++++++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    | 20 ++++++--------
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  7 +++++
 drivers/net/wireless/mediatek/mt76/mt7996/regs.h   |  7 +++++
 7 files changed, 85 insertions(+), 33 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index 19cdd0625c40..90a92c1611f3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -255,6 +255,20 @@ mt7996_mac_init_band(struct mt7996_dev *dev, u8 band)
 	mt76_rmw(dev, MT_WTBLOFF_RSCR(band), mask, set);
 }
 
+static void mt7996_mac_init_basic_rates(struct mt7996_dev *dev)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(mt76_rates); i++) {
+		u16 rate = mt76_rates[i].hw_value;
+		u16 idx = MT7996_BASIC_RATES_TBL + i;
+
+		rate = FIELD_PREP(MT_TX_RATE_MODE, rate >> 8) |
+		       FIELD_PREP(MT_TX_RATE_IDX, rate & GENMASK(7, 0));
+		mt7996_mac_set_fixed_rate_table(dev, idx, rate);
+	}
+}
+
 void mt7996_mac_init(struct mt7996_dev *dev)
 {
 #define HIF_TXD_V2_1	4
@@ -287,6 +301,8 @@ void mt7996_mac_init(struct mt7996_dev *dev)
 
 	for (i = MT_BAND0; i <= MT_BAND2; i++)
 		mt7996_mac_init_band(dev, i);
+
+	mt7996_mac_init_basic_rates(dev);
 }
 
 int mt7996_txbf_init(struct mt7996_dev *dev)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index 1e9452c33e47..c8ada309d73f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -252,6 +252,17 @@ void mt7996_mac_enable_rtscts(struct mt7996_dev *dev,
 		mt76_clear(dev, addr, BIT(5));
 }
 
+void mt7996_mac_set_fixed_rate_table(struct mt7996_dev *dev,
+				     u8 tbl_idx, u16 rate_idx)
+{
+	u32 ctrl = MT_WTBL_ITCR_WR | MT_WTBL_ITCR_EXEC | tbl_idx;
+
+	mt76_wr(dev, MT_WTBL_ITDR0, rate_idx);
+	/* use wtbl spe idx */
+	mt76_wr(dev, MT_WTBL_ITDR1, MT_WTBL_SPE_IDX_SEL);
+	mt76_wr(dev, MT_WTBL_ITCR, ctrl);
+}
+
 static void
 mt7996_mac_decode_he_radiotap_ru(struct mt76_rx_status *status,
 				 struct ieee80211_radiotap_he *he,
@@ -985,7 +996,7 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 {
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_vif *vif = info->control.vif;
-	struct mt76_phy *mphy = &dev->mphy;
+	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
 	u8 band_idx = (info->hw_queue & MT_TX_HW_QUEUE_PHY) >> 2;
 	u8 p_fmt, q_idx, omac_idx = 0, wmm_idx = 0;
 	bool is_8023 = info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP;
@@ -997,15 +1008,11 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 					 BSS_CHANGED_FILS_DISCOVERY));
 
 	if (vif) {
-		struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
-
 		omac_idx = mvif->mt76.omac_idx;
 		wmm_idx = mvif->mt76.wmm_idx;
 		band_idx = mvif->mt76.band_idx;
 	}
 
-	mphy = mt76_dev_phy(&dev->mt76, band_idx);
-
 	if (inband_disc) {
 		p_fmt = MT_TX_TYPE_FW;
 		q_idx = MT_LMAC_ALTX0;
@@ -1063,18 +1070,9 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 		mt7996_mac_write_txwi_80211(dev, txwi, skb, key);
 
 	if (txwi[1] & cpu_to_le32(MT_TXD1_FIXED_RATE)) {
-		/* Fixed rata is available just for 802.11 txd */
-		struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
-		bool multicast = is_multicast_ether_addr(hdr->addr1);
-		u16 rate = mt76_connac2_mac_tx_rate_val(mphy, vif, beacon,
-							multicast);
-
-		/* fix to bw 20 */
-		val = MT_TXD6_FIXED_BW |
-		      FIELD_PREP(MT_TXD6_BW, 0) |
-		      FIELD_PREP(MT_TXD6_TX_RATE, rate);
-
-		txwi[6] |= cpu_to_le32(val);
+		u8 idx = mvif->basic_rates_idx;
+
+		txwi[6] |= FIELD_PREP(MT_TXD6_TX_RATE, idx);
 		txwi[3] |= cpu_to_le32(MT_TXD3_BA_DISABLE);
 	}
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
index 0318e16fcfab..b2bbe0e09e26 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
@@ -224,14 +224,11 @@ enum tx_mgnt_type {
 
 #define MT_TXD6_TX_SRC			GENMASK(31, 30)
 #define MT_TXD6_VTA			BIT(28)
-#define MT_TXD6_FIXED_BW		BIT(25)
-#define MT_TXD6_BW			GENMASK(24, 22)
+#define MT_TXD6_BW			GENMASK(25, 22)
 #define MT_TXD6_TX_RATE			GENMASK(21, 16)
 #define MT_TXD6_TIMESTAMP_OFS_EN	BIT(15)
 #define MT_TXD6_TIMESTAMP_OFS_IDX	GENMASK(14, 10)
 #define MT_TXD6_MSDU_CNT		GENMASK(9, 4)
-#define MT_TXD6_SPE_ID_IDX		BIT(10)
-#define MT_TXD6_ANT_ID			GENMASK(7, 4)
 #define MT_TXD6_DIS_MAT			BIT(3)
 #define MT_TXD6_DAS			BIT(2)
 #define MT_TXD6_AMSDU_CAP		BIT(1)
@@ -245,7 +242,7 @@ enum tx_mgnt_type {
 #define MT_TXD7_UDP_TCP_SUM		BIT(15)
 #define MT_TXD7_TX_TIME			GENMASK(9, 0)
 
-#define MT_TX_RATE_STBC			BIT(13)
+#define MT_TX_RATE_STBC			BIT(14)
 #define MT_TX_RATE_NSS			GENMASK(13, 10)
 #define MT_TX_RATE_MODE			GENMASK(9, 6)
 #define MT_TX_RATE_SU_EXT_TONE		BIT(5)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 7ddfcf68ebd1..1243e5f925ad 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -5,6 +5,7 @@
 
 #include "mt7996.h"
 #include "mcu.h"
+#include "mac.h"
 
 static bool mt7996_dev_running(struct mt7996_dev *dev)
 {
@@ -219,6 +220,11 @@ static int mt7996_add_interface(struct ieee80211_hw *hw,
 		vif->offload_flags = 0;
 	vif->offload_flags |= IEEE80211_OFFLOAD_ENCAP_4ADDR;
 
+	if (phy->mt76->chandef.chan->band != NL80211_BAND_2GHZ)
+		mvif->basic_rates_idx = MT7996_BASIC_RATES_TBL + 4;
+	else
+		mvif->basic_rates_idx = MT7996_BASIC_RATES_TBL;
+
 	mt7996_init_bitrate_mask(vif);
 
 	mt7996_mcu_add_bss_info(phy, vif, true);
@@ -496,11 +502,30 @@ mt7996_update_bss_color(struct ieee80211_hw *hw,
 	}
 }
 
+static u8
+mt7996_get_rates_table(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
+{
+	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
+	struct mt76_phy *mphy = hw->priv;
+	u16 rate;
+	u8 i, idx;
+
+	rate = mt76_connac2_mac_tx_rate_val(mphy, vif, false, false);
+
+	idx = FIELD_GET(MT_TX_RATE_IDX, rate);
+	for (i = 0; i < ARRAY_SIZE(mt76_rates); i++)
+		if ((mt76_rates[i].hw_value & GENMASK(7, 0)) == idx)
+			return MT7996_BASIC_RATES_TBL + i;
+
+	return mvif->basic_rates_idx;
+}
+
 static void mt7996_bss_info_changed(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
 				    struct ieee80211_bss_conf *info,
 				    u64 changed)
 {
+	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
 	struct mt7996_phy *phy = mt7996_hw_phy(hw);
 	struct mt7996_dev *dev = mt7996_hw_dev(hw);
 
@@ -532,6 +557,9 @@ static void mt7996_bss_info_changed(struct ieee80211_hw *hw,
 		}
 	}
 
+	if (changed & BSS_CHANGED_BASIC_RATES)
+		mvif->basic_rates_idx = mt7996_get_rates_table(hw, vif);
+
 	if (changed & BSS_CHANGED_BEACON_ENABLED && info->enable_beacon) {
 		mt7996_mcu_add_bss_info(phy, vif, true);
 		mt7996_mcu_add_sta(dev, vif, NULL, true);
@@ -891,6 +919,7 @@ mt7996_set_antenna(struct ieee80211_hw *hw, u32 tx_ant, u32 rx_ant)
 	mt7996_set_stream_vht_txbf_caps(phy);
 	mt7996_set_stream_he_eht_caps(phy);
 
+	/* TODO: update bmc_wtbl spe_idx when antenna changes */
 	mutex_unlock(&dev->mt76.mutex);
 
 	return 0;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index c99605a1ee6b..3db74ae109dc 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -597,25 +597,23 @@ mt7996_mcu_bss_he_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 }
 
 static void
-mt7996_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt7996_phy *phy)
+mt7996_mcu_bss_bmc_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
+		       struct mt7996_phy *phy)
 {
+	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
 	struct bss_rate_tlv *bmc;
 	struct cfg80211_chan_def *chandef = &phy->mt76->chandef;
 	enum nl80211_band band = chandef->chan->band;
 	struct tlv *tlv;
+	u8 idx = mvif->basic_rates_idx;
 
 	tlv = mt7996_mcu_add_uni_tlv(skb, UNI_BSS_INFO_RATE, sizeof(*bmc));
 
 	bmc = (struct bss_rate_tlv *)tlv;
-	if (band == NL80211_BAND_2GHZ) {
-		bmc->short_preamble = true;
-	} else {
-		bmc->bc_trans = cpu_to_le16(0x8080);
-		bmc->mc_trans = cpu_to_le16(0x8080);
-		bmc->bc_fixed_rate = 1;
-		bmc->mc_fixed_rate = 1;
-		bmc->short_preamble = 1;
-	}
+
+	bmc->short_preamble = (band == NL80211_BAND_2GHZ);
+	bmc->bc_fixed_rate = idx;
+	bmc->mc_fixed_rate = idx;
 }
 
 static void
@@ -823,7 +821,7 @@ int mt7996_mcu_add_bss_info(struct mt7996_phy *phy,
 
 	if (enable) {
 		mt7996_mcu_bss_rfch_tlv(skb, vif, phy);
-		mt7996_mcu_bss_bmc_tlv(skb, phy);
+		mt7996_mcu_bss_bmc_tlv(skb, vif, phy);
 		mt7996_mcu_bss_ra_tlv(skb, vif, phy);
 		mt7996_mcu_bss_txcmd_tlv(skb, true);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 4001cf12b671..5cf5ea8cece9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -43,6 +43,9 @@
 #define MT7996_MAX_STA_TWT_AGRT		8
 #define MT7996_MAX_QUEUE		(__MT_RXQ_MAX +	__MT_MCUQ_MAX + 3)
 
+/* NOTE: used to map mt76_rates. idx may change if firmware expands table */
+#define MT7996_BASIC_RATES_TBL		11
+
 struct mt7996_vif;
 struct mt7996_sta;
 struct mt7996_dfs_pulse;
@@ -120,6 +123,8 @@ struct mt7996_vif {
 
 	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
 	struct cfg80211_bitrate_mask bitrate_mask;
+
+	u8 basic_rates_idx;
 };
 
 /* per-phy stats.  */
@@ -505,6 +510,8 @@ void mt7996_mac_cca_stats_reset(struct mt7996_phy *phy);
 void mt7996_mac_enable_nf(struct mt7996_dev *dev, u8 band);
 void mt7996_mac_enable_rtscts(struct mt7996_dev *dev,
 			      struct ieee80211_vif *vif, bool enable);
+void mt7996_mac_set_fixed_rate_table(struct mt7996_dev *dev,
+				     u8 tbl_idx, u16 rate_idx);
 void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 			   struct sk_buff *skb, struct mt76_wcid *wcid,
 			   struct ieee80211_key_conf *key, int pid,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
index 75f856205559..d1d3d154195d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h
@@ -228,6 +228,13 @@ enum base_rev {
 #define MT_WTBL_UPDATE_ADM_COUNT_CLEAR		BIT(14)
 #define MT_WTBL_UPDATE_BUSY			BIT(31)
 
+#define MT_WTBL_ITCR				MT_WTBLON_TOP(0x3b0)
+#define MT_WTBL_ITCR_WR				BIT(16)
+#define MT_WTBL_ITCR_EXEC			BIT(31)
+#define MT_WTBL_ITDR0				MT_WTBLON_TOP(0x3b8)
+#define MT_WTBL_ITDR1				MT_WTBLON_TOP(0x3bc)
+#define MT_WTBL_SPE_IDX_SEL			BIT(6)
+
 /* WTBL */
 #define MT_WTBL_BASE				0x820d8000
 #define MT_WTBL_LMAC_ID				GENMASK(14, 8)
-- 
cgit v1.2.3-70-g09d2


From ab0eec4bf225c5e78fd10d764d536300376f8445 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Fri, 14 Apr 2023 04:23:32 +0800
Subject: wifi: mt76: mt7996: enable BSS_CHANGED_MCAST_RATE support

Similar to BSS_CHANGED_BASIC_RATES, this enables mcast rate
configuration through fixed rate tables.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Change-Id: Ifc305e8c7de9a7df4ad5f856e2097d721a886aaa
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    |  6 ++++++
 drivers/net/wireless/mediatek/mt76/mt7996/main.c   | 12 +++++++++---
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c    |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  1 +
 4 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index c8ada309d73f..fbb92c39beee 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -1070,8 +1070,14 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 		mt7996_mac_write_txwi_80211(dev, txwi, skb, key);
 
 	if (txwi[1] & cpu_to_le32(MT_TXD1_FIXED_RATE)) {
+		struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+		bool mcast = ieee80211_is_data(hdr->frame_control) &&
+			     is_multicast_ether_addr(hdr->addr1);
 		u8 idx = mvif->basic_rates_idx;
 
+		if (mcast && mvif->mcast_rates_idx)
+			idx = mvif->mcast_rates_idx;
+
 		txwi[6] |= FIELD_PREP(MT_TXD6_TX_RATE, idx);
 		txwi[3] |= cpu_to_le32(MT_TXD3_BA_DISABLE);
 	}
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 1243e5f925ad..f861fbb7de47 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -503,14 +503,15 @@ mt7996_update_bss_color(struct ieee80211_hw *hw,
 }
 
 static u8
-mt7996_get_rates_table(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
+mt7996_get_rates_table(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+		       bool mcast)
 {
 	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
 	struct mt76_phy *mphy = hw->priv;
 	u16 rate;
 	u8 i, idx;
 
-	rate = mt76_connac2_mac_tx_rate_val(mphy, vif, false, false);
+	rate = mt76_connac2_mac_tx_rate_val(mphy, vif, false, mcast);
 
 	idx = FIELD_GET(MT_TX_RATE_IDX, rate);
 	for (i = 0; i < ARRAY_SIZE(mt76_rates); i++)
@@ -557,8 +558,13 @@ static void mt7996_bss_info_changed(struct ieee80211_hw *hw,
 		}
 	}
 
+	if (changed & BSS_CHANGED_MCAST_RATE)
+		mvif->mcast_rates_idx =
+			mt7996_get_rates_table(hw, vif, true);
+
 	if (changed & BSS_CHANGED_BASIC_RATES)
-		mvif->basic_rates_idx = mt7996_get_rates_table(hw, vif);
+		mvif->basic_rates_idx =
+			mt7996_get_rates_table(hw, vif, false);
 
 	if (changed & BSS_CHANGED_BEACON_ENABLED && info->enable_beacon) {
 		mt7996_mcu_add_bss_info(phy, vif, true);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 3db74ae109dc..f96992b55c7a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -605,7 +605,8 @@ mt7996_mcu_bss_bmc_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 	struct cfg80211_chan_def *chandef = &phy->mt76->chandef;
 	enum nl80211_band band = chandef->chan->band;
 	struct tlv *tlv;
-	u8 idx = mvif->basic_rates_idx;
+	u8 idx = mvif->mcast_rates_idx ?
+		 mvif->mcast_rates_idx : mvif->basic_rates_idx;
 
 	tlv = mt7996_mcu_add_uni_tlv(skb, UNI_BSS_INFO_RATE, sizeof(*bmc));
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index 5cf5ea8cece9..a1a6ee0de2ff 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -125,6 +125,7 @@ struct mt7996_vif {
 	struct cfg80211_bitrate_mask bitrate_mask;
 
 	u8 basic_rates_idx;
+	u8 mcast_rates_idx;
 };
 
 /* per-phy stats.  */
-- 
cgit v1.2.3-70-g09d2


From c2171b068beea766311e4c2858ef8497504c6e6d Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Fri, 14 Apr 2023 04:23:33 +0800
Subject: wifi: mt76: mt7996: enable configured beacon tx rate

The user is allowed to change beacon tx rate (HT/VHT/HE) from hostapd.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c    |  2 ++
 drivers/net/wireless/mediatek/mt76/mt7996/main.c   | 26 +++++++++++++++++-----
 drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h |  2 ++
 3 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index fbb92c39beee..1ae2c85e66ac 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -1077,6 +1077,8 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi,
 
 		if (mcast && mvif->mcast_rates_idx)
 			idx = mvif->mcast_rates_idx;
+		else if (beacon && mvif->beacon_rates_idx)
+			idx = mvif->beacon_rates_idx;
 
 		txwi[6] |= FIELD_PREP(MT_TXD6_TX_RATE, idx);
 		txwi[3] |= cpu_to_le32(MT_TXD3_BA_DISABLE);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index f861fbb7de47..0975774fe244 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -504,14 +504,24 @@ mt7996_update_bss_color(struct ieee80211_hw *hw,
 
 static u8
 mt7996_get_rates_table(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-		       bool mcast)
+		       bool beacon, bool mcast)
 {
 	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
 	struct mt76_phy *mphy = hw->priv;
 	u16 rate;
-	u8 i, idx;
+	u8 i, idx, ht;
 
-	rate = mt76_connac2_mac_tx_rate_val(mphy, vif, false, mcast);
+	rate = mt76_connac2_mac_tx_rate_val(mphy, vif, beacon, mcast);
+	ht = FIELD_GET(MT_TX_RATE_MODE, rate) > MT_PHY_TYPE_OFDM;
+
+	if (beacon && ht) {
+		struct mt7996_dev *dev = mt7996_hw_dev(hw);
+
+		/* must odd index */
+		idx = MT7996_BEACON_RATES_TBL + 2 * (mvif->mt76.idx % 20);
+		mt7996_mac_set_fixed_rate_table(dev, idx, rate);
+		return idx;
+	}
 
 	idx = FIELD_GET(MT_TX_RATE_IDX, rate);
 	for (i = 0; i < ARRAY_SIZE(mt76_rates); i++)
@@ -560,11 +570,11 @@ static void mt7996_bss_info_changed(struct ieee80211_hw *hw,
 
 	if (changed & BSS_CHANGED_MCAST_RATE)
 		mvif->mcast_rates_idx =
-			mt7996_get_rates_table(hw, vif, true);
+			mt7996_get_rates_table(hw, vif, false, true);
 
 	if (changed & BSS_CHANGED_BASIC_RATES)
 		mvif->basic_rates_idx =
-			mt7996_get_rates_table(hw, vif, false);
+			mt7996_get_rates_table(hw, vif, false, false);
 
 	if (changed & BSS_CHANGED_BEACON_ENABLED && info->enable_beacon) {
 		mt7996_mcu_add_bss_info(phy, vif, true);
@@ -582,8 +592,12 @@ static void mt7996_bss_info_changed(struct ieee80211_hw *hw,
 		mt7996_update_bss_color(hw, vif, &info->he_bss_color);
 
 	if (changed & (BSS_CHANGED_BEACON |
-		       BSS_CHANGED_BEACON_ENABLED))
+		       BSS_CHANGED_BEACON_ENABLED)) {
+		mvif->beacon_rates_idx =
+			mt7996_get_rates_table(hw, vif, true, false);
+
 		mt7996_mcu_add_beacon(hw, vif, info->enable_beacon);
+	}
 
 	if (changed & BSS_CHANGED_UNSOL_BCAST_PROBE_RESP ||
 	    changed & BSS_CHANGED_FILS_DISCOVERY)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
index a1a6ee0de2ff..4d7dcb95a620 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h
@@ -45,6 +45,7 @@
 
 /* NOTE: used to map mt76_rates. idx may change if firmware expands table */
 #define MT7996_BASIC_RATES_TBL		11
+#define MT7996_BEACON_RATES_TBL		25
 
 struct mt7996_vif;
 struct mt7996_sta;
@@ -126,6 +127,7 @@ struct mt7996_vif {
 
 	u8 basic_rates_idx;
 	u8 mcast_rates_idx;
+	u8 beacon_rates_idx;
 };
 
 /* per-phy stats.  */
-- 
cgit v1.2.3-70-g09d2


From 27db47ab1f47906c2392f9d246e244e412b19278 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Sun, 12 Mar 2023 14:11:25 +0800
Subject: wifi: mt76: mt7996: enable mesh HW amsdu/de-amsdu support

This enables HW offloading amsdu/de-amsdu support for 802.11s mesh
interface.

Co-developed-by: Bo Jiao <bo.jiao@mediatek.com>
Signed-off-by: Bo Jiao <bo.jiao@mediatek.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c  | 24 ++++++++++++++++--------
 drivers/net/wireless/mediatek/mt76/mt7996/mac.h  |  2 ++
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c  |  7 +++++++
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.h  |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mmio.c |  1 +
 5 files changed, 27 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index 1ae2c85e66ac..0639bd3808c6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -641,6 +641,8 @@ mt7996_mac_fill_rx(struct mt7996_dev *dev, struct sk_buff *skb)
 	u32 rxd4 = le32_to_cpu(rxd[4]);
 	u32 csum_mask = MT_RXD0_NORMAL_IP_SUM | MT_RXD0_NORMAL_UDP_TCP_SUM;
 	u32 csum_status = *(u32 *)skb->cb;
+	u32 mesh_mask = MT_RXD0_MESH | MT_RXD0_MHCP;
+	bool is_mesh = (rxd0 & mesh_mask) == mesh_mask;
 	bool unicast, insert_ccmp_hdr = false;
 	u8 remove_pad, amsdu_info, band_idx;
 	u8 mode = 0, qos_ctl = 0;
@@ -832,19 +834,16 @@ mt7996_mac_fill_rx(struct mt7996_dev *dev, struct sk_buff *skb)
 		int pad_start = 0;
 
 		skb_pull(skb, hdr_gap);
-		if (!hdr_trans && status->amsdu) {
+		if (!hdr_trans && status->amsdu && !(ieee80211_has_a4(fc) && is_mesh)) {
 			pad_start = ieee80211_get_hdrlen_from_skb(skb);
-		} else if (hdr_trans && (rxd2 & MT_RXD2_NORMAL_HDR_TRANS_ERROR)) {
+		} else if (hdr_trans && (rxd2 & MT_RXD2_NORMAL_HDR_TRANS_ERROR) &&
+			   get_unaligned_be16(skb->data + pad_start) == ETH_P_8021Q) {
 			/* When header translation failure is indicated,
 			 * the hardware will insert an extra 2-byte field
 			 * containing the data length after the protocol
 			 * type field.
 			 */
-			pad_start = 12;
-			if (get_unaligned_be16(skb->data + pad_start) == ETH_P_8021Q)
-				pad_start += 4;
-			else
-				pad_start = 0;
+			pad_start = 16;
 		}
 
 		if (pad_start) {
@@ -865,8 +864,17 @@ mt7996_mac_fill_rx(struct mt7996_dev *dev, struct sk_buff *skb)
 		hdr = mt76_skb_get_hdr(skb);
 		fc = hdr->frame_control;
 		if (ieee80211_is_data_qos(fc)) {
+			u8 *qos = ieee80211_get_qos_ctl(hdr);
+
 			seq_ctrl = le16_to_cpu(hdr->seq_ctrl);
-			qos_ctl = *ieee80211_get_qos_ctl(hdr);
+			qos_ctl = *qos;
+
+			/* Mesh DA/SA/Length will be stripped after hardware
+			 * de-amsdu, so here needs to clear amsdu present bit
+			 * to mark it as a normal mesh frame.
+			 */
+			if (ieee80211_has_a4(fc) && is_mesh && status->amsdu)
+				*qos &= ~IEEE80211_QOS_CTL_A_MSDU_PRESENT;
 		}
 	} else {
 		status->flag |= RX_FLAG_8023;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
index b2bbe0e09e26..bc4e6c55373e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h
@@ -12,6 +12,8 @@
 #define MT_RXD0_LENGTH			GENMASK(15, 0)
 #define MT_RXD0_PKT_TYPE		GENMASK(31, 27)
 
+#define MT_RXD0_MESH			BIT(18)
+#define MT_RXD0_MHCP			BIT(19)
 #define MT_RXD0_NORMAL_ETH_TYPE_OFS	GENMASK(22, 16)
 #define MT_RXD0_NORMAL_IP_SUM		BIT(23)
 #define MT_RXD0_NORMAL_UDP_TCP_SUM	BIT(24)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index f96992b55c7a..88e2f9d0e513 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -1022,6 +1022,7 @@ mt7996_mcu_sta_amsdu_tlv(struct mt7996_dev *dev, struct sk_buff *skb,
 	struct tlv *tlv;
 
 	if (vif->type != NL80211_IFTYPE_STATION &&
+	    vif->type != NL80211_IFTYPE_MESH_POINT &&
 	    vif->type != NL80211_IFTYPE_AP)
 		return;
 
@@ -1470,6 +1471,12 @@ mt7996_mcu_sta_hdr_trans_tlv(struct mt7996_dev *dev, struct sk_buff *skb,
 		hdr_trans->to_ds = true;
 		hdr_trans->from_ds = true;
 	}
+
+	if (vif->type == NL80211_IFTYPE_MESH_POINT) {
+		hdr_trans->to_ds = true;
+		hdr_trans->from_ds = true;
+		hdr_trans->mesh = true;
+	}
 }
 
 static enum mcu_mmps_mode
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h
index 6f92b9a7d86d..d7075a4d0667 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h
@@ -396,7 +396,7 @@ struct sta_rec_hdr_trans {
 	u8 from_ds;
 	u8 to_ds;
 	u8 dis_rx_hdr_tran;
-	u8 rsv;
+	u8 mesh;
 } __packed;
 
 struct hdr_trans_en {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
index 879b0e90b52e..3a591a7b47ae 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c
@@ -327,6 +327,7 @@ struct mt7996_dev *mt7996_mmio_probe(struct device *pdev,
 		/* txwi_size = txd size + txp size */
 		.txwi_size = MT_TXD_SIZE + sizeof(struct mt76_connac_fw_txp),
 		.drv_flags = MT_DRV_TXWI_NO_FREE |
+			     MT_DRV_AMSDU_OFFLOAD |
 			     MT_DRV_HW_MGMT_TXQ,
 		.survey_flags = SURVEY_INFO_TIME_TX |
 				SURVEY_INFO_TIME_RX |
-- 
cgit v1.2.3-70-g09d2


From e12b2e99b8799f26432528934edc8677888ad72f Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Fri, 11 Nov 2022 19:50:46 +0100
Subject: wifi: mt76: mt7615: increase eeprom size for mt7663

mt7663 efuse has 0x600 bytes instead of 0x400. Increase the size in order
to fix issues with incomplete calibration data

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c | 7 ++++++-
 drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h | 1 +
 3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c
index 6dbaaf95ee38..68e88224b8b1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c
@@ -47,6 +47,9 @@ static int mt7615_efuse_init(struct mt7615_dev *dev, u32 base)
 	void *buf;
 	u32 val;
 
+	if (is_mt7663(&dev->mt76))
+		len = MT7663_EEPROM_SIZE;
+
 	val = mt76_rr(dev, base + MT_EFUSE_BASE_CTRL);
 	if (val & MT_EFUSE_BASE_CTRL_EMPTY)
 		return 0;
@@ -72,6 +75,8 @@ static int mt7615_eeprom_load(struct mt7615_dev *dev, u32 addr)
 {
 	int ret;
 
+	BUILD_BUG_ON(MT7615_EEPROM_FULL_SIZE < MT7663_EEPROM_SIZE);
+
 	ret = mt76_eeprom_init(&dev->mt76, MT7615_EEPROM_FULL_SIZE);
 	if (ret < 0)
 		return ret;
@@ -336,7 +341,7 @@ int mt7615_eeprom_init(struct mt7615_dev *dev, u32 addr)
 	ret = mt7615_check_eeprom(&dev->mt76);
 	if (ret && dev->mt76.otp.data) {
 		memcpy(dev->mt76.eeprom.data, dev->mt76.otp.data,
-		       MT7615_EEPROM_SIZE);
+		       dev->mt76.otp.size);
 	} else {
 		dev->flash_eeprom = true;
 		mt7615_cal_free_data(dev);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h
index a024dee10362..a67fbb90f5b3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h
@@ -46,7 +46,7 @@ enum mt7615_eeprom_field {
 
 	MT7615_EE_MAX =				0x3bf,
 	MT7622_EE_MAX =				0x3db,
-	MT7663_EE_MAX =				0x400,
+	MT7663_EE_MAX =				0x600,
 };
 
 #define MT_EE_RATE_POWER_MASK			GENMASK(5, 0)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
index 932cdad7e7d4..0381c53bc96a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h
@@ -51,6 +51,7 @@
 #define MT7663_FIRMWARE_N9		"mediatek/mt7663_n9_rebb.bin"
 
 #define MT7615_EEPROM_SIZE		1024
+#define MT7663_EEPROM_SIZE		1536
 #define MT7615_TOKEN_SIZE		4096
 
 #define MT_FRAC_SCALE		12
-- 
cgit v1.2.3-70-g09d2


From f4d63a87b527de258eec5bd6e9547f063d472b79 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Sat, 31 Dec 2022 08:44:33 +0100
Subject: wifi: mt76: dma: use napi_build_skb

Improves performance by using bulk allocation

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/dma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c
index 6c0174313d09..465190ebaf1c 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/dma.c
@@ -853,7 +853,7 @@ mt76_dma_rx_process(struct mt76_dev *dev, struct mt76_queue *q, int budget)
 		    !(dev->drv->rx_check(dev, data, len)))
 			goto free_frag;
 
-		skb = build_skb(data, q->buf_size);
+		skb = napi_build_skb(data, q->buf_size);
 		if (!skb)
 			goto free_frag;
 
-- 
cgit v1.2.3-70-g09d2


From 230a167e094770fdcc104481719ef7b1a706fb27 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Thu, 30 Mar 2023 11:07:52 +0200
Subject: wifi: mt76: set NL80211_EXT_FEATURE_CAN_REPLACE_PTK0 on supported
 drivers

Drivers that do not generate IV/PN in software can safely rekey PTK0

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/init.c | 1 +
 drivers/net/wireless/mediatek/mt76/mt7915/init.c | 1 +
 drivers/net/wireless/mediatek/mt76/mt7921/init.c | 1 +
 drivers/net/wireless/mediatek/mt76/mt7996/init.c | 1 +
 4 files changed, 4 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/init.c b/drivers/net/wireless/mediatek/mt76/mt7615/init.c
index 5fa6f097ec30..621e69f07e3c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/init.c
@@ -396,6 +396,7 @@ mt7615_init_wiphy(struct ieee80211_hw *hw)
 
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_SET_SCAN_DWELL);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_VHT_IBSS);
+	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 
 	ieee80211_hw_set(hw, SINGLE_SCAN_ON_ALL_BANDS);
 	ieee80211_hw_set(hw, TX_STATUS_NO_AMPDU_LEN);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
index 538ec0f8e630..ac2049f49bb3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
@@ -369,6 +369,7 @@ mt7915_init_wiphy(struct mt7915_phy *phy)
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_UNSOL_BCAST_PROBE_RESP);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_FILS_DISCOVERY);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT);
+	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 
 	if (!is_mt7915(&dev->mt76))
 		wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_STA_TX_PWR);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
index 5955c7c5b91a..e929f6eb65ce 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
@@ -124,6 +124,7 @@ mt7921_init_wiphy(struct ieee80211_hw *hw)
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_BEACON_RATE_VHT);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_BEACON_RATE_HE);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT);
+	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 
 	ieee80211_hw_set(hw, SINGLE_SCAN_ON_ALL_BANDS);
 	ieee80211_hw_set(hw, HAS_RATE_CONTROL);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
index 90a92c1611f3..f1b48cdda58f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c
@@ -182,6 +182,7 @@ mt7996_init_wiphy(struct ieee80211_hw *hw)
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_UNSOL_BCAST_PROBE_RESP);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_FILS_DISCOVERY);
 	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT);
+	wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 
 	if (!mdev->dev->of_node ||
 	    !of_property_read_bool(mdev->dev->of_node,
-- 
cgit v1.2.3-70-g09d2


From 3b522cadedfe6e9e0e8193d7d4ab5aa8d0c73209 Mon Sep 17 00:00:00 2001
From: Ryder Lee <ryder.lee@mediatek.com>
Date: Wed, 19 Apr 2023 13:58:40 +0800
Subject: wifi: mt76: mt7996: fill txd by host driver

The hardware SDO has issue to fill txd for the moment, so fallback to
driver filling method.

Fixes: 98686cd21624 (wifi: mt76: mt7996: add driver for MediaTek Wi-Fi 7 (802.11be) devices)
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index 0639bd3808c6..130eb7b4fd91 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -1132,11 +1132,8 @@ int mt7996_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 		return id;
 
 	pid = mt76_tx_status_skb_add(mdev, wcid, tx_info->skb);
-	memset(txwi_ptr, 0, MT_TXD_SIZE);
-	/* Transmit non qos data by 802.11 header and need to fill txd by host*/
-	if (!is_8023 || pid >= MT_PACKET_ID_FIRST)
-		mt7996_mac_write_txwi(dev, txwi_ptr, tx_info->skb, wcid, key,
-				      pid, qid, 0);
+	mt7996_mac_write_txwi(dev, txwi_ptr, tx_info->skb, wcid, key,
+			      pid, qid, 0);
 
 	txp = (struct mt76_connac_txp_common *)(txwi + MT_TXD_SIZE);
 	for (i = 0; i < nbuf; i++) {
@@ -1145,10 +1142,8 @@ int mt7996_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 	}
 	txp->fw.nbuf = nbuf;
 
-	txp->fw.flags = cpu_to_le16(MT_CT_INFO_FROM_HOST);
-
-	if (!is_8023 || pid >= MT_PACKET_ID_FIRST)
-		txp->fw.flags |= cpu_to_le16(MT_CT_INFO_APPLY_TXD);
+	txp->fw.flags =
+		cpu_to_le16(MT_CT_INFO_FROM_HOST | MT_CT_INFO_APPLY_TXD);
 
 	if (!key)
 		txp->fw.flags |= cpu_to_le16(MT_CT_INFO_NONE_CIPHER_FRAME);
-- 
cgit v1.2.3-70-g09d2


From 73db1b8f2bb6725b7391e85aab41fdf592b3c0c1 Mon Sep 17 00:00:00 2001
From: Tzung-Bi Shih <tzungbi@kernel.org>
Date: Wed, 19 Apr 2023 13:15:26 +0800
Subject: netfilter: conntrack: fix wrong ct->timeout value

(struct nf_conn)->timeout is an interval before the conntrack
confirmed.  After confirmed, it becomes a timestamp.

It is observed that timeout of an unconfirmed conntrack:
- Set by calling ctnetlink_change_timeout(). As a result,
  `nfct_time_stamp` was wrongly added to `ct->timeout` twice.
- Get by calling ctnetlink_dump_timeout(). As a result,
  `nfct_time_stamp` was wrongly subtracted.

Call Trace:
 <TASK>
 dump_stack_lvl
 ctnetlink_dump_timeout
 __ctnetlink_glue_build
 ctnetlink_glue_build
 __nfqnl_enqueue_packet
 nf_queue
 nf_hook_slow
 ip_mc_output
 ? __pfx_ip_finish_output
 ip_send_skb
 ? __pfx_dst_output
 udp_send_skb
 udp_sendmsg
 ? __pfx_ip_generic_getfrag
 sock_sendmsg

Separate the 2 cases in:
- Setting `ct->timeout` in __nf_ct_set_timeout().
- Getting `ct->timeout` in ctnetlink_dump_timeout().

Pablo appends:

Update ctnetlink to set up the timeout _after_ the IPS_CONFIRMED flag is
set on, otherwise conntrack creation via ctnetlink breaks.

Note that the problem described in this patch occurs since the
introduction of the nfnetlink_queue conntrack support, select a
sufficiently old Fixes: tag for -stable kernel to pick up this fix.

Fixes: a4b4766c3ceb ("netfilter: nfnetlink_queue: rename related to nfqueue attaching conntrack info")
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_conntrack_core.h |  6 +++++-
 net/netfilter/nf_conntrack_netlink.c      | 13 +++++++++----
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h
index 71d1269fe4d4..3384859a8921 100644
--- a/include/net/netfilter/nf_conntrack_core.h
+++ b/include/net/netfilter/nf_conntrack_core.h
@@ -89,7 +89,11 @@ static inline void __nf_ct_set_timeout(struct nf_conn *ct, u64 timeout)
 {
 	if (timeout > INT_MAX)
 		timeout = INT_MAX;
-	WRITE_ONCE(ct->timeout, nfct_time_stamp + (u32)timeout);
+
+	if (nf_ct_is_confirmed(ct))
+		WRITE_ONCE(ct->timeout, nfct_time_stamp + (u32)timeout);
+	else
+		ct->timeout = (u32)timeout;
 }
 
 int __nf_ct_change_timeout(struct nf_conn *ct, u64 cta_timeout);
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index d3ee18854698..6f3b23a6653c 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -176,7 +176,12 @@ nla_put_failure:
 static int ctnetlink_dump_timeout(struct sk_buff *skb, const struct nf_conn *ct,
 				  bool skip_zero)
 {
-	long timeout = nf_ct_expires(ct) / HZ;
+	long timeout;
+
+	if (nf_ct_is_confirmed(ct))
+		timeout = nf_ct_expires(ct) / HZ;
+	else
+		timeout = ct->timeout / HZ;
 
 	if (skip_zero && timeout == 0)
 		return 0;
@@ -2253,9 +2258,6 @@ ctnetlink_create_conntrack(struct net *net,
 	if (!cda[CTA_TIMEOUT])
 		goto err1;
 
-	timeout = (u64)ntohl(nla_get_be32(cda[CTA_TIMEOUT])) * HZ;
-	__nf_ct_set_timeout(ct, timeout);
-
 	rcu_read_lock();
  	if (cda[CTA_HELP]) {
 		char *helpname = NULL;
@@ -2319,6 +2321,9 @@ ctnetlink_create_conntrack(struct net *net,
 	/* we must add conntrack extensions before confirmation. */
 	ct->status |= IPS_CONFIRMED;
 
+	timeout = (u64)ntohl(nla_get_be32(cda[CTA_TIMEOUT])) * HZ;
+	__nf_ct_set_timeout(ct, timeout);
+
 	if (cda[CTA_STATUS]) {
 		err = ctnetlink_change_status(ct, cda);
 		if (err < 0)
-- 
cgit v1.2.3-70-g09d2


From 158441884772a19e840429e210bfd1879094306b Mon Sep 17 00:00:00 2001
From: Jaime Breva <jbreva@nayarsystems.com>
Date: Mon, 17 Apr 2023 08:07:24 +0200
Subject: net: wwan: Expose secondary AT port on DATA1

Our use-case needs two AT ports available:
One for running a ppp daemon, and another one for management

This patch enables a second AT port on DATA1

Signed-off-by: Jaime Breva <jbreva@nayarsystems.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wwan/rpmsg_wwan_ctrl.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wwan/rpmsg_wwan_ctrl.c b/drivers/net/wwan/rpmsg_wwan_ctrl.c
index 06f4b02f1552..86b60aadfa11 100644
--- a/drivers/net/wwan/rpmsg_wwan_ctrl.c
+++ b/drivers/net/wwan/rpmsg_wwan_ctrl.c
@@ -149,6 +149,7 @@ static const struct rpmsg_device_id rpmsg_wwan_ctrl_id_table[] = {
 	/* RPMSG channels for Qualcomm SoCs with integrated modem */
 	{ .name = "DATA5_CNTL", .driver_data = WWAN_PORT_QMI },
 	{ .name = "DATA4", .driver_data = WWAN_PORT_AT },
+	{ .name = "DATA1", .driver_data = WWAN_PORT_AT },
 	{},
 };
 MODULE_DEVICE_TABLE(rpmsg, rpmsg_wwan_ctrl_id_table);
-- 
cgit v1.2.3-70-g09d2


From 3e8b4d6277fd19d98c817576954dd6a4ff3caa2b Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:23 +0200
Subject: net: dsa: qca8k: move qca8k_port_to_phy() to header

Move qca8k_port_to_phy() to qca8k header as it's useful for future
reference in Switch LEDs module since the same logic is applied to get
the right index of the switch port.
Make it inline as it's simple function that just decrease the port.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/qca8k-8xxx.c | 15 ---------------
 drivers/net/dsa/qca/qca8k.h      | 14 ++++++++++++++
 2 files changed, 14 insertions(+), 15 deletions(-)

diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c
index 62810903f1b3..20acbd0292f1 100644
--- a/drivers/net/dsa/qca/qca8k-8xxx.c
+++ b/drivers/net/dsa/qca/qca8k-8xxx.c
@@ -772,21 +772,6 @@ err_clear_skb:
 	return ret;
 }
 
-static u32
-qca8k_port_to_phy(int port)
-{
-	/* From Andrew Lunn:
-	 * Port 0 has no internal phy.
-	 * Port 1 has an internal PHY at MDIO address 0.
-	 * Port 2 has an internal PHY at MDIO address 1.
-	 * ...
-	 * Port 5 has an internal PHY at MDIO address 4.
-	 * Port 6 has no internal PHY.
-	 */
-
-	return port - 1;
-}
-
 static int
 qca8k_mdio_busy_wait(struct mii_bus *bus, u32 reg, u32 mask)
 {
diff --git a/drivers/net/dsa/qca/qca8k.h b/drivers/net/dsa/qca/qca8k.h
index 7996975d29d3..dd7deb9095d3 100644
--- a/drivers/net/dsa/qca/qca8k.h
+++ b/drivers/net/dsa/qca/qca8k.h
@@ -421,6 +421,20 @@ struct qca8k_fdb {
 	u8 mac[6];
 };
 
+static inline u32 qca8k_port_to_phy(int port)
+{
+	/* From Andrew Lunn:
+	 * Port 0 has no internal phy.
+	 * Port 1 has an internal PHY at MDIO address 0.
+	 * Port 2 has an internal PHY at MDIO address 1.
+	 * ...
+	 * Port 5 has an internal PHY at MDIO address 4.
+	 * Port 6 has no internal PHY.
+	 */
+
+	return port - 1;
+}
+
 /* Common setup function */
 extern const struct qca8k_mib_desc ar8327_mib[];
 extern const struct regmap_access_table qca8k_readable_table;
-- 
cgit v1.2.3-70-g09d2


From 1e264f9d2918b5737023c44a23ae04def1095210 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:24 +0200
Subject: net: dsa: qca8k: add LEDs basic support

Add LEDs basic support for qca8k Switch Family by adding basic
brightness_set() support.

Since these LEDs refelect port status, the default label is set to
":port". DT binding should describe the color and function of the
LEDs using standard LEDs api.
Each LED always have the device name as prefix. The device name is
composed from the mii bus id and the PHY addr resulting in example
names like:
- qca8k-0.0:00:amber:lan
- qca8k-0.0:00:white:lan
- qca8k-0.0:01:amber:lan
- qca8k-0.0:01:white:lan

These LEDs supports only blocking variant of the brightness_set()
function since they can sleep during access of the switch leds to set
the brightness.

While at it add to the qca8k header file each mode defined by the Switch
Documentation for future use.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/Kconfig      |   8 ++
 drivers/net/dsa/qca/Makefile     |   3 +
 drivers/net/dsa/qca/qca8k-8xxx.c |   5 +
 drivers/net/dsa/qca/qca8k-leds.c | 239 +++++++++++++++++++++++++++++++++++++++
 drivers/net/dsa/qca/qca8k.h      |  60 ++++++++++
 drivers/net/dsa/qca/qca8k_leds.h |  16 +++
 6 files changed, 331 insertions(+)
 create mode 100644 drivers/net/dsa/qca/qca8k-leds.c
 create mode 100644 drivers/net/dsa/qca/qca8k_leds.h

diff --git a/drivers/net/dsa/qca/Kconfig b/drivers/net/dsa/qca/Kconfig
index ba339747362c..7a86d6d6a246 100644
--- a/drivers/net/dsa/qca/Kconfig
+++ b/drivers/net/dsa/qca/Kconfig
@@ -15,3 +15,11 @@ config NET_DSA_QCA8K
 	help
 	  This enables support for the Qualcomm Atheros QCA8K Ethernet
 	  switch chips.
+
+config NET_DSA_QCA8K_LEDS_SUPPORT
+	bool "Qualcomm Atheros QCA8K Ethernet switch family LEDs support"
+	depends on NET_DSA_QCA8K
+	depends on LEDS_CLASS
+	help
+	  This enabled support for LEDs present on the Qualcomm Atheros
+	  QCA8K Ethernet switch chips.
diff --git a/drivers/net/dsa/qca/Makefile b/drivers/net/dsa/qca/Makefile
index 701f1d199e93..ce66b1984e5f 100644
--- a/drivers/net/dsa/qca/Makefile
+++ b/drivers/net/dsa/qca/Makefile
@@ -2,3 +2,6 @@
 obj-$(CONFIG_NET_DSA_AR9331)	+= ar9331.o
 obj-$(CONFIG_NET_DSA_QCA8K)	+= qca8k.o
 qca8k-y 			+= qca8k-common.o qca8k-8xxx.o
+ifdef CONFIG_NET_DSA_QCA8K_LEDS_SUPPORT
+qca8k-y				+= qca8k-leds.o
+endif
diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c
index 20acbd0292f1..6d5ac7588a69 100644
--- a/drivers/net/dsa/qca/qca8k-8xxx.c
+++ b/drivers/net/dsa/qca/qca8k-8xxx.c
@@ -22,6 +22,7 @@
 #include <linux/dsa/tag_qca.h>
 
 #include "qca8k.h"
+#include "qca8k_leds.h"
 
 static void
 qca8k_split_addr(u32 regaddr, u16 *r1, u16 *r2, u16 *page)
@@ -1782,6 +1783,10 @@ qca8k_setup(struct dsa_switch *ds)
 	if (ret)
 		return ret;
 
+	ret = qca8k_setup_led_ctrl(priv);
+	if (ret)
+		return ret;
+
 	qca8k_setup_pcs(priv, &priv->pcs_port_0, 0);
 	qca8k_setup_pcs(priv, &priv->pcs_port_6, 6);
 
diff --git a/drivers/net/dsa/qca/qca8k-leds.c b/drivers/net/dsa/qca/qca8k-leds.c
new file mode 100644
index 000000000000..0146ee12d34c
--- /dev/null
+++ b/drivers/net/dsa/qca/qca8k-leds.c
@@ -0,0 +1,239 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/regmap.h>
+#include <net/dsa.h>
+
+#include "qca8k.h"
+#include "qca8k_leds.h"
+
+static int
+qca8k_get_enable_led_reg(int port_num, int led_num, struct qca8k_led_pattern_en *reg_info)
+{
+	switch (port_num) {
+	case 0:
+		reg_info->reg = QCA8K_LED_CTRL_REG(led_num);
+		reg_info->shift = QCA8K_LED_PHY0123_CONTROL_RULE_SHIFT;
+		break;
+	case 1:
+	case 2:
+	case 3:
+		/* Port 123 are controlled on a different reg */
+		reg_info->reg = QCA8K_LED_CTRL3_REG;
+		reg_info->shift = QCA8K_LED_PHY123_PATTERN_EN_SHIFT(port_num, led_num);
+		break;
+	case 4:
+		reg_info->reg = QCA8K_LED_CTRL_REG(led_num);
+		reg_info->shift = QCA8K_LED_PHY4_CONTROL_RULE_SHIFT;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+qca8k_led_brightness_set(struct qca8k_led *led,
+			 enum led_brightness brightness)
+{
+	struct qca8k_led_pattern_en reg_info;
+	struct qca8k_priv *priv = led->priv;
+	u32 mask, val;
+
+	qca8k_get_enable_led_reg(led->port_num, led->led_num, &reg_info);
+
+	val = QCA8K_LED_ALWAYS_OFF;
+	if (brightness)
+		val = QCA8K_LED_ALWAYS_ON;
+
+	/* HW regs to control brightness is special and port 1-2-3
+	 * are placed in a different reg.
+	 *
+	 * To control port 0 brightness:
+	 * - the 2 bit (15, 14) of:
+	 *   - QCA8K_LED_CTRL0_REG for led1
+	 *   - QCA8K_LED_CTRL1_REG for led2
+	 *   - QCA8K_LED_CTRL2_REG for led3
+	 *
+	 * To control port 4:
+	 * - the 2 bit (31, 30) of:
+	 *   - QCA8K_LED_CTRL0_REG for led1
+	 *   - QCA8K_LED_CTRL1_REG for led2
+	 *   - QCA8K_LED_CTRL2_REG for led3
+	 *
+	 * To control port 1:
+	 *   - the 2 bit at (9, 8) of QCA8K_LED_CTRL3_REG are used for led1
+	 *   - the 2 bit at (11, 10) of QCA8K_LED_CTRL3_REG are used for led2
+	 *   - the 2 bit at (13, 12) of QCA8K_LED_CTRL3_REG are used for led3
+	 *
+	 * To control port 2:
+	 *   - the 2 bit at (15, 14) of QCA8K_LED_CTRL3_REG are used for led1
+	 *   - the 2 bit at (17, 16) of QCA8K_LED_CTRL3_REG are used for led2
+	 *   - the 2 bit at (19, 18) of QCA8K_LED_CTRL3_REG are used for led3
+	 *
+	 * To control port 3:
+	 *   - the 2 bit at (21, 20) of QCA8K_LED_CTRL3_REG are used for led1
+	 *   - the 2 bit at (23, 22) of QCA8K_LED_CTRL3_REG are used for led2
+	 *   - the 2 bit at (25, 24) of QCA8K_LED_CTRL3_REG are used for led3
+	 *
+	 * To abstract this and have less code, we use the port and led numm
+	 * to calculate the shift and the correct reg due to this problem of
+	 * not having a 1:1 map of LED with the regs.
+	 */
+	if (led->port_num == 0 || led->port_num == 4) {
+		mask = QCA8K_LED_PATTERN_EN_MASK;
+		val <<= QCA8K_LED_PATTERN_EN_SHIFT;
+	} else {
+		mask = QCA8K_LED_PHY123_PATTERN_EN_MASK;
+	}
+
+	return regmap_update_bits(priv->regmap, reg_info.reg,
+				  mask << reg_info.shift,
+				  val << reg_info.shift);
+}
+
+static int
+qca8k_cled_brightness_set_blocking(struct led_classdev *ldev,
+				   enum led_brightness brightness)
+{
+	struct qca8k_led *led = container_of(ldev, struct qca8k_led, cdev);
+
+	return qca8k_led_brightness_set(led, brightness);
+}
+
+static enum led_brightness
+qca8k_led_brightness_get(struct qca8k_led *led)
+{
+	struct qca8k_led_pattern_en reg_info;
+	struct qca8k_priv *priv = led->priv;
+	u32 val;
+	int ret;
+
+	qca8k_get_enable_led_reg(led->port_num, led->led_num, &reg_info);
+
+	ret = regmap_read(priv->regmap, reg_info.reg, &val);
+	if (ret)
+		return 0;
+
+	val >>= reg_info.shift;
+
+	if (led->port_num == 0 || led->port_num == 4) {
+		val &= QCA8K_LED_PATTERN_EN_MASK;
+		val >>= QCA8K_LED_PATTERN_EN_SHIFT;
+	} else {
+		val &= QCA8K_LED_PHY123_PATTERN_EN_MASK;
+	}
+
+	/* Assume brightness ON only when the LED is set to always ON */
+	return val == QCA8K_LED_ALWAYS_ON;
+}
+
+static int
+qca8k_parse_port_leds(struct qca8k_priv *priv, struct fwnode_handle *port, int port_num)
+{
+	struct fwnode_handle *led = NULL, *leds = NULL;
+	struct led_init_data init_data = { };
+	struct dsa_switch *ds = priv->ds;
+	enum led_default_state state;
+	struct qca8k_led *port_led;
+	int led_num, led_index;
+	int ret;
+
+	leds = fwnode_get_named_child_node(port, "leds");
+	if (!leds) {
+		dev_dbg(priv->dev, "No Leds node specified in device tree for port %d!\n",
+			port_num);
+		return 0;
+	}
+
+	fwnode_for_each_child_node(leds, led) {
+		/* Reg represent the led number of the port.
+		 * Each port can have at most 3 leds attached
+		 * Commonly:
+		 * 1. is gigabit led
+		 * 2. is mbit led
+		 * 3. additional status led
+		 */
+		if (fwnode_property_read_u32(led, "reg", &led_num))
+			continue;
+
+		if (led_num >= QCA8K_LED_PORT_COUNT) {
+			dev_warn(priv->dev, "Invalid LED reg %d defined for port %d",
+				 led_num, port_num);
+			continue;
+		}
+
+		led_index = QCA8K_LED_PORT_INDEX(port_num, led_num);
+
+		port_led = &priv->ports_led[led_index];
+		port_led->port_num = port_num;
+		port_led->led_num = led_num;
+		port_led->priv = priv;
+
+		state = led_init_default_state_get(led);
+		switch (state) {
+		case LEDS_DEFSTATE_ON:
+			port_led->cdev.brightness = 1;
+			qca8k_led_brightness_set(port_led, 1);
+			break;
+		case LEDS_DEFSTATE_KEEP:
+			port_led->cdev.brightness =
+					qca8k_led_brightness_get(port_led);
+			break;
+		default:
+			port_led->cdev.brightness = 0;
+			qca8k_led_brightness_set(port_led, 0);
+		}
+
+		port_led->cdev.max_brightness = 1;
+		port_led->cdev.brightness_set_blocking = qca8k_cled_brightness_set_blocking;
+		init_data.default_label = ":port";
+		init_data.fwnode = led;
+		init_data.devname_mandatory = true;
+		init_data.devicename = kasprintf(GFP_KERNEL, "%s:0%d", ds->slave_mii_bus->id,
+						 port_num);
+		if (!init_data.devicename)
+			return -ENOMEM;
+
+		ret = devm_led_classdev_register_ext(priv->dev, &port_led->cdev, &init_data);
+		if (ret)
+			dev_warn(priv->dev, "Failed to init LED %d for port %d", led_num, port_num);
+
+		kfree(init_data.devicename);
+	}
+
+	return 0;
+}
+
+int
+qca8k_setup_led_ctrl(struct qca8k_priv *priv)
+{
+	struct fwnode_handle *ports, *port;
+	int port_num;
+	int ret;
+
+	ports = device_get_named_child_node(priv->dev, "ports");
+	if (!ports) {
+		dev_info(priv->dev, "No ports node specified in device tree!");
+		return 0;
+	}
+
+	fwnode_for_each_child_node(ports, port) {
+		if (fwnode_property_read_u32(port, "reg", &port_num))
+			continue;
+
+		/* Skip checking for CPU port 0 and CPU port 6 as not supported */
+		if (port_num == 0 || port_num == 6)
+			continue;
+
+		/* Each port can have at most 3 different leds attached.
+		 * Switch port starts from 0 to 6, but port 0 and 6 are CPU
+		 * port. The port index needs to be decreased by one to identify
+		 * the correct port for LED setup.
+		 */
+		ret = qca8k_parse_port_leds(priv, port, qca8k_port_to_phy(port_num));
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/dsa/qca/qca8k.h b/drivers/net/dsa/qca/qca8k.h
index dd7deb9095d3..c5cc8a172d65 100644
--- a/drivers/net/dsa/qca/qca8k.h
+++ b/drivers/net/dsa/qca/qca8k.h
@@ -11,6 +11,7 @@
 #include <linux/delay.h>
 #include <linux/regmap.h>
 #include <linux/gpio.h>
+#include <linux/leds.h>
 #include <linux/dsa/tag_qca.h>
 
 #define QCA8K_ETHERNET_MDIO_PRIORITY			7
@@ -85,6 +86,51 @@
 #define   QCA8K_MDIO_MASTER_DATA(x)			FIELD_PREP(QCA8K_MDIO_MASTER_DATA_MASK, x)
 #define   QCA8K_MDIO_MASTER_MAX_PORTS			5
 #define   QCA8K_MDIO_MASTER_MAX_REG			32
+
+/* LED control register */
+#define QCA8K_LED_PORT_COUNT				3
+#define QCA8K_LED_COUNT					((QCA8K_NUM_PORTS - QCA8K_NUM_CPU_PORTS) * QCA8K_LED_PORT_COUNT)
+#define QCA8K_LED_RULE_COUNT				6
+#define QCA8K_LED_RULE_MAX				11
+#define QCA8K_LED_PORT_INDEX(_phy, _led)		(((_phy) * QCA8K_LED_PORT_COUNT) + (_led))
+
+#define QCA8K_LED_PHY123_PATTERN_EN_SHIFT(_phy, _led)	((((_phy) - 1) * 6) + 8 + (2 * (_led)))
+#define QCA8K_LED_PHY123_PATTERN_EN_MASK		GENMASK(1, 0)
+
+#define QCA8K_LED_PHY0123_CONTROL_RULE_SHIFT		0
+#define QCA8K_LED_PHY4_CONTROL_RULE_SHIFT		16
+
+#define QCA8K_LED_CTRL_REG(_i)				(0x050 + (_i) * 4)
+#define QCA8K_LED_CTRL0_REG				0x50
+#define QCA8K_LED_CTRL1_REG				0x54
+#define QCA8K_LED_CTRL2_REG				0x58
+#define QCA8K_LED_CTRL3_REG				0x5C
+#define   QCA8K_LED_CTRL_SHIFT(_i)			(((_i) % 2) * 16)
+#define   QCA8K_LED_CTRL_MASK				GENMASK(15, 0)
+#define QCA8K_LED_RULE_MASK				GENMASK(13, 0)
+#define QCA8K_LED_BLINK_FREQ_MASK			GENMASK(1, 0)
+#define QCA8K_LED_BLINK_FREQ_SHITF			0
+#define   QCA8K_LED_BLINK_2HZ				0
+#define   QCA8K_LED_BLINK_4HZ				1
+#define   QCA8K_LED_BLINK_8HZ				2
+#define   QCA8K_LED_BLINK_AUTO				3
+#define QCA8K_LED_LINKUP_OVER_MASK			BIT(2)
+#define QCA8K_LED_TX_BLINK_MASK				BIT(4)
+#define QCA8K_LED_RX_BLINK_MASK				BIT(5)
+#define QCA8K_LED_COL_BLINK_MASK			BIT(7)
+#define QCA8K_LED_LINK_10M_EN_MASK			BIT(8)
+#define QCA8K_LED_LINK_100M_EN_MASK			BIT(9)
+#define QCA8K_LED_LINK_1000M_EN_MASK			BIT(10)
+#define QCA8K_LED_POWER_ON_LIGHT_MASK			BIT(11)
+#define QCA8K_LED_HALF_DUPLEX_MASK			BIT(12)
+#define QCA8K_LED_FULL_DUPLEX_MASK			BIT(13)
+#define QCA8K_LED_PATTERN_EN_MASK			GENMASK(15, 14)
+#define QCA8K_LED_PATTERN_EN_SHIFT			14
+#define   QCA8K_LED_ALWAYS_OFF				0
+#define   QCA8K_LED_ALWAYS_BLINK_4HZ			1
+#define   QCA8K_LED_ALWAYS_ON				2
+#define   QCA8K_LED_RULE_CONTROLLED			3
+
 #define QCA8K_GOL_MAC_ADDR0				0x60
 #define QCA8K_GOL_MAC_ADDR1				0x64
 #define QCA8K_MAX_FRAME_SIZE				0x78
@@ -382,6 +428,19 @@ struct qca8k_pcs {
 	int port;
 };
 
+struct qca8k_led_pattern_en {
+	u32 reg;
+	u8 shift;
+};
+
+struct qca8k_led {
+	u8 port_num;
+	u8 led_num;
+	u16 old_rule;
+	struct qca8k_priv *priv;
+	struct led_classdev cdev;
+};
+
 struct qca8k_priv {
 	u8 switch_id;
 	u8 switch_revision;
@@ -406,6 +465,7 @@ struct qca8k_priv {
 	struct qca8k_pcs pcs_port_0;
 	struct qca8k_pcs pcs_port_6;
 	const struct qca8k_match_data *info;
+	struct qca8k_led ports_led[QCA8K_LED_COUNT];
 };
 
 struct qca8k_mib_desc {
diff --git a/drivers/net/dsa/qca/qca8k_leds.h b/drivers/net/dsa/qca/qca8k_leds.h
new file mode 100644
index 000000000000..ab367f05b173
--- /dev/null
+++ b/drivers/net/dsa/qca/qca8k_leds.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef __QCA8K_LEDS_H
+#define __QCA8K_LEDS_H
+
+/* Leds Support function */
+#ifdef CONFIG_NET_DSA_QCA8K_LEDS_SUPPORT
+int qca8k_setup_led_ctrl(struct qca8k_priv *priv);
+#else
+static inline int qca8k_setup_led_ctrl(struct qca8k_priv *priv)
+{
+	return 0;
+}
+#endif
+
+#endif /* __QCA8K_LEDS_H */
-- 
cgit v1.2.3-70-g09d2


From 91acadcc6e599dfc62717abcdad58a459cfb1684 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:25 +0200
Subject: net: dsa: qca8k: add LEDs blink_set() support

Add LEDs blink_set() support to qca8k Switch Family.
These LEDs support hw accellerated blinking at a fixed rate
of 4Hz.

Reject any other value since not supported by the LEDs switch.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Pavel Machek <pavel@ucw.cz>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/qca8k-leds.c | 38 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/net/dsa/qca/qca8k-leds.c b/drivers/net/dsa/qca/qca8k-leds.c
index 0146ee12d34c..b883692b7d86 100644
--- a/drivers/net/dsa/qca/qca8k-leds.c
+++ b/drivers/net/dsa/qca/qca8k-leds.c
@@ -127,6 +127,43 @@ qca8k_led_brightness_get(struct qca8k_led *led)
 	return val == QCA8K_LED_ALWAYS_ON;
 }
 
+static int
+qca8k_cled_blink_set(struct led_classdev *ldev,
+		     unsigned long *delay_on,
+		     unsigned long *delay_off)
+{
+	struct qca8k_led *led = container_of(ldev, struct qca8k_led, cdev);
+	u32 mask, val = QCA8K_LED_ALWAYS_BLINK_4HZ;
+	struct qca8k_led_pattern_en reg_info;
+	struct qca8k_priv *priv = led->priv;
+
+	if (*delay_on == 0 && *delay_off == 0) {
+		*delay_on = 125;
+		*delay_off = 125;
+	}
+
+	if (*delay_on != 125 || *delay_off != 125) {
+		/* The hardware only supports blinking at 4Hz. Fall back
+		 * to software implementation in other cases.
+		 */
+		return -EINVAL;
+	}
+
+	qca8k_get_enable_led_reg(led->port_num, led->led_num, &reg_info);
+
+	if (led->port_num == 0 || led->port_num == 4) {
+		mask = QCA8K_LED_PATTERN_EN_MASK;
+		val <<= QCA8K_LED_PATTERN_EN_SHIFT;
+	} else {
+		mask = QCA8K_LED_PHY123_PATTERN_EN_MASK;
+	}
+
+	regmap_update_bits(priv->regmap, reg_info.reg, mask << reg_info.shift,
+			   val << reg_info.shift);
+
+	return 0;
+}
+
 static int
 qca8k_parse_port_leds(struct qca8k_priv *priv, struct fwnode_handle *port, int port_num)
 {
@@ -186,6 +223,7 @@ qca8k_parse_port_leds(struct qca8k_priv *priv, struct fwnode_handle *port, int p
 
 		port_led->cdev.max_brightness = 1;
 		port_led->cdev.brightness_set_blocking = qca8k_cled_brightness_set_blocking;
+		port_led->cdev.blink_set = qca8k_cled_blink_set;
 		init_data.default_label = ":port";
 		init_data.fwnode = led;
 		init_data.devname_mandatory = true;
-- 
cgit v1.2.3-70-g09d2


From e5029edd53937a29801ef507cee12e657ff31ea9 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:26 +0200
Subject: leds: Provide stubs for when CLASS_LED & NEW_LEDS are disabled

Provide stubs for devm_led_classdev_register_ext() and
led_init_default_state_get() so that LED drivers embedded within other
drivers such as PHYs and Ethernet switches still build when LEDS_CLASS
or NEW_LEDS are disabled. This also helps with Kconfig dependencies,
which are somewhat hairy for phylib and mdio and only get worse when
adding a dependency on LED_CLASS.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/leds.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/include/linux/leds.h b/include/linux/leds.h
index d71201a968b6..aa48e643f655 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -82,7 +82,15 @@ struct led_init_data {
 	bool devname_mandatory;
 };
 
+#if IS_ENABLED(CONFIG_NEW_LEDS)
 enum led_default_state led_init_default_state_get(struct fwnode_handle *fwnode);
+#else
+static inline enum led_default_state
+led_init_default_state_get(struct fwnode_handle *fwnode)
+{
+	return LEDS_DEFSTATE_OFF;
+}
+#endif
 
 struct led_hw_trigger_type {
 	int dummy;
@@ -217,9 +225,19 @@ static inline int led_classdev_register(struct device *parent,
 	return led_classdev_register_ext(parent, led_cdev, NULL);
 }
 
+#if IS_ENABLED(CONFIG_LEDS_CLASS)
 int devm_led_classdev_register_ext(struct device *parent,
 					  struct led_classdev *led_cdev,
 					  struct led_init_data *init_data);
+#else
+static inline int
+devm_led_classdev_register_ext(struct device *parent,
+			       struct led_classdev *led_cdev,
+			       struct led_init_data *init_data)
+{
+	return 0;
+}
+#endif
 
 static inline int devm_led_classdev_register(struct device *parent,
 					     struct led_classdev *led_cdev)
-- 
cgit v1.2.3-70-g09d2


From 01e5b728e9e43ae444e0369695a5f72209906464 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:27 +0200
Subject: net: phy: Add a binding for PHY LEDs

Define common binding parsing for all PHY drivers with LEDs using
phylib. Parse the DT as part of the phy_probe and add LEDs to the
linux LED class infrastructure. For the moment, provide a dummy
brightness function, which will later be replaced with a call into the
PHY driver. This allows testing since the LED core might otherwise
reject an LED whose brightness cannot be set.

Add a dependency on LED_CLASS. It either needs to be built in, or not
enabled, since a modular build can result in linker errors.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig      |  1 +
 drivers/net/phy/phy_device.c | 76 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/phy.h          | 16 ++++++++++
 3 files changed, 93 insertions(+)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 6b9525def973..b8cc49820ced 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -18,6 +18,7 @@ menuconfig PHYLIB
 	depends on NETDEVICES
 	select MDIO_DEVICE
 	select MDIO_DEVRES
+	depends on LEDS_CLASS || LEDS_CLASS=n
 	help
 	  Ethernet controllers are usually attached to PHY
 	  devices.  This option provides infrastructure for
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 917ba84105fc..61b971251de5 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -19,10 +19,12 @@
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/mdio.h>
 #include <linux/mii.h>
 #include <linux/mm.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/netdevice.h>
 #include <linux/phy.h>
 #include <linux/phy_led_triggers.h>
@@ -674,6 +676,7 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, u32 phy_id,
 	device_initialize(&mdiodev->dev);
 
 	dev->state = PHY_DOWN;
+	INIT_LIST_HEAD(&dev->leds);
 
 	mutex_init(&dev->lock);
 	INIT_DELAYED_WORK(&dev->state_queue, phy_state_machine);
@@ -2988,6 +2991,74 @@ static bool phy_drv_supports_irq(struct phy_driver *phydrv)
 	return phydrv->config_intr && phydrv->handle_interrupt;
 }
 
+/* Dummy implementation until calls into PHY driver are added */
+static int phy_led_set_brightness(struct led_classdev *led_cdev,
+				  enum led_brightness value)
+{
+	return 0;
+}
+
+static int of_phy_led(struct phy_device *phydev,
+		      struct device_node *led)
+{
+	struct device *dev = &phydev->mdio.dev;
+	struct led_init_data init_data = {};
+	struct led_classdev *cdev;
+	struct phy_led *phyled;
+	int err;
+
+	phyled = devm_kzalloc(dev, sizeof(*phyled), GFP_KERNEL);
+	if (!phyled)
+		return -ENOMEM;
+
+	cdev = &phyled->led_cdev;
+
+	err = of_property_read_u8(led, "reg", &phyled->index);
+	if (err)
+		return err;
+
+	cdev->brightness_set_blocking = phy_led_set_brightness;
+	cdev->max_brightness = 1;
+	init_data.devicename = dev_name(&phydev->mdio.dev);
+	init_data.fwnode = of_fwnode_handle(led);
+	init_data.devname_mandatory = true;
+
+	err = devm_led_classdev_register_ext(dev, cdev, &init_data);
+	if (err)
+		return err;
+
+	list_add(&phyled->list, &phydev->leds);
+
+	return 0;
+}
+
+static int of_phy_leds(struct phy_device *phydev)
+{
+	struct device_node *node = phydev->mdio.dev.of_node;
+	struct device_node *leds, *led;
+	int err;
+
+	if (!IS_ENABLED(CONFIG_OF_MDIO))
+		return 0;
+
+	if (!node)
+		return 0;
+
+	leds = of_get_child_by_name(node, "leds");
+	if (!leds)
+		return 0;
+
+	for_each_available_child_of_node(leds, led) {
+		err = of_phy_led(phydev, led);
+		if (err) {
+			of_node_put(led);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
 /**
  * fwnode_mdio_find_device - Given a fwnode, find the mdio_device
  * @fwnode: pointer to the mdio_device's fwnode
@@ -3183,6 +3254,11 @@ static int phy_probe(struct device *dev)
 	/* Set the state to READY by default */
 	phydev->state = PHY_READY;
 
+	/* Get the LEDs from the device tree, and instantiate standard
+	 * LEDs for them.
+	 */
+	err = of_phy_leds(phydev);
+
 out:
 	/* Re-assert the reset signal on error */
 	if (err)
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 2f83cfc206e5..bd6b5e9bb729 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -14,6 +14,7 @@
 #include <linux/compiler.h>
 #include <linux/spinlock.h>
 #include <linux/ethtool.h>
+#include <linux/leds.h>
 #include <linux/linkmode.h>
 #include <linux/netlink.h>
 #include <linux/mdio.h>
@@ -600,6 +601,7 @@ struct macsec_ops;
  * @phy_num_led_triggers: Number of triggers in @phy_led_triggers
  * @led_link_trigger: LED trigger for link up/down
  * @last_triggered: last LED trigger for link speed
+ * @leds: list of PHY LED structures
  * @master_slave_set: User requested master/slave configuration
  * @master_slave_get: Current master/slave advertisement
  * @master_slave_state: Current master/slave configuration
@@ -699,6 +701,7 @@ struct phy_device {
 
 	struct phy_led_trigger *led_link_trigger;
 #endif
+	struct list_head leds;
 
 	/*
 	 * Interrupt number for this PHY
@@ -834,6 +837,19 @@ struct phy_plca_status {
 	bool pst;
 };
 
+/**
+ * struct phy_led: An LED driven by the PHY
+ *
+ * @list: List of LEDs
+ * @led_cdev: Standard LED class structure
+ * @index: Number of the LED
+ */
+struct phy_led {
+	struct list_head list;
+	struct led_classdev led_cdev;
+	u8 index;
+};
+
 /**
  * struct phy_driver - Driver structure for a particular PHY type
  *
-- 
cgit v1.2.3-70-g09d2


From 684818189b04b095b34964ed4a3ea5249a840eab Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:28 +0200
Subject: net: phy: phy_device: Call into the PHY driver to set LED brightness

Linux LEDs can be software controlled via the brightness file in /sys.
LED drivers need to implement a brightness_set function which the core
will call. Implement an intermediary in phy_device, which will call
into the phy driver if it implements the necessary function.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy_device.c | 15 ++++++++++++---
 include/linux/phy.h          | 13 +++++++++++++
 2 files changed, 25 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 61b971251de5..5c1200160c51 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -2991,11 +2991,18 @@ static bool phy_drv_supports_irq(struct phy_driver *phydrv)
 	return phydrv->config_intr && phydrv->handle_interrupt;
 }
 
-/* Dummy implementation until calls into PHY driver are added */
 static int phy_led_set_brightness(struct led_classdev *led_cdev,
 				  enum led_brightness value)
 {
-	return 0;
+	struct phy_led *phyled = to_phy_led(led_cdev);
+	struct phy_device *phydev = phyled->phydev;
+	int err;
+
+	mutex_lock(&phydev->lock);
+	err = phydev->drv->led_brightness_set(phydev, phyled->index, value);
+	mutex_unlock(&phydev->lock);
+
+	return err;
 }
 
 static int of_phy_led(struct phy_device *phydev,
@@ -3012,12 +3019,14 @@ static int of_phy_led(struct phy_device *phydev,
 		return -ENOMEM;
 
 	cdev = &phyled->led_cdev;
+	phyled->phydev = phydev;
 
 	err = of_property_read_u8(led, "reg", &phyled->index);
 	if (err)
 		return err;
 
-	cdev->brightness_set_blocking = phy_led_set_brightness;
+	if (phydev->drv->led_brightness_set)
+		cdev->brightness_set_blocking = phy_led_set_brightness;
 	cdev->max_brightness = 1;
 	init_data.devicename = dev_name(&phydev->mdio.dev);
 	init_data.fwnode = of_fwnode_handle(led);
diff --git a/include/linux/phy.h b/include/linux/phy.h
index bd6b5e9bb729..f3c7e3c99f24 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -841,15 +841,19 @@ struct phy_plca_status {
  * struct phy_led: An LED driven by the PHY
  *
  * @list: List of LEDs
+ * @phydev: PHY this LED is attached to
  * @led_cdev: Standard LED class structure
  * @index: Number of the LED
  */
 struct phy_led {
 	struct list_head list;
+	struct phy_device *phydev;
 	struct led_classdev led_cdev;
 	u8 index;
 };
 
+#define to_phy_led(d) container_of(d, struct phy_led, led_cdev)
+
 /**
  * struct phy_driver - Driver structure for a particular PHY type
  *
@@ -1072,6 +1076,15 @@ struct phy_driver {
 	/** @get_plca_status: Return the current PLCA status info */
 	int (*get_plca_status)(struct phy_device *dev,
 			       struct phy_plca_status *plca_st);
+
+	/**
+	 * @led_brightness_set: Set a PHY LED brightness. Index
+	 * indicates which of the PHYs led should be set. Value
+	 * follows the standard LED class meaning, e.g. LED_OFF,
+	 * LED_HALF, LED_FULL.
+	 */
+	int (*led_brightness_set)(struct phy_device *dev,
+				  u8 index, enum led_brightness value);
 };
 #define to_phy_driver(d) container_of(to_mdio_common_driver(d),		\
 				      struct phy_driver, mdiodrv)
-- 
cgit v1.2.3-70-g09d2


From 2d3960e58ef7c83fe1dbf952f056b9e906cb6df8 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:29 +0200
Subject: net: phy: marvell: Add software control of the LEDs

Add a brightness function, so the LEDs can be controlled from
software using the standard Linux LED infrastructure.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 45 ++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 40 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 24853e9a889e..8d1eb4c4e1ab 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -144,11 +144,13 @@
 /* WOL Event Interrupt Enable */
 #define MII_88E1318S_PHY_CSIER_WOL_EIE			BIT(7)
 
-/* LED Timer Control Register */
-#define MII_88E1318S_PHY_LED_TCR			0x12
-#define MII_88E1318S_PHY_LED_TCR_FORCE_INT		BIT(15)
-#define MII_88E1318S_PHY_LED_TCR_INTn_ENABLE		BIT(7)
-#define MII_88E1318S_PHY_LED_TCR_INT_ACTIVE_LOW		BIT(11)
+#define MII_88E1318S_PHY_LED_FUNC		0x10
+#define MII_88E1318S_PHY_LED_FUNC_OFF		(0x8)
+#define MII_88E1318S_PHY_LED_FUNC_ON		(0x9)
+#define MII_88E1318S_PHY_LED_TCR		0x12
+#define MII_88E1318S_PHY_LED_TCR_FORCE_INT	BIT(15)
+#define MII_88E1318S_PHY_LED_TCR_INTn_ENABLE	BIT(7)
+#define MII_88E1318S_PHY_LED_TCR_INT_ACTIVE_LOW	BIT(11)
 
 /* Magic Packet MAC address registers */
 #define MII_88E1318S_PHY_MAGIC_PACKET_WORD2		0x17
@@ -2832,6 +2834,34 @@ static int marvell_hwmon_probe(struct phy_device *phydev)
 }
 #endif
 
+static int m88e1318_led_brightness_set(struct phy_device *phydev,
+				       u8 index, enum led_brightness value)
+{
+	int reg;
+
+	reg = phy_read_paged(phydev, MII_MARVELL_LED_PAGE,
+			     MII_88E1318S_PHY_LED_FUNC);
+	if (reg < 0)
+		return reg;
+
+	switch (index) {
+	case 0:
+	case 1:
+	case 2:
+		reg &= ~(0xf << (4 * index));
+		if (value == LED_OFF)
+			reg |= MII_88E1318S_PHY_LED_FUNC_OFF << (4 * index);
+		else
+			reg |= MII_88E1318S_PHY_LED_FUNC_ON << (4 * index);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return phy_write_paged(phydev, MII_MARVELL_LED_PAGE,
+			       MII_88E1318S_PHY_LED_FUNC, reg);
+}
+
 static int marvell_probe(struct phy_device *phydev)
 {
 	struct marvell_priv *priv;
@@ -3081,6 +3111,7 @@ static struct phy_driver marvell_drivers[] = {
 		.get_sset_count = marvell_get_sset_count,
 		.get_strings = marvell_get_strings,
 		.get_stats = marvell_get_stats,
+		.led_brightness_set = m88e1318_led_brightness_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1145,
@@ -3187,6 +3218,7 @@ static struct phy_driver marvell_drivers[] = {
 		.cable_test_start = marvell_vct7_cable_test_start,
 		.cable_test_tdr_start = marvell_vct5_cable_test_tdr_start,
 		.cable_test_get_status = marvell_vct7_cable_test_get_status,
+		.led_brightness_set = m88e1318_led_brightness_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1540,
@@ -3213,6 +3245,7 @@ static struct phy_driver marvell_drivers[] = {
 		.cable_test_start = marvell_vct7_cable_test_start,
 		.cable_test_tdr_start = marvell_vct5_cable_test_tdr_start,
 		.cable_test_get_status = marvell_vct7_cable_test_get_status,
+		.led_brightness_set = m88e1318_led_brightness_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1545,
@@ -3239,6 +3272,7 @@ static struct phy_driver marvell_drivers[] = {
 		.cable_test_start = marvell_vct7_cable_test_start,
 		.cable_test_tdr_start = marvell_vct5_cable_test_tdr_start,
 		.cable_test_get_status = marvell_vct7_cable_test_get_status,
+		.led_brightness_set = m88e1318_led_brightness_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E3016,
@@ -3380,6 +3414,7 @@ static struct phy_driver marvell_drivers[] = {
 		.get_stats = marvell_get_stats,
 		.get_tunable = m88e1540_get_tunable,
 		.set_tunable = m88e1540_set_tunable,
+		.led_brightness_set = m88e1318_led_brightness_set,
 	},
 };
 
-- 
cgit v1.2.3-70-g09d2


From 4e901018432e38eab35d2a352661ce4727795be1 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:30 +0200
Subject: net: phy: phy_device: Call into the PHY driver to set LED blinking

Linux LEDs can be requested to perform hardware accelerated
blinking. Pass this to the PHY driver, if it implements the op.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/phy_device.c | 18 ++++++++++++++++++
 include/linux/phy.h          | 12 ++++++++++++
 2 files changed, 30 insertions(+)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 5c1200160c51..538523a7cd51 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -3005,6 +3005,22 @@ static int phy_led_set_brightness(struct led_classdev *led_cdev,
 	return err;
 }
 
+static int phy_led_blink_set(struct led_classdev *led_cdev,
+			     unsigned long *delay_on,
+			     unsigned long *delay_off)
+{
+	struct phy_led *phyled = to_phy_led(led_cdev);
+	struct phy_device *phydev = phyled->phydev;
+	int err;
+
+	mutex_lock(&phydev->lock);
+	err = phydev->drv->led_blink_set(phydev, phyled->index,
+					 delay_on, delay_off);
+	mutex_unlock(&phydev->lock);
+
+	return err;
+}
+
 static int of_phy_led(struct phy_device *phydev,
 		      struct device_node *led)
 {
@@ -3027,6 +3043,8 @@ static int of_phy_led(struct phy_device *phydev,
 
 	if (phydev->drv->led_brightness_set)
 		cdev->brightness_set_blocking = phy_led_set_brightness;
+	if (phydev->drv->led_blink_set)
+		cdev->blink_set = phy_led_blink_set;
 	cdev->max_brightness = 1;
 	init_data.devicename = dev_name(&phydev->mdio.dev);
 	init_data.fwnode = of_fwnode_handle(led);
diff --git a/include/linux/phy.h b/include/linux/phy.h
index f3c7e3c99f24..c5a0dc829714 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1085,6 +1085,18 @@ struct phy_driver {
 	 */
 	int (*led_brightness_set)(struct phy_device *dev,
 				  u8 index, enum led_brightness value);
+
+	/**
+	 * @led_blink_set: Set a PHY LED brightness.  Index indicates
+	 * which of the PHYs led should be configured to blink. Delays
+	 * are in milliseconds and if both are zero then a sensible
+	 * default should be chosen.  The call should adjust the
+	 * timings in that case and if it can't match the values
+	 * specified exactly.
+	 */
+	int (*led_blink_set)(struct phy_device *dev, u8 index,
+			     unsigned long *delay_on,
+			     unsigned long *delay_off);
 };
 #define to_phy_driver(d) container_of(to_mdio_common_driver(d),		\
 				      struct phy_driver, mdiodrv)
-- 
cgit v1.2.3-70-g09d2


From ea9e86485decb2ac1750005bd96c166c9b780406 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:31 +0200
Subject: net: phy: marvell: Implement led_blink_set()

The Marvell PHY can blink the LEDs, simple on/off. All LEDs blink at
the same rate, and the reset default is 84ms per blink, which is
around 12Hz.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 8d1eb4c4e1ab..cd5d0ed9c5e5 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -147,6 +147,8 @@
 #define MII_88E1318S_PHY_LED_FUNC		0x10
 #define MII_88E1318S_PHY_LED_FUNC_OFF		(0x8)
 #define MII_88E1318S_PHY_LED_FUNC_ON		(0x9)
+#define MII_88E1318S_PHY_LED_FUNC_HI_Z		(0xa)
+#define MII_88E1318S_PHY_LED_FUNC_BLINK		(0xb)
 #define MII_88E1318S_PHY_LED_TCR		0x12
 #define MII_88E1318S_PHY_LED_TCR_FORCE_INT	BIT(15)
 #define MII_88E1318S_PHY_LED_TCR_INTn_ENABLE	BIT(7)
@@ -2862,6 +2864,35 @@ static int m88e1318_led_brightness_set(struct phy_device *phydev,
 			       MII_88E1318S_PHY_LED_FUNC, reg);
 }
 
+static int m88e1318_led_blink_set(struct phy_device *phydev, u8 index,
+				  unsigned long *delay_on,
+				  unsigned long *delay_off)
+{
+	int reg;
+
+	reg = phy_read_paged(phydev, MII_MARVELL_LED_PAGE,
+			     MII_88E1318S_PHY_LED_FUNC);
+	if (reg < 0)
+		return reg;
+
+	switch (index) {
+	case 0:
+	case 1:
+	case 2:
+		reg &= ~(0xf << (4 * index));
+			reg |= MII_88E1318S_PHY_LED_FUNC_BLINK << (4 * index);
+			/* Reset default is 84ms */
+			*delay_on = 84 / 2;
+			*delay_off = 84 / 2;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return phy_write_paged(phydev, MII_MARVELL_LED_PAGE,
+			       MII_88E1318S_PHY_LED_FUNC, reg);
+}
+
 static int marvell_probe(struct phy_device *phydev)
 {
 	struct marvell_priv *priv;
@@ -3112,6 +3143,7 @@ static struct phy_driver marvell_drivers[] = {
 		.get_strings = marvell_get_strings,
 		.get_stats = marvell_get_stats,
 		.led_brightness_set = m88e1318_led_brightness_set,
+		.led_blink_set = m88e1318_led_blink_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1145,
@@ -3219,6 +3251,7 @@ static struct phy_driver marvell_drivers[] = {
 		.cable_test_tdr_start = marvell_vct5_cable_test_tdr_start,
 		.cable_test_get_status = marvell_vct7_cable_test_get_status,
 		.led_brightness_set = m88e1318_led_brightness_set,
+		.led_blink_set = m88e1318_led_blink_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1540,
@@ -3246,6 +3279,7 @@ static struct phy_driver marvell_drivers[] = {
 		.cable_test_tdr_start = marvell_vct5_cable_test_tdr_start,
 		.cable_test_get_status = marvell_vct7_cable_test_get_status,
 		.led_brightness_set = m88e1318_led_brightness_set,
+		.led_blink_set = m88e1318_led_blink_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E1545,
@@ -3273,6 +3307,7 @@ static struct phy_driver marvell_drivers[] = {
 		.cable_test_tdr_start = marvell_vct5_cable_test_tdr_start,
 		.cable_test_get_status = marvell_vct7_cable_test_get_status,
 		.led_brightness_set = m88e1318_led_brightness_set,
+		.led_blink_set = m88e1318_led_blink_set,
 	},
 	{
 		.phy_id = MARVELL_PHY_ID_88E3016,
@@ -3415,6 +3450,7 @@ static struct phy_driver marvell_drivers[] = {
 		.get_tunable = m88e1540_get_tunable,
 		.set_tunable = m88e1540_set_tunable,
 		.led_brightness_set = m88e1318_led_brightness_set,
+		.led_blink_set = m88e1318_led_blink_set,
 	},
 };
 
-- 
cgit v1.2.3-70-g09d2


From 57b6c752c5c05a1f38eb6da9f1618f1b000e1f51 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:32 +0200
Subject: dt-bindings: net: ethernet-controller: Document support for LEDs node

Document support for LEDs node in ethernet-controller.
Ethernet Controller may support different LEDs that can be configured
for different operation like blinking on traffic event or port link.

Also add some Documentation to describe the difference of these nodes
compared to PHY LEDs, since ethernet-controller LEDs are controllable
by the ethernet controller regs and the possible intergated PHY doesn't
have control on them.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../bindings/net/ethernet-controller.yaml          | 35 ++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/ethernet-controller.yaml b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
index 00be387984ac..ebc2646ab5ff 100644
--- a/Documentation/devicetree/bindings/net/ethernet-controller.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
@@ -222,6 +222,41 @@ properties:
         required:
           - speed
 
+  leds:
+    description:
+      Describes the LEDs associated by Ethernet Controller.
+      These LEDs are not integrated in the PHY and PHY doesn't have any
+      control on them. Ethernet Controller regs are used to control
+      these defined LEDs.
+
+    type: object
+
+    properties:
+      '#address-cells':
+        const: 1
+
+      '#size-cells':
+        const: 0
+
+    patternProperties:
+      '^led@[a-f0-9]+$':
+        $ref: /schemas/leds/common.yaml#
+
+        properties:
+          reg:
+            maxItems: 1
+            description:
+              This define the LED index in the PHY or the MAC. It's really
+              driver dependent and required for ports that define multiple
+              LED for the same port.
+
+        required:
+          - reg
+
+        unevaluatedProperties: false
+
+    additionalProperties: false
+
 dependencies:
   pcs-handle-names: [pcs-handle]
 
-- 
cgit v1.2.3-70-g09d2


From ed617bc022f4596dc03815e8f0331e60644d07ee Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:33 +0200
Subject: dt-bindings: net: dsa: qca8k: add LEDs definition example

Add LEDs definition example for qca8k Switch Family to describe how they
should be defined for a correct usage.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../devicetree/bindings/net/dsa/qca8k.yaml         | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/dsa/qca8k.yaml b/Documentation/devicetree/bindings/net/dsa/qca8k.yaml
index fe9ebe285938..df64eebebe18 100644
--- a/Documentation/devicetree/bindings/net/dsa/qca8k.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/qca8k.yaml
@@ -18,6 +18,8 @@ description:
   PHY it is connected to. In this config, an internal mdio-bus is registered and
   the MDIO master is used for communication. Mixed external and internal
   mdio-bus configurations are not supported by the hardware.
+  Each phy has at most 3 LEDs connected and can be declared
+  using the standard LEDs structure.
 
 properties:
   compatible:
@@ -117,6 +119,7 @@ unevaluatedProperties: false
 examples:
   - |
     #include <dt-bindings/gpio/gpio.h>
+    #include <dt-bindings/leds/common.h>
 
     mdio {
         #address-cells = <1>;
@@ -226,6 +229,25 @@ examples:
                     label = "lan1";
                     phy-mode = "internal";
                     phy-handle = <&internal_phy_port1>;
+
+                    leds {
+                        #address-cells = <1>;
+                        #size-cells = <0>;
+
+                        led@0 {
+                            reg = <0>;
+                            color = <LED_COLOR_ID_WHITE>;
+                            function = LED_FUNCTION_LAN;
+                            default-state = "keep";
+                        };
+
+                        led@1 {
+                            reg = <1>;
+                            color = <LED_COLOR_ID_AMBER>;
+                            function = LED_FUNCTION_LAN;
+                            default-state = "keep";
+                        };
+                    };
                 };
 
                 port@2 {
-- 
cgit v1.2.3-70-g09d2


From 939595c79d12bb18e3784bc3faa2ddd2df41f70f Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:34 +0200
Subject: ARM: dts: qcom: ipq8064-rb3011: Drop unevaluated properties in switch
 nodes

IPQ8064 MikroTik RB3011UiAS-RM DT have currently unevaluted properties
in the 2 switch nodes. The bindings #address-cells and #size-cells are
redundant and cause warning for 'Unevaluated properties are not
allowed'.

Drop these bindings to mute these warning as they should not be there
from the start.

Cc: Jonathan McDowell <noodles@earth.li>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Jonathan McDowell <noodles@earth.li>
Tested-by: Jonathan McDowell <noodles@earth.li>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/arm/boot/dts/qcom-ipq8064-rb3011.dts | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts b/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts
index f908889c4f95..47a5d1849c72 100644
--- a/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts
+++ b/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts
@@ -38,8 +38,6 @@
 
 		switch0: switch@10 {
 			compatible = "qca,qca8337";
-			#address-cells = <1>;
-			#size-cells = <0>;
 
 			dsa,member = <0 0>;
 
@@ -105,8 +103,6 @@
 
 		switch1: switch@14 {
 			compatible = "qca,qca8337";
-			#address-cells = <1>;
-			#size-cells = <0>;
 
 			dsa,member = <1 0>;
 
-- 
cgit v1.2.3-70-g09d2


From 09930f1fb8750d5f4e668ab40733ab63cab58ed7 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:35 +0200
Subject: ARM: dts: qcom: ipq8064-rb3011: Add Switch LED for each port

Add Switch LED for each port for MikroTik RB3011UiAS-RM.

MikroTik RB3011UiAS-RM is a 10 port device with 2 qca8337 switch chips
connected.

It was discovered that in the hardware design all 3 Switch LED trace of
the related port is connected to the same LED. This was discovered by
setting to 'always on' the related led in the switch regs and noticing
that all 3 LED for the specific port (for example for port 1) cause the
connected LED for port 1 to turn on. As an extra test we tried enabling
2 different LED for the port resulting in the LED turned off only if
every led in the reg was off.

Aside from this funny and strange hardware implementation, the device
itself have one green LED for each port, resulting in 10 green LED one
for each of the 10 supported port.

Cc: Jonathan McDowell <noodles@earth.li>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/arm/boot/dts/qcom-ipq8064-rb3011.dts | 120 ++++++++++++++++++++++++++++++
 1 file changed, 120 insertions(+)

diff --git a/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts b/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts
index 47a5d1849c72..4d509876294b 100644
--- a/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts
+++ b/arch/arm/boot/dts/qcom-ipq8064-rb3011.dts
@@ -65,26 +65,86 @@
 				port@1 {
 					reg = <1>;
 					label = "sw1";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@2 {
 					reg = <2>;
 					label = "sw2";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@3 {
 					reg = <3>;
 					label = "sw3";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@4 {
 					reg = <4>;
 					label = "sw4";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@5 {
 					reg = <5>;
 					label = "sw5";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 			};
 		};
@@ -130,26 +190,86 @@
 				port@1 {
 					reg = <1>;
 					label = "sw6";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@2 {
 					reg = <2>;
 					label = "sw7";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@3 {
 					reg = <3>;
 					label = "sw8";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@4 {
 					reg = <4>;
 					label = "sw9";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 
 				port@5 {
 					reg = <5>;
 					label = "sw10";
+
+					leds {
+						#address-cells = <1>;
+						#size-cells = <0>;
+
+						led@0 {
+							reg = <0>;
+							color = <LED_COLOR_ID_GREEN>;
+							function = LED_FUNCTION_LAN;
+							default-state = "keep";
+						};
+					};
 				};
 			};
 		};
-- 
cgit v1.2.3-70-g09d2


From 18a24b694a2bfd59ae8a17ffee87f12c65d9f2a6 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Mon, 17 Apr 2023 17:17:36 +0200
Subject: dt-bindings: net: phy: Document support for LEDs node

Document support for LEDs node in phy and add an example for it.
PHY LED will have to match led pattern and should be treated as a
generic led.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../devicetree/bindings/net/ethernet-phy.yaml      | 43 ++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/ethernet-phy.yaml b/Documentation/devicetree/bindings/net/ethernet-phy.yaml
index ac04f8efa35c..4f574532ee13 100644
--- a/Documentation/devicetree/bindings/net/ethernet-phy.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-phy.yaml
@@ -197,6 +197,35 @@ properties:
       PHY's that have configurable TX internal delays. If this property is
       present then the PHY applies the TX delay.
 
+  leds:
+    type: object
+
+    properties:
+      '#address-cells':
+        const: 1
+
+      '#size-cells':
+        const: 0
+
+    patternProperties:
+      '^led@[a-f0-9]+$':
+        $ref: /schemas/leds/common.yaml#
+
+        properties:
+          reg:
+            maxItems: 1
+            description:
+              This define the LED index in the PHY or the MAC. It's really
+              driver dependent and required for ports that define multiple
+              LED for the same port.
+
+        required:
+          - reg
+
+        unevaluatedProperties: false
+
+    additionalProperties: false
+
 required:
   - reg
 
@@ -204,6 +233,8 @@ additionalProperties: true
 
 examples:
   - |
+    #include <dt-bindings/leds/common.h>
+
     ethernet {
         #address-cells = <1>;
         #size-cells = <0>;
@@ -219,5 +250,17 @@ examples:
             reset-gpios = <&gpio1 4 1>;
             reset-assert-us = <1000>;
             reset-deassert-us = <2000>;
+
+            leds {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                led@0 {
+                    reg = <0>;
+                    color = <LED_COLOR_ID_WHITE>;
+                    function = LED_FUNCTION_LAN;
+                    default-state = "keep";
+                };
+            };
         };
     };
-- 
cgit v1.2.3-70-g09d2


From 380a8fe1b2f49f262673485d1c7f1c0da170a926 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:37 +0200
Subject: arm: mvebu: dt: Add PHY LED support for 370-rd WAN port

The WAN port of the 370-RD has a Marvell PHY, with one LED on
the front panel.y List this LED in the device tree.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/arm/boot/dts/armada-370-rd.dts | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/arch/arm/boot/dts/armada-370-rd.dts b/arch/arm/boot/dts/armada-370-rd.dts
index be005c9f42ef..2586f32a3e21 100644
--- a/arch/arm/boot/dts/armada-370-rd.dts
+++ b/arch/arm/boot/dts/armada-370-rd.dts
@@ -20,6 +20,7 @@
 /dts-v1/;
 #include <dt-bindings/input/input.h>
 #include <dt-bindings/interrupt-controller/irq.h>
+#include <dt-bindings/leds/common.h>
 #include <dt-bindings/gpio/gpio.h>
 #include "armada-370.dtsi"
 
@@ -135,6 +136,17 @@
 	pinctrl-names = "default";
 	phy0: ethernet-phy@0 {
 		reg = <0>;
+		leds {
+			#address-cells = <1>;
+			#size-cells = <0>;
+
+			led@0 {
+				reg = <0>;
+				color = <LED_COLOR_ID_WHITE>;
+				function = LED_FUNCTION_WAN;
+				default-state = "keep";
+			};
+		};
 	};
 
 	switch: switch@10 {
-- 
cgit v1.2.3-70-g09d2


From c693ea2fd6e3b43f6caaf618f864236c36a71f70 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Mon, 17 Apr 2023 17:17:38 +0200
Subject: Documentation: LEDs: Describe good names for network LEDs

Network LEDs can exist in both the MAC and the PHY. Naming is
difficult because the netdev name is neither stable or unique, do to
commands like ip link set name eth42 dev eth0, and network
namesspaces.

Give some example names where the MAC and the PHY have unique names
based on device tree nodes, or PCI bus addresses.

Since the LED can be used for anything which Linux supports for LEDs,
avoid using names like activity or link, rather describe the location
on the RJ-45, of what the RJ-45 is expected to be used for, WAN/LAN
etc.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/leds/well-known-leds.txt | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/Documentation/leds/well-known-leds.txt b/Documentation/leds/well-known-leds.txt
index 2160382c86be..e9c30dc75884 100644
--- a/Documentation/leds/well-known-leds.txt
+++ b/Documentation/leds/well-known-leds.txt
@@ -70,3 +70,33 @@ Good: "platform:*:charging" (allwinner sun50i)
 * Screen
 
 Good: ":backlight" (Motorola Droid 4)
+
+* Ethernet LEDs
+
+Currently two types of Network LEDs are support, those controlled by
+the PHY and those by the MAC. In theory both can be present at the
+same time for one Linux netdev, hence the names need to differ between
+MAC and PHY.
+
+Do not use the netdev name, such as eth0, enp1s0. These are not stable
+and are not unique. They also don't differentiate between MAC and PHY.
+
+** MAC LEDs
+
+Good: f1070000.ethernet:white:WAN
+Good: mdio_mux-0.1:00:green:left
+Good: 0000:02:00.0:yellow:top
+
+The first part must uniquely name the MAC controller. Then follows the
+colour.  WAN/LAN should be used for a single LED. If there are
+multiple LEDs, use left/right, or top/bottom to indicate their
+position on the RJ45 socket.
+
+** PHY LEDs
+
+Good: f1072004.mdio-mii:00: white:WAN
+Good: !mdio-mux!mdio@2!switch@0!mdio:01:green:right
+Good: r8169-0-200:00:yellow:bottom
+
+The first part must uniquely name the PHY. This often means uniquely
+identifying the MDIO bus controller, and the address on the bus.
-- 
cgit v1.2.3-70-g09d2


From eb6fba7555a812c07aa984fb9e8e9b151a65ca16 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 17 Apr 2023 08:53:46 -0700
Subject: net: skbuff: hide wifi_acked when CONFIG_WIRELESS not set

Datacenter kernel builds will very likely not include WIRELESS,
so let them shave 2 bits off the skb by hiding the wifi fields.

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 11 +++++++++++
 include/net/sock.h     |  2 +-
 net/core/skbuff.c      |  2 ++
 net/socket.c           |  2 ++
 4 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index a823ec3aa326..513f03b23a73 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -953,8 +953,10 @@ struct sk_buff {
 
 	__u8			l4_hash:1;
 	__u8			sw_hash:1;
+#ifdef CONFIG_WIRELESS
 	__u8			wifi_acked_valid:1;
 	__u8			wifi_acked:1;
+#endif
 	__u8			no_fcs:1;
 	/* Indicates the inner headers are valid in the skbuff. */
 	__u8			encapsulation:1;
@@ -1187,6 +1189,15 @@ static inline unsigned int skb_napi_id(const struct sk_buff *skb)
 #endif
 }
 
+static inline bool skb_wifi_acked_valid(const struct sk_buff *skb)
+{
+#ifdef CONFIG_WIRELESS
+	return skb->wifi_acked_valid;
+#else
+	return 0;
+#endif
+}
+
 /**
  * skb_unref - decrement the skb's reference count
  * @skb: buffer
diff --git a/include/net/sock.h b/include/net/sock.h
index 5edf0038867c..8b7ed7167243 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2697,7 +2697,7 @@ sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
 	else
 		sock_write_timestamp(sk, kt);
 
-	if (sock_flag(sk, SOCK_WIFI_STATUS) && skb->wifi_acked_valid)
+	if (sock_flag(sk, SOCK_WIFI_STATUS) && skb_wifi_acked_valid(skb))
 		__sock_recv_wifi_status(msg, sk, skb);
 }
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index ef81452759be..768f9d04911f 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5189,6 +5189,7 @@ void skb_tstamp_tx(struct sk_buff *orig_skb,
 }
 EXPORT_SYMBOL_GPL(skb_tstamp_tx);
 
+#ifdef CONFIG_WIRELESS
 void skb_complete_wifi_ack(struct sk_buff *skb, bool acked)
 {
 	struct sock *sk = skb->sk;
@@ -5214,6 +5215,7 @@ void skb_complete_wifi_ack(struct sk_buff *skb, bool acked)
 		kfree_skb(skb);
 }
 EXPORT_SYMBOL_GPL(skb_complete_wifi_ack);
+#endif /* CONFIG_WIRELESS */
 
 /**
  * skb_partial_csum_set - set up and verify partial csum values for packet
diff --git a/net/socket.c b/net/socket.c
index 73e493da4589..a7b4b37d86df 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -957,6 +957,7 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
 }
 EXPORT_SYMBOL_GPL(__sock_recv_timestamp);
 
+#ifdef CONFIG_WIRELESS
 void __sock_recv_wifi_status(struct msghdr *msg, struct sock *sk,
 	struct sk_buff *skb)
 {
@@ -972,6 +973,7 @@ void __sock_recv_wifi_status(struct msghdr *msg, struct sock *sk,
 	put_cmsg(msg, SOL_SOCKET, SCM_WIFI_STATUS, sizeof(ack), &ack);
 }
 EXPORT_SYMBOL_GPL(__sock_recv_wifi_status);
+#endif
 
 static inline void sock_recv_drops(struct msghdr *msg, struct sock *sk,
 				   struct sk_buff *skb)
-- 
cgit v1.2.3-70-g09d2


From c24831a13ba2e472f874483525084da2f2feb890 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 17 Apr 2023 08:53:47 -0700
Subject: net: skbuff: hide csum_not_inet when CONFIG_IP_SCTP not set

SCTP is not universally deployed, allow hiding its bit
from the skb.

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 14 ++++++++++++++
 net/core/dev.c         |  3 +--
 net/sched/act_csum.c   |  3 +--
 3 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 513f03b23a73..98d6b48f4dcf 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -983,7 +983,9 @@ struct sk_buff {
 	__u8			decrypted:1;
 #endif
 	__u8			slow_gro:1;
+#if IS_ENABLED(CONFIG_IP_SCTP)
 	__u8			csum_not_inet:1;
+#endif
 
 #ifdef CONFIG_NET_SCHED
 	__u16			tc_index;	/* traffic control index */
@@ -5060,7 +5062,19 @@ static inline void skb_reset_redirect(struct sk_buff *skb)
 
 static inline bool skb_csum_is_sctp(struct sk_buff *skb)
 {
+#if IS_ENABLED(CONFIG_IP_SCTP)
 	return skb->csum_not_inet;
+#else
+	return 0;
+#endif
+}
+
+static inline void skb_reset_csum_not_inet(struct sk_buff *skb)
+{
+	skb->ip_summed = CHECKSUM_NONE;
+#if IS_ENABLED(CONFIG_IP_SCTP)
+	skb->csum_not_inet = 0;
+#endif
 }
 
 static inline void skb_set_kcov_handle(struct sk_buff *skb,
diff --git a/net/core/dev.c b/net/core/dev.c
index 8aea68275172..3fc4dba71f9d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3315,8 +3315,7 @@ int skb_crc32c_csum_help(struct sk_buff *skb)
 						  skb->len - start, ~(__u32)0,
 						  crc32c_csum_stub));
 	*(__le32 *)(skb->data + offset) = crc32c_csum;
-	skb->ip_summed = CHECKSUM_NONE;
-	skb->csum_not_inet = 0;
+	skb_reset_csum_not_inet(skb);
 out:
 	return ret;
 }
diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c
index 95e9304024b7..8ed285023a40 100644
--- a/net/sched/act_csum.c
+++ b/net/sched/act_csum.c
@@ -376,8 +376,7 @@ static int tcf_csum_sctp(struct sk_buff *skb, unsigned int ihl,
 
 	sctph->checksum = sctp_compute_cksum(skb,
 					     skb_network_offset(skb) + ihl);
-	skb->ip_summed = CHECKSUM_NONE;
-	skb->csum_not_inet = 0;
+	skb_reset_csum_not_inet(skb);
 
 	return 1;
 }
-- 
cgit v1.2.3-70-g09d2


From 4398f3f6d1380d9f71a484e1e5d869ba0eaf23d5 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 17 Apr 2023 08:53:48 -0700
Subject: net: skbuff: move alloc_cpu into a potential hole

alloc_cpu is currently between 4 byte fields, so it's almost
guaranteed to create a 2B hole. It has a knock on effect of
creating a 4B hole after @end (and @end and @tail being in
different cachelines).

None of this matters hugely, but for kernel configs which
don't enable all the features there may well be a 2B hole
after the bitfield. Move alloc_cpu there.

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 98d6b48f4dcf..2595b2cfba0d 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -991,6 +991,8 @@ struct sk_buff {
 	__u16			tc_index;	/* traffic control index */
 #endif
 
+	u16			alloc_cpu;
+
 	union {
 		__wsum		csum;
 		struct {
@@ -1014,7 +1016,6 @@ struct sk_buff {
 		unsigned int	sender_cpu;
 	};
 #endif
-	u16			alloc_cpu;
 #ifdef CONFIG_NETWORK_SECMARK
 	__u32		secmark;
 #endif
-- 
cgit v1.2.3-70-g09d2


From 4c60d04c2888a542f96fe77ecbdfc242b484f943 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 17 Apr 2023 08:53:49 -0700
Subject: net: skbuff: push nf_trace down the bitfield

nf_trace is a debug feature, AFAIU, and yet it sits oddly
high in the sk_buff bitfield. Move it down, pushing up
dst_pending_confirm and inner_protocol_type.

Next change will make nf_trace optional (under Kconfig)
and all optional fields should be placed after 2b fields
to avoid 2b fields straddling bytes.

dst_pending_confirm is L3, so it makes sense next to ignore_df.
inner_protocol_type goes up just to keep the balance.

Acked-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 2595b2cfba0d..3ae9e8868afa 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -934,7 +934,7 @@ struct sk_buff {
 	/* public: */
 	__u8			pkt_type:3; /* see PKT_TYPE_MAX */
 	__u8			ignore_df:1;
-	__u8			nf_trace:1;
+	__u8			dst_pending_confirm:1;
 	__u8			ip_summed:2;
 	__u8			ooo_okay:1;
 
@@ -949,7 +949,7 @@ struct sk_buff {
 	__u8			remcsum_offload:1;
 	__u8			csum_complete_sw:1;
 	__u8			csum_level:2;
-	__u8			dst_pending_confirm:1;
+	__u8			inner_protocol_type:1;
 
 	__u8			l4_hash:1;
 	__u8			sw_hash:1;
@@ -967,7 +967,7 @@ struct sk_buff {
 #endif
 
 	__u8			ipvs_property:1;
-	__u8			inner_protocol_type:1;
+	__u8			nf_trace:1;
 #ifdef CONFIG_NET_SWITCHDEV
 	__u8			offload_fwd_mark:1;
 	__u8			offload_l3_fwd_mark:1;
-- 
cgit v1.2.3-70-g09d2


From 48d80c394d3d1afcb49d26398917f5be27bf44cb Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 17 Apr 2023 08:53:50 -0700
Subject: net: skbuff: hide nf_trace and ipvs_property

Accesses to nf_trace and ipvs_property are already wrapped
by ifdefs where necessary. Don't allocate the bits for those
fields at all if possible.

Acked-by: Florian Westphal <fw@strlen.de>
Acked-by: Simon Horman <horms@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 3ae9e8868afa..5f120bbab9da 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -966,8 +966,12 @@ struct sk_buff {
 	__u8			ndisc_nodetype:2;
 #endif
 
+#if IS_ENABLED(CONFIG_IP_VS)
 	__u8			ipvs_property:1;
+#endif
+#if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) || IS_ENABLED(CONFIG_NF_TABLES)
 	__u8			nf_trace:1;
+#endif
 #ifdef CONFIG_NET_SWITCHDEV
 	__u8			offload_fwd_mark:1;
 	__u8			offload_l3_fwd_mark:1;
-- 
cgit v1.2.3-70-g09d2


From 4e195166624851d1da1a4e42110e66ee4faca4a3 Mon Sep 17 00:00:00 2001
From: Corinna Vinschen <vinschen@redhat.com>
Date: Mon, 17 Apr 2023 21:29:03 +0200
Subject: stmmac: fix changing mac address

Without the IFF_LIVE_ADDR_CHANGE flag being set, the network code
disallows changing the mac address while the interface is UP.

Consequences are, for instance, that the interface can't be used
in a failover bond.

Add the missing flag to net_device priv_flags.

Tested on Intel Elkhart Lake with default settings, as well as with
failover and alb mode bonds.

Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 5280256a638f..47534310365a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -7339,6 +7339,8 @@ int stmmac_dvr_probe(struct device *device,
 	if (flow_ctrl)
 		priv->flow_ctrl = FLOW_AUTO;	/* RX/TX pause on */
 
+	ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
+
 	/* Setup channels NAPI */
 	stmmac_napi_add(ndev);
 
-- 
cgit v1.2.3-70-g09d2


From ed09c61eb19d9889780c791cb316ac76468f5186 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Mon, 17 Apr 2023 13:35:00 +0300
Subject: dt-bindings: net: Convert ath10k to YAML

Convert the ath10k bindings to YAML.

Dropped properties that are absent at the current state of mainline:
- qcom,msi_addr
- qcom,msi_base

Somewhat based on the ath11k bindings.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230406-topic-ath10k_bindings-v4-1-9f67a6bb0d56@linaro.org
---
 .../bindings/net/wireless/qcom,ath10k.txt          | 215 -------------
 .../bindings/net/wireless/qcom,ath10k.yaml         | 358 +++++++++++++++++++++
 MAINTAINERS                                        |   2 +-
 3 files changed, 359 insertions(+), 216 deletions(-)
 delete mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
 create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
deleted file mode 100644
index b61c2d5a0ff7..000000000000
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
+++ /dev/null
@@ -1,215 +0,0 @@
-* Qualcomm Atheros ath10k wireless devices
-
-Required properties:
-- compatible: Should be one of the following:
-	* "qcom,ath10k"
-	* "qcom,ipq4019-wifi"
-	* "qcom,wcn3990-wifi"
-
-PCI based devices uses compatible string "qcom,ath10k" and takes calibration
-data along with board specific data via "qcom,ath10k-calibration-data".
-Rest of the properties are not applicable for PCI based devices.
-
-AHB based devices (i.e. ipq4019) uses compatible string "qcom,ipq4019-wifi"
-and also uses most of the properties defined in this doc (except
-"qcom,ath10k-calibration-data"). It uses "qcom,ath10k-pre-calibration-data"
-to carry pre calibration data.
-
-In general, entry "qcom,ath10k-pre-calibration-data" and
-"qcom,ath10k-calibration-data" conflict with each other and only one
-can be provided per device.
-
-SNOC based devices (i.e. wcn3990) uses compatible string "qcom,wcn3990-wifi".
-
-- reg: Address and length of the register set for the device.
-- reg-names: Must include the list of following reg names,
-	     "membase"
-- interrupts: reference to the list of 17 interrupt numbers for "qcom,ipq4019-wifi"
-	      compatible target.
-	      reference to the list of 12 interrupt numbers for "qcom,wcn3990-wifi"
-	      compatible target.
-	      Must contain interrupt-names property per entry for
-	      "qcom,ath10k", "qcom,ipq4019-wifi" compatible targets.
-
-- interrupt-names: Must include the entries for MSI interrupt
-		   names ("msi0" to "msi15") and legacy interrupt
-		   name ("legacy") for "qcom,ath10k", "qcom,ipq4019-wifi"
-		   compatible targets.
-
-Optional properties:
-- resets: Must contain an entry for each entry in reset-names.
-          See ../reset/reseti.txt for details.
-- reset-names: Must include the list of following reset names,
-	       "wifi_cpu_init"
-	       "wifi_radio_srif"
-	       "wifi_radio_warm"
-	       "wifi_radio_cold"
-	       "wifi_core_warm"
-	       "wifi_core_cold"
-- clocks: List of clock specifiers, must contain an entry for each required
-          entry in clock-names.
-- clock-names: Should contain the clock names "wifi_wcss_cmd", "wifi_wcss_ref",
-	       "wifi_wcss_rtc" for "qcom,ipq4019-wifi" compatible target and
-	       "cxo_ref_clk_pin" and optionally "qdss" for "qcom,wcn3990-wifi"
-	       compatible target.
-- qcom,msi_addr: MSI interrupt address.
-- qcom,msi_base: Base value to add before writing MSI data into
-		MSI address register.
-- qcom,ath10k-calibration-variant: string to search for in the board-2.bin
-				   variant list with the same bus and device
-				   specific ids
-- qcom,ath10k-calibration-data : calibration data + board specific data
-				 as an array, the length can vary between
-				 hw versions.
-- qcom,ath10k-pre-calibration-data : pre calibration data as an array,
-				     the length can vary between hw versions.
-- <supply-name>-supply: handle to the regulator device tree node
-			   optional "supply-name" are "vdd-0.8-cx-mx",
-			   "vdd-1.8-xo", "vdd-1.3-rfa", "vdd-3.3-ch0",
-			   and "vdd-3.3-ch1".
-- memory-region:
-	Usage: optional
-	Value type: <phandle>
-	Definition: reference to the reserved-memory for the msa region
-		    used by the wifi firmware running in Q6.
-- iommus:
-	Usage: optional
-	Value type: <prop-encoded-array>
-	Definition: A list of phandle and IOMMU specifier pairs.
-- ext-fem-name:
-	Usage: Optional
-	Value type: string
-	Definition: Name of external front end module used. Some valid FEM names
-		    for example: "microsemi-lx5586", "sky85703-11"
-		    and "sky85803" etc.
-- qcom,snoc-host-cap-8bit-quirk:
-	Usage: Optional
-	Value type: <empty>
-	Definition: Quirk specifying that the firmware expects the 8bit version
-		    of the host capability QMI request
-- qcom,xo-cal-data: xo cal offset to be configured in xo trim register.
-
-- qcom,msa-fixed-perm: Boolean context flag to disable SCM call for statically
-		       mapped msa region.
-
-- qcom,coexist-support : should contain eithr "0" or "1" to indicate coex
-			 support by the hardware.
-- qcom,coexist-gpio-pin : gpio pin number  information to support coex
-			  which will be used by wifi firmware.
-
-* Subnodes
-The ath10k wifi node can contain one optional firmware subnode.
-Firmware subnode is needed when the platform does not have TustZone.
-The firmware subnode must have:
-
-- iommus:
-	Usage: required
-	Value type: <prop-encoded-array>
-	Definition: A list of phandle and IOMMU specifier pairs.
-
-
-Example (to supply PCI based wifi block details):
-
-In this example, the node is defined as child node of the PCI controller.
-
-pci {
-	pcie@0 {
-		reg = <0 0 0 0 0>;
-		#interrupt-cells = <1>;
-		#size-cells = <2>;
-		#address-cells = <3>;
-		device_type = "pci";
-
-		wifi@0,0 {
-			reg = <0 0 0 0 0>;
-			qcom,ath10k-calibration-data = [ 01 02 03 ... ];
-			ext-fem-name = "microsemi-lx5586";
-		};
-	};
-};
-
-Example (to supply ipq4019 SoC wifi block details):
-
-wifi0: wifi@a000000 {
-	compatible = "qcom,ipq4019-wifi";
-	reg = <0xa000000 0x200000>;
-	resets = <&gcc WIFI0_CPU_INIT_RESET>,
-		 <&gcc WIFI0_RADIO_SRIF_RESET>,
-		 <&gcc WIFI0_RADIO_WARM_RESET>,
-		 <&gcc WIFI0_RADIO_COLD_RESET>,
-		 <&gcc WIFI0_CORE_WARM_RESET>,
-		 <&gcc WIFI0_CORE_COLD_RESET>;
-	reset-names = "wifi_cpu_init",
-		      "wifi_radio_srif",
-		      "wifi_radio_warm",
-		      "wifi_radio_cold",
-		      "wifi_core_warm",
-		      "wifi_core_cold";
-	clocks = <&gcc GCC_WCSS2G_CLK>,
-		 <&gcc GCC_WCSS2G_REF_CLK>,
-		 <&gcc GCC_WCSS2G_RTC_CLK>;
-	clock-names = "wifi_wcss_cmd",
-		      "wifi_wcss_ref",
-		      "wifi_wcss_rtc";
-	interrupts = <0 0x20 0x1>,
-		     <0 0x21 0x1>,
-		     <0 0x22 0x1>,
-		     <0 0x23 0x1>,
-		     <0 0x24 0x1>,
-		     <0 0x25 0x1>,
-		     <0 0x26 0x1>,
-		     <0 0x27 0x1>,
-		     <0 0x28 0x1>,
-		     <0 0x29 0x1>,
-		     <0 0x2a 0x1>,
-		     <0 0x2b 0x1>,
-		     <0 0x2c 0x1>,
-		     <0 0x2d 0x1>,
-		     <0 0x2e 0x1>,
-		     <0 0x2f 0x1>,
-		     <0 0xa8 0x0>;
-	interrupt-names = "msi0",  "msi1",  "msi2",  "msi3",
-			  "msi4",  "msi5",  "msi6",  "msi7",
-			  "msi8",  "msi9",  "msi10", "msi11",
-			  "msi12", "msi13", "msi14", "msi15",
-			  "legacy";
-	qcom,msi_addr = <0x0b006040>;
-	qcom,msi_base = <0x40>;
-	qcom,ath10k-pre-calibration-data = [ 01 02 03 ... ];
-	qcom,coexist-support = <1>;
-	qcom,coexist-gpio-pin = <0x33>;
-};
-
-Example (to supply wcn3990 SoC wifi block details):
-
-wifi@18000000 {
-		compatible = "qcom,wcn3990-wifi";
-		reg = <0x18800000 0x800000>;
-		reg-names = "membase";
-		clocks = <&clock_gcc clk_rf_clk2_pin>;
-		clock-names = "cxo_ref_clk_pin";
-		interrupts =
-			<GIC_SPI 414 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 415 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 416 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 417 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 418 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 419 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 420 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 421 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 422 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 423 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 424 IRQ_TYPE_LEVEL_HIGH>,
-			<GIC_SPI 425 IRQ_TYPE_LEVEL_HIGH>;
-		vdd-0.8-cx-mx-supply = <&pm8998_l5>;
-		vdd-1.8-xo-supply = <&vreg_l7a_1p8>;
-		vdd-1.3-rfa-supply = <&vreg_l17a_1p3>;
-		vdd-3.3-ch0-supply = <&vreg_l25a_3p3>;
-		vdd-3.3-ch1-supply = <&vreg_l26a_3p3>;
-		memory-region = <&wifi_msa_mem>;
-		iommus = <&apps_smmu 0x0040 0x1>;
-		qcom,msa-fixed-perm;
-		wifi-firmware {
-			iommus = <&apps_iommu 0xc22 0x1>;
-		};
-};
diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml
new file mode 100644
index 000000000000..c85ed330426d
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml
@@ -0,0 +1,358 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/wireless/qcom,ath10k.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Technologies ath10k wireless devices
+
+maintainers:
+  - Kalle Valo <kvalo@kernel.org>
+
+description:
+  Qualcomm Technologies, Inc. IEEE 802.11ac devices.
+
+properties:
+  compatible:
+    enum:
+      - qcom,ath10k # SDIO-based devices
+      - qcom,ipq4019-wifi
+      - qcom,wcn3990-wifi # SNoC-based devices
+
+  reg:
+    maxItems: 1
+
+  reg-names:
+    items:
+      - const: membase
+
+  interrupts:
+    minItems: 12
+    maxItems: 17
+
+  interrupt-names:
+    minItems: 12
+    maxItems: 17
+
+  memory-region:
+    maxItems: 1
+    description:
+      Reference to the MSA memory region used by the Wi-Fi firmware
+      running on the Q6 core.
+
+  iommus:
+    minItems: 1
+    maxItems: 2
+
+  clocks:
+    minItems: 1
+    maxItems: 3
+
+  clock-names:
+    minItems: 1
+    maxItems: 3
+
+  resets:
+    maxItems: 6
+
+  reset-names:
+    items:
+      - const: wifi_cpu_init
+      - const: wifi_radio_srif
+      - const: wifi_radio_warm
+      - const: wifi_radio_cold
+      - const: wifi_core_warm
+      - const: wifi_core_cold
+
+  ext-fem-name:
+    $ref: /schemas/types.yaml#/definitions/string
+    description: Name of external front end module used.
+    enum:
+      - microsemi-lx5586
+      - sky85703-11
+      - sky85803
+
+  wifi-firmware:
+    type: object
+    additionalProperties: false
+    description: |
+      The ath10k Wi-Fi node can contain one optional firmware subnode.
+      Firmware subnode is needed when the platform does not have Trustzone.
+    properties:
+      iommus:
+        maxItems: 1
+    required:
+      - iommus
+
+  qcom,ath10k-calibration-data:
+    $ref: /schemas/types.yaml#/definitions/uint8-array
+    description:
+      Calibration data + board-specific data as a byte array. The length
+      can vary between hardware versions.
+
+  qcom,ath10k-calibration-variant:
+    $ref: /schemas/types.yaml#/definitions/string
+    description:
+      Unique variant identifier of the calibration data in board-2.bin
+      for designs with colliding bus and device specific ids
+
+  qcom,ath10k-pre-calibration-data:
+    $ref: /schemas/types.yaml#/definitions/uint8-array
+    description:
+      Pre-calibration data as a byte array. The length can vary between
+      hardware versions.
+
+  qcom,coexist-support:
+    $ref: /schemas/types.yaml#/definitions/uint8
+    enum: [0, 1]
+    description:
+      Indicate coex support by the hardware.
+
+  qcom,coexist-gpio-pin:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description:
+      COEX GPIO number provided to the Wi-Fi firmware.
+
+  qcom,msa-fixed-perm:
+    type: boolean
+    description:
+      Whether to skip executing an SCM call that reassigns the memory
+      region ownership.
+
+  qcom,smem-states:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    description: State bits used by the AP to signal the WLAN Q6.
+    items:
+      - description: Signal bits used to enable/disable low power mode
+                     on WCN in the case of WoW (Wake on Wireless).
+
+  qcom,smem-state-names:
+    description: The names of the state bits used for SMP2P output.
+    items:
+      - const: wlan-smp2p-out
+
+  qcom,snoc-host-cap-8bit-quirk:
+    type: boolean
+    description:
+      Quirk specifying that the firmware expects the 8bit version
+      of the host capability QMI request
+
+  qcom,xo-cal-data:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description:
+      XO cal offset to be configured in XO trim register.
+
+  vdd-0.8-cx-mx-supply:
+    description: Main logic power rail
+
+  vdd-1.8-xo-supply:
+    description: Crystal oscillator supply
+
+  vdd-1.3-rfa-supply:
+    description: RFA supply
+
+  vdd-3.3-ch0-supply:
+    description: Primary Wi-Fi antenna supply
+
+  vdd-3.3-ch1-supply:
+    description: Secondary Wi-Fi antenna supply
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,ipq4019-wifi
+    then:
+      properties:
+        interrupts:
+          minItems: 17
+          maxItems: 17
+
+        interrupt-names:
+          items:
+            - const: msi0
+            - const: msi1
+            - const: msi2
+            - const: msi3
+            - const: msi4
+            - const: msi5
+            - const: msi6
+            - const: msi7
+            - const: msi8
+            - const: msi9
+            - const: msi10
+            - const: msi11
+            - const: msi12
+            - const: msi13
+            - const: msi14
+            - const: msi15
+            - const: legacy
+
+        clocks:
+          items:
+            - description: Wi-Fi command clock
+            - description: Wi-Fi reference clock
+            - description: Wi-Fi RTC clock
+
+        clock-names:
+          items:
+            - const: wifi_wcss_cmd
+            - const: wifi_wcss_ref
+            - const: wifi_wcss_rtc
+
+      required:
+        - clocks
+        - clock-names
+        - interrupts
+        - interrupt-names
+        - resets
+        - reset-names
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,wcn3990-wifi
+
+    then:
+      properties:
+        clocks:
+          minItems: 1
+          items:
+            - description: XO reference clock
+            - description: Qualcomm Debug Subsystem clock
+
+        clock-names:
+          minItems: 1
+          items:
+            - const: cxo_ref_clk_pin
+            - const: qdss
+
+        interrupts:
+          items:
+            - description: CE0
+            - description: CE1
+            - description: CE2
+            - description: CE3
+            - description: CE4
+            - description: CE5
+            - description: CE6
+            - description: CE7
+            - description: CE8
+            - description: CE9
+            - description: CE10
+            - description: CE11
+
+        interrupt-names: false
+
+      required:
+        - interrupts
+
+examples:
+  # SNoC
+  - |
+    #include <dt-bindings/clock/qcom,rpmcc.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+
+    wifi@18800000 {
+      compatible = "qcom,wcn3990-wifi";
+      reg = <0x18800000 0x800000>;
+      reg-names = "membase";
+      memory-region = <&wlan_msa_mem>;
+      clocks = <&rpmcc RPM_SMD_RF_CLK2_PIN>;
+      clock-names = "cxo_ref_clk_pin";
+      interrupts = <GIC_SPI 413 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 414 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 415 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 416 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 417 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 418 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 420 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 421 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 422 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 423 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 424 IRQ_TYPE_LEVEL_HIGH>,
+                   <GIC_SPI 425 IRQ_TYPE_LEVEL_HIGH>;
+      iommus = <&anoc2_smmu 0x1900>,
+               <&anoc2_smmu 0x1901>;
+      qcom,snoc-host-cap-8bit-quirk;
+      vdd-0.8-cx-mx-supply = <&vreg_l5a_0p8>;
+      vdd-1.8-xo-supply = <&vreg_l7a_1p8>;
+      vdd-1.3-rfa-supply = <&vreg_l17a_1p3>;
+      vdd-3.3-ch0-supply = <&vreg_l25a_3p3>;
+      vdd-3.3-ch1-supply = <&vreg_l23a_3p3>;
+
+      wifi-firmware {
+        iommus = <&apps_smmu 0x1c02 0x1>;
+      };
+    };
+
+  # AHB
+  - |
+    #include <dt-bindings/clock/qcom,gcc-ipq4019.h>
+
+    wifi@a000000 {
+        compatible = "qcom,ipq4019-wifi";
+        reg = <0xa000000 0x200000>;
+        resets = <&gcc WIFI0_CPU_INIT_RESET>,
+                 <&gcc WIFI0_RADIO_SRIF_RESET>,
+                 <&gcc WIFI0_RADIO_WARM_RESET>,
+                 <&gcc WIFI0_RADIO_COLD_RESET>,
+                 <&gcc WIFI0_CORE_WARM_RESET>,
+                 <&gcc WIFI0_CORE_COLD_RESET>;
+        reset-names = "wifi_cpu_init",
+                      "wifi_radio_srif",
+                      "wifi_radio_warm",
+                      "wifi_radio_cold",
+                      "wifi_core_warm",
+                      "wifi_core_cold";
+        clocks = <&gcc GCC_WCSS2G_CLK>,
+                 <&gcc GCC_WCSS2G_REF_CLK>,
+                 <&gcc GCC_WCSS2G_RTC_CLK>;
+        clock-names = "wifi_wcss_cmd",
+                      "wifi_wcss_ref",
+                      "wifi_wcss_rtc";
+        interrupts = <GIC_SPI 32 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 33 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 34 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 35 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 36 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 37 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 38 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 39 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 40 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 41 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 42 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 43 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 44 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 45 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 46 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 47 IRQ_TYPE_EDGE_RISING>,
+                     <GIC_SPI 168 IRQ_TYPE_LEVEL_HIGH>;
+        interrupt-names =  "msi0",
+                           "msi1",
+                           "msi2",
+                           "msi3",
+                           "msi4",
+                           "msi5",
+                           "msi6",
+                           "msi7",
+                           "msi8",
+                           "msi9",
+                           "msi10",
+                           "msi11",
+                           "msi12",
+                           "msi13",
+                           "msi14",
+                           "msi15",
+                           "legacy";
+      };
diff --git a/MAINTAINERS b/MAINTAINERS
index 8d5bc223f305..c659e12e0c4f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -17196,7 +17196,7 @@ S:	Supported
 W:	https://wireless.wiki.kernel.org/en/users/Drivers/ath10k
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git
 F:	drivers/net/wireless/ath/ath10k/
-F:	Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
+F:	Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml
 
 QUALCOMM ATHEROS ATH11K WIRELESS DRIVER
 M:	Kalle Valo <kvalo@kernel.org>
-- 
cgit v1.2.3-70-g09d2


From 93a91f40c25c3d0e61f8540a7accf105090f9995 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Mon, 17 Apr 2023 13:35:00 +0300
Subject: wifi: ath11k: fix double free of peer rx_tid during reo cmd failure

Peer rx_tid is locally copied thrice during peer_rx_tid_cleanup to
send REO_CMD_UPDATE_RX_QUEUE followed by REO_CMD_FLUSH_CACHE to flush
all aged REO descriptors from HW cache.

When sending REO_CMD_FLUSH_CACHE fails, we do dma unmap of already
mapped rx_tid->vaddr and free it. This is not checked during
reo_cmd_list_cleanup() and dp_reo_cmd_free() before trying to free and
unmap again.

Fix this by setting rx_tid->vaddr NULL in rx tid delete and also
wherever freeing it to check in reo_cmd_list_cleanup() and
reo_cmd_free() before trying to free again.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Sathishkumar Muruganandam <quic_murugana@quicinc.com>
Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403182420.23375-2-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 43 ++++++++++++++++++++++++---------
 1 file changed, 31 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 847da167a41c..1d45aa018df3 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -668,13 +668,18 @@ void ath11k_dp_reo_cmd_list_cleanup(struct ath11k_base *ab)
 	struct ath11k_dp *dp = &ab->dp;
 	struct dp_reo_cmd *cmd, *tmp;
 	struct dp_reo_cache_flush_elem *cmd_cache, *tmp_cache;
+	struct dp_rx_tid *rx_tid;
 
 	spin_lock_bh(&dp->reo_cmd_lock);
 	list_for_each_entry_safe(cmd, tmp, &dp->reo_cmd_list, list) {
 		list_del(&cmd->list);
-		dma_unmap_single(ab->dev, cmd->data.paddr,
-				 cmd->data.size, DMA_BIDIRECTIONAL);
-		kfree(cmd->data.vaddr);
+		rx_tid = &cmd->data;
+		if (rx_tid->vaddr) {
+			dma_unmap_single(ab->dev, rx_tid->paddr,
+					 rx_tid->size, DMA_BIDIRECTIONAL);
+			kfree(rx_tid->vaddr);
+			rx_tid->vaddr = NULL;
+		}
 		kfree(cmd);
 	}
 
@@ -682,9 +687,13 @@ void ath11k_dp_reo_cmd_list_cleanup(struct ath11k_base *ab)
 				 &dp->reo_cmd_cache_flush_list, list) {
 		list_del(&cmd_cache->list);
 		dp->reo_cmd_cache_flush_count--;
-		dma_unmap_single(ab->dev, cmd_cache->data.paddr,
-				 cmd_cache->data.size, DMA_BIDIRECTIONAL);
-		kfree(cmd_cache->data.vaddr);
+		rx_tid = &cmd_cache->data;
+		if (rx_tid->vaddr) {
+			dma_unmap_single(ab->dev, rx_tid->paddr,
+					 rx_tid->size, DMA_BIDIRECTIONAL);
+			kfree(rx_tid->vaddr);
+			rx_tid->vaddr = NULL;
+		}
 		kfree(cmd_cache);
 	}
 	spin_unlock_bh(&dp->reo_cmd_lock);
@@ -698,10 +707,12 @@ static void ath11k_dp_reo_cmd_free(struct ath11k_dp *dp, void *ctx,
 	if (status != HAL_REO_CMD_SUCCESS)
 		ath11k_warn(dp->ab, "failed to flush rx tid hw desc, tid %d status %d\n",
 			    rx_tid->tid, status);
-
-	dma_unmap_single(dp->ab->dev, rx_tid->paddr, rx_tid->size,
-			 DMA_BIDIRECTIONAL);
-	kfree(rx_tid->vaddr);
+	if (rx_tid->vaddr) {
+		dma_unmap_single(dp->ab->dev, rx_tid->paddr, rx_tid->size,
+				 DMA_BIDIRECTIONAL);
+		kfree(rx_tid->vaddr);
+		rx_tid->vaddr = NULL;
+	}
 }
 
 static void ath11k_dp_reo_cache_flush(struct ath11k_base *ab,
@@ -740,6 +751,7 @@ static void ath11k_dp_reo_cache_flush(struct ath11k_base *ab,
 		dma_unmap_single(ab->dev, rx_tid->paddr, rx_tid->size,
 				 DMA_BIDIRECTIONAL);
 		kfree(rx_tid->vaddr);
+		rx_tid->vaddr = NULL;
 	}
 }
 
@@ -792,6 +804,7 @@ free_desc:
 	dma_unmap_single(ab->dev, rx_tid->paddr, rx_tid->size,
 			 DMA_BIDIRECTIONAL);
 	kfree(rx_tid->vaddr);
+	rx_tid->vaddr = NULL;
 }
 
 void ath11k_peer_rx_tid_delete(struct ath11k *ar,
@@ -804,6 +817,8 @@ void ath11k_peer_rx_tid_delete(struct ath11k *ar,
 	if (!rx_tid->active)
 		return;
 
+	rx_tid->active = false;
+
 	cmd.flag = HAL_REO_CMD_FLG_NEED_STATUS;
 	cmd.addr_lo = lower_32_bits(rx_tid->paddr);
 	cmd.addr_hi = upper_32_bits(rx_tid->paddr);
@@ -818,9 +833,11 @@ void ath11k_peer_rx_tid_delete(struct ath11k *ar,
 		dma_unmap_single(ar->ab->dev, rx_tid->paddr, rx_tid->size,
 				 DMA_BIDIRECTIONAL);
 		kfree(rx_tid->vaddr);
+		rx_tid->vaddr = NULL;
 	}
 
-	rx_tid->active = false;
+	rx_tid->paddr = 0;
+	rx_tid->size = 0;
 }
 
 static int ath11k_dp_rx_link_desc_return(struct ath11k_base *ab,
@@ -967,6 +984,7 @@ static void ath11k_dp_rx_tid_mem_free(struct ath11k_base *ab,
 	dma_unmap_single(ab->dev, rx_tid->paddr, rx_tid->size,
 			 DMA_BIDIRECTIONAL);
 	kfree(rx_tid->vaddr);
+	rx_tid->vaddr = NULL;
 
 	rx_tid->active = false;
 
@@ -1067,7 +1085,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 	return ret;
 
 err_mem_free:
-	kfree(vaddr);
+	kfree(rx_tid->vaddr);
+	rx_tid->vaddr = NULL;
 
 	return ret;
 }
-- 
cgit v1.2.3-70-g09d2


From a8ae833657a45746debde85c38bb7f070d344026 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Mon, 17 Apr 2023 13:35:01 +0300
Subject: wifi: ath11k: Prevent REO cmd failures

Prevent REO cmd failures causing double free by increasing REO cmd
ring size and moving REO status ring mask to IRQ group 3 from group
0 to separate from tx completion ring on IRQ group 0 which may delay
reo status processing.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Sathishkumar Muruganandam <quic_murugana@quicinc.com>
Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403182420.23375-3-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp.h | 2 +-
 drivers/net/wireless/ath/ath11k/hw.c | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index be9eafc872b3..6022e27ba91d 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -214,7 +214,7 @@ struct ath11k_pdev_dp {
 #define DP_REO_REINJECT_RING_SIZE	32
 #define DP_RX_RELEASE_RING_SIZE		1024
 #define DP_REO_EXCEPTION_RING_SIZE	128
-#define DP_REO_CMD_RING_SIZE		128
+#define DP_REO_CMD_RING_SIZE		256
 #define DP_REO_STATUS_RING_SIZE		2048
 #define DP_RXDMA_BUF_RING_SIZE		4096
 #define DP_RXDMA_REFILL_RING_SIZE	2048
diff --git a/drivers/net/wireless/ath/ath11k/hw.c b/drivers/net/wireless/ath/ath11k/hw.c
index 6b4355a68e26..eb995f9cf0fa 100644
--- a/drivers/net/wireless/ath/ath11k/hw.c
+++ b/drivers/net/wireless/ath/ath11k/hw.c
@@ -1233,6 +1233,7 @@ const struct ath11k_hw_ring_mask ath11k_hw_ring_mask_ipq8074 = {
 		ATH11K_RX_WBM_REL_RING_MASK_0,
 	},
 	.reo_status = {
+		0, 0, 0,
 		ATH11K_REO_STATUS_RING_MASK_0,
 	},
 	.rxdma2host = {
-- 
cgit v1.2.3-70-g09d2


From 20487cc3ff36bbfa9505f0a078ba98f09abfc717 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Mon, 17 Apr 2023 13:35:01 +0300
Subject: wifi: ath11k: add peer mac information in failure cases

During reo command failure, the peer mac detail for which the reo
command was not successful is unknown. Hence, to improve the
debuggability, add the peer mac information in the failure cases
which would be useful during multi client cases.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Sathishkumar Muruganandam <quic_murugana@quicinc.com>
Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403182420.23375-4-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 1d45aa018df3..a4934bd79969 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1009,7 +1009,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 
 	peer = ath11k_peer_find(ab, vdev_id, peer_mac);
 	if (!peer) {
-		ath11k_warn(ab, "failed to find the peer to set up rx tid\n");
+		ath11k_warn(ab, "failed to find the peer %pM to set up rx tid\n",
+			    peer_mac);
 		spin_unlock_bh(&ab->base_lock);
 		return -ENOENT;
 	}
@@ -1022,7 +1023,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 						    ba_win_sz, ssn, true);
 		spin_unlock_bh(&ab->base_lock);
 		if (ret) {
-			ath11k_warn(ab, "failed to update reo for rx tid %d\n", tid);
+			ath11k_warn(ab, "failed to update reo for peer %pM rx tid %d\n: %d",
+				    peer_mac, tid, ret);
 			return ret;
 		}
 
@@ -1030,8 +1032,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 							     peer_mac, paddr,
 							     tid, 1, ba_win_sz);
 		if (ret)
-			ath11k_warn(ab, "failed to send wmi command to update rx reorder queue, tid :%d (%d)\n",
-				    tid, ret);
+			ath11k_warn(ab, "failed to send wmi rx reorder queue for peer %pM tid %d: %d\n",
+				    peer_mac, tid, ret);
 		return ret;
 	}
 
@@ -1064,6 +1066,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 	ret = dma_mapping_error(ab->dev, paddr);
 	if (ret) {
 		spin_unlock_bh(&ab->base_lock);
+		ath11k_warn(ab, "failed to setup dma map for peer %pM rx tid %d: %d\n",
+			    peer_mac, tid, ret);
 		goto err_mem_free;
 	}
 
@@ -1077,8 +1081,8 @@ int ath11k_peer_rx_tid_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id,
 	ret = ath11k_wmi_peer_rx_reorder_queue_setup(ar, vdev_id, peer_mac,
 						     paddr, tid, 1, ba_win_sz);
 	if (ret) {
-		ath11k_warn(ar->ab, "failed to setup rx reorder queue, tid :%d (%d)\n",
-			    tid, ret);
+		ath11k_warn(ar->ab, "failed to setup rx reorder queue for peer %pM tid %d: %d\n",
+			    peer_mac, tid, ret);
 		ath11k_dp_rx_tid_mem_free(ab, peer_mac, vdev_id, tid);
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 6257c702264c44d74c6b71f0c62a7665da2dc356 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Mon, 17 Apr 2023 13:35:02 +0300
Subject: wifi: ath11k: fix tx status reporting in encap offload mode

ieee80211_tx_status() treats packets in 802.11 frame format and
tries to extract sta address from packet header. When tx encap
offload is enabled, this becomes invalid operation. Hence, switch
to using ieee80211_tx_status_ext() after filling in station
address for handling both 802.11 and 802.3 frames.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403195738.25367-2-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp.h    |  4 ++++
 drivers/net/wireless/ath/ath11k/dp_tx.c | 33 ++++++++++++++++++++++++++++++++-
 drivers/net/wireless/ath/ath11k/dp_tx.h |  1 +
 3 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp.h b/drivers/net/wireless/ath/ath11k/dp.h
index 6022e27ba91d..d04f78ab6b37 100644
--- a/drivers/net/wireless/ath/ath11k/dp.h
+++ b/drivers/net/wireless/ath/ath11k/dp.h
@@ -303,12 +303,16 @@ struct ath11k_dp {
 
 #define HTT_TX_WBM_COMP_STATUS_OFFSET 8
 
+#define HTT_INVALID_PEER_ID	0xffff
+
 /* HTT tx completion is overlaid in wbm_release_ring */
 #define HTT_TX_WBM_COMP_INFO0_STATUS		GENMASK(12, 9)
 #define HTT_TX_WBM_COMP_INFO0_REINJECT_REASON	GENMASK(16, 13)
 #define HTT_TX_WBM_COMP_INFO0_REINJECT_REASON	GENMASK(16, 13)
 
 #define HTT_TX_WBM_COMP_INFO1_ACK_RSSI		GENMASK(31, 24)
+#define HTT_TX_WBM_COMP_INFO2_SW_PEER_ID	GENMASK(15, 0)
+#define HTT_TX_WBM_COMP_INFO2_VALID		BIT(21)
 
 struct htt_tx_wbm_completion {
 	u32 info0;
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 8afbba236935..08a28464eb7a 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -316,10 +316,12 @@ ath11k_dp_tx_htt_tx_complete_buf(struct ath11k_base *ab,
 				 struct dp_tx_ring *tx_ring,
 				 struct ath11k_dp_htt_wbm_tx_status *ts)
 {
+	struct ieee80211_tx_status status = { 0 };
 	struct sk_buff *msdu;
 	struct ieee80211_tx_info *info;
 	struct ath11k_skb_cb *skb_cb;
 	struct ath11k *ar;
+	struct ath11k_peer *peer;
 
 	spin_lock(&tx_ring->tx_idr_lock);
 	msdu = idr_remove(&tx_ring->txbuf_idr, ts->msdu_id);
@@ -341,6 +343,11 @@ ath11k_dp_tx_htt_tx_complete_buf(struct ath11k_base *ab,
 
 	dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
 
+	if (!skb_cb->vif) {
+		dev_kfree_skb_any(msdu);
+		return;
+	}
+
 	memset(&info->status, 0, sizeof(info->status));
 
 	if (ts->acked) {
@@ -355,7 +362,23 @@ ath11k_dp_tx_htt_tx_complete_buf(struct ath11k_base *ab,
 		}
 	}
 
-	ieee80211_tx_status(ar->hw, msdu);
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find_by_id(ab, ts->peer_id);
+	if (!peer || !peer->sta) {
+		ath11k_dbg(ab, ATH11K_DBG_DATA,
+			   "dp_tx: failed to find the peer with peer_id %d\n",
+			    ts->peer_id);
+		spin_unlock_bh(&ab->base_lock);
+		dev_kfree_skb_any(msdu);
+		return;
+	}
+	spin_unlock_bh(&ab->base_lock);
+
+	status.sta = peer->sta;
+	status.info = info;
+	status.skb = msdu;
+
+	ieee80211_tx_status_ext(ar->hw, &status);
 }
 
 static void
@@ -379,7 +402,15 @@ ath11k_dp_tx_process_htt_tx_complete(struct ath11k_base *ab,
 		ts.msdu_id = msdu_id;
 		ts.ack_rssi = FIELD_GET(HTT_TX_WBM_COMP_INFO1_ACK_RSSI,
 					status_desc->info1);
+
+		if (FIELD_GET(HTT_TX_WBM_COMP_INFO2_VALID, status_desc->info2))
+			ts.peer_id = FIELD_GET(HTT_TX_WBM_COMP_INFO2_SW_PEER_ID,
+					       status_desc->info2);
+		else
+			ts.peer_id = HTT_INVALID_PEER_ID;
+
 		ath11k_dp_tx_htt_tx_complete_buf(ab, tx_ring, &ts);
+
 		break;
 	case HAL_WBM_REL_HTT_TX_COMP_STATUS_REINJ:
 	case HAL_WBM_REL_HTT_TX_COMP_STATUS_INSPECT:
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.h b/drivers/net/wireless/ath/ath11k/dp_tx.h
index e87d65bfbf06..68a21ea9b934 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.h
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.h
@@ -13,6 +13,7 @@ struct ath11k_dp_htt_wbm_tx_status {
 	u32 msdu_id;
 	bool acked;
 	int ack_rssi;
+	u16 peer_id;
 };
 
 void ath11k_dp_tx_update_txcompl(struct ath11k *ar, struct hal_tx_status *ts);
-- 
cgit v1.2.3-70-g09d2


From 2f0c9ac8362da09c80f1cd422ef7fd6fa9b252b9 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Mon, 17 Apr 2023 13:35:02 +0300
Subject: wifi: ath11k: Fix incorrect update of radiotap fields

Fix incorrect update of ppdu stats causing incorrect radiotap
fields.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403195738.25367-3-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath11k/hal_rx.c | 4 ++--
 drivers/net/wireless/ath/ath11k/hal_rx.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.c b/drivers/net/wireless/ath/ath11k/hal_rx.c
index b9939c5a4ecd..bb1d40034aa8 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.c
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.c
@@ -1029,7 +1029,7 @@ ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
 		info1 = __le32_to_cpu(vht_sig->info1);
 
 		ppdu_info->ldpc = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_SU_MU_CODING,
-					    info0);
+					    info1);
 		ppdu_info->mcs = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_MCS,
 					   info1);
 		gi_setting = FIELD_GET(HAL_RX_VHT_SIG_A_INFO_INFO1_GI_SETTING,
@@ -1452,7 +1452,7 @@ ath11k_hal_rx_parse_mon_status_tlv(struct ath11k_base *ab,
 		 * PHYRX_OTHER_RECEIVE_INFO TLV.
 		 */
 		ppdu_info->rssi_comb =
-			FIELD_GET(HAL_RX_PHYRX_RSSI_LEGACY_INFO_INFO1_RSSI_COMB,
+			FIELD_GET(HAL_RX_PHYRX_RSSI_LEGACY_INFO_INFO0_RSSI_COMB,
 				  __le32_to_cpu(rssi->info0));
 
 		if (db2dbm) {
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.h b/drivers/net/wireless/ath/ath11k/hal_rx.h
index 47e8208b22e1..61bd8416c4fd 100644
--- a/drivers/net/wireless/ath/ath11k/hal_rx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_rx.h
@@ -385,7 +385,7 @@ struct hal_rx_he_sig_b2_ofdma_info {
 	__le32 info0;
 } __packed;
 
-#define HAL_RX_PHYRX_RSSI_LEGACY_INFO_INFO1_RSSI_COMB	GENMASK(15, 8)
+#define HAL_RX_PHYRX_RSSI_LEGACY_INFO_INFO0_RSSI_COMB	GENMASK(15, 8)
 
 #define HAL_RX_PHYRX_RSSI_PREAMBLE_PRI20	GENMASK(7, 0)
 
-- 
cgit v1.2.3-70-g09d2


From f9fff67d2d7ca6fa8066132003a3deef654c55b1 Mon Sep 17 00:00:00 2001
From: Nagarajan Maran <quic_nmaran@quicinc.com>
Date: Mon, 17 Apr 2023 13:35:02 +0300
Subject: wifi: ath11k: Fix SKB corruption in REO destination ring

While running traffics for a long time, randomly an RX descriptor
filled with value "0" from REO destination ring is received.
This descriptor which is invalid causes the wrong SKB (SKB stored in
the IDR lookup with buffer id "0") to be fetched which in turn
causes SKB memory corruption issue and the same leads to crash
after some time.

Changed the start id for idr allocation to "1" and the buffer id "0"
is reserved for error validation. Introduced Sanity check to validate
the descriptor, before processing the SKB.

Crash Signature :

Unable to handle kernel paging request at virtual address 3f004900
PC points to "b15_dma_inv_range+0x30/0x50"
LR points to "dma_cache_maint_page+0x8c/0x128".
The Backtrace obtained is as follows:
[<8031716c>] (b15_dma_inv_range) from [<80313a4c>] (dma_cache_maint_page+0x8c/0x128)
[<80313a4c>] (dma_cache_maint_page) from [<80313b90>] (__dma_page_dev_to_cpu+0x28/0xcc)
[<80313b90>] (__dma_page_dev_to_cpu) from [<7fb5dd68>] (ath11k_dp_process_rx+0x1e8/0x4a4 [ath11k])
[<7fb5dd68>] (ath11k_dp_process_rx [ath11k]) from [<7fb53c20>] (ath11k_dp_service_srng+0xb0/0x2ac [ath11k])
[<7fb53c20>] (ath11k_dp_service_srng [ath11k]) from [<7f67bba4>] (ath11k_pci_ext_grp_napi_poll+0x1c/0x78 [ath11k_pci])
[<7f67bba4>] (ath11k_pci_ext_grp_napi_poll [ath11k_pci]) from [<807d5cf4>] (__napi_poll+0x28/0xb8)
[<807d5cf4>] (__napi_poll) from [<807d5f28>] (net_rx_action+0xf0/0x280)
[<807d5f28>] (net_rx_action) from [<80302148>] (__do_softirq+0xd0/0x280)
[<80302148>] (__do_softirq) from [<80320408>] (irq_exit+0x74/0xd4)
[<80320408>] (irq_exit) from [<803638a4>] (__handle_domain_irq+0x90/0xb4)
[<803638a4>] (__handle_domain_irq) from [<805bedec>] (gic_handle_irq+0x58/0x90)
[<805bedec>] (gic_handle_irq) from [<80301a78>] (__irq_svc+0x58/0x8c)

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Nagarajan Maran <quic_nmaran@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230403191533.28114-1-quic_nmaran@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index a4934bd79969..f67ce62b2b48 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -389,10 +389,10 @@ int ath11k_dp_rxbufs_replenish(struct ath11k_base *ab, int mac_id,
 			goto fail_free_skb;
 
 		spin_lock_bh(&rx_ring->idr_lock);
-		buf_id = idr_alloc(&rx_ring->bufs_idr, skb, 0,
-				   rx_ring->bufs_max * 3, GFP_ATOMIC);
+		buf_id = idr_alloc(&rx_ring->bufs_idr, skb, 1,
+				   (rx_ring->bufs_max * 3) + 1, GFP_ATOMIC);
 		spin_unlock_bh(&rx_ring->idr_lock);
-		if (buf_id < 0)
+		if (buf_id <= 0)
 			goto fail_dma_unmap;
 
 		desc = ath11k_hal_srng_src_get_next_entry(ab, srng);
@@ -2665,6 +2665,9 @@ try_again:
 				   cookie);
 		mac_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_PDEV_ID, cookie);
 
+		if (unlikely(buf_id == 0))
+			continue;
+
 		ar = ab->pdevs[mac_id].ar;
 		rx_ring = &ar->dp.rx_refill_buf_ring;
 		spin_lock_bh(&rx_ring->idr_lock);
-- 
cgit v1.2.3-70-g09d2


From b956e3110a797a3663f91f9b8935b667cc23fe72 Mon Sep 17 00:00:00 2001
From: Toke Høiland-Jørgensen <toke@toke.dk>
Date: Mon, 17 Apr 2023 13:35:03 +0300
Subject: wifi: ath9k: Don't mark channelmap stack variable read-only in
 ath9k_mci_update_wlan_channels()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This partially reverts commit e161d4b60ae3a5356e07202e0bfedb5fad82c6aa.

Turns out the channelmap variable is not actually read-only, it's modified
through the MCI_GPM_CLR_CHANNEL_BIT() macro further down in the function,
so making it read-only causes page faults when that code is hit.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=217183
Fixes: e161d4b60ae3 ("wifi: ath9k: Make arrays prof_prio and channelmap static const")
Cc: stable@vger.kernel.org
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230413214118.153781-1-toke@toke.dk
---
 drivers/net/wireless/ath/ath9k/mci.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/mci.c b/drivers/net/wireless/ath/ath9k/mci.c
index 3363fc4e8966..a0845002d6fe 100644
--- a/drivers/net/wireless/ath/ath9k/mci.c
+++ b/drivers/net/wireless/ath/ath9k/mci.c
@@ -646,9 +646,7 @@ void ath9k_mci_update_wlan_channels(struct ath_softc *sc, bool allow_all)
 	struct ath_hw *ah = sc->sc_ah;
 	struct ath9k_hw_mci *mci = &ah->btcoex_hw.mci;
 	struct ath9k_channel *chan = ah->curchan;
-	static const u32 channelmap[] = {
-		0x00000000, 0xffff0000, 0xffffffff, 0x7fffffff
-	};
+	u32 channelmap[] = {0x00000000, 0xffff0000, 0xffffffff, 0x7fffffff};
 	int i;
 	s16 chan_start, chan_end;
 	u16 wlan_chan;
-- 
cgit v1.2.3-70-g09d2


From b100722a777f6455d913666a376f81342b2cb995 Mon Sep 17 00:00:00 2001
From: Muna Sinada <quic_msinada@quicinc.com>
Date: Mon, 17 Apr 2023 13:22:27 -0700
Subject: wifi: ath11k: Remove disabling of 80+80 and 160 MHz

This is a regression fix for 80+80 and 160 MHz support bits being
cleared, therefore not adverised. Remove disable of 80+80 and 160 MHz
capability flags and assign valid center frequency 2 similar to
VHT80_80.

Fixes: 38dfe775d0ab ("wifi: ath11k: push MU-MIMO params from hostapd to hardware")
Reported-by: Robert Marko <robert.marko@sartura.hr>
Tested-by: Robert Marko <robert.marko@sartura.hr> # IPQ8074 WLAN.HK.2.9.0.1-01385-QCAHKSWPL_SILICONZ-1
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217299
Co-developed-by: P Praneesh <quic_ppranees@quicinc.com>
Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Signed-off-by: Muna Sinada <quic_msinada@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/1681762947-13882-1-git-send-email-quic_msinada@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 4 ----
 drivers/net/wireless/ath/ath11k/wmi.c | 3 ++-
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 4447fd4bd68f..1c93f1afccc5 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -5585,10 +5585,6 @@ static int ath11k_mac_copy_he_cap(struct ath11k *ar,
 
 		he_cap_elem->mac_cap_info[1] &=
 			IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_MASK;
-		he_cap_elem->phy_cap_info[0] &=
-			~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G;
-		he_cap_elem->phy_cap_info[0] &=
-			~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G;
 
 		he_cap_elem->phy_cap_info[5] &=
 			~IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_MASK;
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index f338d4447f11..d0b59bc2905a 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -871,7 +871,8 @@ static void ath11k_wmi_put_wmi_channel(struct wmi_channel *chan,
 
 		chan->band_center_freq2 = arg->channel.band_center_freq1;
 
-	} else if (arg->channel.mode == MODE_11AC_VHT80_80) {
+	} else if ((arg->channel.mode == MODE_11AC_VHT80_80) ||
+		   (arg->channel.mode == MODE_11AX_HE80_80)) {
 		chan->band_center_freq2 = arg->channel.band_center_freq2;
 	} else {
 		chan->band_center_freq2 = 0;
-- 
cgit v1.2.3-70-g09d2


From 2ddade322925641ee2a75f13665c51f2e74d7791 Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Tue, 18 Apr 2023 16:36:17 +0200
Subject: selftests/xsk: Fix munmap for hugepage allocated umem

Fix the unmapping of hugepage allocated umems so that they are
properly unmapped. The new test referred to in the fixes label,
introduced a test that allocated a umem that is not a multiple of a 2M
hugepage size. This is fine for mmap() that rounds the size up the
nearest multiple of 2M. But munmap() requires the size to be a
multiple of the hugepage size in order for it to unmap the region. The
current behaviour of not properly unmapping the umem, was discovered
when further additions of tests that require hugepages (unaligned mode
tests only) started failing as the system was running out of
hugepages.

Fixes: c0801598e543 ("selftests: xsk: Add test UNALIGNED_INV_DESC_4K1_FRAME_SIZE")
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230418143617.27762-1-magnus.karlsson@gmail.com
---
 tools/testing/selftests/bpf/xskxceiver.c | 19 +++++++++++++++----
 tools/testing/selftests/bpf/xskxceiver.h |  1 +
 2 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index 5a9691e942de..a59d04118842 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -77,6 +77,7 @@
 #include <linux/if_link.h>
 #include <linux/if_ether.h>
 #include <linux/ip.h>
+#include <linux/mman.h>
 #include <linux/udp.h>
 #include <arpa/inet.h>
 #include <net/if.h>
@@ -1286,16 +1287,19 @@ static void thread_common_ops(struct test_spec *test, struct ifobject *ifobject)
 	u64 umem_sz = ifobject->umem->num_frames * ifobject->umem->frame_size;
 	int mmap_flags = MAP_PRIVATE | MAP_ANONYMOUS | MAP_NORESERVE;
 	LIBBPF_OPTS(bpf_xdp_query_opts, opts);
+	off_t mmap_offset = 0;
 	void *bufs;
 	int ret;
 
-	if (ifobject->umem->unaligned_mode)
+	if (ifobject->umem->unaligned_mode) {
 		mmap_flags |= MAP_HUGETLB;
+		mmap_offset = MAP_HUGE_2MB;
+	}
 
 	if (ifobject->shared_umem)
 		umem_sz *= 2;
 
-	bufs = mmap(NULL, umem_sz, PROT_READ | PROT_WRITE, mmap_flags, -1, 0);
+	bufs = mmap(NULL, umem_sz, PROT_READ | PROT_WRITE, mmap_flags, -1, mmap_offset);
 	if (bufs == MAP_FAILED)
 		exit_with_error(errno);
 
@@ -1379,6 +1383,11 @@ static void *worker_testapp_validate_rx(void *arg)
 	pthread_exit(NULL);
 }
 
+static u64 ceil_u64(u64 a, u64 b)
+{
+	return (a + b - 1) / b;
+}
+
 static void testapp_clean_xsk_umem(struct ifobject *ifobj)
 {
 	u64 umem_sz = ifobj->umem->num_frames * ifobj->umem->frame_size;
@@ -1386,6 +1395,7 @@ static void testapp_clean_xsk_umem(struct ifobject *ifobj)
 	if (ifobj->shared_umem)
 		umem_sz *= 2;
 
+	umem_sz = ceil_u64(umem_sz, HUGEPAGE_SIZE) * HUGEPAGE_SIZE;
 	xsk_umem__delete(ifobj->umem->umem);
 	munmap(ifobj->umem->buffer, umem_sz);
 }
@@ -1619,14 +1629,15 @@ static void testapp_stats_fill_empty(struct test_spec *test)
 /* Simple test */
 static bool hugepages_present(struct ifobject *ifobject)
 {
-	const size_t mmap_sz = 2 * ifobject->umem->num_frames * ifobject->umem->frame_size;
+	size_t mmap_sz = 2 * ifobject->umem->num_frames * ifobject->umem->frame_size;
 	void *bufs;
 
 	bufs = mmap(NULL, mmap_sz, PROT_READ | PROT_WRITE,
-		    MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, -1, 0);
+		    MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, -1, MAP_HUGE_2MB);
 	if (bufs == MAP_FAILED)
 		return false;
 
+	mmap_sz = ceil_u64(mmap_sz, HUGEPAGE_SIZE) * HUGEPAGE_SIZE;
 	munmap(bufs, mmap_sz);
 	return true;
 }
diff --git a/tools/testing/selftests/bpf/xskxceiver.h b/tools/testing/selftests/bpf/xskxceiver.h
index 919327807a4e..c535aeab2ca3 100644
--- a/tools/testing/selftests/bpf/xskxceiver.h
+++ b/tools/testing/selftests/bpf/xskxceiver.h
@@ -56,6 +56,7 @@
 #define RX_FULL_RXQSIZE 32
 #define UMEM_HEADROOM_TEST_SIZE 128
 #define XSK_UMEM__INVALID_FRAME_SIZE (XSK_UMEM__DEFAULT_FRAME_SIZE + 1)
+#define HUGEPAGE_SIZE (2 * 1024 * 1024)
 
 #define print_verbose(x...) do { if (opt_verbose) ksft_print_msg(x); } while (0)
 
-- 
cgit v1.2.3-70-g09d2


From 8e4c62c7d980eaf0f64c1c0ef0c80f5685af0fb6 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 17 Apr 2023 08:28:05 -0700
Subject: page_pool: add DMA_ATTR_WEAK_ORDERING on all mappings

Commit c519fe9a4f0d ("bnxt: add dma mapping attributes") added
DMA_ATTR_WEAK_ORDERING to DMA attrs on bnxt. It has since spread
to a few more drivers (possibly as a copy'n'paste).

DMA_ATTR_WEAK_ORDERING only seems to matter on Sparc and PowerPC/cell,
the rarity of these platforms is likely why we never bothered adding
the attribute in the page pool, even though it should be safe to add.

To make the page pool migration in drivers which set this flag less
of a risk (of regressing the precious sparc database workloads or
whatever needed this) let's add DMA_ATTR_WEAK_ORDERING on all
page pool DMA mappings.

We could make this a driver opt-in but frankly I don't think it's
worth complicating the API. I can't think of a reason why device
accesses to packet memory would have to be ordered.

Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/r/20230417152805.331865-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/page_pool.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 2f6bf422ed30..97f20f7ff4fc 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -316,7 +316,8 @@ static bool page_pool_dma_map(struct page_pool *pool, struct page *page)
 	 */
 	dma = dma_map_page_attrs(pool->p.dev, page, 0,
 				 (PAGE_SIZE << pool->p.order),
-				 pool->p.dma_dir, DMA_ATTR_SKIP_CPU_SYNC);
+				 pool->p.dma_dir, DMA_ATTR_SKIP_CPU_SYNC |
+						  DMA_ATTR_WEAK_ORDERING);
 	if (dma_mapping_error(pool->p.dev, dma))
 		return false;
 
@@ -484,7 +485,7 @@ void page_pool_release_page(struct page_pool *pool, struct page *page)
 	/* When page is unmapped, it cannot be returned to our pool */
 	dma_unmap_page_attrs(pool->p.dev, dma,
 			     PAGE_SIZE << pool->p.order, pool->p.dma_dir,
-			     DMA_ATTR_SKIP_CPU_SYNC);
+			     DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
 	page_pool_set_dma_addr(page, 0);
 skip_dma_unmap:
 	page_pool_clear_pp_info(page);
-- 
cgit v1.2.3-70-g09d2


From 91daa4f62ce878b6e1ac5908aceb83550332447f Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sun, 16 Apr 2023 13:08:14 +0100
Subject: net: dsa: mt7530: fix support for MT7531BE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There are two variants of the MT7531 switch IC which got different
features (and pins) regarding port 5:
 * MT7531AE: SGMII/1000Base-X/2500Base-X SerDes PCS
 * MT7531BE: RGMII

Moving the creation of the SerDes PCS from mt753x_setup to mt7530_probe
with commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation
to mt7530_probe function") works fine for MT7531AE which got two
instances of mtk-pcs-lynxi, however, MT7531BE requires mt7531_pll_setup
to setup clocks before the single PCS on port 6 (usually used as CPU
port) starts to work and hence the PCS creation failed on MT7531BE.

Fix this by introducing a pointer to mt7531_create_sgmii function in
struct mt7530_priv and call it again at the end of mt753x_setup like it
was before commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS
creation to mt7530_probe function").

Fixes: 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to mt7530_probe function")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Acked-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Link: https://lore.kernel.org/r/ZDvlLhhqheobUvOK@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/mt7530-mdio.c | 16 ++++++++--------
 drivers/net/dsa/mt7530.c      |  6 ++++++
 drivers/net/dsa/mt7530.h      |  4 ++--
 3 files changed, 16 insertions(+), 10 deletions(-)

diff --git a/drivers/net/dsa/mt7530-mdio.c b/drivers/net/dsa/mt7530-mdio.c
index 34a547b88e49..088533663b83 100644
--- a/drivers/net/dsa/mt7530-mdio.c
+++ b/drivers/net/dsa/mt7530-mdio.c
@@ -81,14 +81,17 @@ static const struct regmap_bus mt7530_regmap_bus = {
 };
 
 static int
-mt7531_create_sgmii(struct mt7530_priv *priv)
+mt7531_create_sgmii(struct mt7530_priv *priv, bool dual_sgmii)
 {
-	struct regmap_config *mt7531_pcs_config[2];
+	struct regmap_config *mt7531_pcs_config[2] = {};
 	struct phylink_pcs *pcs;
 	struct regmap *regmap;
 	int i, ret = 0;
 
-	for (i = 0; i < 2; i++) {
+	/* MT7531AE has two SGMII units for port 5 and port 6
+	 * MT7531BE has only one SGMII unit for port 6
+	 */
+	for (i = dual_sgmii ? 0 : 1; i < 2; i++) {
 		mt7531_pcs_config[i] = devm_kzalloc(priv->dev,
 						    sizeof(struct regmap_config),
 						    GFP_KERNEL);
@@ -208,11 +211,8 @@ mt7530_probe(struct mdio_device *mdiodev)
 	if (IS_ERR(priv->regmap))
 		return PTR_ERR(priv->regmap);
 
-	if (priv->id == ID_MT7531) {
-		ret = mt7531_create_sgmii(priv);
-		if (ret)
-			return ret;
-	}
+	if (priv->id == ID_MT7531)
+		priv->create_sgmii = mt7531_create_sgmii;
 
 	return dsa_register_switch(priv->ds);
 }
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index e4bb5037d352..c680873819b0 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -3018,6 +3018,12 @@ mt753x_setup(struct dsa_switch *ds)
 	if (ret && priv->irq)
 		mt7530_free_irq_common(priv);
 
+	if (priv->create_sgmii) {
+		ret = priv->create_sgmii(priv, mt7531_dual_sgmii_supported(priv));
+		if (ret && priv->irq)
+			mt7530_free_irq(priv);
+	}
+
 	return ret;
 }
 
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index 01db5c9724fa..5084f48a8869 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -748,10 +748,10 @@ struct mt753x_info {
  *			registers
  * @p6_interface	Holding the current port 6 interface
  * @p5_intf_sel:	Holding the current port 5 interface select
- *
  * @irq:		IRQ number of the switch
  * @irq_domain:		IRQ domain of the switch irq_chip
  * @irq_enable:		IRQ enable bits, synced to SYS_INT_EN
+ * @create_sgmii:	Pointer to function creating SGMII PCS instance(s)
  */
 struct mt7530_priv {
 	struct device		*dev;
@@ -770,7 +770,6 @@ struct mt7530_priv {
 	unsigned int		p5_intf_sel;
 	u8			mirror_rx;
 	u8			mirror_tx;
-
 	struct mt7530_port	ports[MT7530_NUM_PORTS];
 	struct mt753x_pcs	pcs[MT7530_NUM_PORTS];
 	/* protect among processes for registers access*/
@@ -778,6 +777,7 @@ struct mt7530_priv {
 	int irq;
 	struct irq_domain *irq_domain;
 	u32 irq_enable;
+	int (*create_sgmii)(struct mt7530_priv *priv, bool dual_sgmii);
 };
 
 struct mt7530_hw_vlan_entry {
-- 
cgit v1.2.3-70-g09d2


From 43bb6100d8d5871d211196034042308c1b1d69c3 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Tue, 18 Apr 2023 13:07:33 +0200
Subject: net: stmmac: dwmac-meson8b: Avoid cast to incompatible function type

Rather than casting clk_disable_unprepare to an incompatible function
type provide a trivial wrapper with the correct signature for the
use-case.

Reported by clang-16 with W=1:

 drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c:276:6: error: cast from 'void (*)(struct clk *)' to 'void (*)(void *)' converts to incompatible function type [-Werror,-Wcast-function-type-strict]
                                        (void(*)(void *))clk_disable_unprepare,
                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
No functional change intended.
Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Link: https://lore.kernel.org/r/20230418-dwmac-meson8b-clk-cb-cast-v1-1-e892b670cbbb@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c
index e8b507f88fbc..f6754e3643f3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c
@@ -263,6 +263,11 @@ static int meson_axg_set_phy_mode(struct meson8b_dwmac *dwmac)
 	return 0;
 }
 
+static void meson8b_clk_disable_unprepare(void *data)
+{
+	clk_disable_unprepare(data);
+}
+
 static int meson8b_devm_clk_prepare_enable(struct meson8b_dwmac *dwmac,
 					   struct clk *clk)
 {
@@ -273,8 +278,7 @@ static int meson8b_devm_clk_prepare_enable(struct meson8b_dwmac *dwmac,
 		return ret;
 
 	return devm_add_action_or_reset(dwmac->dev,
-					(void(*)(void *))clk_disable_unprepare,
-					clk);
+					meson8b_clk_disable_unprepare, clk);
 }
 
 static int meson8b_init_rgmii_delays(struct meson8b_dwmac *dwmac)
-- 
cgit v1.2.3-70-g09d2


From 3e9c0700bf42b6170ae6b6e616cdc2e75a21b506 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Tue, 18 Apr 2023 14:47:13 +0200
Subject: net: micrel: Update the list of supported phys

At the beginning of the file micrel.c there is list of supported PHYs.
Extend this list with the following PHYs lan8841, lan8814 and lan8804,
as these PHYs were added but the list was not updated.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230418124713.2221451-1-horatiu.vultur@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/micrel.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 602e5a46b796..3f81bb8dac44 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -10,13 +10,13 @@
  * Copyright (c) 2014 Johan Hovold <johan@kernel.org>
  *
  * Support : Micrel Phys:
- *		Giga phys: ksz9021, ksz9031, ksz9131
+ *		Giga phys: ksz9021, ksz9031, ksz9131, lan8841, lan8814
  *		100/10 Phys : ksz8001, ksz8721, ksz8737, ksz8041
  *			   ksz8021, ksz8031, ksz8051,
  *			   ksz8081, ksz8091,
  *			   ksz8061,
  *		Switch : ksz8873, ksz886x
- *			 ksz9477
+ *			 ksz9477, lan8804
  */
 
 #include <linux/bitfield.h>
-- 
cgit v1.2.3-70-g09d2


From 84ce730f82dfe62f6f3e76b059ce0f7178322fb9 Mon Sep 17 00:00:00 2001
From: Rob Herring <robh@kernel.org>
Date: Tue, 18 Apr 2023 10:06:27 -0500
Subject: dt-bindings: net: ethernet: Fix JSON pointer references

A JSON pointer reference (the part after the "#") must start with a "/".
Conversely, references to the entire document must not have a trailing "/"
and should be just a "#". The existing jsonschema package allows these,
but coming changes make allowed "$ref" URIs stricter and throw errors on
these references.

Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20230418150628.1528480-1-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/ethernet-controller.yaml | 2 +-
 Documentation/devicetree/bindings/net/ethernet-switch.yaml     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/ethernet-controller.yaml b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
index ebc2646ab5ff..6b0d359367da 100644
--- a/Documentation/devicetree/bindings/net/ethernet-controller.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
@@ -205,7 +205,7 @@ properties:
               duplex is assumed.
 
           pause:
-            $ref: /schemas/types.yaml#definitions/flag
+            $ref: /schemas/types.yaml#/definitions/flag
             description:
               Indicates that pause should be enabled.
 
diff --git a/Documentation/devicetree/bindings/net/ethernet-switch.yaml b/Documentation/devicetree/bindings/net/ethernet-switch.yaml
index 2ceccce6cbd7..f1b9075dc7fb 100644
--- a/Documentation/devicetree/bindings/net/ethernet-switch.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-switch.yaml
@@ -55,7 +55,7 @@ additionalProperties: true
 $defs:
   base:
     description: An ethernet switch without any extra port properties
-    $ref: '#/'
+    $ref: '#'
 
     patternProperties:
       "^(ethernet-)?port@[0-9]+$":
-- 
cgit v1.2.3-70-g09d2


From 2bc42f482bedeafdbe70804fd82bdbf8f7e28b00 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 17 Apr 2023 10:32:19 -0400
Subject: .gitignore: Do not ignore .kunitconfig files

Circumvent the .gitignore wildcard to avoid warnings about ignored
.kunitconfig files. As far as I can tell, the warnings are harmless
and these files are not actually ignored.

Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304142337.jc4oUrov-lkp@intel.com/
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .gitignore | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.gitignore b/.gitignore
index 70ec6037fa7a..51117ba29c88 100644
--- a/.gitignore
+++ b/.gitignore
@@ -105,6 +105,7 @@ modules.order
 !.gitignore
 !.mailmap
 !.rustfmt.toml
+!.kunitconfig
 
 #
 # Generated include files
-- 
cgit v1.2.3-70-g09d2


From 3b3009ea8abb713b022d94fba95ec270cf6e7eae Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 17 Apr 2023 10:32:26 -0400
Subject: net/handshake: Create a NETLINK service for handling handshake
 requests

When a kernel consumer needs a transport layer security session, it
first needs a handshake to negotiate and establish a session. This
negotiation can be done in user space via one of the several
existing library implementations, or it can be done in the kernel.

No in-kernel handshake implementations yet exist. In their absence,
we add a netlink service that can:

a. Notify a user space daemon that a handshake is needed.

b. Once notified, the daemon calls the kernel back via this
   netlink service to get the handshake parameters, including an
   open socket on which to establish the session.

c. Once the handshake is complete, the daemon reports the
   session status and other information via a second netlink
   operation. This operation marks that it is safe for the
   kernel to use the open socket and the security session
   established there.

The notification service uses a multicast group. Each handshake
mechanism (eg, tlshd) adopts its own group number so that the
handshake services are completely independent of one another. The
kernel can then tell via netlink_has_listeners() whether a handshake
service is active and prepared to handle a handshake request.

A new netlink operation, ACCEPT, acts like accept(2) in that it
instantiates a file descriptor in the user space daemon's fd table.
If this operation is successful, the reply carries the fd number,
which can be treated as an open and ready file descriptor.

While user space is performing the handshake, the kernel keeps its
muddy paws off the open socket. A second new netlink operation,
DONE, indicates that the user space daemon is finished with the
socket and it is safe for the kernel to use again. The operation
also indicates whether a session was established successfully.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/handshake.yaml | 122 +++++++++++
 MAINTAINERS                                |   9 +
 include/trace/events/handshake.h           | 159 ++++++++++++++
 include/uapi/linux/handshake.h             |  71 ++++++
 net/Kconfig                                |   5 +
 net/Makefile                               |   1 +
 net/handshake/Makefile                     |  11 +
 net/handshake/genl.c                       |  57 +++++
 net/handshake/genl.h                       |  23 ++
 net/handshake/handshake.h                  |  82 +++++++
 net/handshake/netlink.c                    | 312 ++++++++++++++++++++++++++
 net/handshake/request.c                    | 339 +++++++++++++++++++++++++++++
 net/handshake/trace.c                      |  20 ++
 13 files changed, 1211 insertions(+)
 create mode 100644 Documentation/netlink/specs/handshake.yaml
 create mode 100644 include/trace/events/handshake.h
 create mode 100644 include/uapi/linux/handshake.h
 create mode 100644 net/handshake/Makefile
 create mode 100644 net/handshake/genl.c
 create mode 100644 net/handshake/genl.h
 create mode 100644 net/handshake/handshake.h
 create mode 100644 net/handshake/netlink.c
 create mode 100644 net/handshake/request.c
 create mode 100644 net/handshake/trace.c

diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml
new file mode 100644
index 000000000000..0333d92b1438
--- /dev/null
+++ b/Documentation/netlink/specs/handshake.yaml
@@ -0,0 +1,122 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+#
+# Author: Chuck Lever <chuck.lever@oracle.com>
+#
+# Copyright (c) 2023, Oracle and/or its affiliates.
+#
+
+name: handshake
+
+protocol: genetlink
+
+doc: Netlink protocol to request a transport layer security handshake.
+
+definitions:
+  -
+    type: enum
+    name: handler-class
+    value-start: 0
+    entries: [ none, max ]
+  -
+    type: enum
+    name: msg-type
+    value-start: 0
+    entries: [ unspec, clienthello, serverhello ]
+  -
+    type: enum
+    name: auth
+    value-start: 0
+    entries: [ unspec, unauth, psk, x509 ]
+
+attribute-sets:
+  -
+    name: x509
+    attributes:
+      -
+        name: cert
+        type: u32
+      -
+        name: privkey
+        type: u32
+  -
+    name: accept
+    attributes:
+      -
+        name: sockfd
+        type: u32
+      -
+        name: handler-class
+        type: u32
+        enum: handler-class
+      -
+        name: message-type
+        type: u32
+        enum: msg-type
+      -
+        name: timeout
+        type: u32
+      -
+        name: auth-mode
+        type: u32
+        enum: auth
+      -
+        name: peer-identity
+        type: u32
+        multi-attr: true
+      -
+        name: certificate
+        type: nest
+        nested-attributes: x509
+        multi-attr: true
+  -
+    name: done
+    attributes:
+      -
+        name: status
+        type: u32
+      -
+        name: sockfd
+        type: u32
+      -
+        name: remote-auth
+        type: u32
+        multi-attr: true
+
+operations:
+  list:
+    -
+      name: ready
+      doc: Notify handlers that a new handshake request is waiting
+      notify: accept
+    -
+      name: accept
+      doc: Handler retrieves next queued handshake request
+      attribute-set: accept
+      flags: [ admin-perm ]
+      do:
+        request:
+          attributes:
+            - handler-class
+        reply:
+          attributes:
+            - sockfd
+            - message-type
+            - timeout
+            - auth-mode
+            - peer-identity
+            - certificate
+    -
+      name: done
+      doc: Handler reports handshake completion
+      attribute-set: done
+      do:
+        request:
+          attributes:
+            - status
+            - sockfd
+            - remote-auth
+
+mcast-groups:
+  list:
+    -
+      name: none
diff --git a/MAINTAINERS b/MAINTAINERS
index 4fc57dfd5fd0..cdc7748d15b8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8947,6 +8947,15 @@ Q:	http://patchwork.linuxtv.org/project/linux-media/list/
 T:	git git://linuxtv.org/anttip/media_tree.git
 F:	drivers/media/usb/hackrf/
 
+HANDSHAKE UPCALL FOR TRANSPORT LAYER SECURITY
+M:	Chuck Lever <chuck.lever@oracle.com>
+L:	kernel-tls-handshake@lists.linux.dev
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	Documentation/netlink/specs/handshake.yaml
+F:	include/trace/events/handshake.h
+F:	net/handshake/
+
 HANTRO VPU CODEC DRIVER
 M:	Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
 M:	Philipp Zabel <p.zabel@pengutronix.de>
diff --git a/include/trace/events/handshake.h b/include/trace/events/handshake.h
new file mode 100644
index 000000000000..8dadcab5f12a
--- /dev/null
+++ b/include/trace/events/handshake.h
@@ -0,0 +1,159 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM handshake
+
+#if !defined(_TRACE_HANDSHAKE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_HANDSHAKE_H
+
+#include <linux/net.h>
+#include <linux/tracepoint.h>
+
+DECLARE_EVENT_CLASS(handshake_event_class,
+	TP_PROTO(
+		const struct net *net,
+		const struct handshake_req *req,
+		const struct sock *sk
+	),
+	TP_ARGS(net, req, sk),
+	TP_STRUCT__entry(
+		__field(const void *, req)
+		__field(const void *, sk)
+		__field(unsigned int, netns_ino)
+	),
+	TP_fast_assign(
+		__entry->req = req;
+		__entry->sk = sk;
+		__entry->netns_ino = net->ns.inum;
+	),
+	TP_printk("req=%p sk=%p",
+		__entry->req, __entry->sk
+	)
+);
+#define DEFINE_HANDSHAKE_EVENT(name)				\
+	DEFINE_EVENT(handshake_event_class, name,		\
+		TP_PROTO(					\
+			const struct net *net,			\
+			const struct handshake_req *req,	\
+			const struct sock *sk			\
+		),						\
+		TP_ARGS(net, req, sk))
+
+DECLARE_EVENT_CLASS(handshake_fd_class,
+	TP_PROTO(
+		const struct net *net,
+		const struct handshake_req *req,
+		const struct sock *sk,
+		int fd
+	),
+	TP_ARGS(net, req, sk, fd),
+	TP_STRUCT__entry(
+		__field(const void *, req)
+		__field(const void *, sk)
+		__field(int, fd)
+		__field(unsigned int, netns_ino)
+	),
+	TP_fast_assign(
+		__entry->req = req;
+		__entry->sk = req->hr_sk;
+		__entry->fd = fd;
+		__entry->netns_ino = net->ns.inum;
+	),
+	TP_printk("req=%p sk=%p fd=%d",
+		__entry->req, __entry->sk, __entry->fd
+	)
+);
+#define DEFINE_HANDSHAKE_FD_EVENT(name)				\
+	DEFINE_EVENT(handshake_fd_class, name,			\
+		TP_PROTO(					\
+			const struct net *net,			\
+			const struct handshake_req *req,	\
+			const struct sock *sk,			\
+			int fd					\
+		),						\
+		TP_ARGS(net, req, sk, fd))
+
+DECLARE_EVENT_CLASS(handshake_error_class,
+	TP_PROTO(
+		const struct net *net,
+		const struct handshake_req *req,
+		const struct sock *sk,
+		int err
+	),
+	TP_ARGS(net, req, sk, err),
+	TP_STRUCT__entry(
+		__field(const void *, req)
+		__field(const void *, sk)
+		__field(int, err)
+		__field(unsigned int, netns_ino)
+	),
+	TP_fast_assign(
+		__entry->req = req;
+		__entry->sk = sk;
+		__entry->err = err;
+		__entry->netns_ino = net->ns.inum;
+	),
+	TP_printk("req=%p sk=%p err=%d",
+		__entry->req, __entry->sk, __entry->err
+	)
+);
+#define DEFINE_HANDSHAKE_ERROR(name)				\
+	DEFINE_EVENT(handshake_error_class, name,		\
+		TP_PROTO(					\
+			const struct net *net,			\
+			const struct handshake_req *req,	\
+			const struct sock *sk,			\
+			int err					\
+		),						\
+		TP_ARGS(net, req, sk, err))
+
+
+/*
+ * Request lifetime events
+ */
+
+DEFINE_HANDSHAKE_EVENT(handshake_submit);
+DEFINE_HANDSHAKE_ERROR(handshake_submit_err);
+DEFINE_HANDSHAKE_EVENT(handshake_cancel);
+DEFINE_HANDSHAKE_EVENT(handshake_cancel_none);
+DEFINE_HANDSHAKE_EVENT(handshake_cancel_busy);
+DEFINE_HANDSHAKE_EVENT(handshake_destruct);
+
+
+TRACE_EVENT(handshake_complete,
+	TP_PROTO(
+		const struct net *net,
+		const struct handshake_req *req,
+		const struct sock *sk,
+		int status
+	),
+	TP_ARGS(net, req, sk, status),
+	TP_STRUCT__entry(
+		__field(const void *, req)
+		__field(const void *, sk)
+		__field(int, status)
+		__field(unsigned int, netns_ino)
+	),
+	TP_fast_assign(
+		__entry->req = req;
+		__entry->sk = sk;
+		__entry->status = status;
+		__entry->netns_ino = net->ns.inum;
+	),
+	TP_printk("req=%p sk=%p status=%d",
+		__entry->req, __entry->sk, __entry->status
+	)
+);
+
+/*
+ * Netlink events
+ */
+
+DEFINE_HANDSHAKE_ERROR(handshake_notify_err);
+DEFINE_HANDSHAKE_FD_EVENT(handshake_cmd_accept);
+DEFINE_HANDSHAKE_ERROR(handshake_cmd_accept_err);
+DEFINE_HANDSHAKE_FD_EVENT(handshake_cmd_done);
+DEFINE_HANDSHAKE_ERROR(handshake_cmd_done_err);
+
+#endif /* _TRACE_HANDSHAKE_H */
+
+#include <trace/define_trace.h>
diff --git a/include/uapi/linux/handshake.h b/include/uapi/linux/handshake.h
new file mode 100644
index 000000000000..7f66ff489b87
--- /dev/null
+++ b/include/uapi/linux/handshake.h
@@ -0,0 +1,71 @@
+/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */
+/* Do not edit directly, auto-generated from: */
+/*	Documentation/netlink/specs/handshake.yaml */
+/* YNL-GEN uapi header */
+
+#ifndef _UAPI_LINUX_HANDSHAKE_H
+#define _UAPI_LINUX_HANDSHAKE_H
+
+#define HANDSHAKE_FAMILY_NAME		"handshake"
+#define HANDSHAKE_FAMILY_VERSION	1
+
+enum handshake_handler_class {
+	HANDSHAKE_HANDLER_CLASS_NONE,
+	HANDSHAKE_HANDLER_CLASS_MAX,
+};
+
+enum handshake_msg_type {
+	HANDSHAKE_MSG_TYPE_UNSPEC,
+	HANDSHAKE_MSG_TYPE_CLIENTHELLO,
+	HANDSHAKE_MSG_TYPE_SERVERHELLO,
+};
+
+enum handshake_auth {
+	HANDSHAKE_AUTH_UNSPEC,
+	HANDSHAKE_AUTH_UNAUTH,
+	HANDSHAKE_AUTH_PSK,
+	HANDSHAKE_AUTH_X509,
+};
+
+enum {
+	HANDSHAKE_A_X509_CERT = 1,
+	HANDSHAKE_A_X509_PRIVKEY,
+
+	__HANDSHAKE_A_X509_MAX,
+	HANDSHAKE_A_X509_MAX = (__HANDSHAKE_A_X509_MAX - 1)
+};
+
+enum {
+	HANDSHAKE_A_ACCEPT_SOCKFD = 1,
+	HANDSHAKE_A_ACCEPT_HANDLER_CLASS,
+	HANDSHAKE_A_ACCEPT_MESSAGE_TYPE,
+	HANDSHAKE_A_ACCEPT_TIMEOUT,
+	HANDSHAKE_A_ACCEPT_AUTH_MODE,
+	HANDSHAKE_A_ACCEPT_PEER_IDENTITY,
+	HANDSHAKE_A_ACCEPT_CERTIFICATE,
+
+	__HANDSHAKE_A_ACCEPT_MAX,
+	HANDSHAKE_A_ACCEPT_MAX = (__HANDSHAKE_A_ACCEPT_MAX - 1)
+};
+
+enum {
+	HANDSHAKE_A_DONE_STATUS = 1,
+	HANDSHAKE_A_DONE_SOCKFD,
+	HANDSHAKE_A_DONE_REMOTE_AUTH,
+
+	__HANDSHAKE_A_DONE_MAX,
+	HANDSHAKE_A_DONE_MAX = (__HANDSHAKE_A_DONE_MAX - 1)
+};
+
+enum {
+	HANDSHAKE_CMD_READY = 1,
+	HANDSHAKE_CMD_ACCEPT,
+	HANDSHAKE_CMD_DONE,
+
+	__HANDSHAKE_CMD_MAX,
+	HANDSHAKE_CMD_MAX = (__HANDSHAKE_CMD_MAX - 1)
+};
+
+#define HANDSHAKE_MCGRP_NONE	"none"
+
+#endif /* _UAPI_LINUX_HANDSHAKE_H */
diff --git a/net/Kconfig b/net/Kconfig
index f806722bccf4..4b800706cc76 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -68,6 +68,11 @@ source "net/iucv/Kconfig"
 source "net/smc/Kconfig"
 source "net/xdp/Kconfig"
 
+config NET_HANDSHAKE
+	bool
+	depends on SUNRPC || NVME_TARGET_TCP || NVME_TCP
+	default y
+
 config INET
 	bool "TCP/IP networking"
 	help
diff --git a/net/Makefile b/net/Makefile
index 87592009366f..4c4dc535453d 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -79,3 +79,4 @@ obj-$(CONFIG_NET_NCSI)		+= ncsi/
 obj-$(CONFIG_XDP_SOCKETS)	+= xdp/
 obj-$(CONFIG_MPTCP)		+= mptcp/
 obj-$(CONFIG_MCTP)		+= mctp/
+obj-$(CONFIG_NET_HANDSHAKE)	+= handshake/
diff --git a/net/handshake/Makefile b/net/handshake/Makefile
new file mode 100644
index 000000000000..d38736de45da
--- /dev/null
+++ b/net/handshake/Makefile
@@ -0,0 +1,11 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Makefile for the Generic HANDSHAKE service
+#
+# Author: Chuck Lever <chuck.lever@oracle.com>
+#
+# Copyright (c) 2023, Oracle and/or its affiliates.
+#
+
+obj-y += handshake.o
+handshake-y := genl.o netlink.o request.o trace.o
diff --git a/net/handshake/genl.c b/net/handshake/genl.c
new file mode 100644
index 000000000000..652f37d19bd6
--- /dev/null
+++ b/net/handshake/genl.c
@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+/* Do not edit directly, auto-generated from: */
+/*	Documentation/netlink/specs/handshake.yaml */
+/* YNL-GEN kernel source */
+
+#include <net/netlink.h>
+#include <net/genetlink.h>
+
+#include "genl.h"
+
+#include <linux/handshake.h>
+
+/* HANDSHAKE_CMD_ACCEPT - do */
+static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HANDLER_CLASS + 1] = {
+	[HANDSHAKE_A_ACCEPT_HANDLER_CLASS] = NLA_POLICY_MAX(NLA_U32, 1),
+};
+
+/* HANDSHAKE_CMD_DONE - do */
+static const struct nla_policy handshake_done_nl_policy[HANDSHAKE_A_DONE_REMOTE_AUTH + 1] = {
+	[HANDSHAKE_A_DONE_STATUS] = { .type = NLA_U32, },
+	[HANDSHAKE_A_DONE_SOCKFD] = { .type = NLA_U32, },
+	[HANDSHAKE_A_DONE_REMOTE_AUTH] = { .type = NLA_U32, },
+};
+
+/* Ops table for handshake */
+static const struct genl_split_ops handshake_nl_ops[] = {
+	{
+		.cmd		= HANDSHAKE_CMD_ACCEPT,
+		.doit		= handshake_nl_accept_doit,
+		.policy		= handshake_accept_nl_policy,
+		.maxattr	= HANDSHAKE_A_ACCEPT_HANDLER_CLASS,
+		.flags		= GENL_ADMIN_PERM | GENL_CMD_CAP_DO,
+	},
+	{
+		.cmd		= HANDSHAKE_CMD_DONE,
+		.doit		= handshake_nl_done_doit,
+		.policy		= handshake_done_nl_policy,
+		.maxattr	= HANDSHAKE_A_DONE_REMOTE_AUTH,
+		.flags		= GENL_CMD_CAP_DO,
+	},
+};
+
+static const struct genl_multicast_group handshake_nl_mcgrps[] = {
+	[HANDSHAKE_NLGRP_NONE] = { "none", },
+};
+
+struct genl_family handshake_nl_family __ro_after_init = {
+	.name		= HANDSHAKE_FAMILY_NAME,
+	.version	= HANDSHAKE_FAMILY_VERSION,
+	.netnsok	= true,
+	.parallel_ops	= true,
+	.module		= THIS_MODULE,
+	.split_ops	= handshake_nl_ops,
+	.n_split_ops	= ARRAY_SIZE(handshake_nl_ops),
+	.mcgrps		= handshake_nl_mcgrps,
+	.n_mcgrps	= ARRAY_SIZE(handshake_nl_mcgrps),
+};
diff --git a/net/handshake/genl.h b/net/handshake/genl.h
new file mode 100644
index 000000000000..a1eb7ccccc7f
--- /dev/null
+++ b/net/handshake/genl.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */
+/* Do not edit directly, auto-generated from: */
+/*	Documentation/netlink/specs/handshake.yaml */
+/* YNL-GEN kernel header */
+
+#ifndef _LINUX_HANDSHAKE_GEN_H
+#define _LINUX_HANDSHAKE_GEN_H
+
+#include <net/netlink.h>
+#include <net/genetlink.h>
+
+#include <linux/handshake.h>
+
+int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info);
+int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info);
+
+enum {
+	HANDSHAKE_NLGRP_NONE,
+};
+
+extern struct genl_family handshake_nl_family;
+
+#endif /* _LINUX_HANDSHAKE_GEN_H */
diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h
new file mode 100644
index 000000000000..52568dbe24f1
--- /dev/null
+++ b/net/handshake/handshake.h
@@ -0,0 +1,82 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Generic netlink handshake service
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2023, Oracle and/or its affiliates.
+ */
+
+#ifndef _INTERNAL_HANDSHAKE_H
+#define _INTERNAL_HANDSHAKE_H
+
+/* Per-net namespace context */
+struct handshake_net {
+	spinlock_t		hn_lock;	/* protects next 3 fields */
+	int			hn_pending;
+	int			hn_pending_max;
+	struct list_head	hn_requests;
+
+	unsigned long		hn_flags;
+};
+
+enum hn_flags_bits {
+	HANDSHAKE_F_NET_DRAINING,
+};
+
+struct handshake_proto;
+
+/* One handshake request */
+struct handshake_req {
+	struct list_head		hr_list;
+	struct rhash_head		hr_rhash;
+	unsigned long			hr_flags;
+	const struct handshake_proto	*hr_proto;
+	struct sock			*hr_sk;
+	void				(*hr_odestruct)(struct sock *sk);
+
+	/* Always the last field */
+	char				hr_priv[];
+};
+
+enum hr_flags_bits {
+	HANDSHAKE_F_REQ_COMPLETED,
+};
+
+/* Invariants for all handshake requests for one transport layer
+ * security protocol
+ */
+struct handshake_proto {
+	int			hp_handler_class;
+	size_t			hp_privsize;
+
+	int			(*hp_accept)(struct handshake_req *req,
+					     struct genl_info *info, int fd);
+	void			(*hp_done)(struct handshake_req *req,
+					   unsigned int status,
+					   struct genl_info *info);
+	void			(*hp_destroy)(struct handshake_req *req);
+};
+
+/* netlink.c */
+int handshake_genl_notify(struct net *net, const struct handshake_proto *proto,
+			  gfp_t flags);
+struct nlmsghdr *handshake_genl_put(struct sk_buff *msg,
+				    struct genl_info *info);
+struct handshake_net *handshake_pernet(struct net *net);
+
+/* request.c */
+struct handshake_req *handshake_req_alloc(const struct handshake_proto *proto,
+					  gfp_t flags);
+int handshake_req_hash_init(void);
+void handshake_req_hash_destroy(void);
+void *handshake_req_private(struct handshake_req *req);
+struct handshake_req *handshake_req_hash_lookup(struct sock *sk);
+struct handshake_req *handshake_req_next(struct handshake_net *hn, int class);
+int handshake_req_submit(struct socket *sock, struct handshake_req *req,
+			 gfp_t flags);
+void handshake_complete(struct handshake_req *req, unsigned int status,
+			struct genl_info *info);
+bool handshake_req_cancel(struct sock *sk);
+
+#endif /* _INTERNAL_HANDSHAKE_H */
diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c
new file mode 100644
index 000000000000..7264cac04047
--- /dev/null
+++ b/net/handshake/netlink.c
@@ -0,0 +1,312 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Generic netlink handshake service
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2023, Oracle and/or its affiliates.
+ */
+
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/mm.h>
+
+#include <net/sock.h>
+#include <net/genetlink.h>
+#include <net/netns/generic.h>
+
+#include <uapi/linux/handshake.h>
+#include "handshake.h"
+#include "genl.h"
+
+#include <trace/events/handshake.h>
+
+/**
+ * handshake_genl_notify - Notify handlers that a request is waiting
+ * @net: target network namespace
+ * @proto: handshake protocol
+ * @flags: memory allocation control flags
+ *
+ * Returns zero on success or a negative errno if notification failed.
+ */
+int handshake_genl_notify(struct net *net, const struct handshake_proto *proto,
+			  gfp_t flags)
+{
+	struct sk_buff *msg;
+	void *hdr;
+
+	if (!genl_has_listeners(&handshake_nl_family, net,
+				proto->hp_handler_class))
+		return -ESRCH;
+
+	msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return -ENOMEM;
+
+	hdr = genlmsg_put(msg, 0, 0, &handshake_nl_family, 0,
+			  HANDSHAKE_CMD_READY);
+	if (!hdr)
+		goto out_free;
+
+	if (nla_put_u32(msg, HANDSHAKE_A_ACCEPT_HANDLER_CLASS,
+			proto->hp_handler_class) < 0) {
+		genlmsg_cancel(msg, hdr);
+		goto out_free;
+	}
+
+	genlmsg_end(msg, hdr);
+	return genlmsg_multicast_netns(&handshake_nl_family, net, msg,
+				       0, proto->hp_handler_class, flags);
+
+out_free:
+	nlmsg_free(msg);
+	return -EMSGSIZE;
+}
+
+/**
+ * handshake_genl_put - Create a generic netlink message header
+ * @msg: buffer in which to create the header
+ * @info: generic netlink message context
+ *
+ * Returns a ready-to-use header, or NULL.
+ */
+struct nlmsghdr *handshake_genl_put(struct sk_buff *msg,
+				    struct genl_info *info)
+{
+	return genlmsg_put(msg, info->snd_portid, info->snd_seq,
+			   &handshake_nl_family, 0, info->genlhdr->cmd);
+}
+EXPORT_SYMBOL(handshake_genl_put);
+
+/*
+ * dup() a kernel socket for use as a user space file descriptor
+ * in the current process. The kernel socket must have an
+ * instatiated struct file.
+ *
+ * Implicit argument: "current()"
+ */
+static int handshake_dup(struct socket *sock)
+{
+	struct file *file;
+	int newfd;
+
+	if (!sock->file)
+		return -EBADF;
+
+	file = get_file(sock->file);
+	newfd = get_unused_fd_flags(O_CLOEXEC);
+	if (newfd < 0) {
+		fput(file);
+		return newfd;
+	}
+
+	fd_install(newfd, file);
+	return newfd;
+}
+
+int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info)
+{
+	struct net *net = sock_net(skb->sk);
+	struct handshake_net *hn = handshake_pernet(net);
+	struct handshake_req *req = NULL;
+	struct socket *sock;
+	int class, fd, err;
+
+	err = -EOPNOTSUPP;
+	if (!hn)
+		goto out_status;
+
+	err = -EINVAL;
+	if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_ACCEPT_HANDLER_CLASS))
+		goto out_status;
+	class = nla_get_u32(info->attrs[HANDSHAKE_A_ACCEPT_HANDLER_CLASS]);
+
+	err = -EAGAIN;
+	req = handshake_req_next(hn, class);
+	if (!req)
+		goto out_status;
+
+	sock = req->hr_sk->sk_socket;
+	fd = handshake_dup(sock);
+	if (fd < 0) {
+		err = fd;
+		goto out_complete;
+	}
+	err = req->hr_proto->hp_accept(req, info, fd);
+	if (err)
+		goto out_complete;
+
+	trace_handshake_cmd_accept(net, req, req->hr_sk, fd);
+	return 0;
+
+out_complete:
+	handshake_complete(req, -EIO, NULL);
+	fput(sock->file);
+out_status:
+	trace_handshake_cmd_accept_err(net, req, NULL, err);
+	return err;
+}
+
+int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info)
+{
+	struct net *net = sock_net(skb->sk);
+	struct socket *sock = NULL;
+	struct handshake_req *req;
+	int fd, status, err;
+
+	if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_DONE_SOCKFD))
+		return -EINVAL;
+	fd = nla_get_u32(info->attrs[HANDSHAKE_A_DONE_SOCKFD]);
+
+	err = 0;
+	sock = sockfd_lookup(fd, &err);
+	if (err) {
+		err = -EBADF;
+		goto out_status;
+	}
+
+	req = handshake_req_hash_lookup(sock->sk);
+	if (!req) {
+		err = -EBUSY;
+		fput(sock->file);
+		goto out_status;
+	}
+
+	trace_handshake_cmd_done(net, req, sock->sk, fd);
+
+	status = -EIO;
+	if (info->attrs[HANDSHAKE_A_DONE_STATUS])
+		status = nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]);
+
+	handshake_complete(req, status, info);
+	fput(sock->file);
+	return 0;
+
+out_status:
+	trace_handshake_cmd_done_err(net, req, sock->sk, err);
+	return err;
+}
+
+static unsigned int handshake_net_id;
+
+static int __net_init handshake_net_init(struct net *net)
+{
+	struct handshake_net *hn = net_generic(net, handshake_net_id);
+	unsigned long tmp;
+	struct sysinfo si;
+
+	/*
+	 * Arbitrary limit to prevent handshakes that do not make
+	 * progress from clogging up the system. The cap scales up
+	 * with the amount of physical memory on the system.
+	 */
+	si_meminfo(&si);
+	tmp = si.totalram / (25 * si.mem_unit);
+	hn->hn_pending_max = clamp(tmp, 3UL, 50UL);
+
+	spin_lock_init(&hn->hn_lock);
+	hn->hn_pending = 0;
+	hn->hn_flags = 0;
+	INIT_LIST_HEAD(&hn->hn_requests);
+	return 0;
+}
+
+static void __net_exit handshake_net_exit(struct net *net)
+{
+	struct handshake_net *hn = net_generic(net, handshake_net_id);
+	struct handshake_req *req;
+	LIST_HEAD(requests);
+
+	/*
+	 * Drain the net's pending list. Requests that have been
+	 * accepted and are in progress will be destroyed when
+	 * the socket is closed.
+	 */
+	spin_lock(&hn->hn_lock);
+	set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags);
+	list_splice_init(&requests, &hn->hn_requests);
+	spin_unlock(&hn->hn_lock);
+
+	while (!list_empty(&requests)) {
+		req = list_first_entry(&requests, struct handshake_req, hr_list);
+		list_del(&req->hr_list);
+
+		/*
+		 * Requests on this list have not yet been
+		 * accepted, so they do not have an fd to put.
+		 */
+
+		handshake_complete(req, -ETIMEDOUT, NULL);
+	}
+}
+
+static struct pernet_operations __net_initdata handshake_genl_net_ops = {
+	.init		= handshake_net_init,
+	.exit		= handshake_net_exit,
+	.id		= &handshake_net_id,
+	.size		= sizeof(struct handshake_net),
+};
+
+/**
+ * handshake_pernet - Get the handshake private per-net structure
+ * @net: network namespace
+ *
+ * Returns a pointer to the net's private per-net structure for the
+ * handshake module, or NULL if handshake_init() failed.
+ */
+struct handshake_net *handshake_pernet(struct net *net)
+{
+	return handshake_net_id ?
+		net_generic(net, handshake_net_id) : NULL;
+}
+
+static int __init handshake_init(void)
+{
+	int ret;
+
+	ret = handshake_req_hash_init();
+	if (ret) {
+		pr_warn("handshake: hash initialization failed (%d)\n", ret);
+		return ret;
+	}
+
+	ret = genl_register_family(&handshake_nl_family);
+	if (ret) {
+		pr_warn("handshake: netlink registration failed (%d)\n", ret);
+		handshake_req_hash_destroy();
+		return ret;
+	}
+
+	/*
+	 * ORDER: register_pernet_subsys must be done last.
+	 *
+	 *	If initialization does not make it past pernet_subsys
+	 *	registration, then handshake_net_id will remain 0. That
+	 *	shunts the handshake consumer API to return ENOTSUPP
+	 *	to prevent it from dereferencing something that hasn't
+	 *	been allocated.
+	 */
+	ret = register_pernet_subsys(&handshake_genl_net_ops);
+	if (ret) {
+		pr_warn("handshake: pernet registration failed (%d)\n", ret);
+		genl_unregister_family(&handshake_nl_family);
+		handshake_req_hash_destroy();
+	}
+
+	return ret;
+}
+
+static void __exit handshake_exit(void)
+{
+	unregister_pernet_subsys(&handshake_genl_net_ops);
+	handshake_net_id = 0;
+
+	handshake_req_hash_destroy();
+	genl_unregister_family(&handshake_nl_family);
+}
+
+module_init(handshake_init);
+module_exit(handshake_exit);
diff --git a/net/handshake/request.c b/net/handshake/request.c
new file mode 100644
index 000000000000..d5b2bc6de057
--- /dev/null
+++ b/net/handshake/request.c
@@ -0,0 +1,339 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Handshake request lifetime events
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2023, Oracle and/or its affiliates.
+ */
+
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/inet.h>
+#include <linux/fdtable.h>
+#include <linux/rhashtable.h>
+
+#include <net/sock.h>
+#include <net/genetlink.h>
+#include <net/netns/generic.h>
+
+#include <uapi/linux/handshake.h>
+#include "handshake.h"
+
+#include <trace/events/handshake.h>
+
+/*
+ * We need both a handshake_req -> sock mapping, and a sock ->
+ * handshake_req mapping. Both are one-to-one.
+ *
+ * To avoid adding another pointer field to struct sock, net/handshake
+ * maintains a hash table, indexed by the memory address of @sock, to
+ * find the struct handshake_req outstanding for that socket. The
+ * reverse direction uses a simple pointer field in the handshake_req
+ * struct.
+ */
+
+static struct rhashtable handshake_rhashtbl ____cacheline_aligned_in_smp;
+
+static const struct rhashtable_params handshake_rhash_params = {
+	.key_len		= sizeof_field(struct handshake_req, hr_sk),
+	.key_offset		= offsetof(struct handshake_req, hr_sk),
+	.head_offset		= offsetof(struct handshake_req, hr_rhash),
+	.automatic_shrinking	= true,
+};
+
+int handshake_req_hash_init(void)
+{
+	return rhashtable_init(&handshake_rhashtbl, &handshake_rhash_params);
+}
+
+void handshake_req_hash_destroy(void)
+{
+	rhashtable_destroy(&handshake_rhashtbl);
+}
+
+struct handshake_req *handshake_req_hash_lookup(struct sock *sk)
+{
+	return rhashtable_lookup_fast(&handshake_rhashtbl, &sk,
+				      handshake_rhash_params);
+}
+
+static bool handshake_req_hash_add(struct handshake_req *req)
+{
+	int ret;
+
+	ret = rhashtable_lookup_insert_fast(&handshake_rhashtbl,
+					    &req->hr_rhash,
+					    handshake_rhash_params);
+	return ret == 0;
+}
+
+static void handshake_req_destroy(struct handshake_req *req)
+{
+	if (req->hr_proto->hp_destroy)
+		req->hr_proto->hp_destroy(req);
+	rhashtable_remove_fast(&handshake_rhashtbl, &req->hr_rhash,
+			       handshake_rhash_params);
+	kfree(req);
+}
+
+static void handshake_sk_destruct(struct sock *sk)
+{
+	void (*sk_destruct)(struct sock *sk);
+	struct handshake_req *req;
+
+	req = handshake_req_hash_lookup(sk);
+	if (!req)
+		return;
+
+	trace_handshake_destruct(sock_net(sk), req, sk);
+	sk_destruct = req->hr_odestruct;
+	handshake_req_destroy(req);
+	if (sk_destruct)
+		sk_destruct(sk);
+}
+
+/**
+ * handshake_req_alloc - Allocate a handshake request
+ * @proto: security protocol
+ * @flags: memory allocation flags
+ *
+ * Returns an initialized handshake_req or NULL.
+ */
+struct handshake_req *handshake_req_alloc(const struct handshake_proto *proto,
+					  gfp_t flags)
+{
+	struct handshake_req *req;
+
+	if (!proto)
+		return NULL;
+	if (proto->hp_handler_class <= HANDSHAKE_HANDLER_CLASS_NONE)
+		return NULL;
+	if (proto->hp_handler_class >= HANDSHAKE_HANDLER_CLASS_MAX)
+		return NULL;
+	if (!proto->hp_accept || !proto->hp_done)
+		return NULL;
+
+	req = kzalloc(struct_size(req, hr_priv, proto->hp_privsize), flags);
+	if (!req)
+		return NULL;
+
+	INIT_LIST_HEAD(&req->hr_list);
+	req->hr_proto = proto;
+	return req;
+}
+EXPORT_SYMBOL(handshake_req_alloc);
+
+/**
+ * handshake_req_private - Get per-handshake private data
+ * @req: handshake arguments
+ *
+ */
+void *handshake_req_private(struct handshake_req *req)
+{
+	return (void *)&req->hr_priv;
+}
+EXPORT_SYMBOL(handshake_req_private);
+
+static bool __add_pending_locked(struct handshake_net *hn,
+				 struct handshake_req *req)
+{
+	if (WARN_ON_ONCE(!list_empty(&req->hr_list)))
+		return false;
+	hn->hn_pending++;
+	list_add_tail(&req->hr_list, &hn->hn_requests);
+	return true;
+}
+
+static void __remove_pending_locked(struct handshake_net *hn,
+				    struct handshake_req *req)
+{
+	hn->hn_pending--;
+	list_del_init(&req->hr_list);
+}
+
+/*
+ * Returns %true if the request was found on @net's pending list,
+ * otherwise %false.
+ *
+ * If @req was on a pending list, it has not yet been accepted.
+ */
+static bool remove_pending(struct handshake_net *hn, struct handshake_req *req)
+{
+	bool ret = false;
+
+	spin_lock(&hn->hn_lock);
+	if (!list_empty(&req->hr_list)) {
+		__remove_pending_locked(hn, req);
+		ret = true;
+	}
+	spin_unlock(&hn->hn_lock);
+
+	return ret;
+}
+
+struct handshake_req *handshake_req_next(struct handshake_net *hn, int class)
+{
+	struct handshake_req *req, *pos;
+
+	req = NULL;
+	spin_lock(&hn->hn_lock);
+	list_for_each_entry(pos, &hn->hn_requests, hr_list) {
+		if (pos->hr_proto->hp_handler_class != class)
+			continue;
+		__remove_pending_locked(hn, pos);
+		req = pos;
+		break;
+	}
+	spin_unlock(&hn->hn_lock);
+
+	return req;
+}
+
+/**
+ * handshake_req_submit - Submit a handshake request
+ * @sock: open socket on which to perform the handshake
+ * @req: handshake arguments
+ * @flags: memory allocation flags
+ *
+ * Return values:
+ *   %0: Request queued
+ *   %-EINVAL: Invalid argument
+ *   %-EBUSY: A handshake is already under way for this socket
+ *   %-ESRCH: No handshake agent is available
+ *   %-EAGAIN: Too many pending handshake requests
+ *   %-ENOMEM: Failed to allocate memory
+ *   %-EMSGSIZE: Failed to construct notification message
+ *   %-EOPNOTSUPP: Handshake module not initialized
+ *
+ * A zero return value from handshake_req_submit() means that
+ * exactly one subsequent completion callback is guaranteed.
+ *
+ * A negative return value from handshake_req_submit() means that
+ * no completion callback will be done and that @req has been
+ * destroyed.
+ */
+int handshake_req_submit(struct socket *sock, struct handshake_req *req,
+			 gfp_t flags)
+{
+	struct handshake_net *hn;
+	struct net *net;
+	int ret;
+
+	if (!sock || !req || !sock->file) {
+		kfree(req);
+		return -EINVAL;
+	}
+
+	req->hr_sk = sock->sk;
+	if (!req->hr_sk) {
+		kfree(req);
+		return -EINVAL;
+	}
+	req->hr_odestruct = req->hr_sk->sk_destruct;
+	req->hr_sk->sk_destruct = handshake_sk_destruct;
+
+	ret = -EOPNOTSUPP;
+	net = sock_net(req->hr_sk);
+	hn = handshake_pernet(net);
+	if (!hn)
+		goto out_err;
+
+	ret = -EAGAIN;
+	if (READ_ONCE(hn->hn_pending) >= hn->hn_pending_max)
+		goto out_err;
+
+	spin_lock(&hn->hn_lock);
+	ret = -EOPNOTSUPP;
+	if (test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags))
+		goto out_unlock;
+	ret = -EBUSY;
+	if (!handshake_req_hash_add(req))
+		goto out_unlock;
+	if (!__add_pending_locked(hn, req))
+		goto out_unlock;
+	spin_unlock(&hn->hn_lock);
+
+	ret = handshake_genl_notify(net, req->hr_proto, flags);
+	if (ret) {
+		trace_handshake_notify_err(net, req, req->hr_sk, ret);
+		if (remove_pending(hn, req))
+			goto out_err;
+	}
+
+	/* Prevent socket release while a handshake request is pending */
+	sock_hold(req->hr_sk);
+
+	trace_handshake_submit(net, req, req->hr_sk);
+	return 0;
+
+out_unlock:
+	spin_unlock(&hn->hn_lock);
+out_err:
+	trace_handshake_submit_err(net, req, req->hr_sk, ret);
+	handshake_req_destroy(req);
+	return ret;
+}
+EXPORT_SYMBOL(handshake_req_submit);
+
+void handshake_complete(struct handshake_req *req, unsigned int status,
+			struct genl_info *info)
+{
+	struct sock *sk = req->hr_sk;
+	struct net *net = sock_net(sk);
+
+	if (!test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) {
+		trace_handshake_complete(net, req, sk, status);
+		req->hr_proto->hp_done(req, status, info);
+
+		/* Handshake request is no longer pending */
+		sock_put(sk);
+	}
+}
+
+/**
+ * handshake_req_cancel - Cancel an in-progress handshake
+ * @sk: socket on which there is an ongoing handshake
+ *
+ * Request cancellation races with request completion. To determine
+ * who won, callers examine the return value from this function.
+ *
+ * Return values:
+ *   %true - Uncompleted handshake request was canceled
+ *   %false - Handshake request already completed or not found
+ */
+bool handshake_req_cancel(struct sock *sk)
+{
+	struct handshake_req *req;
+	struct handshake_net *hn;
+	struct net *net;
+
+	net = sock_net(sk);
+	req = handshake_req_hash_lookup(sk);
+	if (!req) {
+		trace_handshake_cancel_none(net, req, sk);
+		return false;
+	}
+
+	hn = handshake_pernet(net);
+	if (hn && remove_pending(hn, req)) {
+		/* Request hadn't been accepted */
+		goto out_true;
+	}
+	if (test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) {
+		/* Request already completed */
+		trace_handshake_cancel_busy(net, req, sk);
+		return false;
+	}
+
+out_true:
+	trace_handshake_cancel(net, req, sk);
+
+	/* Handshake request is no longer pending */
+	sock_put(sk);
+	return true;
+}
+EXPORT_SYMBOL(handshake_req_cancel);
diff --git a/net/handshake/trace.c b/net/handshake/trace.c
new file mode 100644
index 000000000000..1c4d8e27e17a
--- /dev/null
+++ b/net/handshake/trace.c
@@ -0,0 +1,20 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Trace points for transport security layer handshakes.
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2023, Oracle and/or its affiliates.
+ */
+
+#include <linux/types.h>
+
+#include <net/sock.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+
+#include "handshake.h"
+
+#define CREATE_TRACE_POINTS
+
+#include <trace/events/handshake.h>
-- 
cgit v1.2.3-70-g09d2


From 2fd5532044a89d2403b543520b4902e196f7d165 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 17 Apr 2023 10:32:33 -0400
Subject: net/handshake: Add a kernel API for requesting a TLSv1.3 handshake

To enable kernel consumers of TLS to request a TLS handshake, add
support to net/handshake/ to request a handshake upcall.

This patch also acts as a template for adding handshake upcall
support for other kernel transport layer security providers.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/handshake.yaml |   4 +-
 Documentation/networking/index.rst         |   1 +
 Documentation/networking/tls-handshake.rst | 217 +++++++++++++++
 MAINTAINERS                                |   2 +
 include/net/handshake.h                    |  43 +++
 include/uapi/linux/handshake.h             |   2 +
 net/handshake/Makefile                     |   2 +-
 net/handshake/genl.c                       |   3 +-
 net/handshake/genl.h                       |   1 +
 net/handshake/tlshd.c                      | 417 +++++++++++++++++++++++++++++
 10 files changed, 689 insertions(+), 3 deletions(-)
 create mode 100644 Documentation/networking/tls-handshake.rst
 create mode 100644 include/net/handshake.h
 create mode 100644 net/handshake/tlshd.c

diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml
index 0333d92b1438..614f1a585511 100644
--- a/Documentation/netlink/specs/handshake.yaml
+++ b/Documentation/netlink/specs/handshake.yaml
@@ -16,7 +16,7 @@ definitions:
     type: enum
     name: handler-class
     value-start: 0
-    entries: [ none, max ]
+    entries: [ none, tlshd, max ]
   -
     type: enum
     name: msg-type
@@ -120,3 +120,5 @@ mcast-groups:
   list:
     -
       name: none
+    -
+      name: tlshd
diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index 24bb256d6d53..a164ff074356 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -36,6 +36,7 @@ Contents:
    scaling
    tls
    tls-offload
+   tls-handshake
    nfc
    6lowpan
    6pack
diff --git a/Documentation/networking/tls-handshake.rst b/Documentation/networking/tls-handshake.rst
new file mode 100644
index 000000000000..a2817a88e905
--- /dev/null
+++ b/Documentation/networking/tls-handshake.rst
@@ -0,0 +1,217 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=======================
+In-Kernel TLS Handshake
+=======================
+
+Overview
+========
+
+Transport Layer Security (TLS) is a Upper Layer Protocol (ULP) that runs
+over TCP. TLS provides end-to-end data integrity and confidentiality in
+addition to peer authentication.
+
+The kernel's kTLS implementation handles the TLS record subprotocol, but
+does not handle the TLS handshake subprotocol which is used to establish
+a TLS session. Kernel consumers can use the API described here to
+request TLS session establishment.
+
+There are several possible ways to provide a handshake service in the
+kernel. The API described here is designed to hide the details of those
+implementations so that in-kernel TLS consumers do not need to be
+aware of how the handshake gets done.
+
+
+User handshake agent
+====================
+
+As of this writing, there is no TLS handshake implementation in the
+Linux kernel. To provide a handshake service, a handshake agent
+(typically in user space) is started in each network namespace where a
+kernel consumer might require a TLS handshake. Handshake agents listen
+for events sent from the kernel that indicate a handshake request is
+waiting.
+
+An open socket is passed to a handshake agent via a netlink operation,
+which creates a socket descriptor in the agent's file descriptor table.
+If the handshake completes successfully, the handshake agent promotes
+the socket to use the TLS ULP and sets the session information using the
+SOL_TLS socket options. The handshake agent returns the socket to the
+kernel via a second netlink operation.
+
+
+Kernel Handshake API
+====================
+
+A kernel TLS consumer initiates a client-side TLS handshake on an open
+socket by invoking one of the tls_client_hello() functions. First, it
+fills in a structure that contains the parameters of the request:
+
+.. code-block:: c
+
+  struct tls_handshake_args {
+        struct socket   *ta_sock;
+        tls_done_func_t ta_done;
+        void            *ta_data;
+        unsigned int    ta_timeout_ms;
+        key_serial_t    ta_keyring;
+        key_serial_t    ta_my_cert;
+        key_serial_t    ta_my_privkey;
+        unsigned int    ta_num_peerids;
+        key_serial_t    ta_my_peerids[5];
+  };
+
+The @ta_sock field references an open and connected socket. The consumer
+must hold a reference on the socket to prevent it from being destroyed
+while the handshake is in progress. The consumer must also have
+instantiated a struct file in sock->file.
+
+
+@ta_done contains a callback function that is invoked when the handshake
+has completed. Further explanation of this function is in the "Handshake
+Completion" sesction below.
+
+The consumer can fill in the @ta_timeout_ms field to force the servicing
+handshake agent to exit after a number of milliseconds. This enables the
+socket to be fully closed once both the kernel and the handshake agent
+have closed their endpoints.
+
+Authentication material such as x.509 certificates, private certificate
+keys, and pre-shared keys are provided to the handshake agent in keys
+that are instantiated by the consumer before making the handshake
+request. The consumer can provide a private keyring that is linked into
+the handshake agent's process keyring in the @ta_keyring field to prevent
+access of those keys by other subsystems.
+
+To request an x.509-authenticated TLS session, the consumer fills in
+the @ta_my_cert and @ta_my_privkey fields with the serial numbers of
+keys containing an x.509 certificate and the private key for that
+certificate. Then, it invokes this function:
+
+.. code-block:: c
+
+  ret = tls_client_hello_x509(args, gfp_flags);
+
+The function returns zero when the handshake request is under way. A
+zero return guarantees the callback function @ta_done will be invoked
+for this socket. The function returns a negative errno if the handshake
+could not be started. A negative errno guarantees the callback function
+@ta_done will not be invoked on this socket.
+
+
+To initiate a client-side TLS handshake with a pre-shared key, use:
+
+.. code-block:: c
+
+  ret = tls_client_hello_psk(args, gfp_flags);
+
+However, in this case, the consumer fills in the @ta_my_peerids array
+with serial numbers of keys containing the peer identities it wishes
+to offer, and the @ta_num_peerids field with the number of array
+entries it has filled in. The other fields are filled in as above.
+
+
+To initiate an anonymous client-side TLS handshake use:
+
+.. code-block:: c
+
+  ret = tls_client_hello_anon(args, gfp_flags);
+
+The handshake agent presents no peer identity information to the remote
+during this type of handshake. Only server authentication (ie the client
+verifies the server's identity) is performed during the handshake. Thus
+the established session uses encryption only.
+
+
+Consumers that are in-kernel servers use:
+
+.. code-block:: c
+
+  ret = tls_server_hello_x509(args, gfp_flags);
+
+or
+
+.. code-block:: c
+
+  ret = tls_server_hello_psk(args, gfp_flags);
+
+The argument structure is filled in as above.
+
+
+If the consumer needs to cancel the handshake request, say, due to a ^C
+or other exigent event, the consumer can invoke:
+
+.. code-block:: c
+
+  bool tls_handshake_cancel(sock);
+
+This function returns true if the handshake request associated with
+@sock has been canceled. The consumer's handshake completion callback
+will not be invoked. If this function returns false, then the consumer's
+completion callback has already been invoked.
+
+
+Handshake Completion
+====================
+
+When the handshake agent has completed processing, it notifies the
+kernel that the socket may be used by the consumer again. At this point,
+the consumer's handshake completion callback, provided in the @ta_done
+field in the tls_handshake_args structure, is invoked.
+
+The synopsis of this function is:
+
+.. code-block:: c
+
+  typedef void	(*tls_done_func_t)(void *data, int status,
+                                   key_serial_t peerid);
+
+The consumer provides a cookie in the @ta_data field of the
+tls_handshake_args structure that is returned in the @data parameter of
+this callback. The consumer uses the cookie to match the callback to the
+thread waiting for the handshake to complete.
+
+The success status of the handshake is returned via the @status
+parameter:
+
++------------+----------------------------------------------+
+|  status    |  meaning                                     |
++============+==============================================+
+|  0         |  TLS session established successfully        |
++------------+----------------------------------------------+
+|  -EACCESS  |  Remote peer rejected the handshake or       |
+|            |  authentication failed                       |
++------------+----------------------------------------------+
+|  -ENOMEM   |  Temporary resource allocation failure       |
++------------+----------------------------------------------+
+|  -EINVAL   |  Consumer provided an invalid argument       |
++------------+----------------------------------------------+
+|  -ENOKEY   |  Missing authentication material             |
++------------+----------------------------------------------+
+|  -EIO      |  An unexpected fault occurred                |
++------------+----------------------------------------------+
+
+The @peerid parameter contains the serial number of a key containing the
+remote peer's identity or the value TLS_NO_PEERID if the session is not
+authenticated.
+
+A best practice is to close and destroy the socket immediately if the
+handshake failed.
+
+
+Other considerations
+--------------------
+
+While a handshake is under way, the kernel consumer must alter the
+socket's sk_data_ready callback function to ignore all incoming data.
+Once the handshake completion callback function has been invoked, normal
+receive operation can be resumed.
+
+Once a TLS session is established, the consumer must provide a buffer
+for and then examine the control message (CMSG) that is part of every
+subsequent sock_recvmsg(). Each control message indicates whether the
+received message data is TLS record data or session metadata.
+
+See tls.rst for details on how a kTLS consumer recognizes incoming
+(decrypted) application data, alerts, and handshake packets once the
+socket has been promoted to use the TLS ULP.
diff --git a/MAINTAINERS b/MAINTAINERS
index cdc7748d15b8..04ebde8ccb75 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8953,6 +8953,8 @@ L:	kernel-tls-handshake@lists.linux.dev
 L:	netdev@vger.kernel.org
 S:	Maintained
 F:	Documentation/netlink/specs/handshake.yaml
+F:	Documentation/networking/tls-handshake.rst
+F:	include/net/handshake.h
 F:	include/trace/events/handshake.h
 F:	net/handshake/
 
diff --git a/include/net/handshake.h b/include/net/handshake.h
new file mode 100644
index 000000000000..3352b1ab43b3
--- /dev/null
+++ b/include/net/handshake.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Generic netlink HANDSHAKE service.
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2023, Oracle and/or its affiliates.
+ */
+
+#ifndef _NET_HANDSHAKE_H
+#define _NET_HANDSHAKE_H
+
+enum {
+	TLS_NO_KEYRING = 0,
+	TLS_NO_PEERID = 0,
+	TLS_NO_CERT = 0,
+	TLS_NO_PRIVKEY = 0,
+};
+
+typedef void	(*tls_done_func_t)(void *data, int status,
+				   key_serial_t peerid);
+
+struct tls_handshake_args {
+	struct socket		*ta_sock;
+	tls_done_func_t		ta_done;
+	void			*ta_data;
+	unsigned int		ta_timeout_ms;
+	key_serial_t		ta_keyring;
+	key_serial_t		ta_my_cert;
+	key_serial_t		ta_my_privkey;
+	unsigned int		ta_num_peerids;
+	key_serial_t		ta_my_peerids[5];
+};
+
+int tls_client_hello_anon(const struct tls_handshake_args *args, gfp_t flags);
+int tls_client_hello_x509(const struct tls_handshake_args *args, gfp_t flags);
+int tls_client_hello_psk(const struct tls_handshake_args *args, gfp_t flags);
+int tls_server_hello_x509(const struct tls_handshake_args *args, gfp_t flags);
+int tls_server_hello_psk(const struct tls_handshake_args *args, gfp_t flags);
+
+bool tls_handshake_cancel(struct sock *sk);
+
+#endif /* _NET_HANDSHAKE_H */
diff --git a/include/uapi/linux/handshake.h b/include/uapi/linux/handshake.h
index 7f66ff489b87..1de4d0b95325 100644
--- a/include/uapi/linux/handshake.h
+++ b/include/uapi/linux/handshake.h
@@ -11,6 +11,7 @@
 
 enum handshake_handler_class {
 	HANDSHAKE_HANDLER_CLASS_NONE,
+	HANDSHAKE_HANDLER_CLASS_TLSHD,
 	HANDSHAKE_HANDLER_CLASS_MAX,
 };
 
@@ -67,5 +68,6 @@ enum {
 };
 
 #define HANDSHAKE_MCGRP_NONE	"none"
+#define HANDSHAKE_MCGRP_TLSHD	"tlshd"
 
 #endif /* _UAPI_LINUX_HANDSHAKE_H */
diff --git a/net/handshake/Makefile b/net/handshake/Makefile
index d38736de45da..a089f7e3df24 100644
--- a/net/handshake/Makefile
+++ b/net/handshake/Makefile
@@ -8,4 +8,4 @@
 #
 
 obj-y += handshake.o
-handshake-y := genl.o netlink.o request.o trace.o
+handshake-y := genl.o netlink.o request.o tlshd.o trace.o
diff --git a/net/handshake/genl.c b/net/handshake/genl.c
index 652f37d19bd6..9f29efb1493e 100644
--- a/net/handshake/genl.c
+++ b/net/handshake/genl.c
@@ -12,7 +12,7 @@
 
 /* HANDSHAKE_CMD_ACCEPT - do */
 static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HANDLER_CLASS + 1] = {
-	[HANDSHAKE_A_ACCEPT_HANDLER_CLASS] = NLA_POLICY_MAX(NLA_U32, 1),
+	[HANDSHAKE_A_ACCEPT_HANDLER_CLASS] = NLA_POLICY_MAX(NLA_U32, 2),
 };
 
 /* HANDSHAKE_CMD_DONE - do */
@@ -42,6 +42,7 @@ static const struct genl_split_ops handshake_nl_ops[] = {
 
 static const struct genl_multicast_group handshake_nl_mcgrps[] = {
 	[HANDSHAKE_NLGRP_NONE] = { "none", },
+	[HANDSHAKE_NLGRP_TLSHD] = { "tlshd", },
 };
 
 struct genl_family handshake_nl_family __ro_after_init = {
diff --git a/net/handshake/genl.h b/net/handshake/genl.h
index a1eb7ccccc7f..2c1f1aa6a02a 100644
--- a/net/handshake/genl.h
+++ b/net/handshake/genl.h
@@ -16,6 +16,7 @@ int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info);
 
 enum {
 	HANDSHAKE_NLGRP_NONE,
+	HANDSHAKE_NLGRP_TLSHD,
 };
 
 extern struct genl_family handshake_nl_family;
diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c
new file mode 100644
index 000000000000..1b8353296060
--- /dev/null
+++ b/net/handshake/tlshd.c
@@ -0,0 +1,417 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Establish a TLS session for a kernel socket consumer
+ * using the tlshd user space handler.
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2021-2023, Oracle and/or its affiliates.
+ */
+
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/key.h>
+
+#include <net/sock.h>
+#include <net/handshake.h>
+#include <net/genetlink.h>
+
+#include <uapi/linux/keyctl.h>
+#include <uapi/linux/handshake.h>
+#include "handshake.h"
+
+struct tls_handshake_req {
+	void			(*th_consumer_done)(void *data, int status,
+						    key_serial_t peerid);
+	void			*th_consumer_data;
+
+	int			th_type;
+	unsigned int		th_timeout_ms;
+	int			th_auth_mode;
+	key_serial_t		th_keyring;
+	key_serial_t		th_certificate;
+	key_serial_t		th_privkey;
+
+	unsigned int		th_num_peerids;
+	key_serial_t		th_peerid[5];
+};
+
+static struct tls_handshake_req *
+tls_handshake_req_init(struct handshake_req *req,
+		       const struct tls_handshake_args *args)
+{
+	struct tls_handshake_req *treq = handshake_req_private(req);
+
+	treq->th_timeout_ms = args->ta_timeout_ms;
+	treq->th_consumer_done = args->ta_done;
+	treq->th_consumer_data = args->ta_data;
+	treq->th_keyring = args->ta_keyring;
+	treq->th_num_peerids = 0;
+	treq->th_certificate = TLS_NO_CERT;
+	treq->th_privkey = TLS_NO_PRIVKEY;
+	return treq;
+}
+
+static void tls_handshake_remote_peerids(struct tls_handshake_req *treq,
+					 struct genl_info *info)
+{
+	struct nlattr *head = nlmsg_attrdata(info->nlhdr, GENL_HDRLEN);
+	int rem, len = nlmsg_attrlen(info->nlhdr, GENL_HDRLEN);
+	struct nlattr *nla;
+	unsigned int i;
+
+	i = 0;
+	nla_for_each_attr(nla, head, len, rem) {
+		if (nla_type(nla) == HANDSHAKE_A_DONE_REMOTE_AUTH)
+			i++;
+	}
+	if (!i)
+		return;
+	treq->th_num_peerids = min_t(unsigned int, i,
+				     ARRAY_SIZE(treq->th_peerid));
+
+	i = 0;
+	nla_for_each_attr(nla, head, len, rem) {
+		if (nla_type(nla) == HANDSHAKE_A_DONE_REMOTE_AUTH)
+			treq->th_peerid[i++] = nla_get_u32(nla);
+		if (i >= treq->th_num_peerids)
+			break;
+	}
+}
+
+/**
+ * tls_handshake_done - callback to handle a CMD_DONE request
+ * @req: socket on which the handshake was performed
+ * @status: session status code
+ * @info: full results of session establishment
+ *
+ */
+static void tls_handshake_done(struct handshake_req *req,
+			       unsigned int status, struct genl_info *info)
+{
+	struct tls_handshake_req *treq = handshake_req_private(req);
+
+	treq->th_peerid[0] = TLS_NO_PEERID;
+	if (info)
+		tls_handshake_remote_peerids(treq, info);
+
+	treq->th_consumer_done(treq->th_consumer_data, -status,
+			       treq->th_peerid[0]);
+}
+
+#if IS_ENABLED(CONFIG_KEYS)
+static int tls_handshake_private_keyring(struct tls_handshake_req *treq)
+{
+	key_ref_t process_keyring_ref, keyring_ref;
+	int ret;
+
+	if (treq->th_keyring == TLS_NO_KEYRING)
+		return 0;
+
+	process_keyring_ref = lookup_user_key(KEY_SPEC_PROCESS_KEYRING,
+					      KEY_LOOKUP_CREATE,
+					      KEY_NEED_WRITE);
+	if (IS_ERR(process_keyring_ref)) {
+		ret = PTR_ERR(process_keyring_ref);
+		goto out;
+	}
+
+	keyring_ref = lookup_user_key(treq->th_keyring, KEY_LOOKUP_CREATE,
+				      KEY_NEED_LINK);
+	if (IS_ERR(keyring_ref)) {
+		ret = PTR_ERR(keyring_ref);
+		goto out_put_key;
+	}
+
+	ret = key_link(key_ref_to_ptr(process_keyring_ref),
+		       key_ref_to_ptr(keyring_ref));
+
+	key_ref_put(keyring_ref);
+out_put_key:
+	key_ref_put(process_keyring_ref);
+out:
+	return ret;
+}
+#else
+static int tls_handshake_private_keyring(struct tls_handshake_req *treq)
+{
+	return 0;
+}
+#endif
+
+static int tls_handshake_put_peer_identity(struct sk_buff *msg,
+					   struct tls_handshake_req *treq)
+{
+	unsigned int i;
+
+	for (i = 0; i < treq->th_num_peerids; i++)
+		if (nla_put_u32(msg, HANDSHAKE_A_ACCEPT_PEER_IDENTITY,
+				treq->th_peerid[i]) < 0)
+			return -EMSGSIZE;
+	return 0;
+}
+
+static int tls_handshake_put_certificate(struct sk_buff *msg,
+					 struct tls_handshake_req *treq)
+{
+	struct nlattr *entry_attr;
+
+	if (treq->th_certificate == TLS_NO_CERT &&
+	    treq->th_privkey == TLS_NO_PRIVKEY)
+		return 0;
+
+	entry_attr = nla_nest_start(msg, HANDSHAKE_A_ACCEPT_CERTIFICATE);
+	if (!entry_attr)
+		return -EMSGSIZE;
+
+	if (nla_put_u32(msg, HANDSHAKE_A_X509_CERT,
+			treq->th_certificate) ||
+	    nla_put_u32(msg, HANDSHAKE_A_X509_PRIVKEY,
+			treq->th_privkey)) {
+		nla_nest_cancel(msg, entry_attr);
+		return -EMSGSIZE;
+	}
+
+	nla_nest_end(msg, entry_attr);
+	return 0;
+}
+
+/**
+ * tls_handshake_accept - callback to construct a CMD_ACCEPT response
+ * @req: handshake parameters to return
+ * @info: generic netlink message context
+ * @fd: file descriptor to be returned
+ *
+ * Returns zero on success, or a negative errno on failure.
+ */
+static int tls_handshake_accept(struct handshake_req *req,
+				struct genl_info *info, int fd)
+{
+	struct tls_handshake_req *treq = handshake_req_private(req);
+	struct nlmsghdr *hdr;
+	struct sk_buff *msg;
+	int ret;
+
+	ret = tls_handshake_private_keyring(treq);
+	if (ret < 0)
+		goto out;
+
+	ret = -ENOMEM;
+	msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		goto out;
+	hdr = handshake_genl_put(msg, info);
+	if (!hdr)
+		goto out_cancel;
+
+	ret = -EMSGSIZE;
+	ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_SOCKFD, fd);
+	if (ret < 0)
+		goto out_cancel;
+	ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_MESSAGE_TYPE, treq->th_type);
+	if (ret < 0)
+		goto out_cancel;
+	if (treq->th_timeout_ms) {
+		ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_TIMEOUT, treq->th_timeout_ms);
+		if (ret < 0)
+			goto out_cancel;
+	}
+
+	ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_AUTH_MODE,
+			  treq->th_auth_mode);
+	if (ret < 0)
+		goto out_cancel;
+	switch (treq->th_auth_mode) {
+	case HANDSHAKE_AUTH_PSK:
+		ret = tls_handshake_put_peer_identity(msg, treq);
+		if (ret < 0)
+			goto out_cancel;
+		break;
+	case HANDSHAKE_AUTH_X509:
+		ret = tls_handshake_put_certificate(msg, treq);
+		if (ret < 0)
+			goto out_cancel;
+		break;
+	}
+
+	genlmsg_end(msg, hdr);
+	return genlmsg_reply(msg, info);
+
+out_cancel:
+	genlmsg_cancel(msg, hdr);
+out:
+	return ret;
+}
+
+static const struct handshake_proto tls_handshake_proto = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
+	.hp_privsize		= sizeof(struct tls_handshake_req),
+
+	.hp_accept		= tls_handshake_accept,
+	.hp_done		= tls_handshake_done,
+};
+
+/**
+ * tls_client_hello_anon - request an anonymous TLS handshake on a socket
+ * @args: socket and handshake parameters for this request
+ * @flags: memory allocation control flags
+ *
+ * Return values:
+ *   %0: Handshake request enqueue; ->done will be called when complete
+ *   %-ESRCH: No user agent is available
+ *   %-ENOMEM: Memory allocation failed
+ */
+int tls_client_hello_anon(const struct tls_handshake_args *args, gfp_t flags)
+{
+	struct tls_handshake_req *treq;
+	struct handshake_req *req;
+
+	req = handshake_req_alloc(&tls_handshake_proto, flags);
+	if (!req)
+		return -ENOMEM;
+	treq = tls_handshake_req_init(req, args);
+	treq->th_type = HANDSHAKE_MSG_TYPE_CLIENTHELLO;
+	treq->th_auth_mode = HANDSHAKE_AUTH_UNAUTH;
+
+	return handshake_req_submit(args->ta_sock, req, flags);
+}
+EXPORT_SYMBOL(tls_client_hello_anon);
+
+/**
+ * tls_client_hello_x509 - request an x.509-based TLS handshake on a socket
+ * @args: socket and handshake parameters for this request
+ * @flags: memory allocation control flags
+ *
+ * Return values:
+ *   %0: Handshake request enqueue; ->done will be called when complete
+ *   %-ESRCH: No user agent is available
+ *   %-ENOMEM: Memory allocation failed
+ */
+int tls_client_hello_x509(const struct tls_handshake_args *args, gfp_t flags)
+{
+	struct tls_handshake_req *treq;
+	struct handshake_req *req;
+
+	req = handshake_req_alloc(&tls_handshake_proto, flags);
+	if (!req)
+		return -ENOMEM;
+	treq = tls_handshake_req_init(req, args);
+	treq->th_type = HANDSHAKE_MSG_TYPE_CLIENTHELLO;
+	treq->th_auth_mode = HANDSHAKE_AUTH_X509;
+	treq->th_certificate = args->ta_my_cert;
+	treq->th_privkey = args->ta_my_privkey;
+
+	return handshake_req_submit(args->ta_sock, req, flags);
+}
+EXPORT_SYMBOL(tls_client_hello_x509);
+
+/**
+ * tls_client_hello_psk - request a PSK-based TLS handshake on a socket
+ * @args: socket and handshake parameters for this request
+ * @flags: memory allocation control flags
+ *
+ * Return values:
+ *   %0: Handshake request enqueue; ->done will be called when complete
+ *   %-EINVAL: Wrong number of local peer IDs
+ *   %-ESRCH: No user agent is available
+ *   %-ENOMEM: Memory allocation failed
+ */
+int tls_client_hello_psk(const struct tls_handshake_args *args, gfp_t flags)
+{
+	struct tls_handshake_req *treq;
+	struct handshake_req *req;
+	unsigned int i;
+
+	if (!args->ta_num_peerids ||
+	    args->ta_num_peerids > ARRAY_SIZE(treq->th_peerid))
+		return -EINVAL;
+
+	req = handshake_req_alloc(&tls_handshake_proto, flags);
+	if (!req)
+		return -ENOMEM;
+	treq = tls_handshake_req_init(req, args);
+	treq->th_type = HANDSHAKE_MSG_TYPE_CLIENTHELLO;
+	treq->th_auth_mode = HANDSHAKE_AUTH_PSK;
+	treq->th_num_peerids = args->ta_num_peerids;
+	for (i = 0; i < args->ta_num_peerids; i++)
+		treq->th_peerid[i] = args->ta_my_peerids[i];
+
+	return handshake_req_submit(args->ta_sock, req, flags);
+}
+EXPORT_SYMBOL(tls_client_hello_psk);
+
+/**
+ * tls_server_hello_x509 - request a server TLS handshake on a socket
+ * @args: socket and handshake parameters for this request
+ * @flags: memory allocation control flags
+ *
+ * Return values:
+ *   %0: Handshake request enqueue; ->done will be called when complete
+ *   %-ESRCH: No user agent is available
+ *   %-ENOMEM: Memory allocation failed
+ */
+int tls_server_hello_x509(const struct tls_handshake_args *args, gfp_t flags)
+{
+	struct tls_handshake_req *treq;
+	struct handshake_req *req;
+
+	req = handshake_req_alloc(&tls_handshake_proto, flags);
+	if (!req)
+		return -ENOMEM;
+	treq = tls_handshake_req_init(req, args);
+	treq->th_type = HANDSHAKE_MSG_TYPE_SERVERHELLO;
+	treq->th_auth_mode = HANDSHAKE_AUTH_X509;
+	treq->th_certificate = args->ta_my_cert;
+	treq->th_privkey = args->ta_my_privkey;
+
+	return handshake_req_submit(args->ta_sock, req, flags);
+}
+EXPORT_SYMBOL(tls_server_hello_x509);
+
+/**
+ * tls_server_hello_psk - request a server TLS handshake on a socket
+ * @args: socket and handshake parameters for this request
+ * @flags: memory allocation control flags
+ *
+ * Return values:
+ *   %0: Handshake request enqueue; ->done will be called when complete
+ *   %-ESRCH: No user agent is available
+ *   %-ENOMEM: Memory allocation failed
+ */
+int tls_server_hello_psk(const struct tls_handshake_args *args, gfp_t flags)
+{
+	struct tls_handshake_req *treq;
+	struct handshake_req *req;
+
+	req = handshake_req_alloc(&tls_handshake_proto, flags);
+	if (!req)
+		return -ENOMEM;
+	treq = tls_handshake_req_init(req, args);
+	treq->th_type = HANDSHAKE_MSG_TYPE_SERVERHELLO;
+	treq->th_auth_mode = HANDSHAKE_AUTH_PSK;
+	treq->th_num_peerids = 1;
+	treq->th_peerid[0] = args->ta_my_peerids[0];
+
+	return handshake_req_submit(args->ta_sock, req, flags);
+}
+EXPORT_SYMBOL(tls_server_hello_psk);
+
+/**
+ * tls_handshake_cancel - cancel a pending handshake
+ * @sk: socket on which there is an ongoing handshake
+ *
+ * Request cancellation races with request completion. To determine
+ * who won, callers examine the return value from this function.
+ *
+ * Return values:
+ *   %true - Uncompleted handshake request was canceled
+ *   %false - Handshake request already completed or not found
+ */
+bool tls_handshake_cancel(struct sock *sk)
+{
+	return handshake_req_cancel(sk);
+}
+EXPORT_SYMBOL(tls_handshake_cancel);
-- 
cgit v1.2.3-70-g09d2


From 88232ec1ec5ecf4aa5de439cff3d5e2b7adcac93 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 17 Apr 2023 10:32:39 -0400
Subject: net/handshake: Add Kunit tests for the handshake consumer API

These verify the API contracts and help exercise lifetime rules for
consumer sockets and handshake_req structures.

One way to run these tests:

./tools/testing/kunit/kunit.py run --kunitconfig ./net/handshake/.kunitconfig

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/Kconfig                    |  15 ++
 net/handshake/.kunitconfig     |  11 +
 net/handshake/Makefile         |   2 +
 net/handshake/handshake-test.c | 523 +++++++++++++++++++++++++++++++++++++++++
 net/handshake/handshake.h      |   5 +
 net/handshake/netlink.c        |   7 +
 net/handshake/request.c        |   5 +
 net/handshake/tlshd.c          |   1 +
 8 files changed, 569 insertions(+)
 create mode 100644 net/handshake/.kunitconfig
 create mode 100644 net/handshake/handshake-test.c

diff --git a/net/Kconfig b/net/Kconfig
index 4b800706cc76..7d39c1773eb4 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -73,6 +73,21 @@ config NET_HANDSHAKE
 	depends on SUNRPC || NVME_TARGET_TCP || NVME_TCP
 	default y
 
+config NET_HANDSHAKE_KUNIT_TEST
+	tristate "KUnit tests for the handshake upcall mechanism" if !KUNIT_ALL_TESTS
+	default KUNIT_ALL_TESTS
+	depends on KUNIT
+	help
+	  This builds the KUnit tests for the handshake upcall mechanism.
+
+	  KUnit tests run during boot and output the results to the debug
+	  log in TAP format (https://testanything.org/). Only useful for
+	  kernel devs running KUnit test harness and are not for inclusion
+	  into a production build.
+
+	  For more information on KUnit and unit tests in general, refer
+	  to the KUnit documentation in Documentation/dev-tools/kunit/.
+
 config INET
 	bool "TCP/IP networking"
 	help
diff --git a/net/handshake/.kunitconfig b/net/handshake/.kunitconfig
new file mode 100644
index 000000000000..5c48cf4abca2
--- /dev/null
+++ b/net/handshake/.kunitconfig
@@ -0,0 +1,11 @@
+CONFIG_KUNIT=y
+CONFIG_UBSAN=y
+CONFIG_STACKTRACE=y
+CONFIG_NET=y
+CONFIG_NETWORK_FILESYSTEMS=y
+CONFIG_INET=y
+CONFIG_MULTIUSER=y
+CONFIG_NFS_FS=y
+CONFIG_SUNRPC=y
+CONFIG_NET_HANDSHAKE=y
+CONFIG_NET_HANDSHAKE_KUNIT_TEST=y
diff --git a/net/handshake/Makefile b/net/handshake/Makefile
index a089f7e3df24..247d73c6ff6e 100644
--- a/net/handshake/Makefile
+++ b/net/handshake/Makefile
@@ -9,3 +9,5 @@
 
 obj-y += handshake.o
 handshake-y := genl.o netlink.o request.o tlshd.o trace.o
+
+obj-$(CONFIG_NET_HANDSHAKE_KUNIT_TEST) += handshake-test.o
diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c
new file mode 100644
index 000000000000..e6adc5dec11a
--- /dev/null
+++ b/net/handshake/handshake-test.c
@@ -0,0 +1,523 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2023 Oracle and/or its affiliates.
+ *
+ * KUnit test of the handshake upcall mechanism.
+ */
+
+#include <kunit/test.h>
+#include <kunit/visibility.h>
+
+#include <linux/kernel.h>
+
+#include <net/sock.h>
+#include <net/genetlink.h>
+#include <net/netns/generic.h>
+
+#include <uapi/linux/handshake.h>
+#include "handshake.h"
+
+MODULE_IMPORT_NS(EXPORTED_FOR_KUNIT_TESTING);
+
+static int test_accept_func(struct handshake_req *req, struct genl_info *info,
+			    int fd)
+{
+	return 0;
+}
+
+static void test_done_func(struct handshake_req *req, unsigned int status,
+			   struct genl_info *info)
+{
+}
+
+struct handshake_req_alloc_test_param {
+	const char			*desc;
+	struct handshake_proto		*proto;
+	gfp_t				gfp;
+	bool				expect_success;
+};
+
+static struct handshake_proto handshake_req_alloc_proto_2 = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_NONE,
+};
+
+static struct handshake_proto handshake_req_alloc_proto_3 = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_MAX,
+};
+
+static struct handshake_proto handshake_req_alloc_proto_4 = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
+};
+
+static struct handshake_proto handshake_req_alloc_proto_5 = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
+	.hp_accept		= test_accept_func,
+};
+
+static struct handshake_proto handshake_req_alloc_proto_6 = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
+	.hp_privsize		= UINT_MAX,
+	.hp_accept		= test_accept_func,
+	.hp_done		= test_done_func,
+};
+
+static struct handshake_proto handshake_req_alloc_proto_good = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
+	.hp_accept		= test_accept_func,
+	.hp_done		= test_done_func,
+};
+
+static const
+struct handshake_req_alloc_test_param handshake_req_alloc_params[] = {
+	{
+		.desc			= "handshake_req_alloc NULL proto",
+		.proto			= NULL,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= false,
+	},
+	{
+		.desc			= "handshake_req_alloc CLASS_NONE",
+		.proto			= &handshake_req_alloc_proto_2,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= false,
+	},
+	{
+		.desc			= "handshake_req_alloc CLASS_MAX",
+		.proto			= &handshake_req_alloc_proto_3,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= false,
+	},
+	{
+		.desc			= "handshake_req_alloc no callbacks",
+		.proto			= &handshake_req_alloc_proto_4,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= false,
+	},
+	{
+		.desc			= "handshake_req_alloc no done callback",
+		.proto			= &handshake_req_alloc_proto_5,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= false,
+	},
+	{
+		.desc			= "handshake_req_alloc excessive privsize",
+		.proto			= &handshake_req_alloc_proto_6,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= false,
+	},
+	{
+		.desc			= "handshake_req_alloc all good",
+		.proto			= &handshake_req_alloc_proto_good,
+		.gfp			= GFP_KERNEL,
+		.expect_success		= true,
+	},
+};
+
+static void
+handshake_req_alloc_get_desc(const struct handshake_req_alloc_test_param *param,
+			     char *desc)
+{
+	strscpy(desc, param->desc, KUNIT_PARAM_DESC_SIZE);
+}
+
+/* Creates the function handshake_req_alloc_gen_params */
+KUNIT_ARRAY_PARAM(handshake_req_alloc, handshake_req_alloc_params,
+		  handshake_req_alloc_get_desc);
+
+static void handshake_req_alloc_case(struct kunit *test)
+{
+	const struct handshake_req_alloc_test_param *param = test->param_value;
+	struct handshake_req *result;
+
+	/* Arrange */
+
+	/* Act */
+	result = handshake_req_alloc(param->proto, param->gfp);
+
+	/* Assert */
+	if (param->expect_success)
+		KUNIT_EXPECT_NOT_NULL(test, result);
+	else
+		KUNIT_EXPECT_NULL(test, result);
+
+	kfree(result);
+}
+
+static void handshake_req_submit_test1(struct kunit *test)
+{
+	struct socket *sock;
+	int err, result;
+
+	/* Arrange */
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	/* Act */
+	result = handshake_req_submit(sock, NULL, GFP_KERNEL);
+
+	/* Assert */
+	KUNIT_EXPECT_EQ(test, result, -EINVAL);
+
+	sock_release(sock);
+}
+
+static void handshake_req_submit_test2(struct kunit *test)
+{
+	struct handshake_req *req;
+	int result;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	/* Act */
+	result = handshake_req_submit(NULL, req, GFP_KERNEL);
+
+	/* Assert */
+	KUNIT_EXPECT_EQ(test, result, -EINVAL);
+
+	/* handshake_req_submit() destroys @req on error */
+}
+
+static void handshake_req_submit_test3(struct kunit *test)
+{
+	struct handshake_req *req;
+	struct socket *sock;
+	int err, result;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	sock->file = NULL;
+
+	/* Act */
+	result = handshake_req_submit(sock, req, GFP_KERNEL);
+
+	/* Assert */
+	KUNIT_EXPECT_EQ(test, result, -EINVAL);
+
+	/* handshake_req_submit() destroys @req on error */
+	sock_release(sock);
+}
+
+static void handshake_req_submit_test4(struct kunit *test)
+{
+	struct handshake_req *req, *result;
+	struct socket *sock;
+	int err;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+	KUNIT_ASSERT_NOT_NULL(test, sock->sk);
+
+	err = handshake_req_submit(sock, req, GFP_KERNEL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	/* Act */
+	result = handshake_req_hash_lookup(sock->sk);
+
+	/* Assert */
+	KUNIT_EXPECT_NOT_NULL(test, result);
+	KUNIT_EXPECT_PTR_EQ(test, req, result);
+
+	handshake_req_cancel(sock->sk);
+	sock_release(sock);
+}
+
+static void handshake_req_submit_test5(struct kunit *test)
+{
+	struct handshake_req *req;
+	struct handshake_net *hn;
+	struct socket *sock;
+	struct net *net;
+	int saved, err;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+	KUNIT_ASSERT_NOT_NULL(test, sock->sk);
+
+	net = sock_net(sock->sk);
+	hn = handshake_pernet(net);
+	KUNIT_ASSERT_NOT_NULL(test, hn);
+
+	saved = hn->hn_pending;
+	hn->hn_pending = hn->hn_pending_max + 1;
+
+	/* Act */
+	err = handshake_req_submit(sock, req, GFP_KERNEL);
+
+	/* Assert */
+	KUNIT_EXPECT_EQ(test, err, -EAGAIN);
+
+	sock_release(sock);
+	hn->hn_pending = saved;
+}
+
+static void handshake_req_submit_test6(struct kunit *test)
+{
+	struct handshake_req *req1, *req2;
+	struct socket *sock;
+	int err;
+
+	/* Arrange */
+	req1 = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req1);
+	req2 = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req2);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+	KUNIT_ASSERT_NOT_NULL(test, sock->sk);
+
+	/* Act */
+	err = handshake_req_submit(sock, req1, GFP_KERNEL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	err = handshake_req_submit(sock, req2, GFP_KERNEL);
+
+	/* Assert */
+	KUNIT_EXPECT_EQ(test, err, -EBUSY);
+
+	handshake_req_cancel(sock->sk);
+	sock_release(sock);
+}
+
+static void handshake_req_cancel_test1(struct kunit *test)
+{
+	struct handshake_req *req;
+	struct socket *sock;
+	bool result;
+	int err;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+
+	err = handshake_req_submit(sock, req, GFP_KERNEL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	/* NB: handshake_req hasn't been accepted */
+
+	/* Act */
+	result = handshake_req_cancel(sock->sk);
+
+	/* Assert */
+	KUNIT_EXPECT_TRUE(test, result);
+
+	sock_release(sock);
+}
+
+static void handshake_req_cancel_test2(struct kunit *test)
+{
+	struct handshake_req *req, *next;
+	struct handshake_net *hn;
+	struct socket *sock;
+	struct net *net;
+	bool result;
+	int err;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+
+	err = handshake_req_submit(sock, req, GFP_KERNEL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	net = sock_net(sock->sk);
+	hn = handshake_pernet(net);
+	KUNIT_ASSERT_NOT_NULL(test, hn);
+
+	/* Pretend to accept this request */
+	next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD);
+	KUNIT_ASSERT_PTR_EQ(test, req, next);
+
+	/* Act */
+	result = handshake_req_cancel(sock->sk);
+
+	/* Assert */
+	KUNIT_EXPECT_TRUE(test, result);
+
+	sock_release(sock);
+}
+
+static void handshake_req_cancel_test3(struct kunit *test)
+{
+	struct handshake_req *req, *next;
+	struct handshake_net *hn;
+	struct socket *sock;
+	struct net *net;
+	bool result;
+	int err;
+
+	/* Arrange */
+	req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+
+	err = handshake_req_submit(sock, req, GFP_KERNEL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	net = sock_net(sock->sk);
+	hn = handshake_pernet(net);
+	KUNIT_ASSERT_NOT_NULL(test, hn);
+
+	/* Pretend to accept this request */
+	next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD);
+	KUNIT_ASSERT_PTR_EQ(test, req, next);
+
+	/* Pretend to complete this request */
+	handshake_complete(next, -ETIMEDOUT, NULL);
+
+	/* Act */
+	result = handshake_req_cancel(sock->sk);
+
+	/* Assert */
+	KUNIT_EXPECT_FALSE(test, result);
+
+	sock_release(sock);
+}
+
+static struct handshake_req *handshake_req_destroy_test;
+
+static void test_destroy_func(struct handshake_req *req)
+{
+	handshake_req_destroy_test = req;
+}
+
+static struct handshake_proto handshake_req_alloc_proto_destroy = {
+	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
+	.hp_accept		= test_accept_func,
+	.hp_done		= test_done_func,
+	.hp_destroy		= test_destroy_func,
+};
+
+static void handshake_req_destroy_test1(struct kunit *test)
+{
+	struct handshake_req *req;
+	struct socket *sock;
+	int err;
+
+	/* Arrange */
+	handshake_req_destroy_test = NULL;
+
+	req = handshake_req_alloc(&handshake_req_alloc_proto_destroy, GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, req);
+
+	err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP,
+			    &sock, 1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file);
+
+	err = handshake_req_submit(sock, req, GFP_KERNEL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	handshake_req_cancel(sock->sk);
+
+	/* Act */
+	sock_release(sock);
+
+	/* Assert */
+	KUNIT_EXPECT_PTR_EQ(test, handshake_req_destroy_test, req);
+}
+
+static struct kunit_case handshake_api_test_cases[] = {
+	{
+		.name			= "req_alloc API fuzzing",
+		.run_case		= handshake_req_alloc_case,
+		.generate_params	= handshake_req_alloc_gen_params,
+	},
+	{
+		.name			= "req_submit NULL req arg",
+		.run_case		= handshake_req_submit_test1,
+	},
+	{
+		.name			= "req_submit NULL sock arg",
+		.run_case		= handshake_req_submit_test2,
+	},
+	{
+		.name			= "req_submit NULL sock->file",
+		.run_case		= handshake_req_submit_test3,
+	},
+	{
+		.name			= "req_lookup works",
+		.run_case		= handshake_req_submit_test4,
+	},
+	{
+		.name			= "req_submit max pending",
+		.run_case		= handshake_req_submit_test5,
+	},
+	{
+		.name			= "req_submit multiple",
+		.run_case		= handshake_req_submit_test6,
+	},
+	{
+		.name			= "req_cancel before accept",
+		.run_case		= handshake_req_cancel_test1,
+	},
+	{
+		.name			= "req_cancel after accept",
+		.run_case		= handshake_req_cancel_test2,
+	},
+	{
+		.name			= "req_cancel after done",
+		.run_case		= handshake_req_cancel_test3,
+	},
+	{
+		.name			= "req_destroy works",
+		.run_case		= handshake_req_destroy_test1,
+	},
+	{}
+};
+
+static struct kunit_suite handshake_api_suite = {
+       .name                   = "Handshake API tests",
+       .test_cases             = handshake_api_test_cases,
+};
+
+kunit_test_suites(&handshake_api_suite);
+
+MODULE_DESCRIPTION("Test handshake upcall API functions");
+MODULE_LICENSE("GPL");
diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h
index 52568dbe24f1..4dac965c99df 100644
--- a/net/handshake/handshake.h
+++ b/net/handshake/handshake.h
@@ -49,6 +49,7 @@ enum hr_flags_bits {
 struct handshake_proto {
 	int			hp_handler_class;
 	size_t			hp_privsize;
+	unsigned long		hp_flags;
 
 	int			(*hp_accept)(struct handshake_req *req,
 					     struct genl_info *info, int fd);
@@ -58,6 +59,10 @@ struct handshake_proto {
 	void			(*hp_destroy)(struct handshake_req *req);
 };
 
+enum hp_flags_bits {
+	HANDSHAKE_F_PROTO_NOTIFY,
+};
+
 /* netlink.c */
 int handshake_genl_notify(struct net *net, const struct handshake_proto *proto,
 			  gfp_t flags);
diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c
index 7264cac04047..8ea0ff993f9f 100644
--- a/net/handshake/netlink.c
+++ b/net/handshake/netlink.c
@@ -18,6 +18,8 @@
 #include <net/genetlink.h>
 #include <net/netns/generic.h>
 
+#include <kunit/visibility.h>
+
 #include <uapi/linux/handshake.h>
 #include "handshake.h"
 #include "genl.h"
@@ -38,6 +40,10 @@ int handshake_genl_notify(struct net *net, const struct handshake_proto *proto,
 	struct sk_buff *msg;
 	void *hdr;
 
+	/* Disable notifications during unit testing */
+	if (!test_bit(HANDSHAKE_F_PROTO_NOTIFY, &proto->hp_flags))
+		return 0;
+
 	if (!genl_has_listeners(&handshake_nl_family, net,
 				proto->hp_handler_class))
 		return -ESRCH;
@@ -262,6 +268,7 @@ struct handshake_net *handshake_pernet(struct net *net)
 	return handshake_net_id ?
 		net_generic(net, handshake_net_id) : NULL;
 }
+EXPORT_SYMBOL_IF_KUNIT(handshake_pernet);
 
 static int __init handshake_init(void)
 {
diff --git a/net/handshake/request.c b/net/handshake/request.c
index d5b2bc6de057..94d5cef3e048 100644
--- a/net/handshake/request.c
+++ b/net/handshake/request.c
@@ -20,6 +20,8 @@
 #include <net/genetlink.h>
 #include <net/netns/generic.h>
 
+#include <kunit/visibility.h>
+
 #include <uapi/linux/handshake.h>
 #include "handshake.h"
 
@@ -60,6 +62,7 @@ struct handshake_req *handshake_req_hash_lookup(struct sock *sk)
 	return rhashtable_lookup_fast(&handshake_rhashtbl, &sk,
 				      handshake_rhash_params);
 }
+EXPORT_SYMBOL_IF_KUNIT(handshake_req_hash_lookup);
 
 static bool handshake_req_hash_add(struct handshake_req *req)
 {
@@ -192,6 +195,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class)
 
 	return req;
 }
+EXPORT_SYMBOL_IF_KUNIT(handshake_req_next);
 
 /**
  * handshake_req_submit - Submit a handshake request
@@ -293,6 +297,7 @@ void handshake_complete(struct handshake_req *req, unsigned int status,
 		sock_put(sk);
 	}
 }
+EXPORT_SYMBOL_IF_KUNIT(handshake_complete);
 
 /**
  * handshake_req_cancel - Cancel an in-progress handshake
diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c
index 1b8353296060..fcbeb63b4eb1 100644
--- a/net/handshake/tlshd.c
+++ b/net/handshake/tlshd.c
@@ -249,6 +249,7 @@ out:
 static const struct handshake_proto tls_handshake_proto = {
 	.hp_handler_class	= HANDSHAKE_HANDLER_CLASS_TLSHD,
 	.hp_privsize		= sizeof(struct tls_handshake_req),
+	.hp_flags		= BIT(HANDSHAKE_F_PROTO_NOTIFY),
 
 	.hp_accept		= tls_handshake_accept,
 	.hp_done		= tls_handshake_done,
-- 
cgit v1.2.3-70-g09d2


From 2569c7b8726fc06d946a4f999fb1be15b68f3f3c Mon Sep 17 00:00:00 2001
From: Feng Zhou <zhoufeng.zf@bytedance.com>
Date: Thu, 20 Apr 2023 11:27:34 +0800
Subject: bpf: support access variable length array of integer type

After this commit:
bpf: Support variable length array in tracing programs (9c5f8a1008a1)
Trace programs can access variable length array, but for structure
type. This patch adds support for integer type.

Example:
Hook load_balance
struct sched_domain {
	...
	unsigned long span[];
}

The access: sd->span[0].

Co-developed-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Feng Zhou <zhoufeng.zf@bytedance.com>
Link: https://lore.kernel.org/r/20230420032735.27760-2-zhoufeng.zf@bytedance.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 027f9f8a3551..a0887ee44e89 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -6157,11 +6157,13 @@ again:
 		if (off < moff)
 			goto error;
 
-		/* Only allow structure for now, can be relaxed for
-		 * other types later.
-		 */
+		/* allow structure and integer */
 		t = btf_type_skip_modifiers(btf, array_elem->type,
 					    NULL);
+
+		if (btf_type_is_int(t))
+			return WALK_SCALAR;
+
 		if (!btf_type_is_struct(t))
 			goto error;
 
-- 
cgit v1.2.3-70-g09d2


From 5ff54dedf35bfee7a16eb9f8fb7ecadf7d5564cb Mon Sep 17 00:00:00 2001
From: Feng Zhou <zhoufeng.zf@bytedance.com>
Date: Thu, 20 Apr 2023 11:27:35 +0800
Subject: selftests/bpf: Add test to access integer type of variable array

Add prog test for accessing integer type of variable array in tracing
program.
In addition, hook load_balance function to access sd->span[0], only
to confirm whether the load is successful. Because there is no direct
way to trigger load_balance call.

Co-developed-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Feng Zhou <zhoufeng.zf@bytedance.com>
Link: https://lore.kernel.org/r/20230420032735.27760-3-zhoufeng.zf@bytedance.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../testing/selftests/bpf/bpf_testmod/bpf_testmod.c  | 20 ++++++++++++++++++++
 .../selftests/bpf/prog_tests/access_variable_array.c | 16 ++++++++++++++++
 .../selftests/bpf/prog_tests/tracing_struct.c        |  2 ++
 .../selftests/bpf/progs/test_access_variable_array.c | 19 +++++++++++++++++++
 tools/testing/selftests/bpf/progs/tracing_struct.c   | 13 +++++++++++++
 5 files changed, 70 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/access_variable_array.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_access_variable_array.c

diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index fe847ebfb731..52785ba671e6 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -28,6 +28,11 @@ struct bpf_testmod_struct_arg_2 {
 	long b;
 };
 
+struct bpf_testmod_struct_arg_3 {
+	int a;
+	int b[];
+};
+
 __diag_push();
 __diag_ignore_all("-Wmissing-prototypes",
 		  "Global functions as their definitions will be in bpf_testmod.ko BTF");
@@ -63,6 +68,12 @@ bpf_testmod_test_struct_arg_5(void) {
 	return bpf_testmod_test_struct_arg_result;
 }
 
+noinline int
+bpf_testmod_test_struct_arg_6(struct bpf_testmod_struct_arg_3 *a) {
+	bpf_testmod_test_struct_arg_result = a->b[0];
+	return bpf_testmod_test_struct_arg_result;
+}
+
 __bpf_kfunc void
 bpf_testmod_test_mod_kfunc(int i)
 {
@@ -195,6 +206,7 @@ bpf_testmod_test_read(struct file *file, struct kobject *kobj,
 	};
 	struct bpf_testmod_struct_arg_1 struct_arg1 = {10};
 	struct bpf_testmod_struct_arg_2 struct_arg2 = {2, 3};
+	struct bpf_testmod_struct_arg_3 *struct_arg3;
 	int i = 1;
 
 	while (bpf_testmod_return_ptr(i))
@@ -206,6 +218,14 @@ bpf_testmod_test_read(struct file *file, struct kobject *kobj,
 	(void)bpf_testmod_test_struct_arg_4(struct_arg1, 1, 2, 3, struct_arg2);
 	(void)bpf_testmod_test_struct_arg_5();
 
+	struct_arg3 = kmalloc((sizeof(struct bpf_testmod_struct_arg_3) +
+				sizeof(int)), GFP_KERNEL);
+	if (struct_arg3 != NULL) {
+		struct_arg3->b[0] = 1;
+		(void)bpf_testmod_test_struct_arg_6(struct_arg3);
+		kfree(struct_arg3);
+	}
+
 	/* This is always true. Use the check to make sure the compiler
 	 * doesn't remove bpf_testmod_loop_test.
 	 */
diff --git a/tools/testing/selftests/bpf/prog_tests/access_variable_array.c b/tools/testing/selftests/bpf/prog_tests/access_variable_array.c
new file mode 100644
index 000000000000..08131782437c
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/access_variable_array.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2022 Bytedance */
+
+#include <test_progs.h>
+#include "test_access_variable_array.skel.h"
+
+void test_access_variable_array(void)
+{
+	struct test_access_variable_array *skel;
+
+	skel = test_access_variable_array__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "test_access_variable_array__open_and_load"))
+		return;
+
+	test_access_variable_array__destroy(skel);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/tracing_struct.c b/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
index 48dc9472e160..1c75a32186d6 100644
--- a/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
+++ b/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
@@ -53,6 +53,8 @@ static void test_fentry(void)
 
 	ASSERT_EQ(skel->bss->t5_ret, 1, "t5 ret");
 
+	ASSERT_EQ(skel->bss->t6, 1, "t6 ret");
+
 	tracing_struct__detach(skel);
 destroy_skel:
 	tracing_struct__destroy(skel);
diff --git a/tools/testing/selftests/bpf/progs/test_access_variable_array.c b/tools/testing/selftests/bpf/progs/test_access_variable_array.c
new file mode 100644
index 000000000000..808c49b79889
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_access_variable_array.c
@@ -0,0 +1,19 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Bytedance */
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+unsigned long span = 0;
+
+SEC("fentry/load_balance")
+int BPF_PROG(fentry_fentry, int this_cpu, struct rq *this_rq,
+		struct sched_domain *sd)
+{
+	span = sd->span[0];
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/tracing_struct.c b/tools/testing/selftests/bpf/progs/tracing_struct.c
index e718f0ebee7d..c435a3a8328a 100644
--- a/tools/testing/selftests/bpf/progs/tracing_struct.c
+++ b/tools/testing/selftests/bpf/progs/tracing_struct.c
@@ -13,12 +13,18 @@ struct bpf_testmod_struct_arg_2 {
 	long b;
 };
 
+struct bpf_testmod_struct_arg_3 {
+	int a;
+	int b[];
+};
+
 long t1_a_a, t1_a_b, t1_b, t1_c, t1_ret, t1_nregs;
 __u64 t1_reg0, t1_reg1, t1_reg2, t1_reg3;
 long t2_a, t2_b_a, t2_b_b, t2_c, t2_ret;
 long t3_a, t3_b, t3_c_a, t3_c_b, t3_ret;
 long t4_a_a, t4_b, t4_c, t4_d, t4_e_a, t4_e_b, t4_ret;
 long t5_ret;
+int t6;
 
 SEC("fentry/bpf_testmod_test_struct_arg_1")
 int BPF_PROG2(test_struct_arg_1, struct bpf_testmod_struct_arg_2, a, int, b, int, c)
@@ -117,4 +123,11 @@ int BPF_PROG2(test_struct_arg_10, int, ret)
 	return 0;
 }
 
+SEC("fentry/bpf_testmod_test_struct_arg_6")
+int BPF_PROG2(test_struct_arg_11, struct bpf_testmod_struct_arg_3 *, a)
+{
+	t6 = a->b[0];
+	return 0;
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-70-g09d2


From c5050696588660226fe6cc90a360b62d52a87e32 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Tue, 18 Apr 2023 12:28:04 +0300
Subject: wifi: iwlwifi: Update support for b0 version

Add support for B0 version of MAC of MR device

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.5dca1ea7a0cf.I87932e1e216a1940eeae8824071ecb777f4c034f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  | 55 +++++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  5 +++
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c   | 35 +++++++++++++---
 3 files changed, 90 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index 532d14f61253..b6f82510e980 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -50,6 +50,11 @@
 #define IWL_MA_A_GF4_A_FW_PRE		"iwlwifi-ma-a0-gf4-a0-"
 #define IWL_MA_A_MR_A_FW_PRE		"iwlwifi-ma-a0-mr-a0-"
 #define IWL_MA_A_FM_A_FW_PRE		"iwlwifi-ma-a0-fm-a0-"
+#define IWL_MA_B_HR_B_FW_PRE		"iwlwifi-ma-b0-hr-b0-"
+#define IWL_MA_B_GF_A_FW_PRE		"iwlwifi-ma-b0-gf-a0-"
+#define IWL_MA_B_GF4_A_FW_PRE		"iwlwifi-ma-b0-gf4-a0-"
+#define IWL_MA_B_MR_A_FW_PRE		"iwlwifi-ma-b0-mr-a0-"
+#define IWL_MA_B_FM_A_FW_PRE		"iwlwifi-ma-b0-fm-a0-"
 #define IWL_SNJ_A_MR_A_FW_PRE		"iwlwifi-SoSnj-a0-mr-a0-"
 #define IWL_BZ_A_HR_A_FW_PRE		"iwlwifi-bz-a0-hr-b0-"
 #define IWL_BZ_A_HR_B_FW_PRE		"iwlwifi-bz-a0-hr-b0-"
@@ -119,6 +124,16 @@
 	IWL_MA_A_MR_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_MA_A_FM_A_FW_MODULE_FIRMWARE(api)		\
 	IWL_MA_A_FM_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_MA_B_HR_B_FW_MODULE_FIRMWARE(api)		\
+	IWL_MA_B_HR_B_FW_PRE __stringify(api) ".ucode"
+#define IWL_MA_B_GF_A_FW_MODULE_FIRMWARE(api)		\
+	IWL_MA_B_GF_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_MA_B_GF4_A_FW_MODULE_FIRMWARE(api)		\
+	IWL_MA_B_GF4_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_MA_B_MR_A_FW_MODULE_FIRMWARE(api) \
+	IWL_MA_B_MR_A_FW_PRE __stringify(api) ".ucode"
+#define IWL_MA_B_FM_A_FW_MODULE_FIRMWARE(api)		\
+	IWL_MA_B_FM_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_SNJ_A_MR_A_MODULE_FIRMWARE(api) \
 	IWL_SNJ_A_MR_A_FW_PRE __stringify(api) ".ucode"
 #define IWL_BZ_A_HR_A_MODULE_FIRMWARE(api) \
@@ -891,6 +906,41 @@ const struct iwl_cfg iwl_cfg_ma_a0_ms_a0 = {
 	.num_rbds = IWL_NUM_RBDS_AX210_HE,
 };
 
+const struct iwl_cfg iwl_cfg_ma_b0_fm_a0 = {
+	.fw_name_pre = IWL_MA_B_FM_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
+const struct iwl_cfg iwl_cfg_ma_b0_hr_b0 = {
+	.fw_name_pre = IWL_MA_B_HR_B_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
+const struct iwl_cfg iwl_cfg_ma_b0_gf_a0 = {
+	.fw_name_pre = IWL_MA_B_GF_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
+const struct iwl_cfg iwl_cfg_ma_b0_gf4_a0 = {
+	.fw_name_pre = IWL_MA_B_GF4_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
+const struct iwl_cfg iwl_cfg_ma_b0_mr_a0 = {
+	.fw_name_pre = IWL_MA_B_MR_A_FW_PRE,
+	.uhb_supported = true,
+	IWL_DEVICE_AX210,
+	.num_rbds = IWL_NUM_RBDS_AX210_HE,
+};
+
 const struct iwl_cfg iwl_cfg_so_a0_ms_a0 = {
 	.fw_name_pre = IWL_SO_A_MR_A_FW_PRE,
 	.uhb_supported = false,
@@ -1149,6 +1199,11 @@ MODULE_FIRMWARE(IWL_MA_A_GF_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_MA_A_GF4_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_MA_A_MR_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_MA_A_FM_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_MA_B_HR_B_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_MA_B_GF_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_MA_B_GF4_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_MA_B_MR_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
+MODULE_FIRMWARE(IWL_MA_B_FM_A_FW_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_SNJ_A_MR_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_HR_A_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
 MODULE_FIRMWARE(IWL_BZ_A_HR_B_MODULE_FIRMWARE(IWL_22000_UCODE_API_MAX));
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index c4ac951de19f..411b7d4fcc9a 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -640,6 +640,11 @@ extern const struct iwl_cfg iwl_cfg_ma_a0_gf4_a0;
 extern const struct iwl_cfg iwl_cfg_ma_a0_mr_a0;
 extern const struct iwl_cfg iwl_cfg_ma_a0_ms_a0;
 extern const struct iwl_cfg iwl_cfg_ma_a0_fm_a0;
+extern const struct iwl_cfg iwl_cfg_ma_b0_hr_b0;
+extern const struct iwl_cfg iwl_cfg_ma_b0_gf_a0;
+extern const struct iwl_cfg iwl_cfg_ma_b0_gf4_a0;
+extern const struct iwl_cfg iwl_cfg_ma_b0_mr_a0;
+extern const struct iwl_cfg iwl_cfg_ma_b0_fm_a0;
 extern const struct iwl_cfg iwl_cfg_snj_a0_mr_a0;
 extern const struct iwl_cfg iwl_cfg_snj_a0_ms_a0;
 extern const struct iwl_cfg iwl_cfg_so_a0_hr_a0;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 31b575f35410..dba112394838 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1058,27 +1058,27 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 
 /* Ma */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_hr_b0, iwl_ax201_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_gf_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_gf4_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_mr_a0, iwl_ax221_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_A_STEP,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_ma_a0_fm_a0, iwl_ax231_name),
@@ -1087,6 +1087,31 @@ static const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
 		      iwl_cfg_snj_a0_mr_a0, iwl_ax221_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_ma_b0_hr_b0, iwl_ax201_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_ma_b0_gf_a0, iwl_ax211_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_CDB, IWL_CFG_ANY,
+		      iwl_cfg_ma_b0_gf4_a0, iwl_ax211_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_MR, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_ma_b0_mr_a0, iwl_ax221_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_MA, SILICON_B_STEP,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB, IWL_CFG_ANY,
+		      iwl_cfg_ma_b0_fm_a0, iwl_ax231_name),
 
 /* So with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-- 
cgit v1.2.3-70-g09d2


From 3d90d2f4a018fe8cfd65068bc6350b6222be4852 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Apr 2023 12:28:05 +0300
Subject: wifi: iwlwifi: fw: fix memory leak in debugfs

Fix a memory leak that occurs when reading the fw_info
file all the way, since we return NULL indicating no
more data, but don't free the status tracking object.

Fixes: 36dfe9ac6e8b ("iwlwifi: dump api version in yaml format")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.239e501b3b8d.I4268f87809ef91209cbcd748eee0863195e70fa2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/debugfs.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c b/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
index 43e997283db0..607e07ed2477 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
@@ -317,8 +317,10 @@ static void *iwl_dbgfs_fw_info_seq_next(struct seq_file *seq,
 	const struct iwl_fw *fw = priv->fwrt->fw;
 
 	*pos = ++state->pos;
-	if (*pos >= fw->ucode_capa.n_cmd_versions)
+	if (*pos >= fw->ucode_capa.n_cmd_versions) {
+		kfree(state);
 		return NULL;
+	}
 
 	return state;
 }
-- 
cgit v1.2.3-70-g09d2


From 7b41a99ce2d494d01e8fcb25e7113f143a0f2155 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Apr 2023 12:28:06 +0300
Subject: wifi: iwlwifi: mvm: fix MIC removal confusion

The RADA/firmware collaborate on MIC stripping in the following
way:
 - the firmware fills the IWL_RX_MPDU_MFLG1_MIC_CRC_LEN_MASK
   value for how many words need to be removed at the end of
   the frame, CRC and, if decryption was done, MIC
 - if the RADA is active, it will
   - remove that much from the end of the frame
   - zero the value in IWL_RX_MPDU_MFLG1_MIC_CRC_LEN_MASK

As a consequence, the only thing the driver should need to do
is to
 - unconditionally tell mac80211 that the MIC was removed
   if decryption was already done
 - remove as much as IWL_RX_MPDU_MFLG1_MIC_CRC_LEN_MASK says
   at the end of the frame, since either RADA did it and then
   the value is 0, or RADA was disabled and then the value is
   whatever should be removed to strip both CRC & MIC

However, all this code was historically grown and getting a
bit confused. Originally, we were indicating that the MIC was
not stripped, which is the version of the code upstreamed in
commit 780e87c29e77 ("iwlwifi: mvm: add 9000 series RX processing")
which indicated RX_FLAG_DECRYPTED in iwl_mvm_rx_crypto().

We later had a commit to change that to also indicate that the
MIC was stripped, adding RX_FLAG_MIC_STRIPPED. However, this was
then "fixed" later to only do that conditionally on RADA being
enabled, since otherwise RADA didn't strip the MIC bytes yet.
At the time, we were also always including the FCS if the RADA
was not enabled, so that was still broken wrt. the FCS if the
RADA isn't enabled - but that's a pretty rare case. Notably
though, it does happen for management frames, where we do need
to remove the MIC and CRC but the RADA is disabled.

Later, in commit 40a0b38d7a7f ("iwlwifi: mvm: Fix calculation of
frame length"), we changed this again, upstream this was just a
single commit, but internally it was split into first the correct
commit and then an additional fix that reduced the number of bytes
that are removed by crypt_len. Note that this is clearly wrong
since crypt_len indicates the length of the PN header (always 8),
not the length of the MIC (8 or 16 depending on algorithm).
However, this additional fix mostly canceled the other bugs,
apart from the confusion about the size of the MIC.

To fix this correctly, remove all those additional workarounds.
We really should always indicate to mac80211 the MIC was stripped
(it cannot use it anyway if decryption was already done), and also
always actually remove it and the CRC regardless of the RADA being
enabled or not. That's simple though, the value indicated in the
metadata is zeroed by the RADA if it's enabled and used the value,
so there's no need to check if it's enabled or not.

Notably then, this fixes the MIC size confusion, letting us receive
GCMP-256 encrypted management frames correctly that would otherwise
be reported to mac80211 8 bytes too short since the RADA is turned
off for them, crypt_len is 8, but the MIC size is 16, so when we do
the adjustment based on IWL_RX_MPDU_MFLG1_MIC_CRC_LEN_MASK (which
indicates 20 bytes to remove) we remove 12 bytes but indicate then
to mac80211 the MIC is still present, so mac80211 again removes the
MIC of 16 bytes, for an overall removal of 28 rather than 20 bytes.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.81345b6ab0cd.Ibe0348defb6cce11c99929a1f049e60b5cfc150c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 28 +++++----------------------
 1 file changed, 5 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 9dbf14fa0ca7..542f6658f2d4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -106,28 +106,12 @@ static int iwl_mvm_create_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	/*
 	 * For non monitor interface strip the bytes the RADA might not have
-	 * removed. As monitor interface cannot exist with other interfaces
-	 * this removal is safe.
+	 * removed (it might be disabled, e.g. for mgmt frames). As a monitor
+	 * interface cannot exist with other interfaces, this removal is safe
+	 * and sufficient, in monitor mode there's no decryption being done.
 	 */
-	if (mic_crc_len && !ieee80211_hw_check(mvm->hw, RX_INCLUDES_FCS)) {
-		u32 pkt_flags = le32_to_cpu(pkt->len_n_flags);
-
-		/*
-		 * If RADA was not enabled then decryption was not performed so
-		 * the MIC cannot be removed.
-		 */
-		if (!(pkt_flags & FH_RSCSR_RADA_EN)) {
-			if (WARN_ON(crypt_len > mic_crc_len))
-				return -EINVAL;
-
-			mic_crc_len -= crypt_len;
-		}
-
-		if (WARN_ON(mic_crc_len > len))
-			return -EINVAL;
-
+	if (len > mic_crc_len && !ieee80211_hw_check(mvm->hw, RX_INCLUDES_FCS))
 		len -= mic_crc_len;
-	}
 
 	/* If frame is small enough to fit in skb->head, pull it completely.
 	 * If not, only pull ieee80211_hdr (including crypto if present, and
@@ -411,9 +395,7 @@ static int iwl_mvm_rx_crypto(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		if (!(status & IWL_RX_MPDU_STATUS_MIC_OK))
 			return -1;
 
-		stats->flag |= RX_FLAG_DECRYPTED;
-		if (pkt_flags & FH_RSCSR_RADA_EN)
-			stats->flag |= RX_FLAG_MIC_STRIPPED;
+		stats->flag |= RX_FLAG_DECRYPTED | RX_FLAG_MIC_STRIPPED;
 		*crypt_len = IEEE80211_CCMP_HDR_LEN;
 		return 0;
 	case IWL_RX_MPDU_STATUS_SEC_TKIP:
-- 
cgit v1.2.3-70-g09d2


From f4d4ba4613dc92a58a8c051f7b6bcb0276b1fecc Mon Sep 17 00:00:00 2001
From: Alon Giladi <alon.giladi@intel.com>
Date: Tue, 18 Apr 2023 12:28:07 +0300
Subject: wifi: iwlwifi: fw: fix argument to efi.get_variable

We should pass the newly allocated data to fill.

Signed-off-by: Alon Giladi <alon.giladi@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.aaa6d8874442.I734841c71aad9564cb22c50f2737aaff489fadaf@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/uefi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/uefi.c b/drivers/net/wireless/intel/iwlwifi/fw/uefi.c
index 0b6f694cf30d..01afea33c38c 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/uefi.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/uefi.c
@@ -222,7 +222,7 @@ void *iwl_uefi_get_reduced_power(struct iwl_trans *trans, size_t *len)
 		return ERR_PTR(-ENOMEM);
 
 	status = efi.get_variable(IWL_UEFI_REDUCED_POWER_NAME, &IWL_EFI_VAR_GUID,
-				  NULL, &package_size, data);
+				  NULL, &package_size, package);
 	if (status != EFI_SUCCESS) {
 		IWL_DEBUG_FW(trans,
 			     "Reduced Power UEFI variable not found 0x%lx (len %lu)\n",
-- 
cgit v1.2.3-70-g09d2


From 457d7fb03e6c3d73fbb509bd85fc4b02d1ab405e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Apr 2023 12:28:08 +0300
Subject: wifi: iwlwifi: mvm: fix potential memory leak

If we do get multiple notifications from firmware, then
we might have allocated 'notif', but don't free it. Fix
that by checking for duplicates before allocation.

Fixes: 4da46a06d443 ("wifi: iwlwifi: mvm: Add support for wowlan info notification")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.116758321cc4.I8bdbcbb38c89ac637eaa20dda58fa9165b25893a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index 798a7a1299e1..37aa4676dc94 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -2722,6 +2722,13 @@ static bool iwl_mvm_wait_d3_notif(struct iwl_notif_wait_data *notif_wait,
 	case WIDE_ID(PROT_OFFLOAD_GROUP, WOWLAN_INFO_NOTIFICATION): {
 		struct iwl_wowlan_info_notif *notif;
 
+		if (d3_data->notif_received & IWL_D3_NOTIF_WOWLAN_INFO) {
+			/* We might get two notifications due to dual bss */
+			IWL_DEBUG_WOWLAN(mvm,
+					 "Got additional wowlan info notification\n");
+			break;
+		}
+
 		if (wowlan_info_ver < 2) {
 			struct iwl_wowlan_info_notif_v1 *notif_v1 = (void *)pkt->data;
 
@@ -2740,13 +2747,6 @@ static bool iwl_mvm_wait_d3_notif(struct iwl_notif_wait_data *notif_wait,
 			notif = (void *)pkt->data;
 		}
 
-		if (d3_data->notif_received & IWL_D3_NOTIF_WOWLAN_INFO) {
-			/* We might get two notifications due to dual bss */
-			IWL_DEBUG_WOWLAN(mvm,
-					 "Got additional wowlan info notification\n");
-			break;
-		}
-
 		d3_data->notif_received |= IWL_D3_NOTIF_WOWLAN_INFO;
 		len = iwl_rx_packet_payload_len(pkt);
 		iwl_mvm_parse_wowlan_info_notif(mvm, notif, d3_data->status,
-- 
cgit v1.2.3-70-g09d2


From 3e75668be54eff49c86d498623e3cf3d3a23169a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Apr 2023 12:28:09 +0300
Subject: wifi: iwlwifi: mvm: prefer RCU_INIT_POINTER()

For constant values we don't need rcu_assign_pointer(),
use RCU_INIT_POINTER() instead.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.7b400d21a27f.Iccdef9d777677390a9881c88b06c0ed13a83d978@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 4 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c      | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index ac63af549416..51399284e097 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3314,8 +3314,8 @@ void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 						     lockdep_is_held(&mvm->mutex));
 		sta_id = link_sta->sta_id;
 		if (sta == rcu_access_pointer(mvm->fw_id_to_mac_id[sta_id])) {
-			rcu_assign_pointer(mvm->fw_id_to_mac_id[sta_id],
-					   ERR_PTR(-ENOENT));
+			RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta_id],
+					 ERR_PTR(-ENOENT));
 			RCU_INIT_POINTER(mvm->fw_id_to_link_sta[sta_id], NULL);
 		}
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index d323d119c334..5469d634e289 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2152,7 +2152,7 @@ int iwl_mvm_allocate_int_sta(struct iwl_mvm *mvm,
 	sta->type = type;
 
 	/* put a non-NULL value so iterating over the stations won't stop */
-	rcu_assign_pointer(mvm->fw_id_to_mac_id[sta->sta_id], ERR_PTR(-EINVAL));
+	RCU_INIT_POINTER(mvm->fw_id_to_mac_id[sta->sta_id], ERR_PTR(-EINVAL));
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 0120e6b3e33dac3d0812ed93677d872e00dd4564 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Tue, 18 Apr 2023 12:28:10 +0300
Subject: wifi: iwlwifi: mvm: enable support for MLO APIs

Enable driver's support for MLO APIs to unlock this functionality.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://lore.kernel.org/r/20230418122405.0ae0dd6f0481.Iec993cf0f28eacb2483fb9d1e755b0b2fd62e163@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 51399284e097..0f01b62357c6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -305,6 +305,11 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	ieee80211_hw_set(hw, BUFF_MMPDU_TXQ);
 	ieee80211_hw_set(hw, STA_MMPDU_TXQ);
 
+	/* Set this early since we need to have it for the check below */
+	if (mvm->mld_api_is_used &&
+	    mvm->trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ)
+		hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_MLO;
+
 	/* With MLD FW API, it tracks timing by itself,
 	 * no need for any timing from the host
 	 */
-- 
cgit v1.2.3-70-g09d2


From 0d2558838ee815540eb89feb9a468cb88b189661 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Tue, 18 Apr 2023 12:28:11 +0300
Subject: wifi: iwlwifi: modify scan request and results when in link
 protection

When CSME is connected and has link protection set, the driver must
connect to the same AP CSME is connected to.
When in link protection, modify scan request parameters to include
only the channel of the AP CSME is connected to and scan for the
same SSID. In addition, filter the scan results to include only
results from the same AP. This will make sure the driver will connect
to the same AP and will do it fast enough to keep the session alive.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.c1b55de3d704.I3895eebe18b3b672607695c887d728e113fc85ec@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h  | 25 +++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c  |  2 +
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c |  4 +-
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 78 +++++++++++++++++++++++++++
 4 files changed, 107 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 203eb7233c77..6e7470d3a826 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -826,6 +826,12 @@ struct iwl_time_sync_data {
 	bool active;
 };
 
+struct iwl_mei_scan_filter {
+	bool is_mei_limited_scan;
+	struct sk_buff_head scan_res;
+	struct work_struct scan_work;
+};
+
 struct iwl_mvm {
 	/* for logger access */
 	struct device *dev;
@@ -1177,6 +1183,8 @@ struct iwl_mvm {
 	bool pldr_sync;
 
 	struct iwl_time_sync_data time_sync;
+
+	struct iwl_mei_scan_filter mei_scan_filter;
 };
 
 /* Extract MVM priv from op_mode and _hw */
@@ -2304,6 +2312,7 @@ void iwl_mvm_event_frame_timeout_callback(struct iwl_mvm *mvm,
 					  struct ieee80211_vif *vif,
 					  const struct ieee80211_sta *sta,
 					  u16 tid);
+void iwl_mvm_mei_scan_filter_init(struct iwl_mei_scan_filter *mei_scan_filter);
 
 void iwl_mvm_ptp_init(struct iwl_mvm *mvm);
 void iwl_mvm_ptp_remove(struct iwl_mvm *mvm);
@@ -2515,6 +2524,22 @@ static inline void iwl_mvm_mei_set_sw_rfkill_state(struct iwl_mvm *mvm)
 					 sw_rfkill);
 }
 
+static inline bool iwl_mvm_mei_filter_scan(struct iwl_mvm *mvm,
+					   struct sk_buff *skb)
+{
+	struct ieee80211_mgmt *mgmt = (void *)skb->data;
+
+	if (mvm->mei_scan_filter.is_mei_limited_scan &&
+	    (ieee80211_is_probe_resp(mgmt->frame_control) ||
+	     ieee80211_is_beacon(mgmt->frame_control))) {
+		skb_queue_tail(&mvm->mei_scan_filter.scan_res, skb);
+		schedule_work(&mvm->mei_scan_filter.scan_work);
+		return true;
+	}
+
+	return false;
+}
+
 void iwl_mvm_send_roaming_forbidden_event(struct iwl_mvm *mvm,
 					  struct ieee80211_vif *vif,
 					  bool forbidden);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 8bb865e708dc..26c3f00d6f7d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1375,6 +1375,8 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 
 	mvm->mei_registered = !iwl_mei_register(mvm, &mei_ops);
 
+	iwl_mvm_mei_scan_filter_init(&mvm->mei_scan_filter);
+
 	if (iwl_mvm_start_get_nvm(mvm)) {
 		/*
 		 * Getting NVM failed while CSME is the owner, but we are
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 542f6658f2d4..e1d02c260e69 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -2588,10 +2588,10 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 	}
 
 	if (!iwl_mvm_reorder(mvm, napi, queue, sta, skb, desc) &&
-	    likely(!iwl_mvm_time_sync_frame(mvm, skb, hdr->addr2)))
+	    likely(!iwl_mvm_time_sync_frame(mvm, skb, hdr->addr2)) &&
+	    likely(!iwl_mvm_mei_filter_scan(mvm, skb)))
 		iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, queue, sta,
 						link_sta);
-
 out:
 	rcu_read_unlock();
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 1e20f9538640..fe7cb33d5593 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -2624,6 +2624,80 @@ static const struct iwl_scan_umac_handler iwl_scan_umac_handlers[] = {
 	IWL_SCAN_UMAC_HANDLER(12),
 };
 
+static void iwl_mvm_mei_scan_work(struct work_struct *wk)
+{
+	struct iwl_mei_scan_filter *scan_filter =
+		container_of(wk, struct iwl_mei_scan_filter, scan_work);
+	struct iwl_mvm *mvm =
+		container_of(scan_filter, struct iwl_mvm, mei_scan_filter);
+	struct iwl_mvm_csme_conn_info *info;
+	struct sk_buff *skb;
+	u8 bssid[ETH_ALEN];
+
+	mutex_lock(&mvm->mutex);
+	info = iwl_mvm_get_csme_conn_info(mvm);
+	memcpy(bssid, info->conn_info.bssid, ETH_ALEN);
+	mutex_unlock(&mvm->mutex);
+
+	while ((skb = skb_dequeue(&scan_filter->scan_res))) {
+		struct ieee80211_mgmt *mgmt = (void *)skb->data;
+
+		if (!memcmp(mgmt->bssid, bssid, ETH_ALEN))
+			ieee80211_rx_irqsafe(mvm->hw, skb);
+		else
+			kfree_skb(skb);
+	}
+}
+
+void iwl_mvm_mei_scan_filter_init(struct iwl_mei_scan_filter *mei_scan_filter)
+{
+	skb_queue_head_init(&mei_scan_filter->scan_res);
+	INIT_WORK(&mei_scan_filter->scan_work, iwl_mvm_mei_scan_work);
+}
+
+/* In case CSME is connected and has link protection set, this function will
+ * override the scan request to scan only the associated channel and only for
+ * the associated SSID.
+ */
+static void iwl_mvm_mei_limited_scan(struct iwl_mvm *mvm,
+				     struct iwl_mvm_scan_params *params)
+{
+	struct iwl_mvm_csme_conn_info *info = iwl_mvm_get_csme_conn_info(mvm);
+	struct iwl_mei_conn_info *conn_info;
+	struct ieee80211_channel *chan;
+
+	if (!info) {
+		IWL_DEBUG_SCAN(mvm, "mei_limited_scan: no connection info\n");
+		return;
+	}
+
+	conn_info = &info->conn_info;
+	if (!info->conn_info.lp_state || !info->conn_info.ssid_len)
+		return;
+
+	if (!params->n_channels || !params->n_ssids)
+		return;
+
+	mvm->mei_scan_filter.is_mei_limited_scan = true;
+
+	chan = ieee80211_get_channel(mvm->hw->wiphy,
+				     ieee80211_channel_to_frequency(conn_info->channel,
+								    conn_info->band));
+	if (!chan) {
+		IWL_DEBUG_SCAN(mvm,
+			       "Failed to get CSME channel (chan=%u band=%u)\n",
+			       conn_info->channel, conn_info->band);
+		return;
+	}
+
+	params->n_channels = 1;
+	params->channels[0] = chan;
+
+	params->n_ssids = 1;
+	params->ssids[0].ssid_len = conn_info->ssid_len;
+	memcpy(params->ssids[0].ssid, conn_info->ssid, conn_info->ssid_len);
+}
+
 static int iwl_mvm_build_scan_cmd(struct iwl_mvm *mvm,
 				  struct ieee80211_vif *vif,
 				  struct iwl_host_cmd *hcmd,
@@ -2636,6 +2710,8 @@ static int iwl_mvm_build_scan_cmd(struct iwl_mvm *mvm,
 	lockdep_assert_held(&mvm->mutex);
 	memset(mvm->scan_cmd, 0, mvm->scan_cmd_size);
 
+	iwl_mvm_mei_limited_scan(mvm, params);
+
 	if (!fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_UMAC_SCAN)) {
 		hcmd->id = SCAN_OFFLOAD_REQUEST_CMD;
 
@@ -2992,6 +3068,8 @@ void iwl_mvm_rx_umac_scan_complete_notif(struct iwl_mvm *mvm,
 	u32 uid = __le32_to_cpu(notif->uid);
 	bool aborted = (notif->status == IWL_SCAN_OFFLOAD_ABORTED);
 
+	mvm->mei_scan_filter.is_mei_limited_scan = false;
+
 	if (WARN_ON(!(mvm->scan_uid_status[uid] & mvm->scan_status)))
 		return;
 
-- 
cgit v1.2.3-70-g09d2


From 85c78af4e65053defeca958503fcfb0ebc1939e6 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Tue, 18 Apr 2023 12:28:12 +0300
Subject: wifi: iwlwifi: mei: make mei filtered scan more aggressive

When mei filtered scan is performed, it must find the AP on the first
scan, otherwise CSME will take the ownership of the NIC.
Make this scan more aggressive by scanning the channel the AP is
supposed to be on (as reported by CSME) several times.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.47e383b10b18.I14340a118acdb19ecb7214e7ff413054c77bd99c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index fe7cb33d5593..175615755d9d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -45,6 +45,9 @@
 /* minimal number of 2GHz and 5GHz channels in the regular scan request */
 #define IWL_MVM_6GHZ_PASSIVE_SCAN_MIN_CHANS 4
 
+/* Number of iterations on the channel for mei filtered scan */
+#define IWL_MEI_SCAN_NUM_ITER	5U
+
 struct iwl_mvm_scan_timing_params {
 	u32 suspend_time;
 	u32 max_out_time;
@@ -2665,6 +2668,7 @@ static void iwl_mvm_mei_limited_scan(struct iwl_mvm *mvm,
 	struct iwl_mvm_csme_conn_info *info = iwl_mvm_get_csme_conn_info(mvm);
 	struct iwl_mei_conn_info *conn_info;
 	struct ieee80211_channel *chan;
+	int scan_iters, i;
 
 	if (!info) {
 		IWL_DEBUG_SCAN(mvm, "mei_limited_scan: no connection info\n");
@@ -2690,8 +2694,16 @@ static void iwl_mvm_mei_limited_scan(struct iwl_mvm *mvm,
 		return;
 	}
 
-	params->n_channels = 1;
-	params->channels[0] = chan;
+	/* The mei filtered scan must find the AP, otherwise CSME will
+	 * take the NIC ownership. Add several iterations on the channel to
+	 * make the scan more robust.
+	 */
+	scan_iters = min(IWL_MEI_SCAN_NUM_ITER, params->n_channels);
+	params->n_channels = scan_iters;
+	for (i = 0; i < scan_iters; i++)
+		params->channels[i] = chan;
+
+	IWL_DEBUG_SCAN(mvm, "Mei scan: num iterations=%u\n", scan_iters);
 
 	params->n_ssids = 1;
 	params->ssids[0].ssid_len = conn_info->ssid_len;
-- 
cgit v1.2.3-70-g09d2


From 06ce23ad57c8e378b86ef3f439b2e08bcb5d05eb Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Tue, 18 Apr 2023 12:28:13 +0300
Subject: wifi: iwlwifi: mei: re-ask for ownership after it was taken by CSME

When the host disconnects from the AP CSME takes ownership right away.
Since the driver never asks for ownership again wifi is left in rfkill
until CSME releases the NIC, although in many cases the host could
re-connect shortly after the disconnection. To allow the host to
recover from occasional disconnection, re-ask for ownership to let
the host connect again.
Allow one minute before re-asking for ownership to avoid too frequent
ownership transitions.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.a6c6ebc48f2d.I8a17003b86e71b3567521cc69864b9cbe9553ea9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mei/main.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mei/main.c b/drivers/net/wireless/intel/iwlwifi/mei/main.c
index 67dfb77fedf7..f1e0fe806a00 100644
--- a/drivers/net/wireless/intel/iwlwifi/mei/main.c
+++ b/drivers/net/wireless/intel/iwlwifi/mei/main.c
@@ -31,6 +31,11 @@ MODULE_LICENSE("GPL");
 #define MEI_WLAN_UUID UUID_LE(0x13280904, 0x7792, 0x4fcb, \
 			      0xa1, 0xaa, 0x5e, 0x70, 0xcb, 0xb1, 0xe8, 0x65)
 
+/* After CSME takes ownership, it won't release it for 60 seconds to avoid
+ * frequent ownership transitions.
+ */
+#define MEI_OWNERSHIP_RETAKE_TIMEOUT_MS	msecs_to_jiffies(60000)
+
 /*
  * Since iwlwifi calls iwlmei without any context, hold a pointer to the
  * mei_cl_device structure here.
@@ -156,6 +161,8 @@ struct iwl_mei_filters {
  *	accessed without the mutex.
  * @netdev_work: used to defer registering and unregistering of the netdev to
  *	avoid taking the rtnl lock in the SAP messages handlers.
+ * @ownership_dwork: used to re-ask for NIC ownership after ownership was taken
+ *	by CSME or when a previous ownership request failed.
  * @sap_seq_no: the sequence number for the SAP messages
  * @seq_no: the sequence number for the SAP messages
  * @dbgfs_dir: the debugfs dir entry
@@ -179,6 +186,7 @@ struct iwl_mei {
 	bool pldr_active;
 	spinlock_t data_q_lock;
 	struct work_struct netdev_work;
+	struct delayed_work ownership_dwork;
 
 	atomic_t sap_seq_no;
 	atomic_t seq_no;
@@ -833,6 +841,8 @@ static void iwl_mei_handle_csme_taking_ownership(struct mei_cl_device *cldev,
 	} else {
 		iwl_mei_send_sap_msg(cldev,
 				     SAP_MSG_NOTIF_CSME_OWNERSHIP_CONFIRMED);
+		schedule_delayed_work(&mei->ownership_dwork,
+				      MEI_OWNERSHIP_RETAKE_TIMEOUT_MS);
 	}
 }
 
@@ -1447,7 +1457,13 @@ int iwl_mei_get_ownership(void)
 
 	ret = wait_event_timeout(mei->get_ownership_wq,
 				 mei->got_ownership, HZ / 2);
-	return (!ret) ? -ETIMEDOUT : 0;
+	if (!ret) {
+		schedule_delayed_work(&mei->ownership_dwork,
+				      MEI_OWNERSHIP_RETAKE_TIMEOUT_MS);
+		return -ETIMEDOUT;
+	}
+
+	return 0;
 out:
 	mutex_unlock(&iwl_mei_mutex);
 	return ret;
@@ -1738,6 +1754,8 @@ void iwl_mei_device_state(bool up)
 	iwl_mei_send_sap_msg(mei->cldev,
 			     SAP_MSG_NOTIF_CSME_OWNERSHIP_CONFIRMED);
 	mei->csme_taking_ownership = false;
+	schedule_delayed_work(&mei->ownership_dwork,
+			      MEI_OWNERSHIP_RETAKE_TIMEOUT_MS);
 out:
 	mutex_unlock(&iwl_mei_mutex);
 }
@@ -1894,6 +1912,11 @@ static void iwl_mei_dbgfs_unregister(struct iwl_mei *mei) {}
 
 #endif /* CONFIG_DEBUG_FS */
 
+static void iwl_mei_ownership_dwork(struct work_struct *wk)
+{
+	iwl_mei_get_ownership();
+}
+
 #define ALLOC_SHARED_MEM_RETRY_MAX_NUM	3
 
 /*
@@ -1923,6 +1946,7 @@ static int iwl_mei_probe(struct mei_cl_device *cldev,
 	init_waitqueue_head(&mei->pldr_wq);
 	spin_lock_init(&mei->data_q_lock);
 	INIT_WORK(&mei->netdev_work, iwl_mei_netdev_work);
+	INIT_DELAYED_WORK(&mei->ownership_dwork, iwl_mei_ownership_dwork);
 
 	mei_cldev_set_drvdata(cldev, mei);
 	mei->cldev = cldev;
@@ -2105,6 +2129,7 @@ static void iwl_mei_remove(struct mei_cl_device *cldev)
 	cancel_work_sync(&mei->send_csa_msg_wk);
 	cancel_delayed_work_sync(&mei->csa_throttle_end_wk);
 	cancel_work_sync(&mei->netdev_work);
+	cancel_delayed_work_sync(&mei->ownership_dwork);
 
 	/*
 	 * If someone waits for the ownership, let him know that we are going
-- 
cgit v1.2.3-70-g09d2


From 22b68fc6d693e7a2b1c0eb852463f4a72522fa08 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Tue, 18 Apr 2023 12:28:14 +0300
Subject: wifi: iwlwifi: mvm: fix RFKILL report when driver is going down

When CSME takes ownership, the driver sets RFKILL on, and this
triggers driver unload and sending the confirmation SAP message.
However, when IWL_MVM_MEI_REPORT_RFKILL is set, RFKILL was not
reported and as a result, the driver did not confirm the ownership
transition. Fix it.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Link: https://lore.kernel.org/r/20230418122405.29ac3cd3df73.I96b32bc274bfe1e3871e54d3fa29c7ac4f40446f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h |  4 ++--
 drivers/net/wireless/intel/iwlwifi/mei/main.c    | 13 +++++++------
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c     |  6 +++++-
 3 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h b/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
index ae66192feefe..655d95d3a068 100644
--- a/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
+++ b/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * Copyright (C) 2021 Intel Corporation
+ * Copyright (C) 2021 - 2022 Intel Corporation
  */
 
 #ifndef __iwl_mei_h__
@@ -301,7 +301,7 @@ struct iwl_mei_colloc_info {
 struct iwl_mei_ops {
 	void (*me_conn_status)(void *priv,
 			       const struct iwl_mei_conn_info *conn_info);
-	void (*rfkill)(void *priv, bool blocked);
+	void (*rfkill)(void *priv, bool blocked, bool csme_taking_ownership);
 	void (*roaming_forbidden)(void *priv, bool forbidden);
 	void (*sap_connected)(void *priv);
 	void (*nic_stolen)(void *priv);
diff --git a/drivers/net/wireless/intel/iwlwifi/mei/main.c b/drivers/net/wireless/intel/iwlwifi/mei/main.c
index f1e0fe806a00..0a29fb013005 100644
--- a/drivers/net/wireless/intel/iwlwifi/mei/main.c
+++ b/drivers/net/wireless/intel/iwlwifi/mei/main.c
@@ -724,7 +724,7 @@ iwl_mei_handle_conn_status(struct mei_cl_device *cldev,
 						     status->link_prot_state);
 	else
 		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv,
-					  status->link_prot_state);
+					  status->link_prot_state, false);
 }
 
 static void iwl_mei_set_init_conf(struct iwl_mei *mei)
@@ -796,7 +796,7 @@ static void iwl_mei_handle_amt_state(struct mei_cl_device *cldev,
 	if (mei->amt_enabled)
 		iwl_mei_set_init_conf(mei);
 	else if (iwl_mei_cache.ops)
-		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, false);
+		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, false, false);
 
 	schedule_work(&mei->netdev_work);
 
@@ -837,7 +837,7 @@ static void iwl_mei_handle_csme_taking_ownership(struct mei_cl_device *cldev,
 		 */
 		mei->csme_taking_ownership = true;
 
-		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, true);
+		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, true, true);
 	} else {
 		iwl_mei_send_sap_msg(cldev,
 				     SAP_MSG_NOTIF_CSME_OWNERSHIP_CONFIRMED);
@@ -892,7 +892,7 @@ static void iwl_mei_handle_rx_host_own_req(struct mei_cl_device *cldev,
 
 	/* We can now start the connection, unblock rfkill */
 	if (iwl_mei_cache.ops)
-		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, false);
+		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, false, false);
 }
 
 static void iwl_mei_handle_pldr_ack(struct mei_cl_device *cldev,
@@ -1791,7 +1791,8 @@ int iwl_mei_register(void *priv, const struct iwl_mei_ops *ops)
 		if (iwl_mei_is_connected()) {
 			if (mei->amt_enabled)
 				iwl_mei_send_sap_msg(mei->cldev,
-						     SAP_MSG_NOTIF_WIFIDR_UP);
+						     SAP_MSG_NOTIF_WIFIDR_UP,
+						     false);
 			ops->rfkill(priv, mei->link_prot_state);
 		}
 	}
@@ -2111,7 +2112,7 @@ static void iwl_mei_remove(struct mei_cl_device *cldev)
 	spin_unlock_bh(&mei->data_q_lock);
 
 	if (iwl_mei_cache.ops)
-		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, false);
+		iwl_mei_cache.ops->rfkill(iwl_mei_cache.priv, false, false);
 
 	/*
 	 * mei_cldev_disable will return only after all the MEI Rx is done.
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 26c3f00d6f7d..32625bfacaae 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1020,10 +1020,14 @@ static void iwl_mvm_me_conn_status(void *priv, const struct iwl_mei_conn_info *c
 		kfree_rcu(prev_conn_info, rcu_head);
 }
 
-static void iwl_mvm_mei_rfkill(void *priv, bool blocked)
+static void iwl_mvm_mei_rfkill(void *priv, bool blocked,
+			       bool csme_taking_ownership)
 {
 	struct iwl_mvm *mvm = priv;
 
+	if (blocked && !csme_taking_ownership)
+		return;
+
 	mvm->mei_rfkill_blocked = blocked;
 	if (!mvm->hw_registered)
 		return;
-- 
cgit v1.2.3-70-g09d2


From eb74bfcfa99415f1a65ce021b8ac2256b7ed50bc Mon Sep 17 00:00:00 2001
From: Yingsha Xu <ysxu@hust.edu.cn>
Date: Wed, 19 Apr 2023 18:45:47 +0800
Subject: wifi: mac80211: remove return value check of debugfs_create_dir()

Smatch complains that:
debugfs_hw_add() warn: 'statsd' is an error pointer or valid

Debugfs checks are generally not supposed to be checked for errors
and it is not necessary here.

Just delete the dead code.

Signed-off-by: Yingsha Xu <ysxu@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Link: https://lore.kernel.org/r/20230419104548.30124-1-ysxu@hust.edu.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/debugfs.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c
index dfb9f55e2685..207f772bd8ce 100644
--- a/net/mac80211/debugfs.c
+++ b/net/mac80211/debugfs.c
@@ -673,10 +673,6 @@ void debugfs_hw_add(struct ieee80211_local *local)
 
 	statsd = debugfs_create_dir("statistics", phyd);
 
-	/* if the dir failed, don't put all the other things into the root! */
-	if (!statsd)
-		return;
-
 #ifdef CONFIG_MAC80211_DEBUG_COUNTERS
 	DEBUGFS_STATS_ADD(dot11TransmittedFragmentCount);
 	DEBUGFS_STATS_ADD(dot11MulticastTransmittedFrameCount);
-- 
cgit v1.2.3-70-g09d2


From eaddda248483ff78c4d26f1bf420e5f2af436a74 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 14 Apr 2023 16:22:28 +0800
Subject: wifi: rtw89: mac: use regular int as return type of DLE buffer
 request

The function to request DLE (data link engine) buffer uses 'u16' as return
value that mixes error code, so change it to 'int' as regular error code.
Also, treat invalid register value (0xfff) as an error.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414082228.30766-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/debug.c | 13 ++++---------
 drivers/net/wireless/realtek/rtw89/mac.c   | 22 +++++++++++++---------
 drivers/net/wireless/realtek/rtw89/mac.h   |  2 +-
 drivers/net/wireless/realtek/rtw89/reg.h   |  1 +
 4 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c
index 0e0e1483c099..1e5b7a998716 100644
--- a/drivers/net/wireless/realtek/rtw89/debug.c
+++ b/drivers/net/wireless/realtek/rtw89/debug.c
@@ -3069,18 +3069,13 @@ static int rtw89_dbg_trigger_ctrl_error(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_cpuio_ctrl ctrl_para = {0};
 	u16 pkt_id;
+	int ret;
 
 	rtw89_leave_ps_mode(rtwdev);
 
-	pkt_id = rtw89_mac_dle_buf_req(rtwdev, 0x20, true);
-	switch (pkt_id) {
-	case 0xffff:
-		return -ETIMEDOUT;
-	case 0xfff:
-		return -ENOMEM;
-	default:
-		break;
-	}
+	ret = rtw89_mac_dle_buf_req(rtwdev, 0x20, true, &pkt_id);
+	if (ret)
+		return ret;
 
 	/* intentionally, enqueue two pkt, but has only one pkt id */
 	ctrl_para.cmd_type = CPUIO_OP_CMD_ENQ_TO_HEAD;
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 9b47d80b900b..b8019cfc11b2 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -2813,7 +2813,7 @@ int rtw89_mac_resume_sch_tx_v1(struct rtw89_dev *rtwdev, u8 mac_idx, u32 tx_en)
 }
 EXPORT_SYMBOL(rtw89_mac_resume_sch_tx_v1);
 
-u16 rtw89_mac_dle_buf_req(struct rtw89_dev *rtwdev, u16 buf_len, bool wd)
+int rtw89_mac_dle_buf_req(struct rtw89_dev *rtwdev, u16 buf_len, bool wd, u16 *pkt_id)
 {
 	u32 val, reg;
 	int ret;
@@ -2828,9 +2828,13 @@ u16 rtw89_mac_dle_buf_req(struct rtw89_dev *rtwdev, u16 buf_len, bool wd)
 	ret = read_poll_timeout(rtw89_read32, val, val & B_AX_WD_BUF_STAT_DONE,
 				1, 2000, false, rtwdev, reg);
 	if (ret)
-		return 0xffff;
+		return ret;
+
+	*pkt_id = FIELD_GET(B_AX_WD_BUF_STAT_PKTID_MASK, val);
+	if (*pkt_id == S_WD_BUF_STAT_PKTID_INVALID)
+		return -ENOENT;
 
-	return FIELD_GET(B_AX_WD_BUF_STAT_PKTID_MASK, val);
+	return 0;
 }
 
 int rtw89_mac_set_cpuio(struct rtw89_dev *rtwdev,
@@ -2907,10 +2911,10 @@ static int dle_quota_change(struct rtw89_dev *rtwdev, enum rtw89_qta_mode mode)
 
 	dle_quota_cfg(rtwdev, cfg, INVALID_QT_WCPU);
 
-	pkt_id = rtw89_mac_dle_buf_req(rtwdev, 0x20, true);
-	if (pkt_id == 0xffff) {
+	ret = rtw89_mac_dle_buf_req(rtwdev, 0x20, true, &pkt_id);
+	if (ret) {
 		rtw89_err(rtwdev, "[ERR]WDE DLE buf req\n");
-		return -ENOMEM;
+		return ret;
 	}
 
 	ctrl_para.cmd_type = CPUIO_OP_CMD_ENQ_TO_HEAD;
@@ -2925,10 +2929,10 @@ static int dle_quota_change(struct rtw89_dev *rtwdev, enum rtw89_qta_mode mode)
 		return -EFAULT;
 	}
 
-	pkt_id = rtw89_mac_dle_buf_req(rtwdev, 0x20, false);
-	if (pkt_id == 0xffff) {
+	ret = rtw89_mac_dle_buf_req(rtwdev, 0x20, false, &pkt_id);
+	if (ret) {
 		rtw89_err(rtwdev, "[ERR]PLE DLE buf req\n");
-		return -ENOMEM;
+		return ret;
 	}
 
 	ctrl_para.cmd_type = CPUIO_OP_CMD_ENQ_TO_HEAD;
diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h
index 06cbad4b8d62..a8d9847ef0b4 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.h
+++ b/drivers/net/wireless/realtek/rtw89/mac.h
@@ -1149,7 +1149,7 @@ enum rtw89_mac_xtal_si_offset {
 int rtw89_mac_write_xtal_si(struct rtw89_dev *rtwdev, u8 offset, u8 val, u8 mask);
 int rtw89_mac_read_xtal_si(struct rtw89_dev *rtwdev, u8 offset, u8 *val);
 void rtw89_mac_pkt_drop_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif);
-u16 rtw89_mac_dle_buf_req(struct rtw89_dev *rtwdev, u16 buf_len, bool wd);
+int rtw89_mac_dle_buf_req(struct rtw89_dev *rtwdev, u16 buf_len, bool wd, u16 *pkt_id);
 int rtw89_mac_set_cpuio(struct rtw89_dev *rtwdev,
 			struct rtw89_cpuio_ctrl *ctrl_para, bool wd);
 int rtw89_mac_typ_fltr_opt(struct rtw89_dev *rtwdev,
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 7e466ebaec2c..266e4231b5f3 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -1588,6 +1588,7 @@
 #define R_AX_PL_BUF_STATUS 0x9824
 #define B_AX_WD_BUF_STAT_DONE BIT(31)
 #define B_AX_WD_BUF_STAT_PKTID_MASK GENMASK(11, 0)
+#define S_WD_BUF_STAT_PKTID_INVALID GENMASK(11, 0)
 
 #define R_AX_WD_CPUQ_OP_0 0x9810
 #define R_AX_PL_CPUQ_OP_0 0x9830
-- 
cgit v1.2.3-70-g09d2


From f0e741e4ddbc01610ca87167a123702b3fdac51f Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:11:28 +0800
Subject: wifi: rtw88: add bitmap for dynamic port settings

In order to support multiple interfaces, multiple port settings will
be required. Current code always uses port 0 and should be changed.
Declare a bitmap with size equal to hardware port number to record
the current usage.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121135.17828-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/mac80211.c | 13 ++++++++++---
 drivers/net/wireless/realtek/rtw88/main.c     |  1 +
 drivers/net/wireless/realtek/rtw88/main.h     | 10 ++++++++++
 3 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 3b92ac611d3f..d026094a72c4 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -155,25 +155,30 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
 	struct rtw_vif *rtwvif = (struct rtw_vif *)vif->drv_priv;
 	enum rtw_net_type net_type;
 	u32 config = 0;
-	u8 port = 0;
+	u8 port;
 	u8 bcn_ctrl = 0;
 
 	if (rtw_fw_feature_check(&rtwdev->fw, FW_FEATURE_BCN_FILTER))
 		vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER |
 				     IEEE80211_VIF_SUPPORTS_CQM_RSSI;
-	rtwvif->port = port;
 	rtwvif->stats.tx_unicast = 0;
 	rtwvif->stats.rx_unicast = 0;
 	rtwvif->stats.tx_cnt = 0;
 	rtwvif->stats.rx_cnt = 0;
 	rtwvif->scan_req = NULL;
 	memset(&rtwvif->bfee, 0, sizeof(struct rtw_bfee));
-	rtwvif->conf = &rtw_vif_port[port];
 	rtw_txq_init(rtwdev, vif->txq);
 	INIT_LIST_HEAD(&rtwvif->rsvd_page_list);
 
 	mutex_lock(&rtwdev->mutex);
 
+	port = find_first_zero_bit(rtwdev->hw_port, RTW_PORT_NUM);
+	if (port >= RTW_PORT_NUM)
+		return -EINVAL;
+	set_bit(port, rtwdev->hw_port);
+
+	rtwvif->port = port;
+	rtwvif->conf = &rtw_vif_port[port];
 	rtw_leave_lps_deep(rtwdev);
 
 	switch (vif->type) {
@@ -195,6 +200,7 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
 		break;
 	default:
 		WARN_ON(1);
+		clear_bit(rtwvif->port, rtwdev->hw_port);
 		mutex_unlock(&rtwdev->mutex);
 		return -EINVAL;
 	}
@@ -236,6 +242,7 @@ static void rtw_ops_remove_interface(struct ieee80211_hw *hw,
 	rtwvif->bcn_ctrl = 0;
 	config |= PORT_SET_BCN_CTRL;
 	rtw_vif_port_config(rtwdev, rtwvif, config);
+	clear_bit(rtwvif->port, rtwdev->hw_port);
 
 	mutex_unlock(&rtwdev->mutex);
 }
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 1cb553485cff..9fe7e22e31ce 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -610,6 +610,7 @@ free:
 	rcu_read_unlock();
 	rtw_iterate_stas_atomic(rtwdev, rtw_reset_sta_iter, rtwdev);
 	rtw_iterate_vifs_atomic(rtwdev, rtw_reset_vif_iter, rtwdev);
+	bitmap_zero(rtwdev->hw_port, RTW_PORT_NUM);
 	rtw_enter_ips(rtwdev);
 }
 
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index d4a53d556745..efac271497f5 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -395,6 +395,15 @@ enum rtw_snr {
 	RTW_SNR_NUM
 };
 
+enum rtw_port {
+	RTW_PORT_0 = 0,
+	RTW_PORT_1 = 1,
+	RTW_PORT_2 = 2,
+	RTW_PORT_3 = 3,
+	RTW_PORT_4 = 4,
+	RTW_PORT_NUM
+};
+
 enum rtw_wow_flags {
 	RTW_WOW_FLAG_EN_MAGIC_PKT,
 	RTW_WOW_FLAG_EN_REKEY_PKT,
@@ -2036,6 +2045,7 @@ struct rtw_dev {
 	u8 sta_cnt;
 	u32 rts_threshold;
 
+	DECLARE_BITMAP(hw_port, RTW_PORT_NUM);
 	DECLARE_BITMAP(mac_id_map, RTW_MAX_MAC_ID_NUM);
 	DECLARE_BITMAP(flags, NUM_OF_RTW_FLAGS);
 
-- 
cgit v1.2.3-70-g09d2


From ccf73f6e69c0244a979e97eb6c38f80cd6cbc116 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:11:29 +0800
Subject: wifi: rtw88: add port switch for AP mode

Switch port settings if AP mode does not start on port 0 because of
hardware limitation. For some ICs, beacons on ports other than zero
could misbehave and do not issue properly, to fix this we change AP
VIFs to port zero when multiple interfaces is active.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121135.17828-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/mac80211.c |  1 +
 drivers/net/wireless/realtek/rtw88/main.c     | 79 +++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw88/main.h     |  1 +
 3 files changed, 81 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index d026094a72c4..19c4d7c29759 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -212,6 +212,7 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw,
 	rtwvif->bcn_ctrl = bcn_ctrl;
 	config |= PORT_SET_BCN_CTRL;
 	rtw_vif_port_config(rtwdev, rtwvif, config);
+	rtw_core_port_switch(rtwdev, vif);
 
 	mutex_unlock(&rtwdev->mutex);
 
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 9fe7e22e31ce..abebccd0581a 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -2251,6 +2251,85 @@ void rtw_unregister_hw(struct rtw_dev *rtwdev, struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL(rtw_unregister_hw);
 
+static
+void rtw_swap_reg_nbytes(struct rtw_dev *rtwdev, const struct rtw_hw_reg *reg1,
+			 const struct rtw_hw_reg *reg2, u8 nbytes)
+{
+	u8 i;
+
+	for (i = 0; i < nbytes; i++) {
+		u8 v1 = rtw_read8(rtwdev, reg1->addr + i);
+		u8 v2 = rtw_read8(rtwdev, reg2->addr + i);
+
+		rtw_write8(rtwdev, reg1->addr + i, v2);
+		rtw_write8(rtwdev, reg2->addr + i, v1);
+	}
+}
+
+static
+void rtw_swap_reg_mask(struct rtw_dev *rtwdev, const struct rtw_hw_reg *reg1,
+		       const struct rtw_hw_reg *reg2)
+{
+	u32 v1, v2;
+
+	v1 = rtw_read32_mask(rtwdev, reg1->addr, reg1->mask);
+	v2 = rtw_read32_mask(rtwdev, reg2->addr, reg2->mask);
+	rtw_write32_mask(rtwdev, reg2->addr, reg2->mask, v1);
+	rtw_write32_mask(rtwdev, reg1->addr, reg1->mask, v2);
+}
+
+struct rtw_iter_port_switch_data {
+	struct rtw_dev *rtwdev;
+	struct rtw_vif *rtwvif_ap;
+};
+
+static void rtw_port_switch_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
+{
+	struct rtw_iter_port_switch_data *iter_data = data;
+	struct rtw_dev *rtwdev = iter_data->rtwdev;
+	struct rtw_vif *rtwvif_target = (struct rtw_vif *)vif->drv_priv;
+	struct rtw_vif *rtwvif_ap = iter_data->rtwvif_ap;
+	const struct rtw_hw_reg *reg1, *reg2;
+
+	if (rtwvif_target->port != RTW_PORT_0)
+		return;
+
+	rtw_dbg(rtwdev, RTW_DBG_STATE, "AP port switch from %d -> %d\n",
+		rtwvif_ap->port, rtwvif_target->port);
+
+	reg1 = &rtwvif_ap->conf->net_type;
+	reg2 = &rtwvif_target->conf->net_type;
+	rtw_swap_reg_mask(rtwdev, reg1, reg2);
+
+	reg1 = &rtwvif_ap->conf->mac_addr;
+	reg2 = &rtwvif_target->conf->mac_addr;
+	rtw_swap_reg_nbytes(rtwdev, reg1, reg2, ETH_ALEN);
+
+	reg1 = &rtwvif_ap->conf->bssid;
+	reg2 = &rtwvif_target->conf->bssid;
+	rtw_swap_reg_nbytes(rtwdev, reg1, reg2, ETH_ALEN);
+
+	reg1 = &rtwvif_ap->conf->bcn_ctrl;
+	reg2 = &rtwvif_target->conf->bcn_ctrl;
+	rtw_swap_reg_nbytes(rtwdev, reg1, reg2, 1);
+
+	swap(rtwvif_target->port, rtwvif_ap->port);
+	swap(rtwvif_target->conf, rtwvif_ap->conf);
+}
+
+void rtw_core_port_switch(struct rtw_dev *rtwdev, struct ieee80211_vif *vif)
+{
+	struct rtw_vif *rtwvif = (struct rtw_vif *)vif->drv_priv;
+	struct rtw_iter_port_switch_data iter_data;
+
+	if (vif->type != NL80211_IFTYPE_AP || rtwvif->port == RTW_PORT_0)
+		return;
+
+	iter_data.rtwdev = rtwdev;
+	iter_data.rtwvif_ap = rtwvif;
+	rtw_iterate_vifs(rtwdev, rtw_port_switch_iter, &iter_data);
+}
+
 MODULE_AUTHOR("Realtek Corporation");
 MODULE_DESCRIPTION("Realtek 802.11ac wireless core module");
 MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index efac271497f5..790ebf781bc4 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -2198,4 +2198,5 @@ void rtw_set_txrx_1ss(struct rtw_dev *rtwdev, bool config_1ss);
 void rtw_update_channel(struct rtw_dev *rtwdev, u8 center_channel,
 			u8 primary_channel, enum rtw_supported_band band,
 			enum rtw_bandwidth bandwidth);
+void rtw_core_port_switch(struct rtw_dev *rtwdev, struct ieee80211_vif *vif);
 #endif
-- 
cgit v1.2.3-70-g09d2


From ffa71c5477793f41bc7537a60aa54ac40275ab78 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:11:30 +0800
Subject: wifi: rtw88: 8822c: extend reserved page number

Extend 8822c's reserved page number to accommodate additional required
pages. Reserved page is an area of memory in the FIFO dedicated for
special purposes. Previously only one interface is supported so 8 pages
should suffice, extend it so we can support 2 interfaces concurrently.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121135.17828-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/mac.c      | 2 +-
 drivers/net/wireless/realtek/rtw88/main.h     | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8723d.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8821c.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8822b.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8822c.c | 1 +
 6 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 44e07b61b9b9..a168f36c38ec 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -1119,7 +1119,7 @@ static int set_trx_fifo_info(struct rtw_dev *rtwdev)
 	u8 csi_buf_pg_num = chip->csi_buf_pg_num;
 
 	/* config rsvd page num */
-	fifo->rsvd_drv_pg_num = 8;
+	fifo->rsvd_drv_pg_num = chip->rsvd_drv_pg_num;
 	fifo->txff_pg_num = chip->txff_size >> 7;
 	if (rtw_chip_wcpu_11n(rtwdev))
 		fifo->rsvd_pg_num = fifo->rsvd_drv_pg_num;
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 790ebf781bc4..532c56219a5f 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -1177,6 +1177,7 @@ struct rtw_chip_info {
 	u32 txff_size;
 	u32 rxff_size;
 	u32 fw_rxff_size;
+	u16 rsvd_drv_pg_num;
 	u8 band;
 	u8 page_size;
 	u8 csi_buf_pg_num;
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8723d.c b/drivers/net/wireless/realtek/rtw88/rtw8723d.c
index 2d2f768bae2e..06e7454c9ca6 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8723d.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8723d.c
@@ -2743,6 +2743,7 @@ const struct rtw_chip_info rtw8723d_hw_spec = {
 	.ptct_efuse_size = 96 + 1,
 	.txff_size = 32768,
 	.rxff_size = 16384,
+	.rsvd_drv_pg_num = 8,
 	.txgi_factor = 1,
 	.is_pwr_by_rate_dec = true,
 	.max_power_index = 0x3f,
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.c b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
index 7ae0541d7b99..e53836768378 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
@@ -1920,6 +1920,7 @@ const struct rtw_chip_info rtw8821c_hw_spec = {
 	.ptct_efuse_size = 96,
 	.txff_size = 65536,
 	.rxff_size = 16384,
+	.rsvd_drv_pg_num = 8,
 	.txgi_factor = 1,
 	.is_pwr_by_rate_dec = true,
 	.max_power_index = 0x3f,
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822b.c b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
index 531b67787e2e..3017a9760da8 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822b.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
@@ -2540,6 +2540,7 @@ const struct rtw_chip_info rtw8822b_hw_spec = {
 	.txff_size = 262144,
 	.rxff_size = 24576,
 	.fw_rxff_size = 12288,
+	.rsvd_drv_pg_num = 8,
 	.txgi_factor = 1,
 	.is_pwr_by_rate_dec = true,
 	.max_power_index = 0x3f,
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index 5a2c004b12df..cd965edc29ce 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -5358,6 +5358,7 @@ const struct rtw_chip_info rtw8822c_hw_spec = {
 	.txff_size = 262144,
 	.rxff_size = 24576,
 	.fw_rxff_size = 12288,
+	.rsvd_drv_pg_num = 16,
 	.txgi_factor = 2,
 	.is_pwr_by_rate_dec = false,
 	.max_power_index = 0x7f,
-- 
cgit v1.2.3-70-g09d2


From 5ec69129f195f340acb15b8535cb372ccdbcf5d7 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:11:31 +0800
Subject: wifi: rtw88: disallow PS during AP mode

Firmware can't support PS mode during AP mode, so disallow this case.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121135.17828-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/mac80211.c | 13 +++++++++++++
 drivers/net/wireless/realtek/rtw88/main.c     |  2 +-
 drivers/net/wireless/realtek/rtw88/main.h     |  1 +
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 19c4d7c29759..1ced57b09083 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -449,12 +449,24 @@ static int rtw_ops_start_ap(struct ieee80211_hw *hw,
 	const struct rtw_chip_info *chip = rtwdev->chip;
 
 	mutex_lock(&rtwdev->mutex);
+	rtwdev->ap_active = true;
 	chip->ops->phy_calibration(rtwdev);
 	mutex_unlock(&rtwdev->mutex);
 
 	return 0;
 }
 
+static void rtw_ops_stop_ap(struct ieee80211_hw *hw,
+			    struct ieee80211_vif *vif,
+			    struct ieee80211_bss_conf *link_conf)
+{
+	struct rtw_dev *rtwdev = hw->priv;
+
+	mutex_lock(&rtwdev->mutex);
+	rtwdev->ap_active = false;
+	mutex_unlock(&rtwdev->mutex);
+}
+
 static int rtw_ops_conf_tx(struct ieee80211_hw *hw,
 			   struct ieee80211_vif *vif,
 			   unsigned int link_id, u16 ac,
@@ -916,6 +928,7 @@ const struct ieee80211_ops rtw_ops = {
 	.configure_filter	= rtw_ops_configure_filter,
 	.bss_info_changed	= rtw_ops_bss_info_changed,
 	.start_ap		= rtw_ops_start_ap,
+	.stop_ap		= rtw_ops_stop_ap,
 	.conf_tx		= rtw_ops_conf_tx,
 	.sta_add		= rtw_ops_sta_add,
 	.sta_remove		= rtw_ops_sta_remove,
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index abebccd0581a..4537513d0668 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -257,7 +257,7 @@ static void rtw_watch_dog_work(struct work_struct *work)
 	 * threshold.
 	 */
 	if (rtwdev->ps_enabled && data.rtwvif && !ps_active &&
-	    !rtwdev->beacon_loss)
+	    !rtwdev->beacon_loss && !rtwdev->ap_active)
 		rtw_enter_lps(rtwdev, data.rtwvif->port);
 
 	rtwdev->watch_dog_cnt++;
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 532c56219a5f..b04ed190ea5d 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -2058,6 +2058,7 @@ struct rtw_dev {
 
 	bool need_rfk;
 	struct completion fw_scan_density;
+	bool ap_active;
 
 	/* hci related data, must be last */
 	u8 priv[] __aligned(sizeof(void *));
-- 
cgit v1.2.3-70-g09d2


From 982f4a2004f712ce7aed7f7e69fa64e4c06eee7d Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:13:00 +0800
Subject: wifi: rtw88: refine reserved page flow for AP mode

Only gather reserved pages from AP interface after it has started. Or
else ieee80211_beacon_get_*() returns NULL and causes other VIFs'
reserved pages fail to download. Update location of current reserved page
after beacon renews so offsets changed by beacon can be recognized.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121300.17900-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/fw.c       | 5 +++++
 drivers/net/wireless/realtek/rtw88/mac80211.c | 1 +
 2 files changed, 6 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index 82295ac6402e..049473accdb9 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -1393,6 +1393,10 @@ static void rtw_build_rsvd_page_iter(void *data, u8 *mac,
 	struct rtw_vif *rtwvif = (struct rtw_vif *)vif->drv_priv;
 	struct rtw_rsvd_page *rsvd_pkt;
 
+	/* AP not yet started, don't gather its rsvd pages */
+	if (vif->type == NL80211_IFTYPE_AP && !rtwdev->ap_active)
+		return;
+
 	list_for_each_entry(rsvd_pkt, &rtwvif->rsvd_page_list, vif_list) {
 		if (rsvd_pkt->type == RSVD_BEACON)
 			list_add(&rsvd_pkt->build_list,
@@ -1614,6 +1618,7 @@ void rtw_fw_update_beacon_work(struct work_struct *work)
 
 	mutex_lock(&rtwdev->mutex);
 	rtw_fw_download_rsvd_page(rtwdev);
+	rtw_send_rsvd_page_h2c(rtwdev);
 	mutex_unlock(&rtwdev->mutex);
 }
 
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 1ced57b09083..dbd40a26908a 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -417,6 +417,7 @@ static void rtw_ops_bss_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_BEACON) {
 		rtw_set_dtim_period(rtwdev, conf->dtim_period);
 		rtw_fw_download_rsvd_page(rtwdev);
+		rtw_send_rsvd_page_h2c(rtwdev);
 	}
 
 	if (changed & BSS_CHANGED_BEACON_ENABLED) {
-- 
cgit v1.2.3-70-g09d2


From a1b8015da57a783b3ceebd8e114afbe07536bd54 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:13:12 +0800
Subject: wifi: rtw88: prevent scan abort with other VIFs

Only abort scan with current scanning VIF. If we have more than one
interface, we could call rtw_hw_scan_abort() with the wrong VIF as
input. This avoids potential null pointer being accessed when actually
the other VIF is scanning.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121312.17954-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/fw.c       | 4 +++-
 drivers/net/wireless/realtek/rtw88/fw.h       | 2 +-
 drivers/net/wireless/realtek/rtw88/mac80211.c | 7 ++++---
 3 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index 049473accdb9..f2d48091b1e9 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -2163,8 +2163,10 @@ out:
 	return ret;
 }
 
-void rtw_hw_scan_abort(struct rtw_dev *rtwdev, struct ieee80211_vif *vif)
+void rtw_hw_scan_abort(struct rtw_dev *rtwdev)
 {
+	struct ieee80211_vif *vif = rtwdev->scan_info.scanning_vif;
+
 	if (!rtw_fw_feature_check(&rtwdev->fw, FW_FEATURE_SCAN_OFFLOAD))
 		return;
 
diff --git a/drivers/net/wireless/realtek/rtw88/fw.h b/drivers/net/wireless/realtek/rtw88/fw.h
index 0a386e6d6e0d..397cbc3f6af6 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.h
+++ b/drivers/net/wireless/realtek/rtw88/fw.h
@@ -868,5 +868,5 @@ int rtw_hw_scan_offload(struct rtw_dev *rtwdev, struct ieee80211_vif *vif,
 			bool enable);
 void rtw_hw_scan_status_report(struct rtw_dev *rtwdev, struct sk_buff *skb);
 void rtw_hw_scan_chan_switch(struct rtw_dev *rtwdev, struct sk_buff *skb);
-void rtw_hw_scan_abort(struct rtw_dev *rtwdev, struct ieee80211_vif *vif);
+void rtw_hw_scan_abort(struct rtw_dev *rtwdev);
 #endif
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index dbd40a26908a..b3e4c699d425 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -393,7 +393,8 @@ static void rtw_ops_bss_info_changed(struct ieee80211_hw *hw,
 			 * when disconnected by peer
 			 */
 			if (test_bit(RTW_FLAG_SCANNING, rtwdev->flags))
-				rtw_hw_scan_abort(rtwdev, vif);
+				rtw_hw_scan_abort(rtwdev);
+
 		}
 
 		config |= PORT_SET_NET_TYPE;
@@ -870,7 +871,7 @@ static int rtw_ops_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	rtw_hw_scan_start(rtwdev, vif, req);
 	ret = rtw_hw_scan_offload(rtwdev, vif, true);
 	if (ret) {
-		rtw_hw_scan_abort(rtwdev, vif);
+		rtw_hw_scan_abort(rtwdev);
 		rtw_err(rtwdev, "HW scan failed with status: %d\n", ret);
 	}
 	mutex_unlock(&rtwdev->mutex);
@@ -890,7 +891,7 @@ static void rtw_ops_cancel_hw_scan(struct ieee80211_hw *hw,
 		return;
 
 	mutex_lock(&rtwdev->mutex);
-	rtw_hw_scan_abort(rtwdev, vif);
+	rtw_hw_scan_abort(rtwdev);
 	mutex_unlock(&rtwdev->mutex);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 96fbb84de4ffce76d54de8656efe0a580081c037 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:13:23 +0800
Subject: wifi: rtw88: handle station mode concurrent scan with AP mode

This patch allows vifs sharing same hardware with the AP mode vif to
do scan, do note that this could lead to packet loss or disconnection
of the AP's clients. Since we don't have chanctx, update scan info
upon set channel so bandwidth changes won't go unnoticed and get
misconfigured after scan. Download beacon just before scan starts to
allow hardware to get proper content to do beaconing. Last, beacons
should only be transmitted in AP's operating channel. Turn related
beacon functions off while we're in other channels so the receiving
stations won't get confused.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121323.18008-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/fw.c       | 11 +++++++-
 drivers/net/wireless/realtek/rtw88/mac80211.c |  5 +++-
 drivers/net/wireless/realtek/rtw88/main.c     | 39 +++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw88/main.h     |  2 ++
 4 files changed, 55 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/fw.c b/drivers/net/wireless/realtek/rtw88/fw.c
index f2d48091b1e9..2a8ccc8a7f60 100644
--- a/drivers/net/wireless/realtek/rtw88/fw.c
+++ b/drivers/net/wireless/realtek/rtw88/fw.c
@@ -2160,6 +2160,12 @@ int rtw_hw_scan_offload(struct rtw_dev *rtwdev, struct ieee80211_vif *vif,
 	}
 	rtw_fw_set_scan_offload(rtwdev, &cs_option, rtwvif, &chan_list);
 out:
+	if (rtwdev->ap_active) {
+		ret = rtw_download_beacon(rtwdev);
+		if (ret)
+			rtw_err(rtwdev, "HW scan download beacon failed\n");
+	}
+
 	return ret;
 }
 
@@ -2251,6 +2257,7 @@ void rtw_hw_scan_chan_switch(struct rtw_dev *rtwdev, struct sk_buff *skb)
 		if (rtw_is_op_chan(rtwdev, chan)) {
 			rtw_store_op_chan(rtwdev, false);
 			ieee80211_wake_queues(rtwdev->hw);
+			rtw_core_enable_beacon(rtwdev, true);
 		}
 	} else if (id == RTW_SCAN_NOTIFY_ID_PRESWITCH) {
 		if (IS_CH_5G_BAND(chan)) {
@@ -2269,8 +2276,10 @@ void rtw_hw_scan_chan_switch(struct rtw_dev *rtwdev, struct sk_buff *skb)
 		 * if next channel is non-op channel.
 		 */
 		if (!rtw_is_op_chan(rtwdev, chan) &&
-		    rtw_is_op_chan(rtwdev, hal->current_channel))
+		    rtw_is_op_chan(rtwdev, hal->current_channel)) {
+			rtw_core_enable_beacon(rtwdev, false);
 			ieee80211_stop_queues(rtwdev->hw);
+		}
 	}
 
 	rtw_dbg(rtwdev, RTW_DBG_HW_SCAN,
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index b3e4c699d425..7aa6edad0d01 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -404,7 +404,7 @@ static void rtw_ops_bss_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_BSSID) {
 		ether_addr_copy(rtwvif->bssid, conf->bssid);
 		config |= PORT_SET_BSSID;
-		if (is_zero_ether_addr(rtwvif->bssid))
+		if (!rtw_core_check_sta_active(rtwdev))
 			rtw_clear_op_chan(rtwdev);
 		else
 			rtw_store_op_chan(rtwdev, true);
@@ -452,6 +452,7 @@ static int rtw_ops_start_ap(struct ieee80211_hw *hw,
 
 	mutex_lock(&rtwdev->mutex);
 	rtwdev->ap_active = true;
+	rtw_store_op_chan(rtwdev, true);
 	chip->ops->phy_calibration(rtwdev);
 	mutex_unlock(&rtwdev->mutex);
 
@@ -466,6 +467,8 @@ static void rtw_ops_stop_ap(struct ieee80211_hw *hw,
 
 	mutex_lock(&rtwdev->mutex);
 	rtwdev->ap_active = false;
+	if (!rtw_core_check_sta_active(rtwdev))
+		rtw_clear_op_chan(rtwdev);
 	mutex_unlock(&rtwdev->mutex);
 }
 
diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 4537513d0668..c3a91bef13e4 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -830,6 +830,9 @@ void rtw_set_channel(struct rtw_dev *rtwdev)
 
 	rtw_update_channel(rtwdev, center_chan, primary_chan, band, bandwidth);
 
+	if (rtwdev->scan_info.op_chan)
+		rtw_store_op_chan(rtwdev, true);
+
 	chip->ops->set_channel(rtwdev, center_chan, bandwidth,
 			       hal->current_primary_channel_index);
 
@@ -2330,6 +2333,42 @@ void rtw_core_port_switch(struct rtw_dev *rtwdev, struct ieee80211_vif *vif)
 	rtw_iterate_vifs(rtwdev, rtw_port_switch_iter, &iter_data);
 }
 
+static void rtw_check_sta_active_iter(void *data, u8 *mac,
+				      struct ieee80211_vif *vif)
+{
+	struct rtw_vif *rtwvif = (struct rtw_vif *)vif->drv_priv;
+	bool *active = data;
+
+	if (*active)
+		return;
+
+	if (vif->type != NL80211_IFTYPE_STATION)
+		return;
+
+	if (vif->cfg.assoc || !is_zero_ether_addr(rtwvif->bssid))
+		*active = true;
+}
+
+bool rtw_core_check_sta_active(struct rtw_dev *rtwdev)
+{
+	bool sta_active = false;
+
+	rtw_iterate_vifs(rtwdev, rtw_check_sta_active_iter, &sta_active);
+
+	return rtwdev->ap_active || sta_active;
+}
+
+void rtw_core_enable_beacon(struct rtw_dev *rtwdev, bool enable)
+{
+	if (!rtwdev->ap_active)
+		return;
+
+	if (enable)
+		rtw_write32_set(rtwdev, REG_BCN_CTRL, BIT_EN_BCN_FUNCTION);
+	else
+		rtw_write32_clr(rtwdev, REG_BCN_CTRL, BIT_EN_BCN_FUNCTION);
+}
+
 MODULE_AUTHOR("Realtek Corporation");
 MODULE_DESCRIPTION("Realtek 802.11ac wireless core module");
 MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index b04ed190ea5d..621355f84d02 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -2201,4 +2201,6 @@ void rtw_update_channel(struct rtw_dev *rtwdev, u8 center_channel,
 			u8 primary_channel, enum rtw_supported_band band,
 			enum rtw_bandwidth bandwidth);
 void rtw_core_port_switch(struct rtw_dev *rtwdev, struct ieee80211_vif *vif);
+bool rtw_core_check_sta_active(struct rtw_dev *rtwdev);
+void rtw_core_enable_beacon(struct rtw_dev *rtwdev, bool enable);
 #endif
-- 
cgit v1.2.3-70-g09d2


From d16836cdcc3b17734f6aae165268d2f1777a4f74 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 14 Apr 2023 20:13:31 +0800
Subject: wifi: rtw88: 8822c: add iface combination

Allow 8822c to operate two interface concurrently, only 1 AP mode plus
1 station mode under same frequency is supported. Combination of other
types will not be added until requested.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230414121331.18062-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw88/main.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index c3a91bef13e4..0ba81b87aad5 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -103,6 +103,26 @@ static struct ieee80211_rate rtw_ratetable[] = {
 	{.bitrate = 540, .hw_value = 0x0b,},
 };
 
+static const struct ieee80211_iface_limit rtw_iface_limits[] = {
+	{
+		.max = 1,
+		.types = BIT(NL80211_IFTYPE_STATION),
+	},
+	{
+		.max = 1,
+		.types = BIT(NL80211_IFTYPE_AP),
+	}
+};
+
+static const struct ieee80211_iface_combination rtw_iface_combs[] = {
+	{
+		.limits = rtw_iface_limits,
+		.n_limits = ARRAY_SIZE(rtw_iface_limits),
+		.max_interfaces = 2,
+		.num_different_channels = 1,
+	}
+};
+
 u16 rtw_desc_to_bitrate(u8 desc_rate)
 {
 	struct ieee80211_rate rate;
@@ -2205,6 +2225,11 @@ int rtw_register_hw(struct rtw_dev *rtwdev, struct ieee80211_hw *hw)
 	hw->wiphy->max_scan_ssids = RTW_SCAN_MAX_SSIDS;
 	hw->wiphy->max_scan_ie_len = rtw_get_max_scan_ie_len(rtwdev);
 
+	if (rtwdev->chip->id == RTW_CHIP_TYPE_8822C) {
+		hw->wiphy->iface_combinations = rtw_iface_combs;
+		hw->wiphy->n_iface_combinations = ARRAY_SIZE(rtw_iface_combs);
+	}
+
 	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CAN_REPLACE_PTK0);
 	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_SCAN_RANDOM_SN);
 	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_SET_SCAN_DWELL);
-- 
cgit v1.2.3-70-g09d2


From a6f187f92bcc2b17821538b4a11d61764e68b091 Mon Sep 17 00:00:00 2001
From: Sascha Hauer <s.hauer@pengutronix.de>
Date: Mon, 17 Apr 2023 16:03:55 +0200
Subject: wifi: rtw88: usb: fix priority queue to endpoint mapping

The RTW88 chipsets have four different priority queues in hardware. For
the USB type chipsets the packets destined for a specific priority queue
must be sent through the endpoint corresponding to the queue. This was
not fully understood when porting from the RTW88 USB out of tree driver
and thus violated.

This patch implements the qsel to endpoint mapping as in
get_usb_bulkout_id_88xx() in the downstream driver.

Without this the driver often issues "timed out to flush queue 3"
warnings and often TX stalls completely.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Tested-by: ValdikSS <iam@valdikss.org.ru>
Tested-by: Alexandru gagniuc <mr.nuke.me@gmail.com>
Tested-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: stable@vger.kernel.org
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230417140358.2240429-2-s.hauer@pengutronix.de
---
 drivers/net/wireless/realtek/rtw88/usb.c | 70 +++++++++++++++++++++-----------
 1 file changed, 47 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index 68e1b782d199..8e2c99f9c366 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -118,6 +118,22 @@ static void rtw_usb_write32(struct rtw_dev *rtwdev, u32 addr, u32 val)
 	rtw_usb_write(rtwdev, addr, val, 4);
 }
 
+static int dma_mapping_to_ep(enum rtw_dma_mapping dma_mapping)
+{
+	switch (dma_mapping) {
+	case RTW_DMA_MAPPING_HIGH:
+		return 0;
+	case RTW_DMA_MAPPING_NORMAL:
+		return 1;
+	case RTW_DMA_MAPPING_LOW:
+		return 2;
+	case RTW_DMA_MAPPING_EXTRA:
+		return 3;
+	default:
+		return -EINVAL;
+	}
+}
+
 static int rtw_usb_parse(struct rtw_dev *rtwdev,
 			 struct usb_interface *interface)
 {
@@ -129,6 +145,8 @@ static int rtw_usb_parse(struct rtw_dev *rtwdev,
 	int num_out_pipes = 0;
 	int i;
 	u8 num;
+	const struct rtw_chip_info *chip = rtwdev->chip;
+	const struct rtw_rqpn *rqpn;
 
 	for (i = 0; i < interface_desc->bNumEndpoints; i++) {
 		endpoint = &host_interface->endpoint[i].desc;
@@ -183,31 +201,34 @@ static int rtw_usb_parse(struct rtw_dev *rtwdev,
 
 	rtwdev->hci.bulkout_num = num_out_pipes;
 
-	switch (num_out_pipes) {
-	case 4:
-	case 3:
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID0] = 2;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID1] = 2;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID2] = 2;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID3] = 2;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID4] = 1;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID5] = 1;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID6] = 0;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID7] = 0;
-		break;
-	case 2:
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID0] = 1;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID1] = 1;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID2] = 1;
-		rtwusb->qsel_to_ep[TX_DESC_QSEL_TID3] = 1;
-		break;
-	case 1:
-		break;
-	default:
-		rtw_err(rtwdev, "failed to get out_pipes(%d)\n", num_out_pipes);
+	if (num_out_pipes < 1 || num_out_pipes > 4) {
+		rtw_err(rtwdev, "invalid number of endpoints %d\n", num_out_pipes);
 		return -EINVAL;
 	}
 
+	rqpn = &chip->rqpn_table[num_out_pipes];
+
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID0] = dma_mapping_to_ep(rqpn->dma_map_be);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID1] = dma_mapping_to_ep(rqpn->dma_map_bk);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID2] = dma_mapping_to_ep(rqpn->dma_map_bk);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID3] = dma_mapping_to_ep(rqpn->dma_map_be);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID4] = dma_mapping_to_ep(rqpn->dma_map_vi);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID5] = dma_mapping_to_ep(rqpn->dma_map_vi);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID6] = dma_mapping_to_ep(rqpn->dma_map_vo);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID7] = dma_mapping_to_ep(rqpn->dma_map_vo);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID8] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID9] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID10] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID11] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID12] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID13] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID14] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_TID15] = -EINVAL;
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_BEACON] = dma_mapping_to_ep(rqpn->dma_map_hi);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_HIGH] = dma_mapping_to_ep(rqpn->dma_map_hi);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_MGMT] = dma_mapping_to_ep(rqpn->dma_map_mg);
+	rtwusb->qsel_to_ep[TX_DESC_QSEL_H2C] = dma_mapping_to_ep(rqpn->dma_map_hi);
+
 	return 0;
 }
 
@@ -250,7 +271,7 @@ static void rtw_usb_write_port_tx_complete(struct urb *urb)
 static int qsel_to_ep(struct rtw_usb *rtwusb, unsigned int qsel)
 {
 	if (qsel >= ARRAY_SIZE(rtwusb->qsel_to_ep))
-		return 0;
+		return -EINVAL;
 
 	return rtwusb->qsel_to_ep[qsel];
 }
@@ -265,6 +286,9 @@ static int rtw_usb_write_port(struct rtw_dev *rtwdev, u8 qsel, struct sk_buff *s
 	int ret;
 	int ep = qsel_to_ep(rtwusb, qsel);
 
+	if (ep < 0)
+		return ep;
+
 	pipe = usb_sndbulkpipe(usbd, rtwusb->out_ep[ep]);
 	urb = usb_alloc_urb(0, GFP_ATOMIC);
 	if (!urb)
-- 
cgit v1.2.3-70-g09d2


From 14705f969d98187a1cc2682e0c9bd2e230b8098f Mon Sep 17 00:00:00 2001
From: Sascha Hauer <s.hauer@pengutronix.de>
Date: Mon, 17 Apr 2023 16:03:56 +0200
Subject: wifi: rtw88: rtw8821c: Fix rfe_option field width

On my RTW8821CU chipset rfe_option reads as 0x22. Looking at the
vendor driver suggests that the field width of rfe_option is 5 bit,
so rfe_option should be masked with 0x1f.

Without this the rfe_option comparisons with 2 further down the
driver evaluate as false when they should really evaluate as true.
The effect is that 2G channels do not work.

rfe_option is also used as an array index into rtw8821c_rfe_defs[].
rtw8821c_rfe_defs[34] (0x22) was added as part of adding USB support,
likely because rfe_option reads as 0x22. As this now becomes 0x2,
rtw8821c_rfe_defs[34] is no longer used and can be removed.

Note that this might not be the whole truth. In the vendor driver
there are indeed places where the unmasked rfe_option value is used.
However, the driver has several places where rfe_option is tested
with the pattern if (rfe_option == 2 || rfe_option == 0x22) or
if (rfe_option == 4 || rfe_option == 0x24), so that rfe_option BIT(5)
has no influence on the code path taken. We therefore mask BIT(5)
out from rfe_option entirely until this assumption is proved wrong
by some chip variant we do not know yet.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Tested-by: Alexandru gagniuc <mr.nuke.me@gmail.com>
Tested-by: Larry Finger <Larry.Finger@lwfinger.net>
Tested-by: ValdikSS <iam@valdikss.org.ru>
Cc: stable@vger.kernel.org
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230417140358.2240429-3-s.hauer@pengutronix.de
---
 drivers/net/wireless/realtek/rtw88/rtw8821c.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.c b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
index e53836768378..619fb3adb832 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
@@ -53,7 +53,7 @@ static int rtw8821c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 
 	map = (struct rtw8821c_efuse *)log_map;
 
-	efuse->rfe_option = map->rfe_option;
+	efuse->rfe_option = map->rfe_option & 0x1f;
 	efuse->rf_board_option = map->rf_board_option;
 	efuse->crystal_cap = map->xtal_k;
 	efuse->pa_type_2g = map->pa_type;
@@ -1546,7 +1546,6 @@ static const struct rtw_rfe_def rtw8821c_rfe_defs[] = {
 	[2] = RTW_DEF_RFE_EXT(8821c, 0, 0, 2),
 	[4] = RTW_DEF_RFE_EXT(8821c, 0, 0, 2),
 	[6] = RTW_DEF_RFE(8821c, 0, 0),
-	[34] = RTW_DEF_RFE(8821c, 0, 0),
 };
 
 static struct rtw_hw_reg rtw8821c_dig[] = {
-- 
cgit v1.2.3-70-g09d2


From 97c75e1adeda78b3794936c617d8b86e9ebd54f5 Mon Sep 17 00:00:00 2001
From: Sascha Hauer <s.hauer@pengutronix.de>
Date: Mon, 17 Apr 2023 16:03:57 +0200
Subject: wifi: rtw88: set pkg_type correctly for specific rtw8821c variants

According to the vendor driver the pkg_type has to be set to '1'
for some rtw8821c variants. As the pkg_type has been hardcoded to
'0', add a field for it in struct rtw_hal and set this correctly
in the rtw8821c part.
With this parsing of a rtw_table is influenced and check_positive()
in phy.c returns true for some cases here. The same is done in the
vendor driver. However, this has no visible effect on the driver
here.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230417140358.2240429-4-s.hauer@pengutronix.de
---
 drivers/net/wireless/realtek/rtw88/main.c     | 2 +-
 drivers/net/wireless/realtek/rtw88/main.h     | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8821c.c | 3 +++
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c
index 0ba81b87aad5..5bf6b4581557 100644
--- a/drivers/net/wireless/realtek/rtw88/main.c
+++ b/drivers/net/wireless/realtek/rtw88/main.c
@@ -2008,7 +2008,7 @@ static int rtw_chip_board_info_setup(struct rtw_dev *rtwdev)
 	if (!rfe_def)
 		return -ENODEV;
 
-	rtw_phy_setup_phy_cond(rtwdev, 0);
+	rtw_phy_setup_phy_cond(rtwdev, hal->pkg_type);
 
 	rtw_phy_init_tx_power(rtwdev);
 	if (rfe_def->agc_btg_tbl)
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 621355f84d02..a290fb427e96 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -1900,6 +1900,7 @@ struct rtw_hal {
 	u8 cut_version;
 	u8 mp_chip;
 	u8 oem_id;
+	u8 pkg_type;
 	struct rtw_phy_cond phy_cond;
 
 	u8 ps_mode;
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.c b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
index 619fb3adb832..49d6c4ef68c5 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
@@ -47,6 +47,7 @@ enum rtw8821ce_rf_set {
 
 static int rtw8821c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 {
+	struct rtw_hal *hal = &rtwdev->hal;
 	struct rtw_efuse *efuse = &rtwdev->efuse;
 	struct rtw8821c_efuse *map;
 	int i;
@@ -70,6 +71,8 @@ static int rtw8821c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 	efuse->tx_bb_swing_setting_2g = map->tx_bb_swing_setting_2g;
 	efuse->tx_bb_swing_setting_5g = map->tx_bb_swing_setting_5g;
 
+	hal->pkg_type = map->rfe_option & BIT(5) ? 1 : 0;
+
 	for (i = 0; i < 4; i++)
 		efuse->txpwr_idx_table[i] = map->txpwr_idx_table[i];
 
-- 
cgit v1.2.3-70-g09d2


From 172591baa2cc1ec04869a0d798821a6ca9a2f4b4 Mon Sep 17 00:00:00 2001
From: Sascha Hauer <s.hauer@pengutronix.de>
Date: Mon, 17 Apr 2023 16:03:58 +0200
Subject: wifi: rtw88: call rtw8821c_switch_rf_set() according to chip variant

We have to call rtw8821c_switch_rf_set() with SWITCH_TO_WLG or
SWITCH_TO_BTG according to the chip variant as denoted in rfe_option.
The information which argument to use for which variant has been
taken from the vendor driver.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230417140358.2240429-5-s.hauer@pengutronix.de
---
 drivers/net/wireless/realtek/rtw88/main.h     |  1 +
 drivers/net/wireless/realtek/rtw88/rtw8821c.c | 19 +++++++++++++++----
 2 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index a290fb427e96..155525b8df35 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -1902,6 +1902,7 @@ struct rtw_hal {
 	u8 oem_id;
 	u8 pkg_type;
 	struct rtw_phy_cond phy_cond;
+	bool rfe_btg;
 
 	u8 ps_mode;
 	u8 current_channel;
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.c b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
index 49d6c4ef68c5..adf224618a2a 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
@@ -73,6 +73,17 @@ static int rtw8821c_read_efuse(struct rtw_dev *rtwdev, u8 *log_map)
 
 	hal->pkg_type = map->rfe_option & BIT(5) ? 1 : 0;
 
+	switch (efuse->rfe_option) {
+	case 0x2:
+	case 0x4:
+	case 0x7:
+	case 0xa:
+	case 0xc:
+	case 0xf:
+		hal->rfe_btg = true;
+		break;
+	}
+
 	for (i = 0; i < 4; i++)
 		efuse->txpwr_idx_table[i] = map->txpwr_idx_table[i];
 
@@ -298,6 +309,7 @@ static void rtw8821c_switch_rf_set(struct rtw_dev *rtwdev, u8 rf_set)
 
 static void rtw8821c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
 {
+	struct rtw_hal *hal = &rtwdev->hal;
 	u32 rf_reg18;
 
 	rf_reg18 = rtw_read_rf(rtwdev, RF_PATH_A, 0x18, RFREG_MASK);
@@ -329,11 +341,10 @@ static void rtw8821c_set_channel_rf(struct rtw_dev *rtwdev, u8 channel, u8 bw)
 	}
 
 	if (channel <= 14) {
-		if (rtwdev->efuse.rfe_option == 0)
-			rtw8821c_switch_rf_set(rtwdev, SWITCH_TO_WLG);
-		else if (rtwdev->efuse.rfe_option == 2 ||
-			 rtwdev->efuse.rfe_option == 4)
+		if (hal->rfe_btg)
 			rtw8821c_switch_rf_set(rtwdev, SWITCH_TO_BTG);
+		else
+			rtw8821c_switch_rf_set(rtwdev, SWITCH_TO_WLG);
 		rtw_write_rf(rtwdev, RF_PATH_A, RF_LUTDBG, BIT(6), 0x1);
 		rtw_write_rf(rtwdev, RF_PATH_A, 0x64, 0xf, 0xf);
 	} else {
-- 
cgit v1.2.3-70-g09d2


From 59a3a312009723e3e5082899655fdcc420e2b47a Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 17 Apr 2023 11:03:31 -0500
Subject: wifi: rtw88: Fix memory leak in rtw88_usb

Kmemleak shows the following leak arising from routine in the usb
probe routine:

unreferenced object 0xffff895cb29bba00 (size 512):
  comm "(udev-worker)", pid 534, jiffies 4294903932 (age 102751.088s)
  hex dump (first 32 bytes):
    77 30 30 30 00 00 00 00 02 2f 2d 2b 30 00 00 00  w000...../-+0...
    02 00 2a 28 00 00 00 00 ff 55 ff ff ff 00 00 00  ..*(.....U......
  backtrace:
    [<ffffffff9265fa36>] kmalloc_trace+0x26/0x90
    [<ffffffffc17eec41>] rtw_usb_probe+0x2f1/0x680 [rtw_usb]
    [<ffffffffc03e19fd>] usb_probe_interface+0xdd/0x2e0 [usbcore]
    [<ffffffff92b4f2fe>] really_probe+0x18e/0x3d0
    [<ffffffff92b4f5b8>] __driver_probe_device+0x78/0x160
    [<ffffffff92b4f6bf>] driver_probe_device+0x1f/0x90
    [<ffffffff92b4f8df>] __driver_attach+0xbf/0x1b0
    [<ffffffff92b4d350>] bus_for_each_dev+0x70/0xc0
    [<ffffffff92b4e51e>] bus_add_driver+0x10e/0x210
    [<ffffffff92b50935>] driver_register+0x55/0xf0
    [<ffffffffc03e0708>] usb_register_driver+0x88/0x140 [usbcore]
    [<ffffffff92401153>] do_one_initcall+0x43/0x210
    [<ffffffff9254f42a>] do_init_module+0x4a/0x200
    [<ffffffff92551d1c>] __do_sys_finit_module+0xac/0x120
    [<ffffffff92ee6626>] do_syscall_64+0x56/0x80
    [<ffffffff9300006a>] entry_SYSCALL_64_after_hwframe+0x46/0xb0

The leak was verified to be real by unloading the driver, which resulted
in a dangling pointer to the allocation.

The allocated memory is freed in rtw_usb_intf_deinit().

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Ping-Ke Shih <pkshih@realtek.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230417160331.23071-1-Larry.Finger@lwfinger.net
---
 drivers/net/wireless/realtek/rtw88/usb.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index 8e2c99f9c366..44a5fafb9905 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -804,6 +804,7 @@ static void rtw_usb_intf_deinit(struct rtw_dev *rtwdev,
 	struct rtw_usb *rtwusb = rtw_get_usb_priv(rtwdev);
 
 	usb_put_dev(rtwusb->udev);
+	kfree(rtwusb->usb_data);
 	usb_set_intfdata(intf, NULL);
 }
 
-- 
cgit v1.2.3-70-g09d2


From af8678e6c5bd37d99fa37c06a158a263da42efa1 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 17 Apr 2023 20:05:43 +0300
Subject: wifi: rtl8xxxu: Don't print the vendor/product/serial

Most devices have a vendor name, product name, and serial number in the
efuse, but it's pretty useless. It duplicates the information already
printed by the USB subsystem:

   usb 1-4: New USB device found, idVendor=0bda, idProduct=8178, bcdDevice= 2.00
   usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
   usb 1-4: Product: 802.11n WLAN Adapter
   usb 1-4: Manufacturer: Realtek
   usb 1-4: SerialNumber: 00e04c000001
-> usb 1-4: Vendor: Realtek
-> usb 1-4: Product: 802.11n WLAN Adapter

   usb 1-4: New USB device found, idVendor=0bda, idProduct=818b, bcdDevice= 2.00
   usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
   usb 1-4: Product: 802.11n NIC
   usb 1-4: Manufacturer: Realtek
   usb 1-4: SerialNumber: 00e04c000001
-> usb 1-4: Vendor: Realtek
-> usb 1-4: Product: 802.11n NIC
-> usb 1-4: Serial not available.

   usb 1-4: New USB device found, idVendor=0bda, idProduct=f179, bcdDevice= 0.00
   usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
   usb 1-4: Product: 802.11n
   usb 1-4: Manufacturer: Realtek
   usb 1-4: SerialNumber: 002E2DC0041F
-> usb 1-4: Vendor: Realtek
-> usb 1-4: Product: 802.11n

   usb 1-4: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00
   usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
   usb 1-4: Product: 802.11n NIC
   usb 1-4: Manufacturer: Realtek
   usb 1-4: SerialNumber: 00E04C0001
-> usb 1-4: Vendor: Realtek
-> usb 1-4: Product: 802.11n NIC
-> usb 1-4: Serial: 00E04C0001

Also, that data is not interpreted correctly in all cases:

usb 3-1.1.2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00
usb 3-1.1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 3-1.1.2: Product: 802.11n NIC
usb 3-1.1.2: Manufacturer: Realtek
usb 3-1.1.2: Vendor: Realtek
usb 3-1.1.2: Product: \x03802.11n NI
usb 3-1.1.2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff

Link: https://bugzilla.kernel.org/show_bug.cgi?id=217231
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/a2a7d9df-0529-7890-3522-48dce613753f@gmail.com
---
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c |  4 --
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c |  3 --
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c |  5 --
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c | 54 ----------------------
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c |  4 --
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c |  3 --
 6 files changed, 73 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
index af8436070ba7..8986783ae8fa 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c
@@ -568,10 +568,6 @@ static int rtl8188eu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	priv->default_crystal_cap = efuse->xtal_k & 0x3f;
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.11s\n", efuse->device_name);
-	dev_info(&priv->udev->dev, "Serial: %.11s\n", efuse->serial);
-
 	return 0;
 }
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
index dfb250adb168..dbdfd7787465 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188f.c
@@ -734,9 +734,6 @@ static int rtl8188fu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	priv->default_crystal_cap = efuse->xtal_k & 0x3f;
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.7s\n", efuse->device_name);
-
 	return 0;
 }
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
index caeba56241fc..b30a9a513cb8 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
@@ -441,11 +441,6 @@ static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv)
 	       efuse->ht20_max_power_offset,
 	       sizeof(efuse->ht20_max_power_offset));
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
-		 efuse->vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.20s\n",
-		 efuse->device_name);
-
 	priv->power_base = &rtl8192c_power_base;
 
 	if (efuse->rf_regulatory & 0x20) {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
index 4498748164af..fcc2926ea938 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c
@@ -601,43 +601,9 @@ rtl8192e_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 	}
 }
 
-static void rtl8192eu_log_next_device_info(struct rtl8xxxu_priv *priv,
-					   char *record_name,
-					   char *device_info,
-					   unsigned int *record_offset)
-{
-	char *record = device_info + *record_offset;
-
-	/* A record is [ total length | 0x03 | value ] */
-	unsigned char l = record[0];
-
-	/*
-	 * The whole device info section seems to be 80 characters, make sure
-	 * we don't read further.
-	 */
-	if (*record_offset + l > 80) {
-		dev_warn(&priv->udev->dev,
-			 "invalid record length %d while parsing \"%s\" at offset %u.\n",
-			 l, record_name, *record_offset);
-		return;
-	}
-
-	if (l >= 2) {
-		char value[80];
-
-		memcpy(value, &record[2], l - 2);
-		value[l - 2] = '\0';
-		dev_info(&priv->udev->dev, "%s: %s\n", record_name, value);
-		*record_offset = *record_offset + l;
-	} else {
-		dev_info(&priv->udev->dev, "%s not available.\n", record_name);
-	}
-}
-
 static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 {
 	struct rtl8192eu_efuse *efuse = &priv->efuse_wifi.efuse8192eu;
-	unsigned int record_offset;
 	int i;
 
 	if (efuse->rtl_id != cpu_to_le16(0x8129))
@@ -684,26 +650,6 @@ static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	priv->default_crystal_cap = priv->efuse_wifi.efuse8192eu.xtal_k & 0x3f;
 
-	/*
-	 * device_info section seems to be laid out as records
-	 * [ total length | 0x03 | value ] so:
-	 * - vendor length + 2
-	 * - 0x03
-	 * - vendor string (not null terminated)
-	 * - product length + 2
-	 * - 0x03
-	 * - product string (not null terminated)
-	 * Then there is one or 2 0x00 on all the 4 devices I own or found
-	 * dumped online.
-	 * As previous version of the code handled an optional serial
-	 * string, I now assume there may be a third record if the
-	 * length is not 0.
-	 */
-	record_offset = 0;
-	rtl8192eu_log_next_device_info(priv, "Vendor", efuse->device_info, &record_offset);
-	rtl8192eu_log_next_device_info(priv, "Product", efuse->device_info, &record_offset);
-	rtl8192eu_log_next_device_info(priv, "Serial", efuse->device_info, &record_offset);
-
 	return 0;
 }
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c
index d219be19d07f..15a30e496221 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c
@@ -222,10 +222,6 @@ static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	priv->power_base = &rtl8723a_power_base;
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
-		 efuse->vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.41s\n",
-		 efuse->device_name);
 	return 0;
 }
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
index c31c2b52ac77..abc56c7de6f7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c
@@ -494,9 +494,6 @@ static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	priv->default_crystal_cap = priv->efuse_wifi.efuse8723bu.xtal_k & 0x3f;
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.41s\n", efuse->device_name);
-
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From cd85c8b059c54b00e3b509e83fb36c2798f50128 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 17 Apr 2023 20:07:09 +0300
Subject: wifi: rtl8xxxu: Add rtl8xxxu_write{8,16,32}_{set,clear}

Also add rtl8xxxu_write32_mask, rtl8xxxu_write_rfreg_mask.

These helper functions make it easier to modify only parts of a register
by eliminating the call to the register reading function and the bit
manipulations.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/9430b841-1048-b27c-14ec-fca447dc32af@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h   | 12 ++++
 .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c  | 79 ++++++++++++++++++++++
 2 files changed, 91 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 82a0290ccb29..8eafbf1cee71 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1953,10 +1953,22 @@ u32 rtl8xxxu_read32(struct rtl8xxxu_priv *priv, u16 addr);
 int rtl8xxxu_write8(struct rtl8xxxu_priv *priv, u16 addr, u8 val);
 int rtl8xxxu_write16(struct rtl8xxxu_priv *priv, u16 addr, u16 val);
 int rtl8xxxu_write32(struct rtl8xxxu_priv *priv, u16 addr, u32 val);
+int rtl8xxxu_write8_set(struct rtl8xxxu_priv *priv, u16 addr, u8 bits);
+int rtl8xxxu_write8_clear(struct rtl8xxxu_priv *priv, u16 addr, u8 bits);
+int rtl8xxxu_write16_set(struct rtl8xxxu_priv *priv, u16 addr, u16 bits);
+int rtl8xxxu_write16_clear(struct rtl8xxxu_priv *priv, u16 addr, u16 bits);
+int rtl8xxxu_write32_set(struct rtl8xxxu_priv *priv, u16 addr, u32 bits);
+int rtl8xxxu_write32_clear(struct rtl8xxxu_priv *priv, u16 addr, u32 bits);
+int rtl8xxxu_write32_mask(struct rtl8xxxu_priv *priv, u16 addr,
+			  u32 mask, u32 val);
+
 u32 rtl8xxxu_read_rfreg(struct rtl8xxxu_priv *priv,
 			enum rtl8xxxu_rfpath path, u8 reg);
 int rtl8xxxu_write_rfreg(struct rtl8xxxu_priv *priv,
 			 enum rtl8xxxu_rfpath path, u8 reg, u32 data);
+int rtl8xxxu_write_rfreg_mask(struct rtl8xxxu_priv *priv,
+			      enum rtl8xxxu_rfpath path, u8 reg,
+			      u32 mask, u32 val);
 void rtl8xxxu_save_regs(struct rtl8xxxu_priv *priv, const u32 *regs,
 			u32 *backup, int count);
 void rtl8xxxu_restore_regs(struct rtl8xxxu_priv *priv, const u32 *regs,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 6106b47d0c37..03c0aaa9141c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -786,6 +786,85 @@ int rtl8xxxu_write32(struct rtl8xxxu_priv *priv, u16 addr, u32 val)
 	return ret;
 }
 
+int rtl8xxxu_write8_set(struct rtl8xxxu_priv *priv, u16 addr, u8 bits)
+{
+	u8 val8;
+
+	val8 = rtl8xxxu_read8(priv, addr);
+	val8 |= bits;
+	return rtl8xxxu_write8(priv, addr, val8);
+}
+
+int rtl8xxxu_write8_clear(struct rtl8xxxu_priv *priv, u16 addr, u8 bits)
+{
+	u8 val8;
+
+	val8 = rtl8xxxu_read8(priv, addr);
+	val8 &= ~bits;
+	return rtl8xxxu_write8(priv, addr, val8);
+}
+
+int rtl8xxxu_write16_set(struct rtl8xxxu_priv *priv, u16 addr, u16 bits)
+{
+	u16 val16;
+
+	val16 = rtl8xxxu_read16(priv, addr);
+	val16 |= bits;
+	return rtl8xxxu_write16(priv, addr, val16);
+}
+
+int rtl8xxxu_write16_clear(struct rtl8xxxu_priv *priv, u16 addr, u16 bits)
+{
+	u16 val16;
+
+	val16 = rtl8xxxu_read16(priv, addr);
+	val16 &= ~bits;
+	return rtl8xxxu_write16(priv, addr, val16);
+}
+
+int rtl8xxxu_write32_set(struct rtl8xxxu_priv *priv, u16 addr, u32 bits)
+{
+	u32 val32;
+
+	val32 = rtl8xxxu_read32(priv, addr);
+	val32 |= bits;
+	return rtl8xxxu_write32(priv, addr, val32);
+}
+
+int rtl8xxxu_write32_clear(struct rtl8xxxu_priv *priv, u16 addr, u32 bits)
+{
+	u32 val32;
+
+	val32 = rtl8xxxu_read32(priv, addr);
+	val32 &= ~bits;
+	return rtl8xxxu_write32(priv, addr, val32);
+}
+
+int rtl8xxxu_write32_mask(struct rtl8xxxu_priv *priv, u16 addr,
+			  u32 mask, u32 val)
+{
+	u32 orig, new, shift;
+
+	shift = __ffs(mask);
+
+	orig = rtl8xxxu_read32(priv, addr);
+	new = (orig & ~mask) | ((val << shift) & mask);
+	return rtl8xxxu_write32(priv, addr, new);
+}
+
+int rtl8xxxu_write_rfreg_mask(struct rtl8xxxu_priv *priv,
+			      enum rtl8xxxu_rfpath path, u8 reg,
+			      u32 mask, u32 val)
+{
+	u32 orig, new, shift;
+
+	shift = __ffs(mask);
+
+	orig = rtl8xxxu_read_rfreg(priv, path, reg);
+	new = (orig & ~mask) | ((val << shift) & mask);
+	return rtl8xxxu_write_rfreg(priv, path, reg, new);
+}
+
 static int
 rtl8xxxu_writeN(struct rtl8xxxu_priv *priv, u16 addr, u8 *buf, u16 len)
 {
-- 
cgit v1.2.3-70-g09d2


From c8bc3760277965c8eaa90155acea2b16215ebb8a Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 17 Apr 2023 20:08:20 +0300
Subject: wifi: rtl8xxxu: Simplify setting the initial gain

The goal of writing 0x6954341e / 0x6955341e to REG_OFDM0_XA_AGC_CORE1
appears to be setting the initial gain, which is stored in bits 0..6.
Bits 7..31 are the same as what the phy init tables write.

Modify only bits 0..6 so that we don't have to care about the values
of the others. This way we don't have to add another "else if" for the
RTL8192FU.

Why we need to change the initial gain from the default 0x20 to 0x1e?
Not sure. Some of the vendor drivers change it to 0x1e before scanning
and then restore it to the original value after.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/cf91ca69-70e3-4c20-c0b1-e59d452356a1@gmail.com
---
 drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 03c0aaa9141c..fd8c8c6d53d6 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -7034,10 +7034,8 @@ exit:
 	rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0xffff);
 	rtl8xxxu_write16(priv, REG_RXFLTMAP0, 0xffff);
 
-	if (priv->rtl_chip == RTL8188E)
-		rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, 0x6955341e);
-	else
-		rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, 0x6954341e);
+	rtl8xxxu_write32_mask(priv, REG_OFDM0_XA_AGC_CORE1,
+			      OFDM0_X_AGC_CORE1_IGI_MASK, 0x1e);
 
 	return ret;
 
-- 
cgit v1.2.3-70-g09d2


From 09be55585d2785d354b2375e9cf2acdc140d65cb Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 17 Apr 2023 22:51:24 +0200
Subject: wifi: airo: remove ISA_DMA_API dependency

This driver does not actually use the ISA DMA API, it is purely
PIO based, so remove the dependency.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230417205131.1560074-1-arnd@kernel.org
---
 drivers/net/wireless/cisco/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/cisco/Kconfig b/drivers/net/wireless/cisco/Kconfig
index 681bfc2d740a..b40ee25aca99 100644
--- a/drivers/net/wireless/cisco/Kconfig
+++ b/drivers/net/wireless/cisco/Kconfig
@@ -14,7 +14,7 @@ if WLAN_VENDOR_CISCO
 
 config AIRO
 	tristate "Cisco/Aironet 34X/35X/4500/4800 ISA and PCI cards"
-	depends on CFG80211 && ISA_DMA_API && (PCI || BROKEN)
+	depends on CFG80211 && (PCI || BROKEN)
 	select WIRELESS_EXT
 	select CRYPTO
 	select CRYPTO_SKCIPHER
-- 
cgit v1.2.3-70-g09d2


From 6c6d62ae8271bd4b55dd2ba4b7ed552162823880 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Tue, 18 Apr 2023 13:29:22 +0200
Subject: wifi: rtw88: Update spelling in main.h

Update spelling in comments in main.h

Found by inspection.

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20230418-rtw88-starspell-v1-1-70e52a23979b@kernel.org
---
 drivers/net/wireless/realtek/rtw88/main.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 155525b8df35..a563285e90ed 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -88,7 +88,7 @@ enum rtw_supported_band {
 	RTW_BAND_60G = BIT(NL80211_BAND_60GHZ),
 };
 
-/* now, support upto 80M bw */
+/* now, support up to 80M bw */
 #define RTW_MAX_CHANNEL_WIDTH RTW_CHANNEL_WIDTH_80
 
 enum rtw_bandwidth {
@@ -1881,7 +1881,7 @@ enum rtw_sar_bands {
 	RTW_SAR_BAND_NR,
 };
 
-/* the union is reserved for other knids of SAR sources
+/* the union is reserved for other kinds of SAR sources
  * which might not re-use same format with array common.
  */
 union rtw_sar_cfg {
@@ -2032,7 +2032,7 @@ struct rtw_dev {
 	struct rtw_tx_report tx_report;
 
 	struct {
-		/* incicate the mail box to use with fw */
+		/* indicate the mail box to use with fw */
 		u8 last_box_num;
 		u32 seq;
 	} h2c;
-- 
cgit v1.2.3-70-g09d2


From e315e7b83a22043bffee450437d7089ef373cbf6 Mon Sep 17 00:00:00 2001
From: Zhengchao Shao <shaozhengchao@huawei.com>
Date: Tue, 18 Apr 2023 14:54:50 +0800
Subject: net: libwx: fix memory leak in wx_setup_rx_resources

When wx_alloc_page_pool() failed in wx_setup_rx_resources(), it doesn't
release DMA buffer. Add dma_free_coherent() in the error path to release
the DMA buffer.

Fixes: 850b971110b2 ("net: libwx: Allocate Rx and Tx resources")
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230418065450.2268522-1-shaozhengchao@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
index eb89a274083e..1e8d8b7b0c62 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
@@ -1798,10 +1798,13 @@ static int wx_setup_rx_resources(struct wx_ring *rx_ring)
 	ret = wx_alloc_page_pool(rx_ring);
 	if (ret < 0) {
 		dev_err(rx_ring->dev, "Page pool creation failed: %d\n", ret);
-		goto err;
+		goto err_desc;
 	}
 
 	return 0;
+
+err_desc:
+	dma_free_coherent(dev, rx_ring->size, rx_ring->desc, rx_ring->dma);
 err:
 	kvfree(rx_ring->rx_buffer_info);
 	rx_ring->rx_buffer_info = NULL;
-- 
cgit v1.2.3-70-g09d2


From 4f3ed1293feb9502dc254b05802faf1ad3317ac6 Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Sun, 16 Apr 2023 19:12:22 +0000
Subject: ixgbe: Allow flow hash to be set via ethtool

ixgbe currently returns `EINVAL` whenever the flowhash it set by ethtool
because the ethtool code in the kernel passes a non-zero value for hfunc
that ixgbe should allow.

When ethtool is called with `ETHTOOL_SRXFHINDIR`,
`ethtool_set_rxfh_indir` will call ixgbe's set_rxfh function
with `ETH_RSS_HASH_NO_CHANGE`. This value should be accepted.

When ethtool is called with `ETHTOOL_SRSSH`, `ethtool_set_rxfh` will
call ixgbe's set_rxfh function with `rxfh.hfunc`, which appears to be
hardcoded in ixgbe to always be `ETH_RSS_HASH_TOP`. This value should
also be accepted.

Before this patch:

$ sudo ethtool -L eth1 combined 10
$ sudo ethtool -X eth1 default
Cannot set RX flow hash configuration: Invalid argument

After this patch:

$ sudo ethtool -L eth1 combined 10
$ sudo ethtool -X eth1 default
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 10 RX ring(s):
    0:      0     1     2     3     4     5     6     7
    8:      8     9     0     1     2     3     4     5
   16:      6     7     8     9     0     1     2     3
   24:      4     5     6     7     8     9     0     1
   ...

Fixes: 1c7cf0784e4d ("ixgbe: support for ethtool set_rxfh")
Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 6cfc9dc16537..821dfd323fa9 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -3131,8 +3131,8 @@ static int ixgbe_set_rxfh(struct net_device *netdev, const u32 *indir,
 	int i;
 	u32 reta_entries = ixgbe_rss_indir_tbl_entries(adapter);
 
-	if (hfunc)
-		return -EINVAL;
+	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+		return -EOPNOTSUPP;
 
 	/* Fill out the redirection table */
 	if (indir) {
-- 
cgit v1.2.3-70-g09d2


From e85d3d55875f7a1079edfbc4e4e98d6f8aea9ac7 Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Sun, 16 Apr 2023 19:12:23 +0000
Subject: ixgbe: Enable setting RSS table to default values

ethtool uses `ETHTOOL_GRXRINGS` to compute how many queues are supported
by RSS. The driver should return the smaller of either:
  - The maximum number of RSS queues the device supports, OR
  - The number of RX queues configured

Prior to this change, running `ethtool -X $iface default` fails if the
number of queues configured is larger than the number supported by RSS,
even though changing the queue count correctly resets the flowhash to
use all supported queues.

Other drivers (for example, i40e) will succeed but the flow hash will
reset to support the maximum number of queues supported by RSS, even if
that amount is smaller than the configured amount.

Prior to this change:

$ sudo ethtool -L eth1 combined 20
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 20 RX ring(s):
    0:      0     1     2     3     4     5     6     7
    8:      8     9    10    11    12    13    14    15
   16:      0     1     2     3     4     5     6     7
   24:      8     9    10    11    12    13    14    15
   32:      0     1     2     3     4     5     6     7
...

You can see that the flowhash was correctly set to use the maximum
number of queues supported by the driver (16).

However, asking the NIC to reset to "default" fails:

$ sudo ethtool -X eth1 default
Cannot set RX flow hash configuration: Invalid argument

After this change, the flowhash can be reset to default which will use
all of the available RSS queues (16) or the configured queue count,
whichever is smaller.

Starting with eth1 which has 10 queues and a flowhash distributing to
all 10 queues:

$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 10 RX ring(s):
    0:      0     1     2     3     4     5     6     7
    8:      8     9     0     1     2     3     4     5
   16:      6     7     8     9     0     1     2     3
...

Increasing the queue count to 48 resets the flowhash to distribute to 16
queues, as it did before this patch:

$ sudo ethtool -L eth1 combined 48
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 16 RX ring(s):
    0:      0     1     2     3     4     5     6     7
    8:      8     9    10    11    12    13    14    15
   16:      0     1     2     3     4     5     6     7
...

Due to the other bugfix in this series, the flowhash can be set to use
queues 0-5:

$ sudo ethtool -X eth1 equal 5
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 16 RX ring(s):
    0:      0     1     2     3     4     0     1     2
    8:      3     4     0     1     2     3     4     0
   16:      1     2     3     4     0     1     2     3
...

Due to this bugfix, the flowhash can be reset to default and use 16
queues:

$ sudo ethtool -X eth1 default
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 16 RX ring(s):
    0:      0     1     2     3     4     5     6     7
    8:      8     9    10    11    12    13    14    15
   16:      0     1     2     3     4     5     6     7
...

Fixes: 91cd94bfe4f0 ("ixgbe: add basic support for setting and getting nfc controls")
Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 821dfd323fa9..0bbad4a5cc2f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -2665,6 +2665,14 @@ static int ixgbe_get_rss_hash_opts(struct ixgbe_adapter *adapter,
 	return 0;
 }
 
+static int ixgbe_rss_indir_tbl_max(struct ixgbe_adapter *adapter)
+{
+	if (adapter->hw.mac.type < ixgbe_mac_X550)
+		return 16;
+	else
+		return 64;
+}
+
 static int ixgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 			   u32 *rule_locs)
 {
@@ -2673,7 +2681,8 @@ static int ixgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 
 	switch (cmd->cmd) {
 	case ETHTOOL_GRXRINGS:
-		cmd->data = adapter->num_rx_queues;
+		cmd->data = min_t(int, adapter->num_rx_queues,
+				  ixgbe_rss_indir_tbl_max(adapter));
 		ret = 0;
 		break;
 	case ETHTOOL_GRXCLSRLCNT:
@@ -3075,14 +3084,6 @@ static int ixgbe_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 	return ret;
 }
 
-static int ixgbe_rss_indir_tbl_max(struct ixgbe_adapter *adapter)
-{
-	if (adapter->hw.mac.type < ixgbe_mac_X550)
-		return 16;
-	else
-		return 64;
-}
-
 static u32 ixgbe_get_rxfh_key_size(struct net_device *netdev)
 {
 	return IXGBE_RSS_KEY_SIZE;
-- 
cgit v1.2.3-70-g09d2


From 2f8fdcb0a73a1831cc4f205f23493a17c0e5536f Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Wed, 12 Apr 2023 10:19:24 +0200
Subject: ice: do not busy-wait to read GNSS data

The ice-gnss-<dev_name> kernel thread, which reads data from the u-blox
GNSS module, keep a CPU core almost 100% busy. The main reason is that
it busy-waits for data to become available.

A simple improvement would be to replace the "mdelay(10);" in
ice_gnss_read() with sleeping. A better fix is to not do any waiting
directly in the function and just requeue this delayed work as needed.
The advantage is that canceling the work from ice_gnss_exit() becomes
immediate, rather than taking up to ~2.5 seconds (ICE_MAX_UBX_READ_TRIES
* 10 ms).

This lowers the CPU usage of the ice-gnss-<dev_name> thread on my system
from ~90 % to ~8 %.

I am not sure if the larger 0.1 s pause after inserting data into the
gnss subsystem is really necessary, but I'm keeping that as it was.

Of course, ideally the driver would not have to poll at all, but I don't
know if the E810 can watch for GNSS data availability over the i2c bus
by itself and notify the driver.

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_gnss.c | 42 ++++++++++++++-----------------
 drivers/net/ethernet/intel/ice/ice_gnss.h |  3 +--
 2 files changed, 20 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_gnss.c b/drivers/net/ethernet/intel/ice/ice_gnss.c
index 8dec748bb53a..2ea8a2b11bcd 100644
--- a/drivers/net/ethernet/intel/ice/ice_gnss.c
+++ b/drivers/net/ethernet/intel/ice/ice_gnss.c
@@ -117,6 +117,7 @@ static void ice_gnss_read(struct kthread_work *work)
 {
 	struct gnss_serial *gnss = container_of(work, struct gnss_serial,
 						read_work.work);
+	unsigned long delay = ICE_GNSS_POLL_DATA_DELAY_TIME;
 	unsigned int i, bytes_read, data_len, count;
 	struct ice_aqc_link_topo_addr link_topo;
 	struct ice_pf *pf;
@@ -136,11 +137,6 @@ static void ice_gnss_read(struct kthread_work *work)
 		return;
 
 	hw = &pf->hw;
-	buf = (char *)get_zeroed_page(GFP_KERNEL);
-	if (!buf) {
-		err = -ENOMEM;
-		goto exit;
-	}
 
 	memset(&link_topo, 0, sizeof(struct ice_aqc_link_topo_addr));
 	link_topo.topo_params.index = ICE_E810T_GNSS_I2C_BUS;
@@ -151,25 +147,24 @@ static void ice_gnss_read(struct kthread_work *work)
 	i2c_params = ICE_GNSS_UBX_DATA_LEN_WIDTH |
 		     ICE_AQC_I2C_USE_REPEATED_START;
 
-	/* Read data length in a loop, when it's not 0 the data is ready */
-	for (i = 0; i < ICE_MAX_UBX_READ_TRIES; i++) {
-		err = ice_aq_read_i2c(hw, link_topo, ICE_GNSS_UBX_I2C_BUS_ADDR,
-				      cpu_to_le16(ICE_GNSS_UBX_DATA_LEN_H),
-				      i2c_params, (u8 *)&data_len_b, NULL);
-		if (err)
-			goto exit_buf;
+	err = ice_aq_read_i2c(hw, link_topo, ICE_GNSS_UBX_I2C_BUS_ADDR,
+			      cpu_to_le16(ICE_GNSS_UBX_DATA_LEN_H),
+			      i2c_params, (u8 *)&data_len_b, NULL);
+	if (err)
+		goto requeue;
 
-		data_len = be16_to_cpu(data_len_b);
-		if (data_len != 0 && data_len != U16_MAX)
-			break;
+	data_len = be16_to_cpu(data_len_b);
+	if (data_len == 0 || data_len == U16_MAX)
+		goto requeue;
 
-		mdelay(10);
-	}
+	/* The u-blox has data_len bytes for us to read */
 
 	data_len = min_t(typeof(data_len), data_len, PAGE_SIZE);
-	if (!data_len) {
+
+	buf = (char *)get_zeroed_page(GFP_KERNEL);
+	if (!buf) {
 		err = -ENOMEM;
-		goto exit_buf;
+		goto requeue;
 	}
 
 	/* Read received data */
@@ -183,7 +178,7 @@ static void ice_gnss_read(struct kthread_work *work)
 				      cpu_to_le16(ICE_GNSS_UBX_EMPTY_DATA),
 				      bytes_read, &buf[i], NULL);
 		if (err)
-			goto exit_buf;
+			goto free_buf;
 	}
 
 	count = gnss_insert_raw(pf->gnss_dev, buf, i);
@@ -191,10 +186,11 @@ static void ice_gnss_read(struct kthread_work *work)
 		dev_warn(ice_pf_to_dev(pf),
 			 "gnss_insert_raw ret=%d size=%d\n",
 			 count, i);
-exit_buf:
+	delay = ICE_GNSS_TIMER_DELAY_TIME;
+free_buf:
 	free_page((unsigned long)buf);
-	kthread_queue_delayed_work(gnss->kworker, &gnss->read_work,
-				   ICE_GNSS_TIMER_DELAY_TIME);
+requeue:
+	kthread_queue_delayed_work(gnss->kworker, &gnss->read_work, delay);
 exit:
 	if (err)
 		dev_dbg(ice_pf_to_dev(pf), "GNSS failed to read err=%d\n", err);
diff --git a/drivers/net/ethernet/intel/ice/ice_gnss.h b/drivers/net/ethernet/intel/ice/ice_gnss.h
index 4d49e5b0b4b8..640df7411373 100644
--- a/drivers/net/ethernet/intel/ice/ice_gnss.h
+++ b/drivers/net/ethernet/intel/ice/ice_gnss.h
@@ -5,6 +5,7 @@
 #define _ICE_GNSS_H_
 
 #define ICE_E810T_GNSS_I2C_BUS		0x2
+#define ICE_GNSS_POLL_DATA_DELAY_TIME	(HZ / 100) /* poll every 10 ms */
 #define ICE_GNSS_TIMER_DELAY_TIME	(HZ / 10) /* 0.1 second per message */
 #define ICE_GNSS_TTY_WRITE_BUF		250
 #define ICE_MAX_I2C_DATA_SIZE		FIELD_MAX(ICE_AQC_I2C_DATA_SIZE_M)
@@ -20,8 +21,6 @@
  * passed as I2C addr parameter.
  */
 #define ICE_GNSS_UBX_WRITE_BYTES	(ICE_MAX_I2C_WRITE_BYTES + 1)
-#define ICE_MAX_UBX_READ_TRIES		255
-#define ICE_MAX_UBX_ACK_READ_TRIES	4095
 
 struct gnss_write_buf {
 	struct list_head queue;
-- 
cgit v1.2.3-70-g09d2


From 0ec636e571f59b741d1eca47d439181f29f7f083 Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Wed, 12 Apr 2023 10:19:25 +0200
Subject: ice: increase the GNSS data polling interval to 20 ms

Double the GNSS data polling interval from 10 ms to 20 ms.
According to Karol Kolacinski from the Intel team, they have been
planning to make this change.

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_gnss.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_gnss.h b/drivers/net/ethernet/intel/ice/ice_gnss.h
index 640df7411373..b8bb8b63d081 100644
--- a/drivers/net/ethernet/intel/ice/ice_gnss.h
+++ b/drivers/net/ethernet/intel/ice/ice_gnss.h
@@ -5,7 +5,7 @@
 #define _ICE_GNSS_H_
 
 #define ICE_E810T_GNSS_I2C_BUS		0x2
-#define ICE_GNSS_POLL_DATA_DELAY_TIME	(HZ / 100) /* poll every 10 ms */
+#define ICE_GNSS_POLL_DATA_DELAY_TIME	(HZ / 50) /* poll every 20 ms */
 #define ICE_GNSS_TIMER_DELAY_TIME	(HZ / 10) /* 0.1 second per message */
 #define ICE_GNSS_TTY_WRITE_BUF		250
 #define ICE_MAX_I2C_DATA_SIZE		FIELD_MAX(ICE_AQC_I2C_DATA_SIZE_M)
-- 
cgit v1.2.3-70-g09d2


From 84817ab66fd0d9cde371fcb020f610ba808585bb Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Wed, 12 Apr 2023 10:19:26 +0200
Subject: ice: remove ice_ctl_q_info::sq_cmd_timeout

sq_cmd_timeout is initialized to ICE_CTL_Q_SQ_CMD_TIMEOUT and never
changed, so just use the constant directly.

Suggested-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c   | 2 +-
 drivers/net/ethernet/intel/ice/ice_controlq.c | 5 +----
 drivers/net/ethernet/intel/ice/ice_controlq.h | 1 -
 3 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index c2fda4fa4188..f4c256563248 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -2000,7 +2000,7 @@ void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res)
 	/* there are some rare cases when trying to release the resource
 	 * results in an admin queue timeout, so handle them correctly
 	 */
-	while ((status == -EIO) && (total_delay < hw->adminq.sq_cmd_timeout)) {
+	while ((status == -EIO) && (total_delay < ICE_CTL_Q_SQ_CMD_TIMEOUT)) {
 		mdelay(1);
 		status = ice_aq_release_res(hw, res, 0, NULL);
 		total_delay++;
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c
index 6bcfee295991..c8fb10106ec3 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.c
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.c
@@ -637,9 +637,6 @@ static int ice_init_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type)
 		return -EIO;
 	}
 
-	/* setup SQ command write back timeout */
-	cq->sq_cmd_timeout = ICE_CTL_Q_SQ_CMD_TIMEOUT;
-
 	/* allocate the ATQ */
 	ret_code = ice_init_sq(hw, cq);
 	if (ret_code)
@@ -1066,7 +1063,7 @@ ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 
 		udelay(ICE_CTL_Q_SQ_CMD_USEC);
 		total_delay++;
-	} while (total_delay < cq->sq_cmd_timeout);
+	} while (total_delay < ICE_CTL_Q_SQ_CMD_TIMEOUT);
 
 	/* if ready, copy the desc back to temp */
 	if (ice_sq_done(hw, cq)) {
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.h b/drivers/net/ethernet/intel/ice/ice_controlq.h
index c07e9cc9fc6e..e790b2f4e437 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.h
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.h
@@ -87,7 +87,6 @@ struct ice_ctl_q_info {
 	enum ice_ctl_q qtype;
 	struct ice_ctl_q_ring rq;	/* receive queue */
 	struct ice_ctl_q_ring sq;	/* send queue */
-	u32 sq_cmd_timeout;		/* send queue cmd write back timeout */
 	u16 num_rq_entries;		/* receive queue depth */
 	u16 num_sq_entries;		/* send queue depth */
 	u16 rq_buf_size;		/* receive queue buffer size */
-- 
cgit v1.2.3-70-g09d2


From f86d6f9c49f67b05d117495dc440aaa18c66a9da Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Wed, 12 Apr 2023 10:19:27 +0200
Subject: ice: sleep, don't busy-wait, for ICE_CTL_Q_SQ_CMD_TIMEOUT
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver polls for ice_sq_done() with a 100 µs period for up to 1 s
and it uses udelay to do that.

Let's use usleep_range instead. We know sleeping is allowed here,
because we're holding a mutex (cq->sq_lock). To preserve the total
max waiting time, measure the timeout in jiffies.

ICE_CTL_Q_SQ_CMD_TIMEOUT is used also in ice_release_res(), but there
the polling period is 1 ms (i.e. 10 times longer). Since the timeout was
expressed in terms of the number of loops, the total timeout in this
function is 10 s. I do not know if this is intentional. This patch keeps
it.

The patch lowers the CPU usage of the ice-gnss-<dev_name> kernel thread
on my system from ~8 % to less than 1 %.

I received a report of high CPU usage with ptp4l where the busy-waiting
in ice_sq_send_cmd dominated the profile. This patch has been tested in
that usecase too and it made a huge improvement there.

Tested-by: Brent Rowsell <browsell@redhat.com>
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c   | 14 +++++++-------
 drivers/net/ethernet/intel/ice/ice_controlq.c |  9 +++++----
 drivers/net/ethernet/intel/ice/ice_controlq.h |  2 +-
 3 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index f4c256563248..3638598d732b 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1992,19 +1992,19 @@ ice_acquire_res_exit:
  */
 void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res)
 {
-	u32 total_delay = 0;
+	unsigned long timeout;
 	int status;
 
-	status = ice_aq_release_res(hw, res, 0, NULL);
-
 	/* there are some rare cases when trying to release the resource
 	 * results in an admin queue timeout, so handle them correctly
 	 */
-	while ((status == -EIO) && (total_delay < ICE_CTL_Q_SQ_CMD_TIMEOUT)) {
-		mdelay(1);
+	timeout = jiffies + 10 * ICE_CTL_Q_SQ_CMD_TIMEOUT;
+	do {
 		status = ice_aq_release_res(hw, res, 0, NULL);
-		total_delay++;
-	}
+		if (status != -EIO)
+			break;
+		usleep_range(1000, 2000);
+	} while (time_before(jiffies, timeout));
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c
index c8fb10106ec3..d2faf1baad2f 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.c
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.c
@@ -964,7 +964,7 @@ ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 	struct ice_aq_desc *desc_on_ring;
 	bool cmd_completed = false;
 	struct ice_sq_cd *details;
-	u32 total_delay = 0;
+	unsigned long timeout;
 	int status = 0;
 	u16 retval = 0;
 	u32 val = 0;
@@ -1057,13 +1057,14 @@ ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 		cq->sq.next_to_use = 0;
 	wr32(hw, cq->sq.tail, cq->sq.next_to_use);
 
+	timeout = jiffies + ICE_CTL_Q_SQ_CMD_TIMEOUT;
 	do {
 		if (ice_sq_done(hw, cq))
 			break;
 
-		udelay(ICE_CTL_Q_SQ_CMD_USEC);
-		total_delay++;
-	} while (total_delay < ICE_CTL_Q_SQ_CMD_TIMEOUT);
+		usleep_range(ICE_CTL_Q_SQ_CMD_USEC,
+			     ICE_CTL_Q_SQ_CMD_USEC * 3 / 2);
+	} while (time_before(jiffies, timeout));
 
 	/* if ready, copy the desc back to temp */
 	if (ice_sq_done(hw, cq)) {
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.h b/drivers/net/ethernet/intel/ice/ice_controlq.h
index e790b2f4e437..950b7f4a7a05 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.h
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.h
@@ -34,7 +34,7 @@ enum ice_ctl_q {
 };
 
 /* Control Queue timeout settings - max delay 1s */
-#define ICE_CTL_Q_SQ_CMD_TIMEOUT	10000 /* Count 10000 times */
+#define ICE_CTL_Q_SQ_CMD_TIMEOUT	HZ    /* Wait max 1s */
 #define ICE_CTL_Q_SQ_CMD_USEC		100   /* Check every 100usec */
 #define ICE_CTL_Q_ADMIN_INIT_TIMEOUT	10    /* Count 10 times */
 #define ICE_CTL_Q_ADMIN_INIT_MSEC	100   /* Check every 100msec */
-- 
cgit v1.2.3-70-g09d2


From 43a630e37e259fee83ab3fd769c42e2fed97ca81 Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Wed, 12 Apr 2023 10:19:28 +0200
Subject: ice: remove unused buffer copy code in ice_sq_send_cmd_retry()

The 'buf_cpy'-related code in ice_sq_send_cmd_retry() looks broken.
'buf' is nowhere copied into 'buf_cpy'.

The reason this does not cause problems is that all commands for which
'is_cmd_for_retry' is true go with a NULL buf.

Let's remove 'buf_cpy'. Add a WARN_ON in case the assumption no longer
holds in the future.

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 3638598d732b..c6200564304e 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1619,7 +1619,6 @@ ice_sq_send_cmd_retry(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 {
 	struct ice_aq_desc desc_cpy;
 	bool is_cmd_for_retry;
-	u8 *buf_cpy = NULL;
 	u8 idx = 0;
 	u16 opcode;
 	int status;
@@ -1629,11 +1628,8 @@ ice_sq_send_cmd_retry(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 	memset(&desc_cpy, 0, sizeof(desc_cpy));
 
 	if (is_cmd_for_retry) {
-		if (buf) {
-			buf_cpy = kzalloc(buf_size, GFP_KERNEL);
-			if (!buf_cpy)
-				return -ENOMEM;
-		}
+		/* All retryable cmds are direct, without buf. */
+		WARN_ON(buf);
 
 		memcpy(&desc_cpy, desc, sizeof(desc_cpy));
 	}
@@ -1645,17 +1641,12 @@ ice_sq_send_cmd_retry(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 		    hw->adminq.sq_last_status != ICE_AQ_RC_EBUSY)
 			break;
 
-		if (buf_cpy)
-			memcpy(buf, buf_cpy, buf_size);
-
 		memcpy(desc, &desc_cpy, sizeof(desc_cpy));
 
 		mdelay(ICE_SQ_SEND_DELAY_TIME_MS);
 
 	} while (++idx < ICE_SQ_SEND_MAX_EXECUTE);
 
-	kfree(buf_cpy);
-
 	return status;
 }
 
-- 
cgit v1.2.3-70-g09d2


From b488ae52ef9f74155ab358f8c68e74327b45e0e1 Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Wed, 12 Apr 2023 10:19:29 +0200
Subject: ice: sleep, don't busy-wait, in the SQ send retry loop

10 ms is a lot of time to spend busy-waiting. Sleeping is clearly
allowed here, because we have just returned from ice_sq_send_cmd(),
which takes a mutex.

On kernels with HZ=100, this msleep may be twice as long, but I don't
think it matters.
I did not actually observe any retries happening here.

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index c6200564304e..0157f6e98d3e 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1643,7 +1643,7 @@ ice_sq_send_cmd_retry(struct ice_hw *hw, struct ice_ctl_q_info *cq,
 
 		memcpy(desc, &desc_cpy, sizeof(desc_cpy));
 
-		mdelay(ICE_SQ_SEND_DELAY_TIME_MS);
+		msleep(ICE_SQ_SEND_DELAY_TIME_MS);
 
 	} while (++idx < ICE_SQ_SEND_MAX_EXECUTE);
 
-- 
cgit v1.2.3-70-g09d2


From 4b7ef71ac977899cf7bdb09d66052926c7d249ff Mon Sep 17 00:00:00 2001
From: Quentin Monnet <quentin@isovalent.com>
Date: Thu, 20 Apr 2023 01:33:33 +0100
Subject: bpftool: Replace "__fallthrough" by a comment to address merge
 conflict

The recent support for inline annotations in control flow graphs
generated by bpftool introduced the usage of the "__fallthrough" macro
in a switch/case block in btf_dumper.c. This change went through the
bpf-next tree, but resulted in a merge conflict in linux-next, because
this macro has been renamed "fallthrough" (no underscores) in the
meantime.

To address the conflict, we temporarily switch to a simple comment
instead of a macro.

Related: commit f7a858bffcdd ("tools: Rename __fallthrough to fallthrough")

Fixes: 9fd496848b1c ("bpftool: Support inline annotations when dumping the CFG of a program")
Reported-by: Sven Schnelle <svens@linux.ibm.com>
Reported-by: Thomas Richter <tmricht@linux.ibm.com>
Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/all/yt9dttxlwal7.fsf@linux.ibm.com/
Link: https://lore.kernel.org/bpf/20230412123636.2358949-1-tmricht@linux.ibm.com/
Link: https://lore.kernel.org/bpf/20230420003333.90901-1-quentin@isovalent.com
---
 tools/bpf/bpftool/btf_dumper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c
index 6c5e0e82da22..294de231db99 100644
--- a/tools/bpf/bpftool/btf_dumper.c
+++ b/tools/bpf/bpftool/btf_dumper.c
@@ -835,7 +835,7 @@ static void dotlabel_puts(const char *s)
 		case '|':
 		case ' ':
 			putchar('\\');
-			__fallthrough;
+			/* fallthrough */
 		default:
 			putchar(*s);
 		}
-- 
cgit v1.2.3-70-g09d2


From 7c4b96c00043f3b2cad50653f085c178fda92e62 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 02:23:14 +0300
Subject: selftests/bpf: disable program test run for progs/refcounted_kptr.c

Florian Westphal found a bug in test_loader.c processing of __retval
tag. Because of this bug the function test_loader.c:do_prog_test_run()
never executed and all __retval test tags were ignored. This hid an
issue with progs/refcounted_kptr.c tests.

When __retval tag bug is fixed and refcounted_kptr.c tests are run
kernel reports various issues and eventually hangs. Shortest reproducer
is the following command run a few times:

  $ for i in $(seq 1 4); do (./test_progs --allow=refcounted_kptr &); done

Commenting out __retval tags for these tests until this issue is resolved.

Reported-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/bpf/f4c4aee644425842ee6aa8edf1da68f0a8260e7c.camel@gmail.com/T/
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-2-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/refcounted_kptr.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/refcounted_kptr.c b/tools/testing/selftests/bpf/progs/refcounted_kptr.c
index 1d348a225140..b6b2d4f97b19 100644
--- a/tools/testing/selftests/bpf/progs/refcounted_kptr.c
+++ b/tools/testing/selftests/bpf/progs/refcounted_kptr.c
@@ -219,7 +219,7 @@ static long __read_from_unstash(int idx)
 #define INSERT_READ_BOTH(rem_tree, rem_list, desc)			\
 SEC("tc")								\
 __description(desc)							\
-__success __retval(579)							\
+__success /* __retval(579) temporarily disabled */			\
 long insert_and_remove_tree_##rem_tree##_list_##rem_list(void *ctx)	\
 {									\
 	long err, tree_data, list_data;					\
@@ -258,7 +258,7 @@ INSERT_READ_BOTH(false, true, "insert_read_both: remove from list");
 #define INSERT_READ_BOTH(rem_tree, rem_list, desc)			\
 SEC("tc")								\
 __description(desc)							\
-__success __retval(579)							\
+__success /* __retval(579) temporarily disabled */			\
 long insert_and_remove_lf_tree_##rem_tree##_list_##rem_list(void *ctx)	\
 {									\
 	long err, tree_data, list_data;					\
@@ -296,7 +296,7 @@ INSERT_READ_BOTH(false, true, "insert_read_both_list_first: remove from list");
 #define INSERT_DOUBLE_READ_AND_DEL(read_fn, read_root, desc)		\
 SEC("tc")								\
 __description(desc)							\
-__success __retval(-1)							\
+__success /* temporarily __retval(-1) disabled */			\
 long insert_double_##read_fn##_and_del_##read_root(void *ctx)		\
 {									\
 	long err, list_data;						\
@@ -329,7 +329,7 @@ INSERT_DOUBLE_READ_AND_DEL(__read_from_list, head, "insert_double_del: 2x read-a
 #define INSERT_STASH_READ(rem_tree, desc)				\
 SEC("tc")								\
 __description(desc)							\
-__success __retval(84)							\
+__success /* __retval(84) temporarily disabled */			\
 long insert_rbtree_and_stash__del_tree_##rem_tree(void *ctx)		\
 {									\
 	long err, tree_data, map_data;					\
-- 
cgit v1.2.3-70-g09d2


From 7cdddb99e4a69c26881d279472f1487cd67740c4 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 02:23:15 +0300
Subject: selftests/bpf: fix __retval() being always ignored

Florian Westphal found a bug in and suggested a fix for test_loader.c
processing of __retval tag. Because of this bug the function
test_loader.c:do_prog_test_run() never executed and all __retval test
tags were ignored.

If this bug is fixed a number of test cases from
progs/verifier_array_access.c fail with retval not matching the
expected value. This test was recently converted to use test_loader.c
and inline assembly in [1]. When doing the conversion I missed the
important detail of test_verifier.c operation: when it creates
fixup_map_array_ro, fixup_map_array_wo and fixup_map_array_small it
populates these maps with a dummy record.

Disabling the __retval checks for the affected verifier_array_access
in this commit to avoid false-postivies in any potential bisects.
The issue is addressed in the next patch.

I verified that the __retval tags are now respected by changing
expected return values for all tests annotated with __retval, and
checking that these tests started to fail.

[1] https://lore.kernel.org/bpf/20230325025524.144043-1-eddyz87@gmail.com/

Fixes: 19a8e06f5f91 ("selftests/bpf: Tests execution support for test_loader.c")
Reported-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/bpf/f4c4aee644425842ee6aa8edf1da68f0a8260e7c.camel@gmail.com/T/
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-3-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/verifier_array_access.c | 4 ++--
 tools/testing/selftests/bpf/test_loader.c                 | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/verifier_array_access.c b/tools/testing/selftests/bpf/progs/verifier_array_access.c
index 95d7ecc12963..fceeeef78721 100644
--- a/tools/testing/selftests/bpf/progs/verifier_array_access.c
+++ b/tools/testing/selftests/bpf/progs/verifier_array_access.c
@@ -330,7 +330,7 @@ l0_%=:	exit;						\
 
 SEC("socket")
 __description("valid read map access into a read-only array 1")
-__success __success_unpriv __retval(28)
+__success __success_unpriv /* __retval(28) temporarily disable */
 __naked void a_read_only_array_1_1(void)
 {
 	asm volatile ("					\
@@ -351,7 +351,7 @@ l0_%=:	exit;						\
 
 SEC("tc")
 __description("valid read map access into a read-only array 2")
-__success __retval(65507)
+__success /* __retval(65507) temporarily disable */
 __naked void a_read_only_array_2_1(void)
 {
 	asm volatile ("					\
diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index 47e9e076bc8f..e2a1bdc5a570 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -587,7 +587,7 @@ void run_subtest(struct test_loader *tester,
 		/* For some reason test_verifier executes programs
 		 * with all capabilities restored. Do the same here.
 		 */
-		if (!restore_capabilities(&caps))
+		if (restore_capabilities(&caps))
 			goto tobj_cleanup;
 
 		do_prog_test_run(bpf_program__fd(tprog), &retval);
-- 
cgit v1.2.3-70-g09d2


From 5b22f4d1436b2693c7176a76be0ba6d30503bf7c Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 02:23:16 +0300
Subject: selftests/bpf: add pre bpf_prog_test_run_opts() callback for
 test_loader

When a test case is annotated with __retval tag the test_loader engine
would use libbpf's bpf_prog_test_run_opts() to do a test run of the
program and compare retvals.

This commit allows to perform arbitrary actions on bpf object right
before test loader invokes bpf_prog_test_run_opts(). This could be
used to setup some state for program execution, e.g. fill some maps.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-4-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/test_loader.c | 8 ++++++++
 tools/testing/selftests/bpf/test_progs.h  | 9 +++++++++
 2 files changed, 17 insertions(+)

diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index e2a1bdc5a570..40c9b7d532c4 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -590,6 +590,14 @@ void run_subtest(struct test_loader *tester,
 		if (restore_capabilities(&caps))
 			goto tobj_cleanup;
 
+		if (tester->pre_execution_cb) {
+			err = tester->pre_execution_cb(tobj);
+			if (err) {
+				PRINT_FAIL("pre_execution_cb failed: %d\n", err);
+				goto tobj_cleanup;
+			}
+		}
+
 		do_prog_test_run(bpf_program__fd(tprog), &retval);
 		if (retval != subspec->retval && subspec->retval != POINTER_VALUE) {
 			PRINT_FAIL("Unexpected retval: %d != %d\n", retval, subspec->retval);
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 10ba43250668..0ed3134333d4 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -424,14 +424,23 @@ int get_bpf_max_tramp_links(void);
 
 #define BPF_TESTMOD_TEST_FILE "/sys/kernel/bpf_testmod"
 
+typedef int (*pre_execution_cb)(struct bpf_object *obj);
+
 struct test_loader {
 	char *log_buf;
 	size_t log_buf_sz;
 	size_t next_match_pos;
+	pre_execution_cb pre_execution_cb;
 
 	struct bpf_object *obj;
 };
 
+static inline void test_loader__set_pre_execution_cb(struct test_loader *tester,
+						     pre_execution_cb cb)
+{
+	tester->pre_execution_cb = cb;
+}
+
 typedef const void *(*skel_elf_bytes_fn)(size_t *sz);
 
 extern void test_loader__run_subtests(struct test_loader *tester,
-- 
cgit v1.2.3-70-g09d2


From cbb110bc6672f785cab2cb308e9cfefee07af861 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 02:23:17 +0300
Subject: selftests/bpf: populate map_array_ro map for verifier_array_access
 test

Two test cases:
- "valid read map access into a read-only array 1" and
- "valid read map access into a read-only array 2"

Expect that map_array_ro map is filled with mock data. This logic was
not taken into acount during initial test conversion.

This commit modifies prog_tests/verifier.c entry point for this test
to fill the map.

Fixes: a3c830ae0209 ("selftests/bpf: verifier/array_access.c converted to inline assembly")
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-5-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  | 42 ++++++++++++++++++++--
 .../selftests/bpf/progs/verifier_array_access.c    |  4 +--
 2 files changed, 41 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 25bc8958dbfe..7c68d78da9ea 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -44,8 +44,17 @@
 #include "verifier_xdp.skel.h"
 #include "verifier_xdp_direct_packet_access.skel.h"
 
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
 __maybe_unused
-static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_factory)
+static void run_tests_aux(const char *skel_name,
+			  skel_elf_bytes_fn elf_bytes_factory,
+			  pre_execution_cb pre_execution_cb)
 {
 	struct test_loader tester = {};
 	__u64 old_caps;
@@ -58,6 +67,7 @@ static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_fac
 		return;
 	}
 
+	test_loader__set_pre_execution_cb(&tester, pre_execution_cb);
 	test_loader__run_subtests(&tester, skel_name, elf_bytes_factory);
 	test_loader_fini(&tester);
 
@@ -66,10 +76,9 @@ static void run_tests_aux(const char *skel_name, skel_elf_bytes_fn elf_bytes_fac
 		PRINT_FAIL("failed to restore CAP_SYS_ADMIN: %i, %s\n", err, strerror(err));
 }
 
-#define RUN(skel) run_tests_aux(#skel, skel##__elf_bytes)
+#define RUN(skel) run_tests_aux(#skel, skel##__elf_bytes, NULL)
 
 void test_verifier_and(void)                  { RUN(verifier_and); }
-void test_verifier_array_access(void)         { RUN(verifier_array_access); }
 void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
 void test_verifier_bounds_deduction_non_const(void)     { RUN(verifier_bounds_deduction_non_const); }
@@ -108,3 +117,30 @@ void test_verifier_var_off(void)              { RUN(verifier_var_off); }
 void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
 void test_verifier_xdp(void)                  { RUN(verifier_xdp); }
 void test_verifier_xdp_direct_packet_access(void) { RUN(verifier_xdp_direct_packet_access); }
+
+static int init_array_access_maps(struct bpf_object *obj)
+{
+	struct bpf_map *array_ro;
+	struct test_val value = {
+		.index = (6 + 1) * sizeof(int),
+		.foo[6] = 0xabcdef12,
+	};
+	int err, key = 0;
+
+	array_ro = bpf_object__find_map_by_name(obj, "map_array_ro");
+	if (!ASSERT_OK_PTR(array_ro, "lookup map_array_ro"))
+		return -EINVAL;
+
+	err = bpf_map_update_elem(bpf_map__fd(array_ro), &key, &value, 0);
+	if (!ASSERT_OK(err, "map_array_ro update"))
+		return err;
+
+	return 0;
+}
+
+void test_verifier_array_access(void)
+{
+	run_tests_aux("verifier_array_access",
+		      verifier_array_access__elf_bytes,
+		      init_array_access_maps);
+}
diff --git a/tools/testing/selftests/bpf/progs/verifier_array_access.c b/tools/testing/selftests/bpf/progs/verifier_array_access.c
index fceeeef78721..95d7ecc12963 100644
--- a/tools/testing/selftests/bpf/progs/verifier_array_access.c
+++ b/tools/testing/selftests/bpf/progs/verifier_array_access.c
@@ -330,7 +330,7 @@ l0_%=:	exit;						\
 
 SEC("socket")
 __description("valid read map access into a read-only array 1")
-__success __success_unpriv /* __retval(28) temporarily disable */
+__success __success_unpriv __retval(28)
 __naked void a_read_only_array_1_1(void)
 {
 	asm volatile ("					\
@@ -351,7 +351,7 @@ l0_%=:	exit;						\
 
 SEC("tc")
 __description("valid read map access into a read-only array 2")
-__success /* __retval(65507) temporarily disable */
+__success __retval(65507)
 __naked void a_read_only_array_2_1(void)
 {
 	asm volatile ("					\
-- 
cgit v1.2.3-70-g09d2


From 075056005d8cceecbbb8e054d8e3cd2b7519d9c6 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Sun, 1 Jan 2023 00:20:53 +0200
Subject: net/mlx5: DR, Fix dumping of legacy modify_hdr in debug dump

The steering dump parser expects to see 0 as rewrite num of actions
in case pattern/args aren't supported - parsing of legacy modify header
is based on this assumption.
Fix this to align to parser's expectation.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
index 1ff8bde90e1e..ea9f27db4c74 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
@@ -153,13 +153,15 @@ dr_dump_rule_action_mem(struct seq_file *file, const u64 rule_id,
 			   DR_DUMP_REC_TYPE_ACTION_MODIFY_HDR, action_id,
 			   rule_id, action->rewrite->index,
 			   action->rewrite->single_action_opt,
-			   action->rewrite->num_of_actions,
+			   ptrn_arg ? action->rewrite->num_of_actions : 0,
 			   ptrn_arg ? ptrn->index : 0,
 			   ptrn_arg ? mlx5dr_arg_get_obj_id(arg) : 0);
 
-		for (i = 0; i < action->rewrite->num_of_actions; i++) {
-			seq_printf(file, ",0x%016llx",
-				   be64_to_cpu(((__be64 *)rewrite_data)[i]));
+		if (ptrn_arg) {
+			for (i = 0; i < action->rewrite->num_of_actions; i++) {
+				seq_printf(file, ",0x%016llx",
+					   be64_to_cpu(((__be64 *)rewrite_data)[i]));
+			}
 		}
 
 		seq_puts(file, "\n");
-- 
cgit v1.2.3-70-g09d2


From 72b2cff68405e91ee5e772385b68eb4442bcbf43 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Tue, 8 Nov 2022 17:45:03 +0200
Subject: net/mlx5: DR, Calculate sync threshold of each pool according to its
 type

When certain ICM chunk is no longer needed, it needs to be freed.
Fully freeing ICM memory involves issuing FW SYNC_STEERING command.
This is very time consuming, and it is impractical to do it for every
freed chunk.
Instead, we manage these 'freed' chunks in hot list (list of chunks
that are not required by SW any more, but HW might still access them).
When size of the hot list reaches certain threshold, we purge it and
issue SYNC_STEERING FW command.
There is one threshold for all the different ICM types, which is not
optimal, as different ICM types require different approach: STEs pool
is very large, and it is very 'dynamic' in its nature, so letting hot
list to become too large will result in a significant perf hiccup when
purging the hot list. Modify action is much smaller and less dynamic,
so we can let the hot list to grow to almost the size of the whole pool.

This patch fixes this problem: instead of having same hot memory
threshold for all the pools, sync operation will be triggered in
accordance with the ICM type.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_icm_pool.c      | 33 ++++++++++++----------
 1 file changed, 18 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
index 04fc170a6c16..19e9b4d78454 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
@@ -4,7 +4,9 @@
 #include "dr_types.h"
 
 #define DR_ICM_MODIFY_HDR_ALIGN_BASE 64
-#define DR_ICM_POOL_HOT_MEMORY_FRACTION 4
+#define DR_ICM_POOL_STE_HOT_MEM_PERCENT 25
+#define DR_ICM_POOL_MODIFY_HDR_PTRN_HOT_MEM_PERCENT 50
+#define DR_ICM_POOL_MODIFY_ACTION_HOT_MEM_PERCENT 90
 
 struct mlx5dr_icm_hot_chunk {
 	struct mlx5dr_icm_buddy_mem *buddy_mem;
@@ -29,6 +31,8 @@ struct mlx5dr_icm_pool {
 	struct mlx5dr_icm_hot_chunk *hot_chunks_arr;
 	u32 hot_chunks_num;
 	u64 hot_memory_size;
+	/* hot memory size threshold for triggering sync */
+	u64 th;
 };
 
 struct mlx5dr_icm_dm {
@@ -330,15 +334,7 @@ dr_icm_chunk_init(struct mlx5dr_icm_chunk *chunk,
 
 static bool dr_icm_pool_is_sync_required(struct mlx5dr_icm_pool *pool)
 {
-	int allow_hot_size;
-
-	/* sync when hot memory reaches a certain fraction of the pool size */
-	allow_hot_size =
-		mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
-						   pool->icm_type) /
-		DR_ICM_POOL_HOT_MEMORY_FRACTION;
-
-	return pool->hot_memory_size > allow_hot_size;
+	return pool->hot_memory_size > pool->th;
 }
 
 static void dr_icm_pool_clear_hot_chunks_arr(struct mlx5dr_icm_pool *pool)
@@ -503,8 +499,9 @@ void mlx5dr_icm_pool_free_htbl(struct mlx5dr_icm_pool *pool, struct mlx5dr_ste_h
 struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
 					       enum mlx5dr_icm_type icm_type)
 {
-	u32 num_of_chunks, entry_size, max_hot_size;
+	u32 num_of_chunks, entry_size;
 	struct mlx5dr_icm_pool *pool;
+	u32 max_hot_size = 0;
 
 	pool = kvzalloc(sizeof(*pool), GFP_KERNEL);
 	if (!pool)
@@ -520,12 +517,21 @@ struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
 	switch (icm_type) {
 	case DR_ICM_TYPE_STE:
 		pool->max_log_chunk_sz = dmn->info.max_log_sw_icm_sz;
+		max_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
+								  pool->icm_type) *
+			       DR_ICM_POOL_STE_HOT_MEM_PERCENT / 100;
 		break;
 	case DR_ICM_TYPE_MODIFY_ACTION:
 		pool->max_log_chunk_sz = dmn->info.max_log_action_icm_sz;
+		max_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
+								  pool->icm_type) *
+			       DR_ICM_POOL_MODIFY_ACTION_HOT_MEM_PERCENT / 100;
 		break;
 	case DR_ICM_TYPE_MODIFY_HDR_PTRN:
 		pool->max_log_chunk_sz = dmn->info.max_log_modify_hdr_pattern_icm_sz;
+		max_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
+								  pool->icm_type) *
+			       DR_ICM_POOL_MODIFY_HDR_PTRN_HOT_MEM_PERCENT / 100;
 		break;
 	default:
 		WARN_ON(icm_type);
@@ -533,11 +539,8 @@ struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
 
 	entry_size = mlx5dr_icm_pool_dm_type_to_entry_size(pool->icm_type);
 
-	max_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz,
-							  pool->icm_type) /
-		       DR_ICM_POOL_HOT_MEMORY_FRACTION;
-
 	num_of_chunks = DIV_ROUND_UP(max_hot_size, entry_size) + 1;
+	pool->th = max_hot_size;
 
 	pool->hot_chunks_arr = kvcalloc(num_of_chunks,
 					sizeof(struct mlx5dr_icm_hot_chunk),
-- 
cgit v1.2.3-70-g09d2


From cedb6665bc331054a3477ac29ccac5e399c972da Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Wed, 9 Nov 2022 01:35:24 +0200
Subject: net/mlx5: DR, Add more info in domain dbg dump

Add additinal items to domain info dump: Linux version and device name.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
index ea9f27db4c74..552c7857ca1f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
@@ -4,6 +4,7 @@
 #include <linux/debugfs.h>
 #include <linux/kernel.h>
 #include <linux/seq_file.h>
+#include <linux/version.h>
 #include "dr_types.h"
 
 #define DR_DBG_PTR_TO_ID(p) ((u64)(uintptr_t)(p) & 0xFFFFFFFFULL)
@@ -632,9 +633,15 @@ dr_dump_domain(struct seq_file *file, struct mlx5dr_domain *dmn)
 	u64 domain_id = DR_DBG_PTR_TO_ID(dmn);
 	int ret;
 
-	seq_printf(file, "%d,0x%llx,%d,0%x,%d,%s\n", DR_DUMP_REC_TYPE_DOMAIN,
+	seq_printf(file, "%d,0x%llx,%d,0%x,%d,%u.%u.%u,%s,%d\n",
+		   DR_DUMP_REC_TYPE_DOMAIN,
 		   domain_id, dmn->type, dmn->info.caps.gvmi,
-		   dmn->info.supp_sw_steering, pci_name(dmn->mdev->pdev));
+		   dmn->info.supp_sw_steering,
+		   /* package version */
+		   LINUX_VERSION_MAJOR, LINUX_VERSION_PATCHLEVEL,
+		   LINUX_VERSION_SUBLEVEL,
+		   pci_name(dmn->mdev->pdev),
+		   0); /* domain flags */
 
 	ret = dr_dump_domain_info(file, &dmn->info, domain_id);
 	if (ret < 0)
-- 
cgit v1.2.3-70-g09d2


From 57295e069cd8f7ac79f03d732ac17afb9f790ba5 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Wed, 9 Nov 2022 01:39:32 +0200
Subject: net/mlx5: DR, Add memory statistics for domain object

Add counters for number of buddies that are currently in use per domain
per buddy type (STE, MODIFY-HEADER, MODIFY-PATTERN).

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c      | 7 +++++--
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c | 8 +++++++-
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h    | 3 +++
 3 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
index 552c7857ca1f..7e36e1062139 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c
@@ -633,7 +633,7 @@ dr_dump_domain(struct seq_file *file, struct mlx5dr_domain *dmn)
 	u64 domain_id = DR_DBG_PTR_TO_ID(dmn);
 	int ret;
 
-	seq_printf(file, "%d,0x%llx,%d,0%x,%d,%u.%u.%u,%s,%d\n",
+	seq_printf(file, "%d,0x%llx,%d,0%x,%d,%u.%u.%u,%s,%d,%u,%u,%u\n",
 		   DR_DUMP_REC_TYPE_DOMAIN,
 		   domain_id, dmn->type, dmn->info.caps.gvmi,
 		   dmn->info.supp_sw_steering,
@@ -641,7 +641,10 @@ dr_dump_domain(struct seq_file *file, struct mlx5dr_domain *dmn)
 		   LINUX_VERSION_MAJOR, LINUX_VERSION_PATCHLEVEL,
 		   LINUX_VERSION_SUBLEVEL,
 		   pci_name(dmn->mdev->pdev),
-		   0); /* domain flags */
+		   0, /* domain flags */
+		   dmn->num_buddies[DR_ICM_TYPE_STE],
+		   dmn->num_buddies[DR_ICM_TYPE_MODIFY_ACTION],
+		   dmn->num_buddies[DR_ICM_TYPE_MODIFY_HDR_PTRN]);
 
 	ret = dr_dump_domain_info(file, &dmn->info, domain_id);
 	if (ret < 0)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
index 19e9b4d78454..0b5af9f3f605 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c
@@ -288,6 +288,8 @@ static int dr_icm_buddy_create(struct mlx5dr_icm_pool *pool)
 	/* add it to the -start- of the list in order to search in it first */
 	list_add(&buddy->list_node, &pool->buddy_mem_list);
 
+	pool->dmn->num_buddies[pool->icm_type]++;
+
 	return 0;
 
 err_cleanup_buddy:
@@ -301,13 +303,17 @@ free_mr:
 
 static void dr_icm_buddy_destroy(struct mlx5dr_icm_buddy_mem *buddy)
 {
+	enum mlx5dr_icm_type icm_type = buddy->pool->icm_type;
+
 	dr_icm_pool_mr_destroy(buddy->icm_mr);
 
 	mlx5dr_buddy_cleanup(buddy);
 
-	if (buddy->pool->icm_type == DR_ICM_TYPE_STE)
+	if (icm_type == DR_ICM_TYPE_STE)
 		dr_icm_buddy_cleanup_ste_cache(buddy);
 
+	buddy->pool->dmn->num_buddies[icm_type]--;
+
 	kvfree(buddy);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 37b7b1a79f93..678a993ab053 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -72,6 +72,7 @@ enum mlx5dr_icm_type {
 	DR_ICM_TYPE_STE,
 	DR_ICM_TYPE_MODIFY_ACTION,
 	DR_ICM_TYPE_MODIFY_HDR_PTRN,
+	DR_ICM_TYPE_MAX,
 };
 
 static inline enum mlx5dr_icm_chunk_size
@@ -955,6 +956,8 @@ struct mlx5dr_domain {
 	struct list_head dbg_tbl_list;
 	struct mlx5dr_dbg_dump_info dump_info;
 	struct xarray definers_xa;
+	/* memory management statistics */
+	u32 num_buddies[DR_ICM_TYPE_MAX];
 };
 
 struct mlx5dr_table_rx_tx {
-- 
cgit v1.2.3-70-g09d2


From e267b8a52ca5d5e8434929a5e9f5574aed141024 Mon Sep 17 00:00:00 2001
From: Maher Sanalla <msanalla@nvidia.com>
Date: Mon, 20 Mar 2023 19:43:27 +0200
Subject: Revert "net/mlx5: Expose steering dropped packets counter"

This reverts commit 4fe1b3a5f8fe2fdcedcaba9561e5b0ae5cb1d15b, which
exposes the steering dropped packets counter via debugfs. The upcoming
series will expose the counter via devlink health reporter instead
of debugfs.

Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/esw/debugfs.c  | 22 +++-------------------
 1 file changed, 3 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c
index 3d0bbcca1cb9..2db13c71e88c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c
@@ -12,11 +12,10 @@ enum vnic_diag_counter {
 	MLX5_VNIC_DIAG_CQ_OVERRUN,
 	MLX5_VNIC_DIAG_INVALID_COMMAND,
 	MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND,
-	MLX5_VNIC_DIAG_RX_STEERING_DISCARD,
 };
 
 static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_counter counter,
-				    u64 *val)
+				    u32 *val)
 {
 	u32 out[MLX5_ST_SZ_DW(query_vnic_env_out)] = {};
 	u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {};
@@ -58,10 +57,6 @@ static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_cou
 	case MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND:
 		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, quota_exceeded_command);
 		break;
-	case MLX5_VNIC_DIAG_RX_STEERING_DISCARD:
-		*val = MLX5_GET64(vnic_diagnostic_statistics, vnic_diag_out,
-				  nic_receive_steering_discard);
-		break;
 	}
 
 	return 0;
@@ -70,14 +65,14 @@ static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_cou
 static int __show_vnic_diag(struct seq_file *file, struct mlx5_vport *vport,
 			    enum vnic_diag_counter type)
 {
-	u64 val = 0;
+	u32 val = 0;
 	int ret;
 
 	ret = mlx5_esw_query_vnic_diag(vport, type, &val);
 	if (ret)
 		return ret;
 
-	seq_printf(file, "%llu\n", val);
+	seq_printf(file, "%d\n", val);
 	return 0;
 }
 
@@ -117,11 +112,6 @@ static int quota_exceeded_command_show(struct seq_file *file, void *priv)
 	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND);
 }
 
-static int rx_steering_discard_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_RX_STEERING_DISCARD);
-}
-
 DEFINE_SHOW_ATTRIBUTE(total_q_under_processor_handle);
 DEFINE_SHOW_ATTRIBUTE(send_queue_priority_update_flow);
 DEFINE_SHOW_ATTRIBUTE(comp_eq_overrun);
@@ -129,7 +119,6 @@ DEFINE_SHOW_ATTRIBUTE(async_eq_overrun);
 DEFINE_SHOW_ATTRIBUTE(cq_overrun);
 DEFINE_SHOW_ATTRIBUTE(invalid_command);
 DEFINE_SHOW_ATTRIBUTE(quota_exceeded_command);
-DEFINE_SHOW_ATTRIBUTE(rx_steering_discard);
 
 void mlx5_esw_vport_debugfs_destroy(struct mlx5_eswitch *esw, u16 vport_num)
 {
@@ -190,9 +179,4 @@ void mlx5_esw_vport_debugfs_create(struct mlx5_eswitch *esw, u16 vport_num, bool
 	if (MLX5_CAP_GEN(esw->dev, quota_exceeded_count))
 		debugfs_create_file("quota_exceeded_command", 0444, vnic_diag, vport,
 				    &quota_exceeded_command_fops);
-
-	if (MLX5_CAP_GEN(esw->dev, nic_receive_steering_discard))
-		debugfs_create_file("rx_steering_discard", 0444, vnic_diag, vport,
-				    &rx_steering_discard_fops);
-
 }
-- 
cgit v1.2.3-70-g09d2


From 0a431418f685e100c45ff150efaf4a5afa6f1982 Mon Sep 17 00:00:00 2001
From: Maher Sanalla <msanalla@nvidia.com>
Date: Mon, 20 Mar 2023 19:43:47 +0200
Subject: Revert "net/mlx5: Expose vnic diagnostic counters for eswitch managed
 vports"

This reverts commit 606e6a72e29dff9e3341c4cc9b554420e4793f401 which exposes
the vnic diagnostic counters via debugfs. Instead, The upcoming series will
expose the same counters through devlink health reporter.

Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
 .../net/ethernet/mellanox/mlx5/core/esw/debugfs.c  | 182 ---------------------
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c  |   6 -
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |   5 -
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |   3 -
 5 files changed, 1 insertion(+), 197 deletions(-)
 delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index ca3c66cd47ec..68f6a4544f7e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -69,7 +69,7 @@ mlx5_core-$(CONFIG_MLX5_TC_SAMPLE)   += en/tc/sample.o
 #
 mlx5_core-$(CONFIG_MLX5_ESWITCH)   += eswitch.o eswitch_offloads.o eswitch_offloads_termtbl.o \
 				      ecpf.o rdma.o esw/legacy.o \
-				      esw/debugfs.o esw/devlink_port.o esw/vporttbl.o esw/qos.o
+				      esw/devlink_port.o esw/vporttbl.o esw/qos.o
 
 mlx5_core-$(CONFIG_MLX5_ESWITCH)   += esw/acl/helper.o \
 				      esw/acl/egress_lgcy.o esw/acl/egress_ofld.o \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c
deleted file mode 100644
index 2db13c71e88c..000000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c
+++ /dev/null
@@ -1,182 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
-/* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
-
-#include <linux/debugfs.h>
-#include "eswitch.h"
-
-enum vnic_diag_counter {
-	MLX5_VNIC_DIAG_TOTAL_Q_UNDER_PROCESSOR_HANDLE,
-	MLX5_VNIC_DIAG_SEND_QUEUE_PRIORITY_UPDATE_FLOW,
-	MLX5_VNIC_DIAG_COMP_EQ_OVERRUN,
-	MLX5_VNIC_DIAG_ASYNC_EQ_OVERRUN,
-	MLX5_VNIC_DIAG_CQ_OVERRUN,
-	MLX5_VNIC_DIAG_INVALID_COMMAND,
-	MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND,
-};
-
-static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_counter counter,
-				    u32 *val)
-{
-	u32 out[MLX5_ST_SZ_DW(query_vnic_env_out)] = {};
-	u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {};
-	struct mlx5_core_dev *dev = vport->dev;
-	u16 vport_num = vport->vport;
-	void *vnic_diag_out;
-	int err;
-
-	MLX5_SET(query_vnic_env_in, in, opcode, MLX5_CMD_OP_QUERY_VNIC_ENV);
-	MLX5_SET(query_vnic_env_in, in, vport_number, vport_num);
-	if (!mlx5_esw_is_manager_vport(dev->priv.eswitch, vport_num))
-		MLX5_SET(query_vnic_env_in, in, other_vport, 1);
-
-	err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
-	if (err)
-		return err;
-
-	vnic_diag_out = MLX5_ADDR_OF(query_vnic_env_out, out, vport_env);
-	switch (counter) {
-	case MLX5_VNIC_DIAG_TOTAL_Q_UNDER_PROCESSOR_HANDLE:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, total_error_queues);
-		break;
-	case MLX5_VNIC_DIAG_SEND_QUEUE_PRIORITY_UPDATE_FLOW:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out,
-				send_queue_priority_update_flow);
-		break;
-	case MLX5_VNIC_DIAG_COMP_EQ_OVERRUN:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, comp_eq_overrun);
-		break;
-	case MLX5_VNIC_DIAG_ASYNC_EQ_OVERRUN:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, async_eq_overrun);
-		break;
-	case MLX5_VNIC_DIAG_CQ_OVERRUN:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, cq_overrun);
-		break;
-	case MLX5_VNIC_DIAG_INVALID_COMMAND:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, invalid_command);
-		break;
-	case MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND:
-		*val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, quota_exceeded_command);
-		break;
-	}
-
-	return 0;
-}
-
-static int __show_vnic_diag(struct seq_file *file, struct mlx5_vport *vport,
-			    enum vnic_diag_counter type)
-{
-	u32 val = 0;
-	int ret;
-
-	ret = mlx5_esw_query_vnic_diag(vport, type, &val);
-	if (ret)
-		return ret;
-
-	seq_printf(file, "%d\n", val);
-	return 0;
-}
-
-static int total_q_under_processor_handle_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_TOTAL_Q_UNDER_PROCESSOR_HANDLE);
-}
-
-static int send_queue_priority_update_flow_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private,
-				MLX5_VNIC_DIAG_SEND_QUEUE_PRIORITY_UPDATE_FLOW);
-}
-
-static int comp_eq_overrun_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_COMP_EQ_OVERRUN);
-}
-
-static int async_eq_overrun_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_ASYNC_EQ_OVERRUN);
-}
-
-static int cq_overrun_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_CQ_OVERRUN);
-}
-
-static int invalid_command_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_INVALID_COMMAND);
-}
-
-static int quota_exceeded_command_show(struct seq_file *file, void *priv)
-{
-	return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND);
-}
-
-DEFINE_SHOW_ATTRIBUTE(total_q_under_processor_handle);
-DEFINE_SHOW_ATTRIBUTE(send_queue_priority_update_flow);
-DEFINE_SHOW_ATTRIBUTE(comp_eq_overrun);
-DEFINE_SHOW_ATTRIBUTE(async_eq_overrun);
-DEFINE_SHOW_ATTRIBUTE(cq_overrun);
-DEFINE_SHOW_ATTRIBUTE(invalid_command);
-DEFINE_SHOW_ATTRIBUTE(quota_exceeded_command);
-
-void mlx5_esw_vport_debugfs_destroy(struct mlx5_eswitch *esw, u16 vport_num)
-{
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
-
-	debugfs_remove_recursive(vport->dbgfs);
-	vport->dbgfs = NULL;
-}
-
-/* vnic diag dir name is "pf", "ecpf" or "{vf/sf}_xxxx" */
-#define VNIC_DIAG_DIR_NAME_MAX_LEN 8
-
-void mlx5_esw_vport_debugfs_create(struct mlx5_eswitch *esw, u16 vport_num, bool is_sf, u16 sf_num)
-{
-	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);
-	struct dentry *vnic_diag;
-	char dir_name[VNIC_DIAG_DIR_NAME_MAX_LEN];
-	int err;
-
-	if (!MLX5_CAP_GEN(esw->dev, vport_group_manager))
-		return;
-
-	if (vport_num == MLX5_VPORT_PF) {
-		strcpy(dir_name, "pf");
-	} else if (vport_num == MLX5_VPORT_ECPF) {
-		strcpy(dir_name, "ecpf");
-	} else {
-		err = snprintf(dir_name, VNIC_DIAG_DIR_NAME_MAX_LEN, "%s_%d", is_sf ? "sf" : "vf",
-			       is_sf ? sf_num : vport_num - MLX5_VPORT_FIRST_VF);
-		if (WARN_ON(err < 0))
-			return;
-	}
-
-	vport->dbgfs = debugfs_create_dir(dir_name, esw->dbgfs);
-	vnic_diag = debugfs_create_dir("vnic_diag", vport->dbgfs);
-
-	if (MLX5_CAP_GEN(esw->dev, vnic_env_queue_counters)) {
-		debugfs_create_file("total_q_under_processor_handle", 0444, vnic_diag, vport,
-				    &total_q_under_processor_handle_fops);
-		debugfs_create_file("send_queue_priority_update_flow", 0444, vnic_diag, vport,
-				    &send_queue_priority_update_flow_fops);
-	}
-
-	if (MLX5_CAP_GEN(esw->dev, eq_overrun_count)) {
-		debugfs_create_file("comp_eq_overrun", 0444, vnic_diag, vport,
-				    &comp_eq_overrun_fops);
-		debugfs_create_file("async_eq_overrun", 0444, vnic_diag, vport,
-				    &async_eq_overrun_fops);
-	}
-
-	if (MLX5_CAP_GEN(esw->dev, vnic_env_cq_overrun))
-		debugfs_create_file("cq_overrun", 0444, vnic_diag, vport, &cq_overrun_fops);
-
-	if (MLX5_CAP_GEN(esw->dev, invalid_command_count))
-		debugfs_create_file("invalid_command", 0444, vnic_diag, vport,
-				    &invalid_command_fops);
-
-	if (MLX5_CAP_GEN(esw->dev, quota_exceeded_count))
-		debugfs_create_file("quota_exceeded_command", 0444, vnic_diag, vport,
-				    &quota_exceeded_command_fops);
-}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 8bdf28762f41..8d63f5df7646 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -36,7 +36,6 @@
 #include <linux/mlx5/vport.h>
 #include <linux/mlx5/fs.h>
 #include <linux/mlx5/mpfs.h>
-#include <linux/debugfs.h>
 #include "esw/acl/lgcy.h"
 #include "esw/legacy.h"
 #include "esw/qos.h"
@@ -1056,7 +1055,6 @@ int mlx5_eswitch_load_vport(struct mlx5_eswitch *esw, u16 vport_num,
 	if (err)
 		return err;
 
-	mlx5_esw_vport_debugfs_create(esw, vport_num, false, 0);
 	err = esw_offloads_load_rep(esw, vport_num);
 	if (err)
 		goto err_rep;
@@ -1064,7 +1062,6 @@ int mlx5_eswitch_load_vport(struct mlx5_eswitch *esw, u16 vport_num,
 	return err;
 
 err_rep:
-	mlx5_esw_vport_debugfs_destroy(esw, vport_num);
 	mlx5_esw_vport_disable(esw, vport_num);
 	return err;
 }
@@ -1072,7 +1069,6 @@ err_rep:
 void mlx5_eswitch_unload_vport(struct mlx5_eswitch *esw, u16 vport_num)
 {
 	esw_offloads_unload_rep(esw, vport_num);
-	mlx5_esw_vport_debugfs_destroy(esw, vport_num);
 	mlx5_esw_vport_disable(esw, vport_num);
 }
 
@@ -1672,7 +1668,6 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
 	dev->priv.eswitch = esw;
 	BLOCKING_INIT_NOTIFIER_HEAD(&esw->n_head);
 
-	esw->dbgfs = debugfs_create_dir("esw", mlx5_debugfs_get_dev_root(esw->dev));
 	esw_info(dev,
 		 "Total vports %d, per vport: max uc(%d) max mc(%d)\n",
 		 esw->total_vports,
@@ -1696,7 +1691,6 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
 
 	esw_info(esw->dev, "cleanup\n");
 
-	debugfs_remove_recursive(esw->dbgfs);
 	esw->dev->priv.eswitch = NULL;
 	destroy_workqueue(esw->work_queue);
 	WARN_ON(refcount_read(&esw->qos.refcnt));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index e9d68fdf68f5..f8e25ddc066a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -195,7 +195,6 @@ struct mlx5_vport {
 	enum mlx5_eswitch_vport_event enabled_events;
 	int index;
 	struct devlink_port *dl_port;
-	struct dentry *dbgfs;
 };
 
 struct mlx5_esw_indir_table;
@@ -343,7 +342,6 @@ struct mlx5_eswitch {
 		u32             large_group_num;
 	}  params;
 	struct blocking_notifier_head n_head;
-	struct dentry *dbgfs;
 };
 
 void esw_offloads_disable(struct mlx5_eswitch *esw);
@@ -704,9 +702,6 @@ int mlx5_esw_offloads_devlink_port_register(struct mlx5_eswitch *esw, u16 vport_
 void mlx5_esw_offloads_devlink_port_unregister(struct mlx5_eswitch *esw, u16 vport_num);
 struct devlink_port *mlx5_esw_offloads_devlink_port(struct mlx5_eswitch *esw, u16 vport_num);
 
-void mlx5_esw_vport_debugfs_create(struct mlx5_eswitch *esw, u16 vport_num, bool is_sf, u16 sf_num);
-void mlx5_esw_vport_debugfs_destroy(struct mlx5_eswitch *esw, u16 vport_num);
-
 int mlx5_esw_devlink_sf_port_register(struct mlx5_eswitch *esw, struct devlink_port *dl_port,
 				      u16 vport_num, u32 controller, u32 sfnum);
 void mlx5_esw_devlink_sf_port_unregister(struct mlx5_eswitch *esw, u16 vport_num);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index b6e2709c1371..93ece46a0041 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -3828,14 +3828,12 @@ int mlx5_esw_offloads_sf_vport_enable(struct mlx5_eswitch *esw, struct devlink_p
 	if (err)
 		goto devlink_err;
 
-	mlx5_esw_vport_debugfs_create(esw, vport_num, true, sfnum);
 	err = mlx5_esw_offloads_rep_load(esw, vport_num);
 	if (err)
 		goto rep_err;
 	return 0;
 
 rep_err:
-	mlx5_esw_vport_debugfs_destroy(esw, vport_num);
 	mlx5_esw_devlink_sf_port_unregister(esw, vport_num);
 devlink_err:
 	mlx5_esw_vport_disable(esw, vport_num);
@@ -3845,7 +3843,6 @@ devlink_err:
 void mlx5_esw_offloads_sf_vport_disable(struct mlx5_eswitch *esw, u16 vport_num)
 {
 	mlx5_esw_offloads_rep_unload(esw, vport_num);
-	mlx5_esw_vport_debugfs_destroy(esw, vport_num);
 	mlx5_esw_devlink_sf_port_unregister(esw, vport_num);
 	mlx5_esw_vport_disable(esw, vport_num);
 }
-- 
cgit v1.2.3-70-g09d2


From b0bc615df488abd0e95107e4a9ecefb9bf8c250a Mon Sep 17 00:00:00 2001
From: Maher Sanalla <msanalla@nvidia.com>
Date: Tue, 21 Mar 2023 00:10:16 +0200
Subject: net/mlx5: Add vnic devlink health reporter to PFs/VFs

Create a vnic devlink health reporter for PFs/VFs interfaces.
The reporter's diagnose callback displays the values of vNIC/vport
transport debug counters of PFs/VFs, as follows:

$ devlink health diagnose pci/0000:08:00.0 reporter vnic
 vNIC env counters:
    total_error_queues: 0 send_queue_priority_update_flow: 0
    comp_eq_overrun: 0 async_eq_overrun: 0 cq_overrun: 0
    invalid_command: 0 quota_exceeded_command: 0
    nic_receive_steering_discard: 0

Moreover, add documentation on the reporter functionality and the
counters description.

While at it, expose the vNIC counters diagnose function to be used by
the downstream patch, which will reveal the counters for representor
interfaces.

Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/devlink.rst             |  30 +++++
 drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
 .../mellanox/mlx5/core/diag/reporter_vnic.c        | 125 +++++++++++++++++++++
 .../mellanox/mlx5/core/diag/reporter_vnic.h        |  16 +++
 drivers/net/ethernet/mellanox/mlx5/core/health.c   |   4 +
 include/linux/mlx5/driver.h                        |   1 +
 6 files changed, 177 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
index 0995e4e5acd7..ceab18e46456 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
@@ -257,3 +257,33 @@ User commands examples:
     $ devlink health dump show pci/0000:82:00.1 reporter fw_fatal
 
 NOTE: This command can run only on PF.
+
+vnic reporter
+-------------
+The vnic reporter implements only the `diagnose` callback.
+It is responsible for querying the vnic diagnostic counters from fw and displaying
+them in realtime.
+
+Description of the vnic counters:
+total_q_under_processor_handle: number of queues in an error state due to
+an async error or errored command.
+send_queue_priority_update_flow: number of QP/SQ priority/SL update
+events.
+cq_overrun: number of times CQ entered an error state due to an
+overflow.
+async_eq_overrun: number of times an EQ mapped to async events was
+overrun.
+comp_eq_overrun: number of times an EQ mapped to completion events was
+overrun.
+quota_exceeded_command: number of commands issued and failed due to quota
+exceeded.
+invalid_command: number of commands issued and failed dues to any reason
+other than quota exceeded.
+nic_receive_steering_discard: number of packets that completed RX flow
+steering but were discarded due to a mismatch in flow table.
+
+User commands examples:
+- Diagnose PF/VF vnic counters
+        $ devlink health diagnose pci/0000:82:00.1 reporter vnic
+
+NOTE: This command can run only on PF/VF ports.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 68f6a4544f7e..ddf1e352f51d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -16,7 +16,7 @@ mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 		transobj.o vport.o sriov.o fs_cmd.o fs_core.o pci_irq.o \
 		fs_counters.o fs_ft_pool.o rl.o lag/debugfs.o lag/lag.o dev.o events.o wq.o lib/gid.o \
 		lib/devcom.o lib/pci_vsc.o lib/dm.o lib/fs_ttc.o diag/fs_tracepoint.o \
-		diag/fw_tracer.o diag/crdump.o devlink.o diag/rsc_dump.o \
+		diag/fw_tracer.o diag/crdump.o devlink.o diag/rsc_dump.o diag/reporter_vnic.o \
 		fw_reset.o qos.o lib/tout.o lib/aso.o
 
 #
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c
new file mode 100644
index 000000000000..9114661cd967
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.c
@@ -0,0 +1,125 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. */
+
+#include "reporter_vnic.h"
+#include "devlink.h"
+
+#define VNIC_ENV_GET64(vnic_env_stats, c) \
+	MLX5_GET64(query_vnic_env_out, (vnic_env_stats)->query_vnic_env_out, \
+		 vport_env.c)
+
+struct mlx5_vnic_diag_stats {
+	__be64 query_vnic_env_out[MLX5_ST_SZ_QW(query_vnic_env_out)];
+};
+
+int mlx5_reporter_vnic_diagnose_counters(struct mlx5_core_dev *dev,
+					 struct devlink_fmsg *fmsg,
+					 u16 vport_num, bool other_vport)
+{
+	u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {};
+	struct mlx5_vnic_diag_stats vnic;
+	int err;
+
+	MLX5_SET(query_vnic_env_in, in, opcode, MLX5_CMD_OP_QUERY_VNIC_ENV);
+	MLX5_SET(query_vnic_env_in, in, vport_number, vport_num);
+	MLX5_SET(query_vnic_env_in, in, other_vport, !!other_vport);
+
+	err = mlx5_cmd_exec_inout(dev, query_vnic_env, in, &vnic.query_vnic_env_out);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_pair_nest_start(fmsg, "vNIC env counters");
+	if (err)
+		return err;
+
+	err = devlink_fmsg_obj_nest_start(fmsg);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "total_error_queues",
+					VNIC_ENV_GET64(&vnic, total_error_queues));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "send_queue_priority_update_flow",
+					VNIC_ENV_GET64(&vnic, send_queue_priority_update_flow));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "comp_eq_overrun",
+					VNIC_ENV_GET64(&vnic, comp_eq_overrun));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "async_eq_overrun",
+					VNIC_ENV_GET64(&vnic, async_eq_overrun));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "cq_overrun",
+					VNIC_ENV_GET64(&vnic, cq_overrun));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "invalid_command",
+					VNIC_ENV_GET64(&vnic, invalid_command));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "quota_exceeded_command",
+					VNIC_ENV_GET64(&vnic, quota_exceeded_command));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "nic_receive_steering_discard",
+					VNIC_ENV_GET64(&vnic, nic_receive_steering_discard));
+	if (err)
+		return err;
+
+	err = devlink_fmsg_obj_nest_end(fmsg);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_pair_nest_end(fmsg);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int mlx5_reporter_vnic_diagnose(struct devlink_health_reporter *reporter,
+				       struct devlink_fmsg *fmsg,
+				       struct netlink_ext_ack *extack)
+{
+	struct mlx5_core_dev *dev = devlink_health_reporter_priv(reporter);
+
+	return mlx5_reporter_vnic_diagnose_counters(dev, fmsg, 0, false);
+}
+
+static const struct devlink_health_reporter_ops mlx5_reporter_vnic_ops = {
+	.name = "vnic",
+	.diagnose = mlx5_reporter_vnic_diagnose,
+};
+
+void mlx5_reporter_vnic_create(struct mlx5_core_dev *dev)
+{
+	struct mlx5_core_health *health = &dev->priv.health;
+	struct devlink *devlink = priv_to_devlink(dev);
+
+	health->vnic_reporter =
+		devlink_health_reporter_create(devlink,
+					       &mlx5_reporter_vnic_ops,
+					       0, dev);
+	if (IS_ERR(health->vnic_reporter))
+		mlx5_core_warn(dev,
+			       "Failed to create vnic reporter, err = %ld\n",
+			       PTR_ERR(health->vnic_reporter));
+}
+
+void mlx5_reporter_vnic_destroy(struct mlx5_core_dev *dev)
+{
+	struct mlx5_core_health *health = &dev->priv.health;
+
+	if (!IS_ERR_OR_NULL(health->vnic_reporter))
+		devlink_health_reporter_destroy(health->vnic_reporter);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h
new file mode 100644
index 000000000000..eba87a39e9b1
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/reporter_vnic.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+ * Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES.
+ */
+#ifndef __MLX5_REPORTER_VNIC_H
+#define __MLX5_REPORTER_VNIC_H
+
+#include "mlx5_core.h"
+
+void mlx5_reporter_vnic_create(struct mlx5_core_dev *dev);
+void mlx5_reporter_vnic_destroy(struct mlx5_core_dev *dev);
+
+int mlx5_reporter_vnic_diagnose_counters(struct mlx5_core_dev *dev,
+					 struct devlink_fmsg *fmsg,
+					 u16 vport_num, bool other_vport);
+
+#endif /* __MLX5_REPORTER_VNIC_H */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index 016c5f99c470..871c32dda66e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -42,6 +42,7 @@
 #include "lib/pci_vsc.h"
 #include "lib/tout.h"
 #include "diag/fw_tracer.h"
+#include "diag/reporter_vnic.h"
 
 enum {
 	MAX_MISSES			= 3,
@@ -898,6 +899,7 @@ void mlx5_health_cleanup(struct mlx5_core_dev *dev)
 
 	cancel_delayed_work_sync(&health->update_fw_log_ts_work);
 	destroy_workqueue(health->wq);
+	mlx5_reporter_vnic_destroy(dev);
 	mlx5_fw_reporters_destroy(dev);
 }
 
@@ -907,6 +909,7 @@ int mlx5_health_init(struct mlx5_core_dev *dev)
 	char *name;
 
 	mlx5_fw_reporters_create(dev);
+	mlx5_reporter_vnic_create(dev);
 
 	health = &dev->priv.health;
 	name = kmalloc(64, GFP_KERNEL);
@@ -926,6 +929,7 @@ int mlx5_health_init(struct mlx5_core_dev *dev)
 	return 0;
 
 out_err:
+	mlx5_reporter_vnic_destroy(dev);
 	mlx5_fw_reporters_destroy(dev);
 	return -ENOMEM;
 }
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 135a3c8d8237..5d25c4c73046 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -439,6 +439,7 @@ struct mlx5_core_health {
 	struct work_struct		report_work;
 	struct devlink_health_reporter *fw_reporter;
 	struct devlink_health_reporter *fw_fatal_reporter;
+	struct devlink_health_reporter *vnic_reporter;
 	struct delayed_work		update_fw_log_ts_work;
 };
 
-- 
cgit v1.2.3-70-g09d2


From cf14af140a5ad0937d385ce693100f33f02e9c54 Mon Sep 17 00:00:00 2001
From: Maher Sanalla <msanalla@nvidia.com>
Date: Tue, 21 Mar 2023 13:33:00 +0200
Subject: net/mlx5e: Add vnic devlink health reporter to representors

Create a new devlink health reporter for representor interface, which
reports the values of representor vnic diagnostic counters when diagnosed.

This patch will allow admins to monitor VF diagnostic counters through
the representor-interface vnic reporter.

Example of usage:
$ devlink health diagnose pci/0000:08:00.0/65537 reporter vnic
  vNIC env counters:
    total_error_queues: 0 send_queue_priority_update_flow: 0
    comp_eq_overrun: 0 async_eq_overrun: 0 cq_overrun: 0
    invalid_command: 0 quota_exceeded_command: 0
    nic_receive_steering_discard: 0

Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/devlink.rst             |  5 ++-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   | 52 +++++++++++++++++++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.h   |  1 +
 3 files changed, 55 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
index ceab18e46456..3a7a714cc08f 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst
@@ -285,5 +285,8 @@ steering but were discarded due to a mismatch in flow table.
 User commands examples:
 - Diagnose PF/VF vnic counters
         $ devlink health diagnose pci/0000:82:00.1 reporter vnic
+- Diagnose representor vnic counters (performed by supplying devlink port of the
+  representor, which can be obtained via devlink port command)
+        $ devlink health diagnose pci/0000:82:00.1/65537 reporter vnic
 
-NOTE: This command can run only on PF/VF ports.
+NOTE: This command can run over all interfaces such as PF/VF and representor ports.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 8ff654b4e9e1..2d87068f63fd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -53,6 +53,7 @@
 #include "lib/vxlan.h"
 #define CREATE_TRACE_POINTS
 #include "diag/en_rep_tracepoint.h"
+#include "diag/reporter_vnic.h"
 #include "en_accel/ipsec.h"
 #include "en/tc/int_port.h"
 #include "en/ptp.h"
@@ -1294,6 +1295,50 @@ static unsigned int mlx5e_ul_rep_stats_grps_num(struct mlx5e_priv *priv)
 	return ARRAY_SIZE(mlx5e_ul_rep_stats_grps);
 }
 
+static int
+mlx5e_rep_vnic_reporter_diagnose(struct devlink_health_reporter *reporter,
+				 struct devlink_fmsg *fmsg,
+				 struct netlink_ext_ack *extack)
+{
+	struct mlx5e_rep_priv *rpriv = devlink_health_reporter_priv(reporter);
+	struct mlx5_eswitch_rep *rep = rpriv->rep;
+
+	return mlx5_reporter_vnic_diagnose_counters(rep->esw->dev, fmsg,
+						    rep->vport, true);
+}
+
+static const struct devlink_health_reporter_ops mlx5_rep_vnic_reporter_ops = {
+	.name = "vnic",
+	.diagnose = mlx5e_rep_vnic_reporter_diagnose,
+};
+
+static void mlx5e_rep_vnic_reporter_create(struct mlx5e_priv *priv,
+					   struct devlink_port *dl_port)
+{
+	struct mlx5e_rep_priv *rpriv = priv->ppriv;
+	struct devlink_health_reporter *reporter;
+
+	reporter = devl_port_health_reporter_create(dl_port,
+						    &mlx5_rep_vnic_reporter_ops,
+						    0, rpriv);
+	if (IS_ERR(reporter)) {
+		mlx5_core_err(priv->mdev,
+			      "Failed to create representor vnic reporter, err = %ld\n",
+			      PTR_ERR(reporter));
+		return;
+	}
+
+	rpriv->rep_vnic_reporter = reporter;
+}
+
+static void mlx5e_rep_vnic_reporter_destroy(struct mlx5e_priv *priv)
+{
+	struct mlx5e_rep_priv *rpriv = priv->ppriv;
+
+	if (!IS_ERR_OR_NULL(rpriv->rep_vnic_reporter))
+		devl_health_reporter_destroy(rpriv->rep_vnic_reporter);
+}
+
 static const struct mlx5e_profile mlx5e_rep_profile = {
 	.init			= mlx5e_init_rep,
 	.cleanup		= mlx5e_cleanup_rep,
@@ -1394,8 +1439,10 @@ mlx5e_vport_vf_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
 
 	dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch,
 						 rpriv->rep->vport);
-	if (dl_port)
+	if (dl_port) {
 		SET_NETDEV_DEVLINK_PORT(netdev, dl_port);
+		mlx5e_rep_vnic_reporter_create(priv, dl_port);
+	}
 
 	err = register_netdev(netdev);
 	if (err) {
@@ -1408,8 +1455,8 @@ mlx5e_vport_vf_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
 	return 0;
 
 err_detach_netdev:
+	mlx5e_rep_vnic_reporter_destroy(priv);
 	mlx5e_detach_netdev(netdev_priv(netdev));
-
 err_cleanup_profile:
 	priv->profile->cleanup(priv);
 
@@ -1458,6 +1505,7 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep)
 	}
 
 	unregister_netdev(netdev);
+	mlx5e_rep_vnic_reporter_destroy(priv);
 	mlx5e_detach_netdev(priv);
 	priv->profile->cleanup(priv);
 	mlx5e_destroy_netdev(priv);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
index dcfad0bf0f45..80b7f5079a5a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
@@ -118,6 +118,7 @@ struct mlx5e_rep_priv {
 	struct rtnl_link_stats64 prev_vf_vport_stats;
 	struct mlx5_flow_handle *send_to_vport_meta_rule;
 	struct rhashtable tc_ht;
+	struct devlink_health_reporter *rep_vnic_reporter;
 };
 
 static inline
-- 
cgit v1.2.3-70-g09d2


From c8e9090233a70b8bbd8a73521a7a81856695732f Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Thu, 30 Mar 2023 19:37:00 +0300
Subject: net/mlx5e: RX, Fix releasing page_pool pages twice for striding RQ

mlx5e_free_rx_descs is responsible for calling the dealloc_wqe op which
returns pages to the page_pool. This can happen during flush or close.
For XSK, the regular RQ is flushed (when replaced by the XSK RQ) and
also closed later. This is normally not a problem as the wqe list is
empty on a second call to mlx5e_free_rx_descs. However, for striding RQ,
the previously released wqes from the list will appear as missing
and will be released a second time by mlx5e_free_rx_missing_descs.

This patch sets the no release bits on the striding RQ wqes in the
dealloc_wqe op to prevent releasing the pages a second time.

Please note that the bits are set only in the control path during
close and not in the data path.

Fixes: 4c2a13236807 ("net/mlx5e: RX, Defer page release in striding rq for better recycling")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index a8c2ae389d6c..5dc907541094 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -861,6 +861,11 @@ static void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, ix);
 	/* This function is called on rq/netdev close. */
 	mlx5e_free_rx_mpwqe(rq, wi);
+
+	/* Avoid a second release of the wqe pages: dealloc is called also
+	 * for missing wqes on an already flushed RQ.
+	 */
+	bitmap_fill(wi->skip_release_bitmap, rq->mpwqe.pages_per_wqe);
 }
 
 INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq)
-- 
cgit v1.2.3-70-g09d2


From 40afb3b14496afb01d5b3d028444e09d29b95559 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Mon, 3 Apr 2023 20:03:11 +0300
Subject: net/mlx5e: RX, Fix XDP_TX page release for legacy rq nonlinear case

When the XDP handler marks the data for transmission (XDP_TX),
it is incorrect to release the page fragment. Instead, the
fragments should be marked as MLX5E_WQE_FRAG_SKIP_RELEASE
because XDP will release the page directly to the page_pool
(page_pool_put_defragged_page) after TX completion.

The linear case already does this. This patch fixes the
nonlinear part as well.

Also, the looping over the fragments was incorrect: When handling
pages after XDP_TX in the legacy rq nonlinear case, the loop was
skipping the first wqe fragment.

Fixes: 3f93f82988bc ("net/mlx5e: RX, Defer page release in legacy rq for better recycling")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 5dc907541094..69634829558e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1746,10 +1746,10 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi
 	prog = rcu_dereference(rq->xdp_prog);
 	if (prog && mlx5e_xdp_handle(rq, prog, &mxbuf)) {
 		if (test_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) {
-			int i;
+			struct mlx5e_wqe_frag_info *pwi;
 
-			for (i = wi - head_wi; i < rq->wqe.info.num_frags; i++)
-				mlx5e_put_rx_frag(rq, &head_wi[i]);
+			for (pwi = head_wi; pwi < wi; pwi++)
+				pwi->flags |= BIT(MLX5E_WQE_FRAG_SKIP_RELEASE);
 		}
 		return NULL; /* page/packet was consumed by XDP */
 	}
-- 
cgit v1.2.3-70-g09d2


From a880f814739c84fa11e17ac87c91fb711c185610 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Thu, 13 Apr 2023 17:14:05 +0300
Subject: net/mlx5e: RX, Hook NAPIs to page pools

Linking the NAPI to the rq page_pool to improve page_pool cache
usage during skb recycling.

Here are the observed improvements for a iperf single stream
test case:

- For 1500 MTU and legacy rq, seeing a 20% improvement of cache usage.

- For 9K MTU, seeing 33-40 % page_pool cache usage improvements for
both striding and legacy rq (depending if the application is running on
the same core as the rq or not).

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 7eb1eeb115ca..f5504b699fcf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -857,6 +857,7 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
 		pp_params.pool_size = pool_size;
 		pp_params.nid       = node;
 		pp_params.dev       = rq->pdev;
+		pp_params.napi      = rq->cq.napi;
 		pp_params.dma_dir   = rq->buff.map_dir;
 		pp_params.max_len   = PAGE_SIZE;
 
-- 
cgit v1.2.3-70-g09d2


From 45e261b7b821139556d039b69c9101207c152a50 Mon Sep 17 00:00:00 2001
From: Eli Cohen <elic@nvidia.com>
Date: Sun, 2 Apr 2023 15:17:09 +0300
Subject: net/mlx5: Include linux/pci.h for pci_msix_can_alloc_dyn()

Add include directive to assure pci_msix_can_alloc_dyn() prototype.

Fixes: 3354822cde5a ("net/mlx5: Use dynamic msix vectors allocation")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202303291328.sNmTyyWF-lkp@intel.com/
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index e12e528c09f5..2245d3b2f393 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 /* Copyright (c) 2019 Mellanox Technologies. */
 
+#include <linux/pci.h>
 #include <linux/interrupt.h>
 #include <linux/notifier.h>
 #include <linux/mlx5/driver.h>
-- 
cgit v1.2.3-70-g09d2


From 8ca52ada6267fb67d31a5bf7d568a54d793c759e Mon Sep 17 00:00:00 2001
From: Roi Dayan <roid@nvidia.com>
Date: Tue, 21 Mar 2023 10:32:03 +0200
Subject: net/mlx5: E-Switch, Remove redundant dev arg from
 mlx5_esw_vport_alloc()

The passded esw->dev is redundant as esw being passed and esw->dev being
used inside.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 8d63f5df7646..3f25fa2893fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1506,7 +1506,7 @@ out_free:
 	return err;
 }
 
-static int mlx5_esw_vport_alloc(struct mlx5_eswitch *esw, struct mlx5_core_dev *dev,
+static int mlx5_esw_vport_alloc(struct mlx5_eswitch *esw,
 				int index, u16 vport_num)
 {
 	struct mlx5_vport *vport;
@@ -1560,7 +1560,7 @@ static int mlx5_esw_vports_init(struct mlx5_eswitch *esw)
 
 	xa_init(&esw->vports);
 
-	err = mlx5_esw_vport_alloc(esw, dev, idx, MLX5_VPORT_PF);
+	err = mlx5_esw_vport_alloc(esw, idx, MLX5_VPORT_PF);
 	if (err)
 		goto err;
 	if (esw->first_host_vport == MLX5_VPORT_PF)
@@ -1568,7 +1568,7 @@ static int mlx5_esw_vports_init(struct mlx5_eswitch *esw)
 	idx++;
 
 	for (i = 0; i < mlx5_core_max_vfs(dev); i++) {
-		err = mlx5_esw_vport_alloc(esw, dev, idx, idx);
+		err = mlx5_esw_vport_alloc(esw, idx, idx);
 		if (err)
 			goto err;
 		xa_set_mark(&esw->vports, idx, MLX5_ESW_VPT_VF);
@@ -1577,7 +1577,7 @@ static int mlx5_esw_vports_init(struct mlx5_eswitch *esw)
 	}
 	base_sf_num = mlx5_sf_start_function_id(dev);
 	for (i = 0; i < mlx5_sf_max_functions(dev); i++) {
-		err = mlx5_esw_vport_alloc(esw, dev, idx, base_sf_num + i);
+		err = mlx5_esw_vport_alloc(esw, idx, base_sf_num + i);
 		if (err)
 			goto err;
 		xa_set_mark(&esw->vports, base_sf_num + i, MLX5_ESW_VPT_SF);
@@ -1588,7 +1588,7 @@ static int mlx5_esw_vports_init(struct mlx5_eswitch *esw)
 	if (err)
 		goto err;
 	for (i = 0; i < max_host_pf_sfs; i++) {
-		err = mlx5_esw_vport_alloc(esw, dev, idx, base_sf_num + i);
+		err = mlx5_esw_vport_alloc(esw, idx, base_sf_num + i);
 		if (err)
 			goto err;
 		xa_set_mark(&esw->vports, base_sf_num + i, MLX5_ESW_VPT_SF);
@@ -1596,12 +1596,12 @@ static int mlx5_esw_vports_init(struct mlx5_eswitch *esw)
 	}
 
 	if (mlx5_ecpf_vport_exists(dev)) {
-		err = mlx5_esw_vport_alloc(esw, dev, idx, MLX5_VPORT_ECPF);
+		err = mlx5_esw_vport_alloc(esw, idx, MLX5_VPORT_ECPF);
 		if (err)
 			goto err;
 		idx++;
 	}
-	err = mlx5_esw_vport_alloc(esw, dev, idx, MLX5_VPORT_UPLINK);
+	err = mlx5_esw_vport_alloc(esw, idx, MLX5_VPORT_UPLINK);
 	if (err)
 		goto err;
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 38d9a740f68d8bbe92029cdd9eb6259e3974e52a Mon Sep 17 00:00:00 2001
From: Roi Dayan <roid@nvidia.com>
Date: Tue, 21 Mar 2023 15:40:02 +0200
Subject: net/mlx5: E-Switch, Remove unused
 mlx5_esw_offloads_vport_metadata_set()

Remove unused function which also seems a duplicate
of esw_port_metadata_set().

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |  1 -
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 22 ----------------------
 2 files changed, 23 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index f8e25ddc066a..62f01d4600fe 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -354,7 +354,6 @@ mlx5_eswitch_add_send_to_vport_meta_rule(struct mlx5_eswitch *esw, u16 vport_num
 void mlx5_eswitch_del_send_to_vport_meta_rule(struct mlx5_flow_handle *rule);
 
 bool mlx5_esw_vport_match_metadata_supported(const struct mlx5_eswitch *esw);
-int mlx5_esw_offloads_vport_metadata_set(struct mlx5_eswitch *esw, bool enable);
 u32 mlx5_esw_match_metadata_alloc(struct mlx5_eswitch *esw);
 void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 93ece46a0041..12c07a44aa4f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -2939,28 +2939,6 @@ metadata_err:
 	return err;
 }
 
-int mlx5_esw_offloads_vport_metadata_set(struct mlx5_eswitch *esw, bool enable)
-{
-	int err = 0;
-
-	down_write(&esw->mode_lock);
-	if (mlx5_esw_is_fdb_created(esw)) {
-		err = -EBUSY;
-		goto done;
-	}
-	if (!mlx5_esw_vport_match_metadata_supported(esw)) {
-		err = -EOPNOTSUPP;
-		goto done;
-	}
-	if (enable)
-		esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA;
-	else
-		esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA;
-done:
-	up_write(&esw->mode_lock);
-	return err;
-}
-
 int
 esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw,
 				     struct mlx5_vport *vport)
-- 
cgit v1.2.3-70-g09d2


From f9c895a72a390656f9582e048fdcc3d2cec1dd7c Mon Sep 17 00:00:00 2001
From: Roi Dayan <roid@nvidia.com>
Date: Wed, 22 Mar 2023 10:21:46 +0200
Subject: net/mlx5: Update op_mode to op_mod for port selection

To be consistent with the other enum keys use OP_MOD
instead of OP_MODE.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 +-
 include/linux/mlx5/mlx5_ifc.h                  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index a95d1218def9..89a65779494e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -717,7 +717,7 @@ static int handle_hca_cap_port_selection(struct mlx5_core_dev *dev,
 	       MLX5_ST_SZ_BYTES(port_selection_cap));
 	MLX5_SET(port_selection_cap, set_hca_cap, port_select_flow_table_bypass, 1);
 
-	err = set_caps(dev, set_ctx, MLX5_SET_HCA_CAP_OP_MODE_PORT_SELECTION);
+	err = set_caps(dev, set_ctx, MLX5_SET_HCA_CAP_OP_MOD_PORT_SELECTION);
 
 	return err;
 }
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 20d00e09b168..b42696d74c9f 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -69,7 +69,7 @@ enum {
 	MLX5_SET_HCA_CAP_OP_MOD_ATOMIC                = 0x3,
 	MLX5_SET_HCA_CAP_OP_MOD_ROCE                  = 0x4,
 	MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE2       = 0x20,
-	MLX5_SET_HCA_CAP_OP_MODE_PORT_SELECTION       = 0x25,
+	MLX5_SET_HCA_CAP_OP_MOD_PORT_SELECTION        = 0x25,
 };
 
 enum {
-- 
cgit v1.2.3-70-g09d2


From e9fce818fe003b6c527f25517b9ac08eb4661b5d Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Wed, 22 Mar 2023 09:52:26 +0100
Subject: net/mlx5e: Don't clone flow post action attributes second time

The code already clones post action attributes in
mlx5e_clone_flow_attr_for_post_act(). Creating another copy in
mlx5e_tc_post_act_add() is a erroneous leftover from original
implementation. Instead, assign handle->attribute to post_attr provided by
the caller. Note that cloning the attribute second time is not just
wasteful but also causes issues like second copy not being properly updated
in neigh update code which leads to following use-after-free:

Feb 21 09:02:00 c-237-177-40-045 kernel: BUG: KASAN: use-after-free in mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_report+0xbb/0x1a0
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel:  __kasan_kmalloc+0x7a/0x90
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_save_free_info+0x2a/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel:  ____kasan_slab_free+0x11a/0x1b0
Feb 21 09:02:00 c-237-177-40-045 kernel: page dumped because: kasan: bad access detected
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_core 0000:08:00.0: mlx5_cmd_out_err:803:(pid 8833): SET_FLOW_TABLE_ENTRY(0x936) op_mod(0x0) failed, status bad resource state(0x9), syndrome (0xf2ff71), err(-22)
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_core 0000:08:00.0 enp8s0f0: Failed to add post action rule
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_core 0000:08:00.0: mlx5e_tc_encap_flows_add:190:(pid 8833): Failed to update flow post acts, -22
Feb 21 09:02:00 c-237-177-40-045 kernel: Call Trace:
Feb 21 09:02:00 c-237-177-40-045 kernel:  <TASK>
Feb 21 09:02:00 c-237-177-40-045 kernel:  dump_stack_lvl+0x57/0x7d
Feb 21 09:02:00 c-237-177-40-045 kernel:  print_report+0x170/0x471
Feb 21 09:02:00 c-237-177-40-045 kernel:  ? mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_report+0xbb/0x1a0
Feb 21 09:02:00 c-237-177-40-045 kernel:  ? mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  ? __module_address.part.0+0x62/0x200
Feb 21 09:02:00 c-237-177-40-045 kernel:  ? mlx5_cmd_stub_create_flow_table+0xd0/0xd0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  ? __raw_spin_lock_init+0x3b/0x110
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5_cmd_create_fte+0x80/0xb0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  add_rule_fg+0xe80/0x19c0 [mlx5_core]
--
Feb 21 09:02:00 c-237-177-40-045 kernel: Allocated by task 13476:
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel:  __kasan_kmalloc+0x7a/0x90
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5_packet_reformat_alloc+0x7b/0x230 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_tc_tun_create_header_ipv4+0x977/0xf10 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_attach_encap+0x15b4/0x1e10 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  post_process_attr+0x305/0xa30 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_tc_add_fdb_flow+0x4c0/0xcf0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  __mlx5e_add_fdb_flow+0x7cf/0xe90 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_configure_flower+0xcaa/0x4b90 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_rep_setup_tc_cls_flower+0x99/0x1b0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_rep_setup_tc_cb+0x133/0x1e0 [mlx5_core]
--
Feb 21 09:02:00 c-237-177-40-045 kernel: Freed by task 8833:
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel:  kasan_save_free_info+0x2a/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel:  ____kasan_slab_free+0x11a/0x1b0
Feb 21 09:02:00 c-237-177-40-045 kernel:  __kmem_cache_free+0x1de/0x400
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5_packet_reformat_dealloc+0xad/0x100 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_tc_encap_flows_del+0x3c0/0x500 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_rep_update_flows+0x40c/0xa80 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  mlx5e_rep_neigh_update+0x473/0x7a0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel:  process_one_work+0x7c2/0x1310
Feb 21 09:02:00 c-237-177-40-045 kernel:  worker_thread+0x59d/0xec0
Feb 21 09:02:00 c-237-177-40-045 kernel:  kthread+0x28f/0x330

Fixes: 8300f225268b ("net/mlx5e: Create new flow attr for multi table actions")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c | 11 ++---------
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.h |  2 +-
 2 files changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c
index 4e48946c4c2a..0290e0dea539 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.c
@@ -106,22 +106,17 @@ err_rule:
 }
 
 struct mlx5e_post_act_handle *
-mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *attr)
+mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *post_attr)
 {
-	u32 attr_sz = ns_to_attr_sz(post_act->ns_type);
 	struct mlx5e_post_act_handle *handle;
-	struct mlx5_flow_attr *post_attr;
 	int err;
 
 	handle = kzalloc(sizeof(*handle), GFP_KERNEL);
-	post_attr = mlx5_alloc_flow_attr(post_act->ns_type);
-	if (!handle || !post_attr) {
-		kfree(post_attr);
+	if (!handle) {
 		kfree(handle);
 		return ERR_PTR(-ENOMEM);
 	}
 
-	memcpy(post_attr, attr, attr_sz);
 	post_attr->chain = 0;
 	post_attr->prio = 0;
 	post_attr->ft = post_act->ft;
@@ -145,7 +140,6 @@ mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *at
 	return handle;
 
 err_xarray:
-	kfree(post_attr);
 	kfree(handle);
 	return ERR_PTR(err);
 }
@@ -164,7 +158,6 @@ mlx5e_tc_post_act_del(struct mlx5e_post_act *post_act, struct mlx5e_post_act_han
 	if (!IS_ERR_OR_NULL(handle->rule))
 		mlx5e_tc_post_act_unoffload(post_act, handle);
 	xa_erase(&post_act->ids, handle->id);
-	kfree(handle->attr);
 	kfree(handle);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.h
index f476774c0b75..40b8df184af5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_act.h
@@ -19,7 +19,7 @@ void
 mlx5e_tc_post_act_destroy(struct mlx5e_post_act *post_act);
 
 struct mlx5e_post_act_handle *
-mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *attr);
+mlx5e_tc_post_act_add(struct mlx5e_post_act *post_act, struct mlx5_flow_attr *post_attr);
 
 void
 mlx5e_tc_post_act_del(struct mlx5e_post_act *post_act, struct mlx5e_post_act_handle *handle);
-- 
cgit v1.2.3-70-g09d2


From 8ac04a28144cfa89b61be518268233742c23d88d Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Fri, 24 Mar 2023 17:52:17 +0100
Subject: net/mlx5e: Release the label when replacing existing ct entry

Cited commit doesn't release the label mapping when replacing existing ct
entry which leads to following memleak report:

unreferenced object 0xffff8881854cf280 (size 96):
  comm "kworker/u48:74", pid 23093, jiffies 4296664564 (age 175.944s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<000000002722d368>] __kmalloc+0x4b/0x1c0
    [<00000000cc44e18f>] mapping_add+0x6e8/0xc90 [mlx5_core]
    [<000000003ad942a7>] mlx5_get_label_mapping+0x66/0xe0 [mlx5_core]
    [<00000000266308ac>] mlx5_tc_ct_entry_create_mod_hdr+0x1c4/0xf50 [mlx5_core]
    [<000000009a768b4f>] mlx5_tc_ct_entry_add_rule+0x16f/0xaf0 [mlx5_core]
    [<00000000a178f3e5>] mlx5_tc_ct_block_flow_offload_add+0x10cb/0x1f90 [mlx5_core]
    [<000000007b46c496>] mlx5_tc_ct_block_flow_offload+0x14a/0x630 [mlx5_core]
    [<00000000a9a18ac5>] nf_flow_offload_tuple+0x1a3/0x390 [nf_flow_table]
    [<00000000d0881951>] flow_offload_work_handler+0x257/0xd30 [nf_flow_table]
    [<000000009e4935a4>] process_one_work+0x7c2/0x13e0
    [<00000000f5cd36a7>] worker_thread+0x59d/0xec0
    [<00000000baed1daf>] kthread+0x28f/0x330
    [<0000000063d282a4>] ret_from_fork+0x1f/0x30

Fix the issue by correctly releasing the label mapping.

Fixes: 94ceffb48eac ("net/mlx5e: Implement CT entry update")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Paul Blakey <paulb@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index 314983bc6f08..ee49bd2461e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -920,6 +920,7 @@ mlx5_tc_ct_entry_replace_rule(struct mlx5_tc_ct_priv *ct_priv,
 	zone_rule->rule = rule;
 	mlx5_tc_ct_entry_destroy_mod_hdr(ct_priv, old_attr, zone_rule->mh);
 	zone_rule->mh = mh;
+	mlx5_put_label_mapping(ct_priv, old_attr->ct_attr.ct_labels_id);
 
 	kfree(old_attr);
 	kvfree(spec);
-- 
cgit v1.2.3-70-g09d2


From fd745f4c0abe41ebb09d11bf622b054a0f3e7b49 Mon Sep 17 00:00:00 2001
From: Chris Mi <cmi@nvidia.com>
Date: Fri, 10 Mar 2023 11:06:48 +0200
Subject: net/mlx5: E-switch, Create per vport table based on devlink encap
 mode

Currently when creating per vport table, create flags are hardcoded.
Devlink encap mode is set based on user input and HW capability.
Create per vport table based on devlink encap mode.

Fixes: c796bb7cd230 ("net/mlx5: E-switch, Generalize per vport table API")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.c     |  4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/esw/vporttbl.c     | 12 +++++++++++-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h          |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c |  2 +-
 4 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.c
index 558a776359af..5db239cae814 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/sample.c
@@ -14,10 +14,10 @@
 
 #define MLX5_ESW_VPORT_TBL_SIZE_SAMPLE (64 * 1024)
 
-static const struct esw_vport_tbl_namespace mlx5_esw_vport_tbl_sample_ns = {
+static struct esw_vport_tbl_namespace mlx5_esw_vport_tbl_sample_ns = {
 	.max_fte = MLX5_ESW_VPORT_TBL_SIZE_SAMPLE,
 	.max_num_groups = 0,    /* default num of groups */
-	.flags = MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT | MLX5_FLOW_TABLE_TUNNEL_EN_DECAP,
+	.flags = 0,
 };
 
 struct mlx5e_tc_psample {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/vporttbl.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/vporttbl.c
index 9e72118f2e4c..749c3957a128 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/vporttbl.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/vporttbl.c
@@ -11,7 +11,7 @@ struct mlx5_vport_key {
 	u16 prio;
 	u16 vport;
 	u16 vhca_id;
-	const struct esw_vport_tbl_namespace *vport_ns;
+	struct esw_vport_tbl_namespace *vport_ns;
 } __packed;
 
 struct mlx5_vport_table {
@@ -21,6 +21,14 @@ struct mlx5_vport_table {
 	struct mlx5_vport_key key;
 };
 
+static void
+esw_vport_tbl_init(struct mlx5_eswitch *esw, struct esw_vport_tbl_namespace *ns)
+{
+	if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE)
+		ns->flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT |
+			      MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
+}
+
 static struct mlx5_flow_table *
 esw_vport_tbl_create(struct mlx5_eswitch *esw, struct mlx5_flow_namespace *ns,
 		     const struct esw_vport_tbl_namespace *vport_ns)
@@ -80,6 +88,7 @@ mlx5_esw_vporttbl_get(struct mlx5_eswitch *esw, struct mlx5_vport_tbl_attr *attr
 	u32 hkey;
 
 	mutex_lock(&esw->fdb_table.offloads.vports.lock);
+	esw_vport_tbl_init(esw, attr->vport_ns);
 	hkey = flow_attr_to_vport_key(esw, attr, &skey);
 	e = esw_vport_tbl_lookup(esw, &skey, hkey);
 	if (e) {
@@ -127,6 +136,7 @@ mlx5_esw_vporttbl_put(struct mlx5_eswitch *esw, struct mlx5_vport_tbl_attr *attr
 	u32 hkey;
 
 	mutex_lock(&esw->fdb_table.offloads.vports.lock);
+	esw_vport_tbl_init(esw, attr->vport_ns);
 	hkey = flow_attr_to_vport_key(esw, attr, &key);
 	e = esw_vport_tbl_lookup(esw, &key, hkey);
 	if (!e || --e->num_rules)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 19e9a77c4633..9d5a5756a15a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -674,7 +674,7 @@ struct mlx5_vport_tbl_attr {
 	u32 chain;
 	u16 prio;
 	u16 vport;
-	const struct esw_vport_tbl_namespace *vport_ns;
+	struct esw_vport_tbl_namespace *vport_ns;
 };
 
 struct mlx5_flow_table *
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 25a8076a77bf..706746cd10af 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -73,7 +73,7 @@
 
 #define MLX5_ESW_FT_OFFLOADS_DROP_RULE (1)
 
-static const struct esw_vport_tbl_namespace mlx5_esw_vport_tbl_mirror_ns = {
+static struct esw_vport_tbl_namespace mlx5_esw_vport_tbl_mirror_ns = {
 	.max_fte = MLX5_ESW_VPORT_TBL_SIZE,
 	.max_num_groups = MLX5_ESW_VPORT_TBL_NUM_GROUPS,
 	.flags = 0,
-- 
cgit v1.2.3-70-g09d2


From 4c8189302567f75099a336b0efcff8291ec86ff4 Mon Sep 17 00:00:00 2001
From: Chris Mi <cmi@nvidia.com>
Date: Fri, 10 Mar 2023 09:56:08 +0200
Subject: net/mlx5: E-switch, Don't destroy indirect table in split rule

Source port rewrite (forward to ovs internal port or statck device) isn't
supported in the rule of split action. So there is no indirect table in
split rule. The cited commit destroyes indirect table in split rule. The
indirect table for other rules will be destroyed wrongly. It will cause
traffic loss.

Fix it by removing the destroy function in split rule. And also remove
the destroy function in error flow.

Fixes: 10742efc20a4 ("net/mlx5e: VF tunnel TX traffic offloading")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 706746cd10af..c99d208722f5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -760,7 +760,6 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw,
 	kfree(dest);
 	return rule;
 err_chain_src_rewrite:
-	esw_put_dest_tables_loop(esw, attr, 0, i);
 	mlx5_esw_vporttbl_put(esw, &fwd_attr);
 err_get_fwd:
 	mlx5_chains_put_table(chains, attr->chain, attr->prio, 0);
@@ -803,7 +802,6 @@ __mlx5_eswitch_del_rule(struct mlx5_eswitch *esw,
 	if (fwd_rule)  {
 		mlx5_esw_vporttbl_put(esw, &fwd_attr);
 		mlx5_chains_put_table(chains, attr->chain, attr->prio, 0);
-		esw_put_dest_tables_loop(esw, attr, 0, esw_attr->split_count);
 	} else {
 		if (split)
 			mlx5_esw_vporttbl_put(esw, &fwd_attr);
-- 
cgit v1.2.3-70-g09d2


From 4fbef0f8ea6350eaea89b1e3175f9325252913ac Mon Sep 17 00:00:00 2001
From: Chris Mi <cmi@nvidia.com>
Date: Tue, 4 Apr 2023 03:45:04 +0300
Subject: net/mlx5: Release tunnel device after tc update skb

The cited commit causes a regression. Tunnel device is not released
after tc update skb if skb needs to be freed. The following error
message will be printed:

  unregister_netdevice: waiting for vxlan1 to become free. Usage count = 11

Fix it by releasing tunnel device if skb needs to be freed.

Fixes: 93a1ab2c545b ("net/mlx5: Refactor tc miss handling to a single function")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
index 8f7452dc00ee..668fdee9cf05 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c
@@ -715,5 +715,6 @@ forward:
 	return;
 
 free_skb:
+	dev_put(tc_priv.fwd_dev);
 	dev_kfree_skb_any(skb);
 }
-- 
cgit v1.2.3-70-g09d2


From 0a6b069cc60d68d33b4f6e7dd7f1adc3ec749766 Mon Sep 17 00:00:00 2001
From: Roi Dayan <roid@nvidia.com>
Date: Wed, 29 Mar 2023 15:24:32 +0300
Subject: net/mlx5e: Fix error flow in representor failing to add vport rx rule

On representor init rx error flow the flow steering pointer is being
released so mlx5e_attach_netdev() doesn't have a valid fs pointer
in its error flow. Make sure the pointer is nullified when released
and add a check in mlx5e_fs_cleanup() to verify fs is not null
as representor cleanup callback would be called anyway.

Fixes: af8bbf730068 ("net/mlx5e: Convert mlx5e_flow_steering member of mlx5e_priv to pointer")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_fs.c   | 2 ++
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 1 +
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c  | 2 ++
 3 files changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index 05796f8b1d7c..f1dac0244958 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -1490,6 +1490,8 @@ err:
 
 void mlx5e_fs_cleanup(struct mlx5e_flow_steering *fs)
 {
+	if (!fs)
+		return;
 	debugfs_remove_recursive(fs->dfs_root);
 	mlx5e_fs_ethtool_free(fs);
 	mlx5e_fs_tc_free(fs);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 7ca7e9b57607..579c2d217fdc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -5270,6 +5270,7 @@ static void mlx5e_nic_cleanup(struct mlx5e_priv *priv)
 	mlx5e_health_destroy_reporters(priv);
 	mlx5e_ktls_cleanup(priv);
 	mlx5e_fs_cleanup(priv->fs);
+	priv->fs = NULL;
 }
 
 static int mlx5e_init_nic_rx(struct mlx5e_priv *priv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 8ff654b4e9e1..6e18d91c3d76 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -828,6 +828,7 @@ static int mlx5e_init_ul_rep(struct mlx5_core_dev *mdev,
 static void mlx5e_cleanup_rep(struct mlx5e_priv *priv)
 {
 	mlx5e_fs_cleanup(priv->fs);
+	priv->fs = NULL;
 }
 
 static int mlx5e_create_rep_ttc_table(struct mlx5e_priv *priv)
@@ -994,6 +995,7 @@ err_close_drop_rq:
 	priv->rx_res = NULL;
 err_free_fs:
 	mlx5e_fs_cleanup(priv->fs);
+	priv->fs = NULL;
 	return err;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 21608a2cf38e9f743636b5f86507ffbca18ecee7 Mon Sep 17 00:00:00 2001
From: Moshe Shemesh <moshe@nvidia.com>
Date: Sun, 9 Apr 2023 10:44:31 +0300
Subject: Revert "net/mlx5: Remove "recovery" arg from mlx5_load_one()
 function"

This reverts commit 5977ac3910f1cbaf44dca48179118b25c206ac29.

Revert this patch as we need the "recovery" arg back in mlx5_load_one()
function. This arg will be used in the next patch for using recovery
timeout during sync reset flow.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c  | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/main.c      | 9 +++++----
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h | 2 +-
 3 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index 4c2dad9d7cfb..289e915def98 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -167,7 +167,7 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev)
 		if (mlx5_health_wait_pci_up(dev))
 			mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n");
 		else
-			mlx5_load_one(dev);
+			mlx5_load_one(dev, false);
 		devlink_remote_reload_actions_performed(priv_to_devlink(dev), 0,
 							BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
 							BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index f1de152a6113..ad90bf125e94 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1509,13 +1509,13 @@ out:
 	return err;
 }
 
-int mlx5_load_one(struct mlx5_core_dev *dev)
+int mlx5_load_one(struct mlx5_core_dev *dev, bool recovery)
 {
 	struct devlink *devlink = priv_to_devlink(dev);
 	int ret;
 
 	devl_lock(devlink);
-	ret = mlx5_load_one_devl_locked(dev, false);
+	ret = mlx5_load_one_devl_locked(dev, recovery);
 	devl_unlock(devlink);
 	return ret;
 }
@@ -1912,7 +1912,8 @@ static void mlx5_pci_resume(struct pci_dev *pdev)
 
 	mlx5_pci_trace(dev, "Enter, loading driver..\n");
 
-	err = mlx5_load_one(dev);
+	err = mlx5_load_one(dev, false);
+
 	if (!err)
 		devlink_health_reporter_state_update(dev->priv.health.fw_fatal_reporter,
 						     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
@@ -2003,7 +2004,7 @@ static int mlx5_resume(struct pci_dev *pdev)
 {
 	struct mlx5_core_dev *dev = pci_get_drvdata(pdev);
 
-	return mlx5_load_one(dev);
+	return mlx5_load_one(dev, false);
 }
 
 static const struct pci_device_id mlx5_core_pci_table[] = {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index be0785f83083..a3c5c2dab5fd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -321,7 +321,7 @@ int mlx5_init_one(struct mlx5_core_dev *dev);
 void mlx5_uninit_one(struct mlx5_core_dev *dev);
 void mlx5_unload_one(struct mlx5_core_dev *dev, bool suspend);
 void mlx5_unload_one_devl_locked(struct mlx5_core_dev *dev, bool suspend);
-int mlx5_load_one(struct mlx5_core_dev *dev);
+int mlx5_load_one(struct mlx5_core_dev *dev, bool recovery);
 int mlx5_load_one_devl_locked(struct mlx5_core_dev *dev, bool recovery);
 
 int mlx5_vport_set_other_func_cap(struct mlx5_core_dev *dev, const void *hca_cap, u16 function_id,
-- 
cgit v1.2.3-70-g09d2


From dfad99750c0f83b0242572a573afa2c055f85b36 Mon Sep 17 00:00:00 2001
From: Moshe Shemesh <moshe@nvidia.com>
Date: Sun, 2 Apr 2023 06:49:53 +0300
Subject: net/mlx5: Use recovery timeout on sync reset flow

Use the same timeout for sync reset flow and health recovery flow, since
the former involves driver's recovery from firmware reset, which is
similar to health recovery. Otherwise, in some cases, such as a firmware
upgrade on the DPU, the firmware pre-init bit may not be ready within
current timeout and the driver will abort loading back after reset.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Fixes: 37ca95e62ee2 ("net/mlx5: Increase FW pre-init timeout for health recovery")
Reviewed-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/devlink.c  | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index c5d2fdcabd56..e5f03d071a37 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -202,7 +202,7 @@ static int mlx5_devlink_reload_up(struct devlink *devlink, enum devlink_reload_a
 			break;
 		/* On fw_activate action, also driver is reloaded and reinit performed */
 		*actions_performed |= BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT);
-		ret = mlx5_load_one_devl_locked(dev, false);
+		ret = mlx5_load_one_devl_locked(dev, true);
 		break;
 	default:
 		/* Unsupported action should not get to this function */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index 289e915def98..50022e7565f1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -167,7 +167,7 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev)
 		if (mlx5_health_wait_pci_up(dev))
 			mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n");
 		else
-			mlx5_load_one(dev, false);
+			mlx5_load_one(dev, true);
 		devlink_remote_reload_actions_performed(priv_to_devlink(dev), 0,
 							BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) |
 							BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE));
@@ -499,7 +499,7 @@ int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev)
 	err = fw_reset->ret;
 	if (test_and_clear_bit(MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED, &fw_reset->reset_flags)) {
 		mlx5_unload_one_devl_locked(dev, false);
-		mlx5_load_one_devl_locked(dev, false);
+		mlx5_load_one_devl_locked(dev, true);
 	}
 out:
 	clear_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags);
-- 
cgit v1.2.3-70-g09d2


From 1b540decd03acd736693aabb6cb46c71fca38ae6 Mon Sep 17 00:00:00 2001
From: Aya Levin <ayal@nvidia.com>
Date: Sun, 12 Mar 2023 16:37:36 +0200
Subject: net/mlx5e: Nullify table pointer when failing to create

On failing to create promisc flow steering table, the pointer is
returned with an error. Nullify it so unloading the driver won't try to
destroy a non existing table.

Failing to create promisc table may happen over BF devices when the ARM
side is going through a firmware tear down. The host side start a
reload flow. While the driver unloads, it tries to remove the promisc
table. Remove WARN in this state as it is a valid error flow.

Fixes: 1c46d7409f30 ("net/mlx5e: Optimize promiscuous mode")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_fs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index f1dac0244958..33bfe4d7338b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -783,6 +783,7 @@ static int mlx5e_create_promisc_table(struct mlx5e_flow_steering *fs)
 	ft->t = mlx5_create_auto_grouped_flow_table(fs->ns, &ft_attr);
 	if (IS_ERR(ft->t)) {
 		err = PTR_ERR(ft->t);
+		ft->t = NULL;
 		fs_err(fs, "fail to create promisc table err=%d\n", err);
 		return err;
 	}
@@ -810,7 +811,7 @@ static void mlx5e_del_promisc_rule(struct mlx5e_flow_steering *fs)
 
 static void mlx5e_destroy_promisc_table(struct mlx5e_flow_steering *fs)
 {
-	if (WARN(!fs->promisc.ft.t, "Trying to remove non-existing promiscuous table"))
+	if (!fs->promisc.ft.t)
 		return;
 	mlx5e_del_promisc_rule(fs);
 	mlx5_destroy_flow_table(fs->promisc.ft.t);
-- 
cgit v1.2.3-70-g09d2


From 081abcacaf0a9505c0d4cc9780b12e6ce5d33630 Mon Sep 17 00:00:00 2001
From: Vlad Buslov <vladbu@nvidia.com>
Date: Thu, 13 Apr 2023 15:06:59 +0200
Subject: Revert "net/mlx5e: Don't use termination table when redundant"

This reverts commit 14624d7247fcd0f3114a6f5f17b3c8d1020fbbb7.

The termination table usage is requires for DMFS steering mode as firmware
doesn't support mixed table destinations list which causes following
syndrome with hairpin rules:

[81922.283225] mlx5_core 0000:08:00.0: mlx5_cmd_out_err:803:(pid 25977): SET_FLOW_TABLE_ENTRY(0x936) op_mod(0x0) failed, status bad parameter(0x3), syndrome (0xaca205), err(-22)

Fixes: 14624d7247fc ("net/mlx5e: Don't use termination table when redundant")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/eswitch_offloads_termtbl.c  | 32 +++-------------------
 1 file changed, 4 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
index 3a9a6bb9158d..edd910258314 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c
@@ -210,18 +210,6 @@ static bool mlx5_eswitch_offload_is_uplink_port(const struct mlx5_eswitch *esw,
 	return (port_mask & port_value) == MLX5_VPORT_UPLINK;
 }
 
-static bool
-mlx5_eswitch_is_push_vlan_no_cap(struct mlx5_eswitch *esw,
-				 struct mlx5_flow_act *flow_act)
-{
-	if (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH &&
-	    !(mlx5_fs_get_capabilities(esw->dev, MLX5_FLOW_NAMESPACE_FDB) &
-	      MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX))
-		return true;
-
-	return false;
-}
-
 bool
 mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw,
 			      struct mlx5_flow_attr *attr,
@@ -237,7 +225,10 @@ mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw,
 	    (!mlx5_eswitch_offload_is_uplink_port(esw, spec) && !esw_attr->int_port))
 		return false;
 
-	if (mlx5_eswitch_is_push_vlan_no_cap(esw, flow_act))
+	/* push vlan on RX */
+	if (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH &&
+	    !(mlx5_fs_get_capabilities(esw->dev, MLX5_FLOW_NAMESPACE_FDB) &
+	      MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX))
 		return true;
 
 	/* hairpin */
@@ -261,31 +252,19 @@ mlx5_eswitch_add_termtbl_rule(struct mlx5_eswitch *esw,
 	struct mlx5_flow_act term_tbl_act = {};
 	struct mlx5_flow_handle *rule = NULL;
 	bool term_table_created = false;
-	bool is_push_vlan_on_rx;
 	int num_vport_dests = 0;
 	int i, curr_dest;
 
-	is_push_vlan_on_rx = mlx5_eswitch_is_push_vlan_no_cap(esw, flow_act);
 	mlx5_eswitch_termtbl_actions_move(flow_act, &term_tbl_act);
 	term_tbl_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 
 	for (i = 0; i < num_dest; i++) {
 		struct mlx5_termtbl_handle *tt;
-		bool hairpin = false;
 
 		/* only vport destinations can be terminated */
 		if (dest[i].type != MLX5_FLOW_DESTINATION_TYPE_VPORT)
 			continue;
 
-		if (attr->dests[num_vport_dests].rep &&
-		    attr->dests[num_vport_dests].rep->vport == MLX5_VPORT_UPLINK)
-			hairpin = true;
-
-		if (!is_push_vlan_on_rx && !hairpin) {
-			num_vport_dests++;
-			continue;
-		}
-
 		if (attr->dests[num_vport_dests].flags & MLX5_ESW_DEST_ENCAP) {
 			term_tbl_act.action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
 			term_tbl_act.pkt_reformat = attr->dests[num_vport_dests].pkt_reformat;
@@ -333,9 +312,6 @@ revert_changes:
 	for (curr_dest = 0; curr_dest < num_vport_dests; curr_dest++) {
 		struct mlx5_termtbl_handle *tt = attr->dests[curr_dest].termtbl;
 
-		if (!tt)
-			continue;
-
 		attr->dests[curr_dest].termtbl = NULL;
 
 		/* search for the destination associated with the
-- 
cgit v1.2.3-70-g09d2


From 4bb7aac70b5d8a4bddf4ee0791b834f9f56883d2 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 20 Apr 2023 10:45:51 +0200
Subject: net: phy: fix circular LEDS_CLASS dependencies

The CONFIG_PHYLIB symbol is selected by a number of device drivers that
need PHY support, but it now has a dependency on CONFIG_LEDS_CLASS,
which may not be enabled, causing build failures.

Avoid the risk of missing and circular dependencies by guarding the
phylib LED support itself in another Kconfig symbol that can only be
enabled if the dependency is met.

This could be made a hidden symbol and always enabled when both CONFIG_OF
and CONFIG_LEDS_CLASS are reachable from the phylib, but there may be an
advantage in having users see this option when they have a misconfigured
kernel without built-in LED support.

Fixes: 01e5b728e9e4 ("net: phy: Add a binding for PHY LEDs")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230420084624.3005701-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/Kconfig      | 9 ++++++++-
 drivers/net/phy/phy_device.c | 3 ++-
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index b8cc49820ced..513675ae4dd2 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -18,7 +18,6 @@ menuconfig PHYLIB
 	depends on NETDEVICES
 	select MDIO_DEVICE
 	select MDIO_DEVRES
-	depends on LEDS_CLASS || LEDS_CLASS=n
 	help
 	  Ethernet controllers are usually attached to PHY
 	  devices.  This option provides infrastructure for
@@ -45,6 +44,14 @@ config LED_TRIGGER_PHY
 		<Speed in megabits>Mbps OR <Speed in gigabits>Gbps OR link
 		for any speed known to the PHY.
 
+config PHYLIB_LEDS
+	bool "Support probing LEDs from device tree"
+	depends on LEDS_CLASS=y || LEDS_CLASS=PHYLIB
+	depends on OF
+	default y
+	help
+	  When LED class support is enabled, phylib can automatically
+	  probe LED setting from device tree.
 
 config FIXED_PHY
 	tristate "MDIO Bus/PHY emulation with fixed speed/link PHYs"
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 538523a7cd51..d373446ab5ac 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -3284,7 +3284,8 @@ static int phy_probe(struct device *dev)
 	/* Get the LEDs from the device tree, and instantiate standard
 	 * LEDs for them.
 	 */
-	err = of_phy_leds(phydev);
+	if (IS_ENABLED(CONFIG_PHYLIB_LEDS))
+		err = of_phy_leds(phydev);
 
 out:
 	/* Re-assert the reset signal on error */
-- 
cgit v1.2.3-70-g09d2


From dd64b232deb8d48812a2ea739d1fedaeaffb59ed Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 19 Apr 2023 11:20:06 -0700
Subject: page_pool: unlink from napi during destroy

Jesper points out that we must prevent recycling into cache
after page_pool_destroy() is called, because page_pool_destroy()
is not synchronized with recycling (some pages may still be
outstanding when destroy() gets called).

I assumed this will not happen because NAPI can't be scheduled
if its page pool is being destroyed. But I missed the fact that
NAPI may get reused. For instance when user changes ring configuration
driver may allocate a new page pool, stop NAPI, swap, start NAPI,
and then destroy the old pool. The NAPI is running so old page
pool will think it can recycle to the cache, but the consumer
at that point is the destroy() path, not NAPI.

To avoid extra synchronization let the drivers do "unlinking"
during the "swap" stage while NAPI is indeed disabled.

Fixes: 8c48eea3adf3 ("page_pool: allow caching from safely localized NAPI")
Reported-by: Jesper Dangaard Brouer <jbrouer@redhat.com>
Link: https://lore.kernel.org/all/e8df2654-6a5b-3c92-489d-2fe5e444135f@redhat.com/
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/r/20230419182006.719923-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/page_pool.h |  5 +++++
 net/core/page_pool.c    | 18 +++++++++++++++++-
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/include/net/page_pool.h b/include/net/page_pool.h
index 91b808dade82..c8ec2f34722b 100644
--- a/include/net/page_pool.h
+++ b/include/net/page_pool.h
@@ -247,6 +247,7 @@ struct page_pool *page_pool_create(const struct page_pool_params *params);
 struct xdp_mem_info;
 
 #ifdef CONFIG_PAGE_POOL
+void page_pool_unlink_napi(struct page_pool *pool);
 void page_pool_destroy(struct page_pool *pool);
 void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *),
 			   struct xdp_mem_info *mem);
@@ -254,6 +255,10 @@ void page_pool_release_page(struct page_pool *pool, struct page *page);
 void page_pool_put_page_bulk(struct page_pool *pool, void **data,
 			     int count);
 #else
+static inline void page_pool_unlink_napi(struct page_pool *pool)
+{
+}
+
 static inline void page_pool_destroy(struct page_pool *pool)
 {
 }
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 97f20f7ff4fc..e212e9d7edcb 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -839,6 +839,21 @@ void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *),
 	pool->xdp_mem_id = mem->id;
 }
 
+void page_pool_unlink_napi(struct page_pool *pool)
+{
+	if (!pool->p.napi)
+		return;
+
+	/* To avoid races with recycling and additional barriers make sure
+	 * pool and NAPI are unlinked when NAPI is disabled.
+	 */
+	WARN_ON(!test_bit(NAPI_STATE_SCHED, &pool->p.napi->state) ||
+		READ_ONCE(pool->p.napi->list_owner) != -1);
+
+	WRITE_ONCE(pool->p.napi, NULL);
+}
+EXPORT_SYMBOL(page_pool_unlink_napi);
+
 void page_pool_destroy(struct page_pool *pool)
 {
 	if (!pool)
@@ -847,6 +862,7 @@ void page_pool_destroy(struct page_pool *pool)
 	if (!page_pool_put(pool))
 		return;
 
+	page_pool_unlink_napi(pool);
 	page_pool_free_frag(pool);
 
 	if (!page_pool_release(pool))
@@ -900,7 +916,7 @@ bool page_pool_return_skb_page(struct page *page, bool napi_safe)
 	 * in the same context as the consumer would run, so there's
 	 * no possible race.
 	 */
-	napi = pp->p.napi;
+	napi = READ_ONCE(pp->p.napi);
 	allow_direct = napi_safe && napi &&
 		READ_ONCE(napi->list_owner) == smp_processor_id();
 
-- 
cgit v1.2.3-70-g09d2


From e375b503943f512cdd3a7dd12849972b7a006076 Mon Sep 17 00:00:00 2001
From: Jeroen de Borst <jeroendb@google.com>
Date: Wed, 19 Apr 2023 14:05:58 -0700
Subject: gve: update MAINTAINERS

This reflects role changes in our team.

Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Link: https://lore.kernel.org/r/20230419210558.1893400-1-jeroendb@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index c6545eb54104..eefcea6f8757 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8690,7 +8690,7 @@ F:	drivers/input/touchscreen/goodix*
 
 GOOGLE ETHERNET DRIVERS
 M:	Jeroen de Borst <jeroendb@google.com>
-M:	Catherine Sullivan <csully@google.com>
+M:	Praveen Kaligineedi <pkaligineedi@google.com>
 R:	Shailend Chand <shailend@google.com>
 L:	netdev@vger.kernel.org
 S:	Supported
-- 
cgit v1.2.3-70-g09d2


From 8c966a10eb8478dea333a0f7f77b11c3cb3f7dcb Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Wed, 19 Apr 2023 13:34:35 +0200
Subject: flow_dissector: Address kdoc warnings

Address a number of warnings flagged by
./scripts/kernel-doc -none include/net/flow_dissector.h

 include/net/flow_dissector.h:23: warning: Function parameter or member 'addr_type' not described in 'flow_dissector_key_control'
 include/net/flow_dissector.h:23: warning: Function parameter or member 'flags' not described in 'flow_dissector_key_control'
 include/net/flow_dissector.h:46: warning: Function parameter or member 'padding' not described in 'flow_dissector_key_basic'
 include/net/flow_dissector.h:145: warning: Function parameter or member 'tipckey' not described in 'flow_dissector_key_addrs'
 include/net/flow_dissector.h:157: warning: cannot understand function prototype: 'struct flow_dissector_key_arp '
 include/net/flow_dissector.h:171: warning: cannot understand function prototype: 'struct flow_dissector_key_ports '
 include/net/flow_dissector.h:203: warning: cannot understand function prototype: 'struct flow_dissector_key_icmp '

Also improve indentation on adjacent lines to those changed
to address the above.

No functional changes intended.

Signed-off-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230419-flow-dissector-kdoc-v1-1-1aa0cca1118b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/flow_dissector.h | 38 ++++++++++++++++++++------------------
 1 file changed, 20 insertions(+), 18 deletions(-)

diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index 5ccf52ef8809..85b2281576ed 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -14,7 +14,9 @@ struct sk_buff;
 
 /**
  * struct flow_dissector_key_control:
- * @thoff: Transport header offset
+ * @thoff:     Transport header offset
+ * @addr_type: Type of key. One of FLOW_DISSECTOR_KEY_*
+ * @flags:     Key flags. Any of FLOW_DIS_(IS_FRAGMENT|FIRST_FRAGENCAPSULATION)
  */
 struct flow_dissector_key_control {
 	u16	thoff;
@@ -36,8 +38,9 @@ enum flow_dissect_ret {
 
 /**
  * struct flow_dissector_key_basic:
- * @n_proto: Network header protocol (eg. IPv4/IPv6)
+ * @n_proto:  Network header protocol (eg. IPv4/IPv6)
  * @ip_proto: Transport header protocol (eg. TCP/UDP)
+ * @padding:  Unused
  */
 struct flow_dissector_key_basic {
 	__be16	n_proto;
@@ -135,6 +138,7 @@ struct flow_dissector_key_tipc {
  * struct flow_dissector_key_addrs:
  * @v4addrs: IPv4 addresses
  * @v6addrs: IPv6 addresses
+ * @tipckey: TIPC key
  */
 struct flow_dissector_key_addrs {
 	union {
@@ -145,14 +149,12 @@ struct flow_dissector_key_addrs {
 };
 
 /**
- * flow_dissector_key_arp:
- *	@ports: Operation, source and target addresses for an ARP header
- *              for Ethernet hardware addresses and IPv4 protocol addresses
- *		sip: Sender IP address
- *		tip: Target IP address
- *		op:  Operation
- *		sha: Sender hardware address
- *		tpa: Target hardware address
+ * struct flow_dissector_key_arp:
+ * @sip: Sender IP address
+ * @tip: Target IP address
+ * @op:  Operation
+ * @sha: Sender hardware address
+ * @tha: Target hardware address
  */
 struct flow_dissector_key_arp {
 	__u32 sip;
@@ -163,10 +165,10 @@ struct flow_dissector_key_arp {
 };
 
 /**
- * flow_dissector_key_tp_ports:
- *	@ports: port numbers of Transport header
- *		src: source port number
- *		dst: destination port number
+ * struct flow_dissector_key_ports:
+ * @ports: port numbers of Transport header
+ * @src: source port number
+ * @dst: destination port number
  */
 struct flow_dissector_key_ports {
 	union {
@@ -195,10 +197,10 @@ struct flow_dissector_key_ports_range {
 };
 
 /**
- * flow_dissector_key_icmp:
- *		type: ICMP type
- *		code: ICMP code
- *		id:   session identifier
+ * struct flow_dissector_key_icmp:
+ * @type: ICMP type
+ * @code: ICMP code
+ * @id:   Session identifier
  */
 struct flow_dissector_key_icmp {
 	struct {
-- 
cgit v1.2.3-70-g09d2


From 61718206ee5de9cc2c3503c59287c0b968446508 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 19 Apr 2023 18:58:02 -0700
Subject: eth: mlx5: avoid iterator use outside of a loop

Fix the following warning about risky iterator use:

drivers/net/ethernet/mellanox/mlx5/core/eq.c:1010 mlx5_comp_irq_get_affinity_mask() warn: iterator used outside loop: 'eq'

Acked-by: Saeed Mahameed <saeed@kernel.org>
Link: https://lore.kernel.org/r/20230420015802.815362-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index eb41f0abf798..1c35d721a31d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -1070,10 +1070,11 @@ mlx5_comp_irq_get_affinity_mask(struct mlx5_core_dev *dev, int vector)
 
 	list_for_each_entry(eq, &table->comp_eqs_list, list) {
 		if (i++ == vector)
-			break;
+			return mlx5_irq_get_affinity_mask(eq->core.irq);
 	}
 
-	return mlx5_irq_get_affinity_mask(eq->core.irq);
+	WARN_ON_ONCE(1);
+	return NULL;
 }
 EXPORT_SYMBOL(mlx5_comp_irq_get_affinity_mask);
 
-- 
cgit v1.2.3-70-g09d2


From 8fa66e4a1bdd41d55d7842928e60a40fed65715d Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 19 Apr 2023 19:00:05 -0700
Subject: net: skbuff: update and rename __kfree_skb_defer()

__kfree_skb_defer() uses the old naming where "defer" meant
slab bulk free/alloc APIs. In the meantime we also made
__kfree_skb_defer() feed the per-NAPI skb cache, which
implies bulk APIs. So take away the 'defer' and add 'napi'.

While at it add a drop reason. This only matters on the
tx_action path, if the skb has a frag_list. But getting
rid of a SKB_DROP_REASON_NOT_SPECIFIED seems like a net
benefit so why not.

Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230420020005.815854-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/skbuff.h | 2 +-
 net/core/dev.c         | 3 ++-
 net/core/gro.c         | 2 +-
 net/core/skbuff.c      | 4 ++--
 4 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index db5973559042..03aa7ed076f0 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3253,7 +3253,7 @@ static inline struct sk_buff *napi_alloc_skb(struct napi_struct *napi,
 void napi_consume_skb(struct sk_buff *skb, int budget);
 
 void napi_skb_free_stolen_head(struct sk_buff *skb);
-void __kfree_skb_defer(struct sk_buff *skb);
+void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason);
 
 /**
  * __dev_alloc_pages - allocate page for network Rx
diff --git a/net/core/dev.c b/net/core/dev.c
index 3fc4dba71f9d..1551aabac343 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5040,7 +5040,8 @@ static __latent_entropy void net_tx_action(struct softirq_action *h)
 			if (skb->fclone != SKB_FCLONE_UNAVAILABLE)
 				__kfree_skb(skb);
 			else
-				__kfree_skb_defer(skb);
+				__napi_kfree_skb(skb,
+						 get_kfree_skb_cb(skb)->reason);
 		}
 	}
 
diff --git a/net/core/gro.c b/net/core/gro.c
index a606705a0859..2d84165cb4f1 100644
--- a/net/core/gro.c
+++ b/net/core/gro.c
@@ -633,7 +633,7 @@ static gro_result_t napi_skb_finish(struct napi_struct *napi,
 		else if (skb->fclone != SKB_FCLONE_UNAVAILABLE)
 			__kfree_skb(skb);
 		else
-			__kfree_skb_defer(skb);
+			__napi_kfree_skb(skb, SKB_CONSUMED);
 		break;
 
 	case GRO_HELD:
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 768f9d04911f..8764653bede7 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1226,9 +1226,9 @@ static void napi_skb_cache_put(struct sk_buff *skb)
 	}
 }
 
-void __kfree_skb_defer(struct sk_buff *skb)
+void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason)
 {
-	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, true);
+	skb_release_all(skb, reason, true);
 	napi_skb_cache_put(skb);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 787e6144aef74e73f88875cb5d3a5af01701af6f Mon Sep 17 00:00:00 2001
From: Slark Xiao <slark_xiao@163.com>
Date: Thu, 20 Apr 2023 10:36:17 +0800
Subject: wwan: core: add print for wwan port attach/disconnect

Refer to USB serial device or net device, there is a notice to
let end user know the status of device, like attached or
disconnected. Add attach/disconnect print for wwan device as
well.

Signed-off-by: Slark Xiao <slark_xiao@163.com>
Reviewed-by: Loic Poulain <loic.poulain@linaro.org>
Link: https://lore.kernel.org/r/20230420023617.3919569-1-slark_xiao@163.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/wwan/wwan_core.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wwan/wwan_core.c b/drivers/net/wwan/wwan_core.c
index 2e1c01cf00a9..aa54fa6d5f90 100644
--- a/drivers/net/wwan/wwan_core.c
+++ b/drivers/net/wwan/wwan_core.c
@@ -492,6 +492,7 @@ struct wwan_port *wwan_create_port(struct device *parent,
 	if (err)
 		goto error_put_device;
 
+	dev_info(&wwandev->dev, "port %s attached\n", dev_name(&port->dev));
 	return port;
 
 error_put_device:
@@ -517,6 +518,8 @@ void wwan_remove_port(struct wwan_port *port)
 
 	skb_queue_purge(&port->rxq);
 	dev_set_drvdata(&port->dev, NULL);
+
+	dev_info(&wwandev->dev, "port %s disconnected\n", dev_name(&port->dev));
 	device_unregister(&port->dev);
 
 	/* Release related wwan device */
-- 
cgit v1.2.3-70-g09d2


From 59be75db5966f920bf5cccfd2a34b3bf222660d7 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:51 +0300
Subject: net: enetc: fix MAC Merge layer remaining enabled until a link down
 event

Current enetc_set_mm() is designed to set the priv->active_offloads bit
ENETC_F_QBU for enetc_mm_link_state_update() to act on, but if the link
is already up, it modifies the ENETC_MMCSR_ME ("Merge Enable") bit
directly.

The problem is that it only *sets* ENETC_MMCSR_ME if the link is up, it
doesn't *clear* it if needed. So subsequent enetc_get_mm() calls still
see tx-enabled as true, up until a link down event, which is when
enetc_mm_link_state_update() will get called.

This is not a functional issue as far as I can assess. It has only come
up because I'd like to uphold a simple API rule in core ethtool code:
the pMAC cannot be disabled if TX is going to be enabled. Currently,
the fact that TX remains enabled for longer than expected (after the
enetc_set_mm() call that disables it) is going to violate that rule,
which is how it was caught.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index 838750a03cf6..ee1ea71fe79e 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -1041,10 +1041,13 @@ static int enetc_set_mm(struct net_device *ndev, struct ethtool_mm_cfg *cfg,
 	else
 		priv->active_offloads &= ~ENETC_F_QBU;
 
-	/* If link is up, enable MAC Merge right away */
-	if (!!(priv->active_offloads & ENETC_F_QBU) &&
-	    !(val & ENETC_MMCSR_LINK_FAIL))
-		val |= ENETC_MMCSR_ME;
+	/* If link is up, enable/disable MAC Merge right away */
+	if (!(val & ENETC_MMCSR_LINK_FAIL)) {
+		if (!!(priv->active_offloads & ENETC_F_QBU))
+			val |= ENETC_MMCSR_ME;
+		else
+			val &= ~ENETC_MMCSR_ME;
+	}
 
 	val &= ~ENETC_MMCSR_VT_MASK;
 	val |= ENETC_MMCSR_VT(cfg->verify_time);
-- 
cgit v1.2.3-70-g09d2


From 153b5b1d030de3f94ecac4eec73854e4710ca6ba Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:52 +0300
Subject: net: enetc: report mm tx-active based on tx-enabled and verify-status

The MMCSR register contains 2 fields with overlapping meaning:

- LPA (Local preemption active):
This read-only status bit indicates whether preemption is active for
this port. This bit will be set if preemption is both enabled and has
completed the verification process.
- TXSTS (Merge status):
This read-only status field provides the state of the MAC Merge sublayer
transmit status as defined in IEEE Std 802.3-2018 Clause 99.
00 Transmit preemption is inactive
01 Transmit preemption is active
10 Reserved
11 Reserved

However none of these 2 fields offer reliable reporting to software.

When connecting ENETC to a link partner which is not capable of Frame
Preemption, the expectation is that ENETC's verification should fail
(VSTS=4) and its MM TX direction should be inactive (LPA=0, TXSTS=00)
even though the MM TX is enabled (ME=1). But surprise, the LPA bit of
MMCSR stays set even if VSTS=4 and ME=1.

OTOH, the TXSTS field has the opposite problem. I cannot get its value
to change from 0, even when connecting to a link partner capable of
frame preemption, which does respond to its verification frames (ME=1
and VSTS=3, "SUCCEEDED").

The only option with such buggy hardware seems to be to reimplement the
formula for calculating tx-active in software, which is for tx-enabled
to be true, and for the verify-status to be either SUCCEEDED, or
DISABLED.

Without reliable tx-active reporting, we have no good indication when
to commit the preemptible traffic classes to hardware, which makes it
possible (but not desirable) to send preemptible traffic to a link
partner incapable of receiving it. However, currently we do not have the
logic to wait for TX to be active yet, so the impact is limited.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index ee1ea71fe79e..deb674752851 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -976,7 +976,9 @@ static int enetc_get_mm(struct net_device *ndev, struct ethtool_mm_state *state)
 	lafs = ENETC_MMCSR_GET_LAFS(val);
 	state->rx_min_frag_size = ethtool_mm_frag_size_add_to_min(lafs);
 	state->tx_enabled = !!(val & ENETC_MMCSR_LPE); /* mirror of MMCSR_ME */
-	state->tx_active = !!(val & ENETC_MMCSR_LPA);
+	state->tx_active = state->tx_enabled &&
+			   (state->verify_status == ETHTOOL_MM_VERIFY_STATUS_SUCCEEDED ||
+			    state->verify_status == ETHTOOL_MM_VERIFY_STATUS_DISABLED);
 	state->verify_enabled = !(val & ENETC_MMCSR_VDIS);
 	state->verify_time = ENETC_MMCSR_GET_VT(val);
 	/* A verifyTime of 128 ms would exceed the 7 bit width
-- 
cgit v1.2.3-70-g09d2


From 827145392a4aad635b93e5235b7d7fecc2fa31c7 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:53 +0300
Subject: net: enetc: only commit preemptible TCs to hardware when MM TX is
 active

This was left as TODO in commit 01e23b2b3bad ("net: enetc: add support
for preemptible traffic classes") since it's relatively complicated.

Where this makes a difference is with a configuration as follows:

ethtool --set-mm eno0 pmac-enabled on tx-enabled on verify-enabled on

Preemptible packets should only be sent when the MAC Merge TX direction
becomes active (i.o.w. when the verification process succeeds, aka when
the link partner confirms it can process preemptible traffic). But the
tc qdisc with the preemptible traffic classes is offloaded completely
asynchronously w.r.t. the MM becoming active.

The ENETC manual does suggest that this should be handled in the driver:
"On startup, software should wait for the verification process to
complete (MMCSR[VSTS]=011) before initiating traffic".

Adding the necessary logic allows future selftests to uphold the claim
that an inactive or disabled MAC Merge layer should never send data
packets through the pMAC.

This change moves enetc_set_ptcfpr() from enetc.c to enetc_ethtool.c,
where its only caller is now - enetc_mm_commit_preemptible_tcs().

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/enetc/enetc.c       | 23 +++-----
 drivers/net/ethernet/freescale/enetc/enetc.h       |  5 +-
 .../net/ethernet/freescale/enetc/enetc_ethtool.c   | 62 ++++++++++++++++++++++
 drivers/net/ethernet/freescale/enetc/enetc_hw.h    |  3 ++
 4 files changed, 75 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 41c194c1672d..3c4fa26f0f9b 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -25,23 +25,12 @@ void enetc_port_mac_wr(struct enetc_si *si, u32 reg, u32 val)
 }
 EXPORT_SYMBOL_GPL(enetc_port_mac_wr);
 
-void enetc_set_ptcfpr(struct enetc_hw *hw, unsigned long preemptible_tcs)
+static void enetc_change_preemptible_tcs(struct enetc_ndev_priv *priv,
+					 u8 preemptible_tcs)
 {
-	u32 val;
-	int tc;
-
-	for (tc = 0; tc < 8; tc++) {
-		val = enetc_port_rd(hw, ENETC_PTCFPR(tc));
-
-		if (preemptible_tcs & BIT(tc))
-			val |= ENETC_PTCFPR_FPE;
-		else
-			val &= ~ENETC_PTCFPR_FPE;
-
-		enetc_port_wr(hw, ENETC_PTCFPR(tc), val);
-	}
+	priv->preemptible_tcs = preemptible_tcs;
+	enetc_mm_commit_preemptible_tcs(priv);
 }
-EXPORT_SYMBOL_GPL(enetc_set_ptcfpr);
 
 static int enetc_num_stack_tx_queues(struct enetc_ndev_priv *priv)
 {
@@ -2659,7 +2648,7 @@ static void enetc_reset_tc_mqprio(struct net_device *ndev)
 
 	enetc_debug_tx_ring_prios(priv);
 
-	enetc_set_ptcfpr(hw, 0);
+	enetc_change_preemptible_tcs(priv, 0);
 }
 
 int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
@@ -2714,7 +2703,7 @@ int enetc_setup_tc_mqprio(struct net_device *ndev, void *type_data)
 
 	enetc_debug_tx_ring_prios(priv);
 
-	enetc_set_ptcfpr(hw, mqprio->preemptible_tcs);
+	enetc_change_preemptible_tcs(priv, mqprio->preemptible_tcs);
 
 	return 0;
 
diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h
index 143078a9ef16..c97a8e3d7a7f 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc.h
@@ -355,6 +355,9 @@ struct enetc_ndev_priv {
 	u16 rx_bd_count, tx_bd_count;
 
 	u16 msg_enable;
+
+	u8 preemptible_tcs;
+
 	enum enetc_active_offloads active_offloads;
 
 	u32 speed; /* store speed for compare update pspeed */
@@ -433,6 +436,7 @@ int enetc_xdp_xmit(struct net_device *ndev, int num_frames,
 /* ethtool */
 void enetc_set_ethtool_ops(struct net_device *ndev);
 void enetc_mm_link_state_update(struct enetc_ndev_priv *priv, bool link);
+void enetc_mm_commit_preemptible_tcs(struct enetc_ndev_priv *priv);
 
 /* control buffer descriptor ring (CBDR) */
 int enetc_setup_cbdr(struct device *dev, struct enetc_hw *hw, int bd_count,
@@ -486,7 +490,6 @@ static inline void enetc_cbd_free_data_mem(struct enetc_si *si, int size,
 
 void enetc_reset_ptcmsdur(struct enetc_hw *hw);
 void enetc_set_ptcmsdur(struct enetc_hw *hw, u32 *queue_max_sdu);
-void enetc_set_ptcfpr(struct enetc_hw *hw, unsigned long preemptible_tcs);
 
 #ifdef CONFIG_FSL_ENETC_QOS
 int enetc_qos_query_caps(struct net_device *ndev, void *type_data);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index deb674752851..838a92131963 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -991,6 +991,64 @@ static int enetc_get_mm(struct net_device *ndev, struct ethtool_mm_state *state)
 	return 0;
 }
 
+static int enetc_mm_wait_tx_active(struct enetc_hw *hw, int verify_time)
+{
+	int timeout = verify_time * USEC_PER_MSEC * ENETC_MM_VERIFY_RETRIES;
+	u32 val;
+
+	/* This will time out after the standard value of 3 verification
+	 * attempts. To not sleep forever, it relies on a non-zero verify_time,
+	 * guarantee which is provided by the ethtool nlattr policy.
+	 */
+	return read_poll_timeout(enetc_port_rd, val,
+				 ENETC_MMCSR_GET_VSTS(val) == 3,
+				 ENETC_MM_VERIFY_SLEEP_US, timeout,
+				 true, hw, ENETC_MMCSR);
+}
+
+static void enetc_set_ptcfpr(struct enetc_hw *hw, u8 preemptible_tcs)
+{
+	u32 val;
+	int tc;
+
+	for (tc = 0; tc < 8; tc++) {
+		val = enetc_port_rd(hw, ENETC_PTCFPR(tc));
+
+		if (preemptible_tcs & BIT(tc))
+			val |= ENETC_PTCFPR_FPE;
+		else
+			val &= ~ENETC_PTCFPR_FPE;
+
+		enetc_port_wr(hw, ENETC_PTCFPR(tc), val);
+	}
+}
+
+/* ENETC does not have an IRQ to notify changes to the MAC Merge TX status
+ * (active/inactive), but the preemptible traffic classes should only be
+ * committed to hardware once TX is active. Resort to polling.
+ */
+void enetc_mm_commit_preemptible_tcs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	u8 preemptible_tcs = 0;
+	u32 val;
+	int err;
+
+	val = enetc_port_rd(hw, ENETC_MMCSR);
+	if (!(val & ENETC_MMCSR_ME))
+		goto out;
+
+	if (!(val & ENETC_MMCSR_VDIS)) {
+		err = enetc_mm_wait_tx_active(hw, ENETC_MMCSR_GET_VT(val));
+		if (err)
+			goto out;
+	}
+
+	preemptible_tcs = priv->preemptible_tcs;
+out:
+	enetc_set_ptcfpr(hw, preemptible_tcs);
+}
+
 /* FIXME: Workaround for the link partner's verification failing if ENETC
  * priorly received too much express traffic. The documentation doesn't
  * suggest this is needed.
@@ -1061,6 +1119,8 @@ static int enetc_set_mm(struct net_device *ndev, struct ethtool_mm_cfg *cfg,
 
 	enetc_restart_emac_rx(priv->si);
 
+	enetc_mm_commit_preemptible_tcs(priv);
+
 	mutex_unlock(&priv->mm_lock);
 
 	return 0;
@@ -1094,6 +1154,8 @@ void enetc_mm_link_state_update(struct enetc_ndev_priv *priv, bool link)
 
 	enetc_port_wr(hw, ENETC_MMCSR, val);
 
+	enetc_mm_commit_preemptible_tcs(priv);
+
 	mutex_unlock(&priv->mm_lock);
 }
 EXPORT_SYMBOL_GPL(enetc_mm_link_state_update);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
index 36bb2d6d5658..1619943fb263 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h
+++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h
@@ -3,6 +3,9 @@
 
 #include <linux/bitops.h>
 
+#define ENETC_MM_VERIFY_SLEEP_US	USEC_PER_MSEC
+#define ENETC_MM_VERIFY_RETRIES		3
+
 /* ENETC device IDs */
 #define ENETC_DEV_ID_PF		0xe100
 #define ENETC_DEV_ID_VF		0xef00
-- 
cgit v1.2.3-70-g09d2


From 16a2c7634442b48ff24177d04b7ce429840b102e Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:54 +0300
Subject: net: enetc: include MAC Merge / FP registers in register dump

These have been useful in debugging various problems related to frame
preemption, so make them available through ethtool --register-dump for
later too.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index 838a92131963..e993ed04ab57 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -32,6 +32,12 @@ static const u32 enetc_port_regs[] = {
 	ENETC_PM0_CMD_CFG, ENETC_PM0_MAXFRM, ENETC_PM0_IF_MODE
 };
 
+static const u32 enetc_port_mm_regs[] = {
+	ENETC_MMCSR, ENETC_PFPMR, ENETC_PTCFPR(0), ENETC_PTCFPR(1),
+	ENETC_PTCFPR(2), ENETC_PTCFPR(3), ENETC_PTCFPR(4), ENETC_PTCFPR(5),
+	ENETC_PTCFPR(6), ENETC_PTCFPR(7),
+};
+
 static int enetc_get_reglen(struct net_device *ndev)
 {
 	struct enetc_ndev_priv *priv = netdev_priv(ndev);
@@ -45,6 +51,9 @@ static int enetc_get_reglen(struct net_device *ndev)
 	if (hw->port)
 		len += ARRAY_SIZE(enetc_port_regs);
 
+	if (hw->port && !!(priv->si->hw_features & ENETC_SI_F_QBU))
+		len += ARRAY_SIZE(enetc_port_mm_regs);
+
 	len *= sizeof(u32) * 2; /* store 2 entries per reg: addr and value */
 
 	return len;
@@ -90,6 +99,14 @@ static void enetc_get_regs(struct net_device *ndev, struct ethtool_regs *regs,
 		*buf++ = addr;
 		*buf++ = enetc_rd(hw, addr);
 	}
+
+	if (priv->si->hw_features & ENETC_SI_F_QBU) {
+		for (i = 0; i < ARRAY_SIZE(enetc_port_mm_regs); i++) {
+			addr = ENETC_PORT_BASE + enetc_port_mm_regs[i];
+			*buf++ = addr;
+			*buf++ = enetc_rd(hw, addr);
+		}
+	}
 }
 
 static const struct {
-- 
cgit v1.2.3-70-g09d2


From 35b288d6e3d467f6f7bf466fee6184b275999f14 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:55 +0300
Subject: net: ethtool: mm: sanitize some UAPI configurations

The verify-enabled boolean (ETHTOOL_A_MM_VERIFY_ENABLED) was intended to
be a sub-setting of tx-enabled (ETHTOOL_A_MM_TX_ENABLED). IOW, MAC Merge
TX can be enabled with or without verification, but verification with TX
disabled makes no sense.

The pmac-enabled boolean (ETHTOOL_A_MM_PMAC_ENABLED) was intended to be
a global toggle from an API perspective, whereas tx-enabled just handles
the TX direction. IOW, the pMAC can be enabled with or without TX, but
it doesn't make sense to enable TX if the pMAC is not enabled.

Add two checks which sanitize and reject these invalid cases.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/mm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/net/ethtool/mm.c b/net/ethtool/mm.c
index e00d7d5cea7e..4058a557b5a4 100644
--- a/net/ethtool/mm.c
+++ b/net/ethtool/mm.c
@@ -214,6 +214,16 @@ static int ethnl_set_mm(struct ethnl_req_info *req_info, struct genl_info *info)
 		return -ERANGE;
 	}
 
+	if (cfg.verify_enabled && !cfg.tx_enabled) {
+		NL_SET_ERR_MSG(extack, "Verification requires TX enabled");
+		return -EINVAL;
+	}
+
+	if (cfg.tx_enabled && !cfg.pmac_enabled) {
+		NL_SET_ERR_MSG(extack, "TX enabled requires pMAC enabled");
+		return -EINVAL;
+	}
+
 	ret = dev->ethtool_ops->set_mm(dev, &cfg, extack);
 	return ret < 0 ? ret : 1;
 }
-- 
cgit v1.2.3-70-g09d2


From 54e906f1639e195bcb52725aab6814e2af922540 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Tue, 18 Apr 2023 14:14:56 +0300
Subject: selftests: forwarding: sch_tbf_*: Add a pre-run hook

The driver-specific wrappers of these selftests invoke bail_on_lldpad to
make sure that LLDPAD doesn't trample the configuration. The function
bail_on_lldpad is going to move to lib.sh in the next patch. With that, it
won't be visible for the wrappers before sourcing the framework script. And
after sourcing it, it is too late: the selftest will have run by then.

One option might be to source NUM_NETIFS=0 lib.sh from the wrapper, but
even if that worked (it might, it might not), that seems cumbersome. lib.sh
is doing fair amount of stuff, and even if it works today, it does not look
particularly solid as a solution.

Instead, introduce a hook, sch_tbf_pre_hook(), that when available, gets
invoked. Move the bail to the hook.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh  | 6 +++++-
 tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh | 6 +++++-
 tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh | 6 +++++-
 tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh | 4 ++++
 tools/testing/selftests/net/forwarding/sch_tbf_root.sh    | 4 ++++
 5 files changed, 23 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
index c6ce0b448bf3..b9b4cdf14ceb 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
@@ -2,7 +2,11 @@
 # SPDX-License-Identifier: GPL-2.0
 
 source qos_lib.sh
-bail_on_lldpad
+
+sch_tbf_pre_hook()
+{
+	bail_on_lldpad
+}
 
 lib_dir=$(dirname $0)/../../../net/forwarding
 TCFLAGS=skip_sw
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
index 8d245f331619..dff9810ee04f 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
@@ -2,7 +2,11 @@
 # SPDX-License-Identifier: GPL-2.0
 
 source qos_lib.sh
-bail_on_lldpad
+
+sch_tbf_pre_hook()
+{
+	bail_on_lldpad
+}
 
 lib_dir=$(dirname $0)/../../../net/forwarding
 TCFLAGS=skip_sw
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
index 013886061f15..75406bd7036e 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
@@ -2,7 +2,11 @@
 # SPDX-License-Identifier: GPL-2.0
 
 source qos_lib.sh
-bail_on_lldpad
+
+sch_tbf_pre_hook()
+{
+	bail_on_lldpad
+}
 
 lib_dir=$(dirname $0)/../../../net/forwarding
 TCFLAGS=skip_sw
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh b/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh
index 75a37c189ef3..df9bcd6a811a 100644
--- a/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_etsprio.sh
@@ -57,6 +57,10 @@ tbf_root_test()
 	tc qdisc del dev $swp2 root
 }
 
+if type -t sch_tbf_pre_hook >/dev/null; then
+	sch_tbf_pre_hook
+fi
+
 trap cleanup EXIT
 
 setup_prepare
diff --git a/tools/testing/selftests/net/forwarding/sch_tbf_root.sh b/tools/testing/selftests/net/forwarding/sch_tbf_root.sh
index 72aa21ba88c7..96c997be0d03 100755
--- a/tools/testing/selftests/net/forwarding/sch_tbf_root.sh
+++ b/tools/testing/selftests/net/forwarding/sch_tbf_root.sh
@@ -23,6 +23,10 @@ tbf_test()
 	tc qdisc del dev $swp2 root
 }
 
+if type -t sch_tbf_pre_hook >/dev/null; then
+	sch_tbf_pre_hook
+fi
+
 trap cleanup EXIT
 
 setup_prepare
-- 
cgit v1.2.3-70-g09d2


From 8fcac79270cae2e07e5475b682bc425b71e80595 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Tue, 18 Apr 2023 14:14:57 +0300
Subject: selftests: forwarding: generalize bail_on_lldpad from mlxsw

mlxsw selftests often invoke a bail_on_lldpad() helper to make sure LLDPAD
is not running, to prevent conflicts between the QoS configuration applied
through TC or DCB command line tool, and the DCB configuration that LLDPAD
might apply. This helper might be useful to others. Move the function to
lib.sh, and parameterize to make reusable in other contexts.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../selftests/drivers/net/mlxsw/qos_headroom.sh    |  3 +--
 .../testing/selftests/drivers/net/mlxsw/qos_lib.sh | 28 -------------------
 .../testing/selftests/drivers/net/mlxsw/qos_pfc.sh |  3 +--
 .../testing/selftests/drivers/net/mlxsw/sch_ets.sh |  3 +--
 .../selftests/drivers/net/mlxsw/sch_red_core.sh    |  1 -
 .../selftests/drivers/net/mlxsw/sch_red_ets.sh     |  2 +-
 .../selftests/drivers/net/mlxsw/sch_red_root.sh    |  2 +-
 .../selftests/drivers/net/mlxsw/sch_tbf_ets.sh     |  4 +--
 .../selftests/drivers/net/mlxsw/sch_tbf_prio.sh    |  4 +--
 .../selftests/drivers/net/mlxsw/sch_tbf_root.sh    |  4 +--
 tools/testing/selftests/net/forwarding/lib.sh      | 31 ++++++++++++++++++++++
 11 files changed, 39 insertions(+), 46 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_headroom.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_headroom.sh
index 3569ff45f7d5..88162b4027c0 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_headroom.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_headroom.sh
@@ -18,7 +18,6 @@ lib_dir=$(dirname $0)/../../../net/forwarding
 NUM_NETIFS=0
 source $lib_dir/lib.sh
 source $lib_dir/devlink_lib.sh
-source qos_lib.sh
 
 swp=$NETIF_NO_CABLE
 
@@ -371,7 +370,7 @@ test_tc_int_buf()
 	tc qdisc delete dev $swp root
 }
 
-bail_on_lldpad
+bail_on_lldpad "configure DCB" "configure Qdiscs"
 
 trap cleanup EXIT
 setup_wait
diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
index faa51012cdac..5ad092b9bf10 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh
@@ -54,31 +54,3 @@ measure_rate()
 	echo $ir $er
 	return $ret
 }
-
-bail_on_lldpad()
-{
-	if systemctl is-active --quiet lldpad; then
-
-		cat >/dev/stderr <<-EOF
-		WARNING: lldpad is running
-
-			lldpad will likely configure DCB, and this test will
-			configure Qdiscs. mlxsw does not support both at the
-			same time, one of them is arbitrarily going to overwrite
-			the other. That will cause spurious failures (or,
-			unlikely, passes) of this test.
-		EOF
-
-		if [[ -z $ALLOW_LLDPAD ]]; then
-			cat >/dev/stderr <<-EOF
-
-				If you want to run the test anyway, please set
-				an environment variable ALLOW_LLDPAD to a
-				non-empty string.
-			EOF
-			exit 1
-		else
-			return
-		fi
-	fi
-}
diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_pfc.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_pfc.sh
index f9858e221996..42ce602d8d49 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_pfc.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_pfc.sh
@@ -79,7 +79,6 @@ lib_dir=$(dirname $0)/../../../net/forwarding
 NUM_NETIFS=6
 source $lib_dir/lib.sh
 source $lib_dir/devlink_lib.sh
-source qos_lib.sh
 
 _1KB=1000
 _100KB=$((100 * _1KB))
@@ -393,7 +392,7 @@ test_qos_pfc()
 	log_test "PFC"
 }
 
-bail_on_lldpad
+bail_on_lldpad "configure DCB" "configure Qdiscs"
 
 trap cleanup EXIT
 setup_prepare
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh
index ceaa76b17a43..139175fd03e7 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_ets.sh
@@ -5,7 +5,6 @@
 lib_dir=$(dirname $0)/../../../net/forwarding
 source $lib_dir/sch_ets_core.sh
 source $lib_dir/devlink_lib.sh
-source qos_lib.sh
 
 ALL_TESTS="
 	ping_ipv4
@@ -78,5 +77,5 @@ collect_stats()
 	done
 }
 
-bail_on_lldpad
+bail_on_lldpad "configure DCB" "configure Qdiscs"
 ets_run
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh
index 45b41b8f3232..299e06a5808c 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh
@@ -74,7 +74,6 @@ lib_dir=$(dirname $0)/../../../net/forwarding
 source $lib_dir/lib.sh
 source $lib_dir/devlink_lib.sh
 source mlxsw_lib.sh
-source qos_lib.sh
 
 ipaddr()
 {
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh
index 0d01c7cd82a1..8ecddafa79b3 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh
@@ -166,7 +166,7 @@ ecn_mirror_test()
 	uninstall_qdisc
 }
 
-bail_on_lldpad
+bail_on_lldpad "configure DCB" "configure Qdiscs"
 
 trap cleanup EXIT
 setup_prepare
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_red_root.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_red_root.sh
index 860205338e6f..159108d02895 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_red_root.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_red_root.sh
@@ -73,7 +73,7 @@ red_mirror_test()
 	uninstall_qdisc
 }
 
-bail_on_lldpad
+bail_on_lldpad "configure DCB" "configure Qdiscs"
 
 trap cleanup EXIT
 setup_prepare
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
index b9b4cdf14ceb..ecc3664376b3 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_ets.sh
@@ -1,11 +1,9 @@
 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
-source qos_lib.sh
-
 sch_tbf_pre_hook()
 {
-	bail_on_lldpad
+	bail_on_lldpad "configure DCB" "configure Qdiscs"
 }
 
 lib_dir=$(dirname $0)/../../../net/forwarding
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
index dff9810ee04f..2e0a4efb1703 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_prio.sh
@@ -1,11 +1,9 @@
 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
-source qos_lib.sh
-
 sch_tbf_pre_hook()
 {
-	bail_on_lldpad
+	bail_on_lldpad "configure DCB" "configure Qdiscs"
 }
 
 lib_dir=$(dirname $0)/../../../net/forwarding
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
index 75406bd7036e..6679a338dfc4 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_tbf_root.sh
@@ -1,11 +1,9 @@
 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
-source qos_lib.sh
-
 sch_tbf_pre_hook()
 {
-	bail_on_lldpad
+	bail_on_lldpad "configure DCB" "configure Qdiscs"
 }
 
 lib_dir=$(dirname $0)/../../../net/forwarding
diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index d47499ba81c7..efd48e1cadd2 100755
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -1887,3 +1887,34 @@ mldv1_done_get()
 
 	payload_template_expand_checksum "$hbh$icmpv6" $checksum
 }
+
+bail_on_lldpad()
+{
+	local reason1="$1"; shift
+	local reason2="$1"; shift
+
+	if systemctl is-active --quiet lldpad; then
+
+		cat >/dev/stderr <<-EOF
+		WARNING: lldpad is running
+
+			lldpad will likely $reason1, and this test will
+			$reason2. Both are not supported at the same time,
+			one of them is arbitrarily going to overwrite the
+			other. That will cause spurious failures (or, unlikely,
+			passes) of this test.
+		EOF
+
+		if [[ -z $ALLOW_LLDPAD ]]; then
+			cat >/dev/stderr <<-EOF
+
+				If you want to run the test anyway, please set
+				an environment variable ALLOW_LLDPAD to a
+				non-empty string.
+			EOF
+			exit 1
+		else
+			return
+		fi
+	fi
+}
-- 
cgit v1.2.3-70-g09d2


From b5bf7126a6a0e706ef0e6c6cd22b2ebc23bf54a2 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:58 +0300
Subject: selftests: forwarding: introduce helper for standard ethtool counters

Counters for the MAC Merge layer and preemptible MAC have standardized
so far on using structured ethtool stats as opposed to the driver
specific names and meanings.

Benefit from that rare opportunity and introduce a helper to lib.sh for
querying standardized counters, in the hope that these will take off for
other uses as well.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/lib.sh | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index efd48e1cadd2..36e47c9d7cca 100755
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -787,6 +787,17 @@ ethtool_stats_get()
 	ethtool -S $dev | grep "^ *$stat:" | head -n 1 | cut -d: -f2
 }
 
+ethtool_std_stats_get()
+{
+	local dev=$1; shift
+	local grp=$1; shift
+	local name=$1; shift
+	local src=$1; shift
+
+	ethtool --json -S $dev --groups $grp -- --src $src | \
+		jq '.[]."'"$grp"'"."'$name'"'
+}
+
 qdisc_stats_get()
 {
 	local dev=$1; shift
-- 
cgit v1.2.3-70-g09d2


From e6991384ace5c9cbbe453ad1d8998b0cdaf222f5 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 14:14:59 +0300
Subject: selftests: forwarding: add a test for MAC Merge layer

The MAC Merge layer (IEEE 802.3-2018 clause 99) does all the heavy
lifting for Frame Preemption (IEEE 802.1Q-2018 clause 6.7.2), a TSN
feature for minimizing latency.

Preemptible traffic is different on the wire from normal traffic in
incompatible ways. If we send a preemptible packet and the link partner
doesn't support preemption, it will drop it as an error frame and we
will never know. The MAC Merge layer has a control plane of its own,
which can be manipulated (using ethtool) in order to negotiate this
capability with the link partner (through LLDP).

Actually the TLV format for LLDP solves this problem only partly,
because both partners only advertise:
- if they support preemption (RX and TX)
- if they have enabled preemption (TX)
so we cannot tell the link partner what to do - we cannot force it to
enable reception of our preemptible packets.

That is fully solved by the verification feature, where the local device
generates some small probe frames which look like preemptible frames
with no useful content, and the link partner is obliged to respond to
them if it supports the standard. If the verification times out, we know
that preemption isn't active in our TX direction on the link.

Having clarified the definition, this selftest exercises the manual
(ethtool) configuration path of 2 link partners (with and without
verification), and the LLDP code path, using the openlldp project.

The test also verifies the TX activity of the MAC Merge layer by
sending traffic through a traffic class configured as preemptible
(using mqprio). There isn't a good way to make this really portable
(user space cannot find out how many traffic classes there are for
a device), but I chose num_tc 4 here, that should work reasonably well.
I also know that some devices (stmmac) only permit TXQ0 to be
preemptible, so this is why PREEMPTIBLE_PRIO was strategically chosen
as 0. Even if other hardware is more configurable, this test should
cover the baseline.

This is not really a "forwarding" selftest, but I put it near the other
"ethtool" selftests.

$ ./ethtool_mm.sh eno0 swp0
TEST: Manual configuration with verification: eno0 to swp0          [ OK ]
TEST: Manual configuration with verification: swp0 to eno0          [ OK ]
TEST: Manual configuration without verification: eno0 to swp0       [ OK ]
TEST: Manual configuration without verification: swp0 to eno0       [ OK ]
TEST: Manual configuration with failed verification: eno0 to swp0   [ OK ]
TEST: Manual configuration with failed verification: swp0 to eno0   [ OK ]
TEST: LLDP                                                          [ OK ]

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/Makefile    |   1 +
 .../testing/selftests/net/forwarding/ethtool_mm.sh | 288 +++++++++++++++++++++
 tools/testing/selftests/net/forwarding/lib.sh      |  18 ++
 3 files changed, 307 insertions(+)
 create mode 100755 tools/testing/selftests/net/forwarding/ethtool_mm.sh

diff --git a/tools/testing/selftests/net/forwarding/Makefile b/tools/testing/selftests/net/forwarding/Makefile
index 236f6b796a52..a474c60fe348 100644
--- a/tools/testing/selftests/net/forwarding/Makefile
+++ b/tools/testing/selftests/net/forwarding/Makefile
@@ -15,6 +15,7 @@ TEST_PROGS = bridge_igmp.sh \
 	custom_multipath_hash.sh \
 	dual_vxlan_bridge.sh \
 	ethtool_extended_state.sh \
+	ethtool_mm.sh \
 	ethtool.sh \
 	gre_custom_multipath_hash.sh \
 	gre_inner_v4_multipath.sh \
diff --git a/tools/testing/selftests/net/forwarding/ethtool_mm.sh b/tools/testing/selftests/net/forwarding/ethtool_mm.sh
new file mode 100755
index 000000000000..c580ad623848
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/ethtool_mm.sh
@@ -0,0 +1,288 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS="
+	manual_with_verification_h1_to_h2
+	manual_with_verification_h2_to_h1
+	manual_without_verification_h1_to_h2
+	manual_without_verification_h2_to_h1
+	manual_failed_verification_h1_to_h2
+	manual_failed_verification_h2_to_h1
+	lldp
+"
+
+NUM_NETIFS=2
+REQUIRE_MZ=no
+PREEMPTIBLE_PRIO=0
+source lib.sh
+
+traffic_test()
+{
+	local if=$1; shift
+	local src=$1; shift
+	local num_pkts=10000
+	local before=
+	local after=
+	local delta=
+
+	before=$(ethtool_std_stats_get $if "eth-mac" "FramesTransmittedOK" $src)
+
+	$MZ $if -q -c $num_pkts -p 64 -b bcast -t ip -R $PREEMPTIBLE_PRIO
+
+	after=$(ethtool_std_stats_get $if "eth-mac" "FramesTransmittedOK" $src)
+
+	delta=$((after - before))
+
+	# Allow an extra 1% tolerance for random packets sent by the stack
+	[ $delta -ge $num_pkts ] && [ $delta -le $((num_pkts + 100)) ]
+}
+
+manual_with_verification()
+{
+	local tx=$1; shift
+	local rx=$1; shift
+
+	RET=0
+
+	# It isn't completely clear from IEEE 802.3-2018 Figure 99-5: Transmit
+	# Processing state diagram whether the "send_r" variable (send response
+	# to verification frame) should be taken into consideration while the
+	# MAC Merge TX direction is disabled. That being said, at least the
+	# NXP ENETC does not, and requires tx-enabled on in order to respond to
+	# the link partner's verification frames.
+	ethtool --set-mm $rx tx-enabled on
+	ethtool --set-mm $tx verify-enabled on tx-enabled on
+
+	# Wait for verification to finish
+	sleep 1
+
+	ethtool --json --show-mm $tx | jq -r '.[]."verify-status"' | \
+		grep -q 'SUCCEEDED'
+	check_err "$?" "Verification did not succeed"
+
+	ethtool --json --show-mm $tx | jq -r '.[]."tx-active"' | grep -q 'true'
+	check_err "$?" "pMAC TX is not active"
+
+	traffic_test $tx "pmac"
+	check_err "$?" "Traffic did not get sent through $tx's pMAC"
+
+	ethtool --set-mm $tx verify-enabled off tx-enabled off
+	ethtool --set-mm $rx tx-enabled off
+
+	log_test "Manual configuration with verification: $tx to $rx"
+}
+
+manual_with_verification_h1_to_h2()
+{
+	manual_with_verification $h1 $h2
+}
+
+manual_with_verification_h2_to_h1()
+{
+	manual_with_verification $h2 $h1
+}
+
+manual_without_verification()
+{
+	local tx=$1; shift
+	local rx=$1; shift
+
+	RET=0
+
+	ethtool --set-mm $tx verify-enabled off tx-enabled on
+
+	ethtool --json --show-mm $tx | jq -r '.[]."verify-status"' | \
+		grep -q 'DISABLED'
+	check_err "$?" "Verification is not disabled"
+
+	ethtool --json --show-mm $tx | jq -r '.[]."tx-active"' | grep -q 'true'
+	check_err "$?" "pMAC TX is not active"
+
+	traffic_test $tx "pmac"
+	check_err "$?" "Traffic did not get sent through $tx's pMAC"
+
+	ethtool --set-mm $tx verify-enabled off tx-enabled off
+
+	log_test "Manual configuration without verification: $tx to $rx"
+}
+
+manual_without_verification_h1_to_h2()
+{
+	manual_without_verification $h1 $h2
+}
+
+manual_without_verification_h2_to_h1()
+{
+	manual_without_verification $h2 $h1
+}
+
+manual_failed_verification()
+{
+	local tx=$1; shift
+	local rx=$1; shift
+
+	RET=0
+
+	ethtool --set-mm $rx pmac-enabled off
+	ethtool --set-mm $tx verify-enabled on tx-enabled on
+
+	# Wait for verification to time out
+	sleep 1
+
+	ethtool --json --show-mm $tx | jq -r '.[]."verify-status"' | \
+		grep -q 'SUCCEEDED'
+	check_fail "$?" "Verification succeeded when it shouldn't have"
+
+	ethtool --json --show-mm $tx | jq -r '.[]."tx-active"' | grep -q 'true'
+	check_fail "$?" "pMAC TX is active when it shouldn't have"
+
+	traffic_test $tx "emac"
+	check_err "$?" "Traffic did not get sent through $tx's eMAC"
+
+	ethtool --set-mm $tx verify-enabled off tx-enabled off
+	ethtool --set-mm $rx pmac-enabled on
+
+	log_test "Manual configuration with failed verification: $tx to $rx"
+}
+
+manual_failed_verification_h1_to_h2()
+{
+	manual_failed_verification $h1 $h2
+}
+
+manual_failed_verification_h2_to_h1()
+{
+	manual_failed_verification $h2 $h1
+}
+
+lldp_change_add_frag_size()
+{
+	local add_frag_size=$1
+
+	lldptool -T -i $h1 -V addEthCaps addFragSize=$add_frag_size >/dev/null
+	# Wait for TLVs to be received
+	sleep 2
+	lldptool -i $h2 -t -n -V addEthCaps | \
+		grep -q "Additional fragment size: $add_frag_size"
+}
+
+lldp()
+{
+	RET=0
+
+	systemctl start lldpad
+
+	# Configure the interfaces to receive and transmit LLDPDUs
+	lldptool -L -i $h1 adminStatus=rxtx >/dev/null
+	lldptool -L -i $h2 adminStatus=rxtx >/dev/null
+
+	# Enable the transmission of Additional Ethernet Capabilities TLV
+	lldptool -T -i $h1 -V addEthCaps enableTx=yes >/dev/null
+	lldptool -T -i $h2 -V addEthCaps enableTx=yes >/dev/null
+
+	# Wait for TLVs to be received
+	sleep 2
+
+	lldptool -i $h1 -t -n -V addEthCaps | \
+		grep -q "Preemption capability active"
+	check_err "$?" "$h1 pMAC TX is not active"
+
+	lldptool -i $h2 -t -n -V addEthCaps | \
+		grep -q "Preemption capability active"
+	check_err "$?" "$h2 pMAC TX is not active"
+
+	lldp_change_add_frag_size 3
+	check_err "$?" "addFragSize 3"
+
+	lldp_change_add_frag_size 2
+	check_err "$?" "addFragSize 2"
+
+	lldp_change_add_frag_size 1
+	check_err "$?" "addFragSize 1"
+
+	lldp_change_add_frag_size 0
+	check_err "$?" "addFragSize 0"
+
+	traffic_test $h1 "pmac"
+	check_err "$?" "Traffic did not get sent through $h1's pMAC"
+
+	traffic_test $h2 "pmac"
+	check_err "$?" "Traffic did not get sent through $h2's pMAC"
+
+	systemctl stop lldpad
+
+	log_test "LLDP"
+}
+
+h1_create()
+{
+	ip link set dev $h1 up
+
+	tc qdisc add dev $h1 root mqprio num_tc 4 map 0 1 2 3 \
+		queues 1@0 1@1 1@2 1@3 \
+		fp P E E E \
+		hw 1
+
+	ethtool --set-mm $h1 pmac-enabled on tx-enabled off verify-enabled off
+}
+
+h2_create()
+{
+	ip link set dev $h2 up
+
+	ethtool --set-mm $h2 pmac-enabled on tx-enabled off verify-enabled off
+
+	tc qdisc add dev $h2 root mqprio num_tc 4 map 0 1 2 3 \
+		queues 1@0 1@1 1@2 1@3 \
+		fp P E E E \
+		hw 1
+}
+
+h1_destroy()
+{
+	ethtool --set-mm $h1 pmac-enabled off tx-enabled off verify-enabled off
+
+	tc qdisc del dev $h1 root
+
+	ip link set dev $h1 down
+}
+
+h2_destroy()
+{
+	tc qdisc del dev $h2 root
+
+	ethtool --set-mm $h2 pmac-enabled off tx-enabled off verify-enabled off
+
+	ip link set dev $h2 down
+}
+
+setup_prepare()
+{
+	check_ethtool_mm_support
+	check_tc_fp_support
+	require_command lldptool
+	bail_on_lldpad "autoconfigure the MAC Merge layer" "configure it manually"
+
+	h1=${NETIFS[p1]}
+	h2=${NETIFS[p2]}
+
+	h1_create
+	h2_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	h2_destroy
+	h1_destroy
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index 36e47c9d7cca..057c3d0ad620 100755
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -120,6 +120,15 @@ check_tc_action_hw_stats_support()
 	fi
 }
 
+check_tc_fp_support()
+{
+	tc qdisc add dev lo mqprio help 2>&1 | grep -q "fp "
+	if [[ $? -ne 0 ]]; then
+		echo "SKIP: iproute2 too old; tc is missing frame preemption support"
+		exit $ksft_skip
+	fi
+}
+
 check_ethtool_lanes_support()
 {
 	ethtool --help 2>&1| grep lanes &> /dev/null
@@ -129,6 +138,15 @@ check_ethtool_lanes_support()
 	fi
 }
 
+check_ethtool_mm_support()
+{
+	ethtool --help 2>&1| grep -- '--show-mm' &> /dev/null
+	if [[ $? -ne 0 ]]; then
+		echo "SKIP: ethtool too old; it is missing MAC Merge layer support"
+		exit $ksft_skip
+	fi
+}
+
 check_locked_port_support()
 {
 	if ! bridge -d link show | grep -q " locked"; then
-- 
cgit v1.2.3-70-g09d2


From 7ab75456be144a354fbb3df1516d82fc24d3d67d Mon Sep 17 00:00:00 2001
From: Mahesh Bandewar <maheshb@google.com>
Date: Tue, 18 Apr 2023 18:32:38 -0700
Subject: ipv6: add icmpv6_error_anycast_as_unicast for ICMPv6
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

ICMPv6 error packets are not sent to the anycast destinations and this
prevents things like traceroute from working. So create a setting similar
to ECHO when dealing with Anycast sources (icmpv6_echo_ignore_anycast).

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Maciej Żenczykowski <maze@google.com>
Link: https://lore.kernel.org/r/20230419013238.2691167-1-maheshb@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/ip-sysctl.rst |  7 +++++++
 include/net/netns/ipv6.h               |  1 +
 net/ipv6/af_inet6.c                    |  1 +
 net/ipv6/icmp.c                        | 15 +++++++++++++--
 4 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst
index 58a78a316697..6ec06a33688a 100644
--- a/Documentation/networking/ip-sysctl.rst
+++ b/Documentation/networking/ip-sysctl.rst
@@ -2721,6 +2721,13 @@ echo_ignore_anycast - BOOLEAN
 
 	Default: 0
 
+error_anycast_as_unicast - BOOLEAN
+	If set to 1, then the kernel will respond with ICMP Errors
+	resulting from requests sent to it over the IPv6 protocol destined
+	to anycast address essentially treating anycast as unicast.
+
+	Default: 0
+
 xfrm6_gc_thresh - INTEGER
 	(Obsolete since linux-4.14)
 	The threshold at which we will start garbage collecting for IPv6
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index b4af4837d80b..3cceb3e9320b 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -55,6 +55,7 @@ struct netns_sysctl_ipv6 {
 	u64 ioam6_id_wide;
 	bool skip_notify_on_dev_down;
 	u8 fib_notify_on_flag_change;
+	u8 icmpv6_error_anycast_as_unicast;
 };
 
 struct netns_ipv6 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e1b679a590c9..2bbf13216a3d 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -952,6 +952,7 @@ static int __net_init inet6_net_init(struct net *net)
 	net->ipv6.sysctl.icmpv6_echo_ignore_all = 0;
 	net->ipv6.sysctl.icmpv6_echo_ignore_multicast = 0;
 	net->ipv6.sysctl.icmpv6_echo_ignore_anycast = 0;
+	net->ipv6.sysctl.icmpv6_error_anycast_as_unicast = 0;
 
 	/* By default, rate limit error messages.
 	 * Except for pmtu discovery, it would break it.
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 1f53f2a74480..9edf1f45b1ed 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -362,9 +362,10 @@ static struct dst_entry *icmpv6_route_lookup(struct net *net,
 
 	/*
 	 * We won't send icmp if the destination is known
-	 * anycast.
+	 * anycast unless we need to treat anycast as unicast.
 	 */
-	if (ipv6_anycast_destination(dst, &fl6->daddr)) {
+	if (!READ_ONCE(net->ipv6.sysctl.icmpv6_error_anycast_as_unicast) &&
+	    ipv6_anycast_destination(dst, &fl6->daddr)) {
 		net_dbg_ratelimited("icmp6_send: acast source\n");
 		dst_release(dst);
 		return ERR_PTR(-EINVAL);
@@ -1195,6 +1196,15 @@ static struct ctl_table ipv6_icmp_table_template[] = {
 		.mode		= 0644,
 		.proc_handler = proc_do_large_bitmap,
 	},
+	{
+		.procname	= "error_anycast_as_unicast",
+		.data		= &init_net.ipv6.sysctl.icmpv6_error_anycast_as_unicast,
+		.maxlen		= sizeof(u8),
+		.mode		= 0644,
+		.proc_handler	= proc_dou8vec_minmax,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_ONE,
+	},
 	{ },
 };
 
@@ -1212,6 +1222,7 @@ struct ctl_table * __net_init ipv6_icmp_sysctl_init(struct net *net)
 		table[2].data = &net->ipv6.sysctl.icmpv6_echo_ignore_multicast;
 		table[3].data = &net->ipv6.sysctl.icmpv6_echo_ignore_anycast;
 		table[4].data = &net->ipv6.sysctl.icmpv6_ratemask_ptr;
+		table[5].data = &net->ipv6.sysctl.icmpv6_error_anycast_as_unicast;
 	}
 	return table;
 }
-- 
cgit v1.2.3-70-g09d2


From 5b8285cca6fed9bc5baabe2e5699a5a5c0d96371 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 19 Apr 2023 14:52:52 +0200
Subject: net: move dropreason.h to dropreason-core.h

This will, after the next patch, hold only the core
drop reasons and minimal infrastructure. Fix a small
kernel-doc issue while at it, to avoid the move
triggering a checker.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h     |   2 +-
 include/linux/skbuff.h        |   2 +-
 include/net/dropreason-core.h | 364 ++++++++++++++++++++++++++++++++++++++++++
 include/net/dropreason.h      | 363 -----------------------------------------
 include/net/inet_frag.h       |   2 +-
 5 files changed, 367 insertions(+), 366 deletions(-)
 create mode 100644 include/net/dropreason-core.h
 delete mode 100644 include/net/dropreason.h

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 203c0df2046c..a6a3e9457d6c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -52,7 +52,7 @@
 #include <linux/rbtree.h>
 #include <net/net_trackers.h>
 #include <net/net_debug.h>
-#include <net/dropreason.h>
+#include <net/dropreason-core.h>
 
 struct netpoll_info;
 struct device;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 03aa7ed076f0..eb9e7bb76fa6 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -37,7 +37,7 @@
 #include <linux/netfilter/nf_conntrack_common.h>
 #endif
 #include <net/net_debug.h>
-#include <net/dropreason.h>
+#include <net/dropreason-core.h>
 
 /**
  * DOC: skb checksums
diff --git a/include/net/dropreason-core.h b/include/net/dropreason-core.h
new file mode 100644
index 000000000000..ade6d5b9186c
--- /dev/null
+++ b/include/net/dropreason-core.h
@@ -0,0 +1,364 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+
+#ifndef _LINUX_DROPREASON_CORE_H
+#define _LINUX_DROPREASON_CORE_H
+
+#define DEFINE_DROP_REASON(FN, FNe)	\
+	FN(NOT_SPECIFIED)		\
+	FN(NO_SOCKET)			\
+	FN(PKT_TOO_SMALL)		\
+	FN(TCP_CSUM)			\
+	FN(SOCKET_FILTER)		\
+	FN(UDP_CSUM)			\
+	FN(NETFILTER_DROP)		\
+	FN(OTHERHOST)			\
+	FN(IP_CSUM)			\
+	FN(IP_INHDR)			\
+	FN(IP_RPFILTER)			\
+	FN(UNICAST_IN_L2_MULTICAST)	\
+	FN(XFRM_POLICY)			\
+	FN(IP_NOPROTO)			\
+	FN(SOCKET_RCVBUFF)		\
+	FN(PROTO_MEM)			\
+	FN(TCP_MD5NOTFOUND)		\
+	FN(TCP_MD5UNEXPECTED)		\
+	FN(TCP_MD5FAILURE)		\
+	FN(SOCKET_BACKLOG)		\
+	FN(TCP_FLAGS)			\
+	FN(TCP_ZEROWINDOW)		\
+	FN(TCP_OLD_DATA)		\
+	FN(TCP_OVERWINDOW)		\
+	FN(TCP_OFOMERGE)		\
+	FN(TCP_RFC7323_PAWS)		\
+	FN(TCP_INVALID_SEQUENCE)	\
+	FN(TCP_RESET)			\
+	FN(TCP_INVALID_SYN)		\
+	FN(TCP_CLOSE)			\
+	FN(TCP_FASTOPEN)		\
+	FN(TCP_OLD_ACK)			\
+	FN(TCP_TOO_OLD_ACK)		\
+	FN(TCP_ACK_UNSENT_DATA)		\
+	FN(TCP_OFO_QUEUE_PRUNE)		\
+	FN(TCP_OFO_DROP)		\
+	FN(IP_OUTNOROUTES)		\
+	FN(BPF_CGROUP_EGRESS)		\
+	FN(IPV6DISABLED)		\
+	FN(NEIGH_CREATEFAIL)		\
+	FN(NEIGH_FAILED)		\
+	FN(NEIGH_QUEUEFULL)		\
+	FN(NEIGH_DEAD)			\
+	FN(TC_EGRESS)			\
+	FN(QDISC_DROP)			\
+	FN(CPU_BACKLOG)			\
+	FN(XDP)				\
+	FN(TC_INGRESS)			\
+	FN(UNHANDLED_PROTO)		\
+	FN(SKB_CSUM)			\
+	FN(SKB_GSO_SEG)			\
+	FN(SKB_UCOPY_FAULT)		\
+	FN(DEV_HDR)			\
+	FN(DEV_READY)			\
+	FN(FULL_RING)			\
+	FN(NOMEM)			\
+	FN(HDR_TRUNC)			\
+	FN(TAP_FILTER)			\
+	FN(TAP_TXFILTER)		\
+	FN(ICMP_CSUM)			\
+	FN(INVALID_PROTO)		\
+	FN(IP_INADDRERRORS)		\
+	FN(IP_INNOROUTES)		\
+	FN(PKT_TOO_BIG)			\
+	FN(DUP_FRAG)			\
+	FN(FRAG_REASM_TIMEOUT)		\
+	FN(FRAG_TOO_FAR)		\
+	FN(TCP_MINTTL)			\
+	FN(IPV6_BAD_EXTHDR)		\
+	FN(IPV6_NDISC_FRAG)		\
+	FN(IPV6_NDISC_HOP_LIMIT)	\
+	FN(IPV6_NDISC_BAD_CODE)		\
+	FN(IPV6_NDISC_BAD_OPTIONS)	\
+	FN(IPV6_NDISC_NS_OTHERHOST)	\
+	FNe(MAX)
+
+/**
+ * enum skb_drop_reason - the reasons of skb drops
+ *
+ * The reason of skb drop, which is used in kfree_skb_reason().
+ */
+enum skb_drop_reason {
+	/**
+	 * @SKB_NOT_DROPPED_YET: skb is not dropped yet (used for no-drop case)
+	 */
+	SKB_NOT_DROPPED_YET = 0,
+	/** @SKB_CONSUMED: packet has been consumed */
+	SKB_CONSUMED,
+	/** @SKB_DROP_REASON_NOT_SPECIFIED: drop reason is not specified */
+	SKB_DROP_REASON_NOT_SPECIFIED,
+	/** @SKB_DROP_REASON_NO_SOCKET: socket not found */
+	SKB_DROP_REASON_NO_SOCKET,
+	/** @SKB_DROP_REASON_PKT_TOO_SMALL: packet size is too small */
+	SKB_DROP_REASON_PKT_TOO_SMALL,
+	/** @SKB_DROP_REASON_TCP_CSUM: TCP checksum error */
+	SKB_DROP_REASON_TCP_CSUM,
+	/** @SKB_DROP_REASON_SOCKET_FILTER: dropped by socket filter */
+	SKB_DROP_REASON_SOCKET_FILTER,
+	/** @SKB_DROP_REASON_UDP_CSUM: UDP checksum error */
+	SKB_DROP_REASON_UDP_CSUM,
+	/** @SKB_DROP_REASON_NETFILTER_DROP: dropped by netfilter */
+	SKB_DROP_REASON_NETFILTER_DROP,
+	/**
+	 * @SKB_DROP_REASON_OTHERHOST: packet don't belong to current host
+	 * (interface is in promisc mode)
+	 */
+	SKB_DROP_REASON_OTHERHOST,
+	/** @SKB_DROP_REASON_IP_CSUM: IP checksum error */
+	SKB_DROP_REASON_IP_CSUM,
+	/**
+	 * @SKB_DROP_REASON_IP_INHDR: there is something wrong with IP header (see
+	 * IPSTATS_MIB_INHDRERRORS)
+	 */
+	SKB_DROP_REASON_IP_INHDR,
+	/**
+	 * @SKB_DROP_REASON_IP_RPFILTER: IP rpfilter validate failed. see the
+	 * document for rp_filter in ip-sysctl.rst for more information
+	 */
+	SKB_DROP_REASON_IP_RPFILTER,
+	/**
+	 * @SKB_DROP_REASON_UNICAST_IN_L2_MULTICAST: destination address of L2 is
+	 * multicast, but L3 is unicast.
+	 */
+	SKB_DROP_REASON_UNICAST_IN_L2_MULTICAST,
+	/** @SKB_DROP_REASON_XFRM_POLICY: xfrm policy check failed */
+	SKB_DROP_REASON_XFRM_POLICY,
+	/** @SKB_DROP_REASON_IP_NOPROTO: no support for IP protocol */
+	SKB_DROP_REASON_IP_NOPROTO,
+	/** @SKB_DROP_REASON_SOCKET_RCVBUFF: socket receive buff is full */
+	SKB_DROP_REASON_SOCKET_RCVBUFF,
+	/**
+	 * @SKB_DROP_REASON_PROTO_MEM: proto memory limition, such as udp packet
+	 * drop out of udp_memory_allocated.
+	 */
+	SKB_DROP_REASON_PROTO_MEM,
+	/**
+	 * @SKB_DROP_REASON_TCP_MD5NOTFOUND: no MD5 hash and one expected,
+	 * corresponding to LINUX_MIB_TCPMD5NOTFOUND
+	 */
+	SKB_DROP_REASON_TCP_MD5NOTFOUND,
+	/**
+	 * @SKB_DROP_REASON_TCP_MD5UNEXPECTED: MD5 hash and we're not expecting
+	 * one, corresponding to LINUX_MIB_TCPMD5UNEXPECTED
+	 */
+	SKB_DROP_REASON_TCP_MD5UNEXPECTED,
+	/**
+	 * @SKB_DROP_REASON_TCP_MD5FAILURE: MD5 hash and its wrong, corresponding
+	 * to LINUX_MIB_TCPMD5FAILURE
+	 */
+	SKB_DROP_REASON_TCP_MD5FAILURE,
+	/**
+	 * @SKB_DROP_REASON_SOCKET_BACKLOG: failed to add skb to socket backlog (
+	 * see LINUX_MIB_TCPBACKLOGDROP)
+	 */
+	SKB_DROP_REASON_SOCKET_BACKLOG,
+	/** @SKB_DROP_REASON_TCP_FLAGS: TCP flags invalid */
+	SKB_DROP_REASON_TCP_FLAGS,
+	/**
+	 * @SKB_DROP_REASON_TCP_ZEROWINDOW: TCP receive window size is zero,
+	 * see LINUX_MIB_TCPZEROWINDOWDROP
+	 */
+	SKB_DROP_REASON_TCP_ZEROWINDOW,
+	/**
+	 * @SKB_DROP_REASON_TCP_OLD_DATA: the TCP data reveived is already
+	 * received before (spurious retrans may happened), see
+	 * LINUX_MIB_DELAYEDACKLOST
+	 */
+	SKB_DROP_REASON_TCP_OLD_DATA,
+	/**
+	 * @SKB_DROP_REASON_TCP_OVERWINDOW: the TCP data is out of window,
+	 * the seq of the first byte exceed the right edges of receive
+	 * window
+	 */
+	SKB_DROP_REASON_TCP_OVERWINDOW,
+	/**
+	 * @SKB_DROP_REASON_TCP_OFOMERGE: the data of skb is already in the ofo
+	 * queue, corresponding to LINUX_MIB_TCPOFOMERGE
+	 */
+	SKB_DROP_REASON_TCP_OFOMERGE,
+	/**
+	 * @SKB_DROP_REASON_TCP_RFC7323_PAWS: PAWS check, corresponding to
+	 * LINUX_MIB_PAWSESTABREJECTED
+	 */
+	SKB_DROP_REASON_TCP_RFC7323_PAWS,
+	/** @SKB_DROP_REASON_TCP_INVALID_SEQUENCE: Not acceptable SEQ field */
+	SKB_DROP_REASON_TCP_INVALID_SEQUENCE,
+	/** @SKB_DROP_REASON_TCP_RESET: Invalid RST packet */
+	SKB_DROP_REASON_TCP_RESET,
+	/**
+	 * @SKB_DROP_REASON_TCP_INVALID_SYN: Incoming packet has unexpected
+	 * SYN flag
+	 */
+	SKB_DROP_REASON_TCP_INVALID_SYN,
+	/** @SKB_DROP_REASON_TCP_CLOSE: TCP socket in CLOSE state */
+	SKB_DROP_REASON_TCP_CLOSE,
+	/** @SKB_DROP_REASON_TCP_FASTOPEN: dropped by FASTOPEN request socket */
+	SKB_DROP_REASON_TCP_FASTOPEN,
+	/** @SKB_DROP_REASON_TCP_OLD_ACK: TCP ACK is old, but in window */
+	SKB_DROP_REASON_TCP_OLD_ACK,
+	/** @SKB_DROP_REASON_TCP_TOO_OLD_ACK: TCP ACK is too old */
+	SKB_DROP_REASON_TCP_TOO_OLD_ACK,
+	/**
+	 * @SKB_DROP_REASON_TCP_ACK_UNSENT_DATA: TCP ACK for data we haven't
+	 * sent yet
+	 */
+	SKB_DROP_REASON_TCP_ACK_UNSENT_DATA,
+	/** @SKB_DROP_REASON_TCP_OFO_QUEUE_PRUNE: pruned from TCP OFO queue */
+	SKB_DROP_REASON_TCP_OFO_QUEUE_PRUNE,
+	/** @SKB_DROP_REASON_TCP_OFO_DROP: data already in receive queue */
+	SKB_DROP_REASON_TCP_OFO_DROP,
+	/** @SKB_DROP_REASON_IP_OUTNOROUTES: route lookup failed */
+	SKB_DROP_REASON_IP_OUTNOROUTES,
+	/**
+	 * @SKB_DROP_REASON_BPF_CGROUP_EGRESS: dropped by BPF_PROG_TYPE_CGROUP_SKB
+	 * eBPF program
+	 */
+	SKB_DROP_REASON_BPF_CGROUP_EGRESS,
+	/** @SKB_DROP_REASON_IPV6DISABLED: IPv6 is disabled on the device */
+	SKB_DROP_REASON_IPV6DISABLED,
+	/** @SKB_DROP_REASON_NEIGH_CREATEFAIL: failed to create neigh entry */
+	SKB_DROP_REASON_NEIGH_CREATEFAIL,
+	/** @SKB_DROP_REASON_NEIGH_FAILED: neigh entry in failed state */
+	SKB_DROP_REASON_NEIGH_FAILED,
+	/** @SKB_DROP_REASON_NEIGH_QUEUEFULL: arp_queue for neigh entry is full */
+	SKB_DROP_REASON_NEIGH_QUEUEFULL,
+	/** @SKB_DROP_REASON_NEIGH_DEAD: neigh entry is dead */
+	SKB_DROP_REASON_NEIGH_DEAD,
+	/** @SKB_DROP_REASON_TC_EGRESS: dropped in TC egress HOOK */
+	SKB_DROP_REASON_TC_EGRESS,
+	/**
+	 * @SKB_DROP_REASON_QDISC_DROP: dropped by qdisc when packet outputting (
+	 * failed to enqueue to current qdisc)
+	 */
+	SKB_DROP_REASON_QDISC_DROP,
+	/**
+	 * @SKB_DROP_REASON_CPU_BACKLOG: failed to enqueue the skb to the per CPU
+	 * backlog queue. This can be caused by backlog queue full (see
+	 * netdev_max_backlog in net.rst) or RPS flow limit
+	 */
+	SKB_DROP_REASON_CPU_BACKLOG,
+	/** @SKB_DROP_REASON_XDP: dropped by XDP in input path */
+	SKB_DROP_REASON_XDP,
+	/** @SKB_DROP_REASON_TC_INGRESS: dropped in TC ingress HOOK */
+	SKB_DROP_REASON_TC_INGRESS,
+	/** @SKB_DROP_REASON_UNHANDLED_PROTO: protocol not implemented or not supported */
+	SKB_DROP_REASON_UNHANDLED_PROTO,
+	/** @SKB_DROP_REASON_SKB_CSUM: sk_buff checksum computation error */
+	SKB_DROP_REASON_SKB_CSUM,
+	/** @SKB_DROP_REASON_SKB_GSO_SEG: gso segmentation error */
+	SKB_DROP_REASON_SKB_GSO_SEG,
+	/**
+	 * @SKB_DROP_REASON_SKB_UCOPY_FAULT: failed to copy data from user space,
+	 * e.g., via zerocopy_sg_from_iter() or skb_orphan_frags_rx()
+	 */
+	SKB_DROP_REASON_SKB_UCOPY_FAULT,
+	/** @SKB_DROP_REASON_DEV_HDR: device driver specific header/metadata is invalid */
+	SKB_DROP_REASON_DEV_HDR,
+	/**
+	 * @SKB_DROP_REASON_DEV_READY: the device is not ready to xmit/recv due to
+	 * any of its data structure that is not up/ready/initialized,
+	 * e.g., the IFF_UP is not set, or driver specific tun->tfiles[txq]
+	 * is not initialized
+	 */
+	SKB_DROP_REASON_DEV_READY,
+	/** @SKB_DROP_REASON_FULL_RING: ring buffer is full */
+	SKB_DROP_REASON_FULL_RING,
+	/** @SKB_DROP_REASON_NOMEM: error due to OOM */
+	SKB_DROP_REASON_NOMEM,
+	/**
+	 * @SKB_DROP_REASON_HDR_TRUNC: failed to trunc/extract the header from
+	 * networking data, e.g., failed to pull the protocol header from
+	 * frags via pskb_may_pull()
+	 */
+	SKB_DROP_REASON_HDR_TRUNC,
+	/**
+	 * @SKB_DROP_REASON_TAP_FILTER: dropped by (ebpf) filter directly attached
+	 * to tun/tap, e.g., via TUNSETFILTEREBPF
+	 */
+	SKB_DROP_REASON_TAP_FILTER,
+	/**
+	 * @SKB_DROP_REASON_TAP_TXFILTER: dropped by tx filter implemented at
+	 * tun/tap, e.g., check_filter()
+	 */
+	SKB_DROP_REASON_TAP_TXFILTER,
+	/** @SKB_DROP_REASON_ICMP_CSUM: ICMP checksum error */
+	SKB_DROP_REASON_ICMP_CSUM,
+	/**
+	 * @SKB_DROP_REASON_INVALID_PROTO: the packet doesn't follow RFC 2211,
+	 * such as a broadcasts ICMP_TIMESTAMP
+	 */
+	SKB_DROP_REASON_INVALID_PROTO,
+	/**
+	 * @SKB_DROP_REASON_IP_INADDRERRORS: host unreachable, corresponding to
+	 * IPSTATS_MIB_INADDRERRORS
+	 */
+	SKB_DROP_REASON_IP_INADDRERRORS,
+	/**
+	 * @SKB_DROP_REASON_IP_INNOROUTES: network unreachable, corresponding to
+	 * IPSTATS_MIB_INADDRERRORS
+	 */
+	SKB_DROP_REASON_IP_INNOROUTES,
+	/**
+	 * @SKB_DROP_REASON_PKT_TOO_BIG: packet size is too big (maybe exceed the
+	 * MTU)
+	 */
+	SKB_DROP_REASON_PKT_TOO_BIG,
+	/** @SKB_DROP_REASON_DUP_FRAG: duplicate fragment */
+	SKB_DROP_REASON_DUP_FRAG,
+	/** @SKB_DROP_REASON_FRAG_REASM_TIMEOUT: fragment reassembly timeout */
+	SKB_DROP_REASON_FRAG_REASM_TIMEOUT,
+	/**
+	 * @SKB_DROP_REASON_FRAG_TOO_FAR: ipv4 fragment too far.
+	 * (/proc/sys/net/ipv4/ipfrag_max_dist)
+	 */
+	SKB_DROP_REASON_FRAG_TOO_FAR,
+	/**
+	 * @SKB_DROP_REASON_TCP_MINTTL: ipv4 ttl or ipv6 hoplimit below
+	 * the threshold (IP_MINTTL or IPV6_MINHOPCOUNT).
+	 */
+	SKB_DROP_REASON_TCP_MINTTL,
+	/** @SKB_DROP_REASON_IPV6_BAD_EXTHDR: Bad IPv6 extension header. */
+	SKB_DROP_REASON_IPV6_BAD_EXTHDR,
+	/** @SKB_DROP_REASON_IPV6_NDISC_FRAG: invalid frag (suppress_frag_ndisc). */
+	SKB_DROP_REASON_IPV6_NDISC_FRAG,
+	/** @SKB_DROP_REASON_IPV6_NDISC_HOP_LIMIT: invalid hop limit. */
+	SKB_DROP_REASON_IPV6_NDISC_HOP_LIMIT,
+	/** @SKB_DROP_REASON_IPV6_NDISC_BAD_CODE: invalid NDISC icmp6 code. */
+	SKB_DROP_REASON_IPV6_NDISC_BAD_CODE,
+	/** @SKB_DROP_REASON_IPV6_NDISC_BAD_OPTIONS: invalid NDISC options. */
+	SKB_DROP_REASON_IPV6_NDISC_BAD_OPTIONS,
+	/**
+	 * @SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST: NEIGHBOUR SOLICITATION
+	 * for another host.
+	 */
+	SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST,
+	/**
+	 * @SKB_DROP_REASON_MAX: the maximum of drop reason, which shouldn't be
+	 * used as a real 'reason'
+	 */
+	SKB_DROP_REASON_MAX,
+};
+
+#define SKB_DR_INIT(name, reason)				\
+	enum skb_drop_reason name = SKB_DROP_REASON_##reason
+#define SKB_DR(name)						\
+	SKB_DR_INIT(name, NOT_SPECIFIED)
+#define SKB_DR_SET(name, reason)				\
+	(name = SKB_DROP_REASON_##reason)
+#define SKB_DR_OR(name, reason)					\
+	do {							\
+		if (name == SKB_DROP_REASON_NOT_SPECIFIED ||	\
+		    name == SKB_NOT_DROPPED_YET)		\
+			SKB_DR_SET(name, reason);		\
+	} while (0)
+
+extern const char * const drop_reasons[];
+
+#endif
diff --git a/include/net/dropreason.h b/include/net/dropreason.h
deleted file mode 100644
index c0a3ea806cd5..000000000000
--- a/include/net/dropreason.h
+++ /dev/null
@@ -1,363 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-or-later */
-
-#ifndef _LINUX_DROPREASON_H
-#define _LINUX_DROPREASON_H
-
-#define DEFINE_DROP_REASON(FN, FNe)	\
-	FN(NOT_SPECIFIED)		\
-	FN(NO_SOCKET)			\
-	FN(PKT_TOO_SMALL)		\
-	FN(TCP_CSUM)			\
-	FN(SOCKET_FILTER)		\
-	FN(UDP_CSUM)			\
-	FN(NETFILTER_DROP)		\
-	FN(OTHERHOST)			\
-	FN(IP_CSUM)			\
-	FN(IP_INHDR)			\
-	FN(IP_RPFILTER)			\
-	FN(UNICAST_IN_L2_MULTICAST)	\
-	FN(XFRM_POLICY)			\
-	FN(IP_NOPROTO)			\
-	FN(SOCKET_RCVBUFF)		\
-	FN(PROTO_MEM)			\
-	FN(TCP_MD5NOTFOUND)		\
-	FN(TCP_MD5UNEXPECTED)		\
-	FN(TCP_MD5FAILURE)		\
-	FN(SOCKET_BACKLOG)		\
-	FN(TCP_FLAGS)			\
-	FN(TCP_ZEROWINDOW)		\
-	FN(TCP_OLD_DATA)		\
-	FN(TCP_OVERWINDOW)		\
-	FN(TCP_OFOMERGE)		\
-	FN(TCP_RFC7323_PAWS)		\
-	FN(TCP_INVALID_SEQUENCE)	\
-	FN(TCP_RESET)			\
-	FN(TCP_INVALID_SYN)		\
-	FN(TCP_CLOSE)			\
-	FN(TCP_FASTOPEN)		\
-	FN(TCP_OLD_ACK)			\
-	FN(TCP_TOO_OLD_ACK)		\
-	FN(TCP_ACK_UNSENT_DATA)		\
-	FN(TCP_OFO_QUEUE_PRUNE)		\
-	FN(TCP_OFO_DROP)		\
-	FN(IP_OUTNOROUTES)		\
-	FN(BPF_CGROUP_EGRESS)		\
-	FN(IPV6DISABLED)		\
-	FN(NEIGH_CREATEFAIL)		\
-	FN(NEIGH_FAILED)		\
-	FN(NEIGH_QUEUEFULL)		\
-	FN(NEIGH_DEAD)			\
-	FN(TC_EGRESS)			\
-	FN(QDISC_DROP)			\
-	FN(CPU_BACKLOG)			\
-	FN(XDP)				\
-	FN(TC_INGRESS)			\
-	FN(UNHANDLED_PROTO)		\
-	FN(SKB_CSUM)			\
-	FN(SKB_GSO_SEG)			\
-	FN(SKB_UCOPY_FAULT)		\
-	FN(DEV_HDR)			\
-	FN(DEV_READY)			\
-	FN(FULL_RING)			\
-	FN(NOMEM)			\
-	FN(HDR_TRUNC)			\
-	FN(TAP_FILTER)			\
-	FN(TAP_TXFILTER)		\
-	FN(ICMP_CSUM)			\
-	FN(INVALID_PROTO)		\
-	FN(IP_INADDRERRORS)		\
-	FN(IP_INNOROUTES)		\
-	FN(PKT_TOO_BIG)			\
-	FN(DUP_FRAG)			\
-	FN(FRAG_REASM_TIMEOUT)		\
-	FN(FRAG_TOO_FAR)		\
-	FN(TCP_MINTTL)			\
-	FN(IPV6_BAD_EXTHDR)		\
-	FN(IPV6_NDISC_FRAG)		\
-	FN(IPV6_NDISC_HOP_LIMIT)	\
-	FN(IPV6_NDISC_BAD_CODE)		\
-	FN(IPV6_NDISC_BAD_OPTIONS)	\
-	FN(IPV6_NDISC_NS_OTHERHOST)	\
-	FNe(MAX)
-
-/**
- * enum skb_drop_reason - the reasons of skb drops
- *
- * The reason of skb drop, which is used in kfree_skb_reason().
- */
-enum skb_drop_reason {
-	/**
-	 * @SKB_NOT_DROPPED_YET: skb is not dropped yet (used for no-drop case)
-	 */
-	SKB_NOT_DROPPED_YET = 0,
-	/** @SKB_CONSUMED: packet has been consumed */
-	SKB_CONSUMED,
-	/** @SKB_DROP_REASON_NOT_SPECIFIED: drop reason is not specified */
-	SKB_DROP_REASON_NOT_SPECIFIED,
-	/** @SKB_DROP_REASON_NO_SOCKET: socket not found */
-	SKB_DROP_REASON_NO_SOCKET,
-	/** @SKB_DROP_REASON_PKT_TOO_SMALL: packet size is too small */
-	SKB_DROP_REASON_PKT_TOO_SMALL,
-	/** @SKB_DROP_REASON_TCP_CSUM: TCP checksum error */
-	SKB_DROP_REASON_TCP_CSUM,
-	/** @SKB_DROP_REASON_SOCKET_FILTER: dropped by socket filter */
-	SKB_DROP_REASON_SOCKET_FILTER,
-	/** @SKB_DROP_REASON_UDP_CSUM: UDP checksum error */
-	SKB_DROP_REASON_UDP_CSUM,
-	/** @SKB_DROP_REASON_NETFILTER_DROP: dropped by netfilter */
-	SKB_DROP_REASON_NETFILTER_DROP,
-	/**
-	 * @SKB_DROP_REASON_OTHERHOST: packet don't belong to current host
-	 * (interface is in promisc mode)
-	 */
-	SKB_DROP_REASON_OTHERHOST,
-	/** @SKB_DROP_REASON_IP_CSUM: IP checksum error */
-	SKB_DROP_REASON_IP_CSUM,
-	/**
-	 * @SKB_DROP_REASON_IP_INHDR: there is something wrong with IP header (see
-	 * IPSTATS_MIB_INHDRERRORS)
-	 */
-	SKB_DROP_REASON_IP_INHDR,
-	/**
-	 * @SKB_DROP_REASON_IP_RPFILTER: IP rpfilter validate failed. see the
-	 * document for rp_filter in ip-sysctl.rst for more information
-	 */
-	SKB_DROP_REASON_IP_RPFILTER,
-	/**
-	 * @SKB_DROP_REASON_UNICAST_IN_L2_MULTICAST: destination address of L2 is
-	 * multicast, but L3 is unicast.
-	 */
-	SKB_DROP_REASON_UNICAST_IN_L2_MULTICAST,
-	/** @SKB_DROP_REASON_XFRM_POLICY: xfrm policy check failed */
-	SKB_DROP_REASON_XFRM_POLICY,
-	/** @SKB_DROP_REASON_IP_NOPROTO: no support for IP protocol */
-	SKB_DROP_REASON_IP_NOPROTO,
-	/** @SKB_DROP_REASON_SOCKET_RCVBUFF: socket receive buff is full */
-	SKB_DROP_REASON_SOCKET_RCVBUFF,
-	/**
-	 * @SKB_DROP_REASON_PROTO_MEM: proto memory limition, such as udp packet
-	 * drop out of udp_memory_allocated.
-	 */
-	SKB_DROP_REASON_PROTO_MEM,
-	/**
-	 * @SKB_DROP_REASON_TCP_MD5NOTFOUND: no MD5 hash and one expected,
-	 * corresponding to LINUX_MIB_TCPMD5NOTFOUND
-	 */
-	SKB_DROP_REASON_TCP_MD5NOTFOUND,
-	/**
-	 * @SKB_DROP_REASON_TCP_MD5UNEXPECTED: MD5 hash and we're not expecting
-	 * one, corresponding to LINUX_MIB_TCPMD5UNEXPECTED
-	 */
-	SKB_DROP_REASON_TCP_MD5UNEXPECTED,
-	/**
-	 * @SKB_DROP_REASON_TCP_MD5FAILURE: MD5 hash and its wrong, corresponding
-	 * to LINUX_MIB_TCPMD5FAILURE
-	 */
-	SKB_DROP_REASON_TCP_MD5FAILURE,
-	/**
-	 * @SKB_DROP_REASON_SOCKET_BACKLOG: failed to add skb to socket backlog (
-	 * see LINUX_MIB_TCPBACKLOGDROP)
-	 */
-	SKB_DROP_REASON_SOCKET_BACKLOG,
-	/** @SKB_DROP_REASON_TCP_FLAGS: TCP flags invalid */
-	SKB_DROP_REASON_TCP_FLAGS,
-	/**
-	 * @SKB_DROP_REASON_TCP_ZEROWINDOW: TCP receive window size is zero,
-	 * see LINUX_MIB_TCPZEROWINDOWDROP
-	 */
-	SKB_DROP_REASON_TCP_ZEROWINDOW,
-	/**
-	 * @SKB_DROP_REASON_TCP_OLD_DATA: the TCP data reveived is already
-	 * received before (spurious retrans may happened), see
-	 * LINUX_MIB_DELAYEDACKLOST
-	 */
-	SKB_DROP_REASON_TCP_OLD_DATA,
-	/**
-	 * @SKB_DROP_REASON_TCP_OVERWINDOW: the TCP data is out of window,
-	 * the seq of the first byte exceed the right edges of receive
-	 * window
-	 */
-	SKB_DROP_REASON_TCP_OVERWINDOW,
-	/**
-	 * @SKB_DROP_REASON_TCP_OFOMERGE: the data of skb is already in the ofo
-	 * queue, corresponding to LINUX_MIB_TCPOFOMERGE
-	 */
-	SKB_DROP_REASON_TCP_OFOMERGE,
-	/**
-	 * @SKB_DROP_REASON_TCP_RFC7323_PAWS: PAWS check, corresponding to
-	 * LINUX_MIB_PAWSESTABREJECTED
-	 */
-	SKB_DROP_REASON_TCP_RFC7323_PAWS,
-	/** @SKB_DROP_REASON_TCP_INVALID_SEQUENCE: Not acceptable SEQ field */
-	SKB_DROP_REASON_TCP_INVALID_SEQUENCE,
-	/** @SKB_DROP_REASON_TCP_RESET: Invalid RST packet */
-	SKB_DROP_REASON_TCP_RESET,
-	/**
-	 * @SKB_DROP_REASON_TCP_INVALID_SYN: Incoming packet has unexpected
-	 * SYN flag
-	 */
-	SKB_DROP_REASON_TCP_INVALID_SYN,
-	/** @SKB_DROP_REASON_TCP_CLOSE: TCP socket in CLOSE state */
-	SKB_DROP_REASON_TCP_CLOSE,
-	/** @SKB_DROP_REASON_TCP_FASTOPEN: dropped by FASTOPEN request socket */
-	SKB_DROP_REASON_TCP_FASTOPEN,
-	/** @SKB_DROP_REASON_TCP_OLD_ACK: TCP ACK is old, but in window */
-	SKB_DROP_REASON_TCP_OLD_ACK,
-	/** @SKB_DROP_REASON_TCP_TOO_OLD_ACK: TCP ACK is too old */
-	SKB_DROP_REASON_TCP_TOO_OLD_ACK,
-	/**
-	 * @SKB_DROP_REASON_TCP_ACK_UNSENT_DATA: TCP ACK for data we haven't
-	 * sent yet
-	 */
-	SKB_DROP_REASON_TCP_ACK_UNSENT_DATA,
-	/** @SKB_DROP_REASON_TCP_OFO_QUEUE_PRUNE: pruned from TCP OFO queue */
-	SKB_DROP_REASON_TCP_OFO_QUEUE_PRUNE,
-	/** @SKB_DROP_REASON_TCP_OFO_DROP: data already in receive queue */
-	SKB_DROP_REASON_TCP_OFO_DROP,
-	/** @SKB_DROP_REASON_IP_OUTNOROUTES: route lookup failed */
-	SKB_DROP_REASON_IP_OUTNOROUTES,
-	/**
-	 * @SKB_DROP_REASON_BPF_CGROUP_EGRESS: dropped by BPF_PROG_TYPE_CGROUP_SKB
-	 * eBPF program
-	 */
-	SKB_DROP_REASON_BPF_CGROUP_EGRESS,
-	/** @SKB_DROP_REASON_IPV6DISABLED: IPv6 is disabled on the device */
-	SKB_DROP_REASON_IPV6DISABLED,
-	/** @SKB_DROP_REASON_NEIGH_CREATEFAIL: failed to create neigh entry */
-	SKB_DROP_REASON_NEIGH_CREATEFAIL,
-	/** @SKB_DROP_REASON_NEIGH_FAILED: neigh entry in failed state */
-	SKB_DROP_REASON_NEIGH_FAILED,
-	/** @SKB_DROP_REASON_NEIGH_QUEUEFULL: arp_queue for neigh entry is full */
-	SKB_DROP_REASON_NEIGH_QUEUEFULL,
-	/** @SKB_DROP_REASON_NEIGH_DEAD: neigh entry is dead */
-	SKB_DROP_REASON_NEIGH_DEAD,
-	/** @SKB_DROP_REASON_TC_EGRESS: dropped in TC egress HOOK */
-	SKB_DROP_REASON_TC_EGRESS,
-	/**
-	 * @SKB_DROP_REASON_QDISC_DROP: dropped by qdisc when packet outputting (
-	 * failed to enqueue to current qdisc)
-	 */
-	SKB_DROP_REASON_QDISC_DROP,
-	/**
-	 * @SKB_DROP_REASON_CPU_BACKLOG: failed to enqueue the skb to the per CPU
-	 * backlog queue. This can be caused by backlog queue full (see
-	 * netdev_max_backlog in net.rst) or RPS flow limit
-	 */
-	SKB_DROP_REASON_CPU_BACKLOG,
-	/** @SKB_DROP_REASON_XDP: dropped by XDP in input path */
-	SKB_DROP_REASON_XDP,
-	/** @SKB_DROP_REASON_TC_INGRESS: dropped in TC ingress HOOK */
-	SKB_DROP_REASON_TC_INGRESS,
-	/** @SKB_DROP_REASON_UNHANDLED_PROTO: protocol not implemented or not supported */
-	SKB_DROP_REASON_UNHANDLED_PROTO,
-	/** @SKB_DROP_REASON_SKB_CSUM: sk_buff checksum computation error */
-	SKB_DROP_REASON_SKB_CSUM,
-	/** @SKB_DROP_REASON_SKB_GSO_SEG: gso segmentation error */
-	SKB_DROP_REASON_SKB_GSO_SEG,
-	/**
-	 * @SKB_DROP_REASON_SKB_UCOPY_FAULT: failed to copy data from user space,
-	 * e.g., via zerocopy_sg_from_iter() or skb_orphan_frags_rx()
-	 */
-	SKB_DROP_REASON_SKB_UCOPY_FAULT,
-	/** @SKB_DROP_REASON_DEV_HDR: device driver specific header/metadata is invalid */
-	SKB_DROP_REASON_DEV_HDR,
-	/**
-	 * @SKB_DROP_REASON_DEV_READY: the device is not ready to xmit/recv due to
-	 * any of its data structure that is not up/ready/initialized,
-	 * e.g., the IFF_UP is not set, or driver specific tun->tfiles[txq]
-	 * is not initialized
-	 */
-	SKB_DROP_REASON_DEV_READY,
-	/** @SKB_DROP_REASON_FULL_RING: ring buffer is full */
-	SKB_DROP_REASON_FULL_RING,
-	/** @SKB_DROP_REASON_NOMEM: error due to OOM */
-	SKB_DROP_REASON_NOMEM,
-	/**
-	 * @SKB_DROP_REASON_HDR_TRUNC: failed to trunc/extract the header from
-	 * networking data, e.g., failed to pull the protocol header from
-	 * frags via pskb_may_pull()
-	 */
-	SKB_DROP_REASON_HDR_TRUNC,
-	/**
-	 * @SKB_DROP_REASON_TAP_FILTER: dropped by (ebpf) filter directly attached
-	 * to tun/tap, e.g., via TUNSETFILTEREBPF
-	 */
-	SKB_DROP_REASON_TAP_FILTER,
-	/**
-	 * @SKB_DROP_REASON_TAP_TXFILTER: dropped by tx filter implemented at
-	 * tun/tap, e.g., check_filter()
-	 */
-	SKB_DROP_REASON_TAP_TXFILTER,
-	/** @SKB_DROP_REASON_ICMP_CSUM: ICMP checksum error */
-	SKB_DROP_REASON_ICMP_CSUM,
-	/**
-	 * @SKB_DROP_REASON_INVALID_PROTO: the packet doesn't follow RFC 2211,
-	 * such as a broadcasts ICMP_TIMESTAMP
-	 */
-	SKB_DROP_REASON_INVALID_PROTO,
-	/**
-	 * @SKB_DROP_REASON_IP_INADDRERRORS: host unreachable, corresponding to
-	 * IPSTATS_MIB_INADDRERRORS
-	 */
-	SKB_DROP_REASON_IP_INADDRERRORS,
-	/**
-	 * @SKB_DROP_REASON_IP_INNOROUTES: network unreachable, corresponding to
-	 * IPSTATS_MIB_INADDRERRORS
-	 */
-	SKB_DROP_REASON_IP_INNOROUTES,
-	/**
-	 * @SKB_DROP_REASON_PKT_TOO_BIG: packet size is too big (maybe exceed the
-	 * MTU)
-	 */
-	SKB_DROP_REASON_PKT_TOO_BIG,
-	/** @SKB_DROP_REASON_DUP_FRAG: duplicate fragment */
-	SKB_DROP_REASON_DUP_FRAG,
-	/** @SKB_DROP_REASON_FRAG_REASM_TIMEOUT: fragment reassembly timeout */
-	SKB_DROP_REASON_FRAG_REASM_TIMEOUT,
-	/**
-	 * @SKB_DROP_REASON_FRAG_TOO_FAR: ipv4 fragment too far.
-	 * (/proc/sys/net/ipv4/ipfrag_max_dist)
-	 */
-	SKB_DROP_REASON_FRAG_TOO_FAR,
-	/**
-	 * @SKB_DROP_REASON_TCP_MINTTL: ipv4 ttl or ipv6 hoplimit below
-	 * the threshold (IP_MINTTL or IPV6_MINHOPCOUNT).
-	 */
-	SKB_DROP_REASON_TCP_MINTTL,
-	/** @SKB_DROP_REASON_IPV6_BAD_EXTHDR: Bad IPv6 extension header. */
-	SKB_DROP_REASON_IPV6_BAD_EXTHDR,
-	/** @SKB_DROP_REASON_IPV6_NDISC_FRAG: invalid frag (suppress_frag_ndisc). */
-	SKB_DROP_REASON_IPV6_NDISC_FRAG,
-	/** @SKB_DROP_REASON_IPV6_NDISC_HOP_LIMIT: invalid hop limit. */
-	SKB_DROP_REASON_IPV6_NDISC_HOP_LIMIT,
-	/** @SKB_DROP_REASON_IPV6_NDISC_BAD_CODE: invalid NDISC icmp6 code. */
-	SKB_DROP_REASON_IPV6_NDISC_BAD_CODE,
-	/** @SKB_DROP_REASON_IPV6_NDISC_BAD_OPTIONS: invalid NDISC options. */
-	SKB_DROP_REASON_IPV6_NDISC_BAD_OPTIONS,
-	/** @SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST: NEIGHBOUR SOLICITATION
-	 * for another host.
-	 */
-	SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST,
-	/**
-	 * @SKB_DROP_REASON_MAX: the maximum of drop reason, which shouldn't be
-	 * used as a real 'reason'
-	 */
-	SKB_DROP_REASON_MAX,
-};
-
-#define SKB_DR_INIT(name, reason)				\
-	enum skb_drop_reason name = SKB_DROP_REASON_##reason
-#define SKB_DR(name)						\
-	SKB_DR_INIT(name, NOT_SPECIFIED)
-#define SKB_DR_SET(name, reason)				\
-	(name = SKB_DROP_REASON_##reason)
-#define SKB_DR_OR(name, reason)					\
-	do {							\
-		if (name == SKB_DROP_REASON_NOT_SPECIFIED ||	\
-		    name == SKB_NOT_DROPPED_YET)		\
-			SKB_DR_SET(name, reason);		\
-	} while (0)
-
-extern const char * const drop_reasons[];
-
-#endif
diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index b23ddec3cd5c..325ad893f624 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -7,7 +7,7 @@
 #include <linux/in6.h>
 #include <linux/rbtree_types.h>
 #include <linux/refcount.h>
-#include <net/dropreason.h>
+#include <net/dropreason-core.h>
 
 /* Per netns frag queues directory */
 struct fqdir {
-- 
cgit v1.2.3-70-g09d2


From 071c0fc6fb919dcf29c676a842dda08a674877d7 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 19 Apr 2023 14:52:53 +0200
Subject: net: extend drop reasons for multiple subsystems

Extend drop reasons to make them usable by subsystems
other than core by reserving the high 16 bits for a
new subsystem ID, of which 0 of course is used for the
existing reasons immediately.

To still be able to have string reasons, restructure
that code a bit to make the loopup under RCU, the only
user of this (right now) is drop_monitor.

Link: https://lore.kernel.org/netdev/00659771ed54353f92027702c5bbb84702da62ce.camel@sipsolutions.net
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/dropreason-core.h | 14 +++++++---
 include/net/dropreason.h      | 31 +++++++++++++++++++++++
 net/core/drop_monitor.c       | 33 +++++++++++++++++-------
 net/core/skbuff.c             | 59 ++++++++++++++++++++++++++++++++++++++++---
 4 files changed, 121 insertions(+), 16 deletions(-)
 create mode 100644 include/net/dropreason.h

diff --git a/include/net/dropreason-core.h b/include/net/dropreason-core.h
index ade6d5b9186c..a2b953b57689 100644
--- a/include/net/dropreason-core.h
+++ b/include/net/dropreason-core.h
@@ -340,12 +340,20 @@ enum skb_drop_reason {
 	 */
 	SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST,
 	/**
-	 * @SKB_DROP_REASON_MAX: the maximum of drop reason, which shouldn't be
-	 * used as a real 'reason'
+	 * @SKB_DROP_REASON_MAX: the maximum of core drop reasons, which
+	 * shouldn't be used as a real 'reason' - only for tracing code gen
 	 */
 	SKB_DROP_REASON_MAX,
+
+	/**
+	 * @SKB_DROP_REASON_SUBSYS_MASK: subsystem mask in drop reasons,
+	 * see &enum skb_drop_reason_subsys
+	 */
+	SKB_DROP_REASON_SUBSYS_MASK = 0xffff0000,
 };
 
+#define SKB_DROP_REASON_SUBSYS_SHIFT	16
+
 #define SKB_DR_INIT(name, reason)				\
 	enum skb_drop_reason name = SKB_DROP_REASON_##reason
 #define SKB_DR(name)						\
@@ -359,6 +367,4 @@ enum skb_drop_reason {
 			SKB_DR_SET(name, reason);		\
 	} while (0)
 
-extern const char * const drop_reasons[];
-
 #endif
diff --git a/include/net/dropreason.h b/include/net/dropreason.h
new file mode 100644
index 000000000000..f0f2378dbed0
--- /dev/null
+++ b/include/net/dropreason.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+
+#ifndef _LINUX_DROPREASON_H
+#define _LINUX_DROPREASON_H
+#include <net/dropreason-core.h>
+
+/**
+ * enum skb_drop_reason_subsys - subsystem tag for (extended) drop reasons
+ */
+enum skb_drop_reason_subsys {
+	/** @SKB_DROP_REASON_SUBSYS_CORE: core drop reasons defined above */
+	SKB_DROP_REASON_SUBSYS_CORE,
+
+	/** @SKB_DROP_REASON_SUBSYS_NUM: number of subsystems defined */
+	SKB_DROP_REASON_SUBSYS_NUM
+};
+
+struct drop_reason_list {
+	const char * const *reasons;
+	size_t n_reasons;
+};
+
+/* Note: due to dynamic registrations, access must be under RCU */
+extern const struct drop_reason_list __rcu *
+drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_NUM];
+
+void drop_reasons_register_subsys(enum skb_drop_reason_subsys subsys,
+				  const struct drop_reason_list *list);
+void drop_reasons_unregister_subsys(enum skb_drop_reason_subsys subsys);
+
+#endif
diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c
index 5a782d1d8fd3..aff31cd944c2 100644
--- a/net/core/drop_monitor.c
+++ b/net/core/drop_monitor.c
@@ -21,6 +21,7 @@
 #include <linux/workqueue.h>
 #include <linux/netlink.h>
 #include <linux/net_dropmon.h>
+#include <linux/bitfield.h>
 #include <linux/percpu.h>
 #include <linux/timer.h>
 #include <linux/bitops.h>
@@ -29,6 +30,7 @@
 #include <net/genetlink.h>
 #include <net/netevent.h>
 #include <net/flow_offload.h>
+#include <net/dropreason.h>
 #include <net/devlink.h>
 
 #include <trace/events/skb.h>
@@ -504,8 +506,6 @@ static void net_dm_packet_trace_kfree_skb_hit(void *ignore,
 	if (!nskb)
 		return;
 
-	if (unlikely(reason >= SKB_DROP_REASON_MAX || reason <= 0))
-		reason = SKB_DROP_REASON_NOT_SPECIFIED;
 	cb = NET_DM_SKB_CB(nskb);
 	cb->reason = reason;
 	cb->pc = location;
@@ -552,9 +552,9 @@ static size_t net_dm_in_port_size(void)
 }
 
 #define NET_DM_MAX_SYMBOL_LEN 40
+#define NET_DM_MAX_REASON_LEN 50
 
-static size_t net_dm_packet_report_size(size_t payload_len,
-					enum skb_drop_reason reason)
+static size_t net_dm_packet_report_size(size_t payload_len)
 {
 	size_t size;
 
@@ -576,7 +576,7 @@ static size_t net_dm_packet_report_size(size_t payload_len,
 	       /* NET_DM_ATTR_PROTO */
 	       nla_total_size(sizeof(u16)) +
 	       /* NET_DM_ATTR_REASON */
-	       nla_total_size(strlen(drop_reasons[reason]) + 1) +
+	       nla_total_size(NET_DM_MAX_REASON_LEN + 1) +
 	       /* NET_DM_ATTR_PAYLOAD */
 	       nla_total_size(payload_len);
 }
@@ -610,6 +610,8 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb,
 				     size_t payload_len)
 {
 	struct net_dm_skb_cb *cb = NET_DM_SKB_CB(skb);
+	const struct drop_reason_list *list = NULL;
+	unsigned int subsys, subsys_reason;
 	char buf[NET_DM_MAX_SYMBOL_LEN];
 	struct nlattr *attr;
 	void *hdr;
@@ -627,9 +629,24 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb,
 			      NET_DM_ATTR_PAD))
 		goto nla_put_failure;
 
+	rcu_read_lock();
+	subsys = u32_get_bits(cb->reason, SKB_DROP_REASON_SUBSYS_MASK);
+	if (subsys < SKB_DROP_REASON_SUBSYS_NUM)
+		list = rcu_dereference(drop_reasons_by_subsys[subsys]);
+	subsys_reason = cb->reason & ~SKB_DROP_REASON_SUBSYS_MASK;
+	if (!list ||
+	    subsys_reason >= list->n_reasons ||
+	    !list->reasons[subsys_reason] ||
+	    strlen(list->reasons[subsys_reason]) > NET_DM_MAX_REASON_LEN) {
+		list = rcu_dereference(drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_CORE]);
+		subsys_reason = SKB_DROP_REASON_NOT_SPECIFIED;
+	}
 	if (nla_put_string(msg, NET_DM_ATTR_REASON,
-			   drop_reasons[cb->reason]))
+			   list->reasons[subsys_reason])) {
+		rcu_read_unlock();
 		goto nla_put_failure;
+	}
+	rcu_read_unlock();
 
 	snprintf(buf, sizeof(buf), "%pS", cb->pc);
 	if (nla_put_string(msg, NET_DM_ATTR_SYMBOL, buf))
@@ -687,9 +704,7 @@ static void net_dm_packet_report(struct sk_buff *skb)
 	if (net_dm_trunc_len)
 		payload_len = min_t(size_t, net_dm_trunc_len, payload_len);
 
-	msg = nlmsg_new(net_dm_packet_report_size(payload_len,
-						  NET_DM_SKB_CB(skb)->reason),
-			GFP_KERNEL);
+	msg = nlmsg_new(net_dm_packet_report_size(payload_len), GFP_KERNEL);
 	if (!msg)
 		goto out;
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 8764653bede7..0d998806b377 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -58,6 +58,7 @@
 #include <linux/scatterlist.h>
 #include <linux/errqueue.h>
 #include <linux/prefetch.h>
+#include <linux/bitfield.h>
 #include <linux/if_vlan.h>
 #include <linux/mpls.h>
 #include <linux/kcov.h>
@@ -72,6 +73,7 @@
 #include <net/mptcp.h>
 #include <net/mctp.h>
 #include <net/page_pool.h>
+#include <net/dropreason.h>
 
 #include <linux/uaccess.h>
 #include <trace/events/skb.h>
@@ -122,11 +124,59 @@ EXPORT_SYMBOL(sysctl_max_skb_frags);
 
 #undef FN
 #define FN(reason) [SKB_DROP_REASON_##reason] = #reason,
-const char * const drop_reasons[] = {
+static const char * const drop_reasons[] = {
 	[SKB_CONSUMED] = "CONSUMED",
 	DEFINE_DROP_REASON(FN, FN)
 };
-EXPORT_SYMBOL(drop_reasons);
+
+static const struct drop_reason_list drop_reasons_core = {
+	.reasons = drop_reasons,
+	.n_reasons = ARRAY_SIZE(drop_reasons),
+};
+
+const struct drop_reason_list __rcu *
+drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_NUM] = {
+	[SKB_DROP_REASON_SUBSYS_CORE] = RCU_INITIALIZER(&drop_reasons_core),
+};
+EXPORT_SYMBOL(drop_reasons_by_subsys);
+
+/**
+ * drop_reasons_register_subsys - register another drop reason subsystem
+ * @subsys: the subsystem to register, must not be the core
+ * @list: the list of drop reasons within the subsystem, must point to
+ *	a statically initialized list
+ */
+void drop_reasons_register_subsys(enum skb_drop_reason_subsys subsys,
+				  const struct drop_reason_list *list)
+{
+	if (WARN(subsys <= SKB_DROP_REASON_SUBSYS_CORE ||
+		 subsys >= ARRAY_SIZE(drop_reasons_by_subsys),
+		 "invalid subsystem %d\n", subsys))
+		return;
+
+	/* must point to statically allocated memory, so INIT is OK */
+	RCU_INIT_POINTER(drop_reasons_by_subsys[subsys], list);
+}
+EXPORT_SYMBOL_GPL(drop_reasons_register_subsys);
+
+/**
+ * drop_reasons_unregister_subsys - unregister a drop reason subsystem
+ * @subsys: the subsystem to remove, must not be the core
+ *
+ * Note: This will synchronize_rcu() to ensure no users when it returns.
+ */
+void drop_reasons_unregister_subsys(enum skb_drop_reason_subsys subsys)
+{
+	if (WARN(subsys <= SKB_DROP_REASON_SUBSYS_CORE ||
+		 subsys >= ARRAY_SIZE(drop_reasons_by_subsys),
+		 "invalid subsystem %d\n", subsys))
+		return;
+
+	RCU_INIT_POINTER(drop_reasons_by_subsys[subsys], NULL);
+
+	synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(drop_reasons_unregister_subsys);
 
 /**
  *	skb_panic - private function for out-of-line support
@@ -986,7 +1036,10 @@ bool __kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
 	if (unlikely(!skb_unref(skb)))
 		return false;
 
-	DEBUG_NET_WARN_ON_ONCE(reason <= 0 || reason >= SKB_DROP_REASON_MAX);
+	DEBUG_NET_WARN_ON_ONCE(reason == SKB_NOT_DROPPED_YET ||
+			       u32_get_bits(reason,
+					    SKB_DROP_REASON_SUBSYS_MASK) >=
+				SKB_DROP_REASON_SUBSYS_NUM);
 
 	if (reason == SKB_CONSUMED)
 		trace_consume_skb(skb, __builtin_return_address(0));
-- 
cgit v1.2.3-70-g09d2


From baa951a1c1771810f3a378a95fc93e81953027d5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 19 Apr 2023 14:52:54 +0200
Subject: mac80211: use the new drop reasons infrastructure

It can be really hard to analyse or debug why packets are
going missing in mac80211, so add the needed infrastructure
to use use the new per-subsystem drop reasons.

We actually use two drop reason subsystems here because of
the different handling of frames that are dropped but still
go to monitor for old versions of hostapd, and those that
are just completely unusable (e.g. crypto failed.)

Annotate a few reasons here just to illustrate this, we'll
need to go through and annotate more of them later.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/dropreason.h   | 12 ++++++++++
 net/mac80211/drop.h        | 56 ++++++++++++++++++++++++++++++++++++++++++++++
 net/mac80211/ieee80211_i.h |  8 +------
 net/mac80211/main.c        | 31 +++++++++++++++++++++++++
 net/mac80211/rx.c          | 55 +++++++++++++++++++++------------------------
 net/mac80211/wpa.c         | 24 ++++++++++----------
 6 files changed, 138 insertions(+), 48 deletions(-)
 create mode 100644 net/mac80211/drop.h

diff --git a/include/net/dropreason.h b/include/net/dropreason.h
index f0f2378dbed0..685fb37df8e8 100644
--- a/include/net/dropreason.h
+++ b/include/net/dropreason.h
@@ -11,6 +11,18 @@ enum skb_drop_reason_subsys {
 	/** @SKB_DROP_REASON_SUBSYS_CORE: core drop reasons defined above */
 	SKB_DROP_REASON_SUBSYS_CORE,
 
+	/**
+	 * @SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE: mac80211 drop reasons
+	 * for unusable frames, see net/mac80211/drop.h
+	 */
+	SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE,
+
+	/**
+	 * @SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR: mac80211 drop reasons
+	 * for frames still going to monitor, see net/mac80211/drop.h
+	 */
+	SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR,
+
 	/** @SKB_DROP_REASON_SUBSYS_NUM: number of subsystems defined */
 	SKB_DROP_REASON_SUBSYS_NUM
 };
diff --git a/net/mac80211/drop.h b/net/mac80211/drop.h
new file mode 100644
index 000000000000..49dc809cab29
--- /dev/null
+++ b/net/mac80211/drop.h
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * mac80211 drop reason list
+ *
+ * Copyright (C) 2023 Intel Corporation
+ */
+
+#ifndef MAC80211_DROP_H
+#define MAC80211_DROP_H
+#include <net/dropreason.h>
+
+typedef unsigned int __bitwise ieee80211_rx_result;
+
+#define MAC80211_DROP_REASONS_MONITOR(R)	\
+	R(RX_DROP_M_UNEXPECTED_4ADDR_FRAME)	\
+	R(RX_DROP_M_BAD_BCN_KEYIDX)		\
+	R(RX_DROP_M_BAD_MGMT_KEYIDX)		\
+/* this line for the trailing \ - add before this */
+
+#define MAC80211_DROP_REASONS_UNUSABLE(R)	\
+	R(RX_DROP_U_MIC_FAIL)			\
+	R(RX_DROP_U_REPLAY)			\
+	R(RX_DROP_U_BAD_MMIE)			\
+/* this line for the trailing \ - add before this */
+
+/* having two enums allows for checking ieee80211_rx_result use with sparse */
+enum ___mac80211_drop_reason {
+/* if we get to the end of handlers with RX_CONTINUE this will be the reason */
+	___RX_CONTINUE	= SKB_CONSUMED,
+
+/* this never gets used as an argument to kfree_skb_reason() */
+	___RX_QUEUED	= SKB_NOT_DROPPED_YET,
+
+#define ENUM(x) ___ ## x,
+	___RX_DROP_MONITOR = SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR <<
+		SKB_DROP_REASON_SUBSYS_SHIFT,
+	MAC80211_DROP_REASONS_MONITOR(ENUM)
+
+	___RX_DROP_UNUSABLE = SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE <<
+		SKB_DROP_REASON_SUBSYS_SHIFT,
+	MAC80211_DROP_REASONS_UNUSABLE(ENUM)
+#undef ENUM
+};
+
+enum mac80211_drop_reason {
+	RX_CONTINUE	 = (__force ieee80211_rx_result)___RX_CONTINUE,
+	RX_QUEUED	 = (__force ieee80211_rx_result)___RX_QUEUED,
+	RX_DROP_MONITOR	 = (__force ieee80211_rx_result)___RX_DROP_MONITOR,
+	RX_DROP_UNUSABLE = (__force ieee80211_rx_result)___RX_DROP_UNUSABLE,
+#define DEF(x) x = (__force ieee80211_rx_result)___ ## x,
+	MAC80211_DROP_REASONS_MONITOR(DEF)
+	MAC80211_DROP_REASONS_UNUSABLE(DEF)
+#undef DEF
+};
+
+#endif /* MAC80211_DROP_H */
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 9b7e184430b8..a0a7839cb961 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -33,6 +33,7 @@
 #include "key.h"
 #include "sta_info.h"
 #include "debug.h"
+#include "drop.h"
 
 extern const struct cfg80211_ops mac80211_config_ops;
 
@@ -170,13 +171,6 @@ struct ieee80211_tx_data {
 	unsigned int flags;
 };
 
-
-typedef unsigned __bitwise ieee80211_rx_result;
-#define RX_CONTINUE		((__force ieee80211_rx_result) 0u)
-#define RX_DROP_UNUSABLE	((__force ieee80211_rx_result) 1u)
-#define RX_DROP_MONITOR		((__force ieee80211_rx_result) 2u)
-#define RX_QUEUED		((__force ieee80211_rx_result) 3u)
-
 /**
  * enum ieee80211_packet_rx_flags - packet RX flags
  * @IEEE80211_RX_AMSDU: a-MSDU packet
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index ddf2b7811c55..55cdfaef0f5d 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -22,6 +22,7 @@
 #include <linux/bitmap.h>
 #include <linux/inetdevice.h>
 #include <net/net_namespace.h>
+#include <net/dropreason.h>
 #include <net/cfg80211.h>
 #include <net/addrconf.h>
 
@@ -1542,6 +1543,28 @@ void ieee80211_free_hw(struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL(ieee80211_free_hw);
 
+static const char * const drop_reasons_monitor[] = {
+#define V(x)	#x,
+	[0] = "RX_DROP_MONITOR",
+	MAC80211_DROP_REASONS_MONITOR(V)
+};
+
+static struct drop_reason_list drop_reason_list_monitor = {
+	.reasons = drop_reasons_monitor,
+	.n_reasons = ARRAY_SIZE(drop_reasons_monitor),
+};
+
+static const char * const drop_reasons_unusable[] = {
+	[0] = "RX_DROP_UNUSABLE",
+	MAC80211_DROP_REASONS_UNUSABLE(V)
+#undef V
+};
+
+static struct drop_reason_list drop_reason_list_unusable = {
+	.reasons = drop_reasons_unusable,
+	.n_reasons = ARRAY_SIZE(drop_reasons_unusable),
+};
+
 static int __init ieee80211_init(void)
 {
 	struct sk_buff *skb;
@@ -1559,6 +1582,11 @@ static int __init ieee80211_init(void)
 	if (ret)
 		goto err_netdev;
 
+	drop_reasons_register_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR,
+				     &drop_reason_list_monitor);
+	drop_reasons_register_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE,
+				     &drop_reason_list_unusable);
+
 	return 0;
  err_netdev:
 	rc80211_minstrel_exit();
@@ -1574,6 +1602,9 @@ static void __exit ieee80211_exit(void)
 
 	ieee80211_iface_exit();
 
+	drop_reasons_unregister_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR);
+	drop_reasons_unregister_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE);
+
 	rcu_barrier();
 }
 
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index db3451f5f2fb..58222c077898 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1826,7 +1826,7 @@ ieee80211_rx_h_sta_process(struct ieee80211_rx_data *rx)
 				cfg80211_rx_unexpected_4addr_frame(
 					rx->sdata->dev, sta->sta.addr,
 					GFP_ATOMIC);
-			return RX_DROP_MONITOR;
+			return RX_DROP_M_UNEXPECTED_4ADDR_FRAME;
 		}
 		/*
 		 * Update counter and free packet here to avoid
@@ -1961,7 +1961,7 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 				cfg80211_rx_unprot_mlme_mgmt(rx->sdata->dev,
 							     skb->data,
 							     skb->len);
-			return RX_DROP_MONITOR; /* unexpected BIP keyidx */
+			return RX_DROP_M_BAD_BCN_KEYIDX;
 		}
 
 		rx->key = ieee80211_rx_get_bigtk(rx, mmie_keyidx);
@@ -1975,7 +1975,7 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 
 		if (mmie_keyidx < NUM_DEFAULT_KEYS ||
 		    mmie_keyidx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS)
-			return RX_DROP_MONITOR; /* unexpected BIP keyidx */
+			return RX_DROP_M_BAD_MGMT_KEYIDX; /* unexpected BIP keyidx */
 		if (rx->link_sta) {
 			if (ieee80211_is_group_privacy_action(skb) &&
 			    test_sta_flag(rx->sta, WLAN_STA_MFP))
@@ -3960,7 +3960,8 @@ ieee80211_rx_h_mgmt(struct ieee80211_rx_data *rx)
 }
 
 static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx,
-					struct ieee80211_rate *rate)
+					struct ieee80211_rate *rate,
+					ieee80211_rx_result reason)
 {
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_local *local = rx->local;
@@ -4024,42 +4025,38 @@ static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx,
 	}
 
  out_free_skb:
-	dev_kfree_skb(skb);
+	kfree_skb_reason(skb, (__force u32)reason);
 }
 
 static void ieee80211_rx_handlers_result(struct ieee80211_rx_data *rx,
 					 ieee80211_rx_result res)
 {
-	switch (res) {
-	case RX_DROP_MONITOR:
-		I802_DEBUG_INC(rx->sdata->local->rx_handlers_drop);
-		if (rx->sta)
-			rx->link_sta->rx_stats.dropped++;
-		fallthrough;
-	case RX_CONTINUE: {
-		struct ieee80211_rate *rate = NULL;
-		struct ieee80211_supported_band *sband;
-		struct ieee80211_rx_status *status;
-
-		status = IEEE80211_SKB_RXCB((rx->skb));
+	struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb);
+	struct ieee80211_supported_band *sband;
+	struct ieee80211_rate *rate = NULL;
 
-		sband = rx->local->hw.wiphy->bands[status->band];
-		if (status->encoding == RX_ENC_LEGACY)
-			rate = &sband->bitrates[status->rate_idx];
+	if (res == RX_QUEUED) {
+		I802_DEBUG_INC(rx->sdata->local->rx_handlers_queued);
+		return;
+	}
 
-		ieee80211_rx_cooked_monitor(rx, rate);
-		break;
-		}
-	case RX_DROP_UNUSABLE:
+	if (res != RX_CONTINUE) {
 		I802_DEBUG_INC(rx->sdata->local->rx_handlers_drop);
 		if (rx->sta)
 			rx->link_sta->rx_stats.dropped++;
-		dev_kfree_skb(rx->skb);
-		break;
-	case RX_QUEUED:
-		I802_DEBUG_INC(rx->sdata->local->rx_handlers_queued);
-		break;
 	}
+
+	if (u32_get_bits((__force u32)res, SKB_DROP_REASON_SUBSYS_MASK) ==
+			SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE) {
+		kfree_skb_reason(rx->skb, (__force u32)res);
+		return;
+	}
+
+	sband = rx->local->hw.wiphy->bands[status->band];
+	if (status->encoding == RX_ENC_LEGACY)
+		rate = &sband->bitrates[status->rate_idx];
+
+	ieee80211_rx_cooked_monitor(rx, rate, res);
 }
 
 static void ieee80211_rx_handlers(struct ieee80211_rx_data *rx,
diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c
index 20f742b5503b..4133496da378 100644
--- a/net/mac80211/wpa.c
+++ b/net/mac80211/wpa.c
@@ -550,7 +550,7 @@ ieee80211_crypto_ccmp_decrypt(struct ieee80211_rx_data *rx,
 		if (res < 0 ||
 		    (!res && !(status->flag & RX_FLAG_ALLOW_SAME_PN))) {
 			key->u.ccmp.replays++;
-			return RX_DROP_UNUSABLE;
+			return RX_DROP_U_REPLAY;
 		}
 
 		if (!(status->flag & RX_FLAG_DECRYPTED)) {
@@ -564,7 +564,7 @@ ieee80211_crypto_ccmp_decrypt(struct ieee80211_rx_data *rx,
 				    skb->data + hdrlen + IEEE80211_CCMP_HDR_LEN,
 				    data_len,
 				    skb->data + skb->len - mic_len))
-				return RX_DROP_UNUSABLE;
+				return RX_DROP_U_MIC_FAIL;
 		}
 
 		memcpy(key->u.ccmp.rx_pn[queue], pn, IEEE80211_CCMP_PN_LEN);
@@ -746,7 +746,7 @@ ieee80211_crypto_gcmp_decrypt(struct ieee80211_rx_data *rx)
 		if (res < 0 ||
 		    (!res && !(status->flag & RX_FLAG_ALLOW_SAME_PN))) {
 			key->u.gcmp.replays++;
-			return RX_DROP_UNUSABLE;
+			return RX_DROP_U_REPLAY;
 		}
 
 		if (!(status->flag & RX_FLAG_DECRYPTED)) {
@@ -761,7 +761,7 @@ ieee80211_crypto_gcmp_decrypt(struct ieee80211_rx_data *rx)
 				    data_len,
 				    skb->data + skb->len -
 				    IEEE80211_GCMP_MIC_LEN))
-				return RX_DROP_UNUSABLE;
+				return RX_DROP_U_MIC_FAIL;
 		}
 
 		memcpy(key->u.gcmp.rx_pn[queue], pn, IEEE80211_GCMP_PN_LEN);
@@ -930,13 +930,13 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx)
 		(skb->data + skb->len - sizeof(*mmie));
 	if (mmie->element_id != WLAN_EID_MMIE ||
 	    mmie->length != sizeof(*mmie) - 2)
-		return RX_DROP_UNUSABLE; /* Invalid MMIE */
+		return RX_DROP_U_BAD_MMIE; /* Invalid MMIE */
 
 	bip_ipn_swap(ipn, mmie->sequence_number);
 
 	if (memcmp(ipn, key->u.aes_cmac.rx_pn, 6) <= 0) {
 		key->u.aes_cmac.replays++;
-		return RX_DROP_UNUSABLE;
+		return RX_DROP_U_REPLAY;
 	}
 
 	if (!(status->flag & RX_FLAG_DECRYPTED)) {
@@ -946,7 +946,7 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx)
 				   skb->data + 24, skb->len - 24, mic);
 		if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) {
 			key->u.aes_cmac.icverrors++;
-			return RX_DROP_UNUSABLE;
+			return RX_DROP_U_MIC_FAIL;
 		}
 	}
 
@@ -986,7 +986,7 @@ ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx)
 
 	if (memcmp(ipn, key->u.aes_cmac.rx_pn, 6) <= 0) {
 		key->u.aes_cmac.replays++;
-		return RX_DROP_UNUSABLE;
+		return RX_DROP_U_REPLAY;
 	}
 
 	if (!(status->flag & RX_FLAG_DECRYPTED)) {
@@ -996,7 +996,7 @@ ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx)
 				       skb->data + 24, skb->len - 24, mic);
 		if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) {
 			key->u.aes_cmac.icverrors++;
-			return RX_DROP_UNUSABLE;
+			return RX_DROP_U_MIC_FAIL;
 		}
 	}
 
@@ -1079,13 +1079,13 @@ ieee80211_crypto_aes_gmac_decrypt(struct ieee80211_rx_data *rx)
 		(skb->data + skb->len - sizeof(*mmie));
 	if (mmie->element_id != WLAN_EID_MMIE ||
 	    mmie->length != sizeof(*mmie) - 2)
-		return RX_DROP_UNUSABLE; /* Invalid MMIE */
+		return RX_DROP_U_BAD_MMIE; /* Invalid MMIE */
 
 	bip_ipn_swap(ipn, mmie->sequence_number);
 
 	if (memcmp(ipn, key->u.aes_gmac.rx_pn, 6) <= 0) {
 		key->u.aes_gmac.replays++;
-		return RX_DROP_UNUSABLE;
+		return RX_DROP_U_REPLAY;
 	}
 
 	if (!(status->flag & RX_FLAG_DECRYPTED)) {
@@ -1104,7 +1104,7 @@ ieee80211_crypto_aes_gmac_decrypt(struct ieee80211_rx_data *rx)
 		    crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) {
 			key->u.aes_gmac.icverrors++;
 			kfree(mic);
-			return RX_DROP_UNUSABLE;
+			return RX_DROP_U_MIC_FAIL;
 		}
 		kfree(mic);
 	}
-- 
cgit v1.2.3-70-g09d2


From add7370a398930077c6bc257ef5016b040d476eb Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 19 Apr 2023 11:16:28 -0400
Subject: sctp: delete the nested flexible array params

This patch deletes the flexible-array params[] from the structure
sctp_inithdr, sctp_addiphdr and sctp_reconf_chunk to avoid some
sparse warnings:

  # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/
  net/sctp/input.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h):
  ./include/linux/sctp.h:278:29: warning: nested flexible array
  ./include/linux/sctp.h:675:30: warning: nested flexible array

This warning is reported if a structure having a flexible array
member is included by other structures.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sctp.h     |  6 +++---
 include/net/sctp/sctp.h  |  8 ++++----
 net/sctp/input.c         |  2 +-
 net/sctp/sm_make_chunk.c | 18 +++++++++---------
 net/sctp/sm_statefuns.c  |  2 +-
 net/sctp/stream.c        |  2 +-
 6 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/include/linux/sctp.h b/include/linux/sctp.h
index 358dc08e0831..0ff36a2737a3 100644
--- a/include/linux/sctp.h
+++ b/include/linux/sctp.h
@@ -270,7 +270,7 @@ struct sctp_inithdr {
 	__be16 num_outbound_streams;
 	__be16 num_inbound_streams;
 	__be32 initial_tsn;
-	__u8  params[];
+	/* __u8  params[]; */
 };
 
 struct sctp_init_chunk {
@@ -667,7 +667,7 @@ struct sctp_addip_param {
 
 struct sctp_addiphdr {
 	__be32	serial;
-	__u8	params[];
+	/* __u8	params[]; */
 };
 
 struct sctp_addip_chunk {
@@ -742,7 +742,7 @@ struct sctp_infox {
 
 struct sctp_reconf_chunk {
 	struct sctp_chunkhdr chunk_hdr;
-	__u8 params[];
+	/* __u8 params[]; */
 };
 
 struct sctp_strreset_outreq {
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index c335dd01a597..74fae532b944 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -425,11 +425,11 @@ static inline bool sctp_chunk_pending(const struct sctp_chunk *chunk)
  * the chunk length to indicate when to stop.  Make sure
  * there is room for a param header too.
  */
-#define sctp_walk_params(pos, chunk, member)\
-_sctp_walk_params((pos), (chunk), ntohs((chunk)->chunk_hdr.length), member)
+#define sctp_walk_params(pos, chunk)\
+_sctp_walk_params((pos), (chunk), ntohs((chunk)->chunk_hdr.length))
 
-#define _sctp_walk_params(pos, chunk, end, member)\
-for (pos.v = chunk->member;\
+#define _sctp_walk_params(pos, chunk, end)\
+for (pos.v = (u8 *)(chunk + 1);\
      (pos.v + offsetof(struct sctp_paramhdr, length) + sizeof(pos.p->length) <=\
       (void *)chunk + end) &&\
      pos.v <= (void *)chunk + end - ntohs(pos.p->length) &&\
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 127bf28a6033..2613c4d74b16 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -1150,7 +1150,7 @@ static struct sctp_association *__sctp_rcv_init_lookup(struct net *net,
 	init = (struct sctp_init_chunk *)skb->data;
 
 	/* Walk the parameters looking for embedded addresses. */
-	sctp_walk_params(params, init, init_hdr.params) {
+	sctp_walk_params(params, init) {
 
 		/* Note: Ignoring hostname addresses. */
 		af = sctp_get_af_specific(param_type2af(params.p->type));
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index c8f4ec5d5f98..4dbbbc2a7742 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -2306,7 +2306,7 @@ int sctp_verify_init(struct net *net, const struct sctp_endpoint *ep,
 	    ntohl(peer_init->init_hdr.a_rwnd) < SCTP_DEFAULT_MINWINDOW)
 		return sctp_process_inv_mandatory(asoc, chunk, errp);
 
-	sctp_walk_params(param, peer_init, init_hdr.params) {
+	sctp_walk_params(param, peer_init) {
 		if (param.p->type == SCTP_PARAM_STATE_COOKIE)
 			has_cookie = true;
 	}
@@ -2329,7 +2329,7 @@ int sctp_verify_init(struct net *net, const struct sctp_endpoint *ep,
 						  chunk, errp);
 
 	/* Verify all the variable length parameters */
-	sctp_walk_params(param, peer_init, init_hdr.params) {
+	sctp_walk_params(param, peer_init) {
 		result = sctp_verify_param(net, ep, asoc, param, cid,
 					   chunk, errp);
 		switch (result) {
@@ -2381,7 +2381,7 @@ int sctp_process_init(struct sctp_association *asoc, struct sctp_chunk *chunk,
 		src_match = 1;
 
 	/* Process the initialization parameters.  */
-	sctp_walk_params(param, peer_init, init_hdr.params) {
+	sctp_walk_params(param, peer_init) {
 		if (!src_match &&
 		    (param.p->type == SCTP_PARAM_IPV4_ADDRESS ||
 		     param.p->type == SCTP_PARAM_IPV6_ADDRESS)) {
@@ -3202,7 +3202,7 @@ bool sctp_verify_asconf(const struct sctp_association *asoc,
 	union sctp_params param;
 
 	addip = (struct sctp_addip_chunk *)chunk->chunk_hdr;
-	sctp_walk_params(param, addip, addip_hdr.params) {
+	sctp_walk_params(param, addip) {
 		size_t length = ntohs(param.p->length);
 
 		*errp = param.p;
@@ -3215,14 +3215,14 @@ bool sctp_verify_asconf(const struct sctp_association *asoc,
 			/* ensure there is only one addr param and it's in the
 			 * beginning of addip_hdr params, or we reject it.
 			 */
-			if (param.v != addip->addip_hdr.params)
+			if (param.v != (addip + 1))
 				return false;
 			addr_param_seen = true;
 			break;
 		case SCTP_PARAM_IPV6_ADDRESS:
 			if (length != sizeof(struct sctp_ipv6addr_param))
 				return false;
-			if (param.v != addip->addip_hdr.params)
+			if (param.v != (addip + 1))
 				return false;
 			addr_param_seen = true;
 			break;
@@ -3302,7 +3302,7 @@ struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc,
 		goto done;
 
 	/* Process the TLVs contained within the ASCONF chunk. */
-	sctp_walk_params(param, addip, addip_hdr.params) {
+	sctp_walk_params(param, addip) {
 		/* Skip preceeding address parameters. */
 		if (param.p->type == SCTP_PARAM_IPV4_ADDRESS ||
 		    param.p->type == SCTP_PARAM_IPV6_ADDRESS)
@@ -3636,7 +3636,7 @@ static struct sctp_chunk *sctp_make_reconf(const struct sctp_association *asoc,
 		return NULL;
 
 	reconf = (struct sctp_reconf_chunk *)retval->chunk_hdr;
-	retval->param_hdr.v = reconf->params;
+	retval->param_hdr.v = (u8 *)(reconf + 1);
 
 	return retval;
 }
@@ -3878,7 +3878,7 @@ bool sctp_verify_reconf(const struct sctp_association *asoc,
 	__u16 cnt = 0;
 
 	hdr = (struct sctp_reconf_chunk *)chunk->chunk_hdr;
-	sctp_walk_params(param, hdr, params) {
+	sctp_walk_params(param, hdr) {
 		__u16 length = ntohs(param.p->length);
 
 		*errp = param.p;
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index ce5426171206..39d416e7f795 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -4142,7 +4142,7 @@ enum sctp_disposition sctp_sf_do_reconf(struct net *net,
 						  (void *)err_param, commands);
 
 	hdr = (struct sctp_reconf_chunk *)chunk->chunk_hdr;
-	sctp_walk_params(param, hdr, params) {
+	sctp_walk_params(param, hdr) {
 		struct sctp_chunk *reply = NULL;
 		struct sctp_ulpevent *ev = NULL;
 
diff --git a/net/sctp/stream.c b/net/sctp/stream.c
index ee6514af830f..c241cc552e8d 100644
--- a/net/sctp/stream.c
+++ b/net/sctp/stream.c
@@ -491,7 +491,7 @@ static struct sctp_paramhdr *sctp_chunk_lookup_strreset_param(
 		return NULL;
 
 	hdr = (struct sctp_reconf_chunk *)chunk->chunk_hdr;
-	sctp_walk_params(param, hdr, params) {
+	sctp_walk_params(param, hdr) {
 		/* sctp_strreset_tsnreq is actually the basic structure
 		 * of all stream reconf params, so it's safe to use it
 		 * to access request_seq.
-- 
cgit v1.2.3-70-g09d2


From 73175a042955e531ec355a8708585befa67a22db Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 19 Apr 2023 11:16:29 -0400
Subject: sctp: delete the nested flexible array skip

This patch deletes the flexible-array skip[] from the structure
sctp_ifwdtsn/fwdtsn_hdr to avoid some sparse warnings:

  # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/
  net/sctp/stream_interleave.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h):
  ./include/linux/sctp.h:611:32: warning: nested flexible array
  ./include/linux/sctp.h:628:33: warning: nested flexible array

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sctp.h         | 4 ++--
 include/net/sctp/sctp.h      | 4 ++--
 net/sctp/stream_interleave.c | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/linux/sctp.h b/include/linux/sctp.h
index 0ff36a2737a3..9815b801fec0 100644
--- a/include/linux/sctp.h
+++ b/include/linux/sctp.h
@@ -603,7 +603,7 @@ struct sctp_fwdtsn_skip {
 
 struct sctp_fwdtsn_hdr {
 	__be32 new_cum_tsn;
-	struct sctp_fwdtsn_skip skip[];
+	/* struct sctp_fwdtsn_skip skip[]; */
 };
 
 struct sctp_fwdtsn_chunk {
@@ -620,7 +620,7 @@ struct sctp_ifwdtsn_skip {
 
 struct sctp_ifwdtsn_hdr {
 	__be32 new_cum_tsn;
-	struct sctp_ifwdtsn_skip skip[];
+	/* struct sctp_ifwdtsn_skip skip[]; */
 };
 
 struct sctp_ifwdtsn_chunk {
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 74fae532b944..2a67100b2a17 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -452,8 +452,8 @@ for (err = (struct sctp_errhdr *)((void *)chunk_hdr + \
 _sctp_walk_fwdtsn((pos), (chunk), ntohs((chunk)->chunk_hdr->length) - sizeof(struct sctp_fwdtsn_chunk))
 
 #define _sctp_walk_fwdtsn(pos, chunk, end)\
-for (pos = chunk->subh.fwdtsn_hdr->skip;\
-     (void *)pos <= (void *)chunk->subh.fwdtsn_hdr->skip + end - sizeof(struct sctp_fwdtsn_skip);\
+for (pos = (void *)(chunk->subh.fwdtsn_hdr + 1);\
+     (void *)pos <= (void *)(chunk->subh.fwdtsn_hdr + 1) + end - sizeof(struct sctp_fwdtsn_skip);\
      pos++)
 
 /* External references. */
diff --git a/net/sctp/stream_interleave.c b/net/sctp/stream_interleave.c
index b046b11200c9..840f24045ae2 100644
--- a/net/sctp/stream_interleave.c
+++ b/net/sctp/stream_interleave.c
@@ -1153,8 +1153,8 @@ static void sctp_generate_iftsn(struct sctp_outq *q, __u32 ctsn)
 }
 
 #define _sctp_walk_ifwdtsn(pos, chunk, end) \
-	for (pos = chunk->subh.ifwdtsn_hdr->skip; \
-	     (void *)pos <= (void *)chunk->subh.ifwdtsn_hdr->skip + (end) - \
+	for (pos = (void *)(chunk->subh.ifwdtsn_hdr + 1); \
+	     (void *)pos <= (void *)(chunk->subh.ifwdtsn_hdr + 1) + (end) - \
 			    sizeof(struct sctp_ifwdtsn_skip); pos++)
 
 #define sctp_walk_ifwdtsn(pos, ch) \
-- 
cgit v1.2.3-70-g09d2


From 9789c1c6619e0a5eccfc31abe49b1ce5ca3cd11f Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 19 Apr 2023 11:16:30 -0400
Subject: sctp: delete the nested flexible array variable

This patch deletes the flexible-array variable[] from the structure
sctp_sackhdr and sctp_errhdr to avoid some sparse warnings:

  # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/
  net/sctp/sm_statefuns.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h):
  ./include/linux/sctp.h:451:28: warning: nested flexible array
  ./include/linux/sctp.h:393:29: warning: nested flexible array

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sctp.h     |  4 ++--
 net/sctp/outqueue.c      | 11 +++++++----
 net/sctp/sm_sideeffect.c |  3 +--
 net/sctp/sm_statefuns.c  |  2 +-
 4 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/include/linux/sctp.h b/include/linux/sctp.h
index 9815b801fec0..01a0eb7e9fa1 100644
--- a/include/linux/sctp.h
+++ b/include/linux/sctp.h
@@ -385,7 +385,7 @@ struct sctp_sackhdr {
 	__be32 a_rwnd;
 	__be16 num_gap_ack_blocks;
 	__be16 num_dup_tsns;
-	union sctp_sack_variable variable[];
+	/* union sctp_sack_variable variable[]; */
 };
 
 struct sctp_sack_chunk {
@@ -443,7 +443,7 @@ struct sctp_shutdown_chunk {
 struct sctp_errhdr {
 	__be16 cause;
 	__be16 length;
-	__u8  variable[];
+	/* __u8  variable[]; */
 };
 
 struct sctp_operr_chunk {
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 20831079fb09..0dc6b8ab9963 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -1231,7 +1231,7 @@ static void sctp_sack_update_unack_data(struct sctp_association *assoc,
 
 	unack_data = assoc->next_tsn - assoc->ctsn_ack_point - 1;
 
-	frags = sack->variable;
+	frags = (union sctp_sack_variable *)(sack + 1);
 	for (i = 0; i < ntohs(sack->num_gap_ack_blocks); i++) {
 		unack_data -= ((ntohs(frags[i].gab.end) -
 				ntohs(frags[i].gab.start) + 1));
@@ -1252,7 +1252,6 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk)
 	struct sctp_transport *transport;
 	struct sctp_chunk *tchunk = NULL;
 	struct list_head *lchunk, *transport_list, *temp;
-	union sctp_sack_variable *frags = sack->variable;
 	__u32 sack_ctsn, ctsn, tsn;
 	__u32 highest_tsn, highest_new_tsn;
 	__u32 sack_a_rwnd;
@@ -1313,8 +1312,12 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk)
 
 	/* Get the highest TSN in the sack. */
 	highest_tsn = sack_ctsn;
-	if (gap_ack_blocks)
+	if (gap_ack_blocks) {
+		union sctp_sack_variable *frags =
+			(union sctp_sack_variable *)(sack + 1);
+
 		highest_tsn += ntohs(frags[gap_ack_blocks - 1].gab.end);
+	}
 
 	if (TSN_lt(asoc->highest_sacked, highest_tsn))
 		asoc->highest_sacked = highest_tsn;
@@ -1789,7 +1792,7 @@ static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn)
 	 *  Block are assumed to have been received correctly.
 	 */
 
-	frags = sack->variable;
+	frags = (union sctp_sack_variable *)(sack + 1);
 	blocks = ntohs(sack->num_gap_ack_blocks);
 	tsn_offset = tsn - ctsn;
 	for (i = 0; i < blocks; ++i) {
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index 463c4a58d2c3..7fbeb99d8d32 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -984,8 +984,7 @@ static void sctp_cmd_process_operr(struct sctp_cmd_seq *cmds,
 		{
 			struct sctp_chunkhdr *unk_chunk_hdr;
 
-			unk_chunk_hdr = (struct sctp_chunkhdr *)
-							err_hdr->variable;
+			unk_chunk_hdr = (struct sctp_chunkhdr *)(err_hdr + 1);
 			switch (unk_chunk_hdr->type) {
 			/* ADDIP 4.1 A9) If the peer responds to an ASCONF with
 			 * an ERROR chunk reporting that it did not recognized
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 39d416e7f795..8d0cfd689b20 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -1337,7 +1337,7 @@ static int sctp_sf_send_restart_abort(struct net *net, union sctp_addr *ssa,
 	 * throughout the code today.
 	 */
 	errhdr = (struct sctp_errhdr *)buffer;
-	addrparm = (union sctp_addr_param *)errhdr->variable;
+	addrparm = (union sctp_addr_param *)(errhdr + 1);
 
 	/* Copy into a parm format. */
 	len = af->to_addr_param(ssa, addrparm);
-- 
cgit v1.2.3-70-g09d2


From f97278ff346a5f11c68d0ac499999d5ad06a6db2 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 19 Apr 2023 11:16:31 -0400
Subject: sctp: delete the nested flexible array peer_init

This patch deletes the flexible-array peer_init[] from the structure
sctp_cookie to avoid some sparse warnings:

  # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/
  net/sctp/sm_make_chunk.c: note: in included file (through include/net/sctp/sctp.h):
  ./include/net/sctp/structs.h:1588:28: warning: nested flexible array
  ./include/net/sctp/structs.h:343:28: warning: nested flexible array

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h | 2 +-
 net/sctp/associola.c       | 5 +++--
 net/sctp/sm_make_chunk.c   | 4 ++--
 net/sctp/sm_statefuns.c    | 8 +++-----
 4 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 070c9458fff4..5c72d1864dd6 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -332,7 +332,7 @@ struct sctp_cookie {
 	 * the association TCB is re-constructed from the cookie.
 	 */
 	__u32 raw_addr_list_len;
-	struct sctp_init_chunk peer_init[];
+	/* struct sctp_init_chunk peer_init[]; */
 };
 
 
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 63ba5551c13f..796529167e8d 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -1597,9 +1597,10 @@ int sctp_assoc_set_bind_addr_from_cookie(struct sctp_association *asoc,
 					 struct sctp_cookie *cookie,
 					 gfp_t gfp)
 {
-	int var_size2 = ntohs(cookie->peer_init->chunk_hdr.length);
+	struct sctp_init_chunk *peer_init = (struct sctp_init_chunk *)(cookie + 1);
+	int var_size2 = ntohs(peer_init->chunk_hdr.length);
 	int var_size3 = cookie->raw_addr_list_len;
-	__u8 *raw = (__u8 *)cookie->peer_init + var_size2;
+	__u8 *raw = (__u8 *)peer_init + var_size2;
 
 	return sctp_raw_to_bind_addrs(&asoc->base.bind_addr, raw, var_size3,
 				      asoc->ep->base.bind_addr.port, gfp);
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 4dbbbc2a7742..08527d882e56 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1707,11 +1707,11 @@ static struct sctp_cookie_param *sctp_pack_cookie(
 					 ktime_get_real());
 
 	/* Copy the peer's init packet.  */
-	memcpy(&cookie->c.peer_init[0], init_chunk->chunk_hdr,
+	memcpy(cookie + 1, init_chunk->chunk_hdr,
 	       ntohs(init_chunk->chunk_hdr->length));
 
 	/* Copy the raw local address list of the association. */
-	memcpy((__u8 *)&cookie->c.peer_init[0] +
+	memcpy((__u8 *)(cookie + 1) +
 	       ntohs(init_chunk->chunk_hdr->length), raw_addrs, addrs_len);
 
 	if (sctp_sk(ep->base.sk)->hmac) {
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 8d0cfd689b20..7b8eb735fa88 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -794,8 +794,7 @@ enum sctp_disposition sctp_sf_do_5_1D_ce(struct net *net,
 	/* This is a brand-new association, so these are not yet side
 	 * effects--it is safe to run them here.
 	 */
-	peer_init = &chunk->subh.cookie_hdr->c.peer_init[0];
-
+	peer_init = (struct sctp_init_chunk *)(chunk->subh.cookie_hdr + 1);
 	if (!sctp_process_init(new_asoc, chunk,
 			       &chunk->subh.cookie_hdr->c.peer_addr,
 			       peer_init, GFP_ATOMIC))
@@ -1869,8 +1868,7 @@ static enum sctp_disposition sctp_sf_do_dupcook_a(
 	/* new_asoc is a brand-new association, so these are not yet
 	 * side effects--it is safe to run them here.
 	 */
-	peer_init = &chunk->subh.cookie_hdr->c.peer_init[0];
-
+	peer_init = (struct sctp_init_chunk *)(chunk->subh.cookie_hdr + 1);
 	if (!sctp_process_init(new_asoc, chunk, sctp_source(chunk), peer_init,
 			       GFP_ATOMIC))
 		goto nomem;
@@ -1990,7 +1988,7 @@ static enum sctp_disposition sctp_sf_do_dupcook_b(
 	/* new_asoc is a brand-new association, so these are not yet
 	 * side effects--it is safe to run them here.
 	 */
-	peer_init = &chunk->subh.cookie_hdr->c.peer_init[0];
+	peer_init = (struct sctp_init_chunk *)(chunk->subh.cookie_hdr + 1);
 	if (!sctp_process_init(new_asoc, chunk, sctp_source(chunk), peer_init,
 			       GFP_ATOMIC))
 		goto nomem;
-- 
cgit v1.2.3-70-g09d2


From 2ab399a931dddacdf7202cd4b49a5187154623d1 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 19 Apr 2023 11:16:32 -0400
Subject: sctp: delete the nested flexible array hmac

This patch deletes the flexible-array hmac[] from the structure
sctp_authhdr to avoid some sparse warnings:

  # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/
  net/sctp/auth.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h):
  ./include/linux/sctp.h:735:29: warning: nested flexible array

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sctp.h    | 2 +-
 net/sctp/auth.c         | 2 +-
 net/sctp/sm_statefuns.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/sctp.h b/include/linux/sctp.h
index 01a0eb7e9fa1..d182e8c41985 100644
--- a/include/linux/sctp.h
+++ b/include/linux/sctp.h
@@ -727,7 +727,7 @@ struct sctp_addip_chunk {
 struct sctp_authhdr {
 	__be16 shkey_id;
 	__be16 hmac_id;
-	__u8   hmac[];
+	/* __u8   hmac[]; */
 };
 
 struct sctp_auth_chunk {
diff --git a/net/sctp/auth.c b/net/sctp/auth.c
index 34964145514e..c58fffc86a0c 100644
--- a/net/sctp/auth.c
+++ b/net/sctp/auth.c
@@ -738,7 +738,7 @@ void sctp_auth_calculate_hmac(const struct sctp_association *asoc,
 
 	tfm = asoc->ep->auth_hmacs[hmac_id];
 
-	digest = auth->auth_hdr.hmac;
+	digest = (u8 *)(&auth->auth_hdr + 1);
 	if (crypto_shash_setkey(tfm, &asoc_key->data[0], asoc_key->len))
 		goto free;
 
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 7b8eb735fa88..97f1155a2045 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -4391,7 +4391,7 @@ static enum sctp_ierror sctp_sf_authenticate(
 	 *  3. Compute the new digest
 	 *  4. Compare saved and new digests.
 	 */
-	digest = auth_hdr->hmac;
+	digest = (u8 *)(auth_hdr + 1);
 	skb_pull(chunk->skb, sig_len);
 
 	save_digest = kmemdup(digest, sig_len, GFP_ATOMIC);
-- 
cgit v1.2.3-70-g09d2


From dbda0fba7a14f14835c34d59fd329cb90a887862 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Wed, 19 Apr 2023 11:16:33 -0400
Subject: sctp: delete the nested flexible array payload

This patch deletes the flexible-array payload[] from the structure
sctp_datahdr to avoid some sparse warnings:

  # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/
  net/sctp/socket.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h):
  ./include/linux/sctp.h:230:29: warning: nested flexible array

This member is not even used anywhere.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sctp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/sctp.h b/include/linux/sctp.h
index d182e8c41985..836a7e200f39 100644
--- a/include/linux/sctp.h
+++ b/include/linux/sctp.h
@@ -222,7 +222,7 @@ struct sctp_datahdr {
 	__be16 stream;
 	__be16 ssn;
 	__u32 ppid;
-	__u8  payload[];
+	/* __u8  payload[]; */
 };
 
 struct sctp_data_chunk {
-- 
cgit v1.2.3-70-g09d2


From abff3e5e2935f5cc34feb559b352179b13eaa066 Mon Sep 17 00:00:00 2001
From: Emeel Hakim <ehakim@nvidia.com>
Date: Wed, 19 Apr 2023 17:21:22 +0300
Subject: vlan: Add MACsec offload operations for VLAN interface

Add support for MACsec offload operations for VLAN driver
to allow offloading MACsec when VLAN's real device supports
Macsec offload by forwarding the offload request to it.

Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/8021q/vlan_dev.c | 242 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 242 insertions(+)

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 5920544e93e8..870e4935d6e6 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -26,6 +26,7 @@
 #include <linux/ethtool.h>
 #include <linux/phy.h>
 #include <net/arp.h>
+#include <net/macsec.h>
 
 #include "vlan.h"
 #include "vlanproc.h"
@@ -572,6 +573,9 @@ static int vlan_dev_init(struct net_device *dev)
 			   NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC |
 			   NETIF_F_ALL_FCOE;
 
+	if (real_dev->vlan_features & NETIF_F_HW_MACSEC)
+		dev->hw_features |= NETIF_F_HW_MACSEC;
+
 	dev->features |= dev->hw_features | NETIF_F_LLTX;
 	netif_inherit_tso_max(dev, real_dev);
 	if (dev->features & NETIF_F_VLAN_FEATURES)
@@ -803,6 +807,241 @@ static int vlan_dev_fill_forward_path(struct net_device_path_ctx *ctx,
 	return 0;
 }
 
+#if IS_ENABLED(CONFIG_MACSEC)
+
+static const struct macsec_ops *vlan_get_macsec_ops(const struct macsec_context *ctx)
+{
+	return vlan_dev_priv(ctx->netdev)->real_dev->macsec_ops;
+}
+
+static int vlan_macsec_offload(int (* const func)(struct macsec_context *),
+			       struct macsec_context *ctx)
+{
+	if (unlikely(!func))
+		return 0;
+
+	return (*func)(ctx);
+}
+
+static int vlan_macsec_dev_open(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_dev_open, ctx);
+}
+
+static int vlan_macsec_dev_stop(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_dev_stop, ctx);
+}
+
+static int vlan_macsec_add_secy(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_add_secy, ctx);
+}
+
+static int vlan_macsec_upd_secy(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_upd_secy, ctx);
+}
+
+static int vlan_macsec_del_secy(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_del_secy, ctx);
+}
+
+static int vlan_macsec_add_rxsc(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_add_rxsc, ctx);
+}
+
+static int vlan_macsec_upd_rxsc(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_upd_rxsc, ctx);
+}
+
+static int vlan_macsec_del_rxsc(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_del_rxsc, ctx);
+}
+
+static int vlan_macsec_add_rxsa(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_add_rxsa, ctx);
+}
+
+static int vlan_macsec_upd_rxsa(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_upd_rxsa, ctx);
+}
+
+static int vlan_macsec_del_rxsa(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_del_rxsa, ctx);
+}
+
+static int vlan_macsec_add_txsa(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_add_txsa, ctx);
+}
+
+static int vlan_macsec_upd_txsa(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_upd_txsa, ctx);
+}
+
+static int vlan_macsec_del_txsa(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_del_txsa, ctx);
+}
+
+static int vlan_macsec_get_dev_stats(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_get_dev_stats, ctx);
+}
+
+static int vlan_macsec_get_tx_sc_stats(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_get_tx_sc_stats, ctx);
+}
+
+static int vlan_macsec_get_tx_sa_stats(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_get_tx_sa_stats, ctx);
+}
+
+static int vlan_macsec_get_rx_sc_stats(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_get_rx_sc_stats, ctx);
+}
+
+static int vlan_macsec_get_rx_sa_stats(struct macsec_context *ctx)
+{
+	const struct macsec_ops *ops = vlan_get_macsec_ops(ctx);
+
+	if (!ops)
+		return -EOPNOTSUPP;
+
+	return vlan_macsec_offload(ops->mdo_get_rx_sa_stats, ctx);
+}
+
+static const struct macsec_ops macsec_offload_ops = {
+	/* Device wide */
+	.mdo_dev_open = vlan_macsec_dev_open,
+	.mdo_dev_stop = vlan_macsec_dev_stop,
+	/* SecY */
+	.mdo_add_secy = vlan_macsec_add_secy,
+	.mdo_upd_secy = vlan_macsec_upd_secy,
+	.mdo_del_secy = vlan_macsec_del_secy,
+	/* Security channels */
+	.mdo_add_rxsc = vlan_macsec_add_rxsc,
+	.mdo_upd_rxsc = vlan_macsec_upd_rxsc,
+	.mdo_del_rxsc = vlan_macsec_del_rxsc,
+	/* Security associations */
+	.mdo_add_rxsa = vlan_macsec_add_rxsa,
+	.mdo_upd_rxsa = vlan_macsec_upd_rxsa,
+	.mdo_del_rxsa = vlan_macsec_del_rxsa,
+	.mdo_add_txsa = vlan_macsec_add_txsa,
+	.mdo_upd_txsa = vlan_macsec_upd_txsa,
+	.mdo_del_txsa = vlan_macsec_del_txsa,
+	/* Statistics */
+	.mdo_get_dev_stats = vlan_macsec_get_dev_stats,
+	.mdo_get_tx_sc_stats = vlan_macsec_get_tx_sc_stats,
+	.mdo_get_tx_sa_stats = vlan_macsec_get_tx_sa_stats,
+	.mdo_get_rx_sc_stats = vlan_macsec_get_rx_sc_stats,
+	.mdo_get_rx_sa_stats = vlan_macsec_get_rx_sa_stats,
+};
+
+#endif
+
 static const struct ethtool_ops vlan_ethtool_ops = {
 	.get_link_ksettings	= vlan_ethtool_get_link_ksettings,
 	.get_drvinfo	        = vlan_ethtool_get_drvinfo,
@@ -869,6 +1108,9 @@ void vlan_setup(struct net_device *dev)
 	dev->priv_destructor	= vlan_dev_free;
 	dev->ethtool_ops	= &vlan_ethtool_ops;
 
+#if IS_ENABLED(CONFIG_MACSEC)
+	dev->macsec_ops		= &macsec_offload_ops;
+#endif
 	dev->min_mtu		= 0;
 	dev->max_mtu		= ETH_MAX_MTU;
 
-- 
cgit v1.2.3-70-g09d2


From 339ccec8d43d004a377b01e1a9a85487f95e8f13 Mon Sep 17 00:00:00 2001
From: Emeel Hakim <ehakim@nvidia.com>
Date: Wed, 19 Apr 2023 17:21:23 +0300
Subject: net/mlx5: Enable MACsec offload feature for VLAN interface

Enable MACsec offload feature over VLAN by adding NETIF_F_HW_MACSEC
to the device vlan_features.

Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 7eb1eeb115ca..2fda7385de71 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -5109,6 +5109,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 
 	netdev->vlan_features    |= NETIF_F_SG;
 	netdev->vlan_features    |= NETIF_F_HW_CSUM;
+	netdev->vlan_features    |= NETIF_F_HW_MACSEC;
 	netdev->vlan_features    |= NETIF_F_GRO;
 	netdev->vlan_features    |= NETIF_F_TSO;
 	netdev->vlan_features    |= NETIF_F_TSO6;
-- 
cgit v1.2.3-70-g09d2


From 4bba492b04270c29259594d1b9ee66ce96af9466 Mon Sep 17 00:00:00 2001
From: Emeel Hakim <ehakim@nvidia.com>
Date: Wed, 19 Apr 2023 17:21:24 +0300
Subject: net/mlx5: Support MACsec over VLAN

MACsec device may have a VLAN device on top of it.
Detect MACsec state correctly under this condition,
and return the correct net device accordingly.

Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/mellanox/mlx5/core/en_accel/macsec.c  | 42 +++++++++++++---------
 1 file changed, 26 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
index 33b3620ea45c..f3428dbeb298 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
@@ -4,6 +4,7 @@
 #include <linux/mlx5/device.h>
 #include <linux/mlx5/mlx5_ifc.h>
 #include <linux/xarray.h>
+#include <linux/if_vlan.h>
 
 #include "en.h"
 #include "lib/aso.h"
@@ -348,12 +349,21 @@ static void mlx5e_macsec_cleanup_sa(struct mlx5e_macsec *macsec,
 	sa->macsec_rule = NULL;
 }
 
+static struct mlx5e_priv *macsec_netdev_priv(const struct net_device *dev)
+{
+#if IS_ENABLED(CONFIG_VLAN_8021Q)
+	if (is_vlan_dev(dev))
+		return netdev_priv(vlan_dev_priv(dev)->real_dev);
+#endif
+	return netdev_priv(dev);
+}
+
 static int mlx5e_macsec_init_sa(struct macsec_context *ctx,
 				struct mlx5e_macsec_sa *sa,
 				bool encrypt,
 				bool is_tx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	struct mlx5e_macsec *macsec = priv->macsec;
 	struct mlx5_macsec_rule_attrs rule_attrs;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -427,7 +437,7 @@ static int macsec_rx_sa_active_update(struct macsec_context *ctx,
 				      struct mlx5e_macsec_sa *rx_sa,
 				      bool active)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	struct mlx5e_macsec *macsec = priv->macsec;
 	int err = 0;
 
@@ -508,9 +518,9 @@ static void update_macsec_epn(struct mlx5e_macsec_sa *sa, const struct macsec_ke
 
 static int mlx5e_macsec_add_txsa(struct macsec_context *ctx)
 {
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_tx_sc *tx_sc = &ctx->secy->tx_sc;
 	const struct macsec_tx_sa *ctx_tx_sa = ctx->sa.tx_sa;
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	const struct macsec_secy *secy = ctx->secy;
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5_core_dev *mdev = priv->mdev;
@@ -583,9 +593,9 @@ out:
 
 static int mlx5e_macsec_upd_txsa(struct macsec_context *ctx)
 {
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_tx_sc *tx_sc = &ctx->secy->tx_sc;
 	const struct macsec_tx_sa *ctx_tx_sa = ctx->sa.tx_sa;
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	u8 assoc_num = ctx->sa.assoc_num;
 	struct mlx5e_macsec_sa *tx_sa;
@@ -645,7 +655,7 @@ out:
 
 static int mlx5e_macsec_del_txsa(struct macsec_context *ctx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	u8 assoc_num = ctx->sa.assoc_num;
 	struct mlx5e_macsec_sa *tx_sa;
@@ -696,7 +706,7 @@ static u32 mlx5e_macsec_get_sa_from_hashtable(struct rhashtable *sci_hash, sci_t
 static int mlx5e_macsec_add_rxsc(struct macsec_context *ctx)
 {
 	struct mlx5e_macsec_rx_sc_xarray_element *sc_xarray_element;
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_rx_sc *ctx_rx_sc = ctx->rx_sc;
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_rx_sc *rx_sc;
@@ -776,7 +786,7 @@ out:
 
 static int mlx5e_macsec_upd_rxsc(struct macsec_context *ctx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_rx_sc *ctx_rx_sc = ctx->rx_sc;
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_rx_sc *rx_sc;
@@ -854,7 +864,7 @@ static void macsec_del_rxsc_ctx(struct mlx5e_macsec *macsec, struct mlx5e_macsec
 
 static int mlx5e_macsec_del_rxsc(struct macsec_context *ctx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_rx_sc *rx_sc;
 	struct mlx5e_macsec *macsec;
@@ -890,8 +900,8 @@ out:
 
 static int mlx5e_macsec_add_rxsa(struct macsec_context *ctx)
 {
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_rx_sa *ctx_rx_sa = ctx->sa.rx_sa;
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5_core_dev *mdev = priv->mdev;
 	u8 assoc_num = ctx->sa.assoc_num;
@@ -976,8 +986,8 @@ out:
 
 static int mlx5e_macsec_upd_rxsa(struct macsec_context *ctx)
 {
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_rx_sa *ctx_rx_sa = ctx->sa.rx_sa;
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	u8 assoc_num = ctx->sa.assoc_num;
 	struct mlx5e_macsec_rx_sc *rx_sc;
@@ -1033,7 +1043,7 @@ out:
 
 static int mlx5e_macsec_del_rxsa(struct macsec_context *ctx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	sci_t sci = ctx->sa.rx_sa->sc->sci;
 	struct mlx5e_macsec_rx_sc *rx_sc;
@@ -1085,7 +1095,7 @@ out:
 
 static int mlx5e_macsec_add_secy(struct macsec_context *ctx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct net_device *dev = ctx->secy->netdev;
 	const struct net_device *netdev = ctx->netdev;
 	struct mlx5e_macsec_device *macsec_device;
@@ -1137,7 +1147,7 @@ out:
 static int macsec_upd_secy_hw_address(struct macsec_context *ctx,
 				      struct mlx5e_macsec_device *macsec_device)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct net_device *dev = ctx->secy->netdev;
 	struct mlx5e_macsec *macsec = priv->macsec;
 	struct mlx5e_macsec_rx_sc *rx_sc, *tmp;
@@ -1184,8 +1194,8 @@ out:
  */
 static int mlx5e_macsec_upd_secy(struct macsec_context *ctx)
 {
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	const struct macsec_tx_sc *tx_sc = &ctx->secy->tx_sc;
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
 	const struct net_device *dev = ctx->secy->netdev;
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_sa *tx_sa;
@@ -1240,7 +1250,7 @@ out:
 
 static int mlx5e_macsec_del_secy(struct macsec_context *ctx)
 {
-	struct mlx5e_priv *priv = netdev_priv(ctx->netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev);
 	struct mlx5e_macsec_device *macsec_device;
 	struct mlx5e_macsec_rx_sc *rx_sc, *tmp;
 	struct mlx5e_macsec_sa *tx_sa;
@@ -1741,7 +1751,7 @@ void mlx5e_macsec_offload_handle_rx_skb(struct net_device *netdev,
 {
 	struct mlx5e_macsec_rx_sc_xarray_element *sc_xarray_element;
 	u32 macsec_meta_data = be32_to_cpu(cqe->ft_metadata);
-	struct mlx5e_priv *priv = netdev_priv(netdev);
+	struct mlx5e_priv *priv = macsec_netdev_priv(netdev);
 	struct mlx5e_macsec_rx_sc *rx_sc;
 	struct mlx5e_macsec *macsec;
 	u32  fs_id;
-- 
cgit v1.2.3-70-g09d2


From 765f974c7dfd104ed569fae1669adfc0e70ded1b Mon Sep 17 00:00:00 2001
From: Emeel Hakim <ehakim@nvidia.com>
Date: Wed, 19 Apr 2023 17:21:25 +0300
Subject: net/mlx5: Consider VLAN interface in MACsec TX steering rules

Offloading MACsec when its configured over VLAN with current MACsec
TX steering rules will wrongly insert MACsec sec tag after inserting
the VLAN header leading to a ETHERNET | SECTAG | VLAN packet when
ETHERNET | VLAN | SECTAG is configured.

The above issue is due to adding the SECTAG by HW which is a later
stage compared to the VLAN header insertion stage.

Detect such a case and adjust TX steering rules to insert the
SECTAG in the correct place by using reformat_param_0 field in
the packet reformat to indicate the offset of SECTAG from end of
the MAC header to account for VLANs in granularity of 4Bytes.

Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
index 9173b67becef..7fc901a6ec5f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
@@ -4,6 +4,7 @@
 #include <net/macsec.h>
 #include <linux/netdevice.h>
 #include <linux/mlx5/qp.h>
+#include <linux/if_vlan.h>
 #include "fs_core.h"
 #include "en/fs.h"
 #include "en_accel/macsec_fs.h"
@@ -508,6 +509,8 @@ static void macsec_fs_tx_del_rule(struct mlx5e_macsec_fs *macsec_fs,
 	macsec_fs_tx_ft_put(macsec_fs);
 }
 
+#define MLX5_REFORMAT_PARAM_ADD_MACSEC_OFFSET_4_BYTES 1
+
 static union mlx5e_macsec_rule *
 macsec_fs_tx_add_rule(struct mlx5e_macsec_fs *macsec_fs,
 		      const struct macsec_context *macsec_ctx,
@@ -553,6 +556,10 @@ macsec_fs_tx_add_rule(struct mlx5e_macsec_fs *macsec_fs,
 	reformat_params.type = MLX5_REFORMAT_TYPE_ADD_MACSEC;
 	reformat_params.size = reformat_size;
 	reformat_params.data = reformatbf;
+
+	if (is_vlan_dev(macsec_ctx->netdev))
+		reformat_params.param_0 = MLX5_REFORMAT_PARAM_ADD_MACSEC_OFFSET_4_BYTES;
+
 	flow_act.pkt_reformat = mlx5_packet_reformat_alloc(macsec_fs->mdev,
 							   &reformat_params,
 							   MLX5_FLOW_NAMESPACE_EGRESS_MACSEC);
-- 
cgit v1.2.3-70-g09d2


From 7661351a54ec9a6a20203f94fd459a9360049045 Mon Sep 17 00:00:00 2001
From: Emeel Hakim <ehakim@nvidia.com>
Date: Wed, 19 Apr 2023 17:21:26 +0300
Subject: macsec: Don't rely solely on the dst MAC address to identify
 destination MACsec device

Offloading device drivers will mark offloaded MACsec SKBs with the
corresponding SCI in the skb_metadata_dst so the macsec rx handler will
know to which interface to divert those skbs, in case of a marked skb
and a mismatch on the dst MAC address, divert the skb to the macsec
net_device where the macsec rx_handler will be called to consider cases
where relying solely on the dst MAC address is insufficient.

One such instance is when using MACsec with a VLAN as an inner
header, where the packet structure is ETHERNET | SECTAG | VLAN.
In such a scenario, the dst MAC address in the ethernet header
will correspond to the VLAN MAC address, resulting in a mismatch.

Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macsec.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index 25616247d7a5..3427993f94f7 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -1021,8 +1021,12 @@ static enum rx_handler_result handle_not_macsec(struct sk_buff *skb)
 		 * the SecTAG, so we have to deduce which port to deliver to.
 		 */
 		if (macsec_is_offloaded(macsec) && netif_running(ndev)) {
-			if (md_dst && md_dst->type == METADATA_MACSEC &&
-			    (!find_rx_sc(&macsec->secy, md_dst->u.macsec_info.sci)))
+			struct macsec_rx_sc *rx_sc = NULL;
+
+			if (md_dst && md_dst->type == METADATA_MACSEC)
+				rx_sc = find_rx_sc(&macsec->secy, md_dst->u.macsec_info.sci);
+
+			if (md_dst && md_dst->type == METADATA_MACSEC && !rx_sc)
 				continue;
 
 			if (ether_addr_equal_64bits(hdr->h_dest,
@@ -1047,7 +1051,13 @@ static enum rx_handler_result handle_not_macsec(struct sk_buff *skb)
 					nskb->pkt_type = PACKET_MULTICAST;
 
 				__netif_rx(nskb);
+			} else if (rx_sc || ndev->flags & IFF_PROMISC) {
+				skb->dev = ndev;
+				skb->pkt_type = PACKET_HOST;
+				ret = RX_HANDLER_ANOTHER;
+				goto out;
 			}
+
 			continue;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 013a7ce81dd8a5bba31f802053829d91f9a43a8f Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:52 +0300
Subject: bridge: Reorder neighbor suppression check when flooding

The bridge does not flood ARP / NS packets for which a reply was sent to
bridge ports that have neighbor suppression enabled.

Subsequent patches are going to add per-{Port, VLAN} neighbor
suppression, which is going to make it more expensive to check whether
neighbor suppression is enabled since a VLAN lookup will be required.

Therefore, instead of unnecessarily performing this lookup for every
packet, only perform it for ARP / NS packets for which a reply was sent.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_forward.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 02bb620d3b8d..0fe133fa214c 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -224,8 +224,8 @@ void br_flood(struct net_bridge *br, struct sk_buff *skb,
 		/* Do not flood to ports that enable proxy ARP */
 		if (p->flags & BR_PROXYARP)
 			continue;
-		if ((p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS)) &&
-		    BR_INPUT_SKB_CB(skb)->proxyarp_replied)
+		if (BR_INPUT_SKB_CB(skb)->proxyarp_replied &&
+		    (p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS)))
 			continue;
 
 		prev = maybe_deliver(prev, p, skb, local_orig);
-- 
cgit v1.2.3-70-g09d2


From e408336a693e0fd74ba1b959627594b80f1761c3 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:53 +0300
Subject: bridge: Pass VLAN ID to br_flood()

Subsequent patches are going to add per-{Port, VLAN} neighbor
suppression, which will require br_flood() to potentially suppress ARP /
NS packets on a per-{Port, VLAN} basis.

As a preparation, pass the VLAN ID of the packet as another argument to
br_flood().

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_device.c  | 8 ++++----
 net/bridge/br_forward.c | 3 ++-
 net/bridge/br_input.c   | 2 +-
 net/bridge/br_private.h | 3 ++-
 4 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index df47c876230e..8eca8a5c80c6 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -80,10 +80,10 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	dest = eth_hdr(skb)->h_dest;
 	if (is_broadcast_ether_addr(dest)) {
-		br_flood(br, skb, BR_PKT_BROADCAST, false, true);
+		br_flood(br, skb, BR_PKT_BROADCAST, false, true, vid);
 	} else if (is_multicast_ether_addr(dest)) {
 		if (unlikely(netpoll_tx_running(dev))) {
-			br_flood(br, skb, BR_PKT_MULTICAST, false, true);
+			br_flood(br, skb, BR_PKT_MULTICAST, false, true, vid);
 			goto out;
 		}
 		if (br_multicast_rcv(&brmctx, &pmctx_null, vlan, skb, vid)) {
@@ -96,11 +96,11 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 		    br_multicast_querier_exists(brmctx, eth_hdr(skb), mdst))
 			br_multicast_flood(mdst, skb, brmctx, false, true);
 		else
-			br_flood(br, skb, BR_PKT_MULTICAST, false, true);
+			br_flood(br, skb, BR_PKT_MULTICAST, false, true, vid);
 	} else if ((dst = br_fdb_find_rcu(br, dest, vid)) != NULL) {
 		br_forward(dst->dst, skb, false, true);
 	} else {
-		br_flood(br, skb, BR_PKT_UNICAST, false, true);
+		br_flood(br, skb, BR_PKT_UNICAST, false, true, vid);
 	}
 out:
 	rcu_read_unlock();
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 0fe133fa214c..94a8d757ae4e 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -197,7 +197,8 @@ out:
 
 /* called under rcu_read_lock */
 void br_flood(struct net_bridge *br, struct sk_buff *skb,
-	      enum br_pkt_type pkt_type, bool local_rcv, bool local_orig)
+	      enum br_pkt_type pkt_type, bool local_rcv, bool local_orig,
+	      u16 vid)
 {
 	struct net_bridge_port *prev = NULL;
 	struct net_bridge_port *p;
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 3027e8f6be15..fc17b9fd93e6 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -207,7 +207,7 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb
 		br_forward(dst->dst, skb, local_rcv, false);
 	} else {
 		if (!mcast_hit)
-			br_flood(br, skb, pkt_type, local_rcv, false);
+			br_flood(br, skb, pkt_type, local_rcv, false, vid);
 		else
 			br_multicast_flood(mdst, skb, brmctx, local_rcv, false);
 	}
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 7264fd40f82f..1ff4d64ab584 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -849,7 +849,8 @@ void br_forward(const struct net_bridge_port *to, struct sk_buff *skb,
 		bool local_rcv, bool local_orig);
 int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb);
 void br_flood(struct net_bridge *br, struct sk_buff *skb,
-	      enum br_pkt_type pkt_type, bool local_rcv, bool local_orig);
+	      enum br_pkt_type pkt_type, bool local_rcv, bool local_orig,
+	      u16 vid);
 
 /* return true if both source port and dest port are isolated */
 static inline bool br_skb_isolated(const struct net_bridge_port *to,
-- 
cgit v1.2.3-70-g09d2


From a714e3ec230892039b5d5ae6902b58bb084a15c1 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:54 +0300
Subject: bridge: Add internal flags for per-{Port, VLAN} neighbor suppression

Add two internal flags that will be used to enable / disable per-{Port,
VLAN} neighbor suppression:

1. 'BR_NEIGH_VLAN_SUPPRESS': A per-port flag used to indicate that
per-{Port, VLAN} neighbor suppression is enabled on the bridge port.
When set, 'BR_NEIGH_SUPPRESS' has no effect.

2. 'BR_VLFLAG_NEIGH_SUPPRESS_ENABLED': A per-VLAN flag used to indicate
that neighbor suppression is enabled on the given VLAN.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_bridge.h | 1 +
 net/bridge/br_private.h   | 1 +
 2 files changed, 2 insertions(+)

diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h
index 1668ac4d7adc..3ff96ae31bf6 100644
--- a/include/linux/if_bridge.h
+++ b/include/linux/if_bridge.h
@@ -60,6 +60,7 @@ struct br_ip_list {
 #define BR_TX_FWD_OFFLOAD	BIT(20)
 #define BR_PORT_LOCKED		BIT(21)
 #define BR_PORT_MAB		BIT(22)
+#define BR_NEIGH_VLAN_SUPPRESS	BIT(23)
 
 #define BR_DEFAULT_AGEING_TIME	(300 * HZ)
 
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 1ff4d64ab584..b17fc821ecc8 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -178,6 +178,7 @@ enum {
 	BR_VLFLAG_ADDED_BY_SWITCHDEV = BIT(1),
 	BR_VLFLAG_MCAST_ENABLED = BIT(2),
 	BR_VLFLAG_GLOBAL_MCAST_ENABLED = BIT(3),
+	BR_VLFLAG_NEIGH_SUPPRESS_ENABLED = BIT(4),
 };
 
 /**
-- 
cgit v1.2.3-70-g09d2


From 6be42ed0a5f4b26e77b51cd10a09b41d53f97570 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:55 +0300
Subject: bridge: Take per-{Port, VLAN} neighbor suppression into account

The bridge driver gates the neighbor suppression code behind an internal
per-bridge flag called 'BROPT_NEIGH_SUPPRESS_ENABLED'. The flag is set
when at least one bridge port has neighbor suppression enabled.

As a preparation for per-{Port, VLAN} neighbor suppression, make sure
the global flag is also set if per-{Port, VLAN} neighbor suppression is
enabled. That is, when the 'BR_NEIGH_VLAN_SUPPRESS' flag is set on at
least one bridge port.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_arp_nd_proxy.c | 2 +-
 net/bridge/br_if.c           | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c
index b45c00c01dea..016a25a9e444 100644
--- a/net/bridge/br_arp_nd_proxy.c
+++ b/net/bridge/br_arp_nd_proxy.c
@@ -30,7 +30,7 @@ void br_recalculate_neigh_suppress_enabled(struct net_bridge *br)
 	bool neigh_suppress = false;
 
 	list_for_each_entry(p, &br->port_list, list) {
-		if (p->flags & BR_NEIGH_SUPPRESS) {
+		if (p->flags & (BR_NEIGH_SUPPRESS | BR_NEIGH_VLAN_SUPPRESS)) {
 			neigh_suppress = true;
 			break;
 		}
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 24f01ff113f0..3f04b40f6056 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -759,7 +759,7 @@ void br_port_flags_change(struct net_bridge_port *p, unsigned long mask)
 	if (mask & BR_AUTO_MASK)
 		nbp_update_port_count(br);
 
-	if (mask & BR_NEIGH_SUPPRESS)
+	if (mask & (BR_NEIGH_SUPPRESS | BR_NEIGH_VLAN_SUPPRESS))
 		br_recalculate_neigh_suppress_enabled(br);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 3aca683e0654a2f4ff6996341cba98a5396ff050 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:56 +0300
Subject: bridge: Encapsulate data path neighbor suppression logic

Currently, there are various places in the bridge data path that check
whether neighbor suppression is enabled on a given bridge port.

As a preparation for per-{Port, VLAN} neighbor suppression, encapsulate
this logic in a function and pass the VLAN ID of the packet as an
argument.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_arp_nd_proxy.c | 15 ++++++++++-----
 net/bridge/br_forward.c      |  3 ++-
 net/bridge/br_private.h      |  1 +
 3 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c
index 016a25a9e444..16c3a1c5d0ae 100644
--- a/net/bridge/br_arp_nd_proxy.c
+++ b/net/bridge/br_arp_nd_proxy.c
@@ -158,7 +158,7 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br,
 		return;
 
 	if (br_opt_get(br, BROPT_NEIGH_SUPPRESS_ENABLED)) {
-		if (p && (p->flags & BR_NEIGH_SUPPRESS))
+		if (br_is_neigh_suppress_enabled(p, vid))
 			return;
 		if (parp->ar_op != htons(ARPOP_RREQUEST) &&
 		    parp->ar_op != htons(ARPOP_RREPLY) &&
@@ -202,8 +202,8 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br,
 			bool replied = false;
 
 			if ((p && (p->flags & BR_PROXYARP)) ||
-			    (f->dst && (f->dst->flags & (BR_PROXYARP_WIFI |
-							 BR_NEIGH_SUPPRESS)))) {
+			    (f->dst && (f->dst->flags & BR_PROXYARP_WIFI)) ||
+			    br_is_neigh_suppress_enabled(f->dst, vid)) {
 				if (!vid)
 					br_arp_send(br, p, skb->dev, sip, tip,
 						    sha, n->ha, sha, 0, 0);
@@ -407,7 +407,7 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 
 	BR_INPUT_SKB_CB(skb)->proxyarp_replied = 0;
 
-	if (p && (p->flags & BR_NEIGH_SUPPRESS))
+	if (br_is_neigh_suppress_enabled(p, vid))
 		return;
 
 	if (msg->icmph.icmp6_type == NDISC_NEIGHBOUR_ADVERTISEMENT &&
@@ -461,7 +461,7 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 		if (f) {
 			bool replied = false;
 
-			if (f->dst && (f->dst->flags & BR_NEIGH_SUPPRESS)) {
+			if (br_is_neigh_suppress_enabled(f->dst, vid)) {
 				if (vid != 0)
 					br_nd_send(br, p, skb, n,
 						   skb->vlan_proto,
@@ -483,3 +483,8 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 	}
 }
 #endif
+
+bool br_is_neigh_suppress_enabled(const struct net_bridge_port *p, u16 vid)
+{
+	return p && (p->flags & BR_NEIGH_SUPPRESS);
+}
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 94a8d757ae4e..57744704ff69 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -226,7 +226,8 @@ void br_flood(struct net_bridge *br, struct sk_buff *skb,
 		if (p->flags & BR_PROXYARP)
 			continue;
 		if (BR_INPUT_SKB_CB(skb)->proxyarp_replied &&
-		    (p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS)))
+		    ((p->flags & BR_PROXYARP_WIFI) ||
+		     br_is_neigh_suppress_enabled(p, vid)))
 			continue;
 
 		prev = maybe_deliver(prev, p, skb, local_orig);
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index b17fc821ecc8..2119729ded2b 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -2220,4 +2220,5 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br,
 void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 		       u16 vid, struct net_bridge_port *p, struct nd_msg *msg);
 struct nd_msg *br_is_nd_neigh_msg(struct sk_buff *skb, struct nd_msg *m);
+bool br_is_neigh_suppress_enabled(const struct net_bridge_port *p, u16 vid);
 #endif
-- 
cgit v1.2.3-70-g09d2


From 412614b1457af278b00152f3804af5481c6c55ac Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:57 +0300
Subject: bridge: Add per-{Port, VLAN} neighbor suppression data path support

When the bridge is not VLAN-aware (i.e., VLAN ID is 0), determine if
neighbor suppression is enabled on a given bridge port solely based on
the existing 'BR_NEIGH_SUPPRESS' flag.

Otherwise, if the bridge is VLAN-aware, first check if per-{Port, VLAN}
neighbor suppression is enabled on the given bridge port using the
'BR_NEIGH_VLAN_SUPPRESS' flag. If so, look up the VLAN and check whether
it has neighbor suppression enabled based on the per-VLAN
'BR_VLFLAG_NEIGH_SUPPRESS_ENABLED' flag.

If the bridge is VLAN-aware, but the bridge port does not have
per-{Port, VLAN} neighbor suppression enabled, then fallback to
determine neighbor suppression based on the 'BR_NEIGH_SUPPRESS' flag.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_arp_nd_proxy.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c
index 16c3a1c5d0ae..c7869a286df4 100644
--- a/net/bridge/br_arp_nd_proxy.c
+++ b/net/bridge/br_arp_nd_proxy.c
@@ -486,5 +486,21 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 
 bool br_is_neigh_suppress_enabled(const struct net_bridge_port *p, u16 vid)
 {
-	return p && (p->flags & BR_NEIGH_SUPPRESS);
+	if (!p)
+		return false;
+
+	if (!vid)
+		return !!(p->flags & BR_NEIGH_SUPPRESS);
+
+	if (p->flags & BR_NEIGH_VLAN_SUPPRESS) {
+		struct net_bridge_vlan_group *vg = nbp_vlan_group_rcu(p);
+		struct net_bridge_vlan *v;
+
+		v = br_vlan_find(vg, vid);
+		if (!v)
+			return false;
+		return !!(v->priv_flags & BR_VLFLAG_NEIGH_SUPPRESS_ENABLED);
+	} else {
+		return !!(p->flags & BR_NEIGH_SUPPRESS);
+	}
 }
-- 
cgit v1.2.3-70-g09d2


From 83f6d600796c65ab34b08dbddb5795099dfda4d1 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:58 +0300
Subject: bridge: vlan: Allow setting VLAN neighbor suppression state

Add a new VLAN attribute that allows user space to set the neighbor
suppression state of the port VLAN. Example:

 # bridge -d -j -p vlan show dev swp1 vid 10 | jq '.[]["vlans"][]["neigh_suppress"]'
 false
 # bridge vlan set vid 10 dev swp1 neigh_suppress on
 # bridge -d -j -p vlan show dev swp1 vid 10 | jq '.[]["vlans"][]["neigh_suppress"]'
 true
 # bridge vlan set vid 10 dev swp1 neigh_suppress off
 # bridge -d -j -p vlan show dev swp1 vid 10 | jq '.[]["vlans"][]["neigh_suppress"]'
 false

 # bridge vlan set vid 10 dev br0 neigh_suppress on
 Error: bridge: Can't set neigh_suppress for non-port vlans.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_bridge.h |  1 +
 net/bridge/br_vlan.c           |  1 +
 net/bridge/br_vlan_options.c   | 20 +++++++++++++++++++-
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index c9d624f528c5..f95326fce6bb 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -525,6 +525,7 @@ enum {
 	BRIDGE_VLANDB_ENTRY_MCAST_ROUTER,
 	BRIDGE_VLANDB_ENTRY_MCAST_N_GROUPS,
 	BRIDGE_VLANDB_ENTRY_MCAST_MAX_GROUPS,
+	BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS,
 	__BRIDGE_VLANDB_ENTRY_MAX,
 };
 #define BRIDGE_VLANDB_ENTRY_MAX (__BRIDGE_VLANDB_ENTRY_MAX - 1)
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 8a3dbc09ba38..15f44d026e75 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -2134,6 +2134,7 @@ static const struct nla_policy br_vlan_db_policy[BRIDGE_VLANDB_ENTRY_MAX + 1] =
 	[BRIDGE_VLANDB_ENTRY_MCAST_ROUTER]	= { .type = NLA_U8 },
 	[BRIDGE_VLANDB_ENTRY_MCAST_N_GROUPS]	= { .type = NLA_REJECT },
 	[BRIDGE_VLANDB_ENTRY_MCAST_MAX_GROUPS]	= { .type = NLA_U32 },
+	[BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS]	= NLA_POLICY_MAX(NLA_U8, 1),
 };
 
 static int br_vlan_rtm_process_one(struct net_device *dev,
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
index e378c2f3a9e2..8fa89b04ee94 100644
--- a/net/bridge/br_vlan_options.c
+++ b/net/bridge/br_vlan_options.c
@@ -52,7 +52,9 @@ bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v,
 		       const struct net_bridge_port *p)
 {
 	if (nla_put_u8(skb, BRIDGE_VLANDB_ENTRY_STATE, br_vlan_get_state(v)) ||
-	    !__vlan_tun_put(skb, v))
+	    !__vlan_tun_put(skb, v) ||
+	    nla_put_u8(skb, BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS,
+		       !!(v->priv_flags & BR_VLFLAG_NEIGH_SUPPRESS_ENABLED)))
 		return false;
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
@@ -80,6 +82,7 @@ size_t br_vlan_opts_nl_size(void)
 	       + nla_total_size(sizeof(u32)) /* BRIDGE_VLANDB_ENTRY_MCAST_N_GROUPS */
 	       + nla_total_size(sizeof(u32)) /* BRIDGE_VLANDB_ENTRY_MCAST_MAX_GROUPS */
 #endif
+	       + nla_total_size(sizeof(u8)) /* BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS */
 	       + 0;
 }
 
@@ -239,6 +242,21 @@ static int br_vlan_process_one_opts(const struct net_bridge *br,
 	}
 #endif
 
+	if (tb[BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS]) {
+		bool enabled = v->priv_flags & BR_VLFLAG_NEIGH_SUPPRESS_ENABLED;
+		bool val = nla_get_u8(tb[BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS]);
+
+		if (!p) {
+			NL_SET_ERR_MSG_MOD(extack, "Can't set neigh_suppress for non-port vlans");
+			return -EINVAL;
+		}
+
+		if (val != enabled) {
+			v->priv_flags ^= BR_VLFLAG_NEIGH_SUPPRESS_ENABLED;
+			*changed = true;
+		}
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 160656d7201d861a1f2a0bf279a765e8cda2317a Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:34:59 +0300
Subject: bridge: Allow setting per-{Port, VLAN} neighbor suppression state

Add a new bridge port attribute that allows user space to enable
per-{Port, VLAN} neighbor suppression. Example:

 # bridge -d -j -p link show dev swp1 | jq '.[]["neigh_vlan_suppress"]'
 false
 # bridge link set dev swp1 neigh_vlan_suppress on
 # bridge -d -j -p link show dev swp1 | jq '.[]["neigh_vlan_suppress"]'
 true
 # bridge link set dev swp1 neigh_vlan_suppress off
 # bridge -d -j -p link show dev swp1 | jq '.[]["neigh_vlan_suppress"]'
 false

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_link.h | 1 +
 net/bridge/br_netlink.c      | 8 +++++++-
 net/core/rtnetlink.c         | 2 +-
 3 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 8d679688efe0..4ac1000b0ef2 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -569,6 +569,7 @@ enum {
 	IFLA_BRPORT_MAB,
 	IFLA_BRPORT_MCAST_N_GROUPS,
 	IFLA_BRPORT_MCAST_MAX_GROUPS,
+	IFLA_BRPORT_NEIGH_VLAN_SUPPRESS,
 	__IFLA_BRPORT_MAX
 };
 #define IFLA_BRPORT_MAX (__IFLA_BRPORT_MAX - 1)
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index fefb1c0e248b..05c5863d2e20 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -189,6 +189,7 @@ static inline size_t br_port_info_size(void)
 		+ nla_total_size(1)	/* IFLA_BRPORT_ISOLATED */
 		+ nla_total_size(1)	/* IFLA_BRPORT_LOCKED */
 		+ nla_total_size(1)	/* IFLA_BRPORT_MAB */
+		+ nla_total_size(1)	/* IFLA_BRPORT_NEIGH_VLAN_SUPPRESS */
 		+ nla_total_size(sizeof(struct ifla_bridge_id))	/* IFLA_BRPORT_ROOT_ID */
 		+ nla_total_size(sizeof(struct ifla_bridge_id))	/* IFLA_BRPORT_BRIDGE_ID */
 		+ nla_total_size(sizeof(u16))	/* IFLA_BRPORT_DESIGNATED_PORT */
@@ -278,7 +279,9 @@ static int br_port_fill_attrs(struct sk_buff *skb,
 		       !!(p->flags & BR_MRP_LOST_IN_CONT)) ||
 	    nla_put_u8(skb, IFLA_BRPORT_ISOLATED, !!(p->flags & BR_ISOLATED)) ||
 	    nla_put_u8(skb, IFLA_BRPORT_LOCKED, !!(p->flags & BR_PORT_LOCKED)) ||
-	    nla_put_u8(skb, IFLA_BRPORT_MAB, !!(p->flags & BR_PORT_MAB)))
+	    nla_put_u8(skb, IFLA_BRPORT_MAB, !!(p->flags & BR_PORT_MAB)) ||
+	    nla_put_u8(skb, IFLA_BRPORT_NEIGH_VLAN_SUPPRESS,
+		       !!(p->flags & BR_NEIGH_VLAN_SUPPRESS)))
 		return -EMSGSIZE;
 
 	timerval = br_timer_value(&p->message_age_timer);
@@ -891,6 +894,7 @@ static const struct nla_policy br_port_policy[IFLA_BRPORT_MAX + 1] = {
 	[IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT] = { .type = NLA_U32 },
 	[IFLA_BRPORT_MCAST_N_GROUPS] = { .type = NLA_REJECT },
 	[IFLA_BRPORT_MCAST_MAX_GROUPS] = { .type = NLA_U32 },
+	[IFLA_BRPORT_NEIGH_VLAN_SUPPRESS] = NLA_POLICY_MAX(NLA_U8, 1),
 };
 
 /* Change the state of the port and notify spanning tree */
@@ -957,6 +961,8 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[],
 	br_set_port_flag(p, tb, IFLA_BRPORT_ISOLATED, BR_ISOLATED);
 	br_set_port_flag(p, tb, IFLA_BRPORT_LOCKED, BR_PORT_LOCKED);
 	br_set_port_flag(p, tb, IFLA_BRPORT_MAB, BR_PORT_MAB);
+	br_set_port_flag(p, tb, IFLA_BRPORT_NEIGH_VLAN_SUPPRESS,
+			 BR_NEIGH_VLAN_SUPPRESS);
 
 	if ((p->flags & BR_PORT_MAB) &&
 	    (!(p->flags & BR_PORT_LOCKED) || !(p->flags & BR_LEARNING))) {
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index e844d75220fb..653901a1bf75 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -61,7 +61,7 @@
 #include "dev.h"
 
 #define RTNL_MAX_TYPE		50
-#define RTNL_SLAVE_MAX_TYPE	42
+#define RTNL_SLAVE_MAX_TYPE	43
 
 struct rtnl_link {
 	rtnl_doit_func		doit;
-- 
cgit v1.2.3-70-g09d2


From 7648ac72dcd7e22ac1fa5e573e536592773831dc Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Wed, 19 Apr 2023 18:35:00 +0300
Subject: selftests: net: Add bridge neighbor suppression test

Add test cases for bridge neighbor suppression, testing both per-port
and per-{Port, VLAN} neighbor suppression with both ARP and NS packets.

Example truncated output:

 # ./test_bridge_neigh_suppress.sh
 [...]
 Tests passed: 148
 Tests failed:   0

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/Makefile               |   1 +
 .../selftests/net/test_bridge_neigh_suppress.sh    | 862 +++++++++++++++++++++
 2 files changed, 863 insertions(+)
 create mode 100755 tools/testing/selftests/net/test_bridge_neigh_suppress.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 1de34ec99290..c12df57d5539 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -83,6 +83,7 @@ TEST_GEN_FILES += nat6to4.o
 TEST_GEN_FILES += ip_local_port_range
 TEST_GEN_FILES += bind_wildcard
 TEST_PROGS += test_vxlan_mdb.sh
+TEST_PROGS += test_bridge_neigh_suppress.sh
 
 TEST_FILES := settings
 
diff --git a/tools/testing/selftests/net/test_bridge_neigh_suppress.sh b/tools/testing/selftests/net/test_bridge_neigh_suppress.sh
new file mode 100755
index 000000000000..d80f2cd87614
--- /dev/null
+++ b/tools/testing/selftests/net/test_bridge_neigh_suppress.sh
@@ -0,0 +1,862 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# This test is for checking bridge neighbor suppression functionality. The
+# topology consists of two bridges (VTEPs) connected using VXLAN. A single
+# host is connected to each bridge over multiple VLANs. The test checks that
+# ARP/NS messages from the first host are suppressed on the VXLAN port when
+# should.
+#
+# +-----------------------+              +------------------------+
+# | h1                    |              | h2                     |
+# |                       |              |                        |
+# | + eth0.10             |              | + eth0.10              |
+# | | 192.0.2.1/28        |              | | 192.0.2.2/28         |
+# | | 2001:db8:1::1/64    |              | | 2001:db8:1::2/64     |
+# | |                     |              | |                      |
+# | |  + eth0.20          |              | |  + eth0.20           |
+# | \  | 192.0.2.17/28    |              | \  | 192.0.2.18/28     |
+# |  \ | 2001:db8:2::1/64 |              |  \ | 2001:db8:2::2/64  |
+# |   \|                  |              |   \|                   |
+# |    + eth0             |              |    + eth0              |
+# +----|------------------+              +----|-------------------+
+#      |                                      |
+#      |                                      |
+# +----|-------------------------------+ +----|-------------------------------+
+# |    + swp1                   + vx0  | |    + swp1                   + vx0  |
+# |    |                        |      | |    |                        |      |
+# |    |           br0          |      | |    |                        |      |
+# |    +------------+-----------+      | |    +------------+-----------+      |
+# |                 |                  | |                 |                  |
+# |                 |                  | |                 |                  |
+# |             +---+---+              | |             +---+---+              |
+# |             |       |              | |             |       |              |
+# |             |       |              | |             |       |              |
+# |             +       +              | |             +       +              |
+# |          br0.10  br0.20            | |          br0.10  br0.20            |
+# |                                    | |                                    |
+# |                 192.0.2.33         | |                 192.0.2.34         |
+# |                 + lo               | |                 + lo               |
+# |                                    | |                                    |
+# |                                    | |                                    |
+# |                   192.0.2.49/28    | |    192.0.2.50/28                   |
+# |                           veth0 +-------+ veth0                           |
+# |                                    | |                                    |
+# | sw1                                | | sw2                                |
+# +------------------------------------+ +------------------------------------+
+
+ret=0
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+# All tests in this script. Can be overridden with -t option.
+TESTS="
+	neigh_suppress_arp
+	neigh_suppress_ns
+	neigh_vlan_suppress_arp
+	neigh_vlan_suppress_ns
+"
+VERBOSE=0
+PAUSE_ON_FAIL=no
+PAUSE=no
+
+################################################################################
+# Utilities
+
+log_test()
+{
+	local rc=$1
+	local expected=$2
+	local msg="$3"
+
+	if [ ${rc} -eq ${expected} ]; then
+		printf "TEST: %-60s  [ OK ]\n" "${msg}"
+		nsuccess=$((nsuccess+1))
+	else
+		ret=1
+		nfail=$((nfail+1))
+		printf "TEST: %-60s  [FAIL]\n" "${msg}"
+		if [ "$VERBOSE" = "1" ]; then
+			echo "    rc=$rc, expected $expected"
+		fi
+
+		if [ "${PAUSE_ON_FAIL}" = "yes" ]; then
+		echo
+			echo "hit enter to continue, 'q' to quit"
+			read a
+			[ "$a" = "q" ] && exit 1
+		fi
+	fi
+
+	if [ "${PAUSE}" = "yes" ]; then
+		echo
+		echo "hit enter to continue, 'q' to quit"
+		read a
+		[ "$a" = "q" ] && exit 1
+	fi
+
+	[ "$VERBOSE" = "1" ] && echo
+}
+
+run_cmd()
+{
+	local cmd="$1"
+	local out
+	local stderr="2>/dev/null"
+
+	if [ "$VERBOSE" = "1" ]; then
+		printf "COMMAND: $cmd\n"
+		stderr=
+	fi
+
+	out=$(eval $cmd $stderr)
+	rc=$?
+	if [ "$VERBOSE" = "1" -a -n "$out" ]; then
+		echo "    $out"
+	fi
+
+	return $rc
+}
+
+tc_check_packets()
+{
+	local ns=$1; shift
+	local id=$1; shift
+	local handle=$1; shift
+	local count=$1; shift
+	local pkts
+
+	sleep 0.1
+	pkts=$(tc -n $ns -j -s filter show $id \
+		| jq ".[] | select(.options.handle == $handle) | \
+		.options.actions[0].stats.packets")
+	[[ $pkts == $count ]]
+}
+
+################################################################################
+# Setup
+
+setup_topo_ns()
+{
+	local ns=$1; shift
+
+	ip netns add $ns
+	ip -n $ns link set dev lo up
+
+	ip netns exec $ns sysctl -qw net.ipv6.conf.all.keep_addr_on_down=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.default.ignore_routes_with_linkdown=1
+	ip netns exec $ns sysctl -qw net.ipv6.conf.all.accept_dad=0
+	ip netns exec $ns sysctl -qw net.ipv6.conf.default.accept_dad=0
+}
+
+setup_topo()
+{
+	local ns
+
+	for ns in h1 h2 sw1 sw2; do
+		setup_topo_ns $ns
+	done
+
+	ip link add name veth0 type veth peer name veth1
+	ip link set dev veth0 netns h1 name eth0
+	ip link set dev veth1 netns sw1 name swp1
+
+	ip link add name veth0 type veth peer name veth1
+	ip link set dev veth0 netns sw1 name veth0
+	ip link set dev veth1 netns sw2 name veth0
+
+	ip link add name veth0 type veth peer name veth1
+	ip link set dev veth0 netns h2 name eth0
+	ip link set dev veth1 netns sw2 name swp1
+}
+
+setup_host_common()
+{
+	local ns=$1; shift
+	local v4addr1=$1; shift
+	local v4addr2=$1; shift
+	local v6addr1=$1; shift
+	local v6addr2=$1; shift
+
+	ip -n $ns link set dev eth0 up
+	ip -n $ns link add link eth0 name eth0.10 up type vlan id 10
+	ip -n $ns link add link eth0 name eth0.20 up type vlan id 20
+
+	ip -n $ns address add $v4addr1 dev eth0.10
+	ip -n $ns address add $v4addr2 dev eth0.20
+	ip -n $ns address add $v6addr1 dev eth0.10
+	ip -n $ns address add $v6addr2 dev eth0.20
+}
+
+setup_h1()
+{
+	local ns=h1
+	local v4addr1=192.0.2.1/28
+	local v4addr2=192.0.2.17/28
+	local v6addr1=2001:db8:1::1/64
+	local v6addr2=2001:db8:2::1/64
+
+	setup_host_common $ns $v4addr1 $v4addr2 $v6addr1 $v6addr2
+}
+
+setup_h2()
+{
+	local ns=h2
+	local v4addr1=192.0.2.2/28
+	local v4addr2=192.0.2.18/28
+	local v6addr1=2001:db8:1::2/64
+	local v6addr2=2001:db8:2::2/64
+
+	setup_host_common $ns $v4addr1 $v4addr2 $v6addr1 $v6addr2
+}
+
+setup_sw_common()
+{
+	local ns=$1; shift
+	local local_addr=$1; shift
+	local remote_addr=$1; shift
+	local veth_addr=$1; shift
+	local gw_addr=$1; shift
+
+	ip -n $ns address add $local_addr/32 dev lo
+
+	ip -n $ns link set dev veth0 up
+	ip -n $ns address add $veth_addr/28 dev veth0
+	ip -n $ns route add default via $gw_addr
+
+	ip -n $ns link add name br0 up type bridge vlan_filtering 1 \
+		vlan_default_pvid 0 mcast_snooping 0
+
+	ip -n $ns link add link br0 name br0.10 up type vlan id 10
+	bridge -n $ns vlan add vid 10 dev br0 self
+
+	ip -n $ns link add link br0 name br0.20 up type vlan id 20
+	bridge -n $ns vlan add vid 20 dev br0 self
+
+	ip -n $ns link set dev swp1 up master br0
+	bridge -n $ns vlan add vid 10 dev swp1
+	bridge -n $ns vlan add vid 20 dev swp1
+
+	ip -n $ns link add name vx0 up master br0 type vxlan \
+		local $local_addr dstport 4789 nolearning external
+	bridge -n $ns fdb add 00:00:00:00:00:00 dev vx0 self static \
+		dst $remote_addr src_vni 10010
+	bridge -n $ns fdb add 00:00:00:00:00:00 dev vx0 self static \
+		dst $remote_addr src_vni 10020
+	bridge -n $ns link set dev vx0 vlan_tunnel on learning off
+
+	bridge -n $ns vlan add vid 10 dev vx0
+	bridge -n $ns vlan add vid 10 dev vx0 tunnel_info id 10010
+
+	bridge -n $ns vlan add vid 20 dev vx0
+	bridge -n $ns vlan add vid 20 dev vx0 tunnel_info id 10020
+}
+
+setup_sw1()
+{
+	local ns=sw1
+	local local_addr=192.0.2.33
+	local remote_addr=192.0.2.34
+	local veth_addr=192.0.2.49
+	local gw_addr=192.0.2.50
+
+	setup_sw_common $ns $local_addr $remote_addr $veth_addr $gw_addr
+}
+
+setup_sw2()
+{
+	local ns=sw2
+	local local_addr=192.0.2.34
+	local remote_addr=192.0.2.33
+	local veth_addr=192.0.2.50
+	local gw_addr=192.0.2.49
+
+	setup_sw_common $ns $local_addr $remote_addr $veth_addr $gw_addr
+}
+
+setup()
+{
+	set -e
+
+	setup_topo
+	setup_h1
+	setup_h2
+	setup_sw1
+	setup_sw2
+
+	sleep 5
+
+	set +e
+}
+
+cleanup()
+{
+	local ns
+
+	for ns in h1 h2 sw1 sw2; do
+		ip netns del $ns &> /dev/null
+	done
+}
+
+################################################################################
+# Tests
+
+neigh_suppress_arp_common()
+{
+	local vid=$1; shift
+	local sip=$1; shift
+	local tip=$1; shift
+	local h2_mac
+
+	echo
+	echo "Per-port ARP suppression - VLAN $vid"
+	echo "----------------------------------"
+
+	run_cmd "tc -n sw1 qdisc replace dev vx0 clsact"
+	run_cmd "tc -n sw1 filter replace dev vx0 egress pref 1 handle 101 proto 0x0806 flower indev swp1 arp_tip $tip arp_sip $sip arp_op request action pass"
+
+	# Initial state - check that ARP requests are not suppressed and that
+	# ARP replies are received.
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 0 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "ARP suppression"
+
+	# Enable neighbor suppression and check that nothing changes compared
+	# to the initial state.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 0 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 2
+	log_test $? 0 "ARP suppression"
+
+	# Install an FDB entry for the remote host and check that nothing
+	# changes compared to the initial state.
+	h2_mac=$(ip -n h2 -j -p link show eth0.$vid | jq -r '.[]["address"]')
+	run_cmd "bridge -n sw1 fdb replace $h2_mac dev vx0 master static vlan $vid"
+	log_test $? 0 "FDB entry installation"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 0 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 3
+	log_test $? 0 "ARP suppression"
+
+	# Install a neighbor on the matching SVI interface and check that ARP
+	# requests are suppressed.
+	run_cmd "ip -n sw1 neigh replace $tip lladdr $h2_mac nud permanent dev br0.$vid"
+	log_test $? 0 "Neighbor entry installation"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 0 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 3
+	log_test $? 0 "ARP suppression"
+
+	# Take the second host down and check that ARP requests are suppressed
+	# and that ARP replies are received.
+	run_cmd "ip -n h2 link set dev eth0.$vid down"
+	log_test $? 0 "H2 down"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 0 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 3
+	log_test $? 0 "ARP suppression"
+
+	run_cmd "ip -n h2 link set dev eth0.$vid up"
+	log_test $? 0 "H2 up"
+
+	# Disable neighbor suppression and check that ARP requests are no
+	# longer suppressed.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress off"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 0 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 4
+	log_test $? 0 "ARP suppression"
+
+	# Take the second host down and check that ARP requests are not
+	# suppressed and that ARP replies are not received.
+	run_cmd "ip -n h2 link set dev eth0.$vid down"
+	log_test $? 0 "H2 down"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip -I eth0.$vid $tip"
+	log_test $? 1 "arping"
+	tc_check_packets sw1 "dev vx0 egress" 101 5
+	log_test $? 0 "ARP suppression"
+}
+
+neigh_suppress_arp()
+{
+	local vid=10
+	local sip=192.0.2.1
+	local tip=192.0.2.2
+
+	neigh_suppress_arp_common $vid $sip $tip
+
+	vid=20
+	sip=192.0.2.17
+	tip=192.0.2.18
+	neigh_suppress_arp_common $vid $sip $tip
+}
+
+neigh_suppress_ns_common()
+{
+	local vid=$1; shift
+	local saddr=$1; shift
+	local daddr=$1; shift
+	local maddr=$1; shift
+	local h2_mac
+
+	echo
+	echo "Per-port NS suppression - VLAN $vid"
+	echo "---------------------------------"
+
+	run_cmd "tc -n sw1 qdisc replace dev vx0 clsact"
+	run_cmd "tc -n sw1 filter replace dev vx0 egress pref 1 handle 101 proto ipv6 flower indev swp1 ip_proto icmpv6 dst_ip $maddr src_ip $saddr type 135 code 0 action pass"
+
+	# Initial state - check that NS messages are not suppressed and that ND
+	# messages are received.
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 0 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "NS suppression"
+
+	# Enable neighbor suppression and check that nothing changes compared
+	# to the initial state.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 0 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 2
+	log_test $? 0 "NS suppression"
+
+	# Install an FDB entry for the remote host and check that nothing
+	# changes compared to the initial state.
+	h2_mac=$(ip -n h2 -j -p link show eth0.$vid | jq -r '.[]["address"]')
+	run_cmd "bridge -n sw1 fdb replace $h2_mac dev vx0 master static vlan $vid"
+	log_test $? 0 "FDB entry installation"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 0 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 3
+	log_test $? 0 "NS suppression"
+
+	# Install a neighbor on the matching SVI interface and check that NS
+	# messages are suppressed.
+	run_cmd "ip -n sw1 neigh replace $daddr lladdr $h2_mac nud permanent dev br0.$vid"
+	log_test $? 0 "Neighbor entry installation"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 0 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 3
+	log_test $? 0 "NS suppression"
+
+	# Take the second host down and check that NS messages are suppressed
+	# and that ND messages are received.
+	run_cmd "ip -n h2 link set dev eth0.$vid down"
+	log_test $? 0 "H2 down"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 0 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 3
+	log_test $? 0 "NS suppression"
+
+	run_cmd "ip -n h2 link set dev eth0.$vid up"
+	log_test $? 0 "H2 up"
+
+	# Disable neighbor suppression and check that NS messages are no longer
+	# suppressed.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress off"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 0 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 4
+	log_test $? 0 "NS suppression"
+
+	# Take the second host down and check that NS messages are not
+	# suppressed and that ND messages are not received.
+	run_cmd "ip -n h2 link set dev eth0.$vid down"
+	log_test $? 0 "H2 down"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr -w 5000 $daddr eth0.$vid"
+	log_test $? 2 "ndisc6"
+	tc_check_packets sw1 "dev vx0 egress" 101 5
+	log_test $? 0 "NS suppression"
+}
+
+neigh_suppress_ns()
+{
+	local vid=10
+	local saddr=2001:db8:1::1
+	local daddr=2001:db8:1::2
+	local maddr=ff02::1:ff00:2
+
+	neigh_suppress_ns_common $vid $saddr $daddr $maddr
+
+	vid=20
+	saddr=2001:db8:2::1
+	daddr=2001:db8:2::2
+	maddr=ff02::1:ff00:2
+
+	neigh_suppress_ns_common $vid $saddr $daddr $maddr
+}
+
+neigh_vlan_suppress_arp()
+{
+	local vid1=10
+	local vid2=20
+	local sip1=192.0.2.1
+	local sip2=192.0.2.17
+	local tip1=192.0.2.2
+	local tip2=192.0.2.18
+	local h2_mac1
+	local h2_mac2
+
+	echo
+	echo "Per-{Port, VLAN} ARP suppression"
+	echo "--------------------------------"
+
+	run_cmd "tc -n sw1 qdisc replace dev vx0 clsact"
+	run_cmd "tc -n sw1 filter replace dev vx0 egress pref 1 handle 101 proto 0x0806 flower indev swp1 arp_tip $tip1 arp_sip $sip1 arp_op request action pass"
+	run_cmd "tc -n sw1 filter replace dev vx0 egress pref 1 handle 102 proto 0x0806 flower indev swp1 arp_tip $tip2 arp_sip $sip2 arp_op request action pass"
+
+	h2_mac1=$(ip -n h2 -j -p link show eth0.$vid1 | jq -r '.[]["address"]')
+	h2_mac2=$(ip -n h2 -j -p link show eth0.$vid2 | jq -r '.[]["address"]')
+	run_cmd "bridge -n sw1 fdb replace $h2_mac1 dev vx0 master static vlan $vid1"
+	run_cmd "bridge -n sw1 fdb replace $h2_mac2 dev vx0 master static vlan $vid2"
+	run_cmd "ip -n sw1 neigh replace $tip1 lladdr $h2_mac1 nud permanent dev br0.$vid1"
+	run_cmd "ip -n sw1 neigh replace $tip2 lladdr $h2_mac2 nud permanent dev br0.$vid2"
+
+	# Enable per-{Port, VLAN} neighbor suppression and check that ARP
+	# requests are not suppressed and that ARP replies are received.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_vlan_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_vlan_suppress on\""
+	log_test $? 0 "\"neigh_vlan_suppress\" is on"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip1 -I eth0.$vid1 $tip1"
+	log_test $? 0 "arping (VLAN $vid1)"
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip2 -I eth0.$vid2 $tip2"
+	log_test $? 0 "arping (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "ARP suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 1
+	log_test $? 0 "ARP suppression (VLAN $vid2)"
+
+	# Enable neighbor suppression on VLAN 10 and check that only on this
+	# VLAN ARP requests are suppressed.
+	run_cmd "bridge -n sw1 vlan set vid $vid1 dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d vlan show dev vx0 vid $vid1 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on (VLAN $vid1)"
+	run_cmd "bridge -n sw1 -d vlan show dev vx0 vid $vid2 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off (VLAN $vid2)"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip1 -I eth0.$vid1 $tip1"
+	log_test $? 0 "arping (VLAN $vid1)"
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip2 -I eth0.$vid2 $tip2"
+	log_test $? 0 "arping (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "ARP suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 2
+	log_test $? 0 "ARP suppression (VLAN $vid2)"
+
+	# Enable neighbor suppression on the port and check that it has no
+	# effect compared to previous state.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip1 -I eth0.$vid1 $tip1"
+	log_test $? 0 "arping (VLAN $vid1)"
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip2 -I eth0.$vid2 $tip2"
+	log_test $? 0 "arping (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "ARP suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 3
+	log_test $? 0 "ARP suppression (VLAN $vid2)"
+
+	# Disable neighbor suppression on the port and check that it has no
+	# effect compared to previous state.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress off"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip1 -I eth0.$vid1 $tip1"
+	log_test $? 0 "arping (VLAN $vid1)"
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip2 -I eth0.$vid2 $tip2"
+	log_test $? 0 "arping (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "ARP suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 4
+	log_test $? 0 "ARP suppression (VLAN $vid2)"
+
+	# Disable neighbor suppression on VLAN 10 and check that ARP requests
+	# are no longer suppressed on this VLAN.
+	run_cmd "bridge -n sw1 vlan set vid $vid1 dev vx0 neigh_suppress off"
+	run_cmd "bridge -n sw1 -d vlan show dev vx0 vid $vid1 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off (VLAN $vid1)"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip1 -I eth0.$vid1 $tip1"
+	log_test $? 0 "arping (VLAN $vid1)"
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip2 -I eth0.$vid2 $tip2"
+	log_test $? 0 "arping (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 2
+	log_test $? 0 "ARP suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 5
+	log_test $? 0 "ARP suppression (VLAN $vid2)"
+
+	# Disable per-{Port, VLAN} neighbor suppression, enable neighbor
+	# suppression on the port and check that on both VLANs ARP requests are
+	# suppressed.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_vlan_suppress off"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_vlan_suppress off\""
+	log_test $? 0 "\"neigh_vlan_suppress\" is off"
+
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on"
+
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip1 -I eth0.$vid1 $tip1"
+	log_test $? 0 "arping (VLAN $vid1)"
+	run_cmd "ip netns exec h1 arping -q -b -c 1 -w 5 -s $sip2 -I eth0.$vid2 $tip2"
+	log_test $? 0 "arping (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 2
+	log_test $? 0 "ARP suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 5
+	log_test $? 0 "ARP suppression (VLAN $vid2)"
+}
+
+neigh_vlan_suppress_ns()
+{
+	local vid1=10
+	local vid2=20
+	local saddr1=2001:db8:1::1
+	local saddr2=2001:db8:2::1
+	local daddr1=2001:db8:1::2
+	local daddr2=2001:db8:2::2
+	local maddr=ff02::1:ff00:2
+	local h2_mac1
+	local h2_mac2
+
+	echo
+	echo "Per-{Port, VLAN} NS suppression"
+	echo "-------------------------------"
+
+	run_cmd "tc -n sw1 qdisc replace dev vx0 clsact"
+	run_cmd "tc -n sw1 filter replace dev vx0 egress pref 1 handle 101 proto ipv6 flower indev swp1 ip_proto icmpv6 dst_ip $maddr src_ip $saddr1 type 135 code 0 action pass"
+	run_cmd "tc -n sw1 filter replace dev vx0 egress pref 1 handle 102 proto ipv6 flower indev swp1 ip_proto icmpv6 dst_ip $maddr src_ip $saddr2 type 135 code 0 action pass"
+
+	h2_mac1=$(ip -n h2 -j -p link show eth0.$vid1 | jq -r '.[]["address"]')
+	h2_mac2=$(ip -n h2 -j -p link show eth0.$vid2 | jq -r '.[]["address"]')
+	run_cmd "bridge -n sw1 fdb replace $h2_mac1 dev vx0 master static vlan $vid1"
+	run_cmd "bridge -n sw1 fdb replace $h2_mac2 dev vx0 master static vlan $vid2"
+	run_cmd "ip -n sw1 neigh replace $daddr1 lladdr $h2_mac1 nud permanent dev br0.$vid1"
+	run_cmd "ip -n sw1 neigh replace $daddr2 lladdr $h2_mac2 nud permanent dev br0.$vid2"
+
+	# Enable per-{Port, VLAN} neighbor suppression and check that NS
+	# messages are not suppressed and that ND messages are received.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_vlan_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_vlan_suppress on\""
+	log_test $? 0 "\"neigh_vlan_suppress\" is on"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr1 -w 5000 $daddr1 eth0.$vid1"
+	log_test $? 0 "ndisc6 (VLAN $vid1)"
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr2 -w 5000 $daddr2 eth0.$vid2"
+	log_test $? 0 "ndisc6 (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "NS suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 1
+	log_test $? 0 "NS suppression (VLAN $vid2)"
+
+	# Enable neighbor suppression on VLAN 10 and check that only on this
+	# VLAN NS messages are suppressed.
+	run_cmd "bridge -n sw1 vlan set vid $vid1 dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d vlan show dev vx0 vid $vid1 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on (VLAN $vid1)"
+	run_cmd "bridge -n sw1 -d vlan show dev vx0 vid $vid2 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off (VLAN $vid2)"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr1 -w 5000 $daddr1 eth0.$vid1"
+	log_test $? 0 "ndisc6 (VLAN $vid1)"
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr2 -w 5000 $daddr2 eth0.$vid2"
+	log_test $? 0 "ndisc6 (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "NS suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 2
+	log_test $? 0 "NS suppression (VLAN $vid2)"
+
+	# Enable neighbor suppression on the port and check that it has no
+	# effect compared to previous state.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr1 -w 5000 $daddr1 eth0.$vid1"
+	log_test $? 0 "ndisc6 (VLAN $vid1)"
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr2 -w 5000 $daddr2 eth0.$vid2"
+	log_test $? 0 "ndisc6 (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "NS suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 3
+	log_test $? 0 "NS suppression (VLAN $vid2)"
+
+	# Disable neighbor suppression on the port and check that it has no
+	# effect compared to previous state.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress off"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr1 -w 5000 $daddr1 eth0.$vid1"
+	log_test $? 0 "ndisc6 (VLAN $vid1)"
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr2 -w 5000 $daddr2 eth0.$vid2"
+	log_test $? 0 "ndisc6 (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 1
+	log_test $? 0 "NS suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 4
+	log_test $? 0 "NS suppression (VLAN $vid2)"
+
+	# Disable neighbor suppression on VLAN 10 and check that NS messages
+	# are no longer suppressed on this VLAN.
+	run_cmd "bridge -n sw1 vlan set vid $vid1 dev vx0 neigh_suppress off"
+	run_cmd "bridge -n sw1 -d vlan show dev vx0 vid $vid1 | grep \"neigh_suppress off\""
+	log_test $? 0 "\"neigh_suppress\" is off (VLAN $vid1)"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr1 -w 5000 $daddr1 eth0.$vid1"
+	log_test $? 0 "ndisc6 (VLAN $vid1)"
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr2 -w 5000 $daddr2 eth0.$vid2"
+	log_test $? 0 "ndisc6 (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 2
+	log_test $? 0 "NS suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 5
+	log_test $? 0 "NS suppression (VLAN $vid2)"
+
+	# Disable per-{Port, VLAN} neighbor suppression, enable neighbor
+	# suppression on the port and check that on both VLANs NS messages are
+	# suppressed.
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_vlan_suppress off"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_vlan_suppress off\""
+	log_test $? 0 "\"neigh_vlan_suppress\" is off"
+
+	run_cmd "bridge -n sw1 link set dev vx0 neigh_suppress on"
+	run_cmd "bridge -n sw1 -d link show dev vx0 | grep \"neigh_suppress on\""
+	log_test $? 0 "\"neigh_suppress\" is on"
+
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr1 -w 5000 $daddr1 eth0.$vid1"
+	log_test $? 0 "ndisc6 (VLAN $vid1)"
+	run_cmd "ip netns exec h1 ndisc6 -q -r 1 -s $saddr2 -w 5000 $daddr2 eth0.$vid2"
+	log_test $? 0 "ndisc6 (VLAN $vid2)"
+
+	tc_check_packets sw1 "dev vx0 egress" 101 2
+	log_test $? 0 "NS suppression (VLAN $vid1)"
+	tc_check_packets sw1 "dev vx0 egress" 102 5
+	log_test $? 0 "NS suppression (VLAN $vid2)"
+}
+
+################################################################################
+# Usage
+
+usage()
+{
+	cat <<EOF
+usage: ${0##*/} OPTS
+
+        -t <test>   Test(s) to run (default: all)
+                    (options: $TESTS)
+        -p          Pause on fail
+        -P          Pause after each test before cleanup
+        -v          Verbose mode (show commands and output)
+EOF
+}
+
+################################################################################
+# Main
+
+trap cleanup EXIT
+
+while getopts ":t:pPvh" opt; do
+	case $opt in
+		t) TESTS=$OPTARG;;
+		p) PAUSE_ON_FAIL=yes;;
+		P) PAUSE=yes;;
+		v) VERBOSE=$(($VERBOSE + 1));;
+		h) usage; exit 0;;
+		*) usage; exit 1;;
+	esac
+done
+
+# Make sure we don't pause twice.
+[ "${PAUSE}" = "yes" ] && PAUSE_ON_FAIL=no
+
+if [ "$(id -u)" -ne 0 ];then
+	echo "SKIP: Need root privileges"
+	exit $ksft_skip;
+fi
+
+if [ ! -x "$(command -v ip)" ]; then
+	echo "SKIP: Could not run test without ip tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v bridge)" ]; then
+	echo "SKIP: Could not run test without bridge tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v tc)" ]; then
+	echo "SKIP: Could not run test without tc tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v arping)" ]; then
+	echo "SKIP: Could not run test without arping tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v ndisc6)" ]; then
+	echo "SKIP: Could not run test without ndisc6 tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v jq)" ]; then
+	echo "SKIP: Could not run test without jq tool"
+	exit $ksft_skip
+fi
+
+bridge link help 2>&1 | grep -q "neigh_vlan_suppress"
+if [ $? -ne 0 ]; then
+   echo "SKIP: iproute2 bridge too old, missing per-VLAN neighbor suppression support"
+   exit $ksft_skip
+fi
+
+# Start clean.
+cleanup
+
+for t in $TESTS
+do
+	setup; $t; cleanup;
+done
+
+if [ "$TESTS" != "none" ]; then
+	printf "\nTests passed: %3d\n" ${nsuccess}
+	printf "Tests failed: %3d\n"   ${nfail}
+fi
+
+exit $ret
-- 
cgit v1.2.3-70-g09d2


From 55435ea7729accb5b8a330de751836c4be524834 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:14 -0700
Subject: pds_core: initial framework for pds_core PF driver

This is the initial PCI driver framework for the new pds_core device
driver and its family of devices.  This does the very basics of
registering for the new PF PCI device 1dd8:100c, setting up debugfs
entries, and registering with devlink.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../device_drivers/ethernet/amd/pds_core.rst       |  35 ++
 .../networking/device_drivers/ethernet/index.rst   |   1 +
 drivers/net/ethernet/amd/pds_core/Makefile         |   8 +
 drivers/net/ethernet/amd/pds_core/core.h           |  56 ++
 drivers/net/ethernet/amd/pds_core/debugfs.c        |  31 ++
 drivers/net/ethernet/amd/pds_core/main.c           | 277 ++++++++++
 include/linux/pds/pds_common.h                     |  14 +
 include/linux/pds/pds_core_if.h                    | 571 +++++++++++++++++++++
 8 files changed, 993 insertions(+)
 create mode 100644 Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
 create mode 100644 drivers/net/ethernet/amd/pds_core/Makefile
 create mode 100644 drivers/net/ethernet/amd/pds_core/core.h
 create mode 100644 drivers/net/ethernet/amd/pds_core/debugfs.c
 create mode 100644 drivers/net/ethernet/amd/pds_core/main.c
 create mode 100644 include/linux/pds/pds_common.h
 create mode 100644 include/linux/pds/pds_core_if.h

diff --git a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
new file mode 100644
index 000000000000..99a70026f1bc
--- /dev/null
+++ b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
@@ -0,0 +1,35 @@
+.. SPDX-License-Identifier: GPL-2.0+
+
+========================================================
+Linux Driver for the AMD/Pensando(R) DSC adapter family
+========================================================
+
+Copyright(c) 2023 Advanced Micro Devices, Inc
+
+Identifying the Adapter
+=======================
+
+To find if one or more AMD/Pensando PCI Core devices are installed on the
+host, check for the PCI devices::
+
+  # lspci -d 1dd8:100c
+  b5:00.0 Processing accelerators: Pensando Systems Device 100c
+  b6:00.0 Processing accelerators: Pensando Systems Device 100c
+
+If such devices are listed as above, then the pds_core.ko driver should find
+and configure them for use.  There should be log entries in the kernel
+messages such as these::
+
+  $ dmesg | grep pds_core
+  pds_core 0000:b5:00.0: 252.048 Gb/s available PCIe bandwidth (16.0 GT/s PCIe x16 link)
+  pds_core 0000:b5:00.0: FW: 1.60.0-73
+  pds_core 0000:b6:00.0: 252.048 Gb/s available PCIe bandwidth (16.0 GT/s PCIe x16 link)
+  pds_core 0000:b6:00.0: FW: 1.60.0-73
+
+Support
+=======
+
+For general Linux networking support, please use the netdev mailing
+list, which is monitored by AMD/Pensando personnel::
+
+  netdev@vger.kernel.org
diff --git a/Documentation/networking/device_drivers/ethernet/index.rst b/Documentation/networking/device_drivers/ethernet/index.rst
index 6e9e7012d000..417ca514a4d0 100644
--- a/Documentation/networking/device_drivers/ethernet/index.rst
+++ b/Documentation/networking/device_drivers/ethernet/index.rst
@@ -14,6 +14,7 @@ Contents:
    3com/vortex
    amazon/ena
    altera/altera_tse
+   amd/pds_core
    aquantia/atlantic
    chelsio/cxgb
    cirrus/cs89x0
diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
new file mode 100644
index 000000000000..de3bf1d1886c
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Advanced Micro Devices, Inc.
+
+obj-$(CONFIG_PDS_CORE) := pds_core.o
+
+pds_core-y := main.o
+
+pds_core-$(CONFIG_DEBUG_FS) += debugfs.o
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
new file mode 100644
index 000000000000..34ef837e8cfe
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#ifndef _PDSC_H_
+#define _PDSC_H_
+
+#include <linux/debugfs.h>
+#include <net/devlink.h>
+
+#include <linux/pds/pds_common.h>
+#include <linux/pds/pds_core_if.h>
+
+#define PDSC_DRV_DESCRIPTION	"AMD/Pensando Core Driver"
+
+struct pdsc_dev_bar {
+	void __iomem *vaddr;
+	phys_addr_t bus_addr;
+	unsigned long len;
+	int res_index;
+};
+
+/* No state flags set means we are in a steady running state */
+enum pdsc_state_flags {
+	PDSC_S_FW_DEAD,		    /* stopped, wait on startup or recovery */
+	PDSC_S_INITING_DRIVER,	    /* initial startup from probe */
+	PDSC_S_STOPPING_DRIVER,	    /* driver remove */
+
+	/* leave this as last */
+	PDSC_S_STATE_SIZE
+};
+
+struct pdsc {
+	struct pci_dev *pdev;
+	struct dentry *dentry;
+	struct device *dev;
+	struct pdsc_dev_bar bars[PDS_CORE_BARS_MAX];
+	int hw_index;
+	int uid;
+
+	unsigned long state;
+
+	struct pds_core_dev_info_regs __iomem *info_regs;
+	struct pds_core_dev_cmd_regs __iomem *cmd_regs;
+	struct pds_core_intr __iomem *intr_ctrl;
+	u64 __iomem *intr_status;
+	u64 __iomem *db_pages;
+	dma_addr_t phy_db_pages;
+	u64 __iomem *kern_dbpage;
+};
+
+void pdsc_debugfs_create(void);
+void pdsc_debugfs_destroy(void);
+void pdsc_debugfs_add_dev(struct pdsc *pdsc);
+void pdsc_debugfs_del_dev(struct pdsc *pdsc);
+
+#endif /* _PDSC_H_ */
diff --git a/drivers/net/ethernet/amd/pds_core/debugfs.c b/drivers/net/ethernet/amd/pds_core/debugfs.c
new file mode 100644
index 000000000000..b2f7cb795c20
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/debugfs.c
@@ -0,0 +1,31 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/pci.h>
+
+#include "core.h"
+
+static struct dentry *pdsc_dir;
+
+void pdsc_debugfs_create(void)
+{
+	pdsc_dir = debugfs_create_dir(PDS_CORE_DRV_NAME, NULL);
+}
+
+void pdsc_debugfs_destroy(void)
+{
+	debugfs_remove_recursive(pdsc_dir);
+}
+
+void pdsc_debugfs_add_dev(struct pdsc *pdsc)
+{
+	pdsc->dentry = debugfs_create_dir(pci_name(pdsc->pdev), pdsc_dir);
+
+	debugfs_create_ulong("state", 0400, pdsc->dentry, &pdsc->state);
+}
+
+void pdsc_debugfs_del_dev(struct pdsc *pdsc)
+{
+	debugfs_remove_recursive(pdsc->dentry);
+	pdsc->dentry = NULL;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
new file mode 100644
index 000000000000..c2b12f226959
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -0,0 +1,277 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/pci.h>
+
+#include <linux/pds/pds_common.h>
+
+#include "core.h"
+
+MODULE_DESCRIPTION(PDSC_DRV_DESCRIPTION);
+MODULE_AUTHOR("Advanced Micro Devices, Inc");
+MODULE_LICENSE("GPL");
+
+/* Supported devices */
+static const struct pci_device_id pdsc_id_table[] = {
+	{ PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_CORE_PF) },
+	{ 0, }	/* end of table */
+};
+MODULE_DEVICE_TABLE(pci, pdsc_id_table);
+
+static void pdsc_unmap_bars(struct pdsc *pdsc)
+{
+	struct pdsc_dev_bar *bars = pdsc->bars;
+	unsigned int i;
+
+	for (i = 0; i < PDS_CORE_BARS_MAX; i++) {
+		if (bars[i].vaddr)
+			pci_iounmap(pdsc->pdev, bars[i].vaddr);
+	}
+}
+
+static int pdsc_map_bars(struct pdsc *pdsc)
+{
+	struct pdsc_dev_bar *bar = pdsc->bars;
+	struct pci_dev *pdev = pdsc->pdev;
+	struct device *dev = pdsc->dev;
+	struct pdsc_dev_bar *bars;
+	unsigned int i, j;
+	int num_bars = 0;
+	int err;
+	u32 sig;
+
+	bars = pdsc->bars;
+
+	/* Since the PCI interface in the hardware is configurable,
+	 * we need to poke into all the bars to find the set we're
+	 * expecting.
+	 */
+	for (i = 0, j = 0; i < PDS_CORE_BARS_MAX; i++) {
+		if (!(pci_resource_flags(pdev, i) & IORESOURCE_MEM))
+			continue;
+
+		bars[j].len = pci_resource_len(pdev, i);
+		bars[j].bus_addr = pci_resource_start(pdev, i);
+		bars[j].res_index = i;
+
+		/* only map the whole bar 0 */
+		if (j > 0) {
+			bars[j].vaddr = NULL;
+		} else {
+			bars[j].vaddr = pci_iomap(pdev, i, bars[j].len);
+			if (!bars[j].vaddr) {
+				dev_err(dev, "Cannot map BAR %d, aborting\n", i);
+				return -ENODEV;
+			}
+		}
+
+		j++;
+	}
+	num_bars = j;
+
+	/* BAR0: dev_cmd and interrupts */
+	if (num_bars < 1) {
+		dev_err(dev, "No bars found\n");
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	if (bar->len < PDS_CORE_BAR0_SIZE) {
+		dev_err(dev, "Resource bar size %lu too small\n", bar->len);
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	pdsc->info_regs = bar->vaddr + PDS_CORE_BAR0_DEV_INFO_REGS_OFFSET;
+	pdsc->cmd_regs = bar->vaddr + PDS_CORE_BAR0_DEV_CMD_REGS_OFFSET;
+	pdsc->intr_status = bar->vaddr + PDS_CORE_BAR0_INTR_STATUS_OFFSET;
+	pdsc->intr_ctrl = bar->vaddr + PDS_CORE_BAR0_INTR_CTRL_OFFSET;
+
+	sig = ioread32(&pdsc->info_regs->signature);
+	if (sig != PDS_CORE_DEV_INFO_SIGNATURE) {
+		dev_err(dev, "Incompatible firmware signature %x", sig);
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	/* BAR1: doorbells */
+	bar++;
+	if (num_bars < 2) {
+		dev_err(dev, "Doorbell bar missing\n");
+		err = -EFAULT;
+		goto err_out;
+	}
+
+	pdsc->db_pages = bar->vaddr;
+	pdsc->phy_db_pages = bar->bus_addr;
+
+	return 0;
+
+err_out:
+	pdsc_unmap_bars(pdsc);
+	return err;
+}
+
+static int pdsc_init_vf(struct pdsc *vf)
+{
+	return -1;
+}
+
+static int pdsc_init_pf(struct pdsc *pdsc)
+{
+	struct devlink *dl;
+	int err;
+
+	pcie_print_link_status(pdsc->pdev);
+
+	err = pci_request_regions(pdsc->pdev, PDS_CORE_DRV_NAME);
+	if (err) {
+		dev_err(pdsc->dev, "Cannot request PCI regions: %pe\n",
+			ERR_PTR(err));
+		return err;
+	}
+
+	err = pdsc_map_bars(pdsc);
+	if (err)
+		goto err_out_release_regions;
+
+	dl = priv_to_devlink(pdsc);
+	devl_lock(dl);
+	devl_register(dl);
+	devl_unlock(dl);
+
+	return 0;
+
+err_out_release_regions:
+	pci_release_regions(pdsc->pdev);
+
+	return err;
+}
+
+static const struct devlink_ops pdsc_dl_ops = {
+};
+
+static const struct devlink_ops pdsc_dl_vf_ops = {
+};
+
+static DEFINE_IDA(pdsc_ida);
+
+static int pdsc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct device *dev = &pdev->dev;
+	const struct devlink_ops *ops;
+	struct devlink *dl;
+	struct pdsc *pdsc;
+	bool is_pf;
+	int err;
+
+	is_pf = !pdev->is_virtfn;
+	ops = is_pf ? &pdsc_dl_ops : &pdsc_dl_vf_ops;
+	dl = devlink_alloc(ops, sizeof(struct pdsc), dev);
+	if (!dl)
+		return -ENOMEM;
+	pdsc = devlink_priv(dl);
+
+	pdsc->pdev = pdev;
+	pdsc->dev = &pdev->dev;
+	set_bit(PDSC_S_INITING_DRIVER, &pdsc->state);
+	pci_set_drvdata(pdev, pdsc);
+	pdsc_debugfs_add_dev(pdsc);
+
+	err = ida_alloc(&pdsc_ida, GFP_KERNEL);
+	if (err < 0) {
+		dev_err(pdsc->dev, "%s: id alloc failed: %pe\n",
+			__func__, ERR_PTR(err));
+		goto err_out_free_devlink;
+	}
+	pdsc->uid = err;
+
+	/* Query system for DMA addressing limitation for the device. */
+	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(PDS_CORE_ADDR_LEN));
+	if (err) {
+		dev_err(dev, "Unable to obtain 64-bit DMA for consistent allocations, aborting: %pe\n",
+			ERR_PTR(err));
+		goto err_out_free_ida;
+	}
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		dev_err(dev, "Cannot enable PCI device: %pe\n", ERR_PTR(err));
+		goto err_out_free_ida;
+	}
+	pci_set_master(pdev);
+
+	if (is_pf)
+		err = pdsc_init_pf(pdsc);
+	else
+		err = pdsc_init_vf(pdsc);
+	if (err) {
+		dev_err(dev, "Cannot init device: %pe\n", ERR_PTR(err));
+		goto err_out_clear_master;
+	}
+
+	clear_bit(PDSC_S_INITING_DRIVER, &pdsc->state);
+	return 0;
+
+err_out_clear_master:
+	pci_clear_master(pdev);
+	pci_disable_device(pdev);
+err_out_free_ida:
+	ida_free(&pdsc_ida, pdsc->uid);
+err_out_free_devlink:
+	pdsc_debugfs_del_dev(pdsc);
+	devlink_free(dl);
+
+	return err;
+}
+
+static void pdsc_remove(struct pci_dev *pdev)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+	struct devlink *dl;
+
+	/* Unhook the registrations first to be sure there
+	 * are no requests while we're stopping.
+	 */
+	dl = priv_to_devlink(pdsc);
+	devl_lock(dl);
+	devl_unregister(dl);
+	devl_unlock(dl);
+
+	pdsc_unmap_bars(pdsc);
+	pci_release_regions(pdev);
+
+	pci_clear_master(pdev);
+	pci_disable_device(pdev);
+
+	ida_free(&pdsc_ida, pdsc->uid);
+	pdsc_debugfs_del_dev(pdsc);
+	devlink_free(dl);
+}
+
+static struct pci_driver pdsc_driver = {
+	.name = PDS_CORE_DRV_NAME,
+	.id_table = pdsc_id_table,
+	.probe = pdsc_probe,
+	.remove = pdsc_remove,
+};
+
+static int __init pdsc_init_module(void)
+{
+	if (strcmp(KBUILD_MODNAME, PDS_CORE_DRV_NAME))
+		return -EINVAL;
+
+	pdsc_debugfs_create();
+	return pci_register_driver(&pdsc_driver);
+}
+
+static void __exit pdsc_cleanup_module(void)
+{
+	pci_unregister_driver(&pdsc_driver);
+	pdsc_debugfs_destroy();
+}
+
+module_init(pdsc_init_module);
+module_exit(pdsc_cleanup_module);
diff --git a/include/linux/pds/pds_common.h b/include/linux/pds/pds_common.h
new file mode 100644
index 000000000000..bd041a5170a6
--- /dev/null
+++ b/include/linux/pds/pds_common.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR Linux-OpenIB) OR BSD-2-Clause */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc. */
+
+#ifndef _PDS_COMMON_H_
+#define _PDS_COMMON_H_
+
+#define PDS_CORE_DRV_NAME			"pds_core"
+
+/* the device's internal addressing uses up to 52 bits */
+#define PDS_CORE_ADDR_LEN	52
+#define PDS_CORE_ADDR_MASK	(BIT_ULL(PDS_ADDR_LEN) - 1)
+#define PDS_PAGE_SIZE		4096
+
+#endif /* _PDS_COMMON_H_ */
diff --git a/include/linux/pds/pds_core_if.h b/include/linux/pds/pds_core_if.h
new file mode 100644
index 000000000000..e838a2b90440
--- /dev/null
+++ b/include/linux/pds/pds_core_if.h
@@ -0,0 +1,571 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR Linux-OpenIB) OR BSD-2-Clause */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc. */
+
+#ifndef _PDS_CORE_IF_H_
+#define _PDS_CORE_IF_H_
+
+#define PCI_VENDOR_ID_PENSANDO			0x1dd8
+#define PCI_DEVICE_ID_PENSANDO_CORE_PF		0x100c
+#define PCI_DEVICE_ID_VIRTIO_NET_TRANS		0x1000
+#define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_VF	0x1003
+#define PCI_DEVICE_ID_PENSANDO_VDPA_VF		0x100b
+#define PDS_CORE_BARS_MAX			4
+#define PDS_CORE_PCI_BAR_DBELL			1
+
+/* Bar0 */
+#define PDS_CORE_DEV_INFO_SIGNATURE		0x44455649 /* 'DEVI' */
+#define PDS_CORE_BAR0_SIZE			0x8000
+#define PDS_CORE_BAR0_DEV_INFO_REGS_OFFSET	0x0000
+#define PDS_CORE_BAR0_DEV_CMD_REGS_OFFSET	0x0800
+#define PDS_CORE_BAR0_DEV_CMD_DATA_REGS_OFFSET	0x0c00
+#define PDS_CORE_BAR0_INTR_STATUS_OFFSET	0x1000
+#define PDS_CORE_BAR0_INTR_CTRL_OFFSET		0x2000
+#define PDS_CORE_DEV_CMD_DONE			0x00000001
+
+#define PDS_CORE_DEVCMD_TIMEOUT			5
+
+#define PDS_CORE_CLIENT_ID			0
+#define PDS_CORE_ASIC_TYPE_CAPRI		0
+
+/*
+ * enum pds_core_cmd_opcode - Device commands
+ */
+enum pds_core_cmd_opcode {
+	/* Core init */
+	PDS_CORE_CMD_NOP		= 0,
+	PDS_CORE_CMD_IDENTIFY		= 1,
+	PDS_CORE_CMD_RESET		= 2,
+	PDS_CORE_CMD_INIT		= 3,
+
+	PDS_CORE_CMD_FW_DOWNLOAD	= 4,
+	PDS_CORE_CMD_FW_CONTROL		= 5,
+
+	/* SR/IOV commands */
+	PDS_CORE_CMD_VF_GETATTR		= 60,
+	PDS_CORE_CMD_VF_SETATTR		= 61,
+	PDS_CORE_CMD_VF_CTRL		= 62,
+
+	/* Add commands before this line */
+	PDS_CORE_CMD_MAX,
+	PDS_CORE_CMD_COUNT
+};
+
+/*
+ * enum pds_core_status_code - Device command return codes
+ */
+enum pds_core_status_code {
+	PDS_RC_SUCCESS	= 0,	/* Success */
+	PDS_RC_EVERSION	= 1,	/* Incorrect version for request */
+	PDS_RC_EOPCODE	= 2,	/* Invalid cmd opcode */
+	PDS_RC_EIO	= 3,	/* I/O error */
+	PDS_RC_EPERM	= 4,	/* Permission denied */
+	PDS_RC_EQID	= 5,	/* Bad qid */
+	PDS_RC_EQTYPE	= 6,	/* Bad qtype */
+	PDS_RC_ENOENT	= 7,	/* No such element */
+	PDS_RC_EINTR	= 8,	/* operation interrupted */
+	PDS_RC_EAGAIN	= 9,	/* Try again */
+	PDS_RC_ENOMEM	= 10,	/* Out of memory */
+	PDS_RC_EFAULT	= 11,	/* Bad address */
+	PDS_RC_EBUSY	= 12,	/* Device or resource busy */
+	PDS_RC_EEXIST	= 13,	/* object already exists */
+	PDS_RC_EINVAL	= 14,	/* Invalid argument */
+	PDS_RC_ENOSPC	= 15,	/* No space left or alloc failure */
+	PDS_RC_ERANGE	= 16,	/* Parameter out of range */
+	PDS_RC_BAD_ADDR	= 17,	/* Descriptor contains a bad ptr */
+	PDS_RC_DEV_CMD	= 18,	/* Device cmd attempted on AdminQ */
+	PDS_RC_ENOSUPP	= 19,	/* Operation not supported */
+	PDS_RC_ERROR	= 29,	/* Generic error */
+	PDS_RC_ERDMA	= 30,	/* Generic RDMA error */
+	PDS_RC_EVFID	= 31,	/* VF ID does not exist */
+	PDS_RC_BAD_FW	= 32,	/* FW file is invalid or corrupted */
+	PDS_RC_ECLIENT	= 33,   /* No such client id */
+};
+
+/**
+ * struct pds_core_drv_identity - Driver identity information
+ * @drv_type:         Driver type (enum pds_core_driver_type)
+ * @os_dist:          OS distribution, numeric format
+ * @os_dist_str:      OS distribution, string format
+ * @kernel_ver:       Kernel version, numeric format
+ * @kernel_ver_str:   Kernel version, string format
+ * @driver_ver_str:   Driver version, string format
+ */
+struct pds_core_drv_identity {
+	__le32 drv_type;
+	__le32 os_dist;
+	char   os_dist_str[128];
+	__le32 kernel_ver;
+	char   kernel_ver_str[32];
+	char   driver_ver_str[32];
+};
+
+#define PDS_DEV_TYPE_MAX	16
+/**
+ * struct pds_core_dev_identity - Device identity information
+ * @version:	      Version of device identify
+ * @type:	      Identify type (0 for now)
+ * @state:	      Device state
+ * @rsvd:	      Word boundary padding
+ * @nlifs:	      Number of LIFs provisioned
+ * @nintrs:	      Number of interrupts provisioned
+ * @ndbpgs_per_lif:   Number of doorbell pages per LIF
+ * @intr_coal_mult:   Interrupt coalescing multiplication factor
+ *		      Scale user-supplied interrupt coalescing
+ *		      value in usecs to device units using:
+ *		      device units = usecs * mult / div
+ * @intr_coal_div:    Interrupt coalescing division factor
+ *		      Scale user-supplied interrupt coalescing
+ *		      value in usecs to device units using:
+ *		      device units = usecs * mult / div
+ * @vif_types:        How many of each VIF device type is supported
+ */
+struct pds_core_dev_identity {
+	u8     version;
+	u8     type;
+	u8     state;
+	u8     rsvd;
+	__le32 nlifs;
+	__le32 nintrs;
+	__le32 ndbpgs_per_lif;
+	__le32 intr_coal_mult;
+	__le32 intr_coal_div;
+	__le16 vif_types[PDS_DEV_TYPE_MAX];
+};
+
+#define PDS_CORE_IDENTITY_VERSION_1	1
+
+/**
+ * struct pds_core_dev_identify_cmd - Driver/device identify command
+ * @opcode:	Opcode PDS_CORE_CMD_IDENTIFY
+ * @ver:	Highest version of identify supported by driver
+ *
+ * Expects to find driver identification info (struct pds_core_drv_identity)
+ * in cmd_regs->data.  Driver should keep the devcmd interface locked
+ * while preparing the driver info.
+ */
+struct pds_core_dev_identify_cmd {
+	u8 opcode;
+	u8 ver;
+};
+
+/**
+ * struct pds_core_dev_identify_comp - Device identify command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @ver:	Version of identify returned by device
+ *
+ * Device identification info (struct pds_core_dev_identity) can be found
+ * in cmd_regs->data.  Driver should keep the devcmd interface locked
+ * while reading the results.
+ */
+struct pds_core_dev_identify_comp {
+	u8 status;
+	u8 ver;
+};
+
+/**
+ * struct pds_core_dev_reset_cmd - Device reset command
+ * @opcode:	Opcode PDS_CORE_CMD_RESET
+ *
+ * Resets and clears all LIFs, VDevs, and VIFs on the device.
+ */
+struct pds_core_dev_reset_cmd {
+	u8 opcode;
+};
+
+/**
+ * struct pds_core_dev_reset_comp - Reset command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ */
+struct pds_core_dev_reset_comp {
+	u8 status;
+};
+
+/*
+ * struct pds_core_dev_init_data - Pointers and info needed for the Core
+ * initialization PDS_CORE_CMD_INIT command.  The in and out structs are
+ * overlays on the pds_core_dev_cmd_regs.data space for passing data down
+ * to the firmware on init, and then returning initialization results.
+ */
+struct pds_core_dev_init_data_in {
+	__le64 adminq_q_base;
+	__le64 adminq_cq_base;
+	__le64 notifyq_cq_base;
+	__le32 flags;
+	__le16 intr_index;
+	u8     adminq_ring_size;
+	u8     notifyq_ring_size;
+};
+
+struct pds_core_dev_init_data_out {
+	__le32 core_hw_index;
+	__le32 adminq_hw_index;
+	__le32 notifyq_hw_index;
+	u8     adminq_hw_type;
+	u8     notifyq_hw_type;
+};
+
+/**
+ * struct pds_core_dev_init_cmd - Core device initialize
+ * @opcode:          opcode PDS_CORE_CMD_INIT
+ *
+ * Initializes the core device and sets up the AdminQ and NotifyQ.
+ * Expects to find initialization data (struct pds_core_dev_init_data_in)
+ * in cmd_regs->data.  Driver should keep the devcmd interface locked
+ * while preparing the driver info.
+ */
+struct pds_core_dev_init_cmd {
+	u8     opcode;
+};
+
+/**
+ * struct pds_core_dev_init_comp - Core init completion
+ * @status:     Status of the command (enum pds_core_status_code)
+ *
+ * Initialization result data (struct pds_core_dev_init_data_in)
+ * is found in cmd_regs->data.
+ */
+struct pds_core_dev_init_comp {
+	u8     status;
+};
+
+/**
+ * struct pds_core_fw_download_cmd - Firmware download command
+ * @opcode:     opcode
+ * @rsvd:	Word boundary padding
+ * @addr:       DMA address of the firmware buffer
+ * @offset:     offset of the firmware buffer within the full image
+ * @length:     number of valid bytes in the firmware buffer
+ */
+struct pds_core_fw_download_cmd {
+	u8     opcode;
+	u8     rsvd[3];
+	__le32 offset;
+	__le64 addr;
+	__le32 length;
+};
+
+/**
+ * struct pds_core_fw_download_comp - Firmware download completion
+ * @status:     Status of the command (enum pds_core_status_code)
+ */
+struct pds_core_fw_download_comp {
+	u8     status;
+};
+
+/**
+ * enum pds_core_fw_control_oper - FW control operations
+ * @PDS_CORE_FW_INSTALL_ASYNC:     Install firmware asynchronously
+ * @PDS_CORE_FW_INSTALL_STATUS:    Firmware installation status
+ * @PDS_CORE_FW_ACTIVATE_ASYNC:    Activate firmware asynchronously
+ * @PDS_CORE_FW_ACTIVATE_STATUS:   Firmware activate status
+ * @PDS_CORE_FW_UPDATE_CLEANUP:    Cleanup any firmware update leftovers
+ * @PDS_CORE_FW_GET_BOOT:          Return current active firmware slot
+ * @PDS_CORE_FW_SET_BOOT:          Set active firmware slot for next boot
+ * @PDS_CORE_FW_GET_LIST:          Return list of installed firmware images
+ */
+enum pds_core_fw_control_oper {
+	PDS_CORE_FW_INSTALL_ASYNC          = 0,
+	PDS_CORE_FW_INSTALL_STATUS         = 1,
+	PDS_CORE_FW_ACTIVATE_ASYNC         = 2,
+	PDS_CORE_FW_ACTIVATE_STATUS        = 3,
+	PDS_CORE_FW_UPDATE_CLEANUP         = 4,
+	PDS_CORE_FW_GET_BOOT               = 5,
+	PDS_CORE_FW_SET_BOOT               = 6,
+	PDS_CORE_FW_GET_LIST               = 7,
+};
+
+enum pds_core_fw_slot {
+	PDS_CORE_FW_SLOT_INVALID    = 0,
+	PDS_CORE_FW_SLOT_A	    = 1,
+	PDS_CORE_FW_SLOT_B          = 2,
+	PDS_CORE_FW_SLOT_GOLD       = 3,
+};
+
+/**
+ * struct pds_core_fw_control_cmd - Firmware control command
+ * @opcode:    opcode
+ * @rsvd:      Word boundary padding
+ * @oper:      firmware control operation (enum pds_core_fw_control_oper)
+ * @slot:      slot to operate on (enum pds_core_fw_slot)
+ */
+struct pds_core_fw_control_cmd {
+	u8  opcode;
+	u8  rsvd[3];
+	u8  oper;
+	u8  slot;
+};
+
+/**
+ * struct pds_core_fw_control_comp - Firmware control copletion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @rsvd:	Word alignment space
+ * @slot:	Slot number (enum pds_core_fw_slot)
+ * @rsvd1:	Struct padding
+ * @color:	Color bit
+ */
+struct pds_core_fw_control_comp {
+	u8     status;
+	u8     rsvd[3];
+	u8     slot;
+	u8     rsvd1[10];
+	u8     color;
+};
+
+struct pds_core_fw_name_info {
+#define PDS_CORE_FWSLOT_BUFLEN		8
+#define PDS_CORE_FWVERS_BUFLEN		32
+	char   slotname[PDS_CORE_FWSLOT_BUFLEN];
+	char   fw_version[PDS_CORE_FWVERS_BUFLEN];
+};
+
+struct pds_core_fw_list_info {
+#define PDS_CORE_FWVERS_LIST_LEN	16
+	u8 num_fw_slots;
+	struct pds_core_fw_name_info fw_names[PDS_CORE_FWVERS_LIST_LEN];
+} __packed;
+
+enum pds_core_vf_attr {
+	PDS_CORE_VF_ATTR_SPOOFCHK	= 1,
+	PDS_CORE_VF_ATTR_TRUST		= 2,
+	PDS_CORE_VF_ATTR_MAC		= 3,
+	PDS_CORE_VF_ATTR_LINKSTATE	= 4,
+	PDS_CORE_VF_ATTR_VLAN		= 5,
+	PDS_CORE_VF_ATTR_RATE		= 6,
+	PDS_CORE_VF_ATTR_STATSADDR	= 7,
+};
+
+/**
+ * enum pds_core_vf_link_status - Virtual Function link status
+ * @PDS_CORE_VF_LINK_STATUS_AUTO:   Use link state of the uplink
+ * @PDS_CORE_VF_LINK_STATUS_UP:     Link always up
+ * @PDS_CORE_VF_LINK_STATUS_DOWN:   Link always down
+ */
+enum pds_core_vf_link_status {
+	PDS_CORE_VF_LINK_STATUS_AUTO = 0,
+	PDS_CORE_VF_LINK_STATUS_UP   = 1,
+	PDS_CORE_VF_LINK_STATUS_DOWN = 2,
+};
+
+/**
+ * struct pds_core_vf_setattr_cmd - Set VF attributes on the NIC
+ * @opcode:     Opcode
+ * @attr:       Attribute type (enum pds_core_vf_attr)
+ * @vf_index:   VF index
+ * @macaddr:	mac address
+ * @vlanid:	vlan ID
+ * @maxrate:	max Tx rate in Mbps
+ * @spoofchk:	enable address spoof checking
+ * @trust:	enable VF trust
+ * @linkstate:	set link up or down
+ * @stats:	stats addr struct
+ * @stats.pa:	set DMA address for VF stats
+ * @stats.len:	length of VF stats space
+ * @pad:	force union to specific size
+ */
+struct pds_core_vf_setattr_cmd {
+	u8     opcode;
+	u8     attr;
+	__le16 vf_index;
+	union {
+		u8     macaddr[6];
+		__le16 vlanid;
+		__le32 maxrate;
+		u8     spoofchk;
+		u8     trust;
+		u8     linkstate;
+		struct {
+			__le64 pa;
+			__le32 len;
+		} stats;
+		u8     pad[60];
+	} __packed;
+};
+
+struct pds_core_vf_setattr_comp {
+	u8     status;
+	u8     attr;
+	__le16 vf_index;
+	__le16 comp_index;
+	u8     rsvd[9];
+	u8     color;
+};
+
+/**
+ * struct pds_core_vf_getattr_cmd - Get VF attributes from the NIC
+ * @opcode:     Opcode
+ * @attr:       Attribute type (enum pds_core_vf_attr)
+ * @vf_index:   VF index
+ */
+struct pds_core_vf_getattr_cmd {
+	u8     opcode;
+	u8     attr;
+	__le16 vf_index;
+};
+
+struct pds_core_vf_getattr_comp {
+	u8     status;
+	u8     attr;
+	__le16 vf_index;
+	union {
+		u8     macaddr[6];
+		__le16 vlanid;
+		__le32 maxrate;
+		u8     spoofchk;
+		u8     trust;
+		u8     linkstate;
+		__le64 stats_pa;
+		u8     pad[11];
+	} __packed;
+	u8     color;
+};
+
+enum pds_core_vf_ctrl_opcode {
+	PDS_CORE_VF_CTRL_START_ALL	= 0,
+	PDS_CORE_VF_CTRL_START		= 1,
+};
+
+/**
+ * struct pds_core_vf_ctrl_cmd - VF control command
+ * @opcode:         Opcode for the command
+ * @ctrl_opcode:    VF control operation type
+ * @vf_index:       VF Index. It is unused if op START_ALL is used.
+ */
+
+struct pds_core_vf_ctrl_cmd {
+	u8	opcode;
+	u8	ctrl_opcode;
+	__le16	vf_index;
+};
+
+/**
+ * struct pds_core_vf_ctrl_comp - VF_CTRL command completion.
+ * @status:     Status of the command (enum pds_core_status_code)
+ */
+struct pds_core_vf_ctrl_comp {
+	u8	status;
+};
+
+/*
+ * union pds_core_dev_cmd - Overlay of core device command structures
+ */
+union pds_core_dev_cmd {
+	u8     opcode;
+	u32    words[16];
+
+	struct pds_core_dev_identify_cmd identify;
+	struct pds_core_dev_init_cmd     init;
+	struct pds_core_dev_reset_cmd    reset;
+	struct pds_core_fw_download_cmd  fw_download;
+	struct pds_core_fw_control_cmd   fw_control;
+
+	struct pds_core_vf_setattr_cmd   vf_setattr;
+	struct pds_core_vf_getattr_cmd   vf_getattr;
+	struct pds_core_vf_ctrl_cmd      vf_ctrl;
+};
+
+/*
+ * union pds_core_dev_comp - Overlay of core device completion structures
+ */
+union pds_core_dev_comp {
+	u8                                status;
+	u8                                bytes[16];
+
+	struct pds_core_dev_identify_comp identify;
+	struct pds_core_dev_reset_comp    reset;
+	struct pds_core_dev_init_comp     init;
+	struct pds_core_fw_download_comp  fw_download;
+	struct pds_core_fw_control_comp   fw_control;
+
+	struct pds_core_vf_setattr_comp   vf_setattr;
+	struct pds_core_vf_getattr_comp   vf_getattr;
+	struct pds_core_vf_ctrl_comp      vf_ctrl;
+};
+
+/**
+ * struct pds_core_dev_hwstamp_regs - Hardware current timestamp registers
+ * @tick_low:        Low 32 bits of hardware timestamp
+ * @tick_high:       High 32 bits of hardware timestamp
+ */
+struct pds_core_dev_hwstamp_regs {
+	u32    tick_low;
+	u32    tick_high;
+};
+
+/**
+ * struct pds_core_dev_info_regs - Device info register format (read-only)
+ * @signature:       Signature value of 0x44455649 ('DEVI')
+ * @version:         Current version of info
+ * @asic_type:       Asic type
+ * @asic_rev:        Asic revision
+ * @fw_status:       Firmware status
+ *			bit 0   - 1 = fw running
+ *			bit 4-7 - 4 bit generation number, changes on fw restart
+ * @fw_heartbeat:    Firmware heartbeat counter
+ * @serial_num:      Serial number
+ * @fw_version:      Firmware version
+ * @oprom_regs:      oprom_regs to store oprom debug enable/disable and bmp
+ * @rsvd_pad1024:    Struct padding
+ * @hwstamp:         Hardware current timestamp registers
+ * @rsvd_pad2048:    Struct padding
+ */
+struct pds_core_dev_info_regs {
+#define PDS_CORE_DEVINFO_FWVERS_BUFLEN 32
+#define PDS_CORE_DEVINFO_SERIAL_BUFLEN 32
+	u32    signature;
+	u8     version;
+	u8     asic_type;
+	u8     asic_rev;
+#define PDS_CORE_FW_STS_F_STOPPED	0x00
+#define PDS_CORE_FW_STS_F_RUNNING	0x01
+#define PDS_CORE_FW_STS_F_GENERATION	0xF0
+	u8     fw_status;
+	__le32 fw_heartbeat;
+	char   fw_version[PDS_CORE_DEVINFO_FWVERS_BUFLEN];
+	char   serial_num[PDS_CORE_DEVINFO_SERIAL_BUFLEN];
+	u8     oprom_regs[32];     /* reserved */
+	u8     rsvd_pad1024[916];
+	struct pds_core_dev_hwstamp_regs hwstamp;   /* on 1k boundary */
+	u8     rsvd_pad2048[1016];
+} __packed;
+
+/**
+ * struct pds_core_dev_cmd_regs - Device command register format (read-write)
+ * @doorbell:	Device Cmd Doorbell, write-only
+ *              Write a 1 to signal device to process cmd
+ * @done:	Command completed indicator, poll for completion
+ *              bit 0 == 1 when command is complete
+ * @cmd:	Opcode-specific command bytes
+ * @comp:	Opcode-specific response bytes
+ * @rsvd:	Struct padding
+ * @data:	Opcode-specific side-data
+ */
+struct pds_core_dev_cmd_regs {
+	u32                     doorbell;
+	u32                     done;
+	union pds_core_dev_cmd  cmd;
+	union pds_core_dev_comp comp;
+	u8                      rsvd[48];
+	u32                     data[478];
+} __packed;
+
+/**
+ * struct pds_core_dev_regs - Device register format for bar 0 page 0
+ * @info:            Device info registers
+ * @devcmd:          Device command registers
+ */
+struct pds_core_dev_regs {
+	struct pds_core_dev_info_regs info;
+	struct pds_core_dev_cmd_regs  devcmd;
+} __packed;
+
+#ifndef __CHECKER__
+static_assert(sizeof(struct pds_core_drv_identity) <= 1912);
+static_assert(sizeof(struct pds_core_dev_identity) <= 1912);
+static_assert(sizeof(union pds_core_dev_cmd) == 64);
+static_assert(sizeof(union pds_core_dev_comp) == 16);
+static_assert(sizeof(struct pds_core_dev_info_regs) == 2048);
+static_assert(sizeof(struct pds_core_dev_cmd_regs) == 2048);
+static_assert(sizeof(struct pds_core_dev_regs) == 4096);
+#endif /* __CHECKER__ */
+
+#endif /* _PDS_CORE_IF_H_ */
-- 
cgit v1.2.3-70-g09d2


From 523847df1b3718d6286dce0ed1c83742fe0ffa94 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:15 -0700
Subject: pds_core: add devcmd device interfaces

The devcmd interface is the basic connection to the device through the
PCI BAR for low level identification and command services.  This does
the early device initialization and finds the identity data, and adds
devcmd routines to be used by later driver bits.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/Makefile  |   4 +-
 drivers/net/ethernet/amd/pds_core/core.c    |  36 +++
 drivers/net/ethernet/amd/pds_core/core.h    |  50 ++++
 drivers/net/ethernet/amd/pds_core/debugfs.c |  38 +++
 drivers/net/ethernet/amd/pds_core/dev.c     | 348 ++++++++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/main.c    |  33 ++-
 include/linux/pds/pds_common.h              |  30 +++
 include/linux/pds/pds_intr.h                | 163 +++++++++++++
 8 files changed, 699 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/ethernet/amd/pds_core/core.c
 create mode 100644 drivers/net/ethernet/amd/pds_core/dev.c
 create mode 100644 include/linux/pds/pds_intr.h

diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
index de3bf1d1886c..95a6c31e92d2 100644
--- a/drivers/net/ethernet/amd/pds_core/Makefile
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -3,6 +3,8 @@
 
 obj-$(CONFIG_PDS_CORE) := pds_core.o
 
-pds_core-y := main.o
+pds_core-y := main.o \
+	      dev.o \
+	      core.o
 
 pds_core-$(CONFIG_DEBUG_FS) += debugfs.o
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
new file mode 100644
index 000000000000..80d2ecb045df
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include "core.h"
+
+int pdsc_setup(struct pdsc *pdsc, bool init)
+{
+	int err = 0;
+
+	if (init)
+		err = pdsc_dev_init(pdsc);
+	else
+		err = pdsc_dev_reinit(pdsc);
+	if (err)
+		return err;
+
+	clear_bit(PDSC_S_FW_DEAD, &pdsc->state);
+	return 0;
+}
+
+void pdsc_teardown(struct pdsc *pdsc, bool removing)
+{
+	pdsc_devcmd_reset(pdsc);
+
+	if (removing) {
+		kfree(pdsc->intr_info);
+		pdsc->intr_info = NULL;
+	}
+
+	if (pdsc->kern_dbpage) {
+		iounmap(pdsc->kern_dbpage);
+		pdsc->kern_dbpage = NULL;
+	}
+
+	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
+}
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 34ef837e8cfe..fcf6c6545c49 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -9,8 +9,13 @@
 
 #include <linux/pds/pds_common.h>
 #include <linux/pds/pds_core_if.h>
+#include <linux/pds/pds_intr.h>
 
 #define PDSC_DRV_DESCRIPTION	"AMD/Pensando Core Driver"
+#define PDSC_TEARDOWN_RECOVERY	false
+#define PDSC_TEARDOWN_REMOVING	true
+#define PDSC_SETUP_RECOVERY	false
+#define PDSC_SETUP_INIT		true
 
 struct pdsc_dev_bar {
 	void __iomem *vaddr;
@@ -19,6 +24,22 @@ struct pdsc_dev_bar {
 	int res_index;
 };
 
+struct pdsc_devinfo {
+	u8 asic_type;
+	u8 asic_rev;
+	char fw_version[PDS_CORE_DEVINFO_FWVERS_BUFLEN + 1];
+	char serial_num[PDS_CORE_DEVINFO_SERIAL_BUFLEN + 1];
+};
+
+#define PDSC_INTR_NAME_MAX_SZ		32
+
+struct pdsc_intr_info {
+	char name[PDSC_INTR_NAME_MAX_SZ];
+	unsigned int index;
+	unsigned int vector;
+	void *data;
+};
+
 /* No state flags set means we are in a steady running state */
 enum pdsc_state_flags {
 	PDSC_S_FW_DEAD,		    /* stopped, wait on startup or recovery */
@@ -38,7 +59,19 @@ struct pdsc {
 	int uid;
 
 	unsigned long state;
+	u8 fw_status;
+	u8 fw_generation;
+	unsigned long last_fw_time;
+	u32 last_hb;
+
+	struct pdsc_devinfo dev_info;
+	struct pds_core_dev_identity dev_ident;
+	unsigned int nintrs;
+	struct pdsc_intr_info *intr_info;	/* array of nintrs elements */
 
+	unsigned int devcmd_timeout;
+	struct mutex devcmd_lock;	/* lock for dev_cmd operations */
+	struct mutex config_lock;	/* lock for configuration operations */
 	struct pds_core_dev_info_regs __iomem *info_regs;
 	struct pds_core_dev_cmd_regs __iomem *cmd_regs;
 	struct pds_core_intr __iomem *intr_ctrl;
@@ -52,5 +85,22 @@ void pdsc_debugfs_create(void);
 void pdsc_debugfs_destroy(void);
 void pdsc_debugfs_add_dev(struct pdsc *pdsc);
 void pdsc_debugfs_del_dev(struct pdsc *pdsc);
+void pdsc_debugfs_add_ident(struct pdsc *pdsc);
+void pdsc_debugfs_add_irqs(struct pdsc *pdsc);
+
+int pdsc_err_to_errno(enum pds_core_status_code code);
+bool pdsc_is_fw_running(struct pdsc *pdsc);
+bool pdsc_is_fw_good(struct pdsc *pdsc);
+int pdsc_devcmd(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		union pds_core_dev_comp *comp, int max_seconds);
+int pdsc_devcmd_locked(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		       union pds_core_dev_comp *comp, int max_seconds);
+int pdsc_devcmd_init(struct pdsc *pdsc);
+int pdsc_devcmd_reset(struct pdsc *pdsc);
+int pdsc_dev_reinit(struct pdsc *pdsc);
+int pdsc_dev_init(struct pdsc *pdsc);
+
+int pdsc_setup(struct pdsc *pdsc, bool init);
+void pdsc_teardown(struct pdsc *pdsc, bool removing);
 
 #endif /* _PDSC_H_ */
diff --git a/drivers/net/ethernet/amd/pds_core/debugfs.c b/drivers/net/ethernet/amd/pds_core/debugfs.c
index b2f7cb795c20..601431b41abb 100644
--- a/drivers/net/ethernet/amd/pds_core/debugfs.c
+++ b/drivers/net/ethernet/amd/pds_core/debugfs.c
@@ -29,3 +29,41 @@ void pdsc_debugfs_del_dev(struct pdsc *pdsc)
 	debugfs_remove_recursive(pdsc->dentry);
 	pdsc->dentry = NULL;
 }
+
+static int identity_show(struct seq_file *seq, void *v)
+{
+	struct pdsc *pdsc = seq->private;
+	struct pds_core_dev_identity *ident;
+	int vt;
+
+	ident = &pdsc->dev_ident;
+
+	seq_printf(seq, "fw_heartbeat:     0x%x\n",
+		   ioread32(&pdsc->info_regs->fw_heartbeat));
+
+	seq_printf(seq, "nlifs:            %d\n",
+		   le32_to_cpu(ident->nlifs));
+	seq_printf(seq, "nintrs:           %d\n",
+		   le32_to_cpu(ident->nintrs));
+	seq_printf(seq, "ndbpgs_per_lif:   %d\n",
+		   le32_to_cpu(ident->ndbpgs_per_lif));
+	seq_printf(seq, "intr_coal_mult:   %d\n",
+		   le32_to_cpu(ident->intr_coal_mult));
+	seq_printf(seq, "intr_coal_div:    %d\n",
+		   le32_to_cpu(ident->intr_coal_div));
+
+	seq_puts(seq, "vif_types:        ");
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++)
+		seq_printf(seq, "%d ",
+			   le16_to_cpu(pdsc->dev_ident.vif_types[vt]));
+	seq_puts(seq, "\n");
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(identity);
+
+void pdsc_debugfs_add_ident(struct pdsc *pdsc)
+{
+	debugfs_create_file("identity", 0400, pdsc->dentry,
+			    pdsc, &identity_fops);
+}
diff --git a/drivers/net/ethernet/amd/pds_core/dev.c b/drivers/net/ethernet/amd/pds_core/dev.c
new file mode 100644
index 000000000000..f082d69c5128
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/dev.c
@@ -0,0 +1,348 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/utsname.h>
+
+#include "core.h"
+
+int pdsc_err_to_errno(enum pds_core_status_code code)
+{
+	switch (code) {
+	case PDS_RC_SUCCESS:
+		return 0;
+	case PDS_RC_EVERSION:
+	case PDS_RC_EQTYPE:
+	case PDS_RC_EQID:
+	case PDS_RC_EINVAL:
+	case PDS_RC_ENOSUPP:
+		return -EINVAL;
+	case PDS_RC_EPERM:
+		return -EPERM;
+	case PDS_RC_ENOENT:
+		return -ENOENT;
+	case PDS_RC_EAGAIN:
+		return -EAGAIN;
+	case PDS_RC_ENOMEM:
+		return -ENOMEM;
+	case PDS_RC_EFAULT:
+		return -EFAULT;
+	case PDS_RC_EBUSY:
+		return -EBUSY;
+	case PDS_RC_EEXIST:
+		return -EEXIST;
+	case PDS_RC_EVFID:
+		return -ENODEV;
+	case PDS_RC_ECLIENT:
+		return -ECHILD;
+	case PDS_RC_ENOSPC:
+		return -ENOSPC;
+	case PDS_RC_ERANGE:
+		return -ERANGE;
+	case PDS_RC_BAD_ADDR:
+		return -EFAULT;
+	case PDS_RC_EOPCODE:
+	case PDS_RC_EINTR:
+	case PDS_RC_DEV_CMD:
+	case PDS_RC_ERROR:
+	case PDS_RC_ERDMA:
+	case PDS_RC_EIO:
+	default:
+		return -EIO;
+	}
+}
+
+bool pdsc_is_fw_running(struct pdsc *pdsc)
+{
+	pdsc->fw_status = ioread8(&pdsc->info_regs->fw_status);
+	pdsc->last_fw_time = jiffies;
+	pdsc->last_hb = ioread32(&pdsc->info_regs->fw_heartbeat);
+
+	/* Firmware is useful only if the running bit is set and
+	 * fw_status != 0xff (bad PCI read)
+	 */
+	return (pdsc->fw_status != 0xff) &&
+		(pdsc->fw_status & PDS_CORE_FW_STS_F_RUNNING);
+}
+
+bool pdsc_is_fw_good(struct pdsc *pdsc)
+{
+	u8 gen = pdsc->fw_status & PDS_CORE_FW_STS_F_GENERATION;
+
+	return pdsc_is_fw_running(pdsc) && gen == pdsc->fw_generation;
+}
+
+static u8 pdsc_devcmd_status(struct pdsc *pdsc)
+{
+	return ioread8(&pdsc->cmd_regs->comp.status);
+}
+
+static bool pdsc_devcmd_done(struct pdsc *pdsc)
+{
+	return ioread32(&pdsc->cmd_regs->done) & PDS_CORE_DEV_CMD_DONE;
+}
+
+static void pdsc_devcmd_dbell(struct pdsc *pdsc)
+{
+	iowrite32(0, &pdsc->cmd_regs->done);
+	iowrite32(1, &pdsc->cmd_regs->doorbell);
+}
+
+static void pdsc_devcmd_clean(struct pdsc *pdsc)
+{
+	iowrite32(0, &pdsc->cmd_regs->doorbell);
+	memset_io(&pdsc->cmd_regs->cmd, 0, sizeof(pdsc->cmd_regs->cmd));
+}
+
+static const char *pdsc_devcmd_str(int opcode)
+{
+	switch (opcode) {
+	case PDS_CORE_CMD_NOP:
+		return "PDS_CORE_CMD_NOP";
+	case PDS_CORE_CMD_IDENTIFY:
+		return "PDS_CORE_CMD_IDENTIFY";
+	case PDS_CORE_CMD_RESET:
+		return "PDS_CORE_CMD_RESET";
+	case PDS_CORE_CMD_INIT:
+		return "PDS_CORE_CMD_INIT";
+	case PDS_CORE_CMD_FW_DOWNLOAD:
+		return "PDS_CORE_CMD_FW_DOWNLOAD";
+	case PDS_CORE_CMD_FW_CONTROL:
+		return "PDS_CORE_CMD_FW_CONTROL";
+	default:
+		return "PDS_CORE_CMD_UNKNOWN";
+	}
+}
+
+static int pdsc_devcmd_wait(struct pdsc *pdsc, int max_seconds)
+{
+	struct device *dev = pdsc->dev;
+	unsigned long start_time;
+	unsigned long max_wait;
+	unsigned long duration;
+	int timeout = 0;
+	int done = 0;
+	int err = 0;
+	int status;
+	int opcode;
+
+	opcode = ioread8(&pdsc->cmd_regs->cmd.opcode);
+
+	start_time = jiffies;
+	max_wait = start_time + (max_seconds * HZ);
+
+	while (!done && !timeout) {
+		done = pdsc_devcmd_done(pdsc);
+		if (done)
+			break;
+
+		timeout = time_after(jiffies, max_wait);
+		if (timeout)
+			break;
+
+		usleep_range(100, 200);
+	}
+	duration = jiffies - start_time;
+
+	if (done && duration > HZ)
+		dev_dbg(dev, "DEVCMD %d %s after %ld secs\n",
+			opcode, pdsc_devcmd_str(opcode), duration / HZ);
+
+	if (!done || timeout) {
+		dev_err(dev, "DEVCMD %d %s timeout, done %d timeout %d max_seconds=%d\n",
+			opcode, pdsc_devcmd_str(opcode), done, timeout,
+			max_seconds);
+		err = -ETIMEDOUT;
+		pdsc_devcmd_clean(pdsc);
+	}
+
+	status = pdsc_devcmd_status(pdsc);
+	err = pdsc_err_to_errno(status);
+	if (err && err != -EAGAIN)
+		dev_err(dev, "DEVCMD %d %s failed, status=%d err %d %pe\n",
+			opcode, pdsc_devcmd_str(opcode), status, err,
+			ERR_PTR(err));
+
+	return err;
+}
+
+int pdsc_devcmd_locked(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		       union pds_core_dev_comp *comp, int max_seconds)
+{
+	int err;
+
+	memcpy_toio(&pdsc->cmd_regs->cmd, cmd, sizeof(*cmd));
+	pdsc_devcmd_dbell(pdsc);
+	err = pdsc_devcmd_wait(pdsc, max_seconds);
+	memcpy_fromio(comp, &pdsc->cmd_regs->comp, sizeof(*comp));
+
+	return err;
+}
+
+int pdsc_devcmd(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
+		union pds_core_dev_comp *comp, int max_seconds)
+{
+	int err;
+
+	mutex_lock(&pdsc->devcmd_lock);
+	err = pdsc_devcmd_locked(pdsc, cmd, comp, max_seconds);
+	mutex_unlock(&pdsc->devcmd_lock);
+
+	return err;
+}
+
+int pdsc_devcmd_init(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.opcode = PDS_CORE_CMD_INIT,
+	};
+
+	return pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+int pdsc_devcmd_reset(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.reset.opcode = PDS_CORE_CMD_RESET,
+	};
+
+	return pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static int pdsc_devcmd_identify_locked(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.identify.opcode = PDS_CORE_CMD_IDENTIFY,
+		.identify.ver = PDS_CORE_IDENTITY_VERSION_1,
+	};
+
+	return pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static void pdsc_init_devinfo(struct pdsc *pdsc)
+{
+	pdsc->dev_info.asic_type = ioread8(&pdsc->info_regs->asic_type);
+	pdsc->dev_info.asic_rev = ioread8(&pdsc->info_regs->asic_rev);
+	pdsc->fw_generation = PDS_CORE_FW_STS_F_GENERATION &
+			      ioread8(&pdsc->info_regs->fw_status);
+
+	memcpy_fromio(pdsc->dev_info.fw_version,
+		      pdsc->info_regs->fw_version,
+		      PDS_CORE_DEVINFO_FWVERS_BUFLEN);
+	pdsc->dev_info.fw_version[PDS_CORE_DEVINFO_FWVERS_BUFLEN] = 0;
+
+	memcpy_fromio(pdsc->dev_info.serial_num,
+		      pdsc->info_regs->serial_num,
+		      PDS_CORE_DEVINFO_SERIAL_BUFLEN);
+	pdsc->dev_info.serial_num[PDS_CORE_DEVINFO_SERIAL_BUFLEN] = 0;
+
+	dev_dbg(pdsc->dev, "fw_version %s\n", pdsc->dev_info.fw_version);
+}
+
+static int pdsc_identify(struct pdsc *pdsc)
+{
+	struct pds_core_drv_identity drv = {};
+	size_t sz;
+	int err;
+
+	drv.drv_type = cpu_to_le32(PDS_DRIVER_LINUX);
+	snprintf(drv.driver_ver_str, sizeof(drv.driver_ver_str),
+		 "%s %s", PDS_CORE_DRV_NAME, utsname()->release);
+
+	/* Next let's get some info about the device
+	 * We use the devcmd_lock at this level in order to
+	 * get safe access to the cmd_regs->data before anyone
+	 * else can mess it up
+	 */
+	mutex_lock(&pdsc->devcmd_lock);
+
+	sz = min_t(size_t, sizeof(drv), sizeof(pdsc->cmd_regs->data));
+	memcpy_toio(&pdsc->cmd_regs->data, &drv, sz);
+
+	err = pdsc_devcmd_identify_locked(pdsc);
+	if (!err) {
+		sz = min_t(size_t, sizeof(pdsc->dev_ident),
+			   sizeof(pdsc->cmd_regs->data));
+		memcpy_fromio(&pdsc->dev_ident, &pdsc->cmd_regs->data, sz);
+	}
+	mutex_unlock(&pdsc->devcmd_lock);
+
+	if (err) {
+		dev_err(pdsc->dev, "Cannot identify device: %pe\n",
+			ERR_PTR(err));
+		return err;
+	}
+
+	if (isprint(pdsc->dev_info.fw_version[0]) &&
+	    isascii(pdsc->dev_info.fw_version[0]))
+		dev_info(pdsc->dev, "FW: %.*s\n",
+			 (int)(sizeof(pdsc->dev_info.fw_version) - 1),
+			 pdsc->dev_info.fw_version);
+	else
+		dev_info(pdsc->dev, "FW: (invalid string) 0x%02x 0x%02x 0x%02x 0x%02x ...\n",
+			 (u8)pdsc->dev_info.fw_version[0],
+			 (u8)pdsc->dev_info.fw_version[1],
+			 (u8)pdsc->dev_info.fw_version[2],
+			 (u8)pdsc->dev_info.fw_version[3]);
+
+	return 0;
+}
+
+int pdsc_dev_reinit(struct pdsc *pdsc)
+{
+	pdsc_init_devinfo(pdsc);
+
+	return pdsc_identify(pdsc);
+}
+
+int pdsc_dev_init(struct pdsc *pdsc)
+{
+	unsigned int nintrs;
+	int err;
+
+	/* Initial init and reset of device */
+	pdsc_init_devinfo(pdsc);
+	pdsc->devcmd_timeout = PDS_CORE_DEVCMD_TIMEOUT;
+
+	err = pdsc_devcmd_reset(pdsc);
+	if (err)
+		return err;
+
+	err = pdsc_identify(pdsc);
+	if (err)
+		return err;
+
+	pdsc_debugfs_add_ident(pdsc);
+
+	/* Now we can reserve interrupts */
+	nintrs = le32_to_cpu(pdsc->dev_ident.nintrs);
+	nintrs = min_t(unsigned int, num_online_cpus(), nintrs);
+
+	/* Get intr_info struct array for tracking */
+	pdsc->intr_info = kcalloc(nintrs, sizeof(*pdsc->intr_info), GFP_KERNEL);
+	if (!pdsc->intr_info) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	err = pci_alloc_irq_vectors(pdsc->pdev, nintrs, nintrs, PCI_IRQ_MSIX);
+	if (err != nintrs) {
+		dev_err(pdsc->dev, "Can't get %d intrs from OS: %pe\n",
+			nintrs, ERR_PTR(err));
+		err = -ENOSPC;
+		goto err_out;
+	}
+	pdsc->nintrs = nintrs;
+
+	return 0;
+
+err_out:
+	kfree(pdsc->intr_info);
+	pdsc->intr_info = NULL;
+
+	return err;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index c2b12f226959..09afb069dcb3 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -137,6 +137,18 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 	if (err)
 		goto err_out_release_regions;
 
+	mutex_init(&pdsc->devcmd_lock);
+	mutex_init(&pdsc->config_lock);
+
+	mutex_lock(&pdsc->config_lock);
+	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
+
+	err = pdsc_setup(pdsc, PDSC_SETUP_INIT);
+	if (err)
+		goto err_out_unmap_bars;
+
+	mutex_unlock(&pdsc->config_lock);
+
 	dl = priv_to_devlink(pdsc);
 	devl_lock(dl);
 	devl_register(dl);
@@ -144,6 +156,12 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	return 0;
 
+err_out_unmap_bars:
+	mutex_unlock(&pdsc->config_lock);
+	mutex_destroy(&pdsc->config_lock);
+	mutex_destroy(&pdsc->devcmd_lock);
+	pci_free_irq_vectors(pdsc->pdev);
+	pdsc_unmap_bars(pdsc);
 err_out_release_regions:
 	pci_release_regions(pdsc->pdev);
 
@@ -240,8 +258,19 @@ static void pdsc_remove(struct pci_dev *pdev)
 	devl_unregister(dl);
 	devl_unlock(dl);
 
-	pdsc_unmap_bars(pdsc);
-	pci_release_regions(pdev);
+	if (!pdev->is_virtfn) {
+		mutex_lock(&pdsc->config_lock);
+		set_bit(PDSC_S_STOPPING_DRIVER, &pdsc->state);
+
+		pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
+		mutex_unlock(&pdsc->config_lock);
+		mutex_destroy(&pdsc->config_lock);
+		mutex_destroy(&pdsc->devcmd_lock);
+
+		pci_free_irq_vectors(pdev);
+		pdsc_unmap_bars(pdsc);
+		pci_release_regions(pdev);
+	}
 
 	pci_clear_master(pdev);
 	pci_disable_device(pdev);
diff --git a/include/linux/pds/pds_common.h b/include/linux/pds/pds_common.h
index bd041a5170a6..f0798ce01acf 100644
--- a/include/linux/pds/pds_common.h
+++ b/include/linux/pds/pds_common.h
@@ -11,4 +11,34 @@
 #define PDS_CORE_ADDR_MASK	(BIT_ULL(PDS_ADDR_LEN) - 1)
 #define PDS_PAGE_SIZE		4096
 
+enum pds_core_driver_type {
+	PDS_DRIVER_LINUX   = 1,
+	PDS_DRIVER_WIN     = 2,
+	PDS_DRIVER_DPDK    = 3,
+	PDS_DRIVER_FREEBSD = 4,
+	PDS_DRIVER_IPXE    = 5,
+	PDS_DRIVER_ESXI    = 6,
+};
+
+#define PDS_CORE_IFNAMSIZ		16
+
+/**
+ * enum pds_core_logical_qtype - Logical Queue Types
+ * @PDS_CORE_QTYPE_ADMINQ:    Administrative Queue
+ * @PDS_CORE_QTYPE_NOTIFYQ:   Notify Queue
+ * @PDS_CORE_QTYPE_RXQ:       Receive Queue
+ * @PDS_CORE_QTYPE_TXQ:       Transmit Queue
+ * @PDS_CORE_QTYPE_EQ:        Event Queue
+ * @PDS_CORE_QTYPE_MAX:       Max queue type supported
+ */
+enum pds_core_logical_qtype {
+	PDS_CORE_QTYPE_ADMINQ  = 0,
+	PDS_CORE_QTYPE_NOTIFYQ = 1,
+	PDS_CORE_QTYPE_RXQ     = 2,
+	PDS_CORE_QTYPE_TXQ     = 3,
+	PDS_CORE_QTYPE_EQ      = 4,
+
+	PDS_CORE_QTYPE_MAX     = 16   /* don't change - used in struct size */
+};
+
 #endif /* _PDS_COMMON_H_ */
diff --git a/include/linux/pds/pds_intr.h b/include/linux/pds/pds_intr.h
new file mode 100644
index 000000000000..56277c37248c
--- /dev/null
+++ b/include/linux/pds/pds_intr.h
@@ -0,0 +1,163 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR Linux-OpenIB) OR BSD-2-Clause */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc. */
+
+#ifndef _PDS_INTR_H_
+#define _PDS_INTR_H_
+
+/*
+ * Interrupt control register
+ * @coal_init:        Coalescing timer initial value, in
+ *                    device units.  Use @identity->intr_coal_mult
+ *                    and @identity->intr_coal_div to convert from
+ *                    usecs to device units:
+ *
+ *                      coal_init = coal_usecs * coal_mutl / coal_div
+ *
+ *                    When an interrupt is sent the interrupt
+ *                    coalescing timer current value
+ *                    (@coalescing_curr) is initialized with this
+ *                    value and begins counting down.  No more
+ *                    interrupts are sent until the coalescing
+ *                    timer reaches 0.  When @coalescing_init=0
+ *                    interrupt coalescing is effectively disabled
+ *                    and every interrupt assert results in an
+ *                    interrupt.  Reset value: 0
+ * @mask:             Interrupt mask.  When @mask=1 the interrupt
+ *                    resource will not send an interrupt.  When
+ *                    @mask=0 the interrupt resource will send an
+ *                    interrupt if an interrupt event is pending
+ *                    or on the next interrupt assertion event.
+ *                    Reset value: 1
+ * @credits:          Interrupt credits.  This register indicates
+ *                    how many interrupt events the hardware has
+ *                    sent.  When written by software this
+ *                    register atomically decrements @int_credits
+ *                    by the value written.  When @int_credits
+ *                    becomes 0 then the "pending interrupt" bit
+ *                    in the Interrupt Status register is cleared
+ *                    by the hardware and any pending but unsent
+ *                    interrupts are cleared.
+ *                    !!!IMPORTANT!!! This is a signed register.
+ * @flags:            Interrupt control flags
+ *                       @unmask -- When this bit is written with a 1
+ *                       the interrupt resource will set mask=0.
+ *                       @coal_timer_reset -- When this
+ *                       bit is written with a 1 the
+ *                       @coalescing_curr will be reloaded with
+ *                       @coalescing_init to reset the coalescing
+ *                       timer.
+ * @mask_on_assert:   Automatically mask on assertion.  When
+ *                    @mask_on_assert=1 the interrupt resource
+ *                    will set @mask=1 whenever an interrupt is
+ *                    sent.  When using interrupts in Legacy
+ *                    Interrupt mode the driver must select
+ *                    @mask_on_assert=0 for proper interrupt
+ *                    operation.
+ * @coalescing_curr:  Coalescing timer current value, in
+ *                    microseconds.  When this value reaches 0
+ *                    the interrupt resource is again eligible to
+ *                    send an interrupt.  If an interrupt event
+ *                    is already pending when @coalescing_curr
+ *                    reaches 0 the pending interrupt will be
+ *                    sent, otherwise an interrupt will be sent
+ *                    on the next interrupt assertion event.
+ */
+struct pds_core_intr {
+	u32 coal_init;
+	u32 mask;
+	u16 credits;
+	u16 flags;
+#define PDS_CORE_INTR_F_UNMASK		0x0001
+#define PDS_CORE_INTR_F_TIMER_RESET	0x0002
+	u32 mask_on_assert;
+	u32 coalescing_curr;
+	u32 rsvd6[3];
+};
+
+#ifndef __CHECKER__
+static_assert(sizeof(struct pds_core_intr) == 32);
+#endif /* __CHECKER__ */
+
+#define PDS_CORE_INTR_CTRL_REGS_MAX		2048
+#define PDS_CORE_INTR_CTRL_COAL_MAX		0x3F
+#define PDS_CORE_INTR_INDEX_NOT_ASSIGNED	-1
+
+struct pds_core_intr_status {
+	u32 status[2];
+};
+
+/**
+ * enum pds_core_intr_mask_vals - valid values for mask and mask_assert.
+ * @PDS_CORE_INTR_MASK_CLEAR:	unmask interrupt.
+ * @PDS_CORE_INTR_MASK_SET:	mask interrupt.
+ */
+enum pds_core_intr_mask_vals {
+	PDS_CORE_INTR_MASK_CLEAR	= 0,
+	PDS_CORE_INTR_MASK_SET		= 1,
+};
+
+/**
+ * enum pds_core_intr_credits_bits - Bitwise composition of credits values.
+ * @PDS_CORE_INTR_CRED_COUNT:	bit mask of credit count, no shift needed.
+ * @PDS_CORE_INTR_CRED_COUNT_SIGNED: bit mask of credit count, including sign bit.
+ * @PDS_CORE_INTR_CRED_UNMASK:	unmask the interrupt.
+ * @PDS_CORE_INTR_CRED_RESET_COALESCE: reset the coalesce timer.
+ * @PDS_CORE_INTR_CRED_REARM:	unmask the and reset the timer.
+ */
+enum pds_core_intr_credits_bits {
+	PDS_CORE_INTR_CRED_COUNT		= 0x7fffu,
+	PDS_CORE_INTR_CRED_COUNT_SIGNED		= 0xffffu,
+	PDS_CORE_INTR_CRED_UNMASK		= 0x10000u,
+	PDS_CORE_INTR_CRED_RESET_COALESCE	= 0x20000u,
+	PDS_CORE_INTR_CRED_REARM		= (PDS_CORE_INTR_CRED_UNMASK |
+					   PDS_CORE_INTR_CRED_RESET_COALESCE),
+};
+
+static inline void
+pds_core_intr_coal_init(struct pds_core_intr __iomem *intr_ctrl, u32 coal)
+{
+	iowrite32(coal, &intr_ctrl->coal_init);
+}
+
+static inline void
+pds_core_intr_mask(struct pds_core_intr __iomem *intr_ctrl, u32 mask)
+{
+	iowrite32(mask, &intr_ctrl->mask);
+}
+
+static inline void
+pds_core_intr_credits(struct pds_core_intr __iomem *intr_ctrl,
+		      u32 cred, u32 flags)
+{
+	if (WARN_ON_ONCE(cred > PDS_CORE_INTR_CRED_COUNT)) {
+		cred = ioread32(&intr_ctrl->credits);
+		cred &= PDS_CORE_INTR_CRED_COUNT_SIGNED;
+	}
+
+	iowrite32(cred | flags, &intr_ctrl->credits);
+}
+
+static inline void
+pds_core_intr_clean_flags(struct pds_core_intr __iomem *intr_ctrl, u32 flags)
+{
+	u32 cred;
+
+	cred = ioread32(&intr_ctrl->credits);
+	cred &= PDS_CORE_INTR_CRED_COUNT_SIGNED;
+	cred |= flags;
+	iowrite32(cred, &intr_ctrl->credits);
+}
+
+static inline void
+pds_core_intr_clean(struct pds_core_intr __iomem *intr_ctrl)
+{
+	pds_core_intr_clean_flags(intr_ctrl, PDS_CORE_INTR_CRED_RESET_COALESCE);
+}
+
+static inline void
+pds_core_intr_mask_assert(struct pds_core_intr __iomem *intr_ctrl, u32 mask)
+{
+	iowrite32(mask, &intr_ctrl->mask_on_assert);
+}
+
+#endif /* _PDS_INTR_H_ */
-- 
cgit v1.2.3-70-g09d2


From c2dbb0904310362aaff213e8c877e655c745d0d1 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:16 -0700
Subject: pds_core: health timer and workqueue

Add in the periodic health check and the related workqueue,
as well as the handlers for when a FW reset is seen.

The firmware is polled every 5 seconds to be sure that it is
still alive and that the FW generation didn't change.

The alive check looks to see that the PCI bus is still readable
and the fw_status still has the RUNNING bit on.  If not alive,
the driver stops activity and tears things down.  When the FW
recovers and the alive check again succeeds, the driver sets
back up for activity.

The generation check looks at the fw_generation to see if it
has changed, which can happen if the FW crashed and recovered
or was updated in between health checks.  If changed, the
driver counts that as though the alive test failed and forces
the fw_down/fw_up cycle.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/core.c | 61 ++++++++++++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/core.h |  8 +++++
 drivers/net/ethernet/amd/pds_core/dev.c  |  3 ++
 drivers/net/ethernet/amd/pds_core/main.c | 31 ++++++++++++++++
 4 files changed, 103 insertions(+)

diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index 80d2ecb045df..701d27471858 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -34,3 +34,64 @@ void pdsc_teardown(struct pdsc *pdsc, bool removing)
 
 	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
 }
+
+static void pdsc_fw_down(struct pdsc *pdsc)
+{
+	if (test_and_set_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
+		dev_err(pdsc->dev, "%s: already happening\n", __func__);
+		return;
+	}
+
+	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
+}
+
+static void pdsc_fw_up(struct pdsc *pdsc)
+{
+	int err;
+
+	if (!test_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
+		dev_err(pdsc->dev, "%s: fw not dead\n", __func__);
+		return;
+	}
+
+	err = pdsc_setup(pdsc, PDSC_SETUP_RECOVERY);
+	if (err)
+		goto err_out;
+
+	return;
+
+err_out:
+	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
+}
+
+void pdsc_health_thread(struct work_struct *work)
+{
+	struct pdsc *pdsc = container_of(work, struct pdsc, health_work);
+	unsigned long mask;
+	bool healthy;
+
+	mutex_lock(&pdsc->config_lock);
+
+	/* Don't do a check when in a transition state */
+	mask = BIT_ULL(PDSC_S_INITING_DRIVER) |
+	       BIT_ULL(PDSC_S_STOPPING_DRIVER);
+	if (pdsc->state & mask)
+		goto out_unlock;
+
+	healthy = pdsc_is_fw_good(pdsc);
+	dev_dbg(pdsc->dev, "%s: health %d fw_status %#02x fw_heartbeat %d\n",
+		__func__, healthy, pdsc->fw_status, pdsc->last_hb);
+
+	if (test_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
+		if (healthy)
+			pdsc_fw_up(pdsc);
+	} else {
+		if (!healthy)
+			pdsc_fw_down(pdsc);
+	}
+
+	pdsc->fw_generation = pdsc->fw_status & PDS_CORE_FW_STS_F_GENERATION;
+
+out_unlock:
+	mutex_unlock(&pdsc->config_lock);
+}
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index fcf6c6545c49..83c528a2a131 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -12,6 +12,8 @@
 #include <linux/pds/pds_intr.h>
 
 #define PDSC_DRV_DESCRIPTION	"AMD/Pensando Core Driver"
+
+#define PDSC_WATCHDOG_SECS	5
 #define PDSC_TEARDOWN_RECOVERY	false
 #define PDSC_TEARDOWN_REMOVING	true
 #define PDSC_SETUP_RECOVERY	false
@@ -63,12 +65,17 @@ struct pdsc {
 	u8 fw_generation;
 	unsigned long last_fw_time;
 	u32 last_hb;
+	struct timer_list wdtimer;
+	unsigned int wdtimer_period;
+	struct work_struct health_work;
 
 	struct pdsc_devinfo dev_info;
 	struct pds_core_dev_identity dev_ident;
 	unsigned int nintrs;
 	struct pdsc_intr_info *intr_info;	/* array of nintrs elements */
 
+	struct workqueue_struct *wq;
+
 	unsigned int devcmd_timeout;
 	struct mutex devcmd_lock;	/* lock for dev_cmd operations */
 	struct mutex config_lock;	/* lock for configuration operations */
@@ -102,5 +109,6 @@ int pdsc_dev_init(struct pdsc *pdsc);
 
 int pdsc_setup(struct pdsc *pdsc, bool init);
 void pdsc_teardown(struct pdsc *pdsc, bool removing);
+void pdsc_health_thread(struct work_struct *work);
 
 #endif /* _PDSC_H_ */
diff --git a/drivers/net/ethernet/amd/pds_core/dev.c b/drivers/net/ethernet/amd/pds_core/dev.c
index f082d69c5128..f7c597ea5daf 100644
--- a/drivers/net/ethernet/amd/pds_core/dev.c
+++ b/drivers/net/ethernet/amd/pds_core/dev.c
@@ -177,6 +177,9 @@ int pdsc_devcmd_locked(struct pdsc *pdsc, union pds_core_dev_cmd *cmd,
 	err = pdsc_devcmd_wait(pdsc, max_seconds);
 	memcpy_fromio(comp, &pdsc->cmd_regs->comp, sizeof(*comp));
 
+	if (err == -ENXIO || err == -ETIMEDOUT)
+		queue_work(pdsc->wq, &pdsc->health_work);
+
 	return err;
 }
 
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index 09afb069dcb3..c9fbf1d374a7 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -20,6 +20,17 @@ static const struct pci_device_id pdsc_id_table[] = {
 };
 MODULE_DEVICE_TABLE(pci, pdsc_id_table);
 
+static void pdsc_wdtimer_cb(struct timer_list *t)
+{
+	struct pdsc *pdsc = from_timer(pdsc, t, wdtimer);
+
+	dev_dbg(pdsc->dev, "%s: jiffies %ld\n", __func__, jiffies);
+	mod_timer(&pdsc->wdtimer,
+		  round_jiffies(jiffies + pdsc->wdtimer_period));
+
+	queue_work(pdsc->wq, &pdsc->health_work);
+}
+
 static void pdsc_unmap_bars(struct pdsc *pdsc)
 {
 	struct pdsc_dev_bar *bars = pdsc->bars;
@@ -119,8 +130,11 @@ static int pdsc_init_vf(struct pdsc *vf)
 	return -1;
 }
 
+#define PDSC_WQ_NAME_LEN 24
+
 static int pdsc_init_pf(struct pdsc *pdsc)
 {
+	char wq_name[PDSC_WQ_NAME_LEN];
 	struct devlink *dl;
 	int err;
 
@@ -137,6 +151,13 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 	if (err)
 		goto err_out_release_regions;
 
+	/* General workqueue and timer, but don't start timer yet */
+	snprintf(wq_name, sizeof(wq_name), "%s.%d", PDS_CORE_DRV_NAME, pdsc->uid);
+	pdsc->wq = create_singlethread_workqueue(wq_name);
+	INIT_WORK(&pdsc->health_work, pdsc_health_thread);
+	timer_setup(&pdsc->wdtimer, pdsc_wdtimer_cb, 0);
+	pdsc->wdtimer_period = PDSC_WATCHDOG_SECS * HZ;
+
 	mutex_init(&pdsc->devcmd_lock);
 	mutex_init(&pdsc->config_lock);
 
@@ -154,10 +175,16 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 	devl_register(dl);
 	devl_unlock(dl);
 
+	/* Lastly, start the health check timer */
+	mod_timer(&pdsc->wdtimer, round_jiffies(jiffies + pdsc->wdtimer_period));
+
 	return 0;
 
 err_out_unmap_bars:
 	mutex_unlock(&pdsc->config_lock);
+	del_timer_sync(&pdsc->wdtimer);
+	if (pdsc->wq)
+		destroy_workqueue(pdsc->wq);
 	mutex_destroy(&pdsc->config_lock);
 	mutex_destroy(&pdsc->devcmd_lock);
 	pci_free_irq_vectors(pdsc->pdev);
@@ -259,6 +286,10 @@ static void pdsc_remove(struct pci_dev *pdev)
 	devl_unlock(dl);
 
 	if (!pdev->is_virtfn) {
+		del_timer_sync(&pdsc->wdtimer);
+		if (pdsc->wq)
+			destroy_workqueue(pdsc->wq);
+
 		mutex_lock(&pdsc->config_lock);
 		set_bit(PDSC_S_STOPPING_DRIVER, &pdsc->state);
 
-- 
cgit v1.2.3-70-g09d2


From 25b450c05a49e34f2d08ac469ae19d7651abc57c Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:17 -0700
Subject: pds_core: add devlink health facilities

Add devlink health reporting on top of our fw watchdog.

Example:
  # devlink health show pci/0000:2b:00.0 reporter fw
  pci/0000:2b:00.0:
    reporter fw
      state healthy error 0 recover 0
  # devlink health diagnose pci/0000:2b:00.0 reporter fw
   Status: healthy State: 1 Generation: 0 Recoveries: 0

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../device_drivers/ethernet/amd/pds_core.rst       | 12 +++++++
 drivers/net/ethernet/amd/pds_core/Makefile         |  1 +
 drivers/net/ethernet/amd/pds_core/core.c           |  8 ++++-
 drivers/net/ethernet/amd/pds_core/core.h           |  6 ++++
 drivers/net/ethernet/amd/pds_core/devlink.c        | 40 ++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/main.c           | 22 ++++++++++++
 6 files changed, 88 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/amd/pds_core/devlink.c

diff --git a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
index 99a70026f1bc..5b88173a20ff 100644
--- a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
+++ b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
@@ -26,6 +26,18 @@ messages such as these::
   pds_core 0000:b6:00.0: 252.048 Gb/s available PCIe bandwidth (16.0 GT/s PCIe x16 link)
   pds_core 0000:b6:00.0: FW: 1.60.0-73
 
+Health Reporters
+================
+
+The driver supports a devlink health reporter for FW status::
+
+  # devlink health show pci/0000:2b:00.0 reporter fw
+  pci/0000:2b:00.0:
+    reporter fw
+      state healthy error 0 recover 0
+  # devlink health diagnose pci/0000:2b:00.0 reporter fw
+   Status: healthy State: 1 Generation: 0 Recoveries: 0
+
 Support
 =======
 
diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
index 95a6c31e92d2..eaca8557ba66 100644
--- a/drivers/net/ethernet/amd/pds_core/Makefile
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -4,6 +4,7 @@
 obj-$(CONFIG_PDS_CORE) := pds_core.o
 
 pds_core-y := main.o \
+	      devlink.o \
 	      dev.o \
 	      core.o
 
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index 701d27471858..ab8531386226 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -5,7 +5,7 @@
 
 int pdsc_setup(struct pdsc *pdsc, bool init)
 {
-	int err = 0;
+	int err;
 
 	if (init)
 		err = pdsc_dev_init(pdsc);
@@ -42,6 +42,8 @@ static void pdsc_fw_down(struct pdsc *pdsc)
 		return;
 	}
 
+	devlink_health_report(pdsc->fw_reporter, "FW down reported", pdsc);
+
 	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
 }
 
@@ -58,6 +60,10 @@ static void pdsc_fw_up(struct pdsc *pdsc)
 	if (err)
 		goto err_out;
 
+	pdsc->fw_recoveries++;
+	devlink_health_reporter_state_update(pdsc->fw_reporter,
+					     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
+
 	return;
 
 err_out:
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 83c528a2a131..32aa38c40024 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -68,6 +68,8 @@ struct pdsc {
 	struct timer_list wdtimer;
 	unsigned int wdtimer_period;
 	struct work_struct health_work;
+	struct devlink_health_reporter *fw_reporter;
+	u32 fw_recoveries;
 
 	struct pdsc_devinfo dev_info;
 	struct pds_core_dev_identity dev_ident;
@@ -88,6 +90,10 @@ struct pdsc {
 	u64 __iomem *kern_dbpage;
 };
 
+int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
+			      struct devlink_fmsg *fmsg,
+			      struct netlink_ext_ack *extack);
+
 void pdsc_debugfs_create(void);
 void pdsc_debugfs_destroy(void);
 void pdsc_debugfs_add_dev(struct pdsc *pdsc);
diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c
new file mode 100644
index 000000000000..3b05b1af65d1
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/devlink.c
@@ -0,0 +1,40 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include "core.h"
+
+int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
+			      struct devlink_fmsg *fmsg,
+			      struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_health_reporter_priv(reporter);
+	int err;
+
+	mutex_lock(&pdsc->config_lock);
+
+	if (test_bit(PDSC_S_FW_DEAD, &pdsc->state))
+		err = devlink_fmsg_string_pair_put(fmsg, "Status", "dead");
+	else if (!pdsc_is_fw_good(pdsc))
+		err = devlink_fmsg_string_pair_put(fmsg, "Status", "unhealthy");
+	else
+		err = devlink_fmsg_string_pair_put(fmsg, "Status", "healthy");
+
+	mutex_unlock(&pdsc->config_lock);
+
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "State",
+					pdsc->fw_status &
+						~PDS_CORE_FW_STS_F_GENERATION);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "Generation",
+					pdsc->fw_generation >> 4);
+	if (err)
+		return err;
+
+	return devlink_fmsg_u32_pair_put(fmsg, "Recoveries",
+					 pdsc->fw_recoveries);
+}
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index c9fbf1d374a7..54f3aed7adb1 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -130,10 +130,16 @@ static int pdsc_init_vf(struct pdsc *vf)
 	return -1;
 }
 
+static const struct devlink_health_reporter_ops pdsc_fw_reporter_ops = {
+	.name = "fw",
+	.diagnose = pdsc_fw_reporter_diagnose,
+};
+
 #define PDSC_WQ_NAME_LEN 24
 
 static int pdsc_init_pf(struct pdsc *pdsc)
 {
+	struct devlink_health_reporter *hr;
 	char wq_name[PDSC_WQ_NAME_LEN];
 	struct devlink *dl;
 	int err;
@@ -172,6 +178,16 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	dl = priv_to_devlink(pdsc);
 	devl_lock(dl);
+
+	hr = devl_health_reporter_create(dl, &pdsc_fw_reporter_ops, 0, pdsc);
+	if (IS_ERR(hr)) {
+		dev_warn(pdsc->dev, "Failed to create fw reporter: %pe\n", hr);
+		err = PTR_ERR(hr);
+		devl_unlock(dl);
+		goto err_out_teardown;
+	}
+	pdsc->fw_reporter = hr;
+
 	devl_register(dl);
 	devl_unlock(dl);
 
@@ -180,6 +196,8 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	return 0;
 
+err_out_teardown:
+	pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
 err_out_unmap_bars:
 	mutex_unlock(&pdsc->config_lock);
 	del_timer_sync(&pdsc->wdtimer);
@@ -283,6 +301,10 @@ static void pdsc_remove(struct pci_dev *pdev)
 	dl = priv_to_devlink(pdsc);
 	devl_lock(dl);
 	devl_unregister(dl);
+	if (pdsc->fw_reporter) {
+		devl_health_reporter_destroy(pdsc->fw_reporter);
+		pdsc->fw_reporter = NULL;
+	}
 	devl_unlock(dl);
 
 	if (!pdev->is_virtfn) {
-- 
cgit v1.2.3-70-g09d2


From 45d76f492938cdc27ddadc16e1e75103f4cfbf56 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:18 -0700
Subject: pds_core: set up device and adminq

Set up the basic adminq and notifyq queue structures.  These are
used mostly by the client drivers for feature configuration.
These are essentially the same adminq and notifyq as in the
ionic driver.

Part of this includes querying for device identity and FW
information, so we can make that available to devlink dev info.

  $ devlink dev info pci/0000:b5:00.0
  pci/0000:b5:00.0:
    driver pds_core
    serial_number FLM18420073
    versions:
        fixed:
          asic.id 0x0
          asic.rev 0x0
        running:
          fw 1.51.0-73
        stored:
          fw.goldfw 1.15.9-C-22
          fw.mainfwa 1.60.0-73
          fw.mainfwb 1.60.0-57

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../device_drivers/ethernet/amd/pds_core.rst       |  47 ++
 drivers/net/ethernet/amd/pds_core/core.c           | 431 +++++++++++++-
 drivers/net/ethernet/amd/pds_core/core.h           | 151 +++++
 drivers/net/ethernet/amd/pds_core/debugfs.c        |  77 +++
 drivers/net/ethernet/amd/pds_core/devlink.c        |  61 ++
 drivers/net/ethernet/amd/pds_core/main.c           |  17 +-
 include/linux/pds/pds_adminq.h                     | 638 +++++++++++++++++++++
 7 files changed, 1418 insertions(+), 4 deletions(-)
 create mode 100644 include/linux/pds/pds_adminq.h

diff --git a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
index 5b88173a20ff..a48eafb3d0d3 100644
--- a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
+++ b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
@@ -26,6 +26,53 @@ messages such as these::
   pds_core 0000:b6:00.0: 252.048 Gb/s available PCIe bandwidth (16.0 GT/s PCIe x16 link)
   pds_core 0000:b6:00.0: FW: 1.60.0-73
 
+Driver and firmware version information can be gathered with devlink::
+
+  $ devlink dev info pci/0000:b5:00.0
+  pci/0000:b5:00.0:
+    driver pds_core
+    serial_number FLM18420073
+    versions:
+        fixed:
+          asic.id 0x0
+          asic.rev 0x0
+        running:
+          fw 1.51.0-73
+        stored:
+          fw.goldfw 1.15.9-C-22
+          fw.mainfwa 1.60.0-73
+          fw.mainfwb 1.60.0-57
+
+Info versions
+=============
+
+The ``pds_core`` driver reports the following versions
+
+.. list-table:: devlink info versions implemented
+   :widths: 5 5 90
+
+   * - Name
+     - Type
+     - Description
+   * - ``fw``
+     - running
+     - Version of firmware running on the device
+   * - ``fw.goldfw``
+     - stored
+     - Version of firmware stored in the goldfw slot
+   * - ``fw.mainfwa``
+     - stored
+     - Version of firmware stored in the mainfwa slot
+   * - ``fw.mainfwb``
+     - stored
+     - Version of firmware stored in the mainfwb slot
+   * - ``asic.id``
+     - fixed
+     - The ASIC type for this device
+   * - ``asic.rev``
+     - fixed
+     - The revision of the ASIC for this device
+
 Health Reporters
 ================
 
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index ab8531386226..8c0dbdb5efc5 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -1,10 +1,365 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright(c) 2023 Advanced Micro Devices, Inc */
 
+#include <linux/pci.h>
+#include <linux/vmalloc.h>
+
 #include "core.h"
 
+void pdsc_work_thread(struct work_struct *work)
+{
+	/* stub */
+}
+
+irqreturn_t pdsc_adminq_isr(int irq, void *data)
+{
+	/* stub */
+	return IRQ_HANDLED;
+}
+
+void pdsc_intr_free(struct pdsc *pdsc, int index)
+{
+	struct pdsc_intr_info *intr_info;
+
+	if (index >= pdsc->nintrs || index < 0) {
+		WARN(true, "bad intr index %d\n", index);
+		return;
+	}
+
+	intr_info = &pdsc->intr_info[index];
+	if (!intr_info->vector)
+		return;
+	dev_dbg(pdsc->dev, "%s: idx %d vec %d name %s\n",
+		__func__, index, intr_info->vector, intr_info->name);
+
+	pds_core_intr_mask(&pdsc->intr_ctrl[index], PDS_CORE_INTR_MASK_SET);
+	pds_core_intr_clean(&pdsc->intr_ctrl[index]);
+
+	free_irq(intr_info->vector, intr_info->data);
+
+	memset(intr_info, 0, sizeof(*intr_info));
+}
+
+int pdsc_intr_alloc(struct pdsc *pdsc, char *name,
+		    irq_handler_t handler, void *data)
+{
+	struct pdsc_intr_info *intr_info;
+	unsigned int index;
+	int err;
+
+	/* Find the first available interrupt */
+	for (index = 0; index < pdsc->nintrs; index++)
+		if (!pdsc->intr_info[index].vector)
+			break;
+	if (index >= pdsc->nintrs) {
+		dev_warn(pdsc->dev, "%s: no intr, index=%d nintrs=%d\n",
+			 __func__, index, pdsc->nintrs);
+		return -ENOSPC;
+	}
+
+	pds_core_intr_clean_flags(&pdsc->intr_ctrl[index],
+				  PDS_CORE_INTR_CRED_RESET_COALESCE);
+
+	intr_info = &pdsc->intr_info[index];
+
+	intr_info->index = index;
+	intr_info->data = data;
+	strscpy(intr_info->name, name, sizeof(intr_info->name));
+
+	/* Get the OS vector number for the interrupt */
+	err = pci_irq_vector(pdsc->pdev, index);
+	if (err < 0) {
+		dev_err(pdsc->dev, "failed to get intr vector index %d: %pe\n",
+			index, ERR_PTR(err));
+		goto err_out_free_intr;
+	}
+	intr_info->vector = err;
+
+	/* Init the device's intr mask */
+	pds_core_intr_clean(&pdsc->intr_ctrl[index]);
+	pds_core_intr_mask_assert(&pdsc->intr_ctrl[index], 1);
+	pds_core_intr_mask(&pdsc->intr_ctrl[index], PDS_CORE_INTR_MASK_SET);
+
+	/* Register the isr with a name */
+	err = request_irq(intr_info->vector, handler, 0, intr_info->name, data);
+	if (err) {
+		dev_err(pdsc->dev, "failed to get intr irq vector %d: %pe\n",
+			intr_info->vector, ERR_PTR(err));
+		goto err_out_free_intr;
+	}
+
+	return index;
+
+err_out_free_intr:
+	pdsc_intr_free(pdsc, index);
+	return err;
+}
+
+static void pdsc_qcq_intr_free(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	if (!(qcq->flags & PDS_CORE_QCQ_F_INTR) ||
+	    qcq->intx == PDS_CORE_INTR_INDEX_NOT_ASSIGNED)
+		return;
+
+	pdsc_intr_free(pdsc, qcq->intx);
+	qcq->intx = PDS_CORE_INTR_INDEX_NOT_ASSIGNED;
+}
+
+static int pdsc_qcq_intr_alloc(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	char name[PDSC_INTR_NAME_MAX_SZ];
+	int index;
+
+	if (!(qcq->flags & PDS_CORE_QCQ_F_INTR)) {
+		qcq->intx = PDS_CORE_INTR_INDEX_NOT_ASSIGNED;
+		return 0;
+	}
+
+	snprintf(name, sizeof(name), "%s-%d-%s",
+		 PDS_CORE_DRV_NAME, pdsc->pdev->bus->number, qcq->q.name);
+	index = pdsc_intr_alloc(pdsc, name, pdsc_adminq_isr, qcq);
+	if (index < 0)
+		return index;
+	qcq->intx = index;
+
+	return 0;
+}
+
+void pdsc_qcq_free(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	struct device *dev = pdsc->dev;
+
+	if (!(qcq && qcq->pdsc))
+		return;
+
+	pdsc_debugfs_del_qcq(qcq);
+
+	pdsc_qcq_intr_free(pdsc, qcq);
+
+	if (qcq->q_base)
+		dma_free_coherent(dev, qcq->q_size,
+				  qcq->q_base, qcq->q_base_pa);
+
+	if (qcq->cq_base)
+		dma_free_coherent(dev, qcq->cq_size,
+				  qcq->cq_base, qcq->cq_base_pa);
+
+	if (qcq->cq.info)
+		vfree(qcq->cq.info);
+
+	if (qcq->q.info)
+		vfree(qcq->q.info);
+
+	memset(qcq, 0, sizeof(*qcq));
+}
+
+static void pdsc_q_map(struct pdsc_queue *q, void *base, dma_addr_t base_pa)
+{
+	struct pdsc_q_info *cur;
+	unsigned int i;
+
+	q->base = base;
+	q->base_pa = base_pa;
+
+	for (i = 0, cur = q->info; i < q->num_descs; i++, cur++)
+		cur->desc = base + (i * q->desc_size);
+}
+
+static void pdsc_cq_map(struct pdsc_cq *cq, void *base, dma_addr_t base_pa)
+{
+	struct pdsc_cq_info *cur;
+	unsigned int i;
+
+	cq->base = base;
+	cq->base_pa = base_pa;
+
+	for (i = 0, cur = cq->info; i < cq->num_descs; i++, cur++)
+		cur->comp = base + (i * cq->desc_size);
+}
+
+int pdsc_qcq_alloc(struct pdsc *pdsc, unsigned int type, unsigned int index,
+		   const char *name, unsigned int flags, unsigned int num_descs,
+		   unsigned int desc_size, unsigned int cq_desc_size,
+		   unsigned int pid, struct pdsc_qcq *qcq)
+{
+	struct device *dev = pdsc->dev;
+	void *q_base, *cq_base;
+	dma_addr_t cq_base_pa;
+	dma_addr_t q_base_pa;
+	int err;
+
+	qcq->q.info = vzalloc(num_descs * sizeof(*qcq->q.info));
+	if (!qcq->q.info) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	qcq->pdsc = pdsc;
+	qcq->flags = flags;
+	INIT_WORK(&qcq->work, pdsc_work_thread);
+
+	qcq->q.type = type;
+	qcq->q.index = index;
+	qcq->q.num_descs = num_descs;
+	qcq->q.desc_size = desc_size;
+	qcq->q.tail_idx = 0;
+	qcq->q.head_idx = 0;
+	qcq->q.pid = pid;
+	snprintf(qcq->q.name, sizeof(qcq->q.name), "%s%u", name, index);
+
+	err = pdsc_qcq_intr_alloc(pdsc, qcq);
+	if (err)
+		goto err_out_free_q_info;
+
+	qcq->cq.info = vzalloc(num_descs * sizeof(*qcq->cq.info));
+	if (!qcq->cq.info) {
+		err = -ENOMEM;
+		goto err_out_free_irq;
+	}
+
+	qcq->cq.bound_intr = &pdsc->intr_info[qcq->intx];
+	qcq->cq.num_descs = num_descs;
+	qcq->cq.desc_size = cq_desc_size;
+	qcq->cq.tail_idx = 0;
+	qcq->cq.done_color = 1;
+
+	if (flags & PDS_CORE_QCQ_F_NOTIFYQ) {
+		/* q & cq need to be contiguous in case of notifyq */
+		qcq->q_size = PDS_PAGE_SIZE +
+			      ALIGN(num_descs * desc_size, PDS_PAGE_SIZE) +
+			      ALIGN(num_descs * cq_desc_size, PDS_PAGE_SIZE);
+		qcq->q_base = dma_alloc_coherent(dev,
+						 qcq->q_size + qcq->cq_size,
+						 &qcq->q_base_pa,
+						 GFP_KERNEL);
+		if (!qcq->q_base) {
+			err = -ENOMEM;
+			goto err_out_free_cq_info;
+		}
+		q_base = PTR_ALIGN(qcq->q_base, PDS_PAGE_SIZE);
+		q_base_pa = ALIGN(qcq->q_base_pa, PDS_PAGE_SIZE);
+		pdsc_q_map(&qcq->q, q_base, q_base_pa);
+
+		cq_base = PTR_ALIGN(q_base +
+				    ALIGN(num_descs * desc_size, PDS_PAGE_SIZE),
+				    PDS_PAGE_SIZE);
+		cq_base_pa = ALIGN(qcq->q_base_pa +
+				   ALIGN(num_descs * desc_size, PDS_PAGE_SIZE),
+				   PDS_PAGE_SIZE);
+
+	} else {
+		/* q DMA descriptors */
+		qcq->q_size = PDS_PAGE_SIZE + (num_descs * desc_size);
+		qcq->q_base = dma_alloc_coherent(dev, qcq->q_size,
+						 &qcq->q_base_pa,
+						 GFP_KERNEL);
+		if (!qcq->q_base) {
+			err = -ENOMEM;
+			goto err_out_free_cq_info;
+		}
+		q_base = PTR_ALIGN(qcq->q_base, PDS_PAGE_SIZE);
+		q_base_pa = ALIGN(qcq->q_base_pa, PDS_PAGE_SIZE);
+		pdsc_q_map(&qcq->q, q_base, q_base_pa);
+
+		/* cq DMA descriptors */
+		qcq->cq_size = PDS_PAGE_SIZE + (num_descs * cq_desc_size);
+		qcq->cq_base = dma_alloc_coherent(dev, qcq->cq_size,
+						  &qcq->cq_base_pa,
+						  GFP_KERNEL);
+		if (!qcq->cq_base) {
+			err = -ENOMEM;
+			goto err_out_free_q;
+		}
+		cq_base = PTR_ALIGN(qcq->cq_base, PDS_PAGE_SIZE);
+		cq_base_pa = ALIGN(qcq->cq_base_pa, PDS_PAGE_SIZE);
+	}
+
+	pdsc_cq_map(&qcq->cq, cq_base, cq_base_pa);
+	qcq->cq.bound_q = &qcq->q;
+
+	pdsc_debugfs_add_qcq(pdsc, qcq);
+
+	return 0;
+
+err_out_free_q:
+	dma_free_coherent(dev, qcq->q_size, qcq->q_base, qcq->q_base_pa);
+err_out_free_cq_info:
+	vfree(qcq->cq.info);
+err_out_free_irq:
+	pdsc_qcq_intr_free(pdsc, qcq);
+err_out_free_q_info:
+	vfree(qcq->q.info);
+	memset(qcq, 0, sizeof(*qcq));
+err_out:
+	dev_err(dev, "qcq alloc of %s%d failed %d\n", name, index, err);
+	return err;
+}
+
+static int pdsc_core_init(struct pdsc *pdsc)
+{
+	union pds_core_dev_comp comp = {};
+	union pds_core_dev_cmd cmd = {
+		.init.opcode = PDS_CORE_CMD_INIT,
+	};
+	struct pds_core_dev_init_data_out cido;
+	struct pds_core_dev_init_data_in cidi;
+	u32 dbid_count;
+	u32 dbpage_num;
+	size_t sz;
+	int err;
+
+	cidi.adminq_q_base = cpu_to_le64(pdsc->adminqcq.q_base_pa);
+	cidi.adminq_cq_base = cpu_to_le64(pdsc->adminqcq.cq_base_pa);
+	cidi.notifyq_cq_base = cpu_to_le64(pdsc->notifyqcq.cq.base_pa);
+	cidi.flags = cpu_to_le32(PDS_CORE_QINIT_F_IRQ | PDS_CORE_QINIT_F_ENA);
+	cidi.intr_index = cpu_to_le16(pdsc->adminqcq.intx);
+	cidi.adminq_ring_size = ilog2(pdsc->adminqcq.q.num_descs);
+	cidi.notifyq_ring_size = ilog2(pdsc->notifyqcq.q.num_descs);
+
+	mutex_lock(&pdsc->devcmd_lock);
+
+	sz = min_t(size_t, sizeof(cidi), sizeof(pdsc->cmd_regs->data));
+	memcpy_toio(&pdsc->cmd_regs->data, &cidi, sz);
+
+	err = pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+	if (!err) {
+		sz = min_t(size_t, sizeof(cido), sizeof(pdsc->cmd_regs->data));
+		memcpy_fromio(&cido, &pdsc->cmd_regs->data, sz);
+	}
+
+	mutex_unlock(&pdsc->devcmd_lock);
+	if (err) {
+		dev_err(pdsc->dev, "Device init command failed: %pe\n",
+			ERR_PTR(err));
+		return err;
+	}
+
+	pdsc->hw_index = le32_to_cpu(cido.core_hw_index);
+
+	dbid_count = le32_to_cpu(pdsc->dev_ident.ndbpgs_per_lif);
+	dbpage_num = pdsc->hw_index * dbid_count;
+	pdsc->kern_dbpage = pdsc_map_dbpage(pdsc, dbpage_num);
+	if (!pdsc->kern_dbpage) {
+		dev_err(pdsc->dev, "Cannot map dbpage, aborting\n");
+		return -ENOMEM;
+	}
+
+	pdsc->adminqcq.q.hw_type = cido.adminq_hw_type;
+	pdsc->adminqcq.q.hw_index = le32_to_cpu(cido.adminq_hw_index);
+	pdsc->adminqcq.q.dbval = PDS_CORE_DBELL_QID(pdsc->adminqcq.q.hw_index);
+
+	pdsc->notifyqcq.q.hw_type = cido.notifyq_hw_type;
+	pdsc->notifyqcq.q.hw_index = le32_to_cpu(cido.notifyq_hw_index);
+	pdsc->notifyqcq.q.dbval = PDS_CORE_DBELL_QID(pdsc->notifyqcq.q.hw_index);
+
+	pdsc->last_eid = 0;
+
+	return err;
+}
+
 int pdsc_setup(struct pdsc *pdsc, bool init)
 {
+	int numdescs;
 	int err;
 
 	if (init)
@@ -14,17 +369,60 @@ int pdsc_setup(struct pdsc *pdsc, bool init)
 	if (err)
 		return err;
 
+	/* Scale the descriptor ring length based on number of CPUs and VFs */
+	numdescs = max_t(int, PDSC_ADMINQ_MIN_LENGTH, num_online_cpus());
+	numdescs += 2 * pci_sriov_get_totalvfs(pdsc->pdev);
+	numdescs = roundup_pow_of_two(numdescs);
+	err = pdsc_qcq_alloc(pdsc, PDS_CORE_QTYPE_ADMINQ, 0, "adminq",
+			     PDS_CORE_QCQ_F_CORE | PDS_CORE_QCQ_F_INTR,
+			     numdescs,
+			     sizeof(union pds_core_adminq_cmd),
+			     sizeof(union pds_core_adminq_comp),
+			     0, &pdsc->adminqcq);
+	if (err)
+		goto err_out_teardown;
+
+	err = pdsc_qcq_alloc(pdsc, PDS_CORE_QTYPE_NOTIFYQ, 0, "notifyq",
+			     PDS_CORE_QCQ_F_NOTIFYQ,
+			     PDSC_NOTIFYQ_LENGTH,
+			     sizeof(struct pds_core_notifyq_cmd),
+			     sizeof(union pds_core_notifyq_comp),
+			     0, &pdsc->notifyqcq);
+	if (err)
+		goto err_out_teardown;
+
+	/* NotifyQ rides on the AdminQ interrupt */
+	pdsc->notifyqcq.intx = pdsc->adminqcq.intx;
+
+	/* Set up the Core with the AdminQ and NotifyQ info */
+	err = pdsc_core_init(pdsc);
+	if (err)
+		goto err_out_teardown;
+
 	clear_bit(PDSC_S_FW_DEAD, &pdsc->state);
 	return 0;
+
+err_out_teardown:
+	pdsc_teardown(pdsc, init);
+	return err;
 }
 
 void pdsc_teardown(struct pdsc *pdsc, bool removing)
 {
+	int i;
+
 	pdsc_devcmd_reset(pdsc);
+	pdsc_qcq_free(pdsc, &pdsc->notifyqcq);
+	pdsc_qcq_free(pdsc, &pdsc->adminqcq);
+
+	if (pdsc->intr_info) {
+		for (i = 0; i < pdsc->nintrs; i++)
+			pdsc_intr_free(pdsc, i);
 
-	if (removing) {
-		kfree(pdsc->intr_info);
-		pdsc->intr_info = NULL;
+		if (removing) {
+			kfree(pdsc->intr_info);
+			pdsc->intr_info = NULL;
+		}
 	}
 
 	if (pdsc->kern_dbpage) {
@@ -35,6 +433,28 @@ void pdsc_teardown(struct pdsc *pdsc, bool removing)
 	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
 }
 
+int pdsc_start(struct pdsc *pdsc)
+{
+	pds_core_intr_mask(&pdsc->intr_ctrl[pdsc->adminqcq.intx],
+			   PDS_CORE_INTR_MASK_CLEAR);
+
+	return 0;
+}
+
+void pdsc_stop(struct pdsc *pdsc)
+{
+	int i;
+
+	if (!pdsc->intr_info)
+		return;
+
+	/* Mask interrupts that are in use */
+	for (i = 0; i < pdsc->nintrs; i++)
+		if (pdsc->intr_info[i].vector)
+			pds_core_intr_mask(&pdsc->intr_ctrl[i],
+					   PDS_CORE_INTR_MASK_SET);
+}
+
 static void pdsc_fw_down(struct pdsc *pdsc)
 {
 	if (test_and_set_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
@@ -44,6 +464,7 @@ static void pdsc_fw_down(struct pdsc *pdsc)
 
 	devlink_health_report(pdsc->fw_reporter, "FW down reported", pdsc);
 
+	pdsc_stop(pdsc);
 	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
 }
 
@@ -60,6 +481,10 @@ static void pdsc_fw_up(struct pdsc *pdsc)
 	if (err)
 		goto err_out;
 
+	err = pdsc_start(pdsc);
+	if (err)
+		goto err_out;
+
 	pdsc->fw_recoveries++;
 	devlink_health_reporter_state_update(pdsc->fw_reporter,
 					     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 32aa38c40024..7adbd10cb6d9 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -9,11 +9,15 @@
 
 #include <linux/pds/pds_common.h>
 #include <linux/pds/pds_core_if.h>
+#include <linux/pds/pds_adminq.h>
 #include <linux/pds/pds_intr.h>
 
 #define PDSC_DRV_DESCRIPTION	"AMD/Pensando Core Driver"
 
 #define PDSC_WATCHDOG_SECS	5
+#define PDSC_QUEUE_NAME_MAX_SZ  32
+#define PDSC_ADMINQ_MIN_LENGTH	16	/* must be a power of two */
+#define PDSC_NOTIFYQ_LENGTH	64	/* must be a power of two */
 #define PDSC_TEARDOWN_RECOVERY	false
 #define PDSC_TEARDOWN_REMOVING	true
 #define PDSC_SETUP_RECOVERY	false
@@ -33,6 +37,28 @@ struct pdsc_devinfo {
 	char serial_num[PDS_CORE_DEVINFO_SERIAL_BUFLEN + 1];
 };
 
+struct pdsc_queue {
+	struct pdsc_q_info *info;
+	u64 dbval;
+	u16 head_idx;
+	u16 tail_idx;
+	u8 hw_type;
+	unsigned int index;
+	unsigned int num_descs;
+	u64 dbell_count;
+	u64 features;
+	unsigned int type;
+	unsigned int hw_index;
+	union {
+		void *base;
+		struct pds_core_admin_cmd *adminq;
+	};
+	dma_addr_t base_pa;	/* must be page aligned */
+	unsigned int desc_size;
+	unsigned int pid;
+	char name[PDSC_QUEUE_NAME_MAX_SZ];
+};
+
 #define PDSC_INTR_NAME_MAX_SZ		32
 
 struct pdsc_intr_info {
@@ -42,6 +68,61 @@ struct pdsc_intr_info {
 	void *data;
 };
 
+struct pdsc_cq_info {
+	void *comp;
+};
+
+struct pdsc_buf_info {
+	struct page *page;
+	dma_addr_t dma_addr;
+	u32 page_offset;
+	u32 len;
+};
+
+struct pdsc_q_info {
+	union {
+		void *desc;
+		struct pdsc_admin_cmd *adminq_desc;
+	};
+	unsigned int bytes;
+	unsigned int nbufs;
+	struct pdsc_buf_info bufs[PDS_CORE_MAX_FRAGS];
+	struct pdsc_wait_context *wc;
+	void *dest;
+};
+
+struct pdsc_cq {
+	struct pdsc_cq_info *info;
+	struct pdsc_queue *bound_q;
+	struct pdsc_intr_info *bound_intr;
+	u16 tail_idx;
+	bool done_color;
+	unsigned int num_descs;
+	unsigned int desc_size;
+	void *base;
+	dma_addr_t base_pa;	/* must be page aligned */
+} ____cacheline_aligned_in_smp;
+
+struct pdsc_qcq {
+	struct pdsc *pdsc;
+	void *q_base;
+	dma_addr_t q_base_pa;	/* might not be page aligned */
+	void *cq_base;
+	dma_addr_t cq_base_pa;	/* might not be page aligned */
+	u32 q_size;
+	u32 cq_size;
+	bool armed;
+	unsigned int flags;
+
+	struct work_struct work;
+	struct pdsc_queue q;
+	struct pdsc_cq cq;
+	int intx;
+
+	u32 accum_work;
+	struct dentry *dentry;
+};
+
 /* No state flags set means we are in a steady running state */
 enum pdsc_state_flags {
 	PDSC_S_FW_DEAD,		    /* stopped, wait on startup or recovery */
@@ -81,6 +162,7 @@ struct pdsc {
 	unsigned int devcmd_timeout;
 	struct mutex devcmd_lock;	/* lock for dev_cmd operations */
 	struct mutex config_lock;	/* lock for configuration operations */
+	spinlock_t adminq_lock;		/* lock for adminq operations */
 	struct pds_core_dev_info_regs __iomem *info_regs;
 	struct pds_core_dev_cmd_regs __iomem *cmd_regs;
 	struct pds_core_intr __iomem *intr_ctrl;
@@ -88,11 +170,64 @@ struct pdsc {
 	u64 __iomem *db_pages;
 	dma_addr_t phy_db_pages;
 	u64 __iomem *kern_dbpage;
+
+	struct pdsc_qcq adminqcq;
+	struct pdsc_qcq notifyqcq;
+	u64 last_eid;
 };
 
+/** enum pds_core_dbell_bits - bitwise composition of dbell values.
+ *
+ * @PDS_CORE_DBELL_QID_MASK:	unshifted mask of valid queue id bits.
+ * @PDS_CORE_DBELL_QID_SHIFT:	queue id shift amount in dbell value.
+ * @PDS_CORE_DBELL_QID:		macro to build QID component of dbell value.
+ *
+ * @PDS_CORE_DBELL_RING_MASK:	unshifted mask of valid ring bits.
+ * @PDS_CORE_DBELL_RING_SHIFT:	ring shift amount in dbell value.
+ * @PDS_CORE_DBELL_RING:	macro to build ring component of dbell value.
+ *
+ * @PDS_CORE_DBELL_RING_0:	ring zero dbell component value.
+ * @PDS_CORE_DBELL_RING_1:	ring one dbell component value.
+ * @PDS_CORE_DBELL_RING_2:	ring two dbell component value.
+ * @PDS_CORE_DBELL_RING_3:	ring three dbell component value.
+ *
+ * @PDS_CORE_DBELL_INDEX_MASK:	bit mask of valid index bits, no shift needed.
+ */
+enum pds_core_dbell_bits {
+	PDS_CORE_DBELL_QID_MASK		= 0xffffff,
+	PDS_CORE_DBELL_QID_SHIFT		= 24,
+
+#define PDS_CORE_DBELL_QID(n) \
+	(((u64)(n) & PDS_CORE_DBELL_QID_MASK) << PDS_CORE_DBELL_QID_SHIFT)
+
+	PDS_CORE_DBELL_RING_MASK		= 0x7,
+	PDS_CORE_DBELL_RING_SHIFT		= 16,
+
+#define PDS_CORE_DBELL_RING(n) \
+	(((u64)(n) & PDS_CORE_DBELL_RING_MASK) << PDS_CORE_DBELL_RING_SHIFT)
+
+	PDS_CORE_DBELL_RING_0		= 0,
+	PDS_CORE_DBELL_RING_1		= PDS_CORE_DBELL_RING(1),
+	PDS_CORE_DBELL_RING_2		= PDS_CORE_DBELL_RING(2),
+	PDS_CORE_DBELL_RING_3		= PDS_CORE_DBELL_RING(3),
+
+	PDS_CORE_DBELL_INDEX_MASK		= 0xffff,
+};
+
+static inline void pds_core_dbell_ring(u64 __iomem *db_page,
+				       enum pds_core_logical_qtype qtype,
+				       u64 val)
+{
+	writeq(val, &db_page[qtype]);
+}
+
 int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
 			      struct devlink_fmsg *fmsg,
 			      struct netlink_ext_ack *extack);
+int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+		     struct netlink_ext_ack *extack);
+
+void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num);
 
 void pdsc_debugfs_create(void);
 void pdsc_debugfs_destroy(void);
@@ -100,6 +235,8 @@ void pdsc_debugfs_add_dev(struct pdsc *pdsc);
 void pdsc_debugfs_del_dev(struct pdsc *pdsc);
 void pdsc_debugfs_add_ident(struct pdsc *pdsc);
 void pdsc_debugfs_add_irqs(struct pdsc *pdsc);
+void pdsc_debugfs_add_qcq(struct pdsc *pdsc, struct pdsc_qcq *qcq);
+void pdsc_debugfs_del_qcq(struct pdsc_qcq *qcq);
 
 int pdsc_err_to_errno(enum pds_core_status_code code);
 bool pdsc_is_fw_running(struct pdsc *pdsc);
@@ -113,8 +250,22 @@ int pdsc_devcmd_reset(struct pdsc *pdsc);
 int pdsc_dev_reinit(struct pdsc *pdsc);
 int pdsc_dev_init(struct pdsc *pdsc);
 
+int pdsc_intr_alloc(struct pdsc *pdsc, char *name,
+		    irq_handler_t handler, void *data);
+void pdsc_intr_free(struct pdsc *pdsc, int index);
+void pdsc_qcq_free(struct pdsc *pdsc, struct pdsc_qcq *qcq);
+int pdsc_qcq_alloc(struct pdsc *pdsc, unsigned int type, unsigned int index,
+		   const char *name, unsigned int flags, unsigned int num_descs,
+		   unsigned int desc_size, unsigned int cq_desc_size,
+		   unsigned int pid, struct pdsc_qcq *qcq);
 int pdsc_setup(struct pdsc *pdsc, bool init);
 void pdsc_teardown(struct pdsc *pdsc, bool removing);
+int pdsc_start(struct pdsc *pdsc);
+void pdsc_stop(struct pdsc *pdsc);
 void pdsc_health_thread(struct work_struct *work);
 
+void pdsc_process_adminq(struct pdsc_qcq *qcq);
+void pdsc_work_thread(struct work_struct *work);
+irqreturn_t pdsc_adminq_isr(int irq, void *data);
+
 #endif /* _PDSC_H_ */
diff --git a/drivers/net/ethernet/amd/pds_core/debugfs.c b/drivers/net/ethernet/amd/pds_core/debugfs.c
index 601431b41abb..b83e5016644b 100644
--- a/drivers/net/ethernet/amd/pds_core/debugfs.c
+++ b/drivers/net/ethernet/amd/pds_core/debugfs.c
@@ -67,3 +67,80 @@ void pdsc_debugfs_add_ident(struct pdsc *pdsc)
 	debugfs_create_file("identity", 0400, pdsc->dentry,
 			    pdsc, &identity_fops);
 }
+
+static const struct debugfs_reg32 intr_ctrl_regs[] = {
+	{ .name = "coal_init", .offset = 0, },
+	{ .name = "mask", .offset = 4, },
+	{ .name = "credits", .offset = 8, },
+	{ .name = "mask_on_assert", .offset = 12, },
+	{ .name = "coal_timer", .offset = 16, },
+};
+
+void pdsc_debugfs_add_qcq(struct pdsc *pdsc, struct pdsc_qcq *qcq)
+{
+	struct dentry *qcq_dentry, *q_dentry, *cq_dentry;
+	struct dentry *intr_dentry;
+	struct debugfs_regset32 *intr_ctrl_regset;
+	struct pdsc_intr_info *intr = &pdsc->intr_info[qcq->intx];
+	struct pdsc_queue *q = &qcq->q;
+	struct pdsc_cq *cq = &qcq->cq;
+
+	qcq_dentry = debugfs_create_dir(q->name, pdsc->dentry);
+	if (IS_ERR_OR_NULL(qcq_dentry))
+		return;
+	qcq->dentry = qcq_dentry;
+
+	debugfs_create_x64("q_base_pa", 0400, qcq_dentry, &qcq->q_base_pa);
+	debugfs_create_x32("q_size", 0400, qcq_dentry, &qcq->q_size);
+	debugfs_create_x64("cq_base_pa", 0400, qcq_dentry, &qcq->cq_base_pa);
+	debugfs_create_x32("cq_size", 0400, qcq_dentry, &qcq->cq_size);
+	debugfs_create_x32("accum_work", 0400, qcq_dentry, &qcq->accum_work);
+
+	q_dentry = debugfs_create_dir("q", qcq->dentry);
+	if (IS_ERR_OR_NULL(q_dentry))
+		return;
+
+	debugfs_create_u32("index", 0400, q_dentry, &q->index);
+	debugfs_create_u32("num_descs", 0400, q_dentry, &q->num_descs);
+	debugfs_create_u32("desc_size", 0400, q_dentry, &q->desc_size);
+	debugfs_create_u32("pid", 0400, q_dentry, &q->pid);
+
+	debugfs_create_u16("tail", 0400, q_dentry, &q->tail_idx);
+	debugfs_create_u16("head", 0400, q_dentry, &q->head_idx);
+
+	cq_dentry = debugfs_create_dir("cq", qcq->dentry);
+	if (IS_ERR_OR_NULL(cq_dentry))
+		return;
+
+	debugfs_create_x64("base_pa", 0400, cq_dentry, &cq->base_pa);
+	debugfs_create_u32("num_descs", 0400, cq_dentry, &cq->num_descs);
+	debugfs_create_u32("desc_size", 0400, cq_dentry, &cq->desc_size);
+	debugfs_create_bool("done_color", 0400, cq_dentry, &cq->done_color);
+	debugfs_create_u16("tail", 0400, cq_dentry, &cq->tail_idx);
+
+	if (qcq->flags & PDS_CORE_QCQ_F_INTR) {
+		intr_dentry = debugfs_create_dir("intr", qcq->dentry);
+		if (IS_ERR_OR_NULL(intr_dentry))
+			return;
+
+		debugfs_create_u32("index", 0400, intr_dentry, &intr->index);
+		debugfs_create_u32("vector", 0400, intr_dentry, &intr->vector);
+
+		intr_ctrl_regset = kzalloc(sizeof(*intr_ctrl_regset),
+					   GFP_KERNEL);
+		if (!intr_ctrl_regset)
+			return;
+		intr_ctrl_regset->regs = intr_ctrl_regs;
+		intr_ctrl_regset->nregs = ARRAY_SIZE(intr_ctrl_regs);
+		intr_ctrl_regset->base = &pdsc->intr_ctrl[intr->index];
+
+		debugfs_create_regset32("intr_ctrl", 0400, intr_dentry,
+					intr_ctrl_regset);
+	}
+};
+
+void pdsc_debugfs_del_qcq(struct pdsc_qcq *qcq)
+{
+	debugfs_remove_recursive(qcq->dentry);
+	qcq->dentry = NULL;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c
index 3b05b1af65d1..89f23d620d1d 100644
--- a/drivers/net/ethernet/amd/pds_core/devlink.c
+++ b/drivers/net/ethernet/amd/pds_core/devlink.c
@@ -3,6 +3,67 @@
 
 #include "core.h"
 
+static char *fw_slotnames[] = {
+	"fw.goldfw",
+	"fw.mainfwa",
+	"fw.mainfwb",
+};
+
+int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
+		     struct netlink_ext_ack *extack)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = PDS_CORE_FW_GET_LIST,
+	};
+	struct pds_core_fw_list_info fw_list;
+	struct pdsc *pdsc = devlink_priv(dl);
+	union pds_core_dev_comp comp;
+	char buf[16];
+	int listlen;
+	int err;
+	int i;
+
+	mutex_lock(&pdsc->devcmd_lock);
+	err = pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout * 2);
+	memcpy_fromio(&fw_list, pdsc->cmd_regs->data, sizeof(fw_list));
+	mutex_unlock(&pdsc->devcmd_lock);
+	if (err && err != -EIO)
+		return err;
+
+	listlen = fw_list.num_fw_slots;
+	for (i = 0; i < listlen; i++) {
+		if (i < ARRAY_SIZE(fw_slotnames))
+			strscpy(buf, fw_slotnames[i], sizeof(buf));
+		else
+			snprintf(buf, sizeof(buf), "fw.slot_%d", i);
+		err = devlink_info_version_stored_put(req, buf,
+						      fw_list.fw_names[i].fw_version);
+	}
+
+	err = devlink_info_version_running_put(req,
+					       DEVLINK_INFO_VERSION_GENERIC_FW,
+					       pdsc->dev_info.fw_version);
+	if (err)
+		return err;
+
+	snprintf(buf, sizeof(buf), "0x%x", pdsc->dev_info.asic_type);
+	err = devlink_info_version_fixed_put(req,
+					     DEVLINK_INFO_VERSION_GENERIC_ASIC_ID,
+					     buf);
+	if (err)
+		return err;
+
+	snprintf(buf, sizeof(buf), "0x%x", pdsc->dev_info.asic_rev);
+	err = devlink_info_version_fixed_put(req,
+					     DEVLINK_INFO_VERSION_GENERIC_ASIC_REV,
+					     buf);
+	if (err)
+		return err;
+
+	return devlink_info_serial_number_put(req, pdsc->dev_info.serial_num);
+}
+
 int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
 			      struct devlink_fmsg *fmsg,
 			      struct netlink_ext_ack *extack)
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index 54f3aed7adb1..eaff311d4a10 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -125,6 +125,13 @@ err_out:
 	return err;
 }
 
+void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num)
+{
+	return pci_iomap_range(pdsc->pdev,
+			       pdsc->bars[PDS_CORE_PCI_BAR_DBELL].res_index,
+			       (u64)page_num << PAGE_SHIFT, PAGE_SIZE);
+}
+
 static int pdsc_init_vf(struct pdsc *vf)
 {
 	return -1;
@@ -166,6 +173,7 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	mutex_init(&pdsc->devcmd_lock);
 	mutex_init(&pdsc->config_lock);
+	spin_lock_init(&pdsc->adminq_lock);
 
 	mutex_lock(&pdsc->config_lock);
 	set_bit(PDSC_S_FW_DEAD, &pdsc->state);
@@ -173,6 +181,9 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 	err = pdsc_setup(pdsc, PDSC_SETUP_INIT);
 	if (err)
 		goto err_out_unmap_bars;
+	err = pdsc_start(pdsc);
+	if (err)
+		goto err_out_teardown;
 
 	mutex_unlock(&pdsc->config_lock);
 
@@ -184,7 +195,7 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 		dev_warn(pdsc->dev, "Failed to create fw reporter: %pe\n", hr);
 		err = PTR_ERR(hr);
 		devl_unlock(dl);
-		goto err_out_teardown;
+		goto err_out_stop;
 	}
 	pdsc->fw_reporter = hr;
 
@@ -196,6 +207,8 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	return 0;
 
+err_out_stop:
+	pdsc_stop(pdsc);
 err_out_teardown:
 	pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
 err_out_unmap_bars:
@@ -214,6 +227,7 @@ err_out_release_regions:
 }
 
 static const struct devlink_ops pdsc_dl_ops = {
+	.info_get	= pdsc_dl_info_get,
 };
 
 static const struct devlink_ops pdsc_dl_vf_ops = {
@@ -315,6 +329,7 @@ static void pdsc_remove(struct pci_dev *pdev)
 		mutex_lock(&pdsc->config_lock);
 		set_bit(PDSC_S_STOPPING_DRIVER, &pdsc->state);
 
+		pdsc_stop(pdsc);
 		pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
 		mutex_unlock(&pdsc->config_lock);
 		mutex_destroy(&pdsc->config_lock);
diff --git a/include/linux/pds/pds_adminq.h b/include/linux/pds/pds_adminq.h
new file mode 100644
index 000000000000..dd5fbe3ee141
--- /dev/null
+++ b/include/linux/pds/pds_adminq.h
@@ -0,0 +1,638 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#ifndef _PDS_CORE_ADMINQ_H_
+#define _PDS_CORE_ADMINQ_H_
+
+enum pds_core_adminq_flags {
+	PDS_AQ_FLAG_FASTPOLL	= BIT(1),	/* completion poll at 1ms */
+};
+
+/*
+ * enum pds_core_adminq_opcode - AdminQ command opcodes
+ * These commands are only processed on AdminQ, not available in devcmd
+ */
+enum pds_core_adminq_opcode {
+	PDS_AQ_CMD_NOP			= 0,
+
+	/* Client control */
+	PDS_AQ_CMD_CLIENT_REG		= 6,
+	PDS_AQ_CMD_CLIENT_UNREG		= 7,
+	PDS_AQ_CMD_CLIENT_CMD		= 8,
+
+	/* LIF commands */
+	PDS_AQ_CMD_LIF_IDENTIFY		= 20,
+	PDS_AQ_CMD_LIF_INIT		= 21,
+	PDS_AQ_CMD_LIF_RESET		= 22,
+	PDS_AQ_CMD_LIF_GETATTR		= 23,
+	PDS_AQ_CMD_LIF_SETATTR		= 24,
+	PDS_AQ_CMD_LIF_SETPHC		= 25,
+
+	PDS_AQ_CMD_RX_MODE_SET		= 30,
+	PDS_AQ_CMD_RX_FILTER_ADD	= 31,
+	PDS_AQ_CMD_RX_FILTER_DEL	= 32,
+
+	/* Queue commands */
+	PDS_AQ_CMD_Q_IDENTIFY		= 39,
+	PDS_AQ_CMD_Q_INIT		= 40,
+	PDS_AQ_CMD_Q_CONTROL		= 41,
+
+	/* SR/IOV commands */
+	PDS_AQ_CMD_VF_GETATTR		= 60,
+	PDS_AQ_CMD_VF_SETATTR		= 61,
+};
+
+/*
+ * enum pds_core_notifyq_opcode - NotifyQ event codes
+ */
+enum pds_core_notifyq_opcode {
+	PDS_EVENT_LINK_CHANGE		= 1,
+	PDS_EVENT_RESET			= 2,
+	PDS_EVENT_XCVR			= 5,
+	PDS_EVENT_CLIENT		= 6,
+};
+
+#define PDS_COMP_COLOR_MASK  0x80
+
+/**
+ * struct pds_core_notifyq_event - Generic event reporting structure
+ * @eid:   event number
+ * @ecode: event code
+ *
+ * This is the generic event report struct from which the other
+ * actual events will be formed.
+ */
+struct pds_core_notifyq_event {
+	__le64 eid;
+	__le16 ecode;
+};
+
+/**
+ * struct pds_core_link_change_event - Link change event notification
+ * @eid:		event number
+ * @ecode:		event code = PDS_EVENT_LINK_CHANGE
+ * @link_status:	link up/down, with error bits
+ * @link_speed:		speed of the network link
+ *
+ * Sent when the network link state changes between UP and DOWN
+ */
+struct pds_core_link_change_event {
+	__le64 eid;
+	__le16 ecode;
+	__le16 link_status;
+	__le32 link_speed;	/* units of 1Mbps: e.g. 10000 = 10Gbps */
+};
+
+/**
+ * struct pds_core_reset_event - Reset event notification
+ * @eid:		event number
+ * @ecode:		event code = PDS_EVENT_RESET
+ * @reset_code:		reset type
+ * @state:		0=pending, 1=complete, 2=error
+ *
+ * Sent when the NIC or some subsystem is going to be or
+ * has been reset.
+ */
+struct pds_core_reset_event {
+	__le64 eid;
+	__le16 ecode;
+	u8     reset_code;
+	u8     state;
+};
+
+/**
+ * struct pds_core_client_event - Client event notification
+ * @eid:		event number
+ * @ecode:		event code = PDS_EVENT_CLIENT
+ * @client_id:          client to sent event to
+ * @client_event:       wrapped event struct for the client
+ *
+ * Sent when an event needs to be passed on to a client
+ */
+struct pds_core_client_event {
+	__le64 eid;
+	__le16 ecode;
+	__le16 client_id;
+	u8     client_event[54];
+};
+
+/**
+ * struct pds_core_notifyq_cmd - Placeholder for building qcq
+ * @data:      anonymous field for building the qcq
+ */
+struct pds_core_notifyq_cmd {
+	__le32 data;	/* Not used but needed for qcq structure */
+};
+
+/*
+ * union pds_core_notifyq_comp - Overlay of notifyq event structures
+ */
+union pds_core_notifyq_comp {
+	struct {
+		__le64 eid;
+		__le16 ecode;
+	};
+	struct pds_core_notifyq_event     event;
+	struct pds_core_link_change_event link_change;
+	struct pds_core_reset_event       reset;
+	u8     data[64];
+};
+
+#define PDS_DEVNAME_LEN		32
+/**
+ * struct pds_core_client_reg_cmd - Register a new client with DSC
+ * @opcode:         opcode PDS_AQ_CMD_CLIENT_REG
+ * @rsvd:           word boundary padding
+ * @devname:        text name of client device
+ * @vif_type:       what type of device (enum pds_core_vif_types)
+ *
+ * Tell the DSC of the new client, and receive a client_id from DSC.
+ */
+struct pds_core_client_reg_cmd {
+	u8     opcode;
+	u8     rsvd[3];
+	char   devname[PDS_DEVNAME_LEN];
+	u8     vif_type;
+};
+
+/**
+ * struct pds_core_client_reg_comp - Client registration completion
+ * @status:     Status of the command (enum pdc_core_status_code)
+ * @rsvd:       Word boundary padding
+ * @comp_index: Index in the descriptor ring for which this is the completion
+ * @client_id:  New id assigned by DSC
+ * @rsvd1:      Word boundary padding
+ * @color:      Color bit
+ */
+struct pds_core_client_reg_comp {
+	u8     status;
+	u8     rsvd;
+	__le16 comp_index;
+	__le16 client_id;
+	u8     rsvd1[9];
+	u8     color;
+};
+
+/**
+ * struct pds_core_client_unreg_cmd - Unregister a client from DSC
+ * @opcode:     opcode PDS_AQ_CMD_CLIENT_UNREG
+ * @rsvd:       word boundary padding
+ * @client_id:  id of client being removed
+ *
+ * Tell the DSC this client is going away and remove its context
+ * This uses the generic completion.
+ */
+struct pds_core_client_unreg_cmd {
+	u8     opcode;
+	u8     rsvd;
+	__le16 client_id;
+};
+
+/**
+ * struct pds_core_client_request_cmd - Pass along a wrapped client AdminQ cmd
+ * @opcode:     opcode PDS_AQ_CMD_CLIENT_CMD
+ * @rsvd:       word boundary padding
+ * @client_id:  id of client being removed
+ * @client_cmd: the wrapped client command
+ *
+ * Proxy post an adminq command for the client.
+ * This uses the generic completion.
+ */
+struct pds_core_client_request_cmd {
+	u8     opcode;
+	u8     rsvd;
+	__le16 client_id;
+	u8     client_cmd[60];
+};
+
+#define PDS_CORE_MAX_FRAGS		16
+
+#define PDS_CORE_QCQ_F_INITED		BIT(0)
+#define PDS_CORE_QCQ_F_SG		BIT(1)
+#define PDS_CORE_QCQ_F_INTR		BIT(2)
+#define PDS_CORE_QCQ_F_TX_STATS		BIT(3)
+#define PDS_CORE_QCQ_F_RX_STATS		BIT(4)
+#define PDS_CORE_QCQ_F_NOTIFYQ		BIT(5)
+#define PDS_CORE_QCQ_F_CMB_RINGS	BIT(6)
+#define PDS_CORE_QCQ_F_CORE		BIT(7)
+
+enum pds_core_lif_type {
+	PDS_CORE_LIF_TYPE_DEFAULT = 0,
+};
+
+/**
+ * union pds_core_lif_config - LIF configuration
+ * @state:	    LIF state (enum pds_core_lif_state)
+ * @rsvd:           Word boundary padding
+ * @name:	    LIF name
+ * @rsvd2:          Word boundary padding
+ * @features:	    LIF features active (enum pds_core_hw_features)
+ * @queue_count:    Queue counts per queue-type
+ * @words:          Full union buffer size
+ */
+union pds_core_lif_config {
+	struct {
+		u8     state;
+		u8     rsvd[3];
+		char   name[PDS_CORE_IFNAMSIZ];
+		u8     rsvd2[12];
+		__le64 features;
+		__le32 queue_count[PDS_CORE_QTYPE_MAX];
+	} __packed;
+	__le32 words[64];
+};
+
+/**
+ * struct pds_core_lif_status - LIF status register
+ * @eid:	     most recent NotifyQ event id
+ * @rsvd:            full struct size
+ */
+struct pds_core_lif_status {
+	__le64 eid;
+	u8     rsvd[56];
+};
+
+/**
+ * struct pds_core_lif_info - LIF info structure
+ * @config:	LIF configuration structure
+ * @status:	LIF status structure
+ */
+struct pds_core_lif_info {
+	union pds_core_lif_config config;
+	struct pds_core_lif_status status;
+};
+
+/**
+ * struct pds_core_lif_identity - LIF identity information (type-specific)
+ * @features:		LIF features (see enum pds_core_hw_features)
+ * @version:		Identify structure version
+ * @hw_index:		LIF hardware index
+ * @rsvd:		Word boundary padding
+ * @max_nb_sessions:	Maximum number of sessions supported
+ * @rsvd2:		buffer padding
+ * @config:		LIF config struct with features, q counts
+ */
+struct pds_core_lif_identity {
+	__le64 features;
+	u8     version;
+	u8     hw_index;
+	u8     rsvd[2];
+	__le32 max_nb_sessions;
+	u8     rsvd2[120];
+	union pds_core_lif_config config;
+};
+
+/**
+ * struct pds_core_lif_identify_cmd - Get LIF identity info command
+ * @opcode:	Opcode PDS_AQ_CMD_LIF_IDENTIFY
+ * @type:	LIF type (enum pds_core_lif_type)
+ * @client_id:	Client identifier
+ * @ver:	Version of identify returned by device
+ * @rsvd:       Word boundary padding
+ * @ident_pa:	DMA address to receive identity info
+ *
+ * Firmware will copy LIF identity data (struct pds_core_lif_identity)
+ * into the buffer address given.
+ */
+struct pds_core_lif_identify_cmd {
+	u8     opcode;
+	u8     type;
+	__le16 client_id;
+	u8     ver;
+	u8     rsvd[3];
+	__le64 ident_pa;
+};
+
+/**
+ * struct pds_core_lif_identify_comp - LIF identify command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @ver:	Version of identify returned by device
+ * @bytes:	Bytes copied into the buffer
+ * @rsvd:       Word boundary padding
+ * @color:      Color bit
+ */
+struct pds_core_lif_identify_comp {
+	u8     status;
+	u8     ver;
+	__le16 bytes;
+	u8     rsvd[11];
+	u8     color;
+};
+
+/**
+ * struct pds_core_lif_init_cmd - LIF init command
+ * @opcode:	Opcode PDS_AQ_CMD_LIF_INIT
+ * @type:	LIF type (enum pds_core_lif_type)
+ * @client_id:	Client identifier
+ * @rsvd:       Word boundary padding
+ * @info_pa:	Destination address for LIF info (struct pds_core_lif_info)
+ */
+struct pds_core_lif_init_cmd {
+	u8     opcode;
+	u8     type;
+	__le16 client_id;
+	__le32 rsvd;
+	__le64 info_pa;
+};
+
+/**
+ * struct pds_core_lif_init_comp - LIF init command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @rsvd:       Word boundary padding
+ * @hw_index:	Hardware index of the initialized LIF
+ * @rsvd1:      Word boundary padding
+ * @color:      Color bit
+ */
+struct pds_core_lif_init_comp {
+	u8 status;
+	u8 rsvd;
+	__le16 hw_index;
+	u8     rsvd1[11];
+	u8     color;
+};
+
+/**
+ * struct pds_core_lif_reset_cmd - LIF reset command
+ * Will reset only the specified LIF.
+ * @opcode:	Opcode PDS_AQ_CMD_LIF_RESET
+ * @rsvd:       Word boundary padding
+ * @client_id:	Client identifier
+ */
+struct pds_core_lif_reset_cmd {
+	u8     opcode;
+	u8     rsvd;
+	__le16 client_id;
+};
+
+/**
+ * enum pds_core_lif_attr - List of LIF attributes
+ * @PDS_CORE_LIF_ATTR_STATE:		LIF state attribute
+ * @PDS_CORE_LIF_ATTR_NAME:		LIF name attribute
+ * @PDS_CORE_LIF_ATTR_FEATURES:		LIF features attribute
+ * @PDS_CORE_LIF_ATTR_STATS_CTRL:	LIF statistics control attribute
+ */
+enum pds_core_lif_attr {
+	PDS_CORE_LIF_ATTR_STATE		= 0,
+	PDS_CORE_LIF_ATTR_NAME		= 1,
+	PDS_CORE_LIF_ATTR_FEATURES	= 4,
+	PDS_CORE_LIF_ATTR_STATS_CTRL	= 6,
+};
+
+/**
+ * struct pds_core_lif_setattr_cmd - Set LIF attributes on the NIC
+ * @opcode:	Opcode PDS_AQ_CMD_LIF_SETATTR
+ * @attr:	Attribute type (enum pds_core_lif_attr)
+ * @client_id:	Client identifier
+ * @state:	LIF state (enum pds_core_lif_state)
+ * @name:	The name string, 0 terminated
+ * @features:	Features (enum pds_core_hw_features)
+ * @stats_ctl:	Stats control commands (enum pds_core_stats_ctl_cmd)
+ * @rsvd:       Command Buffer padding
+ */
+struct pds_core_lif_setattr_cmd {
+	u8     opcode;
+	u8     attr;
+	__le16 client_id;
+	union {
+		u8      state;
+		char    name[PDS_CORE_IFNAMSIZ];
+		__le64  features;
+		u8      stats_ctl;
+		u8      rsvd[60];
+	} __packed;
+};
+
+/**
+ * struct pds_core_lif_setattr_comp - LIF set attr command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @rsvd:       Word boundary padding
+ * @comp_index: Index in the descriptor ring for which this is the completion
+ * @features:	Features (enum pds_core_hw_features)
+ * @rsvd2:      Word boundary padding
+ * @color:	Color bit
+ */
+struct pds_core_lif_setattr_comp {
+	u8     status;
+	u8     rsvd;
+	__le16 comp_index;
+	union {
+		__le64  features;
+		u8      rsvd2[11];
+	} __packed;
+	u8     color;
+};
+
+/**
+ * struct pds_core_lif_getattr_cmd - Get LIF attributes from the NIC
+ * @opcode:	Opcode PDS_AQ_CMD_LIF_GETATTR
+ * @attr:	Attribute type (enum pds_core_lif_attr)
+ * @client_id:	Client identifier
+ */
+struct pds_core_lif_getattr_cmd {
+	u8     opcode;
+	u8     attr;
+	__le16 client_id;
+};
+
+/**
+ * struct pds_core_lif_getattr_comp - LIF get attr command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @rsvd:       Word boundary padding
+ * @comp_index: Index in the descriptor ring for which this is the completion
+ * @state:	LIF state (enum pds_core_lif_state)
+ * @name:	LIF name string, 0 terminated
+ * @features:	Features (enum pds_core_hw_features)
+ * @rsvd2:      Word boundary padding
+ * @color:	Color bit
+ */
+struct pds_core_lif_getattr_comp {
+	u8     status;
+	u8     rsvd;
+	__le16 comp_index;
+	union {
+		u8      state;
+		__le64  features;
+		u8      rsvd2[11];
+	} __packed;
+	u8     color;
+};
+
+/**
+ * union pds_core_q_identity - Queue identity information
+ * @version:	Queue type version that can be used with FW
+ * @supported:	Bitfield of queue versions, first bit = ver 0
+ * @rsvd:       Word boundary padding
+ * @features:	Queue features
+ * @desc_sz:	Descriptor size
+ * @comp_sz:	Completion descriptor size
+ * @rsvd2:      Word boundary padding
+ */
+struct pds_core_q_identity {
+	u8      version;
+	u8      supported;
+	u8      rsvd[6];
+#define PDS_CORE_QIDENT_F_CQ	0x01	/* queue has completion ring */
+	__le64  features;
+	__le16  desc_sz;
+	__le16  comp_sz;
+	u8      rsvd2[6];
+};
+
+/**
+ * struct pds_core_q_identify_cmd - queue identify command
+ * @opcode:	Opcode PDS_AQ_CMD_Q_IDENTIFY
+ * @type:	Logical queue type (enum pds_core_logical_qtype)
+ * @client_id:	Client identifier
+ * @ver:	Highest queue type version that the driver supports
+ * @rsvd:       Word boundary padding
+ * @ident_pa:   DMA address to receive the data (struct pds_core_q_identity)
+ */
+struct pds_core_q_identify_cmd {
+	u8     opcode;
+	u8     type;
+	__le16 client_id;
+	u8     ver;
+	u8     rsvd[3];
+	__le64 ident_pa;
+};
+
+/**
+ * struct pds_core_q_identify_comp - queue identify command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @rsvd:       Word boundary padding
+ * @comp_index:	Index in the descriptor ring for which this is the completion
+ * @ver:	Queue type version that can be used with FW
+ * @rsvd1:      Word boundary padding
+ * @color:      Color bit
+ */
+struct pds_core_q_identify_comp {
+	u8     status;
+	u8     rsvd;
+	__le16 comp_index;
+	u8     ver;
+	u8     rsvd1[10];
+	u8     color;
+};
+
+/**
+ * struct pds_core_q_init_cmd - Queue init command
+ * @opcode:	  Opcode PDS_AQ_CMD_Q_INIT
+ * @type:	  Logical queue type
+ * @client_id:	  Client identifier
+ * @ver:	  Queue type version
+ * @rsvd:         Word boundary padding
+ * @index:	  (LIF, qtype) relative admin queue index
+ * @intr_index:	  Interrupt control register index, or Event queue index
+ * @pid:	  Process ID
+ * @flags:
+ *    IRQ:	  Interrupt requested on completion
+ *    ENA:	  Enable the queue.  If ENA=0 the queue is initialized
+ *		  but remains disabled, to be later enabled with the
+ *		  Queue Enable command. If ENA=1, then queue is
+ *		  initialized and then enabled.
+ * @cos:	  Class of service for this queue
+ * @ring_size:	  Queue ring size, encoded as a log2(size), in
+ *		  number of descriptors.  The actual ring size is
+ *		  (1 << ring_size).  For example, to select a ring size
+ *		  of 64 descriptors write ring_size = 6. The minimum
+ *		  ring_size value is 2 for a ring of 4 descriptors.
+ *		  The maximum ring_size value is 12 for a ring of 4k
+ *		  descriptors. Values of ring_size <2 and >12 are
+ *		  reserved.
+ * @ring_base:	  Queue ring base address
+ * @cq_ring_base: Completion queue ring base address
+ */
+struct pds_core_q_init_cmd {
+	u8     opcode;
+	u8     type;
+	__le16 client_id;
+	u8     ver;
+	u8     rsvd[3];
+	__le32 index;
+	__le16 pid;
+	__le16 intr_index;
+	__le16 flags;
+#define PDS_CORE_QINIT_F_IRQ	0x01	/* Request interrupt on completion */
+#define PDS_CORE_QINIT_F_ENA	0x02	/* Enable the queue */
+	u8     cos;
+#define PDS_CORE_QSIZE_MIN_LG2	2
+#define PDS_CORE_QSIZE_MAX_LG2	12
+	u8     ring_size;
+	__le64 ring_base;
+	__le64 cq_ring_base;
+} __packed;
+
+/**
+ * struct pds_core_q_init_comp - Queue init command completion
+ * @status:	Status of the command (enum pds_core_status_code)
+ * @rsvd:       Word boundary padding
+ * @comp_index:	Index in the descriptor ring for which this is the completion
+ * @hw_index:	Hardware Queue ID
+ * @hw_type:	Hardware Queue type
+ * @rsvd2:      Word boundary padding
+ * @color:	Color
+ */
+struct pds_core_q_init_comp {
+	u8     status;
+	u8     rsvd;
+	__le16 comp_index;
+	__le32 hw_index;
+	u8     hw_type;
+	u8     rsvd2[6];
+	u8     color;
+};
+
+union pds_core_adminq_cmd {
+	u8     opcode;
+	u8     bytes[64];
+
+	struct pds_core_client_reg_cmd     client_reg;
+	struct pds_core_client_unreg_cmd   client_unreg;
+	struct pds_core_client_request_cmd client_request;
+
+	struct pds_core_lif_identify_cmd  lif_ident;
+	struct pds_core_lif_init_cmd      lif_init;
+	struct pds_core_lif_reset_cmd     lif_reset;
+	struct pds_core_lif_setattr_cmd   lif_setattr;
+	struct pds_core_lif_getattr_cmd   lif_getattr;
+
+	struct pds_core_q_identify_cmd    q_ident;
+	struct pds_core_q_init_cmd        q_init;
+};
+
+union pds_core_adminq_comp {
+	struct {
+		u8     status;
+		u8     rsvd;
+		__le16 comp_index;
+		u8     rsvd2[11];
+		u8     color;
+	};
+	u32    words[4];
+
+	struct pds_core_client_reg_comp   client_reg;
+
+	struct pds_core_lif_identify_comp lif_ident;
+	struct pds_core_lif_init_comp     lif_init;
+	struct pds_core_lif_setattr_comp  lif_setattr;
+	struct pds_core_lif_getattr_comp  lif_getattr;
+
+	struct pds_core_q_identify_comp   q_ident;
+	struct pds_core_q_init_comp       q_init;
+};
+
+#ifndef __CHECKER__
+static_assert(sizeof(union pds_core_adminq_cmd) == 64);
+static_assert(sizeof(union pds_core_adminq_comp) == 16);
+static_assert(sizeof(union pds_core_notifyq_comp) == 64);
+#endif /* __CHECKER__ */
+
+/* The color bit is a 'done' bit for the completion descriptors
+ * where the meaning alternates between '1' and '0' for alternating
+ * passes through the completion descriptor ring.
+ */
+static inline u8 pdsc_color_match(u8 color, u8 done_color)
+{
+	return (!!(color & PDS_COMP_COLOR_MASK)) == done_color;
+}
+#endif /* _PDS_CORE_ADMINQ_H_ */
-- 
cgit v1.2.3-70-g09d2


From 01ba61b55b2041a39c54aefb3153c770dd59a0ef Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:19 -0700
Subject: pds_core: Add adminq processing and commands

Add the service routines for submitting and processing
the adminq messages and for handling notifyq events.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/Makefile |   1 +
 drivers/net/ethernet/amd/pds_core/adminq.c | 288 +++++++++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/core.c   |  11 --
 include/linux/pds/pds_adminq.h             |  11 +-
 4 files changed, 299 insertions(+), 12 deletions(-)
 create mode 100644 drivers/net/ethernet/amd/pds_core/adminq.c

diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
index eaca8557ba66..ef76dcd7fccd 100644
--- a/drivers/net/ethernet/amd/pds_core/Makefile
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -6,6 +6,7 @@ obj-$(CONFIG_PDS_CORE) := pds_core.o
 pds_core-y := main.o \
 	      devlink.o \
 	      dev.o \
+	      adminq.o \
 	      core.o
 
 pds_core-$(CONFIG_DEBUG_FS) += debugfs.o
diff --git a/drivers/net/ethernet/amd/pds_core/adminq.c b/drivers/net/ethernet/amd/pds_core/adminq.c
new file mode 100644
index 000000000000..fb2ba3f62480
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/adminq.c
@@ -0,0 +1,288 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/dynamic_debug.h>
+
+#include "core.h"
+
+struct pdsc_wait_context {
+	struct pdsc_qcq *qcq;
+	struct completion wait_completion;
+};
+
+static int pdsc_process_notifyq(struct pdsc_qcq *qcq)
+{
+	union pds_core_notifyq_comp *comp;
+	struct pdsc *pdsc = qcq->pdsc;
+	struct pdsc_cq *cq = &qcq->cq;
+	struct pdsc_cq_info *cq_info;
+	int nq_work = 0;
+	u64 eid;
+
+	cq_info = &cq->info[cq->tail_idx];
+	comp = cq_info->comp;
+	eid = le64_to_cpu(comp->event.eid);
+	while (eid > pdsc->last_eid) {
+		u16 ecode = le16_to_cpu(comp->event.ecode);
+
+		switch (ecode) {
+		case PDS_EVENT_LINK_CHANGE:
+			dev_info(pdsc->dev, "NotifyQ LINK_CHANGE ecode %d eid %lld\n",
+				 ecode, eid);
+			break;
+
+		case PDS_EVENT_RESET:
+			dev_info(pdsc->dev, "NotifyQ RESET ecode %d eid %lld\n",
+				 ecode, eid);
+			break;
+
+		case PDS_EVENT_XCVR:
+			dev_info(pdsc->dev, "NotifyQ XCVR ecode %d eid %lld\n",
+				 ecode, eid);
+			break;
+
+		default:
+			dev_info(pdsc->dev, "NotifyQ ecode %d eid %lld\n",
+				 ecode, eid);
+			break;
+		}
+
+		pdsc->last_eid = eid;
+		cq->tail_idx = (cq->tail_idx + 1) & (cq->num_descs - 1);
+		cq_info = &cq->info[cq->tail_idx];
+		comp = cq_info->comp;
+		eid = le64_to_cpu(comp->event.eid);
+
+		nq_work++;
+	}
+
+	qcq->accum_work += nq_work;
+
+	return nq_work;
+}
+
+void pdsc_process_adminq(struct pdsc_qcq *qcq)
+{
+	union pds_core_adminq_comp *comp;
+	struct pdsc_queue *q = &qcq->q;
+	struct pdsc *pdsc = qcq->pdsc;
+	struct pdsc_cq *cq = &qcq->cq;
+	struct pdsc_q_info *q_info;
+	unsigned long irqflags;
+	int nq_work = 0;
+	int aq_work = 0;
+	int credits;
+
+	/* Don't process AdminQ when shutting down */
+	if (pdsc->state & BIT_ULL(PDSC_S_STOPPING_DRIVER)) {
+		dev_err(pdsc->dev, "%s: called while PDSC_S_STOPPING_DRIVER\n",
+			__func__);
+		return;
+	}
+
+	/* Check for NotifyQ event */
+	nq_work = pdsc_process_notifyq(&pdsc->notifyqcq);
+
+	/* Check for empty queue, which can happen if the interrupt was
+	 * for a NotifyQ event and there are no new AdminQ completions.
+	 */
+	if (q->tail_idx == q->head_idx)
+		goto credits;
+
+	/* Find the first completion to clean,
+	 * run the callback in the related q_info,
+	 * and continue while we still match done color
+	 */
+	spin_lock_irqsave(&pdsc->adminq_lock, irqflags);
+	comp = cq->info[cq->tail_idx].comp;
+	while (pdsc_color_match(comp->color, cq->done_color)) {
+		q_info = &q->info[q->tail_idx];
+		q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1);
+
+		/* Copy out the completion data */
+		memcpy(q_info->dest, comp, sizeof(*comp));
+
+		complete_all(&q_info->wc->wait_completion);
+
+		if (cq->tail_idx == cq->num_descs - 1)
+			cq->done_color = !cq->done_color;
+		cq->tail_idx = (cq->tail_idx + 1) & (cq->num_descs - 1);
+		comp = cq->info[cq->tail_idx].comp;
+
+		aq_work++;
+	}
+	spin_unlock_irqrestore(&pdsc->adminq_lock, irqflags);
+
+	qcq->accum_work += aq_work;
+
+credits:
+	/* Return the interrupt credits, one for each completion */
+	credits = nq_work + aq_work;
+	if (credits)
+		pds_core_intr_credits(&pdsc->intr_ctrl[qcq->intx],
+				      credits,
+				      PDS_CORE_INTR_CRED_REARM);
+}
+
+void pdsc_work_thread(struct work_struct *work)
+{
+	struct pdsc_qcq *qcq = container_of(work, struct pdsc_qcq, work);
+
+	pdsc_process_adminq(qcq);
+}
+
+irqreturn_t pdsc_adminq_isr(int irq, void *data)
+{
+	struct pdsc_qcq *qcq = data;
+	struct pdsc *pdsc = qcq->pdsc;
+
+	/* Don't process AdminQ when shutting down */
+	if (pdsc->state & BIT_ULL(PDSC_S_STOPPING_DRIVER)) {
+		dev_err(pdsc->dev, "%s: called while PDSC_S_STOPPING_DRIVER\n",
+			__func__);
+		return IRQ_HANDLED;
+	}
+
+	queue_work(pdsc->wq, &qcq->work);
+	pds_core_intr_mask(&pdsc->intr_ctrl[irq], PDS_CORE_INTR_MASK_CLEAR);
+
+	return IRQ_HANDLED;
+}
+
+static int __pdsc_adminq_post(struct pdsc *pdsc,
+			      struct pdsc_qcq *qcq,
+			      union pds_core_adminq_cmd *cmd,
+			      union pds_core_adminq_comp *comp,
+			      struct pdsc_wait_context *wc)
+{
+	struct pdsc_queue *q = &qcq->q;
+	struct pdsc_q_info *q_info;
+	unsigned long irqflags;
+	unsigned int avail;
+	int index;
+	int ret;
+
+	spin_lock_irqsave(&pdsc->adminq_lock, irqflags);
+
+	/* Check for space in the queue */
+	avail = q->tail_idx;
+	if (q->head_idx >= avail)
+		avail += q->num_descs - q->head_idx - 1;
+	else
+		avail -= q->head_idx + 1;
+	if (!avail) {
+		ret = -ENOSPC;
+		goto err_out_unlock;
+	}
+
+	/* Check that the FW is running */
+	if (!pdsc_is_fw_running(pdsc)) {
+		u8 fw_status = ioread8(&pdsc->info_regs->fw_status);
+
+		dev_info(pdsc->dev, "%s: post failed - fw not running %#02x:\n",
+			 __func__, fw_status);
+		ret = -ENXIO;
+
+		goto err_out_unlock;
+	}
+
+	/* Post the request */
+	index = q->head_idx;
+	q_info = &q->info[index];
+	q_info->wc = wc;
+	q_info->dest = comp;
+	memcpy(q_info->desc, cmd, sizeof(*cmd));
+
+	dev_dbg(pdsc->dev, "head_idx %d tail_idx %d\n",
+		q->head_idx, q->tail_idx);
+	dev_dbg(pdsc->dev, "post admin queue command:\n");
+	dynamic_hex_dump("cmd ", DUMP_PREFIX_OFFSET, 16, 1,
+			 cmd, sizeof(*cmd), true);
+
+	q->head_idx = (q->head_idx + 1) & (q->num_descs - 1);
+
+	pds_core_dbell_ring(pdsc->kern_dbpage,
+			    q->hw_type, q->dbval | q->head_idx);
+	ret = index;
+
+err_out_unlock:
+	spin_unlock_irqrestore(&pdsc->adminq_lock, irqflags);
+	return ret;
+}
+
+int pdsc_adminq_post(struct pdsc *pdsc,
+		     union pds_core_adminq_cmd *cmd,
+		     union pds_core_adminq_comp *comp,
+		     bool fast_poll)
+{
+	struct pdsc_wait_context wc = {
+		.wait_completion =
+			COMPLETION_INITIALIZER_ONSTACK(wc.wait_completion),
+	};
+	unsigned long poll_interval = 1;
+	unsigned long poll_jiffies;
+	unsigned long time_limit;
+	unsigned long time_start;
+	unsigned long time_done;
+	unsigned long remaining;
+	int err = 0;
+	int index;
+
+	wc.qcq = &pdsc->adminqcq;
+	index = __pdsc_adminq_post(pdsc, &pdsc->adminqcq, cmd, comp, &wc);
+	if (index < 0) {
+		err = index;
+		goto err_out;
+	}
+
+	time_start = jiffies;
+	time_limit = time_start + HZ * pdsc->devcmd_timeout;
+	do {
+		/* Timeslice the actual wait to catch IO errors etc early */
+		poll_jiffies = msecs_to_jiffies(poll_interval);
+		remaining = wait_for_completion_timeout(&wc.wait_completion,
+							poll_jiffies);
+		if (remaining)
+			break;
+
+		if (!pdsc_is_fw_running(pdsc)) {
+			u8 fw_status = ioread8(&pdsc->info_regs->fw_status);
+
+			dev_dbg(pdsc->dev, "%s: post wait failed - fw not running %#02x:\n",
+				__func__, fw_status);
+			err = -ENXIO;
+			break;
+		}
+
+		/* When fast_poll is not requested, prevent aggressive polling
+		 * on failures due to timeouts by doing exponential back off.
+		 */
+		if (!fast_poll && poll_interval < PDSC_ADMINQ_MAX_POLL_INTERVAL)
+			poll_interval <<= 1;
+	} while (time_before(jiffies, time_limit));
+	time_done = jiffies;
+	dev_dbg(pdsc->dev, "%s: elapsed %d msecs\n",
+		__func__, jiffies_to_msecs(time_done - time_start));
+
+	/* Check the results */
+	if (time_after_eq(time_done, time_limit))
+		err = -ETIMEDOUT;
+
+	dev_dbg(pdsc->dev, "read admin queue completion idx %d:\n", index);
+	dynamic_hex_dump("comp ", DUMP_PREFIX_OFFSET, 16, 1,
+			 comp, sizeof(*comp), true);
+
+	if (remaining && comp->status)
+		err = pdsc_err_to_errno(comp->status);
+
+err_out:
+	if (err) {
+		dev_dbg(pdsc->dev, "%s: opcode %d status %d err %pe\n",
+			__func__, cmd->opcode, comp->status, ERR_PTR(err));
+		if (err == -ENXIO || err == -ETIMEDOUT)
+			queue_work(pdsc->wq, &pdsc->health_work);
+	}
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(pdsc_adminq_post);
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index 8c0dbdb5efc5..59daf8a67ac6 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -6,17 +6,6 @@
 
 #include "core.h"
 
-void pdsc_work_thread(struct work_struct *work)
-{
-	/* stub */
-}
-
-irqreturn_t pdsc_adminq_isr(int irq, void *data)
-{
-	/* stub */
-	return IRQ_HANDLED;
-}
-
 void pdsc_intr_free(struct pdsc *pdsc, int index)
 {
 	struct pdsc_intr_info *intr_info;
diff --git a/include/linux/pds/pds_adminq.h b/include/linux/pds/pds_adminq.h
index dd5fbe3ee141..98a60ce87b92 100644
--- a/include/linux/pds/pds_adminq.h
+++ b/include/linux/pds/pds_adminq.h
@@ -4,6 +4,8 @@
 #ifndef _PDS_CORE_ADMINQ_H_
 #define _PDS_CORE_ADMINQ_H_
 
+#define PDSC_ADMINQ_MAX_POLL_INTERVAL	256
+
 enum pds_core_adminq_flags {
 	PDS_AQ_FLAG_FASTPOLL	= BIT(1),	/* completion poll at 1ms */
 };
@@ -631,8 +633,15 @@ static_assert(sizeof(union pds_core_notifyq_comp) == 64);
  * where the meaning alternates between '1' and '0' for alternating
  * passes through the completion descriptor ring.
  */
-static inline u8 pdsc_color_match(u8 color, u8 done_color)
+static inline bool pdsc_color_match(u8 color, bool done_color)
 {
 	return (!!(color & PDS_COMP_COLOR_MASK)) == done_color;
 }
+
+struct pdsc;
+int pdsc_adminq_post(struct pdsc *pdsc,
+		     union pds_core_adminq_cmd *cmd,
+		     union pds_core_adminq_comp *comp,
+		     bool fast_poll);
+
 #endif /* _PDS_CORE_ADMINQ_H_ */
-- 
cgit v1.2.3-70-g09d2


From 49ce92fbee0b6bb8066dddf37489483b3b6b5c25 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:20 -0700
Subject: pds_core: add FW update feature to devlink

Add in the support for doing firmware updates.  Of the two
main banks available, a and b, this updates the one not in
use and then selects it for the next boot.

Example:
    devlink dev flash pci/0000:b2:00.0 \
	    file pensando/dsc_fw_1.63.0-22.tar

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../device_drivers/ethernet/amd/pds_core.rst       |  10 ++
 drivers/net/ethernet/amd/pds_core/Makefile         |   3 +-
 drivers/net/ethernet/amd/pds_core/core.h           |   5 +
 drivers/net/ethernet/amd/pds_core/devlink.c        |   9 +
 drivers/net/ethernet/amd/pds_core/fw.c             | 194 +++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/main.c           |   1 +
 6 files changed, 221 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/amd/pds_core/fw.c

diff --git a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
index a48eafb3d0d3..932ac03a3359 100644
--- a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
+++ b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
@@ -73,6 +73,16 @@ The ``pds_core`` driver reports the following versions
      - fixed
      - The revision of the ASIC for this device
 
+Firmware Management
+===================
+
+The ``flash`` command can update a the DSC firmware.  The downloaded firmware
+will be saved into either of firmware bank 1 or bank 2, whichever is not
+currently in use, and that bank will used for the next boot::
+
+  # devlink dev flash pci/0000:b5:00.0 \
+            file pensando/dsc_fw_1.63.0-22.tar
+
 Health Reporters
 ================
 
diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
index ef76dcd7fccd..6d1d6c58a1fa 100644
--- a/drivers/net/ethernet/amd/pds_core/Makefile
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -7,6 +7,7 @@ pds_core-y := main.o \
 	      devlink.o \
 	      dev.o \
 	      adminq.o \
-	      core.o
+	      core.o \
+	      fw.o
 
 pds_core-$(CONFIG_DEBUG_FS) += debugfs.o
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 7adbd10cb6d9..7eb02b359f3a 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -226,6 +226,9 @@ int pdsc_fw_reporter_diagnose(struct devlink_health_reporter *reporter,
 			      struct netlink_ext_ack *extack);
 int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
 		     struct netlink_ext_ack *extack);
+int pdsc_dl_flash_update(struct devlink *dl,
+			 struct devlink_flash_update_params *params,
+			 struct netlink_ext_ack *extack);
 
 void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num);
 
@@ -268,4 +271,6 @@ void pdsc_process_adminq(struct pdsc_qcq *qcq);
 void pdsc_work_thread(struct work_struct *work);
 irqreturn_t pdsc_adminq_isr(int irq, void *data);
 
+int pdsc_firmware_update(struct pdsc *pdsc, const struct firmware *fw,
+			 struct netlink_ext_ack *extack);
 #endif /* _PDSC_H_ */
diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c
index 89f23d620d1d..f91d65cc78b5 100644
--- a/drivers/net/ethernet/amd/pds_core/devlink.c
+++ b/drivers/net/ethernet/amd/pds_core/devlink.c
@@ -3,6 +3,15 @@
 
 #include "core.h"
 
+int pdsc_dl_flash_update(struct devlink *dl,
+			 struct devlink_flash_update_params *params,
+			 struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+
+	return pdsc_firmware_update(pdsc, params->fw, extack);
+}
+
 static char *fw_slotnames[] = {
 	"fw.goldfw",
 	"fw.mainfwa",
diff --git a/drivers/net/ethernet/amd/pds_core/fw.c b/drivers/net/ethernet/amd/pds_core/fw.c
new file mode 100644
index 000000000000..90a811f3878a
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/fw.c
@@ -0,0 +1,194 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include "core.h"
+
+/* The worst case wait for the install activity is about 25 minutes when
+ * installing a new CPLD, which is very seldom.  Normal is about 30-35
+ * seconds.  Since the driver can't tell if a CPLD update will happen we
+ * set the timeout for the ugly case.
+ */
+#define PDSC_FW_INSTALL_TIMEOUT	(25 * 60)
+#define PDSC_FW_SELECT_TIMEOUT	30
+
+/* Number of periodic log updates during fw file download */
+#define PDSC_FW_INTERVAL_FRACTION	32
+
+static int pdsc_devcmd_fw_download_locked(struct pdsc *pdsc, u64 addr,
+					  u32 offset, u32 length)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_download.opcode = PDS_CORE_CMD_FW_DOWNLOAD,
+		.fw_download.offset = cpu_to_le32(offset),
+		.fw_download.addr = cpu_to_le64(addr),
+		.fw_download.length = cpu_to_le32(length),
+	};
+	union pds_core_dev_comp comp;
+
+	return pdsc_devcmd_locked(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static int pdsc_devcmd_fw_install(struct pdsc *pdsc)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = PDS_CORE_FW_INSTALL_ASYNC
+	};
+	union pds_core_dev_comp comp;
+	int err;
+
+	err = pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+	if (err < 0)
+		return err;
+
+	return comp.fw_control.slot;
+}
+
+static int pdsc_devcmd_fw_activate(struct pdsc *pdsc,
+				   enum pds_core_fw_slot slot)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = PDS_CORE_FW_ACTIVATE_ASYNC,
+		.fw_control.slot = slot
+	};
+	union pds_core_dev_comp comp;
+
+	return pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+}
+
+static int pdsc_fw_status_long_wait(struct pdsc *pdsc,
+				    const char *label,
+				    unsigned long timeout,
+				    u8 fw_cmd,
+				    struct netlink_ext_ack *extack)
+{
+	union pds_core_dev_cmd cmd = {
+		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
+		.fw_control.oper = fw_cmd,
+	};
+	union pds_core_dev_comp comp;
+	unsigned long start_time;
+	unsigned long end_time;
+	int err;
+
+	/* Ping on the status of the long running async install
+	 * command.  We get EAGAIN while the command is still
+	 * running, else we get the final command status.
+	 */
+	start_time = jiffies;
+	end_time = start_time + (timeout * HZ);
+	do {
+		err = pdsc_devcmd(pdsc, &cmd, &comp, pdsc->devcmd_timeout);
+		msleep(20);
+	} while (time_before(jiffies, end_time) &&
+		 (err == -EAGAIN || err == -ETIMEDOUT));
+
+	if (err == -EAGAIN || err == -ETIMEDOUT) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware wait timed out");
+		dev_err(pdsc->dev, "DEV_CMD firmware wait %s timed out\n",
+			label);
+	} else if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware wait failed");
+	}
+
+	return err;
+}
+
+int pdsc_firmware_update(struct pdsc *pdsc, const struct firmware *fw,
+			 struct netlink_ext_ack *extack)
+{
+	u32 buf_sz, copy_sz, offset;
+	struct devlink *dl;
+	int next_interval;
+	u64 data_addr;
+	int err = 0;
+	int fw_slot;
+
+	dev_info(pdsc->dev, "Installing firmware\n");
+
+	dl = priv_to_devlink(pdsc);
+	devlink_flash_update_status_notify(dl, "Preparing to flash",
+					   NULL, 0, 0);
+
+	buf_sz = sizeof(pdsc->cmd_regs->data);
+
+	dev_dbg(pdsc->dev,
+		"downloading firmware - size %d part_sz %d nparts %lu\n",
+		(int)fw->size, buf_sz, DIV_ROUND_UP(fw->size, buf_sz));
+
+	offset = 0;
+	next_interval = 0;
+	data_addr = offsetof(struct pds_core_dev_cmd_regs, data);
+	while (offset < fw->size) {
+		if (offset >= next_interval) {
+			devlink_flash_update_status_notify(dl, "Downloading",
+							   NULL, offset,
+							   fw->size);
+			next_interval = offset +
+					(fw->size / PDSC_FW_INTERVAL_FRACTION);
+		}
+
+		copy_sz = min_t(unsigned int, buf_sz, fw->size - offset);
+		mutex_lock(&pdsc->devcmd_lock);
+		memcpy_toio(&pdsc->cmd_regs->data, fw->data + offset, copy_sz);
+		err = pdsc_devcmd_fw_download_locked(pdsc, data_addr,
+						     offset, copy_sz);
+		mutex_unlock(&pdsc->devcmd_lock);
+		if (err) {
+			dev_err(pdsc->dev,
+				"download failed offset 0x%x addr 0x%llx len 0x%x: %pe\n",
+				offset, data_addr, copy_sz, ERR_PTR(err));
+			NL_SET_ERR_MSG_MOD(extack, "Segment download failed");
+			goto err_out;
+		}
+		offset += copy_sz;
+	}
+	devlink_flash_update_status_notify(dl, "Downloading", NULL,
+					   fw->size, fw->size);
+
+	devlink_flash_update_timeout_notify(dl, "Installing", NULL,
+					    PDSC_FW_INSTALL_TIMEOUT);
+
+	fw_slot = pdsc_devcmd_fw_install(pdsc);
+	if (fw_slot < 0) {
+		err = fw_slot;
+		dev_err(pdsc->dev, "install failed: %pe\n", ERR_PTR(err));
+		NL_SET_ERR_MSG_MOD(extack, "Failed to start firmware install");
+		goto err_out;
+	}
+
+	err = pdsc_fw_status_long_wait(pdsc, "Installing",
+				       PDSC_FW_INSTALL_TIMEOUT,
+				       PDS_CORE_FW_INSTALL_STATUS,
+				       extack);
+	if (err)
+		goto err_out;
+
+	devlink_flash_update_timeout_notify(dl, "Selecting", NULL,
+					    PDSC_FW_SELECT_TIMEOUT);
+
+	err = pdsc_devcmd_fw_activate(pdsc, fw_slot);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to start firmware select");
+		goto err_out;
+	}
+
+	err = pdsc_fw_status_long_wait(pdsc, "Selecting",
+				       PDSC_FW_SELECT_TIMEOUT,
+				       PDS_CORE_FW_ACTIVATE_STATUS,
+				       extack);
+	if (err)
+		goto err_out;
+
+	dev_info(pdsc->dev, "Firmware update completed, slot %d\n", fw_slot);
+
+err_out:
+	if (err)
+		devlink_flash_update_status_notify(dl, "Flash failed",
+						   NULL, 0, 0);
+	else
+		devlink_flash_update_status_notify(dl, "Flash done",
+						   NULL, 0, 0);
+	return err;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index eaff311d4a10..54aaf213679f 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -228,6 +228,7 @@ err_out_release_regions:
 
 static const struct devlink_ops pdsc_dl_ops = {
 	.info_get	= pdsc_dl_info_get,
+	.flash_update	= pdsc_dl_flash_update,
 };
 
 static const struct devlink_ops pdsc_dl_vf_ops = {
-- 
cgit v1.2.3-70-g09d2


From 65e0185ad764d2801811bb2e7c122e92557208c4 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:21 -0700
Subject: pds_core: set up the VIF definitions and defaults

The Virtual Interfaces (VIFs) supported by the DSC's
configuration (vDPA, Eth, RDMA, etc) are reported in the
dev_ident struct and made visible in debugfs.  At this point
only vDPA is supported in this driver so we only setup
devices for that feature.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/core.c    | 48 +++++++++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/core.h    | 11 +++++++
 drivers/net/ethernet/amd/pds_core/debugfs.c | 24 +++++++++++++++
 include/linux/pds/pds_common.h              | 19 ++++++++++++
 4 files changed, 102 insertions(+)

diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index 59daf8a67ac6..b2fca3b99f02 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -346,6 +346,43 @@ static int pdsc_core_init(struct pdsc *pdsc)
 	return err;
 }
 
+static struct pdsc_viftype pdsc_viftype_defaults[] = {
+	[PDS_DEV_TYPE_VDPA] = { .name = PDS_DEV_TYPE_VDPA_STR,
+				.vif_id = PDS_DEV_TYPE_VDPA,
+				.dl_id = DEVLINK_PARAM_GENERIC_ID_ENABLE_VNET },
+	[PDS_DEV_TYPE_MAX] = {}
+};
+
+static int pdsc_viftypes_init(struct pdsc *pdsc)
+{
+	enum pds_core_vif_types vt;
+
+	pdsc->viftype_status = kzalloc(sizeof(pdsc_viftype_defaults),
+				       GFP_KERNEL);
+	if (!pdsc->viftype_status)
+		return -ENOMEM;
+
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++) {
+		bool vt_support;
+
+		if (!pdsc_viftype_defaults[vt].name)
+			continue;
+
+		/* Grab the defaults */
+		pdsc->viftype_status[vt] = pdsc_viftype_defaults[vt];
+
+		/* See what the Core device has for support */
+		vt_support = !!le16_to_cpu(pdsc->dev_ident.vif_types[vt]);
+		dev_dbg(pdsc->dev, "VIF %s is %ssupported\n",
+			pdsc->viftype_status[vt].name,
+			vt_support ? "" : "not ");
+
+		pdsc->viftype_status[vt].supported = vt_support;
+	}
+
+	return 0;
+}
+
 int pdsc_setup(struct pdsc *pdsc, bool init)
 {
 	int numdescs;
@@ -388,6 +425,14 @@ int pdsc_setup(struct pdsc *pdsc, bool init)
 	if (err)
 		goto err_out_teardown;
 
+	/* Set up the VIFs */
+	err = pdsc_viftypes_init(pdsc);
+	if (err)
+		goto err_out_teardown;
+
+	if (init)
+		pdsc_debugfs_add_viftype(pdsc);
+
 	clear_bit(PDSC_S_FW_DEAD, &pdsc->state);
 	return 0;
 
@@ -404,6 +449,9 @@ void pdsc_teardown(struct pdsc *pdsc, bool removing)
 	pdsc_qcq_free(pdsc, &pdsc->notifyqcq);
 	pdsc_qcq_free(pdsc, &pdsc->adminqcq);
 
+	kfree(pdsc->viftype_status);
+	pdsc->viftype_status = NULL;
+
 	if (pdsc->intr_info) {
 		for (i = 0; i < pdsc->nintrs; i++)
 			pdsc_intr_free(pdsc, i);
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 7eb02b359f3a..ac0480d7f0f1 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -123,6 +123,15 @@ struct pdsc_qcq {
 	struct dentry *dentry;
 };
 
+struct pdsc_viftype {
+	char *name;
+	bool supported;
+	bool enabled;
+	int dl_id;
+	int vif_id;
+	struct pds_auxiliary_dev *padev;
+};
+
 /* No state flags set means we are in a steady running state */
 enum pdsc_state_flags {
 	PDSC_S_FW_DEAD,		    /* stopped, wait on startup or recovery */
@@ -174,6 +183,7 @@ struct pdsc {
 	struct pdsc_qcq adminqcq;
 	struct pdsc_qcq notifyqcq;
 	u64 last_eid;
+	struct pdsc_viftype *viftype_status;
 };
 
 /** enum pds_core_dbell_bits - bitwise composition of dbell values.
@@ -237,6 +247,7 @@ void pdsc_debugfs_destroy(void);
 void pdsc_debugfs_add_dev(struct pdsc *pdsc);
 void pdsc_debugfs_del_dev(struct pdsc *pdsc);
 void pdsc_debugfs_add_ident(struct pdsc *pdsc);
+void pdsc_debugfs_add_viftype(struct pdsc *pdsc);
 void pdsc_debugfs_add_irqs(struct pdsc *pdsc);
 void pdsc_debugfs_add_qcq(struct pdsc *pdsc, struct pdsc_qcq *qcq);
 void pdsc_debugfs_del_qcq(struct pdsc_qcq *qcq);
diff --git a/drivers/net/ethernet/amd/pds_core/debugfs.c b/drivers/net/ethernet/amd/pds_core/debugfs.c
index b83e5016644b..8ec392299b7d 100644
--- a/drivers/net/ethernet/amd/pds_core/debugfs.c
+++ b/drivers/net/ethernet/amd/pds_core/debugfs.c
@@ -68,6 +68,30 @@ void pdsc_debugfs_add_ident(struct pdsc *pdsc)
 			    pdsc, &identity_fops);
 }
 
+static int viftype_show(struct seq_file *seq, void *v)
+{
+	struct pdsc *pdsc = seq->private;
+	int vt;
+
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++) {
+		if (!pdsc->viftype_status[vt].name)
+			continue;
+
+		seq_printf(seq, "%s\t%d supported %d enabled\n",
+			   pdsc->viftype_status[vt].name,
+			   pdsc->viftype_status[vt].supported,
+			   pdsc->viftype_status[vt].enabled);
+	}
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(viftype);
+
+void pdsc_debugfs_add_viftype(struct pdsc *pdsc)
+{
+	debugfs_create_file("viftypes", 0400, pdsc->dentry,
+			    pdsc, &viftype_fops);
+}
+
 static const struct debugfs_reg32 intr_ctrl_regs[] = {
 	{ .name = "coal_init", .offset = 0, },
 	{ .name = "mask", .offset = 4, },
diff --git a/include/linux/pds/pds_common.h b/include/linux/pds/pds_common.h
index f0798ce01acf..b2be14ebadb6 100644
--- a/include/linux/pds/pds_common.h
+++ b/include/linux/pds/pds_common.h
@@ -20,6 +20,25 @@ enum pds_core_driver_type {
 	PDS_DRIVER_ESXI    = 6,
 };
 
+enum pds_core_vif_types {
+	PDS_DEV_TYPE_CORE	= 0,
+	PDS_DEV_TYPE_VDPA	= 1,
+	PDS_DEV_TYPE_VFIO	= 2,
+	PDS_DEV_TYPE_ETH	= 3,
+	PDS_DEV_TYPE_RDMA	= 4,
+	PDS_DEV_TYPE_LM		= 5,
+
+	/* new ones added before this line */
+	PDS_DEV_TYPE_MAX	= 16   /* don't change - used in struct size */
+};
+
+#define PDS_DEV_TYPE_CORE_STR	"Core"
+#define PDS_DEV_TYPE_VDPA_STR	"vDPA"
+#define PDS_DEV_TYPE_VFIO_STR	"VFio"
+#define PDS_DEV_TYPE_ETH_STR	"Eth"
+#define PDS_DEV_TYPE_RDMA_STR	"RDMA"
+#define PDS_DEV_TYPE_LM_STR	"LM"
+
 #define PDS_CORE_IFNAMSIZ		16
 
 /**
-- 
cgit v1.2.3-70-g09d2


From f53d93110aa526ad3df628ad6d9f672c7d5ae285 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:22 -0700
Subject: pds_core: add initial VF device handling

This is the initial VF PCI driver framework for the new
pds_vdpa VF device, which will work in conjunction with an
auxiliary_bus client of the pds_core driver.  This does the
very basics of registering for the new VF device, setting
up debugfs entries, and registering with devlink.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/core.h |  9 ++++++
 drivers/net/ethernet/amd/pds_core/main.c | 48 +++++++++++++++++++++++++++++++-
 2 files changed, 56 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index ac0480d7f0f1..1ec8784773f1 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -30,6 +30,12 @@ struct pdsc_dev_bar {
 	int res_index;
 };
 
+struct pdsc_vf {
+	struct pds_auxiliary_dev *padev;
+	u16     index;
+	__le16  vif_types[PDS_DEV_TYPE_MAX];
+};
+
 struct pdsc_devinfo {
 	u8 asic_type;
 	u8 asic_rev;
@@ -147,6 +153,9 @@ struct pdsc {
 	struct dentry *dentry;
 	struct device *dev;
 	struct pdsc_dev_bar bars[PDS_CORE_BARS_MAX];
+	struct pdsc_vf *vfs;
+	int num_vfs;
+	int vf_id;
 	int hw_index;
 	int uid;
 
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index 54aaf213679f..511cb91a295e 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -16,6 +16,7 @@ MODULE_LICENSE("GPL");
 /* Supported devices */
 static const struct pci_device_id pdsc_id_table[] = {
 	{ PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_CORE_PF) },
+	{ PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_VDPA_VF) },
 	{ 0, }	/* end of table */
 };
 MODULE_DEVICE_TABLE(pci, pdsc_id_table);
@@ -132,9 +133,51 @@ void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num)
 			       (u64)page_num << PAGE_SHIFT, PAGE_SIZE);
 }
 
+static int pdsc_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+	struct pdsc *pdsc = pci_get_drvdata(pdev);
+	struct device *dev = pdsc->dev;
+	int ret = 0;
+
+	if (num_vfs > 0) {
+		pdsc->vfs = kcalloc(num_vfs, sizeof(struct pdsc_vf),
+				    GFP_KERNEL);
+		if (!pdsc->vfs)
+			return -ENOMEM;
+		pdsc->num_vfs = num_vfs;
+
+		ret = pci_enable_sriov(pdev, num_vfs);
+		if (ret) {
+			dev_err(dev, "Cannot enable SRIOV: %pe\n",
+				ERR_PTR(ret));
+			goto no_vfs;
+		}
+
+		return num_vfs;
+	}
+
+no_vfs:
+	pci_disable_sriov(pdev);
+
+	kfree(pdsc->vfs);
+	pdsc->vfs = NULL;
+	pdsc->num_vfs = 0;
+
+	return ret;
+}
+
 static int pdsc_init_vf(struct pdsc *vf)
 {
-	return -1;
+	struct devlink *dl;
+
+	vf->vf_id = pci_iov_vf_id(vf->pdev);
+
+	dl = priv_to_devlink(vf);
+	devl_lock(dl);
+	devl_register(dl);
+	devl_unlock(dl);
+
+	return 0;
 }
 
 static const struct devlink_health_reporter_ops pdsc_fw_reporter_ops = {
@@ -323,6 +366,8 @@ static void pdsc_remove(struct pci_dev *pdev)
 	devl_unlock(dl);
 
 	if (!pdev->is_virtfn) {
+		pdsc_sriov_configure(pdev, 0);
+
 		del_timer_sync(&pdsc->wdtimer);
 		if (pdsc->wq)
 			destroy_workqueue(pdsc->wq);
@@ -354,6 +399,7 @@ static struct pci_driver pdsc_driver = {
 	.id_table = pdsc_id_table,
 	.probe = pdsc_probe,
 	.remove = pdsc_remove,
+	.sriov_configure = pdsc_sriov_configure,
 };
 
 static int __init pdsc_init_module(void)
-- 
cgit v1.2.3-70-g09d2


From 4569cce43bc61e4cdd76597a1cf9b608846c18cc Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:23 -0700
Subject: pds_core: add auxiliary_bus devices

An auxiliary_bus device is created for each vDPA type VF at VF
probe and destroyed at VF remove.  The aux device name comes
from the driver name + VIF type + the unique id assigned at PCI
probe.  The VFs are always removed on PF remove, so there should
be no issues with VFs trying to access missing PF structures.

The auxiliary_device names will look like "pds_core.vDPA.nn"
where 'nn' is the VF's uid.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/Makefile |   1 +
 drivers/net/ethernet/amd/pds_core/auxbus.c | 115 +++++++++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/core.h   |   6 ++
 drivers/net/ethernet/amd/pds_core/main.c   |  36 ++++++++-
 include/linux/pds/pds_auxbus.h             |  14 ++++
 include/linux/pds/pds_common.h             |   1 +
 6 files changed, 171 insertions(+), 2 deletions(-)
 create mode 100644 drivers/net/ethernet/amd/pds_core/auxbus.c
 create mode 100644 include/linux/pds/pds_auxbus.h

diff --git a/drivers/net/ethernet/amd/pds_core/Makefile b/drivers/net/ethernet/amd/pds_core/Makefile
index 6d1d6c58a1fa..0abc33ce826c 100644
--- a/drivers/net/ethernet/amd/pds_core/Makefile
+++ b/drivers/net/ethernet/amd/pds_core/Makefile
@@ -5,6 +5,7 @@ obj-$(CONFIG_PDS_CORE) := pds_core.o
 
 pds_core-y := main.o \
 	      devlink.o \
+	      auxbus.o \
 	      dev.o \
 	      adminq.o \
 	      core.o \
diff --git a/drivers/net/ethernet/amd/pds_core/auxbus.c b/drivers/net/ethernet/amd/pds_core/auxbus.c
new file mode 100644
index 000000000000..adee516b3f0c
--- /dev/null
+++ b/drivers/net/ethernet/amd/pds_core/auxbus.c
@@ -0,0 +1,115 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#include <linux/pci.h>
+
+#include "core.h"
+#include <linux/pds/pds_auxbus.h>
+
+static void pdsc_auxbus_dev_release(struct device *dev)
+{
+	struct pds_auxiliary_dev *padev =
+		container_of(dev, struct pds_auxiliary_dev, aux_dev.dev);
+
+	kfree(padev);
+}
+
+static struct pds_auxiliary_dev *pdsc_auxbus_dev_register(struct pdsc *cf,
+							  struct pdsc *pf,
+							  char *name)
+{
+	struct auxiliary_device *aux_dev;
+	struct pds_auxiliary_dev *padev;
+	int err;
+
+	padev = kzalloc(sizeof(*padev), GFP_KERNEL);
+	if (!padev)
+		return ERR_PTR(-ENOMEM);
+
+	padev->vf_pdev = cf->pdev;
+
+	aux_dev = &padev->aux_dev;
+	aux_dev->name = name;
+	aux_dev->id = cf->uid;
+	aux_dev->dev.parent = cf->dev;
+	aux_dev->dev.release = pdsc_auxbus_dev_release;
+
+	err = auxiliary_device_init(aux_dev);
+	if (err < 0) {
+		dev_warn(cf->dev, "auxiliary_device_init of %s failed: %pe\n",
+			 name, ERR_PTR(err));
+		goto err_out;
+	}
+
+	err = auxiliary_device_add(aux_dev);
+	if (err) {
+		dev_warn(cf->dev, "auxiliary_device_add of %s failed: %pe\n",
+			 name, ERR_PTR(err));
+		goto err_out_uninit;
+	}
+
+	return padev;
+
+err_out_uninit:
+	auxiliary_device_uninit(aux_dev);
+err_out:
+	kfree(padev);
+	return ERR_PTR(err);
+}
+
+int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf)
+{
+	struct pds_auxiliary_dev *padev;
+	int err = 0;
+
+	mutex_lock(&pf->config_lock);
+
+	padev = pf->vfs[cf->vf_id].padev;
+	if (padev) {
+		auxiliary_device_delete(&padev->aux_dev);
+		auxiliary_device_uninit(&padev->aux_dev);
+	}
+	pf->vfs[cf->vf_id].padev = NULL;
+
+	mutex_unlock(&pf->config_lock);
+	return err;
+}
+
+int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf)
+{
+	struct pds_auxiliary_dev *padev;
+	enum pds_core_vif_types vt;
+	u16 vt_support;
+	int err = 0;
+
+	mutex_lock(&pf->config_lock);
+
+	/* We only support vDPA so far, so it is the only one to
+	 * be verified that it is available in the Core device and
+	 * enabled in the devlink param.  In the future this might
+	 * become a loop for several VIF types.
+	 */
+
+	/* Verify that the type is supported and enabled.  It is not
+	 * an error if there is no auxbus device support for this
+	 * VF, it just means something else needs to happen with it.
+	 */
+	vt = PDS_DEV_TYPE_VDPA;
+	vt_support = !!le16_to_cpu(pf->dev_ident.vif_types[vt]);
+	if (!(vt_support &&
+	      pf->viftype_status[vt].supported &&
+	      pf->viftype_status[vt].enabled))
+		goto out_unlock;
+
+	padev = pdsc_auxbus_dev_register(cf, pf,
+					 pf->viftype_status[vt].name);
+	if (IS_ERR(padev)) {
+		err = PTR_ERR(padev);
+		goto out_unlock;
+	}
+	pf->vfs[cf->vf_id].padev = padev;
+
+out_unlock:
+	mutex_unlock(&pf->config_lock);
+	return err;
+}
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 1ec8784773f1..36099d3ac3dd 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -30,8 +30,11 @@ struct pdsc_dev_bar {
 	int res_index;
 };
 
+struct pdsc;
+
 struct pdsc_vf {
 	struct pds_auxiliary_dev *padev;
+	struct pdsc *vf;
 	u16     index;
 	__le16  vif_types[PDS_DEV_TYPE_MAX];
 };
@@ -287,6 +290,9 @@ int pdsc_start(struct pdsc *pdsc);
 void pdsc_stop(struct pdsc *pdsc);
 void pdsc_health_thread(struct work_struct *work);
 
+int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf);
+int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf);
+
 void pdsc_process_adminq(struct pdsc_qcq *qcq);
 void pdsc_work_thread(struct work_struct *work);
 irqreturn_t pdsc_adminq_isr(int irq, void *data);
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index 511cb91a295e..b848f3360fe2 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -169,6 +169,12 @@ no_vfs:
 static int pdsc_init_vf(struct pdsc *vf)
 {
 	struct devlink *dl;
+	struct pdsc *pf;
+	int err;
+
+	pf = pdsc_get_pf_struct(vf->pdev);
+	if (IS_ERR_OR_NULL(pf))
+		return PTR_ERR(pf) ?: -1;
 
 	vf->vf_id = pci_iov_vf_id(vf->pdev);
 
@@ -177,7 +183,15 @@ static int pdsc_init_vf(struct pdsc *vf)
 	devl_register(dl);
 	devl_unlock(dl);
 
-	return 0;
+	pf->vfs[vf->vf_id].vf = vf;
+	err = pdsc_auxbus_dev_add(vf, pf);
+	if (err) {
+		devl_lock(dl);
+		devl_unregister(dl);
+		devl_unlock(dl);
+	}
+
+	return err;
 }
 
 static const struct devlink_health_reporter_ops pdsc_fw_reporter_ops = {
@@ -365,7 +379,19 @@ static void pdsc_remove(struct pci_dev *pdev)
 	}
 	devl_unlock(dl);
 
-	if (!pdev->is_virtfn) {
+	if (pdev->is_virtfn) {
+		struct pdsc *pf;
+
+		pf = pdsc_get_pf_struct(pdsc->pdev);
+		if (!IS_ERR(pf)) {
+			pdsc_auxbus_dev_del(pdsc, pf);
+			pf->vfs[pdsc->vf_id].vf = NULL;
+		}
+	} else {
+		/* Remove the VFs and their aux_bus connections before other
+		 * cleanup so that the clients can use the AdminQ to cleanly
+		 * shut themselves down.
+		 */
 		pdsc_sriov_configure(pdev, 0);
 
 		del_timer_sync(&pdsc->wdtimer);
@@ -402,6 +428,12 @@ static struct pci_driver pdsc_driver = {
 	.sriov_configure = pdsc_sriov_configure,
 };
 
+void *pdsc_get_pf_struct(struct pci_dev *vf_pdev)
+{
+	return pci_iov_get_pf_drvdata(vf_pdev, &pdsc_driver);
+}
+EXPORT_SYMBOL_GPL(pdsc_get_pf_struct);
+
 static int __init pdsc_init_module(void)
 {
 	if (strcmp(KBUILD_MODNAME, PDS_CORE_DRV_NAME))
diff --git a/include/linux/pds/pds_auxbus.h b/include/linux/pds/pds_auxbus.h
new file mode 100644
index 000000000000..493f75b1995e
--- /dev/null
+++ b/include/linux/pds/pds_auxbus.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2023 Advanced Micro Devices, Inc */
+
+#ifndef _PDSC_AUXBUS_H_
+#define _PDSC_AUXBUS_H_
+
+#include <linux/auxiliary_bus.h>
+
+struct pds_auxiliary_dev {
+	struct auxiliary_device aux_dev;
+	struct pci_dev *vf_pdev;
+	u16 client_id;
+};
+#endif /* _PDSC_AUXBUS_H_ */
diff --git a/include/linux/pds/pds_common.h b/include/linux/pds/pds_common.h
index b2be14ebadb6..961b3d02c69f 100644
--- a/include/linux/pds/pds_common.h
+++ b/include/linux/pds/pds_common.h
@@ -60,4 +60,5 @@ enum pds_core_logical_qtype {
 	PDS_CORE_QTYPE_MAX     = 16   /* don't change - used in struct size */
 };
 
+void *pdsc_get_pf_struct(struct pci_dev *vf_pdev);
 #endif /* _PDS_COMMON_H_ */
-- 
cgit v1.2.3-70-g09d2


From 40ced89445364baa8620e92fba5b3fff8d9742b9 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:24 -0700
Subject: pds_core: devlink params for enabling VIF support

Add the devlink parameter switches so the user can enable
the features supported by the VFs.  The only feature supported
at the moment is vDPA.

Example:
    devlink dev param set pci/0000:2b:00.0 \
	    name enable_vnet cmode runtime value true

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../device_drivers/ethernet/amd/pds_core.rst       | 19 ++++++
 drivers/net/ethernet/amd/pds_core/core.h           |  7 +++
 drivers/net/ethernet/amd/pds_core/devlink.c        | 73 ++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/main.c           | 34 ++++++++--
 4 files changed, 127 insertions(+), 6 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
index 932ac03a3359..b9f310de862e 100644
--- a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
+++ b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
@@ -73,6 +73,25 @@ The ``pds_core`` driver reports the following versions
      - fixed
      - The revision of the ASIC for this device
 
+Parameters
+==========
+
+The ``pds_core`` driver implements the following generic
+parameters for controlling the functionality to be made available
+as auxiliary_bus devices.
+
+.. list-table:: Generic parameters implemented
+   :widths: 5 5 8 82
+
+   * - Name
+     - Mode
+     - Type
+     - Description
+   * - ``enable_vnet``
+     - runtime
+     - Boolean
+     - Enables vDPA functionality through an auxiliary_bus device
+
 Firmware Management
 ===================
 
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 36099d3ac3dd..9e01a9ee6868 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -251,6 +251,13 @@ int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
 int pdsc_dl_flash_update(struct devlink *dl,
 			 struct devlink_flash_update_params *params,
 			 struct netlink_ext_ack *extack);
+int pdsc_dl_enable_get(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx);
+int pdsc_dl_enable_set(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx);
+int pdsc_dl_enable_validate(struct devlink *dl, u32 id,
+			    union devlink_param_value val,
+			    struct netlink_ext_ack *extack);
 
 void __iomem *pdsc_map_dbpage(struct pdsc *pdsc, int page_num);
 
diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c
index f91d65cc78b5..9c6b3653c1c7 100644
--- a/drivers/net/ethernet/amd/pds_core/devlink.c
+++ b/drivers/net/ethernet/amd/pds_core/devlink.c
@@ -2,6 +2,79 @@
 /* Copyright(c) 2023 Advanced Micro Devices, Inc */
 
 #include "core.h"
+#include <linux/pds/pds_auxbus.h>
+
+static struct
+pdsc_viftype *pdsc_dl_find_viftype_by_id(struct pdsc *pdsc,
+					 enum devlink_param_type dl_id)
+{
+	int vt;
+
+	for (vt = 0; vt < PDS_DEV_TYPE_MAX; vt++) {
+		if (pdsc->viftype_status[vt].dl_id == dl_id)
+			return &pdsc->viftype_status[vt];
+	}
+
+	return NULL;
+}
+
+int pdsc_dl_enable_get(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+	struct pdsc_viftype *vt_entry;
+
+	vt_entry = pdsc_dl_find_viftype_by_id(pdsc, id);
+	if (!vt_entry)
+		return -ENOENT;
+
+	ctx->val.vbool = vt_entry->enabled;
+
+	return 0;
+}
+
+int pdsc_dl_enable_set(struct devlink *dl, u32 id,
+		       struct devlink_param_gset_ctx *ctx)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+	struct pdsc_viftype *vt_entry;
+	int err = 0;
+	int vf_id;
+
+	vt_entry = pdsc_dl_find_viftype_by_id(pdsc, id);
+	if (!vt_entry || !vt_entry->supported)
+		return -EOPNOTSUPP;
+
+	if (vt_entry->enabled == ctx->val.vbool)
+		return 0;
+
+	vt_entry->enabled = ctx->val.vbool;
+	for (vf_id = 0; vf_id < pdsc->num_vfs; vf_id++) {
+		struct pdsc *vf = pdsc->vfs[vf_id].vf;
+
+		err = ctx->val.vbool ? pdsc_auxbus_dev_add(vf, pdsc) :
+				       pdsc_auxbus_dev_del(vf, pdsc);
+	}
+
+	return err;
+}
+
+int pdsc_dl_enable_validate(struct devlink *dl, u32 id,
+			    union devlink_param_value val,
+			    struct netlink_ext_ack *extack)
+{
+	struct pdsc *pdsc = devlink_priv(dl);
+	struct pdsc_viftype *vt_entry;
+
+	vt_entry = pdsc_dl_find_viftype_by_id(pdsc, id);
+	if (!vt_entry || !vt_entry->supported)
+		return -EOPNOTSUPP;
+
+	if (!pdsc->viftype_status[vt_entry->vif_id].supported)
+		return -ENODEV;
+
+	return 0;
+}
 
 int pdsc_dl_flash_update(struct devlink *dl,
 			 struct devlink_flash_update_params *params,
diff --git a/drivers/net/ethernet/amd/pds_core/main.c b/drivers/net/ethernet/amd/pds_core/main.c
index b848f3360fe2..e2d14b1ca471 100644
--- a/drivers/net/ethernet/amd/pds_core/main.c
+++ b/drivers/net/ethernet/amd/pds_core/main.c
@@ -199,6 +199,14 @@ static const struct devlink_health_reporter_ops pdsc_fw_reporter_ops = {
 	.diagnose = pdsc_fw_reporter_diagnose,
 };
 
+static const struct devlink_param pdsc_dl_params[] = {
+	DEVLINK_PARAM_GENERIC(ENABLE_VNET,
+			      BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			      pdsc_dl_enable_get,
+			      pdsc_dl_enable_set,
+			      pdsc_dl_enable_validate),
+};
+
 #define PDSC_WQ_NAME_LEN 24
 
 static int pdsc_init_pf(struct pdsc *pdsc)
@@ -246,13 +254,19 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	dl = priv_to_devlink(pdsc);
 	devl_lock(dl);
+	err = devl_params_register(dl, pdsc_dl_params,
+				   ARRAY_SIZE(pdsc_dl_params));
+	if (err) {
+		dev_warn(pdsc->dev, "Failed to register devlink params: %pe\n",
+			 ERR_PTR(err));
+		goto err_out_unlock_dl;
+	}
 
 	hr = devl_health_reporter_create(dl, &pdsc_fw_reporter_ops, 0, pdsc);
 	if (IS_ERR(hr)) {
 		dev_warn(pdsc->dev, "Failed to create fw reporter: %pe\n", hr);
 		err = PTR_ERR(hr);
-		devl_unlock(dl);
-		goto err_out_stop;
+		goto err_out_unreg_params;
 	}
 	pdsc->fw_reporter = hr;
 
@@ -264,7 +278,11 @@ static int pdsc_init_pf(struct pdsc *pdsc)
 
 	return 0;
 
-err_out_stop:
+err_out_unreg_params:
+	devl_params_unregister(dl, pdsc_dl_params,
+			       ARRAY_SIZE(pdsc_dl_params));
+err_out_unlock_dl:
+	devl_unlock(dl);
 	pdsc_stop(pdsc);
 err_out_teardown:
 	pdsc_teardown(pdsc, PDSC_TEARDOWN_REMOVING);
@@ -373,9 +391,13 @@ static void pdsc_remove(struct pci_dev *pdev)
 	dl = priv_to_devlink(pdsc);
 	devl_lock(dl);
 	devl_unregister(dl);
-	if (pdsc->fw_reporter) {
-		devl_health_reporter_destroy(pdsc->fw_reporter);
-		pdsc->fw_reporter = NULL;
+	if (!pdev->is_virtfn) {
+		if (pdsc->fw_reporter) {
+			devl_health_reporter_destroy(pdsc->fw_reporter);
+			pdsc->fw_reporter = NULL;
+		}
+		devl_params_unregister(dl, pdsc_dl_params,
+				       ARRAY_SIZE(pdsc_dl_params));
 	}
 	devl_unlock(dl);
 
-- 
cgit v1.2.3-70-g09d2


From 10659034c622738bc1bfab8a76fc576c52d5acce Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:25 -0700
Subject: pds_core: add the aux client API

Add the client API operations for running adminq commands.
The core registers the client with the FW, then the client
has a context for requesting adminq services.  We expect
to add additional operations for other clients, including
requesting additional private adminqs and IRQs, but don't have
the need yet.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/auxbus.c | 151 ++++++++++++++++++++++++++++-
 include/linux/pds/pds_auxbus.h             |   6 ++
 include/linux/pds/pds_common.h             |   2 +
 3 files changed, 158 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amd/pds_core/auxbus.c b/drivers/net/ethernet/amd/pds_core/auxbus.c
index adee516b3f0c..561af8e5b3ea 100644
--- a/drivers/net/ethernet/amd/pds_core/auxbus.c
+++ b/drivers/net/ethernet/amd/pds_core/auxbus.c
@@ -6,6 +6,136 @@
 #include "core.h"
 #include <linux/pds/pds_auxbus.h>
 
+/**
+ * pds_client_register - Link the client to the firmware
+ * @pf_pdev:	ptr to the PF driver struct
+ * @devname:	name that includes service into, e.g. pds_core.vDPA
+ *
+ * Return: 0 on success, or
+ *         negative for error
+ */
+int pds_client_register(struct pci_dev *pf_pdev, char *devname)
+{
+	union pds_core_adminq_comp comp = {};
+	union pds_core_adminq_cmd cmd = {};
+	struct pdsc *pf;
+	int err;
+	u16 ci;
+
+	pf = pci_get_drvdata(pf_pdev);
+	if (pf->state)
+		return -ENXIO;
+
+	cmd.client_reg.opcode = PDS_AQ_CMD_CLIENT_REG;
+	strscpy(cmd.client_reg.devname, devname,
+		sizeof(cmd.client_reg.devname));
+
+	err = pdsc_adminq_post(pf, &cmd, &comp, false);
+	if (err) {
+		dev_info(pf->dev, "register dev_name %s with DSC failed, status %d: %pe\n",
+			 devname, comp.status, ERR_PTR(err));
+		return err;
+	}
+
+	ci = le16_to_cpu(comp.client_reg.client_id);
+	if (!ci) {
+		dev_err(pf->dev, "%s: device returned null client_id\n",
+			__func__);
+		return -EIO;
+	}
+
+	dev_dbg(pf->dev, "%s: device returned client_id %d for %s\n",
+		__func__, ci, devname);
+
+	return ci;
+}
+EXPORT_SYMBOL_GPL(pds_client_register);
+
+/**
+ * pds_client_unregister - Unlink the client from the firmware
+ * @pf_pdev:	ptr to the PF driver struct
+ * @client_id:	id returned from pds_client_register()
+ *
+ * Return: 0 on success, or
+ *         negative for error
+ */
+int pds_client_unregister(struct pci_dev *pf_pdev, u16 client_id)
+{
+	union pds_core_adminq_comp comp = {};
+	union pds_core_adminq_cmd cmd = {};
+	struct pdsc *pf;
+	int err;
+
+	pf = pci_get_drvdata(pf_pdev);
+	if (pf->state)
+		return -ENXIO;
+
+	cmd.client_unreg.opcode = PDS_AQ_CMD_CLIENT_UNREG;
+	cmd.client_unreg.client_id = cpu_to_le16(client_id);
+
+	err = pdsc_adminq_post(pf, &cmd, &comp, false);
+	if (err)
+		dev_info(pf->dev, "unregister client_id %d failed, status %d: %pe\n",
+			 client_id, comp.status, ERR_PTR(err));
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(pds_client_unregister);
+
+/**
+ * pds_client_adminq_cmd - Process an adminq request for the client
+ * @padev:   ptr to the client device
+ * @req:     ptr to buffer with request
+ * @req_len: length of actual struct used for request
+ * @resp:    ptr to buffer where answer is to be copied
+ * @flags:   optional flags from pds_core_adminq_flags
+ *
+ * Return: 0 on success, or
+ *         negative for error
+ *
+ * Client sends pointers to request and response buffers
+ * Core copies request data into pds_core_client_request_cmd
+ * Core sets other fields as needed
+ * Core posts to AdminQ
+ * Core copies completion data into response buffer
+ */
+int pds_client_adminq_cmd(struct pds_auxiliary_dev *padev,
+			  union pds_core_adminq_cmd *req,
+			  size_t req_len,
+			  union pds_core_adminq_comp *resp,
+			  u64 flags)
+{
+	union pds_core_adminq_cmd cmd = {};
+	struct pci_dev *pf_pdev;
+	struct pdsc *pf;
+	size_t cp_len;
+	int err;
+
+	pf_pdev = pci_physfn(padev->vf_pdev);
+	pf = pci_get_drvdata(pf_pdev);
+
+	dev_dbg(pf->dev, "%s: %s opcode %d\n",
+		__func__, dev_name(&padev->aux_dev.dev), req->opcode);
+
+	if (pf->state)
+		return -ENXIO;
+
+	/* Wrap the client's request */
+	cmd.client_request.opcode = PDS_AQ_CMD_CLIENT_CMD;
+	cmd.client_request.client_id = cpu_to_le16(padev->client_id);
+	cp_len = min_t(size_t, req_len, sizeof(cmd.client_request.client_cmd));
+	memcpy(cmd.client_request.client_cmd, req, cp_len);
+
+	err = pdsc_adminq_post(pf, &cmd, resp,
+			       !!(flags & PDS_AQ_FLAG_FASTPOLL));
+	if (err && err != -EAGAIN)
+		dev_info(pf->dev, "client admin cmd failed: %pe\n",
+			 ERR_PTR(err));
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(pds_client_adminq_cmd);
+
 static void pdsc_auxbus_dev_release(struct device *dev)
 {
 	struct pds_auxiliary_dev *padev =
@@ -16,6 +146,7 @@ static void pdsc_auxbus_dev_release(struct device *dev)
 
 static struct pds_auxiliary_dev *pdsc_auxbus_dev_register(struct pdsc *cf,
 							  struct pdsc *pf,
+							  u16 client_id,
 							  char *name)
 {
 	struct auxiliary_device *aux_dev;
@@ -27,6 +158,7 @@ static struct pds_auxiliary_dev *pdsc_auxbus_dev_register(struct pdsc *cf,
 		return ERR_PTR(-ENOMEM);
 
 	padev->vf_pdev = cf->pdev;
+	padev->client_id = client_id;
 
 	aux_dev = &padev->aux_dev;
 	aux_dev->name = name;
@@ -66,8 +198,10 @@ int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf)
 
 	padev = pf->vfs[cf->vf_id].padev;
 	if (padev) {
+		pds_client_unregister(pf->pdev, padev->client_id);
 		auxiliary_device_delete(&padev->aux_dev);
 		auxiliary_device_uninit(&padev->aux_dev);
+		padev->client_id = 0;
 	}
 	pf->vfs[cf->vf_id].padev = NULL;
 
@@ -79,7 +213,9 @@ int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf)
 {
 	struct pds_auxiliary_dev *padev;
 	enum pds_core_vif_types vt;
+	char devname[PDS_DEVNAME_LEN];
 	u16 vt_support;
+	int client_id;
 	int err = 0;
 
 	mutex_lock(&pf->config_lock);
@@ -101,9 +237,22 @@ int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf)
 	      pf->viftype_status[vt].enabled))
 		goto out_unlock;
 
-	padev = pdsc_auxbus_dev_register(cf, pf,
+	/* Need to register with FW and get the client_id before
+	 * creating the aux device so that the aux client can run
+	 * adminq commands as part its probe
+	 */
+	snprintf(devname, sizeof(devname), "%s.%s.%d",
+		 PDS_CORE_DRV_NAME, pf->viftype_status[vt].name, cf->uid);
+	client_id = pds_client_register(pf->pdev, devname);
+	if (client_id < 0) {
+		err = client_id;
+		goto out_unlock;
+	}
+
+	padev = pdsc_auxbus_dev_register(cf, pf, client_id,
 					 pf->viftype_status[vt].name);
 	if (IS_ERR(padev)) {
+		pds_client_unregister(pf->pdev, client_id);
 		err = PTR_ERR(padev);
 		goto out_unlock;
 	}
diff --git a/include/linux/pds/pds_auxbus.h b/include/linux/pds/pds_auxbus.h
index 493f75b1995e..214ef12302d0 100644
--- a/include/linux/pds/pds_auxbus.h
+++ b/include/linux/pds/pds_auxbus.h
@@ -11,4 +11,10 @@ struct pds_auxiliary_dev {
 	struct pci_dev *vf_pdev;
 	u16 client_id;
 };
+
+int pds_client_adminq_cmd(struct pds_auxiliary_dev *padev,
+			  union pds_core_adminq_cmd *req,
+			  size_t req_len,
+			  union pds_core_adminq_comp *resp,
+			  u64 flags);
 #endif /* _PDSC_AUXBUS_H_ */
diff --git a/include/linux/pds/pds_common.h b/include/linux/pds/pds_common.h
index 961b3d02c69f..4b37675fde3e 100644
--- a/include/linux/pds/pds_common.h
+++ b/include/linux/pds/pds_common.h
@@ -61,4 +61,6 @@ enum pds_core_logical_qtype {
 };
 
 void *pdsc_get_pf_struct(struct pci_dev *vf_pdev);
+int pds_client_register(struct pci_dev *pf_pdev, char *devname);
+int pds_client_unregister(struct pci_dev *pf_pdev, u16 client_id);
 #endif /* _PDS_COMMON_H_ */
-- 
cgit v1.2.3-70-g09d2


From d24c28278a01dc4c80d1470533c667cf406f0e88 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:26 -0700
Subject: pds_core: publish events to the clients

When the Core device gets an event from the device, or notices
the device FW to be up or down, it needs to send those events
on to the clients that have an event handler.  Add the code to
pass along the events to the clients.

The entry points pdsc_register_notify() and pdsc_unregister_notify()
are EXPORTed for other drivers that want to listen for these events.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/pds_core/adminq.c |  2 ++
 drivers/net/ethernet/amd/pds_core/core.c   | 32 ++++++++++++++++++++++++++++++
 drivers/net/ethernet/amd/pds_core/core.h   |  3 +++
 include/linux/pds/pds_common.h             |  2 ++
 4 files changed, 39 insertions(+)

diff --git a/drivers/net/ethernet/amd/pds_core/adminq.c b/drivers/net/ethernet/amd/pds_core/adminq.c
index fb2ba3f62480..045fe133f6ee 100644
--- a/drivers/net/ethernet/amd/pds_core/adminq.c
+++ b/drivers/net/ethernet/amd/pds_core/adminq.c
@@ -29,11 +29,13 @@ static int pdsc_process_notifyq(struct pdsc_qcq *qcq)
 		case PDS_EVENT_LINK_CHANGE:
 			dev_info(pdsc->dev, "NotifyQ LINK_CHANGE ecode %d eid %lld\n",
 				 ecode, eid);
+			pdsc_notify(PDS_EVENT_LINK_CHANGE, comp);
 			break;
 
 		case PDS_EVENT_RESET:
 			dev_info(pdsc->dev, "NotifyQ RESET ecode %d eid %lld\n",
 				 ecode, eid);
+			pdsc_notify(PDS_EVENT_RESET, comp);
 			break;
 
 		case PDS_EVENT_XCVR:
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index b2fca3b99f02..483a070d96fa 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -6,6 +6,25 @@
 
 #include "core.h"
 
+static BLOCKING_NOTIFIER_HEAD(pds_notify_chain);
+
+int pdsc_register_notify(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&pds_notify_chain, nb);
+}
+EXPORT_SYMBOL_GPL(pdsc_register_notify);
+
+void pdsc_unregister_notify(struct notifier_block *nb)
+{
+	blocking_notifier_chain_unregister(&pds_notify_chain, nb);
+}
+EXPORT_SYMBOL_GPL(pdsc_unregister_notify);
+
+void pdsc_notify(unsigned long event, void *data)
+{
+	blocking_notifier_call_chain(&pds_notify_chain, event, data);
+}
+
 void pdsc_intr_free(struct pdsc *pdsc, int index)
 {
 	struct pdsc_intr_info *intr_info;
@@ -494,12 +513,19 @@ void pdsc_stop(struct pdsc *pdsc)
 
 static void pdsc_fw_down(struct pdsc *pdsc)
 {
+	union pds_core_notifyq_comp reset_event = {
+		.reset.ecode = cpu_to_le16(PDS_EVENT_RESET),
+		.reset.state = 0,
+	};
+
 	if (test_and_set_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
 		dev_err(pdsc->dev, "%s: already happening\n", __func__);
 		return;
 	}
 
+	/* Notify clients of fw_down */
 	devlink_health_report(pdsc->fw_reporter, "FW down reported", pdsc);
+	pdsc_notify(PDS_EVENT_RESET, &reset_event);
 
 	pdsc_stop(pdsc);
 	pdsc_teardown(pdsc, PDSC_TEARDOWN_RECOVERY);
@@ -507,6 +533,10 @@ static void pdsc_fw_down(struct pdsc *pdsc)
 
 static void pdsc_fw_up(struct pdsc *pdsc)
 {
+	union pds_core_notifyq_comp reset_event = {
+		.reset.ecode = cpu_to_le16(PDS_EVENT_RESET),
+		.reset.state = 1,
+	};
 	int err;
 
 	if (!test_bit(PDSC_S_FW_DEAD, &pdsc->state)) {
@@ -522,9 +552,11 @@ static void pdsc_fw_up(struct pdsc *pdsc)
 	if (err)
 		goto err_out;
 
+	/* Notify clients of fw_up */
 	pdsc->fw_recoveries++;
 	devlink_health_reporter_state_update(pdsc->fw_reporter,
 					     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
+	pdsc_notify(PDS_EVENT_RESET, &reset_event);
 
 	return;
 
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 9e01a9ee6868..e545fafc4819 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -297,6 +297,9 @@ int pdsc_start(struct pdsc *pdsc);
 void pdsc_stop(struct pdsc *pdsc);
 void pdsc_health_thread(struct work_struct *work);
 
+int pdsc_register_notify(struct notifier_block *nb);
+void pdsc_unregister_notify(struct notifier_block *nb);
+void pdsc_notify(unsigned long event, void *data);
 int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf);
 int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf);
 
diff --git a/include/linux/pds/pds_common.h b/include/linux/pds/pds_common.h
index 4b37675fde3e..060331486d50 100644
--- a/include/linux/pds/pds_common.h
+++ b/include/linux/pds/pds_common.h
@@ -60,6 +60,8 @@ enum pds_core_logical_qtype {
 	PDS_CORE_QTYPE_MAX     = 16   /* don't change - used in struct size */
 };
 
+int pdsc_register_notify(struct notifier_block *nb);
+void pdsc_unregister_notify(struct notifier_block *nb);
 void *pdsc_get_pf_struct(struct pci_dev *vf_pdev);
 int pds_client_register(struct pci_dev *pf_pdev, char *devname);
 int pds_client_unregister(struct pci_dev *pf_pdev, u16 client_id);
-- 
cgit v1.2.3-70-g09d2


From ddbcb22055d136f58841c73ca2226dab79eb6101 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Wed, 19 Apr 2023 10:04:27 -0700
Subject: pds_core: Kconfig and pds_core.rst

Remaining documentation and Kconfig hook for building the driver.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../networking/device_drivers/ethernet/amd/pds_core.rst  | 16 ++++++++++++++++
 MAINTAINERS                                              |  9 +++++++++
 drivers/net/ethernet/amd/Kconfig                         | 12 ++++++++++++
 drivers/net/ethernet/amd/Makefile                        |  1 +
 4 files changed, 38 insertions(+)

diff --git a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
index b9f310de862e..9e8a16c44102 100644
--- a/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
+++ b/Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
@@ -114,6 +114,22 @@ The driver supports a devlink health reporter for FW status::
   # devlink health diagnose pci/0000:2b:00.0 reporter fw
    Status: healthy State: 1 Generation: 0 Recoveries: 0
 
+Enabling the driver
+===================
+
+The driver is enabled via the standard kernel configuration system,
+using the make command::
+
+  make oldconfig/menuconfig/etc.
+
+The driver is located in the menu structure at:
+
+  -> Device Drivers
+    -> Network device support (NETDEVICES [=y])
+      -> Ethernet driver support
+        -> AMD devices
+          -> AMD/Pensando Ethernet PDS_CORE Support
+
 Support
 =======
 
diff --git a/MAINTAINERS b/MAINTAINERS
index 85571d5d8209..9420a1599a58 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1041,6 +1041,15 @@ F:	drivers/gpu/drm/amd/include/vi_structs.h
 F:	include/uapi/linux/kfd_ioctl.h
 F:	include/uapi/linux/kfd_sysfs.h
 
+AMD PDS CORE DRIVER
+M:	Shannon Nelson <shannon.nelson@amd.com>
+M:	Brett Creeley <brett.creeley@amd.com>
+L:	netdev@vger.kernel.org
+S:	Supported
+F:	Documentation/networking/device_drivers/ethernet/amd/pds_core.rst
+F:	drivers/net/ethernet/amd/pds_core/
+F:	include/linux/pds/
+
 AMD SPI DRIVER
 M:	Sanjay R Mehta <sanju.mehta@amd.com>
 S:	Maintained
diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig
index ab42f75b9413..235fcacef5c5 100644
--- a/drivers/net/ethernet/amd/Kconfig
+++ b/drivers/net/ethernet/amd/Kconfig
@@ -186,4 +186,16 @@ config AMD_XGBE_HAVE_ECC
 	bool
 	default n
 
+config PDS_CORE
+	tristate "AMD/Pensando Data Systems Core Device Support"
+	depends on 64BIT && PCI
+	help
+	  This enables the support for the AMD/Pensando Core device family of
+	  adapters.  More specific information on this driver can be
+	  found in
+	  <file:Documentation/networking/device_drivers/ethernet/amd/pds_core.rst>.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called pds_core.
+
 endif # NET_VENDOR_AMD
diff --git a/drivers/net/ethernet/amd/Makefile b/drivers/net/ethernet/amd/Makefile
index 42742afe9115..2dcfb84731e1 100644
--- a/drivers/net/ethernet/amd/Makefile
+++ b/drivers/net/ethernet/amd/Makefile
@@ -17,3 +17,4 @@ obj-$(CONFIG_PCNET32) += pcnet32.o
 obj-$(CONFIG_SUN3LANCE) += sun3lance.o
 obj-$(CONFIG_SUNLANCE) += sunlance.o
 obj-$(CONFIG_AMD_XGBE) += xgbe/
+obj-$(CONFIG_PDS_CORE) += pds_core/
-- 
cgit v1.2.3-70-g09d2


From 35226750f7ab9d49140d95bc7d38a2a9b0f4fdfc Mon Sep 17 00:00:00 2001
From: Yan Wang <rk.code@outlook.com>
Date: Wed, 19 Apr 2023 22:13:46 +0800
Subject: net: stmmac:fix system hang when setting up tag_8021q VLAN for DSA
 ports

The system hang because of dsa_tag_8021q_port_setup()->
				stmmac_vlan_rx_add_vid().

I found in stmmac_drv_probe() that cailing pm_runtime_put()
disabled the clock.

First, when the kernel is compiled with CONFIG_PM=y,The stmmac's
resume/suspend is active.

Secondly,stmmac as DSA master,the dsa_tag_8021q_port_setup() function
will callback stmmac_vlan_rx_add_vid when DSA dirver starts. However,
The system is hanged for the stmmac_vlan_rx_add_vid() accesses its
registers after stmmac's clock is closed.

I would suggest adding the pm_runtime_resume_and_get() to the
stmmac_vlan_rx_add_vid().This guarantees that resuming clock output
while in use.

Fixes: b3dcb3127786 ("net: stmmac: correct clocks enabled in stmmac_vlan_rx_kill_vid()")
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Yan Wang <rk.code@outlook.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 47534310365a..f116e4ae293b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -6377,6 +6377,10 @@ static int stmmac_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid
 	bool is_double = false;
 	int ret;
 
+	ret = pm_runtime_resume_and_get(priv->device);
+	if (ret < 0)
+		return ret;
+
 	if (be16_to_cpu(proto) == ETH_P_8021AD)
 		is_double = true;
 
@@ -6384,16 +6388,18 @@ static int stmmac_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid
 	ret = stmmac_vlan_update(priv, is_double);
 	if (ret) {
 		clear_bit(vid, priv->active_vlans);
-		return ret;
+		goto err_pm_put;
 	}
 
 	if (priv->hw->num_vlan) {
 		ret = stmmac_add_hw_vlan_rx_fltr(priv, ndev, priv->hw, proto, vid);
 		if (ret)
-			return ret;
+			goto err_pm_put;
 	}
+err_pm_put:
+	pm_runtime_put(priv->device);
 
-	return 0;
+	return ret;
 }
 
 static int stmmac_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid)
-- 
cgit v1.2.3-70-g09d2


From e239e31ae8026b85e225941eb3f2cd7deb24f0d1 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 20 Apr 2023 11:02:47 +0300
Subject: net/mlx5e: Fix FW error while setting IPsec policy block action

When trying to set IPsec policy block action the following error is
generated:

 mlx5_cmd_out_err:803:(pid 3426): SET_FLOW_TABLE_ENTRY(0x936) op_mod(0x0) failed,
	status bad parameter(0x3), syndrome (0x8708c3), err(-22)

This error means that drop action is not allowed when modify action is
set, so update the code to skip modify header for XFRM_POLICY_BLOCK action.

Fixes: 6721239672fe ("net/mlx5e: Skip IPsec encryption for TX path without matching policy")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c  | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 5a8fcd30fcb1..dbe87bf89c0d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -1252,16 +1252,16 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	setup_fte_no_frags(spec);
 	setup_fte_upper_proto_match(spec, &attrs->upspec);
 
-	if (attrs->reqid) {
+	switch (attrs->action) {
+	case XFRM_POLICY_ALLOW:
+		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+		if (!attrs->reqid)
+			break;
+
 		err = setup_modify_header(mdev, attrs->reqid,
 					  XFRM_DEV_OFFLOAD_OUT, &flow_act);
 		if (err)
 			goto err_mod_header;
-	}
-
-	switch (attrs->action) {
-	case XFRM_POLICY_ALLOW:
-		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 		break;
 	case XFRM_POLICY_BLOCK:
 		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_DROP |
@@ -1273,7 +1273,7 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	default:
 		WARN_ON(true);
 		err = -EINVAL;
-		goto err_action;
+		goto err_mod_header;
 	}
 
 	flow_act.flags |= FLOW_ACT_NO_APPEND;
@@ -1293,7 +1293,7 @@ static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry)
 	return 0;
 
 err_action:
-	if (attrs->reqid)
+	if (flow_act.modify_hdr)
 		mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr);
 err_mod_header:
 	kvfree(spec);
-- 
cgit v1.2.3-70-g09d2


From 697b3518ebfd1d0ccae455d8ac1ac9c88884f4e5 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 20 Apr 2023 11:02:48 +0300
Subject: net/mlx5e: Don't overwrite extack message returned from IPsec SA
 validator

Addition of new err_xfrm label caused to error messages be overwritten.
Fix it by using proper NL_SET_ERR_MSG_WEAK_MOD macro together with change
in a default message.

Fixes: aa8bd0c9518c ("net/mlx5e: Support IPsec acquire default SA")
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 5fd609d1120e..1547d8cda133 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -712,7 +712,7 @@ release_work:
 	kfree(sa_entry->work);
 err_xfrm:
 	kfree(sa_entry);
-	NL_SET_ERR_MSG_MOD(extack, "Device failed to offload this policy");
+	NL_SET_ERR_MSG_WEAK_MOD(extack, "Device failed to offload this state");
 	return err;
 }
 
-- 
cgit v1.2.3-70-g09d2


From 3198ae7d42af844a7d30f6e432c079c134b03852 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 20 Apr 2023 11:02:49 +0300
Subject: net/mlx5e: Compare all fields in IPv6 address

Fix size argument in memcmp to compare whole IPv6 address.

Fixes: b3beba1fb404 ("net/mlx5e: Allow policies with reqid 0, to support IKE policy holes")
Reviewed-by: Raed Salem <raeds@nvidia.com>
Reviewed-by: Emeel Hakim <ehakim@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index f7f7c09d2b32..4e9887171508 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -287,7 +287,7 @@ static inline bool addr6_all_zero(__be32 *addr6)
 {
 	static const __be32 zaddr6[4] = {};
 
-	return !memcmp(addr6, zaddr6, sizeof(*zaddr6));
+	return !memcmp(addr6, zaddr6, sizeof(zaddr6));
 }
 #else
 static inline void mlx5e_ipsec_init(struct mlx5e_priv *priv)
-- 
cgit v1.2.3-70-g09d2


From 94edec448479e302b7c7e18f40cb0d7f32e71682 Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 20 Apr 2023 11:02:50 +0300
Subject: net/mlx5e: Properly release work data structure

There are some flows in which work structure is not allocated at all
and it is needed to be checked prior release of data structure.

 general protection fault, probably for non-canonical address 0xdffffc000000000a: 0000 [#1] SMP KASAN
 KASAN: null-ptr-deref in range [0x0000000000000050-0x0000000000000057]
 CPU: 6 PID: 3486 Comm: kworker/6:0 Not tainted 6.3.0-rc5_for_upstream_debug_2023_04_06_11_01 #1
 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
 Workqueue: events xfrm_state_gc_task
 RIP: 0010:mlx5e_xfrm_free_state+0x177/0x260 [mlx5_core]
 Code: c1 ea 03 80 3c 02 00 0f 85 f5 00 00 00 4c 8b a5 08 01 00 00 48 b8 00 00 00 00 00 fc ff df 49 8d 7c 24 50 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 b7 00 00 00 49 8b 7c 24 50 e8 85 7c 09 e0 4c 89
 RSP: 0018:ffff888137a8fc50 EFLAGS: 00010206
 RAX: dffffc0000000000 RBX: ffff888180398000 RCX: 0000000000000000
 RDX: 000000000000000a RSI: ffffffffa1878227 RDI: 0000000000000050
 RBP: ffff88812a0c8000 R08: ffff888137a8fb60 R09: 0000000000000000
 R10: fffffbfff09aba0c R11: 0000000000000001 R12: 0000000000000000
 R13: ffff88812a0c8108 R14: ffffffff84c63480 R15: ffff8881acb63118
 FS:  0000000000000000(0000) GS:ffff88881eb00000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 00007f667e8bc000 CR3: 0000000004693006 CR4: 0000000000370ea0
 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
 Call Trace:

  ___xfrm_state_destroy+0x3c8/0x5e0
  xfrm_state_gc_task+0xf6/0x140
  ? ___xfrm_state_destroy+0x5e0/0x5e0
  process_one_work+0x7c2/0x1340
  ? lockdep_hardirqs_on_prepare+0x3f0/0x3f0
  ? pwq_dec_nr_in_flight+0x230/0x230
  ? spin_bug+0x1d0/0x1d0
  worker_thread+0x59d/0xec0
  ? __kthread_parkme+0xd9/0x1d0
  ? process_one_work+0x1340/0x1340
  kthread+0x28f/0x330
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x1f/0x30

 Modules linked in: sch_ingress openvswitch nsh mlx5_vdpa vringh vhost_iotlb vdpa mlx5_ib mlx5_core xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi rdma_cm iw_cm ib_umad ib_ipoib ib_cm ib_uverbs ib_core vfio_pci vfio_pci_core vfio_iommu_type1 vfio cuse overlay zram zsmalloc fuse [last unloaded: mlx5_core]
 ---[ end trace 0000000000000000 ]---

Fixes: 4562116f8a56 ("net/mlx5e: Generalize IPsec work structs")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 1547d8cda133..59b9927ac90f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -708,7 +708,8 @@ err_hw_ctx:
 release_dwork:
 	kfree(sa_entry->dwork);
 release_work:
-	kfree(sa_entry->work->data);
+	if (sa_entry->work)
+		kfree(sa_entry->work->data);
 	kfree(sa_entry->work);
 err_xfrm:
 	kfree(sa_entry);
@@ -752,7 +753,8 @@ static void mlx5e_xfrm_free_state(struct xfrm_state *x)
 	mlx5e_accel_ipsec_fs_del_rule(sa_entry);
 	mlx5_ipsec_free_sa_ctx(sa_entry);
 	kfree(sa_entry->dwork);
-	kfree(sa_entry->work->data);
+	if (sa_entry->work)
+		kfree(sa_entry->work->data);
 	kfree(sa_entry->work);
 sa_entry_free:
 	kfree(sa_entry);
-- 
cgit v1.2.3-70-g09d2


From 45fd01f2fbf1119d083931b095ad6d0f13443d0e Mon Sep 17 00:00:00 2001
From: Leon Romanovsky <leonro@nvidia.com>
Date: Thu, 20 Apr 2023 11:02:51 +0300
Subject: net/mlx5e: Refactor duplicated code in mlx5e_ipsec_init_macs

ARP discovery code has same logic for RX and TX flows, but with
different source and destination fields. Instead of duplicating
same code in mlx5e_ipsec_init_macs, let's refactor.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 45 ++++++++++------------
 1 file changed, 21 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
index 59b9927ac90f..55b38544422f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c
@@ -252,6 +252,8 @@ static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	struct net_device *netdev;
 	struct neighbour *n;
 	u8 addr[ETH_ALEN];
+	const void *pkey;
+	u8 *dst, *src;
 
 	if (attrs->mode != XFRM_MODE_TUNNEL ||
 	    attrs->type != XFRM_DEV_OFFLOAD_PACKET)
@@ -262,36 +264,31 @@ static void mlx5e_ipsec_init_macs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	mlx5_query_mac_address(mdev, addr);
 	switch (attrs->dir) {
 	case XFRM_DEV_OFFLOAD_IN:
-		ether_addr_copy(attrs->dmac, addr);
-		n = neigh_lookup(&arp_tbl, &attrs->saddr.a4, netdev);
-		if (!n) {
-			n = neigh_create(&arp_tbl, &attrs->saddr.a4, netdev);
-			if (IS_ERR(n))
-				return;
-			neigh_event_send(n, NULL);
-			attrs->drop = true;
-			break;
-		}
-		neigh_ha_snapshot(addr, n, netdev);
-		ether_addr_copy(attrs->smac, addr);
+		src = attrs->dmac;
+		dst = attrs->smac;
+		pkey = &attrs->saddr.a4;
 		break;
 	case XFRM_DEV_OFFLOAD_OUT:
-		ether_addr_copy(attrs->smac, addr);
-		n = neigh_lookup(&arp_tbl, &attrs->daddr.a4, netdev);
-		if (!n) {
-			n = neigh_create(&arp_tbl, &attrs->daddr.a4, netdev);
-			if (IS_ERR(n))
-				return;
-			neigh_event_send(n, NULL);
-			attrs->drop = true;
-			break;
-		}
-		neigh_ha_snapshot(addr, n, netdev);
-		ether_addr_copy(attrs->dmac, addr);
+		src = attrs->smac;
+		dst = attrs->dmac;
+		pkey = &attrs->daddr.a4;
 		break;
 	default:
 		return;
 	}
+
+	ether_addr_copy(src, addr);
+	n = neigh_lookup(&arp_tbl, pkey, netdev);
+	if (!n) {
+		n = neigh_create(&arp_tbl, pkey, netdev);
+		if (IS_ERR(n))
+			return;
+		neigh_event_send(n, NULL);
+		attrs->drop = true;
+	} else {
+		neigh_ha_snapshot(addr, n, netdev);
+		ether_addr_copy(dst, addr);
+	}
 	neigh_release(n);
 }
 
-- 
cgit v1.2.3-70-g09d2


From dfc39d4026fb2432363c0f77543c4cf3adca4c7b Mon Sep 17 00:00:00 2001
From: Jianfeng Tan <henry.tjf@antgroup.com>
Date: Wed, 19 Apr 2023 15:24:16 +0800
Subject: net/packet: support mergeable feature of virtio

Packet sockets, like tap, can be used as the backend for kernel vhost.
In packet sockets, virtio net header size is currently hardcoded to be
the size of struct virtio_net_hdr, which is 10 bytes; however, it is not
always the case: some virtio features, such as mrg_rxbuf, need virtio
net header to be 12-byte long.

Mergeable buffers, as a virtio feature, is worthy of supporting: packets
that are larger than one-mbuf size will be dropped in vhost worker's
handle_rx if mrg_rxbuf feature is not used, but large packets
cannot be avoided and increasing mbuf's size is not economical.

With this virtio feature enabled by virtio-user, packet sockets with
hardcoded 10-byte virtio net header will parse mac head incorrectly in
packet_snd by taking the last two bytes of virtio net header as part of
mac header.
This incorrect mac header parsing will cause packet to be dropped due to
invalid ether head checking in later under-layer device packet receiving.

By adding extra field vnet_hdr_sz with utilizing holes in struct
packet_sock to record currently used virtio net header size and supporting
extra sockopt PACKET_VNET_HDR_SZ to set specified vnet_hdr_sz, packet
sockets can know the exact length of virtio net header that virtio user
gives.
In packet_snd, tpacket_snd and packet_recvmsg, instead of using
hardcoded virtio net header size, it can get the exact vnet_hdr_sz from
corresponding packet_sock, and parse mac header correctly based on this
information to avoid the packets being mistakenly dropped.

Signed-off-by: Jianfeng Tan <henry.tjf@antgroup.com>
Co-developed-by: Anqi Shen <amy.saq@antgroup.com>
Signed-off-by: Anqi Shen <amy.saq@antgroup.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/if_packet.h |  1 +
 net/packet/af_packet.c         | 95 +++++++++++++++++++++++++-----------------
 net/packet/diag.c              |  2 +-
 net/packet/internal.h          |  2 +-
 4 files changed, 60 insertions(+), 40 deletions(-)

diff --git a/include/uapi/linux/if_packet.h b/include/uapi/linux/if_packet.h
index 78c981d6a9d4..9efc42382fdb 100644
--- a/include/uapi/linux/if_packet.h
+++ b/include/uapi/linux/if_packet.h
@@ -59,6 +59,7 @@ struct sockaddr_ll {
 #define PACKET_ROLLOVER_STATS		21
 #define PACKET_FANOUT_DATA		22
 #define PACKET_IGNORE_OUTGOING		23
+#define PACKET_VNET_HDR_SZ		24
 
 #define PACKET_FANOUT_HASH		0
 #define PACKET_FANOUT_LB		1
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 568f8d76e3c1..6080c0db0814 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2090,18 +2090,18 @@ static unsigned int run_filter(struct sk_buff *skb,
 }
 
 static int packet_rcv_vnet(struct msghdr *msg, const struct sk_buff *skb,
-			   size_t *len)
+			   size_t *len, int vnet_hdr_sz)
 {
-	struct virtio_net_hdr vnet_hdr;
+	struct virtio_net_hdr_mrg_rxbuf vnet_hdr = { .num_buffers = 0 };
 
-	if (*len < sizeof(vnet_hdr))
+	if (*len < vnet_hdr_sz)
 		return -EINVAL;
-	*len -= sizeof(vnet_hdr);
+	*len -= vnet_hdr_sz;
 
-	if (virtio_net_hdr_from_skb(skb, &vnet_hdr, vio_le(), true, 0))
+	if (virtio_net_hdr_from_skb(skb, (struct virtio_net_hdr *)&vnet_hdr, vio_le(), true, 0))
 		return -EINVAL;
 
-	return memcpy_to_msg(msg, (void *)&vnet_hdr, sizeof(vnet_hdr));
+	return memcpy_to_msg(msg, (void *)&vnet_hdr, vnet_hdr_sz);
 }
 
 /*
@@ -2250,7 +2250,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 	__u32 ts_status;
 	bool is_drop_n_account = false;
 	unsigned int slot_id = 0;
-	bool do_vnet = false;
+	int vnet_hdr_sz = 0;
 
 	/* struct tpacket{2,3}_hdr is aligned to a multiple of TPACKET_ALIGNMENT.
 	 * We may add members to them until current aligned size without forcing
@@ -2308,10 +2308,9 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		netoff = TPACKET_ALIGN(po->tp_hdrlen +
 				       (maclen < 16 ? 16 : maclen)) +
 				       po->tp_reserve;
-		if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
-			netoff += sizeof(struct virtio_net_hdr);
-			do_vnet = true;
-		}
+		vnet_hdr_sz = READ_ONCE(po->vnet_hdr_sz);
+		if (vnet_hdr_sz)
+			netoff += vnet_hdr_sz;
 		macoff = netoff - maclen;
 	}
 	if (netoff > USHRT_MAX) {
@@ -2337,7 +2336,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 			snaplen = po->rx_ring.frame_size - macoff;
 			if ((int)snaplen < 0) {
 				snaplen = 0;
-				do_vnet = false;
+				vnet_hdr_sz = 0;
 			}
 		}
 	} else if (unlikely(macoff + snaplen >
@@ -2351,7 +2350,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		if (unlikely((int)snaplen < 0)) {
 			snaplen = 0;
 			macoff = GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len;
-			do_vnet = false;
+			vnet_hdr_sz = 0;
 		}
 	}
 	spin_lock(&sk->sk_receive_queue.lock);
@@ -2367,7 +2366,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		__set_bit(slot_id, po->rx_ring.rx_owner_map);
 	}
 
-	if (do_vnet &&
+	if (vnet_hdr_sz &&
 	    virtio_net_hdr_from_skb(skb, h.raw + macoff -
 				    sizeof(struct virtio_net_hdr),
 				    vio_le(), true, 0)) {
@@ -2551,16 +2550,26 @@ static int __packet_snd_vnet_parse(struct virtio_net_hdr *vnet_hdr, size_t len)
 }
 
 static int packet_snd_vnet_parse(struct msghdr *msg, size_t *len,
-				 struct virtio_net_hdr *vnet_hdr)
+				 struct virtio_net_hdr *vnet_hdr, int vnet_hdr_sz)
 {
-	if (*len < sizeof(*vnet_hdr))
+	int ret;
+
+	if (*len < vnet_hdr_sz)
 		return -EINVAL;
-	*len -= sizeof(*vnet_hdr);
+	*len -= vnet_hdr_sz;
 
 	if (!copy_from_iter_full(vnet_hdr, sizeof(*vnet_hdr), &msg->msg_iter))
 		return -EFAULT;
 
-	return __packet_snd_vnet_parse(vnet_hdr, *len);
+	ret = __packet_snd_vnet_parse(vnet_hdr, *len);
+	if (ret)
+		return ret;
+
+	/* move iter to point to the start of mac header */
+	if (vnet_hdr_sz != sizeof(struct virtio_net_hdr))
+		iov_iter_advance(&msg->msg_iter, vnet_hdr_sz - sizeof(struct virtio_net_hdr));
+
+	return 0;
 }
 
 static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
@@ -2722,6 +2731,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	void *ph;
 	DECLARE_SOCKADDR(struct sockaddr_ll *, saddr, msg->msg_name);
 	bool need_wait = !(msg->msg_flags & MSG_DONTWAIT);
+	int vnet_hdr_sz = READ_ONCE(po->vnet_hdr_sz);
 	unsigned char *addr = NULL;
 	int tp_len, size_max;
 	void *data;
@@ -2779,8 +2789,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	size_max = po->tx_ring.frame_size
 		- (po->tp_hdrlen - sizeof(struct sockaddr_ll));
 
-	if ((size_max > dev->mtu + reserve + VLAN_HLEN) &&
-	    !packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR))
+	if ((size_max > dev->mtu + reserve + VLAN_HLEN) && !vnet_hdr_sz)
 		size_max = dev->mtu + reserve + VLAN_HLEN;
 
 	reinit_completion(&po->skb_completion);
@@ -2809,10 +2818,10 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 		status = TP_STATUS_SEND_REQUEST;
 		hlen = LL_RESERVED_SPACE(dev);
 		tlen = dev->needed_tailroom;
-		if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
+		if (vnet_hdr_sz) {
 			vnet_hdr = data;
-			data += sizeof(*vnet_hdr);
-			tp_len -= sizeof(*vnet_hdr);
+			data += vnet_hdr_sz;
+			tp_len -= vnet_hdr_sz;
 			if (tp_len < 0 ||
 			    __packet_snd_vnet_parse(vnet_hdr, tp_len)) {
 				tp_len = -EINVAL;
@@ -2837,7 +2846,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 					  addr, hlen, copylen, &sockc);
 		if (likely(tp_len >= 0) &&
 		    tp_len > dev->mtu + reserve &&
-		    !packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR) &&
+		    !vnet_hdr_sz &&
 		    !packet_extra_vlan_len_allowed(dev, skb))
 			tp_len = -EMSGSIZE;
 
@@ -2856,7 +2865,7 @@ tpacket_error:
 			}
 		}
 
-		if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
+		if (vnet_hdr_sz) {
 			if (virtio_net_hdr_to_skb(skb, vnet_hdr, vio_le())) {
 				tp_len = -EINVAL;
 				goto tpacket_error;
@@ -2946,7 +2955,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	struct virtio_net_hdr vnet_hdr = { 0 };
 	int offset = 0;
 	struct packet_sock *po = pkt_sk(sk);
-	bool has_vnet_hdr = false;
+	int vnet_hdr_sz = READ_ONCE(po->vnet_hdr_sz);
 	int hlen, tlen, linear;
 	int extra_len = 0;
 
@@ -2990,11 +2999,10 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 
 	if (sock->type == SOCK_RAW)
 		reserve = dev->hard_header_len;
-	if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) {
-		err = packet_snd_vnet_parse(msg, &len, &vnet_hdr);
+	if (vnet_hdr_sz) {
+		err = packet_snd_vnet_parse(msg, &len, &vnet_hdr, vnet_hdr_sz);
 		if (err)
 			goto out_unlock;
-		has_vnet_hdr = true;
 	}
 
 	if (unlikely(sock_flag(sk, SOCK_NOFCS))) {
@@ -3064,11 +3072,11 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 
 	packet_parse_headers(skb, sock);
 
-	if (has_vnet_hdr) {
+	if (vnet_hdr_sz) {
 		err = virtio_net_hdr_to_skb(skb, &vnet_hdr, vio_le());
 		if (err)
 			goto out_free;
-		len += sizeof(vnet_hdr);
+		len += vnet_hdr_sz;
 		virtio_net_hdr_set_proto(skb, &vnet_hdr);
 	}
 
@@ -3408,7 +3416,7 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	struct sock *sk = sock->sk;
 	struct sk_buff *skb;
 	int copied, err;
-	int vnet_hdr_len = 0;
+	int vnet_hdr_len = READ_ONCE(pkt_sk(sk)->vnet_hdr_sz);
 	unsigned int origlen = 0;
 
 	err = -EINVAL;
@@ -3449,11 +3457,10 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 
 	packet_rcv_try_clear_pressure(pkt_sk(sk));
 
-	if (packet_sock_flag(pkt_sk(sk), PACKET_SOCK_HAS_VNET_HDR)) {
-		err = packet_rcv_vnet(msg, skb, &len);
+	if (vnet_hdr_len) {
+		err = packet_rcv_vnet(msg, skb, &len, vnet_hdr_len);
 		if (err)
 			goto out_free;
-		vnet_hdr_len = sizeof(struct virtio_net_hdr);
 	}
 
 	/* You lose any data beyond the buffer you gave. If it worries
@@ -3915,8 +3922,9 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		return 0;
 	}
 	case PACKET_VNET_HDR:
+	case PACKET_VNET_HDR_SZ:
 	{
-		int val;
+		int val, hdr_len;
 
 		if (sock->type != SOCK_RAW)
 			return -EINVAL;
@@ -3925,11 +3933,19 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval,
 		if (copy_from_sockptr(&val, optval, sizeof(val)))
 			return -EFAULT;
 
+		if (optname == PACKET_VNET_HDR_SZ) {
+			if (val && val != sizeof(struct virtio_net_hdr) &&
+			    val != sizeof(struct virtio_net_hdr_mrg_rxbuf))
+				return -EINVAL;
+			hdr_len = val;
+		} else {
+			hdr_len = val ? sizeof(struct virtio_net_hdr) : 0;
+		}
 		lock_sock(sk);
 		if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) {
 			ret = -EBUSY;
 		} else {
-			packet_sock_flag_set(po, PACKET_SOCK_HAS_VNET_HDR, val);
+			WRITE_ONCE(po->vnet_hdr_sz, hdr_len);
 			ret = 0;
 		}
 		release_sock(sk);
@@ -4062,7 +4078,10 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		val = packet_sock_flag(po, PACKET_SOCK_ORIGDEV);
 		break;
 	case PACKET_VNET_HDR:
-		val = packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR);
+		val = !!READ_ONCE(po->vnet_hdr_sz);
+		break;
+	case PACKET_VNET_HDR_SZ:
+		val = READ_ONCE(po->vnet_hdr_sz);
 		break;
 	case PACKET_VERSION:
 		val = po->tp_version;
diff --git a/net/packet/diag.c b/net/packet/diag.c
index de4ced5cf3e8..d0c4eda4cdc6 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -27,7 +27,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
 		pinfo.pdi_flags |= PDI_AUXDATA;
 	if (packet_sock_flag(po, PACKET_SOCK_ORIGDEV))
 		pinfo.pdi_flags |= PDI_ORIGDEV;
-	if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR))
+	if (READ_ONCE(po->vnet_hdr_sz))
 		pinfo.pdi_flags |= PDI_VNETHDR;
 	if (packet_sock_flag(po, PACKET_SOCK_TP_LOSS))
 		pinfo.pdi_flags |= PDI_LOSS;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index 27930f69f368..63f4865202c1 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -118,6 +118,7 @@ struct packet_sock {
 	struct mutex		pg_vec_lock;
 	unsigned long		flags;
 	int			ifindex;	/* bound device		*/
+	u8			vnet_hdr_sz;
 	__be16			num;
 	struct packet_rollover	*rollover;
 	struct packet_mclist	*mclist;
@@ -139,7 +140,6 @@ enum packet_sock_flags {
 	PACKET_SOCK_AUXDATA,
 	PACKET_SOCK_TX_HAS_OFF,
 	PACKET_SOCK_TP_LOSS,
-	PACKET_SOCK_HAS_VNET_HDR,
 	PACKET_SOCK_RUNNING,
 	PACKET_SOCK_PRESSURE,
 	PACKET_SOCK_QDISC_BYPASS,
-- 
cgit v1.2.3-70-g09d2


From acf1c3d68e9a31f10d92bc67ad4673cdae5e8d92 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Thu, 20 Apr 2023 18:49:01 -0700
Subject: bpf: Fix race between btf_put and btf_idr walk.

Florian and Eduard reported hard dead lock:
[   58.433327]  _raw_spin_lock_irqsave+0x40/0x50
[   58.433334]  btf_put+0x43/0x90
[   58.433338]  bpf_find_btf_id+0x157/0x240
[   58.433353]  btf_parse_fields+0x921/0x11c0

This happens since btf->refcount can be 1 at the time of btf_put() and
btf_put() will call btf_free_id() which will try to grab btf_idr_lock
and will dead lock.
Avoid the issue by doing btf_put() without locking.

Fixes: 3d78417b60fb ("bpf: Add bpf_btf_find_by_name_kind() helper.")
Fixes: 1e89106da253 ("bpf: Add bpf_core_add_cands() and wire it into bpf_core_apply_relo_insn().")
Reported-by: Florian Westphal <fw@strlen.de>
Reported-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20230421014901.70908-1-alexei.starovoitov@gmail.com
---
 kernel/bpf/btf.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index a0887ee44e89..7db4ec125fbd 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -577,8 +577,8 @@ static s32 bpf_find_btf_id(const char *name, u32 kind, struct btf **btf_p)
 			*btf_p = btf;
 			return ret;
 		}
-		spin_lock_bh(&btf_idr_lock);
 		btf_put(btf);
+		spin_lock_bh(&btf_idr_lock);
 	}
 	spin_unlock_bh(&btf_idr_lock);
 	return ret;
@@ -8354,12 +8354,10 @@ check_modules:
 		btf_get(mod_btf);
 		spin_unlock_bh(&btf_idr_lock);
 		cands = bpf_core_add_cands(cands, mod_btf, btf_nr_types(main_btf));
-		if (IS_ERR(cands)) {
-			btf_put(mod_btf);
+		btf_put(mod_btf);
+		if (IS_ERR(cands))
 			return ERR_CAST(cands);
-		}
 		spin_lock_bh(&btf_idr_lock);
-		btf_put(mod_btf);
 	}
 	spin_unlock_bh(&btf_idr_lock);
 	/* cands is a pointer to kmalloced memory here if cands->cnt > 0
-- 
cgit v1.2.3-70-g09d2


From 4ab07209d5cc8cb6d2a5324c07b3efc3b2fde494 Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Fri, 21 Apr 2023 00:44:31 -0700
Subject: bpf: Fix bpf_refcount_acquire's refcount_t address calculation

When calculating the address of the refcount_t struct within a local
kptr, bpf_refcount_acquire_impl should add refcount_off bytes to the
address of the local kptr. Due to some missing parens, the function is
incorrectly adding sizeof(refcount_t) * refcount_off bytes. This patch
fixes the calculation.

Due to the incorrect calculation, bpf_refcount_acquire_impl was trying
to refcount_inc some memory well past the end of local kptrs, resulting
in kasan and refcount complaints, as reported in [0]. In that thread,
Florian and Eduard discovered that bpf selftests written in the new
style - with __success and an expected __retval, specifically - were
not actually being run. As a result, selftests added in bpf_refcount
series weren't really exercising this behavior, and thus didn't unearth
the bug.

With this fixed behavior it's safe to revert commit 7c4b96c00043
("selftests/bpf: disable program test run for progs/refcounted_kptr.c"),
this patch does so.

  [0] https://lore.kernel.org/bpf/ZEEp+j22imoN6rn9@strlen.de/

Fixes: 7c50b1cb76ac ("bpf: Add bpf_refcount_acquire kfunc")
Reported-by: Florian Westphal <fw@strlen.de>
Reported-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20230421074431.3548349-1-davemarchevsky@fb.com
---
 kernel/bpf/helpers.c                                | 2 +-
 tools/testing/selftests/bpf/progs/refcounted_kptr.c | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 00e5fb0682ac..8d368fa353f9 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1925,7 +1925,7 @@ __bpf_kfunc void *bpf_refcount_acquire_impl(void *p__refcounted_kptr, void *meta
 	/* Could just cast directly to refcount_t *, but need some code using
 	 * bpf_refcount type so that it is emitted in vmlinux BTF
 	 */
-	ref = (struct bpf_refcount *)p__refcounted_kptr + meta->record->refcount_off;
+	ref = (struct bpf_refcount *)(p__refcounted_kptr + meta->record->refcount_off);
 
 	refcount_inc((refcount_t *)ref);
 	return (void *)p__refcounted_kptr;
diff --git a/tools/testing/selftests/bpf/progs/refcounted_kptr.c b/tools/testing/selftests/bpf/progs/refcounted_kptr.c
index b6b2d4f97b19..1d348a225140 100644
--- a/tools/testing/selftests/bpf/progs/refcounted_kptr.c
+++ b/tools/testing/selftests/bpf/progs/refcounted_kptr.c
@@ -219,7 +219,7 @@ static long __read_from_unstash(int idx)
 #define INSERT_READ_BOTH(rem_tree, rem_list, desc)			\
 SEC("tc")								\
 __description(desc)							\
-__success /* __retval(579) temporarily disabled */			\
+__success __retval(579)							\
 long insert_and_remove_tree_##rem_tree##_list_##rem_list(void *ctx)	\
 {									\
 	long err, tree_data, list_data;					\
@@ -258,7 +258,7 @@ INSERT_READ_BOTH(false, true, "insert_read_both: remove from list");
 #define INSERT_READ_BOTH(rem_tree, rem_list, desc)			\
 SEC("tc")								\
 __description(desc)							\
-__success /* __retval(579) temporarily disabled */			\
+__success __retval(579)							\
 long insert_and_remove_lf_tree_##rem_tree##_list_##rem_list(void *ctx)	\
 {									\
 	long err, tree_data, list_data;					\
@@ -296,7 +296,7 @@ INSERT_READ_BOTH(false, true, "insert_read_both_list_first: remove from list");
 #define INSERT_DOUBLE_READ_AND_DEL(read_fn, read_root, desc)		\
 SEC("tc")								\
 __description(desc)							\
-__success /* temporarily __retval(-1) disabled */			\
+__success __retval(-1)							\
 long insert_double_##read_fn##_and_del_##read_root(void *ctx)		\
 {									\
 	long err, list_data;						\
@@ -329,7 +329,7 @@ INSERT_DOUBLE_READ_AND_DEL(__read_from_list, head, "insert_double_del: 2x read-a
 #define INSERT_STASH_READ(rem_tree, desc)				\
 SEC("tc")								\
 __description(desc)							\
-__success /* __retval(84) temporarily disabled */			\
+__success __retval(84)							\
 long insert_rbtree_and_stash__del_tree_##rem_tree(void *ctx)		\
 {									\
 	long err, tree_data, map_data;					\
-- 
cgit v1.2.3-70-g09d2


From 02e93e0475df21a1091a9833c52c0e241586573a Mon Sep 17 00:00:00 2001
From: Magnus Karlsson <magnus.karlsson@intel.com>
Date: Fri, 21 Apr 2023 08:22:08 +0200
Subject: selftests/xsk: Put MAP_HUGE_2MB in correct argument

Put the flag MAP_HUGE_2MB in the correct flags argument instead of the
wrong offset argument.

Fixes: 2ddade322925 ("selftests/xsk: Fix munmap for hugepage allocated umem")
Reported-by: Kal Cutter Conley <kal.conley@dectris.com>
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230421062208.3772-1-magnus.karlsson@gmail.com
---
 tools/testing/selftests/bpf/xskxceiver.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index a59d04118842..f144d0604ddf 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -1287,19 +1287,16 @@ static void thread_common_ops(struct test_spec *test, struct ifobject *ifobject)
 	u64 umem_sz = ifobject->umem->num_frames * ifobject->umem->frame_size;
 	int mmap_flags = MAP_PRIVATE | MAP_ANONYMOUS | MAP_NORESERVE;
 	LIBBPF_OPTS(bpf_xdp_query_opts, opts);
-	off_t mmap_offset = 0;
 	void *bufs;
 	int ret;
 
-	if (ifobject->umem->unaligned_mode) {
-		mmap_flags |= MAP_HUGETLB;
-		mmap_offset = MAP_HUGE_2MB;
-	}
+	if (ifobject->umem->unaligned_mode)
+		mmap_flags |= MAP_HUGETLB | MAP_HUGE_2MB;
 
 	if (ifobject->shared_umem)
 		umem_sz *= 2;
 
-	bufs = mmap(NULL, umem_sz, PROT_READ | PROT_WRITE, mmap_flags, -1, mmap_offset);
+	bufs = mmap(NULL, umem_sz, PROT_READ | PROT_WRITE, mmap_flags, -1, 0);
 	if (bufs == MAP_FAILED)
 		exit_with_error(errno);
 
@@ -1633,7 +1630,7 @@ static bool hugepages_present(struct ifobject *ifobject)
 	void *bufs;
 
 	bufs = mmap(NULL, mmap_sz, PROT_READ | PROT_WRITE,
-		    MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, -1, MAP_HUGE_2MB);
+		    MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB | MAP_HUGE_2MB, -1, 0);
 	if (bufs == MAP_FAILED)
 		return false;
 
-- 
cgit v1.2.3-70-g09d2


From 00e74ae0863827d944e36e56a4ce1e77e50edb91 Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Tue, 18 Apr 2023 15:53:38 -0700
Subject: bpf: Don't EFAULT for getsockopt with optval=NULL

Some socket options do getsockopt with optval=NULL to estimate the size
of the final buffer (which is returned via optlen). This breaks BPF
getsockopt assumptions about permitted optval buffer size. Let's enforce
these assumptions only when non-NULL optval is provided.

Fixes: 0d01da6afc54 ("bpf: implement getsockopt and setsockopt hooks")
Reported-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/ZD7Js4fj5YyI2oLd@google.com/T/#mb68daf700f87a9244a15d01d00c3f0e5b08f49f7
Link: https://lore.kernel.org/bpf/20230418225343.553806-2-sdf@google.com
---
 kernel/bpf/cgroup.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 53edb8ad2471..a06e118a9be5 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -1921,14 +1921,17 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
 	if (ret < 0)
 		goto out;
 
-	if (ctx.optlen > max_optlen || ctx.optlen < 0) {
+	if (optval && (ctx.optlen > max_optlen || ctx.optlen < 0)) {
 		ret = -EFAULT;
 		goto out;
 	}
 
 	if (ctx.optlen != 0) {
-		if (copy_to_user(optval, ctx.optval, ctx.optlen) ||
-		    put_user(ctx.optlen, optlen)) {
+		if (optval && copy_to_user(optval, ctx.optval, ctx.optlen)) {
+			ret = -EFAULT;
+			goto out;
+		}
+		if (put_user(ctx.optlen, optlen)) {
 			ret = -EFAULT;
 			goto out;
 		}
-- 
cgit v1.2.3-70-g09d2


From 833d67ecdc5f35f1ebf59d0fccc1ce771434be9c Mon Sep 17 00:00:00 2001
From: Stanislav Fomichev <sdf@google.com>
Date: Tue, 18 Apr 2023 15:53:39 -0700
Subject: selftests/bpf: Verify optval=NULL case

Make sure we get optlen exported instead of getting EFAULT.

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230418225343.553806-3-sdf@google.com
---
 .../testing/selftests/bpf/prog_tests/sockopt_sk.c  | 28 ++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/sockopt_sk.c     | 12 ++++++++++
 2 files changed, 40 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c b/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c
index 60d952719d27..4512dd808c33 100644
--- a/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c
+++ b/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c
@@ -3,6 +3,7 @@
 #include "cgroup_helpers.h"
 
 #include <linux/tcp.h>
+#include <linux/netlink.h>
 #include "sockopt_sk.skel.h"
 
 #ifndef SOL_TCP
@@ -183,6 +184,33 @@ static int getsetsockopt(void)
 		goto err;
 	}
 
+	/* optval=NULL case is handled correctly */
+
+	close(fd);
+	fd = socket(AF_NETLINK, SOCK_RAW, 0);
+	if (fd < 0) {
+		log_err("Failed to create AF_NETLINK socket");
+		return -1;
+	}
+
+	buf.u32 = 1;
+	optlen = sizeof(__u32);
+	err = setsockopt(fd, SOL_NETLINK, NETLINK_ADD_MEMBERSHIP, &buf, optlen);
+	if (err) {
+		log_err("Unexpected getsockopt(NETLINK_ADD_MEMBERSHIP) err=%d errno=%d",
+			err, errno);
+		goto err;
+	}
+
+	optlen = 0;
+	err = getsockopt(fd, SOL_NETLINK, NETLINK_LIST_MEMBERSHIPS, NULL, &optlen);
+	if (err) {
+		log_err("Unexpected getsockopt(NETLINK_LIST_MEMBERSHIPS) err=%d errno=%d",
+			err, errno);
+		goto err;
+	}
+	ASSERT_EQ(optlen, 4, "Unexpected NETLINK_LIST_MEMBERSHIPS value");
+
 	free(big_buf);
 	close(fd);
 	return 0;
diff --git a/tools/testing/selftests/bpf/progs/sockopt_sk.c b/tools/testing/selftests/bpf/progs/sockopt_sk.c
index c8d810010a94..fe1df4cd206e 100644
--- a/tools/testing/selftests/bpf/progs/sockopt_sk.c
+++ b/tools/testing/selftests/bpf/progs/sockopt_sk.c
@@ -32,6 +32,12 @@ int _getsockopt(struct bpf_sockopt *ctx)
 	__u8 *optval_end = ctx->optval_end;
 	__u8 *optval = ctx->optval;
 	struct sockopt_sk *storage;
+	struct bpf_sock *sk;
+
+	/* Bypass AF_NETLINK. */
+	sk = ctx->sk;
+	if (sk && sk->family == AF_NETLINK)
+		return 1;
 
 	/* Make sure bpf_get_netns_cookie is callable.
 	 */
@@ -131,6 +137,12 @@ int _setsockopt(struct bpf_sockopt *ctx)
 	__u8 *optval_end = ctx->optval_end;
 	__u8 *optval = ctx->optval;
 	struct sockopt_sk *storage;
+	struct bpf_sock *sk;
+
+	/* Bypass AF_NETLINK. */
+	sk = ctx->sk;
+	if (sk && sk->family == AF_NETLINK)
+		return 1;
 
 	/* Make sure bpf_get_netns_cookie is callable.
 	 */
-- 
cgit v1.2.3-70-g09d2


From 0232b788978652571c4f4e57dc26ff8e4837926a Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 19 Apr 2023 17:28:21 -0700
Subject: bpftool: Register struct_ops with a link.

You can include an optional path after specifying the object name for the
'struct_ops register' subcommand.

Since the commit 226bc6ae6405 ("Merge branch 'Transit between BPF TCP
congestion controls.'") has been accepted, it is now possible to create a
link for a struct_ops. This can be done by defining a struct_ops in
SEC(".struct_ops.link") to make libbpf returns a real link. If we don't pin
the links before leaving bpftool, they will disappear. To instruct bpftool
to pin the links in a directory with the names of the maps, we need to
provide the path of that directory.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230420002822.345222-1-kuifeng@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/common.c     | 14 +++++++++
 tools/bpf/bpftool/main.h       |  3 ++
 tools/bpf/bpftool/prog.c       | 13 --------
 tools/bpf/bpftool/struct_ops.c | 70 ++++++++++++++++++++++++++++++++++--------
 4 files changed, 75 insertions(+), 25 deletions(-)

diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c
index 5a73ccf14332..1360c82ae732 100644
--- a/tools/bpf/bpftool/common.c
+++ b/tools/bpf/bpftool/common.c
@@ -1091,3 +1091,17 @@ const char *bpf_attach_type_input_str(enum bpf_attach_type t)
 	default:	return libbpf_bpf_attach_type_str(t);
 	}
 }
+
+int pathname_concat(char *buf, int buf_sz, const char *path,
+		    const char *name)
+{
+	int len;
+
+	len = snprintf(buf, buf_sz, "%s/%s", path, name);
+	if (len < 0)
+		return -EINVAL;
+	if (len >= buf_sz)
+		return -ENAMETOOLONG;
+
+	return 0;
+}
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 00d11ca6d3f2..9a95788e654d 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -264,4 +264,7 @@ static inline bool hashmap__empty(struct hashmap *map)
 	return map ? hashmap__size(map) == 0 : true;
 }
 
+int pathname_concat(char *buf, int buf_sz, const char *path,
+		    const char *name);
+
 #endif
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index e5b613a7974c..91b6075b2db3 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -1476,19 +1476,6 @@ auto_attach_program(struct bpf_program *prog, const char *path)
 	return err;
 }
 
-static int pathname_concat(char *buf, size_t buf_sz, const char *path, const char *name)
-{
-	int len;
-
-	len = snprintf(buf, buf_sz, "%s/%s", path, name);
-	if (len < 0)
-		return -EINVAL;
-	if ((size_t)len >= buf_sz)
-		return -ENAMETOOLONG;
-
-	return 0;
-}
-
 static int
 auto_attach_programs(struct bpf_object *obj, const char *path)
 {
diff --git a/tools/bpf/bpftool/struct_ops.c b/tools/bpf/bpftool/struct_ops.c
index b389f4830e11..57c3da70aa31 100644
--- a/tools/bpf/bpftool/struct_ops.c
+++ b/tools/bpf/bpftool/struct_ops.c
@@ -475,21 +475,44 @@ static int do_unregister(int argc, char **argv)
 	return cmd_retval(&res, true);
 }
 
+static int pin_link(struct bpf_link *link, const char *pindir,
+		    const char *name)
+{
+	char pinfile[PATH_MAX];
+	int err;
+
+	err = pathname_concat(pinfile, sizeof(pinfile), pindir, name);
+	if (err)
+		return -1;
+
+	return bpf_link__pin(link, pinfile);
+}
+
 static int do_register(int argc, char **argv)
 {
 	LIBBPF_OPTS(bpf_object_open_opts, open_opts);
+	__u32 link_info_len = sizeof(struct bpf_link_info);
+	struct bpf_link_info link_info = {};
 	struct bpf_map_info info = {};
 	__u32 info_len = sizeof(info);
 	int nr_errs = 0, nr_maps = 0;
+	const char *linkdir = NULL;
 	struct bpf_object *obj;
 	struct bpf_link *link;
 	struct bpf_map *map;
 	const char *file;
 
-	if (argc != 1)
+	if (argc != 1 && argc != 2)
 		usage();
 
 	file = GET_ARG();
+	if (argc == 1)
+		linkdir = GET_ARG();
+
+	if (linkdir && mount_bpffs_for_pin(linkdir)) {
+		p_err("can't mount bpffs for pinning");
+		return -1;
+	}
 
 	if (verifier_logs)
 		/* log_level1 + log_level2 + stats, but not stable UAPI */
@@ -519,21 +542,44 @@ static int do_register(int argc, char **argv)
 		}
 		nr_maps++;
 
-		bpf_link__disconnect(link);
-		bpf_link__destroy(link);
-
-		if (!bpf_map_get_info_by_fd(bpf_map__fd(map), &info,
-					    &info_len))
-			p_info("Registered %s %s id %u",
-			       get_kern_struct_ops_name(&info),
-			       bpf_map__name(map),
-			       info.id);
-		else
+		if (bpf_map_get_info_by_fd(bpf_map__fd(map), &info,
+					   &info_len)) {
 			/* Not p_err.  The struct_ops was attached
 			 * successfully.
 			 */
 			p_info("Registered %s but can't find id: %s",
 			       bpf_map__name(map), strerror(errno));
+			goto clean_link;
+		}
+		if (!(bpf_map__map_flags(map) & BPF_F_LINK)) {
+			p_info("Registered %s %s id %u",
+			       get_kern_struct_ops_name(&info),
+			       info.name,
+			       info.id);
+			goto clean_link;
+		}
+		if (bpf_link_get_info_by_fd(bpf_link__fd(link),
+					    &link_info,
+					    &link_info_len)) {
+			p_err("Registered %s but can't find link id: %s",
+			      bpf_map__name(map), strerror(errno));
+			nr_errs++;
+			goto clean_link;
+		}
+		if (linkdir && pin_link(link, linkdir, info.name)) {
+			p_err("can't pin link %u for %s: %s",
+			      link_info.id, info.name,
+			      strerror(errno));
+			nr_errs++;
+			goto clean_link;
+		}
+		p_info("Registered %s %s map id %u link id %u",
+		       get_kern_struct_ops_name(&info),
+		       info.name, info.id, link_info.id);
+
+clean_link:
+		bpf_link__disconnect(link);
+		bpf_link__destroy(link);
 	}
 
 	bpf_object__close(obj);
@@ -562,7 +608,7 @@ static int do_help(int argc, char **argv)
 	fprintf(stderr,
 		"Usage: %1$s %2$s { show | list } [STRUCT_OPS_MAP]\n"
 		"       %1$s %2$s dump [STRUCT_OPS_MAP]\n"
-		"       %1$s %2$s register OBJ\n"
+		"       %1$s %2$s register OBJ [LINK_DIR]\n"
 		"       %1$s %2$s unregister STRUCT_OPS_MAP\n"
 		"       %1$s %2$s help\n"
 		"\n"
-- 
cgit v1.2.3-70-g09d2


From 45cea721ea36d83969473d2abd29bcc2321cacdd Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 19 Apr 2023 17:28:22 -0700
Subject: bpftool: Update doc to explain struct_ops register subcommand.

The "struct_ops register" subcommand now allows for an optional *LINK_DIR*
to be included. This specifies the directory path where bpftool will pin
struct_ops links with the same name as their corresponding map names.

Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230420002822.345222-2-kuifeng@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst b/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst
index ee53a122c0c7..8022b5321dbe 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst
@@ -26,7 +26,7 @@ STRUCT_OPS COMMANDS
 
 |	**bpftool** **struct_ops { show | list }** [*STRUCT_OPS_MAP*]
 |	**bpftool** **struct_ops dump** [*STRUCT_OPS_MAP*]
-|	**bpftool** **struct_ops register** *OBJ*
+|	**bpftool** **struct_ops register** *OBJ* [*LINK_DIR*]
 |	**bpftool** **struct_ops unregister** *STRUCT_OPS_MAP*
 |	**bpftool** **struct_ops help**
 |
@@ -51,10 +51,14 @@ DESCRIPTION
 		  for the given struct_ops.  Otherwise, it dumps all struct_ops
 		  currently existing in the system.
 
-	**bpftool struct_ops register** *OBJ*
+	**bpftool struct_ops register** *OBJ* [*LINK_DIR*]
 		  Register bpf struct_ops from *OBJ*.  All struct_ops under
-		  the ELF section ".struct_ops" will be registered to
-		  its kernel subsystem.
+		  the ELF section ".struct_ops" and ".struct_ops.link" will
+		  be registered to its kernel subsystem.  For each
+		  struct_ops in the ".struct_ops.link" section, a link
+		  will be created.  You can give *LINK_DIR* to provide a
+		  directory path where these links will be pinned with the
+		  same name as their corresponding map name.
 
 	**bpftool struct_ops unregister**  *STRUCT_OPS_MAP*
 		  Unregister the *STRUCT_OPS_MAP* from the kernel subsystem.
-- 
cgit v1.2.3-70-g09d2


From 84601d6ee68ae820dec97450934797046d62db4b Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:02:54 +0200
Subject: bpf: add bpf_link support for BPF_NETFILTER programs

Add bpf_link support skeleton.  To keep this reviewable, no bpf program
can be invoked yet, if a program is attached only a c-stub is called and
not the actual bpf program.

Defaults to 'y' if both netfilter and bpf syscall are enabled in kconfig.

Uapi example usage:
	union bpf_attr attr = { };

	attr.link_create.prog_fd = progfd;
	attr.link_create.attach_type = 0; /* unused */
	attr.link_create.netfilter.pf = PF_INET;
	attr.link_create.netfilter.hooknum = NF_INET_LOCAL_IN;
	attr.link_create.netfilter.priority = -128;

	err = bpf(BPF_LINK_CREATE, &attr, sizeof(attr));

... this would attach progfd to ipv4:input hook.

Such hook gets removed automatically if the calling program exits.

BPF_NETFILTER program invocation is added in followup change.

NF_HOOK_OP_BPF enum will eventually be read from nfnetlink_hook, it
allows to tell userspace which program is attached at the given hook
when user runs 'nft hook list' command rather than just the priority
and not-very-helpful 'this hook runs a bpf prog but I can't tell which
one'.

Will also be used to disallow registration of two bpf programs with
same priority in a followup patch.

v4: arm32 cmpxchg only supports 32bit operand
    s/prio/priority/
v3: restrict prog attachment to ip/ip6 for now, lets lift restrictions if
    more use cases pop up (arptables, ebtables, netdev ingress/egress etc).

Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-2-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/netfilter.h           |   1 +
 include/net/netfilter/nf_bpf_link.h |  10 +++
 include/uapi/linux/bpf.h            |  14 ++++
 kernel/bpf/syscall.c                |   6 ++
 net/netfilter/Kconfig               |   3 +
 net/netfilter/Makefile              |   1 +
 net/netfilter/nf_bpf_link.c         | 159 ++++++++++++++++++++++++++++++++++++
 7 files changed, 194 insertions(+)
 create mode 100644 include/net/netfilter/nf_bpf_link.h
 create mode 100644 net/netfilter/nf_bpf_link.c

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index c8e03bcaecaa..0762444e3767 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -80,6 +80,7 @@ typedef unsigned int nf_hookfn(void *priv,
 enum nf_hook_ops_type {
 	NF_HOOK_OP_UNDEFINED,
 	NF_HOOK_OP_NF_TABLES,
+	NF_HOOK_OP_BPF,
 };
 
 struct nf_hook_ops {
diff --git a/include/net/netfilter/nf_bpf_link.h b/include/net/netfilter/nf_bpf_link.h
new file mode 100644
index 000000000000..eeaeaf3d15de
--- /dev/null
+++ b/include/net/netfilter/nf_bpf_link.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#if IS_ENABLED(CONFIG_NETFILTER_BPF_LINK)
+int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog);
+#else
+static inline int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
+{
+	return -EOPNOTSUPP;
+}
+#endif
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 4b20a7269bee..1bb11a6ee667 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -986,6 +986,7 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_LSM,
 	BPF_PROG_TYPE_SK_LOOKUP,
 	BPF_PROG_TYPE_SYSCALL, /* a program that can execute syscalls */
+	BPF_PROG_TYPE_NETFILTER,
 };
 
 enum bpf_attach_type {
@@ -1050,6 +1051,7 @@ enum bpf_link_type {
 	BPF_LINK_TYPE_PERF_EVENT = 7,
 	BPF_LINK_TYPE_KPROBE_MULTI = 8,
 	BPF_LINK_TYPE_STRUCT_OPS = 9,
+	BPF_LINK_TYPE_NETFILTER = 10,
 
 	MAX_BPF_LINK_TYPE,
 };
@@ -1560,6 +1562,12 @@ union bpf_attr {
 				 */
 				__u64		cookie;
 			} tracing;
+			struct {
+				__u32		pf;
+				__u32		hooknum;
+				__s32		priority;
+				__u32		flags;
+			} netfilter;
 		};
 	} link_create;
 
@@ -6410,6 +6418,12 @@ struct bpf_link_info {
 		struct {
 			__u32 map_id;
 		} struct_ops;
+		struct {
+			__u32 pf;
+			__u32 hooknum;
+			__s32 priority;
+			__u32 flags;
+		} netfilter;
 	};
 } __attribute__((aligned(8)));
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index bcf1a1920ddd..14f39c1e573e 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -35,6 +35,7 @@
 #include <linux/rcupdate_trace.h>
 #include <linux/memcontrol.h>
 #include <linux/trace_events.h>
+#include <net/netfilter/nf_bpf_link.h>
 
 #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \
 			  (map)->map_type == BPF_MAP_TYPE_CGROUP_ARRAY || \
@@ -2462,6 +2463,7 @@ static bool is_net_admin_prog_type(enum bpf_prog_type prog_type)
 	case BPF_PROG_TYPE_CGROUP_SYSCTL:
 	case BPF_PROG_TYPE_SOCK_OPS:
 	case BPF_PROG_TYPE_EXT: /* extends any prog */
+	case BPF_PROG_TYPE_NETFILTER:
 		return true;
 	case BPF_PROG_TYPE_CGROUP_SKB:
 		/* always unpriv */
@@ -4588,6 +4590,7 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr)
 
 	switch (prog->type) {
 	case BPF_PROG_TYPE_EXT:
+	case BPF_PROG_TYPE_NETFILTER:
 		break;
 	case BPF_PROG_TYPE_PERF_EVENT:
 	case BPF_PROG_TYPE_TRACEPOINT:
@@ -4654,6 +4657,9 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr)
 	case BPF_PROG_TYPE_XDP:
 		ret = bpf_xdp_link_attach(attr, prog);
 		break;
+	case BPF_PROG_TYPE_NETFILTER:
+		ret = bpf_nf_link_attach(attr, prog);
+		break;
 #endif
 	case BPF_PROG_TYPE_PERF_EVENT:
 	case BPF_PROG_TYPE_TRACEPOINT:
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index d0bf630482c1..441d1f134110 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -30,6 +30,9 @@ config NETFILTER_FAMILY_BRIDGE
 config NETFILTER_FAMILY_ARP
 	bool
 
+config NETFILTER_BPF_LINK
+	def_bool BPF_SYSCALL
+
 config NETFILTER_NETLINK_HOOK
 	tristate "Netfilter base hook dump support"
 	depends on NETFILTER_ADVANCED
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 5ffef1cd6143..d4958e7e7631 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -22,6 +22,7 @@ nf_conntrack-$(CONFIG_DEBUG_INFO_BTF) += nf_conntrack_bpf.o
 endif
 
 obj-$(CONFIG_NETFILTER) = netfilter.o
+obj-$(CONFIG_NETFILTER_BPF_LINK) += nf_bpf_link.o
 
 obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o
 obj-$(CONFIG_NETFILTER_NETLINK_ACCT) += nfnetlink_acct.o
diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c
new file mode 100644
index 000000000000..efa4f3390742
--- /dev/null
+++ b/net/netfilter/nf_bpf_link.c
@@ -0,0 +1,159 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bpf.h>
+#include <linux/netfilter.h>
+
+#include <net/netfilter/nf_bpf_link.h>
+#include <uapi/linux/netfilter_ipv4.h>
+
+static unsigned int nf_hook_run_bpf(void *bpf_prog, struct sk_buff *skb,
+				    const struct nf_hook_state *s)
+{
+	return NF_ACCEPT;
+}
+
+struct bpf_nf_link {
+	struct bpf_link link;
+	struct nf_hook_ops hook_ops;
+	struct net *net;
+	u32 dead;
+};
+
+static void bpf_nf_link_release(struct bpf_link *link)
+{
+	struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link);
+
+	if (nf_link->dead)
+		return;
+
+	/* prevent hook-not-found warning splat from netfilter core when
+	 * .detach was already called
+	 */
+	if (!cmpxchg(&nf_link->dead, 0, 1))
+		nf_unregister_net_hook(nf_link->net, &nf_link->hook_ops);
+}
+
+static void bpf_nf_link_dealloc(struct bpf_link *link)
+{
+	struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link);
+
+	kfree(nf_link);
+}
+
+static int bpf_nf_link_detach(struct bpf_link *link)
+{
+	bpf_nf_link_release(link);
+	return 0;
+}
+
+static void bpf_nf_link_show_info(const struct bpf_link *link,
+				  struct seq_file *seq)
+{
+	struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link);
+
+	seq_printf(seq, "pf:\t%u\thooknum:\t%u\tprio:\t%d\n",
+		   nf_link->hook_ops.pf, nf_link->hook_ops.hooknum,
+		   nf_link->hook_ops.priority);
+}
+
+static int bpf_nf_link_fill_link_info(const struct bpf_link *link,
+				      struct bpf_link_info *info)
+{
+	struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link);
+
+	info->netfilter.pf = nf_link->hook_ops.pf;
+	info->netfilter.hooknum = nf_link->hook_ops.hooknum;
+	info->netfilter.priority = nf_link->hook_ops.priority;
+	info->netfilter.flags = 0;
+
+	return 0;
+}
+
+static int bpf_nf_link_update(struct bpf_link *link, struct bpf_prog *new_prog,
+			      struct bpf_prog *old_prog)
+{
+	return -EOPNOTSUPP;
+}
+
+static const struct bpf_link_ops bpf_nf_link_lops = {
+	.release = bpf_nf_link_release,
+	.dealloc = bpf_nf_link_dealloc,
+	.detach = bpf_nf_link_detach,
+	.show_fdinfo = bpf_nf_link_show_info,
+	.fill_link_info = bpf_nf_link_fill_link_info,
+	.update_prog = bpf_nf_link_update,
+};
+
+static int bpf_nf_check_pf_and_hooks(const union bpf_attr *attr)
+{
+	switch (attr->link_create.netfilter.pf) {
+	case NFPROTO_IPV4:
+	case NFPROTO_IPV6:
+		if (attr->link_create.netfilter.hooknum >= NF_INET_NUMHOOKS)
+			return -EPROTO;
+		break;
+	default:
+		return -EAFNOSUPPORT;
+	}
+
+	if (attr->link_create.netfilter.flags)
+		return -EOPNOTSUPP;
+
+	/* make sure conntrack confirm is always last.
+	 *
+	 * In the future, if userspace can e.g. request defrag, then
+	 * "defrag_requested && prio before NF_IP_PRI_CONNTRACK_DEFRAG"
+	 * should fail.
+	 */
+	switch (attr->link_create.netfilter.priority) {
+	case NF_IP_PRI_FIRST: return -ERANGE; /* sabotage_in and other warts */
+	case NF_IP_PRI_LAST: return -ERANGE; /* e.g. conntrack confirm */
+	}
+
+	return 0;
+}
+
+int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
+{
+	struct net *net = current->nsproxy->net_ns;
+	struct bpf_link_primer link_primer;
+	struct bpf_nf_link *link;
+	int err;
+
+	if (attr->link_create.flags)
+		return -EINVAL;
+
+	err = bpf_nf_check_pf_and_hooks(attr);
+	if (err)
+		return err;
+
+	link = kzalloc(sizeof(*link), GFP_USER);
+	if (!link)
+		return -ENOMEM;
+
+	bpf_link_init(&link->link, BPF_LINK_TYPE_NETFILTER, &bpf_nf_link_lops, prog);
+
+	link->hook_ops.hook = nf_hook_run_bpf;
+	link->hook_ops.hook_ops_type = NF_HOOK_OP_BPF;
+	link->hook_ops.priv = prog;
+
+	link->hook_ops.pf = attr->link_create.netfilter.pf;
+	link->hook_ops.priority = attr->link_create.netfilter.priority;
+	link->hook_ops.hooknum = attr->link_create.netfilter.hooknum;
+
+	link->net = net;
+	link->dead = false;
+
+	err = bpf_link_prime(&link->link, &link_primer);
+	if (err) {
+		kfree(link);
+		return err;
+	}
+
+	err = nf_register_net_hook(net, &link->hook_ops);
+	if (err) {
+		bpf_link_cleanup(&link_primer);
+		return err;
+	}
+
+	return bpf_link_settle(&link_primer);
+}
-- 
cgit v1.2.3-70-g09d2


From fd9c663b9ad67dedfc9a3fd3429ddd3e83782b4d Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:02:55 +0200
Subject: bpf: minimal support for programs hooked into netfilter framework

This adds minimal support for BPF_PROG_TYPE_NETFILTER bpf programs
that will be invoked via the NF_HOOK() points in the ip stack.

Invocation incurs an indirect call.  This is not a necessity: Its
possible to add 'DEFINE_BPF_DISPATCHER(nf_progs)' and handle the
program invocation with the same method already done for xdp progs.

This isn't done here to keep the size of this chunk down.

Verifier restricts verdicts to either DROP or ACCEPT.

Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-3-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_types.h           |  4 +++
 include/net/netfilter/nf_bpf_link.h |  5 +++
 kernel/bpf/btf.c                    |  6 ++++
 kernel/bpf/verifier.c               |  3 ++
 net/core/filter.c                   |  1 +
 net/netfilter/nf_bpf_link.c         | 70 ++++++++++++++++++++++++++++++++++++-
 6 files changed, 88 insertions(+), 1 deletion(-)

diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h
index d4ee3ccd3753..39a999abb0ce 100644
--- a/include/linux/bpf_types.h
+++ b/include/linux/bpf_types.h
@@ -79,6 +79,10 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_LSM, lsm,
 #endif
 BPF_PROG_TYPE(BPF_PROG_TYPE_SYSCALL, bpf_syscall,
 	      void *, void *)
+#ifdef CONFIG_NETFILTER
+BPF_PROG_TYPE(BPF_PROG_TYPE_NETFILTER, netfilter,
+	      struct bpf_nf_ctx, struct bpf_nf_ctx)
+#endif
 
 BPF_MAP_TYPE(BPF_MAP_TYPE_ARRAY, array_map_ops)
 BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_ARRAY, percpu_array_map_ops)
diff --git a/include/net/netfilter/nf_bpf_link.h b/include/net/netfilter/nf_bpf_link.h
index eeaeaf3d15de..6c984b0ea838 100644
--- a/include/net/netfilter/nf_bpf_link.h
+++ b/include/net/netfilter/nf_bpf_link.h
@@ -1,5 +1,10 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 
+struct bpf_nf_ctx {
+	const struct nf_hook_state *state;
+	struct sk_buff *skb;
+};
+
 #if IS_ENABLED(CONFIG_NETFILTER_BPF_LINK)
 int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog);
 #else
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 7db4ec125fbd..6b682b8e4b50 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -25,6 +25,9 @@
 #include <linux/bsearch.h>
 #include <linux/kobject.h>
 #include <linux/sysfs.h>
+
+#include <net/netfilter/nf_bpf_link.h>
+
 #include <net/sock.h>
 #include "../tools/lib/bpf/relo_core.h"
 
@@ -212,6 +215,7 @@ enum btf_kfunc_hook {
 	BTF_KFUNC_HOOK_SK_SKB,
 	BTF_KFUNC_HOOK_SOCKET_FILTER,
 	BTF_KFUNC_HOOK_LWT,
+	BTF_KFUNC_HOOK_NETFILTER,
 	BTF_KFUNC_HOOK_MAX,
 };
 
@@ -7802,6 +7806,8 @@ static int bpf_prog_type_to_kfunc_hook(enum bpf_prog_type prog_type)
 	case BPF_PROG_TYPE_LWT_XMIT:
 	case BPF_PROG_TYPE_LWT_SEG6LOCAL:
 		return BTF_KFUNC_HOOK_LWT;
+	case BPF_PROG_TYPE_NETFILTER:
+		return BTF_KFUNC_HOOK_NETFILTER;
 	default:
 		return BTF_KFUNC_HOOK_MAX;
 	}
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 1e05355facdc..fc7281d39e46 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -13816,6 +13816,9 @@ static int check_return_code(struct bpf_verifier_env *env)
 		}
 		break;
 
+	case BPF_PROG_TYPE_NETFILTER:
+		range = tnum_range(NF_DROP, NF_ACCEPT);
+		break;
 	case BPF_PROG_TYPE_EXT:
 		/* freplace program can return anything as its return value
 		 * depends on the to-be-replaced kernel func or bpf program.
diff --git a/net/core/filter.c b/net/core/filter.c
index 44fb997434ad..d9ce04ca22ce 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -11717,6 +11717,7 @@ static int __init bpf_kfunc_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_IN, &bpf_kfunc_set_skb);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_XMIT, &bpf_kfunc_set_skb);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_SEG6LOCAL, &bpf_kfunc_set_skb);
+	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_NETFILTER, &bpf_kfunc_set_skb);
 	return ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_XDP, &bpf_kfunc_set_xdp);
 }
 late_initcall(bpf_kfunc_init);
diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c
index efa4f3390742..49cfc5215386 100644
--- a/net/netfilter/nf_bpf_link.c
+++ b/net/netfilter/nf_bpf_link.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/bpf.h>
+#include <linux/filter.h>
 #include <linux/netfilter.h>
 
 #include <net/netfilter/nf_bpf_link.h>
@@ -8,7 +9,13 @@
 static unsigned int nf_hook_run_bpf(void *bpf_prog, struct sk_buff *skb,
 				    const struct nf_hook_state *s)
 {
-	return NF_ACCEPT;
+	const struct bpf_prog *prog = bpf_prog;
+	struct bpf_nf_ctx ctx = {
+		.state = s,
+		.skb = skb,
+	};
+
+	return bpf_prog_run(prog, &ctx);
 }
 
 struct bpf_nf_link {
@@ -157,3 +164,64 @@ int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
 
 	return bpf_link_settle(&link_primer);
 }
+
+const struct bpf_prog_ops netfilter_prog_ops = {
+};
+
+static bool nf_ptr_to_btf_id(struct bpf_insn_access_aux *info, const char *name)
+{
+	struct btf *btf;
+	s32 type_id;
+
+	btf = bpf_get_btf_vmlinux();
+	if (IS_ERR_OR_NULL(btf))
+		return false;
+
+	type_id = btf_find_by_name_kind(btf, name, BTF_KIND_STRUCT);
+	if (WARN_ON_ONCE(type_id < 0))
+		return false;
+
+	info->btf = btf;
+	info->btf_id = type_id;
+	info->reg_type = PTR_TO_BTF_ID | PTR_TRUSTED;
+	return true;
+}
+
+static bool nf_is_valid_access(int off, int size, enum bpf_access_type type,
+			       const struct bpf_prog *prog,
+			       struct bpf_insn_access_aux *info)
+{
+	if (off < 0 || off >= sizeof(struct bpf_nf_ctx))
+		return false;
+
+	if (type == BPF_WRITE)
+		return false;
+
+	switch (off) {
+	case bpf_ctx_range(struct bpf_nf_ctx, skb):
+		if (size != sizeof_field(struct bpf_nf_ctx, skb))
+			return false;
+
+		return nf_ptr_to_btf_id(info, "sk_buff");
+	case bpf_ctx_range(struct bpf_nf_ctx, state):
+		if (size != sizeof_field(struct bpf_nf_ctx, state))
+			return false;
+
+		return nf_ptr_to_btf_id(info, "nf_hook_state");
+	default:
+		return false;
+	}
+
+	return false;
+}
+
+static const struct bpf_func_proto *
+bpf_nf_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
+{
+	return bpf_base_func_proto(func_id);
+}
+
+const struct bpf_verifier_ops netfilter_verifier_ops = {
+	.is_valid_access	= nf_is_valid_access,
+	.get_func_proto		= bpf_nf_func_proto,
+};
-- 
cgit v1.2.3-70-g09d2


From 506a74db7e019a277e987fa65654bdd953859d5b Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:02:56 +0200
Subject: netfilter: nfnetlink hook: dump bpf prog id

This allows userspace ("nft list hooks") to show which bpf program
is attached to which hook.

Without this, user only knows bpf prog is attached at prio
x, y, z at INPUT and FORWARD, but can't tell which program is where.

v4: kdoc fixups (Simon Horman)

Link: https://lore.kernel.org/bpf/ZEELzpNCnYJuZyod@corigine.com/
Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-4-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/uapi/linux/netfilter/nfnetlink_hook.h | 24 +++++++-
 net/netfilter/nfnetlink_hook.c                | 81 ++++++++++++++++++++++-----
 2 files changed, 89 insertions(+), 16 deletions(-)

diff --git a/include/uapi/linux/netfilter/nfnetlink_hook.h b/include/uapi/linux/netfilter/nfnetlink_hook.h
index bbcd285b22e1..84a561a74b98 100644
--- a/include/uapi/linux/netfilter/nfnetlink_hook.h
+++ b/include/uapi/linux/netfilter/nfnetlink_hook.h
@@ -32,8 +32,12 @@ enum nfnl_hook_attributes {
 /**
  * enum nfnl_hook_chain_info_attributes - chain description
  *
- * NFNLA_HOOK_INFO_DESC: nft chain and table name (enum nft_table_attributes) (NLA_NESTED)
- * NFNLA_HOOK_INFO_TYPE: chain type (enum nfnl_hook_chaintype) (NLA_U32)
+ * @NFNLA_HOOK_INFO_DESC: nft chain and table name (NLA_NESTED)
+ * @NFNLA_HOOK_INFO_TYPE: chain type (enum nfnl_hook_chaintype) (NLA_U32)
+ *
+ * NFNLA_HOOK_INFO_DESC depends on NFNLA_HOOK_INFO_TYPE value:
+ *   NFNL_HOOK_TYPE_NFTABLES: enum nft_table_attributes
+ *   NFNL_HOOK_TYPE_BPF: enum nfnl_hook_bpf_attributes
  */
 enum nfnl_hook_chain_info_attributes {
 	NFNLA_HOOK_INFO_UNSPEC,
@@ -55,10 +59,24 @@ enum nfnl_hook_chain_desc_attributes {
 /**
  * enum nfnl_hook_chaintype - chain type
  *
- * @NFNL_HOOK_TYPE_NFTABLES nf_tables base chain
+ * @NFNL_HOOK_TYPE_NFTABLES: nf_tables base chain
+ * @NFNL_HOOK_TYPE_BPF: bpf program
  */
 enum nfnl_hook_chaintype {
 	NFNL_HOOK_TYPE_NFTABLES = 0x1,
+	NFNL_HOOK_TYPE_BPF,
+};
+
+/**
+ * enum nfnl_hook_bpf_attributes - bpf prog description
+ *
+ * @NFNLA_HOOK_BPF_ID: bpf program id (NLA_U32)
+ */
+enum nfnl_hook_bpf_attributes {
+	NFNLA_HOOK_BPF_UNSPEC,
+	NFNLA_HOOK_BPF_ID,
+	__NFNLA_HOOK_BPF_MAX,
 };
+#define NFNLA_HOOK_BPF_MAX (__NFNLA_HOOK_BPF_MAX - 1)
 
 #endif /* _NFNL_HOOK_H */
diff --git a/net/netfilter/nfnetlink_hook.c b/net/netfilter/nfnetlink_hook.c
index 8120aadf6a0f..ade8ee1988b1 100644
--- a/net/netfilter/nfnetlink_hook.c
+++ b/net/netfilter/nfnetlink_hook.c
@@ -5,6 +5,7 @@
  * Author: Florian Westphal <fw@strlen.de>
  */
 
+#include <linux/bpf.h>
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 #include <linux/kernel.h>
@@ -57,35 +58,76 @@ struct nfnl_dump_hook_data {
 	u8 hook;
 };
 
+static struct nlattr *nfnl_start_info_type(struct sk_buff *nlskb, enum nfnl_hook_chaintype t)
+{
+	struct nlattr *nest = nla_nest_start(nlskb, NFNLA_HOOK_CHAIN_INFO);
+	int ret;
+
+	if (!nest)
+		return NULL;
+
+	ret = nla_put_be32(nlskb, NFNLA_HOOK_INFO_TYPE, htonl(t));
+	if (ret == 0)
+		return nest;
+
+	nla_nest_cancel(nlskb, nest);
+	return NULL;
+}
+
+static int nfnl_hook_put_bpf_prog_info(struct sk_buff *nlskb,
+				       const struct nfnl_dump_hook_data *ctx,
+				       unsigned int seq,
+				       const struct bpf_prog *prog)
+{
+	struct nlattr *nest, *nest2;
+	int ret;
+
+	if (!IS_ENABLED(CONFIG_NETFILTER_BPF_LINK))
+		return 0;
+
+	if (WARN_ON_ONCE(!prog))
+		return 0;
+
+	nest = nfnl_start_info_type(nlskb, NFNL_HOOK_TYPE_BPF);
+	if (!nest)
+		return -EMSGSIZE;
+
+	nest2 = nla_nest_start(nlskb, NFNLA_HOOK_INFO_DESC);
+	if (!nest2)
+		goto cancel_nest;
+
+	ret = nla_put_be32(nlskb, NFNLA_HOOK_BPF_ID, htonl(prog->aux->id));
+	if (ret)
+		goto cancel_nest;
+
+	nla_nest_end(nlskb, nest2);
+	nla_nest_end(nlskb, nest);
+	return 0;
+
+cancel_nest:
+	nla_nest_cancel(nlskb, nest);
+	return -EMSGSIZE;
+}
+
 static int nfnl_hook_put_nft_chain_info(struct sk_buff *nlskb,
 					const struct nfnl_dump_hook_data *ctx,
 					unsigned int seq,
-					const struct nf_hook_ops *ops)
+					struct nft_chain *chain)
 {
 	struct net *net = sock_net(nlskb->sk);
 	struct nlattr *nest, *nest2;
-	struct nft_chain *chain;
 	int ret = 0;
 
-	if (ops->hook_ops_type != NF_HOOK_OP_NF_TABLES)
-		return 0;
-
-	chain = ops->priv;
 	if (WARN_ON_ONCE(!chain))
 		return 0;
 
 	if (!nft_is_active(net, chain))
 		return 0;
 
-	nest = nla_nest_start(nlskb, NFNLA_HOOK_CHAIN_INFO);
+	nest = nfnl_start_info_type(nlskb, NFNL_HOOK_TYPE_NFTABLES);
 	if (!nest)
 		return -EMSGSIZE;
 
-	ret = nla_put_be32(nlskb, NFNLA_HOOK_INFO_TYPE,
-			   htonl(NFNL_HOOK_TYPE_NFTABLES));
-	if (ret)
-		goto cancel_nest;
-
 	nest2 = nla_nest_start(nlskb, NFNLA_HOOK_INFO_DESC);
 	if (!nest2)
 		goto cancel_nest;
@@ -171,7 +213,20 @@ static int nfnl_hook_dump_one(struct sk_buff *nlskb,
 	if (ret)
 		goto nla_put_failure;
 
-	ret = nfnl_hook_put_nft_chain_info(nlskb, ctx, seq, ops);
+	switch (ops->hook_ops_type) {
+	case NF_HOOK_OP_NF_TABLES:
+		ret = nfnl_hook_put_nft_chain_info(nlskb, ctx, seq, ops->priv);
+		break;
+	case NF_HOOK_OP_BPF:
+		ret = nfnl_hook_put_bpf_prog_info(nlskb, ctx, seq, ops->priv);
+		break;
+	case NF_HOOK_OP_UNDEFINED:
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		break;
+	}
+
 	if (ret)
 		goto nla_put_failure;
 
-- 
cgit v1.2.3-70-g09d2


From 0bdc6da88f5bac0f3f272cb6f545c2cc70e8c66a Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:02:57 +0200
Subject: netfilter: disallow bpf hook attachment at same priority

This is just to avoid ordering issues between multiple bpf programs,
this could be removed later in case it turns out to be too cautious.

bpf prog could still be shared with non-bpf hook, otherwise we'd have to
make conntrack hook registration fail just because a bpf program has
same priority.

Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-5-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 net/netfilter/core.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index 358220b58521..f0783e42108b 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -119,6 +119,18 @@ nf_hook_entries_grow(const struct nf_hook_entries *old,
 		for (i = 0; i < old_entries; i++) {
 			if (orig_ops[i] != &dummy_ops)
 				alloc_entries++;
+
+			/* Restrict BPF hook type to force a unique priority, not
+			 * shared at attach time.
+			 *
+			 * This is mainly to avoid ordering issues between two
+			 * different bpf programs, this doesn't prevent a normal
+			 * hook at same priority as a bpf one (we don't want to
+			 * prevent defrag, conntrack, iptables etc from attaching).
+			 */
+			if (reg->priority == orig_ops[i]->priority &&
+			    reg->hook_ops_type == NF_HOOK_OP_BPF)
+				return ERR_PTR(-EBUSY);
 		}
 	}
 
-- 
cgit v1.2.3-70-g09d2


From d0fe92fb5e3df6991c640fb9205d880b68603259 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:02:58 +0200
Subject: tools: bpftool: print netfilter link info

Dump protocol family, hook and priority value:
$ bpftool link
2: netfilter  prog 14
        ip input prio -128
        pids install(3264)
5: netfilter  prog 14
        ip6 forward prio 21
        pids a.out(3387)
9: netfilter  prog 14
        ip prerouting prio 123
        pids a.out(5700)
10: netfilter  prog 14
        ip input prio 21
        pids test2(5701)

v2: Quentin Monnet suggested to also add 'bpftool net' support:

$ bpftool net
xdp:

tc:

flow_dissector:

netfilter:

        ip prerouting prio 21 prog_id 14
        ip input prio -128 prog_id 14
        ip input prio 21 prog_id 14
        ip forward prio 21 prog_id 14
        ip output prio 21 prog_id 14
        ip postrouting prio 21 prog_id 14

'bpftool net' only dumps netfilter link type, links are sorted by protocol
family, hook and priority.

v5: fix bpf ci failure: libbpf needs small update to prog_type_name[]
    and probe_prog_load helper.
v4: don't fail with -EOPNOTSUPP in libbpf probe_prog_load, update
    prog_type_name[] with "netfilter" entry (bpf ci)
v3: fix bpf.h copy, 'reserved' member was removed (Alexei)
    use p_err, not fprintf (Quentin)

Suggested-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/bpf/eeeaac99-9053-90c2-aa33-cc1ecb1ae9ca@isovalent.com/
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-6-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/link.c       |  83 ++++++++++++++++++++++++++++++++
 tools/bpf/bpftool/main.h       |   3 ++
 tools/bpf/bpftool/net.c        | 106 +++++++++++++++++++++++++++++++++++++++++
 tools/include/uapi/linux/bpf.h |  14 ++++++
 tools/lib/bpf/libbpf.c         |   3 ++
 tools/lib/bpf/libbpf_probes.c  |   1 +
 6 files changed, 210 insertions(+)

diff --git a/tools/bpf/bpftool/link.c b/tools/bpf/bpftool/link.c
index f985b79cca27..d98dbc50cf4c 100644
--- a/tools/bpf/bpftool/link.c
+++ b/tools/bpf/bpftool/link.c
@@ -3,6 +3,8 @@
 
 #include <errno.h>
 #include <linux/err.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter_arp.h>
 #include <net/if.h>
 #include <stdio.h>
 #include <unistd.h>
@@ -135,6 +137,18 @@ static void show_iter_json(struct bpf_link_info *info, json_writer_t *wtr)
 	}
 }
 
+void netfilter_dump_json(const struct bpf_link_info *info, json_writer_t *wtr)
+{
+	jsonw_uint_field(json_wtr, "pf",
+			 info->netfilter.pf);
+	jsonw_uint_field(json_wtr, "hook",
+			 info->netfilter.hooknum);
+	jsonw_int_field(json_wtr, "prio",
+			 info->netfilter.priority);
+	jsonw_uint_field(json_wtr, "flags",
+			 info->netfilter.flags);
+}
+
 static int get_prog_info(int prog_id, struct bpf_prog_info *info)
 {
 	__u32 len = sizeof(*info);
@@ -195,6 +209,10 @@ static int show_link_close_json(int fd, struct bpf_link_info *info)
 				 info->netns.netns_ino);
 		show_link_attach_type_json(info->netns.attach_type, json_wtr);
 		break;
+	case BPF_LINK_TYPE_NETFILTER:
+		netfilter_dump_json(info, json_wtr);
+		break;
+
 	default:
 		break;
 	}
@@ -263,6 +281,68 @@ static void show_iter_plain(struct bpf_link_info *info)
 	}
 }
 
+static const char * const pf2name[] = {
+	[NFPROTO_INET] = "inet",
+	[NFPROTO_IPV4] = "ip",
+	[NFPROTO_ARP] = "arp",
+	[NFPROTO_NETDEV] = "netdev",
+	[NFPROTO_BRIDGE] = "bridge",
+	[NFPROTO_IPV6] = "ip6",
+};
+
+static const char * const inethook2name[] = {
+	[NF_INET_PRE_ROUTING] = "prerouting",
+	[NF_INET_LOCAL_IN] = "input",
+	[NF_INET_FORWARD] = "forward",
+	[NF_INET_LOCAL_OUT] = "output",
+	[NF_INET_POST_ROUTING] = "postrouting",
+};
+
+static const char * const arphook2name[] = {
+	[NF_ARP_IN] = "input",
+	[NF_ARP_OUT] = "output",
+};
+
+void netfilter_dump_plain(const struct bpf_link_info *info)
+{
+	const char *hookname = NULL, *pfname = NULL;
+	unsigned int hook = info->netfilter.hooknum;
+	unsigned int pf = info->netfilter.pf;
+
+	if (pf < ARRAY_SIZE(pf2name))
+		pfname = pf2name[pf];
+
+	switch (pf) {
+	case NFPROTO_BRIDGE: /* bridge shares numbers with enum nf_inet_hooks */
+	case NFPROTO_IPV4:
+	case NFPROTO_IPV6:
+	case NFPROTO_INET:
+		if (hook < ARRAY_SIZE(inethook2name))
+			hookname = inethook2name[hook];
+		break;
+	case NFPROTO_ARP:
+		if (hook < ARRAY_SIZE(arphook2name))
+			hookname = arphook2name[hook];
+	default:
+		break;
+	}
+
+	if (pfname)
+		printf("\n\t%s", pfname);
+	else
+		printf("\n\tpf: %d", pf);
+
+	if (hookname)
+		printf(" %s", hookname);
+	else
+		printf(", hook %u,", hook);
+
+	printf(" prio %d", info->netfilter.priority);
+
+	if (info->netfilter.flags)
+		printf(" flags 0x%x", info->netfilter.flags);
+}
+
 static int show_link_close_plain(int fd, struct bpf_link_info *info)
 {
 	struct bpf_prog_info prog_info;
@@ -301,6 +381,9 @@ static int show_link_close_plain(int fd, struct bpf_link_info *info)
 		printf("\n\tnetns_ino %u  ", info->netns.netns_ino);
 		show_link_attach_type_plain(info->netns.attach_type);
 		break;
+	case BPF_LINK_TYPE_NETFILTER:
+		netfilter_dump_plain(info);
+		break;
 	default:
 		break;
 	}
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 9a95788e654d..a49534d7eafa 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -267,4 +267,7 @@ static inline bool hashmap__empty(struct hashmap *map)
 int pathname_concat(char *buf, int buf_sz, const char *path,
 		    const char *name);
 
+/* print netfilter bpf_link info */
+void netfilter_dump_plain(const struct bpf_link_info *info);
+void netfilter_dump_json(const struct bpf_link_info *info, json_writer_t *wtr);
 #endif
diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c
index c40e44c938ae..26a49965bf71 100644
--- a/tools/bpf/bpftool/net.c
+++ b/tools/bpf/bpftool/net.c
@@ -647,6 +647,108 @@ static int do_detach(int argc, char **argv)
 	return 0;
 }
 
+static int netfilter_link_compar(const void *a, const void *b)
+{
+	const struct bpf_link_info *nfa = a;
+	const struct bpf_link_info *nfb = b;
+	int delta;
+
+	delta = nfa->netfilter.pf - nfb->netfilter.pf;
+	if (delta)
+		return delta;
+
+	delta = nfa->netfilter.hooknum - nfb->netfilter.hooknum;
+	if (delta)
+		return delta;
+
+	if (nfa->netfilter.priority < nfb->netfilter.priority)
+		return -1;
+	if (nfa->netfilter.priority > nfb->netfilter.priority)
+		return 1;
+
+	return nfa->netfilter.flags - nfb->netfilter.flags;
+}
+
+static void show_link_netfilter(void)
+{
+	unsigned int nf_link_len = 0, nf_link_count = 0;
+	struct bpf_link_info *nf_link_info = NULL;
+	__u32 id = 0;
+
+	while (true) {
+		struct bpf_link_info info;
+		int fd, err;
+		__u32 len;
+
+		err = bpf_link_get_next_id(id, &id);
+		if (err) {
+			if (errno == ENOENT)
+				break;
+			p_err("can't get next link: %s (id %d)", strerror(errno), id);
+			break;
+		}
+
+		fd = bpf_link_get_fd_by_id(id);
+		if (fd < 0) {
+			p_err("can't get link by id (%u): %s", id, strerror(errno));
+			continue;
+		}
+
+		memset(&info, 0, sizeof(info));
+		len = sizeof(info);
+
+		err = bpf_link_get_info_by_fd(fd, &info, &len);
+
+		close(fd);
+
+		if (err) {
+			p_err("can't get link info for fd %d: %s", fd, strerror(errno));
+			continue;
+		}
+
+		if (info.type != BPF_LINK_TYPE_NETFILTER)
+			continue;
+
+		if (nf_link_count >= nf_link_len) {
+			static const unsigned int max_link_count = INT_MAX / sizeof(info);
+			struct bpf_link_info *expand;
+
+			if (nf_link_count > max_link_count) {
+				p_err("cannot handle more than %u links\n", max_link_count);
+				break;
+			}
+
+			nf_link_len += 16;
+
+			expand = realloc(nf_link_info, nf_link_len * sizeof(info));
+			if (!expand) {
+				p_err("realloc: %s",  strerror(errno));
+				break;
+			}
+
+			nf_link_info = expand;
+		}
+
+		nf_link_info[nf_link_count] = info;
+		nf_link_count++;
+	}
+
+	qsort(nf_link_info, nf_link_count, sizeof(*nf_link_info), netfilter_link_compar);
+
+	for (id = 0; id < nf_link_count; id++) {
+		NET_START_OBJECT;
+		if (json_output)
+			netfilter_dump_json(&nf_link_info[id], json_wtr);
+		else
+			netfilter_dump_plain(&nf_link_info[id]);
+
+		NET_DUMP_UINT("id", " prog_id %u", nf_link_info[id].prog_id);
+		NET_END_OBJECT;
+	}
+
+	free(nf_link_info);
+}
+
 static int do_show(int argc, char **argv)
 {
 	struct bpf_attach_info attach_info = {};
@@ -701,6 +803,10 @@ static int do_show(int argc, char **argv)
 		NET_DUMP_UINT("id", "id %u", attach_info.flow_dissector_id);
 	NET_END_ARRAY("\n");
 
+	NET_START_ARRAY("netfilter", "%s:\n");
+	show_link_netfilter();
+	NET_END_ARRAY("\n");
+
 	NET_END_OBJECT;
 	if (json_output)
 		jsonw_end_array(json_wtr);
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 4b20a7269bee..1bb11a6ee667 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -986,6 +986,7 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_LSM,
 	BPF_PROG_TYPE_SK_LOOKUP,
 	BPF_PROG_TYPE_SYSCALL, /* a program that can execute syscalls */
+	BPF_PROG_TYPE_NETFILTER,
 };
 
 enum bpf_attach_type {
@@ -1050,6 +1051,7 @@ enum bpf_link_type {
 	BPF_LINK_TYPE_PERF_EVENT = 7,
 	BPF_LINK_TYPE_KPROBE_MULTI = 8,
 	BPF_LINK_TYPE_STRUCT_OPS = 9,
+	BPF_LINK_TYPE_NETFILTER = 10,
 
 	MAX_BPF_LINK_TYPE,
 };
@@ -1560,6 +1562,12 @@ union bpf_attr {
 				 */
 				__u64		cookie;
 			} tracing;
+			struct {
+				__u32		pf;
+				__u32		hooknum;
+				__s32		priority;
+				__u32		flags;
+			} netfilter;
 		};
 	} link_create;
 
@@ -6410,6 +6418,12 @@ struct bpf_link_info {
 		struct {
 			__u32 map_id;
 		} struct_ops;
+		struct {
+			__u32 pf;
+			__u32 hooknum;
+			__s32 priority;
+			__u32 flags;
+		} netfilter;
 	};
 } __attribute__((aligned(8)));
 
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 2600d8384252..1cbacf9e71f3 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -130,6 +130,7 @@ static const char * const link_type_name[] = {
 	[BPF_LINK_TYPE_PERF_EVENT]		= "perf_event",
 	[BPF_LINK_TYPE_KPROBE_MULTI]		= "kprobe_multi",
 	[BPF_LINK_TYPE_STRUCT_OPS]		= "struct_ops",
+	[BPF_LINK_TYPE_NETFILTER]		= "netfilter",
 };
 
 static const char * const map_type_name[] = {
@@ -201,6 +202,7 @@ static const char * const prog_type_name[] = {
 	[BPF_PROG_TYPE_LSM]			= "lsm",
 	[BPF_PROG_TYPE_SK_LOOKUP]		= "sk_lookup",
 	[BPF_PROG_TYPE_SYSCALL]			= "syscall",
+	[BPF_PROG_TYPE_NETFILTER]		= "netfilter",
 };
 
 static int __base_pr(enum libbpf_print_level level, const char *format,
@@ -8710,6 +8712,7 @@ static const struct bpf_sec_def section_defs[] = {
 	SEC_DEF("struct_ops+",		STRUCT_OPS, 0, SEC_NONE),
 	SEC_DEF("struct_ops.s+",	STRUCT_OPS, 0, SEC_SLEEPABLE),
 	SEC_DEF("sk_lookup",		SK_LOOKUP, BPF_SK_LOOKUP, SEC_ATTACHABLE),
+	SEC_DEF("netfilter",		NETFILTER, 0, SEC_NONE),
 };
 
 static size_t custom_sec_def_cnt;
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index 4f3bc968ff8e..6065f408a59c 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -180,6 +180,7 @@ static int probe_prog_load(enum bpf_prog_type prog_type,
 	case BPF_PROG_TYPE_SK_REUSEPORT:
 	case BPF_PROG_TYPE_FLOW_DISSECTOR:
 	case BPF_PROG_TYPE_CGROUP_SYSCTL:
+	case BPF_PROG_TYPE_NETFILTER:
 		break;
 	default:
 		return -EOPNOTSUPP;
-- 
cgit v1.2.3-70-g09d2


From 2b99ef22e0d237e08bfc437e7d051f78f352aeb2 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:02:59 +0200
Subject: bpf: add test_run support for netfilter program type

add glue code so a bpf program can be run using userspace-provided
netfilter state and packet/skb.

Default is to use ipv4:output hook point, but this can be overridden by
userspace.  Userspace provided netfilter state is restricted, only hook and
protocol families can be overridden and only to ipv4/ipv6.

Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-7-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h         |   3 +
 net/bpf/test_run.c          | 158 ++++++++++++++++++++++++++++++++++++++++++++
 net/netfilter/nf_bpf_link.c |   1 +
 3 files changed, 162 insertions(+)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 18b592fde896..e53ceee1df37 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2264,6 +2264,9 @@ int bpf_prog_test_run_raw_tp(struct bpf_prog *prog,
 int bpf_prog_test_run_sk_lookup(struct bpf_prog *prog,
 				const union bpf_attr *kattr,
 				union bpf_attr __user *uattr);
+int bpf_prog_test_run_nf(struct bpf_prog *prog,
+			 const union bpf_attr *kattr,
+			 union bpf_attr __user *uattr);
 bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 		    const struct bpf_prog *prog,
 		    struct bpf_insn_access_aux *info);
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index f170e8a17974..e79e3a415ca9 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -19,7 +19,9 @@
 #include <linux/error-injection.h>
 #include <linux/smp.h>
 #include <linux/sock_diag.h>
+#include <linux/netfilter.h>
 #include <net/xdp.h>
+#include <net/netfilter/nf_bpf_link.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/bpf_test_run.h>
@@ -1691,6 +1693,162 @@ out:
 	return err;
 }
 
+static int verify_and_copy_hook_state(struct nf_hook_state *state,
+				      const struct nf_hook_state *user,
+				      struct net_device *dev)
+{
+	if (user->in || user->out)
+		return -EINVAL;
+
+	if (user->net || user->sk || user->okfn)
+		return -EINVAL;
+
+	switch (user->pf) {
+	case NFPROTO_IPV4:
+	case NFPROTO_IPV6:
+		switch (state->hook) {
+		case NF_INET_PRE_ROUTING:
+			state->in = dev;
+			break;
+		case NF_INET_LOCAL_IN:
+			state->in = dev;
+			break;
+		case NF_INET_FORWARD:
+			state->in = dev;
+			state->out = dev;
+			break;
+		case NF_INET_LOCAL_OUT:
+			state->out = dev;
+			break;
+		case NF_INET_POST_ROUTING:
+			state->out = dev;
+			break;
+		}
+
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	state->pf = user->pf;
+	state->hook = user->hook;
+
+	return 0;
+}
+
+static __be16 nfproto_eth(int nfproto)
+{
+	switch (nfproto) {
+	case NFPROTO_IPV4:
+		return htons(ETH_P_IP);
+	case NFPROTO_IPV6:
+		break;
+	}
+
+	return htons(ETH_P_IPV6);
+}
+
+int bpf_prog_test_run_nf(struct bpf_prog *prog,
+			 const union bpf_attr *kattr,
+			 union bpf_attr __user *uattr)
+{
+	struct net *net = current->nsproxy->net_ns;
+	struct net_device *dev = net->loopback_dev;
+	struct nf_hook_state *user_ctx, hook_state = {
+		.pf = NFPROTO_IPV4,
+		.hook = NF_INET_LOCAL_OUT,
+	};
+	u32 size = kattr->test.data_size_in;
+	u32 repeat = kattr->test.repeat;
+	struct bpf_nf_ctx ctx = {
+		.state = &hook_state,
+	};
+	struct sk_buff *skb = NULL;
+	u32 retval, duration;
+	void *data;
+	int ret;
+
+	if (kattr->test.flags || kattr->test.cpu || kattr->test.batch_size)
+		return -EINVAL;
+
+	if (size < sizeof(struct iphdr))
+		return -EINVAL;
+
+	data = bpf_test_init(kattr, kattr->test.data_size_in, size,
+			     NET_SKB_PAD + NET_IP_ALIGN,
+			     SKB_DATA_ALIGN(sizeof(struct skb_shared_info)));
+	if (IS_ERR(data))
+		return PTR_ERR(data);
+
+	if (!repeat)
+		repeat = 1;
+
+	user_ctx = bpf_ctx_init(kattr, sizeof(struct nf_hook_state));
+	if (IS_ERR(user_ctx)) {
+		kfree(data);
+		return PTR_ERR(user_ctx);
+	}
+
+	if (user_ctx) {
+		ret = verify_and_copy_hook_state(&hook_state, user_ctx, dev);
+		if (ret)
+			goto out;
+	}
+
+	skb = slab_build_skb(data);
+	if (!skb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	data = NULL; /* data released via kfree_skb */
+
+	skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN);
+	__skb_put(skb, size);
+
+	ret = -EINVAL;
+
+	if (hook_state.hook != NF_INET_LOCAL_OUT) {
+		if (size < ETH_HLEN + sizeof(struct iphdr))
+			goto out;
+
+		skb->protocol = eth_type_trans(skb, dev);
+		switch (skb->protocol) {
+		case htons(ETH_P_IP):
+			if (hook_state.pf == NFPROTO_IPV4)
+				break;
+			goto out;
+		case htons(ETH_P_IPV6):
+			if (size < ETH_HLEN + sizeof(struct ipv6hdr))
+				goto out;
+			if (hook_state.pf == NFPROTO_IPV6)
+				break;
+			goto out;
+		default:
+			ret = -EPROTO;
+			goto out;
+		}
+
+		skb_reset_network_header(skb);
+	} else {
+		skb->protocol = nfproto_eth(hook_state.pf);
+	}
+
+	ctx.skb = skb;
+
+	ret = bpf_test_run(prog, &ctx, repeat, &retval, &duration, false);
+	if (ret)
+		goto out;
+
+	ret = bpf_test_finish(kattr, uattr, NULL, NULL, 0, retval, duration);
+
+out:
+	kfree(user_ctx);
+	kfree_skb(skb);
+	kfree(data);
+	return ret;
+}
+
 static const struct btf_kfunc_id_set bpf_prog_test_kfunc_set = {
 	.owner = THIS_MODULE,
 	.set   = &test_sk_check_kfunc_ids,
diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c
index 49cfc5215386..c36da56d756f 100644
--- a/net/netfilter/nf_bpf_link.c
+++ b/net/netfilter/nf_bpf_link.c
@@ -166,6 +166,7 @@ int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
 }
 
 const struct bpf_prog_ops netfilter_prog_ops = {
+	.test_run = bpf_prog_test_run_nf,
 };
 
 static bool nf_ptr_to_btf_id(struct bpf_insn_access_aux *info, const char *name)
-- 
cgit v1.2.3-70-g09d2


From 006c0e44ed924140d44bc756e6ea36301fcea68d Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 21 Apr 2023 19:03:00 +0200
Subject: selftests/bpf: add missing netfilter return value and ctx access
 tests

Extend prog_tests with two test cases:

 # ./test_progs --allow=verifier_netfilter_retcode
 #278/1   verifier_netfilter_retcode/bpf_exit with invalid return code. test1:OK
 #278/2   verifier_netfilter_retcode/bpf_exit with valid return code. test2:OK
 #278/3   verifier_netfilter_retcode/bpf_exit with valid return code. test3:OK
 #278/4   verifier_netfilter_retcode/bpf_exit with invalid return code. test4:OK
 #278     verifier_netfilter_retcode:OK

This checks that only accept and drop (0,1) are permitted.

NF_QUEUE could be implemented later if we can guarantee that attachment
of such programs can be rejected if they get attached to a pf/hook that
doesn't support async reinjection.

NF_STOLEN could be implemented via trusted helpers that can guarantee
that the skb will eventually be free'd.

v4: test case for bpf_nf_ctx access checks, requested by Alexei Starovoitov.
v5: also check ctx->{state,skb} can be dereferenced (Alexei).

 # ./test_progs --allow=verifier_netfilter_ctx
 #281/1   verifier_netfilter_ctx/netfilter invalid context access, size too short:OK
 #281/2   verifier_netfilter_ctx/netfilter invalid context access, size too short:OK
 #281/3   verifier_netfilter_ctx/netfilter invalid context access, past end of ctx:OK
 #281/4   verifier_netfilter_ctx/netfilter invalid context, write:OK
 #281/5   verifier_netfilter_ctx/netfilter valid context read and invalid write:OK
 #281/6   verifier_netfilter_ctx/netfilter test prog with skb and state read access:OK
 #281/7   verifier_netfilter_ctx/netfilter test prog with skb and state read access @unpriv:OK
 #281     verifier_netfilter_ctx:OK
Summary: 1/7 PASSED, 0 SKIPPED, 0 FAILED

This checks:
1/2: partial reads of ctx->{skb,state} are rejected
3. read access past sizeof(ctx) is rejected
4. write to ctx content, e.g. 'ctx->skb = NULL;' is rejected
5. ctx->state content cannot be altered
6. ctx->state and ctx->skb can be dereferenced
7. ... same program fails for unpriv (CAP_NET_ADMIN needed).

Link: https://lore.kernel.org/bpf/20230419021152.sjq4gttphzzy6b5f@dhcp-172-26-102-232.dhcp.thefacebook.com/
Link: https://lore.kernel.org/bpf/20230420201655.77kkgi3dh7fesoll@MacBook-Pro-6.local/
Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230421170300.24115-8-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   4 +
 .../selftests/bpf/progs/verifier_netfilter_ctx.c   | 121 +++++++++++++++++++++
 .../bpf/progs/verifier_netfilter_retcode.c         |  49 +++++++++
 3 files changed, 174 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_netfilter_retcode.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 7c68d78da9ea..7534f5499d11 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -29,6 +29,8 @@
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
 #include "verifier_meta_access.skel.h"
+#include "verifier_netfilter_ctx.skel.h"
+#include "verifier_netfilter_retcode.skel.h"
 #include "verifier_raw_stack.skel.h"
 #include "verifier_raw_tp_writable.skel.h"
 #include "verifier_reg_equal.skel.h"
@@ -103,6 +105,8 @@ void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
+void test_verifier_netfilter_ctx(void)        { RUN(verifier_netfilter_ctx); }
+void test_verifier_netfilter_retcode(void)    { RUN(verifier_netfilter_retcode); }
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_reg_equal(void)            { RUN(verifier_reg_equal); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c b/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c
new file mode 100644
index 000000000000..65bba330e7e5
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c
@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include "vmlinux.h"
+
+#include "bpf_misc.h"
+
+#include <bpf/bpf_endian.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+
+SEC("netfilter")
+__description("netfilter invalid context access, size too short")
+__failure __msg("invalid bpf_context access")
+__naked void with_invalid_ctx_access_test1(void)
+{
+	asm volatile ("					\
+	r2 = *(u8*)(r1 + %[__bpf_nf_ctx_state]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__bpf_nf_ctx_state, offsetof(struct bpf_nf_ctx, state))
+	: __clobber_all);
+}
+
+SEC("netfilter")
+__description("netfilter invalid context access, size too short")
+__failure __msg("invalid bpf_context access")
+__naked void with_invalid_ctx_access_test2(void)
+{
+	asm volatile ("					\
+	r2 = *(u16*)(r1 + %[__bpf_nf_ctx_skb]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__bpf_nf_ctx_skb, offsetof(struct bpf_nf_ctx, skb))
+	: __clobber_all);
+}
+
+SEC("netfilter")
+__description("netfilter invalid context access, past end of ctx")
+__failure __msg("invalid bpf_context access")
+__naked void with_invalid_ctx_access_test3(void)
+{
+	asm volatile ("					\
+	r2 = *(u64*)(r1 + %[__bpf_nf_ctx_size]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__bpf_nf_ctx_size, sizeof(struct bpf_nf_ctx))
+	: __clobber_all);
+}
+
+SEC("netfilter")
+__description("netfilter invalid context, write")
+__failure __msg("invalid bpf_context access")
+__naked void with_invalid_ctx_access_test4(void)
+{
+	asm volatile ("					\
+	r2 = r1;					\
+	*(u64*)(r2 + 0) = r1;				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm_const(__bpf_nf_ctx_skb, offsetof(struct bpf_nf_ctx, skb))
+	: __clobber_all);
+}
+
+#define NF_DROP 0
+#define NF_ACCEPT 1
+
+SEC("netfilter")
+__description("netfilter valid context read and invalid write")
+__failure __msg("only read is supported")
+int with_invalid_ctx_access_test5(struct bpf_nf_ctx *ctx)
+{
+	struct nf_hook_state *state = (void *)ctx->state;
+
+	state->sk = NULL;
+	return NF_ACCEPT;
+}
+
+extern int bpf_dynptr_from_skb(struct sk_buff *skb, __u64 flags,
+                               struct bpf_dynptr *ptr__uninit) __ksym;
+extern void *bpf_dynptr_slice(const struct bpf_dynptr *ptr, uint32_t offset,
+                                   void *buffer, uint32_t buffer__sz) __ksym;
+
+SEC("netfilter")
+__description("netfilter test prog with skb and state read access")
+__success __failure_unpriv
+__retval(0)
+int with_valid_ctx_access_test6(struct bpf_nf_ctx *ctx)
+{
+	const struct nf_hook_state *state = ctx->state;
+	struct sk_buff *skb = ctx->skb;
+	const struct iphdr *iph;
+	const struct tcphdr *th;
+	u8 buffer_iph[20] = {};
+	u8 buffer_th[40] = {};
+	struct bpf_dynptr ptr;
+	uint8_t ihl;
+
+	if (skb->len <= 20 || bpf_dynptr_from_skb(skb, 0, &ptr))
+		return NF_ACCEPT;
+
+	iph = bpf_dynptr_slice(&ptr, 0, buffer_iph, sizeof(buffer_iph));
+	if (!iph)
+		return NF_ACCEPT;
+
+	if (state->pf != 2)
+		return NF_ACCEPT;
+
+	ihl = iph->ihl << 2;
+
+	th = bpf_dynptr_slice(&ptr, ihl, buffer_th, sizeof(buffer_th));
+	if (!th)
+		return NF_ACCEPT;
+
+	return th->dest == bpf_htons(22) ? NF_ACCEPT : NF_DROP;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/verifier_netfilter_retcode.c b/tools/testing/selftests/bpf/progs/verifier_netfilter_retcode.c
new file mode 100644
index 000000000000..353ae6da00e1
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_netfilter_retcode.c
@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("netfilter")
+__description("bpf_exit with invalid return code. test1")
+__failure __msg("R0 is not a known value")
+__naked void with_invalid_return_code_test1(void)
+{
+	asm volatile ("					\
+	r0 = *(u64*)(r1 + 0);				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("netfilter")
+__description("bpf_exit with valid return code. test2")
+__success
+__naked void with_valid_return_code_test2(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("netfilter")
+__description("bpf_exit with valid return code. test3")
+__success
+__naked void with_valid_return_code_test3(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("netfilter")
+__description("bpf_exit with invalid return code. test4")
+__failure __msg("R0 has value (0x2; 0x0)")
+__naked void with_invalid_return_code_test4(void)
+{
+	asm volatile ("					\
+	r0 = 2;						\
+	exit;						\
+"	::: __clobber_all);
+}
-- 
cgit v1.2.3-70-g09d2


From 63bb645b9da373c29ba4b5bea14e80c49e676694 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:11 +0300
Subject: selftests/bpf: Add notion of auxiliary programs for test_loader

In order to express test cases that use bpf_tail_call() intrinsic it
is necessary to have several programs to be loaded at a time.
This commit adds __auxiliary annotation to the set of annotations
supported by test_loader.c. Programs marked as auxiliary are always
loaded but are not treated as a separate test.

For example:

    void dummy_prog1(void);

    struct {
            __uint(type, BPF_MAP_TYPE_PROG_ARRAY);
            __uint(max_entries, 4);
            __uint(key_size, sizeof(int));
            __array(values, void (void));
    } prog_map SEC(".maps") = {
            .values = {
                    [0] = (void *) &dummy_prog1,
            },
    };

    SEC("tc")
    __auxiliary
    __naked void dummy_prog1(void) {
            asm volatile ("r0 = 42; exit;");
    }

    SEC("tc")
    __description("reference tracking: check reference or tail call")
    __success __retval(0)
    __naked void check_reference_or_tail_call(void)
    {
            asm volatile (
            "r2 = %[prog_map] ll;"
            "r3 = 0;"
            "call %[bpf_tail_call];"
            "r0 = 0;"
            "exit;"
            :: __imm(bpf_tail_call),
            :  __clobber_all);
    }

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-2-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_misc.h |  6 ++
 tools/testing/selftests/bpf/test_loader.c    | 89 +++++++++++++++++++++-------
 2 files changed, 73 insertions(+), 22 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 3b307de8dab9..d3c1217ba79a 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -53,6 +53,10 @@
  *                   - A numeric value.
  *                   Multiple __flag attributes could be specified, the final flags
  *                   value is derived by applying binary "or" to all specified values.
+ *
+ * __auxiliary         Annotated program is not a separate test, but used as auxiliary
+ *                     for some other test cases and should always be loaded.
+ * __auxiliary_unpriv  Same, but load program in unprivileged mode.
  */
 #define __msg(msg)		__attribute__((btf_decl_tag("comment:test_expect_msg=" msg)))
 #define __failure		__attribute__((btf_decl_tag("comment:test_expect_failure")))
@@ -65,6 +69,8 @@
 #define __flag(flag)		__attribute__((btf_decl_tag("comment:test_prog_flags="#flag)))
 #define __retval(val)		__attribute__((btf_decl_tag("comment:test_retval="#val)))
 #define __retval_unpriv(val)	__attribute__((btf_decl_tag("comment:test_retval_unpriv="#val)))
+#define __auxiliary		__attribute__((btf_decl_tag("comment:test_auxiliary")))
+#define __auxiliary_unpriv	__attribute__((btf_decl_tag("comment:test_auxiliary_unpriv")))
 
 /* Convenience macro for use with 'asm volatile' blocks */
 #define __naked __attribute__((naked))
diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index 40c9b7d532c4..b4edd8454934 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -25,6 +25,8 @@
 #define TEST_TAG_DESCRIPTION_PFX "comment:test_description="
 #define TEST_TAG_RETVAL_PFX "comment:test_retval="
 #define TEST_TAG_RETVAL_PFX_UNPRIV "comment:test_retval_unpriv="
+#define TEST_TAG_AUXILIARY "comment:test_auxiliary"
+#define TEST_TAG_AUXILIARY_UNPRIV "comment:test_auxiliary_unpriv"
 
 /* Warning: duplicated in bpf_misc.h */
 #define POINTER_VALUE	0xcafe4all
@@ -59,6 +61,8 @@ struct test_spec {
 	int log_level;
 	int prog_flags;
 	int mode_mask;
+	bool auxiliary;
+	bool valid;
 };
 
 static int tester_init(struct test_loader *tester)
@@ -87,6 +91,11 @@ static void free_test_spec(struct test_spec *spec)
 	free(spec->unpriv.name);
 	free(spec->priv.expect_msgs);
 	free(spec->unpriv.expect_msgs);
+
+	spec->priv.name = NULL;
+	spec->unpriv.name = NULL;
+	spec->priv.expect_msgs = NULL;
+	spec->unpriv.expect_msgs = NULL;
 }
 
 static int push_msg(const char *msg, struct test_subspec *subspec)
@@ -204,6 +213,12 @@ static int parse_test_spec(struct test_loader *tester,
 			spec->unpriv.expect_failure = false;
 			spec->mode_mask |= UNPRIV;
 			has_unpriv_result = true;
+		} else if (strcmp(s, TEST_TAG_AUXILIARY) == 0) {
+			spec->auxiliary = true;
+			spec->mode_mask |= PRIV;
+		} else if (strcmp(s, TEST_TAG_AUXILIARY_UNPRIV) == 0) {
+			spec->auxiliary = true;
+			spec->mode_mask |= UNPRIV;
 		} else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX)) {
 			msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
 			err = push_msg(msg, &spec->priv);
@@ -314,6 +329,8 @@ static int parse_test_spec(struct test_loader *tester,
 		}
 	}
 
+	spec->valid = true;
+
 	return 0;
 
 cleanup:
@@ -516,16 +533,18 @@ void run_subtest(struct test_loader *tester,
 		 struct bpf_object_open_opts *open_opts,
 		 const void *obj_bytes,
 		 size_t obj_byte_cnt,
+		 struct test_spec *specs,
 		 struct test_spec *spec,
 		 bool unpriv)
 {
 	struct test_subspec *subspec = unpriv ? &spec->unpriv : &spec->priv;
+	struct bpf_program *tprog, *tprog_iter;
+	struct test_spec *spec_iter;
 	struct cap_state caps = {};
-	struct bpf_program *tprog;
 	struct bpf_object *tobj;
 	struct bpf_map *map;
-	int retval;
-	int err;
+	int retval, err, i;
+	bool should_load;
 
 	if (!test__start_subtest(subspec->name))
 		return;
@@ -546,15 +565,23 @@ void run_subtest(struct test_loader *tester,
 	if (!ASSERT_OK_PTR(tobj, "obj_open_mem")) /* shouldn't happen */
 		goto subtest_cleanup;
 
-	bpf_object__for_each_program(tprog, tobj)
-		bpf_program__set_autoload(tprog, false);
+	i = 0;
+	bpf_object__for_each_program(tprog_iter, tobj) {
+		spec_iter = &specs[i++];
+		should_load = false;
+
+		if (spec_iter->valid) {
+			if (strcmp(bpf_program__name(tprog_iter), spec->prog_name) == 0) {
+				tprog = tprog_iter;
+				should_load = true;
+			}
 
-	bpf_object__for_each_program(tprog, tobj) {
-		/* only load specified program */
-		if (strcmp(bpf_program__name(tprog), spec->prog_name) == 0) {
-			bpf_program__set_autoload(tprog, true);
-			break;
+			if (spec_iter->auxiliary &&
+			    spec_iter->mode_mask & (unpriv ? UNPRIV : PRIV))
+				should_load = true;
 		}
+
+		bpf_program__set_autoload(tprog_iter, should_load);
 	}
 
 	prepare_case(tester, spec, tobj, tprog);
@@ -617,11 +644,12 @@ static void process_subtest(struct test_loader *tester,
 			    skel_elf_bytes_fn elf_bytes_factory)
 {
 	LIBBPF_OPTS(bpf_object_open_opts, open_opts, .object_name = skel_name);
+	struct test_spec *specs = NULL;
 	struct bpf_object *obj = NULL;
 	struct bpf_program *prog;
 	const void *obj_bytes;
+	int err, i, nr_progs;
 	size_t obj_byte_cnt;
-	int err;
 
 	if (tester_init(tester) < 0)
 		return; /* failed to initialize tester */
@@ -631,25 +659,42 @@ static void process_subtest(struct test_loader *tester,
 	if (!ASSERT_OK_PTR(obj, "obj_open_mem"))
 		return;
 
-	bpf_object__for_each_program(prog, obj) {
-		struct test_spec spec;
+	nr_progs = 0;
+	bpf_object__for_each_program(prog, obj)
+		++nr_progs;
+
+	specs = calloc(nr_progs, sizeof(struct test_spec));
+	if (!ASSERT_OK_PTR(specs, "Can't alloc specs array"))
+		return;
 
-		/* if we can't derive test specification, go to the next test */
-		err = parse_test_spec(tester, obj, prog, &spec);
-		if (err) {
+	i = 0;
+	bpf_object__for_each_program(prog, obj) {
+		/* ignore tests for which  we can't derive test specification */
+		err = parse_test_spec(tester, obj, prog, &specs[i++]);
+		if (err)
 			PRINT_FAIL("Can't parse test spec for program '%s'\n",
 				   bpf_program__name(prog));
+	}
+
+	i = 0;
+	bpf_object__for_each_program(prog, obj) {
+		struct test_spec *spec = &specs[i++];
+
+		if (!spec->valid || spec->auxiliary)
 			continue;
-		}
 
-		if (spec.mode_mask & PRIV)
-			run_subtest(tester, &open_opts, obj_bytes, obj_byte_cnt, &spec, false);
-		if (spec.mode_mask & UNPRIV)
-			run_subtest(tester, &open_opts, obj_bytes, obj_byte_cnt, &spec, true);
+		if (spec->mode_mask & PRIV)
+			run_subtest(tester, &open_opts, obj_bytes, obj_byte_cnt,
+				    specs, spec, false);
+		if (spec->mode_mask & UNPRIV)
+			run_subtest(tester, &open_opts, obj_bytes, obj_byte_cnt,
+				    specs, spec, true);
 
-		free_test_spec(&spec);
 	}
 
+	for (i = 0; i < nr_progs; ++i)
+		free_test_spec(&specs[i]);
+	free(specs);
 	bpf_object__close(obj);
 }
 
-- 
cgit v1.2.3-70-g09d2


From c92336559ac0127250cb3ffa9e6f69f9d7e81779 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:12 +0300
Subject: selftests/bpf: verifier/bounds converted to inline assembly

Test verifier/bounds automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-3-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |    2 +
 .../testing/selftests/bpf/progs/verifier_bounds.c  | 1076 ++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/bounds.c      |  884 ----------------
 3 files changed, 1078 insertions(+), 884 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_bounds.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/bounds.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 7534f5499d11..d71cbd63d94d 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -6,6 +6,7 @@
 #include "verifier_and.skel.h"
 #include "verifier_array_access.skel.h"
 #include "verifier_basic_stack.skel.h"
+#include "verifier_bounds.skel.h"
 #include "verifier_bounds_deduction.skel.h"
 #include "verifier_bounds_deduction_non_const.skel.h"
 #include "verifier_bounds_mix_sign_unsign.skel.h"
@@ -82,6 +83,7 @@ static void run_tests_aux(const char *skel_name,
 
 void test_verifier_and(void)                  { RUN(verifier_and); }
 void test_verifier_basic_stack(void)          { RUN(verifier_basic_stack); }
+void test_verifier_bounds(void)               { RUN(verifier_bounds); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
 void test_verifier_bounds_deduction_non_const(void)     { RUN(verifier_bounds_deduction_non_const); }
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds.c b/tools/testing/selftests/bpf/progs/verifier_bounds.c
new file mode 100644
index 000000000000..c5588a14fe2e
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_bounds.c
@@ -0,0 +1,1076 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/bounds.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("socket")
+__description("subtraction bounds (map value) variant 1")
+__failure __msg("R0 max value is outside of the allowed memory range")
+__failure_unpriv
+__naked void bounds_map_value_variant_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	if r1 > 0xff goto l0_%=;			\
+	r3 = *(u8*)(r0 + 1);				\
+	if r3 > 0xff goto l0_%=;			\
+	r1 -= r3;					\
+	r1 >>= 56;					\
+	r0 += r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("subtraction bounds (map value) variant 2")
+__failure
+__msg("R0 min value is negative, either use unsigned index or do a if (index >=0) check.")
+__msg_unpriv("R1 has unknown scalar with mixed signed bounds")
+__naked void bounds_map_value_variant_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	if r1 > 0xff goto l0_%=;			\
+	r3 = *(u8*)(r0 + 1);				\
+	if r3 > 0xff goto l0_%=;			\
+	r1 -= r3;					\
+	r0 += r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("check subtraction on pointers for unpriv")
+__success __failure_unpriv __msg_unpriv("R9 pointer -= pointer prohibited")
+__retval(0)
+__naked void subtraction_on_pointers_for_unpriv(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = %[map_hash_8b] ll;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r6 = 9;						\
+	*(u64*)(r2 + 0) = r6;				\
+	call %[bpf_map_lookup_elem];			\
+	r9 = r10;					\
+	r9 -= r0;					\
+	r1 = %[map_hash_8b] ll;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r6 = 0;						\
+	*(u64*)(r2 + 0) = r6;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	*(u64*)(r0 + 0) = r9;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check based on zero-extended MOV")
+__success __success_unpriv __retval(0)
+__naked void based_on_zero_extended_mov(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r2 = 0x0000'0000'ffff'ffff */		\
+	w2 = 0xffffffff;				\
+	/* r2 = 0 */					\
+	r2 >>= 32;					\
+	/* no-op */					\
+	r0 += r2;					\
+	/* access at offset 0 */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check based on sign-extended MOV. test1")
+__failure __msg("map_value pointer and 4294967295")
+__failure_unpriv
+__naked void on_sign_extended_mov_test1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r2 = 0xffff'ffff'ffff'ffff */		\
+	r2 = 0xffffffff;				\
+	/* r2 = 0xffff'ffff */				\
+	r2 >>= 32;					\
+	/* r0 = <oob pointer> */			\
+	r0 += r2;					\
+	/* access to OOB pointer */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check based on sign-extended MOV. test2")
+__failure __msg("R0 min value is outside of the allowed memory range")
+__failure_unpriv
+__naked void on_sign_extended_mov_test2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r2 = 0xffff'ffff'ffff'ffff */		\
+	r2 = 0xffffffff;				\
+	/* r2 = 0xfff'ffff */				\
+	r2 >>= 36;					\
+	/* r0 = <oob pointer> */			\
+	r0 += r2;					\
+	/* access to OOB pointer */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("bounds check based on reg_off + var_off + insn_off. test1")
+__failure __msg("value_size=8 off=1073741825")
+__naked void var_off_insn_off_test1(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r6 &= 1;					\
+	r6 += %[__imm_0];				\
+	r0 += r6;					\
+	r0 += %[__imm_0];				\
+l0_%=:	r0 = *(u8*)(r0 + 3);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__imm_0, (1 << 29) - 1),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("bounds check based on reg_off + var_off + insn_off. test2")
+__failure __msg("value 1073741823")
+__naked void var_off_insn_off_test2(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r6 &= 1;					\
+	r6 += %[__imm_0];				\
+	r0 += r6;					\
+	r0 += %[__imm_1];				\
+l0_%=:	r0 = *(u8*)(r0 + 3);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__imm_0, (1 << 30) - 1),
+	  __imm_const(__imm_1, (1 << 29) - 1),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after truncation of non-boundary-crossing range")
+__success __success_unpriv __retval(0)
+__naked void of_non_boundary_crossing_range(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r1 = [0x00, 0xff] */				\
+	r1 = *(u8*)(r0 + 0);				\
+	r2 = 1;						\
+	/* r2 = 0x10'0000'0000 */			\
+	r2 <<= 36;					\
+	/* r1 = [0x10'0000'0000, 0x10'0000'00ff] */	\
+	r1 += r2;					\
+	/* r1 = [0x10'7fff'ffff, 0x10'8000'00fe] */	\
+	r1 += 0x7fffffff;				\
+	/* r1 = [0x00, 0xff] */				\
+	w1 -= 0x7fffffff;				\
+	/* r1 = 0 */					\
+	r1 >>= 8;					\
+	/* no-op */					\
+	r0 += r1;					\
+	/* access at offset 0 */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after truncation of boundary-crossing range (1)")
+__failure
+/* not actually fully unbounded, but the bound is very high */
+__msg("value -4294967168 makes map_value pointer be out of bounds")
+__failure_unpriv
+__naked void of_boundary_crossing_range_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r1 = [0x00, 0xff] */				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 += %[__imm_0];				\
+	/* r1 = [0xffff'ff80, 0x1'0000'007f] */		\
+	r1 += %[__imm_0];				\
+	/* r1 = [0xffff'ff80, 0xffff'ffff] or		\
+	 *      [0x0000'0000, 0x0000'007f]		\
+	 */						\
+	w1 += 0;					\
+	r1 -= %[__imm_0];				\
+	/* r1 = [0x00, 0xff] or				\
+	 *      [0xffff'ffff'0000'0080, 0xffff'ffff'ffff'ffff]\
+	 */						\
+	r1 -= %[__imm_0];				\
+	/* error on OOB pointer computation */		\
+	r0 += r1;					\
+	/* exit */					\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__imm_0, 0xffffff80 >> 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after truncation of boundary-crossing range (2)")
+__failure __msg("value -4294967168 makes map_value pointer be out of bounds")
+__failure_unpriv
+__naked void of_boundary_crossing_range_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r1 = [0x00, 0xff] */				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 += %[__imm_0];				\
+	/* r1 = [0xffff'ff80, 0x1'0000'007f] */		\
+	r1 += %[__imm_0];				\
+	/* r1 = [0xffff'ff80, 0xffff'ffff] or		\
+	 *      [0x0000'0000, 0x0000'007f]		\
+	 * difference to previous test: truncation via MOV32\
+	 * instead of ALU32.				\
+	 */						\
+	w1 = w1;					\
+	r1 -= %[__imm_0];				\
+	/* r1 = [0x00, 0xff] or				\
+	 *      [0xffff'ffff'0000'0080, 0xffff'ffff'ffff'ffff]\
+	 */						\
+	r1 -= %[__imm_0];				\
+	/* error on OOB pointer computation */		\
+	r0 += r1;					\
+	/* exit */					\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b),
+	  __imm_const(__imm_0, 0xffffff80 >> 1)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after wrapping 32-bit addition")
+__success __success_unpriv __retval(0)
+__naked void after_wrapping_32_bit_addition(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r1 = 0x7fff'ffff */				\
+	r1 = 0x7fffffff;				\
+	/* r1 = 0xffff'fffe */				\
+	r1 += 0x7fffffff;				\
+	/* r1 = 0 */					\
+	w1 += 2;					\
+	/* no-op */					\
+	r0 += r1;					\
+	/* access at offset 0 */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after shift with oversized count operand")
+__failure __msg("R0 max value is outside of the allowed memory range")
+__failure_unpriv
+__naked void shift_with_oversized_count_operand(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r2 = 32;					\
+	r1 = 1;						\
+	/* r1 = (u32)1 << (u32)32 = ? */		\
+	w1 <<= w2;					\
+	/* r1 = [0x0000, 0xffff] */			\
+	r1 &= 0xffff;					\
+	/* computes unknown pointer, potentially OOB */	\
+	r0 += r1;					\
+	/* potentially OOB access */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after right shift of maybe-negative number")
+__failure __msg("R0 unbounded memory access")
+__failure_unpriv
+__naked void shift_of_maybe_negative_number(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	/* r1 = [0x00, 0xff] */				\
+	r1 = *(u8*)(r0 + 0);				\
+	/* r1 = [-0x01, 0xfe] */			\
+	r1 -= 1;					\
+	/* r1 = 0 or 0xff'ffff'ffff'ffff */		\
+	r1 >>= 8;					\
+	/* r1 = 0 or 0xffff'ffff'ffff */		\
+	r1 >>= 8;					\
+	/* computes unknown pointer, potentially OOB */	\
+	r0 += r1;					\
+	/* potentially OOB access */			\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	/* exit */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check after 32-bit right shift with 64-bit input")
+__failure __msg("math between map_value pointer and 4294967294 is not allowed")
+__failure_unpriv
+__naked void shift_with_64_bit_input(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 2;						\
+	/* r1 = 1<<32 */				\
+	r1 <<= 31;					\
+	/* r1 = 0 (NOT 2!) */				\
+	w1 >>= 31;					\
+	/* r1 = 0xffff'fffe (NOT 0!) */			\
+	w1 -= 2;					\
+	/* error on computing OOB pointer */		\
+	r0 += r1;					\
+	/* exit */					\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check map access with off+size signed 32bit overflow. test1")
+__failure __msg("map_value pointer and 2147483646")
+__failure_unpriv
+__naked void size_signed_32bit_overflow_test1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r0 += 0x7ffffffe;				\
+	r0 = *(u64*)(r0 + 0);				\
+	goto l1_%=;					\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check map access with off+size signed 32bit overflow. test2")
+__failure __msg("pointer offset 1073741822")
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__naked void size_signed_32bit_overflow_test2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r0 += 0x1fffffff;				\
+	r0 += 0x1fffffff;				\
+	r0 += 0x1fffffff;				\
+	r0 = *(u64*)(r0 + 0);				\
+	goto l1_%=;					\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check map access with off+size signed 32bit overflow. test3")
+__failure __msg("pointer offset -1073741822")
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__naked void size_signed_32bit_overflow_test3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r0 -= 0x1fffffff;				\
+	r0 -= 0x1fffffff;				\
+	r0 = *(u64*)(r0 + 2);				\
+	goto l1_%=;					\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check map access with off+size signed 32bit overflow. test4")
+__failure __msg("map_value pointer and 1000000000000")
+__failure_unpriv
+__naked void size_signed_32bit_overflow_test4(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = 1000000;					\
+	r1 *= 1000000;					\
+	r0 += r1;					\
+	r0 = *(u64*)(r0 + 2);				\
+	goto l1_%=;					\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check mixed 32bit and 64bit arithmetic. test1")
+__success __failure_unpriv __msg_unpriv("R0 invalid mem access 'scalar'")
+__retval(0)
+__naked void _32bit_and_64bit_arithmetic_test1(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = -1;					\
+	r1 <<= 32;					\
+	r1 += 1;					\
+	/* r1 = 0xffffFFFF00000001 */			\
+	if w1 > 1 goto l0_%=;				\
+	/* check ALU64 op keeps 32bit bounds */		\
+	r1 += 1;					\
+	if w1 > 2 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	/* invalid ldx if bounds are lost above */	\
+	r0 = *(u64*)(r0 - 1);				\
+l1_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check mixed 32bit and 64bit arithmetic. test2")
+__success __failure_unpriv __msg_unpriv("R0 invalid mem access 'scalar'")
+__retval(0)
+__naked void _32bit_and_64bit_arithmetic_test2(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = -1;					\
+	r1 <<= 32;					\
+	r1 += 1;					\
+	/* r1 = 0xffffFFFF00000001 */			\
+	r2 = 3;						\
+	/* r1 = 0x2 */					\
+	w1 += 1;					\
+	/* check ALU32 op zero extends 64bit bounds */	\
+	if r1 > r2 goto l0_%=;				\
+	goto l1_%=;					\
+l0_%=:	/* invalid ldx if bounds are lost above */	\
+	r0 = *(u64*)(r0 - 1);				\
+l1_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("assigning 32bit bounds to 64bit for wA = 0, wB = wA")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void for_wa_0_wb_wa(void)
+{
+	asm volatile ("					\
+	r8 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r7 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	w9 = 0;						\
+	w2 = w9;					\
+	r6 = r7;					\
+	r6 += r2;					\
+	r3 = r6;					\
+	r3 += 8;					\
+	if r3 > r8 goto l0_%=;				\
+	r5 = *(u32*)(r6 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg = 0, reg xor 1")
+__success __failure_unpriv
+__msg_unpriv("R0 min value is outside of the allowed memory range")
+__retval(0)
+__naked void reg_0_reg_xor_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = 0;						\
+	r1 ^= 1;					\
+	if r1 != 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg32 = 0, reg32 xor 1")
+__success __failure_unpriv
+__msg_unpriv("R0 min value is outside of the allowed memory range")
+__retval(0)
+__naked void reg32_0_reg32_xor_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	w1 = 0;						\
+	w1 ^= 1;					\
+	if w1 != 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg = 2, reg xor 3")
+__success __failure_unpriv
+__msg_unpriv("R0 min value is outside of the allowed memory range")
+__retval(0)
+__naked void reg_2_reg_xor_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = 2;						\
+	r1 ^= 3;					\
+	if r1 > 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg = any, reg xor 3")
+__failure __msg("invalid access to map value")
+__msg_unpriv("invalid access to map value")
+__naked void reg_any_reg_xor_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = *(u64*)(r0 + 0);				\
+	r1 ^= 3;					\
+	if r1 != 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg32 = any, reg32 xor 3")
+__failure __msg("invalid access to map value")
+__msg_unpriv("invalid access to map value")
+__naked void reg32_any_reg32_xor_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = *(u64*)(r0 + 0);				\
+	w1 ^= 3;					\
+	if w1 != 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg > 0, reg xor 3")
+__success __failure_unpriv
+__msg_unpriv("R0 min value is outside of the allowed memory range")
+__retval(0)
+__naked void reg_0_reg_xor_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = *(u64*)(r0 + 0);				\
+	if r1 <= 0 goto l1_%=;				\
+	r1 ^= 3;					\
+	if r1 >= 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds check for reg32 > 0, reg32 xor 3")
+__success __failure_unpriv
+__msg_unpriv("R0 min value is outside of the allowed memory range")
+__retval(0)
+__naked void reg32_0_reg32_xor_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = *(u64*)(r0 + 0);				\
+	if w1 <= 0 goto l1_%=;				\
+	w1 ^= 3;					\
+	if w1 >= 0 goto l1_%=;				\
+	r0 = *(u64*)(r0 + 8);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks after 32-bit truncation. test 1")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0)
+__naked void _32_bit_truncation_test_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	/* This used to reduce the max bound to 0x7fffffff */\
+	if r1 == 0 goto l1_%=;				\
+	if r1 > 0x7fffffff goto l0_%=;			\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("bounds checks after 32-bit truncation. test 2")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0)
+__naked void _32_bit_truncation_test_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	if r1 s< 1 goto l1_%=;				\
+	if w1 s< 0 goto l0_%=;				\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("bound check with JMP_JLT for crossing 64-bit signed boundary")
+__success __retval(0)
+__naked void crossing_64_bit_signed_boundary_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 1;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = *(u8*)(r2 + 0);				\
+	r0 = 0x7fffffffffffff10 ll;			\
+	r1 += r0;					\
+	r0 = 0x8000000000000000 ll;			\
+l1_%=:	r0 += 1;					\
+	/* r1 unsigned range is [0x7fffffffffffff10, 0x800000000000000f] */\
+	if r0 < r1 goto l1_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("bound check with JMP_JSLT for crossing 64-bit signed boundary")
+__success __retval(0)
+__naked void crossing_64_bit_signed_boundary_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 1;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = *(u8*)(r2 + 0);				\
+	r0 = 0x7fffffffffffff10 ll;			\
+	r1 += r0;					\
+	r2 = 0x8000000000000fff ll;			\
+	r0 = 0x8000000000000000 ll;			\
+l1_%=:	r0 += 1;					\
+	if r0 s> r2 goto l0_%=;				\
+	/* r1 signed range is [S64_MIN, S64_MAX] */	\
+	if r0 s< r1 goto l1_%=;				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("bound check for loop upper bound greater than U32_MAX")
+__success __retval(0)
+__naked void bound_greater_than_u32_max(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 1;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = *(u8*)(r2 + 0);				\
+	r0 = 0x100000000 ll;				\
+	r1 += r0;					\
+	r0 = 0x100000000 ll;				\
+l1_%=:	r0 += 1;					\
+	if r0 < r1 goto l1_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("bound check with JMP32_JLT for crossing 32-bit signed boundary")
+__success __retval(0)
+__naked void crossing_32_bit_signed_boundary_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 1;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = *(u8*)(r2 + 0);				\
+	w0 = 0x7fffff10;				\
+	w1 += w0;					\
+	w0 = 0x80000000;				\
+l1_%=:	w0 += 1;					\
+	/* r1 unsigned range is [0, 0x8000000f] */	\
+	if w0 < w1 goto l1_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("bound check with JMP32_JSLT for crossing 32-bit signed boundary")
+__success __retval(0)
+__naked void crossing_32_bit_signed_boundary_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[xdp_md_data]);		\
+	r3 = *(u32*)(r1 + %[xdp_md_data_end]);		\
+	r1 = r2;					\
+	r1 += 1;					\
+	if r1 > r3 goto l0_%=;				\
+	r1 = *(u8*)(r2 + 0);				\
+	w0 = 0x7fffff10;				\
+	w1 += w0;					\
+	w2 = 0x80000fff;				\
+	w0 = 0x80000000;				\
+l1_%=:	w0 += 1;					\
+	if w0 s> w2 goto l0_%=;				\
+	/* r1 signed range is [S32_MIN, S32_MAX] */	\
+	if w0 s< w1 goto l1_%=;				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(xdp_md_data, offsetof(struct xdp_md, data)),
+	  __imm_const(xdp_md_data_end, offsetof(struct xdp_md, data_end))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/bounds.c b/tools/testing/selftests/bpf/verifier/bounds.c
deleted file mode 100644
index 43942ce8cf15..000000000000
--- a/tools/testing/selftests/bpf/verifier/bounds.c
+++ /dev/null
@@ -1,884 +0,0 @@
-{
-	"subtraction bounds (map value) variant 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_1, 0xff, 7),
-	BPF_LDX_MEM(BPF_B, BPF_REG_3, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, 0xff, 5),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_1, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_1, 56),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 max value is outside of the allowed memory range",
-	.result = REJECT,
-},
-{
-	"subtraction bounds (map value) variant 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_1, 0xff, 6),
-	BPF_LDX_MEM(BPF_B, BPF_REG_3, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_3, 0xff, 4),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_1, BPF_REG_3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 min value is negative, either use unsigned index or do a if (index >=0) check.",
-	.errstr_unpriv = "R1 has unknown scalar with mixed signed bounds",
-	.result = REJECT,
-},
-{
-	"check subtraction on pointers for unpriv",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_LD_MAP_FD(BPF_REG_ARG1, 0),
-	BPF_MOV64_REG(BPF_REG_ARG2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_ARG2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_ARG2, 0, 9),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_FP),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_9, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_ARG1, 0),
-	BPF_MOV64_REG(BPF_REG_ARG2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_ARG2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_ARG2, 0, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_STX_MEM(BPF_DW, BPF_REG_0, BPF_REG_9, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1, 9 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R9 pointer -= pointer prohibited",
-},
-{
-	"bounds check based on zero-extended MOV",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	/* r2 = 0x0000'0000'ffff'ffff */
-	BPF_MOV32_IMM(BPF_REG_2, 0xffffffff),
-	/* r2 = 0 */
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_2, 32),
-	/* no-op */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	/* access at offset 0 */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT
-},
-{
-	"bounds check based on sign-extended MOV. test1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	/* r2 = 0xffff'ffff'ffff'ffff */
-	BPF_MOV64_IMM(BPF_REG_2, 0xffffffff),
-	/* r2 = 0xffff'ffff */
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_2, 32),
-	/* r0 = <oob pointer> */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	/* access to OOB pointer */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "map_value pointer and 4294967295",
-	.result = REJECT
-},
-{
-	"bounds check based on sign-extended MOV. test2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	/* r2 = 0xffff'ffff'ffff'ffff */
-	BPF_MOV64_IMM(BPF_REG_2, 0xffffffff),
-	/* r2 = 0xfff'ffff */
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_2, 36),
-	/* r0 = <oob pointer> */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	/* access to OOB pointer */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 min value is outside of the allowed memory range",
-	.result = REJECT
-},
-{
-	"bounds check based on reg_off + var_off + insn_off. test1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_6, 1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, (1 << 29) - 1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, (1 << 29) - 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 3),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.errstr = "value_size=8 off=1073741825",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"bounds check based on reg_off + var_off + insn_off. test2",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_6, 1),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, (1 << 30) - 1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, (1 << 29) - 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 3),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 4 },
-	.errstr = "value 1073741823",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"bounds check after truncation of non-boundary-crossing range",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	/* r1 = [0x00, 0xff] */
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 1),
-	/* r2 = 0x10'0000'0000 */
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_2, 36),
-	/* r1 = [0x10'0000'0000, 0x10'0000'00ff] */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_2),
-	/* r1 = [0x10'7fff'ffff, 0x10'8000'00fe] */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x7fffffff),
-	/* r1 = [0x00, 0xff] */
-	BPF_ALU32_IMM(BPF_SUB, BPF_REG_1, 0x7fffffff),
-	/* r1 = 0 */
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_1, 8),
-	/* no-op */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* access at offset 0 */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT
-},
-{
-	"bounds check after truncation of boundary-crossing range (1)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	/* r1 = [0x00, 0xff] */
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0xffffff80 >> 1),
-	/* r1 = [0xffff'ff80, 0x1'0000'007f] */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0xffffff80 >> 1),
-	/* r1 = [0xffff'ff80, 0xffff'ffff] or
-	 *      [0x0000'0000, 0x0000'007f]
-	 */
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 0xffffff80 >> 1),
-	/* r1 = [0x00, 0xff] or
-	 *      [0xffff'ffff'0000'0080, 0xffff'ffff'ffff'ffff]
-	 */
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 0xffffff80 >> 1),
-	/* error on OOB pointer computation */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	/* not actually fully unbounded, but the bound is very high */
-	.errstr = "value -4294967168 makes map_value pointer be out of bounds",
-	.result = REJECT,
-},
-{
-	"bounds check after truncation of boundary-crossing range (2)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	/* r1 = [0x00, 0xff] */
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0xffffff80 >> 1),
-	/* r1 = [0xffff'ff80, 0x1'0000'007f] */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0xffffff80 >> 1),
-	/* r1 = [0xffff'ff80, 0xffff'ffff] or
-	 *      [0x0000'0000, 0x0000'007f]
-	 * difference to previous test: truncation via MOV32
-	 * instead of ALU32.
-	 */
-	BPF_MOV32_REG(BPF_REG_1, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 0xffffff80 >> 1),
-	/* r1 = [0x00, 0xff] or
-	 *      [0xffff'ffff'0000'0080, 0xffff'ffff'ffff'ffff]
-	 */
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 0xffffff80 >> 1),
-	/* error on OOB pointer computation */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "value -4294967168 makes map_value pointer be out of bounds",
-	.result = REJECT,
-},
-{
-	"bounds check after wrapping 32-bit addition",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	/* r1 = 0x7fff'ffff */
-	BPF_MOV64_IMM(BPF_REG_1, 0x7fffffff),
-	/* r1 = 0xffff'fffe */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0x7fffffff),
-	/* r1 = 0 */
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_1, 2),
-	/* no-op */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* access at offset 0 */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT
-},
-{
-	"bounds check after shift with oversized count operand",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_IMM(BPF_REG_2, 32),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	/* r1 = (u32)1 << (u32)32 = ? */
-	BPF_ALU32_REG(BPF_LSH, BPF_REG_1, BPF_REG_2),
-	/* r1 = [0x0000, 0xffff] */
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xffff),
-	/* computes unknown pointer, potentially OOB */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* potentially OOB access */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 max value is outside of the allowed memory range",
-	.result = REJECT
-},
-{
-	"bounds check after right shift of maybe-negative number",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	/* r1 = [0x00, 0xff] */
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	/* r1 = [-0x01, 0xfe] */
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 1),
-	/* r1 = 0 or 0xff'ffff'ffff'ffff */
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_1, 8),
-	/* r1 = 0 or 0xffff'ffff'ffff */
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_1, 8),
-	/* computes unknown pointer, potentially OOB */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* potentially OOB access */
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R0 unbounded memory access",
-	.result = REJECT
-},
-{
-	"bounds check after 32-bit right shift with 64-bit input",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	/* r1 = 2 */
-	BPF_MOV64_IMM(BPF_REG_1, 2),
-	/* r1 = 1<<32 */
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 31),
-	/* r1 = 0 (NOT 2!) */
-	BPF_ALU32_IMM(BPF_RSH, BPF_REG_1, 31),
-	/* r1 = 0xffff'fffe (NOT 0!) */
-	BPF_ALU32_IMM(BPF_SUB, BPF_REG_1, 2),
-	/* error on computing OOB pointer */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	/* exit */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "math between map_value pointer and 4294967294 is not allowed",
-	.result = REJECT,
-},
-{
-	"bounds check map access with off+size signed 32bit overflow. test1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 0x7ffffffe),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0),
-	BPF_JMP_A(0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "map_value pointer and 2147483646",
-	.result = REJECT
-},
-{
-	"bounds check map access with off+size signed 32bit overflow. test2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 0x1fffffff),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 0x1fffffff),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 0x1fffffff),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0),
-	BPF_JMP_A(0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "pointer offset 1073741822",
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.result = REJECT
-},
-{
-	"bounds check map access with off+size signed 32bit overflow. test3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_0, 0x1fffffff),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_0, 0x1fffffff),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 2),
-	BPF_JMP_A(0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "pointer offset -1073741822",
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.result = REJECT
-},
-{
-	"bounds check map access with off+size signed 32bit overflow. test4",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_1, 1000000),
-	BPF_ALU64_IMM(BPF_MUL, BPF_REG_1, 1000000),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 2),
-	BPF_JMP_A(0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "map_value pointer and 1000000000000",
-	.result = REJECT
-},
-{
-	"bounds check mixed 32bit and 64bit arithmetic. test1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 32),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	/* r1 = 0xffffFFFF00000001 */
-	BPF_JMP32_IMM(BPF_JGT, BPF_REG_1, 1, 3),
-	/* check ALU64 op keeps 32bit bounds */
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP32_IMM(BPF_JGT, BPF_REG_1, 2, 1),
-	BPF_JMP_A(1),
-	/* invalid ldx if bounds are lost above */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, -1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 invalid mem access 'scalar'",
-	.result_unpriv = REJECT,
-	.result = ACCEPT
-},
-{
-	"bounds check mixed 32bit and 64bit arithmetic. test2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 32),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	/* r1 = 0xffffFFFF00000001 */
-	BPF_MOV64_IMM(BPF_REG_2, 3),
-	/* r1 = 0x2 */
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_1, 1),
-	/* check ALU32 op zero extends 64bit bounds */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_2, 1),
-	BPF_JMP_A(1),
-	/* invalid ldx if bounds are lost above */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, -1),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 invalid mem access 'scalar'",
-	.result_unpriv = REJECT,
-	.result = ACCEPT
-},
-{
-	"assigning 32bit bounds to 64bit for wA = 0, wB = wA",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_MOV32_IMM(BPF_REG_9, 0),
-	BPF_MOV32_REG(BPF_REG_2, BPF_REG_9),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_8, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_5, BPF_REG_6, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"bounds check for reg = 0, reg xor 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_XOR, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 min value is outside of the allowed memory range",
-	.result_unpriv = REJECT,
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-},
-{
-	"bounds check for reg32 = 0, reg32 xor 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_ALU32_IMM(BPF_XOR, BPF_REG_1, 1),
-	BPF_JMP32_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 min value is outside of the allowed memory range",
-	.result_unpriv = REJECT,
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-},
-{
-	"bounds check for reg = 2, reg xor 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_1, 2),
-	BPF_ALU64_IMM(BPF_XOR, BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 min value is outside of the allowed memory range",
-	.result_unpriv = REJECT,
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-},
-{
-	"bounds check for reg = any, reg xor 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_XOR, BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = REJECT,
-	.errstr = "invalid access to map value",
-	.errstr_unpriv = "invalid access to map value",
-},
-{
-	"bounds check for reg32 = any, reg32 xor 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU32_IMM(BPF_XOR, BPF_REG_1, 3),
-	BPF_JMP32_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = REJECT,
-	.errstr = "invalid access to map value",
-	.errstr_unpriv = "invalid access to map value",
-},
-{
-	"bounds check for reg > 0, reg xor 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JLE, BPF_REG_1, 0, 3),
-	BPF_ALU64_IMM(BPF_XOR, BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 min value is outside of the allowed memory range",
-	.result_unpriv = REJECT,
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-},
-{
-	"bounds check for reg32 > 0, reg32 xor 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP32_IMM(BPF_JLE, BPF_REG_1, 0, 3),
-	BPF_ALU32_IMM(BPF_XOR, BPF_REG_1, 3),
-	BPF_JMP32_IMM(BPF_JGE, BPF_REG_1, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R0 min value is outside of the allowed memory range",
-	.result_unpriv = REJECT,
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-},
-{
-	"bounds checks after 32-bit truncation. test 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	/* This used to reduce the max bound to 0x7fffffff */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_1, 0x7fffffff, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"bounds checks after 32-bit truncation. test 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_1, 1, 1),
-	BPF_JMP32_IMM(BPF_JSLT, BPF_REG_1, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"bound check with JMP_JLT for crossing 64-bit signed boundary",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 8),
-
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
-	BPF_LD_IMM64(BPF_REG_0, 0x7fffffffffffff10),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-
-	BPF_LD_IMM64(BPF_REG_0, 0x8000000000000000),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	/* r1 unsigned range is [0x7fffffffffffff10, 0x800000000000000f] */
-	BPF_JMP_REG(BPF_JLT, BPF_REG_0, BPF_REG_1, -2),
-
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"bound check with JMP_JSLT for crossing 64-bit signed boundary",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 13),
-
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
-	BPF_LD_IMM64(BPF_REG_0, 0x7fffffffffffff10),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-
-	BPF_LD_IMM64(BPF_REG_2, 0x8000000000000fff),
-	BPF_LD_IMM64(BPF_REG_0, 0x8000000000000000),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_REG(BPF_JSGT, BPF_REG_0, BPF_REG_2, 3),
-	/* r1 signed range is [S64_MIN, S64_MAX] */
-	BPF_JMP_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -3),
-
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"bound check for loop upper bound greater than U32_MAX",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 8),
-
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
-	BPF_LD_IMM64(BPF_REG_0, 0x100000000),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-
-	BPF_LD_IMM64(BPF_REG_0, 0x100000000),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_0, BPF_REG_1, -2),
-
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"bound check with JMP32_JLT for crossing 32-bit signed boundary",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 6),
-
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
-	BPF_MOV32_IMM(BPF_REG_0, 0x7fffff10),
-	BPF_ALU32_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-
-	BPF_MOV32_IMM(BPF_REG_0, 0x80000000),
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 1),
-	/* r1 unsigned range is [0, 0x8000000f] */
-	BPF_JMP32_REG(BPF_JLT, BPF_REG_0, BPF_REG_1, -2),
-
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"bound check with JMP32_JSLT for crossing 32-bit signed boundary",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct xdp_md, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, offsetof(struct xdp_md, data_end)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 10),
-
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_2, 0),
-	BPF_MOV32_IMM(BPF_REG_0, 0x7fffff10),
-	BPF_ALU32_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-
-	BPF_MOV32_IMM(BPF_REG_2, 0x80000fff),
-	BPF_MOV32_IMM(BPF_REG_0, 0x80000000),
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP32_REG(BPF_JSGT, BPF_REG_0, BPF_REG_2, 3),
-	/* r1 signed range is [S32_MIN, S32_MAX] */
-	BPF_JMP32_REG(BPF_JSLT, BPF_REG_0, BPF_REG_1, -3),
-
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-- 
cgit v1.2.3-70-g09d2


From 965a3f913e723e2da9c0ec21b9825b583b0777e1 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:13 +0300
Subject: selftests/bpf: verifier/bpf_get_stack converted to inline assembly

Test verifier/bpf_get_stack automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-4-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_bpf_get_stack.c   | 124 +++++++++++++++++++++
 .../testing/selftests/bpf/verifier/bpf_get_stack.c |  87 ---------------
 3 files changed, 126 insertions(+), 87 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_bpf_get_stack.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/bpf_get_stack.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index d71cbd63d94d..0319b35241a0 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -10,6 +10,7 @@
 #include "verifier_bounds_deduction.skel.h"
 #include "verifier_bounds_deduction_non_const.skel.h"
 #include "verifier_bounds_mix_sign_unsign.skel.h"
+#include "verifier_bpf_get_stack.skel.h"
 #include "verifier_cfg.skel.h"
 #include "verifier_cgroup_inv_retcode.skel.h"
 #include "verifier_cgroup_skb.skel.h"
@@ -87,6 +88,7 @@ void test_verifier_bounds(void)               { RUN(verifier_bounds); }
 void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction); }
 void test_verifier_bounds_deduction_non_const(void)     { RUN(verifier_bounds_deduction_non_const); }
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
+void test_verifier_bpf_get_stack(void)        { RUN(verifier_bpf_get_stack); }
 void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
 void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
 void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_bpf_get_stack.c b/tools/testing/selftests/bpf/progs/verifier_bpf_get_stack.c
new file mode 100644
index 000000000000..325a2bab4a71
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_bpf_get_stack.c
@@ -0,0 +1,124 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/bpf_get_stack.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+} map_array_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("tracepoint")
+__description("bpf_get_stack return R0 within range")
+__success
+__naked void stack_return_r0_within_range(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	r9 = %[__imm_0];				\
+	r1 = r6;					\
+	r2 = r7;					\
+	r3 = %[__imm_0];				\
+	r4 = 256;					\
+	call %[bpf_get_stack];				\
+	r1 = 0;						\
+	r8 = r0;					\
+	r8 <<= 32;					\
+	r8 s>>= 32;					\
+	if r1 s> r8 goto l0_%=;				\
+	r9 -= r8;					\
+	r2 = r7;					\
+	r2 += r8;					\
+	r1 = r9;					\
+	r1 <<= 32;					\
+	r1 s>>= 32;					\
+	r3 = r2;					\
+	r3 += r1;					\
+	r1 = r7;					\
+	r5 = %[__imm_0];				\
+	r1 += r5;					\
+	if r3 >= r1 goto l0_%=;				\
+	r1 = r6;					\
+	r3 = r9;					\
+	r4 = 0;						\
+	call %[bpf_get_stack];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_stack),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__imm_0, sizeof(struct test_val) / 2)
+	: __clobber_all);
+}
+
+SEC("iter/task")
+__description("bpf_get_task_stack return R0 range is refined")
+__success
+__naked void return_r0_range_is_refined(void)
+{
+	asm volatile ("					\
+	r6 = *(u64*)(r1 + 0);				\
+	r6 = *(u64*)(r6 + 0);		/* ctx->meta->seq */\
+	r7 = *(u64*)(r1 + 8);		/* ctx->task */\
+	r1 = %[map_array_48b] ll;	/* fixup_map_array_48b */\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	if r7 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r1 = r7;					\
+	r2 = r0;					\
+	r9 = r0;			/* keep buf for seq_write */\
+	r3 = 48;					\
+	r4 = 0;						\
+	call %[bpf_get_task_stack];			\
+	if r0 s> 0 goto l2_%=;				\
+	r0 = 0;						\
+	exit;						\
+l2_%=:	r1 = r6;					\
+	r2 = r9;					\
+	r3 = r0;					\
+	call %[bpf_seq_write];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_task_stack),
+	  __imm(bpf_map_lookup_elem),
+	  __imm(bpf_seq_write),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/bpf_get_stack.c b/tools/testing/selftests/bpf/verifier/bpf_get_stack.c
deleted file mode 100644
index 3e024c891178..000000000000
--- a/tools/testing/selftests/bpf/verifier/bpf_get_stack.c
+++ /dev/null
@@ -1,87 +0,0 @@
-{
-	"bpf_get_stack return R0 within range",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 28),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_9, sizeof(struct test_val)/2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_3, sizeof(struct test_val)/2),
-	BPF_MOV64_IMM(BPF_REG_4, 256),
-	BPF_EMIT_CALL(BPF_FUNC_get_stack),
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_8, 32),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_8, 32),
-	BPF_JMP_REG(BPF_JSGT, BPF_REG_1, BPF_REG_8, 16),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_9, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_9),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_1, 32),
-	BPF_ALU64_IMM(BPF_ARSH, BPF_REG_1, 32),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_5, sizeof(struct test_val)/2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_5),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_1, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_9),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_EMIT_CALL(BPF_FUNC_get_stack),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bpf_get_task_stack return R0 range is refined",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_6, 0), // ctx->meta->seq
-	BPF_LDX_MEM(BPF_DW, BPF_REG_7, BPF_REG_1, 8), // ctx->task
-	BPF_LD_MAP_FD(BPF_REG_1, 0), // fixup_map_array_48b
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_7, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_0), // keep buf for seq_write
-	BPF_MOV64_IMM(BPF_REG_3, 48),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_EMIT_CALL(BPF_FUNC_get_task_stack),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_9),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_seq_write),
-
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACING,
-	.expected_attach_type = BPF_TRACE_ITER,
-	.kfunc = "task",
-	.runs = -1, // Don't run, just load
-	.fixup_map_array_48b = { 3 },
-},
-- 
cgit v1.2.3-70-g09d2


From 37467c79e16a7b569fe5c1197e2ed507f72d2ab7 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:14 +0300
Subject: selftests/bpf: verifier/btf_ctx_access converted to inline assembly

Test verifier/btf_ctx_access automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-5-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 ++
 .../selftests/bpf/progs/verifier_btf_ctx_access.c  | 32 ++++++++++++++++++++++
 .../selftests/bpf/verifier/btf_ctx_access.c        | 25 -----------------
 3 files changed, 34 insertions(+), 25 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_btf_ctx_access.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/btf_ctx_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 0319b35241a0..3fb22d2b309a 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -11,6 +11,7 @@
 #include "verifier_bounds_deduction_non_const.skel.h"
 #include "verifier_bounds_mix_sign_unsign.skel.h"
 #include "verifier_bpf_get_stack.skel.h"
+#include "verifier_btf_ctx_access.skel.h"
 #include "verifier_cfg.skel.h"
 #include "verifier_cgroup_inv_retcode.skel.h"
 #include "verifier_cgroup_skb.skel.h"
@@ -89,6 +90,7 @@ void test_verifier_bounds_deduction(void)     { RUN(verifier_bounds_deduction);
 void test_verifier_bounds_deduction_non_const(void)     { RUN(verifier_bounds_deduction_non_const); }
 void test_verifier_bounds_mix_sign_unsign(void) { RUN(verifier_bounds_mix_sign_unsign); }
 void test_verifier_bpf_get_stack(void)        { RUN(verifier_bpf_get_stack); }
+void test_verifier_btf_ctx_access(void)       { RUN(verifier_btf_ctx_access); }
 void test_verifier_cfg(void)                  { RUN(verifier_cfg); }
 void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode); }
 void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_btf_ctx_access.c b/tools/testing/selftests/bpf/progs/verifier_btf_ctx_access.c
new file mode 100644
index 000000000000..a570e48b917a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_btf_ctx_access.c
@@ -0,0 +1,32 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/btf_ctx_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("fentry/bpf_modify_return_test")
+__description("btf_ctx_access accept")
+__success __retval(0)
+__naked void btf_ctx_access_accept(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + 8);		/* load 2nd argument value (int pointer) */\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("fentry/bpf_fentry_test9")
+__description("btf_ctx_access u32 pointer accept")
+__success __retval(0)
+__naked void ctx_access_u32_pointer_accept(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + 0);		/* load 1nd argument value (u32 pointer) */\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/btf_ctx_access.c b/tools/testing/selftests/bpf/verifier/btf_ctx_access.c
deleted file mode 100644
index 0484d3de040d..000000000000
--- a/tools/testing/selftests/bpf/verifier/btf_ctx_access.c
+++ /dev/null
@@ -1,25 +0,0 @@
-{
-	"btf_ctx_access accept",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 8),	/* load 2nd argument value (int pointer) */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACING,
-	.expected_attach_type = BPF_TRACE_FENTRY,
-	.kfunc = "bpf_modify_return_test",
-},
-
-{
-	"btf_ctx_access u32 pointer accept",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),	/* load 1nd argument value (u32 pointer) */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACING,
-	.expected_attach_type = BPF_TRACE_FENTRY,
-	.kfunc = "bpf_fentry_test9",
-},
-- 
cgit v1.2.3-70-g09d2


From fcd36964f22bbe571a07b3cdb3040ec31642ba44 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:15 +0300
Subject: selftests/bpf: verifier/ctx converted to inline assembly

Test verifier/ctx automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-6-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   2 +
 tools/testing/selftests/bpf/progs/verifier_ctx.c  | 221 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/ctx.c        | 186 ------------------
 3 files changed, 223 insertions(+), 186 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_ctx.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/ctx.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 3fb22d2b309a..1a1fd3bca4b8 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -17,6 +17,7 @@
 #include "verifier_cgroup_skb.skel.h"
 #include "verifier_cgroup_storage.skel.h"
 #include "verifier_const_or.skel.h"
+#include "verifier_ctx.skel.h"
 #include "verifier_ctx_sk_msg.skel.h"
 #include "verifier_direct_stack_access_wraparound.skel.h"
 #include "verifier_div0.skel.h"
@@ -96,6 +97,7 @@ void test_verifier_cgroup_inv_retcode(void)   { RUN(verifier_cgroup_inv_retcode)
 void test_verifier_cgroup_skb(void)           { RUN(verifier_cgroup_skb); }
 void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
 void test_verifier_const_or(void)             { RUN(verifier_const_or); }
+void test_verifier_ctx(void)                  { RUN(verifier_ctx); }
 void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
 void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
 void test_verifier_div0(void)                 { RUN(verifier_div0); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_ctx.c b/tools/testing/selftests/bpf/progs/verifier_ctx.c
new file mode 100644
index 000000000000..a83809a1dbbf
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_ctx.c
@@ -0,0 +1,221 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/ctx.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("tc")
+__description("context stores via BPF_ATOMIC")
+__failure __msg("BPF_ATOMIC stores into R1 ctx is not allowed")
+__naked void context_stores_via_bpf_atomic(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	lock *(u32 *)(r1 + %[__sk_buff_mark]) += w0;	\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("arithmetic ops make PTR_TO_CTX unusable")
+__failure __msg("dereference of modified ctx ptr")
+__naked void make_ptr_to_ctx_unusable(void)
+{
+	asm volatile ("					\
+	r1 += %[__imm_0];				\
+	r0 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	exit;						\
+"	:
+	: __imm_const(__imm_0,
+		      offsetof(struct __sk_buff, data) - offsetof(struct __sk_buff, mark)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("pass unmodified ctx pointer to helper")
+__success __retval(0)
+__naked void unmodified_ctx_pointer_to_helper(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	call %[bpf_csum_update];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_update)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("pass modified ctx pointer to helper, 1")
+__failure __msg("negative offset ctx ptr R1 off=-612 disallowed")
+__naked void ctx_pointer_to_helper_1(void)
+{
+	asm volatile ("					\
+	r1 += -612;					\
+	r2 = 0;						\
+	call %[bpf_csum_update];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_update)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("pass modified ctx pointer to helper, 2")
+__failure __msg("negative offset ctx ptr R1 off=-612 disallowed")
+__failure_unpriv __msg_unpriv("negative offset ctx ptr R1 off=-612 disallowed")
+__naked void ctx_pointer_to_helper_2(void)
+{
+	asm volatile ("					\
+	r1 += -612;					\
+	call %[bpf_get_socket_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_socket_cookie)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("pass modified ctx pointer to helper, 3")
+__failure __msg("variable ctx access var_off=(0x0; 0x4)")
+__naked void ctx_pointer_to_helper_3(void)
+{
+	asm volatile ("					\
+	r3 = *(u32*)(r1 + 0);				\
+	r3 &= 4;					\
+	r1 += r3;					\
+	r2 = 0;						\
+	call %[bpf_csum_update];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_csum_update)
+	: __clobber_all);
+}
+
+SEC("cgroup/sendmsg6")
+__description("pass ctx or null check, 1: ctx")
+__success
+__naked void or_null_check_1_ctx(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_netns_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_netns_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/sendmsg6")
+__description("pass ctx or null check, 2: null")
+__success
+__naked void or_null_check_2_null(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	call %[bpf_get_netns_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_netns_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/sendmsg6")
+__description("pass ctx or null check, 3: 1")
+__failure __msg("R1 type=scalar expected=ctx")
+__naked void or_null_check_3_1(void)
+{
+	asm volatile ("					\
+	r1 = 1;						\
+	call %[bpf_get_netns_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_netns_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/sendmsg6")
+__description("pass ctx or null check, 4: ctx - const")
+__failure __msg("negative offset ctx ptr R1 off=-612 disallowed")
+__naked void null_check_4_ctx_const(void)
+{
+	asm volatile ("					\
+	r1 += -612;					\
+	call %[bpf_get_netns_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_netns_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/connect4")
+__description("pass ctx or null check, 5: null (connect)")
+__success
+__naked void null_check_5_null_connect(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	call %[bpf_get_netns_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_netns_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/post_bind4")
+__description("pass ctx or null check, 6: null (bind)")
+__success
+__naked void null_check_6_null_bind(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	call %[bpf_get_netns_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_netns_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/post_bind4")
+__description("pass ctx or null check, 7: ctx (bind)")
+__success
+__naked void null_check_7_ctx_bind(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_socket_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_socket_cookie)
+	: __clobber_all);
+}
+
+SEC("cgroup/post_bind4")
+__description("pass ctx or null check, 8: null (bind)")
+__failure __msg("R1 type=scalar expected=ctx")
+__naked void null_check_8_null_bind(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	call %[bpf_get_socket_cookie];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_socket_cookie)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/ctx.c b/tools/testing/selftests/bpf/verifier/ctx.c
deleted file mode 100644
index 2fd31612c0b8..000000000000
--- a/tools/testing/selftests/bpf/verifier/ctx.c
+++ /dev/null
@@ -1,186 +0,0 @@
-{
-	"context stores via BPF_ATOMIC",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ATOMIC_OP(BPF_W, BPF_ADD, BPF_REG_1, BPF_REG_0, offsetof(struct __sk_buff, mark)),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "BPF_ATOMIC stores into R1 ctx is not allowed",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"arithmetic ops make PTR_TO_CTX unusable",
-	.insns = {
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_1,
-			      offsetof(struct __sk_buff, data) -
-			      offsetof(struct __sk_buff, mark)),
-		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-			    offsetof(struct __sk_buff, mark)),
-		BPF_EXIT_INSN(),
-	},
-	.errstr = "dereference of modified ctx ptr",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"pass unmodified ctx pointer to helper",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_csum_update),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"pass modified ctx pointer to helper, 1",
-	.insns = {
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -612),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_csum_update),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "negative offset ctx ptr R1 off=-612 disallowed",
-},
-{
-	"pass modified ctx pointer to helper, 2",
-	.insns = {
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -612),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_socket_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.result_unpriv = REJECT,
-	.result = REJECT,
-	.errstr_unpriv = "negative offset ctx ptr R1 off=-612 disallowed",
-	.errstr = "negative offset ctx ptr R1 off=-612 disallowed",
-},
-{
-	"pass modified ctx pointer to helper, 3",
-	.insns = {
-		BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, 0),
-		BPF_ALU64_IMM(BPF_AND, BPF_REG_3, 4),
-		BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_3),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_csum_update),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "variable ctx access var_off=(0x0; 0x4)",
-},
-{
-	"pass ctx or null check, 1: ctx",
-	.insns = {
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_netns_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK_ADDR,
-	.expected_attach_type = BPF_CGROUP_UDP6_SENDMSG,
-	.result = ACCEPT,
-},
-{
-	"pass ctx or null check, 2: null",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_1, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_netns_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK_ADDR,
-	.expected_attach_type = BPF_CGROUP_UDP6_SENDMSG,
-	.result = ACCEPT,
-},
-{
-	"pass ctx or null check, 3: 1",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_1, 1),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_netns_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK_ADDR,
-	.expected_attach_type = BPF_CGROUP_UDP6_SENDMSG,
-	.result = REJECT,
-	.errstr = "R1 type=scalar expected=ctx",
-},
-{
-	"pass ctx or null check, 4: ctx - const",
-	.insns = {
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -612),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_netns_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK_ADDR,
-	.expected_attach_type = BPF_CGROUP_UDP6_SENDMSG,
-	.result = REJECT,
-	.errstr = "negative offset ctx ptr R1 off=-612 disallowed",
-},
-{
-	"pass ctx or null check, 5: null (connect)",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_1, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_netns_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK_ADDR,
-	.expected_attach_type = BPF_CGROUP_INET4_CONNECT,
-	.result = ACCEPT,
-},
-{
-	"pass ctx or null check, 6: null (bind)",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_1, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_netns_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-	.expected_attach_type = BPF_CGROUP_INET4_POST_BIND,
-	.result = ACCEPT,
-},
-{
-	"pass ctx or null check, 7: ctx (bind)",
-	.insns = {
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_socket_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-	.expected_attach_type = BPF_CGROUP_INET4_POST_BIND,
-	.result = ACCEPT,
-},
-{
-	"pass ctx or null check, 8: null (bind)",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_1, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
-			     BPF_FUNC_get_socket_cookie),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
-	.expected_attach_type = BPF_CGROUP_INET4_POST_BIND,
-	.result = REJECT,
-	.errstr = "R1 type=scalar expected=ctx",
-},
-- 
cgit v1.2.3-70-g09d2


From 6080280243846f3c5b3715af4334260522bf11f0 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:16 +0300
Subject: selftests/bpf: verifier/d_path converted to inline assembly

Test verifier/d_path automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-7-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../testing/selftests/bpf/progs/verifier_d_path.c  | 48 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/d_path.c      | 37 -----------------
 3 files changed, 50 insertions(+), 37 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_d_path.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/d_path.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 1a1fd3bca4b8..ec4c20150e41 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -19,6 +19,7 @@
 #include "verifier_const_or.skel.h"
 #include "verifier_ctx.skel.h"
 #include "verifier_ctx_sk_msg.skel.h"
+#include "verifier_d_path.skel.h"
 #include "verifier_direct_stack_access_wraparound.skel.h"
 #include "verifier_div0.skel.h"
 #include "verifier_div_overflow.skel.h"
@@ -99,6 +100,7 @@ void test_verifier_cgroup_storage(void)       { RUN(verifier_cgroup_storage); }
 void test_verifier_const_or(void)             { RUN(verifier_const_or); }
 void test_verifier_ctx(void)                  { RUN(verifier_ctx); }
 void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
+void test_verifier_d_path(void)               { RUN(verifier_d_path); }
 void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
 void test_verifier_div0(void)                 { RUN(verifier_div0); }
 void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_d_path.c b/tools/testing/selftests/bpf/progs/verifier_d_path.c
new file mode 100644
index 000000000000..ec79cbcfde91
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_d_path.c
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/d_path.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("fentry/dentry_open")
+__description("d_path accept")
+__success __retval(0)
+__naked void d_path_accept(void)
+{
+	asm volatile ("					\
+	r1 = *(u32*)(r1 + 0);				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r6 = 0;						\
+	*(u64*)(r2 + 0) = r6;				\
+	r3 = 8 ll;					\
+	call %[bpf_d_path];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_d_path)
+	: __clobber_all);
+}
+
+SEC("fentry/d_path")
+__description("d_path reject")
+__failure __msg("helper call is not allowed in probe")
+__naked void d_path_reject(void)
+{
+	asm volatile ("					\
+	r1 = *(u32*)(r1 + 0);				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r6 = 0;						\
+	*(u64*)(r2 + 0) = r6;				\
+	r3 = 8 ll;					\
+	call %[bpf_d_path];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_d_path)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/d_path.c b/tools/testing/selftests/bpf/verifier/d_path.c
deleted file mode 100644
index b988396379a7..000000000000
--- a/tools/testing/selftests/bpf/verifier/d_path.c
+++ /dev/null
@@ -1,37 +0,0 @@
-{
-	"d_path accept",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_6, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_2, BPF_REG_6, 0),
-	BPF_LD_IMM64(BPF_REG_3, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_d_path),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACING,
-	.expected_attach_type = BPF_TRACE_FENTRY,
-	.kfunc = "dentry_open",
-},
-{
-	"d_path reject",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_MOV64_IMM(BPF_REG_6, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_2, BPF_REG_6, 0),
-	BPF_LD_IMM64(BPF_REG_3, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_d_path),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "helper call is not allowed in probe",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACING,
-	.expected_attach_type = BPF_TRACE_FENTRY,
-	.kfunc = "d_path",
-},
-- 
cgit v1.2.3-70-g09d2


From 0a372c9c0812ec5de4a740781fbf183ee523bf4a Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:17 +0300
Subject: selftests/bpf: verifier/direct_packet_access converted to inline
 assembly

Test verifier/direct_packet_access automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-8-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_direct_packet_access.c      | 803 +++++++++++++++++++++
 .../selftests/bpf/verifier/direct_packet_access.c  | 710 ------------------
 3 files changed, 805 insertions(+), 710 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_direct_packet_access.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/direct_packet_access.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index ec4c20150e41..ce2436b4b09f 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -20,6 +20,7 @@
 #include "verifier_ctx.skel.h"
 #include "verifier_ctx_sk_msg.skel.h"
 #include "verifier_d_path.skel.h"
+#include "verifier_direct_packet_access.skel.h"
 #include "verifier_direct_stack_access_wraparound.skel.h"
 #include "verifier_div0.skel.h"
 #include "verifier_div_overflow.skel.h"
@@ -101,6 +102,7 @@ void test_verifier_const_or(void)             { RUN(verifier_const_or); }
 void test_verifier_ctx(void)                  { RUN(verifier_ctx); }
 void test_verifier_ctx_sk_msg(void)           { RUN(verifier_ctx_sk_msg); }
 void test_verifier_d_path(void)               { RUN(verifier_d_path); }
+void test_verifier_direct_packet_access(void) { RUN(verifier_direct_packet_access); }
 void test_verifier_direct_stack_access_wraparound(void) { RUN(verifier_direct_stack_access_wraparound); }
 void test_verifier_div0(void)                 { RUN(verifier_div0); }
 void test_verifier_div_overflow(void)         { RUN(verifier_div_overflow); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_direct_packet_access.c b/tools/testing/selftests/bpf/progs/verifier_direct_packet_access.c
new file mode 100644
index 000000000000..99a23dea8233
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_direct_packet_access.c
@@ -0,0 +1,803 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/direct_packet_access.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("tc")
+__description("pkt_end - pkt_start is allowed")
+__success __retval(TEST_DATA_LEN)
+__naked void end_pkt_start_is_allowed(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r0 -= r2;					\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test1")
+__success __retval(0)
+__naked void direct_packet_access_test1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test2")
+__success __retval(0)
+__naked void direct_packet_access_test2(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	r4 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r3 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r5 = r3;					\
+	r5 += 14;					\
+	if r5 > r4 goto l0_%=;				\
+	r0 = *(u8*)(r3 + 7);				\
+	r4 = *(u8*)(r3 + 12);				\
+	r4 *= 14;					\
+	r3 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 += r4;					\
+	r2 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r2 <<= 49;					\
+	r2 >>= 49;					\
+	r3 += r2;					\
+	r2 = r3;					\
+	r2 += 8;					\
+	r1 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	if r2 > r1 goto l1_%=;				\
+	r1 = *(u8*)(r3 + 4);				\
+l1_%=:	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("direct packet access: test3")
+__failure __msg("invalid bpf_context access off=76")
+__failure_unpriv
+__naked void direct_packet_access_test3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test4 (write)")
+__success __retval(0)
+__naked void direct_packet_access_test4_write(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test5 (pkt_end >= reg, good access)")
+__success __retval(0)
+__naked void pkt_end_reg_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 >= r0 goto l0_%=;				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = *(u8*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test6 (pkt_end >= reg, bad access)")
+__failure __msg("invalid access to packet")
+__naked void pkt_end_reg_bad_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 >= r0 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test7 (pkt_end >= reg, both accesses)")
+__failure __msg("invalid access to packet")
+__naked void pkt_end_reg_both_accesses(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 >= r0 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = *(u8*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test8 (double test, variant 1)")
+__success __retval(0)
+__naked void test8_double_test_variant_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 >= r0 goto l0_%=;				\
+	if r0 > r3 goto l1_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l1_%=:	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = *(u8*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test9 (double test, variant 2)")
+__success __retval(0)
+__naked void test9_double_test_variant_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 >= r0 goto l0_%=;				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	if r0 > r3 goto l1_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l1_%=:	r0 = *(u8*)(r2 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test10 (write invalid)")
+__failure __msg("invalid access to packet")
+__naked void packet_access_test10_write_invalid(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	*(u8*)(r2 + 0) = r2;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test11 (shift, good access)")
+__success __retval(1)
+__naked void access_test11_shift_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 22;					\
+	if r0 > r3 goto l0_%=;				\
+	r3 = 144;					\
+	r5 = r3;					\
+	r5 += 23;					\
+	r5 >>= 3;					\
+	r6 = r2;					\
+	r6 += r5;					\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test12 (and, good access)")
+__success __retval(1)
+__naked void access_test12_and_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 22;					\
+	if r0 > r3 goto l0_%=;				\
+	r3 = 144;					\
+	r5 = r3;					\
+	r5 += 23;					\
+	r5 &= 15;					\
+	r6 = r2;					\
+	r6 += r5;					\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test13 (branches, good access)")
+__success __retval(1)
+__naked void access_test13_branches_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 22;					\
+	if r0 > r3 goto l0_%=;				\
+	r3 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	r4 = 1;						\
+	if r3 > r4 goto l1_%=;				\
+	r3 = 14;					\
+	goto l2_%=;					\
+l1_%=:	r3 = 24;					\
+l2_%=:	r5 = r3;					\
+	r5 += 23;					\
+	r5 &= 15;					\
+	r6 = r2;					\
+	r6 += r5;					\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test14 (pkt_ptr += 0, CONST_IMM, good access)")
+__success __retval(1)
+__naked void _0_const_imm_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 22;					\
+	if r0 > r3 goto l0_%=;				\
+	r5 = 12;					\
+	r5 >>= 4;					\
+	r6 = r2;					\
+	r6 += r5;					\
+	r0 = *(u8*)(r6 + 0);				\
+	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test15 (spill with xadd)")
+__failure __msg("R2 invalid mem access 'scalar'")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void access_test15_spill_with_xadd(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r5 = 4096;					\
+	r4 = r10;					\
+	r4 += -8;					\
+	*(u64*)(r4 + 0) = r2;				\
+	lock *(u64 *)(r4 + 0) += r5;			\
+	r2 = *(u64*)(r4 + 0);				\
+	*(u32*)(r2 + 0) = r5;				\
+	r0 = 0;						\
+l0_%=:	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test16 (arith on data_end)")
+__failure __msg("R3 pointer arithmetic on pkt_end")
+__naked void test16_arith_on_data_end(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	r3 += 16;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test17 (pruning, alignment)")
+__failure __msg("misaligned packet access off 2+(0x0; 0x0)+15+-4 size 4")
+__flag(BPF_F_STRICT_ALIGNMENT)
+__naked void packet_access_test17_pruning_alignment(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r7 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	r0 = r2;					\
+	r0 += 14;					\
+	if r7 > 1 goto l0_%=;				\
+l2_%=:	if r0 > r3 goto l1_%=;				\
+	*(u32*)(r0 - 4) = r0;				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+l0_%=:	r0 += 1;					\
+	goto l2_%=;					\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test18 (imm += pkt_ptr, 1)")
+__success __retval(0)
+__naked void test18_imm_pkt_ptr_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = 8;						\
+	r0 += r2;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test19 (imm += pkt_ptr, 2)")
+__success __retval(0)
+__naked void test19_imm_pkt_ptr_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r4 = 4;						\
+	r4 += r2;					\
+	*(u8*)(r4 + 0) = r4;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test20 (x += pkt_ptr, 1)")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void test20_x_pkt_ptr_1(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = 0xffffffff;				\
+	*(u64*)(r10 - 8) = r0;				\
+	r0 = *(u64*)(r10 - 8);				\
+	r0 &= 0x7fff;					\
+	r4 = r0;					\
+	r4 += r2;					\
+	r5 = r4;					\
+	r4 += %[__imm_0];				\
+	if r4 > r3 goto l0_%=;				\
+	*(u64*)(r5 + 0) = r4;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0x7fff - 1),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test21 (x += pkt_ptr, 2)")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void test21_x_pkt_ptr_2(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r4 = 0xffffffff;				\
+	*(u64*)(r10 - 8) = r4;				\
+	r4 = *(u64*)(r10 - 8);				\
+	r4 &= 0x7fff;					\
+	r4 += r2;					\
+	r5 = r4;					\
+	r4 += %[__imm_0];				\
+	if r4 > r3 goto l0_%=;				\
+	*(u64*)(r5 + 0) = r4;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0x7fff - 1),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test22 (x += pkt_ptr, 3)")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void test22_x_pkt_ptr_3(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	*(u64*)(r10 - 8) = r2;				\
+	*(u64*)(r10 - 16) = r3;				\
+	r3 = *(u64*)(r10 - 16);				\
+	if r0 > r3 goto l0_%=;				\
+	r2 = *(u64*)(r10 - 8);				\
+	r4 = 0xffffffff;				\
+	lock *(u64 *)(r10 - 8) += r4;			\
+	r4 = *(u64*)(r10 - 8);				\
+	r4 >>= 49;					\
+	r4 += r2;					\
+	r0 = r4;					\
+	r0 += 2;					\
+	if r0 > r3 goto l0_%=;				\
+	r2 = 1;						\
+	*(u16*)(r4 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test23 (x += pkt_ptr, 4)")
+__failure __msg("invalid access to packet, off=0 size=8, R5(id=2,off=0,r=0)")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void test23_x_pkt_ptr_4(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	*(u64*)(r10 - 8) = r0;				\
+	r0 = *(u64*)(r10 - 8);				\
+	r0 &= 0xffff;					\
+	r4 = r0;					\
+	r0 = 31;					\
+	r0 += r4;					\
+	r0 += r2;					\
+	r5 = r0;					\
+	r0 += %[__imm_0];				\
+	if r0 > r3 goto l0_%=;				\
+	*(u64*)(r5 + 0) = r0;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0xffff - 1),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test24 (x += pkt_ptr, 5)")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void test24_x_pkt_ptr_5(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = 0xffffffff;				\
+	*(u64*)(r10 - 8) = r0;				\
+	r0 = *(u64*)(r10 - 8);				\
+	r0 &= 0xff;					\
+	r4 = r0;					\
+	r0 = 64;					\
+	r0 += r4;					\
+	r0 += r2;					\
+	r5 = r0;					\
+	r0 += %[__imm_0];				\
+	if r0 > r3 goto l0_%=;				\
+	*(u64*)(r5 + 0) = r0;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__imm_0, 0x7fff - 1),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test25 (marking on <, good access)")
+__success __retval(0)
+__naked void test25_marking_on_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 < r3 goto l0_%=;				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+l0_%=:	r0 = *(u8*)(r2 + 0);				\
+	goto l1_%=;					\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test26 (marking on <, bad access)")
+__failure __msg("invalid access to packet")
+__naked void test26_marking_on_bad_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 < r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l1_%=:	r0 = 0;						\
+	exit;						\
+l0_%=:	goto l1_%=;					\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test27 (marking on <=, good access)")
+__success __retval(1)
+__naked void test27_marking_on_good_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 <= r0 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test28 (marking on <=, bad access)")
+__failure __msg("invalid access to packet")
+__naked void test28_marking_on_bad_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r3 <= r0 goto l0_%=;				\
+l1_%=:	r0 = 1;						\
+	exit;						\
+l0_%=:	r0 = *(u8*)(r2 + 0);				\
+	goto l1_%=;					\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test29 (reg > pkt_end in subprog)")
+__success __retval(0)
+__naked void reg_pkt_end_in_subprog(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r2 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r3 = r6;					\
+	r3 += 8;					\
+	call reg_pkt_end_in_subprog__1;			\
+	if r0 == 0 goto l0_%=;				\
+	r0 = *(u8*)(r6 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void reg_pkt_end_in_subprog__1(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	if r3 > r2 goto l0_%=;				\
+	r0 = 1;						\
+l0_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("direct packet access: test30 (check_id() in regsafe(), bad access)")
+__failure __msg("invalid access to packet, off=0 size=1, R2")
+__flag(BPF_F_TEST_STATE_FREQ)
+__naked void id_in_regsafe_bad_access(void)
+{
+	asm volatile ("					\
+	/* r9 = ctx */					\
+	r9 = r1;					\
+	/* r7 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r7 = r0;					\
+	/* r6 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r6 = r0;					\
+	/* r2 = ctx->data				\
+	 * r3 = ctx->data				\
+	 * r4 = ctx->data_end				\
+	 */						\
+	r2 = *(u32*)(r9 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r9 + %[__sk_buff_data]);		\
+	r4 = *(u32*)(r9 + %[__sk_buff_data_end]);	\
+	/* if r6 > 100 goto exit			\
+	 * if r7 > 100 goto exit			\
+	 */						\
+	if r6 > 100 goto l0_%=;				\
+	if r7 > 100 goto l0_%=;				\
+	/* r2 += r6              ; this forces assignment of ID to r2\
+	 * r2 += 1               ; get some fixed off for r2\
+	 * r3 += r7              ; this forces assignment of ID to r3\
+	 * r3 += 1               ; get some fixed off for r3\
+	 */						\
+	r2 += r6;					\
+	r2 += 1;					\
+	r3 += r7;					\
+	r3 += 1;					\
+	/* if r6 > r7 goto +1    ; no new information about the state is derived from\
+	 *                       ; this check, thus produced verifier states differ\
+	 *                       ; only in 'insn_idx'	\
+	 * r2 = r3               ; optionally share ID between r2 and r3\
+	 */						\
+	if r6 != r7 goto l1_%=;				\
+	r2 = r3;					\
+l1_%=:	/* if r3 > ctx->data_end goto exit */		\
+	if r3 > r4 goto l0_%=;				\
+	/* r5 = *(u8 *) (r2 - 1) ; access packet memory using r2,\
+	 *                       ; this is not always safe\
+	 */						\
+	r5 = *(u8*)(r2 - 1);				\
+l0_%=:	/* exit(0) */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/direct_packet_access.c b/tools/testing/selftests/bpf/verifier/direct_packet_access.c
deleted file mode 100644
index dce2e28aeb43..000000000000
--- a/tools/testing/selftests/bpf/verifier/direct_packet_access.c
+++ /dev/null
@@ -1,710 +0,0 @@
-{
-	"pkt_end - pkt_start is allowed",
-	.insns = {
-		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-			    offsetof(struct __sk_buff, data_end)),
-		BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-			    offsetof(struct __sk_buff, data)),
-		BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_2),
-		BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = TEST_DATA_LEN,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test1",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 14),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_5, BPF_REG_4, 15),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_3, 7),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_3, 12),
-	BPF_ALU64_IMM(BPF_MUL, BPF_REG_4, 14),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_2, 49),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_2, 49),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 8),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_1, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_3, 4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test3",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid bpf_context access off=76",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SOCKET_FILTER,
-},
-{
-	"direct packet access: test4 (write)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test5 (pkt_end >= reg, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test6 (pkt_end >= reg, bad access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_0, 3),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid access to packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test7 (pkt_end >= reg, both accesses)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_0, 3),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid access to packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test8 (double test, variant 1)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_0, 4),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test9 (double test, variant 2)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGE, BPF_REG_3, BPF_REG_0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test10 (write invalid)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid access to packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test11 (shift, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 22),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 144),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 23),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_5, 3),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_5),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 1,
-},
-{
-	"direct packet access: test12 (and, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 22),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 8),
-	BPF_MOV64_IMM(BPF_REG_3, 144),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 23),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_5, 15),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_5),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 1,
-},
-{
-	"direct packet access: test13 (branches, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 22),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 13),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_MOV64_IMM(BPF_REG_4, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_4, 2),
-	BPF_MOV64_IMM(BPF_REG_3, 14),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_3, 24),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_3),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 23),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_5, 15),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_5),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 1,
-},
-{
-	"direct packet access: test14 (pkt_ptr += 0, CONST_IMM, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 22),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 7),
-	BPF_MOV64_IMM(BPF_REG_5, 12),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_5, 4),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_5),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_6, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 1,
-},
-{
-	"direct packet access: test15 (spill with xadd)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 8),
-	BPF_MOV64_IMM(BPF_REG_5, 4096),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_4, BPF_REG_2, 0),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_4, BPF_REG_5, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_4, 0),
-	BPF_STX_MEM(BPF_W, BPF_REG_2, BPF_REG_5, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R2 invalid mem access 'scalar'",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet access: test16 (arith on data_end)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 16),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R3 pointer arithmetic on pkt_end",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test17 (pruning, alignment)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 14),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_7, 1, 4),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, -4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_A(-6),
-	},
-	.errstr = "misaligned packet access off 2+(0x0; 0x0)+15+-4 size 4",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.flags = F_LOAD_WITH_STRICT_ALIGNMENT,
-},
-{
-	"direct packet access: test18 (imm += pkt_ptr, 1)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_IMM(BPF_REG_0, 8),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test19 (imm += pkt_ptr, 2)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 3),
-	BPF_MOV64_IMM(BPF_REG_4, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_2),
-	BPF_STX_MEM(BPF_B, BPF_REG_4, BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test20 (x += pkt_ptr, 1)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_IMM(BPF_REG_0, 0xffffffff),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 0x7fff),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 0x7fff - 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_5, BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet access: test21 (x += pkt_ptr, 2)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 9),
-	BPF_MOV64_IMM(BPF_REG_4, 0xffffffff),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_4, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_10, -8),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_4, 0x7fff),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 0x7fff - 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_5, BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet access: test22 (x += pkt_ptr, 3)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_3, -16),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_10, -16),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 11),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_10, -8),
-	BPF_MOV64_IMM(BPF_REG_4, 0xffffffff),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_10, BPF_REG_4, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_10, -8),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_4, 49),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_4),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 2),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_2, 1),
-	BPF_STX_MEM(BPF_H, BPF_REG_4, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet access: test23 (x += pkt_ptr, 4)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 0xffff),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, 31),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 0xffff - 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_5, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid access to packet, off=0 size=8, R5(id=2,off=0,r=0)",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet access: test24 (x += pkt_ptr, 5)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_IMM(BPF_REG_0, 0xffffffff),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_10, -8),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 0xff),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, 64),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 0x7fff - 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_5, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"direct packet access: test25 (marking on <, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_0, BPF_REG_3, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, -4),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test26 (marking on <, bad access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JLT, BPF_REG_0, BPF_REG_3, 3),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JA, 0, 0, -3),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test27 (marking on <=, good access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_0, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = 1,
-},
-{
-	"direct packet access: test28 (marking on <=, bad access)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JLE, BPF_REG_3, BPF_REG_0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, -4),
-	},
-	.result = REJECT,
-	.errstr = "invalid access to packet",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test29 (reg > pkt_end in subprog)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 4),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_6, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_2, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"direct packet access: test30 (check_id() in regsafe(), bad access)",
-	.insns = {
-	/* r9 = ctx */
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_1),
-	/* r7 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r6 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* r2 = ctx->data
-	 * r3 = ctx->data
-	 * r4 = ctx->data_end
-	 */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_9, offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_9, offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_9, offsetof(struct __sk_buff, data_end)),
-	/* if r6 > 100 goto exit
-	 * if r7 > 100 goto exit
-	 */
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_6, 100, 9),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_7, 100, 8),
-	/* r2 += r6              ; this forces assignment of ID to r2
-	 * r2 += 1               ; get some fixed off for r2
-	 * r3 += r7              ; this forces assignment of ID to r3
-	 * r3 += 1               ; get some fixed off for r3
-	 */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 1),
-	/* if r6 > r7 goto +1    ; no new information about the state is derived from
-	 *                       ; this check, thus produced verifier states differ
-	 *                       ; only in 'insn_idx'
-	 * r2 = r3               ; optionally share ID between r2 and r3
-	 */
-	BPF_JMP_REG(BPF_JNE, BPF_REG_6, BPF_REG_7, 1),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_3),
-	/* if r3 > ctx->data_end goto exit */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_3, BPF_REG_4, 1),
-	/* r5 = *(u8 *) (r2 - 1) ; access packet memory using r2,
-	 *                       ; this is not always safe
-	 */
-	BPF_LDX_MEM(BPF_B, BPF_REG_5, BPF_REG_2, -1),
-	/* exit(0) */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.flags = BPF_F_TEST_STATE_FREQ,
-	.result = REJECT,
-	.errstr = "invalid access to packet, off=0 size=1, R2",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-- 
cgit v1.2.3-70-g09d2


From a5828e3154d17d9fec53910a1aa161d028339ada Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:18 +0300
Subject: selftests/bpf: verifier/jeq_infer_not_null converted to inline
 assembly

Test verifier/jeq_infer_not_null automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-9-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_jeq_infer_not_null.c        | 213 +++++++++++++++++++++
 .../selftests/bpf/verifier/jeq_infer_not_null.c    | 174 -----------------
 3 files changed, 215 insertions(+), 174 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_jeq_infer_not_null.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/jeq_infer_not_null.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index ce2436b4b09f..6585432e088f 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -29,6 +29,7 @@
 #include "verifier_helper_restricted.skel.h"
 #include "verifier_helper_value_access.skel.h"
 #include "verifier_int_ptr.skel.h"
+#include "verifier_jeq_infer_not_null.skel.h"
 #include "verifier_ld_ind.skel.h"
 #include "verifier_leak_ptr.skel.h"
 #include "verifier_map_ptr.skel.h"
@@ -111,6 +112,7 @@ void test_verifier_helper_packet_access(void) { RUN(verifier_helper_packet_acces
 void test_verifier_helper_restricted(void)    { RUN(verifier_helper_restricted); }
 void test_verifier_helper_value_access(void)  { RUN(verifier_helper_value_access); }
 void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
+void test_verifier_jeq_infer_not_null(void)   { RUN(verifier_jeq_infer_not_null); }
 void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_jeq_infer_not_null.c b/tools/testing/selftests/bpf/progs/verifier_jeq_infer_not_null.c
new file mode 100644
index 000000000000..bf16b00502f2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_jeq_infer_not_null.c
@@ -0,0 +1,213 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/jeq_infer_not_null.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_XSKMAP);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+} map_xskmap SEC(".maps");
+
+/* This is equivalent to the following program:
+ *
+ *   r6 = skb->sk;
+ *   r7 = sk_fullsock(r6);
+ *   r0 = sk_fullsock(r6);
+ *   if (r0 == 0) return 0;    (a)
+ *   if (r0 != r7) return 0;   (b)
+ *   *r7->type;                (c)
+ *   return 0;
+ *
+ * It is safe to dereference r7 at point (c), because of (a) and (b).
+ * The test verifies that relation r0 == r7 is propagated from (b) to (c).
+ */
+SEC("cgroup/skb")
+__description("jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL -> PTR_TO_SOCKET for JNE false branch")
+__success __failure_unpriv __msg_unpriv("R7 pointer comparison")
+__retval(0)
+__naked void socket_for_jne_false_branch(void)
+{
+	asm volatile ("					\
+	/* r6 = skb->sk; */				\
+	r6 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	/* if (r6 == 0) return 0; */			\
+	if r6 == 0 goto l0_%=;				\
+	/* r7 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	r7 = r0;					\
+	/* r0 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	/* if (r0 == null) return 0; */			\
+	if r0 == 0 goto l0_%=;				\
+	/* if (r0 == r7) r0 = *(r7->type); */		\
+	if r0 != r7 goto l0_%=;		/* Use ! JNE ! */\
+	r0 = *(u32*)(r7 + %[bpf_sock_type]);		\
+l0_%=:	/* return 0 */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+/* Same as above, but verify that another branch of JNE still
+ * prohibits access to PTR_MAYBE_NULL.
+ */
+SEC("cgroup/skb")
+__description("jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL unchanged for JNE true branch")
+__failure __msg("R7 invalid mem access 'sock_or_null'")
+__failure_unpriv __msg_unpriv("R7 pointer comparison")
+__naked void unchanged_for_jne_true_branch(void)
+{
+	asm volatile ("					\
+	/* r6 = skb->sk */				\
+	r6 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	/* if (r6 == 0) return 0; */			\
+	if r6 == 0 goto l0_%=;				\
+	/* r7 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	r7 = r0;					\
+	/* r0 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	/* if (r0 == null) return 0; */			\
+	if r0 != 0 goto l0_%=;				\
+	/* if (r0 == r7) return 0; */			\
+	if r0 != r7 goto l1_%=;		/* Use ! JNE ! */\
+	goto l0_%=;					\
+l1_%=:	/* r0 = *(r7->type); */				\
+	r0 = *(u32*)(r7 + %[bpf_sock_type]);		\
+l0_%=:	/* return 0 */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+/* Same as a first test, but not null should be inferred for JEQ branch */
+SEC("cgroup/skb")
+__description("jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL -> PTR_TO_SOCKET for JEQ true branch")
+__success __failure_unpriv __msg_unpriv("R7 pointer comparison")
+__retval(0)
+__naked void socket_for_jeq_true_branch(void)
+{
+	asm volatile ("					\
+	/* r6 = skb->sk; */				\
+	r6 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	/* if (r6 == null) return 0; */			\
+	if r6 == 0 goto l0_%=;				\
+	/* r7 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	r7 = r0;					\
+	/* r0 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	/* if (r0 == null) return 0; */			\
+	if r0 == 0 goto l0_%=;				\
+	/* if (r0 != r7) return 0; */			\
+	if r0 == r7 goto l1_%=;		/* Use ! JEQ ! */\
+	goto l0_%=;					\
+l1_%=:	/* r0 = *(r7->type); */				\
+	r0 = *(u32*)(r7 + %[bpf_sock_type]);		\
+l0_%=:	/* return 0; */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+/* Same as above, but verify that another branch of JNE still
+ * prohibits access to PTR_MAYBE_NULL.
+ */
+SEC("cgroup/skb")
+__description("jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL unchanged for JEQ false branch")
+__failure __msg("R7 invalid mem access 'sock_or_null'")
+__failure_unpriv __msg_unpriv("R7 pointer comparison")
+__naked void unchanged_for_jeq_false_branch(void)
+{
+	asm volatile ("					\
+	/* r6 = skb->sk; */				\
+	r6 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	/* if (r6 == null) return 0; */			\
+	if r6 == 0 goto l0_%=;				\
+	/* r7 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	r7 = r0;					\
+	/* r0 = sk_fullsock(skb); */			\
+	r1 = r6;					\
+	call %[bpf_sk_fullsock];			\
+	/* if (r0 == null) return 0; */			\
+	if r0 == 0 goto l0_%=;				\
+	/* if (r0 != r7) r0 = *(r7->type); */		\
+	if r0 == r7 goto l0_%=;		/* Use ! JEQ ! */\
+	r0 = *(u32*)(r7 + %[bpf_sock_type]);		\
+l0_%=:	/* return 0; */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+/* Maps are treated in a different branch of `mark_ptr_not_null_reg`,
+ * so separate test for maps case.
+ */
+SEC("xdp")
+__description("jne/jeq infer not null, PTR_TO_MAP_VALUE_OR_NULL -> PTR_TO_MAP_VALUE")
+__success __retval(0)
+__naked void null_ptr_to_map_value(void)
+{
+	asm volatile ("					\
+	/* r9 = &some stack to use as key */		\
+	r1 = 0;						\
+	*(u32*)(r10 - 8) = r1;				\
+	r9 = r10;					\
+	r9 += -8;					\
+	/* r8 = process local map */			\
+	r8 = %[map_xskmap] ll;				\
+	/* r6 = map_lookup_elem(r8, r9); */		\
+	r1 = r8;					\
+	r2 = r9;					\
+	call %[bpf_map_lookup_elem];			\
+	r6 = r0;					\
+	/* r7 = map_lookup_elem(r8, r9); */		\
+	r1 = r8;					\
+	r2 = r9;					\
+	call %[bpf_map_lookup_elem];			\
+	r7 = r0;					\
+	/* if (r6 == 0) return 0; */			\
+	if r6 == 0 goto l0_%=;				\
+	/* if (r6 != r7) return 0; */			\
+	if r6 != r7 goto l0_%=;				\
+	/* read *r7; */					\
+	r0 = *(u32*)(r7 + %[bpf_xdp_sock_queue_id]);	\
+l0_%=:	/* return 0; */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_xskmap),
+	  __imm_const(bpf_xdp_sock_queue_id, offsetof(struct bpf_xdp_sock, queue_id))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/jeq_infer_not_null.c b/tools/testing/selftests/bpf/verifier/jeq_infer_not_null.c
deleted file mode 100644
index 67a1c07ead34..000000000000
--- a/tools/testing/selftests/bpf/verifier/jeq_infer_not_null.c
+++ /dev/null
@@ -1,174 +0,0 @@
-{
-	/* This is equivalent to the following program:
-	 *
-	 *   r6 = skb->sk;
-	 *   r7 = sk_fullsock(r6);
-	 *   r0 = sk_fullsock(r6);
-	 *   if (r0 == 0) return 0;    (a)
-	 *   if (r0 != r7) return 0;   (b)
-	 *   *r7->type;                (c)
-	 *   return 0;
-	 *
-	 * It is safe to dereference r7 at point (c), because of (a) and (b).
-	 * The test verifies that relation r0 == r7 is propagated from (b) to (c).
-	 */
-	"jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL -> PTR_TO_SOCKET for JNE false branch",
-	.insns = {
-	/* r6 = skb->sk; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	/* if (r6 == 0) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 8),
-	/* r7 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r0 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	/* if (r0 == null) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	/* if (r0 == r7) r0 = *(r7->type); */
-	BPF_JMP_REG(BPF_JNE, BPF_REG_0, BPF_REG_7, 1), /* Use ! JNE ! */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_sock, type)),
-	/* return 0 */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R7 pointer comparison",
-},
-{
-	/* Same as above, but verify that another branch of JNE still
-	 * prohibits access to PTR_MAYBE_NULL.
-	 */
-	"jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL unchanged for JNE true branch",
-	.insns = {
-	/* r6 = skb->sk */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	/* if (r6 == 0) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 9),
-	/* r7 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r0 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	/* if (r0 == null) return 0; */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	/* if (r0 == r7) return 0; */
-	BPF_JMP_REG(BPF_JNE, BPF_REG_0, BPF_REG_7, 1), /* Use ! JNE ! */
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	/* r0 = *(r7->type); */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_sock, type)),
-	/* return 0 */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "R7 invalid mem access 'sock_or_null'",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R7 pointer comparison",
-},
-{
-	/* Same as a first test, but not null should be inferred for JEQ branch */
-	"jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL -> PTR_TO_SOCKET for JEQ true branch",
-	.insns = {
-	/* r6 = skb->sk; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	/* if (r6 == null) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 9),
-	/* r7 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r0 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	/* if (r0 == null) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	/* if (r0 != r7) return 0; */
-	BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_7, 1), /* Use ! JEQ ! */
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	/* r0 = *(r7->type); */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_sock, type)),
-	/* return 0; */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R7 pointer comparison",
-},
-{
-	/* Same as above, but verify that another branch of JNE still
-	 * prohibits access to PTR_MAYBE_NULL.
-	 */
-	"jne/jeq infer not null, PTR_TO_SOCKET_OR_NULL unchanged for JEQ false branch",
-	.insns = {
-	/* r6 = skb->sk; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	/* if (r6 == null) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 8),
-	/* r7 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r0 = sk_fullsock(skb); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	/* if (r0 == null) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	/* if (r0 != r7) r0 = *(r7->type); */
-	BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_7, 1), /* Use ! JEQ ! */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_sock, type)),
-	/* return 0; */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "R7 invalid mem access 'sock_or_null'",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R7 pointer comparison",
-},
-{
-	/* Maps are treated in a different branch of `mark_ptr_not_null_reg`,
-	 * so separate test for maps case.
-	 */
-	"jne/jeq infer not null, PTR_TO_MAP_VALUE_OR_NULL -> PTR_TO_MAP_VALUE",
-	.insns = {
-	/* r9 = &some stack to use as key */
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_9, -8),
-	/* r8 = process local map */
-	BPF_LD_MAP_FD(BPF_REG_8, 0),
-	/* r6 = map_lookup_elem(r8, r9); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_9),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* r7 = map_lookup_elem(r8, r9); */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_9),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* if (r6 == 0) return 0; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 2),
-	/* if (r6 != r7) return 0; */
-	BPF_JMP_REG(BPF_JNE, BPF_REG_6, BPF_REG_7, 1),
-	/* read *r7; */
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_xdp_sock, queue_id)),
-	/* return 0; */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_xskmap = { 3 },
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.result = ACCEPT,
-},
-- 
cgit v1.2.3-70-g09d2


From a6fc14dc5e8d43871eebe5455719de6d7578a3bc Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:19 +0300
Subject: selftests/bpf: verifier/loops1 converted to inline assembly

Test verifier/loops1 automatically converted to use inline assembly.

There are a few modifications for the converted tests.
"tracepoint" programs do not support test execution, change program
type to "xdp" (which supports test execution) for the following tests
that have __retval tags:
- bounded loop, count to 4
- bonded loop containing forward jump

Also, remove the __retval tag for test:
- bounded loop, count from positive unknown to 4

As it's return value is a random number.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-10-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_loops1.c  | 259 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/loops1.c      | 206 ----------------
 3 files changed, 261 insertions(+), 206 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_loops1.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/loops1.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 6585432e088f..632546912627 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -32,6 +32,7 @@
 #include "verifier_jeq_infer_not_null.skel.h"
 #include "verifier_ld_ind.skel.h"
 #include "verifier_leak_ptr.skel.h"
+#include "verifier_loops1.skel.h"
 #include "verifier_map_ptr.skel.h"
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
@@ -115,6 +116,7 @@ void test_verifier_int_ptr(void)              { RUN(verifier_int_ptr); }
 void test_verifier_jeq_infer_not_null(void)   { RUN(verifier_jeq_infer_not_null); }
 void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
+void test_verifier_loops1(void)               { RUN(verifier_loops1); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_loops1.c b/tools/testing/selftests/bpf/progs/verifier_loops1.c
new file mode 100644
index 000000000000..5bc86af80a9a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_loops1.c
@@ -0,0 +1,259 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/loops1.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("xdp")
+__description("bounded loop, count to 4")
+__success __retval(4)
+__naked void bounded_loop_count_to_4(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+l0_%=:	r0 += 1;					\
+	if r0 < 4 goto l0_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded loop, count to 20")
+__success
+__naked void bounded_loop_count_to_20(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+l0_%=:	r0 += 3;					\
+	if r0 < 20 goto l0_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded loop, count from positive unknown to 4")
+__success
+__naked void from_positive_unknown_to_4(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	if r0 s< 0 goto l0_%=;				\
+l1_%=:	r0 += 1;					\
+	if r0 < 4 goto l1_%=;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded loop, count from totally unknown to 4")
+__success
+__naked void from_totally_unknown_to_4(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+l0_%=:	r0 += 1;					\
+	if r0 < 4 goto l0_%=;				\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded loop, count to 4 with equality")
+__success
+__naked void count_to_4_with_equality(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+l0_%=:	r0 += 1;					\
+	if r0 != 4 goto l0_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded loop, start in the middle")
+__failure __msg("back-edge")
+__naked void loop_start_in_the_middle(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	goto l0_%=;					\
+l1_%=:	r0 += 1;					\
+l0_%=:	if r0 < 4 goto l1_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("xdp")
+__description("bounded loop containing a forward jump")
+__success __retval(4)
+__naked void loop_containing_a_forward_jump(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+l1_%=:	r0 += 1;					\
+	if r0 == r0 goto l0_%=;				\
+l0_%=:	if r0 < 4 goto l1_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded loop that jumps out rather than in")
+__success
+__naked void jumps_out_rather_than_in(void)
+{
+	asm volatile ("					\
+	r6 = 0;						\
+l1_%=:	r6 += 1;					\
+	if r6 > 10000 goto l0_%=;			\
+	call %[bpf_get_prandom_u32];			\
+	goto l1_%=;					\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("infinite loop after a conditional jump")
+__failure __msg("program is too large")
+__naked void loop_after_a_conditional_jump(void)
+{
+	asm volatile ("					\
+	r0 = 5;						\
+	if r0 < 4 goto l0_%=;				\
+l1_%=:	r0 += 1;					\
+	goto l1_%=;					\
+l0_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("bounded recursion")
+__failure __msg("back-edge")
+__naked void bounded_recursion(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	call bounded_recursion__1;			\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void bounded_recursion__1(void)
+{
+	asm volatile ("					\
+	r1 += 1;					\
+	r0 = r1;					\
+	if r1 < 4 goto l0_%=;				\
+	exit;						\
+l0_%=:	call bounded_recursion__1;			\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("infinite loop in two jumps")
+__failure __msg("loop detected")
+__naked void infinite_loop_in_two_jumps(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+l1_%=:	goto l0_%=;					\
+l0_%=:	if r0 < 4 goto l1_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("infinite loop: three-jump trick")
+__failure __msg("loop detected")
+__naked void infinite_loop_three_jump_trick(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+l2_%=:	r0 += 1;					\
+	r0 &= 1;					\
+	if r0 < 2 goto l0_%=;				\
+	exit;						\
+l0_%=:	r0 += 1;					\
+	r0 &= 1;					\
+	if r0 < 2 goto l1_%=;				\
+	exit;						\
+l1_%=:	r0 += 1;					\
+	r0 &= 1;					\
+	if r0 < 2 goto l2_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("xdp")
+__description("not-taken loop with back jump to 1st insn")
+__success __retval(123)
+__naked void back_jump_to_1st_insn_1(void)
+{
+	asm volatile ("					\
+l0_%=:	r0 = 123;					\
+	if r0 == 4 goto l0_%=;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("xdp")
+__description("taken loop with back jump to 1st insn")
+__success __retval(55)
+__naked void back_jump_to_1st_insn_2(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	r2 = 0;						\
+	call back_jump_to_1st_insn_2__1;		\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void back_jump_to_1st_insn_2__1(void)
+{
+	asm volatile ("					\
+l0_%=:	r2 += r1;					\
+	r1 -= 1;					\
+	if r1 != 0 goto l0_%=;				\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("xdp")
+__description("taken loop with back jump to 1st insn, 2")
+__success __retval(55)
+__naked void jump_to_1st_insn_2(void)
+{
+	asm volatile ("					\
+	r1 = 10;					\
+	r2 = 0;						\
+	call jump_to_1st_insn_2__1;			\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void jump_to_1st_insn_2__1(void)
+{
+	asm volatile ("					\
+l0_%=:	r2 += r1;					\
+	r1 -= 1;					\
+	if w1 != 0 goto l0_%=;				\
+	r0 = r2;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/loops1.c b/tools/testing/selftests/bpf/verifier/loops1.c
deleted file mode 100644
index 1af37187dc12..000000000000
--- a/tools/testing/selftests/bpf/verifier/loops1.c
+++ /dev/null
@@ -1,206 +0,0 @@
-{
-	"bounded loop, count to 4",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, -2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.retval = 4,
-},
-{
-	"bounded loop, count to 20",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 3),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 20, -2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bounded loop, count from positive unknown to 4",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_0, 0, 2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, -2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.retval = 4,
-},
-{
-	"bounded loop, count from totally unknown to 4",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, -2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bounded loop, count to 4 with equality",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 4, -2),
-		BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bounded loop, start in the middle",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_JMP_A(1),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-		BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, -2),
-		BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "back-edge",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.retval = 4,
-},
-{
-	"bounded loop containing a forward jump",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-		BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, -3),
-		BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.retval = 4,
-},
-{
-	"bounded loop that jumps out rather than in",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_6, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 1),
-	BPF_JMP_IMM(BPF_JGT, BPF_REG_6, 10000, 2),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_JMP_A(-4),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"infinite loop after a conditional jump",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 5),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, 2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_JMP_A(-2),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "program is too large",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"bounded recursion",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 1),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_1, 4, 1),
-	BPF_EXIT_INSN(),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, -5),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "back-edge",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"infinite loop in two jumps",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_JMP_A(0),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 4, -2),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "loop detected",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"infinite loop: three-jump trick",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 2, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 2, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
-	BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 2, -11),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "loop detected",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"not-taken loop with back jump to 1st insn",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 123),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 4, -2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.retval = 123,
-},
-{
-	"taken loop with back jump to 1st insn",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 1),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, -3),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.retval = 55,
-},
-{
-	"taken loop with back jump to 1st insn, 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 10),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_SUB, BPF_REG_1, 1),
-	BPF_JMP32_IMM(BPF_JNE, BPF_REG_1, 0, -3),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.retval = 55,
-},
-- 
cgit v1.2.3-70-g09d2


From b427ca576f83d6853fc323639c5f4608e876527f Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:20 +0300
Subject: selftests/bpf: verifier/lwt converted to inline assembly

Test verifier/lwt automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-11-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   2 +
 tools/testing/selftests/bpf/progs/verifier_lwt.c  | 234 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/lwt.c        | 189 -----------------
 3 files changed, 236 insertions(+), 189 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_lwt.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/lwt.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 632546912627..32e55c8e3e3b 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -33,6 +33,7 @@
 #include "verifier_ld_ind.skel.h"
 #include "verifier_leak_ptr.skel.h"
 #include "verifier_loops1.skel.h"
+#include "verifier_lwt.skel.h"
 #include "verifier_map_ptr.skel.h"
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
@@ -117,6 +118,7 @@ void test_verifier_jeq_infer_not_null(void)   { RUN(verifier_jeq_infer_not_null)
 void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
 void test_verifier_loops1(void)               { RUN(verifier_loops1); }
+void test_verifier_lwt(void)                  { RUN(verifier_lwt); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_lwt.c b/tools/testing/selftests/bpf/progs/verifier_lwt.c
new file mode 100644
index 000000000000..5ab746307309
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_lwt.c
@@ -0,0 +1,234 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/lwt.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("lwt_in")
+__description("invalid direct packet write for LWT_IN")
+__failure __msg("cannot write into packet")
+__naked void packet_write_for_lwt_in(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_out")
+__description("invalid direct packet write for LWT_OUT")
+__failure __msg("cannot write into packet")
+__naked void packet_write_for_lwt_out(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_xmit")
+__description("direct packet write for LWT_XMIT")
+__success __retval(0)
+__naked void packet_write_for_lwt_xmit(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	*(u8*)(r2 + 0) = r2;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("direct packet read for LWT_IN")
+__success __retval(0)
+__naked void packet_read_for_lwt_in(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_out")
+__description("direct packet read for LWT_OUT")
+__success __retval(0)
+__naked void packet_read_for_lwt_out(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_xmit")
+__description("direct packet read for LWT_XMIT")
+__success __retval(0)
+__naked void packet_read_for_lwt_xmit(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r0 = *(u8*)(r2 + 0);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_xmit")
+__description("overlapping checks for direct packet access")
+__success __retval(0)
+__naked void checks_for_direct_packet_access(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 8;					\
+	if r0 > r3 goto l0_%=;				\
+	r1 = r2;					\
+	r1 += 6;					\
+	if r1 > r3 goto l0_%=;				\
+	r0 = *(u16*)(r2 + 6);				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("lwt_xmit")
+__description("make headroom for LWT_XMIT")
+__success __retval(0)
+__naked void make_headroom_for_lwt_xmit(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r2 = 34;					\
+	r3 = 0;						\
+	call %[bpf_skb_change_head];			\
+	/* split for s390 to succeed */			\
+	r1 = r6;					\
+	r2 = 42;					\
+	r3 = 0;						\
+	call %[bpf_skb_change_head];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_skb_change_head)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid access of tc_classid for LWT_IN")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void tc_classid_for_lwt_in(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_tc_classid]);	\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_tc_classid, offsetof(struct __sk_buff, tc_classid))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid access of tc_classid for LWT_OUT")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void tc_classid_for_lwt_out(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_tc_classid]);	\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_tc_classid, offsetof(struct __sk_buff, tc_classid))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid access of tc_classid for LWT_XMIT")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv
+__naked void tc_classid_for_lwt_xmit(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_tc_classid]);	\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_tc_classid, offsetof(struct __sk_buff, tc_classid))
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("check skb->tc_classid half load not permitted for lwt prog")
+__failure __msg("invalid bpf_context access")
+__naked void not_permitted_for_lwt_prog(void)
+{
+	asm volatile (
+	"r0 = 0;"
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+	"r0 = *(u16*)(r1 + %[__sk_buff_tc_classid]);"
+#else
+	"r0 = *(u16*)(r1 + %[__imm_0]);"
+#endif
+	"exit;"
+	:
+	: __imm_const(__imm_0, offsetof(struct __sk_buff, tc_classid) + 2),
+	  __imm_const(__sk_buff_tc_classid, offsetof(struct __sk_buff, tc_classid))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/lwt.c b/tools/testing/selftests/bpf/verifier/lwt.c
deleted file mode 100644
index 5c8944d0b091..000000000000
--- a/tools/testing/selftests/bpf/verifier/lwt.c
+++ /dev/null
@@ -1,189 +0,0 @@
-{
-	"invalid direct packet write for LWT_IN",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "cannot write into packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"invalid direct packet write for LWT_OUT",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "cannot write into packet",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_OUT,
-},
-{
-	"direct packet write for LWT_XMIT",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_STX_MEM(BPF_B, BPF_REG_2, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_XMIT,
-},
-{
-	"direct packet read for LWT_IN",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"direct packet read for LWT_OUT",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_OUT,
-},
-{
-	"direct packet read for LWT_XMIT",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_XMIT,
-},
-{
-	"overlapping checks for direct packet access",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_2, 6),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_XMIT,
-},
-{
-	"make headroom for LWT_XMIT",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_2, 34),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_skb_change_head),
-	/* split for s390 to succeed */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_IMM(BPF_REG_2, 42),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_EMIT_CALL(BPF_FUNC_skb_change_head),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_LWT_XMIT,
-},
-{
-	"invalid access of tc_classid for LWT_IN",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_classid)),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-},
-{
-	"invalid access of tc_classid for LWT_OUT",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_classid)),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-},
-{
-	"invalid access of tc_classid for LWT_XMIT",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_classid)),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-},
-{
-	"check skb->tc_classid half load not permitted for lwt prog",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_classid)),
-#else
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, tc_classid) + 2),
-#endif
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "invalid bpf_context access",
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-- 
cgit v1.2.3-70-g09d2


From 4a400ef9ba416494944db52f95d0f16ea04adfdf Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:21 +0300
Subject: selftests/bpf: verifier/map_in_map converted to inline assembly

Test verifier/map_in_map automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-12-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_map_in_map.c      | 142 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/map_in_map.c  |  96 --------------
 3 files changed, 144 insertions(+), 96 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_map_in_map.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/map_in_map.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 32e55c8e3e3b..a5ba0e198246 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -34,6 +34,7 @@
 #include "verifier_leak_ptr.skel.h"
 #include "verifier_loops1.skel.h"
 #include "verifier_lwt.skel.h"
+#include "verifier_map_in_map.skel.h"
 #include "verifier_map_ptr.skel.h"
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
@@ -119,6 +120,7 @@ void test_verifier_ld_ind(void)               { RUN(verifier_ld_ind); }
 void test_verifier_leak_ptr(void)             { RUN(verifier_leak_ptr); }
 void test_verifier_loops1(void)               { RUN(verifier_loops1); }
 void test_verifier_lwt(void)                  { RUN(verifier_lwt); }
+void test_verifier_map_in_map(void)           { RUN(verifier_map_in_map); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_map_in_map.c b/tools/testing/selftests/bpf/progs/verifier_map_in_map.c
new file mode 100644
index 000000000000..4eaab1468eb7
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_map_in_map.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/map_in_map.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+	__array(values, struct {
+		__uint(type, BPF_MAP_TYPE_ARRAY);
+		__uint(max_entries, 1);
+		__type(key, int);
+		__type(value, int);
+	});
+} map_in_map SEC(".maps");
+
+SEC("socket")
+__description("map in map access")
+__success __success_unpriv __retval(0)
+__naked void map_in_map_access(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_in_map] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = r0;					\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_in_map)
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("map in map state pruning")
+__success __msg("processed 26 insns")
+__log_level(2) __retval(0) __flag(BPF_F_TEST_STATE_FREQ)
+__naked void map_in_map_state_pruning(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r6 = r10;					\
+	r6 += -4;					\
+	r2 = r6;					\
+	r1 = %[map_in_map] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r2 = r6;					\
+	r1 = r0;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l1_%=;				\
+	r2 = r6;					\
+	r1 = %[map_in_map] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l2_%=;				\
+	exit;						\
+l2_%=:	r2 = r6;					\
+	r1 = r0;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r0 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_in_map)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("invalid inner map pointer")
+__failure __msg("R1 pointer arithmetic on map_ptr prohibited")
+__failure_unpriv
+__naked void invalid_inner_map_pointer(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_in_map] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = r0;					\
+	r1 += 8;					\
+	call %[bpf_map_lookup_elem];			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_in_map)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("forgot null checking on the inner map pointer")
+__failure __msg("R1 type=map_value_or_null expected=map_ptr")
+__failure_unpriv
+__naked void on_the_inner_map_pointer(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_in_map] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = r0;					\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_in_map)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/map_in_map.c b/tools/testing/selftests/bpf/verifier/map_in_map.c
deleted file mode 100644
index 128a348b762d..000000000000
--- a/tools/testing/selftests/bpf/verifier/map_in_map.c
+++ /dev/null
@@ -1,96 +0,0 @@
-{
-	"map in map access",
-	.insns = {
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_in_map = { 3 },
-	.result = ACCEPT,
-},
-{
-	"map in map state pruning",
-	.insns = {
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -4),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_6),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 11),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_6),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_in_map = { 4, 14 },
-	.flags = BPF_F_TEST_STATE_FREQ,
-	.result = VERBOSE_ACCEPT,
-	.errstr = "processed 25 insns",
-	.prog_type = BPF_PROG_TYPE_XDP,
-},
-{
-	"invalid inner map pointer",
-	.insns = {
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_in_map = { 3 },
-	.errstr = "R1 pointer arithmetic on map_ptr prohibited",
-	.result = REJECT,
-},
-{
-	"forgot null checking on the inner map pointer",
-	.insns = {
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_ST_MEM(0, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_in_map = { 3 },
-	.errstr = "R1 type=map_value_or_null expected=map_ptr",
-	.result = REJECT,
-},
-- 
cgit v1.2.3-70-g09d2


From aee1779f0deca501e904fbf007ccb7232972e518 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:22 +0300
Subject: selftests/bpf: verifier/map_ptr_mixing converted to inline assembly

Test verifier/map_ptr_mixing automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-13-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_map_ptr_mixing.c  | 265 +++++++++++++++++++++
 .../selftests/bpf/verifier/map_ptr_mixing.c        | 100 --------
 3 files changed, 267 insertions(+), 100 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_map_ptr_mixing.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/map_ptr_mixing.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index a5ba0e198246..58202c3fca07 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -36,6 +36,7 @@
 #include "verifier_lwt.skel.h"
 #include "verifier_map_in_map.skel.h"
 #include "verifier_map_ptr.skel.h"
+#include "verifier_map_ptr_mixing.skel.h"
 #include "verifier_map_ret_val.skel.h"
 #include "verifier_masking.skel.h"
 #include "verifier_meta_access.skel.h"
@@ -122,6 +123,7 @@ void test_verifier_loops1(void)               { RUN(verifier_loops1); }
 void test_verifier_lwt(void)                  { RUN(verifier_lwt); }
 void test_verifier_map_in_map(void)           { RUN(verifier_map_in_map); }
 void test_verifier_map_ptr(void)              { RUN(verifier_map_ptr); }
+void test_verifier_map_ptr_mixing(void)       { RUN(verifier_map_ptr_mixing); }
 void test_verifier_map_ret_val(void)          { RUN(verifier_map_ret_val); }
 void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_map_ptr_mixing.c b/tools/testing/selftests/bpf/progs/verifier_map_ptr_mixing.c
new file mode 100644
index 000000000000..c5a7c1ddc562
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_map_ptr_mixing.c
@@ -0,0 +1,265 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/map_ptr_mixing.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+} map_array_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+	__array(values, struct {
+		__uint(type, BPF_MAP_TYPE_ARRAY);
+		__uint(max_entries, 1);
+		__type(key, int);
+		__type(value, int);
+	});
+} map_in_map SEC(".maps");
+
+void dummy_prog_42_socket(void);
+void dummy_prog_24_socket(void);
+void dummy_prog_loop1_socket(void);
+void dummy_prog_loop2_socket(void);
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 4);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog1_socket SEC(".maps") = {
+	.values = {
+		[0] = (void *)&dummy_prog_42_socket,
+		[1] = (void *)&dummy_prog_loop1_socket,
+		[2] = (void *)&dummy_prog_24_socket,
+	},
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 8);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog2_socket SEC(".maps") = {
+	.values = {
+		[1] = (void *)&dummy_prog_loop2_socket,
+		[2] = (void *)&dummy_prog_24_socket,
+		[7] = (void *)&dummy_prog_42_socket,
+	},
+};
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_42_socket(void)
+{
+	asm volatile ("r0 = 42; exit;");
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_24_socket(void)
+{
+	asm volatile ("r0 = 24; exit;");
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_loop1_socket(void)
+{
+	asm volatile ("			\
+	r3 = 1;				\
+	r2 = %[map_prog1_socket] ll;	\
+	call %[bpf_tail_call];		\
+	r0 = 41;			\
+	exit;				\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_loop2_socket(void)
+{
+	asm volatile ("			\
+	r3 = 1;				\
+	r2 = %[map_prog2_socket] ll;	\
+	call %[bpf_tail_call];		\
+	r0 = 41;			\
+	exit;				\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog2_socket)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("calls: two calls returning different map pointers for lookup (hash, array)")
+__success __retval(1)
+__naked void pointers_for_lookup_hash_array(void)
+{
+	asm volatile ("					\
+	/* main prog */					\
+	if r1 != 0 goto l0_%=;				\
+	call pointers_for_lookup_hash_array__1;		\
+	goto l1_%=;					\
+l0_%=:	call pointers_for_lookup_hash_array__2;		\
+l1_%=:	r1 = r0;					\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+	r0 = 1;						\
+l2_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void pointers_for_lookup_hash_array__1(void)
+{
+	asm volatile ("					\
+	r0 = %[map_hash_48b] ll;			\
+	exit;						\
+"	:
+	: __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void pointers_for_lookup_hash_array__2(void)
+{
+	asm volatile ("					\
+	r0 = %[map_array_48b] ll;			\
+	exit;						\
+"	:
+	: __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("calls: two calls returning different map pointers for lookup (hash, map in map)")
+__failure __msg("only read from bpf_array is supported")
+__naked void lookup_hash_map_in_map(void)
+{
+	asm volatile ("					\
+	/* main prog */					\
+	if r1 != 0 goto l0_%=;				\
+	call lookup_hash_map_in_map__1;			\
+	goto l1_%=;					\
+l0_%=:	call lookup_hash_map_in_map__2;			\
+l1_%=:	r1 = r0;					\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+	r0 = 1;						\
+l2_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void lookup_hash_map_in_map__1(void)
+{
+	asm volatile ("					\
+	r0 = %[map_array_48b] ll;			\
+	exit;						\
+"	:
+	: __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void lookup_hash_map_in_map__2(void)
+{
+	asm volatile ("					\
+	r0 = %[map_in_map] ll;				\
+	exit;						\
+"	:
+	: __imm_addr(map_in_map)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("cond: two branches returning different map pointers for lookup (tail, tail)")
+__success __failure_unpriv __msg_unpriv("tail_call abusing map_ptr")
+__retval(42)
+__naked void pointers_for_lookup_tail_tail_1(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	if r6 != 0 goto l0_%=;				\
+	r2 = %[map_prog2_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r2 = %[map_prog1_socket] ll;			\
+l1_%=:	r3 = 7;						\
+	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_addr(map_prog2_socket),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("cond: two branches returning same map pointers for lookup (tail, tail)")
+__success __success_unpriv __retval(42)
+__naked void pointers_for_lookup_tail_tail_2(void)
+{
+	asm volatile ("					\
+	r6 = *(u32*)(r1 + %[__sk_buff_mark]);		\
+	if r6 == 0 goto l0_%=;				\
+	r2 = %[map_prog2_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r2 = %[map_prog2_socket] ll;			\
+l1_%=:	r3 = 7;						\
+	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog2_socket),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/map_ptr_mixing.c b/tools/testing/selftests/bpf/verifier/map_ptr_mixing.c
deleted file mode 100644
index 1f2b8c4cb26d..000000000000
--- a/tools/testing/selftests/bpf/verifier/map_ptr_mixing.c
+++ /dev/null
@@ -1,100 +0,0 @@
-{
-	"calls: two calls returning different map pointers for lookup (hash, array)",
-	.insns = {
-	/* main prog */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_CALL_REL(11),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_CALL_REL(12),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	/* subprog 1 */
-	BPF_LD_MAP_FD(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	/* subprog 2 */
-	BPF_LD_MAP_FD(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_hash_48b = { 13 },
-	.fixup_map_array_48b = { 16 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"calls: two calls returning different map pointers for lookup (hash, map in map)",
-	.insns = {
-	/* main prog */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_CALL_REL(11),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_CALL_REL(12),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	/* subprog 1 */
-	BPF_LD_MAP_FD(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	/* subprog 2 */
-	BPF_LD_MAP_FD(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.fixup_map_in_map = { 16 },
-	.fixup_map_array_48b = { 13 },
-	.result = REJECT,
-	.errstr = "only read from bpf_array is supported",
-},
-{
-	"cond: two branches returning different map pointers for lookup (tail, tail)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_6, 0, 3),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5 },
-	.fixup_prog2 = { 2 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "tail_call abusing map_ptr",
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"cond: two branches returning same map pointers for lookup (tail, tail)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_1,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 3),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog2 = { 2, 5 },
-	.result_unpriv = ACCEPT,
-	.result = ACCEPT,
-	.retval = 42,
-},
-- 
cgit v1.2.3-70-g09d2


From 8be632795996c198a26b6f39727d9c92007bb6a5 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:25 +0300
Subject: selftests/bpf: verifier/ref_tracking converted to inline assembly

Test verifier/ref_tracking automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-16-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |    2 +
 .../selftests/bpf/progs/verifier_ref_tracking.c    | 1495 ++++++++++++++++++++
 .../testing/selftests/bpf/verifier/ref_tracking.c  | 1082 --------------
 3 files changed, 1497 insertions(+), 1082 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_ref_tracking.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/ref_tracking.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 58202c3fca07..75efbc20fb8e 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -45,6 +45,7 @@
 #include "verifier_raw_stack.skel.h"
 #include "verifier_raw_tp_writable.skel.h"
 #include "verifier_reg_equal.skel.h"
+#include "verifier_ref_tracking.skel.h"
 #include "verifier_ringbuf.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
@@ -132,6 +133,7 @@ void test_verifier_netfilter_retcode(void)    { RUN(verifier_netfilter_retcode);
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_reg_equal(void)            { RUN(verifier_reg_equal); }
+void test_verifier_ref_tracking(void)         { RUN(verifier_ref_tracking); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_ref_tracking.c b/tools/testing/selftests/bpf/progs/verifier_ref_tracking.c
new file mode 100644
index 000000000000..c4c6da21265e
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_ref_tracking.c
@@ -0,0 +1,1495 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/ref_tracking.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "../../../include/linux/filter.h"
+#include "bpf_misc.h"
+
+#define BPF_SK_LOOKUP(func) \
+	/* struct bpf_sock_tuple tuple = {} */ \
+	"r2 = 0;"			\
+	"*(u32*)(r10 - 8) = r2;"	\
+	"*(u64*)(r10 - 16) = r2;"	\
+	"*(u64*)(r10 - 24) = r2;"	\
+	"*(u64*)(r10 - 32) = r2;"	\
+	"*(u64*)(r10 - 40) = r2;"	\
+	"*(u64*)(r10 - 48) = r2;"	\
+	/* sk = func(ctx, &tuple, sizeof tuple, 0, 0) */ \
+	"r2 = r10;"			\
+	"r2 += -48;"			\
+	"r3 = %[sizeof_bpf_sock_tuple];"\
+	"r4 = 0;"			\
+	"r5 = 0;"			\
+	"call %[" #func "];"
+
+struct bpf_key {} __attribute__((preserve_access_index));
+
+extern void bpf_key_put(struct bpf_key *key) __ksym;
+extern struct bpf_key *bpf_lookup_system_key(__u64 id) __ksym;
+extern struct bpf_key *bpf_lookup_user_key(__u32 serial, __u64 flags) __ksym;
+
+/* BTF FUNC records are not generated for kfuncs referenced
+ * from inline assembly. These records are necessary for
+ * libbpf to link the program. The function below is a hack
+ * to ensure that BTF FUNC records are generated.
+ */
+void __kfunc_btf_root(void)
+{
+	bpf_key_put(0);
+	bpf_lookup_system_key(0);
+	bpf_lookup_user_key(0, 0);
+}
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+} map_array_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_RINGBUF);
+	__uint(max_entries, 4096);
+} map_ringbuf SEC(".maps");
+
+void dummy_prog_42_tc(void);
+void dummy_prog_24_tc(void);
+void dummy_prog_loop1_tc(void);
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 4);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog1_tc SEC(".maps") = {
+	.values = {
+		[0] = (void *)&dummy_prog_42_tc,
+		[1] = (void *)&dummy_prog_loop1_tc,
+		[2] = (void *)&dummy_prog_24_tc,
+	},
+};
+
+SEC("tc")
+__auxiliary
+__naked void dummy_prog_42_tc(void)
+{
+	asm volatile ("r0 = 42; exit;");
+}
+
+SEC("tc")
+__auxiliary
+__naked void dummy_prog_24_tc(void)
+{
+	asm volatile ("r0 = 24; exit;");
+}
+
+SEC("tc")
+__auxiliary
+__naked void dummy_prog_loop1_tc(void)
+{
+	asm volatile ("			\
+	r3 = 1;				\
+	r2 = %[map_prog1_tc] ll;	\
+	call %[bpf_tail_call];		\
+	r0 = 41;			\
+	exit;				\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_tc)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: leak potential reference")
+__failure __msg("Unreleased reference")
+__naked void reference_tracking_leak_potential_reference(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;		/* leak reference */	\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: leak potential reference to sock_common")
+__failure __msg("Unreleased reference")
+__naked void potential_reference_to_sock_common_1(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_skc_lookup_tcp)
+"	r6 = r0;		/* leak reference */	\
+	exit;						\
+"	:
+	: __imm(bpf_skc_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: leak potential reference on stack")
+__failure __msg("Unreleased reference")
+__naked void leak_potential_reference_on_stack(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r4 = r10;					\
+	r4 += -8;					\
+	*(u64*)(r4 + 0) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: leak potential reference on stack 2")
+__failure __msg("Unreleased reference")
+__naked void potential_reference_on_stack_2(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r4 = r10;					\
+	r4 += -8;					\
+	*(u64*)(r4 + 0) = r0;				\
+	r0 = 0;						\
+	r1 = 0;						\
+	*(u64*)(r4 + 0) = r1;				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: zero potential reference")
+__failure __msg("Unreleased reference")
+__naked void reference_tracking_zero_potential_reference(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r0 = 0;			/* leak reference */	\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: zero potential reference to sock_common")
+__failure __msg("Unreleased reference")
+__naked void potential_reference_to_sock_common_2(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_skc_lookup_tcp)
+"	r0 = 0;			/* leak reference */	\
+	exit;						\
+"	:
+	: __imm(bpf_skc_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: copy and zero potential references")
+__failure __msg("Unreleased reference")
+__naked void copy_and_zero_potential_references(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r7 = r0;					\
+	r0 = 0;						\
+	r7 = 0;			/* leak reference */	\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: acquire/release user key reference")
+__success
+__naked void acquire_release_user_key_reference(void)
+{
+	asm volatile ("					\
+	r1 = -3;					\
+	r2 = 0;						\
+	call %[bpf_lookup_user_key];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	call %[bpf_key_put];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_key_put),
+	  __imm(bpf_lookup_user_key)
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: acquire/release system key reference")
+__success
+__naked void acquire_release_system_key_reference(void)
+{
+	asm volatile ("					\
+	r1 = 1;						\
+	call %[bpf_lookup_system_key];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r0;					\
+	call %[bpf_key_put];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_key_put),
+	  __imm(bpf_lookup_system_key)
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: release user key reference without check")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
+__naked void user_key_reference_without_check(void)
+{
+	asm volatile ("					\
+	r1 = -3;					\
+	r2 = 0;						\
+	call %[bpf_lookup_user_key];			\
+	r1 = r0;					\
+	call %[bpf_key_put];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_key_put),
+	  __imm(bpf_lookup_user_key)
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: release system key reference without check")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
+__naked void system_key_reference_without_check(void)
+{
+	asm volatile ("					\
+	r1 = 1;						\
+	call %[bpf_lookup_system_key];			\
+	r1 = r0;					\
+	call %[bpf_key_put];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_key_put),
+	  __imm(bpf_lookup_system_key)
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: release with NULL key pointer")
+__failure __msg("Possibly NULL pointer passed to trusted arg0")
+__naked void release_with_null_key_pointer(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	call %[bpf_key_put];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_key_put)
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: leak potential reference to user key")
+__failure __msg("Unreleased reference")
+__naked void potential_reference_to_user_key(void)
+{
+	asm volatile ("					\
+	r1 = -3;					\
+	r2 = 0;						\
+	call %[bpf_lookup_user_key];			\
+	exit;						\
+"	:
+	: __imm(bpf_lookup_user_key)
+	: __clobber_all);
+}
+
+SEC("lsm.s/bpf")
+__description("reference tracking: leak potential reference to system key")
+__failure __msg("Unreleased reference")
+__naked void potential_reference_to_system_key(void)
+{
+	asm volatile ("					\
+	r1 = 1;						\
+	call %[bpf_lookup_system_key];			\
+	exit;						\
+"	:
+	: __imm(bpf_lookup_system_key)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference without check")
+__failure __msg("type=sock_or_null expected=sock")
+__naked void tracking_release_reference_without_check(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	/* reference in r0 may be NULL */		\
+	r1 = r0;					\
+	r2 = 0;						\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference to sock_common without check")
+__failure __msg("type=sock_common_or_null expected=sock")
+__naked void to_sock_common_without_check(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_skc_lookup_tcp)
+"	/* reference in r0 may be NULL */		\
+	r1 = r0;					\
+	r2 = 0;						\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_release),
+	  __imm(bpf_skc_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference")
+__success __retval(0)
+__naked void reference_tracking_release_reference(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference to sock_common")
+__success __retval(0)
+__naked void release_reference_to_sock_common(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_skc_lookup_tcp)
+"	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_release),
+	  __imm(bpf_skc_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference 2")
+__success __retval(0)
+__naked void reference_tracking_release_reference_2(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference twice")
+__failure __msg("type=scalar expected=sock")
+__naked void reference_tracking_release_reference_twice(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	r6 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference twice inside branch")
+__failure __msg("type=scalar expected=sock")
+__naked void release_reference_twice_inside_branch(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	r6 = r0;					\
+	if r0 == 0 goto l0_%=;		/* goto end */	\
+	call %[bpf_sk_release];				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: alloc, check, free in one subbranch")
+__failure __msg("Unreleased reference")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void check_free_in_one_subbranch(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 16;					\
+	/* if (offsetof(skb, mark) > data_len) exit; */	\
+	if r0 <= r3 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = *(u32*)(r2 + %[__sk_buff_mark]);		\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r6 == 0 goto l1_%=;		/* mark == 0? */\
+	/* Leak reference in R0 */			\
+	exit;						\
+l1_%=:	if r0 == 0 goto l2_%=;		/* sk NULL? */	\
+	r1 = r0;					\
+	call %[bpf_sk_release];				\
+l2_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: alloc, check, free in both subbranches")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void check_free_in_both_subbranches(void)
+{
+	asm volatile ("					\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 16;					\
+	/* if (offsetof(skb, mark) > data_len) exit; */	\
+	if r0 <= r3 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = *(u32*)(r2 + %[__sk_buff_mark]);		\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r6 == 0 goto l1_%=;		/* mark == 0? */\
+	if r0 == 0 goto l2_%=;		/* sk NULL? */	\
+	r1 = r0;					\
+	call %[bpf_sk_release];				\
+l2_%=:	exit;						\
+l1_%=:	if r0 == 0 goto l3_%=;		/* sk NULL? */	\
+	r1 = r0;					\
+	call %[bpf_sk_release];				\
+l3_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking in call: free reference in subprog")
+__success __retval(0)
+__naked void call_free_reference_in_subprog(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;	/* unchecked reference */	\
+	call call_free_reference_in_subprog__1;		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void call_free_reference_in_subprog__1(void)
+{
+	asm volatile ("					\
+	/* subprog 1 */					\
+	r2 = r1;					\
+	if r2 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_release)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking in call: free reference in subprog and outside")
+__failure __msg("type=scalar expected=sock")
+__naked void reference_in_subprog_and_outside(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;	/* unchecked reference */	\
+	r6 = r0;					\
+	call reference_in_subprog_and_outside__1;	\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void reference_in_subprog_and_outside__1(void)
+{
+	asm volatile ("					\
+	/* subprog 1 */					\
+	r2 = r1;					\
+	if r2 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_release)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking in call: alloc & leak reference in subprog")
+__failure __msg("Unreleased reference")
+__naked void alloc_leak_reference_in_subprog(void)
+{
+	asm volatile ("					\
+	r4 = r10;					\
+	r4 += -8;					\
+	call alloc_leak_reference_in_subprog__1;	\
+	r1 = r0;					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void alloc_leak_reference_in_subprog__1(void)
+{
+	asm volatile ("					\
+	/* subprog 1 */					\
+	r6 = r4;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	/* spill unchecked sk_ptr into stack of caller */\
+	*(u64*)(r6 + 0) = r0;				\
+	r1 = r0;					\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking in call: alloc in subprog, release outside")
+__success __retval(POINTER_VALUE)
+__naked void alloc_in_subprog_release_outside(void)
+{
+	asm volatile ("					\
+	r4 = r10;					\
+	call alloc_in_subprog_release_outside__1;	\
+	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_release)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void alloc_in_subprog_release_outside__1(void)
+{
+	asm volatile ("					\
+	/* subprog 1 */					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	exit;				/* return sk */	\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking in call: sk_ptr leak into caller stack")
+__failure __msg("Unreleased reference")
+__naked void ptr_leak_into_caller_stack(void)
+{
+	asm volatile ("					\
+	r4 = r10;					\
+	r4 += -8;					\
+	call ptr_leak_into_caller_stack__1;		\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void ptr_leak_into_caller_stack__1(void)
+{
+	asm volatile ("					\
+	/* subprog 1 */					\
+	r5 = r10;					\
+	r5 += -8;					\
+	*(u64*)(r5 + 0) = r4;				\
+	call ptr_leak_into_caller_stack__2;		\
+	/* spill unchecked sk_ptr into stack of caller */\
+	r5 = r10;					\
+	r5 += -8;					\
+	r4 = *(u64*)(r5 + 0);				\
+	*(u64*)(r4 + 0) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void ptr_leak_into_caller_stack__2(void)
+{
+	asm volatile ("					\
+	/* subprog 2 */					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking in call: sk_ptr spill into caller stack")
+__success __retval(0)
+__naked void ptr_spill_into_caller_stack(void)
+{
+	asm volatile ("					\
+	r4 = r10;					\
+	r4 += -8;					\
+	call ptr_spill_into_caller_stack__1;		\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void ptr_spill_into_caller_stack__1(void)
+{
+	asm volatile ("					\
+	/* subprog 1 */					\
+	r5 = r10;					\
+	r5 += -8;					\
+	*(u64*)(r5 + 0) = r4;				\
+	call ptr_spill_into_caller_stack__2;		\
+	/* spill unchecked sk_ptr into stack of caller */\
+	r5 = r10;					\
+	r5 += -8;					\
+	r4 = *(u64*)(r5 + 0);				\
+	*(u64*)(r4 + 0) = r0;				\
+	if r0 == 0 goto l0_%=;				\
+	/* now the sk_ptr is verified, free the reference */\
+	r1 = *(u64*)(r4 + 0);				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_release)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void ptr_spill_into_caller_stack__2(void)
+{
+	asm volatile ("					\
+	/* subprog 2 */					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: allow LD_ABS")
+__success __retval(0)
+__naked void reference_tracking_allow_ld_abs(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	r0 = *(u8*)skb[0];				\
+	r0 = *(u16*)skb[0];				\
+	r0 = *(u32*)skb[0];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: forbid LD_ABS while holding reference")
+__failure __msg("BPF_LD_[ABS|IND] cannot be mixed with socket references")
+__naked void ld_abs_while_holding_reference(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r0 = *(u8*)skb[0];				\
+	r0 = *(u16*)skb[0];				\
+	r0 = *(u32*)skb[0];				\
+	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: allow LD_IND")
+__success __retval(1)
+__naked void reference_tracking_allow_ld_ind(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	r7 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r7;					\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple)),
+	  __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: forbid LD_IND while holding reference")
+__failure __msg("BPF_LD_[ABS|IND] cannot be mixed with socket references")
+__naked void ld_ind_while_holding_reference(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r4 = r0;					\
+	r7 = 1;						\
+	.8byte %[ld_ind];				\
+	r0 = r7;					\
+	r1 = r4;					\
+	if r1 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple)),
+	  __imm_insn(ld_ind, BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: check reference or tail call")
+__success __retval(0)
+__naked void check_reference_or_tail_call(void)
+{
+	asm volatile ("					\
+	r7 = r1;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	/* if (sk) bpf_sk_release() */			\
+	r1 = r0;					\
+	if r1 != 0 goto l0_%=;				\
+	/* bpf_tail_call() */				\
+	r3 = 3;						\
+	r2 = %[map_prog1_tc] ll;			\
+	r1 = r7;					\
+	call %[bpf_tail_call];				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_tc),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: release reference then tail call")
+__success __retval(0)
+__naked void release_reference_then_tail_call(void)
+{
+	asm volatile ("					\
+	r7 = r1;					\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	/* if (sk) bpf_sk_release() */			\
+	r1 = r0;					\
+	if r1 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	/* bpf_tail_call() */				\
+	r3 = 3;						\
+	r2 = %[map_prog1_tc] ll;			\
+	r1 = r7;					\
+	call %[bpf_tail_call];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_tc),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: leak possible reference over tail call")
+__failure __msg("tail_call would lead to reference leak")
+__naked void possible_reference_over_tail_call(void)
+{
+	asm volatile ("					\
+	r7 = r1;					\
+	/* Look up socket and store in REG_6 */		\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	/* bpf_tail_call() */				\
+	r6 = r0;					\
+	r3 = 3;						\
+	r2 = %[map_prog1_tc] ll;			\
+	r1 = r7;					\
+	call %[bpf_tail_call];				\
+	r0 = 0;						\
+	/* if (sk) bpf_sk_release() */			\
+	r1 = r6;					\
+	if r1 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_tc),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: leak checked reference over tail call")
+__failure __msg("tail_call would lead to reference leak")
+__naked void checked_reference_over_tail_call(void)
+{
+	asm volatile ("					\
+	r7 = r1;					\
+	/* Look up socket and store in REG_6 */		\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;					\
+	/* if (!sk) goto end */				\
+	if r0 == 0 goto l0_%=;				\
+	/* bpf_tail_call() */				\
+	r3 = 0;						\
+	r2 = %[map_prog1_tc] ll;			\
+	r1 = r7;					\
+	call %[bpf_tail_call];				\
+	r0 = 0;						\
+	r1 = r6;					\
+l0_%=:	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_tc),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: mangle and release sock_or_null")
+__failure __msg("R1 pointer arithmetic on sock_or_null prohibited")
+__naked void and_release_sock_or_null(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	r1 += 5;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: mangle and release sock")
+__failure __msg("R1 pointer arithmetic on sock prohibited")
+__naked void tracking_mangle_and_release_sock(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r1 += 5;					\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: access member")
+__success __retval(0)
+__naked void reference_tracking_access_member(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r2 = *(u32*)(r0 + 4);				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: write to member")
+__failure __msg("cannot write into sock")
+__naked void reference_tracking_write_to_member(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r1 = r6;					\
+	r2 = 42 ll;					\
+	*(u32*)(r1 + %[bpf_sock_mark]) = r2;		\
+	r1 = r6;					\
+l0_%=:	call %[bpf_sk_release];				\
+	r0 = 0 ll;					\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(bpf_sock_mark, offsetof(struct bpf_sock, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: invalid 64-bit access of member")
+__failure __msg("invalid sock access off=0 size=8")
+__naked void _64_bit_access_of_member(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r2 = *(u64*)(r0 + 0);				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: access after release")
+__failure __msg("!read_ok")
+__naked void reference_tracking_access_after_release(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r1 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+	r2 = *(u32*)(r1 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: direct access for lookup")
+__success __retval(0)
+__naked void tracking_direct_access_for_lookup(void)
+{
+	asm volatile ("					\
+	/* Check that the packet is at least 64B long */\
+	r2 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r3 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r0 = r2;					\
+	r0 += 64;					\
+	if r0 > r3 goto l0_%=;				\
+	/* sk = sk_lookup_tcp(ctx, skb->data, ...) */	\
+	r3 = %[sizeof_bpf_sock_tuple];			\
+	r4 = 0;						\
+	r5 = 0;						\
+	call %[bpf_sk_lookup_tcp];			\
+	r6 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r2 = *(u32*)(r0 + 4);				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: use ptr from bpf_tcp_sock() after release")
+__failure __msg("invalid mem access")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void bpf_tcp_sock_after_release(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r7 = r0;					\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	r0 = *(u32*)(r7 + %[bpf_tcp_sock_snd_cwnd]);	\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tcp_sock),
+	  __imm_const(bpf_tcp_sock_snd_cwnd, offsetof(struct bpf_tcp_sock, snd_cwnd)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: use ptr from bpf_sk_fullsock() after release")
+__failure __msg("invalid mem access")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void bpf_sk_fullsock_after_release(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r7 = r0;					\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	r0 = *(u32*)(r7 + %[bpf_sock_type]);		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: use ptr from bpf_sk_fullsock(tp) after release")
+__failure __msg("invalid mem access")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void sk_fullsock_tp_after_release(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_sk_fullsock];			\
+	r1 = r6;					\
+	r6 = r0;					\
+	call %[bpf_sk_release];				\
+	if r6 != 0 goto l2_%=;				\
+	exit;						\
+l2_%=:	r0 = *(u32*)(r6 + %[bpf_sock_type]);		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tcp_sock),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: use sk after bpf_sk_release(tp)")
+__failure __msg("invalid mem access")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void after_bpf_sk_release_tp(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_sk_release];				\
+	r0 = *(u32*)(r6 + %[bpf_sock_type]);		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tcp_sock),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: use ptr from bpf_get_listener_sock() after bpf_sk_release(sk)")
+__success __retval(0)
+__naked void after_bpf_sk_release_sk(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_get_listener_sock];			\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r1 = r6;					\
+	r6 = r0;					\
+	call %[bpf_sk_release];				\
+	r0 = *(u32*)(r6 + %[bpf_sock_src_port]);	\
+	exit;						\
+"	:
+	: __imm(bpf_get_listener_sock),
+	  __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(bpf_sock_src_port, offsetof(struct bpf_sock, src_port)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: bpf_sk_release(listen_sk)")
+__failure __msg("R1 must be referenced when passed to release function")
+__naked void bpf_sk_release_listen_sk(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_get_listener_sock];			\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_sk_release];				\
+	r0 = *(u32*)(r6 + %[bpf_sock_type]);		\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_get_listener_sock),
+	  __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+/* !bpf_sk_fullsock(sk) is checked but !bpf_tcp_sock(sk) is not checked */
+SEC("tc")
+__description("reference tracking: tp->snd_cwnd after bpf_sk_fullsock(sk) and bpf_tcp_sock(sk)")
+__failure __msg("invalid mem access")
+__naked void and_bpf_tcp_sock_sk(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_sk_fullsock];			\
+	r7 = r0;					\
+	r1 = r6;					\
+	call %[bpf_tcp_sock];				\
+	r8 = r0;					\
+	if r7 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r8 + %[bpf_tcp_sock_snd_cwnd]);	\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_tcp_sock),
+	  __imm_const(bpf_tcp_sock_snd_cwnd, offsetof(struct bpf_tcp_sock, snd_cwnd)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: branch tracking valid pointer null comparison")
+__success __retval(0)
+__naked void tracking_valid_pointer_null_comparison(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;					\
+	r3 = 1;						\
+	if r6 != 0 goto l0_%=;				\
+	r3 = 0;						\
+l0_%=:	if r6 == 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: branch tracking valid pointer value comparison")
+__failure __msg("Unreleased reference")
+__naked void tracking_valid_pointer_value_comparison(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r6 = r0;					\
+	r3 = 1;						\
+	if r6 == 0 goto l0_%=;				\
+	r3 = 0;						\
+	if r6 == 1234 goto l0_%=;			\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: bpf_sk_release(btf_tcp_sock)")
+__success
+__retval(0)
+__naked void sk_release_btf_tcp_sock(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_skc_to_tcp_sock];			\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_skc_to_tcp_sock),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("reference tracking: use ptr from bpf_skc_to_tcp_sock() after release")
+__failure __msg("invalid mem access")
+__naked void to_tcp_sock_after_release(void)
+{
+	asm volatile (
+	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	call %[bpf_skc_to_tcp_sock];			\
+	if r0 != 0 goto l1_%=;				\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	exit;						\
+l1_%=:	r7 = r0;					\
+	r1 = r6;					\
+	call %[bpf_sk_release];				\
+	r0 = *(u8*)(r7 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm(bpf_skc_to_tcp_sock),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("reference tracking: try to leak released ptr reg")
+__success __failure_unpriv __msg_unpriv("R8 !read_ok")
+__retval(0)
+__naked void to_leak_released_ptr_reg(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	*(u32*)(r10 - 4) = r0;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r9 = r0;					\
+	r0 = 0;						\
+	r1 = %[map_ringbuf] ll;				\
+	r2 = 8;						\
+	r3 = 0;						\
+	call %[bpf_ringbuf_reserve];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r8 = r0;					\
+	r1 = r8;					\
+	r2 = 0;						\
+	call %[bpf_ringbuf_discard];			\
+	r0 = 0;						\
+	*(u64*)(r9 + 0) = r8;				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_ringbuf_discard),
+	  __imm(bpf_ringbuf_reserve),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_ringbuf)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/ref_tracking.c b/tools/testing/selftests/bpf/verifier/ref_tracking.c
deleted file mode 100644
index 5a2e154dd1e0..000000000000
--- a/tools/testing/selftests/bpf/verifier/ref_tracking.c
+++ /dev/null
@@ -1,1082 +0,0 @@
-{
-	"reference tracking: leak potential reference",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0), /* leak reference */
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: leak potential reference to sock_common",
-	.insns = {
-	BPF_SK_LOOKUP(skc_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0), /* leak reference */
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: leak potential reference on stack",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_4, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: leak potential reference on stack 2",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_4, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: zero potential reference",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_IMM(BPF_REG_0, 0), /* leak reference */
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: zero potential reference to sock_common",
-	.insns = {
-	BPF_SK_LOOKUP(skc_lookup_tcp),
-	BPF_MOV64_IMM(BPF_REG_0, 0), /* leak reference */
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: copy and zero potential references",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_7, 0), /* leak reference */
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: acquire/release user key reference",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.fixup_kfunc_btf_id = {
-		{ "bpf_lookup_user_key", 2 },
-		{ "bpf_key_put", 5 },
-	},
-	.result = ACCEPT,
-},
-{
-	"reference tracking: acquire/release system key reference",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.fixup_kfunc_btf_id = {
-		{ "bpf_lookup_system_key", 1 },
-		{ "bpf_key_put", 4 },
-	},
-	.result = ACCEPT,
-},
-{
-	"reference tracking: release user key reference without check",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.errstr = "Possibly NULL pointer passed to trusted arg0",
-	.fixup_kfunc_btf_id = {
-		{ "bpf_lookup_user_key", 2 },
-		{ "bpf_key_put", 4 },
-	},
-	.result = REJECT,
-},
-{
-	"reference tracking: release system key reference without check",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.errstr = "Possibly NULL pointer passed to trusted arg0",
-	.fixup_kfunc_btf_id = {
-		{ "bpf_lookup_system_key", 1 },
-		{ "bpf_key_put", 3 },
-	},
-	.result = REJECT,
-},
-{
-	"reference tracking: release with NULL key pointer",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.errstr = "Possibly NULL pointer passed to trusted arg0",
-	.fixup_kfunc_btf_id = {
-		{ "bpf_key_put", 1 },
-	},
-	.result = REJECT,
-},
-{
-	"reference tracking: leak potential reference to user key",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, -3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.errstr = "Unreleased reference",
-	.fixup_kfunc_btf_id = {
-		{ "bpf_lookup_user_key", 2 },
-	},
-	.result = REJECT,
-},
-{
-	"reference tracking: leak potential reference to system key",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_LSM,
-	.kfunc = "bpf",
-	.expected_attach_type = BPF_LSM_MAC,
-	.flags = BPF_F_SLEEPABLE,
-	.errstr = "Unreleased reference",
-	.fixup_kfunc_btf_id = {
-		{ "bpf_lookup_system_key", 1 },
-	},
-	.result = REJECT,
-},
-{
-	"reference tracking: release reference without check",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	/* reference in r0 may be NULL */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "type=sock_or_null expected=sock",
-	.result = REJECT,
-},
-{
-	"reference tracking: release reference to sock_common without check",
-	.insns = {
-	BPF_SK_LOOKUP(skc_lookup_tcp),
-	/* reference in r0 may be NULL */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "type=sock_common_or_null expected=sock",
-	.result = REJECT,
-},
-{
-	"reference tracking: release reference",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: release reference to sock_common",
-	.insns = {
-	BPF_SK_LOOKUP(skc_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: release reference 2",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: release reference twice",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "type=scalar expected=sock",
-	.result = REJECT,
-},
-{
-	"reference tracking: release reference twice inside branch",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3), /* goto end */
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "type=scalar expected=sock",
-	.result = REJECT,
-},
-{
-	"reference tracking: alloc, check, free in one subbranch",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 16),
-	/* if (offsetof(skb, mark) > data_len) exit; */
-	BPF_JMP_REG(BPF_JLE, BPF_REG_0, BPF_REG_3, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_2,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 1), /* mark == 0? */
-	/* Leak reference in R0 */
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2), /* sk NULL? */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"reference tracking: alloc, check, free in both subbranches",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 16),
-	/* if (offsetof(skb, mark) > data_len) exit; */
-	BPF_JMP_REG(BPF_JLE, BPF_REG_0, BPF_REG_3, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_2,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 4), /* mark == 0? */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2), /* sk NULL? */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2), /* sk NULL? */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"reference tracking in call: free reference in subprog",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0), /* unchecked reference */
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	/* subprog 1 */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_2, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking in call: free reference in subprog and outside",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0), /* unchecked reference */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-
-	/* subprog 1 */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_2, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "type=scalar expected=sock",
-	.result = REJECT,
-},
-{
-	"reference tracking in call: alloc & leak reference in subprog",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	/* subprog 1 */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_4),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	/* spill unchecked sk_ptr into stack of caller */
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking in call: alloc in subprog, release outside",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-
-	/* subprog 1 */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_EXIT_INSN(), /* return sk */
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.retval = POINTER_VALUE,
-	.result = ACCEPT,
-},
-{
-	"reference tracking in call: sk_ptr leak into caller stack",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	/* subprog 1 */
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_5, BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 5),
-	/* spill unchecked sk_ptr into stack of caller */
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_5, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_4, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	/* subprog 2 */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking in call: sk_ptr spill into caller stack",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -8),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-
-	/* subprog 1 */
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_5, BPF_REG_4, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 8),
-	/* spill unchecked sk_ptr into stack of caller */
-	BPF_MOV64_REG(BPF_REG_5, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_5, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	/* now the sk_ptr is verified, free the reference */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_4, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-
-	/* subprog 2 */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: allow LD_ABS",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LD_ABS(BPF_B, 0),
-	BPF_LD_ABS(BPF_H, 0),
-	BPF_LD_ABS(BPF_W, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: forbid LD_ABS while holding reference",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_LD_ABS(BPF_B, 0),
-	BPF_LD_ABS(BPF_H, 0),
-	BPF_LD_ABS(BPF_W, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "BPF_LD_[ABS|IND] cannot be mixed with socket references",
-	.result = REJECT,
-},
-{
-	"reference tracking: allow LD_IND",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_7, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_7),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"reference tracking: forbid LD_IND while holding reference",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_7, 1),
-	BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_7),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_4),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "BPF_LD_[ABS|IND] cannot be mixed with socket references",
-	.result = REJECT,
-},
-{
-	"reference tracking: check reference or tail call",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_1),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	/* if (sk) bpf_sk_release() */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 7),
-	/* bpf_tail_call() */
-	BPF_MOV64_IMM(BPF_REG_3, 3),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 17 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: release reference then tail call",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_1),
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	/* if (sk) bpf_sk_release() */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	/* bpf_tail_call() */
-	BPF_MOV64_IMM(BPF_REG_3, 3),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 18 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: leak possible reference over tail call",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_1),
-	/* Look up socket and store in REG_6 */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	/* bpf_tail_call() */
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 3),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	/* if (sk) bpf_sk_release() */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 16 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "tail_call would lead to reference leak",
-	.result = REJECT,
-},
-{
-	"reference tracking: leak checked reference over tail call",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_1),
-	/* Look up socket and store in REG_6 */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* if (!sk) goto end */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	/* bpf_tail_call() */
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 17 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "tail_call would lead to reference leak",
-	.result = REJECT,
-},
-{
-	"reference tracking: mangle and release sock_or_null",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 5),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "R1 pointer arithmetic on sock_or_null prohibited",
-	.result = REJECT,
-},
-{
-	"reference tracking: mangle and release sock",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 5),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "R1 pointer arithmetic on sock prohibited",
-	.result = REJECT,
-},
-{
-	"reference tracking: access member",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: write to member",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_LD_IMM64(BPF_REG_2, 42),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_2,
-		    offsetof(struct bpf_sock, mark)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LD_IMM64(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "cannot write into sock",
-	.result = REJECT,
-},
-{
-	"reference tracking: invalid 64-bit access of member",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "invalid sock access off=0 size=8",
-	.result = REJECT,
-},
-{
-	"reference tracking: access after release",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "!read_ok",
-	.result = REJECT,
-},
-{
-	"reference tracking: direct access for lookup",
-	.insns = {
-	/* Check that the packet is at least 64B long */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 64),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 9),
-	/* sk = sk_lookup_tcp(ctx, skb->data, ...) */
-	BPF_MOV64_IMM(BPF_REG_3, sizeof(struct bpf_sock_tuple)),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_5, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_0, 4),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: use ptr from bpf_tcp_sock() after release",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_tcp_sock, snd_cwnd)),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"reference tracking: use ptr from bpf_sk_fullsock() after release",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_7, offsetof(struct bpf_sock, type)),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"reference tracking: use ptr from bpf_sk_fullsock(tp) after release",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_6, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, offsetof(struct bpf_sock, type)),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"reference tracking: use sk after bpf_sk_release(tp)",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, offsetof(struct bpf_sock, type)),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"reference tracking: use ptr from bpf_get_listener_sock() after bpf_sk_release(sk)",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_listener_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, offsetof(struct bpf_sock, src_port)),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: bpf_sk_release(listen_sk)",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_listener_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, offsetof(struct bpf_sock, type)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "R1 must be referenced when passed to release function",
-},
-{
-	/* !bpf_sk_fullsock(sk) is checked but !bpf_tcp_sock(sk) is not checked */
-	"reference tracking: tp->snd_cwnd after bpf_sk_fullsock(sk) and bpf_tcp_sock(sk)",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_7, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_8, offsetof(struct bpf_tcp_sock, snd_cwnd)),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-},
-{
-	"reference tracking: branch tracking valid pointer null comparison",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_6, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"reference tracking: branch tracking valid pointer value comparison",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 1234, 2),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.errstr = "Unreleased reference",
-	.result = REJECT,
-},
-{
-	"reference tracking: bpf_sk_release(btf_tcp_sock)",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_skc_to_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "unknown func",
-},
-{
-	"reference tracking: use ptr from bpf_skc_to_tcp_sock() after release",
-	.insns = {
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_skc_to_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 3),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "unknown func",
-},
-{
-	"reference tracking: try to leak released ptr reg",
-	.insns = {
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4),
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-		BPF_LD_MAP_FD(BPF_REG_1, 0),
-		BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-		BPF_EXIT_INSN(),
-		BPF_MOV64_REG(BPF_REG_9, BPF_REG_0),
-
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_LD_MAP_FD(BPF_REG_1, 0),
-		BPF_MOV64_IMM(BPF_REG_2, 8),
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-		BPF_EMIT_CALL(BPF_FUNC_ringbuf_reserve),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-		BPF_EXIT_INSN(),
-		BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-
-		BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_EMIT_CALL(BPF_FUNC_ringbuf_discard),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-
-		BPF_STX_MEM(BPF_DW, BPF_REG_9, BPF_REG_8, 0),
-		BPF_EXIT_INSN()
-	},
-	.fixup_map_array_48b = { 4 },
-	.fixup_map_ringbuf = { 11 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R8 !read_ok"
-},
-- 
cgit v1.2.3-70-g09d2


From 16a42573c253ad490a66cd0ca990f07297c0e221 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:26 +0300
Subject: selftests/bpf: verifier/regalloc converted to inline assembly

Test verifier/regalloc automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-17-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_regalloc.c        | 364 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/regalloc.c    | 277 ----------------
 3 files changed, 366 insertions(+), 277 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_regalloc.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/regalloc.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 75efbc20fb8e..990efde06624 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -46,6 +46,7 @@
 #include "verifier_raw_tp_writable.skel.h"
 #include "verifier_reg_equal.skel.h"
 #include "verifier_ref_tracking.skel.h"
+#include "verifier_regalloc.skel.h"
 #include "verifier_ringbuf.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
@@ -134,6 +135,7 @@ void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_reg_equal(void)            { RUN(verifier_reg_equal); }
 void test_verifier_ref_tracking(void)         { RUN(verifier_ref_tracking); }
+void test_verifier_regalloc(void)             { RUN(verifier_regalloc); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_regalloc.c b/tools/testing/selftests/bpf/progs/verifier_regalloc.c
new file mode 100644
index 000000000000..ee5ddea87c91
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_regalloc.c
@@ -0,0 +1,364 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/regalloc.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("tracepoint")
+__description("regalloc basic")
+__success __flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_basic(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r0 s> 20 goto l0_%=;				\
+	if r2 s< 0 goto l0_%=;				\
+	r7 += r0;					\
+	r7 += r2;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc negative")
+__failure __msg("invalid access to map value, value_size=48 off=48 size=1")
+__naked void regalloc_negative(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r0 s> 24 goto l0_%=;				\
+	if r2 s< 0 goto l0_%=;				\
+	r7 += r0;					\
+	r7 += r2;					\
+	r0 = *(u8*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc src_reg mark")
+__success __flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_src_reg_mark(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r0 s> 20 goto l0_%=;				\
+	r3 = 0;						\
+	if r3 s>= r2 goto l0_%=;			\
+	r7 += r0;					\
+	r7 += r2;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc src_reg negative")
+__failure __msg("invalid access to map value, value_size=48 off=44 size=8")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_src_reg_negative(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r0 s> 22 goto l0_%=;				\
+	r3 = 0;						\
+	if r3 s>= r2 goto l0_%=;			\
+	r7 += r0;					\
+	r7 += r2;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc and spill")
+__success __flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_and_spill(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r0 s> 20 goto l0_%=;				\
+	/* r0 has upper bound that should propagate into r2 */\
+	*(u64*)(r10 - 8) = r2;		/* spill r2 */	\
+	r0 = 0;						\
+	r2 = 0;				/* clear r0 and r2 */\
+	r3 = *(u64*)(r10 - 8);		/* fill r3 */	\
+	if r0 s>= r3 goto l0_%=;			\
+	/* r3 has lower and upper bounds */		\
+	r7 += r3;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc and spill negative")
+__failure __msg("invalid access to map value, value_size=48 off=48 size=8")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_and_spill_negative(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r0 s> 48 goto l0_%=;				\
+	/* r0 has upper bound that should propagate into r2 */\
+	*(u64*)(r10 - 8) = r2;		/* spill r2 */	\
+	r0 = 0;						\
+	r2 = 0;				/* clear r0 and r2 */\
+	r3 = *(u64*)(r10 - 8);		/* fill r3 */\
+	if r0 s>= r3 goto l0_%=;			\
+	/* r3 has lower and upper bounds */		\
+	r7 += r3;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc three regs")
+__success __flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_three_regs(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	r4 = r2;					\
+	if r0 s> 12 goto l0_%=;				\
+	if r2 s< 0 goto l0_%=;				\
+	r7 += r0;					\
+	r7 += r2;					\
+	r7 += r4;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc after call")
+__success __flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_after_call(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r8 = r0;					\
+	r9 = r0;					\
+	call regalloc_after_call__1;			\
+	if r8 s> 20 goto l0_%=;				\
+	if r9 s< 0 goto l0_%=;				\
+	r7 += r8;					\
+	r7 += r9;					\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void regalloc_after_call__1(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc in callee")
+__success __flag(BPF_F_ANY_ALIGNMENT)
+__naked void regalloc_in_callee(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r7 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r2 = r0;					\
+	r3 = r7;					\
+	call regalloc_in_callee__1;			\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void regalloc_in_callee__1(void)
+{
+	asm volatile ("					\
+	if r1 s> 20 goto l0_%=;				\
+	if r2 s< 0 goto l0_%=;				\
+	r3 += r1;					\
+	r3 += r2;					\
+	r0 = *(u64*)(r3 + 0);				\
+	exit;						\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("regalloc, spill, JEQ")
+__success
+__naked void regalloc_spill_jeq(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	*(u64*)(r10 - 8) = r0;		/* spill r0 */	\
+	if r0 == 0 goto l0_%=;				\
+l0_%=:	/* The verifier will walk the rest twice with r0 == 0 and r0 == map_value */\
+	call %[bpf_get_prandom_u32];			\
+	r2 = r0;					\
+	if r2 == 20 goto l1_%=;				\
+l1_%=:	/* The verifier will walk the rest two more times with r0 == 20 and r0 == unknown */\
+	r3 = *(u64*)(r10 - 8);		/* fill r3 with map_value */\
+	if r3 == 0 goto l2_%=;		/* skip ldx if map_value == NULL */\
+	/* Buggy verifier will think that r3 == 20 here */\
+	r0 = *(u64*)(r3 + 0);		/* read from map_value */\
+l2_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/regalloc.c b/tools/testing/selftests/bpf/verifier/regalloc.c
deleted file mode 100644
index bb0dd89dd212..000000000000
--- a/tools/testing/selftests/bpf/verifier/regalloc.c
+++ /dev/null
@@ -1,277 +0,0 @@
-{
-	"regalloc basic",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 20, 4),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_2, 0, 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc negative",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 24, 4),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_2, 0, 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_2),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=48 off=48 size=1",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"regalloc src_reg mark",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 20, 5),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_3, BPF_REG_2, 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc src_reg negative",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 22, 5),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_3, BPF_REG_2, 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=48 off=44 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc and spill",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 20, 7),
-	/* r0 has upper bound that should propagate into r2 */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -8), /* spill r2 */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 0), /* clear r0 and r2 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_10, -8), /* fill r3 */
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_0, BPF_REG_3, 2),
-	/* r3 has lower and upper bounds */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_3),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc and spill negative",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 48, 7),
-	/* r0 has upper bound that should propagate into r2 */
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -8), /* spill r2 */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_2, 0), /* clear r0 and r2 */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_10, -8), /* fill r3 */
-	BPF_JMP_REG(BPF_JSGE, BPF_REG_0, BPF_REG_3, 2),
-	/* r3 has lower and upper bounds */
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_3),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = REJECT,
-	.errstr = "invalid access to map value, value_size=48 off=48 size=8",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc three regs",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 10),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_0, 12, 5),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_2, 0, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_4),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc after call",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 10),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 6),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_8, 20, 4),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_9, 0, 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_8),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_7, BPF_REG_9),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc in callee",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_7),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_JMP_IMM(BPF_JSGT, BPF_REG_1, 20, 5),
-	BPF_JMP_IMM(BPF_JSLT, BPF_REG_2, 0, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_3, BPF_REG_2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_3, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"regalloc, spill, JEQ",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_0, -8), /* spill r0 */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 0),
-	/* The verifier will walk the rest twice with r0 == 0 and r0 == map_value */
-	BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_2, 20, 0),
-	/* The verifier will walk the rest two more times with r0 == 20 and r0 == unknown */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_10, -8), /* fill r3 with map_value */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_3, 0, 1), /* skip ldx if map_value == NULL */
-	/* Buggy verifier will think that r3 == 20 here */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_3, 0), /* read from map_value */
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 4 },
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-- 
cgit v1.2.3-70-g09d2


From 65222842ca04be5df505592f9f845acce0463f8b Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:27 +0300
Subject: selftests/bpf: verifier/runtime_jit converted to inline assembly

Test verifier/runtime_jit automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-18-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_runtime_jit.c     | 360 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/runtime_jit.c | 231 -------------
 3 files changed, 362 insertions(+), 231 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_runtime_jit.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/runtime_jit.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 990efde06624..2c1e07b59e8a 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -48,6 +48,7 @@
 #include "verifier_ref_tracking.skel.h"
 #include "verifier_regalloc.skel.h"
 #include "verifier_ringbuf.skel.h"
+#include "verifier_runtime_jit.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_uninit.skel.h"
@@ -137,6 +138,7 @@ void test_verifier_reg_equal(void)            { RUN(verifier_reg_equal); }
 void test_verifier_ref_tracking(void)         { RUN(verifier_ref_tracking); }
 void test_verifier_regalloc(void)             { RUN(verifier_regalloc); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
+void test_verifier_runtime_jit(void)          { RUN(verifier_runtime_jit); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_runtime_jit.c b/tools/testing/selftests/bpf/progs/verifier_runtime_jit.c
new file mode 100644
index 000000000000..27ebfc1fd9ee
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_runtime_jit.c
@@ -0,0 +1,360 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/runtime_jit.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+void dummy_prog_42_socket(void);
+void dummy_prog_24_socket(void);
+void dummy_prog_loop1_socket(void);
+void dummy_prog_loop2_socket(void);
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 4);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog1_socket SEC(".maps") = {
+	.values = {
+		[0] = (void *)&dummy_prog_42_socket,
+		[1] = (void *)&dummy_prog_loop1_socket,
+		[2] = (void *)&dummy_prog_24_socket,
+	},
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 8);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog2_socket SEC(".maps") = {
+	.values = {
+		[1] = (void *)&dummy_prog_loop2_socket,
+		[2] = (void *)&dummy_prog_24_socket,
+		[7] = (void *)&dummy_prog_42_socket,
+	},
+};
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_42_socket(void)
+{
+	asm volatile ("r0 = 42; exit;");
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_24_socket(void)
+{
+	asm volatile ("r0 = 24; exit;");
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_loop1_socket(void)
+{
+	asm volatile ("			\
+	r3 = 1;				\
+	r2 = %[map_prog1_socket] ll;	\
+	call %[bpf_tail_call];		\
+	r0 = 41;			\
+	exit;				\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_loop2_socket(void)
+{
+	asm volatile ("			\
+	r3 = 1;				\
+	r2 = %[map_prog2_socket] ll;	\
+	call %[bpf_tail_call];		\
+	r0 = 41;			\
+	exit;				\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog2_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, prog once")
+__success __success_unpriv __retval(42)
+__naked void call_within_bounds_prog_once(void)
+{
+	asm volatile ("					\
+	r3 = 0;						\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, prog loop")
+__success __success_unpriv __retval(41)
+__naked void call_within_bounds_prog_loop(void)
+{
+	asm volatile ("					\
+	r3 = 1;						\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, no prog")
+__success __success_unpriv __retval(1)
+__naked void call_within_bounds_no_prog(void)
+{
+	asm volatile ("					\
+	r3 = 3;						\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, key 2")
+__success __success_unpriv __retval(24)
+__naked void call_within_bounds_key_2(void)
+{
+	asm volatile ("					\
+	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, key 2 / key 2, first branch")
+__success __success_unpriv __retval(24)
+__naked void _2_key_2_first_branch(void)
+{
+	asm volatile ("					\
+	r0 = 13;					\
+	*(u8*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r0 = *(u8*)(r1 + %[__sk_buff_cb_0]);		\
+	if r0 == 13 goto l0_%=;				\
+	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+l1_%=:	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, key 2 / key 2, second branch")
+__success __success_unpriv __retval(24)
+__naked void _2_key_2_second_branch(void)
+{
+	asm volatile ("					\
+	r0 = 14;					\
+	*(u8*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r0 = *(u8*)(r1 + %[__sk_buff_cb_0]);		\
+	if r0 == 13 goto l0_%=;				\
+	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+l1_%=:	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, key 0 / key 2, first branch")
+__success __success_unpriv __retval(24)
+__naked void _0_key_2_first_branch(void)
+{
+	asm volatile ("					\
+	r0 = 13;					\
+	*(u8*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r0 = *(u8*)(r1 + %[__sk_buff_cb_0]);		\
+	if r0 == 13 goto l0_%=;				\
+	r3 = 0;						\
+	r2 = %[map_prog1_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+l1_%=:	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, key 0 / key 2, second branch")
+__success __success_unpriv __retval(42)
+__naked void _0_key_2_second_branch(void)
+{
+	asm volatile ("					\
+	r0 = 14;					\
+	*(u8*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r0 = *(u8*)(r1 + %[__sk_buff_cb_0]);		\
+	if r0 == 13 goto l0_%=;				\
+	r3 = 0;						\
+	r2 = %[map_prog1_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r3 = 2;						\
+	r2 = %[map_prog1_socket] ll;			\
+l1_%=:	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, different maps, first branch")
+__success __failure_unpriv __msg_unpriv("tail_call abusing map_ptr")
+__retval(1)
+__naked void bounds_different_maps_first_branch(void)
+{
+	asm volatile ("					\
+	r0 = 13;					\
+	*(u8*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r0 = *(u8*)(r1 + %[__sk_buff_cb_0]);		\
+	if r0 == 13 goto l0_%=;				\
+	r3 = 0;						\
+	r2 = %[map_prog1_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r3 = 0;						\
+	r2 = %[map_prog2_socket] ll;			\
+l1_%=:	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_addr(map_prog2_socket),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call within bounds, different maps, second branch")
+__success __failure_unpriv __msg_unpriv("tail_call abusing map_ptr")
+__retval(42)
+__naked void bounds_different_maps_second_branch(void)
+{
+	asm volatile ("					\
+	r0 = 14;					\
+	*(u8*)(r1 + %[__sk_buff_cb_0]) = r0;		\
+	r0 = *(u8*)(r1 + %[__sk_buff_cb_0]);		\
+	if r0 == 13 goto l0_%=;				\
+	r3 = 0;						\
+	r2 = %[map_prog1_socket] ll;			\
+	goto l1_%=;					\
+l0_%=:	r3 = 0;						\
+	r2 = %[map_prog2_socket] ll;			\
+l1_%=:	call %[bpf_tail_call];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket),
+	  __imm_addr(map_prog2_socket),
+	  __imm_const(__sk_buff_cb_0, offsetof(struct __sk_buff, cb[0]))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: tail_call out of bounds")
+__success __success_unpriv __retval(2)
+__naked void tail_call_out_of_bounds(void)
+{
+	asm volatile ("					\
+	r3 = 256;					\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 2;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: pass negative index to tail_call")
+__success __success_unpriv __retval(2)
+__naked void negative_index_to_tail_call(void)
+{
+	asm volatile ("					\
+	r3 = -1;					\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 2;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("runtime/jit: pass > 32bit index to tail_call")
+__success __success_unpriv __retval(42)
+/* Verifier rewrite for unpriv skips tail call here. */
+__retval_unpriv(2)
+__naked void _32bit_index_to_tail_call(void)
+{
+	asm volatile ("					\
+	r3 = 0x100000000 ll;				\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 2;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/runtime_jit.c b/tools/testing/selftests/bpf/verifier/runtime_jit.c
deleted file mode 100644
index 94c399d1faca..000000000000
--- a/tools/testing/selftests/bpf/verifier/runtime_jit.c
+++ /dev/null
@@ -1,231 +0,0 @@
-{
-	"runtime/jit: tail_call within bounds, prog once",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"runtime/jit: tail_call within bounds, prog loop",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.result = ACCEPT,
-	.retval = 41,
-},
-{
-	"runtime/jit: tail_call within bounds, no prog",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, 3),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"runtime/jit: tail_call within bounds, key 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.result = ACCEPT,
-	.retval = 24,
-},
-{
-	"runtime/jit: tail_call within bounds, key 2 / key 2, first branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 13),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 13, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5, 9 },
-	.result = ACCEPT,
-	.retval = 24,
-},
-{
-	"runtime/jit: tail_call within bounds, key 2 / key 2, second branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 14),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 13, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5, 9 },
-	.result = ACCEPT,
-	.retval = 24,
-},
-{
-	"runtime/jit: tail_call within bounds, key 0 / key 2, first branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 13),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 13, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5, 9 },
-	.result = ACCEPT,
-	.retval = 24,
-},
-{
-	"runtime/jit: tail_call within bounds, key 0 / key 2, second branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 14),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 13, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 2),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5, 9 },
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"runtime/jit: tail_call within bounds, different maps, first branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 13),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 13, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5 },
-	.fixup_prog2 = { 9 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "tail_call abusing map_ptr",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"runtime/jit: tail_call within bounds, different maps, second branch",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 14),
-	BPF_STX_MEM(BPF_B, BPF_REG_1, BPF_REG_0,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, cb[0])),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 13, 4),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 5 },
-	.fixup_prog2 = { 9 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "tail_call abusing map_ptr",
-	.result = ACCEPT,
-	.retval = 42,
-},
-{
-	"runtime/jit: tail_call out of bounds",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, 256),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 2),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.result = ACCEPT,
-	.retval = 2,
-},
-{
-	"runtime/jit: pass negative index to tail_call",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, -1),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 2),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.result = ACCEPT,
-	.retval = 2,
-},
-{
-	"runtime/jit: pass > 32bit index to tail_call",
-	.insns = {
-	BPF_LD_IMM64(BPF_REG_3, 0x100000000ULL),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 2),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 2 },
-	.result = ACCEPT,
-	.retval = 42,
-	/* Verifier rewrite for unpriv skips tail call here. */
-	.retval_unpriv = 2,
-},
-- 
cgit v1.2.3-70-g09d2


From 034d9ad25db3569bee306850c6a276bca0d1e12d Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:28 +0300
Subject: selftests/bpf: verifier/search_pruning converted to inline assembly

Test verifier/search_pruning automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-19-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_search_pruning.c  | 339 +++++++++++++++++++++
 .../selftests/bpf/verifier/search_pruning.c        | 266 ----------------
 3 files changed, 341 insertions(+), 266 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_search_pruning.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/search_pruning.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 2c1e07b59e8a..fd4a6e7b41de 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -49,6 +49,7 @@
 #include "verifier_regalloc.skel.h"
 #include "verifier_ringbuf.skel.h"
 #include "verifier_runtime_jit.skel.h"
+#include "verifier_search_pruning.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_uninit.skel.h"
@@ -139,6 +140,7 @@ void test_verifier_ref_tracking(void)         { RUN(verifier_ref_tracking); }
 void test_verifier_regalloc(void)             { RUN(verifier_regalloc); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_runtime_jit(void)          { RUN(verifier_runtime_jit); }
+void test_verifier_search_pruning(void)       { RUN(verifier_search_pruning); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_search_pruning.c b/tools/testing/selftests/bpf/progs/verifier_search_pruning.c
new file mode 100644
index 000000000000..5a14498d352f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_search_pruning.c
@@ -0,0 +1,339 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/search_pruning.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+SEC("socket")
+__description("pointer/scalar confusion in state equality check (way 1)")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr as return value")
+__retval(POINTER_VALUE)
+__naked void state_equality_check_way_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 = *(u64*)(r0 + 0);				\
+	goto l1_%=;					\
+l0_%=:	r0 = r10;					\
+l1_%=:	goto l2_%=;					\
+l2_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("pointer/scalar confusion in state equality check (way 2)")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr as return value")
+__retval(POINTER_VALUE)
+__naked void state_equality_check_way_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	r0 = r10;					\
+	goto l1_%=;					\
+l0_%=:	r0 = *(u64*)(r0 + 0);				\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("lwt_in")
+__description("liveness pruning and write screening")
+__failure __msg("R0 !read_ok")
+__naked void liveness_pruning_and_write_screening(void)
+{
+	asm volatile ("					\
+	/* Get an unknown value */			\
+	r2 = *(u32*)(r1 + 0);				\
+	/* branch conditions teach us nothing about R2 */\
+	if r2 >= 0 goto l0_%=;				\
+	r0 = 0;						\
+l0_%=:	if r2 >= 0 goto l1_%=;				\
+	r0 = 0;						\
+l1_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("varlen_map_value_access pruning")
+__failure __msg("R0 unbounded memory access")
+__failure_unpriv __msg_unpriv("R0 leaks addr")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void varlen_map_value_access_pruning(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r0 + 0);				\
+	w2 = %[max_entries];				\
+	if r2 s> r1 goto l1_%=;				\
+	w1 = 0;						\
+l1_%=:	w1 <<= 2;					\
+	r0 += r1;					\
+	goto l2_%=;					\
+l2_%=:	r1 = %[test_val_foo];				\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(max_entries, MAX_ENTRIES),
+	  __imm_const(test_val_foo, offsetof(struct test_val, foo))
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("search pruning: all branches should be verified (nop operation)")
+__failure __msg("R6 invalid mem access 'scalar'")
+__naked void should_be_verified_nop_operation(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r3 = *(u64*)(r0 + 0);				\
+	if r3 == 0xbeef goto l1_%=;			\
+	r4 = 0;						\
+	goto l2_%=;					\
+l1_%=:	r4 = 1;						\
+l2_%=:	*(u64*)(r10 - 16) = r4;				\
+	call %[bpf_ktime_get_ns];			\
+	r5 = *(u64*)(r10 - 16);				\
+	if r5 == 0 goto l0_%=;				\
+	r6 = 0;						\
+	r1 = 0xdead;					\
+	*(u64*)(r6 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("search pruning: all branches should be verified (invalid stack access)")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid read from stack off -16+0 size 8")
+__retval(0)
+__naked void be_verified_invalid_stack_access(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r3 = *(u64*)(r0 + 0);				\
+	r4 = 0;						\
+	if r3 == 0xbeef goto l1_%=;			\
+	*(u64*)(r10 - 16) = r4;				\
+	goto l2_%=;					\
+l1_%=:	*(u64*)(r10 - 24) = r4;				\
+l2_%=:	call %[bpf_ktime_get_ns];			\
+	r5 = *(u64*)(r10 - 16);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("precision tracking for u32 spill/fill")
+__failure __msg("R0 min value is outside of the allowed memory range")
+__naked void tracking_for_u32_spill_fill(void)
+{
+	asm volatile ("					\
+	r7 = r1;					\
+	call %[bpf_get_prandom_u32];			\
+	w6 = 32;					\
+	if r0 == 0 goto l0_%=;				\
+	w6 = 4;						\
+l0_%=:	/* Additional insns to introduce a pruning point. */\
+	call %[bpf_get_prandom_u32];			\
+	r3 = 0;						\
+	r3 = 0;						\
+	if r0 == 0 goto l1_%=;				\
+	r3 = 0;						\
+l1_%=:	/* u32 spill/fill */				\
+	*(u32*)(r10 - 8) = r6;				\
+	r8 = *(u32*)(r10 - 8);				\
+	/* out-of-bound map value access for r6=32 */	\
+	r1 = 0;						\
+	*(u64*)(r10 - 16) = r1;				\
+	r2 = r10;					\
+	r2 += -16;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r0 += r8;					\
+	r1 = *(u32*)(r0 + 0);				\
+l2_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("precision tracking for u32 spills, u64 fill")
+__failure __msg("div by zero")
+__naked void for_u32_spills_u64_fill(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r6 = r0;					\
+	w7 = 0xffffffff;				\
+	/* Additional insns to introduce a pruning point. */\
+	r3 = 1;						\
+	r3 = 1;						\
+	r3 = 1;						\
+	r3 = 1;						\
+	call %[bpf_get_prandom_u32];			\
+	if r0 == 0 goto l0_%=;				\
+	r3 = 1;						\
+l0_%=:	w3 /= 0;					\
+	/* u32 spills, u64 fill */			\
+	*(u32*)(r10 - 4) = r6;				\
+	*(u32*)(r10 - 8) = r7;				\
+	r8 = *(u64*)(r10 - 8);				\
+	/* if r8 != X goto pc+1  r8 known in fallthrough branch */\
+	if r8 != 0xffffffff goto l1_%=;			\
+	r3 = 1;						\
+l1_%=:	/* if r8 == X goto pc+1  condition always true on first\
+	 * traversal, so starts backtracking to mark r8 as requiring\
+	 * precision. r7 marked as needing precision. r6 not marked\
+	 * since it's not tracked.			\
+	 */						\
+	if r8 == 0xffffffff goto l2_%=;			\
+	/* fails if r8 correctly marked unknown after fill. */\
+	w3 /= 0;					\
+l2_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("allocated_stack")
+__success __msg("processed 15 insns")
+__success_unpriv __msg_unpriv("") __log_level(1) __retval(0)
+__naked void allocated_stack(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	call %[bpf_get_prandom_u32];			\
+	r7 = r0;					\
+	if r0 == 0 goto l0_%=;				\
+	r0 = 0;						\
+	*(u64*)(r10 - 8) = r6;				\
+	r6 = *(u64*)(r10 - 8);				\
+	*(u8*)(r10 - 9) = r7;				\
+	r7 = *(u8*)(r10 - 9);				\
+l0_%=:	if r0 != 0 goto l1_%=;				\
+l1_%=:	if r0 != 0 goto l2_%=;				\
+l2_%=:	if r0 != 0 goto l3_%=;				\
+l3_%=:	if r0 != 0 goto l4_%=;				\
+l4_%=:	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+/* The test performs a conditional 64-bit write to a stack location
+ * fp[-8], this is followed by an unconditional 8-bit write to fp[-8],
+ * then data is read from fp[-8]. This sequence is unsafe.
+ *
+ * The test would be mistakenly marked as safe w/o dst register parent
+ * preservation in verifier.c:copy_register_state() function.
+ *
+ * Note the usage of BPF_F_TEST_STATE_FREQ to force creation of the
+ * checkpoint state after conditional 64-bit assignment.
+ */
+
+SEC("socket")
+__description("write tracking and register parent chain bug")
+/* in privileged mode reads from uninitialized stack locations are permitted */
+__success __failure_unpriv
+__msg_unpriv("invalid read from stack off -8+1 size 8")
+__retval(0) __flag(BPF_F_TEST_STATE_FREQ)
+__naked void and_register_parent_chain_bug(void)
+{
+	asm volatile ("					\
+	/* r6 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r6 = r0;					\
+	/* r0 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	/* if r0 > r6 goto +1 */			\
+	if r0 > r6 goto l0_%=;				\
+	/* *(u64 *)(r10 - 8) = 0xdeadbeef */		\
+	r0 = 0xdeadbeef;				\
+	*(u64*)(r10 - 8) = r0;				\
+l0_%=:	r1 = 42;					\
+	*(u8*)(r10 - 8) = r1;				\
+	r2 = *(u64*)(r10 - 8);				\
+	/* exit(0) */					\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/search_pruning.c b/tools/testing/selftests/bpf/verifier/search_pruning.c
deleted file mode 100644
index 745d6b5842fd..000000000000
--- a/tools/testing/selftests/bpf/verifier/search_pruning.c
+++ /dev/null
@@ -1,266 +0,0 @@
-{
-	"pointer/scalar confusion in state equality check (way 1)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0),
-	BPF_JMP_A(1),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_10),
-	BPF_JMP_A(0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.retval = POINTER_VALUE,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 leaks addr as return value"
-},
-{
-	"pointer/scalar confusion in state equality check (way 2)",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_10),
-	BPF_JMP_A(1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.result = ACCEPT,
-	.retval = POINTER_VALUE,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 leaks addr as return value"
-},
-{
-	"liveness pruning and write screening",
-	.insns = {
-	/* Get an unknown value */
-	BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0),
-	/* branch conditions teach us nothing about R2 */
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_2, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JGE, BPF_REG_2, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "R0 !read_ok",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_LWT_IN,
-},
-{
-	"varlen_map_value_access pruning",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV32_IMM(BPF_REG_2, MAX_ENTRIES),
-	BPF_JMP_REG(BPF_JSGT, BPF_REG_2, BPF_REG_1, 1),
-	BPF_MOV32_IMM(BPF_REG_1, 0),
-	BPF_ALU32_IMM(BPF_LSH, BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, offsetof(struct test_val, foo)),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.errstr = "R0 unbounded memory access",
-	.result_unpriv = REJECT,
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"search pruning: all branches should be verified (nop operation)",
-	.insns = {
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-		BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-		BPF_LD_MAP_FD(BPF_REG_1, 0),
-		BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-		BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_0, 0),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_3, 0xbeef, 2),
-		BPF_MOV64_IMM(BPF_REG_4, 0),
-		BPF_JMP_A(1),
-		BPF_MOV64_IMM(BPF_REG_4, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_4, -16),
-		BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-		BPF_LDX_MEM(BPF_DW, BPF_REG_5, BPF_REG_10, -16),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_5, 0, 2),
-		BPF_MOV64_IMM(BPF_REG_6, 0),
-		BPF_ST_MEM(BPF_DW, BPF_REG_6, 0, 0xdead),
-		BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr = "R6 invalid mem access 'scalar'",
-	.result = REJECT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"search pruning: all branches should be verified (invalid stack access)",
-	.insns = {
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-		BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-		BPF_LD_MAP_FD(BPF_REG_1, 0),
-		BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-		BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_0, 0),
-		BPF_MOV64_IMM(BPF_REG_4, 0),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_3, 0xbeef, 2),
-		BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_4, -16),
-		BPF_JMP_A(1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_4, -24),
-		BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-		BPF_LDX_MEM(BPF_DW, BPF_REG_5, BPF_REG_10, -16),
-		BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr_unpriv = "invalid read from stack off -16+0 size 8",
-	.result_unpriv = REJECT,
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-{
-	"precision tracking for u32 spill/fill",
-	.insns = {
-		BPF_MOV64_REG(BPF_REG_7, BPF_REG_1),
-		BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-		BPF_MOV32_IMM(BPF_REG_6, 32),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-		BPF_MOV32_IMM(BPF_REG_6, 4),
-		/* Additional insns to introduce a pruning point. */
-		BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-		BPF_MOV64_IMM(BPF_REG_3, 0),
-		/* u32 spill/fill */
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_6, -8),
-		BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_10, -8),
-		/* out-of-bound map value access for r6=32 */
-		BPF_ST_MEM(BPF_DW, BPF_REG_10, -16, 0),
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -16),
-		BPF_LD_MAP_FD(BPF_REG_1, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-		BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_8),
-		BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 15 },
-	.result = REJECT,
-	.errstr = "R0 min value is outside of the allowed memory range",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"precision tracking for u32 spills, u64 fill",
-	.insns = {
-		BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-		BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-		BPF_MOV32_IMM(BPF_REG_7, 0xffffffff),
-		/* Additional insns to introduce a pruning point. */
-		BPF_MOV64_IMM(BPF_REG_3, 1),
-		BPF_MOV64_IMM(BPF_REG_3, 1),
-		BPF_MOV64_IMM(BPF_REG_3, 1),
-		BPF_MOV64_IMM(BPF_REG_3, 1),
-		BPF_EMIT_CALL(BPF_FUNC_get_prandom_u32),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-		BPF_MOV64_IMM(BPF_REG_3, 1),
-		BPF_ALU32_IMM(BPF_DIV, BPF_REG_3, 0),
-		/* u32 spills, u64 fill */
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_6, -4),
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_7, -8),
-		BPF_LDX_MEM(BPF_DW, BPF_REG_8, BPF_REG_10, -8),
-		/* if r8 != X goto pc+1  r8 known in fallthrough branch */
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_8, 0xffffffff, 1),
-		BPF_MOV64_IMM(BPF_REG_3, 1),
-		/* if r8 == X goto pc+1  condition always true on first
-		 * traversal, so starts backtracking to mark r8 as requiring
-		 * precision. r7 marked as needing precision. r6 not marked
-		 * since it's not tracked.
-		 */
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_8, 0xffffffff, 1),
-		/* fails if r8 correctly marked unknown after fill. */
-		BPF_ALU32_IMM(BPF_DIV, BPF_REG_3, 0),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "div by zero",
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"allocated_stack",
-	.insns = {
-		BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_1),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-		BPF_ALU64_REG(BPF_MOV, BPF_REG_7, BPF_REG_0),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-		BPF_MOV64_IMM(BPF_REG_0, 0),
-		BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_6, -8),
-		BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_10, -8),
-		BPF_STX_MEM(BPF_B, BPF_REG_10, BPF_REG_7, -9),
-		BPF_LDX_MEM(BPF_B, BPF_REG_7, BPF_REG_10, -9),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 0),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 0),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 0),
-		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 0),
-		BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = ACCEPT,
-	.insn_processed = 15,
-},
-/* The test performs a conditional 64-bit write to a stack location
- * fp[-8], this is followed by an unconditional 8-bit write to fp[-8],
- * then data is read from fp[-8]. This sequence is unsafe.
- *
- * The test would be mistakenly marked as safe w/o dst register parent
- * preservation in verifier.c:copy_register_state() function.
- *
- * Note the usage of BPF_F_TEST_STATE_FREQ to force creation of the
- * checkpoint state after conditional 64-bit assignment.
- */
-{
-	"write tracking and register parent chain bug",
-	.insns = {
-	/* r6 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* r0 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	/* if r0 > r6 goto +1 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_6, 1),
-	/* *(u64 *)(r10 - 8) = 0xdeadbeef */
-	BPF_ST_MEM(BPF_DW, BPF_REG_FP, -8, 0xdeadbeef),
-	/* r1 = 42 */
-	BPF_MOV64_IMM(BPF_REG_1, 42),
-	/* *(u8 *)(r10 - 8) = r1 */
-	BPF_STX_MEM(BPF_B, BPF_REG_FP, BPF_REG_1, -8),
-	/* r2 = *(u64 *)(r10 - 8) */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_FP, -8),
-	/* exit(0) */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.flags = BPF_F_TEST_STATE_FREQ,
-	.errstr_unpriv = "invalid read from stack off -8+1 size 8",
-	.result_unpriv = REJECT,
-	/* in privileged mode reads from uninitialized stack locations are permitted */
-	.result = ACCEPT,
-},
-- 
cgit v1.2.3-70-g09d2


From 426fc0e3fce272c92e804b2d844a41641f9b888e Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:29 +0300
Subject: selftests/bpf: verifier/sock converted to inline assembly

Test verifier/sock automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-20-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c |   2 +
 tools/testing/selftests/bpf/progs/verifier_sock.c | 980 ++++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/sock.c       | 706 ----------------
 3 files changed, 982 insertions(+), 706 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_sock.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/sock.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index fd4a6e7b41de..906f1c532cb9 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -50,6 +50,7 @@
 #include "verifier_ringbuf.skel.h"
 #include "verifier_runtime_jit.skel.h"
 #include "verifier_search_pruning.skel.h"
+#include "verifier_sock.skel.h"
 #include "verifier_spill_fill.skel.h"
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_uninit.skel.h"
@@ -141,6 +142,7 @@ void test_verifier_regalloc(void)             { RUN(verifier_regalloc); }
 void test_verifier_ringbuf(void)              { RUN(verifier_ringbuf); }
 void test_verifier_runtime_jit(void)          { RUN(verifier_runtime_jit); }
 void test_verifier_search_pruning(void)       { RUN(verifier_search_pruning); }
+void test_verifier_sock(void)                 { RUN(verifier_sock); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_sock.c b/tools/testing/selftests/bpf/progs/verifier_sock.c
new file mode 100644
index 000000000000..ee76b51005ab
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_sock.c
@@ -0,0 +1,980 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/sock.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define sizeof_field(TYPE, MEMBER) sizeof((((TYPE *)0)->MEMBER))
+#define offsetofend(TYPE, MEMBER) \
+	(offsetof(TYPE, MEMBER)	+ sizeof_field(TYPE, MEMBER))
+
+struct {
+	__uint(type, BPF_MAP_TYPE_REUSEPORT_SOCKARRAY);
+	__uint(max_entries, 1);
+	__type(key, __u32);
+	__type(value, __u64);
+} map_reuseport_array SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_SOCKHASH);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+} map_sockhash SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_SOCKMAP);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+} map_sockmap SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_XSKMAP);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, int);
+} map_xskmap SEC(".maps");
+
+struct val {
+	int cnt;
+	struct bpf_spin_lock l;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
+	__uint(max_entries, 0);
+	__type(key, int);
+	__type(value, struct val);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+} sk_storage_map SEC(".maps");
+
+SEC("cgroup/skb")
+__description("skb->sk: no NULL check")
+__failure __msg("invalid mem access 'sock_common_or_null'")
+__failure_unpriv
+__naked void skb_sk_no_null_check(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	r0 = *(u32*)(r1 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("skb->sk: sk->family [non fullsock field]")
+__success __success_unpriv __retval(0)
+__naked void sk_family_non_fullsock_field_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	r0 = *(u32*)(r1 + %[bpf_sock_family]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_family, offsetof(struct bpf_sock, family))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("skb->sk: sk->type [fullsock field]")
+__failure __msg("invalid sock_common access")
+__failure_unpriv
+__naked void sk_sk_type_fullsock_field_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	r0 = *(u32*)(r1 + %[bpf_sock_type]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_sk_fullsock(skb->sk): no !skb->sk check")
+__failure __msg("type=sock_common_or_null expected=sock_common")
+__failure_unpriv
+__naked void sk_no_skb_sk_check_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	call %[bpf_sk_fullsock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): no NULL check on ret")
+__failure __msg("invalid mem access 'sock_or_null'")
+__failure_unpriv
+__naked void no_null_check_on_ret_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	r0 = *(u32*)(r0 + %[bpf_sock_type]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->type [fullsock field]")
+__success __success_unpriv __retval(0)
+__naked void sk_sk_type_fullsock_field_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r0 + %[bpf_sock_type]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->family [non fullsock field]")
+__success __success_unpriv __retval(0)
+__naked void sk_family_non_fullsock_field_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r0 + %[bpf_sock_family]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_family, offsetof(struct bpf_sock, family))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->state [narrow load]")
+__success __success_unpriv __retval(0)
+__naked void sk_sk_state_narrow_load(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u8*)(r0 + %[bpf_sock_state]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_state, offsetof(struct bpf_sock, state))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->dst_port [word load] (backward compatibility)")
+__success __success_unpriv __retval(0)
+__naked void port_word_load_backward_compatibility(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r0 + %[bpf_sock_dst_port]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_dst_port, offsetof(struct bpf_sock, dst_port))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->dst_port [half load]")
+__success __success_unpriv __retval(0)
+__naked void sk_dst_port_half_load(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u16*)(r0 + %[bpf_sock_dst_port]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_dst_port, offsetof(struct bpf_sock, dst_port))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->dst_port [half load] (invalid)")
+__failure __msg("invalid sock access")
+__failure_unpriv
+__naked void dst_port_half_load_invalid_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u16*)(r0 + %[__imm_0]);			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__imm_0, offsetof(struct bpf_sock, dst_port) + 2),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->dst_port [byte load]")
+__success __success_unpriv __retval(0)
+__naked void sk_dst_port_byte_load(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r2 = *(u8*)(r0 + %[bpf_sock_dst_port]);		\
+	r2 = *(u8*)(r0 + %[__imm_0]);			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__imm_0, offsetof(struct bpf_sock, dst_port) + 1),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_dst_port, offsetof(struct bpf_sock, dst_port))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->dst_port [byte load] (invalid)")
+__failure __msg("invalid sock access")
+__failure_unpriv
+__naked void dst_port_byte_load_invalid(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u8*)(r0 + %[__imm_0]);			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__imm_0, offsetof(struct bpf_sock, dst_port) + 2),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): past sk->dst_port [half load] (invalid)")
+__failure __msg("invalid sock access")
+__failure_unpriv
+__naked void dst_port_half_load_invalid_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u16*)(r0 + %[bpf_sock_dst_port__end]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_dst_port__end, offsetofend(struct bpf_sock, dst_port))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->dst_ip6 [load 2nd byte]")
+__success __success_unpriv __retval(0)
+__naked void dst_ip6_load_2nd_byte(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u8*)(r0 + %[__imm_0]);			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__imm_0, offsetof(struct bpf_sock, dst_ip6[0]) + 1),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->type [narrow load]")
+__success __success_unpriv __retval(0)
+__naked void sk_sk_type_narrow_load(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u8*)(r0 + %[bpf_sock_type]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): sk->protocol [narrow load]")
+__success __success_unpriv __retval(0)
+__naked void sk_sk_protocol_narrow_load(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u8*)(r0 + %[bpf_sock_protocol]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_protocol, offsetof(struct bpf_sock, protocol))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("sk_fullsock(skb->sk): beyond last field")
+__failure __msg("invalid sock access")
+__failure_unpriv
+__naked void skb_sk_beyond_last_field_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r0 + %[bpf_sock_rx_queue_mapping__end]);\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_sock_rx_queue_mapping__end, offsetofend(struct bpf_sock, rx_queue_mapping))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_tcp_sock(skb->sk): no !skb->sk check")
+__failure __msg("type=sock_common_or_null expected=sock_common")
+__failure_unpriv
+__naked void sk_no_skb_sk_check_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	call %[bpf_tcp_sock];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_tcp_sock(skb->sk): no NULL check on ret")
+__failure __msg("invalid mem access 'tcp_sock_or_null'")
+__failure_unpriv
+__naked void no_null_check_on_ret_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_tcp_sock];				\
+	r0 = *(u32*)(r0 + %[bpf_tcp_sock_snd_cwnd]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_tcp_sock_snd_cwnd, offsetof(struct bpf_tcp_sock, snd_cwnd))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_tcp_sock(skb->sk): tp->snd_cwnd")
+__success __success_unpriv __retval(0)
+__naked void skb_sk_tp_snd_cwnd_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r0 = *(u32*)(r0 + %[bpf_tcp_sock_snd_cwnd]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_tcp_sock_snd_cwnd, offsetof(struct bpf_tcp_sock, snd_cwnd))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_tcp_sock(skb->sk): tp->bytes_acked")
+__success __success_unpriv __retval(0)
+__naked void skb_sk_tp_bytes_acked(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r0 = *(u64*)(r0 + %[bpf_tcp_sock_bytes_acked]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_tcp_sock_bytes_acked, offsetof(struct bpf_tcp_sock, bytes_acked))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_tcp_sock(skb->sk): beyond last field")
+__failure __msg("invalid tcp_sock access")
+__failure_unpriv
+__naked void skb_sk_beyond_last_field_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r0 = *(u64*)(r0 + %[bpf_tcp_sock_bytes_acked__end]);\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_tcp_sock_bytes_acked__end, offsetofend(struct bpf_tcp_sock, bytes_acked))
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("bpf_tcp_sock(bpf_sk_fullsock(skb->sk)): tp->snd_cwnd")
+__success __success_unpriv __retval(0)
+__naked void skb_sk_tp_snd_cwnd_2(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l2_%=;				\
+	exit;						\
+l2_%=:	r0 = *(u32*)(r0 + %[bpf_tcp_sock_snd_cwnd]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk)),
+	  __imm_const(bpf_tcp_sock_snd_cwnd, offsetof(struct bpf_tcp_sock, snd_cwnd))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("bpf_sk_release(skb->sk)")
+__failure __msg("R1 must be referenced when passed to release function")
+__naked void bpf_sk_release_skb_sk(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 == 0 goto l0_%=;				\
+	call %[bpf_sk_release];				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_release),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("bpf_sk_release(bpf_sk_fullsock(skb->sk))")
+__failure __msg("R1 must be referenced when passed to release function")
+__naked void bpf_sk_fullsock_skb_sk(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_sk_release];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_release),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("bpf_sk_release(bpf_tcp_sock(skb->sk))")
+__failure __msg("R1 must be referenced when passed to release function")
+__naked void bpf_tcp_sock_skb_sk(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_tcp_sock];				\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r1 = r0;					\
+	call %[bpf_sk_release];				\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_release),
+	  __imm(bpf_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("sk_storage_get(map, skb->sk, NULL, 0): value == NULL")
+__success __retval(0)
+__naked void sk_null_0_value_null(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r4 = 0;						\
+	r3 = 0;						\
+	r2 = r0;					\
+	r1 = %[sk_storage_map] ll;			\
+	call %[bpf_sk_storage_get];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_storage_get),
+	  __imm_addr(sk_storage_map),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("sk_storage_get(map, skb->sk, 1, 1): value == 1")
+__failure __msg("R3 type=scalar expected=fp")
+__naked void sk_1_1_value_1(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r4 = 1;						\
+	r3 = 1;						\
+	r2 = r0;					\
+	r1 = %[sk_storage_map] ll;			\
+	call %[bpf_sk_storage_get];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_storage_get),
+	  __imm_addr(sk_storage_map),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("sk_storage_get(map, skb->sk, &stack_value, 1): stack_value")
+__success __retval(0)
+__naked void stack_value_1_stack_value(void)
+{
+	asm volatile ("					\
+	r2 = 0;						\
+	*(u64*)(r10 - 8) = r2;				\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	call %[bpf_sk_fullsock];			\
+	if r0 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r4 = 1;						\
+	r3 = r10;					\
+	r3 += -8;					\
+	r2 = r0;					\
+	r1 = %[sk_storage_map] ll;			\
+	call %[bpf_sk_storage_get];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_fullsock),
+	  __imm(bpf_sk_storage_get),
+	  __imm_addr(sk_storage_map),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("bpf_map_lookup_elem(smap, &key)")
+__failure __msg("cannot pass map_type 24 into func bpf_map_lookup_elem")
+__naked void map_lookup_elem_smap_key(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[sk_storage_map] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(sk_storage_map)
+	: __clobber_all);
+}
+
+SEC("xdp")
+__description("bpf_map_lookup_elem(xskmap, &key); xs->queue_id")
+__success __retval(0)
+__naked void xskmap_key_xs_queue_id(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_xskmap] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r0 = *(u32*)(r0 + %[bpf_xdp_sock_queue_id]);	\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_xskmap),
+	  __imm_const(bpf_xdp_sock_queue_id, offsetof(struct bpf_xdp_sock, queue_id))
+	: __clobber_all);
+}
+
+SEC("sk_skb")
+__description("bpf_map_lookup_elem(sockmap, &key)")
+__failure __msg("Unreleased reference id=2 alloc_insn=6")
+__naked void map_lookup_elem_sockmap_key(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_sockmap] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_sockmap)
+	: __clobber_all);
+}
+
+SEC("sk_skb")
+__description("bpf_map_lookup_elem(sockhash, &key)")
+__failure __msg("Unreleased reference id=2 alloc_insn=6")
+__naked void map_lookup_elem_sockhash_key(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_sockhash] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_sockhash)
+	: __clobber_all);
+}
+
+SEC("sk_skb")
+__description("bpf_map_lookup_elem(sockmap, &key); sk->type [fullsock field]; bpf_sk_release(sk)")
+__success
+__naked void field_bpf_sk_release_sk_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_sockmap] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = r0;					\
+	r0 = *(u32*)(r0 + %[bpf_sock_type]);		\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_sk_release),
+	  __imm_addr(map_sockmap),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+SEC("sk_skb")
+__description("bpf_map_lookup_elem(sockhash, &key); sk->type [fullsock field]; bpf_sk_release(sk)")
+__success
+__naked void field_bpf_sk_release_sk_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_sockhash] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = r0;					\
+	r0 = *(u32*)(r0 + %[bpf_sock_type]);		\
+	call %[bpf_sk_release];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_sk_release),
+	  __imm_addr(map_sockhash),
+	  __imm_const(bpf_sock_type, offsetof(struct bpf_sock, type))
+	: __clobber_all);
+}
+
+SEC("sk_reuseport")
+__description("bpf_sk_select_reuseport(ctx, reuseport_array, &key, flags)")
+__success
+__naked void ctx_reuseport_array_key_flags(void)
+{
+	asm volatile ("					\
+	r4 = 0;						\
+	r2 = 0;						\
+	*(u32*)(r10 - 4) = r2;				\
+	r3 = r10;					\
+	r3 += -4;					\
+	r2 = %[map_reuseport_array] ll;			\
+	call %[bpf_sk_select_reuseport];		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_select_reuseport),
+	  __imm_addr(map_reuseport_array)
+	: __clobber_all);
+}
+
+SEC("sk_reuseport")
+__description("bpf_sk_select_reuseport(ctx, sockmap, &key, flags)")
+__success
+__naked void reuseport_ctx_sockmap_key_flags(void)
+{
+	asm volatile ("					\
+	r4 = 0;						\
+	r2 = 0;						\
+	*(u32*)(r10 - 4) = r2;				\
+	r3 = r10;					\
+	r3 += -4;					\
+	r2 = %[map_sockmap] ll;				\
+	call %[bpf_sk_select_reuseport];		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_select_reuseport),
+	  __imm_addr(map_sockmap)
+	: __clobber_all);
+}
+
+SEC("sk_reuseport")
+__description("bpf_sk_select_reuseport(ctx, sockhash, &key, flags)")
+__success
+__naked void reuseport_ctx_sockhash_key_flags(void)
+{
+	asm volatile ("					\
+	r4 = 0;						\
+	r2 = 0;						\
+	*(u32*)(r10 - 4) = r2;				\
+	r3 = r10;					\
+	r3 += -4;					\
+	r2 = %[map_sockmap] ll;				\
+	call %[bpf_sk_select_reuseport];		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_select_reuseport),
+	  __imm_addr(map_sockmap)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("mark null check on return value of bpf_skc_to helpers")
+__failure __msg("invalid mem access")
+__naked void of_bpf_skc_to_helpers(void)
+{
+	asm volatile ("					\
+	r1 = *(u64*)(r1 + %[__sk_buff_sk]);		\
+	if r1 != 0 goto l0_%=;				\
+	r0 = 0;						\
+	exit;						\
+l0_%=:	r6 = r1;					\
+	call %[bpf_skc_to_tcp_sock];			\
+	r7 = r0;					\
+	r1 = r6;					\
+	call %[bpf_skc_to_tcp_request_sock];		\
+	r8 = r0;					\
+	if r8 != 0 goto l1_%=;				\
+	r0 = 0;						\
+	exit;						\
+l1_%=:	r0 = *(u8*)(r7 + 0);				\
+	exit;						\
+"	:
+	: __imm(bpf_skc_to_tcp_request_sock),
+	  __imm(bpf_skc_to_tcp_sock),
+	  __imm_const(__sk_buff_sk, offsetof(struct __sk_buff, sk))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/sock.c b/tools/testing/selftests/bpf/verifier/sock.c
deleted file mode 100644
index 108dd3ee1edd..000000000000
--- a/tools/testing/selftests/bpf/verifier/sock.c
+++ /dev/null
@@ -1,706 +0,0 @@
-{
-	"skb->sk: no NULL check",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid mem access 'sock_common_or_null'",
-},
-{
-	"skb->sk: sk->family [non fullsock field]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, offsetof(struct bpf_sock, family)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"skb->sk: sk->type [fullsock field]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, offsetof(struct bpf_sock, type)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid sock_common access",
-},
-{
-	"bpf_sk_fullsock(skb->sk): no !skb->sk check",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "type=sock_common_or_null expected=sock_common",
-},
-{
-	"sk_fullsock(skb->sk): no NULL check on ret",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, type)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid mem access 'sock_or_null'",
-},
-{
-	"sk_fullsock(skb->sk): sk->type [fullsock field]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, type)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->family [non fullsock field]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, family)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->state [narrow load]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, state)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->dst_port [word load] (backward compatibility)",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->dst_port [half load]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->dst_port [half load] (invalid)",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid sock access",
-},
-{
-	"sk_fullsock(skb->sk): sk->dst_port [byte load]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_2, BPF_REG_0, offsetof(struct bpf_sock, dst_port)),
-	BPF_LDX_MEM(BPF_B, BPF_REG_2, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->dst_port [byte load] (invalid)",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid sock access",
-},
-{
-	"sk_fullsock(skb->sk): past sk->dst_port [half load] (invalid)",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_0, offsetofend(struct bpf_sock, dst_port)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid sock access",
-},
-{
-	"sk_fullsock(skb->sk): sk->dst_ip6 [load 2nd byte]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_ip6[0]) + 1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->type [narrow load]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, type)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): sk->protocol [narrow load]",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, protocol)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"sk_fullsock(skb->sk): beyond last field",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetofend(struct bpf_sock, rx_queue_mapping)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid sock access",
-},
-{
-	"bpf_tcp_sock(skb->sk): no !skb->sk check",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "type=sock_common_or_null expected=sock_common",
-},
-{
-	"bpf_tcp_sock(skb->sk): no NULL check on ret",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_tcp_sock, snd_cwnd)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid mem access 'tcp_sock_or_null'",
-},
-{
-	"bpf_tcp_sock(skb->sk): tp->snd_cwnd",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_tcp_sock, snd_cwnd)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"bpf_tcp_sock(skb->sk): tp->bytes_acked",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_tcp_sock, bytes_acked)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"bpf_tcp_sock(skb->sk): beyond last field",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, offsetofend(struct bpf_tcp_sock, bytes_acked)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = REJECT,
-	.errstr = "invalid tcp_sock access",
-},
-{
-	"bpf_tcp_sock(bpf_sk_fullsock(skb->sk)): tp->snd_cwnd",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_tcp_sock, snd_cwnd)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.result = ACCEPT,
-},
-{
-	"bpf_sk_release(skb->sk)",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "R1 must be referenced when passed to release function",
-},
-{
-	"bpf_sk_release(bpf_sk_fullsock(skb->sk))",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "R1 must be referenced when passed to release function",
-},
-{
-	"bpf_sk_release(bpf_tcp_sock(skb->sk))",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_tcp_sock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "R1 must be referenced when passed to release function",
-},
-{
-	"sk_storage_get(map, skb->sk, NULL, 0): value == NULL",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_storage_get),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_sk_storage_map = { 11 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"sk_storage_get(map, skb->sk, 1, 1): value == 1",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_4, 1),
-	BPF_MOV64_IMM(BPF_REG_3, 1),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_storage_get),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_sk_storage_map = { 11 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "R3 type=scalar expected=fp",
-},
-{
-	"sk_storage_get(map, skb->sk, &stack_value, 1): stack_value",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_2, -8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_EMIT_CALL(BPF_FUNC_sk_fullsock),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_4, 1),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_storage_get),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_sk_storage_map = { 14 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-},
-{
-	"bpf_map_lookup_elem(smap, &key)",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_sk_storage_map = { 3 },
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "cannot pass map_type 24 into func bpf_map_lookup_elem",
-},
-{
-	"bpf_map_lookup_elem(xskmap, &key); xs->queue_id",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_xdp_sock, queue_id)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_xskmap = { 3 },
-	.prog_type = BPF_PROG_TYPE_XDP,
-	.result = ACCEPT,
-},
-{
-	"bpf_map_lookup_elem(sockmap, &key)",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_sockmap = { 3 },
-	.prog_type = BPF_PROG_TYPE_SK_SKB,
-	.result = REJECT,
-	.errstr = "Unreleased reference id=2 alloc_insn=5",
-},
-{
-	"bpf_map_lookup_elem(sockhash, &key)",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_sockhash = { 3 },
-	.prog_type = BPF_PROG_TYPE_SK_SKB,
-	.result = REJECT,
-	.errstr = "Unreleased reference id=2 alloc_insn=5",
-},
-{
-	"bpf_map_lookup_elem(sockmap, &key); sk->type [fullsock field]; bpf_sk_release(sk)",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, type)),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_sockmap = { 3 },
-	.prog_type = BPF_PROG_TYPE_SK_SKB,
-	.result = ACCEPT,
-},
-{
-	"bpf_map_lookup_elem(sockhash, &key); sk->type [fullsock field]; bpf_sk_release(sk)",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, type)),
-	BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_sockhash = { 3 },
-	.prog_type = BPF_PROG_TYPE_SK_SKB,
-	.result = ACCEPT,
-},
-{
-	"bpf_sk_select_reuseport(ctx, reuseport_array, &key, flags)",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, -4),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_select_reuseport),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_reuseport_array = { 4 },
-	.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,
-	.result = ACCEPT,
-},
-{
-	"bpf_sk_select_reuseport(ctx, sockmap, &key, flags)",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, -4),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_select_reuseport),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_sockmap = { 4 },
-	.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,
-	.result = ACCEPT,
-},
-{
-	"bpf_sk_select_reuseport(ctx, sockhash, &key, flags)",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_4, 0),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, -4),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_EMIT_CALL(BPF_FUNC_sk_select_reuseport),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_sockmap = { 4 },
-	.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,
-	.result = ACCEPT,
-},
-{
-	"mark null check on return value of bpf_skc_to helpers",
-	.insns = {
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_EMIT_CALL(BPF_FUNC_skc_to_tcp_sock),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_EMIT_CALL(BPF_FUNC_skc_to_tcp_request_sock),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_8, 0, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = REJECT,
-	.errstr = "invalid mem access",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "unknown func",
-},
-- 
cgit v1.2.3-70-g09d2


From f323a81806bd79adafa6e88607b388166e92a1b0 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:30 +0300
Subject: selftests/bpf: verifier/spin_lock converted to inline assembly

Test verifier/spin_lock automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-21-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_spin_lock.c       | 533 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/spin_lock.c   | 447 -----------------
 3 files changed, 535 insertions(+), 447 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_spin_lock.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/spin_lock.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 906f1c532cb9..1582cb5bdcc4 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -52,6 +52,7 @@
 #include "verifier_search_pruning.skel.h"
 #include "verifier_sock.skel.h"
 #include "verifier_spill_fill.skel.h"
+#include "verifier_spin_lock.skel.h"
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_uninit.skel.h"
 #include "verifier_value_adj_spill.skel.h"
@@ -144,6 +145,7 @@ void test_verifier_runtime_jit(void)          { RUN(verifier_runtime_jit); }
 void test_verifier_search_pruning(void)       { RUN(verifier_search_pruning); }
 void test_verifier_sock(void)                 { RUN(verifier_sock); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
+void test_verifier_spin_lock(void)            { RUN(verifier_spin_lock); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_spin_lock.c b/tools/testing/selftests/bpf/progs/verifier_spin_lock.c
new file mode 100644
index 000000000000..9c1aa69650f8
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_spin_lock.c
@@ -0,0 +1,533 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/spin_lock.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct val {
+	int cnt;
+	struct bpf_spin_lock l;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct val);
+} map_spin_lock SEC(".maps");
+
+SEC("cgroup/skb")
+__description("spin_lock: test1 success")
+__success __failure_unpriv __msg_unpriv("")
+__retval(0)
+__naked void spin_lock_test1_success(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u32*)(r6 + 0);				\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test2 direct ld/st")
+__failure __msg("cannot be accessed directly")
+__failure_unpriv __msg_unpriv("")
+__naked void lock_test2_direct_ld_st(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u32*)(r1 + 0);				\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test3 direct ld/st")
+__failure __msg("cannot be accessed directly")
+__failure_unpriv __msg_unpriv("")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void lock_test3_direct_ld_st(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u32*)(r6 + 1);				\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test4 direct ld/st")
+__failure __msg("cannot be accessed directly")
+__failure_unpriv __msg_unpriv("")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void lock_test4_direct_ld_st(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u16*)(r6 + 3);				\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test5 call within a locked region")
+__failure __msg("calls are not allowed")
+__failure_unpriv __msg_unpriv("")
+__naked void call_within_a_locked_region(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r6;					\
+	r1 += 4;					\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test6 missing unlock")
+__failure __msg("unlock is missing")
+__failure_unpriv __msg_unpriv("")
+__naked void spin_lock_test6_missing_unlock(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u32*)(r6 + 0);				\
+	if r0 != 0 goto l1_%=;				\
+	call %[bpf_spin_unlock];			\
+l1_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test7 unlock without lock")
+__failure __msg("without taking a lock")
+__failure_unpriv __msg_unpriv("")
+__naked void lock_test7_unlock_without_lock(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	if r1 != 0 goto l1_%=;				\
+	call %[bpf_spin_lock];				\
+l1_%=:	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u32*)(r6 + 0);				\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test8 double lock")
+__failure __msg("calls are not allowed")
+__failure_unpriv __msg_unpriv("")
+__naked void spin_lock_test8_double_lock(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r6;					\
+	r1 += 4;					\
+	r0 = *(u32*)(r6 + 0);				\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test9 different lock")
+__failure __msg("unlock of different lock")
+__failure_unpriv __msg_unpriv("")
+__naked void spin_lock_test9_different_lock(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r7 = r0;					\
+	r1 = r6;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r1 = r7;					\
+	r1 += 4;					\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("cgroup/skb")
+__description("spin_lock: test10 lock in subprog without unlock")
+__failure __msg("unlock is missing")
+__failure_unpriv __msg_unpriv("")
+__naked void lock_in_subprog_without_unlock(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r6 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call lock_in_subprog_without_unlock__1;		\
+	r1 = r6;					\
+	r1 += 4;					\
+	call %[bpf_spin_unlock];			\
+	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+static __naked __noinline __attribute__((used))
+void lock_in_subprog_without_unlock__1(void)
+{
+	asm volatile ("					\
+	call %[bpf_spin_lock];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_spin_lock)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("spin_lock: test11 ld_abs under lock")
+__failure __msg("inside bpf_spin_lock")
+__naked void test11_ld_abs_under_lock(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r7 = r0;					\
+	r1 = r0;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r0 = *(u8*)skb[0];				\
+	r1 = r7;					\
+	r1 += 4;					\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("spin_lock: regsafe compare reg->id for map value")
+__failure __msg("bpf_spin_unlock of different lock")
+__flag(BPF_F_TEST_STATE_FREQ)
+__naked void reg_id_for_map_value(void)
+{
+	asm volatile ("					\
+	r6 = r1;					\
+	r6 = *(u32*)(r6 + %[__sk_buff_mark]);		\
+	r1 = %[map_spin_lock] ll;			\
+	r9 = r1;					\
+	r2 = 0;						\
+	*(u32*)(r10 - 4) = r2;				\
+	r2 = r10;					\
+	r2 += -4;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r7 = r0;					\
+	r1 = r9;					\
+	r2 = r10;					\
+	r2 += -4;					\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l1_%=;				\
+	exit;						\
+l1_%=:	r8 = r0;					\
+	r1 = r7;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	if r6 == 0 goto l2_%=;				\
+	goto l3_%=;					\
+l2_%=:	r7 = r8;					\
+l3_%=:	r1 = r7;					\
+	r1 += 4;					\
+	call %[bpf_spin_unlock];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+/* Make sure that regsafe() compares ids for spin lock records using
+ * check_ids():
+ *  1: r9 = map_lookup_elem(...)  ; r9.id == 1
+ *  2: r8 = map_lookup_elem(...)  ; r8.id == 2
+ *  3: r7 = ktime_get_ns()
+ *  4: r6 = ktime_get_ns()
+ *  5: if r6 > r7 goto <9>
+ *  6: spin_lock(r8)
+ *  7: r9 = r8
+ *  8: goto <10>
+ *  9: spin_lock(r9)
+ * 10: spin_unlock(r9)             ; r9.id == 1 || r9.id == 2 and lock is active,
+ *                                 ; second visit to (10) should be considered safe
+ *                                 ; if check_ids() is used.
+ * 11: exit(0)
+ */
+
+SEC("cgroup/skb")
+__description("spin_lock: regsafe() check_ids() similar id mappings")
+__success __msg("29: safe")
+__failure_unpriv __msg_unpriv("")
+__log_level(2) __retval(0) __flag(BPF_F_TEST_STATE_FREQ)
+__naked void check_ids_similar_id_mappings(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u32*)(r10 - 4) = r1;				\
+	/* r9 = map_lookup_elem(...) */			\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r9 = r0;					\
+	/* r8 = map_lookup_elem(...) */			\
+	r2 = r10;					\
+	r2 += -4;					\
+	r1 = %[map_spin_lock] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l1_%=;				\
+	r8 = r0;					\
+	/* r7 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r7 = r0;					\
+	/* r6 = ktime_get_ns() */			\
+	call %[bpf_ktime_get_ns];			\
+	r6 = r0;					\
+	/* if r6 > r7 goto +5      ; no new information about the state is derived from\
+	 *                         ; this check, thus produced verifier states differ\
+	 *                         ; only in 'insn_idx'	\
+	 * spin_lock(r8)				\
+	 * r9 = r8					\
+	 * goto unlock					\
+	 */						\
+	if r6 > r7 goto l2_%=;				\
+	r1 = r8;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+	r9 = r8;					\
+	goto l3_%=;					\
+l2_%=:	/* spin_lock(r9) */				\
+	r1 = r9;					\
+	r1 += 4;					\
+	call %[bpf_spin_lock];				\
+l3_%=:	/* spin_unlock(r9) */				\
+	r1 = r9;					\
+	r1 += 4;					\
+	call %[bpf_spin_unlock];			\
+l0_%=:	/* exit(0) */					\
+	r0 = 0;						\
+l1_%=:	exit;						\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm(bpf_map_lookup_elem),
+	  __imm(bpf_spin_lock),
+	  __imm(bpf_spin_unlock),
+	  __imm_addr(map_spin_lock)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/spin_lock.c b/tools/testing/selftests/bpf/verifier/spin_lock.c
deleted file mode 100644
index eaf114f07e2e..000000000000
--- a/tools/testing/selftests/bpf/verifier/spin_lock.c
+++ /dev/null
@@ -1,447 +0,0 @@
-{
-	"spin_lock: test1 success",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test2 direct ld/st",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "cannot be accessed directly",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test3 direct ld/st",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "cannot be accessed directly",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"spin_lock: test4 direct ld/st",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_6, 3),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "cannot be accessed directly",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"spin_lock: test5 call within a locked region",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "calls are not allowed",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test6 missing unlock",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "unlock is missing",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test7 unlock without lock",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "without taking a lock",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test8 double lock",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "calls are not allowed",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test9 different lock",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3, 11 },
-	.result = REJECT,
-	.errstr = "unlock of different lock",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test10 lock in subprog without unlock",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 1, 0, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3 },
-	.result = REJECT,
-	.errstr = "unlock is missing",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-},
-{
-	"spin_lock: test11 ld_abs under lock",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_LD_ABS(BPF_B, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 4 },
-	.result = REJECT,
-	.errstr = "inside bpf_spin_lock",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"spin_lock: regsafe compare reg->id for map value",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_6, BPF_REG_6, offsetof(struct __sk_buff, mark)),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_1),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_9),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_lock),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_6, 0, 1),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_spin_unlock),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 2 },
-	.result = REJECT,
-	.errstr = "bpf_spin_unlock of different lock",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.flags = BPF_F_TEST_STATE_FREQ,
-},
-/* Make sure that regsafe() compares ids for spin lock records using
- * check_ids():
- *  1: r9 = map_lookup_elem(...)  ; r9.id == 1
- *  2: r8 = map_lookup_elem(...)  ; r8.id == 2
- *  3: r7 = ktime_get_ns()
- *  4: r6 = ktime_get_ns()
- *  5: if r6 > r7 goto <9>
- *  6: spin_lock(r8)
- *  7: r9 = r8
- *  8: goto <10>
- *  9: spin_lock(r9)
- * 10: spin_unlock(r9)             ; r9.id == 1 || r9.id == 2 and lock is active,
- *                                 ; second visit to (10) should be considered safe
- *                                 ; if check_ids() is used.
- * 11: exit(0)
- */
-{
-	"spin_lock: regsafe() check_ids() similar id mappings",
-	.insns = {
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -4, 0),
-	/* r9 = map_lookup_elem(...) */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 24),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_0),
-	/* r8 = map_lookup_elem(...) */
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
-	BPF_LD_MAP_FD(BPF_REG_1,
-		      0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 18),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
-	/* r7 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	/* r6 = ktime_get_ns() */
-	BPF_EMIT_CALL(BPF_FUNC_ktime_get_ns),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	/* if r6 > r7 goto +5      ; no new information about the state is derived from
-	 *                         ; this check, thus produced verifier states differ
-	 *                         ; only in 'insn_idx'
-	 * spin_lock(r8)
-	 * r9 = r8
-	 * goto unlock
-	 */
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_7, 5),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_EMIT_CALL(BPF_FUNC_spin_lock),
-	BPF_MOV64_REG(BPF_REG_9, BPF_REG_8),
-	BPF_JMP_A(3),
-	/* spin_lock(r9) */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_9),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_EMIT_CALL(BPF_FUNC_spin_lock),
-	/* spin_unlock(r9) */
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_9),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 4),
-	BPF_EMIT_CALL(BPF_FUNC_spin_unlock),
-	/* exit(0) */
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_spin_lock = { 3, 10 },
-	.result = VERBOSE_ACCEPT,
-	.errstr = "28: safe",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "",
-	.prog_type = BPF_PROG_TYPE_CGROUP_SKB,
-	.flags = BPF_F_TEST_STATE_FREQ,
-},
-- 
cgit v1.2.3-70-g09d2


From 81d1d6dd4037755b98bf9b9f9d0dbd715b1734e5 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:31 +0300
Subject: selftests/bpf: verifier/subreg converted to inline assembly

Test verifier/subreg automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-22-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../testing/selftests/bpf/progs/verifier_subreg.c  | 673 +++++++++++++++++++++
 tools/testing/selftests/bpf/verifier/subreg.c      | 533 ----------------
 3 files changed, 675 insertions(+), 533 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_subreg.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/subreg.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 1582cb5bdcc4..8eb36cbf00bd 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -54,6 +54,7 @@
 #include "verifier_spill_fill.skel.h"
 #include "verifier_spin_lock.skel.h"
 #include "verifier_stack_ptr.skel.h"
+#include "verifier_subreg.skel.h"
 #include "verifier_uninit.skel.h"
 #include "verifier_value_adj_spill.skel.h"
 #include "verifier_value.skel.h"
@@ -147,6 +148,7 @@ void test_verifier_sock(void)                 { RUN(verifier_sock); }
 void test_verifier_spill_fill(void)           { RUN(verifier_spill_fill); }
 void test_verifier_spin_lock(void)            { RUN(verifier_spin_lock); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
+void test_verifier_subreg(void)               { RUN(verifier_subreg); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
 void test_verifier_value(void)                { RUN(verifier_value); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_subreg.c b/tools/testing/selftests/bpf/progs/verifier_subreg.c
new file mode 100644
index 000000000000..8613ea160dcd
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_subreg.c
@@ -0,0 +1,673 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/subreg.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+/* This file contains sub-register zero extension checks for insns defining
+ * sub-registers, meaning:
+ *   - All insns under BPF_ALU class. Their BPF_ALU32 variants or narrow width
+ *     forms (BPF_END) could define sub-registers.
+ *   - Narrow direct loads, BPF_B/H/W | BPF_LDX.
+ *   - BPF_LD is not exposed to JIT back-ends, so no need for testing.
+ *
+ * "get_prandom_u32" is used to initialize low 32-bit of some registers to
+ * prevent potential optimizations done by verifier or JIT back-ends which could
+ * optimize register back into constant when range info shows one register is a
+ * constant.
+ */
+
+SEC("socket")
+__description("add32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void add32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = 0x100000000 ll;				\
+	w0 += w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("add32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void add32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	/* An insn could have no effect on the low 32-bit, for example:\
+	 *   a = a + 0					\
+	 *   a = a | 0					\
+	 *   a = a & -1					\
+	 * But, they should still zero high 32-bit.	\
+	 */						\
+	w0 += 0;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 += -2;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("sub32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void sub32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = 0x1ffffffff ll;				\
+	w0 -= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("sub32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void sub32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 -= 0;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 -= 1;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("mul32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void mul32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = 0x100000001 ll;				\
+	w0 *= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("mul32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void mul32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 *= 1;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 *= -1;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("div32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void div32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = -1;					\
+	w0 /= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("div32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void div32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 /= 1;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 /= 2;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("or32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void or32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = 0x100000001 ll;				\
+	w0 |= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("or32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void or32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 |= 0;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 |= 1;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("and32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void and32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x100000000 ll;				\
+	r1 |= r0;					\
+	r0 = 0x1ffffffff ll;				\
+	w0 &= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("and32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void and32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 &= -1;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 &= -2;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("lsh32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void lsh32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x100000000 ll;				\
+	r0 |= r1;					\
+	r1 = 1;						\
+	w0 <<= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("lsh32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void lsh32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 <<= 0;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 <<= 1;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("rsh32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void rsh32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	r1 = 1;						\
+	w0 >>= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("rsh32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void rsh32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 >>= 0;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 >>= 1;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("neg32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void neg32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 = -w0;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("mod32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void mod32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = -1;					\
+	w0 %%= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("mod32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void mod32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 %%= 1;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 %%= 2;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("xor32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void xor32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = r0;					\
+	r0 = 0x100000000 ll;				\
+	w0 ^= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("xor32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void xor32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 ^= 1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("mov32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void mov32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x100000000 ll;				\
+	r1 |= r0;					\
+	r0 = 0x100000000 ll;				\
+	w0 = w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("mov32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void mov32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 = 0;						\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 = 1;						\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("arsh32 reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void arsh32_reg_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	r1 = 1;						\
+	w0 s>>= w1;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("arsh32 imm zero extend check")
+__success __success_unpriv __retval(0)
+__naked void arsh32_imm_zero_extend_check(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 s>>= 0;					\
+	r0 >>= 32;					\
+	r6 = r0;					\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	w0 s>>= 1;					\
+	r0 >>= 32;					\
+	r0 |= r6;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("end16 (to_le) reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void le_reg_zero_extend_check_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r6 = r0;					\
+	r6 <<= 32;					\
+	call %[bpf_get_prandom_u32];			\
+	r0 |= r6;					\
+	r0 = le16 r0;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("end32 (to_le) reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void le_reg_zero_extend_check_2(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r6 = r0;					\
+	r6 <<= 32;					\
+	call %[bpf_get_prandom_u32];			\
+	r0 |= r6;					\
+	r0 = le32 r0;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("end16 (to_be) reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void be_reg_zero_extend_check_1(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r6 = r0;					\
+	r6 <<= 32;					\
+	call %[bpf_get_prandom_u32];			\
+	r0 |= r6;					\
+	r0 = be16 r0;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("end32 (to_be) reg zero extend check")
+__success __success_unpriv __retval(0)
+__naked void be_reg_zero_extend_check_2(void)
+{
+	asm volatile ("					\
+	call %[bpf_get_prandom_u32];			\
+	r6 = r0;					\
+	r6 <<= 32;					\
+	call %[bpf_get_prandom_u32];			\
+	r0 |= r6;					\
+	r0 = be32 r0;					\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ldx_b zero extend check")
+__success __success_unpriv __retval(0)
+__naked void ldx_b_zero_extend_check(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -4;					\
+	r7 = 0xfaceb00c;				\
+	*(u32*)(r6 + 0) = r7;				\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	r0 = *(u8*)(r6 + 0);				\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ldx_h zero extend check")
+__success __success_unpriv __retval(0)
+__naked void ldx_h_zero_extend_check(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -4;					\
+	r7 = 0xfaceb00c;				\
+	*(u32*)(r6 + 0) = r7;				\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	r0 = *(u16*)(r6 + 0);				\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("ldx_w zero extend check")
+__success __success_unpriv __retval(0)
+__naked void ldx_w_zero_extend_check(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -4;					\
+	r7 = 0xfaceb00c;				\
+	*(u32*)(r6 + 0) = r7;				\
+	call %[bpf_get_prandom_u32];			\
+	r1 = 0x1000000000 ll;				\
+	r0 |= r1;					\
+	r0 = *(u32*)(r6 + 0);				\
+	r0 >>= 32;					\
+	exit;						\
+"	:
+	: __imm(bpf_get_prandom_u32)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/subreg.c b/tools/testing/selftests/bpf/verifier/subreg.c
deleted file mode 100644
index 4c4133c80440..000000000000
--- a/tools/testing/selftests/bpf/verifier/subreg.c
+++ /dev/null
@@ -1,533 +0,0 @@
-/* This file contains sub-register zero extension checks for insns defining
- * sub-registers, meaning:
- *   - All insns under BPF_ALU class. Their BPF_ALU32 variants or narrow width
- *     forms (BPF_END) could define sub-registers.
- *   - Narrow direct loads, BPF_B/H/W | BPF_LDX.
- *   - BPF_LD is not exposed to JIT back-ends, so no need for testing.
- *
- * "get_prandom_u32" is used to initialize low 32-bit of some registers to
- * prevent potential optimizations done by verifier or JIT back-ends which could
- * optimize register back into constant when range info shows one register is a
- * constant.
- */
-{
-	"add32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x100000000ULL),
-	BPF_ALU32_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"add32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	/* An insn could have no effect on the low 32-bit, for example:
-	 *   a = a + 0
-	 *   a = a | 0
-	 *   a = a & -1
-	 * But, they should still zero high 32-bit.
-	 */
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, -2),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"sub32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x1ffffffffULL),
-	BPF_ALU32_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"sub32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_SUB, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_SUB, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"mul32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x100000001ULL),
-	BPF_ALU32_REG(BPF_MUL, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"mul32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_MUL, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_MUL, BPF_REG_0, -1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"div32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, -1),
-	BPF_ALU32_REG(BPF_DIV, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"div32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_DIV, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_DIV, BPF_REG_0, 2),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"or32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x100000001ULL),
-	BPF_ALU32_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"or32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_OR, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_OR, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"and32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x100000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x1ffffffffULL),
-	BPF_ALU32_REG(BPF_AND, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"and32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_AND, BPF_REG_0, -1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_AND, BPF_REG_0, -2),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"lsh32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x100000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_ALU32_REG(BPF_LSH, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"lsh32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_LSH, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_LSH, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"rsh32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_ALU32_REG(BPF_RSH, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"rsh32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_RSH, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_RSH, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"neg32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_NEG, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"mod32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, -1),
-	BPF_ALU32_REG(BPF_MOD, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"mod32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_MOD, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_MOD, BPF_REG_0, 2),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"xor32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x100000000ULL),
-	BPF_ALU32_REG(BPF_XOR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"xor32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_XOR, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"mov32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x100000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_1, BPF_REG_0),
-	BPF_LD_IMM64(BPF_REG_0, 0x100000000ULL),
-	BPF_MOV32_REG(BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"mov32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_MOV32_IMM(BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_MOV32_IMM(BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"arsh32 reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_ALU32_REG(BPF_ARSH, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"arsh32 imm zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_ARSH, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_ALU32_IMM(BPF_ARSH, BPF_REG_0, 1),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"end16 (to_le) reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_6, 32),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_ENDIAN(BPF_TO_LE, BPF_REG_0, 16),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"end32 (to_le) reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_6, 32),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_ENDIAN(BPF_TO_LE, BPF_REG_0, 32),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"end16 (to_be) reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_6, 32),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_ENDIAN(BPF_TO_BE, BPF_REG_0, 16),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"end32 (to_be) reg zero extend check",
-	.insns = {
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_ALU64_IMM(BPF_LSH, BPF_REG_6, 32),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_6),
-	BPF_ENDIAN(BPF_TO_BE, BPF_REG_0, 32),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"ldx_b zero extend check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -4),
-	BPF_ST_MEM(BPF_W, BPF_REG_6, 0, 0xfaceb00c),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_6, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"ldx_h zero extend check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -4),
-	BPF_ST_MEM(BPF_W, BPF_REG_6, 0, 0xfaceb00c),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_6, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"ldx_w zero extend check",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -4),
-	BPF_ST_MEM(BPF_W, BPF_REG_6, 0, 0xfaceb00c),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
-	BPF_LD_IMM64(BPF_REG_1, 0x1000000000ULL),
-	BPF_ALU64_REG(BPF_OR, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 0),
-	BPF_ALU64_IMM(BPF_RSH, BPF_REG_0, 32),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = 0,
-},
-- 
cgit v1.2.3-70-g09d2


From 82887c2568e4f78d5cfcb85bf3cdc056dfd97c85 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:32 +0300
Subject: selftests/bpf: verifier/unpriv converted to inline assembly

Test verifier/unpriv semi-automatically converted to use inline assembly.

The verifier/unpriv.c had to be split in two parts:
- the bulk of the tests is in the progs/verifier_unpriv.c;
- the single test that needs `struct bpf_perf_event_data`
  definition is in the progs/verifier_unpriv_perf.c.

The tests above can't be in a single file because:
- first requires inclusion of the filter.h header
  (to get access to BPF_ST_MEM macro, inline assembler does
   not support this isntruction);
- the second requires vmlinux.h, which contains definitions
  conflicting with filter.h.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-23-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   4 +
 .../testing/selftests/bpf/progs/verifier_unpriv.c  | 726 +++++++++++++++++++++
 .../selftests/bpf/progs/verifier_unpriv_perf.c     |  34 +
 tools/testing/selftests/bpf/verifier/unpriv.c      | 562 ----------------
 4 files changed, 764 insertions(+), 562 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_unpriv.c
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_unpriv_perf.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/unpriv.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 8eb36cbf00bd..a2754ea49b19 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -56,6 +56,8 @@
 #include "verifier_stack_ptr.skel.h"
 #include "verifier_subreg.skel.h"
 #include "verifier_uninit.skel.h"
+#include "verifier_unpriv.skel.h"
+#include "verifier_unpriv_perf.skel.h"
 #include "verifier_value_adj_spill.skel.h"
 #include "verifier_value.skel.h"
 #include "verifier_value_or_null.skel.h"
@@ -150,6 +152,8 @@ void test_verifier_spin_lock(void)            { RUN(verifier_spin_lock); }
 void test_verifier_stack_ptr(void)            { RUN(verifier_stack_ptr); }
 void test_verifier_subreg(void)               { RUN(verifier_subreg); }
 void test_verifier_uninit(void)               { RUN(verifier_uninit); }
+void test_verifier_unpriv(void)               { RUN(verifier_unpriv); }
+void test_verifier_unpriv_perf(void)          { RUN(verifier_unpriv_perf); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
 void test_verifier_value(void)                { RUN(verifier_value); }
 void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_unpriv.c b/tools/testing/selftests/bpf/progs/verifier_unpriv.c
new file mode 100644
index 000000000000..7ea535bfbacd
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_unpriv.c
@@ -0,0 +1,726 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/unpriv.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "../../../include/linux/filter.h"
+#include "bpf_misc.h"
+
+#define BPF_SK_LOOKUP(func) \
+	/* struct bpf_sock_tuple tuple = {} */ \
+	"r2 = 0;"			\
+	"*(u32*)(r10 - 8) = r2;"	\
+	"*(u64*)(r10 - 16) = r2;"	\
+	"*(u64*)(r10 - 24) = r2;"	\
+	"*(u64*)(r10 - 32) = r2;"	\
+	"*(u64*)(r10 - 40) = r2;"	\
+	"*(u64*)(r10 - 48) = r2;"	\
+	/* sk = func(ctx, &tuple, sizeof tuple, 0, 0) */ \
+	"r2 = r10;"			\
+	"r2 += -48;"			\
+	"r3 = %[sizeof_bpf_sock_tuple];"\
+	"r4 = 0;"			\
+	"r5 = 0;"			\
+	"call %[" #func "];"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, long long);
+} map_hash_8b SEC(".maps");
+
+void dummy_prog_42_socket(void);
+void dummy_prog_24_socket(void);
+void dummy_prog_loop1_socket(void);
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 4);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog1_socket SEC(".maps") = {
+	.values = {
+		[0] = (void *)&dummy_prog_42_socket,
+		[1] = (void *)&dummy_prog_loop1_socket,
+		[2] = (void *)&dummy_prog_24_socket,
+	},
+};
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_42_socket(void)
+{
+	asm volatile ("r0 = 42; exit;");
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_24_socket(void)
+{
+	asm volatile ("r0 = 24; exit;");
+}
+
+SEC("socket")
+__auxiliary __auxiliary_unpriv
+__naked void dummy_prog_loop1_socket(void)
+{
+	asm volatile ("			\
+	r3 = 1;				\
+	r2 = %[map_prog1_socket] ll;	\
+	call %[bpf_tail_call];		\
+	r0 = 41;			\
+	exit;				\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: return pointer")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(POINTER_VALUE)
+__naked void unpriv_return_pointer(void)
+{
+	asm volatile ("					\
+	r0 = r10;					\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: add const to pointer")
+__success __success_unpriv __retval(0)
+__naked void unpriv_add_const_to_pointer(void)
+{
+	asm volatile ("					\
+	r1 += 8;					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: add pointer to pointer")
+__failure __msg("R1 pointer += pointer")
+__failure_unpriv
+__naked void unpriv_add_pointer_to_pointer(void)
+{
+	asm volatile ("					\
+	r1 += r10;					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: neg pointer")
+__success __failure_unpriv __msg_unpriv("R1 pointer arithmetic")
+__retval(0)
+__naked void unpriv_neg_pointer(void)
+{
+	asm volatile ("					\
+	r1 = -r1;					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: cmp pointer with const")
+__success __failure_unpriv __msg_unpriv("R1 pointer comparison")
+__retval(0)
+__naked void unpriv_cmp_pointer_with_const(void)
+{
+	asm volatile ("					\
+	if r1 == 0 goto l0_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: cmp pointer with pointer")
+__success __failure_unpriv __msg_unpriv("R10 pointer comparison")
+__retval(0)
+__naked void unpriv_cmp_pointer_with_pointer(void)
+{
+	asm volatile ("					\
+	if r1 == r10 goto l0_%=;			\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tracepoint")
+__description("unpriv: check that printk is disallowed")
+__success
+__naked void check_that_printk_is_disallowed(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r1 = r10;					\
+	r1 += -8;					\
+	r2 = 8;						\
+	r3 = r1;					\
+	call %[bpf_trace_printk];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_trace_printk)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: pass pointer to helper function")
+__success __failure_unpriv __msg_unpriv("R4 leaks addr")
+__retval(0)
+__naked void pass_pointer_to_helper_function(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	r3 = r2;					\
+	r4 = r2;					\
+	call %[bpf_map_update_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_update_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: indirectly pass pointer on stack to helper function")
+__success __failure_unpriv
+__msg_unpriv("invalid indirect read from stack R2 off -8+0 size 8")
+__retval(0)
+__naked void on_stack_to_helper_function(void)
+{
+	asm volatile ("					\
+	*(u64*)(r10 - 8) = r10;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: mangle pointer on stack 1")
+__success __failure_unpriv __msg_unpriv("attempt to corrupt spilled")
+__retval(0)
+__naked void mangle_pointer_on_stack_1(void)
+{
+	asm volatile ("					\
+	*(u64*)(r10 - 8) = r10;				\
+	r0 = 0;						\
+	*(u32*)(r10 - 8) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: mangle pointer on stack 2")
+__success __failure_unpriv __msg_unpriv("attempt to corrupt spilled")
+__retval(0)
+__naked void mangle_pointer_on_stack_2(void)
+{
+	asm volatile ("					\
+	*(u64*)(r10 - 8) = r10;				\
+	r0 = 0;						\
+	*(u8*)(r10 - 1) = r0;				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: read pointer from stack in small chunks")
+__failure __msg("invalid size")
+__failure_unpriv
+__naked void from_stack_in_small_chunks(void)
+{
+	asm volatile ("					\
+	*(u64*)(r10 - 8) = r10;				\
+	r0 = *(u32*)(r10 - 8);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: write pointer into ctx")
+__failure __msg("invalid bpf_context access")
+__failure_unpriv __msg_unpriv("R1 leaks addr")
+__naked void unpriv_write_pointer_into_ctx(void)
+{
+	asm volatile ("					\
+	*(u64*)(r1 + 0) = r1;				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: spill/fill of ctx")
+__success __success_unpriv __retval(0)
+__naked void unpriv_spill_fill_of_ctx(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	*(u64*)(r6 + 0) = r1;				\
+	r1 = *(u64*)(r6 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of ctx 2")
+__success __retval(0)
+__naked void spill_fill_of_ctx_2(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	*(u64*)(r6 + 0) = r1;				\
+	r1 = *(u64*)(r6 + 0);				\
+	call %[bpf_get_hash_recalc];			\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_get_hash_recalc)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of ctx 3")
+__failure __msg("R1 type=fp expected=ctx")
+__naked void spill_fill_of_ctx_3(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	*(u64*)(r6 + 0) = r1;				\
+	*(u64*)(r6 + 0) = r10;				\
+	r1 = *(u64*)(r6 + 0);				\
+	call %[bpf_get_hash_recalc];			\
+	exit;						\
+"	:
+	: __imm(bpf_get_hash_recalc)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of ctx 4")
+__failure __msg("R1 type=scalar expected=ctx")
+__naked void spill_fill_of_ctx_4(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	*(u64*)(r6 + 0) = r1;				\
+	r0 = 1;						\
+	lock *(u64 *)(r10 - 8) += r0;			\
+	r1 = *(u64*)(r6 + 0);				\
+	call %[bpf_get_hash_recalc];			\
+	exit;						\
+"	:
+	: __imm(bpf_get_hash_recalc)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of different pointers stx")
+__failure __msg("same insn cannot be used with different pointers")
+__naked void fill_of_different_pointers_stx(void)
+{
+	asm volatile ("					\
+	r3 = 42;					\
+	r6 = r10;					\
+	r6 += -8;					\
+	if r1 == 0 goto l0_%=;				\
+	r2 = r10;					\
+	r2 += -16;					\
+	*(u64*)(r6 + 0) = r2;				\
+l0_%=:	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l1_%=:	r1 = *(u64*)(r6 + 0);				\
+	*(u32*)(r1 + %[__sk_buff_mark]) = r3;		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark))
+	: __clobber_all);
+}
+
+/* Same as above, but use BPF_ST_MEM to save 42
+ * instead of BPF_STX_MEM.
+ */
+SEC("tc")
+__description("unpriv: spill/fill of different pointers st")
+__failure __msg("same insn cannot be used with different pointers")
+__naked void fill_of_different_pointers_st(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	if r1 == 0 goto l0_%=;				\
+	r2 = r10;					\
+	r2 += -16;					\
+	*(u64*)(r6 + 0) = r2;				\
+l0_%=:	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l1_%=:	r1 = *(u64*)(r6 + 0);				\
+	.8byte %[st_mem];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_insn(st_mem,
+		     BPF_ST_MEM(BPF_W, BPF_REG_1, offsetof(struct __sk_buff, mark), 42))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of different pointers stx - ctx and sock")
+__failure __msg("type=ctx expected=sock")
+__naked void pointers_stx_ctx_and_sock(void)
+{
+	asm volatile ("					\
+	r8 = r1;					\
+	/* struct bpf_sock *sock = bpf_sock_lookup(...); */\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r2 = r0;					\
+	/* u64 foo; */					\
+	/* void *target = &foo; */			\
+	r6 = r10;					\
+	r6 += -8;					\
+	r1 = r8;					\
+	/* if (skb == NULL) *target = sock; */		\
+	if r1 == 0 goto l0_%=;				\
+	*(u64*)(r6 + 0) = r2;				\
+l0_%=:	/* else *target = skb; */			\
+	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l1_%=:	/* struct __sk_buff *skb = *target; */		\
+	r1 = *(u64*)(r6 + 0);				\
+	/* skb->mark = 42; */				\
+	r3 = 42;					\
+	*(u32*)(r1 + %[__sk_buff_mark]) = r3;		\
+	/* if (sk) bpf_sk_release(sk) */		\
+	if r1 == 0 goto l2_%=;				\
+	call %[bpf_sk_release];				\
+l2_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of different pointers stx - leak sock")
+__failure
+//.errstr = "same insn cannot be used with different pointers",
+__msg("Unreleased reference")
+__naked void different_pointers_stx_leak_sock(void)
+{
+	asm volatile ("					\
+	r8 = r1;					\
+	/* struct bpf_sock *sock = bpf_sock_lookup(...); */\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r2 = r0;					\
+	/* u64 foo; */					\
+	/* void *target = &foo; */			\
+	r6 = r10;					\
+	r6 += -8;					\
+	r1 = r8;					\
+	/* if (skb == NULL) *target = sock; */		\
+	if r1 == 0 goto l0_%=;				\
+	*(u64*)(r6 + 0) = r2;				\
+l0_%=:	/* else *target = skb; */			\
+	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l1_%=:	/* struct __sk_buff *skb = *target; */		\
+	r1 = *(u64*)(r6 + 0);				\
+	/* skb->mark = 42; */				\
+	r3 = 42;					\
+	*(u32*)(r1 + %[__sk_buff_mark]) = r3;		\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm_const(__sk_buff_mark, offsetof(struct __sk_buff, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of different pointers stx - sock and ctx (read)")
+__failure __msg("same insn cannot be used with different pointers")
+__naked void stx_sock_and_ctx_read(void)
+{
+	asm volatile ("					\
+	r8 = r1;					\
+	/* struct bpf_sock *sock = bpf_sock_lookup(...); */\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r2 = r0;					\
+	/* u64 foo; */					\
+	/* void *target = &foo; */			\
+	r6 = r10;					\
+	r6 += -8;					\
+	r1 = r8;					\
+	/* if (skb) *target = skb */			\
+	if r1 == 0 goto l0_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l0_%=:	/* else *target = sock */			\
+	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r2;				\
+l1_%=:	/* struct bpf_sock *sk = *target; */		\
+	r1 = *(u64*)(r6 + 0);				\
+	/* if (sk) u32 foo = sk->mark; bpf_sk_release(sk); */\
+	if r1 == 0 goto l2_%=;				\
+	r3 = *(u32*)(r1 + %[bpf_sock_mark]);		\
+	call %[bpf_sk_release];				\
+l2_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(bpf_sock_mark, offsetof(struct bpf_sock, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("unpriv: spill/fill of different pointers stx - sock and ctx (write)")
+__failure
+//.errstr = "same insn cannot be used with different pointers",
+__msg("cannot write into sock")
+__naked void stx_sock_and_ctx_write(void)
+{
+	asm volatile ("					\
+	r8 = r1;					\
+	/* struct bpf_sock *sock = bpf_sock_lookup(...); */\
+"	BPF_SK_LOOKUP(bpf_sk_lookup_tcp)
+"	r2 = r0;					\
+	/* u64 foo; */					\
+	/* void *target = &foo; */			\
+	r6 = r10;					\
+	r6 += -8;					\
+	r1 = r8;					\
+	/* if (skb) *target = skb */			\
+	if r1 == 0 goto l0_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l0_%=:	/* else *target = sock */			\
+	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r2;				\
+l1_%=:	/* struct bpf_sock *sk = *target; */		\
+	r1 = *(u64*)(r6 + 0);				\
+	/* if (sk) sk->mark = 42; bpf_sk_release(sk); */\
+	if r1 == 0 goto l2_%=;				\
+	r3 = 42;					\
+	*(u32*)(r1 + %[bpf_sock_mark]) = r3;		\
+	call %[bpf_sk_release];				\
+l2_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_sk_lookup_tcp),
+	  __imm(bpf_sk_release),
+	  __imm_const(bpf_sock_mark, offsetof(struct bpf_sock, mark)),
+	  __imm_const(sizeof_bpf_sock_tuple, sizeof(struct bpf_sock_tuple))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: write pointer into map elem value")
+__success __failure_unpriv __msg_unpriv("R0 leaks addr")
+__retval(0)
+__naked void pointer_into_map_elem_value(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_hash_8b] ll;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	*(u64*)(r0 + 0) = r0;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("alu32: mov u32 const")
+__success __failure_unpriv __msg_unpriv("R7 invalid mem access 'scalar'")
+__retval(0)
+__naked void alu32_mov_u32_const(void)
+{
+	asm volatile ("					\
+	w7 = 0;						\
+	w7 &= 1;					\
+	w0 = w7;					\
+	if r0 == 0 goto l0_%=;				\
+	r0 = *(u64*)(r7 + 0);				\
+l0_%=:	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: partial copy of pointer")
+__success __failure_unpriv __msg_unpriv("R10 partial copy")
+__retval(0)
+__naked void unpriv_partial_copy_of_pointer(void)
+{
+	asm volatile ("					\
+	w1 = w10;					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: pass pointer to tail_call")
+__success __failure_unpriv __msg_unpriv("R3 leaks addr into helper")
+__retval(0)
+__naked void pass_pointer_to_tail_call(void)
+{
+	asm volatile ("					\
+	r3 = r1;					\
+	r2 = %[map_prog1_socket] ll;			\
+	call %[bpf_tail_call];				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_tail_call),
+	  __imm_addr(map_prog1_socket)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: cmp map pointer with zero")
+__success __failure_unpriv __msg_unpriv("R1 pointer comparison")
+__retval(0)
+__naked void cmp_map_pointer_with_zero(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	r1 = %[map_hash_8b] ll;				\
+	if r1 == 0 goto l0_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_addr(map_hash_8b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: write into frame pointer")
+__failure __msg("frame pointer is read only")
+__failure_unpriv
+__naked void unpriv_write_into_frame_pointer(void)
+{
+	asm volatile ("					\
+	r10 = r1;					\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: spill/fill frame pointer")
+__failure __msg("frame pointer is read only")
+__failure_unpriv
+__naked void unpriv_spill_fill_frame_pointer(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	*(u64*)(r6 + 0) = r10;				\
+	r10 = *(u64*)(r6 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: cmp of frame pointer")
+__success __failure_unpriv __msg_unpriv("R10 pointer comparison")
+__retval(0)
+__naked void unpriv_cmp_of_frame_pointer(void)
+{
+	asm volatile ("					\
+	if r10 == 0 goto l0_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: adding of fp, reg")
+__success __failure_unpriv
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__retval(0)
+__naked void unpriv_adding_of_fp_reg(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = 0;						\
+	r1 += r10;					\
+	*(u64*)(r1 - 8) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: adding of fp, imm")
+__success __failure_unpriv
+__msg_unpriv("R1 stack pointer arithmetic goes out of range")
+__retval(0)
+__naked void unpriv_adding_of_fp_imm(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = r10;					\
+	r1 += 0;					\
+	*(u64*)(r1 - 8) = r0;				\
+	exit;						\
+"	::: __clobber_all);
+}
+
+SEC("socket")
+__description("unpriv: cmp of stack pointer")
+__success __failure_unpriv __msg_unpriv("R2 pointer comparison")
+__retval(0)
+__naked void unpriv_cmp_of_stack_pointer(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r2 == 0 goto l0_%=;				\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	::: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/verifier_unpriv_perf.c b/tools/testing/selftests/bpf/progs/verifier_unpriv_perf.c
new file mode 100644
index 000000000000..4d77407a0a79
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_unpriv_perf.c
@@ -0,0 +1,34 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/unpriv.c */
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+SEC("perf_event")
+__description("unpriv: spill/fill of different pointers ldx")
+__failure __msg("same insn cannot be used with different pointers")
+__naked void fill_of_different_pointers_ldx(void)
+{
+	asm volatile ("					\
+	r6 = r10;					\
+	r6 += -8;					\
+	if r1 == 0 goto l0_%=;				\
+	r2 = r10;					\
+	r2 += %[__imm_0];				\
+	*(u64*)(r6 + 0) = r2;				\
+l0_%=:	if r1 != 0 goto l1_%=;				\
+	*(u64*)(r6 + 0) = r1;				\
+l1_%=:	r1 = *(u64*)(r6 + 0);				\
+	r1 = *(u64*)(r1 + %[sample_period]);		\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__imm_0,
+		      -(__s32) offsetof(struct bpf_perf_event_data, sample_period) - 8),
+	  __imm_const(sample_period,
+		      offsetof(struct bpf_perf_event_data, sample_period))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/unpriv.c b/tools/testing/selftests/bpf/verifier/unpriv.c
deleted file mode 100644
index af0c0f336625..000000000000
--- a/tools/testing/selftests/bpf/verifier/unpriv.c
+++ /dev/null
@@ -1,562 +0,0 @@
-{
-	"unpriv: return pointer",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_10),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 leaks addr",
-	.retval = POINTER_VALUE,
-},
-{
-	"unpriv: add const to pointer",
-	.insns = {
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-},
-{
-	"unpriv: add pointer to pointer",
-	.insns = {
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_10),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R1 pointer += pointer",
-},
-{
-	"unpriv: neg pointer",
-	.insns = {
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 pointer arithmetic",
-},
-{
-	"unpriv: cmp pointer with const",
-	.insns = {
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 pointer comparison",
-},
-{
-	"unpriv: cmp pointer with pointer",
-	.insns = {
-	BPF_JMP_REG(BPF_JEQ, BPF_REG_1, BPF_REG_10, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R10 pointer comparison",
-},
-{
-	"unpriv: check that printk is disallowed",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
-	BPF_MOV64_IMM(BPF_REG_2, 8),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_1),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_trace_printk),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "unknown func bpf_trace_printk#6",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
-},
-{
-	"unpriv: pass pointer to helper function",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_2),
-	BPF_MOV64_REG(BPF_REG_4, BPF_REG_2),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_update_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr_unpriv = "R4 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: indirectly pass pointer on stack to helper function",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_10, -8),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr_unpriv = "invalid indirect read from stack R2 off -8+0 size 8",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: mangle pointer on stack 1",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_10, -8),
-	BPF_ST_MEM(BPF_W, BPF_REG_10, -8, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "attempt to corrupt spilled",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: mangle pointer on stack 2",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_10, -8),
-	BPF_ST_MEM(BPF_B, BPF_REG_10, -1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "attempt to corrupt spilled",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: read pointer from stack in small chunks",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_10, -8),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_10, -8),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "invalid size",
-	.result = REJECT,
-},
-{
-	"unpriv: write pointer into ctx",
-	.insns = {
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 leaks addr",
-	.result_unpriv = REJECT,
-	.errstr = "invalid bpf_context access",
-	.result = REJECT,
-},
-{
-	"unpriv: spill/fill of ctx",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-},
-{
-	"unpriv: spill/fill of ctx 2",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_hash_recalc),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of ctx 3",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_10, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_hash_recalc),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R1 type=fp expected=ctx",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of ctx 4",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_RAW_INSN(BPF_STX | BPF_ATOMIC | BPF_DW,
-		     BPF_REG_10, BPF_REG_0, -8, BPF_ADD),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_hash_recalc),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "R1 type=scalar expected=ctx",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of different pointers stx",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_3, 42),
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_3,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "same insn cannot be used with different pointers",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	/* Same as above, but use BPF_ST_MEM to save 42
-	 * instead of BPF_STX_MEM.
-	 */
-	"unpriv: spill/fill of different pointers st",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -16),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_ST_MEM(BPF_W, BPF_REG_1, offsetof(struct __sk_buff, mark), 42),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "same insn cannot be used with different pointers",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of different pointers stx - ctx and sock",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	/* struct bpf_sock *sock = bpf_sock_lookup(...); */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	/* u64 foo; */
-	/* void *target = &foo; */
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	/* if (skb == NULL) *target = sock; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	/* else *target = skb; */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	/* struct __sk_buff *skb = *target; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	/* skb->mark = 42; */
-	BPF_MOV64_IMM(BPF_REG_3, 42),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_3,
-		    offsetof(struct __sk_buff, mark)),
-	/* if (sk) bpf_sk_release(sk) */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-		BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "type=ctx expected=sock",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of different pointers stx - leak sock",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	/* struct bpf_sock *sock = bpf_sock_lookup(...); */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	/* u64 foo; */
-	/* void *target = &foo; */
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	/* if (skb == NULL) *target = sock; */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	/* else *target = skb; */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	/* struct __sk_buff *skb = *target; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	/* skb->mark = 42; */
-	BPF_MOV64_IMM(BPF_REG_3, 42),
-	BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_3,
-		    offsetof(struct __sk_buff, mark)),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	//.errstr = "same insn cannot be used with different pointers",
-	.errstr = "Unreleased reference",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of different pointers stx - sock and ctx (read)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	/* struct bpf_sock *sock = bpf_sock_lookup(...); */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	/* u64 foo; */
-	/* void *target = &foo; */
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	/* if (skb) *target = skb */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	/* else *target = sock */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	/* struct bpf_sock *sk = *target; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	/* if (sk) u32 foo = sk->mark; bpf_sk_release(sk); */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 2),
-		BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
-			    offsetof(struct bpf_sock, mark)),
-		BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "same insn cannot be used with different pointers",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of different pointers stx - sock and ctx (write)",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_1),
-	/* struct bpf_sock *sock = bpf_sock_lookup(...); */
-	BPF_SK_LOOKUP(sk_lookup_tcp),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	/* u64 foo; */
-	/* void *target = &foo; */
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_8),
-	/* if (skb) *target = skb */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	/* else *target = sock */
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	/* struct bpf_sock *sk = *target; */
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	/* if (sk) sk->mark = 42; bpf_sk_release(sk); */
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
-		BPF_MOV64_IMM(BPF_REG_3, 42),
-		BPF_STX_MEM(BPF_W, BPF_REG_1, BPF_REG_3,
-			    offsetof(struct bpf_sock, mark)),
-		BPF_EMIT_CALL(BPF_FUNC_sk_release),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	//.errstr = "same insn cannot be used with different pointers",
-	.errstr = "cannot write into sock",
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-},
-{
-	"unpriv: spill/fill of different pointers ldx",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2,
-		      -(__s32)offsetof(struct bpf_perf_event_data,
-				       sample_period) - 8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_2, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1,
-		    offsetof(struct bpf_perf_event_data, sample_period)),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.result = REJECT,
-	.errstr = "same insn cannot be used with different pointers",
-	.prog_type = BPF_PROG_TYPE_PERF_EVENT,
-},
-{
-	"unpriv: write pointer into map elem value",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 3 },
-	.errstr_unpriv = "R0 leaks addr",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"alu32: mov u32 const",
-	.insns = {
-	BPF_MOV32_IMM(BPF_REG_7, 0),
-	BPF_ALU32_IMM(BPF_AND, BPF_REG_7, 1),
-	BPF_MOV32_REG(BPF_REG_0, BPF_REG_7),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_7, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R7 invalid mem access 'scalar'",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-	.retval = 0,
-},
-{
-	"unpriv: partial copy of pointer",
-	.insns = {
-	BPF_MOV32_REG(BPF_REG_1, BPF_REG_10),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R10 partial copy",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: pass pointer to tail_call",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_3, BPF_REG_1),
-	BPF_LD_MAP_FD(BPF_REG_2, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_tail_call),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog1 = { 1 },
-	.errstr_unpriv = "R3 leaks addr into helper",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: cmp map pointer with zero",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_8b = { 1 },
-	.errstr_unpriv = "R1 pointer comparison",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: write into frame pointer",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_10, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "frame pointer is read only",
-	.result = REJECT,
-},
-{
-	"unpriv: spill/fill frame pointer",
-	.insns = {
-	BPF_ALU64_REG(BPF_MOV, BPF_REG_6, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_10, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_10, BPF_REG_6, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr = "frame pointer is read only",
-	.result = REJECT,
-},
-{
-	"unpriv: cmp of frame pointer",
-	.insns = {
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_10, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R10 pointer comparison",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: adding of fp, reg",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_1, 0),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_10),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, -8),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: adding of fp, imm",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 0),
-	BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, -8),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R1 stack pointer arithmetic goes out of range",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-{
-	"unpriv: cmp of stack pointer",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_2, 0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.errstr_unpriv = "R2 pointer comparison",
-	.result_unpriv = REJECT,
-	.result = ACCEPT,
-},
-- 
cgit v1.2.3-70-g09d2


From efe25a330b106f39e5d27760cc57def3e8c5a3f9 Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:33 +0300
Subject: selftests/bpf: verifier/value_illegal_alu converted to inline
 assembly

Test verifier/value_illegal_alu automatically converted to use inline assembly.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-24-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../bpf/progs/verifier_value_illegal_alu.c         | 149 +++++++++++++++++++++
 .../selftests/bpf/verifier/value_illegal_alu.c     |  95 -------------
 3 files changed, 151 insertions(+), 95 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_value_illegal_alu.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/value_illegal_alu.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index a2754ea49b19..8695d39f84dd 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -60,6 +60,7 @@
 #include "verifier_unpriv_perf.skel.h"
 #include "verifier_value_adj_spill.skel.h"
 #include "verifier_value.skel.h"
+#include "verifier_value_illegal_alu.skel.h"
 #include "verifier_value_or_null.skel.h"
 #include "verifier_var_off.skel.h"
 #include "verifier_xadd.skel.h"
@@ -156,6 +157,7 @@ void test_verifier_unpriv(void)               { RUN(verifier_unpriv); }
 void test_verifier_unpriv_perf(void)          { RUN(verifier_unpriv_perf); }
 void test_verifier_value_adj_spill(void)      { RUN(verifier_value_adj_spill); }
 void test_verifier_value(void)                { RUN(verifier_value); }
+void test_verifier_value_illegal_alu(void)    { RUN(verifier_value_illegal_alu); }
 void test_verifier_value_or_null(void)        { RUN(verifier_value_or_null); }
 void test_verifier_var_off(void)              { RUN(verifier_var_off); }
 void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_value_illegal_alu.c b/tools/testing/selftests/bpf/progs/verifier_value_illegal_alu.c
new file mode 100644
index 000000000000..71814a753216
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_value_illegal_alu.c
@@ -0,0 +1,149 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/value_illegal_alu.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("socket")
+__description("map element value illegal alu op, 1")
+__failure __msg("R0 bitwise operator &= on pointer")
+__failure_unpriv
+__naked void value_illegal_alu_op_1(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 &= 8;					\
+	r1 = 22;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value illegal alu op, 2")
+__failure __msg("R0 32-bit pointer arithmetic prohibited")
+__failure_unpriv
+__naked void value_illegal_alu_op_2(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	w0 += 0;					\
+	r1 = 22;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value illegal alu op, 3")
+__failure __msg("R0 pointer arithmetic with /= operator")
+__failure_unpriv
+__naked void value_illegal_alu_op_3(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 /= 42;					\
+	r1 = 22;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value illegal alu op, 4")
+__failure __msg("invalid mem access 'scalar'")
+__failure_unpriv __msg_unpriv("R0 pointer arithmetic prohibited")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void value_illegal_alu_op_4(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 = be64 r0;					\
+	r1 = 22;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map element value illegal alu op, 5")
+__failure __msg("R0 invalid mem access 'scalar'")
+__msg_unpriv("leaking pointer from stack off -8")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void value_illegal_alu_op_5(void)
+{
+	asm volatile ("					\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = 0;						\
+	*(u64*)(r2 + 0) = r1;				\
+	r1 = %[map_hash_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r3 = 4096;					\
+	r2 = r10;					\
+	r2 += -8;					\
+	*(u64*)(r2 + 0) = r0;				\
+	lock *(u64 *)(r2 + 0) += r3;			\
+	r0 = *(u64*)(r2 + 0);				\
+	r1 = 22;					\
+	*(u64*)(r0 + 0) = r1;				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_hash_48b)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/value_illegal_alu.c b/tools/testing/selftests/bpf/verifier/value_illegal_alu.c
deleted file mode 100644
index d6f29eb4bd57..000000000000
--- a/tools/testing/selftests/bpf/verifier/value_illegal_alu.c
+++ /dev/null
@@ -1,95 +0,0 @@
-{
-	"map element value illegal alu op, 1",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 22),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 bitwise operator &= on pointer",
-	.result = REJECT,
-},
-{
-	"map element value illegal alu op, 2",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU32_IMM(BPF_ADD, BPF_REG_0, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 22),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 32-bit pointer arithmetic prohibited",
-	.result = REJECT,
-},
-{
-	"map element value illegal alu op, 3",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU64_IMM(BPF_DIV, BPF_REG_0, 42),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 22),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr = "R0 pointer arithmetic with /= operator",
-	.result = REJECT,
-},
-{
-	"map element value illegal alu op, 4",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ENDIAN(BPF_FROM_BE, BPF_REG_0, 64),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 22),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "R0 pointer arithmetic prohibited",
-	.errstr = "invalid mem access 'scalar'",
-	.result = REJECT,
-	.result_unpriv = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map element value illegal alu op, 5",
-	.insns = {
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_ST_MEM(BPF_DW, BPF_REG_2, 0, 0),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_IMM(BPF_REG_3, 4096),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_STX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
-	BPF_ATOMIC_OP(BPF_DW, BPF_ADD, BPF_REG_2, BPF_REG_3, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_0, 0, 22),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 3 },
-	.errstr_unpriv = "leaking pointer from stack off -8",
-	.errstr = "R0 invalid mem access 'scalar'",
-	.result = REJECT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-- 
cgit v1.2.3-70-g09d2


From 4db10a8243df665ced10b027c2d4862173a7a3ec Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 20:42:34 +0300
Subject: selftests/bpf: verifier/value_ptr_arith converted to inline assembly

Test verifier/value_ptr_arith automatically converted to use inline assembly.

Test cases "sanitation: alu with different scalars 2" and
"sanitation: alu with different scalars 3" are updated to
avoid -ENOENT as return value, as __retval() annotation
only supports numeric literals.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421174234.2391278-25-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   34 +-
 .../selftests/bpf/progs/verifier_value_ptr_arith.c | 1423 ++++++++++++++++++++
 .../selftests/bpf/verifier/value_ptr_arith.c       | 1140 ----------------
 3 files changed, 1451 insertions(+), 1146 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_value_ptr_arith.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/value_ptr_arith.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index 8695d39f84dd..c8bab8b1a6a4 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -62,6 +62,7 @@
 #include "verifier_value.skel.h"
 #include "verifier_value_illegal_alu.skel.h"
 #include "verifier_value_or_null.skel.h"
+#include "verifier_value_ptr_arith.skel.h"
 #include "verifier_var_off.skel.h"
 #include "verifier_xadd.skel.h"
 #include "verifier_xdp.skel.h"
@@ -164,29 +165,50 @@ void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
 void test_verifier_xdp(void)                  { RUN(verifier_xdp); }
 void test_verifier_xdp_direct_packet_access(void) { RUN(verifier_xdp_direct_packet_access); }
 
-static int init_array_access_maps(struct bpf_object *obj)
+static int init_test_val_map(struct bpf_object *obj, char *map_name)
 {
-	struct bpf_map *array_ro;
 	struct test_val value = {
 		.index = (6 + 1) * sizeof(int),
 		.foo[6] = 0xabcdef12,
 	};
+	struct bpf_map *map;
 	int err, key = 0;
 
-	array_ro = bpf_object__find_map_by_name(obj, "map_array_ro");
-	if (!ASSERT_OK_PTR(array_ro, "lookup map_array_ro"))
+	map = bpf_object__find_map_by_name(obj, map_name);
+	if (!map) {
+		PRINT_FAIL("Can't find map '%s'\n", map_name);
 		return -EINVAL;
+	}
 
-	err = bpf_map_update_elem(bpf_map__fd(array_ro), &key, &value, 0);
-	if (!ASSERT_OK(err, "map_array_ro update"))
+	err = bpf_map_update_elem(bpf_map__fd(map), &key, &value, 0);
+	if (err) {
+		PRINT_FAIL("Error while updating map '%s': %d\n", map_name, err);
 		return err;
+	}
 
 	return 0;
 }
 
+static int init_array_access_maps(struct bpf_object *obj)
+{
+	return init_test_val_map(obj, "map_array_ro");
+}
+
 void test_verifier_array_access(void)
 {
 	run_tests_aux("verifier_array_access",
 		      verifier_array_access__elf_bytes,
 		      init_array_access_maps);
 }
+
+static int init_value_ptr_arith_maps(struct bpf_object *obj)
+{
+	return init_test_val_map(obj, "map_array_48b");
+}
+
+void test_verifier_value_ptr_arith(void)
+{
+	run_tests_aux("verifier_value_ptr_arith",
+		      verifier_value_ptr_arith__elf_bytes,
+		      init_value_ptr_arith_maps);
+}
diff --git a/tools/testing/selftests/bpf/progs/verifier_value_ptr_arith.c b/tools/testing/selftests/bpf/progs/verifier_value_ptr_arith.c
new file mode 100644
index 000000000000..5ba6e53571c8
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_value_ptr_arith.c
@@ -0,0 +1,1423 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/value_ptr_arith.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <errno.h>
+#include "bpf_misc.h"
+
+#define MAX_ENTRIES 11
+
+struct test_val {
+	unsigned int index;
+	int foo[MAX_ENTRIES];
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct test_val);
+} map_array_48b SEC(".maps");
+
+struct other_val {
+	long long foo;
+	long long bar;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct other_val);
+} map_hash_16b SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 1);
+	__type(key, long long);
+	__type(value, struct test_val);
+} map_hash_48b SEC(".maps");
+
+SEC("socket")
+__description("map access: known scalar += value_ptr unknown vs const")
+__success __failure_unpriv
+__msg_unpriv("R1 tried to add from different maps, paths or scalars")
+__retval(1)
+__naked void value_ptr_unknown_vs_const(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x7;					\
+	goto l4_%=;					\
+l3_%=:	r1 = 3;						\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr const vs unknown")
+__success __failure_unpriv
+__msg_unpriv("R1 tried to add from different maps, paths or scalars")
+__retval(1)
+__naked void value_ptr_const_vs_unknown(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 3;						\
+	goto l4_%=;					\
+l3_%=:	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x7;					\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr const vs const (ne)")
+__success __failure_unpriv
+__msg_unpriv("R1 tried to add from different maps, paths or scalars")
+__retval(1)
+__naked void ptr_const_vs_const_ne(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 3;						\
+	goto l4_%=;					\
+l3_%=:	r1 = 5;						\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr const vs const (eq)")
+__success __success_unpriv __retval(1)
+__naked void ptr_const_vs_const_eq(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 5;						\
+	goto l4_%=;					\
+l3_%=:	r1 = 5;						\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr unknown vs unknown (eq)")
+__success __success_unpriv __retval(1)
+__naked void ptr_unknown_vs_unknown_eq(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x7;					\
+	goto l4_%=;					\
+l3_%=:	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x7;					\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr unknown vs unknown (lt)")
+__success __failure_unpriv
+__msg_unpriv("R1 tried to add from different maps, paths or scalars")
+__retval(1)
+__naked void ptr_unknown_vs_unknown_lt(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x3;					\
+	goto l4_%=;					\
+l3_%=:	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x7;					\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr unknown vs unknown (gt)")
+__success __failure_unpriv
+__msg_unpriv("R1 tried to add from different maps, paths or scalars")
+__retval(1)
+__naked void ptr_unknown_vs_unknown_gt(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r4 = *(u8*)(r0 + 0);				\
+	if r4 == 1 goto l3_%=;				\
+	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x7;					\
+	goto l4_%=;					\
+l3_%=:	r1 = 6;						\
+	r1 = -r1;					\
+	r1 &= 0x3;					\
+l4_%=:	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr from different maps")
+__success __success_unpriv __retval(1)
+__naked void value_ptr_from_different_maps(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r1 = 4;						\
+	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= known scalar from different maps")
+__success __failure_unpriv
+__msg_unpriv("R0 min value is outside of the allowed memory range")
+__retval(1)
+__naked void known_scalar_from_different_maps(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_16b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r1 = 4;						\
+	r0 -= r1;					\
+	r0 += r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_16b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr from different maps, but same value properties")
+__success __success_unpriv __retval(1)
+__naked void maps_but_same_value_properties(void)
+{
+	asm volatile ("					\
+	r0 = *(u32*)(r1 + %[__sk_buff_len]);		\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	if r0 == 1 goto l0_%=;				\
+	r1 = %[map_hash_48b] ll;			\
+	if r0 != 1 goto l1_%=;				\
+l0_%=:	r1 = %[map_array_48b] ll;			\
+l1_%=:	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l2_%=;				\
+	r1 = 4;						\
+	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l2_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_addr(map_hash_48b),
+	  __imm_const(__sk_buff_len, offsetof(struct __sk_buff, len))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: mixing value pointer and scalar, 1")
+__success __failure_unpriv __msg_unpriv("R2 pointer comparison prohibited")
+__retval(0)
+__naked void value_pointer_and_scalar_1(void)
+{
+	asm volatile ("					\
+	/* load map value pointer into r0 and r2 */	\
+	r0 = 1;						\
+	r1 = %[map_array_48b] ll;			\
+	r2 = r10;					\
+	r2 += -16;					\
+	r6 = 0;						\
+	*(u64*)(r10 - 16) = r6;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	/* load some number from the map into r1 */	\
+	r1 = *(u8*)(r0 + 0);				\
+	/* depending on r1, branch: */			\
+	if r1 != 0 goto l1_%=;				\
+	/* branch A */					\
+	r2 = r0;					\
+	r3 = 0;						\
+	goto l2_%=;					\
+l1_%=:	/* branch B */					\
+	r2 = 0;						\
+	r3 = 0x100000;					\
+l2_%=:	/* common instruction */			\
+	r2 += r3;					\
+	/* depending on r1, branch: */			\
+	if r1 != 0 goto l3_%=;				\
+	/* branch A */					\
+	goto l4_%=;					\
+l3_%=:	/* branch B */					\
+	r0 = 0x13371337;				\
+	/* verifier follows fall-through */		\
+	if r2 != 0x100000 goto l4_%=;			\
+	r0 = 0;						\
+	exit;						\
+l4_%=:	/* fake-dead code; targeted from branch A to	\
+	 * prevent dead code sanitization		\
+	 */						\
+	r0 = *(u8*)(r0 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: mixing value pointer and scalar, 2")
+__success __failure_unpriv __msg_unpriv("R0 invalid mem access 'scalar'")
+__retval(0)
+__naked void value_pointer_and_scalar_2(void)
+{
+	asm volatile ("					\
+	/* load map value pointer into r0 and r2 */	\
+	r0 = 1;						\
+	r1 = %[map_array_48b] ll;			\
+	r2 = r10;					\
+	r2 += -16;					\
+	r6 = 0;						\
+	*(u64*)(r10 - 16) = r6;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	/* load some number from the map into r1 */	\
+	r1 = *(u8*)(r0 + 0);				\
+	/* depending on r1, branch: */			\
+	if r1 == 0 goto l1_%=;				\
+	/* branch A */					\
+	r2 = 0;						\
+	r3 = 0x100000;					\
+	goto l2_%=;					\
+l1_%=:	/* branch B */					\
+	r2 = r0;					\
+	r3 = 0;						\
+l2_%=:	/* common instruction */			\
+	r2 += r3;					\
+	/* depending on r1, branch: */			\
+	if r1 != 0 goto l3_%=;				\
+	/* branch A */					\
+	goto l4_%=;					\
+l3_%=:	/* branch B */					\
+	r0 = 0x13371337;				\
+	/* verifier follows fall-through */		\
+	if r2 != 0x100000 goto l4_%=;			\
+	r0 = 0;						\
+	exit;						\
+l4_%=:	/* fake-dead code; targeted from branch A to	\
+	 * prevent dead code sanitization, rejected	\
+	 * via branch B however				\
+	 */						\
+	r0 = *(u8*)(r0 + 0);				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("sanitation: alu with different scalars 1")
+__success __success_unpriv __retval(0x100000)
+__naked void alu_with_different_scalars_1(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	r1 = %[map_array_48b] ll;			\
+	r2 = r10;					\
+	r2 += -16;					\
+	r6 = 0;						\
+	*(u64*)(r10 - 16) = r6;				\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r1 = *(u32*)(r0 + 0);				\
+	if r1 == 0 goto l1_%=;				\
+	r2 = 0;						\
+	r3 = 0x100000;					\
+	goto l2_%=;					\
+l1_%=:	r2 = 42;					\
+	r3 = 0x100001;					\
+l2_%=:	r2 += r3;					\
+	r0 = r2;					\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("sanitation: alu with different scalars 2")
+__success __success_unpriv __retval(0)
+__naked void alu_with_different_scalars_2(void)
+{
+	asm volatile ("					\
+	r0 = 1;						\
+	r1 = %[map_array_48b] ll;			\
+	r6 = r1;					\
+	r2 = r10;					\
+	r2 += -16;					\
+	r7 = 0;						\
+	*(u64*)(r10 - 16) = r7;				\
+	call %[bpf_map_delete_elem];			\
+	r7 = r0;					\
+	r1 = r6;					\
+	r2 = r10;					\
+	r2 += -16;					\
+	call %[bpf_map_delete_elem];			\
+	r6 = r0;					\
+	r8 = r6;					\
+	r8 += r7;					\
+	r0 = r8;					\
+	r0 += %[einval];				\
+	r0 += %[einval];				\
+	exit;						\
+"	:
+	: __imm(bpf_map_delete_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_const(einval, EINVAL)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("sanitation: alu with different scalars 3")
+__success __success_unpriv __retval(0)
+__naked void alu_with_different_scalars_3(void)
+{
+	asm volatile ("					\
+	r0 = %[einval];					\
+	r0 *= -1;					\
+	r7 = r0;					\
+	r0 = %[einval];					\
+	r0 *= -1;					\
+	r6 = r0;					\
+	r8 = r6;					\
+	r8 += r7;					\
+	r0 = r8;					\
+	r0 += %[einval];				\
+	r0 += %[einval];				\
+	exit;						\
+"	:
+	: __imm_const(einval, EINVAL)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, upper oob arith, test 1")
+__success __failure_unpriv
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__retval(1)
+__naked void upper_oob_arith_test_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 48;					\
+	r0 += r1;					\
+	r0 -= r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, upper oob arith, test 2")
+__success __failure_unpriv
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__retval(1)
+__naked void upper_oob_arith_test_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 49;					\
+	r0 += r1;					\
+	r0 -= r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, upper oob arith, test 3")
+__success __success_unpriv __retval(1)
+__naked void upper_oob_arith_test_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 47;					\
+	r0 += r1;					\
+	r0 -= r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= known scalar, lower oob arith, test 1")
+__failure __msg("R0 min value is outside of the allowed memory range")
+__failure_unpriv
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__naked void lower_oob_arith_test_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 47;					\
+	r0 += r1;					\
+	r1 = 48;					\
+	r0 -= r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= known scalar, lower oob arith, test 2")
+__success __failure_unpriv
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__retval(1)
+__naked void lower_oob_arith_test_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 47;					\
+	r0 += r1;					\
+	r1 = 48;					\
+	r0 -= r1;					\
+	r1 = 1;						\
+	r0 += r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= known scalar, lower oob arith, test 3")
+__success __success_unpriv __retval(1)
+__naked void lower_oob_arith_test_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 47;					\
+	r0 += r1;					\
+	r1 = 47;					\
+	r0 -= r1;					\
+	r0 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar += value_ptr")
+__success __success_unpriv __retval(1)
+__naked void access_known_scalar_value_ptr_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 4;						\
+	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, 1")
+__success __success_unpriv __retval(1)
+__naked void value_ptr_known_scalar_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 4;						\
+	r0 += r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, 2")
+__failure __msg("invalid access to map value")
+__failure_unpriv
+__naked void value_ptr_known_scalar_2_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 49;					\
+	r0 += r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, 3")
+__failure __msg("invalid access to map value")
+__failure_unpriv
+__naked void value_ptr_known_scalar_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = -1;					\
+	r0 += r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, 4")
+__success __success_unpriv __retval(1)
+__naked void value_ptr_known_scalar_4(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 5;						\
+	r0 += r1;					\
+	r1 = -2;					\
+	r0 += r1;					\
+	r1 = -1;					\
+	r0 += r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, 5")
+__success __success_unpriv __retval(0xabcdef12)
+__naked void value_ptr_known_scalar_5(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = %[__imm_0];				\
+	r1 += r0;					\
+	r0 = *(u32*)(r1 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_const(__imm_0, (6 + 1) * sizeof(int))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += known scalar, 6")
+__success __success_unpriv __retval(0xabcdef12)
+__naked void value_ptr_known_scalar_6(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = %[__imm_0];				\
+	r0 += r1;					\
+	r1 = %[__imm_1];				\
+	r0 += r1;					\
+	r0 = *(u32*)(r0 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b),
+	  __imm_const(__imm_0, (3 + 1) * sizeof(int)),
+	  __imm_const(__imm_1, 3 * sizeof(int))
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += N, value_ptr -= N known scalar")
+__success __success_unpriv __retval(0x12345678)
+__naked void value_ptr_n_known_scalar(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	w1 = 0x12345678;				\
+	*(u32*)(r0 + 0) = r1;				\
+	r0 += 2;					\
+	r1 = 2;						\
+	r0 -= r1;					\
+	r0 = *(u32*)(r0 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: unknown scalar += value_ptr, 1")
+__success __success_unpriv __retval(1)
+__naked void unknown_scalar_value_ptr_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= 0xf;					\
+	r1 += r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: unknown scalar += value_ptr, 2")
+__success __success_unpriv __retval(0xabcdef12) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void unknown_scalar_value_ptr_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	r1 &= 31;					\
+	r1 += r0;					\
+	r0 = *(u32*)(r1 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: unknown scalar += value_ptr, 3")
+__success __failure_unpriv
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__retval(0xabcdef12) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void unknown_scalar_value_ptr_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = -1;					\
+	r0 += r1;					\
+	r1 = 1;						\
+	r0 += r1;					\
+	r1 = *(u32*)(r0 + 0);				\
+	r1 &= 31;					\
+	r1 += r0;					\
+	r0 = *(u32*)(r1 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: unknown scalar += value_ptr, 4")
+__failure __msg("R1 max value is outside of the allowed memory range")
+__msg_unpriv("R1 pointer arithmetic of map value goes out of range")
+__flag(BPF_F_ANY_ALIGNMENT)
+__naked void unknown_scalar_value_ptr_4(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 19;					\
+	r0 += r1;					\
+	r1 = *(u32*)(r0 + 0);				\
+	r1 &= 31;					\
+	r1 += r0;					\
+	r0 = *(u32*)(r1 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += unknown scalar, 1")
+__success __success_unpriv __retval(1)
+__naked void value_ptr_unknown_scalar_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= 0xf;					\
+	r0 += r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += unknown scalar, 2")
+__success __success_unpriv __retval(0xabcdef12) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void value_ptr_unknown_scalar_2_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u32*)(r0 + 0);				\
+	r1 &= 31;					\
+	r0 += r1;					\
+	r0 = *(u32*)(r0 + 0);				\
+l0_%=:	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += unknown scalar, 3")
+__success __success_unpriv __retval(1)
+__naked void value_ptr_unknown_scalar_3(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u64*)(r0 + 0);				\
+	r2 = *(u64*)(r0 + 8);				\
+	r3 = *(u64*)(r0 + 16);				\
+	r1 &= 0xf;					\
+	r3 &= 1;					\
+	r3 |= 1;					\
+	if r2 > r3 goto l0_%=;				\
+	r0 += r3;					\
+	r0 = *(u8*)(r0 + 0);				\
+	r0 = 1;						\
+l1_%=:	exit;						\
+l0_%=:	r0 = 2;						\
+	goto l1_%=;					\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr += value_ptr")
+__failure __msg("R0 pointer += pointer prohibited")
+__failure_unpriv
+__naked void access_value_ptr_value_ptr_1(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 += r0;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: known scalar -= value_ptr")
+__failure __msg("R1 tried to subtract pointer from scalar")
+__failure_unpriv
+__naked void access_known_scalar_value_ptr_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 4;						\
+	r1 -= r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= known scalar")
+__failure __msg("R0 min value is outside of the allowed memory range")
+__failure_unpriv
+__naked void access_value_ptr_known_scalar(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 4;						\
+	r0 -= r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= known scalar, 2")
+__success __success_unpriv __retval(1)
+__naked void value_ptr_known_scalar_2_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = 6;						\
+	r2 = 4;						\
+	r0 += r1;					\
+	r0 -= r2;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: unknown scalar -= value_ptr")
+__failure __msg("R1 tried to subtract pointer from scalar")
+__failure_unpriv
+__naked void access_unknown_scalar_value_ptr(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= 0xf;					\
+	r1 -= r0;					\
+	r0 = *(u8*)(r1 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= unknown scalar")
+__failure __msg("R0 min value is negative")
+__failure_unpriv
+__naked void access_value_ptr_unknown_scalar(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= 0xf;					\
+	r0 -= r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= unknown scalar, 2")
+__success __failure_unpriv
+__msg_unpriv("R0 pointer arithmetic of map value goes out of range")
+__retval(1)
+__naked void value_ptr_unknown_scalar_2_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= 0xf;					\
+	r1 |= 0x7;					\
+	r0 += r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+	r1 &= 0x7;					\
+	r0 -= r1;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: value_ptr -= value_ptr")
+__failure __msg("R0 invalid mem access 'scalar'")
+__msg_unpriv("R0 pointer -= pointer prohibited")
+__naked void access_value_ptr_value_ptr_2(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 == 0 goto l0_%=;				\
+	r0 -= r0;					\
+	r1 = *(u8*)(r0 + 0);				\
+l0_%=:	r0 = 1;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("map access: trying to leak tainted dst reg")
+__failure __msg("math between map_value pointer and 4294967295 is not allowed")
+__failure_unpriv
+__naked void to_leak_tainted_dst_reg(void)
+{
+	asm volatile ("					\
+	r0 = 0;						\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_array_48b] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	if r0 != 0 goto l0_%=;				\
+	exit;						\
+l0_%=:	r2 = r0;					\
+	w1 = 0xFFFFFFFF;				\
+	w1 = w1;					\
+	r2 -= r1;					\
+	*(u64*)(r0 + 0) = r2;				\
+	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_array_48b)
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("32bit pkt_ptr -= scalar")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void _32bit_pkt_ptr_scalar(void)
+{
+	asm volatile ("					\
+	r8 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r7 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r6 = r7;					\
+	r6 += 40;					\
+	if r6 > r8 goto l0_%=;				\
+	w4 = w7;					\
+	w6 -= w4;					\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+SEC("tc")
+__description("32bit scalar -= pkt_ptr")
+__success __retval(0) __flag(BPF_F_ANY_ALIGNMENT)
+__naked void _32bit_scalar_pkt_ptr(void)
+{
+	asm volatile ("					\
+	r8 = *(u32*)(r1 + %[__sk_buff_data_end]);	\
+	r7 = *(u32*)(r1 + %[__sk_buff_data]);		\
+	r6 = r7;					\
+	r6 += 40;					\
+	if r6 > r8 goto l0_%=;				\
+	w4 = w6;					\
+	w4 -= w7;					\
+l0_%=:	r0 = 0;						\
+	exit;						\
+"	:
+	: __imm_const(__sk_buff_data, offsetof(struct __sk_buff, data)),
+	  __imm_const(__sk_buff_data_end, offsetof(struct __sk_buff, data_end))
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/value_ptr_arith.c b/tools/testing/selftests/bpf/verifier/value_ptr_arith.c
deleted file mode 100644
index 249187d3c530..000000000000
--- a/tools/testing/selftests/bpf/verifier/value_ptr_arith.c
+++ /dev/null
@@ -1,1140 +0,0 @@
-{
-	"map access: known scalar += value_ptr unknown vs const",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_1, 3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr const vs unknown",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 2),
-	BPF_MOV64_IMM(BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr const vs const (ne)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 2),
-	BPF_MOV64_IMM(BPF_REG_1, 3),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_1, 5),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr const vs const (eq)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 2),
-	BPF_MOV64_IMM(BPF_REG_1, 5),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 1),
-	BPF_MOV64_IMM(BPF_REG_1, 5),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr unknown vs unknown (eq)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr unknown vs unknown (lt)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x3),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr unknown vs unknown (gt)",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_JMP_IMM(BPF_JA, 0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x3),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr from different maps",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr -= known scalar from different maps",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_16b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 min value is outside of the allowed memory range",
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr from different maps, but same value properties",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
-		    offsetof(struct __sk_buff, len)),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_hash_48b = { 5 },
-	.fixup_map_array_48b = { 8 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: mixing value pointer and scalar, 1",
-	.insns = {
-	// load map value pointer into r0 and r2
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_LD_MAP_FD(BPF_REG_ARG1, 0),
-	BPF_MOV64_REG(BPF_REG_ARG2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_ARG2, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_FP, -16, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	// load some number from the map into r1
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	// depending on r1, branch:
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 3),
-	// branch A
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	BPF_JMP_A(2),
-	// branch B
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0x100000),
-	// common instruction
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	// depending on r1, branch:
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	// branch A
-	BPF_JMP_A(4),
-	// branch B
-	BPF_MOV64_IMM(BPF_REG_0, 0x13371337),
-	// verifier follows fall-through
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_2, 0x100000, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	// fake-dead code; targeted from branch A to
-	// prevent dead code sanitization
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R2 pointer comparison prohibited",
-	.retval = 0,
-},
-{
-	"map access: mixing value pointer and scalar, 2",
-	.insns = {
-	// load map value pointer into r0 and r2
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_LD_MAP_FD(BPF_REG_ARG1, 0),
-	BPF_MOV64_REG(BPF_REG_ARG2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_ARG2, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_FP, -16, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	// load some number from the map into r1
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	// depending on r1, branch:
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
-	// branch A
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0x100000),
-	BPF_JMP_A(2),
-	// branch B
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_3, 0),
-	// common instruction
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	// depending on r1, branch:
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 1),
-	// branch A
-	BPF_JMP_A(4),
-	// branch B
-	BPF_MOV64_IMM(BPF_REG_0, 0x13371337),
-	// verifier follows fall-through
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_2, 0x100000, 2),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	// fake-dead code; targeted from branch A to
-	// prevent dead code sanitization, rejected
-	// via branch B however
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 invalid mem access 'scalar'",
-	.retval = 0,
-},
-{
-	"sanitation: alu with different scalars 1",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_LD_MAP_FD(BPF_REG_ARG1, 0),
-	BPF_MOV64_REG(BPF_REG_ARG2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_ARG2, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_FP, -16, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_3, 0x100000),
-	BPF_JMP_A(2),
-	BPF_MOV64_IMM(BPF_REG_2, 42),
-	BPF_MOV64_IMM(BPF_REG_3, 0x100001),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_2, BPF_REG_3),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result = ACCEPT,
-	.retval = 0x100000,
-},
-{
-	"sanitation: alu with different scalars 2",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -16),
-	BPF_ST_MEM(BPF_DW, BPF_REG_FP, -16, 0),
-	BPF_EMIT_CALL(BPF_FUNC_map_delete_elem),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_FP),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -16),
-	BPF_EMIT_CALL(BPF_FUNC_map_delete_elem),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_6),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_8, BPF_REG_7),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_8),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 1 },
-	.result = ACCEPT,
-	.retval = -EINVAL * 2,
-},
-{
-	"sanitation: alu with different scalars 3",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, EINVAL),
-	BPF_ALU64_IMM(BPF_MUL, BPF_REG_0, -1),
-	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
-	BPF_MOV64_IMM(BPF_REG_0, EINVAL),
-	BPF_ALU64_IMM(BPF_MUL, BPF_REG_0, -1),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_0),
-	BPF_MOV64_REG(BPF_REG_8, BPF_REG_6),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_8, BPF_REG_7),
-	BPF_MOV64_REG(BPF_REG_0, BPF_REG_8),
-	BPF_EXIT_INSN(),
-	},
-	.result = ACCEPT,
-	.retval = -EINVAL * 2,
-},
-{
-	"map access: value_ptr += known scalar, upper oob arith, test 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 48),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.retval = 1,
-},
-{
-	"map access: value_ptr += known scalar, upper oob arith, test 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 49),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.retval = 1,
-},
-{
-	"map access: value_ptr += known scalar, upper oob arith, test 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_MOV64_IMM(BPF_REG_1, 47),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr -= known scalar, lower oob arith, test 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_IMM(BPF_REG_1, 47),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 48),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R0 min value is outside of the allowed memory range",
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-},
-{
-	"map access: value_ptr -= known scalar, lower oob arith, test 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_IMM(BPF_REG_1, 47),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 48),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.retval = 1,
-},
-{
-	"map access: value_ptr -= known scalar, lower oob arith, test 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_IMM(BPF_REG_1, 47),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 47),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: known scalar += value_ptr",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr += known scalar, 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr += known scalar, 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 49),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "invalid access to map value",
-},
-{
-	"map access: value_ptr += known scalar, 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "invalid access to map value",
-},
-{
-	"map access: value_ptr += known scalar, 4",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
-	BPF_MOV64_IMM(BPF_REG_1, 5),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, -2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr += known scalar, 5",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, (6 + 1) * sizeof(int)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 0xabcdef12,
-},
-{
-	"map access: value_ptr += known scalar, 6",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_IMM(BPF_REG_1, (3 + 1) * sizeof(int)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 3 * sizeof(int)),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 0xabcdef12,
-},
-{
-	"map access: value_ptr += N, value_ptr -= N known scalar",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV32_IMM(BPF_REG_1, 0x12345678),
-	BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 2),
-	BPF_MOV64_IMM(BPF_REG_1, 2),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 0x12345678,
-},
-{
-	"map access: unknown scalar += value_ptr, 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xf),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: unknown scalar += value_ptr, 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 31),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 0xabcdef12,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map access: unknown scalar += value_ptr, 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_MOV64_IMM(BPF_REG_1, -1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_MOV64_IMM(BPF_REG_1, 1),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 31),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.retval = 0xabcdef12,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map access: unknown scalar += value_ptr, 4",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 6),
-	BPF_MOV64_IMM(BPF_REG_1, 19),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 31),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 max value is outside of the allowed memory range",
-	.errstr_unpriv = "R1 pointer arithmetic of map value goes out of range",
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map access: value_ptr += unknown scalar, 1",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xf),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr += unknown scalar, 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 31),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 0xabcdef12,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"map access: value_ptr += unknown scalar, 3",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_0, 0),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 8),
-	BPF_LDX_MEM(BPF_DW, BPF_REG_3, BPF_REG_0, 16),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xf),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_3, 1),
-	BPF_ALU64_IMM(BPF_OR, BPF_REG_3, 1),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_2, BPF_REG_3, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_3),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_IMM(BPF_REG_0, 2),
-	BPF_JMP_IMM(BPF_JA, 0, 0, -3),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: value_ptr += value_ptr",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R0 pointer += pointer prohibited",
-},
-{
-	"map access: known scalar -= value_ptr",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 tried to subtract pointer from scalar",
-},
-{
-	"map access: value_ptr -= known scalar",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 3),
-	BPF_MOV64_IMM(BPF_REG_1, 4),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R0 min value is outside of the allowed memory range",
-},
-{
-	"map access: value_ptr -= known scalar, 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 5),
-	BPF_MOV64_IMM(BPF_REG_1, 6),
-	BPF_MOV64_IMM(BPF_REG_2, 4),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_2),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.retval = 1,
-},
-{
-	"map access: unknown scalar -= value_ptr",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xf),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_1, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R1 tried to subtract pointer from scalar",
-},
-{
-	"map access: value_ptr -= unknown scalar",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 4),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xf),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R0 min value is negative",
-},
-{
-	"map access: value_ptr -= unknown scalar, 2",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 8),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0xf),
-	BPF_ALU64_IMM(BPF_OR, BPF_REG_1, 0x7),
-	BPF_ALU64_REG(BPF_ADD, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_1),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
-	.retval = 1,
-},
-{
-	"map access: value_ptr -= value_ptr",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_0, BPF_REG_0),
-	BPF_LDX_MEM(BPF_B, BPF_REG_1, BPF_REG_0, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 1),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 3 },
-	.result = REJECT,
-	.errstr = "R0 invalid mem access 'scalar'",
-	.errstr_unpriv = "R0 pointer -= pointer prohibited",
-},
-{
-	"map access: trying to leak tainted dst reg",
-	.insns = {
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
-	BPF_EXIT_INSN(),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_0),
-	BPF_MOV32_IMM(BPF_REG_1, 0xFFFFFFFF),
-	BPF_MOV32_REG(BPF_REG_1, BPF_REG_1),
-	BPF_ALU64_REG(BPF_SUB, BPF_REG_2, BPF_REG_1),
-	BPF_STX_MEM(BPF_DW, BPF_REG_0, BPF_REG_2, 0),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_array_48b = { 4 },
-	.result = REJECT,
-	.errstr = "math between map_value pointer and 4294967295 is not allowed",
-},
-{
-	"32bit pkt_ptr -= scalar",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 40),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_8, 2),
-	BPF_ALU32_REG(BPF_MOV, BPF_REG_4, BPF_REG_7),
-	BPF_ALU32_REG(BPF_SUB, BPF_REG_6, BPF_REG_4),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-{
-	"32bit scalar -= pkt_ptr",
-	.insns = {
-	BPF_LDX_MEM(BPF_W, BPF_REG_8, BPF_REG_1,
-		    offsetof(struct __sk_buff, data_end)),
-	BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
-		    offsetof(struct __sk_buff, data)),
-	BPF_MOV64_REG(BPF_REG_6, BPF_REG_7),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_6, 40),
-	BPF_JMP_REG(BPF_JGT, BPF_REG_6, BPF_REG_8, 2),
-	BPF_ALU32_REG(BPF_MOV, BPF_REG_4, BPF_REG_6),
-	BPF_ALU32_REG(BPF_SUB, BPF_REG_4, BPF_REG_7),
-	BPF_MOV64_IMM(BPF_REG_0, 0),
-	BPF_EXIT_INSN(),
-	},
-	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
-	.result = ACCEPT,
-	.flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS,
-},
-- 
cgit v1.2.3-70-g09d2


From e38fbfa972eb0944ce4387caf23d166c1df4b114 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 11 Apr 2023 16:29:45 +0200
Subject: netfilter: nf_tables: merge nft_rules_old structure and end of
 ruleblob marker

In order to free the rules in a chain via call_rcu, the rule array used
to stash a rcu_head and space for a pointer at the end of the rule array.

When the current nft_rule_dp blob format got added in
2c865a8a28a1 ("netfilter: nf_tables: add rule blob layout"), this results
in a double-trailer:

  size (unsigned long)
  struct nft_rule_dp
    struct nft_expr
         ...
    struct nft_rule_dp
     struct nft_expr
         ...
    struct nft_rule_dp (is_last=1) // Trailer

The trailer, struct nft_rule_dp (is_last=1), is not accounted for in size,
so it can be located via start_addr + size.

Because the rcu_head is stored after 'start+size' as well this means the
is_last trailer is *aliased* to the rcu_head (struct nft_rules_old).

This is harmless, because at this time the nft_do_chain function never
evaluates/accesses the trailer, it only checks the address boundary:

        for (; rule < last_rule; rule = nft_rule_next(rule)) {
...

But this way the last_rule address has to be stashed in the jump
structure to restore it after returning from a chain.

nft_do_chain stack usage has become way too big, so put it on a diet.

Without this patch is impossible to use
        for (; !rule->is_last; rule = nft_rule_next(rule)) {

... because on free, the needed update of the rcu_head will clobber the
nft_rule_dp is_last bit.

Furthermore, also stash the chain pointer in the trailer, this allows
to recover the original chain structure from nf_tables_trace infra
without a need to place them in the jump struct.

After this patch it is trivial to diet the jump stack structure,
done in the next two patches.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 55 +++++++++++++++++++++----------------------
 1 file changed, 27 insertions(+), 28 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index e48ab8dfb541..79848a27e640 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -2110,38 +2110,41 @@ static void nft_chain_release_hook(struct nft_chain_hook *hook)
 	module_put(hook->type->owner);
 }
 
-struct nft_rules_old {
+struct nft_rule_dp_last {
+	struct nft_rule_dp end;	/* end of nft_rule_blob marker */
 	struct rcu_head h;
 	struct nft_rule_blob *blob;
+	const struct nft_chain *chain;	/* for tracing */
 };
 
-static void nft_last_rule(struct nft_rule_blob *blob, const void *ptr)
+static void nft_last_rule(const struct nft_chain *chain, const void *ptr)
 {
-	struct nft_rule_dp *prule;
+	struct nft_rule_dp_last *lrule;
+
+	BUILD_BUG_ON(offsetof(struct nft_rule_dp_last, end) != 0);
 
-	prule = (struct nft_rule_dp *)ptr;
-	prule->is_last = 1;
+	lrule = (struct nft_rule_dp_last *)ptr;
+	lrule->end.is_last = 1;
+	lrule->chain = chain;
 	/* blob size does not include the trailer rule */
 }
 
-static struct nft_rule_blob *nf_tables_chain_alloc_rules(unsigned int size)
+static struct nft_rule_blob *nf_tables_chain_alloc_rules(const struct nft_chain *chain,
+							 unsigned int size)
 {
 	struct nft_rule_blob *blob;
 
-	/* size must include room for the last rule */
-	if (size < offsetof(struct nft_rule_dp, data))
-		return NULL;
-
-	size += sizeof(struct nft_rule_blob) + sizeof(struct nft_rules_old);
 	if (size > INT_MAX)
 		return NULL;
 
+	size += sizeof(struct nft_rule_blob) + sizeof(struct nft_rule_dp_last);
+
 	blob = kvmalloc(size, GFP_KERNEL_ACCOUNT);
 	if (!blob)
 		return NULL;
 
 	blob->size = 0;
-	nft_last_rule(blob, blob->data);
+	nft_last_rule(chain, blob->data);
 
 	return blob;
 }
@@ -2220,7 +2223,6 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask,
 	struct nft_rule_blob *blob;
 	struct nft_trans *trans;
 	struct nft_chain *chain;
-	unsigned int data_size;
 	int err;
 
 	if (table->use == UINT_MAX)
@@ -2308,8 +2310,7 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask,
 		chain->udlen = nla_len(nla[NFTA_CHAIN_USERDATA]);
 	}
 
-	data_size = offsetof(struct nft_rule_dp, data);	/* last rule */
-	blob = nf_tables_chain_alloc_rules(data_size);
+	blob = nf_tables_chain_alloc_rules(chain, 0);
 	if (!blob) {
 		err = -ENOMEM;
 		goto err_destroy_chain;
@@ -8817,9 +8818,8 @@ static int nf_tables_commit_chain_prepare(struct net *net, struct nft_chain *cha
 				return -ENOMEM;
 		}
 	}
-	data_size += offsetof(struct nft_rule_dp, data);	/* last rule */
 
-	chain->blob_next = nf_tables_chain_alloc_rules(data_size);
+	chain->blob_next = nf_tables_chain_alloc_rules(chain, data_size);
 	if (!chain->blob_next)
 		return -ENOMEM;
 
@@ -8864,12 +8864,11 @@ static int nf_tables_commit_chain_prepare(struct net *net, struct nft_chain *cha
 		chain->blob_next->size += (unsigned long)(data - (void *)prule);
 	}
 
-	prule = (struct nft_rule_dp *)data;
-	data += offsetof(struct nft_rule_dp, data);
 	if (WARN_ON_ONCE(data > data_boundary))
 		return -ENOMEM;
 
-	nft_last_rule(chain->blob_next, prule);
+	prule = (struct nft_rule_dp *)data;
+	nft_last_rule(chain, prule);
 
 	return 0;
 }
@@ -8890,22 +8889,22 @@ static void nf_tables_commit_chain_prepare_cancel(struct net *net)
 	}
 }
 
-static void __nf_tables_commit_chain_free_rules_old(struct rcu_head *h)
+static void __nf_tables_commit_chain_free_rules(struct rcu_head *h)
 {
-	struct nft_rules_old *o = container_of(h, struct nft_rules_old, h);
+	struct nft_rule_dp_last *l = container_of(h, struct nft_rule_dp_last, h);
 
-	kvfree(o->blob);
+	kvfree(l->blob);
 }
 
 static void nf_tables_commit_chain_free_rules_old(struct nft_rule_blob *blob)
 {
-	struct nft_rules_old *old;
+	struct nft_rule_dp_last *last;
 
-	/* rcu_head is after end marker */
-	old = (void *)blob + sizeof(*blob) + blob->size;
-	old->blob = blob;
+	/* last rule trailer is after end marker */
+	last = (void *)blob + sizeof(*blob) + blob->size;
+	last->blob = blob;
 
-	call_rcu(&old->h, __nf_tables_commit_chain_free_rules_old);
+	call_rcu(&last->h, __nf_tables_commit_chain_free_rules);
 }
 
 static void nf_tables_commit_chain(struct net *net, struct nft_chain *chain)
-- 
cgit v1.2.3-70-g09d2


From d4d89e6546e0d1ac09cb9dd353f0cb31c8a8deb1 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 11 Apr 2023 16:29:46 +0200
Subject: netfilter: nf_tables: don't store address of last rule on jump

Walk the rule headers until the trailer one (last_bit flag set) instead
of stopping at last_rule address.

This avoids the need to store the address when jumping to another chain.

This cuts size of jumpstack array by one third, on 64bit from
384 to 256 bytes.  Still, stack usage is still quite large:

scripts/stackusage:
nf_tables_core.c:258 nft_do_chain    496     static

Next patch will also remove chain pointer.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_core.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index 6ecd0ba2e546..ec3bab751092 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -205,7 +205,6 @@ static noinline void nft_update_chain_stats(const struct nft_chain *chain,
 struct nft_jumpstack {
 	const struct nft_chain *chain;
 	const struct nft_rule_dp *rule;
-	const struct nft_rule_dp *last_rule;
 };
 
 static void expr_call_ops_eval(const struct nft_expr *expr,
@@ -259,9 +258,9 @@ unsigned int
 nft_do_chain(struct nft_pktinfo *pkt, void *priv)
 {
 	const struct nft_chain *chain = priv, *basechain = chain;
-	const struct nft_rule_dp *rule, *last_rule;
 	const struct net *net = nft_net(pkt);
 	const struct nft_expr *expr, *last;
+	const struct nft_rule_dp *rule;
 	struct nft_regs regs = {};
 	unsigned int stackptr = 0;
 	struct nft_jumpstack jumpstack[NFT_JUMP_STACK_SIZE];
@@ -279,10 +278,9 @@ do_chain:
 		blob = rcu_dereference(chain->blob_gen_0);
 
 	rule = (struct nft_rule_dp *)blob->data;
-	last_rule = (void *)blob->data + blob->size;
 next_rule:
 	regs.verdict.code = NFT_CONTINUE;
-	for (; rule < last_rule; rule = nft_rule_next(rule)) {
+	for (; !rule->is_last ; rule = nft_rule_next(rule)) {
 		nft_rule_dp_for_each_expr(expr, last, rule) {
 			if (expr->ops == &nft_cmp_fast_ops)
 				nft_cmp_fast_eval(expr, &regs);
@@ -327,7 +325,6 @@ next_rule:
 			return NF_DROP;
 		jumpstack[stackptr].chain = chain;
 		jumpstack[stackptr].rule = nft_rule_next(rule);
-		jumpstack[stackptr].last_rule = last_rule;
 		stackptr++;
 		fallthrough;
 	case NFT_GOTO:
@@ -344,7 +341,6 @@ next_rule:
 		stackptr--;
 		chain = jumpstack[stackptr].chain;
 		rule = jumpstack[stackptr].rule;
-		last_rule = jumpstack[stackptr].last_rule;
 		goto next_rule;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 63e9bbbcca60333490e13744ae736d8f988e4950 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 11 Apr 2023 16:29:47 +0200
Subject: netfilter: nf_tables: don't store chain address on jump

Now that the rule trailer/end marker and the rcu head reside in the
same structure, we no longer need to save/restore the chain pointer
when performing/returning from a jump.

We can simply let the trace infra walk the evaluated rule until it
hits the end marker and then fetch the chain pointer from there.

When the rule is NULL (policy tracing), then chain and basechain
pointers were already identical, so just use the basechain.

This cuts size of jumpstack in half, from 256 to 128 bytes in 64bit,
scripts/stackusage says:

nf_tables_core.c:251 nft_do_chain    328     static

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h | 14 ++++++++++++--
 net/netfilter/nf_tables_api.c     |  7 -------
 net/netfilter/nf_tables_core.c    | 21 ++++++---------------
 net/netfilter/nf_tables_trace.c   | 30 ++++++++++++++++++++++++++----
 4 files changed, 44 insertions(+), 28 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 1b8e305bb54a..f476fd030626 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1046,6 +1046,18 @@ struct nft_rule_dp {
 		__attribute__((aligned(__alignof__(struct nft_expr))));
 };
 
+struct nft_rule_dp_last {
+	struct nft_rule_dp end;		/* end of nft_rule_blob marker */
+	struct rcu_head h;		/* call_rcu head */
+	struct nft_rule_blob *blob;	/* ptr to free via call_rcu */
+	const struct nft_chain *chain;	/* for nftables tracing */
+};
+
+static inline const struct nft_rule_dp *nft_rule_next(const struct nft_rule_dp *rule)
+{
+	return (void *)rule + sizeof(*rule) + rule->dlen;
+}
+
 struct nft_rule_blob {
 	unsigned long			size;
 	unsigned char			data[]
@@ -1396,7 +1408,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type);
  *	@packet_dumped: packet headers sent in a previous traceinfo message
  *	@pkt: pktinfo currently processed
  *	@basechain: base chain currently processed
- *	@chain: chain currently processed
  *	@rule:  rule that was evaluated
  *	@verdict: verdict given by rule
  */
@@ -1408,7 +1419,6 @@ struct nft_traceinfo {
 	u32				skbid;
 	const struct nft_pktinfo	*pkt;
 	const struct nft_base_chain	*basechain;
-	const struct nft_chain		*chain;
 	const struct nft_rule_dp	*rule;
 	const struct nft_verdict	*verdict;
 };
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 79848a27e640..0e1c86bb51a2 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -2110,13 +2110,6 @@ static void nft_chain_release_hook(struct nft_chain_hook *hook)
 	module_put(hook->type->owner);
 }
 
-struct nft_rule_dp_last {
-	struct nft_rule_dp end;	/* end of nft_rule_blob marker */
-	struct rcu_head h;
-	struct nft_rule_blob *blob;
-	const struct nft_chain *chain;	/* for tracing */
-};
-
 static void nft_last_rule(const struct nft_chain *chain, const void *ptr)
 {
 	struct nft_rule_dp_last *lrule;
diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index ec3bab751092..89c05b64c2a2 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -42,13 +42,11 @@ static inline void nf_skip_indirect_calls_enable(void) { }
 #endif
 
 static noinline void __nft_trace_packet(struct nft_traceinfo *info,
-					const struct nft_chain *chain,
 					enum nft_trace_types type)
 {
 	if (!info->trace || !info->nf_trace)
 		return;
 
-	info->chain = chain;
 	info->type = type;
 
 	nft_trace_notify(info);
@@ -56,14 +54,13 @@ static noinline void __nft_trace_packet(struct nft_traceinfo *info,
 
 static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
 				    struct nft_traceinfo *info,
-				    const struct nft_chain *chain,
 				    const struct nft_rule_dp *rule,
 				    enum nft_trace_types type)
 {
 	if (static_branch_unlikely(&nft_trace_enabled)) {
 		info->nf_trace = pkt->skb->nf_trace;
 		info->rule = rule;
-		__nft_trace_packet(info, chain, type);
+		__nft_trace_packet(info, type);
 	}
 }
 
@@ -111,7 +108,6 @@ static void nft_cmp16_fast_eval(const struct nft_expr *expr,
 }
 
 static noinline void __nft_trace_verdict(struct nft_traceinfo *info,
-					 const struct nft_chain *chain,
 					 const struct nft_regs *regs)
 {
 	enum nft_trace_types type;
@@ -133,17 +129,16 @@ static noinline void __nft_trace_verdict(struct nft_traceinfo *info,
 		break;
 	}
 
-	__nft_trace_packet(info, chain, type);
+	__nft_trace_packet(info, type);
 }
 
 static inline void nft_trace_verdict(struct nft_traceinfo *info,
-				     const struct nft_chain *chain,
 				     const struct nft_rule_dp *rule,
 				     const struct nft_regs *regs)
 {
 	if (static_branch_unlikely(&nft_trace_enabled)) {
 		info->rule = rule;
-		__nft_trace_verdict(info, chain, regs);
+		__nft_trace_verdict(info, regs);
 	}
 }
 
@@ -203,7 +198,6 @@ static noinline void nft_update_chain_stats(const struct nft_chain *chain,
 }
 
 struct nft_jumpstack {
-	const struct nft_chain *chain;
 	const struct nft_rule_dp *rule;
 };
 
@@ -247,7 +241,6 @@ indirect_call:
 #define nft_rule_expr_first(rule)	(struct nft_expr *)&rule->data[0]
 #define nft_rule_expr_next(expr)	((void *)expr) + expr->ops->size
 #define nft_rule_expr_last(rule)	(struct nft_expr *)&rule->data[rule->dlen]
-#define nft_rule_next(rule)		(void *)rule + sizeof(*rule) + rule->dlen
 
 #define nft_rule_dp_for_each_expr(expr, last, rule) \
         for ((expr) = nft_rule_expr_first(rule), (last) = nft_rule_expr_last(rule); \
@@ -302,14 +295,14 @@ next_rule:
 			nft_trace_copy_nftrace(pkt, &info);
 			continue;
 		case NFT_CONTINUE:
-			nft_trace_packet(pkt, &info, chain, rule,
+			nft_trace_packet(pkt, &info, rule,
 					 NFT_TRACETYPE_RULE);
 			continue;
 		}
 		break;
 	}
 
-	nft_trace_verdict(&info, chain, rule, &regs);
+	nft_trace_verdict(&info, rule, &regs);
 
 	switch (regs.verdict.code & NF_VERDICT_MASK) {
 	case NF_ACCEPT:
@@ -323,7 +316,6 @@ next_rule:
 	case NFT_JUMP:
 		if (WARN_ON_ONCE(stackptr >= NFT_JUMP_STACK_SIZE))
 			return NF_DROP;
-		jumpstack[stackptr].chain = chain;
 		jumpstack[stackptr].rule = nft_rule_next(rule);
 		stackptr++;
 		fallthrough;
@@ -339,12 +331,11 @@ next_rule:
 
 	if (stackptr > 0) {
 		stackptr--;
-		chain = jumpstack[stackptr].chain;
 		rule = jumpstack[stackptr].rule;
 		goto next_rule;
 	}
 
-	nft_trace_packet(pkt, &info, basechain, NULL, NFT_TRACETYPE_POLICY);
+	nft_trace_packet(pkt, &info, NULL, NFT_TRACETYPE_POLICY);
 
 	if (static_branch_unlikely(&nft_counters_enabled))
 		nft_update_chain_stats(basechain, pkt);
diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c
index 1163ba9c1401..3d9b83d84a84 100644
--- a/net/netfilter/nf_tables_trace.c
+++ b/net/netfilter/nf_tables_trace.c
@@ -164,9 +164,29 @@ static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info)
 	return true;
 }
 
+static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *info)
+{
+	const struct nft_rule_dp *rule = info->rule;
+	const struct nft_rule_dp_last *last;
+
+	if (!rule)
+		return &info->basechain->chain;
+
+	while (!rule->is_last)
+		rule = nft_rule_next(rule);
+
+	last = (const struct nft_rule_dp_last *)rule;
+
+	if (WARN_ON_ONCE(!last->chain))
+		return &info->basechain->chain;
+
+	return last->chain;
+}
+
 void nft_trace_notify(struct nft_traceinfo *info)
 {
 	const struct nft_pktinfo *pkt = info->pkt;
+	const struct nft_chain *chain;
 	struct nlmsghdr *nlh;
 	struct sk_buff *skb;
 	unsigned int size;
@@ -176,9 +196,11 @@ void nft_trace_notify(struct nft_traceinfo *info)
 	if (!nfnetlink_has_listeners(nft_net(pkt), NFNLGRP_NFTRACE))
 		return;
 
+	chain = nft_trace_get_chain(info);
+
 	size = nlmsg_total_size(sizeof(struct nfgenmsg)) +
-		nla_total_size(strlen(info->chain->table->name)) +
-		nla_total_size(strlen(info->chain->name)) +
+		nla_total_size(strlen(chain->table->name)) +
+		nla_total_size(strlen(chain->name)) +
 		nla_total_size_64bit(sizeof(__be64)) +	/* rule handle */
 		nla_total_size(sizeof(__be32)) +	/* trace type */
 		nla_total_size(0) +			/* VERDICT, nested */
@@ -217,10 +239,10 @@ void nft_trace_notify(struct nft_traceinfo *info)
 	if (nla_put_u32(skb, NFTA_TRACE_ID, info->skbid))
 		goto nla_put_failure;
 
-	if (nla_put_string(skb, NFTA_TRACE_CHAIN, info->chain->name))
+	if (nla_put_string(skb, NFTA_TRACE_CHAIN, chain->name))
 		goto nla_put_failure;
 
-	if (nla_put_string(skb, NFTA_TRACE_TABLE, info->chain->table->name))
+	if (nla_put_string(skb, NFTA_TRACE_TABLE, chain->table->name))
 		goto nla_put_failure;
 
 	if (nf_trace_fill_rule_info(skb, info))
-- 
cgit v1.2.3-70-g09d2


From 9a32e9850686599ed194ccdceb6cd3dd56b2d9b9 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 13 Apr 2023 17:13:19 +0200
Subject: netfilter: nf_tables: don't write table validation state without
 mutex

The ->cleanup callback needs to be removed, this doesn't work anymore as
the transaction mutex is already released in the ->abort function.

Just do it after a successful validation pass, this either happens
from commit or abort phases where transaction mutex is held.

Fixes: f102d66b335a ("netfilter: nf_tables: use dedicated mutex to guard transactions")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/linux/netfilter/nfnetlink.h | 1 -
 net/netfilter/nf_tables_api.c       | 8 ++------
 net/netfilter/nfnetlink.c           | 2 --
 3 files changed, 2 insertions(+), 9 deletions(-)

diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h
index 241e005f290a..e9a9ab34a7cc 100644
--- a/include/linux/netfilter/nfnetlink.h
+++ b/include/linux/netfilter/nfnetlink.h
@@ -45,7 +45,6 @@ struct nfnetlink_subsystem {
 	int (*commit)(struct net *net, struct sk_buff *skb);
 	int (*abort)(struct net *net, struct sk_buff *skb,
 		     enum nfnl_abort_action action);
-	void (*cleanup)(struct net *net);
 	bool (*valid_genid)(struct net *net, u32 genid);
 };
 
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 0e1c86bb51a2..21eb273d2740 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -8639,6 +8639,8 @@ static int nf_tables_validate(struct net *net)
 			if (nft_table_validate(net, table) < 0)
 				return -EAGAIN;
 		}
+
+		nft_validate_state_update(net, NFT_VALIDATE_SKIP);
 		break;
 	}
 
@@ -9578,11 +9580,6 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 	return 0;
 }
 
-static void nf_tables_cleanup(struct net *net)
-{
-	nft_validate_state_update(net, NFT_VALIDATE_SKIP);
-}
-
 static int nf_tables_abort(struct net *net, struct sk_buff *skb,
 			   enum nfnl_abort_action action)
 {
@@ -9616,7 +9613,6 @@ static const struct nfnetlink_subsystem nf_tables_subsys = {
 	.cb		= nf_tables_cb,
 	.commit		= nf_tables_commit,
 	.abort		= nf_tables_abort,
-	.cleanup	= nf_tables_cleanup,
 	.valid_genid	= nf_tables_valid_genid,
 	.owner		= THIS_MODULE,
 };
diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c
index 81c7737c803a..ae7146475d17 100644
--- a/net/netfilter/nfnetlink.c
+++ b/net/netfilter/nfnetlink.c
@@ -590,8 +590,6 @@ done:
 			goto replay_abort;
 		}
 	}
-	if (ss->cleanup)
-		ss->cleanup(net);
 
 	nfnl_err_deliver(&err_list, oskb);
 	kfree_skb(skb);
-- 
cgit v1.2.3-70-g09d2


From 00c320f9b75560628e840bef027a27c746706759 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 13 Apr 2023 17:13:20 +0200
Subject: netfilter: nf_tables: make validation state per table

We only need to validate tables that saw changes in the current
transaction.

The existing code revalidates all tables, but this isn't needed as
cross-table jumps are not allowed (chains have table scope).

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  3 ++-
 net/netfilter/nf_tables_api.c     | 38 ++++++++++++++++++--------------------
 2 files changed, 20 insertions(+), 21 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index f476fd030626..ec347d9cff9e 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1209,6 +1209,7 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv);
  *	@genmask: generation mask
  *	@afinfo: address family info
  *	@name: name of the table
+ *	@validate_state: internal, set when transaction adds jumps
  */
 struct nft_table {
 	struct list_head		list;
@@ -1227,6 +1228,7 @@ struct nft_table {
 	char				*name;
 	u16				udlen;
 	u8				*udata;
+	u8				validate_state;
 };
 
 static inline bool nft_table_has_owner(const struct nft_table *table)
@@ -1698,7 +1700,6 @@ struct nftables_pernet {
 	struct mutex		commit_mutex;
 	u64			table_handle;
 	unsigned int		base_seq;
-	u8			validate_state;
 };
 
 extern unsigned int nf_tables_net_id;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 21eb273d2740..44ebc5f9598e 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -102,11 +102,9 @@ static const u8 nft2audit_op[NFT_MSG_MAX] = { // enum nf_tables_msg_types
 	[NFT_MSG_DELFLOWTABLE]	= AUDIT_NFT_OP_FLOWTABLE_UNREGISTER,
 };
 
-static void nft_validate_state_update(struct net *net, u8 new_validate_state)
+static void nft_validate_state_update(struct nft_table *table, u8 new_validate_state)
 {
-	struct nftables_pernet *nft_net = nft_pernet(net);
-
-	switch (nft_net->validate_state) {
+	switch (table->validate_state) {
 	case NFT_VALIDATE_SKIP:
 		WARN_ON_ONCE(new_validate_state == NFT_VALIDATE_DO);
 		break;
@@ -117,7 +115,7 @@ static void nft_validate_state_update(struct net *net, u8 new_validate_state)
 			return;
 	}
 
-	nft_net->validate_state = new_validate_state;
+	table->validate_state = new_validate_state;
 }
 static void nf_tables_trans_destroy_work(struct work_struct *w);
 static DECLARE_WORK(trans_destroy_work, nf_tables_trans_destroy_work);
@@ -1224,6 +1222,7 @@ static int nf_tables_newtable(struct sk_buff *skb, const struct nfnl_info *info,
 	if (table == NULL)
 		goto err_kzalloc;
 
+	table->validate_state = NFT_VALIDATE_SKIP;
 	table->name = nla_strdup(attr, GFP_KERNEL_ACCOUNT);
 	if (table->name == NULL)
 		goto err_strdup;
@@ -3660,7 +3659,7 @@ static int nf_tables_newrule(struct sk_buff *skb, const struct nfnl_info *info,
 		}
 
 		if (expr_info[i].ops->validate)
-			nft_validate_state_update(net, NFT_VALIDATE_NEED);
+			nft_validate_state_update(table, NFT_VALIDATE_NEED);
 
 		expr_info[i].ops = NULL;
 		expr = nft_expr_next(expr);
@@ -3710,7 +3709,7 @@ static int nf_tables_newrule(struct sk_buff *skb, const struct nfnl_info *info,
 	if (flow)
 		nft_trans_flow_rule(trans) = flow;
 
-	if (nft_net->validate_state == NFT_VALIDATE_DO)
+	if (table->validate_state == NFT_VALIDATE_DO)
 		return nft_table_validate(net, table);
 
 	return 0;
@@ -6312,7 +6311,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 			if (desc.type == NFT_DATA_VERDICT &&
 			    (elem.data.val.verdict.code == NFT_GOTO ||
 			     elem.data.val.verdict.code == NFT_JUMP))
-				nft_validate_state_update(ctx->net,
+				nft_validate_state_update(ctx->table,
 							  NFT_VALIDATE_NEED);
 		}
 
@@ -6437,7 +6436,6 @@ static int nf_tables_newsetelem(struct sk_buff *skb,
 				const struct nfnl_info *info,
 				const struct nlattr * const nla[])
 {
-	struct nftables_pernet *nft_net = nft_pernet(info->net);
 	struct netlink_ext_ack *extack = info->extack;
 	u8 genmask = nft_genmask_next(info->net);
 	u8 family = info->nfmsg->nfgen_family;
@@ -6476,7 +6474,7 @@ static int nf_tables_newsetelem(struct sk_buff *skb,
 		}
 	}
 
-	if (nft_net->validate_state == NFT_VALIDATE_DO)
+	if (table->validate_state == NFT_VALIDATE_DO)
 		return nft_table_validate(net, table);
 
 	return 0;
@@ -8628,19 +8626,20 @@ static int nf_tables_validate(struct net *net)
 	struct nftables_pernet *nft_net = nft_pernet(net);
 	struct nft_table *table;
 
-	switch (nft_net->validate_state) {
-	case NFT_VALIDATE_SKIP:
-		break;
-	case NFT_VALIDATE_NEED:
-		nft_validate_state_update(net, NFT_VALIDATE_DO);
-		fallthrough;
-	case NFT_VALIDATE_DO:
-		list_for_each_entry(table, &nft_net->tables, list) {
+	list_for_each_entry(table, &nft_net->tables, list) {
+		switch (table->validate_state) {
+		case NFT_VALIDATE_SKIP:
+			continue;
+		case NFT_VALIDATE_NEED:
+			nft_validate_state_update(table, NFT_VALIDATE_DO);
+			fallthrough;
+		case NFT_VALIDATE_DO:
 			if (nft_table_validate(net, table) < 0)
 				return -EAGAIN;
+
+			nft_validate_state_update(table, NFT_VALIDATE_SKIP);
 		}
 
-		nft_validate_state_update(net, NFT_VALIDATE_SKIP);
 		break;
 	}
 
@@ -10355,7 +10354,6 @@ static int __net_init nf_tables_init_net(struct net *net)
 	INIT_LIST_HEAD(&nft_net->notify_list);
 	mutex_init(&nft_net->commit_mutex);
 	nft_net->base_seq = 1;
-	nft_net->validate_state = NFT_VALIDATE_SKIP;
 
 	return 0;
 }
-- 
cgit v1.2.3-70-g09d2


From 2a1d6abd7ebeafc45f2fb5fabfd53b8b3beaa170 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 14 Apr 2023 15:01:31 +0200
Subject: netfilter: nf_tables: remove unneeded conditional

This helper is inlined, so keep it as small as possible.

If the static key is true, there is only a very small chance
that info->trace is false:

1. tracing was enabled at this very moment, the static key was
   updated to active right after nft_do_table was called.

2. tracing was disabled at this very moment.
   trace->info is already false, the static key is about to
   be patched to false soon.

In both cases, no event will be sent because info->trace
is false (checked in noinline slowpath). info->nf_trace is irrelevant.

The nf_trace update is redunant in this case, but this will only
happen for short duration, when static key flips.

       text  data   bss   dec   hex filename
old:   2980   192    32  3204   c84 nf_tables_core.o
new:   2964   192    32  3188   c74i nf_tables_core.o

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index 89c05b64c2a2..bed855638050 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -67,10 +67,8 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
 static inline void nft_trace_copy_nftrace(const struct nft_pktinfo *pkt,
 					  struct nft_traceinfo *info)
 {
-	if (static_branch_unlikely(&nft_trace_enabled)) {
-		if (info->trace)
-			info->nf_trace = pkt->skb->nf_trace;
-	}
+	if (static_branch_unlikely(&nft_trace_enabled))
+		info->nf_trace = pkt->skb->nf_trace;
 }
 
 static void nft_bitwise_fast_eval(const struct nft_expr *expr,
-- 
cgit v1.2.3-70-g09d2


From 698bb828a6c20c86e30b307175be1827c071ce23 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 14 Apr 2023 15:01:32 +0200
Subject: netfilter: nf_tables: do not store pktinfo in traceinfo structure

pass it as argument.  No change in object size.

stack usage decreases by 8 byte:
 nf_tables_core.c:254  nft_do_chain       320     static

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  5 ++---
 net/netfilter/nf_tables_core.c    | 21 ++++++++++++---------
 net/netfilter/nf_tables_trace.c   |  5 ++---
 3 files changed, 16 insertions(+), 15 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index ec347d9cff9e..cab351928cd2 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1408,7 +1408,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type);
  *	@type: event type (enum nft_trace_types)
  *	@skbid: hash of skb to be used as trace id
  *	@packet_dumped: packet headers sent in a previous traceinfo message
- *	@pkt: pktinfo currently processed
  *	@basechain: base chain currently processed
  *	@rule:  rule that was evaluated
  *	@verdict: verdict given by rule
@@ -1419,7 +1418,6 @@ struct nft_traceinfo {
 	bool				packet_dumped;
 	enum nft_trace_types		type:8;
 	u32				skbid;
-	const struct nft_pktinfo	*pkt;
 	const struct nft_base_chain	*basechain;
 	const struct nft_rule_dp	*rule;
 	const struct nft_verdict	*verdict;
@@ -1429,7 +1427,8 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
 		    const struct nft_verdict *verdict,
 		    const struct nft_chain *basechain);
 
-void nft_trace_notify(struct nft_traceinfo *info);
+void nft_trace_notify(const struct nft_pktinfo *pkt,
+		      struct nft_traceinfo *info);
 
 #define MODULE_ALIAS_NFT_CHAIN(family, name) \
 	MODULE_ALIAS("nft-chain-" __stringify(family) "-" name)
diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index bed855638050..776eb2b9f632 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -41,7 +41,8 @@ static inline bool nf_skip_indirect_calls(void) { return false; }
 static inline void nf_skip_indirect_calls_enable(void) { }
 #endif
 
-static noinline void __nft_trace_packet(struct nft_traceinfo *info,
+static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt,
+					struct nft_traceinfo *info,
 					enum nft_trace_types type)
 {
 	if (!info->trace || !info->nf_trace)
@@ -49,7 +50,7 @@ static noinline void __nft_trace_packet(struct nft_traceinfo *info,
 
 	info->type = type;
 
-	nft_trace_notify(info);
+	nft_trace_notify(pkt, info);
 }
 
 static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
@@ -60,7 +61,7 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
 	if (static_branch_unlikely(&nft_trace_enabled)) {
 		info->nf_trace = pkt->skb->nf_trace;
 		info->rule = rule;
-		__nft_trace_packet(info, type);
+		__nft_trace_packet(pkt, info, type);
 	}
 }
 
@@ -105,7 +106,8 @@ static void nft_cmp16_fast_eval(const struct nft_expr *expr,
 	regs->verdict.code = NFT_BREAK;
 }
 
-static noinline void __nft_trace_verdict(struct nft_traceinfo *info,
+static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt,
+					 struct nft_traceinfo *info,
 					 const struct nft_regs *regs)
 {
 	enum nft_trace_types type;
@@ -123,20 +125,21 @@ static noinline void __nft_trace_verdict(struct nft_traceinfo *info,
 		type = NFT_TRACETYPE_RULE;
 
 		if (info->trace)
-			info->nf_trace = info->pkt->skb->nf_trace;
+			info->nf_trace = pkt->skb->nf_trace;
 		break;
 	}
 
-	__nft_trace_packet(info, type);
+	__nft_trace_packet(pkt, info, type);
 }
 
-static inline void nft_trace_verdict(struct nft_traceinfo *info,
+static inline void nft_trace_verdict(const struct nft_pktinfo *pkt,
+				     struct nft_traceinfo *info,
 				     const struct nft_rule_dp *rule,
 				     const struct nft_regs *regs)
 {
 	if (static_branch_unlikely(&nft_trace_enabled)) {
 		info->rule = rule;
-		__nft_trace_verdict(info, regs);
+		__nft_trace_verdict(pkt, info, regs);
 	}
 }
 
@@ -300,7 +303,7 @@ next_rule:
 		break;
 	}
 
-	nft_trace_verdict(&info, rule, &regs);
+	nft_trace_verdict(pkt, &info, rule, &regs);
 
 	switch (regs.verdict.code & NF_VERDICT_MASK) {
 	case NF_ACCEPT:
diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c
index 3d9b83d84a84..0a0dcf2587fd 100644
--- a/net/netfilter/nf_tables_trace.c
+++ b/net/netfilter/nf_tables_trace.c
@@ -183,9 +183,9 @@ static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *i
 	return last->chain;
 }
 
-void nft_trace_notify(struct nft_traceinfo *info)
+void nft_trace_notify(const struct nft_pktinfo *pkt,
+		      struct nft_traceinfo *info)
 {
-	const struct nft_pktinfo *pkt = info->pkt;
 	const struct nft_chain *chain;
 	struct nlmsghdr *nlh;
 	struct sk_buff *skb;
@@ -305,7 +305,6 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
 	info->trace = true;
 	info->nf_trace = pkt->skb->nf_trace;
 	info->packet_dumped = false;
-	info->pkt = pkt;
 	info->verdict = verdict;
 
 	net_get_random_once(&trace_key, sizeof(trace_key));
-- 
cgit v1.2.3-70-g09d2


From 0a202145d5f9277dd24474aca8245731d030f29e Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 14 Apr 2023 15:01:33 +0200
Subject: netfilter: nf_tables: do not store verdict in traceinfo structure

Just pass it as argument to nft_trace_notify. Stack is reduced by 8 bytes:

nf_tables_core.c:256 nft_do_chain    312     static

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  4 +---
 net/netfilter/nf_tables_core.c    | 14 ++++++++------
 net/netfilter/nf_tables_trace.c   | 21 +++++++++++----------
 3 files changed, 20 insertions(+), 19 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index cab351928cd2..693469ecfa54 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1410,7 +1410,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type);
  *	@packet_dumped: packet headers sent in a previous traceinfo message
  *	@basechain: base chain currently processed
  *	@rule:  rule that was evaluated
- *	@verdict: verdict given by rule
  */
 struct nft_traceinfo {
 	bool				trace;
@@ -1420,14 +1419,13 @@ struct nft_traceinfo {
 	u32				skbid;
 	const struct nft_base_chain	*basechain;
 	const struct nft_rule_dp	*rule;
-	const struct nft_verdict	*verdict;
 };
 
 void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
-		    const struct nft_verdict *verdict,
 		    const struct nft_chain *basechain);
 
 void nft_trace_notify(const struct nft_pktinfo *pkt,
+		      const struct nft_verdict *verdict,
 		      struct nft_traceinfo *info);
 
 #define MODULE_ALIAS_NFT_CHAIN(family, name) \
diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index 776eb2b9f632..6debe8b2623f 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -42,6 +42,7 @@ static inline void nf_skip_indirect_calls_enable(void) { }
 #endif
 
 static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt,
+					const struct nft_verdict *verdict,
 					struct nft_traceinfo *info,
 					enum nft_trace_types type)
 {
@@ -50,10 +51,11 @@ static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt,
 
 	info->type = type;
 
-	nft_trace_notify(pkt, info);
+	nft_trace_notify(pkt, verdict, info);
 }
 
 static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
+				    struct nft_verdict *verdict,
 				    struct nft_traceinfo *info,
 				    const struct nft_rule_dp *rule,
 				    enum nft_trace_types type)
@@ -61,7 +63,7 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
 	if (static_branch_unlikely(&nft_trace_enabled)) {
 		info->nf_trace = pkt->skb->nf_trace;
 		info->rule = rule;
-		__nft_trace_packet(pkt, info, type);
+		__nft_trace_packet(pkt, verdict, info, type);
 	}
 }
 
@@ -129,7 +131,7 @@ static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt,
 		break;
 	}
 
-	__nft_trace_packet(pkt, info, type);
+	__nft_trace_packet(pkt, &regs->verdict, info, type);
 }
 
 static inline void nft_trace_verdict(const struct nft_pktinfo *pkt,
@@ -264,7 +266,7 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv)
 
 	info.trace = false;
 	if (static_branch_unlikely(&nft_trace_enabled))
-		nft_trace_init(&info, pkt, &regs.verdict, basechain);
+		nft_trace_init(&info, pkt, basechain);
 do_chain:
 	if (genbit)
 		blob = rcu_dereference(chain->blob_gen_1);
@@ -296,7 +298,7 @@ next_rule:
 			nft_trace_copy_nftrace(pkt, &info);
 			continue;
 		case NFT_CONTINUE:
-			nft_trace_packet(pkt, &info, rule,
+			nft_trace_packet(pkt, &regs.verdict,  &info, rule,
 					 NFT_TRACETYPE_RULE);
 			continue;
 		}
@@ -336,7 +338,7 @@ next_rule:
 		goto next_rule;
 	}
 
-	nft_trace_packet(pkt, &info, NULL, NFT_TRACETYPE_POLICY);
+	nft_trace_packet(pkt, &regs.verdict, &info, NULL, NFT_TRACETYPE_POLICY);
 
 	if (static_branch_unlikely(&nft_counters_enabled))
 		nft_update_chain_stats(basechain, pkt);
diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c
index 0a0dcf2587fd..e635104a42be 100644
--- a/net/netfilter/nf_tables_trace.c
+++ b/net/netfilter/nf_tables_trace.c
@@ -124,6 +124,7 @@ static int nf_trace_fill_pkt_info(struct sk_buff *nlskb,
 }
 
 static int nf_trace_fill_rule_info(struct sk_buff *nlskb,
+				   const struct nft_verdict *verdict,
 				   const struct nft_traceinfo *info)
 {
 	if (!info->rule || info->rule->is_last)
@@ -135,7 +136,7 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb,
 	 * Since no rule matched, the ->rule pointer is invalid.
 	 */
 	if (info->type == NFT_TRACETYPE_RETURN &&
-	    info->verdict->code == NFT_CONTINUE)
+	    verdict->code == NFT_CONTINUE)
 		return 0;
 
 	return nla_put_be64(nlskb, NFTA_TRACE_RULE_HANDLE,
@@ -143,7 +144,8 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb,
 			    NFTA_TRACE_PAD);
 }
 
-static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info)
+static bool nft_trace_have_verdict_chain(const struct nft_verdict *verdict,
+					 struct nft_traceinfo *info)
 {
 	switch (info->type) {
 	case NFT_TRACETYPE_RETURN:
@@ -153,7 +155,7 @@ static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info)
 		return false;
 	}
 
-	switch (info->verdict->code) {
+	switch (verdict->code) {
 	case NFT_JUMP:
 	case NFT_GOTO:
 		break;
@@ -184,6 +186,7 @@ static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *i
 }
 
 void nft_trace_notify(const struct nft_pktinfo *pkt,
+		      const struct nft_verdict *verdict,
 		      struct nft_traceinfo *info)
 {
 	const struct nft_chain *chain;
@@ -217,8 +220,8 @@ void nft_trace_notify(const struct nft_pktinfo *pkt,
 		nla_total_size(sizeof(u32)) +		/* nfproto */
 		nla_total_size(sizeof(u32));		/* policy */
 
-	if (nft_trace_have_verdict_chain(info))
-		size += nla_total_size(strlen(info->verdict->chain->name)); /* jump target */
+	if (nft_trace_have_verdict_chain(verdict, info))
+		size += nla_total_size(strlen(verdict->chain->name)); /* jump target */
 
 	skb = nlmsg_new(size, GFP_ATOMIC);
 	if (!skb)
@@ -245,7 +248,7 @@ void nft_trace_notify(const struct nft_pktinfo *pkt,
 	if (nla_put_string(skb, NFTA_TRACE_TABLE, chain->table->name))
 		goto nla_put_failure;
 
-	if (nf_trace_fill_rule_info(skb, info))
+	if (nf_trace_fill_rule_info(skb, verdict, info))
 		goto nla_put_failure;
 
 	switch (info->type) {
@@ -254,11 +257,11 @@ void nft_trace_notify(const struct nft_pktinfo *pkt,
 		break;
 	case NFT_TRACETYPE_RETURN:
 	case NFT_TRACETYPE_RULE:
-		if (nft_verdict_dump(skb, NFTA_TRACE_VERDICT, info->verdict))
+		if (nft_verdict_dump(skb, NFTA_TRACE_VERDICT, verdict))
 			goto nla_put_failure;
 
 		/* pkt->skb undefined iff NF_STOLEN, disable dump */
-		if (info->verdict->code == NF_STOLEN)
+		if (verdict->code == NF_STOLEN)
 			info->packet_dumped = true;
 		else
 			mark = pkt->skb->mark;
@@ -295,7 +298,6 @@ void nft_trace_notify(const struct nft_pktinfo *pkt,
 }
 
 void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
-		    const struct nft_verdict *verdict,
 		    const struct nft_chain *chain)
 {
 	static siphash_key_t trace_key __read_mostly;
@@ -305,7 +307,6 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
 	info->trace = true;
 	info->nf_trace = pkt->skb->nf_trace;
 	info->packet_dumped = false;
-	info->verdict = verdict;
 
 	net_get_random_once(&trace_key, sizeof(trace_key));
 
-- 
cgit v1.2.3-70-g09d2


From 46df417544f4f7fa3438caada0dc0e534a694343 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Fri, 14 Apr 2023 15:01:34 +0200
Subject: netfilter: nf_tables: do not store rule in traceinfo structure

pass it as argument instead.  This reduces size of traceinfo to
16 bytes.  Total stack usage:

 nf_tables_core.c:252 nft_do_chain    304     static

While its possible to also pass basechain as argument, doing so
increases nft_do_chaininfo function size.

Unlike pktinfo/verdict/rule the basechain info isn't used in
the expression evaluation path. gcc places it on the stack, which
results in extra push/pop when it gets passed to the trace helpers
as argument rather than as part of the traceinfo structure.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  3 +--
 net/netfilter/nf_tables_core.c    | 15 +++++++--------
 net/netfilter/nf_tables_trace.c   | 14 ++++++++------
 3 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 693469ecfa54..58a4d217faaf 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1409,7 +1409,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type);
  *	@skbid: hash of skb to be used as trace id
  *	@packet_dumped: packet headers sent in a previous traceinfo message
  *	@basechain: base chain currently processed
- *	@rule:  rule that was evaluated
  */
 struct nft_traceinfo {
 	bool				trace;
@@ -1418,7 +1417,6 @@ struct nft_traceinfo {
 	enum nft_trace_types		type:8;
 	u32				skbid;
 	const struct nft_base_chain	*basechain;
-	const struct nft_rule_dp	*rule;
 };
 
 void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
@@ -1426,6 +1424,7 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
 
 void nft_trace_notify(const struct nft_pktinfo *pkt,
 		      const struct nft_verdict *verdict,
+		      const struct nft_rule_dp *rule,
 		      struct nft_traceinfo *info);
 
 #define MODULE_ALIAS_NFT_CHAIN(family, name) \
diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index 6debe8b2623f..4d0ce12221f6 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -43,6 +43,7 @@ static inline void nf_skip_indirect_calls_enable(void) { }
 
 static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt,
 					const struct nft_verdict *verdict,
+					const struct nft_rule_dp *rule,
 					struct nft_traceinfo *info,
 					enum nft_trace_types type)
 {
@@ -51,7 +52,7 @@ static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt,
 
 	info->type = type;
 
-	nft_trace_notify(pkt, verdict, info);
+	nft_trace_notify(pkt, verdict, rule, info);
 }
 
 static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
@@ -62,8 +63,7 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt,
 {
 	if (static_branch_unlikely(&nft_trace_enabled)) {
 		info->nf_trace = pkt->skb->nf_trace;
-		info->rule = rule;
-		__nft_trace_packet(pkt, verdict, info, type);
+		__nft_trace_packet(pkt, verdict, rule, info, type);
 	}
 }
 
@@ -110,6 +110,7 @@ static void nft_cmp16_fast_eval(const struct nft_expr *expr,
 
 static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt,
 					 struct nft_traceinfo *info,
+					 const struct nft_rule_dp *rule,
 					 const struct nft_regs *regs)
 {
 	enum nft_trace_types type;
@@ -131,7 +132,7 @@ static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt,
 		break;
 	}
 
-	__nft_trace_packet(pkt, &regs->verdict, info, type);
+	__nft_trace_packet(pkt, &regs->verdict, rule, info, type);
 }
 
 static inline void nft_trace_verdict(const struct nft_pktinfo *pkt,
@@ -139,10 +140,8 @@ static inline void nft_trace_verdict(const struct nft_pktinfo *pkt,
 				     const struct nft_rule_dp *rule,
 				     const struct nft_regs *regs)
 {
-	if (static_branch_unlikely(&nft_trace_enabled)) {
-		info->rule = rule;
-		__nft_trace_verdict(pkt, info, regs);
-	}
+	if (static_branch_unlikely(&nft_trace_enabled))
+		__nft_trace_verdict(pkt, info, rule, regs);
 }
 
 static bool nft_payload_fast_eval(const struct nft_expr *expr,
diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c
index e635104a42be..6d41c0bd3d78 100644
--- a/net/netfilter/nf_tables_trace.c
+++ b/net/netfilter/nf_tables_trace.c
@@ -125,9 +125,10 @@ static int nf_trace_fill_pkt_info(struct sk_buff *nlskb,
 
 static int nf_trace_fill_rule_info(struct sk_buff *nlskb,
 				   const struct nft_verdict *verdict,
+				   const struct nft_rule_dp *rule,
 				   const struct nft_traceinfo *info)
 {
-	if (!info->rule || info->rule->is_last)
+	if (!rule || rule->is_last)
 		return 0;
 
 	/* a continue verdict with ->type == RETURN means that this is
@@ -140,7 +141,7 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb,
 		return 0;
 
 	return nla_put_be64(nlskb, NFTA_TRACE_RULE_HANDLE,
-			    cpu_to_be64(info->rule->handle),
+			    cpu_to_be64(rule->handle),
 			    NFTA_TRACE_PAD);
 }
 
@@ -166,9 +167,9 @@ static bool nft_trace_have_verdict_chain(const struct nft_verdict *verdict,
 	return true;
 }
 
-static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *info)
+static const struct nft_chain *nft_trace_get_chain(const struct nft_rule_dp *rule,
+						   const struct nft_traceinfo *info)
 {
-	const struct nft_rule_dp *rule = info->rule;
 	const struct nft_rule_dp_last *last;
 
 	if (!rule)
@@ -187,6 +188,7 @@ static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *i
 
 void nft_trace_notify(const struct nft_pktinfo *pkt,
 		      const struct nft_verdict *verdict,
+		      const struct nft_rule_dp *rule,
 		      struct nft_traceinfo *info)
 {
 	const struct nft_chain *chain;
@@ -199,7 +201,7 @@ void nft_trace_notify(const struct nft_pktinfo *pkt,
 	if (!nfnetlink_has_listeners(nft_net(pkt), NFNLGRP_NFTRACE))
 		return;
 
-	chain = nft_trace_get_chain(info);
+	chain = nft_trace_get_chain(rule, info);
 
 	size = nlmsg_total_size(sizeof(struct nfgenmsg)) +
 		nla_total_size(strlen(chain->table->name)) +
@@ -248,7 +250,7 @@ void nft_trace_notify(const struct nft_pktinfo *pkt,
 	if (nla_put_string(skb, NFTA_TRACE_TABLE, chain->table->name))
 		goto nla_put_failure;
 
-	if (nf_trace_fill_rule_info(skb, verdict, info))
+	if (nf_trace_fill_rule_info(skb, verdict, rule, info))
 		goto nla_put_failure;
 
 	switch (info->type) {
-- 
cgit v1.2.3-70-g09d2


From e3478c68f6704638d08f437cbc552ca5970c151a Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Mon, 17 Apr 2023 17:10:45 +0200
Subject: ipvs: Update width of source for ip_vs_sync_conn_options

In ip_vs_sync_conn_v0() copy is made to struct ip_vs_sync_conn_options.
That structure looks like this:

struct ip_vs_sync_conn_options {
        struct ip_vs_seq        in_seq;
        struct ip_vs_seq        out_seq;
};

The source of the copy is the in_seq field of struct ip_vs_conn.  Whose
type is struct ip_vs_seq. Thus we can see that the source - is not as
wide as the amount of data copied, which is the width of struct
ip_vs_sync_conn_option.

The copy is safe because the next field in is another struct ip_vs_seq.
Make use of struct_group() to annotate this.

Flagged by gcc-13 as:

 In file included from ./include/linux/string.h:254,
                  from ./include/linux/bitmap.h:11,
                  from ./include/linux/cpumask.h:12,
                  from ./arch/x86/include/asm/paravirt.h:17,
                  from ./arch/x86/include/asm/cpuid.h:62,
                  from ./arch/x86/include/asm/processor.h:19,
                  from ./arch/x86/include/asm/timex.h:5,
                  from ./include/linux/timex.h:67,
                  from ./include/linux/time32.h:13,
                  from ./include/linux/time.h:60,
                  from ./include/linux/stat.h:19,
                  from ./include/linux/module.h:13,
                  from net/netfilter/ipvs/ip_vs_sync.c:38:
 In function 'fortify_memcpy_chk',
     inlined from 'ip_vs_sync_conn_v0' at net/netfilter/ipvs/ip_vs_sync.c:606:3:
 ./include/linux/fortify-string.h:529:25: error: call to '__read_overflow2_field' declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror=attribute-warning]
   529 |                         __read_overflow2_field(q_size_field, size);
       |

Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/ip_vs.h             | 6 ++++--
 net/netfilter/ipvs/ip_vs_sync.c | 2 +-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 6d71a5ff52df..e20f1f92066d 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -630,8 +630,10 @@ struct ip_vs_conn {
 	 */
 	struct ip_vs_app        *app;           /* bound ip_vs_app object */
 	void                    *app_data;      /* Application private data */
-	struct ip_vs_seq        in_seq;         /* incoming seq. struct */
-	struct ip_vs_seq        out_seq;        /* outgoing seq. struct */
+	struct_group(sync_conn_opt,
+		struct ip_vs_seq  in_seq;       /* incoming seq. struct */
+		struct ip_vs_seq  out_seq;      /* outgoing seq. struct */
+	);
 
 	const struct ip_vs_pe	*pe;
 	char			*pe_data;
diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c
index 4963fec815da..d4fe7bb4f853 100644
--- a/net/netfilter/ipvs/ip_vs_sync.c
+++ b/net/netfilter/ipvs/ip_vs_sync.c
@@ -603,7 +603,7 @@ static void ip_vs_sync_conn_v0(struct netns_ipvs *ipvs, struct ip_vs_conn *cp,
 	if (cp->flags & IP_VS_CONN_F_SEQ_MASK) {
 		struct ip_vs_sync_conn_options *opt =
 			(struct ip_vs_sync_conn_options *)&s[1];
-		memcpy(opt, &cp->in_seq, sizeof(*opt));
+		memcpy(opt, &cp->sync_conn_opt, sizeof(*opt));
 	}
 
 	m->nr_conns++;
-- 
cgit v1.2.3-70-g09d2


From 280654932e341bf2860503af49bd5cb8e2b78522 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Mon, 17 Apr 2023 17:10:46 +0200
Subject: ipvs: Consistently use array_size() in ip_vs_conn_init()

Consistently use array_size() to calculate the size of ip_vs_conn_tab
in bytes.

Flagged by Coccinelle:
 WARNING: array_size is already used (line 1498) to compute the same size

No functional change intended.
Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/ipvs/ip_vs_conn.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index 13534e02346c..928e64653837 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -1481,6 +1481,7 @@ void __net_exit ip_vs_conn_net_cleanup(struct netns_ipvs *ipvs)
 
 int __init ip_vs_conn_init(void)
 {
+	size_t tab_array_size;
 	int idx;
 
 	/* Compute size and mask */
@@ -1494,8 +1495,9 @@ int __init ip_vs_conn_init(void)
 	/*
 	 * Allocate the connection hash table and initialize its list heads
 	 */
-	ip_vs_conn_tab = vmalloc(array_size(ip_vs_conn_tab_size,
-					    sizeof(*ip_vs_conn_tab)));
+	tab_array_size = array_size(ip_vs_conn_tab_size,
+				    sizeof(*ip_vs_conn_tab));
+	ip_vs_conn_tab = vmalloc(tab_array_size);
 	if (!ip_vs_conn_tab)
 		return -ENOMEM;
 
@@ -1508,10 +1510,8 @@ int __init ip_vs_conn_init(void)
 		return -ENOMEM;
 	}
 
-	pr_info("Connection hash table configured "
-		"(size=%d, memory=%ldKbytes)\n",
-		ip_vs_conn_tab_size,
-		(long)(ip_vs_conn_tab_size*sizeof(*ip_vs_conn_tab))/1024);
+	pr_info("Connection hash table configured (size=%d, memory=%zdKbytes)\n",
+		ip_vs_conn_tab_size, tab_array_size / 1024);
 	IP_VS_DBG(0, "Each connection entry needs %zd bytes at least\n",
 		  sizeof(struct ip_vs_conn));
 
-- 
cgit v1.2.3-70-g09d2


From 210ffe4a74caead4d6790747d32b63c6152c70b7 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Mon, 17 Apr 2023 17:10:47 +0200
Subject: ipvs: Remove {Enter,Leave}Function

Remove EnterFunction and LeaveFunction.

These debugging macros seem well past their use-by date.  And seem to
have little value these days. Removing them allows some trivial cleanup
of some exit paths for some functions. These are also included in this
patch. There is likely scope for further cleanup of both debugging and
unwind paths. But let's leave that for another day.

Only intended to change debug output, and only when CONFIG_IP_VS_DEBUG
is enabled. Compile tested only.

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/ip_vs.h             | 20 -------------
 net/netfilter/ipvs/ip_vs_core.c |  8 ------
 net/netfilter/ipvs/ip_vs_ctl.c  | 26 +----------------
 net/netfilter/ipvs/ip_vs_sync.c |  5 ----
 net/netfilter/ipvs/ip_vs_xmit.c | 62 ++++++-----------------------------------
 5 files changed, 9 insertions(+), 112 deletions(-)

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index e20f1f92066d..a3adc246ee31 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -265,26 +265,6 @@ static inline const char *ip_vs_dbg_addr(int af, char *buf, size_t buf_len,
 			pr_err(msg, ##__VA_ARGS__);			\
 	} while (0)
 
-#ifdef CONFIG_IP_VS_DEBUG
-#define EnterFunction(level)						\
-	do {								\
-		if (level <= ip_vs_get_debug_level())			\
-			printk(KERN_DEBUG				\
-			       pr_fmt("Enter: %s, %s line %i\n"),	\
-			       __func__, __FILE__, __LINE__);		\
-	} while (0)
-#define LeaveFunction(level)						\
-	do {								\
-		if (level <= ip_vs_get_debug_level())			\
-			printk(KERN_DEBUG				\
-			       pr_fmt("Leave: %s, %s line %i\n"),	\
-			       __func__, __FILE__, __LINE__);		\
-	} while (0)
-#else
-#define EnterFunction(level)   do {} while (0)
-#define LeaveFunction(level)   do {} while (0)
-#endif
-
 /* The port number of FTP service (in network order). */
 #define FTPPORT  cpu_to_be16(21)
 #define FTPDATA  cpu_to_be16(20)
diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 2fcc26507d69..cb83ca506c5c 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -1140,7 +1140,6 @@ struct ip_vs_conn *ip_vs_new_conn_out(struct ip_vs_service *svc,
 	__be16 vport;
 	unsigned int flags;
 
-	EnterFunction(12);
 	vaddr = &svc->addr;
 	vport = svc->port;
 	daddr = &iph->saddr;
@@ -1208,7 +1207,6 @@ struct ip_vs_conn *ip_vs_new_conn_out(struct ip_vs_service *svc,
 		      IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport),
 		      IP_VS_DBG_ADDR(cp->af, &cp->daddr), ntohs(cp->dport),
 		      cp->flags, refcount_read(&cp->refcnt));
-	LeaveFunction(12);
 	return cp;
 }
 
@@ -1316,13 +1314,11 @@ after_nat:
 		ip_vs_update_conntrack(skb, cp, 0);
 	ip_vs_conn_put(cp);
 
-	LeaveFunction(11);
 	return NF_ACCEPT;
 
 drop:
 	ip_vs_conn_put(cp);
 	kfree_skb(skb);
-	LeaveFunction(11);
 	return NF_STOLEN;
 }
 
@@ -1341,8 +1337,6 @@ ip_vs_out_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *stat
 	int af = state->pf;
 	struct sock *sk;
 
-	EnterFunction(11);
-
 	/* Already marked as IPVS request or reply? */
 	if (skb->ipvs_property)
 		return NF_ACCEPT;
@@ -2365,7 +2359,6 @@ static void __net_exit __ip_vs_dev_cleanup_batch(struct list_head *net_list)
 	struct netns_ipvs *ipvs;
 	struct net *net;
 
-	EnterFunction(2);
 	list_for_each_entry(net, net_list, exit_list) {
 		ipvs = net_ipvs(net);
 		ip_vs_unregister_hooks(ipvs, AF_INET);
@@ -2374,7 +2367,6 @@ static void __net_exit __ip_vs_dev_cleanup_batch(struct list_head *net_list)
 		smp_wmb();
 		ip_vs_sync_net_cleanup(ipvs);
 	}
-	LeaveFunction(2);
 }
 
 static struct pernet_operations ipvs_core_ops = {
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 2a5ed71c82c3..62606fb44d02 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -1061,8 +1061,6 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	unsigned int atype;
 	int ret;
 
-	EnterFunction(2);
-
 #ifdef CONFIG_IP_VS_IPV6
 	if (udest->af == AF_INET6) {
 		atype = ipv6_addr_type(&udest->addr.in6);
@@ -1111,7 +1109,6 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	spin_lock_init(&dest->dst_lock);
 	__ip_vs_update_dest(svc, dest, udest, 1);
 
-	LeaveFunction(2);
 	return 0;
 
 err_stats:
@@ -1134,8 +1131,6 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	__be16 dport = udest->port;
 	int ret;
 
-	EnterFunction(2);
-
 	if (udest->weight < 0) {
 		pr_err("%s(): server weight less than zero\n", __func__);
 		return -ERANGE;
@@ -1183,7 +1178,7 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 
 		ret = ip_vs_start_estimator(svc->ipvs, &dest->stats);
 		if (ret < 0)
-			goto err;
+			return ret;
 		__ip_vs_update_dest(svc, dest, udest, 1);
 	} else {
 		/*
@@ -1192,9 +1187,6 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 		ret = ip_vs_new_dest(svc, udest);
 	}
 
-err:
-	LeaveFunction(2);
-
 	return ret;
 }
 
@@ -1209,8 +1201,6 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	union nf_inet_addr daddr;
 	__be16 dport = udest->port;
 
-	EnterFunction(2);
-
 	if (udest->weight < 0) {
 		pr_err("%s(): server weight less than zero\n", __func__);
 		return -ERANGE;
@@ -1242,7 +1232,6 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	}
 
 	__ip_vs_update_dest(svc, dest, udest, 0);
-	LeaveFunction(2);
 
 	return 0;
 }
@@ -1317,8 +1306,6 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	struct ip_vs_dest *dest;
 	__be16 dport = udest->port;
 
-	EnterFunction(2);
-
 	/* We use function that requires RCU lock */
 	rcu_read_lock();
 	dest = ip_vs_lookup_dest(svc, udest->af, &udest->addr, dport);
@@ -1339,8 +1326,6 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 	 */
 	__ip_vs_del_dest(svc->ipvs, dest, false);
 
-	LeaveFunction(2);
-
 	return 0;
 }
 
@@ -1746,7 +1731,6 @@ void ip_vs_service_nets_cleanup(struct list_head *net_list)
 	struct netns_ipvs *ipvs;
 	struct net *net;
 
-	EnterFunction(2);
 	/* Check for "full" addressed entries */
 	mutex_lock(&__ip_vs_mutex);
 	list_for_each_entry(net, net_list, exit_list) {
@@ -1754,7 +1738,6 @@ void ip_vs_service_nets_cleanup(struct list_head *net_list)
 		ip_vs_flush(ipvs, true);
 	}
 	mutex_unlock(&__ip_vs_mutex);
-	LeaveFunction(2);
 }
 
 /* Put all references for device (dst_cache) */
@@ -1792,7 +1775,6 @@ static int ip_vs_dst_event(struct notifier_block *this, unsigned long event,
 	if (event != NETDEV_DOWN || !ipvs)
 		return NOTIFY_DONE;
 	IP_VS_DBG(3, "%s() dev=%s\n", __func__, dev->name);
-	EnterFunction(2);
 	mutex_lock(&__ip_vs_mutex);
 	for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) {
 		hlist_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) {
@@ -1821,7 +1803,6 @@ static int ip_vs_dst_event(struct notifier_block *this, unsigned long event,
 	}
 	spin_unlock_bh(&ipvs->dest_trash_lock);
 	mutex_unlock(&__ip_vs_mutex);
-	LeaveFunction(2);
 	return NOTIFY_DONE;
 }
 
@@ -4537,8 +4518,6 @@ int __init ip_vs_control_init(void)
 	int idx;
 	int ret;
 
-	EnterFunction(2);
-
 	/* Initialize svc_table, ip_vs_svc_fwm_table */
 	for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) {
 		INIT_HLIST_HEAD(&ip_vs_svc_table[idx]);
@@ -4551,15 +4530,12 @@ int __init ip_vs_control_init(void)
 	if (ret < 0)
 		return ret;
 
-	LeaveFunction(2);
 	return 0;
 }
 
 
 void ip_vs_control_cleanup(void)
 {
-	EnterFunction(2);
 	unregister_netdevice_notifier(&ip_vs_dst_notifier);
 	/* relying on common rcu_barrier() in ip_vs_cleanup() */
-	LeaveFunction(2);
 }
diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c
index d4fe7bb4f853..264f2f87a437 100644
--- a/net/netfilter/ipvs/ip_vs_sync.c
+++ b/net/netfilter/ipvs/ip_vs_sync.c
@@ -1582,13 +1582,11 @@ ip_vs_send_async(struct socket *sock, const char *buffer, const size_t length)
 	struct kvec	iov;
 	int		len;
 
-	EnterFunction(7);
 	iov.iov_base     = (void *)buffer;
 	iov.iov_len      = length;
 
 	len = kernel_sendmsg(sock, &msg, &iov, 1, (size_t)(length));
 
-	LeaveFunction(7);
 	return len;
 }
 
@@ -1614,15 +1612,12 @@ ip_vs_receive(struct socket *sock, char *buffer, const size_t buflen)
 	struct kvec		iov = {buffer, buflen};
 	int			len;
 
-	EnterFunction(7);
-
 	/* Receive a packet */
 	iov_iter_kvec(&msg.msg_iter, ITER_DEST, &iov, 1, buflen);
 	len = sock_recvmsg(sock, &msg, MSG_DONTWAIT);
 	if (len < 0)
 		return len;
 
-	LeaveFunction(7);
 	return len;
 }
 
diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
index 99c349c0d968..feb1d7fcb09f 100644
--- a/net/netfilter/ipvs/ip_vs_xmit.c
+++ b/net/netfilter/ipvs/ip_vs_xmit.c
@@ -706,8 +706,6 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 {
 	struct iphdr  *iph = ip_hdr(skb);
 
-	EnterFunction(10);
-
 	if (__ip_vs_get_out_rt(cp->ipvs, cp->af, skb, NULL, iph->daddr,
 			       IP_VS_RT_MODE_NON_LOCAL, NULL, ipvsh) < 0)
 		goto tx_error;
@@ -719,12 +717,10 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0);
 
-	LeaveFunction(10);
 	return NF_STOLEN;
 
  tx_error:
 	kfree_skb(skb);
-	LeaveFunction(10);
 	return NF_STOLEN;
 }
 
@@ -735,8 +731,6 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 {
 	struct ipv6hdr *iph = ipv6_hdr(skb);
 
-	EnterFunction(10);
-
 	if (__ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, NULL,
 				  &iph->daddr, NULL,
 				  ipvsh, 0, IP_VS_RT_MODE_NON_LOCAL) < 0)
@@ -747,12 +741,10 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0);
 
-	LeaveFunction(10);
 	return NF_STOLEN;
 
  tx_error:
 	kfree_skb(skb);
-	LeaveFunction(10);
 	return NF_STOLEN;
 }
 #endif
@@ -768,8 +760,6 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	struct rtable *rt;		/* Route to the other host */
 	int local, rc, was_input;
 
-	EnterFunction(10);
-
 	/* check if it is a connection of no-client-port */
 	if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT)) {
 		__be16 _pt, *p;
@@ -839,12 +829,10 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local);
 
-	LeaveFunction(10);
 	return rc;
 
   tx_error:
 	kfree_skb(skb);
-	LeaveFunction(10);
 	return NF_STOLEN;
 }
 
@@ -856,8 +844,6 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	struct rt6_info *rt;		/* Route to the other host */
 	int local, rc;
 
-	EnterFunction(10);
-
 	/* check if it is a connection of no-client-port */
 	if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT && !ipvsh->fragoffs)) {
 		__be16 _pt, *p;
@@ -927,11 +913,9 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local);
 
-	LeaveFunction(10);
 	return rc;
 
 tx_error:
-	LeaveFunction(10);
 	kfree_skb(skb);
 	return NF_STOLEN;
 }
@@ -1149,8 +1133,6 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	int tun_type, gso_type;
 	int tun_flags;
 
-	EnterFunction(10);
-
 	local = __ip_vs_get_out_rt(ipvs, cp->af, skb, cp->dest, cp->daddr.ip,
 				   IP_VS_RT_MODE_LOCAL |
 				   IP_VS_RT_MODE_NON_LOCAL |
@@ -1199,7 +1181,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 					 &next_protocol, NULL, &dsfield,
 					 &ttl, dfp);
 	if (IS_ERR(skb))
-		goto tx_error;
+		return NF_STOLEN;
 
 	gso_type = __tun_gso_type_mask(AF_INET, cp->af);
 	if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
@@ -1267,14 +1249,10 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	else if (ret == NF_DROP)
 		kfree_skb(skb);
 
-	LeaveFunction(10);
-
 	return NF_STOLEN;
 
   tx_error:
-	if (!IS_ERR(skb))
-		kfree_skb(skb);
-	LeaveFunction(10);
+	kfree_skb(skb);
 	return NF_STOLEN;
 }
 
@@ -1298,8 +1276,6 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	int tun_type, gso_type;
 	int tun_flags;
 
-	EnterFunction(10);
-
 	local = __ip_vs_get_out_rt_v6(ipvs, cp->af, skb, cp->dest,
 				      &cp->daddr.in6,
 				      &saddr, ipvsh, 1,
@@ -1347,7 +1323,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 					 &next_protocol, &payload_len,
 					 &dsfield, &ttl, NULL);
 	if (IS_ERR(skb))
-		goto tx_error;
+		return NF_STOLEN;
 
 	gso_type = __tun_gso_type_mask(AF_INET6, cp->af);
 	if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) {
@@ -1414,14 +1390,10 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	else if (ret == NF_DROP)
 		kfree_skb(skb);
 
-	LeaveFunction(10);
-
 	return NF_STOLEN;
 
 tx_error:
-	if (!IS_ERR(skb))
-		kfree_skb(skb);
-	LeaveFunction(10);
+	kfree_skb(skb);
 	return NF_STOLEN;
 }
 #endif
@@ -1437,8 +1409,6 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 {
 	int local;
 
-	EnterFunction(10);
-
 	local = __ip_vs_get_out_rt(cp->ipvs, cp->af, skb, cp->dest, cp->daddr.ip,
 				   IP_VS_RT_MODE_LOCAL |
 				   IP_VS_RT_MODE_NON_LOCAL |
@@ -1455,12 +1425,10 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0);
 
-	LeaveFunction(10);
 	return NF_STOLEN;
 
   tx_error:
 	kfree_skb(skb);
-	LeaveFunction(10);
 	return NF_STOLEN;
 }
 
@@ -1471,8 +1439,6 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 {
 	int local;
 
-	EnterFunction(10);
-
 	local = __ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, cp->dest,
 				      &cp->daddr.in6,
 				      NULL, ipvsh, 0,
@@ -1489,12 +1455,10 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0);
 
-	LeaveFunction(10);
 	return NF_STOLEN;
 
 tx_error:
 	kfree_skb(skb);
-	LeaveFunction(10);
 	return NF_STOLEN;
 }
 #endif
@@ -1514,8 +1478,6 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	int local;
 	int rt_mode, was_input;
 
-	EnterFunction(10);
-
 	/* The ICMP packet for VS/TUN, VS/DR and LOCALNODE will be
 	   forwarded directly here, because there is no need to
 	   translate address/port back */
@@ -1526,7 +1488,7 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 			rc = NF_ACCEPT;
 		/* do not touch skb anymore */
 		atomic_inc(&cp->in_pkts);
-		goto out;
+		return rc;
 	}
 
 	/*
@@ -1582,14 +1544,11 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->ignore_df = 1;
 
-	rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local);
-	goto out;
+	return ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local);
 
   tx_error:
 	kfree_skb(skb);
 	rc = NF_STOLEN;
-  out:
-	LeaveFunction(10);
 	return rc;
 }
 
@@ -1604,8 +1563,6 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	int local;
 	int rt_mode;
 
-	EnterFunction(10);
-
 	/* The ICMP packet for VS/TUN, VS/DR and LOCALNODE will be
 	   forwarded directly here, because there is no need to
 	   translate address/port back */
@@ -1616,7 +1573,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 			rc = NF_ACCEPT;
 		/* do not touch skb anymore */
 		atomic_inc(&cp->in_pkts);
-		goto out;
+		return rc;
 	}
 
 	/*
@@ -1671,14 +1628,11 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->ignore_df = 1;
 
-	rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local);
-	goto out;
+	return ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local);
 
 tx_error:
 	kfree_skb(skb);
 	rc = NF_STOLEN;
-out:
-	LeaveFunction(10);
 	return rc;
 }
 #endif
-- 
cgit v1.2.3-70-g09d2


From c7d15aaa105a9484b5385e5c391ea5203347d6b0 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Mon, 17 Apr 2023 17:10:48 +0200
Subject: ipvs: Correct spelling in comments

Correct some spelling errors flagged by codespell and found by inspection.

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/ip_vs.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index a3adc246ee31..ff406ef4fd4a 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -584,7 +584,7 @@ struct ip_vs_conn {
 	spinlock_t              lock;           /* lock for state transition */
 	volatile __u16          state;          /* state info */
 	volatile __u16          old_state;      /* old state, to be used for
-						 * state transition triggerd
+						 * state transition triggered
 						 * synchronization
 						 */
 	__u32			fwmark;		/* Fire wall mark from skb */
@@ -635,7 +635,7 @@ struct ip_vs_service_user_kern {
 	u16			protocol;
 	union nf_inet_addr	addr;		/* virtual ip address */
 	__be16			port;
-	u32			fwmark;		/* firwall mark of service */
+	u32			fwmark;		/* firewall mark of service */
 
 	/* virtual service options */
 	char			*sched_name;
@@ -1036,7 +1036,7 @@ struct netns_ipvs {
 	struct ipvs_sync_daemon_cfg	bcfg;	/* Backup Configuration */
 	/* net name space ptr */
 	struct net		*net;            /* Needed by timer routines */
-	/* Number of heterogeneous destinations, needed becaus heterogeneous
+	/* Number of heterogeneous destinations, needed because heterogeneous
 	 * are not supported when synchronization is enabled.
 	 */
 	unsigned int		mixed_address_family_dests;
-- 
cgit v1.2.3-70-g09d2


From c3c060adc0249355411a93e61888051e6902b8a1 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Fri, 21 Apr 2023 00:34:28 +0200
Subject: netfilter: nf_tables: extended netlink error reporting for netdevice

Flowtable and netdev chains are bound to one or several netdevice,
extend netlink error reporting to specify the the netdevice that
triggers the error.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 38 ++++++++++++++++++++++++--------------
 1 file changed, 24 insertions(+), 14 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 44ebc5f9598e..0fd4e28fbb60 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1954,7 +1954,8 @@ static struct nft_hook *nft_hook_list_find(struct list_head *hook_list,
 
 static int nf_tables_parse_netdev_hooks(struct net *net,
 					const struct nlattr *attr,
-					struct list_head *hook_list)
+					struct list_head *hook_list,
+					struct netlink_ext_ack *extack)
 {
 	struct nft_hook *hook, *next;
 	const struct nlattr *tmp;
@@ -1968,10 +1969,12 @@ static int nf_tables_parse_netdev_hooks(struct net *net,
 
 		hook = nft_netdev_hook_alloc(net, tmp);
 		if (IS_ERR(hook)) {
+			NL_SET_BAD_ATTR(extack, tmp);
 			err = PTR_ERR(hook);
 			goto err_hook;
 		}
 		if (nft_hook_list_find(hook_list, hook)) {
+			NL_SET_BAD_ATTR(extack, tmp);
 			kfree(hook);
 			err = -EEXIST;
 			goto err_hook;
@@ -2004,20 +2007,23 @@ struct nft_chain_hook {
 
 static int nft_chain_parse_netdev(struct net *net,
 				  struct nlattr *tb[],
-				  struct list_head *hook_list)
+				  struct list_head *hook_list,
+				  struct netlink_ext_ack *extack)
 {
 	struct nft_hook *hook;
 	int err;
 
 	if (tb[NFTA_HOOK_DEV]) {
 		hook = nft_netdev_hook_alloc(net, tb[NFTA_HOOK_DEV]);
-		if (IS_ERR(hook))
+		if (IS_ERR(hook)) {
+			NL_SET_BAD_ATTR(extack, tb[NFTA_HOOK_DEV]);
 			return PTR_ERR(hook);
+		}
 
 		list_add_tail(&hook->list, hook_list);
 	} else if (tb[NFTA_HOOK_DEVS]) {
 		err = nf_tables_parse_netdev_hooks(net, tb[NFTA_HOOK_DEVS],
-						   hook_list);
+						   hook_list, extack);
 		if (err < 0)
 			return err;
 
@@ -2085,7 +2091,7 @@ static int nft_chain_parse_hook(struct net *net,
 
 	INIT_LIST_HEAD(&hook->list);
 	if (nft_base_chain_netdev(family, hook->num)) {
-		err = nft_chain_parse_netdev(net, ha, &hook->list);
+		err = nft_chain_parse_netdev(net, ha, &hook->list, extack);
 		if (err < 0) {
 			module_put(type->owner);
 			return err;
@@ -7560,7 +7566,8 @@ static const struct nla_policy nft_flowtable_hook_policy[NFTA_FLOWTABLE_HOOK_MAX
 static int nft_flowtable_parse_hook(const struct nft_ctx *ctx,
 				    const struct nlattr *attr,
 				    struct nft_flowtable_hook *flowtable_hook,
-				    struct nft_flowtable *flowtable, bool add)
+				    struct nft_flowtable *flowtable,
+				    struct netlink_ext_ack *extack, bool add)
 {
 	struct nlattr *tb[NFTA_FLOWTABLE_HOOK_MAX + 1];
 	struct nft_hook *hook;
@@ -7607,7 +7614,8 @@ static int nft_flowtable_parse_hook(const struct nft_ctx *ctx,
 	if (tb[NFTA_FLOWTABLE_HOOK_DEVS]) {
 		err = nf_tables_parse_netdev_hooks(ctx->net,
 						   tb[NFTA_FLOWTABLE_HOOK_DEVS],
-						   &flowtable_hook->list);
+						   &flowtable_hook->list,
+						   extack);
 		if (err < 0)
 			return err;
 	}
@@ -7750,7 +7758,8 @@ static void nft_flowtable_hooks_destroy(struct list_head *hook_list)
 }
 
 static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh,
-				struct nft_flowtable *flowtable)
+				struct nft_flowtable *flowtable,
+				struct netlink_ext_ack *extack)
 {
 	const struct nlattr * const *nla = ctx->nla;
 	struct nft_flowtable_hook flowtable_hook;
@@ -7761,7 +7770,7 @@ static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh,
 	int err;
 
 	err = nft_flowtable_parse_hook(ctx, nla[NFTA_FLOWTABLE_HOOK],
-				       &flowtable_hook, flowtable, false);
+				       &flowtable_hook, flowtable, extack, false);
 	if (err < 0)
 		return err;
 
@@ -7866,7 +7875,7 @@ static int nf_tables_newflowtable(struct sk_buff *skb,
 
 		nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla);
 
-		return nft_flowtable_update(&ctx, info->nlh, flowtable);
+		return nft_flowtable_update(&ctx, info->nlh, flowtable, extack);
 	}
 
 	nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla);
@@ -7907,7 +7916,7 @@ static int nf_tables_newflowtable(struct sk_buff *skb,
 		goto err3;
 
 	err = nft_flowtable_parse_hook(&ctx, nla[NFTA_FLOWTABLE_HOOK],
-				       &flowtable_hook, flowtable, true);
+				       &flowtable_hook, flowtable, extack, true);
 	if (err < 0)
 		goto err4;
 
@@ -7959,7 +7968,8 @@ static void nft_flowtable_hook_release(struct nft_flowtable_hook *flowtable_hook
 }
 
 static int nft_delflowtable_hook(struct nft_ctx *ctx,
-				 struct nft_flowtable *flowtable)
+				 struct nft_flowtable *flowtable,
+				 struct netlink_ext_ack *extack)
 {
 	const struct nlattr * const *nla = ctx->nla;
 	struct nft_flowtable_hook flowtable_hook;
@@ -7969,7 +7979,7 @@ static int nft_delflowtable_hook(struct nft_ctx *ctx,
 	int err;
 
 	err = nft_flowtable_parse_hook(ctx, nla[NFTA_FLOWTABLE_HOOK],
-				       &flowtable_hook, flowtable, false);
+				       &flowtable_hook, flowtable, extack, false);
 	if (err < 0)
 		return err;
 
@@ -8051,7 +8061,7 @@ static int nf_tables_delflowtable(struct sk_buff *skb,
 	nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla);
 
 	if (nla[NFTA_FLOWTABLE_HOOK])
-		return nft_delflowtable_hook(&ctx, flowtable);
+		return nft_delflowtable_hook(&ctx, flowtable, extack);
 
 	if (flowtable->use > 0) {
 		NL_SET_BAD_ATTR(extack, attr);
-- 
cgit v1.2.3-70-g09d2


From 28339b21a365a8cb9edc6956cbfd679f569daa81 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Fri, 21 Apr 2023 00:34:29 +0200
Subject: netfilter: nf_tables: do not send complete notification of deletions

In most cases, table, name and handle is sufficient for userspace to
identify an object that has been deleted. Skipping unneeded fields in
the netlink attributes in the message saves bandwidth (ie. less chances
of hitting ENOBUFS).

Rules are an exception: the existing userspace monitor code relies on
the rule definition. This exception can be removed by implementing a
rule cache in userspace, this is already supported by the tracing
infrastructure.

Regarding flowtables, incremental deletion of devices is possible.
Skipping a full notification allows userspace to differentiate between
flowtable removal and incremental removal of devices.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 70 +++++++++++++++++++++++++++++++------------
 1 file changed, 51 insertions(+), 19 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 0fd4e28fbb60..3f95fb24a0bc 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -819,12 +819,20 @@ static int nf_tables_fill_table_info(struct sk_buff *skb, struct net *net,
 		goto nla_put_failure;
 
 	if (nla_put_string(skb, NFTA_TABLE_NAME, table->name) ||
-	    nla_put_be32(skb, NFTA_TABLE_FLAGS,
-			 htonl(table->flags & NFT_TABLE_F_MASK)) ||
 	    nla_put_be32(skb, NFTA_TABLE_USE, htonl(table->use)) ||
 	    nla_put_be64(skb, NFTA_TABLE_HANDLE, cpu_to_be64(table->handle),
 			 NFTA_TABLE_PAD))
 		goto nla_put_failure;
+
+	if (event == NFT_MSG_DELTABLE) {
+		nlmsg_end(skb, nlh);
+		return 0;
+	}
+
+	if (nla_put_be32(skb, NFTA_TABLE_FLAGS,
+			 htonl(table->flags & NFT_TABLE_F_MASK)))
+		goto nla_put_failure;
+
 	if (nft_table_has_owner(table) &&
 	    nla_put_be32(skb, NFTA_TABLE_OWNER, htonl(table->nlpid)))
 		goto nla_put_failure;
@@ -1626,13 +1634,16 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net,
 	if (!nlh)
 		goto nla_put_failure;
 
-	if (nla_put_string(skb, NFTA_CHAIN_TABLE, table->name))
-		goto nla_put_failure;
-	if (nla_put_be64(skb, NFTA_CHAIN_HANDLE, cpu_to_be64(chain->handle),
+	if (nla_put_string(skb, NFTA_CHAIN_TABLE, table->name) ||
+	    nla_put_string(skb, NFTA_CHAIN_NAME, chain->name) ||
+	    nla_put_be64(skb, NFTA_CHAIN_HANDLE, cpu_to_be64(chain->handle),
 			 NFTA_CHAIN_PAD))
 		goto nla_put_failure;
-	if (nla_put_string(skb, NFTA_CHAIN_NAME, chain->name))
-		goto nla_put_failure;
+
+	if (event == NFT_MSG_DELCHAIN) {
+		nlmsg_end(skb, nlh);
+		return 0;
+	}
 
 	if (nft_is_base_chain(chain)) {
 		const struct nft_base_chain *basechain = nft_base_chain(chain);
@@ -4150,6 +4161,12 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx,
 	if (nla_put_be64(skb, NFTA_SET_HANDLE, cpu_to_be64(set->handle),
 			 NFTA_SET_PAD))
 		goto nla_put_failure;
+
+	if (event == NFT_MSG_DELSET) {
+		nlmsg_end(skb, nlh);
+		return 0;
+	}
+
 	if (set->flags != 0)
 		if (nla_put_be32(skb, NFTA_SET_FLAGS, htonl(set->flags)))
 			goto nla_put_failure;
@@ -7154,13 +7171,20 @@ static int nf_tables_fill_obj_info(struct sk_buff *skb, struct net *net,
 
 	if (nla_put_string(skb, NFTA_OBJ_TABLE, table->name) ||
 	    nla_put_string(skb, NFTA_OBJ_NAME, obj->key.name) ||
-	    nla_put_be32(skb, NFTA_OBJ_TYPE, htonl(obj->ops->type->type)) ||
-	    nla_put_be32(skb, NFTA_OBJ_USE, htonl(obj->use)) ||
-	    nft_object_dump(skb, NFTA_OBJ_DATA, obj, reset) ||
 	    nla_put_be64(skb, NFTA_OBJ_HANDLE, cpu_to_be64(obj->handle),
 			 NFTA_OBJ_PAD))
 		goto nla_put_failure;
 
+	if (event == NFT_MSG_DELOBJ) {
+		nlmsg_end(skb, nlh);
+		return 0;
+	}
+
+	if (nla_put_be32(skb, NFTA_OBJ_TYPE, htonl(obj->ops->type->type)) ||
+	    nla_put_be32(skb, NFTA_OBJ_USE, htonl(obj->use)) ||
+	    nft_object_dump(skb, NFTA_OBJ_DATA, obj, reset))
+		goto nla_put_failure;
+
 	if (obj->udata &&
 	    nla_put(skb, NFTA_OBJ_USERDATA, obj->udlen, obj->udata))
 		goto nla_put_failure;
@@ -8089,9 +8113,16 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net,
 
 	if (nla_put_string(skb, NFTA_FLOWTABLE_TABLE, flowtable->table->name) ||
 	    nla_put_string(skb, NFTA_FLOWTABLE_NAME, flowtable->name) ||
-	    nla_put_be32(skb, NFTA_FLOWTABLE_USE, htonl(flowtable->use)) ||
 	    nla_put_be64(skb, NFTA_FLOWTABLE_HANDLE, cpu_to_be64(flowtable->handle),
-			 NFTA_FLOWTABLE_PAD) ||
+			 NFTA_FLOWTABLE_PAD))
+		goto nla_put_failure;
+
+	if (event == NFT_MSG_DELFLOWTABLE && !hook_list) {
+		nlmsg_end(skb, nlh);
+		return 0;
+	}
+
+	if (nla_put_be32(skb, NFTA_FLOWTABLE_USE, htonl(flowtable->use)) ||
 	    nla_put_be32(skb, NFTA_FLOWTABLE_FLAGS, htonl(flowtable->data.flags)))
 		goto nla_put_failure;
 
@@ -8106,6 +8137,9 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net,
 	if (!nest_devs)
 		goto nla_put_failure;
 
+	if (!hook_list)
+		hook_list = &flowtable->hook_list;
+
 	list_for_each_entry_rcu(hook, hook_list, list) {
 		if (nla_put_string(skb, NFTA_DEVICE_NAME, hook->ops.dev->name))
 			goto nla_put_failure;
@@ -8162,8 +8196,7 @@ static int nf_tables_dump_flowtable(struct sk_buff *skb,
 							  NFT_MSG_NEWFLOWTABLE,
 							  NLM_F_MULTI | NLM_F_APPEND,
 							  table->family,
-							  flowtable,
-							  &flowtable->hook_list) < 0)
+							  flowtable, NULL) < 0)
 				goto done;
 
 			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
@@ -8258,7 +8291,7 @@ static int nf_tables_getflowtable(struct sk_buff *skb,
 	err = nf_tables_fill_flowtable_info(skb2, net, NETLINK_CB(skb).portid,
 					    info->nlh->nlmsg_seq,
 					    NFT_MSG_NEWFLOWTABLE, 0, family,
-					    flowtable, &flowtable->hook_list);
+					    flowtable, NULL);
 	if (err < 0)
 		goto err_fill_flowtable_info;
 
@@ -8271,8 +8304,7 @@ err_fill_flowtable_info:
 
 static void nf_tables_flowtable_notify(struct nft_ctx *ctx,
 				       struct nft_flowtable *flowtable,
-				       struct list_head *hook_list,
-				       int event)
+				       struct list_head *hook_list, int event)
 {
 	struct nftables_pernet *nft_net = nft_pernet(ctx->net);
 	struct sk_buff *skb;
@@ -9333,7 +9365,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				nft_clear(net, nft_trans_flowtable(trans));
 				nf_tables_flowtable_notify(&trans->ctx,
 							   nft_trans_flowtable(trans),
-							   &nft_trans_flowtable(trans)->hook_list,
+							   NULL,
 							   NFT_MSG_NEWFLOWTABLE);
 			}
 			nft_trans_destroy(trans);
@@ -9351,7 +9383,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				list_del_rcu(&nft_trans_flowtable(trans)->list);
 				nf_tables_flowtable_notify(&trans->ctx,
 							   nft_trans_flowtable(trans),
-							   &nft_trans_flowtable(trans)->hook_list,
+							   NULL,
 							   trans->msg_type);
 				nft_unregister_flowtable_net_hooks(net,
 						&nft_trans_flowtable(trans)->hook_list);
-- 
cgit v1.2.3-70-g09d2


From cdc32546632354305afdcf399a5431138a31c9e0 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Fri, 21 Apr 2023 00:34:30 +0200
Subject: netfilter: nf_tables: rename function to destroy hook list

Rename nft_flowtable_hooks_destroy() by nft_hooks_destroy() to prepare
for netdev chain device updates.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 3f95fb24a0bc..77975b4b0fdc 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -7771,7 +7771,7 @@ err_unregister_net_hooks:
 	return err;
 }
 
-static void nft_flowtable_hooks_destroy(struct list_head *hook_list)
+static void nft_hooks_destroy(struct list_head *hook_list)
 {
 	struct nft_hook *hook, *next;
 
@@ -7952,7 +7952,7 @@ static int nf_tables_newflowtable(struct sk_buff *skb,
 					       &flowtable->hook_list,
 					       flowtable);
 	if (err < 0) {
-		nft_flowtable_hooks_destroy(&flowtable->hook_list);
+		nft_hooks_destroy(&flowtable->hook_list);
 		goto err4;
 	}
 
@@ -8789,7 +8789,7 @@ static void nft_commit_release(struct nft_trans *trans)
 	case NFT_MSG_DELFLOWTABLE:
 	case NFT_MSG_DESTROYFLOWTABLE:
 		if (nft_trans_flowtable_update(trans))
-			nft_flowtable_hooks_destroy(&nft_trans_flowtable_hooks(trans));
+			nft_hooks_destroy(&nft_trans_flowtable_hooks(trans));
 		else
 			nf_tables_flowtable_destroy(nft_trans_flowtable(trans));
 		break;
@@ -9440,7 +9440,7 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 		break;
 	case NFT_MSG_NEWFLOWTABLE:
 		if (nft_trans_flowtable_update(trans))
-			nft_flowtable_hooks_destroy(&nft_trans_flowtable_hooks(trans));
+			nft_hooks_destroy(&nft_trans_flowtable_hooks(trans));
 		else
 			nf_tables_flowtable_destroy(nft_trans_flowtable(trans));
 		break;
-- 
cgit v1.2.3-70-g09d2


From b9703ed44ffbfba85c103b9de01886a225e14b38 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Fri, 21 Apr 2023 00:34:31 +0200
Subject: netfilter: nf_tables: support for adding new devices to an existing
 netdev chain

This patch allows users to add devices to an existing netdev chain.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |   6 ++
 net/netfilter/nf_tables_api.c     | 217 ++++++++++++++++++++++++--------------
 2 files changed, 142 insertions(+), 81 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 58a4d217faaf..3ed21d2d5659 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1609,6 +1609,8 @@ struct nft_trans_chain {
 	struct nft_stats __percpu	*stats;
 	u8				policy;
 	u32				chain_id;
+	struct nft_base_chain		*basechain;
+	struct list_head		hook_list;
 };
 
 #define nft_trans_chain_update(trans)	\
@@ -1621,6 +1623,10 @@ struct nft_trans_chain {
 	(((struct nft_trans_chain *)trans->data)->policy)
 #define nft_trans_chain_id(trans)	\
 	(((struct nft_trans_chain *)trans->data)->chain_id)
+#define nft_trans_basechain(trans)	\
+	(((struct nft_trans_chain *)trans->data)->basechain)
+#define nft_trans_chain_hooks(trans)	\
+	(((struct nft_trans_chain *)trans->data)->hook_list)
 
 struct nft_trans_table {
 	bool				update;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 77975b4b0fdc..f8d8cace0c7d 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1582,7 +1582,8 @@ nla_put_failure:
 }
 
 static int nft_dump_basechain_hook(struct sk_buff *skb, int family,
-				   const struct nft_base_chain *basechain)
+				   const struct nft_base_chain *basechain,
+				   const struct list_head *hook_list)
 {
 	const struct nf_hook_ops *ops = &basechain->ops;
 	struct nft_hook *hook, *first = NULL;
@@ -1599,7 +1600,11 @@ static int nft_dump_basechain_hook(struct sk_buff *skb, int family,
 
 	if (nft_base_chain_netdev(family, ops->hooknum)) {
 		nest_devs = nla_nest_start_noflag(skb, NFTA_HOOK_DEVS);
-		list_for_each_entry(hook, &basechain->hook_list, list) {
+
+		if (!hook_list)
+			hook_list = &basechain->hook_list;
+
+		list_for_each_entry(hook, hook_list, list) {
 			if (!first)
 				first = hook;
 
@@ -1624,7 +1629,8 @@ nla_put_failure:
 static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net,
 				     u32 portid, u32 seq, int event, u32 flags,
 				     int family, const struct nft_table *table,
-				     const struct nft_chain *chain)
+				     const struct nft_chain *chain,
+				     const struct list_head *hook_list)
 {
 	struct nlmsghdr *nlh;
 
@@ -1649,7 +1655,7 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net,
 		const struct nft_base_chain *basechain = nft_base_chain(chain);
 		struct nft_stats __percpu *stats;
 
-		if (nft_dump_basechain_hook(skb, family, basechain))
+		if (nft_dump_basechain_hook(skb, family, basechain, hook_list))
 			goto nla_put_failure;
 
 		if (nla_put_be32(skb, NFTA_CHAIN_POLICY,
@@ -1684,7 +1690,8 @@ nla_put_failure:
 	return -1;
 }
 
-static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event)
+static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event,
+				   const struct list_head *hook_list)
 {
 	struct nftables_pernet *nft_net;
 	struct sk_buff *skb;
@@ -1704,7 +1711,7 @@ static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event)
 
 	err = nf_tables_fill_chain_info(skb, ctx->net, ctx->portid, ctx->seq,
 					event, flags, ctx->family, ctx->table,
-					ctx->chain);
+					ctx->chain, hook_list);
 	if (err < 0) {
 		kfree_skb(skb);
 		goto err;
@@ -1750,7 +1757,7 @@ static int nf_tables_dump_chains(struct sk_buff *skb,
 						      NFT_MSG_NEWCHAIN,
 						      NLM_F_MULTI,
 						      table->family, table,
-						      chain) < 0)
+						      chain, NULL) < 0)
 				goto done;
 
 			nl_dump_check_consistent(cb, nlmsg_hdr(skb));
@@ -1804,7 +1811,7 @@ static int nf_tables_getchain(struct sk_buff *skb, const struct nfnl_info *info,
 
 	err = nf_tables_fill_chain_info(skb2, net, NETLINK_CB(skb).portid,
 					info->nlh->nlmsg_seq, NFT_MSG_NEWCHAIN,
-					0, family, table, chain);
+					0, family, table, chain, NULL);
 	if (err < 0)
 		goto err_fill_chain_info;
 
@@ -2048,9 +2055,10 @@ static int nft_chain_parse_netdev(struct net *net,
 }
 
 static int nft_chain_parse_hook(struct net *net,
+				struct nft_base_chain *basechain,
 				const struct nlattr * const nla[],
 				struct nft_chain_hook *hook, u8 family,
-				struct netlink_ext_ack *extack, bool autoload)
+				struct netlink_ext_ack *extack)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
 	struct nlattr *ha[NFTA_HOOK_MAX + 1];
@@ -2066,31 +2074,46 @@ static int nft_chain_parse_hook(struct net *net,
 	if (err < 0)
 		return err;
 
-	if (ha[NFTA_HOOK_HOOKNUM] == NULL ||
-	    ha[NFTA_HOOK_PRIORITY] == NULL)
-		return -EINVAL;
+	if (!basechain) {
+		if (!ha[NFTA_HOOK_HOOKNUM] ||
+		    !ha[NFTA_HOOK_PRIORITY])
+			return -EINVAL;
 
-	hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM]));
-	hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY]));
+		hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM]));
+		hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY]));
 
-	type = __nft_chain_type_get(family, NFT_CHAIN_T_DEFAULT);
-	if (!type)
-		return -EOPNOTSUPP;
+		type = __nft_chain_type_get(family, NFT_CHAIN_T_DEFAULT);
+		if (!type)
+			return -EOPNOTSUPP;
 
-	if (nla[NFTA_CHAIN_TYPE]) {
-		type = nf_tables_chain_type_lookup(net, nla[NFTA_CHAIN_TYPE],
-						   family, autoload);
-		if (IS_ERR(type)) {
-			NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_TYPE]);
-			return PTR_ERR(type);
+		if (nla[NFTA_CHAIN_TYPE]) {
+			type = nf_tables_chain_type_lookup(net, nla[NFTA_CHAIN_TYPE],
+							   family, true);
+			if (IS_ERR(type)) {
+				NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_TYPE]);
+				return PTR_ERR(type);
+			}
 		}
-	}
-	if (hook->num >= NFT_MAX_HOOKS || !(type->hook_mask & (1 << hook->num)))
-		return -EOPNOTSUPP;
+		if (hook->num >= NFT_MAX_HOOKS || !(type->hook_mask & (1 << hook->num)))
+			return -EOPNOTSUPP;
 
-	if (type->type == NFT_CHAIN_T_NAT &&
-	    hook->priority <= NF_IP_PRI_CONNTRACK)
-		return -EOPNOTSUPP;
+		if (type->type == NFT_CHAIN_T_NAT &&
+		    hook->priority <= NF_IP_PRI_CONNTRACK)
+			return -EOPNOTSUPP;
+	} else {
+		if (ha[NFTA_HOOK_HOOKNUM]) {
+			hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM]));
+			if (hook->num != basechain->ops.hooknum)
+				return -EOPNOTSUPP;
+		}
+		if (ha[NFTA_HOOK_PRIORITY]) {
+			hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY]));
+			if (hook->priority != basechain->ops.priority)
+				return -EOPNOTSUPP;
+		}
+
+		type = basechain->type;
+	}
 
 	if (!try_module_get(type->owner)) {
 		if (nla[NFTA_CHAIN_TYPE])
@@ -2184,12 +2207,8 @@ static int nft_basechain_init(struct nft_base_chain *basechain, u8 family,
 		list_splice_init(&hook->list, &basechain->hook_list);
 		list_for_each_entry(h, &basechain->hook_list, list)
 			nft_basechain_hook_init(&h->ops, family, hook, chain);
-
-		basechain->ops.hooknum	= hook->num;
-		basechain->ops.priority	= hook->priority;
-	} else {
-		nft_basechain_hook_init(&basechain->ops, family, hook, chain);
 	}
+	nft_basechain_hook_init(&basechain->ops, family, hook, chain);
 
 	chain->flags |= NFT_CHAIN_BASE | flags;
 	basechain->policy = NF_ACCEPT;
@@ -2239,13 +2258,13 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask,
 
 	if (nla[NFTA_CHAIN_HOOK]) {
 		struct nft_stats __percpu *stats = NULL;
-		struct nft_chain_hook hook;
+		struct nft_chain_hook hook = {};
 
 		if (flags & NFT_CHAIN_BINDING)
 			return -EOPNOTSUPP;
 
-		err = nft_chain_parse_hook(net, nla, &hook, family, extack,
-					   true);
+		err = nft_chain_parse_hook(net, NULL, nla, &hook, family,
+					   extack);
 		if (err < 0)
 			return err;
 
@@ -2359,65 +2378,57 @@ err_destroy_chain:
 	return err;
 }
 
-static bool nft_hook_list_equal(struct list_head *hook_list1,
-				struct list_head *hook_list2)
-{
-	struct nft_hook *hook;
-	int n = 0, m = 0;
-
-	n = 0;
-	list_for_each_entry(hook, hook_list2, list) {
-		if (!nft_hook_list_find(hook_list1, hook))
-			return false;
-
-		n++;
-	}
-	list_for_each_entry(hook, hook_list1, list)
-		m++;
-
-	return n == m;
-}
-
 static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 			      u32 flags, const struct nlattr *attr,
 			      struct netlink_ext_ack *extack)
 {
 	const struct nlattr * const *nla = ctx->nla;
+	struct nft_base_chain *basechain = NULL;
 	struct nft_table *table = ctx->table;
 	struct nft_chain *chain = ctx->chain;
-	struct nft_base_chain *basechain;
+	struct nft_chain_hook hook = {};
 	struct nft_stats *stats = NULL;
-	struct nft_chain_hook hook;
+	struct nft_hook *h, *next;
 	struct nf_hook_ops *ops;
 	struct nft_trans *trans;
+	bool unregister = false;
 	int err;
 
 	if (chain->flags ^ flags)
 		return -EOPNOTSUPP;
 
+	INIT_LIST_HEAD(&hook.list);
+
 	if (nla[NFTA_CHAIN_HOOK]) {
 		if (!nft_is_base_chain(chain)) {
 			NL_SET_BAD_ATTR(extack, attr);
 			return -EEXIST;
 		}
-		err = nft_chain_parse_hook(ctx->net, nla, &hook, ctx->family,
-					   extack, false);
+
+		basechain = nft_base_chain(chain);
+		err = nft_chain_parse_hook(ctx->net, basechain, nla, &hook,
+					   ctx->family, extack);
 		if (err < 0)
 			return err;
 
-		basechain = nft_base_chain(chain);
 		if (basechain->type != hook.type) {
 			nft_chain_release_hook(&hook);
 			NL_SET_BAD_ATTR(extack, attr);
 			return -EEXIST;
 		}
 
-		if (nft_base_chain_netdev(ctx->family, hook.num)) {
-			if (!nft_hook_list_equal(&basechain->hook_list,
-						 &hook.list)) {
-				nft_chain_release_hook(&hook);
-				NL_SET_BAD_ATTR(extack, attr);
-				return -EEXIST;
+		if (nft_base_chain_netdev(ctx->family, basechain->ops.hooknum)) {
+			list_for_each_entry_safe(h, next, &hook.list, list) {
+				h->ops.pf	= basechain->ops.pf;
+				h->ops.hooknum	= basechain->ops.hooknum;
+				h->ops.priority	= basechain->ops.priority;
+				h->ops.priv	= basechain->ops.priv;
+				h->ops.hook	= basechain->ops.hook;
+
+				if (nft_hook_list_find(&basechain->hook_list, h)) {
+					list_del(&h->list);
+					kfree(h);
+				}
 			}
 		} else {
 			ops = &basechain->ops;
@@ -2428,7 +2439,6 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 				return -EEXIST;
 			}
 		}
-		nft_chain_release_hook(&hook);
 	}
 
 	if (nla[NFTA_CHAIN_HANDLE] &&
@@ -2439,24 +2449,43 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 					  nla[NFTA_CHAIN_NAME], genmask);
 		if (!IS_ERR(chain2)) {
 			NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_NAME]);
-			return -EEXIST;
+			err = -EEXIST;
+			goto err_hooks;
 		}
 	}
 
 	if (nla[NFTA_CHAIN_COUNTERS]) {
-		if (!nft_is_base_chain(chain))
-			return -EOPNOTSUPP;
+		if (!nft_is_base_chain(chain)) {
+			err = -EOPNOTSUPP;
+			goto err_hooks;
+		}
 
 		stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]);
-		if (IS_ERR(stats))
-			return PTR_ERR(stats);
+		if (IS_ERR(stats)) {
+			err = PTR_ERR(stats);
+			goto err_hooks;
+		}
 	}
 
+	if (!(table->flags & NFT_TABLE_F_DORMANT) &&
+	    nft_is_base_chain(chain) &&
+	    !list_empty(&hook.list)) {
+		basechain = nft_base_chain(chain);
+		ops = &basechain->ops;
+
+		if (nft_base_chain_netdev(table->family, basechain->ops.hooknum)) {
+			err = nft_netdev_register_hooks(ctx->net, &hook.list);
+			if (err < 0)
+				goto err_hooks;
+		}
+	}
+
+	unregister = true;
 	err = -ENOMEM;
 	trans = nft_trans_alloc(ctx, NFT_MSG_NEWCHAIN,
 				sizeof(struct nft_trans_chain));
 	if (trans == NULL)
-		goto err;
+		goto err_trans;
 
 	nft_trans_chain_stats(trans) = stats;
 	nft_trans_chain_update(trans) = true;
@@ -2475,7 +2504,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 		err = -ENOMEM;
 		name = nla_strdup(nla[NFTA_CHAIN_NAME], GFP_KERNEL_ACCOUNT);
 		if (!name)
-			goto err;
+			goto err_trans;
 
 		err = -EEXIST;
 		list_for_each_entry(tmp, &nft_net->commit_list, list) {
@@ -2486,18 +2515,35 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 			    strcmp(name, nft_trans_chain_name(tmp)) == 0) {
 				NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_NAME]);
 				kfree(name);
-				goto err;
+				goto err_trans;
 			}
 		}
 
 		nft_trans_chain_name(trans) = name;
 	}
+
+	nft_trans_basechain(trans) = basechain;
+	INIT_LIST_HEAD(&nft_trans_chain_hooks(trans));
+	list_splice(&hook.list, &nft_trans_chain_hooks(trans));
+
 	nft_trans_commit_list_add_tail(ctx->net, trans);
 
 	return 0;
-err:
+
+err_trans:
 	free_percpu(stats);
 	kfree(trans);
+err_hooks:
+	if (nla[NFTA_CHAIN_HOOK]) {
+		list_for_each_entry_safe(h, next, &hook.list, list) {
+			if (unregister)
+				nf_unregister_net_hook(ctx->net, &h->ops);
+			list_del(&h->list);
+			kfree_rcu(h, rcu);
+		}
+		module_put(hook.type->owner);
+	}
+
 	return err;
 }
 
@@ -9244,19 +9290,22 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		case NFT_MSG_NEWCHAIN:
 			if (nft_trans_chain_update(trans)) {
 				nft_chain_commit_update(trans);
-				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN);
+				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN,
+						       &nft_trans_chain_hooks(trans));
+				list_splice(&nft_trans_chain_hooks(trans),
+					    &nft_trans_basechain(trans)->hook_list);
 				/* trans destroyed after rcu grace period */
 			} else {
 				nft_chain_commit_drop_policy(trans);
 				nft_clear(net, trans->ctx.chain);
-				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN);
+				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN, NULL);
 				nft_trans_destroy(trans);
 			}
 			break;
 		case NFT_MSG_DELCHAIN:
 		case NFT_MSG_DESTROYCHAIN:
 			nft_chain_del(trans->ctx.chain);
-			nf_tables_chain_notify(&trans->ctx, trans->msg_type);
+			nf_tables_chain_notify(&trans->ctx, trans->msg_type, NULL);
 			nf_tables_unregister_hook(trans->ctx.net,
 						  trans->ctx.table,
 						  trans->ctx.chain);
@@ -9423,7 +9472,10 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 		nf_tables_table_destroy(&trans->ctx);
 		break;
 	case NFT_MSG_NEWCHAIN:
-		nf_tables_chain_destroy(&trans->ctx);
+		if (nft_trans_chain_update(trans))
+			nft_hooks_destroy(&nft_trans_chain_hooks(trans));
+		else
+			nf_tables_chain_destroy(&trans->ctx);
 		break;
 	case NFT_MSG_NEWRULE:
 		nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans));
@@ -9486,6 +9538,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			break;
 		case NFT_MSG_NEWCHAIN:
 			if (nft_trans_chain_update(trans)) {
+				nft_netdev_unregister_hooks(net,
+							    &nft_trans_chain_hooks(trans),
+							    true);
 				free_percpu(nft_trans_chain_stats(trans));
 				kfree(nft_trans_chain_name(trans));
 				nft_trans_destroy(trans);
-- 
cgit v1.2.3-70-g09d2


From 7d937b107108f918906fc867ea070cc3d0cbcd32 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Fri, 21 Apr 2023 00:34:32 +0200
Subject: netfilter: nf_tables: support for deleting devices in an existing
 netdev chain

This patch allows for deleting devices in an existing netdev chain.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 99 ++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 88 insertions(+), 11 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index f8d8cace0c7d..c55ac3d57c51 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1646,7 +1646,7 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net,
 			 NFTA_CHAIN_PAD))
 		goto nla_put_failure;
 
-	if (event == NFT_MSG_DELCHAIN) {
+	if (event == NFT_MSG_DELCHAIN && !hook_list) {
 		nlmsg_end(skb, nlh);
 		return 0;
 	}
@@ -2667,6 +2667,59 @@ static int nf_tables_newchain(struct sk_buff *skb, const struct nfnl_info *info,
 	return nf_tables_addchain(&ctx, family, genmask, policy, flags, extack);
 }
 
+static int nft_delchain_hook(struct nft_ctx *ctx, struct nft_chain *chain,
+			     struct netlink_ext_ack *extack)
+{
+	const struct nlattr * const *nla = ctx->nla;
+	struct nft_chain_hook chain_hook = {};
+	struct nft_base_chain *basechain;
+	struct nft_hook *this, *hook;
+	LIST_HEAD(chain_del_list);
+	struct nft_trans *trans;
+	int err;
+
+	if (!nft_is_base_chain(chain))
+		return -EOPNOTSUPP;
+
+	basechain = nft_base_chain(chain);
+	err = nft_chain_parse_hook(ctx->net, basechain, nla, &chain_hook,
+				   ctx->family, extack);
+	if (err < 0)
+		return err;
+
+	list_for_each_entry(this, &chain_hook.list, list) {
+		hook = nft_hook_list_find(&basechain->hook_list, this);
+		if (!hook) {
+			err = -ENOENT;
+			goto err_chain_del_hook;
+		}
+		list_move(&hook->list, &chain_del_list);
+	}
+
+	trans = nft_trans_alloc(ctx, NFT_MSG_DELCHAIN,
+				sizeof(struct nft_trans_chain));
+	if (!trans) {
+		err = -ENOMEM;
+		goto err_chain_del_hook;
+	}
+
+	nft_trans_basechain(trans) = basechain;
+	nft_trans_chain_update(trans) = true;
+	INIT_LIST_HEAD(&nft_trans_chain_hooks(trans));
+	list_splice(&chain_del_list, &nft_trans_chain_hooks(trans));
+	nft_chain_release_hook(&chain_hook);
+
+	nft_trans_commit_list_add_tail(ctx->net, trans);
+
+	return 0;
+
+err_chain_del_hook:
+	list_splice(&chain_del_list, &basechain->hook_list);
+	nft_chain_release_hook(&chain_hook);
+
+	return err;
+}
+
 static int nf_tables_delchain(struct sk_buff *skb, const struct nfnl_info *info,
 			      const struct nlattr * const nla[])
 {
@@ -2707,12 +2760,19 @@ static int nf_tables_delchain(struct sk_buff *skb, const struct nfnl_info *info,
 		return PTR_ERR(chain);
 	}
 
+	nft_ctx_init(&ctx, net, skb, info->nlh, family, table, chain, nla);
+
+	if (nla[NFTA_CHAIN_HOOK]) {
+		if (chain->flags & NFT_CHAIN_HW_OFFLOAD)
+			return -EOPNOTSUPP;
+
+		return nft_delchain_hook(&ctx, chain, extack);
+	}
+
 	if (info->nlh->nlmsg_flags & NLM_F_NONREC &&
 	    chain->use > 0)
 		return -EBUSY;
 
-	nft_ctx_init(&ctx, net, skb, info->nlh, family, table, chain, nla);
-
 	use = chain->use;
 	list_for_each_entry(rule, &chain->rules, list) {
 		if (!nft_is_active_next(net, rule))
@@ -8812,7 +8872,10 @@ static void nft_commit_release(struct nft_trans *trans)
 		break;
 	case NFT_MSG_DELCHAIN:
 	case NFT_MSG_DESTROYCHAIN:
-		nf_tables_chain_destroy(&trans->ctx);
+		if (nft_trans_chain_update(trans))
+			nft_hooks_destroy(&nft_trans_chain_hooks(trans));
+		else
+			nf_tables_chain_destroy(&trans->ctx);
 		break;
 	case NFT_MSG_DELRULE:
 	case NFT_MSG_DESTROYRULE:
@@ -9304,11 +9367,20 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			break;
 		case NFT_MSG_DELCHAIN:
 		case NFT_MSG_DESTROYCHAIN:
-			nft_chain_del(trans->ctx.chain);
-			nf_tables_chain_notify(&trans->ctx, trans->msg_type, NULL);
-			nf_tables_unregister_hook(trans->ctx.net,
-						  trans->ctx.table,
-						  trans->ctx.chain);
+			if (nft_trans_chain_update(trans)) {
+				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
+						       &nft_trans_chain_hooks(trans));
+				nft_netdev_unregister_hooks(net,
+							    &nft_trans_chain_hooks(trans),
+							    true);
+			} else {
+				nft_chain_del(trans->ctx.chain);
+				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
+						       NULL);
+				nf_tables_unregister_hook(trans->ctx.net,
+							  trans->ctx.table,
+							  trans->ctx.chain);
+			}
 			break;
 		case NFT_MSG_NEWRULE:
 			nft_clear(trans->ctx.net, nft_trans_rule(trans));
@@ -9558,8 +9630,13 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			break;
 		case NFT_MSG_DELCHAIN:
 		case NFT_MSG_DESTROYCHAIN:
-			trans->ctx.table->use++;
-			nft_clear(trans->ctx.net, trans->ctx.chain);
+			if (nft_trans_chain_update(trans)) {
+				list_splice(&nft_trans_chain_hooks(trans),
+					    &nft_trans_basechain(trans)->hook_list);
+			} else {
+				trans->ctx.table->use++;
+				nft_clear(trans->ctx.net, trans->ctx.chain);
+			}
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_NEWRULE:
-- 
cgit v1.2.3-70-g09d2


From 207296f1a03bfead0110ffc4f192f242100ce4ff Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Fri, 21 Apr 2023 00:34:33 +0200
Subject: netfilter: nf_tables: allow to create netdev chain without device

Relax netdev chain creation to allow for loading the ruleset, then
adding/deleting devices at a later stage. Hardware offload does not
support for this feature yet.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 23 +++++++++++------------
 1 file changed, 11 insertions(+), 12 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index c55ac3d57c51..09542951656c 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -2023,10 +2023,9 @@ struct nft_chain_hook {
 	struct list_head		list;
 };
 
-static int nft_chain_parse_netdev(struct net *net,
-				  struct nlattr *tb[],
+static int nft_chain_parse_netdev(struct net *net, struct nlattr *tb[],
 				  struct list_head *hook_list,
-				  struct netlink_ext_ack *extack)
+				  struct netlink_ext_ack *extack, u32 flags)
 {
 	struct nft_hook *hook;
 	int err;
@@ -2045,12 +2044,12 @@ static int nft_chain_parse_netdev(struct net *net,
 		if (err < 0)
 			return err;
 
-		if (list_empty(hook_list))
-			return -EINVAL;
-	} else {
-		return -EINVAL;
 	}
 
+	if (flags & NFT_CHAIN_HW_OFFLOAD &&
+	    list_empty(hook_list))
+		return -EINVAL;
+
 	return 0;
 }
 
@@ -2058,7 +2057,7 @@ static int nft_chain_parse_hook(struct net *net,
 				struct nft_base_chain *basechain,
 				const struct nlattr * const nla[],
 				struct nft_chain_hook *hook, u8 family,
-				struct netlink_ext_ack *extack)
+				u32 flags, struct netlink_ext_ack *extack)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
 	struct nlattr *ha[NFTA_HOOK_MAX + 1];
@@ -2125,7 +2124,7 @@ static int nft_chain_parse_hook(struct net *net,
 
 	INIT_LIST_HEAD(&hook->list);
 	if (nft_base_chain_netdev(family, hook->num)) {
-		err = nft_chain_parse_netdev(net, ha, &hook->list, extack);
+		err = nft_chain_parse_netdev(net, ha, &hook->list, extack, flags);
 		if (err < 0) {
 			module_put(type->owner);
 			return err;
@@ -2263,7 +2262,7 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask,
 		if (flags & NFT_CHAIN_BINDING)
 			return -EOPNOTSUPP;
 
-		err = nft_chain_parse_hook(net, NULL, nla, &hook, family,
+		err = nft_chain_parse_hook(net, NULL, nla, &hook, family, flags,
 					   extack);
 		if (err < 0)
 			return err;
@@ -2407,7 +2406,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 
 		basechain = nft_base_chain(chain);
 		err = nft_chain_parse_hook(ctx->net, basechain, nla, &hook,
-					   ctx->family, extack);
+					   ctx->family, flags, extack);
 		if (err < 0)
 			return err;
 
@@ -2683,7 +2682,7 @@ static int nft_delchain_hook(struct nft_ctx *ctx, struct nft_chain *chain,
 
 	basechain = nft_base_chain(chain);
 	err = nft_chain_parse_hook(ctx->net, basechain, nla, &chain_hook,
-				   ctx->family, extack);
+				   ctx->family, chain->flags, extack);
 	if (err < 0)
 		return err;
 
-- 
cgit v1.2.3-70-g09d2


From f3b766d981310989d524b076065a93b8f11bcab2 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Tue, 18 Apr 2023 22:01:41 +0300
Subject: net: phy: add basic driver for NXP CBTX PHY

The CBTX PHY is a Fast Ethernet PHY integrated into the SJA1110 A/B/C
automotive Ethernet switches.

It was hoped it would work with the Generic PHY driver, but alas, it
doesn't. The most important reason why is that the PHY is powered down
by default, and it needs a vendor register to power it on.

It has a linear memory map that is accessed over SPI by the SJA1110
switch driver, which exposes a fake MDIO controller. It has the
following (and only the following) standard clause 22 registers:

0x0: MII_BMCR
0x1: MII_BMSR
0x2: MII_PHYSID1
0x3: MII_PHYSID2
0x4: MII_ADVERTISE
0x5: MII_LPA
0x6: MII_EXPANSION
0x7: the missing MII_NPAGE for Next Page Transmit Register

Every other register is vendor-defined.

The register map expands the standard clause 22 5-bit address space of
0x20 registers, however the driver does not need to access the extra
registers for now (and hopefully never). If it ever needs to do that, it
is possible to implement a fake (software) page switching mechanism
between the PHY driver and the SJA1110 MDIO controller driver.

Also, Auto-MDIX is turned off by default in hardware, the driver turns
it on by default and reports the current status. I've tested this with a
VSC8514 link partner and a crossover cable, by forcing the mode on the
link partner, and seeing that the CBTX PHY always sees the reverse of
the mode forced on the VSC8514 (and that traffic works). The link
doesn't come up (as expected) if MDI modes are forced on both ends in
the same way (with the cross-over cable, that is).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230418190141.1040562-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/Kconfig    |   6 ++
 drivers/net/phy/Makefile   |   1 +
 drivers/net/phy/nxp-cbtx.c | 227 +++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 234 insertions(+)
 create mode 100644 drivers/net/phy/nxp-cbtx.c

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 513675ae4dd2..bcfba077b80e 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -273,6 +273,12 @@ config NATIONAL_PHY
 	help
 	  Currently supports the DP83865 PHY.
 
+config NXP_CBTX_PHY
+	tristate "NXP 100BASE-TX PHYs"
+	help
+	  Support the 100BASE-TX PHY integrated on the SJA1110 automotive
+	  switch family.
+
 config NXP_C45_TJA11XX_PHY
 	tristate "NXP C45 TJA11XX PHYs"
 	depends on PTP_1588_CLOCK_OPTIONAL
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index b5138066ba04..ae11bf20b46e 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -79,6 +79,7 @@ obj-$(CONFIG_MOTORCOMM_PHY)	+= motorcomm.o
 obj-$(CONFIG_NATIONAL_PHY)	+= national.o
 obj-$(CONFIG_NCN26000_PHY)	+= ncn26000.o
 obj-$(CONFIG_NXP_C45_TJA11XX_PHY)	+= nxp-c45-tja11xx.o
+obj-$(CONFIG_NXP_CBTX_PHY)	+= nxp-cbtx.o
 obj-$(CONFIG_NXP_TJA11XX_PHY)	+= nxp-tja11xx.o
 obj-$(CONFIG_QSEMI_PHY)		+= qsemi.o
 obj-$(CONFIG_REALTEK_PHY)	+= realtek.o
diff --git a/drivers/net/phy/nxp-cbtx.c b/drivers/net/phy/nxp-cbtx.c
new file mode 100644
index 000000000000..145703f0a406
--- /dev/null
+++ b/drivers/net/phy/nxp-cbtx.c
@@ -0,0 +1,227 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Driver for 100BASE-TX PHY embedded into NXP SJA1110 switch
+ *
+ * Copyright 2022-2023 NXP
+ */
+
+#include <linux/kernel.h>
+#include <linux/mii.h>
+#include <linux/module.h>
+#include <linux/phy.h>
+
+#define PHY_ID_CBTX_SJA1110			0x001bb020
+
+/* Registers */
+#define  CBTX_MODE_CTRL_STAT			0x11
+#define  CBTX_PDOWN_CTRL			0x18
+#define  CBTX_RX_ERR_COUNTER			0x1a
+#define  CBTX_IRQ_STAT				0x1d
+#define  CBTX_IRQ_ENABLE			0x1e
+
+/* Fields */
+#define CBTX_MODE_CTRL_STAT_AUTO_MDIX_EN	BIT(7)
+#define CBTX_MODE_CTRL_STAT_MDIX_MODE		BIT(6)
+
+#define CBTX_PDOWN_CTL_TRUE_PDOWN		BIT(0)
+
+#define CBTX_IRQ_ENERGYON			BIT(7)
+#define CBTX_IRQ_AN_COMPLETE			BIT(6)
+#define CBTX_IRQ_REM_FAULT			BIT(5)
+#define CBTX_IRQ_LINK_DOWN			BIT(4)
+#define CBTX_IRQ_AN_LP_ACK			BIT(3)
+#define CBTX_IRQ_PARALLEL_DETECT_FAULT		BIT(2)
+#define CBTX_IRQ_AN_PAGE_RECV			BIT(1)
+
+static int cbtx_soft_reset(struct phy_device *phydev)
+{
+	int ret;
+
+	/* Can't soft reset unless we remove PHY from true power down mode */
+	ret = phy_clear_bits(phydev, CBTX_PDOWN_CTRL,
+			     CBTX_PDOWN_CTL_TRUE_PDOWN);
+	if (ret)
+		return ret;
+
+	return genphy_soft_reset(phydev);
+}
+
+static int cbtx_config_init(struct phy_device *phydev)
+{
+	/* Wait for cbtx_config_aneg() to kick in and apply this */
+	phydev->mdix_ctrl = ETH_TP_MDI_AUTO;
+
+	return 0;
+}
+
+static int cbtx_mdix_status(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_read(phydev, CBTX_MODE_CTRL_STAT);
+	if (ret < 0)
+		return ret;
+
+	if (ret & CBTX_MODE_CTRL_STAT_MDIX_MODE)
+		phydev->mdix = ETH_TP_MDI_X;
+	else
+		phydev->mdix = ETH_TP_MDI;
+
+	return 0;
+}
+
+static int cbtx_read_status(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = cbtx_mdix_status(phydev);
+	if (ret)
+		return ret;
+
+	return genphy_read_status(phydev);
+}
+
+static int cbtx_mdix_config(struct phy_device *phydev)
+{
+	int ret;
+
+	switch (phydev->mdix_ctrl) {
+	case ETH_TP_MDI_AUTO:
+		return phy_set_bits(phydev, CBTX_MODE_CTRL_STAT,
+				    CBTX_MODE_CTRL_STAT_AUTO_MDIX_EN);
+	case ETH_TP_MDI:
+		ret = phy_clear_bits(phydev, CBTX_MODE_CTRL_STAT,
+				     CBTX_MODE_CTRL_STAT_AUTO_MDIX_EN);
+		if (ret)
+			return ret;
+
+		return phy_clear_bits(phydev, CBTX_MODE_CTRL_STAT,
+				      CBTX_MODE_CTRL_STAT_MDIX_MODE);
+	case ETH_TP_MDI_X:
+		ret = phy_clear_bits(phydev, CBTX_MODE_CTRL_STAT,
+				     CBTX_MODE_CTRL_STAT_AUTO_MDIX_EN);
+		if (ret)
+			return ret;
+
+		return phy_set_bits(phydev, CBTX_MODE_CTRL_STAT,
+				    CBTX_MODE_CTRL_STAT_MDIX_MODE);
+	}
+
+	return 0;
+}
+
+static int cbtx_config_aneg(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = cbtx_mdix_config(phydev);
+	if (ret)
+		return ret;
+
+	return genphy_config_aneg(phydev);
+}
+
+static int cbtx_ack_interrupts(struct phy_device *phydev)
+{
+	return phy_read(phydev, CBTX_IRQ_STAT);
+}
+
+static int cbtx_config_intr(struct phy_device *phydev)
+{
+	int ret;
+
+	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
+		ret = cbtx_ack_interrupts(phydev);
+		if (ret < 0)
+			return ret;
+
+		ret = phy_write(phydev, CBTX_IRQ_ENABLE, CBTX_IRQ_LINK_DOWN |
+				CBTX_IRQ_AN_COMPLETE | CBTX_IRQ_ENERGYON);
+		if (ret)
+			return ret;
+	} else {
+		ret = phy_write(phydev, CBTX_IRQ_ENABLE, 0);
+		if (ret)
+			return ret;
+
+		ret = cbtx_ack_interrupts(phydev);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+static irqreturn_t cbtx_handle_interrupt(struct phy_device *phydev)
+{
+	int irq_stat, irq_enabled;
+
+	irq_stat = cbtx_ack_interrupts(phydev);
+	if (irq_stat < 0) {
+		phy_error(phydev);
+		return IRQ_NONE;
+	}
+
+	irq_enabled = phy_read(phydev, CBTX_IRQ_ENABLE);
+	if (irq_enabled < 0) {
+		phy_error(phydev);
+		return IRQ_NONE;
+	}
+
+	if (!(irq_enabled & irq_stat))
+		return IRQ_NONE;
+
+	phy_trigger_machine(phydev);
+
+	return IRQ_HANDLED;
+}
+
+static int cbtx_get_sset_count(struct phy_device *phydev)
+{
+	return 1;
+}
+
+static void cbtx_get_strings(struct phy_device *phydev, u8 *data)
+{
+	strncpy(data, "100btx_rx_err", ETH_GSTRING_LEN);
+}
+
+static void cbtx_get_stats(struct phy_device *phydev,
+			   struct ethtool_stats *stats, u64 *data)
+{
+	int ret;
+
+	ret = phy_read(phydev, CBTX_RX_ERR_COUNTER);
+	data[0] = (ret < 0) ? U64_MAX : ret;
+}
+
+static struct phy_driver cbtx_driver[] = {
+	{
+		PHY_ID_MATCH_MODEL(PHY_ID_CBTX_SJA1110),
+		.name			= "NXP CBTX (SJA1110)",
+		/* PHY_BASIC_FEATURES */
+		.soft_reset		= cbtx_soft_reset,
+		.config_init		= cbtx_config_init,
+		.suspend		= genphy_suspend,
+		.resume			= genphy_resume,
+		.config_intr		= cbtx_config_intr,
+		.handle_interrupt	= cbtx_handle_interrupt,
+		.read_status		= cbtx_read_status,
+		.config_aneg		= cbtx_config_aneg,
+		.get_sset_count		= cbtx_get_sset_count,
+		.get_strings		= cbtx_get_strings,
+		.get_stats		= cbtx_get_stats,
+	},
+};
+
+module_phy_driver(cbtx_driver);
+
+static struct mdio_device_id __maybe_unused cbtx_tbl[] = {
+	{ PHY_ID_MATCH_MODEL(PHY_ID_CBTX_SJA1110) },
+	{ },
+};
+
+MODULE_DEVICE_TABLE(mdio, cbtx_tbl);
+
+MODULE_AUTHOR("Vladimir Oltean <vladimir.oltean@nxp.com>");
+MODULE_DESCRIPTION("NXP CBTX PHY driver");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-70-g09d2


From 461bb5b97049a149278f2c27a3aa12af16da6a2e Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 20 Apr 2023 15:36:07 +0300
Subject: net: dpaa: Fix uninitialized variable in dpaa_stop()

The return value is not initialized on the success path.

Fixes: 901bdff2f529 ("net: fman: Change return type of disable to void")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Reviewed-by: Sean Anderson <sean.anderson@seco.com>
Link: https://lore.kernel.org/r/8c9dc377-8495-495f-a4e5-4d2d0ee12f0c@kili.mountain
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 9318a2554056..f96196617121 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -299,7 +299,8 @@ static int dpaa_stop(struct net_device *net_dev)
 {
 	struct mac_device *mac_dev;
 	struct dpaa_priv *priv;
-	int i, err, error;
+	int i, error;
+	int err = 0;
 
 	priv = netdev_priv(net_dev);
 	mac_dev = priv->mac_dev;
-- 
cgit v1.2.3-70-g09d2


From 63cfd210034c772fad047afa13dd5a4664b0a72e Mon Sep 17 00:00:00 2001
From: Huanhuan Wang <huanhuan.wang@corigine.com>
Date: Thu, 20 Apr 2023 16:01:25 +0200
Subject: nfp: fix incorrect pointer deference when offloading IPsec with
 bonding

There are two pointers in struct xfrm_dev_offload, *dev, *real_dev.
The *dev points whether bonding interface or real interface, if
bonding IPsec offload is used, it points bonding interface; if not,
it points real interface. And *real_dev always points real interface.
So nfp should always use real_dev instead of dev.

Prior to this change the system becomes unresponsive when offloading
IPsec for a device which is a lower device to a bonding device.

Fixes: 859a497fe80c ("nfp: implement xfrm callbacks and expose ipsec offload feature to upper layer")
CC: stable@vger.kernel.org
Signed-off-by: Huanhuan Wang <huanhuan.wang@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Link: https://lore.kernel.org/r/20230420140125.38521-1-louis.peens@corigine.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/netronome/nfp/crypto/ipsec.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/crypto/ipsec.c b/drivers/net/ethernet/netronome/nfp/crypto/ipsec.c
index c0dcce8ae437..b1f026b81dea 100644
--- a/drivers/net/ethernet/netronome/nfp/crypto/ipsec.c
+++ b/drivers/net/ethernet/netronome/nfp/crypto/ipsec.c
@@ -269,7 +269,7 @@ static void set_sha2_512hmac(struct nfp_ipsec_cfg_add_sa *cfg, int *trunc_len)
 static int nfp_net_xfrm_add_state(struct xfrm_state *x,
 				  struct netlink_ext_ack *extack)
 {
-	struct net_device *netdev = x->xso.dev;
+	struct net_device *netdev = x->xso.real_dev;
 	struct nfp_ipsec_cfg_mssg msg = {};
 	int i, key_len, trunc_len, err = 0;
 	struct nfp_ipsec_cfg_add_sa *cfg;
@@ -513,7 +513,7 @@ static void nfp_net_xfrm_del_state(struct xfrm_state *x)
 		.cmd = NFP_IPSEC_CFG_MSSG_INV_SA,
 		.sa_idx = x->xso.offload_handle - 1,
 	};
-	struct net_device *netdev = x->xso.dev;
+	struct net_device *netdev = x->xso.real_dev;
 	struct nfp_net *nn;
 	int err;
 
-- 
cgit v1.2.3-70-g09d2


From 7041101ff6c3073fd8f2e99920f535b111c929cb Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Thu, 20 Apr 2023 16:59:46 +0200
Subject: net/sched: sch_fq: fix integer overflow of "credit"

if sch_fq is configured with "initial quantum" having values greater than
INT_MAX, the first assignment of "credit" does signed integer overflow to
a very negative value.
In this situation, the syzkaller script provided by Cristoph triggers the
CPU soft-lockup warning even with few sockets. It's not an infinite loop,
but "credit" wasn't probably meant to be minus 2Gb for each new flow.
Capping "initial quantum" to INT_MAX proved to fix the issue.

v2: validation of "initial quantum" is done in fq_policy, instead of open
    coding in fq_change() _ suggested by Jakub Kicinski

Reported-by: Christoph Paasch <cpaasch@apple.com>
Link: https://github.com/multipath-tcp/mptcp_net-next/issues/377
Fixes: afe4fd062416 ("pkt_sched: fq: Fair Queue packet scheduler")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/7b3a3c7e36d03068707a021760a194a8eb5ad41a.1682002300.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/sch_fq.c                                 |  6 +++++-
 .../selftests/tc-testing/tc-tests/qdiscs/fq.json   | 22 ++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index 48d14fb90ba0..f59a2cb2c803 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -779,13 +779,17 @@ static int fq_resize(struct Qdisc *sch, u32 log)
 	return 0;
 }
 
+static struct netlink_range_validation iq_range = {
+	.max = INT_MAX,
+};
+
 static const struct nla_policy fq_policy[TCA_FQ_MAX + 1] = {
 	[TCA_FQ_UNSPEC]			= { .strict_start_type = TCA_FQ_TIMER_SLACK },
 
 	[TCA_FQ_PLIMIT]			= { .type = NLA_U32 },
 	[TCA_FQ_FLOW_PLIMIT]		= { .type = NLA_U32 },
 	[TCA_FQ_QUANTUM]		= { .type = NLA_U32 },
-	[TCA_FQ_INITIAL_QUANTUM]	= { .type = NLA_U32 },
+	[TCA_FQ_INITIAL_QUANTUM]	= NLA_POLICY_FULL_RANGE(NLA_U32, &iq_range),
 	[TCA_FQ_RATE_ENABLE]		= { .type = NLA_U32 },
 	[TCA_FQ_FLOW_DEFAULT_RATE]	= { .type = NLA_U32 },
 	[TCA_FQ_FLOW_MAX_RATE]		= { .type = NLA_U32 },
diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json
index 8acb904d1419..3593fb8f79ad 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json
@@ -114,6 +114,28 @@
             "$IP link del dev $DUMMY type dummy"
         ]
     },
+    {
+        "id": "10f7",
+        "name": "Create FQ with invalid initial_quantum setting",
+        "category": [
+            "qdisc",
+            "fq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true"
+        ],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root fq initial_quantum 0x80000000",
+        "expExitCode": "2",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "qdisc fq 1: root.*initial_quantum 2048Mb",
+        "matchCount": "0",
+        "teardown": [
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
     {
         "id": "9398",
         "name": "Create FQ with maxrate setting",
-- 
cgit v1.2.3-70-g09d2


From 6aa445e39693bff9c98b12f960e66b4e18c7378b Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Thu, 20 Apr 2023 19:37:23 +0200
Subject: net/handshake: Fix section mismatch in handshake_exit

If CONFIG_NET_NS=n (e.g. m68k/defconfig):

    WARNING: modpost: vmlinux.o: section mismatch in reference: handshake_exit (section: .exit.text) -> handshake_genl_net_ops (section: .init.data)
    ERROR: modpost: Section mismatches detected.

Fix this by dropping the __net_initdata tag from handshake_genl_net_ops.

Fixes: 3b3009ea8abb713b ("net/handshake: Create a NETLINK service for handling handshake requests")
Reported-by: noreply@ellerman.id.au
Closes: http://kisskb.ellerman.id.au/kisskb/buildresult/14912987
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Link: https://lore.kernel.org/r/20230420173723.3773434-1-geert@linux-m68k.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/handshake/netlink.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c
index 8ea0ff993f9f..35c9c445e0b8 100644
--- a/net/handshake/netlink.c
+++ b/net/handshake/netlink.c
@@ -249,7 +249,7 @@ static void __net_exit handshake_net_exit(struct net *net)
 	}
 }
 
-static struct pernet_operations __net_initdata handshake_genl_net_ops = {
+static struct pernet_operations handshake_genl_net_ops = {
 	.init		= handshake_net_init,
 	.exit		= handshake_net_exit,
 	.id		= &handshake_net_id,
-- 
cgit v1.2.3-70-g09d2


From 2cc8a008d62f3c04eeb7ec6fe59e542802bb8df3 Mon Sep 17 00:00:00 2001
From: Ivan Vecera <ivecera@redhat.com>
Date: Thu, 20 Apr 2023 20:36:33 +0200
Subject: net/sched: cls_api: Initialize miss_cookie_node when action miss is
 not used

Function tcf_exts_init_ex() sets exts->miss_cookie_node ptr only
when use_action_miss is true so it assumes in other case that
the field is set to NULL by the caller. If not then the field
contains garbage and subsequent tcf_exts_destroy() call results
in a crash.
Ensure that the field .miss_cookie_node pointer is NULL when
use_action_miss parameter is false to avoid this potential scenario.

Fixes: 80cd22c35c90 ("net/sched: cls_api: Support hardware miss to tc action")
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Reviewed-by: Pedro Tammela <pctammela@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230420183634.1139391-1-ivecera@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/cls_api.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 35785a36c802..3c3629c9e7b6 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -3211,6 +3211,7 @@ int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action,
 #ifdef CONFIG_NET_CLS_ACT
 	exts->type = 0;
 	exts->nr_actions = 0;
+	exts->miss_cookie_node = NULL;
 	/* Note: we do not own yet a reference on net.
 	 * This reference might be taken later from tcf_exts_get_net().
 	 */
-- 
cgit v1.2.3-70-g09d2


From 33c1af8e2c75ab601524e25efeeed0173d703ce7 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 20 Apr 2023 23:36:31 +0200
Subject: net: dsa: qca8k: fix LEDS_CLASS dependency

With LEDS_CLASS=m, a built-in qca8k driver fails to link:

arm-linux-gnueabi-ld: drivers/net/dsa/qca/qca8k-leds.o: in function `qca8k_setup_led_ctrl':
qca8k-leds.c:(.text+0x1ea): undefined reference to `devm_led_classdev_register_ext'

Change the dependency to avoid the broken configuration.

Fixes: 1e264f9d2918 ("net: dsa: qca8k: add LEDs basic support")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230420213639.2243388-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/qca/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/qca/Kconfig b/drivers/net/dsa/qca/Kconfig
index 7a86d6d6a246..4347b42c50fd 100644
--- a/drivers/net/dsa/qca/Kconfig
+++ b/drivers/net/dsa/qca/Kconfig
@@ -19,7 +19,7 @@ config NET_DSA_QCA8K
 config NET_DSA_QCA8K_LEDS_SUPPORT
 	bool "Qualcomm Atheros QCA8K Ethernet switch family LEDs support"
 	depends on NET_DSA_QCA8K
-	depends on LEDS_CLASS
+	depends on LEDS_CLASS=y || LEDS_CLASS=NET_DSA_QCA8K
 	help
 	  This enabled support for LEDs present on the Qualcomm Atheros
 	  QCA8K Ethernet switch chips.
-- 
cgit v1.2.3-70-g09d2


From 418a73074da9182f571e467eaded03ea501f3281 Mon Sep 17 00:00:00 2001
From: Maxime Bizon <mbizon@freebox.fr>
Date: Thu, 20 Apr 2023 20:25:08 +0200
Subject: net: dst: fix missing initialization of rt_uncached

xfrm_alloc_dst() followed by xfrm4_dst_destroy(), without a
xfrm4_fill_dst() call in between, causes the following BUG:

 BUG: spinlock bad magic on CPU#0, fbxhostapd/732
  lock: 0x890b7668, .magic: 890b7668, .owner: <none>/-1, .owner_cpu: 0
 CPU: 0 PID: 732 Comm: fbxhostapd Not tainted 6.3.0-rc6-next-20230414-00613-ge8de66369925-dirty #9
 Hardware name: Marvell Kirkwood (Flattened Device Tree)
  unwind_backtrace from show_stack+0x10/0x14
  show_stack from dump_stack_lvl+0x28/0x30
  dump_stack_lvl from do_raw_spin_lock+0x20/0x80
  do_raw_spin_lock from rt_del_uncached_list+0x30/0x64
  rt_del_uncached_list from xfrm4_dst_destroy+0x3c/0xbc
  xfrm4_dst_destroy from dst_destroy+0x5c/0xb0
  dst_destroy from rcu_process_callbacks+0xc4/0xec
  rcu_process_callbacks from __do_softirq+0xb4/0x22c
  __do_softirq from call_with_stack+0x1c/0x24
  call_with_stack from do_softirq+0x60/0x6c
  do_softirq from __local_bh_enable_ip+0xa0/0xcc

Patch "net: dst: Prevent false sharing vs. dst_entry:: __refcnt" moved
rt_uncached and rt_uncached_list fields from rtable struct to dst
struct, so they are more zeroed by memset_after(xdst, 0, u.dst) in
xfrm_alloc_dst().

Note that rt_uncached (list_head) was never properly initialized at
alloc time, but xfrm[46]_dst_destroy() is written in such a way that
it was not an issue thanks to the memset:

	if (xdst->u.rt.dst.rt_uncached_list)
		rt_del_uncached_list(&xdst->u.rt);

The route code does it the other way around: rt_uncached_list is
assumed to be valid IIF rt_uncached list_head is not empty:

void rt_del_uncached_list(struct rtable *rt)
{
        if (!list_empty(&rt->dst.rt_uncached)) {
                struct uncached_list *ul = rt->dst.rt_uncached_list;

                spin_lock_bh(&ul->lock);
                list_del_init(&rt->dst.rt_uncached);
                spin_unlock_bh(&ul->lock);
        }
}

This patch adds mandatory rt_uncached list_head initialization in
generic dst_init(), and adapt xfrm[46]_dst_destroy logic to match the
rest of the code.

Fixes: d288a162dd1c ("net: dst: Prevent false sharing vs. dst_entry:: __refcnt")
Reported-by: kernel test robot <oliver.sang@intel.com>
Link: https://lore.kernel.org/oe-lkp/202304162125.18b7bcdd-oliver.sang@intel.com
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
CC: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Maxime Bizon <mbizon@freebox.fr>
Link: https://lore.kernel.org/r/20230420182508.2417582-1-mbizon@freebox.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/dst.c          | 1 +
 net/ipv4/route.c        | 4 ----
 net/ipv4/xfrm4_policy.c | 4 +---
 net/ipv6/route.c        | 1 -
 net/ipv6/xfrm6_policy.c | 4 +---
 5 files changed, 3 insertions(+), 11 deletions(-)

diff --git a/net/core/dst.c b/net/core/dst.c
index 3247e84045ca..79d9306ad1ee 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -67,6 +67,7 @@ void dst_init(struct dst_entry *dst, struct dst_ops *ops,
 #endif
 	dst->lwtstate = NULL;
 	rcuref_init(&dst->__rcuref, initial_ref);
+	INIT_LIST_HEAD(&dst->rt_uncached);
 	dst->__use = 0;
 	dst->lastuse = jiffies;
 	dst->flags = flags;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 2a3d14d95ada..98d7e6ba7493 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1644,7 +1644,6 @@ struct rtable *rt_dst_alloc(struct net_device *dev,
 		rt->rt_uses_gateway = 0;
 		rt->rt_gw_family = 0;
 		rt->rt_gw4 = 0;
-		INIT_LIST_HEAD(&rt->dst.rt_uncached);
 
 		rt->dst.output = ip_output;
 		if (flags & RTCF_LOCAL)
@@ -1675,7 +1674,6 @@ struct rtable *rt_dst_clone(struct net_device *dev, struct rtable *rt)
 			new_rt->rt_gw4 = rt->rt_gw4;
 		else if (rt->rt_gw_family == AF_INET6)
 			new_rt->rt_gw6 = rt->rt_gw6;
-		INIT_LIST_HEAD(&new_rt->dst.rt_uncached);
 
 		new_rt->dst.input = rt->dst.input;
 		new_rt->dst.output = rt->dst.output;
@@ -2858,8 +2856,6 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or
 			rt->rt_gw4 = ort->rt_gw4;
 		else if (rt->rt_gw_family == AF_INET6)
 			rt->rt_gw6 = ort->rt_gw6;
-
-		INIT_LIST_HEAD(&rt->dst.rt_uncached);
 	}
 
 	dst_release(dst_orig);
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 47861c8b7340..9403bbaf1b61 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -91,7 +91,6 @@ static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
 		xdst->u.rt.rt_gw6 = rt->rt_gw6;
 	xdst->u.rt.rt_pmtu = rt->rt_pmtu;
 	xdst->u.rt.rt_mtu_locked = rt->rt_mtu_locked;
-	INIT_LIST_HEAD(&xdst->u.rt.dst.rt_uncached);
 	rt_add_uncached_list(&xdst->u.rt);
 
 	return 0;
@@ -121,8 +120,7 @@ static void xfrm4_dst_destroy(struct dst_entry *dst)
 	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
 
 	dst_destroy_metrics_generic(dst);
-	if (xdst->u.rt.dst.rt_uncached_list)
-		rt_del_uncached_list(&xdst->u.rt);
+	rt_del_uncached_list(&xdst->u.rt);
 	xfrm_dst_destroy(xdst);
 }
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 35085fc0cf15..e3aec46bd466 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -334,7 +334,6 @@ static const struct rt6_info ip6_blk_hole_entry_template = {
 static void rt6_info_init(struct rt6_info *rt)
 {
 	memset_after(rt, 0, dst);
-	INIT_LIST_HEAD(&rt->dst.rt_uncached);
 }
 
 /* allocate dst with ip6_dst_ops */
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 2b493f8d0091..eecc5e59da17 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -89,7 +89,6 @@ static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
 	xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway;
 	xdst->u.rt6.rt6i_dst = rt->rt6i_dst;
 	xdst->u.rt6.rt6i_src = rt->rt6i_src;
-	INIT_LIST_HEAD(&xdst->u.rt6.dst.rt_uncached);
 	rt6_uncached_list_add(&xdst->u.rt6);
 
 	return 0;
@@ -121,8 +120,7 @@ static void xfrm6_dst_destroy(struct dst_entry *dst)
 	if (likely(xdst->u.rt6.rt6i_idev))
 		in6_dev_put(xdst->u.rt6.rt6i_idev);
 	dst_destroy_metrics_generic(dst);
-	if (xdst->u.rt6.dst.rt_uncached_list)
-		rt6_uncached_list_del(&xdst->u.rt6);
+	rt6_uncached_list_del(&xdst->u.rt6);
 	xfrm_dst_destroy(xdst);
 }
 
-- 
cgit v1.2.3-70-g09d2


From 99e5acae193e369b71217efe6f1dad42f3f18815 Mon Sep 17 00:00:00 2001
From: Ziyang Xuan <william.xuanziyang@huawei.com>
Date: Thu, 20 Apr 2023 20:40:35 +0800
Subject: ipv4: Fix potential uninit variable access bug in __ip_make_skb()

Like commit ea30388baebc ("ipv6: Fix an uninit variable access bug in
__ip6_make_skb()"). icmphdr does not in skb linear region under the
scenario of SOCK_RAW socket. Access icmp_hdr(skb)->type directly will
trigger the uninit variable access bug.

Use a local variable icmp_type to carry the correct value in different
scenarios.

Fixes: 96793b482540 ("[IPV4]: Add ICMPMsgStats MIB (RFC 4293)")
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Ziyang Xuan <william.xuanziyang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/ip_output.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 4e4e308c3230..89bd7872c629 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1570,9 +1570,19 @@ struct sk_buff *__ip_make_skb(struct sock *sk,
 	cork->dst = NULL;
 	skb_dst_set(skb, &rt->dst);
 
-	if (iph->protocol == IPPROTO_ICMP)
-		icmp_out_count(net, ((struct icmphdr *)
-			skb_transport_header(skb))->type);
+	if (iph->protocol == IPPROTO_ICMP) {
+		u8 icmp_type;
+
+		/* For such sockets, transhdrlen is zero when do ip_append_data(),
+		 * so icmphdr does not in skb linear region and can not get icmp_type
+		 * by icmp_hdr(skb)->type.
+		 */
+		if (sk->sk_type == SOCK_RAW && !inet_sk(sk)->hdrincl)
+			icmp_type = fl4->fl4_icmp_type;
+		else
+			icmp_type = icmp_hdr(skb)->type;
+		icmp_out_count(net, icmp_type);
+	}
 
 	ip_cork_release(cork);
 out:
-- 
cgit v1.2.3-70-g09d2


From cf88231d973909dc8d578138509973e062aba3d7 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Thu, 20 Apr 2023 17:04:23 +0100
Subject: dt-bindings: net: mediatek: add WED RX binding for MT7981 eth driver

Add compatible string for mediatek,mt7981-wed as MT7981 also supports
RX WED just like MT7986, but needs a different firmware file.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/arm/mediatek/mediatek,mt7622-wed.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt7622-wed.yaml b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt7622-wed.yaml
index 5c223cb063d4..f7d578a171a4 100644
--- a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt7622-wed.yaml
+++ b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt7622-wed.yaml
@@ -20,6 +20,7 @@ properties:
     items:
       - enum:
           - mediatek,mt7622-wed
+          - mediatek,mt7981-wed
           - mediatek,mt7986-wed
       - const: syscon
 
-- 
cgit v1.2.3-70-g09d2


From 86ce0d09e42463816f13766ac286012ec7b71fdc Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Thu, 20 Apr 2023 17:05:05 +0100
Subject: net: ethernet: mtk_eth_soc: use WO firmware for MT7981

In order to support wireless offloading on MT7981 we need to load the
appropriate firmware. Recognize MT7981 and load mt7981_wo.bin.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mediatek/mtk_wed_mcu.c | 7 ++++++-
 drivers/net/ethernet/mediatek/mtk_wed_wo.h  | 1 +
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_wed_mcu.c b/drivers/net/ethernet/mediatek/mtk_wed_mcu.c
index 6bad0d262f28..071ed3dea860 100644
--- a/drivers/net/ethernet/mediatek/mtk_wed_mcu.c
+++ b/drivers/net/ethernet/mediatek/mtk_wed_mcu.c
@@ -326,7 +326,11 @@ mtk_wed_mcu_load_firmware(struct mtk_wed_wo *wo)
 		wo->hw->index + 1);
 
 	/* load firmware */
-	fw_name = wo->hw->index ? MT7986_FIRMWARE_WO1 : MT7986_FIRMWARE_WO0;
+	if (of_device_is_compatible(wo->hw->node, "mediatek,mt7981-wed"))
+		fw_name = MT7981_FIRMWARE_WO;
+	else
+		fw_name = wo->hw->index ? MT7986_FIRMWARE_WO1 : MT7986_FIRMWARE_WO0;
+
 	ret = request_firmware(&fw, fw_name, wo->hw->dev);
 	if (ret)
 		return ret;
@@ -386,5 +390,6 @@ int mtk_wed_mcu_init(struct mtk_wed_wo *wo)
 				  100, MTK_FW_DL_TIMEOUT);
 }
 
+MODULE_FIRMWARE(MT7981_FIRMWARE_WO);
 MODULE_FIRMWARE(MT7986_FIRMWARE_WO0);
 MODULE_FIRMWARE(MT7986_FIRMWARE_WO1);
diff --git a/drivers/net/ethernet/mediatek/mtk_wed_wo.h b/drivers/net/ethernet/mediatek/mtk_wed_wo.h
index dbcf42ce9173..7a1a2a28f1ac 100644
--- a/drivers/net/ethernet/mediatek/mtk_wed_wo.h
+++ b/drivers/net/ethernet/mediatek/mtk_wed_wo.h
@@ -88,6 +88,7 @@ enum mtk_wed_dummy_cr_idx {
 	MTK_WED_DUMMY_CR_WO_STATUS,
 };
 
+#define MT7981_FIRMWARE_WO	"mediatek/mt7981_wo.bin"
 #define MT7986_FIRMWARE_WO0	"mediatek/mt7986_wo_0.bin"
 #define MT7986_FIRMWARE_WO1	"mediatek/mt7986_wo_1.bin"
 
-- 
cgit v1.2.3-70-g09d2


From e0416e7d33361d2ad0bf9f007428346579ac854a Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 21 Apr 2023 23:03:46 +0100
Subject: rxrpc: Fix potential race in error handling in afs_make_call()

If the rxrpc call set up by afs_make_call() receives an error whilst it is
transmitting the request, there's the possibility that it may get to the
point the rxrpc call is ended (after the error_kill_call label) just as the
call is queued for async processing.

This could manifest itself as call->rxcall being seen as NULL in
afs_deliver_to_call() when it tries to lock the call.

Fix this by splitting rxrpc_kernel_end_call() into a function to shut down
an rxrpc call and a function to release the caller's reference and calling
the latter only when we get to afs_put_call().

Reported-by: Jeffrey Altman <jaltman@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Tested-by: kafs-testing+fedora36_64checkkafs-build-306@auristor.com
cc: Marc Dionne <marc.dionne@auristor.com>
cc: "David S. Miller" <davem@davemloft.net>
cc: Eric Dumazet <edumazet@google.com>
cc: Jakub Kicinski <kuba@kernel.org>
cc: Paolo Abeni <pabeni@redhat.com>
cc: linux-afs@lists.infradead.org
cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/rxrpc.rst | 17 ++++++++++++-----
 fs/afs/rxrpc.c                     |  9 ++++-----
 include/net/af_rxrpc.h             |  3 ++-
 net/rxrpc/af_rxrpc.c               | 37 +++++++++++++++++++++++++------------
 net/rxrpc/rxperf.c                 |  3 ++-
 5 files changed, 45 insertions(+), 24 deletions(-)

diff --git a/Documentation/networking/rxrpc.rst b/Documentation/networking/rxrpc.rst
index ec1323d92c96..e807e18ba32a 100644
--- a/Documentation/networking/rxrpc.rst
+++ b/Documentation/networking/rxrpc.rst
@@ -848,14 +848,21 @@ The kernel interface functions are as follows:
      returned.  The caller now holds a reference on this and it must be
      properly ended.
 
- (#) End a client call::
+ (#) Shut down a client call::
 
-	void rxrpc_kernel_end_call(struct socket *sock,
+	void rxrpc_kernel_shutdown_call(struct socket *sock,
+					struct rxrpc_call *call);
+
+     This is used to shut down a previously begun call.  The user_call_ID is
+     expunged from AF_RXRPC's knowledge and will not be seen again in
+     association with the specified call.
+
+ (#) Release the ref on a client call::
+
+	void rxrpc_kernel_put_call(struct socket *sock,
 				   struct rxrpc_call *call);
 
-     This is used to end a previously begun call.  The user_call_ID is expunged
-     from AF_RXRPC's knowledge and will not be seen again in association with
-     the specified call.
+     This is used to release the caller's ref on an rxrpc call.
 
  (#) Send data through a call::
 
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
index 7817e2b860e5..e08b850c3e6d 100644
--- a/fs/afs/rxrpc.c
+++ b/fs/afs/rxrpc.c
@@ -179,7 +179,8 @@ void afs_put_call(struct afs_call *call)
 		ASSERT(call->type->name != NULL);
 
 		if (call->rxcall) {
-			rxrpc_kernel_end_call(net->socket, call->rxcall);
+			rxrpc_kernel_shutdown_call(net->socket, call->rxcall);
+			rxrpc_kernel_put_call(net->socket, call->rxcall);
 			call->rxcall = NULL;
 		}
 		if (call->type->destructor)
@@ -420,10 +421,8 @@ error_kill_call:
 	 * The call, however, might be queued on afs_async_calls and we need to
 	 * make sure we don't get any more notifications that might requeue it.
 	 */
-	if (call->rxcall) {
-		rxrpc_kernel_end_call(call->net->socket, call->rxcall);
-		call->rxcall = NULL;
-	}
+	if (call->rxcall)
+		rxrpc_kernel_shutdown_call(call->net->socket, call->rxcall);
 	if (call->async) {
 		if (cancel_work_sync(&call->async_work))
 			afs_put_call(call);
diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h
index ba717eac0229..01a35e113ab9 100644
--- a/include/net/af_rxrpc.h
+++ b/include/net/af_rxrpc.h
@@ -57,7 +57,8 @@ int rxrpc_kernel_recv_data(struct socket *, struct rxrpc_call *,
 			   struct iov_iter *, size_t *, bool, u32 *, u16 *);
 bool rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *,
 			     u32, int, enum rxrpc_abort_reason);
-void rxrpc_kernel_end_call(struct socket *, struct rxrpc_call *);
+void rxrpc_kernel_shutdown_call(struct socket *sock, struct rxrpc_call *call);
+void rxrpc_kernel_put_call(struct socket *sock, struct rxrpc_call *call);
 void rxrpc_kernel_get_peer(struct socket *, struct rxrpc_call *,
 			   struct sockaddr_rxrpc *);
 bool rxrpc_kernel_get_srtt(struct socket *, struct rxrpc_call *, u32 *);
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 102f5cbff91a..c32b164206f9 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -342,31 +342,44 @@ static void rxrpc_dummy_notify_rx(struct sock *sk, struct rxrpc_call *rxcall,
 }
 
 /**
- * rxrpc_kernel_end_call - Allow a kernel service to end a call it was using
+ * rxrpc_kernel_shutdown_call - Allow a kernel service to shut down a call it was using
  * @sock: The socket the call is on
  * @call: The call to end
  *
- * Allow a kernel service to end a call it was using.  The call must be
+ * Allow a kernel service to shut down a call it was using.  The call must be
  * complete before this is called (the call should be aborted if necessary).
  */
-void rxrpc_kernel_end_call(struct socket *sock, struct rxrpc_call *call)
+void rxrpc_kernel_shutdown_call(struct socket *sock, struct rxrpc_call *call)
 {
 	_enter("%d{%d}", call->debug_id, refcount_read(&call->ref));
 
 	mutex_lock(&call->user_mutex);
-	rxrpc_release_call(rxrpc_sk(sock->sk), call);
-
-	/* Make sure we're not going to call back into a kernel service */
-	if (call->notify_rx) {
-		spin_lock(&call->notify_lock);
-		call->notify_rx = rxrpc_dummy_notify_rx;
-		spin_unlock(&call->notify_lock);
+	if (!test_bit(RXRPC_CALL_RELEASED, &call->flags)) {
+		rxrpc_release_call(rxrpc_sk(sock->sk), call);
+
+		/* Make sure we're not going to call back into a kernel service */
+		if (call->notify_rx) {
+			spin_lock(&call->notify_lock);
+			call->notify_rx = rxrpc_dummy_notify_rx;
+			spin_unlock(&call->notify_lock);
+		}
 	}
-
 	mutex_unlock(&call->user_mutex);
+}
+EXPORT_SYMBOL(rxrpc_kernel_shutdown_call);
+
+/**
+ * rxrpc_kernel_put_call - Release a reference to a call
+ * @sock: The socket the call is on
+ * @call: The call to put
+ *
+ * Drop the application's ref on an rxrpc call.
+ */
+void rxrpc_kernel_put_call(struct socket *sock, struct rxrpc_call *call)
+{
 	rxrpc_put_call(call, rxrpc_call_put_kernel);
 }
-EXPORT_SYMBOL(rxrpc_kernel_end_call);
+EXPORT_SYMBOL(rxrpc_kernel_put_call);
 
 /**
  * rxrpc_kernel_check_life - Check to see whether a call is still alive
diff --git a/net/rxrpc/rxperf.c b/net/rxrpc/rxperf.c
index 4a2e90015ca7..085e7892d310 100644
--- a/net/rxrpc/rxperf.c
+++ b/net/rxrpc/rxperf.c
@@ -342,7 +342,8 @@ static void rxperf_deliver_to_call(struct work_struct *work)
 call_complete:
 	rxperf_set_call_complete(call, ret, remote_abort);
 	/* The call may have been requeued */
-	rxrpc_kernel_end_call(rxperf_socket, call->rxcall);
+	rxrpc_kernel_shutdown_call(rxperf_socket, call->rxcall);
+	rxrpc_kernel_put_call(rxperf_socket, call->rxcall);
 	cancel_work(&call->work);
 	kfree(call);
 }
-- 
cgit v1.2.3-70-g09d2


From 6d26d985eeda89faedabbcf6607c37454b9691b0 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sat, 22 Apr 2023 09:35:44 +0200
Subject: bpf: fix link failure with NETFILTER=y INET=n

Explicitly check if NETFILTER_BPF_LINK is enabled, else configs
that have NETFILTER=y but CONFIG_INET=n fail to link:

> kernel/bpf/syscall.o: undefined reference to `netfilter_prog_ops'
> kernel/bpf/verifier.o: undefined reference to `netfilter_verifier_ops'

Fixes: fd9c663b9ad6 ("bpf: minimal support for programs hooked into netfilter framework")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304220903.fRZTJtxe-lkp@intel.com/
Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230422073544.17634-1-fw@strlen.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf_types.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h
index 39a999abb0ce..fc0d6f32c687 100644
--- a/include/linux/bpf_types.h
+++ b/include/linux/bpf_types.h
@@ -79,7 +79,7 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_LSM, lsm,
 #endif
 BPF_PROG_TYPE(BPF_PROG_TYPE_SYSCALL, bpf_syscall,
 	      void *, void *)
-#ifdef CONFIG_NETFILTER
+#ifdef CONFIG_NETFILTER_BPF_LINK
 BPF_PROG_TYPE(BPF_PROG_TYPE_NETFILTER, netfilter,
 	      struct bpf_nf_ctx, struct bpf_nf_ctx)
 #endif
-- 
cgit v1.2.3-70-g09d2


From 35150203e30b52d657165e325e3abc3b29c2086d Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Fri, 21 Apr 2023 23:45:14 +0300
Subject: selftests/bpf: verifier/prevent_map_lookup converted to inline
 assembly

Test verifier/prevent_map_lookup automatically converted to use inline assembly.

This was a part of a series [1] but could not be applied becuase
another patch from a series had to be witheld.

[1] https://lore.kernel.org/bpf/20230421174234.2391278-1-eddyz87@gmail.com/

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230421204514.2450907-1-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |  2 +
 .../bpf/progs/verifier_prevent_map_lookup.c        | 61 ++++++++++++++++++++++
 .../selftests/bpf/verifier/prevent_map_lookup.c    | 29 ----------
 3 files changed, 63 insertions(+), 29 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_prevent_map_lookup.c
 delete mode 100644 tools/testing/selftests/bpf/verifier/prevent_map_lookup.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index c8bab8b1a6a4..2497716ee379 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -42,6 +42,7 @@
 #include "verifier_meta_access.skel.h"
 #include "verifier_netfilter_ctx.skel.h"
 #include "verifier_netfilter_retcode.skel.h"
+#include "verifier_prevent_map_lookup.skel.h"
 #include "verifier_raw_stack.skel.h"
 #include "verifier_raw_tp_writable.skel.h"
 #include "verifier_reg_equal.skel.h"
@@ -140,6 +141,7 @@ void test_verifier_masking(void)              { RUN(verifier_masking); }
 void test_verifier_meta_access(void)          { RUN(verifier_meta_access); }
 void test_verifier_netfilter_ctx(void)        { RUN(verifier_netfilter_ctx); }
 void test_verifier_netfilter_retcode(void)    { RUN(verifier_netfilter_retcode); }
+void test_verifier_prevent_map_lookup(void)   { RUN(verifier_prevent_map_lookup); }
 void test_verifier_raw_stack(void)            { RUN(verifier_raw_stack); }
 void test_verifier_raw_tp_writable(void)      { RUN(verifier_raw_tp_writable); }
 void test_verifier_reg_equal(void)            { RUN(verifier_reg_equal); }
diff --git a/tools/testing/selftests/bpf/progs/verifier_prevent_map_lookup.c b/tools/testing/selftests/bpf/progs/verifier_prevent_map_lookup.c
new file mode 100644
index 000000000000..8d27c780996f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_prevent_map_lookup.c
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Converted from tools/testing/selftests/bpf/verifier/prevent_map_lookup.c */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+
+struct {
+	__uint(type, BPF_MAP_TYPE_STACK_TRACE);
+	__uint(max_entries, 1);
+	__type(key, __u32);
+	__type(value, __u64);
+} map_stacktrace SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_PROG_ARRAY);
+	__uint(max_entries, 8);
+	__uint(key_size, sizeof(int));
+	__array(values, void (void));
+} map_prog2_socket SEC(".maps");
+
+SEC("perf_event")
+__description("prevent map lookup in stack trace")
+__failure __msg("cannot pass map_type 7 into func bpf_map_lookup_elem")
+__naked void map_lookup_in_stack_trace(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_stacktrace] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_stacktrace)
+	: __clobber_all);
+}
+
+SEC("socket")
+__description("prevent map lookup in prog array")
+__failure __msg("cannot pass map_type 3 into func bpf_map_lookup_elem")
+__failure_unpriv
+__naked void map_lookup_in_prog_array(void)
+{
+	asm volatile ("					\
+	r1 = 0;						\
+	*(u64*)(r10 - 8) = r1;				\
+	r2 = r10;					\
+	r2 += -8;					\
+	r1 = %[map_prog2_socket] ll;			\
+	call %[bpf_map_lookup_elem];			\
+	exit;						\
+"	:
+	: __imm(bpf_map_lookup_elem),
+	  __imm_addr(map_prog2_socket)
+	: __clobber_all);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/verifier/prevent_map_lookup.c b/tools/testing/selftests/bpf/verifier/prevent_map_lookup.c
deleted file mode 100644
index fc4e301260f6..000000000000
--- a/tools/testing/selftests/bpf/verifier/prevent_map_lookup.c
+++ /dev/null
@@ -1,29 +0,0 @@
-{
-	"prevent map lookup in stack trace",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_map_stacktrace = { 3 },
-	.result = REJECT,
-	.errstr = "cannot pass map_type 7 into func bpf_map_lookup_elem",
-	.prog_type = BPF_PROG_TYPE_PERF_EVENT,
-},
-{
-	"prevent map lookup in prog array",
-	.insns = {
-	BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
-	BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-	BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
-	BPF_LD_MAP_FD(BPF_REG_1, 0),
-	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-	BPF_EXIT_INSN(),
-	},
-	.fixup_prog2 = { 3 },
-	.result = REJECT,
-	.errstr = "cannot pass map_type 3 into func bpf_map_lookup_elem",
-},
-- 
cgit v1.2.3-70-g09d2


From 4eaeca1fc43a4060fa79d7a071572e89181a725e Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Thu, 20 Apr 2023 22:06:42 +0100
Subject: net: mtk_eth_soc: mediatek: fix ppe flow accounting for v1 hardware

Older chips (like MT7622) use a different bit in ib2 to enable hardware
counter support. Add macros for both and select the appropriate bit.

Fixes: 3fbe4d8c0e53 ("net: ethernet: mtk_eth_soc: ppe: add support for flow accounting")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mediatek/mtk_ppe.c | 10 ++++++++--
 drivers/net/ethernet/mediatek/mtk_ppe.h |  3 ++-
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index dd9581334b05..9129821f3ab8 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -601,6 +601,7 @@ __mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 	struct mtk_eth *eth = ppe->eth;
 	u16 timestamp = mtk_eth_timestamp(eth);
 	struct mtk_foe_entry *hwe;
+	u32 val;
 
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_NETSYS_V2)) {
 		entry->ib1 &= ~MTK_FOE_IB1_BIND_TIMESTAMP_V2;
@@ -617,8 +618,13 @@ __mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 	wmb();
 	hwe->ib1 = entry->ib1;
 
-	if (ppe->accounting)
-		*mtk_foe_entry_ib2(eth, hwe) |= MTK_FOE_IB2_MIB_CNT;
+	if (ppe->accounting) {
+		if (MTK_HAS_CAPS(eth->soc->caps, MTK_NETSYS_V2))
+			val = MTK_FOE_IB2_MIB_CNT_V2;
+		else
+			val = MTK_FOE_IB2_MIB_CNT;
+		*mtk_foe_entry_ib2(eth, hwe) |= val;
+	}
 
 	dma_wmb();
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index e1aab2e8e262..e51de31a52ec 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -55,9 +55,10 @@ enum {
 #define MTK_FOE_IB2_PSE_QOS		BIT(4)
 #define MTK_FOE_IB2_DEST_PORT		GENMASK(7, 5)
 #define MTK_FOE_IB2_MULTICAST		BIT(8)
+#define MTK_FOE_IB2_MIB_CNT		BIT(10)
 
 #define MTK_FOE_IB2_WDMA_QID2		GENMASK(13, 12)
-#define MTK_FOE_IB2_MIB_CNT		BIT(15)
+#define MTK_FOE_IB2_MIB_CNT_V2		BIT(15)
 #define MTK_FOE_IB2_WDMA_DEVIDX		BIT(16)
 #define MTK_FOE_IB2_WDMA_WINFO		BIT(17)
 
-- 
cgit v1.2.3-70-g09d2


From e8e1ce8454c9cc8ad2e4422bef346428e52455e3 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 21 Apr 2023 09:43:53 +0000
Subject: net: add debugging checks in skb_attempt_defer_free()

Make sure skbs that are stored in softnet_data.defer_list
do not have a dst attached.

Also make sure the the skb was orphaned.

Link: https://lore.kernel.org/netdev/CANn89iJuEVe72bPmEftyEJHLzzN=QNR2yueFjTxYXCEpS5S8HQ@mail.gmail.com/T/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 0d998806b377..bd815a00d2af 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -6881,6 +6881,9 @@ nodefer:	__kfree_skb(skb);
 		return;
 	}
 
+	DEBUG_NET_WARN_ON_ONCE(skb_dst(skb));
+	DEBUG_NET_WARN_ON_ONCE(skb->destructor);
+
 	sd = &per_cpu(softnet_data, cpu);
 	defer_max = READ_ONCE(sysctl_skb_defer_max);
 	if (READ_ONCE(sd->defer_count) >= defer_max)
-- 
cgit v1.2.3-70-g09d2


From 931e93bdf8ca71cef1f8759c43bc2c5385392b8b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 21 Apr 2023 09:43:54 +0000
Subject: net: do not provide hard irq safety for sd->defer_lock

kfree_skb() can be called from hard irq handlers,
but skb_attempt_defer_free() is meant to be used
from process or BH contexts, and skb_defer_free_flush()
is meant to be called from BH contexts.

Not having to mask hard irq can save some cycles.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c    | 4 ++--
 net/core/skbuff.c | 5 ++---
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 1551aabac343..d15568f5a44f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6632,11 +6632,11 @@ static void skb_defer_free_flush(struct softnet_data *sd)
 	if (!READ_ONCE(sd->defer_list))
 		return;
 
-	spin_lock_irq(&sd->defer_lock);
+	spin_lock(&sd->defer_lock);
 	skb = sd->defer_list;
 	sd->defer_list = NULL;
 	sd->defer_count = 0;
-	spin_unlock_irq(&sd->defer_lock);
+	spin_unlock(&sd->defer_lock);
 
 	while (skb != NULL) {
 		next = skb->next;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index bd815a00d2af..304a966164d8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -6870,7 +6870,6 @@ void skb_attempt_defer_free(struct sk_buff *skb)
 {
 	int cpu = skb->alloc_cpu;
 	struct softnet_data *sd;
-	unsigned long flags;
 	unsigned int defer_max;
 	bool kick;
 
@@ -6889,7 +6888,7 @@ nodefer:	__kfree_skb(skb);
 	if (READ_ONCE(sd->defer_count) >= defer_max)
 		goto nodefer;
 
-	spin_lock_irqsave(&sd->defer_lock, flags);
+	spin_lock_bh(&sd->defer_lock);
 	/* Send an IPI every time queue reaches half capacity. */
 	kick = sd->defer_count == (defer_max >> 1);
 	/* Paired with the READ_ONCE() few lines above */
@@ -6898,7 +6897,7 @@ nodefer:	__kfree_skb(skb);
 	skb->next = sd->defer_list;
 	/* Paired with READ_ONCE() in skb_defer_free_flush() */
 	WRITE_ONCE(sd->defer_list, skb);
-	spin_unlock_irqrestore(&sd->defer_lock, flags);
+	spin_unlock_bh(&sd->defer_lock);
 
 	/* Make sure to trigger NET_RX_SOFTIRQ on the remote CPU
 	 * if we are unlucky enough (this seems very unlikely).
-- 
cgit v1.2.3-70-g09d2


From e6f50edfef046cf8ad541b4d5972bf38fdcdec39 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 21 Apr 2023 09:43:55 +0000
Subject: net: move skb_defer_free_flush() up

We plan using skb_defer_free_flush() from napi_threaded_poll()
in the following patch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c | 42 +++++++++++++++++++++---------------------
 1 file changed, 21 insertions(+), 21 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index d15568f5a44f..81ded215731b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6598,6 +6598,27 @@ static int napi_thread_wait(struct napi_struct *napi)
 	return -1;
 }
 
+static void skb_defer_free_flush(struct softnet_data *sd)
+{
+	struct sk_buff *skb, *next;
+
+	/* Paired with WRITE_ONCE() in skb_attempt_defer_free() */
+	if (!READ_ONCE(sd->defer_list))
+		return;
+
+	spin_lock(&sd->defer_lock);
+	skb = sd->defer_list;
+	sd->defer_list = NULL;
+	sd->defer_count = 0;
+	spin_unlock(&sd->defer_lock);
+
+	while (skb != NULL) {
+		next = skb->next;
+		napi_consume_skb(skb, 1);
+		skb = next;
+	}
+}
+
 static int napi_threaded_poll(void *data)
 {
 	struct napi_struct *napi = data;
@@ -6624,27 +6645,6 @@ static int napi_threaded_poll(void *data)
 	return 0;
 }
 
-static void skb_defer_free_flush(struct softnet_data *sd)
-{
-	struct sk_buff *skb, *next;
-
-	/* Paired with WRITE_ONCE() in skb_attempt_defer_free() */
-	if (!READ_ONCE(sd->defer_list))
-		return;
-
-	spin_lock(&sd->defer_lock);
-	skb = sd->defer_list;
-	sd->defer_list = NULL;
-	sd->defer_count = 0;
-	spin_unlock(&sd->defer_lock);
-
-	while (skb != NULL) {
-		next = skb->next;
-		napi_consume_skb(skb, 1);
-		skb = next;
-	}
-}
-
 static __latent_entropy void net_rx_action(struct softirq_action *h)
 {
 	struct softnet_data *sd = this_cpu_ptr(&softnet_data);
-- 
cgit v1.2.3-70-g09d2


From a1aaee7f8f79d1b0595e24f8c3caed24630d6cb6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 21 Apr 2023 09:43:56 +0000
Subject: net: make napi_threaded_poll() aware of sd->defer_list

If we call skb_defer_free_flush() from napi_threaded_poll(),
we can avoid to raise IPI from skb_attempt_defer_free()
when the list becomes too big.

This allows napi_threaded_poll() to rely less on softirqs,
and lowers latency caused by a too big list.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/core/dev.c b/net/core/dev.c
index 81ded215731b..7d9ec23f97c6 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6622,6 +6622,7 @@ static void skb_defer_free_flush(struct softnet_data *sd)
 static int napi_threaded_poll(void *data)
 {
 	struct napi_struct *napi = data;
+	struct softnet_data *sd;
 	void *have;
 
 	while (!napi_thread_wait(napi)) {
@@ -6629,11 +6630,13 @@ static int napi_threaded_poll(void *data)
 			bool repoll = false;
 
 			local_bh_disable();
+			sd = this_cpu_ptr(&softnet_data);
 
 			have = netpoll_poll_lock(napi);
 			__napi_poll(napi, &repoll);
 			netpoll_poll_unlock(have);
 
+			skb_defer_free_flush(sd);
 			local_bh_enable();
 
 			if (!repoll)
-- 
cgit v1.2.3-70-g09d2


From 87eff2ec57b6d68d294013d8dd21e839a1175e3a Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 21 Apr 2023 09:43:57 +0000
Subject: net: optimize napi_threaded_poll() vs RPS/RFS

We use napi_threaded_poll() in order to reduce our softirq dependency.

We can add a followup of 821eba962d95 ("net: optimize napi_schedule_rps()")
to further remove the need of firing NET_RX_SOFTIRQ whenever
RPS/RFS are used.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  3 +++
 net/core/dev.c            | 12 ++++++++++--
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a6a3e9457d6c..08fbd4622ccf 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3194,7 +3194,10 @@ struct softnet_data {
 #ifdef CONFIG_RPS
 	struct softnet_data	*rps_ipi_list;
 #endif
+
 	bool			in_net_rx_action;
+	bool			in_napi_threaded_poll;
+
 #ifdef CONFIG_NET_FLOW_LIMIT
 	struct sd_flow_limit __rcu *flow_limit;
 #endif
diff --git a/net/core/dev.c b/net/core/dev.c
index 7d9ec23f97c6..735096d42c1d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4603,10 +4603,10 @@ static void napi_schedule_rps(struct softnet_data *sd)
 		sd->rps_ipi_next = mysd->rps_ipi_list;
 		mysd->rps_ipi_list = sd;
 
-		/* If not called from net_rx_action()
+		/* If not called from net_rx_action() or napi_threaded_poll()
 		 * we have to raise NET_RX_SOFTIRQ.
 		 */
-		if (!mysd->in_net_rx_action)
+		if (!mysd->in_net_rx_action && !mysd->in_napi_threaded_poll)
 			__raise_softirq_irqoff(NET_RX_SOFTIRQ);
 		return;
 	}
@@ -6631,11 +6631,19 @@ static int napi_threaded_poll(void *data)
 
 			local_bh_disable();
 			sd = this_cpu_ptr(&softnet_data);
+			sd->in_napi_threaded_poll = true;
 
 			have = netpoll_poll_lock(napi);
 			__napi_poll(napi, &repoll);
 			netpoll_poll_unlock(have);
 
+			sd->in_napi_threaded_poll = false;
+			barrier();
+
+			if (sd_has_rps_ipi_waiting(sd)) {
+				local_irq_disable();
+				net_rps_action_and_irq_enable(sd);
+			}
 			skb_defer_free_flush(sd);
 			local_bh_enable();
 
-- 
cgit v1.2.3-70-g09d2


From 788352191c85326beb74f518ba761ed38b82f533 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Fri, 21 Apr 2023 14:08:53 +0100
Subject: rxrpc: Replace fake flex-array with flexible-array member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length arrays as fake flexible arrays are deprecated and we are
moving towards adopting C99 flexible-array members instead.

Transform zero-length array into flexible-array member in struct
rxrpc_ackpacket.

Address the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
net/rxrpc/call_event.c:149:38: warning: array subscript i is outside array bounds of ‘uint8_t[0]’ {aka ‘unsigned char[]’} [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/263
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Jeffrey Altman <jaltman@auristor.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: "David S. Miller" <davem@davemloft.net>
cc: Eric Dumazet <edumazet@google.com>
cc: Jakub Kicinski <kuba@kernel.org>
cc: Paolo Abeni <pabeni@redhat.com>
cc: linux-afs@lists.infradead.org
cc: netdev@vger.kernel.org
cc: linux-hardening@vger.kernel.org
Link: https://lore.kernel.org/r/ZAZT11n4q5bBttW0@work/
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/rxrpc/protocol.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/rxrpc/protocol.h b/net/rxrpc/protocol.h
index 6760cb99c6d6..e8ee4af43ca8 100644
--- a/net/rxrpc/protocol.h
+++ b/net/rxrpc/protocol.h
@@ -126,7 +126,7 @@ struct rxrpc_ackpacket {
 	uint8_t		nAcks;		/* number of ACKs */
 #define RXRPC_MAXACKS	255
 
-	uint8_t		acks[0];	/* list of ACK/NAKs */
+	uint8_t		acks[];		/* list of ACK/NAKs */
 #define RXRPC_ACK_TYPE_NACK		0
 #define RXRPC_ACK_TYPE_ACK		1
 
-- 
cgit v1.2.3-70-g09d2


From b148b9abc8444c9e0579a6bbe47b2da6adf5a150 Mon Sep 17 00:00:00 2001
From: Wang Zhang <silver_code@hust.edu.cn>
Date: Fri, 21 Apr 2023 23:10:09 +0800
Subject: net: ethernet: mediatek: remove return value check of
 `mtk_wed_hw_add_debugfs`

Smatch complains that:
mtk_wed_hw_add_debugfs() warn: 'dir' is an error pointer or valid

Debugfs checks are generally not supposed to be checked
for errors and it is not necessary here.

fix it by just deleting the dead code.

Signed-off-by: Wang Zhang <silver_code@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mediatek/mtk_wed_debugfs.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c b/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
index 56f663439721..b244c02c5b51 100644
--- a/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
+++ b/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
@@ -252,8 +252,6 @@ void mtk_wed_hw_add_debugfs(struct mtk_wed_hw *hw)
 
 	snprintf(hw->dirname, sizeof(hw->dirname), "wed%d", hw->index);
 	dir = debugfs_create_dir(hw->dirname, NULL);
-	if (!dir)
-		return;
 
 	hw->debugfs_dir = dir;
 	debugfs_create_u32("regidx", 0600, dir, &hw->debugfs_reg);
-- 
cgit v1.2.3-70-g09d2


From fadfc57cc8047080a123b16f288b7138524fb1e2 Mon Sep 17 00:00:00 2001
From: Marc Dionne <marc.dionne@auristor.com>
Date: Fri, 21 Apr 2023 17:16:17 +0100
Subject: rxrpc: Fix error when reading rxrpc tokens

When converting from ASSERTCMP to WARN_ON, the tested condition must
be inverted, which was missed for this case.

This would cause an EIO error when trying to read an rxrpc token, for
instance when trying to display tokens with AuriStor's "tokens" command.

Fixes: 84924aac08a4 ("rxrpc: Fix checker warning")
Signed-off-by: Marc Dionne <marc.dionne@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: "David S. Miller" <davem@davemloft.net>
cc: Eric Dumazet <edumazet@google.com>
cc: Jakub Kicinski <kuba@kernel.org>
cc: Paolo Abeni <pabeni@redhat.com>
cc: linux-afs@lists.infradead.org
cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/rxrpc/key.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/rxrpc/key.c b/net/rxrpc/key.c
index 8d53aded09c4..33e8302a79e3 100644
--- a/net/rxrpc/key.c
+++ b/net/rxrpc/key.c
@@ -680,7 +680,7 @@ static long rxrpc_read(const struct key *key,
 			return -ENOPKG;
 		}
 
-		if (WARN_ON((unsigned long)xdr - (unsigned long)oldxdr ==
+		if (WARN_ON((unsigned long)xdr - (unsigned long)oldxdr !=
 			    toksize))
 			return -EIO;
 	}
-- 
cgit v1.2.3-70-g09d2


From 4d2bd2581c3bc73e9fdf5f82b25c359212567457 Mon Sep 17 00:00:00 2001
From: Ramón Nordin Rodriguez <ramon.nordin.rodriguez@ferroamp.se>
Date: Fri, 21 Apr 2023 18:38:54 +0200
Subject: drivers/net/phy: add driver for Microchip LAN867x 10BASE-T1S PHY
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds support for the Microchip LAN867x 10BASE-T1S family
(LAN8670/1/2). The driver supports P2MP with PLCA.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Ramón Nordin Rodriguez <ramon.nordin.rodriguez@ferroamp.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/Kconfig         |   5 ++
 drivers/net/phy/Makefile        |   1 +
 drivers/net/phy/microchip_t1s.c | 138 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 144 insertions(+)
 create mode 100644 drivers/net/phy/microchip_t1s.c

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index bcfba077b80e..2f3ddc446cbb 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -244,6 +244,11 @@ config MICREL_PHY
 	help
 	  Supports the KSZ9021, VSC8201, KS8001 PHYs.
 
+config MICROCHIP_T1S_PHY
+	tristate "Microchip 10BASE-T1S Ethernet PHY"
+	help
+	  Currently supports the LAN8670, LAN8671, LAN8672
+
 config MICROCHIP_PHY
 	tristate "Microchip PHYs"
 	help
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index ae11bf20b46e..f289ab16a1da 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -74,6 +74,7 @@ obj-$(CONFIG_MICREL_KS8995MA)	+= spi_ks8995.o
 obj-$(CONFIG_MICREL_PHY)	+= micrel.o
 obj-$(CONFIG_MICROCHIP_PHY)	+= microchip.o
 obj-$(CONFIG_MICROCHIP_T1_PHY)	+= microchip_t1.o
+obj-$(CONFIG_MICROCHIP_T1S_PHY) += microchip_t1s.o
 obj-$(CONFIG_MICROSEMI_PHY)	+= mscc/
 obj-$(CONFIG_MOTORCOMM_PHY)	+= motorcomm.o
 obj-$(CONFIG_NATIONAL_PHY)	+= national.o
diff --git a/drivers/net/phy/microchip_t1s.c b/drivers/net/phy/microchip_t1s.c
new file mode 100644
index 000000000000..094967b3c111
--- /dev/null
+++ b/drivers/net/phy/microchip_t1s.c
@@ -0,0 +1,138 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Driver for Microchip 10BASE-T1S LAN867X PHY
+ *
+ * Support: Microchip Phys:
+ *  lan8670, lan8671, lan8672
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/phy.h>
+
+#define PHY_ID_LAN867X 0x0007C160
+
+#define LAN867X_REG_IRQ_1_CTL 0x001C
+#define LAN867X_REG_IRQ_2_CTL 0x001D
+
+/* The arrays below are pulled from the following table from AN1699
+ * Access MMD Address Value Mask
+ * RMW 0x1F 0x00D0 0x0002 0x0E03
+ * RMW 0x1F 0x00D1 0x0000 0x0300
+ * RMW 0x1F 0x0084 0x3380 0xFFC0
+ * RMW 0x1F 0x0085 0x0006 0x000F
+ * RMW 0x1F 0x008A 0xC000 0xF800
+ * RMW 0x1F 0x0087 0x801C 0x801C
+ * RMW 0x1F 0x0088 0x033F 0x1FFF
+ * W   0x1F 0x008B 0x0404 ------
+ * RMW 0x1F 0x0080 0x0600 0x0600
+ * RMW 0x1F 0x00F1 0x2400 0x7F00
+ * RMW 0x1F 0x0096 0x2000 0x2000
+ * W   0x1F 0x0099 0x7F80 ------
+ */
+
+static const int lan867x_fixup_registers[12] = {
+	0x00D0, 0x00D1, 0x0084, 0x0085,
+	0x008A, 0x0087, 0x0088, 0x008B,
+	0x0080, 0x00F1, 0x0096, 0x0099,
+};
+
+static const int lan867x_fixup_values[12] = {
+	0x0002, 0x0000, 0x3380, 0x0006,
+	0xC000, 0x801C, 0x033F, 0x0404,
+	0x0600, 0x2400, 0x2000, 0x7F80,
+};
+
+static const int lan867x_fixup_masks[12] = {
+	0x0E03, 0x0300, 0xFFC0, 0x000F,
+	0xF800, 0x801C, 0x1FFF, 0xFFFF,
+	0x0600, 0x7F00, 0x2000, 0xFFFF,
+};
+
+static int lan867x_config_init(struct phy_device *phydev)
+{
+	/* HW quirk: Microchip states in the application note (AN1699) for the phy
+	 * that a set of read-modify-write (rmw) operations has to be performed
+	 * on a set of seemingly magic registers.
+	 * The result of these operations is just described as 'optimal performance'
+	 * Microchip gives no explanation as to what these mmd regs do,
+	 * in fact they are marked as reserved in the datasheet.
+	 * It is unclear if phy_modify_mmd would be safe to use or if a write
+	 * really has to happen to each register.
+	 * In order to exactly conform to what is stated in the AN phy_write_mmd is
+	 * used, which might then write the same value back as read + modified.
+	 */
+
+	int reg_value;
+	int err;
+	int reg;
+
+	/* Read-Modified Write Pseudocode (from AN1699)
+	 * current_val = read_register(mmd, addr) // Read current register value
+	 * new_val = current_val AND (NOT mask) // Clear bit fields to be written
+	 * new_val = new_val OR value // Set bits
+	 * write_register(mmd, addr, new_val) // Write back updated register value
+	 */
+	for (int i = 0; i < ARRAY_SIZE(lan867x_fixup_registers); i++) {
+		reg = lan867x_fixup_registers[i];
+		reg_value = phy_read_mmd(phydev, MDIO_MMD_VEND2, reg);
+		reg_value &= ~lan867x_fixup_masks[i];
+		reg_value |= lan867x_fixup_values[i];
+		err = phy_write_mmd(phydev, MDIO_MMD_VEND2, reg, reg_value);
+		if (err != 0)
+			return err;
+	}
+
+	/* None of the interrupts in the lan867x phy seem relevant.
+	 * Other phys inspect the link status and call phy_trigger_machine
+	 * in the interrupt handler.
+	 * This phy does not support link status, and thus has no interrupt
+	 * for it either.
+	 * So we'll just disable all interrupts on the chip.
+	 */
+	err = phy_write_mmd(phydev, MDIO_MMD_VEND2, LAN867X_REG_IRQ_1_CTL, 0xFFFF);
+	if (err != 0)
+		return err;
+	return phy_write_mmd(phydev, MDIO_MMD_VEND2, LAN867X_REG_IRQ_2_CTL, 0xFFFF);
+}
+
+static int lan867x_read_status(struct phy_device *phydev)
+{
+	/* The phy has some limitations, namely:
+	 *  - always reports link up
+	 *  - only supports 10MBit half duplex
+	 *  - does not support auto negotiate
+	 */
+	phydev->link = 1;
+	phydev->duplex = DUPLEX_HALF;
+	phydev->speed = SPEED_10;
+	phydev->autoneg = AUTONEG_DISABLE;
+
+	return 0;
+}
+
+static struct phy_driver lan867x_driver[] = {
+	{
+		PHY_ID_MATCH_MODEL(PHY_ID_LAN867X),
+		.name               = "LAN867X",
+		.features           = PHY_BASIC_T1S_P2MP_FEATURES,
+		.config_init        = lan867x_config_init,
+		.read_status        = lan867x_read_status,
+		.get_plca_cfg	    = genphy_c45_plca_get_cfg,
+		.set_plca_cfg	    = genphy_c45_plca_set_cfg,
+		.get_plca_status    = genphy_c45_plca_get_status,
+	}
+};
+
+module_phy_driver(lan867x_driver);
+
+static struct mdio_device_id __maybe_unused tbl[] = {
+	{ PHY_ID_MATCH_MODEL(PHY_ID_LAN867X) },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(mdio, tbl);
+
+MODULE_DESCRIPTION("Microchip 10BASE-T1S lan867x Phy driver");
+MODULE_AUTHOR("Ramón Nordin Rodriguez");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-70-g09d2


From f90615ada0b1e21a9d93ff89b04549fd7a92c92b Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:53 +0300
Subject: net: vlan: don't adjust MAC header in __vlan_insert_inner_tag()
 unless set

This is a preparatory change for the deletion of skb_reset_mac_header(skb)
from __dev_queue_xmit(). After that deletion, skb_mac_header(skb) will
no longer be set in TX paths, from which __vlan_insert_inner_tag() can
still be called (perhaps indirectly).

If we don't make this change, then an unset MAC header (equal to ~0U)
will become set after the adjustment with VLAN_HLEN.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 6864b89ef868..90b76d63c11c 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -351,7 +351,8 @@ static inline int __vlan_insert_inner_tag(struct sk_buff *skb,
 	/* Move the mac header sans proto to the beginning of the new header. */
 	if (likely(mac_len > ETH_TLEN))
 		memmove(skb->data, skb->data + VLAN_HLEN, mac_len - ETH_TLEN);
-	skb->mac_header -= VLAN_HLEN;
+	if (skb_mac_header_was_set(skb))
+		skb->mac_header -= VLAN_HLEN;
 
 	veth = (struct vlan_ethhdr *)(skb->data + mac_len - ETH_HLEN);
 
-- 
cgit v1.2.3-70-g09d2


From 1f5020acb33f926030f62563c86dffca35c7b701 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:54 +0300
Subject: net: vlan: introduce skb_vlan_eth_hdr()

Similar to skb_eth_hdr() introduced in commit 96cc4b69581d ("macvlan: do
not assume mac_header is set in macvlan_broadcast()"), let's introduce a
skb_vlan_eth_hdr() helper which can be used in TX-only code paths to get
to the VLAN header based on skb->data rather than based on the
skb_mac_header(skb).

We also consolidate the drivers that dereference skb->data to go through
this helper.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c      |  3 +--
 drivers/net/ethernet/emulex/benet/be_main.c          |  2 +-
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c      |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c          |  2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c        |  2 +-
 drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c |  2 +-
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c       |  4 ++--
 drivers/net/ethernet/sfc/tx_tso.c                    |  2 +-
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c    |  7 ++-----
 drivers/staging/gdm724x/gdm_lte.c                    |  4 ++--
 include/linux/if_vlan.h                              | 12 ++++++++++--
 net/batman-adv/soft-interface.c                      |  2 +-
 12 files changed, 24 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 12083b9679b5..6ea5521074d3 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -1935,8 +1935,7 @@ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb,
 
 		/* Skip VLAN tag if present */
 		if (ether_type == ETH_P_8021Q) {
-			struct vlan_ethhdr *vhdr =
-				(struct vlan_ethhdr *)skb->data;
+			struct vlan_ethhdr *vhdr = skb_vlan_eth_hdr(skb);
 
 			ether_type = ntohs(vhdr->h_vlan_encapsulated_proto);
 		}
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index aed1b622f51f..7e408bcc88de 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1124,7 +1124,7 @@ static struct sk_buff *be_lancer_xmit_workarounds(struct be_adapter *adapter,
 						  struct be_wrb_params
 						  *wrb_params)
 {
-	struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data;
+	struct vlan_ethhdr *veh = skb_vlan_eth_hdr(skb);
 	unsigned int eth_hdr_len;
 	struct iphdr *ip;
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 5caea154362f..7356ad965487 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -1532,7 +1532,7 @@ static int hns3_handle_vtags(struct hns3_enet_ring *tx_ring,
 	if (unlikely(rc < 0))
 		return rc;
 
-	vhdr = (struct vlan_ethhdr *)skb->data;
+	vhdr = skb_vlan_eth_hdr(skb);
 	vhdr->h_vlan_TCI |= cpu_to_be16((skb->priority << VLAN_PRIO_SHIFT)
 					 & VLAN_PRIO_MASK);
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index c8c2cbaa0ede..8b8bf4880faa 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -3063,7 +3063,7 @@ static inline int i40e_tx_prepare_vlan_flags(struct sk_buff *skb,
 			rc = skb_cow_head(skb, 0);
 			if (rc < 0)
 				return rc;
-			vhdr = (struct vlan_ethhdr *)skb->data;
+			vhdr = skb_vlan_eth_hdr(skb);
 			vhdr->h_vlan_TCI = htons(tx_flags >>
 						 I40E_TX_FLAGS_VLAN_SHIFT);
 		} else {
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index f2604fc05991..e961ef4bbf4d 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8798,7 +8798,7 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb,
 
 			if (skb_cow_head(skb, 0))
 				goto out_drop;
-			vhdr = (struct vlan_ethhdr *)skb->data;
+			vhdr = skb_vlan_eth_hdr(skb);
 			vhdr->h_vlan_TCI = htons(tx_flags >>
 						 IXGBE_TX_FLAGS_VLAN_SHIFT);
 		} else {
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index 59d0dd862fd1..1d1e183d3a8b 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -1854,7 +1854,7 @@ netxen_tso_check(struct net_device *netdev,
 
 	if (protocol == cpu_to_be16(ETH_P_8021Q)) {
 
-		vh = (struct vlan_ethhdr *)skb->data;
+		vh = skb_vlan_eth_hdr(skb);
 		protocol = vh->h_vlan_encapsulated_proto;
 		flags = FLAGS_VLAN_TAGGED;
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
index 92930a055cbc..41894d154013 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
@@ -318,7 +318,7 @@ static void qlcnic_send_filter(struct qlcnic_adapter *adapter,
 
 	if (adapter->flags & QLCNIC_VLAN_FILTERING) {
 		if (protocol == ETH_P_8021Q) {
-			vh = (struct vlan_ethhdr *)skb->data;
+			vh = skb_vlan_eth_hdr(skb);
 			vlan_id = ntohs(vh->h_vlan_TCI);
 		} else if (skb_vlan_tag_present(skb)) {
 			vlan_id = skb_vlan_tag_get(skb);
@@ -468,7 +468,7 @@ static int qlcnic_tx_pkt(struct qlcnic_adapter *adapter,
 	u32 producer = tx_ring->producer;
 
 	if (protocol == ETH_P_8021Q) {
-		vh = (struct vlan_ethhdr *)skb->data;
+		vh = skb_vlan_eth_hdr(skb);
 		flags = QLCNIC_FLAGS_VLAN_TAGGED;
 		vlan_tci = ntohs(vh->h_vlan_TCI);
 		protocol = ntohs(vh->h_vlan_encapsulated_proto);
diff --git a/drivers/net/ethernet/sfc/tx_tso.c b/drivers/net/ethernet/sfc/tx_tso.c
index 898e5c61d908..d381d8164f07 100644
--- a/drivers/net/ethernet/sfc/tx_tso.c
+++ b/drivers/net/ethernet/sfc/tx_tso.c
@@ -147,7 +147,7 @@ static __be16 efx_tso_check_protocol(struct sk_buff *skb)
 	EFX_WARN_ON_ONCE_PARANOID(((struct ethhdr *)skb->data)->h_proto !=
 				  protocol);
 	if (protocol == htons(ETH_P_8021Q)) {
-		struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data;
+		struct vlan_ethhdr *veh = skb_vlan_eth_hdr(skb);
 
 		protocol = veh->h_vlan_encapsulated_proto;
 	}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index f116e4ae293b..0fca81507a77 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4569,13 +4569,10 @@ dma_map_err:
 
 static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb)
 {
-	struct vlan_ethhdr *veth;
-	__be16 vlan_proto;
+	struct vlan_ethhdr *veth = skb_vlan_eth_hdr(skb);
+	__be16 vlan_proto = veth->h_vlan_proto;
 	u16 vlanid;
 
-	veth = (struct vlan_ethhdr *)skb->data;
-	vlan_proto = veth->h_vlan_proto;
-
 	if ((vlan_proto == htons(ETH_P_8021Q) &&
 	     dev->features & NETIF_F_HW_VLAN_CTAG_RX) ||
 	    (vlan_proto == htons(ETH_P_8021AD) &&
diff --git a/drivers/staging/gdm724x/gdm_lte.c b/drivers/staging/gdm724x/gdm_lte.c
index 671ee8843c88..5703a9ddb6d0 100644
--- a/drivers/staging/gdm724x/gdm_lte.c
+++ b/drivers/staging/gdm724x/gdm_lte.c
@@ -349,7 +349,7 @@ static s32 gdm_lte_tx_nic_type(struct net_device *dev, struct sk_buff *skb)
 	/* Get ethernet protocol */
 	eth = (struct ethhdr *)skb->data;
 	if (ntohs(eth->h_proto) == ETH_P_8021Q) {
-		vlan_eth = (struct vlan_ethhdr *)skb->data;
+		vlan_eth = skb_vlan_eth_hdr(skb);
 		mac_proto = ntohs(vlan_eth->h_vlan_encapsulated_proto);
 		network_data = skb->data + VLAN_ETH_HLEN;
 		nic_type |= NIC_TYPE_F_VLAN;
@@ -435,7 +435,7 @@ static netdev_tx_t gdm_lte_tx(struct sk_buff *skb, struct net_device *dev)
 	 * driver based on the NIC mac
 	 */
 	if (nic_type & NIC_TYPE_F_VLAN) {
-		struct vlan_ethhdr *vlan_eth = (struct vlan_ethhdr *)skb->data;
+		struct vlan_ethhdr *vlan_eth = skb_vlan_eth_hdr(skb);
 
 		nic->vlan_id = ntohs(vlan_eth->h_vlan_TCI) & VLAN_VID_MASK;
 		data_buf = skb->data + (VLAN_ETH_HLEN - ETH_HLEN);
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 90b76d63c11c..3698f2b391cd 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -62,6 +62,14 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb)
 	return (struct vlan_ethhdr *)skb_mac_header(skb);
 }
 
+/* Prefer this version in TX path, instead of
+ * skb_reset_mac_header() + vlan_eth_hdr()
+ */
+static inline struct vlan_ethhdr *skb_vlan_eth_hdr(const struct sk_buff *skb)
+{
+	return (struct vlan_ethhdr *)skb->data;
+}
+
 #define VLAN_PRIO_MASK		0xe000 /* Priority Code Point */
 #define VLAN_PRIO_SHIFT		13
 #define VLAN_CFI_MASK		0x1000 /* Canonical Format Indicator / Drop Eligible Indicator */
@@ -529,7 +537,7 @@ static inline void __vlan_hwaccel_put_tag(struct sk_buff *skb,
  */
 static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
-	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)skb->data;
+	struct vlan_ethhdr *veth = skb_vlan_eth_hdr(skb);
 
 	if (!eth_type_vlan(veth->h_vlan_proto))
 		return -EINVAL;
@@ -713,7 +721,7 @@ static inline bool skb_vlan_tagged_multi(struct sk_buff *skb)
 		if (unlikely(!pskb_may_pull(skb, VLAN_ETH_HLEN)))
 			return false;
 
-		veh = (struct vlan_ethhdr *)skb->data;
+		veh = skb_vlan_eth_hdr(skb);
 		protocol = veh->h_vlan_encapsulated_proto;
 	}
 
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 125f4628687c..d3fdf82282af 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -439,7 +439,7 @@ void batadv_interface_rx(struct net_device *soft_iface,
 		if (!pskb_may_pull(skb, VLAN_ETH_HLEN))
 			goto dropped;
 
-		vhdr = (struct vlan_ethhdr *)skb->data;
+		vhdr = skb_vlan_eth_hdr(skb);
 
 		/* drop batman-in-batman packets to prevent loops */
 		if (vhdr->h_vlan_encapsulated_proto != htons(ETH_P_BATMAN))
-- 
cgit v1.2.3-70-g09d2


From e2fdfd7119128b49d81972a75e74169802aa8ae0 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:55 +0300
Subject: net: dpaa: avoid one skb_reset_mac_header() in dpaa_enable_tx_csum()

It appears that dpaa_enable_tx_csum() only calls skb_reset_mac_header()
to get to the VLAN header using skb_mac_header().

We can use skb_vlan_eth_hdr() to get to the VLAN header based on
skb->data directly. This avoids spending a few cycles to set
skb->mac_header.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 9318a2554056..1fa676308c5e 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -1482,13 +1482,8 @@ static int dpaa_enable_tx_csum(struct dpaa_priv *priv,
 	parse_result = (struct fman_prs_result *)parse_results;
 
 	/* If we're dealing with VLAN, get the real Ethernet type */
-	if (ethertype == ETH_P_8021Q) {
-		/* We can't always assume the MAC header is set correctly
-		 * by the stack, so reset to beginning of skb->data
-		 */
-		skb_reset_mac_header(skb);
-		ethertype = ntohs(vlan_eth_hdr(skb)->h_vlan_encapsulated_proto);
-	}
+	if (ethertype == ETH_P_8021Q)
+		ethertype = ntohs(skb_vlan_eth_hdr(skb)->h_vlan_encapsulated_proto);
 
 	/* Fill in the relevant L3 parse result fields
 	 * and read the L4 protocol type
-- 
cgit v1.2.3-70-g09d2


From eabb1494c9f20362ae53a9991481a1523be4f4b7 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:56 +0300
Subject: net: dsa: tag_ocelot: do not rely on skb_mac_header() for VLAN xmit

skb_mac_header() will no longer be available in the TX path when
reverting commit 6d1ccff62780 ("net: reset mac header in
dev_start_xmit()"). As preparation for that, let's use
skb_vlan_eth_hdr() to get to the VLAN header instead, which assumes it's
located at skb->data (assumption which holds true here).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/tag_ocelot.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c
index 28ebecafdd24..73ee09de1a3a 100644
--- a/net/dsa/tag_ocelot.c
+++ b/net/dsa/tag_ocelot.c
@@ -26,7 +26,7 @@ static void ocelot_xmit_get_vlan_info(struct sk_buff *skb, struct dsa_port *dp,
 		return;
 	}
 
-	hdr = (struct vlan_ethhdr *)skb_mac_header(skb);
+	hdr = skb_vlan_eth_hdr(skb);
 	br_vlan_get_proto(br, &proto);
 
 	if (ntohs(hdr->h_vlan_proto) == proto) {
-- 
cgit v1.2.3-70-g09d2


From 499b2491d550677b824b828ff431e4ef4d1d3b9d Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:57 +0300
Subject: net: dsa: tag_ksz: do not rely on skb_mac_header() in TX paths

skb_mac_header() will no longer be available in the TX path when
reverting commit 6d1ccff62780 ("net: reset mac header in
dev_start_xmit()"). As preparation for that, let's use skb_eth_hdr() to
get to the Ethernet header's MAC DA instead, helper which assumes this
header is located at skb->data (assumption which holds true here).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/tag_ksz.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c
index 0eb1c7784c3d..ea100bd25939 100644
--- a/net/dsa/tag_ksz.c
+++ b/net/dsa/tag_ksz.c
@@ -120,18 +120,18 @@ static struct sk_buff *ksz_common_rcv(struct sk_buff *skb,
 static struct sk_buff *ksz8795_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct ethhdr *hdr;
 	u8 *tag;
-	u8 *addr;
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))
 		return NULL;
 
 	/* Tag encoding */
 	tag = skb_put(skb, KSZ_INGRESS_TAG_LEN);
-	addr = skb_mac_header(skb);
+	hdr = skb_eth_hdr(skb);
 
 	*tag = 1 << dp->index;
-	if (is_link_local_ether_addr(addr))
+	if (is_link_local_ether_addr(hdr->h_dest))
 		*tag |= KSZ8795_TAIL_TAG_OVERRIDE;
 
 	return skb;
@@ -273,8 +273,8 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb,
 	u16 queue_mapping = skb_get_queue_mapping(skb);
 	u8 prio = netdev_txq_to_tc(dev, queue_mapping);
 	struct dsa_port *dp = dsa_slave_to_port(dev);
+	struct ethhdr *hdr;
 	__be16 *tag;
-	u8 *addr;
 	u16 val;
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))
@@ -284,13 +284,13 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb,
 	ksz_xmit_timestamp(dp, skb);
 
 	tag = skb_put(skb, KSZ9477_INGRESS_TAG_LEN);
-	addr = skb_mac_header(skb);
+	hdr = skb_eth_hdr(skb);
 
 	val = BIT(dp->index);
 
 	val |= FIELD_PREP(KSZ9477_TAIL_TAG_PRIO, prio);
 
-	if (is_link_local_ether_addr(addr))
+	if (is_link_local_ether_addr(hdr->h_dest))
 		val |= KSZ9477_TAIL_TAG_OVERRIDE;
 
 	*tag = cpu_to_be16(val);
@@ -337,7 +337,7 @@ static struct sk_buff *ksz9893_xmit(struct sk_buff *skb,
 	u16 queue_mapping = skb_get_queue_mapping(skb);
 	u8 prio = netdev_txq_to_tc(dev, queue_mapping);
 	struct dsa_port *dp = dsa_slave_to_port(dev);
-	u8 *addr;
+	struct ethhdr *hdr;
 	u8 *tag;
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))
@@ -347,13 +347,13 @@ static struct sk_buff *ksz9893_xmit(struct sk_buff *skb,
 	ksz_xmit_timestamp(dp, skb);
 
 	tag = skb_put(skb, KSZ_INGRESS_TAG_LEN);
-	addr = skb_mac_header(skb);
+	hdr = skb_eth_hdr(skb);
 
 	*tag = BIT(dp->index);
 
 	*tag |= FIELD_PREP(KSZ9893_TAIL_TAG_PRIO, prio);
 
-	if (is_link_local_ether_addr(addr))
+	if (is_link_local_ether_addr(hdr->h_dest))
 		*tag |= KSZ9893_TAIL_TAG_OVERRIDE;
 
 	return ksz_defer_xmit(dp, skb);
-- 
cgit v1.2.3-70-g09d2


From f9346f00b5af0b32503dcd74072392d9592163ef Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:58 +0300
Subject: net: dsa: tag_sja1105: don't rely on skb_mac_header() in TX paths

skb_mac_header() will no longer be available in the TX path when
reverting commit 6d1ccff62780 ("net: reset mac header in
dev_start_xmit()"). As preparation for that, let's use
skb_vlan_eth_hdr() to get to the VLAN header instead, which assumes it's
located at skb->data (assumption which holds true here).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/tag_sja1105.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 1c2ceba4771b..a7ca97b7ac9e 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -256,7 +256,7 @@ static struct sk_buff *sja1105_pvid_tag_control_pkt(struct dsa_port *dp,
 			return NULL;
 	}
 
-	hdr = (struct vlan_ethhdr *)skb_mac_header(skb);
+	hdr = skb_vlan_eth_hdr(skb);
 
 	/* If skb is already VLAN-tagged, leave that VLAN ID in place */
 	if (hdr->h_vlan_proto == xmit_tpid)
-- 
cgit v1.2.3-70-g09d2


From b5653b157e55a25ec76506e8cd31353e0ed4944c Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:55:59 +0300
Subject: net: dsa: tag_sja1105: replace skb_mac_header() with vlan_eth_hdr()

This is a cosmetic patch which consolidates the code to use the helper
function offered by if_vlan.h.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/tag_sja1105.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index a7ca97b7ac9e..a5f3b73da417 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -516,7 +516,7 @@ static bool sja1110_skb_has_inband_control_extension(const struct sk_buff *skb)
 static void sja1105_vlan_rcv(struct sk_buff *skb, int *source_port,
 			     int *switch_id, int *vbid, u16 *vid)
 {
-	struct vlan_ethhdr *hdr = (struct vlan_ethhdr *)skb_mac_header(skb);
+	struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
 	u16 vlan_tci;
 
 	if (skb_vlan_tag_present(skb))
-- 
cgit v1.2.3-70-g09d2


From f0a9d563064c3040a2d7b32317b37a6bcb1099b2 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:56:00 +0300
Subject: net: dsa: update TX path comments to not mention skb_mac_header()

Once commit 6d1ccff62780 ("net: reset mac header in dev_start_xmit()")
will be reverted, it will no longer be true that skb->data points at
skb_mac_header(skb) - since the skb->mac_header will not be set - so
stop saying that, and just say that it points to the MAC header.

I've reviewed vlan_insert_tag() and it does not *actually* depend on
skb_mac_header(), so reword that to avoid the confusion.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/tag.h       | 2 +-
 net/dsa/tag_8021q.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/dsa/tag.h b/net/dsa/tag.h
index 7cfbca824f1c..32d12f4a9d73 100644
--- a/net/dsa/tag.h
+++ b/net/dsa/tag.h
@@ -229,7 +229,7 @@ static inline void *dsa_etype_header_pos_rx(struct sk_buff *skb)
 	return skb->data - 2;
 }
 
-/* On TX, skb->data points to skb_mac_header(skb), which means that EtherType
+/* On TX, skb->data points to the MAC header, which means that EtherType
  * header taggers start exactly where the EtherType is (the EtherType is
  * treated as part of the DSA header).
  */
diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index 5ee9ef00954e..cbdfc392f7e0 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -461,8 +461,8 @@ EXPORT_SYMBOL_GPL(dsa_tag_8021q_unregister);
 struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev,
 			       u16 tpid, u16 tci)
 {
-	/* skb->data points at skb_mac_header, which
-	 * is fine for vlan_insert_tag.
+	/* skb->data points at the MAC header, which is fine
+	 * for vlan_insert_tag().
 	 */
 	return vlan_insert_tag(skb, htons(tpid), tci);
 }
-- 
cgit v1.2.3-70-g09d2


From 0bcf2e4aca6c29a07555b713f2fb461dc38d5977 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Fri, 21 Apr 2023 01:56:01 +0300
Subject: net: dsa: tag_ocelot: call only the relevant portion of
 __skb_vlan_pop() on TX

ocelot_xmit_get_vlan_info() calls __skb_vlan_pop() as the most
appropriate helper I could find which strips away a VLAN header.
That's all I need it to do, but __skb_vlan_pop() has more logic, which
will become incompatible with the future revert of commit 6d1ccff62780
("net: reset mac header in dev_start_xmit()").

Namely, it performs a sanity check on skb_mac_header(), which will stop
being set after the above revert, so it will return an error instead of
removing the VLAN tag.

ocelot_xmit_get_vlan_info() gets called in 2 circumstances:

(1) the port is under a VLAN-aware bridge and the bridge sends
    VLAN-tagged packets

(2) the port is under a VLAN-aware bridge and somebody else (an 8021q
    upper) sends VLAN-tagged packets (using a VID that isn't in the
    bridge vlan tables)

In case (1), there is actually no bug to defend against, because
br_dev_xmit() calls skb_reset_mac_header() and things continue to work.

However, in case (2), illustrated using the commands below, it can be
seen that our intervention is needed, since __skb_vlan_pop() complains:

$ ip link add br0 type bridge vlan_filtering 1 && ip link set br0 up
$ ip link set $eth master br0 && ip link set $eth up
$ ip link add link $eth name $eth.100 type vlan id 100 && ip link set $eth.100 up
$ ip addr add 192.168.100.1/24 dev $eth.100

I could fend off the checks in __skb_vlan_pop() with some
skb_mac_header_was_set() calls, but seeing how few callers of
__skb_vlan_pop() there are from TX paths, that seems rather
unproductive.

As an alternative solution, extract the bare minimum logic to strip a
VLAN header, and move it to a new helper named vlan_remove_tag(), close
to the definition of vlan_insert_tag(). Document it appropriately and
make ocelot_xmit_get_vlan_info() call this smaller helper instead.

Seeing that it doesn't appear illegal to test skb->protocol in the TX
path, I guess it would be a good for vlan_remove_tag() to also absorb
the vlan_set_encap_proto() function call.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 21 +++++++++++++++++++++
 net/core/skbuff.c       |  8 +-------
 net/dsa/tag_ocelot.c    |  2 +-
 3 files changed, 23 insertions(+), 8 deletions(-)

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 3698f2b391cd..0f40f379d75c 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -685,6 +685,27 @@ static inline void vlan_set_encap_proto(struct sk_buff *skb,
 		skb->protocol = htons(ETH_P_802_2);
 }
 
+/**
+ * vlan_remove_tag - remove outer VLAN tag from payload
+ * @skb: skbuff to remove tag from
+ * @vlan_tci: buffer to store value
+ *
+ * Expects the skb to contain a VLAN tag in the payload, and to have skb->data
+ * pointing at the MAC header.
+ *
+ * Returns a new pointer to skb->data, or NULL on failure to pull.
+ */
+static inline void *vlan_remove_tag(struct sk_buff *skb, u16 *vlan_tci)
+{
+	struct vlan_hdr *vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN);
+
+	*vlan_tci = ntohs(vhdr->h_vlan_TCI);
+
+	memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN);
+	vlan_set_encap_proto(skb, vhdr);
+	return __skb_pull(skb, VLAN_HLEN);
+}
+
 /**
  * skb_vlan_tagged - check if skb is vlan tagged.
  * @skb: skbuff to query
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 304a966164d8..c7c141f6fc14 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5996,7 +5996,6 @@ EXPORT_SYMBOL(skb_ensure_writable);
  */
 int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci)
 {
-	struct vlan_hdr *vhdr;
 	int offset = skb->data - skb_mac_header(skb);
 	int err;
 
@@ -6012,13 +6011,8 @@ int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci)
 
 	skb_postpull_rcsum(skb, skb->data + (2 * ETH_ALEN), VLAN_HLEN);
 
-	vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN);
-	*vlan_tci = ntohs(vhdr->h_vlan_TCI);
-
-	memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN);
-	__skb_pull(skb, VLAN_HLEN);
+	vlan_remove_tag(skb, vlan_tci);
 
-	vlan_set_encap_proto(skb, vhdr);
 	skb->mac_header += VLAN_HLEN;
 
 	if (skb_network_offset(skb) < ETH_HLEN)
diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c
index 73ee09de1a3a..20bf7074d5a6 100644
--- a/net/dsa/tag_ocelot.c
+++ b/net/dsa/tag_ocelot.c
@@ -30,7 +30,7 @@ static void ocelot_xmit_get_vlan_info(struct sk_buff *skb, struct dsa_port *dp,
 	br_vlan_get_proto(br, &proto);
 
 	if (ntohs(hdr->h_vlan_proto) == proto) {
-		__skb_vlan_pop(skb, &tci);
+		vlan_remove_tag(skb, &tci);
 		*vlan_tci = tci;
 	} else {
 		rcu_read_lock();
-- 
cgit v1.2.3-70-g09d2


From 2f0f9465ad9fa9c93f30009184c10da0f504f313 Mon Sep 17 00:00:00 2001
From: Yajun Deng <yajun.deng@linux.dev>
Date: Fri, 21 Apr 2023 16:26:06 +0800
Subject: net: sched: Print msecs when transmit queue time out

The kernel will print several warnings in a short period of time
when it stalls. Like this:

First warning:
[ 7100.097547] ------------[ cut here ]------------
[ 7100.097550] NETDEV WATCHDOG: eno2 (xxx): transmit queue 8 timed out
[ 7100.097571] WARNING: CPU: 8 PID: 0 at net/sched/sch_generic.c:467
                       dev_watchdog+0x260/0x270
...

Second warning:
[ 7147.756952] rcu: INFO: rcu_preempt self-detected stall on CPU
[ 7147.756958] rcu:   24-....: (59999 ticks this GP) idle=546/1/0x400000000000000
                      softirq=367      3137/3673146 fqs=13844
[ 7147.756960]        (t=60001 jiffies g=4322709 q=133381)
[ 7147.756962] NMI backtrace for cpu 24
...

We calculate that the transmit queue start stall should occur before
7095s according to watchdog_timeo, the rcu start stall at 7087s.
These two times are close together, it is difficult to confirm which
happened first.

To let users know the exact time the stall started, print msecs when
the transmit queue time out.

Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_generic.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index a9aadc4e6858..37e41f972f69 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -502,7 +502,7 @@ static void dev_watchdog(struct timer_list *t)
 		if (netif_device_present(dev) &&
 		    netif_running(dev) &&
 		    netif_carrier_ok(dev)) {
-			int some_queue_timedout = 0;
+			unsigned int timedout_ms = 0;
 			unsigned int i;
 			unsigned long trans_start;
 
@@ -514,16 +514,16 @@ static void dev_watchdog(struct timer_list *t)
 				if (netif_xmit_stopped(txq) &&
 				    time_after(jiffies, (trans_start +
 							 dev->watchdog_timeo))) {
-					some_queue_timedout = 1;
+					timedout_ms = jiffies_to_msecs(jiffies - trans_start);
 					atomic_long_inc(&txq->trans_timeout);
 					break;
 				}
 			}
 
-			if (unlikely(some_queue_timedout)) {
+			if (unlikely(timedout_ms)) {
 				trace_net_dev_xmit_timeout(dev, i);
-				WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n",
-				       dev->name, netdev_drivername(dev), i);
+				WARN_ONCE(1, "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out %u ms\n",
+					  dev->name, netdev_drivername(dev), i, timedout_ms);
 				netif_freeze_queues(dev);
 				dev->netdev_ops->ndo_tx_timeout(dev, i);
 				netif_unfreeze_queues(dev);
-- 
cgit v1.2.3-70-g09d2


From 5036034572b79daa6d6600338e8e8229e2a44b09 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Fri, 21 Apr 2023 18:25:13 -0300
Subject: net/sched: act_pedit: use NLA_POLICY for parsing 'ex' keys

Transform two checks in the 'ex' key parsing into netlink policies
removing extra if checks.

Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_pedit.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index 4559a1507ea5..45201f75e896 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -30,8 +30,9 @@ static const struct nla_policy pedit_policy[TCA_PEDIT_MAX + 1] = {
 };
 
 static const struct nla_policy pedit_key_ex_policy[TCA_PEDIT_KEY_EX_MAX + 1] = {
-	[TCA_PEDIT_KEY_EX_HTYPE]  = { .type = NLA_U16 },
-	[TCA_PEDIT_KEY_EX_CMD]	  = { .type = NLA_U16 },
+	[TCA_PEDIT_KEY_EX_HTYPE] =
+		NLA_POLICY_MAX(NLA_U16, TCA_PEDIT_HDR_TYPE_MAX),
+	[TCA_PEDIT_KEY_EX_CMD] = NLA_POLICY_MAX(NLA_U16, TCA_PEDIT_CMD_MAX),
 };
 
 static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla,
@@ -81,12 +82,6 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla,
 		k->htype = nla_get_u16(tb[TCA_PEDIT_KEY_EX_HTYPE]);
 		k->cmd = nla_get_u16(tb[TCA_PEDIT_KEY_EX_CMD]);
 
-		if (k->htype > TCA_PEDIT_HDR_TYPE_MAX ||
-		    k->cmd > TCA_PEDIT_CMD_MAX) {
-			err = -EINVAL;
-			goto err_out;
-		}
-
 		k++;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 0c83c5210e18763ec54252b990d270748c99b01d Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Fri, 21 Apr 2023 18:25:14 -0300
Subject: net/sched: act_pedit: use extack in 'ex' parsing errors

We have extack available when parsing 'ex' keys, so pass it to
tcf_pedit_keys_ex_parse and add more detailed error messages.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_pedit.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index 45201f75e896..24976cd4e4a2 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -36,7 +36,7 @@ static const struct nla_policy pedit_key_ex_policy[TCA_PEDIT_KEY_EX_MAX + 1] = {
 };
 
 static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla,
-							u8 n)
+							u8 n, struct netlink_ext_ack *extack)
 {
 	struct tcf_pedit_key_ex *keys_ex;
 	struct tcf_pedit_key_ex *k;
@@ -57,12 +57,14 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla,
 		struct nlattr *tb[TCA_PEDIT_KEY_EX_MAX + 1];
 
 		if (!n) {
+			NL_SET_ERR_MSG_MOD(extack, "Can't parse more extended keys than requested");
 			err = -EINVAL;
 			goto err_out;
 		}
 		n--;
 
 		if (nla_type(ka) != TCA_PEDIT_KEY_EX) {
+			NL_SET_ERR_MSG_ATTR(extack, ka, "Unknown attribute, expected extended key");
 			err = -EINVAL;
 			goto err_out;
 		}
@@ -73,8 +75,14 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla,
 		if (err)
 			goto err_out;
 
-		if (!tb[TCA_PEDIT_KEY_EX_HTYPE] ||
-		    !tb[TCA_PEDIT_KEY_EX_CMD]) {
+		if (NL_REQ_ATTR_CHECK(extack, nla, tb, TCA_PEDIT_KEY_EX_HTYPE)) {
+			NL_SET_ERR_MSG(extack, "Missing required attribute");
+			err = -EINVAL;
+			goto err_out;
+		}
+
+		if (NL_REQ_ATTR_CHECK(extack, nla, tb, TCA_PEDIT_KEY_EX_CMD)) {
+			NL_SET_ERR_MSG(extack, "Missing required attribute");
 			err = -EINVAL;
 			goto err_out;
 		}
@@ -86,6 +94,7 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla,
 	}
 
 	if (n) {
+		NL_SET_ERR_MSG_MOD(extack, "Not enough extended keys to parse");
 		err = -EINVAL;
 		goto err_out;
 	}
@@ -217,7 +226,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla,
 	}
 
 	nparms->tcfp_keys_ex =
-		tcf_pedit_keys_ex_parse(tb[TCA_PEDIT_KEYS_EX], parm->nkeys);
+		tcf_pedit_keys_ex_parse(tb[TCA_PEDIT_KEYS_EX], parm->nkeys, extack);
 	if (IS_ERR(nparms->tcfp_keys_ex)) {
 		ret = PTR_ERR(nparms->tcfp_keys_ex);
 		goto out_free;
-- 
cgit v1.2.3-70-g09d2


From e1201bc781c28766720e78a5e099ffa568be4d74 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Fri, 21 Apr 2023 18:25:15 -0300
Subject: net/sched: act_pedit: check static offsets a priori

Static key offsets should always be on 32 bit boundaries. Validate them on
create/update time for static offsets and move the datapath validation
for runtime offsets only.

iproute2 already errors out if a given offset and data size cannot be
packed to a 32 bit boundary. This change will make sure users which
create/update pedit instances directly via netlink also error out,
instead of finding out when packets are traversing.

Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_pedit.c | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index 24976cd4e4a2..cc4dfb01c6c7 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -251,8 +251,16 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla,
 	memcpy(nparms->tcfp_keys, parm->keys, ksize);
 
 	for (i = 0; i < nparms->tcfp_nkeys; ++i) {
+		u32 offmask = nparms->tcfp_keys[i].offmask;
 		u32 cur = nparms->tcfp_keys[i].off;
 
+		/* The AT option can be added to static offsets in the datapath */
+		if (!offmask && cur % 4) {
+			NL_SET_ERR_MSG_MOD(extack, "Offsets must be on 32bit boundaries");
+			ret = -EINVAL;
+			goto put_chain;
+		}
+
 		/* sanitize the shift value for any later use */
 		nparms->tcfp_keys[i].shift = min_t(size_t,
 						   BITS_PER_TYPE(int) - 1,
@@ -261,7 +269,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla,
 		/* The AT option can read a single byte, we can bound the actual
 		 * value with uchar max.
 		 */
-		cur += (0xff & nparms->tcfp_keys[i].offmask) >> nparms->tcfp_keys[i].shift;
+		cur += (0xff & offmask) >> nparms->tcfp_keys[i].shift;
 
 		/* Each key touches 4 bytes starting from the computed offset */
 		nparms->tcfp_off_max_hint =
@@ -411,12 +419,12 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
 					       sizeof(_d), &_d);
 			if (!d)
 				goto bad;
-			offset += (*d & tkey->offmask) >> tkey->shift;
-		}
 
-		if (offset % 4) {
-			pr_info("tc action pedit offset must be on 32 bit boundaries\n");
-			goto bad;
+			offset += (*d & tkey->offmask) >> tkey->shift;
+			if (offset % 4) {
+				pr_info("tc action pedit offset must be on 32 bit boundaries\n");
+				goto bad;
+			}
 		}
 
 		if (!offset_valid(skb, hoffset + offset)) {
-- 
cgit v1.2.3-70-g09d2


From 577140180ba28d0d37bc898c7bd6702c83aa106f Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Fri, 21 Apr 2023 18:25:16 -0300
Subject: net/sched: act_pedit: remove extra check for key type

The netlink parsing already validates the key 'htype'.
Remove the datapath check as it's redundant.

Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_pedit.c | 29 +++++++----------------------
 1 file changed, 7 insertions(+), 22 deletions(-)

diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index cc4dfb01c6c7..2fec4473d800 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -325,37 +325,28 @@ static bool offset_valid(struct sk_buff *skb, int offset)
 	return true;
 }
 
-static int pedit_skb_hdr_offset(struct sk_buff *skb,
-				enum pedit_header_type htype, int *hoffset)
+static void pedit_skb_hdr_offset(struct sk_buff *skb,
+				 enum pedit_header_type htype, int *hoffset)
 {
-	int ret = -EINVAL;
-
+	/* 'htype' is validated in the netlink parsing */
 	switch (htype) {
 	case TCA_PEDIT_KEY_EX_HDR_TYPE_ETH:
-		if (skb_mac_header_was_set(skb)) {
+		if (skb_mac_header_was_set(skb))
 			*hoffset = skb_mac_offset(skb);
-			ret = 0;
-		}
 		break;
 	case TCA_PEDIT_KEY_EX_HDR_TYPE_NETWORK:
 	case TCA_PEDIT_KEY_EX_HDR_TYPE_IP4:
 	case TCA_PEDIT_KEY_EX_HDR_TYPE_IP6:
 		*hoffset = skb_network_offset(skb);
-		ret = 0;
 		break;
 	case TCA_PEDIT_KEY_EX_HDR_TYPE_TCP:
 	case TCA_PEDIT_KEY_EX_HDR_TYPE_UDP:
-		if (skb_transport_header_was_set(skb)) {
+		if (skb_transport_header_was_set(skb))
 			*hoffset = skb_transport_offset(skb);
-			ret = 0;
-		}
 		break;
 	default:
-		ret = -EINVAL;
 		break;
 	}
-
-	return ret;
 }
 
 TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
@@ -388,10 +379,9 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
 
 	for (i = parms->tcfp_nkeys; i > 0; i--, tkey++) {
 		int offset = tkey->off;
+		int hoffset = 0;
 		u32 *ptr, hdata;
-		int hoffset;
 		u32 val;
-		int rc;
 
 		if (tkey_ex) {
 			htype = tkey_ex->htype;
@@ -400,12 +390,7 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
 			tkey_ex++;
 		}
 
-		rc = pedit_skb_hdr_offset(skb, htype, &hoffset);
-		if (rc) {
-			pr_info("tc action pedit bad header type specified (0x%x)\n",
-				htype);
-			goto bad;
-		}
+		pedit_skb_hdr_offset(skb, htype, &hoffset);
 
 		if (tkey->offmask) {
 			u8 *d, _d;
-- 
cgit v1.2.3-70-g09d2


From e3c9673e2f6e1b3aa4bb87c570336e10f364c28a Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Fri, 21 Apr 2023 18:25:17 -0300
Subject: net/sched: act_pedit: rate limit datapath messages

Unbounded info messages in the pedit datapath can flood the printk
ring buffer quite easily depending on the action created.
As these messages are informational, usually printing some, not all,
is enough to bring attention to the real issue.

Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_pedit.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index 2fec4473d800..fb93d4c1faca 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -396,8 +396,8 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
 			u8 *d, _d;
 
 			if (!offset_valid(skb, hoffset + tkey->at)) {
-				pr_info("tc action pedit 'at' offset %d out of bounds\n",
-					hoffset + tkey->at);
+				pr_info_ratelimited("tc action pedit 'at' offset %d out of bounds\n",
+						    hoffset + tkey->at);
 				goto bad;
 			}
 			d = skb_header_pointer(skb, hoffset + tkey->at,
@@ -407,14 +407,13 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
 
 			offset += (*d & tkey->offmask) >> tkey->shift;
 			if (offset % 4) {
-				pr_info("tc action pedit offset must be on 32 bit boundaries\n");
+				pr_info_ratelimited("tc action pedit offset must be on 32 bit boundaries\n");
 				goto bad;
 			}
 		}
 
 		if (!offset_valid(skb, hoffset + offset)) {
-			pr_info("tc action pedit offset %d out of bounds\n",
-				hoffset + offset);
+			pr_info_ratelimited("tc action pedit offset %d out of bounds\n", hoffset + offset);
 			goto bad;
 		}
 
@@ -431,8 +430,7 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb,
 			val = (*ptr + tkey->val) & ~tkey->mask;
 			break;
 		default:
-			pr_info("tc action pedit bad command (%d)\n",
-				cmd);
+			pr_info_ratelimited("tc action pedit bad command (%d)\n", cmd);
 			goto bad;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From 807cfded92b0a2e8be9c078c693075790c7c4131 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Sat, 22 Apr 2023 12:56:09 -0300
Subject: net/sched: sch_htb: use extack on errors messages

Some error messages are still being printed to dmesg.
Since extack is available, provide error messages there.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_htb.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 92f2975b6a82..8aef7dd9fb88 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1786,7 +1786,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		goto failure;
 
 	err = nla_parse_nested_deprecated(tb, TCA_HTB_MAX, opt, htb_policy,
-					  NULL);
+					  extack);
 	if (err < 0)
 		goto failure;
 
@@ -1858,7 +1858,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 
 		/* check maximal depth */
 		if (parent && parent->parent && parent->parent->level < 2) {
-			pr_err("htb: tree is too deep\n");
+			NL_SET_ERR_MSG_MOD(extack, "tree is too deep");
 			goto failure;
 		}
 		err = -ENOBUFS;
@@ -1917,8 +1917,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 			};
 			err = htb_offload(dev, &offload_opt);
 			if (err) {
-				pr_err("htb: TC_HTB_LEAF_ALLOC_QUEUE failed with err = %d\n",
-				       err);
+				NL_SET_ERR_MSG_WEAK(extack,
+						    "Failed to offload TC_HTB_LEAF_ALLOC_QUEUE");
 				goto err_kill_estimator;
 			}
 			dev_queue = netdev_get_tx_queue(dev, offload_opt.qid);
@@ -1937,8 +1937,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 			};
 			err = htb_offload(dev, &offload_opt);
 			if (err) {
-				pr_err("htb: TC_HTB_LEAF_TO_INNER failed with err = %d\n",
-				       err);
+				NL_SET_ERR_MSG_WEAK(extack,
+						    "Failed to offload TC_HTB_LEAF_TO_INNER");
 				htb_graft_helper(dev_queue, old_q);
 				goto err_kill_estimator;
 			}
@@ -2067,8 +2067,9 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 	qdisc_put(parent_qdisc);
 
 	if (warn)
-		pr_warn("HTB: quantum of class %X is %s. Consider r2q change.\n",
-			    cl->common.classid, (warn == -1 ? "small" : "big"));
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "quantum of class %X is %s. Consider r2q change.",
+				       cl->common.classid, (warn == -1 ? "small" : "big"));
 
 	qdisc_class_hash_grow(sch, &q->clhash);
 
-- 
cgit v1.2.3-70-g09d2


From c69a9b023f65b73068a17f2f4eb6a1b6e257f5bf Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Sat, 22 Apr 2023 12:56:10 -0300
Subject: net/sched: sch_qfq: use extack on errors messages

Some error messages are still being printed to dmesg.
Since extack is available, provide error messages there.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_qfq.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 02098a02943e..abcc48087831 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -402,8 +402,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	int err;
 	int delta_w;
 
-	if (tca[TCA_OPTIONS] == NULL) {
-		pr_notice("qfq: no options\n");
+	if (NL_REQ_ATTR_CHECK(extack, NULL, tca, TCA_OPTIONS)) {
+		NL_SET_ERR_MSG_MOD(extack, "missing options");
 		return -EINVAL;
 	}
 
@@ -442,8 +442,9 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	delta_w = weight - (cl ? cl->agg->class_weight : 0);
 
 	if (q->wsum + delta_w > QFQ_MAX_WSUM) {
-		pr_notice("qfq: total weight out of range (%d + %u)\n",
-			  delta_w, q->wsum);
+		NL_SET_ERR_MSG_FMT_MOD(extack,
+				       "total weight out of range (%d + %u)\n",
+				       delta_w, q->wsum);
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 25369891fcef373540f8b4e0b3bccf77a04490d5 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Sat, 22 Apr 2023 12:56:11 -0300
Subject: net/sched: sch_qfq: refactor parsing of netlink parameters

Two parameters can be transformed into netlink policies and
validated while parsing the netlink message.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_qfq.c | 25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)

diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index abcc48087831..dfd9a99e6257 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -113,6 +113,7 @@
 
 #define QFQ_MTU_SHIFT		16	/* to support TSO/GSO */
 #define QFQ_MIN_LMAX		512	/* see qfq_slot_insert */
+#define QFQ_MAX_LMAX		(1UL << QFQ_MTU_SHIFT)
 
 #define QFQ_MAX_AGG_CLASSES	8 /* max num classes per aggregate allowed */
 
@@ -214,9 +215,14 @@ static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid)
 	return container_of(clc, struct qfq_class, common);
 }
 
+static struct netlink_range_validation lmax_range = {
+	.min = QFQ_MIN_LMAX,
+	.max = QFQ_MAX_LMAX,
+};
+
 static const struct nla_policy qfq_policy[TCA_QFQ_MAX + 1] = {
-	[TCA_QFQ_WEIGHT] = { .type = NLA_U32 },
-	[TCA_QFQ_LMAX] = { .type = NLA_U32 },
+	[TCA_QFQ_WEIGHT] = NLA_POLICY_RANGE(NLA_U32, 1, QFQ_MAX_WEIGHT),
+	[TCA_QFQ_LMAX] = NLA_POLICY_FULL_RANGE(NLA_U32, &lmax_range),
 };
 
 /*
@@ -408,17 +414,13 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	}
 
 	err = nla_parse_nested_deprecated(tb, TCA_QFQ_MAX, tca[TCA_OPTIONS],
-					  qfq_policy, NULL);
+					  qfq_policy, extack);
 	if (err < 0)
 		return err;
 
-	if (tb[TCA_QFQ_WEIGHT]) {
+	if (tb[TCA_QFQ_WEIGHT])
 		weight = nla_get_u32(tb[TCA_QFQ_WEIGHT]);
-		if (!weight || weight > (1UL << QFQ_MAX_WSHIFT)) {
-			pr_notice("qfq: invalid weight %u\n", weight);
-			return -EINVAL;
-		}
-	} else
+	else
 		weight = 1;
 
 	if (tb[TCA_QFQ_LMAX])
@@ -426,11 +428,6 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	else
 		lmax = psched_mtu(qdisc_dev(sch));
 
-	if (lmax < QFQ_MIN_LMAX || lmax > (1UL << QFQ_MTU_SHIFT)) {
-		pr_notice("qfq: invalid max length %u\n", lmax);
-		return -EINVAL;
-	}
-
 	inv_w = ONE_FP / weight;
 	weight = ONE_FP / inv_w;
 
-- 
cgit v1.2.3-70-g09d2


From 7eb060a51a3ba44acefafefd242dcbf12dd91fb9 Mon Sep 17 00:00:00 2001
From: Pedro Tammela <pctammela@mojatatu.com>
Date: Sat, 22 Apr 2023 12:56:12 -0300
Subject: selftests: tc-testing: add more tests for sch_qfq

The QFQ qdisc class has parameter bounds that are not being
checked for correctness.

Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/tc-testing/tc-tests/qdiscs/qfq.json  | 72 ++++++++++++++++++++++
 1 file changed, 72 insertions(+)

diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/qfq.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/qfq.json
index 330f1a25e0ab..147899a868d3 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/qfq.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/qfq.json
@@ -46,6 +46,30 @@
             "$IP link del dev $DUMMY type dummy"
         ]
     },
+    {
+        "id": "d364",
+        "name": "Test QFQ with max class weight setting",
+        "category": [
+            "qdisc",
+            "qfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root qfq"
+        ],
+        "cmdUnderTest": "$TC class add dev $DUMMY parent 1: classid 1:1 qfq weight 9999",
+        "expExitCode": "2",
+        "verifyCmd": "$TC class show dev $DUMMY",
+        "matchPattern": "class qfq 1:1 root weight 9999 maxpkt",
+        "matchCount": "0",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
     {
         "id": "8452",
         "name": "Create QFQ with class maxpkt setting",
@@ -70,6 +94,54 @@
             "$IP link del dev $DUMMY type dummy"
         ]
     },
+    {
+        "id": "22df",
+        "name": "Test QFQ class maxpkt setting lower bound",
+        "category": [
+            "qdisc",
+            "qfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root qfq"
+        ],
+        "cmdUnderTest": "$TC class add dev $DUMMY parent 1: classid 1:1 qfq maxpkt 128",
+        "expExitCode": "2",
+        "verifyCmd": "$TC class show dev $DUMMY",
+        "matchPattern": "class qfq 1:1 root weight 1 maxpkt 128",
+        "matchCount": "0",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
+    {
+        "id": "92ee",
+        "name": "Test QFQ class maxpkt setting upper bound",
+        "category": [
+            "qdisc",
+            "qfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link add dev $DUMMY type dummy || /bin/true",
+            "$TC qdisc add dev $DUMMY handle 1: root qfq"
+        ],
+        "cmdUnderTest": "$TC class add dev $DUMMY parent 1: classid 1:1 qfq maxpkt 99999",
+        "expExitCode": "2",
+        "verifyCmd": "$TC class show dev $DUMMY",
+        "matchPattern": "class qfq 1:1 root weight 1 maxpkt 99999",
+        "matchCount": "0",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP link del dev $DUMMY type dummy"
+        ]
+    },
     {
         "id": "d920",
         "name": "Create QFQ with multiple class setting",
-- 
cgit v1.2.3-70-g09d2


From a80d2c545ded86d0350b9a870735565d8b749786 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 13 Feb 2023 14:28:55 -0800
Subject: Bluetooth: MGMT: Use BIT macro when defining bitfields

This makes use of BIT macro when defining bitfields which makes it
clearer what bit it is toggling.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/mgmt.h | 80 ++++++++++++++++++++++----------------------
 1 file changed, 40 insertions(+), 40 deletions(-)

diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h
index e18a927669c0..a5801649f619 100644
--- a/include/net/bluetooth/mgmt.h
+++ b/include/net/bluetooth/mgmt.h
@@ -91,26 +91,26 @@ struct mgmt_rp_read_index_list {
 #define MGMT_MAX_NAME_LENGTH		(HCI_MAX_NAME_LENGTH + 1)
 #define MGMT_MAX_SHORT_NAME_LENGTH	(HCI_MAX_SHORT_NAME_LENGTH + 1)
 
-#define MGMT_SETTING_POWERED		0x00000001
-#define MGMT_SETTING_CONNECTABLE	0x00000002
-#define MGMT_SETTING_FAST_CONNECTABLE	0x00000004
-#define MGMT_SETTING_DISCOVERABLE	0x00000008
-#define MGMT_SETTING_BONDABLE		0x00000010
-#define MGMT_SETTING_LINK_SECURITY	0x00000020
-#define MGMT_SETTING_SSP		0x00000040
-#define MGMT_SETTING_BREDR		0x00000080
-#define MGMT_SETTING_HS			0x00000100
-#define MGMT_SETTING_LE			0x00000200
-#define MGMT_SETTING_ADVERTISING	0x00000400
-#define MGMT_SETTING_SECURE_CONN	0x00000800
-#define MGMT_SETTING_DEBUG_KEYS		0x00001000
-#define MGMT_SETTING_PRIVACY		0x00002000
-#define MGMT_SETTING_CONFIGURATION	0x00004000
-#define MGMT_SETTING_STATIC_ADDRESS	0x00008000
-#define MGMT_SETTING_PHY_CONFIGURATION	0x00010000
-#define MGMT_SETTING_WIDEBAND_SPEECH	0x00020000
-#define MGMT_SETTING_CIS_CENTRAL	0x00040000
-#define MGMT_SETTING_CIS_PERIPHERAL	0x00080000
+#define MGMT_SETTING_POWERED		BIT(0)
+#define MGMT_SETTING_CONNECTABLE	BIT(1)
+#define MGMT_SETTING_FAST_CONNECTABLE	BIT(2)
+#define MGMT_SETTING_DISCOVERABLE	BIT(3)
+#define MGMT_SETTING_BONDABLE		BIT(4)
+#define MGMT_SETTING_LINK_SECURITY	BIT(5)
+#define MGMT_SETTING_SSP		BIT(6)
+#define MGMT_SETTING_BREDR		BIT(7)
+#define MGMT_SETTING_HS			BIT(8)
+#define MGMT_SETTING_LE			BIT(9)
+#define MGMT_SETTING_ADVERTISING	BIT(10)
+#define MGMT_SETTING_SECURE_CONN	BIT(11)
+#define MGMT_SETTING_DEBUG_KEYS		BIT(12)
+#define MGMT_SETTING_PRIVACY		BIT(13)
+#define MGMT_SETTING_CONFIGURATION	BIT(14)
+#define MGMT_SETTING_STATIC_ADDRESS	BIT(15)
+#define MGMT_SETTING_PHY_CONFIGURATION	BIT(16)
+#define MGMT_SETTING_WIDEBAND_SPEECH	BIT(17)
+#define MGMT_SETTING_CIS_CENTRAL	BIT(18)
+#define MGMT_SETTING_CIS_PERIPHERAL	BIT(19)
 
 #define MGMT_OP_READ_INFO		0x0004
 #define MGMT_READ_INFO_SIZE		0
@@ -635,21 +635,21 @@ struct mgmt_rp_get_phy_configuration {
 } __packed;
 #define MGMT_GET_PHY_CONFIGURATION_SIZE	0
 
-#define MGMT_PHY_BR_1M_1SLOT	0x00000001
-#define MGMT_PHY_BR_1M_3SLOT	0x00000002
-#define MGMT_PHY_BR_1M_5SLOT	0x00000004
-#define MGMT_PHY_EDR_2M_1SLOT	0x00000008
-#define MGMT_PHY_EDR_2M_3SLOT	0x00000010
-#define MGMT_PHY_EDR_2M_5SLOT	0x00000020
-#define MGMT_PHY_EDR_3M_1SLOT	0x00000040
-#define MGMT_PHY_EDR_3M_3SLOT	0x00000080
-#define MGMT_PHY_EDR_3M_5SLOT	0x00000100
-#define MGMT_PHY_LE_1M_TX		0x00000200
-#define MGMT_PHY_LE_1M_RX		0x00000400
-#define MGMT_PHY_LE_2M_TX		0x00000800
-#define MGMT_PHY_LE_2M_RX		0x00001000
-#define MGMT_PHY_LE_CODED_TX	0x00002000
-#define MGMT_PHY_LE_CODED_RX	0x00004000
+#define MGMT_PHY_BR_1M_1SLOT		BIT(0)
+#define MGMT_PHY_BR_1M_3SLOT		BIT(1)
+#define MGMT_PHY_BR_1M_5SLOT		BIT(2)
+#define MGMT_PHY_EDR_2M_1SLOT		BIT(3)
+#define MGMT_PHY_EDR_2M_3SLOT		BIT(4)
+#define MGMT_PHY_EDR_2M_5SLOT		BIT(5)
+#define MGMT_PHY_EDR_3M_1SLOT		BIT(6)
+#define MGMT_PHY_EDR_3M_3SLOT		BIT(7)
+#define MGMT_PHY_EDR_3M_5SLOT		BIT(8)
+#define MGMT_PHY_LE_1M_TX		BIT(9)
+#define MGMT_PHY_LE_1M_RX		BIT(10)
+#define MGMT_PHY_LE_2M_TX		BIT(11)
+#define MGMT_PHY_LE_2M_RX		BIT(12)
+#define MGMT_PHY_LE_CODED_TX		BIT(13)
+#define MGMT_PHY_LE_CODED_RX		BIT(14)
 
 #define MGMT_PHY_BREDR_MASK (MGMT_PHY_BR_1M_1SLOT | MGMT_PHY_BR_1M_3SLOT | \
 			     MGMT_PHY_BR_1M_5SLOT | MGMT_PHY_EDR_2M_1SLOT | \
@@ -974,11 +974,11 @@ struct mgmt_ev_auth_failed {
 	__u8	status;
 } __packed;
 
-#define MGMT_DEV_FOUND_CONFIRM_NAME		0x01
-#define MGMT_DEV_FOUND_LEGACY_PAIRING		0x02
-#define MGMT_DEV_FOUND_NOT_CONNECTABLE		0x04
-#define MGMT_DEV_FOUND_INITIATED_CONN		0x08
-#define MGMT_DEV_FOUND_NAME_REQUEST_FAILED	0x10
+#define MGMT_DEV_FOUND_CONFIRM_NAME		BIT(0)
+#define MGMT_DEV_FOUND_LEGACY_PAIRING		BIT(1)
+#define MGMT_DEV_FOUND_NOT_CONNECTABLE		BIT(2)
+#define MGMT_DEV_FOUND_INITIATED_CONN		BIT(3)
+#define MGMT_DEV_FOUND_NAME_REQUEST_FAILED	BIT(4)
 
 #define MGMT_EV_DEVICE_FOUND		0x0012
 struct mgmt_ev_device_found {
-- 
cgit v1.2.3-70-g09d2


From c13380a55522bf14e54779a142487c224509db95 Mon Sep 17 00:00:00 2001
From: Tomasz Moń <tomasz.mon@nordicsemi.no>
Date: Tue, 7 Feb 2023 12:57:41 +0100
Subject: Bluetooth: btusb: Do not require hardcoded interface numbers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove hardcoded interface number check because Bluetooth specification
since version 4.0 only recommends and no longer requires specific
interface numbers.

While earlier Bluetooth versions, i.e. 2.1 + EDR and 3.0 + HS, contain
required configuration table in Volume 4 - Host Controller Interface
Part B - USB Transport Layer, Bluetooth Core Specification Addendum 2
changes the table from required to recommended configuration.

Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 5c536151ef83..4ca91c033d2f 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3831,13 +3831,9 @@ static int btusb_probe(struct usb_interface *intf,
 
 	BT_DBG("intf %p id %p", intf, id);
 
-	/* interface numbers are hardcoded in the spec */
-	if (intf->cur_altsetting->desc.bInterfaceNumber != 0) {
-		if (!(id->driver_info & BTUSB_IFNUM_2))
-			return -ENODEV;
-		if (intf->cur_altsetting->desc.bInterfaceNumber != 2)
-			return -ENODEV;
-	}
+	if ((id->driver_info & BTUSB_IFNUM_2) &&
+	    (intf->cur_altsetting->desc.bInterfaceNumber != 2))
+		return -ENODEV;
 
 	ifnum_base = intf->cur_altsetting->desc.bInterfaceNumber;
 
-- 
cgit v1.2.3-70-g09d2


From c43132abfbe9d2f3befdbddfc3097d30905dc3d1 Mon Sep 17 00:00:00 2001
From: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Date: Mon, 13 Feb 2023 13:09:22 +0100
Subject: dt-bindings: bluetooth: marvell: add 88W8997

Update the documentation with the device tree binding for the Marvell
88W8997 bluetooth device.

Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 Documentation/devicetree/bindings/net/marvell-bluetooth.yaml | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml b/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
index 6aa7a078faa2..65e09476b048 100644
--- a/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
@@ -15,7 +15,9 @@ maintainers:
 
 properties:
   compatible:
-    const: mrvl,88w8897
+    enum:
+      - mrvl,88w8897
+      - mrvl,88w8997
 
 required:
   - compatible
-- 
cgit v1.2.3-70-g09d2


From 552705fd891b6ed6dda340567ce3ffff943f1f3b Mon Sep 17 00:00:00 2001
From: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Date: Mon, 13 Feb 2023 13:09:23 +0100
Subject: dt-bindings: bluetooth: marvell: add max-speed property

The 88W8997 bluetooth module supports setting the max-speed property.

Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../devicetree/bindings/net/marvell-bluetooth.yaml       | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml b/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
index 65e09476b048..188a42ca6ceb 100644
--- a/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/marvell-bluetooth.yaml
@@ -19,9 +19,25 @@ properties:
       - mrvl,88w8897
       - mrvl,88w8997
 
+  max-speed:
+    description: see Documentation/devicetree/bindings/serial/serial.yaml
+
 required:
   - compatible
 
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: mrvl,88w8997
+    then:
+      properties:
+        max-speed: true
+    else:
+      properties:
+        max-speed: false
+
 additionalProperties: false
 
 examples:
-- 
cgit v1.2.3-70-g09d2


From 069690a92ffdda399350af072ea62640bcd370df Mon Sep 17 00:00:00 2001
From: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Date: Mon, 13 Feb 2023 13:09:24 +0100
Subject: Bluetooth: hci_mrvl: use maybe_unused macro for device tree ids

Use the maybe_unused macro for the device tree ids instead of #ifdef
CONFIG_OF. This makes it easier to add support for new devices.

Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_mrvl.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/bluetooth/hci_mrvl.c b/drivers/bluetooth/hci_mrvl.c
index fbc3f7c3a5c7..eaa9c51cacfa 100644
--- a/drivers/bluetooth/hci_mrvl.c
+++ b/drivers/bluetooth/hci_mrvl.c
@@ -414,13 +414,11 @@ static void mrvl_serdev_remove(struct serdev_device *serdev)
 	hci_uart_unregister_device(&mrvldev->hu);
 }
 
-#ifdef CONFIG_OF
-static const struct of_device_id mrvl_bluetooth_of_match[] = {
+static const struct of_device_id __maybe_unused mrvl_bluetooth_of_match[] = {
 	{ .compatible = "mrvl,88w8897" },
 	{ },
 };
 MODULE_DEVICE_TABLE(of, mrvl_bluetooth_of_match);
-#endif
 
 static struct serdev_device_driver mrvl_serdev_driver = {
 	.probe = mrvl_serdev_probe,
-- 
cgit v1.2.3-70-g09d2


From a860c50f7aef5186fcbb73aff224ae6f5e51cecc Mon Sep 17 00:00:00 2001
From: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Date: Mon, 13 Feb 2023 13:09:25 +0100
Subject: Bluetooth: hci_mrvl: Add serdev support for 88W8997

Add serdev support for the 88W8997 from NXP (previously Marvell). It
includes support for changing the baud rate. The command to change the
baud rate is taken from the user manual UM11483 Rev. 9 in section 7
(Bring-up of Bluetooth interfaces) from NXP.

Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_mrvl.c | 86 ++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 79 insertions(+), 7 deletions(-)

diff --git a/drivers/bluetooth/hci_mrvl.c b/drivers/bluetooth/hci_mrvl.c
index eaa9c51cacfa..e08222395772 100644
--- a/drivers/bluetooth/hci_mrvl.c
+++ b/drivers/bluetooth/hci_mrvl.c
@@ -27,10 +27,12 @@
 #define MRVL_ACK 0x5A
 #define MRVL_NAK 0xBF
 #define MRVL_RAW_DATA 0x1F
+#define MRVL_SET_BAUDRATE 0xFC09
 
 enum {
 	STATE_CHIP_VER_PENDING,
 	STATE_FW_REQ_PENDING,
+	STATE_FW_LOADED,
 };
 
 struct mrvl_data {
@@ -254,6 +256,14 @@ static int mrvl_recv(struct hci_uart *hu, const void *data, int count)
 	if (!test_bit(HCI_UART_REGISTERED, &hu->flags))
 		return -EUNATCH;
 
+	/* We might receive some noise when there is no firmware loaded. Therefore,
+	 * we drop data if the firmware is not loaded yet and if there is no fw load
+	 * request pending.
+	 */
+	if (!test_bit(STATE_FW_REQ_PENDING, &mrvl->flags) &&
+				!test_bit(STATE_FW_LOADED, &mrvl->flags))
+		return count;
+
 	mrvl->rx_skb = h4_recv_buf(hu->hdev, mrvl->rx_skb, data, count,
 				    mrvl_recv_pkts,
 				    ARRAY_SIZE(mrvl_recv_pkts));
@@ -354,6 +364,7 @@ static int mrvl_load_firmware(struct hci_dev *hdev, const char *name)
 static int mrvl_setup(struct hci_uart *hu)
 {
 	int err;
+	struct mrvl_data *mrvl = hu->priv;
 
 	hci_uart_set_flow_control(hu, true);
 
@@ -367,9 +378,9 @@ static int mrvl_setup(struct hci_uart *hu)
 	hci_uart_wait_until_sent(hu);
 
 	if (hu->serdev)
-		serdev_device_set_baudrate(hu->serdev, 3000000);
+		serdev_device_set_baudrate(hu->serdev, hu->oper_speed);
 	else
-		hci_uart_set_baudrate(hu, 3000000);
+		hci_uart_set_baudrate(hu, hu->oper_speed);
 
 	hci_uart_set_flow_control(hu, false);
 
@@ -377,13 +388,54 @@ static int mrvl_setup(struct hci_uart *hu)
 	if (err)
 		return err;
 
+	set_bit(STATE_FW_LOADED, &mrvl->flags);
+
+	return 0;
+}
+
+static int mrvl_set_baudrate(struct hci_uart *hu, unsigned int speed)
+{
+	int err;
+	struct mrvl_data *mrvl = hu->priv;
+	__le32 speed_le = cpu_to_le32(speed);
+
+	/* The firmware might be loaded by the Wifi driver over SDIO. We wait
+	 * up to 10s for the CTS to go up. Afterward, we know that the firmware
+	 * is ready.
+	 */
+	err = serdev_device_wait_for_cts(hu->serdev, true, 10000);
+	if (err) {
+		bt_dev_err(hu->hdev, "Wait for CTS failed with %d\n", err);
+		return err;
+	}
+
+	set_bit(STATE_FW_LOADED, &mrvl->flags);
+
+	err = __hci_cmd_sync_status(hu->hdev, MRVL_SET_BAUDRATE,
+				    sizeof(speed_le), &speed_le,
+				    HCI_INIT_TIMEOUT);
+	if (err) {
+		bt_dev_err(hu->hdev, "send command failed: %d", err);
+		return err;
+	}
+
+	serdev_device_set_baudrate(hu->serdev, speed);
+
+	/* We forcefully have to send a command to the bluetooth module so that
+	 * the driver detects it after a baudrate change. This is foreseen by
+	 * hci_serdev by setting HCI_UART_VND_DETECT which then causes a dummy
+	 * local version read.
+	 */
+	set_bit(HCI_UART_VND_DETECT, &hu->hdev_flags);
+
 	return 0;
 }
 
-static const struct hci_uart_proto mrvl_proto = {
+static const struct hci_uart_proto mrvl_proto_8897 = {
 	.id		= HCI_UART_MRVL,
 	.name		= "Marvell",
 	.init_speed	= 115200,
+	.oper_speed	= 3000000,
 	.open		= mrvl_open,
 	.close		= mrvl_close,
 	.flush		= mrvl_flush,
@@ -393,18 +445,37 @@ static const struct hci_uart_proto mrvl_proto = {
 	.dequeue	= mrvl_dequeue,
 };
 
+static const struct hci_uart_proto mrvl_proto_8997 = {
+	.id		= HCI_UART_MRVL,
+	.name		= "Marvell 8997",
+	.init_speed	= 115200,
+	.oper_speed	= 3000000,
+	.open		= mrvl_open,
+	.close		= mrvl_close,
+	.flush		= mrvl_flush,
+	.set_baudrate   = mrvl_set_baudrate,
+	.recv		= mrvl_recv,
+	.enqueue	= mrvl_enqueue,
+	.dequeue	= mrvl_dequeue,
+};
+
 static int mrvl_serdev_probe(struct serdev_device *serdev)
 {
 	struct mrvl_serdev *mrvldev;
+	const struct hci_uart_proto *mrvl_proto = device_get_match_data(&serdev->dev);
 
 	mrvldev = devm_kzalloc(&serdev->dev, sizeof(*mrvldev), GFP_KERNEL);
 	if (!mrvldev)
 		return -ENOMEM;
 
+	mrvldev->hu.oper_speed = mrvl_proto->oper_speed;
+	if (mrvl_proto->set_baudrate)
+		of_property_read_u32(serdev->dev.of_node, "max-speed", &mrvldev->hu.oper_speed);
+
 	mrvldev->hu.serdev = serdev;
 	serdev_device_set_drvdata(serdev, mrvldev);
 
-	return hci_uart_register_device(&mrvldev->hu, &mrvl_proto);
+	return hci_uart_register_device(&mrvldev->hu, mrvl_proto);
 }
 
 static void mrvl_serdev_remove(struct serdev_device *serdev)
@@ -415,7 +486,8 @@ static void mrvl_serdev_remove(struct serdev_device *serdev)
 }
 
 static const struct of_device_id __maybe_unused mrvl_bluetooth_of_match[] = {
-	{ .compatible = "mrvl,88w8897" },
+	{ .compatible = "mrvl,88w8897", .data = &mrvl_proto_8897},
+	{ .compatible = "mrvl,88w8997", .data = &mrvl_proto_8997},
 	{ },
 };
 MODULE_DEVICE_TABLE(of, mrvl_bluetooth_of_match);
@@ -433,12 +505,12 @@ int __init mrvl_init(void)
 {
 	serdev_device_driver_register(&mrvl_serdev_driver);
 
-	return hci_uart_register_proto(&mrvl_proto);
+	return hci_uart_register_proto(&mrvl_proto_8897);
 }
 
 int __exit mrvl_deinit(void)
 {
 	serdev_device_driver_unregister(&mrvl_serdev_driver);
 
-	return hci_uart_unregister_proto(&mrvl_proto);
+	return hci_uart_unregister_proto(&mrvl_proto_8897);
 }
-- 
cgit v1.2.3-70-g09d2


From fe4b71b59ed5067ae591a33d512947160fd9aa89 Mon Sep 17 00:00:00 2001
From: Victor Hassan <victor@allwinnertech.com>
Date: Wed, 22 Feb 2023 17:47:20 +0800
Subject: Bluetooth: btrtl: Add support for RTL8852BS

Add the support for RTL8852BS BT controller on UART interface.
The necessary firmware file will be submitted to linux-firmware.

Signed-off-by: Victor Hassan <victor@allwinnertech.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index 69c3fe649ca7..aee75f278e60 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -190,6 +190,14 @@ static const struct id_table ic_id_table[] = {
 	  .fw_name  = "rtl_bt/rtl8852au_fw.bin",
 	  .cfg_name = "rtl_bt/rtl8852au_config" },
 
+	/* 8852B with UART interface */
+	{ IC_INFO(RTL_ROM_LMP_8852A, 0xb, 0xb, HCI_UART),
+	  .config_needed = true,
+	  .has_rom_version = true,
+	  .has_msft_ext = true,
+	  .fw_name  = "rtl_bt/rtl8852bs_fw.bin",
+	  .cfg_name = "rtl_bt/rtl8852bs_config" },
+
 	/* 8852B */
 	{ IC_INFO(RTL_ROM_LMP_8852A, 0xb, 0xb, HCI_USB),
 	  .config_needed = false,
@@ -963,6 +971,8 @@ MODULE_FIRMWARE("rtl_bt/rtl8822b_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8822b_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852au_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852au_config.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8852bs_fw.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8852bs_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852bu_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852bu_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852cu_fw.bin");
-- 
cgit v1.2.3-70-g09d2


From d4b20f0b8491bfb3238ebf7df4b13243189620d2 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Fri, 24 Feb 2023 15:27:33 -0800
Subject: Bluetooth: hci_core: Make hci_conn_hash_add append to the list

This makes hci_conn_hash_add append to the tail of the conn_hash so it
matches the order they are created, this is required if the controller
attempts to match the order of ACL with CIS which uses append logic
when programming the CIS ids on the CIG.

The result of this change affects Create CIS:

Before:

< HCI Command: LE Create Connected Isochronous Stream (0x08|0x0064) plen 9
        Number of CIS: 2
        CIS Handle: 2560
        ACL Handle: 3586
        CIS Handle: 2561
        ACL Handle: 3585

After:

< HCI Command: LE Create Connected Isochronous Stream (0x08|0x0064) plen 9
        Number of CIS: 2
        CIS Handle: 2560
        ACL Handle: 3585
        CIS Handle: 2561
        ACL Handle: 3586

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index d5311ceb21c6..e22e45fbe8db 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -979,7 +979,7 @@ static inline bool hci_conn_sc_enabled(struct hci_conn *conn)
 static inline void hci_conn_hash_add(struct hci_dev *hdev, struct hci_conn *c)
 {
 	struct hci_conn_hash *h = &hdev->conn_hash;
-	list_add_rcu(&c->list, &h->list);
+	list_add_tail_rcu(&c->list, &h->list);
 	switch (c->type) {
 	case ACL_LINK:
 		h->acl_num++;
-- 
cgit v1.2.3-70-g09d2


From ae051b044ebf5938940741c25a8989e07a5d7e88 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 27 Feb 2023 15:50:58 -0800
Subject: Bluetooth: hci_sync: Fix smatch warning

This fixes the following new warning:

net/bluetooth/hci_sync.c:2403 hci_pause_addr_resolution() warn: missing
error code? 'err'

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/r/202302251952.xryXOegd-lkp@intel.com/
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_sync.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 632be1267288..b64af017ab99 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -2403,7 +2403,7 @@ static int hci_pause_addr_resolution(struct hci_dev *hdev)
 
 	/* Return if address resolution is disabled and RPA is not used. */
 	if (!err && scan_use_rpa(hdev))
-		return err;
+		return 0;
 
 	hci_resume_advertising_sync(hdev);
 	return err;
-- 
cgit v1.2.3-70-g09d2


From 13209415d0e88396d99d346b184864834d70d68a Mon Sep 17 00:00:00 2001
From: Meng Tang <tangmeng@uniontech.com>
Date: Tue, 28 Feb 2023 13:55:17 +0800
Subject: Bluetooth: btusb: Add new PID/VID 04ca:3801 for MT7663

This bluetooth device is found in a combo WLAN/BT card
for a MediaTek 7663.

Tested on Acer Aspire A315-24P Notebook

The device information:

T:  Bus=01 Lev=01 Prnt=01 Port=01 Cnt=01 Dev#=  2 Spd=480  MxCh= 0
D:  Ver= 2.10 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=04ca ProdID=3801 Rev= 1.00
S:  Manufacturer=MediaTek Inc.
S:  Product=Wireless_Device
S:  SerialNumber=000000000
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
A:  FirstIf#= 0 IfCount= 2 Cls=e0(wlcon) Sub=01 Prot=01
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=125us
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:  If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  63 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  63 Ivl=1ms

Signed-off-by: Meng Tang <tangmeng@uniontech.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 4ca91c033d2f..372007f78467 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -558,6 +558,9 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x043e, 0x310c), .driver_info = BTUSB_MEDIATEK |
 						     BTUSB_WIDEBAND_SPEECH |
 						     BTUSB_VALID_LE_STATES },
+	{ USB_DEVICE(0x04ca, 0x3801), .driver_info = BTUSB_MEDIATEK |
+						     BTUSB_WIDEBAND_SPEECH |
+						     BTUSB_VALID_LE_STATES },
 
 	/* Additional MediaTek MT7668 Bluetooth devices */
 	{ USB_DEVICE(0x043e, 0x3109), .driver_info = BTUSB_MEDIATEK |
-- 
cgit v1.2.3-70-g09d2


From 95ee3a93239e172b592dc4613c403de2410e4a82 Mon Sep 17 00:00:00 2001
From: Chris Morgan <macromorgan@hotmail.com>
Date: Tue, 28 Feb 2023 09:22:03 -0600
Subject: dt-bindings: net: realtek-bluetooth: Add RTL8821CS

Add compatible string for RTL8821CS for existing Realtek Bluetooth
driver.

Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Alistair Francis <alistair@alistair23.me>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../devicetree/bindings/net/realtek-bluetooth.yaml | 24 ++++++++++++++--------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml b/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml
index 143b5667abad..8cc2b9924680 100644
--- a/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml
@@ -4,24 +4,30 @@
 $id: http://devicetree.org/schemas/net/realtek-bluetooth.yaml#
 $schema: http://devicetree.org/meta-schemas/core.yaml#
 
-title: RTL8723BS/RTL8723CS/RTL8822CS Bluetooth
+title: RTL8723BS/RTL8723CS/RTL8821CS/RTL8822CS Bluetooth
 
 maintainers:
   - Vasily Khoruzhick <anarsoul@gmail.com>
   - Alistair Francis <alistair@alistair23.me>
 
 description:
-  RTL8723CS/RTL8723CS/RTL8822CS is WiFi + BT chip. WiFi part is connected over
-  SDIO, while BT is connected over serial. It speaks H5 protocol with few
-  extra commands to upload firmware and change module speed.
+  RTL8723CS/RTL8723CS/RTL8821CS/RTL8822CS is a WiFi + BT chip. WiFi part
+  is connected over SDIO, while BT is connected over serial. It speaks
+  H5 protocol with few extra commands to upload firmware and change
+  module speed.
 
 properties:
   compatible:
-    enum:
-      - realtek,rtl8723bs-bt
-      - realtek,rtl8723cs-bt
-      - realtek,rtl8723ds-bt
-      - realtek,rtl8822cs-bt
+    oneOf:
+      - enum:
+          - realtek,rtl8723bs-bt
+          - realtek,rtl8723cs-bt
+          - realtek,rtl8723ds-bt
+          - realtek,rtl8822cs-bt
+      - items:
+          - enum:
+              - realtek,rtl8821cs-bt
+          - const: realtek,rtl8822cs-bt
 
   device-wake-gpios:
     maxItems: 1
-- 
cgit v1.2.3-70-g09d2


From 10c9e0104e3bffa31ddc653619f993d6842955d6 Mon Sep 17 00:00:00 2001
From: Chris Morgan <macromorgan@hotmail.com>
Date: Tue, 28 Feb 2023 09:22:04 -0600
Subject: Bluetooth: hci_h5: btrtl: Add support for RTL8821CS

RTL8821CS is a WiFi + Bluetooth combo chip from Realtek that provides
WiFi A/B/G/N/AC over an SDIO interface and Bluetooth 4.2 over a UART
interface.

Note that the firmware this was tested with was firmware version
0x75b8f098.

Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index aee75f278e60..0931f27d2b83 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -128,6 +128,14 @@ static const struct id_table ic_id_table[] = {
 	  .fw_name  = "rtl_bt/rtl8821c_fw.bin",
 	  .cfg_name = "rtl_bt/rtl8821c_config" },
 
+	/* 8821CS */
+	{ IC_INFO(RTL_ROM_LMP_8821A, 0xc, 0x8, HCI_UART),
+	  .config_needed = true,
+	  .has_rom_version = true,
+	  .has_msft_ext = true,
+	  .fw_name  = "rtl_bt/rtl8821cs_fw.bin",
+	  .cfg_name = "rtl_bt/rtl8821cs_config" },
+
 	/* 8761A */
 	{ IC_INFO(RTL_ROM_LMP_8761A, 0xa, 0x6, HCI_USB),
 	  .config_needed = false,
-- 
cgit v1.2.3-70-g09d2


From b6986b7920bb50ed64e96ddd8c4da43bca289ecc Mon Sep 17 00:00:00 2001
From: Chris Morgan <macromorgan@hotmail.com>
Date: Tue, 28 Feb 2023 09:22:05 -0600
Subject: arm64: dts: rockchip: Update compatible for bluetooth

Update the compatible for the Realtek RTL8821CS bluetooth node.

Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 arch/arm64/boot/dts/rockchip/rk3566-anbernic-rgxx3.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/boot/dts/rockchip/rk3566-anbernic-rgxx3.dtsi b/arch/arm64/boot/dts/rockchip/rk3566-anbernic-rgxx3.dtsi
index 41262a69d33e..8fadd8afb190 100644
--- a/arch/arm64/boot/dts/rockchip/rk3566-anbernic-rgxx3.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3566-anbernic-rgxx3.dtsi
@@ -716,7 +716,7 @@
 	status = "okay";
 
 	bluetooth {
-		compatible = "realtek,rtl8821cs-bt";
+		compatible = "realtek,rtl8821cs-bt", "realtek,rtl8822cs-bt";
 		device-wake-gpios = <&gpio4 4 GPIO_ACTIVE_HIGH>;
 		enable-gpios = <&gpio4 3 GPIO_ACTIVE_HIGH>;
 		host-wake-gpios = <&gpio4 5 GPIO_ACTIVE_HIGH>;
-- 
cgit v1.2.3-70-g09d2


From 125ede3afe5830c3e748dbc66098b7b640899321 Mon Sep 17 00:00:00 2001
From: Meng Tang <tangmeng@uniontech.com>
Date: Mon, 6 Mar 2023 09:44:59 +0800
Subject: Bluetooth: Add VID/PID 0489/e0e4 for MediaTek MT7922

Tested on Acer Nitro AN515-47 Notebook

output from /sys/kernel/debug/usb/devices:

T:  Bus=03 Lev=01 Prnt=01 Port=02 Cnt=02 Dev#=  2 Spd=480  MxCh= 0
D:  Ver= 2.10 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=0489 ProdID=e0e4 Rev= 1.00
S:  Manufacturer=MediaTek Inc.
S:  Product=Wireless_Device
S:  SerialNumber=000000000
C:* #Ifs= 3 Cfg#= 1 Atr=e0 MxPwr=100mA
A:  FirstIf#= 0 IfCount= 3 Cls=e0(wlcon) Sub=01 Prot=01
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=125us
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:  If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=83(I) Atr=01(Isoc) MxPS=  63 Ivl=1ms
E:  Ad=03(O) Atr=01(Isoc) MxPS=  63 Ivl=1ms
I:* If#= 2 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none)
E:  Ad=8a(I) Atr=03(Int.) MxPS=  64 Ivl=125us
E:  Ad=0a(O) Atr=03(Int.) MxPS=  64 Ivl=125us
I:  If#= 2 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none)
E:  Ad=8a(I) Atr=03(Int.) MxPS= 512 Ivl=125us
E:  Ad=0a(O) Atr=03(Int.) MxPS= 512 Ivl=125us

Signed-off-by: Meng Tang <tangmeng@uniontech.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 372007f78467..0c94cd7f4af3 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -615,6 +615,9 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x0489, 0xe0e2), .driver_info = BTUSB_MEDIATEK |
 						     BTUSB_WIDEBAND_SPEECH |
 						     BTUSB_VALID_LE_STATES },
+	{ USB_DEVICE(0x0489, 0xe0e4), .driver_info = BTUSB_MEDIATEK |
+						     BTUSB_WIDEBAND_SPEECH |
+						     BTUSB_VALID_LE_STATES },
 	{ USB_DEVICE(0x0489, 0xe0f2), .driver_info = BTUSB_MEDIATEK |
 						     BTUSB_WIDEBAND_SPEECH |
 						     BTUSB_VALID_LE_STATES },
-- 
cgit v1.2.3-70-g09d2


From f5d13029a41d6504831ab99da0a9a0332fa7ea19 Mon Sep 17 00:00:00 2001
From: Archie Pusaka <apusaka@chromium.org>
Date: Mon, 6 Mar 2023 17:07:07 +0800
Subject: Bluetooth: hci_sync: Don't wait peer's reply when powering off

Currently, when we initiate disconnection, we will wait for the peer's
reply unless when we are suspending, where we fire and forget the
disconnect request.

A similar case is when adapter is powering off. However, we still wait
for the peer's reply in this case. Therefore, if the peer is
unresponsive, the command will time out and the power off sequence
will fail, causing "bluetooth powered on by itself" to users.

This patch makes the host doesn't wait for the peer's reply when the
disconnection reason is powering off.

Signed-off-by: Archie Pusaka <apusaka@chromium.org>
Reviewed-by: Abhishek Pandit-Subedi <abhishekpandit@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_sync.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index b64af017ab99..eb5fd2fe1dc2 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -5108,10 +5108,12 @@ static int hci_disconnect_sync(struct hci_dev *hdev, struct hci_conn *conn,
 	cp.handle = cpu_to_le16(conn->handle);
 	cp.reason = reason;
 
-	/* Wait for HCI_EV_DISCONN_COMPLETE not HCI_EV_CMD_STATUS when not
-	 * suspending.
+	/* Wait for HCI_EV_DISCONN_COMPLETE, not HCI_EV_CMD_STATUS, when the
+	 * reason is anything but HCI_ERROR_REMOTE_POWER_OFF. This reason is
+	 * used when suspending or powering off, where we don't want to wait
+	 * for the peer's response.
 	 */
-	if (!hdev->suspended)
+	if (reason != HCI_ERROR_REMOTE_POWER_OFF)
 		return __hci_cmd_sync_status_sk(hdev, HCI_OP_DISCONNECT,
 						sizeof(cp), &cp,
 						HCI_EV_DISCONN_COMPLETE,
-- 
cgit v1.2.3-70-g09d2


From 493ea69970e2a4e4102fdf0be6d262456f564c3a Mon Sep 17 00:00:00 2001
From: Brian Gix <brian.gix@intel.com>
Date: Fri, 24 Feb 2023 09:56:32 -0800
Subject: Bluetooth: Convert MSFT filter HCI cmd to hci_sync

The msft_set_filter_enable() command was using the deprecated
hci_request mechanism rather than hci_sync. This caused the warning error:
hci0: HCI_REQ-0xfcf0

Signed-off-by: Brian Gix <brian.gix@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/msft.c | 36 +++++++++++-------------------------
 1 file changed, 11 insertions(+), 25 deletions(-)

diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c
index bee6a4c656be..bf5cee48916c 100644
--- a/net/bluetooth/msft.c
+++ b/net/bluetooth/msft.c
@@ -743,17 +743,12 @@ __u64 msft_get_features(struct hci_dev *hdev)
 }
 
 static void msft_le_set_advertisement_filter_enable_cb(struct hci_dev *hdev,
-						       u8 status, u16 opcode,
-						       struct sk_buff *skb)
+						       void *user_data,
+						       u8 status)
 {
-	struct msft_cp_le_set_advertisement_filter_enable *cp;
-	struct msft_rp_le_set_advertisement_filter_enable *rp;
+	struct msft_cp_le_set_advertisement_filter_enable *cp = user_data;
 	struct msft_data *msft = hdev->msft_data;
 
-	rp = (struct msft_rp_le_set_advertisement_filter_enable *)skb->data;
-	if (skb->len < sizeof(*rp))
-		return;
-
 	/* Error 0x0C would be returned if the filter enabled status is
 	 * already set to whatever we were trying to set.
 	 * Although the default state should be disabled, some controller set
@@ -766,7 +761,6 @@ static void msft_le_set_advertisement_filter_enable_cb(struct hci_dev *hdev,
 
 	hci_dev_lock(hdev);
 
-	cp = hci_sent_cmd_data(hdev, hdev->msft_opcode);
 	msft->filter_enabled = cp->enable;
 
 	if (status == 0x0C)
@@ -804,31 +798,23 @@ int msft_remove_monitor(struct hci_dev *hdev, struct adv_monitor *monitor)
 	return msft_remove_monitor_sync(hdev, monitor);
 }
 
-void msft_req_add_set_filter_enable(struct hci_request *req, bool enable)
-{
-	struct hci_dev *hdev = req->hdev;
-	struct msft_cp_le_set_advertisement_filter_enable cp;
-
-	cp.sub_opcode = MSFT_OP_LE_SET_ADVERTISEMENT_FILTER_ENABLE;
-	cp.enable = enable;
-
-	hci_req_add(req, hdev->msft_opcode, sizeof(cp), &cp);
-}
-
 int msft_set_filter_enable(struct hci_dev *hdev, bool enable)
 {
-	struct hci_request req;
+	struct msft_cp_le_set_advertisement_filter_enable cp;
 	struct msft_data *msft = hdev->msft_data;
 	int err;
 
 	if (!msft)
 		return -EOPNOTSUPP;
 
-	hci_req_init(&req, hdev);
-	msft_req_add_set_filter_enable(&req, enable);
-	err = hci_req_run_skb(&req, msft_le_set_advertisement_filter_enable_cb);
+	cp.sub_opcode = MSFT_OP_LE_SET_ADVERTISEMENT_FILTER_ENABLE;
+	cp.enable = enable;
+	err = __hci_cmd_sync_status(hdev, hdev->msft_opcode, sizeof(cp), &cp,
+				    HCI_CMD_TIMEOUT);
+
+	msft_le_set_advertisement_filter_enable_cb(hdev, &cp, err);
 
-	return err;
+	return 0;
 }
 
 bool msft_curve_validity(struct hci_dev *hdev)
-- 
cgit v1.2.3-70-g09d2


From 9f5ae8fcae1227e9bae3d973fcbef8319a1b25c7 Mon Sep 17 00:00:00 2001
From: Inga Stotland <inga.stotland@intel.com>
Date: Wed, 8 Mar 2023 10:32:35 -0800
Subject: Bluetooth: hci_sync: Remove duplicate statement

This removes the following duplicate statement in
hci_le_ext_directed_advertising_sync():
cp.own_addr_type = own_addr_type;

Signed-off-by: Inga Stotland <inga.stotland@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_sync.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index eb5fd2fe1dc2..31231f0e4a28 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -5855,7 +5855,6 @@ static int hci_le_ext_directed_advertising_sync(struct hci_dev *hdev,
 	memset(&cp, 0, sizeof(cp));
 
 	cp.evt_properties = cpu_to_le16(LE_LEGACY_ADV_DIRECT_IND);
-	cp.own_addr_type = own_addr_type;
 	cp.channel_map = hdev->le_adv_channel_map;
 	cp.tx_power = HCI_TX_POWER_INVALID;
 	cp.primary_phy = HCI_ADV_PHY_1M;
-- 
cgit v1.2.3-70-g09d2


From b8b23001b8025a61f0979578884a74faa825023e Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 8 Mar 2023 16:16:31 -0800
Subject: Bluetooth: L2CAP: Delay identity address updates

This delays the identity address updates to give time for userspace to
process the new address otherwise there is a risk that userspace
creates a duplicated device if the MGMT event is delayed for some
reason.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/l2cap.h | 2 +-
 net/bluetooth/l2cap_core.c    | 7 +++----
 net/bluetooth/smp.c           | 9 ++++++++-
 3 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h
index 2f766e3437ce..cf393e72d6ed 100644
--- a/include/net/bluetooth/l2cap.h
+++ b/include/net/bluetooth/l2cap.h
@@ -694,7 +694,7 @@ struct l2cap_conn {
 	struct sk_buff_head	pending_rx;
 	struct work_struct	pending_rx_work;
 
-	struct work_struct	id_addr_update_work;
+	struct delayed_work	id_addr_timer;
 
 	__u8			disc_reason;
 
diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 55a7226233f9..5cc95fd17f7d 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -745,7 +745,7 @@ EXPORT_SYMBOL_GPL(l2cap_chan_list);
 static void l2cap_conn_update_id_addr(struct work_struct *work)
 {
 	struct l2cap_conn *conn = container_of(work, struct l2cap_conn,
-					       id_addr_update_work);
+					       id_addr_timer.work);
 	struct hci_conn *hcon = conn->hcon;
 	struct l2cap_chan *chan;
 
@@ -1907,8 +1907,7 @@ static void l2cap_conn_del(struct hci_conn *hcon, int err)
 	if (work_pending(&conn->pending_rx_work))
 		cancel_work_sync(&conn->pending_rx_work);
 
-	if (work_pending(&conn->id_addr_update_work))
-		cancel_work_sync(&conn->id_addr_update_work);
+	cancel_delayed_work_sync(&conn->id_addr_timer);
 
 	l2cap_unregister_all_users(conn);
 
@@ -7874,7 +7873,7 @@ static struct l2cap_conn *l2cap_conn_add(struct hci_conn *hcon)
 
 	skb_queue_head_init(&conn->pending_rx);
 	INIT_WORK(&conn->pending_rx_work, process_pending_rx);
-	INIT_WORK(&conn->id_addr_update_work, l2cap_conn_update_id_addr);
+	INIT_DELAYED_WORK(&conn->id_addr_timer, l2cap_conn_update_id_addr);
 
 	conn->disc_reason = HCI_ERROR_REMOTE_USER_TERM;
 
diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c
index 70663229b3cc..f1a9fc0012f0 100644
--- a/net/bluetooth/smp.c
+++ b/net/bluetooth/smp.c
@@ -58,6 +58,8 @@
 
 #define SMP_TIMEOUT	msecs_to_jiffies(30000)
 
+#define ID_ADDR_TIMEOUT	msecs_to_jiffies(200)
+
 #define AUTH_REQ_MASK(dev)	(hci_dev_test_flag(dev, HCI_SC_ENABLED) ? \
 				 0x3f : 0x07)
 #define KEY_DIST_MASK		0x07
@@ -1067,7 +1069,12 @@ static void smp_notify_keys(struct l2cap_conn *conn)
 		if (hcon->type == LE_LINK) {
 			bacpy(&hcon->dst, &smp->remote_irk->bdaddr);
 			hcon->dst_type = smp->remote_irk->addr_type;
-			queue_work(hdev->workqueue, &conn->id_addr_update_work);
+			/* Use a short delay to make sure the new address is
+			 * propagated _before_ the channels.
+			 */
+			queue_delayed_work(hdev->workqueue,
+					   &conn->id_addr_timer,
+					   ID_ADDR_TIMEOUT);
 		}
 	}
 
-- 
cgit v1.2.3-70-g09d2


From cd97d47f3069ef578f3b3ddb8c7ae3d6ce38b61a Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 12:13:51 +0100
Subject: Bluetooth: hci_ll: drop of_match_ptr for ID table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it might not be relevant here).

  drivers/bluetooth/hci_ll.c:769:34: error: ‘hci_ti_of_match’ defined
  but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_ll.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/hci_ll.c b/drivers/bluetooth/hci_ll.c
index 5abc01a2acf7..4a0b5c3160c2 100644
--- a/drivers/bluetooth/hci_ll.c
+++ b/drivers/bluetooth/hci_ll.c
@@ -786,7 +786,7 @@ MODULE_DEVICE_TABLE(of, hci_ti_of_match);
 static struct serdev_device_driver hci_ti_drv = {
 	.driver		= {
 		.name	= "hci-ti",
-		.of_match_table = of_match_ptr(hci_ti_of_match),
+		.of_match_table = hci_ti_of_match,
 	},
 	.probe	= hci_ti_probe,
 	.remove	= hci_ti_remove,
-- 
cgit v1.2.3-70-g09d2


From 04909d7bf847bc0b939a5571a1a1fcbe5f2e2ead Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 12:13:52 +0100
Subject: Bluetooth: btmrvl_sdio: mark OF related data as maybe unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can be compile tested with !CONFIG_OF making certain data
unused:

  drivers/bluetooth/btmrvl_sdio.c:43:34: error:
  ‘btmrvl_sdio_of_match_table’ defined but not used
  [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmrvl_sdio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c
index ba057ebfda5c..d76c799553aa 100644
--- a/drivers/bluetooth/btmrvl_sdio.c
+++ b/drivers/bluetooth/btmrvl_sdio.c
@@ -40,7 +40,7 @@ static struct memory_type_mapping mem_type_mapping_tbl[] = {
 	{"EXTLAST", NULL, 0, 0xFE},
 };
 
-static const struct of_device_id btmrvl_sdio_of_match_table[] = {
+static const struct of_device_id btmrvl_sdio_of_match_table[] __maybe_unused = {
 	{ .compatible = "marvell,sd8897-bt" },
 	{ .compatible = "marvell,sd8997-bt" },
 	{ }
-- 
cgit v1.2.3-70-g09d2


From 44fac8a2fd2f72ee98ee41e6bc9ecc7765b5d3cc Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 12:13:53 +0100
Subject: Bluetooth: hci_qca: mark OF related data as maybe unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can be compile tested with !CONFIG_OF making certain data
unused:

  drivers/bluetooth/hci_qca.c:1869:37: error: ‘qca_soc_data_wcn6750’
  defined but not used [-Werror=unused-const-variable=]
  drivers/bluetooth/hci_qca.c:1853:37: error: ‘qca_soc_data_wcn3998’
  defined but not used [-Werror=unused-const-variable=]
  drivers/bluetooth/hci_qca.c:1841:37: error: ‘qca_soc_data_wcn3991’
  defined but not used [-Werror=unused-const-variable=]
  drivers/bluetooth/hci_qca.c:1830:37: error: ‘qca_soc_data_wcn3990’
  defined but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 3df8c3606e93..a199ed52d76d 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -1827,7 +1827,7 @@ static const struct hci_uart_proto qca_proto = {
 	.dequeue	= qca_dequeue,
 };
 
-static const struct qca_device_data qca_soc_data_wcn3990 = {
+static const struct qca_device_data qca_soc_data_wcn3990 __maybe_unused = {
 	.soc_type = QCA_WCN3990,
 	.vregs = (struct qca_vreg []) {
 		{ "vddio", 15000  },
@@ -1838,7 +1838,7 @@ static const struct qca_device_data qca_soc_data_wcn3990 = {
 	.num_vregs = 4,
 };
 
-static const struct qca_device_data qca_soc_data_wcn3991 = {
+static const struct qca_device_data qca_soc_data_wcn3991 __maybe_unused = {
 	.soc_type = QCA_WCN3991,
 	.vregs = (struct qca_vreg []) {
 		{ "vddio", 15000  },
@@ -1850,7 +1850,7 @@ static const struct qca_device_data qca_soc_data_wcn3991 = {
 	.capabilities = QCA_CAP_WIDEBAND_SPEECH | QCA_CAP_VALID_LE_STATES,
 };
 
-static const struct qca_device_data qca_soc_data_wcn3998 = {
+static const struct qca_device_data qca_soc_data_wcn3998 __maybe_unused = {
 	.soc_type = QCA_WCN3998,
 	.vregs = (struct qca_vreg []) {
 		{ "vddio", 10000  },
@@ -1861,12 +1861,12 @@ static const struct qca_device_data qca_soc_data_wcn3998 = {
 	.num_vregs = 4,
 };
 
-static const struct qca_device_data qca_soc_data_qca6390 = {
+static const struct qca_device_data qca_soc_data_qca6390 __maybe_unused = {
 	.soc_type = QCA_QCA6390,
 	.num_vregs = 0,
 };
 
-static const struct qca_device_data qca_soc_data_wcn6750 = {
+static const struct qca_device_data qca_soc_data_wcn6750 __maybe_unused = {
 	.soc_type = QCA_WCN6750,
 	.vregs = (struct qca_vreg []) {
 		{ "vddio", 5000 },
-- 
cgit v1.2.3-70-g09d2


From a88ea43b73f109e446f6027df237c8f868127a78 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sat, 11 Mar 2023 12:13:54 +0100
Subject: Bluetooth: btmtkuart: mark OF related data as maybe unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver can be compile tested with !CONFIG_OF making certain data
unused:

  drivers/bluetooth/btmtkuart.c:971:36: error: ‘mt7668_data’ defined
  but not used [-Werror=unused-const-variable=]
  drivers/bluetooth/btmtkuart.c:966:36: error: ‘mt7663_data’ defined
  but not used [-Werror=unused-const-variable=]
  drivers/bluetooth/btmtkuart.c:962:36: error: ‘mt7622_data’ defined
  but not used [-Werror=unused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtkuart.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/bluetooth/btmtkuart.c b/drivers/bluetooth/btmtkuart.c
index c98691cdbbd5..7680c67cdb35 100644
--- a/drivers/bluetooth/btmtkuart.c
+++ b/drivers/bluetooth/btmtkuart.c
@@ -959,16 +959,16 @@ static void btmtkuart_remove(struct serdev_device *serdev)
 	hci_free_dev(hdev);
 }
 
-static const struct btmtkuart_data mt7622_data = {
+static const struct btmtkuart_data mt7622_data __maybe_unused = {
 	.fwname = FIRMWARE_MT7622,
 };
 
-static const struct btmtkuart_data mt7663_data = {
+static const struct btmtkuart_data mt7663_data __maybe_unused = {
 	.flags = BTMTKUART_FLAG_STANDALONE_HW,
 	.fwname = FIRMWARE_MT7663,
 };
 
-static const struct btmtkuart_data mt7668_data = {
+static const struct btmtkuart_data mt7668_data __maybe_unused = {
 	.flags = BTMTKUART_FLAG_STANDALONE_HW,
 	.fwname = FIRMWARE_MT7668,
 };
-- 
cgit v1.2.3-70-g09d2


From 8194f1ef5a815aea815a91daf2c721eab2674f1f Mon Sep 17 00:00:00 2001
From: Vasily Khoruzhick <anarsoul@gmail.com>
Date: Tue, 7 Mar 2023 23:17:30 +0100
Subject: Bluetooth: Add new quirk for broken local ext features page 2

Some adapters (e.g. RTL8723CS) advertise that they have more than
2 pages for local ext features, but they don't support any features
declared in these pages. RTL8723CS reports max_page = 2 and declares
support for sync train and secure connection, but it responds with
either garbage or with error in status on corresponding commands.

Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Bastian Germann <bage@debian.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci.h | 7 +++++++
 net/bluetooth/hci_event.c   | 9 +++++++--
 2 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 400f8a7d0c3f..997107bfc0b1 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -294,6 +294,13 @@ enum {
 	 * during the hdev->setup vendor callback.
 	 */
 	HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG,
+
+	/* When this quirk is set, max_page for local extended features
+	 * is set to 1, even if controller reports higher number. Some
+	 * controllers (e.g. RTL8723CS) report more pages, but they
+	 * don't actually support features declared there.
+	 */
+	HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2,
 };
 
 /* HCI device flags */
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index e87c928c9e17..51f13518dba9 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -886,8 +886,13 @@ static u8 hci_cc_read_local_ext_features(struct hci_dev *hdev, void *data,
 	if (rp->status)
 		return rp->status;
 
-	if (hdev->max_page < rp->max_page)
-		hdev->max_page = rp->max_page;
+	if (hdev->max_page < rp->max_page) {
+		if (test_bit(HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2,
+			     &hdev->quirks))
+			bt_dev_warn(hdev, "broken local ext features page 2");
+		else
+			hdev->max_page = rp->max_page;
+	}
 
 	if (rp->page < HCI_MAX_PAGES)
 		memcpy(hdev->features[rp->page], rp->features, 8);
-- 
cgit v1.2.3-70-g09d2


From c0123cb6c4c7fc2a42ead6cd7d3e82b8e1c25c6f Mon Sep 17 00:00:00 2001
From: Vasily Khoruzhick <anarsoul@gmail.com>
Date: Tue, 7 Mar 2023 23:17:31 +0100
Subject: Bluetooth: btrtl: add support for the RTL8723CS

The Realtek RTL8723CS is a SDIO WiFi chip. It also contains a Bluetooth
module which is connected via UART to the host.

It shares lmp subversion with 8703B, so Realtek's userspace
initialization tool (rtk_hciattach) differentiates varieties of RTL8723CS
(CG, VF, XX) with RTL8703B using vendor's command to read chip type.

Also this chip declares support for some features it doesn't support
so add a quirk to indicate that these features are broken.

Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Bastian Germann <bage@debian.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c  | 120 +++++++++++++++++++++++++++++++++++++++++++--
 drivers/bluetooth/btrtl.h  |   5 ++
 drivers/bluetooth/hci_h5.c |   4 ++
 3 files changed, 125 insertions(+), 4 deletions(-)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index 0931f27d2b83..8fa7733812ad 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -17,7 +17,11 @@
 
 #define VERSION "0.1"
 
+#define RTL_CHIP_8723CS_CG	3
+#define RTL_CHIP_8723CS_VF	4
+#define RTL_CHIP_8723CS_XX	5
 #define RTL_EPATCH_SIGNATURE	"Realtech"
+#define RTL_ROM_LMP_8703B	0x8703
 #define RTL_ROM_LMP_8723A	0x1200
 #define RTL_ROM_LMP_8723B	0x8723
 #define RTL_ROM_LMP_8821A	0x8821
@@ -30,6 +34,7 @@
 #define IC_MATCH_FL_HCIREV	(1 << 1)
 #define IC_MATCH_FL_HCIVER	(1 << 2)
 #define IC_MATCH_FL_HCIBUS	(1 << 3)
+#define IC_MATCH_FL_CHIP_TYPE	(1 << 4)
 #define IC_INFO(lmps, hcir, hciv, bus) \
 	.match_flags = IC_MATCH_FL_LMPSUBV | IC_MATCH_FL_HCIREV | \
 		       IC_MATCH_FL_HCIVER | IC_MATCH_FL_HCIBUS, \
@@ -59,6 +64,7 @@ struct id_table {
 	__u16 hci_rev;
 	__u8 hci_ver;
 	__u8 hci_bus;
+	__u8 chip_type;
 	bool config_needed;
 	bool has_rom_version;
 	bool has_msft_ext;
@@ -99,6 +105,39 @@ static const struct id_table ic_id_table[] = {
 	  .fw_name  = "rtl_bt/rtl8723b_fw.bin",
 	  .cfg_name = "rtl_bt/rtl8723b_config" },
 
+	/* 8723CS-CG */
+	{ .match_flags = IC_MATCH_FL_LMPSUBV | IC_MATCH_FL_CHIP_TYPE |
+			 IC_MATCH_FL_HCIBUS,
+	  .lmp_subver = RTL_ROM_LMP_8703B,
+	  .chip_type = RTL_CHIP_8723CS_CG,
+	  .hci_bus = HCI_UART,
+	  .config_needed = true,
+	  .has_rom_version = true,
+	  .fw_name  = "rtl_bt/rtl8723cs_cg_fw.bin",
+	  .cfg_name = "rtl_bt/rtl8723cs_cg_config" },
+
+	/* 8723CS-VF */
+	{ .match_flags = IC_MATCH_FL_LMPSUBV | IC_MATCH_FL_CHIP_TYPE |
+			 IC_MATCH_FL_HCIBUS,
+	  .lmp_subver = RTL_ROM_LMP_8703B,
+	  .chip_type = RTL_CHIP_8723CS_VF,
+	  .hci_bus = HCI_UART,
+	  .config_needed = true,
+	  .has_rom_version = true,
+	  .fw_name  = "rtl_bt/rtl8723cs_vf_fw.bin",
+	  .cfg_name = "rtl_bt/rtl8723cs_vf_config" },
+
+	/* 8723CS-XX */
+	{ .match_flags = IC_MATCH_FL_LMPSUBV | IC_MATCH_FL_CHIP_TYPE |
+			 IC_MATCH_FL_HCIBUS,
+	  .lmp_subver = RTL_ROM_LMP_8703B,
+	  .chip_type = RTL_CHIP_8723CS_XX,
+	  .hci_bus = HCI_UART,
+	  .config_needed = true,
+	  .has_rom_version = true,
+	  .fw_name  = "rtl_bt/rtl8723cs_xx_fw.bin",
+	  .cfg_name = "rtl_bt/rtl8723cs_xx_config" },
+
 	/* 8723D */
 	{ IC_INFO(RTL_ROM_LMP_8723B, 0xd, 0x8, HCI_USB),
 	  .config_needed = true,
@@ -224,7 +263,8 @@ static const struct id_table ic_id_table[] = {
 	};
 
 static const struct id_table *btrtl_match_ic(u16 lmp_subver, u16 hci_rev,
-					     u8 hci_ver, u8 hci_bus)
+					     u8 hci_ver, u8 hci_bus,
+					     u8 chip_type)
 {
 	int i;
 
@@ -241,6 +281,9 @@ static const struct id_table *btrtl_match_ic(u16 lmp_subver, u16 hci_rev,
 		if ((ic_id_table[i].match_flags & IC_MATCH_FL_HCIBUS) &&
 		    (ic_id_table[i].hci_bus != hci_bus))
 			continue;
+		if ((ic_id_table[i].match_flags & IC_MATCH_FL_CHIP_TYPE) &&
+		    (ic_id_table[i].chip_type != chip_type))
+			continue;
 
 		break;
 	}
@@ -323,6 +366,7 @@ static int rtlbt_parse_firmware(struct hci_dev *hdev,
 		{ RTL_ROM_LMP_8723B, 1 },
 		{ RTL_ROM_LMP_8821A, 2 },
 		{ RTL_ROM_LMP_8761A, 3 },
+		{ RTL_ROM_LMP_8703B, 7 },
 		{ RTL_ROM_LMP_8822B, 8 },
 		{ RTL_ROM_LMP_8723B, 9 },	/* 8723D */
 		{ RTL_ROM_LMP_8821A, 10 },	/* 8821C */
@@ -603,6 +647,48 @@ out:
 	return ret;
 }
 
+static bool rtl_has_chip_type(u16 lmp_subver)
+{
+	switch (lmp_subver) {
+	case RTL_ROM_LMP_8703B:
+		return true;
+	default:
+		break;
+	}
+
+	return  false;
+}
+
+static int rtl_read_chip_type(struct hci_dev *hdev, u8 *type)
+{
+	struct rtl_chip_type_evt *chip_type;
+	struct sk_buff *skb;
+	const unsigned char cmd_buf[] = {0x00, 0x94, 0xa0, 0x00, 0xb0};
+
+	/* Read RTL chip type command */
+	skb = __hci_cmd_sync(hdev, 0xfc61, 5, cmd_buf, HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		rtl_dev_err(hdev, "Read chip type failed (%ld)",
+			    PTR_ERR(skb));
+		return PTR_ERR(skb);
+	}
+
+	chip_type = skb_pull_data(skb, sizeof(*chip_type));
+	if (!chip_type) {
+		rtl_dev_err(hdev, "RTL chip type event length mismatch");
+		kfree_skb(skb);
+		return -EIO;
+	}
+
+	rtl_dev_info(hdev, "chip_type status=%x type=%x",
+		     chip_type->status, chip_type->type);
+
+	*type = chip_type->type & 0x0f;
+
+	kfree_skb(skb);
+	return 0;
+}
+
 void btrtl_free(struct btrtl_device_info *btrtl_dev)
 {
 	kvfree(btrtl_dev->fw_data);
@@ -619,7 +705,7 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 	struct hci_rp_read_local_version *resp;
 	char cfg_name[40];
 	u16 hci_rev, lmp_subver;
-	u8 hci_ver;
+	u8 hci_ver, chip_type = 0;
 	int ret;
 	u16 opcode;
 	u8 cmd[2];
@@ -645,8 +731,14 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 	hci_rev = le16_to_cpu(resp->hci_rev);
 	lmp_subver = le16_to_cpu(resp->lmp_subver);
 
+	if (rtl_has_chip_type(lmp_subver)) {
+		ret = rtl_read_chip_type(hdev, &chip_type);
+		if (ret)
+			goto err_free;
+	}
+
 	btrtl_dev->ic_info = btrtl_match_ic(lmp_subver, hci_rev, hci_ver,
-					    hdev->bus);
+					    hdev->bus, chip_type);
 
 	if (!btrtl_dev->ic_info)
 		btrtl_dev->drop_fw = true;
@@ -689,7 +781,7 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 		lmp_subver = le16_to_cpu(resp->lmp_subver);
 
 		btrtl_dev->ic_info = btrtl_match_ic(lmp_subver, hci_rev, hci_ver,
-						    hdev->bus);
+						    hdev->bus, chip_type);
 	}
 out_free:
 	kfree_skb(skb);
@@ -771,6 +863,7 @@ int btrtl_download_firmware(struct hci_dev *hdev,
 	case RTL_ROM_LMP_8761A:
 	case RTL_ROM_LMP_8822B:
 	case RTL_ROM_LMP_8852A:
+	case RTL_ROM_LMP_8703B:
 		return btrtl_setup_rtl8723b(hdev, btrtl_dev);
 	default:
 		rtl_dev_info(hdev, "assuming no firmware upload needed");
@@ -811,6 +904,19 @@ void btrtl_set_quirks(struct hci_dev *hdev, struct btrtl_device_info *btrtl_dev)
 		rtl_dev_dbg(hdev, "WBS supported not enabled.");
 		break;
 	}
+
+	switch (btrtl_dev->ic_info->lmp_subver) {
+	case RTL_ROM_LMP_8703B:
+		/* 8723CS reports two pages for local ext features,
+		 * but it doesn't support any features from page 2 -
+		 * it either responds with garbage or with error status
+		 */
+		set_bit(HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2,
+			&hdev->quirks);
+		break;
+	default:
+		break;
+	}
 }
 EXPORT_SYMBOL_GPL(btrtl_set_quirks);
 
@@ -969,6 +1075,12 @@ MODULE_FIRMWARE("rtl_bt/rtl8723b_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8723b_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8723bs_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8723bs_config.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8723cs_cg_fw.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8723cs_cg_config.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8723cs_vf_fw.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8723cs_vf_config.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8723cs_xx_fw.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8723cs_xx_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8723ds_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8723ds_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8761a_fw.bin");
diff --git a/drivers/bluetooth/btrtl.h b/drivers/bluetooth/btrtl.h
index ebf0101c959b..349d72ee571b 100644
--- a/drivers/bluetooth/btrtl.h
+++ b/drivers/bluetooth/btrtl.h
@@ -14,6 +14,11 @@
 
 struct btrtl_device_info;
 
+struct rtl_chip_type_evt {
+	__u8 status;
+	__u8 type;
+} __packed;
+
 struct rtl_download_cmd {
 	__u8 index;
 	__u8 data[RTL_FRAG_LEN];
diff --git a/drivers/bluetooth/hci_h5.c b/drivers/bluetooth/hci_h5.c
index 6455bc4fb5bb..e90670955df2 100644
--- a/drivers/bluetooth/hci_h5.c
+++ b/drivers/bluetooth/hci_h5.c
@@ -936,6 +936,8 @@ static int h5_btrtl_setup(struct h5 *h5)
 	err = btrtl_download_firmware(h5->hu->hdev, btrtl_dev);
 	/* Give the device some time before the hci-core sends it a reset */
 	usleep_range(10000, 20000);
+	if (err)
+		goto out_free;
 
 	btrtl_set_quirks(h5->hu->hdev, btrtl_dev);
 
@@ -1100,6 +1102,8 @@ static const struct of_device_id rtl_bluetooth_of_match[] = {
 	  .data = (const void *)&h5_data_rtl8822cs },
 	{ .compatible = "realtek,rtl8723bs-bt",
 	  .data = (const void *)&h5_data_rtl8723bs },
+	{ .compatible = "realtek,rtl8723cs-bt",
+	  .data = (const void *)&h5_data_rtl8723bs },
 	{ .compatible = "realtek,rtl8723ds-bt",
 	  .data = (const void *)&h5_data_rtl8723bs },
 #endif
-- 
cgit v1.2.3-70-g09d2


From 7c2b2d2d0cb658aa543e11e90ae95621d3cb5fe6 Mon Sep 17 00:00:00 2001
From: Raul Cheleguini <rcheleguini@google.com>
Date: Fri, 10 Mar 2023 15:14:10 +0000
Subject: Bluetooth: Improve support for Actions Semi ATS2851 based devices

Add two more quirks to resume the device initialization and basic
operation as the device seems not to support "Read Transmit Power"
and "Set Extended Scan Parameters".

< HCI Command: LE Read Transmit Power (0x08|0x004b) plen 0
> HCI Event: Command Status (0x0f) plen 4
      LE Read Transmit Power (0x08|0x004b) ncmd 1
        Status: Unknown HCI Command (0x01)

< HCI Command: LE Set Extended Scan Parameters (0x08|0x0041) plen 8
        Own address type: Random (0x01)
        Filter policy: Accept all advertisement (0x00)
        PHYs: 0x01
        Entry 0: LE 1M
          Type: Active (0x01)
          Interval: 11.250 msec (0x0012)
          Window: 11.250 msec (0x0012)
> HCI Event: Command Status (0x0f) plen 4
      LE Set Extended Scan Parameters (0x08|0x0041) ncmd 1
        Status: Unknown HCI Command (0x01)

Signed-off-by: Raul Cheleguini <rcheleguini@google.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 0c94cd7f4af3..7382b021f3df 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -4104,6 +4104,8 @@ static int btusb_probe(struct usb_interface *intf,
 	if (id->driver_info & BTUSB_ACTIONS_SEMI) {
 		/* Support is advertised, but not implemented */
 		set_bit(HCI_QUIRK_BROKEN_ERR_DATA_REPORTING, &hdev->quirks);
+		set_bit(HCI_QUIRK_BROKEN_READ_TRANSMIT_POWER, &hdev->quirks);
+		set_bit(HCI_QUIRK_BROKEN_EXT_SCAN, &hdev->quirks);
 	}
 
 	if (!reset)
-- 
cgit v1.2.3-70-g09d2


From 586e154f98ac826233c2be5dc80e101303b4b2bd Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Date: Thu, 9 Mar 2023 10:20:30 +0200
Subject: Bluetooth: hci_ldisc: Fix tty_set_termios() return value assumptions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

tty_set_termios() never returns anything else than 0. Make the debug
prints to look directly into the new termios instead to check CRTSCTS
state.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_ldisc.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c
index 865112e96ff9..efdda2c3fce8 100644
--- a/drivers/bluetooth/hci_ldisc.c
+++ b/drivers/bluetooth/hci_ldisc.c
@@ -323,9 +323,9 @@ void hci_uart_set_flow_control(struct hci_uart *hu, bool enable)
 		/* Disable hardware flow control */
 		ktermios = tty->termios;
 		ktermios.c_cflag &= ~CRTSCTS;
-		status = tty_set_termios(tty, &ktermios);
+		tty_set_termios(tty, &ktermios);
 		BT_DBG("Disabling hardware flow control: %s",
-		       status ? "failed" : "success");
+		       (tty->termios.c_cflag & CRTSCTS) ? "failed" : "success");
 
 		/* Clear RTS to prevent the device from sending */
 		/* Most UARTs need OUT2 to enable interrupts */
@@ -357,9 +357,9 @@ void hci_uart_set_flow_control(struct hci_uart *hu, bool enable)
 		/* Re-enable hardware flow control */
 		ktermios = tty->termios;
 		ktermios.c_cflag |= CRTSCTS;
-		status = tty_set_termios(tty, &ktermios);
+		tty_set_termios(tty, &ktermios);
 		BT_DBG("Enabling hardware flow control: %s",
-		       status ? "failed" : "success");
+		       !(tty->termios.c_cflag & CRTSCTS) ? "failed" : "success");
 	}
 }
 
-- 
cgit v1.2.3-70-g09d2


From 29f93a687f3c435e94d026ee1fc8ad18ce56b7fb Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Thu, 16 Mar 2023 22:52:11 +0530
Subject: serdev: Replace all instances of ENOTSUPP with EOPNOTSUPP

This replaces all instances of ENOTSUPP with EOPNOTSUPP since ENOTSUPP
is not a standard error code. This will help maintain consistency in
error codes when new serdev API's are added.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/tty/serdev/core.c           | 6 +++---
 drivers/tty/serdev/serdev-ttyport.c | 4 ++--
 include/linux/serdev.h              | 4 ++--
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/tty/serdev/core.c b/drivers/tty/serdev/core.c
index 678014253b7b..078fe64625cb 100644
--- a/drivers/tty/serdev/core.c
+++ b/drivers/tty/serdev/core.c
@@ -366,7 +366,7 @@ int serdev_device_set_parity(struct serdev_device *serdev,
 	struct serdev_controller *ctrl = serdev->ctrl;
 
 	if (!ctrl || !ctrl->ops->set_parity)
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	return ctrl->ops->set_parity(ctrl, parity);
 }
@@ -388,7 +388,7 @@ int serdev_device_get_tiocm(struct serdev_device *serdev)
 	struct serdev_controller *ctrl = serdev->ctrl;
 
 	if (!ctrl || !ctrl->ops->get_tiocm)
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	return ctrl->ops->get_tiocm(ctrl);
 }
@@ -399,7 +399,7 @@ int serdev_device_set_tiocm(struct serdev_device *serdev, int set, int clear)
 	struct serdev_controller *ctrl = serdev->ctrl;
 
 	if (!ctrl || !ctrl->ops->set_tiocm)
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	return ctrl->ops->set_tiocm(ctrl, set, clear);
 }
diff --git a/drivers/tty/serdev/serdev-ttyport.c b/drivers/tty/serdev/serdev-ttyport.c
index d367803e2044..f26ff82723f1 100644
--- a/drivers/tty/serdev/serdev-ttyport.c
+++ b/drivers/tty/serdev/serdev-ttyport.c
@@ -231,7 +231,7 @@ static int ttyport_get_tiocm(struct serdev_controller *ctrl)
 	struct tty_struct *tty = serport->tty;
 
 	if (!tty->ops->tiocmget)
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	return tty->ops->tiocmget(tty);
 }
@@ -242,7 +242,7 @@ static int ttyport_set_tiocm(struct serdev_controller *ctrl, unsigned int set, u
 	struct tty_struct *tty = serport->tty;
 
 	if (!tty->ops->tiocmset)
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	return tty->ops->tiocmset(tty, set, clear);
 }
diff --git a/include/linux/serdev.h b/include/linux/serdev.h
index 5f6bfe4f6d95..babedb13785b 100644
--- a/include/linux/serdev.h
+++ b/include/linux/serdev.h
@@ -250,11 +250,11 @@ static inline int serdev_device_write_buf(struct serdev_device *serdev,
 static inline void serdev_device_wait_until_sent(struct serdev_device *sdev, long timeout) {}
 static inline int serdev_device_get_tiocm(struct serdev_device *serdev)
 {
-	return -ENOTSUPP;
+	return -EOPNOTSUPP;
 }
 static inline int serdev_device_set_tiocm(struct serdev_device *serdev, int set, int clear)
 {
-	return -ENOTSUPP;
+	return -EOPNOTSUPP;
 }
 static inline int serdev_device_write(struct serdev_device *sdev, const unsigned char *buf,
 				      size_t count, unsigned long timeout)
-- 
cgit v1.2.3-70-g09d2


From 8eaf839e4ac4feadf06e03eeff34059795450712 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Thu, 16 Mar 2023 22:52:12 +0530
Subject: serdev: Add method to assert break signal over tty UART port

Adds serdev_device_break_ctl() and an implementation for ttyport.
This function simply calls the break_ctl in tty layer, which can
assert a break signal over UART-TX line, if the tty and the
underlying platform and UART peripheral supports this operation.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/tty/serdev/core.c           | 11 +++++++++++
 drivers/tty/serdev/serdev-ttyport.c | 12 ++++++++++++
 include/linux/serdev.h              |  6 ++++++
 3 files changed, 29 insertions(+)

diff --git a/drivers/tty/serdev/core.c b/drivers/tty/serdev/core.c
index 078fe64625cb..e7d663901c07 100644
--- a/drivers/tty/serdev/core.c
+++ b/drivers/tty/serdev/core.c
@@ -405,6 +405,17 @@ int serdev_device_set_tiocm(struct serdev_device *serdev, int set, int clear)
 }
 EXPORT_SYMBOL_GPL(serdev_device_set_tiocm);
 
+int serdev_device_break_ctl(struct serdev_device *serdev, int break_state)
+{
+	struct serdev_controller *ctrl = serdev->ctrl;
+
+	if (!ctrl || !ctrl->ops->break_ctl)
+		return -EOPNOTSUPP;
+
+	return ctrl->ops->break_ctl(ctrl, break_state);
+}
+EXPORT_SYMBOL_GPL(serdev_device_break_ctl);
+
 static int serdev_drv_probe(struct device *dev)
 {
 	const struct serdev_device_driver *sdrv = to_serdev_device_driver(dev->driver);
diff --git a/drivers/tty/serdev/serdev-ttyport.c b/drivers/tty/serdev/serdev-ttyport.c
index f26ff82723f1..8033ef19669c 100644
--- a/drivers/tty/serdev/serdev-ttyport.c
+++ b/drivers/tty/serdev/serdev-ttyport.c
@@ -247,6 +247,17 @@ static int ttyport_set_tiocm(struct serdev_controller *ctrl, unsigned int set, u
 	return tty->ops->tiocmset(tty, set, clear);
 }
 
+static int ttyport_break_ctl(struct serdev_controller *ctrl, unsigned int break_state)
+{
+	struct serport *serport = serdev_controller_get_drvdata(ctrl);
+	struct tty_struct *tty = serport->tty;
+
+	if (!tty->ops->break_ctl)
+		return -EOPNOTSUPP;
+
+	return tty->ops->break_ctl(tty, break_state);
+}
+
 static const struct serdev_controller_ops ctrl_ops = {
 	.write_buf = ttyport_write_buf,
 	.write_flush = ttyport_write_flush,
@@ -259,6 +270,7 @@ static const struct serdev_controller_ops ctrl_ops = {
 	.wait_until_sent = ttyport_wait_until_sent,
 	.get_tiocm = ttyport_get_tiocm,
 	.set_tiocm = ttyport_set_tiocm,
+	.break_ctl = ttyport_break_ctl,
 };
 
 struct device *serdev_tty_port_register(struct tty_port *port,
diff --git a/include/linux/serdev.h b/include/linux/serdev.h
index babedb13785b..f5f97fa25e8a 100644
--- a/include/linux/serdev.h
+++ b/include/linux/serdev.h
@@ -93,6 +93,7 @@ struct serdev_controller_ops {
 	void (*wait_until_sent)(struct serdev_controller *, long);
 	int (*get_tiocm)(struct serdev_controller *);
 	int (*set_tiocm)(struct serdev_controller *, unsigned int, unsigned int);
+	int (*break_ctl)(struct serdev_controller *ctrl, unsigned int break_state);
 };
 
 /**
@@ -203,6 +204,7 @@ int serdev_device_write_buf(struct serdev_device *, const unsigned char *, size_
 void serdev_device_wait_until_sent(struct serdev_device *, long);
 int serdev_device_get_tiocm(struct serdev_device *);
 int serdev_device_set_tiocm(struct serdev_device *, int, int);
+int serdev_device_break_ctl(struct serdev_device *serdev, int break_state);
 void serdev_device_write_wakeup(struct serdev_device *);
 int serdev_device_write(struct serdev_device *, const unsigned char *, size_t, long);
 void serdev_device_write_flush(struct serdev_device *);
@@ -256,6 +258,10 @@ static inline int serdev_device_set_tiocm(struct serdev_device *serdev, int set,
 {
 	return -EOPNOTSUPP;
 }
+static inline int serdev_device_break_ctl(struct serdev_device *serdev, int break_state)
+{
+	return -EOPNOTSUPP;
+}
 static inline int serdev_device_write(struct serdev_device *sdev, const unsigned char *buf,
 				      size_t count, unsigned long timeout)
 {
-- 
cgit v1.2.3-70-g09d2


From ab3a769b4dccec2cf60f0a0700b140991bf9afc8 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Thu, 16 Mar 2023 22:52:13 +0530
Subject: dt-bindings: net: bluetooth: Add NXP bluetooth support

Add binding document for NXP bluetooth chipsets attached over UART.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../bindings/net/bluetooth/nxp,88w8987-bt.yaml     | 45 ++++++++++++++++++++++
 MAINTAINERS                                        |  6 +++
 2 files changed, 51 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml

diff --git a/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml b/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
new file mode 100644
index 000000000000..57e4c87cb00b
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
@@ -0,0 +1,45 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/bluetooth/nxp,88w8987-bt.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: NXP Bluetooth chips
+
+description:
+  This binding describes UART-attached NXP bluetooth chips. These chips
+  are dual-radio chips supporting WiFi and Bluetooth. The bluetooth
+  works on standard H4 protocol over 4-wire UART. The RTS and CTS lines
+  are used during FW download. To enable power save mode, the host
+  asserts break signal over UART-TX line to put the chip into power save
+  state. De-asserting break wakes up the BT chip.
+
+maintainers:
+  - Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
+
+properties:
+  compatible:
+    enum:
+      - nxp,88w8987-bt
+      - nxp,88w8997-bt
+
+  fw-init-baudrate:
+    description:
+      Chip baudrate after FW is downloaded and initialized.
+      This property depends on the module vendor's
+      configuration. If this property is not specified,
+      115200 is set as default.
+
+required:
+  - compatible
+
+additionalProperties: false
+
+examples:
+  - |
+    serial {
+        bluetooth {
+            compatible = "nxp,88w8987-bt";
+            fw-init-baudrate = <3000000>;
+        };
+    };
diff --git a/MAINTAINERS b/MAINTAINERS
index 6ac562e0381e..ee743e25b096 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -23237,6 +23237,12 @@ L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/zswap.c
 
+NXP BLUETOOTH WIRELESS DRIVERS
+M:	Amitkumar Karwar <amitkumar.karwar@nxp.com>
+M:	Neeraj Kale <neeraj.sanjaykale@nxp.com>
+S:	Maintained
+F:	Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
+
 THE REST
 M:	Linus Torvalds <torvalds@linux-foundation.org>
 L:	linux-kernel@vger.kernel.org
-- 
cgit v1.2.3-70-g09d2


From 689ca16e523278470c38832a3010645a78c544d8 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Thu, 16 Mar 2023 22:52:14 +0530
Subject: Bluetooth: NXP: Add protocol support for NXP Bluetooth chipsets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This adds a driver based on serdev driver for the NXP BT serial protocol
based on running H:4, which can enable the built-in Bluetooth device
inside an NXP BT chip.

This driver has Power Save feature that will put the chip into sleep state
whenever there is no activity for 2000ms, and will be woken up when any
activity is to be initiated over UART.

This driver enables the power save feature by default by sending the vendor
specific commands to the chip during setup.

During setup, the driver checks if a FW is already running on the chip
by waiting for the bootloader signature, and downloads device specific FW
file into the chip over UART if bootloader signature is received..

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 MAINTAINERS                   |    1 +
 drivers/bluetooth/Kconfig     |   12 +
 drivers/bluetooth/Makefile    |    1 +
 drivers/bluetooth/btnxpuart.c | 1297 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 1311 insertions(+)
 create mode 100644 drivers/bluetooth/btnxpuart.c

diff --git a/MAINTAINERS b/MAINTAINERS
index ee743e25b096..c2f3ecdb47e0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -23242,6 +23242,7 @@ M:	Amitkumar Karwar <amitkumar.karwar@nxp.com>
 M:	Neeraj Kale <neeraj.sanjaykale@nxp.com>
 S:	Maintained
 F:	Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
+F:	drivers/bluetooth/btnxpuart.c
 
 THE REST
 M:	Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index 5a1a7bec3c42..0703bdd44140 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -465,4 +465,16 @@ config BT_VIRTIO
 	  Say Y here to compile support for HCI over Virtio into the
 	  kernel or say M to compile as a module.
 
+config BT_NXPUART
+	tristate "NXP protocol support"
+	depends on SERIAL_DEV_BUS
+	select CRC32
+	help
+	  NXP is serial driver required for NXP Bluetooth
+	  devices with UART interface.
+
+	  Say Y here to compile support for NXP Bluetooth UART device into
+	  the kernel, or say M here to compile as a module (btnxpuart).
+
+
 endmenu
diff --git a/drivers/bluetooth/Makefile b/drivers/bluetooth/Makefile
index e0b261f24fc9..7a5967e9ac48 100644
--- a/drivers/bluetooth/Makefile
+++ b/drivers/bluetooth/Makefile
@@ -29,6 +29,7 @@ obj-$(CONFIG_BT_QCA)		+= btqca.o
 obj-$(CONFIG_BT_MTK)		+= btmtk.o
 
 obj-$(CONFIG_BT_VIRTIO)		+= virtio_bt.o
+obj-$(CONFIG_BT_NXPUART)	+= btnxpuart.o
 
 obj-$(CONFIG_BT_HCIUART_NOKIA)	+= hci_nokia.o
 
diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
new file mode 100644
index 000000000000..5f641466d695
--- /dev/null
+++ b/drivers/bluetooth/btnxpuart.c
@@ -0,0 +1,1297 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  NXP Bluetooth driver
+ *  Copyright 2023 NXP
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+
+#include <linux/serdev.h>
+#include <linux/of.h>
+#include <linux/skbuff.h>
+#include <asm/unaligned.h>
+#include <linux/firmware.h>
+#include <linux/string.h>
+#include <linux/crc8.h>
+#include <linux/crc32.h>
+#include <linux/string_helpers.h>
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+#include "h4_recv.h"
+
+#define MANUFACTURER_NXP		37
+
+#define BTNXPUART_TX_STATE_ACTIVE	1
+#define BTNXPUART_FW_DOWNLOADING	2
+#define BTNXPUART_CHECK_BOOT_SIGNATURE	3
+#define BTNXPUART_SERDEV_OPEN		4
+
+#define FIRMWARE_W8987	"nxp/uartuart8987_bt.bin"
+#define FIRMWARE_W8997	"nxp/uartuart8997_bt_v4.bin"
+#define FIRMWARE_W9098	"nxp/uartuart9098_bt_v1.bin"
+#define FIRMWARE_IW416	"nxp/uartiw416_bt_v0.bin"
+#define FIRMWARE_IW612	"nxp/uartspi_n61x_v1.bin.se"
+
+#define CHIP_ID_W9098		0x5c03
+#define CHIP_ID_IW416		0x7201
+#define CHIP_ID_IW612		0x7601
+
+#define HCI_NXP_PRI_BAUDRATE	115200
+#define HCI_NXP_SEC_BAUDRATE	3000000
+
+#define MAX_FW_FILE_NAME_LEN    50
+
+/* Default ps timeout period in milliseconds */
+#define PS_DEFAULT_TIMEOUT_PERIOD_MS     2000
+
+/* wakeup methods */
+#define WAKEUP_METHOD_DTR       0
+#define WAKEUP_METHOD_BREAK     1
+#define WAKEUP_METHOD_EXT_BREAK 2
+#define WAKEUP_METHOD_RTS       3
+#define WAKEUP_METHOD_INVALID   0xff
+
+/* power save mode status */
+#define PS_MODE_DISABLE         0
+#define PS_MODE_ENABLE          1
+
+/* Power Save Commands to ps_work_func  */
+#define PS_CMD_EXIT_PS          1
+#define PS_CMD_ENTER_PS         2
+
+/* power save state */
+#define PS_STATE_AWAKE          0
+#define PS_STATE_SLEEP          1
+
+/* Bluetooth vendor command : Sleep mode */
+#define HCI_NXP_AUTO_SLEEP_MODE	0xfc23
+/* Bluetooth vendor command : Wakeup method */
+#define HCI_NXP_WAKEUP_METHOD	0xfc53
+/* Bluetooth vendor command : Set operational baudrate */
+#define HCI_NXP_SET_OPER_SPEED	0xfc09
+/* Bluetooth vendor command: Independent Reset */
+#define HCI_NXP_IND_RESET	0xfcfc
+
+/* Bluetooth Power State : Vendor cmd params */
+#define BT_PS_ENABLE			0x02
+#define BT_PS_DISABLE			0x03
+
+/* Bluetooth Host Wakeup Methods */
+#define BT_HOST_WAKEUP_METHOD_NONE      0x00
+#define BT_HOST_WAKEUP_METHOD_DTR       0x01
+#define BT_HOST_WAKEUP_METHOD_BREAK     0x02
+#define BT_HOST_WAKEUP_METHOD_GPIO      0x03
+
+/* Bluetooth Chip Wakeup Methods */
+#define BT_CTRL_WAKEUP_METHOD_DSR       0x00
+#define BT_CTRL_WAKEUP_METHOD_BREAK     0x01
+#define BT_CTRL_WAKEUP_METHOD_GPIO      0x02
+#define BT_CTRL_WAKEUP_METHOD_EXT_BREAK 0x04
+#define BT_CTRL_WAKEUP_METHOD_RTS       0x05
+
+struct ps_data {
+	u8    target_ps_mode;	/* ps mode to be set */
+	u8    cur_psmode;	/* current ps_mode */
+	u8    ps_state;		/* controller's power save state */
+	u8    ps_cmd;
+	u8    h2c_wakeupmode;
+	u8    cur_h2c_wakeupmode;
+	u8    c2h_wakeupmode;
+	u8    c2h_wakeup_gpio;
+	u8    h2c_wakeup_gpio;
+	bool  driver_sent_cmd;
+	u16   h2c_ps_interval;
+	u16   c2h_ps_interval;
+	struct hci_dev *hdev;
+	struct work_struct work;
+	struct timer_list ps_timer;
+};
+
+struct wakeup_cmd_payload {
+	u8 c2h_wakeupmode;
+	u8 c2h_wakeup_gpio;
+	u8 h2c_wakeupmode;
+	u8 h2c_wakeup_gpio;
+} __packed;
+
+struct psmode_cmd_payload {
+	u8 ps_cmd;
+	__le16 c2h_ps_interval;
+} __packed;
+
+struct btnxpuart_data {
+	bool fw_download_3M_baudrate;
+	const char *fw_name;
+};
+
+struct btnxpuart_dev {
+	struct hci_dev *hdev;
+	struct serdev_device *serdev;
+
+	struct work_struct tx_work;
+	unsigned long tx_state;
+	struct sk_buff_head txq;
+	struct sk_buff *rx_skb;
+
+	const struct firmware *fw;
+	u8 fw_name[MAX_FW_FILE_NAME_LEN];
+	u32 fw_dnld_v1_offset;
+	u32 fw_v1_sent_bytes;
+	u32 fw_v3_offset_correction;
+	u32 fw_v1_expected_len;
+	wait_queue_head_t fw_dnld_done_wait_q;
+	wait_queue_head_t check_boot_sign_wait_q;
+
+	u32 new_baudrate;
+	u32 current_baudrate;
+	u32 fw_init_baudrate;
+	bool timeout_changed;
+	bool baudrate_changed;
+
+	struct ps_data psdata;
+	struct btnxpuart_data *nxp_data;
+};
+
+#define NXP_V1_FW_REQ_PKT	0xa5
+#define NXP_V1_CHIP_VER_PKT	0xaa
+#define NXP_V3_FW_REQ_PKT	0xa7
+#define NXP_V3_CHIP_VER_PKT	0xab
+
+#define NXP_ACK_V1		0x5a
+#define NXP_NAK_V1		0xbf
+#define NXP_ACK_V3		0x7a
+#define NXP_NAK_V3		0x7b
+#define NXP_CRC_ERROR_V3	0x7c
+
+#define HDR_LEN			16
+
+#define NXP_RECV_FW_REQ_V1 \
+	.type = NXP_V1_FW_REQ_PKT, \
+	.hlen = 4, \
+	.loff = 0, \
+	.lsize = 0, \
+	.maxlen = 4
+
+#define NXP_RECV_CHIP_VER_V3 \
+	.type = NXP_V3_CHIP_VER_PKT, \
+	.hlen = 4, \
+	.loff = 0, \
+	.lsize = 0, \
+	.maxlen = 4
+
+#define NXP_RECV_FW_REQ_V3 \
+	.type = NXP_V3_FW_REQ_PKT, \
+	.hlen = 9, \
+	.loff = 0, \
+	.lsize = 0, \
+	.maxlen = 9
+
+struct v1_data_req {
+	__le16 len;
+	__le16 len_comp;
+} __packed;
+
+struct v3_data_req {
+	__le16 len;
+	__le32 offset;
+	__le16 error;
+	u8 crc;
+} __packed;
+
+struct v3_start_ind {
+	__le16 chip_id;
+	u8 loader_ver;
+	u8 crc;
+} __packed;
+
+/* UART register addresses of BT chip */
+#define CLKDIVADDR	0x7f00008f
+#define UARTDIVADDR	0x7f000090
+#define UARTMCRADDR	0x7f000091
+#define UARTREINITADDR	0x7f000092
+#define UARTICRADDR	0x7f000093
+#define UARTFCRADDR	0x7f000094
+
+#define MCR		0x00000022
+#define INIT		0x00000001
+#define ICR		0x000000c7
+#define FCR		0x000000c7
+
+#define POLYNOMIAL8	0x07
+
+struct uart_reg {
+	__le32 address;
+	__le32 value;
+} __packed;
+
+struct uart_config {
+	struct uart_reg clkdiv;
+	struct uart_reg uartdiv;
+	struct uart_reg mcr;
+	struct uart_reg re_init;
+	struct uart_reg icr;
+	struct uart_reg fcr;
+	__be32 crc;
+} __packed;
+
+struct nxp_bootloader_cmd {
+	__le32 header;
+	__le32 arg;
+	__le32 payload_len;
+	__be32 crc;
+} __packed;
+
+static u8 crc8_table[CRC8_TABLE_SIZE];
+
+/* Default configurations */
+#define DEFAULT_H2C_WAKEUP_MODE	WAKEUP_METHOD_BREAK
+#define DEFAULT_PS_MODE		PS_MODE_ENABLE
+#define FW_INIT_BAUDRATE	HCI_NXP_PRI_BAUDRATE
+
+static struct sk_buff *nxp_drv_send_cmd(struct hci_dev *hdev, u16 opcode,
+					u32 plen,
+					void *param)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+	struct sk_buff *skb;
+
+	/* set flag to prevent nxp_enqueue from parsing values from this command and
+	 * calling hci_cmd_sync_queue() again.
+	 */
+	psdata->driver_sent_cmd = true;
+	skb = __hci_cmd_sync(hdev, opcode, plen, param, HCI_CMD_TIMEOUT);
+	psdata->driver_sent_cmd = false;
+
+	return skb;
+}
+
+static void btnxpuart_tx_wakeup(struct btnxpuart_dev *nxpdev)
+{
+	if (schedule_work(&nxpdev->tx_work))
+		set_bit(BTNXPUART_TX_STATE_ACTIVE, &nxpdev->tx_state);
+}
+
+/* NXP Power Save Feature */
+static void ps_start_timer(struct btnxpuart_dev *nxpdev)
+{
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	if (!psdata)
+		return;
+
+	if (psdata->cur_psmode == PS_MODE_ENABLE)
+		mod_timer(&psdata->ps_timer, jiffies + msecs_to_jiffies(psdata->h2c_ps_interval));
+}
+
+static void ps_cancel_timer(struct btnxpuart_dev *nxpdev)
+{
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	flush_work(&psdata->work);
+	del_timer_sync(&psdata->ps_timer);
+}
+
+static void ps_control(struct hci_dev *hdev, u8 ps_state)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+	int status;
+
+	if (psdata->ps_state == ps_state ||
+	    !test_bit(BTNXPUART_SERDEV_OPEN, &nxpdev->tx_state))
+		return;
+
+	switch (psdata->cur_h2c_wakeupmode) {
+	case WAKEUP_METHOD_DTR:
+		if (ps_state == PS_STATE_AWAKE)
+			status = serdev_device_set_tiocm(nxpdev->serdev, TIOCM_DTR, 0);
+		else
+			status = serdev_device_set_tiocm(nxpdev->serdev, 0, TIOCM_DTR);
+		break;
+	case WAKEUP_METHOD_BREAK:
+	default:
+		if (ps_state == PS_STATE_AWAKE)
+			status = serdev_device_break_ctl(nxpdev->serdev, 0);
+		else
+			status = serdev_device_break_ctl(nxpdev->serdev, -1);
+		bt_dev_dbg(hdev, "Set UART break: %s, status=%d",
+			   str_on_off(ps_state == PS_STATE_SLEEP), status);
+		break;
+	}
+	if (!status)
+		psdata->ps_state = ps_state;
+	if (ps_state == PS_STATE_AWAKE)
+		btnxpuart_tx_wakeup(nxpdev);
+}
+
+static void ps_work_func(struct work_struct *work)
+{
+	struct ps_data *data = container_of(work, struct ps_data, work);
+
+	if (data->ps_cmd == PS_CMD_ENTER_PS && data->cur_psmode == PS_MODE_ENABLE)
+		ps_control(data->hdev, PS_STATE_SLEEP);
+	else if (data->ps_cmd == PS_CMD_EXIT_PS)
+		ps_control(data->hdev, PS_STATE_AWAKE);
+}
+
+static void ps_timeout_func(struct timer_list *t)
+{
+	struct ps_data *data = from_timer(data, t, ps_timer);
+	struct hci_dev *hdev = data->hdev;
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+
+	if (test_bit(BTNXPUART_TX_STATE_ACTIVE, &nxpdev->tx_state)) {
+		ps_start_timer(nxpdev);
+	} else {
+		data->ps_cmd = PS_CMD_ENTER_PS;
+		schedule_work(&data->work);
+	}
+}
+
+static int ps_init_work(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	psdata->h2c_ps_interval = PS_DEFAULT_TIMEOUT_PERIOD_MS;
+	psdata->ps_state = PS_STATE_AWAKE;
+	psdata->target_ps_mode = DEFAULT_PS_MODE;
+	psdata->hdev = hdev;
+	psdata->c2h_wakeupmode = BT_HOST_WAKEUP_METHOD_NONE;
+	psdata->c2h_wakeup_gpio = 0xff;
+
+	switch (DEFAULT_H2C_WAKEUP_MODE) {
+	case WAKEUP_METHOD_DTR:
+		psdata->h2c_wakeupmode = WAKEUP_METHOD_DTR;
+		break;
+	case WAKEUP_METHOD_BREAK:
+	default:
+		psdata->h2c_wakeupmode = WAKEUP_METHOD_BREAK;
+		break;
+	}
+	psdata->cur_psmode = PS_MODE_DISABLE;
+	psdata->cur_h2c_wakeupmode = WAKEUP_METHOD_INVALID;
+	INIT_WORK(&psdata->work, ps_work_func);
+
+	return 0;
+}
+
+static void ps_init_timer(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	timer_setup(&psdata->ps_timer, ps_timeout_func, 0);
+}
+
+static void ps_wakeup(struct btnxpuart_dev *nxpdev)
+{
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	if (psdata->ps_state != PS_STATE_AWAKE) {
+		psdata->ps_cmd = PS_CMD_EXIT_PS;
+		schedule_work(&psdata->work);
+	}
+}
+
+static int send_ps_cmd(struct hci_dev *hdev, void *data)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+	struct psmode_cmd_payload pcmd;
+	struct sk_buff *skb;
+	u8 *status;
+
+	if (psdata->target_ps_mode == PS_MODE_ENABLE)
+		pcmd.ps_cmd = BT_PS_ENABLE;
+	else
+		pcmd.ps_cmd = BT_PS_DISABLE;
+	pcmd.c2h_ps_interval = __cpu_to_le16(psdata->c2h_ps_interval);
+
+	skb = nxp_drv_send_cmd(hdev, HCI_NXP_AUTO_SLEEP_MODE, sizeof(pcmd), &pcmd);
+	if (IS_ERR(skb)) {
+		bt_dev_err(hdev, "Setting Power Save mode failed (%ld)", PTR_ERR(skb));
+		return PTR_ERR(skb);
+	}
+
+	status = skb_pull_data(skb, 1);
+	if (status) {
+		if (!*status)
+			psdata->cur_psmode = psdata->target_ps_mode;
+		else
+			psdata->target_ps_mode = psdata->cur_psmode;
+		if (psdata->cur_psmode == PS_MODE_ENABLE)
+			ps_start_timer(nxpdev);
+		else
+			ps_wakeup(nxpdev);
+		bt_dev_dbg(hdev, "Power Save mode response: status=%d, ps_mode=%d",
+			   *status, psdata->cur_psmode);
+	}
+	kfree_skb(skb);
+
+	return 0;
+}
+
+static int send_wakeup_method_cmd(struct hci_dev *hdev, void *data)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+	struct wakeup_cmd_payload pcmd;
+	struct sk_buff *skb;
+	u8 *status;
+
+	pcmd.c2h_wakeupmode = psdata->c2h_wakeupmode;
+	pcmd.c2h_wakeup_gpio = psdata->c2h_wakeup_gpio;
+	switch (psdata->h2c_wakeupmode) {
+	case WAKEUP_METHOD_DTR:
+		pcmd.h2c_wakeupmode = BT_CTRL_WAKEUP_METHOD_DSR;
+		break;
+	case WAKEUP_METHOD_BREAK:
+	default:
+		pcmd.h2c_wakeupmode = BT_CTRL_WAKEUP_METHOD_BREAK;
+		break;
+	}
+	pcmd.h2c_wakeup_gpio = 0xff;
+
+	skb = nxp_drv_send_cmd(hdev, HCI_NXP_WAKEUP_METHOD, sizeof(pcmd), &pcmd);
+	if (IS_ERR(skb)) {
+		bt_dev_err(hdev, "Setting wake-up method failed (%ld)", PTR_ERR(skb));
+		return PTR_ERR(skb);
+	}
+
+	status = skb_pull_data(skb, 1);
+	if (status) {
+		if (*status == 0)
+			psdata->cur_h2c_wakeupmode = psdata->h2c_wakeupmode;
+		else
+			psdata->h2c_wakeupmode = psdata->cur_h2c_wakeupmode;
+		bt_dev_dbg(hdev, "Set Wakeup Method response: status=%d, h2c_wakeupmode=%d",
+			   *status, psdata->cur_h2c_wakeupmode);
+	}
+	kfree_skb(skb);
+
+	return 0;
+}
+
+static void ps_init(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	serdev_device_set_tiocm(nxpdev->serdev, 0, TIOCM_RTS);
+	usleep_range(5000, 10000);
+	serdev_device_set_tiocm(nxpdev->serdev, TIOCM_RTS, 0);
+	usleep_range(5000, 10000);
+
+	switch (psdata->h2c_wakeupmode) {
+	case WAKEUP_METHOD_DTR:
+		serdev_device_set_tiocm(nxpdev->serdev, 0, TIOCM_DTR);
+		serdev_device_set_tiocm(nxpdev->serdev, TIOCM_DTR, 0);
+		break;
+	case WAKEUP_METHOD_BREAK:
+	default:
+		serdev_device_break_ctl(nxpdev->serdev, -1);
+		usleep_range(5000, 10000);
+		serdev_device_break_ctl(nxpdev->serdev, 0);
+		usleep_range(5000, 10000);
+		break;
+	}
+	if (psdata->cur_h2c_wakeupmode != psdata->h2c_wakeupmode)
+		hci_cmd_sync_queue(hdev, send_wakeup_method_cmd, NULL, NULL);
+	if (psdata->cur_psmode != psdata->target_ps_mode)
+		hci_cmd_sync_queue(hdev, send_ps_cmd, NULL, NULL);
+}
+
+/* NXP Firmware Download Feature */
+static int nxp_download_firmware(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	int err = 0;
+
+	nxpdev->fw_dnld_v1_offset = 0;
+	nxpdev->fw_v1_sent_bytes = 0;
+	nxpdev->fw_v1_expected_len = HDR_LEN;
+	nxpdev->fw_v3_offset_correction = 0;
+	nxpdev->baudrate_changed = false;
+	nxpdev->timeout_changed = false;
+
+	serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_PRI_BAUDRATE);
+	serdev_device_set_flow_control(nxpdev->serdev, 0);
+	nxpdev->current_baudrate = HCI_NXP_PRI_BAUDRATE;
+
+	/* Wait till FW is downloaded and CTS becomes low */
+	err = wait_event_interruptible_timeout(nxpdev->fw_dnld_done_wait_q,
+					       !test_bit(BTNXPUART_FW_DOWNLOADING,
+							 &nxpdev->tx_state),
+					       msecs_to_jiffies(60000));
+	if (err == 0) {
+		bt_dev_err(hdev, "FW Download Timeout.");
+		return -ETIMEDOUT;
+	}
+
+	serdev_device_set_flow_control(nxpdev->serdev, 1);
+	err = serdev_device_wait_for_cts(nxpdev->serdev, 1, 60000);
+	if (err < 0) {
+		bt_dev_err(hdev, "CTS is still high. FW Download failed.");
+		return err;
+	}
+	release_firmware(nxpdev->fw);
+	memset(nxpdev->fw_name, 0, sizeof(nxpdev->fw_name));
+
+	/* Allow the downloaded FW to initialize */
+	usleep_range(800 * USEC_PER_MSEC, 1 * USEC_PER_SEC);
+
+	return 0;
+}
+
+static void nxp_send_ack(u8 ack, struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	u8 ack_nak[2];
+	int len = 1;
+
+	ack_nak[0] = ack;
+	if (ack == NXP_ACK_V3) {
+		ack_nak[1] = crc8(crc8_table, ack_nak, 1, 0xff);
+		len = 2;
+	}
+	serdev_device_write_buf(nxpdev->serdev, ack_nak, len);
+}
+
+static bool nxp_fw_change_baudrate(struct hci_dev *hdev, u16 req_len)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct nxp_bootloader_cmd nxp_cmd5;
+	struct uart_config uart_config;
+
+	if (req_len == sizeof(nxp_cmd5)) {
+		nxp_cmd5.header = __cpu_to_le32(5);
+		nxp_cmd5.arg = 0;
+		nxp_cmd5.payload_len = __cpu_to_le32(sizeof(uart_config));
+		/* FW expects swapped CRC bytes */
+		nxp_cmd5.crc = __cpu_to_be32(crc32_be(0UL, (char *)&nxp_cmd5,
+						      sizeof(nxp_cmd5) - 4));
+
+		serdev_device_write_buf(nxpdev->serdev, (u8 *)&nxp_cmd5, sizeof(nxp_cmd5));
+		nxpdev->fw_v3_offset_correction += req_len;
+	} else if (req_len == sizeof(uart_config)) {
+		uart_config.clkdiv.address = __cpu_to_le32(CLKDIVADDR);
+		uart_config.clkdiv.value = __cpu_to_le32(0x00c00000);
+		uart_config.uartdiv.address = __cpu_to_le32(UARTDIVADDR);
+		uart_config.uartdiv.value = __cpu_to_le32(1);
+		uart_config.mcr.address = __cpu_to_le32(UARTMCRADDR);
+		uart_config.mcr.value = __cpu_to_le32(MCR);
+		uart_config.re_init.address = __cpu_to_le32(UARTREINITADDR);
+		uart_config.re_init.value = __cpu_to_le32(INIT);
+		uart_config.icr.address = __cpu_to_le32(UARTICRADDR);
+		uart_config.icr.value = __cpu_to_le32(ICR);
+		uart_config.fcr.address = __cpu_to_le32(UARTFCRADDR);
+		uart_config.fcr.value = __cpu_to_le32(FCR);
+		/* FW expects swapped CRC bytes */
+		uart_config.crc = __cpu_to_be32(crc32_be(0UL, (char *)&uart_config,
+							 sizeof(uart_config) - 4));
+
+		serdev_device_write_buf(nxpdev->serdev, (u8 *)&uart_config, sizeof(uart_config));
+		serdev_device_wait_until_sent(nxpdev->serdev, 0);
+		nxpdev->fw_v3_offset_correction += req_len;
+		return true;
+	}
+	return false;
+}
+
+static bool nxp_fw_change_timeout(struct hci_dev *hdev, u16 req_len)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct nxp_bootloader_cmd nxp_cmd7;
+
+	if (req_len != sizeof(nxp_cmd7))
+		return false;
+
+	nxp_cmd7.header = __cpu_to_le32(7);
+	nxp_cmd7.arg = __cpu_to_le32(0x70);
+	nxp_cmd7.payload_len = 0;
+	/* FW expects swapped CRC bytes */
+	nxp_cmd7.crc = __cpu_to_be32(crc32_be(0UL, (char *)&nxp_cmd7,
+					      sizeof(nxp_cmd7) - 4));
+	serdev_device_write_buf(nxpdev->serdev, (u8 *)&nxp_cmd7, sizeof(nxp_cmd7));
+	serdev_device_wait_until_sent(nxpdev->serdev, 0);
+	nxpdev->fw_v3_offset_correction += req_len;
+	return true;
+}
+
+static u32 nxp_get_data_len(const u8 *buf)
+{
+	struct nxp_bootloader_cmd *hdr = (struct nxp_bootloader_cmd *)buf;
+
+	return __le32_to_cpu(hdr->payload_len);
+}
+
+static bool is_fw_downloading(struct btnxpuart_dev *nxpdev)
+{
+	return test_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+}
+
+static bool process_boot_signature(struct btnxpuart_dev *nxpdev)
+{
+	if (test_bit(BTNXPUART_CHECK_BOOT_SIGNATURE, &nxpdev->tx_state)) {
+		clear_bit(BTNXPUART_CHECK_BOOT_SIGNATURE, &nxpdev->tx_state);
+		wake_up_interruptible(&nxpdev->check_boot_sign_wait_q);
+		return false;
+	}
+	return is_fw_downloading(nxpdev);
+}
+
+static int nxp_request_firmware(struct hci_dev *hdev, const char *fw_name)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	int err = 0;
+
+	if (!strlen(nxpdev->fw_name)) {
+		snprintf(nxpdev->fw_name, MAX_FW_FILE_NAME_LEN, "%s", fw_name);
+
+		bt_dev_dbg(hdev, "Request Firmware: %s", nxpdev->fw_name);
+		err = request_firmware(&nxpdev->fw, nxpdev->fw_name, &hdev->dev);
+		if (err < 0) {
+			bt_dev_err(hdev, "Firmware file %s not found", nxpdev->fw_name);
+			clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+		}
+	}
+	return err;
+}
+
+/* for legacy chipsets with V1 bootloader */
+static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct btnxpuart_data *nxp_data = nxpdev->nxp_data;
+	struct v1_data_req *req;
+	u32 requested_len;
+
+	if (!process_boot_signature(nxpdev))
+		goto free_skb;
+
+	req = (struct v1_data_req *)skb_pull_data(skb, sizeof(struct v1_data_req));
+	if (!req)
+		goto free_skb;
+
+	if ((req->len ^ req->len_comp) != 0xffff) {
+		bt_dev_dbg(hdev, "ERR: Send NAK");
+		nxp_send_ack(NXP_NAK_V1, hdev);
+		goto free_skb;
+	}
+	nxp_send_ack(NXP_ACK_V1, hdev);
+
+	if (nxp_data->fw_download_3M_baudrate) {
+		if (!nxpdev->timeout_changed) {
+			nxpdev->timeout_changed = nxp_fw_change_timeout(hdev, req->len);
+			goto free_skb;
+		}
+		if (!nxpdev->baudrate_changed) {
+			nxpdev->baudrate_changed = nxp_fw_change_baudrate(hdev, req->len);
+			if (nxpdev->baudrate_changed) {
+				serdev_device_set_baudrate(nxpdev->serdev,
+							   HCI_NXP_SEC_BAUDRATE);
+				serdev_device_set_flow_control(nxpdev->serdev, 1);
+				nxpdev->current_baudrate = HCI_NXP_SEC_BAUDRATE;
+			}
+			goto free_skb;
+		}
+	}
+
+	if (nxp_request_firmware(hdev, nxp_data->fw_name))
+		goto free_skb;
+
+	requested_len = req->len;
+	if (requested_len == 0) {
+		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes", nxpdev->fw->size);
+		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
+		goto free_skb;
+	}
+	if (requested_len & 0x01) {
+		/* The CRC did not match at the other end.
+		 * Simply send the same bytes again.
+		 */
+		requested_len = nxpdev->fw_v1_sent_bytes;
+		bt_dev_dbg(hdev, "CRC error. Resend %d bytes of FW.", requested_len);
+	} else {
+		nxpdev->fw_dnld_v1_offset += nxpdev->fw_v1_sent_bytes;
+
+		/* The FW bin file is made up of many blocks of
+		 * 16 byte header and payload data chunks. If the
+		 * FW has requested a header, read the payload length
+		 * info from the header, before sending the header.
+		 * In the next iteration, the FW should request the
+		 * payload data chunk, which should be equal to the
+		 * payload length read from header. If there is a
+		 * mismatch, clearly the driver and FW are out of sync,
+		 * and we need to re-send the previous header again.
+		 */
+		if (requested_len == nxpdev->fw_v1_expected_len) {
+			if (requested_len == HDR_LEN)
+				nxpdev->fw_v1_expected_len = nxp_get_data_len(nxpdev->fw->data +
+									nxpdev->fw_dnld_v1_offset);
+			else
+				nxpdev->fw_v1_expected_len = HDR_LEN;
+		} else if (requested_len == HDR_LEN) {
+			/* FW download out of sync. Send previous chunk again */
+			nxpdev->fw_dnld_v1_offset -= nxpdev->fw_v1_sent_bytes;
+			nxpdev->fw_v1_expected_len = HDR_LEN;
+		}
+	}
+
+	if (nxpdev->fw_dnld_v1_offset + requested_len <= nxpdev->fw->size)
+		serdev_device_write_buf(nxpdev->serdev,
+					nxpdev->fw->data + nxpdev->fw_dnld_v1_offset,
+					requested_len);
+	nxpdev->fw_v1_sent_bytes = requested_len;
+
+free_skb:
+	kfree_skb(skb);
+	return 0;
+}
+
+static char *nxp_get_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid)
+{
+	char *fw_name = NULL;
+
+	switch (chipid) {
+	case CHIP_ID_W9098:
+		fw_name = FIRMWARE_W9098;
+		break;
+	case CHIP_ID_IW416:
+		fw_name = FIRMWARE_IW416;
+		break;
+	case CHIP_ID_IW612:
+		fw_name = FIRMWARE_IW612;
+		break;
+	default:
+		bt_dev_err(hdev, "Unknown chip signature %04x", chipid);
+		break;
+	}
+	return fw_name;
+}
+
+static int nxp_recv_chip_ver_v3(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct v3_start_ind *req = skb_pull_data(skb, sizeof(struct v3_start_ind));
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+
+	if (!process_boot_signature(nxpdev))
+		goto free_skb;
+
+	if (!nxp_request_firmware(hdev, nxp_get_fw_name_from_chipid(hdev, req->chip_id)))
+		nxp_send_ack(NXP_ACK_V3, hdev);
+
+free_skb:
+	kfree_skb(skb);
+	return 0;
+}
+
+static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct v3_data_req *req;
+
+	if (!process_boot_signature(nxpdev))
+		goto free_skb;
+
+	req = (struct v3_data_req *)skb_pull_data(skb, sizeof(struct v3_data_req));
+	if (!req || !nxpdev->fw)
+		goto free_skb;
+
+	nxp_send_ack(NXP_ACK_V3, hdev);
+
+	if (!nxpdev->timeout_changed) {
+		nxpdev->timeout_changed = nxp_fw_change_timeout(hdev, req->len);
+		goto free_skb;
+	}
+
+	if (!nxpdev->baudrate_changed) {
+		nxpdev->baudrate_changed = nxp_fw_change_baudrate(hdev, req->len);
+		if (nxpdev->baudrate_changed) {
+			serdev_device_set_baudrate(nxpdev->serdev,
+						   HCI_NXP_SEC_BAUDRATE);
+			serdev_device_set_flow_control(nxpdev->serdev, 1);
+			nxpdev->current_baudrate = HCI_NXP_SEC_BAUDRATE;
+		}
+		goto free_skb;
+	}
+
+	if (req->len == 0) {
+		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes", nxpdev->fw->size);
+		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
+		goto free_skb;
+	}
+	if (req->error)
+		bt_dev_dbg(hdev, "FW Download received err 0x%02x from chip", req->error);
+
+	if (req->offset < nxpdev->fw_v3_offset_correction) {
+		/* This scenario should ideally never occur. But if it ever does,
+		 * FW is out of sync and needs a power cycle.
+		 */
+		bt_dev_err(hdev, "Something went wrong during FW download. Please power cycle and try again");
+		goto free_skb;
+	}
+
+	serdev_device_write_buf(nxpdev->serdev,
+				nxpdev->fw->data + req->offset - nxpdev->fw_v3_offset_correction,
+				req->len);
+
+free_skb:
+	kfree_skb(skb);
+	return 0;
+}
+
+static int nxp_set_baudrate_cmd(struct hci_dev *hdev, void *data)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	__le32 new_baudrate = __cpu_to_le32(nxpdev->new_baudrate);
+	struct ps_data *psdata = &nxpdev->psdata;
+	struct sk_buff *skb;
+	u8 *status;
+
+	if (!psdata)
+		return 0;
+
+	skb = nxp_drv_send_cmd(hdev, HCI_NXP_SET_OPER_SPEED, 4, (u8 *)&new_baudrate);
+	if (IS_ERR(skb)) {
+		bt_dev_err(hdev, "Setting baudrate failed (%ld)", PTR_ERR(skb));
+		return PTR_ERR(skb);
+	}
+
+	status = (u8 *)skb_pull_data(skb, 1);
+	if (status) {
+		if (*status == 0) {
+			serdev_device_set_baudrate(nxpdev->serdev, nxpdev->new_baudrate);
+			nxpdev->current_baudrate = nxpdev->new_baudrate;
+		}
+		bt_dev_dbg(hdev, "Set baudrate response: status=%d, baudrate=%d",
+			   *status, nxpdev->new_baudrate);
+	}
+	kfree_skb(skb);
+
+	return 0;
+}
+
+static int nxp_set_ind_reset(struct hci_dev *hdev, void *data)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct sk_buff *skb;
+	u8 *status;
+	u8 pcmd = 0;
+	int err = 0;
+
+	skb = nxp_drv_send_cmd(hdev, HCI_NXP_IND_RESET, 1, &pcmd);
+	if (IS_ERR(skb))
+		return PTR_ERR(skb);
+
+	status = skb_pull_data(skb, 1);
+	if (!status || *status)
+		goto free_skb;
+
+	set_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+	err = nxp_download_firmware(hdev);
+	if (err < 0)
+		goto free_skb;
+	serdev_device_set_baudrate(nxpdev->serdev, nxpdev->fw_init_baudrate);
+	nxpdev->current_baudrate = nxpdev->fw_init_baudrate;
+	if (nxpdev->current_baudrate != HCI_NXP_SEC_BAUDRATE) {
+		nxpdev->new_baudrate = HCI_NXP_SEC_BAUDRATE;
+		nxp_set_baudrate_cmd(hdev, NULL);
+	}
+	hci_cmd_sync_queue(hdev, send_wakeup_method_cmd, NULL, NULL);
+	hci_cmd_sync_queue(hdev, send_ps_cmd, NULL, NULL);
+
+free_skb:
+	kfree_skb(skb);
+	return err;
+}
+
+/* NXP protocol */
+static int nxp_check_boot_sign(struct btnxpuart_dev *nxpdev)
+{
+	serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_PRI_BAUDRATE);
+	serdev_device_set_flow_control(nxpdev->serdev, 0);
+	set_bit(BTNXPUART_CHECK_BOOT_SIGNATURE, &nxpdev->tx_state);
+
+	return wait_event_interruptible_timeout(nxpdev->check_boot_sign_wait_q,
+					       !test_bit(BTNXPUART_CHECK_BOOT_SIGNATURE,
+							 &nxpdev->tx_state),
+					       msecs_to_jiffies(1000));
+}
+
+static int nxp_setup(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	int err = 0;
+
+	set_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+	init_waitqueue_head(&nxpdev->fw_dnld_done_wait_q);
+	init_waitqueue_head(&nxpdev->check_boot_sign_wait_q);
+
+	if (nxp_check_boot_sign(nxpdev)) {
+		bt_dev_dbg(hdev, "Need FW Download.");
+		err = nxp_download_firmware(hdev);
+		if (err < 0)
+			return err;
+	} else {
+		bt_dev_dbg(hdev, "FW already running.");
+		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+	}
+
+	serdev_device_set_flow_control(nxpdev->serdev, 1);
+	device_property_read_u32(&nxpdev->serdev->dev, "fw-init-baudrate",
+				 &nxpdev->fw_init_baudrate);
+	if (!nxpdev->fw_init_baudrate)
+		nxpdev->fw_init_baudrate = FW_INIT_BAUDRATE;
+	serdev_device_set_baudrate(nxpdev->serdev, nxpdev->fw_init_baudrate);
+	nxpdev->current_baudrate = nxpdev->fw_init_baudrate;
+
+	if (nxpdev->current_baudrate != HCI_NXP_SEC_BAUDRATE) {
+		nxpdev->new_baudrate = HCI_NXP_SEC_BAUDRATE;
+		hci_cmd_sync_queue(hdev, nxp_set_baudrate_cmd, NULL, NULL);
+	}
+
+	ps_init(hdev);
+
+	return 0;
+}
+
+static int btnxpuart_queue_skb(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+
+	/* Prepend skb with frame type */
+	memcpy(skb_push(skb, 1), &hci_skb_pkt_type(skb), 1);
+	skb_queue_tail(&nxpdev->txq, skb);
+	btnxpuart_tx_wakeup(nxpdev);
+	return 0;
+}
+
+static int nxp_enqueue(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct ps_data *psdata = &nxpdev->psdata;
+	struct hci_command_hdr *hdr;
+	struct psmode_cmd_payload ps_parm;
+	struct wakeup_cmd_payload wakeup_parm;
+	__le32 baudrate_parm;
+
+	/* if vendor commands are received from user space (e.g. hcitool), update
+	 * driver flags accordingly and ask driver to re-send the command to FW.
+	 * In case the payload for any command does not match expected payload
+	 * length, let the firmware and user space program handle it, or throw
+	 * an error.
+	 */
+	if (bt_cb(skb)->pkt_type == HCI_COMMAND_PKT && !psdata->driver_sent_cmd) {
+		hdr = (struct hci_command_hdr *)skb->data;
+		if (hdr->plen != (skb->len - HCI_COMMAND_HDR_SIZE))
+			return btnxpuart_queue_skb(hdev, skb);
+
+		switch (__le16_to_cpu(hdr->opcode)) {
+		case HCI_NXP_AUTO_SLEEP_MODE:
+			if (hdr->plen == sizeof(ps_parm)) {
+				memcpy(&ps_parm, skb->data + HCI_COMMAND_HDR_SIZE, hdr->plen);
+				if (ps_parm.ps_cmd == BT_PS_ENABLE)
+					psdata->target_ps_mode = PS_MODE_ENABLE;
+				else if (ps_parm.ps_cmd == BT_PS_DISABLE)
+					psdata->target_ps_mode = PS_MODE_DISABLE;
+				psdata->c2h_ps_interval = __le16_to_cpu(ps_parm.c2h_ps_interval);
+				hci_cmd_sync_queue(hdev, send_ps_cmd, NULL, NULL);
+				goto free_skb;
+			}
+			break;
+		case HCI_NXP_WAKEUP_METHOD:
+			if (hdr->plen == sizeof(wakeup_parm)) {
+				memcpy(&wakeup_parm, skb->data + HCI_COMMAND_HDR_SIZE, hdr->plen);
+				psdata->c2h_wakeupmode = wakeup_parm.c2h_wakeupmode;
+				psdata->c2h_wakeup_gpio = wakeup_parm.c2h_wakeup_gpio;
+				psdata->h2c_wakeup_gpio = wakeup_parm.h2c_wakeup_gpio;
+				switch (wakeup_parm.h2c_wakeupmode) {
+				case BT_CTRL_WAKEUP_METHOD_DSR:
+					psdata->h2c_wakeupmode = WAKEUP_METHOD_DTR;
+					break;
+				case BT_CTRL_WAKEUP_METHOD_BREAK:
+				default:
+					psdata->h2c_wakeupmode = WAKEUP_METHOD_BREAK;
+					break;
+				}
+				hci_cmd_sync_queue(hdev, send_wakeup_method_cmd, NULL, NULL);
+				goto free_skb;
+			}
+			break;
+		case HCI_NXP_SET_OPER_SPEED:
+			if (hdr->plen == sizeof(baudrate_parm)) {
+				memcpy(&baudrate_parm, skb->data + HCI_COMMAND_HDR_SIZE, hdr->plen);
+				nxpdev->new_baudrate = __le32_to_cpu(baudrate_parm);
+				hci_cmd_sync_queue(hdev, nxp_set_baudrate_cmd, NULL, NULL);
+				goto free_skb;
+			}
+			break;
+		case HCI_NXP_IND_RESET:
+			if (hdr->plen == 1) {
+				hci_cmd_sync_queue(hdev, nxp_set_ind_reset, NULL, NULL);
+				goto free_skb;
+			}
+			break;
+		default:
+			break;
+		}
+	}
+
+	return btnxpuart_queue_skb(hdev, skb);
+
+free_skb:
+	kfree_skb(skb);
+	return 0;
+}
+
+static struct sk_buff *nxp_dequeue(void *data)
+{
+	struct btnxpuart_dev *nxpdev = (struct btnxpuart_dev *)data;
+
+	ps_wakeup(nxpdev);
+	ps_start_timer(nxpdev);
+	return skb_dequeue(&nxpdev->txq);
+}
+
+/* btnxpuart based on serdev */
+static void btnxpuart_tx_work(struct work_struct *work)
+{
+	struct btnxpuart_dev *nxpdev = container_of(work, struct btnxpuart_dev,
+						   tx_work);
+	struct serdev_device *serdev = nxpdev->serdev;
+	struct hci_dev *hdev = nxpdev->hdev;
+	struct sk_buff *skb;
+	int len;
+
+	while ((skb = nxp_dequeue(nxpdev))) {
+		len = serdev_device_write_buf(serdev, skb->data, skb->len);
+		hdev->stat.byte_tx += len;
+
+		skb_pull(skb, len);
+		if (skb->len > 0) {
+			skb_queue_head(&nxpdev->txq, skb);
+			break;
+		}
+
+		switch (hci_skb_pkt_type(skb)) {
+		case HCI_COMMAND_PKT:
+			hdev->stat.cmd_tx++;
+			break;
+		case HCI_ACLDATA_PKT:
+			hdev->stat.acl_tx++;
+			break;
+		case HCI_SCODATA_PKT:
+			hdev->stat.sco_tx++;
+			break;
+		}
+
+		kfree_skb(skb);
+	}
+	clear_bit(BTNXPUART_TX_STATE_ACTIVE, &nxpdev->tx_state);
+}
+
+static int btnxpuart_open(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	int err = 0;
+
+	err = serdev_device_open(nxpdev->serdev);
+	if (err) {
+		bt_dev_err(hdev, "Unable to open UART device %s",
+			   dev_name(&nxpdev->serdev->dev));
+	} else {
+		set_bit(BTNXPUART_SERDEV_OPEN, &nxpdev->tx_state);
+	}
+	return err;
+}
+
+static int btnxpuart_close(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+
+	serdev_device_close(nxpdev->serdev);
+	clear_bit(BTNXPUART_SERDEV_OPEN, &nxpdev->tx_state);
+	return 0;
+}
+
+static int btnxpuart_flush(struct hci_dev *hdev)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+
+	/* Flush any pending characters */
+	serdev_device_write_flush(nxpdev->serdev);
+	skb_queue_purge(&nxpdev->txq);
+
+	cancel_work_sync(&nxpdev->tx_work);
+
+	kfree_skb(nxpdev->rx_skb);
+	nxpdev->rx_skb = NULL;
+
+	return 0;
+}
+
+static const struct h4_recv_pkt nxp_recv_pkts[] = {
+	{ H4_RECV_ACL,          .recv = hci_recv_frame },
+	{ H4_RECV_SCO,          .recv = hci_recv_frame },
+	{ H4_RECV_EVENT,        .recv = hci_recv_frame },
+	{ NXP_RECV_FW_REQ_V1,   .recv = nxp_recv_fw_req_v1 },
+	{ NXP_RECV_CHIP_VER_V3, .recv = nxp_recv_chip_ver_v3 },
+	{ NXP_RECV_FW_REQ_V3,   .recv = nxp_recv_fw_req_v3 },
+};
+
+static bool is_valid_bootloader_signature(const u8 *data, size_t count)
+{
+	if ((*data == NXP_V1_FW_REQ_PKT && count == sizeof(struct v1_data_req) + 1) ||
+	    (*data == NXP_V3_FW_REQ_PKT && count == sizeof(struct v3_data_req) + 1) ||
+	    (*data == NXP_V3_CHIP_VER_PKT && count == sizeof(struct v3_start_ind) + 1))
+		return true;
+	else
+		return false;
+}
+
+static int btnxpuart_receive_buf(struct serdev_device *serdev, const u8 *data,
+				 size_t count)
+{
+	struct btnxpuart_dev *nxpdev = serdev_device_get_drvdata(serdev);
+
+	if (is_fw_downloading(nxpdev) && !is_valid_bootloader_signature(data, count)) {
+		/* Unknown bootloader signature, skip without returning error */
+		return count;
+	}
+
+	ps_start_timer(nxpdev);
+
+	nxpdev->rx_skb = h4_recv_buf(nxpdev->hdev, nxpdev->rx_skb, data, count,
+				     nxp_recv_pkts, ARRAY_SIZE(nxp_recv_pkts));
+	if (IS_ERR(nxpdev->rx_skb)) {
+		int err = PTR_ERR(nxpdev->rx_skb);
+
+		bt_dev_err(nxpdev->hdev, "Frame reassembly failed (%d)", err);
+		nxpdev->rx_skb = NULL;
+		return err;
+	}
+	nxpdev->hdev->stat.byte_rx += count;
+	return count;
+}
+
+static void btnxpuart_write_wakeup(struct serdev_device *serdev)
+{
+	serdev_device_write_wakeup(serdev);
+}
+
+static const struct serdev_device_ops btnxpuart_client_ops = {
+	.receive_buf = btnxpuart_receive_buf,
+	.write_wakeup = btnxpuart_write_wakeup,
+};
+
+static int nxp_serdev_probe(struct serdev_device *serdev)
+{
+	struct hci_dev *hdev;
+	struct btnxpuart_dev *nxpdev;
+
+	nxpdev = devm_kzalloc(&serdev->dev, sizeof(*nxpdev), GFP_KERNEL);
+	if (!nxpdev)
+		return -ENOMEM;
+
+	nxpdev->nxp_data = (struct btnxpuart_data *)device_get_match_data(&serdev->dev);
+
+	nxpdev->serdev = serdev;
+	serdev_device_set_drvdata(serdev, nxpdev);
+
+	serdev_device_set_client_ops(serdev, &btnxpuart_client_ops);
+
+	INIT_WORK(&nxpdev->tx_work, btnxpuart_tx_work);
+	skb_queue_head_init(&nxpdev->txq);
+
+	crc8_populate_msb(crc8_table, POLYNOMIAL8);
+
+	/* Initialize and register HCI device */
+	hdev = hci_alloc_dev();
+	if (!hdev) {
+		dev_err(&serdev->dev, "Can't allocate HCI device\n");
+		return -ENOMEM;
+	}
+
+	nxpdev->hdev = hdev;
+
+	hdev->bus = HCI_UART;
+	hci_set_drvdata(hdev, nxpdev);
+
+	hdev->manufacturer = MANUFACTURER_NXP;
+	hdev->open  = btnxpuart_open;
+	hdev->close = btnxpuart_close;
+	hdev->flush = btnxpuart_flush;
+	hdev->setup = nxp_setup;
+	hdev->send  = nxp_enqueue;
+	SET_HCIDEV_DEV(hdev, &serdev->dev);
+
+	if (hci_register_dev(hdev) < 0) {
+		dev_err(&serdev->dev, "Can't register HCI device\n");
+		hci_free_dev(hdev);
+		return -ENODEV;
+	}
+
+	ps_init_work(hdev);
+	ps_init_timer(hdev);
+
+	return 0;
+}
+
+static void nxp_serdev_remove(struct serdev_device *serdev)
+{
+	struct btnxpuart_dev *nxpdev = serdev_device_get_drvdata(serdev);
+	struct hci_dev *hdev = nxpdev->hdev;
+
+	/* Restore FW baudrate to fw_init_baudrate if changed.
+	 * This will ensure FW baudrate is in sync with
+	 * driver baudrate in case this driver is re-inserted.
+	 */
+	if (nxpdev->current_baudrate != nxpdev->fw_init_baudrate) {
+		nxpdev->new_baudrate = nxpdev->fw_init_baudrate;
+		nxp_set_baudrate_cmd(hdev, NULL);
+	}
+
+	ps_cancel_timer(nxpdev);
+	hci_unregister_dev(hdev);
+	hci_free_dev(hdev);
+}
+
+static struct btnxpuart_data w8987_data = {
+	.fw_download_3M_baudrate = true,
+	.fw_name = FIRMWARE_W8987,
+};
+
+static struct btnxpuart_data w8997_data = {
+	.fw_download_3M_baudrate = false,
+	.fw_name = FIRMWARE_W8997,
+};
+
+static const struct of_device_id nxpuart_of_match_table[] = {
+	{ .compatible = "nxp,88w8987-bt", .data = &w8987_data },
+	{ .compatible = "nxp,88w8997-bt", .data = &w8997_data },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, nxpuart_of_match_table);
+
+static struct serdev_device_driver nxp_serdev_driver = {
+	.probe = nxp_serdev_probe,
+	.remove = nxp_serdev_remove,
+	.driver = {
+		.name = "btnxpuart",
+		.of_match_table = of_match_ptr(nxpuart_of_match_table),
+	},
+};
+
+module_serdev_device_driver(nxp_serdev_driver);
+
+MODULE_AUTHOR("Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>");
+MODULE_DESCRIPTION("NXP Bluetooth Serial driver");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-70-g09d2


From 11e29b4a3fda975f4427800285b7e4544f4e9152 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 22 Mar 2023 12:16:27 +0100
Subject: Bluetooth: NXP: select CONFIG_CRC8
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The driver needs both CRC32 and CRC8 to link correctly:

ld: drivers/bluetooth/btnxpuart.o: in function `nxp_serdev_probe':
drivers/bluetooth/btnxpuart.c:1214: undefined reference to
`crc8_populate_msb'
ld: drivers/bluetooth/btnxpuart.o: in function `nxp_send_ack':
drivers/bluetooth/btnxpuart.c:559: undefined reference to `crc8'
ld: drivers/bluetooth/btnxpuart.c:559: undefined reference to `crc8'

Fixes: 3e662aa4453a ("Bluetooth: NXP: Add protocol support for NXP Bluetooth chipsets")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index 0703bdd44140..f84322d375de 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -469,6 +469,7 @@ config BT_NXPUART
 	tristate "NXP protocol support"
 	depends on SERIAL_DEV_BUS
 	select CRC32
+	select CRC8
 	help
 	  NXP is serial driver required for NXP Bluetooth
 	  devices with UART interface.
-- 
cgit v1.2.3-70-g09d2


From 730a1d1a93a3e30c3723f87af97a8517334b2203 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Wed, 22 Mar 2023 19:52:02 -0500
Subject: bluetooth: Add device 0bda:887b to device tables

This device is part of a Realtek RTW8852BE chip.

The device table entry is as follows:

T:  Bus=03 Lev=01 Prnt=01 Port=12 Cnt=02 Dev#=  3 Spd=12   MxCh= 0
D:  Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=0bda ProdID=887b Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=Bluetooth Radio
S:  SerialNumber=00e04c000001
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 7382b021f3df..128deab609e4 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -540,6 +540,8 @@ static const struct usb_device_id blacklist_table[] = {
 	/* Realtek 8852BE Bluetooth devices */
 	{ USB_DEVICE(0x0cb8, 0xc559), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
+	{ USB_DEVICE(0x0bda, 0x887b), .driver_info = BTUSB_REALTEK |
+						     BTUSB_WIDEBAND_SPEECH },
 
 	/* Realtek Bluetooth devices */
 	{ USB_VENDOR_AND_INTERFACE_INFO(0x0bda, 0xe0, 0x01, 0x01),
-- 
cgit v1.2.3-70-g09d2


From 069f534247bb6db4f8c2c2ea8e9155abf495c37e Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Wed, 22 Mar 2023 19:52:03 -0500
Subject: bluetooth: Add device 13d3:3571 to device tables

This device is part of a Realtek RTW8852BE chip. The device table is as follows:

T:  Bus=03 Lev=01 Prnt=01 Port=01 Cnt=01 Dev#=  2 Spd=12   MxCh= 0
D:  Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=13d3 ProdID=3571 Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=Bluetooth Radio
S:  SerialNumber=00e04c000001
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 128deab609e4..9bf1d4ee74c2 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -542,6 +542,8 @@ static const struct usb_device_id blacklist_table[] = {
 						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x0bda, 0x887b), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
+	{ USB_DEVICE(0x13d3, 0x3571), .driver_info = BTUSB_REALTEK |
+						     BTUSB_WIDEBAND_SPEECH },
 
 	/* Realtek Bluetooth devices */
 	{ USB_VENDOR_AND_INTERFACE_INFO(0x0bda, 0xe0, 0x01, 0x01),
-- 
cgit v1.2.3-70-g09d2


From 253cf30e8d3d001850a95c4729d668f916b037ab Mon Sep 17 00:00:00 2001
From: Max Chou <max.chou@realtek.com>
Date: Tue, 21 Mar 2023 19:48:26 +0800
Subject: Bluetooth: btrtl: check for NULL in btrtl_set_quirks()

The btrtl_set_quirks() has accessed btrtl_dev->ic_info->lmp_subver since
b8e482d02513. However, if installing a Realtek Bluetooth controller
without the driver supported, it will hit the NULL point accessed.

Add a check for NULL to avoid the Kernel Oops.

Signed-off-by: Max Chou <max.chou@realtek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index 8fa7733812ad..0f256a8abef4 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -905,6 +905,9 @@ void btrtl_set_quirks(struct hci_dev *hdev, struct btrtl_device_info *btrtl_dev)
 		break;
 	}
 
+	if (!btrtl_dev->ic_info)
+		return;
+
 	switch (btrtl_dev->ic_info->lmp_subver) {
 	case RTL_ROM_LMP_8703B:
 		/* 8723CS reports two pages for local ext features,
-- 
cgit v1.2.3-70-g09d2


From b1d00baaa0298a5e033bc00a92522e4fd07900a8 Mon Sep 17 00:00:00 2001
From: Steev Klimaszewski <steev@kali.org>
Date: Sun, 26 Mar 2023 18:38:09 -0500
Subject: dt-bindings: net: Add WCN6855 Bluetooth

Add bindings for the QTI WCN6855 chipset.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Steev Klimaszewski <steev@kali.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../bindings/net/bluetooth/qualcomm-bluetooth.yaml      | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml b/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
index a6a6b0e4df7a..68f78b90d23a 100644
--- a/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
@@ -23,6 +23,7 @@ properties:
       - qcom,wcn3998-bt
       - qcom,qca6390-bt
       - qcom,wcn6750-bt
+      - qcom,wcn6855-bt
 
   enable-gpios:
     maxItems: 1
@@ -133,6 +134,22 @@ allOf:
         - vddrfa1p7-supply
         - vddrfa1p2-supply
         - vddasd-supply
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,wcn6855-bt
+    then:
+      required:
+        - enable-gpios
+        - swctrl-gpios
+        - vddio-supply
+        - vddbtcxmx-supply
+        - vddrfacmn-supply
+        - vddrfa0p8-supply
+        - vddrfa1p2-supply
+        - vddrfa1p7-supply
 
 examples:
   - |
-- 
cgit v1.2.3-70-g09d2


From 095327fede005f4b14d40b2183b2f7965c739dbd Mon Sep 17 00:00:00 2001
From: Steev Klimaszewski <steev@kali.org>
Date: Sun, 26 Mar 2023 18:38:10 -0500
Subject: Bluetooth: hci_qca: Add support for QTI Bluetooth chip wcn6855

Add regulators, GPIOs and changes required to power on/off wcn6855.
Add support for firmware download for wcn6855 which is in the
linux-firmware repository as hpbtfw21.tlv and hpnv21.bin.

Based on the assumption that this is similar to the wcn6750

Tested-on: BTFW.HSP.2.1.0-00538-VER_PATCHZ-1

Signed-off-by: Steev Klimaszewski <steev@kali.org>
Reviewed-by: Bjorn Andersson <andersson@kernel.org>
Tested-by: Bjorn Andersson <andersson@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btqca.c   | 14 ++++++++++-
 drivers/bluetooth/btqca.h   | 10 ++++++++
 drivers/bluetooth/hci_qca.c | 57 ++++++++++++++++++++++++++++++++++-----------
 3 files changed, 66 insertions(+), 15 deletions(-)

diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c
index c9064d34d830..fd0941fe8608 100644
--- a/drivers/bluetooth/btqca.c
+++ b/drivers/bluetooth/btqca.c
@@ -614,6 +614,9 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate,
 		config.type = ELF_TYPE_PATCH;
 		snprintf(config.fwname, sizeof(config.fwname),
 			 "qca/msbtfw%02x.mbn", rom_ver);
+	} else if (soc_type == QCA_WCN6855) {
+		snprintf(config.fwname, sizeof(config.fwname),
+			 "qca/hpbtfw%02x.tlv", rom_ver);
 	} else {
 		snprintf(config.fwname, sizeof(config.fwname),
 			 "qca/rampatch_%08x.bin", soc_ver);
@@ -648,6 +651,9 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate,
 	else if (soc_type == QCA_WCN6750)
 		snprintf(config.fwname, sizeof(config.fwname),
 			 "qca/msnv%02x.bin", rom_ver);
+	else if (soc_type == QCA_WCN6855)
+		snprintf(config.fwname, sizeof(config.fwname),
+			 "qca/hpnv%02x.bin", rom_ver);
 	else
 		snprintf(config.fwname, sizeof(config.fwname),
 			 "qca/nvm_%08x.bin", soc_ver);
@@ -685,11 +691,17 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate,
 		return err;
 	}
 
-	if (soc_type == QCA_WCN3991 || soc_type == QCA_WCN6750) {
+	switch (soc_type) {
+	case QCA_WCN3991:
+	case QCA_WCN6750:
+	case QCA_WCN6855:
 		/* get fw build info */
 		err = qca_read_fw_build_info(hdev);
 		if (err < 0)
 			return err;
+		break;
+	default:
+		break;
 	}
 
 	bt_dev_info(hdev, "QCA setup on UART is completed");
diff --git a/drivers/bluetooth/btqca.h b/drivers/bluetooth/btqca.h
index 61e9a50e66ae..b884095bcd9d 100644
--- a/drivers/bluetooth/btqca.h
+++ b/drivers/bluetooth/btqca.h
@@ -147,6 +147,7 @@ enum qca_btsoc_type {
 	QCA_WCN3991,
 	QCA_QCA6390,
 	QCA_WCN6750,
+	QCA_WCN6855,
 };
 
 #if IS_ENABLED(CONFIG_BT_QCA)
@@ -168,6 +169,10 @@ static inline bool qca_is_wcn6750(enum qca_btsoc_type soc_type)
 {
 	return soc_type == QCA_WCN6750;
 }
+static inline bool qca_is_wcn6855(enum qca_btsoc_type soc_type)
+{
+	return soc_type == QCA_WCN6855;
+}
 
 #else
 
@@ -206,6 +211,11 @@ static inline bool qca_is_wcn6750(enum qca_btsoc_type soc_type)
 	return false;
 }
 
+static inline bool qca_is_wcn6855(enum qca_btsoc_type soc_type)
+{
+	return false;
+}
+
 static inline int qca_send_pre_shutdown_cmd(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index a199ed52d76d..1597797ff169 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -1317,7 +1317,8 @@ static int qca_set_baudrate(struct hci_dev *hdev, uint8_t baudrate)
 
 	/* Give the controller time to process the request */
 	if (qca_is_wcn399x(qca_soc_type(hu)) ||
-	    qca_is_wcn6750(qca_soc_type(hu)))
+	    qca_is_wcn6750(qca_soc_type(hu)) ||
+	    qca_is_wcn6855(qca_soc_type(hu)))
 		usleep_range(1000, 10000);
 	else
 		msleep(300);
@@ -1394,7 +1395,8 @@ static unsigned int qca_get_speed(struct hci_uart *hu,
 static int qca_check_speeds(struct hci_uart *hu)
 {
 	if (qca_is_wcn399x(qca_soc_type(hu)) ||
-	    qca_is_wcn6750(qca_soc_type(hu))) {
+	    qca_is_wcn6750(qca_soc_type(hu)) ||
+	    qca_is_wcn6855(qca_soc_type(hu))) {
 		if (!qca_get_speed(hu, QCA_INIT_SPEED) &&
 		    !qca_get_speed(hu, QCA_OPER_SPEED))
 			return -EINVAL;
@@ -1428,7 +1430,8 @@ static int qca_set_speed(struct hci_uart *hu, enum qca_speed_type speed_type)
 		 * changing the baudrate of chip and host.
 		 */
 		if (qca_is_wcn399x(soc_type) ||
-		    qca_is_wcn6750(soc_type))
+		    qca_is_wcn6750(soc_type) ||
+		    qca_is_wcn6855(soc_type))
 			hci_uart_set_flow_control(hu, true);
 
 		if (soc_type == QCA_WCN3990) {
@@ -1446,7 +1449,8 @@ static int qca_set_speed(struct hci_uart *hu, enum qca_speed_type speed_type)
 
 error:
 		if (qca_is_wcn399x(soc_type) ||
-		    qca_is_wcn6750(soc_type))
+		    qca_is_wcn6750(soc_type) ||
+		    qca_is_wcn6855(soc_type))
 			hci_uart_set_flow_control(hu, false);
 
 		if (soc_type == QCA_WCN3990) {
@@ -1682,7 +1686,8 @@ static int qca_power_on(struct hci_dev *hdev)
 		return 0;
 
 	if (qca_is_wcn399x(soc_type) ||
-	    qca_is_wcn6750(soc_type)) {
+	    qca_is_wcn6750(soc_type) ||
+	    qca_is_wcn6855(soc_type)) {
 		ret = qca_regulator_init(hu);
 	} else {
 		qcadev = serdev_device_get_drvdata(hu->serdev);
@@ -1723,7 +1728,8 @@ static int qca_setup(struct hci_uart *hu)
 
 	bt_dev_info(hdev, "setting up %s",
 		qca_is_wcn399x(soc_type) ? "wcn399x" :
-		(soc_type == QCA_WCN6750) ? "wcn6750" : "ROME/QCA6390");
+		(soc_type == QCA_WCN6750) ? "wcn6750" :
+		(soc_type == QCA_WCN6855) ? "wcn6855" : "ROME/QCA6390");
 
 	qca->memdump_state = QCA_MEMDUMP_IDLE;
 
@@ -1735,7 +1741,8 @@ retry:
 	clear_bit(QCA_SSR_TRIGGERED, &qca->flags);
 
 	if (qca_is_wcn399x(soc_type) ||
-	    qca_is_wcn6750(soc_type)) {
+	    qca_is_wcn6750(soc_type) ||
+	    qca_is_wcn6855(soc_type)) {
 		set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);
 		hci_set_aosp_capable(hdev);
 
@@ -1757,7 +1764,8 @@ retry:
 	}
 
 	if (!(qca_is_wcn399x(soc_type) ||
-	     qca_is_wcn6750(soc_type))) {
+	      qca_is_wcn6750(soc_type) ||
+	      qca_is_wcn6855(soc_type))) {
 		/* Get QCA version information */
 		ret = qca_read_soc_version(hdev, &ver, soc_type);
 		if (ret)
@@ -1883,6 +1891,20 @@ static const struct qca_device_data qca_soc_data_wcn6750 __maybe_unused = {
 	.capabilities = QCA_CAP_WIDEBAND_SPEECH | QCA_CAP_VALID_LE_STATES,
 };
 
+static const struct qca_device_data qca_soc_data_wcn6855 = {
+	.soc_type = QCA_WCN6855,
+	.vregs = (struct qca_vreg []) {
+		{ "vddio", 5000 },
+		{ "vddbtcxmx", 126000 },
+		{ "vddrfacmn", 12500 },
+		{ "vddrfa0p8", 102000 },
+		{ "vddrfa1p7", 302000 },
+		{ "vddrfa1p2", 257000 },
+	},
+	.num_vregs = 6,
+	.capabilities = QCA_CAP_WIDEBAND_SPEECH | QCA_CAP_VALID_LE_STATES,
+};
+
 static void qca_power_shutdown(struct hci_uart *hu)
 {
 	struct qca_serdev *qcadev;
@@ -1912,7 +1934,7 @@ static void qca_power_shutdown(struct hci_uart *hu)
 		host_set_baudrate(hu, 2400);
 		qca_send_power_pulse(hu, false);
 		qca_regulator_disable(qcadev);
-	} else if (soc_type == QCA_WCN6750) {
+	} else if (soc_type == QCA_WCN6750 || soc_type == QCA_WCN6855) {
 		gpiod_set_value_cansleep(qcadev->bt_en, 0);
 		msleep(100);
 		qca_regulator_disable(qcadev);
@@ -2047,7 +2069,8 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 
 	if (data &&
 	    (qca_is_wcn399x(data->soc_type) ||
-	    qca_is_wcn6750(data->soc_type))) {
+	     qca_is_wcn6750(data->soc_type) ||
+	     qca_is_wcn6855(data->soc_type))) {
 		qcadev->btsoc_type = data->soc_type;
 		qcadev->bt_power = devm_kzalloc(&serdev->dev,
 						sizeof(struct qca_power),
@@ -2067,14 +2090,18 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
 					       GPIOD_OUT_LOW);
-		if (IS_ERR_OR_NULL(qcadev->bt_en) && data->soc_type == QCA_WCN6750) {
+		if (IS_ERR_OR_NULL(qcadev->bt_en) &&
+		    (data->soc_type == QCA_WCN6750 ||
+		     data->soc_type == QCA_WCN6855)) {
 			dev_err(&serdev->dev, "failed to acquire BT_EN gpio\n");
 			power_ctrl_enabled = false;
 		}
 
 		qcadev->sw_ctrl = devm_gpiod_get_optional(&serdev->dev, "swctrl",
 					       GPIOD_IN);
-		if (IS_ERR_OR_NULL(qcadev->sw_ctrl) && data->soc_type == QCA_WCN6750)
+		if (IS_ERR_OR_NULL(qcadev->sw_ctrl) &&
+		    (data->soc_type == QCA_WCN6750 ||
+		     data->soc_type == QCA_WCN6855))
 			dev_warn(&serdev->dev, "failed to acquire SW_CTRL gpio\n");
 
 		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
@@ -2150,8 +2177,9 @@ static void qca_serdev_remove(struct serdev_device *serdev)
 	struct qca_power *power = qcadev->bt_power;
 
 	if ((qca_is_wcn399x(qcadev->btsoc_type) ||
-	     qca_is_wcn6750(qcadev->btsoc_type)) &&
-	     power->vregs_on)
+	     qca_is_wcn6750(qcadev->btsoc_type) ||
+	     qca_is_wcn6855(qcadev->btsoc_type)) &&
+	    power->vregs_on)
 		qca_power_shutdown(&qcadev->serdev_hu);
 	else if (qcadev->susclk)
 		clk_disable_unprepare(qcadev->susclk);
@@ -2335,6 +2363,7 @@ static const struct of_device_id qca_bluetooth_of_match[] = {
 	{ .compatible = "qcom,wcn3991-bt", .data = &qca_soc_data_wcn3991},
 	{ .compatible = "qcom,wcn3998-bt", .data = &qca_soc_data_wcn3998},
 	{ .compatible = "qcom,wcn6750-bt", .data = &qca_soc_data_wcn6750},
+	{ .compatible = "qcom,wcn6855-bt", .data = &qca_soc_data_wcn6855},
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, qca_bluetooth_of_match);
-- 
cgit v1.2.3-70-g09d2


From 77f542b10c535c9a93bf8afdd2665524935807c2 Mon Sep 17 00:00:00 2001
From: Chethan T N <chethan.tumkur.narayan@intel.com>
Date: Tue, 21 Mar 2023 10:03:10 +0530
Subject: Bluetooth: btintel: Add LE States quirk support

Basically all Intel controllers support both Central/Peripheral
LE states.

This patch enables the LE States quirk by default on all
Solar and Magnertor Intel controllers.

Signed-off-by: Chethan T N <chethan.tumkur.narayan@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index af774688f1c0..7a6dc05553f1 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -2684,9 +2684,8 @@ static int btintel_setup_combined(struct hci_dev *hdev)
 		 */
 		set_bit(HCI_QUIRK_WIDEBAND_SPEECH_SUPPORTED, &hdev->quirks);
 
-		/* Valid LE States quirk for GfP */
-		if (INTEL_HW_VARIANT(ver_tlv.cnvi_bt) == 0x18)
-			set_bit(HCI_QUIRK_VALID_LE_STATES, &hdev->quirks);
+		/* Apply LE States quirk from solar onwards */
+		set_bit(HCI_QUIRK_VALID_LE_STATES, &hdev->quirks);
 
 		/* Setup MSFT Extension support */
 		btintel_set_msft_opcode(hdev,
-- 
cgit v1.2.3-70-g09d2


From 9695ef876fd122cb7bbc04a4a93b8727d2e36bda Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Thu, 30 Mar 2023 09:58:23 -0700
Subject: Bluetooth: Add support for hci devcoredump

Add devcoredump APIs to hci core so that drivers only have to provide
the dump skbs instead of managing the synchronization and timeouts.

The devcoredump APIs should be used in the following manner:
 - hci_devcoredump_init is called to allocate the dump.
 - hci_devcoredump_append is called to append any skbs with dump data
   OR hci_devcoredump_append_pattern is called to insert a pattern.
 - hci_devcoredump_complete is called when all dump packets have been
   sent OR hci_devcoredump_abort is called to indicate an error and
   cancel an ongoing dump collection.

The high level APIs just prepare some skbs with the appropriate data and
queue it for the dump to process. Packets part of the crashdump can be
intercepted in the driver in interrupt context and forwarded directly to
the devcoredump APIs.

Internally, there are 5 states for the dump: idle, active, complete,
abort and timeout. A devcoredump will only be in active state after it
has been initialized. Once active, it accepts data to be appended,
patterns to be inserted (i.e. memset) and a completion event or an abort
event to generate a devcoredump. The timeout is initialized at the same
time the dump is initialized (defaulting to 10s) and will be cleared
either when the timeout occurs or the dump is complete or aborted.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Manish Mandlik <mmandlik@google.com>
Reviewed-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/coredump.h | 116 +++++++++
 include/net/bluetooth/hci_core.h |  14 +
 net/bluetooth/Makefile           |   2 +
 net/bluetooth/coredump.c         | 535 +++++++++++++++++++++++++++++++++++++++
 net/bluetooth/hci_core.c         |   1 +
 net/bluetooth/hci_sync.c         |   2 +
 6 files changed, 670 insertions(+)
 create mode 100644 include/net/bluetooth/coredump.h
 create mode 100644 net/bluetooth/coredump.c

diff --git a/include/net/bluetooth/coredump.h b/include/net/bluetooth/coredump.h
new file mode 100644
index 000000000000..72f51b587a04
--- /dev/null
+++ b/include/net/bluetooth/coredump.h
@@ -0,0 +1,116 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2022 Google Corporation
+ */
+
+#ifndef __COREDUMP_H
+#define __COREDUMP_H
+
+#define DEVCOREDUMP_TIMEOUT	msecs_to_jiffies(10000)	/* 10 sec */
+
+typedef void (*coredump_t)(struct hci_dev *hdev);
+typedef void (*dmp_hdr_t)(struct hci_dev *hdev, struct sk_buff *skb);
+typedef void (*notify_change_t)(struct hci_dev *hdev, int state);
+
+/* struct hci_devcoredump - Devcoredump state
+ *
+ * @supported: Indicates if FW dump collection is supported by driver
+ * @state: Current state of dump collection
+ * @timeout: Indicates a timeout for collecting the devcoredump
+ *
+ * @alloc_size: Total size of the dump
+ * @head: Start of the dump
+ * @tail: Pointer to current end of dump
+ * @end: head + alloc_size for easy comparisons
+ *
+ * @dump_q: Dump queue for state machine to process
+ * @dump_rx: Devcoredump state machine work
+ * @dump_timeout: Devcoredump timeout work
+ *
+ * @coredump: Called from the driver's .coredump() function.
+ * @dmp_hdr: Create a dump header to identify controller/fw/driver info
+ * @notify_change: Notify driver when devcoredump state has changed
+ */
+struct hci_devcoredump {
+	bool		supported;
+
+	enum devcoredump_state {
+		HCI_DEVCOREDUMP_IDLE,
+		HCI_DEVCOREDUMP_ACTIVE,
+		HCI_DEVCOREDUMP_DONE,
+		HCI_DEVCOREDUMP_ABORT,
+		HCI_DEVCOREDUMP_TIMEOUT,
+	} state;
+
+	unsigned long	timeout;
+
+	size_t		alloc_size;
+	char		*head;
+	char		*tail;
+	char		*end;
+
+	struct sk_buff_head	dump_q;
+	struct work_struct	dump_rx;
+	struct delayed_work	dump_timeout;
+
+	coredump_t		coredump;
+	dmp_hdr_t		dmp_hdr;
+	notify_change_t		notify_change;
+};
+
+#ifdef CONFIG_DEV_COREDUMP
+
+void hci_devcd_reset(struct hci_dev *hdev);
+void hci_devcd_rx(struct work_struct *work);
+void hci_devcd_timeout(struct work_struct *work);
+
+int hci_devcd_register(struct hci_dev *hdev, coredump_t coredump,
+		       dmp_hdr_t dmp_hdr, notify_change_t notify_change);
+int hci_devcd_init(struct hci_dev *hdev, u32 dump_size);
+int hci_devcd_append(struct hci_dev *hdev, struct sk_buff *skb);
+int hci_devcd_append_pattern(struct hci_dev *hdev, u8 pattern, u32 len);
+int hci_devcd_complete(struct hci_dev *hdev);
+int hci_devcd_abort(struct hci_dev *hdev);
+
+#else
+
+static inline void hci_devcd_reset(struct hci_dev *hdev) {}
+static inline void hci_devcd_rx(struct work_struct *work) {}
+static inline void hci_devcd_timeout(struct work_struct *work) {}
+
+static inline int hci_devcd_register(struct hci_dev *hdev, coredump_t coredump,
+				     dmp_hdr_t dmp_hdr,
+				     notify_change_t notify_change)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int hci_devcd_init(struct hci_dev *hdev, u32 dump_size)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int hci_devcd_append(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int hci_devcd_append_pattern(struct hci_dev *hdev,
+					   u8 pattern, u32 len)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int hci_devcd_complete(struct hci_dev *hdev)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int hci_devcd_abort(struct hci_dev *hdev)
+{
+	return -EOPNOTSUPP;
+}
+
+#endif /* CONFIG_DEV_COREDUMP */
+
+#endif /* __COREDUMP_H */
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index e22e45fbe8db..a461335beacf 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -32,6 +32,7 @@
 #include <net/bluetooth/hci.h>
 #include <net/bluetooth/hci_sync.h>
 #include <net/bluetooth/hci_sock.h>
+#include <net/bluetooth/coredump.h>
 
 /* HCI priority */
 #define HCI_PRIO_MAX	7
@@ -590,6 +591,10 @@ struct hci_dev {
 	const char		*fw_info;
 	struct dentry		*debugfs;
 
+#ifdef CONFIG_DEV_COREDUMP
+	struct hci_devcoredump	dump;
+#endif
+
 	struct device		dev;
 
 	struct rfkill		*rfkill;
@@ -1497,6 +1502,15 @@ static inline void hci_set_aosp_capable(struct hci_dev *hdev)
 #endif
 }
 
+static inline void hci_devcd_setup(struct hci_dev *hdev)
+{
+#ifdef CONFIG_DEV_COREDUMP
+	INIT_WORK(&hdev->dump.dump_rx, hci_devcd_rx);
+	INIT_DELAYED_WORK(&hdev->dump.dump_timeout, hci_devcd_timeout);
+	skb_queue_head_init(&hdev->dump.dump_q);
+#endif
+}
+
 int hci_dev_open(__u16 dev);
 int hci_dev_close(__u16 dev);
 int hci_dev_do_close(struct hci_dev *hdev);
diff --git a/net/bluetooth/Makefile b/net/bluetooth/Makefile
index 0e7b7db42750..141ac1fda0bf 100644
--- a/net/bluetooth/Makefile
+++ b/net/bluetooth/Makefile
@@ -17,6 +17,8 @@ bluetooth-y := af_bluetooth.o hci_core.o hci_conn.o hci_event.o mgmt.o \
 	ecdh_helper.o hci_request.o mgmt_util.o mgmt_config.o hci_codec.o \
 	eir.o hci_sync.o
 
+bluetooth-$(CONFIG_DEV_COREDUMP) += coredump.o
+
 bluetooth-$(CONFIG_BT_BREDR) += sco.o
 bluetooth-$(CONFIG_BT_LE) += iso.o
 bluetooth-$(CONFIG_BT_HS) += a2mp.o amp.o
diff --git a/net/bluetooth/coredump.c b/net/bluetooth/coredump.c
new file mode 100644
index 000000000000..08fa98505454
--- /dev/null
+++ b/net/bluetooth/coredump.c
@@ -0,0 +1,535 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2023 Google Corporation
+ */
+
+#include <linux/devcoredump.h>
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+enum hci_devcoredump_pkt_type {
+	HCI_DEVCOREDUMP_PKT_INIT,
+	HCI_DEVCOREDUMP_PKT_SKB,
+	HCI_DEVCOREDUMP_PKT_PATTERN,
+	HCI_DEVCOREDUMP_PKT_COMPLETE,
+	HCI_DEVCOREDUMP_PKT_ABORT,
+};
+
+struct hci_devcoredump_skb_cb {
+	u16 pkt_type;
+};
+
+struct hci_devcoredump_skb_pattern {
+	u8 pattern;
+	u32 len;
+} __packed;
+
+#define hci_dmp_cb(skb)	((struct hci_devcoredump_skb_cb *)((skb)->cb))
+
+#define DBG_UNEXPECTED_STATE() \
+	bt_dev_dbg(hdev, \
+		   "Unexpected packet (%d) for state (%d). ", \
+		   hci_dmp_cb(skb)->pkt_type, hdev->dump.state)
+
+#define MAX_DEVCOREDUMP_HDR_SIZE	512	/* bytes */
+
+static int hci_devcd_update_hdr_state(char *buf, size_t size, int state)
+{
+	int len = 0;
+
+	if (!buf)
+		return 0;
+
+	len = scnprintf(buf, size, "Bluetooth devcoredump\nState: %d\n", state);
+
+	return len + 1; /* scnprintf adds \0 at the end upon state rewrite */
+}
+
+/* Call with hci_dev_lock only. */
+static int hci_devcd_update_state(struct hci_dev *hdev, int state)
+{
+	bt_dev_dbg(hdev, "Updating devcoredump state from %d to %d.",
+		   hdev->dump.state, state);
+
+	hdev->dump.state = state;
+
+	return hci_devcd_update_hdr_state(hdev->dump.head,
+					  hdev->dump.alloc_size, state);
+}
+
+static int hci_devcd_mkheader(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	char dump_start[] = "--- Start dump ---\n";
+	char hdr[80];
+	int hdr_len;
+
+	hdr_len = hci_devcd_update_hdr_state(hdr, sizeof(hdr),
+					     HCI_DEVCOREDUMP_IDLE);
+	skb_put_data(skb, hdr, hdr_len);
+
+	if (hdev->dump.dmp_hdr)
+		hdev->dump.dmp_hdr(hdev, skb);
+
+	skb_put_data(skb, dump_start, strlen(dump_start));
+
+	return skb->len;
+}
+
+/* Do not call with hci_dev_lock since this calls driver code. */
+static void hci_devcd_notify(struct hci_dev *hdev, int state)
+{
+	if (hdev->dump.notify_change)
+		hdev->dump.notify_change(hdev, state);
+}
+
+/* Call with hci_dev_lock only. */
+void hci_devcd_reset(struct hci_dev *hdev)
+{
+	hdev->dump.head = NULL;
+	hdev->dump.tail = NULL;
+	hdev->dump.alloc_size = 0;
+
+	hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_IDLE);
+
+	cancel_delayed_work(&hdev->dump.dump_timeout);
+	skb_queue_purge(&hdev->dump.dump_q);
+}
+
+/* Call with hci_dev_lock only. */
+static void hci_devcd_free(struct hci_dev *hdev)
+{
+	if (hdev->dump.head)
+		vfree(hdev->dump.head);
+
+	hci_devcd_reset(hdev);
+}
+
+/* Call with hci_dev_lock only. */
+static int hci_devcd_alloc(struct hci_dev *hdev, u32 size)
+{
+	hdev->dump.head = vmalloc(size);
+	if (!hdev->dump.head)
+		return -ENOMEM;
+
+	hdev->dump.alloc_size = size;
+	hdev->dump.tail = hdev->dump.head;
+	hdev->dump.end = hdev->dump.head + size;
+
+	hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_IDLE);
+
+	return 0;
+}
+
+/* Call with hci_dev_lock only. */
+static bool hci_devcd_copy(struct hci_dev *hdev, char *buf, u32 size)
+{
+	if (hdev->dump.tail + size > hdev->dump.end)
+		return false;
+
+	memcpy(hdev->dump.tail, buf, size);
+	hdev->dump.tail += size;
+
+	return true;
+}
+
+/* Call with hci_dev_lock only. */
+static bool hci_devcd_memset(struct hci_dev *hdev, u8 pattern, u32 len)
+{
+	if (hdev->dump.tail + len > hdev->dump.end)
+		return false;
+
+	memset(hdev->dump.tail, pattern, len);
+	hdev->dump.tail += len;
+
+	return true;
+}
+
+/* Call with hci_dev_lock only. */
+static int hci_devcd_prepare(struct hci_dev *hdev, u32 dump_size)
+{
+	struct sk_buff *skb;
+	int dump_hdr_size;
+	int err = 0;
+
+	skb = alloc_skb(MAX_DEVCOREDUMP_HDR_SIZE, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	dump_hdr_size = hci_devcd_mkheader(hdev, skb);
+
+	if (hci_devcd_alloc(hdev, dump_hdr_size + dump_size)) {
+		err = -ENOMEM;
+		goto hdr_free;
+	}
+
+	/* Insert the device header */
+	if (!hci_devcd_copy(hdev, skb->data, skb->len)) {
+		bt_dev_err(hdev, "Failed to insert header");
+		hci_devcd_free(hdev);
+
+		err = -ENOMEM;
+		goto hdr_free;
+	}
+
+hdr_free:
+	kfree_skb(skb);
+
+	return err;
+}
+
+static void hci_devcd_handle_pkt_init(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	u32 *dump_size;
+
+	if (hdev->dump.state != HCI_DEVCOREDUMP_IDLE) {
+		DBG_UNEXPECTED_STATE();
+		return;
+	}
+
+	if (skb->len != sizeof(*dump_size)) {
+		bt_dev_dbg(hdev, "Invalid dump init pkt");
+		return;
+	}
+
+	dump_size = skb_pull_data(skb, sizeof(*dump_size));
+	if (!*dump_size) {
+		bt_dev_err(hdev, "Zero size dump init pkt");
+		return;
+	}
+
+	if (hci_devcd_prepare(hdev, *dump_size)) {
+		bt_dev_err(hdev, "Failed to prepare for dump");
+		return;
+	}
+
+	hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_ACTIVE);
+	queue_delayed_work(hdev->workqueue, &hdev->dump.dump_timeout,
+			   hdev->dump.timeout);
+}
+
+static void hci_devcd_handle_pkt_skb(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) {
+		DBG_UNEXPECTED_STATE();
+		return;
+	}
+
+	if (!hci_devcd_copy(hdev, skb->data, skb->len))
+		bt_dev_dbg(hdev, "Failed to insert skb");
+}
+
+static void hci_devcd_handle_pkt_pattern(struct hci_dev *hdev,
+					 struct sk_buff *skb)
+{
+	struct hci_devcoredump_skb_pattern *pattern;
+
+	if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) {
+		DBG_UNEXPECTED_STATE();
+		return;
+	}
+
+	if (skb->len != sizeof(*pattern)) {
+		bt_dev_dbg(hdev, "Invalid pattern skb");
+		return;
+	}
+
+	pattern = skb_pull_data(skb, sizeof(*pattern));
+
+	if (!hci_devcd_memset(hdev, pattern->pattern, pattern->len))
+		bt_dev_dbg(hdev, "Failed to set pattern");
+}
+
+static void hci_devcd_handle_pkt_complete(struct hci_dev *hdev,
+					  struct sk_buff *skb)
+{
+	u32 dump_size;
+
+	if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) {
+		DBG_UNEXPECTED_STATE();
+		return;
+	}
+
+	hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_DONE);
+	dump_size = hdev->dump.tail - hdev->dump.head;
+
+	bt_dev_dbg(hdev, "complete with size %u (expect %zu)", dump_size,
+		   hdev->dump.alloc_size);
+
+	dev_coredumpv(&hdev->dev, hdev->dump.head, dump_size, GFP_KERNEL);
+}
+
+static void hci_devcd_handle_pkt_abort(struct hci_dev *hdev,
+				       struct sk_buff *skb)
+{
+	u32 dump_size;
+
+	if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) {
+		DBG_UNEXPECTED_STATE();
+		return;
+	}
+
+	hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_ABORT);
+	dump_size = hdev->dump.tail - hdev->dump.head;
+
+	bt_dev_dbg(hdev, "aborted with size %u (expect %zu)", dump_size,
+		   hdev->dump.alloc_size);
+
+	/* Emit a devcoredump with the available data */
+	dev_coredumpv(&hdev->dev, hdev->dump.head, dump_size, GFP_KERNEL);
+}
+
+/* Bluetooth devcoredump state machine.
+ *
+ * Devcoredump states:
+ *
+ *      HCI_DEVCOREDUMP_IDLE: The default state.
+ *
+ *      HCI_DEVCOREDUMP_ACTIVE: A devcoredump will be in this state once it has
+ *              been initialized using hci_devcd_init(). Once active, the driver
+ *              can append data using hci_devcd_append() or insert a pattern
+ *              using hci_devcd_append_pattern().
+ *
+ *      HCI_DEVCOREDUMP_DONE: Once the dump collection is complete, the drive
+ *              can signal the completion using hci_devcd_complete(). A
+ *              devcoredump is generated indicating the completion event and
+ *              then the state machine is reset to the default state.
+ *
+ *      HCI_DEVCOREDUMP_ABORT: The driver can cancel ongoing dump collection in
+ *              case of any error using hci_devcd_abort(). A devcoredump is
+ *              still generated with the available data indicating the abort
+ *              event and then the state machine is reset to the default state.
+ *
+ *      HCI_DEVCOREDUMP_TIMEOUT: A timeout timer for HCI_DEVCOREDUMP_TIMEOUT sec
+ *              is started during devcoredump initialization. Once the timeout
+ *              occurs, the driver is notified, a devcoredump is generated with
+ *              the available data indicating the timeout event and then the
+ *              state machine is reset to the default state.
+ *
+ * The driver must register using hci_devcd_register() before using the hci
+ * devcoredump APIs.
+ */
+void hci_devcd_rx(struct work_struct *work)
+{
+	struct hci_dev *hdev = container_of(work, struct hci_dev, dump.dump_rx);
+	struct sk_buff *skb;
+	int start_state;
+
+	while ((skb = skb_dequeue(&hdev->dump.dump_q))) {
+		/* Return if timeout occurs. The timeout handler function
+		 * hci_devcd_timeout() will report the available dump data.
+		 */
+		if (hdev->dump.state == HCI_DEVCOREDUMP_TIMEOUT) {
+			kfree_skb(skb);
+			return;
+		}
+
+		hci_dev_lock(hdev);
+		start_state = hdev->dump.state;
+
+		switch (hci_dmp_cb(skb)->pkt_type) {
+		case HCI_DEVCOREDUMP_PKT_INIT:
+			hci_devcd_handle_pkt_init(hdev, skb);
+			break;
+
+		case HCI_DEVCOREDUMP_PKT_SKB:
+			hci_devcd_handle_pkt_skb(hdev, skb);
+			break;
+
+		case HCI_DEVCOREDUMP_PKT_PATTERN:
+			hci_devcd_handle_pkt_pattern(hdev, skb);
+			break;
+
+		case HCI_DEVCOREDUMP_PKT_COMPLETE:
+			hci_devcd_handle_pkt_complete(hdev, skb);
+			break;
+
+		case HCI_DEVCOREDUMP_PKT_ABORT:
+			hci_devcd_handle_pkt_abort(hdev, skb);
+			break;
+
+		default:
+			bt_dev_dbg(hdev, "Unknown packet (%d) for state (%d). ",
+				   hci_dmp_cb(skb)->pkt_type, hdev->dump.state);
+			break;
+		}
+
+		hci_dev_unlock(hdev);
+		kfree_skb(skb);
+
+		/* Notify the driver about any state changes before resetting
+		 * the state machine
+		 */
+		if (start_state != hdev->dump.state)
+			hci_devcd_notify(hdev, hdev->dump.state);
+
+		/* Reset the state machine if the devcoredump is complete */
+		hci_dev_lock(hdev);
+		if (hdev->dump.state == HCI_DEVCOREDUMP_DONE ||
+		    hdev->dump.state == HCI_DEVCOREDUMP_ABORT)
+			hci_devcd_reset(hdev);
+		hci_dev_unlock(hdev);
+	}
+}
+EXPORT_SYMBOL(hci_devcd_rx);
+
+void hci_devcd_timeout(struct work_struct *work)
+{
+	struct hci_dev *hdev = container_of(work, struct hci_dev,
+					    dump.dump_timeout.work);
+	u32 dump_size;
+
+	hci_devcd_notify(hdev, HCI_DEVCOREDUMP_TIMEOUT);
+
+	hci_dev_lock(hdev);
+
+	cancel_work(&hdev->dump.dump_rx);
+
+	hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_TIMEOUT);
+
+	dump_size = hdev->dump.tail - hdev->dump.head;
+	bt_dev_dbg(hdev, "timeout with size %u (expect %zu)", dump_size,
+		   hdev->dump.alloc_size);
+
+	/* Emit a devcoredump with the available data */
+	dev_coredumpv(&hdev->dev, hdev->dump.head, dump_size, GFP_KERNEL);
+
+	hci_devcd_reset(hdev);
+
+	hci_dev_unlock(hdev);
+}
+EXPORT_SYMBOL(hci_devcd_timeout);
+
+int hci_devcd_register(struct hci_dev *hdev, coredump_t coredump,
+		       dmp_hdr_t dmp_hdr, notify_change_t notify_change)
+{
+	/* Driver must implement coredump() and dmp_hdr() functions for
+	 * bluetooth devcoredump. The coredump() should trigger a coredump
+	 * event on the controller when the device's coredump sysfs entry is
+	 * written to. The dmp_hdr() should create a dump header to identify
+	 * the controller/fw/driver info.
+	 */
+	if (!coredump || !dmp_hdr)
+		return -EINVAL;
+
+	hci_dev_lock(hdev);
+	hdev->dump.coredump = coredump;
+	hdev->dump.dmp_hdr = dmp_hdr;
+	hdev->dump.notify_change = notify_change;
+	hdev->dump.supported = true;
+	hdev->dump.timeout = DEVCOREDUMP_TIMEOUT;
+	hci_dev_unlock(hdev);
+
+	return 0;
+}
+EXPORT_SYMBOL(hci_devcd_register);
+
+static inline bool hci_devcd_enabled(struct hci_dev *hdev)
+{
+	return hdev->dump.supported;
+}
+
+int hci_devcd_init(struct hci_dev *hdev, u32 dump_size)
+{
+	struct sk_buff *skb;
+
+	if (!hci_devcd_enabled(hdev))
+		return -EOPNOTSUPP;
+
+	skb = alloc_skb(sizeof(dump_size), GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_INIT;
+	skb_put_data(skb, &dump_size, sizeof(dump_size));
+
+	skb_queue_tail(&hdev->dump.dump_q, skb);
+	queue_work(hdev->workqueue, &hdev->dump.dump_rx);
+
+	return 0;
+}
+EXPORT_SYMBOL(hci_devcd_init);
+
+int hci_devcd_append(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	if (!skb)
+		return -ENOMEM;
+
+	if (!hci_devcd_enabled(hdev)) {
+		kfree_skb(skb);
+		return -EOPNOTSUPP;
+	}
+
+	hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_SKB;
+
+	skb_queue_tail(&hdev->dump.dump_q, skb);
+	queue_work(hdev->workqueue, &hdev->dump.dump_rx);
+
+	return 0;
+}
+EXPORT_SYMBOL(hci_devcd_append);
+
+int hci_devcd_append_pattern(struct hci_dev *hdev, u8 pattern, u32 len)
+{
+	struct hci_devcoredump_skb_pattern p;
+	struct sk_buff *skb;
+
+	if (!hci_devcd_enabled(hdev))
+		return -EOPNOTSUPP;
+
+	skb = alloc_skb(sizeof(p), GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	p.pattern = pattern;
+	p.len = len;
+
+	hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_PATTERN;
+	skb_put_data(skb, &p, sizeof(p));
+
+	skb_queue_tail(&hdev->dump.dump_q, skb);
+	queue_work(hdev->workqueue, &hdev->dump.dump_rx);
+
+	return 0;
+}
+EXPORT_SYMBOL(hci_devcd_append_pattern);
+
+int hci_devcd_complete(struct hci_dev *hdev)
+{
+	struct sk_buff *skb;
+
+	if (!hci_devcd_enabled(hdev))
+		return -EOPNOTSUPP;
+
+	skb = alloc_skb(0, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_COMPLETE;
+
+	skb_queue_tail(&hdev->dump.dump_q, skb);
+	queue_work(hdev->workqueue, &hdev->dump.dump_rx);
+
+	return 0;
+}
+EXPORT_SYMBOL(hci_devcd_complete);
+
+int hci_devcd_abort(struct hci_dev *hdev)
+{
+	struct sk_buff *skb;
+
+	if (!hci_devcd_enabled(hdev))
+		return -EOPNOTSUPP;
+
+	skb = alloc_skb(0, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_ABORT;
+
+	skb_queue_tail(&hdev->dump.dump_q, skb);
+	queue_work(hdev->workqueue, &hdev->dump.dump_rx);
+
+	return 0;
+}
+EXPORT_SYMBOL(hci_devcd_abort);
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 334e308451f5..393b317ae68f 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -2544,6 +2544,7 @@ struct hci_dev *hci_alloc_dev_priv(int sizeof_priv)
 	INIT_DELAYED_WORK(&hdev->cmd_timer, hci_cmd_timeout);
 	INIT_DELAYED_WORK(&hdev->ncmd_timer, hci_ncmd_timeout);
 
+	hci_devcd_setup(hdev);
 	hci_request_setup(hdev);
 
 	hci_init_sysfs(hdev);
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 31231f0e4a28..f21497ebc814 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -4727,6 +4727,8 @@ int hci_dev_open_sync(struct hci_dev *hdev)
 		goto done;
 	}
 
+	hci_devcd_reset(hdev);
+
 	set_bit(HCI_RUNNING, &hdev->flags);
 	hci_sock_dev_event(hdev, HCI_DEV_OPEN);
 
-- 
cgit v1.2.3-70-g09d2


From ab4e4380d4e158486e595013a2635190e07e28ce Mon Sep 17 00:00:00 2001
From: Manish Mandlik <mmandlik@google.com>
Date: Thu, 30 Mar 2023 09:58:24 -0700
Subject: Bluetooth: Add vhci devcoredump support

Add devcoredump support for vhci that creates forcce_devcoredump debugfs
entry. This is used for mgmt-tester tests.

Signed-off-by: Manish Mandlik <mmandlik@google.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/Kconfig    |  1 +
 drivers/bluetooth/hci_vhci.c | 97 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 98 insertions(+)

diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index f84322d375de..bc211c324206 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -363,6 +363,7 @@ config BT_HCIBLUECARD
 
 config BT_HCIVHCI
 	tristate "HCI VHCI (Virtual HCI device) driver"
+	select WANT_DEV_COREDUMP
 	help
 	  Bluetooth Virtual HCI device driver.
 	  This driver is required if you want to use HCI Emulation software.
diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c
index c443c3b0a4da..691fe93b1976 100644
--- a/drivers/bluetooth/hci_vhci.c
+++ b/drivers/bluetooth/hci_vhci.c
@@ -278,6 +278,100 @@ static int vhci_setup(struct hci_dev *hdev)
 	return 0;
 }
 
+static void vhci_coredump(struct hci_dev *hdev)
+{
+	/* No need to do anything */
+}
+
+static void vhci_coredump_hdr(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	char buf[80];
+
+	snprintf(buf, sizeof(buf), "Controller Name: vhci_ctrl\n");
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Firmware Version: vhci_fw\n");
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Driver: vhci_drv\n");
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Vendor: vhci\n");
+	skb_put_data(skb, buf, strlen(buf));
+}
+
+#define MAX_COREDUMP_LINE_LEN	40
+
+struct devcoredump_test_data {
+	enum devcoredump_state state;
+	unsigned int timeout;
+	char data[MAX_COREDUMP_LINE_LEN];
+};
+
+static inline void force_devcd_timeout(struct hci_dev *hdev,
+				       unsigned int timeout)
+{
+#ifdef CONFIG_DEV_COREDUMP
+	hdev->dump.timeout = msecs_to_jiffies(timeout * 1000);
+#endif
+}
+
+static ssize_t force_devcd_write(struct file *file, const char __user *user_buf,
+				 size_t count, loff_t *ppos)
+{
+	struct vhci_data *data = file->private_data;
+	struct hci_dev *hdev = data->hdev;
+	struct sk_buff *skb = NULL;
+	struct devcoredump_test_data dump_data;
+	int ret;
+
+	ret = simple_write_to_buffer(&dump_data, sizeof(dump_data), ppos,
+				     user_buf, count);
+	if (ret < count)
+		return ret;
+
+	skb = alloc_skb(sizeof(dump_data.data), GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+	skb_put_data(skb, &dump_data.data, sizeof(dump_data.data));
+
+	hci_devcd_register(hdev, vhci_coredump, vhci_coredump_hdr, NULL);
+
+	/* Force the devcoredump timeout */
+	if (dump_data.timeout)
+		force_devcd_timeout(hdev, dump_data.timeout);
+
+	ret = hci_devcd_init(hdev, skb->len);
+	if (ret) {
+		BT_ERR("Failed to generate devcoredump");
+		kfree_skb(skb);
+		return ret;
+	}
+
+	hci_devcd_append(hdev, skb);
+
+	switch (dump_data.state) {
+	case HCI_DEVCOREDUMP_DONE:
+		hci_devcd_complete(hdev);
+		break;
+	case HCI_DEVCOREDUMP_ABORT:
+		hci_devcd_abort(hdev);
+		break;
+	case HCI_DEVCOREDUMP_TIMEOUT:
+		/* Do nothing */
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return count;
+}
+
+static const struct file_operations force_devcoredump_fops = {
+	.open		= simple_open,
+	.write		= force_devcd_write,
+};
+
 static int __vhci_create_device(struct vhci_data *data, __u8 opcode)
 {
 	struct hci_dev *hdev;
@@ -355,6 +449,9 @@ static int __vhci_create_device(struct vhci_data *data, __u8 opcode)
 		debugfs_create_file("aosp_capable", 0644, hdev->debugfs, data,
 				    &aosp_capable_fops);
 
+	debugfs_create_file("force_devcoredump", 0644, hdev->debugfs, data,
+			    &force_devcoredump_fops);
+
 	hci_skb_pkt_type(skb) = HCI_VENDOR_PKT;
 
 	skb_put_u8(skb, 0xff);
-- 
cgit v1.2.3-70-g09d2


From 4f9c1a089623ac606e79a7886646939e0e2f89bf Mon Sep 17 00:00:00 2001
From: Manish Mandlik <mmandlik@google.com>
Date: Thu, 30 Mar 2023 09:58:25 -0700
Subject: Bluetooth: btusb: Add btusb devcoredump support

This patch implements the btusb driver side .coredump() callback to
trigger a devcoredump via sysfs.

Signed-off-by: Manish Mandlik <mmandlik@google.com>
Reviewed-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 9bf1d4ee74c2..498e13102367 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -4397,6 +4397,17 @@ done:
 }
 #endif
 
+#ifdef CONFIG_DEV_COREDUMP
+static void btusb_coredump(struct device *dev)
+{
+	struct btusb_data *data = dev_get_drvdata(dev);
+	struct hci_dev *hdev = data->hdev;
+
+	if (hdev->dump.coredump)
+		hdev->dump.coredump(hdev);
+}
+#endif
+
 static struct usb_driver btusb_driver = {
 	.name		= "btusb",
 	.probe		= btusb_probe,
@@ -4408,6 +4419,14 @@ static struct usb_driver btusb_driver = {
 	.id_table	= btusb_table,
 	.supports_autosuspend = 1,
 	.disable_hub_initiated_lpm = 1,
+
+#ifdef CONFIG_DEV_COREDUMP
+	.drvwrap = {
+		.driver = {
+			.coredump = btusb_coredump,
+		},
+	},
+#endif
 };
 
 module_usb_driver(btusb_driver);
-- 
cgit v1.2.3-70-g09d2


From af395330abed142a2685bf3d17a938544816bf3c Mon Sep 17 00:00:00 2001
From: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Date: Thu, 30 Mar 2023 09:58:26 -0700
Subject: Bluetooth: btintel: Add Intel devcoredump support

Intercept debug exception events from the controller and put them into
a devcoredump using hci devcoredump APIs. The debug exception contains
data in a TLV format and it will be parsed in userspace.

Signed-off-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Signed-off-by: Manish Mandlik <mmandlik@google.com>
Reviewed-by: Abhishek Pandit-Subedi <abhishekpandit@chromium.org>
Reviewed-by: Chethan Tumkur Narayan <chethan.tumkur.narayan@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 72 ++++++++++++++++++++++++++++++++++++++++++++-
 drivers/bluetooth/btintel.h | 12 ++++++--
 drivers/bluetooth/btusb.c   | 54 +++++++++++++++++++++++++++++-----
 3 files changed, 128 insertions(+), 10 deletions(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 7a6dc05553f1..d9349ba48281 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -43,6 +43,12 @@ struct cmd_write_boot_params {
 	u8  fw_build_yy;
 } __packed;
 
+static struct {
+	const char *driver_name;
+	u8         hw_variant;
+	u32        fw_build_num;
+} coredump_info;
+
 int btintel_check_bdaddr(struct hci_dev *hdev)
 {
 	struct hci_rp_read_bd_addr *bda;
@@ -315,6 +321,9 @@ int btintel_version_info(struct hci_dev *hdev, struct intel_version *ver)
 		return -EINVAL;
 	}
 
+	coredump_info.hw_variant = ver->hw_variant;
+	coredump_info.fw_build_num = ver->fw_build_num;
+
 	bt_dev_info(hdev, "%s revision %u.%u build %u week %u %u",
 		    variant, ver->fw_revision >> 4, ver->fw_revision & 0x0f,
 		    ver->fw_build_num, ver->fw_build_ww,
@@ -509,6 +518,9 @@ static int btintel_version_info_tlv(struct hci_dev *hdev,
 		return -EINVAL;
 	}
 
+	coredump_info.hw_variant = INTEL_HW_VARIANT(version->cnvi_bt);
+	coredump_info.fw_build_num = version->build_num;
+
 	bt_dev_info(hdev, "%s timestamp %u.%u buildtype %u build %u", variant,
 		    2000 + (version->timestamp >> 8), version->timestamp & 0xff,
 		    version->build_type, version->build_num);
@@ -1462,6 +1474,59 @@ int btintel_set_quality_report(struct hci_dev *hdev, bool enable)
 }
 EXPORT_SYMBOL_GPL(btintel_set_quality_report);
 
+static void btintel_coredump(struct hci_dev *hdev)
+{
+	struct sk_buff *skb;
+
+	skb = __hci_cmd_sync(hdev, 0xfc4e, 0, NULL, HCI_CMD_TIMEOUT);
+	if (IS_ERR(skb)) {
+		bt_dev_err(hdev, "Coredump failed (%ld)", PTR_ERR(skb));
+		return;
+	}
+
+	kfree_skb(skb);
+}
+
+static void btintel_dmp_hdr(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	char buf[80];
+
+	snprintf(buf, sizeof(buf), "Controller Name: 0x%X\n",
+		 coredump_info.hw_variant);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Firmware Version: 0x%X\n",
+		 coredump_info.fw_build_num);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Driver: %s\n", coredump_info.driver_name);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Vendor: Intel\n");
+	skb_put_data(skb, buf, strlen(buf));
+}
+
+static int btintel_register_devcoredump_support(struct hci_dev *hdev)
+{
+	struct intel_debug_features features;
+	int err;
+
+	err = btintel_read_debug_features(hdev, &features);
+	if (err) {
+		bt_dev_info(hdev, "Error reading debug features");
+		return err;
+	}
+
+	if (!(features.page1[0] & 0x3f)) {
+		bt_dev_dbg(hdev, "Telemetry exception format not supported");
+		return -EOPNOTSUPP;
+	}
+
+	hci_devcd_register(hdev, btintel_coredump, btintel_dmp_hdr, NULL);
+
+	return err;
+}
+
 static const struct firmware *btintel_legacy_rom_get_fw(struct hci_dev *hdev,
 					       struct intel_version *ver)
 {
@@ -2597,6 +2662,7 @@ static int btintel_setup_combined(struct hci_dev *hdev)
 			btintel_set_msft_opcode(hdev, ver.hw_variant);
 
 			err = btintel_bootloader_setup(hdev, &ver);
+			btintel_register_devcoredump_support(hdev);
 			break;
 		default:
 			bt_dev_err(hdev, "Unsupported Intel hw variant (%u)",
@@ -2670,6 +2736,7 @@ static int btintel_setup_combined(struct hci_dev *hdev)
 		btintel_set_msft_opcode(hdev, ver.hw_variant);
 
 		err = btintel_bootloader_setup(hdev, &ver);
+		btintel_register_devcoredump_support(hdev);
 		break;
 	case 0x17:
 	case 0x18:
@@ -2692,6 +2759,7 @@ static int btintel_setup_combined(struct hci_dev *hdev)
 					INTEL_HW_VARIANT(ver_tlv.cnvi_bt));
 
 		err = btintel_bootloader_setup_tlv(hdev, &ver_tlv);
+		btintel_register_devcoredump_support(hdev);
 		break;
 	default:
 		bt_dev_err(hdev, "Unsupported Intel hw variant (%u)",
@@ -2741,7 +2809,7 @@ static int btintel_shutdown_combined(struct hci_dev *hdev)
 	return 0;
 }
 
-int btintel_configure_setup(struct hci_dev *hdev)
+int btintel_configure_setup(struct hci_dev *hdev, const char *driver_name)
 {
 	hdev->manufacturer = 2;
 	hdev->setup = btintel_setup_combined;
@@ -2750,6 +2818,8 @@ int btintel_configure_setup(struct hci_dev *hdev)
 	hdev->set_diag = btintel_set_diag_combined;
 	hdev->set_bdaddr = btintel_set_bdaddr;
 
+	coredump_info.driver_name = driver_name;
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(btintel_configure_setup);
diff --git a/drivers/bluetooth/btintel.h b/drivers/bluetooth/btintel.h
index 8fdb65b66315..d6a1dc8d8a82 100644
--- a/drivers/bluetooth/btintel.h
+++ b/drivers/bluetooth/btintel.h
@@ -143,6 +143,13 @@ struct btintel_loc_aware_reg {
 	__le32 delta;
 } __packed;
 
+#define INTEL_TLV_TYPE_ID		0x01
+
+#define INTEL_TLV_SYSTEM_EXCEPTION	0x00
+#define INTEL_TLV_FATAL_EXCEPTION	0x01
+#define INTEL_TLV_DEBUG_EXCEPTION	0x02
+#define INTEL_TLV_TEST_EXCEPTION	0xDE
+
 #define INTEL_HW_PLATFORM(cnvx_bt)	((u8)(((cnvx_bt) & 0x0000ff00) >> 8))
 #define INTEL_HW_VARIANT(cnvx_bt)	((u8)(((cnvx_bt) & 0x003f0000) >> 16))
 #define INTEL_CNVX_TOP_TYPE(cnvx_top)	((cnvx_top) & 0x00000fff)
@@ -212,7 +219,7 @@ int btintel_read_boot_params(struct hci_dev *hdev,
 			     struct intel_boot_params *params);
 int btintel_download_firmware(struct hci_dev *dev, struct intel_version *ver,
 			      const struct firmware *fw, u32 *boot_param);
-int btintel_configure_setup(struct hci_dev *hdev);
+int btintel_configure_setup(struct hci_dev *hdev, const char *driver_name);
 void btintel_bootup(struct hci_dev *hdev, const void *ptr, unsigned int len);
 void btintel_secure_send_result(struct hci_dev *hdev,
 				const void *ptr, unsigned int len);
@@ -293,7 +300,8 @@ static inline int btintel_download_firmware(struct hci_dev *dev,
 	return -EOPNOTSUPP;
 }
 
-static inline int btintel_configure_setup(struct hci_dev *hdev)
+static inline int btintel_configure_setup(struct hci_dev *hdev,
+					  const char *driver_name)
 {
 	return -ENODEV;
 }
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 498e13102367..2303b0a66323 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2386,16 +2386,47 @@ static int btusb_recv_bulk_intel(struct btusb_data *data, void *buffer,
 	return btusb_recv_bulk(data, buffer, count);
 }
 
+static int btusb_intel_diagnostics(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct intel_tlv *tlv = (void *)&skb->data[5];
+
+	/* The first event is always an event type TLV */
+	if (tlv->type != INTEL_TLV_TYPE_ID)
+		goto recv_frame;
+
+	switch (tlv->val[0]) {
+	case INTEL_TLV_SYSTEM_EXCEPTION:
+	case INTEL_TLV_FATAL_EXCEPTION:
+	case INTEL_TLV_DEBUG_EXCEPTION:
+	case INTEL_TLV_TEST_EXCEPTION:
+		/* Generate devcoredump from exception */
+		if (!hci_devcd_init(hdev, skb->len)) {
+			hci_devcd_append(hdev, skb);
+			hci_devcd_complete(hdev);
+		} else {
+			bt_dev_err(hdev, "Failed to generate devcoredump");
+			kfree_skb(skb);
+		}
+		return 0;
+	default:
+		bt_dev_err(hdev, "Invalid exception type %02X", tlv->val[0]);
+	}
+
+recv_frame:
+	return hci_recv_frame(hdev, skb);
+}
+
 static int btusb_recv_event_intel(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	if (btintel_test_flag(hdev, INTEL_BOOTLOADER)) {
-		struct hci_event_hdr *hdr = (void *)skb->data;
+	struct hci_event_hdr *hdr = (void *)skb->data;
+	const char diagnostics_hdr[] = { 0x87, 0x80, 0x03 };
 
-		if (skb->len > HCI_EVENT_HDR_SIZE && hdr->evt == 0xff &&
-		    hdr->plen > 0) {
-			const void *ptr = skb->data + HCI_EVENT_HDR_SIZE + 1;
-			unsigned int len = skb->len - HCI_EVENT_HDR_SIZE - 1;
+	if (skb->len > HCI_EVENT_HDR_SIZE && hdr->evt == 0xff &&
+	    hdr->plen > 0) {
+		const void *ptr = skb->data + HCI_EVENT_HDR_SIZE + 1;
+		unsigned int len = skb->len - HCI_EVENT_HDR_SIZE - 1;
 
+		if (btintel_test_flag(hdev, INTEL_BOOTLOADER)) {
 			switch (skb->data[2]) {
 			case 0x02:
 				/* When switching to the operational firmware
@@ -2414,6 +2445,15 @@ static int btusb_recv_event_intel(struct hci_dev *hdev, struct sk_buff *skb)
 				break;
 			}
 		}
+
+		/* Handle all diagnostics events separately. May still call
+		 * hci_recv_frame.
+		 */
+		if (len >= sizeof(diagnostics_hdr) &&
+		    memcmp(&skb->data[2], diagnostics_hdr,
+			   sizeof(diagnostics_hdr)) == 0) {
+			return btusb_intel_diagnostics(hdev, skb);
+		}
 	}
 
 	return hci_recv_frame(hdev, skb);
@@ -4018,7 +4058,7 @@ static int btusb_probe(struct usb_interface *intf,
 
 	/* Combined Intel Device setup to support multiple setup routine */
 	if (id->driver_info & BTUSB_INTEL_COMBINED) {
-		err = btintel_configure_setup(hdev);
+		err = btintel_configure_setup(hdev, btusb_driver.name);
 		if (err)
 			goto out_free_dev;
 
-- 
cgit v1.2.3-70-g09d2


From 0fe8c8d071343fa9278980ce4b6f8e6ea24a2ed1 Mon Sep 17 00:00:00 2001
From: Iulia Tanasescu <iulia.tanasescu@nxp.com>
Date: Fri, 31 Mar 2023 18:38:01 +0300
Subject: Bluetooth: Split bt_iso_qos into dedicated structures

Split bt_iso_qos into dedicated unicast and broadcast
structures and add additional broadcast parameters.

Fixes: eca0ae4aea66 ("Bluetooth: Add initial implementation of BIS connections")
Signed-off-by: Iulia Tanasescu <iulia.tanasescu@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/bluetooth.h |  43 +++++++---
 include/net/bluetooth/hci_core.h  |   9 ++-
 net/bluetooth/hci_conn.c          | 162 ++++++++++++++++++++------------------
 net/bluetooth/hci_event.c         |  33 ++++----
 net/bluetooth/iso.c               | 125 +++++++++++++++++++++++------
 5 files changed, 237 insertions(+), 135 deletions(-)

diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h
index bcc5a4cd2c17..1b4230cd42a3 100644
--- a/include/net/bluetooth/bluetooth.h
+++ b/include/net/bluetooth/bluetooth.h
@@ -1,6 +1,7 @@
 /*
    BlueZ - Bluetooth protocol stack for Linux
    Copyright (C) 2000-2001 Qualcomm Incorporated
+   Copyright 2023 NXP
 
    Written 2000,2001 by Maxim Krasnyansky <maxk@qualcomm.com>
 
@@ -171,23 +172,39 @@ struct bt_iso_io_qos {
 	__u8  rtn;
 };
 
-struct bt_iso_qos {
-	union {
-		__u8  cig;
-		__u8  big;
-	};
-	union {
-		__u8  cis;
-		__u8  bis;
-	};
-	union {
-		__u8  sca;
-		__u8  sync_interval;
-	};
+struct bt_iso_ucast_qos {
+	__u8  cig;
+	__u8  cis;
+	__u8  sca;
+	__u8  packing;
+	__u8  framing;
+	struct bt_iso_io_qos in;
+	struct bt_iso_io_qos out;
+};
+
+struct bt_iso_bcast_qos {
+	__u8  big;
+	__u8  bis;
+	__u8  sync_interval;
 	__u8  packing;
 	__u8  framing;
 	struct bt_iso_io_qos in;
 	struct bt_iso_io_qos out;
+	__u8  encryption;
+	__u8  bcode[16];
+	__u8  options;
+	__u16 skip;
+	__u16 sync_timeout;
+	__u8  sync_cte_type;
+	__u8  mse;
+	__u16 timeout;
+};
+
+struct bt_iso_qos {
+	union {
+		struct bt_iso_ucast_qos ucast;
+		struct bt_iso_bcast_qos bcast;
+	};
 };
 
 #define BT_ISO_PHY_1M		0x01
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index a461335beacf..f11689284112 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -1,6 +1,7 @@
 /*
    BlueZ - Bluetooth protocol stack for Linux
    Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved.
+   Copyright 2023 NXP
 
    Written 2000,2001 by Maxim Krasnyansky <maxk@qualcomm.com>
 
@@ -1096,7 +1097,7 @@ static inline struct hci_conn *hci_conn_hash_lookup_bis(struct hci_dev *hdev,
 		if (bacmp(&c->dst, ba) || c->type != ISO_LINK)
 			continue;
 
-		if (c->iso_qos.big == big && c->iso_qos.bis == bis) {
+		if (c->iso_qos.bcast.big == big && c->iso_qos.bcast.bis == bis) {
 			rcu_read_unlock();
 			return c;
 		}
@@ -1205,7 +1206,7 @@ static inline struct hci_conn *hci_conn_hash_lookup_cig(struct hci_dev *hdev,
 		if (c->type != ISO_LINK)
 			continue;
 
-		if (handle == c->iso_qos.cig) {
+		if (handle == c->iso_qos.ucast.cig) {
 			rcu_read_unlock();
 			return c;
 		}
@@ -1228,7 +1229,7 @@ static inline struct hci_conn *hci_conn_hash_lookup_big(struct hci_dev *hdev,
 		if (bacmp(&c->dst, BDADDR_ANY) || c->type != ISO_LINK)
 			continue;
 
-		if (handle == c->iso_qos.big) {
+		if (handle == c->iso_qos.bcast.big) {
 			rcu_read_unlock();
 			return c;
 		}
@@ -1337,7 +1338,7 @@ struct hci_conn *hci_connect_bis(struct hci_dev *hdev, bdaddr_t *dst,
 				 __u8 dst_type, struct bt_iso_qos *qos,
 				 __u8 data_len, __u8 *data);
 int hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type,
-		       __u8 sid);
+		       __u8 sid, struct bt_iso_qos *qos);
 int hci_le_big_create_sync(struct hci_dev *hdev, struct bt_iso_qos *qos,
 			   __u16 sync_handle, __u8 num_bis, __u8 bis[]);
 int hci_conn_check_link_mode(struct hci_conn *conn);
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 8455ba141ee6..5672b4924572 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -1,6 +1,7 @@
 /*
    BlueZ - Bluetooth protocol stack for Linux
    Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved.
+   Copyright 2023 NXP
 
    Written 2000,2001 by Maxim Krasnyansky <maxk@qualcomm.com>
 
@@ -795,8 +796,8 @@ static void bis_list(struct hci_conn *conn, void *data)
 	if (bacmp(&conn->dst, BDADDR_ANY))
 		return;
 
-	if (d->big != conn->iso_qos.big || d->bis == BT_ISO_QOS_BIS_UNSET ||
-	    d->bis != conn->iso_qos.bis)
+	if (d->big != conn->iso_qos.bcast.big || d->bis == BT_ISO_QOS_BIS_UNSET ||
+	    d->bis != conn->iso_qos.bcast.bis)
 		return;
 
 	d->count++;
@@ -916,10 +917,10 @@ static void bis_cleanup(struct hci_conn *conn)
 		if (!test_and_clear_bit(HCI_CONN_PER_ADV, &conn->flags))
 			return;
 
-		hci_le_terminate_big(hdev, conn->iso_qos.big,
-				     conn->iso_qos.bis);
+		hci_le_terminate_big(hdev, conn->iso_qos.bcast.big,
+				     conn->iso_qos.bcast.bis);
 	} else {
-		hci_le_big_terminate(hdev, conn->iso_qos.big,
+		hci_le_big_terminate(hdev, conn->iso_qos.bcast.big,
 				     conn->sync_handle);
 	}
 }
@@ -959,7 +960,7 @@ static void cis_cleanup(struct hci_conn *conn)
 	struct iso_list_data d;
 
 	memset(&d, 0, sizeof(d));
-	d.cig = conn->iso_qos.cig;
+	d.cig = conn->iso_qos.ucast.cig;
 
 	/* Check if ISO connection is a CIS and remove CIG if there are
 	 * no other connections using it.
@@ -968,7 +969,7 @@ static void cis_cleanup(struct hci_conn *conn)
 	if (d.count)
 		return;
 
-	hci_le_remove_cig(hdev, conn->iso_qos.cig);
+	hci_le_remove_cig(hdev, conn->iso_qos.ucast.cig);
 }
 
 struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst,
@@ -1411,7 +1412,7 @@ static int qos_set_big(struct hci_dev *hdev, struct bt_iso_qos *qos)
 	struct iso_list_data data;
 
 	/* Allocate a BIG if not set */
-	if (qos->big == BT_ISO_QOS_BIG_UNSET) {
+	if (qos->bcast.big == BT_ISO_QOS_BIG_UNSET) {
 		for (data.big = 0x00; data.big < 0xef; data.big++) {
 			data.count = 0;
 			data.bis = 0xff;
@@ -1426,7 +1427,7 @@ static int qos_set_big(struct hci_dev *hdev, struct bt_iso_qos *qos)
 			return -EADDRNOTAVAIL;
 
 		/* Update BIG */
-		qos->big = data.big;
+		qos->bcast.big = data.big;
 	}
 
 	return 0;
@@ -1437,7 +1438,7 @@ static int qos_set_bis(struct hci_dev *hdev, struct bt_iso_qos *qos)
 	struct iso_list_data data;
 
 	/* Allocate BIS if not set */
-	if (qos->bis == BT_ISO_QOS_BIS_UNSET) {
+	if (qos->bcast.bis == BT_ISO_QOS_BIS_UNSET) {
 		/* Find an unused adv set to advertise BIS, skip instance 0x00
 		 * since it is reserved as general purpose set.
 		 */
@@ -1455,7 +1456,7 @@ static int qos_set_bis(struct hci_dev *hdev, struct bt_iso_qos *qos)
 			return -EADDRNOTAVAIL;
 
 		/* Update BIS */
-		qos->bis = data.bis;
+		qos->bcast.bis = data.bis;
 	}
 
 	return 0;
@@ -1484,8 +1485,8 @@ static struct hci_conn *hci_add_bis(struct hci_dev *hdev, bdaddr_t *dst,
 	if (err)
 		return ERR_PTR(err);
 
-	data.big = qos->big;
-	data.bis = qos->bis;
+	data.big = qos->bcast.big;
+	data.bis = qos->bcast.bis;
 	data.count = 0;
 
 	/* Check if there is already a matching BIG/BIS */
@@ -1493,7 +1494,7 @@ static struct hci_conn *hci_add_bis(struct hci_dev *hdev, bdaddr_t *dst,
 	if (data.count)
 		return ERR_PTR(-EADDRINUSE);
 
-	conn = hci_conn_hash_lookup_bis(hdev, dst, qos->big, qos->bis);
+	conn = hci_conn_hash_lookup_bis(hdev, dst, qos->bcast.big, qos->bcast.bis);
 	if (conn)
 		return ERR_PTR(-EADDRINUSE);
 
@@ -1648,13 +1649,13 @@ static void cis_add(struct iso_list_data *d, struct bt_iso_qos *qos)
 {
 	struct hci_cis_params *cis = &d->pdu.cis[d->pdu.cp.num_cis];
 
-	cis->cis_id = qos->cis;
-	cis->c_sdu  = cpu_to_le16(qos->out.sdu);
-	cis->p_sdu  = cpu_to_le16(qos->in.sdu);
-	cis->c_phy  = qos->out.phy ? qos->out.phy : qos->in.phy;
-	cis->p_phy  = qos->in.phy ? qos->in.phy : qos->out.phy;
-	cis->c_rtn  = qos->out.rtn;
-	cis->p_rtn  = qos->in.rtn;
+	cis->cis_id = qos->ucast.cis;
+	cis->c_sdu  = cpu_to_le16(qos->ucast.out.sdu);
+	cis->p_sdu  = cpu_to_le16(qos->ucast.in.sdu);
+	cis->c_phy  = qos->ucast.out.phy ? qos->ucast.out.phy : qos->ucast.in.phy;
+	cis->p_phy  = qos->ucast.in.phy ? qos->ucast.in.phy : qos->ucast.out.phy;
+	cis->c_rtn  = qos->ucast.out.rtn;
+	cis->p_rtn  = qos->ucast.in.rtn;
 
 	d->pdu.cp.num_cis++;
 }
@@ -1667,8 +1668,8 @@ static void cis_list(struct hci_conn *conn, void *data)
 	if (!bacmp(&conn->dst, BDADDR_ANY))
 		return;
 
-	if (d->cig != conn->iso_qos.cig || d->cis == BT_ISO_QOS_CIS_UNSET ||
-	    d->cis != conn->iso_qos.cis)
+	if (d->cig != conn->iso_qos.ucast.cig || d->cis == BT_ISO_QOS_CIS_UNSET ||
+	    d->cis != conn->iso_qos.ucast.cis)
 		return;
 
 	d->count++;
@@ -1687,17 +1688,18 @@ static int hci_le_create_big(struct hci_conn *conn, struct bt_iso_qos *qos)
 
 	memset(&cp, 0, sizeof(cp));
 
-	cp.handle = qos->big;
-	cp.adv_handle = qos->bis;
+	cp.handle = qos->bcast.big;
+	cp.adv_handle = qos->bcast.bis;
 	cp.num_bis  = 0x01;
-	hci_cpu_to_le24(qos->out.interval, cp.bis.sdu_interval);
-	cp.bis.sdu = cpu_to_le16(qos->out.sdu);
-	cp.bis.latency =  cpu_to_le16(qos->out.latency);
-	cp.bis.rtn  = qos->out.rtn;
-	cp.bis.phy  = qos->out.phy;
-	cp.bis.packing = qos->packing;
-	cp.bis.framing = qos->framing;
-	cp.bis.encryption = 0x00;
+	hci_cpu_to_le24(qos->bcast.out.interval, cp.bis.sdu_interval);
+	cp.bis.sdu = cpu_to_le16(qos->bcast.out.sdu);
+	cp.bis.latency =  cpu_to_le16(qos->bcast.out.latency);
+	cp.bis.rtn  = qos->bcast.out.rtn;
+	cp.bis.phy  = qos->bcast.out.phy;
+	cp.bis.packing = qos->bcast.packing;
+	cp.bis.framing = qos->bcast.framing;
+	cp.bis.encryption = qos->bcast.encryption;
+	memcpy(cp.bis.bcode, qos->bcast.bcode, sizeof(cp.bis.bcode));
 	memset(&cp.bis.bcode, 0, sizeof(cp.bis.bcode));
 
 	return hci_send_cmd(hdev, HCI_OP_LE_CREATE_BIG, sizeof(cp), &cp);
@@ -1711,7 +1713,7 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos)
 	memset(&data, 0, sizeof(data));
 
 	/* Allocate a CIG if not set */
-	if (qos->cig == BT_ISO_QOS_CIG_UNSET) {
+	if (qos->ucast.cig == BT_ISO_QOS_CIG_UNSET) {
 		for (data.cig = 0x00; data.cig < 0xff; data.cig++) {
 			data.count = 0;
 			data.cis = 0xff;
@@ -1731,22 +1733,22 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos)
 			return false;
 
 		/* Update CIG */
-		qos->cig = data.cig;
+		qos->ucast.cig = data.cig;
 	}
 
-	data.pdu.cp.cig_id = qos->cig;
-	hci_cpu_to_le24(qos->out.interval, data.pdu.cp.c_interval);
-	hci_cpu_to_le24(qos->in.interval, data.pdu.cp.p_interval);
-	data.pdu.cp.sca = qos->sca;
-	data.pdu.cp.packing = qos->packing;
-	data.pdu.cp.framing = qos->framing;
-	data.pdu.cp.c_latency = cpu_to_le16(qos->out.latency);
-	data.pdu.cp.p_latency = cpu_to_le16(qos->in.latency);
+	data.pdu.cp.cig_id = qos->ucast.cig;
+	hci_cpu_to_le24(qos->ucast.out.interval, data.pdu.cp.c_interval);
+	hci_cpu_to_le24(qos->ucast.in.interval, data.pdu.cp.p_interval);
+	data.pdu.cp.sca = qos->ucast.sca;
+	data.pdu.cp.packing = qos->ucast.packing;
+	data.pdu.cp.framing = qos->ucast.framing;
+	data.pdu.cp.c_latency = cpu_to_le16(qos->ucast.out.latency);
+	data.pdu.cp.p_latency = cpu_to_le16(qos->ucast.in.latency);
 
-	if (qos->cis != BT_ISO_QOS_CIS_UNSET) {
+	if (qos->ucast.cis != BT_ISO_QOS_CIS_UNSET) {
 		data.count = 0;
-		data.cig = qos->cig;
-		data.cis = qos->cis;
+		data.cig = qos->ucast.cig;
+		data.cis = qos->ucast.cis;
 
 		hci_conn_hash_list_state(hdev, cis_list, ISO_LINK, BT_BOUND,
 					 &data);
@@ -1757,7 +1759,7 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos)
 	}
 
 	/* Reprogram all CIS(s) with the same CIG */
-	for (data.cig = qos->cig, data.cis = 0x00; data.cis < 0x11;
+	for (data.cig = qos->ucast.cig, data.cis = 0x00; data.cis < 0x11;
 	     data.cis++) {
 		data.count = 0;
 
@@ -1767,14 +1769,14 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos)
 			continue;
 
 		/* Allocate a CIS if not set */
-		if (qos->cis == BT_ISO_QOS_CIS_UNSET) {
+		if (qos->ucast.cis == BT_ISO_QOS_CIS_UNSET) {
 			/* Update CIS */
-			qos->cis = data.cis;
+			qos->ucast.cis = data.cis;
 			cis_add(&data, qos);
 		}
 	}
 
-	if (qos->cis == BT_ISO_QOS_CIS_UNSET || !data.pdu.cp.num_cis)
+	if (qos->ucast.cis == BT_ISO_QOS_CIS_UNSET || !data.pdu.cp.num_cis)
 		return false;
 
 	if (hci_send_cmd(hdev, HCI_OP_LE_SET_CIG_PARAMS,
@@ -1809,32 +1811,32 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst,
 		return cis;
 
 	/* Update LINK PHYs according to QoS preference */
-	cis->le_tx_phy = qos->out.phy;
-	cis->le_rx_phy = qos->in.phy;
+	cis->le_tx_phy = qos->ucast.out.phy;
+	cis->le_rx_phy = qos->ucast.in.phy;
 
 	/* If output interval is not set use the input interval as it cannot be
 	 * 0x000000.
 	 */
-	if (!qos->out.interval)
-		qos->out.interval = qos->in.interval;
+	if (!qos->ucast.out.interval)
+		qos->ucast.out.interval = qos->ucast.in.interval;
 
 	/* If input interval is not set use the output interval as it cannot be
 	 * 0x000000.
 	 */
-	if (!qos->in.interval)
-		qos->in.interval = qos->out.interval;
+	if (!qos->ucast.in.interval)
+		qos->ucast.in.interval = qos->ucast.out.interval;
 
 	/* If output latency is not set use the input latency as it cannot be
 	 * 0x0000.
 	 */
-	if (!qos->out.latency)
-		qos->out.latency = qos->in.latency;
+	if (!qos->ucast.out.latency)
+		qos->ucast.out.latency = qos->ucast.in.latency;
 
 	/* If input latency is not set use the output latency as it cannot be
 	 * 0x0000.
 	 */
-	if (!qos->in.latency)
-		qos->in.latency = qos->out.latency;
+	if (!qos->ucast.in.latency)
+		qos->ucast.in.latency = qos->ucast.out.latency;
 
 	if (!hci_le_set_cig_params(cis, qos)) {
 		hci_conn_drop(cis);
@@ -1854,7 +1856,7 @@ bool hci_iso_setup_path(struct hci_conn *conn)
 
 	memset(&cmd, 0, sizeof(cmd));
 
-	if (conn->iso_qos.out.sdu) {
+	if (conn->iso_qos.ucast.out.sdu) {
 		cmd.handle = cpu_to_le16(conn->handle);
 		cmd.direction = 0x00; /* Input (Host to Controller) */
 		cmd.path = 0x00; /* HCI path if enabled */
@@ -1865,7 +1867,7 @@ bool hci_iso_setup_path(struct hci_conn *conn)
 			return false;
 	}
 
-	if (conn->iso_qos.in.sdu) {
+	if (conn->iso_qos.ucast.in.sdu) {
 		cmd.handle = cpu_to_le16(conn->handle);
 		cmd.direction = 0x01; /* Output (Controller to Host) */
 		cmd.path = 0x00; /* HCI path if enabled */
@@ -1892,7 +1894,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 	cmd.cis[0].acl_handle = cpu_to_le16(conn->link->handle);
 	cmd.cis[0].cis_handle = cpu_to_le16(conn->handle);
 	cmd.cp.num_cis++;
-	cig = conn->iso_qos.cig;
+	cig = conn->iso_qos.ucast.cig;
 
 	hci_dev_lock(hdev);
 
@@ -1902,7 +1904,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 		struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis];
 
 		if (conn == data || conn->type != ISO_LINK ||
-		    conn->state == BT_CONNECTED || conn->iso_qos.cig != cig)
+		    conn->state == BT_CONNECTED || conn->iso_qos.ucast.cig != cig)
 			continue;
 
 		/* Check if all CIS(s) belonging to a CIG are ready */
@@ -2002,8 +2004,8 @@ static void hci_bind_bis(struct hci_conn *conn,
 			 struct bt_iso_qos *qos)
 {
 	/* Update LINK PHYs according to QoS preference */
-	conn->le_tx_phy = qos->out.phy;
-	conn->le_tx_phy = qos->out.phy;
+	conn->le_tx_phy = qos->bcast.out.phy;
+	conn->le_tx_phy = qos->bcast.out.phy;
 	conn->iso_qos = *qos;
 	conn->state = BT_BOUND;
 }
@@ -2016,16 +2018,16 @@ static int create_big_sync(struct hci_dev *hdev, void *data)
 	u32 flags = 0;
 	int err;
 
-	if (qos->out.phy == 0x02)
+	if (qos->bcast.out.phy == 0x02)
 		flags |= MGMT_ADV_FLAG_SEC_2M;
 
 	/* Align intervals */
-	interval = qos->out.interval / 1250;
+	interval = qos->bcast.out.interval / 1250;
 
-	if (qos->bis)
-		sync_interval = qos->sync_interval * 1600;
+	if (qos->bcast.bis)
+		sync_interval = qos->bcast.sync_interval * 1600;
 
-	err = hci_start_per_adv_sync(hdev, qos->bis, conn->le_per_adv_data_len,
+	err = hci_start_per_adv_sync(hdev, qos->bcast.bis, conn->le_per_adv_data_len,
 				     conn->le_per_adv_data, flags, interval,
 				     interval, sync_interval);
 	if (err)
@@ -2062,7 +2064,7 @@ static int create_pa_sync(struct hci_dev *hdev, void *data)
 }
 
 int hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type,
-		       __u8 sid)
+		       __u8 sid, struct bt_iso_qos *qos)
 {
 	struct hci_cp_le_pa_create_sync *cp;
 
@@ -2075,9 +2077,13 @@ int hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type,
 		return -ENOMEM;
 	}
 
+	cp->options = qos->bcast.options;
 	cp->sid = sid;
 	cp->addr_type = dst_type;
 	bacpy(&cp->addr, dst);
+	cp->skip = cpu_to_le16(qos->bcast.skip);
+	cp->sync_timeout = cpu_to_le16(qos->bcast.sync_timeout);
+	cp->sync_cte_type = qos->bcast.sync_cte_type;
 
 	/* Queue start pa_create_sync and scan */
 	return hci_cmd_sync_queue(hdev, create_pa_sync, cp, create_pa_complete);
@@ -2100,8 +2106,12 @@ int hci_le_big_create_sync(struct hci_dev *hdev, struct bt_iso_qos *qos,
 		return err;
 
 	memset(&pdu, 0, sizeof(pdu));
-	pdu.cp.handle = qos->big;
+	pdu.cp.handle = qos->bcast.big;
 	pdu.cp.sync_handle = cpu_to_le16(sync_handle);
+	pdu.cp.encryption = qos->bcast.encryption;
+	memcpy(pdu.cp.bcode, qos->bcast.bcode, sizeof(pdu.cp.bcode));
+	pdu.cp.mse = qos->bcast.mse;
+	pdu.cp.timeout = cpu_to_le16(qos->bcast.timeout);
 	pdu.cp.num_bis = num_bis;
 	memcpy(pdu.bis, bis, num_bis);
 
@@ -2151,7 +2161,7 @@ struct hci_conn *hci_connect_bis(struct hci_dev *hdev, bdaddr_t *dst,
 		return ERR_PTR(err);
 	}
 
-	hci_iso_qos_setup(hdev, conn, &qos->out,
+	hci_iso_qos_setup(hdev, conn, &qos->bcast.out,
 			  conn->le_tx_phy ? conn->le_tx_phy :
 			  hdev->le_tx_def_phys);
 
@@ -2177,9 +2187,9 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst,
 	if (IS_ERR(le))
 		return le;
 
-	hci_iso_qos_setup(hdev, le, &qos->out,
+	hci_iso_qos_setup(hdev, le, &qos->ucast.out,
 			  le->le_tx_phy ? le->le_tx_phy : hdev->le_tx_def_phys);
-	hci_iso_qos_setup(hdev, le, &qos->in,
+	hci_iso_qos_setup(hdev, le, &qos->ucast.in,
 			  le->le_rx_phy ? le->le_rx_phy : hdev->le_rx_def_phys);
 
 	cis = hci_bind_cis(hdev, dst, dst_type, qos);
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 51f13518dba9..0e0a93cc1218 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -1,6 +1,7 @@
 /*
    BlueZ - Bluetooth protocol stack for Linux
    Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved.
+   Copyright 2023 NXP
 
    Written 2000,2001 by Maxim Krasnyansky <maxk@qualcomm.com>
 
@@ -3833,7 +3834,7 @@ static u8 hci_cc_le_set_cig_params(struct hci_dev *hdev, void *data,
 	rcu_read_lock();
 
 	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		if (conn->type != ISO_LINK || conn->iso_qos.cig != rp->cig_id ||
+		if (conn->type != ISO_LINK || conn->iso_qos.ucast.cig != rp->cig_id ||
 		    conn->state == BT_CONNECTED)
 			continue;
 
@@ -3890,7 +3891,7 @@ static u8 hci_cc_le_setup_iso_path(struct hci_dev *hdev, void *data,
 	/* Input (Host to Controller) */
 	case 0x00:
 		/* Only confirm connection if output only */
-		if (conn->iso_qos.out.sdu && !conn->iso_qos.in.sdu)
+		if (conn->iso_qos.ucast.out.sdu && !conn->iso_qos.ucast.in.sdu)
 			hci_connect_cfm(conn, rp->status);
 		break;
 	/* Output (Controller to Host) */
@@ -6818,15 +6819,15 @@ static void hci_le_cis_estabilished_evt(struct hci_dev *hdev, void *data,
 		memset(&interval, 0, sizeof(interval));
 
 		memcpy(&interval, ev->c_latency, sizeof(ev->c_latency));
-		conn->iso_qos.in.interval = le32_to_cpu(interval);
+		conn->iso_qos.ucast.in.interval = le32_to_cpu(interval);
 		memcpy(&interval, ev->p_latency, sizeof(ev->p_latency));
-		conn->iso_qos.out.interval = le32_to_cpu(interval);
-		conn->iso_qos.in.latency = le16_to_cpu(ev->interval);
-		conn->iso_qos.out.latency = le16_to_cpu(ev->interval);
-		conn->iso_qos.in.sdu = le16_to_cpu(ev->c_mtu);
-		conn->iso_qos.out.sdu = le16_to_cpu(ev->p_mtu);
-		conn->iso_qos.in.phy = ev->c_phy;
-		conn->iso_qos.out.phy = ev->p_phy;
+		conn->iso_qos.ucast.out.interval = le32_to_cpu(interval);
+		conn->iso_qos.ucast.in.latency = le16_to_cpu(ev->interval);
+		conn->iso_qos.ucast.out.latency = le16_to_cpu(ev->interval);
+		conn->iso_qos.ucast.in.sdu = le16_to_cpu(ev->c_mtu);
+		conn->iso_qos.ucast.out.sdu = le16_to_cpu(ev->p_mtu);
+		conn->iso_qos.ucast.in.phy = ev->c_phy;
+		conn->iso_qos.ucast.out.phy = ev->p_phy;
 	}
 
 	if (!ev->status) {
@@ -6900,8 +6901,8 @@ static void hci_le_cis_req_evt(struct hci_dev *hdev, void *data,
 		cis->handle = cis_handle;
 	}
 
-	cis->iso_qos.cig = ev->cig_id;
-	cis->iso_qos.cis = ev->cis_id;
+	cis->iso_qos.ucast.cig = ev->cig_id;
+	cis->iso_qos.ucast.cis = ev->cis_id;
 
 	if (!(flags & HCI_PROTO_DEFER)) {
 		hci_le_accept_cis(hdev, ev->cis_handle);
@@ -6988,13 +6989,13 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data,
 			bis->handle = handle;
 		}
 
-		bis->iso_qos.big = ev->handle;
+		bis->iso_qos.bcast.big = ev->handle;
 		memset(&interval, 0, sizeof(interval));
 		memcpy(&interval, ev->latency, sizeof(ev->latency));
-		bis->iso_qos.in.interval = le32_to_cpu(interval);
+		bis->iso_qos.bcast.in.interval = le32_to_cpu(interval);
 		/* Convert ISO Interval (1.25 ms slots) to latency (ms) */
-		bis->iso_qos.in.latency = le16_to_cpu(ev->interval) * 125 / 100;
-		bis->iso_qos.in.sdu = le16_to_cpu(ev->max_pdu);
+		bis->iso_qos.bcast.in.latency = le16_to_cpu(ev->interval) * 125 / 100;
+		bis->iso_qos.bcast.in.sdu = le16_to_cpu(ev->max_pdu);
 
 		hci_iso_setup_path(bis);
 	}
diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c
index 8d136a730163..74117df03a3f 100644
--- a/net/bluetooth/iso.c
+++ b/net/bluetooth/iso.c
@@ -3,6 +3,7 @@
  * BlueZ - Bluetooth protocol stack for Linux
  *
  * Copyright (C) 2022 Intel Corporation
+ * Copyright 2023 NXP
  */
 
 #include <linux/module.h>
@@ -59,11 +60,17 @@ struct iso_pinfo {
 	__u16			sync_handle;
 	__u32			flags;
 	struct bt_iso_qos	qos;
+	bool			qos_user_set;
 	__u8			base_len;
 	__u8			base[BASE_MAX_LENGTH];
 	struct iso_conn		*conn;
 };
 
+static struct bt_iso_qos default_qos;
+
+static bool check_ucast_qos(struct bt_iso_qos *qos);
+static bool check_bcast_qos(struct bt_iso_qos *qos);
+
 /* ---- ISO timers ---- */
 #define ISO_CONN_TIMEOUT	(HZ * 40)
 #define ISO_DISCONN_TIMEOUT	(HZ * 2)
@@ -264,8 +271,15 @@ static int iso_connect_bis(struct sock *sk)
 		goto unlock;
 	}
 
+	/* Fail if user set invalid QoS */
+	if (iso_pi(sk)->qos_user_set && !check_bcast_qos(&iso_pi(sk)->qos)) {
+		iso_pi(sk)->qos = default_qos;
+		err = -EINVAL;
+		goto unlock;
+	}
+
 	/* Fail if out PHYs are marked as disabled */
-	if (!iso_pi(sk)->qos.out.phy) {
+	if (!iso_pi(sk)->qos.bcast.out.phy) {
 		err = -EINVAL;
 		goto unlock;
 	}
@@ -336,8 +350,15 @@ static int iso_connect_cis(struct sock *sk)
 		goto unlock;
 	}
 
+	/* Fail if user set invalid QoS */
+	if (iso_pi(sk)->qos_user_set && !check_ucast_qos(&iso_pi(sk)->qos)) {
+		iso_pi(sk)->qos = default_qos;
+		err = -EINVAL;
+		goto unlock;
+	}
+
 	/* Fail if either PHYs are marked as disabled */
-	if (!iso_pi(sk)->qos.in.phy && !iso_pi(sk)->qos.out.phy) {
+	if (!iso_pi(sk)->qos.ucast.in.phy && !iso_pi(sk)->qos.ucast.out.phy) {
 		err = -EINVAL;
 		goto unlock;
 	}
@@ -417,7 +438,7 @@ static int iso_send_frame(struct sock *sk, struct sk_buff *skb)
 
 	BT_DBG("sk %p len %d", sk, skb->len);
 
-	if (skb->len > qos->out.sdu)
+	if (skb->len > qos->ucast.out.sdu)
 		return -EMSGSIZE;
 
 	len = skb->len;
@@ -680,13 +701,23 @@ static struct proto iso_proto = {
 }
 
 static struct bt_iso_qos default_qos = {
-	.cig		= BT_ISO_QOS_CIG_UNSET,
-	.cis		= BT_ISO_QOS_CIS_UNSET,
-	.sca		= 0x00,
-	.packing	= 0x00,
-	.framing	= 0x00,
-	.in		= DEFAULT_IO_QOS,
-	.out		= DEFAULT_IO_QOS,
+	.bcast = {
+		.big			= BT_ISO_QOS_BIG_UNSET,
+		.bis			= BT_ISO_QOS_BIS_UNSET,
+		.sync_interval		= 0x00,
+		.packing		= 0x00,
+		.framing		= 0x00,
+		.in			= DEFAULT_IO_QOS,
+		.out			= DEFAULT_IO_QOS,
+		.encryption		= 0x00,
+		.bcode			= {0x00},
+		.options		= 0x00,
+		.skip			= 0x0000,
+		.sync_timeout		= 0x4000,
+		.sync_cte_type		= 0x00,
+		.mse			= 0x00,
+		.timeout		= 0x4000,
+	},
 };
 
 static struct sock *iso_sock_alloc(struct net *net, struct socket *sock,
@@ -893,9 +924,15 @@ static int iso_listen_bis(struct sock *sk)
 	if (!hdev)
 		return -EHOSTUNREACH;
 
+	/* Fail if user set invalid QoS */
+	if (iso_pi(sk)->qos_user_set && !check_bcast_qos(&iso_pi(sk)->qos)) {
+		iso_pi(sk)->qos = default_qos;
+		return -EINVAL;
+	}
+
 	err = hci_pa_create_sync(hdev, &iso_pi(sk)->dst,
 				 le_addr_type(iso_pi(sk)->dst_type),
-				 iso_pi(sk)->bc_sid);
+				 iso_pi(sk)->bc_sid, &iso_pi(sk)->qos);
 
 	hci_dev_put(hdev);
 
@@ -1154,21 +1191,62 @@ static bool check_io_qos(struct bt_iso_io_qos *qos)
 	return true;
 }
 
-static bool check_qos(struct bt_iso_qos *qos)
+static bool check_ucast_qos(struct bt_iso_qos *qos)
 {
-	if (qos->sca > 0x07)
+	if (qos->ucast.sca > 0x07)
 		return false;
 
-	if (qos->packing > 0x01)
+	if (qos->ucast.packing > 0x01)
 		return false;
 
-	if (qos->framing > 0x01)
+	if (qos->ucast.framing > 0x01)
 		return false;
 
-	if (!check_io_qos(&qos->in))
+	if (!check_io_qos(&qos->ucast.in))
 		return false;
 
-	if (!check_io_qos(&qos->out))
+	if (!check_io_qos(&qos->ucast.out))
+		return false;
+
+	return true;
+}
+
+static bool check_bcast_qos(struct bt_iso_qos *qos)
+{
+	if (qos->bcast.sync_interval > 0x07)
+		return false;
+
+	if (qos->bcast.packing > 0x01)
+		return false;
+
+	if (qos->bcast.framing > 0x01)
+		return false;
+
+	if (!check_io_qos(&qos->bcast.in))
+		return false;
+
+	if (!check_io_qos(&qos->bcast.out))
+		return false;
+
+	if (qos->bcast.encryption > 0x01)
+		return false;
+
+	if (qos->bcast.options > 0x07)
+		return false;
+
+	if (qos->bcast.skip > 0x01f3)
+		return false;
+
+	if (qos->bcast.sync_timeout < 0x000a || qos->bcast.sync_timeout > 0x4000)
+		return false;
+
+	if (qos->bcast.sync_cte_type > 0x1f)
+		return false;
+
+	if (qos->bcast.mse > 0x1f)
+		return false;
+
+	if (qos->bcast.timeout < 0x000a || qos->bcast.timeout > 0x4000)
 		return false;
 
 	return true;
@@ -1179,7 +1257,7 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname,
 {
 	struct sock *sk = sock->sk;
 	int len, err = 0;
-	struct bt_iso_qos qos;
+	struct bt_iso_qos qos = default_qos;
 	u32 opt;
 
 	BT_DBG("sk %p", sk);
@@ -1212,24 +1290,19 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname,
 		}
 
 		len = min_t(unsigned int, sizeof(qos), optlen);
-		if (len != sizeof(qos)) {
-			err = -EINVAL;
-			break;
-		}
-
-		memset(&qos, 0, sizeof(qos));
 
 		if (copy_from_sockptr(&qos, optval, len)) {
 			err = -EFAULT;
 			break;
 		}
 
-		if (!check_qos(&qos)) {
+		if (len == sizeof(qos.ucast) && !check_ucast_qos(&qos)) {
 			err = -EINVAL;
 			break;
 		}
 
 		iso_pi(sk)->qos = qos;
+		iso_pi(sk)->qos_user_set = true;
 
 		break;
 
@@ -1419,7 +1492,7 @@ static bool iso_match_big(struct sock *sk, void *data)
 {
 	struct hci_evt_le_big_sync_estabilished *ev = data;
 
-	return ev->handle == iso_pi(sk)->qos.big;
+	return ev->handle == iso_pi(sk)->qos.bcast.big;
 }
 
 static void iso_conn_ready(struct iso_conn *conn)
-- 
cgit v1.2.3-70-g09d2


From 288c90224eec55d13e786844b7954ef060752089 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 19 Dec 2022 13:37:02 -0800
Subject: Bluetooth: Enable all supported LE PHY by default

This enables 2M and Coded PHY by default if they are marked as supported
in the LE features bits.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h |  4 ++++
 net/bluetooth/hci_sync.c         | 28 ++++++++++++++++++++++++----
 net/bluetooth/mgmt.c             |  4 ++--
 3 files changed, 30 insertions(+), 6 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index f11689284112..827e67159523 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -1683,9 +1683,13 @@ void hci_conn_del_sysfs(struct hci_conn *conn);
 #define scan_1m(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_1M) || \
 		      ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_1M))
 
+#define le_2m_capable(dev) (((dev)->le_features[1] & HCI_LE_PHY_2M))
+
 #define scan_2m(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_2M) || \
 		      ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_2M))
 
+#define le_coded_capable(dev) (((dev)->le_features[1] & HCI_LE_PHY_CODED))
+
 #define scan_coded(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_CODED) || \
 			 ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_CODED))
 
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index f21497ebc814..00017f75cd41 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -4414,18 +4414,38 @@ static int hci_le_set_write_def_data_len_sync(struct hci_dev *hdev)
 				     sizeof(cp), &cp, HCI_CMD_TIMEOUT);
 }
 
-/* Set Default PHY parameters if command is supported */
+/* Set Default PHY parameters if command is supported, enables all supported
+ * PHYs according to the LE Features bits.
+ */
 static int hci_le_set_default_phy_sync(struct hci_dev *hdev)
 {
 	struct hci_cp_le_set_default_phy cp;
 
-	if (!(hdev->commands[35] & 0x20))
+	if (!(hdev->commands[35] & 0x20)) {
+		/* If the command is not supported it means only 1M PHY is
+		 * supported.
+		 */
+		hdev->le_tx_def_phys = HCI_LE_SET_PHY_1M;
+		hdev->le_rx_def_phys = HCI_LE_SET_PHY_1M;
 		return 0;
+	}
 
 	memset(&cp, 0, sizeof(cp));
 	cp.all_phys = 0x00;
-	cp.tx_phys = hdev->le_tx_def_phys;
-	cp.rx_phys = hdev->le_rx_def_phys;
+	cp.tx_phys = HCI_LE_SET_PHY_1M;
+	cp.rx_phys = HCI_LE_SET_PHY_1M;
+
+	/* Enables 2M PHY if supported */
+	if (le_2m_capable(hdev)) {
+		cp.tx_phys |= HCI_LE_SET_PHY_2M;
+		cp.rx_phys |= HCI_LE_SET_PHY_2M;
+	}
+
+	/* Enables Coded PHY if supported */
+	if (le_coded_capable(hdev)) {
+		cp.tx_phys |= HCI_LE_SET_PHY_CODED;
+		cp.rx_phys |= HCI_LE_SET_PHY_CODED;
+	}
 
 	return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_DEFAULT_PHY,
 				     sizeof(cp), &cp, HCI_CMD_TIMEOUT);
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 249dc6777fb4..5f8c144c84b8 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -8393,10 +8393,10 @@ static u32 get_supported_adv_flags(struct hci_dev *hdev)
 		flags |= MGMT_ADV_FLAG_HW_OFFLOAD;
 		flags |= MGMT_ADV_FLAG_CAN_SET_TX_POWER;
 
-		if (hdev->le_features[1] & HCI_LE_PHY_2M)
+		if (le_2m_capable(hdev))
 			flags |= MGMT_ADV_FLAG_SEC_2M;
 
-		if (hdev->le_features[1] & HCI_LE_PHY_CODED)
+		if (le_coded_capable(hdev))
 			flags |= MGMT_ADV_FLAG_SEC_CODED;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 3c0d41f141c97e516d986f0571b0d7a745a4e2a8 Mon Sep 17 00:00:00 2001
From: Qiqi Zhang <eddy.zhang@rock-chips.com>
Date: Tue, 4 Apr 2023 10:27:11 +0800
Subject: Bluetooth: hci_h5: Complements reliable packet processing logic

As shown in the schematic diagram below.There may be a critical
scenario in the current code. If the device does not receive an
pure ack sent by the host due to insufficient receive buffer or
other reasons and triggers a retransmission, the host will always
be in an 'out-of-order' state.The state machine will get stuck.

       host                 device
     SEQ3,ACK4 --------->
               <--------- SEQ4,ACK4
     pure ACK  ---------> (not received)
(out-of-order) <--------- SEQ4,ACK4(retransmission)
                ........
(out-of-order) <--------- SEQ4,ACK4(retransmission)

According to the description in the core specification: "whenever
a reliable packet is received, an acknowledgment shall be generated."
So set H5_TX_ACK_REQ bit to trigger retransmission of pure ack packet
when "out-of-order" occurs.

Signed-off-by: Qiqi Zhang <eddy.zhang@rock-chips.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_h5.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/hci_h5.c b/drivers/bluetooth/hci_h5.c
index e90670955df2..fefc37b98b4a 100644
--- a/drivers/bluetooth/hci_h5.c
+++ b/drivers/bluetooth/hci_h5.c
@@ -463,6 +463,8 @@ static int h5_rx_3wire_hdr(struct hci_uart *hu, unsigned char c)
 	if (H5_HDR_RELIABLE(hdr) && H5_HDR_SEQ(hdr) != h5->tx_ack) {
 		bt_dev_err(hu->hdev, "Out-of-order packet arrived (%u != %u)",
 			   H5_HDR_SEQ(hdr), h5->tx_ack);
+		set_bit(H5_TX_ACK_REQ, &h5->flags);
+		hci_uart_tx_wakeup(hu);
 		h5_reset_rx(h5);
 		return 0;
 	}
-- 
cgit v1.2.3-70-g09d2


From 0d218c3642b9ccf71f44987cd03c19320f3bd918 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 31 Mar 2023 23:11:21 +0200
Subject: Bluetooth: hci_bcm: Fall back to getting bdaddr from EFI if not set

On some devices the BCM Bluetooth adapter does not have a valid bdaddr set.

btbcm.c currently sets HCI_QUIRK_INVALID_BDADDR to indicate when this is
the case. But this requires users to manual setup a btaddr, by doing e.g.:

btmgmt -i hci0 public-addr 'B0:F1:EC:82:1D:B3'

Which means that Bluetooth will not work out of the box on such devices.
To avoid this (where possible) hci_bcm sets: HCI_QUIRK_USE_BDADDR_PROPERTY
which tries to get the bdaddr from devicetree.

But this only works on devicetree platforms. On UEFI based platforms
there is a special Broadcom UEFI variable which when present contains
the devices bdaddr, just like how there is another UEFI variable which
contains wifi nvram contents including the wifi MAC address.

Add support for getting the bdaddr from this Broadcom UEFI variable,
so that Bluetooth will work OOTB for users on devices where this
UEFI variable is present.

This fixes Bluetooth not working on for example Asus T100HA 2-in-1s.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btbcm.c | 47 ++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 44 insertions(+), 3 deletions(-)

diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c
index 43e98a598bd9..de2ea589aa49 100644
--- a/drivers/bluetooth/btbcm.c
+++ b/drivers/bluetooth/btbcm.c
@@ -6,6 +6,7 @@
  *  Copyright (C) 2015  Intel Corporation
  */
 
+#include <linux/efi.h>
 #include <linux/module.h>
 #include <linux/firmware.h>
 #include <linux/dmi.h>
@@ -34,6 +35,43 @@
 /* For kmalloc-ing the fw-name array instead of putting it on the stack */
 typedef char bcm_fw_name[BCM_FW_NAME_LEN];
 
+#ifdef CONFIG_EFI
+static int btbcm_set_bdaddr_from_efi(struct hci_dev *hdev)
+{
+	efi_guid_t guid = EFI_GUID(0x74b00bd9, 0x805a, 0x4d61, 0xb5, 0x1f,
+				   0x43, 0x26, 0x81, 0x23, 0xd1, 0x13);
+	bdaddr_t efi_bdaddr, bdaddr;
+	efi_status_t status;
+	unsigned long len;
+	int ret;
+
+	if (!efi_rt_services_supported(EFI_RT_SUPPORTED_GET_VARIABLE))
+		return -EOPNOTSUPP;
+
+	len = sizeof(efi_bdaddr);
+	status = efi.get_variable(L"BDADDR", &guid, NULL, &len, &efi_bdaddr);
+	if (status != EFI_SUCCESS)
+		return -ENXIO;
+
+	if (len != sizeof(efi_bdaddr))
+		return -EIO;
+
+	baswap(&bdaddr, &efi_bdaddr);
+
+	ret = btbcm_set_bdaddr(hdev, &bdaddr);
+	if (ret)
+		return ret;
+
+	bt_dev_info(hdev, "BCM: Using EFI device address (%pMR)", &bdaddr);
+	return 0;
+}
+#else
+static int btbcm_set_bdaddr_from_efi(struct hci_dev *hdev)
+{
+	return -EOPNOTSUPP;
+}
+#endif
+
 int btbcm_check_bdaddr(struct hci_dev *hdev)
 {
 	struct hci_rp_read_bd_addr *bda;
@@ -87,9 +125,12 @@ int btbcm_check_bdaddr(struct hci_dev *hdev)
 	    !bacmp(&bda->bdaddr, BDADDR_BCM4345C5) ||
 	    !bacmp(&bda->bdaddr, BDADDR_BCM43430A0) ||
 	    !bacmp(&bda->bdaddr, BDADDR_BCM43341B)) {
-		bt_dev_info(hdev, "BCM: Using default device address (%pMR)",
-			    &bda->bdaddr);
-		set_bit(HCI_QUIRK_INVALID_BDADDR, &hdev->quirks);
+		/* Try falling back to BDADDR EFI variable */
+		if (btbcm_set_bdaddr_from_efi(hdev) != 0) {
+			bt_dev_info(hdev, "BCM: Using default device address (%pMR)",
+				    &bda->bdaddr);
+			set_bit(HCI_QUIRK_INVALID_BDADDR, &hdev->quirks);
+		}
 	}
 
 	kfree_skb(skb);
-- 
cgit v1.2.3-70-g09d2


From ce439473acdf0a3aecbda616a0f92973ab1c2e52 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 31 Mar 2023 23:11:22 +0200
Subject: Bluetooth: hci_bcm: Limit bcm43430a0 / bcm43430a1 baudrate to 2000000

The bcm43430a0 and bcm43430a1 BT does not support the 0xfc45 command
to set the UART clock to 48 MHz and it also does not work at 4000000
baud without this command as some newer models do.

These chips are found on ACPI/x86 devices where the operating baudrate
does not come from the firmware but is hardcoded at 4000000, which does
not work.

Add a max_baudrate value to struct bcm_device_data and set this
to 2000000 on all known ACPI hardware-ids for the bcm43430a0
and the bcm43430a1.

Note this also adds the BCM2E9F ACPI HID which was missing until now.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_bcm.c | 37 ++++++++++++++++++++++++-------------
 1 file changed, 24 insertions(+), 13 deletions(-)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index 2b6c0e1922cb..4a3ef844df5d 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -55,12 +55,14 @@
  * @drive_rts_on_open: drive RTS signal on ->open() when platform requires it
  * @no_uart_clock_set: UART clock set command for >3Mbps mode is unavailable
  * @max_autobaud_speed: max baudrate supported by device in autobaud mode
+ * @max_speed: max baudrate supported
  */
 struct bcm_device_data {
 	bool	no_early_set_baudrate;
 	bool	drive_rts_on_open;
 	bool	no_uart_clock_set;
 	u32	max_autobaud_speed;
+	u32	max_speed;
 };
 
 /**
@@ -1300,6 +1302,12 @@ static const struct hci_uart_proto bcm_proto = {
 };
 
 #ifdef CONFIG_ACPI
+
+/* bcm43430a0/a1 BT does not support 48MHz UART clock, limit to 2000000 baud */
+static struct bcm_device_data bcm43430_device_data = {
+	.max_speed = 2000000,
+};
+
 static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E00" },
 	{ "BCM2E01" },
@@ -1414,19 +1422,19 @@ static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E71" },
 	{ "BCM2E72" },
 	{ "BCM2E73" },
-	{ "BCM2E74" },
-	{ "BCM2E75" },
+	{ "BCM2E74", (long)&bcm43430_device_data },
+	{ "BCM2E75", (long)&bcm43430_device_data },
 	{ "BCM2E76" },
 	{ "BCM2E77" },
 	{ "BCM2E78" },
 	{ "BCM2E79" },
 	{ "BCM2E7A" },
-	{ "BCM2E7B" },
+	{ "BCM2E7B", (long)&bcm43430_device_data },
 	{ "BCM2E7C" },
 	{ "BCM2E7D" },
 	{ "BCM2E7E" },
 	{ "BCM2E7F" },
-	{ "BCM2E80" },
+	{ "BCM2E80", (long)&bcm43430_device_data },
 	{ "BCM2E81" },
 	{ "BCM2E82" },
 	{ "BCM2E83" },
@@ -1435,7 +1443,7 @@ static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E86" },
 	{ "BCM2E87" },
 	{ "BCM2E88" },
-	{ "BCM2E89" },
+	{ "BCM2E89", (long)&bcm43430_device_data },
 	{ "BCM2E8A" },
 	{ "BCM2E8B" },
 	{ "BCM2E8C" },
@@ -1444,29 +1452,30 @@ static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E90" },
 	{ "BCM2E92" },
 	{ "BCM2E93" },
-	{ "BCM2E94" },
+	{ "BCM2E94", (long)&bcm43430_device_data },
 	{ "BCM2E95" },
 	{ "BCM2E96" },
 	{ "BCM2E97" },
 	{ "BCM2E98" },
-	{ "BCM2E99" },
+	{ "BCM2E99", (long)&bcm43430_device_data },
 	{ "BCM2E9A" },
-	{ "BCM2E9B" },
+	{ "BCM2E9B", (long)&bcm43430_device_data },
 	{ "BCM2E9C" },
 	{ "BCM2E9D" },
+	{ "BCM2E9F", (long)&bcm43430_device_data },
 	{ "BCM2EA0" },
 	{ "BCM2EA1" },
-	{ "BCM2EA2" },
-	{ "BCM2EA3" },
+	{ "BCM2EA2", (long)&bcm43430_device_data },
+	{ "BCM2EA3", (long)&bcm43430_device_data },
 	{ "BCM2EA4" },
 	{ "BCM2EA5" },
 	{ "BCM2EA6" },
 	{ "BCM2EA7" },
 	{ "BCM2EA8" },
 	{ "BCM2EA9" },
-	{ "BCM2EAA" },
-	{ "BCM2EAB" },
-	{ "BCM2EAC" },
+	{ "BCM2EAA", (long)&bcm43430_device_data },
+	{ "BCM2EAB", (long)&bcm43430_device_data },
+	{ "BCM2EAC", (long)&bcm43430_device_data },
 	{ },
 };
 MODULE_DEVICE_TABLE(acpi, bcm_acpi_match);
@@ -1535,6 +1544,8 @@ static int bcm_serdev_probe(struct serdev_device *serdev)
 		bcmdev->no_early_set_baudrate = data->no_early_set_baudrate;
 		bcmdev->drive_rts_on_open = data->drive_rts_on_open;
 		bcmdev->no_uart_clock_set = data->no_uart_clock_set;
+		if (data->max_speed && bcmdev->oper_speed > data->max_speed)
+			bcmdev->oper_speed = data->max_speed;
 	}
 
 	return hci_uart_register_device(&bcmdev->serdev_hu, &bcm_proto);
-- 
cgit v1.2.3-70-g09d2


From 9a09460274d311466b44743c281c81f36c3e4df4 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 31 Mar 2023 23:11:23 +0200
Subject: Bluetooth: hci_bcm: Add Lenovo Yoga Tablet 2 830 / 1050 to the
 bcm_broken_irq_dmi_table

The DSDT for the Lenovo Yoga Tablet 2 830 / 1050 models (which share
the same mainboard) specifies a IOAPIC IRQ for the HCI -> host IRQ but
this is not correct.

Like the Asus TF103C these tablets use pin 17 of the INT33FC:02 GPIO
controller for the IRQ and this pin is _not_ configured in direct IRQ
mode by the firmware.

Add a DMI match for this, re-using the Asus TF103C gpiod_lookup_table,
to fix bluetooth not working on these tablets.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_bcm.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index 4a3ef844df5d..eb9e1c858e91 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -890,7 +890,7 @@ unlock:
 #endif
 
 /* Some firmware reports an IRQ which does not work (wrong pin in fw table?) */
-static struct gpiod_lookup_table asus_tf103c_irq_gpios = {
+static struct gpiod_lookup_table irq_on_int33fc02_pin17_gpios = {
 	.dev_id = "serial0-0",
 	.table = {
 		GPIO_LOOKUP("INT33FC:02", 17, "host-wakeup-alt", GPIO_ACTIVE_HIGH),
@@ -905,7 +905,18 @@ static const struct dmi_system_id bcm_broken_irq_dmi_table[] = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
 			DMI_MATCH(DMI_PRODUCT_NAME, "TF103C"),
 		},
-		.driver_data = &asus_tf103c_irq_gpios,
+		.driver_data = &irq_on_int33fc02_pin17_gpios,
+	},
+	{
+		.ident = "Lenovo Yoga Tablet 2 830F/L / 1050F/L",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Intel Corp."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "VALLEYVIEW C0 PLATFORM"),
+			DMI_MATCH(DMI_BOARD_NAME, "BYT-T FFD8"),
+			/* Partial match on beginning of BIOS version */
+			DMI_MATCH(DMI_BIOS_VERSION, "BLADE_21"),
+		},
+		.driver_data = &irq_on_int33fc02_pin17_gpios,
 	},
 	{
 		.ident = "Meegopad T08",
-- 
cgit v1.2.3-70-g09d2


From 09df5a91e620a2d21f7f410938e5e28364fae0bf Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 31 Mar 2023 23:11:24 +0200
Subject: Bluetooth: hci_bcm: Add Acer Iconia One 7 B1-750 to the
 bcm_broken_irq_dmi_table

The DSDT for the Acer Iconia One 7 B1-750 models (which share
the same mainboard) specifies a IOAPIC IRQ for the HCI -> host IRQ but
this is not correct.

Like the Asus TF103C these tablets use pin 17 of the INT33FC:02 GPIO
controller for the IRQ and this pin is _not_ configured in direct IRQ
mode by the firmware.

Add a DMI match for this, re-using the Asus TF103C gpiod_lookup_table,
to fix bluetooth not working on these tablets.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_bcm.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index eb9e1c858e91..83bf5d4330c4 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -899,6 +899,14 @@ static struct gpiod_lookup_table irq_on_int33fc02_pin17_gpios = {
 };
 
 static const struct dmi_system_id bcm_broken_irq_dmi_table[] = {
+	{
+		.ident = "Acer Iconia One 7 B1-750",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Insyde"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "VESPA2"),
+		},
+		.driver_data = &irq_on_int33fc02_pin17_gpios,
+	},
 	{
 		.ident = "Asus TF103C",
 		.matches = {
-- 
cgit v1.2.3-70-g09d2


From 38a4f83d802caf5e7ec93a187d211d87b6d67800 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Mon, 3 Apr 2023 17:54:27 +0530
Subject: Bluetooth: btnxpuart: Add support to download helper FW file for
 w8997

This adds support to download helper FW file for the legacy NXP chipset
88w8997 for the btnxpuart driver. This helper FW file is necessary to
set the bootloader baudrate to 3000000 after which the actual BT FW file
can be downloaded. This change helps bring the FW download time from
around 10 sec to less than 2 sec for 88w8997 chip. For newer chipsets,
both V1 and V3 bootloader, driver sends the cmd5 and cmd7 to the chip
bootloader, and does not need a helper FW file.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 67 +++++++++++++++++++++++++++++++++++++------
 1 file changed, 59 insertions(+), 8 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 5f641466d695..32e22044b6b2 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -34,6 +34,7 @@
 #define FIRMWARE_W9098	"nxp/uartuart9098_bt_v1.bin"
 #define FIRMWARE_IW416	"nxp/uartiw416_bt_v0.bin"
 #define FIRMWARE_IW612	"nxp/uartspi_n61x_v1.bin.se"
+#define FIRMWARE_HELPER	"nxp/helper_uart_3000000.bin"
 
 #define CHIP_ID_W9098		0x5c03
 #define CHIP_ID_IW416		0x7201
@@ -123,7 +124,7 @@ struct psmode_cmd_payload {
 } __packed;
 
 struct btnxpuart_data {
-	bool fw_download_3M_baudrate;
+	const char *helper_fw_name;
 	const char *fw_name;
 };
 
@@ -150,6 +151,7 @@ struct btnxpuart_dev {
 	u32 fw_init_baudrate;
 	bool timeout_changed;
 	bool baudrate_changed;
+	bool helper_downloaded;
 
 	struct ps_data psdata;
 	struct btnxpuart_data *nxp_data;
@@ -168,6 +170,13 @@ struct btnxpuart_dev {
 
 #define HDR_LEN			16
 
+#define NXP_RECV_CHIP_VER_V1 \
+	.type = NXP_V1_CHIP_VER_PKT, \
+	.hlen = 4, \
+	.loff = 0, \
+	.lsize = 0, \
+	.maxlen = 4
+
 #define NXP_RECV_FW_REQ_V1 \
 	.type = NXP_V1_FW_REQ_PKT, \
 	.hlen = 4, \
@@ -194,6 +203,11 @@ struct v1_data_req {
 	__le16 len_comp;
 } __packed;
 
+struct v1_start_ind {
+	__le16 chip_id;
+	__le16 chip_id_comp;
+} __packed;
+
 struct v3_data_req {
 	__le16 len;
 	__le32 offset;
@@ -518,6 +532,7 @@ static int nxp_download_firmware(struct hci_dev *hdev)
 	nxpdev->fw_v3_offset_correction = 0;
 	nxpdev->baudrate_changed = false;
 	nxpdev->timeout_changed = false;
+	nxpdev->helper_downloaded = false;
 
 	serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_PRI_BAUDRATE);
 	serdev_device_set_flow_control(nxpdev->serdev, 0);
@@ -664,6 +679,29 @@ static int nxp_request_firmware(struct hci_dev *hdev, const char *fw_name)
 }
 
 /* for legacy chipsets with V1 bootloader */
+static int nxp_recv_chip_ver_v1(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	struct v1_start_ind *req;
+
+	req = (struct v1_start_ind *)skb_pull_data(skb, sizeof(struct v1_start_ind));
+	if (!req)
+		goto free_skb;
+
+	if ((req->chip_id ^ req->chip_id_comp) == 0xffff) {
+		nxpdev->fw_dnld_v1_offset = 0;
+		nxpdev->fw_v1_sent_bytes = 0;
+		nxpdev->fw_v1_expected_len = HDR_LEN;
+		release_firmware(nxpdev->fw);
+		memset(nxpdev->fw_name, 0, sizeof(nxpdev->fw_name));
+		nxp_send_ack(NXP_ACK_V1, hdev);
+	}
+
+free_skb:
+	kfree_skb(skb);
+	return 0;
+}
+
 static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
@@ -685,7 +723,7 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 	}
 	nxp_send_ack(NXP_ACK_V1, hdev);
 
-	if (nxp_data->fw_download_3M_baudrate) {
+	if (!nxp_data->helper_fw_name) {
 		if (!nxpdev->timeout_changed) {
 			nxpdev->timeout_changed = nxp_fw_change_timeout(hdev, req->len);
 			goto free_skb;
@@ -702,14 +740,26 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 		}
 	}
 
-	if (nxp_request_firmware(hdev, nxp_data->fw_name))
-		goto free_skb;
+	if (!nxp_data->helper_fw_name || nxpdev->helper_downloaded) {
+		if (nxp_request_firmware(hdev, nxp_data->fw_name))
+			goto free_skb;
+	} else if (nxp_data->helper_fw_name && !nxpdev->helper_downloaded) {
+		if (nxp_request_firmware(hdev, nxp_data->helper_fw_name))
+			goto free_skb;
+	}
 
 	requested_len = req->len;
 	if (requested_len == 0) {
 		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes", nxpdev->fw->size);
-		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
-		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
+		if (nxp_data->helper_fw_name && !nxpdev->helper_downloaded) {
+			nxpdev->helper_downloaded = true;
+			serdev_device_wait_until_sent(nxpdev->serdev, 0);
+			serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_SEC_BAUDRATE);
+			serdev_device_set_flow_control(nxpdev->serdev, 1);
+		} else {
+			clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+			wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
+		}
 		goto free_skb;
 	}
 	if (requested_len & 0x01) {
@@ -1142,6 +1192,7 @@ static const struct h4_recv_pkt nxp_recv_pkts[] = {
 	{ H4_RECV_ACL,          .recv = hci_recv_frame },
 	{ H4_RECV_SCO,          .recv = hci_recv_frame },
 	{ H4_RECV_EVENT,        .recv = hci_recv_frame },
+	{ NXP_RECV_CHIP_VER_V1, .recv = nxp_recv_chip_ver_v1 },
 	{ NXP_RECV_FW_REQ_V1,   .recv = nxp_recv_fw_req_v1 },
 	{ NXP_RECV_CHIP_VER_V3, .recv = nxp_recv_chip_ver_v3 },
 	{ NXP_RECV_FW_REQ_V3,   .recv = nxp_recv_fw_req_v3 },
@@ -1265,12 +1316,12 @@ static void nxp_serdev_remove(struct serdev_device *serdev)
 }
 
 static struct btnxpuart_data w8987_data = {
-	.fw_download_3M_baudrate = true,
+	.helper_fw_name = NULL,
 	.fw_name = FIRMWARE_W8987,
 };
 
 static struct btnxpuart_data w8997_data = {
-	.fw_download_3M_baudrate = false,
+	.helper_fw_name = FIRMWARE_HELPER,
 	.fw_name = FIRMWARE_W8997,
 };
 
-- 
cgit v1.2.3-70-g09d2


From 86d55f124b52de2ba0d066d89b766bcc0387fd72 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Mon, 3 Apr 2023 17:54:28 +0530
Subject: Bluetooth: btnxpuart: Deasset UART break before closing serdev device

This adds a call to ps_wakeup() before closing the serdev device, to
de-assert UART break.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 32e22044b6b2..12cc53c0b88b 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -1167,6 +1167,7 @@ static int btnxpuart_close(struct hci_dev *hdev)
 {
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
 
+	ps_wakeup(nxpdev);
 	serdev_device_close(nxpdev->serdev);
 	clear_bit(BTNXPUART_SERDEV_OPEN, &nxpdev->tx_state);
 	return 0;
-- 
cgit v1.2.3-70-g09d2


From 893410b221f4966890e5bc33be08d38d4bc968b2 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Mon, 3 Apr 2023 17:54:29 +0530
Subject: Bluetooth: btnxpuart: Disable Power Save feature on startup

This sets the default power save mode setting to disabled.
With this setting, this driver will behave like a normal h4 driver.
If user needs to use the power save feature, it can be enabled
using the following vendor command:
hcitool cmd 3f 23 02 00 00 (HCI_NXP_AUTO_SLEEP_MODE)

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 12cc53c0b88b..a2335fe3e264 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -262,7 +262,7 @@ static u8 crc8_table[CRC8_TABLE_SIZE];
 
 /* Default configurations */
 #define DEFAULT_H2C_WAKEUP_MODE	WAKEUP_METHOD_BREAK
-#define DEFAULT_PS_MODE		PS_MODE_ENABLE
+#define DEFAULT_PS_MODE		PS_MODE_DISABLE
 #define FW_INIT_BAUDRATE	HCI_NXP_PRI_BAUDRATE
 
 static struct sk_buff *nxp_drv_send_cmd(struct hci_dev *hdev, u16 opcode,
-- 
cgit v1.2.3-70-g09d2


From 305d6b6e485e46a7a3e85d5bfecda9af2294f423 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Mon, 3 Apr 2023 17:54:30 +0530
Subject: Bluetooth: btnxpuart: No need to check the received bootloader
 signature

We can never assume the uart will deliver a complete packet to the BT
layer at once, the expected packet may be divided into several parts by
uart as uart doesn't know the received packet size, the received data
count may mismatch with the expected packet size, so here
is_valid_bootloader_signature() check may always return false.

Even we remove the count check in is_valid_bootloader_signature(), then
the first part of the data which includes the packet type can pass the
is_valid_bootloader_signature() check, but the remaining parts don't
have the packet type data still cannot pass the check, here return
directly will cause the data loss.

So need to remove the received bootloader signature check here, the
h4_recv_buf() can help us combine the different data received into one
packet. If any out-of-sync or incomplete bootloader signature is received,
it is safe to ignore and discard it, and process the next bootloader
signature.

Co-developed-by: Sherry Sun <sherry.sun@nxp.com>
Signed-off-by: Sherry Sun <sherry.sun@nxp.com>
Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 19 +++----------------
 1 file changed, 3 insertions(+), 16 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index a2335fe3e264..93f3afc0c0c8 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -1199,33 +1199,20 @@ static const struct h4_recv_pkt nxp_recv_pkts[] = {
 	{ NXP_RECV_FW_REQ_V3,   .recv = nxp_recv_fw_req_v3 },
 };
 
-static bool is_valid_bootloader_signature(const u8 *data, size_t count)
-{
-	if ((*data == NXP_V1_FW_REQ_PKT && count == sizeof(struct v1_data_req) + 1) ||
-	    (*data == NXP_V3_FW_REQ_PKT && count == sizeof(struct v3_data_req) + 1) ||
-	    (*data == NXP_V3_CHIP_VER_PKT && count == sizeof(struct v3_start_ind) + 1))
-		return true;
-	else
-		return false;
-}
-
 static int btnxpuart_receive_buf(struct serdev_device *serdev, const u8 *data,
 				 size_t count)
 {
 	struct btnxpuart_dev *nxpdev = serdev_device_get_drvdata(serdev);
 
-	if (is_fw_downloading(nxpdev) && !is_valid_bootloader_signature(data, count)) {
-		/* Unknown bootloader signature, skip without returning error */
-		return count;
-	}
-
 	ps_start_timer(nxpdev);
 
 	nxpdev->rx_skb = h4_recv_buf(nxpdev->hdev, nxpdev->rx_skb, data, count,
 				     nxp_recv_pkts, ARRAY_SIZE(nxp_recv_pkts));
 	if (IS_ERR(nxpdev->rx_skb)) {
 		int err = PTR_ERR(nxpdev->rx_skb);
-
+		/* Safe to ignore out-of-sync bootloader signatures */
+		if (is_fw_downloading(nxpdev))
+			return count;
 		bt_dev_err(nxpdev->hdev, "Frame reassembly failed (%d)", err);
 		nxpdev->rx_skb = NULL;
 		return err;
-- 
cgit v1.2.3-70-g09d2


From 3c690a0d64f5964977050a91b6bba8de09c13d94 Mon Sep 17 00:00:00 2001
From: Lanzhe Li <u202212060@hust.edu.cn>
Date: Sun, 9 Apr 2023 21:02:29 +0800
Subject: Bluetooth: fix inconsistent indenting

Fixed a wrong indentation before "return".This line uses a 7 space
indent instead of a tab.

Signed-off-by: Lanzhe Li <u202212060@hust.edu.cn>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_debugfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c
index b7f682922a16..ec0df2f9188e 100644
--- a/net/bluetooth/hci_debugfs.c
+++ b/net/bluetooth/hci_debugfs.c
@@ -189,7 +189,7 @@ static int uuids_show(struct seq_file *f, void *p)
 	}
 	hci_dev_unlock(hdev);
 
-       return 0;
+	return 0;
 }
 
 DEFINE_SHOW_ATTRIBUTE(uuids);
-- 
cgit v1.2.3-70-g09d2


From 9e3c2ea67a4c24f7a7517e863e3adc89c4e3b66d Mon Sep 17 00:00:00 2001
From: Iulia Tanasescu <iulia.tanasescu@nxp.com>
Date: Tue, 11 Apr 2023 10:41:35 +0300
Subject: Bluetooth: hci_conn: remove extra line in hci_le_big_create_sync

Remove extra line setting the broadcast code parameter of the
hci_cp_le_create_big struct to 0. The broadcast code is copied
from the QoS struct.

Signed-off-by: Iulia Tanasescu <iulia.tanasescu@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_conn.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 5672b4924572..01e0b7255174 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -1700,7 +1700,6 @@ static int hci_le_create_big(struct hci_conn *conn, struct bt_iso_qos *qos)
 	cp.bis.framing = qos->bcast.framing;
 	cp.bis.encryption = qos->bcast.encryption;
 	memcpy(cp.bis.bcode, qos->bcast.bcode, sizeof(cp.bis.bcode));
-	memset(&cp.bis.bcode, 0, sizeof(cp.bis.bcode));
 
 	return hci_send_cmd(hdev, HCI_OP_LE_CREATE_BIG, sizeof(cp), &cp);
 }
-- 
cgit v1.2.3-70-g09d2


From 0811ff480978a2122f0465cb8682d9f7a2f6a169 Mon Sep 17 00:00:00 2001
From: Steev Klimaszewski <steev@kali.org>
Date: Wed, 12 Apr 2023 18:58:17 -0500
Subject: Bluetooth: hci_qca: mark OF related data as maybe unused

The driver can be compile tested with !CONFIG_OF making certain data
unused.

Signed-off-by: Steev Klimaszewski <steev@kali.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 1597797ff169..1b064504b388 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -1891,7 +1891,7 @@ static const struct qca_device_data qca_soc_data_wcn6750 __maybe_unused = {
 	.capabilities = QCA_CAP_WIDEBAND_SPEECH | QCA_CAP_VALID_LE_STATES,
 };
 
-static const struct qca_device_data qca_soc_data_wcn6855 = {
+static const struct qca_device_data qca_soc_data_wcn6855 __maybe_unused = {
 	.soc_type = QCA_WCN6855,
 	.vregs = (struct qca_vreg []) {
 		{ "vddio", 5000 },
-- 
cgit v1.2.3-70-g09d2


From e4eea890369c00dd58d97b1c066dc2bddf0da2c7 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Thu, 6 Apr 2023 11:55:17 +0300
Subject: Bluetooth: vhci: Fix info leak in force_devcd_write()

There are a number of bugs here:

1) If "count" is less than sizeof(dump_data.data) then it copies
   uninitialized data.
2) If simple_write_to_buffer() returns -EFAULT then we run into a
   problem "ret < count" comparison.  "count" is an unsigned long so the
   comparison is type promoted to unsigned long and the negative returns
   become high positive values.  That also results in copying
   uninitialized data.
3) If "*ppos" is non-zero then the first part of the dump_data
   buffer is uninitialized.  Using copy_from_user() instead of
   simple_write_to_buffer() is more appropriate here.

Fixes: d5d5df6da0aa ("Bluetooth: Add vhci devcoredump support")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_vhci.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c
index 691fe93b1976..40e2b9fa11a2 100644
--- a/drivers/bluetooth/hci_vhci.c
+++ b/drivers/bluetooth/hci_vhci.c
@@ -323,17 +323,21 @@ static ssize_t force_devcd_write(struct file *file, const char __user *user_buf,
 	struct hci_dev *hdev = data->hdev;
 	struct sk_buff *skb = NULL;
 	struct devcoredump_test_data dump_data;
+	size_t data_size;
 	int ret;
 
-	ret = simple_write_to_buffer(&dump_data, sizeof(dump_data), ppos,
-				     user_buf, count);
-	if (ret < count)
-		return ret;
+	if (count < offsetof(struct devcoredump_test_data, data) ||
+	    count > sizeof(dump_data))
+		return -EINVAL;
+
+	if (copy_from_user(&dump_data, user_buf, count))
+		return -EFAULT;
 
-	skb = alloc_skb(sizeof(dump_data.data), GFP_ATOMIC);
+	data_size = count - offsetof(struct devcoredump_test_data, data);
+	skb = alloc_skb(data_size, GFP_ATOMIC);
 	if (!skb)
 		return -ENOMEM;
-	skb_put_data(skb, &dump_data.data, sizeof(dump_data.data));
+	skb_put_data(skb, &dump_data.data, data_size);
 
 	hci_devcd_register(hdev, vhci_coredump, vhci_coredump_hdr, NULL);
 
-- 
cgit v1.2.3-70-g09d2


From 06149746e7203d5ffe2d6faf9799ee36203aa8b8 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Tue, 11 Apr 2023 16:02:22 -0700
Subject: Bluetooth: hci_conn: Add support for linking multiple hcon

Since it is required for some configurations to have multiple CIS with
the same peer which is now covered by iso-tester in the following test
cases:

    ISO AC 6(i) - Success
    ISO AC 7(i) - Success
    ISO AC 8(i) - Success
    ISO AC 9(i) - Success
    ISO AC 11(i) - Success

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h |  14 +++-
 net/bluetooth/hci_conn.c         | 155 ++++++++++++++++++++++++++++-----------
 net/bluetooth/hci_event.c        |  92 +++++++++++------------
 net/bluetooth/iso.c              |   8 +-
 4 files changed, 172 insertions(+), 97 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 827e67159523..4fe1e71cb9d8 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -770,7 +770,10 @@ struct hci_conn {
 	void		*iso_data;
 	struct amp_mgr	*amp_mgr;
 
-	struct hci_conn	*link;
+	struct list_head link_list;
+	struct hci_conn	*parent;
+	struct hci_link *link;
+
 	struct bt_codec codec;
 
 	void (*connect_cfm_cb)	(struct hci_conn *conn, u8 status);
@@ -780,6 +783,11 @@ struct hci_conn {
 	void (*cleanup)(struct hci_conn *conn);
 };
 
+struct hci_link {
+	struct list_head list;
+	struct hci_conn *conn;
+};
+
 struct hci_chan {
 	struct list_head list;
 	__u16 handle;
@@ -1383,12 +1391,14 @@ static inline void hci_conn_put(struct hci_conn *conn)
 	put_device(&conn->dev);
 }
 
-static inline void hci_conn_hold(struct hci_conn *conn)
+static inline struct hci_conn *hci_conn_hold(struct hci_conn *conn)
 {
 	BT_DBG("hcon %p orig refcnt %d", conn, atomic_read(&conn->refcnt));
 
 	atomic_inc(&conn->refcnt);
 	cancel_delayed_work(&conn->disc_work);
+
+	return conn;
 }
 
 static inline void hci_conn_drop(struct hci_conn *conn)
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 01e0b7255174..d8466abbb36a 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -330,8 +330,11 @@ static void hci_add_sco(struct hci_conn *conn, __u16 handle)
 static bool find_next_esco_param(struct hci_conn *conn,
 				 const struct sco_param *esco_param, int size)
 {
+	if (!conn->parent)
+		return false;
+
 	for (; conn->attempt <= size; conn->attempt++) {
-		if (lmp_esco_2m_capable(conn->link) ||
+		if (lmp_esco_2m_capable(conn->parent) ||
 		    (esco_param[conn->attempt - 1].pkt_type & ESCO_2EV3))
 			break;
 		BT_DBG("hcon %p skipped attempt %d, eSCO 2M not supported",
@@ -461,7 +464,7 @@ static int hci_enhanced_setup_sync(struct hci_dev *hdev, void *data)
 		break;
 
 	case BT_CODEC_CVSD:
-		if (lmp_esco_capable(conn->link)) {
+		if (conn->parent && lmp_esco_capable(conn->parent)) {
 			if (!find_next_esco_param(conn, esco_param_cvsd,
 						  ARRAY_SIZE(esco_param_cvsd)))
 				return -EINVAL;
@@ -531,7 +534,7 @@ static bool hci_setup_sync_conn(struct hci_conn *conn, __u16 handle)
 		param = &esco_param_msbc[conn->attempt - 1];
 		break;
 	case SCO_AIRMODE_CVSD:
-		if (lmp_esco_capable(conn->link)) {
+		if (conn->parent && lmp_esco_capable(conn->parent)) {
 			if (!find_next_esco_param(conn, esco_param_cvsd,
 						  ARRAY_SIZE(esco_param_cvsd)))
 				return false;
@@ -637,21 +640,22 @@ void hci_le_start_enc(struct hci_conn *conn, __le16 ediv, __le64 rand,
 /* Device _must_ be locked */
 void hci_sco_setup(struct hci_conn *conn, __u8 status)
 {
-	struct hci_conn *sco = conn->link;
+	struct hci_link *link;
 
-	if (!sco)
+	link = list_first_entry_or_null(&conn->link_list, struct hci_link, list);
+	if (!link || !link->conn)
 		return;
 
 	BT_DBG("hcon %p", conn);
 
 	if (!status) {
 		if (lmp_esco_capable(conn->hdev))
-			hci_setup_sync(sco, conn->handle);
+			hci_setup_sync(link->conn, conn->handle);
 		else
-			hci_add_sco(sco, conn->handle);
+			hci_add_sco(link->conn, conn->handle);
 	} else {
-		hci_connect_cfm(sco, status);
-		hci_conn_del(sco);
+		hci_connect_cfm(link->conn, status);
+		hci_conn_del(link->conn);
 	}
 }
 
@@ -1042,6 +1046,7 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst,
 	skb_queue_head_init(&conn->data_q);
 
 	INIT_LIST_HEAD(&conn->chan_list);
+	INIT_LIST_HEAD(&conn->link_list);
 
 	INIT_DELAYED_WORK(&conn->disc_work, hci_conn_timeout);
 	INIT_DELAYED_WORK(&conn->auto_accept_work, hci_conn_auto_accept);
@@ -1069,15 +1074,39 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst,
 	return conn;
 }
 
-static bool hci_conn_unlink(struct hci_conn *conn)
+static void hci_conn_unlink(struct hci_conn *conn)
 {
+	struct hci_dev *hdev = conn->hdev;
+
+	bt_dev_dbg(hdev, "hcon %p", conn);
+
+	if (!conn->parent) {
+		struct hci_link *link, *t;
+
+		list_for_each_entry_safe(link, t, &conn->link_list, list)
+			hci_conn_unlink(link->conn);
+
+		return;
+	}
+
 	if (!conn->link)
-		return false;
+		return;
+
+	hci_conn_put(conn->parent);
+	conn->parent = NULL;
 
-	conn->link->link = NULL;
+	list_del_rcu(&conn->link->list);
+	synchronize_rcu();
+
+	kfree(conn->link);
 	conn->link = NULL;
 
-	return true;
+	/* Due to race, SCO connection might be not established
+	 * yet at this point. Delete it now, otherwise it is
+	 * possible for it to be stuck and can't be deleted.
+	 */
+	if (conn->handle == HCI_CONN_HANDLE_UNSET)
+		hci_conn_del(conn);
 }
 
 int hci_conn_del(struct hci_conn *conn)
@@ -1091,18 +1120,7 @@ int hci_conn_del(struct hci_conn *conn)
 	cancel_delayed_work_sync(&conn->idle_work);
 
 	if (conn->type == ACL_LINK) {
-		struct hci_conn *link = conn->link;
-
-		if (link) {
-			hci_conn_unlink(conn);
-			/* Due to race, SCO connection might be not established
-			 * yet at this point. Delete it now, otherwise it is
-			 * possible for it to be stuck and can't be deleted.
-			 */
-			if (link->handle == HCI_CONN_HANDLE_UNSET)
-				hci_conn_del(link);
-		}
-
+		hci_conn_unlink(conn);
 		/* Unacked frames */
 		hdev->acl_cnt += conn->sent;
 	} else if (conn->type == LE_LINK) {
@@ -1113,7 +1131,7 @@ int hci_conn_del(struct hci_conn *conn)
 		else
 			hdev->acl_cnt += conn->sent;
 	} else {
-		struct hci_conn *acl = conn->link;
+		struct hci_conn *acl = conn->parent;
 
 		if (acl) {
 			hci_conn_unlink(conn);
@@ -1600,11 +1618,40 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 	return acl;
 }
 
+static struct hci_link *hci_conn_link(struct hci_conn *parent,
+				      struct hci_conn *conn)
+{
+	struct hci_dev *hdev = parent->hdev;
+	struct hci_link *link;
+
+	bt_dev_dbg(hdev, "parent %p hcon %p", parent, conn);
+
+	if (conn->link)
+		return conn->link;
+
+	if (conn->parent)
+		return NULL;
+
+	link = kzalloc(sizeof(*link), GFP_KERNEL);
+	if (!link)
+		return NULL;
+
+	link->conn = hci_conn_hold(conn);
+	conn->link = link;
+	conn->parent = hci_conn_get(parent);
+
+	/* Use list_add_tail_rcu append to the list */
+	list_add_tail_rcu(&link->list, &parent->link_list);
+
+	return link;
+}
+
 struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst,
 				 __u16 setting, struct bt_codec *codec)
 {
 	struct hci_conn *acl;
 	struct hci_conn *sco;
+	struct hci_link *link;
 
 	acl = hci_connect_acl(hdev, dst, BT_SECURITY_LOW, HCI_AT_NO_BONDING,
 			      CONN_REASON_SCO_CONNECT);
@@ -1620,10 +1667,12 @@ struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst,
 		}
 	}
 
-	acl->link = sco;
-	sco->link = acl;
-
-	hci_conn_hold(sco);
+	link = hci_conn_link(acl, sco);
+	if (!link) {
+		hci_conn_drop(acl);
+		hci_conn_drop(sco);
+		return NULL;
+	}
 
 	sco->setting = setting;
 	sco->codec = *codec;
@@ -1890,7 +1939,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 	u8 cig;
 
 	memset(&cmd, 0, sizeof(cmd));
-	cmd.cis[0].acl_handle = cpu_to_le16(conn->link->handle);
+	cmd.cis[0].acl_handle = cpu_to_le16(conn->parent->handle);
 	cmd.cis[0].cis_handle = cpu_to_le16(conn->handle);
 	cmd.cp.num_cis++;
 	cig = conn->iso_qos.ucast.cig;
@@ -1903,11 +1952,12 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 		struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis];
 
 		if (conn == data || conn->type != ISO_LINK ||
-		    conn->state == BT_CONNECTED || conn->iso_qos.ucast.cig != cig)
+		    conn->state == BT_CONNECTED ||
+		    conn->iso_qos.ucast.cig != cig)
 			continue;
 
 		/* Check if all CIS(s) belonging to a CIG are ready */
-		if (!conn->link || conn->link->state != BT_CONNECTED ||
+		if (!conn->parent || conn->parent->state != BT_CONNECTED ||
 		    conn->state != BT_CONNECT) {
 			cmd.cp.num_cis = 0;
 			break;
@@ -1924,7 +1974,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 		 * command have been generated, the Controller shall return the
 		 * error code Command Disallowed (0x0C).
 		 */
-		cis->acl_handle = cpu_to_le16(conn->link->handle);
+		cis->acl_handle = cpu_to_le16(conn->parent->handle);
 		cis->cis_handle = cpu_to_le16(conn->handle);
 		cmd.cp.num_cis++;
 	}
@@ -1943,15 +1993,33 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 int hci_le_create_cis(struct hci_conn *conn)
 {
 	struct hci_conn *cis;
+	struct hci_link *link, *t;
 	struct hci_dev *hdev = conn->hdev;
 	int err;
 
+	bt_dev_dbg(hdev, "hcon %p", conn);
+
 	switch (conn->type) {
 	case LE_LINK:
-		if (!conn->link || conn->state != BT_CONNECTED)
+		if (conn->state != BT_CONNECTED || list_empty(&conn->link_list))
 			return -EINVAL;
-		cis = conn->link;
-		break;
+
+		cis = NULL;
+
+		/* hci_conn_link uses list_add_tail_rcu so the list is in
+		 * the same order as the connections are requested.
+		 */
+		list_for_each_entry_safe(link, t, &conn->link_list, list) {
+			if (link->conn->state == BT_BOUND) {
+				err = hci_le_create_cis(link->conn);
+				if (err)
+					return err;
+
+				cis = link->conn;
+			}
+		}
+
+		return cis ? 0 : -EINVAL;
 	case ISO_LINK:
 		cis = conn;
 		break;
@@ -2172,6 +2240,7 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst,
 {
 	struct hci_conn *le;
 	struct hci_conn *cis;
+	struct hci_link *link;
 
 	if (hci_dev_test_flag(hdev, HCI_ADVERTISING))
 		le = hci_connect_le(hdev, dst, dst_type, false,
@@ -2197,16 +2266,18 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst,
 		return cis;
 	}
 
-	le->link = cis;
-	cis->link = le;
-
-	hci_conn_hold(cis);
+	link = hci_conn_link(le, cis);
+	if (!link) {
+		hci_conn_drop(le);
+		hci_conn_drop(cis);
+		return NULL;
+	}
 
 	/* If LE is already connected and CIS handle is already set proceed to
 	 * Create CIS immediately.
 	 */
 	if (le->state == BT_CONNECTED && cis->handle != HCI_CONN_HANDLE_UNSET)
-		hci_le_create_cis(le);
+		hci_le_create_cis(cis);
 
 	return cis;
 }
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 0e0a93cc1218..d00ef6e3fc45 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -2345,7 +2345,8 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status)
 static void hci_cs_add_sco(struct hci_dev *hdev, __u8 status)
 {
 	struct hci_cp_add_sco *cp;
-	struct hci_conn *acl, *sco;
+	struct hci_conn *acl;
+	struct hci_link *link;
 	__u16 handle;
 
 	bt_dev_dbg(hdev, "status 0x%2.2x", status);
@@ -2365,12 +2366,13 @@ static void hci_cs_add_sco(struct hci_dev *hdev, __u8 status)
 
 	acl = hci_conn_hash_lookup_handle(hdev, handle);
 	if (acl) {
-		sco = acl->link;
-		if (sco) {
-			sco->state = BT_CLOSED;
+		link = list_first_entry_or_null(&acl->link_list,
+						struct hci_link, list);
+		if (link && link->conn) {
+			link->conn->state = BT_CLOSED;
 
-			hci_connect_cfm(sco, status);
-			hci_conn_del(sco);
+			hci_connect_cfm(link->conn, status);
+			hci_conn_del(link->conn);
 		}
 	}
 
@@ -2637,74 +2639,61 @@ static void hci_cs_read_remote_ext_features(struct hci_dev *hdev, __u8 status)
 	hci_dev_unlock(hdev);
 }
 
-static void hci_cs_setup_sync_conn(struct hci_dev *hdev, __u8 status)
+static void hci_setup_sync_conn_status(struct hci_dev *hdev, __u16 handle,
+				       __u8 status)
 {
-	struct hci_cp_setup_sync_conn *cp;
-	struct hci_conn *acl, *sco;
-	__u16 handle;
-
-	bt_dev_dbg(hdev, "status 0x%2.2x", status);
-
-	if (!status)
-		return;
-
-	cp = hci_sent_cmd_data(hdev, HCI_OP_SETUP_SYNC_CONN);
-	if (!cp)
-		return;
+	struct hci_conn *acl;
+	struct hci_link *link;
 
-	handle = __le16_to_cpu(cp->handle);
-
-	bt_dev_dbg(hdev, "handle 0x%4.4x", handle);
+	bt_dev_dbg(hdev, "handle 0x%4.4x status 0x%2.2x", handle, status);
 
 	hci_dev_lock(hdev);
 
 	acl = hci_conn_hash_lookup_handle(hdev, handle);
 	if (acl) {
-		sco = acl->link;
-		if (sco) {
-			sco->state = BT_CLOSED;
+		link = list_first_entry_or_null(&acl->link_list,
+						struct hci_link, list);
+		if (link && link->conn) {
+			link->conn->state = BT_CLOSED;
 
-			hci_connect_cfm(sco, status);
-			hci_conn_del(sco);
+			hci_connect_cfm(link->conn, status);
+			hci_conn_del(link->conn);
 		}
 	}
 
 	hci_dev_unlock(hdev);
 }
 
-static void hci_cs_enhanced_setup_sync_conn(struct hci_dev *hdev, __u8 status)
+static void hci_cs_setup_sync_conn(struct hci_dev *hdev, __u8 status)
 {
-	struct hci_cp_enhanced_setup_sync_conn *cp;
-	struct hci_conn *acl, *sco;
-	__u16 handle;
+	struct hci_cp_setup_sync_conn *cp;
 
 	bt_dev_dbg(hdev, "status 0x%2.2x", status);
 
 	if (!status)
 		return;
 
-	cp = hci_sent_cmd_data(hdev, HCI_OP_ENHANCED_SETUP_SYNC_CONN);
+	cp = hci_sent_cmd_data(hdev, HCI_OP_SETUP_SYNC_CONN);
 	if (!cp)
 		return;
 
-	handle = __le16_to_cpu(cp->handle);
+	hci_setup_sync_conn_status(hdev, __le16_to_cpu(cp->handle), status);
+}
 
-	bt_dev_dbg(hdev, "handle 0x%4.4x", handle);
+static void hci_cs_enhanced_setup_sync_conn(struct hci_dev *hdev, __u8 status)
+{
+	struct hci_cp_enhanced_setup_sync_conn *cp;
 
-	hci_dev_lock(hdev);
+	bt_dev_dbg(hdev, "status 0x%2.2x", status);
 
-	acl = hci_conn_hash_lookup_handle(hdev, handle);
-	if (acl) {
-		sco = acl->link;
-		if (sco) {
-			sco->state = BT_CLOSED;
+	if (!status)
+		return;
 
-			hci_connect_cfm(sco, status);
-			hci_conn_del(sco);
-		}
-	}
+	cp = hci_sent_cmd_data(hdev, HCI_OP_ENHANCED_SETUP_SYNC_CONN);
+	if (!cp)
+		return;
 
-	hci_dev_unlock(hdev);
+	hci_setup_sync_conn_status(hdev, __le16_to_cpu(cp->handle), status);
 }
 
 static void hci_cs_sniff_mode(struct hci_dev *hdev, __u8 status)
@@ -3834,19 +3823,20 @@ static u8 hci_cc_le_set_cig_params(struct hci_dev *hdev, void *data,
 	rcu_read_lock();
 
 	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		if (conn->type != ISO_LINK || conn->iso_qos.ucast.cig != rp->cig_id ||
+		if (conn->type != ISO_LINK ||
+		    conn->iso_qos.ucast.cig != rp->cig_id ||
 		    conn->state == BT_CONNECTED)
 			continue;
 
 		conn->handle = __le16_to_cpu(rp->handle[i++]);
 
-		bt_dev_dbg(hdev, "%p handle 0x%4.4x link %p", conn,
-			   conn->handle, conn->link);
+		bt_dev_dbg(hdev, "%p handle 0x%4.4x parent %p", conn,
+			   conn->handle, conn->parent);
 
 		/* Create CIS if LE is already connected */
-		if (conn->link && conn->link->state == BT_CONNECTED) {
+		if (conn->parent && conn->parent->state == BT_CONNECTED) {
 			rcu_read_unlock();
-			hci_le_create_cis(conn->link);
+			hci_le_create_cis(conn);
 			rcu_read_lock();
 		}
 
@@ -5031,7 +5021,7 @@ static void hci_sync_conn_complete_evt(struct hci_dev *hdev, void *data,
 		if (conn->out) {
 			conn->pkt_type = (hdev->esco_type & SCO_ESCO_MASK) |
 					(hdev->esco_type & EDR_ESCO_MASK);
-			if (hci_setup_sync(conn, conn->link->handle))
+			if (hci_setup_sync(conn, conn->parent->handle))
 				goto unlock;
 		}
 		fallthrough;
diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c
index 74117df03a3f..34d55a85d8f6 100644
--- a/net/bluetooth/iso.c
+++ b/net/bluetooth/iso.c
@@ -1657,8 +1657,12 @@ static void iso_connect_cfm(struct hci_conn *hcon, __u8 status)
 
 		/* Check if LE link has failed */
 		if (status) {
-			if (hcon->link)
-				iso_conn_del(hcon->link, bt_to_errno(status));
+			struct hci_link *link, *t;
+
+			list_for_each_entry_safe(link, t, &hcon->link_list,
+						 list)
+				iso_conn_del(link->conn, bt_to_errno(status));
+
 			return;
 		}
 
-- 
cgit v1.2.3-70-g09d2


From c14516faede33c2c31da45cf950d55dbff42962e Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Tue, 11 Apr 2023 16:14:25 -0700
Subject: Bluetooth: hci_conn: Fix not matching by CIS ID

This fixes only matching CIS by address which prevents creating new hcon
if upper layer is requesting a specific CIS ID.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h | 12 +++++++++++-
 net/bluetooth/hci_conn.c         |  3 ++-
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 4fe1e71cb9d8..a6c8aee2f256 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -1180,7 +1180,9 @@ static inline struct hci_conn *hci_conn_hash_lookup_le(struct hci_dev *hdev,
 
 static inline struct hci_conn *hci_conn_hash_lookup_cis(struct hci_dev *hdev,
 							bdaddr_t *ba,
-							__u8 ba_type)
+							__u8 ba_type,
+							__u8 cig,
+							__u8 id)
 {
 	struct hci_conn_hash *h = &hdev->conn_hash;
 	struct hci_conn  *c;
@@ -1191,6 +1193,14 @@ static inline struct hci_conn *hci_conn_hash_lookup_cis(struct hci_dev *hdev,
 		if (c->type != ISO_LINK)
 			continue;
 
+		/* Match CIG ID if set */
+		if (cig != BT_ISO_QOS_CIG_UNSET && cig != c->iso_qos.ucast.cig)
+			continue;
+
+		/* Match CIS ID if set */
+		if (id != BT_ISO_QOS_CIS_UNSET && id != c->iso_qos.ucast.cis)
+			continue;
+
 		if (ba_type == c->dst_type && !bacmp(&c->dst, ba)) {
 			rcu_read_unlock();
 			return c;
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index d8466abbb36a..c215e983e287 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -1841,7 +1841,8 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst,
 {
 	struct hci_conn *cis;
 
-	cis = hci_conn_hash_lookup_cis(hdev, dst, dst_type);
+	cis = hci_conn_hash_lookup_cis(hdev, dst, dst_type, qos->ucast.cig,
+				       qos->ucast.cis);
 	if (!cis) {
 		cis = hci_conn_add(hdev, ISO_LINK, dst, HCI_ROLE_MASTER);
 		if (!cis)
-- 
cgit v1.2.3-70-g09d2


From c09b80be6ffc338634b2f5f8cfa12b6843410834 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 12 Apr 2023 17:45:51 -0700
Subject: Bluetooth: hci_conn: Fix not waiting for HCI_EVT_LE_CIS_ESTABLISHED

When submitting HCI_OP_LE_CREATE_CIS the code shall wait for
HCI_EVT_LE_CIS_ESTABLISHED thus enforcing the serialization of
HCI_OP_LE_CREATE_CIS as the Core spec does not allow to send them in
parallel:

  BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E page 2566:

  If the Host issues this command before all the HCI_LE_CIS_Established
  events from the previous use of the command have been generated, the
  Controller shall return the error code Command Disallowed (0x0C).

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sync.h |  2 ++
 net/bluetooth/hci_conn.c         | 58 +----------------------------------
 net/bluetooth/hci_sync.c         | 65 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 68 insertions(+), 57 deletions(-)

diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index 17f5a4c32f36..f61b249787fc 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -122,6 +122,8 @@ int hci_abort_conn_sync(struct hci_dev *hdev, struct hci_conn *conn, u8 reason);
 
 int hci_le_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn);
 
+int hci_le_create_cis_sync(struct hci_dev *hdev, struct hci_conn *conn);
+
 int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle);
 
 int hci_le_terminate_big_sync(struct hci_dev *hdev, u8 handle, u8 reason);
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index c215e983e287..640b951bf40a 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -1932,63 +1932,7 @@ bool hci_iso_setup_path(struct hci_conn *conn)
 
 static int hci_create_cis_sync(struct hci_dev *hdev, void *data)
 {
-	struct {
-		struct hci_cp_le_create_cis cp;
-		struct hci_cis cis[0x1f];
-	} cmd;
-	struct hci_conn *conn = data;
-	u8 cig;
-
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.cis[0].acl_handle = cpu_to_le16(conn->parent->handle);
-	cmd.cis[0].cis_handle = cpu_to_le16(conn->handle);
-	cmd.cp.num_cis++;
-	cig = conn->iso_qos.ucast.cig;
-
-	hci_dev_lock(hdev);
-
-	rcu_read_lock();
-
-	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis];
-
-		if (conn == data || conn->type != ISO_LINK ||
-		    conn->state == BT_CONNECTED ||
-		    conn->iso_qos.ucast.cig != cig)
-			continue;
-
-		/* Check if all CIS(s) belonging to a CIG are ready */
-		if (!conn->parent || conn->parent->state != BT_CONNECTED ||
-		    conn->state != BT_CONNECT) {
-			cmd.cp.num_cis = 0;
-			break;
-		}
-
-		/* Group all CIS with state BT_CONNECT since the spec don't
-		 * allow to send them individually:
-		 *
-		 * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E
-		 * page 2566:
-		 *
-		 * If the Host issues this command before all the
-		 * HCI_LE_CIS_Established events from the previous use of the
-		 * command have been generated, the Controller shall return the
-		 * error code Command Disallowed (0x0C).
-		 */
-		cis->acl_handle = cpu_to_le16(conn->parent->handle);
-		cis->cis_handle = cpu_to_le16(conn->handle);
-		cmd.cp.num_cis++;
-	}
-
-	rcu_read_unlock();
-
-	hci_dev_unlock(hdev);
-
-	if (!cmd.cp.num_cis)
-		return 0;
-
-	return hci_send_cmd(hdev, HCI_OP_LE_CREATE_CIS, sizeof(cmd.cp) +
-			    sizeof(cmd.cis[0]) * cmd.cp.num_cis, &cmd);
+	return hci_le_create_cis_sync(hdev, data);
 }
 
 int hci_le_create_cis(struct hci_conn *conn)
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 00017f75cd41..6f060d00a70a 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -6137,6 +6137,71 @@ done:
 	return err;
 }
 
+int hci_le_create_cis_sync(struct hci_dev *hdev, struct hci_conn *conn)
+{
+	struct {
+		struct hci_cp_le_create_cis cp;
+		struct hci_cis cis[0x1f];
+	} cmd;
+	u8 cig;
+	struct hci_conn *hcon = conn;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.cis[0].acl_handle = cpu_to_le16(conn->parent->handle);
+	cmd.cis[0].cis_handle = cpu_to_le16(conn->handle);
+	cmd.cp.num_cis++;
+	cig = conn->iso_qos.ucast.cig;
+
+	hci_dev_lock(hdev);
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
+		struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis];
+
+		if (conn == hcon || conn->type != ISO_LINK ||
+		    conn->state == BT_CONNECTED ||
+		    conn->iso_qos.ucast.cig != cig)
+			continue;
+
+		/* Check if all CIS(s) belonging to a CIG are ready */
+		if (!conn->parent || conn->parent->state != BT_CONNECTED ||
+		    conn->state != BT_CONNECT) {
+			cmd.cp.num_cis = 0;
+			break;
+		}
+
+		/* Group all CIS with state BT_CONNECT since the spec don't
+		 * allow to send them individually:
+		 *
+		 * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E
+		 * page 2566:
+		 *
+		 * If the Host issues this command before all the
+		 * HCI_LE_CIS_Established events from the previous use of the
+		 * command have been generated, the Controller shall return the
+		 * error code Command Disallowed (0x0C).
+		 */
+		cis->acl_handle = cpu_to_le16(conn->parent->handle);
+		cis->cis_handle = cpu_to_le16(conn->handle);
+		cmd.cp.num_cis++;
+	}
+
+	rcu_read_unlock();
+
+	hci_dev_unlock(hdev);
+
+	if (!cmd.cp.num_cis)
+		return 0;
+
+	/* Wait for HCI_LE_CIS_Established */
+	return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CREATE_CIS,
+					sizeof(cmd.cp) + sizeof(cmd.cis[0]) *
+					cmd.cp.num_cis, &cmd,
+					HCI_EVT_LE_CIS_ESTABLISHED,
+					conn->conn_timeout, NULL);
+}
+
 int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle)
 {
 	struct hci_cp_le_remove_cig cp;
-- 
cgit v1.2.3-70-g09d2


From 91b6d02ddcd113352bdd895990b252065c596de7 Mon Sep 17 00:00:00 2001
From: Raul Cheleguini <raul.cheleguini@gmail.com>
Date: Thu, 23 Mar 2023 10:45:39 -0300
Subject: Bluetooth: Add new quirk for broken set random RPA timeout for
 ATS2851

The ATS2851 based controller advertises support for command "LE Set Random
Private Address Timeout" but does not actually implement it, impeding the
controller initialization.

Add the quirk HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT to unblock the controller
initialization.

< HCI Command: LE Set Resolvable Private... (0x08|0x002e) plen 2
        Timeout: 900 seconds
> HCI Event: Command Status (0x0f) plen 4
      LE Set Resolvable Private Address Timeout (0x08|0x002e) ncmd 1
        Status: Unknown HCI Command (0x01)

Co-developed-by: imoc <wzj9912@gmail.com>
Signed-off-by: imoc <wzj9912@gmail.com>
Signed-off-by: Raul Cheleguini <raul.cheleguini@gmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c   | 1 +
 include/net/bluetooth/hci.h | 8 ++++++++
 net/bluetooth/hci_sync.c    | 6 +++++-
 3 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 2303b0a66323..3aa189b1986d 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -4149,6 +4149,7 @@ static int btusb_probe(struct usb_interface *intf,
 		/* Support is advertised, but not implemented */
 		set_bit(HCI_QUIRK_BROKEN_ERR_DATA_REPORTING, &hdev->quirks);
 		set_bit(HCI_QUIRK_BROKEN_READ_TRANSMIT_POWER, &hdev->quirks);
+		set_bit(HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT, &hdev->quirks);
 		set_bit(HCI_QUIRK_BROKEN_EXT_SCAN, &hdev->quirks);
 	}
 
diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 997107bfc0b1..07df96c47ef4 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -301,6 +301,14 @@ enum {
 	 * don't actually support features declared there.
 	 */
 	HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2,
+
+	/*
+	 * When this quirk is set, the HCI_OP_LE_SET_RPA_TIMEOUT command is
+	 * skipped during initialization. This is required for the Actions
+	 * Semiconductor ATS2851 based controllers, which erroneously claims
+	 * to support it.
+	 */
+	HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT,
 };
 
 /* HCI device flags */
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 6f060d00a70a..771aaa808967 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -4093,7 +4093,8 @@ static int hci_le_set_rpa_timeout_sync(struct hci_dev *hdev)
 {
 	__le16 timeout = cpu_to_le16(hdev->rpa_timeout);
 
-	if (!(hdev->commands[35] & 0x04))
+	if (!(hdev->commands[35] & 0x04) ||
+	    test_bit(HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT, &hdev->quirks))
 		return 0;
 
 	return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_RPA_TIMEOUT,
@@ -4553,6 +4554,9 @@ static const struct {
 			 "HCI Set Event Filter command not supported."),
 	HCI_QUIRK_BROKEN(ENHANCED_SETUP_SYNC_CONN,
 			 "HCI Enhanced Setup Synchronous Connection command is "
+			 "advertised, but not supported."),
+	HCI_QUIRK_BROKEN(SET_RPA_TIMEOUT,
+			 "HCI LE Set Random Private Address Timeout command is "
 			 "advertised, but not supported.")
 };
 
-- 
cgit v1.2.3-70-g09d2


From db2bf510bd5d57f064d9e1db395ed86a08320c54 Mon Sep 17 00:00:00 2001
From: Liu Jian <liujian56@huawei.com>
Date: Fri, 14 Apr 2023 18:30:06 +0800
Subject: Revert "Bluetooth: btsdio: fix use after free bug in btsdio_remove
 due to unfinished work"

This reverts commit 1e9ac114c4428fdb7ff4635b45d4f46017e8916f.

This patch introduces a possible null-ptr-def problem. Revert it. And the
fixed bug by this patch have resolved by commit 73f7b171b7c0 ("Bluetooth:
btsdio: fix use after free bug in btsdio_remove due to race condition").

Fixes: 1e9ac114c442 ("Bluetooth: btsdio: fix use after free bug in btsdio_remove due to unfinished work")
Signed-off-by: Liu Jian <liujian56@huawei.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btsdio.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/bluetooth/btsdio.c b/drivers/bluetooth/btsdio.c
index 51000320e1ea..f19d31ee37ea 100644
--- a/drivers/bluetooth/btsdio.c
+++ b/drivers/bluetooth/btsdio.c
@@ -354,7 +354,6 @@ static void btsdio_remove(struct sdio_func *func)
 
 	BT_DBG("func %p", func);
 
-	cancel_work_sync(&data->work);
 	if (!data)
 		return;
 
-- 
cgit v1.2.3-70-g09d2


From 000c2fa2c144c499c881a101819cf1936a1f7cf2 Mon Sep 17 00:00:00 2001
From: Ruihan Li <lrh2000@pku.edu.cn>
Date: Sun, 16 Apr 2023 16:02:51 +0800
Subject: bluetooth: Add cmd validity checks at the start of hci_sock_ioctl()

Previously, channel open messages were always sent to monitors on the first
ioctl() call for unbound HCI sockets, even if the command and arguments
were completely invalid. This can leave an exploitable hole with the abuse
of invalid ioctl calls.

This commit hardens the ioctl processing logic by first checking if the
command is valid, and immediately returning with an ENOIOCTLCMD error code
if it is not. This ensures that ioctl calls with invalid commands are free
of side effects, and increases the difficulty of further exploitation by
forcing exploitation to find a way to pass a valid command first.

Signed-off-by: Ruihan Li <lrh2000@pku.edu.cn>
Co-developed-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_sock.c | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index 06581223238c..8d8950101cee 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -987,6 +987,34 @@ static int hci_sock_ioctl(struct socket *sock, unsigned int cmd,
 
 	BT_DBG("cmd %x arg %lx", cmd, arg);
 
+	/* Make sure the cmd is valid before doing anything */
+	switch (cmd) {
+	case HCIGETDEVLIST:
+	case HCIGETDEVINFO:
+	case HCIGETCONNLIST:
+	case HCIDEVUP:
+	case HCIDEVDOWN:
+	case HCIDEVRESET:
+	case HCIDEVRESTAT:
+	case HCISETSCAN:
+	case HCISETAUTH:
+	case HCISETENCRYPT:
+	case HCISETPTYPE:
+	case HCISETLINKPOL:
+	case HCISETLINKMODE:
+	case HCISETACLMTU:
+	case HCISETSCOMTU:
+	case HCIINQUIRY:
+	case HCISETRAW:
+	case HCIGETCONNINFO:
+	case HCIGETAUTHINFO:
+	case HCIBLOCKADDR:
+	case HCIUNBLOCKADDR:
+		break;
+	default:
+		return -ENOIOCTLCMD;
+	}
+
 	lock_sock(sk);
 
 	if (hci_pi(sk)->channel != HCI_CHANNEL_RAW) {
-- 
cgit v1.2.3-70-g09d2


From 25e97f7b1866e6b8503be349eeea44bb52d661ce Mon Sep 17 00:00:00 2001
From: Min Li <lm0963hack@gmail.com>
Date: Mon, 17 Apr 2023 10:27:54 +0800
Subject: Bluetooth: L2CAP: fix "bad unlock balance" in l2cap_disconnect_rsp

conn->chan_lock isn't acquired before l2cap_get_chan_by_scid,
if l2cap_get_chan_by_scid returns NULL, then 'bad unlock balance'
is triggered.

Reported-by: syzbot+9519d6b5b79cf7787cf3@syzkaller.appspotmail.com
Link: https://lore.kernel.org/all/000000000000894f5f05f95e9f4d@google.com/
Signed-off-by: Min Li <lm0963hack@gmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/l2cap_core.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 5cc95fd17f7d..376b523c7b26 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -4693,7 +4693,6 @@ static inline int l2cap_disconnect_rsp(struct l2cap_conn *conn,
 
 	chan = l2cap_get_chan_by_scid(conn, scid);
 	if (!chan) {
-		mutex_unlock(&conn->chan_lock);
 		return 0;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 25c150ac103a4ebeed0319994c742a90634ddf18 Mon Sep 17 00:00:00 2001
From: Ruihan Li <lrh2000@pku.edu.cn>
Date: Sun, 16 Apr 2023 16:14:04 +0800
Subject: bluetooth: Perform careful capability checks in hci_sock_ioctl()

Previously, capability was checked using capable(), which verified that the
caller of the ioctl system call had the required capability. In addition,
the result of the check would be stored in the HCI_SOCK_TRUSTED flag,
making it persistent for the socket.

However, malicious programs can abuse this approach by deliberately sharing
an HCI socket with a privileged task. The HCI socket will be marked as
trusted when the privileged task occasionally makes an ioctl call.

This problem can be solved by using sk_capable() to check capability, which
ensures that not only the current task but also the socket opener has the
specified capability, thus reducing the risk of privilege escalation
through the previously identified vulnerability.

Cc: stable@vger.kernel.org
Fixes: f81f5b2db869 ("Bluetooth: Send control open and close messages for HCI raw sockets")
Signed-off-by: Ruihan Li <lrh2000@pku.edu.cn>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_sock.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index 8d8950101cee..1d249d839819 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -1031,7 +1031,14 @@ static int hci_sock_ioctl(struct socket *sock, unsigned int cmd,
 	if (hci_sock_gen_cookie(sk)) {
 		struct sk_buff *skb;
 
-		if (capable(CAP_NET_ADMIN))
+		/* Perform careful checks before setting the HCI_SOCK_TRUSTED
+		 * flag. Make sure that not only the current task but also
+		 * the socket opener has the required capability, since
+		 * privileged programs can be tricked into making ioctl calls
+		 * on HCI sockets, and the socket should not be marked as
+		 * trusted simply because the ioctl caller is privileged.
+		 */
+		if (sk_capable(sk, CAP_NET_ADMIN))
 			hci_sock_set_flag(sk, HCI_SOCK_TRUSTED);
 
 		/* Send event to monitor */
-- 
cgit v1.2.3-70-g09d2


From 0ab905c3be8c2be368821dd2bd1f8d126064d993 Mon Sep 17 00:00:00 2001
From: Zijun Hu <quic_zijuhu@quicinc.com>
Date: Mon, 17 Apr 2023 17:39:59 +0800
Subject: Bluetooth: Devcoredump: Fix storing u32 without specifying byte order
 issue

API hci_devcd_init() stores its u32 type parameter @dump_size into
skb, but it does not specify which byte order is used to store the
integer, let us take little endian to store and parse the integer.

Fixes: f5cc609d09d4 ("Bluetooth: Add support for hci devcoredump")
Signed-off-by: Zijun Hu <quic_zijuhu@quicinc.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/coredump.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/net/bluetooth/coredump.c b/net/bluetooth/coredump.c
index 08fa98505454..d2d2624ec708 100644
--- a/net/bluetooth/coredump.c
+++ b/net/bluetooth/coredump.c
@@ -5,6 +5,7 @@
 
 #include <linux/devcoredump.h>
 
+#include <asm/unaligned.h>
 #include <net/bluetooth/bluetooth.h>
 #include <net/bluetooth/hci_core.h>
 
@@ -180,25 +181,25 @@ hdr_free:
 
 static void hci_devcd_handle_pkt_init(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	u32 *dump_size;
+	u32 dump_size;
 
 	if (hdev->dump.state != HCI_DEVCOREDUMP_IDLE) {
 		DBG_UNEXPECTED_STATE();
 		return;
 	}
 
-	if (skb->len != sizeof(*dump_size)) {
+	if (skb->len != sizeof(dump_size)) {
 		bt_dev_dbg(hdev, "Invalid dump init pkt");
 		return;
 	}
 
-	dump_size = skb_pull_data(skb, sizeof(*dump_size));
-	if (!*dump_size) {
+	dump_size = get_unaligned_le32(skb_pull_data(skb, 4));
+	if (!dump_size) {
 		bt_dev_err(hdev, "Zero size dump init pkt");
 		return;
 	}
 
-	if (hci_devcd_prepare(hdev, *dump_size)) {
+	if (hci_devcd_prepare(hdev, dump_size)) {
 		bt_dev_err(hdev, "Failed to prepare for dump");
 		return;
 	}
@@ -441,7 +442,7 @@ int hci_devcd_init(struct hci_dev *hdev, u32 dump_size)
 		return -ENOMEM;
 
 	hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_INIT;
-	skb_put_data(skb, &dump_size, sizeof(dump_size));
+	put_unaligned_le32(dump_size, skb_put(skb, 4));
 
 	skb_queue_tail(&hdev->dump.dump_q, skb);
 	queue_work(hdev->workqueue, &hdev->dump.dump_rx);
-- 
cgit v1.2.3-70-g09d2


From 9a24ce5e29b15c4c6b0c89c04f9df6ce14addefa Mon Sep 17 00:00:00 2001
From: Max Chou <max.chou@realtek.com>
Date: Mon, 17 Apr 2023 10:33:58 +0800
Subject: Bluetooth: btrtl: Firmware format v2 support

Realtek changed the format of the firmware file as v2. The driver
should implement the patch to extract the firmware data from the
firmware file. The future chips must apply this patch for firmware loading.
This patch is compatible with the both previous format and v2 as well.

Signed-off-by: Allen Chen <allen_chen@realsil.com.cn>
Signed-off-by: Alex Lu <alex_lu@realsil.com.cn>
Tested-by: Hilda Wu <hildawu@realtek.com>
Signed-off-by: Max Chou <max.chou@realtek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c | 350 ++++++++++++++++++++++++++++++++++++++++------
 drivers/bluetooth/btrtl.h |  53 ++++++-
 2 files changed, 362 insertions(+), 41 deletions(-)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index 0f256a8abef4..2ced74f8f87d 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -21,6 +21,7 @@
 #define RTL_CHIP_8723CS_VF	4
 #define RTL_CHIP_8723CS_XX	5
 #define RTL_EPATCH_SIGNATURE	"Realtech"
+#define RTL_EPATCH_SIGNATURE_V2	"RTBTCore"
 #define RTL_ROM_LMP_8703B	0x8703
 #define RTL_ROM_LMP_8723A	0x1200
 #define RTL_ROM_LMP_8723B	0x8723
@@ -43,6 +44,14 @@
 	.hci_ver = (hciv), \
 	.hci_bus = (bus)
 
+#define	RTL_CHIP_SUBVER (&(struct rtl_vendor_cmd) {{0x10, 0x38, 0x04, 0x28, 0x80}})
+#define	RTL_CHIP_REV    (&(struct rtl_vendor_cmd) {{0x10, 0x3A, 0x04, 0x28, 0x80}})
+#define	RTL_SEC_PROJ    (&(struct rtl_vendor_cmd) {{0x10, 0xA4, 0x0D, 0x00, 0xb0}})
+
+#define RTL_PATCH_SNIPPETS		0x01
+#define RTL_PATCH_DUMMY_HEADER		0x02
+#define RTL_PATCH_SECURITY_HEADER	0x03
+
 enum btrtl_chip_id {
 	CHIP_ID_8723A,
 	CHIP_ID_8723B,
@@ -81,6 +90,8 @@ struct btrtl_device_info {
 	int cfg_len;
 	bool drop_fw;
 	int project_id;
+	u8 key_id;
+	struct list_head patch_subsecs;
 };
 
 static const struct id_table ic_id_table[] = {
@@ -343,6 +354,227 @@ static int rtl_read_rom_version(struct hci_dev *hdev, u8 *version)
 	return 0;
 }
 
+static int btrtl_vendor_read_reg16(struct hci_dev *hdev,
+				   struct rtl_vendor_cmd *cmd, u8 *rp)
+{
+	struct sk_buff *skb;
+	int err = 0;
+
+	skb = __hci_cmd_sync(hdev, 0xfc61, sizeof(*cmd), cmd,
+			     HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		err = PTR_ERR(skb);
+		rtl_dev_err(hdev, "RTL: Read reg16 failed (%d)", err);
+		return err;
+	}
+
+	if (skb->len != 3 || skb->data[0]) {
+		bt_dev_err(hdev, "RTL: Read reg16 length mismatch");
+		kfree_skb(skb);
+		return -EIO;
+	}
+
+	if (rp)
+		memcpy(rp, skb->data + 1, 2);
+
+	kfree_skb(skb);
+
+	return 0;
+}
+
+static void *rtl_iov_pull_data(struct rtl_iovec *iov, u32 len)
+{
+	void *data = iov->data;
+
+	if (iov->len < len)
+		return NULL;
+
+	iov->data += len;
+	iov->len  -= len;
+
+	return data;
+}
+
+static void btrtl_insert_ordered_subsec(struct rtl_subsection *node,
+					struct btrtl_device_info *btrtl_dev)
+{
+	struct list_head *pos;
+	struct list_head *next;
+	struct rtl_subsection *subsec;
+
+	list_for_each_safe(pos, next, &btrtl_dev->patch_subsecs) {
+		subsec = list_entry(pos, struct rtl_subsection, list);
+		if (subsec->prio >= node->prio)
+			break;
+	}
+	__list_add(&node->list, pos->prev, pos);
+}
+
+static int btrtl_parse_section(struct hci_dev *hdev,
+			       struct btrtl_device_info *btrtl_dev, u32 opcode,
+			       u8 *data, u32 len)
+{
+	struct rtl_section_hdr *hdr;
+	struct rtl_subsection *subsec;
+	struct rtl_common_subsec *common_subsec;
+	struct rtl_sec_hdr *sec_hdr;
+	int i;
+	u8 *ptr;
+	u16 num_subsecs;
+	u32 subsec_len;
+	int rc = 0;
+	struct rtl_iovec iov = {
+		.data = data,
+		.len  = len,
+	};
+
+	hdr = rtl_iov_pull_data(&iov, sizeof(*hdr));
+	if (!hdr)
+		return -EINVAL;
+	num_subsecs = le16_to_cpu(hdr->num);
+
+	for (i = 0; i < num_subsecs; i++) {
+		common_subsec = rtl_iov_pull_data(&iov, sizeof(*common_subsec));
+		if (!common_subsec)
+			break;
+		subsec_len = le32_to_cpu(common_subsec->len);
+
+		rtl_dev_dbg(hdev, "subsec, eco 0x%02x, len %08x",
+			    common_subsec->eco, subsec_len);
+
+		ptr = rtl_iov_pull_data(&iov, subsec_len);
+		if (!ptr)
+			break;
+
+		if (common_subsec->eco != btrtl_dev->rom_version + 1)
+			continue;
+
+		switch (opcode) {
+		case RTL_PATCH_SECURITY_HEADER:
+			sec_hdr = (void *)common_subsec;
+			if (sec_hdr->key_id != btrtl_dev->key_id)
+				continue;
+			break;
+		}
+
+		subsec = kzalloc(sizeof(*subsec), GFP_KERNEL);
+		if (!subsec)
+			return -ENOMEM;
+		subsec->opcode = opcode;
+		subsec->prio = common_subsec->prio;
+		subsec->len  = subsec_len;
+		subsec->data = ptr;
+		btrtl_insert_ordered_subsec(subsec, btrtl_dev);
+		rc  += subsec_len;
+	}
+
+	return rc;
+}
+
+static int rtlbt_parse_firmware_v2(struct hci_dev *hdev,
+				   struct btrtl_device_info *btrtl_dev,
+				   unsigned char **_buf)
+{
+	struct rtl_epatch_header_v2 *hdr;
+	int rc;
+	u8 reg_val[2];
+	u8 key_id;
+	u32 num_sections;
+	struct rtl_section *section;
+	struct rtl_subsection *entry, *tmp;
+	u32 section_len;
+	u32 opcode;
+	int len = 0;
+	int i;
+	u8 *ptr;
+	struct rtl_iovec iov = {
+		.data = btrtl_dev->fw_data,
+		.len  = btrtl_dev->fw_len - 7, /* Cut the tail */
+	};
+
+	rc = btrtl_vendor_read_reg16(hdev, RTL_SEC_PROJ, reg_val);
+	if (rc < 0)
+		return -EIO;
+	key_id = reg_val[0];
+
+	rtl_dev_dbg(hdev, "%s: key id %u", __func__, key_id);
+
+	btrtl_dev->key_id = key_id;
+
+	hdr = rtl_iov_pull_data(&iov, sizeof(*hdr));
+	if (!hdr)
+		return -EINVAL;
+	num_sections = le32_to_cpu(hdr->num_sections);
+
+	rtl_dev_dbg(hdev, "FW version %08x-%08x", *((u32 *)hdr->fw_version),
+		    *((u32 *)(hdr->fw_version + 4)));
+
+	for (i = 0; i < num_sections; i++) {
+		section = rtl_iov_pull_data(&iov, sizeof(*section));
+		if (!section)
+			break;
+		section_len = le32_to_cpu(section->len);
+		opcode      = le32_to_cpu(section->opcode);
+
+		rtl_dev_dbg(hdev, "opcode 0x%04x", section->opcode);
+
+		ptr = rtl_iov_pull_data(&iov, section_len);
+		if (!ptr)
+			break;
+
+		switch (opcode) {
+		case RTL_PATCH_SNIPPETS:
+			rc = btrtl_parse_section(hdev, btrtl_dev, opcode,
+						 ptr, section_len);
+			break;
+		case RTL_PATCH_SECURITY_HEADER:
+			/* If key_id from chip is zero, ignore all security
+			 * headers.
+			 */
+			if (!key_id)
+				break;
+			rc = btrtl_parse_section(hdev, btrtl_dev, opcode,
+						 ptr, section_len);
+			break;
+		case RTL_PATCH_DUMMY_HEADER:
+			rc = btrtl_parse_section(hdev, btrtl_dev, opcode,
+						 ptr, section_len);
+			break;
+		default:
+			rc = 0;
+			break;
+		}
+		if (rc < 0) {
+			rtl_dev_err(hdev, "RTL: Parse section (%u) err %d",
+				    opcode, rc);
+			return rc;
+		}
+		len += rc;
+	}
+
+	if (!len)
+		return -ENODATA;
+
+	/* Allocate mem and copy all found subsecs. */
+	ptr = kvmalloc(len, GFP_KERNEL);
+	if (!ptr)
+		return -ENOMEM;
+
+	len = 0;
+	list_for_each_entry_safe(entry, tmp, &btrtl_dev->patch_subsecs, list) {
+		rtl_dev_dbg(hdev, "RTL: opcode %08x, addr %p, len 0x%x",
+			    entry->opcode, entry->data, entry->len);
+		memcpy(ptr + len, entry->data, entry->len);
+		len += entry->len;
+	}
+
+	if (!len)
+		return -EPERM;
+
+	*_buf = ptr;
+	return len;
+}
+
 static int rtlbt_parse_firmware(struct hci_dev *hdev,
 				struct btrtl_device_info *btrtl_dev,
 				unsigned char **_buf)
@@ -377,7 +609,18 @@ static int rtlbt_parse_firmware(struct hci_dev *hdev,
 		{ RTL_ROM_LMP_8852A, 25 },	/* 8852C */
 	};
 
-	min_size = sizeof(struct rtl_epatch_header) + sizeof(extension_sig) + 3;
+	if (btrtl_dev->fw_len <= 8)
+		return -EINVAL;
+
+	if (!memcmp(btrtl_dev->fw_data, RTL_EPATCH_SIGNATURE, 8))
+		min_size = sizeof(struct rtl_epatch_header) +
+				sizeof(extension_sig) + 3;
+	else if (!memcmp(btrtl_dev->fw_data, RTL_EPATCH_SIGNATURE_V2, 8))
+		min_size = sizeof(struct rtl_epatch_header_v2) +
+				sizeof(extension_sig) + 3;
+	else
+		return -EINVAL;
+
 	if (btrtl_dev->fw_len < min_size)
 		return -EINVAL;
 
@@ -442,12 +685,14 @@ static int rtlbt_parse_firmware(struct hci_dev *hdev,
 		return -EINVAL;
 	}
 
-	epatch_info = (struct rtl_epatch_header *)btrtl_dev->fw_data;
-	if (memcmp(epatch_info->signature, RTL_EPATCH_SIGNATURE, 8) != 0) {
+	if (memcmp(btrtl_dev->fw_data, RTL_EPATCH_SIGNATURE, 8) != 0) {
+		if (!memcmp(btrtl_dev->fw_data, RTL_EPATCH_SIGNATURE_V2, 8))
+			return rtlbt_parse_firmware_v2(hdev, btrtl_dev, _buf);
 		rtl_dev_err(hdev, "bad EPATCH signature");
 		return -EINVAL;
 	}
 
+	epatch_info = (struct rtl_epatch_header *)btrtl_dev->fw_data;
 	num_patches = le16_to_cpu(epatch_info->num_patches);
 	BT_DBG("fw_version=%x, num_patches=%d",
 	       le32_to_cpu(epatch_info->fw_version), num_patches);
@@ -511,6 +756,7 @@ static int rtl_download_firmware(struct hci_dev *hdev,
 	int frag_len = RTL_FRAG_LEN;
 	int ret = 0;
 	int i;
+	int j = 0;
 	struct sk_buff *skb;
 	struct hci_rp_read_local_version *rp;
 
@@ -521,17 +767,16 @@ static int rtl_download_firmware(struct hci_dev *hdev,
 	for (i = 0; i < frag_num; i++) {
 		struct sk_buff *skb;
 
-		BT_DBG("download fw (%d/%d)", i, frag_num);
-
-		if (i > 0x7f)
-			dl_cmd->index = (i & 0x7f) + 1;
-		else
-			dl_cmd->index = i;
+		dl_cmd->index = j++;
+		if (dl_cmd->index == 0x7f)
+			j = 1;
 
 		if (i == (frag_num - 1)) {
 			dl_cmd->index |= 0x80; /* data end */
 			frag_len = fw_len % RTL_FRAG_LEN;
 		}
+		rtl_dev_dbg(hdev, "download fw (%d/%d). index = %d", i,
+				frag_num, dl_cmd->index);
 		memcpy(dl_cmd->data, data, frag_len);
 
 		/* Send download command */
@@ -691,8 +936,16 @@ static int rtl_read_chip_type(struct hci_dev *hdev, u8 *type)
 
 void btrtl_free(struct btrtl_device_info *btrtl_dev)
 {
+	struct rtl_subsection *entry, *tmp;
+
 	kvfree(btrtl_dev->fw_data);
 	kvfree(btrtl_dev->cfg_data);
+
+	list_for_each_entry_safe(entry, tmp, &btrtl_dev->patch_subsecs, list) {
+		list_del(&entry->list);
+		kfree(entry);
+	}
+
 	kfree(btrtl_dev);
 }
 EXPORT_SYMBOL_GPL(btrtl_free);
@@ -705,10 +958,11 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 	struct hci_rp_read_local_version *resp;
 	char cfg_name[40];
 	u16 hci_rev, lmp_subver;
-	u8 hci_ver, chip_type = 0;
+	u8 hci_ver, lmp_ver, chip_type = 0;
 	int ret;
 	u16 opcode;
 	u8 cmd[2];
+	u8 reg_val[2];
 
 	btrtl_dev = kzalloc(sizeof(*btrtl_dev), GFP_KERNEL);
 	if (!btrtl_dev) {
@@ -716,6 +970,31 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 		goto err_alloc;
 	}
 
+	INIT_LIST_HEAD(&btrtl_dev->patch_subsecs);
+
+check_version:
+	ret = btrtl_vendor_read_reg16(hdev, RTL_CHIP_SUBVER, reg_val);
+	if (ret < 0)
+		goto err_free;
+	lmp_subver = get_unaligned_le16(reg_val);
+
+	if (lmp_subver == RTL_ROM_LMP_8822B) {
+		ret = btrtl_vendor_read_reg16(hdev, RTL_CHIP_REV, reg_val);
+		if (ret < 0)
+			goto err_free;
+		hci_rev = get_unaligned_le16(reg_val);
+
+		/* 8822E */
+		if (hci_rev == 0x000e) {
+			hci_ver = 0x0c;
+			lmp_ver = 0x0c;
+			btrtl_dev->ic_info = btrtl_match_ic(lmp_subver, hci_rev,
+							    hci_ver, hdev->bus,
+							    chip_type);
+			goto next;
+		}
+	}
+
 	skb = btrtl_read_local_version(hdev);
 	if (IS_ERR(skb)) {
 		ret = PTR_ERR(skb);
@@ -723,14 +1002,14 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 	}
 
 	resp = (struct hci_rp_read_local_version *)skb->data;
-	rtl_dev_info(hdev, "examining hci_ver=%02x hci_rev=%04x lmp_ver=%02x lmp_subver=%04x",
-		     resp->hci_ver, resp->hci_rev,
-		     resp->lmp_ver, resp->lmp_subver);
 
-	hci_ver = resp->hci_ver;
-	hci_rev = le16_to_cpu(resp->hci_rev);
+	hci_ver    = resp->hci_ver;
+	hci_rev    = le16_to_cpu(resp->hci_rev);
+	lmp_ver    = resp->lmp_ver;
 	lmp_subver = le16_to_cpu(resp->lmp_subver);
 
+	kfree_skb(skb);
+
 	if (rtl_has_chip_type(lmp_subver)) {
 		ret = rtl_read_chip_type(hdev, &chip_type);
 		if (ret)
@@ -740,8 +1019,15 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 	btrtl_dev->ic_info = btrtl_match_ic(lmp_subver, hci_rev, hci_ver,
 					    hdev->bus, chip_type);
 
-	if (!btrtl_dev->ic_info)
+next:
+	rtl_dev_info(hdev, "examining hci_ver=%02x hci_rev=%04x lmp_ver=%02x lmp_subver=%04x",
+		     hci_ver, hci_rev,
+		     lmp_ver, lmp_subver);
+
+	if (!btrtl_dev->ic_info && !btrtl_dev->drop_fw)
 		btrtl_dev->drop_fw = true;
+	else
+		btrtl_dev->drop_fw = false;
 
 	if (btrtl_dev->drop_fw) {
 		opcode = hci_opcode_pack(0x3f, 0x66);
@@ -750,41 +1036,25 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 
 		skb = bt_skb_alloc(sizeof(cmd), GFP_KERNEL);
 		if (!skb)
-			goto out_free;
+			goto err_free;
 
 		skb_put_data(skb, cmd, sizeof(cmd));
 		hci_skb_pkt_type(skb) = HCI_COMMAND_PKT;
 
-		hdev->send(hdev, skb);
+		ret = hdev->send(hdev, skb);
+		if (ret < 0) {
+			bt_dev_err(hdev, "sending frame failed (%d)", ret);
+			kfree_skb(skb);
+			goto err_free;
+		}
 
 		/* Ensure the above vendor command is sent to controller and
 		 * process has done.
 		 */
 		msleep(200);
 
-		/* Read the local version again. Expect to have the vanilla
-		 * version as cold boot.
-		 */
-		skb = btrtl_read_local_version(hdev);
-		if (IS_ERR(skb)) {
-			ret = PTR_ERR(skb);
-			goto err_free;
-		}
-
-		resp = (struct hci_rp_read_local_version *)skb->data;
-		rtl_dev_info(hdev, "examining hci_ver=%02x hci_rev=%04x lmp_ver=%02x lmp_subver=%04x",
-			     resp->hci_ver, resp->hci_rev,
-			     resp->lmp_ver, resp->lmp_subver);
-
-		hci_ver = resp->hci_ver;
-		hci_rev = le16_to_cpu(resp->hci_rev);
-		lmp_subver = le16_to_cpu(resp->lmp_subver);
-
-		btrtl_dev->ic_info = btrtl_match_ic(lmp_subver, hci_rev, hci_ver,
-						    hdev->bus, chip_type);
+		goto check_version;
 	}
-out_free:
-	kfree_skb(skb);
 
 	if (!btrtl_dev->ic_info) {
 		rtl_dev_info(hdev, "unknown IC info, lmp subver %04x, hci rev %04x, hci ver %04x",
diff --git a/drivers/bluetooth/btrtl.h b/drivers/bluetooth/btrtl.h
index 349d72ee571b..adb4c2c9abc5 100644
--- a/drivers/bluetooth/btrtl.h
+++ b/drivers/bluetooth/btrtl.h
@@ -49,7 +49,58 @@ struct rtl_vendor_config_entry {
 struct rtl_vendor_config {
 	__le32 signature;
 	__le16 total_len;
-	struct rtl_vendor_config_entry entry[];
+	__u8 entry[];
+} __packed;
+
+struct rtl_epatch_header_v2 {
+	__u8   signature[8];
+	__u8   fw_version[8];
+	__le32 num_sections;
+} __packed;
+
+struct rtl_section {
+	__le32 opcode;
+	__le32 len;
+	u8     data[];
+} __packed;
+
+struct rtl_section_hdr {
+	__le16 num;
+	__le16 reserved;
+} __packed;
+
+struct rtl_common_subsec {
+	__u8   eco;
+	__u8   prio;
+	__u8   cb[2];
+	__le32 len;
+	__u8   data[];
+};
+
+struct rtl_sec_hdr {
+	__u8   eco;
+	__u8   prio;
+	__u8   key_id;
+	__u8   reserved;
+	__le32 len;
+	__u8   data[];
+} __packed;
+
+struct rtl_subsection {
+	struct list_head list;
+	u32 opcode;
+	u32 len;
+	u8 prio;
+	u8 *data;
+};
+
+struct rtl_iovec {
+	u8  *data;
+	u32 len;
+};
+
+struct rtl_vendor_cmd {
+	__u8 param[5];
 } __packed;
 
 enum {
-- 
cgit v1.2.3-70-g09d2


From 9e080b53dafae52759c87a105abcb87aa0592cf4 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 17 Apr 2023 17:01:16 -0700
Subject: Bluetooth: btnxpuart: Fix sparse warnings

This fixes the following sparse warnings:

   drivers/bluetooth/btnxpuart.c:681:23: sparse: sparse:
   restricted __le16 degrades to integer
   drivers/bluetooth/btnxpuart.c:690:82: sparse:
   sparse: incorrect type in argument 2 (different base types)
   @@     expected unsigned short [usertype] req_len
   @@     got restricted __le16 [usertype] len @@
   drivers/bluetooth/btnxpuart.c:690:82: sparse:
   expected unsigned short [usertype] req_len
   drivers/bluetooth/btnxpuart.c:690:82: sparse:
   got restricted __le16 [usertype] len
   drivers/bluetooth/btnxpuart.c:694:84: sparse:
   sparse: incorrect type in argument 2 (different base types)
   @@     expected unsigned short [usertype] req_len
   @@     got restricted __le16 [usertype] len @@
   drivers/bluetooth/btnxpuart.c:694:84: sparse:
   expected unsigned short [usertype] req_len
   drivers/bluetooth/btnxpuart.c:694:84: sparse:
   got restricted __le16 [usertype] len
   drivers/bluetooth/btnxpuart.c:708:23: sparse:
   sparse: incorrect type in assignment (different base types)
   @@     expected unsigned int [usertype] requested_len
   @@     got restricted __le16 [usertype] len @@
   drivers/bluetooth/btnxpuart.c:708:23: sparse:
   expected unsigned int [usertype] requested_len
   drivers/bluetooth/btnxpuart.c:708:23: sparse:
   got restricted __le16 [usertype] len
   drivers/bluetooth/btnxpuart.c:787:78: sparse:
   sparse: incorrect type in argument 2 (different base types)
   @@     expected unsigned short [usertype] chipid
   @@     got restricted __le16 [usertype] chip_id @@
   drivers/bluetooth/btnxpuart.c:787:78: sparse:
   expected unsigned short [usertype] chipid
   drivers/bluetooth/btnxpuart.c:787:78: sparse:
   got restricted __le16 [usertype] chip_id
   drivers/bluetooth/btnxpuart.c:810:74: sparse:
   sparse: incorrect type in argument 2 (different base types)
   @@     expected unsigned short [usertype] req_len
   @@     got restricted __le16 [usertype] len @@
   drivers/bluetooth/btnxpuart.c:810:74: sparse:
   expected unsigned short [usertype] req_len
   drivers/bluetooth/btnxpuart.c:810:74: sparse:
   got restricted __le16 [usertype] len
   drivers/bluetooth/btnxpuart.c:815:76: sparse:
   sparse: incorrect type in argument 2 (different base types)
   @@     expected unsigned short [usertype] req_len
   @@     got restricted __le16 [usertype] len @@
   drivers/bluetooth/btnxpuart.c:815:76: sparse:
   expected unsigned short [usertype] req_len
   drivers/bluetooth/btnxpuart.c:815:76: sparse:
   got restricted __le16 [usertype] len
   drivers/bluetooth/btnxpuart.c:834:16: sparse:
   sparse: restricted __le32 degrades to integer
   drivers/bluetooth/btnxpuart.c:843:55: sparse:
   sparse: restricted __le32 degrades to integer
   drivers/bluetooth/btnxpuart.c:844:36: sparse:
   sparse: incorrect type in argument 3 (different base types)
   @@     expected unsigned long [usertype]
   @@     got restricted __le16 [usertype] len @@
   drivers/bluetooth/btnxpuart.c:844:36: sparse:
   expected unsigned long [usertype]
   drivers/bluetooth/btnxpuart.c:844:36: sparse:
   got restricted __le16 [usertype] len

Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304160736.Tsa0zTBU-lkp@intel.com/
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 85 ++++++++++++++++++++++++++-----------------
 1 file changed, 51 insertions(+), 34 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 93f3afc0c0c8..31046f26e712 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -683,12 +683,14 @@ static int nxp_recv_chip_ver_v1(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
 	struct v1_start_ind *req;
+	__u16 chip_id;
 
-	req = (struct v1_start_ind *)skb_pull_data(skb, sizeof(struct v1_start_ind));
+	req = skb_pull_data(skb, sizeof(*req));
 	if (!req)
 		goto free_skb;
 
-	if ((req->chip_id ^ req->chip_id_comp) == 0xffff) {
+	chip_id = le16_to_cpu(req->chip_id ^ req->chip_id_comp);
+	if (chip_id == 0xffff) {
 		nxpdev->fw_dnld_v1_offset = 0;
 		nxpdev->fw_v1_sent_bytes = 0;
 		nxpdev->fw_v1_expected_len = HDR_LEN;
@@ -707,29 +709,34 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
 	struct btnxpuart_data *nxp_data = nxpdev->nxp_data;
 	struct v1_data_req *req;
-	u32 requested_len;
+	__u16 len;
 
 	if (!process_boot_signature(nxpdev))
 		goto free_skb;
 
-	req = (struct v1_data_req *)skb_pull_data(skb, sizeof(struct v1_data_req));
+	req = skb_pull_data(skb, sizeof(*req));
 	if (!req)
 		goto free_skb;
 
-	if ((req->len ^ req->len_comp) != 0xffff) {
+	len = __le16_to_cpu(req->len ^ req->len_comp);
+	if (len != 0xffff) {
 		bt_dev_dbg(hdev, "ERR: Send NAK");
 		nxp_send_ack(NXP_NAK_V1, hdev);
 		goto free_skb;
 	}
 	nxp_send_ack(NXP_ACK_V1, hdev);
 
+	len = __le16_to_cpu(req->len);
+
 	if (!nxp_data->helper_fw_name) {
 		if (!nxpdev->timeout_changed) {
-			nxpdev->timeout_changed = nxp_fw_change_timeout(hdev, req->len);
+			nxpdev->timeout_changed = nxp_fw_change_timeout(hdev,
+									len);
 			goto free_skb;
 		}
 		if (!nxpdev->baudrate_changed) {
-			nxpdev->baudrate_changed = nxp_fw_change_baudrate(hdev, req->len);
+			nxpdev->baudrate_changed = nxp_fw_change_baudrate(hdev,
+									  len);
 			if (nxpdev->baudrate_changed) {
 				serdev_device_set_baudrate(nxpdev->serdev,
 							   HCI_NXP_SEC_BAUDRATE);
@@ -748,13 +755,14 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 			goto free_skb;
 	}
 
-	requested_len = req->len;
-	if (requested_len == 0) {
-		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes", nxpdev->fw->size);
+	if (!len) {
+		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes",
+			   nxpdev->fw->size);
 		if (nxp_data->helper_fw_name && !nxpdev->helper_downloaded) {
 			nxpdev->helper_downloaded = true;
 			serdev_device_wait_until_sent(nxpdev->serdev, 0);
-			serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_SEC_BAUDRATE);
+			serdev_device_set_baudrate(nxpdev->serdev,
+						   HCI_NXP_SEC_BAUDRATE);
 			serdev_device_set_flow_control(nxpdev->serdev, 1);
 		} else {
 			clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
@@ -762,12 +770,12 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 		}
 		goto free_skb;
 	}
-	if (requested_len & 0x01) {
+	if (len & 0x01) {
 		/* The CRC did not match at the other end.
 		 * Simply send the same bytes again.
 		 */
-		requested_len = nxpdev->fw_v1_sent_bytes;
-		bt_dev_dbg(hdev, "CRC error. Resend %d bytes of FW.", requested_len);
+		len = nxpdev->fw_v1_sent_bytes;
+		bt_dev_dbg(hdev, "CRC error. Resend %d bytes of FW.", len);
 	} else {
 		nxpdev->fw_dnld_v1_offset += nxpdev->fw_v1_sent_bytes;
 
@@ -781,24 +789,23 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 		 * mismatch, clearly the driver and FW are out of sync,
 		 * and we need to re-send the previous header again.
 		 */
-		if (requested_len == nxpdev->fw_v1_expected_len) {
-			if (requested_len == HDR_LEN)
+		if (len == nxpdev->fw_v1_expected_len) {
+			if (len == HDR_LEN)
 				nxpdev->fw_v1_expected_len = nxp_get_data_len(nxpdev->fw->data +
 									nxpdev->fw_dnld_v1_offset);
 			else
 				nxpdev->fw_v1_expected_len = HDR_LEN;
-		} else if (requested_len == HDR_LEN) {
+		} else if (len == HDR_LEN) {
 			/* FW download out of sync. Send previous chunk again */
 			nxpdev->fw_dnld_v1_offset -= nxpdev->fw_v1_sent_bytes;
 			nxpdev->fw_v1_expected_len = HDR_LEN;
 		}
 	}
 
-	if (nxpdev->fw_dnld_v1_offset + requested_len <= nxpdev->fw->size)
-		serdev_device_write_buf(nxpdev->serdev,
-					nxpdev->fw->data + nxpdev->fw_dnld_v1_offset,
-					requested_len);
-	nxpdev->fw_v1_sent_bytes = requested_len;
+	if (nxpdev->fw_dnld_v1_offset + len <= nxpdev->fw->size)
+		serdev_device_write_buf(nxpdev->serdev, nxpdev->fw->data +
+					nxpdev->fw_dnld_v1_offset, len);
+	nxpdev->fw_v1_sent_bytes = len;
 
 free_skb:
 	kfree_skb(skb);
@@ -828,13 +835,16 @@ static char *nxp_get_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid)
 
 static int nxp_recv_chip_ver_v3(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	struct v3_start_ind *req = skb_pull_data(skb, sizeof(struct v3_start_ind));
+	struct v3_start_ind *req = skb_pull_data(skb, sizeof(*req));
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	u16 chip_id;
 
 	if (!process_boot_signature(nxpdev))
 		goto free_skb;
 
-	if (!nxp_request_firmware(hdev, nxp_get_fw_name_from_chipid(hdev, req->chip_id)))
+	chip_id = le16_to_cpu(req->chip_id);
+	if (!nxp_request_firmware(hdev, nxp_get_fw_name_from_chipid(hdev,
+								    chip_id)))
 		nxp_send_ack(NXP_ACK_V3, hdev);
 
 free_skb:
@@ -846,23 +856,27 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
 	struct v3_data_req *req;
+	__u16 len;
+	__u32 offset;
 
 	if (!process_boot_signature(nxpdev))
 		goto free_skb;
 
-	req = (struct v3_data_req *)skb_pull_data(skb, sizeof(struct v3_data_req));
+	req = skb_pull_data(skb, sizeof(*req));
 	if (!req || !nxpdev->fw)
 		goto free_skb;
 
 	nxp_send_ack(NXP_ACK_V3, hdev);
 
+	len = __le16_to_cpu(req->len);
+
 	if (!nxpdev->timeout_changed) {
-		nxpdev->timeout_changed = nxp_fw_change_timeout(hdev, req->len);
+		nxpdev->timeout_changed = nxp_fw_change_timeout(hdev, len);
 		goto free_skb;
 	}
 
 	if (!nxpdev->baudrate_changed) {
-		nxpdev->baudrate_changed = nxp_fw_change_baudrate(hdev, req->len);
+		nxpdev->baudrate_changed = nxp_fw_change_baudrate(hdev, len);
 		if (nxpdev->baudrate_changed) {
 			serdev_device_set_baudrate(nxpdev->serdev,
 						   HCI_NXP_SEC_BAUDRATE);
@@ -873,25 +887,28 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 	}
 
 	if (req->len == 0) {
-		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes", nxpdev->fw->size);
+		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes",
+			   nxpdev->fw->size);
 		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
 		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
 		goto free_skb;
 	}
 	if (req->error)
-		bt_dev_dbg(hdev, "FW Download received err 0x%02x from chip", req->error);
+		bt_dev_dbg(hdev, "FW Download received err 0x%02x from chip",
+			   req->error);
 
-	if (req->offset < nxpdev->fw_v3_offset_correction) {
+	offset = __le32_to_cpu(req->offset);
+	if (offset < nxpdev->fw_v3_offset_correction) {
 		/* This scenario should ideally never occur. But if it ever does,
 		 * FW is out of sync and needs a power cycle.
 		 */
-		bt_dev_err(hdev, "Something went wrong during FW download. Please power cycle and try again");
+		bt_dev_err(hdev, "Something went wrong during FW download");
+		bt_dev_err(hdev, "Please power cycle and try again");
 		goto free_skb;
 	}
 
-	serdev_device_write_buf(nxpdev->serdev,
-				nxpdev->fw->data + req->offset - nxpdev->fw_v3_offset_correction,
-				req->len);
+	serdev_device_write_buf(nxpdev->serdev, nxpdev->fw->data + offset -
+				nxpdev->fw_v3_offset_correction, len);
 
 free_skb:
 	kfree_skb(skb);
-- 
cgit v1.2.3-70-g09d2


From 7948fe1c92d92313eea5453f83deb7f0141355e8 Mon Sep 17 00:00:00 2001
From: Max Chou <max.chou@realtek.com>
Date: Tue, 18 Apr 2023 13:43:54 +0800
Subject: Bluetooth: btrtl: Add the support for RTL8851B

Add the support for RTL8851B BT controller on USB interface.
The necessary firmware will be submitted to linux-firmware project.

Note that the Bluetooth devices WITH the VID=0x0bda would be set the
feature quirk in btrtl_setup_realtek(). It's able to ignore the
feature flag set for the specific VID and PID in blacklist_table[] of
btusb.c. (check [1])

If Realtek Bluetooth chips WITHOUT the VID=0x0bda, it shall be added
the feature flag for the specific VID and PID in blacklist_table[] of
btusb.c. (check [2])

[1] '9ab9235fe5cf ("Bluetooth: btrtl: Enable WBS for the specific
    Realtek devices")'
[2] '73280f13c9bb ("Bluetooth: btusb: Add the more support IDs for
    Realtek RTL8822CE")'

The device info from /sys/kernel/debug/usb/devices as below.

T:  Bus=03 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 33 Spd=480  MxCh= 0
D:  Ver= 2.00 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=0bda ProdID=b851 Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=802.11ax WLAN Adapter
S:  SerialNumber=00E04C885A01
C:* #Ifs= 3 Cfg#= 1 Atr=80 MxPwr=500mA
A:  FirstIf#= 0 IfCount= 2 Cls=e0(wlcon) Sub=01 Prot=01
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:* If#= 2 Alt= 0 #EPs= 8 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=05(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=06(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=07(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=09(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0a(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0b(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0c(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms

Signed-off-by: Max Chou <max.chou@realtek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index 2ced74f8f87d..2915c82d719d 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -29,6 +29,7 @@
 #define RTL_ROM_LMP_8761A	0x8761
 #define RTL_ROM_LMP_8822B	0x8822
 #define RTL_ROM_LMP_8852A	0x8852
+#define RTL_ROM_LMP_8851B	0x8851
 #define RTL_CONFIG_MAGIC	0x8723ab55
 
 #define IC_MATCH_FL_LMPSUBV	(1 << 0)
@@ -65,6 +66,7 @@ enum btrtl_chip_id {
 	CHIP_ID_8852A = 18,
 	CHIP_ID_8852B = 20,
 	CHIP_ID_8852C = 25,
+	CHIP_ID_8851B = 36,
 };
 
 struct id_table {
@@ -271,6 +273,14 @@ static const struct id_table ic_id_table[] = {
 	  .has_msft_ext = true,
 	  .fw_name  = "rtl_bt/rtl8852cu_fw.bin",
 	  .cfg_name = "rtl_bt/rtl8852cu_config" },
+
+	/* 8851B */
+	{ IC_INFO(RTL_ROM_LMP_8851B, 0xb, 0xc, HCI_USB),
+	  .config_needed = false,
+	  .has_rom_version = true,
+	  .has_msft_ext = false,
+	  .fw_name  = "rtl_bt/rtl8851bu_fw.bin",
+	  .cfg_name = "rtl_bt/rtl8851bu_config" },
 	};
 
 static const struct id_table *btrtl_match_ic(u16 lmp_subver, u16 hci_rev,
@@ -607,6 +617,7 @@ static int rtlbt_parse_firmware(struct hci_dev *hdev,
 		{ RTL_ROM_LMP_8852A, 18 },	/* 8852A */
 		{ RTL_ROM_LMP_8852A, 20 },	/* 8852B */
 		{ RTL_ROM_LMP_8852A, 25 },	/* 8852C */
+		{ RTL_ROM_LMP_8851B, 36 },	/* 8851B */
 	};
 
 	if (btrtl_dev->fw_len <= 8)
@@ -1134,6 +1145,7 @@ int btrtl_download_firmware(struct hci_dev *hdev,
 	case RTL_ROM_LMP_8822B:
 	case RTL_ROM_LMP_8852A:
 	case RTL_ROM_LMP_8703B:
+	case RTL_ROM_LMP_8851B:
 		return btrtl_setup_rtl8723b(hdev, btrtl_dev);
 	default:
 		rtl_dev_info(hdev, "assuming no firmware upload needed");
@@ -1158,6 +1170,7 @@ void btrtl_set_quirks(struct hci_dev *hdev, struct btrtl_device_info *btrtl_dev)
 	case CHIP_ID_8852A:
 	case CHIP_ID_8852B:
 	case CHIP_ID_8852C:
+	case CHIP_ID_8851B:
 		set_bit(HCI_QUIRK_VALID_LE_STATES, &hdev->quirks);
 		set_bit(HCI_QUIRK_WIDEBAND_SPEECH_SUPPORTED, &hdev->quirks);
 
@@ -1370,3 +1383,5 @@ MODULE_FIRMWARE("rtl_bt/rtl8852bu_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852bu_config.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852cu_fw.bin");
 MODULE_FIRMWARE("rtl_bt/rtl8852cu_config.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8851bu_fw.bin");
+MODULE_FIRMWARE("rtl_bt/rtl8851bu_config.bin");
-- 
cgit v1.2.3-70-g09d2


From f419863588217f76eaf754e1dfce21ea7fcb026d Mon Sep 17 00:00:00 2001
From: Archie Pusaka <apusaka@chromium.org>
Date: Thu, 20 Apr 2023 20:23:36 +0800
Subject: Bluetooth: Cancel sync command before suspend and power off

Some of the sync commands might take a long time to complete, e.g.
LE Create Connection when the peer device isn't responding might take
20 seconds before it times out. If suspend command is issued during
this time, it will need to wait for completion since both commands are
using the same sync lock.

This patch cancel any running sync commands before attempting to
suspend or adapter power off.

Signed-off-by: Archie Pusaka <apusaka@chromium.org>
Reviewed-by: Ying Hsu <yinghsu@chromium.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_core.c | 3 +++
 net/bluetooth/mgmt.c     | 4 ++++
 2 files changed, 7 insertions(+)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 393b317ae68f..a856b1051d35 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -2803,6 +2803,9 @@ int hci_suspend_dev(struct hci_dev *hdev)
 	if (mgmt_powering_down(hdev))
 		return 0;
 
+	/* Cancel potentially blocking sync operation before suspend */
+	__hci_cmd_sync_cancel(hdev, -EHOSTDOWN);
+
 	hci_req_sync_lock(hdev);
 	ret = hci_suspend_sync(hdev);
 	hci_req_sync_unlock(hdev);
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 5f8c144c84b8..13c745876b39 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -1399,6 +1399,10 @@ static int set_powered(struct sock *sk, struct hci_dev *hdev, void *data,
 		goto failed;
 	}
 
+	/* Cancel potentially blocking sync operation before power off */
+	if (cp->val == 0x00)
+		__hci_cmd_sync_cancel(hdev, -EHOSTDOWN);
+
 	err = hci_cmd_sync_queue(hdev, set_powered_sync, cmd,
 				 mgmt_set_powered_complete);
 
-- 
cgit v1.2.3-70-g09d2


From b0310d6ed684b862dd29049d7b9cca5589555654 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Wed, 19 Apr 2023 20:54:13 +0530
Subject: Bluetooth: btnxpuart: Enable flow control before checking boot
 signature

This enables flow control before checking for bootloader signature and
deciding whether FW download is needed or not. In case of V1 bootloader
chips w8987 and w8997, it is observed that if WLAN FW is downloaded first
and power save is enabled in wlan core, bootloader signatures are not
emitted by the BT core when the chip is put to sleep. As a result, the
driver skips FW download and subsequent HCI commands get timeout errors
in dmesg as shown below:

[  112.898867] Bluetooth: hci0: Opcode 0x c03 failed: -110
[  114.914865] Bluetooth: hci0: Setting baudrate failed (-110)
[  116.930856] Bluetooth: hci0: Setting wake-up method failed (-110)

By enabling the flow control, the host enables its RTS pin, and an
interrupt in chip's UART peripheral causes the bootloader to wake up,
enabling the bootloader signatures, which then helps in downloading
the bluetooth FW file.

This changes all instances of 0/1 for serdev_device_set_flow_control()
to false/true.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 31046f26e712..3a34d7c1475b 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -535,7 +535,7 @@ static int nxp_download_firmware(struct hci_dev *hdev)
 	nxpdev->helper_downloaded = false;
 
 	serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_PRI_BAUDRATE);
-	serdev_device_set_flow_control(nxpdev->serdev, 0);
+	serdev_device_set_flow_control(nxpdev->serdev, false);
 	nxpdev->current_baudrate = HCI_NXP_PRI_BAUDRATE;
 
 	/* Wait till FW is downloaded and CTS becomes low */
@@ -548,7 +548,7 @@ static int nxp_download_firmware(struct hci_dev *hdev)
 		return -ETIMEDOUT;
 	}
 
-	serdev_device_set_flow_control(nxpdev->serdev, 1);
+	serdev_device_set_flow_control(nxpdev->serdev, true);
 	err = serdev_device_wait_for_cts(nxpdev->serdev, 1, 60000);
 	if (err < 0) {
 		bt_dev_err(hdev, "CTS is still high. FW Download failed.");
@@ -740,7 +740,7 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 			if (nxpdev->baudrate_changed) {
 				serdev_device_set_baudrate(nxpdev->serdev,
 							   HCI_NXP_SEC_BAUDRATE);
-				serdev_device_set_flow_control(nxpdev->serdev, 1);
+				serdev_device_set_flow_control(nxpdev->serdev, true);
 				nxpdev->current_baudrate = HCI_NXP_SEC_BAUDRATE;
 			}
 			goto free_skb;
@@ -763,7 +763,7 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 			serdev_device_wait_until_sent(nxpdev->serdev, 0);
 			serdev_device_set_baudrate(nxpdev->serdev,
 						   HCI_NXP_SEC_BAUDRATE);
-			serdev_device_set_flow_control(nxpdev->serdev, 1);
+			serdev_device_set_flow_control(nxpdev->serdev, true);
 		} else {
 			clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
 			wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
@@ -880,7 +880,7 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 		if (nxpdev->baudrate_changed) {
 			serdev_device_set_baudrate(nxpdev->serdev,
 						   HCI_NXP_SEC_BAUDRATE);
-			serdev_device_set_flow_control(nxpdev->serdev, 1);
+			serdev_device_set_flow_control(nxpdev->serdev, true);
 			nxpdev->current_baudrate = HCI_NXP_SEC_BAUDRATE;
 		}
 		goto free_skb;
@@ -984,7 +984,7 @@ free_skb:
 static int nxp_check_boot_sign(struct btnxpuart_dev *nxpdev)
 {
 	serdev_device_set_baudrate(nxpdev->serdev, HCI_NXP_PRI_BAUDRATE);
-	serdev_device_set_flow_control(nxpdev->serdev, 0);
+	serdev_device_set_flow_control(nxpdev->serdev, true);
 	set_bit(BTNXPUART_CHECK_BOOT_SIGNATURE, &nxpdev->tx_state);
 
 	return wait_event_interruptible_timeout(nxpdev->check_boot_sign_wait_q,
@@ -1012,7 +1012,6 @@ static int nxp_setup(struct hci_dev *hdev)
 		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
 	}
 
-	serdev_device_set_flow_control(nxpdev->serdev, 1);
 	device_property_read_u32(&nxpdev->serdev->dev, "fw-init-baudrate",
 				 &nxpdev->fw_init_baudrate);
 	if (!nxpdev->fw_init_baudrate)
-- 
cgit v1.2.3-70-g09d2


From 20981ce2d5a5f79d17da1ace3a93df7b3c6ba3d6 Mon Sep 17 00:00:00 2001
From: Tim Jiang <quic_tjiang@quicinc.com>
Date: Wed, 19 Apr 2023 11:38:05 +0800
Subject: Bluetooth: btusb: Add WCN6855 devcoredump support

WCN6855 will report memdump via ACL data or HCI event when
it get crashed, so we collect memdump to debug firmware.

Signed-off-by: Tim Jiang <quic_tjiang@quicinc.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 222 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 222 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 3aa189b1986d..2a8e2bb038f5 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -733,6 +733,16 @@ static const struct dmi_system_id btusb_needs_reset_resume_table[] = {
 	{}
 };
 
+struct qca_dump_info {
+	/* fields for dump collection */
+	u16 id_vendor;
+	u16 id_product;
+	u32 fw_version;
+	u32 controller_id;
+	u32 ram_dump_size;
+	u16 ram_dump_seqno;
+};
+
 #define BTUSB_MAX_ISOC_FRAMES	10
 
 #define BTUSB_INTR_RUNNING	0
@@ -752,6 +762,7 @@ static const struct dmi_system_id btusb_needs_reset_resume_table[] = {
 #define BTUSB_WAKEUP_AUTOSUSPEND	14
 #define BTUSB_USE_ALT3_FOR_WBS	15
 #define BTUSB_ALT6_CONTINUOUS_TX	16
+#define BTUSB_HW_SSR_ACTIVE	17
 
 struct btusb_data {
 	struct hci_dev       *hdev;
@@ -814,6 +825,8 @@ struct btusb_data {
 
 	int oob_wake_irq;   /* irq for out-of-band wake-on-bt */
 	unsigned cmd_timeout_cnt;
+
+	struct qca_dump_info qca_dump;
 };
 
 static void btusb_reset(struct hci_dev *hdev)
@@ -904,6 +917,11 @@ static void btusb_qca_cmd_timeout(struct hci_dev *hdev)
 	struct btusb_data *data = hci_get_drvdata(hdev);
 	struct gpio_desc *reset_gpio = data->reset_gpio;
 
+	if (test_bit(BTUSB_HW_SSR_ACTIVE, &data->flags)) {
+		bt_dev_info(hdev, "Ramdump in progress, defer cmd_timeout");
+		return;
+	}
+
 	if (++data->cmd_timeout_cnt < 5)
 		return;
 
@@ -3294,6 +3312,202 @@ static int btusb_set_bdaddr_wcn6855(struct hci_dev *hdev,
 	return 0;
 }
 
+#define QCA_MEMDUMP_ACL_HANDLE 0x2EDD
+#define QCA_MEMDUMP_SIZE_MAX  0x100000
+#define QCA_MEMDUMP_VSE_CLASS 0x01
+#define QCA_MEMDUMP_MSG_TYPE 0x08
+#define QCA_MEMDUMP_PKT_SIZE 248
+#define QCA_LAST_SEQUENCE_NUM 0xffff
+
+struct qca_dump_hdr {
+	u8 vse_class;
+	u8 msg_type;
+	__le16 seqno;
+	u8 reserved;
+	union {
+		u8 data[0];
+		struct {
+			__le32 ram_dump_size;
+			u8 data0[0];
+		} __packed;
+	};
+} __packed;
+
+
+static void btusb_dump_hdr_qca(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	char buf[128];
+	struct btusb_data *btdata = hci_get_drvdata(hdev);
+
+	snprintf(buf, sizeof(buf), "Controller Name: 0x%x\n",
+			btdata->qca_dump.controller_id);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Firmware Version: 0x%x\n",
+			btdata->qca_dump.fw_version);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Driver: %s\nVendor: qca\n",
+			btusb_driver.name);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "VID: 0x%x\nPID:0x%x\n",
+			btdata->qca_dump.id_vendor, btdata->qca_dump.id_product);
+	skb_put_data(skb, buf, strlen(buf));
+
+	snprintf(buf, sizeof(buf), "Lmp Subversion: 0x%x\n",
+			hdev->lmp_subver);
+	skb_put_data(skb, buf, strlen(buf));
+}
+
+static void btusb_coredump_qca(struct hci_dev *hdev)
+{
+	static const u8 param[] = { 0x26 };
+	struct sk_buff *skb;
+
+	skb = __hci_cmd_sync(hdev, 0xfc0c, 1, param, HCI_CMD_TIMEOUT);
+	if (IS_ERR(skb))
+		bt_dev_err(hdev, "%s: triggle crash failed (%ld)", __func__, PTR_ERR(skb));
+	kfree_skb(skb);
+}
+
+/*
+ * ==0: not a dump pkt.
+ * < 0: fails to handle a dump pkt
+ * > 0: otherwise.
+ */
+static int handle_dump_pkt_qca(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	int ret = 1;
+	u8 pkt_type;
+	u8 *sk_ptr;
+	unsigned int sk_len;
+	u16 seqno;
+	u32 dump_size;
+
+	struct hci_event_hdr *event_hdr;
+	struct hci_acl_hdr *acl_hdr;
+	struct qca_dump_hdr *dump_hdr;
+	struct btusb_data *btdata = hci_get_drvdata(hdev);
+	struct usb_device *udev = btdata->udev;
+
+	pkt_type = hci_skb_pkt_type(skb);
+	sk_ptr = skb->data;
+	sk_len = skb->len;
+
+	if (pkt_type == HCI_ACLDATA_PKT) {
+		acl_hdr = hci_acl_hdr(skb);
+		if (le16_to_cpu(acl_hdr->handle) != QCA_MEMDUMP_ACL_HANDLE)
+			return 0;
+		sk_ptr += HCI_ACL_HDR_SIZE;
+		sk_len -= HCI_ACL_HDR_SIZE;
+		event_hdr = (struct hci_event_hdr *)sk_ptr;
+	} else {
+		event_hdr = hci_event_hdr(skb);
+	}
+
+	if ((event_hdr->evt != HCI_VENDOR_PKT)
+		|| (event_hdr->plen != (sk_len - HCI_EVENT_HDR_SIZE)))
+		return 0;
+
+	sk_ptr += HCI_EVENT_HDR_SIZE;
+	sk_len -= HCI_EVENT_HDR_SIZE;
+
+	dump_hdr = (struct qca_dump_hdr *)sk_ptr;
+	if ((sk_len < offsetof(struct qca_dump_hdr, data))
+		|| (dump_hdr->vse_class != QCA_MEMDUMP_VSE_CLASS)
+	    || (dump_hdr->msg_type != QCA_MEMDUMP_MSG_TYPE))
+		return 0;
+
+	/*it is dump pkt now*/
+	seqno = le16_to_cpu(dump_hdr->seqno);
+	if (seqno == 0) {
+		set_bit(BTUSB_HW_SSR_ACTIVE, &btdata->flags);
+		dump_size = le32_to_cpu(dump_hdr->ram_dump_size);
+		if (!dump_size || (dump_size > QCA_MEMDUMP_SIZE_MAX)) {
+			ret = -EILSEQ;
+			bt_dev_err(hdev, "Invalid memdump size(%u)",
+				   dump_size);
+			goto out;
+		}
+
+		ret = hci_devcd_init(hdev, dump_size);
+		if (ret < 0) {
+			bt_dev_err(hdev, "memdump init error(%d)", ret);
+			goto out;
+		}
+
+		btdata->qca_dump.ram_dump_size = dump_size;
+		btdata->qca_dump.ram_dump_seqno = 0;
+		sk_ptr += offsetof(struct qca_dump_hdr, data0);
+		sk_len -= offsetof(struct qca_dump_hdr, data0);
+
+		usb_disable_autosuspend(udev);
+		bt_dev_info(hdev, "%s memdump size(%u)\n",
+			    (pkt_type == HCI_ACLDATA_PKT) ? "ACL" : "event",
+			    dump_size);
+	} else {
+		sk_ptr += offsetof(struct qca_dump_hdr, data);
+		sk_len -= offsetof(struct qca_dump_hdr, data);
+	}
+
+	if (!btdata->qca_dump.ram_dump_size) {
+		ret = -EINVAL;
+		bt_dev_err(hdev, "memdump is not active");
+		goto out;
+	}
+
+	if ((seqno > btdata->qca_dump.ram_dump_seqno + 1) && (seqno != QCA_LAST_SEQUENCE_NUM)) {
+		dump_size = QCA_MEMDUMP_PKT_SIZE * (seqno - btdata->qca_dump.ram_dump_seqno - 1);
+		hci_devcd_append_pattern(hdev, 0x0, dump_size);
+		bt_dev_err(hdev,
+			   "expected memdump seqno(%u) is not received(%u)\n",
+			   btdata->qca_dump.ram_dump_seqno, seqno);
+		btdata->qca_dump.ram_dump_seqno = seqno;
+		kfree_skb(skb);
+		return ret;
+	}
+
+	skb_pull(skb, skb->len - sk_len);
+	hci_devcd_append(hdev, skb);
+	btdata->qca_dump.ram_dump_seqno++;
+	if (seqno == QCA_LAST_SEQUENCE_NUM) {
+		bt_dev_info(hdev,
+				"memdump done: pkts(%u), total(%u)\n",
+				btdata->qca_dump.ram_dump_seqno, btdata->qca_dump.ram_dump_size);
+
+		hci_devcd_complete(hdev);
+		goto out;
+	}
+	return ret;
+
+out:
+	if (btdata->qca_dump.ram_dump_size)
+		usb_enable_autosuspend(udev);
+	btdata->qca_dump.ram_dump_size = 0;
+	btdata->qca_dump.ram_dump_seqno = 0;
+	clear_bit(BTUSB_HW_SSR_ACTIVE, &btdata->flags);
+
+	if (ret < 0)
+		kfree_skb(skb);
+	return ret;
+}
+
+static int btusb_recv_acl_qca(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	if (handle_dump_pkt_qca(hdev, skb))
+		return 0;
+	return hci_recv_frame(hdev, skb);
+}
+
+static int btusb_recv_evt_qca(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	if (handle_dump_pkt_qca(hdev, skb))
+		return 0;
+	return hci_recv_frame(hdev, skb);
+}
+
+
 #define QCA_DFU_PACKET_LEN	4096
 
 #define QCA_GET_TARGET_VERSION	0x09
@@ -3628,6 +3842,9 @@ static int btusb_setup_qca(struct hci_dev *hdev)
 	if (err < 0)
 		return err;
 
+	btdata->qca_dump.fw_version = le32_to_cpu(ver.patch_version);
+	btdata->qca_dump.controller_id = le32_to_cpu(ver.rom_version);
+
 	if (!(status & QCA_SYSCFG_UPDATED)) {
 		err = btusb_setup_qca_load_nvm(hdev, &ver, info);
 		if (err < 0)
@@ -4117,6 +4334,11 @@ static int btusb_probe(struct usb_interface *intf,
 	}
 
 	if (id->driver_info & BTUSB_QCA_WCN6855) {
+		data->qca_dump.id_vendor = id->idVendor;
+		data->qca_dump.id_product = id->idProduct;
+		data->recv_event = btusb_recv_evt_qca;
+		data->recv_acl = btusb_recv_acl_qca;
+		hci_devcd_register(hdev, btusb_coredump_qca, btusb_dump_hdr_qca, NULL);
 		data->setup_on_usb = btusb_setup_qca;
 		hdev->shutdown = btusb_shutdown_qca;
 		hdev->set_bdaddr = btusb_set_bdaddr_wcn6855;
-- 
cgit v1.2.3-70-g09d2


From d883a4669a1def6d121ccf5e64ad28260d1c9531 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Fri, 21 Apr 2023 11:37:55 -0700
Subject: Bluetooth: hci_sync: Only allow hci_cmd_sync_queue if running

This makes sure hci_cmd_sync_queue only queue new work if HCI_RUNNING
has been set otherwise there is a risk of commands being sent while
turning off.

Because hci_cmd_sync_queue can no longer queue work while HCI_RUNNING is
not set it cannot be used to power on adapters so instead
hci_cmd_sync_submit is introduced which bypass the HCI_RUNNING check, so
it behaves like the old implementation.

Link: https://lore.kernel.org/all/CAB4PzUpDMvdc8j2MdeSAy1KkAE-D3woprCwAdYWeOc-3v3c9Sw@mail.gmail.com/
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sync.h |  2 ++
 net/bluetooth/hci_sync.c         | 25 +++++++++++++++++++++++--
 net/bluetooth/mgmt.c             | 12 ++++++++----
 3 files changed, 33 insertions(+), 6 deletions(-)

diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index f61b249787fc..2495be4d8b82 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -41,6 +41,8 @@ void hci_cmd_sync_clear(struct hci_dev *hdev);
 void hci_cmd_sync_cancel(struct hci_dev *hdev, int err);
 void __hci_cmd_sync_cancel(struct hci_dev *hdev, int err);
 
+int hci_cmd_sync_submit(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+			void *data, hci_cmd_sync_work_destroy_t destroy);
 int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 		       void *data, hci_cmd_sync_work_destroy_t destroy);
 
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 771aaa808967..647a8ce54062 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -684,8 +684,12 @@ void hci_cmd_sync_cancel(struct hci_dev *hdev, int err)
 }
 EXPORT_SYMBOL(hci_cmd_sync_cancel);
 
-int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
-		       void *data, hci_cmd_sync_work_destroy_t destroy)
+/* Submit HCI command to be run in as cmd_sync_work:
+ *
+ * - hdev must _not_ be unregistered
+ */
+int hci_cmd_sync_submit(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+			void *data, hci_cmd_sync_work_destroy_t destroy)
 {
 	struct hci_cmd_sync_work_entry *entry;
 
@@ -708,6 +712,23 @@ int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
 
 	return 0;
 }
+EXPORT_SYMBOL(hci_cmd_sync_submit);
+
+/* Queue HCI command:
+ *
+ * - hdev must be running
+ */
+int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func,
+		       void *data, hci_cmd_sync_work_destroy_t destroy)
+{
+	/* Only queue command if hdev is running which means it had been opened
+	 * and is either on init phase or is already up.
+	 */
+	if (!test_bit(HCI_RUNNING, &hdev->flags))
+		return -ENETDOWN;
+
+	return hci_cmd_sync_submit(hdev, func, data, destroy);
+}
 EXPORT_SYMBOL(hci_cmd_sync_queue);
 
 int hci_update_eir_sync(struct hci_dev *hdev)
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 13c745876b39..f7b2d0971f24 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -1400,11 +1400,15 @@ static int set_powered(struct sock *sk, struct hci_dev *hdev, void *data,
 	}
 
 	/* Cancel potentially blocking sync operation before power off */
-	if (cp->val == 0x00)
+	if (cp->val == 0x00) {
 		__hci_cmd_sync_cancel(hdev, -EHOSTDOWN);
-
-	err = hci_cmd_sync_queue(hdev, set_powered_sync, cmd,
-				 mgmt_set_powered_complete);
+		err = hci_cmd_sync_queue(hdev, set_powered_sync, cmd,
+					 mgmt_set_powered_complete);
+	} else {
+		/* Use hci_cmd_sync_submit since hdev might not be running */
+		err = hci_cmd_sync_submit(hdev, set_powered_sync, cmd,
+					  mgmt_set_powered_complete);
+	}
 
 	if (err < 0)
 		mgmt_pending_remove(cmd);
-- 
cgit v1.2.3-70-g09d2


From 60fd497c99b336e466db0aaeff44c7d12a1fc10a Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@ozlabs.org>
Date: Mon, 24 Apr 2023 09:02:32 +1000
Subject: MAINTAINERS: Remove PPP maintainer

I am not currently maintaining the kernel PPP code, so remove my
address from the MAINTAINERS entry for it.

Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index eefcea6f8757..623a4e76fd09 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -16771,9 +16771,8 @@ F:	include/uapi/linux/if_pppol2tp.h
 F:	net/l2tp/l2tp_ppp.c
 
 PPP PROTOCOL DRIVERS AND COMPRESSORS
-M:	Paul Mackerras <paulus@samba.org>
 L:	linux-ppp@vger.kernel.org
-S:	Maintained
+S:	Orphan
 F:	drivers/net/ppp/ppp_*
 
 PPS SUPPORT
-- 
cgit v1.2.3-70-g09d2


From 7deca5eae83389ca40ac1b1bde96e4af17cca84f Mon Sep 17 00:00:00 2001
From: Dave Marchevsky <davemarchevsky@fb.com>
Date: Mon, 24 Apr 2023 13:43:21 -0700
Subject: bpf: Disable bpf_refcount_acquire kfunc calls until race conditions
 are fixed

As reported by Kumar in [0], the shared ownership implementation for BPF
programs has some race conditions which need to be addressed before it
can safely be used. This patch does so in a minimal way instead of
ripping out shared ownership entirely, as proper fixes for the issues
raised will follow ASAP, at which point this patch's commit can be
reverted to re-enable shared ownership.

The patch removes the ability to call bpf_refcount_acquire_impl from BPF
programs. Programs can only bump refcount and obtain a new owning
reference using this kfunc, so removing the ability to call it
effectively disables shared ownership.

Instead of changing success / failure expectations for
bpf_refcount-related selftests, this patch just disables them from
running for now.

  [0]: https://lore.kernel.org/bpf/d7hyspcow5wtjcmw4fugdgyp3fwhljwuscp3xyut5qnwivyeru@ysdq543otzv2/

Reported-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20230424204321.2680232-1-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c                                    | 5 ++++-
 tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c | 2 --
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 0d73139ee4d8..5c4aa393f65a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -10509,7 +10509,10 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 				verbose(env, "arg#%d doesn't point to a type with bpf_refcount field\n", i);
 				return -EINVAL;
 			}
-
+			if (rec->refcount_off >= 0) {
+				verbose(env, "bpf_refcount_acquire calls are disabled for now\n");
+				return -EINVAL;
+			}
 			meta->arg_refcount_acquire.btf = reg->btf;
 			meta->arg_refcount_acquire.btf_id = reg->btf_id;
 			break;
diff --git a/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c b/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c
index 2ab23832062d..595cbf92bff5 100644
--- a/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c
+++ b/tools/testing/selftests/bpf/prog_tests/refcounted_kptr.c
@@ -9,10 +9,8 @@
 
 void test_refcounted_kptr(void)
 {
-	RUN_TESTS(refcounted_kptr);
 }
 
 void test_refcounted_kptr_fail(void)
 {
-	RUN_TESTS(refcounted_kptr_fail);
 }
-- 
cgit v1.2.3-70-g09d2


From a0c109dcafb15b8bee187c49fb746779374f60f0 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Mon, 24 Apr 2023 16:11:03 +0000
Subject: bpf: Add __rcu_read_{lock,unlock} into btf id deny list

The tracing recursion prevention mechanism must be protected by rcu, that
leaves __rcu_read_{lock,unlock} unprotected by this mechanism. If we trace
them, the recursion will happen. Let's add them into the btf id deny list.

When CONFIG_PREEMPT_RCU is enabled, it can be reproduced with a simple bpf
program as such:
  SEC("fentry/__rcu_read_lock")
  int fentry_run()
  {
      return 0;
  }

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230424161104.3737-2-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 5c4aa393f65a..fbcf5a4e2fcd 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18671,6 +18671,10 @@ BTF_ID(func, rcu_read_unlock_strict)
 BTF_ID(func, preempt_count_add)
 BTF_ID(func, preempt_count_sub)
 #endif
+#ifdef CONFIG_PREEMPT_RCU
+BTF_ID(func, __rcu_read_lock)
+BTF_ID(func, __rcu_read_unlock)
+#endif
 BTF_SET_END(btf_id_deny)
 
 static bool can_be_sleepable(struct bpf_prog *prog)
-- 
cgit v1.2.3-70-g09d2


From be7dbd275dc6b911a5b9a22c4f9cb71b2c7fd847 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 24 Apr 2023 16:51:28 -0700
Subject: selftests/bpf: avoid mark_all_scalars_precise() trigger in one of
 iter tests

iter_pass_iter_ptr_to_subprog subtest is relying on actual array size
being passed as subprog parameter. This combined with recent fixes to
precision tracking in conditional jumps ([0]) is now causing verifier to
backtrack all the way to the point where sum() and fill() subprogs are
called, at which point precision backtrack bails out and forces all the
states to have precise SCALAR registers. This in turn causes each
possible value of i within fill() and sum() subprogs to cause
a different non-equivalent state, preventing iterator code to converge.

For now, change the test to assume fixed size of passed in array. Once
BPF verifier supports precision tracking across subprogram calls, these
changes will be reverted as unnecessary.

  [0] 71b547f56124 ("bpf: Fix incorrect verifier pruning due to missing register precision taints")

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230424235128.1941726-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/iters.c | 26 +++++++++++++++-----------
 1 file changed, 15 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/iters.c b/tools/testing/selftests/bpf/progs/iters.c
index 6b9b3c56f009..be16143ae292 100644
--- a/tools/testing/selftests/bpf/progs/iters.c
+++ b/tools/testing/selftests/bpf/progs/iters.c
@@ -651,25 +651,29 @@ int iter_stack_array_loop(const void *ctx)
 	return sum;
 }
 
-static __noinline void fill(struct bpf_iter_num *it, int *arr, __u32 n, int mul)
+#define ARR_SZ 16
+
+static __noinline void fill(struct bpf_iter_num *it, int *arr, int mul)
 {
-	int *t, i;
+	int *t;
+	__u64 i;
 
 	while ((t = bpf_iter_num_next(it))) {
 		i = *t;
-		if (i >= n)
+		if (i >= ARR_SZ)
 			break;
 		arr[i] =  i * mul;
 	}
 }
 
-static __noinline int sum(struct bpf_iter_num *it, int *arr, __u32 n)
+static __noinline int sum(struct bpf_iter_num *it, int *arr)
 {
-	int *t, i, sum = 0;;
+	int *t, sum = 0;;
+	__u64 i;
 
 	while ((t = bpf_iter_num_next(it))) {
 		i = *t;
-		if (i >= n)
+		if (i >= ARR_SZ)
 			break;
 		sum += arr[i];
 	}
@@ -681,7 +685,7 @@ SEC("raw_tp")
 __success
 int iter_pass_iter_ptr_to_subprog(const void *ctx)
 {
-	int arr1[16], arr2[32];
+	int arr1[ARR_SZ], arr2[ARR_SZ];
 	struct bpf_iter_num it;
 	int n, sum1, sum2;
 
@@ -690,25 +694,25 @@ int iter_pass_iter_ptr_to_subprog(const void *ctx)
 	/* fill arr1 */
 	n = ARRAY_SIZE(arr1);
 	bpf_iter_num_new(&it, 0, n);
-	fill(&it, arr1, n, 2);
+	fill(&it, arr1, 2);
 	bpf_iter_num_destroy(&it);
 
 	/* fill arr2 */
 	n = ARRAY_SIZE(arr2);
 	bpf_iter_num_new(&it, 0, n);
-	fill(&it, arr2, n, 10);
+	fill(&it, arr2, 10);
 	bpf_iter_num_destroy(&it);
 
 	/* sum arr1 */
 	n = ARRAY_SIZE(arr1);
 	bpf_iter_num_new(&it, 0, n);
-	sum1 = sum(&it, arr1, n);
+	sum1 = sum(&it, arr1);
 	bpf_iter_num_destroy(&it);
 
 	/* sum arr2 */
 	n = ARRAY_SIZE(arr2);
 	bpf_iter_num_new(&it, 0, n);
-	sum2 = sum(&it, arr2, n);
+	sum2 = sum(&it, arr2);
 	bpf_iter_num_destroy(&it);
 
 	bpf_printk("sum1=%d, sum2=%d", sum1, sum2);
-- 
cgit v1.2.3-70-g09d2


From d913d32cc2707e9cd24fe6fa6d7d470e9c728980 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Fri, 21 Apr 2023 11:52:55 -0700
Subject: netlink: Use copy_to_user() for optval in netlink_getsockopt().

Brad Spencer provided a detailed report [0] that when calling getsockopt()
for AF_NETLINK, some SOL_NETLINK options set only 1 byte even though such
options require at least sizeof(int) as length.

The options return a flag value that fits into 1 byte, but such behaviour
confuses users who do not initialise the variable before calling
getsockopt() and do not strictly check the returned value as char.

Currently, netlink_getsockopt() uses put_user() to copy data to optlen and
optval, but put_user() casts the data based on the pointer, char *optval.
As a result, only 1 byte is set to optval.

To avoid this behaviour, we need to use copy_to_user() or cast optval for
put_user().

Note that this changes the behaviour on big-endian systems, but we document
that the size of optval is int in the man page.

  $ man 7 netlink
  ...
  Socket options
       To set or get a netlink socket option, call getsockopt(2) to read
       or setsockopt(2) to write the option with the option level argument
       set to SOL_NETLINK.  Unless otherwise noted, optval is a pointer to
       an int.

Fixes: 9a4595bc7e67 ("[NETLINK]: Add set/getsockopt options to support more than 32 groups")
Fixes: be0c22a46cfb ("netlink: add NETLINK_BROADCAST_ERROR socket option")
Fixes: 38938bfe3489 ("netlink: add NETLINK_NO_ENOBUFS socket flag")
Fixes: 0a6a3a23ea6e ("netlink: add NETLINK_CAP_ACK socket option")
Fixes: 2d4bc93368f5 ("netlink: extended ACK reporting")
Fixes: 89d35528d17d ("netlink: Add new socket option to enable strict checking on dumps")
Reported-by: Brad Spencer <bspencer@blackberry.com>
Link: https://lore.kernel.org/netdev/ZD7VkNWFfp22kTDt@datsun.rim.net/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Link: https://lore.kernel.org/r/20230421185255.94606-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/netlink/af_netlink.c | 75 +++++++++++++++---------------------------------
 1 file changed, 23 insertions(+), 52 deletions(-)

diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index f365dfdd672d..9b6eb28e6e94 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1742,7 +1742,8 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname,
 {
 	struct sock *sk = sock->sk;
 	struct netlink_sock *nlk = nlk_sk(sk);
-	int len, val, err;
+	unsigned int flag;
+	int len, val;
 
 	if (level != SOL_NETLINK)
 		return -ENOPROTOOPT;
@@ -1754,39 +1755,17 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname,
 
 	switch (optname) {
 	case NETLINK_PKTINFO:
-		if (len < sizeof(int))
-			return -EINVAL;
-		len = sizeof(int);
-		val = nlk->flags & NETLINK_F_RECV_PKTINFO ? 1 : 0;
-		if (put_user(len, optlen) ||
-		    put_user(val, optval))
-			return -EFAULT;
-		err = 0;
+		flag = NETLINK_F_RECV_PKTINFO;
 		break;
 	case NETLINK_BROADCAST_ERROR:
-		if (len < sizeof(int))
-			return -EINVAL;
-		len = sizeof(int);
-		val = nlk->flags & NETLINK_F_BROADCAST_SEND_ERROR ? 1 : 0;
-		if (put_user(len, optlen) ||
-		    put_user(val, optval))
-			return -EFAULT;
-		err = 0;
+		flag = NETLINK_F_BROADCAST_SEND_ERROR;
 		break;
 	case NETLINK_NO_ENOBUFS:
-		if (len < sizeof(int))
-			return -EINVAL;
-		len = sizeof(int);
-		val = nlk->flags & NETLINK_F_RECV_NO_ENOBUFS ? 1 : 0;
-		if (put_user(len, optlen) ||
-		    put_user(val, optval))
-			return -EFAULT;
-		err = 0;
+		flag = NETLINK_F_RECV_NO_ENOBUFS;
 		break;
 	case NETLINK_LIST_MEMBERSHIPS: {
-		int pos, idx, shift;
+		int pos, idx, shift, err = 0;
 
-		err = 0;
 		netlink_lock_table();
 		for (pos = 0; pos * 8 < nlk->ngroups; pos += sizeof(u32)) {
 			if (len - pos < sizeof(u32))
@@ -1803,40 +1782,32 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname,
 		if (put_user(ALIGN(nlk->ngroups / 8, sizeof(u32)), optlen))
 			err = -EFAULT;
 		netlink_unlock_table();
-		break;
+		return err;
 	}
 	case NETLINK_CAP_ACK:
-		if (len < sizeof(int))
-			return -EINVAL;
-		len = sizeof(int);
-		val = nlk->flags & NETLINK_F_CAP_ACK ? 1 : 0;
-		if (put_user(len, optlen) ||
-		    put_user(val, optval))
-			return -EFAULT;
-		err = 0;
+		flag = NETLINK_F_CAP_ACK;
 		break;
 	case NETLINK_EXT_ACK:
-		if (len < sizeof(int))
-			return -EINVAL;
-		len = sizeof(int);
-		val = nlk->flags & NETLINK_F_EXT_ACK ? 1 : 0;
-		if (put_user(len, optlen) || put_user(val, optval))
-			return -EFAULT;
-		err = 0;
+		flag = NETLINK_F_EXT_ACK;
 		break;
 	case NETLINK_GET_STRICT_CHK:
-		if (len < sizeof(int))
-			return -EINVAL;
-		len = sizeof(int);
-		val = nlk->flags & NETLINK_F_STRICT_CHK ? 1 : 0;
-		if (put_user(len, optlen) || put_user(val, optval))
-			return -EFAULT;
-		err = 0;
+		flag = NETLINK_F_STRICT_CHK;
 		break;
 	default:
-		err = -ENOPROTOOPT;
+		return -ENOPROTOOPT;
 	}
-	return err;
+
+	if (len < sizeof(int))
+		return -EINVAL;
+
+	len = sizeof(int);
+	val = nlk->flags & flag ? 1 : 0;
+
+	if (put_user(len, optlen) ||
+	    copy_to_user(optval, &val, len))
+		return -EFAULT;
+
+	return 0;
 }
 
 static void netlink_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb)
-- 
cgit v1.2.3-70-g09d2


From 0ebab78cbcbfd698d40bd307b68337bdf1b47e9b Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Sat, 22 Apr 2023 20:54:32 +0200
Subject: net: veth: add page_pool for page recycling

Introduce page_pool support in veth driver in order to recycle pages
in veth_convert_skb_to_xdp_buff routine and avoid reallocating the skb
through the page allocator.
The patch has been tested sending tcp traffic to a veth pair where the
remote peer is running a simple xdp program just returning xdp_pass:

veth upstream codebase:
MTU 1500B: ~ 8Gbps
MTU 8000B: ~ 13.9Gbps

veth upstream codebase + pp support:
MTU 1500B: ~ 9.2Gbps
MTU 8000B: ~ 16.2Gbps

Tested-by: Maryam Tahhan <mtahhan@redhat.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/Kconfig |  1 +
 drivers/net/veth.c  | 48 ++++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 45 insertions(+), 4 deletions(-)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index c34bd432da27..368c6f5b327e 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -402,6 +402,7 @@ config TUN_VNET_CROSS_LE
 
 config VETH
 	tristate "Virtual ethernet pair device"
+	select PAGE_POOL
 	help
 	  This device is a local ethernet tunnel. Devices are created in pairs.
 	  When one end receives the packet it appears on its pair and vice
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 4b3c6647edc6..35d2285dec25 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -26,6 +26,7 @@
 #include <linux/ptr_ring.h>
 #include <linux/bpf_trace.h>
 #include <linux/net_tstamp.h>
+#include <net/page_pool.h>
 
 #define DRV_NAME	"veth"
 #define DRV_VERSION	"1.0"
@@ -65,6 +66,7 @@ struct veth_rq {
 	bool			rx_notify_masked;
 	struct ptr_ring		xdp_ring;
 	struct xdp_rxq_info	xdp_rxq;
+	struct page_pool	*page_pool;
 };
 
 struct veth_priv {
@@ -727,17 +729,20 @@ static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq,
 			goto drop;
 
 		/* Allocate skb head */
-		page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
+		page = page_pool_dev_alloc_pages(rq->page_pool);
 		if (!page)
 			goto drop;
 
 		nskb = build_skb(page_address(page), PAGE_SIZE);
 		if (!nskb) {
-			put_page(page);
+			page_pool_put_full_page(rq->page_pool, page, true);
 			goto drop;
 		}
 
 		skb_reserve(nskb, VETH_XDP_HEADROOM);
+		skb_copy_header(nskb, skb);
+		skb_mark_for_recycle(nskb);
+
 		size = min_t(u32, skb->len, max_head_size);
 		if (skb_copy_bits(skb, 0, nskb->data, size)) {
 			consume_skb(nskb);
@@ -745,7 +750,6 @@ static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq,
 		}
 		skb_put(nskb, size);
 
-		skb_copy_header(nskb, skb);
 		head_off = skb_headroom(nskb) - skb_headroom(skb);
 		skb_headers_offset_update(nskb, head_off);
 
@@ -754,7 +758,7 @@ static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq,
 		len = skb->len - off;
 
 		for (i = 0; i < MAX_SKB_FRAGS && off < skb->len; i++) {
-			page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
+			page = page_pool_dev_alloc_pages(rq->page_pool);
 			if (!page) {
 				consume_skb(nskb);
 				goto drop;
@@ -1002,11 +1006,37 @@ static int veth_poll(struct napi_struct *napi, int budget)
 	return done;
 }
 
+static int veth_create_page_pool(struct veth_rq *rq)
+{
+	struct page_pool_params pp_params = {
+		.order = 0,
+		.pool_size = VETH_RING_SIZE,
+		.nid = NUMA_NO_NODE,
+		.dev = &rq->dev->dev,
+	};
+
+	rq->page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(rq->page_pool)) {
+		int err = PTR_ERR(rq->page_pool);
+
+		rq->page_pool = NULL;
+		return err;
+	}
+
+	return 0;
+}
+
 static int __veth_napi_enable_range(struct net_device *dev, int start, int end)
 {
 	struct veth_priv *priv = netdev_priv(dev);
 	int err, i;
 
+	for (i = start; i < end; i++) {
+		err = veth_create_page_pool(&priv->rq[i]);
+		if (err)
+			goto err_page_pool;
+	}
+
 	for (i = start; i < end; i++) {
 		struct veth_rq *rq = &priv->rq[i];
 
@@ -1027,6 +1057,11 @@ static int __veth_napi_enable_range(struct net_device *dev, int start, int end)
 err_xdp_ring:
 	for (i--; i >= start; i--)
 		ptr_ring_cleanup(&priv->rq[i].xdp_ring, veth_ptr_free);
+err_page_pool:
+	for (i = start; i < end; i++) {
+		page_pool_destroy(priv->rq[i].page_pool);
+		priv->rq[i].page_pool = NULL;
+	}
 
 	return err;
 }
@@ -1056,6 +1091,11 @@ static void veth_napi_del_range(struct net_device *dev, int start, int end)
 		rq->rx_notify_masked = false;
 		ptr_ring_cleanup(&rq->xdp_ring, veth_ptr_free);
 	}
+
+	for (i = start; i < end; i++) {
+		page_pool_destroy(priv->rq[i].page_pool);
+		priv->rq[i].page_pool = NULL;
+	}
 }
 
 static void veth_napi_del(struct net_device *dev)
-- 
cgit v1.2.3-70-g09d2


From 4fc418053ec7b839d98533d0d4ee83e11ffbaf53 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Sat, 22 Apr 2023 20:54:33 +0200
Subject: net: veth: add page_pool stats

Introduce page_pool stats support to report info about local page_pool
through ethtool

Tested-by: Maryam Tahhan <mtahhan@redhat.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/Kconfig |  1 +
 drivers/net/veth.c  | 20 +++++++++++++++++---
 2 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 368c6f5b327e..d0a1ed216d15 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -403,6 +403,7 @@ config TUN_VNET_CROSS_LE
 config VETH
 	tristate "Virtual ethernet pair device"
 	select PAGE_POOL
+	select PAGE_POOL_STATS
 	help
 	  This device is a local ethernet tunnel. Devices are created in pairs.
 	  When one end receives the packet it appears on its pair and vice
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 35d2285dec25..dce9f9d63e04 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -157,6 +157,8 @@ static void veth_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
 			for (j = 0; j < VETH_TQ_STATS_LEN; j++)
 				ethtool_sprintf(&p, "tx_queue_%u_%.18s",
 						i, veth_tq_stats_desc[j].desc);
+
+		page_pool_ethtool_stats_get_strings(p);
 		break;
 	}
 }
@@ -167,7 +169,8 @@ static int veth_get_sset_count(struct net_device *dev, int sset)
 	case ETH_SS_STATS:
 		return ARRAY_SIZE(ethtool_stats_keys) +
 		       VETH_RQ_STATS_LEN * dev->real_num_rx_queues +
-		       VETH_TQ_STATS_LEN * dev->real_num_tx_queues;
+		       VETH_TQ_STATS_LEN * dev->real_num_tx_queues +
+		       page_pool_ethtool_stats_get_count();
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -178,7 +181,8 @@ static void veth_get_ethtool_stats(struct net_device *dev,
 {
 	struct veth_priv *rcv_priv, *priv = netdev_priv(dev);
 	struct net_device *peer = rtnl_dereference(priv->peer);
-	int i, j, idx;
+	struct page_pool_stats pp_stats = {};
+	int i, j, idx, pp_idx;
 
 	data[0] = peer ? peer->ifindex : 0;
 	idx = 1;
@@ -197,9 +201,10 @@ static void veth_get_ethtool_stats(struct net_device *dev,
 		} while (u64_stats_fetch_retry(&rq_stats->syncp, start));
 		idx += VETH_RQ_STATS_LEN;
 	}
+	pp_idx = idx;
 
 	if (!peer)
-		return;
+		goto page_pool_stats;
 
 	rcv_priv = netdev_priv(peer);
 	for (i = 0; i < peer->real_num_rx_queues; i++) {
@@ -216,7 +221,16 @@ static void veth_get_ethtool_stats(struct net_device *dev,
 				data[tx_idx + j] += *(u64 *)(base + offset);
 			}
 		} while (u64_stats_fetch_retry(&rq_stats->syncp, start));
+		pp_idx = tx_idx + VETH_TQ_STATS_LEN;
+	}
+
+page_pool_stats:
+	for (i = 0; i < dev->real_num_rx_queues; i++) {
+		if (!priv->rq[i].page_pool)
+			continue;
+		page_pool_get_stats(priv->rq[i].page_pool, &pp_stats);
 	}
+	page_pool_ethtool_stats_get(&data[pp_idx], &pp_stats);
 }
 
 static void veth_get_channels(struct net_device *dev,
-- 
cgit v1.2.3-70-g09d2


From 5c74064f43c291d9add2b436a2d70205b71a7cc7 Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Fri, 21 Apr 2023 10:06:57 -0700
Subject: net: mana: Rename mana_refill_rxoob and remove some empty lines

Rename mana_refill_rxoob for naming consistency.
And remove some empty lines between function call and error
checking.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index cabecbfa1102..db2887e25714 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -563,7 +563,6 @@ static int mana_pre_alloc_rxbufs(struct mana_port_context *mpc, int new_mtu)
 
 		da = dma_map_single(dev, va + mpc->rxbpre_headroom,
 				    mpc->rxbpre_datasize, DMA_FROM_DEVICE);
-
 		if (dma_mapping_error(dev, da)) {
 			put_page(virt_to_head_page(va));
 			goto error;
@@ -1515,7 +1514,6 @@ static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 
 	*da = dma_map_single(dev, va + rxq->headroom, rxq->datasize,
 			     DMA_FROM_DEVICE);
-
 	if (dma_mapping_error(dev, *da)) {
 		put_page(virt_to_head_page(va));
 		return NULL;
@@ -1525,14 +1523,13 @@ static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 }
 
 /* Allocate frag for rx buffer, and save the old buf */
-static void mana_refill_rxoob(struct device *dev, struct mana_rxq *rxq,
-			      struct mana_recv_buf_oob *rxoob, void **old_buf)
+static void mana_refill_rx_oob(struct device *dev, struct mana_rxq *rxq,
+			       struct mana_recv_buf_oob *rxoob, void **old_buf)
 {
 	dma_addr_t da;
 	void *va;
 
 	va = mana_get_rxfrag(rxq, dev, &da, true);
-
 	if (!va)
 		return;
 
@@ -1597,7 +1594,7 @@ static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	rxbuf_oob = &rxq->rx_oobs[curr];
 	WARN_ON_ONCE(rxbuf_oob->wqe_inf.wqe_size_in_bu != 1);
 
-	mana_refill_rxoob(dev, rxq, rxbuf_oob, &old_buf);
+	mana_refill_rx_oob(dev, rxq, rxbuf_oob, &old_buf);
 
 	/* Unsuccessful refill will have old_buf == NULL.
 	 * In this case, mana_rx_skb() will drop the packet.
-- 
cgit v1.2.3-70-g09d2


From df18f2da302f169e1a29098c6ca3b474f1b0269e Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Fri, 21 Apr 2023 10:06:58 -0700
Subject: net: mana: Check if netdev/napi_alloc_frag returns single page

netdev/napi_alloc_frag() may fall back to single page which is smaller
than the requested size.
Add error checking to avoid memory overwritten.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index db2887e25714..06d6292e09b3 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -553,6 +553,14 @@ static int mana_pre_alloc_rxbufs(struct mana_port_context *mpc, int new_mtu)
 			va = netdev_alloc_frag(mpc->rxbpre_alloc_size);
 			if (!va)
 				goto error;
+
+			page = virt_to_head_page(va);
+			/* Check if the frag falls back to single page */
+			if (compound_order(page) <
+			    get_order(mpc->rxbpre_alloc_size)) {
+				put_page(page);
+				goto error;
+			}
 		} else {
 			page = dev_alloc_page();
 			if (!page)
@@ -1504,6 +1512,13 @@ static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 
 		if (!va)
 			return NULL;
+
+		page = virt_to_head_page(va);
+		/* Check if the frag falls back to single page */
+		if (compound_order(page) < get_order(rxq->alloc_size)) {
+			put_page(page);
+			return NULL;
+		}
 	} else {
 		page = dev_alloc_page();
 		if (!page)
-- 
cgit v1.2.3-70-g09d2


From 00d0f31a1ec8eb3e4d692bdf2455daf0fb865bb3 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 20 Apr 2023 16:33:02 -0700
Subject: net: ethtool: coalesce: try to make user settings stick twice

SET_COALESCE may change operation mode and parameters in one call.
Changing operation mode may cause the driver to reset the parameter
values to what is a reasonable default for new operation mode.

Since driver does not know which parameters come from user and which
are echoed back from ->get, driver may ignore the parameters when
switching operation modes.

This used to be inevitable for ioctl() but in netlink we know which
parameters are actually specified by the user.

We could inform which parameters were set by the user but this would
lead to a lot of code duplication in the drivers. Instead try to call
the drivers twice if both mode and params are changed. The set method
already checks if any params need updating so in case the driver did
the right thing the first time around - there will be no second call
to it's ->set method (only an extra call to ->get()).

For mlx5 for example before this patch we'd see:

 # ethtool -C eth0 adaptive-rx on  adaptive-tx on
 # ethtool -C eth0 adaptive-rx off adaptive-tx off \
		   tx-usecs 123 rx-usecs 123
 Adaptive RX: off  TX: off
 rx-usecs: 3
 rx-frames: 32
 tx-usecs: 16
 tx-frames: 32
 [...]

After the change:

 # ethtool -C eth0 adaptive-rx on  adaptive-tx on
 # ethtool -C eth0 adaptive-rx off adaptive-tx off \
		   tx-usecs 123 rx-usecs 123
 Adaptive RX: off  TX: off
 rx-usecs: 123
 rx-frames: 32
 tx-usecs: 123
 tx-frames: 32
 [...]

This only works for netlink, so it's a small discrepancy between
netlink and ioctl(). Since we anticipate most users to move to
netlink I believe it's worth making their lives easier.

Link: https://lore.kernel.org/r/20230420233302.944382-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/ethtool-netlink.rst |  4 +++
 net/ethtool/coalesce.c                       | 54 ++++++++++++++++++++++------
 2 files changed, 47 insertions(+), 11 deletions(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index cd0973d4ba01..2540c70952ff 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -1099,6 +1099,10 @@ such that the corresponding bit in ``ethtool_ops::supported_coalesce_params``
 is not set), regardless of their values. Driver may impose additional
 constraints on coalescing parameters and their values.
 
+Compared to requests issued via the ``ioctl()`` netlink version of this request
+will try harder to make sure that values specified by the user have been applied
+and may call the driver twice.
+
 
 PAUSE_GET
 =========
diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c
index 443e7e642c96..01a59ce211c8 100644
--- a/net/ethtool/coalesce.c
+++ b/net/ethtool/coalesce.c
@@ -254,13 +254,14 @@ ethnl_set_coalesce_validate(struct ethnl_req_info *req_info,
 }
 
 static int
-ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info)
+__ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info,
+		     bool *dual_change)
 {
 	struct kernel_ethtool_coalesce kernel_coalesce = {};
 	struct net_device *dev = req_info->dev;
 	struct ethtool_coalesce coalesce = {};
+	bool mod_mode = false, mod = false;
 	struct nlattr **tb = info->attrs;
-	bool mod = false;
 	int ret;
 
 	ret = dev->ethtool_ops->get_coalesce(dev, &coalesce, &kernel_coalesce,
@@ -268,6 +269,7 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info)
 	if (ret < 0)
 		return ret;
 
+	/* Update values */
 	ethnl_update_u32(&coalesce.rx_coalesce_usecs,
 			 tb[ETHTOOL_A_COALESCE_RX_USECS], &mod);
 	ethnl_update_u32(&coalesce.rx_max_coalesced_frames,
@@ -286,10 +288,6 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info)
 			 tb[ETHTOOL_A_COALESCE_TX_MAX_FRAMES_IRQ], &mod);
 	ethnl_update_u32(&coalesce.stats_block_coalesce_usecs,
 			 tb[ETHTOOL_A_COALESCE_STATS_BLOCK_USECS], &mod);
-	ethnl_update_bool32(&coalesce.use_adaptive_rx_coalesce,
-			    tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_RX], &mod);
-	ethnl_update_bool32(&coalesce.use_adaptive_tx_coalesce,
-			    tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_TX], &mod);
 	ethnl_update_u32(&coalesce.pkt_rate_low,
 			 tb[ETHTOOL_A_COALESCE_PKT_RATE_LOW], &mod);
 	ethnl_update_u32(&coalesce.rx_coalesce_usecs_low,
@@ -312,17 +310,25 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info)
 			 tb[ETHTOOL_A_COALESCE_TX_MAX_FRAMES_HIGH], &mod);
 	ethnl_update_u32(&coalesce.rate_sample_interval,
 			 tb[ETHTOOL_A_COALESCE_RATE_SAMPLE_INTERVAL], &mod);
-	ethnl_update_u8(&kernel_coalesce.use_cqe_mode_tx,
-			tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_TX], &mod);
-	ethnl_update_u8(&kernel_coalesce.use_cqe_mode_rx,
-			tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_RX], &mod);
 	ethnl_update_u32(&kernel_coalesce.tx_aggr_max_bytes,
 			 tb[ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES], &mod);
 	ethnl_update_u32(&kernel_coalesce.tx_aggr_max_frames,
 			 tb[ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES], &mod);
 	ethnl_update_u32(&kernel_coalesce.tx_aggr_time_usecs,
 			 tb[ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS], &mod);
-	if (!mod)
+
+	/* Update operation modes */
+	ethnl_update_bool32(&coalesce.use_adaptive_rx_coalesce,
+			    tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_RX], &mod_mode);
+	ethnl_update_bool32(&coalesce.use_adaptive_tx_coalesce,
+			    tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_TX], &mod_mode);
+	ethnl_update_u8(&kernel_coalesce.use_cqe_mode_tx,
+			tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_TX], &mod_mode);
+	ethnl_update_u8(&kernel_coalesce.use_cqe_mode_rx,
+			tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_RX], &mod_mode);
+
+	*dual_change = mod && mod_mode;
+	if (!mod && !mod_mode)
 		return 0;
 
 	ret = dev->ethtool_ops->set_coalesce(dev, &coalesce, &kernel_coalesce,
@@ -330,6 +336,32 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info)
 	return ret < 0 ? ret : 1;
 }
 
+static int
+ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info)
+{
+	bool dual_change;
+	int err, ret;
+
+	/* SET_COALESCE may change operation mode and parameters in one call.
+	 * Changing operation mode may cause the driver to reset the parameter
+	 * values, and therefore ignore user input (driver does not know which
+	 * parameters come from user and which are echoed back from ->get).
+	 * To not complicate the drivers if user tries to change both the mode
+	 * and parameters at once - call the driver twice.
+	 */
+	err = __ethnl_set_coalesce(req_info, info, &dual_change);
+	if (err < 0)
+		return err;
+	ret = err;
+
+	if (ret && dual_change) {
+		err = __ethnl_set_coalesce(req_info, info, &dual_change);
+		if (err < 0)
+			return err;
+	}
+	return ret;
+}
+
 const struct ethnl_request_ops ethnl_coalesce_request_ops = {
 	.request_cmd		= ETHTOOL_MSG_COALESCE_GET,
 	.reply_cmd		= ETHTOOL_MSG_COALESCE_GET_REPLY,
-- 
cgit v1.2.3-70-g09d2


From 86c2b51f203ec89adc18492a46343472055fd3c8 Mon Sep 17 00:00:00 2001
From: wuych <yunchuan@nfschina.com>
Date: Mon, 24 Apr 2023 18:15:50 +0800
Subject: net: phy: dp83867: Remove unnecessary (void*) conversions

Pointer variables of void * type do not require type cast.

Signed-off-by: wuych <yunchuan@nfschina.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230424101550.664319-1-yunchuan@nfschina.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/dp83867.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c
index 5821f04c69dc..4376fc2aa22c 100644
--- a/drivers/net/phy/dp83867.c
+++ b/drivers/net/phy/dp83867.c
@@ -468,8 +468,7 @@ static int dp83867_set_tunable(struct phy_device *phydev,
 
 static int dp83867_config_port_mirroring(struct phy_device *phydev)
 {
-	struct dp83867_private *dp83867 =
-		(struct dp83867_private *)phydev->priv;
+	struct dp83867_private *dp83867 = phydev->priv;
 
 	if (dp83867->port_mirroring == DP83867_PORT_MIRROING_EN)
 		phy_set_bits_mmd(phydev, DP83867_DEVADDR, DP83867_CFG4,
-- 
cgit v1.2.3-70-g09d2


From e515c330d7e22692d03475740fa45d179281cdf7 Mon Sep 17 00:00:00 2001
From: Jianuo Kuang <u202110722@hust.edu.cn>
Date: Mon, 24 Apr 2023 10:41:40 +0800
Subject: drivers: nfc: nfcsim: remove return value check of `dev_dir`

Smatch complains that:
nfcsim_debugfs_init_dev() warn: 'dev_dir' is an error pointer or valid

According to the documentation of the debugfs_create_dir() function,
there is no need to check the return value of this function.
Just delete the dead code.

Signed-off-by: Jianuo Kuang <u202110722@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20230424024140.34607-1-u202110722@hust.edu.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/nfc/nfcsim.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/nfc/nfcsim.c b/drivers/nfc/nfcsim.c
index 85bf8d586c70..44eeb17ae48d 100644
--- a/drivers/nfc/nfcsim.c
+++ b/drivers/nfc/nfcsim.c
@@ -367,11 +367,6 @@ static void nfcsim_debugfs_init_dev(struct nfcsim *dev)
 	}
 
 	dev_dir = debugfs_create_dir(devname, nfcsim_debugfs_root);
-	if (!dev_dir) {
-		NFCSIM_ERR(dev, "Could not create debugfs entries for nfc%d\n",
-			   idx);
-		return;
-	}
 
 	debugfs_create_u8("dropframe", 0664, dev_dir, &dev->dropframe);
 }
-- 
cgit v1.2.3-70-g09d2


From aed8fdad2152d946add50bec00a6b07c457bdcdf Mon Sep 17 00:00:00 2001
From: Alexander Stein <alexander.stein@ew.tq-group.com>
Date: Mon, 24 Apr 2023 16:16:48 +0200
Subject: net: phy: Fix reading LED reg property

'reg' is always encoded in 32 bits, thus it has to be read using the
function with the corresponding bit width.

Fixes: 01e5b728e9e4 ("net: phy: Add a binding for PHY LEDs")
Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20230424141648.317944-1-alexander.stein@ew.tq-group.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phy_device.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index d373446ab5ac..17d0d0555a79 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -3028,6 +3028,7 @@ static int of_phy_led(struct phy_device *phydev,
 	struct led_init_data init_data = {};
 	struct led_classdev *cdev;
 	struct phy_led *phyled;
+	u32 index;
 	int err;
 
 	phyled = devm_kzalloc(dev, sizeof(*phyled), GFP_KERNEL);
@@ -3037,10 +3038,13 @@ static int of_phy_led(struct phy_device *phydev,
 	cdev = &phyled->led_cdev;
 	phyled->phydev = phydev;
 
-	err = of_property_read_u8(led, "reg", &phyled->index);
+	err = of_property_read_u32(led, "reg", &index);
 	if (err)
 		return err;
+	if (index > U8_MAX)
+		return -EINVAL;
 
+	phyled->index = index;
 	if (phydev->drv->led_brightness_set)
 		cdev->brightness_set_blocking = phy_led_set_brightness;
 	if (phydev->drv->led_blink_set)
-- 
cgit v1.2.3-70-g09d2


From 938f65adc4202da8ed3c85ef5a40dd3b8ee328d9 Mon Sep 17 00:00:00 2001
From: Alexander Stein <alexander.stein@ew.tq-group.com>
Date: Mon, 24 Apr 2023 15:46:25 +0200
Subject: net: phy: dp83867: Add led_brightness_set support

Up to 4 LEDs can be attached to the PHY, add support for setting
brightness manually.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230424134625.303957-1-alexander.stein@ew.tq-group.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/dp83867.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c
index 4376fc2aa22c..d75f526a20a4 100644
--- a/drivers/net/phy/dp83867.c
+++ b/drivers/net/phy/dp83867.c
@@ -26,6 +26,8 @@
 #define MII_DP83867_MICR	0x12
 #define MII_DP83867_ISR		0x13
 #define DP83867_CFG2		0x14
+#define DP83867_LEDCR1		0x18
+#define DP83867_LEDCR2		0x19
 #define DP83867_CFG3		0x1e
 #define DP83867_CTRL		0x1f
 
@@ -150,6 +152,12 @@
 /* FLD_THR_CFG */
 #define DP83867_FLD_THR_CFG_ENERGY_LOST_THR_MASK	0x7
 
+#define DP83867_LED_COUNT	4
+
+/* LED_DRV bits */
+#define DP83867_LED_DRV_EN(x)	BIT((x) * 4)
+#define DP83867_LED_DRV_VAL(x)	BIT((x) * 4 + 1)
+
 enum {
 	DP83867_PORT_MIRROING_KEEP,
 	DP83867_PORT_MIRROING_EN,
@@ -969,6 +977,27 @@ static int dp83867_loopback(struct phy_device *phydev, bool enable)
 			  enable ? BMCR_LOOPBACK : 0);
 }
 
+static int
+dp83867_led_brightness_set(struct phy_device *phydev,
+			   u8 index, enum led_brightness brightness)
+{
+	u32 val;
+
+	if (index >= DP83867_LED_COUNT)
+		return -EINVAL;
+
+	/* DRV_EN==1: output is DRV_VAL */
+	val = DP83867_LED_DRV_EN(index);
+
+	if (brightness)
+		val |= DP83867_LED_DRV_VAL(index);
+
+	return phy_modify(phydev, DP83867_LEDCR2,
+			  DP83867_LED_DRV_VAL(index) |
+			  DP83867_LED_DRV_EN(index),
+			  val);
+}
+
 static struct phy_driver dp83867_driver[] = {
 	{
 		.phy_id		= DP83867_PHY_ID,
@@ -996,6 +1025,8 @@ static struct phy_driver dp83867_driver[] = {
 
 		.link_change_notify = dp83867_link_change_notify,
 		.set_loopback	= dp83867_loopback,
+
+		.led_brightness_set = dp83867_led_brightness_set,
 	},
 };
 module_phy_driver(dp83867_driver);
-- 
cgit v1.2.3-70-g09d2


From 42fb2962b4a6f8b46aa1c141748319e01d0c7c39 Mon Sep 17 00:00:00 2001
From: Gerhard Engleder <gerhard@engleder-embedded.com>
Date: Fri, 21 Apr 2023 21:46:51 +0200
Subject: tsnep: Replace modulo operation with mask

TX/RX ring size is static and power of 2 to enable compiler to optimize
modulo operation to mask operation. Make this optimization already in
the code and don't rely on the compiler.

CPU utilisation during high packet rate has not changed. So no
performance improvement has been measured. But it is best practice to
prevent modulo operations.

Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep.h      |  1 +
 drivers/net/ethernet/engleder/tsnep_main.c | 28 ++++++++++++++--------------
 2 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep.h b/drivers/net/ethernet/engleder/tsnep.h
index 058c2bcf31a7..1de26aec78d3 100644
--- a/drivers/net/ethernet/engleder/tsnep.h
+++ b/drivers/net/ethernet/engleder/tsnep.h
@@ -18,6 +18,7 @@
 #define TSNEP "tsnep"
 
 #define TSNEP_RING_SIZE 256
+#define TSNEP_RING_MASK (TSNEP_RING_SIZE - 1)
 #define TSNEP_RING_RX_REFILL 16
 #define TSNEP_RING_RX_REUSE (TSNEP_RING_SIZE - TSNEP_RING_SIZE / 4)
 #define TSNEP_RING_ENTRIES_PER_PAGE (PAGE_SIZE / TSNEP_DESC_SIZE)
diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index ed1b6102cfeb..3d15e673894a 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -292,7 +292,7 @@ static int tsnep_tx_ring_init(struct tsnep_tx *tx)
 	}
 	for (i = 0; i < TSNEP_RING_SIZE; i++) {
 		entry = &tx->entry[i];
-		next_entry = &tx->entry[(i + 1) % TSNEP_RING_SIZE];
+		next_entry = &tx->entry[(i + 1) & TSNEP_RING_MASK];
 		entry->desc->next = __cpu_to_le64(next_entry->desc_dma);
 	}
 
@@ -381,7 +381,7 @@ static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count)
 	int i;
 
 	for (i = 0; i < count; i++) {
-		entry = &tx->entry[(tx->write + i) % TSNEP_RING_SIZE];
+		entry = &tx->entry[(tx->write + i) & TSNEP_RING_MASK];
 
 		if (!i) {
 			len = skb_headlen(skb);
@@ -419,7 +419,7 @@ static int tsnep_tx_unmap(struct tsnep_tx *tx, int index, int count)
 	int i;
 
 	for (i = 0; i < count; i++) {
-		entry = &tx->entry[(index + i) % TSNEP_RING_SIZE];
+		entry = &tx->entry[(index + i) & TSNEP_RING_MASK];
 
 		if (entry->len) {
 			if (entry->type & TSNEP_TX_TYPE_SKB)
@@ -481,9 +481,9 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
 		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 
 	for (i = 0; i < count; i++)
-		tsnep_tx_activate(tx, (tx->write + i) % TSNEP_RING_SIZE, length,
+		tsnep_tx_activate(tx, (tx->write + i) & TSNEP_RING_MASK, length,
 				  i == count - 1);
-	tx->write = (tx->write + count) % TSNEP_RING_SIZE;
+	tx->write = (tx->write + count) & TSNEP_RING_MASK;
 
 	skb_tx_timestamp(skb);
 
@@ -516,7 +516,7 @@ static int tsnep_xdp_tx_map(struct xdp_frame *xdpf, struct tsnep_tx *tx,
 	frag = NULL;
 	len = xdpf->len;
 	for (i = 0; i < count; i++) {
-		entry = &tx->entry[(tx->write + i) % TSNEP_RING_SIZE];
+		entry = &tx->entry[(tx->write + i) & TSNEP_RING_MASK];
 		if (type & TSNEP_TX_TYPE_XDP_NDO) {
 			data = unlikely(frag) ? skb_frag_address(frag) :
 						xdpf->data;
@@ -589,9 +589,9 @@ static bool tsnep_xdp_xmit_frame_ring(struct xdp_frame *xdpf,
 	length = retval;
 
 	for (i = 0; i < count; i++)
-		tsnep_tx_activate(tx, (tx->write + i) % TSNEP_RING_SIZE, length,
+		tsnep_tx_activate(tx, (tx->write + i) & TSNEP_RING_MASK, length,
 				  i == count - 1);
-	tx->write = (tx->write + count) % TSNEP_RING_SIZE;
+	tx->write = (tx->write + count) & TSNEP_RING_MASK;
 
 	/* descriptor properties shall be valid before hardware is notified */
 	dma_wmb();
@@ -691,7 +691,7 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
 		/* xdpf is union with skb */
 		entry->skb = NULL;
 
-		tx->read = (tx->read + count) % TSNEP_RING_SIZE;
+		tx->read = (tx->read + count) & TSNEP_RING_MASK;
 
 		tx->packets++;
 		tx->bytes += length + ETH_FCS_LEN;
@@ -839,7 +839,7 @@ static int tsnep_rx_ring_init(struct tsnep_rx *rx)
 
 	for (i = 0; i < TSNEP_RING_SIZE; i++) {
 		entry = &rx->entry[i];
-		next_entry = &rx->entry[(i + 1) % TSNEP_RING_SIZE];
+		next_entry = &rx->entry[(i + 1) & TSNEP_RING_MASK];
 		entry->desc->next = __cpu_to_le64(next_entry->desc_dma);
 	}
 
@@ -925,7 +925,7 @@ static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
 	int retval;
 
 	for (i = 0; i < count && !alloc_failed; i++) {
-		index = (rx->write + i) % TSNEP_RING_SIZE;
+		index = (rx->write + i) & TSNEP_RING_MASK;
 
 		retval = tsnep_rx_alloc_buffer(rx, index);
 		if (unlikely(retval)) {
@@ -945,7 +945,7 @@ static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
 	}
 
 	if (enable) {
-		rx->write = (rx->write + i) % TSNEP_RING_SIZE;
+		rx->write = (rx->write + i) & TSNEP_RING_MASK;
 
 		/* descriptor properties shall be valid before hardware is
 		 * notified
@@ -1090,7 +1090,7 @@ static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
 				 * empty RX ring, thus buffer cannot be used for
 				 * RX processing
 				 */
-				rx->read = (rx->read + 1) % TSNEP_RING_SIZE;
+				rx->read = (rx->read + 1) & TSNEP_RING_MASK;
 				desc_available++;
 
 				rx->dropped++;
@@ -1117,7 +1117,7 @@ static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
 		 */
 		length -= TSNEP_RX_INLINE_METADATA_SIZE;
 
-		rx->read = (rx->read + 1) % TSNEP_RING_SIZE;
+		rx->read = (rx->read + 1) & TSNEP_RING_MASK;
 		desc_available++;
 
 		if (prog) {
-- 
cgit v1.2.3-70-g09d2


From 33b0ee02c84c962a9701107be5683f7a600ebffd Mon Sep 17 00:00:00 2001
From: Gerhard Engleder <gerhard@engleder-embedded.com>
Date: Fri, 21 Apr 2023 21:46:52 +0200
Subject: tsnep: Rework TX/RX queue initialization

Make initialization of TX and RX queues less dynamic by moving some
initialization from netdev open/close to device probing.

Additionally, move some initialization code to separate functions to
enable future use in other execution paths.

This is done as preparation for queue reconfigure at runtime, which is
necessary for XSK zero-copy support.

Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep_main.c | 94 ++++++++++++++++--------------
 1 file changed, 51 insertions(+), 43 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 3d15e673894a..095d36e953fc 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -265,7 +265,7 @@ static void tsnep_tx_ring_cleanup(struct tsnep_tx *tx)
 	}
 }
 
-static int tsnep_tx_ring_init(struct tsnep_tx *tx)
+static int tsnep_tx_ring_create(struct tsnep_tx *tx)
 {
 	struct device *dmadev = tx->adapter->dmadev;
 	struct tsnep_tx_entry *entry;
@@ -288,6 +288,7 @@ static int tsnep_tx_ring_init(struct tsnep_tx *tx)
 			entry->desc = (struct tsnep_tx_desc *)
 				(((u8 *)entry->desc_wb) + TSNEP_DESC_OFFSET);
 			entry->desc_dma = tx->page_dma[i] + TSNEP_DESC_SIZE * j;
+			entry->owner_user_flag = false;
 		}
 	}
 	for (i = 0; i < TSNEP_RING_SIZE; i++) {
@@ -303,6 +304,19 @@ alloc_failed:
 	return retval;
 }
 
+static void tsnep_tx_init(struct tsnep_tx *tx)
+{
+	dma_addr_t dma;
+
+	dma = tx->entry[0].desc_dma | TSNEP_RESET_OWNER_COUNTER;
+	iowrite32(DMA_ADDR_LOW(dma), tx->addr + TSNEP_TX_DESC_ADDR_LOW);
+	iowrite32(DMA_ADDR_HIGH(dma), tx->addr + TSNEP_TX_DESC_ADDR_HIGH);
+	tx->write = 0;
+	tx->read = 0;
+	tx->owner_counter = 1;
+	tx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+}
+
 static void tsnep_tx_activate(struct tsnep_tx *tx, int index, int length,
 			      bool last)
 {
@@ -731,26 +745,15 @@ static bool tsnep_tx_pending(struct tsnep_tx *tx)
 	return pending;
 }
 
-static int tsnep_tx_open(struct tsnep_adapter *adapter, void __iomem *addr,
-			 int queue_index, struct tsnep_tx *tx)
+static int tsnep_tx_open(struct tsnep_tx *tx)
 {
-	dma_addr_t dma;
 	int retval;
 
-	memset(tx, 0, sizeof(*tx));
-	tx->adapter = adapter;
-	tx->addr = addr;
-	tx->queue_index = queue_index;
-
-	retval = tsnep_tx_ring_init(tx);
+	retval = tsnep_tx_ring_create(tx);
 	if (retval)
 		return retval;
 
-	dma = tx->entry[0].desc_dma | TSNEP_RESET_OWNER_COUNTER;
-	iowrite32(DMA_ADDR_LOW(dma), tx->addr + TSNEP_TX_DESC_ADDR_LOW);
-	iowrite32(DMA_ADDR_HIGH(dma), tx->addr + TSNEP_TX_DESC_ADDR_HIGH);
-	tx->owner_counter = 1;
-	tx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+	tsnep_tx_init(tx);
 
 	return 0;
 }
@@ -795,7 +798,7 @@ static void tsnep_rx_ring_cleanup(struct tsnep_rx *rx)
 	}
 }
 
-static int tsnep_rx_ring_init(struct tsnep_rx *rx)
+static int tsnep_rx_ring_create(struct tsnep_rx *rx)
 {
 	struct device *dmadev = rx->adapter->dmadev;
 	struct tsnep_rx_entry *entry;
@@ -850,6 +853,19 @@ failed:
 	return retval;
 }
 
+static void tsnep_rx_init(struct tsnep_rx *rx)
+{
+	dma_addr_t dma;
+
+	dma = rx->entry[0].desc_dma | TSNEP_RESET_OWNER_COUNTER;
+	iowrite32(DMA_ADDR_LOW(dma), rx->addr + TSNEP_RX_DESC_ADDR_LOW);
+	iowrite32(DMA_ADDR_HIGH(dma), rx->addr + TSNEP_RX_DESC_ADDR_HIGH);
+	rx->write = 0;
+	rx->read = 0;
+	rx->owner_counter = 1;
+	rx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+}
+
 static int tsnep_rx_desc_available(struct tsnep_rx *rx)
 {
 	if (rx->read <= rx->write)
@@ -1181,26 +1197,15 @@ static bool tsnep_rx_pending(struct tsnep_rx *rx)
 	return false;
 }
 
-static int tsnep_rx_open(struct tsnep_adapter *adapter, void __iomem *addr,
-			 int queue_index, struct tsnep_rx *rx)
+static int tsnep_rx_open(struct tsnep_rx *rx)
 {
-	dma_addr_t dma;
 	int retval;
 
-	memset(rx, 0, sizeof(*rx));
-	rx->adapter = adapter;
-	rx->addr = addr;
-	rx->queue_index = queue_index;
-
-	retval = tsnep_rx_ring_init(rx);
+	retval = tsnep_rx_ring_create(rx);
 	if (retval)
 		return retval;
 
-	dma = rx->entry[0].desc_dma | TSNEP_RESET_OWNER_COUNTER;
-	iowrite32(DMA_ADDR_LOW(dma), rx->addr + TSNEP_RX_DESC_ADDR_LOW);
-	iowrite32(DMA_ADDR_HIGH(dma), rx->addr + TSNEP_RX_DESC_ADDR_HIGH);
-	rx->owner_counter = 1;
-	rx->increment_owner_counter = TSNEP_RING_SIZE - 1;
+	tsnep_rx_init(rx);
 
 	tsnep_rx_refill(rx, tsnep_rx_desc_available(rx), false);
 
@@ -1335,8 +1340,6 @@ static int tsnep_queue_open(struct tsnep_adapter *adapter,
 	struct tsnep_tx *tx = queue->tx;
 	int retval;
 
-	queue->adapter = adapter;
-
 	netif_napi_add(adapter->netdev, &queue->napi, tsnep_poll);
 
 	if (rx) {
@@ -1377,27 +1380,18 @@ failed:
 static int tsnep_netdev_open(struct net_device *netdev)
 {
 	struct tsnep_adapter *adapter = netdev_priv(netdev);
-	int tx_queue_index = 0;
-	int rx_queue_index = 0;
-	void __iomem *addr;
 	int i, retval;
 
 	for (i = 0; i < adapter->num_queues; i++) {
 		if (adapter->queue[i].tx) {
-			addr = adapter->addr + TSNEP_QUEUE(tx_queue_index);
-			retval = tsnep_tx_open(adapter, addr, tx_queue_index,
-					       adapter->queue[i].tx);
+			retval = tsnep_tx_open(adapter->queue[i].tx);
 			if (retval)
 				goto failed;
-			tx_queue_index++;
 		}
 		if (adapter->queue[i].rx) {
-			addr = adapter->addr + TSNEP_QUEUE(rx_queue_index);
-			retval = tsnep_rx_open(adapter, addr, rx_queue_index,
-					       adapter->queue[i].rx);
+			retval = tsnep_rx_open(adapter->queue[i].rx);
 			if (retval)
 				goto failed;
-			rx_queue_index++;
 		}
 
 		retval = tsnep_queue_open(adapter, &adapter->queue[i], i == 0);
@@ -1796,9 +1790,16 @@ static int tsnep_queue_init(struct tsnep_adapter *adapter, int queue_count)
 	adapter->num_tx_queues = 1;
 	adapter->num_rx_queues = 1;
 	adapter->num_queues = 1;
+	adapter->queue[0].adapter = adapter;
 	adapter->queue[0].irq = retval;
 	adapter->queue[0].tx = &adapter->tx[0];
+	adapter->queue[0].tx->adapter = adapter;
+	adapter->queue[0].tx->addr = adapter->addr + TSNEP_QUEUE(0);
+	adapter->queue[0].tx->queue_index = 0;
 	adapter->queue[0].rx = &adapter->rx[0];
+	adapter->queue[0].rx->adapter = adapter;
+	adapter->queue[0].rx->addr = adapter->addr + TSNEP_QUEUE(0);
+	adapter->queue[0].rx->queue_index = 0;
 	adapter->queue[0].irq_mask = irq_mask;
 	adapter->queue[0].irq_delay_addr = adapter->addr + ECM_INT_DELAY;
 	retval = tsnep_set_irq_coalesce(&adapter->queue[0],
@@ -1820,9 +1821,16 @@ static int tsnep_queue_init(struct tsnep_adapter *adapter, int queue_count)
 		adapter->num_tx_queues++;
 		adapter->num_rx_queues++;
 		adapter->num_queues++;
+		adapter->queue[i].adapter = adapter;
 		adapter->queue[i].irq = retval;
 		adapter->queue[i].tx = &adapter->tx[i];
+		adapter->queue[i].tx->adapter = adapter;
+		adapter->queue[i].tx->addr = adapter->addr + TSNEP_QUEUE(i);
+		adapter->queue[i].tx->queue_index = i;
 		adapter->queue[i].rx = &adapter->rx[i];
+		adapter->queue[i].rx->adapter = adapter;
+		adapter->queue[i].rx->addr = adapter->addr + TSNEP_QUEUE(i);
+		adapter->queue[i].rx->queue_index = i;
 		adapter->queue[i].irq_mask =
 			irq_mask << (ECM_INT_TXRX_SHIFT * i);
 		adapter->queue[i].irq_delay_addr =
-- 
cgit v1.2.3-70-g09d2


From 2ea0a282ba09d29915f1d59b22c38207b5c7ed62 Mon Sep 17 00:00:00 2001
From: Gerhard Engleder <gerhard@engleder-embedded.com>
Date: Fri, 21 Apr 2023 21:46:53 +0200
Subject: tsnep: Add functions for queue enable/disable

Move queue enable and disable code to separate functions. This way the
activation and deactivation of the queues are defined actions, which can
be used in future execution paths.

This functions will be used for the queue reconfiguration at runtime,
which is necessary for XSK zero-copy support.

Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep_main.c | 97 ++++++++++++++++++++----------
 1 file changed, 64 insertions(+), 33 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 095d36e953fc..039629af6a43 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -866,6 +866,24 @@ static void tsnep_rx_init(struct tsnep_rx *rx)
 	rx->increment_owner_counter = TSNEP_RING_SIZE - 1;
 }
 
+static void tsnep_rx_enable(struct tsnep_rx *rx)
+{
+	/* descriptor properties shall be valid before hardware is notified */
+	dma_wmb();
+
+	iowrite32(TSNEP_CONTROL_RX_ENABLE, rx->addr + TSNEP_CONTROL);
+}
+
+static void tsnep_rx_disable(struct tsnep_rx *rx)
+{
+	u32 val;
+
+	iowrite32(TSNEP_CONTROL_RX_DISABLE, rx->addr + TSNEP_CONTROL);
+	readx_poll_timeout(ioread32, rx->addr + TSNEP_CONTROL, val,
+			   ((val & TSNEP_CONTROL_RX_ENABLE) == 0), 10000,
+			   1000000);
+}
+
 static int tsnep_rx_desc_available(struct tsnep_rx *rx)
 {
 	if (rx->read <= rx->write)
@@ -932,19 +950,15 @@ static void tsnep_rx_activate(struct tsnep_rx *rx, int index)
 	entry->desc->properties = __cpu_to_le32(entry->properties);
 }
 
-static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
+static int tsnep_rx_alloc(struct tsnep_rx *rx, int count, bool reuse)
 {
-	int index;
 	bool alloc_failed = false;
-	bool enable = false;
-	int i;
-	int retval;
+	int i, index;
 
 	for (i = 0; i < count && !alloc_failed; i++) {
 		index = (rx->write + i) & TSNEP_RING_MASK;
 
-		retval = tsnep_rx_alloc_buffer(rx, index);
-		if (unlikely(retval)) {
+		if (unlikely(tsnep_rx_alloc_buffer(rx, index))) {
 			rx->alloc_failed++;
 			alloc_failed = true;
 
@@ -956,22 +970,23 @@ static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
 		}
 
 		tsnep_rx_activate(rx, index);
-
-		enable = true;
 	}
 
-	if (enable) {
+	if (i)
 		rx->write = (rx->write + i) & TSNEP_RING_MASK;
 
-		/* descriptor properties shall be valid before hardware is
-		 * notified
-		 */
-		dma_wmb();
+	return i;
+}
 
-		iowrite32(TSNEP_CONTROL_RX_ENABLE, rx->addr + TSNEP_CONTROL);
-	}
+static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
+{
+	int desc_refilled;
 
-	return i;
+	desc_refilled = tsnep_rx_alloc(rx, count, reuse);
+	if (desc_refilled)
+		tsnep_rx_enable(rx);
+
+	return desc_refilled;
 }
 
 static bool tsnep_xdp_run_prog(struct tsnep_rx *rx, struct bpf_prog *prog,
@@ -1199,6 +1214,7 @@ static bool tsnep_rx_pending(struct tsnep_rx *rx)
 
 static int tsnep_rx_open(struct tsnep_rx *rx)
 {
+	int desc_available;
 	int retval;
 
 	retval = tsnep_rx_ring_create(rx);
@@ -1207,20 +1223,19 @@ static int tsnep_rx_open(struct tsnep_rx *rx)
 
 	tsnep_rx_init(rx);
 
-	tsnep_rx_refill(rx, tsnep_rx_desc_available(rx), false);
+	desc_available = tsnep_rx_desc_available(rx);
+	retval = tsnep_rx_alloc(rx, desc_available, false);
+	if (retval != desc_available) {
+		tsnep_rx_ring_cleanup(rx);
+
+		return -ENOMEM;
+	}
 
 	return 0;
 }
 
 static void tsnep_rx_close(struct tsnep_rx *rx)
 {
-	u32 val;
-
-	iowrite32(TSNEP_CONTROL_RX_DISABLE, rx->addr + TSNEP_CONTROL);
-	readx_poll_timeout(ioread32, rx->addr + TSNEP_CONTROL, val,
-			   ((val & TSNEP_CONTROL_RX_ENABLE) == 0), 10000,
-			   1000000);
-
 	tsnep_rx_ring_cleanup(rx);
 }
 
@@ -1377,6 +1392,27 @@ failed:
 	return retval;
 }
 
+static void tsnep_queue_enable(struct tsnep_queue *queue)
+{
+	napi_enable(&queue->napi);
+	tsnep_enable_irq(queue->adapter, queue->irq_mask);
+
+	if (queue->rx)
+		tsnep_rx_enable(queue->rx);
+}
+
+static void tsnep_queue_disable(struct tsnep_queue *queue)
+{
+	napi_disable(&queue->napi);
+	tsnep_disable_irq(queue->adapter, queue->irq_mask);
+
+	/* disable RX after NAPI polling has been disabled, because RX can be
+	 * enabled during NAPI polling
+	 */
+	if (queue->rx)
+		tsnep_rx_disable(queue->rx);
+}
+
 static int tsnep_netdev_open(struct net_device *netdev)
 {
 	struct tsnep_adapter *adapter = netdev_priv(netdev);
@@ -1413,11 +1449,8 @@ static int tsnep_netdev_open(struct net_device *netdev)
 	if (retval)
 		goto phy_failed;
 
-	for (i = 0; i < adapter->num_queues; i++) {
-		napi_enable(&adapter->queue[i].napi);
-
-		tsnep_enable_irq(adapter, adapter->queue[i].irq_mask);
-	}
+	for (i = 0; i < adapter->num_queues; i++)
+		tsnep_queue_enable(&adapter->queue[i]);
 
 	return 0;
 
@@ -1444,9 +1477,7 @@ static int tsnep_netdev_close(struct net_device *netdev)
 	tsnep_phy_close(adapter);
 
 	for (i = 0; i < adapter->num_queues; i++) {
-		tsnep_disable_irq(adapter, adapter->queue[i].irq_mask);
-
-		napi_disable(&adapter->queue[i].napi);
+		tsnep_queue_disable(&adapter->queue[i]);
 
 		tsnep_queue_close(&adapter->queue[i], i == 0);
 
-- 
cgit v1.2.3-70-g09d2


From c2d64697f41b0c188690f4f5841045ed86243134 Mon Sep 17 00:00:00 2001
From: Gerhard Engleder <gerhard@engleder-embedded.com>
Date: Fri, 21 Apr 2023 21:46:54 +0200
Subject: tsnep: Move skb receive action to separate function

The function tsnep_rx_poll() is already pretty long and the skb receive
action can be reused for XSK zero-copy support. Move page based skb
receive to separate function.

Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep_main.c | 39 ++++++++++++++++++------------
 1 file changed, 23 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 039629af6a43..2db94b96a1f0 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -1076,6 +1076,28 @@ static struct sk_buff *tsnep_build_skb(struct tsnep_rx *rx, struct page *page,
 	return skb;
 }
 
+static void tsnep_rx_page(struct tsnep_rx *rx, struct napi_struct *napi,
+			  struct page *page, int length)
+{
+	struct sk_buff *skb;
+
+	skb = tsnep_build_skb(rx, page, length);
+	if (skb) {
+		page_pool_release_page(rx->page_pool, page);
+
+		rx->packets++;
+		rx->bytes += length;
+		if (skb->pkt_type == PACKET_MULTICAST)
+			rx->multicast++;
+
+		napi_gro_receive(napi, skb);
+	} else {
+		page_pool_recycle_direct(rx->page_pool, page);
+
+		rx->dropped++;
+	}
+}
+
 static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
 			 int budget)
 {
@@ -1085,7 +1107,6 @@ static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
 	struct netdev_queue *tx_nq;
 	struct bpf_prog *prog;
 	struct xdp_buff xdp;
-	struct sk_buff *skb;
 	struct tsnep_tx *tx;
 	int desc_available;
 	int xdp_status = 0;
@@ -1170,21 +1191,7 @@ static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
 			}
 		}
 
-		skb = tsnep_build_skb(rx, entry->page, length);
-		if (skb) {
-			page_pool_release_page(rx->page_pool, entry->page);
-
-			rx->packets++;
-			rx->bytes += length;
-			if (skb->pkt_type == PACKET_MULTICAST)
-				rx->multicast++;
-
-			napi_gro_receive(napi, skb);
-		} else {
-			page_pool_recycle_direct(rx->page_pool, entry->page);
-
-			rx->dropped++;
-		}
+		tsnep_rx_page(rx, napi, entry->page, length);
 		entry->page = NULL;
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 3fc2333933fdf1148b694d15db824e10449ecbc1 Mon Sep 17 00:00:00 2001
From: Gerhard Engleder <gerhard@engleder-embedded.com>
Date: Fri, 21 Apr 2023 21:46:55 +0200
Subject: tsnep: Add XDP socket zero-copy RX support

Add support for XSK zero-copy to RX path. The setup of the XSK pool can
be done at runtime. If the netdev is running, then the queue must be
disabled and enabled during reconfiguration. This can be done easily
with functions introduced in previous commits.

A more important property is that, if the netdev is running, then the
setup of the XSK pool shall not stop the netdev in case of errors. A
broken netdev after a failed XSK pool setup is bad behavior. Therefore,
the allocation and setup of resources during XSK pool setup is done only
before any queue is disabled. Additionally, freeing and later allocation
of resources is eliminated in some cases. Page pool entries are kept for
later use. Two memory models are registered in parallel. As a result,
the XSK pool setup cannot fail during queue reconfiguration.

In contrast to other drivers, XSK pool setup and XDP BPF program setup
are separate actions. XSK pool setup can be done without any XDP BPF
program. The XDP BPF program can be added, removed or changed without
any reconfiguration of the XSK pool.

Test results with A53 1.2GHz:

xdpsock rxdrop copy mode, 64 byte frames:
                   pps            pkts           1.00
rx                 856,054        10,625,775
Two CPUs with both 100% utilization.

xdpsock rxdrop zero-copy mode, 64 byte frames:
                   pps            pkts           1.00
rx                 889,388        4,615,284
Two CPUs with 100% and 20% utilization.

Packet rate increases and CPU utilization is reduced.

100% CPU load seems to the base load. This load is consumed by ksoftirqd
just for dropping the generated packets without xdpsock running.

Using batch API reduced CPU utilization slightly, but measurements are
not stable enough to provide meaningful numbers.

Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep.h      |  13 +-
 drivers/net/ethernet/engleder/tsnep_main.c | 491 ++++++++++++++++++++++++++++-
 drivers/net/ethernet/engleder/tsnep_xdp.c  |  66 ++++
 3 files changed, 556 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep.h b/drivers/net/ethernet/engleder/tsnep.h
index 1de26aec78d3..d0bea605a1d1 100644
--- a/drivers/net/ethernet/engleder/tsnep.h
+++ b/drivers/net/ethernet/engleder/tsnep.h
@@ -101,7 +101,10 @@ struct tsnep_rx_entry {
 
 	u32 properties;
 
-	struct page *page;
+	union {
+		struct page *page;
+		struct xdp_buff *xdp;
+	};
 	size_t len;
 	dma_addr_t dma;
 };
@@ -121,6 +124,9 @@ struct tsnep_rx {
 	u32 owner_counter;
 	int increment_owner_counter;
 	struct page_pool *page_pool;
+	struct page **page_buffer;
+	struct xsk_buff_pool *xsk_pool;
+	struct xdp_buff **xdp_batch;
 
 	u32 packets;
 	u32 bytes;
@@ -129,6 +135,7 @@ struct tsnep_rx {
 	u32 alloc_failed;
 
 	struct xdp_rxq_info xdp_rxq;
+	struct xdp_rxq_info xdp_rxq_zc;
 };
 
 struct tsnep_queue {
@@ -214,6 +221,8 @@ int tsnep_rxnfc_del_rule(struct tsnep_adapter *adapter,
 
 int tsnep_xdp_setup_prog(struct tsnep_adapter *adapter, struct bpf_prog *prog,
 			 struct netlink_ext_ack *extack);
+int tsnep_xdp_setup_pool(struct tsnep_adapter *adapter,
+			 struct xsk_buff_pool *pool, u16 queue_id);
 
 #if IS_ENABLED(CONFIG_TSNEP_SELFTESTS)
 int tsnep_ethtool_get_test_count(void);
@@ -242,5 +251,7 @@ static inline void tsnep_ethtool_self_test(struct net_device *dev,
 void tsnep_get_system_time(struct tsnep_adapter *adapter, u64 *time);
 int tsnep_set_irq_coalesce(struct tsnep_queue *queue, u32 usecs);
 u32 tsnep_get_irq_coalesce(struct tsnep_queue *queue);
+int tsnep_enable_xsk(struct tsnep_queue *queue, struct xsk_buff_pool *pool);
+void tsnep_disable_xsk(struct tsnep_queue *queue);
 
 #endif /* _TSNEP_H */
diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 2db94b96a1f0..f300e2963667 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -28,11 +28,16 @@
 #include <linux/iopoll.h>
 #include <linux/bpf.h>
 #include <linux/bpf_trace.h>
+#include <net/xdp_sock_drv.h>
 
 #define TSNEP_RX_OFFSET (max(NET_SKB_PAD, XDP_PACKET_HEADROOM) + NET_IP_ALIGN)
 #define TSNEP_HEADROOM ALIGN(TSNEP_RX_OFFSET, 4)
 #define TSNEP_MAX_RX_BUF_SIZE (PAGE_SIZE - TSNEP_HEADROOM - \
 			       SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+/* XSK buffer shall store at least Q-in-Q frame */
+#define TSNEP_XSK_RX_BUF_SIZE (ALIGN(TSNEP_RX_INLINE_METADATA_SIZE + \
+				     ETH_FRAME_LEN + ETH_FCS_LEN + \
+				     VLAN_HLEN * 2, 4))
 
 #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
 #define DMA_ADDR_HIGH(dma_addr) ((u32)(((dma_addr) >> 32) & 0xFFFFFFFF))
@@ -777,9 +782,12 @@ static void tsnep_rx_ring_cleanup(struct tsnep_rx *rx)
 
 	for (i = 0; i < TSNEP_RING_SIZE; i++) {
 		entry = &rx->entry[i];
-		if (entry->page)
+		if (!rx->xsk_pool && entry->page)
 			page_pool_put_full_page(rx->page_pool, entry->page,
 						false);
+		if (rx->xsk_pool && entry->xdp)
+			xsk_buff_free(entry->xdp);
+		/* xdp is union with page */
 		entry->page = NULL;
 	}
 
@@ -892,6 +900,40 @@ static int tsnep_rx_desc_available(struct tsnep_rx *rx)
 		return rx->read - rx->write - 1;
 }
 
+static void tsnep_rx_free_page_buffer(struct tsnep_rx *rx)
+{
+	struct page **page;
+
+	/* last entry of page_buffer is always zero, because ring cannot be
+	 * filled completely
+	 */
+	page = rx->page_buffer;
+	while (*page) {
+		page_pool_put_full_page(rx->page_pool, *page, false);
+		*page = NULL;
+		page++;
+	}
+}
+
+static int tsnep_rx_alloc_page_buffer(struct tsnep_rx *rx)
+{
+	int i;
+
+	/* alloc for all ring entries except the last one, because ring cannot
+	 * be filled completely
+	 */
+	for (i = 0; i < TSNEP_RING_SIZE - 1; i++) {
+		rx->page_buffer[i] = page_pool_dev_alloc_pages(rx->page_pool);
+		if (!rx->page_buffer[i]) {
+			tsnep_rx_free_page_buffer(rx);
+
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
 static void tsnep_rx_set_page(struct tsnep_rx *rx, struct tsnep_rx_entry *entry,
 			      struct page *page)
 {
@@ -927,7 +969,7 @@ static void tsnep_rx_activate(struct tsnep_rx *rx, int index)
 {
 	struct tsnep_rx_entry *entry = &rx->entry[index];
 
-	/* TSNEP_MAX_RX_BUF_SIZE is a multiple of 4 */
+	/* TSNEP_MAX_RX_BUF_SIZE and TSNEP_XSK_RX_BUF_SIZE are multiple of 4 */
 	entry->properties = entry->len & TSNEP_DESC_LENGTH_MASK;
 	entry->properties |= TSNEP_DESC_INTERRUPT_FLAG;
 	if (index == rx->increment_owner_counter) {
@@ -989,6 +1031,76 @@ static int tsnep_rx_refill(struct tsnep_rx *rx, int count, bool reuse)
 	return desc_refilled;
 }
 
+static void tsnep_rx_set_xdp(struct tsnep_rx *rx, struct tsnep_rx_entry *entry,
+			     struct xdp_buff *xdp)
+{
+	entry->xdp = xdp;
+	entry->len = TSNEP_XSK_RX_BUF_SIZE;
+	entry->dma = xsk_buff_xdp_get_dma(entry->xdp);
+	entry->desc->rx = __cpu_to_le64(entry->dma);
+}
+
+static void tsnep_rx_reuse_buffer_zc(struct tsnep_rx *rx, int index)
+{
+	struct tsnep_rx_entry *entry = &rx->entry[index];
+	struct tsnep_rx_entry *read = &rx->entry[rx->read];
+
+	tsnep_rx_set_xdp(rx, entry, read->xdp);
+	read->xdp = NULL;
+}
+
+static int tsnep_rx_alloc_zc(struct tsnep_rx *rx, int count, bool reuse)
+{
+	u32 allocated;
+	int i;
+
+	allocated = xsk_buff_alloc_batch(rx->xsk_pool, rx->xdp_batch, count);
+	for (i = 0; i < allocated; i++) {
+		int index = (rx->write + i) & TSNEP_RING_MASK;
+		struct tsnep_rx_entry *entry = &rx->entry[index];
+
+		tsnep_rx_set_xdp(rx, entry, rx->xdp_batch[i]);
+		tsnep_rx_activate(rx, index);
+	}
+	if (i == 0) {
+		rx->alloc_failed++;
+
+		if (reuse) {
+			tsnep_rx_reuse_buffer_zc(rx, rx->write);
+			tsnep_rx_activate(rx, rx->write);
+		}
+	}
+
+	if (i)
+		rx->write = (rx->write + i) & TSNEP_RING_MASK;
+
+	return i;
+}
+
+static void tsnep_rx_free_zc(struct tsnep_rx *rx)
+{
+	int i;
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		struct tsnep_rx_entry *entry = &rx->entry[i];
+
+		if (entry->xdp)
+			xsk_buff_free(entry->xdp);
+		entry->xdp = NULL;
+	}
+}
+
+static int tsnep_rx_refill_zc(struct tsnep_rx *rx, int count, bool reuse)
+{
+	int desc_refilled;
+
+	desc_refilled = tsnep_rx_alloc_zc(rx, count, reuse);
+	if (desc_refilled)
+		tsnep_rx_enable(rx);
+
+	return desc_refilled;
+}
+
 static bool tsnep_xdp_run_prog(struct tsnep_rx *rx, struct bpf_prog *prog,
 			       struct xdp_buff *xdp, int *status,
 			       struct netdev_queue *tx_nq, struct tsnep_tx *tx)
@@ -1000,11 +1112,6 @@ static bool tsnep_xdp_run_prog(struct tsnep_rx *rx, struct bpf_prog *prog,
 	length = xdp->data_end - xdp->data_hard_start - XDP_PACKET_HEADROOM;
 
 	act = bpf_prog_run_xdp(prog, xdp);
-
-	/* Due xdp_adjust_tail: DMA sync for_device cover max len CPU touch */
-	sync = xdp->data_end - xdp->data_hard_start - XDP_PACKET_HEADROOM;
-	sync = max(sync, length);
-
 	switch (act) {
 	case XDP_PASS:
 		return false;
@@ -1026,12 +1133,56 @@ out_failure:
 		trace_xdp_exception(rx->adapter->netdev, prog, act);
 		fallthrough;
 	case XDP_DROP:
+		/* Due xdp_adjust_tail: DMA sync for_device cover max len CPU
+		 * touch
+		 */
+		sync = xdp->data_end - xdp->data_hard_start -
+		       XDP_PACKET_HEADROOM;
+		sync = max(sync, length);
 		page_pool_put_page(rx->page_pool, virt_to_head_page(xdp->data),
 				   sync, true);
 		return true;
 	}
 }
 
+static bool tsnep_xdp_run_prog_zc(struct tsnep_rx *rx, struct bpf_prog *prog,
+				  struct xdp_buff *xdp, int *status,
+				  struct netdev_queue *tx_nq,
+				  struct tsnep_tx *tx)
+{
+	u32 act;
+
+	act = bpf_prog_run_xdp(prog, xdp);
+
+	/* XDP_REDIRECT is the main action for zero-copy */
+	if (likely(act == XDP_REDIRECT)) {
+		if (xdp_do_redirect(rx->adapter->netdev, xdp, prog) < 0)
+			goto out_failure;
+		*status |= TSNEP_XDP_REDIRECT;
+		return true;
+	}
+
+	switch (act) {
+	case XDP_PASS:
+		return false;
+	case XDP_TX:
+		if (!tsnep_xdp_xmit_back(rx->adapter, xdp, tx_nq, tx))
+			goto out_failure;
+		*status |= TSNEP_XDP_TX;
+		return true;
+	default:
+		bpf_warn_invalid_xdp_action(rx->adapter->netdev, prog, act);
+		fallthrough;
+	case XDP_ABORTED:
+out_failure:
+		trace_xdp_exception(rx->adapter->netdev, prog, act);
+		fallthrough;
+	case XDP_DROP:
+		xsk_buff_free(xdp);
+		return true;
+	}
+}
+
 static void tsnep_finalize_xdp(struct tsnep_adapter *adapter, int status,
 			       struct netdev_queue *tx_nq, struct tsnep_tx *tx)
 {
@@ -1204,6 +1355,124 @@ static int tsnep_rx_poll(struct tsnep_rx *rx, struct napi_struct *napi,
 	return done;
 }
 
+static int tsnep_rx_poll_zc(struct tsnep_rx *rx, struct napi_struct *napi,
+			    int budget)
+{
+	struct tsnep_rx_entry *entry;
+	struct netdev_queue *tx_nq;
+	struct bpf_prog *prog;
+	struct tsnep_tx *tx;
+	int desc_available;
+	int xdp_status = 0;
+	struct page *page;
+	int done = 0;
+	int length;
+
+	desc_available = tsnep_rx_desc_available(rx);
+	prog = READ_ONCE(rx->adapter->xdp_prog);
+	if (prog) {
+		tx_nq = netdev_get_tx_queue(rx->adapter->netdev,
+					    rx->tx_queue_index);
+		tx = &rx->adapter->tx[rx->tx_queue_index];
+	}
+
+	while (likely(done < budget) && (rx->read != rx->write)) {
+		entry = &rx->entry[rx->read];
+		if ((__le32_to_cpu(entry->desc_wb->properties) &
+		     TSNEP_DESC_OWNER_COUNTER_MASK) !=
+		    (entry->properties & TSNEP_DESC_OWNER_COUNTER_MASK))
+			break;
+		done++;
+
+		if (desc_available >= TSNEP_RING_RX_REFILL) {
+			bool reuse = desc_available >= TSNEP_RING_RX_REUSE;
+
+			desc_available -= tsnep_rx_refill_zc(rx, desc_available,
+							     reuse);
+			if (!entry->xdp) {
+				/* buffer has been reused for refill to prevent
+				 * empty RX ring, thus buffer cannot be used for
+				 * RX processing
+				 */
+				rx->read = (rx->read + 1) & TSNEP_RING_MASK;
+				desc_available++;
+
+				rx->dropped++;
+
+				continue;
+			}
+		}
+
+		/* descriptor properties shall be read first, because valid data
+		 * is signaled there
+		 */
+		dma_rmb();
+
+		prefetch(entry->xdp->data);
+		length = __le32_to_cpu(entry->desc_wb->properties) &
+			 TSNEP_DESC_LENGTH_MASK;
+		xsk_buff_set_size(entry->xdp, length);
+		xsk_buff_dma_sync_for_cpu(entry->xdp, rx->xsk_pool);
+
+		/* RX metadata with timestamps is in front of actual data,
+		 * subtract metadata size to get length of actual data and
+		 * consider metadata size as offset of actual data during RX
+		 * processing
+		 */
+		length -= TSNEP_RX_INLINE_METADATA_SIZE;
+
+		rx->read = (rx->read + 1) & TSNEP_RING_MASK;
+		desc_available++;
+
+		if (prog) {
+			bool consume;
+
+			entry->xdp->data += TSNEP_RX_INLINE_METADATA_SIZE;
+			entry->xdp->data_meta += TSNEP_RX_INLINE_METADATA_SIZE;
+
+			consume = tsnep_xdp_run_prog_zc(rx, prog, entry->xdp,
+							&xdp_status, tx_nq, tx);
+			if (consume) {
+				rx->packets++;
+				rx->bytes += length;
+
+				entry->xdp = NULL;
+
+				continue;
+			}
+		}
+
+		page = page_pool_dev_alloc_pages(rx->page_pool);
+		if (page) {
+			memcpy(page_address(page) + TSNEP_RX_OFFSET,
+			       entry->xdp->data - TSNEP_RX_INLINE_METADATA_SIZE,
+			       length + TSNEP_RX_INLINE_METADATA_SIZE);
+			tsnep_rx_page(rx, napi, page, length);
+		} else {
+			rx->dropped++;
+		}
+		xsk_buff_free(entry->xdp);
+		entry->xdp = NULL;
+	}
+
+	if (xdp_status)
+		tsnep_finalize_xdp(rx->adapter, xdp_status, tx_nq, tx);
+
+	if (desc_available)
+		desc_available -= tsnep_rx_refill_zc(rx, desc_available, false);
+
+	if (xsk_uses_need_wakeup(rx->xsk_pool)) {
+		if (desc_available)
+			xsk_set_rx_need_wakeup(rx->xsk_pool);
+		else
+			xsk_clear_rx_need_wakeup(rx->xsk_pool);
+
+		return done;
+	}
+
+	return desc_available ? budget : done;
+}
+
 static bool tsnep_rx_pending(struct tsnep_rx *rx)
 {
 	struct tsnep_rx_entry *entry;
@@ -1231,21 +1500,113 @@ static int tsnep_rx_open(struct tsnep_rx *rx)
 	tsnep_rx_init(rx);
 
 	desc_available = tsnep_rx_desc_available(rx);
-	retval = tsnep_rx_alloc(rx, desc_available, false);
+	if (rx->xsk_pool)
+		retval = tsnep_rx_alloc_zc(rx, desc_available, false);
+	else
+		retval = tsnep_rx_alloc(rx, desc_available, false);
 	if (retval != desc_available) {
-		tsnep_rx_ring_cleanup(rx);
+		retval = -ENOMEM;
 
-		return -ENOMEM;
+		goto alloc_failed;
+	}
+
+	/* prealloc pages to prevent allocation failures when XSK pool is
+	 * disabled at runtime
+	 */
+	if (rx->xsk_pool) {
+		retval = tsnep_rx_alloc_page_buffer(rx);
+		if (retval)
+			goto alloc_failed;
 	}
 
 	return 0;
+
+alloc_failed:
+	tsnep_rx_ring_cleanup(rx);
+	return retval;
 }
 
 static void tsnep_rx_close(struct tsnep_rx *rx)
 {
+	if (rx->xsk_pool)
+		tsnep_rx_free_page_buffer(rx);
+
 	tsnep_rx_ring_cleanup(rx);
 }
 
+static void tsnep_rx_reopen(struct tsnep_rx *rx)
+{
+	struct page **page = rx->page_buffer;
+	int i;
+
+	tsnep_rx_init(rx);
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		struct tsnep_rx_entry *entry = &rx->entry[i];
+
+		/* defined initial values for properties are required for
+		 * correct owner counter checking
+		 */
+		entry->desc->properties = 0;
+		entry->desc_wb->properties = 0;
+
+		/* prevent allocation failures by reusing kept pages */
+		if (*page) {
+			tsnep_rx_set_page(rx, entry, *page);
+			tsnep_rx_activate(rx, rx->write);
+			rx->write++;
+
+			*page = NULL;
+			page++;
+		}
+	}
+}
+
+static void tsnep_rx_reopen_xsk(struct tsnep_rx *rx)
+{
+	struct page **page = rx->page_buffer;
+	u32 allocated;
+	int i;
+
+	tsnep_rx_init(rx);
+
+	/* alloc all ring entries except the last one, because ring cannot be
+	 * filled completely, as many buffers as possible is enough as wakeup is
+	 * done if new buffers are available
+	 */
+	allocated = xsk_buff_alloc_batch(rx->xsk_pool, rx->xdp_batch,
+					 TSNEP_RING_SIZE - 1);
+
+	for (i = 0; i < TSNEP_RING_SIZE; i++) {
+		struct tsnep_rx_entry *entry = &rx->entry[i];
+
+		/* keep pages to prevent allocation failures when xsk is
+		 * disabled
+		 */
+		if (entry->page) {
+			*page = entry->page;
+			entry->page = NULL;
+
+			page++;
+		}
+
+		/* defined initial values for properties are required for
+		 * correct owner counter checking
+		 */
+		entry->desc->properties = 0;
+		entry->desc_wb->properties = 0;
+
+		if (allocated) {
+			tsnep_rx_set_xdp(rx, entry,
+					 rx->xdp_batch[allocated - 1]);
+			tsnep_rx_activate(rx, rx->write);
+			rx->write++;
+
+			allocated--;
+		}
+	}
+}
+
 static bool tsnep_pending(struct tsnep_queue *queue)
 {
 	if (queue->tx && tsnep_tx_pending(queue->tx))
@@ -1268,7 +1629,9 @@ static int tsnep_poll(struct napi_struct *napi, int budget)
 		complete = tsnep_tx_poll(queue->tx, budget);
 
 	if (queue->rx) {
-		done = tsnep_rx_poll(queue->rx, napi, budget);
+		done = queue->rx->xsk_pool ?
+		       tsnep_rx_poll_zc(queue->rx, napi, budget) :
+		       tsnep_rx_poll(queue->rx, napi, budget);
 		if (done >= budget)
 			complete = false;
 	}
@@ -1349,8 +1712,12 @@ static void tsnep_queue_close(struct tsnep_queue *queue, bool first)
 
 	tsnep_free_irq(queue, first);
 
-	if (rx && xdp_rxq_info_is_reg(&rx->xdp_rxq))
-		xdp_rxq_info_unreg(&rx->xdp_rxq);
+	if (rx) {
+		if (xdp_rxq_info_is_reg(&rx->xdp_rxq))
+			xdp_rxq_info_unreg(&rx->xdp_rxq);
+		if (xdp_rxq_info_is_reg(&rx->xdp_rxq_zc))
+			xdp_rxq_info_unreg(&rx->xdp_rxq_zc);
+	}
 
 	netif_napi_del(&queue->napi);
 }
@@ -1373,6 +1740,10 @@ static int tsnep_queue_open(struct tsnep_adapter *adapter,
 		else
 			rx->tx_queue_index = 0;
 
+		/* prepare both memory models to eliminate possible registration
+		 * errors when memory model is switched between page pool and
+		 * XSK pool during runtime
+		 */
 		retval = xdp_rxq_info_reg(&rx->xdp_rxq, adapter->netdev,
 					  rx->queue_index, queue->napi.napi_id);
 		if (retval)
@@ -1382,6 +1753,17 @@ static int tsnep_queue_open(struct tsnep_adapter *adapter,
 						    rx->page_pool);
 		if (retval)
 			goto failed;
+		retval = xdp_rxq_info_reg(&rx->xdp_rxq_zc, adapter->netdev,
+					  rx->queue_index, queue->napi.napi_id);
+		if (retval)
+			goto failed;
+		retval = xdp_rxq_info_reg_mem_model(&rx->xdp_rxq_zc,
+						    MEM_TYPE_XSK_BUFF_POOL,
+						    NULL);
+		if (retval)
+			goto failed;
+		if (rx->xsk_pool)
+			xsk_pool_set_rxq_info(rx->xsk_pool, &rx->xdp_rxq_zc);
 	}
 
 	retval = tsnep_request_irq(queue, first);
@@ -1497,6 +1879,67 @@ static int tsnep_netdev_close(struct net_device *netdev)
 	return 0;
 }
 
+int tsnep_enable_xsk(struct tsnep_queue *queue, struct xsk_buff_pool *pool)
+{
+	bool running = netif_running(queue->adapter->netdev);
+	u32 frame_size;
+
+	frame_size = xsk_pool_get_rx_frame_size(pool);
+	if (frame_size < TSNEP_XSK_RX_BUF_SIZE)
+		return -EOPNOTSUPP;
+
+	queue->rx->page_buffer = kcalloc(TSNEP_RING_SIZE,
+					 sizeof(*queue->rx->page_buffer),
+					 GFP_KERNEL);
+	if (!queue->rx->page_buffer)
+		return -ENOMEM;
+	queue->rx->xdp_batch = kcalloc(TSNEP_RING_SIZE,
+				       sizeof(*queue->rx->xdp_batch),
+				       GFP_KERNEL);
+	if (!queue->rx->xdp_batch) {
+		kfree(queue->rx->page_buffer);
+		queue->rx->page_buffer = NULL;
+
+		return -ENOMEM;
+	}
+
+	xsk_pool_set_rxq_info(pool, &queue->rx->xdp_rxq_zc);
+
+	if (running)
+		tsnep_queue_disable(queue);
+
+	queue->rx->xsk_pool = pool;
+
+	if (running) {
+		tsnep_rx_reopen_xsk(queue->rx);
+		tsnep_queue_enable(queue);
+	}
+
+	return 0;
+}
+
+void tsnep_disable_xsk(struct tsnep_queue *queue)
+{
+	bool running = netif_running(queue->adapter->netdev);
+
+	if (running)
+		tsnep_queue_disable(queue);
+
+	tsnep_rx_free_zc(queue->rx);
+
+	queue->rx->xsk_pool = NULL;
+
+	if (running) {
+		tsnep_rx_reopen(queue->rx);
+		tsnep_queue_enable(queue);
+	}
+
+	kfree(queue->rx->xdp_batch);
+	queue->rx->xdp_batch = NULL;
+	kfree(queue->rx->page_buffer);
+	queue->rx->page_buffer = NULL;
+}
+
 static netdev_tx_t tsnep_netdev_xmit_frame(struct sk_buff *skb,
 					   struct net_device *netdev)
 {
@@ -1646,6 +2089,9 @@ static int tsnep_netdev_bpf(struct net_device *dev, struct netdev_bpf *bpf)
 	switch (bpf->command) {
 	case XDP_SETUP_PROG:
 		return tsnep_xdp_setup_prog(adapter, bpf->prog, bpf->extack);
+	case XDP_SETUP_XSK_POOL:
+		return tsnep_xdp_setup_pool(adapter, bpf->xsk.pool,
+					    bpf->xsk.queue_id);
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -1700,6 +2146,24 @@ static int tsnep_netdev_xdp_xmit(struct net_device *dev, int n,
 	return nxmit;
 }
 
+static int tsnep_netdev_xsk_wakeup(struct net_device *dev, u32 queue_id,
+				   u32 flags)
+{
+	struct tsnep_adapter *adapter = netdev_priv(dev);
+	struct tsnep_queue *queue;
+
+	if (queue_id >= adapter->num_rx_queues ||
+	    queue_id >= adapter->num_tx_queues)
+		return -EINVAL;
+
+	queue = &adapter->queue[queue_id];
+
+	if (!napi_if_scheduled_mark_missed(&queue->napi))
+		napi_schedule(&queue->napi);
+
+	return 0;
+}
+
 static const struct net_device_ops tsnep_netdev_ops = {
 	.ndo_open = tsnep_netdev_open,
 	.ndo_stop = tsnep_netdev_close,
@@ -1713,6 +2177,7 @@ static const struct net_device_ops tsnep_netdev_ops = {
 	.ndo_setup_tc = tsnep_tc_setup,
 	.ndo_bpf = tsnep_netdev_bpf,
 	.ndo_xdp_xmit = tsnep_netdev_xdp_xmit,
+	.ndo_xsk_wakeup = tsnep_netdev_xsk_wakeup,
 };
 
 static int tsnep_mac_init(struct tsnep_adapter *adapter)
diff --git a/drivers/net/ethernet/engleder/tsnep_xdp.c b/drivers/net/ethernet/engleder/tsnep_xdp.c
index 4d14cb1fd772..c0513848c547 100644
--- a/drivers/net/ethernet/engleder/tsnep_xdp.c
+++ b/drivers/net/ethernet/engleder/tsnep_xdp.c
@@ -17,3 +17,69 @@ int tsnep_xdp_setup_prog(struct tsnep_adapter *adapter, struct bpf_prog *prog,
 
 	return 0;
 }
+
+static int tsnep_xdp_enable_pool(struct tsnep_adapter *adapter,
+				 struct xsk_buff_pool *pool, u16 queue_id)
+{
+	struct tsnep_queue *queue;
+	int retval;
+
+	if (queue_id >= adapter->num_rx_queues ||
+	    queue_id >= adapter->num_tx_queues)
+		return -EINVAL;
+
+	queue = &adapter->queue[queue_id];
+	if (queue->rx->queue_index != queue_id ||
+	    queue->tx->queue_index != queue_id) {
+		netdev_err(adapter->netdev,
+			   "XSK support only for TX/RX queue pairs\n");
+
+		return -EOPNOTSUPP;
+	}
+
+	retval = xsk_pool_dma_map(pool, adapter->dmadev,
+				  DMA_ATTR_SKIP_CPU_SYNC);
+	if (retval) {
+		netdev_err(adapter->netdev, "failed to map XSK pool\n");
+
+		return retval;
+	}
+
+	retval = tsnep_enable_xsk(queue, pool);
+	if (retval) {
+		xsk_pool_dma_unmap(pool, DMA_ATTR_SKIP_CPU_SYNC);
+
+		return retval;
+	}
+
+	return 0;
+}
+
+static int tsnep_xdp_disable_pool(struct tsnep_adapter *adapter, u16 queue_id)
+{
+	struct xsk_buff_pool *pool;
+	struct tsnep_queue *queue;
+
+	if (queue_id >= adapter->num_rx_queues ||
+	    queue_id >= adapter->num_tx_queues)
+		return -EINVAL;
+
+	pool = xsk_get_pool_from_qid(adapter->netdev, queue_id);
+	if (!pool)
+		return -EINVAL;
+
+	queue = &adapter->queue[queue_id];
+
+	tsnep_disable_xsk(queue);
+
+	xsk_pool_dma_unmap(pool, DMA_ATTR_SKIP_CPU_SYNC);
+
+	return 0;
+}
+
+int tsnep_xdp_setup_pool(struct tsnep_adapter *adapter,
+			 struct xsk_buff_pool *pool, u16 queue_id)
+{
+	return pool ? tsnep_xdp_enable_pool(adapter, pool, queue_id) :
+		      tsnep_xdp_disable_pool(adapter, queue_id);
+}
-- 
cgit v1.2.3-70-g09d2


From cd275c236b3f342cb7d9437ef5adcdeaf42fdd2c Mon Sep 17 00:00:00 2001
From: Gerhard Engleder <gerhard@engleder-embedded.com>
Date: Fri, 21 Apr 2023 21:46:56 +0200
Subject: tsnep: Add XDP socket zero-copy TX support

Send and complete XSK pool frames within TX NAPI context. NAPI context
is triggered by ndo_xsk_wakeup.

Test results with A53 1.2GHz:

xdpsock txonly copy mode, 64 byte frames:
                   pps            pkts           1.00
tx                 284,409        11,398,144
Two CPUs with 100% and 10% utilization.

xdpsock txonly zero-copy mode, 64 byte frames:
                   pps            pkts           1.00
tx                 511,929        5,890,368
Two CPUs with 100% and 1% utilization.

xdpsock l2fwd copy mode, 64 byte frames:
                   pps            pkts           1.00
rx                 248,985        7,315,885
tx                 248,921        7,315,885
Two CPUs with 100% and 10% utilization.

xdpsock l2fwd zero-copy mode, 64 byte frames:
                   pps            pkts           1.00
rx                 254,735        3,039,456
tx                 254,735        3,039,456
Two CPUs with 100% and 4% utilization.

Packet rate increases and CPU utilization is reduced in both cases.

Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep.h      |   2 +
 drivers/net/ethernet/engleder/tsnep_main.c | 130 ++++++++++++++++++++++++++---
 2 files changed, 121 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep.h b/drivers/net/ethernet/engleder/tsnep.h
index d0bea605a1d1..11b29f56aaf9 100644
--- a/drivers/net/ethernet/engleder/tsnep.h
+++ b/drivers/net/ethernet/engleder/tsnep.h
@@ -70,6 +70,7 @@ struct tsnep_tx_entry {
 	union {
 		struct sk_buff *skb;
 		struct xdp_frame *xdpf;
+		bool zc;
 	};
 	size_t len;
 	DEFINE_DMA_UNMAP_ADDR(dma);
@@ -88,6 +89,7 @@ struct tsnep_tx {
 	int read;
 	u32 owner_counter;
 	int increment_owner_counter;
+	struct xsk_buff_pool *xsk_pool;
 
 	u32 packets;
 	u32 bytes;
diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index f300e2963667..84751bb303a6 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -54,6 +54,8 @@
 #define TSNEP_TX_TYPE_SKB_FRAG	BIT(1)
 #define TSNEP_TX_TYPE_XDP_TX	BIT(2)
 #define TSNEP_TX_TYPE_XDP_NDO	BIT(3)
+#define TSNEP_TX_TYPE_XDP	(TSNEP_TX_TYPE_XDP_TX | TSNEP_TX_TYPE_XDP_NDO)
+#define TSNEP_TX_TYPE_XSK	BIT(4)
 
 #define TSNEP_XDP_TX		BIT(0)
 #define TSNEP_XDP_REDIRECT	BIT(1)
@@ -322,13 +324,47 @@ static void tsnep_tx_init(struct tsnep_tx *tx)
 	tx->increment_owner_counter = TSNEP_RING_SIZE - 1;
 }
 
+static void tsnep_tx_enable(struct tsnep_tx *tx)
+{
+	struct netdev_queue *nq;
+
+	nq = netdev_get_tx_queue(tx->adapter->netdev, tx->queue_index);
+
+	__netif_tx_lock_bh(nq);
+	netif_tx_wake_queue(nq);
+	__netif_tx_unlock_bh(nq);
+}
+
+static void tsnep_tx_disable(struct tsnep_tx *tx, struct napi_struct *napi)
+{
+	struct netdev_queue *nq;
+	u32 val;
+
+	nq = netdev_get_tx_queue(tx->adapter->netdev, tx->queue_index);
+
+	__netif_tx_lock_bh(nq);
+	netif_tx_stop_queue(nq);
+	__netif_tx_unlock_bh(nq);
+
+	/* wait until TX is done in hardware */
+	readx_poll_timeout(ioread32, tx->addr + TSNEP_CONTROL, val,
+			   ((val & TSNEP_CONTROL_TX_ENABLE) == 0), 10000,
+			   1000000);
+
+	/* wait until TX is also done in software */
+	while (READ_ONCE(tx->read) != tx->write) {
+		napi_schedule(napi);
+		napi_synchronize(napi);
+	}
+}
+
 static void tsnep_tx_activate(struct tsnep_tx *tx, int index, int length,
 			      bool last)
 {
 	struct tsnep_tx_entry *entry = &tx->entry[index];
 
 	entry->properties = 0;
-	/* xdpf is union with skb */
+	/* xdpf and zc are union with skb */
 	if (entry->skb) {
 		entry->properties = length & TSNEP_DESC_LENGTH_MASK;
 		entry->properties |= TSNEP_DESC_INTERRUPT_FLAG;
@@ -646,10 +682,69 @@ static bool tsnep_xdp_xmit_back(struct tsnep_adapter *adapter,
 	return xmit;
 }
 
+static int tsnep_xdp_tx_map_zc(struct xdp_desc *xdpd, struct tsnep_tx *tx)
+{
+	struct tsnep_tx_entry *entry;
+	dma_addr_t dma;
+
+	entry = &tx->entry[tx->write];
+	entry->zc = true;
+
+	dma = xsk_buff_raw_get_dma(tx->xsk_pool, xdpd->addr);
+	xsk_buff_raw_dma_sync_for_device(tx->xsk_pool, dma, xdpd->len);
+
+	entry->type = TSNEP_TX_TYPE_XSK;
+	entry->len = xdpd->len;
+
+	entry->desc->tx = __cpu_to_le64(dma);
+
+	return xdpd->len;
+}
+
+static void tsnep_xdp_xmit_frame_ring_zc(struct xdp_desc *xdpd,
+					 struct tsnep_tx *tx)
+{
+	int length;
+
+	length = tsnep_xdp_tx_map_zc(xdpd, tx);
+
+	tsnep_tx_activate(tx, tx->write, length, true);
+	tx->write = (tx->write + 1) & TSNEP_RING_MASK;
+}
+
+static void tsnep_xdp_xmit_zc(struct tsnep_tx *tx)
+{
+	int desc_available = tsnep_tx_desc_available(tx);
+	struct xdp_desc *descs = tx->xsk_pool->tx_descs;
+	int batch, i;
+
+	/* ensure that TX ring is not filled up by XDP, always MAX_SKB_FRAGS
+	 * will be available for normal TX path and queue is stopped there if
+	 * necessary
+	 */
+	if (desc_available <= (MAX_SKB_FRAGS + 1))
+		return;
+	desc_available -= MAX_SKB_FRAGS + 1;
+
+	batch = xsk_tx_peek_release_desc_batch(tx->xsk_pool, desc_available);
+	for (i = 0; i < batch; i++)
+		tsnep_xdp_xmit_frame_ring_zc(&descs[i], tx);
+
+	if (batch) {
+		/* descriptor properties shall be valid before hardware is
+		 * notified
+		 */
+		dma_wmb();
+
+		tsnep_xdp_xmit_flush(tx);
+	}
+}
+
 static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
 {
 	struct tsnep_tx_entry *entry;
 	struct netdev_queue *nq;
+	int xsk_frames = 0;
 	int budget = 128;
 	int length;
 	int count;
@@ -676,7 +771,7 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
 		if ((entry->type & TSNEP_TX_TYPE_SKB) &&
 		    skb_shinfo(entry->skb)->nr_frags > 0)
 			count += skb_shinfo(entry->skb)->nr_frags;
-		else if (!(entry->type & TSNEP_TX_TYPE_SKB) &&
+		else if ((entry->type & TSNEP_TX_TYPE_XDP) &&
 			 xdp_frame_has_frags(entry->xdpf))
 			count += xdp_get_shared_info_from_frame(entry->xdpf)->nr_frags;
 
@@ -705,9 +800,11 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
 
 		if (entry->type & TSNEP_TX_TYPE_SKB)
 			napi_consume_skb(entry->skb, napi_budget);
-		else
+		else if (entry->type & TSNEP_TX_TYPE_XDP)
 			xdp_return_frame_rx_napi(entry->xdpf);
-		/* xdpf is union with skb */
+		else
+			xsk_frames++;
+		/* xdpf and zc are union with skb */
 		entry->skb = NULL;
 
 		tx->read = (tx->read + count) & TSNEP_RING_MASK;
@@ -718,6 +815,14 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
 		budget--;
 	} while (likely(budget));
 
+	if (tx->xsk_pool) {
+		if (xsk_frames)
+			xsk_tx_completed(tx->xsk_pool, xsk_frames);
+		if (xsk_uses_need_wakeup(tx->xsk_pool))
+			xsk_set_tx_need_wakeup(tx->xsk_pool);
+		tsnep_xdp_xmit_zc(tx);
+	}
+
 	if ((tsnep_tx_desc_available(tx) >= ((MAX_SKB_FRAGS + 1) * 2)) &&
 	    netif_tx_queue_stopped(nq)) {
 		netif_tx_wake_queue(nq);
@@ -765,12 +870,6 @@ static int tsnep_tx_open(struct tsnep_tx *tx)
 
 static void tsnep_tx_close(struct tsnep_tx *tx)
 {
-	u32 val;
-
-	readx_poll_timeout(ioread32, tx->addr + TSNEP_CONTROL, val,
-			   ((val & TSNEP_CONTROL_TX_ENABLE) == 0), 10000,
-			   1000000);
-
 	tsnep_tx_ring_cleanup(tx);
 }
 
@@ -1786,12 +1885,18 @@ static void tsnep_queue_enable(struct tsnep_queue *queue)
 	napi_enable(&queue->napi);
 	tsnep_enable_irq(queue->adapter, queue->irq_mask);
 
+	if (queue->tx)
+		tsnep_tx_enable(queue->tx);
+
 	if (queue->rx)
 		tsnep_rx_enable(queue->rx);
 }
 
 static void tsnep_queue_disable(struct tsnep_queue *queue)
 {
+	if (queue->tx)
+		tsnep_tx_disable(queue->tx, &queue->napi);
+
 	napi_disable(&queue->napi);
 	tsnep_disable_irq(queue->adapter, queue->irq_mask);
 
@@ -1908,6 +2013,7 @@ int tsnep_enable_xsk(struct tsnep_queue *queue, struct xsk_buff_pool *pool)
 	if (running)
 		tsnep_queue_disable(queue);
 
+	queue->tx->xsk_pool = pool;
 	queue->rx->xsk_pool = pool;
 
 	if (running) {
@@ -1928,6 +2034,7 @@ void tsnep_disable_xsk(struct tsnep_queue *queue)
 	tsnep_rx_free_zc(queue->rx);
 
 	queue->rx->xsk_pool = NULL;
+	queue->tx->xsk_pool = NULL;
 
 	if (running) {
 		tsnep_rx_reopen(queue->rx);
@@ -2438,7 +2545,8 @@ static int tsnep_probe(struct platform_device *pdev)
 
 	netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
 			       NETDEV_XDP_ACT_NDO_XMIT |
-			       NETDEV_XDP_ACT_NDO_XMIT_SG;
+			       NETDEV_XDP_ACT_NDO_XMIT_SG |
+			       NETDEV_XDP_ACT_XSK_ZEROCOPY;
 
 	/* carrier off reporting is important to ethtool even BEFORE open */
 	netif_carrier_off(netdev);
-- 
cgit v1.2.3-70-g09d2


From 700f11eb2cbea349bda2599b4b676b49d43b4175 Mon Sep 17 00:00:00 2001
From: Horatiu Vultur <horatiu.vultur@microchip.com>
Date: Sat, 22 Apr 2023 16:23:44 +0200
Subject: lan966x: Don't use xdp_frame when action is XDP_TX

When the action of an xdp program was XDP_TX, lan966x was creating
a xdp_frame and use this one to send the frame back. But it is also
possible to send back the frame without needing a xdp_frame, because
it is possible to send it back using the page.
And then once the frame is transmitted is possible to use directly
page_pool_recycle_direct as lan966x is using page pools.
This would save some CPU usage on this path, which results in higher
number of transmitted frames. Bellow are the statistics:
Frame size:    Improvement:
64                ~8%
256              ~11%
512               ~8%
1000              ~0%
1500              ~0%

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230422142344.3630602-1-horatiu.vultur@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../net/ethernet/microchip/lan966x/lan966x_fdma.c  | 35 ++++++++++++++--------
 .../net/ethernet/microchip/lan966x/lan966x_main.h  |  6 ++--
 .../net/ethernet/microchip/lan966x/lan966x_xdp.c   | 10 ++-----
 3 files changed, 28 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c b/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c
index 2ed76bb61a73..bd72fbc2220f 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c
@@ -390,6 +390,7 @@ static void lan966x_fdma_stop_netdev(struct lan966x *lan966x)
 static void lan966x_fdma_tx_clear_buf(struct lan966x *lan966x, int weight)
 {
 	struct lan966x_tx *tx = &lan966x->tx;
+	struct lan966x_rx *rx = &lan966x->rx;
 	struct lan966x_tx_dcb_buf *dcb_buf;
 	struct xdp_frame_bulk bq;
 	struct lan966x_db *db;
@@ -432,7 +433,8 @@ static void lan966x_fdma_tx_clear_buf(struct lan966x *lan966x, int weight)
 			if (dcb_buf->xdp_ndo)
 				xdp_return_frame_bulk(dcb_buf->data.xdpf, &bq);
 			else
-				xdp_return_frame_rx_napi(dcb_buf->data.xdpf);
+				page_pool_recycle_direct(rx->page_pool,
+							 dcb_buf->data.page);
 		}
 
 		clear = true;
@@ -699,15 +701,14 @@ static void lan966x_fdma_tx_start(struct lan966x_tx *tx, int next_to_use)
 	tx->last_in_use = next_to_use;
 }
 
-int lan966x_fdma_xmit_xdpf(struct lan966x_port *port,
-			   struct xdp_frame *xdpf,
-			   struct page *page,
-			   bool dma_map)
+int lan966x_fdma_xmit_xdpf(struct lan966x_port *port, void *ptr, u32 len)
 {
 	struct lan966x *lan966x = port->lan966x;
 	struct lan966x_tx_dcb_buf *next_dcb_buf;
 	struct lan966x_tx *tx = &lan966x->tx;
+	struct xdp_frame *xdpf;
 	dma_addr_t dma_addr;
+	struct page *page;
 	int next_to_use;
 	__be32 *ifh;
 	int ret = 0;
@@ -722,8 +723,13 @@ int lan966x_fdma_xmit_xdpf(struct lan966x_port *port,
 		goto out;
 	}
 
+	/* Get the next buffer */
+	next_dcb_buf = &tx->dcbs_buf[next_to_use];
+
 	/* Generate new IFH */
-	if (dma_map) {
+	if (!len) {
+		xdpf = ptr;
+
 		if (xdpf->headroom < IFH_LEN_BYTES) {
 			ret = NETDEV_TX_OK;
 			goto out;
@@ -743,11 +749,16 @@ int lan966x_fdma_xmit_xdpf(struct lan966x_port *port,
 			goto out;
 		}
 
+		next_dcb_buf->data.xdpf = xdpf;
+		next_dcb_buf->len = xdpf->len + IFH_LEN_BYTES;
+
 		/* Setup next dcb */
 		lan966x_fdma_tx_setup_dcb(tx, next_to_use,
 					  xdpf->len + IFH_LEN_BYTES,
 					  dma_addr);
 	} else {
+		page = ptr;
+
 		ifh = page_address(page) + XDP_PACKET_HEADROOM;
 		memset(ifh, 0x0, sizeof(__be32) * IFH_LEN);
 		lan966x_ifh_set_bypass(ifh, 1);
@@ -756,21 +767,21 @@ int lan966x_fdma_xmit_xdpf(struct lan966x_port *port,
 		dma_addr = page_pool_get_dma_addr(page);
 		dma_sync_single_for_device(lan966x->dev,
 					   dma_addr + XDP_PACKET_HEADROOM,
-					   xdpf->len + IFH_LEN_BYTES,
+					   len + IFH_LEN_BYTES,
 					   DMA_TO_DEVICE);
 
+		next_dcb_buf->data.page = page;
+		next_dcb_buf->len = len + IFH_LEN_BYTES;
+
 		/* Setup next dcb */
 		lan966x_fdma_tx_setup_dcb(tx, next_to_use,
-					  xdpf->len + IFH_LEN_BYTES,
+					  len + IFH_LEN_BYTES,
 					  dma_addr + XDP_PACKET_HEADROOM);
 	}
 
 	/* Fill up the buffer */
-	next_dcb_buf = &tx->dcbs_buf[next_to_use];
 	next_dcb_buf->use_skb = false;
-	next_dcb_buf->data.xdpf = xdpf;
-	next_dcb_buf->xdp_ndo = dma_map;
-	next_dcb_buf->len = xdpf->len + IFH_LEN_BYTES;
+	next_dcb_buf->xdp_ndo = !len;
 	next_dcb_buf->dma_addr = dma_addr;
 	next_dcb_buf->used = true;
 	next_dcb_buf->ptp = false;
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
index 851afb0166b1..c977c70abc3d 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h
@@ -243,6 +243,7 @@ struct lan966x_tx_dcb_buf {
 	union {
 		struct sk_buff *skb;
 		struct xdp_frame *xdpf;
+		struct page *page;
 	} data;
 	u32 len;
 	u32 used : 1;
@@ -541,10 +542,7 @@ int lan966x_ptp_setup_traps(struct lan966x_port *port, struct ifreq *ifr);
 int lan966x_ptp_del_traps(struct lan966x_port *port);
 
 int lan966x_fdma_xmit(struct sk_buff *skb, __be32 *ifh, struct net_device *dev);
-int lan966x_fdma_xmit_xdpf(struct lan966x_port *port,
-			   struct xdp_frame *frame,
-			   struct page *page,
-			   bool dma_map);
+int lan966x_fdma_xmit_xdpf(struct lan966x_port *port, void *ptr, u32 len);
 int lan966x_fdma_change_mtu(struct lan966x *lan966x);
 void lan966x_fdma_netdev_init(struct lan966x *lan966x, struct net_device *dev);
 void lan966x_fdma_netdev_deinit(struct lan966x *lan966x, struct net_device *dev);
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_xdp.c b/drivers/net/ethernet/microchip/lan966x/lan966x_xdp.c
index 2e6f486ec67d..9ee61db8690b 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_xdp.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_xdp.c
@@ -62,7 +62,7 @@ int lan966x_xdp_xmit(struct net_device *dev,
 		struct xdp_frame *xdpf = frames[i];
 		int err;
 
-		err = lan966x_fdma_xmit_xdpf(port, xdpf, NULL, true);
+		err = lan966x_fdma_xmit_xdpf(port, xdpf, 0);
 		if (err)
 			break;
 
@@ -76,7 +76,6 @@ int lan966x_xdp_run(struct lan966x_port *port, struct page *page, u32 data_len)
 {
 	struct bpf_prog *xdp_prog = port->xdp_prog;
 	struct lan966x *lan966x = port->lan966x;
-	struct xdp_frame *xdpf;
 	struct xdp_buff xdp;
 	u32 act;
 
@@ -90,11 +89,8 @@ int lan966x_xdp_run(struct lan966x_port *port, struct page *page, u32 data_len)
 	case XDP_PASS:
 		return FDMA_PASS;
 	case XDP_TX:
-		xdpf = xdp_convert_buff_to_frame(&xdp);
-		if (!xdpf)
-			return FDMA_DROP;
-
-		return lan966x_fdma_xmit_xdpf(port, xdpf, page, false) ?
+		return lan966x_fdma_xmit_xdpf(port, page,
+					      data_len - IFH_LEN_BYTES) ?
 		       FDMA_DROP : FDMA_TX;
 	case XDP_REDIRECT:
 		if (xdp_do_redirect(port->dev, &xdp, xdp_prog))
-- 
cgit v1.2.3-70-g09d2


From 4774ad841bef97cc51df90195338c5b2573dd4cb Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Sun, 23 Apr 2023 19:28:00 +0200
Subject: net: phy: marvell: Fix inconsistent indenting in led_blink_set

Fix inconsistent indeinting in m88e1318_led_blink_set reported by kernel
test robot, probably done by the presence of an if condition dropped in
later revision of the same code.

Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304240007.0VEX8QYG-lkp@intel.com/
Fixes: ea9e86485dec ("net: phy: marvell: Implement led_blink_set()")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230423172800.3470-1-ansuelsmth@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/marvell.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index cd5d0ed9c5e5..43b6cb725551 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -2880,10 +2880,10 @@ static int m88e1318_led_blink_set(struct phy_device *phydev, u8 index,
 	case 1:
 	case 2:
 		reg &= ~(0xf << (4 * index));
-			reg |= MII_88E1318S_PHY_LED_FUNC_BLINK << (4 * index);
-			/* Reset default is 84ms */
-			*delay_on = 84 / 2;
-			*delay_off = 84 / 2;
+		reg |= MII_88E1318S_PHY_LED_FUNC_BLINK << (4 * index);
+		/* Reset default is 84ms */
+		*delay_on = 84 / 2;
+		*delay_off = 84 / 2;
 		break;
 	default:
 		return -EINVAL;
-- 
cgit v1.2.3-70-g09d2


From d325c34d9e7e38d371c0a299d415e9b07f66a1fb Mon Sep 17 00:00:00 2001
From: Gencen Gan <gangecen@hust.edu.cn>
Date: Mon, 24 Apr 2023 23:28:01 +0800
Subject: net: amd: Fix link leak when verifying config failed

After failing to verify configuration, it returns directly without
releasing link, which may cause memory leak.

Paolo Abeni thinks that the whole code of this driver is quite
"suboptimal" and looks unmainatained since at least ~15y, so he
suggests that we could simply remove the whole driver, please
take it into consideration.

Simon Horman suggests that the fix label should be set to
"Linux-2.6.12-rc2" considering that the problem has existed
since the driver was introduced and the commit above doesn't
seem to exist in net/net-next.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Gan Gecen <gangecen@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/amd/nmclan_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/amd/nmclan_cs.c b/drivers/net/ethernet/amd/nmclan_cs.c
index 823a329a921f..0dd391c84c13 100644
--- a/drivers/net/ethernet/amd/nmclan_cs.c
+++ b/drivers/net/ethernet/amd/nmclan_cs.c
@@ -651,7 +651,7 @@ static int nmclan_config(struct pcmcia_device *link)
     } else {
       pr_notice("mace id not found: %x %x should be 0x40 0x?9\n",
 		sig[0], sig[1]);
-      return -ENODEV;
+      goto failed;
     }
   }
 
-- 
cgit v1.2.3-70-g09d2


From 50749f2dd6854a41830996ad302aef2ffaf011d8 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Mon, 24 Apr 2023 15:20:22 -0700
Subject: tcp/udp: Fix memleaks of sk and zerocopy skbs with TX timestamp.

syzkaller reported [0] memory leaks of an UDP socket and ZEROCOPY
skbs.  We can reproduce the problem with these sequences:

  sk = socket(AF_INET, SOCK_DGRAM, 0)
  sk.setsockopt(SOL_SOCKET, SO_TIMESTAMPING, SOF_TIMESTAMPING_TX_SOFTWARE)
  sk.setsockopt(SOL_SOCKET, SO_ZEROCOPY, 1)
  sk.sendto(b'', MSG_ZEROCOPY, ('127.0.0.1', 53))
  sk.close()

sendmsg() calls msg_zerocopy_alloc(), which allocates a skb, sets
skb->cb->ubuf.refcnt to 1, and calls sock_hold().  Here, struct
ubuf_info_msgzc indirectly holds a refcnt of the socket.  When the
skb is sent, __skb_tstamp_tx() clones it and puts the clone into
the socket's error queue with the TX timestamp.

When the original skb is received locally, skb_copy_ubufs() calls
skb_unclone(), and pskb_expand_head() increments skb->cb->ubuf.refcnt.
This additional count is decremented while freeing the skb, but struct
ubuf_info_msgzc still has a refcnt, so __msg_zerocopy_callback() is
not called.

The last refcnt is not released unless we retrieve the TX timestamped
skb by recvmsg().  Since we clear the error queue in inet_sock_destruct()
after the socket's refcnt reaches 0, there is a circular dependency.
If we close() the socket holding such skbs, we never call sock_put()
and leak the count, sk, and skb.

TCP has the same problem, and commit e0c8bccd40fc ("net: stream:
purge sk_error_queue in sk_stream_kill_queues()") tried to fix it
by calling skb_queue_purge() during close().  However, there is a
small chance that skb queued in a qdisc or device could be put
into the error queue after the skb_queue_purge() call.

In __skb_tstamp_tx(), the cloned skb should not have a reference
to the ubuf to remove the circular dependency, but skb_clone() does
not call skb_copy_ubufs() for zerocopy skb.  So, we need to call
skb_orphan_frags_rx() for the cloned skb to call skb_copy_ubufs().

[0]:
BUG: memory leak
unreferenced object 0xffff88800c6d2d00 (size 1152):
  comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 cd af e8 81 00 00 00 00  ................
    02 00 07 40 00 00 00 00 00 00 00 00 00 00 00 00  ...@............
  backtrace:
    [<0000000055636812>] sk_prot_alloc+0x64/0x2a0 net/core/sock.c:2024
    [<0000000054d77b7a>] sk_alloc+0x3b/0x800 net/core/sock.c:2083
    [<0000000066f3c7e0>] inet_create net/ipv4/af_inet.c:319 [inline]
    [<0000000066f3c7e0>] inet_create+0x31e/0xe40 net/ipv4/af_inet.c:245
    [<000000009b83af97>] __sock_create+0x2ab/0x550 net/socket.c:1515
    [<00000000b9b11231>] sock_create net/socket.c:1566 [inline]
    [<00000000b9b11231>] __sys_socket_create net/socket.c:1603 [inline]
    [<00000000b9b11231>] __sys_socket_create net/socket.c:1588 [inline]
    [<00000000b9b11231>] __sys_socket+0x138/0x250 net/socket.c:1636
    [<000000004fb45142>] __do_sys_socket net/socket.c:1649 [inline]
    [<000000004fb45142>] __se_sys_socket net/socket.c:1647 [inline]
    [<000000004fb45142>] __x64_sys_socket+0x73/0xb0 net/socket.c:1647
    [<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
    [<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
    [<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

BUG: memory leak
unreferenced object 0xffff888017633a00 (size 240):
  comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 2d 6d 0c 80 88 ff ff  .........-m.....
  backtrace:
    [<000000002b1c4368>] __alloc_skb+0x229/0x320 net/core/skbuff.c:497
    [<00000000143579a6>] alloc_skb include/linux/skbuff.h:1265 [inline]
    [<00000000143579a6>] sock_omalloc+0xaa/0x190 net/core/sock.c:2596
    [<00000000be626478>] msg_zerocopy_alloc net/core/skbuff.c:1294 [inline]
    [<00000000be626478>] msg_zerocopy_realloc+0x1ce/0x7f0 net/core/skbuff.c:1370
    [<00000000cbfc9870>] __ip_append_data+0x2adf/0x3b30 net/ipv4/ip_output.c:1037
    [<0000000089869146>] ip_make_skb+0x26c/0x2e0 net/ipv4/ip_output.c:1652
    [<00000000098015c2>] udp_sendmsg+0x1bac/0x2390 net/ipv4/udp.c:1253
    [<0000000045e0e95e>] inet_sendmsg+0x10a/0x150 net/ipv4/af_inet.c:819
    [<000000008d31bfde>] sock_sendmsg_nosec net/socket.c:714 [inline]
    [<000000008d31bfde>] sock_sendmsg+0x141/0x190 net/socket.c:734
    [<0000000021e21aa4>] __sys_sendto+0x243/0x360 net/socket.c:2117
    [<00000000ac0af00c>] __do_sys_sendto net/socket.c:2129 [inline]
    [<00000000ac0af00c>] __se_sys_sendto net/socket.c:2125 [inline]
    [<00000000ac0af00c>] __x64_sys_sendto+0xe1/0x1c0 net/socket.c:2125
    [<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
    [<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
    [<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

Fixes: f214f915e7db ("tcp: enable MSG_ZEROCOPY")
Fixes: b5947e5d1e71 ("udp: msg_zerocopy")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 4c0879798eb8..2f9bb98170ab 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5162,6 +5162,9 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb,
 			skb = alloc_skb(0, GFP_ATOMIC);
 	} else {
 		skb = skb_clone(orig_skb, GFP_ATOMIC);
+
+		if (skb_orphan_frags_rx(skb, GFP_ATOMIC))
+			return;
 	}
 	if (!skb)
 		return;
-- 
cgit v1.2.3-70-g09d2


From 28b17f6270f182e22cdad5a0fdc4979031e4486a Mon Sep 17 00:00:00 2001
From: wuych <yunchuan@nfschina.com>
Date: Tue, 25 Apr 2023 13:15:32 +0800
Subject: net: phy: marvell-88x2222: remove unnecessary (void*) conversions

Pointer variables of void * type do not require type cast.

Signed-off-by: wuych <yunchuan@nfschina.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/marvell-88x2222.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/marvell-88x2222.c b/drivers/net/phy/marvell-88x2222.c
index fd9ad4820192..f83cae64585d 100644
--- a/drivers/net/phy/marvell-88x2222.c
+++ b/drivers/net/phy/marvell-88x2222.c
@@ -487,7 +487,7 @@ static int mv2222_sfp_insert(void *upstream, const struct sfp_eeprom_id *id)
 
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(sfp_supported) = { 0, };
 
-	priv = (struct mv2222_data *)phydev->priv;
+	priv = phydev->priv;
 	dev = &phydev->mdio.dev;
 
 	sfp_parse_support(phydev->sfp_bus, id, sfp_supported, interfaces);
@@ -524,7 +524,7 @@ static void mv2222_sfp_remove(void *upstream)
 	struct phy_device *phydev = upstream;
 	struct mv2222_data *priv;
 
-	priv = (struct mv2222_data *)phydev->priv;
+	priv = phydev->priv;
 
 	priv->line_interface = PHY_INTERFACE_MODE_NA;
 	linkmode_zero(priv->supported);
-- 
cgit v1.2.3-70-g09d2


From 9b78d919632b7149d311aaad5a977e4b48b10321 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Wed, 26 Apr 2023 10:15:31 +0200
Subject: net: phy: hide the PHYLIB_LEDS knob

commit 4bb7aac70b5d ("net: phy: fix circular LEDS_CLASS dependencies")
solved a build failure, but introduces a new config knob with a default
'y' value: PHYLIB_LEDS.

The latter is against the current new config policy. The exception
was raised to allow the user to catch bad configurations without led
support.

Anyway the current definition of PHYLIB_LEDS does not fit the above
goal: if LEDS_CLASS is disabled, the new config will be available
only with PHYLIB disabled, too.

Hide the mentioned config, to preserve the randconfig testing done so
far, while respecting the mentioned policy.

Suggested-by: Andrew Lunn <andrew@lunn.ch>
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/d82489be8ed911c383c3447e9abf469995ccf39a.1682496488.git.pabeni@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/phy/Kconfig | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 2f3ddc446cbb..93b8efc79227 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -45,10 +45,8 @@ config LED_TRIGGER_PHY
 		for any speed known to the PHY.
 
 config PHYLIB_LEDS
-	bool "Support probing LEDs from device tree"
+	def_bool OF
 	depends on LEDS_CLASS=y || LEDS_CLASS=PHYLIB
-	depends on OF
-	default y
 	help
 	  When LED class support is enabled, phylib can automatically
 	  probe LED setting from device tree.
-- 
cgit v1.2.3-70-g09d2